diff --git a/config.json b/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..894c25000fca1fcbfdc4633e05386c7c15f1f2da
--- /dev/null
+++ b/config.json
@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "/fs/archive/share/yulan/data/aa_mini/output/miniyulan-2B-final-stage13-from-125.75k/checkpoint-126516-rms_norm",
+  "architectures": [
+    "MiniYuLanModelForCausalLM"
+  ],
+  "attention_bias": true,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dim_model_base": 1920,
+  "dim_model_base_attn": 64,
+  "dim_model_base_init": null,
+  "dim_model_base_lmh": 1,
+  "dim_model_base_logits": 1920.0,
+  "dim_model_base_lr": 256.0,
+  "down_proj_alpha": 0.03450327796711771,
+  "embed_tokens_alpha": 1,
+  "embedding_ln": false,
+  "embedding_rmsln": false,
+  "eos_token_id": 2,
+  "gate_up_proj_alpha": 0.3651483716701107,
+  "gradient_checkpointing_step": 11,
+  "hidden_act": "silu",
+  "hidden_size": 1920,
+  "hidden_states_shrink": 0.18708286933869706,
+  "init_scale_o": 1,
+  "initializer_range": 5e-05,
+  "input_layernorm_alpha": 1.0,
+  "intermediate_size": 4800,
+  "k_proj_alpha": 0.3651483716701107,
+  "layer_norm_eps": 1e-06,
+  "lm_head_alpha": 1.0,
+  "ln_scale": 1,
+  "max_position_embeddings": 4096,
+  "model_reproduce": "transformer",
+  "model_type": "miniyulan",
+  "norm_alpha": 1.0,
+  "num_attention_heads": 30,
+  "num_hidden_layers": 56,
+  "num_key_value_heads": 6,
+  "o_proj_alpha": 0.03450327796711771,
+  "post_attention_layernorm_alpha": 1.0,
+  "q_proj_alpha": 0.3651483716701107,
+  "qk_layernorm": false,
+  "rms_norm_eps": 1e-06,
+  "rms_type": "llama",
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "scale_emb": 10.0,
+  "shrink_alpha": 1,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.44.0",
+  "use_cache": false,
+  "use_emb_alpha": true,
+  "use_liger": true,
+  "use_norm_alpha": true,
+  "use_sliding_window": false,
+  "v_proj_alpha": 0.3651483716701107,
+  "vocab_size": 99000,
+  "wesar_weights": true,
+  "z_loss": 0.0001
+}
diff --git a/global_step136346/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2db5cad78117a6290a7a4fcc58de03add3b4066c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f34097767d7d6e4c72a0adc0bd8679b507783c5dbf96ff1576b6e689e128ec4c
+size 558554482
diff --git a/global_step136346/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7655c47474ac9d1bf8962061bec2459f19c84c46
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6b0d15069d4f8d83e05739a903ee9eef85705753da8bdc9d5ef26513063f9f6
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ab41d1c1dafa9106c7aa09952d6354aa12682f1b
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2308b5f24cdc1e447d28eaa75c383d2997f343743746dd3a2ee33797f61d2da1
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d391548e08f54e62d643358d2eb529a5a74d36ec
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a83f324b0c27abe3cc94b58d9e4a6ec8207ecc4fff0d2e19087e50ff1c09a717
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..776a422a1746511e870de9e8321331164f186cb8
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53d2bc4d7983d82e41e5cc1dde8b642a26a197af6267a366d37d29e256f99d70
+size 558554434
diff --git a/global_step136346/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ea8090b780ddd985d4e9a0fccebdb7678a27d2a2
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:639c09f2309e583c004044a6586d61a4dac4561106556722d979e4e5bd9c3b52
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bb1d039e4cffae2f84d4465be4cfe851ad0ac99c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b2f33fa08471542722411c5f49270015b421414dd63576b7b108ccbeacd0606
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cebd0525348a4a371b2448841684400e87020139
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a05a154f41e993f9b800ca86b00f78fa57f1840aba12895679d9f77f84d0f481
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..84c8505bc62b4e1a2d976f4bdbfc171605ecaae8
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c42e045762d2e5a2f7864c11538a1c6bdefbca07d53a87e1c12ede79bd1c7444
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..458984aca29bff37efc67c674b8a04e914d256d9
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adc7e18ac518bfa23ab3a18175ea333ed8e14beda20a746987679e7432d77b0a
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..52d1792d325ac4a648b3a158bc33324e2db121e6
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:496e1592455204dbf5aa7eba7d9206807b097cebdfdfc75f343bf9e0e6c62c37
+size 558554434
diff --git a/global_step136346/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f083a2ad18955cb13a14165a5c4243de1b69278c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24cf033eb3fff9bbfd40f46386a9452466be42b93184a27ad3825dd492599467
+size 558554418
diff --git a/global_step136346/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..06e373911d0e17ff354b49466847535116a1f505
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43a450871737799f2d698565f32bdeb8c2b7387fd09f7e1e142e45a4bebd6877
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1c5a5e0cc9bc7a27c9dc11dea4b9e28d666097fc
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3da3d8a6aa0bae7e739c7a0f1b809e0b52b654b9cc4e3a2d7f62db4812d657a
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5ba3714971eb911e981612925386d836705e7b50
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2741e12ebe2b62d5d74f0368871b0c3e467a8fa892b1830bf88caf3ee4bcf87b
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..44ef347b2c9ece1dac0d9870de0dfad817dd0dbc
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a17cb4ff34e93ac4860dfb405413c35c6dfb6055a8d5f22b0b71bde872562aec
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..386d6df7aeeafb9f4e91665fb5add49ac07838e9
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c47b02f0d69dc1676c3260c8f4d26d2d03f8bc8b1ad12dee88aac85b1d93bf14
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0bc49b2f1128e2fa7de7ed3b932e7c80c25a9788
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ef871538a04ac6df290bdb2ba8b9f4dda2a00af597f22b5838df57c9e633989
+size 558554434
diff --git a/global_step136346/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e3a8bedcfe76b1c868f5b9fc4e49ae7d2bbdd659
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9f9e98e551f84e37121e22d57bbd4465c68d2f935189d47ee71259d23be77b4
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..57ee686b7f9a66bc9df11fad00530a5402d492d6
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e7c55218f29e5e55cd358de2c7061d3fca118f77f5887a1b656eb8e8288144
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..80858285834b2c7d82fefa2c49e1581203e79785
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4760ac03174ffe9c108b8d25882aab1eb69678adff70af61d854a40763eb2c2c
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c0f5e51e60ab13432fb8d5214415c580691a0ac9
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:035414df309d2b3b017a1529dc59c47965cb75ef530a889b9bd1514200f98300
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..eb4b6791552b21b5ecb4a359ddf9b5232931e2ac
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53796fe793a1d8ff30af8d83f0a730e89ff3edf1dd10ee956c7281ce6fc581dc
+size 558554290
diff --git a/global_step136346/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..41be3edc260c118d4cbec37caca51df9c53c6d0c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f409ff7010900545bba25ae7c757240677afb2a7eaf3eb1e58e405e86b5bf420
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..aec2d6ddfe6632a7ab09ae108383650944bed58e
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d57238ade3be1b5f0be7764bc7710a0526ef7af7c34463afd67c7fe77a27aa0
+size 558554434
diff --git a/global_step136346/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..513675113b929171bb682a35b442b4fb55aec720
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6461684f09b0243ebbe8c39e582aa3e970023aff9cb469e49209f691fb5f53d
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f9823d8da23864f90e19984a40b49788c43e506c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd8a07889268f15934cea78f5db66225260ee00dae12ca643faf3d3f7e3e629d
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8d3cff634f25dd804d2cc9bdeaa9faa1cbd9db13
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44fbcfd64c622f06c3ad50708be605de66b41401d8e3d87cfad082b12dab07b3
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fe7ccc877918fb27557c2ada23326d04d19d4d6b
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:687ebe86d6a60944fffad668b5e897049e4766ad7848d7a0cbb1b64ca4065907
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3011fcaf96407be7f57d65f2b5c307eedac9c3fe
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80275eff07c93cb4ac2e3a010d310d427afa29069302695935e459c434ad418b
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bc59a3f60d14c63fac7ca7a8a9ec88eb4ccbee29
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1d81d202eb5be5c24981095486f9219d25e7661e0d2de3699668c963913c712
+size 558554434
diff --git a/global_step136346/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..583edf1fbda7080772d6ea9562de98686ee9b260
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee9be424f3480983e62e7bca7dcab1ae241dd5cff6b8d941791e3c714f2f401b
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..19fa84bb33c26b28a97481ca63a2edfdf7b8e395
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:077c07b7029e3998c8bc0a0f79c6ad0bf9c54c7348a4a037275061ccd5f77766
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..32b177cd275fdbca91e8e7e7e1de37304fa23dd1
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa2fb459f5f956a954c13fd686bab86dcfa8f63818f18a7c43a6aeff7b09b591
+size 558554354
diff --git a/global_step136346/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cd16b22f5d426690811f3d43cdc97c74de4e571b
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32c5c764e9eff34c626e97b06108f905ac1aecd6ac62b6ec7709df151aa226bc
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..092d9e0f17614e1785b75620d43f0fa6eeb8e396
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b26b015ddd74f40a9aa00af42c8b9dcf97b4a0d351424a00b49e5d8c7f430eff
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1e17905e0fef0c813495a82e3e30c5231cddcd0c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71fec012bbd30141624763b3e656fe197adbfc1f25cc84a995f6e9a285ca95a0
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1c0e78c8e3a4c94f4151378c91a63618f73f06ec
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6a5f10030a6aa54d9f7aa2f609f85d9b7d5511059f1416477e3fef40f448d6d
+size 558554434
diff --git a/global_step136346/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1f77d36704055ff1c332c885ef98bc5866fb0da3
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00d55fcfa5caa7f95a9cf70f27dcd7a5b98c02776c4df633d52da7e490dd2296
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3aabfc894e3c8af3c852291f9dc5e4d9d6f1f3c0
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daae8287146bf55a33457e939b3d6643b688a20f4cf4fa520fcb7bea08a7f79d
+size 558554370
diff --git a/global_step136346/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8e829ab2a068aff8079151cae85fef6839e9b5cb
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea00d8bd81883857849f0e279c57571c9a61453bcce7b13757725d4cdbc443fa
+size 558554306
diff --git a/global_step136346/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d9f12a0dd8c7ea2bc85def975ca3a9e9ec793233
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bdbe75d350746f14ec3bc81f921f0108196fc298dedab9c1969a1aacc13502b
+size 558610626
diff --git a/global_step136346/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e89990c528eca387ae9d9b82553ff268551c588c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:022ef3a43dd59f0b63806197917d0432fcbf15c5997bdc2978657d134971a33e
+size 558554290
diff --git a/global_step136346/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..75cc051366ba642db3058cb86a940dc0c64ac781
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7845ce21cbdf5c911f35ea6041cc5e21120624a9eee8e75f1d3227f325086edb
+size 558554290
diff --git a/global_step136346/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e42a4b2ffbf1205637fee5a66d5bcca346f2c672
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fd6b07a6f23b9ccdfcac6f52ace80da810c4f74a7a4878f488cb6f7d6b220dd
+size 558554354
diff --git a/global_step136346/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fb68ab681184301ba557cec46173b2c63098f9a0
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e300e5cfe2bc4132197449cdc84e6af1af25d39401b9d017454dc05aa9ca15e
+size 558554418
diff --git a/global_step136346/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..de7bb8bf59d0dfcda891ed17e85fecf1cd65b50c
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdc6b140efc2274ee6fc41c0e7098650a8f8addb648d303cda5b40237b9913e3
+size 558554290
diff --git a/global_step136346/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt b/global_step136346/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..42787e1b684ec4659372f50c4d5c4de5b3fe7c53
--- /dev/null
+++ b/global_step136346/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c56d5dc7491422748cb9439e46af2c10fd719de93fbf8cdb79fdcfd43ca40ac1
+size 558554354
diff --git a/global_step136346/mp_rank_00_model_states.pt b/global_step136346/mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..32fe3d02636e50e75d34fad262de562989db6f28
--- /dev/null
+++ b/global_step136346/mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d6db194266f82b58cdbe4310e08b09779068b7173aac56957574d4cb9c0541e
+size 4468641136
diff --git a/latest b/latest
new file mode 100644
index 0000000000000000000000000000000000000000..21dc937643b5bf07f4d1a2423e60aae12233fbe0
--- /dev/null
+++ b/latest
@@ -0,0 +1 @@
+global_step136346
\ No newline at end of file
diff --git a/model.safetensors b/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0efc084fbeb01b460d06bacaa325501d8b23ab42
--- /dev/null
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8b48e046519cf41172ac814bad216fca331bea034efda161c14467dc9aefb65
+size 4848661852
diff --git a/special_tokens_map.json b/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..a782b2f1cdab4d0bacb2dc0f85d02c4b1e31f0bd
--- /dev/null
+++ b/special_tokens_map.json
@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/tokenizer.json b/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..93efa0d862564bfac68cec7e18d2b451fa56f1aa
--- /dev/null
+++ b/tokenizer.json
@@ -0,0 +1,226270 @@
+{
+  "version": "1.0",
+  "truncation": null,
+  "padding": null,
+  "added_tokens": [
+    {
+      "id": 0,
+      "content": "<unk>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 1,
+      "content": "<s>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 2,
+      "content": "</s>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 102,
+      "content": "<pad>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  ],
+  "normalizer": {
+    "type": "Sequence",
+    "normalizers": [
+      {
+        "type": "Prepend",
+        "prepend": "▁"
+      },
+      {
+        "type": "Replace",
+        "pattern": {
+          "String": " "
+        },
+        "content": "▁"
+      },
+      {
+        "type": "Replace",
+        "pattern": {
+          "String": "　"
+        },
+        "content": "▁"
+      }
+    ]
+  },
+  "pre_tokenizer": null,
+  "post_processor": {
+    "type": "TemplateProcessing",
+    "single": [
+      {
+        "SpecialToken": {
+          "id": "<s>",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      }
+    ],
+    "pair": [
+      {
+        "SpecialToken": {
+          "id": "<s>",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "SpecialToken": {
+          "id": "<s>",
+          "type_id": 1
+        }
+      },
+      {
+        "Sequence": {
+          "id": "B",
+          "type_id": 1
+        }
+      }
+    ],
+    "special_tokens": {
+      "<s>": {
+        "id": "<s>",
+        "ids": [
+          1
+        ],
+        "tokens": [
+          "<s>"
+        ]
+      }
+    }
+  },
+  "decoder": {
+    "type": "Sequence",
+    "decoders": [
+      {
+        "type": "Replace",
+        "pattern": {
+          "String": "▁"
+        },
+        "content": " "
+      },
+      {
+        "type": "ByteFallback"
+      },
+      {
+        "type": "Fuse"
+      },
+      {
+        "type": "Strip",
+        "content": " ",
+        "start": 1,
+        "stop": 0
+      }
+    ]
+  },
+  "model": {
+    "type": "BPE",
+    "dropout": null,
+    "unk_token": "<unk>",
+    "continuing_subword_prefix": null,
+    "end_of_word_suffix": null,
+    "fuse_unk": true,
+    "byte_fallback": true,
+    "ignore_merges": false,
+    "vocab": {
+      "<unk>": 0,
+      "<s>": 1,
+      "</s>": 2,
+      "<SEP>": 3,
+      "<CLS>": 4,
+      "\n": 5,
+      "\t": 6,
+      "<img>": 7,
+      "<img/>": 8,
+      "</img>": 9,
+      "<h2>": 10,
+      "<h2/>": 11,
+      "</h2>": 12,
+      "<td>": 13,
+      "<td/>": 14,
+      "</td>": 15,
+      "<strong>": 16,
+      "<strong/>": 17,
+      "</strong>": 18,
+      "<table>": 19,
+      "<table/>": 20,
+      "</table>": 21,
+      "<tr>": 22,
+      "<tr/>": 23,
+      "</tr>": 24,
+      "<li>": 25,
+      "<li/>": 26,
+      "</li>": 27,
+      "<b>": 28,
+      "<b/>": 29,
+      "</b>": 30,
+      "<h3>": 31,
+      "<h3/>": 32,
+      "</h3>": 33,
+      "<br>": 34,
+      "<br/>": 35,
+      "</br>": 36,
+      "<h4>": 37,
+      "<h4/>": 38,
+      "</h4>": 39,
+      "<h5>": 40,
+      "<h5/>": 41,
+      "</h5>": 42,
+      "<p>": 43,
+      "<p/>": 44,
+      "</p>": 45,
+      "<h1>": 46,
+      "<h1/>": 47,
+      "</h1>": 48,
+      "<tbody>": 49,
+      "<tbody/>": 50,
+      "</tbody>": 51,
+      "０": 52,
+      "１": 53,
+      "２": 54,
+      "３": 55,
+      "４": 56,
+      "５": 57,
+      "６": 58,
+      "７": 59,
+      "８": 60,
+      "９": 61,
+      "+": 62,
+      "-": 63,
+      "=": 64,
+      "，": 65,
+      "。": 66,
+      "！": 67,
+      "？": 68,
+      "、": 69,
+      "：": 70,
+      "￥": 71,
+      ".": 72,
+      "!": 73,
+      "?": 74,
+      "...": 75,
+      "。。。": 76,
+      "。。。。。。": 77,
+      "《": 78,
+      "》": 79,
+      "【": 80,
+      "】": 81,
+      "『": 82,
+      "』": 83,
+      "```": 84,
+      "<!--": 85,
+      "-->": 86,
+      "---": 87,
+      "<!DOCTYPE>": 88,
+      "；": 89,
+      "．": 90,
+      "＝": 91,
+      "＜": 92,
+      "＞": 93,
+      "－": 94,
+      "＋": 95,
+      "％": 96,
+      "‼": 97,
+      "㊣": 98,
+      "／": 99,
+      "｜": 100,
+      "<url>": 101,
+      "<pad>": 102,
+      "<reserved_14>": 103,
+      "<reserved_15>": 104,
+      "<reserved_16>": 105,
+      "<reserved_17>": 106,
+      "<reserved_18>": 107,
+      "<reserved_19>": 108,
+      "<reserved_20>": 109,
+      "<reserved_21>": 110,
+      "<reserved_22>": 111,
+      "<reserved_23>": 112,
+      "<reserved_24>": 113,
+      "<reserved_25>": 114,
+      "<reserved_26>": 115,
+      "<reserved_27>": 116,
+      "<reserved_28>": 117,
+      "<reserved_29>": 118,
+      "<reserved_30>": 119,
+      "<reserved_31>": 120,
+      "<reserved_32>": 121,
+      "<reserved_33>": 122,
+      "<reserved_34>": 123,
+      "<reserved_35>": 124,
+      "<reserved_36>": 125,
+      "<reserved_37>": 126,
+      "<reserved_38>": 127,
+      "<reserved_39>": 128,
+      "<reserved_40>": 129,
+      "<reserved_41>": 130,
+      "<reserved_42>": 131,
+      "<reserved_43>": 132,
+      "<reserved_44>": 133,
+      "<reserved_45>": 134,
+      "<reserved_46>": 135,
+      "<reserved_47>": 136,
+      "<reserved_48>": 137,
+      "<reserved_49>": 138,
+      "<reserved_50>": 139,
+      "<reserved_51>": 140,
+      "<reserved_52>": 141,
+      "<reserved_53>": 142,
+      "<reserved_54>": 143,
+      "<reserved_55>": 144,
+      "<reserved_56>": 145,
+      "<reserved_57>": 146,
+      "<reserved_58>": 147,
+      "<reserved_59>": 148,
+      "<reserved_60>": 149,
+      "<reserved_61>": 150,
+      "<reserved_62>": 151,
+      "<reserved_63>": 152,
+      "<reserved_64>": 153,
+      "<reserved_65>": 154,
+      "<reserved_66>": 155,
+      "<reserved_67>": 156,
+      "<reserved_68>": 157,
+      "<reserved_69>": 158,
+      "<reserved_70>": 159,
+      "<reserved_71>": 160,
+      "<reserved_72>": 161,
+      "<reserved_73>": 162,
+      "<reserved_74>": 163,
+      "<reserved_75>": 164,
+      "<reserved_76>": 165,
+      "<reserved_77>": 166,
+      "<reserved_78>": 167,
+      "<reserved_79>": 168,
+      "<reserved_80>": 169,
+      "<reserved_81>": 170,
+      "<reserved_82>": 171,
+      "<reserved_83>": 172,
+      "<reserved_84>": 173,
+      "<reserved_85>": 174,
+      "<reserved_86>": 175,
+      "<reserved_87>": 176,
+      "<reserved_88>": 177,
+      "<reserved_89>": 178,
+      "<reserved_90>": 179,
+      "<reserved_91>": 180,
+      "<reserved_92>": 181,
+      "<reserved_93>": 182,
+      "<reserved_94>": 183,
+      "<reserved_95>": 184,
+      "<reserved_96>": 185,
+      "<reserved_97>": 186,
+      "<reserved_98>": 187,
+      "<reserved_99>": 188,
+      "<reserved_100>": 189,
+      "<reserved_101>": 190,
+      "<reserved_102>": 191,
+      "<reserved_103>": 192,
+      "<reserved_104>": 193,
+      "<reserved_105>": 194,
+      "<reserved_106>": 195,
+      "<reserved_107>": 196,
+      "<reserved_108>": 197,
+      "<reserved_109>": 198,
+      "<reserved_110>": 199,
+      "<reserved_111>": 200,
+      "<reserved_112>": 201,
+      "<reserved_113>": 202,
+      "<reserved_114>": 203,
+      "<reserved_115>": 204,
+      "<reserved_116>": 205,
+      "<reserved_117>": 206,
+      "<reserved_118>": 207,
+      "<reserved_119>": 208,
+      "<reserved_120>": 209,
+      "<reserved_121>": 210,
+      "<reserved_122>": 211,
+      "<reserved_123>": 212,
+      "<reserved_124>": 213,
+      "<reserved_125>": 214,
+      "<reserved_126>": 215,
+      "<reserved_127>": 216,
+      "<reserved_128>": 217,
+      "<reserved_129>": 218,
+      "<reserved_130>": 219,
+      "<reserved_131>": 220,
+      "<reserved_132>": 221,
+      "<reserved_133>": 222,
+      "<reserved_134>": 223,
+      "<reserved_135>": 224,
+      "<reserved_136>": 225,
+      "<reserved_137>": 226,
+      "<reserved_138>": 227,
+      "<reserved_139>": 228,
+      "<reserved_140>": 229,
+      "<reserved_141>": 230,
+      "<reserved_142>": 231,
+      "<reserved_143>": 232,
+      "<reserved_144>": 233,
+      "<reserved_145>": 234,
+      "<reserved_146>": 235,
+      "<reserved_147>": 236,
+      "<reserved_148>": 237,
+      "<reserved_149>": 238,
+      "<reserved_150>": 239,
+      "<reserved_151>": 240,
+      "<reserved_152>": 241,
+      "<reserved_153>": 242,
+      "<reserved_154>": 243,
+      "<reserved_155>": 244,
+      "<reserved_156>": 245,
+      "<reserved_157>": 246,
+      "<reserved_158>": 247,
+      "<reserved_159>": 248,
+      "<reserved_160>": 249,
+      "<reserved_161>": 250,
+      "<reserved_162>": 251,
+      "<reserved_163>": 252,
+      "<reserved_164>": 253,
+      "<reserved_165>": 254,
+      "<reserved_166>": 255,
+      "<reserved_167>": 256,
+      "<reserved_168>": 257,
+      "<reserved_169>": 258,
+      "<reserved_170>": 259,
+      "<reserved_171>": 260,
+      "<reserved_172>": 261,
+      "<reserved_173>": 262,
+      "<reserved_174>": 263,
+      "<reserved_175>": 264,
+      "<reserved_176>": 265,
+      "<reserved_177>": 266,
+      "<reserved_178>": 267,
+      "<reserved_179>": 268,
+      "<reserved_180>": 269,
+      "<reserved_181>": 270,
+      "<reserved_182>": 271,
+      "<reserved_183>": 272,
+      "<reserved_184>": 273,
+      "<reserved_185>": 274,
+      "<reserved_186>": 275,
+      "<reserved_187>": 276,
+      "<reserved_188>": 277,
+      "<reserved_189>": 278,
+      "<reserved_190>": 279,
+      "<reserved_191>": 280,
+      "<reserved_192>": 281,
+      "<reserved_193>": 282,
+      "<reserved_194>": 283,
+      "<reserved_195>": 284,
+      "<reserved_196>": 285,
+      "<reserved_197>": 286,
+      "<reserved_198>": 287,
+      "<reserved_199>": 288,
+      "<reserved_200>": 289,
+      "<reserved_201>": 290,
+      "<reserved_202>": 291,
+      "<reserved_203>": 292,
+      "<reserved_204>": 293,
+      "<reserved_205>": 294,
+      "<reserved_206>": 295,
+      "<reserved_207>": 296,
+      "<reserved_208>": 297,
+      "<reserved_209>": 298,
+      "<reserved_210>": 299,
+      "<reserved_211>": 300,
+      "<reserved_212>": 301,
+      "<reserved_213>": 302,
+      "<reserved_214>": 303,
+      "<reserved_215>": 304,
+      "<reserved_216>": 305,
+      "<reserved_217>": 306,
+      "<reserved_218>": 307,
+      "<reserved_219>": 308,
+      "<reserved_220>": 309,
+      "<reserved_221>": 310,
+      "<reserved_222>": 311,
+      "<reserved_223>": 312,
+      "<reserved_224>": 313,
+      "<reserved_225>": 314,
+      "<reserved_226>": 315,
+      "<reserved_227>": 316,
+      "<reserved_228>": 317,
+      "<reserved_229>": 318,
+      "<reserved_230>": 319,
+      "<reserved_231>": 320,
+      "<reserved_232>": 321,
+      "<reserved_233>": 322,
+      "<reserved_234>": 323,
+      "<reserved_235>": 324,
+      "<reserved_236>": 325,
+      "<reserved_237>": 326,
+      "<reserved_238>": 327,
+      "<reserved_239>": 328,
+      "<reserved_240>": 329,
+      "<reserved_241>": 330,
+      "<reserved_242>": 331,
+      "<reserved_243>": 332,
+      "<reserved_244>": 333,
+      "<reserved_245>": 334,
+      "<reserved_246>": 335,
+      "<reserved_247>": 336,
+      "<reserved_248>": 337,
+      "<reserved_249>": 338,
+      "<reserved_250>": 339,
+      "<reserved_251>": 340,
+      "<reserved_252>": 341,
+      "<reserved_253>": 342,
+      "<reserved_254>": 343,
+      "<reserved_255>": 344,
+      "<reserved_256>": 345,
+      "<reserved_257>": 346,
+      "<reserved_258>": 347,
+      "<reserved_259>": 348,
+      "<reserved_260>": 349,
+      "<reserved_261>": 350,
+      "<reserved_262>": 351,
+      "<reserved_263>": 352,
+      "<reserved_264>": 353,
+      "<reserved_265>": 354,
+      "<reserved_266>": 355,
+      "<reserved_267>": 356,
+      "<reserved_268>": 357,
+      "<reserved_269>": 358,
+      "<reserved_270>": 359,
+      "<reserved_271>": 360,
+      "<reserved_272>": 361,
+      "<reserved_273>": 362,
+      "<reserved_274>": 363,
+      "<reserved_275>": 364,
+      "<reserved_276>": 365,
+      "<reserved_277>": 366,
+      "<reserved_278>": 367,
+      "<reserved_279>": 368,
+      "<reserved_280>": 369,
+      "<reserved_281>": 370,
+      "<reserved_282>": 371,
+      "<reserved_283>": 372,
+      "<reserved_284>": 373,
+      "<reserved_285>": 374,
+      "<reserved_286>": 375,
+      "<reserved_287>": 376,
+      "<reserved_288>": 377,
+      "<reserved_289>": 378,
+      "<reserved_290>": 379,
+      "<reserved_291>": 380,
+      "<reserved_292>": 381,
+      "<reserved_293>": 382,
+      "<reserved_294>": 383,
+      "<reserved_295>": 384,
+      "<reserved_296>": 385,
+      "<reserved_297>": 386,
+      "<reserved_298>": 387,
+      "<reserved_299>": 388,
+      "<reserved_300>": 389,
+      "<reserved_301>": 390,
+      "<reserved_302>": 391,
+      "<reserved_303>": 392,
+      "<reserved_304>": 393,
+      "<reserved_305>": 394,
+      "<reserved_306>": 395,
+      "<reserved_307>": 396,
+      "<reserved_308>": 397,
+      "<reserved_309>": 398,
+      "<reserved_310>": 399,
+      "<reserved_311>": 400,
+      "<reserved_312>": 401,
+      "<reserved_313>": 402,
+      "<reserved_314>": 403,
+      "<reserved_315>": 404,
+      "<reserved_316>": 405,
+      "<reserved_317>": 406,
+      "<reserved_318>": 407,
+      "<reserved_319>": 408,
+      "<reserved_320>": 409,
+      "<reserved_321>": 410,
+      "<reserved_322>": 411,
+      "<reserved_323>": 412,
+      "<reserved_324>": 413,
+      "<reserved_325>": 414,
+      "<reserved_326>": 415,
+      "<reserved_327>": 416,
+      "<reserved_328>": 417,
+      "<reserved_329>": 418,
+      "<reserved_330>": 419,
+      "<reserved_331>": 420,
+      "<reserved_332>": 421,
+      "<reserved_333>": 422,
+      "<reserved_334>": 423,
+      "<reserved_335>": 424,
+      "<reserved_336>": 425,
+      "<reserved_337>": 426,
+      "<reserved_338>": 427,
+      "<reserved_339>": 428,
+      "<reserved_340>": 429,
+      "<reserved_341>": 430,
+      "<reserved_342>": 431,
+      "<reserved_343>": 432,
+      "<reserved_344>": 433,
+      "<reserved_345>": 434,
+      "<reserved_346>": 435,
+      "<reserved_347>": 436,
+      "<reserved_348>": 437,
+      "<reserved_349>": 438,
+      "<reserved_350>": 439,
+      "<reserved_351>": 440,
+      "<reserved_352>": 441,
+      "<reserved_353>": 442,
+      "<reserved_354>": 443,
+      "<reserved_355>": 444,
+      "<reserved_356>": 445,
+      "<reserved_357>": 446,
+      "<reserved_358>": 447,
+      "<reserved_359>": 448,
+      "<reserved_360>": 449,
+      "<reserved_361>": 450,
+      "<reserved_362>": 451,
+      "<reserved_363>": 452,
+      "<reserved_364>": 453,
+      "<reserved_365>": 454,
+      "<reserved_366>": 455,
+      "<reserved_367>": 456,
+      "<reserved_368>": 457,
+      "<reserved_369>": 458,
+      "<reserved_370>": 459,
+      "<reserved_371>": 460,
+      "<reserved_372>": 461,
+      "<reserved_373>": 462,
+      "<reserved_374>": 463,
+      "<reserved_375>": 464,
+      "<reserved_376>": 465,
+      "<reserved_377>": 466,
+      "<reserved_378>": 467,
+      "<reserved_379>": 468,
+      "<reserved_380>": 469,
+      "<reserved_381>": 470,
+      "<reserved_382>": 471,
+      "<reserved_383>": 472,
+      "<reserved_384>": 473,
+      "<reserved_385>": 474,
+      "<reserved_386>": 475,
+      "<reserved_387>": 476,
+      "<reserved_388>": 477,
+      "<reserved_389>": 478,
+      "<reserved_390>": 479,
+      "<reserved_391>": 480,
+      "<reserved_392>": 481,
+      "<reserved_393>": 482,
+      "<reserved_394>": 483,
+      "<reserved_395>": 484,
+      "<reserved_396>": 485,
+      "<reserved_397>": 486,
+      "<reserved_398>": 487,
+      "<reserved_399>": 488,
+      "<reserved_400>": 489,
+      "<reserved_401>": 490,
+      "<reserved_402>": 491,
+      "<reserved_403>": 492,
+      "<reserved_404>": 493,
+      "<reserved_405>": 494,
+      "<reserved_406>": 495,
+      "<reserved_407>": 496,
+      "<reserved_408>": 497,
+      "<reserved_409>": 498,
+      "<reserved_410>": 499,
+      "<reserved_411>": 500,
+      "<reserved_412>": 501,
+      "<reserved_413>": 502,
+      "<reserved_414>": 503,
+      "<reserved_415>": 504,
+      "<reserved_416>": 505,
+      "<reserved_417>": 506,
+      "<reserved_418>": 507,
+      "<reserved_419>": 508,
+      "<reserved_420>": 509,
+      "<reserved_421>": 510,
+      "<reserved_422>": 511,
+      "<reserved_423>": 512,
+      "<reserved_424>": 513,
+      "<reserved_425>": 514,
+      "<reserved_426>": 515,
+      "<reserved_427>": 516,
+      "<reserved_428>": 517,
+      "<reserved_429>": 518,
+      "<reserved_430>": 519,
+      "<reserved_431>": 520,
+      "<reserved_432>": 521,
+      "<reserved_433>": 522,
+      "<reserved_434>": 523,
+      "<reserved_435>": 524,
+      "<reserved_436>": 525,
+      "<reserved_437>": 526,
+      "<reserved_438>": 527,
+      "<reserved_439>": 528,
+      "<reserved_440>": 529,
+      "<reserved_441>": 530,
+      "<reserved_442>": 531,
+      "<reserved_443>": 532,
+      "<reserved_444>": 533,
+      "<reserved_445>": 534,
+      "<reserved_446>": 535,
+      "<reserved_447>": 536,
+      "<reserved_448>": 537,
+      "<reserved_449>": 538,
+      "<reserved_450>": 539,
+      "<reserved_451>": 540,
+      "<reserved_452>": 541,
+      "<reserved_453>": 542,
+      "<reserved_454>": 543,
+      "<reserved_455>": 544,
+      "<reserved_456>": 545,
+      "<reserved_457>": 546,
+      "<reserved_458>": 547,
+      "<reserved_459>": 548,
+      "<reserved_460>": 549,
+      "<reserved_461>": 550,
+      "<reserved_462>": 551,
+      "<reserved_463>": 552,
+      "<reserved_464>": 553,
+      "<reserved_465>": 554,
+      "<reserved_466>": 555,
+      "<reserved_467>": 556,
+      "<reserved_468>": 557,
+      "<reserved_469>": 558,
+      "<reserved_470>": 559,
+      "<reserved_471>": 560,
+      "<reserved_472>": 561,
+      "<reserved_473>": 562,
+      "<reserved_474>": 563,
+      "<reserved_475>": 564,
+      "<reserved_476>": 565,
+      "<reserved_477>": 566,
+      "<reserved_478>": 567,
+      "<reserved_479>": 568,
+      "<reserved_480>": 569,
+      "<reserved_481>": 570,
+      "<reserved_482>": 571,
+      "<reserved_483>": 572,
+      "<reserved_484>": 573,
+      "<reserved_485>": 574,
+      "<reserved_486>": 575,
+      "<reserved_487>": 576,
+      "<reserved_488>": 577,
+      "<reserved_489>": 578,
+      "<reserved_490>": 579,
+      "<reserved_491>": 580,
+      "<reserved_492>": 581,
+      "<reserved_493>": 582,
+      "<reserved_494>": 583,
+      "<reserved_495>": 584,
+      "<reserved_496>": 585,
+      "<reserved_497>": 586,
+      "<reserved_498>": 587,
+      "<reserved_499>": 588,
+      "<reserved_500>": 589,
+      "<reserved_501>": 590,
+      "<reserved_502>": 591,
+      "<reserved_503>": 592,
+      "<reserved_504>": 593,
+      "<reserved_505>": 594,
+      "<reserved_506>": 595,
+      "<reserved_507>": 596,
+      "<reserved_508>": 597,
+      "<reserved_509>": 598,
+      "<reserved_510>": 599,
+      "<reserved_511>": 600,
+      "<reserved_512>": 601,
+      "<reserved_513>": 602,
+      "<reserved_514>": 603,
+      "<reserved_515>": 604,
+      "<reserved_516>": 605,
+      "<reserved_517>": 606,
+      "<reserved_518>": 607,
+      "<reserved_519>": 608,
+      "<reserved_520>": 609,
+      "<reserved_521>": 610,
+      "<reserved_522>": 611,
+      "<reserved_523>": 612,
+      "<reserved_524>": 613,
+      "<reserved_525>": 614,
+      "<reserved_526>": 615,
+      "<reserved_527>": 616,
+      "<reserved_528>": 617,
+      "<reserved_529>": 618,
+      "<reserved_530>": 619,
+      "<reserved_531>": 620,
+      "<reserved_532>": 621,
+      "<reserved_533>": 622,
+      "<reserved_534>": 623,
+      "<reserved_535>": 624,
+      "<reserved_536>": 625,
+      "<reserved_537>": 626,
+      "<reserved_538>": 627,
+      "<reserved_539>": 628,
+      "<reserved_540>": 629,
+      "<reserved_541>": 630,
+      "<reserved_542>": 631,
+      "<reserved_543>": 632,
+      "<reserved_544>": 633,
+      "<reserved_545>": 634,
+      "<reserved_546>": 635,
+      "<reserved_547>": 636,
+      "<reserved_548>": 637,
+      "<reserved_549>": 638,
+      "<reserved_550>": 639,
+      "<reserved_551>": 640,
+      "<reserved_552>": 641,
+      "<reserved_553>": 642,
+      "<reserved_554>": 643,
+      "<reserved_555>": 644,
+      "<reserved_556>": 645,
+      "<reserved_557>": 646,
+      "<reserved_558>": 647,
+      "<reserved_559>": 648,
+      "<reserved_560>": 649,
+      "<reserved_561>": 650,
+      "<reserved_562>": 651,
+      "<reserved_563>": 652,
+      "<reserved_564>": 653,
+      "<reserved_565>": 654,
+      "<reserved_566>": 655,
+      "<reserved_567>": 656,
+      "<reserved_568>": 657,
+      "<reserved_569>": 658,
+      "<reserved_570>": 659,
+      "<reserved_571>": 660,
+      "<reserved_572>": 661,
+      "<reserved_573>": 662,
+      "<reserved_574>": 663,
+      "<reserved_575>": 664,
+      "<reserved_576>": 665,
+      "<reserved_577>": 666,
+      "<reserved_578>": 667,
+      "<reserved_579>": 668,
+      "<reserved_580>": 669,
+      "<reserved_581>": 670,
+      "<reserved_582>": 671,
+      "<reserved_583>": 672,
+      "<reserved_584>": 673,
+      "<reserved_585>": 674,
+      "<reserved_586>": 675,
+      "<reserved_587>": 676,
+      "<reserved_588>": 677,
+      "<reserved_589>": 678,
+      "<reserved_590>": 679,
+      "<reserved_591>": 680,
+      "<reserved_592>": 681,
+      "<reserved_593>": 682,
+      "<reserved_594>": 683,
+      "<reserved_595>": 684,
+      "<reserved_596>": 685,
+      "<reserved_597>": 686,
+      "<reserved_598>": 687,
+      "<reserved_599>": 688,
+      "<reserved_600>": 689,
+      "<reserved_601>": 690,
+      "<reserved_602>": 691,
+      "<reserved_603>": 692,
+      "<reserved_604>": 693,
+      "<reserved_605>": 694,
+      "<reserved_606>": 695,
+      "<reserved_607>": 696,
+      "<reserved_608>": 697,
+      "<reserved_609>": 698,
+      "<reserved_610>": 699,
+      "<reserved_611>": 700,
+      "<reserved_612>": 701,
+      "<reserved_613>": 702,
+      "<reserved_614>": 703,
+      "<reserved_615>": 704,
+      "<reserved_616>": 705,
+      "<reserved_617>": 706,
+      "<reserved_618>": 707,
+      "<reserved_619>": 708,
+      "<reserved_620>": 709,
+      "<reserved_621>": 710,
+      "<reserved_622>": 711,
+      "<reserved_623>": 712,
+      "<reserved_624>": 713,
+      "<reserved_625>": 714,
+      "<reserved_626>": 715,
+      "<reserved_627>": 716,
+      "<reserved_628>": 717,
+      "<reserved_629>": 718,
+      "<reserved_630>": 719,
+      "<reserved_631>": 720,
+      "<reserved_632>": 721,
+      "<reserved_633>": 722,
+      "<reserved_634>": 723,
+      "<reserved_635>": 724,
+      "<reserved_636>": 725,
+      "<reserved_637>": 726,
+      "<reserved_638>": 727,
+      "<reserved_639>": 728,
+      "<reserved_640>": 729,
+      "<reserved_641>": 730,
+      "<reserved_642>": 731,
+      "<reserved_643>": 732,
+      "<reserved_644>": 733,
+      "<reserved_645>": 734,
+      "<reserved_646>": 735,
+      "<reserved_647>": 736,
+      "<reserved_648>": 737,
+      "<reserved_649>": 738,
+      "<reserved_650>": 739,
+      "<reserved_651>": 740,
+      "<reserved_652>": 741,
+      "<reserved_653>": 742,
+      "<reserved_654>": 743,
+      "<reserved_655>": 744,
+      "<reserved_656>": 745,
+      "<reserved_657>": 746,
+      "<reserved_658>": 747,
+      "<reserved_659>": 748,
+      "<reserved_660>": 749,
+      "<reserved_661>": 750,
+      "<reserved_662>": 751,
+      "<reserved_663>": 752,
+      "<reserved_664>": 753,
+      "<reserved_665>": 754,
+      "<reserved_666>": 755,
+      "<reserved_667>": 756,
+      "<reserved_668>": 757,
+      "<reserved_669>": 758,
+      "<reserved_670>": 759,
+      "<reserved_671>": 760,
+      "<reserved_672>": 761,
+      "<reserved_673>": 762,
+      "<reserved_674>": 763,
+      "<reserved_675>": 764,
+      "<reserved_676>": 765,
+      "<reserved_677>": 766,
+      "<reserved_678>": 767,
+      "<reserved_679>": 768,
+      "<reserved_680>": 769,
+      "<reserved_681>": 770,
+      "<reserved_682>": 771,
+      "<reserved_683>": 772,
+      "<reserved_684>": 773,
+      "<reserved_685>": 774,
+      "<reserved_686>": 775,
+      "<reserved_687>": 776,
+      "<reserved_688>": 777,
+      "<reserved_689>": 778,
+      "<reserved_690>": 779,
+      "<reserved_691>": 780,
+      "<reserved_692>": 781,
+      "<reserved_693>": 782,
+      "<reserved_694>": 783,
+      "<reserved_695>": 784,
+      "<reserved_696>": 785,
+      "<reserved_697>": 786,
+      "<reserved_698>": 787,
+      "<reserved_699>": 788,
+      "<reserved_700>": 789,
+      "<reserved_701>": 790,
+      "<reserved_702>": 791,
+      "<reserved_703>": 792,
+      "<reserved_704>": 793,
+      "<reserved_705>": 794,
+      "<reserved_706>": 795,
+      "<reserved_707>": 796,
+      "<reserved_708>": 797,
+      "<reserved_709>": 798,
+      "<reserved_710>": 799,
+      "<reserved_711>": 800,
+      "<reserved_712>": 801,
+      "<reserved_713>": 802,
+      "<reserved_714>": 803,
+      "<reserved_715>": 804,
+      "<reserved_716>": 805,
+      "<reserved_717>": 806,
+      "<reserved_718>": 807,
+      "<reserved_719>": 808,
+      "<reserved_720>": 809,
+      "<reserved_721>": 810,
+      "<reserved_722>": 811,
+      "<reserved_723>": 812,
+      "<reserved_724>": 813,
+      "<reserved_725>": 814,
+      "<reserved_726>": 815,
+      "<reserved_727>": 816,
+      "<reserved_728>": 817,
+      "<reserved_729>": 818,
+      "<reserved_730>": 819,
+      "<reserved_731>": 820,
+      "<reserved_732>": 821,
+      "<reserved_733>": 822,
+      "<reserved_734>": 823,
+      "<reserved_735>": 824,
+      "<reserved_736>": 825,
+      "<reserved_737>": 826,
+      "<reserved_738>": 827,
+      "<reserved_739>": 828,
+      "<reserved_740>": 829,
+      "<reserved_741>": 830,
+      "<reserved_742>": 831,
+      "<reserved_743>": 832,
+      "<reserved_744>": 833,
+      "<reserved_745>": 834,
+      "<reserved_746>": 835,
+      "<reserved_747>": 836,
+      "<reserved_748>": 837,
+      "<reserved_749>": 838,
+      "<reserved_750>": 839,
+      "<reserved_751>": 840,
+      "<reserved_752>": 841,
+      "<reserved_753>": 842,
+      "<reserved_754>": 843,
+      "<reserved_755>": 844,
+      "<reserved_756>": 845,
+      "<reserved_757>": 846,
+      "<reserved_758>": 847,
+      "<reserved_759>": 848,
+      "<reserved_760>": 849,
+      "<reserved_761>": 850,
+      "<reserved_762>": 851,
+      "<reserved_763>": 852,
+      "<reserved_764>": 853,
+      "<reserved_765>": 854,
+      "<reserved_766>": 855,
+      "<reserved_767>": 856,
+      "<reserved_768>": 857,
+      "<reserved_769>": 858,
+      "<reserved_770>": 859,
+      "<reserved_771>": 860,
+      "<reserved_772>": 861,
+      "<reserved_773>": 862,
+      "<reserved_774>": 863,
+      "<reserved_775>": 864,
+      "<reserved_776>": 865,
+      "<reserved_777>": 866,
+      "<reserved_778>": 867,
+      "<reserved_779>": 868,
+      "<reserved_780>": 869,
+      "<reserved_781>": 870,
+      "<reserved_782>": 871,
+      "<reserved_783>": 872,
+      "<reserved_784>": 873,
+      "<reserved_785>": 874,
+      "<reserved_786>": 875,
+      "<reserved_787>": 876,
+      "<reserved_788>": 877,
+      "<reserved_789>": 878,
+      "<reserved_790>": 879,
+      "<reserved_791>": 880,
+      "<reserved_792>": 881,
+      "<reserved_793>": 882,
+      "<reserved_794>": 883,
+      "<reserved_795>": 884,
+      "<reserved_796>": 885,
+      "<reserved_797>": 886,
+      "<reserved_798>": 887,
+      "<reserved_799>": 888,
+      "<reserved_800>": 889,
+      "<reserved_801>": 890,
+      "<reserved_802>": 891,
+      "<reserved_803>": 892,
+      "<reserved_804>": 893,
+      "<reserved_805>": 894,
+      "<reserved_806>": 895,
+      "<reserved_807>": 896,
+      "<reserved_808>": 897,
+      "<reserved_809>": 898,
+      "<reserved_810>": 899,
+      "<reserved_811>": 900,
+      "<reserved_812>": 901,
+      "<reserved_813>": 902,
+      "<reserved_814>": 903,
+      "<reserved_815>": 904,
+      "<reserved_816>": 905,
+      "<reserved_817>": 906,
+      "<reserved_818>": 907,
+      "<reserved_819>": 908,
+      "<reserved_820>": 909,
+      "<reserved_821>": 910,
+      "<reserved_822>": 911,
+      "<reserved_823>": 912,
+      "<reserved_824>": 913,
+      "<reserved_825>": 914,
+      "<reserved_826>": 915,
+      "<reserved_827>": 916,
+      "<reserved_828>": 917,
+      "<reserved_829>": 918,
+      "<reserved_830>": 919,
+      "<reserved_831>": 920,
+      "<reserved_832>": 921,
+      "<reserved_833>": 922,
+      "<reserved_834>": 923,
+      "<reserved_835>": 924,
+      "<reserved_836>": 925,
+      "<reserved_837>": 926,
+      "<reserved_838>": 927,
+      "<reserved_839>": 928,
+      "<reserved_840>": 929,
+      "<reserved_841>": 930,
+      "<reserved_842>": 931,
+      "<reserved_843>": 932,
+      "<reserved_844>": 933,
+      "<reserved_845>": 934,
+      "<reserved_846>": 935,
+      "<reserved_847>": 936,
+      "<reserved_848>": 937,
+      "<reserved_849>": 938,
+      "<reserved_850>": 939,
+      "<reserved_851>": 940,
+      "<reserved_852>": 941,
+      "<reserved_853>": 942,
+      "<reserved_854>": 943,
+      "<reserved_855>": 944,
+      "<reserved_856>": 945,
+      "<reserved_857>": 946,
+      "<reserved_858>": 947,
+      "<reserved_859>": 948,
+      "<reserved_860>": 949,
+      "<reserved_861>": 950,
+      "<reserved_862>": 951,
+      "<reserved_863>": 952,
+      "<reserved_864>": 953,
+      "<reserved_865>": 954,
+      "<reserved_866>": 955,
+      "<reserved_867>": 956,
+      "<reserved_868>": 957,
+      "<reserved_869>": 958,
+      "<reserved_870>": 959,
+      "<reserved_871>": 960,
+      "<reserved_872>": 961,
+      "<reserved_873>": 962,
+      "<reserved_874>": 963,
+      "<reserved_875>": 964,
+      "<reserved_876>": 965,
+      "<reserved_877>": 966,
+      "<reserved_878>": 967,
+      "<reserved_879>": 968,
+      "<reserved_880>": 969,
+      "<reserved_881>": 970,
+      "<reserved_882>": 971,
+      "<reserved_883>": 972,
+      "<reserved_884>": 973,
+      "<reserved_885>": 974,
+      "<reserved_886>": 975,
+      "<reserved_887>": 976,
+      "<reserved_888>": 977,
+      "<reserved_889>": 978,
+      "<reserved_890>": 979,
+      "<reserved_891>": 980,
+      "<reserved_892>": 981,
+      "<reserved_893>": 982,
+      "<reserved_894>": 983,
+      "<reserved_895>": 984,
+      "<reserved_896>": 985,
+      "<reserved_897>": 986,
+      "<reserved_898>": 987,
+      "<reserved_899>": 988,
+      "<reserved_900>": 989,
+      "<reserved_901>": 990,
+      "<reserved_902>": 991,
+      "<reserved_903>": 992,
+      "<reserved_904>": 993,
+      "<reserved_905>": 994,
+      "<reserved_906>": 995,
+      "<reserved_907>": 996,
+      "<reserved_908>": 997,
+      "<reserved_909>": 998,
+      "<reserved_910>": 999,
+      "<reserved_911>": 1000,
+      "<reserved_912>": 1001,
+      "<reserved_913>": 1002,
+      "<reserved_914>": 1003,
+      "<reserved_915>": 1004,
+      "<reserved_916>": 1005,
+      "<reserved_917>": 1006,
+      "<reserved_918>": 1007,
+      "<reserved_919>": 1008,
+      "<reserved_920>": 1009,
+      "<reserved_921>": 1010,
+      "<reserved_922>": 1011,
+      "<reserved_923>": 1012,
+      "<reserved_924>": 1013,
+      "<reserved_925>": 1014,
+      "<reserved_926>": 1015,
+      "<reserved_927>": 1016,
+      "<reserved_928>": 1017,
+      "<reserved_929>": 1018,
+      "<reserved_930>": 1019,
+      "<reserved_931>": 1020,
+      "<reserved_932>": 1021,
+      "<reserved_933>": 1022,
+      "<reserved_934>": 1023,
+      "<reserved_935>": 1024,
+      "<reserved_936>": 1025,
+      "<reserved_937>": 1026,
+      "<reserved_938>": 1027,
+      "<reserved_939>": 1028,
+      "<reserved_940>": 1029,
+      "<reserved_941>": 1030,
+      "<reserved_942>": 1031,
+      "<reserved_943>": 1032,
+      "<reserved_944>": 1033,
+      "<reserved_945>": 1034,
+      "<reserved_946>": 1035,
+      "<reserved_947>": 1036,
+      "<reserved_948>": 1037,
+      "<reserved_949>": 1038,
+      "<reserved_950>": 1039,
+      "<reserved_951>": 1040,
+      "<reserved_952>": 1041,
+      "<reserved_953>": 1042,
+      "<reserved_954>": 1043,
+      "<reserved_955>": 1044,
+      "<reserved_956>": 1045,
+      "<reserved_957>": 1046,
+      "<reserved_958>": 1047,
+      "<reserved_959>": 1048,
+      "<reserved_960>": 1049,
+      "<reserved_961>": 1050,
+      "<reserved_962>": 1051,
+      "<reserved_963>": 1052,
+      "<reserved_964>": 1053,
+      "<reserved_965>": 1054,
+      "<reserved_966>": 1055,
+      "<reserved_967>": 1056,
+      "<reserved_968>": 1057,
+      "<reserved_969>": 1058,
+      "<reserved_970>": 1059,
+      "<reserved_971>": 1060,
+      "<reserved_972>": 1061,
+      "<reserved_973>": 1062,
+      "<reserved_974>": 1063,
+      "<reserved_975>": 1064,
+      "<reserved_976>": 1065,
+      "<reserved_977>": 1066,
+      "<reserved_978>": 1067,
+      "<reserved_979>": 1068,
+      "<reserved_980>": 1069,
+      "<reserved_981>": 1070,
+      "<reserved_982>": 1071,
+      "<reserved_983>": 1072,
+      "<reserved_984>": 1073,
+      "<reserved_985>": 1074,
+      "<reserved_986>": 1075,
+      "<reserved_987>": 1076,
+      "<reserved_988>": 1077,
+      "<reserved_989>": 1078,
+      "<reserved_990>": 1079,
+      "<reserved_991>": 1080,
+      "<reserved_992>": 1081,
+      "<reserved_993>": 1082,
+      "<reserved_994>": 1083,
+      "<reserved_995>": 1084,
+      "<reserved_996>": 1085,
+      "<reserved_997>": 1086,
+      "<reserved_998>": 1087,
+      "<reserved_999>": 1088,
+      "<0x00>": 1089,
+      "<0x01>": 1090,
+      "<0x02>": 1091,
+      "<0x03>": 1092,
+      "<0x04>": 1093,
+      "<0x05>": 1094,
+      "<0x06>": 1095,
+      "<0x07>": 1096,
+      "<0x08>": 1097,
+      "<0x09>": 1098,
+      "<0x0A>": 1099,
+      "<0x0B>": 1100,
+      "<0x0C>": 1101,
+      "<0x0D>": 1102,
+      "<0x0E>": 1103,
+      "<0x0F>": 1104,
+      "<0x10>": 1105,
+      "<0x11>": 1106,
+      "<0x12>": 1107,
+      "<0x13>": 1108,
+      "<0x14>": 1109,
+      "<0x15>": 1110,
+      "<0x16>": 1111,
+      "<0x17>": 1112,
+      "<0x18>": 1113,
+      "<0x19>": 1114,
+      "<0x1A>": 1115,
+      "<0x1B>": 1116,
+      "<0x1C>": 1117,
+      "<0x1D>": 1118,
+      "<0x1E>": 1119,
+      "<0x1F>": 1120,
+      "<0x20>": 1121,
+      "<0x21>": 1122,
+      "<0x22>": 1123,
+      "<0x23>": 1124,
+      "<0x24>": 1125,
+      "<0x25>": 1126,
+      "<0x26>": 1127,
+      "<0x27>": 1128,
+      "<0x28>": 1129,
+      "<0x29>": 1130,
+      "<0x2A>": 1131,
+      "<0x2B>": 1132,
+      "<0x2C>": 1133,
+      "<0x2D>": 1134,
+      "<0x2E>": 1135,
+      "<0x2F>": 1136,
+      "<0x30>": 1137,
+      "<0x31>": 1138,
+      "<0x32>": 1139,
+      "<0x33>": 1140,
+      "<0x34>": 1141,
+      "<0x35>": 1142,
+      "<0x36>": 1143,
+      "<0x37>": 1144,
+      "<0x38>": 1145,
+      "<0x39>": 1146,
+      "<0x3A>": 1147,
+      "<0x3B>": 1148,
+      "<0x3C>": 1149,
+      "<0x3D>": 1150,
+      "<0x3E>": 1151,
+      "<0x3F>": 1152,
+      "<0x40>": 1153,
+      "<0x41>": 1154,
+      "<0x42>": 1155,
+      "<0x43>": 1156,
+      "<0x44>": 1157,
+      "<0x45>": 1158,
+      "<0x46>": 1159,
+      "<0x47>": 1160,
+      "<0x48>": 1161,
+      "<0x49>": 1162,
+      "<0x4A>": 1163,
+      "<0x4B>": 1164,
+      "<0x4C>": 1165,
+      "<0x4D>": 1166,
+      "<0x4E>": 1167,
+      "<0x4F>": 1168,
+      "<0x50>": 1169,
+      "<0x51>": 1170,
+      "<0x52>": 1171,
+      "<0x53>": 1172,
+      "<0x54>": 1173,
+      "<0x55>": 1174,
+      "<0x56>": 1175,
+      "<0x57>": 1176,
+      "<0x58>": 1177,
+      "<0x59>": 1178,
+      "<0x5A>": 1179,
+      "<0x5B>": 1180,
+      "<0x5C>": 1181,
+      "<0x5D>": 1182,
+      "<0x5E>": 1183,
+      "<0x5F>": 1184,
+      "<0x60>": 1185,
+      "<0x61>": 1186,
+      "<0x62>": 1187,
+      "<0x63>": 1188,
+      "<0x64>": 1189,
+      "<0x65>": 1190,
+      "<0x66>": 1191,
+      "<0x67>": 1192,
+      "<0x68>": 1193,
+      "<0x69>": 1194,
+      "<0x6A>": 1195,
+      "<0x6B>": 1196,
+      "<0x6C>": 1197,
+      "<0x6D>": 1198,
+      "<0x6E>": 1199,
+      "<0x6F>": 1200,
+      "<0x70>": 1201,
+      "<0x71>": 1202,
+      "<0x72>": 1203,
+      "<0x73>": 1204,
+      "<0x74>": 1205,
+      "<0x75>": 1206,
+      "<0x76>": 1207,
+      "<0x77>": 1208,
+      "<0x78>": 1209,
+      "<0x79>": 1210,
+      "<0x7A>": 1211,
+      "<0x7B>": 1212,
+      "<0x7C>": 1213,
+      "<0x7D>": 1214,
+      "<0x7E>": 1215,
+      "<0x7F>": 1216,
+      "<0x80>": 1217,
+      "<0x81>": 1218,
+      "<0x82>": 1219,
+      "<0x83>": 1220,
+      "<0x84>": 1221,
+      "<0x85>": 1222,
+      "<0x86>": 1223,
+      "<0x87>": 1224,
+      "<0x88>": 1225,
+      "<0x89>": 1226,
+      "<0x8A>": 1227,
+      "<0x8B>": 1228,
+      "<0x8C>": 1229,
+      "<0x8D>": 1230,
+      "<0x8E>": 1231,
+      "<0x8F>": 1232,
+      "<0x90>": 1233,
+      "<0x91>": 1234,
+      "<0x92>": 1235,
+      "<0x93>": 1236,
+      "<0x94>": 1237,
+      "<0x95>": 1238,
+      "<0x96>": 1239,
+      "<0x97>": 1240,
+      "<0x98>": 1241,
+      "<0x99>": 1242,
+      "<0x9A>": 1243,
+      "<0x9B>": 1244,
+      "<0x9C>": 1245,
+      "<0x9D>": 1246,
+      "<0x9E>": 1247,
+      "<0x9F>": 1248,
+      "<0xA0>": 1249,
+      "<0xA1>": 1250,
+      "<0xA2>": 1251,
+      "<0xA3>": 1252,
+      "<0xA4>": 1253,
+      "<0xA5>": 1254,
+      "<0xA6>": 1255,
+      "<0xA7>": 1256,
+      "<0xA8>": 1257,
+      "<0xA9>": 1258,
+      "<0xAA>": 1259,
+      "<0xAB>": 1260,
+      "<0xAC>": 1261,
+      "<0xAD>": 1262,
+      "<0xAE>": 1263,
+      "<0xAF>": 1264,
+      "<0xB0>": 1265,
+      "<0xB1>": 1266,
+      "<0xB2>": 1267,
+      "<0xB3>": 1268,
+      "<0xB4>": 1269,
+      "<0xB5>": 1270,
+      "<0xB6>": 1271,
+      "<0xB7>": 1272,
+      "<0xB8>": 1273,
+      "<0xB9>": 1274,
+      "<0xBA>": 1275,
+      "<0xBB>": 1276,
+      "<0xBC>": 1277,
+      "<0xBD>": 1278,
+      "<0xBE>": 1279,
+      "<0xBF>": 1280,
+      "<0xC0>": 1281,
+      "<0xC1>": 1282,
+      "<0xC2>": 1283,
+      "<0xC3>": 1284,
+      "<0xC4>": 1285,
+      "<0xC5>": 1286,
+      "<0xC6>": 1287,
+      "<0xC7>": 1288,
+      "<0xC8>": 1289,
+      "<0xC9>": 1290,
+      "<0xCA>": 1291,
+      "<0xCB>": 1292,
+      "<0xCC>": 1293,
+      "<0xCD>": 1294,
+      "<0xCE>": 1295,
+      "<0xCF>": 1296,
+      "<0xD0>": 1297,
+      "<0xD1>": 1298,
+      "<0xD2>": 1299,
+      "<0xD3>": 1300,
+      "<0xD4>": 1301,
+      "<0xD5>": 1302,
+      "<0xD6>": 1303,
+      "<0xD7>": 1304,
+      "<0xD8>": 1305,
+      "<0xD9>": 1306,
+      "<0xDA>": 1307,
+      "<0xDB>": 1308,
+      "<0xDC>": 1309,
+      "<0xDD>": 1310,
+      "<0xDE>": 1311,
+      "<0xDF>": 1312,
+      "<0xE0>": 1313,
+      "<0xE1>": 1314,
+      "<0xE2>": 1315,
+      "<0xE3>": 1316,
+      "<0xE4>": 1317,
+      "<0xE5>": 1318,
+      "<0xE6>": 1319,
+      "<0xE7>": 1320,
+      "<0xE8>": 1321,
+      "<0xE9>": 1322,
+      "<0xEA>": 1323,
+      "<0xEB>": 1324,
+      "<0xEC>": 1325,
+      "<0xED>": 1326,
+      "<0xEE>": 1327,
+      "<0xEF>": 1328,
+      "<0xF0>": 1329,
+      "<0xF1>": 1330,
+      "<0xF2>": 1331,
+      "<0xF3>": 1332,
+      "<0xF4>": 1333,
+      "<0xF5>": 1334,
+      "<0xF6>": 1335,
+      "<0xF7>": 1336,
+      "<0xF8>": 1337,
+      "<0xF9>": 1338,
+      "<0xFA>": 1339,
+      "<0xFB>": 1340,
+      "<0xFC>": 1341,
+      "<0xFD>": 1342,
+      "<0xFE>": 1343,
+      "<0xFF>": 1344,
+      "▁▁": 1345,
+      "▁t": 1346,
+      "in": 1347,
+      "▁a": 1348,
+      "er": 1349,
+      "on": 1350,
+      "he": 1351,
+      "re": 1352,
+      "at": 1353,
+      "▁▁▁▁": 1354,
+      "or": 1355,
+      "en": 1356,
+      "▁s": 1357,
+      "▁the": 1358,
+      "es": 1359,
+      "is": 1360,
+      "▁c": 1361,
+      "it": 1362,
+      "an": 1363,
+      "al": 1364,
+      "▁o": 1365,
+      "▁w": 1366,
+      "ed": 1367,
+      "▁f": 1368,
+      "▁p": 1369,
+      "ar": 1370,
+      "ing": 1371,
+      "nd": 1372,
+      "ion": 1373,
+      "ic": 1374,
+      "ou": 1375,
+      "▁b": 1376,
+      "▁in": 1377,
+      "le": 1378,
+      "▁of": 1379,
+      "▁m": 1380,
+      "ro": 1381,
+      "as": 1382,
+      "▁d": 1383,
+      "▁and": 1384,
+      "▁to": 1385,
+      "ent": 1386,
+      "ct": 1387,
+      "et": 1388,
+      "st": 1389,
+      "om": 1390,
+      "el": 1391,
+      "▁h": 1392,
+      "▁th": 1393,
+      "il": 1394,
+      "▁▁▁▁▁▁▁▁": 1395,
+      "▁re": 1396,
+      "▁n": 1397,
+      "am": 1398,
+      "▁l": 1399,
+      "im": 1400,
+      "▁e": 1401,
+      "id": 1402,
+      "ut": 1403,
+      "ol": 1404,
+      "ac": 1405,
+      "us": 1406,
+      "ur": 1407,
+      "▁T": 1408,
+      "ation": 1409,
+      "▁is": 1410,
+      "ot": 1411,
+      "ve": 1412,
+      "▁S": 1413,
+      "▁g": 1414,
+      "\\\\": 1415,
+      "ig": 1416,
+      "▁I": 1417,
+      "\\'": 1418,
+      "▁C": 1419,
+      "▁A": 1420,
+      "▁for": 1421,
+      "ad": 1422,
+      "ul": 1423,
+      "ow": 1424,
+      "▁|": 1425,
+      "▁(": 1426,
+      "▁y": 1427,
+      ";\\": 1428,
+      "ith": 1429,
+      "ver": 1430,
+      "ay": 1431,
+      "▁on": 1432,
+      "▁st": 1433,
+      "ly": 1434,
+      "se": 1435,
+      "if": 1436,
+      "ce": 1437,
+      "od": 1438,
+      "ch": 1439,
+      "th": 1440,
+      "▁be": 1441,
+      "ag": 1442,
+      "ter": 1443,
+      "un": 1444,
+      "▁M": 1445,
+      "ate": 1446,
+      "est": 1447,
+      "ir": 1448,
+      "▁you": 1449,
+      "▁with": 1450,
+      "um": 1451,
+      "▁P": 1452,
+      "▁pro": 1453,
+      "em": 1454,
+      ")\\": 1455,
+      "ab": 1456,
+      "▁that": 1457,
+      "▁an": 1458,
+      "ort": 1459,
+      "ri": 1460,
+      "ap": 1461,
+      "qu": 1462,
+      "▁con": 1463,
+      "op": 1464,
+      "ers": 1465,
+      "ist": 1466,
+      "▁as": 1467,
+      "ect": 1468,
+      "▁B": 1469,
+      "res": 1470,
+      "▁v": 1471,
+      "os": 1472,
+      "oc": 1473,
+      "▁al": 1474,
+      "rom": 1475,
+      "▁it": 1476,
+      "▁wh": 1477,
+      "ht": 1478,
+      "▁D": 1479,
+      "▁com": 1480,
+      "ess": 1481,
+      "ment": 1482,
+      "and": 1483,
+      "ub": 1484,
+      "ain": 1485,
+      "ere": 1486,
+      "ill": 1487,
+      "▁r": 1488,
+      "ame": 1489,
+      "▁R": 1490,
+      "▁H": 1491,
+      "▁ex": 1492,
+      "▁F": 1493,
+      "▁L": 1494,
+      "▁W": 1495,
+      "▁\"": 1496,
+      "up": 1497,
+      "iv": 1498,
+      "ity": 1499,
+      "port": 1500,
+      "ec": 1501,
+      "▁are": 1502,
+      "pl": 1503,
+      "▁{": 1504,
+      "ew": 1505,
+      "ex": 1506,
+      "▁The": 1507,
+      "▁or": 1508,
+      "ke": 1509,
+      "▁at": 1510,
+      "ue": 1511,
+      "ore": 1512,
+      "nt": 1513,
+      "og": 1514,
+      "▁N": 1515,
+      "ant": 1516,
+      "end": 1517,
+      "act": 1518,
+      "ive": 1519,
+      "▁by": 1520,
+      "ef": 1521,
+      "art": 1522,
+      "▁was": 1523,
+      "pt": 1524,
+      "▁from": 1525,
+      ",\\": 1526,
+      "▁G": 1527,
+      "ud": 1528,
+      "▁E": 1529,
+      ":\\": 1530,
+      "▁us": 1531,
+      "ies": 1532,
+      "age": 1533,
+      "ine": 1534,
+      "▁▁▁▁▁▁▁▁▁▁▁▁": 1535,
+      "▁this": 1536,
+      "ass": 1537,
+      "▁$": 1538,
+      "ure": 1539,
+      "ld": 1540,
+      "all": 1541,
+      "}\\": 1542,
+      "ave": 1543,
+      "▁he": 1544,
+      ");\\": 1545,
+      "ial": 1546,
+      "com": 1547,
+      "▁ch": 1548,
+      "▁O": 1549,
+      "ight": 1550,
+      "cl": 1551,
+      "ction": 1552,
+      "nT": 1553,
+      "▁we": 1554,
+      "##": 1555,
+      "ated": 1556,
+      "iz": 1557,
+      "ack": 1558,
+      "ult": 1559,
+      "▁tr": 1560,
+      "ge": 1561,
+      "ip": 1562,
+      "our": 1563,
+      "▁de": 1564,
+      "int": 1565,
+      "//": 1566,
+      "for": 1567,
+      "ell": 1568,
+      "ust": 1569,
+      "▁sh": 1570,
+      "▁can": 1571,
+      "ost": 1572,
+      "▁not": 1573,
+      "out": 1574,
+      "▁\\\\": 1575,
+      "per": 1576,
+      "▁se": 1577,
+      "ard": 1578,
+      "are": 1579,
+      "▁have": 1580,
+      "▁im": 1581,
+      "ase": 1582,
+      "ical": 1583,
+      "▁cl": 1584,
+      "ind": 1585,
+      "tr": 1586,
+      "▁{\\": 1587,
+      "**": 1588,
+      "rou": 1589,
+      "pe": 1590,
+      "av": 1591,
+      "ions": 1592,
+      "▁le": 1593,
+      "▁pl": 1594,
+      "ff": 1595,
+      "con": 1596,
+      "able": 1597,
+      "ath": 1598,
+      "ib": 1599,
+      "ide": 1600,
+      "▁will": 1601,
+      "▁comp": 1602,
+      "ile": 1603,
+      "ire": 1604,
+      "urn": 1605,
+      "ast": 1606,
+      "ok": 1607,
+      "▁res": 1608,
+      "▁J": 1609,
+      "ork": 1610,
+      "ans": 1611,
+      "▁U": 1612,
+      "ice": 1613,
+      "▁*": 1614,
+      "▁your": 1615,
+      "▁j": 1616,
+      "▁ad": 1617,
+      "▁sp": 1618,
+      "ould": 1619,
+      "ich": 1620,
+      "▁k": 1621,
+      "ak": 1622,
+      "ign": 1623,
+      "ime": 1624,
+      "du": 1625,
+      "ret": 1626,
+      "ia": 1627,
+      "▁ab": 1628,
+      "ary": 1629,
+      "ail": 1630,
+      "▁\\": 1631,
+      "▁Th": 1632,
+      "ace": 1633,
+      "执行": 1634,
+      "ong": 1635,
+      "ep": 1636,
+      "elf": 1637,
+      "ents": 1638,
+      "one": 1639,
+      "ber": 1640,
+      "ations": 1641,
+      "ear": 1642,
+      "ens": 1643,
+      "pro": 1644,
+      "▁all": 1645,
+      "▁In": 1646,
+      "ence": 1647,
+      "▁ap": 1648,
+      "▁has": 1649,
+      "▁int": 1650,
+      "ext": 1651,
+      ">\\": 1652,
+      "ome": 1653,
+      "nI": 1654,
+      "ord": 1655,
+      "so": 1656,
+      "ang": 1657,
+      "form": 1658,
+      "ance": 1659,
+      "und": 1660,
+      "lic": 1661,
+      "▁cont": 1662,
+      "ata": 1663,
+      "▁me": 1664,
+      "cc": 1665,
+      "▁|\\": 1666,
+      "fer": 1667,
+      "hen": 1668,
+      "()": 1669,
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁": 1670,
+      "ite": 1671,
+      "人民": 1672,
+      "enc": 1673,
+      "ress": 1674,
+      "▁do": 1675,
+      "nA": 1676,
+      "iss": 1677,
+      "ject": 1678,
+      "ous": 1679,
+      "erv": 1680,
+      "▁K": 1681,
+      "ther": 1682,
+      "str": 1683,
+      "ach": 1684,
+      "lo": 1685,
+      "get": 1686,
+      "ry": 1687,
+      "name": 1688,
+      "tp": 1689,
+      "▁V": 1690,
+      "▁ne": 1691,
+      "vel": 1692,
+      "ory": 1693,
+      "▁un": 1694,
+      "gh": 1695,
+      "▁but": 1696,
+      "▁which": 1697,
+      "ors": 1698,
+      "ll": 1699,
+      "nThe": 1700,
+      "ition": 1701,
+      "nS": 1702,
+      "▁per": 1703,
+      "▁▁▁▁▁▁": 1704,
+      "▁en": 1705,
+      "公司": 1706,
+      "▁out": 1707,
+      "fr": 1708,
+      "▁[": 1709,
+      "ake": 1710,
+      "▁\\'": 1711,
+      "▁up": 1712,
+      "▁qu": 1713,
+      "ount": 1714,
+      "▁new": 1715,
+      "▁St": 1716,
+      "▁go": 1717,
+      "ook": 1718,
+      "ug": 1719,
+      "return": 1720,
+      "ose": 1721,
+      "▁ar": 1722,
+      "ublic": 1723,
+      "import": 1724,
+      "ft": 1725,
+      "▁dis": 1726,
+      "▁were": 1727,
+      "ple": 1728,
+      "ys": 1729,
+      "vers": 1730,
+      "</": 1731,
+      "\"\\": 1732,
+      "▁so": 1733,
+      "ode": 1734,
+      "ild": 1735,
+      "ally": 1736,
+      "ype": 1737,
+      "ph": 1738,
+      "te": 1739,
+      "▁str": 1740,
+      "ree": 1741,
+      "ks": 1742,
+      "),": 1743,
+      "一个": 1744,
+      "▁one": 1745,
+      "\":": 1746,
+      "ound": 1747,
+      "▁get": 1748,
+      "pon": 1749,
+      "own": 1750,
+      "▁also": 1751,
+      "▁inter": 1752,
+      "▁more": 1753,
+      "self": 1754,
+      "gr": 1755,
+      "{\\": 1756,
+      "▁sc": 1757,
+      "▁work": 1758,
+      "▁pre": 1759,
+      "oth": 1760,
+      "clud": 1761,
+      "被告": 1762,
+      "ont": 1763,
+      "]\\": 1764,
+      "ra": 1765,
+      "▁their": 1766,
+      "reat": 1767,
+      "unction": 1768,
+      "ely": 1769,
+      "ian": 1770,
+      "eth": 1771,
+      "▁'": 1772,
+      "row": 1773,
+      "very": 1774,
+      "yst": 1775,
+      "▁ev": 1776,
+      "nimport": 1777,
+      "ie": 1778,
+      "__": 1779,
+      "nR": 1780,
+      "\",": 1781,
+      "▁man": 1782,
+      "ish": 1783,
+      "ific": 1784,
+      "rib": 1785,
+      "▁<": 1786,
+      "we": 1787,
+      "▁Y": 1788,
+      "▁part": 1789,
+      "ob": 1790,
+      "ms": 1791,
+      "▁te": 1792,
+      "▁Ch": 1793,
+      "igh": 1794,
+      "本院": 1795,
+      "▁if": 1796,
+      "low": 1797,
+      "ings": 1798,
+      "odel": 1799,
+      "iew": 1800,
+      "}{": 1801,
+      "可以": 1802,
+      "lect": 1803,
+      "ov": 1804,
+      "mer": 1805,
+      "under": 1806,
+      "▁our": 1807,
+      "ark": 1808,
+      "ick": 1809,
+      "ond": 1810,
+      "原告": 1811,
+      "://": 1812,
+      "▁li": 1813,
+      "pr": 1814,
+      "irst": 1815,
+      "▁su": 1816,
+      "▁time": 1817,
+      "ystem": 1818,
+      "▁stud": 1819,
+      "▁pr": 1820,
+      "ict": 1821,
+      "ert": 1822,
+      "ates": 1823,
+      "wo": 1824,
+      "▁ro": 1825,
+      "▁It": 1826,
+      "ock": 1827,
+      "▁they": 1828,
+      "ays": 1829,
+      "▁des": 1830,
+      "申请": 1831,
+      "(\"": 1832,
+      "nC": 1833,
+      "ener": 1834,
+      "▁about": 1835,
+      "user": 1836,
+      "ink": 1837,
+      "▁class": 1838,
+      "io": 1839,
+      "ption": 1840,
+      "ax": 1841,
+      "▁any": 1842,
+      "▁been": 1843,
+      "ble": 1844,
+      "▁ag": 1845,
+      "nW": 1846,
+      "ient": 1847,
+      "use": 1848,
+      "old": 1849,
+      "ool": 1850,
+      "ize": 1851,
+      "put": 1852,
+      "ng": 1853,
+      "frac": 1854,
+      "uc": 1855,
+      "▁year": 1856,
+      "olog": 1857,
+      "ular": 1858,
+      "▁his": 1859,
+      "fter": 1860,
+      "ne": 1861,
+      "ater": 1862,
+      "▁my": 1863,
+      "ual": 1864,
+      "▁over": 1865,
+      "ove": 1866,
+      "alue": 1867,
+      "▁i": 1868,
+      "erm": 1869,
+      "ics": 1870,
+      "ased": 1871,
+      "▁who": 1872,
+      "ull": 1873,
+      "▁We": 1874,
+      "▁other": 1875,
+      "▁add": 1876,
+      "▁under": 1877,
+      "▁kn": 1878,
+      "▁ind": 1879,
+      "tern": 1880,
+      "▁bet": 1881,
+      "▁when": 1882,
+      "▁comm": 1883,
+      "▁col": 1884,
+      "ss": 1885,
+      "ix": 1886,
+      "▁there": 1887,
+      "les": 1888,
+      "hed": 1889,
+      "执行人": 1890,
+      "▁am": 1891,
+      "eld": 1892,
+      "sc": 1893,
+      "anc": 1894,
+      "▁some": 1895,
+      "▁use": 1896,
+      "rit": 1897,
+      "app": 1898,
+      "▁set": 1899,
+      "▁This": 1900,
+      "▁rec": 1901,
+      "):\\": 1902,
+      "▁dif": 1903,
+      "eg": 1904,
+      "ative": 1905,
+      "▁ass": 1906,
+      "ied": 1907,
+      "▁act": 1908,
+      "sh": 1909,
+      "▁than": 1910,
+      "中国": 1911,
+      "▁exp": 1912,
+      "ange": 1913,
+      "▁function": 1914,
+      "▁num": 1915,
+      "▁ob": 1916,
+      "roup": 1917,
+      "oint": 1918,
+      "▁no": 1919,
+      "ari": 1920,
+      "▁:": 1921,
+      "xa": 1922,
+      "nP": 1923,
+      "▁sup": 1924,
+      "http": 1925,
+      "▁sub": 1926,
+      "aw": 1927,
+      "法院": 1928,
+      "▁need": 1929,
+      "▁&": 1930,
+      "▁import": 1931,
+      "▁its": 1932,
+      "▁first": 1933,
+      "▁prov": 1934,
+      "sp": 1935,
+      "▁reg": 1936,
+      "ield": 1937,
+      "nf": 1938,
+      "▁two": 1939,
+      "▁ph": 1940,
+      "自己": 1941,
+      "ution": 1942,
+      "▁pe": 1943,
+      "ments": 1944,
+      "▁acc": 1945,
+      "arch": 1946,
+      "cre": 1947,
+      "▁eff": 1948,
+      "▁x": 1949,
+      "als": 1950,
+      "▁into": 1951,
+      "▁would": 1952,
+      "▁fe": 1953,
+      "ility": 1954,
+      "▁co": 1955,
+      "^{": 1956,
+      "第一": 1957,
+      "ade": 1958,
+      "ating": 1959,
+      "▁spec": 1960,
+      "ty": 1961,
+      "def": 1962,
+      "issue": 1963,
+      "velop": 1964,
+      "arg": 1965,
+      "ning": 1966,
+      "let": 1967,
+      "public": 1968,
+      "err": 1969,
+      "▁includ": 1970,
+      "▁trans": 1971,
+      "▁like": 1972,
+      "▁rel": 1973,
+      "ople": 1974,
+      "工作": 1975,
+      "▁pat": 1976,
+      "▁ac": 1977,
+      "有限": 1978,
+      "pos": 1979,
+      "▁how": 1980,
+      "▁high": 1981,
+      "▁may": 1982,
+      "▁def": 1983,
+      "par": 1984,
+      "▁requ": 1985,
+      "ns": 1986,
+      "nM": 1987,
+      "ational": 1988,
+      "问题": 1989,
+      "▁produ": 1990,
+      "▁–": 1991,
+      "rap": 1992,
+      "人民法院": 1993,
+      "orm": 1994,
+      "▁inv": 1995,
+      "line": 1996,
+      "urre": 1997,
+      "ise": 1998,
+      "我们": 1999,
+      "ER": 2000,
+      "▁sa": 2001,
+      "▁off": 2002,
+      "▁had": 2003,
+      "oy": 2004,
+      "nF": 2005,
+      "进行": 2006,
+      "nc": 2007,
+      "▁differ": 2008,
+      "▁these": 2009,
+      "Ex": 2010,
+      "$$": 2011,
+      "plic": 2012,
+      "cess": 2013,
+      "lp": 2014,
+      "uct": 2015,
+      "to": 2016,
+      "▁play": 2017,
+      "▁data": 2018,
+      "ON": 2019,
+      "▁pos": 2020,
+      "ures": 2021,
+      "inal": 2022,
+      "▁Un": 2023,
+      "▁what": 2024,
+      "▁only": 2025,
+      "()\\": 2026,
+      "ences": 2027,
+      "set": 2028,
+      "_{": 2029,
+      "ities": 2030,
+      "hat": 2031,
+      "▁det": 2032,
+      "::": 2033,
+      "nB": 2034,
+      "▁result": 2035,
+      "ev": 2036,
+      "ues": 2037,
+      "round": 2038,
+      "uild": 2039,
+      "nH": 2040,
+      "没有": 2041,
+      "▁just": 2042,
+      "▁cons": 2043,
+      "hes": 2044,
+      "▁op": 2045,
+      "诉讼": 2046,
+      "Pro": 2047,
+      "riv": 2048,
+      "ym": 2049,
+      "St": 2050,
+      "▁em": 2051,
+      "▁dec": 2052,
+      "ices": 2053,
+      "▁them": 2054,
+      "▁after": 2055,
+      "(\\'": 2056,
+      "ages": 2057,
+      "log": 2058,
+      "ident": 2059,
+      "有限公司": 2060,
+      "rough": 2061,
+      "什么": 2062,
+      "▁He": 2063,
+      "oid": 2064,
+      "ible": 2065,
+      "ser": 2066,
+      "▁using": 2067,
+      "▁sign": 2068,
+      "ense": 2069,
+      "▁system": 2070,
+      "▁form": 2071,
+      "ons": 2072,
+      "ator": 2073,
+      "\",\\": 2074,
+      "ames": 2075,
+      "▁test": 2076,
+      "▁loc": 2077,
+      "▁Pro": 2078,
+      "its": 2079,
+      "时间": 2080,
+      "uring": 2081,
+      "right": 2082,
+      "▁develop": 2083,
+      "ethod": 2084,
+      "▁serv": 2085,
+      "▁used": 2086,
+      "▁show": 2087,
+      "key": 2088,
+      "nL": 2089,
+      "username": 2090,
+      "ollow": 2091,
+      "ox": 2092,
+      "nTh": 2093,
+      "▁sec": 2094,
+      "ween": 2095,
+      "ients": 2096,
+      "IN": 2097,
+      "uth": 2098,
+      "▁}": 2099,
+      "▁her": 2100,
+      "规定": 2101,
+      "atch": 2102,
+      "cont": 2103,
+      "▁people": 2104,
+      "▁pres": 2105,
+      "▁dist": 2106,
+      "nD": 2107,
+      "ck": 2108,
+      "就是": 2109,
+      "▁then": 2110,
+      "▁help": 2111,
+      "hem": 2112,
+      "this": 2113,
+      "read": 2114,
+      "ause": 2115,
+      "ash": 2116,
+      "cept": 2117,
+      "eng": 2118,
+      "▁well": 2119,
+      "ner": 2120,
+      "▁most": 2121,
+      "▁sim": 2122,
+      "▁br": 2123,
+      "atic": 2124,
+      "ting": 2125,
+      "yp": 2126,
+      "▁know": 2127,
+      "▁through": 2128,
+      "min": 2129,
+      "tring": 2130,
+      "▁between": 2131,
+      "▁where": 2132,
+      "▁cell": 2133,
+      "der": 2134,
+      "▁ret": 2135,
+      "▁pol": 2136,
+      "▁@": 2137,
+      "irect": 2138,
+      "ited": 2139,
+      "▁rem": 2140,
+      "();\\": 2141,
+      "▁Com": 2142,
+      "▁cre": 2143,
+      "▁make": 2144,
+      "\\\\(": 2145,
+      "air": 2146,
+      "▁should": 2147,
+      "▁`": 2148,
+      "work": 2149,
+      "这个": 2150,
+      "▁exper": 2151,
+      "fig": 2152,
+      "In": 2153,
+      "▁sur": 2154,
+      "被执行人": 2155,
+      "管理": 2156,
+      "ood": 2157,
+      "▁model": 2158,
+      "urrent": 2159,
+      "ale": 2160,
+      "ute": 2161,
+      "son": 2162,
+      "▁find": 2163,
+      "▁end": 2164,
+      "\\',": 2165,
+      "rt": 2166,
+      "▁sol": 2167,
+      "ily": 2168,
+      "裁定": 2169,
+      "oot": 2170,
+      "amp": 2171,
+      "发展": 2172,
+      "RE": 2173,
+      "ars": 2174,
+      "ward": 2175,
+      "ower": 2176,
+      "}\\\\": 2177,
+      "ious": 2178,
+      "▁ser": 2179,
+      "▁fil": 2180,
+      "over": 2181,
+      "class": 2182,
+      "▁att": 2183,
+      "fore": 2184,
+      "▁follow": 2185,
+      "any": 2186,
+      "ines": 2187,
+      "raph": 2188,
+      "▁number": 2189,
+      "ough": 2190,
+      "ml": 2191,
+      "▁sm": 2192,
+      "▁look": 2193,
+      "▁bl": 2194,
+      "nO": 2195,
+      "▁inst": 2196,
+      "▁gr": 2197,
+      "▁see": 2198,
+      "ee": 2199,
+      "ature": 2200,
+      "itle": 2201,
+      "ized": 2202,
+      "vent": 2203,
+      "▁min": 2204,
+      "服务": 2205,
+      "gram": 2206,
+      "ource": 2207,
+      "Qu": 2208,
+      "day": 2209,
+      "▁vari": 2210,
+      "br": 2211,
+      "ery": 2212,
+      "▁such": 2213,
+      "AT": 2214,
+      "az": 2215,
+      "法律": 2216,
+      "企业": 2217,
+      "▁effect": 2218,
+      "The": 2219,
+      "▁gener": 2220,
+      "中华": 2221,
+      "cent": 2222,
+      "▁different": 2223,
+      "formation": 2224,
+      "ts": 2225,
+      "orld": 2226,
+      "▁ent": 2227,
+      "estion": 2228,
+      "view": 2229,
+      "ier": 2230,
+      "ank": 2231,
+      "ton": 2232,
+      "ins": 2233,
+      "ann": 2234,
+      "▁want": 2235,
+      "▁group": 2236,
+      "▁back": 2237,
+      "▁/": 2238,
+      "oci": 2239,
+      "第二": 2240,
+      "####": 2241,
+      "▁cor": 2242,
+      "▁years": 2243,
+      "blem": 2244,
+      "▁par": 2245,
+      "▁inf": 2246,
+      "ection": 2247,
+      "const": 2248,
+      "('": 2249,
+      "chn": 2250,
+      "cri": 2251,
+      "）\\": 2252,
+      "pect": 2253,
+      "技术": 2254,
+      "ider": 2255,
+      "▁di": 2256,
+      "和国": 2257,
+      "▁name": 2258,
+      "comment": 2259,
+      "val": 2260,
+      "ren": 2261,
+      "▁mon": 2262,
+      "/\\": 2263,
+      "▁said": 2264,
+      "lease": 2265,
+      "ined": 2266,
+      "SU": 2267,
+      "IT": 2268,
+      "ases": 2269,
+      "▁Al": 2270,
+      "▁she": 2271,
+      "land": 2272,
+      "如果": 2273,
+      "▁bec": 2274,
+      "NA": 2275,
+      "▁self": 2276,
+      "ired": 2277,
+      "▁Tr": 2278,
+      "▁start": 2279,
+      "https": 2280,
+      "▁med": 2281,
+      "data": 2282,
+      "nIn": 2283,
+      "div": 2284,
+      "ful": 2285,
+      "▁could": 2286,
+      "▁fr": 2287,
+      "▁long": 2288,
+      "add": 2289,
+      "▁equ": 2290,
+      "▁every": 2291,
+      "▁You": 2292,
+      "共和国": 2293,
+      "ify": 2294,
+      "▁fl": 2295,
+      "ric": 2296,
+      "ement": 2297,
+      "▁incre": 2298,
+      "ample": 2299,
+      "ality": 2300,
+      "ince": 2301,
+      "▁many": 2302,
+      "▁found": 2303,
+      "ife": 2304,
+      "gan": 2305,
+      "hip": 2306,
+      "nN": 2307,
+      "ists": 2308,
+      "alth": 2309,
+      "ever": 2310,
+      "人民共和国": 2311,
+      "ST": 2312,
+      "ault": 2313,
+      "▁call": 2314,
+      "中华人民共和国": 2315,
+      "▁mod": 2316,
+      "wn": 2317,
+      "ced": 2318,
+      "▁both": 2319,
+      "ument": 2320,
+      "▁mem": 2321,
+      "ray": 2322,
+      "ically": 2323,
+      "AR": 2324,
+      "ole": 2325,
+      "ology": 2326,
+      "red": 2327,
+      "))": 2328,
+      "ody": 2329,
+      "民事": 2330,
+      "file": 2331,
+      "erson": 2332,
+      "▁▁▁▁▁▁▁▁▁▁": 2333,
+      "heck": 2334,
+      "▁anal": 2335,
+      "▁now": 2336,
+      "uch": 2337,
+      "imes": 2338,
+      "▁app": 2339,
+      "▁char": 2340,
+      "rol": 2341,
+      "',": 2342,
+      "EN": 2343,
+      "lection": 2344,
+      "▁value": 2345,
+      "comm": 2346,
+      "判决": 2347,
+      "ead": 2348,
+      "itt": 2349,
+      "led": 2350,
+      "▁each": 2351,
+      "▁right": 2352,
+      "研究": 2353,
+      "▁fact": 2354,
+      "orn": 2355,
+      "ization": 2356,
+      "▁main": 2357,
+      "▁appro": 2358,
+      "ron": 2359,
+      "▁design": 2360,
+      "▁process": 2361,
+      "▁very": 2362,
+      "requ": 2363,
+      "▁ed": 2364,
+      "ivers": 2365,
+      "▁$$": 2366,
+      "▁study": 2367,
+      "▁car": 2368,
+      "▁way": 2369,
+      "▁av": 2370,
+      "ik": 2371,
+      "▁method": 2372,
+      "▁does": 2373,
+      "▁val": 2374,
+      "sup": 2375,
+      "▁read": 2376,
+      "ience": 2377,
+      "▁respon": 2378,
+      "ets": 2379,
+      "](": 2380,
+      "▁list": 2381,
+      "of": 2382,
+      "▁gu": 2383,
+      "text": 2384,
+      "▁post": 2385,
+      "load": 2386,
+      "▁Ex": 2387,
+      "▁patients": 2388,
+      "alse": 2389,
+      "Name": 2390,
+      "col": 2391,
+      "ision": 2392,
+      "ule": 2393,
+      "chool": 2394,
+      "aking": 2395,
+      "sub": 2396,
+      "version": 2397,
+      "需要": 2398,
+      "▁run": 2399,
+      "▁ext": 2400,
+      "ware": 2401,
+      "▁est": 2402,
+      "ttp": 2403,
+      "▁New": 2404,
+      "ten": 2405,
+      "ared": 2406,
+      "▁great": 2407,
+      "的人": 2408,
+      "ulation": 2409,
+      "OR": 2410,
+      "other": 2411,
+      "▁point": 2412,
+      "way": 2413,
+      "meric": 2414,
+      "▁disc": 2415,
+      "search": 2416,
+      "ants": 2417,
+      "ected": 2418,
+      "▁Sh": 2419,
+      "ah": 2420,
+      "因为": 2421,
+      "ason": 2422,
+      "不是": 2423,
+      "▁don": 2424,
+      "iqu": 2425,
+      "▁An": 2426,
+      "信息": 2427,
+      "efer": 2428,
+      "dd": 2429,
+      "essage": 2430,
+      "Cont": 2431,
+      "使用": 2432,
+      "▁$\\\\": 2433,
+      "▁information": 2434,
+      "fil": 2435,
+      "ional": 2436,
+      "▁rep": 2437,
+      "▁here": 2438,
+      "▁fam": 2439,
+      "iness": 2440,
+      "ouse": 2441,
+      "▁If": 2442,
+      "ential": 2443,
+      "path": 2444,
+      "the": 2445,
+      "▁art": 2446,
+      "yn": 2447,
+      "**\\": 2448,
+      "根据": 2449,
+      "man": 2450,
+      "\"\"": 2451,
+      "▁activ": 2452,
+      "ps": 2453,
+      "系统": 2454,
+      "产品": 2455,
+      "▁support": 2456,
+      "日期": 2457,
+      "▁giv": 2458,
+      "▁during": 2459,
+      "opy": 2460,
+      "▁same": 2461,
+      "ateg": 2462,
+      "▁file": 2463,
+      "type": 2464,
+      "通过": 2465,
+      "print": 2466,
+      "ished": 2467,
+      "ral": 2468,
+      "ek": 2469,
+      "审判": 2470,
+      "▁good": 2471,
+      "IG": 2472,
+      "▁line": 2473,
+      "所以": 2474,
+      "date": 2475,
+      "ural": 2476,
+      "icense": 2477,
+      "▁even": 2478,
+      "SE": 2479,
+      "###": 2480,
+      "中文": 2481,
+      "else": 2482,
+      "nE": 2483,
+      "\")": 2484,
+      "uss": 2485,
+      "}$": 2486,
+      "▁type": 2487,
+      "ency": 2488,
+      "▁level": 2489,
+      "de": 2490,
+      "her": 2491,
+      "认为": 2492,
+      "▁because": 2493,
+      "yl": 2494,
+      "▁bel": 2495,
+      "▁>": 2496,
+      "cer": 2497,
+      "▁%": 2498,
+      "aining": 2499,
+      "url": 2500,
+      "ME": 2501,
+      "主要": 2502,
+      "▁techn": 2503,
+      "nfrom": 2504,
+      "▁perform": 2505,
+      "▁state": 2506,
+      "▁cent": 2507,
+      "ium": 2508,
+      "oh": 2509,
+      "▁met": 2510,
+      "rivate": 2511,
+      "bers": 2512,
+      "▁program": 2513,
+      "ified": 2514,
+      "▁project": 2515,
+      "提供": 2516,
+      "▁again": 2517,
+      "▁Re": 2518,
+      "ask": 2519,
+      "action": 2520,
+      "ains": 2521,
+      "▁For": 2522,
+      "wh": 2523,
+      "▁down": 2524,
+      "时候": 2525,
+      "教育": 2526,
+      "▁mat": 2527,
+      "ces": 2528,
+      "roll": 2529,
+      "▁current": 2530,
+      "资料": 2531,
+      "play": 2532,
+      "▁before": 2533,
+      "▁Cl": 2534,
+      "▁err": 2535,
+      "lex": 2536,
+      "Id": 2537,
+      "bs": 2538,
+      "选择": 2539,
+      "▁z": 2540,
+      "AN": 2541,
+      "▁partic": 2542,
+      "如下": 2543,
+      "ustom": 2544,
+      "ters": 2545,
+      "pp": 2546,
+      "▁gen": 2547,
+      "▁Ar": 2548,
+      "aj": 2549,
+      "ved": 2550,
+      "▁mult": 2551,
+      "ams": 2552,
+      "back": 2553,
+      "▁second": 2554,
+      "▁dep": 2555,
+      "],": 2556,
+      "ually": 2557,
+      "▁best": 2558,
+      "Con": 2559,
+      "▁being": 2560,
+      "test": 2561,
+      "eferences": 2562,
+      "▁made": 2563,
+      "bl": 2564,
+      "usiness": 2565,
+      "▁three": 2566,
+      "参考": 2567,
+      "▁care": 2568,
+      "▁provid": 2569,
+      "大学": 2570,
+      "Com": 2571,
+      "▁Z": 2572,
+      "▁“": 2573,
+      "****": 2574,
+      "String": 2575,
+      "ior": 2576,
+      "ID": 2577,
+      "ote": 2578,
+      "▁did": 2579,
+      "ression": 2580,
+      "▁mark": 2581,
+      "可能": 2582,
+      "ism": 2583,
+      "▁Con": 2584,
+      "dfrac": 2585,
+      "CT": 2586,
+      "▁hand": 2587,
+      "yle": 2588,
+      "▁direct": 2589,
+      "ribut": 2590,
+      "ital": 2591,
+      "社会": 2592,
+      "ability": 2593,
+      "ccess": 2594,
+      "▁product": 2595,
+      "▁while": 2596,
+      "sq": 2597,
+      "wer": 2598,
+      "▁count": 2599,
+      "他们": 2600,
+      "发生": 2601,
+      "nReferences": 2602,
+      "▁dr": 2603,
+      "国家": 2604,
+      "ithub": 2605,
+      "ackage": 2606,
+      "▁lead": 2607,
+      "▁those": 2608,
+      "生活": 2609,
+      "▁build": 2610,
+      "ane": 2611,
+      "方法": 2612,
+      "▁fin": 2613,
+      "LE": 2614,
+      "acter": 2615,
+      "clude": 2616,
+      "▁open": 2617,
+      "▁*/\\": 2618,
+      "arn": 2619,
+      ",\"": 2620,
+      "审理": 2621,
+      "情况": 2622,
+      "ends": 2623,
+      "▁results": 2624,
+      "abel": 2625,
+      "ball": 2626,
+      "但是": 2627,
+      "nG": 2628,
+      "▁think": 2629,
+      "py": 2630,
+      "using": 2631,
+      "nThis": 2632,
+      "诉讼法": 2633,
+      "现在": 2634,
+      "▁non": 2635,
+      "▁much": 2636,
+      "▁ins": 2637,
+      "ross": 2638,
+      "经济": 2639,
+      "的一": 2640,
+      "arm": 2641,
+      "be": 2642,
+      "var": 2643,
+      "▁child": 2644,
+      "by": 2645,
+      "net": 2646,
+      "org": 2647,
+      "face": 2648,
+      "是一": 2649,
+      "▁report": 2650,
+      "alc": 2651,
+      "▁last": 2652,
+      "▁#": 2653,
+      "提出": 2654,
+      "▁How": 2655,
+      "▁commun": 2656,
+      "icro": 2657,
+      "itive": 2658,
+      "▁cap": 2659,
+      "ength": 2660,
+      "▁take": 2661,
+      "ained": 2662,
+      "lish": 2663,
+      "▁del": 2664,
+      "\")\\": 2665,
+      "atus": 2666,
+      "ograph": 2667,
+      "已经": 2668,
+      "设计": 2669,
+      "▁poss": 2670,
+      "▁present": 2671,
+      "ta": 2672,
+      "▁invest": 2673,
+      "come": 2674,
+      "▁order": 2675,
+      "▁day": 2676,
+      "IC": 2677,
+      "▁书": 2678,
+      "▁oper": 2679,
+      "上诉": 2680,
+      "ows": 2681,
+      "equ": 2682,
+      "ger": 2683,
+      "▁som": 2684,
+      "▁As": 2685,
+      "▁research": 2686,
+      "分析": 2687,
+      "出生": 2688,
+      "stand": 2689,
+      "▁opt": 2690,
+      "ervice": 2691,
+      "\\\\)": 2692,
+      "time": 2693,
+      "arget": 2694,
+      "▁iss": 2695,
+      "vir": 2696,
+      "▁home": 2697,
+      "还是": 2698,
+      "▁vis": 2699,
+      "ather": 2700,
+      "▁own": 2701,
+      "▁return": 2702,
+      "▁ant": 2703,
+      "github": 2704,
+      "▁top": 2705,
+      "代表": 2706,
+      "▁Americ": 2707,
+      "uthor": 2708,
+      "▁sk": 2709,
+      "ailable": 2710,
+      "html": 2711,
+      "▁http": 2712,
+      "math": 2713,
+      "ness": 2714,
+      "▁treat": 2715,
+      "Tr": 2716,
+      "▁proper": 2717,
+      "▁small": 2718,
+      "引用": 2719,
+      "▁cr": 2720,
+      "))\\": 2721,
+      "▁applic": 2722,
+      "EF": 2723,
+      "汉族": 2724,
+      "内容": 2725,
+      "left": 2726,
+      "rr": 2727,
+      "ma": 2728,
+      "ugg": 2729,
+      "▁chang": 2730,
+      "ively": 2731,
+      "▁Sc": 2732,
+      "duct": 2733,
+      "oad": 2734,
+      "util": 2735,
+      "ception": 2736,
+      "iversity": 2737,
+      "▁And": 2738,
+      "Th": 2739,
+      "要求": 2740,
+      "new": 2741,
+      "ream": 2742,
+      "市场": 2743,
+      "▁pop": 2744,
+      "▁sing": 2745,
+      "orth": 2746,
+      "Cl": 2747,
+      "▁De": 2748,
+      "ES": 2749,
+      "▁without": 2750,
+      "akes": 2751,
+      "▁ident": 2752,
+      "出版": 2753,
+      "▁including": 2754,
+      "AS": 2755,
+      "ems": 2756,
+      "ysis": 2757,
+      "▁Col": 2758,
+      "▁true": 2759,
+      "▁health": 2760,
+      "过程": 2761,
+      "其他": 2762,
+      "reg": 2763,
+      "▁Ind": 2764,
+      "▁_": 2765,
+      "med": 2766,
+      "string": 2767,
+      "bin": 2768,
+      "申请执行人": 2769,
+      "ister": 2770,
+      "}}": 2771,
+      "▁string": 2772,
+      "引用日期": 2773,
+      "dit": 2774,
+      "start": 2775,
+      "uro": 2776,
+      "lement": 2777,
+      "Ch": 2778,
+      "▁path": 2779,
+      "▁elect": 2780,
+      "List": 2781,
+      "▁tri": 2782,
+      "▁cle": 2783,
+      "代理": 2784,
+      "▁X": 2785,
+      "▁based": 2786,
+      "▁person": 2787,
+      "AL": 2788,
+      "alk": 2789,
+      "ats": 2790,
+      "cription": 2791,
+      "\">": 2792,
+      "include": 2793,
+      "\">\\": 2794,
+      "民事诉讼法": 2795,
+      "erg": 2796,
+      "ava": 2797,
+      "tract": 2798,
+      "”\\": 2799,
+      "▁following": 2800,
+      "yt": 2801,
+      "ives": 2802,
+      "Type": 2803,
+      "Res": 2804,
+      "审判员": 2805,
+      "reen": 2806,
+      "ths": 2807,
+      "ocument": 2808,
+      "▁world": 2809,
+      "▁life": 2810,
+      "arge": 2811,
+      "oin": 2812,
+      "中华人民共和国民事诉讼法": 2813,
+      "▁Wh": 2814,
+      "ivid": 2815,
+      "li": 2816,
+      "word": 2817,
+      "▁signific": 2818,
+      "ety": 2819,
+      "▁ke": 2820,
+      "omen": 2821,
+      "less": 2822,
+      "▁中文": 2823,
+      "sqrt": 2824,
+      "ey": 2825,
+      "▁meas": 2826,
+      "pose": 2827,
+      "▁const": 2828,
+      "irc": 2829,
+      "▁Res": 2830,
+      "olution": 2831,
+      "▁struct": 2832,
+      "amet": 2833,
+      "▁business": 2834,
+      "▁aut": 2835,
+      "▁Pl": 2836,
+      "▁eas": 2837,
+      "▁ear": 2838,
+      "都是": 2839,
+      "▁team": 2840,
+      "ref": 2841,
+      "por": 2842,
+      "▁pass": 2843,
+      "▁case": 2844,
+      "文化": 2845,
+      "ven": 2846,
+      "ury": 2847,
+      "▁bu": 2848,
+      "▁free": 2849,
+      "ration": 2850,
+      "aces": 2851,
+      "▁writ": 2852,
+      "很多": 2853,
+      "中的": 2854,
+      "财产": 2855,
+      "oard": 2856,
+      "ways": 2857,
+      "▁adv": 2858,
+      "项目": 2859,
+      "人的": 2860,
+      "▁Fr": 2861,
+      "▁License": 2862,
+      "▁中文名": 2863,
+      "▁field": 2864,
+      "▁rece": 2865,
+      "ission": 2866,
+      "学习": 2867,
+      "二〇": 2868,
+      "},\\": 2869,
+      "▁Sp": 2870,
+      "aut": 2871,
+      "▁available": 2872,
+      "▁control": 2873,
+      "au": 2874,
+      "urs": 2875,
+      "aper": 2876,
+      "rem": 2877,
+      "SUP": 2878,
+      "▁contin": 2879,
+      "▁cells": 2880,
+      "np": 2881,
+      "学生": 2882,
+      "二〇一": 2883,
+      "数据": 2884,
+      "▁place": 2885,
+      "nY": 2886,
+      "或者": 2887,
+      "ina": 2888,
+      "▁provide": 2889,
+      "依照": 2890,
+      "dis": 2891,
+      "ache": 2892,
+      "建设": 2893,
+      "专业": 2894,
+      "mod": 2895,
+      "ape": 2896,
+      "不同": 2897,
+      "ternal": 2898,
+      "▁Is": 2899,
+      "lob": 2900,
+      "开始": 2901,
+      "▁real": 2902,
+      "ush": 2903,
+      "][": 2904,
+      "▁problem": 2905,
+      "▁dem": 2906,
+      "向本院": 2907,
+      "index": 2908,
+      "icle": 2909,
+      "rc": 2910,
+      "▁still": 2911,
+      "ived": 2912,
+      "作者": 2913,
+      "▁around": 2914,
+      "▁tem": 2915,
+      "js": 2916,
+      "FIG": 2917,
+      "lete": 2918,
+      "▁cour": 2919,
+      "▁stand": 2920,
+      "ope": 2921,
+      "书记": 2922,
+      "\\'\\": 2923,
+      "viron": 2924,
+      "世界": 2925,
+      "aster": 2926,
+      "▁hum": 2927,
+      "not": 2928,
+      "▁organ": 2929,
+      "▁expl": 2930,
+      "vert": 2931,
+      "这样": 2932,
+      "outh": 2933,
+      "cm": 2934,
+      "ploy": 2935,
+      "inks": 2936,
+      "til": 2937,
+      "活动": 2938,
+      "rite": 2939,
+      "▁key": 2940,
+      "知道": 2941,
+      "参考资料": 2942,
+      "tem": 2943,
+      "▁Ph": 2944,
+      "▁log": 2945,
+      "ww": 2946,
+      "▁significant": 2947,
+      "resent": 2948,
+      "工程": 2949,
+      "init": 2950,
+      "ails": 2951,
+      "银行": 2952,
+      "{\\\\": 2953,
+      "▁book": 2954,
+      "▁grow": 2955,
+      "▁development": 2956,
+      "▁area": 2957,
+      "关于": 2958,
+      "▁impro": 2959,
+      "▁ve": 2960,
+      "▁determ": 2961,
+      "重要": 2962,
+      "▁lim": 2963,
+      "iving": 2964,
+      "▁Let": 2965,
+      "att": 2966,
+      "aving": 2967,
+      "els": 2968,
+      "list": 2969,
+      "以及": 2970,
+      "arent": 2971,
+      "▁There": 2972,
+      "的是": 2973,
+      "ots": 2974,
+      "ndef": 2975,
+      "ification": 2976,
+      "▁sever": 2977,
+      "af": 2978,
+      "安全": 2979,
+      "▁important": 2980,
+      "rel": 2981,
+      "angu": 2982,
+      "关系": 2983,
+      "oss": 2984,
+      "www": 2985,
+      "方式": 2986,
+      "好的": 2987,
+      "ait": 2988,
+      "▁Se": 2989,
+      "本案": 2990,
+      "▁mus": 2991,
+      "游戏": 2992,
+      "▁redu": 2993,
+      "▁family": 2994,
+      "▁him": 2995,
+      "crib": 2996,
+      "▁conf": 2997,
+      "▁bi": 2998,
+      "中心": 2999,
+      "mun": 3000,
+      "部分": 3001,
+      "▁addition": 3002,
+      "na": 3003,
+      "value": 3004,
+      "▁head": 3005,
+      "▁定": 3006,
+      "opt": 3007,
+      "nal": 3008,
+      "),\\": 3009,
+      "依法": 3010,
+      "▁local": 3011,
+      "比较": 3012,
+      "▁Le": 3013,
+      "lu": 3014,
+      "▁To": 3015,
+      "ott": 3016,
+      "struct": 3017,
+      "影响": 3018,
+      "▁within": 3019,
+      "▁change": 3020,
+      "▁since": 3021,
+      "uck": 3022,
+      "——": 3023,
+      "ique": 3024,
+      "自己的": 3025,
+      "lished": 3026,
+      "ately": 3027,
+      "不能": 3028,
+      "ght": 3029,
+      "▁All": 3030,
+      "▁object": 3031,
+      "发现": 3032,
+      "▁manag": 3033,
+      "同时": 3034,
+      "▁imp": 3035,
+      "▁consider": 3036,
+      "怎么": 3037,
+      "vious": 3038,
+      "model": 3039,
+      "(\\": 3040,
+      "▁given": 3041,
+      "rror": 3042,
+      "▁associ": 3043,
+      "ION": 3044,
+      "相关": 3045,
+      "▁,": 3046,
+      "IS": 3047,
+      "生产": 3048,
+      "▁public": 3049,
+      "▁beh": 3050,
+      "etwork": 3051,
+      "▁Q": 3052,
+      "▁pot": 3053,
+      "▁bas": 3054,
+      "▁So": 3055,
+      "egr": 3056,
+      "Data": 3057,
+      "▁custom": 3058,
+      "▁service": 3059,
+      "▁user": 3060,
+      "ler": 3061,
+      "nEx": 3062,
+      "▁University": 3063,
+      "',\\": 3064,
+      "▁void": 3065,
+      "UT": 3066,
+      "▁What": 3067,
+      "计算": 3068,
+      "▁require": 3069,
+      "▁next": 3070,
+      "num": 3071,
+      "▁sit": 3072,
+      "▁bre": 3073,
+      "第一百": 3074,
+      "ember": 3075,
+      "fess": 3076,
+      "▁specific": 3077,
+      "▁code": 3078,
+      "▁try": 3079,
+      "ently": 3080,
+      "▁cou": 3081,
+      "AC": 3082,
+      "北京": 3083,
+      "▁power": 3084,
+      "ird": 3085,
+      "▁ide": 3086,
+      "\"\"\"\\": 3087,
+      "mp": 3088,
+      "▁view": 3089,
+      "能力": 3090,
+      "方面": 3091,
+      "umm": 3092,
+      "Question": 3093,
+      "▁must": 3094,
+      "字第": 3095,
+      "ftware": 3096,
+      "ern": 3097,
+      "▁really": 3098,
+      "被告人": 3099,
+      "ename": 3100,
+      "/**\\": 3101,
+      "lev": 3102,
+      "▁interest": 3103,
+      "也是": 3104,
+      "▁school": 3105,
+      "head": 3106,
+      "然后": 3107,
+      "一定": 3108,
+      "▁sl": 3109,
+      "▁hist": 3110,
+      "nWe": 3111,
+      "export": 3112,
+      "结果": 3113,
+      "▁let": 3114,
+      "osed": 3115,
+      "co": 3116,
+      "▁appe": 3117,
+      "参加": 3118,
+      "met": 3119,
+      "如何": 3120,
+      "epend": 3121,
+      "▁understand": 3122,
+      "网络": 3123,
+      "作为": 3124,
+      "▁pur": 3125,
+      "▁water": 3126,
+      "ples": 3127,
+      "go": 3128,
+      "▁week": 3129,
+      "AP": 3130,
+      "RO": 3131,
+      "履行": 3132,
+      "▁version": 3133,
+      "环境": 3134,
+      "的时候": 3135,
+      "itor": 3136,
+      "纠纷": 3137,
+      "▁range": 3138,
+      "\\')\\": 3139,
+      "ological": 3140,
+      "▁At": 3141,
+      "uff": 3142,
+      "ormal": 3143,
+      "\\',\\": 3144,
+      "▁create": 3145,
+      "pace": 3146,
+      "nclass": 3147,
+      "unc": 3148,
+      "期间": 3149,
+      "支持": 3150,
+      "▁four": 3151,
+      "aff": 3152,
+      "order": 3153,
+      "▁On": 3154,
+      "▁few": 3155,
+      "\");\\": 3156,
+      "idth": 3157,
+      "nREF": 3158,
+      "ured": 3159,
+      "iver": 3160,
+      "thon": 3161,
+      "▁better": 3162,
+      "▁But": 3163,
+      "程序": 3164,
+      "chan": 3165,
+      "];\\": 3166,
+      "▁prote": 3167,
+      "ules": 3168,
+      "ially": 3169,
+      "Col": 3170,
+      "▁prot": 3171,
+      "▁init": 3172,
+      "raw": 3173,
+      "vironment": 3174,
+      "▁record": 3175,
+      "▁offer": 3176,
+      "ET": 3177,
+      "▁company": 3178,
+      "表示": 3179,
+      "▁text": 3180,
+      "▁arg": 3181,
+      "大家": 3182,
+      "ither": 3183,
+      "▁event": 3184,
+      "组织": 3185,
+      "▁dise": 3186,
+      "▁character": 3187,
+      "这些": 3188,
+      "了一": 3189,
+      "rop": 3190,
+      "idd": 3191,
+      "孩子": 3192,
+      "▁check": 3193,
+      "合同": 3194,
+      "▁conn": 3195,
+      "ference": 3196,
+      "ething": 3197,
+      "yd": 3198,
+      "大的": 3199,
+      "▁dev": 3200,
+      "类型": 3201,
+      "政府": 3202,
+      "▁ref": 3203,
+      "ittle": 3204,
+      "▁occ": 3205,
+      "alcul": 3206,
+      "ene": 3207,
+      "那么": 3208,
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁▁": 3209,
+      "isk": 3210,
+      "▁foot": 3211,
+      "个人": 3212,
+      "ideo": 3213,
+      "科学": 3214,
+      "eb": 3215,
+      "所有": 3216,
+      "▁treatment": 3217,
+      "答案": 3218,
+      "第三": 3219,
+      "▁going": 3220,
+      "▁sure": 3221,
+      ")$": 3222,
+      "aterial": 3223,
+      "▁Ap": 3224,
+      "开发": 3225,
+      "选项": 3226,
+      "▁Ad": 3227,
+      "hib": 3228,
+      "atures": 3229,
+      "loc": 3230,
+      "ajor": 3231,
+      "uf": 3232,
+      "▁old": 3233,
+      "nU": 3234,
+      "▁access": 3235,
+      "▁null": 3236,
+      "▁cost": 3237,
+      "▁裁": 3238,
+      "ides": 3239,
+      "一案": 3240,
+      "裁定如下": 3241,
+      "学校": 3242,
+      "ai": 3243,
+      "基本": 3244,
+      "▁typ": 3245,
+      "AD": 3246,
+      "urther": 3247,
+      "▁orig": 3248,
+      "▁less": 3249,
+      "▁low": 3250,
+      "受理": 3251,
+      "nJ": 3252,
+      "对于": 3253,
+      "anguage": 3254,
+      "▁say": 3255,
+      "▁analysis": 3256,
+      "▁happ": 3257,
+      "▁prom": 3258,
+      "▁possible": 3259,
+      "ibr": 3260,
+      "▁__": 3261,
+      "bject": 3262,
+      "▁known": 3263,
+      "作品": 3264,
+      "▁too": 3265,
+      "▁size": 3266,
+      "acy": 3267,
+      "▁example": 3268,
+      "ves": 3269,
+      "▁cho": 3270,
+      "▁Or": 3271,
+      "px": 3272,
+      "plement": 3273,
+      "平台": 3274,
+      "▁sugg": 3275,
+      "ours": 3276,
+      "案件": 3277,
+      "某某": 3278,
+      "treturn": 3279,
+      "应用": 3280,
+      "[\\'": 3281,
+      "▁large": 3282,
+      "▁String": 3283,
+      "▁players": 3284,
+      "▁law": 3285,
+      "▁tot": 3286,
+      "inter": 3287,
+      "▁Te": 3288,
+      "▁mov": 3289,
+      "ocial": 3290,
+      "▁ans": 3291,
+      "▁against": 3292,
+      "ump": 3293,
+      "serv": 3294,
+      "▁American": 3295,
+      "▁investig": 3296,
+      "size": 3297,
+      "▁common": 3298,
+      "▁game": 3299,
+      "ung": 3300,
+      "cle": 3301,
+      "日出生": 3302,
+      "▁site": 3303,
+      "▁full": 3304,
+      "ume": 3305,
+      "iod": 3306,
+      "):": 3307,
+      "▁She": 3308,
+      "▁System": 3309,
+      "▁es": 3310,
+      "▁exam": 3311,
+      "处理": 3312,
+      "assert": 3313,
+      "非常": 3314,
+      "ised": 3315,
+      "alu": 3316,
+      "▁Br": 3317,
+      "▁Mar": 3318,
+      "▁fac": 3319,
+      "单位": 3320,
+      "imal": 3321,
+      "▁children": 3322,
+      "科技": 3323,
+      "IL": 3324,
+      "ering": 3325,
+      "▁gl": 3326,
+      "▁Pr": 3327,
+      "▁success": 3328,
+      "▁vol": 3329,
+      "▁They": 3330,
+      "▁single": 3331,
+      "▁spe": 3332,
+      "uction": 3333,
+      "▁request": 3334,
+      "ources": 3335,
+      "具有": 3336,
+      "ales": 3337,
+      "ctor": 3338,
+      "thing": 3339,
+      "ental": 3340,
+      "▁mean": 3341,
+      "这种": 3342,
+      "▁plan": 3343,
+      "▁eng": 3344,
+      "喜欢": 3345,
+      "▁Eng": 3346,
+      "▁propos": 3347,
+      "▁These": 3348,
+      "des": 3349,
+      "uary": 3350,
+      "▁another": 3351,
+      "ators": 3352,
+      "ponse": 3353,
+      "上海": 3354,
+      "ination": 3355,
+      "ED": 3356,
+      "inc": 3357,
+      ");": 3358,
+      "\",\"": 3359,
+      "wit": 3360,
+      "▁question": 3361,
+      "ny": 3362,
+      "▁rad": 3363,
+      "▁links": 3364,
+      "pi": 3365,
+      "上的": 3366,
+      "还有": 3367,
+      "mit": 3368,
+      "View": 3369,
+      "▁put": 3370,
+      "atory": 3371,
+      "到了": 3372,
+      "朋友": 3373,
+      "▁able": 3374,
+      "▁something": 3375,
+      "▁suggest": 3376,
+      "成为": 3377,
+      "▁light": 3378,
+      "▁phot": 3379,
+      "▁aff": 3380,
+      "▁market": 3381,
+      "▁include": 3382,
+      "nIt": 3383,
+      "▁u": 3384,
+      "▁experience": 3385,
+      "▁keep": 3386,
+      "▁pract": 3387,
+      "ended": 3388,
+      "ley": 3389,
+      "Error": 3390,
+      "mber": 3391,
+      "▁services": 3392,
+      "▁gover": 3393,
+      "Ar": 3394,
+      "><": 3395,
+      "agn": 3396,
+      "功能": 3397,
+      "ables": 3398,
+      "▁page": 3399,
+      "州市": 3400,
+      "\";\\": 3401,
+      "▁give": 3402,
+      "不会": 3403,
+      "private": 3404,
+      "即可": 3405,
+      "rix": 3406,
+      "Test": 3407,
+      "基础": 3408,
+      "AB": 3409,
+      "▁always": 3410,
+      "anks": 3411,
+      "▁studies": 3412,
+      "▁et": 3413,
+      "觉得": 3414,
+      "ories": 3415,
+      "▁whe": 3416,
+      "ries": 3417,
+      "arr": 3418,
+      "出现": 3419,
+      "iter": 3420,
+      "ivity": 3421,
+      "▁allow": 3422,
+      "ilar": 3423,
+      "▁days": 3424,
+      "MA": 3425,
+      "▁Can": 3426,
+      "▁might": 3427,
+      "▁left": 3428,
+      "▁working": 3429,
+      "▁invol": 3430,
+      "责任": 3431,
+      "▁install": 3432,
+      "ands": 3433,
+      "omet": 3434,
+      "ties": 3435,
+      "nIf": 3436,
+      "nusing": 3437,
+      "终结": 3438,
+      "ought": 3439,
+      "▁el": 3440,
+      "▁human": 3441,
+      "used": 3442,
+      "but": 3443,
+      "▁Cent": 3444,
+      "dict": 3445,
+      "▁false": 3446,
+      "▁prob": 3447,
+      "ouble": 3448,
+      "▁Count": 3449,
+      "▁changes": 3450,
+      "andom": 3451,
+      "▁la": 3452,
+      "▁id": 3453,
+      "HE": 3454,
+      "▁response": 3455,
+      "ground": 3456,
+      "To": 3457,
+      "gen": 3458,
+      "UR": 3459,
+      "($": 3460,
+      "'\\": 3461,
+      "▁error": 3462,
+      "成立": 3463,
+      "ividual": 3464,
+      "美国": 3465,
+      "▁quality": 3466,
+      "amespace": 3467,
+      "},": 3468,
+      "me": 3469,
+      "▁role": 3470,
+      "人员": 3471,
+      "两个": 3472,
+      "ript": 3473,
+      "▁Bl": 3474,
+      "▁special": 3475,
+      "直接": 3476,
+      "▁bro": 3477,
+      "rec": 3478,
+      "其中": 3479,
+      "国际": 3480,
+      "],\\": 3481,
+      "▁final": 3482,
+      "config": 3483,
+      "ici": 3484,
+      "ored": 3485,
+      "纠纷一案": 3486,
+      "mon": 3487,
+      "▁*\\": 3488,
+      "uration": 3489,
+      "申请人": 3490,
+      "▁May": 3491,
+      "eff": 3492,
+      "ling": 3493,
+      "ertain": 3494,
+      "▁ann": 3495,
+      "条件": 3496,
+      "▁mechan": 3497,
+      "inf": 3498,
+      "▁target": 3499,
+      "▁describ": 3500,
+      "iple": 3501,
+      "amb": 3502,
+      "一些": 3503,
+      "with": 3504,
+      "▁several": 3505,
+      "送达": 3506,
+      "▁paper": 3507,
+      "db": 3508,
+      "▁Gr": 3509,
+      "▁big": 3510,
+      "▁clin": 3511,
+      "▁series": 3512,
+      "你的": 3513,
+      "umn": 3514,
+      "▁little": 3515,
+      "▁action": 3516,
+      "▁obt": 3517,
+      "日起": 3518,
+      "▁However": 3519,
+      "apt": 3520,
+      "简介": 3521,
+      "城市": 3522,
+      "合作": 3523,
+      "uture": 3524,
+      "事实": 3525,
+      "▁ter": 3526,
+      "▁review": 3527,
+      "HO": 3528,
+      "rect": 3529,
+      "▁models": 3530,
+      "▁love": 3531,
+      "▁United": 3532,
+      "提高": 3533,
+      "ids": 3534,
+      "▁max": 3535,
+      "aps": 3536,
+      "▁inc": 3537,
+      "▁hard": 3538,
+      "';\\": 3539,
+      "idence": 3540,
+      "ording": 3541,
+      "▁US": 3542,
+      "▁potential": 3543,
+      "args": 3544,
+      "▁women": 3545,
+      "{{": 3546,
+      "▁rest": 3547,
+      "field": 3548,
+      "我的": 3549,
+      "知识": 3550,
+      "ending": 3551,
+      "\":\"": 3552,
+      "oph": 3553,
+      "▁short": 3554,
+      "fe": 3555,
+      "ably": 3556,
+      "▁estab": 3557,
+      "MP": 3558,
+      "范围": 3559,
+      "my": 3560,
+      "投资": 3561,
+      "▁foc": 3562,
+      "▁Sup": 3563,
+      "▁profess": 3564,
+      "ances": 3565,
+      "▁**": 3566,
+      "}\\\\)": 3567,
+      "▁major": 3568,
+      "▁either": 3569,
+      "▁ca": 3570,
+      "▁integr": 3571,
+      "raft": 3572,
+      "法定": 3573,
+      "▁bo": 3574,
+      "▁pay": 3575,
+      "api": 3576,
+      "▁required": 3577,
+      "sy": 3578,
+      "▁bir": 3579,
+      "▁exc": 3580,
+      "▁select": 3581,
+      "nn": 3582,
+      "▁season": 3583,
+      "ift": 3584,
+      "tt": 3585,
+      "▁card": 3586,
+      "sl": 3587,
+      "▁along": 3588,
+      "ention": 3589,
+      "一般": 3590,
+      "tif": 3591,
+      "▁evalu": 3592,
+      "▁represent": 3593,
+      "▁Be": 3594,
+      "rist": 3595,
+      "▁got": 3596,
+      "▁space": 3597,
+      "///": 3598,
+      "来的": 3599,
+      "ux": 3600,
+      "plate": 3601,
+      "目前": 3602,
+      "▁判": 3603,
+      "最高": 3604,
+      "适用": 3605,
+      "nter": 3606,
+      "▁default": 3607,
+      "治疗": 3608,
+      "▁}\\": 3609,
+      "借款": 3610,
+      "▁come": 3611,
+      "▁age": 3612,
+      "▁phys": 3613,
+      "▁step": 3614,
+      "ension": 3615,
+      "▁total": 3616,
+      "小说": 3617,
+      "利用": 3618,
+      "ival": 3619,
+      "▁fri": 3620,
+      "▁network": 3621,
+      "▁ben": 3622,
+      "▁事": 3623,
+      "anag": 3624,
+      "质量": 3625,
+      "书记员": 3626,
+      "为了": 3627,
+      "▁things": 3628,
+      "book": 3629,
+      "})": 3630,
+      "最后": 3631,
+      "▁org": 3632,
+      "委托": 3633,
+      "▁webs": 3634,
+      "标准": 3635,
+      "▁individual": 3636,
+      "ponent": 3637,
+      "▁called": 3638,
+      "▁until": 3639,
+      "cript": 3640,
+      "rict": 3641,
+      "▁due": 3642,
+      "▁static": 3643,
+      "ready": 3644,
+      "osis": 3645,
+      "leg": 3646,
+      "设备": 3647,
+      "};\\": 3648,
+      "▁der": 3649,
+      "▁students": 3650,
+      "▁times": 3651,
+      "介绍": 3652,
+      "count": 3653,
+      "二十": 3654,
+      "state": 3655,
+      "OS": 3656,
+      "机关": 3657,
+      "OT": 3658,
+      "▁turn": 3659,
+      "▁content": 3660,
+      "▁later": 3661,
+      "ply": 3662,
+      "作用": 3663,
+      "资源": 3664,
+      "illion": 3665,
+      "实现": 3666,
+      "});\\": 3667,
+      "off": 3668,
+      "}^{": 3669,
+      "▁leg": 3670,
+      "▁feel": 3671,
+      "nexport": 3672,
+      "\\')": 3673,
+      "\\':": 3674,
+      "ieve": 3675,
+      "cular": 3676,
+      "律师": 3677,
+      "定的": 3678,
+      "▁near": 3679,
+      "ogn": 3680,
+      "filename": 3681,
+      "▁sy": 3682,
+      "▁pub": 3683,
+      "▁Car": 3684,
+      "包括": 3685,
+      "▁else": 3686,
+      "▁complet": 3687,
+      "存在": 3688,
+      "param": 3689,
+      "abor": 3690,
+      "code": 3691,
+      "gin": 3692,
+      "▁comput": 3693,
+      "Problem": 3694,
+      "json": 3695,
+      "▁rate": 3696,
+      "▁disease": 3697,
+      "▁https": 3698,
+      "经营": 3699,
+      "▁side": 3700,
+      "行政": 3701,
+      "ING": 3702,
+      "而且": 3703,
+      "File": 3704,
+      "roid": 3705,
+      "sin": 3706,
+      "NAME": 3707,
+      "▁energ": 3708,
+      "Model": 3709,
+      "estions": 3710,
+      "▁Cont": 3711,
+      "起诉": 3712,
+      "连载": 3713,
+      "▁period": 3714,
+      "uit": 3715,
+      "▁ask": 3716,
+      "▁term": 3717,
+      "▁lot": 3718,
+      "verage": 3719,
+      "……": 3720,
+      "▁No": 3721,
+      "mb": 3722,
+      "Title": 3723,
+      "▁micro": 3724,
+      "$\\": 3725,
+      "lobal": 3726,
+      "ession": 3727,
+      "ribution": 3728,
+      "▁tool": 3729,
+      "input": 3730,
+      "cd": 3731,
+      "▁indic": 3732,
+      "解决": 3733,
+      "▁online": 3734,
+      "▁meet": 3735,
+      "▁enc": 3736,
+      "struction": 3737,
+      "五十": 3738,
+      "▁general": 3739,
+      "mm": 3740,
+      "符合": 3741,
+      "oney": 3742,
+      "一次": 3743,
+      "function": 3744,
+      "wards": 3745,
+      "效力": 3746,
+      "Sc": 3747,
+      "color": 3748,
+      "void": 3749,
+      "▁pain": 3750,
+      "▁standard": 3751,
+      "应该": 3752,
+      "ayer": 3753,
+      "ogle": 3754,
+      "教学": 3755,
+      "▁values": 3756,
+      "▁complex": 3757,
+      "arning": 3758,
+      "title": 3759,
+      "之规定": 3760,
+      "hel": 3761,
+      "content": 3762,
+      "urity": 3763,
+      "bum": 3764,
+      "max": 3765,
+      "▁image": 3766,
+      "本院认为": 3767,
+      "no": 3768,
+      "控制": 3769,
+      "ae": 3770,
+      "▁exist": 3771,
+      "tings": 3772,
+      "ography": 3773,
+      "pec": 3774,
+      "▁account": 3775,
+      "uk": 3776,
+      "nconst": 3777,
+      "▁community": 3778,
+      "md": 3779,
+      "▁discuss": 3780,
+      "本裁定": 3781,
+      "面积": 3782,
+      "ones": 3783,
+      "▁compet": 3784,
+      "行业": 3785,
+      "nder": 3786,
+      "▁environment": 3787,
+      "map": 3788,
+      "代理人": 3789,
+      "第二百": 3790,
+      "▁respect": 3791,
+      "民初": 3792,
+      "▁prim": 3793,
+      "▁football": 3794,
+      "▁body": 3795,
+      "▁associated": 3796,
+      "▁men": 3797,
+      "ising": 3798,
+      "ober": 3799,
+      "之后": 3800,
+      "US": 3801,
+      "材料": 3802,
+      "▁AN": 3803,
+      "▁Our": 3804,
+      "▁tre": 3805,
+      "▁above": 3806,
+      "函数": 3807,
+      "])\\": 3808,
+      "▁effects": 3809,
+      "▁sym": 3810,
+      "side": 3811,
+      "能够": 3812,
+      "▁activity": 3813,
+      "ney": 3814,
+      "▁ep": 3815,
+      "▁risk": 3816,
+      "aring": 3817,
+      "▁mar": 3818,
+      "▁solution": 3819,
+      ")/": 3820,
+      "▁material": 3821,
+      "▁sens": 3822,
+      "▁IS": 3823,
+      "usion": 3824,
+      "▁particular": 3825,
+      "error": 3826,
+      "▁air": 3827,
+      "品牌": 3828,
+      "▁early": 3829,
+      "nment": 3830,
+      "▁performance": 3831,
+      "lib": 3832,
+      "真的": 3833,
+      "mat": 3834,
+      "▁link": 3835,
+      "▁among": 3836,
+      "Par": 3837,
+      "全国": 3838,
+      "oms": 3839,
+      "from": 3840,
+      "ibility": 3841,
+      "出版社": 3842,
+      "▁issue": 3843,
+      "▁crit": 3844,
+      "box": 3845,
+      "ze": 3846,
+      "的规定": 3847,
+      "▁word": 3848,
+      "完成": 3849,
+      "olic": 3850,
+      "▁dat": 3851,
+      "********": 3852,
+      "inary": 3853,
+      "▁energy": 3854,
+      "正确": 3855,
+      "ero": 3856,
+      "▁base": 3857,
+      "▁port": 3858,
+      "▁move": 3859,
+      "ilt": 3860,
+      "une": 3861,
+      "▁previous": 3862,
+      "▁looking": 3863,
+      "asing": 3864,
+      "ender": 3865,
+      "odes": 3866,
+      "▁With": 3867,
+      "▁Jan": 3868,
+      "▁educ": 3869,
+      "▁similar": 3870,
+      "▁below": 3871,
+      "Set": 3872,
+      "Value": 3873,
+      "(\\\\": 3874,
+      "▁relations": 3875,
+      "plied": 3876,
+      "▁respons": 3877,
+      "一起": 3878,
+      "▁mot": 3879,
+      "inst": 3880,
+      "▁why": 3881,
+      "其实": 3882,
+      "▁course": 3883,
+      "icient": 3884,
+      "roller": 3885,
+      "▁fix": 3886,
+      "ores": 3887,
+      "▁County": 3888,
+      "cos": 3889,
+      "mission": 3890,
+      "▁creat": 3891,
+      "价格": 3892,
+      "ipp": 3893,
+      "▁approach": 3894,
+      "▁products": 3895,
+      "▁making": 3896,
+      "Service": 3897,
+      "top": 3898,
+      "代表人": 3899,
+      "四十": 3900,
+      "Im": 3901,
+      "da": 3902,
+      "nExternal": 3903,
+      "不要": 3904,
+      "委员": 3905,
+      "fl": 3906,
+      "ights": 3907,
+      "▁systems": 3908,
+      "支付": 3909,
+      "▁today": 3910,
+      "▁flo": 3911,
+      "▁software": 3912,
+      "▁means": 3913,
+      "rodu": 3914,
+      "▁indust": 3915,
+      "建议": 3916,
+      "main": 3917,
+      "▁config": 3918,
+      "▁dom": 3919,
+      "▁observ": 3920,
+      "CO": 3921,
+      "▁needs": 3922,
+      "▁source": 3923,
+      "Field": 3924,
+      "▁improve": 3925,
+      "tra": 3926,
+      "有效": 3927,
+      "Requ": 3928,
+      "地区": 3929,
+      "imum": 3930,
+      "▁having": 3931,
+      "▁lo": 3932,
+      "▁—": 3933,
+      "gor": 3934,
+      "过程中": 3935,
+      "建筑": 3936,
+      "pass": 3937,
+      "oken": 3938,
+      "Sh": 3939,
+      "pha": 3940,
+      "label": 3941,
+      "▁already": 3942,
+      "▁ther": 3943,
+      "ica": 3944,
+      "▁never": 3945,
+      "▁address": 3946,
+      "trans": 3947,
+      "证据": 3948,
+      "IP": 3949,
+      "pre": 3950,
+      "▁done": 3951,
+      "▁levels": 3952,
+      "oper": 3953,
+      "nb": 3954,
+      "istic": 3955,
+      "utions": 3956,
+      "indows": 3957,
+      "▁job": 3958,
+      "▁members": 3959,
+      "▁offic": 3960,
+      "Int": 3961,
+      "▁nec": 3962,
+      "职业": 3963,
+      "▁whether": 3964,
+      "ague": 3965,
+      "result": 3966,
+      "ribute": 3967,
+      "▁Joh": 3968,
+      "hold": 3969,
+      "dir": 3970,
+      "ino": 3971,
+      "nYou": 3972,
+      "▁Ass": 3973,
+      "一下": 3974,
+      "采用": 3975,
+      "一直": 3976,
+      "▁answer": 3977,
+      "▁circ": 3978,
+      "▁behav": 3979,
+      "ributed": 3980,
+      "▁toget": 3981,
+      "php": 3982,
+      "iation": 3983,
+      "▁together": 3984,
+      "▁document": 3985,
+      "▁death": 3986,
+      "Exception": 3987,
+      "一样": 3988,
+      "ports": 3989,
+      "表现": 3990,
+      "▁expression": 3991,
+      "▁employ": 3992,
+      "操作": 3993,
+      "▁lik": 3994,
+      "▁groups": 3995,
+      "Text": 3996,
+      "行为": 3997,
+      "▁least": 3998,
+      "▁press": 3999,
+      "▁video": 4000,
+      "▁Par": 4001,
+      "▁win": 4002,
+      "alf": 4003,
+      "DE": 4004,
+      "ills": 4005,
+      "▁–\\": 4006,
+      "▁application": 4007,
+      "artment": 4008,
+      "结构": 4009,
+      "进行了": 4010,
+      "▁National": 4011,
+      "based": 4012,
+      "thers": 4013,
+      "▁copy": 4014,
+      "▁Im": 4015,
+      "▁various": 4016,
+      "earch": 4017,
+      "▁experim": 4018,
+      "▁features": 4019,
+      "atter": 4020,
+      "▁often": 4021,
+      "各种": 4022,
+      "▁website": 4023,
+      "▁States": 4024,
+      "ainer": 4025,
+      "▁fail": 4026,
+      "place": 4027,
+      "User": 4028,
+      "▁paramet": 4029,
+      "▁higher": 4030,
+      "▁sequ": 4031,
+      "arrow": 4032,
+      "▁social": 4033,
+      "之间": 4034,
+      "▁vir": 4035,
+      "Object": 4036,
+      "▁//": 4037,
+      "oor": 4038,
+      "了解": 4039,
+      "获得": 4040,
+      "lor": 4041,
+      "/*": 4042,
+      "决定": 4043,
+      "ounc": 4044,
+      "Key": 4045,
+      "▁trad": 4046,
+      "▁cur": 4047,
+      "▁related": 4048,
+      "oo": 4049,
+      "罪犯": 4050,
+      "ummary": 4051,
+      "now": 4052,
+      "item": 4053,
+      "▁cases": 4054,
+      "▁implement": 4055,
+      "bb": 4056,
+      "osp": 4057,
+      "以下": 4058,
+      "▁Aust": 4059,
+      "健康": 4060,
+      "oud": 4061,
+      "▁Fe": 4062,
+      "▁dig": 4063,
+      "▁simple": 4064,
+      "utes": 4065,
+      "自然": 4066,
+      "▁Med": 4067,
+      "ging": 4068,
+      "第一款": 4069,
+      "交通": 4070,
+      "ogen": 4071,
+      "▁across": 4072,
+      "以上": 4073,
+      "法定代表人": 4074,
+      "积极": 4075,
+      "法律效力": 4076,
+      "nWhat": 4077,
+      "▁won": 4078,
+      "egin": 4079,
+      "run": 4080,
+      "▁员": 4081,
+      "Time": 4082,
+      "注意": 4083,
+      "ditions": 4084,
+      "由于": 4085,
+      "▁sum": 4086,
+      "▁Des": 4087,
+      "ags": 4088,
+      "▁None": 4089,
+      "gorith": 4090,
+      "▁flow": 4091,
+      "erc": 4092,
+      "amed": 4093,
+      "amm": 4094,
+      "omic": 4095,
+      "lim": 4096,
+      "▁Reg": 4097,
+      "nV": 4098,
+      "request": 4099,
+      "▁film": 4100,
+      "bar": 4101,
+      "▁Int": 4102,
+      "▁clear": 4103,
+      "▁input": 4104,
+      "Pl": 4105,
+      "blems": 4106,
+      "ches": 4107,
+      "▁past": 4108,
+      "▁()": 4109,
+      "▁cult": 4110,
+      "▁fund": 4111,
+      "学院": 4112,
+      "▁house": 4113,
+      "year": 4114,
+      "stit": 4115,
+      "pen": 4116,
+      "SS": 4117,
+      "▁Min": 4118,
+      "作出": 4119,
+      "▁compon": 4120,
+      "point": 4121,
+      "▁further": 4122,
+      "▁host": 4123,
+      "ison": 4124,
+      "ude": 4125,
+      "~~": 4126,
+      "▁please": 4127,
+      "发生法律效力": 4128,
+      "part": 4129,
+      "plit": 4130,
+      "style": 4131,
+      "▁normal": 4132,
+      "▁increase": 4133,
+      "分钟": 4134,
+      "▁fun": 4135,
+      "aim": 4136,
+      "运动": 4137,
+      "▁When": 4138,
+      "lin": 4139,
+      "万元": 4140,
+      "ips": 4141,
+      "avel": 4142,
+      "xy": 4143,
+      "comp": 4144,
+      "body": 4145,
+      "plication": 4146,
+      "▁points": 4147,
+      "dom": 4148,
+      "个月": 4149,
+      "▁contact": 4150,
+      "▁abs": 4151,
+      "True": 4152,
+      "▁Co": 4153,
+      "▁che": 4154,
+      "▁Jul": 4155,
+      "▁works": 4156,
+      "icult": 4157,
+      "▁dri": 4158,
+      "((": 4159,
+      "dition": 4160,
+      "▁My": 4161,
+      "产业": 4162,
+      "▁factors": 4163,
+      "▁assess": 4164,
+      "ride": 4165,
+      "irm": 4166,
+      "▁South": 4167,
+      "amework": 4168,
+      "#\\": 4169,
+      "价值": 4170,
+      "ving": 4171,
+      "▁months": 4172,
+      "▁subject": 4173,
+      "义务": 4174,
+      "indow": 4175,
+      "▁multiple": 4176,
+      "▁World": 4177,
+      "▁clinical": 4178,
+      "cin": 4179,
+      "▁IN": 4180,
+      "▁deg": 4181,
+      "lege": 4182,
+      "这是": 4183,
+      "▁cancer": 4184,
+      "▁food": 4185,
+      "希望": 4186,
+      "不过": 4187,
+      "▁terms": 4188,
+      "特别": 4189,
+      "atform": 4190,
+      "down": 4191,
+      "▁style": 4192,
+      "用户": 4193,
+      "▁bit": 4194,
+      "患者": 4195,
+      "条第一款": 4196,
+      "▁express": 4197,
+      "▁search": 4198,
+      "▁State": 4199,
+      "▁strong": 4200,
+      "aging": 4201,
+      "组成": 4202,
+      "▁measure": 4203,
+      "予以": 4204,
+      "IV": 4205,
+      "########": 4206,
+      "));\\": 4207,
+      "▁fore": 4208,
+      "机构": 4209,
+      "std": 4210,
+      "group": 4211,
+      "起来": 4212,
+      "astic": 4213,
+      "src": 4214,
+      "▁larg": 4215,
+      "imate": 4216,
+      "actions": 4217,
+      "verse": 4218,
+      "▁Am": 4219,
+      "We": 4220,
+      "fect": 4221,
+      "onstr": 4222,
+      "df": 4223,
+      "▁diff": 4224,
+      "▁conditions": 4225,
+      "这一": 4226,
+      "实际": 4227,
+      "ilit": 4228,
+      "效果": 4229,
+      "ector": 4230,
+      "▁Gener": 4231,
+      "irl": 4232,
+      "▁School": 4233,
+      "nint": 4234,
+      "gress": 4235,
+      "perty": 4236,
+      "ford": 4237,
+      "▁Euro": 4238,
+      "▁mass": 4239,
+      "ython": 4240,
+      "历史": 4241,
+      "手机": 4242,
+      "▁future": 4243,
+      "\\']": 4244,
+      "▁methods": 4245,
+      "综合": 4246,
+      "eful": 4247,
+      "Log": 4248,
+      "是否": 4249,
+      "ama": 4250,
+      "▁particip": 4251,
+      "State": 4252,
+      "CH": 4253,
+      "▁Gu": 4254,
+      "为什么": 4255,
+      "PI": 4256,
+      "▁far": 4257,
+      "保护": 4258,
+      "joy": 4259,
+      "▁<<": 4260,
+      "apy": 4261,
+      "ada": 4262,
+      "▁music": 4263,
+      "住所": 4264,
+      "酒店": 4265,
+      "级人民法院": 4266,
+      "This": 4267,
+      "Get": 4268,
+      "ana": 4269,
+      "empt": 4270,
+      "▁protein": 4271,
+      "iate": 4272,
+      "base": 4273,
+      "多少": 4274,
+      "▁&&": 4275,
+      "▁doesn": 4276,
+      "▁stre": 4277,
+      "▁kind": 4278,
+      "▁load": 4279,
+      "▁increased": 4280,
+      "nFor": 4281,
+      "老师": 4282,
+      "abase": 4283,
+      "UL": 4284,
+      "▁Qu": 4285,
+      "ynam": 4286,
+      "▁compared": 4287,
+      "状态": 4288,
+      "▁message": 4289,
+      "▁miss": 4290,
+      "▁continu": 4291,
+      "只有": 4292,
+      "▁print": 4293,
+      "得到": 4294,
+      "length": 4295,
+      "though": 4296,
+      "集团": 4297,
+      "cho": 4298,
+      "oute": 4299,
+      "nK": 4300,
+      "▁easy": 4301,
+      "▁table": 4302,
+      "▁Oct": 4303,
+      "isc": 4304,
+      "▁album": 4305,
+      "board": 4306,
+      "因此": 4307,
+      "▁aim": 4308,
+      "▁position": 4309,
+      "▁strateg": 4310,
+      "object": 4311,
+      "▁makes": 4312,
+      "image": 4313,
+      "leased": 4314,
+      "atural": 4315,
+      "create": 4316,
+      "▁visit": 4317,
+      "▁benef": 4318,
+      "rior": 4319,
+      "atal": 4320,
+      "link": 4321,
+      "▁Apr": 4322,
+      "eta": 4323,
+      "ication": 4324,
+      "sec": 4325,
+      "nusername": 4326,
+      "销售": 4327,
+      "▁training": 4328,
+      "看到": 4329,
+      "▁lower": 4330,
+      "ependent": 4331,
+      "▁coll": 4332,
+      "CL": 4333,
+      "▁cy": 4334,
+      "感觉": 4335,
+      "▁os": 4336,
+      "▁learn": 4337,
+      "▁ach": 4338,
+      "▁polit": 4339,
+      "eger": 4340,
+      "▁city": 4341,
+      "era": 4342,
+      "lood": 4343,
+      "anced": 4344,
+      "不断": 4345,
+      "build": 4346,
+      "ilities": 4347,
+      ">>": 4348,
+      "▁limit": 4349,
+      "▁color": 4350,
+      "年的": 4351,
+      "日内": 4352,
+      "依据": 4353,
+      "aken": 4354,
+      "▁become": 4355,
+      "ibrary": 4356,
+      "bit": 4357,
+      "利息": 4358,
+      "etic": 4359,
+      "股份": 4360,
+      "原因": 4361,
+      "十二": 4362,
+      "▁production": 4363,
+      "ersion": 4364,
+      "应当": 4365,
+      "劳动": 4366,
+      "oman": 4367,
+      "▁via": 4368,
+      "▁track": 4369,
+      "部门": 4370,
+      "ategory": 4371,
+      "ounds": 4372,
+      "面的": 4373,
+      "$$\\": 4374,
+      "enn": 4375,
+      "annel": 4376,
+      "core": 4377,
+      "plac": 4378,
+      "负责": 4379,
+      "){\\": 4380,
+      "bre": 4381,
+      "▁relationship": 4382,
+      "nce": 4383,
+      "创新": 4384,
+      "▁squ": 4385,
+      "▁Intern": 4386,
+      "▁structure": 4387,
+      "orage": 4388,
+      "AM": 4389,
+      "▁Me": 4390,
+      "▁frequ": 4391,
+      "ires": 4392,
+      "oll": 4393,
+      "▁Go": 4394,
+      "▁Ge": 4395,
+      "南省": 4396,
+      "计划": 4397,
+      "Config": 4398,
+      "▁added": 4399,
+      "▁II": 4400,
+      "▁correct": 4401,
+      "虽然": 4402,
+      "▁作者": 4403,
+      "▁cover": 4404,
+      "▁Add": 4405,
+      "▁Rec": 4406,
+      "解释": 4407,
+      "ledge": 4408,
+      "stars": 4409,
+      "eral": 4410,
+      "住所地": 4411,
+      "▁events": 4412,
+      "nswer": 4413,
+      "width": 4414,
+      "▁immun": 4415,
+      "ops": 4416,
+      "default": 4417,
+      "$,": 4418,
+      "▁demonstr": 4419,
+      "▁hy": 4420,
+      "▁problems": 4421,
+      "omy": 4422,
+      "▁growth": 4423,
+      "有关": 4424,
+      "▁连载": 4425,
+      "▁red": 4426,
+      "地方": 4427,
+      "▁Rep": 4428,
+      "▁received": 4429,
+      "ding": 4430,
+      "ament": 4431,
+      "▁positive": 4432,
+      "的话": 4433,
+      "begin": 4434,
+      "▁York": 4435,
+      "tpublic": 4436,
+      "Re": 4437,
+      "安装": 4438,
+      "▁java": 4439,
+      "▁That": 4440,
+      "成功": 4441,
+      "iles": 4442,
+      "spec": 4443,
+      "▁feature": 4444,
+      "▁language": 4445,
+      "IR": 4446,
+      "回答": 4447,
+      "受理费": 4448,
+      "▁Dr": 4449,
+      "▁country": 4450,
+      "▁length": 4451,
+      "It": 4452,
+      "iddle": 4453,
+      "▁thought": 4454,
+      "▁factor": 4455,
+      "▁hom": 4456,
+      "package": 4457,
+      "ENT": 4458,
+      "医院": 4459,
+      "▁admin": 4460,
+      "▁away": 4461,
+      "水平": 4462,
+      "也不": 4463,
+      "▁hyp": 4464,
+      "本次": 4465,
+      "\\\\(\\\\": 4466,
+      "post": 4467,
+      "Un": 4468,
+      "▁Part": 4469,
+      "▁population": 4470,
+      "nTo": 4471,
+      "▁refer": 4472,
+      "▁neg": 4473,
+      "▁block": 4474,
+      "▁output": 4475,
+      "istics": 4476,
+      "▁proposed": 4477,
+      "▁management": 4478,
+      "most": 4479,
+      "que": 4480,
+      "消费": 4481,
+      "▁questions": 4482,
+      "医学": 4483,
+      "▁drug": 4484,
+      "Request": 4485,
+      "▁third": 4486,
+      "▁five": 4487,
+      "艺术": 4488,
+      "必须": 4489,
+      "array": 4490,
+      "uments": 4491,
+      "eter": 4492,
+      "▁web": 4493,
+      "国内": 4494,
+      "分别": 4495,
+      "▁option": 4496,
+      "▁Do": 4497,
+      "sk": 4498,
+      "sole": 4499,
+      "之一": 4500,
+      "▁One": 4501,
+      "▁amount": 4502,
+      "来说": 4503,
+      "▁certain": 4504,
+      "ites": 4505,
+      "只是": 4506,
+      "方程": 4507,
+      "十一": 4508,
+      "open": 4509,
+      "▁provides": 4510,
+      "▁created": 4511,
+      "▁Aug": 4512,
+      "▁North": 4513,
+      "▁hold": 4514,
+      "▁Christ": 4515,
+      "▁song": 4516,
+      "请求": 4517,
+      "▁List": 4518,
+      "▁located": 4519,
+      "repon": 4520,
+      "▁determine": 4521,
+      "▁mor": 4522,
+      "▁included": 4523,
+      "AG": 4524,
+      "▁developed": 4525,
+      "电子": 4526,
+      "▁author": 4527,
+      "▁land": 4528,
+      "format": 4529,
+      "stract": 4530,
+      "▁)": 4531,
+      "▁ang": 4532,
+      "portun": 4533,
+      "reponame": 4534,
+      "ners": 4535,
+      "▁history": 4536,
+      "elling": 4537,
+      "▁young": 4538,
+      "▁rat": 4539,
+      "▁chall": 4540,
+      "Path": 4541,
+      "uses": 4542,
+      "▁Cal": 4543,
+      "▁complete": 4544,
+      "他的": 4545,
+      "▁mole": 4546,
+      "▁month": 4547,
+      "press": 4548,
+      "itory": 4549,
+      "▁predict": 4550,
+      "umb": 4551,
+      "izing": 4552,
+      "mail": 4553,
+      "▁calcul": 4554,
+      "npublic": 4555,
+      "doc": 4556,
+      "ried": 4557,
+      "撤诉": 4558,
+      "生效": 4559,
+      "unch": 4560,
+      "▁thing": 4561,
+      "精神": 4562,
+      "aling": 4563,
+      "node": 4564,
+      "▁ens": 4565,
+      "▁money": 4566,
+      "UN": 4567,
+      "itions": 4568,
+      "▁index": 4569,
+      "Or": 4570,
+      "▁options": 4571,
+      "PS": 4572,
+      "结合": 4573,
+      "▁sat": 4574,
+      "一步": 4575,
+      "LO": 4576,
+      "▁valid": 4577,
+      "中级人民法院": 4578,
+      "▁provided": 4579,
+      "sequ": 4580,
+      "▁issues": 4581,
+      "汽车": 4582,
+      "section": 4583,
+      "无法": 4584,
+      "▁once": 4585,
+      "山东": 4586,
+      "ula": 4587,
+      "▁Ab": 4588,
+      "▁OF": 4589,
+      "icles": 4590,
+      "▁Pol": 4591,
+      "MS": 4592,
+      "▁[\\": 4593,
+      "istance": 4594,
+      "By": 4595,
+      "▁$$\\\\": 4596,
+      "▁camp": 4597,
+      "▁City": 4598,
+      "▁property": 4599,
+      "ike": 4600,
+      "▁impact": 4601,
+      "▁started": 4602,
+      "Add": 4603,
+      "alt": 4604,
+      "▁building": 4605,
+      "itional": 4606,
+      "增加": 4607,
+      "hest": 4608,
+      "[]": 4609,
+      "▁original": 4610,
+      "▁Not": 4611,
+      "urch": 4612,
+      "info": 4613,
+      "这里": 4614,
+      "angle": 4615,
+      "irection": 4616,
+      "ui": 4617,
+      "双方": 4618,
+      "amples": 4619,
+      "itted": 4620,
+      "▁Europe": 4621,
+      "ered": 4622,
+      "ults": 4623,
+      "script": 4624,
+      "▁parent": 4625,
+      "▁ver": 4626,
+      "模式": 4627,
+      "ram": 4628,
+      "▁percent": 4629,
+      "▁context": 4630,
+      "判决书": 4631,
+      "▁mag": 4632,
+      "____": 4633,
+      "message": 4634,
+      "iction": 4635,
+      "▁teach": 4636,
+      "▁Ser": 4637,
+      "▁q": 4638,
+      "anies": 4639,
+      "▁super": 4640,
+      "▁files": 4641,
+      "ango": 4642,
+      "active": 4643,
+      "目标": 4644,
+      "政策": 4645,
+      "display": 4646,
+      "▁others": 4647,
+      "▁sex": 4648,
+      "▁technology": 4649,
+      "oding": 4650,
+      "▁Art": 4651,
+      "文件": 4652,
+      "icon": 4653,
+      "`\\": 4654,
+      "▁release": 4655,
+      "▁mix": 4656,
+      "▁match": 4657,
+      "$\\\\": 4658,
+      "▁areas": 4659,
+      "可供": 4660,
+      "atest": 4661,
+      "▁random": 4662,
+      "▁hot": 4663,
+      "olve": 4664,
+      "系列": 4665,
+      "[\"": 4666,
+      "▁CO": 4667,
+      "客户": 4668,
+      "while": 4669,
+      "软件": 4670,
+      "▁div": 4671,
+      "我国": 4672,
+      "▁opportun": 4673,
+      "这么": 4674,
+      "继续": 4675,
+      "lements": 4676,
+      "▁functions": 4677,
+      "|\\": 4678,
+      "▁hours": 4679,
+      "▁yet": 4680,
+      "uation": 4681,
+      "ulated": 4682,
+      "▁grad": 4683,
+      "Form": 4684,
+      "▁currently": 4685,
+      "anch": 4686,
+      "osph": 4687,
+      "sum": 4688,
+      "ensive": 4689,
+      "▁species": 4690,
+      "▁engine": 4691,
+      "img": 4692,
+      "▁government": 4693,
+      "▁article": 4694,
+      "iff": 4695,
+      "▁live": 4696,
+      "▁types": 4697,
+      "工业": 4698,
+      "avor": 4699,
+      "位置": 4700,
+      "▁reason": 4701,
+      "▁member": 4702,
+      "▁pick": 4703,
+      "Map": 4704,
+      "进入": 4705,
+      "▁Stud": 4706,
+      "▁enough": 4707,
+      "ayout": 4708,
+      "▁regard": 4709,
+      "mark": 4710,
+      "▁date": 4711,
+      "Ser": 4712,
+      "▁clos": 4713,
+      "▁exec": 4714,
+      "}}{": 4715,
+      "▁Austral": 4716,
+      "▁getting": 4717,
+      "▁though": 4718,
+      "▁After": 4719,
+      "conn": 4720,
+      "')\\": 4721,
+      "农村": 4722,
+      "▁price": 4723,
+      "inding": 4724,
+      "expect": 4725,
+      "room": 4726,
+      "urg": 4727,
+      "▁comes": 4728,
+      "pat": 4729,
+      "})\\": 4730,
+      "urb": 4731,
+      "▁numbers": 4732,
+      "rm": 4733,
+      "出来": 4734,
+      "案件受理费": 4735,
+      "opyright": 4736,
+      "▁room": 4737,
+      "ML": 4738,
+      "▁enjoy": 4739,
+      "obile": 4740,
+      "简单": 4741,
+      "▁designed": 4742,
+      "▁inhib": 4743,
+      "证明": 4744,
+      "▁six": 4745,
+      "股份有限公司": 4746,
+      "▁quant": 4747,
+      "自动": 4748,
+      "');\\": 4749,
+      "nLet": 4750,
+      "电影": 4751,
+      "~\\": 4752,
+      "准许": 4753,
+      "▁seen": 4754,
+      "▁personal": 4755,
+      "▁doing": 4756,
+      "▁Sec": 4757,
+      "▁sw": 4758,
+      "▁minutes": 4759,
+      "▁autom": 4760,
+      "where": 4761,
+      "▁pattern": 4762,
+      "ming": 4763,
+      "umber": 4764,
+      "▁quick": 4765,
+      "▁necess": 4766,
+      "ese": 4767,
+      "▁OR": 4768,
+      "${": 4769,
+      "按照": 4770,
+      "Inf": 4771,
+      "▁properties": 4772,
+      "ony": 4773,
+      "reate": 4774,
+      "产生": 4775,
+      "更多": 4776,
+      "▁graph": 4777,
+      "dev": 4778,
+      "▁actually": 4779,
+      "▁close": 4780,
+      "\\';\\": 4781,
+      "完全": 4782,
+      "通知": 4783,
+      "▁influ": 4784,
+      "▁unique": 4785,
+      "浙江": 4786,
+      "{\"": 4787,
+      "▁million": 4788,
+      "▁bar": 4789,
+      "▁initial": 4790,
+      "soft": 4791,
+      "▁surface": 4792,
+      "▁mom": 4793,
+      "▁perfect": 4794,
+      "istry": 4795,
+      "ges": 4796,
+      "sing": 4797,
+      "▁introdu": 4798,
+      "▁diagn": 4799,
+      "▁John": 4800,
+      "▁clean": 4801,
+      "CC": 4802,
+      "▁cut": 4803,
+      "▁std": 4804,
+      "gn": 4805,
+      "▁heart": 4806,
+      "▁users": 4807,
+      "新的": 4808,
+      "之前": 4809,
+      "▁decre": 4810,
+      "▁details": 4811,
+      "nAs": 4812,
+      "▁Per": 4813,
+      "host": 4814,
+      "▁buy": 4815,
+      "▁High": 4816,
+      "PO": 4817,
+      "▁separ": 4818,
+      "nm": 4819,
+      "领导": 4820,
+      "位于": 4821,
+      "Al": 4822,
+      "上述": 4823,
+      "inally": 4824,
+      "oma": 4825,
+      "东西": 4826,
+      "web": 4827,
+      "容易": 4828,
+      "ius": 4829,
+      "▁econ": 4830,
+      "经过": 4831,
+      "▁shows": 4832,
+      "共同": 4833,
+      "平方": 4834,
+      "resh": 4835,
+      "▁comb": 4836,
+      "业务": 4837,
+      "日本": 4838,
+      "▁arch": 4839,
+      "▁novel": 4840,
+      "PE": 4841,
+      "nSt": 4842,
+      "ises": 4843,
+      "▁beaut": 4844,
+      "▁tum": 4845,
+      "▁intern": 4846,
+      "▁begin": 4847,
+      "理论": 4848,
+      "开展": 4849,
+      "信用": 4850,
+      "Item": 4851,
+      "▁title": 4852,
+      "委托代理人": 4853,
+      "▁break": 4854,
+      "table": 4855,
+      "ucle": 4856,
+      "▁blood": 4857,
+      "▁concent": 4858,
+      "传统": 4859,
+      "hand": 4860,
+      "之日起": 4861,
+      "最高人民法院": 4862,
+      "网站": 4863,
+      "env": 4864,
+      "处罚": 4865,
+      "Mod": 4866,
+      "▁games": 4867,
+      "cast": 4868,
+      "▁took": 4869,
+      "onal": 4870,
+      "▁conduct": 4871,
+      "▁loss": 4872,
+      "的大": 4873,
+      "valid": 4874,
+      "家庭": 4875,
+      "▁reported": 4876,
+      "月二十": 4877,
+      "具体": 4878,
+      "ugin": 4879,
+      "istration": 4880,
+      "▁polic": 4881,
+      "委员会": 4882,
+      "ydro": 4883,
+      "ached": 4884,
+      "nWh": 4885,
+      "领域": 4886,
+      "▁recomm": 4887,
+      "▁distributed": 4888,
+      "append": 4889,
+      "light": 4890,
+      "意义": 4891,
+      "Context": 4892,
+      "BC": 4893,
+      "▁successful": 4894,
+      "encies": 4895,
+      "县人民法院": 4896,
+      "isms": 4897,
+      "▁extends": 4898,
+      "ube": 4899,
+      "idge": 4900,
+      "检察": 4901,
+      "client": 4902,
+      "有一": 4903,
+      "撤回": 4904,
+      "▁prevent": 4905,
+      "▁needed": 4906,
+      "▁Cor": 4907,
+      "▁significantly": 4908,
+      "▁showed": 4909,
+      "政治": 4910,
+      "inate": 4911,
+      "▁temper": 4912,
+      "▁released": 4913,
+      "▁En": 4914,
+      "形成": 4915,
+      "▁region": 4916,
+      "ged": 4917,
+      "目的": 4918,
+      "ief": 4919,
+      "的事": 4920,
+      "module": 4921,
+      "nown": 4922,
+      "▁media": 4923,
+      "▁screen": 4924,
+      "oring": 4925,
+      "jango": 4926,
+      "ync": 4927,
+      "▁equation": 4928,
+      "aid": 4929,
+      "▁email": 4930,
+      "▁according": 4931,
+      "主任": 4932,
+      "▁says": 4933,
+      "prot": 4934,
+      "▁industry": 4935,
+      "aur": 4936,
+      "▁collection": 4937,
+      "rest": 4938,
+      "ros": 4939,
+      "adem": 4940,
+      "现代": 4941,
+      "bt": 4942,
+      "▁however": 4943,
+      "▁knowledge": 4944,
+      "▁ess": 4945,
+      "nThere": 4946,
+      "aches": 4947,
+      "▁former": 4948,
+      "图书": 4949,
+      "lected": 4950,
+      "ring": 4951,
+      "gorithm": 4952,
+      "▁share": 4953,
+      "illed": 4954,
+      "▁professional": 4955,
+      "▁played": 4956,
+      "▁chem": 4957,
+      "dat": 4958,
+      "▁learning": 4959,
+      "actory": 4960,
+      "承担": 4961,
+      "需求": 4962,
+      "▁ever": 4963,
+      "▁recent": 4964,
+      "电视": 4965,
+      "ampions": 4966,
+      "status": 4967,
+      "当事": 4968,
+      "负担": 4969,
+      "%)": 4970,
+      "显示": 4971,
+      "▁||": 4972,
+      "▁Dist": 4973,
+      "▁night": 4974,
+      "▁stop": 4975,
+      "igned": 4976,
+      "rew": 4977,
+      "▁Get": 4978,
+      "anager": 4979,
+      "努力": 4980,
+      "▁THE": 4981,
+      "属于": 4982,
+      "▁talk": 4983,
+      "underline": 4984,
+      "可供执行": 4985,
+      "▁March": 4986,
+      "会议": 4987,
+      "▁career": 4988,
+      "Ph": 4989,
+      "nHe": 4990,
+      "教师": 4991,
+      "今天": 4992,
+      "不到": 4993,
+      "rows": 4994,
+      "接受": 4995,
+      "Event": 4996,
+      "try": 4997,
+      "▁patient": 4998,
+      "▁agre": 4999,
+      "nuse": 5000,
+      "▁mut": 5001,
+      "数的": 5002,
+      "change": 5003,
+      "任何": 5004,
+      "小时": 5005,
+      "iol": 5006,
+      "▁built": 5007,
+      "pring": 5008,
+      "▁idea": 5009,
+      "公里": 5010,
+      "▁writing": 5011,
+      "人民币": 5012,
+      "flow": 5013,
+      "▁published": 5014,
+      "帮助": 5015,
+      ",”": 5016,
+      "['": 5017,
+      "提升": 5018,
+      "收入": 5019,
+      "导致": 5020,
+      "最大": 5021,
+      "▁shown": 5022,
+      "▁story": 5023,
+      "登记": 5024,
+      "iated": 5025,
+      "ects": 5026,
+      "TE": 5027,
+      "确定": 5028,
+      "position": 5029,
+      "ha": 5030,
+      "▁goal": 5031,
+      "ched": 5032,
+      "olean": 5033,
+      "ronic": 5034,
+      "有的": 5035,
+      "CE": 5036,
+      "▁array": 5037,
+      "毕业": 5038,
+      "▁cross": 5039,
+      "imens": 5040,
+      "lying": 5041,
+      "▁pack": 5042,
+      "故事": 5043,
+      "▁Brit": 5044,
+      "apan": 5045,
+      "▁half": 5046,
+      "▁update": 5047,
+      "])": 5048,
+      "▁ir": 5049,
+      "▁expected": 5050,
+      "eed": 5051,
+      "普通": 5052,
+      "HT": 5053,
+      "hesis": 5054,
+      "▁pict": 5055,
+      "born": 5056,
+      "query": 5057,
+      "▁whole": 5058,
+      "▁companies": 5059,
+      "▁inj": 5060,
+      "comes": 5061,
+      "AA": 5062,
+      "static": 5063,
+      "▁primary": 5064,
+      "▁write": 5065,
+      "重庆": 5066,
+      "iday": 5067,
+      "▁natural": 5068,
+      "CR": 5069,
+      "Class": 5070,
+      "任务": 5071,
+      "▁spect": 5072,
+      "ele": 5073,
+      "▁recogn": 5074,
+      "▁effective": 5075,
+      "▁map": 5076,
+      "\\\\\\\\": 5077,
+      "illi": 5078,
+      "甚至": 5079,
+      "ocol": 5080,
+      "▁By": 5081,
+      "▁described": 5082,
+      "送达后": 5083,
+      "SUB": 5084,
+      "▁href": 5085,
+      "▁status": 5086,
+      "thern": 5087,
+      "▁Comp": 5088,
+      "▁average": 5089,
+      "▁April": 5090,
+      "▁tw": 5091,
+      "bol": 5092,
+      "空间": 5093,
+      "do": 5094,
+      "term": 5095,
+      "与被告": 5096,
+      "▁Sim": 5097,
+      "ANT": 5098,
+      "server": 5099,
+      "▁Off": 5100,
+      "▁Germ": 5101,
+      "author": 5102,
+      "▁▁▁": 5103,
+      "iven": 5104,
+      ")(": 5105,
+      "▁ey": 5106,
+      "nthe": 5107,
+      "▁stat": 5108,
+      "▁Test": 5109,
+      "__(": 5110,
+      "▁additional": 5111,
+      "本判决": 5112,
+      "iful": 5113,
+      "▁appear": 5114,
+      "ln": 5115,
+      "mand": 5116,
+      "▁black": 5117,
+      "▁indu": 5118,
+      "▁condition": 5119,
+      "▁births": 5120,
+      "刑事": 5121,
+      "▁ago": 5122,
+      "百科": 5123,
+      "▁center": 5124,
+      "ades": 5125,
+      "());\\": 5126,
+      "▁seems": 5127,
+      "▁Dep": 5128,
+      "▁dam": 5129,
+      "uz": 5130,
+      "lick": 5131,
+      "tx": 5132,
+      "金融": 5133,
+      "span": 5134,
+      "旅游": 5135,
+      "cs": 5136,
+      "IM": 5137,
+      "思想": 5138,
+      "\"\"\"": 5139,
+      "▁fem": 5140,
+      "ided": 5141,
+      "unt": 5142,
+      "▁January": 5143,
+      "▁tour": 5144,
+      "▁concer": 5145,
+      "推荐": 5146,
+      "▁running": 5147,
+      "上诉人": 5148,
+      "变化": 5149,
+      "文章": 5150,
+      "▁mach": 5151,
+      "sm": 5152,
+      "aries": 5153,
+      "pack": 5154,
+      "▁likely": 5155,
+      "▁weight": 5156,
+      "▁pers": 5157,
+      "▁sent": 5158,
+      "▁Sept": 5159,
+      "▁depend": 5160,
+      "▁Bar": 5161,
+      "Over": 5162,
+      "培训": 5163,
+      "Node": 5164,
+      "保险": 5165,
+      "▁command": 5166,
+      "日向本院": 5167,
+      "四川": 5168,
+      "农业": 5169,
+      "rown": 5170,
+      "times": 5171,
+      "output": 5172,
+      "▁Man": 5173,
+      "▁didn": 5174,
+      "nSo": 5175,
+      "▁display": 5176,
+      "▁established": 5177,
+      "▁distribution": 5178,
+      "nHow": 5179,
+      "关键": 5180,
+      "ospital": 5181,
+      "▁gene": 5182,
+      "ensity": 5183,
+      "just": 5184,
+      "nSee": 5185,
+      "For": 5186,
+      "case": 5187,
+      "nAl": 5188,
+      "▁fig": 5189,
+      "▁behavior": 5190,
+      "××": 5191,
+      "pecially": 5192,
+      "eks": 5193,
+      "▁June": 5194,
+      "like": 5195,
+      "local": 5196,
+      "cal": 5197,
+      "时代": 5198,
+      "▁etc": 5199,
+      "Ind": 5200,
+      "▁ge": 5201,
+      "ille": 5202,
+      "▁consist": 5203,
+      "ournal": 5204,
+      "明显": 5205,
+      "▁includes": 5206,
+      "▁identify": 5207,
+      "随着": 5208,
+      "是一个": 5209,
+      "▁reve": 5210,
+      "aph": 5211,
+      "OD": 5212,
+      "不服": 5213,
+      "视频": 5214,
+      "▁App": 5215,
+      "Message": 5216,
+      "fort": 5217,
+      "▁deal": 5218,
+      "▁trying": 5219,
+      "xim": 5220,
+      "▁enh": 5221,
+      "的问题": 5222,
+      "png": 5223,
+      "▁front": 5224,
+      "kg": 5225,
+      "Ent": 5226,
+      "▁instead": 5227,
+      "经验": 5228,
+      "▁held": 5229,
+      "▁War": 5230,
+      "vis": 5231,
+      "▁involved": 5232,
+      "事务": 5233,
+      "▁exact": 5234,
+      "icrosoft": 5235,
+      "江苏": 5236,
+      "nCh": 5237,
+      "▁item": 5238,
+      "aker": 5239,
+      "description": 5240,
+      "▁wr": 5241,
+      "音乐": 5242,
+      "▁bring": 5243,
+      "当事人": 5244,
+      "▁Mon": 5245,
+      "gy": 5246,
+      "▁came": 5247,
+      "▁accept": 5248,
+      "Ad": 5249,
+      "▁aud": 5250,
+      "typ": 5251,
+      "▁util": 5252,
+      "▁taking": 5253,
+      "重点": 5254,
+      "关注": 5255,
+      "▁aw": 5256,
+      "ining": 5257,
+      "istory": 5258,
+      "性的": 5259,
+      "eople": 5260,
+      "la": 5261,
+      "ECT": 5262,
+      "条之规定": 5263,
+      "summary": 5264,
+      "▁ill": 5265,
+      "费用": 5266,
+      "一点": 5267,
+      "aged": 5268,
+      "respon": 5269,
+      "▁Mat": 5270,
+      "▁effort": 5271,
+      "Fr": 5272,
+      "▁offers": 5273,
+      "距离": 5274,
+      "▁happen": 5275,
+      "current": 5276,
+      "▁taken": 5277,
+      "▁El": 5278,
+      "apter": 5279,
+      "▁Mc": 5280,
+      "▁feed": 5281,
+      "Se": 5282,
+      "cing": 5283,
+      "▁Trans": 5284,
+      "ovember": 5285,
+      "igr": 5286,
+      "正常": 5287,
+      "project": 5288,
+      "达到": 5289,
+      "Info": 5290,
+      "](/": 5291,
+      "witch": 5292,
+      "CD": 5293,
+      "▁multi": 5294,
+      "的小": 5295,
+      "▁General": 5296,
+      "▁combin": 5297,
+      "▁ra": 5298,
+      "▁lines": 5299,
+      "rad": 5300,
+      "context": 5301,
+      "unk": 5302,
+      "airs": 5303,
+      "oles": 5304,
+      "(),": 5305,
+      "▁someone": 5306,
+      "▁West": 5307,
+      "制作": 5308,
+      "▁useful": 5309,
+      "TR": 5310,
+      "▁Sm": 5311,
+      "nov": 5312,
+      "pri": 5313,
+      "All": 5314,
+      "}_{": 5315,
+      "ios": 5316,
+      "▁记": 5317,
+      "():\\": 5318,
+      "▁tell": 5319,
+      "商品": 5320,
+      "▁difficult": 5321,
+      "oe": 5322,
+      "▁understanding": 5323,
+      "广东": 5324,
+      "Hand": 5325,
+      "调查": 5326,
+      "法律规定": 5327,
+      "och": 5328,
+      "▁box": 5329,
+      "▁element": 5330,
+      "▁label": 5331,
+      "find": 5332,
+      "▁became": 5333,
+      "▁Afr": 5334,
+      "▁written": 5335,
+      "号民事": 5336,
+      "alpha": 5337,
+      "▁consum": 5338,
+      "▁compl": 5339,
+      "icip": 5340,
+      "}$\\": 5341,
+      "Equ": 5342,
+      "otes": 5343,
+      "uid": 5344,
+      "▁experi": 5345,
+      ")\\\\": 5346,
+      "▁entire": 5347,
+      "▁white": 5348,
+      "权利": 5349,
+      "方案": 5350,
+      "ouch": 5351,
+      "主义": 5352,
+      "▁national": 5353,
+      "lation": 5354,
+      "不仅": 5355,
+      "page": 5356,
+      "▁Inter": 5357,
+      "font": 5358,
+      "iting": 5359,
+      "▁performed": 5360,
+      "aint": 5361,
+      "▁np": 5362,
+      "▁upd": 5363,
+      "▁mind": 5364,
+      "你们": 5365,
+      "商业": 5366,
+      "公开": 5367,
+      "理由": 5368,
+      "▁insp": 5369,
+      "▁double": 5370,
+      "立案": 5371,
+      "▁prior": 5372,
+      "以后": 5373,
+      "▁其": 5374,
+      "▁estim": 5375,
+      "alls": 5376,
+      "▁Mark": 5377,
+      "▁\\\\\\\\": 5378,
+      "check": 5379,
+      "bo": 5380,
+      "▁famil": 5381,
+      "oved": 5382,
+      "cember": 5383,
+      "安徽": 5384,
+      "特点": 5385,
+      "▁July": 5386,
+      "发布": 5387,
+      "Date": 5388,
+      "▁connect": 5389,
+      "rip": 5390,
+      "一种": 5391,
+      "代码": 5392,
+      "检查": 5393,
+      "▁purch": 5394,
+      "文学": 5395,
+      "enced": 5396,
+      "aves": 5397,
+      "ham": 5398,
+      "Up": 5399,
+      "related": 5400,
+      "▁active": 5401,
+      "uel": 5402,
+      "离婚": 5403,
+      "▁October": 5404,
+      "▁education": 5405,
+      "造成": 5406,
+      "▁行": 5407,
+      "App": 5408,
+      "▁especially": 5409,
+      "atives": 5410,
+      "▁und": 5411,
+      "ufact": 5412,
+      "实验": 5413,
+      "Is": 5414,
+      "Code": 5415,
+      "▁suff": 5416,
+      "▁news": 5417,
+      "ani": 5418,
+      "TP": 5419,
+      "减刑": 5420,
+      "▁den": 5421,
+      "ottom": 5422,
+      "拥有": 5423,
+      "▁club": 5424,
+      "▁sem": 5425,
+      "▁replac": 5426,
+      "名称": 5427,
+      "▁medical": 5428,
+      "▁Red": 5429,
+      "aily": 5430,
+      "▁cannot": 5431,
+      "▁Park": 5432,
+      "▁Ac": 5433,
+      "upt": 5434,
+      "▁Act": 5435,
+      "▁claim": 5436,
+      "语言": 5437,
+      "现场": 5438,
+      "▁tax": 5439,
+      "卫生": 5440,
+      "miss": 5441,
+      "▁saf": 5442,
+      "▁expect": 5443,
+      "▁August": 5444,
+      "rypt": 5445,
+      "▁focus": 5446,
+      "len": 5447,
+      "▁pret": 5448,
+      "▁dest": 5449,
+      "alid": 5450,
+      "▁regular": 5451,
+      "▁fast": 5452,
+      "aled": 5453,
+      "▁Center": 5454,
+      "▁films": 5455,
+      "OL": 5456,
+      "▁September": 5457,
+      "ees": 5458,
+      "write": 5459,
+      "▁location": 5460,
+      "▁(\\": 5461,
+      "▁recept": 5462,
+      "▁round": 5463,
+      "▁difference": 5464,
+      "process": 5465,
+      "▁cause": 5466,
+      "最高人民法院关于": 5467,
+      "files": 5468,
+      "▁speed": 5469,
+      "▁private": 5470,
+      "经理": 5471,
+      "join": 5472,
+      "ancial": 5473,
+      "物业": 5474,
+      "aven": 5475,
+      "▁tests": 5476,
+      "hemat": 5477,
+      "形式": 5478,
+      "align": 5479,
+      "ceed": 5480,
+      "▁platform": 5481,
+      "bor": 5482,
+      "ards": 5483,
+      "▁practice": 5484,
+      "▁bound": 5485,
+      "▁client": 5486,
+      "▁observed": 5487,
+      "比如": 5488,
+      "▁probably": 5489,
+      "十五": 5490,
+      "保证": 5491,
+      "对方": 5492,
+      "roph": 5493,
+      "▁images": 5494,
+      "▁office": 5495,
+      "andid": 5496,
+      "▁contract": 5497,
+      "▁player": 5498,
+      "▁rather": 5499,
+      "▁considered": 5500,
+      "rial": 5501,
+      "▁living": 5502,
+      "efore": 5503,
+      "otal": 5504,
+      "▁root": 5505,
+      "▁weeks": 5506,
+      "▁vill": 5507,
+      "方向": 5508,
+      "▁unit": 5509,
+      "▁band": 5510,
+      "今年": 5511,
+      "▁activities": 5512,
+      "▁hope": 5513,
+      "本院于": 5514,
+      "全部": 5515,
+      "iet": 5516,
+      "vey": 5517,
+      "▁English": 5518,
+      "ailed": 5519,
+      "leep": 5520,
+      "nif": 5521,
+      "call": 5522,
+      "next": 5523,
+      "En": 5524,
+      "▁fit": 5525,
+      "Build": 5526,
+      "▁defined": 5527,
+      "BA": 5528,
+      "▁went": 5529,
+      "event": 5530,
+      "ping": 5531,
+      "ij": 5532,
+      "▁friend": 5533,
+      "全面": 5534,
+      "ocal": 5535,
+      "▁ability": 5536,
+      "▁solutions": 5537,
+      "数字": 5538,
+      "并且": 5539,
+      "▁Sub": 5540,
+      "izer": 5541,
+      "iance": 5542,
+      "才能": 5543,
+      "nnamespace": 5544,
+      "ken": 5545,
+      "▁More": 5546,
+      "▁Group": 5547,
+      "▁format": 5548,
+      "首先": 5549,
+      "左右": 5550,
+      "▁accur": 5551,
+      "REF": 5552,
+      "lock": 5553,
+      "aul": 5554,
+      "▁obtained": 5555,
+      "char": 5556,
+      "▁face": 5557,
+      "农民": 5558,
+      "nes": 5559,
+      "Des": 5560,
+      "▁House": 5561,
+      "▁Pres": 5562,
+      "bed": 5563,
+      "atur": 5564,
+      "null": 5565,
+      "▁mid": 5566,
+      "He": 5567,
+      "ective": 5568,
+      "response": 5569,
+      "身体": 5570,
+      "▁November": 5571,
+      "▁stress": 5572,
+      "▁sil": 5573,
+      "iques": 5574,
+      "▁maint": 5575,
+      "▁everything": 5576,
+      "▁Public": 5577,
+      "ocket": 5578,
+      "当然": 5579,
+      "ii": 5580,
+      "▁prec": 5581,
+      "▁Pat": 5582,
+      "▁analy": 5583,
+      "ql": 5584,
+      "一年": 5585,
+      "都有": 5586,
+      "Client": 5587,
+      "')": 5588,
+      "augh": 5589,
+      "▁compos": 5590,
+      "Not": 5591,
+      "现已": 5592,
+      "▁parameters": 5593,
+      "loat": 5594,
+      "ools": 5595,
+      "([": 5596,
+      "准备": 5597,
+      "▁regul": 5598,
+      "▁staff": 5599,
+      "▁node": 5600,
+      "▁popular": 5601,
+      "inition": 5602,
+      "EL": 5603,
+      "▁ut": 5604,
+      "nOn": 5605,
+      "▁Ag": 5606,
+      "▁allows": 5607,
+      "▁ful": 5608,
+      "▁pressure": 5609,
+      "▁See": 5610,
+      "TH": 5611,
+      "apping": 5612,
+      "儿童": 5613,
+      "acc": 5614,
+      "▁La": 5615,
+      "argin": 5616,
+      "▁recently": 5617,
+      "▁True": 5618,
+      "ado": 5619,
+      "▁]": 5620,
+      "说明": 5621,
+      "▁highly": 5622,
+      "▁store": 5623,
+      "enu": 5624,
+      "▁global": 5625,
+      "▁dynam": 5626,
+      "SD": 5627,
+      "阅读": 5628,
+      "last": 5629,
+      "参数": 5630,
+      "▁applications": 5631,
+      "Response": 5632,
+      "heme": 5633,
+      "就会": 5634,
+      "che": 5635,
+      "▁named": 5636,
+      "PL": 5637,
+      "▁len": 5638,
+      "▁sus": 5639,
+      "\\');\\": 5640,
+      "source": 5641,
+      "dec": 5642,
+      "协议": 5643,
+      "▁late": 5644,
+      "gs": 5645,
+      "Component": 5646,
+      "智能": 5647,
+      "▁forward": 5648,
+      "▁surg": 5649,
+      "enses": 5650,
+      "也有": 5651,
+      "bug": 5652,
+      "十九": 5653,
+      "本次执行": 5654,
+      "充分": 5655,
+      "▁Sw": 5656,
+      "▁December": 5657,
+      "▁customers": 5658,
+      "实施": 5659,
+      "▁matrix": 5660,
+      "Controller": 5661,
+      "ico": 5662,
+      "▁intr": 5663,
+      "进一步": 5664,
+      "▁inside": 5665,
+      "▁directly": 5666,
+      "▁League": 5667,
+      "({": 5668,
+      "Inter": 5669,
+      "▁stay": 5670,
+      "▁server": 5671,
+      "▁package": 5672,
+      "////": 5673,
+      "▁click": 5674,
+      "commit": 5675,
+      "创建": 5676,
+      "▁acid": 5677,
+      "ampionship": 5678,
+      "加强": 5679,
+      "▁vers": 5680,
+      "▁official": 5681,
+      "button": 5682,
+      "超过": 5683,
+      "hood": 5684,
+      "十三": 5685,
+      "▁concept": 5686,
+      "▁town": 5687,
+      "▁section": 5688,
+      "cell": 5689,
+      "人们": 5690,
+      "models": 5691,
+      "民族": 5692,
+      "▁walk": 5693,
+      "▁girl": 5694,
+      "rl": 5695,
+      "rench": 5696,
+      "▁international": 5697,
+      "▁bed": 5698,
+      "eline": 5699,
+      "▁Japan": 5700,
+      "规划": 5701,
+      "level": 5702,
+      "':": 5703,
+      "建立": 5704,
+      "▁continue": 5705,
+      "是在": 5706,
+      "审查": 5707,
+      "下的": 5708,
+      "同学": 5709,
+      "ipe": 5710,
+      "办公": 5711,
+      "true": 5712,
+      "IF": 5713,
+      "满足": 5714,
+      "▁attempt": 5715,
+      "公安": 5716,
+      "优秀": 5717,
+      "rome": 5718,
+      "older": 5719,
+      "System": 5720,
+      "ypes": 5721,
+      "▁ways": 5722,
+      "ondon": 5723,
+      "rote": 5724,
+      "▁uses": 5725,
+      "Def": 5726,
+      "▁长": 5727,
+      "▁anything": 5728,
+      "gl": 5729,
+      "▁security": 5730,
+      "▁Ne": 5731,
+      "isf": 5732,
+      "几个": 5733,
+      "▁increasing": 5734,
+      "email": 5735,
+      "▁ensure": 5736,
+      "▁almost": 5737,
+      "closed": 5738,
+      "▁download": 5739,
+      "评价": 5740,
+      "OP": 5741,
+      "▁liter": 5742,
+      "▁Post": 5743,
+      "ds": 5744,
+      "措施": 5745,
+      "ifier": 5746,
+      "▁hyper": 5747,
+      "判断": 5748,
+      "long": 5749,
+      "▁sites": 5750,
+      "只要": 5751,
+      "未来": 5752,
+      "coming": 5753,
+      "reet": 5754,
+      "MM": 5755,
+      "期限": 5756,
+      "▁Cr": 5757,
+      "七年": 5758,
+      "▁‘": 5759,
+      "▁comment": 5760,
+      "Pr": 5761,
+      "acity": 5762,
+      "abled": 5763,
+      "问题的": 5764,
+      "考虑": 5765,
+      "体系": 5766,
+      "驾驶": 5767,
+      "▁surv": 5768,
+      "每天": 5769,
+      "bruary": 5770,
+      "▁contains": 5771,
+      "eration": 5772,
+      "▁draw": 5773,
+      "▁leading": 5774,
+      "De": 5775,
+      "第四": 5776,
+      "报告": 5777,
+      "Array": 5778,
+      "第一百四十": 5779,
+      "With": 5780,
+      "▁ste": 5781,
+      "bon": 5782,
+      "每个": 5783,
+      "性能": 5784,
+      "▁constant": 5785,
+      "▁quite": 5786,
+      "▁veh": 5787,
+      "地址": 5788,
+      "▁proced": 5789,
+      "▁wond": 5790,
+      "▁Here": 5791,
+      "▁told": 5792,
+      "平均": 5793,
+      "react": 5794,
+      "izes": 5795,
+      "调解": 5796,
+      "nAnd": 5797,
+      "▁evidence": 5798,
+      "▁road": 5799,
+      "除了": 5800,
+      "▁takes": 5801,
+      "▁seem": 5802,
+      "给出": 5803,
+      "▁station": 5804,
+      "公诉": 5805,
+      "输入": 5806,
+      "▁arr": 5807,
+      "▁simply": 5808,
+      "▁device": 5809,
+      "fit": 5810,
+      "ambda": 5811,
+      "▁ground": 5812,
+      "ocker": 5813,
+      "known": 5814,
+      "▁Up": 5815,
+      "▁Set": 5816,
+      "ara": 5817,
+      ")*": 5818,
+      "▁interface": 5819,
+      "bf": 5820,
+      "▁CD": 5821,
+      "▁neuro": 5822,
+      "▁PR": 5823,
+      "SC": 5824,
+      "oon": 5825,
+      "检测": 5826,
+      "▁sequence": 5827,
+      "▁wait": 5828,
+      "制度": 5829,
+      "▁Dec": 5830,
+      "▁items": 5831,
+      "pty": 5832,
+      "▁necessary": 5833,
+      "交流": 5834,
+      "rightarrow": 5835,
+      "sych": 5836,
+      "▁contr": 5837,
+      "ociation": 5838,
+      "不知道": 5839,
+      "ocr": 5840,
+      "▁believe": 5841,
+      "培养": 5842,
+      "define": 5843,
+      "▁gas": 5844,
+      "▁College": 5845,
+      "有些": 5846,
+      "▁travel": 5847,
+      "徒刑": 5848,
+      "立即": 5849,
+      "▁brain": 5850,
+      "▁therapy": 5851,
+      "▁Data": 5852,
+      "参与": 5853,
+      "考试": 5854,
+      "What": 5855,
+      "But": 5856,
+      "As": 5857,
+      "▁Russ": 5858,
+      "▁elements": 5859,
+      "期徒刑": 5860,
+      "▁que": 5861,
+      "五年": 5862,
+      "那个": 5863,
+      "▁capt": 5864,
+      "Be": 5865,
+      "▁projects": 5866,
+      "oke": 5867,
+      "本院在": 5868,
+      "不可": 5869,
+      "▁sample": 5870,
+      "arter": 5871,
+      "合同纠纷一案": 5872,
+      "▁respectively": 5873,
+      "otic": 5874,
+      "本裁定送达后": 5875,
+      "css": 5876,
+      "▁Then": 5877,
+      "联网": 5878,
+      "人物": 5879,
+      "只能": 5880,
+      "%,": 5881,
+      "▁friends": 5882,
+      "▁His": 5883,
+      "merc": 5884,
+      "程度": 5885,
+      "ican": 5886,
+      "▁direction": 5887,
+      "hema": 5888,
+      "())\\": 5889,
+      "▁Ang": 5890,
+      "▁anim": 5891,
+      "▁choose": 5892,
+      "New": 5893,
+      "▁commit": 5894,
+      "当时": 5895,
+      "事务所": 5896,
+      "的财产": 5897,
+      "机械": 5898,
+      "六年": 5899,
+      "▁worked": 5900,
+      "lymp": 5901,
+      "root": 5902,
+      "PC": 5903,
+      "速度": 5904,
+      "数量": 5905,
+      "II": 5906,
+      "▁produced": 5907,
+      "atively": 5908,
+      "改革": 5909,
+      "itude": 5910,
+      "raz": 5911,
+      "▁International": 5912,
+      "市中级人民法院": 5913,
+      "]:": 5914,
+      "▁board": 5915,
+      "▁latest": 5916,
+      "Of": 5917,
+      "▁beautiful": 5918,
+      "iforn": 5919,
+      "avid": 5920,
+      "ster": 5921,
+      "nWhen": 5922,
+      "tual": 5923,
+      "estern": 5924,
+      "Element": 5925,
+      "igen": 5926,
+      "本次执行程序": 5927,
+      "▁Ed": 5928,
+      "mathbb": 5929,
+      "urance": 5930,
+      "▁deriv": 5931,
+      "clus": 5932,
+      "olar": 5933,
+      "▁instance": 5934,
+      "▁looks": 5935,
+      ",\\\\": 5936,
+      "more": 5937,
+      "ington": 5938,
+      "ancy": 5939,
+      "▁emer": 5940,
+      "认定": 5941,
+      "课程": 5942,
+      "▁establish": 5943,
+      "▁temperature": 5944,
+      "电话": 5945,
+      "ued": 5946,
+      "ians": 5947,
+      "自治": 5948,
+      "▁soon": 5949,
+      "▁Copyright": 5950,
+      "制造": 5951,
+      "grad": 5952,
+      "理解": 5953,
+      "avy": 5954,
+      "}{\\\\": 5955,
+      "▁isn": 5956,
+      "exp": 5957,
+      "▁prep": 5958,
+      "心理": 5959,
+      "媒体": 5960,
+      "英语": 5961,
+      "▁bad": 5962,
+      "▁Please": 5963,
+      "成本": 5964,
+      "▁altern": 5965,
+      "▁Sch": 5966,
+      "nTr": 5967,
+      "▁plac": 5968,
+      "▁>>": 5969,
+      "保持": 5970,
+      "▁basis": 5971,
+      "ociety": 5972,
+      "坚持": 5973,
+      "lections": 5974,
+      "分享": 5975,
+      "▁blog": 5976,
+      "▁broad": 5977,
+      "idget": 5978,
+      "ocks": 5979,
+      "下来": 5980,
+      "适合": 5981,
+      "▁tim": 5982,
+      "▁favor": 5983,
+      "oes": 5984,
+      "▁receive": 5985,
+      "▁District": 5986,
+      "elta": 5987,
+      "▁compar": 5988,
+      "▁obtain": 5989,
+      "针对": 5990,
+      "哪些": 5991,
+      "▁variety": 5992,
+      "ellow": 5993,
+      "▁India": 5994,
+      "update": 5995,
+      "▁wide": 5996,
+      "▁fire": 5997,
+      "service": 5998,
+      "社区": 5999,
+      "▁send": 6000,
+      "An": 6001,
+      "第六": 6002,
+      "▁presented": 6003,
+      "semb": 6004,
+      "}(": 6005,
+      "▁degree": 6006,
+      "▁transfer": 6007,
+      "▁identified": 6008,
+      "的情况": 6009,
+      "ifornia": 6010,
+      "▁parts": 6011,
+      "wise": 6012,
+      "SP": 6013,
+      "风险": 6014,
+      "Manager": 6015,
+      "icians": 6016,
+      "cel": 6017,
+      "许多": 6018,
+      "▁party": 6019,
+      "▁王": 6020,
+      "hing": 6021,
+      "中央": 6022,
+      "eader": 6023,
+      "联合": 6024,
+      "持续": 6025,
+      "Input": 6026,
+      "▁resid": 6027,
+      "▁attack": 6028,
+      "etime": 6029,
+      "▁sort": 6030,
+      "▁ult": 6031,
+      "严重": 6032,
+      "资金": 6033,
+      "▁computer": 6034,
+      "orph": 6035,
+      "auth": 6036,
+      "交易": 6037,
+      "▁greater": 6038,
+      "▁gives": 6039,
+      "\"]": 6040,
+      "▁stage": 6041,
+      "加入": 6042,
+      "Base": 6043,
+      "▁numer": 6044,
+      "ka": 6045,
+      "▁rates": 6046,
+      "丰富": 6047,
+      "▁qual": 6048,
+      "iers": 6049,
+      "▁From": 6050,
+      "▁born": 6051,
+      "▁usually": 6052,
+      "▁physical": 6053,
+      "深圳": 6054,
+      "意见": 6055,
+      "▁throughout": 6056,
+      "itect": 6057,
+      "指导": 6058,
+      "▁Mus": 6059,
+      "CP": 6060,
+      "内容简介": 6061,
+      "比赛": 6062,
+      "▁ax": 6063,
+      "parse": 6064,
+      "▁Str": 6065,
+      "▁Dis": 6066,
+      "▁moment": 6067,
+      "duc": 6068,
+      "咨询": 6069,
+      "▁poly": 6070,
+      "初字第": 6071,
+      "ship": 6072,
+      "合法": 6073,
+      "▁February": 6074,
+      "▁hypot": 6075,
+      "iciency": 6076,
+      "▁Just": 6077,
+      "全球": 6078,
+      "▁tools": 6079,
+      "▁secret": 6080,
+      "▁deliver": 6081,
+      "None": 6082,
+      "▁studied": 6083,
+      "增长": 6084,
+      "收取": 6085,
+      "▁theory": 6086,
+      "▁male": 6087,
+      "▁longer": 6088,
+      "oz": 6089,
+      "▁oil": 6090,
+      "▁watch": 6091,
+      "▁district": 6092,
+      "▁matter": 6093,
+      "设置": 6094,
+      "Trans": 6095,
+      "▁RE": 6096,
+      "▁London": 6097,
+      "▁existing": 6098,
+      "不得": 6099,
+      "▁wor": 6100,
+      "asons": 6101,
+      "▁skin": 6102,
+      "itation": 6103,
+      "感情": 6104,
+      "erve": 6105,
+      "▁Inc": 6106,
+      "java": 6107,
+      "▁CON": 6108,
+      "den": 6109,
+      "tic": 6110,
+      "▁manufact": 6111,
+      "北省": 6112,
+      "document": 6113,
+      "▁Rev": 6114,
+      "abs": 6115,
+      "Override": 6116,
+      "▁Char": 6117,
+      "hy": 6118,
+      "method": 6119,
+      "nw": 6120,
+      "▁construct": 6121,
+      "target": 6122,
+      "Rep": 6123,
+      "▁reading": 6124,
+      "msg": 6125,
+      "nBut": 6126,
+      "anel": 6127,
+      "stream": 6128,
+      "细胞": 6129,
+      "▁resources": 6130,
+      "▁King": 6131,
+      "▁anti": 6132,
+      "▁states": 6133,
+      "▁Spec": 6134,
+      "DS": 6135,
+      "river": 6136,
+      "▁upon": 6137,
+      "医学百科": 6138,
+      "▁Your": 6139,
+      "▁words": 6140,
+      "更加": 6141,
+      "itc": 6142,
+      "▁rights": 6143,
+      "生物": 6144,
+      "\\\\_": 6145,
+      "▁Air": 6146,
+      "正在": 6147,
+      "except": 6148,
+      "稳定": 6149,
+      "检察院": 6150,
+      "▁PM": 6151,
+      "TS": 6152,
+      "三角": 6153,
+      "查明": 6154,
+      "特色": 6155,
+      "律师事务所": 6156,
+      "ias": 6157,
+      "这样的": 6158,
+      "越来": 6159,
+      "▁Inst": 6160,
+      "▁nucle": 6161,
+      "取得": 6162,
+      "~~~\\": 6163,
+      "西省": 6164,
+      "有期徒刑": 6165,
+      "gener": 6166,
+      "If": 6167,
+      "oice": 6168,
+      "asks": 6169,
+      "Ap": 6170,
+      "▁plant": 6171,
+      "▁approxim": 6172,
+      "ptions": 6173,
+      "监狱": 6174,
+      "▁${": 6175,
+      "▁techniques": 6176,
+      "▁save": 6177,
+      "▁equal": 6178,
+      "▁nil": 6179,
+      "nh": 6180,
+      "▁::": 6181,
+      "▁Now": 6182,
+      "保障": 6183,
+      "Gener": 6184,
+      "▁AM": 6185,
+      "▁force": 6186,
+      "▁except": 6187,
+      "pm": 6188,
+      "▁materials": 6189,
+      "越来越": 6190,
+      "购买": 6191,
+      "区域": 6192,
+      "样的": 6193,
+      "▁linear": 6194,
+      "td": 6195,
+      "目录": 6196,
+      "▁essential": 6197,
+      "体验": 6198,
+      "▁optim": 6199,
+      "用于": 6200,
+      "专家": 6201,
+      "▁Health": 6202,
+      "git": 6203,
+      "ighest": 6204,
+      "提交": 6205,
+      "代理审判员": 6206,
+      "▁outside": 6207,
+      "点击": 6208,
+      "运行": 6209,
+      "▁challeng": 6210,
+      "▁First": 6211,
+      "▁China": 6212,
+      "新闻": 6213,
+      "阶段": 6214,
+      "▁fall": 6215,
+      "▁algorithm": 6216,
+      "ATE": 6217,
+      "▁extra": 6218,
+      "因素": 6219,
+      "那些": 6220,
+      "▁Acc": 6221,
+      "求出": 6222,
+      "▁wrong": 6223,
+      "后的": 6224,
+      "为主": 6225,
+      "▁Pre": 6226,
+      "输出": 6227,
+      "▁DNA": 6228,
+      "广告": 6229,
+      "▁British": 6230,
+      "▁pretty": 6231,
+      "cdot": 6232,
+      "elves": 6233,
+      "改造": 6234,
+      "▁phone": 6235,
+      "▁Thanks": 6236,
+      "▁testing": 6237,
+      "▁\\\\\\": 6238,
+      "iece": 6239,
+      "▁李": 6240,
+      "alle": 6241,
+      "▁mechanisms": 6242,
+      "合理": 6243,
+      "第七": 6244,
+      "▁remain": 6245,
+      "▁basic": 6246,
+      "▁Har": 6247,
+      "▁Out": 6248,
+      "促进": 6249,
+      "▁served": 6250,
+      "出了": 6251,
+      "CA": 6252,
+      "▁distance": 6253,
+      "▁task": 6254,
+      "▁selected": 6255,
+      "aign": 6256,
+      "erved": 6257,
+      "plications": 6258,
+      "第五": 6259,
+      "人才": 6260,
+      "RA": 6261,
+      "▁America": 6262,
+      "▁Some": 6263,
+      "▁birth": 6264,
+      "▁promot": 6265,
+      "▁reach": 6266,
+      "食品": 6267,
+      "受到": 6268,
+      "uccess": 6269,
+      "▁programs": 6270,
+      "pg": 6271,
+      "▁sn": 6272,
+      "▁God": 6273,
+      "ran": 6274,
+      "▁apply": 6275,
+      "改变": 6276,
+      "▁url": 6277,
+      "▁Def": 6278,
+      "▁overall": 6279,
+      "▁$(": 6280,
+      "vider": 6281,
+      "▁bal": 6282,
+      "有人": 6283,
+      "▁central": 6284,
+      "merican": 6285,
+      "]);\\": 6286,
+      "▁itself": 6287,
+      "测试": 6288,
+      "▁providing": 6289,
+      "不同的": 6290,
+      "stitute": 6291,
+      "▁California": 6292,
+      "options": 6293,
+      "▁Mich": 6294,
+      "▁critical": 6295,
+      "▁hit": 6296,
+      "员工": 6297,
+      "▁fib": 6298,
+      "TO": 6299,
+      "century": 6300,
+      "GB": 6301,
+      "▁phase": 6302,
+      "KE": 6303,
+      "▁policy": 6304,
+      "▁transform": 6305,
+      "解答": 6306,
+      "都会": 6307,
+      "ooth": 6308,
+      "levant": 6309,
+      "▁decision": 6310,
+      "▁San": 6311,
+      "宣传": 6312,
+      "Inst": 6313,
+      "▁doc": 6314,
+      "犯罪": 6315,
+      "yles": 6316,
+      "▁negative": 6317,
+      "统计": 6318,
+      "以来": 6319,
+      "设施": 6320,
+      "▁entry": 6321,
+      "▁Class": 6322,
+      "ponents": 6323,
+      "▁presence": 6324,
+      "我们的": 6325,
+      "eps": 6326,
+      "▁component": 6327,
+      "win": 6328,
+      "量的": 6329,
+      "And": 6330,
+      "▁individuals": 6331,
+      "admin": 6332,
+      "▁mer": 6333,
+      "mercial": 6334,
+      "▁height": 6335,
+      "优势": 6336,
+      "uge": 6337,
+      "())": 6338,
+      "▁张": 6339,
+      "民间": 6340,
+      "plies": 6341,
+      "ca": 6342,
+      "▁choice": 6343,
+      "人民检察院": 6344,
+      "▁vit": 6345,
+      "的发展": 6346,
+      "▁variables": 6347,
+      "block": 6348,
+      "直线": 6349,
+      "▁cal": 6350,
+      "▁avoid": 6351,
+      "utils": 6352,
+      "▁Google": 6353,
+      "快速": 6354,
+      "恢复": 6355,
+      "see": 6356,
+      "levision": 6357,
+      "▁applied": 6358,
+      "eturn": 6359,
+      "▁candid": 6360,
+      "身份": 6361,
+      "▁Fl": 6362,
+      "Size": 6363,
+      "性质": 6364,
+      "ords": 6365,
+      "是一种": 6366,
+      "记者": 6367,
+      "▁everyone": 6368,
+      "ulations": 6369,
+      "images": 6370,
+      "▁sal": 6371,
+      "number": 6372,
+      "▁src": 6373,
+      "nCom": 6374,
+      "▁satisf": 6375,
+      "▁easily": 6376,
+      "▁synt": 6377,
+      "hent": 6378,
+      "▁sound": 6379,
+      "ounced": 6380,
+      "成绩": 6381,
+      "土地": 6382,
+      "▁Bro": 6383,
+      "▁Found": 6384,
+      "协会": 6385,
+      "体育": 6386,
+      "VER": 6387,
+      "▁limited": 6388,
+      "enge": 6389,
+      "▁reduce": 6390,
+      "▁stim": 6391,
+      "免费": 6392,
+      "▁Work": 6393,
+      "▁progress": 6394,
+      "▁deep": 6395,
+      "nand": 6396,
+      "show": 6397,
+      "ytes": 6398,
+      "▁places": 6399,
+      "办法": 6400,
+      "的高": 6401,
+      "申请执行": 6402,
+      "▁Open": 6403,
+      "▁Det": 6404,
+      "告诉": 6405,
+      "▁transport": 6406,
+      "别人": 6407,
+      "▁financial": 6408,
+      "▁son": 6409,
+      "height": 6410,
+      "▁shop": 6411,
+      "▁cam": 6412,
+      "再次": 6413,
+      "▁divid": 6414,
+      "\"></": 6415,
+      "utive": 6416,
+      "python": 6417,
+      "服刑": 6418,
+      "ares": 6419,
+      "▁coming": 6420,
+      "▁highest": 6421,
+      "▁sense": 6422,
+      "▁Math": 6423,
+      "实践": 6424,
+      "vention": 6425,
+      "Ass": 6426,
+      "elt": 6427,
+      "▁volume": 6428,
+      "▁modern": 6429,
+      "▁variable": 6430,
+      "theta": 6431,
+      "weet": 6432,
+      "找到": 6433,
+      "Out": 6434,
+      "▁ready": 6435,
+      "/(": 6436,
+      "▁Indian": 6437,
+      "nAn": 6438,
+      "▁independent": 6439,
+      "▁UK": 6440,
+      "▁join": 6441,
+      "icket": 6442,
+      "yr": 6443,
+      "经常": 6444,
+      "▁cook": 6445,
+      "医疗": 6446,
+      "▁components": 6447,
+      "用的": 6448,
+      "essages": 6449,
+      "工具": 6450,
+      "意思": 6451,
+      "▁led": 6452,
+      "DF": 6453,
+      "orge": 6454,
+      "地产": 6455,
+      "bra": 6456,
+      "查封": 6457,
+      "****************": 6458,
+      "No": 6459,
+      "▁anyone": 6460,
+      "}$$": 6461,
+      "QL": 6462,
+      "XX": 6463,
+      "▁correspon": 6464,
+      "ael": 6465,
+      "npackage": 6466,
+      "▁namespace": 6467,
+      "//\\": 6468,
+      "▁safety": 6469,
+      "联系": 6470,
+      "Sub": 6471,
+      "二〇一七年": 6472,
+      "▁mode": 6473,
+      "▁achieve": 6474,
+      "创作": 6475,
+      "<<": 6476,
+      "▁molecular": 6477,
+      "是什么": 6478,
+      "施工": 6479,
+      "types": 6480,
+      "Image": 6481,
+      "▁Sk": 6482,
+      "强制": 6483,
+      "▁Nor": 6484,
+      "错误": 6485,
+      "amin": 6486,
+      "▁missing": 6487,
+      "▁vector": 6488,
+      "互联网": 6489,
+      "▁behind": 6490,
+      "ba": 6491,
+      "ocation": 6492,
+      "split": 6493,
+      "etch": 6494,
+      "DA": 6495,
+      "十八": 6496,
+      "真正": 6497,
+      "ameter": 6498,
+      "▁functional": 6499,
+      "▁countries": 6500,
+      "到庭": 6501,
+      "▁Em": 6502,
+      "▁Ital": 6503,
+      "▁hon": 6504,
+      "▁assum": 6505,
+      "edia": 6506,
+      "ously": 6507,
+      "NU": 6508,
+      "({\\": 6509,
+      "ala": 6510,
+      "▁Ant": 6511,
+      "first": 6512,
+      "icens": 6513,
+      "▁\\\\)": 6514,
+      "▁hor": 6515,
+      "机会": 6516,
+      "space": 6517,
+      "ipment": 6518,
+      "▁}}": 6519,
+      "Index": 6520,
+      "▁rap": 6521,
+      "▁driv": 6522,
+      "iring": 6523,
+      "▁API": 6524,
+      "贷款": 6525,
+      "SON": 6526,
+      "ste": 6527,
+      "生态": 6528,
+      "▁类型": 6529,
+      "▁conc": 6530,
+      "▁enter": 6531,
+      "uster": 6532,
+      "▁rev": 6533,
+      "都不": 6534,
+      "osure": 6535,
+      "▁Cons": 6536,
+      "审理终结": 6537,
+      "False": 6538,
+      "▁skills": 6539,
+      "SA": 6540,
+      "▁streng": 6541,
+      "整个": 6542,
+      "▁student": 6543,
+      "istr": 6544,
+      "orial": 6545,
+      "开庭": 6546,
+      "▁proteins": 6547,
+      "认识": 6548,
+      "▁pull": 6549,
+      "▁Port": 6550,
+      "从而": 6551,
+      "▁script": 6552,
+      "▁ideas": 6553,
+      "▁Her": 6554,
+      "▁princ": 6555,
+      "product": 6556,
+      "▁scient": 6557,
+      "▁args": 6558,
+      "减少": 6559,
+      "▁Department": 6560,
+      "年来": 6561,
+      "tes": 6562,
+      "console": 6563,
+      "mt": 6564,
+      "▁immed": 6565,
+      "模型": 6566,
+      "确认": 6567,
+      "小学": 6568,
+      "Count": 6569,
+      "▁appropri": 6570,
+      "▁Dem": 6571,
+      "福建": 6572,
+      "nPro": 6573,
+      "▁asked": 6574,
+      "arily": 6575,
+      "You": 6576,
+      "Result": 6577,
+      "生命": 6578,
+      "经历": 6579,
+      "临床": 6580,
+      "Page": 6581,
+      "▁completed": 6582,
+      "▁coord": 6583,
+      "律师事务所律师": 6584,
+      "witter": 6585,
+      "pite": 6586,
+      "leq": 6587,
+      "核心": 6588,
+      "BN": 6589,
+      "ida": 6590,
+      "ecause": 6591,
+      "wood": 6592,
+      "▁isol": 6593,
+      "▁dog": 6594,
+      "select": 6595,
+      "▁President": 6596,
+      "det": 6597,
+      "ections": 6598,
+      "▁opportunity": 6599,
+      "zym": 6600,
+      "如不服": 6601,
+      "▁followed": 6602,
+      "▁register": 6603,
+      "velopment": 6604,
+      "ivery": 6605,
+      "Action": 6606,
+      "人口": 6607,
+      "▁processes": 6608,
+      "▁digital": 6609,
+      "最终": 6610,
+      "ening": 6611,
+      "▁recommend": 6612,
+      "▁solve": 6613,
+      "▁reduced": 6614,
+      "▁reference": 6615,
+      "here": 6616,
+      "阳市": 6617,
+      "At": 6618,
+      "▁Val": 6619,
+      "由原告": 6620,
+      "移动": 6621,
+      "posed": 6622,
+      "调整": 6623,
+      "▁visual": 6624,
+      "Reg": 6625,
+      "mitted": 6626,
+      "降低": 6627,
+      "记录": 6628,
+      "sv": 6629,
+      "▁contribut": 6630,
+      "common": 6631,
+      "▁hydro": 6632,
+      "template": 6633,
+      "ologies": 6634,
+      "想要": 6635,
+      "▁ANY": 6636,
+      "山东省": 6637,
+      "attle": 6638,
+      "▁Of": 6639,
+      "riter": 6640,
+      "IONS": 6641,
+      "▁clients": 6642,
+      "体的": 6643,
+      "ulate": 6644,
+      "▁customer": 6645,
+      "▁brand": 6646,
+      "▁extrem": 6647,
+      "ycl": 6648,
+      "▁antib": 6649,
+      "▁began": 6650,
+      "lands": 6651,
+      "rupt": 6652,
+      "本院依法": 6653,
+      "IB": 6654,
+      "dist": 6655,
+      "现已审理终结": 6656,
+      "及其": 6657,
+      "password": 6658,
+      "Handler": 6659,
+      "▁fields": 6660,
+      "▁Log": 6661,
+      "alculate": 6662,
+      "▁wanted": 6663,
+      "▁tried": 6664,
+      "▁Day": 6665,
+      "版本": 6666,
+      "▁maximum": 6667,
+      "Le": 6668,
+      "车辆": 6669,
+      "band": 6670,
+      "事件": 6671,
+      "ager": 6672,
+      "▁attention": 6673,
+      "▁acqu": 6674,
+      "里的": 6675,
+      "及时": 6676,
+      "压力": 6677,
+      "▁assist": 6678,
+      "▁samples": 6679,
+      "▁units": 6680,
+      "▁meeting": 6681,
+      "▁exhib": 6682,
+      "token": 6683,
+      "ris": 6684,
+      "认真": 6685,
+      "就可以": 6686,
+      "▁histor": 6687,
+      "运用": 6688,
+      "▁Australia": 6689,
+      "▁construction": 6690,
+      "amily": 6691,
+      "parent": 6692,
+      "▁tumor": 6693,
+      "文书": 6694,
+      "\"/": 6695,
+      "▁playing": 6696,
+      "Rec": 6697,
+      "Color": 6698,
+      "▁lost": 6699,
+      "即发生法律效力": 6700,
+      "▁discover": 6701,
+      "▁tissue": 6702,
+      "至今": 6703,
+      "illing": 6704,
+      "▁row": 6705,
+      "mathrm": 6706,
+      "出的": 6707,
+      "另外": 6708,
+      "不好": 6709,
+      "▁TV": 6710,
+      "▁第": 6711,
+      "▁slow": 6712,
+      "omb": 6713,
+      "nfunc": 6714,
+      "ario": 6715,
+      "▁Free": 6716,
+      "gu": 6717,
+      "ervices": 6718,
+      "▁User": 6719,
+      "方便": 6720,
+      "清楚": 6721,
+      "▁square": 6722,
+      "天津": 6723,
+      "▁domain": 6724,
+      "▁Are": 6725,
+      "▁Fin": 6726,
+      "▁forms": 6727,
+      "二〇一六年": 6728,
+      "bot": 6729,
+      "电脑": 6730,
+      "▁Mod": 6731,
+      "break": 6732,
+      "定义": 6733,
+      "被申请人": 6734,
+      "约定": 6735,
+      "tf": 6736,
+      "工艺": 6737,
+      "▁While": 6738,
+      "▁village": 6739,
+      "za": 6740,
+      "▁efficient": 6741,
+      "▁instit": 6742,
+      "Status": 6743,
+      "ya": 6744,
+      "ascular": 6745,
+      "做好": 6746,
+      "elcome": 6747,
+      "otype": 6748,
+      "nLiving": 6749,
+      "fty": 6750,
+      "三个": 6751,
+      "plot": 6752,
+      "lier": 6753,
+      "Te": 6754,
+      "应的": 6755,
+      "债务": 6756,
+      "疾病": 6757,
+      "nSolution": 6758,
+      "可在": 6759,
+      "▁purpose": 6760,
+      "master": 6761,
+      "omas": 6762,
+      "▁starting": 6763,
+      "▁determined": 6764,
+      "▁theore": 6765,
+      "icial": 6766,
+      "▁Research": 6767,
+      "监督": 6768,
+      "▁memory": 6769,
+      "▁frequency": 6770,
+      "竞争": 6771,
+      "加工": 6772,
+      "演员": 6773,
+      "长期": 6774,
+      "background": 6775,
+      "GF": 6776,
+      "tag": 6777,
+      "▁books": 6778,
+      "▁Aut": 6779,
+      "CS": 6780,
+      "▁Black": 6781,
+      "表达": 6782,
+      "注册": 6783,
+      "nl": 6784,
+      "▁Direct": 6785,
+      "acing": 6786,
+      "▁cop": 6787,
+      "filter": 6788,
+      "▁subst": 6789,
+      "结论": 6790,
+      "七条": 6791,
+      "▁nice": 6792,
+      "日生": 6793,
+      "▁wall": 6794,
+      "▁Techn": 6795,
+      "▁WAR": 6796,
+      "push": 6797,
+      "▁differences": 6798,
+      "▁setting": 6799,
+      "▁daily": 6800,
+      "▁closed": 6801,
+      "igation": 6802,
+      "▁leave": 6803,
+      "生的": 6804,
+      "▁machine": 6805,
+      "elebr": 6806,
+      "下面": 6807,
+      "▁previously": 6808,
+      "配置": 6809,
+      "▁nothing": 6810,
+      "asc": 6811,
+      "omin": 6812,
+      "▁charg": 6813,
+      "▁</": 6814,
+      "过去": 6815,
+      "child": 6816,
+      "第二百五十": 6817,
+      "▁fully": 6818,
+      "▁award": 6819,
+      "更新": 6820,
+      "▁statement": 6821,
+      "▁European": 6822,
+      "▁外": 6823,
+      "Content": 6824,
+      "▁po": 6825,
+      "oto": 6826,
+      "厘米": 6827,
+      "营销": 6828,
+      "▁prefer": 6829,
+      "查询": 6830,
+      "rowser": 6831,
+      "ieved": 6832,
+      "副本": 6833,
+      "▁[]\\": 6834,
+      "基金": 6835,
+      "▁culture": 6836,
+      "思路": 6837,
+      "clusion": 6838,
+      "Http": 6839,
+      "事业": 6840,
+      "银行股份有限公司": 6841,
+      "群众": 6842,
+      "▁Commun": 6843,
+      "▁mice": 6844,
+      "真实": 6845,
+      "▁changed": 6846,
+      "▁thus": 6847,
+      "nam": 6848,
+      "▁ratio": 6849,
+      "提起": 6850,
+      "DB": 6851,
+      "它的": 6852,
+      "▁Answer": 6853,
+      "▁Sen": 6854,
+      "▁remove": 6855,
+      "▁Championship": 6856,
+      "unit": 6857,
+      "▁Mr": 6858,
+      "正式": 6859,
+      "▁Cour": 6860,
+      "▁psych": 6861,
+      "看看": 6862,
+      "▁phen": 6863,
+      "MB": 6864,
+      "▁mobile": 6865,
+      "▁database": 6866,
+      "iable": 6867,
+      "rep": 6868,
+      "▁cool": 6869,
+      "rolled": 6870,
+      "作者是": 6871,
+      "撤诉申请": 6872,
+      "missions": 6873,
+      "▁Hol": 6874,
+      "十四": 6875,
+      "也可以": 6876,
+      "osing": 6877,
+      "也会": 6878,
+      "▁couple": 6879,
+      "植物": 6880,
+      "▁background": 6881,
+      "房屋": 6882,
+      "▁mechanism": 6883,
+      "化的": 6884,
+      "▁AND": 6885,
+      "意识": 6886,
+      "民初字第": 6887,
+      "团队": 6888,
+      "▁lack": 6889,
+      "▁leaders": 6890,
+      "▁storage": 6891,
+      "后来": 6892,
+      "fs": 6893,
+      "anger": 6894,
+      "▁charge": 6895,
+      "ned": 6896,
+      "▁safe": 6897,
+      "izations": 6898,
+      "▁Project": 6899,
+      "明确": 6900,
+      "ota": 6901,
+      "来自": 6902,
+      "atin": 6903,
+      "▁Mac": 6904,
+      "▁generated": 6905,
+      "▁yourself": 6906,
+      "▁Olymp": 6907,
+      "▁module": 6908,
+      "▁Service": 6909,
+      "考查": 6910,
+      "▁False": 6911,
+      "azing": 6912,
+      "▁note": 6913,
+      "▁ones": 6914,
+      "▁expert": 6915,
+      "存款": 6916,
+      "▁)\\": 6917,
+      "▁ten": 6918,
+      "hi": 6919,
+      "给付": 6920,
+      "oined": 6921,
+      "先进": 6922,
+      "words": 6923,
+      ")^": 6924,
+      "zer": 6925,
+      "overy": 6926,
+      "原审": 6927,
+      "▁excell": 6928,
+      "▁findings": 6929,
+      "mar": 6930,
+      "ga": 6931,
+      "Act": 6932,
+      "借贷": 6933,
+      "▁vs": 6934,
+      "▁metab": 6935,
+      "▁filter": 6936,
+      "obj": 6937,
+      "▁fixed": 6938,
+      "▁fine": 6939,
+      "多的": 6940,
+      "▁Cup": 6941,
+      "▁mention": 6942,
+      "浙江省": 6943,
+      "▁library": 6944,
+      "DO": 6945,
+      "txt": 6946,
+      "▁traditional": 6947,
+      "▁suit": 6948,
+      "▁nature": 6949,
+      "only": 6950,
+      "infty": 6951,
+      "▁East": 6952,
+      "正确的": 6953,
+      "IO": 6954,
+      "onic": 6955,
+      "rees": 6956,
+      "isions": 6957,
+      "ialog": 6958,
+      "结婚": 6959,
+      "减半": 6960,
+      "hens": 6961,
+      "Line": 6962,
+      "can": 6963,
+      "下载": 6964,
+      "ometry": 6965,
+      "▁collabor": 6966,
+      "Group": 6967,
+      "Url": 6968,
+      "题目": 6969,
+      "anged": 6970,
+      "▁costs": 6971,
+      "▁WIT": 6972,
+      "战略": 6973,
+      "▁Met": 6974,
+      "▁French": 6975,
+      "command": 6976,
+      "framework": 6977,
+      "EX": 6978,
+      "▁prem": 6979,
+      "FT": 6980,
+      "▁died": 6981,
+      "rate": 6982,
+      "esis": 6983,
+      "自由": 6984,
+      "▁rot": 6985,
+      "▁Web": 6986,
+      "已知": 6987,
+      "LIC": 6988,
+      "▁happy": 6989,
+      "▁Canada": 6990,
+      "▁vert": 6991,
+      "▁mac": 6992,
+      "URL": 6993,
+      "▁operation": 6994,
+      "atically": 6995,
+      "NS": 6996,
+      "Button": 6997,
+      "pir": 6998,
+      "▁auth": 6999,
+      "▁aspect": 7000,
+      "################": 7001,
+      "ivo": 7002,
+      "actor": 7003,
+      "若干": 7004,
+      "docs": 7005,
+      "rd": 7006,
+      "mem": 7007,
+      "ville": 7008,
+      "▁cyt": 7009,
+      "ENSE": 7010,
+      "不错": 7011,
+      "sembly": 7012,
+      "▁width": 7013,
+      "amma": 7014,
+      "▁Also": 7015,
+      "icks": 7016,
+      "orpor": 7017,
+      "usr": 7018,
+      "xx": 7019,
+      "董事": 7020,
+      "ao": 7021,
+      "反应": 7022,
+      "itors": 7023,
+      "▁settings": 7024,
+      "▁Windows": 7025,
+      "▁eth": 7026,
+      "统一": 7027,
+      "inct": 7028,
+      "▁core": 7029,
+      "事故": 7030,
+      "▁description": 7031,
+      "▁president": 7032,
+      "▁prime": 7033,
+      "▁viol": 7034,
+      "▁neigh": 7035,
+      "▁mount": 7036,
+      "itis": 7037,
+      "▁zero": 7038,
+      "最近": 7039,
+      "鉴定": 7040,
+      "干部": 7041,
+      "▁summer": 7042,
+      "conf": 7043,
+      "▁requirements": 7044,
+      "▁Rem": 7045,
+      "▁window": 7046,
+      "动物": 7047,
+      "▁interesting": 7048,
+      "▁ow": 7049,
+      "▁whose": 7050,
+      "肯定": 7051,
+      "ls": 7052,
+      "▁scale": 7053,
+      "安徽省": 7054,
+      "protected": 7055,
+      "▁ded": 7056,
+      "▁families": 7057,
+      "ilitary": 7058,
+      "▁sets": 7059,
+      "▁exactly": 7060,
+      "require": 7061,
+      "▁Don": 7062,
+      "发表": 7063,
+      "▁heat": 7064,
+      "▁teams": 7065,
+      "来了": 7066,
+      "▁comfort": 7067,
+      "▁ID": 7068,
+      "dule": 7069,
+      "apache": 7070,
+      "Conn": 7071,
+      "▁seg": 7072,
+      "父母": 7073,
+      "复杂": 7074,
+      "▁measured": 7075,
+      "▁combination": 7076,
+      "▁Phil": 7077,
+      "idden": 7078,
+      "▁absol": 7079,
+      "▁exerc": 7080,
+      "▁drive": 7081,
+      "▁hop": 7082,
+      "大小": 7083,
+      "十分": 7084,
+      "▁Since": 7085,
+      "▁collect": 7086,
+      "▁monitor": 7087,
+      "▁completely": 7088,
+      "Property": 7089,
+      "close": 7090,
+      "▁war": 7091,
+      "▁cycl": 7092,
+      "▁fre": 7093,
+      "▁sold": 7094,
+      "的解释": 7095,
+      "wide": 7096,
+      "▁solid": 7097,
+      "▁/>\\": 7098,
+      "▁cardi": 7099,
+      "▁net": 7100,
+      "RANT": 7101,
+      "▁produce": 7102,
+      "▁regarding": 7103,
+      "道路": 7104,
+      "annot": 7105,
+      "▁tag": 7106,
+      "的重要": 7107,
+      "PA": 7108,
+      "vest": 7109,
+      "ounter": 7110,
+      "▁tree": 7111,
+      "打开": 7112,
+      "形的": 7113,
+      "graph": 7114,
+      "ught": 7115,
+      "plant": 7116,
+      "引起": 7117,
+      "tee": 7118,
+      "观察": 7119,
+      "▁Republic": 7120,
+      "uffer": 7121,
+      "▁sympt": 7122,
+      "matrix": 7123,
+      "▁wood": 7124,
+      "▁json": 7125,
+      "文名": 7126,
+      "▁Version": 7127,
+      "▁handle": 7128,
+      "▁towards": 7129,
+      "特征": 7130,
+      "描述": 7131,
+      "sem": 7132,
+      "▁deaths": 7133,
+      "▁organization": 7134,
+      "冻结": 7135,
+      "必要": 7136,
+      "▁park": 7137,
+      "我是": 7138,
+      "温度": 7139,
+      "▁names": 7140,
+      "▁innov": 7141,
+      "技能": 7142,
+      "▁expand": 7143,
+      "▁ox": 7144,
+      "▁channel": 7145,
+      "rier": 7146,
+      "▁hospital": 7147,
+      "ART": 7148,
+      "▁signal": 7149,
+      "高的": 7150,
+      "▁inform": 7151,
+      "▁sales": 7152,
+      "▁selection": 7153,
+      "ensor": 7154,
+      "iki": 7155,
+      "▁Willi": 7156,
+      "Us": 7157,
+      "▁technique": 7158,
+      "▁ball": 7159,
+      "▁damage": 7160,
+      "▁resource": 7161,
+      "化学": 7162,
+      "▁Hel": 7163,
+      "params": 7164,
+      "▁Program": 7165,
+      "icated": 7166,
+      "▁advant": 7167,
+      "header": 7168,
+      "From": 7169,
+      "perties": 7170,
+      "▁nor": 7171,
+      "▁political": 7172,
+      "EM": 7173,
+      "▁var": 7174,
+      "inner": 7175,
+      "▁bank": 7176,
+      "日立案": 7177,
+      "▁worth": 7178,
+      "▁interview": 7179,
+      "▁ur": 7180,
+      "▁affect": 7181,
+      "ician": 7182,
+      "nhttps": 7183,
+      "规模": 7184,
+      "▁lif": 7185,
+      "▁flex": 7186,
+      "里面": 7187,
+      "Str": 7188,
+      "▁Form": 7189,
+      "▁Software": 7190,
+      "▁genes": 7191,
+      "rid": 7192,
+      "▁structures": 7193,
+      "nvar": 7194,
+      "▁Di": 7195,
+      "has": 7196,
+      "成了": 7197,
+      "世纪": 7198,
+      "icensed": 7199,
+      "相对": 7200,
+      "▁campaign": 7201,
+      "Command": 7202,
+      "并不": 7203,
+      "▁police": 7204,
+      "▁页": 7205,
+      "之间的": 7206,
+      "FL": 7207,
+      "▁River": 7208,
+      "inese": 7209,
+      "iment": 7210,
+      "itch": 7211,
+      "▁Mor": 7212,
+      "ho": 7213,
+      "▁score": 7214,
+      "Factory": 7215,
+      "从事": 7216,
+      "▁restr": 7217,
+      "ymph": 7218,
+      "Module": 7219,
+      "Let": 7220,
+      "nfor": 7221,
+      "一份": 7222,
+      "pret": 7223,
+      "beta": 7224,
+      "free": 7225,
+      "igration": 7226,
+      "uate": 7227,
+      "▁auto": 7228,
+      "▁goals": 7229,
+      "stack": 7230,
+      "撤回起诉": 7231,
+      "Att": 7232,
+      "▁gold": 7233,
+      "▁devices": 7234,
+      "ishing": 7235,
+      "▁Law": 7236,
+      "分为": 7237,
+      "rive": 7238,
+      "las": 7239,
+      "weight": 7240,
+      "▁sell": 7241,
+      "▁goes": 7242,
+      "▁strategy": 7243,
+      "减半收取": 7244,
+      "判处": 7245,
+      "▁creating": 7246,
+      "▁correl": 7247,
+      "平方米": 7248,
+      "▁announced": 7249,
+      "起点": 7250,
+      "成长": 7251,
+      "夫妻": 7252,
+      "香港": 7253,
+      "台湾": 7254,
+      "acebook": 7255,
+      "help": 7256,
+      "margin": 7257,
+      "人类": 7258,
+      "nAnswer": 7259,
+      "确定的": 7260,
+      "▁green": 7261,
+      "nAfter": 7262,
+      "时期": 7263,
+      "home": 7264,
+      "▁appropriate": 7265,
+      "ederal": 7266,
+      "▁catch": 7267,
+      "工资": 7268,
+      "spring": 7269,
+      "▁Suppose": 7270,
+      "adding": 7271,
+      "useum": 7272,
+      "房地产": 7273,
+      "istor": 7274,
+      "▁Bel": 7275,
+      "如此": 7276,
+      "▁quickly": 7277,
+      "nel": 7278,
+      "Token": 7279,
+      "子的": 7280,
+      "CI": 7281,
+      "有点": 7282,
+      "采取": 7283,
+      "On": 7284,
+      "▁labor": 7285,
+      "相信": 7286,
+      "不少": 7287,
+      "办理": 7288,
+      "aine": 7289,
+      "▁Vir": 7290,
+      "▁rout": 7291,
+      "Sup": 7292,
+      "▁adapt": 7293,
+      "▁HT": 7294,
+      "▁gets": 7295,
+      "有限责任": 7296,
+      "▁interested": 7297,
+      "▁relevant": 7298,
+      "RNA": 7299,
+      "商务": 7300,
+      "▁Bo": 7301,
+      "▁Street": 7302,
+      "roy": 7303,
+      "nUn": 7304,
+      "adian": 7305,
+      "RI": 7306,
+      "lamm": 7307,
+      "▁contain": 7308,
+      "ummer": 7309,
+      "ests": 7310,
+      "第一百四十五": 7311,
+      "super": 7312,
+      "imer": 7313,
+      "Query": 7314,
+      "CON": 7315,
+      "Aut": 7316,
+      "adata": 7317,
+      "▁Ear": 7318,
+      "推进": 7319,
+      "解题": 7320,
+      "居民": 7321,
+      "ini": 7322,
+      "▁connection": 7323,
+      "RL": 7324,
+      "重要的": 7325,
+      "exec": 7326,
+      "有限责任公司": 7327,
+      "▁infection": 7328,
+      "教授": 7329,
+      "对象": 7330,
+      "妈妈": 7331,
+      "hedul": 7332,
+      "▁views": 7333,
+      "▁Sing": 7334,
+      "▁wonder": 7335,
+      "ographic": 7336,
+      "接口": 7337,
+      "▁steps": 7338,
+      "▁Div": 7339,
+      "▁Mal": 7340,
+      "opic": 7341,
+      "▁les": 7342,
+      "▁lives": 7343,
+      "▁requires": 7344,
+      "▁treated": 7345,
+      "确实": 7346,
+      "手术": 7347,
+      "▁relative": 7348,
+      "gebra": 7349,
+      "▁Over": 7350,
+      "▁WARRANT": 7351,
+      "ago": 7352,
+      "陪审": 7353,
+      "▁NOT": 7354,
+      "parser": 7355,
+      "▁Fil": 7356,
+      "Builder": 7357,
+      "▁appoint": 7358,
+      "▁influence": 7359,
+      "结束": 7360,
+      "urt": 7361,
+      "▁evaluate": 7362,
+      "条的规定": 7363,
+      "▁larger": 7364,
+      "▁moved": 7365,
+      "issues": 7366,
+      "详细": 7367,
+      "aby": 7368,
+      "▁Ben": 7369,
+      "FF": 7370,
+      "lambda": 7371,
+      "训练": 7372,
+      "ki": 7373,
+      "到庭参加": 7374,
+      "ni": 7375,
+      "▁healthy": 7376,
+      "Dis": 7377,
+      "enty": 7378,
+      "ano": 7379,
+      "UI": 7380,
+      "igma": 7381,
+      "nhttp": 7382,
+      "Entity": 7383,
+      "规范": 7384,
+      "▁caused": 7385,
+      "当前": 7386,
+      "▁particularly": 7387,
+      "陪审员": 7388,
+      "掌握": 7389,
+      "▁implementation": 7390,
+      "nprint": 7391,
+      "▁plas": 7392,
+      "▁commercial": 7393,
+      "▁hol": 7394,
+      "reated": 7395,
+      "▁Home": 7396,
+      "涉及": 7397,
+      "erate": 7398,
+      "▁session": 7399,
+      "▁pul": 7400,
+      "nth": 7401,
+      "odies": 7402,
+      "报道": 7403,
+      "netic": 7404,
+      "公共": 7405,
+      "nvoid": 7406,
+      "的最": 7407,
+      "▁er": 7408,
+      "kw": 7409,
+      "尤其": 7410,
+      "Equal": 7411,
+      "ayers": 7412,
+      "以前": 7413,
+      "▁architect": 7414,
+      "Run": 7415,
+      "▁reports": 7416,
+      "ria": 7417,
+      "日止": 7418,
+      "危险": 7419,
+      "▁conver": 7420,
+      "▁pair": 7421,
+      "本文": 7422,
+      "▁injury": 7423,
+      "公式": 7424,
+      "iat": 7425,
+      "instance": 7426,
+      "▁tested": 7427,
+      "上诉状": 7428,
+      "UE": 7429,
+      "nav": 7430,
+      "她的": 7431,
+      "nPeople": 7432,
+      "ester": 7433,
+      "Util": 7434,
+      "▁plans": 7435,
+      "onday": 7436,
+      "▁Find": 7437,
+      "避免": 7438,
+      "▁legal": 7439,
+      "的值": 7440,
+      "circ": 7441,
+      "agon": 7442,
+      "本裁定送达后即发生法律效力": 7443,
+      "▁female": 7444,
+      "▁revealed": 7445,
+      "▁Solution": 7446,
+      "▁celebr": 7447,
+      "行的": 7448,
+      "EO": 7449,
+      "▁poor": 7450,
+      "偿还": 7451,
+      "ouncil": 7452,
+      "▁derivative": 7453,
+      "人民陪审员": 7454,
+      "▁Ste": 7455,
+      "微信": 7456,
+      "iding": 7457,
+      "astr": 7458,
+      "河南省": 7459,
+      "mas": 7460,
+      "▁capacity": 7461,
+      "▁television": 7462,
+      "▁classes": 7463,
+      "四川省": 7464,
+      "▁serious": 7465,
+      "▁python": 7466,
+      "ITY": 7467,
+      "运输": 7468,
+      "▁equipment": 7469,
+      "▁Vol": 7470,
+      "Create": 7471,
+      "本书": 7472,
+      "▁planning": 7473,
+      "独立": 7474,
+      "nWith": 7475,
+      "ped": 7476,
+      "的新": 7477,
+      "四年": 7478,
+      "sim": 7479,
+      "bi": 7480,
+      "审判长": 7481,
+      "▁train": 7482,
+      "▁Using": 7483,
+      "Stream": 7484,
+      "▁fa": 7485,
+      "绝对": 7486,
+      "▁rules": 7487,
+      "uced": 7488,
+      "xml": 7489,
+      "▁Association": 7490,
+      "▁Ill": 7491,
+      "▁developing": 7492,
+      "▁receptor": 7493,
+      "创造": 7494,
+      "面对": 7495,
+      "▁formation": 7496,
+      "produ": 7497,
+      "机制": 7498,
+      "▁sen": 7499,
+      "{%": 7500,
+      "▁Friday": 7501,
+      "force": 7502,
+      "力量": 7503,
+      "▁enzym": 7504,
+      "ibly": 7505,
+      "orders": 7506,
+      "维护": 7507,
+      "▁density": 7508,
+      "▁公司": 7509,
+      "深入": 7510,
+      "文明": 7511,
+      "ci": 7512,
+      "▁generate": 7513,
+      "▁shape": 7514,
+      "nms": 7515,
+      "▁science": 7516,
+      "fix": 7517,
+      "▁Foundation": 7518,
+      "Version": 7519,
+      "▁moving": 7520,
+      "ITE": 7521,
+      "整体": 7522,
+      "▁David": 7523,
+      "▁($": 7524,
+      "▁syn": 7525,
+      "ifically": 7526,
+      "▁hair": 7527,
+      "▁benefits": 7528,
+      "Web": 7529,
+      "消费者": 7530,
+      "delete": 7531,
+      "card": 7532,
+      "ses": 7533,
+      "NO": 7534,
+      "▁middle": 7535,
+      "▁picture": 7536,
+      "Load": 7537,
+      "皮肤": 7538,
+      "理念": 7539,
+      "▁parents": 7540,
+      "Table": 7541,
+      "▁TO": 7542,
+      "指控": 7543,
+      "undle": 7544,
+      "▁stream": 7545,
+      "EC": 7546,
+      "$$\\\\": 7547,
+      "▁cert": 7548,
+      "sg": 7549,
+      "▁largest": 7550,
+      "erry": 7551,
+      "ua": 7552,
+      "贸易": 7553,
+      "▁Time": 7554,
+      "nIs": 7555,
+      "rum": 7556,
+      "sw": 7557,
+      "自身": 7558,
+      "hedule": 7559,
+      "▁reduction": 7560,
+      "▁Trump": 7561,
+      "itchen": 7562,
+      "分布": 7563,
+      "讨论": 7564,
+      "▁Great": 7565,
+      "▁Anal": 7566,
+      "▁electric": 7567,
+      "特殊": 7568,
+      "ryst": 7569,
+      "acters": 7570,
+      "▁Model": 7571,
+      "vector": 7572,
+      "▁cat": 7573,
+      "▁fresh": 7574,
+      "一定要": 7575,
+      "req": 7576,
+      "▁actual": 7577,
+      "▁responses": 7578,
+      "PD": 7579,
+      "▁repe": 7580,
+      "float": 7581,
+      "▁although": 7582,
+      "▁jud": 7583,
+      "nAll": 7584,
+      "▁Wash": 7585,
+      "uy": 7586,
+      "包含": 7587,
+      "▁figure": 7588,
+      "▁formula": 7589,
+      "▁Use": 7590,
+      "角色": 7591,
+      "▁bus": 7592,
+      "▁bud": 7593,
+      "▁serve": 7594,
+      "多次": 7595,
+      "南京": 7596,
+      "合议": 7597,
+      "ello": 7598,
+      "▁wrote": 7599,
+      "重新": 7600,
+      "lines": 7601,
+      "false": 7602,
+      "eria": 7603,
+      "利率": 7604,
+      "地点": 7605,
+      "ien": 7606,
+      "google": 7607,
+      "了一个": 7608,
+      "▁England": 7609,
+      "▁fat": 7610,
+      "▁py": 7611,
+      "▁binding": 7612,
+      "推动": 7613,
+      "▁allowed": 7614,
+      "它们": 7615,
+      "sys": 7616,
+      "MO": 7617,
+      "oom": 7618,
+      "▁sources": 7619,
+      "▁remember": 7620,
+      "发挥": 7621,
+      "▁chance": 7622,
+      "cases": 7623,
+      "ky": 7624,
+      "property": 7625,
+      "计算机": 7626,
+      "▁cm": 7627,
+      "风格": 7628,
+      "ideos": 7629,
+      "▁Book": 7630,
+      "先生": 7631,
+      "▁measures": 7632,
+      "displaystyle": 7633,
+      "emplate": 7634,
+      "限制": 7635,
+      "olf": 7636,
+      "osc": 7637,
+      "▁processing": 7638,
+      "▁saw": 7639,
+      "十日内": 7640,
+      "武汉": 7641,
+      "相同": 7642,
+      "▁surgery": 7643,
+      "▁decided": 7644,
+      "▁sou": 7645,
+      "IES": 7646,
+      "▁button": 7647,
+      "store": 7648,
+      "负责人": 7649,
+      "slant": 7650,
+      "center": 7651,
+      "▁rock": 7652,
+      "▁Tex": 7653,
+      "无论": 7654,
+      "▁Council": 7655,
+      "资产": 7656,
+      "LL": 7657,
+      "▁extract": 7658,
+      "▁improved": 7659,
+      "king": 7660,
+      "berg": 7661,
+      "▁experimental": 7662,
+      "▁woman": 7663,
+      "三年": 7664,
+      "nAt": 7665,
+      "▁master": 7666,
+      "ivil": 7667,
+      "合议庭": 7668,
+      "▁ST": 7669,
+      "▁therefore": 7670,
+      "▁Del": 7671,
+      "▁delivery": 7672,
+      "▁mys": 7673,
+      "▁touch": 7674,
+      "▁partners": 7675,
+      "数是": 7676,
+      "▁listed": 7677,
+      "▁divided": 7678,
+      "nThanks": 7679,
+      "esh": 7680,
+      "营养": 7681,
+      "▁transl": 7682,
+      "消息": 7683,
+      "▁Wed": 7684,
+      "▁helps": 7685,
+      "▁giving": 7686,
+      "normal": 7687,
+      "ifying": 7688,
+      "▁Institute": 7689,
+      "patch": 7690,
+      "▁encour": 7691,
+      "Number": 7692,
+      "▁vict": 7693,
+      "rief": 7694,
+      "年轻": 7695,
+      "▁definit": 7696,
+      "现象": 7697,
+      "可是": 7698,
+      "Fil": 7699,
+      "cout": 7700,
+      "图片": 7701,
+      "异议": 7702,
+      "My": 7703,
+      "开放": 7704,
+      "坐标": 7705,
+      "▁attract": 7706,
+      "二〇一五年": 7707,
+      "props": 7708,
+      "▁Object": 7709,
+      "让人": 7710,
+      "\\\\\"": 7711,
+      "then": 7712,
+      "▁roll": 7713,
+      "arant": 7714,
+      "发展的": 7715,
+      "▁Nov": 7716,
+      "▁hous": 7717,
+      "▁Further": 7718,
+      "▁外文名": 7719,
+      "ston": 7720,
+      "▁characteristics": 7721,
+      "▁calc": 7722,
+      "▁mort": 7723,
+      "照片": 7724,
+      "ATH": 7725,
+      "▁sleep": 7726,
+      "phone": 7727,
+      "oyal": 7728,
+      "PR": 7729,
+      "▁schools": 7730,
+      "创业": 7731,
+      "▁连载平台": 7732,
+      "亿元": 7733,
+      "▁adjust": 7734,
+      "haps": 7735,
+      "▁中国": 7736,
+      "antic": 7737,
+      "▁sun": 7738,
+      "anges": 7739,
+      "asket": 7740,
+      "▁thanks": 7741,
+      "少年": 7742,
+      "基于": 7743,
+      "▁responsible": 7744,
+      "▁Sund": 7745,
+      "▁rapid": 7746,
+      "pan": 7747,
+      "option": 7748,
+      "▁define": 7749,
+      "▁Cam": 7750,
+      "▁sys": 7751,
+      "▁Two": 7752,
+      "Char": 7753,
+      "▁internal": 7754,
+      "江苏省": 7755,
+      "Server": 7756,
+      "▁growing": 7757,
+      "▁Prov": 7758,
+      "▁TH": 7759,
+      "range": 7760,
+      "logger": 7761,
+      "▁resistance": 7762,
+      "install": 7763,
+      "分类": 7764,
+      "该公司": 7765,
+      "etimes": 7766,
+      "isation": 7767,
+      "▁equations": 7768,
+      "成都": 7769,
+      "▁Mag": 7770,
+      "▁sports": 7771,
+      "会有": 7772,
+      "▁neur": 7773,
+      "ESS": 7774,
+      "而是": 7775,
+      "rim": 7776,
+      "元素": 7777,
+      "我觉得": 7778,
+      "▁analyz": 7779,
+      "▁excellent": 7780,
+      "▁occur": 7781,
+      "效率": 7782,
+      "仔细": 7783,
+      "▁phosph": 7784,
+      "▁刘": 7785,
+      "▁century": 7786,
+      "连续": 7787,
+      "遇到": 7788,
+      "▁Any": 7789,
+      "等等": 7790,
+      "▁android": 7791,
+      "abet": 7792,
+      "▁profile": 7793,
+      "attr": 7794,
+      "行动": 7795,
+      "▁eight": 7796,
+      "▁lib": 7797,
+      "▁Pal": 7798,
+      "数学": 7799,
+      "car": 7800,
+      "▁counter": 7801,
+      "igger": 7802,
+      "不了": 7803,
+      "值得": 7804,
+      "比例": 7805,
+      "mselves": 7806,
+      "安排": 7807,
+      "PT": 7808,
+      "最高人民法院关于适用": 7809,
+      "▁operations": 7810,
+      "重庆市": 7811,
+      "rete": 7812,
+      ")}": 7813,
+      "▁star": 7814,
+      "▁vent": 7815,
+      "▁reaction": 7816,
+      "▁React": 7817,
+      "▁piece": 7818,
+      "青年": 7819,
+      "▁Paul": 7820,
+      "▁trust": 7821,
+      "▁kin": 7822,
+      "▁stock": 7823,
+      "ributes": 7824,
+      "▁situation": 7825,
+      "items": 7826,
+      "构成": 7827,
+      "▁membr": 7828,
+      "nNow": 7829,
+      "力的": 7830,
+      "rical": 7831,
+      "icine": 7832,
+      "▁Date": 7833,
+      "广东省": 7834,
+      "▁Que": 7835,
+      "符合法律规定": 7836,
+      "▁alter": 7837,
+      "aud": 7838,
+      "▁pen": 7839,
+      "可得": 7840,
+      "▁\\\\(": 7841,
+      "Json": 7842,
+      "▁firm": 7843,
+      "arden": 7844,
+      "▁Ke": 7845,
+      "▁posted": 7846,
+      "aughter": 7847,
+      "根本": 7848,
+      "raise": 7849,
+      "进行了审理": 7850,
+      "▁adding": 7851,
+      "保全": 7852,
+      "北京市": 7853,
+      "▁approximately": 7854,
+      "nor": 7855,
+      "aser": 7856,
+      "ufeff": 7857,
+      "▁containing": 7858,
+      "委会": 7859,
+      "oday": 7860,
+      "▁combined": 7861,
+      "▁older": 7862,
+      "▁Run": 7863,
+      "幸福": 7864,
+      "工商": 7865,
+      "▁earlier": 7866,
+      "▁Green": 7867,
+      "主题": 7868,
+      "▁remains": 7869,
+      "van": 7870,
+      "▁morning": 7871,
+      "akers": 7872,
+      "▁themselves": 7873,
+      "ech": 7874,
+      "fa": 7875,
+      "ona": 7876,
+      "values": 7877,
+      "▁carry": 7878,
+      "▁German": 7879,
+      "的一个": 7880,
+      "ository": 7881,
+      "▁column": 7882,
+      "rey": 7883,
+      "▁strategies": 7884,
+      "吉林": 7885,
+      "▁dim": 7886,
+      "▁Club": 7887,
+      "的影响": 7888,
+      "icate": 7889,
+      "▁license": 7890,
+      "成果": 7891,
+      "球队": 7892,
+      "▁north": 7893,
+      "▁communication": 7894,
+      "法律文书": 7895,
+      "试验": 7896,
+      "▁race": 7897,
+      "▁supply": 7898,
+      "apeut": 7899,
+      "ded": 7900,
+      "ucture": 7901,
+      "▁comments": 7902,
+      "▁conducted": 7903,
+      "house": 7904,
+      "rian": 7905,
+      "▁Although": 7906,
+      "大多": 7907,
+      "▁credit": 7908,
+      "现实": 7909,
+      "^{\\\\": 7910,
+      "▁meaning": 7911,
+      "▁aware": 7912,
+      "▁activation": 7913,
+      "EMA": 7914,
+      "▁vehicle": 7915,
+      "▁errors": 7916,
+      "本身": 7917,
+      "apped": 7918,
+      "▁appears": 7919,
+      "连接": 7920,
+      "▁photos": 7921,
+      "▁lip": 7922,
+      "omial": 7923,
+      "urrenc": 7924,
+      "▁detection": 7925,
+      "▁Hall": 7926,
+      "▁django": 7927,
+      "▁boolean": 7928,
+      "porary": 7929,
+      "di": 7930,
+      "▁kids": 7931,
+      "ils": 7932,
+      "▁animals": 7933,
+      "▁Microsoft": 7934,
+      "▁subjects": 7935,
+      "▁Gover": 7936,
+      "蛋白": 7937,
+      "差异": 7938,
+      "▁insurance": 7939,
+      "▁south": 7940,
+      "原则": 7941,
+      ")/(": 7942,
+      ":\"": 7943,
+      "non": 7944,
+      "▁hour": 7945,
+      "算法": 7946,
+      "所在": 7947,
+      "FI": 7948,
+      "iveness": 7949,
+      "的工作": 7950,
+      "▁convert": 7951,
+      "完善": 7952,
+      "▁Check": 7953,
+      "▁genetic": 7954,
+      "民间借贷": 7955,
+      "▁chemical": 7956,
+      "▁parameter": 7957,
+      "send": 7958,
+      "▁investigate": 7959,
+      "证实": 7960,
+      "undred": 7961,
+      "▁tal": 7962,
+      "自治区": 7963,
+      "']": 7964,
+      "▁directory": 7965,
+      "AV": 7966,
+      "ker": 7967,
+      "机器": 7968,
+      "ULL": 7969,
+      "▁plus": 7970,
+      "背景": 7971,
+      "▁demand": 7972,
+      "改善": 7973,
+      "年代": 7974,
+      "▁sometimes": 7975,
+      "ategories": 7976,
+      "▁dark": 7977,
+      "Call": 7978,
+      "百度": 7979,
+      "rought": 7980,
+      "uman": 7981,
+      "▁economic": 7982,
+      "本题": 7983,
+      "CK": 7984,
+      "nPl": 7985,
+      "▁concentration": 7986,
+      "组成合议庭": 7987,
+      "▁AD": 7988,
+      "▁Rel": 7989,
+      "which": 7990,
+      "▁Sur": 7991,
+      "▁rank": 7992,
+      "overrightarrow": 7993,
+      "▁apprec": 7994,
+      "以为": 7995,
+      "▁stri": 7996,
+      "运营": 7997,
+      "作出的": 7998,
+      "▁ord": 7999,
+      "▁info": 8000,
+      "▁coach": 8001,
+      "女性": 8002,
+      "cp": 8003,
+      "欢迎": 8004,
+      "推广": 8005,
+      "▁tra": 8006,
+      "▁epis": 8007,
+      "事情": 8008,
+      "来源": 8009,
+      "▁pi": 8010,
+      "▁demonstrated": 8011,
+      "▁Most": 8012,
+      "Link": 8013,
+      "olved": 8014,
+      "▁Design": 8015,
+      "▁mm": 8016,
+      "inity": 8017,
+      "上海市": 8018,
+      "arts": 8019,
+      "pol": 8020,
+      "幼儿": 8021,
+      "▁gave": 8022,
+      "▁pages": 8023,
+      "▁{\\\\": 8024,
+      "eal": 8025,
+      "▁notice": 8026,
+      "的时间": 8027,
+      "▁smaller": 8028,
+      "▁rule": 8029,
+      "▁Bay": 8030,
+      "爱情": 8031,
+      "SI": 8032,
+      "illa": 8033,
+      "▁failure": 8034,
+      "▁float": 8035,
+      "Hz": 8036,
+      "我就": 8037,
+      "▁mg": 8038,
+      "▁supported": 8039,
+      "不用": 8040,
+      "dep": 8041,
+      "▁situ": 8042,
+      "▁consult": 8043,
+      "}/": 8044,
+      "▁bur": 8045,
+      "front": 8046,
+      "}^": 8047,
+      "▁Arch": 8048,
+      "abilities": 8049,
+      "hemical": 8050,
+      "></": 8051,
+      "gg": 8052,
+      "第一百五十": 8053,
+      "▁blue": 8054,
+      "十六": 8055,
+      "ette": 8056,
+      "▁className": 8057,
+      "mu": 8058,
+      "▁query": 8059,
+      "▁layer": 8060,
+      "毕竟": 8061,
+      "Pos": 8062,
+      "▁seven": 8063,
+      "▁guid": 8064,
+      "娱乐": 8065,
+      "内的": 8066,
+      "Options": 8067,
+      "医生": 8068,
+      "dx": 8069,
+      "▁agreed": 8070,
+      "▁objects": 8071,
+      "▁framework": 8072,
+      "▁surround": 8073,
+      "多种": 8074,
+      "很好": 8075,
+      "ella": 8076,
+      "▁amazing": 8077,
+      "▁surpr": 8078,
+      "▁White": 8079,
+      "不足": 8080,
+      "pop": 8081,
+      "▁decl": 8082,
+      "▁director": 8083,
+      "iginal": 8084,
+      "打造": 8085,
+      "▁reflect": 8086,
+      "▁extr": 8087,
+      "▁updated": 8088,
+      "cf": 8089,
+      "TER": 8090,
+      "▁决": 8091,
+      "▁floor": 8092,
+      "疫情": 8093,
+      "HOUT": 8094,
+      "LA": 8095,
+      "到底": 8096,
+      "Types": 8097,
+      "▁FOR": 8098,
+      "人民政府": 8099,
+      "颜色": 8100,
+      "▁concentr": 8101,
+      "aturday": 8102,
+      "广泛": 8103,
+      "onym": 8104,
+      "container": 8105,
+      "▁capital": 8106,
+      "▁returned": 8107,
+      "ipped": 8108,
+      "▁interaction": 8109,
+      "日向本院提出": 8110,
+      "jpg": 8111,
+      "党员": 8112,
+      "▁Us": 8113,
+      "cap": 8114,
+      "年龄": 8115,
+      "一切": 8116,
+      "imensional": 8117,
+      "grade": 8118,
+      "CM": 8119,
+      "▁yield": 8120,
+      "一个人": 8121,
+      "cy": 8122,
+      "she": 8123,
+      "物质": 8124,
+      "分子": 8125,
+      "scrib": 8126,
+      "选择题": 8127,
+      "ethyl": 8128,
+      "▁diagnosis": 8129,
+      "家的": 8130,
+      "主动": 8131,
+      "ASS": 8132,
+      "▁Science": 8133,
+      "▁imm": 8134,
+      "▁advanced": 8135,
+      "nMy": 8136,
+      "▁definition": 8137,
+      "faces": 8138,
+      "▁Type": 8139,
+      "▁election": 8140,
+      "▁finding": 8141,
+      "ili": 8142,
+      "▁investigated": 8143,
+      "▁fair": 8144,
+      "ctors": 8145,
+      "nFrom": 8146,
+      "▁failed": 8147,
+      "▁Facebook": 8148,
+      "Valid": 8149,
+      "▁trend": 8150,
+      "Provider": 8151,
+      "小区": 8152,
+      "ancing": 8153,
+      "▁diseases": 8154,
+      "他们的": 8155,
+      "▁radio": 8156,
+      "fully": 8157,
+      "▁eye": 8158,
+      "▁\"/": 8159,
+      "▁motor": 8160,
+      "Source": 8161,
+      "rics": 8162,
+      "概念": 8163,
+      "▁evaluation": 8164,
+      "(\"%": 8165,
+      "▁Read": 8166,
+      "pers": 8167,
+      "auto": 8168,
+      "学会": 8169,
+      "优化": 8170,
+      "▁nut": 8171,
+      "让我": 8172,
+      "address": 8173,
+      "Box": 8174,
+      "▁Gl": 8175,
+      "▁flu": 8176,
+      "▁alone": 8177,
+      "▁generation": 8178,
+      "表明": 8179,
+      "▁spread": 8180,
+      "包装": 8181,
+      "render": 8182,
+      "▁mother": 8183,
+      "号民事判决书": 8184,
+      "▁patterns": 8185,
+      "▁earn": 8186,
+      "▁remov": 8187,
+      "bstract": 8188,
+      "aset": 8189,
+      "▁exposure": 8190,
+      "structor": 8191,
+      "Point": 8192,
+      "wers": 8193,
+      "azon": 8194,
+      "▁Build": 8195,
+      "rs": 8196,
+      "内部": 8197,
+      "▁applicable": 8198,
+      "pository": 8199,
+      "▁Che": 8200,
+      "▁recorded": 8201,
+      "经审理": 8202,
+      "pub": 8203,
+      "▁returns": 8204,
+      "▁出生": 8205,
+      "final": 8206,
+      "border": 8207,
+      "▁origin": 8208,
+      "how": 8209,
+      "▁molec": 8210,
+      "ifications": 8211,
+      "▁Rich": 8212,
+      "▁benefit": 8213,
+      "▁emb": 8214,
+      "GE": 8215,
+      "▁outcomes": 8216,
+      "anagement": 8217,
+      "总经理": 8218,
+      "agram": 8219,
+      "aight": 8220,
+      "研发": 8221,
+      "nend": 8222,
+      "▁Monday": 8223,
+      "▁Washington": 8224,
+      "伤害": 8225,
+      "phi": 8226,
+      "\\'\\'": 8227,
+      "iform": 8228,
+      "reek": 8229,
+      "总结": 8230,
+      "▁thinking": 8231,
+      "习惯": 8232,
+      "▁chronic": 8233,
+      "dy": 8234,
+      "iately": 8235,
+      "▁stuff": 8236,
+      "▁Phys": 8237,
+      "PU": 8238,
+      "▁;": 8239,
+      "▁mit": 8240,
+      "▁college": 8241,
+      "▁wild": 8242,
+      "启动": 8243,
+      "▁bool": 8244,
+      "点的": 8245,
+      "据此": 8246,
+      "rapper": 8247,
+      "Det": 8248,
+      "▁Gold": 8249,
+      "system": 8250,
+      "▁metal": 8251,
+      "▁Texas": 8252,
+      "▁marketing": 8253,
+      "公开开庭": 8254,
+      "执字第": 8255,
+      "▁beginning": 8256,
+      "▁enhance": 8257,
+      "▁chain": 8258,
+      "▁protect": 8259,
+      "nin": 8260,
+      "ements": 8261,
+      "icity": 8262,
+      "unicip": 8263,
+      "▁experiments": 8264,
+      "▁James": 8265,
+      "题意": 8266,
+      "▁category": 8267,
+      "▁inject": 8268,
+      "plicit": 8269,
+      "▁Best": 8270,
+      "▁prices": 8271,
+      "*\\": 8272,
+      "nOur": 8273,
+      "▁Rad": 8274,
+      "▁characters": 8275,
+      "加上": 8276,
+      "▁integer": 8277,
+      "▁father": 8278,
+      "▁constit": 8279,
+      "site": 8280,
+      "▁approaches": 8281,
+      "anging": 8282,
+      "▁implied": 8283,
+      "IND": 8284,
+      "asis": 8285,
+      "▁favorite": 8286,
+      "aved": 8287,
+      "▁Road": 8288,
+      "▁Python": 8289,
+      "allel": 8290,
+      "这次": 8291,
+      "Instance": 8292,
+      "资本": 8293,
+      "▁math": 8294,
+      "▁Chinese": 8295,
+      "ulin": 8296,
+      "nAmerican": 8297,
+      "如果你": 8298,
+      "▁purchase": 8299,
+      "▁Party": 8300,
+      "\\\\((": 8301,
+      "gent": 8302,
+      "▁bottom": 8303,
+      "▁frame": 8304,
+      "▁render": 8305,
+      "有什么": 8306,
+      "▁court": 8307,
+      "▁bacter": 8308,
+      "▁Games": 8309,
+      "IA": 8310,
+      "HP": 8311,
+      "▁Sol": 8312,
+      "coh": 8313,
+      "▁~": 8314,
+      "相当": 8315,
+      "▁trade": 8316,
+      "英国": 8317,
+      "▁follows": 8318,
+      "session": 8319,
+      "读书": 8320,
+      "▁alternative": 8321,
+      "uesday": 8322,
+      "▁cast": 8323,
+      "[:": 8324,
+      "pet": 8325,
+      "▁HIV": 8326,
+      "vec": 8327,
+      "gest": 8328,
+      "▁File": 8329,
+      "▁separate": 8330,
+      "▁Mem": 8331,
+      "▁behavi": 8332,
+      "▁photo": 8333,
+      "induced": 8334,
+      "licenses": 8335,
+      "的方法": 8336,
+      "obert": 8337,
+      "▁((": 8338,
+      "plete": 8339,
+      "DIR": 8340,
+      "▁vel": 8341,
+      "anned": 8342,
+      "entity": 8343,
+      "日常": 8344,
+      "▁publish": 8345,
+      "高级": 8346,
+      "山市": 8347,
+      "所有的": 8348,
+      "▁dro": 8349,
+      "▁Lin": 8350,
+      "thlet": 8351,
+      "最大的": 8352,
+      "▁opened": 8353,
+      "▁News": 8354,
+      "人生": 8355,
+      "inder": 8356,
+      "BS": 8357,
+      "▁compliance": 8358,
+      "▁France": 8359,
+      "▁chrom": 8360,
+      "▁challenge": 8361,
+      "▁profession": 8362,
+      "tfor": 8363,
+      "▁minimum": 8364,
+      "after": 8365,
+      "riend": 8366,
+      "▁Base": 8367,
+      "在线": 8368,
+      "尺寸": 8369,
+      "click": 8370,
+      "ru": 8371,
+      "EMAIL": 8372,
+      "▁fab": 8373,
+      "▁Method": 8374,
+      "▁prop": 8375,
+      "▁huge": 8376,
+      "▁upper": 8377,
+      "nfunction": 8378,
+      "ootball": 8379,
+      "▁examples": 8380,
+      "▁probability": 8381,
+      "▁movie": 8382,
+      "两种": 8383,
+      "刑期": 8384,
+      "▁bone": 8385,
+      "▁exists": 8386,
+      "整数": 8387,
+      "▁repair": 8388,
+      "ora": 8389,
+      "anner": 8390,
+      "至少": 8391,
+      "▁sch": 8392,
+      "▁guide": 8393,
+      "二次": 8394,
+      "▁instr": 8395,
+      "的地方": 8396,
+      "ture": 8397,
+      "oration": 8398,
+      "component": 8399,
+      "角度": 8400,
+      "▁imag": 8401,
+      "save": 8402,
+      "▁smooth": 8403,
+      "▁Bre": 8404,
+      "ether": 8405,
+      "care": 8406,
+      "▁sale": 8407,
+      "能源": 8408,
+      "contin": 8409,
+      "hematics": 8410,
+      "给你": 8411,
+      "▁prepar": 8412,
+      "ORT": 8413,
+      "▁coding": 8414,
+      "uzz": 8415,
+      "▁equival": 8416,
+      "ameters": 8417,
+      "下列": 8418,
+      "▁Hot": 8419,
+      "▁lymph": 8420,
+      "足球": 8421,
+      "han": 8422,
+      "ocity": 8423,
+      "▁af": 8424,
+      "▁scen": 8425,
+      "ervation": 8426,
+      "▁sust": 8427,
+      "sey": 8428,
+      "▁administration": 8429,
+      "▁traff": 8430,
+      "▁glass": 8431,
+      "specific": 8432,
+      "Core": 8433,
+      "arp": 8434,
+      "是不是": 8435,
+      "探讨": 8436,
+      "落实": 8437,
+      "▁Lou": 8438,
+      "困难": 8439,
+      "▁mill": 8440,
+      "tim": 8441,
+      "广西": 8442,
+      "▁annual": 8443,
+      "▁soft": 8444,
+      "▁SO": 8445,
+      "▁restaur": 8446,
+      "asure": 8447,
+      "location": 8448,
+      "中学": 8449,
+      "estival": 8450,
+      "ATION": 8451,
+      "▁angle": 8452,
+      "▁Mart": 8453,
+      "going": 8454,
+      "▁replacement": 8455,
+      "损失": 8456,
+      "▁Church": 8457,
+      "nx": 8458,
+      "Em": 8459,
+      "▁easier": 8460,
+      "▁survey": 8461,
+      "中医": 8462,
+      "▁prove": 8463,
+      "▁Company": 8464,
+      "Case": 8465,
+      "▁Pen": 8466,
+      "刑法": 8467,
+      "判决如下": 8468,
+      "▁Africa": 8469,
+      "/*\\": 8470,
+      "▁straight": 8471,
+      "ursday": 8472,
+      "irmed": 8473,
+      "ications": 8474,
+      "▁occurre": 8475,
+      "iced": 8476,
+      "▁trial": 8477,
+      "的事情": 8478,
+      "金属": 8479,
+      "发现被执行人": 8480,
+      "一条": 8481,
+      "▁dry": 8482,
+      "▁symptoms": 8483,
+      "▁academ": 8484,
+      "▁Sal": 8485,
+      "▁door": 8486,
+      "▁template": 8487,
+      "然而": 8488,
+      "▁ISBN": 8489,
+      "▁finished": 8490,
+      "Null": 8491,
+      "vas": 8492,
+      "LC": 8493,
+      "example": 8494,
+      "Method": 8495,
+      "▁plays": 8496,
+      "一日": 8497,
+      "▁electron": 8498,
+      "▁saying": 8499,
+      "▁joint": 8500,
+      "▁mis": 8501,
+      "iny": 8502,
+      "带来": 8503,
+      "增强": 8504,
+      "▁thous": 8505,
+      "▁rare": 8506,
+      "▁Ca": 8507,
+      "▁employees": 8508,
+      "公诉机关": 8509,
+      "apers": 8510,
+      "▁hands": 8511,
+      "▁Thus": 8512,
+      "keys": 8513,
+      "▁connected": 8514,
+      "▁Network": 8515,
+      "$(": 8516,
+      "itivity": 8517,
+      "▁Sunday": 8518,
+      "粉丝": 8519,
+      "cat": 8520,
+      "▁generally": 8521,
+      "都在": 8522,
+      "状况": 8523,
+      "▁proof": 8524,
+      "DR": 8525,
+      "▁efficiency": 8526,
+      "▁indicate": 8527,
+      "ja": 8528,
+      "评论": 8529,
+      "第一百五十四": 8530,
+      "▁Mult": 8531,
+      "▁nan": 8532,
+      "components": 8533,
+      "▁contrast": 8534,
+      "担保": 8535,
+      "meta": 8536,
+      "▁{}\\": 8537,
+      "于是": 8538,
+      "▁debut": 8539,
+      "还要": 8540,
+      "▁bath": 8541,
+      "umns": 8542,
+      "年度": 8543,
+      "sign": 8544,
+      "▁wave": 8545,
+      "control": 8546,
+      "▁Summer": 8547,
+      "▁placed": 8548,
+      "▁wish": 8549,
+      "ashion": 8550,
+      "VID": 8551,
+      "▁smart": 8552,
+      "▁die": 8553,
+      "settings": 8554,
+      "nesday": 8555,
+      "afe": 8556,
+      "▁Mex": 8557,
+      "▁diet": 8558,
+      "itar": 8559,
+      "学术": 8560,
+      "展示": 8561,
+      "▁shared": 8562,
+      "法规": 8563,
+      "▁importance": 8564,
+      "▁association": 8565,
+      "tenance": 8566,
+      "▁muscle": 8567,
+      "oses": 8568,
+      "��": 8569,
+      "▁spent": 8570,
+      "品质": 8571,
+      "kin": 8572,
+      "准许原告": 8573,
+      "▁adult": 8574,
+      "game": 8575,
+      "▁switch": 8576,
+      "由被告": 8577,
+      "pc": 8578,
+      "▁grant": 8579,
+      "送达之日起": 8580,
+      "aa": 8581,
+      "istant": 8582,
+      "▁Central": 8583,
+      "就能": 8584,
+      "▁subsequ": 8585,
+      "SH": 8586,
+      "出口": 8587,
+      "▁literature": 8588,
+      "▁church": 8589,
+      "km": 8590,
+      "▁opportunities": 8591,
+      "公布": 8592,
+      "▁environmental": 8593,
+      "▁ideal": 8594,
+      "论文": 8595,
+      "当地": 8596,
+      "▁fight": 8597,
+      "▁Ber": 8598,
+      ":\\\\": 8599,
+      "ão": 8600,
+      ")^{": 8601,
+      "▁dead": 8602,
+      "▁joined": 8603,
+      "环保": 8604,
+      "诉讼代理人": 8605,
+      "aints": 8606,
+      "▁exception": 8607,
+      "▁configuration": 8608,
+      "▁Miss": 8609,
+      "▁bul": 8610,
+      "快乐": 8611,
+      "pling": 8612,
+      "▁hundred": 8613,
+      "一天": 8614,
+      "举办": 8615,
+      "诉被告": 8616,
+      "▁WITHOUT": 8617,
+      "▁experienced": 8618,
+      "▁Question": 8619,
+      "▁IT": 8620,
+      "网上": 8621,
+      "▁evaluated": 8622,
+      "nThat": 8623,
+      "nHere": 8624,
+      "▁resulting": 8625,
+      "▁passed": 8626,
+      "▁maybe": 8627,
+      "▁Apache": 8628,
+      "Hel": 8629,
+      "▁Year": 8630,
+      "很多人": 8631,
+      "本院在审理": 8632,
+      "tml": 8633,
+      "▁arriv": 8634,
+      "确保": 8635,
+      "encing": 8636,
+      "nRe": 8637,
+      "License": 8638,
+      "▁Elect": 8639,
+      "▁immediately": 8640,
+      "bour": 8641,
+      "lang": 8642,
+      "▁thread": 8643,
+      "▁Cong": 8644,
+      "▁push": 8645,
+      "▁Other": 8646,
+      "▁Each": 8647,
+      "▁fill": 8648,
+      "inger": 8649,
+      "▁introduced": 8650,
+      "FA": 8651,
+      "可供执行的财产": 8652,
+      "▁Leg": 8653,
+      "holder": 8654,
+      "PM": 8655,
+      "思考": 8656,
+      "办公室": 8657,
+      "▁actions": 8658,
+      "▁breast": 8659,
+      "arest": 8660,
+      "▁Jun": 8661,
+      "atabase": 8662,
+      "Api": 8663,
+      "`,": 8664,
+      "▁describe": 8665,
+      "Sec": 8666,
+      "▁regions": 8667,
+      "▁signed": 8668,
+      "SO": 8669,
+      "▁drugs": 8670,
+      "▁drop": 8671,
+      "cks": 8672,
+      "register": 8673,
+      "▁Saturday": 8674,
+      "▁proposal": 8675,
+      "Dec": 8676,
+      "ato": 8677,
+      "党委": 8678,
+      "uan": 8679,
+      "▁guess": 8680,
+      "是指": 8681,
+      "hab": 8682,
+      "▁extension": 8683,
+      "大会": 8684,
+      "抚养": 8685,
+      "说的": 8686,
+      "同意": 8687,
+      "▁competition": 8688,
+      "▁Fore": 8689,
+      "▁simpl": 8690,
+      "到庭参加诉讼": 8691,
+      "Repository": 8692,
+      "ialize": 8693,
+      "▁norm": 8694,
+      "osen": 8695,
+      "▁jour": 8696,
+      "▁出版": 8697,
+      "兴趣": 8698,
+      "出具": 8699,
+      "▁bug": 8700,
+      "▁therm": 8701,
+      "ensions": 8702,
+      "ducation": 8703,
+      "▁Flor": 8704,
+      "▁Engine": 8705,
+      "▁songs": 8706,
+      "式的": 8707,
+      "imary": 8708,
+      "▁Jack": 8709,
+      "aste": 8710,
+      "▁har": 8711,
+      "▁wife": 8712,
+      "动力": 8713,
+      "▁cos": 8714,
+      "利益": 8715,
+      "API": 8716,
+      "突然": 8717,
+      "街道": 8718,
+      "▁ple": 8719,
+      "▁vac": 8720,
+      "Ref": 8721,
+      "美元": 8722,
+      "▁rob": 8723,
+      "▁setup": 8724,
+      "Default": 8725,
+      "▁wid": 8726,
+      "委托诉讼代理人": 8727,
+      "▁Vis": 8728,
+      "cr": 8729,
+      "悔改": 8730,
+      "▁sett": 8731,
+      "iffer": 8732,
+      "成员": 8733,
+      "▁protocol": 8734,
+      "▁efforts": 8735,
+      "ailway": 8736,
+      "▁detect": 8737,
+      "某甲": 8738,
+      "Read": 8739,
+      "放在": 8740,
+      "▁nearly": 8741,
+      "▁dynamic": 8742,
+      "▁partner": 8743,
+      "▁imaging": 8744,
+      "合适": 8745,
+      "partial": 8746,
+      "▁propose": 8747,
+      "▁Long": 8748,
+      "不等": 8749,
+      "hash": 8750,
+      "到的": 8751,
+      "▁experiment": 8752,
+      "大部分": 8753,
+      "执行机关": 8754,
+      "yond": 8755,
+      "ita": 8756,
+      "ensus": 8757,
+      "Select": 8758,
+      "升级": 8759,
+      "▁开": 8760,
+      "食物": 8761,
+      "uns": 8762,
+      "来看": 8763,
+      "▁db": 8764,
+      "ownload": 8765,
+      "▁finally": 8766,
+      "▁mouse": 8767,
+      "▁\"^": 8768,
+      "▁journal": 8769,
+      "apeutic": 8770,
+      "expected": 8771,
+      "▁Create": 8772,
+      "常见": 8773,
+      "login": 8774,
+      "添加": 8775,
+      "look": 8776,
+      "▁assessment": 8777,
+      "▁directed": 8778,
+      "scale": 8779,
+      "董事长": 8780,
+      "▁procedure": 8781,
+      "GA": 8782,
+      "绿色": 8783,
+      "死亡": 8784,
+      "cor": 8785,
+      "重大": 8786,
+      "▁password": 8787,
+      "kwargs": 8788,
+      "uly": 8789,
+      "fields": 8790,
+      "买卖": 8791,
+      "▁edge": 8792,
+      "▁runs": 8793,
+      "anish": 8794,
+      "▁Bi": 8795,
+      "▁distinct": 8796,
+      "aneous": 8797,
+      "Phone": 8798,
+      "▁exercise": 8799,
+      "standing": 8800,
+      "递交": 8801,
+      "▁participants": 8802,
+      "表面": 8803,
+      "▁inflamm": 8804,
+      "科技有限公司": 8805,
+      "本院予以": 8806,
+      "▁movement": 8807,
+      "▁陈": 8808,
+      "▁Germany": 8809,
+      "▁discussed": 8810,
+      "▁View": 8811,
+      "olicy": 8812,
+      "第一次": 8813,
+      "eces": 8814,
+      "annels": 8815,
+      "▁analys": 8816,
+      "▁virus": 8817,
+      "▁Academ": 8818,
+      "angular": 8819,
+      "▁owners": 8820,
+      "▁African": 8821,
+      "公安局": 8822,
+      "lif": 8823,
+      "▁slight": 8824,
+      "通常": 8825,
+      "集中": 8826,
+      "▁pool": 8827,
+      "▁aims": 8828,
+      "▁dou": 8829,
+      "如不服本判决": 8830,
+      "▁Society": 8831,
+      "▁explore": 8832,
+      "广州": 8833,
+      "玩家": 8834,
+      "sql": 8835,
+      "job": 8836,
+      "▁myself": 8837,
+      "ean": 8838,
+      "ted": 8839,
+      "▁medium": 8840,
+      "itative": 8841,
+      "▁agents": 8842,
+      "lig": 8843,
+      "▁Court": 8844,
+      "west": 8845,
+      "▁priv": 8846,
+      "▁Id": 8847,
+      "▁military": 8848,
+      "emic": 8849,
+      "peed": 8850,
+      "的基础": 8851,
+      "▁Australian": 8852,
+      "苹果": 8853,
+      "对照": 8854,
+      "Ps": 8855,
+      "是个": 8856,
+      "men": 8857,
+      "nSub": 8858,
+      "▁Ev": 8859,
+      "▁hear": 8860,
+      "▁Award": 8861,
+      "▁tit": 8862,
+      "▁survival": 8863,
+      "▁NO": 8864,
+      "所谓": 8865,
+      "sort": 8866,
+      "▁During": 8867,
+      "▁manager": 8868,
+      "▁collected": 8869,
+      "▁cycle": 8870,
+      "ulf": 8871,
+      "有了": 8872,
+      "的主要": 8873,
+      "▁threat": 8874,
+      "职工": 8875,
+      "赔偿": 8876,
+      "▁Power": 8877,
+      "基地": 8878,
+      "asic": 8879,
+      "Sty": 8880,
+      "简称": 8881,
+      "▁Super": 8882,
+      "LI": 8883,
+      "greg": 8884,
+      "福建省": 8885,
+      "▁documentation": 8886,
+      "指标": 8887,
+      "科研": 8888,
+      "▁interactions": 8889,
+      "pression": 8890,
+      "▁Mount": 8891,
+      "强制执行": 8892,
+      "▁communities": 8893,
+      "曾经": 8894,
+      "real": 8895,
+      "▁gain": 8896,
+      "▁arter": 8897,
+      "心的": 8898,
+      "▁induced": 8899,
+      "▁spir": 8900,
+      "One": 8901,
+      "发出": 8902,
+      "确有": 8903,
+      "书面": 8904,
+      "▁increases": 8905,
+      "形象": 8906,
+      "诉讼请求": 8907,
+      "Check": 8908,
+      "▁authors": 8909,
+      "条第一款第": 8910,
+      "bert": 8911,
+      "asy": 8912,
+      "▁Democr": 8913,
+      "▁Top": 8914,
+      "▁vitro": 8915,
+      "家长": 8916,
+      "remove": 8917,
+      "*/\\": 8918,
+      "感染": 8919,
+      "▁protection": 8920,
+      "▁spaces": 8921,
+      "市人民法院": 8922,
+      "TA": 8923,
+      "症状": 8924,
+      "▁showing": 8925,
+      "▁farm": 8926,
+      "orry": 8927,
+      "每次": 8928,
+      "match": 8929,
+      "▁Main": 8930,
+      "也就是": 8931,
+      "ytic": 8932,
+      "▁plasma": 8933,
+      "();": 8934,
+      "sync": 8935,
+      "▁carried": 8936,
+      "▁lung": 8937,
+      "有一个": 8938,
+      "大概": 8939,
+      "是不": 8940,
+      "严格": 8941,
+      "空气": 8942,
+      "▁shoot": 8943,
+      "▁continuous": 8944,
+      "\\']\\": 8945,
+      "▁beyond": 8946,
+      "的女": 8947,
+      "▁Internet": 8948,
+      "▁Therefore": 8949,
+      "▁tox": 8950,
+      "就业": 8951,
+      "博士": 8952,
+      "▁examined": 8953,
+      "operative": 8954,
+      "者的": 8955,
+      "局长": 8956,
+      "UD": 8957,
+      "▁Mary": 8958,
+      "xi": 8959,
+      "track": 8960,
+      "▁Town": 8961,
+      "allery": 8962,
+      "dk": 8963,
+      "▁records": 8964,
+      "▁videos": 8965,
+      "蒙古": 8966,
+      "▁animal": 8967,
+      "bool": 8968,
+      "oster": 8969,
+      "secon": 8970,
+      "最好": 8971,
+      "elif": 8972,
+      "▁seed": 8973,
+      "ALL": 8974,
+      "▁nation": 8975,
+      "▁dose": 8976,
+      "▁Kore": 8977,
+      "学科": 8978,
+      "▁vivo": 8979,
+      "▁route": 8980,
+      "沟通": 8981,
+      "ounce": 8982,
+      "▁incorpor": 8983,
+      "药物": 8984,
+      "ometric": 8985,
+      "宝宝": 8986,
+      "oted": 8987,
+      "▁await": 8988,
+      "▁Big": 8989,
+      "▁Island": 8990,
+      "osh": 8991,
+      "▁hypothesis": 8992,
+      "▁pharm": 8993,
+      "nPlease": 8994,
+      "▁Pet": 8995,
+      "去了": 8996,
+      "命题": 8997,
+      "悔改表现": 8998,
+      "Start": 8999,
+      "oj": 9000,
+      "▁nav": 9001,
+      "tan": 9002,
+      "umni": 9003,
+      "▁degrees": 9004,
+      "▁challenges": 9005,
+      "quad": 9006,
+      "看着": 9007,
+      "▁expressed": 9008,
+      "▁PC": 9009,
+      "驳回": 9010,
+      "▁continued": 9011,
+      "▁strength": 9012,
+      "▁Pe": 9013,
+      "▁Text": 9014,
+      "rb": 9015,
+      "支行": 9016,
+      "▁emot": 9017,
+      "modules": 9018,
+      "cache": 9019,
+      "享受": 9020,
+      "传播": 9021,
+      "▁limitations": 9022,
+      "▁warm": 9023,
+      "Eng": 9024,
+      "ola": 9025,
+      "▁Rob": 9026,
+      "▁none": 9027,
+      "▁bond": 9028,
+      "提供的": 9029,
+      "▁resolution": 9030,
+      "(){\\": 9031,
+      "▁Union": 9032,
+      "TY": 9033,
+      "编辑": 9034,
+      "▁formed": 9035,
+      "▁uns": 9036,
+      "iler": 9037,
+      "nament": 9038,
+      "nOne": 9039,
+      "管理局": 9040,
+      "ón": 9041,
+      "End": 9042,
+      "▁ing": 9043,
+      "▁Results": 9044,
+      "▁rats": 9045,
+      "经审理查明": 9046,
+      "debug": 9047,
+      "▁married": 9048,
+      "vari": 9049,
+      "resents": 9050,
+      "▁reasons": 9051,
+      "ror": 9052,
+      "不管": 9053,
+      "tthis": 9054,
+      "inated": 9055,
+      "城区": 9056,
+      "well": 9057,
+      "ostic": 9058,
+      "▁launch": 9059,
+      "▁agent": 9060,
+      "ellig": 9061,
+      "cohol": 9062,
+      "▁infr": 9063,
+      "▁recip": 9064,
+      "FC": 9065,
+      "▁installed": 9066,
+      "度的": 9067,
+      "举行": 9068,
+      "▁Oh": 9069,
+      "ancer": 9070,
+      "▁letter": 9071,
+      "Script": 9072,
+      "mut": 9073,
+      "▁technical": 9074,
+      "iar": 9075,
+      "▁empty": 9076,
+      ">(": 9077,
+      "▁conv": 9078,
+      "anta": 9079,
+      "▁ign": 9080,
+      "另一": 9081,
+      "▁explain": 9082,
+      "▁compre": 9083,
+      "WA": 9084,
+      "期间的": 9085,
+      "▁pregn": 9086,
+      "此外": 9087,
+      "DIT": 9088,
+      "▁Services": 9089,
+      "▁operator": 9090,
+      "▁wed": 9091,
+      "▁symbol": 9092,
+      "abin": 9093,
+      "▁guarant": 9094,
+      "▁minor": 9095,
+      "▁Show": 9096,
+      "▁budget": 9097,
+      "婚姻": 9098,
+      "动车": 9099,
+      "革命": 9100,
+      "▁Business": 9101,
+      "歌曲": 9102,
+      "▁Its": 9103,
+      "draw": 9104,
+      "mid": 9105,
+      "▁Key": 9106,
+      "▁unknown": 9107,
+      "▁compare": 9108,
+      "公示": 9109,
+      "▁liver": 9110,
+      "▁corresponding": 9111,
+      "▁delay": 9112,
+      "▁Michael": 9113,
+      "▁sin": 9114,
+      "广大": 9115,
+      "elines": 9116,
+      "装修": 9117,
+      "inite": 9118,
+      "民间借贷纠纷一案": 9119,
+      "Sim": 9120,
+      "的人数": 9121,
+      "▁becomes": 9122,
+      "ura": 9123,
+      "▁affected": 9124,
+      "ROM": 9125,
+      "循环": 9126,
+      "多个": 9127,
+      "下降": 9128,
+      "八年": 9129,
+      "▁branch": 9130,
+      "eli": 9131,
+      "▁AR": 9132,
+      "中华人民共和国刑法": 9133,
+      "▁instruction": 9134,
+      "plements": 9135,
+      "▁Sam": 9136,
+      "▁Development": 9137,
+      "Post": 9138,
+      "文字": 9139,
+      "#####": 9140,
+      "微博": 9141,
+      "▁businesses": 9142,
+      "required": 9143,
+      "生育": 9144,
+      "▁pictures": 9145,
+      "几乎": 9146,
+      "road": 9147,
+      "outer": 9148,
+      "▁deb": 9149,
+      "评估": 9150,
+      "选出": 9151,
+      "direct": 9152,
+      "Fl": 9153,
+      "英雄": 9154,
+      "▁Line": 9155,
+      "▁decreased": 9156,
+      "▁Good": 9157,
+      "ORD": 9158,
+      "搜索": 9159,
+      "erman": 9160,
+      "plan": 9161,
+      "▁statist": 9162,
+      "language": 9163,
+      "是因为": 9164,
+      "azine": 9165,
+      "▁pet": 9166,
+      "▁neut": 9167,
+      "▁stories": 9168,
+      "ustr": 9169,
+      "pay": 9170,
+      ")\\\\)": 9171,
+      "▁offered": 9172,
+      "▁[\"": 9173,
+      "▁border": 9174,
+      "shire": 9175,
+      "▁GNU": 9176,
+      "▁networks": 9177,
+      "良好的": 9178,
+      "应予": 9179,
+      "blog": 9180,
+      "Red": 9181,
+      "▁controls": 9182,
+      "▁appeared": 9183,
+      "大量": 9184,
+      "的好": 9185,
+      "上了": 9186,
+      "total": 9187,
+      "高校": 9188,
+      "具备": 9189,
+      "▁Board": 9190,
+      "▁internet": 9191,
+      "▁investment": 9192,
+      "贵州": 9193,
+      "向本院申请": 9194,
+      "▁Jer": 9195,
+      "▁dress": 9196,
+      "▁finish": 9197,
+      "del": 9198,
+      "主席": 9199,
+      "▁feedback": 9200,
+      "blob": 9201,
+      "plugin": 9202,
+      "▁Franc": 9203,
+      "▁external": 9204,
+      "▁fish": 9205,
+      "接触": 9206,
+      "iences": 9207,
+      "▁adults": 9208,
+      "自我": 9209,
+      "▁workers": 9210,
+      "生长": 9211,
+      "boot": 9212,
+      "agen": 9213,
+      "▁furn": 9214,
+      "▁选项": 9215,
+      "▁[\\'": 9216,
+      "▁examine": 9217,
+      "在服刑": 9218,
+      "神经": 9219,
+      "▁elev": 9220,
+      "BO": 9221,
+      "▁argument": 9222,
+      "▁Angel": 9223,
+      "导演": 9224,
+      "唯一": 9225,
+      "对其": 9226,
+      "▁(\"": 9227,
+      "▁opening": 9228,
+      "▁Step": 9229,
+      "▁Tuesday": 9230,
+      "WS": 9231,
+      "roke": 9232,
+      "▁investigation": 9233,
+      "▁department": 9234,
+      "Bl": 9235,
+      "▁detail": 9236,
+      "refix": 9237,
+      "ups": 9238,
+      "\\':\\": 9239,
+      "▁Att": 9240,
+      "挑战": 9241,
+      "▁Sy": 9242,
+      "eries": 9243,
+      "iture": 9244,
+      "▁shot": 9245,
+      "inates": 9246,
+      "▁sides": 9247,
+      "▁du": 9248,
+      "▁mil": 9249,
+      "通知书": 9250,
+      "男人": 9251,
+      "▁removed": 9252,
+      "Open": 9253,
+      "users": 9254,
+      "nSh": 9255,
+      "▁demonstrate": 9256,
+      "会计": 9257,
+      "connect": 9258,
+      "ignment": 9259,
+      "▁wind": 9260,
+      "irt": 9261,
+      "amil": 9262,
+      "▁coeff": 9263,
+      "cmd": 9264,
+      "ander": 9265,
+      "roc": 9266,
+      "▁insert": 9267,
+      "▁immune": 9268,
+      "▁camera": 9269,
+      "线索": 9270,
+      "▁Second": 9271,
+      "▁Sl": 9272,
+      "ulty": 9273,
+      "▁pan": 9274,
+      "公路": 9275,
+      "aks": 9276,
+      "ilon": 9277,
+      "西安": 9278,
+      "一定的": 9279,
+      "▁fol": 9280,
+      "▁Ann": 9281,
+      "dr": 9282,
+      "▁Carol": 9283,
+      "esome": 9284,
+      "下午": 9285,
+      "给我": 9286,
+      "▁confirm": 9287,
+      "▁objective": 9288,
+      "should": 9289,
+      "▁interpret": 9290,
+      "数列": 9291,
+      "▁brought": 9292,
+      "ums": 9293,
+      "▁Thank": 9294,
+      "LICENSE": 9295,
+      "▁spot": 9296,
+      "▁seconds": 9297,
+      "▁regulation": 9298,
+      "本院在审理原告": 9299,
+      "ipping": 9300,
+      "edy": 9301,
+      "Resource": 9302,
+      "月份": 9303,
+      "▁comparison": 9304,
+      "向原告": 9305,
+      "▁letters": 9306,
+      "bal": 9307,
+      "受理后": 9308,
+      "▁出版社": 9309,
+      "▁acute": 9310,
+      "ened": 9311,
+      "Integer": 9312,
+      "asketball": 9313,
+      "供应": 9314,
+      "终结本次执行程序": 9315,
+      "▁Olympics": 9316,
+      "aming": 9317,
+      "考核": 9318,
+      "放弃": 9319,
+      "ena": 9320,
+      "含量": 9321,
+      "▁changing": 9322,
+      "▁advantage": 9323,
+      "▁plants": 9324,
+      "▁articles": 9325,
+      "mathcal": 9326,
+      "tn": 9327,
+      "▁integral": 9328,
+      "before": 9329,
+      "重视": 9330,
+      "▁Tur": 9331,
+      "ACK": 9332,
+      "▁cit": 9333,
+      "activ": 9334,
+      "step": 9335,
+      "▁leads": 9336,
+      "管辖": 9337,
+      "ursor": 9338,
+      "▁grand": 9339,
+      "做出": 9340,
+      "allow": 9341,
+      "▁income": 9342,
+      "ologic": 9343,
+      "愿意": 9344,
+      "更好": 9345,
+      "▁aspects": 9346,
+      "▁Bank": 9347,
+      "▁async": 9348,
+      "▁heavy": 9349,
+      "色的": 9350,
+      "在一起": 9351,
+      "种植": 9352,
+      "感受": 9353,
+      "iro": 9354,
+      "▁ring": 9355,
+      "▁offering": 9356,
+      "Project": 9357,
+      "*(": 9358,
+      "第一百四十五条第一款": 9359,
+      "▁magn": 9360,
+      "PRO": 9361,
+      "▁catal": 9362,
+      "EE": 9363,
+      "nQ": 9364,
+      "Per": 9365,
+      "ye": 9366,
+      "agment": 9367,
+      "Edit": 9368,
+      "amine": 9369,
+      "▁synd": 9370,
+      "Pass": 9371,
+      "▁trip": 9372,
+      "fill": 9373,
+      "random": 9374,
+      "▁liqu": 9375,
+      "▁carbon": 9376,
+      "▁Thursday": 9377,
+      "电池": 9378,
+      "▁confirmed": 9379,
+      "▁Code": 9380,
+      "▁beta": 9381,
+      "▁Av": 9382,
+      "组合": 9383,
+      "▁consequ": 9384,
+      "▁Lake": 9385,
+      "▁attend": 9386,
+      "▁operating": 9387,
+      "哈哈": 9388,
+      "河北省": 9389,
+      "▁van": 9390,
+      "nCon": 9391,
+      "▁driver": 9392,
+      "▁Equ": 9393,
+      "vol": 9394,
+      "方面的": 9395,
+      "节目": 9396,
+      "▁kitchen": 9397,
+      "▁drink": 9398,
+      "▁topic": 9399,
+      "做到": 9400,
+      "▁specified": 9401,
+      "________": 9402,
+      "▁prepared": 9403,
+      "▁Mad": 9404,
+      "▁Western": 9405,
+      "category": 9406,
+      "▁constr": 9407,
+      "cloud": 9408,
+      "趋势": 9409,
+      "▁Every": 9410,
+      "▁causes": 9411,
+      "会员": 9412,
+      "配合": 9413,
+      "cut": 9414,
+      "固定": 9415,
+      "故选": 9416,
+      "▁metabol": 9417,
+      "经典": 9418,
+      "▁focused": 9419,
+      "▁的": 9420,
+      "venue": 9421,
+      "▁footballers": 9422,
+      "ying": 9423,
+      "▁Call": 9424,
+      "▁Color": 9425,
+      "iling": 9426,
+      "dt": 9427,
+      "▁Tor": 9428,
+      "▁rent": 9429,
+      "▁membrane": 9430,
+      "(),\\": 9431,
+      "article": 9432,
+      "素质": 9433,
+      "▁successfully": 9434,
+      "声音": 9435,
+      "▁Music": 9436,
+      "本金": 9437,
+      "unte": 9438,
+      "fast": 9439,
+      "网络小说": 9440,
+      "▁relationships": 9441,
+      "▁Ob": 9442,
+      "▁editor": 9443,
+      "////////": 9444,
+      "▁eyes": 9445,
+      "▁enable": 9446,
+      "▁weekend": 9447,
+      "▁outcome": 9448,
+      "▁detailed": 9449,
+      "▁wants": 9450,
+      "brid": 9451,
+      "tension": 9452,
+      "ogenesis": 9453,
+      "entry": 9454,
+      "zen": 9455,
+      "▁suggested": 9456,
+      ")),\\": 9457,
+      "cribe": 9458,
+      "态度": 9459,
+      "投入": 9460,
+      "优质": 9461,
+      "那种": 9462,
+      "吸引": 9463,
+      ")$\\": 9464,
+      "▁alt": 9465,
+      "▁主要": 9466,
+      "Update": 9467,
+      "准确": 9468,
+      "▁cas": 9469,
+      "▁wasn": 9470,
+      "▁Conn": 9471,
+      "经本院": 9472,
+      "▁Wednesday": 9473,
+      "ingu": 9474,
+      "的方式": 9475,
+      "rie": 9476,
+      "怎样": 9477,
+      "estamp": 9478,
+      "装饰": 9479,
+      "▁advance": 9480,
+      "的一种": 9481,
+      "BU": 9482,
+      "sect": 9483,
+      "▁transition": 9484,
+      "▁Roman": 9485,
+      "实力": 9486,
+      "orter": 9487,
+      "▁Office": 9488,
+      "▁Tour": 9489,
+      "▁cloud": 9490,
+      "研究生": 9491,
+      "romise": 9492,
+      "▁Management": 9493,
+      "othing": 9494,
+      "追求": 9495,
+      "django": 9496,
+      "您的": 9497,
+      "▁feet": 9498,
+      "认罪": 9499,
+      "逐渐": 9500,
+      "▁maintain": 9501,
+      "例如": 9502,
+      "湖南省": 9503,
+      "▁batter": 9504,
+      "tprivate": 9505,
+      "物流": 9506,
+      "oul": 9507,
+      "▁requests": 9508,
+      "▁Press": 9509,
+      "▁traffic": 9510,
+      "▁Requ": 9511,
+      "springframework": 9512,
+      "▁measurements": 9513,
+      "anal": 9514,
+      "我也": 9515,
+      "nHistory": 9516,
+      "号刑事": 9517,
+      "▁Ver": 9518,
+      "inant": 9519,
+      "与被执行人": 9520,
+      "▁covered": 9521,
+      "task": 9522,
+      "arc": 9523,
+      "▁Cle": 9524,
+      "上市": 9525,
+      "资格": 9526,
+      "quare": 9527,
+      "高速": 9528,
+      "▁thick": 9529,
+      "让你": 9530,
+      "found": 9531,
+      "▁paid": 9532,
+      "▁出版时间": 9533,
+      "lets": 9534,
+      "▁Mass": 9535,
+      "▁biggest": 9536,
+      "▁severe": 9537,
+      "收到": 9538,
+      "▁Comput": 9539,
+      "▁otherwise": 9540,
+      "▁powerful": 9541,
+      "最小": 9542,
+      "公司的": 9543,
+      "Net": 9544,
+      "tm": 9545,
+      "去年": 9546,
+      "%\\": 9547,
+      "▁Tom": 9548,
+      "▁artist": 9549,
+      "永远": 9550,
+      "asion": 9551,
+      "▁Dev": 9552,
+      "dem": 9553,
+      "▁Sil": 9554,
+      "认证": 9555,
+      "clusive": 9556,
+      "▁URL": 9557,
+      "称为": 9558,
+      "▁onto": 9559,
+      "nergy": 9560,
+      "▁£": 9561,
+      "指数": 9562,
+      "Fe": 9563,
+      "污染": 9564,
+      "自行": 9565,
+      "▁pip": 9566,
+      "▁unc": 9567,
+      "传奇": 9568,
+      "▁cellular": 9569,
+      "美术": 9570,
+      "geq": 9571,
+      "iller": 9572,
+      "ooks": 9573,
+      "原理": 9574,
+      "工作的": 9575,
+      "points": 9576,
+      "房产": 9577,
+      "▁▁▁▁▁": 9578,
+      "▁游戏": 9579,
+      "指出": 9580,
+      "▁kid": 9581,
+      "mathbf": 9582,
+      "osa": 9583,
+      ")$$": 9584,
+      "名单": 9585,
+      "检验": 9586,
+      "Acc": 9587,
+      "▁specifically": 9588,
+      "▁token": 9589,
+      "OK": 9590,
+      "简易": 9591,
+      "mg": 9592,
+      "riz": 9593,
+      "▁George": 9594,
+      "▁derived": 9595,
+      "▁plate": 9596,
+      "cur": 9597,
+      "▁calls": 9598,
+      "提供了": 9599,
+      "yan": 9600,
+      "护理": 9601,
+      "获取": 9602,
+      "▁\"\"": 9603,
+      "豆瓣": 9604,
+      "▁Canadian": 9605,
+      "▁alpha": 9606,
+      "▁Well": 9607,
+      "▁Commit": 9608,
+      "▁Scott": 9609,
+      "多年": 9610,
+      "缴纳": 9611,
+      "我想": 9612,
+      "▁ice": 9613,
+      "access": 9614,
+      "▁majority": 9615,
+      "▁Why": 9616,
+      "▁extremely": 9617,
+      "big": 9618,
+      "▁exchange": 9619,
+      "▁comprehens": 9620,
+      "ansion": 9621,
+      "▁gar": 9622,
+      "ictor": 9623,
+      "▁automatically": 9624,
+      "有着": 9625,
+      "▁Both": 9626,
+      "主持": 9627,
+      "burg": 9628,
+      "iant": 9629,
+      "▁Exper": 9630,
+      "格式": 9631,
+      "Work": 9632,
+      "power": 9633,
+      "Val": 9634,
+      "就不": 9635,
+      "广场": 9636,
+      "▁obvious": 9637,
+      "▁daughter": 9638,
+      "KEY": 9639,
+      "▁WARRANTIES": 9640,
+      "一位": 9641,
+      "修改": 9642,
+      "throw": 9643,
+      "▁theme": 9644,
+      "官方": 9645,
+      "▁Oper": 9646,
+      "校园": 9647,
+      "▁owner": 9648,
+      "▁%}\\": 9649,
+      ">:": 9650,
+      "frame": 9651,
+      "▁cash": 9652,
+      "▁Division": 9653,
+      "mor": 9654,
+      "▁conference": 9655,
+      "出版的": 9656,
+      "agement": 9657,
+      "▁corpor": 9658,
+      "umer": 9659,
+      "||": 9660,
+      "PP": 9661,
+      "▁albums": 9662,
+      "aml": 9663,
+      "urrency": 9664,
+      "ali": 9665,
+      "▁facilit": 9666,
+      "批准": 9667,
+      "他人": 9668,
+      "nto": 9669,
+      "replace": 9670,
+      "名的": 9671,
+      "▁definitely": 9672,
+      "有哪些": 9673,
+      "lab": 9674,
+      "▁Under": 9675,
+      "ady": 9676,
+      "uce": 9677,
+      "▁ven": 9678,
+      "▁turned": 9679,
+      "反映": 9680,
+      "▁accuracy": 9681,
+      "fol": 9682,
+      "▁国": 9683,
+      "爱的": 9684,
+      "▁elim": 9685,
+      "▁achieved": 9686,
+      "”“": 9687,
+      "▁maintenance": 9688,
+      "▁balance": 9689,
+      "▁deploy": 9690,
+      "上面": 9691,
+      "scri": 9692,
+      "欧洲": 9693,
+      "ims": 9694,
+      "▁advice": 9695,
+      "ocytes": 9696,
+      "▁advert": 9697,
+      "▁integration": 9698,
+      "美的": 9699,
+      "oder": 9700,
+      "至于": 9701,
+      "nNo": 9702,
+      "▁strict": 9703,
+      "▁spend": 9704,
+      "被害": 9705,
+      "nCan": 9706,
+      "avis": 9707,
+      "pdf": 9708,
+      "总部": 9709,
+      "▁hab": 9710,
+      "▁heard": 9711,
+      "rint": 9712,
+      "正当": 9713,
+      "inn": 9714,
+      "担任": 9715,
+      "nSup": 9716,
+      "晚上": 9717,
+      "▁concentrations": 9718,
+      "▁represents": 9719,
+      "好了": 9720,
+      "如今": 9721,
+      "ami": 9722,
+      "▁Women": 9723,
+      "eck": 9724,
+      "MC": 9725,
+      "▁billion": 9726,
+      "▁relax": 9727,
+      "▁mixed": 9728,
+      "多了": 9729,
+      "▁restrict": 9730,
+      "▁IL": 9731,
+      "ynomial": 9732,
+      "]]": 9733,
+      "▁Cap": 9734,
+      "rot": 9735,
+      "确定的义务": 9736,
+      "double": 9737,
+      "▁morph": 9738,
+      "的不": 9739,
+      "▁emerg": 9740,
+      "(_": 9741,
+      "▁permissions": 9742,
+      "enter": 9743,
+      "▁Florida": 9744,
+      "职务": 9745,
+      "ett": 9746,
+      "essions": 9747,
+      "▁organizations": 9748,
+      "abetes": 9749,
+      "▁therapeutic": 9750,
+      "ested": 9751,
+      "▁William": 9752,
+      "此次": 9753,
+      "**(": 9754,
+      "TC": 9755,
+      "递交上诉状": 9756,
+      "河南": 9757,
+      "性和": 9758,
+      "▁harm": 9759,
+      "太阳": 9760,
+      "▁split": 9761,
+      "发行": 9762,
+      "同样": 9763,
+      "▁Japanese": 9764,
+      "所属": 9765,
+      "公园": 9766,
+      "▁Many": 9767,
+      "▁Sep": 9768,
+      "都能": 9769,
+      "Task": 9770,
+      "▁wire": 9771,
+      "oded": 9772,
+      "司法": 9773,
+      "Listener": 9774,
+      "▁candidate": 9775,
+      "德国": 9776,
+      "▁involve": 9777,
+      "DI": 9778,
+      "▁weather": 9779,
+      "iders": 9780,
+      "▁motion": 9781,
+      "avour": 9782,
+      "linear": 9783,
+      "VE": 9784,
+      "Pre": 9785,
+      "适应": 9786,
+      "▁leader": 9787,
+      "Sp": 9788,
+      "println": 9789,
+      "各项": 9790,
+      "▁^": 9791,
+      "atever": 9792,
+      "mic": 9793,
+      "即使": 9794,
+      "家里": 9795,
+      "选项中": 9796,
+      "ipl": 9797,
+      "▁decor": 9798,
+      "▁respond": 9799,
+      "ventional": 9800,
+      "高度": 9801,
+      "相比": 9802,
+      "▁analyzed": 9803,
+      "总是": 9804,
+      "▁agreement": 9805,
+      "▁accel": 9806,
+      "实行": 9807,
+      "▁manage": 9808,
+      "▁ped": 9809,
+      "▁girls": 9810,
+      "▁Robert": 9811,
+      "night": 9812,
+      "▁belong": 9813,
+      "Thread": 9814,
+      "▁nearest": 9815,
+      "▁plot": 9816,
+      "door": 9817,
+      "▁afford": 9818,
+      "▁dedicated": 9819,
+      "How": 9820,
+      "想到": 9821,
+      "遵守": 9822,
+      "▁Mont": 9823,
+      "ellular": 9824,
+      "▁Sun": 9825,
+      "VD": 9826,
+      "FO": 9827,
+      "贡献": 9828,
+      "▁font": 9829,
+      "我不": 9830,
+      "▁practices": 9831,
+      "▁Child": 9832,
+      "▁conven": 9833,
+      "udo": 9834,
+      "\\\\)\\": 9835,
+      "▁Christmas": 9836,
+      "MI": 9837,
+      "▁improvement": 9838,
+      "▁Gen": 9839,
+      "一段": 9840,
+      "{}": 9841,
+      "手段": 9842,
+      "edit": 9843,
+      "提示": 9844,
+      "▁decrease": 9845,
+      "make": 9846,
+      "▁Because": 9847,
+      "▁stable": 9848,
+      "相互": 9849,
+      "村民": 9850,
+      "其它": 9851,
+      "clear": 9852,
+      "战争": 9853,
+      "▁Alex": 9854,
+      "▁estimated": 9855,
+      "▁scheme": 9856,
+      "▁messages": 9857,
+      "玻璃": 9858,
+      "Click": 9859,
+      "▁thank": 9860,
+      "&#": 9861,
+      "custom": 9862,
+      "Back": 9863,
+      "nDes": 9864,
+      "▁native": 9865,
+      "▁driving": 9866,
+      "Filter": 9867,
+      "▁interval": 9868,
+      "罗斯": 9869,
+      "hr": 9870,
+      "▁doll": 9871,
+      "▁Profess": 9872,
+      "▁finite": 9873,
+      "▁);\\": 9874,
+      "确有悔改表现": 9875,
+      "▁AS": 9876,
+      "▁Will": 9877,
+      "▁{}": 9878,
+      "names": 9879,
+      "▁Grand": 9880,
+      "流程": 9881,
+      "▁schedul": 9882,
+      "gorithms": 9883,
+      "▁alcohol": 9884,
+      "aked": 9885,
+      "endif": 9886,
+      "▁reviews": 9887,
+      "uted": 9888,
+      "才是": 9889,
+      "Function": 9890,
+      "上诉于": 9891,
+      "▁Russian": 9892,
+      "acks": 9893,
+      "各类": 9894,
+      "▁DO": 9895,
+      "……\\": 9896,
+      "▁Inf": 9897,
+      "▁delete": 9898,
+      "Logger": 9899,
+      "就像": 9900,
+      "anning": 9901,
+      "iological": 9902,
+      "最新": 9903,
+      "Application": 9904,
+      "▁reached": 9905,
+      "▁identity": 9906,
+      "razil": 9907,
+      "系统的": 9908,
+      "iled": 9909,
+      "▁calculate": 9910,
+      "很难": 9911,
+      "AGE": 9912,
+      "管理有限公司": 9913,
+      "过的": 9914,
+      "dated": 9915,
+      "oi": 9916,
+      "辽宁": 9917,
+      "▁Make": 9918,
+      "▁tend": 9919,
+      "变得": 9920,
+      "志愿": 9921,
+      "信号": 9922,
+      "▁Apple": 9923,
+      "对方当事人": 9924,
+      "opath": 9925,
+      "▁MA": 9926,
+      "▁teaching": 9927,
+      "▁virtual": 9928,
+      "▁himself": 9929,
+      "pa": 9930,
+      "element": 9931,
+      "stop": 9932,
+      "lam": 9933,
+      "▁mission": 9934,
+      "unately": 9935,
+      "变成": 9936,
+      "shop": 9937,
+      "▁structural": 9938,
+      "▁pal": 9939,
+      "onymous": 9940,
+      "离开": 9941,
+      "▁ON": 9942,
+      "哪里": 9943,
+      "▁happens": 9944,
+      "▁End": 9945,
+      "▁relation": 9946,
+      "最佳": 9947,
+      "▁contribute": 9948,
+      "▁facilities": 9949,
+      "▁secure": 9950,
+      "▁hotel": 9951,
+      "LS": 9952,
+      "▁serum": 9953,
+      "▁represented": 9954,
+      "大陆": 9955,
+      "INE": 9956,
+      "结案": 9957,
+      "nis": 9958,
+      "扩大": 9959,
+      "▁diss": 9960,
+      "ernel": 9961,
+      "思维": 9962,
+      "品种": 9963,
+      "▁flag": 9964,
+      "full": 9965,
+      "违反": 9966,
+      "扣押": 9967,
+      "▁buildings": 9968,
+      "Spec": 9969,
+      "Co": 9970,
+      "otherapy": 9971,
+      "时的": 9972,
+      "大型": 9973,
+      "helial": 9974,
+      "▁USA": 9975,
+      "▁detected": 9976,
+      "tage": 9977,
+      "▁Journal": 9978,
+      "等于": 9979,
+      "▁bow": 9980,
+      "▁numerous": 9981,
+      "▁throw": 9982,
+      "本科": 9983,
+      "进步": 9984,
+      "▁relatively": 9985,
+      "很大": 9986,
+      "padding": 9987,
+      "▁[]": 9988,
+      "}$$\\": 9989,
+      "ulating": 9990,
+      "还能": 9991,
+      "二〇一四年": 9992,
+      "▁secondary": 9993,
+      "▁flat": 9994,
+      "经营范围": 9995,
+      "身边": 9996,
+      "策略": 9997,
+      "▁discussion": 9998,
+      "ipal": 9999,
+      "mode": 10000,
+      "exit": 10001,
+      "tree": 10002,
+      "▁suitable": 10003,
+      "ader": 10004,
+      "集体": 10005,
+      "▁toward": 10006,
+      "依然": 10007,
+      "▁governing": 10008,
+      "reshold": 10009,
+      "▁homes": 10010,
+      "山区": 10011,
+      "GET": 10012,
+      "盗窃": 10013,
+      "院校": 10014,
+      "往往": 10015,
+      "▁Stand": 10016,
+      "unched": 10017,
+      "▁modify": 10018,
+      "▁equivalent": 10019,
+      "财政": 10020,
+      "▁Virgin": 10021,
+      "容量": 10022,
+      "规律": 10023,
+      "TV": 10024,
+      "CITE": 10025,
+      "原来": 10026,
+      "iology": 10027,
+      "▁whereas": 10028,
+      "▁pow": 10029,
+      "gt": 10030,
+      "上述事实": 10031,
+      "oral": 10032,
+      "并按": 10033,
+      "istan": 10034,
+      "_\\": 10035,
+      "▁ok": 10036,
+      "▁experiences": 10037,
+      "ras": 10038,
+      "▁occup": 10039,
+      "名字": 10040,
+      "父亲": 10041,
+      "astructure": 10042,
+      "▁schedule": 10043,
+      "的前": 10044,
+      "▁Even": 10045,
+      "uv": 10046,
+      "people": 10047,
+      "ported": 10048,
+      "▁partial": 10049,
+      "眼睛": 10050,
+      "屏幕": 10051,
+      "▁gift": 10052,
+      "物理": 10053,
+      "▁artists": 10054,
+      "▁Team": 10055,
+      "▁colors": 10056,
+      "▁Star": 10057,
+      "Serial": 10058,
+      "减去": 10059,
+      "person": 10060,
+      "idae": 10061,
+      "▁Error": 10062,
+      "icing": 10063,
+      "styles": 10064,
+      ">,": 10065,
+      "▁intervention": 10066,
+      "天下": 10067,
+      "acts": 10068,
+      "Wh": 10069,
+      "END": 10070,
+      "病毒": 10071,
+      "▁Mel": 10072,
+      "▁suggests": 10073,
+      "电源": 10074,
+      "▁ahead": 10075,
+      "▁quarter": 10076,
+      "▁Lear": 10077,
+      "▁shift": 10078,
+      "efined": 10079,
+      "kl": 10080,
+      "▁Hill": 10081,
+      "日向本院提出撤诉申请": 10082,
+      "▁helped": 10083,
+      "▁mal": 10084,
+      "好像": 10085,
+      "▁Card": 10086,
+      "先后": 10087,
+      "地理": 10088,
+      "中共": 10089,
+      "Windows": 10090,
+      "处于": 10091,
+      "的生活": 10092,
+      "▁administr": 10093,
+      "作业": 10094,
+      "▁dys": 10095,
+      "▁mentioned": 10096,
+      "未按": 10097,
+      "▁monitoring": 10098,
+      "终于": 10099,
+      "不知": 10100,
+      "▁Camp": 10101,
+      "模拟": 10102,
+      "udio": 10103,
+      "\":\\": 10104,
+      "▁wear": 10105,
+      "zip": 10106,
+      "▁standards": 10107,
+      "▁senior": 10108,
+      "▁Zeal": 10109,
+      "ogene": 10110,
+      "TI": 10111,
+      "idential": 10112,
+      "application": 10113,
+      "▁extended": 10114,
+      "tests": 10115,
+      "是由": 10116,
+      "organ": 10117,
+      "▁substant": 10118,
+      "▁mental": 10119,
+      "▁连载状态": 10120,
+      "UB": 10121,
+      "处罚金": 10122,
+      "规则": 10123,
+      "▁helping": 10124,
+      "▁haven": 10125,
+      "女生": 10126,
+      "grid": 10127,
+      "在这": 10128,
+      "▁centre": 10129,
+      "urban": 10130,
+      "▁magnetic": 10131,
+      "Off": 10132,
+      "uri": 10133,
+      "scope": 10134,
+      ")]": 10135,
+      "▁Histor": 10136,
+      "▁obj": 10137,
+      "▁beg": 10138,
+      "▁roles": 10139,
+      "的心": 10140,
+      "nAcc": 10141,
+      "高中": 10142,
+      "etry": 10143,
+      "下面的": 10144,
+      "Bar": 10145,
+      "▁absolute": 10146,
+      "▁Calculate": 10147,
+      "eq": 10148,
+      "存储": 10149,
+      "het": 10150,
+      "▁Start": 10151,
+      "可能会": 10152,
+      "情绪": 10153,
+      "isl": 10154,
+      "▁Italian": 10155,
+      "定位": 10156,
+      "resource": 10157,
+      "）<": 10158,
+      "xygen": 10159,
+      "▁tasks": 10160,
+      "ickets": 10161,
+      "zh": 10162,
+      "▁Game": 10163,
+      "dependent": 10164,
+      "aws": 10165,
+      "Output": 10166,
+      "icago": 10167,
+      "▁loop": 10168,
+      "rich": 10169,
+      "Met": 10170,
+      "看了": 10171,
+      "nOr": 10172,
+      "▁Review": 10173,
+      "▁notes": 10174,
+      "igrations": 10175,
+      "▁raised": 10176,
+      "▁Android": 10177,
+      "在于": 10178,
+      "动作": 10179,
+      "▁mainly": 10180,
+      "account": 10181,
+      "▁KIND": 10182,
+      "SQL": 10183,
+      "美丽": 10184,
+      "▁Education": 10185,
+      "▁division": 10186,
+      "有所": 10187,
+      "▁fans": 10188,
+      "▁replace": 10189,
+      "就要": 10190,
+      "▁Fire": 10191,
+      "umin": 10192,
+      "▁IP": 10193,
+      "▁edit": 10194,
+      "printf": 10195,
+      "tmp": 10196,
+      "compl": 10197,
+      "Props": 10198,
+      "▁residents": 10199,
+      "Parser": 10200,
+      "公告": 10201,
+      "▁Back": 10202,
+      "▁Zealand": 10203,
+      "RC": 10204,
+      "▁velocity": 10205,
+      "▁spring": 10206,
+      "长度": 10207,
+      "azz": 10208,
+      "步骤": 10209,
+      "财务": 10210,
+      "▁laws": 10211,
+      "探索": 10212,
+      "▁Exception": 10213,
+      "▁成立": 10214,
+      "▁Twitter": 10215,
+      "▁aren": 10216,
+      "业有限公司": 10217,
+      "传唤": 10218,
+      "plates": 10219,
+      "uries": 10220,
+      "po": 10221,
+      "明白": 10222,
+      "儿子": 10223,
+      "人家": 10224,
+      "▁disorders": 10225,
+      "▁Academy": 10226,
+      "handle": 10227,
+      "window": 10228,
+      "kr": 10229,
+      "▁buff": 10230,
+      "\\',\\'": 10231,
+      "湖北省": 10232,
+      "▁hel": 10233,
+      "▁cold": 10234,
+      "▁signaling": 10235,
+      "龙江": 10236,
+      "迅速": 10237,
+      "▁fourth": 10238,
+      "日被": 10239,
+      "▁sand": 10240,
+      "dest": 10241,
+      "▁perspect": 10242,
+      "ognitive": 10243,
+      "ugins": 10244,
+      "▁sexual": 10245,
+      "arrant": 10246,
+      "Rem": 10247,
+      "▁oxid": 10248,
+      "▁measurement": 10249,
+      "''": 10250,
+      "十年": 10251,
+      "igan": 10252,
+      "成就": 10253,
+      "▁Ir": 10254,
+      "ishes": 10255,
+      "MD": 10256,
+      "ymbol": 10257,
+      "▁Lim": 10258,
+      "_{\\\\": 10259,
+      "特性": 10260,
+      "深圳市": 10261,
+      "reens": 10262,
+      "线的": 10263,
+      "ffee": 10264,
+      "▁cryst": 10265,
+      "Block": 10266,
+      "life": 10267,
+      "nShe": 10268,
+      "Container": 10269,
+      "▁cards": 10270,
+      "轻松": 10271,
+      "完毕": 10272,
+      "▁promote": 10273,
+      "▁agree": 10274,
+      "本人": 10275,
+      "▁Max": 10276,
+      "▁mostly": 10277,
+      "▁rich": 10278,
+      "▁isolated": 10279,
+      "▁tun": 10280,
+      "在中国": 10281,
+      "ostream": 10282,
+      "▁Life": 10283,
+      "一家": 10284,
+      "ERS": 10285,
+      "iot": 10286,
+      "LED": 10287,
+      "体现": 10288,
+      "日作出": 10289,
+      "奖励": 10290,
+      "omer": 10291,
+      "▁baby": 10292,
+      "▁fashion": 10293,
+      "判处有期徒刑": 10294,
+      "▁starts": 10295,
+      "▁Royal": 10296,
+      "每年": 10297,
+      "▁weak": 10298,
+      "原告的": 10299,
+      "▁technologies": 10300,
+      "积极参加": 10301,
+      "▁ion": 10302,
+      "▁Function": 10303,
+      "▁CA": 10304,
+      "强的": 10305,
+      "▁hus": 10306,
+      "▁gly": 10307,
+      "▁cart": 10308,
+      "omes": 10309,
+      "▁sea": 10310,
+      "▁scientific": 10311,
+      "客房": 10312,
+      "▁engineering": 10313,
+      "向本院递交上诉状": 10314,
+      "▁mel": 10315,
+      "▁Director": 10316,
+      "],[": 10317,
+      "ipher": 10318,
+      "omal": 10319,
+      "▁jobs": 10320,
+      "▁modified": 10321,
+      "empty": 10322,
+      "▁tub": 10323,
+      "▁highlight": 10324,
+      "HS": 10325,
+      "▁locations": 10326,
+      "Return": 10327,
+      "▁mist": 10328,
+      "命令": 10329,
+      "async": 10330,
+      "句话": 10331,
+      "ogenic": 10332,
+      "▁recovery": 10333,
+      "▁cred": 10334,
+      "网友": 10335,
+      "bur": 10336,
+      "▁brief": 10337,
+      "forms": 10338,
+      "▁Map": 10339,
+      "Oper": 10340,
+      "▁picked": 10341,
+      "aria": 10342,
+      "Address": 10343,
+      "bind": 10344,
+      "▁scene": 10345,
+      "▁Sign": 10346,
+      "是我": 10347,
+      "▁experts": 10348,
+      "各个": 10349,
+      "▁书名": 10350,
+      "▁consistent": 10351,
+      "▁practical": 10352,
+      "▁{\"": 10353,
+      "pped": 10354,
+      "文献": 10355,
+      "ads": 10356,
+      "▁Information": 10357,
+      "▁CI": 10358,
+      "emia": 10359,
+      "annotation": 10360,
+      "执行过程中": 10361,
+      "▁Which": 10362,
+      "▁opin": 10363,
+      "第一百零": 10364,
+      "▁Market": 10365,
+      "▁consists": 10366,
+      "cons": 10367,
+      "domain": 10368,
+      "▁ship": 10369,
+      "▁lock": 10370,
+      "cean": 10371,
+      "▁Java": 10372,
+      "理想": 10373,
+      "▁da": 10374,
+      "interface": 10375,
+      ")*(": 10376,
+      "请你": 10377,
+      "▁Mot": 10378,
+      "韩国": 10379,
+      "记得": 10380,
+      "▁stations": 10381,
+      "▁documents": 10382,
+      "三角形": 10383,
+      "uma": 10384,
+      "也没有": 10385,
+      "liament": 10386,
+      "▁Care": 10387,
+      "▁Christian": 10388,
+      "▁miles": 10389,
+      "▁tor": 10390,
+      "几年": 10391,
+      "▁relig": 10392,
+      "Change": 10393,
+      "▁Image": 10394,
+      "Mat": 10395,
+      "▁lab": 10396,
+      "▁odd": 10397,
+      "▁Mo": 10398,
+      "▁susp": 10399,
+      "▁procedures": 10400,
+      "队伍": 10401,
+      "endar": 10402,
+      "不予": 10403,
+      "▁accurate": 10404,
+      "▁Message": 10405,
+      "保存": 10406,
+      "居住": 10407,
+      "▁parallel": 10408,
+      "await": 10409,
+      "hot": 10410,
+      "▁coverage": 10411,
+      "▁correlation": 10412,
+      "满意": 10413,
+      "player": 10414,
+      "DP": 10415,
+      "ented": 10416,
+      "success": 10417,
+      "共有": 10418,
+      "mult": 10419,
+      "▁remaining": 10420,
+      "▁certainly": 10421,
+      "▁Admin": 10422,
+      "中间": 10423,
+      "▁Minister": 10424,
+      "的水": 10425,
+      "▁plane": 10426,
+      "▁summ": 10427,
+      "▁somew": 10428,
+      "▁photograph": 10429,
+      "even": 10430,
+      "大利": 10431,
+      "idx": 10432,
+      "▁Play": 10433,
+      "▁Feb": 10434,
+      "foot": 10435,
+      "adr": 10436,
+      "公众": 10437,
+      "ohn": 10438,
+      "nRes": 10439,
+      "▁synthesis": 10440,
+      "召开": 10441,
+      "▁simulation": 10442,
+      "▁becoming": 10443,
+      "estic": 10444,
+      "▁menu": 10445,
+      "igital": 10446,
+      "机的": 10447,
+      "▁cort": 10448,
+      "▁pieces": 10449,
+      "▁biological": 10450,
+      "▁welcome": 10451,
+      "▁despite": 10452,
+      "hether": 10453,
+      "Process": 10454,
+      "简易程序": 10455,
+      "▁Fre": 10456,
+      "▁PRO": 10457,
+      "▁Smith": 10458,
+      "];": 10459,
+      "▁por": 10460,
+      "▁voice": 10461,
+      "诊断": 10462,
+      "▁esc": 10463,
+      "情形": 10464,
+      "维持": 10465,
+      "下去": 10466,
+      "▁panel": 10467,
+      "▁Note": 10468,
+      "Layout": 10469,
+      "▁integrated": 10470,
+      "▁lots": 10471,
+      "▁Lib": 10472,
+      "▁proport": 10473,
+      "▁forces": 10474,
+      "母亲": 10475,
+      "[\\": 10476,
+      "▁implemented": 10477,
+      "appro": 10478,
+      "▁linked": 10479,
+      "▁occurs": 10480,
+      "achine": 10481,
+      "慢慢": 10482,
+      "尤其是": 10483,
+      "onald": 10484,
+      "▁controlled": 10485,
+      "▁vary": 10486,
+      "urer": 10487,
+      "GL": 10488,
+      "相应的": 10489,
+      "▁:\\": 10490,
+      "▁felt": 10491,
+      "water": 10492,
+      "▁Finally": 10493,
+      "hematical": 10494,
+      "strap": 10495,
+      "So": 10496,
+      "▁greatest": 10497,
+      "plicated": 10498,
+      "Column": 10499,
+      "icacy": 10500,
+      "智慧": 10501,
+      "时尚": 10502,
+      "自愿": 10503,
+      "▁Register": 10504,
+      "主张": 10505,
+      "▁versions": 10506,
+      "▁matches": 10507,
+      "▁sharing": 10508,
+      "▁Love": 10509,
+      "青春": 10510,
+      "ILITY": 10511,
+      "心里": 10512,
+      "kit": 10513,
+      "▁Den": 10514,
+      "技巧": 10515,
+      "▁manif": 10516,
+      "▁Bill": 10517,
+      "▁arrest": 10518,
+      "There": 10519,
+      "四个": 10520,
+      "▁absor": 10521,
+      "▁minim": 10522,
+      "anda": 10523,
+      "Vis": 10524,
+      "▁federal": 10525,
+      "▁underlying": 10526,
+      "的执行": 10527,
+      "▁Commission": 10528,
+      "scopy": 10529,
+      "Auth": 10530,
+      "argv": 10531,
+      "▁optimal": 10532,
+      "▁foreign": 10533,
+      "HER": 10534,
+      "▁Isra": 10535,
+      "▁binary": 10536,
+      "而言": 10537,
+      "▁league": 10538,
+      "签订": 10539,
+      "▁Mer": 10540,
+      "的情况下": 10541,
+      "突出": 10542,
+      "emporary": 10543,
+      "▁occurred": 10544,
+      "女人": 10545,
+      "利润": 10546,
+      "停止": 10547,
+      "类似": 10548,
+      "▁managed": 10549,
+      "制定": 10550,
+      "▁alumni": 10551,
+      "▁convers": 10552,
+      "friend": 10553,
+      "▁Committee": 10554,
+      "未履行": 10555,
+      "vm": 10556,
+      "法国": 10557,
+      "▁meant": 10558,
+      "Order": 10559,
+      "岗位": 10560,
+      "▁Order": 10561,
+      "监管": 10562,
+      "▁Mil": 10563,
+      "hav": 10564,
+      "更是": 10565,
+      "解得": 10566,
+      "▁architecture": 10567,
+      "▁segment": 10568,
+      "iatric": 10569,
+      "的事实": 10570,
+      "Go": 10571,
+      "▁stability": 10572,
+      "▁icon": 10573,
+      "cluded": 10574,
+      "▁Gal": 10575,
+      "▁allowing": 10576,
+      "IE": 10577,
+      "放入": 10578,
+      "▁prosp": 10579,
+      "bound": 10580,
+      "企业的": 10581,
+      "等证据": 10582,
+      "shape": 10583,
+      "Hub": 10584,
+      "科学院": 10585,
+      "▁成立时间": 10586,
+      "地位": 10587,
+      "dimensional": 10588,
+      ":(": 10589,
+      "Format": 10590,
+      "▁clearly": 10591,
+      "Me": 10592,
+      "▁digit": 10593,
+      "dot": 10594,
+      "npm": 10595,
+      "▁feeling": 10596,
+      "流量": 10597,
+      "查看": 10598,
+      "空调": 10599,
+      "Search": 10600,
+      "Arg": 10601,
+      "等级": 10602,
+      "▁estate": 10603,
+      "▁·": 10604,
+      "被上诉人": 10605,
+      "▁confidence": 10606,
+      "给予": 10607,
+      "突破": 10608,
+      "NT": 10609,
+      "还款": 10610,
+      "Right": 10611,
+      "▁Water": 10612,
+      "得出": 10613,
+      "记忆": 10614,
+      "Br": 10615,
+      "▁guys": 10616,
+      "ABLE": 10617,
+      "强调": 10618,
+      "▁emph": 10619,
+      "utter": 10620,
+      "axis": 10621,
+      "▁Core": 10622,
+      "▁fant": 10623,
+      "▁gluc": 10624,
+      "▁vacc": 10625,
+      "▁funding": 10626,
+      "ctx": 10627,
+      "▁estimate": 10628,
+      "HA": 10629,
+      "▁slightly": 10630,
+      "▁parse": 10631,
+      "▁throws": 10632,
+      "日在": 10633,
+      "DL": 10634,
+      "▁perhaps": 10635,
+      "显着": 10636,
+      "人数": 10637,
+      "复议": 10638,
+      "posite": 10639,
+      "▁posts": 10640,
+      "▁programming": 10641,
+      "Write": 10642,
+      "profile": 10643,
+      "iger": 10644,
+      "icious": 10645,
+      "Store": 10646,
+      "成熟": 10647,
+      "▁typically": 10648,
+      "▁bill": 10649,
+      "▁logging": 10650,
+      "layout": 10651,
+      "垃圾": 10652,
+      "▁regional": 10653,
+      "▁sector": 10654,
+      "分配": 10655,
+      "ione": 10656,
+      "词条": 10657,
+      "装备": 10658,
+      "学历": 10659,
+      "created": 10660,
+      "并处罚金": 10661,
+      "矛盾": 10662,
+      ")]\\": 10663,
+      "▁Wil": 10664,
+      "Interface": 10665,
+      "▁electro": 10666,
+      "▁reserved": 10667,
+      "late": 10668,
+      "▁tele": 10669,
+      "摄影": 10670,
+      "▁sensitivity": 10671,
+      "观众": 10672,
+      "▁Louis": 10673,
+      "预防": 10674,
+      "▁academic": 10675,
+      "▁Frank": 10676,
+      "▁professionals": 10677,
+      "lt": 10678,
+      "完整": 10679,
+      "▁Custom": 10680,
+      "va": 10681,
+      "访问": 10682,
+      "writ": 10683,
+      "区别": 10684,
+      "感谢": 10685,
+      "uint": 10686,
+      "views": 10687,
+      "Dev": 10688,
+      "▁utf": 10689,
+      "▁retail": 10690,
+      "atial": 10691,
+      "▁founded": 10692,
+      "▁ast": 10693,
+      "Length": 10694,
+      "ucky": 10695,
+      "\"]\\": 10696,
+      "十七": 10697,
+      "衣服": 10698,
+      "▁payment": 10699,
+      "▁wonderful": 10700,
+      "ontal": 10701,
+      "▁Rock": 10702,
+      "Label": 10703,
+      "ega": 10704,
+      "青岛": 10705,
+      "做的": 10706,
+      "▁syndrome": 10707,
+      "▁Francis": 10708,
+      "使得": 10709,
+      "▁mortality": 10710,
+      "▁thousands": 10711,
+      "anguages": 10712,
+      "▁Wood": 10713,
+      "▁mur": 10714,
+      "ATA": 10715,
+      "长的": 10716,
+      "▁duration": 10717,
+      "aters": 10718,
+      "tself": 10719,
+      "Part": 10720,
+      "本题考查": 10721,
+      "▁Furthermore": 10722,
+      "一张": 10723,
+      "astern": 10724,
+      "▁Land": 10725,
+      "▁guitar": 10726,
+      "测量": 10727,
+      "omorph": 10728,
+      "▁Point": 10729,
+      "▁Ins": 10730,
+      "▁Licensed": 10731,
+      "▁noise": 10732,
+      "plt": 10733,
+      "▁Technology": 10734,
+      "充满": 10735,
+      "arian": 10736,
+      "bose": 10737,
+      "adow": 10738,
+      "▁vote": 10739,
+      "exports": 10740,
+      "▁BAS": 10741,
+      "黑龙江": 10742,
+      "news": 10743,
+      "▁differential": 10744,
+      "▁leaves": 10745,
+      "也没": 10746,
+      "ugar": 10747,
+      "uminate": 10748,
+      "APP": 10749,
+      "bits": 10750,
+      "湖南": 10751,
+      "MT": 10752,
+      "阳光": 10753,
+      "▁curve": 10754,
+      "可能是": 10755,
+      "▁cars": 10756,
+      "出版的图书": 10757,
+      "/**": 10758,
+      "▁noted": 10759,
+      "eless": 10760,
+      "▁strugg": 10761,
+      "发现被执行人有": 10762,
+      "itionally": 10763,
+      "在服刑期间": 10764,
+      "铁路": 10765,
+      "▁concerns": 10766,
+      "tw": 10767,
+      "不再": 10768,
+      "sigma": 10769,
+      "教材": 10770,
+      "成分": 10771,
+      "▁DE": 10772,
+      "nBy": 10773,
+      "▁Football": 10774,
+      "展开": 10775,
+      "Max": 10776,
+      "营业": 10777,
+      "chron": 10778,
+      "出去": 10779,
+      "周期": 10780,
+      "▁comprehensive": 10781,
+      "▁Ent": 10782,
+      "See": 10783,
+      "实在": 10784,
+      "可供执行财产": 10785,
+      "证书": 10786,
+      "对比": 10787,
+      "unct": 10788,
+      "▁civil": 10789,
+      "▁calculated": 10790,
+      "▁creative": 10791,
+      "RS": 10792,
+      "之外": 10793,
+      "media": 10794,
+      "▁appearance": 10795,
+      "ressed": 10796,
+      "引导": 10797,
+      "▁Product": 10798,
+      "BI": 10799,
+      "gricult": 10800,
+      "对称": 10801,
+      "AY": 10802,
+      "▁Thomas": 10803,
+      "▁whom": 10804,
+      "ading": 10805,
+      "▁jav": 10806,
+      "▁sq": 10807,
+      "▁transmission": 10808,
+      "成的": 10809,
+      "来到": 10810,
+      "duction": 10811,
+      "小组": 10812,
+      "▁rect": 10813,
+      "▁society": 10814,
+      "▁liquid": 10815,
+      "nWhile": 10816,
+      "▁intended": 10817,
+      "▁##": 10818,
+      "Header": 10819,
+      "▁OS": 10820,
+      "▁Real": 10821,
+      "▁Enter": 10822,
+      "uplic": 10823,
+      "▁adj": 10824,
+      "大于": 10825,
+      "裁定书": 10826,
+      "▁extensive": 10827,
+      "英文": 10828,
+      "平面": 10829,
+      "▁researchers": 10830,
+      "已经发生法律效力": 10831,
+      "ockey": 10832,
+      "itud": 10833,
+      "▁indicated": 10834,
+      "的原因": 10835,
+      "的作用": 10836,
+      "abil": 10837,
+      "▁remote": 10838,
+      "▁writer": 10839,
+      "逾期": 10840,
+      "分公司": 10841,
+      "黄金": 10842,
+      "uty": 10843,
+      "构建": 10844,
+      "heet": 10845,
+      "▁Social": 10846,
+      "▁mo": 10847,
+      "▁algebra": 10848,
+      "AI": 10849,
+      "Dir": 10850,
+      "做法": 10851,
+      "并按对方当事人": 10852,
+      "并不是": 10853,
+      "▁AT": 10854,
+      "ateral": 10855,
+      "▁gather": 10856,
+      "人士": 10857,
+      "▁characterized": 10858,
+      "▁…": 10859,
+      "▁continues": 10860,
+      "不如": 10861,
+      "十五日内": 10862,
+      "▁answers": 10863,
+      "重量": 10864,
+      "一致": 10865,
+      "人工": 10866,
+      "▁Value": 10867,
+      "我在": 10868,
+      "ben": 10869,
+      "▁talking": 10870,
+      "▁Control": 10871,
+      "▁carcin": 10872,
+      "▁fraction": 10873,
+      "▁Hand": 10874,
+      "▁marked": 10875,
+      "dc": 10876,
+      "▁circle": 10877,
+      "▁RNA": 10878,
+      "transform": 10879,
+      "▁{{": 10880,
+      "▁seeing": 10881,
+      "▁sam": 10882,
+      "▁Number": 10883,
+      "▁discovered": 10884,
+      "排名": 10885,
+      "违法": 10886,
+      "▁Name": 10887,
+      "▁nit": 10888,
+      "▁recommended": 10889,
+      "chain": 10890,
+      "▁upload": 10891,
+      "yg": 10892,
+      "amental": 10893,
+      "▁epit": 10894,
+      "动态": 10895,
+      "▁manner": 10896,
+      "集合": 10897,
+      "ourse": 10898,
+      "经审查": 10899,
+      "olit": 10900,
+      "一十九": 10901,
+      "是一款": 10902,
+      "提出了": 10903,
+      "还可以": 10904,
+      "▁内容简介": 10905,
+      "ktop": 10906,
+      "防止": 10907,
+      "▁Hen": 10908,
+      "▁Author": 10909,
+      "就算": 10910,
+      "▁properly": 10911,
+      "产品的": 10912,
+      "▁rain": 10913,
+      "提出副本": 10914,
+      "iostream": 10915,
+      "Min": 10916,
+      "▁Array": 10917,
+      "此时": 10918,
+      "▁diabetes": 10919,
+      "▁EX": 10920,
+      "▁burn": 10921,
+      "▁polynomial": 10922,
+      "杭州": 10923,
+      "wiki": 10924,
+      "强化": 10925,
+      "▁pept": 10926,
+      "子女": 10927,
+      "▁steel": 10928,
+      "reland": 10929,
+      "▁voc": 10930,
+      "Level": 10931,
+      "▁Museum": 10932,
+      "▁facility": 10933,
+      "fin": 10934,
+      "▁evolution": 10935,
+      "生效法律文书": 10936,
+      "▁wedding": 10937,
+      "leqslant": 10938,
+      "▁famous": 10939,
+      "Configuration": 10940,
+      "▁国籍": 10941,
+      "▁Field": 10942,
+      "▁installation": 10943,
+      "imensions": 10944,
+      "被害人": 10945,
+      "都要": 10946,
+      "通道": 10947,
+      "utch": 10948,
+      "hentication": 10949,
+      "单元": 10950,
+      "▁submit": 10951,
+      "▁eat": 10952,
+      "▁serving": 10953,
+      "acle": 10954,
+      "▁enzyme": 10955,
+      "适用简易程序": 10956,
+      "▁},\\": 10957,
+      "Assert": 10958,
+      "▁nodes": 10959,
+      "SL": 10960,
+      "出来的": 10961,
+      "▁husband": 10962,
+      "允许": 10963,
+      "▁Series": 10964,
+      "▁Community": 10965,
+      "▁Wal": 10966,
+      "nCl": 10967,
+      "带着": 10968,
+      "bc": 10969,
+      "ogg": 10970,
+      "equal": 10971,
+      "metric": 10972,
+      "氧化": 10973,
+      "解析": 10974,
+      "tab": 10975,
+      "攻击": 10976,
+      "▁sustain": 10977,
+      "high": 10978,
+      "second": 10979,
+      "PATH": 10980,
+      "▁合作": 10981,
+      "相等": 10982,
+      "补充": 10983,
+      "独特": 10984,
+      "ongo": 10985,
+      "还在": 10986,
+      "不多": 10987,
+      "类别": 10988,
+      "▁electronic": 10989,
+      "▁depth": 10990,
+      "俄罗斯": 10991,
+      "▁（": 10992,
+      "▁depending": 10993,
+      "各地": 10994,
+      "func": 10995,
+      "nap": 10996,
+      "▁Pan": 10997,
+      "\"><": 10998,
+      "pled": 10999,
+      "▁apparent": 11000,
+      "▁nuclear": 11001,
+      "▁Blue": 11002,
+      "伙伴": 11003,
+      "▁browser": 11004,
+      "▁volunte": 11005,
+      "\\\\{": 11006,
+      "副主任": 11007,
+      "▁tail": 11008,
+      "ira": 11009,
+      "案例": 11010,
+      "▁chosen": 11011,
+      "NN": 11012,
+      "主体": 11013,
+      "通信": 11014,
+      "▁widely": 11015,
+      "scape": 11016,
+      "rogen": 11017,
+      "▁presents": 11018,
+      "▁Pub": 11019,
+      "ilation": 11020,
+      "▁Tem": 11021,
+      "plex": 11022,
+      "天然": 11023,
+      "}}\\": 11024,
+      "packages": 11025,
+      "近日": 11026,
+      "并给出": 11027,
+      "▁transcription": 11028,
+      "▁neighbor": 11029,
+      "战斗": 11030,
+      "btn": 11031,
+      "▁Tim": 11032,
+      "身上": 11033,
+      "条第二": 11034,
+      "债权": 11035,
+      "▁il": 11036,
+      "▁People": 11037,
+      "▁Congress": 11038,
+      "depend": 11039,
+      "Description": 11040,
+      "▁fif": 11041,
+      "▁Fort": 11042,
+      "Control": 11043,
+      "▁yes": 11044,
+      "Init": 11045,
+      "fold": 11046,
+      "temp": 11047,
+      "umes": 11048,
+      "不变": 11049,
+      "▁heter": 11050,
+      "DC": 11051,
+      "▁warrant": 11052,
+      "ilot": 11053,
+      "▁purs": 11054,
+      "平衡": 11055,
+      "▁stack": 11056,
+      "▁fabric": 11057,
+      "基因": 11058,
+      "▁sufficient": 11059,
+      "▁helpful": 11060,
+      "▁Event": 11061,
+      "svg": 11062,
+      "▁economy": 11063,
+      "▁provider": 11064,
+      "芯片": 11065,
+      "▁folder": 11066,
+      "争议": 11067,
+      "▁eventually": 11068,
+      "ador": 11069,
+      "特别是": 11070,
+      "▁waiting": 11071,
+      "nNot": 11072,
+      "▁household": 11073,
+      "utf": 11074,
+      "▁NULL": 11075,
+      "▁vision": 11076,
+      "▁launched": 11077,
+      "▁youth": 11078,
+      "第一个": 11079,
+      "非法": 11080,
+      "▁io": 11081,
+      "搭配": 11082,
+      "▁channels": 11083,
+      "▁在": 11084,
+      "▁updates": 11085,
+      "▁municip": 11086,
+      "▁assume": 11087,
+      "二〇一八年": 11088,
+      "▁faster": 11089,
+      "▁recon": 11090,
+      "▁assessed": 11091,
+      "queue": 11092,
+      "▁container": 11093,
+      "tc": 11094,
+      "role": 11095,
+      "▁parties": 11096,
+      "▁shall": 11097,
+      "▁ment": 11098,
+      "▁gradu": 11099,
+      "▁implements": 11100,
+      "nv": 11101,
+      "bad": 11102,
+      "▁Ref": 11103,
+      "ko": 11104,
+      "期待": 11105,
+      "▁oxygen": 11106,
+      "▁learned": 11107,
+      "global": 11108,
+      "学生的": 11109,
+      "eft": 11110,
+      "ptr": 11111,
+      "▁horm": 11112,
+      "▁Russia": 11113,
+      "▁anc": 11114,
+      "▁trials": 11115,
+      "LOG": 11116,
+      "▁boy": 11117,
+      "month": 11118,
+      "Eff": 11119,
+      "(:": 11120,
+      "当代": 11121,
+      "EA": 11122,
+      "天的": 11123,
+      "▁criteria": 11124,
+      "▁winning": 11125,
+      "csv": 11126,
+      "▁arguments": 11127,
+      "Mode": 11128,
+      "kins": 11129,
+      "物品": 11130,
+      "直播": 11131,
+      "都可以": 11132,
+      "bridge": 11133,
+      "号码": 11134,
+      "assets": 11135,
+      "insert": 11136,
+      "▁Medical": 11137,
+      "▁fan": 11138,
+      "ji": 11139,
+      "▁Disc": 11140,
+      "▁Based": 11141,
+      "▁supports": 11142,
+      "▁medal": 11143,
+      "▁axis": 11144,
+      "ws": 11145,
+      "ears": 11146,
+      "▁recognition": 11147,
+      "▁motiv": 11148,
+      "Utils": 11149,
+      "交纳": 11150,
+      "▁Once": 11151,
+      "▁pathway": 11152,
+      "▁truly": 11153,
+      "(*": 11154,
+      "neg": 11155,
+      "▁杨": 11156,
+      "▁minute": 11157,
+      "manager": 11158,
+      "▁looked": 11159,
+      "'),\\": 11160,
+      "是有": 11161,
+      "哪个": 11162,
+      "▁teacher": 11163,
+      "▁representation": 11164,
+      "▁rise": 11165,
+      "▁sport": 11166,
+      "usercontent": 11167,
+      "▁出生日期": 11168,
+      "人体": 11169,
+      "▁twice": 11170,
+      "▁According": 11171,
+      "atre": 11172,
+      "▁Page": 11173,
+      "nmodule": 11174,
+      "浏览": 11175,
+      "FILE": 11176,
+      "性格": 11177,
+      "▁chair": 11178,
+      "▁fluid": 11179,
+      "▁Georg": 11180,
+      "poses": 11181,
+      "最多": 11182,
+      "ictionary": 11183,
+      "练习": 11184,
+      "channel": 11185,
+      "okes": 11186,
+      "▁override": 11187,
+      "▁comfortable": 11188,
+      "▁CH": 11189,
+      "device": 11190,
+      "▁loved": 11191,
+      "▁laboratory": 11192,
+      "delta": 11193,
+      "▁CONDIT": 11194,
+      "Tag": 11195,
+      "硕士": 11196,
+      "▁Mexico": 11197,
+      "▁windows": 11198,
+      "权益": 11199,
+      "hester": 11200,
+      "的东西": 11201,
+      "gression": 11202,
+      "echo": 11203,
+      "▁sweet": 11204,
+      "▁originally": 11205,
+      "审批": 11206,
+      "金钱": 11207,
+      "▁Los": 11208,
+      "▁biom": 11209,
+      "顺利": 11210,
+      "ochemical": 11211,
+      "▁coffee": 11212,
+      "的基本": 11213,
+      "▁Cath": 11214,
+      "租赁": 11215,
+      "后面": 11216,
+      "人在": 11217,
+      "覆盖": 11218,
+      "提取": 11219,
+      "初始": 11220,
+      "▁happened": 11221,
+      "▁III": 11222,
+      "nThese": 11223,
+      "▁Paris": 11224,
+      "本案现已审理终结": 11225,
+      "▁designs": 11226,
+      "nSince": 11227,
+      "大多数": 11228,
+      "还会": 11229,
+      "▁leaving": 11230,
+      "▁reliable": 11231,
+      "▁analyses": 11232,
+      "nNew": 11233,
+      "推出": 11234,
+      "nAccording": 11235,
+      "请回答": 11236,
+      "▁Peter": 11237,
+      "▁molecules": 11238,
+      "menu": 11239,
+      "▁Chicago": 11240,
+      "▁stick": 11241,
+      "▁echo": 11242,
+      "▁multip": 11243,
+      "▁Victor": 11244,
+      "东方": 11245,
+      "▁bio": 11246,
+      "\"),": 11247,
+      "inds": 11248,
+      "人大": 11249,
+      "▁Bet": 11250,
+      "▁Valley": 11251,
+      "▁theorem": 11252,
+      "bottom": 11253,
+      "▁Their": 11254,
+      "release": 11255,
+      "的人数提出副本": 11256,
+      "日前": 11257,
+      "ACT": 11258,
+      "▁Sou": 11259,
+      "▁ha": 11260,
+      "world": 11261,
+      "成年": 11262,
+      "nit": 11263,
+      "▁mail": 11264,
+      "报名": 11265,
+      "▁Earth": 11266,
+      "更多的": 11267,
+      "objects": 11268,
+      "▁spirit": 11269,
+      "有没有": 11270,
+      "▁alleg": 11271,
+      "rooms": 11272,
+      "第二百四十": 11273,
+      "机场": 11274,
+      "证券": 11275,
+      "回家": 11276,
+      "维修": 11277,
+      "lers": 11278,
+      "▁effectively": 11279,
+      "Attribute": 11280,
+      "iddleware": 11281,
+      "uits": 11282,
+      "elled": 11283,
+      "Args": 11284,
+      "▁BASIS": 11285,
+      "ila": 11286,
+      "eference": 11287,
+      "▁initi": 11288,
+      "▁Army": 11289,
+      "▁CONDITIONS": 11290,
+      "形态": 11291,
+      "协调": 11292,
+      "日起至": 11293,
+      "▁preval": 11294,
+      "▁nic": 11295,
+      "▁conflic": 11296,
+      "▁plastic": 11297,
+      "iana": 11298,
+      "晋江": 11299,
+      "column": 11300,
+      "▁copyright": 11301,
+      "▁univers": 11302,
+      "也许": 11303,
+      "新疆": 11304,
+      "Stack": 11305,
+      "bles": 11306,
+      "百分": 11307,
+      "项之规定": 11308,
+      "rat": 11309,
+      "▁involving": 11310,
+      "▁industrial": 11311,
+      "ifies": 11312,
+      "自主": 11313,
+      "▁raw": 11314,
+      "良好": 11315,
+      "▁jump": 11316,
+      "▁Hy": 11317,
+      "合并": 11318,
+      "▁unless": 11319,
+      "▁Brazil": 11320,
+      "股票": 11321,
+      "▁calling": 11322,
+      "▁educational": 11323,
+      "▁classic": 11324,
+      "▁连载中": 11325,
+      "和谐": 11326,
+      "Connection": 11327,
+      "▁Light": 11328,
+      "▁Special": 11329,
+      "▁Government": 11330,
+      "▁loan": 11331,
+      "Sl": 11332,
+      ">();\\": 11333,
+      "ema": 11334,
+      "提前": 11335,
+      "yte": 11336,
+      "nj": 11337,
+      "Menu": 11338,
+      "也能": 11339,
+      "▁nine": 11340,
+      "(__": 11341,
+      "▁kept": 11342,
+      "▁instrument": 11343,
+      "RT": 11344,
+      "▁Current": 11345,
+      "oking": 11346,
+      "ceived": 11347,
+      "utorial": 11348,
+      "仲裁": 11349,
+      "▁Hospital": 11350,
+      "平时": 11351,
+      "▁compounds": 11352,
+      "▁earth": 11353,
+      "▁decisions": 11354,
+      "nHowever": 11355,
+      "results": 11356,
+      "面临": 11357,
+      "Access": 11358,
+      "edom": 11359,
+      "AND": 11360,
+      "▁Only": 11361,
+      "读者": 11362,
+      "▁spin": 11363,
+      "▁Virginia": 11364,
+      "nList": 11365,
+      "▁Na": 11366,
+      "copy": 11367,
+      "▁climate": 11368,
+      "overs": 11369,
+      "leton": 11370,
+      "▁seat": 11371,
+      "▁frequently": 11372,
+      "▁Ma": 11373,
+      "Mon": 11374,
+      "并没有": 11375,
+      "(&": 11376,
+      "Product": 11377,
+      "▁bird": 11378,
+      ";\\\\": 11379,
+      "▁couldn": 11380,
+      "大学生": 11381,
+      "▁Carolina": 11382,
+      "▁oral": 11383,
+      "红色": 11384,
+      "▁algorithms": 11385,
+      "不够": 11386,
+      "▁Sand": 11387,
+      "ailability": 11388,
+      "Comp": 11389,
+      "▁tab": 11390,
+      "cles": 11391,
+      "▁street": 11392,
+      "commod": 11393,
+      "orough": 11394,
+      "engine": 11395,
+      "▁winter": 11396,
+      "scribe": 11397,
+      "▁'@": 11398,
+      "▁pure": 11399,
+      "Collections": 11400,
+      "等奖": 11401,
+      "整理": 11402,
+      "vents": 11403,
+      "▁asking": 11404,
+      "ABILITY": 11405,
+      "▁depends": 11406,
+      "▁Jose": 11407,
+      "区间": 11408,
+      "ixed": 11409,
+      "screen": 11410,
+      "▁sensitive": 11411,
+      "策划": 11412,
+      "asp": 11413,
+      "配套": 11414,
+      "原料": 11415,
+      "▁forest": 11416,
+      "授权": 11417,
+      "route": 11418,
+      "IGHT": 11419,
+      "ruit": 11420,
+      "▁positions": 11421,
+      "▁Swed": 11422,
+      "▁\"<": 11423,
+      "也很": 11424,
+      "▁elected": 11425,
+      "逻辑": 11426,
+      "▁shell": 11427,
+      "▁targets": 11428,
+      "cluding": 11429,
+      "anes": 11430,
+      "故答案": 11431,
+      "有个": 11432,
+      "▁commonly": 11433,
+      "Delta": 11434,
+      "▁cities": 11435,
+      "icular": 11436,
+      "————": 11437,
+      "JSON": 11438,
+      "():": 11439,
+      "▁delet": 11440,
+      "同志": 11441,
+      "▁receiving": 11442,
+      "▁humans": 11443,
+      "Admin": 11444,
+      "stitution": 11445,
+      "法的": 11446,
+      "吉林省": 11447,
+      "天津市": 11448,
+      "公开开庭进行了审理": 11449,
+      "Option": 11450,
+      "▁composition": 11451,
+      "住房": 11452,
+      "融合": 11453,
+      "可以在": 11454,
+      "令人": 11455,
+      "利于": 11456,
+      "har": 11457,
+      "ntype": 11458,
+      "▁orient": 11459,
+      "向本院提出": 11460,
+      "一个月": 11461,
+      "▁abstract": 11462,
+      "▁trees": 11463,
+      "▁['": 11464,
+      "ollowing": 11465,
+      "hal": 11466,
+      "▁Yes": 11467,
+      "优惠": 11468,
+      "▁circum": 11469,
+      "▁keeping": 11470,
+      "argument": 11471,
+      "▁holds": 11472,
+      "▁receptors": 11473,
+      "▁innovative": 11474,
+      "服装": 11475,
+      "tvar": 11476,
+      "始终": 11477,
+      "Found": 11478,
+      "制品": 11479,
+      "两年": 11480,
+      "论坛": 11481,
+      "formance": 11482,
+      "▁±": 11483,
+      "也要": 11484,
+      "▁regression": 11485,
+      "}_": 11486,
+      "类的": 11487,
+      "拍摄": 11488,
+      "▁corner": 11489,
+      "Widget": 11490,
+      "▁Jes": 11491,
+      "▁iron": 11492,
+      "imb": 11493,
+      "周围": 11494,
+      "▁circuit": 11495,
+      "▁JSON": 11496,
+      "老年": 11497,
+      "eper": 11498,
+      "nab": 11499,
+      "▁cere": 11500,
+      "wd": 11501,
+      "▁numpy": 11502,
+      "Reader": 11503,
+      "Port": 11504,
+      "▁Init": 11505,
+      "nDo": 11506,
+      "上午": 11507,
+      "情节": 11508,
+      "lose": 11509,
+      "▁craft": 11510,
+      "▁journey": 11511,
+      "▁页数": 11512,
+      "Hash": 11513,
+      "女儿": 11514,
+      "▁grid": 11515,
+      "poss": 11516,
+      "朋友圈": 11517,
+      "▁reality": 11518,
+      "不想": 11519,
+      "istent": 11520,
+      "▁vehicles": 11521,
+      "▁faculty": 11522,
+      "▁Old": 11523,
+      ")))\\": 11524,
+      "▁km": 11525,
+      "千万": 11526,
+      "寻找": 11527,
+      "房间": 11528,
+      "utely": 11529,
+      "▁expansion": 11530,
+      "纤维": 11531,
+      "内蒙古": 11532,
+      "▁studio": 11533,
+      "血管": 11534,
+      "activity": 11535,
+      "cers": 11536,
+      "drop": 11537,
+      "▁Aim": 11538,
+      "▁Records": 11539,
+      "▁Organ": 11540,
+      "▁treatments": 11541,
+      "她们": 11542,
+      "预测": 11543,
+      "▁dict": 11544,
+      "family": 11545,
+      "rog": 11546,
+      "产权": 11547,
+      "范围内": 11548,
+      "ikip": 11549,
+      "Window": 11550,
+      "中文网": 11551,
+      "装置": 11552,
+      "inical": 11553,
+      "Template": 11554,
+      "随机": 11555,
+      "eding": 11556,
+      "接到": 11557,
+      "pool": 11558,
+      "▁Analysis": 11559,
+      "ati": 11560,
+      "▁conventional": 11561,
+      "陈述": 11562,
+      "Exec": 11563,
+      "apes": 11564,
+      "▁edition": 11565,
+      "本来": 11566,
+      "Session": 11567,
+      "prise": 11568,
+      "▁reprodu": 11569,
+      "Unit": 11570,
+      "riage": 11571,
+      "▁contribution": 11572,
+      "▁trail": 11573,
+      "▁possibly": 11574,
+      "▁correctly": 11575,
+      "小的": 11576,
+      "Tree": 11577,
+      "itz": 11578,
+      "机动车": 11579,
+      "arks": 11580,
+      "ificate": 11581,
+      "rf": 11582,
+      "▁header": 11583,
+      "edge": 11584,
+      "的内容": 11585,
+      "Equals": 11586,
+      "对照组": 11587,
+      "半年": 11588,
+      "也就": 11589,
+      "▁п": 11590,
+      "▁colour": 11591,
+      "▁usage": 11592,
+      "▁bag": 11593,
+      "函数的": 11594,
+      "nNote": 11595,
+      "附近": 11596,
+      "NF": 11597,
+      "电力": 11598,
+      "▁forget": 11599,
+      "▁scores": 11600,
+      "Lin": 11601,
+      "新区": 11602,
+      "▁replaced": 11603,
+      "▁killed": 11604,
+      "Icon": 11605,
+      "▁Energy": 11606,
+      "高等": 11607,
+      "ufeffusing": 11608,
+      "thread": 11609,
+      "▁accompl": 11610,
+      "you": 11611,
+      "近年来": 11612,
+      "impl": 11613,
+      "超级": 11614,
+      "天气": 11615,
+      "Direct": 11616,
+      "▁fract": 11617,
+      "▁plugin": 11618,
+      "▁courses": 11619,
+      "高效": 11620,
+      "▁Ter": 11621,
+      "▁identification": 11622,
+      "地图": 11623,
+      "ById": 11624,
+      "吸收": 11625,
+      "▁cultural": 11626,
+      "▁laser": 11627,
+      "做了": 11628,
+      "apse": 11629,
+      "账户": 11630,
+      "IVE": 11631,
+      "▁garden": 11632,
+      "▁west": 11633,
+      "翻译": 11634,
+      "▁coast": 11635,
+      "▁nurs": 11636,
+      "▁cer": 11637,
+      "rize": 11638,
+      "idity": 11639,
+      "▁Conference": 11640,
+      "▁gone": 11641,
+      "▁excited": 11642,
+      "千克": 11643,
+      "条规定": 11644,
+      "▁Richard": 11645,
+      "▁serial": 11646,
+      "中国的": 11647,
+      "▁Does": 11648,
+      "出发": 11649,
+      "cence": 11650,
+      "周边": 11651,
+      "▁agency": 11652,
+      "aching": 11653,
+      "ito": 11654,
+      "▁surgical": 11655,
+      "aniel": 11656,
+      "▁Moreover": 11657,
+      "▁colon": 11658,
+      "uls": 11659,
+      "ointer": 11660,
+      "房子": 11661,
+      "▁Jew": 11662,
+      "介绍了": 11663,
+      "JECT": 11664,
+      "acket": 11665,
+      "▁lawy": 11666,
+      "train": 11667,
+      "$$,": 11668,
+      "▁Bur": 11669,
+      "▁consumption": 11670,
+      "▁arm": 11671,
+      "SM": 11672,
+      "strip": 11673,
+      "HR": 11674,
+      "▁Brown": 11675,
+      "葡萄": 11676,
+      "明星": 11677,
+      "zz": 11678,
+      "银行存款": 11679,
+      "▁Secret": 11680,
+      "三十": 11681,
+      "基层": 11682,
+      "环节": 11683,
+      "▁brother": 11684,
+      "中国人民": 11685,
+      "oston": 11686,
+      "network": 11687,
+      "▁Spring": 11688,
+      "▁Kingdom": 11689,
+      "▁coron": 11690,
+      "POST": 11691,
+      "渠道": 11692,
+      "nInd": 11693,
+      "AF": 11694,
+      "达成": 11695,
+      "现任": 11696,
+      "▁Stat": 11697,
+      "女孩": 11698,
+      "用品": 11699,
+      "▁Corpor": 11700,
+      "▁friendly": 11701,
+      "开心": 11702,
+      "一名": 11703,
+      "scription": 11704,
+      "转化": 11705,
+      "▁soc": 11706,
+      "交付": 11707,
+      "▁reject": 11708,
+      "休闲": 11709,
+      "mosph": 11710,
+      "enger": 11711,
+      "ikipedia": 11712,
+      "▁claims": 11713,
+      "▁Centre": 11714,
+      "▁efficacy": 11715,
+      "▁university": 11716,
+      "Style": 11717,
+      "▁presentation": 11718,
+      "▁enhanced": 11719,
+      "▁divers": 11720,
+      "▁signs": 11721,
+      "fn": 11722,
+      "国土": 11723,
+      "▁supporting": 11724,
+      "UP": 11725,
+      "▁trigger": 11726,
+      "▁chart": 11727,
+      "▁officials": 11728,
+      "]{": 11729,
+      "hu": 11730,
+      "omics": 11731,
+      "▁island": 11732,
+      "ntitle": 11733,
+      "▁funds": 11734,
+      "______": 11735,
+      "▁radius": 11736,
+      "▁retro": 11737,
+      "uality": 11738,
+      "nAbstract": 11739,
+      "nk": 11740,
+      "治理": 11741,
+      "Card": 11742,
+      "▁Command": 11743,
+      "▁attached": 11744,
+      "widget": 11745,
+      "▁substr": 11746,
+      "▁Tre": 11747,
+      "难以": 11748,
+      "调节": 11749,
+      "▁Pass": 11750,
+      "剧情": 11751,
+      "刺激": 11752,
+      "▁neurons": 11753,
+      "▁policies": 11754,
+      "供述": 11755,
+      "EG": 11756,
+      "▁ended": 11757,
+      "▁UN": 11758,
+      "▁east": 11759,
+      "▁Search": 11760,
+      "▁pathways": 11761,
+      "foo": 11762,
+      "isor": 11763,
+      "OM": 11764,
+      "▁Lo": 11765,
+      "nu": 11766,
+      "专门": 11767,
+      "nThey": 11768,
+      "▁valu": 11769,
+      "▁sounds": 11770,
+      "▁emergency": 11771,
+      "▁bright": 11772,
+      "由此": 11773,
+      "icode": 11774,
+      "这个问题": 11775,
+      "失败": 11776,
+      "acellular": 11777,
+      "▁creation": 11778,
+      "scan": 11779,
+      "Frame": 11780,
+      "aired": 11781,
+      "When": 11782,
+      "license": 11783,
+      "公交": 11784,
+      "mi": 11785,
+      "▁statistics": 11786,
+      "▁expertise": 11787,
+      "▁percentage": 11788,
+      "什么是": 11789,
+      "货币": 11790,
+      "▁accomp": 11791,
+      "(\\'/": 11792,
+      "饮食": 11793,
+      "本地": 11794,
+      "▁Wall": 11795,
+      "Settings": 11796,
+      "▁dynamics": 11797,
+      "▁registered": 11798,
+      "强度": 11799,
+      "iability": 11800,
+      "▁embed": 11801,
+      "▁Ireland": 11802,
+      "转移": 11803,
+      "好好": 11804,
+      "▁inner": 11805,
+      "▁sensor": 11806,
+      "合同纠纷一案中": 11807,
+      "▁apps": 11808,
+      "混合": 11809,
+      "ERR": 11810,
+      "▁featured": 11811,
+      "▁voltage": 11812,
+      "nZ": 11813,
+      "现金": 11814,
+      "▁episode": 11815,
+      "▁Bal": 11816,
+      "▁minimal": 11817,
+      "▁fold": 11818,
+      "Do": 11819,
+      "roud": 11820,
+      "▁Championships": 11821,
+      "外的": 11822,
+      "▁Media": 11823,
+      "Are": 11824,
+      "▁dependent": 11825,
+      "▁citiz": 11826,
+      "▁grade": 11827,
+      "▁incred": 11828,
+      "▁considering": 11829,
+      "运算": 11830,
+      "nr": 11831,
+      "Schema": 11832,
+      "▁leadership": 11833,
+      "怎么办": 11834,
+      "说是": 11835,
+      "体积": 11836,
+      "第一百五十四条第一款第": 11837,
+      "荣誉": 11838,
+      "▁Italy": 11839,
+      "mal": 11840,
+      "gamma": 11841,
+      "决策": 11842,
+      "▁suspect": 11843,
+      "高考": 11844,
+      "IST": 11845,
+      "▁conj": 11846,
+      "▁clar": 11847,
+      "▁History": 11848,
+      "▁Scot": 11849,
+      "▁Atl": 11850,
+      "prop": 11851,
+      "white": 11852,
+      "▁audience": 11853,
+      "nThank": 11854,
+      "▁Three": 11855,
+      "tools": 11856,
+      "▁Amazon": 11857,
+      "children": 11858,
+      "冠军": 11859,
+      "为由": 11860,
+      "收益": 11861,
+      "▁Ret": 11862,
+      "主要是": 11863,
+      "▁fuel": 11864,
+      "内存": 11865,
+      "器的": 11866,
+      "▁fundamental": 11867,
+      "▁HD": 11868,
+      "实用": 11869,
+      "率为": 11870,
+      "▁Library": 11871,
+      "否则": 11872,
+      "▁ris": 11873,
+      "aka": 11874,
+      "avity": 11875,
+      "▁Iran": 11876,
+      "事项": 11877,
+      "▁shopping": 11878,
+      "▁Security": 11879,
+      "aturing": 11880,
+      "▁versus": 11881,
+      "▁knew": 11882,
+      "▁institution": 11883,
+      "▁subset": 11884,
+      "▁stem": 11885,
+      "路径": 11886,
+      "estyle": 11887,
+      "▁advent": 11888,
+      "▁screening": 11889,
+      "▁Pay": 11890,
+      "\"),\\": 11891,
+      "Game": 11892,
+      "▁epid": 11893,
+      "▁doctor": 11894,
+      "呈现": 11895,
+      "▁swim": 11896,
+      "combin": 11897,
+      "mac": 11898,
+      "代入": 11899,
+      "elect": 11900,
+      "▁neural": 11901,
+      "加快": 11902,
+      "室内": 11903,
+      "最好的": 11904,
+      "▁sequences": 11905,
+      "有多": 11906,
+      "▁Unless": 11907,
+      "ventions": 11908,
+      "再审": 11909,
+      "▁Non": 11910,
+      "SV": 11911,
+      "umps": 11912,
+      "▁teachers": 11913,
+      "ima": 11914,
+      "工作人员": 11915,
+      "一体": 11916,
+      "迟延": 11917,
+      "ODO": 11918,
+      "回到": 11919,
+      "指定的": 11920,
+      "仍然": 11921,
+      "▁stars": 11922,
+      "爸爸": 11923,
+      "▁Arts": 11924,
+      "过来": 11925,
+      "上下": 11926,
+      "在此": 11927,
+      "\\'),\\": 11928,
+      "▁raise": 11929,
+      "▁keys": 11930,
+      "▁whatever": 11931,
+      "tags": 11932,
+      "金额": 11933,
+      "都没有": 11934,
+      "eds": 11935,
+      "▁Spanish": 11936,
+      "▁人物": 11937,
+      "▁cardiac": 11938,
+      "SUL": 11939,
+      "ibraries": 11940,
+      "▁draft": 11941,
+      "▁TR": 11942,
+      "▁defe": 11943,
+      "omega": 11944,
+      "untu": 11945,
+      "▁wouldn": 11946,
+      "creen": 11947,
+      "\"{{": 11948,
+      "▁soil": 11949,
+      "色彩": 11950,
+      "izontal": 11951,
+      "onia": 11952,
+      "▁Server": 11953,
+      "byte": 11954,
+      "wan": 11955,
+      "};": 11956,
+      "▁proceed": 11957,
+      "很有": 11958,
+      "nAss": 11959,
+      "细节": 11960,
+      "发动": 11961,
+      "Empty": 11962,
+      "bro": 11963,
+      "▁Aff": 11964,
+      "博物": 11965,
+      "Buffer": 11966,
+      "▁tips": 11967,
+      "开发区": 11968,
+      "▁evening": 11969,
+      "uple": 11970,
+      "▁vot": 11971,
+      "▁crow": 11972,
+      "▁discount": 11973,
+      "有很多": 11974,
+      "iy": 11975,
+      "▁passion": 11976,
+      "}$,": 11977,
+      "甘肃": 11978,
+      "▁appointed": 11979,
+      "ipt": 11980,
+      "controller": 11981,
+      "知名": 11982,
+      "omatic": 11983,
+      "}}\\\\": 11984,
+      "▁extent": 11985,
+      "claim": 11986,
+      "▁Ha": 11987,
+      "▁stores": 11988,
+      "▁phil": 11989,
+      "mean": 11990,
+      "▁pharmac": 11991,
+      "信用卡": 11992,
+      "▁salt": 11993,
+      "internal": 11994,
+      "的同时": 11995,
+      "▁cognitive": 11996,
+      "故答案为": 11997,
+      "登录": 11998,
+      "▁signals": 11999,
+      "download": 12000,
+      "学位": 12001,
+      "回来": 12002,
+      "性别": 12003,
+      "},{\"": 12004,
+      "▁potentially": 12005,
+      "正当理由": 12006,
+      "▁na": 12007,
+      "精彩": 12008,
+      "omain": 12009,
+      "▁packages": 12010,
+      "▁basketball": 12011,
+      "数据库": 12012,
+      "▁CR": 12013,
+      "▁Asian": 12014,
+      "▁typical": 12015,
+      "▁candidates": 12016,
+      "▁hem": 12017,
+      "▁Cross": 12018,
+      "Generic": 12019,
+      "▁hearing": 12020,
+      "印度": 12021,
+      "oves": 12022,
+      "entials": 12023,
+      "TeX": 12024,
+      "tical": 12025,
+      "宁波": 12026,
+      "Ob": 12027,
+      "▁historical": 12028,
+      "▁optical": 12029,
+      "的关键": 12030,
+      "Can": 12031,
+      "▁Config": 12032,
+      "刑初字第": 12033,
+      "丰富的": 12034,
+      "▁recognized": 12035,
+      "专业的": 12036,
+      "预交": 12037,
+      "▁formal": 12038,
+      "lar": 12039,
+      "EFA": 12040,
+      "▁Abstract": 12041,
+      "▁ce": 12042,
+      "情况下": 12043,
+      "▁accum": 12044,
+      "▁populations": 12045,
+      "▁mir": 12046,
+      "上升": 12047,
+      "dig": 12048,
+      "▁variation": 12049,
+      "▁Prem": 12050,
+      "▁uniform": 12051,
+      "ifts": 12052,
+      "lint": 12053,
+      "▁Update": 12054,
+      "师范": 12055,
+      "▁mic": 12056,
+      "老人": 12057,
+      "房地产开发": 12058,
+      "▁boot": 12059,
+      "解题过程": 12060,
+      "▁diverse": 12061,
+      "▁dream": 12062,
+      "▁tumors": 12063,
+      "▁Kar": 12064,
+      "梦想": 12065,
+      "▁speech": 12066,
+      "▁params": 12067,
+      "未能": 12068,
+      "不起": 12069,
+      "共享": 12070,
+      "Body": 12071,
+      "▁veget": 12072,
+      "eller": 12073,
+      "化工": 12074,
+      "▁accommod": 12075,
+      "取消": 12076,
+      "型的": 12077,
+      "GR": 12078,
+      "▁holiday": 12079,
+      "idents": 12080,
+      "▁hash": 12081,
+      "tri": 12082,
+      "▁union": 12083,
+      "迟延履行": 12084,
+      "WORD": 12085,
+      "▁cleaning": 12086,
+      "▁disorder": 12087,
+      "▁Angeles": 12088,
+      "答辩": 12089,
+      "▁dir": 12090,
+      "地球": 12091,
+      "▁Online": 12092,
+      "▁bott": 12093,
+      "▁Global": 12094,
+      "▁urban": 12095,
+      "contrib": 12096,
+      "connection": 12097,
+      "▁closely": 12098,
+      "▁ww": 12099,
+      "hydro": 12100,
+      "▁remained": 12101,
+      "▁surrounding": 12102,
+      "▁lig": 12103,
+      "ennis": 12104,
+      "BD": 12105,
+      "SION": 12106,
+      "▁rooms": 12107,
+      "▁compact": 12108,
+      "互动": 12109,
+      "的研究": 12110,
+      "HTML": 12111,
+      "dl": 12112,
+      "agr": 12113,
+      "uous": 12114,
+      "emon": 12115,
+      "▁analog": 12116,
+      "生活中": 12117,
+      "第八": 12118,
+      "uke": 12119,
+      "oving": 12120,
+      "▁NE": 12121,
+      "▁toler": 12122,
+      "▁decide": 12123,
+      "artments": 12124,
+      "▁Eff": 12125,
+      "▁rub": 12126,
+      "INT": 12127,
+      "▁gender": 12128,
+      "▁roof": 12129,
+      "▁env": 12130,
+      "IX": 12131,
+      "▁audio": 12132,
+      "新华": 12133,
+      "▁Ep": 12134,
+      "审核": 12135,
+      "▁chief": 12136,
+      "▁wine": 12137,
+      "^\\\\": 12138,
+      "▁amb": 12139,
+      "▁accounts": 12140,
+      "iano": 12141,
+      "免疫": 12142,
+      "▁organic": 12143,
+      "▁pod": 12144,
+      "android": 12145,
+      "▁Adv": 12146,
+      "道德": 12147,
+      "▁newly": 12148,
+      "▁Cast": 12149,
+      "属性": 12150,
+      "ucket": 12151,
+      "sol": 12152,
+      "现在的": 12153,
+      "docker": 12154,
+      "cher": 12155,
+      "县人": 12156,
+      "无正当理由": 12157,
+      "▁radiation": 12158,
+      "▁Asia": 12159,
+      "▁reverse": 12160,
+      "nFil": 12161,
+      "▁Festival": 12162,
+      "mock": 12163,
+      "图书馆": 12164,
+      "东北": 12165,
+      "▁stages": 12166,
+      "直到": 12167,
+      "isters": 12168,
+      "许可": 12169,
+      "nThen": 12170,
+      "Child": 12171,
+      "▁river": 12172,
+      "▁instant": 12173,
+      "ascript": 12174,
+      "▁intellig": 12175,
+      "示范": 12176,
+      "HD": 12177,
+      "▁filled": 12178,
+      "▁Men": 12179,
+      "收费": 12180,
+      "Collection": 12181,
+      "的其他": 12182,
+      "▁×": 12183,
+      "归还": 12184,
+      "▁rac": 12185,
+      "▁classification": 12186,
+      "MR": 12187,
+      "看待": 12188,
+      "▁transformation": 12189,
+      "工作者": 12190,
+      "▁snow": 12191,
+      "nRec": 12192,
+      "▁Action": 12193,
+      "每一个": 12194,
+      "ixel": 12195,
+      "▁covers": 12196,
+      "▁tissues": 12197,
+      "▁seasons": 12198,
+      "▁adopt": 12199,
+      ")}{": 12200,
+      "▁YO": 12201,
+      "▁compat": 12202,
+      "每一": 12203,
+      "Position": 12204,
+      "▁prepare": 12205,
+      "▁scope": 12206,
+      "nex": 12207,
+      "▁Bas": 12208,
+      "故意": 12209,
+      "▁resulted": 12210,
+      "▁poll": 12211,
+      "进口": 12212,
+      "幼儿园": 12213,
+      "▁cris": 12214,
+      "▁explan": 12215,
+      "monary": 12216,
+      "河北": 12217,
+      "▁ton": 12218,
+      "▁compr": 12219,
+      "责令": 12220,
+      "▁peak": 12221,
+      "在我": 12222,
+      "▁Return": 12223,
+      "感到": 12224,
+      "不良": 12225,
+      "Program": 12226,
+      "ashboard": 12227,
+      "▁Times": 12228,
+      "ais": 12229,
+      "▁renal": 12230,
+      "Users": 12231,
+      "pred": 12232,
+      "▁owned": 12233,
+      "▁crypt": 12234,
+      "该犯": 12235,
+      "jo": 12236,
+      "▁markets": 12237,
+      "▁Tw": 12238,
+      "SR": 12239,
+      "ugby": 12240,
+      "roduction": 12241,
+      "▁dimension": 12242,
+      "loor": 12243,
+      "估计": 12244,
+      "场景": 12245,
+      "▁medicine": 12246,
+      "▁complement": 12247,
+      "▁mess": 12248,
+      "nLast": 12249,
+      "cha": 12250,
+      "羁押": 12251,
+      "▁CT": 12252,
+      "▁limits": 12253,
+      "ulum": 12254,
+      "▁Province": 12255,
+      "orney": 12256,
+      "本院在执行": 12257,
+      "大道": 12258,
+      "Database": 12259,
+      "能量": 12260,
+      "Enum": 12261,
+      "BP": 12262,
+      "iples": 12263,
+      "干净": 12264,
+      "真是": 12265,
+      "nAdd": 12266,
+      "▁Prot": 12267,
+      "▁cluster": 12268,
+      "▁destroy": 12269,
+      "▁silver": 12270,
+      "执行通知书": 12271,
+      "等方面": 12272,
+      "时刻": 12273,
+      "▁possibility": 12274,
+      "▁Solve": 12275,
+      "和平": 12276,
+      "表演": 12277,
+      "lat": 12278,
+      "▁remainder": 12279,
+      "WE": 12280,
+      "监测": 12281,
+      "Print": 12282,
+      "uis": 12283,
+      "解放": 12284,
+      "▁behaviour": 12285,
+      "购物": 12286,
+      "(\"/": 12287,
+      "▁Table": 12288,
+      "▁LO": 12289,
+      "Route": 12290,
+      "▁recover": 12291,
+      "▁console": 12292,
+      "判决书送达之日起": 12293,
+      "bel": 12294,
+      "appy": 12295,
+      "testing": 12296,
+      "缺乏": 12297,
+      "nson": 12298,
+      "terior": 12299,
+      "reatment": 12300,
+      "轴承": 12301,
+      "ICE": 12302,
+      "▁restaurant": 12303,
+      "江西省": 12304,
+      "UM": 12305,
+      "%;\\": 12306,
+      "READ": 12307,
+      "食用": 12308,
+      "▁topics": 12309,
+      "的产品": 12310,
+      "条第二款": 12311,
+      "area": 12312,
+      "联盟": 12313,
+      "▁primarily": 12314,
+      "▁vul": 12315,
+      "▁accident": 12316,
+      "适当": 12317,
+      "▁hybrid": 12318,
+      "似乎": 12319,
+      "餐厅": 12320,
+      "errors": 12321,
+      "CF": 12322,
+      "▁Git": 12323,
+      "宠物": 12324,
+      "男生": 12325,
+      "aped": 12326,
+      "▁ta": 12327,
+      "▁Arab": 12328,
+      "▁stated": 12329,
+      "▁listing": 12330,
+      "过了": 12331,
+      "▁Exec": 12332,
+      "二条": 12333,
+      "mes": 12334,
+      "汉语": 12335,
+      "咖啡": 12336,
+      "▁prolif": 12337,
+      "▁Ram": 12338,
+      "ULT": 12339,
+      "担心": 12340,
+      "▁Last": 12341,
+      "▁reducing": 12342,
+      "▁Ken": 12343,
+      "▁ER": 12344,
+      "rh": 12345,
+      "▁\"\\\\": 12346,
+      "航空": 12347,
+      "▁describes": 12348,
+      "社会主义": 12349,
+      "rawn": 12350,
+      "▁electrical": 12351,
+      "▁perspective": 12352,
+      "▁merge": 12353,
+      "求解": 12354,
+      "shot": 12355,
+      "utor": 12356,
+      "▁polar": 12357,
+      "▁Edit": 12358,
+      "ocard": 12359,
+      "atar": 12360,
+      "共产": 12361,
+      "▁awarded": 12362,
+      "▁Language": 12363,
+      "nEng": 12364,
+      "▁www": 12365,
+      "Storage": 12366,
+      "ycle": 12367,
+      "▁ran": 12368,
+      "前的": 12369,
+      "穿越": 12370,
+      "nFirst": 12371,
+      "部长": 12372,
+      "▁cruc": 12373,
+      "ova": 12374,
+      "现状": 12375,
+      "▁speak": 12376,
+      "大众": 12377,
+      "erts": 12378,
+      "▁independ": 12379,
+      "某些": 12380,
+      "期的": 12381,
+      "含有": 12382,
+      "▁interact": 12383,
+      "▁improving": 12384,
+      "ст": 12385,
+      "Each": 12386,
+      "▁sizes": 12387,
+      "▁所属": 12388,
+      "辅助": 12389,
+      "合适的": 12390,
+      "舒适": 12391,
+      "▁pp": 12392,
+      "services": 12393,
+      "▁Pop": 12394,
+      "prefix": 12395,
+      "etics": 12396,
+      "较大": 12397,
+      "▁Pac": 12398,
+      "tdef": 12399,
+      "ti": 12400,
+      "PG": 12401,
+      "▁javax": 12402,
+      "下面的问题": 12403,
+      "star": 12404,
+      "una": 12405,
+      "厦门": 12406,
+      "imation": 12407,
+      "Custom": 12408,
+      "nCont": 12409,
+      "不在": 12410,
+      "ircraft": 12411,
+      "这款": 12412,
+      "▁median": 12413,
+      "schema": 12414,
+      "湖北": 12415,
+      "美食": 12416,
+      "lf": 12417,
+      "▁Master": 12418,
+      "位数": 12419,
+      "▁flight": 12420,
+      "协助": 12421,
+      "老板": 12422,
+      "▁spectrum": 12423,
+      "iography": 12424,
+      "zone": 12425,
+      "平行": 12426,
+      "▁Click": 12427,
+      "▁phenomen": 12428,
+      "生存": 12429,
+      "异常": 12430,
+      "▁para": 12431,
+      "▁strongly": 12432,
+      "生效后": 12433,
+      "▁accepted": 12434,
+      "单调": 12435,
+      "想法": 12436,
+      "请回答下面的问题": 12437,
+      "▁Link": 12438,
+      "分别为": 12439,
+      "第九": 12440,
+      "plus": 12441,
+      "ONE": 12442,
+      "lower": 12443,
+      "ynamic": 12444,
+      "▁familiar": 12445,
+      "Que": 12446,
+      "bling": 12447,
+      "兄弟": 12448,
+      "▁knows": 12449,
+      "一旦": 12450,
+      "NG": 12451,
+      "nBe": 12452,
+      "Player": 12453,
+      "zy": 12454,
+      "的要求": 12455,
+      "\\\\()": 12456,
+      "▁effectiveness": 12457,
+      "▁favour": 12458,
+      "密码": 12459,
+      "embers": 12460,
+      "▁blocks": 12461,
+      "▁api": 12462,
+      "这就是": 12463,
+      "take": 12464,
+      "相应": 12465,
+      "就在": 12466,
+      "ospit": 12467,
+      "指定": 12468,
+      "▁guy": 12469,
+      "▁Est": 12470,
+      "▁executive": 12471,
+      "▁BE": 12472,
+      "▁rough": 12473,
+      "▁announce": 12474,
+      "陕西省": 12475,
+      "太多": 12476,
+      "有机": 12477,
+      "riev": 12478,
+      "▁delivered": 12479,
+      "债务利息": 12480,
+      "▁competitive": 12481,
+      "认可": 12482,
+      "白色": 12483,
+      "▁prevention": 12484,
+      "▁incidence": 12485,
+      "▁prompt": 12486,
+      "▁infrastructure": 12487,
+      "▁Martin": 12488,
+      "▁solar": 12489,
+      "链接": 12490,
+      "LECT": 12491,
+      "▁tables": 12492,
+      "问题的答案": 12493,
+      "rub": 12494,
+      "▁insulin": 12495,
+      "课堂": 12496,
+      "Width": 12497,
+      "适用于": 12498,
+      "Ab": 12499,
+      "部署": 12500,
+      "方式结案": 12501,
+      "尝试": 12502,
+      "▁Vari": 12503,
+      "▁mechanical": 12504,
+      "用来": 12505,
+      "集成": 12506,
+      "▁battery": 12507,
+      "规格": 12508,
+      "▁ingred": 12509,
+      "项的规定": 12510,
+      "▁cup": 12511,
+      "党的": 12512,
+      "采购": 12513,
+      "dp": 12514,
+      "Car": 12515,
+      "▁styles": 12516,
+      "irus": 12517,
+      "纪念": 12518,
+      "不住": 12519,
+      "围绕": 12520,
+      "▁charges": 12521,
+      "▁institutions": 12522,
+      "router": 12523,
+      "▁Human": 12524,
+      "元及": 12525,
+      "▁trained": 12526,
+      "nSome": 12527,
+      "全体": 12528,
+      "▁latter": 12529,
+      "uh": 12530,
+      "▁Beach": 12531,
+      "解除": 12532,
+      "或许": 12533,
+      "ylvan": 12534,
+      "orrect": 12535,
+      "拒绝": 12536,
+      "cpp": 12537,
+      "leases": 12538,
+      "▁categories": 12539,
+      "▁beauty": 12540,
+      "本裁定书": 12541,
+      "sa": 12542,
+      "都市": 12543,
+      "Home": 12544,
+      "▁selling": 12545,
+      "▁danger": 12546,
+      "Graph": 12547,
+      "▁enum": 12548,
+      "nAlso": 12549,
+      "清洁": 12550,
+      "ansas": 12551,
+      "出版社出版的图书": 12552,
+      "nAny": 12553,
+      "anth": 12554,
+      "人力": 12555,
+      "▁exposed": 12556,
+      "▁出": 12557,
+      "▁inhibition": 12558,
+      "▁decades": 12559,
+      "身份证": 12560,
+      "味道": 12561,
+      "bsite": 12562,
+      "说话": 12563,
+      "我会": 12564,
+      "Cache": 12565,
+      "旅行": 12566,
+      "观点": 12567,
+      "▁lat": 12568,
+      "▁excess": 12569,
+      "▁tournament": 12570,
+      "等待": 12571,
+      "▁DC": 12572,
+      "▁stimul": 12573,
+      "▁dating": 12574,
+      "▁furniture": 12575,
+      "Vari": 12576,
+      "不一样": 12577,
+      "收藏": 12578,
+      "▁subsequent": 12579,
+      "rollers": 12580,
+      "▁html": 12581,
+      "▁Charles": 12582,
+      "nDe": 12583,
+      "▁committed": 12584,
+      "▁msg": 12585,
+      "Password": 12586,
+      "晋江文学": 12587,
+      "▁lit": 12588,
+      "edu": 12589,
+      "▁aggreg": 12590,
+      "▁simult": 12591,
+      "▁beat": 12592,
+      "▁deals": 12593,
+      "▁perman": 12594,
+      "服务器": 12595,
+      "nic": 12596,
+      "宣布": 12597,
+      "plet": 12598,
+      "▁choices": 12599,
+      "upy": 12600,
+      "▁Head": 12601,
+      "邀请": 12602,
+      "×××": 12603,
+      "tool": 12604,
+      "nSuppose": 12605,
+      "▁Mur": 12606,
+      "▁valuable": 12607,
+      "tex": 12608,
+      "▁athlet": 12609,
+      "▁bought": 12610,
+      "两人": 12611,
+      "应该是": 12612,
+      "可靠": 12613,
+      "家人": 12614,
+      "华为": 12615,
+      "ersonal": 12616,
+      "▁ends": 12617,
+      "▁scored": 12618,
+      "▁asym": 12619,
+      "note": 12620,
+      "▁hardware": 12621,
+      "itness": 12622,
+      "▁Film": 12623,
+      "开启": 12624,
+      "生素": 12625,
+      "achment": 12626,
+      "logging": 12627,
+      "vals": 12628,
+      "识别": 12629,
+      "ressive": 12630,
+      "▁pig": 12631,
+      "）（": 12632,
+      "生于": 12633,
+      "blue": 12634,
+      "▁particles": 12635,
+      "ussian": 12636,
+      "说了": 12637,
+      "▁vess": 12638,
+      "Am": 12639,
+      "▁reson": 12640,
+      "撤销": 12641,
+      "▁tempor": 12642,
+      "▁collaboration": 12643,
+      "总体": 12644,
+      "nSubject": 12645,
+      "第五十": 12646,
+      "准予": 12647,
+      "▁thr": 12648,
+      "onds": 12649,
+      "cludes": 12650,
+      "writer": 12651,
+      "乡村": 12652,
+      "▁spons": 12653,
+      "▁spatial": 12654,
+      "▁seek": 12655,
+      "aya": 12656,
+      "handler": 12657,
+      "第二百五十七条": 12658,
+      "▁Study": 12659,
+      "legal": 12660,
+      "urable": 12661,
+      "▁assistance": 12662,
+      "融资": 12663,
+      "heast": 12664,
+      "PH": 12665,
+      "手续": 12666,
+      "ifferent": 12667,
+      "▁robust": 12668,
+      "决定书": 12669,
+      "当年": 12670,
+      "更好的": 12671,
+      "物业服务": 12672,
+      "▁exciting": 12673,
+      "中小": 12674,
+      "▁coh": 12675,
+      "osoph": 12676,
+      "ea": 12677,
+      "pull": 12678,
+      "的关系": 12679,
+      "足够": 12680,
+      "▁NS": 12681,
+      "▁datetime": 12682,
+      "▁rural": 12683,
+      "这道": 12684,
+      "▁proud": 12685,
+      "▁maps": 12686,
+      "立即生效": 12687,
+      "评审": 12688,
+      "▁lin": 12689,
+      "笔记": 12690,
+      "Mark": 12691,
+      "▁inher": 12692,
+      "▁Young": 12693,
+      "Art": 12694,
+      "uing": 12695,
+      "Helper": 12696,
+      "▁transplant": 12697,
+      "icken": 12698,
+      "以调解": 12699,
+      "▁hypothes": 12700,
+      "公民": 12701,
+      "转换": 12702,
+      "▁Invest": 12703,
+      "正确的答案": 12704,
+      "olds": 12705,
+      "▁purposes": 12706,
+      "▁Indust": 12707,
+      "▁Historic": 12708,
+      "各位": 12709,
+      "war": 12710,
+      "接下来": 12711,
+      "▁approved": 12712,
+      "obs": 12713,
+      "ión": 12714,
+      "以调解方式结案": 12715,
+      "盗窃罪": 12716,
+      "▁statistical": 12717,
+      "之中": 12718,
+      "▁职业": 12719,
+      "▁Py": 12720,
+      "▁aged": 12721,
+      "判令": 12722,
+      "▁expensive": 12723,
+      "watch": 12724,
+      "CV": 12725,
+      "sha": 12726,
+      "▁guests": 12727,
+      "元的": 12728,
+      "▁atmosph": 12729,
+      "古代": 12730,
+      "是对": 12731,
+      "虚拟": 12732,
+      "HOU": 12733,
+      "zure": 12734,
+      "▁dogs": 12735,
+      "▁partnership": 12736,
+      "▁stone": 12737,
+      "asts": 12738,
+      "▁complications": 12739,
+      "▁examination": 12740,
+      "POSE": 12741,
+      "Sign": 12742,
+      "▁\\'/": 12743,
+      "wa": 12744,
+      "日子": 12745,
+      "▁Description": 12746,
+      "cb": 12747,
+      "停车": 12748,
+      "▁suc": 12749,
+      "同一": 12750,
+      "合格": 12751,
+      "Author": 12752,
+      "ooking": 12753,
+      "怎么样": 12754,
+      "并按对方当事人的人数提出副本": 12755,
+      "▁attended": 12756,
+      "Engine": 12757,
+      "曲线": 12758,
+      "写作": 12759,
+      "众多": 12760,
+      "▁Net": 12761,
+      "▁capable": 12762,
+      "}]": 12763,
+      "车程": 12764,
+      "ORM": 12765,
+      "▁cin": 12766,
+      "vironments": 12767,
+      "nYour": 12768,
+      "▁listen": 12769,
+      "▁usual": 12770,
+      "▁officer": 12771,
+      "arly": 12772,
+      "ategor": 12773,
+      "▁黄": 12774,
+      "以调解方式结案的": 12775,
+      "▁establishments": 12776,
+      "判决书送达之日起十五日内": 12777,
+      "▁shut": 12778,
+      "▁boundary": 12779,
+      "▁mRNA": 12780,
+      "保留": 12781,
+      "headers": 12782,
+      "官网": 12783,
+      "Account": 12784,
+      "面试": 12785,
+      "胜利": 12786,
+      "毫米": 12787,
+      "血液": 12788,
+      "▁broken": 12789,
+      "▁nom": 12790,
+      "exists": 12791,
+      "▁Pak": 12792,
+      "▁Path": 12793,
+      "▁主": 12794,
+      "TD": 12795,
+      "▁MS": 12796,
+      "isted": 12797,
+      "ylvania": 12798,
+      "黑色": 12799,
+      "eah": 12800,
+      "彻底": 12801,
+      "eric": 12802,
+      "ongs": 12803,
+      "标的": 12804,
+      "笔录": 12805,
+      "节点": 12806,
+      "▁tow": 12807,
+      "sample": 12808,
+      "危害": 12809,
+      "outs": 12810,
+      "&&": 12811,
+      "▁sed": 12812,
+      "恋爱": 12813,
+      "经营范围包括": 12814,
+      "Activity": 12815,
+      "民主": 12816,
+      "%),": 12817,
+      "report": 12818,
+      "创作的": 12819,
+      "▁Local": 12820,
+      "▁explicit": 12821,
+      "进程": 12822,
+      "▁waste": 12823,
+      "aylor": 12824,
+      "Models": 12825,
+      "▁brings": 12826,
+      "orem": 12827,
+      "▁neighborhood": 12828,
+      "etal": 12829,
+      "会上": 12830,
+      "▁HTML": 12831,
+      "▁regulatory": 12832,
+      "▁worldwide": 12833,
+      "Current": 12834,
+      "▁Haw": 12835,
+      "征收": 12836,
+      "▁extend": 12837,
+      "platform": 12838,
+      "rand": 12839,
+      "gi": 12840,
+      "在这里": 12841,
+      "▁entity": 12842,
+      "Bo": 12843,
+      "不等式": 12844,
+      "aced": 12845,
+      "执行员": 12846,
+      "▁trav": 12847,
+      "|\\\\": 12848,
+      "▁Ohio": 12849,
+      "动画": 12850,
+      "▁uncertain": 12851,
+      "ND": 12852,
+      "▁assay": 12853,
+      "▁interior": 12854,
+      "▁IF": 12855,
+      "\"))\\": 12856,
+      "RON": 12857,
+      "考生": 12858,
+      "▁plenty": 12859,
+      "▁repeated": 12860,
+      "▁Access": 12861,
+      "iration": 12862,
+      "▁vascular": 12863,
+      "toString": 12864,
+      "▁Ga": 12865,
+      "完美": 12866,
+      "▁quantum": 12867,
+      "security": 12868,
+      "▁Hi": 12869,
+      "Plugin": 12870,
+      "的故事": 12871,
+      "igin": 12872,
+      "▁SD": 12873,
+      "usic": 12874,
+      "▁explained": 12875,
+      "ovascular": 12876,
+      "▁stored": 12877,
+      "Document": 12878,
+      "普遍": 12879,
+      "夫妻感情": 12880,
+      "情感": 12881,
+      "损伤": 12882,
+      "▁tight": 12883,
+      "注重": 12884,
+      "▁Process": 12885,
+      "▁Tri": 12886,
+      "▁persons": 12887,
+      "向被执行人": 12888,
+      "车站": 12889,
+      "▁Four": 12890,
+      "加倍": 12891,
+      "▁rig": 12892,
+      "jav": 12893,
+      "▁chat": 12894,
+      "uilt": 12895,
+      "一场": 12896,
+      "型号": 12897,
+      "▁stret": 12898,
+      "▁Install": 12899,
+      "▁zone": 12900,
+      "▁instructions": 12901,
+      "▁Northern": 12902,
+      "pin": 12903,
+      "▁writers": 12904,
+      "stood": 12905,
+      "uite": 12906,
+      "消防": 12907,
+      "二年": 12908,
+      "▁作品": 12909,
+      "▁Li": 12910,
+      "▁Environment": 12911,
+      "pages": 12912,
+      "▁bin": 12913,
+      "▁incident": 12914,
+      "fire": 12915,
+      "▁Lee": 12916,
+      "尽量": 12917,
+      "region": 12918,
+      "tor": 12919,
+      "一边": 12920,
+      "ifiers": 12921,
+      "▁Boston": 12922,
+      "▁participate": 12923,
+      "区人民检察院": 12924,
+      "ancel": 12925,
+      "有限公司于": 12926,
+      "频率": 12927,
+      "▁Nav": 12928,
+      "熟悉": 12929,
+      "公司经营范围包括": 12930,
+      ")[": 12931,
+      "▁aimed": 12932,
+      "▁abnormal": 12933,
+      "▁beach": 12934,
+      "一路": 12935,
+      "Mult": 12936,
+      "户籍": 12937,
+      "发的": 12938,
+      "▁hidden": 12939,
+      "ela": 12940,
+      "ndes": 12941,
+      "▁manual": 12942,
+      "▁providers": 12943,
+      "综上": 12944,
+      "并在": 12945,
+      "▁diagnostic": 12946,
+      "▁Assembly": 12947,
+      "深度": 12948,
+      "ACE": 12949,
+      "▁Systems": 12950,
+      "▁Ltd": 12951,
+      "数组": 12952,
+      "▁eld": 12953,
+      "Ge": 12954,
+      "▁constructed": 12955,
+      "诉讼费用": 12956,
+      "便宜": 12957,
+      "▁differentiation": 12958,
+      "▁profit": 12959,
+      "▁drivers": 12960,
+      "ucks": 12961,
+      "▁awesome": 12962,
+      "另一个": 12963,
+      "cover": 12964,
+      "”,": 12965,
+      "▁Help": 12966,
+      "一审": 12967,
+      "adium": 12968,
+      "诉称": 12969,
+      "Target": 12970,
+      "▁assigned": 12971,
+      "的设计": 12972,
+      "▁scheduled": 12973,
+      "nAr": 12974,
+      "Enc": 12975,
+      "时光": 12976,
+      "▁Catholic": 12977,
+      "DNA": 12978,
+      "▁Jim": 12979,
+      "▁mes": 12980,
+      "SER": 12981,
+      "云南省": 12982,
+      "▁spending": 12983,
+      "复合": 12984,
+      "▁particle": 12985,
+      "人间": 12986,
+      "的基础上": 12987,
+      "▁func": 12988,
+      "▁pand": 12989,
+      "▁Rest": 12990,
+      "驱动": 12991,
+      "▁depos": 12992,
+      "毒品": 12993,
+      "gers": 12994,
+      "▁Columb": 12995,
+      "▁distribut": 12996,
+      "{(": 12997,
+      "▁entered": 12998,
+      "deg": 12999,
+      "rolling": 13000,
+      "blank": 13001,
+      "森林": 13002,
+      "国外": 13003,
+      "▁housing": 13004,
+      "▁fruit": 13005,
+      "oids": 13006,
+      "安市": 13007,
+      "字符": 13008,
+      "▁illustr": 13009,
+      "mo": 13010,
+      "支撑": 13011,
+      "而不是": 13012,
+      "mediate": 13013,
+      "不上": 13014,
+      "▁ec": 13015,
+      "▁walking": 13016,
+      "score": 13017,
+      "▁involves": 13018,
+      "▁sessions": 13019,
+      "▁layout": 13020,
+      "▁PDF": 13021,
+      "▁aid": 13022,
+      "▁trading": 13023,
+      "▁merg": 13024,
+      "说法": 13025,
+      "▁repository": 13026,
+      "不太": 13027,
+      "agger": 13028,
+      "▁Ro": 13029,
+      "▁pm": 13030,
+      "Man": 13031,
+      "operator": 13032,
+      "reme": 13033,
+      "▁desired": 13034,
+      "▁undefined": 13035,
+      "operation": 13036,
+      "▁analyze": 13037,
+      "▁planned": 13038,
+      "博物馆": 13039,
+      "▁progression": 13040,
+      "塑料": 13041,
+      "video": 13042,
+      "nRE": 13043,
+      "sur": 13044,
+      "ulator": 13045,
+      "▁serves": 13046,
+      "metadata": 13047,
+      "▁corporate": 13048,
+      "现住": 13049,
+      "▁complexity": 13050,
+      "▁conversion": 13051,
+      "分解": 13052,
+      "理事": 13053,
+      "▁Plan": 13054,
+      "调解书": 13055,
+      "▁recording": 13056,
+      "▁inspired": 13057,
+      "ez": 13058,
+      "轻处罚": 13059,
+      "(`": 13060,
+      "▁utility": 13061,
+      "gence": 13062,
+      "prov": 13063,
+      "层次": 13064,
+      "破坏": 13065,
+      "foreach": 13066,
+      "随时": 13067,
+      "record": 13068,
+      "dirname": 13069,
+      "▁pump": 13070,
+      "epsilon": 13071,
+      "顾客": 13072,
+      "oir": 13073,
+      "▁iter": 13074,
+      "merce": 13075,
+      "▁gun": 13076,
+      "ells": 13077,
+      "▁lies": 13078,
+      "借条": 13079,
+      "▁Econ": 13080,
+      "▁NY": 13081,
+      "Location": 13082,
+      "贯彻": 13083,
+      "▁putting": 13084,
+      "icker": 13085,
+      "ferences": 13086,
+      "▁commission": 13087,
+      "▁cheap": 13088,
+      "Main": 13089,
+      "▁}}\\": 13090,
+      "ampion": 13091,
+      "Record": 13092,
+      "ocus": 13093,
+      "▁capture": 13094,
+      "拼音": 13095,
+      "▁AL": 13096,
+      "重复": 13097,
+      "让我们": 13098,
+      "杂志": 13099,
+      "同年": 13100,
+      "的一些": 13101,
+      "中国科学院": 13102,
+      "▁定价": 13103,
+      "人均": 13104,
+      "▁Have": 13105,
+      "Params": 13106,
+      "buffer": 13107,
+      "awa": 13108,
+      "▁lose": 13109,
+      "▁manufacturing": 13110,
+      "▁mamm": 13111,
+      "▁ongoing": 13112,
+      "▁initially": 13113,
+      "First": 13114,
+      "BL": 13115,
+      "蔬菜": 13116,
+      "cli": 13117,
+      "▁analyt": 13118,
+      "short": 13119,
+      "打车": 13120,
+      "对外": 13121,
+      "▁edited": 13122,
+      "constructor": 13123,
+      "▁genus": 13124,
+      "▁supposed": 13125,
+      "▁其一": 13126,
+      "日至": 13127,
+      "▁politician": 13128,
+      "别的": 13129,
+      "▁logic": 13130,
+      "▁boost": 13131,
+      "▁Sar": 13132,
+      "ighter": 13133,
+      "Match": 13134,
+      "tprint": 13135,
+      "possible": 13136,
+      "▁lists": 13137,
+      "▁glucose": 13138,
+      "▁outdoor": 13139,
+      "▁thin": 13140,
+      "▁optimization": 13141,
+      "▁Mi": 13142,
+      "▁copies": 13143,
+      "▁translation": 13144,
+      "家具": 13145,
+      "▁scal": 13146,
+      "期间的债务利息": 13147,
+      "oa": 13148,
+      "ен": 13149,
+      "有效的": 13150,
+      "乐部": 13151,
+      "▁charged": 13152,
+      "图像": 13153,
+      "inois": 13154,
+      "Email": 13155,
+      "▁Bul": 13156,
+      "▁triangle": 13157,
+      "▁lesions": 13158,
+      "▁crim": 13159,
+      "▁contained": 13160,
+      "▁removal": 13161,
+      "bec": 13162,
+      "验证": 13163,
+      "线上": 13164,
+      "▁turns": 13165,
+      "▁radi": 13166,
+      "▁preparation": 13167,
+      "▁buying": 13168,
+      "▁Application": 13169,
+      "ibilities": 13170,
+      "▁Michigan": 13171,
+      "hetic": 13172,
+      "FS": 13173,
+      "setup": 13174,
+      "nsylvania": 13175,
+      "metry": 13176,
+      "▁vertical": 13177,
+      "upload": 13178,
+      "▁guidelines": 13179,
+      "▁existence": 13180,
+      "▁controller": 13181,
+      "small": 13182,
+      "That": 13183,
+      "样例": 13184,
+      "▁classical": 13185,
+      "▁Bob": 13186,
+      "你要": 13187,
+      "gian": 13188,
+      "lik": 13189,
+      "▁Girl": 13190,
+      "esterday": 13191,
+      "▁Bud": 13192,
+      "▁MO": 13193,
+      "▁porn": 13194,
+      "orship": 13195,
+      "▁truth": 13196,
+      "chem": 13197,
+      "live": 13198,
+      "▁orders": 13199,
+      "俱乐部": 13200,
+      "nGe": 13201,
+      "▁Chief": 13202,
+      "nTrue": 13203,
+      "ocyte": 13204,
+      "▁employed": 13205,
+      "足以": 13206,
+      "▁kg": 13207,
+      "▁contents": 13208,
+      "PY": 13209,
+      "删除": 13210,
+      "▁thoughts": 13211,
+      "iph": 13212,
+      "如果是": 13213,
+      "▁align": 13214,
+      "▁平": 13215,
+      "会在": 13216,
+      "VAL": 13217,
+      "Local": 13218,
+      "▁cache": 13219,
+      "pres": 13220,
+      "不动": 13221,
+      "件事": 13222,
+      "Show": 13223,
+      "istence": 13224,
+      "▁fear": 13225,
+      "▁semi": 13226,
+      "现有": 13227,
+      "▁musical": 13228,
+      "▁fellow": 13229,
+      "▁commands": 13230,
+      "▁pairs": 13231,
+      "均匀": 13232,
+      "首次": 13233,
+      "stud": 13234,
+      "破裂": 13235,
+      "▁ri": 13236,
+      "是一家": 13237,
+      "▁encoding": 13238,
+      "▁milk": 13239,
+      "提醒": 13240,
+      "交通事故": 13241,
+      "▁swe": 13242,
+      "▁Like": 13243,
+      "▁Node": 13244,
+      "▁willing": 13245,
+      "offset": 13246,
+      "SET": 13247,
+      "共计": 13248,
+      "▁Determ": 13249,
+      "处分": 13250,
+      "▁protected": 13251,
+      "VA": 13252,
+      "起点中文网": 13253,
+      "lg": 13254,
+      "▁consumer": 13255,
+      "nof": 13256,
+      "贵州省": 13257,
+      "riers": 13258,
+      "最低": 13259,
+      "本院依法组成合议庭": 13260,
+      "生成": 13261,
+      "uj": 13262,
+      "pons": 13263,
+      "Ext": 13264,
+      "▁dram": 13265,
+      "▁displayed": 13266,
+      "肿瘤": 13267,
+      "nPost": 13268,
+      "▁Circ": 13269,
+      "▁indicates": 13270,
+      "▁Support": 13271,
+      "▁Option": 13272,
+      "primary": 13273,
+      "egers": 13274,
+      "▁公司名称": 13275,
+      "国民": 13276,
+      "货物": 13277,
+      "无人": 13278,
+      "continue": 13279,
+      "返回": 13280,
+      "▁reporting": 13281,
+      "when": 13282,
+      "▁observations": 13283,
+      "▁threshold": 13284,
+      "国务": 13285,
+      "花园": 13286,
+      "零售": 13287,
+      "execute": 13288,
+      "▁worry": 13289,
+      "意味": 13290,
+      "Row": 13291,
+      "Entry": 13292,
+      "印象": 13293,
+      "irms": 13294,
+      "uber": 13295,
+      "▁fee": 13296,
+      "folio": 13297,
+      "▁diameter": 13298,
+      "▁hundreds": 13299,
+      "缓刑": 13300,
+      "▁rational": 13301,
+      "cnt": 13302,
+      "ociated": 13303,
+      "▁Fed": 13304,
+      "▁Word": 13305,
+      "哲学": 13306,
+      "深刻": 13307,
+      "groups": 13308,
+      "▁antibodies": 13309,
+      "涉嫌": 13310,
+      "▁languages": 13311,
+      "endor": 13312,
+      "执法": 13313,
+      "▁SC": 13314,
+      "▁strain": 13315,
+      "yped": 13316,
+      "▁offset": 13317,
+      "院长": 13318,
+      "▁HE": 13319,
+      "▁Area": 13320,
+      "▁lived": 13321,
+      "oxy": 13322,
+      "▁dimensions": 13323,
+      "▁sections": 13324,
+      "▁counc": 13325,
+      "ban": 13326,
+      "述了": 13327,
+      "▁brown": 13328,
+      "pare": 13329,
+      "nthetic": 13330,
+      "stone": 13331,
+      "Render": 13332,
+      "Bundle": 13333,
+      "▁lux": 13334,
+      "UND": 13335,
+      "▁Wild": 13336,
+      "ра": 13337,
+      "▁opinion": 13338,
+      "高于": 13339,
+      "▁});\\": 13340,
+      "样子": 13341,
+      "▁\",": 13342,
+      "▁Er": 13343,
+      "▁Engineering": 13344,
+      "▁pin": 13345,
+      "igate": 13346,
+      "▁aer": 13347,
+      "监规": 13348,
+      "在这个": 13349,
+      "刚刚": 13350,
+      "▁CEO": 13351,
+      "▁quadr": 13352,
+      "鼓励": 13353,
+      "gov": 13354,
+      "Next": 13355,
+      "运动员": 13356,
+      "erves": 13357,
+      "连载于": 13358,
+      "▁Way": 13359,
+      "appe": 13360,
+      "▁locally": 13361,
+      "的感觉": 13362,
+      "互相": 13363,
+      "▁\"$": 13364,
+      "▁representing": 13365,
+      "流行": 13366,
+      "▁dual": 13367,
+      "▁mapping": 13368,
+      "She": 13369,
+      "Enron": 13370,
+      "\":{\"": 13371,
+      "▁git": 13372,
+      "▁absence": 13373,
+      "▁composite": 13374,
+      "▁fly": 13375,
+      "Top": 13376,
+      "物业管理有限公司": 13377,
+      "▁ads": 13378,
+      "DM": 13379,
+      "▁Lab": 13380,
+      "PASS": 13381,
+      "that": 13382,
+      "即将": 13383,
+      "这个问题的答案": 13384,
+      "▁metabolism": 13385,
+      "▁taste": 13386,
+      "Rel": 13387,
+      "▁Van": 13388,
+      "otox": 13389,
+      "bell": 13390,
+      "▁Give": 13391,
+      "▁Today": 13392,
+      "cam": 13393,
+      "aneously": 13394,
+      "就有": 13395,
+      "故障": 13396,
+      "▁columns": 13397,
+      "悔罪": 13398,
+      "sis": 13399,
+      "▁Through": 13400,
+      "▁Input": 13401,
+      "▁rom": 13402,
+      "▁Family": 13403,
+      "▁concern": 13404,
+      "▁watching": 13405,
+      "▁cabin": 13406,
+      ",,": 13407,
+      "行驶": 13408,
+      "▁renew": 13409,
+      "(()": 13410,
+      "eh": 13411,
+      "千米": 13412,
+      "▁metast": 13413,
+      "▁visitors": 13414,
+      "喜欢的": 13415,
+      "vin": 13416,
+      "▁Food": 13417,
+      "Dialog": 13418,
+      "看见": 13419,
+      "▁Next": 13420,
+      "专辑": 13421,
+      "▁SE": 13422,
+      "进而": 13423,
+      "无限": 13424,
+      "▁blank": 13425,
+      "GO": 13426,
+      "▁ordered": 13427,
+      "▁depression": 13428,
+      "写的": 13429,
+      "▁dance": 13430,
+      "难度": 13431,
+      "▁演员": 13432,
+      "我说": 13433,
+      "档案": 13434,
+      "黑龙江省": 13435,
+      "调研": 13436,
+      "▁Police": 13437,
+      "▁pregnancy": 13438,
+      "olver": 13439,
+      "▁chapter": 13440,
+      "▁functionality": 13441,
+      "实数": 13442,
+      "▁AC": 13443,
+      "▁assign": 13444,
+      "NET": 13445,
+      "▁其二": 13446,
+      "▁actor": 13447,
+      "oun": 13448,
+      "▁denomin": 13449,
+      "▁smallest": 13450,
+      "对手": 13451,
+      "▁matching": 13452,
+      "▁Mid": 13453,
+      "virtual": 13454,
+      "▁introduction": 13455,
+      "▁Jeff": 13456,
+      "正方": 13457,
+      "▁btn": 13458,
+      "清晰": 13459,
+      "▁IV": 13460,
+      "flu": 13461,
+      "▁Israel": 13462,
+      "两次": 13463,
+      "第五百": 13464,
+      "▁Jud": 13465,
+      "这个问题的答案是": 13466,
+      "▁creates": 13467,
+      "▁seeking": 13468,
+      "预计": 13469,
+      "▁gap": 13470,
+      "画面": 13471,
+      "▁tickets": 13472,
+      "inv": 13473,
+      "▁mutations": 13474,
+      "URE": 13475,
+      "▁inn": 13476,
+      "▁基本": 13477,
+      "▁Die": 13478,
+      "Async": 13479,
+      "王某": 13480,
+      "▁元": 13481,
+      "▁Common": 13482,
+      "rection": 13483,
+      "住宅": 13484,
+      "▁papers": 13485,
+      "porter": 13486,
+      "可知": 13487,
+      "week": 13488,
+      "▁manip": 13489,
+      "pand": 13490,
+      "oplas": 13491,
+      "步行": 13492,
+      "▁desc": 13493,
+      "PV": 13494,
+      "got": 13495,
+      "▁totally": 13496,
+      "名为": 13497,
+      "减刑建议": 13498,
+      "人群": 13499,
+      "实际上": 13500,
+      "odo": 13501,
+      "▁induction": 13502,
+      "WN": 13503,
+      "returns": 13504,
+      "Sm": 13505,
+      "风景": 13506,
+      "BM": 13507,
+      "nsel": 13508,
+      "migrations": 13509,
+      "收集": 13510,
+      "▁absolutely": 13511,
+      "▁anywhere": 13512,
+      "usep": 13513,
+      "的热": 13514,
+      "▁Graph": 13515,
+      "▁mand": 13516,
+      "wait": 13517,
+      "▁Sem": 13518,
+      "Num": 13519,
+      "功率": 13520,
+      "iginally": 13521,
+      "▁Jones": 13522,
+      "agent": 13523,
+      "▁injuries": 13524,
+      "▁Index": 13525,
+      "rent": 13526,
+      "▁newsp": 13527,
+      "▁permission": 13528,
+      "▁referred": 13529,
+      "的起诉": 13530,
+      "研究所": 13531,
+      "飞机": 13532,
+      "不需要": 13533,
+      "▁Mic": 13534,
+      "▁websites": 13535,
+      "GS": 13536,
+      "简单的": 13537,
+      "▁Bang": 13538,
+      "Ke": 13539,
+      "地的": 13540,
+      "▁Studies": 13541,
+      "▁habit": 13542,
+      "迟延履行期间的债务利息": 13543,
+      "展现": 13544,
+      "ework": 13545,
+      "ARY": 13546,
+      "▁yellow": 13547,
+      "▁Southern": 13548,
+      "▁battle": 13549,
+      "▁performing": 13550,
+      "▁viral": 13551,
+      "▁acquired": 13552,
+      "对罪犯": 13553,
+      "▁Week": 13554,
+      "▁compute": 13555,
+      "ortunately": 13556,
+      "▁exit": 13557,
+      "标志": 13558,
+      "▁в": 13559,
+      "ASE": 13560,
+      "孩子的": 13561,
+      "▁crucial": 13562,
+      "▁therap": 13563,
+      "database": 13564,
+      "localhost": 13565,
+      "storage": 13566,
+      "哈尔": 13567,
+      "hpp": 13568,
+      "onto": 13569,
+      "减去有期徒刑": 13570,
+      "▁sqrt": 13571,
+      "▁bridge": 13572,
+      "小说网": 13573,
+      "flex": 13574,
+      "告诉我": 13575,
+      "▁singer": 13576,
+      "▁actress": 13577,
+      "条文": 13578,
+      "Nav": 13579,
+      "小于": 13580,
+      "尽管": 13581,
+      "charge": 13582,
+      "▁assets": 13583,
+      "清单": 13584,
+      "▁Download": 13585,
+      "电器": 13586,
+      "▁Live": 13587,
+      "secret": 13588,
+      "中有": 13589,
+      "平安": 13590,
+      "TYPE": 13591,
+      "流动": 13592,
+      "gb": 13593,
+      "none": 13594,
+      "lie": 13595,
+      "▁Full": 13596,
+      "latest": 13597,
+      "▁Report": 13598,
+      "你是": 13599,
+      "到达": 13600,
+      "今日": 13601,
+      "创意": 13602,
+      "know": 13603,
+      "▁recruit": 13604,
+      "ocated": 13605,
+      "▁encounter": 13606,
+      "云南": 13607,
+      "军事": 13608,
+      "▁modules": 13609,
+      "▁gran": 13610,
+      "十七条": 13611,
+      "▁bigger": 13612,
+      "▁studying": 13613,
+      "▁lowest": 13614,
+      "不出": 13615,
+      "考察": 13616,
+      "▁Task": 13617,
+      "Host": 13618,
+      "▁Awards": 13619,
+      "usepackage": 13620,
+      "future": 13621,
+      "个体": 13622,
+      "▁Coast": 13623,
+      "▁bench": 13624,
+      "▁visible": 13625,
+      "▁Ox": 13626,
+      "pkg": 13627,
+      "地下": 13628,
+      "身高": 13629,
+      "▁;\\": 13630,
+      "▁Gar": 13631,
+      "▁introduce": 13632,
+      "工人": 13633,
+      "~~~~": 13634,
+      "专利": 13635,
+      "损害": 13636,
+      "east": 13637,
+      "▁thermal": 13638,
+      "岁的": 13639,
+      "bitr": 13640,
+      "▁sugar": 13641,
+      "▁infections": 13642,
+      "而已": 13643,
+      "Long": 13644,
+      "西方": 13645,
+      "▁Response": 13646,
+      "▁login": 13647,
+      "▁codes": 13648,
+      "是以": 13649,
+      "济南": 13650,
+      "microsoft": 13651,
+      "分局": 13652,
+      "rick": 13653,
+      "▁Force": 13654,
+      "几天": 13655,
+      "▁Linux": 13656,
+      "▁magazine": 13657,
+      "作家": 13658,
+      "ori": 13659,
+      "▁bon": 13660,
+      "的行为": 13661,
+      "实习": 13662,
+      "任意": 13663,
+      "▁Po": 13664,
+      "长沙": 13665,
+      "lay": 13666,
+      "市民": 13667,
+      "▁Source": 13668,
+      "警察": 13669,
+      "▁abuse": 13670,
+      "告知": 13671,
+      "▁mob": 13672,
+      "占地": 13673,
+      "BR": 13674,
+      "▁Fund": 13675,
+      "▁•": 13676,
+      "▁wal": 13677,
+      "properties": 13678,
+      "所得": 13679,
+      "efully": 13680,
+      "清洗": 13681,
+      "▁Station": 13682,
+      "▁suggesting": 13683,
+      "ilib": 13684,
+      "▁stroke": 13685,
+      "变量": 13686,
+      "asi": 13687,
+      "ecraft": 13688,
+      "Cs": 13689,
+      "▁superior": 13690,
+      "ando": 13691,
+      "▁appreciate": 13692,
+      "有时": 13693,
+      "▁healthcare": 13694,
+      "脂肪": 13695,
+      "低于": 13696,
+      "nJust": 13697,
+      "▁theoretical": 13698,
+      "路上": 13699,
+      "▁PART": 13700,
+      "石油": 13701,
+      "爱好": 13702,
+      "日报": 13703,
+      "LY": 13704,
+      "hol": 13705,
+      "vely": 13706,
+      "▁critic": 13707,
+      "▁winner": 13708,
+      "广州市": 13709,
+      "nCol": 13710,
+      "▁Creat": 13711,
+      "长方": 13712,
+      "QU": 13713,
+      "▁consumers": 13714,
+      "设定": 13715,
+      "边的": 13716,
+      "每个人": 13717,
+      "▁singles": 13718,
+      "nCON": 13719,
+      "▁aircraft": 13720,
+      "是最": 13721,
+      "▁roots": 13722,
+      "MENT": 13723,
+      "▁broadcast": 13724,
+      "teen": 13725,
+      "een": 13726,
+      "▁anonymous": 13727,
+      "▁oste": 13728,
+      "tons": 13729,
+      "▁ju": 13730,
+      "Math": 13731,
+      "kt": 13732,
+      "▁Member": 13733,
+      "▁compens": 13734,
+      "▁clust": 13735,
+      "行政处罚": 13736,
+      "▁Mike": 13737,
+      "▁Http": 13738,
+      "}`": 13739,
+      "办事": 13740,
+      "实验室": 13741,
+      "▁entertain": 13742,
+      "▁tracks": 13743,
+      "Deb": 13744,
+      "nan": 13745,
+      "美好": 13746,
+      "▁Pennsylvania": 13747,
+      "Tests": 13748,
+      "将其": 13749,
+      "▁southern": 13750,
+      "效益": 13751,
+      "nCare": 13752,
+      "一代": 13753,
+      "▁hack": 13754,
+      "▁govern": 13755,
+      "想象": 13756,
+      "unning": 13757,
+      "repository": 13758,
+      "公寓": 13759,
+      "副书记": 13760,
+      "大队": 13761,
+      "▁rul": 13762,
+      "家族": 13763,
+      "▁Chris": 13764,
+      "▁buffer": 13765,
+      "打印": 13766,
+      "urers": 13767,
+      "utable": 13768,
+      "适宜": 13769,
+      "Vector": 13770,
+      "La": 13771,
+      "▁vectors": 13772,
+      "callback": 13773,
+      "▁Appro": 13774,
+      "URPOSE": 13775,
+      "罚金": 13776,
+      "redu": 13777,
+      "WARE": 13778,
+      "▁migr": 13779,
+      "▁Jersey": 13780,
+      "编号": 13781,
+      "▁filename": 13782,
+      "price": 13783,
+      "\\'\\'\\'\\": 13784,
+      "onom": 13785,
+      "rium": 13786,
+      "uts": 13787,
+      "▁forced": 13788,
+      "▁reader": 13789,
+      "在执行过程中": 13790,
+      "iverse": 13791,
+      "▁figures": 13792,
+      "▁Hard": 13793,
+      "▁Request": 13794,
+      "ZE": 13795,
+      "▁movies": 13796,
+      "omp": 13797,
+      "reams": 13798,
+      "▁cream": 13799,
+      "override": 13800,
+      "他说": 13801,
+      "▁prevalence": 13802,
+      "▁faith": 13803,
+      "课题": 13804,
+      "▁adds": 13805,
+      "▁Rh": 13806,
+      "anna": 13807,
+      "模块": 13808,
+      "布局": 13809,
+      "体重": 13810,
+      "新技术": 13811,
+      "▁exceed": 13812,
+      "女子": 13813,
+      "mega": 13814,
+      "tk": 13815,
+      "rons": 13816,
+      "▁sister": 13817,
+      "防控": 13818,
+      "你可以": 13819,
+      "▁Cost": 13820,
+      "resources": 13821,
+      "▁contemporary": 13822,
+      "很快": 13823,
+      "鸡蛋": 13824,
+      "▁装": 13825,
+      "▁Lord": 13826,
+      "留下": 13827,
+      "▁kinds": 13828,
+      "ushed": 13829,
+      "▁modeling": 13830,
+      "很好的": 13831,
+      "外国": 13832,
+      "▁Spain": 13833,
+      "▁principal": 13834,
+      "▁Bow": 13835,
+      "▁FC": 13836,
+      "▁paint": 13837,
+      "页面": 13838,
+      "fd": 13839,
+      "riate": 13840,
+      "真的是": 13841,
+      "repre": 13842,
+      "本院受理后": 13843,
+      "造型": 13844,
+      "监控": 13845,
+      "builder": 13846,
+      "▁qualified": 13847,
+      "}\"": 13848,
+      "▁illness": 13849,
+      "bat": 13850,
+      "▁sav": 13851,
+      "体制": 13852,
+      "▁houses": 13853,
+      "▁Jesus": 13854,
+      "Last": 13855,
+      "▁wheel": 13856,
+      "基本上": 13857,
+      "▁county": 13858,
+      "dim": 13859,
+      "oned": 13860,
+      "hyth": 13861,
+      "▁encourage": 13862,
+      "工厂": 13863,
+      "的主": 13864,
+      "▁Ell": 13865,
+      "▁Integr": 13866,
+      "名下": 13867,
+      "rev": 13868,
+      "的应用": 13869,
+      "ategy": 13870,
+      "▁fell": 13871,
+      "▁succeed": 13872,
+      "知乎": 13873,
+      "当中": 13874,
+      "Java": 13875,
+      "应予准许": 13876,
+      "电动": 13877,
+      "assertEquals": 13878,
+      "剥夺": 13879,
+      "ovie": 13880,
+      "▁ticket": 13881,
+      "顺序": 13882,
+      "天天": 13883,
+      "▁featuring": 13884,
+      "第五百一十九": 13885,
+      "commun": 13886,
+      "▁Methods": 13887,
+      "▁Phot": 13888,
+      "nFalse": 13889,
+      "pip": 13890,
+      ">>>": 13891,
+      "▁Take": 13892,
+      "▁patch": 13893,
+      "将会": 13894,
+      "nwith": 13895,
+      "▁correlated": 13896,
+      "nset": 13897,
+      "▁unable": 13898,
+      "ployment": 13899,
+      "▁employee": 13900,
+      "▁holding": 13901,
+      "});": 13902,
+      "▁orth": 13903,
+      "计划生育": 13904,
+      "Gr": 13905,
+      "▁bron": 13906,
+      "Height": 13907,
+      "program": 13908,
+      "通讯": 13909,
+      "amber": 13910,
+      "▁concepts": 13911,
+      "▁Where": 13912,
+      "▁teen": 13913,
+      "nEnglish": 13914,
+      "azure": 13915,
+      "nWhy": 13916,
+      "▁Hom": 13917,
+      "▁ride": 13918,
+      "▁massive": 13919,
+      "药品": 13920,
+      "改进": 13921,
+      "CG": 13922,
+      "ologist": 13923,
+      "▁insight": 13924,
+      "▁vulner": 13925,
+      "无线": 13926,
+      "▁Integer": 13927,
+      "noon": 13928,
+      "▁opposite": 13929,
+      "指南": 13930,
+      "分数": 13931,
+      "fun": 13932,
+      "题主": 13933,
+      "▁dataset": 13934,
+      "oprote": 13935,
+      "ishop": 13936,
+      "▁earned": 13937,
+      "漂亮": 13938,
+      "▁substantial": 13939,
+      "▁α": 13940,
+      "uset": 13941,
+      "','": 13942,
+      "nFind": 13943,
+      "▁officers": 13944,
+      "▁situations": 13945,
+      "园区": 13946,
+      "我要": 13947,
+      "▁kick": 13948,
+      "▁mine": 13949,
+      "▁Manager": 13950,
+      "评为": 13951,
+      "lan": 13952,
+      "loop": 13953,
+      "mann": 13954,
+      "▁scientists": 13955,
+      "▁largely": 13956,
+      "▁consec": 13957,
+      "Parameter": 13958,
+      "reprene": 13959,
+      "stat": 13960,
+      "分离": 13961,
+      "days": 13962,
+      "▁passing": 13963,
+      "▁challenging": 13964,
+      "▁suppress": 13965,
+      "clusions": 13966,
+      "景观": 13967,
+      "对应": 13968,
+      "防治": 13969,
+      "sf": 13970,
+      "适量": 13971,
+      "topic": 13972,
+      "障碍": 13973,
+      "▁lect": 13974,
+      "▁PA": 13975,
+      "▁Prop": 13976,
+      "▁shock": 13977,
+      "▁nearby": 13978,
+      "▁composed": 13979,
+      "▁temp": 13980,
+      "▁FROM": 13981,
+      "▁females": 13982,
+      "▁rapidly": 13983,
+      "也不是": 13984,
+      "▁intensity": 13985,
+      "▁Senate": 13986,
+      "izable": 13987,
+      "▁invent": 13988,
+      "▁Standard": 13989,
+      "对此": 13990,
+      "ummy": 13991,
+      "▁Hor": 13992,
+      "车程约": 13993,
+      "随后": 13994,
+      "三月": 13995,
+      "付款": 13996,
+      "▁AP": 13997,
+      "▁Opt": 13998,
+      "好看": 13999,
+      "▁drama": 14000,
+      "nYes": 14001,
+      "一级": 14002,
+      "▁normally": 14003,
+      "yled": 14004,
+      "▁PO": 14005,
+      "▁guest": 14006,
+      "odium": 14007,
+      "ULAR": 14008,
+      "秘书": 14009,
+      "途径": 14010,
+      "不见": 14011,
+      "海洋": 14012,
+      "致力": 14013,
+      "lass": 14014,
+      "ogether": 14015,
+      "第一节": 14016,
+      "▁с": 14017,
+      "男子": 14018,
+      "▁Wis": 14019,
+      "cn": 14020,
+      "diff": 14021,
+      "▁environments": 14022,
+      "▁resolve": 14023,
+      "itage": 14024,
+      "意外": 14025,
+      "加倍支付": 14026,
+      "推理": 14027,
+      "ownt": 14028,
+      "Book": 14029,
+      "硬件": 14030,
+      "▁(\\'": 14031,
+      "awn": 14032,
+      "▁portion": 14033,
+      "▁Americans": 14034,
+      "El": 14035,
+      "场所": 14036,
+      "餐饮": 14037,
+      "▁Hotel": 14038,
+      "▁acting": 14039,
+      "▁prelim": 14040,
+      "书法": 14041,
+      "逐步": 14042,
+      "三大": 14043,
+      "李某": 14044,
+      "看来": 14045,
+      "如不服本裁定": 14046,
+      "American": 14047,
+      "▁Dub": 14048,
+      "▁iPhone": 14049,
+      "▁coordinates": 14050,
+      "acr": 14051,
+      "▁issued": 14052,
+      "REE": 14053,
+      "养殖": 14054,
+      "▁attacks": 14055,
+      "关心": 14056,
+      "▁prison": 14057,
+      "▁wa": 14058,
+      "▁Jac": 14059,
+      "isa": 14060,
+      "▁MIT": 14061,
+      "群体": 14062,
+      "team": 14063,
+      "小编": 14064,
+      "▁saved": 14065,
+      "▁begins": 14066,
+      "mond": 14067,
+      "从来": 14068,
+      "概率": 14069,
+      "▁northern": 14070,
+      "关节": 14071,
+      "玉米": 14072,
+      "接近": 14073,
+      "个数": 14074,
+      "▁causing": 14075,
+      "▁{\\'": 14076,
+      "▁understood": 14077,
+      "第二百零": 14078,
+      "yes": 14079,
+      "修复": 14080,
+      "layer": 14081,
+      "预算": 14082,
+      "Please": 14083,
+      "▁entr": 14084,
+      "▁Illinois": 14085,
+      "相关的": 14086,
+      "▁flowers": 14087,
+      "个性": 14088,
+      "▁san": 14089,
+      "那样": 14090,
+      "▁targeted": 14091,
+      "也不会": 14092,
+      "▁${\\\\": 14093,
+      "Pay": 14094,
+      "andy": 14095,
+      "▁monthly": 14096,
+      "USER": 14097,
+      "▁personnel": 14098,
+      "▁availability": 14099,
+      "第二节": 14100,
+      "elly": 14101,
+      "阳县": 14102,
+      "材质": 14103,
+      "东南": 14104,
+      "水果": 14105,
+      "▁Corporation": 14106,
+      "▁producer": 14107,
+      "▁narrow": 14108,
+      "azy": 14109,
+      "▁estimates": 14110,
+      "▁Contact": 14111,
+      "以上的": 14112,
+      "cert": 14113,
+      "一道": 14114,
+      "▁ble": 14115,
+      "▁references": 14116,
+      "还没": 14117,
+      "▁eating": 14118,
+      "医师": 14119,
+      "lights": 14120,
+      "可爱": 14121,
+      "▁significance": 14122,
+      "▁Physics": 14123,
+      "Writer": 14124,
+      "asm": 14125,
+      "▁grown": 14126,
+      "VERSION": 14127,
+      "辽宁省": 14128,
+      "竟然": 14129,
+      "▁strings": 14130,
+      "etc": 14131,
+      "其次": 14132,
+      "▁honest": 14133,
+      "spect": 14134,
+      "UID": 14135,
+      "▁distingu": 14136,
+      "公益": 14137,
+      "nIm": 14138,
+      "limit": 14139,
+      "Doc": 14140,
+      "▁username": 14141,
+      "▁pH": 14142,
+      "必然": 14143,
+      "的申请": 14144,
+      "nAd": 14145,
+      "▁operators": 14146,
+      ")\"": 14147,
+      "icients": 14148,
+      "/\\\\": 14149,
+      "交换": 14150,
+      "otion": 14151,
+      "Properties": 14152,
+      "▁IO": 14153,
+      "陕西": 14154,
+      "▁convenient": 14155,
+      "▁responsibility": 14156,
+      "为此": 14157,
+      "▁foundation": 14158,
+      "▁Children": 14159,
+      "esting": 14160,
+      "amic": 14161,
+      "▁revers": 14162,
+      "▁summary": 14163,
+      "保健": 14164,
+      "▁spl": 14165,
+      "Category": 14166,
+      "▁Korea": 14167,
+      "▁Wales": 14168,
+      "ificial": 14169,
+      "▁MP": 14170,
+      "▁###": 14171,
+      "▁integers": 14172,
+      "UTH": 14173,
+      "emy": 14174,
+      "alian": 14175,
+      "超市": 14176,
+      "▁antigen": 14177,
+      "▁Radio": 14178,
+      "▁privacy": 14179,
+      "▁handling": 14180,
+      "▁stimulation": 14181,
+      "▁peace": 14182,
+      "▁Georgia": 14183,
+      "把握": 14184,
+      "PORT": 14185,
+      "▁arrang": 14186,
+      "氛围": 14187,
+      "nfortunately": 14188,
+      "▁platforms": 14189,
+      "呼吸": 14190,
+      "NC": 14191,
+      "▁ours": 14192,
+      "▁Jo": 14193,
+      "▁Kh": 14194,
+      "▁quantitative": 14195,
+      "▁Maybe": 14196,
+      "的中": 14197,
+      "▁occasion": 14198,
+      "HL": 14199,
+      "▁concerned": 14200,
+      "▁tracking": 14201,
+      "国务院": 14202,
+      "释放": 14203,
+      "\"#": 14204,
+      "▁reactions": 14205,
+      "CharField": 14206,
+      "HC": 14207,
+      "nQu": 14208,
+      "▁polymer": 14209,
+      "▁risks": 14210,
+      "很少": 14211,
+      "张某": 14212,
+      "bash": 14213,
+      "fetch": 14214,
+      "你会": 14215,
+      "姐姐": 14216,
+      "▁parking": 14217,
+      "小米": 14218,
+      "▁Micro": 14219,
+      "eting": 14220,
+      "resolve": 14221,
+      "▁Store": 14222,
+      "verter": 14223,
+      "▁reviewed": 14224,
+      "▁Saint": 14225,
+      "▁guarantee": 14226,
+      "取值": 14227,
+      "iest": 14228,
+      "▁implications": 14229,
+      "的能力": 14230,
+      "系数": 14231,
+      "结果表明": 14232,
+      "从轻处罚": 14233,
+      "▁purchased": 14234,
+      "ERROR": 14235,
+      "erly": 14236,
+      "▁busy": 14237,
+      "▁hence": 14238,
+      "▁Francisco": 14239,
+      "▁assembly": 14240,
+      "大连": 14241,
+      "▁总部": 14242,
+      "不错的": 14243,
+      "▁/>": 14244,
+      "▁programme": 14245,
+      "携带": 14246,
+      "exe": 14247,
+      "详情": 14248,
+      "large": 14249,
+      "部位": 14250,
+      "▁Who": 14251,
+      "comput": 14252,
+      "imately": 14253,
+      "▁compound": 14254,
+      "▁Mah": 14255,
+      "▁Saf": 14256,
+      "▁Ul": 14257,
+      "tint": 14258,
+      "NESS": 14259,
+      "inals": 14260,
+      "hemistry": 14261,
+      "ieves": 14262,
+      "可见": 14263,
+      "也在": 14264,
+      "都没": 14265,
+      "▁Unit": 14266,
+      "▁callback": 14267,
+      "▁clubs": 14268,
+      "真正的": 14269,
+      "EFAULT": 14270,
+      "▁respir": 14271,
+      "▁flexible": 14272,
+      "加速": 14273,
+      "带来的": 14274,
+      "舞台": 14275,
+      "和解": 14276,
+      "▁interests": 14277,
+      "▁destination": 14278,
+      "累计": 14279,
+      "erior": 14280,
+      "/>\\": 14281,
+      "▁proved": 14282,
+      "actic": 14283,
+      "▁bathroom": 14284,
+      "sex": 14285,
+      "承诺": 14286,
+      "▁Space": 14287,
+      "五月": 14288,
+      "▁react": 14289,
+      "劳务": 14290,
+      "播放": 14291,
+      "节能": 14292,
+      "▁citizens": 14293,
+      "▁pitch": 14294,
+      "ista": 14295,
+      "▁managers": 14296,
+      "游客": 14297,
+      "尊重": 14298,
+      "▁entirely": 14299,
+      "▁tip": 14300,
+      "▁dates": 14301,
+      "▁applying": 14302,
+      "可在判决书送达之日起十五日内": 14303,
+      "危机": 14304,
+      "▁commitment": 14305,
+      "▁machines": 14306,
+      "地铁": 14307,
+      "▁entries": 14308,
+      "▁surfaces": 14309,
+      "展览": 14310,
+      "▁Video": 14311,
+      "ipheral": 14312,
+      "▁\"\"\\": 14313,
+      "▁Kent": 14314,
+      "nas": 14315,
+      "▁Johnson": 14316,
+      "▁Illuminate": 14317,
+      "十二月": 14318,
+      "▁Pacific": 14319,
+      "为何": 14320,
+      "▁周": 14321,
+      "间的": 14322,
+      "shared": 14323,
+      "接受教育": 14324,
+      "加大": 14325,
+      "▁violence": 14326,
+      "▁mountain": 14327,
+      "▁Democratic": 14328,
+      "namespace": 14329,
+      "▁domestic": 14330,
+      "小数": 14331,
+      "ochond": 14332,
+      "十八条": 14333,
+      "▁coc": 14334,
+      "第二百五十三": 14335,
+      "▁dental": 14336,
+      "▁Low": 14337,
+      "视觉": 14338,
+      "junit": 14339,
+      "▁Problem": 14340,
+      "▁enabled": 14341,
+      "近期": 14342,
+      "nTe": 14343,
+      "国的": 14344,
+      "▁eg": 14345,
+      "iner": 14346,
+      "抚养费": 14347,
+      "\")]\\": 14348,
+      "站在": 14349,
+      "Tube": 14350,
+      "initial": 14351,
+      "ometimes": 14352,
+      "▁fle": 14353,
+      "战士": 14354,
+      "ju": 14355,
+      "ellite": 14356,
+      "总统": 14357,
+      "▁surve": 14358,
+      "的本次执行程序": 14359,
+      "▁assuming": 14360,
+      "常规": 14361,
+      "bd": 14362,
+      "assertEqual": 14363,
+      "III": 14364,
+      "▁bacteria": 14365,
+      "▁ABC": 14366,
+      "▁weap": 14367,
+      "bian": 14368,
+      "法律法规": 14369,
+      "▁antibody": 14370,
+      "Fi": 14371,
+      "Mock": 14372,
+      "外观": 14373,
+      "电压": 14374,
+      "养老": 14375,
+      "istered": 14376,
+      "设立": 14377,
+      "▁Ide": 14378,
+      "补偿": 14379,
+      "▁pert": 14380,
+      "得到了": 14381,
+      "mir": 14382,
+      "Ne": 14383,
+      "▁inflammatory": 14384,
+      "vc": 14385,
+      "▁Mathematics": 14386,
+      "isition": 14387,
+      "plugins": 14388,
+      "婚后": 14389,
+      "▁museum": 14390,
+      "混凝": 14391,
+      "明的": 14392,
+      "不然": 14393,
+      "消失": 14394,
+      "▁Far": 14395,
+      "bury": 14396,
+      "ibli": 14397,
+      "▁verify": 14398,
+      "SULTS": 14399,
+      "VM": 14400,
+      "▁SU": 14401,
+      "▁Little": 14402,
+      "▁export": 14403,
+      "xiety": 14404,
+      "ido": 14405,
+      "联赛": 14406,
+      "black": 14407,
+      "长春": 14408,
+      "▁sheet": 14409,
+      "▁Level": 14410,
+      "▁factory": 14411,
+      "▁terminal": 14412,
+      "申报": 14413,
+      "▁awards": 14414,
+      "▁About": 14415,
+      "疗效": 14416,
+      "优点": 14417,
+      "▁registration": 14418,
+      "icted": 14419,
+      "▁总部地点": 14420,
+      "Our": 14421,
+      "广播": 14422,
+      "iox": 14423,
+      "clean": 14424,
+      "术后": 14425,
+      "SF": 14426,
+      "▁Parliament": 14427,
+      "▁inflammation": 14428,
+      "▁我": 14429,
+      "rowse": 14430,
+      "▁Irish": 14431,
+      "acent": 14432,
+      "本院审理": 14433,
+      "▁concrete": 14434,
+      "inton": 14435,
+      "DateTime": 14436,
+      "▁Studio": 14437,
+      "▁reveal": 14438,
+      "review": 14439,
+      "▁Sports": 14440,
+      "▁NC": 14441,
+      "▁administrative": 14442,
+      "八日": 14443,
+      "▁ages": 14444,
+      "▁immediate": 14445,
+      "扩展": 14446,
+      "▁railway": 14447,
+      "▁Example": 14448,
+      "▁injection": 14449,
+      "inson": 14450,
+      "material": 14451,
+      "▁warranty": 14452,
+      "面前": 14453,
+      "圆的": 14454,
+      "▁▁▁▁▁▁▁": 14455,
+      "▁Ig": 14456,
+      "▁hospit": 14457,
+      "账号": 14458,
+      "rices": 14459,
+      "和社会": 14460,
+      "▁EU": 14461,
+      "江西": 14462,
+      "oggle": 14463,
+      "▁Coll": 14464,
+      "vo": 14465,
+      "▁preliminary": 14466,
+      "▁Account": 14467,
+      "承包": 14468,
+      "▁fantastic": 14469,
+      "native": 14470,
+      "▁SP": 14471,
+      "illy": 14472,
+      "▁Represent": 14473,
+      "aus": 14474,
+      "▁loading": 14475,
+      "那里": 14476,
+      "Head": 14477,
+      "▁defense": 14478,
+      "应对": 14479,
+      "nCareer": 14480,
+      "▁accessible": 14481,
+      "拘留": 14482,
+      "的第二": 14483,
+      "市委": 14484,
+      "Items": 14485,
+      "工程师": 14486,
+      "▁leuk": 14487,
+      "期限内": 14488,
+      "休息": 14489,
+      "▁bringing": 14490,
+      "内心": 14491,
+      "loader": 14492,
+      "八月": 14493,
+      "指挥": 14494,
+      "▁ster": 14495,
+      "▁emotional": 14496,
+      "▁enjoyed": 14497,
+      "ih": 14498,
+      "▁Promise": 14499,
+      "▁UI": 14500,
+      "指控被告人": 14501,
+      "▁baseline": 14502,
+      "图形": 14503,
+      "CLUS": 14504,
+      "▁Middle": 14505,
+      "nthat": 14506,
+      "▁characteristic": 14507,
+      "forward": 14508,
+      "▁breath": 14509,
+      "仪式": 14510,
+      "geqslant": 14511,
+      "▁pulmonary": 14512,
+      "提出上诉": 14513,
+      "ов": 14514,
+      "货款": 14515,
+      "孩子们": 14516,
+      "麻烦": 14517,
+      "▁census": 14518,
+      "气候": 14519,
+      "ocolate": 14520,
+      "商标": 14521,
+      "▁professor": 14522,
+      "gar": 14523,
+      "▁producing": 14524,
+      "▁um": 14525,
+      "tar": 14526,
+      "▁pup": 14527,
+      "著名": 14528,
+      "\\'],": 14529,
+      "▁recipe": 14530,
+      "远程": 14531,
+      "客观": 14532,
+      "aze": 14533,
+      "▁contributions": 14534,
+      "▁fault": 14535,
+      "▁amino": 14536,
+      "▁validate": 14537,
+      "irth": 14538,
+      "▁affordable": 14539,
+      "▁Labor": 14540,
+      "▁é": 14541,
+      "海外": 14542,
+      "iser": 14543,
+      "▁cohort": 14544,
+      "垂直": 14545,
+      "合肥": 14546,
+      "ifer": 14547,
+      "武器": 14548,
+      "▁metabolic": 14549,
+      "/)": 14550,
+      "submit": 14551,
+      "健身": 14552,
+      "近代": 14553,
+      "▁markers": 14554,
+      "subject": 14555,
+      "学的": 14556,
+      "该村": 14557,
+      "asters": 14558,
+      "年底": 14559,
+      "如实": 14560,
+      "校长": 14561,
+      "▁LI": 14562,
+      "▁younger": 14563,
+      "▁feas": 14564,
+      "▁interventions": 14565,
+      "anol": 14566,
+      "Queue": 14567,
+      "▁acids": 14568,
+      "▁agricult": 14569,
+      "于本判决": 14570,
+      "原审被告": 14571,
+      "LP": 14572,
+      "bral": 14573,
+      "chat": 14574,
+      "二审": 14575,
+      "▁赵": 14576,
+      "▁layers": 14577,
+      "▁awareness": 14578,
+      "▁calcium": 14579,
+      "\"}": 14580,
+      "▁attribute": 14581,
+      "维生素": 14582,
+      "产量": 14583,
+      "▁consequences": 14584,
+      "Ac": 14585,
+      "▁veter": 14586,
+      "▁migration": 14587,
+      "_\\\\": 14588,
+      "hom": 14589,
+      "纷纷": 14590,
+      "sid": 14591,
+      "▁Load": 14592,
+      "▁brands": 14593,
+      "enders": 14594,
+      "\"<": 14595,
+      "▁crime": 14596,
+      "▁fluores": 14597,
+      "▁trends": 14598,
+      "icons": 14599,
+      "▁endot": 14600,
+      "▁Williams": 14601,
+      "▁hepat": 14602,
+      "很大的": 14603,
+      "hus": 14604,
+      "投资者": 14605,
+      "NE": 14606,
+      "▁doubt": 14607,
+      "complete": 14608,
+      "▁supplement": 14609,
+      "▁greatly": 14610,
+      "▁Try": 14611,
+      "吃饭": 14612,
+      "图象": 14613,
+      "▁Iss": 14614,
+      "后期": 14615,
+      "紧张": 14616,
+      "▁noticed": 14617,
+      "Use": 14618,
+      "而不": 14619,
+      "▁trouble": 14620,
+      "▁regularly": 14621,
+      "▁Put": 14622,
+      "▁nutr": 14623,
+      "PASSWORD": 14624,
+      "纠纷一案中": 14625,
+      "▁Ray": 14626,
+      "ologists": 14627,
+      "合同法": 14628,
+      "▁tags": 14629,
+      "演唱": 14630,
+      "看起来": 14631,
+      "出现了": 14632,
+      "▁tells": 14633,
+      "▁males": 14634,
+      "▁refr": 14635,
+      "生活的": 14636,
+      "***": 14637,
+      "市场的": 14638,
+      "解决方案": 14639,
+      "▁Kim": 14640,
+      "▁Hon": 14641,
+      "▁operated": 14642,
+      "▁insights": 14643,
+      "条款": 14644,
+      "LD": 14645,
+      "▁acts": 14646,
+      "occ": 14647,
+      "izers": 14648,
+      "isting": 14649,
+      "▁actors": 14650,
+      "Gu": 14651,
+      "esc": 14652,
+      "odec": 14653,
+      "▁mixture": 14654,
+      "nrequire": 14655,
+      "车的": 14656,
+      "%;": 14657,
+      "血压": 14658,
+      "▁Ukr": 14659,
+      "neq": 14660,
+      "▁readers": 14661,
+      "wall": 14662,
+      "集团有限公司": 14663,
+      "▁Ren": 14664,
+      "issions": 14665,
+      "unting": 14666,
+      "▁missed": 14667,
+      "je": 14668,
+      "▁indeed": 14669,
+      "fc": 14670,
+      "▁terr": 14671,
+      "温暖": 14672,
+      "▁reput": 14673,
+      "和分析": 14674,
+      "不宜": 14675,
+      "是中国": 14676,
+      "星级": 14677,
+      "asty": 14678,
+      "强烈": 14679,
+      "▁Kn": 14680,
+      "苏州": 14681,
+      "团体": 14682,
+      "oga": 14683,
+      "▁connections": 14684,
+      "文艺": 14685,
+      "▁strains": 14686,
+      "食材": 14687,
+      "estinal": 14688,
+      "▁elections": 14689,
+      "Fore": 14690,
+      "六月": 14691,
+      "界面": 14692,
+      "Delete": 14693,
+      "zero": 14694,
+      "▁ms": 14695,
+      "▁visiting": 14696,
+      "bet": 14697,
+      "▁cutting": 14698,
+      "▁labels": 14699,
+      "命运": 14700,
+      "▁Whether": 14701,
+      "▁fro": 14702,
+      "otypes": 14703,
+      "▁FIT": 14704,
+      "cls": 14705,
+      "混凝土": 14706,
+      "▁Expl": 14707,
+      "izz": 14708,
+      "新型": 14709,
+      "不到庭": 14710,
+      "有时候": 14711,
+      "▁discrim": 14712,
+      "catch": 14713,
+      "After": 14714,
+      "▁facing": 14715,
+      "STR": 14716,
+      "或是": 14717,
+      "Files": 14718,
+      "▁committee": 14719,
+      "Python": 14720,
+      "▁collections": 14721,
+      "▁COVID": 14722,
+      "▁diversity": 14723,
+      "▁Es": 14724,
+      "山西": 14725,
+      "▁boys": 14726,
+      "接受教育改造": 14727,
+      "▁(\\\\": 14728,
+      "意味着": 14729,
+      "危险驾驶": 14730,
+      "▁TODO": 14731,
+      "reference": 14732,
+      "加倍支付迟延履行期间的债务利息": 14733,
+      "everal": 14734,
+      "ulture": 14735,
+      "events": 14736,
+      "Callback": 14737,
+      "oked": 14738,
+      "encode": 14739,
+      "是一道": 14740,
+      "▁pilot": 14741,
+      "▁Cur": 14742,
+      "▁proven": 14743,
+      "▁Cloud": 14744,
+      "▁req": 14745,
+      "\\'))\\": 14746,
+      "同步": 14747,
+      "诉讼费": 14748,
+      "\"));\\": 14749,
+      "上涨": 14750,
+      "▁developers": 14751,
+      "Non": 14752,
+      "ateway": 14753,
+      "repo": 14754,
+      "▁centers": 14755,
+      "Root": 14756,
+      "较为": 14757,
+      "法律条文": 14758,
+      "udden": 14759,
+      "支部": 14760,
+      "▁Vill": 14761,
+      "contact": 14762,
+      "走了": 14763,
+      "都很": 14764,
+      "线路": 14765,
+      "世界上": 14766,
+      "▁dial": 14767,
+      "aki": 14768,
+      "每月": 14769,
+      "▁politicians": 14770,
+      "▁freedom": 14771,
+      "opes": 14772,
+      "厨房": 14773,
+      "▁Sea": 14774,
+      "[[": 14775,
+      "刑事诉讼法": 14776,
+      "大师": 14777,
+      "▁painting": 14778,
+      "使用的": 14779,
+      "▁solving": 14780,
+      "▁moth": 14781,
+      "berry": 14782,
+      "▁routine": 14783,
+      "▁rid": 14784,
+      "电路": 14785,
+      "▁Learning": 14786,
+      "▁upcoming": 14787,
+      "▁Among": 14788,
+      "▁apopt": 14789,
+      "▁walls": 14790,
+      "瞬间": 14791,
+      "处的": 14792,
+      "\\'])\\": 14793,
+      "▁Ham": 14794,
+      "letter": 14795,
+      "▁mini": 14796,
+      "漫画": 14797,
+      "主角": 14798,
+      "green": 14799,
+      "ouston": 14800,
+      "▁shipping": 14801,
+      "传票": 14802,
+      "Pad": 14803,
+      "▁saving": 14804,
+      "相反": 14805,
+      "▁Auto": 14806,
+      "灵活": 14807,
+      "给了": 14808,
+      "▁Luc": 14809,
+      "▁descript": 14810,
+      "学者": 14811,
+      "Param": 14812,
+      "▁glob": 14813,
+      "▁Sym": 14814,
+      "福利": 14815,
+      "▁edges": 14816,
+      "▁realize": 14817,
+      "▁ingredients": 14818,
+      "▁teeth": 14819,
+      "月至": 14820,
+      "▁panc": 14821,
+      "about": 14822,
+      "▁Patients": 14823,
+      "物的": 14824,
+      "▁bot": 14825,
+      "优先": 14826,
+      "▁vocals": 14827,
+      "▁nerve": 14828,
+      "details": 14829,
+      "入住": 14830,
+      "forcement": 14831,
+      "审计": 14832,
+      "申请撤诉": 14833,
+      "▁maintained": 14834,
+      "\\\\((\\\\)": 14835,
+      "认知": 14836,
+      "履行给付": 14837,
+      "▁Such": 14838,
+      "▁denominator": 14839,
+      "▁gal": 14840,
+      "水的": 14841,
+      "▁closer": 14842,
+      "国内外": 14843,
+      "政治权利": 14844,
+      "▁Box": 14845,
+      "▁MR": 14846,
+      "▁magnitude": 14847,
+      "电机": 14848,
+      "linux": 14849,
+      "▁Emp": 14850,
+      "▁increasingly": 14851,
+      "}>\\": 14852,
+      "▁装帧": 14853,
+      "EP": 14854,
+      "又是": 14855,
+      "▁rows": 14856,
+      "▁paths": 14857,
+      "学员": 14858,
+      "scripts": 14859,
+      "▁standing": 14860,
+      "▁ban": 14861,
+      "move": 14862,
+      "feature": 14863,
+      "▁clock": 14864,
+      "bul": 14865,
+      "更有": 14866,
+      "男性": 14867,
+      "nBack": 14868,
+      "nPh": 14869,
+      "上一": 14870,
+      "▁Federal": 14871,
+      "除以": 14872,
+      "endl": 14873,
+      "tv": 14874,
+      "选手": 14875,
+      "▁bodies": 14876,
+      "▁warning": 14877,
+      "▁believed": 14878,
+      "▁Carl": 14879,
+      "▁behavioral": 14880,
+      "(@": 14881,
+      "▁dent": 14882,
+      "分行": 14883,
+      "▁conclusion": 14884,
+      "fall": 14885,
+      "▁Places": 14886,
+      "▁superv": 14887,
+      "民警": 14888,
+      "▁Another": 14889,
+      "quality": 14890,
+      "OB": 14891,
+      "全市": 14892,
+      "tle": 14893,
+      "▁Cond": 14894,
+      "他在": 14895,
+      "▁cro": 14896,
+      "▁margin": 14897,
+      "教程": 14898,
+      "根据题意": 14899,
+      "抑制": 14900,
+      "号刑事判决": 14901,
+      "Mar": 14902,
+      "西南": 14903,
+      "paper": 14904,
+      "▁shares": 14905,
+      "▁Spe": 14906,
+      "▁Sciences": 14907,
+      "▁elig": 14908,
+      "▁characterize": 14909,
+      "▁fiber": 14910,
+      "▁Op": 14911,
+      "▁PARTIC": 14912,
+      "▁Cou": 14913,
+      "CHANT": 14914,
+      "▁prediction": 14915,
+      "▁guidance": 14916,
+      "痛苦": 14917,
+      "知识的": 14918,
+      "▁repo": 14919,
+      "▁historic": 14920,
+      "了吗": 14921,
+      "hemes": 14922,
+      "bootstrap": 14923,
+      "▁organized": 14924,
+      "▁sole": 14925,
+      "▁bedroom": 14926,
+      "▁高": 14927,
+      "百年": 14928,
+      "▁promin": 14929,
+      "upyter": 14930,
+      "articles": 14931,
+      "DD": 14932,
+      "▁mouth": 14933,
+      "usetts": 14934,
+      "idal": 14935,
+      "▁amounts": 14936,
+      "▁polym": 14937,
+      "oly": 14938,
+      "achusetts": 14939,
+      "那就": 14940,
+      "▁Henry": 14941,
+      "nDuring": 14942,
+      "▁execute": 14943,
+      "▁goods": 14944,
+      "土壤": 14945,
+      "▁facilitate": 14946,
+      "nSc": 14947,
+      "▁complicated": 14948,
+      "▁domains": 14949,
+      "TF": 14950,
+      "▁Dan": 14951,
+      "积累": 14952,
+      "oust": 14953,
+      "hour": 14954,
+      "锻炼": 14955,
+      "Range": 14956,
+      "eras": 14957,
+      "nHi": 14958,
+      "▁scan": 14959,
+      "▁publication": 14960,
+      "剥夺政治权利": 14961,
+      "▁involvement": 14962,
+      "oop": 14963,
+      "▁attributes": 14964,
+      "icture": 14965,
+      "ANCE": 14966,
+      "依赖": 14967,
+      "Pat": 14968,
+      "omething": 14969,
+      "▁Jon": 14970,
+      "▁合作作品": 14971,
+      "专项": 14972,
+      "▁lay": 14973,
+      "stein": 14974,
+      "▁Walk": 14975,
+      "▁afternoon": 14976,
+      "▁religious": 14977,
+      "UC": 14978,
+      "Mapping": 14979,
+      "ANG": 14980,
+      "还没有": 14981,
+      "Git": 14982,
+      "拒不到庭": 14983,
+      "▁arbitr": 14984,
+      "浓度": 14985,
+      "▁validation": 14986,
+      "面向": 14987,
+      "▁中": 14988,
+      "ilateral": 14989,
+      "▁mi": 14990,
+      "▁Document": 14991,
+      "公主": 14992,
+      "ERCHANT": 14993,
+      "▁underw": 14994,
+      "design": 14995,
+      "五日": 14996,
+      "▁aden": 14997,
+      "ika": 14998,
+      "▁focuses": 14999,
+      "chi": 15000,
+      "▁pit": 15001,
+      "九日": 15002,
+      "不受": 15003,
+      "从小": 15004,
+      "▁Component": 15005,
+      "iga": 15006,
+      "▁predicted": 15007,
+      "▁Auth": 15008,
+      "中华人民共和国行政": 15009,
+      "关闭": 15010,
+      "▁mathematics": 15011,
+      "人为": 15012,
+      "无异议": 15013,
+      "谢谢": 15014,
+      "▁HN": 15015,
+      "▁searching": 15016,
+      "▁signature": 15017,
+      "▁dinner": 15018,
+      "▁perfectly": 15019,
+      "▁advertising": 15020,
+      "acer": 15021,
+      "▁enables": 15022,
+      "言情": 15023,
+      "%%": 15024,
+      "nbsp": 15025,
+      "yaml": 15026,
+      "▁maxim": 15027,
+      "away": 15028,
+      "▁numerical": 15029,
+      "acker": 15030,
+      "▁小": 15031,
+      "▁Fair": 15032,
+      "▁agencies": 15033,
+      "Values": 15034,
+      "▁neck": 15035,
+      "▁PURPOSE": 15036,
+      "▁Joe": 15037,
+      "接着": 15038,
+      "农户": 15039,
+      "财产可供执行": 15040,
+      "同期": 15041,
+      "burgh": 15042,
+      "serial": 15043,
+      "consin": 15044,
+      "的朋友": 15045,
+      "对被告": 15046,
+      "support": 15047,
+      "上去": 15048,
+      "山西省": 15049,
+      "赚钱": 15050,
+      "driver": 15051,
+      "▁Andrew": 15052,
+      "▁discovery": 15053,
+      "▁priority": 15054,
+      "反正": 15055,
+      "▁profiles": 15056,
+      "▁championship": 15057,
+      "▁kil": 15058,
+      "▁constitu": 15059,
+      "rophy": 15060,
+      "▁warn": 15061,
+      "▁inhibitors": 15062,
+      "相似": 15063,
+      "▁Greek": 15064,
+      "▁proliferation": 15065,
+      "提出的": 15066,
+      "定理": 15067,
+      "theme": 15068,
+      "各自": 15069,
+      "▁marriage": 15070,
+      "hor": 15071,
+      "quir": 15072,
+      "的技术": 15073,
+      "▁stands": 15074,
+      "▁revenue": 15075,
+      "▁Digital": 15076,
+      "\\\\,": 15077,
+      "第二百六": 15078,
+      "▁exclusive": 15079,
+      "aver": 15080,
+      "outes": 15081,
+      "大大": 15082,
+      "▁degrad": 15083,
+      "flag": 15084,
+      "desc": 15085,
+      "▁Dou": 15086,
+      "▁Joseph": 15087,
+      "▁produces": 15088,
+      "▁attr": 15089,
+      "意大利": 15090,
+      "ropy": 15091,
+      "▁malign": 15092,
+      "辅导": 15093,
+      "框架": 15094,
+      "▁mathematical": 15095,
+      "万人": 15096,
+      "scanf": 15097,
+      "危险驾驶罪": 15098,
+      "enth": 15099,
+      "][\\'": 15100,
+      "ricular": 15101,
+      "▁drawn": 15102,
+      "上班": 15103,
+      "大赛": 15104,
+      "\"],": 15105,
+      "▁doors": 15106,
+      "question": 15107,
+      "若干问题": 15108,
+      "早期": 15109,
+      "▁concerning": 15110,
+      "九月": 15111,
+      "sen": 15112,
+      "切实": 15113,
+      "餐馆": 15114,
+      "▁bass": 15115,
+      "Login": 15116,
+      "worth": 15117,
+      "▁appointment": 15118,
+      "密度": 15119,
+      "▁'/": 15120,
+      "fusion": 15121,
+      "jar": 15122,
+      "}\\\\)\\": 15123,
+      "ERCHANTABILITY": 15124,
+      "mediated": 15125,
+      "plays": 15126,
+      "▁endl": 15127,
+      "depth": 15128,
+      "化妆": 15129,
+      "▁beam": 15130,
+      "ynt": 15131,
+      "Ident": 15132,
+      "▁plain": 15133,
+      "develop": 15134,
+      "一是": 15135,
+      "hard": 15136,
+      "named": 15137,
+      "FR": 15138,
+      "居然": 15139,
+      "工程有限公司": 15140,
+      "▁Bol": 15141,
+      "观念": 15142,
+      "ods": 15143,
+      "▁crisis": 15144,
+      "▁temperatures": 15145,
+      "nOther": 15146,
+      "纳入": 15147,
+      "▁dx": 15148,
+      "ulpt": 15149,
+      "stage": 15150,
+      "eness": 15151,
+      "Cor": 15152,
+      "▁Connect": 15153,
+      "на": 15154,
+      "整合": 15155,
+      "▁principles": 15156,
+      "业主": 15157,
+      "京东": 15158,
+      "有权": 15159,
+      "本质": 15160,
+      "▁transaction": 15161,
+      "serve": 15162,
+      "这两": 15163,
+      "▁wet": 15164,
+      "library": 15165,
+      "▁coefficient": 15166,
+      "中华人民共和国刑事诉讼法": 15167,
+      "\"`\\": 15168,
+      "\"))": 15169,
+      "▁dollars": 15170,
+      "FTWARE": 15171,
+      "ivari": 15172,
+      "▁negot": 15173,
+      "JS": 15174,
+      "▁rein": 15175,
+      "电视剧": 15176,
+      "欠款": 15177,
+      "▁PARTICULAR": 15178,
+      "▁operate": 15179,
+      "mys": 15180,
+      "('/": 15181,
+      "便利": 15182,
+      "irement": 15183,
+      "▁improvements": 15184,
+      "▁divide": 15185,
+      "立方": 15186,
+      "pur": 15187,
+      "▁scholar": 15188,
+      "▁Win": 15189,
+      "placement": 15190,
+      "▁redirect": 15191,
+      "▁Import": 15192,
+      "▁checked": 15193,
+      "形势": 15194,
+      "▁adoles": 15195,
+      "用途": 15196,
+      "▁drawing": 15197,
+      "▁optional": 15198,
+      "▁footballer": 15199,
+      "agues": 15200,
+      "起了": 15201,
+      "传递": 15202,
+      "nSe": 15203,
+      "▁geometry": 15204,
+      "▁feels": 15205,
+      "country": 15206,
+      "esides": 15207,
+      "▁observation": 15208,
+      "联社": 15209,
+      "小心": 15210,
+      "▁Islam": 15211,
+      "agraph": 15212,
+      "婴儿": 15213,
+      "feed": 15214,
+      "▁coefficients": 15215,
+      "▁fort": 15216,
+      "]:\\": 15217,
+      "▁DA": 15218,
+      "称号": 15219,
+      "州市中级人民法院": 15220,
+      "▁kinase": 15221,
+      "族自治": 15222,
+      "nAre": 15223,
+      "nNotes": 15224,
+      "gressive": 15225,
+      "▁campus": 15226,
+      "``": 15227,
+      "马上": 15228,
+      "▁capabilities": 15229,
+      "七日": 15230,
+      "求得": 15231,
+      "用了": 15232,
+      "▁suppl": 15233,
+      "▁Af": 15234,
+      "Role": 15235,
+      "▁chromos": 15236,
+      "▁decade": 15237,
+      "恢复执行": 15238,
+      "tb": 15239,
+      "▁大": 15240,
+      "elen": 15241,
+      "▁mock": 15242,
+      "▁\"#": 15243,
+      "aga": 15244,
+      "▁arrived": 15245,
+      "▁Taylor": 15246,
+      "Adapter": 15247,
+      "▁scatter": 15248,
+      "▁reasonable": 15249,
+      "▁boat": 15250,
+      "▁moves": 15251,
+      "▁FA": 15252,
+      "功效": 15253,
+      "▁permanent": 15254,
+      "Ver": 15255,
+      "▁Navy": 15256,
+      "▁fees": 15257,
+      "消化": 15258,
+      "如何看待": 15259,
+      "助理": 15260,
+      "elle": 15261,
+      "vs": 15262,
+      "致力于": 15263,
+      "▁completion": 15264,
+      "EST": 15265,
+      "▁tough": 15266,
+      "▁rib": 15267,
+      "express": 15268,
+      "orrow": 15269,
+      "▁Pot": 15270,
+      "▁Britain": 15271,
+      "不可能": 15272,
+      "回收": 15273,
+      "){": 15274,
+      "同学们": 15275,
+      "▁Change": 15276,
+      "iley": 15277,
+      "▁Scotland": 15278,
+      "视为": 15279,
+      "▁butter": 15280,
+      "算是": 15281,
+      "CRE": 15282,
+      "部队": 15283,
+      "▁Trust": 15284,
+      "▁landscape": 15285,
+      "定期": 15286,
+      "▁Cro": 15287,
+      "将在": 15288,
+      "yroid": 15289,
+      "▁king": 15290,
+      "NP": 15291,
+      "▁Univers": 15292,
+      "既然": 15293,
+      "▁ele": 15294,
+      "▁hall": 15295,
+      "factory": 15296,
+      "yo": 15297,
+      "ailing": 15298,
+      "▁SH": 15299,
+      "▁diagnosed": 15300,
+      "▁computing": 15301,
+      "▁infinite": 15302,
+      "庭审": 15303,
+      "ulous": 15304,
+      "Abstract": 15305,
+      "Original": 15306,
+      "大量的": 15307,
+      "上诉的": 15308,
+      "▁Fac": 15309,
+      "▁graduate": 15310,
+      "▁eleg": 15311,
+      "禁止": 15312,
+      "并处罚金人民币": 15313,
+      "力度": 15314,
+      "依法组成合议庭": 15315,
+      "▁bytes": 15316,
+      "▁Dar": 15317,
+      "timestamp": 15318,
+      "十一月": 15319,
+      "不行": 15320,
+      "▁Mean": 15321,
+      "▁dining": 15322,
+      "▁medi": 15323,
+      "▁Ont": 15324,
+      "▁kill": 15325,
+      "semble": 15326,
+      "▁Santa": 15327,
+      "赛季": 15328,
+      "方的": 15329,
+      "人都": 15330,
+      "▁intent": 15331,
+      "走向": 15332,
+      "orb": 15333,
+      "▁proportion": 15334,
+      "axy": 15335,
+      "Display": 15336,
+      "▁ultimately": 15337,
+      "▁taught": 15338,
+      "牺牲": 15339,
+      "借款本金": 15340,
+      "oured": 15341,
+      "hex": 15342,
+      "▁licensed": 15343,
+      "owa": 15344,
+      "scious": 15345,
+      "▁Should": 15346,
+      "▁participation": 15347,
+      "▁upgrade": 15348,
+      "▁residential": 15349,
+      "rovers": 15350,
+      "▁indicating": 15351,
+      "▁promising": 15352,
+      "回归": 15353,
+      "selected": 15354,
+      "CTION": 15355,
+      "ookie": 15356,
+      "▁kidney": 15357,
+      "▁Despite": 15358,
+      "rank": 15359,
+      "▁requirement": 15360,
+      "pu": 15361,
+      "}}$": 15362,
+      "\"_": 15363,
+      "Grid": 15364,
+      "▁Mir": 15365,
+      "nwhere": 15366,
+      "▁Vector": 15367,
+      "Impl": 15368,
+      "▁Medicine": 15369,
+      "依旧": 15370,
+      "▁Poland": 15371,
+      "▁recommendations": 15372,
+      "inations": 15373,
+      "的特点": 15374,
+      "▁Airport": 15375,
+      "▁artery": 15376,
+      "承认": 15377,
+      "传输": 15378,
+      "教育部": 15379,
+      "秘密": 15380,
+      "▁western": 15381,
+      "影视": 15382,
+      "引发": 15383,
+      "▁kne": 15384,
+      "▁Night": 15385,
+      "多项": 15386,
+      "▁debt": 15387,
+      "返还": 15388,
+      "随便": 15389,
+      "ald": 15390,
+      "季度": 15391,
+      "Selector": 15392,
+      "▁carcinoma": 15393,
+      "全省": 15394,
+      "ETHO": 15395,
+      "▁embry": 15396,
+      "▁authorities": 15397,
+      "PER": 15398,
+      "生涯": 15399,
+      "▁Aud": 15400,
+      "▁hockey": 15401,
+      "职责": 15402,
+      "区的": 15403,
+      "▁Default": 15404,
+      "▁开本": 15405,
+      "aux": 15406,
+      "Effect": 15407,
+      "ORS": 15408,
+      "pping": 15409,
+      "专题": 15410,
+      "▁explo": 15411,
+      "Device": 15412,
+      "▁eigen": 15413,
+      "影片": 15414,
+      "分享到": 15415,
+      "evin": 15416,
+      ")$,": 15417,
+      "MAX": 15418,
+      "▁innovation": 15419,
+      "▁ease": 15420,
+      "同比": 15421,
+      "▁quote": 15422,
+      "▁WH": 15423,
+      "▁assert": 15424,
+      "stra": 15425,
+      "▁scr": 15426,
+      "▁episodes": 15427,
+      "以下简称": 15428,
+      "一直在": 15429,
+      "wt": 15430,
+      "等证据证实": 15431,
+      "新生": 15432,
+      "▁apart": 15433,
+      "▁adequ": 15434,
+      "▁Cat": 15435,
+      "two": 15436,
+      "tain": 15437,
+      "演出": 15438,
+      "▁attempts": 15439,
+      "▁representative": 15440,
+      "▁Simple": 15441,
+      "▁metric": 15442,
+      "传统的": 15443,
+      "▁asset": 15444,
+      "房地产开发有限公司": 15445,
+      "…\\": 15446,
+      "iber": 15447,
+      "十三条": 15448,
+      "rosis": 15449,
+      "RESS": 15450,
+      "▁LIC": 15451,
+      "answer": 15452,
+      "bytes": 15453,
+      "Active": 15454,
+      "emp": 15455,
+      "▁ham": 15456,
+      "领先": 15457,
+      "初中": 15458,
+      "ambig": 15459,
+      "▁alongside": 15460,
+      "ucid": 15461,
+      "▁submitted": 15462,
+      "▁Express": 15463,
+      "▁Bern": 15464,
+      "▁styl": 15465,
+      "进去": 15466,
+      "▁Obama": 15467,
+      "自信": 15468,
+      "▁conversation": 15469,
+      "则是": 15470,
+      "政协": 15471,
+      "▁mask": 15472,
+      "反而": 15473,
+      "显得": 15474,
+      "gypt": 15475,
+      "▁periods": 15476,
+      "cred": 15477,
+      "amount": 15478,
+      "加盟": 15479,
+      "三日": 15480,
+      "▁Title": 15481,
+      "▁honor": 15482,
+      "▁hosted": 15483,
+      "他是": 15484,
+      "▁tech": 15485,
+      "赢得": 15486,
+      "▁weekly": 15487,
+      "]$": 15488,
+      "▁sitting": 15489,
+      "▁trace": 15490,
+      "分辨": 15491,
+      "▁remark": 15492,
+      "全村": 15493,
+      "▁stopped": 15494,
+      "directory": 15495,
+      "▁reward": 15496,
+      "▁dysfunction": 15497,
+      "▁addresses": 15498,
+      "▁gon": 15499,
+      "▁Region": 15500,
+      "▁EV": 15501,
+      "▁Stan": 15502,
+      "requency": 15503,
+      "原式": 15504,
+      "寿命": 15505,
+      "▁Ba": 15506,
+      "books": 15507,
+      "▁meetings": 15508,
+      "▁Sir": 15509,
+      "▁Rights": 15510,
+      "听到": 15511,
+      "敏感": 15512,
+      "▁Republican": 15513,
+      "▁explanation": 15514,
+      "nPop": 15515,
+      "Ed": 15516,
+      "inline": 15517,
+      "三条": 15518,
+      "若干问题的规定": 15519,
+      "owner": 15520,
+      "▁myst": 15521,
+      "OH": 15522,
+      "产生的": 15523,
+      "▁Sus": 15524,
+      "odynam": 15525,
+      "▁Eth": 15526,
+      "▁throm": 15527,
+      "▁马": 15528,
+      "▁skill": 15529,
+      "缺点": 15530,
+      "metic": 15531,
+      "违约": 15532,
+      "▁belief": 15533,
+      "▁biology": 15534,
+      "▁knock": 15535,
+      "wrap": 15536,
+      "acking": 15537,
+      "adel": 15538,
+      "族自治区": 15539,
+      "▁Spr": 15540,
+      "▁placeholder": 15541,
+      "}^{\\\\": 15542,
+      "▁Tree": 15543,
+      "了本案": 15544,
+      "▁Cas": 15545,
+      "▁satisfaction": 15546,
+      "▁screens": 15547,
+      "rine": 15548,
+      "rho": 15549,
+      "wing": 15550,
+      "▁chlor": 15551,
+      "▁duplic": 15552,
+      "GT": 15553,
+      "atomic": 15554,
+      "高端": 15555,
+      "boolean": 15556,
+      "onomy": 15557,
+      "\\'],\\": 15558,
+      "▁Chair": 15559,
+      "▁combine": 15560,
+      "▁carefully": 15561,
+      "liter": 15562,
+      "ruption": 15563,
+      "argo": 15564,
+      "▁peripheral": 15565,
+      "ois": 15566,
+      "▁rotation": 15567,
+      "▁Der": 15568,
+      "▁caught": 15569,
+      "▁mild": 15570,
+      "▁FITNESS": 15571,
+      "▁suscept": 15572,
+      "的那": 15573,
+      "Dr": 15574,
+      "大力": 15575,
+      "鼠标": 15576,
+      "ierarch": 15577,
+      "才会": 15578,
+      "Const": 15579,
+      "planation": 15580,
+      "的过程": 15581,
+      "cler": 15582,
+      "▁von": 15583,
+      "▁physics": 15584,
+      "hs": 15585,
+      "▁lights": 15586,
+      "ipse": 15587,
+      "▁consecutive": 15588,
+      "让他": 15589,
+      "延长": 15590,
+      "不敢": 15591,
+      "▁PD": 15592,
+      "仪器": 15593,
+      "fish": 15594,
+      "nesota": 15595,
+      "▁{@": 15596,
+      "规定的": 15597,
+      "▁吴": 15598,
+      "遭遇": 15599,
+      "▁Bad": 15600,
+      "地震": 15601,
+      "▁Creek": 15602,
+      "正是": 15603,
+      "▁Additionally": 15604,
+      "▁premium": 15605,
+      "orer": 15606,
+      "不懂": 15607,
+      "nGet": 15608,
+      "宇宙": 15609,
+      "无正当理由拒不到庭": 15610,
+      "▁pars": 15611,
+      "人力资源": 15612,
+      "▁Watch": 15613,
+      "▁Rub": 15614,
+      "▁decom": 15615,
+      "转变": 15616,
+      "BUG": 15617,
+      "member": 15618,
+      "的数": 15619,
+      "zo": 15620,
+      "指定的期间": 15621,
+      "▁synthes": 15622,
+      "▁employment": 15623,
+      "▁ancient": 15624,
+      "表扬": 15625,
+      "▁permit": 15626,
+      "正确答案": 15627,
+      "ologically": 15628,
+      "yard": 15629,
+      "▁Massachusetts": 15630,
+      "星期": 15631,
+      "很容易": 15632,
+      "▁activated": 15633,
+      "▁elevated": 15634,
+      "的企业": 15635,
+      "▁docker": 15636,
+      "dart": 15637,
+      "一只": 15638,
+      "起点网": 15639,
+      "▁nm": 15640,
+      "rule": 15641,
+      "太平": 15642,
+      "▁authent": 15643,
+      "▁printed": 15644,
+      "更为": 15645,
+      "▁expanded": 15646,
+      "发病": 15647,
+      "validate": 15648,
+      "▁SQL": 15649,
+      "▁interactive": 15650,
+      "iva": 15651,
+      "és": 15652,
+      "���": 15653,
+      "▁yesterday": 15654,
+      "继承": 15655,
+      "▁promise": 15656,
+      "的结果": 15657,
+      "▁cardiovascular": 15658,
+      "usher": 15659,
+      "▁Round": 15660,
+      "lot": 15661,
+      "▁Plus": 15662,
+      "ogeneous": 15663,
+      "iem": 15664,
+      "直径": 15665,
+      "巨大": 15666,
+      "classes": 15667,
+      "▁Jackson": 15668,
+      "▁fifth": 15669,
+      "▁Kong": 15670,
+      "▁Korean": 15671,
+      "havior": 15672,
+      "▁worse": 15673,
+      "百姓": 15674,
+      "▁airport": 15675,
+      "犯盗窃罪": 15676,
+      "▁precision": 15677,
+      "▁birthday": 15678,
+      "▁Brook": 15679,
+      "nyou": 15680,
+      "▁quantity": 15681,
+      "收获": 15682,
+      "▁Emer": 15683,
+      "▁Instead": 15684,
+      "企业信用": 15685,
+      "面上": 15686,
+      "舞蹈": 15687,
+      "要求被告": 15688,
+      "▁territ": 15689,
+      "▁restaurants": 15690,
+      "▁Mill": 15691,
+      "天地": 15692,
+      "部的": 15693,
+      "▁inverse": 15694,
+      "传说": 15695,
+      "radius": 15696,
+      "分析了": 15697,
+      "telse": 15698,
+      "▁je": 15699,
+      "前往": 15700,
+      "Argument": 15701,
+      "amps": 15702,
+      "发育": 15703,
+      "▁Sat": 15704,
+      "razy": 15705,
+      "▁planet": 15706,
+      "▁loaded": 15707,
+      "Anal": 15708,
+      "写出": 15709,
+      "olecular": 15710,
+      "▁discipl": 15711,
+      "效应": 15712,
+      "▁cricket": 15713,
+      "直接向": 15714,
+      "变形": 15715,
+      "镜头": 15716,
+      "斯特": 15717,
+      "▁tack": 15718,
+      "这位": 15719,
+      "▁readonly": 15720,
+      "若干问题的": 15721,
+      "▁Prime": 15722,
+      "▁Professor": 15723,
+      "著作": 15724,
+      "次数": 15725,
+      "电商": 15726,
+      "▁specify": 15727,
+      "onomic": 15728,
+      "▁investigators": 15729,
+      "▁parser": 15730,
+      "▁radical": 15731,
+      "obe": 15732,
+      "▁glo": 15733,
+      "lywood": 15734,
+      "▁basically": 15735,
+      "▁ske": 15736,
+      "疼痛": 15737,
+      "▁ng": 15738,
+      "报送": 15739,
+      "▁uint": 15740,
+      "ocate": 15741,
+      "获奖": 15742,
+      "bserv": 15743,
+      "reset": 15744,
+      "▁investors": 15745,
+      "▁grew": 15746,
+      "号民事调解书": 15747,
+      "datetime": 15748,
+      "▁circumst": 15749,
+      "▁OK": 15750,
+      "萝卜": 15751,
+      "▁rating": 15752,
+      "adi": 15753,
+      "三种": 15754,
+      "买了": 15755,
+      "十月": 15756,
+      "anches": 15757,
+      "▁checking": 15758,
+      "导师": 15759,
+      "▁};\\": 15760,
+      "aturally": 15761,
+      "▁truck": 15762,
+      "▁philosoph": 15763,
+      "napp": 15764,
+      "转型": 15765,
+      "ogenous": 15766,
+      "▁objectives": 15767,
+      "考研": 15768,
+      "▁murder": 15769,
+      "应急": 15770,
+      "老公": 15771,
+      "七月": 15772,
+      "▁trick": 15773,
+      "六日": 15774,
+      "Find": 15775,
+      "▁abund": 15776,
+      "的形式": 15777,
+      "▁sending": 15778,
+      "pattern": 15779,
+      "公务": 15780,
+      "ermal": 15781,
+      "▁heav": 15782,
+      "Mapper": 15783,
+      "▁anxiety": 15784,
+      "▁batch": 15785,
+      "▁vital": 15786,
+      "讲解": 15787,
+      "▁gained": 15788,
+      "沒有": 15789,
+      "▁implant": 15790,
+      "请您": 15791,
+      "vy": 15792,
+      "nYear": 15793,
+      "▁hyd": 15794,
+      "▁lovely": 15795,
+      "豆瓣读书": 15796,
+      "心情": 15797,
+      "▁dialog": 15798,
+      "对被执行人": 15799,
+      "▁supplies": 15800,
+      "▁Place": 15801,
+      "environment": 15802,
+      "管道": 15803,
+      "Profile": 15804,
+      "▁Cancer": 15805,
+      "▁sle": 15806,
+      "仓库": 15807,
+      "▁principle": 15808,
+      "Loader": 15809,
+      "adv": 15810,
+      "alo": 15811,
+      "▁requested": 15812,
+      "▁Right": 15813,
+      "▁Steve": 15814,
+      "▁criminal": 15815,
+      "▁Ball": 15816,
+      "▁knowing": 15817,
+      "Debug": 15818,
+      "▁Less": 15819,
+      "一片": 15820,
+      "Microsoft": 15821,
+      "暴力": 15822,
+      "共产党": 15823,
+      "热情": 15824,
+      "VS": 15825,
+      "▁lic": 15826,
+      "▁desire": 15827,
+      "heel": 15828,
+      "ydney": 15829,
+      "Big": 15830,
+      "测定": 15831,
+      "取值范围": 15832,
+      "▁libraries": 15833,
+      "ção": 15834,
+      "▁childhood": 15835,
+      "Rest": 15836,
+      "本裁定书送达后": 15837,
+      "▁closest": 15838,
+      "第一百四十四": 15839,
+      "igue": 15840,
+      "ieving": 15841,
+      "THER": 15842,
+      "xs": 15843,
+      "nX": 15844,
+      "大约": 15845,
+      "差不多": 15846,
+      "▁})": 15847,
+      "adesh": 15848,
+      "少女": 15849,
+      "bus": 15850,
+      "财富": 15851,
+      "每日": 15852,
+      "tail": 15853,
+      "回复": 15854,
+      "Ins": 15855,
+      "algorithm": 15856,
+      "▁rab": 15857,
+      "▁Cert": 15858,
+      "▁aging": 15859,
+      "bas": 15860,
+      "是连载于": 15861,
+      "▁advantages": 15862,
+      "stats": 15863,
+      "AK": 15864,
+      "[\\\\": 15865,
+      "▁frequent": 15866,
+      "履行给付金钱": 15867,
+      "nMay": 15868,
+      "▁上": 15869,
+      "shell": 15870,
+      "▁managing": 15871,
+      "Channel": 15872,
+      "种类": 15873,
+      "西北": 15874,
+      "ponential": 15875,
+      "▁unus": 15876,
+      "农村商业": 15877,
+      "▁Determine": 15878,
+      "农村信用": 15879,
+      "ervative": 15880,
+      "原创": 15881,
+      "▁decimal": 15882,
+      "inating": 15883,
+      "▁automatic": 15884,
+      "subset": 15885,
+      "了一些": 15886,
+      "logo": 15887,
+      "主办": 15888,
+      "▁alleged": 15889,
+      "▁forecast": 15890,
+      "▁mitochond": 15891,
+      "▁happening": 15892,
+      "▁glut": 15893,
+      "PK": 15894,
+      "离子": 15895,
+      "▁robot": 15896,
+      "▁comedy": 15897,
+      "税务": 15898,
+      "加床": 15899,
+      "XT": 15900,
+      "oli": 15901,
+      "利亚": 15902,
+      "病人": 15903,
+      "▁spr": 15904,
+      "Console": 15905,
+      "无数": 15906,
+      "▁meat": 15907,
+      "▁mining": 15908,
+      "▁necessarily": 15909,
+      "▁tea": 15910,
+      "缓解": 15911,
+      "done": 15912,
+      "▁thorough": 15913,
+      "▁outstanding": 15914,
+      "摄像": 15915,
+      "▁anticip": 15916,
+      "▁Case": 15917,
+      "▁Status": 15918,
+      "arry": 15919,
+      "▁smoking": 15920,
+      "▁execution": 15921,
+      "OC": 15922,
+      "▁driven": 15923,
+      "拍卖": 15924,
+      "一套": 15925,
+      "配件": 15926,
+      "舒服": 15927,
+      "中华人民共和国合同法": 15928,
+      "▁neutral": 15929,
+      "▁Zbl": 15930,
+      "同事": 15931,
+      "Hello": 15932,
+      "alert": 15933,
+      "▁LICENSE": 15934,
+      "带领": 15935,
+      "▁Edition": 15936,
+      "▁retired": 15937,
+      "▁meta": 15938,
+      "宿舍": 15939,
+      "审理过程中": 15940,
+      "▁Wisconsin": 15941,
+      "apos": 15942,
+      "多人": 15943,
+      "第三款": 15944,
+      "▁Colorado": 15945,
+      "▁seemed": 15946,
+      "aska": 15947,
+      "监狱服刑": 15948,
+      "▁designer": 15949,
+      "▁redist": 15950,
+      "▁Before": 15951,
+      "判决执行": 15952,
+      "▁conflict": 15953,
+      "ussion": 15954,
+      "窗口": 15955,
+      "Rev": 15956,
+      "办学": 15957,
+      "退休": 15958,
+      "fb": 15959,
+      "一句": 15960,
+      "ustral": 15961,
+      "小型": 15962,
+      "回忆": 15963,
+      "主管": 15964,
+      "▁informed": 15965,
+      "科学技术": 15966,
+      "▁CM": 15967,
+      "▁victory": 15968,
+      "ugs": 15969,
+      "▁fo": 15970,
+      "如实供述": 15971,
+      "▁granted": 15972,
+      "场上": 15973,
+      "▁measuring": 15974,
+      "▁sampling": 15975,
+      "nlet": 15976,
+      "Sing": 15977,
+      "▁Room": 15978,
+      "人事": 15979,
+      "appoint": 15980,
+      "▁wearing": 15981,
+      "▁era": 15982,
+      "nMembers": 15983,
+      "▁progn": 15984,
+      "////////////////": 15985,
+      "▁BC": 15986,
+      "提供被执行人": 15987,
+      "▁lighting": 15988,
+      "▁rail": 15989,
+      "认罪悔罪": 15990,
+      "后果": 15991,
+      "▁intersection": 15992,
+      "▁checks": 15993,
+      "城镇": 15994,
+      "payload": 15995,
+      "Note": 15996,
+      "▁Building": 15997,
+      "▁Contin": 15998,
+      "分享到朋友圈": 15999,
+      "VC": 16000,
+      "sdk": 16001,
+      "nReg": 16002,
+      "跟着": 16003,
+      "▁matters": 16004,
+      "▁Bon": 16005,
+      "▁Shop": 16006,
+      "前面": 16007,
+      "▁stretch": 16008,
+      "nMore": 16009,
+      "小孩": 16010,
+      "▁Ash": 16011,
+      "三角形的": 16012,
+      "▁Small": 16013,
+      "▁Columbia": 16014,
+      "名词": 16015,
+      "Tool": 16016,
+      "▁Element": 16017,
+      "对应的": 16018,
+      "▁hydrogen": 16019,
+      "▁Turn": 16020,
+      "▁Bed": 16021,
+      "amilton": 16022,
+      "▁lie": 16023,
+      "长时间": 16024,
+      "新建": 16025,
+      "▁determination": 16026,
+      "新鲜": 16027,
+      "高兴": 16028,
+      "allenge": 16029,
+      "▁Fox": 16030,
+      "▁speaking": 16031,
+      "激光": 16032,
+      "nAm": 16033,
+      "apore": 16034,
+      "▁HTTP": 16035,
+      "thal": 16036,
+      "▁Queen": 16037,
+      "▁trig": 16038,
+      "hematic": 16039,
+      "出席": 16040,
+      "球员": 16041,
+      "Av": 16042,
+      "condition": 16043,
+      "questions": 16044,
+      "打击": 16045,
+      "▁spectro": 16046,
+      "rase": 16047,
+      "▁statements": 16048,
+      "itus": 16049,
+      "▁WARRANTY": 16050,
+      "美容": 16051,
+      "▁kernel": 16052,
+      "eties": 16053,
+      "nClick": 16054,
+      "▁tube": 16055,
+      "▁entertainment": 16056,
+      "▁expectations": 16057,
+      "▁SOFTWARE": 16058,
+      "pton": 16059,
+      "哥哥": 16060,
+      "▁zip": 16061,
+      "印刷": 16062,
+      "hm": 16063,
+      "之日起计算": 16064,
+      "▁Daniel": 16065,
+      "speed": 16066,
+      "无论是": 16067,
+      "▁impossible": 16068,
+      "socket": 16069,
+      "拓展": 16070,
+      "▁filed": 16071,
+      "▁paying": 16072,
+      "▁ty": 16073,
+      "又有": 16074,
+      "▁respiratory": 16075,
+      "atom": 16076,
+      "Stud": 16077,
+      "cue": 16078,
+      "▁Ut": 16079,
+      "amond": 16080,
+      "将文章": 16081,
+      "▁hole": 16082,
+      "国庆": 16083,
+      "电子商务": 16084,
+      "前后": 16085,
+      "messages": 16086,
+      "生日": 16087,
+      "▁contest": 16088,
+      "profit": 16089,
+      "▁affects": 16090,
+      "Left": 16091,
+      "▁sorry": 16092,
+      "ilst": 16093,
+      "冲击": 16094,
+      "单纯": 16095,
+      "▁Block": 16096,
+      "onda": 16097,
+      "fmt": 16098,
+      "▁meal": 16099,
+      "▁moments": 16100,
+      "kes": 16101,
+      "athan": 16102,
+      "▁Hop": 16103,
+      "▁coronary": 16104,
+      "少数": 16105,
+      "照顾": 16106,
+      "\\\\\\": 16107,
+      "生效法律文书确定的义务": 16108,
+      "腾讯": 16109,
+      "审理了本案": 16110,
+      "OUT": 16111,
+      "nPre": 16112,
+      "财产保全": 16113,
+      "▁banks": 16114,
+      "▁determining": 16115,
+      "团结": 16116,
+      "addr": 16117,
+      "的就是": 16118,
+      "▁diagram": 16119,
+      "▁judge": 16120,
+      "eph": 16121,
+      "▁routes": 16122,
+      "▁Ok": 16123,
+      "机器人": 16124,
+      "的有": 16125,
+      "▁solved": 16126,
+      "之路": 16127,
+      "了吧": 16128,
+      "Any": 16129,
+      "Mem": 16130,
+      "▁IC": 16131,
+      "所需": 16132,
+      "将文章分享到朋友圈": 16133,
+      "▁Latin": 16134,
+      "arters": 16135,
+      "BB": 16136,
+      "stable": 16137,
+      "半径": 16138,
+      "多元": 16139,
+      "collection": 16140,
+      "▁nerv": 16141,
+      "▁ip": 16142,
+      "▁Print": 16143,
+      "的效果": 16144,
+      "▁简介": 16145,
+      "otyp": 16146,
+      "▁cov": 16147,
+      "inx": 16148,
+      "▁Ly": 16149,
+      "hh": 16150,
+      "并非": 16151,
+      "acon": 16152,
+      "switch": 16153,
+      "粮食": 16154,
+      "▁stuck": 16155,
+      "匹配": 16156,
+      "第三节": 16157,
+      "▁claimed": 16158,
+      "合金": 16159,
+      "▁Rod": 16160,
+      "psi": 16161,
+      "▁suggestions": 16162,
+      "江南": 16163,
+      "legate": 16164,
+      "▁recognize": 16165,
+      "arios": 16166,
+      "windows": 16167,
+      "▁repeat": 16168,
+      "Auto": 16169,
+      "男女": 16170,
+      "▁debug": 16171,
+      "哈尔滨": 16172,
+      "abeth": 16173,
+      "orous": 16174,
+      "▁coal": 16175,
+      "▁breaking": 16176,
+      "seud": 16177,
+      "▁Know": 16178,
+      "▁Training": 16179,
+      "▁leaf": 16180,
+      "和其他": 16181,
+      "六个月": 16182,
+      "▁province": 16183,
+      "▁alert": 16184,
+      "姑娘": 16185,
+      "▁Enc": 16186,
+      "▁dealing": 16187,
+      "tip": 16188,
+      "保修": 16189,
+      "未按本判决": 16190,
+      "▁``": 16191,
+      "adder": 16192,
+      "▁keeps": 16193,
+      "昨天": 16194,
+      "要是": 16195,
+      "what": 16196,
+      "信息公示": 16197,
+      "同类": 16198,
+      "▁sudden": 16199,
+      "生意": 16200,
+      "的解": 16201,
+      "下了": 16202,
+      "公斤": 16203,
+      "贫困": 16204,
+      "Step": 16205,
+      "aret": 16206,
+      "履行给付金钱义务": 16207,
+      "▁moderate": 16208,
+      "acher": 16209,
+      "regon": 16210,
+      "刑罚": 16211,
+      "直角": 16212,
+      "▁lunch": 16213,
+      "zech": 16214,
+      "失去": 16215,
+      "忘记": 16216,
+      "成人": 16217,
+      "当天": 16218,
+      "▁Pakistan": 16219,
+      "▁Harr": 16220,
+      "▁inhibitor": 16221,
+      "▁Eastern": 16222,
+      "scill": 16223,
+      "ulp": 16224,
+      "▁hormone": 16225,
+      "▁Justice": 16226,
+      "']\\": 16227,
+      "二日": 16228,
+      "▁wireless": 16229,
+      "▁estimation": 16230,
+      "▁Donald": 16231,
+      "▁flash": 16232,
+      "▁picking": 16233,
+      "所在地": 16234,
+      "▁stead": 16235,
+      "这两个": 16236,
+      "▁modification": 16237,
+      "▁bacterial": 16238,
+      "▁ourselves": 16239,
+      "batch": 16240,
+      "种子": 16241,
+      "ecl": 16242,
+      "▁maintaining": 16243,
+      "上有": 16244,
+      "Err": 16245,
+      "▁HO": 16246,
+      "▁Sa": 16247,
+      "到期": 16248,
+      "发表于": 16249,
+      "的一部": 16250,
+      "▁crew": 16251,
+      "▁arts": 16252,
+      "▁lipid": 16253,
+      "充电": 16254,
+      "bean": 16255,
+      "▁wrap": 16256,
+      "▁Gi": 16257,
+      "\");": 16258,
+      "▁Stock": 16259,
+      "gz": 16260,
+      "▁Dutch": 16261,
+      "nCONCLUS": 16262,
+      "医药": 16263,
+      "▁logo": 16264,
+      "▁covering": 16265,
+      "▁Winter": 16266,
+      "▁joining": 16267,
+      "▁constraints": 16268,
+      "integr": 16269,
+      "olester": 16270,
+      "including": 16271,
+      "主编": 16272,
+      "▁prefix": 16273,
+      "四月": 16274,
+      "fi": 16275,
+      "▁Select": 16276,
+      "诚信": 16277,
+      "IME": 16278,
+      "▁Jr": 16279,
+      "Common": 16280,
+      "▁politics": 16281,
+      "hidden": 16282,
+      "vi": 16283,
+      "vue": 16284,
+      "语文": 16285,
+      "周长": 16286,
+      "▁atmosphere": 16287,
+      "GC": 16288,
+      "ENRON": 16289,
+      "道理": 16290,
+      "▁Os": 16291,
+      "▁infected": 16292,
+      "反对": 16293,
+      "西亚": 16294,
+      "复印": 16295,
+      "精准": 16296,
+      "▁Theory": 16297,
+      "▁meth": 16298,
+      "奋斗": 16299,
+      "sleep": 16300,
+      "常用": 16301,
+      "压缩": 16302,
+      "存在的": 16303,
+      "信息化": 16304,
+      "▁wat": 16305,
+      "中药": 16306,
+      "水泥": 16307,
+      "原本": 16308,
+      "自动化": 16309,
+      "▁imagine": 16310,
+      "▁vaccine": 16311,
+      "厂商": 16312,
+      "▁Given": 16313,
+      "社交": 16314,
+      "防护": 16315,
+      "tures": 16316,
+      "▁Short": 16317,
+      "▁Section": 16318,
+      "▁competed": 16319,
+      "▁Ak": 16320,
+      "▁equipped": 16321,
+      "▁underwent": 16322,
+      "▁bind": 16323,
+      "▁lift": 16324,
+      "nPer": 16325,
+      "▁Route": 16326,
+      "尚未": 16327,
+      "labels": 16328,
+      "▁terror": 16329,
+      "▁fed": 16330,
+      "irty": 16331,
+      "▁Hong": 16332,
+      "▁cock": 16333,
+      "▁council": 16334,
+      "典型": 16335,
+      "▁Broad": 16336,
+      "的相关": 16337,
+      "反复": 16338,
+      "uting": 16339,
+      "旋转": 16340,
+      "▁allerg": 16341,
+      "▁tutorial": 16342,
+      "一行": 16343,
+      "▁vast": 16344,
+      "▁graduated": 16345,
+      "dam": 16346,
+      "javascript": 16347,
+      "神秘": 16348,
+      "其行为": 16349,
+      "▁lens": 16350,
+      "键盘": 16351,
+      "NOT": 16352,
+      "▁[[": 16353,
+      "keep": 16354,
+      "▁Kr": 16355,
+      "README": 16356,
+      "ifi": 16357,
+      "Job": 16358,
+      "复制": 16359,
+      "ogram": 16360,
+      "Reference": 16361,
+      "地质": 16362,
+      "成立于": 16363,
+      "esity": 16364,
+      "▁identifying": 16365,
+      "twitter": 16366,
+      "▁tro": 16367,
+      "xad": 16368,
+      "olid": 16369,
+      "小游戏": 16370,
+      "▁strip": 16371,
+      "eek": 16372,
+      "▁membership": 16373,
+      "tconst": 16374,
+      "一生": 16375,
+      "活性": 16376,
+      "地理位置": 16377,
+      "▁visited": 16378,
+      "Rule": 16379,
+      "一款": 16380,
+      "prom": 16381,
+      "景区": 16382,
+      "患者的": 16383,
+      "厉害": 16384,
+      "资讯": 16385,
+      "主人": 16386,
+      "▁Ext": 16387,
+      "▁entreprene": 16388,
+      "▁shel": 16389,
+      "Close": 16390,
+      "所示": 16391,
+      "公平": 16392,
+      "▁à": 16393,
+      "▁iOS": 16394,
+      "pload": 16395,
+      "changed": 16396,
+      "二级": 16397,
+      "▁colle": 16398,
+      "Many": 16399,
+      "▁wealth": 16400,
+      "epage": 16401,
+      "parameters": 16402,
+      "asset": 16403,
+      "究竟": 16404,
+      "▁fuck": 16405,
+      "▁possess": 16406,
+      "▁econom": 16407,
+      "国有": 16408,
+      "出庭": 16409,
+      "男孩": 16410,
+      "AMP": 16411,
+      "▁tall": 16412,
+      "局部": 16413,
+      "▁randomized": 16414,
+      "Press": 16415,
+      "创作的网络小说": 16416,
+      "Required": 16417,
+      "的两": 16418,
+      "妇女": 16419,
+      "uda": 16420,
+      "▁faces": 16421,
+      "▁Computer": 16422,
+      "山县": 16423,
+      "ashed": 16424,
+      "▁reliability": 16425,
+      "路线": 16426,
+      "马克": 16427,
+      "专用": 16428,
+      "▁dropped": 16429,
+      "▁Boy": 16430,
+      "▁assistant": 16431,
+      "▁Pract": 16432,
+      "through": 16433,
+      "cost": 16434,
+      "营造": 16435,
+      "截至": 16436,
+      "inator": 16437,
+      "▁Country": 16438,
+      "▁Down": 16439,
+      "你说": 16440,
+      "▁calculation": 16441,
+      "▁Valid": 16442,
+      "▁lifestyle": 16443,
+      "projects": 16444,
+      "SN": 16445,
+      "▁molecule": 16446,
+      "holders": 16447,
+      "信息公示系统": 16448,
+      "nOver": 16449,
+      "▁millions": 16450,
+      "▁urg": 16451,
+      "▁affil": 16452,
+      "▁isinstance": 16453,
+      "▁reply": 16454,
+      "▁Content": 16455,
+      "▁LED": 16456,
+      "each": 16457,
+      "▁shooting": 16458,
+      "对我": 16459,
+      "UST": 16460,
+      "周年": 16461,
+      "有利于": 16462,
+      "▁seats": 16463,
+      "飞行": 16464,
+      "▁产品": 16465,
+      "▁horse": 16466,
+      "nWhich": 16467,
+      "▁Ontario": 16468,
+      "▁gallery": 16469,
+      "▁sql": 16470,
+      "的条件": 16471,
+      "使命": 16472,
+      "加油": 16473,
+      "▁muc": 16474,
+      "▁engagement": 16475,
+      "▁complexes": 16476,
+      "▁preferred": 16477,
+      "这段": 16478,
+      "atile": 16479,
+      "seq": 16480,
+      "aternal": 16481,
+      "gment": 16482,
+      "是要": 16483,
+      "研究院": 16484,
+      "加热": 16485,
+      "▁choosing": 16486,
+      "▁Conf": 16487,
+      "▁Toronto": 16488,
+      "▁controvers": 16489,
+      "考验": 16490,
+      "▁tank": 16491,
+      "沈阳": 16492,
+      "企业信用信息公示系统": 16493,
+      "▁waves": 16494,
+      "级的": 16495,
+      "yield": 16496,
+      "▁Nic": 16497,
+      "宁夏": 16498,
+      "▁shouldn": 16499,
+      "家居": 16500,
+      "绿化": 16501,
+      "▁cytok": 16502,
+      "ante": 16503,
+      "对该": 16504,
+      "▁Self": 16505,
+      "BE": 16506,
+      "▁parad": 16507,
+      "▁golf": 16508,
+      "negative": 16509,
+      "欠条": 16510,
+      "▁emerging": 16511,
+      "▁adopted": 16512,
+      "nDef": 16513,
+      "▁bike": 16514,
+      "转让": 16515,
+      "▁Arg": 16516,
+      "▁fairly": 16517,
+      "的选择": 16518,
+      "▁nonlinear": 16519,
+      "apply": 16520,
+      "▁平装": 16521,
+      ",$": 16522,
+      "▁fitness": 16523,
+      "教练": 16524,
+      "取得了": 16525,
+      "elihood": 16526,
+      "▁YouTube": 16527,
+      "商家": 16528,
+      "给大家": 16529,
+      "你就": 16530,
+      "▁____": 16531,
+      "▁pred": 16532,
+      "聊天": 16533,
+      "▁Rs": 16534,
+      "▁addressed": 16535,
+      "▁precise": 16536,
+      "▁和": 16537,
+      "传承": 16538,
+      "▁rounded": 16539,
+      "请仔细": 16540,
+      "▁sustainable": 16541,
+      "OLD": 16542,
+      "ibration": 16543,
+      "文物": 16544,
+      "loom": 16545,
+      "也可": 16546,
+      "rait": 16547,
+      "强大": 16548,
+      "immun": 16549,
+      "▁witness": 16550,
+      "opl": 16551,
+      "Details": 16552,
+      "▁Collection": 16553,
+      "▁Has": 16554,
+      "TestCase": 16555,
+      "kind": 16556,
+      "▁chicken": 16557,
+      "Parameters": 16558,
+      "▁sulf": 16559,
+      "▁forum": 16560,
+      "▁drinking": 16561,
+      "▁bread": 16562,
+      "▁minister": 16563,
+      "▁horizontal": 16564,
+      "▁]\\": 16565,
+      "capt": 16566,
+      "业绩": 16567,
+      "开了": 16568,
+      "独任": 16569,
+      "▁guard": 16570,
+      "的变化": 16571,
+      "▁YOU": 16572,
+      "▁\"\\": 16573,
+      "▁losing": 16574,
+      "百万": 16575,
+      "邮政": 16576,
+      "▁bear": 16577,
+      "▁delicious": 16578,
+      "▁prof": 16579,
+      "▁quiet": 16580,
+      "显卡": 16581,
+      "厂家": 16582,
+      "Ag": 16583,
+      "mk": 16584,
+      "▁fetch": 16585,
+      "▁ranging": 16586,
+      "之下": 16587,
+      "▁los": 16588,
+      "招生": 16589,
+      "大家都": 16590,
+      "Cons": 16591,
+      "领域的": 16592,
+      "并将": 16593,
+      "Cell": 16594,
+      "proxy": 16595,
+      "oose": 16596,
+      "Vert": 16597,
+      "enda": 16598,
+      "expression": 16599,
+      "▁festival": 16600,
+      "法师": 16601,
+      "节奏": 16602,
+      "city": 16603,
+      "ldots": 16604,
+      "硬盘": 16605,
+      "强大的": 16606,
+      "eneration": 16607,
+      "Prot": 16608,
+      "▁appreciated": 16609,
+      "▁cable": 16610,
+      "▁代码": 16611,
+      "▁titles": 16612,
+      "▁queue": 16613,
+      "▁μ": 16614,
+      "nWell": 16615,
+      "norm": 16616,
+      "▁firms": 16617,
+      "国家企业信用信息公示系统": 16618,
+      "▁crash": 16619,
+      "nLe": 16620,
+      "Gen": 16621,
+      "▁adverse": 16622,
+      "\":[": 16623,
+      "▁marks": 16624,
+      "▁Material": 16625,
+      "地上": 16626,
+      "▁ampl": 16627,
+      "▁returning": 16628,
+      "provider": 16629,
+      "社会的": 16630,
+      "bing": 16631,
+      "ilibrium": 16632,
+      "rizona": 16633,
+      "fail": 16634,
+      "yz": 16635,
+      "▁Posted": 16636,
+      "▁hint": 16637,
+      "正本": 16638,
+      "MIT": 16639,
+      "▁\"%": 16640,
+      "Sync": 16641,
+      "▁grass": 16642,
+      "本裁定送达后立即生效": 16643,
+      "▁identical": 16644,
+      "abling": 16645,
+      "sn": 16646,
+      "建筑面积": 16647,
+      "exception": 16648,
+      "起诉书": 16649,
+      "gage": 16650,
+      "▁Lead": 16651,
+      "anti": 16652,
+      "▁separated": 16653,
+      "▁methodology": 16654,
+      "▁图书": 16655,
+      "lov": 16656,
+      "▁acquisition": 16657,
+      "▁Minnesota": 16658,
+      "▁extreme": 16659,
+      "▁consideration": 16660,
+      "▁Chapter": 16661,
+      "发送": 16662,
+      "▁whenever": 16663,
+      "Dep": 16664,
+      "▁physiological": 16665,
+      "毕业于": 16666,
+      "▁everyday": 16667,
+      "called": 16668,
+      "▁genome": 16669,
+      "aren": 16670,
+      "▁nursing": 16671,
+      "▁OTHER": 16672,
+      "▁peptide": 16673,
+      "自觉": 16674,
+      "▁Lew": 16675,
+      "此题": 16676,
+      "rin": 16677,
+      "Network": 16678,
+      "默认": 16679,
+      "▁Mountain": 16680,
+      "叫做": 16681,
+      "▁propag": 16682,
+      "消除": 16683,
+      "未成年": 16684,
+      "▁neighb": 16685,
+      "▁Victoria": 16686,
+      "地面": 16687,
+      "建成": 16688,
+      "▁coordinate": 16689,
+      "án": 16690,
+      "▁sentence": 16691,
+      "FP": 16692,
+      "▁interviews": 16693,
+      "▁acet": 16694,
+      "▁clot": 16695,
+      "equality": 16696,
+      "href": 16697,
+      "modal": 16698,
+      "▁cyl": 16699,
+      "\";": 16700,
+      "▁»": 16701,
+      "的外": 16702,
+      "▁somewhat": 16703,
+      "▁tang": 16704,
+      "▁realized": 16705,
+      "▁simultaneously": 16706,
+      "▁wound": 16707,
+      "奥运": 16708,
+      "▁disk": 16709,
+      "过滤": 16710,
+      "▁Ministry": 16711,
+      "▁Portug": 16712,
+      "pler": 16713,
+      "LR": 16714,
+      "ennes": 16715,
+      "发放": 16716,
+      "难道": 16717,
+      "igure": 16718,
+      "Names": 16719,
+      "▁appl": 16720,
+      "▁curves": 16721,
+      "enum": 16722,
+      "编制": 16723,
+      "owntown": 16724,
+      "TT": 16725,
+      "出租": 16726,
+      "otted": 16727,
+      "证人": 16728,
+      "ulus": 16729,
+      "▁Clean": 16730,
+      "▁props": 16731,
+      "▁fails": 16732,
+      "▁moist": 16733,
+      "▁converted": 16734,
+      "nSports": 16735,
+      "▁closing": 16736,
+      "▁民族": 16737,
+      "agic": 16738,
+      "从选项中": 16739,
+      "定制": 16740,
+      "nCons": 16741,
+      "▁Band": 16742,
+      "▁nam": 16743,
+      "▁Lat": 16744,
+      "▁AB": 16745,
+      "AE": 16746,
+      "▁Egypt": 16747,
+      "▁Six": 16748,
+      "▁是": 16749,
+      "best": 16750,
+      "冲突": 16751,
+      "郑州": 16752,
+      "rangle": 16753,
+      "anz": 16754,
+      "协商": 16755,
+      "▁inputs": 16756,
+      "▁bits": 16757,
+      "▁participated": 16758,
+      "▁disturb": 16759,
+      "传媒": 16760,
+      "ouri": 16761,
+      "▁focusing": 16762,
+      "高温": 16763,
+      "▁marg": 16764,
+      "竞赛": 16765,
+      "▁turning": 16766,
+      "ouses": 16767,
+      "▁bat": 16768,
+      "▁cultures": 16769,
+      "的价格": 16770,
+      "ordan": 16771,
+      "祖国": 16772,
+      "▁Pos": 16773,
+      "▁reduces": 16774,
+      "▁penal": 16775,
+      "▁Client": 16776,
+      "▁behaviors": 16777,
+      "截止": 16778,
+      "超越": 16779,
+      "找出": 16780,
+      "nplt": 16781,
+      "▁transportation": 16782,
+      ",\\'": 16783,
+      "folder": 16784,
+      "nodes": 16785,
+      "Rad": 16786,
+      "▁invited": 16787,
+      "oda": 16788,
+      "一方": 16789,
+      "武汉市": 16790,
+      "ermission": 16791,
+      "浪费": 16792,
+      "这部": 16793,
+      "▁outer": 16794,
+      "▁matrices": 16795,
+      "IDS": 16796,
+      "▁Mobile": 16797,
+      "明天": 16798,
+      "enron": 16799,
+      "▁manifest": 16800,
+      "▁thickness": 16801,
+      "▁Specific": 16802,
+      "antly": 16803,
+      "bourne": 16804,
+      "▁hook": 16805,
+      "▁directions": 16806,
+      "▁enzymes": 16807,
+      "的数据": 16808,
+      "信任": 16809,
+      "▁offices": 16810,
+      "▁Price": 16811,
+      "▁boxes": 16812,
+      "电气": 16813,
+      "ropical": 16814,
+      "成都市": 16815,
+      "▁tomor": 16816,
+      "剩余": 16817,
+      "si": 16818,
+      "▁neither": 16819,
+      "▁developer": 16820,
+      "黄色": 16821,
+      "made": 16822,
+      "ür": 16823,
+      "消耗": 16824,
+      "Na": 16825,
+      "反馈": 16826,
+      "acet": 16827,
+      "▁MD": 16828,
+      "▁Develop": 16829,
+      "ifndef": 16830,
+      "顾问": 16831,
+      "nRESULTS": 16832,
+      "▁somewhere": 16833,
+      "发动机": 16834,
+      "不得不": 16835,
+      "▁Philipp": 16836,
+      "▁arrested": 16837,
+      "то": 16838,
+      "第一章": 16839,
+      "▁一": 16840,
+      "青少年": 16841,
+      "phal": 16842,
+      "ques": 16843,
+      "▁payments": 16844,
+      "更换": 16845,
+      "▁Need": 16846,
+      "▁vibr": 16847,
+      "▁fer": 16848,
+      "don": 16849,
+      "▁meters": 16850,
+      "的服务": 16851,
+      "▁lawyer": 16852,
+      "▁variations": 16853,
+      "uer": 16854,
+      "就行": 16855,
+      "nOut": 16856,
+      "▁continuing": 16857,
+      "大厦": 16858,
+      "明明": 16859,
+      "▁variant": 16860,
+      "aron": 16861,
+      "▁socket": 16862,
+      "权的": 16863,
+      "nre": 16864,
+      "起到": 16865,
+      "肌肉": 16866,
+      "overflow": 16867,
+      "▁correction": 16868,
+      "sky": 16869,
+      "招聘": 16870,
+      "妻子": 16871,
+      "▁gall": 16872,
+      "▁dil": 16873,
+      "▁attractive": 16874,
+      "▁cub": 16875,
+      "甘肃省": 16876,
+      "ushing": 16877,
+      "bg": 16878,
+      "多样": 16879,
+      "Only": 16880,
+      "ectors": 16881,
+      "agonal": 16882,
+      "World": 16883,
+      "技术的": 16884,
+      "▁exhibit": 16885,
+      "▁hosting": 16886,
+      "▁consisting": 16887,
+      "▁ignore": 16888,
+      "▁decline": 16889,
+      "部件": 16890,
+      "positive": 16891,
+      "▁徐": 16892,
+      "你在": 16893,
+      "▁adhes": 16894,
+      "▁Princ": 16895,
+      "HERE": 16896,
+      "▁turb": 16897,
+      "law": 16898,
+      "neum": 16899,
+      "▁tomorrow": 16900,
+      "▁bab": 16901,
+      "经济发展": 16902,
+      "▁Subject": 16903,
+      "公众号": 16904,
+      "nUse": 16905,
+      "some": 16906,
+      "▁subsequently": 16907,
+      "▁DB": 16908,
+      "nDon": 16909,
+      "农村商业银行股份有限公司": 16910,
+      "不开": 16911,
+      "干燥": 16912,
+      "▁bold": 16913,
+      "发言": 16914,
+      "share": 16915,
+      "_,": 16916,
+      "▁baseball": 16917,
+      "illary": 16918,
+      "ucl": 16919,
+      "▁sear": 16920,
+      "obby": 16921,
+      "Down": 16922,
+      "歌手": 16923,
+      "引进": 16924,
+      "合法权益": 16925,
+      "▁graft": 16926,
+      "▁keyboard": 16927,
+      "IH": 16928,
+      "▁intake": 16929,
+      "▁tensor": 16930,
+      "Generator": 16931,
+      "彼此": 16932,
+      "atalog": 16933,
+      "合成": 16934,
+      "这类": 16935,
+      "▁rear": 16936,
+      "▁strategic": 16937,
+      "差距": 16938,
+      "engers": 16939,
+      "nTwo": 16940,
+      "▁infer": 16941,
+      "▁byte": 16942,
+      "ARD": 16943,
+      "links": 16944,
+      "ortic": 16945,
+      "▁Raj": 16946,
+      "apl": 16947,
+      "HTTP": 16948,
+      "▁Visual": 16949,
+      "发达": 16950,
+      "假设": 16951,
+      "城乡": 16952,
+      "▁substance": 16953,
+      "▁Sex": 16954,
+      "东莞": 16955,
+      "通过本院": 16956,
+      "editor": 16957,
+      "mn": 16958,
+      "角的": 16959,
+      "otor": 16960,
+      "▁inspiration": 16961,
+      "▁Pur": 16962,
+      "▁calculations": 16963,
+      "欣赏": 16964,
+      "气温": 16965,
+      "▁foods": 16966,
+      "▁apoptosis": 16967,
+      "扮演": 16968,
+      "imit": 16969,
+      "icer": 16970,
+      "▁finds": 16971,
+      "县人民检察院": 16972,
+      "footer": 16973,
+      "▁Created": 16974,
+      "▁Avenue": 16975,
+      "aug": 16976,
+      "EV": 16977,
+      "$)": 16978,
+      "▁regardless": 16979,
+      "▁−": 16980,
+      "Wrapper": 16981,
+      "living": 16982,
+      "imental": 16983,
+      "没什么": 16984,
+      "equiv": 16985,
+      "▁synthetic": 16986,
+      "▁Civil": 16987,
+      "▁Sydney": 16988,
+      "我院": 16989,
+      "Sk": 16990,
+      "▁Clinical": 16991,
+      "▁severity": 16992,
+      "▁beer": 16993,
+      "binding": 16994,
+      "astro": 16995,
+      "▁administered": 16996,
+      "吃的": 16997,
+      "实现了": 16998,
+      "ashes": 16999,
+      "▁healing": 17000,
+      "oviet": 17001,
+      "outube": 17002,
+      "给他": 17003,
+      "打算": 17004,
+      "naire": 17005,
+      "▁Daily": 17006,
+      "▁nervous": 17007,
+      "耕地": 17008,
+      "▁glad": 17009,
+      "职业技术": 17010,
+      "魅力": 17011,
+      "unique": 17012,
+      "itled": 17013,
+      "igm": 17014,
+      "▁Forest": 17015,
+      "本院经审查": 17016,
+      "psy": 17017,
+      "▁elucid": 17018,
+      "▁momentum": 17019,
+      "▁dependencies": 17020,
+      "▁Les": 17021,
+      "条例": 17022,
+      "慢性": 17023,
+      "lean": 17024,
+      "nAlthough": 17025,
+      "▁wondering": 17026,
+      "常务": 17027,
+      "gal": 17028,
+      "urls": 17029,
+      "being": 17030,
+      "bow": 17031,
+      "▁generator": 17032,
+      "▁intelligence": 17033,
+      "是多少": 17034,
+      "▁recall": 17035,
+      "ETHODS": 17036,
+      "tau": 17037,
+      "▁cord": 17038,
+      "compile": 17039,
+      "国土资源": 17040,
+      "我和": 17041,
+      "但在": 17042,
+      "odd": 17043,
+      "热度": 17044,
+      "season": 17045,
+      "nOnce": 17046,
+      "▁Secretary": 17047,
+      "毫无": 17048,
+      "两组": 17049,
+      "icide": 17050,
+      "各级": 17051,
+      "选择的": 17052,
+      "▁myocard": 17053,
+      "▁substrate": 17054,
+      "对话": 17055,
+      "Rightarrow": 17056,
+      "ometer": 17057,
+      "▁/*": 17058,
+      "的历史": 17059,
+      "价比": 17060,
+      "▁soci": 17061,
+      "▁compatible": 17062,
+      "▁finger": 17063,
+      "iary": 17064,
+      "罪名": 17065,
+      "▁wel": 17066,
+      "earing": 17067,
+      "▁beneficial": 17068,
+      "▁Bridge": 17069,
+      "专注": 17070,
+      "暂无": 17071,
+      "之日止": 17072,
+      "▁hypertension": 17073,
+      "▁ell": 17074,
+      "%(": 17075,
+      "▁Guide": 17076,
+      "▁promotion": 17077,
+      "▁arms": 17078,
+      "nOff": 17079,
+      "jet": 17080,
+      "要么": 17081,
+      "批发": 17082,
+      "▁luck": 17083,
+      "较高": 17084,
+      "arse": 17085,
+      "▁expat": 17086,
+      "(\"\\\\": 17087,
+      "▁Beaut": 17088,
+      "农村信用合作": 17089,
+      "bial": 17090,
+      "▁passes": 17091,
+      "present": 17092,
+      "▁Major": 17093,
+      "糖尿": 17094,
+      "一线": 17095,
+      "朝鲜": 17096,
+      "▁comparing": 17097,
+      "▁Dam": 17098,
+      "▁dop": 17099,
+      "▁generic": 17100,
+      "老婆": 17101,
+      "▁adjacent": 17102,
+      "marks": 17103,
+      "▁opens": 17104,
+      "pected": 17105,
+      "▁phosphory": 17106,
+      "▁incredible": 17107,
+      "▁worst": 17108,
+      "了很多": 17109,
+      "cursor": 17110,
+      "注射": 17111,
+      "一开始": 17112,
+      "▁Argent": 17113,
+      "极速": 17114,
+      "▁texture": 17115,
+      "全新": 17116,
+      "wrapper": 17117,
+      "宗旨": 17118,
+      "子里": 17119,
+      "亿美元": 17120,
+      "orts": 17121,
+      "的使用": 17122,
+      "▁Similar": 17123,
+      "Mac": 17124,
+      "▁Imp": 17125,
+      "They": 17126,
+      "师范大学": 17127,
+      "圆形": 17128,
+      "不大": 17129,
+      "ships": 17130,
+      "nMost": 17131,
+      "imet": 17132,
+      "▁Oxford": 17133,
+      "ritis": 17134,
+      "▁Output": 17135,
+      "▁©": 17136,
+      "▁Ther": 17137,
+      "的取值范围": 17138,
+      "股权": 17139,
+      "▁tiny": 17140,
+      "▁assumed": 17141,
+      "toBe": 17142,
+      "▁Premier": 17143,
+      "▁Hence": 17144,
+      "▁generating": 17145,
+      "进展": 17146,
+      "▁simulations": 17147,
+      "号民事判决": 17148,
+      "常见的": 17149,
+      "▁mutation": 17150,
+      "▁communications": 17151,
+      "▁till": 17152,
+      "simple": 17153,
+      "▁shower": 17154,
+      "▁acceler": 17155,
+      "边形": 17156,
+      "▁attorney": 17157,
+      "▁prospective": 17158,
+      "心中": 17159,
+      "电流": 17160,
+      "精度": 17161,
+      "\\'),": 17162,
+      "蛋白质": 17163,
+      "▁asc": 17164,
+      "tein": 17165,
+      "PUT": 17166,
+      "▁spokes": 17167,
+      "基础上": 17168,
+      "▁correspond": 17169,
+      "▁macro": 17170,
+      "▁GitHub": 17171,
+      "unsigned": 17172,
+      "前期": 17173,
+      "pal": 17174,
+      "▁sculpt": 17175,
+      "Success": 17176,
+      "实体": 17177,
+      "dm": 17178,
+      "reader": 17179,
+      "▁cancel": 17180,
+      "demo": 17181,
+      "后续": 17182,
+      "▁approval": 17183,
+      "我们可以": 17184,
+      "enable": 17185,
+      "▁carrying": 17186,
+      "otton": 17187,
+      "的人生": 17188,
+      "▁cum": 17189,
+      "redit": 17190,
+      "诉至": 17191,
+      "▁kit": 17192,
+      "overset": 17193,
+      "▁Those": 17194,
+      "▁virt": 17195,
+      "应当依照": 17196,
+      "采访": 17197,
+      "Play": 17198,
+      "oker": 17199,
+      "积分": 17200,
+      "大幅": 17201,
+      "▁computational": 17202,
+      "纺织": 17203,
+      "而在": 17204,
+      "位的": 17205,
+      "▁Heart": 17206,
+      "门口": 17207,
+      "分支": 17208,
+      "▁toxic": 17209,
+      "aha": 17210,
+      "▁emphas": 17211,
+      "发现被执行人有可供执行财产": 17212,
+      "detail": 17213,
+      "▁altered": 17214,
+      "”）": 17215,
+      "vex": 17216,
+      "▁Keep": 17217,
+      "heng": 17218,
+      "▁Berlin": 17219,
+      "二是": 17220,
+      "▁ul": 17221,
+      "某个": 17222,
+      "ERT": 17223,
+      "▁talks": 17224,
+      "▁dangerous": 17225,
+      "▁reproduce": 17226,
+      "▁counsel": 17227,
+      "会长": 17228,
+      "▁blow": 17229,
+      "▁prominent": 17230,
+      "收录": 17231,
+      "▁crystal": 17232,
+      "▁circumstances": 17233,
+      "▁Turkey": 17234,
+      "opher": 17235,
+      "icking": 17236,
+      "▁Total": 17237,
+      "▁performances": 17238,
+      "▁cooking": 17239,
+      "▁*/": 17240,
+      "▁USB": 17241,
+      "▁Double": 17242,
+      "长江": 17243,
+      "省级": 17244,
+      "构造": 17245,
+      "出血": 17246,
+      "▁AI": 17247,
+      "tech": 17248,
+      "的位置": 17249,
+      "▁Official": 17250,
+      "nAnother": 17251,
+      "buf": 17252,
+      "▁rend": 17253,
+      "▁soph": 17254,
+      "精品": 17255,
+      "的最大": 17256,
+      "nMany": 17257,
+      "▁dish": 17258,
+      "▁unexpected": 17259,
+      "▁instances": 17260,
+      "▁systemic": 17261,
+      "▁microscopy": 17262,
+      "支出": 17263,
+      "▁winners": 17264,
+      "如何评价": 17265,
+      "▁hers": 17266,
+      "▁dot": 17267,
+      "▁apartment": 17268,
+      "▁concert": 17269,
+      "四日": 17270,
+      "▁Site": 17271,
+      "▁debate": 17272,
+      "缺席": 17273,
+      "▁Player": 17274,
+      "orient": 17275,
+      "▁Ni": 17276,
+      "气的": 17277,
+      "course": 17278,
+      "▁integrate": 17279,
+      "股东": 17280,
+      "making": 17281,
+      "▁MERCHANTABILITY": 17282,
+      "一人": 17283,
+      "▁solo": 17284,
+      "▁tf": 17285,
+      "orne": 17286,
+      "威胁": 17287,
+      "▁Indones": 17288,
+      "林业": 17289,
+      "会的": 17290,
+      "卫星": 17291,
+      "▁bands": 17292,
+      "Socket": 17293,
+      "▁Publish": 17294,
+      "▁justice": 17295,
+      "angel": 17296,
+      "▁scenario": 17297,
+      "▁defects": 17298,
+      "这时": 17299,
+      "并发": 17300,
+      "▁hire": 17301,
+      "▁manufacturer": 17302,
+      "▁hits": 17303,
+      "exper": 17304,
+      "ntyped": 17305,
+      "传票传唤": 17306,
+      "▁impressive": 17307,
+      "温馨": 17308,
+      "▁infants": 17309,
+      "灵魂": 17310,
+      "简直": 17311,
+      "▁apparently": 17312,
+      "不喜欢": 17313,
+      "们的": 17314,
+      "▁escape": 17315,
+      "总裁": 17316,
+      "日向本院提起": 17317,
+      "指定的期间履行给付金钱义务": 17318,
+      "Qt": 17319,
+      "Matrix": 17320,
+      "▁disappe": 17321,
+      "丛书": 17322,
+      "▁constantly": 17323,
+      "▁BY": 17324,
+      "暂时": 17325,
+      "第二百五十三条之规定": 17326,
+      "▁volunteers": 17327,
+      "肌肤": 17328,
+      "▁votes": 17329,
+      "urname": 17330,
+      "分米": 17331,
+      "▁admitted": 17332,
+      "礼物": 17333,
+      "equals": 17334,
+      "▁三": 17335,
+      "▁intra": 17336,
+      "oline": 17337,
+      "ambiguation": 17338,
+      "冰箱": 17339,
+      "▁drag": 17340,
+      "dependencies": 17341,
+      "▁potent": 17342,
+      "▁Son": 17343,
+      "tvoid": 17344,
+      "板块": 17345,
+      "扶贫": 17346,
+      "▁pd": 17347,
+      "CB": 17348,
+      "▁Bell": 17349,
+      "▁sick": 17350,
+      "不明": 17351,
+      "▁Fix": 17352,
+      "正文": 17353,
+      "dots": 17354,
+      "第二百五十七条第": 17355,
+      "▁Oregon": 17356,
+      "助于": 17357,
+      "ternational": 17358,
+      "▁anyway": 17359,
+      "MIN": 17360,
+      "igo": 17361,
+      "▁出生地": 17362,
+      "终端": 17363,
+      "▁hosts": 17364,
+      "itter": 17365,
+      "排列": 17366,
+      "独任审判": 17367,
+      "化简": 17368,
+      "▁Pa": 17369,
+      "毕业生": 17370,
+      "全力": 17371,
+      "▁narr": 17372,
+      "Screen": 17373,
+      "水利": 17374,
+      "▁fert": 17375,
+      "死了": 17376,
+      "▁onset": 17377,
+      "▁handler": 17378,
+      "▁Administration": 17379,
+      "日向本院申请": 17380,
+      "▁variance": 17381,
+      "▁newspaper": 17382,
+      "▁IMP": 17383,
+      "▁impair": 17384,
+      "riends": 17385,
+      "▁manually": 17386,
+      "财产保险": 17387,
+      "Tim": 17388,
+      "nEarly": 17389,
+      "▁computed": 17390,
+      "▁transferred": 17391,
+      "Iter": 17392,
+      "园林": 17393,
+      "民一": 17394,
+      "akh": 17395,
+      "▁talent": 17396,
+      "日的": 17397,
+      "里面的": 17398,
+      "\\\\()\\\\)": 17399,
+      "ophys": 17400,
+      "便是": 17401,
+      "yml": 17402,
+      "▁tur": 17403,
+      ")](": 17404,
+      "有趣": 17405,
+      "减肥": 17406,
+      "季节": 17407,
+      "减刑建议书": 17408,
+      "提交上诉状": 17409,
+      "triangle": 17410,
+      "温柔": 17411,
+      "清华": 17412,
+      "行业的": 17413,
+      "▁posit": 17414,
+      "ORK": 17415,
+      "▁PS": 17416,
+      "apol": 17417,
+      "▁Houston": 17418,
+      "▁gradient": 17419,
+      "绝对值": 17420,
+      "overline": 17421,
+      "▁graphics": 17422,
+      "摩托": 17423,
+      "▁engage": 17424,
+      "Directory": 17425,
+      "features": 17426,
+      "标题": 17427,
+      "nst": 17428,
+      "chant": 17429,
+      "Simple": 17430,
+      "asets": 17431,
+      "ateur": 17432,
+      "不符合": 17433,
+      "▁hur": 17434,
+      "▁stabil": 17435,
+      "检察员": 17436,
+      "▁Rose": 17437,
+      "▁targeting": 17438,
+      "正本一份": 17439,
+      "▁merged": 17440,
+      "字母": 17441,
+      "pid": 17442,
+      "alities": 17443,
+      "defined": 17444,
+      "но": 17445,
+      "千元": 17446,
+      "abetic": 17447,
+      "islative": 17448,
+      "▁luxury": 17449,
+      "▁Senior": 17450,
+      "▁viewed": 17451,
+      "ée": 17452,
+      "▁Si": 17453,
+      "英寸": 17454,
+      "ritish": 17455,
+      "能否": 17456,
+      "▁surprise": 17457,
+      "▁reset": 17458,
+      "etherlands": 17459,
+      "序列": 17460,
+      "▁workshop": 17461,
+      "体会": 17462,
+      "ceptor": 17463,
+      "内置": 17464,
+      "统计学": 17465,
+      "▁grab": 17466,
+      "nDr": 17467,
+      "了一下": 17468,
+      "TEST": 17469,
+      "▁overview": 17470,
+      "从此": 17471,
+      "Cal": 17472,
+      "▁hospitals": 17473,
+      "▁temporary": 17474,
+      "▁Ros": 17475,
+      "▁founder": 17476,
+      "▁jun": 17477,
+      "IGN": 17478,
+      "IAL": 17479,
+      "篮球": 17480,
+      "▁Did": 17481,
+      "ogs": 17482,
+      "▁目录": 17483,
+      "original": 17484,
+      "Framework": 17485,
+      "runtime": 17486,
+      "常常": 17487,
+      "timeout": 17488,
+      "nDescription": 17489,
+      "引擎": 17490,
+      "较好": 17491,
+      "▁Square": 17492,
+      "yll": 17493,
+      "Invalid": 17494,
+      "nAssociation": 17495,
+      "糖尿病": 17496,
+      "早上": 17497,
+      "家庄": 17498,
+      "牛奶": 17499,
+      "rock": 17500,
+      "nTrans": 17501,
+      "▁hip": 17502,
+      "刘某": 17503,
+      "预期": 17504,
+      "▁lake": 17505,
+      "农村信用合作联社": 17506,
+      "是为了": 17507,
+      "书面上诉的": 17508,
+      "izabeth": 17509,
+      "▁incorrect": 17510,
+      "▁portfolio": 17511,
+      "仅仅": 17512,
+      "司机": 17513,
+      "mc": 17514,
+      "刑诉": 17515,
+      "▁accomplish": 17516,
+      "WI": 17517,
+      "helper": 17518,
+      "▁Machine": 17519,
+      "▁sodium": 17520,
+      "counter": 17521,
+      "▁spac": 17522,
+      "▁slope": 17523,
+      "olving": 17524,
+      "also": 17525,
+      "ntypedef": 17526,
+      "▁biochemical": 17527,
+      "西部": 17528,
+      "年轻人": 17529,
+      "▁haz": 17530,
+      "▁append": 17531,
+      "VR": 17532,
+      "▁McC": 17533,
+      "▁powder": 17534,
+      "habilit": 17535,
+      "ро": 17536,
+      "anny": 17537,
+      "权限": 17538,
+      "▁Sab": 17539,
+      "密切": 17540,
+      "▁Rog": 17541,
+      "ittest": 17542,
+      "islation": 17543,
+      "nso": 17544,
+      "▁essay": 17545,
+      "Pool": 17546,
+      "uuid": 17547,
+      "动的": 17548,
+      "房价": 17549,
+      "▁mile": 17550,
+      "sens": 17551,
+      "请问": 17552,
+      "▁\\'__": 17553,
+      "\\\\}": 17554,
+      "▁Olympic": 17555,
+      "更高": 17556,
+      "MAT": 17557,
+      "Services": 17558,
+      "▁eligible": 17559,
+      "▁Theatre": 17560,
+      "▁municipality": 17561,
+      "例子": 17562,
+      "Media": 17563,
+      "▁jew": 17564,
+      "一流": 17565,
+      "未来的": 17566,
+      "▁cyst": 17567,
+      "Meta": 17568,
+      "▁chose": 17569,
+      "接收": 17570,
+      "nStep": 17571,
+      "▁surge": 17572,
+      "▁gel": 17573,
+      "▁}{": 17574,
+      "一辆": 17575,
+      "stdout": 17576,
+      "▁nomin": 17577,
+      "可在接到": 17578,
+      "受伤": 17579,
+      "▁Scottish": 17580,
+      ">\"": 17581,
+      "颗粒": 17582,
+      "▁ischem": 17583,
+      "▁stake": 17584,
+      "▁selective": 17585,
+      "活动的": 17586,
+      "怎样的": 17587,
+      "斗争": 17588,
+      "perm": 17589,
+      "aded": 17590,
+      "idine": 17591,
+      "▁Tok": 17592,
+      "▁capac": 17593,
+      "updated": 17594,
+      "▁variants": 17595,
+      "参观": 17596,
+      "响应": 17597,
+      "我很": 17598,
+      "▁counts": 17599,
+      "▁essentially": 17600,
+      "▁sad": 17601,
+      "}[": 17602,
+      "▁Never": 17603,
+      "县委": 17604,
+      "临时": 17605,
+      "▁Wi": 17606,
+      "日起十日内": 17607,
+      "▁fn": 17608,
+      "形状": 17609,
+      "▁第二": 17610,
+      "▁Ven": 17611,
+      "容器": 17612,
+      "▁separation": 17613,
+      "raq": 17614,
+      "▁dust": 17615,
+      "▁Policy": 17616,
+      "实时": 17617,
+      "频道": 17618,
+      "▁marker": 17619,
+      "Vol": 17620,
+      "lectric": 17621,
+      "▁GO": 17622,
+      "▁Ol": 17623,
+      "决赛": 17624,
+      "车型": 17625,
+      "':\\": 17626,
+      "▁ub": 17627,
+      "Parse": 17628,
+      "ikes": 17629,
+      "监察": 17630,
+      "抛物": 17631,
+      "宁市": 17632,
+      "▁doses": 17633,
+      "▁性别": 17634,
+      "但被执行人": 17635,
+      "▁undert": 17636,
+      "▁shoes": 17637,
+      "▁ultimate": 17638,
+      "▁sought": 17639,
+      "apps": 17640,
+      "▁invalid": 17641,
+      "加拿": 17642,
+      "▁Ford": 17643,
+      "签约": 17644,
+      "经济学": 17645,
+      "agnetic": 17646,
+      "Copyright": 17647,
+      "▁ME": 17648,
+      "▁vice": 17649,
+      "lined": 17650,
+      "nOlymp": 17651,
+      "▁suffering": 17652,
+      "▁Jean": 17653,
+      "斯坦": 17654,
+      "phia": 17655,
+      "ders": 17656,
+      "▁Gro": 17657,
+      "▁temporal": 17658,
+      "均为": 17659,
+      "ctic": 17660,
+      "▁resonance": 17661,
+      "通过对": 17662,
+      "▁interpretation": 17663,
+      "▁NA": 17664,
+      "Del": 17665,
+      "和计划生育": 17666,
+      "▁inches": 17667,
+      "Py": 17668,
+      "▁Instagram": 17669,
+      "▁breakfast": 17670,
+      "▁overl": 17671,
+      "睡觉": 17672,
+      "ados": 17673,
+      "▁Matt": 17674,
+      "▁naturally": 17675,
+      "ENV": 17676,
+      "夏天": 17677,
+      "stylesheet": 17678,
+      "的需求": 17679,
+      "▁fragment": 17680,
+      "▁magic": 17681,
+      "nEvery": 17682,
+      "Support": 17683,
+      "▁celebrate": 17684,
+      "UG": 17685,
+      "扫描": 17686,
+      "▁twenty": 17687,
+      "oprotein": 17688,
+      "▁epithelial": 17689,
+      "独特的": 17690,
+      "pend": 17691,
+      "了我": 17692,
+      "itudes": 17693,
+      "真的很": 17694,
+      "▁clone": 17695,
+      "▁gam": 17696,
+      "巨大的": 17697,
+      "▁Ident": 17698,
+      "▁lifetime": 17699,
+      "派出": 17700,
+      "bolic": 17701,
+      "closure": 17702,
+      "keywords": 17703,
+      "▁flav": 17704,
+      "▁cir": 17705,
+      "▁angles": 17706,
+      "▁Prom": 17707,
+      "阿里": 17708,
+      "imp": 17709,
+      "▁elderly": 17710,
+      ",'": 17711,
+      "jan": 17712,
+      "▁strengthen": 17713,
+      "▁injured": 17714,
+      "走进": 17715,
+      "标签": 17716,
+      "无效": 17717,
+      "在家": 17718,
+      "verbose": 17719,
+      "处置": 17720,
+      "结算": 17721,
+      "ocom": 17722,
+      "▁[];\\": 17723,
+      ":/": 17724,
+      "ership": 17725,
+      "VP": 17726,
+      "dispatch": 17727,
+      "▁arbitrary": 17728,
+      "nAp": 17729,
+      "fox": 17730,
+      "怀疑": 17731,
+      "拿到": 17732,
+      "▁heating": 17733,
+      "Boolean": 17734,
+      "lon": 17735,
+      "处理器": 17736,
+      "晋江文学网": 17737,
+      "▁printing": 17738,
+      "▁obviously": 17739,
+      "Router": 17740,
+      "slug": 17741,
+      "由题意": 17742,
+      "▁Executive": 17743,
+      "▁chocolate": 17744,
+      "▁Capt": 17745,
+      "建材": 17746,
+      "hello": 17747,
+      "▁Phase": 17748,
+      "▁Muslim": 17749,
+      "今后": 17750,
+      "Serializer": 17751,
+      "书画": 17752,
+      "报警": 17753,
+      "辈子": 17754,
+      "▁acknow": 17755,
+      "▁Fish": 17756,
+      "idays": 17757,
+      "▁birds": 17758,
+      "号起诉书": 17759,
+      "className": 17760,
+      "▁js": 17761,
+      "级别": 17762,
+      "itable": 17763,
+      "▁sorted": 17764,
+      "荣获": 17765,
+      "▁coupled": 17766,
+      "▁heads": 17767,
+      "现代化": 17768,
+      "不当": 17769,
+      "▁Rom": 17770,
+      "足以认定": 17771,
+      "监护": 17772,
+      "▁situated": 17773,
+      "arring": 17774,
+      "国家的": 17775,
+      "▁efficiently": 17776,
+      "▁absorption": 17777,
+      "▁headers": 17778,
+      "irk": 17779,
+      "▁Pers": 17780,
+      "的大小": 17781,
+      "Op": 17782,
+      "信心": 17783,
+      "待遇": 17784,
+      "decode": 17785,
+      "dataset": 17786,
+      "▁filters": 17787,
+      "olesterol": 17788,
+      "▁Resource": 17789,
+      "▁schema": 17790,
+      "cluster": 17791,
+      "▁ending": 17792,
+      "▁LLC": 17793,
+      "▁oscill": 17794,
+      "▁Lie": 17795,
+      "agma": 17796,
+      "mask": 17797,
+      "▁и": 17798,
+      "generate": 17799,
+      "enabled": 17800,
+      "传感": 17801,
+      "▁THIS": 17802,
+      "电信": 17803,
+      "皇帝": 17804,
+      "Expression": 17805,
+      ")**": 17806,
+      "▁rely": 17807,
+      "透明": 17808,
+      "▁Cambridge": 17809,
+      "▁bounded": 17810,
+      "▁asympt": 17811,
+      "party": 17812,
+      "CN": 17813,
+      "\"@": 17814,
+      "的定义": 17815,
+      "▁Tele": 17816,
+      "中华人民共和国行政诉讼法": 17817,
+      "留学": 17818,
+      "▁tom": 17819,
+      "story": 17820,
+      "Timeout": 17821,
+      "▁Singapore": 17822,
+      "▁Quant": 17823,
+      "▁recombin": 17824,
+      "这也是": 17825,
+      "abases": 17826,
+      "的学生": 17827,
+      "▁hat": 17828,
+      "本院予以支持": 17829,
+      "▁relief": 17830,
+      "讲话": 17831,
+      "figure": 17832,
+      "urrence": 17833,
+      "▁\"__": 17834,
+      "▁发": 17835,
+      "▁enthus": 17836,
+      "▁classified": 17837,
+      "Ret": 17838,
+      "adelphia": 17839,
+      "已发生法律效力": 17840,
+      "▁Supreme": 17841,
+      "npub": 17842,
+      "▁cond": 17843,
+      "▁decreasing": 17844,
+      "▁Property": 17845,
+      "▁WITH": 17846,
+      "要注意": 17847,
+      "coll": 17848,
+      "笔记本": 17849,
+      "性价比": 17850,
+      "▁Ale": 17851,
+      "相当于": 17852,
+      "江湖": 17853,
+      "▁delta": 17854,
+      "▁intermediate": 17855,
+      "▁Silver": 17856,
+      "▁illust": 17857,
+      "▁expressions": 17858,
+      "明显的": 17859,
+      "▁lessons": 17860,
+      "Tab": 17861,
+      "一块": 17862,
+      "▁officially": 17863,
+      "▁param": 17864,
+      "城县": 17865,
+      "inos": 17866,
+      "内涵": 17867,
+      "▁si": 17868,
+      "ulatory": 17869,
+      "iments": 17870,
+      "▁fusion": 17871,
+      "▁Hamilton": 17872,
+      "▁regulations": 17873,
+      "browser": 17874,
+      "▁hal": 17875,
+      "TM": 17876,
+      "ABC": 17877,
+      "▁nur": 17878,
+      "ORY": 17879,
+      "ixin": 17880,
+      "▁reaching": 17881,
+      "▁Consider": 17882,
+      "清理": 17883,
+      "▁explains": 17884,
+      "还不": 17885,
+      "BASE": 17886,
+      "usal": 17887,
+      "▁Soc": 17888,
+      "ensis": 17889,
+      "商场": 17890,
+      "的过程中": 17891,
+      "notation": 17892,
+      "▁artificial": 17893,
+      "ige": 17894,
+      "▁Arizona": 17895,
+      "Arr": 17896,
+      "viously": 17897,
+      "▁therapies": 17898,
+      "▁conclude": 17899,
+      "▁delight": 17900,
+      "Policy": 17901,
+      "耳机": 17902,
+      "▁▁▁▁▁▁▁▁▁": 17903,
+      "▁telling": 17904,
+      "▁manufacturers": 17905,
+      "Med": 17906,
+      "疫苗": 17907,
+      "Transform": 17908,
+      "▁savings": 17909,
+      "邮件": 17910,
+      "▁calendar": 17911,
+      "STRA": 17912,
+      "High": 17913,
+      "▁derivatives": 17914,
+      "▁vertex": 17915,
+      "▁instruments": 17916,
+      "不但": 17917,
+      "▁Multi": 17918,
+      "▁Ocean": 17919,
+      "▁giant": 17920,
+      "Now": 17921,
+      "nWork": 17922,
+      "auge": 17923,
+      "nJohn": 17924,
+      "▁Jewish": 17925,
+      "失信": 17926,
+      "▁Indiana": 17927,
+      "archive": 17928,
+      "avelength": 17929,
+      "组成员": 17930,
+      "▁endothelial": 17931,
+      "纪律": 17932,
+      "▁engaged": 17933,
+      "开关": 17934,
+      "associated": 17935,
+      "年在": 17936,
+      "▁Anton": 17937,
+      "十二条第二款": 17938,
+      "成交": 17939,
+      "*;\\": 17940,
+      "SELECT": 17941,
+      "QUE": 17942,
+      "VI": 17943,
+      "ía": 17944,
+      "cfg": 17945,
+      "头发": 17946,
+      "angers": 17947,
+      "▁respective": 17948,
+      "▁frag": 17949,
+      "Android": 17950,
+      "线性": 17951,
+      "三个月": 17952,
+      "ritical": 17953,
+      "▁CF": 17954,
+      "▁modes": 17955,
+      "销量": 17956,
+      "nz": 17957,
+      "密封": 17958,
+      "estab": 17959,
+      "validation": 17960,
+      "oustic": 17961,
+      "▁implies": 17962,
+      "▁champion": 17963,
+      "▁Transport": 17964,
+      "▁Smart": 17965,
+      "PAR": 17966,
+      "▁Questions": 17967,
+      "物资": 17968,
+      "nHis": 17969,
+      "▁Students": 17970,
+      "已有": 17971,
+      "\\'):\\": 17972,
+      "▁athletes": 17973,
+      "▁HR": 17974,
+      "▁mature": 17975,
+      "standard": 17976,
+      "终审": 17977,
+      "etary": 17978,
+      "pk": 17979,
+      "▁accompan": 17980,
+      "cv": 17981,
+      "回顾": 17982,
+      "▁arth": 17983,
+      "▁Final": 17984,
+      "Package": 17985,
+      "hythm": 17986,
+      "rieve": 17987,
+      "别名": 17988,
+      "health": 17989,
+      "square": 17990,
+      "这篇": 17991,
+      "▁pic": 17992,
+      "▁industries": 17993,
+      "终结执行": 17994,
+      "南方": 17995,
+      "▁photography": 17996,
+      "下一": 17997,
+      "Serv": 17998,
+      "ellect": 17999,
+      "fly": 18000,
+      "▁gang": 18001,
+      "▁stronger": 18002,
+      "圆柱": 18003,
+      "nMed": 18004,
+      "并从": 18005,
+      "▁wins": 18006,
+      "▁intracellular": 18007,
+      "中国人": 18008,
+      "cribed": 18009,
+      "▁apt": 18010,
+      "▁ownership": 18011,
+      "杭州市": 18012,
+      "disambiguation": 18013,
+      "Day": 18014,
+      "地区的": 18015,
+      "朝阳": 18016,
+      "向上": 18017,
+      "▁arrive": 18018,
+      "perature": 18019,
+      "▁支持": 18020,
+      "▁cheese": 18021,
+      "动脉": 18022,
+      "元及利息": 18023,
+      "▁Cop": 18024,
+      "江市": 18025,
+      "Domain": 18026,
+      "生理": 18027,
+      "treated": 18028,
+      "声明": 18029,
+      "▁Wat": 18030,
+      "在一个": 18031,
+      "▁Hash": 18032,
+      "概述": 18033,
+      "▁Boolean": 18034,
+      "eros": 18035,
+      "▁Farm": 18036,
+      "▁Could": 18037,
+      "iors": 18038,
+      "Bytes": 18039,
+      "rivial": 18040,
+      "问题和": 18041,
+      "▁gest": 18042,
+      "设计师": 18043,
+      "▁Database": 18044,
+      "过于": 18045,
+      "▁flows": 18046,
+      "民众": 18047,
+      "补贴": 18048,
+      "enz": 18049,
+      "nNational": 18050,
+      "实例": 18051,
+      "Editor": 18052,
+      "▁Natural": 18053,
+      "unes": 18054,
+      "▁concluded": 18055,
+      "Bean": 18056,
+      "▁resistant": 18057,
+      "Obj": 18058,
+      "▁docs": 18059,
+      "▁('": 18060,
+      "▁promoting": 18061,
+      "works": 18062,
+      "▁EN": 18063,
+      "▁embedded": 18064,
+      "▁discrete": 18065,
+      "网页": 18066,
+      "logical": 18067,
+      "▁displays": 18068,
+      "etr": 18069,
+      "▁LA": 18070,
+      "columns": 18071,
+      "Sort": 18072,
+      "辐射": 18073,
+      "porters": 18074,
+      "的世界": 18075,
+      "排除": 18076,
+      "________________": 18077,
+      "▁randomly": 18078,
+      "乘法": 18079,
+      "▁sky": 18080,
+      "SW": 18081,
+      "Operation": 18082,
+      "▁finance": 18083,
+      "▁degradation": 18084,
+      "参考文献": 18085,
+      "'],\\": 18086,
+      "▁anten": 18087,
+      "海南": 18088,
+      "▁wrt": 18089,
+      "inking": 18090,
+      "configuration": 18091,
+      "Definition": 18092,
+      "▁inequality": 18093,
+      "linked": 18094,
+      "▁google": 18095,
+      "▁partially": 18096,
+      "▁img": 18097,
+      "的知识": 18098,
+      "▁eggs": 18099,
+      "isations": 18100,
+      "orption": 18101,
+      "Depend": 18102,
+      "▁worker": 18103,
+      "izza": 18104,
+      "creat": 18105,
+      "nSent": 18106,
+      "害怕": 18107,
+      "plify": 18108,
+      "▁specificity": 18109,
+      "asive": 18110,
+      "oen": 18111,
+      "市长": 18112,
+      "▁hoping": 18113,
+      "▁peer": 18114,
+      "▁unusual": 18115,
+      "▁intervals": 18116,
+      "▁cancell": 18117,
+      "日上午": 18118,
+      "nArt": 18119,
+      "nEn": 18120,
+      "▁о": 18121,
+      "▁carb": 18122,
+      "▁swimming": 18123,
+      "peat": 18124,
+      "秩序": 18125,
+      "important": 18126,
+      "▁孙": 18127,
+      "作物": 18128,
+      "无锡": 18129,
+      "▁refresh": 18130,
+      "unter": 18131,
+      "▁disabled": 18132,
+      "▁developmental": 18133,
+      "工作中": 18134,
+      "ivariate": 18135,
+      "▁Writ": 18136,
+      "▁orientation": 18137,
+      "*)": 18138,
+      "▁egg": 18139,
+      "appa": 18140,
+      "▁methyl": 18141,
+      "▁favourite": 18142,
+      "新冠": 18143,
+      "▁verbose": 18144,
+      "Di": 18145,
+      "▁reputation": 18146,
+      "证言": 18147,
+      "提高了": 18148,
+      "彩票": 18149,
+      "attributes": 18150,
+      "Omega": 18151,
+      "副局长": 18152,
+      "撤回上诉": 18153,
+      "音频": 18154,
+      "▁antagon": 18155,
+      "attribute": 18156,
+      "equence": 18157,
+      "reation": 18158,
+      "ichael": 18159,
+      "nOlympic": 18160,
+      "▁adventure": 18161,
+      "拘役": 18162,
+      "国产": 18163,
+      "幸运": 18164,
+      "aba": 18165,
+      "▁empir": 18166,
+      "odge": 18167,
+      "▁rh": 18168,
+      "诈骗": 18169,
+      "▁Cub": 18170,
+      "见过": 18171,
+      "判决书的第二": 18172,
+      "第一百七": 18173,
+      "单独": 18174,
+      "因被执行人": 18175,
+      "▁reform": 18176,
+      "gres": 18177,
+      "百度百科": 18178,
+      "▁Brand": 18179,
+      "Handle": 18180,
+      "Google": 18181,
+      "别人的": 18182,
+      "▁出处": 18183,
+      "coord": 18184,
+      "spective": 18185,
+      "▁processed": 18186,
+      "优良": 18187,
+      "它是": 18188,
+      "各大": 18189,
+      ">;\\": 18190,
+      "▁systematic": 18191,
+      "能不能": 18192,
+      "将军": 18193,
+      "nUsing": 18194,
+      "Total": 18195,
+      "一系列": 18196,
+      "IMIT": 18197,
+      "▁Viet": 18198,
+      "construct": 18199,
+      "▁chemotherapy": 18200,
+      "ighth": 18201,
+      "▁seriously": 18202,
+      "编写": 18203,
+      "patient": 18204,
+      "▁RO": 18205,
+      "收购": 18206,
+      "▁prostate": 18207,
+      "适用法律": 18208,
+      "riculum": 18209,
+      "看的": 18210,
+      "美女": 18211,
+      "的专业": 18212,
+      "iversary": 18213,
+      "▁nob": 18214,
+      "监督管理局": 18215,
+      "wikipedia": 18216,
+      "▁Edward": 18217,
+      "Don": 18218,
+      "激励": 18219,
+      "我知道": 18220,
+      "看法": 18221,
+      "▁certified": 18222,
+      "necho": 18223,
+      "▁authority": 18224,
+      "▁Look": 18225,
+      "住安徽省": 18226,
+      "▁obesity": 18227,
+      "▁tens": 18228,
+      "背后": 18229,
+      "▁assays": 18230,
+      "复习": 18231,
+      "判令被告": 18232,
+      "本院经审查认为": 18233,
+      "锈钢": 18234,
+      "不久": 18235,
+      "履行义务": 18236,
+      "口味": 18237,
+      "清偿": 18238,
+      "built": 18239,
+      "nGiven": 18240,
+      "订单": 18241,
+      "Example": 18242,
+      "▁confident": 18243,
+      "▁listening": 18244,
+      "▁uptake": 18245,
+      "要素": 18246,
+      "direction": 18247,
+      "nAb": 18248,
+      "loss": 18249,
+      "是非常": 18250,
+      "▁Tag": 18251,
+      "▁Hung": 18252,
+      "▁Write": 18253,
+      "dh": 18254,
+      "睡眠": 18255,
+      "▁confused": 18256,
+      "YPE": 18257,
+      "▁nums": 18258,
+      "riority": 18259,
+      "细菌": 18260,
+      "▁medication": 18261,
+      "USE": 18262,
+      "▁\\'@": 18263,
+      "uclear": 18264,
+      "▁\\\\\"": 18265,
+      "▁rose": 18266,
+      "▁Sound": 18267,
+      "高质量": 18268,
+      "▁riv": 18269,
+      "'),": 18270,
+      "公开开庭审理了本案": 18271,
+      "▁Davis": 18272,
+      "gly": 18273,
+      "ocyt": 18274,
+      "▁promoted": 18275,
+      "erd": 18276,
+      "▁CPU": 18277,
+      "洗衣": 18278,
+      "▁occas": 18279,
+      "wal": 18280,
+      "▁spectral": 18281,
+      "▁公司类型": 18282,
+      "▁span": 18283,
+      "▁Cy": 18284,
+      "癫痫": 18285,
+      "plotlib": 18286,
+      "新增": 18287,
+      "anga": 18288,
+      "▁SA": 18289,
+      "▁Athlet": 18290,
+      "disc": 18291,
+      "▁eastern": 18292,
+      "▁joy": 18293,
+      "aza": 18294,
+      "Loc": 18295,
+      "▁Quality": 18296,
+      "Cr": 18297,
+      "pag": 18298,
+      "▁posting": 18299,
+      "工会": 18300,
+      "▁pour": 18301,
+      "尼亚": 18302,
+      "Changed": 18303,
+      "imen": 18304,
+      "引领": 18305,
+      "▁_{": 18306,
+      "wich": 18307,
+      "因涉嫌": 18308,
+      "▁surprising": 18309,
+      "交付执行": 18310,
+      "初步": 18311,
+      "买卖合同纠纷一案": 18312,
+      "试行": 18313,
+      "▁拼音": 18314,
+      "▁powers": 18315,
+      "▁Ra": 18316,
+      "▁toxicity": 18317,
+      "被评为": 18318,
+      "▁electricity": 18319,
+      "▁likes": 18320,
+      "uto": 18321,
+      "▁capability": 18322,
+      "这是一个": 18323,
+      "▁\\\\{": 18324,
+      "项目的": 18325,
+      "YS": 18326,
+      "learn": 18327,
+      "▁preced": 18328,
+      "emony": 18329,
+      "人文": 18330,
+      "▁influenced": 18331,
+      "▁cute": 18332,
+      "▁Channel": 18333,
+      "昆明": 18334,
+      "快递": 18335,
+      "mx": 18336,
+      "期刊": 18337,
+      "ryption": 18338,
+      "Created": 18339,
+      "质证": 18340,
+      "▁pep": 18341,
+      "available": 18342,
+      "搅拌": 18343,
+      "▁personality": 18344,
+      "isely": 18345,
+      "▁graphs": 18346,
+      "orable": 18347,
+      "nbut": 18348,
+      "世界的": 18349,
+      "凤凰": 18350,
+      "▁Remove": 18351,
+      "erty": 18352,
+      "haust": 18353,
+      "所谓的": 18354,
+      "▁extraction": 18355,
+      "mysql": 18356,
+      "▁fighting": 18357,
+      "没想到": 18358,
+      "eman": 18359,
+      "letal": 18360,
+      "▁gear": 18361,
+      "电视台": 18362,
+      "▁Calcul": 18363,
+      "报送本院审理": 18364,
+      "▁dos": 18365,
+      "▁arc": 18366,
+      "Global": 18367,
+      "值为": 18368,
+      "宣告": 18369,
+      "编剧": 18370,
+      "一大": 18371,
+      "西班": 18372,
+      "▁tract": 18373,
+      "▁用": 18374,
+      "▁bid": 18375,
+      "▁exhibition": 18376,
+      "analysis": 18377,
+      "回应": 18378,
+      "▁clinic": 18379,
+      "▁deck": 18380,
+      "SIZE": 18381,
+      "树立": 18382,
+      "good": 18383,
+      "▁bowl": 18384,
+      "刑执字第": 18385,
+      "sit": 18386,
+      "TIME": 18387,
+      "话题": 18388,
+      "▁settlement": 18389,
+      "▁legislation": 18390,
+      "▁Governor": 18391,
+      "一共": 18392,
+      "依靠": 18393,
+      "▁caps": 18394,
+      "▁automated": 18395,
+      "完了": 18396,
+      "▁streets": 18397,
+      "▁\\'\\\\": 18398,
+      "▁matched": 18399,
+      "信息的": 18400,
+      "▁Golden": 18401,
+      "ivation": 18402,
+      "}|": 18403,
+      "ulo": 18404,
+      "▁})\\": 18405,
+      "<%": 18406,
+      "GER": 18407,
+      "phase": 18408,
+      "▁integrity": 18409,
+      "▁synchron": 18410,
+      "seed": 18411,
+      "▁Following": 18412,
+      "▁Cell": 18413,
+      "▁reconstruction": 18414,
+      "深化": 18415,
+      "trace": 18416,
+      "▁engineer": 18417,
+      "▁careful": 18418,
+      "▁emission": 18419,
+      "显然": 18420,
+      "▁loves": 18421,
+      "保养": 18422,
+      "西班牙": 18423,
+      "mol": 18424,
+      "apters": 18425,
+      "▁Greg": 18426,
+      "进度": 18427,
+      "ни": 18428,
+      "parts": 18429,
+      "CLUD": 18430,
+      "publish": 18431,
+      "▁gym": 18432,
+      "▁rod": 18433,
+      "普及": 18434,
+      "icit": 18435,
+      "▁Netherlands": 18436,
+      "▁disappoint": 18437,
+      "领取": 18438,
+      "▁alle": 18439,
+      "disable": 18440,
+      "病例": 18441,
+      "关联": 18442,
+      "Byte": 18443,
+      "宗教": 18444,
+      "▁distributions": 18445,
+      "▁induce": 18446,
+      "nMar": 18447,
+      "▁herself": 18448,
+      "理工": 18449,
+      "边缘": 18450,
+      "通用": 18451,
+      "▁tolerance": 18452,
+      "▁Agency": 18453,
+      "▁approximation": 18454,
+      "Space": 18455,
+      "ulas": 18456,
+      "▁suffered": 18457,
+      "判决书的第二日起十日内": 18458,
+      "nFilms": 18459,
+      "▁bundle": 18460,
+      "quis": 18461,
+      "Inv": 18462,
+      "▁ships": 18463,
+      "Ang": 18464,
+      "▁protective": 18465,
+      "ardens": 18466,
+      "商贸": 18467,
+      "▁fishing": 18468,
+      "▁Ban": 18469,
+      "▁Ped": 18470,
+      "中医药": 18471,
+      "Cloud": 18472,
+      "消毒": 18473,
+      "短期": 18474,
+      "大街": 18475,
+      "急性": 18476,
+      "疯狂": 18477,
+      "▁CS": 18478,
+      "──": 18479,
+      "anchester": 18480,
+      "▁DVD": 18481,
+      "▁exhibited": 18482,
+      "▁lets": 18483,
+      "BT": 18484,
+      "your": 18485,
+      "ко": 18486,
+      "▁stunning": 18487,
+      "▁rental": 18488,
+      "代表作品": 18489,
+      "十四条": 18490,
+      "▁mediated": 18491,
+      "▁segments": 18492,
+      "抓住": 18493,
+      "prod": 18494,
+      "▁amplit": 18495,
+      "师生": 18496,
+      "▁pandemic": 18497,
+      "一件": 18498,
+      "表情": 18499,
+      "听说": 18500,
+      "▁legs": 18501,
+      "▁considerable": 18502,
+      "▁rice": 18503,
+      "▁namely": 18504,
+      "▁satellite": 18505,
+      "Cam": 18506,
+      "▁flux": 18507,
+      "▁weights": 18508,
+      "commands": 18509,
+      "notes": 18510,
+      "心脏": 18511,
+      "日下午": 18512,
+      "文化的": 18513,
+      "Enter": 18514,
+      "其余": 18515,
+      "▁wer": 18516,
+      "stan": 18517,
+      "大了": 18518,
+      "▁racing": 18519,
+      "▁chest": 18520,
+      "Float": 18521,
+      "▁mistake": 18522,
+      "▁emphasis": 18523,
+      "▁regulated": 18524,
+      "▁cmd": 18525,
+      "posts": 18526,
+      "▁slowly": 18527,
+      "▁sensors": 18528,
+      "日成立": 18529,
+      "原始": 18530,
+      "wegian": 18531,
+      "▁contributed": 18532,
+      "ourier": 18533,
+      "号起诉书指控被告人": 18534,
+      "的管理": 18535,
+      "▁Pack": 18536,
+      "▁Iowa": 18537,
+      "▁secretion": 18538,
+      "mits": 18539,
+      "▁psychological": 18540,
+      "之处": 18541,
+      "的手": 18542,
+      "挖掘": 18543,
+      "RED": 18544,
+      "Opt": 18545,
+      "了个": 18546,
+      "作品简介": 18547,
+      "(\"#": 18548,
+      "▁Sweden": 18549,
+      "▁Eval": 18550,
+      "▁dependence": 18551,
+      "tl": 18552,
+      "▁Kal": 18553,
+      ")}^{": 18554,
+      "ceive": 18555,
+      "▁Article": 18556,
+      "plementation": 18557,
+      "shift": 18558,
+      "▁Tra": 18559,
+      "▁servers": 18560,
+      "丈夫": 18561,
+      "▁DI": 18562,
+      "▁Guard": 18563,
+      "理财": 18564,
+      "cup": 18565,
+      "纪录": 18566,
+      "Panel": 18567,
+      "▁predom": 18568,
+      "nPar": 18569,
+      "▁OUT": 18570,
+      "队员": 18571,
+      "常委": 18572,
+      "▁contributing": 18573,
+      "相结合": 18574,
+      "around": 18575,
+      "nBackground": 18576,
+      "▁utilize": 18577,
+      "岁月": 18578,
+      "avirus": 18579,
+      "typeof": 18580,
+      "是很": 18581,
+      "设计的": 18582,
+      "western": 18583,
+      "yy": 18584,
+      "苏联": 18585,
+      "RES": 18586,
+      "付出": 18587,
+      "ancies": 18588,
+      "和我": 18589,
+      "/)\\": 18590,
+      "带动": 18591,
+      "loading": 18592,
+      "orgeous": 18593,
+      "inch": 18594,
+      "▁snap": 18595,
+      "oni": 18596,
+      "有可能": 18597,
+      "▁Student": 18598,
+      "▁ocean": 18599,
+      "中国农业": 18600,
+      "模板": 18601,
+      "▁musicians": 18602,
+      "并对": 18603,
+      "▁macroph": 18604,
+      "代谢": 18605,
+      "▁Modern": 18606,
+      "无需": 18607,
+      "rient": 18608,
+      "▁protocols": 18609,
+      "▁conservation": 18610,
+      "▁probe": 18611,
+      "▁Queens": 18612,
+      "nSim": 18613,
+      "intage": 18614,
+      "▁statistically": 18615,
+      "ployee": 18616,
+      "▁scoring": 18617,
+      "▁Unfortunately": 18618,
+      "▁Buy": 18619,
+      "nat": 18620,
+      "新能源": 18621,
+      "sheet": 18622,
+      "录取": 18623,
+      "hook": 18624,
+      "▁surf": 18625,
+      "phere": 18626,
+      "Db": 18627,
+      "hma": 18628,
+      "千年": 18629,
+      "看出": 18630,
+      "▁overcome": 18631,
+      "▁restricted": 18632,
+      "nWhere": 18633,
+      "几十": 18634,
+      "财产保险股份有限公司": 18635,
+      "我们在": 18636,
+      "▁Corp": 18637,
+      "刑期自": 18638,
+      "诉讼权利": 18639,
+      "Report": 18640,
+      "▁varying": 18641,
+      "▁Share": 18642,
+      "▁crowd": 18643,
+      "端口": 18644,
+      "▁copper": 18645,
+      "Pub": 18646,
+      "豆腐": 18647,
+      "▁appeal": 18648,
+      "想着": 18649,
+      "▁losses": 18650,
+      "符号": 18651,
+      "▁doctors": 18652,
+      "▁computation": 18653,
+      "▁Gh": 18654,
+      "omat": 18655,
+      "ortion": 18656,
+      "achelor": 18657,
+      "▁Nig": 18658,
+      "usc": 18659,
+      "认真遵守": 18660,
+      "▁retirement": 18661,
+      "bes": 18662,
+      "Layer": 18663,
+      "▁utilized": 18664,
+      "enant": 18665,
+      "▁squad": 18666,
+      "redirect": 18667,
+      "对本案": 18668,
+      "Middleware": 18669,
+      "词汇": 18670,
+      "jupyter": 18671,
+      "branch": 18672,
+      "▁cv": 18673,
+      "已构成": 18674,
+      "▁\"@": 18675,
+      "公安机关": 18676,
+      "内外": 18677,
+      "aptic": 18678,
+      "▁victim": 18679,
+      "lyn": 18680,
+      "爱国": 18681,
+      "convert": 18682,
+      "▁specialist": 18683,
+      "▁Basic": 18684,
+      "恐怖": 18685,
+      "▁Gra": 18686,
+      "▁Thread": 18687,
+      "▁Malays": 18688,
+      "激发": 18689,
+      "法律服务": 18690,
+      "▁movements": 18691,
+      "UTF": 18692,
+      "合理的": 18693,
+      "▁mutant": 18694,
+      "▁paras": 18695,
+      "▁Algebra": 18696,
+      "▁Islands": 18697,
+      "全程": 18698,
+      "×（": 18699,
+      "ingham": 18700,
+      "Price": 18701,
+      "driv": 18702,
+      "LINE": 18703,
+      "tcase": 18704,
+      "▁declared": 18705,
+      "前景": 18706,
+      "女士": 18707,
+      "方程为": 18708,
+      "▁certificate": 18709,
+      "▁Random": 18710,
+      "▁gate": 18711,
+      "history": 18712,
+      "FM": 18713,
+      "第三章": 18714,
+      "▁tooth": 18715,
+      "memory": 18716,
+      "▁bonus": 18717,
+      "▁Swedish": 18718,
+      "▁Garden": 18719,
+      "已经发生法律效力的": 18720,
+      "ILL": 18721,
+      "otrop": 18722,
+      "▁marine": 18723,
+      "rend": 18724,
+      "▁demo": 18725,
+      "再次申请执行": 18726,
+      "nGo": 18727,
+      "澳门": 18728,
+      "▁trim": 18729,
+      "aping": 18730,
+      "▁thereby": 18731,
+      "▁victims": 18732,
+      "Returns": 18733,
+      "ifact": 18734,
+      "▁faced": 18735,
+      "对你": 18736,
+      "aptop": 18737,
+      "编程": 18738,
+      ">)": 18739,
+      "什么的": 18740,
+      "▁Empire": 18741,
+      "▁entering": 18742,
+      "分泌": 18743,
+      "▁redistribute": 18744,
+      "Dem": 18745,
+      "▁difficulty": 18746,
+      "▁CL": 18747,
+      "▁Age": 18748,
+      "这就": 18749,
+      "▁Nick": 18750,
+      "可能性": 18751,
+      "▁garage": 18752,
+      "pad": 18753,
+      "先行": 18754,
+      "AGS": 18755,
+      "有两个": 18756,
+      "pic": 18757,
+      "tz": 18758,
+      "▁quot": 18759,
+      "distance": 18760,
+      "▁pricing": 18761,
+      "心态": 18762,
+      "▁publications": 18763,
+      "▁flexibility": 18764,
+      "一百": 18765,
+      "▁pleased": 18766,
+      "▁transactions": 18767,
+      "volume": 18768,
+      "nFurther": 18769,
+      "气体": 18770,
+      "优秀的": 18771,
+      "▁anymore": 18772,
+      "▁Il": 18773,
+      "▁\"\"\"\\": 18774,
+      "pot": 18775,
+      "一月": 18776,
+      "▁BBC": 18777,
+      "▁bias": 18778,
+      "\\\\;": 18779,
+      "pair": 18780,
+      "▁dominant": 18781,
+      "Has": 18782,
+      "内核": 18783,
+      "▁visits": 18784,
+      "/#": 18785,
+      "asant": 18786,
+      "stock": 18787,
+      "dfs": 18788,
+      "▁ensuring": 18789,
+      "nCalculate": 18790,
+      "公务员": 18791,
+      "需要的": 18792,
+      "Creat": 18793,
+      "ytics": 18794,
+      "friendly": 18795,
+      ":%": 18796,
+      "sets": 18797,
+      ";//": 18798,
+      "Fragment": 18799,
+      "middleware": 18800,
+      "▁leather": 18801,
+      "sb": 18802,
+      "attice": 18803,
+      "ка": 18804,
+      "nTrack": 18805,
+      "inar": 18806,
+      "▁affecting": 18807,
+      "▁angular": 18808,
+      "回去": 18809,
+      "引入": 18810,
+      "▁provision": 18811,
+      "([\\": 18812,
+      "哪怕": 18813,
+      "高达": 18814,
+      "分割": 18815,
+      "ench": 18816,
+      "▁Single": 18817,
+      "▁LaTeX": 18818,
+      "▁fewer": 18819,
+      "jection": 18820,
+      "rus": 18821,
+      "头的": 18822,
+      "浏览器": 18823,
+      "▁blind": 18824,
+      "stdio": 18825,
+      "▁suite": 18826,
+      "▁skip": 18827,
+      "▁Fall": 18828,
+      "教室": 18829,
+      "▁falls": 18830,
+      "是怎么": 18831,
+      "blast": 18832,
+      "激烈": 18833,
+      "ARCH": 18834,
+      "撤回对被告": 18835,
+      "▁metrics": 18836,
+      "acht": 18837,
+      "作为一个": 18838,
+      "▁oxide": 18839,
+      "▁folks": 18840,
+      "▁dies": 18841,
+      "▁acceleration": 18842,
+      "售后": 18843,
+      "INK": 18844,
+      "flat": 18845,
+      "Metadata": 18846,
+      "变更": 18847,
+      "▁bulk": 18848,
+      "▁specimens": 18849,
+      "▁physician": 18850,
+      "▁Lind": 18851,
+      "出现在": 18852,
+      "提交上诉状正本一份": 18853,
+      "▁captured": 18854,
+      "▁deleted": 18855,
+      "气象": 18856,
+      "opathy": 18857,
+      "▁Record": 18858,
+      "ер": 18859,
+      "ols": 18860,
+      "对待": 18861,
+      "GM": 18862,
+      "▁Sav": 18863,
+      "lear": 18864,
+      "▁decay": 18865,
+      "▁\\'\\'\\": 18866,
+      "▁communicate": 18867,
+      "ceptions": 18868,
+      "▁cake": 18869,
+      "▁bare": 18870,
+      "opts": 18871,
+      "▁Brad": 18872,
+      "▁IR": 18873,
+      "nPart": 18874,
+      "▁syntax": 18875,
+      "▁removing": 18876,
+      "浪漫": 18877,
+      "中国共产党": 18878,
+      "rowth": 18879,
+      "forEach": 18880,
+      "外出": 18881,
+      "▁Several": 18882,
+      "操作系统": 18883,
+      "Dto": 18884,
+      "Var": 18885,
+      "vard": 18886,
+      "选用": 18887,
+      "GRO": 18888,
+      "因式": 18889,
+      "力和": 18890,
+      "uum": 18891,
+      "▁Missouri": 18892,
+      "▁medalists": 18893,
+      "▁payload": 18894,
+      "坐在": 18895,
+      "▁investigations": 18896,
+      ">&": 18897,
+      "分辨率": 18898,
+      "yal": 18899,
+      "类型的": 18900,
+      "▁operational": 18901,
+      "connected": 18902,
+      "焦点": 18903,
+      "▁exponential": 18904,
+      "▁Ta": 18905,
+      "iral": 18906,
+      "▁restriction": 18907,
+      "iO": 18908,
+      "izon": 18909,
+      "graduate": 18910,
+      "奶奶": 18911,
+      "sym": 18912,
+      "▁PropTypes": 18913,
+      "Dat": 18914,
+      "▁fra": 18915,
+      "▁inclusion": 18916,
+      "安全生产": 18917,
+      "▁朱": 18918,
+      "▁spark": 18919,
+      "▁Cart": 18920,
+      "prototype": 18921,
+      "▁dipl": 18922,
+      "scopic": 18923,
+      "nQuestion": 18924,
+      "▁pathogenesis": 18925,
+      "▁林": 18926,
+      "▁descent": 18927,
+      "▁swit": 18928,
+      "出门": 18929,
+      "lux": 18930,
+      "quer": 18931,
+      "▁Volume": 18932,
+      "疫情防控": 18933,
+      "Save": 18934,
+      "不含": 18935,
+      "▁tradition": 18936,
+      "权威": 18937,
+      "▁Item": 18938,
+      "▁Disney": 18939,
+      "我现在": 18940,
+      "▁Corn": 18941,
+      "nbool": 18942,
+      "管理的": 18943,
+      "贵的": 18944,
+      "▁revolution": 18945,
+      "▁vo": 18946,
+      "▁PL": 18947,
+      "▁个": 18948,
+      "▁controlling": 18949,
+      "aco": 18950,
+      "hd": 18951,
+      "mers": 18952,
+      "arma": 18953,
+      "▁spinal": 18954,
+      "数为": 18955,
+      "Tasks": 18956,
+      "Sw": 18957,
+      "▁Vo": 18958,
+      "shops": 18959,
+      "▁applies": 18960,
+      "▁Prince": 18961,
+      "▁Moh": 18962,
+      "erving": 18963,
+      "不锈钢": 18964,
+      "字符串": 18965,
+      "Draw": 18966,
+      "inic": 18967,
+      "市公安局": 18968,
+      "职能": 18969,
+      "病情": 18970,
+      "MW": 18971,
+      "▁aug": 18972,
+      "there": 18973,
+      "▁participating": 18974,
+      "▁regulate": 18975,
+      "分别是": 18976,
+      "Kit": 18977,
+      "附带": 18978,
+      "abc": 18979,
+      "dump": 18980,
+      "nUniversity": 18981,
+      "▁phenomenon": 18982,
+      "USB": 18983,
+      "一样的": 18984,
+      "\"],\\": 18985,
+      "kish": 18986,
+      "▁polymorph": 18987,
+      "信息均": 18988,
+      "outing": 18989,
+      "nec": 18990,
+      "▁forgot": 18991,
+      "▁symbols": 18992,
+      "南昌": 18993,
+      "backend": 18994,
+      "▁utilizing": 18995,
+      "的三": 18996,
+      "指派": 18997,
+      "▁rugby": 18998,
+      "志愿者": 18999,
+      "房型": 19000,
+      "日电": 19001,
+      "▁puts": 19002,
+      "▁bottle": 19003,
+      "ntr": 19004,
+      "Full": 19005,
+      "轿车": 19006,
+      "▁Active": 19007,
+      "▁independently": 19008,
+      "商初字第": 19009,
+      "▁Wars": 19010,
+      "▁disable": 19011,
+      "一批": 19012,
+      "▁Move": 19013,
+      "▁Person": 19014,
+      "妹妹": 19015,
+      "▁Due": 19016,
+      "学校的": 19017,
+      "▁ultr": 19018,
+      "不清": 19019,
+      "▁scenes": 19020,
+      "▁timing": 19021,
+      "▁PHP": 19022,
+      "授予": 19023,
+      "▁sight": 19024,
+      "▁Bath": 19025,
+      "只会": 19026,
+      "▁gastr": 19027,
+      "prove": 19028,
+      "▁Iraq": 19029,
+      "▁tweet": 19030,
+      "▁currency": 19031,
+      "cu": 19032,
+      "▁MT": 19033,
+      "因而": 19034,
+      "▁trib": 19035,
+      "▁barrier": 19036,
+      "成为了": 19037,
+      "▁obtaining": 19038,
+      "▁Tal": 19039,
+      "Font": 19040,
+      "日益": 19041,
+      "▁Railway": 19042,
+      "爆炸": 19043,
+      "Public": 19044,
+      "洗净": 19045,
+      "outheast": 19046,
+      "▁duty": 19047,
+      "hover": 19048,
+      "▁alumin": 19049,
+      "绘画": 19050,
+      "采集": 19051,
+      "▁transcript": 19052,
+      "nals": 19053,
+      "▁\\'\\'": 19054,
+      "cycle": 19055,
+      "公诉机关指控": 19056,
+      "▁circular": 19057,
+      "▁transf": 19058,
+      "▁progressive": 19059,
+      "▁clothing": 19060,
+      "出行": 19061,
+      "当初": 19062,
+      "终身": 19063,
+      "▁themes": 19064,
+      "nOh": 19065,
+      "▁cookies": 19066,
+      "不易": 19067,
+      "人和": 19068,
+      "本院依据": 19069,
+      "合计": 19070,
+      "▁Pu": 19071,
+      "Power": 19072,
+      "burn": 19073,
+      "分手": 19074,
+      "姓名": 19075,
+      "▁meets": 19076,
+      "文本": 19077,
+      "▁cerebral": 19078,
+      "tok": 19079,
+      "万里": 19080,
+      "▁Tel": 19081,
+      "malink": 19082,
+      "▁associations": 19083,
+      "▁bars": 19084,
+      "控股": 19085,
+      "春节": 19086,
+      "▁Hay": 19087,
+      "▁satisfy": 19088,
+      "这么多": 19089,
+      "dn": 19090,
+      "引自": 19091,
+      "▁Associ": 19092,
+      "esian": 19093,
+      "▁demands": 19094,
+      "科大": 19095,
+      "▁RT": 19096,
+      "iper": 19097,
+      "法庭": 19098,
+      "▁tel": 19099,
+      "▁varied": 19100,
+      "▁alk": 19101,
+      "Random": 19102,
+      "pb": 19103,
+      "▁frames": 19104,
+      "占地面积": 19105,
+      "加拿大": 19106,
+      "中文化": 19107,
+      "但我": 19108,
+      "▁twe": 19109,
+      "zi": 19110,
+      "▁constants": 19111,
+      "▁FL": 19112,
+      "牛肉": 19113,
+      "▁Kansas": 19114,
+      "▁Pi": 19115,
+      "▁legend": 19116,
+      "illance": 19117,
+      "职称": 19118,
+      "nauthor": 19119,
+      "▁pregnant": 19120,
+      "▁partition": 19121,
+      "▁accumulation": 19122,
+      "▁sauce": 19123,
+      "▁vacation": 19124,
+      "ISO": 19125,
+      "atted": 19126,
+      "▁gamma": 19127,
+      "▁breaks": 19128,
+      "▁Front": 19129,
+      "▁navigation": 19130,
+      "化为": 19131,
+      "▁grap": 19132,
+      "}{{": 19133,
+      "▁导演": 19134,
+      ")):\\": 19135,
+      "▁retriev": 19136,
+      "▁facts": 19137,
+      "▁pink": 19138,
+      "}\"\\": 19139,
+      "一段时间": 19140,
+      "nom": 19141,
+      "可以通过": 19142,
+      "accept": 19143,
+      "olor": 19144,
+      "▁Kat": 19145,
+      "更好地": 19146,
+      "dates": 19147,
+      "▁approx": 19148,
+      "Feature": 19149,
+      "▁convergence": 19150,
+      "panel": 19151,
+      "▁conform": 19152,
+      "▁documented": 19153,
+      "olute": 19154,
+      "感动": 19155,
+      "Modal": 19156,
+      "▁Wilson": 19157,
+      "^(": 19158,
+      "设有": 19159,
+      "▁sync": 19160,
+      "▁highlights": 19161,
+      "▁Financial": 19162,
+      "▁trou": 19163,
+      "▁pipeline": 19164,
+      "itcoin": 19165,
+      "▁desktop": 19166,
+      "emat": 19167,
+      "轨道": 19168,
+      "opter": 19169,
+      "ignore": 19170,
+      "的机会": 19171,
+      "看过": 19172,
+      "Conf": 19173,
+      "▁gent": 19174,
+      "询问": 19175,
+      "▁drain": 19176,
+      "参加了": 19177,
+      "▁depart": 19178,
+      "旗下": 19179,
+      "▁Future": 19180,
+      "examples": 19181,
+      "地板": 19182,
+      "本题考查了": 19183,
+      "办事处": 19184,
+      "▁translate": 19185,
+      "habilitation": 19186,
+      "ixels": 19187,
+      "itting": 19188,
+      "FORM": 19189,
+      "生效之日起": 19190,
+      "魔法": 19191,
+      "▁releases": 19192,
+      "▁Hu": 19193,
+      "大数据": 19194,
+      "▁Cook": 19195,
+      "▁anch": 19196,
+      "▁illegal": 19197,
+      "▁combinations": 19198,
+      "▁ALL": 19199,
+      "entral": 19200,
+      "▁elastic": 19201,
+      "▁Sn": 19202,
+      "▁pel": 19203,
+      "olitan": 19204,
+      "▁personally": 19205,
+      "▁vitamin": 19206,
+      "▁charts": 19207,
+      "到位": 19208,
+      "COL": 19209,
+      "▁hasn": 19210,
+      "市政府": 19211,
+      "LOW": 19212,
+      "口腔": 19213,
+      "▁餐馆": 19214,
+      "发明": 19215,
+      "▁EP": 19216,
+      "tg": 19217,
+      "▁clothes": 19218,
+      "▁tast": 19219,
+      "▁executed": 19220,
+      "▁CC": 19221,
+      "男朋友": 19222,
+      "防水": 19223,
+      "批评": 19224,
+      "凭借": 19225,
+      "▁vertices": 19226,
+      "*\\\\": 19227,
+      "▁mood": 19228,
+      "▁enterprise": 19229,
+      "对自己": 19230,
+      "上是": 19231,
+      "sed": 19232,
+      "▁adaptive": 19233,
+      "重的": 19234,
+      "▁PT": 19235,
+      "诞生": 19236,
+      "▁→": 19237,
+      "Suppose": 19238,
+      "netes": 19239,
+      "▁editing": 19240,
+      "▁Again": 19241,
+      "▁CP": 19242,
+      "itudinal": 19243,
+      "rypto": 19244,
+      "▁waters": 19245,
+      "▁hang": 19246,
+      "培育": 19247,
+      "放心": 19248,
+      "▁Eric": 19249,
+      "ennessee": 19250,
+      "ILD": 19251,
+      "ials": 19252,
+      "▁xml": 19253,
+      "二首": 19254,
+      "ingly": 19255,
+      "__\\':\\": 19256,
+      "▁resolved": 19257,
+      "排序": 19258,
+      "optional": 19259,
+      "})^{": 19260,
+      "urches": 19261,
+      "fx": 19262,
+      "contains": 19263,
+      "covery": 19264,
+      "Msg": 19265,
+      "事实清楚": 19266,
+      "otive": 19267,
+      "abad": 19268,
+      "thur": 19269,
+      "可在接到判决书的第二日起十日内": 19270,
+      "精心": 19271,
+      "那就是": 19272,
+      "一项": 19273,
+      "▁Tax": 19274,
+      "peg": 19275,
+      "ultural": 19276,
+      "▁ma": 19277,
+      "增加了": 19278,
+      "较大的": 19279,
+      "▁Const": 19280,
+      "军队": 19281,
+      "CAD": 19282,
+      "Two": 19283,
+      "olo": 19284,
+      "chrome": 19285,
+      "nwe": 19286,
+      "▁appearances": 19287,
+      "eur": 19288,
+      "▁Cu": 19289,
+      "入口": 19290,
+      "新华社": 19291,
+      "▁lateral": 19292,
+      "aton": 19293,
+      "▁comparable": 19294,
+      "▁Sport": 19295,
+      "▁有": 19296,
+      "静脉": 19297,
+      "▁Nations": 19298,
+      "▁:)": 19299,
+      "住院": 19300,
+      "▁neuronal": 19301,
+      "进行分析": 19302,
+      "▁refers": 19303,
+      "按摩": 19304,
+      "steps": 19305,
+      "▁crack": 19306,
+      "带有": 19307,
+      "basic": 19308,
+      "▁equally": 19309,
+      "诗人": 19310,
+      "oscopic": 19311,
+      "▁新": 19312,
+      "▁lambda": 19313,
+      "▁Performance": 19314,
+      "▁Township": 19315,
+      "SB": 19316,
+      "一部": 19317,
+      "县城": 19318,
+      "▁cry": 19319,
+      "duration": 19320,
+      "游泳": 19321,
+      "nthis": 19322,
+      "文档": 19323,
+      "词语": 19324,
+      "合法传唤": 19325,
+      "北方": 19326,
+      "▁muscles": 19327,
+      "那么多": 19328,
+      "▁seeds": 19329,
+      ")),": 19330,
+      "▁accompanied": 19331,
+      "psilon": 19332,
+      "多数": 19333,
+      "nums": 19334,
+      "▁everywhere": 19335,
+      "水中": 19336,
+      "apsed": 19337,
+      "▁shops": 19338,
+      "▁computers": 19339,
+      "▁Capital": 19340,
+      "ERE": 19341,
+      "▁spont": 19342,
+      "常委会": 19343,
+      "▁stain": 19344,
+      "▁damaged": 19345,
+      "两边": 19346,
+      "ghan": 19347,
+      "发电": 19348,
+      "aware": 19349,
+      "▁spots": 19350,
+      "▁Sher": 19351,
+      "海拔": 19352,
+      "▁crazy": 19353,
+      "▁Ah": 19354,
+      "▁districts": 19355,
+      "▁notation": 19356,
+      "▁withdraw": 19357,
+      "是你": 19358,
+      "brand": 19359,
+      "inyl": 19360,
+      "▁Would": 19361,
+      "场的": 19362,
+      "amer": 19363,
+      "ania": 19364,
+      "▁flavor": 19365,
+      "▁analytical": 19366,
+      "获得了": 19367,
+      "经纪": 19368,
+      "ectomy": 19369,
+      "gate": 19370,
+      "登记机关": 19371,
+      "▁Stephen": 19372,
+      "▁coaches": 19373,
+      "IBUT": 19374,
+      "▁Limited": 19375,
+      "itecture": 19376,
+      "▁funny": 19377,
+      "▁ps": 19378,
+      "expand": 19379,
+      "▁forth": 19380,
+      "▁Village": 19381,
+      "starts": 19382,
+      "▁races": 19383,
+      "merge": 19384,
+      "▁closure": 19385,
+      "quant": 19386,
+      "为止": 19387,
+      "\\'\"": 19388,
+      "发生了": 19389,
+      "▁storm": 19390,
+      "contents": 19391,
+      "mL": 19392,
+      "为例": 19393,
+      "▁Insp": 19394,
+      "▁github": 19395,
+      "审批表": 19396,
+      "▁funded": 19397,
+      "▁ru": 19398,
+      "▁restore": 19399,
+      "▁金": 19400,
+      "某种": 19401,
+      "}(\\\\": 19402,
+      "urations": 19403,
+      "喜爱": 19404,
+      "融入": 19405,
+      "▁metadata": 19406,
+      "oped": 19407,
+      "数额": 19408,
+      "增大": 19409,
+      "感兴趣": 19410,
+      "▁aver": 19411,
+      "▁unclear": 19412,
+      "二月": 19413,
+      "整治": 19414,
+      "爆发": 19415,
+      "▁Dise": 19416,
+      "cdots": 19417,
+      "ibliography": 19418,
+      "Member": 19419,
+      "用药": 19420,
+      "好多": 19421,
+      "ynchron": 19422,
+      "终字第": 19423,
+      "进出口": 19424,
+      "{\\'": 19425,
+      "创建通道": 19426,
+      "symbol": 19427,
+      "词条初始": 19428,
+      "oming": 19429,
+      "信息均引自": 19430,
+      "词条初始信息均引自": 19431,
+      "acco": 19432,
+      "▁bunch": 19433,
+      "企业词条": 19434,
+      "企业词条极速": 19435,
+      "企业词条极速创建通道": 19436,
+      "百度百科企业词条极速创建通道": 19437,
+      "▁Solutions": 19438,
+      "▁hide": 19439,
+      "▁destr": 19440,
+      "▁Currently": 19441,
+      "▁feeding": 19442,
+      "▁slot": 19443,
+      "▁neon": 19444,
+      "▁assignment": 19445,
+      "▁submission": 19446,
+      "ubernetes": 19447,
+      "company": 19448,
+      "adies": 19449,
+      "▁surprised": 19450,
+      "}}{\\\\": 19451,
+      "▁Amb": 19452,
+      "▁households": 19453,
+      "两位": 19454,
+      "▁Veg": 19455,
+      "May": 19456,
+      "Dict": 19457,
+      "的天": 19458,
+      "蛋糕": 19459,
+      "nerg": 19460,
+      "赶紧": 19461,
+      "nSpec": 19462,
+      "▁towns": 19463,
+      "的信息": 19464,
+      "▁gay": 19465,
+      "ignKey": 19466,
+      ">）": 19467,
+      "▁Scient": 19468,
+      "ocl": 19469,
+      "▁myel": 19470,
+      "isons": 19471,
+      "▁anth": 19472,
+      "herence": 19473,
+      "逮捕": 19474,
+      "PCR": 19475,
+      "▁compete": 19476,
+      "工商行政": 19477,
+      "Ca": 19478,
+      "▁Philadelphia": 19479,
+      "▁Forum": 19480,
+      "形成了": 19481,
+      "▁Alexander": 19482,
+      "_(": 19483,
+      "aceut": 19484,
+      "anted": 19485,
+      "▁cycles": 19486,
+      "▁wash": 19487,
+      "▁equilibrium": 19488,
+      "\\\\/": 19489,
+      "▁inhibited": 19490,
+      "▁accurately": 19491,
+      "▁Cos": 19492,
+      ":**": 19493,
+      "▁logger": 19494,
+      "▁Hal": 19495,
+      "陶瓷": 19496,
+      "irts": 19497,
+      "财经": 19498,
+      "skip": 19499,
+      "▁Flow": 19500,
+      "▁whilst": 19501,
+      "病的": 19502,
+      "asia": 19503,
+      "厚的": 19504,
+      "社会抚养费": 19505,
+      "▁北京": 19506,
+      "▁rising": 19507,
+      "法人": 19508,
+      "可以说": 19509,
+      "纪委": 19510,
+      "隐藏": 19511,
+      "▁fest": 19512,
+      "LEV": 19513,
+      "看到了": 19514,
+      "▁pulse": 19515,
+      "ntest": 19516,
+      "时许": 19517,
+      "▁Stone": 19518,
+      "最为": 19519,
+      "▁fractions": 19520,
+      "爱你": 19521,
+      "在外": 19522,
+      "▁sharp": 19523,
+      "nBuild": 19524,
+      "cod": 19525,
+      "残疾": 19526,
+      "▁Overall": 19527,
+      "书籍": 19528,
+      "精确": 19529,
+      "▁prize": 19530,
+      "高压": 19531,
+      "▁soul": 19532,
+      "▁tou": 19533,
+      "Flow": 19534,
+      "宝贝": 19535,
+      "▁recipient": 19536,
+      "▁universal": 19537,
+      "大气": 19538,
+      "十一日": 19539,
+      "▁tagged": 19540,
+      "▁Enron": 19541,
+      "nDoes": 19542,
+      "中山": 19543,
+      "assign": 19544,
+      "nia": 19545,
+      "▁Result": 19546,
+      "▁hopes": 19547,
+      "责令被执行人": 19548,
+      "▁explored": 19549,
+      "气质": 19550,
+      "▁atom": 19551,
+      "▁歌曲": 19552,
+      "rules": 19553,
+      "ureau": 19554,
+      "Func": 19555,
+      "▁Brian": 19556,
+      "uity": 19557,
+      "▁Diego": 19558,
+      "prime": 19559,
+      "安县": 19560,
+      "▁Professional": 19561,
+      "法则": 19562,
+      "州区": 19563,
+      "替代": 19564,
+      "iking": 19565,
+      "▁男": 19566,
+      "▁compile": 19567,
+      "▁intention": 19568,
+      "▁loans": 19569,
+      "动漫": 19570,
+      "▁satisfied": 19571,
+      "发射": 19572,
+      "被人": 19573,
+      "身为": 19574,
+      "opro": 19575,
+      "▁indices": 19576,
+      "▁Miller": 19577,
+      "]]\\": 19578,
+      "▁Battle": 19579,
+      "sea": 19580,
+      "▁initiative": 19581,
+      "inery": 19582,
+      "▁diver": 19583,
+      "▁pneum": 19584,
+      "服务的": 19585,
+      "Xml": 19586,
+      "造成的": 19587,
+      "▁Elizabeth": 19588,
+      "tasks": 19589,
+      "▁Living": 19590,
+      "入手": 19591,
+      "▁specialized": 19592,
+      "lem": 19593,
+      "▁deform": 19594,
+      "上网": 19595,
+      "▁Compar": 19596,
+      "敌人": 19597,
+      "▁Gallery": 19598,
+      "讨厌": 19599,
+      "▁irre": 19600,
+      "下来的": 19601,
+      "▁logs": 19602,
+      "▁Fun": 19603,
+      "偶尔": 19604,
+      "obic": 19605,
+      "▁Returns": 19606,
+      "STAT": 19607,
+      "illiant": 19608,
+      "▁venue": 19609,
+      "▁酒店": 19610,
+      "▁lang": 19611,
+      "▁Ho": 19612,
+      "reement": 19613,
+      "未知": 19614,
+      "留言": 19615,
+      "▁shed": 19616,
+      "changes": 19617,
+      "Detail": 19618,
+      "limited": 19619,
+      "▁Czech": 19620,
+      "ovo": 19621,
+      "相同的": 19622,
+      "不及": 19623,
+      "相关部门": 19624,
+      "▁Sort": 19625,
+      "康复": 19626,
+      "embed": 19627,
+      "▁implementing": 19628,
+      "的距离": 19629,
+      "第二百六十二条第二款": 19630,
+      "▁Unity": 19631,
+      "mount": 19632,
+      "▁transplantation": 19633,
+      "iotic": 19634,
+      "remote": 19635,
+      "opsy": 19636,
+      "▁*)": 19637,
+      "不对": 19638,
+      "usage": 19639,
+      "以此": 19640,
+      "}}}": 19641,
+      "oning": 19642,
+      "而成": 19643,
+      "▁Story": 19644,
+      "▁knee": 19645,
+      "▁expatriate": 19646,
+      "▁cholesterol": 19647,
+      "的第一": 19648,
+      "ASH": 19649,
+      "一方面": 19650,
+      "▁shapes": 19651,
+      "的目的": 19652,
+      "▁attending": 19653,
+      "假释": 19654,
+      "osity": 19655,
+      "describe": 19656,
+      "protocol": 19657,
+      "minute": 19658,
+      "▁Production": 19659,
+      "亚洲": 19660,
+      "奇怪": 19661,
+      "extension": 19662,
+      "▁consistently": 19663,
+      "▁Very": 19664,
+      "乡镇": 19665,
+      "豪华": 19666,
+      "ete": 19667,
+      "▁tone": 19668,
+      "▁spher": 19669,
+      "▁adequate": 19670,
+      "▁Pit": 19671,
+      "coin": 19672,
+      "▁volumes": 19673,
+      "▁infect": 19674,
+      "▁invari": 19675,
+      "Variable": 19676,
+      "▁vessels": 19677,
+      "想想": 19678,
+      "▁Ryan": 19679,
+      "RECT": 19680,
+      "▁chemistry": 19681,
+      "▁nick": 19682,
+      "节约": 19683,
+      "pletion": 19684,
+      "▁Polish": 19685,
+      "Word": 19686,
+      "]))\\": 19687,
+      "▁tile": 19688,
+      "owers": 19689,
+      "条件下": 19690,
+      "BF": 19691,
+      "▁las": 19692,
+      "白云": 19693,
+      "▁designing": 19694,
+      "立体": 19695,
+      "/{": 19696,
+      "nest": 19697,
+      "配有": 19698,
+      "陈某": 19699,
+      "移民": 19700,
+      "内蒙古自治区": 19701,
+      "▁templates": 19702,
+      "▁wing": 19703,
+      "▁spoke": 19704,
+      "▁Adam": 19705,
+      "▁tennis": 19706,
+      "▁BUT": 19707,
+      "▁Tool": 19708,
+      "▁β": 19709,
+      "umbn": 19710,
+      "▁д": 19711,
+      "将于": 19712,
+      "▁equals": 19713,
+      "management": 19714,
+      "[(": 19715,
+      "村委会": 19716,
+      "▁uncertainty": 19717,
+      "offs": 19718,
+      "INFO": 19719,
+      "▁discussions": 19720,
+      "第四十": 19721,
+      "ubble": 19722,
+      "为一": 19723,
+      "进攻": 19724,
+      "▁excit": 19725,
+      "▁organisation": 19726,
+      "一次性": 19727,
+      "aque": 19728,
+      "oku": 19729,
+      ">>\\": 19730,
+      "Some": 19731,
+      "▁Lewis": 19732,
+      "iw": 19733,
+      "Section": 19734,
+      "▁styled": 19735,
+      "paths": 19736,
+      "bmatrix": 19737,
+      "客服": 19738,
+      "▁productivity": 19739,
+      "坚定": 19740,
+      "▁imper": 19741,
+      "向原告借款": 19742,
+      "members": 19743,
+      "ethe": 19744,
+      "占有": 19745,
+      "trl": 19746,
+      "compat": 19747,
+      "Offset": 19748,
+      "cycl": 19749,
+      "uler": 19750,
+      "arth": 19751,
+      "▁Song": 19752,
+      "二手": 19753,
+      "▁Buff": 19754,
+      "▁replic": 19755,
+      "完成了": 19756,
+      "▁requiring": 19757,
+      "▁carrier": 19758,
+      "发出执行通知书": 19759,
+      "▁Website": 19760,
+      "▁burden": 19761,
+      "▁physicians": 19762,
+      "▁rings": 19763,
+      "▁delayed": 19764,
+      "组件": 19765,
+      "语音": 19766,
+      "随意": 19767,
+      "Ax": 19768,
+      "的安全": 19769,
+      "▁scrap": 19770,
+      "▁mirror": 19771,
+      "▁hex": 19772,
+      "▁Early": 19773,
+      "herit": 19774,
+      "inction": 19775,
+      "是谁": 19776,
+      "Transaction": 19777,
+      "变成了": 19778,
+      "nHave": 19779,
+      "售价": 19780,
+      "风雨": 19781,
+      "▁clip": 19782,
+      "快捷": 19783,
+      "chnology": 19784,
+      "▁atoms": 19785,
+      "▁Tech": 19786,
+      "的美": 19787,
+      "▁dietary": 19788,
+      "首发": 19789,
+      "▁Drive": 19790,
+      "energy": 19791,
+      "amination": 19792,
+      "iac": 19793,
+      "加载": 19794,
+      "颜值": 19795,
+      "▁connecting": 19796,
+      "▁consultation": 19797,
+      "ahoma": 19798,
+      "▁inventory": 19799,
+      "nOf": 19800,
+      "QQ": 19801,
+      "▁developments": 19802,
+      "列表": 19803,
+      "▁Las": 19804,
+      "nreturn": 19805,
+      "phones": 19806,
+      "酒精": 19807,
+      "▁clinically": 19808,
+      "▁胡": 19809,
+      "▁({": 19810,
+      "▁treating": 19811,
+      "abama": 19812,
+      "emet": 19813,
+      "▁stere": 19814,
+      "▁pun": 19815,
+      "ellectual": 19816,
+      "▁taxes": 19817,
+      "可惜": 19818,
+      "罗马": 19819,
+      "犯危险驾驶罪": 19820,
+      "全身": 19821,
+      "然后再": 19822,
+      "NING": 19823,
+      "或者直接向": 19824,
+      "Download": 19825,
+      "amsung": 19826,
+      "党组": 19827,
+      "lad": 19828,
+      "▁engines": 19829,
+      "扎实": 19830,
+      "Identifier": 19831,
+      "付费": 19832,
+      "▁fatty": 19833,
+      "▁Gas": 19834,
+      "▁entrance": 19835,
+      "▁colleagues": 19836,
+      "第四章": 19837,
+      "▁coat": 19838,
+      "rupted": 19839,
+      "▁注": 19840,
+      "aceutical": 19841,
+      "结局": 19842,
+      "争取": 19843,
+      "制的": 19844,
+      "的一切": 19845,
+      "patic": 19846,
+      "主流": 19847,
+      "▁pancreat": 19848,
+      "urope": 19849,
+      "▁Exchange": 19850,
+      "Library": 19851,
+      "Register": 19852,
+      "第二次": 19853,
+      "essment": 19854,
+      "Pop": 19855,
+      "厚度": 19856,
+      "▁Cit": 19857,
+      "Validator": 19858,
+      "gnu": 19859,
+      "像素": 19860,
+      "uzzy": 19861,
+      "条件的": 19862,
+      "aire": 19863,
+      "▁disp": 19864,
+      "▁obst": 19865,
+      "▁USE": 19866,
+      "▁believes": 19867,
+      "盈利": 19868,
+      "Parent": 19869,
+      "缺陷": 19870,
+      "▁trauma": 19871,
+      "nPopulated": 19872,
+      "以被告人": 19873,
+      "Remove": 19874,
+      "创办": 19875,
+      "essed": 19876,
+      "▁dependency": 19877,
+      "cancel": 19878,
+      "▁digits": 19879,
+      "llib": 19880,
+      "FE": 19881,
+      "Dist": 19882,
+      "▁yields": 19883,
+      "一把": 19884,
+      "▁orbit": 19885,
+      "的精神": 19886,
+      "locale": 19887,
+      "人心": 19888,
+      "nRem": 19889,
+      "Results": 19890,
+      "一看": 19891,
+      "▁sevent": 19892,
+      "抵押": 19893,
+      "日向本院提起公诉": 19894,
+      "▁clicking": 19895,
+      "▁broke": 19896,
+      "▁metres": 19897,
+      "{}\\": 19898,
+      "omorphic": 19899,
+      "▁incorporated": 19900,
+      "▁mesh": 19901,
+      "▁soccer": 19902,
+      "▁cnt": 19903,
+      "劳动任务": 19904,
+      "KEN": 19905,
+      "▁mineral": 19906,
+      "estone": 19907,
+      "▁Mexican": 19908,
+      "机遇": 19909,
+      "▁tmp": 19910,
+      "proof": 19911,
+      "裁决": 19912,
+      "kan": 19913,
+      "自行车": 19914,
+      "爷爷": 19915,
+      "▁aqu": 19916,
+      "oys": 19917,
+      "▁pill": 19918,
+      "上课": 19919,
+      "不论": 19920,
+      "▁Mu": 19921,
+      "▁schemes": 19922,
+      "Extension": 19923,
+      "criptor": 19924,
+      "▁Oak": 19925,
+      "▁Complex": 19926,
+      "#[": 19927,
+      "ет": 19928,
+      "}</": 19929,
+      "▁hate": 19930,
+      "abric": 19931,
+      "▁Maryland": 19932,
+      "fg": 19933,
+      "▁fung": 19934,
+      "▁reserve": 19935,
+      "▁occurring": 19936,
+      "▁suffer": 19937,
+      "别墅": 19938,
+      "nPr": 19939,
+      "可持续": 19940,
+      "热水": 19941,
+      "ugging": 19942,
+      "▁melan": 19943,
+      "▁fits": 19944,
+      "▁remind": 19945,
+      "发票": 19946,
+      "▁Soviet": 19947,
+      "就好": 19948,
+      "无奈": 19949,
+      "分的": 19950,
+      "菜单": 19951,
+      "▁plug": 19952,
+      "▁gifts": 19953,
+      "▁myocardial": 19954,
+      "who": 19955,
+      "▁advances": 19956,
+      "样本": 19957,
+      "▁phones": 19958,
+      "提起诉讼": 19959,
+      "▁postoperative": 19960,
+      "▁diffusion": 19961,
+      "能力的": 19962,
+      "products": 19963,
+      "热爱": 19964,
+      "▁conjug": 19965,
+      "店铺": 19966,
+      "▁compensation": 19967,
+      "是否有": 19968,
+      "职员": 19969,
+      "limits": 19970,
+      "▁ranges": 19971,
+      "▁dy": 19972,
+      "nCurrent": 19973,
+      "IPT": 19974,
+      "Free": 19975,
+      "知识产权": 19976,
+      "▁Regional": 19977,
+      "AAAA": 19978,
+      "expl": 19979,
+      "▁accounting": 19980,
+      "▁Dark": 19981,
+      "rible": 19982,
+      "igens": 19983,
+      "ventory": 19984,
+      "的环境": 19985,
+      "南北": 19986,
+      "nResults": 19987,
+      "allas": 19988,
+      "resp": 19989,
+      "nEach": 19990,
+      "赛事": 19991,
+      "一定是": 19992,
+      "scal": 19993,
+      "▁impacts": 19994,
+      "通过本院或者直接向": 19995,
+      "▁pleasure": 19996,
+      "▁spectroscopy": 19997,
+      "uzzle": 19998,
+      "组织机构": 19999,
+      "▁\\\\,": 20000,
+      "倍数": 20001,
+      "▁Rail": 20002,
+      "▁exploring": 20003,
+      "帝国": 20004,
+      "格局": 20005,
+      "同时也": 20006,
+      "beat": 20007,
+      "hell": 20008,
+      "▁shorter": 20009,
+      "nCheck": 20010,
+      "▁extracted": 20011,
+      "iop": 20012,
+      "tassert": 20013,
+      "▁Term": 20014,
+      "为终审": 20015,
+      "endo": 20016,
+      "▁Ship": 20017,
+      "咱们": 20018,
+      "▁renov": 20019,
+      "▁practition": 20020,
+      "nSec": 20021,
+      "之类的": 20022,
+      "cross": 20023,
+      "减轻": 20024,
+      "▁premier": 20025,
+      "因数": 20026,
+      "可以用": 20027,
+      "▁accord": 20028,
+      "mons": 20029,
+      "依次": 20030,
+      "楼盘": 20031,
+      "教育的": 20032,
+      "▁Sky": 20033,
+      "/%": 20034,
+      "iso": 20035,
+      "▁Ess": 20036,
+      "的老": 20037,
+      "▁overs": 20038,
+      "海军": 20039,
+      "记住": 20040,
+      "按钮": 20041,
+      "MBER": 20042,
+      "▁exclus": 20043,
+      "▁recycl": 20044,
+      "▁branches": 20045,
+      "sters": 20046,
+      "没人": 20047,
+      "pragma": 20048,
+      "▁heavily": 20049,
+      "agu": 20050,
+      "第二章": 20051,
+      "▁resident": 20052,
+      "肺炎": 20053,
+      "当你": 20054,
+      "/_": 20055,
+      "怀孕": 20056,
+      "活力": 20057,
+      "▁Marsh": 20058,
+      "▁idx": 20059,
+      "作文": 20060,
+      "▁Paper": 20061,
+      "▁Melbourne": 20062,
+      "搭建": 20063,
+      "笔者": 20064,
+      "▁possibilities": 20065,
+      "▁Qt": 20066,
+      "月二十日": 20067,
+      "第七十八条": 20068,
+      "最高人民法院关于人民法院": 20069,
+      "development": 20070,
+      "早餐": 20071,
+      "Your": 20072,
+      "仿真": 20073,
+      "▁roads": 20074,
+      "▁presenting": 20075,
+      "nstruct": 20076,
+      "mont": 20077,
+      "外交": 20078,
+      "▁pa": 20079,
+      "什么时候": 20080,
+      "▁assumption": 20081,
+      "连锁": 20082,
+      "iel": 20083,
+      "要有": 20084,
+      "▁guaranteed": 20085,
+      "▁pointed": 20086,
+      "ulse": 20087,
+      "▁chip": 20088,
+      "Environment": 20089,
+      "发生法律效力后": 20090,
+      "发生的": 20091,
+      "不停": 20092,
+      "旨在": 20093,
+      "regular": 20094,
+      "toggle": 20095,
+      "火箭": 20096,
+      "gif": 20097,
+      "这一点": 20098,
+      "▁hemat": 20099,
+      "刑事拘留": 20100,
+      "依法适用简易程序": 20101,
+      "▁inline": 20102,
+      "▁orange": 20103,
+      "住宿": 20104,
+      "onder": 20105,
+      "传感器": 20106,
+      "▁thousand": 20107,
+      "GP": 20108,
+      "▁grey": 20109,
+      "▁Insurance": 20110,
+      "仿佛": 20111,
+      "▁Ross": 20112,
+      "since": 20113,
+      "▁alterations": 20114,
+      "商店": 20115,
+      "总之": 20116,
+      "igs": 20117,
+      "series": 20118,
+      "atern": 20119,
+      "derived": 20120,
+      "▁mixing": 20121,
+      "稳定性": 20122,
+      "▁Municip": 20123,
+      "ercise": 20124,
+      "levance": 20125,
+      "▁FREE": 20126,
+      "▁decreases": 20127,
+      "年前": 20128,
+      "招商": 20129,
+      "units": 20130,
+      "######": 20131,
+      "▁catalog": 20132,
+      "一定会": 20133,
+      "hedral": 20134,
+      "引起的": 20135,
+      "抓获": 20136,
+      "▁pediatric": 20137,
+      "▁insect": 20138,
+      "Ant": 20139,
+      "▁builder": 20140,
+      "总收入": 20141,
+      "▁Manchester": 20142,
+      "预订": 20143,
+      "subscribe": 20144,
+      "竞争力": 20145,
+      "mitter": 20146,
+      "外科": 20147,
+      "▁\\\\)\\": 20148,
+      "▁bags": 20149,
+      "小朋友": 20150,
+      "▁morb": 20151,
+      "__\\": 20152,
+      "▁pulled": 20153,
+      "条第": 20154,
+      "了一种": 20155,
+      "稳定的": 20156,
+      "出售": 20157,
+      "▁ventricular": 20158,
+      "integer": 20159,
+      "EXT": 20160,
+      "不存在": 20161,
+      "▁Jordan": 20162,
+      "amar": 20163,
+      "▁npm": 20164,
+      "▁frequencies": 20165,
+      "或者是": 20166,
+      "的证据": 20167,
+      "charg": 20168,
+      "working": 20169,
+      "▁cancers": 20170,
+      "▁Kevin": 20171,
+      "熟练": 20172,
+      "▁Psych": 20173,
+      "▁gro": 20174,
+      "via": 20175,
+      "craft": 20176,
+      "nInt": 20177,
+      "税收": 20178,
+      "▁Email": 20179,
+      "▁sphere": 20180,
+      "cents": 20181,
+      "▁Marg": 20182,
+      "党委书记": 20183,
+      "十日": 20184,
+      "用水": 20185,
+      "ADD": 20186,
+      "▁heading": 20187,
+      "除数": 20188,
+      "▁leak": 20189,
+      "▁tower": 20190,
+      "▁ceremony": 20191,
+      "spot": 20192,
+      "单词": 20193,
+      "▁Norm": 20194,
+      "▁UV": 20195,
+      "石家庄": 20196,
+      "▁Five": 20197,
+      "聪明": 20198,
+      "选举": 20199,
+      "▁Properties": 20200,
+      "承受": 20201,
+      "阿尔": 20202,
+      "▁definitions": 20203,
+      "▁behalf": 20204,
+      "otlin": 20205,
+      "▁accessories": 20206,
+      "DT": 20207,
+      "▁steady": 20208,
+      "▁Junior": 20209,
+      "▁Yet": 20210,
+      "▁headqu": 20211,
+      "▁difficulties": 20212,
+      "prob": 20213,
+      "各国": 20214,
+      "▁Pas": 20215,
+      "▁Prob": 20216,
+      "oths": 20217,
+      "感受到": 20218,
+      "什么样的": 20219,
+      "流水": 20220,
+      "一部分": 20221,
+      "iPhone": 20222,
+      "uble": 20223,
+      "▁Lady": 20224,
+      "agg": 20225,
+      ")))": 20226,
+      "▁prolong": 20227,
+      "▁Row": 20228,
+      "WR": 20229,
+      "'''\\": 20230,
+      "ographical": 20231,
+      "▁GL": 20232,
+      "的学习": 20233,
+      "ucc": 20234,
+      "Trace": 20235,
+      "▁falling": 20236,
+      "▁filling": 20237,
+      "▁influences": 20238,
+      "▁strange": 20239,
+      "合物": 20240,
+      "▁opposed": 20241,
+      "▁signing": 20242,
+      "出现的": 20243,
+      "▁Ow": 20244,
+      "Proxy": 20245,
+      "开拓": 20246,
+      "iances": 20247,
+      "▁documentary": 20248,
+      "所述": 20249,
+      "extra": 20250,
+      "倒入": 20251,
+      "▁geometric": 20252,
+      "▁Atlantic": 20253,
+      "Video": 20254,
+      "▁Polit": 20255,
+      "第十": 20256,
+      "▁NAS": 20257,
+      "▁如果": 20258,
+      "▁Azure": 20259,
+      "▁evaluating": 20260,
+      "的总": 20261,
+      "ellar": 20262,
+      "Linq": 20263,
+      "phys": 20264,
+      "▁Primary": 20265,
+      "▁piano": 20266,
+      "作战": 20267,
+      "pread": 20268,
+      "清水": 20269,
+      "▁replication": 20270,
+      "▁predictive": 20271,
+      "刑初": 20272,
+      "候审": 20273,
+      "▁pron": 20274,
+      "iations": 20275,
+      "隔离": 20276,
+      "五个": 20277,
+      "视角": 20278,
+      "▁coupling": 20279,
+      "▁NFL": 20280,
+      "▁Gree": 20281,
+      "民一初字第": 20282,
+      "▁collaborative": 20283,
+      "alg": 20284,
+      "▁glyc": 20285,
+      "▁accused": 20286,
+      "钢铁": 20287,
+      "好友": 20288,
+      "辛苦": 20289,
+      "hub": 20290,
+      "你认为": 20291,
+      "▁speaker": 20292,
+      "▁aband": 20293,
+      "只需要": 20294,
+      "正确的选项": 20295,
+      "▁chromat": 20296,
+      "支持公诉": 20297,
+      "RIGHT": 20298,
+      "▁fiction": 20299,
+      "rust": 20300,
+      "▁laid": 20301,
+      "几何": 20302,
+      "▁educated": 20303,
+      "成功的": 20304,
+      "是可以": 20305,
+      "观看": 20306,
+      "vet": 20307,
+      "HOME": 20308,
+      "方位": 20309,
+      "记载": 20310,
+      "我们要": 20311,
+      "izard": 20312,
+      "angles": 20313,
+      "上传": 20314,
+      "thony": 20315,
+      "取保": 20316,
+      "ields": 20317,
+      "▁Save": 20318,
+      "▁suppose": 20319,
+      "一周": 20320,
+      "logic": 20321,
+      "运作": 20322,
+      "▁demonstrates": 20323,
+      "▁Meta": 20324,
+      "wr": 20325,
+      "hyd": 20326,
+      "▁eliminate": 20327,
+      "▁voted": 20328,
+      "ＸＸ": 20329,
+      "千里": 20330,
+      "的社会": 20331,
+      "▁ery": 20332,
+      "Contract": 20333,
+      "▁forever": 20334,
+      "重要的是": 20335,
+      "▁viewing": 20336,
+      "▁contracts": 20337,
+      "再加上": 20338,
+      "拉丁": 20339,
+      "STRACT": 20340,
+      "景点": 20341,
+      "玩具": 20342,
+      "▁ratios": 20343,
+      "有助于": 20344,
+      "olas": 20345,
+      "▁nanop": 20346,
+      "▁Clinton": 20347,
+      "入门": 20348,
+      "erto": 20349,
+      "▁Sample": 20350,
+      "▁algebraic": 20351,
+      "▁compression": 20352,
+      "▁\"${": 20353,
+      "▁Sin": 20354,
+      "女主": 20355,
+      "ibl": 20356,
+      "aman": 20357,
+      "aro": 20358,
+      "▁summar": 20359,
+      "▁vocal": 20360,
+      "▁Grad": 20361,
+      "nInter": 20362,
+      "正好": 20363,
+      "▁classroom": 20364,
+      "▁actively": 20365,
+      "collect": 20366,
+      "nRUN": 20367,
+      "▁flying": 20368,
+      "▁queries": 20369,
+      "▁preparing": 20370,
+      "▁MRI": 20371,
+      "oxide": 20372,
+      "comments": 20373,
+      "多家": 20374,
+      "江区": 20375,
+      "口感": 20376,
+      "辣椒": 20377,
+      "建于": 20378,
+      "▁magnet": 20379,
+      "▁reserv": 20380,
+      "▁scripts": 20381,
+      "日作出的": 20382,
+      "▁perception": 20383,
+      "▁Steel": 20384,
+      "actors": 20385,
+      "▁yards": 20386,
+      "▁MB": 20387,
+      "▁Mit": 20388,
+      "▁输入": 20389,
+      "▁exhaust": 20390,
+      "多年的": 20391,
+      "▁tension": 20392,
+      "▁mitochondrial": 20393,
+      "坚决": 20394,
+      ")}\\": 20395,
+      "▁Ukraine": 20396,
+      "▁Memorial": 20397,
+      "ifest": 20398,
+      "getElement": 20399,
+      "▁intensive": 20400,
+      "▁CSS": 20401,
+      "PN": 20402,
+      "▁advis": 20403,
+      "如果未按本判决": 20404,
+      "橡胶": 20405,
+      "elli": 20406,
+      "ndata": 20407,
+      "▁nin": 20408,
+      "▁democr": 20409,
+      "可用": 20410,
+      "abe": 20411,
+      "并未": 20412,
+      "▁clusters": 20413,
+      "▁rarely": 20414,
+      "也不能": 20415,
+      "bank": 20416,
+      "ooter": 20417,
+      "出庭支持公诉": 20418,
+      "ToString": 20419,
+      "▁lap": 20420,
+      "umor": 20421,
+      "▁作者简介": 20422,
+      "▁Deep": 20423,
+      "nut": 20424,
+      "两个人": 20425,
+      "▁Arm": 20426,
+      "▁Survey": 20427,
+      "▁designated": 20428,
+      "▁functioning": 20429,
+      "焦虑": 20430,
+      "的功能": 20431,
+      "▁Linear": 20432,
+      "▁reveals": 20433,
+      "▁Tennessee": 20434,
+      "▁adaptation": 20435,
+      "是一部": 20436,
+      "▁Safety": 20437,
+      "▁spectra": 20438,
+      "▁disability": 20439,
+      "semin": 20440,
+      "▁Ark": 20441,
+      "Fix": 20442,
+      "▁screw": 20443,
+      "▁Ten": 20444,
+      "等地": 20445,
+      "曝光": 20446,
+      "▁localization": 20447,
+      ">\\\\": 20448,
+      "bone": 20449,
+      "buntu": 20450,
+      "Before": 20451,
+      "▁Dun": 20452,
+      "'(": 20453,
+      "越南": 20454,
+      "举报": 20455,
+      "为人": 20456,
+      "▁pdf": 20457,
+      "▁admission": 20458,
+      "贩卖": 20459,
+      "面包": 20460,
+      "▁Oil": 20461,
+      "▁mol": 20462,
+      "▁combining": 20463,
+      "▁switching": 20464,
+      "minecraft": 20465,
+      "olk": 20466,
+      "欠原告": 20467,
+      "以便": 20468,
+      "hop": 20469,
+      "警方": 20470,
+      "▁:)\\": 20471,
+      "▁pipe": 20472,
+      "Html": 20473,
+      "▁Originally": 20474,
+      "不去": 20475,
+      "伟大": 20476,
+      "▁Members": 20477,
+      "omm": 20478,
+      "▁extensions": 20479,
+      "说道": 20480,
+      "▁Norway": 20481,
+      "会不会": 20482,
+      "应用于": 20483,
+      "▁Democrats": 20484,
+      "▁Jen": 20485,
+      "dor": 20486,
+      "▁alive": 20487,
+      "下落": 20488,
+      "荣耀": 20489,
+      "事业单位": 20490,
+      "ARS": 20491,
+      "几次": 20492,
+      "帮我": 20493,
+      "应当提交上诉状正本一份": 20494,
+      "饭店": 20495,
+      "Events": 20496,
+      "nBritish": 20497,
+      "遗产": 20498,
+      "▁liked": 20499,
+      "▁phases": 20500,
+      "时代的": 20501,
+      "弟弟": 20502,
+      "▁flower": 20503,
+      "▁tap": 20504,
+      "持有": 20505,
+      "计量": 20506,
+      "▁startup": 20507,
+      "origin": 20508,
+      "冬季": 20509,
+      "__)\\": 20510,
+      "几种": 20511,
+      "ли": 20512,
+      "cosystem": 20513,
+      "两条": 20514,
+      "risk": 20515,
+      "▁Electric": 20516,
+      "好处": 20517,
+      "single": 20518,
+      "影像": 20519,
+      "是一名": 20520,
+      "的多": 20521,
+      "▁army": 20522,
+      "▁epidem": 20523,
+      "▁junior": 20524,
+      "十二条": 20525,
+      "编码": 20526,
+      "航天": 20527,
+      "erver": 20528,
+      "ationship": 20529,
+      "▁placement": 20530,
+      "eno": 20531,
+      "没有任何": 20532,
+      "PF": 20533,
+      "腐蚀": 20534,
+      "▁Template": 20535,
+      "XXX": 20536,
+      "▁roughly": 20537,
+      "至今未履行": 20538,
+      "黑暗": 20539,
+      "服务中心": 20540,
+      "▁weird": 20541,
+      "inflamm": 20542,
+      "有多少": 20543,
+      "▁pri": 20544,
+      "月出生": 20545,
+      "elve": 20546,
+      "取保候审": 20547,
+      "民生": 20548,
+      "▁scanning": 20549,
+      "generated": 20550,
+      "▁presidential": 20551,
+      "▁Highway": 20552,
+      "▁Kentucky": 20553,
+      "generator": 20554,
+      "▁Intel": 20555,
+      "▁loads": 20556,
+      "▁adip": 20557,
+      "▁wavelength": 20558,
+      "▁RA": 20559,
+      "▁junction": 20560,
+      "▁emails": 20561,
+      "住址": 20562,
+      "▁categor": 20563,
+      "▁restrictions": 20564,
+      "故意伤害": 20565,
+      "▁inch": 20566,
+      "▁起点中文网": 20567,
+      "党支部": 20568,
+      "▁attach": 20569,
+      "▁killing": 20570,
+      "sch": 20571,
+      "因子": 20572,
+      "报考": 20573,
+      "▁fake": 20574,
+      "transfer": 20575,
+      "brew": 20576,
+      "切换": 20577,
+      "avit": 20578,
+      "原审原告": 20579,
+      "▁optimize": 20580,
+      "蓝色": 20581,
+      "不超过": 20582,
+      "主机": 20583,
+      "▁Rome": 20584,
+      "▁dictionary": 20585,
+      "▁Hind": 20586,
+      "ursive": 20587,
+      "国土资源局": 20588,
+      "▁climb": 20589,
+      "▁ML": 20590,
+      "▁Context": 20591,
+      "▁Issue": 20592,
+      "▁quadratic": 20593,
+      "▁Fast": 20594,
+      "▁Delhi": 20595,
+      "ein": 20596,
+      "辖区": 20597,
+      "饮料": 20598,
+      "与其": 20599,
+      "的钱": 20600,
+      "▁Rot": 20601,
+      "高新技术": 20602,
+      "▁Prize": 20603,
+      "▁exceptional": 20604,
+      "OF": 20605,
+      "rence": 20606,
+      "LES": 20607,
+      "却是": 20608,
+      "▁defend": 20609,
+      "▁ions": 20610,
+      "▁couples": 20611,
+      "▁Blog": 20612,
+      "▁DR": 20613,
+      "apor": 20614,
+      "▁encouraged": 20615,
+      "照明": 20616,
+      "离婚诉讼": 20617,
+      "▁discharge": 20618,
+      "▁strike": 20619,
+      "我自己": 20620,
+      "入选": 20621,
+      "MG": 20622,
+      "的光": 20623,
+      "oment": 20624,
+      "imeter": 20625,
+      "pleted": 20626,
+      "igator": 20627,
+      "▁yours": 20628,
+      "期货": 20629,
+      "Lib": 20630,
+      "依法由": 20631,
+      "▁\\'<": 20632,
+      "extract": 20633,
+      "scroll": 20634,
+      "在使用": 20635,
+      "放松": 20636,
+      "DES": 20637,
+      "▁deficiency": 20638,
+      "▁Youth": 20639,
+      "Area": 20640,
+      "▁memories": 20641,
+      ">`": 20642,
+      "手动": 20643,
+      "▁Independent": 20644,
+      "▁convenience": 20645,
+      "variable": 20646,
+      "▁occurrence": 20647,
+      "解读": 20648,
+      "除外": 20649,
+      "▁cerv": 20650,
+      "青海": 20651,
+      "遗传": 20652,
+      "▁ratings": 20653,
+      "erator": 20654,
+      "▁gorgeous": 20655,
+      "配备": 20656,
+      "ptic": 20657,
+      "看完": 20658,
+      "ophil": 20659,
+      "▁shots": 20660,
+      "▁perceived": 20661,
+      "▁biomark": 20662,
+      "Term": 20663,
+      "我认为": 20664,
+      "xt": 20665,
+      "号楼": 20666,
+      "提到": 20667,
+      "▁Motor": 20668,
+      "▁maternal": 20669,
+      "行长": 20670,
+      "祝福": 20671,
+      "东京": 20672,
+      "▁Draw": 20673,
+      "监狱于": 20674,
+      "▁delivering": 20675,
+      "▁widget": 20676,
+      "神奇": 20677,
+      "干扰": 20678,
+      "▁smoke": 20679,
+      "心灵": 20680,
+      "正规": 20681,
+      "theless": 20682,
+      "▁Inv": 20683,
+      "▁Bat": 20684,
+      "functions": 20685,
+      "▁ml": 20686,
+      "▁Burn": 20687,
+      "nNext": 20688,
+      "租金": 20689,
+      "▁sophistic": 20690,
+      "生产的": 20691,
+      "▁cortex": 20692,
+      "▁disput": 20693,
+      "▁multiply": 20694,
+      "很高": 20695,
+      "基础设施": 20696,
+      "▁variability": 20697,
+      "▁hierarch": 20698,
+      "东风": 20699,
+      "years": 20700,
+      "发现被执行人有可供执行财产的": 20701,
+      "资质": 20702,
+      "▁Photo": 20703,
+      "▁animation": 20704,
+      "market": 20705,
+      "period": 20706,
+      "cart": 20707,
+      "▁Observ": 20708,
+      "买的": 20709,
+      "initely": 20710,
+      "▁universities": 20711,
+      "Rect": 20712,
+      "▁franch": 20713,
+      "▁mobility": 20714,
+      "nUp": 20715,
+      "栽培": 20716,
+      "▁anterior": 20717,
+      "大学出版社": 20718,
+      "▁Recent": 20719,
+      "▁Mechan": 20720,
+      "rett": 20721,
+      "nSolve": 20722,
+      "▁establishment": 20723,
+      "▁Chrome": 20724,
+      "mos": 20725,
+      "建设的": 20726,
+      "摄像头": 20727,
+      "rik": 20728,
+      "▁unf": 20729,
+      "振兴": 20730,
+      "有一种": 20731,
+      "▁Ast": 20732,
+      "antom": 20733,
+      "▁fs": 20734,
+      "ysts": 20735,
+      "▁chains": 20736,
+      "▁Rap": 20737,
+      "供应商": 20738,
+      "acial": 20739,
+      "长安": 20740,
+      "lla": 20741,
+      "▁然后": 20742,
+      "在大": 20743,
+      "文件夹": 20744,
+      "MF": 20745,
+      "带来了": 20746,
+      "▁asthma": 20747,
+      "▁workflow": 20748,
+      "▁backup": 20749,
+      "不仅仅": 20750,
+      "复印件": 20751,
+      "稍微": 20752,
+      ")|": 20753,
+      "路由": 20754,
+      "▁describing": 20755,
+      "ulative": 20756,
+      "▁router": 20757,
+      "▁engineers": 20758,
+      "架构": 20759,
+      "recip": 20760,
+      "享有": 20761,
+      "快的": 20762,
+      "Comment": 20763,
+      "interest": 20764,
+      "ingers": 20765,
+      "严重的": 20766,
+      "服饰": 20767,
+      "▁FO": 20768,
+      "▁arrival": 20769,
+      "难题": 20770,
+      "▁stats": 20771,
+      "activation": 20772,
+      "wp": 20773,
+      "身材": 20774,
+      "Bit": 20775,
+      "价值观": 20776,
+      "改为": 20777,
+      "▁Size": 20778,
+      "流通": 20779,
+      "pathy": 20780,
+      "▁observe": 20781,
+      "emetery": 20782,
+      "ithmetic": 20783,
+      "tit": 20784,
+      "sequence": 20785,
+      ">{": 20786,
+      "风机": 20787,
+      "▁Mun": 20788,
+      "▁farmers": 20789,
+      "▁Treatment": 20790,
+      "某乙": 20791,
+      "罚款": 20792,
+      "高层": 20793,
+      "▁plt": 20794,
+      "English": 20795,
+      "▁arterial": 20796,
+      "▁Profile": 20797,
+      "张家": 20798,
+      "广西壮": 20799,
+      "通报": 20800,
+      "▁bronze": 20801,
+      "交叉": 20802,
+      "▁builds": 20803,
+      "模具": 20804,
+      "的发生": 20805,
+      "Security": 20806,
+      "▁kw": 20807,
+      "rous": 20808,
+      "overty": 20809,
+      "▁Native": 20810,
+      "以外": 20811,
+      "体内": 20812,
+      "▁indirect": 20813,
+      "▁traveling": 20814,
+      "十九条": 20815,
+      "▁smile": 20816,
+      "▁designers": 20817,
+      "▁meaningful": 20818,
+      "博客": 20819,
+      "诸多": 20820,
+      "▁scales": 20821,
+      "隐患": 20822,
+      "水分": 20823,
+      "一会": 20824,
+      "witzer": 20825,
+      "yrics": 20826,
+      "掉了": 20827,
+      "▁determin": 20828,
+      "▁Tai": 20829,
+      "Lock": 20830,
+      "plane": 20831,
+      "挑选": 20832,
+      "透露": 20833,
+      "秘书长": 20834,
+      "brace": 20835,
+      "nAct": 20836,
+      "TRIBUT": 20837,
+      "▁Miami": 20838,
+      "ahoo": 20839,
+      "▁ultra": 20840,
+      "nExample": 20841,
+      "向被告": 20842,
+      "▁Maria": 20843,
+      "▁separately": 20844,
+      "DATA": 20845,
+      "最早": 20846,
+      "▁Resources": 20847,
+      "▁vessel": 20848,
+      "研讨": 20849,
+      "管辖权": 20850,
+      "澳大利": 20851,
+      "▁quotes": 20852,
+      "getLogger": 20853,
+      "wind": 20854,
+      "▁DJ": 20855,
+      "▁defeated": 20856,
+      "(\\'#": 20857,
+      "▁topology": 20858,
+      "▁harvest": 20859,
+      "拍照": 20860,
+      "nBiography": 20861,
+      "▁holes": 20862,
+      "▁divisor": 20863,
+      "nBefore": 20864,
+      "zhe": 20865,
+      "▁thesis": 20866,
+      "gmail": 20867,
+      "Enabled": 20868,
+      "phosph": 20869,
+      "▁laugh": 20870,
+      "▁SR": 20871,
+      "▁AA": 20872,
+      "roit": 20873,
+      "stim": 20874,
+      "山水": 20875,
+      "walk": 20876,
+      "))/": 20877,
+      "\"/>\\": 20878,
+      "relation": 20879,
+      "广西壮族自治区": 20880,
+      "让自己": 20881,
+      "clerosis": 20882,
+      "▁interfer": 20883,
+      "澳大利亚": 20884,
+      "▁Still": 20885,
+      "有关法律规定": 20886,
+      "ال": 20887,
+      "▁answered": 20888,
+      "▁Third": 20889,
+      "BV": 20890,
+      "就没有": 20891,
+      "▁hub": 20892,
+      "▁raising": 20893,
+      "▁coli": 20894,
+      "▁oldest": 20895,
+      "\\'))": 20896,
+      "icator": 20897,
+      "Join": 20898,
+      "较低": 20899,
+      "otide": 20900,
+      "nCar": 20901,
+      "▁adapted": 20902,
+      "短信": 20903,
+      "开设": 20904,
+      "对原告": 20905,
+      "[:,": 20906,
+      "▁JS": 20907,
+      "擅长": 20908,
+      "多么": 20909,
+      "▁Style": 20910,
+      "▁expanding": 20911,
+      "flags": 20912,
+      "紧急": 20913,
+      "▁shortly": 20914,
+      "▁Marketing": 20915,
+      "整改": 20916,
+      "社会信用": 20917,
+      "consum": 20918,
+      "▁scenarios": 20919,
+      "的质量": 20920,
+      "本院不予": 20921,
+      "▁speakers": 20922,
+      "Scope": 20923,
+      "quot": 20924,
+      "napshot": 20925,
+      "▁singular": 20926,
+      "▁Staff": 20927,
+      "templates": 20928,
+      "rup": 20929,
+      "相识": 20930,
+      "audi": 20931,
+      "\\'][\\'": 20932,
+      "外部": 20933,
+      "oscow": 20934,
+      "dialog": 20935,
+      "数码": 20936,
+      "redis": 20937,
+      "为什么要": 20938,
+      "懂得": 20939,
+      "Eval": 20940,
+      "bx": 20941,
+      "▁fetal": 20942,
+      "▁Advanced": 20943,
+      "▁ordinary": 20944,
+      "Send": 20945,
+      "▁(@": 20946,
+      "▁subsid": 20947,
+      "的最小": 20948,
+      "失信被执行人": 20949,
+      "▁privile": 20950,
+      "▁dump": 20951,
+      "能在": 20952,
+      "这条": 20953,
+      "一半": 20954,
+      "▁desk": 20955,
+      "饲料": 20956,
+      "▁finishing": 20957,
+      "lav": 20958,
+      "▁PI": 20959,
+      "binary": 20960,
+      ";;": 20961,
+      "▁ranked": 20962,
+      "▁graphic": 20963,
+      "▁cuts": 20964,
+      "▁phenotype": 20965,
+      "iden": 20966,
+      "▁flags": 20967,
+      "▁curriculum": 20968,
+      "▁Hawai": 20969,
+      "阿拉": 20970,
+      "▁Hope": 20971,
+      "▁assault": 20972,
+      "附加": 20973,
+      "▁enrolled": 20974,
+      "▁ATP": 20975,
+      "▁vec": 20976,
+      "▁substit": 20977,
+      "▁Harry": 20978,
+      "跟我": 20979,
+      "▁Hart": 20980,
+      "▁inhab": 20981,
+      "Win": 20982,
+      "Cond": 20983,
+      "▁panels": 20984,
+      "boy": 20985,
+      "odynamic": 20986,
+      "裁判": 20987,
+      "相关法律条文": 20988,
+      "▁不": 20989,
+      "▁downtown": 20990,
+      "stdc": 20991,
+      "▁philosophy": 20992,
+      "淘宝": 20993,
+      "oton": 20994,
+      "储蓄": 20995,
+      "aceae": 20996,
+      "越来越多": 20997,
+      "iev": 20998,
+      "▁宋": 20999,
+      "前进": 21000,
+      "段时间": 21001,
+      "选答案": 21002,
+      "此前": 21003,
+      "Runtime": 21004,
+      "may": 21005,
+      "另一方面": 21006,
+      "▁Private": 21007,
+      "▁deviation": 21008,
+      "▁Moon": 21009,
+      "▁Trade": 21010,
+      "人大常委会": 21011,
+      "opens": 21012,
+      "▁aside": 21013,
+      "▁bounds": 21014,
+      "ocon": 21015,
+      "will": 21016,
+      "orses": 21017,
+      "▁elegant": 21018,
+      "▁receiver": 21019,
+      "▁afr": 21020,
+      "()`": 21021,
+      "▁residence": 21022,
+      "delet": 21023,
+      "前提": 21024,
+      "可供执行的财产线索": 21025,
+      "CPU": 21026,
+      "▁blend": 21027,
+      "诗词": 21028,
+      "▁ArrayList": 21029,
+      "▁红": 21030,
+      "火车": 21031,
+      "学名": 21032,
+      "组织的": 21033,
+      "▁wooden": 21034,
+      "摩托车": 21035,
+      "▁Hello": 21036,
+      "脱贫": 21037,
+      "▁universe": 21038,
+      "非洲": 21039,
+      "供给": 21040,
+      "▁poorly": 21041,
+      "jquery": 21042,
+      "依托": 21043,
+      "such": 21044,
+      "▁Ever": 21045,
+      "▁undergoing": 21046,
+      "党组成员": 21047,
+      "▁Comment": 21048,
+      "▁endpoint": 21049,
+      "▁weapons": 21050,
+      "▁announcement": 21051,
+      "MHz": 21052,
+      "▁Alabama": 21053,
+      "附件": 21054,
+      "▁}^{": 21055,
+      "出场": 21056,
+      "落地": 21057,
+      "是从": 21058,
+      "witzerland": 21059,
+      "ForeignKey": 21060,
+      "▁Dat": 21061,
+      "▁perf": 21062,
+      "▁proxy": 21063,
+      "spe": 21064,
+      "Just": 21065,
+      "▁combat": 21066,
+      "]*": 21067,
+      "亦未": 21068,
+      "▁Lev": 21069,
+      "第二天": 21070,
+      "▁Advis": 21071,
+      "ski": 21072,
+      "▁Federation": 21073,
+      "▁Stadium": 21074,
+      "加以": 21075,
+      "大地": 21076,
+      "理性": 21077,
+      "▁genu": 21078,
+      "▁conscious": 21079,
+      "▁enforcement": 21080,
+      "▁登记机关": 21081,
+      "rypted": 21082,
+      "▁supplied": 21083,
+      "▁prototype": 21084,
+      "olin": 21085,
+      "wave": 21086,
+      "见到": 21087,
+      "巴黎": 21088,
+      "▁preview": 21089,
+      "▁generalized": 21090,
+      "×××××": 21091,
+      "净化": 21092,
+      "会被": 21093,
+      "电梯": 21094,
+      "idad": 21095,
+      "uties": 21096,
+      "Driver": 21097,
+      "线下": 21098,
+      "Light": 21099,
+      "▁Sant": 21100,
+      "我都": 21101,
+      "之所以": 21102,
+      "四大": 21103,
+      "参赛": 21104,
+      "无关": 21105,
+      "▁configure": 21106,
+      "吃了": 21107,
+      "▁deployment": 21108,
+      "sent": 21109,
+      "的长": 21110,
+      "amous": 21111,
+      "klahoma": 21112,
+      "忘了": 21113,
+      "▁infant": 21114,
+      "第四百": 21115,
+      "Here": 21116,
+      "在北京": 21117,
+      "▁intense": 21118,
+      "经查": 21119,
+      "Fail": 21120,
+      "CRIPT": 21121,
+      "辩称": 21122,
+      "退出": 21123,
+      "Install": 21124,
+      "错了": 21125,
+      "Exp": 21126,
+      "▁feelings": 21127,
+      "Pol": 21128,
+      "▁notable": 21129,
+      "▁Han": 21130,
+      "▁ft": 21131,
+      "括号": 21132,
+      "NULL": 21133,
+      "▁resort": 21134,
+      "其他的": 21135,
+      "▁hurt": 21136,
+      "小姐": 21137,
+      "▁dip": 21138,
+      "友好": 21139,
+      "omorphism": 21140,
+      "AVE": 21141,
+      "▁unicode": 21142,
+      "▁Du": 21143,
+      "▁操作": 21144,
+      "▁tok": 21145,
+      "▁oc": 21146,
+      "借鉴": 21147,
+      "▁reads": 21148,
+      "effective": 21149,
+      "Linux": 21150,
+      "视野": 21151,
+      "progress": 21152,
+      "活跃": 21153,
+      "ROOT": 21154,
+      "chart": 21155,
+      "▁Jay": 21156,
+      "▁agricultural": 21157,
+      "建造": 21158,
+      "排放": 21159,
+      "Cap": 21160,
+      "dark": 21161,
+      "ör": 21162,
+      "▁Pier": 21163,
+      "界的": 21164,
+      "▁Json": 21165,
+      "ATED": 21166,
+      "十五日": 21167,
+      "▁pointer": 21168,
+      "ammar": 21169,
+      "▁Learn": 21170,
+      "民间借贷纠纷一案中": 21171,
+      "Validation": 21172,
+      "零件": 21173,
+      "keeper": 21174,
+      "款项": 21175,
+      "手册": 21176,
+      "▁Job": 21177,
+      "▁precisely": 21178,
+      "嫌疑": 21179,
+      "▁Indonesia": 21180,
+      "iaz": 21181,
+      "▁Cape": 21182,
+      "exchange": 21183,
+      "▁Chen": 21184,
+      "was": 21185,
+      "高级人民法院": 21186,
+      "mobile": 21187,
+      "▁attempted": 21188,
+      "党建": 21189,
+      "出具的": 21190,
+      "expr": 21191,
+      "▁Officer": 21192,
+      "▁boundaries": 21193,
+      "打破": 21194,
+      "重组": 21195,
+      "不容易": 21196,
+      "国家级": 21197,
+      "有一天": 21198,
+      "▁nucleus": 21199,
+      "▁somehow": 21200,
+      "攻坚": 21201,
+      "▁collagen": 21202,
+      "langle": 21203,
+      "▁hotels": 21204,
+      "总局": 21205,
+      "▁spray": 21206,
+      "▁timeout": 21207,
+      "第五章": 21208,
+      "▁wider": 21209,
+      "Dest": 21210,
+      "▁indoor": 21211,
+      "机电": 21212,
+      "工商行政管理局": 21213,
+      "▁defining": 21214,
+      "▁bases": 21215,
+      "▁unfort": 21216,
+      "▁improves": 21217,
+      "▁Theorem": 21218,
+      "▁amplitude": 21219,
+      "nMr": 21220,
+      "insic": 21221,
+      "分成": 21222,
+      "▁phosphorylation": 21223,
+      "已成为": 21224,
+      "EEE": 21225,
+      "▁retain": 21226,
+      "如果有": 21227,
+      "弹性": 21228,
+      "helpers": 21229,
+      "getElementById": 21230,
+      "象限": 21231,
+      "举证": 21232,
+      "▁una": 21233,
+      "▁Spect": 21234,
+      "醉酒": 21235,
+      "▁df": 21236,
+      "▁lawyers": 21237,
+      "▁dp": 21238,
+      "▁photographs": 21239,
+      "uris": 21240,
+      "执行费": 21241,
+      "urious": 21242,
+      "▁dt": 21243,
+      "▁chances": 21244,
+      "Agent": 21245,
+      "▁struggle": 21246,
+      "纽约": 21247,
+      "美好的": 21248,
+      "关系的": 21249,
+      "Contact": 21250,
+      "找不到": 21251,
+      "▁tie": 21252,
+      "尽快": 21253,
+      "▁validated": 21254,
+      "▁Az": 21255,
+      "▁metall": 21256,
+      "浦东": 21257,
+      "平方公里": 21258,
+      "▁Books": 21259,
+      "ogens": 21260,
+      "透过": 21261,
+      "posing": 21262,
+      "一本": 21263,
+      "▁lucky": 21264,
+      "打电话": 21265,
+      "Programming": 21266,
+      "▁questionnaire": 21267,
+      "▁overw": 21268,
+      "▁ski": 21269,
+      "完成劳动任务": 21270,
+      "mates": 21271,
+      "pick": 21272,
+      "nRead": 21273,
+      "▁peptides": 21274,
+      "▁Liter": 21275,
+      "西安市": 21276,
+      "tainment": 21277,
+      "据悉": 21278,
+      "DU": 21279,
+      "▁cout": 21280,
+      "▁meals": 21281,
+      "coln": 21282,
+      "养生": 21283,
+      "▁subscription": 21284,
+      "▁bic": 21285,
+      "nBl": 21286,
+      "也未": 21287,
+      "这件事": 21288,
+      "光明": 21289,
+      "移送": 21290,
+      "afka": 21291,
+      "走到": 21292,
+      "thew": 21293,
+      "给自己": 21294,
+      "渐渐": 21295,
+      "micro": 21296,
+      "hai": 21297,
+      "经本院合法传唤": 21298,
+      "▁reception": 21299,
+      "▁publishing": 21300,
+      "▁expenses": 21301,
+      "罪名成立": 21302,
+      "▁olig": 21303,
+      "▁GR": 21304,
+      "nBecause": 21305,
+      "ioxid": 21306,
+      "once": 21307,
+      "▁residual": 21308,
+      "的范围": 21309,
+      "▁symmetric": 21310,
+      "▁enhancement": 21311,
+      "过度": 21312,
+      "youtube": 21313,
+      "▁ctx": 21314,
+      "师傅": 21315,
+      "抑郁": 21316,
+      "▁bes": 21317,
+      "▁Altern": 21318,
+      "elastic": 21319,
+      "结束后": 21320,
+      "已被": 21321,
+      "LIB": 21322,
+      "▁cooperation": 21323,
+      "▁_,": 21324,
+      "▁transient": 21325,
+      "▁transformed": 21326,
+      "▁BO": 21327,
+      "▁dont": 21328,
+      "ulates": 21329,
+      "▁Da": 21330,
+      "▁cant": 21331,
+      "▁Michel": 21332,
+      "事实上": 21333,
+      "textcolor": 21334,
+      "到来": 21335,
+      "Ptr": 21336,
+      "ación": 21337,
+      "Failed": 21338,
+      "Chat": 21339,
+      "▁Affairs": 21340,
+      "▁Magazine": 21341,
+      "▁likelihood": 21342,
+      ";&": 21343,
+      "▁cul": 21344,
+      "▁anest": 21345,
+      "▁Bit": 21346,
+      "▁streaming": 21347,
+      "代表大会": 21348,
+      "▁PK": 21349,
+      "期限为": 21350,
+      "▁verified": 21351,
+      "物业管理": 21352,
+      "ordon": 21353,
+      "尴尬": 21354,
+      "▁lum": 21355,
+      "我家": 21356,
+      "的项目": 21357,
+      "▁Cover": 21358,
+      "解答即可": 21359,
+      "▁packaging": 21360,
+      "▁torch": 21361,
+      "▁arranged": 21362,
+      "likely": 21363,
+      "采纳": 21364,
+      "▁Controller": 21365,
+      "▁gonna": 21366,
+      "珍惜": 21367,
+      "▁colours": 21368,
+      "Blue": 21369,
+      "▁empirical": 21370,
+      "hl": 21371,
+      "▁deeper": 21372,
+      "▁Nev": 21373,
+      "idespread": 21374,
+      "三次": 21375,
+      "OSS": 21376,
+      "▁Gil": 21377,
+      "▁Statistics": 21378,
+      "__\":\\": 21379,
+      "▁balls": 21380,
+      "▁preventing": 21381,
+      "派出所": 21382,
+      "四条": 21383,
+      "那是": 21384,
+      "▁Factor": 21385,
+      "Keys": 21386,
+      "▁上海": 21387,
+      "输送": 21388,
+      "nToday": 21389,
+      "andra": 21390,
+      "▁Mas": 21391,
+      "从业": 21392,
+      "干预": 21393,
+      "穿着": 21394,
+      "nRel": 21395,
+      "人民法院裁定": 21396,
+      "▁SM": 21397,
+      "Holder": 21398,
+      "MODE": 21399,
+      "third": 21400,
+      "▁Spir": 21401,
+      "▁hero": 21402,
+      "▁rated": 21403,
+      "▁sixth": 21404,
+      "三星": 21405,
+      "hyper": 21406,
+      "science": 21407,
+      "turn": 21408,
+      "▁exclusively": 21409,
+      "▁по": 21410,
+      "olly": 21411,
+      "▁Hil": 21412,
+      "▁abilities": 21413,
+      "▁compiled": 21414,
+      "▁dreams": 21415,
+      "Double": 21416,
+      "Gui": 21417,
+      "▁prosec": 21418,
+      "▁explicitly": 21419,
+      "eu": 21420,
+      "poly": 21421,
+      "讲述": 21422,
+      "▁EC": 21423,
+      "▁entities": 21424,
+      "▁abnormalities": 21425,
+      "审美": 21426,
+      "不履行": 21427,
+      "▁Hollywood": 21428,
+      "▁dollar": 21429,
+      "本院已": 21430,
+      "我还": 21431,
+      "海市": 21432,
+      "可以再次申请执行": 21433,
+      "上线": 21434,
+      "extend": 21435,
+      "青山": 21436,
+      "Import": 21437,
+      "方针": 21438,
+      "财产线索": 21439,
+      "obacter": 21440,
+      "▁entitled": 21441,
+      "leges": 21442,
+      "放大": 21443,
+      "的意义": 21444,
+      "的市场": 21445,
+      "▁tries": 21446,
+      "▁predominant": 21447,
+      "▁transparent": 21448,
+      "定价": 21449,
+      "Stat": 21450,
+      "rose": 21451,
+      "▁Stream": 21452,
+      "▁impl": 21453,
+      "▁achieving": 21454,
+      "sr": 21455,
+      "clock": 21456,
+      "▁safely": 21457,
+      "执行工作": 21458,
+      "▁Body": 21459,
+      "▁aggressive": 21460,
+      "▁vulnerable": 21461,
+      "empl": 21462,
+      "nWomen": 21463,
+      "少的": 21464,
+      "▁科": 21465,
+      "▁affinity": 21466,
+      "▁substitute": 21467,
+      "▁surname": 21468,
+      "iating": 21469,
+      "犯罪事实": 21470,
+      "▁literally": 21471,
+      "社会信用代码": 21472,
+      "以往": 21473,
+      "含义": 21474,
+      "iliary": 21475,
+      "Controllers": 21476,
+      "矩阵": 21477,
+      "都知道": 21478,
+      "试试": 21479,
+      "upper": 21480,
+      "协同": 21481,
+      "评分": 21482,
+      "acchar": 21483,
+      "▁midfield": 21484,
+      "燃烧": 21485,
+      "▁fixes": 21486,
+      "精致": 21487,
+      "▁Challenge": 21488,
+      "▁theatre": 21489,
+      "▁Castle": 21490,
+      "▁acceptable": 21491,
+      "▁resume": 21492,
+      "户外": 21493,
+      "▁responded": 21494,
+      "anim": 21495,
+      "天使": 21496,
+      "▁Sales": 21497,
+      "▁donor": 21498,
+      "▁oxidative": 21499,
+      "爱心": 21500,
+      "▁posterior": 21501,
+      "tlog": 21502,
+      "▁Austin": 21503,
+      "▁},": 21504,
+      "ognition": 21505,
+      "ropolitan": 21506,
+      "▁typing": 21507,
+      "nDisc": 21508,
+      "全世界": 21509,
+      "▁emissions": 21510,
+      "▁diabetic": 21511,
+      "▁fluorescence": 21512,
+      "说不": 21513,
+      "▁nas": 21514,
+      "terminal": 21515,
+      "▁Category": 21516,
+      "过敏": 21517,
+      "原判": 21518,
+      "并有": 21519,
+      "▁Fa": 21520,
+      "第十三": 21521,
+      "▁Doc": 21522,
+      "▁汉族": 21523,
+      "分工": 21524,
+      "差别": 21525,
+      "▁onc": 21526,
+      "NotFound": 21527,
+      "▁Different": 21528,
+      "HOST": 21529,
+      "法律服务所": 21530,
+      "genic": 21531,
+      "bled": 21532,
+      "Ow": 21533,
+      "Star": 21534,
+      "contract": 21535,
+      "淀粉": 21536,
+      "▁electroph": 21537,
+      "MOD": 21538,
+      "歌词": 21539,
+      "▁Sri": 21540,
+      "▁placebo": 21541,
+      "▁nutrition": 21542,
+      "到现在": 21543,
+      "牙齿": 21544,
+      "冬天": 21545,
+      "athon": 21546,
+      "babel": 21547,
+      "▁cod": 21548,
+      "▁Season": 21549,
+      "为你": 21550,
+      ":`": 21551,
+      "Sur": 21552,
+      "马克思": 21553,
+      "▁cameras": 21554,
+      "会出现": 21555,
+      "▁cow": 21556,
+      "积极参加劳动": 21557,
+      "等问题": 21558,
+      "▁Som": 21559,
+      "SK": 21560,
+      "原来的": 21561,
+      "pipe": 21562,
+      "¶\\": 21563,
+      "水电": 21564,
+      "法律工作者": 21565,
+      "▁notion": 21566,
+      "CLUDING": 21567,
+      "心理学": 21568,
+      "▁optimized": 21569,
+      "二维": 21570,
+      "眼前": 21571,
+      "▁intact": 21572,
+      "parameter": 21573,
+      "apple": 21574,
+      "统一社会信用代码": 21575,
+      "▁operates": 21576,
+      "冒险": 21577,
+      "▁Nigeria": 21578,
+      "▁telephone": 21579,
+      "打工": 21580,
+      "nPol": 21581,
+      "olves": 21582,
+      "▁Liver": 21583,
+      "▁yard": 21584,
+      "mate": 21585,
+      "▁notification": 21586,
+      "知识的选择题": 21587,
+      "eto": 21588,
+      "Fin": 21589,
+      "附录": 21590,
+      "▁(*": 21591,
+      "本院在执行申请执行人": 21592,
+      "nama": 21593,
+      "resentation": 21594,
+      "登记结婚": 21595,
+      "最后一": 21596,
+      "行情": 21597,
+      "月二十八日": 21598,
+      "导向": 21599,
+      "▁cats": 21600,
+      "spl": 21601,
+      "DEBUG": 21602,
+      "onical": 21603,
+      "/>": 21604,
+      "ntil": 21605,
+      "▁frozen": 21606,
+      "ochemistry": 21607,
+      "有序": 21608,
+      "▁surviv": 21609,
+      "▁elevation": 21610,
+      "摘要": 21611,
+      "XML": 21612,
+      "▁chairman": 21613,
+      "▁recruitment": 21614,
+      "▁Travel": 21615,
+      "▁Vice": 21616,
+      "Pal": 21617,
+      "也就是说": 21618,
+      "▁lesson": 21619,
+      "▁augment": 21620,
+      "陆续": 21621,
+      "每人": 21622,
+      "▁predictions": 21623,
+      "你能": 21624,
+      "南京市": 21625,
+      "▁investigating": 21626,
+      "▁spiritual": 21627,
+      "我们将": 21628,
+      "▁tons": 21629,
+      "▁Hur": 21630,
+      "▁Target": 21631,
+      "encoding": 21632,
+      "INGS": 21633,
+      "▁intestinal": 21634,
+      "孤独": 21635,
+      "nie": 21636,
+      "▁massage": 21637,
+      "▁relating": 21638,
+      "历程": 21639,
+      "执行完毕": 21640,
+      "▁nurses": 21641,
+      "绩效": 21642,
+      "数值": 21643,
+      "▁directors": 21644,
+      "▁PhD": 21645,
+      "▁类别": 21646,
+      "▁tar": 21647,
+      "的通知": 21648,
+      "升高": 21649,
+      "为准": 21650,
+      "来讲": 21651,
+      "的状态": 21652,
+      "日以": 21653,
+      "▁地理位置": 21654,
+      "谅解": 21655,
+      "▁Representatives": 21656,
+      "的和": 21657,
+      "▁forming": 21658,
+      "▁undergo": 21659,
+      "▁Jam": 21660,
+      "特定": 21661,
+      "▁enabling": 21662,
+      "▁Side": 21663,
+      "MMMM": 21664,
+      "mart": 21665,
+      "Real": 21666,
+      "▁explos": 21667,
+      "oco": 21668,
+      "留在": 21669,
+      "▁agreements": 21670,
+      "▁什么是": 21671,
+      "那一": 21672,
+      "▁incredibly": 21673,
+      "erk": 21674,
+      ")$$\\": 21675,
+      "orse": 21676,
+      "▁malignant": 21677,
+      "小程序": 21678,
+      "Servlet": 21679,
+      "orus": 21680,
+      "conv": 21681,
+      "nik": 21682,
+      "▁attitude": 21683,
+      "块钱": 21684,
+      "dominal": 21685,
+      "夏季": 21686,
+      "▁attachment": 21687,
+      "取决": 21688,
+      "的生产": 21689,
+      "registry": 21690,
+      "nPersonal": 21691,
+      "▁coaching": 21692,
+      "▁replacing": 21693,
+      "▁Autom": 21694,
+      "ortheast": 21695,
+      "对策": 21696,
+      "▁Stop": 21697,
+      ")<": 21698,
+      "驳回原告": 21699,
+      "关键词": 21700,
+      "ardo": 21701,
+      "▁Newton": 21702,
+      "▁flights": 21703,
+      "▁painted": 21704,
+      "nelse": 21705,
+      "▁Marc": 21706,
+      "station": 21707,
+      "————————": 21708,
+      "UNT": 21709,
+      "你想": 21710,
+      "▁Kenn": 21711,
+      "期望": 21712,
+      "两国": 21713,
+      "▁proposes": 21714,
+      "ascin": 21715,
+      "克斯": 21716,
+      "人人": 21717,
+      "科技大学": 21718,
+      "▁pocket": 21719,
+      "tration": 21720,
+      "人工智能": 21721,
+      "的中国": 21722,
+      "▁advoc": 21723,
+      "的所有": 21724,
+      "▁bios": 21725,
+      "▁boss": 21726,
+      "周末": 21727,
+      "从选项中选出": 21728,
+      "相机": 21729,
+      "在新": 21730,
+      "▁harder": 21731,
+      "posal": 21732,
+      "焊接": 21733,
+      "▁fraud": 21734,
+      "▁{};\\": 21735,
+      "/$": 21736,
+      "后再": 21737,
+      "为您": 21738,
+      "oro": 21739,
+      "▁flour": 21740,
+      "orms": 21741,
+      "▁workshops": 21742,
+      "▁databases": 21743,
+      "让她": 21744,
+      "万吨": 21745,
+      "比亚": 21746,
+      "icut": 21747,
+      "▁Dave": 21748,
+      "▁Bry": 21749,
+      "▁bomb": 21750,
+      "平移": 21751,
+      "回事": 21752,
+      "▁Sometimes": 21753,
+      "你不": 21754,
+      "▁recurrence": 21755,
+      "hou": 21756,
+      "azines": 21757,
+      "▁constructor": 21758,
+      "▁fmt": 21759,
+      "▁sew": 21760,
+      "▁pace": 21761,
+      "▁morbidity": 21762,
+      "bird": 21763,
+      "Theme": 21764,
+      "Gamma": 21765,
+      "多久": 21766,
+      "▁Though": 21767,
+      "▁Speed": 21768,
+      "lessly": 21769,
+      "齐全": 21770,
+      "▁Having": 21771,
+      "凭证": 21772,
+      "▁Hills": 21773,
+      "▁ecosystem": 21774,
+      "和发展": 21775,
+      "轻轻": 21776,
+      "▁neighbour": 21777,
+      "下一步": 21778,
+      "school": 21779,
+      "▁writes": 21780,
+      "防范": 21781,
+      "▁archive": 21782,
+      "婚礼": 21783,
+      "▁nod": 21784,
+      "nname": 21785,
+      "使其": 21786,
+      "▁fibr": 21787,
+      "▁modifications": 21788,
+      "可怜": 21789,
+      "elson": 21790,
+      "法官": 21791,
+      "实话": 21792,
+      "nFr": 21793,
+      "▁electrons": 21794,
+      "▁argparse": 21795,
+      "▁blocked": 21796,
+      "玫瑰": 21797,
+      "NR": 21798,
+      "▁jet": 21799,
+      "structure": 21800,
+      "▁Docker": 21801,
+      "includes": 21802,
+      "加坡": 21803,
+      "就开始": 21804,
+      "▁resist": 21805,
+      "▁bol": 21806,
+      "jack": 21807,
+      "路口": 21808,
+      "波动": 21809,
+      "了诉讼": 21810,
+      "▁gre": 21811,
+      "▁barriers": 21812,
+      "当庭": 21813,
+      "很久": 21814,
+      "实战": 21815,
+      "▁producers": 21816,
+      "一带": 21817,
+      "在开庭": 21818,
+      "初中文化": 21819,
+      "标准化": 21820,
+      "节日": 21821,
+      "▁minimize": 21822,
+      "▁immigr": 21823,
+      "由被告承担": 21824,
+      "Threading": 21825,
+      "▁theories": 21826,
+      "▁nitrogen": 21827,
+      "factor": 21828,
+      "层的": 21829,
+      "▁composer": 21830,
+      "▁destroyed": 21831,
+      "延伸": 21832,
+      "prev": 21833,
+      "YP": 21834,
+      "渗透": 21835,
+      "注意事项": 21836,
+      "▁ensures": 21837,
+      "ailand": 21838,
+      "▁organisations": 21839,
+      "▁Organization": 21840,
+      "显著": 21841,
+      "经销": 21842,
+      "▁Host": 21843,
+      "▁XML": 21844,
+      "▁Kitchen": 21845,
+      "借款合同纠纷一案": 21846,
+      "orporated": 21847,
+      "▁isolation": 21848,
+      "Rob": 21849,
+      "▁(<": 21850,
+      "▁Simon": 21851,
+      "NotNull": 21852,
+      "igious": 21853,
+      "最初": 21854,
+      "character": 21855,
+      "▁reflection": 21856,
+      "ouver": 21857,
+      "县公安局": 21858,
+      "点评": 21859,
+      "}:": 21860,
+      "桃花": 21861,
+      "▁Jane": 21862,
+      "ecture": 21863,
+      "driven": 21864,
+      "的权利": 21865,
+      "nHigh": 21866,
+      "演讲": 21867,
+      "组成的": 21868,
+      "anka": 21869,
+      "▁parliament": 21870,
+      "▁≤": 21871,
+      "▁软件": 21872,
+      "▁clim": 21873,
+      "高速公路": 21874,
+      "向本院提起": 21875,
+      "▁titled": 21876,
+      "▁projection": 21877,
+      "客人": 21878,
+      "▁gravity": 21879,
+      "▁Mand": 21880,
+      "▁sees": 21881,
+      "▁prohib": 21882,
+      "deep": 21883,
+      "等差": 21884,
+      "ás": 21885,
+      "▁Files": 21886,
+      "▁sustained": 21887,
+      "▁arrays": 21888,
+      "▁ultras": 21889,
+      "照时间": 21890,
+      "▁Notes": 21891,
+      "isode": 21892,
+      "▁Ur": 21893,
+      "badge": 21894,
+      "▁Eag": 21895,
+      "effect": 21896,
+      "以下的": 21897,
+      "未经": 21898,
+      "▁enjoying": 21899,
+      "fp": 21900,
+      "essary": 21901,
+      "▁recipes": 21902,
+      "一对": 21903,
+      "learning": 21904,
+      "▁Bitcoin": 21905,
+      "年纪": 21906,
+      "精力": 21907,
+      "ль": 21908,
+      "▁cooling": 21909,
+      "▁defines": 21910,
+      "平等": 21911,
+      "▁villages": 21912,
+      "校区": 21913,
+      "▁prev": 21914,
+      "amy": 21915,
+      "插入": 21916,
+      "▁Dor": 21917,
+      "▁overwhel": 21918,
+      "▁territory": 21919,
+      "cstring": 21920,
+      "▁pounds": 21921,
+      "▁Leon": 21922,
+      "▁belongs": 21923,
+      "etition": 21924,
+      "激素": 21925,
+      "点赞": 21926,
+      "React": 21927,
+      "中共党员": 21928,
+      "登记成立": 21929,
+      "▁anniversary": 21930,
+      "▁odds": 21931,
+      "▁Patrick": 21932,
+      "贝尔": 21933,
+      "ntry": 21934,
+      "nInst": 21935,
+      "▁Storage": 21936,
+      "audio": 21937,
+      "▁failing": 21938,
+      "▁holidays": 21939,
+      "guide": 21940,
+      "incial": 21941,
+      "▁Without": 21942,
+      "nDem": 21943,
+      "▁pursue": 21944,
+      "Move": 21945,
+      "▁passengers": 21946,
+      "助力": 21947,
+      "camp": 21948,
+      "minus": 21949,
+      "取决于": 21950,
+      "▁buttons": 21951,
+      "▁Philippines": 21952,
+      "Ill": 21953,
+      "forming": 21954,
+      "invalid": 21955,
+      "损坏": 21956,
+      "gba": 21957,
+      "▁oven": 21958,
+      "▁wake": 21959,
+      "的性质": 21960,
+      "▁Large": 21961,
+      "▁irrad": 21962,
+      "▁borrow": 21963,
+      "假如": 21964,
+      ")\"\\": 21965,
+      "nlayout": 21966,
+      "▁$$(": 21967,
+      "umatic": 21968,
+      "▁distribute": 21969,
+      "selector": 21970,
+      "▁Dead": 21971,
+      "▁scroll": 21972,
+      "▁Character": 21973,
+      "导航": 21974,
+      "plain": 21975,
+      "Fields": 21976,
+      "▁Bus": 21977,
+      "olars": 21978,
+      "▁tied": 21979,
+      "报酬": 21980,
+      "▁Emb": 21981,
+      "程度上": 21982,
+      "▁mortgage": 21983,
+      "colors": 21984,
+      "副主席": 21985,
+      "高血压": 21986,
+      "▁Come": 21987,
+      "供电": 21988,
+      "Why": 21989,
+      "Stop": 21990,
+      "hide": 21991,
+      "▁pas": 21992,
+      "([\\'": 21993,
+      "nSportspeople": 21994,
+      "▁Girls": 21995,
+      "inflammatory": 21996,
+      "oup": 21997,
+      "▁bonds": 21998,
+      "MPL": 21999,
+      "▁qualitative": 22000,
+      "▁argue": 22001,
+      "abb": 22002,
+      "▁restoration": 22003,
+      "omon": 22004,
+      "ensively": 22005,
+      "▁purchasing": 22006,
+      "▁Vegas": 22007,
+      "时间内": 22008,
+      "nch": 22009,
+      "做一个": 22010,
+      "Ask": 22011,
+      "伴随着": 22012,
+      "frequency": 22013,
+      "\"},{\"": 22014,
+      "▁PUB": 22015,
+      "▁casual": 22016,
+      "去除": 22017,
+      "shaped": 22018,
+      "组织机构代码": 22019,
+      "遇见": 22020,
+      "某的": 22021,
+      "kh": 22022,
+      "权力": 22023,
+      "▁indicator": 22024,
+      "▁Lyn": 22025,
+      "straint": 22026,
+      "▁Collect": 22027,
+      "更大": 22028,
+      "ndescribe": 22029,
+      "小时候": 22030,
+      "真相": 22031,
+      "到庭参加了诉讼": 22032,
+      "▁Foreign": 22033,
+      "▁prost": 22034,
+      "▁Duke": 22035,
+      "▁Follow": 22036,
+      "▁Seattle": 22037,
+      "▁widespread": 22038,
+      "发展有限公司": 22039,
+      "并于": 22040,
+      "阶级": 22041,
+      "▁surveillance": 22042,
+      "民事诉讼": 22043,
+      "More": 22044,
+      "看守": 22045,
+      "nadd": 22046,
+      "▁modulation": 22047,
+      "Pattern": 22048,
+      "▁susceptibility": 22049,
+      "▁simplify": 22050,
+      "和计划生育局": 22051,
+      "▁persistent": 22052,
+      "有关部门": 22053,
+      "piece": 22054,
+      "documentclass": 22055,
+      "ssh": 22056,
+      "▁Blu": 22057,
+      "aying": 22058,
+      "幅度": 22059,
+      "▁Range": 22060,
+      "▁riding": 22061,
+      "▁[],\\": 22062,
+      "▁initiatives": 22063,
+      "多少钱": 22064,
+      "warning": 22065,
+      "该公司董事长": 22066,
+      "▁parish": 22067,
+      "▁listings": 22068,
+      "onde": 22069,
+      "▁impaired": 22070,
+      "▁consisted": 22071,
+      "眼镜": 22072,
+      "▁fibro": 22073,
+      "▁divor": 22074,
+      "▁Boot": 22075,
+      "旁边": 22076,
+      "▁processor": 22077,
+      "▁extracellular": 22078,
+      "失信被执行人名单": 22079,
+      "▁accompany": 22080,
+      "Attributes": 22081,
+      "atio": 22082,
+      "▁Details": 22083,
+      "abi": 22084,
+      "最重要的": 22085,
+      "ippi": 22086,
+      "owed": 22087,
+      "热门": 22088,
+      "未到庭参加诉讼": 22089,
+      "nEven": 22090,
+      "奉献": 22091,
+      "防火": 22092,
+      "nPolit": 22093,
+      "need": 22094,
+      "▁EL": 22095,
+      "统计学意义": 22096,
+      "▁Finance": 22097,
+      "天空": 22098,
+      "predict": 22099,
+      "▁elsewhere": 22100,
+      "^^": 22101,
+      "ENCE": 22102,
+      "Loading": 22103,
+      "▁associate": 22104,
+      "▁emerged": 22105,
+      "相处": 22106,
+      "COMM": 22107,
+      "戏剧": 22108,
+      "nTax": 22109,
+      "song": 22110,
+      "▁kinetic": 22111,
+      "BJECT": 22112,
+      "▁initialize": 22113,
+      "▁remodel": 22114,
+      "超声": 22115,
+      "▁canvas": 22116,
+      "▁compilation": 22117,
+      "退还": 22118,
+      "▁competitions": 22119,
+      "婚姻法": 22120,
+      "tempt": 22121,
+      "eep": 22122,
+      "旅客": 22123,
+      "▁buyers": 22124,
+      "MAG": 22125,
+      "处处": 22126,
+      "期间不": 22127,
+      "hang": 22128,
+      "生态环境": 22129,
+      "初期": 22130,
+      "enez": 22131,
+      "▁sizeof": 22132,
+      "▁invasive": 22133,
+      "宁波市": 22134,
+      "列出": 22135,
+      "▁有限责任公司": 22136,
+      "Initial": 22137,
+      "▁Fourier": 22138,
+      "Registry": 22139,
+      "煤炭": 22140,
+      "▁小说": 22141,
+      "inea": 22142,
+      "tre": 22143,
+      "▁oxidation": 22144,
+      "▁Bible": 22145,
+      "▁sid": 22146,
+      "▁laptop": 22147,
+      "保温": 22148,
+      "Year": 22149,
+      "约束": 22150,
+      "▁Balt": 22151,
+      "进了": 22152,
+      "elenium": 22153,
+      "▁FI": 22154,
+      "cdn": 22155,
+      "warn": 22156,
+      "anion": 22157,
+      "附属": 22158,
+      "▁Dur": 22159,
+      "▁representatives": 22160,
+      "▁gaming": 22161,
+      "产地": 22162,
+      "▁WE": 22163,
+      "▁Tokyo": 22164,
+      "女朋友": 22165,
+      "▁preferences": 22166,
+      "和你": 22167,
+      "交警": 22168,
+      "onomous": 22169,
+      "家属": 22170,
+      "▁GA": 22171,
+      "第六十七条": 22172,
+      "守护": 22173,
+      "Collect": 22174,
+      "▁voters": 22175,
+      "▁selecting": 22176,
+      "▁rolling": 22177,
+      "▁PE": 22178,
+      "▁Plant": 22179,
+      "These": 22180,
+      "translation": 22181,
+      "服从": 22182,
+      "市场上": 22183,
+      "stroke": 22184,
+      "台北": 22185,
+      "日立案执行": 22186,
+      "高峰": 22187,
+      "美丽的": 22188,
+      "})$": 22189,
+      "▁Jacob": 22190,
+      "全县": 22191,
+      "只不过": 22192,
+      "花费": 22193,
+      "本案在执行过程中": 22194,
+      "协作": 22195,
+      "places": 22196,
+      "火锅": 22197,
+      "种种": 22198,
+      "家用": 22199,
+      "building": 22200,
+      "▁郑": 22201,
+      "▁handled": 22202,
+      "问我": 22203,
+      "▁Additional": 22204,
+      "电缆": 22205,
+      "\"\\\\": 22206,
+      "▁Nik": 22207,
+      "alam": 22208,
+      "▁Roy": 22209,
+      "ockets": 22210,
+      "▁Success": 22211,
+      "▁inspection": 22212,
+      "▁\"\",\\": 22213,
+      "便捷": 22214,
+      "科普": 22215,
+      "▁amongst": 22216,
+      "查找": 22217,
+      "▁ARE": 22218,
+      "▁contacts": 22219,
+      "arith": 22220,
+      "▁spectrom": 22221,
+      "VO": 22222,
+      "▁Particip": 22223,
+      "骨折": 22224,
+      "请给出": 22225,
+      "enh": 22226,
+      "rban": 22227,
+      "▁triple": 22228,
+      "▁GET": 22229,
+      "▁Holy": 22230,
+      "▁Dak": 22231,
+      "environ": 22232,
+      "acles": 22233,
+      "▁Module": 22234,
+      "abstract": 22235,
+      "▁indicators": 22236,
+      "紧密": 22237,
+      "policy": 22238,
+      "充足": 22239,
+      "ViewModel": 22240,
+      "▁hepatic": 22241,
+      "}{(": 22242,
+      "是我们": 22243,
+      "▁eq": 22244,
+      "社保": 22245,
+      "鉴于": 22246,
+      "复杂度": 22247,
+      "▁recru": 22248,
+      "气氛": 22249,
+      "▁HP": 22250,
+      "▁atomic": 22251,
+      "▁thyroid": 22252,
+      "▁trem": 22253,
+      "▁longitudinal": 22254,
+      "▁suggestion": 22255,
+      "家中": 22256,
+      "eval": 22257,
+      "▁maximal": 22258,
+      "被称为": 22259,
+      "▁traject": 22260,
+      "民国": 22261,
+      "flix": 22262,
+      "nMETHODS": 22263,
+      "布置": 22264,
+      "第一百四十五条第一款之规定": 22265,
+      "▁Atlanta": 22266,
+      "four": 22267,
+      "大战": 22268,
+      "ackets": 22269,
+      "arguments": 22270,
+      "下跌": 22271,
+      "幻想": 22272,
+      "PIO": 22273,
+      "四边形": 22274,
+      "▁identifier": 22275,
+      "*/": 22276,
+      "▁郭": 22277,
+      "▁deposit": 22278,
+      "▁elementary": 22279,
+      "elfare": 22280,
+      "▁Explanation": 22281,
+      "复杂的": 22282,
+      "考查了": 22283,
+      "本院予以确认": 22284,
+      "填写": 22285,
+      "侵权": 22286,
+      "▁bull": 22287,
+      "假期": 22288,
+      "艺术家": 22289,
+      "▁exercises": 22290,
+      "还是要": 22291,
+      "Make": 22292,
+      "′(": 22293,
+      "nFe": 22294,
+      "Converter": 22295,
+      "▁volunteer": 22296,
+      "少许": 22297,
+      "王子": 22298,
+      "sembl": 22299,
+      "retion": 22300,
+      "isp": 22301,
+      "eland": 22302,
+      "▁Tony": 22303,
+      "}}$\\": 22304,
+      "给人": 22305,
+      "▁Assistant": 22306,
+      "陷入": 22307,
+      "iscal": 22308,
+      "Person": 22309,
+      "首都": 22310,
+      "▁gut": 22311,
+      "oting": 22312,
+      "▁promoter": 22313,
+      "不违反": 22314,
+      "是真的": 22315,
+      "terr": 22316,
+      "▁crystall": 22317,
+      "▁Ger": 22318,
+      "quote": 22319,
+      "▁religion": 22320,
+      "stderr": 22321,
+      "建立了": 22322,
+      "ić": 22323,
+      "后悔": 22324,
+      "inine": 22325,
+      "宣判": 22326,
+      "rivacy": 22327,
+      "rophic": 22328,
+      "overnment": 22329,
+      "有钱": 22330,
+      "月底": 22331,
+      "▁revision": 22332,
+      "▁Ty": 22333,
+      "▁JavaScript": 22334,
+      "▁recordings": 22335,
+      "▁BB": 22336,
+      "和解协议": 22337,
+      "▁Temple": 22338,
+      "库存": 22339,
+      "琐事": 22340,
+      "▁∴": 22341,
+      "▁completing": 22342,
+      "▁chromosome": 22343,
+      "▁Stre": 22344,
+      "▁romantic": 22345,
+      "同一个": 22346,
+      "指令": 22347,
+      "微笑": 22348,
+      "erable": 22349,
+      "▁发照时间": 22350,
+      "▁cotton": 22351,
+      "▁kilomet": 22352,
+      "▁independence": 22353,
+      "▁gastric": 22354,
+      "▁derive": 22355,
+      "▁Credit": 22356,
+      "osin": 22357,
+      "其行为已构成": 22358,
+      "▁overse": 22359,
+      "lee": 22360,
+      "▁competing": 22361,
+      "的教学": 22362,
+      "鉴定表": 22363,
+      "▁acoustic": 22364,
+      "mq": 22365,
+      "一遍": 22366,
+      "本案按": 22367,
+      "全书": 22368,
+      "惊喜": 22369,
+      "iang": 22370,
+      "▁podcast": 22371,
+      "▁authentication": 22372,
+      "▁Poly": 22373,
+      "ogy": 22374,
+      "opsis": 22375,
+      "▁Switzerland": 22376,
+      "插件": 22377,
+      "▁balanced": 22378,
+      "二份": 22379,
+      "试点": 22380,
+      "▁Anderson": 22381,
+      "▁arrangement": 22382,
+      "Score": 22383,
+      "▁Products": 22384,
+      "▁Bureau": 22385,
+      "▁activate": 22386,
+      "▁squares": 22387,
+      "▁LIMIT": 22388,
+      "▁networking": 22389,
+      "▁Cov": 22390,
+      "▁Moscow": 22391,
+      "▁defect": 22392,
+      "面粉": 22393,
+      "细致": 22394,
+      "▁Ms": 22395,
+      "二被告": 22396,
+      "指示": 22397,
+      "acion": 22398,
+      "▁novels": 22399,
+      "说我": 22400,
+      "财物": 22401,
+      "太原": 22402,
+      "▁Iron": 22403,
+      "若干问题的解释": 22404,
+      "度假": 22405,
+      "▁acquire": 22406,
+      "脸上": 22407,
+      "bench": 22408,
+      "▁remarkable": 22409,
+      "Sql": 22410,
+      "频繁": 22411,
+      "igg": 22412,
+      "▁wrapper": 22413,
+      "善良": 22414,
+      "▁bills": 22415,
+      "▁marrow": 22416,
+      "▁interference": 22417,
+      "▁Fred": 22418,
+      "▁exploration": 22419,
+      "▁ov": 22420,
+      "eslint": 22421,
+      "一声": 22422,
+      "▁\\'%": 22423,
+      "▁journals": 22424,
+      "olip": 22425,
+      "成效": 22426,
+      "▁cru": 22427,
+      "assertTrue": 22428,
+      "oT": 22429,
+      "域名": 22430,
+      "对面": 22431,
+      "aly": 22432,
+      "▁Authors": 22433,
+      "jobs": 22434,
+      "rated": 22435,
+      "nSw": 22436,
+      "▁rehabilitation": 22437,
+      "mus": 22438,
+      "日提出": 22439,
+      "担当": 22440,
+      "第二百四十二条": 22441,
+      "▁dé": 22442,
+      "▁Images": 22443,
+      "▁Pred": 22444,
+      "▁Ali": 22445,
+      "▁drives": 22446,
+      "▁inhibitory": 22447,
+      "注销": 22448,
+      "第二个": 22449,
+      "LETE": 22450,
+      "Compar": 22451,
+      "正常的": 22452,
+      "只需": 22453,
+      "▁opposition": 22454,
+      "▁Kit": 22455,
+      "Language": 22456,
+      "▁intellectual": 22457,
+      "AX": 22458,
+      "▁facial": 22459,
+      "▁accomplished": 22460,
+      "Inject": 22461,
+      "正义": 22462,
+      "▁解释": 22463,
+      "判定": 22464,
+      "uru": 22465,
+      "忽略": 22466,
+      "必要的": 22467,
+      "▁skilled": 22468,
+      "▁feasibility": 22469,
+      "履历": 22470,
+      "接待": 22471,
+      "信息技术": 22472,
+      "▁Greece": 22473,
+      "微博热度": 22474,
+      "并发症": 22475,
+      "▁cabinet": 22476,
+      "▁Fif": 22477,
+      "地说": 22478,
+      "nConsider": 22479,
+      "冷却": 22480,
+      "▁Mix": 22481,
+      "▁OP": 22482,
+      "」\\": 22483,
+      "▁COPY": 22484,
+      "▁rum": 22485,
+      "▁Editor": 22486,
+      "ugu": 22487,
+      "▁sensory": 22488,
+      "得到的": 22489,
+      "▁Applications": 22490,
+      "▁Made": 22491,
+      "mith": 22492,
+      "URI": 22493,
+      "▁valve": 22494,
+      "▁invite": 22495,
+      "alone": 22496,
+      "的原则": 22497,
+      "▁regime": 22498,
+      "创始": 22499,
+      "信仰": 22500,
+      "逝世": 22501,
+      "▁coin": 22502,
+      "itories": 22503,
+      "ationally": 22504,
+      "第三方": 22505,
+      "备案": 22506,
+      "▁upgrad": 22507,
+      "第四节": 22508,
+      "▁drinks": 22509,
+      "早已": 22510,
+      "▁charset": 22511,
+      "/\"": 22512,
+      "Prov": 22513,
+      "▁sexy": 22514,
+      "自身的": 22515,
+      "▁formats": 22516,
+      "▁Button": 22517,
+      "▁ES": 22518,
+      "剩下": 22519,
+      "],\"": 22520,
+      "就被": 22521,
+      "izational": 22522,
+      "和他": 22523,
+      "▁Allen": 22524,
+      "vare": 22525,
+      "▁shoulder": 22526,
+      "的确": 22527,
+      "nBel": 22528,
+      "▁urinary": 22529,
+      "▁Display": 22530,
+      "dj": 22531,
+      "productive": 22532,
+      "ijing": 22533,
+      "lated": 22534,
+      "▁Moore": 22535,
+      "Every": 22536,
+      "▁strictly": 22537,
+      "▁finest": 22538,
+      "认同": 22539,
+      "转化为": 22540,
+      "▁aest": 22541,
+      "均有": 22542,
+      "▁urine": 22543,
+      "MAP": 22544,
+      "▁motivation": 22545,
+      "samples": 22546,
+      "边长": 22547,
+      "amide": 22548,
+      "amazon": 22549,
+      "▁cater": 22550,
+      "▁contributes": 22551,
+      "incess": 22552,
+      "恶心": 22553,
+      "▁Consult": 22554,
+      "▁Partners": 22555,
+      "▁validity": 22556,
+      "▁gem": 22557,
+      "提请": 22558,
+      "题材": 22559,
+      "▁tape": 22560,
+      "市场监督管理局": 22561,
+      "▁adoption": 22562,
+      "版权": 22563,
+      "投影": 22564,
+      "▁dramatic": 22565,
+      "▁mothers": 22566,
+      "▁relate": 22567,
+      "ACH": 22568,
+      "▁海": 22569,
+      "IDE": 22570,
+      "大脑": 22571,
+      "普通的": 22572,
+      "▁Course": 22573,
+      "真心": 22574,
+      "three": 22575,
+      "pus": 22576,
+      "istical": 22577,
+      "优雅": 22578,
+      "防御": 22579,
+      "▁ISO": 22580,
+      "▁agon": 22581,
+      "▁coinc": 22582,
+      "▁alignment": 22583,
+      "▁SW": 22584,
+      "▁subgroup": 22585,
+      "泰国": 22586,
+      "▁equity": 22587,
+      "符合有关法律规定": 22588,
+      "▁viruses": 22589,
+      "▁overlook": 22590,
+      "也将": 22591,
+      "绍兴": 22592,
+      "通风": 22593,
+      "▁utilization": 22594,
+      "▁continuously": 22595,
+      "液体": 22596,
+      "shadow": 22597,
+      "古典": 22598,
+      "arding": 22599,
+      "曾任": 22600,
+      "tery": 22601,
+      "这场": 22602,
+      "▁paragraph": 22603,
+      "nGener": 22604,
+      "qq": 22605,
+      "号为": 22606,
+      "▁survive": 22607,
+      "的具体": 22608,
+      "的目标": 22609,
+      "ustralian": 22610,
+      "日志": 22611,
+      "阳性": 22612,
+      "agan": 22613,
+      "History": 22614,
+      "▁departments": 22615,
+      "▁membranes": 22616,
+      "scene": 22617,
+      "▁periodic": 22618,
+      "▁booking": 22619,
+      "idi": 22620,
+      "▁Malaysia": 22621,
+      "这种情况": 22622,
+      "去做": 22623,
+      "▁Quick": 22624,
+      "除法": 22625,
+      "格兰": 22626,
+      "▁Brun": 22627,
+      "想起": 22628,
+      "optim": 22629,
+      "拿着": 22630,
+      "Spring": 22631,
+      "代替": 22632,
+      "diction": 22633,
+      "▁boards": 22634,
+      "茶叶": 22635,
+      "的样子": 22636,
+      "但是我": 22637,
+      "▁sha": 22638,
+      "两点": 22639,
+      "民政": 22640,
+      "两天": 22641,
+      "^*": 22642,
+      "dashboard": 22643,
+      "ivities": 22644,
+      "▁oct": 22645,
+      "▁Barn": 22646,
+      "categories": 22647,
+      "▁entrepreneur": 22648,
+      "ogr": 22649,
+      "Based": 22650,
+      "▁Enjoy": 22651,
+      "定了": 22652,
+      "▁Oklahoma": 22653,
+      "▁preference": 22654,
+      "▁smartphone": 22655,
+      "Ms": 22656,
+      "全新的": 22657,
+      "EY": 22658,
+      "▁antioxid": 22659,
+      "regulation": 22660,
+      "谈判": 22661,
+      "▁Mrs": 22662,
+      "▁slide": 22663,
+      "▁Ideas": 22664,
+      "▁Neuro": 22665,
+      "nView": 22666,
+      "inus": 22667,
+      "▁matplotlib": 22668,
+      "nRoute": 22669,
+      "帮忙": 22670,
+      "▁retention": 22671,
+      "判决发生法律效力后": 22672,
+      "▁combines": 22673,
+      "王某某": 22674,
+      "▁headed": 22675,
+      "一台": 22676,
+      "annotations": 22677,
+      "▁>\\": 22678,
+      "modified": 22679,
+      "不说": 22680,
+      "▁curious": 22681,
+      "▁limitation": 22682,
+      "实地": 22683,
+      "▁图书目录": 22684,
+      "Renderer": 22685,
+      "人与": 22686,
+      "单一": 22687,
+      "synt": 22688,
+      "▁Stack": 22689,
+      "disci": 22690,
+      "料的": 22691,
+      "两者": 22692,
+      "住了": 22693,
+      "▁Guys": 22694,
+      "书中": 22695,
+      "▁tonight": 22696,
+      "Mail": 22697,
+      "▁dense": 22698,
+      "▁Specifically": 22699,
+      "▁θ": 22700,
+      "场地": 22701,
+      "▁encountered": 22702,
+      "proc": 22703,
+      "▁Kelly": 22704,
+      "ат": 22705,
+      "▁reactive": 22706,
+      "增值": 22707,
+      "市区": 22708,
+      "把自己": 22709,
+      "年初": 22710,
+      "执行工作若干问题的规定": 22711,
+      "▁cc": 22712,
+      "respons": 22713,
+      "▁adjusted": 22714,
+      ">\\'": 22715,
+      "▁Andre": 22716,
+      "▁pushed": 22717,
+      "idase": 22718,
+      "loid": 22719,
+      "Arch": 22720,
+      "▁Entertainment": 22721,
+      "生气": 22722,
+      "表彰": 22723,
+      "▁packed": 22724,
+      "ermat": 22725,
+      "▁caption": 22726,
+      "的声音": 22727,
+      "健全": 22728,
+      "等差数列": 22729,
+      "▁relaxation": 22730,
+      "六条": 22731,
+      "discipl": 22732,
+      "凌晨": 22733,
+      "相遇": 22734,
+      "▁bearing": 22735,
+      "▁何": 22736,
+      "ncc": 22737,
+      "该公司总经理": 22738,
+      "▁Summary": 22739,
+      "▁别名": 22740,
+      "在美国": 22741,
+      "在互联网": 22742,
+      "第三人": 22743,
+      "rob": 22744,
+      "▁runner": 22745,
+      "nInit": 22746,
+      "遗憾": 22747,
+      "▁Norwegian": 22748,
+      "▁lesion": 22749,
+      "▁carpet": 22750,
+      "Notification": 22751,
+      "新浪": 22752,
+      "▁()\\": 22753,
+      "▁enthusi": 22754,
+      "发酵": 22755,
+      "将来": 22756,
+      "▁Must": 22757,
+      "积极参加思想": 22758,
+      "春风": 22759,
+      "inherit": 22760,
+      "的意思": 22761,
+      "▁Laure": 22762,
+      "wealth": 22763,
+      "▁vegetables": 22764,
+      "Folder": 22765,
+      "nexports": 22766,
+      "production": 22767,
+      "airy": 22768,
+      "▁Depend": 22769,
+      "案后": 22770,
+      "customer": 22771,
+      "delay": 22772,
+      "长大": 22773,
+      "▁soldiers": 22774,
+      "uition": 22775,
+      "褐色": 22776,
+      "nTHE": 22777,
+      "iative": 22778,
+      "webpack": 22779,
+      "▁Mathematical": 22780,
+      "▁Communications": 22781,
+      "▁Edge": 22782,
+      "便于": 22783,
+      "gun": 22784,
+      "▁calculator": 22785,
+      "▁Hold": 22786,
+      "iotics": 22787,
+      "中国人民银行": 22788,
+      "nCast": 22789,
+      "▁gastro": 22790,
+      "▁Alber": 22791,
+      "一元": 22792,
+      "月利率": 22793,
+      "udge": 22794,
+      "mill": 22795,
+      "后台": 22796,
+      "aline": 22797,
+      "▁______": 22798,
+      "的爱": 22799,
+      "职业技术教育": 22800,
+      "开幕": 22801,
+      "美味": 22802,
+      "▁ovar": 22803,
+      "Prefix": 22804,
+      "▁Blood": 22805,
+      "signal": 22806,
+      "疑问": 22807,
+      "▁sophisticated": 22808,
+      "▁convex": 22809,
+      "nge": 22810,
+      "车间": 22811,
+      "纳税": 22812,
+      "medium": 22813,
+      "▁grain": 22814,
+      "John": 22815,
+      "ductor": 22816,
+      "▁portal": 22817,
+      "人民法院认为": 22818,
+      "简历": 22819,
+      "backs": 22820,
+      "▁Gall": 22821,
+      "acular": 22822,
+      "▁invasion": 22823,
+      "fo": 22824,
+      "▁vas": 22825,
+      "▁basal": 22826,
+      "▁sender": 22827,
+      "Welcome": 22828,
+      "lists": 22829,
+      "你知道": 22830,
+      "递增": 22831,
+      "▁Harris": 22832,
+      "▁collision": 22833,
+      "▁Bor": 22834,
+      "▁impairment": 22835,
+      "▁pap": 22836,
+      "▁Examples": 22837,
+      "较高的": 22838,
+      "▁admit": 22839,
+      "罪行": 22840,
+      "▁Ans": 22841,
+      "金华": 22842,
+      "facebook": 22843,
+      "▁Ax": 22844,
+      "跟踪": 22845,
+      "wx": 22846,
+      "oft": 22847,
+      "addle": 22848,
+      "▁BR": 22849,
+      "aran": 22850,
+      "连带": 22851,
+      "邮箱": 22852,
+      "ét": 22853,
+      "]),": 22854,
+      "ilty": 22855,
+      "olec": 22856,
+      "▁consent": 22857,
+      "Disc": 22858,
+      "第六章": 22859,
+      "▁csv": 22860,
+      "Calculate": 22861,
+      "alling": 22862,
+      "ergic": 22863,
+      "▁heritage": 22864,
+      "bold": 22865,
+      "▁utils": 22866,
+      "▁mu": 22867,
+      "}),\\": 22868,
+      "▁Perhaps": 22869,
+      "ussy": 22870,
+      "Extensions": 22871,
+      "以下问题": 22872,
+      "为基础": 22873,
+      "▁Protection": 22874,
+      "去看": 22875,
+      "有些人": 22876,
+      "▁tet": 22877,
+      "▁Ott": 22878,
+      "▁题目": 22879,
+      "▁fitted": 22880,
+      "nCommun": 22881,
+      "\\'])": 22882,
+      "▁moon": 22883,
+      "计算即可": 22884,
+      "污水": 22885,
+      "▁ValueError": 22886,
+      "平米": 22887,
+      "heim": 22888,
+      "▁lon": 22889,
+      "Black": 22890,
+      "的两个": 22891,
+      "▁simulated": 22892,
+      "▁ultrasound": 22893,
+      "▁其三": 22894,
+      "▁Measure": 22895,
+      "要点": 22896,
+      "▁Islamic": 22897,
+      "▁legacy": 22898,
+      "▁voting": 22899,
+      "借助": 22900,
+      "ngit": 22901,
+      "经济技术": 22902,
+      "签字": 22903,
+      "rystal": 22904,
+      "执业": 22905,
+      "▁lact": 22906,
+      "emale": 22907,
+      "淄博": 22908,
+      "讲座": 22909,
+      "ifting": 22910,
+      "农产品": 22911,
+      "bras": 22912,
+      "▁refuge": 22913,
+      "▁Vietnam": 22914,
+      "nDis": 22915,
+      "录音": 22916,
+      "▁importantly": 22917,
+      "nun": 22918,
+      "三国": 22919,
+      "▁reb": 22920,
+      "▁conducting": 22921,
+      "▁surrounded": 22922,
+      "聚集": 22923,
+      "救援": 22924,
+      "▁toile": 22925,
+      "不动产": 22926,
+      "数据的": 22927,
+      "▁Dallas": 22928,
+      "研究中心": 22929,
+      "▁Khan": 22930,
+      "债券": 22931,
+      "DEFAULT": 22932,
+      "一等奖": 22933,
+      "▁conservative": 22934,
+      "▁Pear": 22935,
+      "▁代表作品": 22936,
+      "▁thromb": 22937,
+      "灾害": 22938,
+      "▁duct": 22939,
+      "▁pros": 22940,
+      "nwhich": 22941,
+      "月二十五日": 22942,
+      "▁engaging": 22943,
+      "▁时间": 22944,
+      "现有的": 22945,
+      "▁charity": 22946,
+      "Should": 22947,
+      "plicity": 22948,
+      "▁celebration": 22949,
+      "剩下的": 22950,
+      "▁Reference": 22951,
+      "ordered": 22952,
+      "uscript": 22953,
+      "▁plates": 22954,
+      "▁已": 22955,
+      "巴西": 22956,
+      "▁SS": 22957,
+      "Stand": 22958,
+      "▁cyclic": 22959,
+      "▁Forward": 22960,
+      "}\\'": 22961,
+      "Authentication": 22962,
+      "▁Grant": 22963,
+      "张某某": 22964,
+      "okin": 22965,
+      "▁Trail": 22966,
+      "▁venues": 22967,
+      "倾向": 22968,
+      "试题": 22969,
+      "nil": 22970,
+      "▁Annual": 22971,
+      "精英": 22972,
+      "enza": 22973,
+      "leet": 22974,
+      "▁rescue": 22975,
+      "塑造": 22976,
+      "nPublic": 22977,
+      "Rate": 22978,
+      "安卓": 22979,
+      "纳米": 22980,
+      "Actions": 22981,
+      "三分": 22982,
+      "的风": 22983,
+      "Ok": 22984,
+      "▁uniqu": 22985,
+      "▁xx": 22986,
+      "▁symmetry": 22987,
+      "▁政": 22988,
+      "黄河": 22989,
+      "cz": 22990,
+      "▁alarm": 22991,
+      "▁detector": 22992,
+      "quiry": 22993,
+      "▁structured": 22994,
+      "▁translated": 22995,
+      "设备的": 22996,
+      "loads": 22997,
+      "uther": 22998,
+      "dll": 22999,
+      "▁struggling": 23000,
+      "wick": 23001,
+      "esa": 23002,
+      "▁brothers": 23003,
+      "emit": 23004,
+      "▁READ": 23005,
+      "▁Dream": 23006,
+      "▁Rol": 23007,
+      "#{": 23008,
+      "▁suicide": 23009,
+      "接种": 23010,
+      "Div": 23011,
+      "执行中": 23012,
+      "枚举": 23013,
+      "道的": 23014,
+      "eder": 23015,
+      "▁multid": 23016,
+      "▁相关": 23017,
+      "vars": 23018,
+      "▁Incre": 23019,
+      "局局长": 23020,
+      "▁Hans": 23021,
+      "中午": 23022,
+      "两岸": 23023,
+      "▁年": 23024,
+      "▁ET": 23025,
+      "trim": 23026,
+      "新兴": 23027,
+      "▁输出": 23028,
+      "▁mold": 23029,
+      "执行申请": 23030,
+      "▁Mail": 23031,
+      "手指": 23032,
+      "这一切": 23033,
+      "PECT": 23034,
+      "oca": 23035,
+      "▁Brazilian": 23036,
+      "▁seeks": 23037,
+      "ictures": 23038,
+      "开户": 23039,
+      "▁morphology": 23040,
+      "nFrench": 23041,
+      "就这样": 23042,
+      "▁lengths": 23043,
+      "ochastic": 23044,
+      "▁天": 23045,
+      "玄幻": 23046,
+      "`;\\": 23047,
+      "用心": 23048,
+      "椭圆": 23049,
+      "▁discusses": 23050,
+      "PDATE": 23051,
+      "评审鉴定表": 23052,
+      "▁Within": 23053,
+      "血清": 23054,
+      "上学": 23055,
+      "▁Ky": 23056,
+      "子宫": 23057,
+      "deploy": 23058,
+      "urities": 23059,
+      "▁Industry": 23060,
+      "家园": 23061,
+      "▁proc": 23062,
+      "zheimer": 23063,
+      "▁antenna": 23064,
+      "▁staying": 23065,
+      "平凡": 23066,
+      "▁Logger": 23067,
+      "▁stops": 23068,
+      "udd": 23069,
+      "▁residues": 23070,
+      "▁generates": 23071,
+      "▁Num": 23072,
+      "General": 23073,
+      "▁SK": 23074,
+      "▁gray": 23075,
+      "▁arom": 23076,
+      "uming": 23077,
+      "transaction": 23078,
+      "▁conjunction": 23079,
+      "WT": 23080,
+      "日出生于": 23081,
+      "职位": 23082,
+      "▁alternatives": 23083,
+      "坐落": 23084,
+      "▁Bishop": 23085,
+      "遵循": 23086,
+      "▁Turkish": 23087,
+      "▁annually": 23088,
+      "▁chemicals": 23089,
+      "▁adolescents": 23090,
+      "五金": 23091,
+      "Selected": 23092,
+      "iri": 23093,
+      "▁Feature": 23094,
+      "还需要": 23095,
+      "human": 23096,
+      "Memory": 23097,
+      "封闭": 23098,
+      "ArrayList": 23099,
+      "告诉你": 23100,
+      "▁poverty": 23101,
+      "▁Alb": 23102,
+      "▁ligand": 23103,
+      "gium": 23104,
+      "slice": 23105,
+      "时间的": 23106,
+      "杀人": 23107,
+      "▁deeply": 23108,
+      "rio": 23109,
+      "在开庭审理过程中": 23110,
+      "▁crusher": 23111,
+      "▁quantities": 23112,
+      "违规": 23113,
+      "法律服务所法律工作者": 23114,
+      "▁twin": 23115,
+      "▁Hamp": 23116,
+      "▁inhibit": 23117,
+      "safe": 23118,
+      "Single": 23119,
+      "nAc": 23120,
+      "让他们": 23121,
+      "▁Platform": 23122,
+      "▁unlikely": 23123,
+      "formed": 23124,
+      "chr": 23125,
+      "mass": 23126,
+      "▁experiencing": 23127,
+      "▁rust": 23128,
+      "winning": 23129,
+      "小额": 23130,
+      "▁outputs": 23131,
+      "就没": 23132,
+      "ancell": 23133,
+      "汇报": 23134,
+      "processing": 23135,
+      "▁jazz": 23136,
+      "代数": 23137,
+      "CODE": 23138,
+      "侵犯": 23139,
+      "的角度": 23140,
+      "备用": 23141,
+      "要想": 23142,
+      "的其他情形": 23143,
+      "队的": 23144,
+      "▁Doctor": 23145,
+      "彩色": 23146,
+      "电磁": 23147,
+      "▁runtime": 23148,
+      "裁定书送达之日起": 23149,
+      "COMP": 23150,
+      "▁repairs": 23151,
+      "requires": 23152,
+      "▁Disease": 23153,
+      "复议一次": 23154,
+      "▁亚": 23155,
+      "writing": 23156,
+      "aval": 23157,
+      "▁Construction": 23158,
+      "品牌的": 23159,
+      "ENG": 23160,
+      "▁weigh": 23161,
+      "SY": 23162,
+      "▁relevance": 23163,
+      "▁conceptual": 23164,
+      "子女抚养": 23165,
+      "特朗": 23166,
+      "爱好者": 23167,
+      "OG": 23168,
+      "owned": 23169,
+      "生动": 23170,
+      "▁stom": 23171,
+      "▁minus": 23172,
+      "▁fibers": 23173,
+      "为终审裁定": 23174,
+      "力学": 23175,
+      "▁dishes": 23176,
+      "▁antigens": 23177,
+      "▁Domin": 23178,
+      "▁Mort": 23179,
+      "▁NF": 23180,
+      "▁unsigned": 23181,
+      "繁殖": 23182,
+      "老师的": 23183,
+      "▁talked": 23184,
+      "还原": 23185,
+      "ternet": 23186,
+      "▁Scale": 23187,
+      "recated": 23188,
+      "▁journalist": 23189,
+      "老年人": 23190,
+      "regulated": 23191,
+      "何处": 23192,
+      "Migration": 23193,
+      "xxx": 23194,
+      "▁incent": 23195,
+      "▁Swiss": 23196,
+      "interval": 23197,
+      "回头": 23198,
+      "热点": 23199,
+      "▁jewel": 23200,
+      "解决了": 23201,
+      "velope": 23202,
+      "OA": 23203,
+      "容积": 23204,
+      "放置": 23205,
+      "lc": 23206,
+      "▁blast": 23207,
+      "osomal": 23208,
+      "▁Wel": 23209,
+      "▁Working": 23210,
+      "形容": 23211,
+      "itan": 23212,
+      "二人": 23213,
+      "ifference": 23214,
+      "本裁定为终审裁定": 23215,
+      "加班": 23216,
+      "multi": 23217,
+      "▁attributed": 23218,
+      "气息": 23219,
+      "▁collective": 23220,
+      "▁fired": 23221,
+      "inject": 23222,
+      "抛物线": 23223,
+      "▁restart": 23224,
+      "edical": 23225,
+      "\")[": 23226,
+      "液压": 23227,
+      "纵横": 23228,
+      "回报": 23229,
+      "kee": 23230,
+      "血糖": 23231,
+      "▁Cer": 23232,
+      "国防": 23233,
+      "等人": 23234,
+      "▁sequencing": 23235,
+      "癜风": 23236,
+      "leasing": 23237,
+      "发起": 23238,
+      "辩护": 23239,
+      "据了解": 23240,
+      "▁opinions": 23241,
+      "▁okay": 23242,
+      "▁Bah": 23243,
+      "延续": 23244,
+      "▁cig": 23245,
+      "written": 23246,
+      "undefined": 23247,
+      "▁emotion": 23248,
+      "lia": 23249,
+      "▁Chile": 23250,
+      "▁sectors": 23251,
+      "桥梁": 23252,
+      "▁mistakes": 23253,
+      "▁%}": 23254,
+      "因与": 23255,
+      "花生": 23256,
+      "▁preserve": 23257,
+      "én": 23258,
+      "▁Bu": 23259,
+      "nOpen": 23260,
+      "▁pixel": 23261,
+      "▁arrangements": 23262,
+      "isible": 23263,
+      "resolution": 23264,
+      "Interval": 23265,
+      "▁Address": 23266,
+      "amento": 23267,
+      "为啥": 23268,
+      "的诉讼请求": 23269,
+      "activated": 23270,
+      "十字": 23271,
+      "▁floors": 23272,
+      "heter": 23273,
+      "投诉": 23274,
+      "inge": 23275,
+      "和小": 23276,
+      "▁astr": 23277,
+      "向量": 23278,
+      "identifier": 23279,
+      "umbnail": 23280,
+      "班级": 23281,
+      "helium": 23282,
+      "不算": 23283,
+      "ocar": 23284,
+      "▁cultured": 23285,
+      "▁ward": 23286,
+      "▁fascin": 23287,
+      "出来了": 23288,
+      "▁Kam": 23289,
+      "▁pride": 23290,
+      "▁brid": 23291,
+      "▁installing": 23292,
+      "Drop": 23293,
+      "oven": 23294,
+      "作了": 23295,
+      "Multi": 23296,
+      "▁synthesized": 23297,
+      "Da": 23298,
+      "▁sacr": 23299,
+      "▁scalar": 23300,
+      "▁abroad": 23301,
+      "▁罗": 23302,
+      "▁Initial": 23303,
+      "\">{{": 23304,
+      "▁regards": 23305,
+      "沈阳市": 23306,
+      "▁Nag": 23307,
+      "▁clearance": 23308,
+      "gray": 23309,
+      "多媒体": 23310,
+      "学习的": 23311,
+      "▁Ice": 23312,
+      "如果没有": 23313,
+      "earn": 23314,
+      "worker": 23315,
+      "▁ethnic": 23316,
+      "客户端": 23317,
+      "▁accommodation": 23318,
+      "立即执行": 23319,
+      "我校": 23320,
+      "nJoined": 23321,
+      "HECK": 23322,
+      "新时代": 23323,
+      "陌生": 23324,
+      "▁TE": 23325,
+      "IZ": 23326,
+      "ompson": 23327,
+      "▁gauge": 23328,
+      "▁workplace": 23329,
+      "除非": 23330,
+      "uper": 23331,
+      "▁poetry": 23332,
+      "▁Bull": 23333,
+      "在一": 23334,
+      "当下": 23335,
+      "▁perturb": 23336,
+      "卓越": 23337,
+      "eor": 23338,
+      "Normal": 23339,
+      "以免": 23340,
+      "白癜风": 23341,
+      "这个时候": 23342,
+      "▁Util": 23343,
+      "土豆": 23344,
+      "李某某": 23345,
+      "ente": 23346,
+      "nDet": 23347,
+      "主演": 23348,
+      "arity": 23349,
+      "余人": 23350,
+      "▁mountains": 23351,
+      "亦无异议": 23352,
+      "当日": 23353,
+      "▁Contract": 23354,
+      "▁acknowled": 23355,
+      "ementia": 23356,
+      "▁Antonio": 23357,
+      "▁consequence": 23358,
+      "heart": 23359,
+      "开口": 23360,
+      "comfort": 23361,
+      "dale": 23362,
+      "护士": 23363,
+      "TEXT": 23364,
+      "▁Thailand": 23365,
+      "ри": 23366,
+      "介质": 23367,
+      "充分发挥": 23368,
+      "▁mechanics": 23369,
+      "好评": 23370,
+      "服用": 23371,
+      "seconds": 23372,
+      "佐证": 23373,
+      "▁representations": 23374,
+      "不管是": 23375,
+      "有效地": 23376,
+      "治愈": 23377,
+      "too": 23378,
+      "辉煌": 23379,
+      "▁dismiss": 23380,
+      "▁compiler": 23381,
+      "▁reflected": 23382,
+      "▁Send": 23383,
+      "▁publisher": 23384,
+      "官员": 23385,
+      "▁Dog": 23386,
+      "▁batt": 23387,
+      "▁immunity": 23388,
+      "mia": 23389,
+      "▁wrest": 23390,
+      "▁Chap": 23391,
+      "▁incorporate": 23392,
+      "logs": 23393,
+      "传染": 23394,
+      "▁hearts": 23395,
+      "scala": 23396,
+      "▁recovered": 23397,
+      "因为我": 23398,
+      "得以": 23399,
+      "uese": 23400,
+      "之前的": 23401,
+      "）×": 23402,
+      "十六条": 23403,
+      "▁Beautiful": 23404,
+      "交通运输": 23405,
+      "▁loving": 23406,
+      "▁realistic": 23407,
+      "nsub": 23408,
+      "▁Lincoln": 23409,
+      "▁gathered": 23410,
+      "▁Economic": 23411,
+      "宾馆": 23412,
+      "层面": 23413,
+      "▁initiated": 23414,
+      "issippi": 23415,
+      "iere": 23416,
+      "Objects": 23417,
+      "kle": 23418,
+      "UV": 23419,
+      "水量": 23420,
+      "诗歌": 23421,
+      "▁],\\": 23422,
+      "队友": 23423,
+      "▁lattice": 23424,
+      "▁bip": 23425,
+      "eine": 23426,
+      "▁Join": 23427,
+      "▁connectivity": 23428,
+      "民二": 23429,
+      "▁singing": 23430,
+      "到处": 23431,
+      "中止": 23432,
+      "▁Nature": 23433,
+      "乘坐": 23434,
+      "第一时间": 23435,
+      "ateful": 23436,
+      "▁Samsung": 23437,
+      "节省": 23438,
+      "予以证实": 23439,
+      "▁stir": 23440,
+      "tu": 23441,
+      "实的": 23442,
+      "sun": 23443,
+      "同样的": 23444,
+      "fred": 23445,
+      "CCESS": 23446,
+      "\\\\}$": 23447,
+      "ancouver": 23448,
+      "钢琴": 23449,
+      "▁examines": 23450,
+      "日公开开庭进行了审理": 23451,
+      "▁buried": 23452,
+      "coma": 23453,
+      "要在": 23454,
+      "▁tropical": 23455,
+      "bucket": 23456,
+      "▁scattering": 23457,
+      "▁Was": 23458,
+      "商城": 23459,
+      "nHello": 23460,
+      "▁golden": 23461,
+      "友谊": 23462,
+      "法律法规及": 23463,
+      "烦恼": 23464,
+      "ritic": 23465,
+      "traction": 23466,
+      "▁investments": 23467,
+      "▁specialists": 23468,
+      "烟台": 23469,
+      "▁Queensland": 23470,
+      "juana": 23471,
+      "▁grows": 23472,
+      "提问": 23473,
+      "▁PCR": 23474,
+      "▁Together": 23475,
+      "之和": 23476,
+      "如有": 23477,
+      "▁trips": 23478,
+      "Tools": 23479,
+      "在哪里": 23480,
+      "nGeography": 23481,
+      "Where": 23482,
+      "时效": 23483,
+      "nInternational": 23484,
+      "占用": 23485,
+      "做什么": 23486,
+      "▁Gib": 23487,
+      "▁clarify": 23488,
+      "中等": 23489,
+      "nMat": 23490,
+      "的最佳": 23491,
+      "▁anomal": 23492,
+      "▁declar": 23493,
+      "清除": 23494,
+      "arijuana": 23495,
+      "elements": 23496,
+      "REG": 23497,
+      "▁tob": 23498,
+      "rowave": 23499,
+      "▁hopefully": 23500,
+      "人性": 23501,
+      "iously": 23502,
+      "ynthesis": 23503,
+      "晋江文学城": 23504,
+      "▁empower": 23505,
+      "▁specification": 23506,
+      "▁governor": 23507,
+      "综合性": 23508,
+      "你好": 23509,
+      "athers": 23510,
+      "oler": 23511,
+      "erals": 23512,
+      "▁Info": 23513,
+      "▁medications": 23514,
+      "▁Login": 23515,
+      "strument": 23516,
+      "作风": 23517,
+      "variant": 23518,
+      "××××": 23519,
+      "▁ure": 23520,
+      "▁nose": 23521,
+      "▁perme": 23522,
+      "的现象": 23523,
+      "▁aspir": 23524,
+      "visual": 23525,
+      "strings": 23526,
+      "▁powered": 23527,
+      "在了": 23528,
+      "larg": 23529,
+      "fessional": 23530,
+      "的游戏": 23531,
+      "DK": 23532,
+      "Dispatch": 23533,
+      "verify": 23534,
+      "ynamics": 23535,
+      "circle": 23536,
+      "▁fluct": 23537,
+      "这也": 23538,
+      "▁Afghan": 23539,
+      "▁substitution": 23540,
+      "我们就": 23541,
+      "▁Physical": 23542,
+      "热带": 23543,
+      "相邻": 23544,
+      "其中的": 23545,
+      "▁Matrix": 23546,
+      "have": 23547,
+      "▁slic": 23548,
+      "又称": 23549,
+      "▁increment": 23550,
+      "还好": 23551,
+      "男主": 23552,
+      "▁differently": 23553,
+      "▁Utah": 23554,
+      "▁passenger": 23555,
+      "民法": 23556,
+      "较多": 23557,
+      "▁vag": 23558,
+      "私人": 23559,
+      "▁assumptions": 23560,
+      "每周": 23561,
+      "ikz": 23562,
+      "姐妹": 23563,
+      "ISA": 23564,
+      "尽可能": 23565,
+      "能力和": 23566,
+      "▁reportedly": 23567,
+      "总理": 23568,
+      "▁renamed": 23569,
+      "▁arise": 23570,
+      "网易": 23571,
+      "nVill": 23572,
+      "phabet": 23573,
+      "▁ranking": 23574,
+      "陪伴": 23575,
+      "▁ignored": 23576,
+      "划拨": 23577,
+      "切成": 23578,
+      "临沂": 23579,
+      "▁Personal": 23580,
+      "你看": 23581,
+      "空中": 23582,
+      "▁certification": 23583,
+      "▁mim": 23584,
+      "▁IOException": 23585,
+      "▁chamber": 23586,
+      "▁cervical": 23587,
+      "用户的": 23588,
+      "▁Session": 23589,
+      "▁Ron": 23590,
+      "期间不停止": 23591,
+      "▁pil": 23592,
+      "境内": 23593,
+      "副本二份": 23594,
+      "canvas": 23595,
+      "▁MC": 23596,
+      "strong": 23597,
+      "▁packet": 23598,
+      "▁reflects": 23599,
+      "▁pyl": 23600,
+      "说到": 23601,
+      "▁Happy": 23602,
+      "实行独任审判": 23603,
+      "不同意": 23604,
+      "十大": 23605,
+      "▁makeup": 23606,
+      "耐心": 23607,
+      "Dao": 23608,
+      "▁白": 23609,
+      "▁Aqu": 23610,
+      "▁cand": 23611,
+      "标记": 23612,
+      "饰演": 23613,
+      "▁Aw": 23614,
+      "一口": 23615,
+      "再生": 23616,
+      "▁GT": 23617,
+      "大专": 23618,
+      "投票": 23619,
+      "调解协议": 23620,
+      "▁PUBLIC": 23621,
+      "▁infar": 23622,
+      "发作": 23623,
+      "▁Lu": 23624,
+      "▁pancreatic": 23625,
+      "投标": 23626,
+      "▁ethanol": 23627,
+      "号线": 23628,
+      "▁Mut": 23629,
+      "赋予": 23630,
+      "▁lear": 23631,
+      "MV": 23632,
+      "ów": 23633,
+      "nbl": 23634,
+      "去世": 23635,
+      "有一些": 23636,
+      "走势": 23637,
+      "▁oblig": 23638,
+      "▁iPad": 23639,
+      "▁tact": 23640,
+      "▁cube": 23641,
+      "无疑": 23642,
+      "▁portions": 23643,
+      "▁positively": 23644,
+      "届满": 23645,
+      "▁Bron": 23646,
+      "▁pione": 23647,
+      "南部": 23648,
+      "Visual": 23649,
+      "线段": 23650,
+      "strict": 23651,
+      "▁cytotox": 23652,
+      "atories": 23653,
+      "<<\"": 23654,
+      "那时候": 23655,
+      "nTake": 23656,
+      "▁trains": 23657,
+      "▁accommodate": 23658,
+      "▁dispatch": 23659,
+      "▁NIH": 23660,
+      "▁cyber": 23661,
+      "▁Monte": 23662,
+      "务工": 23663,
+      "石头": 23664,
+      "一个是": 23665,
+      "Comput": 23666,
+      "复议期间不停止": 23667,
+      "▁loyal": 23668,
+      "▁Sale": 23669,
+      "月二十七日": 23670,
+      "unks": 23671,
+      "自然科学": 23672,
+      "phy": 23673,
+      "▁Metro": 23674,
+      "▁Singh": 23675,
+      "igenous": 23676,
+      "龙头": 23677,
+      "均衡": 23678,
+      "▁downloaded": 23679,
+      "▁backend": 23680,
+      "花钱": 23681,
+      "▁brilliant": 23682,
+      "▁unlike": 23683,
+      "▁无": 23684,
+      "▁Republicans": 23685,
+      "储备": 23686,
+      "口罩": 23687,
+      "lit": 23688,
+      "▁Flo": 23689,
+      "功夫": 23690,
+      "化妆品": 23691,
+      "▁Toy": 23692,
+      "中华人民共和国婚姻法": 23693,
+      "安静": 23694,
+      "fund": 23695,
+      "▁paintings": 23696,
+      "ascade": 23697,
+      "ifacts": 23698,
+      "出台": 23699,
+      "吐槽": 23700,
+      "runner": 23701,
+      "▁traits": 23702,
+      "nSimpl": 23703,
+      "▁Intellig": 23704,
+      "Design": 23705,
+      "ainian": 23706,
+      "▁Fellow": 23707,
+      "也不知道": 23708,
+      "▁establishing": 23709,
+      "▁purified": 23710,
+      "alloc": 23711,
+      "▁Hyper": 23712,
+      "mv": 23713,
+      "▁Beng": 23714,
+      "▁tokens": 23715,
+      "决心": 23716,
+      "量刑": 23717,
+      "这时候": 23718,
+      "glas": 23719,
+      "本案受理费": 23720,
+      "▁Environmental": 23721,
+      "▁institutional": 23722,
+      "专科": 23723,
+      "默默": 23724,
+      "的思路": 23725,
+      "▁watched": 23726,
+      "▁isolates": 23727,
+      "小明": 23728,
+      "的路": 23729,
+      "支公司": 23730,
+      "▁circulation": 23731,
+      "running": 23732,
+      "负数": 23733,
+      "▁reporter": 23734,
+      "extensions": 23735,
+      "Uri": 23736,
+      "在国内": 23737,
+      "▁casino": 23738,
+      "佛山市": 23739,
+      "sequently": 23740,
+      "▁popularity": 23741,
+      "诉至法院": 23742,
+      "▁Whit": 23743,
+      "film": 23744,
+      "新手": 23745,
+      "社团": 23746,
+      "▁制作": 23747,
+      "Team": 23748,
+      "nEl": 23749,
+      "▁afraid": 23750,
+      "methyl": 23751,
+      "变换": 23752,
+      "组的": 23753,
+      "试图": 23754,
+      "\"):\\": 23755,
+      "▁updating": 23756,
+      "▁suppliers": 23757,
+      "▁drums": 23758,
+      "▁Later": 23759,
+      "ifferential": 23760,
+      "不成": 23761,
+      "▁gentle": 23762,
+      "heat": 23763,
+      "▁incub": 23764,
+      "dependency": 23765,
+      "体检": 23766,
+      "本院依据已经发生法律效力的": 23767,
+      "携手": 23768,
+      "mouth": 23769,
+      "▁venture": 23770,
+      "social": 23771,
+      "Resources": 23772,
+      "▁Connecticut": 23773,
+      "verted": 23774,
+      "▁bleeding": 23775,
+      "▁grounds": 23776,
+      "▁wanting": 23777,
+      "▁Options": 23778,
+      "靠谱": 23779,
+      "▁Sarah": 23780,
+      "▁trusted": 23781,
+      "▁Optim": 23782,
+      "▁walked": 23783,
+      "模糊": 23784,
+      "编译": 23785,
+      "给原告": 23786,
+      "balance": 23787,
+      "operatorname": 23788,
+      "▁asks": 23789,
+      ":#": 23790,
+      "富有": 23791,
+      "▁genomic": 23792,
+      "透支": 23793,
+      "交往": 23794,
+      "十八日": 23795,
+      "高职": 23796,
+      "▁cha": 23797,
+      "▁registry": 23798,
+      ">\",\\": 23799,
+      "偶像": 23800,
+      "▁formulas": 23801,
+      "configure": 23802,
+      "▁datasets": 23803,
+      "▁Laboratory": 23804,
+      "ordinary": 23805,
+      "plemented": 23806,
+      "企业管理": 23807,
+      "选择了": 23808,
+      "竞技": 23809,
+      "常用的": 23810,
+      "▁addressing": 23811,
+      "nUnited": 23812,
+      "很简单": 23813,
+      "深的": 23814,
+      "RF": 23815,
+      "filled": 23816,
+      "研制": 23817,
+      "涵盖": 23818,
+      "形成的": 23819,
+      "▁highlighted": 23820,
+      "▁Customer": 23821,
+      "▁Framework": 23822,
+      "▁tomat": 23823,
+      "亲自": 23824,
+      "itone": 23825,
+      "▁salary": 23826,
+      "Short": 23827,
+      "看守所": 23828,
+      "▁habitat": 23829,
+      "▁cone": 23830,
+      "▁Reserved": 23831,
+      "▁generations": 23832,
+      "▁Column": 23833,
+      "南海": 23834,
+      "translate": 23835,
+      "故意伤害罪": 23836,
+      "▁recipients": 23837,
+      "特朗普": 23838,
+      "approx": 23839,
+      "photo": 23840,
+      "▁labeled": 23841,
+      "民营": 23842,
+      "立足": 23843,
+      "apis": 23844,
+      "▁fitting": 23845,
+      "的我": 23846,
+      "Sum": 23847,
+      "actual": 23848,
+      "▁pseud": 23849,
+      "备选答案": 23850,
+      "▁suspected": 23851,
+      "▁[`": 23852,
+      "它们的": 23853,
+      "▁Tools": 23854,
+      "▁ordering": 23855,
+      "寻求": 23856,
+      "▁urllib": 23857,
+      "▁Revolution": 23858,
+      "▁credentials": 23859,
+      "▁ceiling": 23860,
+      "▁assessing": 23861,
+      "见面": 23862,
+      "郑州市": 23863,
+      "月二十六日": 23864,
+      "科目": 23865,
+      "▁Mach": 23866,
+      "Fire": 23867,
+      "联合国": 23868,
+      "destroy": 23869,
+      "▁Bos": 23870,
+      "和技术": 23871,
+      "endors": 23872,
+      "年后": 23873,
+      "asa": 23874,
+      "多多": 23875,
+      "LV": 23876,
+      "▁Authority": 23877,
+      "的表现": 23878,
+      "ruby": 23879,
+      "第一百三": 23880,
+      "▁convention": 23881,
+      "主要有": 23882,
+      "oriented": 23883,
+      "romagnetic": 23884,
+      "的孩子": 23885,
+      "▁Lower": 23886,
+      "▁stood": 23887,
+      "olph": 23888,
+      "nNe": 23889,
+      "▁ED": 23890,
+      "仪表": 23891,
+      "十几": 23892,
+      "三十日": 23893,
+      "NECT": 23894,
+      "bara": 23895,
+      "Basic": 23896,
+      "新加坡": 23897,
+      "olutions": 23898,
+      "QUEST": 23899,
+      "varphi": 23900,
+      "许可证": 23901,
+      "ре": 23902,
+      "”（": 23903,
+      "之际": 23904,
+      "对接": 23905,
+      "时空": 23906,
+      "人员的": 23907,
+      "▁bugs": 23908,
+      "pine": 23909,
+      "解决问题": 23910,
+      "bn": 23911,
+      "期内": 23912,
+      "▁juice": 23913,
+      "▁multiplication": 23914,
+      "azi": 23915,
+      "▁SV": 23916,
+      "▁irr": 23917,
+      "润滑": 23918,
+      "载体": 23919,
+      "十一章": 23920,
+      "中介": 23921,
+      "对抗": 23922,
+      "*}": 23923,
+      "把它": 23924,
+      "releases": 23925,
+      "cpu": 23926,
+      "建设工程": 23927,
+      "▁Rural": 23928,
+      "▁competitors": 23929,
+      "溶液": 23930,
+      "▁crop": 23931,
+      "▁evident": 23932,
+      "Constructor": 23933,
+      "不必": 23934,
+      "routes": 23935,
+      "▁Culture": 23936,
+      "人们的": 23937,
+      "pd": 23938,
+      "qual": 23939,
+      "seudo": 23940,
+      "graphics": 23941,
+      "▁elucidate": 23942,
+      "▁enumerate": 23943,
+      "oenix": 23944,
+      "▁shadow": 23945,
+      "顶点": 23946,
+      "▁Between": 23947,
+      "▁timestamp": 23948,
+      "▁deput": 23949,
+      "Constants": 23950,
+      "▁restored": 23951,
+      "选取": 23952,
+      "Elements": 23953,
+      "Sample": 23954,
+      "▁franchise": 23955,
+      "第一百四十四条之规定": 23956,
+      "▁campaigns": 23957,
+      "▁declare": 23958,
+      "▁governments": 23959,
+      "▁Innov": 23960,
+      "Character": 23961,
+      "早就": 23962,
+      "reach": 23963,
+      "▁Anthony": 23964,
+      "美人": 23965,
+      "resistant": 23966,
+      "▁analyzing": 23967,
+      "的点": 23968,
+      "uters": 23969,
+      "远远": 23970,
+      "目的是": 23971,
+      "▁Bra": 23972,
+      "navbar": 23973,
+      "▁ministry": 23974,
+      "不适": 23975,
+      "nDist": 23976,
+      "itored": 23977,
+      "▁Normal": 23978,
+      "▁defeat": 23979,
+      "▁brush": 23980,
+      "▁banking": 23981,
+      "至今未": 23982,
+      "开发商": 23983,
+      "上来": 23984,
+      "▁个人": 23985,
+      "▁minority": 23986,
+      "越大": 23987,
+      "▁denied": 23988,
+      "ilde": 23989,
+      "♪\\": 23990,
+      "News": 23991,
+      "组建": 23992,
+      "nmy": 23993,
+      "▁nu": 23994,
+      "淘汰": 23995,
+      "▁auction": 23996,
+      "市中级人民法院提出上诉": 23997,
+      "任免": 23998,
+      "年级": 23999,
+      "iu": 24000,
+      "ACC": 24001,
+      "fraction": 24002,
+      "hn": 24003,
+      "▁comic": 24004,
+      "个小时": 24005,
+      "sd": 24006,
+      "▁RF": 24007,
+      "Progress": 24008,
+      "▁双": 24009,
+      "催要": 24010,
+      "没有什么": 24011,
+      "立刻": 24012,
+      "giving": 24013,
+      "▁Query": 24014,
+      "▁Tournament": 24015,
+      "▁fastest": 24016,
+      "▁Match": 24017,
+      "urgical": 24018,
+      "金融机构": 24019,
+      "在校": 24020,
+      "sembled": 24021,
+      "▁pollution": 24022,
+      "\"{": 24023,
+      "▁Bird": 24024,
+      "先进的": 24025,
+      "south": 24026,
+      "[](": 24027,
+      "▁visitor": 24028,
+      "▁Lang": 24029,
+      "Upload": 24030,
+      "choice": 24031,
+      "▁parks": 24032,
+      "得很": 24033,
+      "…………": 24034,
+      "区块": 24035,
+      "极其": 24036,
+      "▁Su": 24037,
+      "▁Jason": 24038,
+      "Health": 24039,
+      "▁steam": 24040,
+      "激情": 24041,
+      "▁rounds": 24042,
+      "Runner": 24043,
+      "的小说": 24044,
+      "价的": 24045,
+      "eron": 24046,
+      "▁seam": 24047,
+      "潜力": 24048,
+      "▁instruct": 24049,
+      "▁lyrics": 24050,
+      "即便": 24051,
+      "等到": 24052,
+      "Scale": 24053,
+      "▁Sorry": 24054,
+      "▁Filter": 24055,
+      "istol": 24056,
+      "▁courts": 24057,
+      "Cred": 24058,
+      "▁lymphocytes": 24059,
+      "ongoose": 24060,
+      "多年来": 24061,
+      "服务有限公司": 24062,
+      "reak": 24063,
+      "riving": 24064,
+      "relative": 24065,
+      "▁Baseball": 24066,
+      "rell": 24067,
+      "额外": 24068,
+      "▁€": 24069,
+      "▁Agricult": 24070,
+      "▁constraint": 24071,
+      "▁Pap": 24072,
+      "ributions": 24073,
+      "▁discontin": 24074,
+      "▁duplicate": 24075,
+      "gui": 24076,
+      "incorporated": 24077,
+      "nMe": 24078,
+      "▁Kind": 24079,
+      "▁hill": 24080,
+      "▁advertis": 24081,
+      "▁distinguish": 24082,
+      "▁declined": 24083,
+      "▁Argentina": 24084,
+      "家电": 24085,
+      "的部分": 24086,
+      "XXXX": 24087,
+      "图案": 24088,
+      "▁Users": 24089,
+      "个人的": 24090,
+      "▁yeast": 24091,
+      "▁justify": 24092,
+      "▁scratch": 24093,
+      "▁mos": 24094,
+      "▁Lag": 24095,
+      "▁exterior": 24096,
+      "置业有限公司": 24097,
+      "icy": 24098,
+      "释义": 24099,
+      "Fac": 24100,
+      "Connect": 24101,
+      "▁subscrib": 24102,
+      "nLa": 24103,
+      "▁lady": 24104,
+      "▁sheets": 24105,
+      "籍贯": 24106,
+      "活动中": 24107,
+      "▁rubber": 24108,
+      "Topic": 24109,
+      "散热": 24110,
+      "合同纠纷": 24111,
+      "▁precip": 24112,
+      "▁disaster": 24113,
+      "▁tomography": 24114,
+      "▁Orange": 24115,
+      "佛教": 24116,
+      "▁Usage": 24117,
+      "▁\\\\;": 24118,
+      "截图": 24119,
+      "▁fet": 24120,
+      "loqu": 24121,
+      "深受": 24122,
+      "高铁": 24123,
+      "aden": 24124,
+      "plier": 24125,
+      "▁silent": 24126,
+      "▁Nation": 24127,
+      "▁virtually": 24128,
+      "kernel": 24129,
+      "loaded": 24130,
+      "▁artistic": 24131,
+      "优越": 24132,
+      "dynamic": 24133,
+      "发布的": 24134,
+      "像是": 24135,
+      "预约": 24136,
+      "调用": 24137,
+      "市人民检察院": 24138,
+      "uni": 24139,
+      "itol": 24140,
+      "▁initiation": 24141,
+      "XY": 24142,
+      "您认为": 24143,
+      "▁Marine": 24144,
+      "▁reaches": 24145,
+      "GI": 24146,
+      "▁Bh": 24147,
+      "水平的": 24148,
+      "记数": 24149,
+      "▁acceptance": 24150,
+      "不高": 24151,
+      "nSouth": 24152,
+      "Platform": 24153,
+      "UK": 24154,
+      "▁pad": 24155,
+      "▁Estate": 24156,
+      "▁Formula": 24157,
+      "▁唐": 24158,
+      "GD": 24159,
+      "▁Spirit": 24160,
+      "创立": 24161,
+      "▁congr": 24162,
+      "▁publicly": 24163,
+      "▁penalty": 24164,
+      "▁phrase": 24165,
+      "mag": 24166,
+      "这项": 24167,
+      "前言": 24168,
+      "制造业": 24169,
+      "flash": 24170,
+      "inance": 24171,
+      "▁Alzheimer": 24172,
+      "ippet": 24173,
+      "西藏": 24174,
+      "pes": 24175,
+      "▁Israeli": 24176,
+      "Secret": 24177,
+      "没有人": 24178,
+      "▁landing": 24179,
+      "的概念": 24180,
+      "招标": 24181,
+      "口碑": 24182,
+      "较好的": 24183,
+      "▁Introduction": 24184,
+      "oji": 24185,
+      "▁newest": 24186,
+      "recision": 24187,
+      "达到了": 24188,
+      "▁Easy": 24189,
+      "▁hypert": 24190,
+      "住四川省": 24191,
+      "TOKEN": 24192,
+      "liest": 24193,
+      "visible": 24194,
+      "明年": 24195,
+      "▁GP": 24196,
+      "可达": 24197,
+      "ndate": 24198,
+      "▁discrimination": 24199,
+      "▁Flu": 24200,
+      "▁vacuum": 24201,
+      "源于": 24202,
+      "▁Glass": 24203,
+      "▁Pain": 24204,
+      "▁Added": 24205,
+      "▁使用": 24206,
+      "的一部分": 24207,
+      "▁Wikipedia": 24208,
+      "▁Nap": 24209,
+      "工作室": 24210,
+      "Tem": 24211,
+      "▁stimuli": 24212,
+      "Span": 24213,
+      "mind": 24214,
+      "nCo": 24215,
+      "市政": 24216,
+      "职场": 24217,
+      "宏观": 24218,
+      "asures": 24219,
+      "LIED": 24220,
+      "▁Cut": 24221,
+      "Scene": 24222,
+      "▁sink": 24223,
+      "▁consistency": 24224,
+      "从中": 24225,
+      "nFull": 24226,
+      "▁TABLE": 24227,
+      "演技": 24228,
+      "▁pra": 24229,
+      "▁encoded": 24230,
+      "▁oh": 24231,
+      "▁backing": 24232,
+      "▁tackle": 24233,
+      "克服": 24234,
+      "▁bis": 24235,
+      "都不会": 24236,
+      "ridge": 24237,
+      "Hi": 24238,
+      "小白": 24239,
+      "readonly": 24240,
+      "▁Reserve": 24241,
+      "▁Bangl": 24242,
+      "\"])\\": 24243,
+      "病变": 24244,
+      "CONFIG": 24245,
+      "artz": 24246,
+      "▁dissemin": 24247,
+      "Si": 24248,
+      "nRev": 24249,
+      "▁fracture": 24250,
+      "完结": 24251,
+      "()[": 24252,
+      "▁talented": 24253,
+      "伦敦": 24254,
+      "法制": 24255,
+      "当今": 24256,
+      "百分之": 24257,
+      "巩固": 24258,
+      "开通": 24259,
+      "ourses": 24260,
+      "器材": 24261,
+      "transition": 24262,
+      "▁Pin": 24263,
+      "▁scar": 24264,
+      "经济的": 24265,
+      "▁kwargs": 24266,
+      "偿还原告": 24267,
+      "聚焦": 24268,
+      "本院作出的": 24269,
+      "nGerman": 24270,
+      "▁surveys": 24271,
+      "走出": 24272,
+      "▁FM": 24273,
+      "▁analysed": 24274,
+      "sized": 24275,
+      "▁brick": 24276,
+      "▁protest": 24277,
+      "▁Optional": 24278,
+      "缺少": 24279,
+      "特别授权": 24280,
+      "科学家": 24281,
+      "▁iso": 24282,
+      "第六十七条第三款": 24283,
+      "Ign": 24284,
+      "getName": 24285,
+      "Coll": 24286,
+      "▁Virtual": 24287,
+      "Usage": 24288,
+      "▁Him": 24289,
+      "▁Guid": 24290,
+      "▁spontaneous": 24291,
+      "ostics": 24292,
+      "efficient": 24293,
+      "apolis": 24294,
+      "Texture": 24295,
+      "概况": 24296,
+      "▁Cow": 24297,
+      "▁pathology": 24298,
+      "ARK": 24299,
+      "治疗组": 24300,
+      "▁Palest": 24301,
+      "客厅": 24302,
+      "▁organs": 24303,
+      "偿还借款": 24304,
+      "joint": 24305,
+      "nGood": 24306,
+      "建国": 24307,
+      "▁mayor": 24308,
+      "olen": 24309,
+      "同行": 24310,
+      "▁lam": 24311,
+      "`)": 24312,
+      "煤矿": 24313,
+      "lobin": 24314,
+      "ombok": 24315,
+      "申请符合法律规定": 24316,
+      "划分": 24317,
+      "odb": 24318,
+      "▁Maine": 24319,
+      "Bot": 24320,
+      "隧道": 24321,
+      "▁Nob": 24322,
+      "filters": 24323,
+      "▁settled": 24324,
+      "本报": 24325,
+      "extends": 24326,
+      "月二十三日": 24327,
+      "▁unw": 24328,
+      "▁gains": 24329,
+      "nGr": 24330,
+      "▁approximate": 24331,
+      "Yes": 24332,
+      "▁verification": 24333,
+      "▁proportional": 24334,
+      "▁technological": 24335,
+      "这句话": 24336,
+      "▁nights": 24337,
+      "▁SEO": 24338,
+      "anean": 24339,
+      "▁mutants": 24340,
+      "▁creativity": 24341,
+      "的标准": 24342,
+      "ajax": 24343,
+      "▁royal": 24344,
+      "▁manufactured": 24345,
+      "▁timely": 24346,
+      "▁▁▁▁▁▁▁▁▁▁▁": 24347,
+      "ptide": 24348,
+      "combinator": 24349,
+      "nGreat": 24350,
+      "Prop": 24351,
+      "判决生效后": 24352,
+      "一群": 24353,
+      "扩张": 24354,
+      "▁详情": 24355,
+      "▁Practice": 24356,
+      "地域": 24357,
+      "总数": 24358,
+      "amel": 24359,
+      "确诊": 24360,
+      "振动": 24361,
+      "及其他": 24362,
+      "▁merc": 24363,
+      "umat": 24364,
+      "▁Bag": 24365,
+      "▁Clark": 24366,
+      "▁Think": 24367,
+      "务必": 24368,
+      "手工": 24369,
+      "▁unnec": 24370,
+      "го": 24371,
+      "mel": 24372,
+      "毒品罪": 24373,
+      "▁Baron": 24374,
+      "▁texts": 24375,
+      "▁Multiple": 24376,
+      "▁churches": 24377,
+      "市人": 24378,
+      "nAB": 24379,
+      "家协会": 24380,
+      "▁Gab": 24381,
+      "▁Stage": 24382,
+      "otropic": 24383,
+      "把我": 24384,
+      "火车站": 24385,
+      "umar": 24386,
+      "的高度": 24387,
+      "▁secretary": 24388,
+      "▁collecting": 24389,
+      "球的": 24390,
+      "亮点": 24391,
+      "uffle": 24392,
+      "▁на": 24393,
+      "entially": 24394,
+      "▁feasible": 24395,
+      "inputs": 24396,
+      "哈哈哈": 24397,
+      "▁belt": 24398,
+      "nPosts": 24399,
+      "法律法规及监规": 24400,
+      "列车": 24401,
+      "Styles": 24402,
+      "▁nominated": 24403,
+      "}\\\\\\\\": 24404,
+      "网址": 24405,
+      "圆心": 24406,
+      "意的": 24407,
+      "PDF": 24408,
+      "▁Anne": 24409,
+      "Ids": 24410,
+      "▁comparative": 24411,
+      "▁convey": 24412,
+      "得了": 24413,
+      "▁cookie": 24414,
+      "追踪": 24415,
+      "▁sensing": 24416,
+      "▁floating": 24417,
+      "ulent": 24418,
+      "项和": 24419,
+      "组成部分": 24420,
+      "▁circuits": 24421,
+      "产物": 24422,
+      "有几": 24423,
+      "lovak": 24424,
+      "~~~~~~~~": 24425,
+      "▁prints": 24426,
+      "XR": 24427,
+      "孕妇": 24428,
+      "nKey": 24429,
+      "▁medic": 24430,
+      "nby": 24431,
+      "▁biopsy": 24432,
+      "▁Conc": 24433,
+      "环境保护": 24434,
+      "▁flask": 24435,
+      "▁Classic": 24436,
+      "▁legit": 24437,
+      "统治": 24438,
+      "▁hanging": 24439,
+      "▁illustrate": 24440,
+      "感冒": 24441,
+      "月参加": 24442,
+      "▁tobacco": 24443,
+      "▁passionate": 24444,
+      "▁bil": 24445,
+      "ercul": 24446,
+      "▁Release": 24447,
+      "▁emotions": 24448,
+      "Env": 24449,
+      "▁Rain": 24450,
+      "▁starring": 24451,
+      "▁threats": 24452,
+      "兴奋": 24453,
+      "质疑": 24454,
+      "sal": 24455,
+      "安置": 24456,
+      "eer": 24457,
+      "great": 24458,
+      "▁Role": 24459,
+      "▁substrates": 24460,
+      "外人": 24461,
+      "alias": 24462,
+      "'))": 24463,
+      "▁aluminum": 24464,
+      "▁confirmation": 24465,
+      "监护的": 24466,
+      "▁prox": 24467,
+      "EventListener": 24468,
+      "劳动力": 24469,
+      "尼斯": 24470,
+      "▁adhesion": 24471,
+      "MAGE": 24472,
+      "▁SDK": 24473,
+      "▁prospect": 24474,
+      "gp": 24475,
+      "用地": 24476,
+      "都被": 24477,
+      "uli": 24478,
+      "离婚纠纷一案": 24479,
+      "▁briefly": 24480,
+      "在全国": 24481,
+      "timer": 24482,
+      "▁Mother": 24483,
+      "▁Seg": 24484,
+      "Material": 24485,
+      "▁thrown": 24486,
+      "向本院申请复议一次": 24487,
+      "▁Bio": 24488,
+      "科学记数": 24489,
+      "▁Ka": 24490,
+      "中小企业": 24491,
+      "illiam": 24492,
+      "舆论": 24493,
+      "▁retrospective": 24494,
+      "比较好": 24495,
+      "▁hipp": 24496,
+      "▁islands": 24497,
+      "▁vintage": 24498,
+      "取出": 24499,
+      "Symbol": 24500,
+      "▁Settings": 24501,
+      "夫人": 24502,
+      "Feed": 24503,
+      "容纳": 24504,
+      "法治": 24505,
+      "话说": 24506,
+      "▁SER": 24507,
+      "IOException": 24508,
+      "▁contributor": 24509,
+      "apanese": 24510,
+      "也好": 24511,
+      "的作品": 24512,
+      "涉及未成年": 24513,
+      "▁everybody": 24514,
+      "对错": 24515,
+      "▁Advent": 24516,
+      "▁productive": 24517,
+      "克兰": 24518,
+      "<>();\\": 24519,
+      "controlled": 24520,
+      "暴露": 24521,
+      "iptic": 24522,
+      "rst": 24523,
+      "得知": 24524,
+      "▁nations": 24525,
+      "▁Vin": 24526,
+      "▁logical": 24527,
+      "▁Tun": 24528,
+      "Wait": 24529,
+      "▁fatal": 24530,
+      "▁Morgan": 24531,
+      "vg": 24532,
+      "Have": 24533,
+      "Optional": 24534,
+      "这点": 24535,
+      "sam": 24536,
+      "▁(,": 24537,
+      "的经济": 24538,
+      "厦门市": 24539,
+      "▁attempting": 24540,
+      "分散": 24541,
+      "ovies": 24542,
+      "▁Mayor": 24543,
+      "▁м": 24544,
+      "▁adjustment": 24545,
+      "又不": 24546,
+      "nMon": 24547,
+      "你自己": 24548,
+      "ceiver": 24549,
+      "的范围内": 24550,
+      "ividually": 24551,
+      "hire": 24552,
+      "▁dimin": 24553,
+      "▁Taiwan": 24554,
+      "▁portable": 24555,
+      "涉及未成年子女抚养": 24556,
+      "NY": 24557,
+      "可行": 24558,
+      "disabled": 24559,
+      "nType": 24560,
+      "为大家": 24561,
+      "离婚诉讼或者": 24562,
+      "▁Golf": 24563,
+      "▁socio": 24564,
+      "atum": 24565,
+      "nStart": 24566,
+      "离婚诉讼或者涉及未成年子女抚养": 24567,
+      "▁readily": 24568,
+      "▁sour": 24569,
+      "compass": 24570,
+      "atorial": 24571,
+      "这本书": 24572,
+      "▁formulation": 24573,
+      "成品": 24574,
+      "issa": 24575,
+      "maps": 24576,
+      "兼容": 24577,
+      "无力": 24578,
+      "没事": 24579,
+      "更大的": 24580,
+      "班子": 24581,
+      "Site": 24582,
+      "▁eryth": 24583,
+      "实质": 24584,
+      "精选": 24585,
+      "LOAD": 24586,
+      "▁refund": 24587,
+      "计算至": 24588,
+      "▁suppression": 24589,
+      "ulu": 24590,
+      "给出的": 24591,
+      "忽视": 24592,
+      "年开始": 24593,
+      "bundle": 24594,
+      "利的": 24595,
+      "kx": 24596,
+      "就可以了": 24597,
+      "提供担保": 24598,
+      "边界": 24599,
+      "▁expressing": 24600,
+      "本院依法组成合议庭进行了审理": 24601,
+      "economic": 24602,
+      "合伙": 24603,
+      "向前": 24604,
+      "ocamp": 24605,
+      "▁proton": 24606,
+      "精华": 24607,
+      "▁$[": 24608,
+      "annon": 24609,
+      "hesia": 24610,
+      "▁tourism": 24611,
+      "wei": 24612,
+      "comb": 24613,
+      "南通": 24614,
+      "▁Format": 24615,
+      "▁equality": 24616,
+      "▁nullable": 24617,
+      "Background": 24618,
+      "▁Rick": 24619,
+      "▁pole": 24620,
+      "叶片": 24621,
+      "包含了": 24622,
+      "同比增长": 24623,
+      "▁highway": 24624,
+      "▁swing": 24625,
+      "osterone": 24626,
+      "▁BSD": 24627,
+      "▁Script": 24628,
+      "Cmd": 24629,
+      "完整的": 24630,
+      "▁Things": 24631,
+      "▁Related": 24632,
+      "'))\\": 24633,
+      "▁WHERE": 24634,
+      "▁Dublin": 24635,
+      "凝聚": 24636,
+      "Duration": 24637,
+      "▁Industrial": 24638,
+      "▁subprocess": 24639,
+      "▁Louisiana": 24640,
+      "▁macrophages": 24641,
+      "▁Summ": 24642,
+      "nPub": 24643,
+      "▁Pharm": 24644,
+      "WD": 24645,
+      "owski": 24646,
+      "家乡": 24647,
+      "ipy": 24648,
+      "quet": 24649,
+      "DTO": 24650,
+      "▁Bush": 24651,
+      "urgery": 24652,
+      "▁hired": 24653,
+      "原子": 24654,
+      "▁Wy": 24655,
+      "▁permitted": 24656,
+      "bie": 24657,
+      "Messages": 24658,
+      "～\\": 24659,
+      "VICE": 24660,
+      "二次方程": 24661,
+      "nLike": 24662,
+      "ла": 24663,
+      "▁landsc": 24664,
+      "一起来": 24665,
+      "▁Phill": 24666,
+      "ogeneity": 24667,
+      "▁SF": 24668,
+      "▁coordination": 24669,
+      "%）": 24670,
+      "chunk": 24671,
+      "GV": 24672,
+      "词典": 24673,
+      "的前提": 24674,
+      "▁hypotheses": 24675,
+      "▁Nep": 24676,
+      "GitHub": 24677,
+      "▁chromatography": 24678,
+      "▁\"\",": 24679,
+      "/${": 24680,
+      "语法": 24681,
+      "itations": 24682,
+      "▁mounted": 24683,
+      "hosts": 24684,
+      "▁endogenous": 24685,
+      "十六日": 24686,
+      "可怕": 24687,
+      "签署": 24688,
+      "▁Definition": 24689,
+      "▁pepper": 24690,
+      "▁deployed": 24691,
+      "▁^{": 24692,
+      "不断的": 24693,
+      "▁SET": 24694,
+      "▁contrad": 24695,
+      "idel": 24696,
+      "oidal": 24697,
+      "▁impression": 24698,
+      "▁gross": 24699,
+      "monitor": 24700,
+      "▁keyword": 24701,
+      "▁cylinder": 24702,
+      "迎来": 24703,
+      "再根据": 24704,
+      "在他": 24705,
+      "business": 24706,
+      "▁stainless": 24707,
+      "未到庭": 24708,
+      "▁controversial": 24709,
+      "ubuntu": 24710,
+      "▁Works": 24711,
+      "▁Harvard": 24712,
+      "nbe": 24713,
+      "urchase": 24714,
+      "prises": 24715,
+      "▁Medal": 24716,
+      "HI": 24717,
+      "Watch": 24718,
+      "agers": 24719,
+      "\",&": 24720,
+      "下一个": 24721,
+      "组长": 24722,
+      "由原告负担": 24723,
+      "requests": 24724,
+      "irs": 24725,
+      "的公司": 24726,
+      "album": 24727,
+      "▁moisture": 24728,
+      "剂量": 24729,
+      "服法": 24730,
+      "▁Snow": 24731,
+      "commons": 24732,
+      "ographics": 24733,
+      "小麦": 24734,
+      "uki": 24735,
+      "oblast": 24736,
+      "活着": 24737,
+      "compose": 24738,
+      "▁varieties": 24739,
+      "局面": 24740,
+      "执行的": 24741,
+      "这家": 24742,
+      "▁thinks": 24743,
+      "ificant": 24744,
+      "一一": 24745,
+      "hots": 24746,
+      "▁lar": 24747,
+      "长春市": 24748,
+      "队长": 24749,
+      "▁longest": 24750,
+      "开车": 24751,
+      "▁fulfill": 24752,
+      "▁Bey": 24753,
+      "极限": 24754,
+      "双方当事人": 24755,
+      "▁employer": 24756,
+      "▁gradually": 24757,
+      "▁mammalian": 24758,
+      "▁infectious": 24759,
+      "▁Risk": 24760,
+      "▁slice": 24761,
+      "▁polynomials": 24762,
+      "王者": 24763,
+      "'],": 24764,
+      "采用了": 24765,
+      "▁corn": 24766,
+      "▁slip": 24767,
+      "验收": 24768,
+      "metrics": 24769,
+      "▁bund": 24770,
+      "▁Portland": 24771,
+      "潮流": 24772,
+      "▁chunk": 24773,
+      "',['": 24774,
+      "遍历": 24775,
+      "mix": 24776,
+      "二等奖": 24777,
+      "Good": 24778,
+      "aura": 24779,
+      "mitt": 24780,
+      "Submit": 24781,
+      "东路": 24782,
+      "IENT": 24783,
+      "村民委员会": 24784,
+      "卫生间": 24785,
+      "imore": 24786,
+      "▁linking": 24787,
+      "▁attitudes": 24788,
+      "日向本院申请执行": 24789,
+      "▁XX": 24790,
+      "aca": 24791,
+      "nOk": 24792,
+      "borne": 24793,
+      "because": 24794,
+      "承办": 24795,
+      "重生": 24796,
+      "glob": 24797,
+      "sil": 24798,
+      "座谈": 24799,
+      "▁Ju": 24800,
+      "▁trailer": 24801,
+      "▁weren": 24802,
+      "atility": 24803,
+      "死的": 24804,
+      "加深": 24805,
+      "▁globe": 24806,
+      "海上": 24807,
+      "nHer": 24808,
+      "远离": 24809,
+      "CES": 24810,
+      "出于": 24811,
+      "▁\\\\((": 24812,
+      "▁Wind": 24813,
+      "rx": 24814,
+      "tfmt": 24815,
+      "城市的": 24816,
+      "prite": 24817,
+      "Master": 24818,
+      "中原": 24819,
+      "published": 24820,
+      "▁dens": 24821,
+      "▁hazard": 24822,
+      "amo": 24823,
+      "Formatter": 24824,
+      "▁fluor": 24825,
+      "▁bandwidth": 24826,
+      "▁benchmark": 24827,
+      "Pi": 24828,
+      "▁Abs": 24829,
+      "iti": 24830,
+      "▁Josh": 24831,
+      "osine": 24832,
+      "牡丹": 24833,
+      "▁AWS": 24834,
+      "▁Alliance": 24835,
+      "ku": 24836,
+      "gender": 24837,
+      "成语": 24838,
+      "ndescription": 24839,
+      "town": 24840,
+      "▁Gaussian": 24841,
+      "▁interfaces": 24842,
+      "Second": 24843,
+      "▁Ach": 24844,
+      "anne": 24845,
+      "evalu": 24846,
+      "不为": 24847,
+      "的有效": 24848,
+      "▁examining": 24849,
+      "不一定": 24850,
+      "▁Cool": 24851,
+      "▁recurrent": 24852,
+      "Pack": 24853,
+      "日立案后": 24854,
+      "isson": 24855,
+      "▁durable": 24856,
+      "劳动争议": 24857,
+      "意思表示": 24858,
+      "了几": 24859,
+      "▁gotten": 24860,
+      "▁Schools": 24861,
+      "_;\\": 24862,
+      "▁题目描述": 24863,
+      "Che": 24864,
+      "\\'/": 24865,
+      "解析式": 24866,
+      "十二日": 24867,
+      "索引": 24868,
+      "过渡": 24869,
+      "▁SI": 24870,
+      "Bad": 24871,
+      "▁Tab": 24872,
+      "▁nurse": 24873,
+      "▁Immun": 24874,
+      "nChar": 24875,
+      "▁Stew": 24876,
+      "clinical": 24877,
+      "骑士": 24878,
+      "正面": 24879,
+      "约定的": 24880,
+      "需要注意": 24881,
+      "▁simpler": 24882,
+      "过多": 24883,
+      "限于": 24884,
+      "percent": 24885,
+      "摩擦": 24886,
+      "issing": 24887,
+      "▁Chairman": 24888,
+      "▁«": 24889,
+      "涂料": 24890,
+      "listen": 24891,
+      "▁Complete": 24892,
+      "本土": 24893,
+      "就已经": 24894,
+      "▁germ": 24895,
+      "精灵": 24896,
+      "Struct": 24897,
+      "▁promotes": 24898,
+      "▁hypox": 24899,
+      "▁fork": 24900,
+      "▁Token": 24901,
+      "从未": 24902,
+      "LAB": 24903,
+      "▁prognosis": 24904,
+      "▁physically": 24905,
+      "付清": 24906,
+      "欢乐": 24907,
+      "constants": 24908,
+      "▁Being": 24909,
+      "眼里": 24910,
+      "▁Bot": 24911,
+      "▁Coord": 24912,
+      "▁struck": 24913,
+      "猪肉": 24914,
+      "▁paste": 24915,
+      "nBest": 24916,
+      "nEnter": 24917,
+      "WriteLine": 24918,
+      "▁instanceof": 24919,
+      "PLC": 24920,
+      "七日内": 24921,
+      "▁diagonal": 24922,
+      "▁Nan": 24923,
+      "月二十二日": 24924,
+      "▁referring": 24925,
+      "TB": 24926,
+      "去的": 24927,
+      "的图象": 24928,
+      "该如何": 24929,
+      "手法": 24930,
+      ">\"\\": 24931,
+      "第七章": 24932,
+      "）÷": 24933,
+      "▁shaped": 24934,
+      "▁cortical": 24935,
+      "nMake": 24936,
+      "养成": 24937,
+      "▁regarded": 24938,
+      "▁displacement": 24939,
+      "placeholder": 24940,
+      "Internal": 24941,
+      "产生了": 24942,
+      "实际情况": 24943,
+      "▁°": 24944,
+      "说过": 24945,
+      "▁Hub": 24946,
+      "▁LIMITED": 24947,
+      "▁loose": 24948,
+      "一个小": 24949,
+      "湿度": 24950,
+      "▁Administr": 24951,
+      "一层": 24952,
+      "武装": 24953,
+      "给定": 24954,
+      "我有": 24955,
+      "nComp": 24956,
+      "▁captain": 24957,
+      "▁pixels": 24958,
+      "▁racial": 24959,
+      "一组": 24960,
+      "外面": 24961,
+      "AST": 24962,
+      "▁Baby": 24963,
+      "▁dimensional": 24964,
+      "日晚": 24965,
+      "登场": 24966,
+      "▁tunnel": 24967,
+      "trigger": 24968,
+      "▁Labour": 24969,
+      "不止": 24970,
+      "north": 24971,
+      "▁Howard": 24972,
+      "工委": 24973,
+      "▁bash": 24974,
+      "物业服务有限公司": 24975,
+      "▁az": 24976,
+      "最后的": 24977,
+      "第三百": 24978,
+      "了他": 24979,
+      "美观": 24980,
+      "▁Saudi": 24981,
+      "▁stayed": 24982,
+      "▁Parent": 24983,
+      "▁Technical": 24984,
+      "意志": 24985,
+      "▁dispar": 24986,
+      "▁cultiv": 24987,
+      "▁[(": 24988,
+      "折抵": 24989,
+      "half": 24990,
+      "'));\\": 24991,
+      "之旅": 24992,
+      "▁Location": 24993,
+      "▁arrange": 24994,
+      "added": 24995,
+      "ophag": 24996,
+      "北部": 24997,
+      "▁outbreak": 24998,
+      "▁confusion": 24999,
+      "spir": 25000,
+      "▁Ask": 25001,
+      "▁presentations": 25002,
+      "拉斯": 25003,
+      "里斯": 25004,
+      "bps": 25005,
+      "受害": 25006,
+      "than": 25007,
+      "ioid": 25008,
+      "▁Roc": 25009,
+      "dw": 25010,
+      "▁elabor": 25011,
+      "Compat": 25012,
+      "▁(_": 25013,
+      "Copy": 25014,
+      "▁consensus": 25015,
+      "▁tired": 25016,
+      "于本判决生效后": 25017,
+      "乃至": 25018,
+      "宽度": 25019,
+      "刑更": 25020,
+      "▁distr": 25021,
+      "▁Mock": 25022,
+      "▁credits": 25023,
+      "▁timer": 25024,
+      "Release": 25025,
+      "doi": 25026,
+      "立法": 25027,
+      "▁achievement": 25028,
+      "才有": 25029,
+      "aler": 25030,
+      "▁Got": 25031,
+      "▁stiff": 25032,
+      "拿出": 25033,
+      "Thanks": 25034,
+      "▁placing": 25035,
+      "荷兰": 25036,
+      "了解到": 25037,
+      "wear": 25038,
+      "nSimplify": 25039,
+      "panic": 25040,
+      "▁hiring": 25041,
+      "ologous": 25042,
+      "Strategy": 25043,
+      "▁devoted": 25044,
+      "移植": 25045,
+      "nSet": 25046,
+      "合作社": 25047,
+      "▁prolonged": 25048,
+      "ncy": 25049,
+      "▁Galaxy": 25050,
+      "iao": 25051,
+      "▁amen": 25052,
+      "▁presum": 25053,
+      "\\\\()\\\\)\\": 25054,
+      "改革开放": 25055,
+      "联想": 25056,
+      "claimed": 25057,
+      "lr": 25058,
+      "isco": 25059,
+      "iologic": 25060,
+      "▁formerly": 25061,
+      "▁letting": 25062,
+      "重启": 25063,
+      "]\\\\": 25064,
+      "nCONCLUSIONS": 25065,
+      "rans": 25066,
+      "寂寞": 25067,
+      "▁suspended": 25068,
+      "太大": 25069,
+      "感恩": 25070,
+      "HF": 25071,
+      "病理": 25072,
+      "的态度": 25073,
+      "语种": 25074,
+      "▁Mississippi": 25075,
+      "救助": 25076,
+      "▁Recently": 25077,
+      "放到": 25078,
+      "▁Package": 25079,
+      "student": 25080,
+      "▁\"\";\\": 25081,
+      "八条": 25082,
+      "平常": 25083,
+      "选出正确的答案": 25084,
+      "有理": 25085,
+      "▁Rab": 25086,
+      "统筹": 25087,
+      "makers": 25088,
+      "梅花": 25089,
+      "缩小": 25090,
+      "▁varies": 25091,
+      "考虑到": 25092,
+      "▁TX": 25093,
+      "throws": 25094,
+      "象征": 25095,
+      "欧盟": 25096,
+      "内容介绍": 25097,
+      "▁protecting": 25098,
+      "tensor": 25099,
+      "▁Runtime": 25100,
+      "命名": 25101,
+      "▁din": 25102,
+      "izoph": 25103,
+      "▁DateTime": 25104,
+      "setting": 25105,
+      "占据": 25106,
+      "@@": 25107,
+      "人民医院": 25108,
+      "problem": 25109,
+      "performance": 25110,
+      "▁Vel": 25111,
+      "▁platelet": 25112,
+      "目光": 25113,
+      "▁Nut": 25114,
+      "▁aux": 25115,
+      "是没有": 25116,
+      "nBoth": 25117,
+      "Integr": 25118,
+      "nObject": 25119,
+      "special": 25120,
+      "▁organisms": 25121,
+      "▁entert": 25122,
+      "▁Sud": 25123,
+      "▁abundance": 25124,
+      "排水": 25125,
+      "▁accordance": 25126,
+      "这是一": 25127,
+      "▁Rule": 25128,
+      "▁Gun": 25129,
+      "GBT": 25130,
+      "▁suddenly": 25131,
+      "HEAD": 25132,
+      "的首": 25133,
+      "emed": 25134,
+      "compare": 25135,
+      "attrs": 25136,
+      "▁dB": 25137,
+      "面板": 25138,
+      "▁THAT": 25139,
+      "gebras": 25140,
+      "tax": 25141,
+      "▁Orth": 25142,
+      "confirm": 25143,
+      "重建": 25144,
+      "▁excessive": 25145,
+      "▁阿": 25146,
+      "▁introducing": 25147,
+      "登陆": 25148,
+      "船舶": 25149,
+      "proto": 25150,
+      "Lab": 25151,
+      "映时间": 25152,
+      "nstatic": 25153,
+      "▁counting": 25154,
+      "我个人": 25155,
+      "而来": 25156,
+      "▁begun": 25157,
+      "班主任": 25158,
+      "▁crown": 25159,
+      "书写": 25160,
+      "Binding": 25161,
+      "▁carries": 25162,
+      "▁headquarters": 25163,
+      "自从": 25164,
+      "▁jo": 25165,
+      "▁decent": 25166,
+      "▁ovarian": 25167,
+      "月二十九日": 25168,
+      "▁Jos": 25169,
+      "▁Django": 25170,
+      "商品房": 25171,
+      "十七日": 25172,
+      "经费": 25173,
+      "▁explores": 25174,
+      "真诚": 25175,
+      "▁Lam": 25176,
+      "▁foss": 25177,
+      "▁Veter": 25178,
+      "新人": 25179,
+      "heduler": 25180,
+      "Side": 25181,
+      "觉得自己": 25182,
+      "党校": 25183,
+      "into": 25184,
+      "VALUE": 25185,
+      "isi": 25186,
+      "经济社会": 25187,
+      "aurant": 25188,
+      "`,\\": 25189,
+      "的不同": 25190,
+      "▁SL": 25191,
+      "FLAGS": 25192,
+      "▁survived": 25193,
+      "▁Rober": 25194,
+      "如同": 25195,
+      "十二章": 25196,
+      "\\');": 25197,
+      "CTYPE": 25198,
+      "▁Philip": 25199,
+      "公积": 25200,
+      "▁Detroit": 25201,
+      "十三日": 25202,
+      "▁ein": 25203,
+      "▁Manufact": 25204,
+      "amped": 25205,
+      "reverse": 25206,
+      "▁Events": 25207,
+      "▁frustr": 25208,
+      "具体的": 25209,
+      "Information": 25210,
+      "▁donation": 25211,
+      "music": 25212,
+      "▁Available": 25213,
+      "学士": 25214,
+      "监会": 25215,
+      "irable": 25216,
+      "时不": 25217,
+      "▁Conclusions": 25218,
+      "▁bedrooms": 25219,
+      "ounding": 25220,
+      "恐惧": 25221,
+      "abul": 25222,
+      "的全": 25223,
+      "眼神": 25224,
+      "配方": 25225,
+      "在本": 25226,
+      "separ": 25227,
+      "▁stocks": 25228,
+      "▁Alan": 25229,
+      "▁Heritage": 25230,
+      "disciplinary": 25231,
+      "着力": 25232,
+      "lag": 25233,
+      "▁monitored": 25234,
+      "][\"": 25235,
+      "▁将": 25236,
+      "▁beds": 25237,
+      "▁occasionally": 25238,
+      "▁Hockey": 25239,
+      "▁striking": 25240,
+      "太平洋": 25241,
+      "▁Mini": 25242,
+      "▁specifications": 25243,
+      "▁Talk": 25244,
+      "▁localized": 25245,
+      "▁tablet": 25246,
+      "▁similarity": 25247,
+      "▁tear": 25248,
+      "公元": 25249,
+      "后卫": 25250,
+      "▁streams": 25251,
+      "▁bypass": 25252,
+      "▁abdominal": 25253,
+      "▁fragments": 25254,
+      "发性": 25255,
+      "▁Wire": 25256,
+      "usk": 25257,
+      "学家": 25258,
+      "ogonal": 25259,
+      "▁guilty": 25260,
+      "▁merely": 25261,
+      "▁artwork": 25262,
+      "并无": 25263,
+      "▁addiction": 25264,
+      "啤酒": 25265,
+      "ipeline": 25266,
+      "Sch": 25267,
+      "online": 25268,
+      "ufficient": 25269,
+      "▁distances": 25270,
+      "双手": 25271,
+      "▁Albert": 25272,
+      "addy": 25273,
+      "ifiable": 25274,
+      "pmatrix": 25275,
+      "细腻": 25276,
+      "调控": 25277,
+      "和管理": 25278,
+      "Mixin": 25279,
+      "ainers": 25280,
+      "▁Colomb": 25281,
+      "▁refused": 25282,
+      "▁propri": 25283,
+      "▁musician": 25284,
+      "ohyd": 25285,
+      "▁burst": 25286,
+      "▁Remember": 25287,
+      "▁Serial": 25288,
+      "bp": 25289,
+      "明确的": 25290,
+      "▁infinity": 25291,
+      "减刑条件": 25292,
+      "▁$|": 25293,
+      "▁ana": 25294,
+      "immune": 25295,
+      "Processor": 25296,
+      "▁substances": 25297,
+      "外形": 25298,
+      "破碎": 25299,
+      "判处拘役": 25300,
+      "向本院提起上诉": 25301,
+      "▁decomposition": 25302,
+      "给我们": 25303,
+      "▁photographer": 25304,
+      "▁Kin": 25305,
+      "▁Walker": 25306,
+      "▁belonging": 25307,
+      "借款人": 25308,
+      "illo": 25309,
+      "▁blocking": 25310,
+      "向本院申请执行": 25311,
+      "就是一个": 25312,
+      "▁Carib": 25313,
+      "▁shirt": 25314,
+      "regex": 25315,
+      "▁Break": 25316,
+      "[$": 25317,
+      "计数": 25318,
+      "▁Fixed": 25319,
+      "一度": 25320,
+      "繁荣": 25321,
+      "一体化": 25322,
+      "甲基": 25323,
+      "▁padding": 25324,
+      "▁Arthur": 25325,
+      "高等教育": 25326,
+      "tokens": 25327,
+      "标准的": 25328,
+      "Protocol": 25329,
+      "弘扬": 25330,
+      "▁Dick": 25331,
+      "▁Entity": 25332,
+      "用人": 25333,
+      "▁hell": 25334,
+      "▁tours": 25335,
+      "举措": 25336,
+      "))/(": 25337,
+      "▁hesit": 25338,
+      "concept": 25339,
+      "▁Past": 25340,
+      "owners": 25341,
+      ":\\'": 25342,
+      "ylation": 25343,
+      "ilization": 25344,
+      "▁sons": 25345,
+      "▁Belgium": 25346,
+      "▁pushing": 25347,
+      "碰撞": 25348,
+      "的速度": 25349,
+      "▁rendering": 25350,
+      "▁horses": 25351,
+      "▁运动": 25352,
+      "▁Merc": 25353,
+      "变革": 25354,
+      "ycombinator": 25355,
+      "nVillages": 25356,
+      "▁regulating": 25357,
+      "认罪服法": 25358,
+      "är": 25359,
+      "Does": 25360,
+      "▁Mars": 25361,
+      "▁recombinant": 25362,
+      "tox": 25363,
+      "Center": 25364,
+      "年间": 25365,
+      "日立案受理": 25366,
+      "体质": 25367,
+      "十四日": 25368,
+      "▁Kids": 25369,
+      "▁Track": 25370,
+      "有一定的": 25371,
+      "▁aggregate": 25372,
+      "▁occasions": 25373,
+      "▁resection": 25374,
+      "酒店的": 25375,
+      "涉案": 25376,
+      "▁Finland": 25377,
+      "▁hypothesize": 25378,
+      "leveland": 25379,
+      "▁swap": 25380,
+      "▁Previous": 25381,
+      "促销": 25382,
+      "分析和": 25383,
+      "▁Dakota": 25384,
+      "阐述": 25385,
+      "▁god": 25386,
+      "▁rejected": 25387,
+      "份额": 25388,
+      "▁Agreement": 25389,
+      "两侧": 25390,
+      "▁Choose": 25391,
+      "hole": 25392,
+      "izophren": 25393,
+      "Direction": 25394,
+      "▁Death": 25395,
+      "chell": 25396,
+      "VED": 25397,
+      "leyball": 25398,
+      "creens": 25399,
+      "ANGE": 25400,
+      "▁Rio": 25401,
+      "external": 25402,
+      "nTherefore": 25403,
+      "气管": 25404,
+      "▁tuple": 25405,
+      "给她": 25406,
+      "puts": 25407,
+      "ohist": 25408,
+      "▁Tests": 25409,
+      "▁lombok": 25410,
+      "enerative": 25411,
+      "]}": 25412,
+      "以上事实": 25413,
+      "网卡": 25414,
+      "行政处罚决定书": 25415,
+      "LOC": 25416,
+      "▁reproductive": 25417,
+      "BY": 25418,
+      "用量": 25419,
+      "注入": 25420,
+      "▁Lane": 25421,
+      "▁基本信息": 25422,
+      "看似": 25423,
+      "heastern": 25424,
+      "属实": 25425,
+      "phot": 25426,
+      "▁Qual": 25427,
+      "▁licenses": 25428,
+      "nUnder": 25429,
+      "▁enemy": 25430,
+      "▁Denmark": 25431,
+      "电动车": 25432,
+      "▁weapon": 25433,
+      "anche": 25434,
+      "配送": 25435,
+      "GROUND": 25436,
+      "▁Bowl": 25437,
+      "endpoint": 25438,
+      "应当按照": 25439,
+      "▁AG": 25440,
+      "厕所": 25441,
+      "时间为": 25442,
+      "▁seiz": 25443,
+      "▁Franklin": 25444,
+      "▁employers": 25445,
+      "▁liv": 25446,
+      "人气": 25447,
+      "约为": 25448,
+      "▁NCAA": 25449,
+      "ATCH": 25450,
+      "麦克": 25451,
+      "▁Dom": 25452,
+      "▁\\\\$": 25453,
+      "▁Nash": 25454,
+      "▁Lux": 25455,
+      "ны": 25456,
+      "▁\"[": 25457,
+      "▁poet": 25458,
+      "桂林": 25459,
+      "oba": 25460,
+      "▁constitution": 25461,
+      "iii": 25462,
+      "homepage": 25463,
+      "欧美": 25464,
+      "▁revised": 25465,
+      "nde": 25466,
+      "▁focal": 25467,
+      "下方": 25468,
+      "▁Root": 25469,
+      "▁Vancouver": 25470,
+      "▁permut": 25471,
+      "▁Behavior": 25472,
+      "面貌": 25473,
+      "later": 25474,
+      "说出": 25475,
+      "催化": 25476,
+      "百货": 25477,
+      "Live": 25478,
+      "▁Trib": 25479,
+      "за": 25480,
+      "\\'\\\\": 25481,
+      "\"}\\": 25482,
+      "▁confront": 25483,
+      "▁CB": 25484,
+      "▁cls": 25485,
+      "▁maximize": 25486,
+      "osures": 25487,
+      "解放军": 25488,
+      "▁CV": 25489,
+      "发现了": 25490,
+      "nOBJECT": 25491,
+      "▁navigate": 25492,
+      "▁loud": 25493,
+      "iche": 25494,
+      "▁conditioning": 25495,
+      "Ra": 25496,
+      "▁时": 25497,
+      "进球": 25498,
+      "▁defensive": 25499,
+      "tsburgh": 25500,
+      "火灾": 25501,
+      "▁Movie": 25502,
+      "▁ather": 25503,
+      "▁Hz": 25504,
+      "第一百一十九": 25505,
+      "▁Zone": 25506,
+      "established": 25507,
+      "bital": 25508,
+      "launch": 25509,
+      "inburgh": 25510,
+      "enezuel": 25511,
+      "样品": 25512,
+      "再说": 25513,
+      "法律保护": 25514,
+      "▁Money": 25515,
+      "nAbout": 25516,
+      "的教育": 25517,
+      "月二十一日": 25518,
+      "上半年": 25519,
+      "nCanadian": 25520,
+      "▁globally": 25521,
+      "等比": 25522,
+      "青岛市": 25523,
+      "▁hung": 25524,
+      "▁Arkansas": 25525,
+      "贯彻落实": 25526,
+      "不下": 25527,
+      "▁Quebec": 25528,
+      "tM": 25529,
+      "区分": 25530,
+      "▁stomach": 25531,
+      "▁fruits": 25532,
+      "▁logged": 25533,
+      "圆锥": 25534,
+      "▁Sel": 25535,
+      "后面的": 25536,
+      "forced": 25537,
+      "▁Flash": 25538,
+      "的通": 25539,
+      "nFollowing": 25540,
+      "股市": 25541,
+      "▁Jenn": 25542,
+      "道路上": 25543,
+      "ttry": 25544,
+      "▁scaling": 25545,
+      "是大": 25546,
+      "存在的问题": 25547,
+      "努力完成劳动任务": 25548,
+      "community": 25549,
+      "▁Magic": 25550,
+      "without": 25551,
+      "前列": 25552,
+      "▁Chat": 25553,
+      "▁coins": 25554,
+      "春天": 25555,
+      "▁foo": 25556,
+      "nem": 25557,
+      "ени": 25558,
+      "forge": 25559,
+      "nAwards": 25560,
+      "▁merch": 25561,
+      "明月": 25562,
+      "必须要": 25563,
+      "persistence": 25564,
+      "依法由审判员": 25565,
+      "路面": 25566,
+      "▁bucket": 25567,
+      "▁DEFAULT": 25568,
+      "▁narrative": 25569,
+      "发了": 25570,
+      "希腊": 25571,
+      "疲劳": 25572,
+      "观测": 25573,
+      "买房": 25574,
+      "师资": 25575,
+      "▁Verm": 25576,
+      "ocrine": 25577,
+      "▁agenda": 25578,
+      "Analysis": 25579,
+      "▁ancest": 25580,
+      "▁yoga": 25581,
+      "Amount": 25582,
+      "和服务": 25583,
+      "floor": 25584,
+      "事实和": 25585,
+      "nextern": 25586,
+      "▁Struct": 25587,
+      "▁Military": 25588,
+      "物体": 25589,
+      "▁wheels": 25590,
+      "▁helper": 25591,
+      "名人": 25592,
+      "▁supporters": 25593,
+      "erranean": 25594,
+      "满满": 25595,
+      "scriber": 25596,
+      "储存": 25597,
+      "▁gathering": 25598,
+      "桌面": 25599,
+      "COM": 25600,
+      "father": 25601,
+      "草原": 25602,
+      "decor": 25603,
+      "▁LL": 25604,
+      "Speed": 25605,
+      "▁Clear": 25606,
+      "FD": 25607,
+      "吸毒": 25608,
+      "erald": 25609,
+      "残疾人": 25610,
+      "▁Numer": 25611,
+      ">,\\": 25612,
+      "▁somebody": 25613,
+      "评选": 25614,
+      "▁Pick": 25615,
+      "UA": 25616,
+      "仅供": 25617,
+      "ighters": 25618,
+      "\"\\'": 25619,
+      "子上": 25620,
+      "白天": 25621,
+      "一时": 25622,
+      "▁Audio": 25623,
+      "Ga": 25624,
+      "▁handles": 25625,
+      "ionic": 25626,
+      "tuple": 25627,
+      "缩短": 25628,
+      "olang": 25629,
+      "▁receives": 25630,
+      "以其": 25631,
+      "福州": 25632,
+      "▁flood": 25633,
+      "Resolver": 25634,
+      "▁Austria": 25635,
+      "勇敢": 25636,
+      "tom": 25637,
+      "▁celebrated": 25638,
+      "月二十四日": 25639,
+      "ONG": 25640,
+      "单价": 25641,
+      "号刑事裁定": 25642,
+      "ozilla": 25643,
+      "▁Mind": 25644,
+      "▁temple": 25645,
+      "内在": 25646,
+      "中毒": 25647,
+      "▁mutual": 25648,
+      "▁metals": 25649,
+      "▁nanoparticles": 25650,
+      "▁Experience": 25651,
+      "▁configured": 25652,
+      "先锋": 25653,
+      "教研": 25654,
+      "otypic": 25655,
+      "▁secrets": 25656,
+      "就读": 25657,
+      "exceptions": 25658,
+      "▁Vent": 25659,
+      "About": 25660,
+      "内地": 25661,
+      "herent": 25662,
+      "▁coastal": 25663,
+      "膨胀": 25664,
+      "offee": 25665,
+      "encer": 25666,
+      "toEqual": 25667,
+      "▁skeletal": 25668,
+      "户型": 25669,
+      "▁#{": 25670,
+      "olia": 25671,
+      "qa": 25672,
+      "▁leukemia": 25673,
+      "▁Pod": 25674,
+      "▁Bart": 25675,
+      "tier": 25676,
+      "▁lean": 25677,
+      "encil": 25678,
+      "▁Memory": 25679,
+      "▁Transform": 25680,
+      "我对": 25681,
+      "redux": 25682,
+      "真空": 25683,
+      "TL": 25684,
+      "▁dur": 25685,
+      "▁Attorney": 25686,
+      "感的": 25687,
+      "Flag": 25688,
+      "▁supplier": 25689,
+      "十一条": 25690,
+      "yer": 25691,
+      ">}": 25692,
+      "与此": 25693,
+      "▁customized": 25694,
+      "▁发行": 25695,
+      "▁guided": 25696,
+      "发热": 25697,
+      "月生": 25698,
+      "GHz": 25699,
+      "▁vi": 25700,
+      "},\\\\": 25701,
+      "私服": 25702,
+      "▁succeeded": 25703,
+      "乐趣": 25704,
+      "▁Portugal": 25705,
+      "美国的": 25706,
+      "▁limiting": 25707,
+      "导弹": 25708,
+      "(['": 25709,
+      "cov": 25710,
+      "nMark": 25711,
+      "erculosis": 25712,
+      "总额": 25713,
+      "pher": 25714,
+      "oscopy": 25715,
+      "vendor": 25716,
+      "标识": 25717,
+      "这份": 25718,
+      "▁pose": 25719,
+      "左边": 25720,
+      "氨基": 25721,
+      "▁trivial": 25722,
+      "▁Otherwise": 25723,
+      "▁suspension": 25724,
+      "ctl": 25725,
+      "▁literary": 25726,
+      "nick": 25727,
+      "我只": 25728,
+      "▁individually": 25729,
+      "liver": 25730,
+      "Sequence": 25731,
+      "▁Lawrence": 25732,
+      "aceful": 25733,
+      "栏目": 25734,
+      "nPlace": 25735,
+      "抱着": 25736,
+      "uario": 25737,
+      "▁VS": 25738,
+      "nim": 25739,
+      "▁obl": 25740,
+      "东部": 25741,
+      "十九日": 25742,
+      "URCE": 25743,
+      "姆斯": 25744,
+      "▁genera": 25745,
+      "nMin": 25746,
+      "嘉宾": 25747,
+      "orters": 25748,
+      "picture": 25749,
+      "tString": 25750,
+      "总会": 25751,
+      "正确的是": 25752,
+      "▁bullet": 25753,
+      "chrom": 25754,
+      "▁meter": 25755,
+      "▁personalized": 25756,
+      "综艺": 25757,
+      "hill": 25758,
+      "isan": 25759,
+      "su": 25760,
+      "潜在": 25761,
+      "▁`${": 25762,
+      "护肤": 25763,
+      "▁beliefs": 25764,
+      "▁dementia": 25765,
+      "advant": 25766,
+      "▁solely": 25767,
+      "▁surely": 25768,
+      "她说": 25769,
+      "Flags": 25770,
+      "公顷": 25771,
+      "▁stadium": 25772,
+      "’,": 25773,
+      "▁mart": 25774,
+      "repeat": 25775,
+      "开奖": 25776,
+      "▁CE": 25777,
+      "AUTH": 25778,
+      "InputStream": 25779,
+      "Jan": 25780,
+      "hene": 25781,
+      "选择的依据": 25782,
+      "精细": 25783,
+      "compute": 25784,
+      "▁practitioners": 25785,
+      "▁suck": 25786,
+      "▁lacking": 25787,
+      "我妈": 25788,
+      "nStud": 25789,
+      "raig": 25790,
+      "WAR": 25791,
+      "酱油": 25792,
+      "▁NP": 25793,
+      "aughters": 25794,
+      "▁printer": 25795,
+      "▁consolid": 25796,
+      "物料": 25797,
+      "▁wrapped": 25798,
+      "全民": 25799,
+      "▁Pun": 25800,
+      "▁chron": 25801,
+      "nwhile": 25802,
+      "约翰": 25803,
+      "▁locate": 25804,
+      "etically": 25805,
+      "的花": 25806,
+      "Bus": 25807,
+      "就是在": 25808,
+      "裁定的执行": 25809,
+      "rawer": 25810,
+      "▁Inn": 25811,
+      "勇气": 25812,
+      "gon": 25813,
+      "ayan": 25814,
+      "henyl": 25815,
+      "初级": 25816,
+      "▁qualify": 25817,
+      "yset": 25818,
+      "▁Defense": 25819,
+      "▁incl": 25820,
+      "ère": 25821,
+      "ITH": 25822,
+      "JsonProperty": 25823,
+      "Permission": 25824,
+      "five": 25825,
+      "▁Nord": 25826,
+      "▁Testing": 25827,
+      "攻略": 25828,
+      "▁Nich": 25829,
+      "exc": 25830,
+      "▁Bulg": 25831,
+      "▁Grade": 25832,
+      "中国农业银行股份有限公司": 25833,
+      "激活": 25834,
+      "▁lands": 25835,
+      "调味": 25836,
+      "事物": 25837,
+      "▁worship": 25838,
+      "▁Migration": 25839,
+      "创始人": 25840,
+      "▁ves": 25841,
+      "utures": 25842,
+      "▁transcriptional": 25843,
+      "▁VA": 25844,
+      "▁enorm": 25845,
+      "▁Mission": 25846,
+      "长得": 25847,
+      "eton": 25848,
+      "差的": 25849,
+      "▁batteries": 25850,
+      "libs": 25851,
+      "研究了": 25852,
+      "▁TNF": 25853,
+      "▁Changes": 25854,
+      "▁grateful": 25855,
+      "问题和选项": 25856,
+      "clone": 25857,
+      "她是": 25858,
+      "出国": 25859,
+      "公正": 25860,
+      "▁enroll": 25861,
+      "▁magazines": 25862,
+      "联络": 25863,
+      "男友": 25864,
+      "memset": 25865,
+      "▁Danish": 25866,
+      "▁tan": 25867,
+      "DOCTYPE": 25868,
+      "▁delivers": 25869,
+      "▁Imm": 25870,
+      "nClub": 25871,
+      "能认罪悔罪": 25872,
+      "之地": 25873,
+      "adic": 25874,
+      "全年": 25875,
+      "在其": 25876,
+      "公积金": 25877,
+      "▁judges": 25878,
+      "▁marijuana": 25879,
+      "▁colleges": 25880,
+      "▁Linked": 25881,
+      "▁containers": 25882,
+      "▁noteb": 25883,
+      "inae": 25884,
+      "▁McG": 25885,
+      "ivals": 25886,
+      "与其他": 25887,
+      "Descriptor": 25888,
+      "不怕": 25889,
+      "还清": 25890,
+      "▁retinal": 25891,
+      "情的": 25892,
+      "数字化": 25893,
+      "▁assessments": 25894,
+      "▁Mos": 25895,
+      "encia": 25896,
+      "谨慎": 25897,
+      "Autow": 25898,
+      "▁Month": 25899,
+      "▁lasting": 25900,
+      "▁ware": 25901,
+      "▁imped": 25902,
+      "▁orche": 25903,
+      "就很": 25904,
+      "dose": 25905,
+      "低温": 25906,
+      "南县": 25907,
+      "▁spoken": 25908,
+      "政务": 25909,
+      "Lower": 25910,
+      "▁authentic": 25911,
+      "etooth": 25912,
+      "▁Planning": 25913,
+      "▁Kath": 25914,
+      "▁annoy": 25915,
+      "ruct": 25916,
+      "▁Configuration": 25917,
+      "uplicate": 25918,
+      "▁contributors": 25919,
+      "材料的": 25920,
+      "的银行存款": 25921,
+      "general": 25922,
+      "五日内": 25923,
+      "积极分子": 25924,
+      "▁Cole": 25925,
+      "nMedal": 25926,
+      "▁Networks": 25927,
+      "▁execut": 25928,
+      "违约金": 25929,
+      "详细的": 25930,
+      "▁seasonal": 25931,
+      "versely": 25932,
+      "就好了": 25933,
+      "▁wings": 25934,
+      "▁Roll": 25935,
+      "▁cytokines": 25936,
+      "roles": 25937,
+      "Pod": 25938,
+      "iquid": 25939,
+      "▁Twenty": 25940,
+      "▁Label": 25941,
+      "vd": 25942,
+      "cop": 25943,
+      "▁straightforward": 25944,
+      "户口": 25945,
+      "剧本": 25946,
+      "完美的": 25947,
+      "▁Gon": 25948,
+      "信念": 25949,
+      "icht": 25950,
+      "镜像": 25951,
+      "▁Besides": 25952,
+      "не": 25953,
+      "▁performs": 25954,
+      "就不会": 25955,
+      "textbf": 25956,
+      "▁Champions": 25957,
+      "酌情": 25958,
+      "▁AF": 25959,
+      "▁tort": 25960,
+      "versions": 25961,
+      "▁plugins": 25962,
+      "ferred": 25963,
+      "▁genre": 25964,
+      "身心": 25965,
+      "▁appearing": 25966,
+      "RY": 25967,
+      "▁toys": 25968,
+      "▁([": 25969,
+      "oplasmic": 25970,
+      "$_": 25971,
+      "▁Cin": 25972,
+      "路人": 25973,
+      "电网": 25974,
+      "▁typeof": 25975,
+      "▁dad": 25976,
+      "▁substantially": 25977,
+      "交互": 25978,
+      "▁Pages": 25979,
+      "江县": 25980,
+      "▁destruction": 25981,
+      "ssl": 25982,
+      "白酒": 25983,
+      "jax": 25984,
+      "repos": 25985,
+      "▁工作": 25986,
+      "欲望": 25987,
+      "影响力": 25988,
+      "中华民族": 25989,
+      "情报": 25990,
+      "为其": 25991,
+      "被执行人王": 25992,
+      "servers": 25993,
+      "▁Posts": 25994,
+      "的一款": 25995,
+      "世界杯": 25996,
+      "▁Functions": 25997,
+      "之上": 25998,
+      "▁threads": 25999,
+      "▁enhancing": 26000,
+      "City": 26001,
+      "在你": 26002,
+      "罢了": 26003,
+      "登记表": 26004,
+      "▁theater": 26005,
+      "▁pointing": 26006,
+      "手的": 26007,
+      "安区": 26008,
+      "第二百一": 26009,
+      "Shell": 26010,
+      "一同": 26011,
+      "可以说是": 26012,
+      "arin": 26013,
+      "上市公司": 26014,
+      "nWorld": 26015,
+      "斯拉": 26016,
+      "安全性": 26017,
+      "▁implicit": 26018,
+      "▁jail": 26019,
+      "地处": 26020,
+      "leen": 26021,
+      "一眼": 26022,
+      "加密": 26023,
+      "▁lith": 26024,
+      "Did": 26025,
+      "ALTH": 26026,
+      "平稳": 26027,
+      "▁Nam": 26028,
+      "▁encouraging": 26029,
+      "▁lightweight": 26030,
+      "Oct": 26031,
+      "等多种": 26032,
+      "ammad": 26033,
+      "▁organizational": 26034,
+      "nOfficial": 26035,
+      "nRep": 26036,
+      "▁alternate": 26037,
+      "加重": 26038,
+      "rors": 26039,
+      "nSources": 26040,
+      "▁Unc": 26041,
+      "Sn": 26042,
+      "抖音": 26043,
+      "▁Fab": 26044,
+      "gel": 26045,
+      "Asset": 26046,
+      "沙发": 26047,
+      "▁[],": 26048,
+      "▁Wolf": 26049,
+      "▁infusion": 26050,
+      "▁habits": 26051,
+      "向本院提出诉讼请求": 26052,
+      "▁Font": 26053,
+      "购房": 26054,
+      "▁dramatically": 26055,
+      "▁undergraduate": 26056,
+      "历史上": 26057,
+      "assador": 26058,
+      "开展了": 26059,
+      "iquit": 26060,
+      "极大": 26061,
+      "学期": 26062,
+      "业内": 26063,
+      "concat": 26064,
+      "▁prest": 26065,
+      "画家": 26066,
+      "任职": 26067,
+      "Iterator": 26068,
+      "行程": 26069,
+      "Pair": 26070,
+      "Username": 26071,
+      "即可得出": 26072,
+      "▁quit": 26073,
+      "法学": 26074,
+      "▁champions": 26075,
+      "ousing": 26076,
+      "后者": 26077,
+      "转向": 26078,
+      "▁Gre": 26079,
+      "implies": 26080,
+      "的活动": 26081,
+      ";}\\": 26082,
+      "Customer": 26083,
+      "stringify": 26084,
+      "▁investigator": 26085,
+      "▁migrations": 26086,
+      "▁moder": 26087,
+      "▁poster": 26088,
+      "▁算法": 26089,
+      "▁BL": 26090,
+      "▁sac": 26091,
+      "▁Hero": 26092,
+      "▁hepatitis": 26093,
+      "ducer": 26094,
+      "Bal": 26095,
+      "微软": 26096,
+      "Static": 26097,
+      "▁aortic": 26098,
+      "▁induces": 26099,
+      "▁Original": 26100,
+      "▁intraven": 26101,
+      "▁rabbit": 26102,
+      "▁Sens": 26103,
+      "基准": 26104,
+      "▁reper": 26105,
+      "▁downstream": 26106,
+      "内角": 26107,
+      "的利息": 26108,
+      "▁friction": 26109,
+      "的日子": 26110,
+      "洛阳": 26111,
+      "水稻": 26112,
+      "为中心": 26113,
+      "西湖": 26114,
+      "在职": 26115,
+      "神话": 26116,
+      "意思是": 26117,
+      "▁Delta": 26118,
+      "ococc": 26119,
+      "Switch": 26120,
+      "nFinally": 26121,
+      "ogly": 26122,
+      "')),\\": 26123,
+      "▁Technologies": 26124,
+      "大桥": 26125,
+      "▁dc": 26126,
+      "▁baking": 26127,
+      "芝麻": 26128,
+      "ninterface": 26129,
+      "▁intelligent": 26130,
+      "KS": 26131,
+      "▁corresponds": 26132,
+      "both": 26133,
+      "▁immigration": 26134,
+      "upiter": 26135,
+      "YY": 26136,
+      "sender": 26137,
+      "▁Dance": 26138,
+      "海关": 26139,
+      "borough": 26140,
+      "长城": 26141,
+      "▁(;": 26142,
+      "――": 26143,
+      "▁Gordon": 26144,
+      "▁curl": 26145,
+      "Autowired": 26146,
+      "▁McK": 26147,
+      "Capt": 26148,
+      "名义": 26149,
+      "visit": 26150,
+      "遗址": 26151,
+      "太阳能": 26152,
+      "一般的": 26153,
+      "庆祝": 26154,
+      "▁topological": 26155,
+      "▁fatigue": 26156,
+      "器械": 26157,
+      "ishment": 26158,
+      "▁broader": 26159,
+      "nDespite": 26160,
+      "▁Connection": 26161,
+      "payment": 26162,
+      "▁painter": 26163,
+      "一颗": 26164,
+      "再用": 26165,
+      "▁attractions": 26166,
+      "参演": 26167,
+      "列入": 26168,
+      "▁pie": 26169,
+      "entries": 26170,
+      "Timer": 26171,
+      "▁Nothing": 26172,
+      "nIN": 26173,
+      "▁chairs": 26174,
+      "▁unlock": 26175,
+      "▁sail": 26176,
+      "▁charging": 26177,
+      "pendicular": 26178,
+      "▁COPYRIGHT": 26179,
+      "Tags": 26180,
+      "著名的": 26181,
+      "▁inhabit": 26182,
+      "nen": 26183,
+      "▁Mess": 26184,
+      "ascul": 26185,
+      "nIndian": 26186,
+      "▁Hum": 26187,
+      "要把": 26188,
+      "▁cited": 26189,
+      "万平方米": 26190,
+      "celona": 26191,
+      "▁euro": 26192,
+      "▁piv": 26193,
+      "补助": 26194,
+      "▁locom": 26195,
+      "tprotected": 26196,
+      "切除": 26197,
+      "阵容": 26198,
+      "毕业院校": 26199,
+      "▁Grid": 26200,
+      "向本院申请恢复执行": 26201,
+      "阳台": 26202,
+      "▁walks": 26203,
+      "不是很": 26204,
+      "▁subtle": 26205,
+      "nAtt": 26206,
+      "▁epile": 26207,
+      "▁subjected": 26208,
+      "Subject": 26209,
+      "atherine": 26210,
+      "自首": 26211,
+      "▁sciences": 26212,
+      "▁ال": 26213,
+      "▁第三": 26214,
+      "折抵刑期": 26215,
+      "▁duties": 26216,
+      "▁Conclusion": 26217,
+      "旗舰": 26218,
+      "错过": 26219,
+      "顶级": 26220,
+      "iates": 26221,
+      "▁Urban": 26222,
+      "する": 26223,
+      "场合": 26224,
+      "▁newer": 26225,
+      "没收": 26226,
+      "选购": 26227,
+      "poser": 26228,
+      "▁Finn": 26229,
+      "▁≥": 26230,
+      "▁$_": 26231,
+      "▁fav": 26232,
+      "审定": 26233,
+      "矫正": 26234,
+      "`](": 26235,
+      "tele": 26236,
+      "▁hitting": 26237,
+      "这边": 26238,
+      "▁coloring": 26239,
+      "▁disag": 26240,
+      "Chart": 26241,
+      "▁multic": 26242,
+      "国家和": 26243,
+      "▁induct": 26244,
+      "圆满": 26245,
+      "建筑工程": 26246,
+      "▁Chel": 26247,
+      "的上": 26248,
+      "▁recommendation": 26249,
+      "▁scheduling": 26250,
+      "▁complaint": 26251,
+      "▁Bab": 26252,
+      "\\'\\'\\'": 26253,
+      "▁fibrosis": 26254,
+      "▁♪\\": 26255,
+      "PRESS": 26256,
+      "▁troops": 26257,
+      "入学": 26258,
+      "初心": 26259,
+      "unicode": 26260,
+      "失望": 26261,
+      "▁earnings": 26262,
+      "emen": 26263,
+      "▁Visit": 26264,
+      "▁centuries": 26265,
+      "▁exhibits": 26266,
+      "修正": 26267,
+      "▁undertaken": 26268,
+      "小儿": 26269,
+      "问题是": 26270,
+      "invas": 26271,
+      "香味": 26272,
+      "(\\'\\\\": 26273,
+      "触摸": 26274,
+      "guard": 26275,
+      "女友": 26276,
+      "▁integrating": 26277,
+      "变动": 26278,
+      "fits": 26279,
+      "▁uncover": 26280,
+      "米的": 26281,
+      "游戏中": 26282,
+      "▁constituency": 26283,
+      "▁mast": 26284,
+      "▁Hug": 26285,
+      "▁tradem": 26286,
+      "Illuminate": 26287,
+      "Seg": 26288,
+      "▁extensively": 26289,
+      "▁arthritis": 26290,
+      "▁Manag": 26291,
+      "多名": 26292,
+      "leading": 26293,
+      "四季": 26294,
+      "Help": 26295,
+      "▁circles": 26296,
+      "months": 26297,
+      "▁Od": 26298,
+      "▁SN": 26299,
+      "▁manuscript": 26300,
+      "应力": 26301,
+      "nFilm": 26302,
+      "撤诉处理": 26303,
+      "暂停": 26304,
+      "负荷": 26305,
+      "送到": 26306,
+      "}^\\\\": 26307,
+      "必备": 26308,
+      "夫妻关系": 26309,
+      "▁二": 26310,
+      "▁Sie": 26311,
+      "来源于": 26312,
+      "驾驶机动车": 26313,
+      "Phys": 26314,
+      "oclonal": 26315,
+      "南路": 26316,
+      "食堂": 26317,
+      "endment": 26318,
+      "乐观": 26319,
+      "极为": 26320,
+      "▁tailored": 26321,
+      "同等": 26322,
+      "▁Tro": 26323,
+      "▁allocation": 26324,
+      "yster": 26325,
+      "乙醇": 26326,
+      "▁agriculture": 26327,
+      "▁conditional": 26328,
+      "Review": 26329,
+      "▁suited": 26330,
+      "nTop": 26331,
+      "urring": 26332,
+      "▁inaug": 26333,
+      "可谓": 26334,
+      "codes": 26335,
+      "ugged": 26336,
+      "upid": 26337,
+      "先行羁押": 26338,
+      "分管": 26339,
+      "boost": 26340,
+      "icits": 26341,
+      "unches": 26342,
+      "▁华": 26343,
+      "清新": 26344,
+      "▁Lore": 26345,
+      "▁speeds": 26346,
+      "▁Phone": 26347,
+      "▁thoroughly": 26348,
+      "uffix": 26349,
+      "安全的": 26350,
+      "▁broker": 26351,
+      "人是": 26352,
+      "▁depict": 26353,
+      "natal": 26354,
+      "的数量": 26355,
+      "▁Convert": 26356,
+      "运转": 26357,
+      "我市": 26358,
+      ";\"": 26359,
+      "所确定的义务": 26360,
+      "scriptions": 26361,
+      "复议期间不停止裁定的执行": 26362,
+      "Kind": 26363,
+      "指派检察员": 26364,
+      "travis": 26365,
+      "▁rectangle": 26366,
+      "▁liability": 26367,
+      "WM": 26368,
+      "ACKGROUND": 26369,
+      "▁IMPLIED": 26370,
+      "灯光": 26371,
+      "▁Solar": 26372,
+      "发生在": 26373,
+      "▁wildlife": 26374,
+      "Cart": 26375,
+      "operations": 26376,
+      "oglobin": 26377,
+      "卧室": 26378,
+      "▁electronics": 26379,
+      "情侣": 26380,
+      "robot": 26381,
+      "▁arrow": 26382,
+      "machine": 26383,
+      "艰难": 26384,
+      "▁bladder": 26385,
+      "阐述了": 26386,
+      "▁crossing": 26387,
+      "▁Dal": 26388,
+      "▁Syria": 26389,
+      "avatar": 26390,
+      "ducible": 26391,
+      "▁gaps": 26392,
+      "▁Cabin": 26393,
+      "▁Captain": 26394,
+      "▁locked": 26395,
+      "博览": 26396,
+      "三是": 26397,
+      "奖金": 26398,
+      "▁masses": 26399,
+      "▁concurrent": 26400,
+      "价值的": 26401,
+      "日语": 26402,
+      "首歌": 26403,
+      "▁zones": 26404,
+      "▁neurological": 26405,
+      "imon": 26406,
+      "▁animated": 26407,
+      "▁Api": 26408,
+      "▁unders": 26409,
+      "▁od": 26410,
+      "▁rocks": 26411,
+      "▁babies": 26412,
+      "{\\\\,\\\\": 26413,
+      "近似": 26414,
+      "outine": 26415,
+      "▁advise": 26416,
+      "判决执行之日起计算": 26417,
+      "▁calculating": 26418,
+      "▁useState": 26419,
+      "enta": 26420,
+      "priate": 26421,
+      "patriate": 26422,
+      "省委": 26423,
+      "geometry": 26424,
+      "三级": 26425,
+      "▁digest": 26426,
+      "erial": 26427,
+      "Statement": 26428,
+      "联邦": 26429,
+      "风云": 26430,
+      ">$": 26431,
+      "一分": 26432,
+      "antry": 26433,
+      "▁burning": 26434,
+      "平板": 26435,
+      "ativity": 26436,
+      "西路": 26437,
+      "▁Alaska": 26438,
+      "出自": 26439,
+      "aho": 26440,
+      "orted": 26441,
+      "▁worn": 26442,
+      "INST": 26443,
+      "▁ruling": 26444,
+      "Raw": 26445,
+      "yi": 26446,
+      "永久": 26447,
+      "▁spell": 26448,
+      "▁warnings": 26449,
+      "TRACT": 26450,
+      "今年的": 26451,
+      "反弹": 26452,
+      "情人": 26453,
+      "参加诉讼": 26454,
+      "Donald": 26455,
+      "潍坊": 26456,
+      "住重庆市": 26457,
+      "▁complaints": 26458,
+      "▁reconstruct": 26459,
+      "这不是": 26460,
+      "rying": 26461,
+      "▁Fine": 26462,
+      "大学的": 26463,
+      "▁modular": 26464,
+      "frak": 26465,
+      "dB": 26466,
+      "▁Activity": 26467,
+      "刚开始": 26468,
+      "生前": 26469,
+      "imated": 26470,
+      "▁Chamber": 26471,
+      "iko": 26472,
+      "ificates": 26473,
+      "一支": 26474,
+      "kov": 26475,
+      "orac": 26476,
+      "▁genuine": 26477,
+      "落后": 26478,
+      "ching": 26479,
+      "抢劫": 26480,
+      "▁参考文献": 26481,
+      "不妨": 26482,
+      "▁une": 26483,
+      "early": 26484,
+      "▁Anna": 26485,
+      "Identity": 26486,
+      "MRI": 26487,
+      "Counter": 26488,
+      "轻易": 26489,
+      "▁Late": 26490,
+      "▁Neg": 26491,
+      "▁Dean": 26492,
+      "单项": 26493,
+      "nSing": 26494,
+      "▁Animal": 26495,
+      "▁evil": 26496,
+      "▁grades": 26497,
+      "quick": 26498,
+      "▁Gulf": 26499,
+      "在法定": 26500,
+      "▁Upper": 26501,
+      "nApp": 26502,
+      "incre": 26503,
+      "▁weakness": 26504,
+      "进来": 26505,
+      "找到了": 26506,
+      "▁cricketers": 26507,
+      "即自": 26508,
+      "▁Prior": 26509,
+      "▁cavity": 26510,
+      "retty": 26511,
+      "变为": 26512,
+      "▁cycling": 26513,
+      "RR": 26514,
+      "▁apple": 26515,
+      "素养": 26516,
+      "硬化": 26517,
+      "Edge": 26518,
+      "▁Beat": 26519,
+      "供应链": 26520,
+      "风吹": 26521,
+      "为主题": 26522,
+      "▁carriers": 26523,
+      "▁deposition": 26524,
+      "野生": 26525,
+      "▁drops": 26526,
+      "▁DM": 26527,
+      "▁honey": 26528,
+      "▁expense": 26529,
+      "focus": 26530,
+      "eda": 26531,
+      "Pred": 26532,
+      "交通事故责任": 26533,
+      "▁Reading": 26534,
+      "▁为": 26535,
+      "webkit": 26536,
+      "▁forests": 26537,
+      "▁stopping": 26538,
+      ")`": 26539,
+      "gulp": 26540,
+      "我看": 26541,
+      "教堂": 26542,
+      "notations": 26543,
+      "首席": 26544,
+      "▁Mom": 26545,
+      "▁correlations": 26546,
+      "kn": 26547,
+      "复兴": 26548,
+      "gem": 26549,
+      "▁prevents": 26550,
+      "▁sponsored": 26551,
+      "lotte": 26552,
+      "▁Puerto": 26553,
+      "existing": 26554,
+      "nextInt": 26555,
+      "▁hypoth": 26556,
+      "相交": 26557,
+      "forall": 26558,
+      "性强": 26559,
+      "}×": 26560,
+      "urd": 26561,
+      "健康的": 26562,
+      "▁Ring": 26563,
+      "中路": 26564,
+      "▁passive": 26565,
+      "Pers": 26566,
+      "OO": 26567,
+      "▁Dictionary": 26568,
+      "()))\\": 26569,
+      "orect": 26570,
+      "在开庭审理过程中亦无异议": 26571,
+      "始建于": 26572,
+      "aq": 26573,
+      "cards": 26574,
+      "车主": 26575,
+      "rowing": 26576,
+      "LIST": 26577,
+      "▁extending": 26578,
+      "ndo": 26579,
+      "▁extracts": 26580,
+      "KA": 26581,
+      "▁responsive": 26582,
+      ":\\\\(": 26583,
+      "oka": 26584,
+      "产品质量": 26585,
+      "achel": 26586,
+      "▁explaining": 26587,
+      "INTER": 26588,
+      "▁covari": 26589,
+      "Vir": 26590,
+      "疾病的": 26591,
+      "scr": 26592,
+      "▁journalists": 26593,
+      "所以在": 26594,
+      "▁Delete": 26595,
+      "▁psychiatric": 26596,
+      "”(": 26597,
+      "gow": 26598,
+      "▁地址": 26599,
+      "astrop": 26600,
+      "Private": 26601,
+      "▁Exp": 26602,
+      "fixed": 26603,
+      "▁locks": 26604,
+      "▁fever": 26605,
+      "▁conject": 26606,
+      "建筑物": 26607,
+      "切割": 26608,
+      "rene": 26609,
+      "正数": 26610,
+      "边形的": 26611,
+      "的义务": 26612,
+      "▁Judge": 26613,
+      "copyright": 26614,
+      "placed": 26615,
+      "几点": 26616,
+      "报价": 26617,
+      "isure": 26618,
+      "▁Pad": 26619,
+      "树脂": 26620,
+      "▁androidx": 26621,
+      "▁DES": 26622,
+      "简体": 26623,
+      "胡萝卜": 26624,
+      "起来的": 26625,
+      "Look": 26626,
+      "▁Years": 26627,
+      "nABSTRACT": 26628,
+      "methods": 26629,
+      "▁donors": 26630,
+      "▁angi": 26631,
+      "门诊": 26632,
+      "iolet": 26633,
+      "nExpatriate": 26634,
+      "Coord": 26635,
+      "归来": 26636,
+      "▁Legislative": 26637,
+      "▁shr": 26638,
+      "ieval": 26639,
+      "▁shifts": 26640,
+      "▁Pradesh": 26641,
+      "▁Lock": 26642,
+      "做事": 26643,
+      "logy": 26644,
+      "▁prescription": 26645,
+      "dog": 26646,
+      "▁cd": 26647,
+      "***\\": 26648,
+      "▁Bou": 26649,
+      "Convert": 26650,
+      "更改": 26651,
+      "直至": 26652,
+      "▁Had": 26653,
+      "新一": 26654,
+      "▁Figure": 26655,
+      "床上": 26656,
+      "squ": 26657,
+      "itarian": 26658,
+      "当当": 26659,
+      "贷款利率": 26660,
+      "digit": 26661,
+      "▁tangent": 26662,
+      "▁Gets": 26663,
+      "势力": 26664,
+      "GFR": 26665,
+      "流畅": 26666,
+      "union": 26667,
+      "▁simultaneous": 26668,
+      "uplex": 26669,
+      "ikh": 26670,
+      "flutter": 26671,
+      "据说": 26672,
+      "listed": 26673,
+      "去掉": 26674,
+      "nBuildings": 26675,
+      "nMult": 26676,
+      "▁architectural": 26677,
+      "rag": 26678,
+      "蓝牙": 26679,
+      "▁bay": 26680,
+      "Exper": 26681,
+      "筛选": 26682,
+      "▁absent": 26683,
+      "▁Studios": 26684,
+      "▁coronavirus": 26685,
+      "说实话": 26686,
+      "▁municipal": 26687,
+      "综上所述": 26688,
+      "▁Gran": 26689,
+      "▁tent": 26690,
+      "▁Indeed": 26691,
+      "▁unnecessary": 26692,
+      "先行羁押的": 26693,
+      "两大": 26694,
+      "液晶": 26695,
+      "▁illustrated": 26696,
+      "▁Chart": 26697,
+      "▁Sony": 26698,
+      "本院将": 26699,
+      "人民群众": 26700,
+      "imals": 26701,
+      "nix": 26702,
+      "Audio": 26703,
+      "▁Window": 26704,
+      "微生物": 26705,
+      "variables": 26706,
+      "NH": 26707,
+      "irmingham": 26708,
+      "调试": 26709,
+      "GH": 26710,
+      "字典": 26711,
+      "那天": 26712,
+      "▁Fly": 26713,
+      "rig": 26714,
+      "▁Cant": 26715,
+      "▁thym": 26716,
+      "▁synaptic": 26717,
+      "三天": 26718,
+      "情怀": 26719,
+      "▁intro": 26720,
+      "Unknown": 26721,
+      "应在": 26722,
+      "温泉": 26723,
+      "支付利息": 26724,
+      "单位的": 26725,
+      "secure": 26726,
+      "上面的": 26727,
+      "号之一": 26728,
+      "我还是": 26729,
+      ")/((": 26730,
+      "▁guides": 26731,
+      "▁scholars": 26732,
+      "▁hi": 26733,
+      "错误的": 26734,
+      "▁Matthew": 26735,
+      "▁searched": 26736,
+      "nAustralian": 26737,
+      "▁Bangladesh": 26738,
+      "号案件": 26739,
+      "iam": 26740,
+      "iza": 26741,
+      "▁Excel": 26742,
+      "▁deleg": 26743,
+      "▁legisl": 26744,
+      "nall": 26745,
+      "▁苏": 26746,
+      "对了": 26747,
+      "▁compatibility": 26748,
+      "最具": 26749,
+      "查获": 26750,
+      "▁Device": 26751,
+      "▁Typ": 26752,
+      "假日": 26753,
+      "tend": 26754,
+      "caption": 26755,
+      "现原告": 26756,
+      "▁%>\\": 26757,
+      "▁Caribbean": 26758,
+      "▁Cel": 26759,
+      "制备": 26760,
+      "▁其他": 26761,
+      "十条": 26762,
+      "▁Cab": 26763,
+      "ierra": 26764,
+      "nSecond": 26765,
+      "processor": 26766,
+      "▁injected": 26767,
+      "▁Medic": 26768,
+      "▁financing": 26769,
+      "▁arithmetic": 26770,
+      "Pa": 26771,
+      "▁triggered": 26772,
+      "▁multim": 26773,
+      "羁押一日": 26774,
+      "▁Used": 26775,
+      "agine": 26776,
+      "uated": 26777,
+      "分开": 26778,
+      "▁Cre": 26779,
+      "▁programmes": 26780,
+      "热烈": 26781,
+      "▁offensive": 26782,
+      "贸易有限公司": 26783,
+      "して": 26784,
+      "产业发展": 26785,
+      "conduct": 26786,
+      "▁partly": 26787,
+      "当场": 26788,
+      "的存在": 26789,
+      "头部": 26790,
+      "股价": 26791,
+      "gre": 26792,
+      "最大值": 26793,
+      "等比数列": 26794,
+      "没办法": 26795,
+      "▁claiming": 26796,
+      "▁toy": 26797,
+      "的力量": 26798,
+      "资助": 26799,
+      "看到的": 26800,
+      "inely": 26801,
+      "▁careers": 26802,
+      "上级": 26803,
+      "▁retrieve": 26804,
+      "研究员": 26805,
+      "▁unre": 26806,
+      "实业有限公司": 26807,
+      "▁Beijing": 26808,
+      "中所": 26809,
+      "驳回上诉": 26810,
+      "parency": 26811,
+      "拆除": 26812,
+      "术语": 26813,
+      "danger": 26814,
+      "preview": 26815,
+      "ICAL": 26816,
+      "却被": 26817,
+      "nBorn": 26818,
+      "▁motivated": 26819,
+      "▁Edinburgh": 26820,
+      "分期": 26821,
+      "▁\"{{": 26822,
+      "新城": 26823,
+      "激动": 26824,
+      "▁viable": 26825,
+      "Mo": 26826,
+      "▁calm": 26827,
+      "最多的": 26828,
+      "▁folk": 26829,
+      "▁verte": 26830,
+      "典型的": 26831,
+      "arium": 26832,
+      "▁hind": 26833,
+      "thead": 26834,
+      "▁Drug": 26835,
+      "▁occupied": 26836,
+      "门户": 26837,
+      "第一百一": 26838,
+      "这张": 26839,
+      "▁MI": 26840,
+      "states": 26841,
+      ")));\\": 26842,
+      "▁pathological": 26843,
+      "Self": 26844,
+      "误差": 26845,
+      "▁cheaper": 26846,
+      "▁Valent": 26847,
+      "除此": 26848,
+      "quin": 26849,
+      "光学": 26850,
+      "▁rhe": 26851,
+      "再审申请人": 26852,
+      "▁Switch": 26853,
+      "减去有期徒刑一年": 26854,
+      "htm": 26855,
+      "▁pharmaceutical": 26856,
+      "要看": 26857,
+      "▁ACT": 26858,
+      "▁fantasy": 26859,
+      "的面积": 26860,
+      "正方形": 26861,
+      "aras": 26862,
+      "Article": 26863,
+      "▁moral": 26864,
+      "training": 26865,
+      "▁evolved": 26866,
+      "动手": 26867,
+      "无可": 26868,
+      "模范": 26869,
+      "折抵刑期一日": 26870,
+      "礼品": 26871,
+      "▁Left": 26872,
+      "rays": 26873,
+      "rena": 26874,
+      "经历了": 26875,
+      "lining": 26876,
+      "▁Fern": 26877,
+      "▁Mountains": 26878,
+      "▁quest": 26879,
+      "calc": 26880,
+      "▁西": 26881,
+      "红旗": 26882,
+      "▁subsets": 26883,
+      "▁Hampshire": 26884,
+      "▁cement": 26885,
+      "▁makers": 26886,
+      "▁vein": 26887,
+      "靠近": 26888,
+      "▁curv": 26889,
+      "▁conserved": 26890,
+      "精密": 26891,
+      "office": 26892,
+      "nSource": 26893,
+      "▁Much": 26894,
+      "leans": 26895,
+      "▁configurations": 26896,
+      "bach": 26897,
+      "frastructure": 26898,
+      "于本判决生效之日起": 26899,
+      "▁analytics": 26900,
+      "tT": 26901,
+      "各界": 26902,
+      "TON": 26903,
+      "notification": 26904,
+      "andal": 26905,
+      "渴望": 26906,
+      "你有": 26907,
+      "▁actresses": 26908,
+      "物种": 26909,
+      "行政区": 26910,
+      "▁tiles": 26911,
+      "红包": 26912,
+      "West": 26913,
+      "▁Excell": 26914,
+      "▁Pow": 26915,
+      "▁tourist": 26916,
+      "omi": 26917,
+      "▁VM": 26918,
+      "▁Dest": 26919,
+      "▁outfit": 26920,
+      "▁stylish": 26921,
+      "▁Springs": 26922,
+      "ANS": 26923,
+      "客户的": 26924,
+      "WF": 26925,
+      "▁pizza": 26926,
+      "▁confidential": 26927,
+      "愿望": 26928,
+      "cht": 26929,
+      "显存": 26930,
+      "refresh": 26931,
+      "wifi": 26932,
+      "ften": 26933,
+      "▁figured": 26934,
+      "{{\\\\": 26935,
+      "▁neighbors": 26936,
+      "]\"": 26937,
+      "WIN": 26938,
+      "accur": 26939,
+      "▁Berg": 26940,
+      "管理工作": 26941,
+      "avi": 26942,
+      "scar": 26943,
+      "▁MSC": 26944,
+      "▁advised": 26945,
+      "▁donations": 26946,
+      "▁(`": 26947,
+      "大规模": 26948,
+      "也是一": 26949,
+      "quit": 26950,
+      "~/": 26951,
+      "▁sheep": 26952,
+      "▁Kings": 26953,
+      "下次": 26954,
+      "]/": 26955,
+      "大人": 26956,
+      "抵抗": 26957,
+      "mands": 26958,
+      "oil": 26959,
+      "rche": 26960,
+      "志愿服务": 26961,
+      "▁Greater": 26962,
+      "istically": 26963,
+      "结构的": 26964,
+      "终结本院": 26965,
+      "▁EVENT": 26966,
+      "Headers": 26967,
+      "版的": 26968,
+      "nEnd": 26969,
+      "indent": 26970,
+      "avage": 26971,
+      "▁weighted": 26972,
+      "▁compliment": 26973,
+      "考点": 26974,
+      "fony": 26975,
+      "▁Bec": 26976,
+      "▁Brooklyn": 26977,
+      "你怎么": 26978,
+      "▁accessed": 26979,
+      "在卷": 26980,
+      "nabla": 26981,
+      "▁spine": 26982,
+      "▁subtract": 26983,
+      "开始了": 26984,
+      "Software": 26985,
+      "在互联网公布": 26986,
+      "PROJECT": 26987,
+      "▁worried": 26988,
+      "吸附": 26989,
+      "第二百零六条": 26990,
+      "kappa": 26991,
+      "unate": 26992,
+      "一座": 26993,
+      "天然气": 26994,
+      "doors": 26995,
+      "▁staining": 26996,
+      "还将": 26997,
+      "▁passage": 26998,
+      "ellers": 26999,
+      "隐私": 27000,
+      "▁Vit": 27001,
+      "民二初字第": 27002,
+      "claimer": 27003,
+      "我去": 27004,
+      "ALSE": 27005,
+      "已经成为": 27006,
+      "▁collisions": 27007,
+      "及相关": 27008,
+      "Station": 27009,
+      "ér": 27010,
+      "全方位": 27011,
+      "Limit": 27012,
+      "ulence": 27013,
+      "▁ports": 27014,
+      "▁demonstration": 27015,
+      "▁cursor": 27016,
+      "▁Condition": 27017,
+      "毕业后": 27018,
+      "经济损失": 27019,
+      "十足": 27020,
+      "gets": 27021,
+      "▁Mall": 27022,
+      "如果不": 27023,
+      "insp": 27024,
+      "▁php": 27025,
+      "制冷": 27026,
+      "口市": 27027,
+      "▁Days": 27028,
+      "善于": 27029,
+      "▁antibiotics": 27030,
+      "▁consulting": 27031,
+      "▁Archive": 27032,
+      "及利息": 27033,
+      "”)": 27034,
+      "就把": 27035,
+      "刑期从": 27036,
+      "▁fuzzy": 27037,
+      "▁GB": 27038,
+      "最新的": 27039,
+      "▁flip": 27040,
+      "主力": 27041,
+      "等腰": 27042,
+      "▁stimulus": 27043,
+      "一句话": 27044,
+      "▁comprised": 27045,
+      "infected": 27046,
+      "有的人": 27047,
+      "▁marginal": 27048,
+      "看得": 27049,
+      "▁商品": 27050,
+      "▁centres": 27051,
+      "nCONCLUSION": 27052,
+      "▁critically": 27053,
+      "▁similarly": 27054,
+      "▁Enterprise": 27055,
+      "分居": 27056,
+      "参照": 27057,
+      "▁chips": 27058,
+      "会导致": 27059,
+      "▁Drop": 27060,
+      "妊娠": 27061,
+      "maker": 27062,
+      "breaking": 27063,
+      "刚好": 27064,
+      "▁escap": 27065,
+      "▁patent": 27066,
+      "化合物": 27067,
+      "▁argued": 27068,
+      "▁centered": 27069,
+      "startswith": 27070,
+      "说什么": 27071,
+      "ducational": 27072,
+      "射击": 27073,
+      "nFC": 27074,
+      "iterator": 27075,
+      "▁ink": 27076,
+      "assium": 27077,
+      "废弃": 27078,
+      "▁entropy": 27079,
+      "▁Trad": 27080,
+      "严谨": 27081,
+      "的比": 27082,
+      "联合会": 27083,
+      "▁considers": 27084,
+      "▁intrinsic": 27085,
+      "kar": 27086,
+      "▁satisfying": 27087,
+      "fff": 27088,
+      "▁complementary": 27089,
+      "Quant": 27090,
+      "HB": 27091,
+      "好奇": 27092,
+      "反射": 27093,
+      "骨干": 27094,
+      "between": 27095,
+      "所有人": 27096,
+      "▁Kaz": 27097,
+      "可以使用": 27098,
+      "▁Everything": 27099,
+      "三维": 27100,
+      "▁做法": 27101,
+      "те": 27102,
+      "▁Madrid": 27103,
+      "受益": 27104,
+      "▁hunting": 27105,
+      "不宜在互联网公布": 27106,
+      "▁ethical": 27107,
+      "政法": 27108,
+      "nSp": 27109,
+      "▁Esc": 27110,
+      "▁Slo": 27111,
+      "之一的": 27112,
+      "▁Gary": 27113,
+      "▁analytic": 27114,
+      "gency": 27115,
+      "▁cust": 27116,
+      "▁resp": 27117,
+      "在服刑期间确有悔改表现": 27118,
+      "**:": 27119,
+      "大鼠": 27120,
+      "nCopyright": 27121,
+      "fluid": 27122,
+      "▁Effect": 27123,
+      "entities": 27124,
+      "Gl": 27125,
+      "▁bg": 27126,
+      "▁NT": 27127,
+      "nOriginally": 27128,
+      "时时": 27129,
+      "▁Cycl": 27130,
+      "立场": 27131,
+      "▁für": 27132,
+      "淋浴": 27133,
+      "sensitive": 27134,
+      "▁discussing": 27135,
+      "acs": 27136,
+      "MAKE": 27137,
+      "▁Deg": 27138,
+      "▁Occ": 27139,
+      "Circ": 27140,
+      "番茄": 27141,
+      "签名": 27142,
+      "▁Buck": 27143,
+      "▁Wait": 27144,
+      "nRailway": 27145,
+      "Cu": 27146,
+      "是用": 27147,
+      "Weight": 27148,
+      "▁thro": 27149,
+      "▁assisted": 27150,
+      "到大": 27151,
+      "口服": 27152,
+      "区块链": 27153,
+      "说说": 27154,
+      "▁ker": 27155,
+      "经传票传唤": 27156,
+      "OPT": 27157,
+      "产业链": 27158,
+      "▁failures": 27159,
+      "高手": 27160,
+      "▁Meeting": 27161,
+      "tha": 27162,
+      "▁毕业院校": 27163,
+      "有一定": 27164,
+      "Best": 27165,
+      "iour": 27166,
+      "▁grat": 27167,
+      "特种": 27168,
+      "mapping": 27169,
+      "▁governed": 27170,
+      "▁travelling": 27171,
+      "▁stakeholders": 27172,
+      "或查封": 27173,
+      "▁beef": 27174,
+      "▁cure": 27175,
+      "utenant": 27176,
+      "▁Corps": 27177,
+      "▁instantly": 27178,
+      "▁hippocamp": 27179,
+      "极端": 27180,
+      "▁rivers": 27181,
+      "arms": 27182,
+      "跑步": 27183,
+      "总监": 27184,
+      "GG": 27185,
+      "申请人的": 27186,
+      "▁interpol": 27187,
+      "第一百四十五条第一款的规定": 27188,
+      "heading": 27189,
+      "自学": 27190,
+      "脾气": 27191,
+      "▁BP": 27192,
+      "▁ties": 27193,
+      "质保": 27194,
+      "riel": 27195,
+      "rex": 27196,
+      "▁compass": 27197,
+      "▁Circle": 27198,
+      "▁gardens": 27199,
+      "▁ladies": 27200,
+      "斯科": 27201,
+      "nos": 27202,
+      "▁adren": 27203,
+      "自来": 27204,
+      "唯一的": 27205,
+      "developer": 27206,
+      "rame": 27207,
+      "▁Kay": 27208,
+      "▁proposals": 27209,
+      "普通人": 27210,
+      "根据您的": 27211,
+      "第七十九条": 27212,
+      "有力": 27213,
+      "▁Bruce": 27214,
+      "Components": 27215,
+      "防腐": 27216,
+      "▁isot": 27217,
+      "▁profound": 27218,
+      "染色": 27219,
+      "此类": 27220,
+      "职权": 27221,
+      "▁Kir": 27222,
+      "▁HERE": 27223,
+      "东莞市": 27224,
+      "▁founding": 27225,
+      "▁deadline": 27226,
+      "的各种": 27227,
+      "▁backed": 27228,
+      "▁stimulated": 27229,
+      "Setting": 27230,
+      "hp": 27231,
+      "调料": 27232,
+      "bums": 27233,
+      "裁定书送达之日起十日内": 27234,
+      "▁hormones": 27235,
+      "▁sustainability": 27236,
+      "userId": 27237,
+      "extr": 27238,
+      "▁Thompson": 27239,
+      "帮你": 27240,
+      "apon": 27241,
+      "▁reservoir": 27242,
+      "resolved": 27243,
+      "▁holder": 27244,
+      "▁Colon": 27245,
+      "dumps": 27246,
+      "很重要": 27247,
+      "这道题": 27248,
+      "lett": 27249,
+      "▁Decor": 27250,
+      "一身": 27251,
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁": 27252,
+      "诱导": 27253,
+      "▁nov": 27254,
+      "▁Perfect": 27255,
+      "iah": 27256,
+      "ISH": 27257,
+      "特殊的": 27258,
+      "▁twelve": 27259,
+      "㈰㈰": 27260,
+      "果然": 27261,
+      "amping": 27262,
+      "resses": 27263,
+      "▁purchases": 27264,
+      "mma": 27265,
+      "给定的": 27266,
+      "拯救": 27267,
+      "nWho": 27268,
+      "族自治县": 27269,
+      "correct": 27270,
+      "Bind": 27271,
+      "organization": 27272,
+      "CSS": 27273,
+      "nAuthor": 27274,
+      "号的": 27275,
+      "▁schizophren": 27276,
+      "演绎": 27277,
+      "▁confusing": 27278,
+      "locks": 27279,
+      "▁Friends": 27280,
+      "▁stochastic": 27281,
+      "起码": 27282,
+      "osomes": 27283,
+      "}&": 27284,
+      "commerce": 27285,
+      ":“": 27286,
+      "支队": 27287,
+      "nReturn": 27288,
+      "▁nutritional": 27289,
+      "春秋": 27290,
+      "▁Bug": 27291,
+      "携带宠物": 27292,
+      "手中": 27293,
+      "▁opts": 27294,
+      "hattan": 27295,
+      "器件": 27296,
+      "OW": 27297,
+      "orp": 27298,
+      "▁paired": 27299,
+      "站点": 27300,
+      "被执行人李": 27301,
+      "kers": 27302,
+      "ewise": 27303,
+      "▁Published": 27304,
+      "▁anticipated": 27305,
+      "客车": 27306,
+      "主持人": 27307,
+      "blocks": 27308,
+      "▁recruited": 27309,
+      "▁Ubuntu": 27310,
+      "website": 27311,
+      "sectional": 27312,
+      "intestinal": 27313,
+      "▁breathing": 27314,
+      "▁infil": 27315,
+      "▁investing": 27316,
+      "▁GPS": 27317,
+      "▁Assessment": 27318,
+      "▁unfortunately": 27319,
+      "都需要": 27320,
+      "阻止": 27321,
+      "|||": 27322,
+      "ucer": 27323,
+      "即可求出": 27324,
+      "▁rhythm": 27325,
+      "准则": 27326,
+      "示例": 27327,
+      "向人民法院": 27328,
+      "hibition": 27329,
+      "▁preserved": 27330,
+      "且有": 27331,
+      "恒成立": 27332,
+      "▁Creative": 27333,
+      "▁rotate": 27334,
+      "无穷": 27335,
+      "▁scholarship": 27336,
+      "身体健康": 27337,
+      "▁rendered": 27338,
+      "▁Feel": 27339,
+      "ashi": 27340,
+      "matic": 27341,
+      "▁versatile": 27342,
+      "证件": 27343,
+      "▁fundra": 27344,
+      "▁paradigm": 27345,
+      "▁Electron": 27346,
+      "▁Jess": 27347,
+      "REFIX": 27348,
+      "▁contractors": 27349,
+      "ocent": 27350,
+      "▁Screen": 27351,
+      "还得": 27352,
+      "谷歌": 27353,
+      "selection": 27354,
+      "▁Arc": 27355,
+      "markdown": 27356,
+      "可以选择": 27357,
+      "▁Bachelor": 27358,
+      "Rot": 27359,
+      "▁GM": 27360,
+      "olished": 27361,
+      "幽默": 27362,
+      "uitive": 27363,
+      "医学院": 27364,
+      "▁vendors": 27365,
+      "进行的": 27366,
+      "utors": 27367,
+      "研究的": 27368,
+      "▁ERR": 27369,
+      "▁AIDS": 27370,
+      "RET": 27371,
+      "beans": 27372,
+      "▁Installation": 27373,
+      "▁conversations": 27374,
+      "肯定是": 27375,
+      "ristian": 27376,
+      "▁Unlike": 27377,
+      "▁homework": 27378,
+      "tile": 27379,
+      "▁metastasis": 27380,
+      "nLife": 27381,
+      "Boot": 27382,
+      "▁phosphat": 27383,
+      "▁fur": 27384,
+      "▁descending": 27385,
+      "\"$": 27386,
+      "的存款": 27387,
+      "▁flooring": 27388,
+      "▁biomarkers": 27389,
+      "▁主演": 27390,
+      "的人物": 27391,
+      "用法": 27392,
+      "▁au": 27393,
+      "▁Hawaii": 27394,
+      "单身": 27395,
+      "王国": 27396,
+      "拆迁": 27397,
+      "▁dors": 27398,
+      "都说": 27399,
+      "崛起": 27400,
+      "▁crimes": 27401,
+      "▁gland": 27402,
+      "取代": 27403,
+      "▁audit": 27404,
+      "▁shelter": 27405,
+      "合约": 27406,
+      "▁tender": 27407,
+      "▁Liverpool": 27408,
+      "dings": 27409,
+      "数位": 27410,
+      "月经": 27411,
+      "浴室": 27412,
+      "免费的": 27413,
+      "一个个": 27414,
+      "cond": 27415,
+      "亲戚": 27416,
+      "的同学": 27417,
+      "▁zinc": 27418,
+      "当选": 27419,
+      "smart": 27420,
+      "肚子": 27421,
+      "不允许": 27422,
+      "器官": 27423,
+      "▁MY": 27424,
+      "▁phenomena": 27425,
+      "▁Expression": 27426,
+      "▁interpreted": 27427,
+      "立案执行": 27428,
+      "▁dialogue": 27429,
+      "▁ts": 27430,
+      "的重要性": 27431,
+      "FileName": 27432,
+      "世间": 27433,
+      "童话": 27434,
+      "续航": 27435,
+      "才知道": 27436,
+      "ntemplate": 27437,
+      "CUR": 27438,
+      "olesale": 27439,
+      "▁pg": 27440,
+      "转载": 27441,
+      "被执行人张": 27442,
+      "Shared": 27443,
+      "乘客": 27444,
+      "▁Falls": 27445,
+      "▁perspectives": 27446,
+      "▁Elementary": 27447,
+      "Abs": 27448,
+      "Stats": 27449,
+      "▁Ward": 27450,
+      "▁automation": 27451,
+      "数十": 27452,
+      "秉承": 27453,
+      "花了": 27454,
+      "赛车": 27455,
+      "▁excellence": 27456,
+      "orneys": 27457,
+      "▁Cass": 27458,
+      "上门": 27459,
+      "kill": 27460,
+      "ophage": 27461,
+      "高清": 27462,
+      "▁chapters": 27463,
+      "芜湖": 27464,
+      "一笑": 27465,
+      "immer": 27466,
+      "iemann": 27467,
+      "▁mad": 27468,
+      "所以我": 27469,
+      "protein": 27470,
+      "大家的": 27471,
+      "▁visualization": 27472,
+      "▁mp": 27473,
+      "▁MPs": 27474,
+      "usive": 27475,
+      "▁Monitor": 27476,
+      "▁Lem": 27477,
+      "▁attenu": 27478,
+      "nprivate": 27479,
+      "▁respondents": 27480,
+      "central": 27481,
+      "▁Glas": 27482,
+      "ecycle": 27483,
+      "▁Peters": 27484,
+      "ERY": 27485,
+      "zens": 27486,
+      "David": 27487,
+      "▁Cemetery": 27488,
+      "图文": 27489,
+      "▁IBM": 27490,
+      "▁draws": 27491,
+      "对中国": 27492,
+      "▁outline": 27493,
+      "Pres": 27494,
+      "▁PET": 27495,
+      "▁sorts": 27496,
+      "nPerson": 27497,
+      "主观": 27498,
+      "PB": 27499,
+      "abolic": 27500,
+      "羁押一日折抵刑期一日": 27501,
+      "为主要": 27502,
+      "rost": 27503,
+      "vic": 27504,
+      "▁exceptions": 27505,
+      ">_": 27506,
+      "▁airway": 27507,
+      "election": 27508,
+      "▁tends": 27509,
+      "▁TeX": 27510,
+      "修订": 27511,
+      "▁renown": 27512,
+      "目的探讨": 27513,
+      "zeros": 27514,
+      "▁Mode": 27515,
+      "▁Father": 27516,
+      "defaults": 27517,
+      "过后": 27518,
+      "▁detecting": 27519,
+      "inian": 27520,
+      "agrams": 27521,
+      "▁Croat": 27522,
+      "少了": 27523,
+      "ICS": 27524,
+      "▁amateur": 27525,
+      "区委": 27526,
+      "▁abandoned": 27527,
+      "星星": 27528,
+      "naments": 27529,
+      "年中国": 27530,
+      "umbai": 27531,
+      "▁neph": 27532,
+      "urally": 27533,
+      "▁剧": 27534,
+      "水库": 27535,
+      "ocese": 27536,
+      "▁作品目录": 27537,
+      "▁conflicts": 27538,
+      "预警": 27539,
+      "itects": 27540,
+      "Scroll": 27541,
+      "▁strat": 27542,
+      "▁colored": 27543,
+      "▁logarith": 27544,
+      "ococcus": 27545,
+      "权利人": 27546,
+      "▁网络小说": 27547,
+      "▁helic": 27548,
+      "Present": 27549,
+      "та": 27550,
+      "果实": 27551,
+      "淋巴": 27552,
+      "让孩子": 27553,
+      "▁tremend": 27554,
+      "率先": 27555,
+      "▁welfare": 27556,
+      "keley": 27557,
+      "Mobile": 27558,
+      "▁inspire": 27559,
+      "▁attracted": 27560,
+      "avan": 27561,
+      "ongue": 27562,
+      "▁Gram": 27563,
+      "道士": 27564,
+      "Selection": 27565,
+      "乌克兰": 27566,
+      "▁accidents": 27567,
+      "▁fleet": 27568,
+      "▁Ot": 27569,
+      "IFA": 27570,
+      "▁Kap": 27571,
+      "sto": 27572,
+      "hedules": 27573,
+      "▁kinetics": 27574,
+      "高新区": 27575,
+      "▁plots": 27576,
+      "时机": 27577,
+      "▁Repair": 27578,
+      "间隔": 27579,
+      "▁md": 27580,
+      "▁das": 27581,
+      "▁Tob": 27582,
+      "代价": 27583,
+      "▁inserted": 27584,
+      "素材": 27585,
+      "般的": 27586,
+      "▁Juan": 27587,
+      "nSongs": 27588,
+      "sizeof": 27589,
+      "▁vendor": 27590,
+      "华人": 27591,
+      "▁Abb": 27592,
+      "新开": 27593,
+      "▁Birth": 27594,
+      "▁grants": 27595,
+      "畜牧": 27596,
+      "▁imported": 27597,
+      "▁expecting": 27598,
+      "▁onClick": 27599,
+      "电子产品": 27600,
+      "varepsilon": 27601,
+      "原文": 27602,
+      "▁jewelry": 27603,
+      "▁soup": 27604,
+      "anners": 27605,
+      "▁manipulation": 27606,
+      "奥运会": 27607,
+      "的地": 27608,
+      "unittest": 27609,
+      "合肥市": 27610,
+      "atsby": 27611,
+      "▁Medit": 27612,
+      "zing": 27613,
+      "shore": 27614,
+      "综合运用": 27615,
+      "▁Better": 27616,
+      "▁Freder": 27617,
+      "点点": 27618,
+      "▁footage": 27619,
+      "grades": 27620,
+      "▁dfs": 27621,
+      "▁seventh": 27622,
+      "▁encryption": 27623,
+      "MAGES": 27624,
+      "qquad": 27625,
+      "nare": 27626,
+      "月的": 27627,
+      "等着": 27628,
+      "▁Sum": 27629,
+      "▁transitions": 27630,
+      "异性": 27631,
+      "ennial": 27632,
+      "▁Ry": 27633,
+      "应当在": 27634,
+      "ninclude": 27635,
+      "天涯": 27636,
+      "▁Dir": 27637,
+      "ablab": 27638,
+      "nMoths": 27639,
+      "客运": 27640,
+      "▁tick": 27641,
+      "本院不予支持": 27642,
+      "held": 27643,
+      "▁ruled": 27644,
+      "但他": 27645,
+      "▁lawsuit": 27646,
+      "▁antibiotic": 27647,
+      "毫升": 27648,
+      "▁barrel": 27649,
+      "Endpoint": 27650,
+      "副教授": 27651,
+      "案发": 27652,
+      "▁Prim": 27653,
+      "何时": 27654,
+      "严肃": 27655,
+      "的女人": 27656,
+      "▁Bloom": 27657,
+      "▁metastatic": 27658,
+      "但是在": 27659,
+      "▁Tower": 27660,
+      "涉及到": 27661,
+      "▁overnight": 27662,
+      "nHon": 27663,
+      "温州": 27664,
+      "援助": 27665,
+      "▁Cash": 27666,
+      "倡导": 27667,
+      "对他": 27668,
+      "酒吧": 27669,
+      "osal": 27670,
+      "▁Adams": 27671,
+      "oons": 27672,
+      "*,": 27673,
+      "大佬": 27674,
+      "我没": 27675,
+      "boards": 27676,
+      "▁comprom": 27677,
+      "▁demographic": 27678,
+      "被告的": 27679,
+      "While": 27680,
+      "()));\\": 27681,
+      "执行期间": 27682,
+      "▁Ay": 27683,
+      "▁spectacular": 27684,
+      "▁Abd": 27685,
+      "tel": 27686,
+      "▁Pool": 27687,
+      "▁Close": 27688,
+      "▁internationally": 27689,
+      "▁Metal": 27690,
+      "比赛中": 27691,
+      "▁considerably": 27692,
+      "今天的": 27693,
+      "▁excluded": 27694,
+      "妹子": 27695,
+      "Scanner": 27696,
+      "Animation": 27697,
+      "三人": 27698,
+      "暖气": 27699,
+      "▁lys": 27700,
+      "▁labour": 27701,
+      "关怀": 27702,
+      "主导": 27703,
+      "曾在": 27704,
+      "玩的": 27705,
+      "▁iteration": 27706,
+      "atable": 27707,
+      "怀特": 27708,
+      "▁shear": 27709,
+      "workflow": 27710,
+      "▁Shel": 27711,
+      "有害": 27712,
+      "IEW": 27713,
+      "为中国": 27714,
+      "体现了": 27715,
+      "▁Lot": 27716,
+      "▁repeatedly": 27717,
+      "肇事": 27718,
+      "▁KEY": 27719,
+      "▁fingers": 27720,
+      "大使": 27721,
+      "▁Nelson": 27722,
+      "的文化": 27723,
+      "mathfrak": 27724,
+      "▁bilateral": 27725,
+      "▁incidents": 27726,
+      "见证": 27727,
+      "▁sketch": 27728,
+      "承载": 27729,
+      "广阔": 27730,
+      "下落不明": 27731,
+      "▁introduces": 27732,
+      "nginx": 27733,
+      "足球运动员": 27734,
+      "▁seal": 27735,
+      "▁promises": 27736,
+      "▁vine": 27737,
+      "imension": 27738,
+      "▁sportspeople": 27739,
+      "▁devast": 27740,
+      "锅炉": 27741,
+      "▁Faculty": 27742,
+      "趣味": 27743,
+      "▁prevented": 27744,
+      "▁HEALTH": 27745,
+      "您的选择": 27746,
+      "▁Constitution": 27747,
+      "▁myth": 27748,
+      "▁cargo": 27749,
+      "ABA": 27750,
+      "党组书记": 27751,
+      "备注": 27752,
+      "brief": 27753,
+      "nexcept": 27754,
+      "typename": 27755,
+      "值的": 27756,
+      "判断对错": 27757,
+      "▁rival": 27758,
+      "▁brow": 27759,
+      "反思": 27760,
+      "layers": 27761,
+      "▁intens": 27762,
+      "▁ber": 27763,
+      "真实的": 27764,
+      "riteria": 27765,
+      "动员": 27766,
+      "▁References": 27767,
+      "▁veteran": 27768,
+      "▁pets": 27769,
+      "大奖": 27770,
+      "nDid": 27771,
+      "▁Denver": 27772,
+      "umbing": 27773,
+      "制作的": 27774,
+      "▁Furn": 27775,
+      "Region": 27776,
+      "anghai": 27777,
+      "▁handy": 27778,
+      "▁extraordinary": 27779,
+      "▁”": 27780,
+      "▁nationwide": 27781,
+      "▁TB": 27782,
+      "/:": 27783,
+      "轨迹": 27784,
+      "小红": 27785,
+      "轮胎": 27786,
+      "▁criter": 27787,
+      "▁HA": 27788,
+      "ancers": 27789,
+      "▁transgenic": 27790,
+      "▁browse": 27791,
+      "杨某": 27792,
+      "▁melt": 27793,
+      "enberg": 27794,
+      "mesh": 27795,
+      "identity": 27796,
+      "表现为": 27797,
+      "▁acres": 27798,
+      "▁Douglas": 27799,
+      "收缩": 27800,
+      "ULE": 27801,
+      "Shop": 27802,
+      "▁Boys": 27803,
+      "乌鲁": 27804,
+      "合适的答案": 27805,
+      "阅读选项": 27806,
+      "queous": 27807,
+      "▁iconic": 27808,
+      "▁reasoning": 27809,
+      "▁commissioned": 27810,
+      "噪音": 27811,
+      "该案": 27812,
+      "▁deserve": 27813,
+      "WH": 27814,
+      "军人": 27815,
+      "在中": 27816,
+      "many": 27817,
+      "科长": 27818,
+      "这里的": 27819,
+      "▁MAT": 27820,
+      "热的": 27821,
+      "年至": 27822,
+      "lora": 27823,
+      "▁Welcome": 27824,
+      "▁propagation": 27825,
+      "pril": 27826,
+      "rients": 27827,
+      "▁authorized": 27828,
+      "伊朗": 27829,
+      "发展和": 27830,
+      "▁epic": 27831,
+      "▁tune": 27832,
+      "最高的": 27833,
+      "shields": 27834,
+      "▁singers": 27835,
+      "无聊": 27836,
+      "战术": 27837,
+      "超出": 27838,
+      "北路": 27839,
+      "ncan": 27840,
+      "集团公司": 27841,
+      "▁Scientific": 27842,
+      "ablabla": 27843,
+      "▁prayer": 27844,
+      "▁mystery": 27845,
+      "▁Andy": 27846,
+      "cookie": 27847,
+      "▁collapse": 27848,
+      "iOS": 27849,
+      "▁differentiate": 27850,
+      "Cert": 27851,
+      "pytest": 27852,
+      "导入": 27853,
+      "▁horror": 27854,
+      "hum": 27855,
+      "的一次": 27856,
+      ">/": 27857,
+      "Development": 27858,
+      "jen": 27859,
+      "nPlot": 27860,
+      "▁Dig": 27861,
+      "Colum": 27862,
+      "▁slides": 27863,
+      "要求的": 27864,
+      "▁torque": 27865,
+      "nMan": 27866,
+      "SERVER": 27867,
+      "jamin": 27868,
+      "姿态": 27869,
+      "▁allegedly": 27870,
+      "大楼": 27871,
+      "irates": 27872,
+      "orectal": 27873,
+      "▁murine": 27874,
+      "▁limb": 27875,
+      "本息": 27876,
+      "聚会": 27877,
+      "天赋": 27878,
+      "非常的": 27879,
+      "▁ischemia": 27880,
+      "丽的": 27881,
+      "CLA": 27882,
+      "排查": 27883,
+      "Sigma": 27884,
+      "全是": 27885,
+      "▁Heat": 27886,
+      "uisine": 27887,
+      "常识": 27888,
+      "▁IM": 27889,
+      "▁RSS": 27890,
+      "▁css": 27891,
+      "▁Higher": 27892,
+      "Demo": 27893,
+      "emor": 27894,
+      "▁yeah": 27895,
+      "口头": 27896,
+      "面部": 27897,
+      "▁encode": 27898,
+      "的核心": 27899,
+      "▁NBA": 27900,
+      "▁Marie": 27901,
+      "▁sits": 27902,
+      "▁Cleveland": 27903,
+      "一辈子": 27904,
+      "undry": 27905,
+      "流传": 27906,
+      "▁aria": 27907,
+      "▁Fer": 27908,
+      "得分": 27909,
+      "言论": 27910,
+      "▁anybody": 27911,
+      "▁secured": 27912,
+      "otimes": 27913,
+      "osome": 27914,
+      "其在": 27915,
+      "偶然": 27916,
+      "▁企业": 27917,
+      "强势": 27918,
+      "▁vinyl": 27919,
+      "未发现": 27920,
+      "degree": 27921,
+      "▁Montreal": 27922,
+      "螺旋": 27923,
+      "Mill": 27924,
+      "ivan": 27925,
+      "▁Sure": 27926,
+      "日立案受理后": 27927,
+      "▁toilet": 27928,
+      "▁ray": 27929,
+      "奖项": 27930,
+      "Series": 27931,
+      "▁Tamil": 27932,
+      "受法律保护": 27933,
+      "▁Commonwealth": 27934,
+      "▁Hat": 27935,
+      "[]\\": 27936,
+      "▁Graham": 27937,
+      "▁Nevada": 27938,
+      "▁checkout": 27939,
+      "▁运动项目": 27940,
+      "日记": 27941,
+      "燃料": 27942,
+      "表面积": 27943,
+      "即可得到": 27944,
+      "▁Pref": 27945,
+      "▁uploaded": 27946,
+      "婚生": 27947,
+      "观赏": 27948,
+      "eva": 27949,
+      "▁happiness": 27950,
+      "偶数": 27951,
+      "用电": 27952,
+      "▁perpendicular": 27953,
+      "▁Christopher": 27954,
+      "RP": 27955,
+      "▁地区": 27956,
+      "读音": 27957,
+      "mentation": 27958,
+      "▁Lar": 27959,
+      "占比": 27960,
+      "▁calculus": 27961,
+      "柔软": 27962,
+      "阴阳": 27963,
+      "人才培养": 27964,
+      "▁Census": 27965,
+      "▁Factory": 27966,
+      "▁simplified": 27967,
+      "LOCK": 27968,
+      "手里": 27969,
+      "马来": 27970,
+      "他们在": 27971,
+      "Super": 27972,
+      "不满": 27973,
+      "都不是": 27974,
+      "▁patio": 27975,
+      "报错": 27976,
+      "▁Manual": 27977,
+      "人民代表大会": 27978,
+      "▁cryptoc": 27979,
+      "▁inflation": 27980,
+      "无敌": 27981,
+      "此后": 27982,
+      "hist": 27983,
+      "▁buck": 27984,
+      "currency": 27985,
+      "▁sleeping": 27986,
+      "uations": 27987,
+      "之为": 27988,
+      "是我国": 27989,
+      "nBibliography": 27990,
+      "▁aggregation": 27991,
+      "中最": 27992,
+      "annes": 27993,
+      "oauth": 27994,
+      "ATIONS": 27995,
+      "空白": 27996,
+      "诉请": 27997,
+      "chlor": 27998,
+      "▁Equations": 27999,
+      "relim": 28000,
+      "locfile": 28001,
+      "▁{'": 28002,
+      "▁contamin": 28003,
+      "▁Period": 28004,
+      "▁attendance": 28005,
+      "▁distinctive": 28006,
+      "好吃": 28007,
+      "▁maker": 28008,
+      "validator": 28009,
+      "▁comparisons": 28010,
+      "伟大的": 28011,
+      "昨日": 28012,
+      "委员会委员": 28013,
+      "▁valley": 28014,
+      "▁alike": 28015,
+      "完成后": 28016,
+      "reply": 28017,
+      "▁torn": 28018,
+      "▁Looking": 28019,
+      "越多": 28020,
+      "本院予以准许": 28021,
+      "柠檬": 28022,
+      "ahl": 28023,
+      "史上": 28024,
+      "Binary": 28025,
+      "▁shade": 28026,
+      "▁Experimental": 28027,
+      "▁上映时间": 28028,
+      "羡慕": 28029,
+      "坐标为": 28030,
+      "大致": 28031,
+      "tility": 28032,
+      "▁submitting": 28033,
+      "poral": 28034,
+      "▁relay": 28035,
+      "▁wishes": 28036,
+      "▁earthqu": 28037,
+      "就是要": 28038,
+      "▁incomplete": 28039,
+      "▁develops": 28040,
+      "▁Calc": 28041,
+      "打了": 28042,
+      "▁fiscal": 28043,
+      "sources": 28044,
+      "▁Hydro": 28045,
+      "外汇": 28046,
+      "▁sinus": 28047,
+      "运气": 28048,
+      "▁LP": 28049,
+      "搭载": 28050,
+      "ylon": 28051,
+      "surface": 28052,
+      "疗法": 28053,
+      "fonts": 28054,
+      "的思想": 28055,
+      "▁Universal": 28056,
+      "▁否": 28057,
+      "▁indent": 28058,
+      "也被": 28059,
+      "thlete": 28060,
+      "▁sne": 28061,
+      "年以来": 28062,
+      "notify": 28063,
+      "▁participant": 28064,
+      "▁Wang": 28065,
+      "▁Materials": 28066,
+      "▁reven": 28067,
+      "前端": 28068,
+      "宁县": 28069,
+      "其实是": 28070,
+      "的价值": 28071,
+      "▁dans": 28072,
+      "▁Internal": 28073,
+      "tSystem": 28074,
+      "直观": 28075,
+      "一波": 28076,
+      "▁Wright": 28077,
+      "首届": 28078,
+      "为客户": 28079,
+      "▁upstream": 28080,
+      "▁VAL": 28081,
+      "▁Bund": 28082,
+      "'])\\": 28083,
+      "Zone": 28084,
+      "▁blockchain": 28085,
+      "有线": 28086,
+      "梳理": 28087,
+      "▁prognostic": 28088,
+      "奇数": 28089,
+      "▁MAX": 28090,
+      "▁\");\\": 28091,
+      "qs": 28092,
+      "传达": 28093,
+      "MET": 28094,
+      "▁overseas": 28095,
+      "之间的关系": 28096,
+      "燃气": 28097,
+      "[{": 28098,
+      "Hg": 28099,
+      "}\")\\": 28100,
+      "NUMBER": 28101,
+      "追究": 28102,
+      "\\\\|": 28103,
+      "还是很": 28104,
+      "да": 28105,
+      "▁Russell": 28106,
+      "的理由": 28107,
+      "风光": 28108,
+      "initions": 28109,
+      "TW": 28110,
+      "▁criticism": 28111,
+      "▁\\'#": 28112,
+      "▁Baltimore": 28113,
+      "Ali": 28114,
+      "的优势": 28115,
+      "▁经营范围": 28116,
+      "星座": 28117,
+      "sterdam": 28118,
+      "usp": 28119,
+      "▁thirty": 28120,
+      "compiler": 28121,
+      "尔多": 28122,
+      "纠结": 28123,
+      "的空间": 28124,
+      "rosine": 28125,
+      "▁Walter": 28126,
+      "literals": 28127,
+      "缓存": 28128,
+      "ichi": 28129,
+      "frames": 28130,
+      "▁interf": 28131,
+      "▁distinguished": 28132,
+      "nStr": 28133,
+      "不详": 28134,
+      "ahn": 28135,
+      "受欢迎": 28136,
+      "Volume": 28137,
+      "警告": 28138,
+      "ffff": 28139,
+      "camera": 28140,
+      "justify": 28141,
+      "散文": 28142,
+      "▁armed": 28143,
+      "uba": 28144,
+      "\\'];\\": 28145,
+      "Hy": 28146,
+      "xl": 28147,
+      "▁wis": 28148,
+      "numpy": 28149,
+      "▁CNN": 28150,
+      "▁microwave": 28151,
+      "UDE": 28152,
+      "hea": 28153,
+      "人介绍": 28154,
+      "来自于": 28155,
+      "▁damages": 28156,
+      "▁''": 28157,
+      "stick": 28158,
+      "▁breat": 28159,
+      "structions": 28160,
+      "nGu": 28161,
+      "chair": 28162,
+      "制成": 28163,
+      "分会": 28164,
+      "([]": 28165,
+      "开业": 28166,
+      "不愿意": 28167,
+      "▁Eight": 28168,
+      "▁英": 28169,
+      "▁releasing": 28170,
+      "▁Hallow": 28171,
+      "爱上": 28172,
+      "国王": 28173,
+      "acted": 28174,
+      "▁predicting": 28175,
+      "▁biomedical": 28176,
+      "semantic": 28177,
+      "写了": 28178,
+      "software": 28179,
+      "▁Rate": 28180,
+      "▁NH": 28181,
+      "▁pending": 28182,
+      "港口": 28183,
+      "cester": 28184,
+      "农药": 28185,
+      "▁pandas": 28186,
+      "▁Steven": 28187,
+      ">（": 28188,
+      "▁pent": 28189,
+      "nDavid": 28190,
+      "抓好": 28191,
+      "▁encompass": 28192,
+      "▁mandatory": 28193,
+      "门市": 28194,
+      "nThree": 28195,
+      "▁newsletter": 28196,
+      "▁predominantly": 28197,
+      "▁\\',": 28198,
+      "▁listener": 28199,
+      "▁retained": 28200,
+      "乳腺": 28201,
+      "实业": 28202,
+      "嫌疑人": 28203,
+      "nVol": 28204,
+      "Other": 28205,
+      "▁converter": 28206,
+      "Dependencies": 28207,
+      "▁Catal": 28208,
+      "rod": 28209,
+      "十四章": 28210,
+      "permission": 28211,
+      "以内": 28212,
+      "▁Kon": 28213,
+      "希望能": 28214,
+      "▁impressed": 28215,
+      "做过": 28216,
+      "Points": 28217,
+      "▁PROVID": 28218,
+      "▁portray": 28219,
+      "Board": 28220,
+      "另行": 28221,
+      "核算": 28222,
+      "谢邀": 28223,
+      "个单位": 28224,
+      "Expand": 28225,
+      "▁Floor": 28226,
+      "nwhat": 28227,
+      "都是一": 28228,
+      "LEVANCE": 28229,
+      "▁Branch": 28230,
+      "的双": 28231,
+      "▁{(": 28232,
+      "不等式的": 28233,
+      "rv": 28234,
+      "其他人": 28235,
+      "neys": 28236,
+      "terns": 28237,
+      "nPat": 28238,
+      "▁π": 28239,
+      "▁pneumonia": 28240,
+      "刘某某": 28241,
+      "wash": 28242,
+      "Track": 28243,
+      "▁amid": 28244,
+      "简洁": 28245,
+      "▁PH": 28246,
+      "▁Loc": 28247,
+      "▁Ti": 28248,
+      "▁cannab": 28249,
+      "▁cab": 28250,
+      "Mer": 28251,
+      "▁routing": 28252,
+      "Air": 28253,
+      "dbc": 28254,
+      "sig": 28255,
+      "▁rgba": 28256,
+      "▁vaccines": 28257,
+      "▁machinery": 28258,
+      "小伙伴": 28259,
+      "学习和": 28260,
+      "▁}}\"": 28261,
+      "月在": 28262,
+      "秋季": 28263,
+      "▁Alberta": 28264,
+      "spath": 28265,
+      "▁Pictures": 28266,
+      "▁ll": 28267,
+      "搞笑": 28268,
+      "▁ranks": 28269,
+      "否定": 28270,
+      "谈话": 28271,
+      "方程组": 28272,
+      "听力": 28273,
+      "证据确实": 28274,
+      "Println": 28275,
+      "的车": 28276,
+      "破产": 28277,
+      "▁oils": 28278,
+      "讲究": 28279,
+      "▁valued": 28280,
+      "▁basement": 28281,
+      "▁amenities": 28282,
+      "经销商": 28283,
+      "▁borders": 28284,
+      "不像": 28285,
+      "▁endif": 28286,
+      "门槛": 28287,
+      "▁filtering": 28288,
+      "原告诉称": 28289,
+      "atinum": 28290,
+      "▁estrogen": 28291,
+      "智力": 28292,
+      "▁coating": 28293,
+      "▁standardized": 28294,
+      "▁demonstrating": 28295,
+      "有问题": 28296,
+      "搜索引擎": 28297,
+      "作曲": 28298,
+      "的说": 28299,
+      "asse": 28300,
+      "球迷": 28301,
+      "Dig": 28302,
+      "scode": 28303,
+      "▁wiring": 28304,
+      "ngo": 28305,
+      "doctor": 28306,
+      "▁Costa": 28307,
+      "▁searches": 28308,
+      "hostname": 28309,
+      "FB": 28310,
+      "nWhether": 28311,
+      "nonatomic": 28312,
+      "人格": 28313,
+      "爸妈": 28314,
+      "合同履行": 28315,
+      "untary": 28316,
+      "▁foster": 28317,
+      "▁morphological": 28318,
+      "厂房": 28319,
+      "木齐": 28320,
+      "▁latex": 28321,
+      "生平": 28322,
+      "中部": 28323,
+      "▁overhead": 28324,
+      "▁divorce": 28325,
+      "este": 28326,
+      "▁Intelligence": 28327,
+      "complex": 28328,
+      "的临床": 28329,
+      "年起": 28330,
+      "nLooking": 28331,
+      "nThus": 28332,
+      "探测": 28333,
+      "▁游戏类型": 28334,
+      "自带": 28335,
+      "▁diamond": 28336,
+      "的海": 28337,
+      "/\"\\": 28338,
+      "icrobial": 28339,
+      "▁distress": 28340,
+      "中西": 28341,
+      "\"},": 28342,
+      "▁AV": 28343,
+      "Standard": 28344,
+      "▁Comb": 28345,
+      "containing": 28346,
+      "圣诞": 28347,
+      "Mouse": 28348,
+      "▁RAM": 28349,
+      "▁stones": 28350,
+      "进制": 28351,
+      "碰到": 28352,
+      "▁Hend": 28353,
+      "对着": 28354,
+      "大批": 28355,
+      "▁UP": 28356,
+      "机动车交通事故责任": 28357,
+      "与服务": 28358,
+      "Leg": 28359,
+      "▁Kenya": 28360,
+      "第十五": 28361,
+      "six": 28362,
+      "政府的": 28363,
+      "▁Bright": 28364,
+      "fac": 28365,
+      "nVis": 28366,
+      "▁韩": 28367,
+      "generic": 28368,
+      "意识到": 28369,
+      "足球俱乐部": 28370,
+      "▁Century": 28371,
+      "▁keen": 28372,
+      "${\\\\": 28373,
+      "usions": 28374,
+      "异的": 28375,
+      "▁Deputy": 28376,
+      "▁shallow": 28377,
+      "司马": 28378,
+      "▁cytos": 28379,
+      "▁organize": 28380,
+      "Ctrl": 28381,
+      "花开": 28382,
+      "fname": 28383,
+      "套房": 28384,
+      "▁Lanka": 28385,
+      "nMod": 28386,
+      "▁lag": 28387,
+      "▁Anti": 28388,
+      "月任": 28389,
+      "schedule": 28390,
+      "************": 28391,
+      "why": 28392,
+      "▁Syn": 28393,
+      "▁fascinating": 28394,
+      "\\\\)\\\\(": 28395,
+      "跟你": 28396,
+      "nar": 28397,
+      "}}{{": 28398,
+      "▁unem": 28399,
+      "乌鲁木齐": 28400,
+      "));": 28401,
+      "▁Similarly": 28402,
+      "欧阳": 28403,
+      "▁本": 28404,
+      "社会科学": 28405,
+      "道歉": 28406,
+      "\"${": 28407,
+      "一根": 28408,
+      "xxxx": 28409,
+      "▁Communication": 28410,
+      "离心": 28411,
+      "SSL": 28412,
+      "▁Operations": 28413,
+      "▁Ja": 28414,
+      "▁USD": 28415,
+      "扶持": 28416,
+      "致使": 28417,
+      "TPS": 28418,
+      "screens": 28419,
+      "▁deletion": 28420,
+      "礼仪": 28421,
+      ":\",": 28422,
+      "isec": 28423,
+      "网络安全": 28424,
+      "即为": 28425,
+      "只有一个": 28426,
+      "generation": 28427,
+      "▁Chi": 28428,
+      "shal": 28429,
+      "▁Chemistry": 28430,
+      "yntax": 28431,
+      "Camera": 28432,
+      "▁lease": 28433,
+      "基金会": 28434,
+      "▁sperm": 28435,
+      "经营者": 28436,
+      "KB": 28437,
+      "脱离": 28438,
+      "hydroxy": 28439,
+      "▁switched": 28440,
+      "▁Appe": 28441,
+      "▁Tags": 28442,
+      "(/": 28443,
+      "▁subscribe": 28444,
+      "信贷": 28445,
+      "ienna": 28446,
+      "▁Porn": 28447,
+      "这个人": 28448,
+      "执行案件": 28449,
+      "opa": 28450,
+      "等工作": 28451,
+      "nFormer": 28452,
+      "▁Orleans": 28453,
+      ">()": 28454,
+      "▁repro": 28455,
+      "segment": 28456,
+      "nea": 28457,
+      "▁fate": 28458,
+      "species": 28459,
+      "不忘": 28460,
+      "brain": 28461,
+      "▁dancing": 28462,
+      "▁discipline": 28463,
+      "`)\\": 28464,
+      "▁workforce": 28465,
+      "选定": 28466,
+      "不好的": 28467,
+      "更容易": 28468,
+      "▁Menu": 28469,
+      "▁ho": 28470,
+      "▁Say": 28471,
+      "Foundation": 28472,
+      "▁Mediterranean": 28473,
+      "▁安": 28474,
+      "▁对": 28475,
+      "管理和": 28476,
+      "可以看到": 28477,
+      "▁chick": 28478,
+      "▁underground": 28479,
+      "癌症": 28480,
+      "三位": 28481,
+      "消灭": 28482,
+      "兰州": 28483,
+      "seg": 28484,
+      "▁optimum": 28485,
+      "▁locality": 28486,
+      "▁Soft": 28487,
+      "activate": 28488,
+      "nconsole": 28489,
+      "▁subunit": 28490,
+      "▁Gray": 28491,
+      "Errorf": 28492,
+      "▁Barcelona": 28493,
+      "特色的": 28494,
+      "channels": 28495,
+      "nblablabla": 28496,
+      "▁CN": 28497,
+      "▁Joy": 28498,
+      "▁wallet": 28499,
+      "izar": 28500,
+      "无比": 28501,
+      "更高的": 28502,
+      "isol": 28503,
+      "重阳": 28504,
+      "恶性": 28505,
+      "▁quasi": 28506,
+      "▁congen": 28507,
+      "▁monoclonal": 28508,
+      "胡椒": 28509,
+      "▁RELEVANCE": 28510,
+      "▁descriptions": 28511,
+      "抗日": 28512,
+      "电阻": 28513,
+      "纯粹": 28514,
+      "放下": 28515,
+      "torch": 28516,
+      "▁infarction": 28517,
+      "乐园": 28518,
+      "aja": 28519,
+      "马克思主义": 28520,
+      "▁Rank": 28521,
+      "喝酒": 28522,
+      "Shape": 28523,
+      "公司在": 28524,
+      "▁om": 28525,
+      "那时": 28526,
+      "▁defin": 28527,
+      "出租车": 28528,
+      "UUID": 28529,
+      "修理": 28530,
+      "▁disabilities": 28531,
+      "▁beet": 28532,
+      "ograms": 28533,
+      "▁IDE": 28534,
+      "党委副书记": 28535,
+      "follow": 28536,
+      "▁redis": 28537,
+      "isEmpty": 28538,
+      "xc": 28539,
+      "aku": 28540,
+      "▁主要食材": 28541,
+      "Iss": 28542,
+      "尔斯": 28543,
+      "nMean": 28544,
+      "▁apolog": 28545,
+      "BUILD": 28546,
+      "权证": 28547,
+      "得更": 28548,
+      "扩散": 28549,
+      "应用程序": 28550,
+      "▁fractures": 28551,
+      "Tech": 28552,
+      "▁Phoenix": 28553,
+      "迎接": 28554,
+      "▁transmitted": 28555,
+      "cou": 28556,
+      "人生的": 28557,
+      "▁Hyp": 28558,
+      "相等的": 28559,
+      "▁IFN": 28560,
+      "▁strengths": 28561,
+      "ки": 28562,
+      "▁followers": 28563,
+      "县长": 28564,
+      "管理系统": 28565,
+      "钻石": 28566,
+      "等相关": 28567,
+      "▁transit": 28568,
+      "风扇": 28569,
+      "nSta": 28570,
+      "▁smell": 28571,
+      "ibe": 28572,
+      "raf": 28573,
+      "▁Convention": 28574,
+      "tbreak": 28575,
+      "久的": 28576,
+      "▁steep": 28577,
+      "Plus": 28578,
+      "第四百八": 28579,
+      "报纸": 28580,
+      "红外": 28581,
+      "▁Venezuel": 28582,
+      "▁caring": 28583,
+      "▁troubles": 28584,
+      "bai": 28585,
+      "▁MM": 28586,
+      "Split": 28587,
+      "传统文化": 28588,
+      "▁antim": 28589,
+      "woman": 28590,
+      "▁Spot": 28591,
+      "之家": 28592,
+      "sat": 28593,
+      "DOM": 28594,
+      "▁matt": 28595,
+      "展会": 28596,
+      "▁threatened": 28597,
+      "IFI": 28598,
+      "设备有限公司": 28599,
+      "▁Copy": 28600,
+      "▁renowned": 28601,
+      "typescript": 28602,
+      "▁appliances": 28603,
+      "▁determines": 28604,
+      "▁Pittsburgh": 28605,
+      "呵呵": 28606,
+      "▁Seven": 28607,
+      "恶意": 28608,
+      "failed": 28609,
+      "不确定": 28610,
+      "Browser": 28611,
+      "和大": 28612,
+      "KT": 28613,
+      "归纳": 28614,
+      "为目的": 28615,
+      "向申请执行人": 28616,
+      "这个世界": 28617,
+      "mouse": 28618,
+      "RD": 28619,
+      "▁Netflix": 28620,
+      "▁profits": 28621,
+      "▁Basketball": 28622,
+      "Put": 28623,
+      "▁Compared": 28624,
+      "村庄": 28625,
+      "克力": 28626,
+      "▁Niger": 28627,
+      "definition": 28628,
+      "▁translational": 28629,
+      "之内": 28630,
+      "讲师": 28631,
+      "▁ischemic": 28632,
+      "情景": 28633,
+      "技术和": 28634,
+      "▁Stewart": 28635,
+      "Solve": 28636,
+      "▁所属地区": 28637,
+      "Summary": 28638,
+      "wb": 28639,
+      "相加": 28640,
+      "用到": 28641,
+      "Cur": 28642,
+      "外语": 28643,
+      "天堂": 28644,
+      "专属": 28645,
+      "MED": 28646,
+      "的治疗": 28647,
+      "立方米": 28648,
+      "言情小说": 28649,
+      "▁南": 28650,
+      "}},": 28651,
+      "遭到": 28652,
+      "拉克": 28653,
+      "▁Poss": 28654,
+      "ooting": 28655,
+      "Force": 28656,
+      "▁Simply": 28657,
+      "基础题": 28658,
+      "▁neonatal": 28659,
+      "edges": 28660,
+      "认真遵守法律法规及监规": 28661,
+      "ogue": 28662,
+      "▁Meg": 28663,
+      "▁phy": 28664,
+      "nHey": 28665,
+      "年利率": 28666,
+      "▁游戏目标": 28667,
+      "▁survivors": 28668,
+      "巅峰": 28669,
+      "可以直接": 28670,
+      "▁fluorescent": 28671,
+      "防守": 28672,
+      "腹部": 28673,
+      "assertThat": 28674,
+      "▁contractor": 28675,
+      "终点": 28676,
+      "▁'#": 28677,
+      "▁Hunter": 28678,
+      ":**\\": 28679,
+      "▁consultant": 28680,
+      "▁invariant": 28681,
+      "签证": 28682,
+      "boss": 28683,
+      "困扰": 28684,
+      "增多": 28685,
+      "▁wheat": 28686,
+      "Diff": 28687,
+      "rating": 28688,
+      "▁motif": 28689,
+      "前来": 28690,
+      "▁disadvant": 28691,
+      "再来": 28692,
+      "指针": 28693,
+      "已经是": 28694,
+      "anic": 28695,
+      "▁clustering": 28696,
+      "▁accordingly": 28697,
+      "rocytes": 28698,
+      "出身": 28699,
+      "▁relies": 28700,
+      "▁Ruby": 28701,
+      "▁implicated": 28702,
+      "真实意思表示": 28703,
+      "▁stimulate": 28704,
+      "▁filing": 28705,
+      "▁Bear": 28706,
+      "▁spherical": 28707,
+      "微波": 28708,
+      "bia": 28709,
+      "▁Individual": 28710,
+      "简体中文": 28711,
+      "流浪": 28712,
+      "▁Birmingham": 28713,
+      "▁metabolites": 28714,
+      "▁nude": 28715,
+      "▁soluble": 28716,
+      "▁品牌": 28717,
+      "的建议": 28718,
+      "▁regeneration": 28719,
+      "说明书": 28720,
+      "iply": 28721,
+      "产能": 28722,
+      "治安": 28723,
+      "▁人物履历": 28724,
+      "▁viewers": 28725,
+      "又能": 28726,
+      "▁breakdown": 28727,
+      "名下的": 28728,
+      "▁juris": 28729,
+      "▁Legal": 28730,
+      "▁lying": 28731,
+      "asted": 28732,
+      "套路": 28733,
+      "▁Impact": 28734,
+      "uter": 28735,
+      "▁pathogen": 28736,
+      "带头": 28737,
+      "▁Fame": 28738,
+      "nlocal": 28739,
+      "会对": 28740,
+      "感激": 28741,
+      "他就": 28742,
+      "▁spokesman": 28743,
+      "Plan": 28744,
+      "▁Pel": 28745,
+      "every": 28746,
+      "▁cubic": 28747,
+      "clip": 28748,
+      "▁cad": 28749,
+      "arious": 28750,
+      "宝石": 28751,
+      "previous": 28752,
+      "posable": 28753,
+      "北区": 28754,
+      "河流": 28755,
+      "▁LIABILITY": 28756,
+      "向下": 28757,
+      "▁Sequ": 28758,
+      "▁poison": 28759,
+      "▁lever": 28760,
+      "耳朵": 28761,
+      "Which": 28762,
+      "White": 28763,
+      "月参加工作": 28764,
+      "忍不住": 28765,
+      "▁Shah": 28766,
+      "LT": 28767,
+      "Visible": 28768,
+      "表格": 28769,
+      "▁Models": 28770,
+      "毫不": 28771,
+      "▁Ground": 28772,
+      "othy": 28773,
+      "▁Hud": 28774,
+      "▁Afghanistan": 28775,
+      "犹豫": 28776,
+      "UES": 28777,
+      "▁mainstream": 28778,
+      "▁Want": 28779,
+      "金山": 28780,
+      "▁careg": 28781,
+      "▁PV": 28782,
+      "holm": 28783,
+      "▁violent": 28784,
+      "▁compress": 28785,
+      "lip": 28786,
+      "ichard": 28787,
+      "CRIPTION": 28788,
+      "ITS": 28789,
+      "hend": 28790,
+      "wheel": 28791,
+      "▁pathogens": 28792,
+      "▁remodeling": 28793,
+      "▁peers": 28794,
+      "encephal": 28795,
+      "pH": 28796,
+      "nsudo": 28797,
+      "syn": 28798,
+      "▁Ze": 28799,
+      "▁modelling": 28800,
+      "oct": 28801,
+      "idea": 28802,
+      "▁Wol": 28803,
+      "▁greet": 28804,
+      "nMus": 28805,
+      "checked": 28806,
+      "▁eliminated": 28807,
+      "获悉": 28808,
+      "▁transmit": 28809,
+      "了好": 28810,
+      "Tex": 28811,
+      "▁}\\\\": 28812,
+      "哈哈哈哈": 28813,
+      "▁hemor": 28814,
+      "生死": 28815,
+      "▁indication": 28816,
+      "▁conferences": 28817,
+      "Native": 28818,
+      "itance": 28819,
+      "▁retailers": 28820,
+      "isy": 28821,
+      "全局": 28822,
+      "ehicle": 28823,
+      "▁Easter": 28824,
+      "▁oriented": 28825,
+      "binom": 28826,
+      "▁washing": 28827,
+      "导体": 28828,
+      "一九": 28829,
+      "但不": 28830,
+      "oric": 28831,
+      "▁favorable": 28832,
+      "▁cytokine": 28833,
+      "nPosted": 28834,
+      "ulia": 28835,
+      "个小": 28836,
+      "alym": 28837,
+      "▁economics": 28838,
+      "Docker": 28839,
+      "▁exped": 28840,
+      "\\\\\":": 28841,
+      "▁Maur": 28842,
+      "▁distinction": 28843,
+      "▁Hey": 28844,
+      "进化": 28845,
+      "▁Vert": 28846,
+      "▁Peace": 28847,
+      "▁Rugby": 28848,
+      "发现被执行人有可供执行的财产": 28849,
+      "▁communic": 28850,
+      "▁Invalid": 28851,
+      "外界": 28852,
+      "失业": 28853,
+      "▁Guy": 28854,
+      "nsure": 28855,
+      "appings": 28856,
+      "▁Susan": 28857,
+      "▁earliest": 28858,
+      "matches": 28859,
+      "▁Tig": 28860,
+      "队列": 28861,
+      "adjust": 28862,
+      "▁Ki": 28863,
+      "慈善": 28864,
+      "阿姨": 28865,
+      "aths": 28866,
+      "instein": 28867,
+      "环境下": 28868,
+      "理事长": 28869,
+      "建华": 28870,
+      "▁lecture": 28871,
+      "这个数": 28872,
+      "存放": 28873,
+      "▁commented": 28874,
+      "▁discounts": 28875,
+      "住福建省": 28876,
+      "DIS": 28877,
+      "▁enormous": 28878,
+      "▁dominated": 28879,
+      "你觉得": 28880,
+      "▁NMR": 28881,
+      "农场": 28882,
+      "抽取": 28883,
+      "nthen": 28884,
+      "模仿": 28885,
+      "▁exports": 28886,
+      "以前先行羁押的": 28887,
+      "Oh": 28888,
+      "ucker": 28889,
+      "▁visibility": 28890,
+      "▁`/": 28891,
+      "▁bra": 28892,
+      "ugust": 28893,
+      "ustain": 28894,
+      "▁buyer": 28895,
+      "多彩": 28896,
+      "▁VR": 28897,
+      "一万": 28898,
+      "fre": 28899,
+      "执行情况": 28900,
+      "▁Bour": 28901,
+      "▁Halloween": 28902,
+      "▁negatively": 28903,
+      "防疫": 28904,
+      "food": 28905,
+      "▁stays": 28906,
+      "常态": 28907,
+      "保安": 28908,
+      "hemy": 28909,
+      "少量": 28910,
+      "故原告": 28911,
+      "Remote": 28912,
+      "物联网": 28913,
+      "▁trait": 28914,
+      "cutaneous": 28915,
+      "多项式": 28916,
+      "▁cocaine": 28917,
+      "合作伙伴": 28918,
+      "▁Upon": 28919,
+      "▁Combin": 28920,
+      "▁criterion": 28921,
+      "▁Kan": 28922,
+      "Decimal": 28923,
+      "▁profil": 28924,
+      "▁counties": 28925,
+      "通则": 28926,
+      "▁easiest": 28927,
+      "某一": 28928,
+      "汉字": 28929,
+      "TAG": 28930,
+      "并与": 28931,
+      "Bay": 28932,
+      "ushes": 28933,
+      "马拉": 28934,
+      "▁purple": 28935,
+      "odeship": 28936,
+      "querySelector": 28937,
+      "有用": 28938,
+      "▁数据": 28939,
+      "背包": 28940,
+      "了自己的": 28941,
+      "大事": 28942,
+      "MQ": 28943,
+      "▁primes": 28944,
+      "▁ligands": 28945,
+      "我喜欢": 28946,
+      "##\\": 28947,
+      "observ": 28948,
+      "nullable": 28949,
+      "▁Herm": 28950,
+      "何况": 28951,
+      "▁rally": 28952,
+      "度和": 28953,
+      "是他": 28954,
+      "▁Friend": 28955,
+      "先把": 28956,
+      "的国家": 28957,
+      "▁Along": 28958,
+      "▁dozen": 28959,
+      "straints": 28960,
+      "▁CLI": 28961,
+      "听取": 28962,
+      "外包": 28963,
+      "存在着": 28964,
+      "etween": 28965,
+      "跟随": 28966,
+      "Mesh": 28967,
+      "改编": 28968,
+      "拖欠": 28969,
+      "▁~/": 28970,
+      "▁reversed": 28971,
+      "烈士": 28972,
+      "▁crypto": 28973,
+      "难点": 28974,
+      "的答案": 28975,
+      "▁exert": 28976,
+      "pod": 28977,
+      "oval": 28978,
+      "上帝": 28979,
+      "▁fought": 28980,
+      "征收社会抚养费": 28981,
+      "▁terrible": 28982,
+      "游戏的": 28983,
+      "▁fabulous": 28984,
+      "Vertex": 28985,
+      "▁Prix": 28986,
+      "▁Soccer": 28987,
+      "转发": 28988,
+      "Vec": 28989,
+      "第一百四十五条": 28990,
+      "婴儿床": 28991,
+      "▁buf": 28992,
+      "nFootball": 28993,
+      "▁multivariate": 28994,
+      "▁ly": 28995,
+      "就知道": 28996,
+      "珠宝": 28997,
+      "nHence": 28998,
+      "▁Container": 28999,
+      "智能化": 29000,
+      "没啥": 29001,
+      "▁blot": 29002,
+      "nWill": 29003,
+      "等多": 29004,
+      "▁Race": 29005,
+      "▁archae": 29006,
+      "▁puzzle": 29007,
+      "\"%": 29008,
+      "日向被执行人": 29009,
+      "▁enrich": 29010,
+      "▁harmful": 29011,
+      "ATOR": 29012,
+      "▁glasses": 29013,
+      "刑期从判决执行之日起计算": 29014,
+      "▁airline": 29015,
+      "Complete": 29016,
+      "▁bronch": 29017,
+      "朋友们": 29018,
+      ")>": 29019,
+      "泡沫": 29020,
+      "(**": 29021,
+      "▁slower": 29022,
+      "▁Alice": 29023,
+      "▁Features": 29024,
+      "▁donated": 29025,
+      "蜂蜜": 29026,
+      "rowsers": 29027,
+      "▁withdrawal": 29028,
+      "▁Voiv": 29029,
+      "▁spare": 29030,
+      "▁STR": 29031,
+      "传奇私服": 29032,
+      "▁Commercial": 29033,
+      ",&": 29034,
+      "不准": 29035,
+      "anç": 29036,
+      "▁defence": 29037,
+      "中小学": 29038,
+      "▁distal": 29039,
+      "ARGET": 29040,
+      "▁README": 29041,
+      "etitive": 29042,
+      "Delegate": 29043,
+      "▁cannabis": 29044,
+      "当事人的": 29045,
+      "▁剧名": 29046,
+      "羽毛": 29047,
+      "还有一个": 29048,
+      "▁deputy": 29049,
+      "▁bust": 29050,
+      "经济技术开发区": 29051,
+      "ioned": 29052,
+      "老的": 29053,
+      "Sharp": 29054,
+      "▁Nad": 29055,
+      "ygon": 29056,
+      "▁Operation": 29057,
+      "麻醉": 29058,
+      "azed": 29059,
+      "开发的": 29060,
+      "教育局": 29061,
+      "olytic": 29062,
+      "▁aligned": 29063,
+      "女神": 29064,
+      "影响到": 29065,
+      "▁cattle": 29066,
+      "▁Architecture": 29067,
+      "mr": 29068,
+      "思想政治": 29069,
+      "▁sends": 29070,
+      "▁月": 29071,
+      "Ste": 29072,
+      "中美": 29073,
+      "不一样的": 29074,
+      "▁radar": 29075,
+      "oria": 29076,
+      "眼泪": 29077,
+      "▁GC": 29078,
+      "insula": 29079,
+      "唱歌": 29080,
+      "▁Tar": 29081,
+      "▁guns": 29082,
+      "reduce": 29083,
+      "▁escort": 29084,
+      "▁Ric": 29085,
+      "▁cis": 29086,
+      "carbon": 29087,
+      "▁DIS": 29088,
+      "投放": 29089,
+      "pract": 29090,
+      "战场": 29091,
+      "和分析思路": 29092,
+      "invasive": 29093,
+      "自杀": 29094,
+      "Life": 29095,
+      "▁polymerase": 29096,
+      "找个": 29097,
+      "▁Rat": 29098,
+      "▁lazy": 29099,
+      "Redirect": 29100,
+      "塞尔": 29101,
+      "有关规定": 29102,
+      "理论与": 29103,
+      "一般来说": 29104,
+      "▁YOUR": 29105,
+      "跨越": 29106,
+      "personal": 29107,
+      "Asp": 29108,
+      "的风险": 29109,
+      "pective": 29110,
+      "▁daughters": 29111,
+      "不堪": 29112,
+      "减半收取计": 29113,
+      "▁Alpha": 29114,
+      "▁affiliate": 29115,
+      "aligned": 29116,
+      "澳洲": 29117,
+      "▁Ghana": 29118,
+      "我真的": 29119,
+      "年毕业于": 29120,
+      "瑞士": 29121,
+      "让学生": 29122,
+      "如图": 29123,
+      "▁cous": 29124,
+      "▁bless": 29125,
+      "arded": 29126,
+      "ictional": 29127,
+      "钥匙": 29128,
+      "udes": 29129,
+      "▁&&\\": 29130,
+      "比如说": 29131,
+      "rition": 29132,
+      "▁Jonathan": 29133,
+      "▁Stars": 29134,
+      "管制": 29135,
+      "仅供参考": 29136,
+      "▁castle": 29137,
+      "分裂": 29138,
+      "atomy": 29139,
+      "▁Statement": 29140,
+      "▁pine": 29141,
+      "Perm": 29142,
+      "acting": 29143,
+      "起步": 29144,
+      "]))": 29145,
+      "shine": 29146,
+      "▁harmonic": 29147,
+      "中外": 29148,
+      "紧紧": 29149,
+      "fed": 29150,
+      "tot": 29151,
+      "出生于": 29152,
+      "深深": 29153,
+      "技术人员": 29154,
+      "▁hemorrh": 29155,
+      "▁deficits": 29156,
+      "▁commune": 29157,
+      "的原": 29158,
+      "▁prescribed": 29159,
+      "浸泡": 29160,
+      "▁sized": 29161,
+      "▁joints": 29162,
+      "江山": 29163,
+      "luence": 29164,
+      "知识和": 29165,
+      "vision": 29166,
+      "较强的": 29167,
+      "▁$\\\\{": 29168,
+      "▁Icon": 29169,
+      "▁intrac": 29170,
+      "心动": 29171,
+      "▁({\\": 29172,
+      "twhile": 29173,
+      "炎症": 29174,
+      "jun": 29175,
+      "▁costly": 29176,
+      "they": 29177,
+      "这道选择题": 29178,
+      "PEG": 29179,
+      "permalink": 29180,
+      "OTE": 29181,
+      "laration": 29182,
+      "adapter": 29183,
+      "nItalian": 29184,
+      "endant": 29185,
+      "▁adherence": 29186,
+      "▁dend": 29187,
+      "ICATION": 29188,
+      "▁Tan": 29189,
+      "▁Warren": 29190,
+      "▁ambit": 29191,
+      "不受申请执行": 29192,
+      "feedback": 29193,
+      "independent": 29194,
+      "准予原告": 29195,
+      "▁Problems": 29196,
+      "度过": 29197,
+      "▁progen": 29198,
+      "▁permits": 29199,
+      "▁uter": 29200,
+      "due": 29201,
+      "nMaybe": 29202,
+      "▁milit": 29203,
+      "leaf": 29204,
+      "▁TGF": 29205,
+      "Share": 29206,
+      "▁Palace": 29207,
+      "亏损": 29208,
+      "▁注音": 29209,
+      "▁Fried": 29210,
+      "有益": 29211,
+      "LEASE": 29212,
+      "okers": 29213,
+      "▁assured": 29214,
+      "制片": 29215,
+      "灰色": 29216,
+      "对自己的": 29217,
+      "▁Improve": 29218,
+      "adays": 29219,
+      "▁homeost": 29220,
+      "▁Sent": 29221,
+      "Dynamic": 29222,
+      "▁spectrometry": 29223,
+      "▁relaxing": 29224,
+      "仔细阅读": 29225,
+      "▁Kle": 29226,
+      "▁Portuguese": 29227,
+      "/\\'": 29228,
+      "hooks": 29229,
+      "着的": 29230,
+      "▁irrit": 29231,
+      "若是": 29232,
+      "▁dopamine": 29233,
+      "nax": 29234,
+      "▁Campbell": 29235,
+      "flush": 29236,
+      "▁penet": 29237,
+      "nJan": 29238,
+      "▁LC": 29239,
+      "▁nested": 29240,
+      "Expected": 29241,
+      "Management": 29242,
+      "▁ventilation": 29243,
+      "▁ug": 29244,
+      "▁bout": 29245,
+      "supported": 29246,
+      "▁energies": 29247,
+      "建军": 29248,
+      "▁aqueous": 29249,
+      "亲子": 29250,
+      "越高": 29251,
+      "ORE": 29252,
+      "postgres": 29253,
+      "MY": 29254,
+      "核实": 29255,
+      "失去了": 29256,
+      "▁naked": 29257,
+      "]['": 29258,
+      "行包含": 29259,
+      "haviour": 29260,
+      "▁Ce": 29261,
+      "的交": 29262,
+      "arct": 29263,
+      "matched": 29264,
+      "▁crystals": 29265,
+      "}}(": 29266,
+      "unknown": 29267,
+      "▁Weight": 29268,
+      "这对": 29269,
+      "Warning": 29270,
+      "▁rejection": 29271,
+      "createElement": 29272,
+      "▁resil": 29273,
+      "▁blogs": 29274,
+      "▁shelf": 29275,
+      "nST": 29276,
+      "缓慢": 29277,
+      "ximity": 29278,
+      "▁ranged": 29279,
+      "佩戴": 29280,
+      "opol": 29281,
+      "enes": 29282,
+      "▁gaining": 29283,
+      "▁pics": 29284,
+      "▁predictors": 29285,
+      "uropean": 29286,
+      "滨海": 29287,
+      "银川": 29288,
+      "独自": 29289,
+      "▁Ku": 29290,
+      "▁ou": 29291,
+      "▁STE": 29292,
+      "▁Salt": 29293,
+      "arians": 29294,
+      "Cluster": 29295,
+      "assy": 29296,
+      "摆脱": 29297,
+      "rain": 29298,
+      "▁shipped": 29299,
+      "总量": 29300,
+      "mall": 29301,
+      "▁imports": 29302,
+      "▁anchor": 29303,
+      "▁Homework": 29304,
+      "opedia": 29305,
+      "▁homogeneous": 29306,
+      "走的": 29307,
+      "yler": 29308,
+      "achers": 29309,
+      "花的": 29310,
+      "clipse": 29311,
+      "▁Domain": 29312,
+      "JP": 29313,
+      "本裁定书送达后立即生效": 29314,
+      "▁accompanying": 29315,
+      "die": 29316,
+      "判决执行以前先行羁押的": 29317,
+      "负面": 29318,
+      "▁Fal": 29319,
+      "▁Entry": 29320,
+      "▁freely": 29321,
+      "∈[": 29322,
+      "递减": 29323,
+      "Arrays": 29324,
+      "imgur": 29325,
+      "前锋": 29326,
+      "▁expend": 29327,
+      "推行": 29328,
+      "Exists": 29329,
+      "改造积极分子": 29330,
+      "的销售": 29331,
+      "分母": 29332,
+      "若干问题的意见": 29333,
+      "▁expectation": 29334,
+      "▁existed": 29335,
+      "▁scientist": 29336,
+      "▁Documentation": 29337,
+      "延迟": 29338,
+      "中秋": 29339,
+      "▁keywords": 29340,
+      "nSer": 29341,
+      "智商": 29342,
+      "▁satisfies": 29343,
+      "reliminary": 29344,
+      "▁silicon": 29345,
+      "战胜": 29346,
+      "amate": 29347,
+      "authent": 29348,
+      "▁cooked": 29349,
+      "typed": 29350,
+      "▁Carbon": 29351,
+      "▁Voivodeship": 29352,
+      "她在": 29353,
+      "▁SEC": 29354,
+      "▁Fant": 29355,
+      "▁isolate": 29356,
+      "各有": 29357,
+      "艺人": 29358,
+      "企业家": 29359,
+      "▁Rud": 29360,
+      "缓刑考验": 29361,
+      "wire": 29362,
+      "fen": 29363,
+      "电站": 29364,
+      "▁connects": 29365,
+      "▁%(": 29366,
+      "的一项": 29367,
+      "▁Dav": 29368,
+      "▁Ts": 29369,
+      "功能的": 29370,
+      "▁Squ": 29371,
+      "attach": 29372,
+      "▁tuberculosis": 29373,
+      "蝴蝶": 29374,
+      "TextField": 29375,
+      "▁intersect": 29376,
+      "进行了审查": 29377,
+      "借款合同纠纷一案中": 29378,
+      "avitational": 29379,
+      "得多": 29380,
+      "Authorization": 29381,
+      "刊物": 29382,
+      "所作": 29383,
+      "iffs": 29384,
+      "奖惩": 29385,
+      "专注于": 29386,
+      "▁symptom": 29387,
+      "radio": 29388,
+      "肠道": 29389,
+      "备考": 29390,
+      "tat": 29391,
+      "sever": 29392,
+      "罪犯评审鉴定表": 29393,
+      "▁heterogeneous": 29394,
+      "▁disrupt": 29395,
+      "DH": 29396,
+      "cells": 29397,
+      "▁Dynamic": 29398,
+      "▁vibrant": 29399,
+      "Errors": 29400,
+      ">#": 29401,
+      "沉默": 29402,
+      "▁shore": 29403,
+      "astics": 29404,
+      "ENS": 29405,
+      "bay": 29406,
+      "drive": 29407,
+      "▁Woman": 29408,
+      "▁Commons": 29409,
+      "是为": 29410,
+      "▁Ster": 29411,
+      "ICK": 29412,
+      "MAC": 29413,
+      "▁possession": 29414,
+      "优于": 29415,
+      "不可以": 29416,
+      "▁nons": 29417,
+      "▁ubiquit": 29418,
+      "中级": 29419,
+      "\"},\\": 29420,
+      "andidate": 29421,
+      "omo": 29422,
+      "ourt": 29423,
+      "otomy": 29424,
+      "▁promised": 29425,
+      "omers": 29426,
+      "▁desper": 29427,
+      "▁IgG": 29428,
+      "境界": 29429,
+      "ratio": 29430,
+      "▁wallpaper": 29431,
+      "不合": 29432,
+      "▁INCLUDING": 29433,
+      "▁affairs": 29434,
+      "▁rectangular": 29435,
+      "混乱": 29436,
+      "看作": 29437,
+      "▁comprises": 29438,
+      "▁dess": 29439,
+      "▁antioxidant": 29440,
+      "▁Ukrainian": 29441,
+      "受到了": 29442,
+      "▁slots": 29443,
+      "周岁": 29444,
+      "pit": 29445,
+      "▁pressures": 29446,
+      "▁chop": 29447,
+      "▁researcher": 29448,
+      "nRed": 29449,
+      "genase": 29450,
+      "▁residue": 29451,
+      "▁操作指南": 29452,
+      "▁dirty": 29453,
+      "∞)": 29454,
+      "侵害": 29455,
+      "love": 29456,
+      "▁probes": 29457,
+      "YO": 29458,
+      "刷新": 29459,
+      "▁Sets": 29460,
+      "▁endemic": 29461,
+      "▁Fel": 29462,
+      "平县": 29463,
+      "济南市": 29464,
+      "一阵": 29465,
+      "标本": 29466,
+      "▁cs": 29467,
+      "cuts": 29468,
+      "▁Athletics": 29469,
+      "暑假": 29470,
+      "▁NASA": 29471,
+      "▁LE": 29472,
+      "▁dashboard": 29473,
+      "支援": 29474,
+      "alph": 29475,
+      "abulary": 29476,
+      "全面的": 29477,
+      "▁receipt": 29478,
+      "▁Transfer": 29479,
+      "▁elimination": 29480,
+      "jl": 29481,
+      "aptist": 29482,
+      "fm": 29483,
+      "口语": 29484,
+      "和高": 29485,
+      "▁beautifully": 29486,
+      "UX": 29487,
+      "icensing": 29488,
+      "▁logistic": 29489,
+      "计生": 29490,
+      "各部门": 29491,
+      "我不知道": 29492,
+      "▁ram": 29493,
+      "CAS": 29494,
+      "EMENT": 29495,
+      "木材": 29496,
+      "▁清": 29497,
+      "▁unve": 29498,
+      "▁whit": 29499,
+      "白菜": 29500,
+      "mong": 29501,
+      "▁trucks": 29502,
+      "第一百三十三条": 29503,
+      "所长": 29504,
+      "辨认": 29505,
+      "▁{},\\": 29506,
+      "▁loops": 29507,
+      "Primary": 29508,
+      "在意": 29509,
+      "REQUEST": 29510,
+      "▁relaxed": 29511,
+      "▁inclusive": 29512,
+      "纳入失信被执行人名单": 29513,
+      "赌博": 29514,
+      "▁demanding": 29515,
+      "到期后": 29516,
+      "▁Gene": 29517,
+      "▁spacious": 29518,
+      "等的": 29519,
+      "▁Ele": 29520,
+      "可爱的": 29521,
+      "Pointer": 29522,
+      "线程": 29523,
+      "▁ore": 29524,
+      "你还": 29525,
+      "婆婆": 29526,
+      "PRE": 29527,
+      "如实供述自己的": 29528,
+      "的方程": 29529,
+      "支付原告": 29530,
+      "castle": 29531,
+      "门店": 29532,
+      "一点点": 29533,
+      "▁界": 29534,
+      "黑白": 29535,
+      "▁${{": 29536,
+      "▁MAP": 29537,
+      "长沙市": 29538,
+      "Ignore": 29539,
+      "▁irradiation": 29540,
+      "uclide": 29541,
+      "cook": 29542,
+      "▁bankrupt": 29543,
+      "unge": 29544,
+      "开学": 29545,
+      "POS": 29546,
+      "hips": 29547,
+      "▁priz": 29548,
+      "seen": 29549,
+      "▁blo": 29550,
+      "咨询服务": 29551,
+      "heap": 29552,
+      "▁governance": 29553,
+      "华夏": 29554,
+      "formatics": 29555,
+      "▁Thai": 29556,
+      "assembly": 29557,
+      "▁样例": 29558,
+      "章节": 29559,
+      "uron": 29560,
+      "▁Cox": 29561,
+      "hero": 29562,
+      "▁WordPress": 29563,
+      "弥补": 29564,
+      "▁accred": 29565,
+      "▁implants": 29566,
+      "每一位": 29567,
+      "▁delays": 29568,
+      "nName": 29569,
+      "▁determinants": 29570,
+      "▁announc": 29571,
+      "inda": 29572,
+      "倒是": 29573,
+      "▁needle": 29574,
+      "▁throwing": 29575,
+      "iq": 29576,
+      "发光": 29577,
+      "的时代": 29578,
+      "▁neat": 29579,
+      "otid": 29580,
+      "▁coupon": 29581,
+      "rpc": 29582,
+      "▁分类": 29583,
+      "过程中的": 29584,
+      "suite": 29585,
+      "Loop": 29586,
+      "投资有限公司": 29587,
+      "）[": 29588,
+      "resize": 29589,
+      "▁Everyone": 29590,
+      "alib": 29591,
+      "东区": 29592,
+      "AML": 29593,
+      "的水景": 29594,
+      "ivic": 29595,
+      "▁Hit": 29596,
+      "导致的": 29597,
+      "▁Ald": 29598,
+      "scss": 29599,
+      "▁seller": 29600,
+      "▁Ferr": 29601,
+      "▁Below": 29602,
+      "information": 29603,
+      ")_": 29604,
+      "▁Cooper": 29605,
+      "均可": 29606,
+      "▁bones": 29607,
+      "▁Je": 29608,
+      "市级": 29609,
+      "▁Publishing": 29610,
+      "\\\\%": 29611,
+      "货车": 29612,
+      "nBACKGROUND": 29613,
+      "した": 29614,
+      "突发": 29615,
+      "提交的": 29616,
+      "观光": 29617,
+      "eneath": 29618,
+      "▁rearr": 29619,
+      "nStaff": 29620,
+      "HH": 29621,
+      "一杯": 29622,
+      "既有": 29623,
+      "▁phosphate": 29624,
+      "ounters": 29625,
+      "IK": 29626,
+      "有人说": 29627,
+      "▁Carlo": 29628,
+      "numbers": 29629,
+      "▁deterior": 29630,
+      "关村": 29631,
+      "小学生": 29632,
+      "▁olive": 29633,
+      "▁regulates": 29634,
+      "SHA": 29635,
+      "▁Liberal": 29636,
+      "▁overwhelming": 29637,
+      "▁precious": 29638,
+      "▁vaccination": 29639,
+      "安慰": 29640,
+      "水质": 29641,
+      "▁painful": 29642,
+      "就行了": 29643,
+      "不应": 29644,
+      "扮演角色": 29645,
+      "▁rand": 29646,
+      "之道": 29647,
+      "auer": 29648,
+      "▁\"\\'": 29649,
+      "▁departure": 29650,
+      "▁earning": 29651,
+      "▁insertion": 29652,
+      "▁affiliated": 29653,
+      "▁ist": 29654,
+      "subseteq": 29655,
+      "ISING": 29656,
+      "nGroup": 29657,
+      "或其他": 29658,
+      "Bi": 29659,
+      "sudo": 29660,
+      "缺席判决": 29661,
+      "illes": 29662,
+      "最近的": 29663,
+      "▁inhabitants": 29664,
+      "党组织": 29665,
+      "ampa": 29666,
+      "▁opera": 29667,
+      "Nullable": 29668,
+      "▁Counter": 29669,
+      "▁Minor": 29670,
+      "▁Knowledge": 29671,
+      "▁converges": 29672,
+      "一个数": 29673,
+      "奇迹": 29674,
+      "有球队": 29675,
+      "▁showcase": 29676,
+      "urred": 29677,
+      "initialize": 29678,
+      "▁BA": 29679,
+      "urrencies": 29680,
+      "icul": 29681,
+      "attery": 29682,
+      "Operator": 29683,
+      "▁Secondary": 29684,
+      "食品安全": 29685,
+      "葡萄酒": 29686,
+      "与此同时": 29687,
+      "iasis": 29688,
+      "▁Tam": 29689,
+      "▁Firefox": 29690,
+      "▁Participants": 29691,
+      "▁Suite": 29692,
+      "▁Municipality": 29693,
+      "环境的": 29694,
+      "ctrine": 29695,
+      "▁circulating": 29696,
+      "alis": 29697,
+      "▁manifold": 29698,
+      "▁bridges": 29699,
+      "▁preparations": 29700,
+      "初始化": 29701,
+      "▁通": 29702,
+      "▁qualifying": 29703,
+      "▁Robinson": 29704,
+      "好事": 29705,
+      "▁Jews": 29706,
+      "▁pale": 29707,
+      "▁Constant": 29708,
+      "COUNT": 29709,
+      "▁critics": 29710,
+      "——\\": 29711,
+      "信誉": 29712,
+      "▁nobody": 29713,
+      "▁职务": 29714,
+      "▁bump": 29715,
+      "▁HOLD": 29716,
+      "NB": 29717,
+      "减小": 29718,
+      "过去的": 29719,
+      "▁Alt": 29720,
+      "▁abundant": 29721,
+      "看好": 29722,
+      "Enable": 29723,
+      "你的遭遇": 29724,
+      "▁INTO": 29725,
+      "nlog": 29726,
+      "有过": 29727,
+      "Azure": 29728,
+      "▁weld": 29729,
+      "▁Score": 29730,
+      "uent": 29731,
+      "县级": 29732,
+      "阳区": 29733,
+      "判决确定": 29734,
+      "的值为": 29735,
+      "▁Circuit": 29736,
+      "饱和": 29737,
+      "vertical": 29738,
+      "排行": 29739,
+      "▁hun": 29740,
+      "另一种": 29741,
+      "Intent": 29742,
+      "▁Writing": 29743,
+      "RENT": 29744,
+      "分式": 29745,
+      "▁Attribute": 29746,
+      "▁soap": 29747,
+      "▁ESP": 29748,
+      "▁theta": 29749,
+      "▁tt": 29750,
+      "▁AUTH": 29751,
+      "▁Feed": 29752,
+      "▁vig": 29753,
+      "▁zoom": 29754,
+      "▁minds": 29755,
+      "▁Hell": 29756,
+      "▁Wayne": 29757,
+      "▁cDNA": 29758,
+      "Sheet": 29759,
+      "▁desert": 29760,
+      "Online": 29761,
+      "▁catalytic": 29762,
+      "▁potentials": 29763,
+      "··": 29764,
+      "START": 29765,
+      "▁disciplines": 29766,
+      "spaces": 29767,
+      "喜剧": 29768,
+      "▁梁": 29769,
+      "etric": 29770,
+      "▁exponent": 29771,
+      "老百姓": 29772,
+      "▁venous": 29773,
+      "Future": 29774,
+      "▁Meanwhile": 29775,
+      "赤峰": 29776,
+      "braska": 29777,
+      "小区的水景": 29778,
+      "届全国": 29779,
+      "Like": 29780,
+      "多少小区的水景": 29781,
+      "酒后": 29782,
+      "是如何": 29783,
+      "▁mL": 29784,
+      "一个人的": 29785,
+      "不违反法律规定": 29786,
+      "▁atheros": 29787,
+      "▁Ec": 29788,
+      "▁orthogonal": 29789,
+      "Ev": 29790,
+      "WISE": 29791,
+      "▁reviewing": 29792,
+      "▁specially": 29793,
+      "▁Knight": 29794,
+      "播出": 29795,
+      "▁PROVIDED": 29796,
+      "lb": 29797,
+      "年内": 29798,
+      "战役": 29799,
+      "速率": 29800,
+      "我已经": 29801,
+      "数据显示": 29802,
+      "▁ger": 29803,
+      "▁apartments": 29804,
+      "与发展": 29805,
+      "nFinal": 29806,
+      "太多了": 29807,
+      "▁lob": 29808,
+      "额定": 29809,
+      "animation": 29810,
+      "学费": 29811,
+      "uptools": 29812,
+      "Ah": 29813,
+      "知识点": 29814,
+      "▁spreading": 29815,
+      "▁Strateg": 29816,
+      "▁Zero": 29817,
+      "去找": 29818,
+      "葫芦": 29819,
+      "因其": 29820,
+      "苏州市": 29821,
+      "▁disruption": 29822,
+      "一篇": 29823,
+      "\"/>": 29824,
+      "▁''\\": 29825,
+      "navigation": 29826,
+      "ncon": 29827,
+      "ributor": 29828,
+      "▁disappear": 29829,
+      "▁ende": 29830,
+      "allowed": 29831,
+      "真题": 29832,
+      "▁radial": 29833,
+      "▁lamp": 29834,
+      "▁incons": 29835,
+      "constant": 29836,
+      "▁Password": 29837,
+      "tikz": 29838,
+      "副院长": 29839,
+      "履行完毕": 29840,
+      "▁Evaluation": 29841,
+      "▁#\\": 29842,
+      "▁beaches": 29843,
+      "单调性": 29844,
+      "第五百一十九条的规定": 29845,
+      "▁Morris": 29846,
+      "▁denote": 29847,
+      "小数点": 29848,
+      "离不开": 29849,
+      "▁Pull": 29850,
+      "▁specialty": 29851,
+      "水晶": 29852,
+      "▁microbi": 29853,
+      "annah": 29854,
+      "▁prevalent": 29855,
+      "▁mathematic": 29856,
+      "▁dub": 29857,
+      "▁Nevertheless": 29858,
+      "婚前": 29859,
+      "尔夫": 29860,
+      "▁Draft": 29861,
+      "的食物": 29862,
+      "joys": 29863,
+      "curl": 29864,
+      "经纪人": 29865,
+      "自定义": 29866,
+      "汉语词语": 29867,
+      "之日": 29868,
+      "较小": 29869,
+      "贩卖毒品罪": 29870,
+      "终止": 29871,
+      "xyz": 29872,
+      "appropriate": 29873,
+      "侧面": 29874,
+      "建模": 29875,
+      "▁mapped": 29876,
+      "▁Lanc": 29877,
+      "idy": 29878,
+      "▁synth": 29879,
+      "放射": 29880,
+      "线条": 29881,
+      "nWould": 29882,
+      "harm": 29883,
+      "ulsion": 29884,
+      "SUCCESS": 29885,
+      "杆菌": 29886,
+      "MN": 29887,
+      "士兵": 29888,
+      "esium": 29889,
+      "机身": 29890,
+      "不爱": 29891,
+      "或打车": 29892,
+      "我没有": 29893,
+      "玩法": 29894,
+      "Photo": 29895,
+      "Mut": 29896,
+      "▁扮演角色": 29897,
+      "▁cruise": 29898,
+      "必需": 29899,
+      "无所": 29900,
+      "长为": 29901,
+      "期满": 29902,
+      "甲醛": 29903,
+      "菜品": 29904,
+      "▁Surface": 29905,
+      "classmethod": 29906,
+      "tset": 29907,
+      "港市": 29908,
+      "▁lemon": 29909,
+      "▁melanoma": 29910,
+      "十章": 29911,
+      "▁ech": 29912,
+      "Machine": 29913,
+      "Today": 29914,
+      "▁dirt": 29915,
+      "下滑": 29916,
+      "priority": 29917,
+      "absolute": 29918,
+      "年被": 29919,
+      "底部": 29920,
+      "路段": 29921,
+      "▁pussy": 29922,
+      "opus": 29923,
+      "▁当": 29924,
+      "bage": 29925,
+      "Since": 29926,
+      "▁Synt": 29927,
+      "▁Distribution": 29928,
+      "显示屏": 29929,
+      "achi": 29930,
+      "▁nicely": 29931,
+      "▁你": 29932,
+      "十万": 29933,
+      "▁psychology": 29934,
+      "▁instability": 29935,
+      "bread": 29936,
+      "资源的": 29937,
+      "▁formally": 29938,
+      "▁instrumental": 29939,
+      "mailto": 29940,
+      "犯故意伤害罪": 29941,
+      "正弦": 29942,
+      "▁outs": 29943,
+      "▁transformations": 29944,
+      "▁ash": 29945,
+      "▁trails": 29946,
+      "▁tuned": 29947,
+      "▁comed": 29948,
+      "▁pleasant": 29949,
+      "授课": 29950,
+      "漏洞": 29951,
+      "这块": 29952,
+      "途中": 29953,
+      "inet": 29954,
+      "忽然": 29955,
+      "密集": 29956,
+      "umi": 29957,
+      "没错": 29958,
+      "退房": 29959,
+      "▁dtype": 29960,
+      "繁华": 29961,
+      "';": 29962,
+      "▁'<": 29963,
+      "▁contacted": 29964,
+      "temperature": 29965,
+      "nTime": 29966,
+      "▁sparse": 29967,
+      "▁discre": 29968,
+      "门窗": 29969,
+      ")\",": 29970,
+      "tC": 29971,
+      "经贸": 29972,
+      "的表达": 29973,
+      "revision": 29974,
+      "Lambda": 29975,
+      "▁Charlotte": 29976,
+      "不让": 29977,
+      "活泼": 29978,
+      "▁intend": 29979,
+      "tstd": 29980,
+      "arness": 29981,
+      "▁modest": 29982,
+      "大火": 29983,
+      "▁Nurs": 29984,
+      "▁gi": 29985,
+      "Publish": 29986,
+      "▁settle": 29987,
+      "agar": 29988,
+      "脑子": 29989,
+      "▁Pic": 29990,
+      "▁Letter": 29991,
+      "标注": 29992,
+      "nFree": 29993,
+      "心肌": 29994,
+      "义乌": 29995,
+      "▁Sweet": 29996,
+      "▁runners": 29997,
+      "太空": 29998,
+      "ograft": 29999,
+      "▁Setup": 30000,
+      "他还": 30001,
+      "虚假": 30002,
+      "总面积": 30003,
+      "▁toss": 30004,
+      "在上": 30005,
+      "nnot": 30006,
+      "▁首发": 30007,
+      "▁Television": 30008,
+      "没了": 30009,
+      "▁tat": 30010,
+      "▁painters": 30011,
+      "inces": 30012,
+      "amilies": 30013,
+      "谁能": 30014,
+      "程度的": 30015,
+      "RNAs": 30016,
+      "ounded": 30017,
+      "被动": 30018,
+      "除此之外": 30019,
+      "▁sequel": 30020,
+      "中华人民共和国民法": 30021,
+      "nEducation": 30022,
+      "tesy": 30023,
+      "▁audiences": 30024,
+      "NAL": 30025,
+      "▁Lisa": 30026,
+      "▁籍贯": 30027,
+      "ETHER": 30028,
+      "▁susceptible": 30029,
+      "开工": 30030,
+      "▁homeless": 30031,
+      "peror": 30032,
+      "average": 30033,
+      "五条": 30034,
+      "抽象": 30035,
+      "本案中": 30036,
+      "理工大学": 30037,
+      "▁reopen": 30038,
+      "▁warned": 30039,
+      "关爱": 30040,
+      "Tom": 30041,
+      "perial": 30042,
+      "think": 30043,
+      "▁原": 30044,
+      "着重": 30045,
+      "▁intravenous": 30046,
+      "▁Therapy": 30047,
+      "网红": 30048,
+      "▁lingu": 30049,
+      "▁maintains": 30050,
+      "mother": 30051,
+      "▁leverage": 30052,
+      "万物": 30053,
+      "时期的": 30054,
+      "三等奖": 30055,
+      "女的": 30056,
+      "nget": 30057,
+      "央视": 30058,
+      "▁sentences": 30059,
+      "贝壳": 30060,
+      "▁eager": 30061,
+      "fact": 30062,
+      "Organ": 30063,
+      "套装": 30064,
+      "▁uma": 30065,
+      "不利": 30066,
+      "▁Evans": 30067,
+      "▁Samuel": 30068,
+      "▁Sad": 30069,
+      "Plot": 30070,
+      "▁Fresh": 30071,
+      "▁boats": 30072,
+      "▁curvature": 30073,
+      "背景下": 30074,
+      "▁EXPRESS": 30075,
+      "习题": 30076,
+      "▁Agent": 30077,
+      "开水": 30078,
+      "STRING": 30079,
+      "在我国": 30080,
+      "越好": 30081,
+      "atoes": 30082,
+      "▁Enum": 30083,
+      "▁launching": 30084,
+      "开机": 30085,
+      "lando": 30086,
+      "▁fancy": 30087,
+      "nhow": 30088,
+      "etter": 30089,
+      "nMusic": 30090,
+      "▁contexts": 30091,
+      "▁refriger": 30092,
+      "机床": 30093,
+      "uart": 30094,
+      "自助": 30095,
+      "▁unt": 30096,
+      "obi": 30097,
+      "▁offline": 30098,
+      "▁punct": 30099,
+      "authors": 30100,
+      "Executor": 30101,
+      "▁corners": 30102,
+      "介入": 30103,
+      "▁hardly": 30104,
+      "咳嗽": 30105,
+      "YW": 30106,
+      "{'": 30107,
+      "沉淀": 30108,
+      "[];\\": 30109,
+      "nsup": 30110,
+      "一直是": 30111,
+      "raul": 30112,
+      "▁Tes": 30113,
+      "▁supportive": 30114,
+      "邻居": 30115,
+      "Worker": 30116,
+      "pipeline": 30117,
+      "FROM": 30118,
+      "▁inev": 30119,
+      "右边": 30120,
+      "▁Lex": 30121,
+      "基部": 30122,
+      "▁neutroph": 30123,
+      "Using": 30124,
+      "▁handed": 30125,
+      "Axis": 30126,
+      "enic": 30127,
+      "crypto": 30128,
+      "▁б": 30129,
+      "全区": 30130,
+      "▁uk": 30131,
+      "▁sealed": 30132,
+      ">{\\": 30133,
+      "Place": 30134,
+      "▁psy": 30135,
+      "▁Grey": 30136,
+      "depends": 30137,
+      "▁catheter": 30138,
+      "▁sequential": 30139,
+      "得好": 30140,
+      "secondary": 30141,
+      "进取": 30142,
+      "女孩子": 30143,
+      "Month": 30144,
+      "integration": 30145,
+      "我们会": 30146,
+      "▁Generated": 30147,
+      "▁immunos": 30148,
+      "▁unittest": 30149,
+      "见的": 30150,
+      "kubernetes": 30151,
+      "▁zu": 30152,
+      "bag": 30153,
+      "▁EM": 30154,
+      "Nodes": 30155,
+      "elian": 30156,
+      "▁vulnerability": 30157,
+      "▁),": 30158,
+      "kdir": 30159,
+      "ín": 30160,
+      "指向": 30161,
+      "▁安装": 30162,
+      "在该": 30163,
+      "最有": 30164,
+      "▁单": 30165,
+      "▁Joint": 30166,
+      "码头": 30167,
+      "fare": 30168,
+      "▁Bond": 30169,
+      "Push": 30170,
+      "arus": 30171,
+      "仲裁委员会": 30172,
+      "插槽": 30173,
+      "球场": 30174,
+      "▁Om": 30175,
+      "▁notify": 30176,
+      "Intel": 30177,
+      "ophageal": 30178,
+      "刚才": 30179,
+      "干嘛": 30180,
+      "Len": 30181,
+      "iline": 30182,
+      "论述": 30183,
+      "buy": 30184,
+      "现羁押": 30185,
+      "巴巴": 30186,
+      "ributors": 30187,
+      "Adv": 30188,
+      "▁newspapers": 30189,
+      "▁entertaining": 30190,
+      "vid": 30191,
+      "坚守": 30192,
+      "▁Dragon": 30193,
+      "成型": 30194,
+      "▁Scholar": 30195,
+      "生肖": 30196,
+      "▁Programming": 30197,
+      "▁horn": 30198,
+      "▁stationary": 30199,
+      "}_{\\\\": 30200,
+      "▁obs": 30201,
+      "▁pulling": 30202,
+      "号民事判决书的本次执行程序": 30203,
+      "▁projected": 30204,
+      "丧失": 30205,
+      "与被告离婚": 30206,
+      "▁Murray": 30207,
+      "的无": 30208,
+      "不属于": 30209,
+      "adequ": 30210,
+      "小镇": 30211,
+      "▁twist": 30212,
+      "▁trunc": 30213,
+      "▁Amsterdam": 30214,
+      "▁Door": 30215,
+      "技术服务": 30216,
+      "▁proximal": 30217,
+      "▁许": 30218,
+      "▁sensit": 30219,
+      "▁foam": 30220,
+      "mbed": 30221,
+      "preter": 30222,
+      "▁responding": 30223,
+      "定量": 30224,
+      "▁conclusions": 30225,
+      ")}\\\\": 30226,
+      "▁Brew": 30227,
+      "▁cylind": 30228,
+      "loyd": 30229,
+      "▁inherited": 30230,
+      "行政处罚决定": 30231,
+      "servlet": 30232,
+      "▁谢": 30233,
+      "▁killer": 30234,
+      "次韵": 30235,
+      "足够的": 30236,
+      "中的应用": 30237,
+      "▁Lith": 30238,
+      "responsive": 30239,
+      "▁relatives": 30240,
+      "资深": 30241,
+      "▁NR": 30242,
+      "▁deposits": 30243,
+      "▁Mumbai": 30244,
+      "▁Commerce": 30245,
+      "▁serialize": 30246,
+      "▁Hem": 30247,
+      "emies": 30248,
+      "步伐": 30249,
+      "todo": 30250,
+      "leuk": 30251,
+      "▁locus": 30252,
+      "天才": 30253,
+      "▁Orche": 30254,
+      "Execution": 30255,
+      "西兰": 30256,
+      "▁Glasgow": 30257,
+      "借款合同": 30258,
+      "管理部门": 30259,
+      "你也": 30260,
+      "Cost": 30261,
+      "▁cave": 30262,
+      "▁qui": 30263,
+      "又名": 30264,
+      "年经": 30265,
+      "Insert": 30266,
+      "▁icons": 30267,
+      "▁heated": 30268,
+      "Lo": 30269,
+      "▁picks": 30270,
+      "nSpe": 30271,
+      "▁follic": 30272,
+      "ographer": 30273,
+      "而出": 30274,
+      "▁Diff": 30275,
+      "▁literacy": 30276,
+      "▁\"_": 30277,
+      "▁raises": 30278,
+      "anium": 30279,
+      "▁Armen": 30280,
+      "▁infrared": 30281,
+      "ELL": 30282,
+      "▁DAMAGES": 30283,
+      "交给": 30284,
+      "Setup": 30285,
+      "ede": 30286,
+      "heses": 30287,
+      "Ns": 30288,
+      "争吵": 30289,
+      "▁Ment": 30290,
+      "▁Gay": 30291,
+      "余额": 30292,
+      "Buff": 30293,
+      "▁↑": 30294,
+      "][]": 30295,
+      "chen": 30296,
+      "行为的": 30297,
+      "科技创新": 30298,
+      "狮子": 30299,
+      "▁centr": 30300,
+      "▁sums": 30301,
+      "bm": 30302,
+      "nold": 30303,
+      "▁garlic": 30304,
+      "▁overlap": 30305,
+      "▁duo": 30306,
+      "▁doubles": 30307,
+      "▁unchanged": 30308,
+      "▁poem": 30309,
+      "▁regulator": 30310,
+      "儿的": 30311,
+      "▁Validation": 30312,
+      "行政诉讼": 30313,
+      "▁sufficiently": 30314,
+      "▁fence": 30315,
+      "锅中": 30316,
+      "司法鉴定": 30317,
+      "▁arising": 30318,
+      "▁peaks": 30319,
+      "为核心": 30320,
+      "等方面的": 30321,
+      "支付宝": 30322,
+      "▁Nova": 30323,
+      "越来越多的": 30324,
+      "卡片": 30325,
+      "间接": 30326,
+      "▁simulate": 30327,
+      "nContact": 30328,
+      "spy": 30329,
+      "▁lymphoma": 30330,
+      "SED": 30331,
+      "blasts": 30332,
+      "▁Directory": 30333,
+      "行星": 30334,
+      "的城市": 30335,
+      "▁envelope": 30336,
+      "外卖": 30337,
+      "▁于": 30338,
+      "多吃": 30339,
+      "▁crate": 30340,
+      "▁Geometry": 30341,
+      "rez": 30342,
+      "alchemy": 30343,
+      "▁hydrox": 30344,
+      "▁bottles": 30345,
+      "我来": 30346,
+      "▁breeding": 30347,
+      "▁executives": 30348,
+      "▁Understanding": 30349,
+      "用手": 30350,
+      "是其": 30351,
+      "enue": 30352,
+      "ampton": 30353,
+      "▁hesitate": 30354,
+      "Forms": 30355,
+      "直属": 30356,
+      ")}$": 30357,
+      "icion": 30358,
+      "▁Economics": 30359,
+      "▁NEW": 30360,
+      "期限届满": 30361,
+      "光滑": 30362,
+      "ATAB": 30363,
+      "prints": 30364,
+      "lm": 30365,
+      "▁Rand": 30366,
+      "▁outlined": 30367,
+      "▁companion": 30368,
+      "紫外": 30369,
+      "▁nail": 30370,
+      "▁avoiding": 30371,
+      "开门": 30372,
+      "▁Krist": 30373,
+      "▁Concept": 30374,
+      "▁Lif": 30375,
+      "▁Diamond": 30376,
+      "Express": 30377,
+      "▁Ug": 30378,
+      "北大": 30379,
+      "Accept": 30380,
+      "向本院申请强制执行": 30381,
+      "▁genetics": 30382,
+      "▁pytest": 30383,
+      "▁imply": 30384,
+      "interfaces": 30385,
+      "人民的": 30386,
+      "▁Nat": 30387,
+      "alter": 30388,
+      "pathetic": 30389,
+      "Methods": 30390,
+      "遭受": 30391,
+      "nMen": 30392,
+      "▁Alp": 30393,
+      "statement": 30394,
+      "▁knife": 30395,
+      "考古": 30396,
+      "Exit": 30397,
+      "▁CAN": 30398,
+      "证监会": 30399,
+      "sson": 30400,
+      "▁Blues": 30401,
+      "nEd": 30402,
+      "来进行": 30403,
+      "▁Wa": 30404,
+      "某某的": 30405,
+      "itic": 30406,
+      "个别": 30407,
+      "底层": 30408,
+      "spark": 30409,
+      "专栏": 30410,
+      "▁mb": 30411,
+      "不断地": 30412,
+      "▁Ted": 30413,
+      "▁congress": 30414,
+      "吾尔": 30415,
+      "在我们": 30416,
+      "▁consumed": 30417,
+      "首先要": 30418,
+      "Payment": 30419,
+      "品的": 30420,
+      "推出的": 30421,
+      "eyer": 30422,
+      "▁Mak": 30423,
+      "▁immunohist": 30424,
+      "替换": 30425,
+      "▁wisdom": 30426,
+      "土木": 30427,
+      "过去了": 30428,
+      "▁Nice": 30429,
+      "▁determinant": 30430,
+      "▁reminder": 30431,
+      "oser": 30432,
+      "illus": 30433,
+      "▁enjoys": 30434,
+      "▁autoimmune": 30435,
+      "▁PP": 30436,
+      "jp": 30437,
+      "Execute": 30438,
+      "绝望": 30439,
+      "感觉到": 30440,
+      "益智": 30441,
+      "▁Chrom": 30442,
+      "▁actin": 30443,
+      "仓储": 30444,
+      "▁pathophys": 30445,
+      "▁Tu": 30446,
+      "了起来": 30447,
+      "的分析": 30448,
+      "周知": 30449,
+      "▁IE": 30450,
+      "▁nest": 30451,
+      "▁Manhattan": 30452,
+      "fu": 30453,
+      "▁como": 30454,
+      "Micro": 30455,
+      "回国": 30456,
+      "▁JO": 30457,
+      "▁converting": 30458,
+      "上方": 30459,
+      "感应": 30460,
+      "nOutput": 30461,
+      "每小时": 30462,
+      "▁canal": 30463,
+      "▁Making": 30464,
+      "我就是": 30465,
+      "Will": 30466,
+      "please": 30467,
+      "Namespace": 30468,
+      "▁sanction": 30469,
+      "皇家": 30470,
+      "Chain": 30471,
+      "▁creator": 30472,
+      "热量": 30473,
+      "Pages": 30474,
+      "▁colony": 30475,
+      "cas": 30476,
+      "▁种": 30477,
+      "▁sans": 30478,
+      "▁PG": 30479,
+      "我又": 30480,
+      "tcp": 30481,
+      "oples": 30482,
+      "▁numerator": 30483,
+      "▁Morning": 30484,
+      "▁Brig": 30485,
+      "网游": 30486,
+      "有机会": 30487,
+      "极大的": 30488,
+      "given": 30489,
+      "▁Yellow": 30490,
+      "▁defaults": 30491,
+      "▁patches": 30492,
+      "▁Coach": 30493,
+      "▁POST": 30494,
+      "▁correlate": 30495,
+      ">\",": 30496,
+      "▁Yan": 30497,
+      "▁Bookmark": 30498,
+      "▁polarization": 30499,
+      "第一百四十五条第一款规定": 30500,
+      "ncd": 30501,
+      "ishops": 30502,
+      "为企业": 30503,
+      "▁benign": 30504,
+      "剧院": 30505,
+      "issance": 30506,
+      "公证": 30507,
+      ":]\\": 30508,
+      "不仅是": 30509,
+      "rocery": 30510,
+      "并给出你的": 30511,
+      "的长度": 30512,
+      "▁notifications": 30513,
+      "itual": 30514,
+      "▁Honor": 30515,
+      "▁Lesser": 30516,
+      "▁Getting": 30517,
+      "保定": 30518,
+      "的人员": 30519,
+      "▁Yang": 30520,
+      "▁CONTRACT": 30521,
+      "▁Charlie": 30522,
+      "▁veterans": 30523,
+      "▁complication": 30524,
+      "人不": 30525,
+      "Music": 30526,
+      "▁Opera": 30527,
+      "▁assists": 30528,
+      "▁tendency": 30529,
+      "平静": 30530,
+      "itime": 30531,
+      "▁liberal": 30532,
+      "恶魔": 30533,
+      "density": 30534,
+      "Zero": 30535,
+      "死刑": 30536,
+      "降价": 30537,
+      "HashMap": 30538,
+      "▁workout": 30539,
+      "isch": 30540,
+      "▁FDA": 30541,
+      "号民事判决书已经发生法律效力": 30542,
+      "里有": 30543,
+      "uitar": 30544,
+      "噪声": 30545,
+      "方可": 30546,
+      "TOP": 30547,
+      "借据": 30548,
+      "笑容": 30549,
+      "wy": 30550,
+      "控制器": 30551,
+      "加入中国共产党": 30552,
+      "DIRECT": 30553,
+      "交易所": 30554,
+      "▁automotive": 30555,
+      "▁calibration": 30556,
+      "eway": 30557,
+      "恶劣": 30558,
+      "nUs": 30559,
+      "汉典": 30560,
+      "注定": 30561,
+      "组织部": 30562,
+      "的对": 30563,
+      "维吾尔": 30564,
+      "安心": 30565,
+      "nCurrently": 30566,
+      "tdata": 30567,
+      "催告": 30568,
+      "angan": 30569,
+      "并通过": 30570,
+      "▁experimentally": 30571,
+      "EQ": 30572,
+      "qt": 30573,
+      "Fs": 30574,
+      "usa": 30575,
+      "岁以下的": 30576,
+      "nEdit": 30577,
+      "申请执行人于": 30578,
+      "PHP": 30579,
+      "mina": 30580,
+      "困境": 30581,
+      "▁SHALL": 30582,
+      "▁Masters": 30583,
+      "▁premature": 30584,
+      "浩特": 30585,
+      "rawl": 30586,
+      "▁nuclei": 30587,
+      "▁redund": 30588,
+      "▁afterwards": 30589,
+      "▁Cold": 30590,
+      "▁succ": 30591,
+      "之时": 30592,
+      "▁Brow": 30593,
+      "▁editors": 30594,
+      "ля": 30595,
+      "▁Baker": 30596,
+      "▁Bass": 30597,
+      "reason": 30598,
+      "Timestamp": 30599,
+      "玩笑": 30600,
+      "▁wise": 30601,
+      "业的": 30602,
+      "售后服务": 30603,
+      "▁notably": 30604,
+      "▁Kil": 30605,
+      "▁pharmacological": 30606,
+      "和国家": 30607,
+      "PLAY": 30608,
+      "▁nicot": 30609,
+      "Distance": 30610,
+      "检索": 30611,
+      "destination": 30612,
+      "nDep": 30613,
+      "之后的": 30614,
+      "▁instructor": 30615,
+      "有无": 30616,
+      "▁generous": 30617,
+      "▁diox": 30618,
+      "月初": 30619,
+      "▁renewable": 30620,
+      "▁trap": 30621,
+      "▁rewards": 30622,
+      "调度": 30623,
+      "prim": 30624,
+      "▁soy": 30625,
+      "PREFIX": 30626,
+      "Dictionary": 30627,
+      "etta": 30628,
+      "莫名": 30629,
+      "othe": 30630,
+      "▁Train": 30631,
+      "▁til": 30632,
+      "affold": 30633,
+      "主板": 30634,
+      "Wi": 30635,
+      "盛世": 30636,
+      "的方向": 30637,
+      "▁Molecular": 30638,
+      "▁Und": 30639,
+      "商贸有限公司": 30640,
+      "indexOf": 30641,
+      "▁Driver": 30642,
+      "▁violation": 30643,
+      "温和": 30644,
+      "产后": 30645,
+      "永恒": 30646,
+      "▁subproject": 30647,
+      "raits": 30648,
+      "nTH": 30649,
+      "Office": 30650,
+      "更快": 30651,
+      "▁perox": 30652,
+      "▁quoted": 30653,
+      "随之": 30654,
+      "内科": 30655,
+      "zyme": 30656,
+      "▁Projects": 30657,
+      "恐怕": 30658,
+      "负债": 30659,
+      "奶茶": 30660,
+      "十六章": 30661,
+      "▁Panel": 30662,
+      "给出的选项中": 30663,
+      "▁strive": 30664,
+      "法律文书确定的义务": 30665,
+      "坚信": 30666,
+      "ATABASE": 30667,
+      "::\\": 30668,
+      "化成": 30669,
+      "▁rigid": 30670,
+      "也很好": 30671,
+      "▁Carter": 30672,
+      "最少": 30673,
+      "ifdef": 30674,
+      "rab": 30675,
+      "ountain": 30676,
+      "的人才": 30677,
+      "▁inspiring": 30678,
+      "▁formatting": 30679,
+      "▁destinations": 30680,
+      "热血": 30681,
+      "走在": 30682,
+      "enario": 30683,
+      "nNorth": 30684,
+      "东海": 30685,
+      "weather": 30686,
+      "大胆": 30687,
+      "▁argv": 30688,
+      "▁Brain": 30689,
+      "▁neglig": 30690,
+      "脚步": 30691,
+      "▁screened": 30692,
+      "简约": 30693,
+      "做出了": 30694,
+      "eler": 30695,
+      "▁jer": 30696,
+      "▁owns": 30697,
+      "纪检": 30698,
+      "案外人": 30699,
+      "Repo": 30700,
+      "感知": 30701,
+      "Batch": 30702,
+      "不容": 30703,
+      "▁COMP": 30704,
+      "asting": 30705,
+      "▁Statistical": 30706,
+      "▁carbohyd": 30707,
+      "▁fi": 30708,
+      "ewater": 30709,
+      "represent": 30710,
+      "▁Marshall": 30711,
+      "实物": 30712,
+      "停车场": 30713,
+      "第一百零二条": 30714,
+      "rong": 30715,
+      "狗狗": 30716,
+      "▁potassium": 30717,
+      "/@": 30718,
+      "场比赛": 30719,
+      "asma": 30720,
+      "nMedalists": 30721,
+      "▁PER": 30722,
+      "▁hunt": 30723,
+      "▁Choice": 30724,
+      "▁anthrop": 30725,
+      "▁attacked": 30726,
+      "可在裁定书送达之日起十日内": 30727,
+      "力求": 30728,
+      "难得": 30729,
+      "nNor": 30730,
+      "银行卡": 30731,
+      "▁Historical": 30732,
+      "▁dealer": 30733,
+      "意义的": 30734,
+      "▁Raw": 30735,
+      "▁challenged": 30736,
+      "▁interacting": 30737,
+      "XP": 30738,
+      "九年": 30739,
+      "缴费": 30740,
+      "互为": 30741,
+      "我会给你": 30742,
+      "▁wherever": 30743,
+      "wen": 30744,
+      "▁Iranian": 30745,
+      "hours": 30746,
+      "▁sorting": 30747,
+      "▁severely": 30748,
+      "鼻子": 30749,
+      "▁primitive": 30750,
+      "因为他": 30751,
+      "▁uid": 30752,
+      "也都": 30753,
+      "只能说": 30754,
+      "tstatic": 30755,
+      "遇到了": 30756,
+      "▁herb": 30757,
+      "▁clinicians": 30758,
+      "rise": 30759,
+      "rah": 30760,
+      "▁Arabia": 30761,
+      "missing": 30762,
+      "中考": 30763,
+      "所致": 30764,
+      "asonic": 30765,
+      "修建": 30766,
+      "filepath": 30767,
+      "培训班": 30768,
+      "Alert": 30769,
+      "打包": 30770,
+      "▁Barbara": 30771,
+      "轴对称": 30772,
+      "▁enhances": 30773,
+      "历任": 30774,
+      "包裹": 30775,
+      "环球": 30776,
+      "income": 30777,
+      "皇后": 30778,
+      "erset": 30779,
+      "二〇一三年": 30780,
+      "▁inspect": 30781,
+      "很多的": 30782,
+      "achy": 30783,
+      "买卖合同纠纷一案中": 30784,
+      "▁Interesting": 30785,
+      "日生育": 30786,
+      "lua": 30787,
+      "贵族": 30788,
+      "处分自己的": 30789,
+      "转为": 30790,
+      "▁necrosis": 30791,
+      "要用": 30792,
+      ":$": 30793,
+      "抱怨": 30794,
+      "sten": 30795,
+      "▁robots": 30796,
+      "rivers": 30797,
+      "▁Iter": 30798,
+      "真人": 30799,
+      "▁VE": 30800,
+      "交纳办法": 30801,
+      "quo": 30802,
+      "科院": 30803,
+      "alled": 30804,
+      "ifice": 30805,
+      "▁condens": 30806,
+      "▁employing": 30807,
+      "▁decorated": 30808,
+      "最小的": 30809,
+      "▁Fat": 30810,
+      "ipes": 30811,
+      "ftime": 30812,
+      "▁Magn": 30813,
+      "▁OH": 30814,
+      "及以上": 30815,
+      "的事实清楚": 30816,
+      "▁Leban": 30817,
+      "▁travels": 30818,
+      "sche": 30819,
+      "Li": 30820,
+      "ICES": 30821,
+      "▁togg": 30822,
+      "▁definite": 30823,
+      "热闹": 30824,
+      "邯郸": 30825,
+      "Fetch": 30826,
+      "认定事实": 30827,
+      "▁Forces": 30828,
+      "莲花": 30829,
+      "饲养": 30830,
+      "学金": 30831,
+      "实事": 30832,
+      "这几": 30833,
+      "mbox": 30834,
+      "领导干部": 30835,
+      "exclude": 30836,
+      "芙蓉": 30837,
+      "ões": 30838,
+      "▁accepting": 30839,
+      "你认为正确的答案": 30840,
+      "▁Ple": 30841,
+      "开发者": 30842,
+      "division": 30843,
+      "▁Chemical": 30844,
+      "▁Tak": 30845,
+      "celand": 30846,
+      "▁tourists": 30847,
+      "idade": 30848,
+      "西瓜": 30849,
+      "usters": 30850,
+      "▁UnityEngine": 30851,
+      "传动": 30852,
+      "movie": 30853,
+      "▁quantify": 30854,
+      "同类项": 30855,
+      "SOURCE": 30856,
+      "park": 30857,
+      "▁Romania": 30858,
+      "▁Argument": 30859,
+      "▁FBI": 30860,
+      "货运": 30861,
+      "geo": 30862,
+      "进行审理": 30863,
+      "▁urgent": 30864,
+      "▁graduates": 30865,
+      "红军": 30866,
+      "▁Jazz": 30867,
+      "▁HS": 30868,
+      "开创": 30869,
+      "显示器": 30870,
+      "▁pressing": 30871,
+      "▁qualities": 30872,
+      "接下来的": 30873,
+      "合同履行地": 30874,
+      "▁outlet": 30875,
+      "▁Progress": 30876,
+      "pts": 30877,
+      "osition": 30878,
+      "▁UC": 30879,
+      "▁|||": 30880,
+      "一千": 30881,
+      "proved": 30882,
+      "powered": 30883,
+      "evens": 30884,
+      "dao": 30885,
+      "Bro": 30886,
+      "▁应用": 30887,
+      "loquent": 30888,
+      "▁priorit": 30889,
+      "▁rigorous": 30890,
+      "▁lobby": 30891,
+      "▁Present": 30892,
+      "▁microbial": 30893,
+      "分区": 30894,
+      "▁pools": 30895,
+      "▁Interior": 30896,
+      "长方形": 30897,
+      "出处": 30898,
+      "受损": 30899,
+      "整形": 30900,
+      "经济开发区": 30901,
+      "▁Justin": 30902,
+      "▁unlimited": 30903,
+      "addEventListener": 30904,
+      "ully": 30905,
+      "uria": 30906,
+      "▁gli": 30907,
+      "大哥": 30908,
+      "省市": 30909,
+      "rac": 30910,
+      "nPrevious": 30911,
+      "▁É": 30912,
+      "FOR": 30913,
+      "最小值": 30914,
+      "kel": 30915,
+      "▁evolutionary": 30916,
+      "bye": 30917,
+      "▁Leader": 30918,
+      "▁distant": 30919,
+      "▁manufacture": 30920,
+      "主播": 30921,
+      "gluc": 30922,
+      "offer": 30923,
+      "县农村信用合作联社": 30924,
+      "▁reporters": 30925,
+      "▁irregular": 30926,
+      "bu": 30927,
+      "的结构": 30928,
+      "▁considerations": 30929,
+      "▁electromagnetic": 30930,
+      "庆市": 30931,
+      "Green": 30932,
+      "▁hood": 30933,
+      "▁straw": 30934,
+      "奶粉": 30935,
+      "dst": 30936,
+      "阅读网": 30937,
+      "Room": 30938,
+      "▁问题": 30939,
+      "第一项": 30940,
+      "itoneal": 30941,
+      "His": 30942,
+      "可持续发展": 30943,
+      "▁房型": 30944,
+      "▁Carr": 30945,
+      "THE": 30946,
+      "▁RV": 30947,
+      "▁photon": 30948,
+      "electron": 30949,
+      "▁og": 30950,
+      "tsuper": 30951,
+      "DVD": 30952,
+      "▁Yeah": 30953,
+      "体裁": 30954,
+      "angi": 30955,
+      "▁crossed": 30956,
+      "▁invested": 30957,
+      "▁ambient": 30958,
+      "Js": 30959,
+      "三百": 30960,
+      "▁Generation": 30961,
+      "▁NK": 30962,
+      "ptember": 30963,
+      "摄入": 30964,
+      "djang": 30965,
+      "▁Territ": 30966,
+      "ANK": 30967,
+      "君子": 30968,
+      "▁voices": 30969,
+      "▁Extension": 30970,
+      "▁autonomous": 30971,
+      "是的": 30972,
+      "credentials": 30973,
+      "静态": 30974,
+      "▁Yorkshire": 30975,
+      "Tile": 30976,
+      "positions": 30977,
+      "姿势": 30978,
+      "threshold": 30979,
+      "nOBJECTIVE": 30980,
+      "staticmethod": 30981,
+      "Packet": 30982,
+      "NECTION": 30983,
+      "▁dock": 30984,
+      "倾斜": 30985,
+      "普通话": 30986,
+      "abspath": 30987,
+      "▁reserves": 30988,
+      "▁DS": 30989,
+      "公子": 30990,
+      "▁Kate": 30991,
+      "▁Salv": 30992,
+      "金刚": 30993,
+      "umper": 30994,
+      "▁pray": 30995,
+      "但也": 30996,
+      "ovirus": 30997,
+      "),(": 30998,
+      "▁citizen": 30999,
+      "▁chloride": 31000,
+      "该怎么": 31001,
+      "▁shield": 31002,
+      "水市": 31003,
+      "equalities": 31004,
+      "▁Riemann": 31005,
+      "添香": 31006,
+      "MAIN": 31007,
+      "▁Reviews": 31008,
+      "▁nucleotide": 31009,
+      "lookup": 31010,
+      "其间": 31011,
+      "▁alias": 31012,
+      "gene": 31013,
+      "▁Ton": 31014,
+      "按时": 31015,
+      "▁足球": 31016,
+      "开的": 31017,
+      "▁rif": 31018,
+      "People": 31019,
+      "神仙": 31020,
+      "街道办事处": 31021,
+      "日照": 31022,
+      "Tok": 31023,
+      "▁Begin": 31024,
+      "▁Position": 31025,
+      "▁turnover": 31026,
+      "▁Structure": 31027,
+      "concurrent": 31028,
+      "▁seemingly": 31029,
+      "王朝": 31030,
+      "adows": 31031,
+      "techn": 31032,
+      "coding": 31033,
+      "▁cancelled": 31034,
+      "nCreate": 31035,
+      "▁nutrients": 31036,
+      "华东": 31037,
+      "保全措施": 31038,
+      "数控": 31039,
+      "千万不要": 31040,
+      "▁SY": 31041,
+      "▁lenses": 31042,
+      "出道": 31043,
+      "否认": 31044,
+      "estrian": 31045,
+      "▁eslint": 31046,
+      "童年": 31047,
+      "▁Jimmy": 31048,
+      "dto": 31049,
+      "了不少": 31050,
+      "nWrit": 31051,
+      "▁ont": 31052,
+      "▁CK": 31053,
+      "journal": 31054,
+      "fake": 31055,
+      "finals": 31056,
+      "差不": 31057,
+      "▁banned": 31058,
+      "对不起": 31059,
+      "相对于": 31060,
+      "erialize": 31061,
+      "▁ba": 31062,
+      "垄断": 31063,
+      "公交车": 31064,
+      "▁ought": 31065,
+      "Encoder": 31066,
+      "民终": 31067,
+      "▁Roger": 31068,
+      "nPres": 31069,
+      "▁rotating": 31070,
+      "▁pronounced": 31071,
+      "▁genetically": 31072,
+      "不着": 31073,
+      "那边": 31074,
+      "Sem": 31075,
+      "▁若": 31076,
+      "hit": 31077,
+      "你们的": 31078,
+      "发布会": 31079,
+      "▁WHETHER": 31080,
+      "Condition": 31081,
+      "样式": 31082,
+      "▁Always": 31083,
+      "▁barely": 31084,
+      "▁sporting": 31085,
+      "▁stub": 31086,
+      "▁beans": 31087,
+      "▁dairy": 31088,
+      "▁Define": 31089,
+      "▁cooper": 31090,
+      "uta": 31091,
+      "▁pants": 31092,
+      "▁divisions": 31093,
+      "▁enjoyable": 31094,
+      "QUI": 31095,
+      "counts": 31096,
+      "▁incorporating": 31097,
+      "遵守监规": 31098,
+      "▁Oracle": 31099,
+      "是否准许": 31100,
+      "compatible": 31101,
+      "logout": 31102,
+      "transl": 31103,
+      "并承担": 31104,
+      "▁Ru": 31105,
+      "未果": 31106,
+      "▁Champion": 31107,
+      "经本院传票传唤": 31108,
+      "▁mt": 31109,
+      "▁harmon": 31110,
+      "▁Billboard": 31111,
+      "nSte": 31112,
+      "湖区": 31113,
+      "ospel": 31114,
+      "viewer": 31115,
+      "▁ensemble": 31116,
+      "范畴": 31117,
+      "lest": 31118,
+      "普通程序": 31119,
+      "▁priced": 31120,
+      "忠诚": 31121,
+      "▁jaw": 31122,
+      "ricane": 31123,
+      "将自己": 31124,
+      "club": 31125,
+      "▁atmospheric": 31126,
+      "ostat": 31127,
+      "grams": 31128,
+      "started": 31129,
+      "▁都市": 31130,
+      "▁stolen": 31131,
+      "绝缘": 31132,
+      "也比较": 31133,
+      "▁Face": 31134,
+      "▁Wheel": 31135,
+      "▁suits": 31136,
+      "nClass": 31137,
+      "preced": 31138,
+      "珍珠": 31139,
+      "▁opioid": 31140,
+      "柴油": 31141,
+      "▁schizophrenia": 31142,
+      "▁TS": 31143,
+      "底面": 31144,
+      "nIS": 31145,
+      "riched": 31146,
+      "pow": 31147,
+      "Could": 31148,
+      "Study": 31149,
+      "\"},\"": 31150,
+      "Cancel": 31151,
+      "修炼": 31152,
+      "生命的": 31153,
+      "▁Clar": 31154,
+      "FW": 31155,
+      "给付原告": 31156,
+      "推出了": 31157,
+      "anine": 31158,
+      "▁Console": 31159,
+      "iciencies": 31160,
+      "的情形": 31161,
+      "▁rack": 31162,
+      "[],": 31163,
+      "所以说": 31164,
+      "控制系统": 31165,
+      "Bottom": 31166,
+      "纳金": 31167,
+      "租房": 31168,
+      "讲述了": 31169,
+      "odic": 31170,
+      "经济效益": 31171,
+      "▁anytime": 31172,
+      "÷（": 31173,
+      "头上": 31174,
+      "▁seating": 31175,
+      "sz": 31176,
+      "入侵": 31177,
+      "▁mud": 31178,
+      "▁sediment": 31179,
+      "▁deficit": 31180,
+      "汇率": 31181,
+      "披露": 31182,
+      "时间和": 31183,
+      "measure": 31184,
+      "▁desirable": 31185,
+      "优美": 31186,
+      "提供证据": 31187,
+      "▁additionally": 31188,
+      "▁ears": 31189,
+      "lead": 31190,
+      "birth": 31191,
+      "年年": 31192,
+      "nIts": 31193,
+      "几百": 31194,
+      "dal": 31195,
+      "▁Coffee": 31196,
+      "▁genotype": 31197,
+      "ahan": 31198,
+      "▁Kennedy": 31199,
+      "girl": 31200,
+      "▁Rico": 31201,
+      "EB": 31202,
+      "原点": 31203,
+      "平台的": 31204,
+      "Reset": 31205,
+      "眼光": 31206,
+      "月亮": 31207,
+      "发布了": 31208,
+      "egu": 31209,
+      "表现出": 31210,
+      "保险公司": 31211,
+      "的那种": 31212,
+      "认为是": 31213,
+      "的名字": 31214,
+      "▁Ian": 31215,
+      "▁Nepal": 31216,
+      "comed": 31217,
+      "▁besides": 31218,
+      "▁priorities": 31219,
+      "pn": 31220,
+      "实在是": 31221,
+      "ocused": 31222,
+      "Ni": 31223,
+      "化和": 31224,
+      "▁五": 31225,
+      "光电": 31226,
+      "▁sul": 31227,
+      "出具借条": 31228,
+      "▁memorial": 31229,
+      "Sam": 31230,
+      "钓鱼": 31231,
+      "▁Mg": 31232,
+      "Tunes": 31233,
+      "▁Craft": 31234,
+      "看上去": 31235,
+      "▁Own": 31236,
+      "▁Doll": 31237,
+      "▁spacing": 31238,
+      "▁trajectory": 31239,
+      "进率": 31240,
+      "▁planes": 31241,
+      "▁continuity": 31242,
+      "▁yarn": 31243,
+      "▁Carlos": 31244,
+      "求职": 31245,
+      "评定": 31246,
+      "第一百六": 31247,
+      "▁Calculus": 31248,
+      "fw": 31249,
+      "\\'^": 31250,
+      "正整数": 31251,
+      "▁ecological": 31252,
+      "▁technicians": 31253,
+      "歧视": 31254,
+      "▁Background": 31255,
+      "gas": 31256,
+      "▁Pok": 31257,
+      "▁dated": 31258,
+      "▁adapter": 31259,
+      "丢失": 31260,
+      "左侧": 31261,
+      "各方": 31262,
+      "\":[{\"": 31263,
+      "答应": 31264,
+      "Hot": 31265,
+      "psych": 31266,
+      "▁Algorithm": 31267,
+      "poll": 31268,
+      "▁reflex": 31269,
+      "日向本院申请强制执行": 31270,
+      "比起": 31271,
+      "iae": 31272,
+      "legend": 31273,
+      "冷冻": 31274,
+      "行李": 31275,
+      "适当的": 31276,
+      "orch": 31277,
+      "现羁押于": 31278,
+      "▁\";\\": 31279,
+      "▁fallen": 31280,
+      "选拔": 31281,
+      "难受": 31282,
+      "\":\"<": 31283,
+      "▁Chand": 31284,
+      "一面": 31285,
+      "脱贫攻坚": 31286,
+      "setminus": 31287,
+      "次的": 31288,
+      "amd": 31289,
+      "▁Sha": 31290,
+      "codec": 31291,
+      "强行": 31292,
+      "▁Patri": 31293,
+      "中华人民共和国民法通则": 31294,
+      "料酒": 31295,
+      "▁auf": 31296,
+      "▁LIABLE": 31297,
+      "permissions": 31298,
+      "Ds": 31299,
+      "▁NHL": 31300,
+      "▁corrected": 31301,
+      "▁Cruz": 31302,
+      "▁tongue": 31303,
+      "自收到": 31304,
+      "制约": 31305,
+      "▁origins": 31306,
+      "日向": 31307,
+      "该公司经理": 31308,
+      "平原": 31309,
+      "▁GH": 31310,
+      "▁Protocol": 31311,
+      "汇总": 31312,
+      "illin": 31313,
+      "\"(": 31314,
+      "二维码": 31315,
+      "不断提高": 31316,
+      "Alter": 31317,
+      "▁preservation": 31318,
+      "▁Brothers": 31319,
+      "大理": 31320,
+      "glut": 31321,
+      "▁水": 31322,
+      "▁regret": 31323,
+      "▁bubble": 31324,
+      "MESS": 31325,
+      "▁Treat": 31326,
+      "▁Vision": 31327,
+      "▁customize": 31328,
+      "reflect": 31329,
+      "▁congenital": 31330,
+      "▁asp": 31331,
+      "esters": 31332,
+      "documentation": 31333,
+      "odont": 31334,
+      "▁Hungary": 31335,
+      "空格": 31336,
+      "▁precursor": 31337,
+      "hew": 31338,
+      "九龙": 31339,
+      "为自己": 31340,
+      "▁Senator": 31341,
+      "▁jack": 31342,
+      "成一个": 31343,
+      "▁clever": 31344,
+      "▁cinema": 31345,
+      "甲状": 31346,
+      "▁prospects": 31347,
+      "▁TM": 31348,
+      "哈尔滨市": 31349,
+      "又要": 31350,
+      "▁INTER": 31351,
+      "▁Proof": 31352,
+      "其他可供执行的财产": 31353,
+      "▁fu": 31354,
+      "WORK": 31355,
+      "▁incon": 31356,
+      "核桃": 31357,
+      "DEX": 31358,
+      "▁trainees": 31359,
+      "▁Instit": 31360,
+      "一夜": 31361,
+      "▁Carm": 31362,
+      "工序": 31363,
+      "逾期不": 31364,
+      "▁reproduction": 31365,
+      "应承担": 31366,
+      "目前的": 31367,
+      "STATUS": 31368,
+      "▁outdoors": 31369,
+      "管理体系": 31370,
+      "灌溉": 31371,
+      "风味": 31372,
+      "他也": 31373,
+      "▁pseudo": 31374,
+      "imo": 31375,
+      "在网上": 31376,
+      "▁Victorian": 31377,
+      "swap": 31378,
+      "▁dispers": 31379,
+      "▁cumulative": 31380,
+      "来临": 31381,
+      "jab": 31382,
+      "宁夏回": 31383,
+      "Implemented": 31384,
+      "娃娃": 31385,
+      "蒸汽": 31386,
+      "机理": 31387,
+      "眼中": 31388,
+      "▁wiki": 31389,
+      "▁athlete": 31390,
+      "tin": 31391,
+      "再度": 31392,
+      "▁copied": 31393,
+      "▁Camera": 31394,
+      "科技进步": 31395,
+      "扬州": 31396,
+      "票据": 31397,
+      "ousel": 31398,
+      "▁Beta": 31399,
+      "▁parsed": 31400,
+      "ем": 31401,
+      "Fact": 31402,
+      "▁Mathematic": 31403,
+      "NBA": 31404,
+      "▁Ridge": 31405,
+      "▁comply": 31406,
+      "手表": 31407,
+      "SDK": 31408,
+      "原审法院": 31409,
+      "▁riders": 31410,
+      "等式": 31411,
+      "▁cups": 31412,
+      "agination": 31413,
+      "umble": 31414,
+      "猫咪": 31415,
+      "▁expects": 31416,
+      "弟子": 31417,
+      "Export": 31418,
+      "▁intimate": 31419,
+      "▁electrode": 31420,
+      "兼职": 31421,
+      "seek": 31422,
+      "▁brew": 31423,
+      "▁Ve": 31424,
+      "可以从": 31425,
+      "nac": 31426,
+      "▁Coal": 31427,
+      "辞职": 31428,
+      "ntime": 31429,
+      "thy": 31430,
+      "nett": 31431,
+      "▁resign": 31432,
+      "renderer": 31433,
+      "isb": 31434,
+      "▁offerings": 31435,
+      "前款": 31436,
+      "Business": 31437,
+      "充满了": 31438,
+      "Commun": 31439,
+      "phil": 31440,
+      "inja": 31441,
+      "▁educators": 31442,
+      "▁逝世": 31443,
+      "可靠性": 31444,
+      "nOnly": 31445,
+      "▁toll": 31446,
+      "▁Diagn": 31447,
+      "rgb": 31448,
+      "Clear": 31449,
+      "ensin": 31450,
+      "共识": 31451,
+      "太极": 31452,
+      "▁TORT": 31453,
+      "▁门": 31454,
+      "eling": 31455,
+      "▁breed": 31456,
+      "deb": 31457,
+      "backup": 31458,
+      "事儿": 31459,
+      "ELD": 31460,
+      "项公式": 31461,
+      "▁Tol": 31462,
+      "南市": 31463,
+      "比特": 31464,
+      "\":[\"": 31465,
+      "邮政储蓄": 31466,
+      "▁proving": 31467,
+      "▁Freedom": 31468,
+      "▁uniformly": 31469,
+      "nScott": 31470,
+      "▁fixing": 31471,
+      "Bank": 31472,
+      "▁Sof": 31473,
+      "顺便": 31474,
+      "ril": 31475,
+      "olysis": 31476,
+      "▁dozens": 31477,
+      "▁markedly": 31478,
+      "DN": 31479,
+      "▁Medicare": 31480,
+      "opacity": 31481,
+      "ERN": 31482,
+      "▁rolled": 31483,
+      "会让": 31484,
+      "党史": 31485,
+      "▁alphabet": 31486,
+      "▁transparency": 31487,
+      "简便": 31488,
+      "▁occupational": 31489,
+      "▁\"*": 31490,
+      "▁printf": 31491,
+      "签订的": 31492,
+      "elo": 31493,
+      "obar": 31494,
+      "shirt": 31495,
+      "勇士": 31496,
+      "口号": 31497,
+      "▁ARISING": 31498,
+      "妙的": 31499,
+      "▁tran": 31500,
+      "▁uuid": 31501,
+      "▁entrepreneurs": 31502,
+      "连云": 31503,
+      "nnpm": 31504,
+      "▁surgeon": 31505,
+      "▁notebook": 31506,
+      "died": 31507,
+      "▁dying": 31508,
+      "▁advocate": 31509,
+      "关税": 31510,
+      "风情": 31511,
+      "▁GPU": 31512,
+      "宁夏回族自治区": 31513,
+      "workspace": 31514,
+      "▁invitation": 31515,
+      "▁龙": 31516,
+      "之人": 31517,
+      "▁IGF": 31518,
+      "▁tremendous": 31519,
+      "Scan": 31520,
+      "iplinary": 31521,
+      "众人": 31522,
+      "▁SPE": 31523,
+      "骄傲": 31524,
+      "▁CLA": 31525,
+      "▁Fig": 31526,
+      "curr": 31527,
+      "▁其四": 31528,
+      "日受理后": 31529,
+      "Owner": 31530,
+      "▁tablets": 31531,
+      "▁elliptic": 31532,
+      "▁endorse": 31533,
+      "看书": 31534,
+      "lst": 31535,
+      "伤心": 31536,
+      "袖添香": 31537,
+      "middle": 31538,
+      "增速": 31539,
+      "ativ": 31540,
+      "ulators": 31541,
+      "▁memorable": 31542,
+      "烹饪": 31543,
+      "有多种": 31544,
+      "▁inert": 31545,
+      "ocs": 31546,
+      "▁材料": 31547,
+      "latex": 31548,
+      "实训": 31549,
+      "quisite": 31550,
+      "▁tumour": 31551,
+      "本届": 31552,
+      "翻炒": 31553,
+      "nCould": 31554,
+      "腐败": 31555,
+      "具体情况": 31556,
+      "光伏": 31557,
+      "媳妇": 31558,
+      "manifest": 31559,
+      "院士": 31560,
+      "Children": 31561,
+      "etheless": 31562,
+      "▁forwards": 31563,
+      "▁cleaned": 31564,
+      "▁stupid": 31565,
+      "▁activists": 31566,
+      "▁concentrated": 31567,
+      "国民党": 31568,
+      "演奏": 31569,
+      "oping": 31570,
+      "airo": 31571,
+      "▁nomine": 31572,
+      "银河": 31573,
+      "controllers": 31574,
+      "▁counseling": 31575,
+      "三千": 31576,
+      "syntax": 31577,
+      "Student": 31578,
+      "中关村": 31579,
+      "的取值范围是": 31580,
+      "▁endless": 31581,
+      "款式": 31582,
+      "logen": 31583,
+      "CU": 31584,
+      "kb": 31585,
+      "▁retreat": 31586,
+      "坚强": 31587,
+      "德市": 31588,
+      "辅料": 31589,
+      "前沿": 31590,
+      "▁Horn": 31591,
+      "ceedings": 31592,
+      "▁attorneys": 31593,
+      "我能": 31594,
+      "下属": 31595,
+      "廉政": 31596,
+      "对称轴": 31597,
+      "查明被执行人": 31598,
+      "▁grammar": 31599,
+      "Effects": 31600,
+      "nparser": 31601,
+      "▁celebrating": 31602,
+      "科大学": 31603,
+      "fh": 31604,
+      "百家": 31605,
+      "▁cu": 31606,
+      "▁Collins": 31607,
+      "▁allegations": 31608,
+      "Monitor": 31609,
+      "nResearch": 31610,
+      "五大": 31611,
+      "沿海": 31612,
+      "法和": 31613,
+      "▁Applied": 31614,
+      "▁portrait": 31615,
+      "▁allocated": 31616,
+      "各方面": 31617,
+      "▁farming": 31618,
+      "太子": 31619,
+      "把你": 31620,
+      "Bool": 31621,
+      "nYeah": 31622,
+      "▁IEEE": 31623,
+      "当作": 31624,
+      "花卉": 31625,
+      "▁Vermont": 31626,
+      "采信": 31627,
+      "ensitive": 31628,
+      "illation": 31629,
+      "▁VI": 31630,
+      "▁Header": 31631,
+      "联动": 31632,
+      "▁petition": 31633,
+      "的成绩": 31634,
+      "香蕉": 31635,
+      "'];\\": 31636,
+      "pyplot": 31637,
+      "▁Cricket": 31638,
+      "▁excitement": 31639,
+      "吸烟": 31640,
+      "通话": 31641,
+      "▁Griff": 31642,
+      "▁incoming": 31643,
+      "nEm": 31644,
+      "▁organised": 31645,
+      "夕阳": 31646,
+      "的话题": 31647,
+      "▁compromise": 31648,
+      "衡量": 31649,
+      "aptive": 31650,
+      "把他": 31651,
+      "通过了": 31652,
+      "▁orbital": 31653,
+      "奇幻": 31654,
+      "绘制": 31655,
+      "▁gig": 31656,
+      "▁Earl": 31657,
+      "▁getattr": 31658,
+      "▁delighted": 31659,
+      "pert": 31660,
+      "或者其他": 31661,
+      "▁utilities": 31662,
+      "自有": 31663,
+      "两个月": 31664,
+      "▁tex": 31665,
+      "nq": 31666,
+      "▁microm": 31667,
+      "▁pigs": 31668,
+      "▁ot": 31669,
+      "ingo": 31670,
+      "▁blame": 31671,
+      "▁artifacts": 31672,
+      "▁opponent": 31673,
+      "▁stdout": 31674,
+      "▁bom": 31675,
+      "通俗": 31676,
+      "▁Fan": 31677,
+      "awesome": 31678,
+      "庭审中": 31679,
+      "specially": 31680,
+      "▁highlighting": 31681,
+      "▁impress": 31682,
+      "▁renewal": 31683,
+      "头脑": 31684,
+      "▁Types": 31685,
+      "orious": 31686,
+      "▁achievements": 31687,
+      "▁Cultural": 31688,
+      "周转": 31689,
+      "出一个": 31690,
+      "artifact": 31691,
+      "▁arteries": 31692,
+      "▁plaque": 31693,
+      "▁mysql": 31694,
+      "▁Vi": 31695,
+      "LINK": 31696,
+      "reas": 31697,
+      "isors": 31698,
+      "▁Voice": 31699,
+      "nAnt": 31700,
+      "nTry": 31701,
+      "▁Schema": 31702,
+      "▁festivals": 31703,
+      "▁创作": 31704,
+      "▁zeros": 31705,
+      "▁resc": 31706,
+      "▁MySQL": 31707,
+      "▁Certificate": 31708,
+      "▁ascending": 31709,
+      "投稿": 31710,
+      "▁tv": 31711,
+      "▁dough": 31712,
+      "▁endomet": 31713,
+      "▁全": 31714,
+      "巧克力": 31715,
+      "研究会": 31716,
+      "Under": 31717,
+      "nWelcome": 31718,
+      "▁packing": 31719,
+      "▁Integration": 31720,
+      "inology": 31721,
+      "▁currents": 31722,
+      "urpose": 31723,
+      "▁Rou": 31724,
+      "必要条件": 31725,
+      "▁setUp": 31726,
+      "子公司": 31727,
+      "滞纳金": 31728,
+      "▁repos": 31729,
+      "回首": 31730,
+      "通行": 31731,
+      "▁\\\\\\\\\\": 31732,
+      "san": 31733,
+      "工地": 31734,
+      "▁Dubai": 31735,
+      "是一位": 31736,
+      "▁Craig": 31737,
+      "▁subjective": 31738,
+      "▁informative": 31739,
+      "非常好": 31740,
+      "▁jacket": 31741,
+      "obil": 31742,
+      "变的": 31743,
+      "右侧": 31744,
+      "▁sig": 31745,
+      "fulness": 31746,
+      "▁hydroly": 31747,
+      "▁bite": 31748,
+      "▁parsing": 31749,
+      "▁solvent": 31750,
+      "向往": 31751,
+      "当时的": 31752,
+      "钢筋": 31753,
+      "▁descriptive": 31754,
+      "urtle": 31755,
+      "不幸": 31756,
+      "▁ln": 31757,
+      "ENER": 31758,
+      "诉讼费用交纳办法": 31759,
+      "nTelevision": 31760,
+      "▁probabilities": 31761,
+      "书院": 31762,
+      "机型": 31763,
+      "▁scanner": 31764,
+      "伤口": 31765,
+      "候选": 31766,
+      "ozyg": 31767,
+      "县政府": 31768,
+      "等方式": 31769,
+      "县人民政府": 31770,
+      "基督": 31771,
+      "本品": 31772,
+      "CHECK": 31773,
+      "▁missions": 31774,
+      "Testing": 31775,
+      "常数": 31776,
+      "钱的": 31777,
+      "▁approached": 31778,
+      "语句": 31779,
+      "▁saves": 31780,
+      "是通过": 31781,
+      "的大型": 31782,
+      "山路": 31783,
+      "}\",": 31784,
+      "arten": 31785,
+      "founder": 31786,
+      "▁Gard": 31787,
+      "▁Stanley": 31788,
+      "Ig": 31789,
+      "▁LGBT": 31790,
+      "最优": 31791,
+      "退役": 31792,
+      "十七章": 31793,
+      "▁lid": 31794,
+      "stackoverflow": 31795,
+      "ivalent": 31796,
+      "▁legislative": 31797,
+      "一期": 31798,
+      "打扮": 31799,
+      "▁KB": 31800,
+      "▁Sr": 31801,
+      "sear": 31802,
+      "▁needing": 31803,
+      "fy": 31804,
+      "▁uri": 31805,
+      "▁charter": 31806,
+      "或者说": 31807,
+      "▁colorectal": 31808,
+      "ogo": 31809,
+      "原被告": 31810,
+      "▁Crow": 31811,
+      "春季": 31812,
+      "USA": 31813,
+      "▁backgrounds": 31814,
+      "rases": 31815,
+      "▁Nebraska": 31816,
+      "帮扶": 31817,
+      "二百": 31818,
+      "Studio": 31819,
+      "ibular": 31820,
+      "是比较": 31821,
+      "▁Colombia": 31822,
+      "末尾": 31823,
+      "嵌入": 31824,
+      "▁ii": 31825,
+      "▁scans": 31826,
+      "___": 31827,
+      "nMc": 31828,
+      "▁Princess": 31829,
+      "▁semantic": 31830,
+      "问卷": 31831,
+      "AAA": 31832,
+      "▁COMM": 31833,
+      "记功": 31834,
+      "这件": 31835,
+      "CLUDE": 31836,
+      "就需要": 31837,
+      "▁Ko": 31838,
+      "异界": 31839,
+      "▁Adult": 31840,
+      "在哪": 31841,
+      "▁bell": 31842,
+      "imiter": 31843,
+      "▁immigrants": 31844,
+      "uka": 31845,
+      "rically": 31846,
+      "nCPU": 31847,
+      "扣除": 31848,
+      "特效": 31849,
+      "请求法院": 31850,
+      "anship": 31851,
+      "詹姆斯": 31852,
+      "▁释义": 31853,
+      "▁Remote": 31854,
+      "▁equivalence": 31855,
+      "▁у": 31856,
+      "▁investor": 31857,
+      "▁Simulation": 31858,
+      "原因是": 31859,
+      "▁spam": 31860,
+      "▁Milan": 31861,
+      "▁Approach": 31862,
+      "▁Transportation": 31863,
+      "抢救": 31864,
+      "▁speaks": 31865,
+      "对角": 31866,
+      "意愿": 31867,
+      "alar": 31868,
+      "arta": 31869,
+      "早在": 31870,
+      "愤怒": 31871,
+      "自然人": 31872,
+      "▁constitute": 31873,
+      "伦理": 31874,
+      "二次根": 31875,
+      "Extract": 31876,
+      "▁marketplace": 31877,
+      "otics": 31878,
+      "Consumer": 31879,
+      "小学文化": 31880,
+      "outputs": 31881,
+      "▁从": 31882,
+      "乐队": 31883,
+      "▁simplest": 31884,
+      "▁assembled": 31885,
+      "耐磨": 31886,
+      "道具": 31887,
+      "nHaving": 31888,
+      "▁叶": 31889,
+      "衍生": 31890,
+      "▁Emergency": 31891,
+      "▁逝世日期": 31892,
+      "▁Wonder": 31893,
+      "▁magical": 31894,
+      "▁regulators": 31895,
+      "适配": 31896,
+      "农田": 31897,
+      "chol": 31898,
+      "感和": 31899,
+      "▁Associate": 31900,
+      "▁provincial": 31901,
+      "rov": 31902,
+      "nhave": 31903,
+      "确立": 31904,
+      "▁项目": 31905,
+      "▁}}\"\\": 31906,
+      "▁Strong": 31907,
+      "▁influential": 31908,
+      "saved": 31909,
+      "▁meg": 31910,
+      "元件": 31911,
+      "ovine": 31912,
+      "▁Euler": 31913,
+      "ximately": 31914,
+      "igraph": 31915,
+      "oplastic": 31916,
+      "▁clinics": 31917,
+      "特別": 31918,
+      "fiction": 31919,
+      "▁Beyond": 31920,
+      "▁locals": 31921,
+      "深处": 31922,
+      "CEO": 31923,
+      "的患者": 31924,
+      "Sep": 31925,
+      "代书记员": 31926,
+      "▁Pope": 31927,
+      "nSorry": 31928,
+      "第一百五十四条第一款": 31929,
+      "关键是": 31930,
+      "地名": 31931,
+      "oplast": 31932,
+      "setState": 31933,
+      "reational": 31934,
+      "场上位置": 31935,
+      "▁Amy": 31936,
+      "▁drew": 31937,
+      "▁crops": 31938,
+      "▁Running": 31939,
+      "与中国": 31940,
+      "德里": 31941,
+      "没法": 31942,
+      "莆田": 31943,
+      "nJames": 31944,
+      "冶金": 31945,
+      "人类的": 31946,
+      "uclid": 31947,
+      "▁Welsh": 31948,
+      "SES": 31949,
+      "odi": 31950,
+      "LF": 31951,
+      "pointer": 31952,
+      "▁uniquely": 31953,
+      "▁Montana": 31954,
+      "},\"": 31955,
+      "指的是": 31956,
+      "同居": 31957,
+      "零部件": 31958,
+      "Factor": 31959,
+      "▁canonical": 31960,
+      "立马": 31961,
+      "▁adm": 31962,
+      "▁ethn": 31963,
+      "▁mism": 31964,
+      "新颖": 31965,
+      "▁hello": 31966,
+      "主要的": 31967,
+      "anonymous": 31968,
+      "方法的": 31969,
+      "onClick": 31970,
+      "▁drainage": 31971,
+      "Attr": 31972,
+      "基本的": 31973,
+      "Filename": 31974,
+      "▁traveled": 31975,
+      "打的": 31976,
+      "好吧": 31977,
+      "Corp": 31978,
+      "实践中": 31979,
+      "▁beneath": 31980,
+      "在不": 31981,
+      "consumer": 31982,
+      "▁aesthetic": 31983,
+      "诠释": 31984,
+      "pel": 31985,
+      "▁tire": 31986,
+      "chanical": 31987,
+      "武术": 31988,
+      "已经被": 31989,
+      "▁furnished": 31990,
+      "ASK": 31991,
+      "▁Biology": 31992,
+      "▁positioning": 31993,
+      "▁Eye": 31994,
+      "乙烯": 31995,
+      "\\\\:": 31996,
+      "境外": 31997,
+      "异地": 31998,
+      "转账": 31999,
+      "首批": 32000,
+      "综合征": 32001,
+      "looking": 32002,
+      "▁generators": 32003,
+      "▁recycling": 32004,
+      "plasia": 32005,
+      "ieu": 32006,
+      "主管部门": 32007,
+      "▁identifies": 32008,
+      "并已": 32009,
+      "tabs": 32010,
+      "▁TRUE": 32011,
+      "本着": 32012,
+      "CAP": 32013,
+      "major": 32014,
+      "▁Perm": 32015,
+      "▁Chem": 32016,
+      "中期": 32017,
+      "袭击": 32018,
+      "党委委员": 32019,
+      "▁transfers": 32020,
+      "▁attraction": 32021,
+      "相亲": 32022,
+      "下班": 32023,
+      "▁attain": 32024,
+      "▁属": 32025,
+      "▁pac": 32026,
+      "▁clay": 32027,
+      "化解": 32028,
+      "有望": 32029,
+      "▁OFF": 32030,
+      "▁applicants": 32031,
+      "determ": 32032,
+      "vae": 32033,
+      "archar": 32034,
+      "包容": 32035,
+      "心疼": 32036,
+      "模特": 32037,
+      "▁Poll": 32038,
+      "▁OTHERWISE": 32039,
+      "味精": 32040,
+      "劳动者": 32041,
+      "▁signatures": 32042,
+      "▁traditionally": 32043,
+      "愉快": 32044,
+      "▁permalink": 32045,
+      "Desc": 32046,
+      "远的": 32047,
+      "orr": 32048,
+      "nData": 32049,
+      "▁comprising": 32050,
+      "阀门": 32051,
+      "生物学": 32052,
+      "▁Resort": 32053,
+      "▁subgroups": 32054,
+      "cot": 32055,
+      "大门": 32056,
+      "▁latency": 32057,
+      "inist": 32058,
+      "▁Arena": 32059,
+      "胶囊": 32060,
+      "dar": 32061,
+      "本案诉讼费": 32062,
+      "▁conce": 32063,
+      "transport": 32064,
+      "可视": 32065,
+      "awk": 32066,
+      "▁身高": 32067,
+      "不适合": 32068,
+      "oshi": 32069,
+      "weights": 32070,
+      "▁perceptions": 32071,
+      "我可以": 32072,
+      "ovich": 32073,
+      "andbox": 32074,
+      "choices": 32075,
+      "▁NJ": 32076,
+      "▁exams": 32077,
+      "Ep": 32078,
+      "故乡": 32079,
+      "▁influenza": 32080,
+      "▁LV": 32081,
+      "▁AMP": 32082,
+      "▁flaw": 32083,
+      "配音": 32084,
+      "▁Finnish": 32085,
+      "lette": 32086,
+      "▁elite": 32087,
+      "无业": 32088,
+      "Simplify": 32089,
+      "{:": 32090,
+      "▁hass": 32091,
+      "▁wage": 32092,
+      "三包": 32093,
+      "低的": 32094,
+      "▁dermat": 32095,
+      "路程": 32096,
+      "tables": 32097,
+      "opathic": 32098,
+      "▁proximity": 32099,
+      "ва": 32100,
+      "第八章": 32101,
+      "▁toString": 32102,
+      "serif": 32103,
+      "▁intuitive": 32104,
+      "adas": 32105,
+      "角落": 32106,
+      "▁Growth": 32107,
+      "▁museums": 32108,
+      "长达": 32109,
+      "▁Effects": 32110,
+      "相思": 32111,
+      "▁Liga": 32112,
+      "ocytic": 32113,
+      "▁Allow": 32114,
+      "Three": 32115,
+      "三点": 32116,
+      "教学中": 32117,
+      "imag": 32118,
+      "▁diesel": 32119,
+      "云中": 32120,
+      "odox": 32121,
+      "▁sliding": 32122,
+      "并以": 32123,
+      "态势": 32124,
+      "如下图": 32125,
+      "阿拉伯": 32126,
+      "▁Parse": 32127,
+      "才发现": 32128,
+      "▁observer": 32129,
+      "众所": 32130,
+      "住河南省": 32131,
+      "aten": 32132,
+      "nout": 32133,
+      "大神": 32134,
+      "▁hereby": 32135,
+      "▁Bristol": 32136,
+      "抵达": 32137,
+      "▁COL": 32138,
+      "▁pays": 32139,
+      "往来": 32140,
+      "▁anger": 32141,
+      "▁Peru": 32142,
+      "Promise": 32143,
+      "▁Var": 32144,
+      "▁sweep": 32145,
+      "杀手": 32146,
+      "frontend": 32147,
+      "▁tier": 32148,
+      "第一百五十四条": 32149,
+      "六个": 32150,
+      "旋律": 32151,
+      "雷达": 32152,
+      "▁Mitchell": 32153,
+      "RM": 32154,
+      "可选": 32155,
+      "▁assignments": 32156,
+      "头痛": 32157,
+      "满了": 32158,
+      "保守": 32159,
+      "amos": 32160,
+      "▁refin": 32161,
+      "▁pupils": 32162,
+      "▁inhibits": 32163,
+      "个性化": 32164,
+      "称之为": 32165,
+      "chapter": 32166,
+      "napi": 32167,
+      "▁recognised": 32168,
+      "的责任": 32169,
+      "的建设": 32170,
+      "吵架": 32171,
+      "▁Too": 32172,
+      "▁Permission": 32173,
+      "写字": 32174,
+      "的书": 32175,
+      "向本院提出撤诉申请": 32176,
+      "▁Something": 32177,
+      "▁ips": 32178,
+      "了自己": 32179,
+      "sites": 32180,
+      "ocarcin": 32181,
+      "部落": 32182,
+      "IMP": 32183,
+      "不出来": 32184,
+      "行人": 32185,
+      "▁Cob": 32186,
+      "iblings": 32187,
+      "GN": 32188,
+      "齿轮": 32189,
+      "▁conjecture": 32190,
+      "hind": 32191,
+      "▁obese": 32192,
+      "Throw": 32193,
+      "▁HAVE": 32194,
+      "▁deformation": 32195,
+      "▁Ave": 32196,
+      "▁Pent": 32197,
+      "▁ethics": 32198,
+      "如下所示": 32199,
+      "▁basics": 32200,
+      "▁opponents": 32201,
+      "abol": 32202,
+      "ilia": 32203,
+      "巡视": 32204,
+      "担任职务": 32205,
+      "诉求": 32206,
+      "坐落于": 32207,
+      "温州市": 32208,
+      "岁以下的儿童": 32209,
+      "和对": 32210,
+      "▁Crown": 32211,
+      "偷偷": 32212,
+      "▁treats": 32213,
+      "▁Patient": 32214,
+      "▁activist": 32215,
+      "降低了": 32216,
+      "▁Half": 32217,
+      "▁fonts": 32218,
+      "生产线": 32219,
+      "Expl": 32220,
+      "座位": 32221,
+      "新冠肺炎": 32222,
+      "中年": 32223,
+      "ighbor": 32224,
+      "BACK": 32225,
+      "明亮": 32226,
+      "夜晚": 32227,
+      "▁\")\\": 32228,
+      "▁upgraded": 32229,
+      "▁sigma": 32230,
+      "无财产可供执行": 32231,
+      "▁Vienna": 32232,
+      "失踪": 32233,
+      "▁Steps": 32234,
+      "上诉案件受理费": 32235,
+      "主治": 32236,
+      "会展": 32237,
+      "▁Upload": 32238,
+      "的申请符合法律规定": 32239,
+      "无正当理由拒不到庭参加诉讼": 32240,
+      "▁knowledgeable": 32241,
+      "地狱": 32242,
+      "▁vic": 32243,
+      "▁abortion": 32244,
+      "inent": 32245,
+      "▁enterprises": 32246,
+      "总的": 32247,
+      "▁quotient": 32248,
+      "▁Interface": 32249,
+      "刻苦": 32250,
+      "Groups": 32251,
+      "悲剧": 32252,
+      "▁Flore": 32253,
+      "月中": 32254,
+      "▁electoral": 32255,
+      "试卷": 32256,
+      "内有": 32257,
+      "educ": 32258,
+      "▁Wedding": 32259,
+      "▁antagonist": 32260,
+      "能让": 32261,
+      "buff": 32262,
+      "▁Esp": 32263,
+      "StackTrace": 32264,
+      "期间的限制": 32265,
+      "Virtual": 32266,
+      "CIAL": 32267,
+      "ZX": 32268,
+      "▁曹": 32269,
+      "DEF": 32270,
+      "tees": 32271,
+      "▁doi": 32272,
+      "▁lesser": 32273,
+      "▁discharg": 32274,
+      "做个": 32275,
+      "明朝": 32276,
+      "▁teens": 32277,
+      "casting": 32278,
+      "▁Innovation": 32279,
+      "unny": 32280,
+      "事实和理由": 32281,
+      "实务": 32282,
+      "▁`<": 32283,
+      "还不错": 32284,
+      "最短": 32285,
+      "▁embryonic": 32286,
+      "VALID": 32287,
+      "mkdir": 32288,
+      "折扣": 32289,
+      "▁Len": 32290,
+      "checkbox": 32291,
+      "▁Metropolitan": 32292,
+      "林地": 32293,
+      "大厅": 32294,
+      "生活在": 32295,
+      "体力": 32296,
+      "饮用": 32297,
+      "urst": 32298,
+      "yzer": 32299,
+      "然后根据": 32300,
+      "▁acetyl": 32301,
+      "functional": 32302,
+      "批判": 32303,
+      "亲切": 32304,
+      "▁Pure": 32305,
+      "震撼": 32306,
+      "sqlite": 32307,
+      "过年": 32308,
+      "Eloquent": 32309,
+      "nDec": 32310,
+      "nPort": 32311,
+      "avorite": 32312,
+      "Rs": 32313,
+      "▁fasc": 32314,
+      "▁Touch": 32315,
+      "四周": 32316,
+      "▁triangles": 32317,
+      "▁pkg": 32318,
+      "ynasty": 32319,
+      "lod": 32320,
+      "出资": 32321,
+      "抽样": 32322,
+      "imir": 32323,
+      "nJoin": 32324,
+      "▁Racing": 32325,
+      "▁CNS": 32326,
+      "▁boots": 32327,
+      "▁consume": 32328,
+      "hey": 32329,
+      "了中国": 32330,
+      "在一定": 32331,
+      "▁Xbox": 32332,
+      "▁人": 32333,
+      "▁ui": 32334,
+      "▁Appl": 32335,
+      "vt": 32336,
+      "nInput": 32337,
+      "▁laboratories": 32338,
+      "▁Rever": 32339,
+      "brane": 32340,
+      "▁Strategy": 32341,
+      "▁disappointed": 32342,
+      "Well": 32343,
+      ",(": 32344,
+      "业界": 32345,
+      "促使": 32346,
+      "STM": 32347,
+      "八个月": 32348,
+      "▁aired": 32349,
+      "sections": 32350,
+      "▁RC": 32351,
+      "▁chef": 32352,
+      "同人": 32353,
+      "arel": 32354,
+      "arl": 32355,
+      "spacing": 32356,
+      "reck": 32357,
+      "可分为": 32358,
+      "▁Tell": 32359,
+      "娱乐圈": 32360,
+      "宽带": 32361,
+      "nArch": 32362,
+      "压器": 32363,
+      "Most": 32364,
+      "温馨提示": 32365,
+      "人对": 32366,
+      "写成": 32367,
+      "▁hadn": 32368,
+      "▁tyrosine": 32369,
+      "▁Sean": 32370,
+      "▁squared": 32371,
+      "▁nationally": 32372,
+      "科学的": 32373,
+      "▁shit": 32374,
+      "▁uncon": 32375,
+      "刑期至": 32376,
+      "atomical": 32377,
+      "nThrough": 32378,
+      "\\')[": 32379,
+      "▁maturation": 32380,
+      "urated": 32381,
+      "▁municipalities": 32382,
+      "Phi": 32383,
+      "ints": 32384,
+      "uniform": 32385,
+      "荧光": 32386,
+      "探究": 32387,
+      "▁Palm": 32388,
+      "▁Wiley": 32389,
+      "美洲": 32390,
+      "金牌": 32391,
+      "eves": 32392,
+      "增产": 32393,
+      "widgets": 32394,
+      "由中国": 32395,
+      "GDP": 32396,
+      "▁Stri": 32397,
+      "科学记数法": 32398,
+      "ndf": 32399,
+      "▁Beck": 32400,
+      "▁Wave": 32401,
+      "▁illum": 32402,
+      "▁shifted": 32403,
+      "▁Asp": 32404,
+      "()),\\": 32405,
+      "▁Hist": 32406,
+      "流失": 32407,
+      "作为一名": 32408,
+      "Thank": 32409,
+      "▁steroid": 32410,
+      "▁国家": 32411,
+      "▁sisters": 32412,
+      "▁请": 32413,
+      "海鲜": 32414,
+      "▁farms": 32415,
+      "▁Richmond": 32416,
+      "zb": 32417,
+      "▁Crim": 32418,
+      "itution": 32419,
+      "IntegerField": 32420,
+      "小平": 32421,
+      "▁mant": 32422,
+      "双向": 32423,
+      "RU": 32424,
+      "bris": 32425,
+      "原判决": 32426,
+      "Cent": 32427,
+      "将被执行人": 32428,
+      "▁Router": 32429,
+      "Navigation": 32430,
+      "差异有": 32431,
+      "▁Sleep": 32432,
+      "▁superb": 32433,
+      "淡淡": 32434,
+      "既是": 32435,
+      "▁supervision": 32436,
+      "cols": 32437,
+      "▁Belgian": 32438,
+      "rones": 32439,
+      "nMeanwhile": 32440,
+      "历史的": 32441,
+      "Bs": 32442,
+      "令牌": 32443,
+      "描写": 32444,
+      "▁Yam": 32445,
+      "常年": 32446,
+      "ammed": 32447,
+      "和社会保障": 32448,
+      "▁summit": 32449,
+      "▁nullptr": 32450,
+      "rish": 32451,
+      "商业银行": 32452,
+      "CY": 32453,
+      "__,": 32454,
+      "万美元": 32455,
+      "执行标的": 32456,
+      "索尼": 32457,
+      "NK": 32458,
+      "▁displaying": 32459,
+      "副部长": 32460,
+      "jpeg": 32461,
+      "看不到": 32462,
+      "泽东": 32463,
+      "管控": 32464,
+      "^{(": 32465,
+      "\\'][": 32466,
+      "ocha": 32467,
+      "▁relates": 32468,
+      "青睐": 32469,
+      "原有": 32470,
+      "proj": 32471,
+      "运营商": 32472,
+      "现已审查": 32473,
+      "▁ACTION": 32474,
+      "lazy": 32475,
+      "quest": 32476,
+      "玩意": 32477,
+      "▁Layout": 32478,
+      "oproject": 32479,
+      "▁Mann": 32480,
+      "▁inadequ": 32481,
+      "▁forcing": 32482,
+      "▁corruption": 32483,
+      "增添": 32484,
+      "向本院提交": 32485,
+      "▁safer": 32486,
+      "irlines": 32487,
+      "Contents": 32488,
+      "▁perfusion": 32489,
+      "出入": 32490,
+      "▁ol": 32491,
+      "▁firing": 32492,
+      "▁beating": 32493,
+      "▁stating": 32494,
+      "宣判前": 32495,
+      "ortex": 32496,
+      "▁Fashion": 32497,
+      "Old": 32498,
+      "getInstance": 32499,
+      "本案案件受理费": 32500,
+      "▁Curt": 32501,
+      "uo": 32502,
+      "▁Sit": 32503,
+      "catalog": 32504,
+      "▁handful": 32505,
+      "高档": 32506,
+      "▁Gem": 32507,
+      "noise": 32508,
+      "▁humor": 32509,
+      "Annotation": 32510,
+      "▁lakes": 32511,
+      "▁wines": 32512,
+      "▁Kid": 32513,
+      "长方形的": 32514,
+      "▁prestigious": 32515,
+      "▁Electronic": 32516,
+      "▁CBS": 32517,
+      "▁shoe": 32518,
+      "▁inherent": 32519,
+      "ATP": 32520,
+      "▁amy": 32521,
+      "▁Shanghai": 32522,
+      "欺负": 32523,
+      "▁pt": 32524,
+      "▁Sanders": 32525,
+      "开源": 32526,
+      "旅馆": 32527,
+      "▁traditions": 32528,
+      "▁partnerships": 32529,
+      "旅行社": 32530,
+      "买卖合同": 32531,
+      "▁Movement": 32532,
+      "▁Orchestra": 32533,
+      "▁深圳市": 32534,
+      "日常生活": 32535,
+      "▁textbook": 32536,
+      "子和": 32537,
+      "Seed": 32538,
+      "inced": 32539,
+      "pf": 32540,
+      "▁rs": 32541,
+      "edly": 32542,
+      "一枚": 32543,
+      "排队": 32544,
+      "消费者的": 32545,
+      "Apple": 32546,
+      "▁roster": 32547,
+      "▁diffraction": 32548,
+      "亮相": 32549,
+      "▁normalized": 32550,
+      "▁PAR": 32551,
+      "▁homeostasis": 32552,
+      "Who": 32553,
+      "▁ns": 32554,
+      "choose": 32555,
+      "▁Classification": 32556,
+      "完善的": 32557,
+      "▁eighth": 32558,
+      "香菇": 32559,
+      "面积为": 32560,
+      "▁Mes": 32561,
+      "▁Drag": 32562,
+      "▁Plugin": 32563,
+      "rea": 32564,
+      "不具备": 32565,
+      "▁este": 32566,
+      "trm": 32567,
+      "▁Clay": 32568,
+      "▁cancellation": 32569,
+      "\\';": 32570,
+      "未在": 32571,
+      "nWilliam": 32572,
+      "不小心": 32573,
+      "storm": 32574,
+      "▁cleared": 32575,
+      "REMO": 32576,
+      "▁dried": 32577,
+      "▁Gov": 32578,
+      "ploys": 32579,
+      "有关的": 32580,
+      "▁cref": 32581,
+      "南区": 32582,
+      "▁auditory": 32583,
+      "一中": 32584,
+      "乘积": 32585,
+      "的真": 32586,
+      "▁起点网": 32587,
+      "hemas": 32588,
+      "sorted": 32589,
+      "uitary": 32590,
+      "▁myster": 32591,
+      "vascular": 32592,
+      "▁drum": 32593,
+      "▁twitter": 32594,
+      "Trigger": 32595,
+      "之心": 32596,
+      "official": 32597,
+      "▁多": 32598,
+      "誉为": 32599,
+      "首页": 32600,
+      "▁perpet": 32601,
+      "剪辑": 32602,
+      "职业教育": 32603,
+      "▁fixture": 32604,
+      "nmain": 32605,
+      "▁Athletic": 32606,
+      "▁permeability": 32607,
+      "嘉兴": 32608,
+      "市中心": 32609,
+      "针对性": 32610,
+      "Mart": 32611,
+      "▁Jama": 32612,
+      "▁endblock": 32613,
+      "▁Kel": 32614,
+      "zerbai": 32615,
+      "两地": 32616,
+      "驾驶员": 32617,
+      "专业知识": 32618,
+      "走过": 32619,
+      "rates": 32620,
+      "▁eigenvalues": 32621,
+      "岩石": 32622,
+      "教训": 32623,
+      "yman": 32624,
+      "imming": 32625,
+      "▁climbing": 32626,
+      "ло": 32627,
+      "阴谋": 32628,
+      "▁contraction": 32629,
+      "Img": 32630,
+      "▁altitude": 32631,
+      "做题": 32632,
+      "ntags": 32633,
+      "zm": 32634,
+      "▁commod": 32635,
+      "▁additive": 32636,
+      ">'": 32637,
+      "新车": 32638,
+      "▁Plate": 32639,
+      "Ui": 32640,
+      "的人们": 32641,
+      "评论区": 32642,
+      "▁gambling": 32643,
+      "▁ко": 32644,
+      "talk": 32645,
+      "▁forgotten": 32646,
+      "排出": 32647,
+      "Try": 32648,
+      "为主的": 32649,
+      "▁Prof": 32650,
+      "gradle": 32651,
+      ");//": 32652,
+      "tutorial": 32653,
+      "runs": 32654,
+      "但这": 32655,
+      "惩罚": 32656,
+      "CONT": 32657,
+      "mapper": 32658,
+      "▁winds": 32659,
+      "▁folding": 32660,
+      "高位": 32661,
+      "▁charm": 32662,
+      "收拾": 32663,
+      "▁ceramic": 32664,
+      "载明": 32665,
+      "侦查": 32666,
+      "▁heterogeneity": 32667,
+      "▁Wor": 32668,
+      "▁Flex": 32669,
+      "▁geomet": 32670,
+      "parseInt": 32671,
+      "▁女": 32672,
+      "在案": 32673,
+      "怎么说": 32674,
+      "▁simplicity": 32675,
+      "▁intercept": 32676,
+      "甲状腺": 32677,
+      "▁Parish": 32678,
+      "▁lex": 32679,
+      "自在": 32680,
+      "算式": 32681,
+      "▁Pand": 32682,
+      "▁shar": 32683,
+      "▁flavors": 32684,
+      "弯曲": 32685,
+      "▁ya": 32686,
+      "▁organism": 32687,
+      "CAR": 32688,
+      "系统中": 32689,
+      "ilage": 32690,
+      "决定的": 32691,
+      "可以让": 32692,
+      "unda": 32693,
+      "nVideo": 32694,
+      "▁implementations": 32695,
+      "港澳": 32696,
+      "零食": 32697,
+      "老家": 32698,
+      "NonNull": 32699,
+      "架空": 32700,
+      "▁Theme": 32701,
+      "▁segmentation": 32702,
+      "▁volcan": 32703,
+      "企业在": 32704,
+      "内容的": 32705,
+      "价钱": 32706,
+      "执恢": 32707,
+      "崩溃": 32708,
+      "bey": 32709,
+      "▁legendary": 32710,
+      "相差": 32711,
+      "结尾": 32712,
+      "tS": 32713,
+      "▁jury": 32714,
+      "我相信": 32715,
+      "▁Margaret": 32716,
+      "高原": 32717,
+      "不做": 32718,
+      "推送": 32719,
+      "要求被执行人": 32720,
+      "▁Toyota": 32721,
+      "▁resting": 32722,
+      "Bbb": 32723,
+      "了多少": 32724,
+      "某某与被告": 32725,
+      "▁分布": 32726,
+      "半导体": 32727,
+      "▁lawn": 32728,
+      "唿吸": 32729,
+      "肝脏": 32730,
+      "▁nd": 32731,
+      "▁exploit": 32732,
+      "▁schedules": 32733,
+      "外贸": 32734,
+      "高度重视": 32735,
+      "anas": 32736,
+      "mini": 32737,
+      "▁Δ": 32738,
+      "Done": 32739,
+      "钢管": 32740,
+      "▁salad": 32741,
+      "不安": 32742,
+      "几个月": 32743,
+      "same": 32744,
+      "▁Lip": 32745,
+      "QR": 32746,
+      "▁accelerate": 32747,
+      "养护": 32748,
+      "▁axes": 32749,
+      "reject": 32750,
+      "定点": 32751,
+      "醒来": 32752,
+      "today": 32753,
+      "ctrl": 32754,
+      "▁tale": 32755,
+      "▁Storm": 32756,
+      "小小": 32757,
+      "ffic": 32758,
+      "仲裁裁决": 32759,
+      "isset": 32760,
+      "一堆": 32761,
+      "University": 32762,
+      "Std": 32763,
+      "▁NBC": 32764,
+      "年限": 32765,
+      "▁ambig": 32766,
+      "悲伤": 32767,
+      "证据充分": 32768,
+      "▁colorful": 32769,
+      "有一次": 32770,
+      "incipal": 32771,
+      "jud": 32772,
+      "我一直": 32773,
+      "▁交通": 32774,
+      "▁spa": 32775,
+      "ensen": 32776,
+      "oslav": 32777,
+      "▁finishes": 32778,
+      "带上": 32779,
+      "生产任务": 32780,
+      "TX": 32781,
+      "译者": 32782,
+      "北京大学": 32783,
+      "过得": 32784,
+      "▁Anc": 32785,
+      "棉花": 32786,
+      "sep": 32787,
+      "▁Parker": 32788,
+      "▁graduating": 32789,
+      "▁cuisine": 32790,
+      "沉积": 32791,
+      "▁Camb": 32792,
+      "你去": 32793,
+      "▁\\'\\": 32794,
+      "riors": 32795,
+      "揭示": 32796,
+      "maven": 32797,
+      "执照": 32798,
+      "ylase": 32799,
+      "▁cleaner": 32800,
+      "▁Processing": 32801,
+      "kk": 32802,
+      "美学": 32803,
+      "定时": 32804,
+      "STEM": 32805,
+      "还是有": 32806,
+      "▁editorial": 32807,
+      "▁FE": 32808,
+      "amas": 32809,
+      "hyde": 32810,
+      "网民": 32811,
+      "▁Stanford": 32812,
+      "十八章": 32813,
+      "▁kits": 32814,
+      "keyword": 32815,
+      "expensive": 32816,
+      "电台": 32817,
+      "VIP": 32818,
+      "业务的": 32819,
+      "选项中挑": 32820,
+      "▁naming": 32821,
+      "agne": 32822,
+      "户籍证明": 32823,
+      "洗澡": 32824,
+      "gain": 32825,
+      "司令": 32826,
+      "▁harsh": 32827,
+      "nRelated": 32828,
+      "▁antimicrobial": 32829,
+      "保卫": 32830,
+      "▁Swift": 32831,
+      "一番": 32832,
+      "▁SB": 32833,
+      "Blog": 32834,
+      "Arguments": 32835,
+      "人身": 32836,
+      "lli": 32837,
+      "ursion": 32838,
+      "▁adolescent": 32839,
+      "而非": 32840,
+      "▁sudo": 32841,
+      "露出": 32842,
+      "nShow": 32843,
+      "setText": 32844,
+      "keleton": 32845,
+      "各样的": 32846,
+      "土耳": 32847,
+      "▁wx": 32848,
+      "▁Foot": 32849,
+      "Encoding": 32850,
+      "室外": 32851,
+      "rosso": 32852,
+      "peer": 32853,
+      "repr": 32854,
+      "化市": 32855,
+      "起的": 32856,
+      "奶牛": 32857,
+      "nBr": 32858,
+      "高品质": 32859,
+      "▁switches": 32860,
+      "▁administrator": 32861,
+      "▁Barry": 32862,
+      "▁betting": 32863,
+      "前者": 32864,
+      "太太": 32865,
+      "ashing": 32866,
+      "▁Madison": 32867,
+      "▁Commissioner": 32868,
+      "▁dispute": 32869,
+      "TG": 32870,
+      "uchy": 32871,
+      "▁Southeast": 32872,
+      "▁所": 32873,
+      "▁prol": 32874,
+      "片的": 32875,
+      "istani": 32876,
+      "ます": 32877,
+      "amins": 32878,
+      "▁Crypt": 32879,
+      "▁Jerry": 32880,
+      "古城": 32881,
+      "氨酸": 32882,
+      "直流": 32883,
+      "▁Squad": 32884,
+      "▁linux": 32885,
+      "肥胖": 32886,
+      "在上海": 32887,
+      "▁judgment": 32888,
+      "▁Equipment": 32889,
+      "▁endpoints": 32890,
+      "▁aneur": 32891,
+      "▁DNS": 32892,
+      "▁Eston": 32893,
+      "▁finals": 32894,
+      "tun": 32895,
+      "quire": 32896,
+      "workers": 32897,
+      "▁compassion": 32898,
+      "▁compelling": 32899,
+      "cke": 32900,
+      "高科技": 32901,
+      "ustralia": 32902,
+      "picker": 32903,
+      "结束了": 32904,
+      "acies": 32905,
+      "styled": 32906,
+      "▁Larry": 32907,
+      "乡村振兴": 32908,
+      "▁Artist": 32909,
+      "▁graduation": 32910,
+      "试用": 32911,
+      "▁reactivity": 32912,
+      "最长": 32913,
+      "结石": 32914,
+      "aky": 32915,
+      "▁}}</": 32916,
+      "bast": 32917,
+      "forum": 32918,
+      "▁Grace": 32919,
+      "▁drove": 32920,
+      "天生": 32921,
+      "一手": 32922,
+      "▁donate": 32923,
+      "而又": 32924,
+      "还可": 32925,
+      "fragment": 32926,
+      "bud": 32927,
+      "▁nasal": 32928,
+      "▁mediate": 32929,
+      "相连": 32930,
+      "一会儿": 32931,
+      "▁glow": 32932,
+      "▁Groups": 32933,
+      "▁sar": 32934,
+      "oietic": 32935,
+      "核酸": 32936,
+      "▁Laura": 32937,
+      "晚饭": 32938,
+      "▁dropping": 32939,
+      "第一百三十三条之一": 32940,
+      "有几个": 32941,
+      "▁dioxide": 32942,
+      "Great": 32943,
+      "Prom": 32944,
+      "DER": 32945,
+      "送达了": 32946,
+      "一般是": 32947,
+      "▁HttpResponse": 32948,
+      "ichia": 32949,
+      "▁担任职务": 32950,
+      "起来了": 32951,
+      "isha": 32952,
+      "phyl": 32953,
+      "▁Same": 32954,
+      "agents": 32955,
+      "▁Fields": 32956,
+      "▁phenotypes": 32957,
+      "dz": 32958,
+      "▁gram": 32959,
+      "▁rocket": 32960,
+      "▁insufficient": 32961,
+      "族自治州": 32962,
+      "disk": 32963,
+      "如果说": 32964,
+      "gold": 32965,
+      "乒乓": 32966,
+      "▁engineered": 32967,
+      "再审申请": 32968,
+      "▁suppressed": 32969,
+      "▁Cleaning": 32970,
+      "sensor": 32971,
+      "这样一个": 32972,
+      "nDate": 32973,
+      "▁brackets": 32974,
+      "主要以": 32975,
+      "▁township": 32976,
+      "▁万": 32977,
+      "的一点": 32978,
+      "earning": 32979,
+      "tube": 32980,
+      "▁rush": 32981,
+      "precedented": 32982,
+      "▁accelerated": 32983,
+      "WC": 32984,
+      "卫视": 32985,
+      "光盘": 32986,
+      "Excel": 32987,
+      "▁Markov": 32988,
+      "和中国": 32989,
+      "▁clarity": 32990,
+      "▁clips": 32991,
+      "约会": 32992,
+      "nreq": 32993,
+      "副会长": 32994,
+      "案发后": 32995,
+      "incinn": 32996,
+      "副作用": 32997,
+      "enosis": 32998,
+      "▁emergence": 32999,
+      "▁negotiations": 33000,
+      "▁identities": 33001,
+      "即时": 33002,
+      "视力": 33003,
+      "nPut": 33004,
+      "了两": 33005,
+      "着名": 33006,
+      "▁hybridization": 33007,
+      "JO": 33008,
+      "想过": 33009,
+      "由人民法院裁定": 33010,
+      "的女生": 33011,
+      "▁Pink": 33012,
+      "一小": 33013,
+      "Linear": 33014,
+      "▁allev": 33015,
+      "风水": 33016,
+      "下半年": 33017,
+      "▁Inside": 33018,
+      "Changes": 33019,
+      "赠送": 33020,
+      "甜蜜": 33021,
+      "预先": 33022,
+      "▁convinced": 33023,
+      "\\'));\\": 33024,
+      "▁方": 33025,
+      "ucas": 33026,
+      "▁lapar": 33027,
+      "intendo": 33028,
+      "▁slices": 33029,
+      "督促": 33030,
+      "asha": 33031,
+      "research": 33032,
+      "▁Fail": 33033,
+      "单调递增": 33034,
+      "职业生涯": 33035,
+      "短暂": 33036,
+      "Picker": 33037,
+      "▁Binary": 33038,
+      "▁Jennifer": 33039,
+      "二者": 33040,
+      "毛泽东": 33041,
+      "regn": 33042,
+      "局限": 33043,
+      "套餐": 33044,
+      "领袖": 33045,
+      "调动": 33046,
+      "选出正确的选项": 33047,
+      "▁imagination": 33048,
+      "白的": 33049,
+      "机构的": 33050,
+      "在内的": 33051,
+      "nKeep": 33052,
+      "LM": 33053,
+      "固体": 33054,
+      "自然的": 33055,
+      "▁tempt": 33056,
+      "▁peaceful": 33057,
+      "价位": 33058,
+      "▁Theater": 33059,
+      "▁XP": 33060,
+      "剥夺政治权利终身": 33061,
+      "里程": 33062,
+      "▁pyr": 33063,
+      "NI": 33064,
+      "冷静": 33065,
+      "orton": 33066,
+      "上看": 33067,
+      "NEW": 33068,
+      "按键": 33069,
+      "avg": 33070,
+      "▁BU": 33071,
+      "Payload": 33072,
+      "nBased": 33073,
+      "▁adsorption": 33074,
+      "文中": 33075,
+      "▁Vers": 33076,
+      "基数": 33077,
+      "已破裂": 33078,
+      "▁forums": 33079,
+      "Dependency": 33080,
+      "raulic": 33081,
+      "坦克": 33082,
+      "▁assumes": 33083,
+      "▁Forwarded": 33084,
+      "少儿": 33085,
+      "Age": 33086,
+      "▁inference": 33087,
+      "为我": 33088,
+      "▁Fib": 33089,
+      "以色": 33090,
+      "cwd": 33091,
+      "toc": 33092,
+      "ngth": 33093,
+      "epoch": 33094,
+      "正如": 33095,
+      "elist": 33096,
+      "Tri": 33097,
+      "xia": 33098,
+      "▁Quantum": 33099,
+      "溃疡": 33100,
+      "▁Dh": 33101,
+      "▁manages": 33102,
+      "想去": 33103,
+      "都是在": 33104,
+      "ectin": 33105,
+      "独立的": 33106,
+      "▁Glen": 33107,
+      "展望": 33108,
+      "usalem": 33109,
+      "▁grip": 33110,
+      "▁crude": 33111,
+      "▁fractional": 33112,
+      "身影": 33113,
+      "Contains": 33114,
+      "触发": 33115,
+      "▁Edd": 33116,
+      "比较多": 33117,
+      "changing": 33118,
+      "▁dye": 33119,
+      "aphyl": 33120,
+      "\\\\$": 33121,
+      "▁Astr": 33122,
+      "▁romance": 33123,
+      "Sal": 33124,
+      "▁Ng": 33125,
+      "▁Jet": 33126,
+      "enerate": 33127,
+      "▁lovers": 33128,
+      "久了": 33129,
+      "IDENT": 33130,
+      "▁Flat": 33131,
+      "▁Weather": 33132,
+      "▁Benjamin": 33133,
+      "▁Relevant": 33134,
+      "夜间": 33135,
+      "达标": 33136,
+      "Est": 33137,
+      "▁Et": 33138,
+      "▁杜": 33139,
+      "▁deciding": 33140,
+      "▁Prefecture": 33141,
+      "▁aneurys": 33142,
+      "–\\": 33143,
+      "复数": 33144,
+      "tlet": 33145,
+      "FX": 33146,
+      "`);\\": 33147,
+      "encoded": 33148,
+      "andidates": 33149,
+      "▁unstable": 33150,
+      "▁unrelated": 33151,
+      "事宜": 33152,
+      "版本的": 33153,
+      "他会": 33154,
+      "▁welcomed": 33155,
+      "itate": 33156,
+      "$(\"#": 33157,
+      "▁epithelium": 33158,
+      ">::": 33159,
+      "土耳其": 33160,
+      "▁offense": 33161,
+      "▁atrial": 33162,
+      "整齐": 33163,
+      "emb": 33164,
+      "idation": 33165,
+      "沙漠": 33166,
+      "▁peri": 33167,
+      "▁vibration": 33168,
+      "分布于": 33169,
+      "nNotable": 33170,
+      "▁promotional": 33171,
+      "djangoproject": 33172,
+      "webs": 33173,
+      "▁inquiry": 33174,
+      "代表团": 33175,
+      "sembles": 33176,
+      "脉冲": 33177,
+      "酸盐": 33178,
+      "▁Suz": 33179,
+      "富含": 33180,
+      "say": 33181,
+      "最后一个": 33182,
+      "Classes": 33183,
+      "复发": 33184,
+      "底线": 33185,
+      "捐赠": 33186,
+      "istoric": 33187,
+      "▁Motion": 33188,
+      "相继": 33189,
+      "▁Kur": 33190,
+      "非要": 33191,
+      "cx": 33192,
+      "▁deemed": 33193,
+      "nPersonnel": 33194,
+      "的任务": 33195,
+      "lv": 33196,
+      "据此解答": 33197,
+      "▁Pri": 33198,
+      "▁benz": 33199,
+      "ortium": 33200,
+      "▁Batt": 33201,
+      "下车": 33202,
+      "只好": 33203,
+      "▁harass": 33204,
+      "tie": 33205,
+      "发扬": 33206,
+      "在对": 33207,
+      "大同": 33208,
+      "难免": 33209,
+      "▁retin": 33210,
+      "meth": 33211,
+      "ophila": 33212,
+      "▁lighter": 33213,
+      "黄瓜": 33214,
+      "监督管理": 33215,
+      "▁decode": 33216,
+      "形式的": 33217,
+      "nPublished": 33218,
+      "从严": 33219,
+      "信息系统": 33220,
+      "▁WA": 33221,
+      "Backend": 33222,
+      "他对": 33223,
+      "的后": 33224,
+      "nsh": 33225,
+      "tid": 33226,
+      "undo": 33227,
+      "大酒店": 33228,
+      "▁lately": 33229,
+      "隶属于": 33230,
+      "▁miR": 33231,
+      "少数民族": 33232,
+      "starter": 33233,
+      "车身": 33234,
+      "较少": 33235,
+      "▁provisions": 33236,
+      "面膜": 33237,
+      "Images": 33238,
+      "一事": 33239,
+      "Ann": 33240,
+      "▁protects": 33241,
+      "保管": 33242,
+      "ounge": 33243,
+      "ти": 33244,
+      "原油": 33245,
+      "acin": 33246,
+      "mented": 33247,
+      "Reducer": 33248,
+      "▁approaching": 33249,
+      "ollo": 33250,
+      "▁banner": 33251,
+      "▁strand": 33252,
+      "兑换": 33253,
+      "Ray": 33254,
+      "DELETE": 33255,
+      "磨损": 33256,
+      "▁reign": 33257,
+      "▁labeling": 33258,
+      "rys": 33259,
+      "▁Hun": 33260,
+      "irements": 33261,
+      "一首": 33262,
+      "机械设备": 33263,
+      "▁hoped": 33264,
+      "”[": 33265,
+      "病因": 33266,
+      "VT": 33267,
+      "冷水": 33268,
+      "衔接": 33269,
+      "关注的": 33270,
+      "教育教学": 33271,
+      "年获": 33272,
+      "▁concentrate": 33273,
+      "▁acted": 33274,
+      "判决确定之日起计算": 33275,
+      "MAN": 33276,
+      "sidebar": 33277,
+      "造价": 33278,
+      "▁Maj": 33279,
+      "第一步": 33280,
+      "dan": 33281,
+      "Radius": 33282,
+      "artist": 33283,
+      "doctoral": 33284,
+      "▁systematically": 33285,
+      "▁phon": 33286,
+      "分级": 33287,
+      "吉祥": 33288,
+      "四个月": 33289,
+      "代数式": 33290,
+      "的倍数": 33291,
+      "▁diets": 33292,
+      "▁leisure": 33293,
+      "▁Furniture": 33294,
+      "双眼": 33295,
+      "无期徒刑": 33296,
+      "▁Apply": 33297,
+      "▁controversy": 33298,
+      "无意": 33299,
+      "填充": 33300,
+      "等你": 33301,
+      "▁diagrams": 33302,
+      "可以去": 33303,
+      "▁organizing": 33304,
+      "incinnati": 33305,
+      "风流": 33306,
+      "▁,\\": 33307,
+      "nTransport": 33308,
+      "blind": 33309,
+      "▁Robin": 33310,
+      "Temp": 33311,
+      "jackson": 33312,
+      "▁avoided": 33313,
+      "▁viability": 33314,
+      "▁transforming": 33315,
+      "▁Comments": 33316,
+      "履行生效法律文书确定的义务": 33317,
+      "住在": 33318,
+      "部分的": 33319,
+      "▁mM": 33320,
+      "▁Luke": 33321,
+      "▁Homes": 33322,
+      "DataSource": 33323,
+      "剖析": 33324,
+      "gom": 33325,
+      "▁Rust": 33326,
+      "▁sponsor": 33327,
+      "原件": 33328,
+      "▁pall": 33329,
+      "患儿": 33330,
+      "秋天": 33331,
+      "tname": 33332,
+      "▁defending": 33333,
+      "NAB": 33334,
+      "phant": 33335,
+      "###\\": 33336,
+      "EGIN": 33337,
+      "在法定期限内": 33338,
+      "▁predictor": 33339,
+      "▁Christians": 33340,
+      "▁Investment": 33341,
+      "不尽": 33342,
+      "cms": 33343,
+      "没必要": 33344,
+      "给孩子": 33345,
+      "我不是": 33346,
+      "isciplinary": 33347,
+      "浓郁": 33348,
+      "▁所以": 33349,
+      "▁engr": 33350,
+      "▁funeral": 33351,
+      "▁Jour": 33352,
+      "▁cows": 33353,
+      "reading": 33354,
+      "▁Michelle": 33355,
+      "▁geographic": 33356,
+      "leukin": 33357,
+      "企业和": 33358,
+      "▁Berkeley": 33359,
+      "▁\\'{": 33360,
+      "鲜花": 33361,
+      "的味道": 33362,
+      "信访": 33363,
+      "▁hints": 33364,
+      "tnew": 33365,
+      "▁nutrient": 33366,
+      "读取": 33367,
+      "▁Meet": 33368,
+      "▁denotes": 33369,
+      "年华": 33370,
+      "▁设施": 33371,
+      "▁parental": 33372,
+      "(<": 33373,
+      "大方": 33374,
+      "发表于起点网": 33375,
+      "▁incorporation": 33376,
+      "应诉": 33377,
+      "薪资": 33378,
+      "▁cob": 33379,
+      "▁arena": 33380,
+      "车牌": 33381,
+      "辞典": 33382,
+      "的支持": 33383,
+      "sock": 33384,
+      "stri": 33385,
+      "▁judg": 33386,
+      "商会": 33387,
+      "strategy": 33388,
+      "uu": 33389,
+      "产值": 33390,
+      "支架": 33391,
+      "nThose": 33392,
+      "▁dressing": 33393,
+      "召唤": 33394,
+      "bid": 33395,
+      "cmp": 33396,
+      "分析过程": 33397,
+      "▁unity": 33398,
+      "▁remed": 33399,
+      "▁Nur": 33400,
+      "▁Recomm": 33401,
+      "▁brace": 33402,
+      "vit": 33403,
+      "ivia": 33404,
+      "▁visa": 33405,
+      "yang": 33406,
+      "就去": 33407,
+      "践行": 33408,
+      "建设和": 33409,
+      "▁imposed": 33410,
+      "▁occupation": 33411,
+      "▁exclude": 33412,
+      "亲属": 33413,
+      "心中的": 33414,
+      "arette": 33415,
+      "▁angry": 33416,
+      "▁intrig": 33417,
+      "▁Bengal": 33418,
+      "▁明": 33419,
+      "纠正": 33420,
+      "▁consuming": 33421,
+      "rochemical": 33422,
+      "性与": 33423,
+      "持久": 33424,
+      "$(\\'#": 33425,
+      "加减": 33426,
+      "▁laundry": 33427,
+      ":]": 33428,
+      "RG": 33429,
+      "但由于": 33430,
+      "must": 33431,
+      "fileName": 33432,
+      "▁Computing": 33433,
+      "的品牌": 33434,
+      "注明": 33435,
+      "▁bowel": 33436,
+      "▁Length": 33437,
+      "▁Exercise": 33438,
+      "信赖": 33439,
+      "techo": 33440,
+      "ophosph": 33441,
+      "tick": 33442,
+      "就说": 33443,
+      "Constraint": 33444,
+      "▁trafficking": 33445,
+      "落在": 33446,
+      "posium": 33447,
+      "▁girlfriend": 33448,
+      "这套": 33449,
+      "AIN": 33450,
+      "▁\"{": 33451,
+      "▁Explorer": 33452,
+      "金融借款合同纠纷一案": 33453,
+      "▁physiology": 33454,
+      "▁priest": 33455,
+      "COLOR": 33456,
+      "▁cortic": 33457,
+      "UTO": 33458,
+      "▁awk": 33459,
+      "▁granul": 33460,
+      "Cat": 33461,
+      "本案的": 33462,
+      "▁FIFA": 33463,
+      "▁Luis": 33464,
+      "iliates": 33465,
+      "Period": 33466,
+      "ndecl": 33467,
+      "NaN": 33468,
+      "▁accepts": 33469,
+      "领导小组": 33470,
+      "在这种": 33471,
+      "字体": 33472,
+      "当我": 33473,
+      "▁coordinated": 33474,
+      "▁ga": 33475,
+      "大概率": 33476,
+      "nfi": 33477,
+      "ebra": 33478,
+      "▁neighbourhood": 33479,
+      "($_": 33480,
+      "does": 33481,
+      "有利": 33482,
+      "饮水": 33483,
+      "▁Cad": 33484,
+      "▁Helper": 33485,
+      "恰好": 33486,
+      "高三": 33487,
+      "▁lumin": 33488,
+      "石化": 33489,
+      "制药": 33490,
+      "树的": 33491,
+      "诉前": 33492,
+      "▁argues": 33493,
+      "做得": 33494,
+      "冲洗": 33495,
+      "行使": 33496,
+      "dehyde": 33497,
+      "▁refined": 33498,
+      "▁nomination": 33499,
+      "履行债务": 33500,
+      "acious": 33501,
+      "▁niche": 33502,
+      "▁tubes": 33503,
+      "对的": 33504,
+      "入住时间": 33505,
+      "▁mell": 33506,
+      "当成": 33507,
+      "{{{": 33508,
+      "▁ONE": 33509,
+      "▁aeros": 33510,
+      "uclidean": 33511,
+      "▁democracy": 33512,
+      "▁TIME": 33513,
+      "▁accessibility": 33514,
+      "书的": 33515,
+      "古人": 33516,
+      "向右": 33517,
+      "离职": 33518,
+      "eni": 33519,
+      "在前": 33520,
+      "原材料": 33521,
+      "▁Fri": 33522,
+      "▁NON": 33523,
+      "▁tears": 33524,
+      "山的": 33525,
+      "菩萨": 33526,
+      "新生儿": 33527,
+      "indices": 33528,
+      "▁warming": 33529,
+      "UPDATE": 33530,
+      "▁за": 33531,
+      "▁commut": 33532,
+      "nExplanation": 33533,
+      "很小": 33534,
+      "雕塑": 33535,
+      "民俗": 33536,
+      "点到": 33537,
+      "▁surre": 33538,
+      "▁Accept": 33539,
+      "悠悠": 33540,
+      "答题": 33541,
+      "elong": 33542,
+      "▁Clinic": 33543,
+      "农历": 33544,
+      "ojo": 33545,
+      "一下子": 33546,
+      "的认识": 33547,
+      "自家": 33548,
+      "可以从轻处罚": 33549,
+      "宣传部": 33550,
+      "防晒": 33551,
+      "▁];": 33552,
+      "▁Ign": 33553,
+      "dotnet": 33554,
+      "gue": 33555,
+      "nChe": 33556,
+      "▁作品名称": 33557,
+      "▁northeast": 33558,
+      "门票": 33559,
+      "申请撤回起诉": 33560,
+      "信托": 33561,
+      "▁人物经历": 33562,
+      "▁balloon": 33563,
+      "solution": 33564,
+      "东南亚": 33565,
+      "neath": 33566,
+      "▁physic": 33567,
+      "特尔": 33568,
+      "▁Billy": 33569,
+      "Ti": 33570,
+      "▁ja": 33571,
+      "meter": 33572,
+      "毛病": 33573,
+      "慰问": 33574,
+      "▁Bin": 33575,
+      "wright": 33576,
+      "▁相关条款": 33577,
+      "manual": 33578,
+      "检公诉": 33579,
+      "SEO": 33580,
+      "我国的": 33581,
+      "观察组": 33582,
+      "经济增长": 33583,
+      "▁premises": 33584,
+      "aac": 33585,
+      "▁sofa": 33586,
+      "▁accent": 33587,
+      "扣划": 33588,
+      "▁cocon": 33589,
+      "清华大学": 33590,
+      "onChange": 33591,
+      "▁traded": 33592,
+      "合一": 33593,
+      "▁gloss": 33594,
+      "records": 33595,
+      "▁东": 33596,
+      "其中有": 33597,
+      "▁mush": 33598,
+      "▁snapshot": 33599,
+      "▁compositions": 33600,
+      "Seq": 33601,
+      "nState": 33602,
+      "▁tricks": 33603,
+      "▁semester": 33604,
+      "当局": 33605,
+      "▁Evidence": 33606,
+      "▁kilometers": 33607,
+      "第五十二条": 33608,
+      "国际化": 33609,
+      "▁beloved": 33610,
+      "STATE": 33611,
+      "▁RS": 33612,
+      "举行的": 33613,
+      "▁convicted": 33614,
+      "▁integrals": 33615,
+      "分组": 33616,
+      "▁Watson": 33617,
+      "瑞典": 33618,
+      "ighting": 33619,
+      "▁lectures": 33620,
+      "▁四": 33621,
+      "▁vapor": 33622,
+      "qi": 33623,
+      "昆虫": 33624,
+      "媒介": 33625,
+      "Once": 33626,
+      "了解一下": 33627,
+      "▁allergic": 33628,
+      "▁outlets": 33629,
+      "走访": 33630,
+      "topics": 33631,
+      "▁waist": 33632,
+      "夫妇": 33633,
+      "就连": 33634,
+      "nOrgan": 33635,
+      "开花": 33636,
+      "LAY": 33637,
+      "Transfer": 33638,
+      "Inc": 33639,
+      "▁maj": 33640,
+      "▁prizes": 33641,
+      "是被": 33642,
+      "▁filmm": 33643,
+      "友情": 33644,
+      "我们需要": 33645,
+      ";(": 33646,
+      "决定了": 33647,
+      "他们还": 33648,
+      "探讨了": 33649,
+      "lict": 33650,
+      "日登记结婚": 33651,
+      "仍有": 33652,
+      "向东": 33653,
+      "▁Lions": 33654,
+      "命中": 33655,
+      "▁controllers": 33656,
+      "面料": 33657,
+      "宪法": 33658,
+      "▁Julia": 33659,
+      "▁ti": 33660,
+      "▁favorites": 33661,
+      "太过": 33662,
+      "拒不": 33663,
+      "本周": 33664,
+      "▁hed": 33665,
+      "大全": 33666,
+      "\\\\\",": 33667,
+      "oglob": 33668,
+      "▁algebras": 33669,
+      "痕迹": 33670,
+      "羊肉": 33671,
+      "也可能": 33672,
+      "▁motorcycle": 33673,
+      "▁interviewed": 33674,
+      "走路": 33675,
+      "nSign": 33676,
+      "▁Chron": 33677,
+      "▁essays": 33678,
+      "▁ramp": 33679,
+      "ocytosis": 33680,
+      "共建": 33681,
+      "近几年": 33682,
+      "itatively": 33683,
+      "ermark": 33684,
+      "▁positioned": 33685,
+      "绝大多数": 33686,
+      "▁timed": 33687,
+      "feld": 33688,
+      "raham": 33689,
+      "每一次": 33690,
+      "dv": 33691,
+      "洗衣机": 33692,
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁": 33693,
+      "▁cele": 33694,
+      "众所周知": 33695,
+      "listener": 33696,
+      "▁WHO": 33697,
+      "▁appreciation": 33698,
+      "▁GHz": 33699,
+      "▁transduction": 33700,
+      "充分利用": 33701,
+      "将成为": 33702,
+      "▁Roberts": 33703,
+      "▁Emeritus": 33704,
+      "▁gravitational": 33705,
+      "很喜欢": 33706,
+      "▁Enh": 33707,
+      "▁Rather": 33708,
+      "▁purely": 33709,
+      "▁mailing": 33710,
+      "开封": 33711,
+      "必定": 33712,
+      "▁Chain": 33713,
+      "成立了": 33714,
+      "强烈的": 33715,
+      "秋风": 33716,
+      "▁obstacles": 33717,
+      "yc": 33718,
+      "奥斯": 33719,
+      "▁appet": 33720,
+      "▁%>": 33721,
+      "engu": 33722,
+      "马来西亚": 33723,
+      "▁Tigers": 33724,
+      "大海": 33725,
+      "清代": 33726,
+      "arently": 33727,
+      "digital": 33728,
+      "▁Luck": 33729,
+      "opoietic": 33730,
+      "▁dressed": 33731,
+      "老大": 33732,
+      "打卡": 33733,
+      "空军": 33734,
+      "老鼠": 33735,
+      "itary": 33736,
+      "▁honour": 33737,
+      "acterial": 33738,
+      "Typ": 33739,
+      "总书记": 33740,
+      "对该犯": 33741,
+      "▁practicing": 33742,
+      "far": 33743,
+      "ocular": 33744,
+      "▁Rivers": 33745,
+      "男的": 33746,
+      "writers": 33747,
+      "manage": 33748,
+      "▁bean": 33749,
+      "fixture": 33750,
+      "▁Kirk": 33751,
+      "▁supplements": 33752,
+      "nPS": 33753,
+      "elles": 33754,
+      "pressed": 33755,
+      "年第": 33756,
+      "▁accessing": 33757,
+      "街头": 33758,
+      "了被执行人": 33759,
+      "▁PB": 33760,
+      "▁Democrat": 33761,
+      "<\\": 33762,
+      "卖出": 33763,
+      "Commands": 33764,
+      "nVery": 33765,
+      "▁flowing": 33766,
+      "▁friendship": 33767,
+      "IPS": 33768,
+      "}}^{": 33769,
+      "▁rgb": 33770,
+      "▁Buddh": 33771,
+      "研究生院": 33772,
+      "touch": 33773,
+      "▁Extract": 33774,
+      "▁Iceland": 33775,
+      "时长": 33776,
+      "论证": 33777,
+      "CLASS": 33778,
+      "▁initiate": 33779,
+      "tub": 33780,
+      "▁cooperative": 33781,
+      "FilePath": 33782,
+      "老虎": 33783,
+      "给出了": 33784,
+      "▁Chelsea": 33785,
+      "▁HOLDERS": 33786,
+      "▁fundament": 33787,
+      "市中": 33788,
+      "监理": 33789,
+      "Sound": 33790,
+      "本院依法适用简易程序": 33791,
+      "ecom": 33792,
+      "▁throughput": 33793,
+      "emade": 33794,
+      "gateway": 33795,
+      "瓷砖": 33796,
+      "faster": 33797,
+      "▁honored": 33798,
+      "▁Expected": 33799,
+      "▁):\\": 33800,
+      "硕士学位": 33801,
+      "▁拉丁": 33802,
+      "azole": 33803,
+      "eclipse": 33804,
+      "科室": 33805,
+      "一律": 33806,
+      "想说": 33807,
+      "AVA": 33808,
+      "医科大学": 33809,
+      "可以根据": 33810,
+      "乘以": 33811,
+      "igating": 33812,
+      "erget": 33813,
+      "差点": 33814,
+      "贯穿": 33815,
+      "KO": 33816,
+      "▁spat": 33817,
+      "看你": 33818,
+      "的比例": 33819,
+      "cedure": 33820,
+      "▁Fisher": 33821,
+      "▁whites": 33822,
+      "中途": 33823,
+      "我也是": 33824,
+      "▁Notice": 33825,
+      "怪物": 33826,
+      "▁Stay": 33827,
+      "coverage": 33828,
+      "South": 33829,
+      "▁absorb": 33830,
+      "games": 33831,
+      "Low": 33832,
+      "洋葱": 33833,
+      "▁Until": 33834,
+      "▁Bernard": 33835,
+      "▁adventures": 33836,
+      "叔叔": 33837,
+      "报表": 33838,
+      "essa": 33839,
+      "农村经济": 33840,
+      "Desktop": 33841,
+      "大米": 33842,
+      "duino": 33843,
+      "区人民检察院以": 33844,
+      "▁Various": 33845,
+      "▁appropriately": 33846,
+      "生殖": 33847,
+      "想的": 33848,
+      "imm": 33849,
+      "refs": 33850,
+      "▁REST": 33851,
+      "wi": 33852,
+      "▁Prep": 33853,
+      "▁Github": 33854,
+      "▁Picture": 33855,
+      "DateTimeField": 33856,
+      "nPaul": 33857,
+      "不来": 33858,
+      "人为本": 33859,
+      "▁Ran": 33860,
+      "▁athletic": 33861,
+      "价款": 33862,
+      "已经在": 33863,
+      "phantom": 33864,
+      "▁Baptist": 33865,
+      "▁ids": 33866,
+      "handlers": 33867,
+      "▁Regular": 33868,
+      "▁Extra": 33869,
+      "Ma": 33870,
+      "天上": 33871,
+      "▁Equation": 33872,
+      "退货": 33873,
+      "NUM": 33874,
+      "▁caf": 33875,
+      "管理员": 33876,
+      "quantity": 33877,
+      "已经有": 33878,
+      "的实现": 33879,
+      "▁Austrian": 33880,
+      "哪种": 33881,
+      "所求": 33882,
+      "用到的": 33883,
+      "viewport": 33884,
+      "▁hypoxia": 33885,
+      "▁guarantees": 33886,
+      "eward": 33887,
+      "signature": 33888,
+      "▁Pil": 33889,
+      "还是比较": 33890,
+      "icked": 33891,
+      "WP": 33892,
+      "▁冯": 33893,
+      "Lines": 33894,
+      "▁Agriculture": 33895,
+      "我最": 33896,
+      "缺失": 33897,
+      "以色列": 33898,
+      "xit": 33899,
+      "结晶": 33900,
+      "副总经理": 33901,
+      "▁storing": 33902,
+      "▁NM": 33903,
+      "Running": 33904,
+      "▁tx": 33905,
+      "▁\\\\(\\\\": 33906,
+      "海岸": 33907,
+      "▁trainer": 33908,
+      "渲染": 33909,
+      "文案": 33910,
+      "这些都是": 33911,
+      "ussels": 33912,
+      "他用": 33913,
+      "robe": 33914,
+      "cad": 33915,
+      "再见": 33916,
+      "▁burg": 33917,
+      "▁Differential": 33918,
+      "▁surprisingly": 33919,
+      "▁unprecedented": 33920,
+      "Small": 33921,
+      "外资": 33922,
+      "方法和": 33923,
+      "▁Harm": 33924,
+      "▁evolving": 33925,
+      "usted": 33926,
+      "▁Points": 33927,
+      "Sil": 33928,
+      "▁qualification": 33929,
+      "外地": 33930,
+      "失眠": 33931,
+      "▁Updated": 33932,
+      "上映": 33933,
+      "NODE": 33934,
+      "▁Nar": 33935,
+      "accounts": 33936,
+      "▁drawings": 33937,
+      "]>": 33938,
+      "CONF": 33939,
+      "的艺术": 33940,
+      "执行条件": 33941,
+      "%~": 33942,
+      "%%%%": 33943,
+      "Then": 33944,
+      "staff": 33945,
+      "轮回": 33946,
+      "uvant": 33947,
+      "▁Plot": 33948,
+      "ци": 33949,
+      "nJapanese": 33950,
+      "二个月": 33951,
+      "▁pressed": 33952,
+      "号案": 33953,
+      "venile": 33954,
+      "督导": 33955,
+      "注意力": 33956,
+      "▁Political": 33957,
+      "▁specimen": 33958,
+      "亚马": 33959,
+      "的一般": 33960,
+      "表示为": 33961,
+      "iji": 33962,
+      "Fast": 33963,
+      "▁buses": 33964,
+      "▁alloy": 33965,
+      "▁examinations": 33966,
+      "▁referral": 33967,
+      "▁Handle": 33968,
+      "诊疗": 33969,
+      "儿女": 33970,
+      "▁Consumer": 33971,
+      "▁excitation": 33972,
+      "不愿": 33973,
+      "SUV": 33974,
+      "termin": 33975,
+      "ISS": 33976,
+      "arrass": 33977,
+      "▁Surgery": 33978,
+      "零点": 33979,
+      "▁medals": 33980,
+      "▁protests": 33981,
+      "脑袋": 33982,
+      "elp": 33983,
+      "选项错误": 33984,
+      "本院于同": 33985,
+      "▁fibroblasts": 33986,
+      "▁Holiday": 33987,
+      "▁coil": 33988,
+      "▁lungs": 33989,
+      "autiful": 33990,
+      "科员": 33991,
+      "ramid": 33992,
+      "▁Schedule": 33993,
+      "的意见": 33994,
+      "▁temporarily": 33995,
+      "nWatch": 33996,
+      "积极性": 33997,
+      "等因素": 33998,
+      "onucle": 33999,
+      "▁Observable": 34000,
+      "nAv": 34001,
+      "▁dots": 34002,
+      "条第一款之规定": 34003,
+      "visor": 34004,
+      "ulmonary": 34005,
+      "▁proceeds": 34006,
+      "▁annotation": 34007,
+      "七八": 34008,
+      "▁Gate": 34009,
+      "▁notified": 34010,
+      "▁Competition": 34011,
+      "离家": 34012,
+      "办公厅": 34013,
+      "▁theft": 34014,
+      "yu": 34015,
+      "事先": 34016,
+      "简化": 34017,
+      "▁perc": 34018,
+      "定向": 34019,
+      "▁aberr": 34020,
+      "▁cores": 34021,
+      "▁TA": 34022,
+      "▁synerg": 34023,
+      "篇小说": 34024,
+      "▁obsc": 34025,
+      "establish": 34026,
+      "verts": 34027,
+      "nSchool": 34028,
+      "▁soundtrack": 34029,
+      "的体积": 34030,
+      "的理论": 34031,
+      "研究方向": 34032,
+      "艰苦": 34033,
+      "▁√": 34034,
+      "idding": 34035,
+      "和一个": 34036,
+      "▁Orlando": 34037,
+      "长久": 34038,
+      "▁Solid": 34039,
+      "整天": 34040,
+      "十九章": 34041,
+      "下游": 34042,
+      "味的": 34043,
+      "将继续": 34044,
+      "posure": 34045,
+      "▁candy": 34046,
+      "Describe": 34047,
+      "freq": 34048,
+      "▁spotted": 34049,
+      "orum": 34050,
+      "▁Haz": 34051,
+      "desktop": 34052,
+      "信用卡中心": 34053,
+      "prising": 34054,
+      "▁hierarchy": 34055,
+      "射线": 34056,
+      "hall": 34057,
+      "共同生活": 34058,
+      "ievable": 34059,
+      "分明": 34060,
+      "▁directories": 34061,
+      "使用了": 34062,
+      "nMichael": 34063,
+      "▁overflow": 34064,
+      "▁clouds": 34065,
+      "▁worlds": 34066,
+      "▁Turner": 34067,
+      "▁Generate": 34068,
+      "bib": 34069,
+      "很高的": 34070,
+      "的用户": 34071,
+      "Country": 34072,
+      "rieved": 34073,
+      "买到": 34074,
+      "enton": 34075,
+      "itten": 34076,
+      "SULT": 34077,
+      "▁punch": 34078,
+      "▁enlarg": 34079,
+      "▁Academic": 34080,
+      "Mg": 34081,
+      "▁Steam": 34082,
+      "高低": 34083,
+      "处在": 34084,
+      "getValue": 34085,
+      "可以将": 34086,
+      "▁Certified": 34087,
+      "所得税": 34088,
+      "▁xen": 34089,
+      "报告财产": 34090,
+      "重要参数": 34091,
+      "▁reflecting": 34092,
+      "埃及": 34093,
+      "OFF": 34094,
+      "检公诉刑诉": 34095,
+      "imi": 34096,
+      "▁barg": 34097,
+      "▁Hack": 34098,
+      "▁corporations": 34099,
+      "okia": 34100,
+      "}}\\\\)": 34101,
+      "UTE": 34102,
+      "的可能性": 34103,
+      "icketer": 34104,
+      "▁Collabor": 34105,
+      "辩护人": 34106,
+      "Updated": 34107,
+      "▁rm": 34108,
+      "▁minerals": 34109,
+      "为本": 34110,
+      "测评": 34111,
+      "筹码": 34112,
+      "cedes": 34113,
+      "▁sweat": 34114,
+      "尔德": 34115,
+      "徐州": 34116,
+      "▁energet": 34117,
+      "溶解": 34118,
+      "行走": 34119,
+      "这座": 34120,
+      "mine": 34121,
+      "▁Hab": 34122,
+      "正方形的": 34123,
+      "lishes": 34124,
+      "▁informal": 34125,
+      "夸张": 34126,
+      "LAN": 34127,
+      "▁Adobe": 34128,
+      "iton": 34129,
+      "▁Eagles": 34130,
+      "之类": 34131,
+      "白糖": 34132,
+      "▁Shar": 34133,
+      "organic": 34134,
+      "彰显": 34135,
+      "▁fires": 34136,
+      "涨幅": 34137,
+      "风暴": 34138,
+      "不吃": 34139,
+      "tB": 34140,
+      "不良反应": 34141,
+      "佛山": 34142,
+      "▁Jav": 34143,
+      "▁Johnny": 34144,
+      "sharp": 34145,
+      "▁grocery": 34146,
+      "了的": 34147,
+      "也不要": 34148,
+      "nRound": 34149,
+      "ENTS": 34150,
+      "▁resigned": 34151,
+      "不该": 34152,
+      "ardless": 34153,
+      "交通安全": 34154,
+      "▁terrain": 34155,
+      "悄悄": 34156,
+      "anco": 34157,
+      "▁Viol": 34158,
+      "▁Advisory": 34159,
+      "▁Broadway": 34160,
+      "InstanceState": 34161,
+      "历经": 34162,
+      "▁retrieval": 34163,
+      "▁precipitation": 34164,
+      "▁Vue": 34165,
+      "nUnit": 34166,
+      "▁Redist": 34167,
+      "Snapshot": 34168,
+      "▁essence": 34169,
+      "▁contamination": 34170,
+      "layouts": 34171,
+      "ymal": 34172,
+      "▁luxurious": 34173,
+      "中共中央": 34174,
+      "\\']:\\": 34175,
+      "▁applicant": 34176,
+      "▁declaration": 34177,
+      "▁APIs": 34178,
+      "侵略": 34179,
+      "原唱": 34180,
+      "▁HLA": 34181,
+      "▁Setting": 34182,
+      "▁methylation": 34183,
+      "的操作": 34184,
+      "aconda": 34185,
+      "eri": 34186,
+      "Grad": 34187,
+      "thand": 34188,
+      "▁Looks": 34189,
+      "预报": 34190,
+      "▁stressed": 34191,
+      "▁MAC": 34192,
+      "科幻": 34193,
+      "nLeg": 34194,
+      "服务业": 34195,
+      "的前提下": 34196,
+      "ADDRESS": 34197,
+      "Mc": 34198,
+      "([\"": 34199,
+      "▁Err": 34200,
+      "失效": 34201,
+      "▁Sz": 34202,
+      "▁ib": 34203,
+      "分之一": 34204,
+      "▁Speaker": 34205,
+      "▁prompted": 34206,
+      "▁referenced": 34207,
+      "pyp": 34208,
+      "▁Kos": 34209,
+      "▁evaluations": 34210,
+      "ottage": 34211,
+      "▁ni": 34212,
+      "inkle": 34213,
+      "veolar": 34214,
+      "Research": 34215,
+      "owns": 34216,
+      "归案后": 34217,
+      "brev": 34218,
+      "getId": 34219,
+      "wig": 34220,
+      "了一次": 34221,
+      "因与被上诉人": 34222,
+      "Pin": 34223,
+      "district": 34224,
+      "▁unfold": 34225,
+      "▁Companies": 34226,
+      "PAGE": 34227,
+      "▁analyse": 34228,
+      "▁jumping": 34229,
+      "灵敏": 34230,
+      "为进一步": 34231,
+      "一脸": 34232,
+      "菊花": 34233,
+      "幂的": 34234,
+      "演艺": 34235,
+      "▁Vec": 34236,
+      "▁proves": 34237,
+      "文化和": 34238,
+      "微微": 34239,
+      "▁Karl": 34240,
+      "点的距离": 34241,
+      "appers": 34242,
+      "Michael": 34243,
+      "nFour": 34244,
+      "未作": 34245,
+      "聚合": 34246,
+      "多功能": 34247,
+      "快速发展": 34248,
+      "Stage": 34249,
+      "creator": 34250,
+      "▁removes": 34251,
+      "exam": 34252,
+      "▁licensing": 34253,
+      "清醒": 34254,
+      "▁ministers": 34255,
+      "hline": 34256,
+      "▁STAT": 34257,
+      "▁formulated": 34258,
+      "DG": 34259,
+      "▁nuts": 34260,
+      "先天": 34261,
+      "南山": 34262,
+      "Lat": 34263,
+      "主要包括": 34264,
+      "▁Dental": 34265,
+      "▁fossil": 34266,
+      "ém": 34267,
+      "ginx": 34268,
+      "预付": 34269,
+      "beit": 34270,
+      "▁Associated": 34271,
+      "备受": 34272,
+      "Final": 34273,
+      "▁modeled": 34274,
+      "阻碍": 34275,
+      "第一百四十五条之规定": 34276,
+      "▁compartment": 34277,
+      "片段": 34278,
+      "nrouter": 34279,
+      "▁Maz": 34280,
+      "制造有限公司": 34281,
+      "出手": 34282,
+      "的撤诉申请": 34283,
+      "▁enemies": 34284,
+      "▁石": 34285,
+      "面积是": 34286,
+      "▁Moroc": 34287,
+      "uckland": 34288,
+      "istle": 34289,
+      "认真遵守监规": 34290,
+      "书店": 34291,
+      "▁pest": 34292,
+      "▁reductions": 34293,
+      "互联": 34294,
+      "](#": 34295,
+      "很多时候": 34296,
+      "boxes": 34297,
+      "▁artic": 34298,
+      "床铺": 34299,
+      "▁Cha": 34300,
+      "Writ": 34301,
+      "conda": 34302,
+      "ei": 34303,
+      "▁Barr": 34304,
+      "每个月": 34305,
+      "▁conductivity": 34306,
+      "月起": 34307,
+      "Delay": 34308,
+      "▁blade": 34309,
+      "▁Syrian": 34310,
+      "▁fucking": 34311,
+      "▁epidemic": 34312,
+      "光泽": 34313,
+      "▁Lion": 34314,
+      "全文": 34315,
+      "导数": 34316,
+      "调理": 34317,
+      "钢材": 34318,
+      "著作权": 34319,
+      "经原告多次": 34320,
+      "▁vegetable": 34321,
+      "▁知": 34322,
+      "packet": 34323,
+      "▁chromatin": 34324,
+      "学报": 34325,
+      "滴滴": 34326,
+      "nTaxa": 34327,
+      "市人民政府": 34328,
+      "▁Resume": 34329,
+      "Facebook": 34330,
+      "Generated": 34331,
+      "▁Beth": 34332,
+      "▁Summit": 34333,
+      "另查明": 34334,
+      "若干规定": 34335,
+      "▁Browser": 34336,
+      "医保": 34337,
+      "是真": 34338,
+      "落叶": 34339,
+      "OPY": 34340,
+      "ainted": 34341,
+      "料理": 34342,
+      "实用的": 34343,
+      "▁queen": 34344,
+      "▁Murphy": 34345,
+      "质数": 34346,
+      "▁pec": 34347,
+      "结婚登记": 34348,
+      "otrans": 34349,
+      "▁dresses": 34350,
+      "▁(),": 34351,
+      "basename": 34352,
+      "}>": 34353,
+      "nCall": 34354,
+      "}}\"": 34355,
+      "很强的": 34356,
+      "▁embrace": 34357,
+      "却又": 34358,
+      "▁Universe": 34359,
+      "nSummary": 34360,
+      "来到了": 34361,
+      "Paul": 34362,
+      "唤醒": 34363,
+      "ematic": 34364,
+      "▁Hispanic": 34365,
+      "瑜伽": 34366,
+      "博士学位": 34367,
+      "acre": 34368,
+      "暗示": 34369,
+      "vim": 34370,
+      "信用社": 34371,
+      "下列情形": 34372,
+      "respond": 34373,
+      "▁tribute": 34374,
+      "工程建设": 34375,
+      "▁glomer": 34376,
+      "▁Paralym": 34377,
+      "▁regimen": 34378,
+      "余数": 34379,
+      "发货": 34380,
+      "代言": 34381,
+      "大一": 34382,
+      "▁semin": 34383,
+      "斯基": 34384,
+      "▁Sci": 34385,
+      "Fixed": 34386,
+      "减速": 34387,
+      "仍未": 34388,
+      "假的": 34389,
+      "是非": 34390,
+      "efan": 34391,
+      "\\']);\\": 34392,
+      "▁badge": 34393,
+      "▁encrypted": 34394,
+      "Times": 34395,
+      "一体的": 34396,
+      "▁Crime": 34397,
+      "氨基酸": 34398,
+      "along": 34399,
+      "ATER": 34400,
+      "▁fertil": 34401,
+      "举办的": 34402,
+      "▁addict": 34403,
+      "技术开发": 34404,
+      "星级酒店": 34405,
+      "brit": 34406,
+      "▁revealing": 34407,
+      "▁cheer": 34408,
+      "▁Yahoo": 34409,
+      "tically": 34410,
+      "播种": 34411,
+      "nthere": 34412,
+      "▁Principal": 34413,
+      "▁Issues": 34414,
+      "▁innate": 34415,
+      "的出现": 34416,
+      "名家": 34417,
+      "Mask": 34418,
+      "▁Rules": 34419,
+      "▁explor": 34420,
+      "▁fluids": 34421,
+      "▁dividing": 34422,
+      "不仅仅是": 34423,
+      "▁hypothesized": 34424,
+      "▁Vik": 34425,
+      "▁ammon": 34426,
+      "▁antit": 34427,
+      "▁joins": 34428,
+      "农作物": 34429,
+      "的经营": 34430,
+      "▁crust": 34431,
+      "ynchronous": 34432,
+      "造成了": 34433,
+      "ITLE": 34434,
+      "nFollow": 34435,
+      "▁Signal": 34436,
+      "美股": 34437,
+      "民事权利": 34438,
+      "nUnfortunately": 34439,
+      "VIEW": 34440,
+      "等领域": 34441,
+      "sided": 34442,
+      "▁Parks": 34443,
+      "edar": 34444,
+      "▁differs": 34445,
+      "▁insects": 34446,
+      "川区": 34447,
+      "▁Keith": 34448,
+      "▁calories": 34449,
+      "转速": 34450,
+      "劳动合同": 34451,
+      "▁fixtures": 34452,
+      "tos": 34453,
+      "▁annoying": 34454,
+      "小龙": 34455,
+      "▁Guinea": 34456,
+      "余篇": 34457,
+      "ndeclare": 34458,
+      "nCO": 34459,
+      "▁Kl": 34460,
+      "▁iv": 34461,
+      "与原告": 34462,
+      "科学发展": 34463,
+      "饱满": 34464,
+      "Redis": 34465,
+      "▁Marvel": 34466,
+      "▁Hungarian": 34467,
+      "yx": 34468,
+      "实数根": 34469,
+      "▁regex": 34470,
+      "摆放": 34471,
+      "▁dorsal": 34472,
+      "好吗": 34473,
+      "ambers": 34474,
+      "▁metallic": 34475,
+      "一丝": 34476,
+      "中档": 34477,
+      "]),\\": 34478,
+      "▁Prevention": 34479,
+      "答主": 34480,
+      "谈恋爱": 34481,
+      "▁Tow": 34482,
+      "▁Serv": 34483,
+      "各式": 34484,
+      "keeping": 34485,
+      "法定代表人王": 34486,
+      "iculous": 34487,
+      "ednesday": 34488,
+      "三十日内": 34489,
+      "▁规格": 34490,
+      "局副局长": 34491,
+      "▁grams": 34492,
+      "海南省": 34493,
+      "urom": 34494,
+      "▁ali": 34495,
+      "▁Spark": 34496,
+      "▁美国": 34497,
+      "▁hills": 34498,
+      "itives": 34499,
+      "Company": 34500,
+      "/](": 34501,
+      "区长": 34502,
+      "质感": 34503,
+      "RGB": 34504,
+      "▁LG": 34505,
+      "Mr": 34506,
+      "溪县": 34507,
+      "cellular": 34508,
+      "后勤": 34509,
+      "自己在": 34510,
+      "好人": 34511,
+      "新年": 34512,
+      "izen": 34513,
+      "种族": 34514,
+      "})\\\\": 34515,
+      "▁hover": 34516,
+      "▁idi": 34517,
+      "▁Numbers": 34518,
+      "这些人": 34519,
+      "stad": 34520,
+      "意图": 34521,
+      "集群": 34522,
+      "▁seq": 34523,
+      "▁ulcer": 34524,
+      "▁Egyptian": 34525,
+      "业余": 34526,
+      "▁Yu": 34527,
+      "▁multif": 34528,
+      "名校": 34529,
+      "短短": 34530,
+      "nCal": 34531,
+      "odynamics": 34532,
+      "极致": 34533,
+      "nAthlet": 34534,
+      "晶体": 34535,
+      "申请执行人可": 34536,
+      "最强": 34537,
+      "的根": 34538,
+      "ende": 34539,
+      "做人": 34540,
+      "▁Mol": 34541,
+      "▁numbered": 34542,
+      "降雨": 34543,
+      "▁Dry": 34544,
+      "Failure": 34545,
+      "为了让": 34546,
+      "ANGU": 34547,
+      "▁Oblast": 34548,
+      "phan": 34549,
+      "finished": 34550,
+      "国企": 34551,
+      "的综合": 34552,
+      "指引": 34553,
+      "时光网": 34554,
+      "▁Northwest": 34555,
+      "牌子": 34556,
+      "这不": 34557,
+      "判断出": 34558,
+      "zoom": 34559,
+      "▁unsuccess": 34560,
+      "ACHE": 34561,
+      "gatsby": 34562,
+      "berries": 34563,
+      "▁bonding": 34564,
+      "▁dispersion": 34565,
+      "▁mentor": 34566,
+      "pw": 34567,
+      ";\">": 34568,
+      "dos": 34569,
+      "等情况": 34570,
+      "水准": 34571,
+      "红的": 34572,
+      "起源": 34573,
+      "▁medieval": 34574,
+      "别说": 34575,
+      "Locale": 34576,
+      "▁pursuing": 34577,
+      "自己是": 34578,
+      "工商银行": 34579,
+      "\\':\\'": 34580,
+      "自律": 34581,
+      "Front": 34582,
+      "▁Shell": 34583,
+      "▁Rice": 34584,
+      "▁Conservation": 34585,
+      "SESS": 34586,
+      "▁midfielder": 34587,
+      "'/": 34588,
+      "进入了": 34589,
+      "平方厘米": 34590,
+      "reactive": 34591,
+      "ococ": 34592,
+      "▁expose": 34593,
+      "跟他": 34594,
+      "▁hr": 34595,
+      "总收入的": 34596,
+      "SECRE": 34597,
+      "▁cloned": 34598,
+      "hin": 34599,
+      "撰写": 34600,
+      "vote": 34601,
+      "▁MOD": 34602,
+      "iliation": 34603,
+      "▁总": 34604,
+      "残留": 34605,
+      "▁selectively": 34606,
+      "市中级人民法院于": 34607,
+      "▁fabricated": 34608,
+      "变速": 34609,
+      "笔试": 34610,
+      "文化遗产": 34611,
+      "▁Mining": 34612,
+      "processors": 34613,
+      "▁neoplas": 34614,
+      "珠海": 34615,
+      "甚至是": 34616,
+      "▁tert": 34617,
+      "dropdown": 34618,
+      "▁Analytics": 34619,
+      "的最新": 34620,
+      "▁Safe": 34621,
+      "ivated": 34622,
+      "▁estimating": 34623,
+      "长约": 34624,
+      "▁jar": 34625,
+      "icol": 34626,
+      "亮度": 34627,
+      "原价": 34628,
+      "▁Nu": 34629,
+      "▁GUI": 34630,
+      "所提供的": 34631,
+      "入了": 34632,
+      "半小时": 34633,
+      "▁Either": 34634,
+      "也为": 34635,
+      "含量为": 34636,
+      "固定的": 34637,
+      "helm": 34638,
+      "▁Dop": 34639,
+      "▁Mack": 34640,
+      "▁Vlad": 34641,
+      "人口和计划生育局": 34642,
+      "轻微": 34643,
+      "小额贷款": 34644,
+      "熟练掌握": 34645,
+      "▁Workshop": 34646,
+      "号召": 34647,
+      "河区": 34648,
+      "▁Bathroom": 34649,
+      "slf": 34650,
+      "顶尖": 34651,
+      "▁fears": 34652,
+      "▁Alfred": 34653,
+      "▁文": 34654,
+      "的经验": 34655,
+      "nFlora": 34656,
+      "MON": 34657,
+      "▁Perry": 34658,
+      "▁pharmacokin": 34659,
+      "graphql": 34660,
+      "参阅": 34661,
+      "Bel": 34662,
+      "上述事实有": 34663,
+      "▁Consequently": 34664,
+      "▁Luther": 34665,
+      "gie": 34666,
+      "所在的": 34667,
+      ")\",\\": 34668,
+      "比分": 34669,
+      "▁freeze": 34670,
+      "四种": 34671,
+      "强迫": 34672,
+      "checks": 34673,
+      "▁Starting": 34674,
+      "晋升": 34675,
+      "牛逼": 34676,
+      "脚本": 34677,
+      "承担连带": 34678,
+      "essel": 34679,
+      "▁starter": 34680,
+      "rage": 34681,
+      "▁Graduate": 34682,
+      "第一百七十": 34683,
+      "nLove": 34684,
+      "▁modal": 34685,
+      "▁Construct": 34686,
+      "二氧化": 34687,
+      "机械有限公司": 34688,
+      "▁Loading": 34689,
+      "studio": 34690,
+      "▁Sid": 34691,
+      "▁erg": 34692,
+      "otyping": 34693,
+      "和好": 34694,
+      "坏了": 34695,
+      "▁antiv": 34696,
+      "▁locale": 34697,
+      "▁Variable": 34698,
+      "小火": 34699,
+      "布鲁": 34700,
+      "多学科": 34701,
+      "▁recursive": 34702,
+      "▁TW": 34703,
+      "留学生": 34704,
+      "▁revis": 34705,
+      "的你": 34706,
+      "tfloat": 34707,
+      "申请强制执行": 34708,
+      "cid": 34709,
+      "▁emigr": 34710,
+      "▁grace": 34711,
+      "▁appealing": 34712,
+      "]][": 34713,
+      "▁cyto": 34714,
+      "nLearn": 34715,
+      "▁embryos": 34716,
+      "▁anesthesia": 34717,
+      "出色": 34718,
+      "vor": 34719,
+      "会影响": 34720,
+      "nDownload": 34721,
+      "▁cabinets": 34722,
+      "▁commander": 34723,
+      "但却": 34724,
+      "再去": 34725,
+      "十三章": 34726,
+      "odal": 34727,
+      "予以确认": 34728,
+      "nFile": 34729,
+      "}$$,": 34730,
+      "▁stamp": 34731,
+      "宜昌": 34732,
+      "▁reasonably": 34733,
+      "形态特征": 34734,
+      "▁newborn": 34735,
+      "心思": 34736,
+      "▁escal": 34737,
+      "▁Glo": 34738,
+      "▁Calendar": 34739,
+      "▁Relations": 34740,
+      "易于": 34741,
+      "lare": 34742,
+      "▁sells": 34743,
+      "失误": 34744,
+      "▁epilepsy": 34745,
+      "兔子": 34746,
+      "沿着": 34747,
+      "▁Crystal": 34748,
+      "底数": 34749,
+      "{{$": 34750,
+      "上一个": 34751,
+      "教会": 34752,
+      "neal": 34753,
+      "▁Render": 34754,
+      "▁点击": 34755,
+      "▁Mono": 34756,
+      "rien": 34757,
+      "▁Arabic": 34758,
+      "▁browsing": 34759,
+      "拿下": 34760,
+      "letters": 34761,
+      "▁tuning": 34762,
+      "▁trademark": 34763,
+      "dns": 34764,
+      "累积": 34765,
+      "的真实": 34766,
+      "gone": 34767,
+      "Rules": 34768,
+      "▁Sloven": 34769,
+      "书城": 34770,
+      "着我": 34771,
+      "mot": 34772,
+      "zel": 34773,
+      "的区别": 34774,
+      "老实": 34775,
+      "▁staging": 34776,
+      "▁teaches": 34777,
+      "▁Preliminary": 34778,
+      "的很": 34779,
+      "抄袭": 34780,
+      "跨境": 34781,
+      "低碳": 34782,
+      "墨西": 34783,
+      "够了": 34784,
+      "带给": 34785,
+      "早晨": 34786,
+      "▁Rodr": 34787,
+      "▁shame": 34788,
+      "参展": 34789,
+      "▁allele": 34790,
+      "▁日": 34791,
+      "兵团": 34792,
+      "了许多": 34793,
+      "▁march": 34794,
+      "▁VP": 34795,
+      "天马": 34796,
+      "此刻": 34797,
+      "▁CRE": 34798,
+      "▁dive": 34799,
+      "结核": 34800,
+      "bidden": 34801,
+      "Americ": 34802,
+      "▁Leadership": 34803,
+      "橄榄": 34804,
+      "▁Bir": 34805,
+      "Shader": 34806,
+      "▁Bayesian": 34807,
+      "▁Jerusalem": 34808,
+      "requirements": 34809,
+      "判决发生法律效力后交付执行": 34810,
+      "罪犯减刑": 34811,
+      "▁Legend": 34812,
+      "牢记": 34813,
+      "greater": 34814,
+      "封建": 34815,
+      "的需要": 34816,
+      "之子": 34817,
+      "▁merchand": 34818,
+      "▁elaborate": 34819,
+      "TODO": 34820,
+      "inging": 34821,
+      "▁Adding": 34822,
+      "▁browsers": 34823,
+      "▁姜": 34824,
+      "模样": 34825,
+      "为一体的": 34826,
+      "▁Villa": 34827,
+      "县人民法院于": 34828,
+      "▁basket": 34829,
+      "笑话": 34830,
+      "▁bankruptcy": 34831,
+      "Touch": 34832,
+      "▁Parkinson": 34833,
+      "不觉": 34834,
+      "小吃": 34835,
+      "袖添香网": 34836,
+      "▁Border": 34837,
+      "▁earthquake": 34838,
+      "illery": 34839,
+      "日提出减刑建议书": 34840,
+      "的一天": 34841,
+      "▁HCV": 34842,
+      "的计算": 34843,
+      "结婚证": 34844,
+      "▁Eld": 34845,
+      "adoop": 34846,
+      "Week": 34847,
+      "rosion": 34848,
+      "▁sklearn": 34849,
+      "obuf": 34850,
+      "▁Todd": 34851,
+      "全国各地": 34852,
+      "cherichia": 34853,
+      "央行": 34854,
+      "不应该": 34855,
+      "被迫": 34856,
+      "refer": 34857,
+      "▁appointments": 34858,
+      "▁ionic": 34859,
+      "的女孩": 34860,
+      "opal": 34861,
+      "▁recreational": 34862,
+      "▁coconut": 34863,
+      "▁southwest": 34864,
+      "▁cloth": 34865,
+      "swagger": 34866,
+      "ertation": 34867,
+      "▁differentiated": 34868,
+      "IOS": 34869,
+      "▁Soul": 34870,
+      "ClassName": 34871,
+      "依职权": 34872,
+      "▁shades": 34873,
+      "▁Imperial": 34874,
+      "物价": 34875,
+      "IPE": 34876,
+      "▁wax": 34877,
+      "▁northwest": 34878,
+      "▁WW": 34879,
+      "autom": 34880,
+      "光驱": 34881,
+      "cro": 34882,
+      "escape": 34883,
+      "▁advisor": 34884,
+      "viral": 34885,
+      "ès": 34886,
+      "在学校": 34887,
+      "使用寿命": 34888,
+      "▁Partial": 34889,
+      "滑动": 34890,
+      "首选": 34891,
+      "钢板": 34892,
+      "用人单位": 34893,
+      "Origin": 34894,
+      "▁durability": 34895,
+      "▁Sierra": 34896,
+      "nSen": 34897,
+      "光谱": 34898,
+      "▁Kumar": 34899,
+      "▁FR": 34900,
+      "我们都": 34901,
+      "▁poems": 34902,
+      "▁casting": 34903,
+      "▁courage": 34904,
+      "tres": 34905,
+      "▁cran": 34906,
+      "▁legally": 34907,
+      "斜率": 34908,
+      "致富": 34909,
+      "▁volleyball": 34910,
+      "克拉": 34911,
+      "Imp": 34912,
+      "违法行为": 34913,
+      "oside": 34914,
+      "▁Garc": 34915,
+      "▁heaven": 34916,
+      "▁analyst": 34917,
+      "雕刻": 34918,
+      "改变了": 34919,
+      "Graphics": 34920,
+      "额度": 34921,
+      "▁warehouse": 34922,
+      "tbackground": 34923,
+      "ipper": 34924,
+      "▁Trinity": 34925,
+      "▁strongest": 34926,
+      "半岛": 34927,
+      "报刊": 34928,
+      "▁Diet": 34929,
+      "▁custody": 34930,
+      "枸杞": 34931,
+      "Diagn": 34932,
+      "UTC": 34933,
+      "运动会": 34934,
+      "平市": 34935,
+      "ondo": 34936,
+      "▁Oliver": 34937,
+      "}]\\": 34938,
+      "ographs": 34939,
+      "▁Penn": 34940,
+      "▁Soph": 34941,
+      "mil": 34942,
+      "INIT": 34943,
+      "▁Eli": 34944,
+      "Tensor": 34945,
+      "▁semicon": 34946,
+      "▁Wine": 34947,
+      "五百": 34948,
+      "排骨": 34949,
+      "▁dealt": 34950,
+      "nAlumni": 34951,
+      "Ip": 34952,
+      "odef": 34953,
+      "铸造": 34954,
+      "军区": 34955,
+      "▁Fu": 34956,
+      "aukee": 34957,
+      "endsection": 34958,
+      "nReception": 34959,
+      "上次": 34960,
+      "管理人员": 34961,
+      "▁delicate": 34962,
+      "维度": 34963,
+      "的爱情": 34964,
+      "的小伙伴": 34965,
+      "ployees": 34966,
+      "▁margins": 34967,
+      "任命": 34968,
+      "多万": 34969,
+      "▁explanations": 34970,
+      "国家队": 34971,
+      "▁gad": 34972,
+      "structured": 34973,
+      "全家": 34974,
+      "Smart": 34975,
+      "oracic": 34976,
+      "rais": 34977,
+      "▁lane": 34978,
+      "▁dedication": 34979,
+      "iotensin": 34980,
+      "▁contrary": 34981,
+      "的电影": 34982,
+      "自然村": 34983,
+      "变卖": 34984,
+      "oresis": 34985,
+      "▁Builder": 34986,
+      "eo": 34987,
+      "不信": 34988,
+      "开头": 34989,
+      "▁toggle": 34990,
+      "▁etiology": 34991,
+      "与人": 34992,
+      "同上": 34993,
+      "oko": 34994,
+      "folk": 34995,
+      "薄膜": 34996,
+      "福州市": 34997,
+      "民办": 34998,
+      "akt": 34999,
+      "▁Expert": 35000,
+      "中新": 35001,
+      "这是一道": 35002,
+      "▁vest": 35003,
+      "▁wires": 35004,
+      "▁hypers": 35005,
+      "诸葛": 35006,
+      "▁Aaron": 35007,
+      "nSimilar": 35008,
+      "Il": 35009,
+      "▁pc": 35010,
+      "成立的": 35011,
+      "nCOPY": 35012,
+      "▁Trip": 35013,
+      "高新技术产业": 35014,
+      "▁Surv": 35015,
+      "▁premiere": 35016,
+      "听了": 35017,
+      "▁Gol": 35018,
+      "交通肇事": 35019,
+      "kotlin": 35020,
+      "工业园区": 35021,
+      "▁IA": 35022,
+      "已于": 35023,
+      "**:\\": 35024,
+      "▁squee": 35025,
+      "▁boring": 35026,
+      "▁mysterious": 35027,
+      "更能": 35028,
+      "▁gently": 35029,
+      "▁wellness": 35030,
+      "▁trash": 35031,
+      "的政策": 35032,
+      "▁touchdown": 35033,
+      "其前": 35034,
+      "atz": 35035,
+      "▁analysts": 35036,
+      "南瓜": 35037,
+      "富贵": 35038,
+      "抛弃": 35039,
+      "机组": 35040,
+      "牢固": 35041,
+      "近视": 35042,
+      "geom": 35043,
+      "▁Grove": 35044,
+      "▁colonial": 35045,
+      "台阶": 35046,
+      "▁oak": 35047,
+      "lycer": 35048,
+      "▁Hair": 35049,
+      "▁absorbed": 35050,
+      "▁adjustable": 35051,
+      "▁projections": 35052,
+      "▁gates": 35053,
+      "啊啊": 35054,
+      "水县": 35055,
+      "环境中": 35056,
+      "屏幕尺寸": 35057,
+      "▁Maps": 35058,
+      "▁Treas": 35059,
+      "▁Affero": 35060,
+      "宝马": 35061,
+      "▁joke": 35062,
+      "▁reversible": 35063,
+      "tfont": 35064,
+      "▁Average": 35065,
+      "wof": 35066,
+      "破解": 35067,
+      "化学品": 35068,
+      "加强对": 35069,
+      "▁resin": 35070,
+      "Ben": 35071,
+      "bys": 35072,
+      "婚纱": 35073,
+      "水流": 35074,
+      "▁Pav": 35075,
+      "▁CAD": 35076,
+      "amente": 35077,
+      "▁aster": 35078,
+      "▁Cathedral": 35079,
+      "共赢": 35080,
+      "ALLY": 35081,
+      "▁youngest": 35082,
+      "合资": 35083,
+      "惊艳": 35084,
+      "予以支持": 35085,
+      "▁DATA": 35086,
+      "ustration": 35087,
+      "明珠": 35088,
+      "运费": 35089,
+      "▁{:": 35090,
+      "▁disclosure": 35091,
+      "场面": 35092,
+      "被子": 35093,
+      "▁sax": 35094,
+      "养老保险": 35095,
+      "▁Delaware": 35096,
+      "网格": 35097,
+      "▁fare": 35098,
+      "▁steering": 35099,
+      "aments": 35100,
+      "▁cartilage": 35101,
+      "再利用": 35102,
+      "inters": 35103,
+      "指导下": 35104,
+      "▁persistence": 35105,
+      "领导人": 35106,
+      "▁Simplify": 35107,
+      "▁tops": 35108,
+      "▁ankle": 35109,
+      "水面": 35110,
+      "脆弱": 35111,
+      "音响": 35112,
+      "▁Keywords": 35113,
+      "▁hiking": 35114,
+      "▁remembered": 35115,
+      "▁Worth": 35116,
+      "迁移": 35117,
+      "▁carp": 35118,
+      "▁(#": 35119,
+      "有两": 35120,
+      "也需要": 35121,
+      "Allow": 35122,
+      "▁obstruction": 35123,
+      "列式": 35124,
+      "的全部": 35125,
+      "Transport": 35126,
+      "概括": 35127,
+      "ipart": 35128,
+      "▁基本参数": 35129,
+      "▁viewer": 35130,
+      "▁Venezuela": 35131,
+      "▁clones": 35132,
+      "▁APP": 35133,
+      "uristic": 35134,
+      "parallel": 35135,
+      ">[": 35136,
+      "eve": 35137,
+      "多元化": 35138,
+      "总公司": 35139,
+      "▁grep": 35140,
+      "▁microscopic": 35141,
+      "▁inequalities": 35142,
+      "合同约定": 35143,
+      "量子": 35144,
+      "不清楚": 35145,
+      "公布的": 35146,
+      "Family": 35147,
+      "熬夜": 35148,
+      "着手": 35149,
+      "霉素": 35150,
+      "▁preoperative": 35151,
+      "Addr": 35152,
+      "inth": 35153,
+      "▁Mouse": 35154,
+      "整除": 35155,
+      "香气": 35156,
+      "▁Pierre": 35157,
+      "askat": 35158,
+      "elem": 35159,
+      "新媒体": 35160,
+      "yzed": 35161,
+      "nsc": 35162,
+      "▁Peak": 35163,
+      "▁attendees": 35164,
+      "\\\\(\\\\{": 35165,
+      "▁Idaho": 35166,
+      "▁pipes": 35167,
+      "ANN": 35168,
+      "{},": 35169,
+      "墨西哥": 35170,
+      "启发": 35171,
+      "户籍地": 35172,
+      "各项工作": 35173,
+      "▁phospholip": 35174,
+      "nCr": 35175,
+      "▁chrome": 35176,
+      "▁Campaign": 35177,
+      "Mi": 35178,
+      "独家": 35179,
+      "had": 35180,
+      "▁\\'\"": 35181,
+      "上限": 35182,
+      "描绘": 35183,
+      "▁filtered": 35184,
+      "▁embr": 35185,
+      "tilde": 35186,
+      "双人": 35187,
+      "做成": 35188,
+      "▁Terry": 35189,
+      "▁professionally": 35190,
+      "似的": 35191,
+      "便秘": 35192,
+      "acci": 35193,
+      "▁CMS": 35194,
+      "▁não": 35195,
+      "之战": 35196,
+      "米饭": 35197,
+      "配料": 35198,
+      "▁Limit": 35199,
+      "▁occasional": 35200,
+      "陪同": 35201,
+      "Suite": 35202,
+      "▁constructs": 35203,
+      "nome": 35204,
+      "nBar": 35205,
+      "▁appar": 35206,
+      "▁knockout": 35207,
+      "▁modulate": 35208,
+      "DX": 35209,
+      "接入": 35210,
+      "税务局": 35211,
+      "nPhil": 35212,
+      "▁upset": 35213,
+      "事后": 35214,
+      "转动": 35215,
+      "brook": 35216,
+      "▁pond": 35217,
+      "▁Mongo": 35218,
+      "hb": 35219,
+      "那你": 35220,
+      "tro": 35221,
+      "yme": 35222,
+      "卖家": 35223,
+      "▁Lik": 35224,
+      "大纲": 35225,
+      "要比": 35226,
+      "机体": 35227,
+      "树木": 35228,
+      "cmake": 35229,
+      "▁Stra": 35230,
+      "查处": 35231,
+      "▁loci": 35232,
+      "▁rides": 35233,
+      "已知函数": 35234,
+      "▁Loss": 35235,
+      "caster": 35236,
+      "▁logistics": 35237,
+      "但其": 35238,
+      "我们是": 35239,
+      "voc": 35240,
+      "重要作用": 35241,
+      "Contin": 35242,
+      "outline": 35243,
+      "▁sexually": 35244,
+      "▁coated": 35245,
+      "▁correspondence": 35246,
+      "变频": 35247,
+      "网点": 35248,
+      "ETA": 35249,
+      "▁Dol": 35250,
+      "▁noting": 35251,
+      "Tx": 35252,
+      "要去": 35253,
+      "▁经营": 35254,
+      "第五节": 35255,
+      "osion": 35256,
+      "/<": 35257,
+      "stere": 35258,
+      "agrant": 35259,
+      "nChapter": 35260,
+      "keletal": 35261,
+      "▁continually": 35262,
+      "将是": 35263,
+      "street": 35264,
+      "▁pathogenic": 35265,
+      "▁Vas": 35266,
+      "化疗": 35267,
+      "Hook": 35268,
+      "▁strips": 35269,
+      "左右的": 35270,
+      "▁learners": 35271,
+      "▁Au": 35272,
+      "▁linkage": 35273,
+      "雅思": 35274,
+      "▁MORE": 35275,
+      "getString": 35276,
+      "背后的": 35277,
+      "党政": 35278,
+      "前行": 35279,
+      "拥抱": 35280,
+      "polit": 35281,
+      "sound": 35282,
+      "▁analges": 35283,
+      "六十": 35284,
+      "arnings": 35285,
+      "eye": 35286,
+      "▁NYC": 35287,
+      "海淀": 35288,
+      "座椅": 35289,
+      "蚂蚁": 35290,
+      "▁FT": 35291,
+      "党员干部": 35292,
+      "nMajor": 35293,
+      "直角坐标": 35294,
+      "iron": 35295,
+      "解决方法": 35296,
+      "BG": 35297,
+      "ле": 35298,
+      "的健康": 35299,
+      "▁Mans": 35300,
+      "▁syst": 35301,
+      "▁surroundings": 35302,
+      "Large": 35303,
+      "▁创作年代": 35304,
+      "OutputStream": 35305,
+      "击败": 35306,
+      "映射": 35307,
+      "▁Restaurant": 35308,
+      "▁salv": 35309,
+      "CREATE": 35310,
+      "decoration": 35311,
+      "发掘": 35312,
+      "迷茫": 35313,
+      "灾难": 35314,
+      "回来了": 35315,
+      "heduled": 35316,
+      "电视机": 35317,
+      "计分": 35318,
+      "March": 35319,
+      "Available": 35320,
+      "往事": 35321,
+      "切线": 35322,
+      "nForm": 35323,
+      "▁degeneration": 35324,
+      "班的": 35325,
+      "▁Hunt": 35326,
+      "ributing": 35327,
+      "宋代": 35328,
+      "山镇": 35329,
+      "gap": 35330,
+      "tmargin": 35331,
+      "十九大": 35332,
+      "▁Reform": 35333,
+      "▁sake": 35334,
+      "▁defender": 35335,
+      "拼搏": 35336,
+      "Den": 35337,
+      "SECRET": 35338,
+      "▁夏": 35339,
+      "容留": 35340,
+      "▁任": 35341,
+      "草莓": 35342,
+      "hetics": 35343,
+      "▁Finding": 35344,
+      "▁shaft": 35345,
+      "再到": 35346,
+      "回族": 35347,
+      "vertex": 35348,
+      "gradient": 35349,
+      "▁stepped": 35350,
+      "制造商": 35351,
+      "▁shifting": 35352,
+      "庄村": 35353,
+      "RUN": 35354,
+      "▁bail": 35355,
+      "banner": 35356,
+      "Serializable": 35357,
+      "eca": 35358,
+      "感悟": 35359,
+      "▁tabs": 35360,
+      "formula": 35361,
+      "mercially": 35362,
+      "▁NAT": 35363,
+      "nGeneral": 35364,
+      "被执行人刘": 35365,
+      "与之": 35366,
+      "▁\\\\:": 35367,
+      "▁metaph": 35368,
+      "不顾": 35369,
+      "▁Skin": 35370,
+      "▁Teen": 35371,
+      "▁trunk": 35372,
+      "▁transverse": 35373,
+      "因本案": 35374,
+      "检出": 35375,
+      "▁WAS": 35376,
+      "地带": 35377,
+      "▁alg": 35378,
+      "▁cush": 35379,
+      "ть": 35380,
+      "▁Casino": 35381,
+      "▁Conservative": 35382,
+      "助手": 35383,
+      "TreeNode": 35384,
+      "大国": 35385,
+      "▁Reyn": 35386,
+      "▁Buffalo": 35387,
+      "▁汉": 35388,
+      "图标": 35389,
+      "希望大家": 35390,
+      "uctive": 35391,
+      "▁preventive": 35392,
+      "猜测": 35393,
+      "无锡市": 35394,
+      "pixel": 35395,
+      "tresult": 35396,
+      "nthey": 35397,
+      "▁沈": 35398,
+      "▁tricky": 35399,
+      "年份": 35400,
+      "TARGET": 35401,
+      "▁lookup": 35402,
+      "lox": 35403,
+      "办公室主任": 35404,
+      "gc": 35405,
+      "处长": 35406,
+      "的家": 35407,
+      "AWS": 35408,
+      "认识到": 35409,
+      "▁prone": 35410,
+      "笑着": 35411,
+      "▁Holland": 35412,
+      "▁ingredient": 35413,
+      "音像": 35414,
+      "▁verb": 35415,
+      "alleng": 35416,
+      "遵义": 35417,
+      "\\\\,\\\\": 35418,
+      "▁conductor": 35419,
+      "几率": 35420,
+      "的消息": 35421,
+      "▁mono": 35422,
+      "▁panic": 35423,
+      "▁monument": 35424,
+      "举例": 35425,
+      "年出生": 35426,
+      "chip": 35427,
+      "▁Indians": 35428,
+      "的歌": 35429,
+      "以人为本": 35430,
+      "ilo": 35431,
+      "latitude": 35432,
+      "▁voluntary": 35433,
+      "apeutics": 35434,
+      "▁bitcoin": 35435,
+      "Gal": 35436,
+      "洗涤": 35437,
+      "腹泻": 35438,
+      "▁Das": 35439,
+      "还有很多": 35440,
+      "▁dummy": 35441,
+      "▁Gardens": 35442,
+      "如果您": 35443,
+      "▁他": 35444,
+      "四首": 35445,
+      "固醇": 35446,
+      "terms": 35447,
+      "▁José": 35448,
+      "▁tires": 35449,
+      "出土": 35450,
+      "▁på": 35451,
+      "▁landed": 35452,
+      "▁historian": 35453,
+      "乾坤": 35454,
+      "运算法": 35455,
+      "Links": 35456,
+      "▁neutron": 35457,
+      "▁Terms": 35458,
+      "我爱": 35459,
+      "Cookie": 35460,
+      "▁commits": 35461,
+      "atisf": 35462,
+      "▁rabbits": 35463,
+      "▁businessman": 35464,
+      "折叠": 35465,
+      "合法的": 35466,
+      "zhou": 35467,
+      "Assembly": 35468,
+      "▁taxi": 35469,
+      "jug": 35470,
+      "▁AK": 35471,
+      "▁Really": 35472,
+      "的概率": 35473,
+      "rylic": 35474,
+      "escope": 35475,
+      "subprocess": 35476,
+      "机动": 35477,
+      "▁Gift": 35478,
+      "平行四边形": 35479,
+      "▁grinding": 35480,
+      "▁closet": 35481,
+      "在乎": 35482,
+      "iston": 35483,
+      "▁Logic": 35484,
+      "还款义务": 35485,
+      "个位": 35486,
+      "个百分": 35487,
+      "Serialize": 35488,
+      "▁countless": 35489,
+      "▁Gonz": 35490,
+      "▁technically": 35491,
+      "nRobert": 35492,
+      "日月": 35493,
+      "nCount": 35494,
+      "月加入中国共产党": 35495,
+      "avo": 35496,
+      "oron": 35497,
+      "▁disposal": 35498,
+      "评议": 35499,
+      "山中": 35500,
+      "性好": 35501,
+      "者和": 35502,
+      "的做法": 35503,
+      "▁decides": 35504,
+      "▁merchant": 35505,
+      "在那": 35506,
+      "胎儿": 35507,
+      "▁CAS": 35508,
+      "假释案件": 35509,
+      "广播电视": 35510,
+      "与管理": 35511,
+      "yellow": 35512,
+      "deletion": 35513,
+      "性分析": 35514,
+      "减为": 35515,
+      "POINT": 35516,
+      "providers": 35517,
+      "▁Tat": 35518,
+      "literal": 35519,
+      "▁roller": 35520,
+      "是连载于起点中文网": 35521,
+      "赵某": 35522,
+      "满意度": 35523,
+      "nPolice": 35524,
+      "▁polymorphism": 35525,
+      "Append": 35526,
+      "演练": 35527,
+      "gary": 35528,
+      "electric": 35529,
+      "要知道": 35530,
+      ")}{\\\\": 35531,
+      "calendar": 35532,
+      "上周": 35533,
+      "矩形": 35534,
+      "▁Literature": 35535,
+      ")){\\": 35536,
+      "again": 35537,
+      "三首": 35538,
+      "生病": 35539,
+      "TRIBUTING": 35540,
+      "▁southeast": 35541,
+      "▁permanently": 35542,
+      "▁bovine": 35543,
+      "▁relapse": 35544,
+      "▁glycos": 35545,
+      "tvirtual": 35546,
+      "▁Rect": 35547,
+      "▁Mason": 35548,
+      "Love": 35549,
+      "低调": 35550,
+      "Clean": 35551,
+      "▁microscope": 35552,
+      "逍遥": 35553,
+      "可根据": 35554,
+      "老子": 35555,
+      "affe": 35556,
+      "▁Housing": 35557,
+      "ADE": 35558,
+      "zeta": 35559,
+      "▁Hep": 35560,
+      "一号": 35561,
+      "▁Symbol": 35562,
+      "▁anime": 35563,
+      "▁Alexand": 35564,
+      "走上": 35565,
+      "供水": 35566,
+      "亚马逊": 35567,
+      "▁emerge": 35568,
+      "负载": 35569,
+      "tal": 35570,
+      "▁Tit": 35571,
+      "▁splitting": 35572,
+      "会计师": 35573,
+      "▁Aims": 35574,
+      "彩虹": 35575,
+      "也非常": 35576,
+      "▁delegate": 35577,
+      "haus": 35578,
+      "auc": 35579,
+      "灵感": 35580,
+      "nLocation": 35581,
+      "每位": 35582,
+      "老化": 35583,
+      "迷你": 35584,
+      "制度的": 35585,
+      "▁histological": 35586,
+      "▁tu": 35587,
+      "前面的": 35588,
+      "ophyll": 35589,
+      "百合": 35590,
+      "(){": 35591,
+      "抢劫罪": 35592,
+      "▁metastases": 35593,
+      "ан": 35594,
+      "应付": 35595,
+      "▁Queue": 35596,
+      "端的": 35597,
+      "▁stiffness": 35598,
+      "备份": 35599,
+      "▁iTunes": 35600,
+      "▁Wes": 35601,
+      "▁coalition": 35602,
+      "村镇": 35603,
+      "ako": 35604,
+      "课堂教学": 35605,
+      "一股": 35606,
+      "▁visually": 35607,
+      "’\\": 35608,
+      "心病": 35609,
+      "Incre": 35610,
+      "Extra": 35611,
+      "设计和": 35612,
+      "▁terrorist": 35613,
+      "ascii": 35614,
+      "phthal": 35615,
+      "长寿": 35616,
+      "如何在": 35617,
+      "▁malaria": 35618,
+      "应由": 35619,
+      "做饭": 35620,
+      "组织实施": 35621,
+      "▁***": 35622,
+      "istries": 35623,
+      "▁fluctuations": 35624,
+      "五一": 35625,
+      "注释": 35626,
+      "SPACE": 35627,
+      "▁sprint": 35628,
+      "▁transforms": 35629,
+      "该行": 35630,
+      "身后": 35631,
+      "▁fel": 35632,
+      "动机": 35633,
+      "湿地": 35634,
+      "▁Snap": 35635,
+      "▁pian": 35636,
+      "nReview": 35637,
+      "▁sulfate": 35638,
+      "续行": 35639,
+      "imony": 35640,
+      "cstdio": 35641,
+      "suffix": 35642,
+      "▁cardinal": 35643,
+      "NetCore": 35644,
+      "External": 35645,
+      "isen": 35646,
+      "▁linearly": 35647,
+      "///\\": 35648,
+      "在法律规定": 35649,
+      "仅有": 35650,
+      "农机": 35651,
+      "▁TF": 35652,
+      "perform": 35653,
+      "▁bishop": 35654,
+      "▁larvae": 35655,
+      "中专": 35656,
+      "anto": 35657,
+      "▁Kash": 35658,
+      "▁Thor": 35659,
+      "▁Taking": 35660,
+      "▁causal": 35661,
+      "▁Warner": 35662,
+      "▁Corporate": 35663,
+      "加法": 35664,
+      "actional": 35665,
+      "▁numeric": 35666,
+      "帽子": 35667,
+      "▁BMI": 35668,
+      "主要用于": 35669,
+      "▁exchanges": 35670,
+      "犯贩卖毒品罪": 35671,
+      "hyp": 35672,
+      "▁axi": 35673,
+      "ndocker": 35674,
+      "一经": 35675,
+      "enti": 35676,
+      "▁lover": 35677,
+      "▁triggers": 35678,
+      "xf": 35679,
+      "▁innovations": 35680,
+      "这段话": 35681,
+      "prepare": 35682,
+      "ALS": 35683,
+      "中山市": 35684,
+      "▁intermed": 35685,
+      "во": 35686,
+      "▁freel": 35687,
+      "▁motors": 35688,
+      "▁composers": 35689,
+      "的视频": 35690,
+      "批量": 35691,
+      "青海省": 35692,
+      "布拉": 35693,
+      "手游": 35694,
+      "spin": 35695,
+      "油脂": 35696,
+      "熊猫": 35697,
+      "我感觉": 35698,
+      "为什么会": 35699,
+      "nGold": 35700,
+      "▁Borough": 35701,
+      "率高": 35702,
+      "ulos": 35703,
+      "智能手机": 35704,
+      "▁accounted": 35705,
+      "▁homeowners": 35706,
+      "DATE": 35707,
+      "Need": 35708,
+      "▁Lost": 35709,
+      "人了": 35710,
+      "football": 35711,
+      "▁encourages": 35712,
+      "Alpha": 35713,
+      "▁Lakes": 35714,
+      "▁vegetation": 35715,
+      "▁implantation": 35716,
+      "啥的": 35717,
+      "▁ré": 35718,
+      "Special": 35719,
+      "▁pension": 35720,
+      "▁impedance": 35721,
+      "▁bracket": 35722,
+      "▁necessity": 35723,
+      "本市": 35724,
+      "▁Flight": 35725,
+      "退回": 35726,
+      "textarea": 35727,
+      "▁indigenous": 35728,
+      "▁Contemporary": 35729,
+      "Paths": 35730,
+      "政策的": 35731,
+      "▁compressed": 35732,
+      "▁constructing": 35733,
+      "神的": 35734,
+      "▁）": 35735,
+      "而后": 35736,
+      "Authent": 35737,
+      "▁nitric": 35738,
+      "过大": 35739,
+      "▁yielded": 35740,
+      "authentication": 35741,
+      "▁đ": 35742,
+      "니다": 35743,
+      "▁basin": 35744,
+      "▁dehydro": 35745,
+      "]<": 35746,
+      "亲密": 35747,
+      "执政": 35748,
+      "演示": 35749,
+      "▁bub": 35750,
+      "▁shine": 35751,
+      "种植业": 35752,
+      "▁haem": 35753,
+      "▁todo": 35754,
+      "▁potatoes": 35755,
+      "INS": 35756,
+      "agle": 35757,
+      "agulation": 35758,
+      "▁screenshot": 35759,
+      "▁unemployment": 35760,
+      "韭菜": 35761,
+      "Signal": 35762,
+      "▁pursuit": 35763,
+      "委屈": 35764,
+      "}}$$": 35765,
+      "在此基础上": 35766,
+      "Vs": 35767,
+      "诱惑": 35768,
+      "▁Privacy": 35769,
+      "▁Trading": 35770,
+      "▁Identity": 35771,
+      "containers": 35772,
+      "送给": 35773,
+      "nTotal": 35774,
+      "▁asymptotic": 35775,
+      "研究和": 35776,
+      "▁VIP": 35777,
+      "▁withd": 35778,
+      "nProv": 35779,
+      "inh": 35780,
+      "平方根": 35781,
+      "study": 35782,
+      "Xiv": 35783,
+      "prehens": 35784,
+      "▁humidity": 35785,
+      "▁costume": 35786,
+      "▁Regiment": 35787,
+      "ptive": 35788,
+      "AQ": 35789,
+      "硬度": 35790,
+      "▁injections": 35791,
+      "有余": 35792,
+      "着急": 35793,
+      "cgi": 35794,
+      "los": 35795,
+      "andem": 35796,
+      "争夺": 35797,
+      "▁boyfriend": 35798,
+      "鲜明": 35799,
+      "研讨会": 35800,
+      "peak": 35801,
+      "Logging": 35802,
+      "▁Comparison": 35803,
+      "清明": 35804,
+      "lime": 35805,
+      "▁END": 35806,
+      "的处理": 35807,
+      "nNon": 35808,
+      "▁installations": 35809,
+      "askell": 35810,
+      "▁enters": 35811,
+      "乐器": 35812,
+      "▁imagery": 35813,
+      "▁courtesy": 35814,
+      "室友": 35815,
+      "隆重": 35816,
+      "▁nano": 35817,
+      "和方法": 35818,
+      "▁Springer": 35819,
+      "Mount": 35820,
+      "一处": 35821,
+      "nai": 35822,
+      "▁Nashville": 35823,
+      "Credentials": 35824,
+      "欧元": 35825,
+      "nCommunes": 35826,
+      "emph": 35827,
+      "mess": 35828,
+      "▁financ": 35829,
+      "ovan": 35830,
+      "只是一个": 35831,
+      "acerb": 35832,
+      "Unique": 35833,
+      "十三条之规定": 35834,
+      "▁Saints": 35835,
+      "▁mere": 35836,
+      "横向": 35837,
+      "▁DIY": 35838,
+      "华南": 35839,
+      "▁tournaments": 35840,
+      "了对": 35841,
+      "尤为": 35842,
+      "可以看出": 35843,
+      "nchar": 35844,
+      "Round": 35845,
+      "erala": 35846,
+      "parents": 35847,
+      "实际应用": 35848,
+      "substr": 35849,
+      "模式的": 35850,
+      "▁Aber": 35851,
+      "▁saturation": 35852,
+      "FUN": 35853,
+      "fas": 35854,
+      "我非常": 35855,
+      "nimage": 35856,
+      "▁Vincent": 35857,
+      "▁Float": 35858,
+      "leted": 35859,
+      "▁Carn": 35860,
+      "▁granite": 35861,
+      "早日": 35862,
+      "过高": 35863,
+      "▁Symfony": 35864,
+      "▁吕": 35865,
+      "psis": 35866,
+      "▁neighboring": 35867,
+      "▁(>": 35868,
+      "▁Kol": 35869,
+      "规范化": 35870,
+      "▁emp": 35871,
+      "▁Planet": 35872,
+      "▁FP": 35873,
+      "omed": 35874,
+      "档次": 35875,
+      "盲目": 35876,
+      "▁Mend": 35877,
+      "▁traumatic": 35878,
+      "icin": 35879,
+      "▁lineup": 35880,
+      "shots": 35881,
+      "▁arriving": 35882,
+      "▁embedding": 35883,
+      "▁田": 35884,
+      "我省": 35885,
+      "震荡": 35886,
+      "心血管": 35887,
+      "Alex": 35888,
+      "boro": 35889,
+      "▁IoT": 35890,
+      "另有": 35891,
+      "规章": 35892,
+      "innerHTML": 35893,
+      "arial": 35894,
+      "特斯拉": 35895,
+      "Commit": 35896,
+      "终极": 35897,
+      "ependence": 35898,
+      "▁Archives": 35899,
+      "antics": 35900,
+      "▁celebrity": 35901,
+      "▁波": 35902,
+      "鞋子": 35903,
+      "▁Parad": 35904,
+      "中场": 35905,
+      "锁定": 35906,
+      "拉开": 35907,
+      "随机分为": 35908,
+      "estine": 35909,
+      "▁messaging": 35910,
+      "断裂": 35911,
+      "的反": 35912,
+      "chedule": 35913,
+      "▁unwanted": 35914,
+      "▁nonprofit": 35915,
+      "▁frameworks": 35916,
+      "▁horizon": 35917,
+      "▁neuron": 35918,
+      "柏林": 35919,
+      "▁remarks": 35920,
+      "将以": 35921,
+      "▁Mul": 35922,
+      "cached": 35923,
+      "enames": 35924,
+      "中国人民银行同期": 35925,
+      "▁rolls": 35926,
+      "Initialize": 35927,
+      "公关": 35928,
+      "体外": 35929,
+      "GPS": 35930,
+      "类似于": 35931,
+      "▁URI": 35932,
+      "Inline": 35933,
+      "Controls": 35934,
+      "▁Sheriff": 35935,
+      "沃尔": 35936,
+      "directed": 35937,
+      "▁$(\\\\": 35938,
+      "驾驶证": 35939,
+      "iseconds": 35940,
+      "改良": 35941,
+      "道路交通事故": 35942,
+      "战斗力": 35943,
+      "▁简体中文": 35944,
+      "▁Tas": 35945,
+      "▁populated": 35946,
+      "(%": 35947,
+      "boys": 35948,
+      "▁NHS": 35949,
+      "▁agonist": 35950,
+      "ussia": 35951,
+      "moment": 35952,
+      "不被": 35953,
+      "肠胃": 35954,
+      "装配": 35955,
+      "hack": 35956,
+      "▁Functional": 35957,
+      "▁comics": 35958,
+      "▁averaged": 35959,
+      "奶油": 35960,
+      "紫色": 35961,
+      "冷藏": 35962,
+      "(\"[": 35963,
+      "直角三角形": 35964,
+      "普查": 35965,
+      "▁floral": 35966,
+      "regation": 35967,
+      "parsed": 35968,
+      "▁Premium": 35969,
+      "第五百一十九条规定": 35970,
+      "agh": 35971,
+      "总价": 35972,
+      "VIDEO": 35973,
+      "raform": 35974,
+      "▁untreated": 35975,
+      "▁setuptools": 35976,
+      "多为": 35977,
+      "RAY": 35978,
+      "内心的": 35979,
+      "▁fabrication": 35980,
+      "盒子": 35981,
+      "经鉴定": 35982,
+      "lington": 35983,
+      "Answer": 35984,
+      "薪酬": 35985,
+      "有意思": 35986,
+      "▁pile": 35987,
+      "utdown": 35988,
+      "▁Trophy": 35989,
+      "apses": 35990,
+      "waukee": 35991,
+      "团委": 35992,
+      "程序员": 35993,
+      "▁SMS": 35994,
+      "▁plumbing": 35995,
+      "▁Wu": 35996,
+      "orig": 35997,
+      "案件适用法律": 35998,
+      "quiries": 35999,
+      "▁Beauty": 36000,
+      "▁SHO": 36001,
+      "▁decorative": 36002,
+      "第九章": 36003,
+      "▁rip": 36004,
+      "▁Naval": 36005,
+      "一提": 36006,
+      "考证": 36007,
+      "Calendar": 36008,
+      "医疗器械": 36009,
+      "▁Celebr": 36010,
+      "▁legitimate": 36011,
+      "东营": 36012,
+      "第五百一十九条之规定": 36013,
+      "▁problematic": 36014,
+      "之二": 36015,
+      "万一": 36016,
+      "Hor": 36017,
+      "authorized": 36018,
+      "挂牌": 36019,
+      "AMS": 36020,
+      "San": 36021,
+      "▁monol": 36022,
+      "▁accumulated": 36023,
+      "▁eliminating": 36024,
+      "▁Lud": 36025,
+      "Embed": 36026,
+      "grunt": 36027,
+      "世的": 36028,
+      "Exist": 36029,
+      "nSpan": 36030,
+      "手上": 36031,
+      "CADE": 36032,
+      "∈(": 36033,
+      "▁UTC": 36034,
+      "所能": 36035,
+      "并不能": 36036,
+      "acao": 36037,
+      "serializer": 36038,
+      "设计与": 36039,
+      "马尔": 36040,
+      "都不能": 36041,
+      "竣工": 36042,
+      "衰老": 36043,
+      "言语": 36044,
+      "gro": 36045,
+      "组数据": 36046,
+      "▁concerts": 36047,
+      "▁HF": 36048,
+      "新疆维吾尔": 36049,
+      "▁respected": 36050,
+      "光荣": 36051,
+      "▁Rogers": 36052,
+      "▁Azerbai": 36053,
+      "大豆": 36054,
+      "obl": 36055,
+      "isse": 36056,
+      "CSF": 36057,
+      "nwill": 36058,
+      "unday": 36059,
+      "▁Hg": 36060,
+      "申请恢复执行": 36061,
+      "Po": 36062,
+      "▁Maple": 36063,
+      "▁underneath": 36064,
+      "jest": 36065,
+      "Rank": 36066,
+      "▁satisfactory": 36067,
+      "原审被告人": 36068,
+      "未必": 36069,
+      "▁Divis": 36070,
+      "trained": 36071,
+      "▁similarities": 36072,
+      "腹腔": 36073,
+      "被执行人陈": 36074,
+      "▁adjusting": 36075,
+      "GPIO": 36076,
+      "}}$$\\": 36077,
+      "和她": 36078,
+      "关注我": 36079,
+      "▁Schw": 36080,
+      "slot": 36081,
+      "▁Mario": 36082,
+      "minimum": 36083,
+      "▁bears": 36084,
+      "coordinates": 36085,
+      "▁mul": 36086,
+      "▁Sequence": 36087,
+      "nHar": 36088,
+      "▁Fle": 36089,
+      "这篇文章": 36090,
+      "▁bile": 36091,
+      "▁kiss": 36092,
+      "卫生和计划生育局": 36093,
+      "上演": 36094,
+      "zhong": 36095,
+      "▁rays": 36096,
+      "▁arises": 36097,
+      "▁Fifth": 36098,
+      "▁symbolic": 36099,
+      "特定的": 36100,
+      "▁downloading": 36101,
+      "的歌曲": 36102,
+      "采取保全措施": 36103,
+      "mf": 36104,
+      "Observer": 36105,
+      "契机": 36106,
+      "自然资源": 36107,
+      "▁palm": 36108,
+      "▁Valentine": 36109,
+      "是无": 36110,
+      "推荐了": 36111,
+      "▁Benef": 36112,
+      "bic": 36113,
+      "▁scipy": 36114,
+      "▁silly": 36115,
+      "▁dissolved": 36116,
+      "让大家": 36117,
+      "tap": 36118,
+      "的平均": 36119,
+      "aggreg": 36120,
+      "伊斯": 36121,
+      "▁cleanup": 36122,
+      "criptions": 36123,
+      "成像": 36124,
+      "▁debris": 36125,
+      "喜好": 36126,
+      "纯收入": 36127,
+      "▁Payment": 36128,
+      "Ul": 36129,
+      "新一代": 36130,
+      "▁elong": 36131,
+      "▁fireplace": 36132,
+      "ificent": 36133,
+      "日后": 36134,
+      "的事件": 36135,
+      "▁fool": 36136,
+      "▁pockets": 36137,
+      "Syntax": 36138,
+      "▁Abu": 36139,
+      "工作经验": 36140,
+      "▁incorporates": 36141,
+      "▁indexes": 36142,
+      "不会有": 36143,
+      "osit": 36144,
+      "worthy": 36145,
+      "康熙": 36146,
+      "▁enrollment": 36147,
+      "汽油": 36148,
+      "▁lac": 36149,
+      "可以帮助": 36150,
+      "▁Expand": 36151,
+      "▁江": 36152,
+      "Canvas": 36153,
+      "playing": 36154,
+      "nRemember": 36155,
+      "启用": 36156,
+      "的四": 36157,
+      "的照片": 36158,
+      "▁Actually": 36159,
+      "若有": 36160,
+      "▁已完结": 36161,
+      "▁crafted": 36162,
+      "▁Discovery": 36163,
+      "共青": 36164,
+      "exist": 36165,
+      "▁Oscar": 36166,
+      "南宁": 36167,
+      "▁honestly": 36168,
+      "再把": 36169,
+      "特此": 36170,
+      "保证金": 36171,
+      "与被申请人": 36172,
+      "money": 36173,
+      "▁Prove": 36174,
+      "固定资产": 36175,
+      "工作会议": 36176,
+      "oproteins": 36177,
+      "▁inferior": 36178,
+      "勇于": 36179,
+      "<\\\\": 36180,
+      "▁penetration": 36181,
+      "▁supern": 36182,
+      "nwhen": 36183,
+      "▁chore": 36184,
+      "▁threw": 36185,
+      "▁Others": 36186,
+      "患有": 36187,
+      "▁hiding": 36188,
+      "▁Mari": 36189,
+      "▁Karen": 36190,
+      "在地": 36191,
+      "难过": 36192,
+      "▁Answers": 36193,
+      "疑惑": 36194,
+      "nRussian": 36195,
+      "的人群": 36196,
+      "复仇": 36197,
+      "Bound": 36198,
+      "▁diffuse": 36199,
+      "▁Population": 36200,
+      "锂电池": 36201,
+      "当事人有权": 36202,
+      "▁electroly": 36203,
+      "Elect": 36204,
+      "▁Sah": 36205,
+      "▁conjugate": 36206,
+      "奖励审批表": 36207,
+      "▁Ker": 36208,
+      "反映了": 36209,
+      "北海": 36210,
+      "▁Escherichia": 36211,
+      "▁lined": 36212,
+      "▁decoration": 36213,
+      "蜘蛛": 36214,
+      "fers": 36215,
+      "▁可": 36216,
+      "确已破裂": 36217,
+      "▁humanity": 36218,
+      "▁imaginary": 36219,
+      "ijk": 36220,
+      "iosity": 36221,
+      "ranial": 36222,
+      "Requests": 36223,
+      "快来": 36224,
+      "▁ubuntu": 36225,
+      "▁internally": 36226,
+      "也无法": 36227,
+      "创造了": 36228,
+      "nRivers": 36229,
+      "▁cytotoxic": 36230,
+      "▁termination": 36231,
+      "创建了": 36232,
+      "▁disappeared": 36233,
+      "▁Pra": 36234,
+      "▁sampled": 36235,
+      "微量": 36236,
+      "起床": 36237,
+      "adj": 36238,
+      "的依据": 36239,
+      "▁educate": 36240,
+      "▁methodologies": 36241,
+      "戏曲": 36242,
+      "▁背景": 36243,
+      "Mal": 36244,
+      "所说的": 36245,
+      "完成生产任务": 36246,
+      "流域": 36247,
+      "isle": 36248,
+      "▁HashMap": 36249,
+      "▁dentist": 36250,
+      "武侠": 36251,
+      "门的": 36252,
+      "▁defer": 36253,
+      "atan": 36254,
+      "onian": 36255,
+      "▁caus": 36256,
+      "Signature": 36257,
+      "pics": 36258,
+      "▁Tensor": 36259,
+      "震惊": 36260,
+      "矿山": 36261,
+      "▁wrist": 36262,
+      "的理念": 36263,
+      "Linked": 36264,
+      "▁binds": 36265,
+      "▁utter": 36266,
+      "麦克风": 36267,
+      "外表": 36268,
+      "Exchange": 36269,
+      "って": 36270,
+      "急救": 36271,
+      "门前": 36272,
+      "\\'>": 36273,
+      "方言": 36274,
+      "▁Delivery": 36275,
+      "有意": 36276,
+      "已在": 36277,
+      "衣柜": 36278,
+      "OST": 36279,
+      "▁Mob": 36280,
+      "▁prohibited": 36281,
+      "▁Citation": 36282,
+      "▁creatures": 36283,
+      "▁ROS": 36284,
+      "amino": 36285,
+      "▁Dynamics": 36286,
+      "googleapis": 36287,
+      "至此": 36288,
+      "paste": 36289,
+      "▁exotic": 36290,
+      "▁Thunder": 36291,
+      "“\\": 36292,
+      "▁GE": 36293,
+      "▁AMD": 36294,
+      "▁Vista": 36295,
+      "▁commentary": 36296,
+      "▁micros": 36297,
+      "jupiter": 36298,
+      "▁certificates": 36299,
+      "半天": 36300,
+      "▁persu": 36301,
+      "▁目": 36302,
+      "有一点": 36303,
+      "闺蜜": 36304,
+      "▁Neil": 36305,
+      "ncol": 36306,
+      "▁Por": 36307,
+      "▁folders": 36308,
+      "过分": 36309,
+      "enos": 36310,
+      "▁traces": 36311,
+      "是目前": 36312,
+      "演唱会": 36313,
+      "xe": 36314,
+      "烧烤": 36315,
+      "▁antis": 36316,
+      "Hex": 36317,
+      "▁ci": 36318,
+      "▁Medium": 36319,
+      "▁Slovak": 36320,
+      "stackexchange": 36321,
+      "勉强": 36322,
+      "占领": 36323,
+      "效力的": 36324,
+      "▁norms": 36325,
+      "locations": 36326,
+      "▁与": 36327,
+      "叶子": 36328,
+      "▁siblings": 36329,
+      "▁cb": 36330,
+      "sery": 36331,
+      "senal": 36332,
+      "▁Horse": 36333,
+      "affected": 36334,
+      "▁Synd": 36335,
+      "▁alien": 36336,
+      "▁nerves": 36337,
+      "▁Required": 36338,
+      "▁fertility": 36339,
+      "▁INT": 36340,
+      "▁Registration": 36341,
+      "▁NotImplemented": 36342,
+      "只为": 36343,
+      "笑了": 36344,
+      "invoke": 36345,
+      "本裁定书送达后即发生法律效力": 36346,
+      "▁METHODS": 36347,
+      "日本的": 36348,
+      "▁{}\"": 36349,
+      "▁forg": 36350,
+      "nShould": 36351,
+      "▁suffix": 36352,
+      "▁epigen": 36353,
+      "之作": 36354,
+      "呈现出": 36355,
+      "行审": 36356,
+      "静电": 36357,
+      "▁Norman": 36358,
+      "转身": 36359,
+      "▁Lan": 36360,
+      "ylene": 36361,
+      "不负": 36362,
+      "每个人的": 36363,
+      "▁Cuba": 36364,
+      "第二届": 36365,
+      "nAng": 36366,
+      "▁ASC": 36367,
+      "▁Ras": 36368,
+      "也是一个": 36369,
+      "乾隆": 36370,
+      "根廷": 36371,
+      "nasync": 36372,
+      "▁galaxy": 36373,
+      "▁putative": 36374,
+      "头市": 36375,
+      "▁Sax": 36376,
+      "河市": 36377,
+      "”——": 36378,
+      "▁Tib": 36379,
+      "threads": 36380,
+      "▁推荐": 36381,
+      "▁uncommon": 36382,
+      "权重": 36383,
+      "梦幻": 36384,
+      "▁Done": 36385,
+      "▁dynamically": 36386,
+      "公约": 36387,
+      "干什么": 36388,
+      "axies": 36389,
+      "ARI": 36390,
+      "treatment": 36391,
+      "▁impacted": 36392,
+      "阻力": 36393,
+      "nmodel": 36394,
+      "向法院": 36395,
+      "▁HPV": 36396,
+      "registration": 36397,
+      "半夜": 36398,
+      "aternity": 36399,
+      "wg": 36400,
+      "jing": 36401,
+      "recipe": 36402,
+      "▁mucosal": 36403,
+      "▁weekends": 36404,
+      "']);\\": 36405,
+      "显微": 36406,
+      "黄某": 36407,
+      "家里的": 36408,
+      "▁Standards": 36409,
+      "吉林市": 36410,
+      "宜宾": 36411,
+      "牛仔": 36412,
+      "CLIENT": 36413,
+      "▁codec": 36414,
+      "即是": 36415,
+      "nShare": 36416,
+      "▁Elastic": 36417,
+      "▁holders": 36418,
+      "窗户": 36419,
+      "九个月": 36420,
+      "aspx": 36421,
+      "▁Exam": 36422,
+      "nAlbums": 36423,
+      "能认罪服法": 36424,
+      "▁Raman": 36425,
+      "▁assembl": 36426,
+      "▁Especially": 36427,
+      "▁Male": 36428,
+      "bullet": 36429,
+      "▁mutagen": 36430,
+      "▁traders": 36431,
+      "▁proceedings": 36432,
+      "Entities": 36433,
+      "台州市": 36434,
+      "菲律": 36435,
+      "▁Engineer": 36436,
+      "▁exposures": 36437,
+      "▁曾": 36438,
+      "▁lounge": 36439,
+      "▁Frame": 36440,
+      "Screens": 36441,
+      "▁Scanner": 36442,
+      "SERT": 36443,
+      "comit": 36444,
+      "一心": 36445,
+      "总数的": 36446,
+      "▁uro": 36447,
+      "晚会": 36448,
+      "法兰": 36449,
+      "▁charming": 36450,
+      "Jack": 36451,
+      "基础知识": 36452,
+      "actin": 36453,
+      "heets": 36454,
+      "▁####": 36455,
+      "时常": 36456,
+      "吹风机": 36457,
+      "\\'(": 36458,
+      "继续执行": 36459,
+      "中心支公司": 36460,
+      "▁aiming": 36461,
+      "[@": 36462,
+      "之美": 36463,
+      "抗体": 36464,
+      "elaide": 36465,
+      "▁adjustments": 36466,
+      "质的": 36467,
+      "开发有限公司": 36468,
+      "上游": 36469,
+      "合法有效": 36470,
+      "tget": 36471,
+      "▁meantime": 36472,
+      "ру": 36473,
+      "时报": 36474,
+      "portal": 36475,
+      "▁Discover": 36476,
+      "▁converge": 36477,
+      "新疆维吾尔自治区": 36478,
+      "终究": 36479,
+      "USD": 36480,
+      "digest": 36481,
+      "nMilitary": 36482,
+      "担忧": 36483,
+      "▁Degree": 36484,
+      "]],": 36485,
+      "请您务必": 36486,
+      "▁Cabinet": 36487,
+      "警惕": 36488,
+      "一带一路": 36489,
+      "▁Airlines": 36490,
+      "▁debugging": 36491,
+      "Guard": 36492,
+      "▁Fitz": 36493,
+      "▁inexpensive": 36494,
+      "典礼": 36495,
+      "流行的": 36496,
+      "检修": 36497,
+      "动力学": 36498,
+      "orie": 36499,
+      "▁smoothly": 36500,
+      "台上": 36501,
+      "▁ect": 36502,
+      "mathop": 36503,
+      "并告诉我": 36504,
+      "不准许": 36505,
+      "在道路上": 36506,
+      "IMAGE": 36507,
+      "省份": 36508,
+      "▁TN": 36509,
+      "▁arrives": 36510,
+      "相逢": 36511,
+      "ratulations": 36512,
+      "▁Ivan": 36513,
+      "▁卡": 36514,
+      "npython": 36515,
+      "nSan": 36516,
+      "▁歌曲原唱": 36517,
+      "nWorks": 36518,
+      "imeters": 36519,
+      "checkout": 36520,
+      "éd": 36521,
+      "ubes": 36522,
+      "▁Players": 36523,
+      "▁mechanistic": 36524,
+      "菲律宾": 36525,
+      "营业时间": 36526,
+      "选定正确答案": 36527,
+      "ONT": 36528,
+      "osex": 36529,
+      "ATT": 36530,
+      "▁\\'_": 36531,
+      "▁Muslims": 36532,
+      "▁escorts": 36533,
+      "cite": 36534,
+      "▁Hilbert": 36535,
+      "▁Prest": 36536,
+      "▁Ethiop": 36537,
+      "ArgumentParser": 36538,
+      "分化": 36539,
+      "ALT": 36540,
+      "SPE": 36541,
+      "Rows": 36542,
+      "影子": 36543,
+      "清热": 36544,
+      "都非常": 36545,
+      "▁Hamm": 36546,
+      "路的": 36547,
+      "ViewController": 36548,
+      "▁丁": 36549,
+      "僵尸": 36550,
+      "选择题时": 36551,
+      "▁Climate": 36552,
+      "Pe": 36553,
+      "头像": 36554,
+      "▁Crusher": 36555,
+      "▁persist": 36556,
+      "▁爱": 36557,
+      "nyour": 36558,
+      "Sys": 36559,
+      "五入": 36560,
+      "净水": 36561,
+      "用以": 36562,
+      "▁puzz": 36563,
+      "$(\\'": 36564,
+      "computer": 36565,
+      "▁cleavage": 36566,
+      "▁radiotherapy": 36567,
+      "▁lips": 36568,
+      "占总收入的": 36569,
+      "▁Album": 36570,
+      "▁gases": 36571,
+      "▁cope": 36572,
+      "ophone": 36573,
+      "无非": 36574,
+      "▁Bak": 36575,
+      "▁Ord": 36576,
+      "▁timeline": 36577,
+      "Wal": 36578,
+      "▁mitochondria": 36579,
+      "票房": 36580,
+      "andro": 36581,
+      "发音": 36582,
+      "副校长": 36583,
+      "展示了": 36584,
+      "▁tweets": 36585,
+      "▁smartphones": 36586,
+      "▁depriv": 36587,
+      "▁obstruct": 36588,
+      "nPolitical": 36589,
+      "xiv": 36590,
+      "▁cel": 36591,
+      "bourg": 36592,
+      "▁NAME": 36593,
+      "substring": 36594,
+      "nnew": 36595,
+      "▁NOW": 36596,
+      "million": 36597,
+      "▁lifting": 36598,
+      "忙碌": 36599,
+      "nTIME": 36600,
+      "并做出": 36601,
+      "▁Shaw": 36602,
+      "出版社出版": 36603,
+      "▁opener": 36604,
+      "something": 36605,
+      "踏实": 36606,
+      "培训机构": 36607,
+      "▁ч": 36608,
+      "常州": 36609,
+      "nRoman": 36610,
+      "▁advisory": 36611,
+      "均价": 36612,
+      "▁caution": 36613,
+      "严峻": 36614,
+      "▁FAQ": 36615,
+      "unwrap": 36616,
+      "现为": 36617,
+      "csrf": 36618,
+      "劳动报酬": 36619,
+      "居家": 36620,
+      "信息公开": 36621,
+      "contr": 36622,
+      "务实": 36623,
+      "对方的": 36624,
+      "为期": 36625,
+      "率和": 36626,
+      "▁strikes": 36627,
+      "accharide": 36628,
+      "保密": 36629,
+      "thumbnail": 36630,
+      "newcommand": 36631,
+      "nversion": 36632,
+      "本院在执行过程中": 36633,
+      "▁merging": 36634,
+      "▁Focus": 36635,
+      "▁时间复杂度": 36636,
+      "▁embarrass": 36637,
+      "▁convolution": 36638,
+      "品味": 36639,
+      "感慨": 36640,
+      "面团": 36641,
+      "担保人": 36642,
+      "威尔": 36643,
+      "落的": 36644,
+      "▁Bluetooth": 36645,
+      "▁Pep": 36646,
+      "▁tenant": 36647,
+      "urement": 36648,
+      "unar": 36649,
+      "▁Lapl": 36650,
+      "圈子": 36651,
+      "以下备选答案": 36652,
+      "artered": 36653,
+      "▁acknowledged": 36654,
+      "▁ds": 36655,
+      "网络营销": 36656,
+      "▁eleven": 36657,
+      "▁saline": 36658,
+      "不通": 36659,
+      "想知道": 36660,
+      "arynge": 36661,
+      "之力": 36662,
+      "selling": 36663,
+      "▁Directors": 36664,
+      "nBelow": 36665,
+      "委托代理人王": 36666,
+      "▁honors": 36667,
+      "distributed": 36668,
+      "楼主": 36669,
+      "FET": 36670,
+      "执行通知": 36671,
+      "代理书记员": 36672,
+      "▁welcoming": 36673,
+      "与会": 36674,
+      "cit": 36675,
+      "我当时": 36676,
+      "粗糙": 36677,
+      "美军": 36678,
+      "鉴别": 36679,
+      "▁och": 36680,
+      "▁Cache": 36681,
+      "磨练": 36682,
+      "▁catalyst": 36683,
+      "精美": 36684,
+      "颁奖": 36685,
+      "▁Mathematica": 36686,
+      "▁reservation": 36687,
+      "▁LD": 36688,
+      "诈骗罪": 36689,
+      "▁reass": 36690,
+      "画的": 36691,
+      "作出了": 36692,
+      "cuador": 36693,
+      "▁Nuclear": 36694,
+      "▁biosynthesis": 36695,
+      "怎么做": 36696,
+      "▁corporation": 36697,
+      "▁MN": 36698,
+      "▁Ottawa": 36699,
+      "▁Uganda": 36700,
+      "▁mellitus": 36701,
+      "▁Sor": 36702,
+      "▁sung": 36703,
+      "操控": 36704,
+      "十个月": 36705,
+      "▁CSV": 36706,
+      "Generate": 36707,
+      "▁dielectric": 36708,
+      "文件的": 36709,
+      "▁fixation": 36710,
+      "▁surviving": 36711,
+      "中法": 36712,
+      "cuda": 36713,
+      "However": 36714,
+      "▁investigates": 36715,
+      "一直以来": 36716,
+      "口袋": 36717,
+      "▁\\\\*": 36718,
+      "isbane": 36719,
+      "jk": 36720,
+      "用力": 36721,
+      "nQuote": 36722,
+      "▁chess": 36723,
+      "▁mammals": 36724,
+      "▁Provider": 36725,
+      "▁subsidiary": 36726,
+      "KC": 36727,
+      "▁Deal": 36728,
+      "nError": 36729,
+      "▁NAD": 36730,
+      "▁Electrical": 36731,
+      "▁Poisson": 36732,
+      "才可以": 36733,
+      "▁rewrite": 36734,
+      "▁村": 36735,
+      "使用权": 36736,
+      "罪犯李": 36737,
+      "REEN": 36738,
+      "先进单位": 36739,
+      "撤回上诉处理": 36740,
+      "▁Hudson": 36741,
+      "▁answering": 36742,
+      "还未": 36743,
+      "看一下": 36744,
+      "市场经济": 36745,
+      "nTom": 36746,
+      "这一次": 36747,
+      "神器": 36748,
+      "mund": 36749,
+      "nmake": 36750,
+      "▁facebook": 36751,
+      "▁filtration": 36752,
+      "▁litigation": 36753,
+      ":{": 36754,
+      "IRST": 36755,
+      "bill": 36756,
+      "▁chap": 36757,
+      "designed": 36758,
+      "hospital": 36759,
+      "会使": 36760,
+      "BSD": 36761,
+      "▁rupt": 36762,
+      "质地": 36763,
+      "日被逮捕": 36764,
+      "nHome": 36765,
+      "的重": 36766,
+      ";\">\\": 36767,
+      "rice": 36768,
+      "车上": 36769,
+      "▁Gy": 36770,
+      "▁correlates": 36771,
+      "]],\\": 36772,
+      "amonds": 36773,
+      "▁booked": 36774,
+      "▁heap": 36775,
+      "▁vanilla": 36776,
+      "▁utilizes": 36777,
+      "创伤": 36778,
+      "回答以下问题": 36779,
+      "▁marble": 36780,
+      "▁Wireless": 36781,
+      "晚期": 36782,
+      "服务平台": 36783,
+      "naires": 36784,
+      "distribution": 36785,
+      "列为": 36786,
+      "Flash": 36787,
+      "stairs": 36788,
+      "的正": 36789,
+      "粉碎": 36790,
+      "olt": 36791,
+      "Creator": 36792,
+      "挺好的": 36793,
+      "▁ghost": 36794,
+      "▁Generator": 36795,
+      "中学生": 36796,
+      "Upper": 36797,
+      "▁collector": 36798,
+      "▁cryptocurrency": 36799,
+      "鲁迅": 36800,
+      "Structure": 36801,
+      "▁Rah": 36802,
+      "系其": 36803,
+      "▁timber": 36804,
+      "controls": 36805,
+      "参演电影": 36806,
+      "▁expans": 36807,
+      "Actor": 36808,
+      "▁не": 36809,
+      "erek": 36810,
+      "文化艺术": 36811,
+      "pires": 36812,
+      "▁migrate": 36813,
+      "▁Institution": 36814,
+      "临沂市": 36815,
+      "减少了": 36816,
+      "知名度": 36817,
+      "▁Near": 36818,
+      "▁yaml": 36819,
+      "▁proto": 36820,
+      "为之": 36821,
+      "▁Ahmed": 36822,
+      "▁commercially": 36823,
+      "见了": 36824,
+      "有许多": 36825,
+      "▁CAR": 36826,
+      "uppet": 36827,
+      "▁Munich": 36828,
+      "其中一个": 36829,
+      "▁subunits": 36830,
+      "Ter": 36831,
+      "最重要": 36832,
+      "▁addr": 36833,
+      "idency": 36834,
+      "prevent": 36835,
+      "Agg": 36836,
+      "▁dosage": 36837,
+      "出演": 36838,
+      "nRad": 36839,
+      "▁nep": 36840,
+      "就来": 36841,
+      "林木": 36842,
+      "▁blues": 36843,
+      "▁exacerb": 36844,
+      "▁spite": 36845,
+      "▁refreshing": 36846,
+      "▁Eve": 36847,
+      "PTION": 36848,
+      "一元二次方程": 36849,
+      "▁Ori": 36850,
+      "▁tin": 36851,
+      "▁advers": 36852,
+      "互助": 36853,
+      "火爆": 36854,
+      "▁SPD": 36855,
+      "nProblem": 36856,
+      "▁Cameron": 36857,
+      "心得": 36858,
+      "内存容量": 36859,
+      "▁Jar": 36860,
+      "荣誉称号": 36861,
+      "tbool": 36862,
+      "量化": 36863,
+      "▁Scre": 36864,
+      "umbent": 36865,
+      "Projects": 36866,
+      "显着性": 36867,
+      "具体应用": 36868,
+      "▁sept": 36869,
+      "compar": 36870,
+      "两名": 36871,
+      "▁vim": 36872,
+      "▁shutdown": 36873,
+      "ña": 36874,
+      "endent": 36875,
+      "Os": 36876,
+      "的说法": 36877,
+      "▁Jak": 36878,
+      "▁trades": 36879,
+      "nAmong": 36880,
+      "德育": 36881,
+      "火焰": 36882,
+      "▁autism": 36883,
+      "和新": 36884,
+      "虚拟机": 36885,
+      "▁Interest": 36886,
+      "ILE": 36887,
+      "ся": 36888,
+      "▁Campus": 36889,
+      "{#": 36890,
+      "楼梯": 36891,
+      "▁vaccin": 36892,
+      "的一生": 36893,
+      "inas": 36894,
+      "执行董事": 36895,
+      "▁backwards": 36896,
+      "`:": 36897,
+      "anyon": 36898,
+      "鸡肉": 36899,
+      "igi": 36900,
+      "我也不": 36901,
+      "▁violations": 36902,
+      "区政府": 36903,
+      "iker": 36904,
+      "▁worthy": 36905,
+      "stub": 36906,
+      "小康": 36907,
+      "unn": 36908,
+      "ischer": 36909,
+      "▁soils": 36910,
+      "osocial": 36911,
+      "阴道": 36912,
+      "▁sometime": 36913,
+      "▁Cincinnati": 36914,
+      "体温": 36915,
+      "不同于": 36916,
+      "▁cents": 36917,
+      "Observ": 36918,
+      "仍在": 36919,
+      "▁measurable": 36920,
+      "大夫": 36921,
+      "▁xs": 36922,
+      "▁latent": 36923,
+      "因为在": 36924,
+      "微信公众号": 36925,
+      "犯罪嫌疑人": 36926,
+      "▁defenders": 36927,
+      "得起": 36928,
+      "▁MIN": 36929,
+      "Ro": 36930,
+      "和中": 36931,
+      "▁Demo": 36932,
+      "installed": 36933,
+      "半个": 36934,
+      "Hey": 36935,
+      "{()": 36936,
+      "aryngeal": 36937,
+      "周某": 36938,
+      "农民工": 36939,
+      "▁SSL": 36940,
+      "就不能": 36941,
+      "市场营销": 36942,
+      "▁Null": 36943,
+      "的家庭": 36944,
+      "明代": 36945,
+      "▁Detection": 36946,
+      "▁Optimization": 36947,
+      "筹备": 36948,
+      "好不好": 36949,
+      "自来水": 36950,
+      "China": 36951,
+      "inel": 36952,
+      "改革的": 36953,
+      "本科生": 36954,
+      "ardi": 36955,
+      "▁Terr": 36956,
+      "但被告": 36957,
+      "▁tanks": 36958,
+      "▁Repository": 36959,
+      "从不": 36960,
+      "唐代": 36961,
+      "近平": 36962,
+      "拥有一": 36963,
+      "▁wounds": 36964,
+      "zu": 36965,
+      "ische": 36966,
+      "十五章": 36967,
+      "▁Honda": 36968,
+      "协议书": 36969,
+      "所需的": 36970,
+      "bars": 36971,
+      "加盖": 36972,
+      "委托代理人张": 36973,
+      "mul": 36974,
+      "▁skew": 36975,
+      "portion": 36976,
+      "屁股": 36977,
+      "▁diarr": 36978,
+      "Metrics": 36979,
+      "emption": 36980,
+      "failure": 36981,
+      "商用": 36982,
+      "apa": 36983,
+      "管理办法": 36984,
+      "因为你": 36985,
+      "paid": 36986,
+      "对被申请人": 36987,
+      "现有的床铺": 36988,
+      "▁intim": 36989,
+      "小鼠": 36990,
+      "强劲": 36991,
+      "云计算": 36992,
+      "FORMAT": 36993,
+      "highlight": 36994,
+      "mentioned": 36995,
+      "▁bachelor": 36996,
+      "接地": 36997,
+      "Radio": 36998,
+      "▁fifty": 36999,
+      "回合": 37000,
+      "ité": 37001,
+      "~~\\": 37002,
+      "上科大": 37003,
+      "主义的": 37004,
+      "▁drift": 37005,
+      "九条": 37006,
+      "可能的": 37007,
+      "了一场": 37008,
+      "▁BMW": 37009,
+      "Toggle": 37010,
+      "商铺": 37011,
+      "为维护": 37012,
+      "▁Ultra": 37013,
+      "粒子": 37014,
+      "nSpanish": 37015,
+      "▁Cohen": 37016,
+      "四舍": 37017,
+      "发展战略": 37018,
+      "▁burned": 37019,
+      "遥控": 37020,
+      "dtype": 37021,
+      "▁unions": 37022,
+      "his": 37023,
+      "铝合金": 37024,
+      "▁Vic": 37025,
+      "onaut": 37026,
+      "}));\\": 37027,
+      "Fixture": 37028,
+      "限度": 37029,
+      "明白了": 37030,
+      "▁inspir": 37031,
+      "▁amplification": 37032,
+      "▁人口": 37033,
+      "nsys": 37034,
+      "▁Push": 37035,
+      "▁poets": 37036,
+      "nThough": 37037,
+      "▁Neither": 37038,
+      "ictory": 37039,
+      "▁touches": 37040,
+      "printStackTrace": 37041,
+      "欺骗": 37042,
+      "fat": 37043,
+      "申请执行人王": 37044,
+      "vh": 37045,
+      "项规定": 37046,
+      "退房时间": 37047,
+      "▁badly": 37048,
+      "起诉人": 37049,
+      "壮大": 37050,
+      "|}\\": 37051,
+      "▁multiplied": 37052,
+      "包装清单": 37053,
+      "scores": 37054,
+      "▁monetary": 37055,
+      "diagn": 37056,
+      "multiple": 37057,
+      "▁cosmetic": 37058,
+      "支球队": 37059,
+      "▁sculpture": 37060,
+      "多余": 37061,
+      "是小": 37062,
+      "被除数": 37063,
+      "▁Travis": 37064,
+      "▁silence": 37065,
+      "举行了": 37066,
+      "到最后": 37067,
+      "▁Columbus": 37068,
+      "还是在": 37069,
+      "CHAR": 37070,
+      "▁brut": 37071,
+      "世人": 37072,
+      "▁snippet": 37073,
+      "如果在": 37074,
+      "ipro": 37075,
+      "▁preserving": 37076,
+      "平生": 37077,
+      "▁ss": 37078,
+      "systems": 37079,
+      "▁workspace": 37080,
+      "ereum": 37081,
+      "▁downloads": 37082,
+      "national": 37083,
+      "▁continuation": 37084,
+      "就这么": 37085,
+      "▁nos": 37086,
+      "▁startups": 37087,
+      "教导": 37088,
+      "▁获奖": 37089,
+      "Focus": 37090,
+      "眼的": 37091,
+      "年生": 37092,
+      "▁Supply": 37093,
+      "▁Probability": 37094,
+      "cats": 37095,
+      "▁devastating": 37096,
+      "相反数": 37097,
+      "▁doub": 37098,
+      "本身的": 37099,
+      "说起": 37100,
+      "▁anis": 37101,
+      "下单": 37102,
+      "ighb": 37103,
+      "▁Pattern": 37104,
+      "▁proprietary": 37105,
+      "绑定": 37106,
+      "▁divisible": 37107,
+      "▁poker": 37108,
+      "▁cartoon": 37109,
+      "▁spelling": 37110,
+      "原谅": 37111,
+      "VELO": 37112,
+      "▁Commander": 37113,
+      "检刑诉": 37114,
+      "▁Edwards": 37115,
+      "很不": 37116,
+      "得上": 37117,
+      "新品": 37118,
+      "ycler": 37119,
+      "cosity": 37120,
+      "INF": 37121,
+      "博览会": 37122,
+      "slide": 37123,
+      "▁employs": 37124,
+      "日制": 37125,
+      "▁Nancy": 37126,
+      "▁binomial": 37127,
+      "▁toxin": 37128,
+      "thood": 37129,
+      "清风": 37130,
+      "商品的": 37131,
+      "赢得了": 37132,
+      "该怎么办": 37133,
+      "▁ugly": 37134,
+      "就想": 37135,
+      "icus": 37136,
+      "其中包括": 37137,
+      "▁mall": 37138,
+      "吃完": 37139,
+      "奈何": 37140,
+      "进修": 37141,
+      "median": 37142,
+      "换算": 37143,
+      "生化": 37144,
+      "需求的": 37145,
+      "阳县人民法院": 37146,
+      "和浩特": 37147,
+      "影响了": 37148,
+      "▁geographical": 37149,
+      "▁inadequate": 37150,
+      "规矩": 37151,
+      "除尘": 37152,
+      "宫颈": 37153,
+      "ersed": 37154,
+      "▁dear": 37155,
+      "▁thumb": 37156,
+      "博主": 37157,
+      "就诊": 37158,
+      "nable": 37159,
+      "▁Mills": 37160,
+      "▁Representative": 37161,
+      "告别": 37162,
+      "▁Brid": 37163,
+      "▁Harrison": 37164,
+      "提名": 37165,
+      "灿烂": 37166,
+      "马斯": 37167,
+      "▁Tang": 37168,
+      "▁seroton": 37169,
+      "▁Trial": 37170,
+      "星球": 37171,
+      "不影响": 37172,
+      "ishi": 37173,
+      "getMessage": 37174,
+      "海报": 37175,
+      "polar": 37176,
+      "omology": 37177,
+      "家常": 37178,
+      "wik": 37179,
+      "▁cousin": 37180,
+      "▁neurodeg": 37181,
+      "帐号": 37182,
+      "▁curved": 37183,
+      "矿业": 37184,
+      "orean": 37185,
+      "Take": 37186,
+      "bund": 37187,
+      "▁foundations": 37188,
+      "▁brass": 37189,
+      "▁potato": 37190,
+      "太高": 37191,
+      "nDirect": 37192,
+      "肖战": 37193,
+      "▁neighborhoods": 37194,
+      "外壳": 37195,
+      "▁Hob": 37196,
+      "党中央": 37197,
+      "▁heal": 37198,
+      "mathit": 37199,
+      "▁replies": 37200,
+      "▁tam": 37201,
+      "第一百条": 37202,
+      "clients": 37203,
+      "八十": 37204,
+      "rides": 37205,
+      "▁influencing": 37206,
+      "把这个": 37207,
+      "▁Generic": 37208,
+      "▁Singles": 37209,
+      "boat": 37210,
+      "奔驰": 37211,
+      "North": 37212,
+      "▁wool": 37213,
+      "▁camping": 37214,
+      "句子": 37215,
+      "▁evac": 37216,
+      "颁发": 37217,
+      "保鲜": 37218,
+      "量为": 37219,
+      "一顿": 37220,
+      "提倡": 37221,
+      "crypt": 37222,
+      "ibles": 37223,
+      "iropr": 37224,
+      "▁Wildlife": 37225,
+      "abwe": 37226,
+      "▁drought": 37227,
+      "最快": 37228,
+      "这样做": 37229,
+      "航班": 37230,
+      "Tip": 37231,
+      "ungen": 37232,
+      "numeric": 37233,
+      "石材": 37234,
+      "▁HER": 37235,
+      "nWrite": 37236,
+      "▁broadly": 37237,
+      "▁Operating": 37238,
+      "图的": 37239,
+      "富裕": 37240,
+      "注意到": 37241,
+      "▁ket": 37242,
+      "▁Clare": 37243,
+      "冻结被申请人": 37244,
+      "▁scattered": 37245,
+      "计算出": 37246,
+      "吴某": 37247,
+      "术前": 37248,
+      "HIV": 37249,
+      "oplus": 37250,
+      "▁tomato": 37251,
+      "deployment": 37252,
+      "地块": 37253,
+      "onc": 37254,
+      "ausal": 37255,
+      "▁societies": 37256,
+      "申请的": 37257,
+      "▁dan": 37258,
+      "▁statistic": 37259,
+      "方法对": 37260,
+      "▁Auckland": 37261,
+      "▁RES": 37262,
+      "eme": 37263,
+      "装箱": 37264,
+      "针灸": 37265,
+      "jin": 37266,
+      "▁Burg": 37267,
+      "今晚": 37268,
+      "你对": 37269,
+      "实录": 37270,
+      "烤箱": 37271,
+      "提供被执行人其他": 37272,
+      "▁beams": 37273,
+      "▁trapped": 37274,
+      "▁INC": 37275,
+      "firebase": 37276,
+      "决议": 37277,
+      "▁Bac": 37278,
+      "▁Serbia": 37279,
+      "▁prophyl": 37280,
+      "人际": 37281,
+      "女王": 37282,
+      "板的": 37283,
+      "xpath": 37284,
+      "▁topped": 37285,
+      "▁submissions": 37286,
+      "和生活": 37287,
+      "▁attrs": 37288,
+      "蘑菇": 37289,
+      "onn": 37290,
+      "▁Stevens": 37291,
+      "都将": 37292,
+      "归还借款": 37293,
+      "Geometry": 37294,
+      "枢纽": 37295,
+      "下列范围": 37296,
+      "工程施工": 37297,
+      "▁Zhang": 37298,
+      "的都是": 37299,
+      "中医药学": 37300,
+      "▁releg": 37301,
+      "nKeywords": 37302,
+      "启示": 37303,
+      "▁coupons": 37304,
+      "▁upgrades": 37305,
+      "▁Frederick": 37306,
+      "从判决确定之日起计算": 37307,
+      "长相": 37308,
+      "▁语种": 37309,
+      "▁tut": 37310,
+      "swift": 37311,
+      "▁shortest": 37312,
+      "olerance": 37313,
+      "▁stresses": 37314,
+      "▁mx": 37315,
+      "HY": 37316,
+      "▁Safari": 37317,
+      "想不到": 37318,
+      "主营": 37319,
+      "utral": 37320,
+      "时要": 37321,
+      "▁counterparts": 37322,
+      "HPP": 37323,
+      "bio": 37324,
+      "▁Flag": 37325,
+      "禁忌": 37326,
+      "例患者": 37327,
+      "Tele": 37328,
+      "April": 37329,
+      "双重": 37330,
+      "▁Protein": 37331,
+      "TCP": 37332,
+      "意见书": 37333,
+      "▁clue": 37334,
+      "▁Built": 37335,
+      "杰出": 37336,
+      "与中国科学院": 37337,
+      "能够在": 37338,
+      "▁gateway": 37339,
+      "右键": 37340,
+      "Sym": 37341,
+      "nWant": 37342,
+      "▁soldier": 37343,
+      "▁refugees": 37344,
+      "ない": 37345,
+      "bons": 37346,
+      "▁gard": 37347,
+      "硕士研究生": 37348,
+      "▁talents": 37349,
+      "Course": 37350,
+      "operatively": 37351,
+      "olicies": 37352,
+      "▁LCD": 37353,
+      "从来没有": 37354,
+      "▁Sever": 37355,
+      "不已": 37356,
+      "Gold": 37357,
+      "▁evolve": 37358,
+      "一无": 37359,
+      "升降": 37360,
+      "变迁": 37361,
+      "着你": 37362,
+      "▁rheumat": 37363,
+      "ovy": 37364,
+      "▁Va": 37365,
+      "等部门": 37366,
+      "研究成果": 37367,
+      "NSString": 37368,
+      "中断": 37369,
+      "iator": 37370,
+      "plots": 37371,
+      "Es": 37372,
+      "不难": 37373,
+      "jb": 37374,
+      "ROP": 37375,
+      "Safe": 37376,
+      "▁consciousness": 37377,
+      "▁manifestations": 37378,
+      "比利": 37379,
+      "▁Gill": 37380,
+      "▁Centers": 37381,
+      "堵塞": 37382,
+      "▁蔡": 37383,
+      "珍贵": 37384,
+      "▁boasts": 37385,
+      "银行股份有限公司信用卡中心": 37386,
+      "▁championships": 37387,
+      "卸载": 37388,
+      "▁Thanksgiving": 37389,
+      "吸引了": 37390,
+      "广泛的": 37391,
+      "辅导员": 37392,
+      "内容包括": 37393,
+      "发改": 37394,
+      "nDue": 37395,
+      "Facades": 37396,
+      "paragraph": 37397,
+      "marine": 37398,
+      "▁sympathetic": 37399,
+      "▁transported": 37400,
+      "一键": 37401,
+      "▁Implementation": 37402,
+      "无情": 37403,
+      "行车": 37404,
+      "WER": 37405,
+      "▁FF": 37406,
+      "▁Ultr": 37407,
+      "oshop": 37408,
+      "tol": 37409,
+      "辨认笔录": 37410,
+      "高等学校": 37411,
+      "luent": 37412,
+      "until": 37413,
+      "MATH": 37414,
+      "▁Conserv": 37415,
+      "▁bathrooms": 37416,
+      "▁DDR": 37417,
+      "呕吐": 37418,
+      "开始游戏": 37419,
+      "▁incentives": 37420,
+      "▁subscribers": 37421,
+      "录像": 37422,
+      "预售": 37423,
+      "的理解": 37424,
+      "▁credited": 37425,
+      "篇文章": 37426,
+      "声器": 37427,
+      "▁typed": 37428,
+      "▁praise": 37429,
+      "吉他": 37430,
+      "城堡": 37431,
+      "Unity": 37432,
+      "不过是": 37433,
+      "nSeveral": 37434,
+      "▁Registry": 37435,
+      "平分": 37436,
+      "encv": 37437,
+      "不准许撤诉": 37438,
+      "itated": 37439,
+      "▁reduct": 37440,
+      "▁homology": 37441,
+      "实惠": 37442,
+      "▁TC": 37443,
+      "texture": 37444,
+      "买家": 37445,
+      "prison": 37446,
+      "Fn": 37447,
+      "EES": 37448,
+      "心理健康": 37449,
+      "assisted": 37450,
+      "▁µ": 37451,
+      "二审案件受理费": 37452,
+      "花椒": 37453,
+      "intercept": 37454,
+      "▁dyst": 37455,
+      "▁Muh": 37456,
+      "▁counted": 37457,
+      "▁Tea": 37458,
+      "▁Usually": 37459,
+      "▁monkeys": 37460,
+      "得不到": 37461,
+      "淄博市": 37462,
+      "▁Brisbane": 37463,
+      "我了": 37464,
+      "顶端": 37465,
+      "领导班子": 37466,
+      "▁Woods": 37467,
+      "ivative": 37468,
+      "市面上": 37469,
+      "misc": 37470,
+      "CONTRIBUTING": 37471,
+      "解方程": 37472,
+      "▁Eug": 37473,
+      "维权": 37474,
+      "scient": 37475,
+      "▁predec": 37476,
+      "台风": 37477,
+      "▁amp": 37478,
+      "merged": 37479,
+      "MX": 37480,
+      "公社": 37481,
+      "口的": 37482,
+      "▁spiral": 37483,
+      "▁inappropriate": 37484,
+      "▁Tot": 37485,
+      "readme": 37486,
+      "▁stair": 37487,
+      "▁Associates": 37488,
+      "iasm": 37489,
+      "成为一个": 37490,
+      "▁Maximum": 37491,
+      "▁amyloid": 37492,
+      "▁BS": 37493,
+      "angelog": 37494,
+      "▁Frances": 37495,
+      "▁unified": 37496,
+      "▁Veterans": 37497,
+      "▁晋江文学网": 37498,
+      "图中": 37499,
+      "处理的": 37500,
+      "▁{},": 37501,
+      "骨头": 37502,
+      "特色服务": 37503,
+      "▁ImportError": 37504,
+      "敬业": 37505,
+      "▁Direction": 37506,
+      "▁specializes": 37507,
+      "thumb": 37508,
+      "▁volt": 37509,
+      "法定代表人张": 37510,
+      "切片": 37511,
+      "逃离": 37512,
+      "▁\"(": 37513,
+      "▁pestic": 37514,
+      "字数": 37515,
+      "前三": 37516,
+      "▁styling": 37517,
+      "▁Healthcare": 37518,
+      "bis": 37519,
+      "stdin": 37520,
+      "主场": 37521,
+      "▁cues": 37522,
+      "Cursor": 37523,
+      "▁backyard": 37524,
+      "▁Parameters": 37525,
+      "utm": 37526,
+      "horizontal": 37527,
+      "书证": 37528,
+      "nEX": 37529,
+      "▁Guild": 37530,
+      "formatter": 37531,
+      "▁'__": 37532,
+      "paces": 37533,
+      "▁deserves": 37534,
+      "▁disturbance": 37535,
+      "剧中": 37536,
+      "))*": 37537,
+      "有两种": 37538,
+      "日经": 37539,
+      "▁CBD": 37540,
+      "这些问题": 37541,
+      "▁assure": 37542,
+      "▁Midd": 37543,
+      "▁outpatient": 37544,
+      "就看": 37545,
+      "指纹": 37546,
+      "anya": 37547,
+      "ré": 37548,
+      "▁sug": 37549,
+      "▁mate": 37550,
+      "▁Saskat": 37551,
+      "MMMMMMMM": 37552,
+      "要做": 37553,
+      "nse": 37554,
+      "按月利率": 37555,
+      "acional": 37556,
+      "▁acknowledge": 37557,
+      "▁practically": 37558,
+      "畅销": 37559,
+      "Src": 37560,
+      "说着": 37561,
+      "jsx": 37562,
+      "INCLUDE": 37563,
+      "执着": 37564,
+      "排行榜": 37565,
+      "cool": 37566,
+      "izens": 37567,
+      "▁adul": 37568,
+      "▁galleries": 37569,
+      "▁hammer": 37570,
+      "▁defective": 37571,
+      "rp": 37572,
+      "火山": 37573,
+      "baz": 37574,
+      "亲爱的": 37575,
+      "▁pear": 37576,
+      "▁bundles": 37577,
+      "▁seminar": 37578,
+      "marker": 37579,
+      "▁rankings": 37580,
+      "Guid": 37581,
+      "价值的财产": 37582,
+      "是将": 37583,
+      "些什么": 37584,
+      "'$": 37585,
+      "诊治": 37586,
+      "nCustom": 37587,
+      "▁formatted": 37588,
+      "对上述": 37589,
+      "atics": 37590,
+      "理想的": 37591,
+      "毁灭": 37592,
+      "Care": 37593,
+      "FFFF": 37594,
+      "vectors": 37595,
+      "▁Workers": 37596,
+      "倒数": 37597,
+      "agi": 37598,
+      "找房": 37599,
+      "会议室": 37600,
+      "网站的": 37601,
+      "奖惩审批表": 37602,
+      "在审理过程中": 37603,
+      "网络科技有限公司": 37604,
+      "现将": 37605,
+      "理智": 37606,
+      "手术治疗": 37607,
+      "年平均": 37608,
+      "easy": 37609,
+      "▁Vac": 37610,
+      "ippets": 37611,
+      "于本判决生效之日起十日内": 37612,
+      "▁潘": 37613,
+      "所选": 37614,
+      "二期": 37615,
+      "新产品": 37616,
+      ";&#": 37617,
+      "▁Launch": 37618,
+      "不胜": 37619,
+      "总线": 37620,
+      "增值税": 37621,
+      "▁Teacher": 37622,
+      "保湿": 37623,
+      "June": 37624,
+      "▁strom": 37625,
+      "astically": 37626,
+      "南宁市": 37627,
+      "日常生活中": 37628,
+      "▁remot": 37629,
+      "硫酸": 37630,
+      "第一条": 37631,
+      "▁Aur": 37632,
+      "应有": 37633,
+      "oux": 37634,
+      "国科大": 37635,
+      "扬子": 37636,
+      "nod": 37637,
+      "cale": 37638,
+      "▁flies": 37639,
+      "这三个": 37640,
+      "▁caregivers": 37641,
+      "与社会": 37642,
+      "▁inoc": 37643,
+      "nRight": 37644,
+      "roduce": 37645,
+      "▁additions": 37646,
+      "▁coherent": 37647,
+      "一来": 37648,
+      "故人": 37649,
+      "▁Imaging": 37650,
+      "发表于晋江文学网": 37651,
+      "奢侈": 37652,
+      "生导师": 37653,
+      "知道了": 37654,
+      "下标": 37655,
+      "lord": 37656,
+      "nWin": 37657,
+      "▁wanna": 37658,
+      "▁Anyone": 37659,
+      "聚餐": 37660,
+      "▁sine": 37661,
+      "执导": 37662,
+      "膳食": 37663,
+      "▁autumn": 37664,
+      "▁polymorphisms": 37665,
+      "匆匆": 37666,
+      "我爸": 37667,
+      "jdk": 37668,
+      "Seconds": 37669,
+      "▁Peninsula": 37670,
+      "滋味": 37671,
+      "▁重要参数": 37672,
+      "▁immunosup": 37673,
+      "志强": 37674,
+      "axes": 37675,
+      "ANGUAGE": 37676,
+      "▁Adelaide": 37677,
+      "vb": 37678,
+      "邮寄": 37679,
+      "玩游戏": 37680,
+      "opor": 37681,
+      "▁drill": 37682,
+      "Confirm": 37683,
+      "▁bicycle": 37684,
+      "▁Recovery": 37685,
+      "复旦": 37686,
+      "CLO": 37687,
+      "中档题": 37688,
+      "Modules": 37689,
+      "iversal": 37690,
+      "溶剂": 37691,
+      "▁fist": 37692,
+      "houses": 37693,
+      "无人机": 37694,
+      "igon": 37695,
+      "Begin": 37696,
+      "万事": 37697,
+      "财产调查": 37698,
+      "getClass": 37699,
+      "惊讶": 37700,
+      "▁wore": 37701,
+      "▁Actor": 37702,
+      "▁backward": 37703,
+      "▁aure": 37704,
+      "▁stems": 37705,
+      "▁spanning": 37706,
+      "发型": 37707,
+      "▁Newcastle": 37708,
+      "方差": 37709,
+      "明细": 37710,
+      "Missing": 37711,
+      "极具": 37712,
+      "海峡": 37713,
+      "▁CYP": 37714,
+      "effects": 37715,
+      "▁albumin": 37716,
+      "endum": 37717,
+      "▁behave": 37718,
+      "▁dw": 37719,
+      "▁bootstrap": 37720,
+      "▁competence": 37721,
+      "URN": 37722,
+      "降温": 37723,
+      "先进个人": 37724,
+      "▁quart": 37725,
+      "变异": 37726,
+      "FAIL": 37727,
+      "▁illustrations": 37728,
+      "nCong": 37729,
+      "▁farmer": 37730,
+      "▁recruiting": 37731,
+      "时针": 37732,
+      "至关": 37733,
+      "conditions": 37734,
+      "溪市": 37735,
+      "▁drying": 37736,
+      "▁魏": 37737,
+      "躺在": 37738,
+      "▁theat": 37739,
+      "nDesign": 37740,
+      "▁twisted": 37741,
+      "区内": 37742,
+      "很强": 37743,
+      "▁Jobs": 37744,
+      "▁酒店地址": 37745,
+      "▁verbal": 37746,
+      "▁depressed": 37747,
+      "力争": 37748,
+      "能力强": 37749,
+      "Drive": 37750,
+      "名师": 37751,
+      "them": 37752,
+      "▁Lloyd": 37753,
+      "▁deadly": 37754,
+      "▁patience": 37755,
+      "eterm": 37756,
+      "开辟": 37757,
+      "先用": 37758,
+      "天猫": 37759,
+      "ocarcinoma": 37760,
+      "tcolor": 37761,
+      "从前": 37762,
+      "此事": 37763,
+      "首个": 37764,
+      "现场勘": 37765,
+      "个百分点": 37766,
+      "▁WILL": 37767,
+      "buttons": 37768,
+      "▁meditation": 37769,
+      "colon": 37770,
+      "ogical": 37771,
+      "▁Split": 37772,
+      "Refresh": 37773,
+      "▁condem": 37774,
+      "▁Heights": 37775,
+      "出厂": 37776,
+      "演变": 37777,
+      "也不用": 37778,
+      "ustion": 37779,
+      "▁Illegal": 37780,
+      "aram": 37781,
+      "酒店提供": 37782,
+      "Getter": 37783,
+      "▁Perth": 37784,
+      "ahu": 37785,
+      "变压器": 37786,
+      "所有权": 37787,
+      "第一百一十八条": 37788,
+      "▁phylogen": 37789,
+      "常州市": 37790,
+      "二次函数": 37791,
+      "▁Dennis": 37792,
+      "公立": 37793,
+      "▁advocacy": 37794,
+      "▁occlusion": 37795,
+      "▁tutorials": 37796,
+      "Sun": 37797,
+      "kon": 37798,
+      "实现的": 37799,
+      "▁ritual": 37800,
+      "imbabwe": 37801,
+      "▁Export": 37802,
+      "▁Linda": 37803,
+      "▁predecess": 37804,
+      "罪犯王": 37805,
+      "▁snack": 37806,
+      "山村": 37807,
+      "合法性": 37808,
+      "aris": 37809,
+      "▁前": 37810,
+      "校友": 37811,
+      "fon": 37812,
+      "\\'}\\": 37813,
+      "▁lev": 37814,
+      "中国特色": 37815,
+      "▁Experiment": 37816,
+      "hw": 37817,
+      "▁plasticity": 37818,
+      "▁fundraising": 37819,
+      "bugs": 37820,
+      "nimpl": 37821,
+      "▁pork": 37822,
+      "▁insulation": 37823,
+      "edo": 37824,
+      "nok": 37825,
+      "必须在": 37826,
+      "▁rhyth": 37827,
+      "式子": 37828,
+      "guid": 37829,
+      "Preview": 37830,
+      "号民事判决书已发生法律效力": 37831,
+      "(\"<": 37832,
+      "▁PF": 37833,
+      "nWest": 37834,
+      "ullivan": 37835,
+      "教科": 37836,
+      "Land": 37837,
+      "门外": 37838,
+      "Cross": 37839,
+      "▁Nintendo": 37840,
+      "did": 37841,
+      "enery": 37842,
+      "▁smokers": 37843,
+      "▁tasty": 37844,
+      "帖子": 37845,
+      "Docs": 37846,
+      "umen": 37847,
+      "可以缺席判决": 37848,
+      "这几个": 37849,
+      "SESSION": 37850,
+      "到你": 37851,
+      "一般情况下": 37852,
+      "})(": 37853,
+      "uded": 37854,
+      "astruct": 37855,
+      "凸显": 37856,
+      "录制": 37857,
+      "▁Nigerian": 37858,
+      "▁prolifer": 37859,
+      "不给": 37860,
+      "团购": 37861,
+      "▁anx": 37862,
+      "细胞的": 37863,
+      "▁编剧": 37864,
+      "▁planted": 37865,
+      "▁Election": 37866,
+      "▁古": 37867,
+      "ARM": 37868,
+      "ixon": 37869,
+      "oseph": 37870,
+      "▁Heavy": 37871,
+      "▁salmon": 37872,
+      "▁Duration": 37873,
+      "isRequired": 37874,
+      "残酷": 37875,
+      "Expr": 37876,
+      "jective": 37877,
+      "妇科": 37878,
+      "▁testosterone": 37879,
+      "对被告人": 37880,
+      "oices": 37881,
+      "▁beats": 37882,
+      "例外": 37883,
+      "▁onion": 37884,
+      "号刑事判决书": 37885,
+      "falls": 37886,
+      "▁Coin": 37887,
+      "▁aviation": 37888,
+      "音箱": 37889,
+      "省政府": 37890,
+      "▁TCP": 37891,
+      "Documents": 37892,
+      "以前的": 37893,
+      "然后在": 37894,
+      "然后用": 37895,
+      "等内容": 37896,
+      "充实": 37897,
+      "国籍": 37898,
+      "Ready": 37899,
+      "recogn": 37900,
+      "习近平": 37901,
+      "儿童和": 37902,
+      "一年的": 37903,
+      "▁NOTE": 37904,
+      "西区": 37905,
+      "Numer": 37906,
+      "▁'',\\": 37907,
+      "海底": 37908,
+      "客场": 37909,
+      "nHom": 37910,
+      "▁marking": 37911,
+      "Bul": 37912,
+      "要不要": 37913,
+      "▁volatility": 37914,
+      "veh": 37915,
+      "MySQL": 37916,
+      "▁Liberty": 37917,
+      "tis": 37918,
+      "▁Naz": 37919,
+      "人民日报": 37920,
+      "▁encaps": 37921,
+      "修养": 37922,
+      "▁feeds": 37923,
+      "Helpers": 37924,
+      "寒冷": 37925,
+      "值班": 37926,
+      "颁布": 37927,
+      "\"][\"": 37928,
+      "idyl": 37929,
+      "▁餐馆名称": 37930,
+      "学业": 37931,
+      "的吗": 37932,
+      "xsl": 37933,
+      "▁planar": 37934,
+      "DESCRIPTION": 37935,
+      "增进": 37936,
+      "只要你": 37937,
+      "▁creep": 37938,
+      "压迫": 37939,
+      "重型": 37940,
+      "中方": 37941,
+      "是一本": 37942,
+      "ENGTH": 37943,
+      "乒乓球": 37944,
+      "▁presumably": 37945,
+      "\\'<": 37946,
+      "这两种": 37947,
+      "▁Hus": 37948,
+      "nio": 37949,
+      "新闻网": 37950,
+      "最多容纳": 37951,
+      "▁Eagle": 37952,
+      "▁executing": 37953,
+      "▁LPS": 37954,
+      "高质量发展": 37955,
+      "▁pumps": 37956,
+      "PET": 37957,
+      "干事": 37958,
+      "▁selector": 37959,
+      "思念": 37960,
+      "Social": 37961,
+      "▁iterations": 37962,
+      "opard": 37963,
+      "especially": 37964,
+      "MK": 37965,
+      "还不如": 37966,
+      "deps": 37967,
+      "如果不是": 37968,
+      "▁simulator": 37969,
+      "坚实": 37970,
+      "契约": 37971,
+      "longitude": 37972,
+      "月下": 37973,
+      "说他": 37974,
+      "▁植物": 37975,
+      "毫克": 37976,
+      "难忘": 37977,
+      "现已审查终结": 37978,
+      "▁xrange": 37979,
+      "▁cascade": 37980,
+      "▁stimulating": 37981,
+      "}⋅": 37982,
+      "朝阳区": 37983,
+      "▁Soon": 37984,
+      "▁Links": 37985,
+      "不予受理": 37986,
+      "油田": 37987,
+      "▁主要成就": 37988,
+      "最重要的是": 37989,
+      "▁carotid": 37990,
+      "二元": 37991,
+      "▁contour": 37992,
+      "▁underway": 37993,
+      "选中": 37994,
+      "中型": 37995,
+      "当做": 37996,
+      "海水": 37997,
+      "可行性": 37998,
+      "独资": 37999,
+      "纪录片": 38000,
+      "▁Reports": 38001,
+      "▁prosper": 38002,
+      "▁tensorflow": 38003,
+      "废物": 38004,
+      "▁healthier": 38005,
+      "▁伟": 38006,
+      "▁cp": 38007,
+      "▁sqlalchemy": 38008,
+      "HK": 38009,
+      "不懈": 38010,
+      "Hom": 38011,
+      "▁cherry": 38012,
+      "▁profitable": 38013,
+      "习俗": 38014,
+      "▁适宜": 38015,
+      "▁grave": 38016,
+      "▁nicotine": 38017,
+      "▁程": 38018,
+      "财政部": 38019,
+      "身上的": 38020,
+      "▁Ana": 38021,
+      "▁Lok": 38022,
+      "▁BT": 38023,
+      "Books": 38024,
+      "▁unsure": 38025,
+      "▁compares": 38026,
+      "▁Jefferson": 38027,
+      "受众": 38028,
+      "▁Laser": 38029,
+      "▁executable": 38030,
+      "▁Born": 38031,
+      "vr": 38032,
+      "在全球": 38033,
+      "urbs": 38034,
+      "Called": 38035,
+      "机箱": 38036,
+      "过往": 38037,
+      "gomery": 38038,
+      "隔壁": 38039,
+      "▁normalize": 38040,
+      "nLook": 38041,
+      "▁orchestr": 38042,
+      "同意终结本次执行程序": 38043,
+      "得解": 38044,
+      "▁anatomy": 38045,
+      "成熟的": 38046,
+      "icky": 38047,
+      "▁spouse": 38048,
+      "▁Shipping": 38049,
+      "▁volatile": 38050,
+      "▁Birthday": 38051,
+      "周一": 38052,
+      "才华": 38053,
+      "水位": 38054,
+      "PCI": 38055,
+      "▁房间": 38056,
+      "▁depletion": 38057,
+      "ouis": 38058,
+      "ellows": 38059,
+      "▁touched": 38060,
+      "收款": 38061,
+      "教职工": 38062,
+      "bies": 38063,
+      "credit": 38064,
+      "又会": 38065,
+      "挣扎": 38066,
+      "敢于": 38067,
+      "Gs": 38068,
+      "cord": 38069,
+      "ettes": 38070,
+      "uator": 38071,
+      "平民": 38072,
+      "▁Sustain": 38073,
+      "▁biomass": 38074,
+      "取名": 38075,
+      "可以是": 38076,
+      "▁sentenced": 38077,
+      "的法律": 38078,
+      "raper": 38079,
+      "阴影": 38080,
+      "第七十八": 38081,
+      "Dataset": 38082,
+      "比为": 38083,
+      "\\\\*": 38084,
+      "日被刑事拘留": 38085,
+      "technology": 38086,
+      "我所": 38087,
+      "可用于": 38088,
+      "是国内": 38089,
+      "▁railroad": 38090,
+      "▁GOP": 38091,
+      "EXPECT": 38092,
+      "UserId": 38093,
+      "▁densities": 38094,
+      "格尔": 38095,
+      "natural": 38096,
+      "Constant": 38097,
+      "▁threading": 38098,
+      "彩礼": 38099,
+      "▁TEST": 38100,
+      "跳舞": 38101,
+      "tpl": 38102,
+      "中国经济": 38103,
+      "▁alloc": 38104,
+      "▁Salmon": 38105,
+      "流感": 38106,
+      "rossover": 38107,
+      "▁cfg": 38108,
+      "▁cation": 38109,
+      "ViewById": 38110,
+      "ди": 38111,
+      "fasterxml": 38112,
+      "▁Nas": 38113,
+      "的白": 38114,
+      "奖学金": 38115,
+      "管理中心": 38116,
+      "▁commemor": 38117,
+      "nUnincorporated": 38118,
+      "▁Guardian": 38119,
+      "▁userId": 38120,
+      "殖民": 38121,
+      "▁CAP": 38122,
+      "▁Reader": 38123,
+      "▁reactor": 38124,
+      "自古": 38125,
+      "▁Career": 38126,
+      "▁anatomical": 38127,
+      "国道": 38128,
+      "而且还": 38129,
+      "公诉机关指控的": 38130,
+      "dirs": 38131,
+      "二代": 38132,
+      "▁stap": 38133,
+      "▁Athens": 38134,
+      "comfortable": 38135,
+      "nOct": 38136,
+      "verb": 38137,
+      "抗菌": 38138,
+      "aspberry": 38139,
+      "清朝": 38140,
+      "nTest": 38141,
+      "▁harness": 38142,
+      "偏差": 38143,
+      "印发": 38144,
+      "并为": 38145,
+      "▁founders": 38146,
+      "▁understands": 38147,
+      "▁deline": 38148,
+      "government": 38149,
+      "▁Ny": 38150,
+      "▁generalization": 38151,
+      "海域": 38152,
+      "IGH": 38153,
+      "▁Cluster": 38154,
+      "▁Hour": 38155,
+      "第一百零三": 38156,
+      "▁resear": 38157,
+      "Multiple": 38158,
+      "两家": 38159,
+      "▁Gam": 38160,
+      "mania": 38161,
+      "准予强制执行": 38162,
+      "采样": 38163,
+      "sessions": 38164,
+      "▁Nursing": 38165,
+      "nid": 38166,
+      "oya": 38167,
+      "▁mas": 38168,
+      "的了": 38169,
+      "逃避": 38170,
+      "的主题": 38171,
+      "的症状": 38172,
+      "nSur": 38173,
+      "▁Often": 38174,
+      "▁drunk": 38175,
+      "▁explosion": 38176,
+      "光线": 38177,
+      "threat": 38178,
+      "▁seizures": 38179,
+      "▁deprecated": 38180,
+      "▁rebuild": 38181,
+      "一副": 38182,
+      "限期": 38183,
+      "▁bos": 38184,
+      "cerpt": 38185,
+      "▁Defence": 38186,
+      "二〇一五年十二月": 38187,
+      "胆固醇": 38188,
+      "erie": 38189,
+      "'''": 38190,
+      "nBen": 38191,
+      "▁Paulo": 38192,
+      "和工作": 38193,
+      "▁collaborate": 38194,
+      "人之": 38195,
+      "要使": 38196,
+      "npro": 38197,
+      "暴雨": 38198,
+      "▁rope": 38199,
+      "离婚纠纷一案中": 38200,
+      "Solution": 38201,
+      "代书": 38202,
+      "蒸发": 38203,
+      "▁Lis": 38204,
+      "户籍所在地": 38205,
+      "ctuary": 38206,
+      "▁Brother": 38207,
+      "Her": 38208,
+      "gang": 38209,
+      "iphone": 38210,
+      "rogate": 38211,
+      "▁fibrin": 38212,
+      "世界各地": 38213,
+      "不同类型的": 38214,
+      "不和": 38215,
+      "估算": 38216,
+      "时时彩": 38217,
+      "▁Byte": 38218,
+      "▁unused": 38219,
+      ":<": 38220,
+      "冠状": 38221,
+      "联系方式": 38222,
+      "▁Hillary": 38223,
+      "可以把": 38224,
+      "影响的": 38225,
+      "▁capturing": 38226,
+      "▁dismissed": 38227,
+      "这事": 38228,
+      "限定": 38229,
+      "iggs": 38230,
+      "▁pituitary": 38231,
+      "下辖": 38232,
+      "生产经营": 38233,
+      "▁pharmacy": 38234,
+      "▁successive": 38235,
+      "抱歉": 38236,
+      "电线": 38237,
+      "麒麟": 38238,
+      "批准后": 38239,
+      "Deep": 38240,
+      "▁Lac": 38241,
+      "▁extern": 38242,
+      "▁modifying": 38243,
+      "ın": 38244,
+      "尺度": 38245,
+      "督查": 38246,
+      "自愿认罪": 38247,
+      "捐款": 38248,
+      "合唱": 38249,
+      "▁型号": 38250,
+      "体现在": 38251,
+      "ursors": 38252,
+      "被盗": 38253,
+      "日依法": 38254,
+      "▁Cru": 38255,
+      "一点的": 38256,
+      "fram": 38257,
+      "oxic": 38258,
+      "中国科学院大学": 38259,
+      "analytics": 38260,
+      "▁Developer": 38261,
+      "▁HV": 38262,
+      "▁Tesla": 38263,
+      "曲线的": 38264,
+      "▁monster": 38265,
+      "菏泽": 38266,
+      "▁affiliates": 38267,
+      "华盛": 38268,
+      "任何人": 38269,
+      "特别的": 38270,
+      "▁innings": 38271,
+      "故宫": 38272,
+      "▁ssh": 38273,
+      "roleum": 38274,
+      "万的": 38275,
+      "获利": 38276,
+      "衰竭": 38277,
+      "LET": 38278,
+      "▁Sak": 38279,
+      "partner": 38280,
+      "nDefunct": 38281,
+      "就得": 38282,
+      "华北": 38283,
+      "▁Static": 38284,
+      "idas": 38285,
+      "▁inaugural": 38286,
+      "市新": 38287,
+      "犹如": 38288,
+      "▁spy": 38289,
+      "▁seniors": 38290,
+      "派遣": 38291,
+      "MODEL": 38292,
+      "▁wast": 38293,
+      "▁Neural": 38294,
+      "学子": 38295,
+      "都会有": 38296,
+      "经人介绍": 38297,
+      "▁青": 38298,
+      "该项目": 38299,
+      "macro": 38300,
+      "产的": 38301,
+      "nId": 38302,
+      "amboo": 38303,
+      "handed": 38304,
+      "▁rises": 38305,
+      "▁sepsis": 38306,
+      "▁realm": 38307,
+      "▁lipids": 38308,
+      "粉末": 38309,
+      "▁DOI": 38310,
+      "brities": 38311,
+      "▁spokesperson": 38312,
+      "村里": 38313,
+      "lesh": 38314,
+      "▁mounting": 38315,
+      "面子": 38316,
+      "npmjs": 38317,
+      "smith": 38318,
+      "率的": 38319,
+      "理论和": 38320,
+      "中和": 38321,
+      "就医": 38322,
+      "▁Municipal": 38323,
+      "高新": 38324,
+      "▁Fit": 38325,
+      "fabric": 38326,
+      "▁Asset": 38327,
+      "博弈": 38328,
+      "▁modulus": 38329,
+      "很可能": 38330,
+      "的职业": 38331,
+      "钢结构": 38332,
+      "nDiv": 38333,
+      "▁perimeter": 38334,
+      "开采": 38335,
+      "▁sper": 38336,
+      "堪称": 38337,
+      "yze": 38338,
+      "与你": 38339,
+      "前途": 38340,
+      "数目": 38341,
+      "角是": 38342,
+      "itas": 38343,
+      "▁flush": 38344,
+      "应为": 38345,
+      "行政执法": 38346,
+      "▁chaos": 38347,
+      "▁Programs": 38348,
+      "前列腺": 38349,
+      "支付令": 38350,
+      "otides": 38351,
+      "ccc": 38352,
+      "inj": 38353,
+      "nLocal": 38354,
+      "▁Pseud": 38355,
+      "▁recombination": 38356,
+      "▁Mong": 38357,
+      "eredReader": 38358,
+      "特意": 38359,
+      "太多的": 38360,
+      "▁headache": 38361,
+      "诉至本院": 38362,
+      "食品有限公司": 38363,
+      "▁merger": 38364,
+      "出品": 38365,
+      "应聘": 38366,
+      "时有": 38367,
+      "▁Videos": 38368,
+      "▁approve": 38369,
+      "▁starred": 38370,
+      "名额": 38371,
+      "如实供述自己的罪行": 38372,
+      "底下": 38373,
+      "▁flap": 38374,
+      "▁晋江文学城": 38375,
+      "植株": 38376,
+      "osphere": 38377,
+      "velocity": 38378,
+      "的神": 38379,
+      "él": 38380,
+      "觉醒": 38381,
+      "▁Debug": 38382,
+      "ān": 38383,
+      "交通肇事罪": 38384,
+      "▁alerts": 38385,
+      "▁pygame": 38386,
+      "nDemographics": 38387,
+      "产业化": 38388,
+      "ulsive": 38389,
+      "法定代表人李": 38390,
+      "徐州市": 38391,
+      "的对象": 38392,
+      "azor": 38393,
+      "▁vet": 38394,
+      "Human": 38395,
+      "ensure": 38396,
+      "\\\\\\\\\\": 38397,
+      "▁snake": 38398,
+      "▁Motors": 38399,
+      "▁congreg": 38400,
+      "▁已知": 38401,
+      "heads": 38402,
+      "磷酸": 38403,
+      "▁rainfall": 38404,
+      "的自": 38405,
+      "ées": 38406,
+      "的文章": 38407,
+      "▁lept": 38408,
+      "▁loader": 38409,
+      "▁atherosclerosis": 38410,
+      "整洁": 38411,
+      "ilio": 38412,
+      "▁boil": 38413,
+      "▁Latino": 38414,
+      "ushi": 38415,
+      "inski": 38416,
+      "和水": 38417,
+      "杀死": 38418,
+      "▁Outdoor": 38419,
+      "真理": 38420,
+      "色素": 38421,
+      "停车位": 38422,
+      "▁enforce": 38423,
+      "NL": 38424,
+      "asta": 38425,
+      "nell": 38426,
+      "datas": 38427,
+      "交点": 38428,
+      "BIT": 38429,
+      "▁WiFi": 38430,
+      "▁fabrics": 38431,
+      "▁Phillips": 38432,
+      "哪儿": 38433,
+      "驳回起诉": 38434,
+      "▁carbox": 38435,
+      "▁blessed": 38436,
+      "▁filmed": 38437,
+      "好用": 38438,
+      "就是这样": 38439,
+      "▁Colors": 38440,
+      "ndefined": 38441,
+      "iba": 38442,
+      "▁pins": 38443,
+      "▁renewed": 38444,
+      "月光": 38445,
+      "你需要": 38446,
+      "马路": 38447,
+      "他人的": 38448,
+      "共青团": 38449,
+      "的角色": 38450,
+      "共同举办": 38451,
+      "▁Parameter": 38452,
+      "▁Fro": 38453,
+      "reload": 38454,
+      "一审被告": 38455,
+      "裁定适用于": 38456,
+      "labeled": 38457,
+      "sel": 38458,
+      "年月": 38459,
+      "起义": 38460,
+      "modern": 38461,
+      "的个数": 38462,
+      "▁scala": 38463,
+      "的重要组成部分": 38464,
+      "Products": 38465,
+      "anni": 38466,
+      "happy": 38467,
+      "▁lasted": 38468,
+      "▁indications": 38469,
+      "▁accessory": 38470,
+      "▁debuted": 38471,
+      "▁observational": 38472,
+      ":'": 38473,
+      "议员": 38474,
+      "▁Mik": 38475,
+      "在向": 38476,
+      "的内": 38477,
+      "▁Butter": 38478,
+      "会儿": 38479,
+      "ictions": 38480,
+      "House": 38481,
+      "在水": 38482,
+      "性疾病": 38483,
+      "LowerCase": 38484,
+      "近年": 38485,
+      "▁crashes": 38486,
+      "▁apparatus": 38487,
+      "但如果": 38488,
+      "官方网站": 38489,
+      "狭窄": 38490,
+      "的成功": 38491,
+      "ação": 38492,
+      "GU": 38493,
+      "Same": 38494,
+      "简要": 38495,
+      "bee": 38496,
+      "的语言": 38497,
+      "igrate": 38498,
+      "▁彭": 38499,
+      "有力的": 38500,
+      "的一首": 38501,
+      "Validate": 38502,
+      "▁hemoglobin": 38503,
+      "IFF": 38504,
+      "DJ": 38505,
+      "河县": 38506,
+      "Mvc": 38507,
+      "rosophila": 38508,
+      "ocortic": 38509,
+      "▁Weekly": 38510,
+      "▁lymphocyte": 38511,
+      "utt": 38512,
+      "acier": 38513,
+      "流转": 38514,
+      "龙岩": 38515,
+      "任何一个": 38516,
+      "▁pasta": 38517,
+      "▁comorb": 38518,
+      "▁jumped": 38519,
+      "任教": 38520,
+      "行政机关": 38521,
+      "▁pound": 38522,
+      "会产生": 38523,
+      "新中国的": 38524,
+      "terone": 38525,
+      "▁Cyber": 38526,
+      "未作答辩": 38527,
+      "collapse": 38528,
+      "的限制": 38529,
+      "leigh": 38530,
+      "▁administrators": 38531,
+      "oubt": 38532,
+      "nHead": 38533,
+      "▁constrained": 38534,
+      "kafka": 38535,
+      "finish": 38536,
+      "ERC": 38537,
+      "gev": 38538,
+      "销售额": 38539,
+      "stown": 38540,
+      "缓刑考验期限": 38541,
+      "ametric": 38542,
+      "▁medial": 38543,
+      "冲动": 38544,
+      "孵化": 38545,
+      "杀菌": 38546,
+      "roe": 38547,
+      "高尔夫": 38548,
+      "其实就是": 38549,
+      "▁Latest": 38550,
+      "▁institute": 38551,
+      "▁董": 38552,
+      "\"];\\": 38553,
+      "▁halt": 38554,
+      "▁形态特征": 38555,
+      "▁antip": 38556,
+      "▁antagonists": 38557,
+      "erness": 38558,
+      "▁cytoplasmic": 38559,
+      "mov": 38560,
+      "起伏": 38561,
+      "▁Raf": 38562,
+      "cores": 38563,
+      ")%": 38564,
+      "▁disclaimer": 38565,
+      "CAL": 38566,
+      "▁sclerosis": 38567,
+      "reb": 38568,
+      "eight": 38569,
+      "mongodb": 38570,
+      "亲人": 38571,
+      "▁fridge": 38572,
+      "▁crushing": 38573,
+      "科学与": 38574,
+      "▁urged": 38575,
+      "▁Philosoph": 38576,
+      "▁anticipate": 38577,
+      "nIns": 38578,
+      "▁jest": 38579,
+      "ductory": 38580,
+      "裁定适用于下列范围": 38581,
+      "▁glutamate": 38582,
+      "第三次": 38583,
+      "Printf": 38584,
+      "▁poses": 38585,
+      "▁permutation": 38586,
+      "gm": 38587,
+      "华侨": 38588,
+      "▁diving": 38589,
+      "封面": 38590,
+      "的想法": 38591,
+      "▁broadband": 38592,
+      "▁内": 38593,
+      "others": 38594,
+      "▁sentiment": 38595,
+      "▁&\\\\": 38596,
+      "请仔细阅读": 38597,
+      "▁masks": 38598,
+      "发行的": 38599,
+      "papers": 38600,
+      "▁dialysis": 38601,
+      "▁exceeded": 38602,
+      "▁Formation": 38603,
+      "迎来了": 38604,
+      "▁Sara": 38605,
+      "algebra": 38606,
+      "▁neurotrans": 38607,
+      "centered": 38608,
+      "▁hierarchical": 38609,
+      "建有": 38610,
+      "泉州": 38611,
+      "Fill": 38612,
+      "▁mattress": 38613,
+      "▁outreach": 38614,
+      "女性的": 38615,
+      "不好意思": 38616,
+      "▁fibre": 38617,
+      "TextView": 38618,
+      "▁deposited": 38619,
+      "Camp": 38620,
+      "▁transcripts": 38621,
+      "光辉": 38622,
+      "ointers": 38623,
+      "rotation": 38624,
+      "Align": 38625,
+      "网路": 38626,
+      "一共有": 38627,
+      "deleted": 38628,
+      "▁cleans": 38629,
+      "▁originated": 38630,
+      "▁zh": 38631,
+      "PHONE": 38632,
+      "oden": 38633,
+      "的一年": 38634,
+      "▁hungry": 38635,
+      "etti": 38636,
+      "▁SAP": 38637,
+      "技术创新": 38638,
+      "Columns": 38639,
+      "盯着": 38640,
+      "辩解": 38641,
+      "ylan": 38642,
+      "▁Depending": 38643,
+      "AZ": 38644,
+      "Cover": 38645,
+      "▁struggled": 38646,
+      "丰田": 38647,
+      "可乐": 38648,
+      "的处分": 38649,
+      "▁fade": 38650,
+      "▁triangular": 38651,
+      "\"\"\\": 38652,
+      "▁topical": 38653,
+      "▁Buch": 38654,
+      "arctic": 38655,
+      "本案诉讼费用": 38656,
+      "▁enclosed": 38657,
+      "▁profiling": 38658,
+      "▁Sans": 38659,
+      "▁listeners": 38660,
+      "▁symptomatic": 38661,
+      "Der": 38662,
+      "▁Rd": 38663,
+      "目的地": 38664,
+      "▁Than": 38665,
+      ">•": 38666,
+      "名牌": 38667,
+      "塑胶": 38668,
+      "为我们": 38669,
+      "▁AZ": 38670,
+      "大概是": 38671,
+      "▁architects": 38672,
+      "白了": 38673,
+      "的实际": 38674,
+      "▁Novel": 38675,
+      "although": 38676,
+      "▁sailing": 38677,
+      "▁nominations": 38678,
+      "▁oz": 38679,
+      "同理": 38680,
+      "报导": 38681,
+      "集中在": 38682,
+      "共同财产": 38683,
+      "社会保险": 38684,
+      "nDraw": 38685,
+      "具有良好的": 38686,
+      "▁interleukin": 38687,
+      "营业执照": 38688,
+      "▁adop": 38689,
+      "Matcher": 38690,
+      "▁privilege": 38691,
+      "nBlack": 38692,
+      "imports": 38693,
+      "▁reminded": 38694,
+      "油画": 38695,
+      "第一天": 38696,
+      "▁paralle": 38697,
+      "扣留": 38698,
+      "▁lol": 38699,
+      "▁Giants": 38700,
+      "▁cardiomy": 38701,
+      "▁inactive": 38702,
+      "▁CONNECTION": 38703,
+      "jwt": 38704,
+      "▁性质": 38705,
+      "▁Serge": 38706,
+      "tW": 38707,
+      "狐狸": 38708,
+      "薄弱": 38709,
+      "▁历史": 38710,
+      "收回": 38711,
+      "时节": 38712,
+      "▁Wells": 38713,
+      "层层": 38714,
+      "确实是": 38715,
+      "Cast": 38716,
+      "▁Nine": 38717,
+      "▁doubled": 38718,
+      "工程学院": 38719,
+      "▁tastes": 38720,
+      "ltimate": 38721,
+      "▁pulses": 38722,
+      "的黑": 38723,
+      "GIS": 38724,
+      "Notes": 38725,
+      "ugoslav": 38726,
+      "▁multiplying": 38727,
+      "观音": 38728,
+      "Angle": 38729,
+      "phosphate": 38730,
+      "刷卡": 38731,
+      "/~": 38732,
+      "工伤": 38733,
+      "在国际": 38734,
+      "▁shrub": 38735,
+      "feeding": 38736,
+      "▁م": 38737,
+      "相乘": 38738,
+      "odos": 38739,
+      "▁eco": 38740,
+      "hagen": 38741,
+      "▁mongoose": 38742,
+      "▁guitarist": 38743,
+      "▁incubation": 38744,
+      "经历过": 38745,
+      "还请": 38746,
+      "unders": 38747,
+      "▁Angular": 38748,
+      "体系的": 38749,
+      "▁solver": 38750,
+      "Pb": 38751,
+      "Sensor": 38752,
+      "▁cables": 38753,
+      "▁tribut": 38754,
+      "马鞍": 38755,
+      "Ops": 38756,
+      "▁tenure": 38757,
+      "历代": 38758,
+      "▁体重": 38759,
+      "ipation": 38760,
+      "▁swelling": 38761,
+      "]`": 38762,
+      "问问": 38763,
+      "住所地安徽省": 38764,
+      "的红": 38765,
+      "▁loyalty": 38766,
+      "▁我们": 38767,
+      "▁LOG": 38768,
+      "reeze": 38769,
+      "▁Songs": 38770,
+      "acl": 38771,
+      "▁glue": 38772,
+      "Members": 38773,
+      "并经": 38774,
+      "要说": 38775,
+      "reddit": 38776,
+      "▁balcon": 38777,
+      "Comments": 38778,
+      "Sent": 38779,
+      "Strings": 38780,
+      "Pipeline": 38781,
+      "胰岛": 38782,
+      "▁SHA": 38783,
+      "结果显示": 38784,
+      "只想": 38785,
+      "nup": 38786,
+      "年全国": 38787,
+      "判断即可": 38788,
+      "我将": 38789,
+      "NOS": 38790,
+      "▁revenues": 38791,
+      "地形": 38792,
+      "作为一种": 38793,
+      "的那个": 38794,
+      "人均纯收入": 38795,
+      "MPLATE": 38796,
+      "堆积": 38797,
+      "▁branding": 38798,
+      "▁interpolation": 38799,
+      "stem": 38800,
+      "▁а": 38801,
+      "查控": 38802,
+      "▁ACC": 38803,
+      "有所不同": 38804,
+      "投案": 38805,
+      "内容为": 38806,
+      "女方": 38807,
+      "先求出": 38808,
+      "当然是": 38809,
+      "朋友的": 38810,
+      "CMAKE": 38811,
+      "brains": 38812,
+      "刻意": 38813,
+      "又在": 38814,
+      "孔子": 38815,
+      "尸体": 38816,
+      "明日": 38817,
+      "irteen": 38818,
+      "EAR": 38819,
+      "dro": 38820,
+      "随访": 38821,
+      "▁silk": 38822,
+      "治疗后": 38823,
+      "有大": 38824,
+      "zt": 38825,
+      "▁σ": 38826,
+      "▁Stephan": 38827,
+      "nExpand": 38828,
+      "Movie": 38829,
+      "Stock": 38830,
+      "符合减刑条件": 38831,
+      "▁definitive": 38832,
+      "田园": 38833,
+      "nany": 38834,
+      "▁enriched": 38835,
+      "▁manga": 38836,
+      "▁黑": 38837,
+      "Aud": 38838,
+      "▁Photos": 38839,
+      "废水": 38840,
+      "▁ocular": 38841,
+      "的事项": 38842,
+      "arlier": 38843,
+      "rotate": 38844,
+      "▁Brexit": 38845,
+      "▁Throw": 38846,
+      "▁connector": 38847,
+      "chor": 38848,
+      "onna": 38849,
+      "Shift": 38850,
+      "nHope": 38851,
+      "▁webpack": 38852,
+      "▁historically": 38853,
+      "游戏大小": 38854,
+      "保罗": 38855,
+      "市的": 38856,
+      "本期": 38857,
+      "气味": 38858,
+      "slider": 38859,
+      "▁camps": 38860,
+      "minutes": 38861,
+      "▁supplementation": 38862,
+      "入职": 38863,
+      "}}</": 38864,
+      "▁Mine": 38865,
+      "▁Rac": 38866,
+      "Christ": 38867,
+      "▁stained": 38868,
+      "▁Discussion": 38869,
+      "脱落": 38870,
+      "▁GMT": 38871,
+      "scheme": 38872,
+      "热线": 38873,
+      "还是会": 38874,
+      "▁Scar": 38875,
+      "的他": 38876,
+      "fib": 38877,
+      "lings": 38878,
+      "prompt": 38879,
+      "valued": 38880,
+      "预交案件受理费": 38881,
+      "减法": 38882,
+      "nSam": 38883,
+      "▁TreeNode": 38884,
+      "Subscription": 38885,
+      "局势": 38886,
+      "的积": 38887,
+      "orescence": 38888,
+      "借贷关系": 38889,
+      "无一": 38890,
+      "omitem": 38891,
+      "▁\"~": 38892,
+      "amation": 38893,
+      "GW": 38894,
+      "ça": 38895,
+      "中信": 38896,
+      "兴起": 38897,
+      "▁Bee": 38898,
+      "Sender": 38899,
+      "▁excav": 38900,
+      "不分": 38901,
+      "网的": 38902,
+      "无所谓": 38903,
+      "▁bikes": 38904,
+      "杂交": 38905,
+      "▁Guest": 38906,
+      ",,,,": 38907,
+      "ções": 38908,
+      "amburg": 38909,
+      "第四百八十七条": 38910,
+      "▁heights": 38911,
+      "被打": 38912,
+      "uploads": 38913,
+      "弘历": 38914,
+      "intent": 38915,
+      "▁exempl": 38916,
+      "▁proofs": 38917,
+      "consider": 38918,
+      "crumb": 38919,
+      "▁Alban": 38920,
+      "▁aromatic": 38921,
+      "买入": 38922,
+      "▁subt": 38923,
+      "▁offspring": 38924,
+      "小小的": 38925,
+      "nLong": 38926,
+      "▁纲": 38927,
+      "它在": 38928,
+      "nno": 38929,
+      "▁是否": 38930,
+      "__()\\": 38931,
+      "ymphony": 38932,
+      "怪兽": 38933,
+      "优质的": 38934,
+      "Soft": 38935,
+      "▁intentions": 38936,
+      "墙面": 38937,
+      "案经过": 38938,
+      "▁Leaf": 38939,
+      "身体的": 38940,
+      "priv": 38941,
+      "▁Selection": 38942,
+      "联通": 38943,
+      "的一面": 38944,
+      "▁bulb": 38945,
+      "stead": 38946,
+      "▁monkey": 38947,
+      "伴随": 38948,
+      "引流": 38949,
+      "装的": 38950,
+      "otional": 38951,
+      "omitempty": 38952,
+      "▁exclusion": 38953,
+      "▁quad": 38954,
+      "▁Florence": 38955,
+      "罪犯张": 38956,
+      "atex": 38957,
+      "歌唱": 38958,
+      "▁MMP": 38959,
+      "的脸": 38960,
+      "▁ai": 38961,
+      "ergy": 38962,
+      "reshape": 38963,
+      "nRun": 38964,
+      "▁lod": 38965,
+      "nSearch": 38966,
+      "▁sooner": 38967,
+      "▁garbage": 38968,
+      "互补": 38969,
+      "ellij": 38970,
+      "nsource": 38971,
+      "▁shelves": 38972,
+      "▁cann": 38973,
+      "身份证号码": 38974,
+      "▁Thirty": 38975,
+      "申请执行人李": 38976,
+      "审议": 38977,
+      "红楼": 38978,
+      "°\\": 38979,
+      "负有": 38980,
+      "nFROM": 38981,
+      "▁Teaching": 38982,
+      "},{": 38983,
+      "座谈会": 38984,
+      "bins": 38985,
+      "▁LOVE": 38986,
+      "▁所属专辑": 38987,
+      "icz": 38988,
+      "▁BD": 38989,
+      "▁lithium": 38990,
+      "▁playground": 38991,
+      "过户": 38992,
+      "flies": 38993,
+      "▁Tile": 38994,
+      "Algorithm": 38995,
+      "饺子": 38996,
+      "▁dich": 38997,
+      "我把": 38998,
+      "了一条": 38999,
+      "▁hemis": 39000,
+      ":[": 39001,
+      "最美": 39002,
+      "了一系列": 39003,
+      "stdlib": 39004,
+      "lap": 39005,
+      "莫斯科": 39006,
+      "▁Zen": 39007,
+      "▁Views": 39008,
+      "ourcing": 39009,
+      "应届": 39010,
+      "▁retailer": 39011,
+      "兼顾": 39012,
+      "▁}$": 39013,
+      "▁Papers": 39014,
+      "nSometimes": 39015,
+      "▁struggles": 39016,
+      "乘方": 39017,
+      "▁External": 39018,
+      "▁monitors": 39019,
+      "▁obligation": 39020,
+      "▁日本": 39021,
+      "很多人都": 39022,
+      "ismic": 39023,
+      "ployed": 39024,
+      "▁optic": 39025,
+      "nProduction": 39026,
+      "▁illustration": 39027,
+      "Github": 39028,
+      "ialysis": 39029,
+      "该校": 39030,
+      "Measure": 39031,
+      "▁percussion": 39032,
+      "解毒": 39033,
+      "▁equip": 39034,
+      "nGeorge": 39035,
+      "几千": 39036,
+      "索要": 39037,
+      "一案中": 39038,
+      "理事会": 39039,
+      "的功效": 39040,
+      "尚欠": 39041,
+      "蓝天": 39042,
+      "ingle": 39043,
+      "二〇一五年十一月": 39044,
+      "CAT": 39045,
+      "mob": 39046,
+      "▁Iv": 39047,
+      "专职": 39048,
+      "▁retrieved": 39049,
+      "HttpServlet": 39050,
+      "个股": 39051,
+      "zes": 39052,
+      "平方分米": 39053,
+      "▁jointly": 39054,
+      "三四": 39055,
+      "是家": 39056,
+      "Pet": 39057,
+      "大连市": 39058,
+      "▁Sons": 39059,
+      "▁Tall": 39060,
+      "▁operative": 39061,
+      "▁Counsel": 39062,
+      "冰雪": 39063,
+      "nBusiness": 39064,
+      "▁subtypes": 39065,
+      "此款": 39066,
+      "MMARY": 39067,
+      "rails": 39068,
+      "ogan": 39069,
+      "Routes": 39070,
+      "Compiler": 39071,
+      "▁adrenal": 39072,
+      "Dom": 39073,
+      "▁electrodes": 39074,
+      "硬币": 39075,
+      "ndon": 39076,
+      "▁azure": 39077,
+      "beam": 39078,
+      "▁ambitious": 39079,
+      "enson": 39080,
+      "具体应用法律": 39081,
+      "ellington": 39082,
+      "serialize": 39083,
+      "▁sponsors": 39084,
+      "wedge": 39085,
+      "▁Encycl": 39086,
+      "向左": 39087,
+      "▁ninth": 39088,
+      "▁Lithuan": 39089,
+      "用过": 39090,
+      "站长": 39091,
+      "是一件": 39092,
+      "yarn": 39093,
+      "giene": 39094,
+      "nRequ": 39095,
+      "▁terminals": 39096,
+      "▁complimentary": 39097,
+      "▁Kun": 39098,
+      "生活中的": 39099,
+      "▁fighter": 39100,
+      "▁conserv": 39101,
+      "▁preceding": 39102,
+      "▁余": 39103,
+      "▁Gang": 39104,
+      "▁vide": 39105,
+      "▁manifolds": 39106,
+      "}\\\\}": 39107,
+      "▁MHz": 39108,
+      "uating": 39109,
+      "往下": 39110,
+      "都区": 39111,
+      "▁xxx": 39112,
+      "▁Harbor": 39113,
+      "▁leakage": 39114,
+      "▁archives": 39115,
+      "▁邓": 39116,
+      "一类": 39117,
+      "比我": 39118,
+      "abl": 39119,
+      "是我的": 39120,
+      "▁synthase": 39121,
+      "▁associates": 39122,
+      "WEB": 39123,
+      "民政局": 39124,
+      "▁Lit": 39125,
+      "免疫力": 39126,
+      "▁Layer": 39127,
+      "▁lamin": 39128,
+      "要求与被告离婚": 39129,
+      "urrences": 39130,
+      "▁selectivity": 39131,
+      "▁(%": 39132,
+      "▁MAN": 39133,
+      "nSelect": 39134,
+      "诚实": 39135,
+      "dummy": 39136,
+      "▁房型介绍": 39137,
+      "委托代理人李": 39138,
+      "唐山": 39139,
+      "文科": 39140,
+      "▁Around": 39141,
+      "▁stenosis": 39142,
+      "▁merchandise": 39143,
+      "奇葩": 39144,
+      "上海浦东": 39145,
+      "▁adorable": 39146,
+      "Turn": 39147,
+      "corn": 39148,
+      "工业大学": 39149,
+      "ometown": 39150,
+      "\\'>\\": 39151,
+      "▁rape": 39152,
+      "▁Butler": 39153,
+      "光源": 39154,
+      "克尔": 39155,
+      "▁QU": 39156,
+      "amba": 39157,
+      "osaic": 39158,
+      "▁vars": 39159,
+      "Sprintf": 39160,
+      "的范围内处分自己的": 39161,
+      "nCollect": 39162,
+      "单位和": 39163,
+      "唯有": 39164,
+      "$/": 39165,
+      "还挺": 39166,
+      "Wrap": 39167,
+      "▁dece": 39168,
+      "▁fifteen": 39169,
+      "▁footprint": 39170,
+      "无声": 39171,
+      "uncture": 39172,
+      "▁slopes": 39173,
+      "icates": 39174,
+      "▁catching": 39175,
+      "KG": 39176,
+      "rystall": 39177,
+      "MESSAGE": 39178,
+      "▁Respon": 39179,
+      "▁fictional": 39180,
+      "追逐": 39181,
+      "风格的": 39182,
+      "ospor": 39183,
+      "nExper": 39184,
+      "至上": 39185,
+      "fan": 39186,
+      "grav": 39187,
+      "请你根据": 39188,
+      "mile": 39189,
+      "刑事责任": 39190,
+      "余弦": 39191,
+      "▁peaked": 39192,
+      "reported": 39193,
+      "在自己": 39194,
+      "▁Cord": 39195,
+      "▁buzz": 39196,
+      "▁Catalog": 39197,
+      "oelectric": 39198,
+      "▁slave": 39199,
+      "▁statue": 39200,
+      "▁palette": 39201,
+      "▁markdown": 39202,
+      "▁manipulate": 39203,
+      "hspace": 39204,
+      "▁reminds": 39205,
+      "叙述": 39206,
+      "always": 39207,
+      "scribed": 39208,
+      "▁dissemination": 39209,
+      "到一个": 39210,
+      "metal": 39211,
+      "vscode": 39212,
+      "discord": 39213,
+      "▁이": 39214,
+      "Dot": 39215,
+      "的生命": 39216,
+      "oflu": 39217,
+      "▁drone": 39218,
+      "第十六": 39219,
+      "▁classifier": 39220,
+      "▁Experiments": 39221,
+      "碎片": 39222,
+      "AH": 39223,
+      "▁sunny": 39224,
+      "▁invert": 39225,
+      "勤奋": 39226,
+      "水费": 39227,
+      "▁excretion": 39228,
+      "▁processors": 39229,
+      "nad": 39230,
+      "支气管": 39231,
+      "▁hike": 39232,
+      "热搜": 39233,
+      "▁successor": 39234,
+      "什么都": 39235,
+      "bler": 39236,
+      "▁performers": 39237,
+      "strand": 39238,
+      "▁molar": 39239,
+      "strftime": 39240,
+      "抬头": 39241,
+      "警示": 39242,
+      "▁Lucas": 39243,
+      "▁mentoring": 39244,
+      "LB": 39245,
+      "常住": 39246,
+      "▁Kill": 39247,
+      "在刑罚": 39248,
+      "的心情": 39249,
+      "▁antic": 39250,
+      "dg": 39251,
+      "▁shake": 39252,
+      "ilingual": 39253,
+      "▁hospitality": 39254,
+      "▁肖": 39255,
+      "国人": 39256,
+      "主人公": 39257,
+      "都配有": 39258,
+      "正式批准": 39259,
+      "Registration": 39260,
+      "以太": 39261,
+      "有好": 39262,
+      "锦标": 39263,
+      "胜利的": 39264,
+      "∞）": 39265,
+      "清算": 39266,
+      "nTour": 39267,
+      "▁Nicholas": 39268,
+      "▁constitutional": 39269,
+      "军的": 39270,
+      "TRA": 39271,
+      "监视": 39272,
+      "选择一个": 39273,
+      "▁feat": 39274,
+      "中心的": 39275,
+      "ufeffnamespace": 39276,
+      "oeing": 39277,
+      "▁packs": 39278,
+      "interpret": 39279,
+      "使人": 39280,
+      "贴近": 39281,
+      "犯罪的": 39282,
+      "▁DEAL": 39283,
+      "scatter": 39284,
+      "▁surgeons": 39285,
+      "▁augmented": 39286,
+      "专业化": 39287,
+      "▁Rachel": 39288,
+      "▁Roof": 39289,
+      "いる": 39290,
+      "▁capillary": 39291,
+      "副市长": 39292,
+      "▁Lub": 39293,
+      "Calcul": 39294,
+      "履行生效法律文书": 39295,
+      "▁Resolution": 39296,
+      "▁parliamentary": 39297,
+      "何必": 39298,
+      "比较大": 39299,
+      "▁Hopkins": 39300,
+      "无正当理由拒不到庭的": 39301,
+      "vere": 39302,
+      "▁Contest": 39303,
+      "▁grouped": 39304,
+      "▁Initiative": 39305,
+      "入户": 39306,
+      "木质": 39307,
+      "风采": 39308,
+      "第二条": 39309,
+      "技术有限公司": 39310,
+      "▁Provide": 39311,
+      "求解即可": 39312,
+      "▁retina": 39313,
+      "▁wholesale": 39314,
+      "Enumerable": 39315,
+      "▁vm": 39316,
+      "去医院": 39317,
+      "附近的": 39318,
+      "itely": 39319,
+      "nsetup": 39320,
+      "▁PATH": 39321,
+      "▁scary": 39322,
+      "还有就是": 39323,
+      "▁Fill": 39324,
+      "tdefer": 39325,
+      "▁grandfather": 39326,
+      "抑郁症": 39327,
+      "▁forty": 39328,
+      "之日起五日内": 39329,
+      "NER": 39330,
+      "nOb": 39331,
+      "▁modalities": 39332,
+      "色谱": 39333,
+      "管的": 39334,
+      "重症": 39335,
+      "icers": 39336,
+      "汇聚": 39337,
+      "(\\'%": 39338,
+      "▁confined": 39339,
+      "▁contested": 39340,
+      "▁estimator": 39341,
+      "但被执行人至今未履行": 39342,
+      "不相": 39343,
+      "疑似": 39344,
+      "wallet": 39345,
+      "abilistic": 39346,
+      "▁readings": 39347,
+      "▁enzymatic": 39348,
+      "▁visualize": 39349,
+      "ICA": 39350,
+      "▁Timer": 39351,
+      "▁hassle": 39352,
+      "nShort": 39353,
+      "transferase": 39354,
+      "还算": 39355,
+      "▁deficient": 39356,
+      "白银": 39357,
+      "银行账户": 39358,
+      "▁Distance": 39359,
+      "chet": 39360,
+      "ticket": 39361,
+      "▁Railroad": 39362,
+      "悠久": 39363,
+      "月息": 39364,
+      "自然数": 39365,
+      "▁Plain": 39366,
+      "▁Kerala": 39367,
+      "wordpress": 39368,
+      "▁facilitates": 39369,
+      "Pen": 39370,
+      "物业公司": 39371,
+      "icaid": 39372,
+      "った": 39373,
+      "换成": 39374,
+      "▁Flask": 39375,
+      "▁discrep": 39376,
+      "▁nowhere": 39377,
+      "▁thrilled": 39378,
+      "翡翠": 39379,
+      "静静": 39380,
+      "一个整数": 39381,
+      "▁Danny": 39382,
+      "的本": 39383,
+      "衣物": 39384,
+      "打印机": 39385,
+      "▁iod": 39386,
+      "传送": 39387,
+      "贝壳找房": 39388,
+      "▁NOTICE": 39389,
+      "ät": 39390,
+      "sted": 39391,
+      "▁英文": 39392,
+      "▁captures": 39393,
+      "关于多学科": 39394,
+      "idelity": 39395,
+      "关于多学科知识的选择题": 39396,
+      "伯特": 39397,
+      "▁taxp": 39398,
+      "outheastern": 39399,
+      "lify": 39400,
+      "▁Parts": 39401,
+      "流体": 39402,
+      "您需要": 39403,
+      "等特点": 39404,
+      "▁jam": 39405,
+      "▁battles": 39406,
+      "广电": 39407,
+      "不合格": 39408,
+      "学生在": 39409,
+      "ablo": 39410,
+      "nAlex": 39411,
+      "▁Harvey": 39412,
+      "Modified": 39413,
+      "▁vaginal": 39414,
+      "判处有期徒刑一年": 39415,
+      "最高人民法院关于执行": 39416,
+      "感叹": 39417,
+      "行驶至": 39418,
+      "▁dispens": 39419,
+      "▁quarterback": 39420,
+      "人们对": 39421,
+      "▁Mental": 39422,
+      "当我们": 39423,
+      "▁bats": 39424,
+      "▁Stories": 39425,
+      "▁Shot": 39426,
+      "██": 39427,
+      "cine": 39428,
+      "▁餐馆简介": 39429,
+      "▁Transaction": 39430,
+      "▁illustrates": 39431,
+      "Footer": 39432,
+      "nHonours": 39433,
+      "会是": 39434,
+      "▁Dot": 39435,
+      "▁Sed": 39436,
+      "▁Spa": 39437,
+      "▁GABA": 39438,
+      "▁卢": 39439,
+      "nAir": 39440,
+      "akespe": 39441,
+      "▁再": 39442,
+      "古今": 39443,
+      "折磨": 39444,
+      "exus": 39445,
+      "否决": 39446,
+      "deal": 39447,
+      "nMal": 39448,
+      "解决问题的": 39449,
+      "穿过": 39450,
+      "关系式": 39451,
+      "可作为": 39452,
+      "▁Arctic": 39453,
+      "和的": 39454,
+      "编著": 39455,
+      "延期": 39456,
+      "▁dire": 39457,
+      "▁histopath": 39458,
+      "ethylene": 39459,
+      "▁Developers": 39460,
+      "{$": 39461,
+      "nSem": 39462,
+      "▁hay": 39463,
+      "npy": 39464,
+      "▁Led": 39465,
+      "以下这道": 39466,
+      "▁prisoners": 39467,
+      "下面这道选择题": 39468,
+      "▁或": 39469,
+      "belongs": 39470,
+      "▁Describe": 39471,
+      "预告": 39472,
+      "部部长": 39473,
+      "健康发展": 39474,
+      "由上海市": 39475,
+      "▁秦": 39476,
+      "保暖": 39477,
+      "战国": 39478,
+      "Energy": 39479,
+      "aaS": 39480,
+      "devices": 39481,
+      "▁corpus": 39482,
+      "▁ablation": 39483,
+      "▁stressful": 39484,
+      "▁renovation": 39485,
+      "\\\\_{": 39486,
+      "▁herbs": 39487,
+      "▁meteor": 39488,
+      "▁Calgary": 39489,
+      "滚动": 39490,
+      "Destroy": 39491,
+      "俗称": 39492,
+      "三十一日": 39493,
+      "▁Desktop": 39494,
+      "▁corneal": 39495,
+      "案件的": 39496,
+      "▁disturbances": 39497,
+      "▁è": 39498,
+      "Pac": 39499,
+      "ungle": 39500,
+      "▁eval": 39501,
+      "▁Canyon": 39502,
+      "▁clicked": 39503,
+      "eon": 39504,
+      "▁lint": 39505,
+      "▁neglect": 39506,
+      "村民小组": 39507,
+      "▁coordinator": 39508,
+      "▁Rangers": 39509,
+      "▁Dad": 39510,
+      "▁kingdom": 39511,
+      "丙烯": 39512,
+      "选题": 39513,
+      "mology": 39514,
+      "沥青": 39515,
+      "肾脏": 39516,
+      "outed": 39517,
+      "路由器": 39518,
+      "▁Interestingly": 39519,
+      "综合运用多": 39520,
+      "Surface": 39521,
+      "的公": 39522,
+      "并从给出的选项中": 39523,
+      "▁Programme": 39524,
+      "下半": 39525,
+      "年参加": 39526,
+      "▁pier": 39527,
+      "▁premiered": 39528,
+      "}`);\\": 39529,
+      "▁网络": 39530,
+      "共同建设": 39531,
+      "ctype": 39532,
+      "被执行人有": 39533,
+      "管委会": 39534,
+      "renew": 39535,
+      "▁fond": 39536,
+      "AspNetCore": 39537,
+      "yro": 39538,
+      "(`${": 39539,
+      "▁wounded": 39540,
+      "单个": 39541,
+      "duplicate": 39542,
+      "清爽": 39543,
+      "▁一个": 39544,
+      "earth": 39545,
+      "aurus": 39546,
+      "熟悉的": 39547,
+      "▁landscapes": 39548,
+      "nurl": 39549,
+      "附带民事": 39550,
+      "angered": 39551,
+      "VIS": 39552,
+      "是需要": 39553,
+      "NOW": 39554,
+      "请你仔细": 39555,
+      "gesterone": 39556,
+      "ocene": 39557,
+      "tthrow": 39558,
+      "National": 39559,
+      "良心": 39560,
+      "自动撤回上诉处理": 39561,
+      "专著": 39562,
+      "olis": 39563,
+      "****\\": 39564,
+      "▁内容介绍": 39565,
+      "低价": 39566,
+      "毒素": 39567,
+      "▁URLs": 39568,
+      "▁sera": 39569,
+      "▁housed": 39570,
+      "▁lining": 39571,
+      "北美": 39572,
+      "行政复议": 39573,
+      "▁Alger": 39574,
+      "Percent": 39575,
+      "产妇": 39576,
+      "nISBN": 39577,
+      "用科学记数法": 39578,
+      "thro": 39579,
+      "▁Popular": 39580,
+      "传真": 39581,
+      "▁memo": 39582,
+      "▁Plans": 39583,
+      "酒店的房型": 39584,
+      "manuel": 39585,
+      "ophysical": 39586,
+      "陈某某": 39587,
+      "了就": 39588,
+      "将近": 39589,
+      "▁表示": 39590,
+      "为一体": 39591,
+      "为您提供": 39592,
+      "变化的": 39593,
+      "征集": 39594,
+      "▁Banks": 39595,
+      "樱花": 39596,
+      "▁Bog": 39597,
+      "差异有统计学意义": 39598,
+      "▁responsiveness": 39599,
+      "lich": 39600,
+      "▁TypeError": 39601,
+      "不禁": 39602,
+      "oucher": 39603,
+      "▁GD": 39604,
+      "▁wages": 39605,
+      "多位": 39606,
+      "完全平方": 39607,
+      "▁paral": 39608,
+      "制品有限公司": 39609,
+      "▁Doppler": 39610,
+      "▁'\\\\": 39611,
+      "▁cass": 39612,
+      "▁touring": 39613,
+      "▁wrestling": 39614,
+      "▁Hamiltonian": 39615,
+      "新中国": 39616,
+      "获得的": 39617,
+      "▁curr": 39618,
+      "▁Croatia": 39619,
+      "▁threatening": 39620,
+      "nStill": 39621,
+      "▁sacrific": 39622,
+      "Wall": 39623,
+      "▁Nit": 39624,
+      "多大": 39625,
+      "LCJ": 39626,
+      "ylum": 39627,
+      "文化传播": 39628,
+      "▁Karn": 39629,
+      "▁barn": 39630,
+      "nHouses": 39631,
+      "木耳": 39632,
+      "▁axial": 39633,
+      "原名": 39634,
+      "CLC": 39635,
+      "elia": 39636,
+      "▁Volunte": 39637,
+      "千金": 39638,
+      "敏捷": 39639,
+      "订阅": 39640,
+      "aru": 39641,
+      "今后的": 39642,
+      "终结本案": 39643,
+      "bishop": 39644,
+      "▁steal": 39645,
+      "颈椎": 39646,
+      "Include": 39647,
+      "没钱": 39648,
+      "▁fost": 39649,
+      "▁axios": 39650,
+      "借口": 39651,
+      "umab": 39652,
+      "杀了": 39653,
+      "▁Kas": 39654,
+      "逾期利息": 39655,
+      "几个选项": 39656,
+      "▁aerial": 39657,
+      "▁pickle": 39658,
+      "Men": 39659,
+      "phinx": 39660,
+      "▁metro": 39661,
+      "▁superf": 39662,
+      "{→": 39663,
+      "по": 39664,
+      "应按": 39665,
+      ">)\\": 39666,
+      "SIG": 39667,
+      "县支行": 39668,
+      "因式分解": 39669,
+      "▁Cells": 39670,
+      "▁hasattr": 39671,
+      "▁GDP": 39672,
+      "▁abort": 39673,
+      "波兰": 39674,
+      "ungal": 39675,
+      "▁Emily": 39676,
+      "镜子": 39677,
+      "的传统": 39678,
+      "的心态": 39679,
+      "jiang": 39680,
+      "▁epoch": 39681,
+      "▁tactics": 39682,
+      "nme": 39683,
+      "找工作": 39684,
+      "Tra": 39685,
+      "nBase": 39686,
+      "把手": 39687,
+      "▁Dialog": 39688,
+      "▁beings": 39689,
+      "签下": 39690,
+      "色泽": 39691,
+      "▁REF": 39692,
+      "建设项目": 39693,
+      "登记信息": 39694,
+      "erton": 39695,
+      "年经教育部": 39696,
+      "▁phrases": 39697,
+      "▁depressive": 39698,
+      "泉水": 39699,
+      "离开了": 39700,
+      "bred": 39701,
+      "▁Moz": 39702,
+      "▁subspace": 39703,
+      "communications": 39704,
+      "▁authorization": 39705,
+      "分钟后": 39706,
+      "IDTH": 39707,
+      "▁cn": 39708,
+      "第十一": 39709,
+      "▁Sugar": 39710,
+      "▁witnessed": 39711,
+      "FRING": 39712,
+      "snapshot": 39713,
+      "车道": 39714,
+      "othic": 39715,
+      "额外的": 39716,
+      "Internet": 39717,
+      "iman": 39718,
+      "roadcast": 39719,
+      "右手": 39720,
+      "▁coping": 39721,
+      "▁Educational": 39722,
+      "同心": 39723,
+      "Slot": 39724,
+      "REMOVED": 39725,
+      "▁Partner": 39726,
+      "材料有限公司": 39727,
+      "▁praised": 39728,
+      "因被告": 39729,
+      "▁hug": 39730,
+      "醉酒驾驶机动车": 39731,
+      "置业": 39732,
+      "nergic": 39733,
+      "▁translations": 39734,
+      "届中国": 39735,
+      "nStand": 39736,
+      "▁rewarding": 39737,
+      "▁retrospect": 39738,
+      "蓬勃": 39739,
+      "新农村": 39740,
+      "VIDIA": 39741,
+      "cube": 39742,
+      "▁sacrifice": 39743,
+      "ernate": 39744,
+      "中奖": 39745,
+      "电竞": 39746,
+      "PACK": 39747,
+      "端正": 39748,
+      "的商品": 39749,
+      "▁kicked": 39750,
+      "中市": 39751,
+      "串行": 39752,
+      "azer": 39753,
+      "▁excel": 39754,
+      "elasticsearch": 39755,
+      "了点": 39756,
+      "不建议": 39757,
+      "Develop": 39758,
+      "茅台": 39759,
+      "]);": 39760,
+      "Named": 39761,
+      "▁crafts": 39762,
+      "反之": 39763,
+      "precision": 39764,
+      "▁workaround": 39765,
+      "▁productions": 39766,
+      "前辈": 39767,
+      "Face": 39768,
+      "Phil": 39769,
+      "▁homemade": 39770,
+      "▁medicines": 39771,
+      "Pane": 39772,
+      "uras": 39773,
+      "可通过": 39774,
+      "群岛": 39775,
+      "▁Sold": 39776,
+      "博会": 39777,
+      "▁consultants": 39778,
+      "南昌市": 39779,
+      "这个城市": 39780,
+      "血脂": 39781,
+      "▁vault": 39782,
+      "系的": 39783,
+      "Metric": 39784,
+      "湾区": 39785,
+      "timezone": 39786,
+      "在使用现有的床铺": 39787,
+      "无语": 39788,
+      "此处": 39789,
+      "IDs": 39790,
+      "▁房间数量": 39791,
+      "▁Suit": 39792,
+      "nobject": 39793,
+      "Mass": 39794,
+      "Phot": 39795,
+      "calcul": 39796,
+      "flower": 39797,
+      "▁alliance": 39798,
+      "中国大陆": 39799,
+      "matter": 39800,
+      "Features": 39801,
+      "▁asymmetric": 39802,
+      "字段": 39803,
+      "employee": 39804,
+      "▁tertiary": 39805,
+      "agem": 39806,
+      "▁雷": 39807,
+      "给您": 39808,
+      "▁convince": 39809,
+      "TRUE": 39810,
+      "▁Hindu": 39811,
+      "▁mucosa": 39812,
+      "蚌埠": 39813,
+      "▁第四十": 39814,
+      "▁Stein": 39815,
+      "ño": 39816,
+      "▁超": 39817,
+      "▁ls": 39818,
+      "uang": 39819,
+      "managed": 39820,
+      "▁Arnold": 39821,
+      "能被": 39822,
+      "该院": 39823,
+      "▁elic": 39824,
+      "▁vegan": 39825,
+      "▁intermitt": 39826,
+      "otechnology": 39827,
+      "DAO": 39828,
+      "dead": 39829,
+      "rophys": 39830,
+      "▁gp": 39831,
+      "有帮助": 39832,
+      "alla": 39833,
+      "ipur": 39834,
+      "寻常": 39835,
+      "就此": 39836,
+      "愉悦": 39837,
+      "icester": 39838,
+      "▁Episode": 39839,
+      "DataFrame": 39840,
+      "对角线": 39841,
+      "给他们": 39842,
+      "▁dice": 39843,
+      "antine": 39844,
+      "phrine": 39845,
+      "▁Winds": 39846,
+      "friends": 39847,
+      "▁usefulness": 39848,
+      "糯米": 39849,
+      "间隙": 39850,
+      "~\\\\": 39851,
+      "申请执行人张": 39852,
+      "▁τ": 39853,
+      "▁创建": 39854,
+      "▁spo": 39855,
+      "▁QtGui": 39856,
+      "卵巢": 39857,
+      "重重": 39858,
+      "▁Dy": 39859,
+      "双方的": 39860,
+      "▁Shore": 39861,
+      "第十章": 39862,
+      "▁scream": 39863,
+      "发回": 39864,
+      "情况的": 39865,
+      "程序合法": 39866,
+      "▁Pars": 39867,
+      "▁brains": 39868,
+      "▁Hopefully": 39869,
+      "rer": 39870,
+      "▁junk": 39871,
+      "▁Matter": 39872,
+      "▁whereby": 39873,
+      "到底是": 39874,
+      "查询被执行人": 39875,
+      "nPerhaps": 39876,
+      "习性": 39877,
+      "渔业": 39878,
+      "▁sep": 39879,
+      "▁assisting": 39880,
+      "剧场": 39881,
+      "wid": 39882,
+      "▁FALSE": 39883,
+      "ensitivity": 39884,
+      "({\\'": 39885,
+      "▁Biden": 39886,
+      "FRINGEMENT": 39887,
+      "War": 39888,
+      "并支付": 39889,
+      "utton": 39890,
+      "类网络小说": 39891,
+      "▁pelvic": 39892,
+      "▁Catherine": 39893,
+      "政权": 39894,
+      "▁文件": 39895,
+      "ASSERT": 39896,
+      "nComput": 39897,
+      "深刻的": 39898,
+      "的平台": 39899,
+      "pain": 39900,
+      "▁replicate": 39901,
+      "从选项中选出你认为正确的答案": 39902,
+      "▁vom": 39903,
+      "不同意离婚": 39904,
+      "合理的解释": 39905,
+      "照射": 39906,
+      "Also": 39907,
+      "故本": 39908,
+      "ttes": 39909,
+      "▁mint": 39910,
+      "▁journalism": 39911,
+      "修行": 39912,
+      "千古": 39913,
+      "需要对": 39914,
+      "assay": 39915,
+      "▁couch": 39916,
+      "CORE": 39917,
+      "▁Initialize": 39918,
+      "宁静": 39919,
+      "工业园": 39920,
+      "质量的": 39921,
+      "▁lowering": 39922,
+      "板材": 39923,
+      "病症": 39924,
+      "贵阳": 39925,
+      "▁groom": 39926,
+      "▁Chapel": 39927,
+      "execution": 39928,
+      "明了": 39929,
+      "财务管理": 39930,
+      "▁Congo": 39931,
+      "deficient": 39932,
+      "▁Gym": 39933,
+      "现行": 39934,
+      "▁noble": 39935,
+      "▁licence": 39936,
+      "▁emphasize": 39937,
+      "万年": 39938,
+      "▁Dual": 39939,
+      "opausal": 39940,
+      "▁exhibitions": 39941,
+      "总共": 39942,
+      "回答是": 39943,
+      "▁touching": 39944,
+      "▁袁": 39945,
+      "twidth": 39946,
+      "ashington": 39947,
+      "该方法": 39948,
+      "▁AUTHORS": 39949,
+      "功耗": 39950,
+      "烟草": 39951,
+      "gpu": 39952,
+      "五个月": 39953,
+      "邮政储蓄银行股份有限公司": 39954,
+      "加固": 39955,
+      "▁Female": 39956,
+      "stereotype": 39957,
+      "▁energetic": 39958,
+      "换个": 39959,
+      "有效率": 39960,
+      "这才是": 39961,
+      "ussions": 39962,
+      "liga": 39963,
+      "▁abelian": 39964,
+      "endregion": 39965,
+      "第一名": 39966,
+      "级别的": 39967,
+      "runch": 39968,
+      "州市人民法院": 39969,
+      "物业服务合同纠纷一案": 39970,
+      "二战": 39971,
+      "却不": 39972,
+      "被誉为": 39973,
+      "Walk": 39974,
+      "nwas": 39975,
+      "▁Mold": 39976,
+      "▁disparities": 39977,
+      "夏日": 39978,
+      "扫码": 39979,
+      "▁Mercedes": 39980,
+      "也算": 39981,
+      "福田": 39982,
+      "unted": 39983,
+      "▁tomatoes": 39984,
+      "▁fame": 39985,
+      "在做": 39986,
+      "▁Ranch": 39987,
+      "喜悦": 39988,
+      "▁下列": 39989,
+      "发现自己": 39990,
+      "hift": 39991,
+      "▁ups": 39992,
+      "第一百零七条": 39993,
+      "▁Ecuador": 39994,
+      "EU": 39995,
+      "nte": 39996,
+      "咨询有限公司": 39997,
+      "准备好": 39998,
+      "全场": 39999,
+      "nda": 40000,
+      "并进行": 40001,
+      "nopen": 40002,
+      "DOWN": 40003,
+      "weets": 40004,
+      "MODULE": 40005,
+      "oides": 40006,
+      "▁suburb": 40007,
+      "▁accidentally": 40008,
+      "icio": 40009,
+      "LG": 40010,
+      "发表论文": 40011,
+      "梦中": 40012,
+      "alore": 40013,
+      "芒果": 40014,
+      "legiate": 40015,
+      "▁Drosophila": 40016,
+      "▁detectable": 40017,
+      "▁selections": 40018,
+      "居住地": 40019,
+      "nChrist": 40020,
+      "城的": 40021,
+      "米兰": 40022,
+      "▁pag": 40023,
+      "ttext": 40024,
+      "我一": 40025,
+      "nNumber": 40026,
+      "▁theorems": 40027,
+      "renergic": 40028,
+      "_%": 40029,
+      "mph": 40030,
+      "▁xmlns": 40031,
+      "在内": 40032,
+      "在区间": 40033,
+      "扬声器": 40034,
+      "▁yearly": 40035,
+      "▁crosses": 40036,
+      "株洲": 40037,
+      "深夜": 40038,
+      "▁reuse": 40039,
+      "▁mentions": 40040,
+      "忍受": 40041,
+      "电影作品": 40042,
+      "▁tightly": 40043,
+      "▁tracked": 40044,
+      "跨国": 40045,
+      "结构住房": 40046,
+      "▁Rhode": 40047,
+      "有多种选择": 40048,
+      "glass": 40049,
+      "▁butt": 40050,
+      "nMount": 40051,
+      "▁settlements": 40052,
+      "draft": 40053,
+      "ukary": 40054,
+      "DAY": 40055,
+      "▁ic": 40056,
+      "anton": 40057,
+      "olv": 40058,
+      "的重点": 40059,
+      "处方": 40060,
+      "长途": 40061,
+      "nNeed": 40062,
+      "▁Reynolds": 40063,
+      "▁Chef": 40064,
+      "万千": 40065,
+      "uno": 40066,
+      "August": 40067,
+      "▁’": 40068,
+      "▁useless": 40069,
+      "徘徊": 40070,
+      "的颜色": 40071,
+      "Unless": 40072,
+      "七十": 40073,
+      "▁apples": 40074,
+      "第一百四十四条": 40075,
+      "▁homepage": 40076,
+      "能为": 40077,
+      "nsic": 40078,
+      "▁Filip": 40079,
+      "▁aureus": 40080,
+      "▁purification": 40081,
+      "▁circumference": 40082,
+      "具有一定的": 40083,
+      "▁Mig": 40084,
+      "▁memor": 40085,
+      "▁Throughout": 40086,
+      "缺血": 40087,
+      "▁diagnose": 40088,
+      "▁passwords": 40089,
+      "▁consequently": 40090,
+      "▁clearing": 40091,
+      "工作效率": 40092,
+      "▁Include": 40093,
+      "▁cricketer": 40094,
+      "化肥": 40095,
+      "许多人": 40096,
+      "July": 40097,
+      "population": 40098,
+      "莫名其": 40099,
+      "▁transc": 40100,
+      "万余": 40101,
+      "就是个": 40102,
+      "法执字第": 40103,
+      "▁beetle": 40104,
+      "▁specifying": 40105,
+      "▁Thom": 40106,
+      "Listen": 40107,
+      "hong": 40108,
+      "▁Shared": 40109,
+      "Dashboard": 40110,
+      "▁shoulders": 40111,
+      "▁Industries": 40112,
+      "结点": 40113,
+      "和解题": 40114,
+      "医护": 40115,
+      "opsies": 40116,
+      "nProfessional": 40117,
+      "之星": 40118,
+      "店的": 40119,
+      "推荐过": 40120,
+      "labs": 40121,
+      "nfile": 40122,
+      "Segment": 40123,
+      "还有一些": 40124,
+      "Reply": 40125,
+      "跑了": 40126,
+      "轴上": 40127,
+      "服务于": 40128,
+      "▁hygiene": 40129,
+      "总投资": 40130,
+      "犯抢劫罪": 40131,
+      "▁sidew": 40132,
+      "omonas": 40133,
+      "▁Coron": 40134,
+      "bearing": 40135,
+      "时段": 40136,
+      "省内": 40137,
+      "▁glory": 40138,
+      "▁Compare": 40139,
+      "巨头": 40140,
+      "换了": 40141,
+      "zeg": 40142,
+      "关于适用": 40143,
+      "▁kinda": 40144,
+      "▁Palestinian": 40145,
+      "陆军": 40146,
+      "▁middleware": 40147,
+      "留意": 40148,
+      "▁travelers": 40149,
+      "ZW": 40150,
+      "cemia": 40151,
+      "▁electrophoresis": 40152,
+      "\"`": 40153,
+      "▁Awesome": 40154,
+      "▁peoples": 40155,
+      "▁Separ": 40156,
+      "长远": 40157,
+      "在不同": 40158,
+      "的一半": 40159,
+      "▁Sabha": 40160,
+      "世上": 40161,
+      "▁定义": 40162,
+      "医生的": 40163,
+      "▁termed": 40164,
+      "▁expired": 40165,
+      "望着": 40166,
+      "本法": 40167,
+      "nDistribution": 40168,
+      "随即": 40169,
+      "▁polls": 40170,
+      "▁rotational": 40171,
+      "AU": 40172,
+      "掌控": 40173,
+      "维持原判": 40174,
+      "ket": 40175,
+      "grand": 40176,
+      "而成的": 40177,
+      "▁recreation": 40178,
+      "▁Ghost": 40179,
+      "Preferences": 40180,
+      "七个月": 40181,
+      "催化剂": 40182,
+      "acked": 40183,
+      "ffield": 40184,
+      "HV": 40185,
+      "itat": 40186,
+      "poke": 40187,
+      "ATURE": 40188,
+      "良性": 40189,
+      "重心": 40190,
+      "▁Films": 40191,
+      "错的": 40192,
+      "▁flooding": 40193,
+      "▁resilience": 40194,
+      "▁Tennis": 40195,
+      "▁symmet": 40196,
+      "▁automobile": 40197,
+      "歌舞": 40198,
+      "Circle": 40199,
+      "▁remotely": 40200,
+      "▁distortion": 40201,
+      "刹车": 40202,
+      "火的": 40203,
+      "▁het": 40204,
+      "romes": 40205,
+      "背面": 40206,
+      "产业的": 40207,
+      "财政局": 40208,
+      "esta": 40209,
+      "ograp": 40210,
+      "营运": 40211,
+      "要不": 40212,
+      "▁Turb": 40213,
+      "▁Covid": 40214,
+      "GPT": 40215,
+      "uran": 40216,
+      "阿里巴巴": 40217,
+      "daily": 40218,
+      "tD": 40219,
+      "顿时": 40220,
+      "提升了": 40221,
+      "顶部": 40222,
+      "alent": 40223,
+      "号民事裁定": 40224,
+      "细分": 40225,
+      "Water": 40226,
+      "的通项公式": 40227,
+      "onacci": 40228,
+      "▁posed": 40229,
+      "▁trigon": 40230,
+      "ttar": 40231,
+      "focused": 40232,
+      "akespeare": 40233,
+      "▁Bulgaria": 40234,
+      "awi": 40235,
+      "nmore": 40236,
+      "仍是": 40237,
+      "指责": 40238,
+      "以来的": 40239,
+      "▁cocktail": 40240,
+      "nombre": 40241,
+      "匿名": 40242,
+      "全日制": 40243,
+      "下水": 40244,
+      "▁Sudan": 40245,
+      "▁luc": 40246,
+      "pending": 40247,
+      "ArgsConstructor": 40248,
+      "下图": 40249,
+      "upgrade": 40250,
+      "川县": 40251,
+      "frican": 40252,
+      "▁hyperbolic": 40253,
+      "唐山市": 40254,
+      "iane": 40255,
+      "▁Helen": 40256,
+      "▁Creates": 40257,
+      "nDocument": 40258,
+      "▁出场": 40259,
+      "borg": 40260,
+      "▁aus": 40261,
+      "▁Grande": 40262,
+      "▁bother": 40263,
+      "▁capacitor": 40264,
+      "Aw": 40265,
+      "xls": 40266,
+      "我们也": 40267,
+      "▁Chang": 40268,
+      "▁brake": 40269,
+      "ellation": 40270,
+      "▁Alternatively": 40271,
+      "来回": 40272,
+      "dry": 40273,
+      "▁Thr": 40274,
+      "lasses": 40275,
+      "▁lesbian": 40276,
+      "▁surprises": 40277,
+      "全校": 40278,
+      "对她": 40279,
+      "him": 40280,
+      "▁throat": 40281,
+      "▁travelled": 40282,
+      "weak": 40283,
+      "Border": 40284,
+      "▁dynamical": 40285,
+      "伴有": 40286,
+      "名誉": 40287,
+      "等多个": 40288,
+      "▁effector": 40289,
+      "推销": 40290,
+      "wang": 40291,
+      "▁蒋": 40292,
+      "活的": 40293,
+      "xref": 40294,
+      "▁stellar": 40295,
+      "我这": 40296,
+      "肯定会": 40297,
+      "▁Rosen": 40298,
+      "nabout": 40299,
+      "▁McDonald": 40300,
+      "tP": 40301,
+      "▁„": 40302,
+      "▁Gir": 40303,
+      "▁obligations": 40304,
+      "明智": 40305,
+      "aaaa": 40306,
+      "▁Deb": 40307,
+      "暂无财产可供执行": 40308,
+      "投保": 40309,
+      "追加": 40310,
+      "记者从": 40311,
+      "ocomplete": 40312,
+      "农家": 40313,
+      "出示": 40314,
+      "知的": 40315,
+      "保证了": 40316,
+      "没有了": 40317,
+      "▁Bash": 40318,
+      "▁ping": 40319,
+      "Er": 40320,
+      "楼层": 40321,
+      "风湿": 40322,
+      "没问题": 40323,
+      "酒店的房型有多种选择": 40324,
+      "仙人": 40325,
+      "光纤": 40326,
+      "▁flame": 40327,
+      "▁infinitely": 40328,
+      "▁tweeted": 40329,
+      "卡尔": 40330,
+      "摄影师": 40331,
+      "▁Saw": 40332,
+      "▁spor": 40333,
+      "技艺": 40334,
+      "计时": 40335,
+      "时间长": 40336,
+      "的数字": 40337,
+      "▁Bench": 40338,
+      "他和": 40339,
+      "我心": 40340,
+      "茫茫": 40341,
+      "▁政治": 40342,
+      "Replace": 40343,
+      "▁outlook": 40344,
+      "INFRINGEMENT": 40345,
+      "亚太": 40346,
+      "礼貌": 40347,
+      "▁Lak": 40348,
+      "▁citing": 40349,
+      "▁broadcasting": 40350,
+      "大声": 40351,
+      "角为": 40352,
+      "iches": 40353,
+      "高频": 40354,
+      "安全隐患": 40355,
+      "rawler": 40356,
+      "▁Values": 40357,
+      "▁misunder": 40358,
+      "tL": 40359,
+      "干的": 40360,
+      "nInstead": 40361,
+      "加剧": 40362,
+      "▁打开": 40363,
+      "事迹": 40364,
+      "建平": 40365,
+      "tconsole": 40366,
+      "拼命": 40367,
+      "aidu": 40368,
+      "▁Trend": 40369,
+      "▁embryo": 40370,
+      "独具": 40371,
+      "随身": 40372,
+      "▁PCI": 40373,
+      "▁Html": 40374,
+      "仙侠": 40375,
+      "▁galaxies": 40376,
+      "▁hydraulic": 40377,
+      "以非法": 40378,
+      "▁breach": 40379,
+      "aluable": 40380,
+      "▁Moving": 40381,
+      "▁Myst": 40382,
+      "▁midfielders": 40383,
+      "▁woods": 40384,
+      "therefore": 40385,
+      "土家族": 40386,
+      "▁reforms": 40387,
+      "▁Employee": 40388,
+      "hc": 40389,
+      "无缝": 40390,
+      "游览": 40391,
+      "盐水": 40392,
+      "moz": 40393,
+      "的决定": 40394,
+      "'^": 40395,
+      "引起了": 40396,
+      "▁Excellence": 40397,
+      "subscription": 40398,
+      "搬家": 40399,
+      "nElect": 40400,
+      "大盘": 40401,
+      "放弃了": 40402,
+      "▁hazards": 40403,
+      "mare": 40404,
+      "▁GMAT": 40405,
+      "▁mirrors": 40406,
+      "▁freezing": 40407,
+      "::{": 40408,
+      "▁discourse": 40409,
+      "addon": 40410,
+      "▁sellers": 40411,
+      "▁graphene": 40412,
+      "▁enthusiasm": 40413,
+      "ис": 40414,
+      "关卡": 40415,
+      "是国家": 40416,
+      "为什么不": 40417,
+      "▁sunlight": 40418,
+      "奥林": 40419,
+      "还真": 40420,
+      "Idx": 40421,
+      "▁ka": 40422,
+      "的心理": 40423,
+      "▁reconc": 40424,
+      "fav": 40425,
+      "的音乐": 40426,
+      "▁Wing": 40427,
+      "sta": 40428,
+      "Given": 40429,
+      "第二百五十七": 40430,
+      "▁routinely": 40431,
+      "Chrome": 40432,
+      "问道": 40433,
+      "Aim": 40434,
+      "•\\": 40435,
+      "较长": 40436,
+      "▁skiing": 40437,
+      "生活方式": 40438,
+      "nEval": 40439,
+      "endants": 40440,
+      "先将": 40441,
+      "aders": 40442,
+      "▁aper": 40443,
+      "有色": 40444,
+      "▁haul": 40445,
+      "▁achieves": 40446,
+      "怀念": 40447,
+      "阿根廷": 40448,
+      "▁advocates": 40449,
+      "▁mediating": 40450,
+      "▁mixtures": 40451,
+      "一站": 40452,
+      "\\'%": 40453,
+      "为目标": 40454,
+      "洛阳市": 40455,
+      "▁Vehicle": 40456,
+      "取得的": 40457,
+      "的网络": 40458,
+      "电子邮件": 40459,
+      "▁[^": 40460,
+      "说的是": 40461,
+      "growth": 40462,
+      "compress": 40463,
+      "▁Guidelines": 40464,
+      "弹簧": 40465,
+      "题型": 40466,
+      "▁macrophage": 40467,
+      "Eq": 40468,
+      "企业文化": 40469,
+      "TEGER": 40470,
+      "▁Leave": 40471,
+      "人情": 40472,
+      "你做出": 40473,
+      "oretical": 40474,
+      "▁mitigate": 40475,
+      "▁启动": 40476,
+      "relationship": 40477,
+      "▁Oakland": 40478,
+      "▁publishers": 40479,
+      "并购": 40480,
+      "uffy": 40481,
+      "▁Therm": 40482,
+      "▁refuse": 40483,
+      "programming": 40484,
+      "▁hematopoietic": 40485,
+      "见解": 40486,
+      "车位": 40487,
+      "ICT": 40488,
+      "▁nost": 40489,
+      "▁Charg": 40490,
+      "前方": 40491,
+      "▁capsule": 40492,
+      "▁cytochrome": 40493,
+      "特有的": 40494,
+      "运动的": 40495,
+      "▁protease": 40496,
+      "▁克": 40497,
+      "Choice": 40498,
+      "urry": 40499,
+      "▁Buffer": 40500,
+      "axios": 40501,
+      "配资": 40502,
+      "nSort": 40503,
+      "▁landmark": 40504,
+      "▁seminars": 40505,
+      "▁derivation": 40506,
+      "oter": 40507,
+      "予以准许": 40508,
+      "Suppress": 40509,
+      "火星": 40510,
+      "出具了": 40511,
+      "添加剂": 40512,
+      "的一条": 40513,
+      "Jo": 40514,
+      "情境": 40515,
+      ")//": 40516,
+      "Ts": 40517,
+      "的一家": 40518,
+      "Logic": 40519,
+      "▁sanctions": 40520,
+      "▁vocabulary": 40521,
+      "作词": 40522,
+      "螺丝": 40523,
+      "陷阱": 40524,
+      "▁Unique": 40525,
+      "ROS": 40526,
+      "nDi": 40527,
+      "▁Cred": 40528,
+      "▁Hook": 40529,
+      "午餐": 40530,
+      "如上": 40531,
+      "愈合": 40532,
+      "▁maturity": 40533,
+      "nWow": 40534,
+      "▁skull": 40535,
+      "▁supermark": 40536,
+      "三观": 40537,
+      "请从": 40538,
+      "有兴趣": 40539,
+      "胡椒粉": 40540,
+      "Words": 40541,
+      "pwd": 40542,
+      "▁mph": 40543,
+      "总支": 40544,
+      "▁tau": 40545,
+      "▁clicks": 40546,
+      "manif": 40547,
+      "versation": 40548,
+      "涂抹": 40549,
+      "邢台": 40550,
+      "▁Kick": 40551,
+      "▁crush": 40552,
+      "▁guessing": 40553,
+      "▁SUB": 40554,
+      "▁Syl": 40555,
+      "符合法定": 40556,
+      "nSection": 40557,
+      "▁flavour": 40558,
+      "▁breakthrough": 40559,
+      "▁cigarette": 40560,
+      "宣讲": 40561,
+      "▁Nak": 40562,
+      "edish": 40563,
+      "umann": 40564,
+      "亲情": 40565,
+      "唱片": 40566,
+      "▁Apart": 40567,
+      "▁affair": 40568,
+      "却依然": 40569,
+      "ptides": 40570,
+      "▁lethal": 40571,
+      "▁Continue": 40572,
+      "花序": 40573,
+      "即使是": 40574,
+      "kv": 40575,
+      "就不是": 40576,
+      "阶段的": 40577,
+      "▁bod": 40578,
+      "EventArgs": 40579,
+      "▁Argentine": 40580,
+      "高山": 40581,
+      "▁shrink": 40582,
+      "昆明市": 40583,
+      "nMain": 40584,
+      "▁Tampa": 40585,
+      "▁parasite": 40586,
+      "Buy": 40587,
+      "DEP": 40588,
+      "ipients": 40589,
+      "▁hottest": 40590,
+      "▁studios": 40591,
+      "JavaScript": 40592,
+      "有趣的": 40593,
+      "ibus": 40594,
+      "▁classify": 40595,
+      "swick": 40596,
+      "▁bif": 40597,
+      "▁Punjab": 40598,
+      "▁thrombosis": 40599,
+      "反击": 40600,
+      "有何": 40601,
+      "发言人": 40602,
+      "拿起": 40603,
+      "htra": 40604,
+      "▁assurance": 40605,
+      "{→}{": 40606,
+      "富士": 40607,
+      "紧凑": 40608,
+      "TNF": 40609,
+      "一道题": 40610,
+      "的解集": 40611,
+      "▁altogether": 40612,
+      "散发": 40613,
+      "无忧": 40614,
+      "▁ample": 40615,
+      "可使": 40616,
+      "行政行为": 40617,
+      "maximum": 40618,
+      "牵引": 40619,
+      "atti": 40620,
+      "分许": 40621,
+      "Psi": 40622,
+      "▁LH": 40623,
+      "nDel": 40624,
+      "出具欠条": 40625,
+      "▁fled": 40626,
+      "▁Objective": 40627,
+      "得的": 40628,
+      "不舒服": 40629,
+      "▁Koh": 40630,
+      "▁grains": 40631,
+      "▁converts": 40632,
+      "▁snacks": 40633,
+      "▁Milwaukee": 40634,
+      "[/": 40635,
+      "口气": 40636,
+      "▁compose": 40637,
+      "▁muscular": 40638,
+      "跑到": 40639,
+      "luc": 40640,
+      "全集": 40641,
+      "名片": 40642,
+      "有必要": 40643,
+      "duplex": 40644,
+      "nowled": 40645,
+      "Gateway": 40646,
+      "中国人民财产保险股份有限公司": 40647,
+      "好转": 40648,
+      "*}\\": 40649,
+      "公告费": 40650,
+      "njust": 40651,
+      "边境": 40652,
+      "equation": 40653,
+      "%\\\\": 40654,
+      "para": 40655,
+      "人大代表": 40656,
+      "经营管理": 40657,
+      "notice": 40658,
+      "▁cooler": 40659,
+      "山药": 40660,
+      "手续费": 40661,
+      "▁competent": 40662,
+      "され": 40663,
+      "隶属": 40664,
+      "签订了": 40665,
+      "otopy": 40666,
+      "▁princip": 40667,
+      "▁multimedia": 40668,
+      "INA": 40669,
+      "才能够": 40670,
+      "▁wors": 40671,
+      "players": 40672,
+      "▁Photography": 40673,
+      "▁发展": 40674,
+      "▁||\\": 40675,
+      "FILES": 40676,
+      "▁Leeds": 40677,
+      "▁}_{": 40678,
+      "因为他们": 40679,
+      "npart": 40680,
+      "叙事": 40681,
+      "nEp": 40682,
+      "▁Zn": 40683,
+      "又一次": 40684,
+      "nGoogle": 40685,
+      "▁Uber": 40686,
+      "▁promo": 40687,
+      "▁overlapping": 40688,
+      "▁军": 40689,
+      "商丘": 40690,
+      "▁记员": 40691,
+      "Tw": 40692,
+      "尊敬": 40693,
+      "就是为了": 40694,
+      "United": 40695,
+      "▁hydroph": 40696,
+      "的男人": 40697,
+      "▁Geo": 40698,
+      "▁pause": 40699,
+      "▁physi": 40700,
+      "tingham": 40701,
+      "▁Jeremy": 40702,
+      "畸形": 40703,
+      "锦标赛": 40704,
+      "▁CU": 40705,
+      "通知后": 40706,
+      "▁airports": 40707,
+      "▁conting": 40708,
+      "完后": 40709,
+      "强制法": 40710,
+      "ivate": 40711,
+      "▁Lodge": 40712,
+      "ifferences": 40713,
+      "▁副": 40714,
+      "反抗": 40715,
+      "二十四": 40716,
+      "▁Quarter": 40717,
+      "情商": 40718,
+      "GUI": 40719,
+      "▁proportions": 40720,
+      "∞,": 40721,
+      "urling": 40722,
+      "pressure": 40723,
+      "▁parametric": 40724,
+      "战队": 40725,
+      "缓缓": 40726,
+      "▁amph": 40727,
+      "intellij": 40728,
+      "到时候": 40729,
+      "▁environmentally": 40730,
+      "整机": 40731,
+      "梦见": 40732,
+      "▁){\\": 40733,
+      "▁appliance": 40734,
+      "却在": 40735,
+      "用作": 40736,
+      "Har": 40737,
+      "nText": 40738,
+      "Labels": 40739,
+      "▁airlines": 40740,
+      "osi": 40741,
+      "pent": 40742,
+      "▁Liu": 40743,
+      "earrange": 40744,
+      "▁analogue": 40745,
+      "西门": 40746,
+      "▁sor": 40747,
+      "那个时候": 40748,
+      "一头": 40749,
+      "oug": 40750,
+      "choline": 40751,
+      "有种": 40752,
+      "▁xy": 40753,
+      "▁Tyler": 40754,
+      "▁electrophys": 40755,
+      "▁电": 40756,
+      "表达式": 40757,
+      "都已经": 40758,
+      "▁Pont": 40759,
+      "▁dign": 40760,
+      "CASCADE": 40761,
+      "▁Against": 40762,
+      "债权人": 40763,
+      "Spe": 40764,
+      "orte": 40765,
+      "icone": 40766,
+      "pars": 40767,
+      "源头": 40768,
+      "▁accredited": 40769,
+      "征服": 40770,
+      "抗生素": 40771,
+      "▁Emma": 40772,
+      "捕捉": 40773,
+      "致命": 40774,
+      "'\\\\": 40775,
+      "商学院": 40776,
+      "净利润": 40777,
+      "到手": 40778,
+      "与应用": 40779,
+      "▁Bever": 40780,
+      "抛物线的": 40781,
+      "dating": 40782,
+      "▁echocard": 40783,
+      "灭火": 40784,
+      "▁promptly": 40785,
+      "Evaluation": 40786,
+      "aci": 40787,
+      "▁concludes": 40788,
+      "Ij": 40789,
+      "兴业": 40790,
+      "▁LDL": 40791,
+      "▁porch": 40792,
+      "▁glycol": 40793,
+      "挣钱": 40794,
+      "▁pean": 40795,
+      "▁slug": 40796,
+      "miral": 40797,
+      "闪光": 40798,
+      "▁HC": 40799,
+      "▁tracker": 40800,
+      "▁Lines": 40801,
+      "▁interferon": 40802,
+      "想要的": 40803,
+      "▁INF": 40804,
+      "▁tib": 40805,
+      "出院": 40806,
+      "▁lineage": 40807,
+      "nInf": 40808,
+      "▁deer": 40809,
+      "入驻": 40810,
+      "▁Cream": 40811,
+      "ographies": 40812,
+      "Translation": 40813,
+      "迄今": 40814,
+      "umed": 40815,
+      "))/((": 40816,
+      "笔画": 40817,
+      "▁Spl": 40818,
+      "▁Closing": 40819,
+      "▁seamless": 40820,
+      "▁Mn": 40821,
+      "其前身为": 40822,
+      "▁Redis": 40823,
+      "▁advancing": 40824,
+      "▁Relationship": 40825,
+      "无名": 40826,
+      "opia": 40827,
+      "文的": 40828,
+      "▁Proc": 40829,
+      "▁offshore": 40830,
+      "左手": 40831,
+      "纸张": 40832,
+      "私信": 40833,
+      "违背": 40834,
+      "versal": 40835,
+      "下雨": 40836,
+      "▁Whatever": 40837,
+      "诱发": 40838,
+      "增函数": 40839,
+      "乡市": 40840,
+      "打架": 40841,
+      "的系统": 40842,
+      "考上": 40843,
+      "屏蔽": 40844,
+      "期权": 40845,
+      "并按照": 40846,
+      "二〇一七年十二月": 40847,
+      "icals": 40848,
+      "伴侣": 40849,
+      "列举": 40850,
+      "育人": 40851,
+      "辩论": 40852,
+      "新中国的诞生": 40853,
+      "filling": 40854,
+      "▁uncomfortable": 40855,
+      "三峡": 40856,
+      "跳跃": 40857,
+      "▁IB": 40858,
+      "lasting": 40859,
+      "▁Jessica": 40860,
+      "▁desires": 40861,
+      "nSchedule": 40862,
+      "▁Cd": 40863,
+      "幸运的是": 40864,
+      "stanbul": 40865,
+      "▁predicts": 40866,
+      "▁compromised": 40867,
+      "一间": 40868,
+      "geon": 40869,
+      "insky": 40870,
+      "▁进球": 40871,
+      "ulose": 40872,
+      "好看的": 40873,
+      "▁PDE": 40874,
+      "maybe": 40875,
+      "▁balancing": 40876,
+      "并向": 40877,
+      "SEC": 40878,
+      "服务等": 40879,
+      "minster": 40880,
+      "万亩": 40881,
+      "路边": 40882,
+      "Pixel": 40883,
+      "正确答案和": 40884,
+      "交通大学": 40885,
+      "▁chim": 40886,
+      "Launch": 40887,
+      "invest": 40888,
+      "▁specs": 40889,
+      "▁traged": 40890,
+      "操纵": 40891,
+      "来做": 40892,
+      "Assets": 40893,
+      "Loaded": 40894,
+      "▁corrosion": 40895,
+      "城乡建设": 40896,
+      "执行长": 40897,
+      "即可求解": 40898,
+      "nthanks": 40899,
+      "▁clause": 40900,
+      "▁shortage": 40901,
+      "▁MW": 40902,
+      "住浙江省": 40903,
+      "▁unveiled": 40904,
+      "lift": 40905,
+      "ocyst": 40906,
+      "critical": 40907,
+      "▁vesicles": 40908,
+      "叠加": 40909,
+      "复苏": 40910,
+      "准确的": 40911,
+      "投资人": 40912,
+      "金牛": 40913,
+      "AAC": 40914,
+      "先生的": 40915,
+      "集装箱": 40916,
+      "midt": 40917,
+      "Sprite": 40918,
+      "▁Emperor": 40919,
+      "▁hemorrhage": 40920,
+      "zig": 40921,
+      "慢慢的": 40922,
+      "▁FILE": 40923,
+      "铁道": 40924,
+      "应予支持": 40925,
+      "chin": 40926,
+      "积极参与": 40927,
+      "工业出版社": 40928,
+      "▁attenuated": 40929,
+      "nFilmography": 40930,
+      "新能源汽车": 40931,
+      "rocket": 40932,
+      "▁strengthening": 40933,
+      "▁Terminal": 40934,
+      "恰当": 40935,
+      "选育": 40936,
+      "bery": 40937,
+      "guided": 40938,
+      "缓冲": 40939,
+      "一双": 40940,
+      "egg": 40941,
+      "nPhys": 40942,
+      "coords": 40943,
+      "▁Chester": 40944,
+      "亮的": 40945,
+      "们在": 40946,
+      "第三项": 40947,
+      "合作关系": 40948,
+      "▁Audience": 40949,
+      "骨骼": 40950,
+      "▁\")": 40951,
+      "▁planets": 40952,
+      "Fit": 40953,
+      "完全可以": 40954,
+      "iframe": 40955,
+      "制订": 40956,
+      "糟糕": 40957,
+      "斯的": 40958,
+      "樱桃": 40959,
+      "▁convergent": 40960,
+      "分层": 40961,
+      "余年": 40962,
+      "▁Kre": 40963,
+      "arius": 40964,
+      "altern": 40965,
+      "投注": 40966,
+      "音效": 40967,
+      "▁Mt": 40968,
+      "卫生院": 40969,
+      "toLowerCase": 40970,
+      "食欲": 40971,
+      "▁Almost": 40972,
+      "sass": 40973,
+      "▁IRS": 40974,
+      "▁Bound": 40975,
+      "processed": 40976,
+      "DNS": 40977,
+      "Hard": 40978,
+      "rocal": 40979,
+      "起重": 40980,
+      "chewan": 40981,
+      "LaTeX": 40982,
+      "Optim": 40983,
+      "▁ital": 40984,
+      "▁revel": 40985,
+      "athione": 40986,
+      "不便": 40987,
+      "也让": 40988,
+      "ekyll": 40989,
+      "capacity": 40990,
+      "▁Engineers": 40991,
+      "bond": 40992,
+      "▁Reed": 40993,
+      "ViewHolder": 40994,
+      "和解题的思路": 40995,
+      "spiracy": 40996,
+      "▁cerebell": 40997,
+      "印章": 40998,
+      "发电机": 40999,
+      "我只是": 41000,
+      "}\\\\,": 41001,
+      "琐事发生": 41002,
+      "重要的作用": 41003,
+      "你了": 41004,
+      "打出": 41005,
+      "经验的": 41006,
+      "维亚": 41007,
+      "tron": 41008,
+      "▁dar": 41009,
+      "▁fut": 41010,
+      "水肿": 41011,
+      "Sci": 41012,
+      "Less": 41013,
+      "intro": 41014,
+      "第一百零八条": 41015,
+      "较强": 41016,
+      "幸福的": 41017,
+      "▁percept": 41018,
+      "Vue": 41019,
+      "Fatal": 41020,
+      "最爱": 41021,
+      "积极的": 41022,
+      "titles": 41023,
+      "女人的": 41024,
+      "▁EEG": 41025,
+      "告知申请执行人": 41026,
+      "全天": 41027,
+      "▁covariance": 41028,
+      "困惑": 41029,
+      "aji": 41030,
+      "身份号码": 41031,
+      "actly": 41032,
+      "де": 41033,
+      "权纠纷一案": 41034,
+      "本次执行程序终结": 41035,
+      "伴随着新中国的诞生": 41036,
+      "▁Gaz": 41037,
+      "请给出你的": 41038,
+      "▁veloc": 41039,
+      "必不可": 41040,
+      "更名为": 41041,
+      "ByName": 41042,
+      "golang": 41043,
+      "▁magnesium": 41044,
+      "▁jurisdiction": 41045,
+      "检察机关": 41046,
+      "▁inhibiting": 41047,
+      "▁Pinterest": 41048,
+      "▁contacting": 41049,
+      "KL": 41050,
+      "jj": 41051,
+      "flight": 41052,
+      "▁photographers": 41053,
+      "导致了": 41054,
+      "aceous": 41055,
+      "第一百五十四条第一款第十一": 41056,
+      "化县": 41057,
+      "搬运": 41058,
+      "美术馆": 41059,
+      "某与被告": 41060,
+      "▁Letters": 41061,
+      "▁magnificent": 41062,
+      "充值": 41063,
+      "奇函数": 41064,
+      "▁Hughes": 41065,
+      "▁Brussels": 41066,
+      "该款": 41067,
+      "avin": 41068,
+      "▁uncle": 41069,
+      "nAL": 41070,
+      "不充分": 41071,
+      ">\",\"": 41072,
+      "▁lacks": 41073,
+      "定律": 41074,
+      "格外": 41075,
+      "的费用": 41076,
+      "呼和浩特": 41077,
+      "▁Yale": 41078,
+      "步入": 41079,
+      "信息查询": 41080,
+      "teenth": 41081,
+      "▁Notable": 41082,
+      "▁inability": 41083,
+      "再次提出": 41084,
+      "▁requesting": 41085,
+      "▁moll": 41086,
+      "耐用": 41087,
+      "▁jeans": 41088,
+      "反驳": 41089,
+      "IFT": 41090,
+      "idem": 41091,
+      "联保": 41092,
+      "▁Compute": 41093,
+      "▁静": 41094,
+      "▁mum": 41095,
+      "▁服务": 41096,
+      "▁Pete": 41097,
+      "▁metropolitan": 41098,
+      "ITION": 41099,
+      "▁Tiger": 41100,
+      "targets": 41101,
+      "▁Sevent": 41102,
+      "主打": 41103,
+      "BIN": 41104,
+      "▁functionally": 41105,
+      "fra": 41106,
+      "获荣誉": 41107,
+      "▁Dict": 41108,
+      "▁Sacr": 41109,
+      "▁Soci": 41110,
+      "▁Positive": 41111,
+      "工贸": 41112,
+      "▁],": 41113,
+      "完成的": 41114,
+      "和使用": 41115,
+      "oche": 41116,
+      "▁cAMP": 41117,
+      "十二条之规定": 41118,
+      "▁Prec": 41119,
+      "▁penalties": 41120,
+      "经营活动": 41121,
+      "▁tenth": 41122,
+      "▁fungal": 41123,
+      "▁undergone": 41124,
+      "Mag": 41125,
+      "▁获取": 41126,
+      "奠定了": 41127,
+      "▁hepar": 41128,
+      "肝炎": 41129,
+      "zhi": 41130,
+      "▁acidic": 41131,
+      "iliar": 41132,
+      "Comparison": 41133,
+      "inheritdoc": 41134,
+      "交际": 41135,
+      "▁isomorphism": 41136,
+      "▁Kom": 41137,
+      "▁Sebast": 41138,
+      "来电": 41139,
+      "▁Shir": 41140,
+      "▁microM": 41141,
+      "于本判决生效后十日内": 41142,
+      "vn": 41143,
+      "万达": 41144,
+      "胸部": 41145,
+      "publisher": 41146,
+      "']['": 41147,
+      "▁tilt": 41148,
+      "Views": 41149,
+      "▁Pine": 41150,
+      "tenant": 41151,
+      "▁Bring": 41152,
+      "▁solic": 41153,
+      "▁Fourth": 41154,
+      "▁Publications": 41155,
+      "▁Gand": 41156,
+      "▁recycled": 41157,
+      "上天": 41158,
+      "美白": 41159,
+      "▁设施与服务": 41160,
+      "开本": 41161,
+      "▁hil": 41162,
+      "▁femoral": 41163,
+      "多层": 41164,
+      "不明确": 41165,
+      "▁excuse": 41166,
+      "composer": 41167,
+      "Functions": 41168,
+      "▁polygon": 41169,
+      "▁Azerbaijan": 41170,
+      "平台上": 41171,
+      "二〇一六年十一月": 41172,
+      "和解题过程": 41173,
+      "原数": 41174,
+      "实证": 41175,
+      "boarding": 41176,
+      "主要从事": 41177,
+      "compact": 41178,
+      "▁Ing": 41179,
+      "▁São": 41180,
+      "▁disestablished": 41181,
+      "自卑": 41182,
+      "让您": 41183,
+      "不死": 41184,
+      "::$": 41185,
+      "教育厅": 41186,
+      "▁Tap": 41187,
+      "▁Faith": 41188,
+      "担保有限公司": 41189,
+      "▁lawmakers": 41190,
+      "买个": 41191,
+      "以至于": 41192,
+      "Mbps": 41193,
+      "岁时": 41194,
+      "燃油": 41195,
+      "pie": 41196,
+      "▁Mull": 41197,
+      "▁noon": 41198,
+      "▁stabilization": 41199,
+      "▁Lamb": 41200,
+      "美联": 41201,
+      "nsp": 41202,
+      "ABCD": 41203,
+      "▁decorating": 41204,
+      "nAsk": 41205,
+      "verting": 41206,
+      "▁Princeton": 41207,
+      "▁Palmer": 41208,
+      "你做出选择的依据": 41209,
+      "▁Markdown": 41210,
+      "▁Stockholm": 41211,
+      "你做出选择的依据和分析": 41212,
+      "稳步": 41213,
+      "两被告": 41214,
+      "Visitor": 41215,
+      "Categories": 41216,
+      "NM": 41217,
+      "严禁": 41218,
+      "gcd": 41219,
+      "LENGTH": 41220,
+      "无尽": 41221,
+      "▁CX": 41222,
+      "此外请给出": 41223,
+      "▁nicht": 41224,
+      "▁Stuart": 41225,
+      "▁surveyed": 41226,
+      "很长": 41227,
+      "肺癌": 41228,
+      "详见": 41229,
+      "一对一": 41230,
+      "▁iterative": 41231,
+      "▁neutr": 41232,
+      "风的": 41233,
+      "▁\":": 41234,
+      "▁spike": 41235,
+      "▁solub": 41236,
+      "广东省广州市": 41237,
+      "第一百一十九条": 41238,
+      "nDiscography": 41239,
+      "帐户": 41240,
+      "天文": 41241,
+      "Poly": 41242,
+      "每个客房": 41243,
+      "▁ERROR": 41244,
+      "▁Govern": 41245,
+      "宏伟": 41246,
+      "▁diminished": 41247,
+      "QA": 41248,
+      "单车": 41249,
+      "AIDS": 41250,
+      "pered": 41251,
+      "尔多斯": 41252,
+      "tracks": 41253,
+      "貌似": 41254,
+      "anchor": 41255,
+      "(\",": 41256,
+      "就不要": 41257,
+      "nop": 41258,
+      "ibling": 41259,
+      "▁inducing": 41260,
+      "加州": 41261,
+      "农民收入": 41262,
+      "forest": 41263,
+      "ylabel": 41264,
+      "▁VALUES": 41265,
+      "%\"": 41266,
+      "▁武": 41267,
+      "('#": 41268,
+      "grpc": 41269,
+      "过程的": 41270,
+      "申请再审": 41271,
+      "中国科学院成立": 41272,
+      "fordshire": 41273,
+      "availability": 41274,
+      "▁CSF": 41275,
+      "并给出相应的": 41276,
+      "封装": 41277,
+      "我从": 41278,
+      "顺德": 41279,
+      "▁ballot": 41280,
+      "▁Eg": 41281,
+      "etheus": 41282,
+      "▁damaging": 41283,
+      "▁quarters": 41284,
+      "浴缸": 41285,
+      "陈列": 41286,
+      "ACS": 41287,
+      "超过了": 41288,
+      "▁drafted": 41289,
+      "▁printable": 41290,
+      "优异": 41291,
+      "转会": 41292,
+      "▁facilitating": 41293,
+      "毒性": 41294,
+      "说不定": 41295,
+      "▁stent": 41296,
+      "前两": 41297,
+      "四级": 41298,
+      "nmod": 41299,
+      "四方": 41300,
+      "沉浸": 41301,
+      "nSE": 41302,
+      "漂亮的": 41303,
+      "▁nephro": 41304,
+      "组装": 41305,
+      "▁playoffs": 41306,
+      "NV": 41307,
+      "便携": 41308,
+      "滑雪": 41309,
+      "给的": 41310,
+      "tec": 41311,
+      "▁Traditional": 41312,
+      "locked": 41313,
+      "我希望": 41314,
+      "身边的": 41315,
+      "WB": 41316,
+      "发现的": 41317,
+      "每个客房都配有": 41318,
+      "completed": 41319,
+      "astery": 41320,
+      "▁graphical": 41321,
+      "瀑布": 41322,
+      "是自己": 41323,
+      "▁conduction": 41324,
+      "锐角": 41325,
+      "openhagen": 41326,
+      "▁Administrative": 41327,
+      "并用": 41328,
+      "▁va": 41329,
+      "ndis": 41330,
+      "其他可供执行财产": 41331,
+      "违纪": 41332,
+      "hao": 41333,
+      "Dark": 41334,
+      "资本主义": 41335,
+      "vac": 41336,
+      "rives": 41337,
+      "色调": 41338,
+      "▁evenly": 41339,
+      "这篇日志": 41340,
+      "{`": 41341,
+      "海尔": 41342,
+      "高雄": 41343,
+      "长方体": 41344,
+      "长方体的": 41345,
+      "Colors": 41346,
+      "▁greenhouse": 41347,
+      "ARC": 41348,
+      "▁systolic": 41349,
+      "冷漠": 41350,
+      "数的和": 41351,
+      "▁Evolution": 41352,
+      "▁privately": 41353,
+      "▁questionnaires": 41354,
+      "清晨": 41355,
+      "在日本": 41356,
+      "件的": 41357,
+      "由来": 41358,
+      "Issue": 41359,
+      "▁(/": 41360,
+      "可以向本院申请复议一次": 41361,
+      "▁territories": 41362,
+      "AMD": 41363,
+      "华中": 41364,
+      "算了": 41365,
+      "鱼类": 41366,
+      "▁Doug": 41367,
+      "患病": 41368,
+      "\"],\"": 41369,
+      "Behavior": 41370,
+      "交汇": 41371,
+      "只见": 41372,
+      "▁Sang": 41373,
+      "▁juvenile": 41374,
+      "全自动": 41375,
+      "主要作品": 41376,
+      "▁benefici": 41377,
+      "THON": 41378,
+      "▁Yoga": 41379,
+      "Schedule": 41380,
+      "▁Amazing": 41381,
+      "水上": 41382,
+      "▁Wag": 41383,
+      "▁thereafter": 41384,
+      "▁shaping": 41385,
+      "oglobulin": 41386,
+      "放假": 41387,
+      "现货": 41388,
+      "畅通": 41389,
+      "比喻": 41390,
+      "相传": 41391,
+      "()))": 41392,
+      "▁remission": 41393,
+      "面临的": 41394,
+      "pson": 41395,
+      "供热": 41396,
+      "制裁": 41397,
+      "Previous": 41398,
+      "versible": 41399,
+      "手感": 41400,
+      "▁stro": 41401,
+      "▁Bundle": 41402,
+      "▁Possible": 41403,
+      "我公司": 41404,
+      "oplasm": 41405,
+      "▁Hindi": 41406,
+      "社会保障": 41407,
+      "人民出版社": 41408,
+      "▁leagues": 41409,
+      "upus": 41410,
+      "经济社会发展": 41411,
+      "▁Insert": 41412,
+      "万元的": 41413,
+      "econ": 41414,
+      "lite": 41415,
+      "▁beads": 41416,
+      "▁botan": 41417,
+      "▁mines": 41418,
+      "gallery": 41419,
+      "▁decorations": 41420,
+      "|$": 41421,
+      "国立": 41422,
+      "美景": 41423,
+      "营收": 41424,
+      "吸引力": 41425,
+      "ismiss": 41426,
+      "▁Reuters": 41427,
+      "Hol": 41428,
+      "[])": 41429,
+      "timately": 41430,
+      "▁collaborations": 41431,
+      "由上海市人民政府": 41432,
+      "ancellor": 41433,
+      "西游": 41434,
+      "他们还推荐了": 41435,
+      "这篇日志的人": 41436,
+      "推荐过这篇日志的人": 41437,
+      "宿迁": 41438,
+      "▁VPN": 41439,
+      "素质教育": 41440,
+      "arming": 41441,
+      "▁confirms": 41442,
+      "鄂尔多斯": 41443,
+      "▁polyp": 41444,
+      "▁Proceedings": 41445,
+      "▁sandwich": 41446,
+      "原生": 41447,
+      "气势": 41448,
+      "高消费": 41449,
+      "▁Titan": 41450,
+      "▁Lebanon": 41451,
+      "岁以下的儿童在使用现有的床铺": 41452,
+      "▁있": 41453,
+      "教你": 41454,
+      "比值": 41455,
+      "▁expenditure": 41456,
+      "格林": 41457,
+      "生姜": 41458,
+      "▁输入样例": 41459,
+      "如意": 41460,
+      "atta": 41461,
+      "LEVEL": 41462,
+      "ophilic": 41463,
+      "▁Foster": 41464,
+      "Employee": 41465,
+      "脖子": 41466,
+      "追索": 41467,
+      "▁nond": 41468,
+      "attachment": 41469,
+      "第二百六十二条第二款之规定": 41470,
+      "自豪": 41471,
+      "nTheir": 41472,
+      "▁CLAIM": 41473,
+      "▁Dynam": 41474,
+      "▁refrigerator": 41475,
+      "家伙": 41476,
+      "▁glu": 41477,
+      "▁Communist": 41478,
+      "话语": 41479,
+      "监狱提出": 41480,
+      "sales": 41481,
+      "依法向被执行人": 41482,
+      "gd": 41483,
+      "×\\": 41484,
+      "寓意": 41485,
+      "▁gymn": 41486,
+      "湘潭": 41487,
+      "guy": 41488,
+      "第六十四": 41489,
+      "模型的": 41490,
+      "stre": 41491,
+      "▁Advert": 41492,
+      "▁Montgomery": 41493,
+      "▁Tanz": 41494,
+      "▁denoted": 41495,
+      "gra": 41496,
+      "acry": 41497,
+      "ochrom": 41498,
+      "是根据": 41499,
+      "}}}{": 41500,
+      "▁Mead": 41501,
+      "umbled": 41502,
+      "第一位": 41503,
+      "ducers": 41504,
+      "年经教育部正式批准": 41505,
+      "您好": 41506,
+      "村的": 41507,
+      "申诉": 41508,
+      "▁todd": 41509,
+      "▁拉丁学名": 41510,
+      "一般在": 41511,
+      "▁discomfort": 41512,
+      "粘贴": 41513,
+      "chars": 41514,
+      "▁editions": 41515,
+      "名学生": 41516,
+      "▁cheapest": 41517,
+      "华丽": 41518,
+      "喝水": 41519,
+      "认定的": 41520,
+      "nJul": 41521,
+      "▁textures": 41522,
+      "▁dendritic": 41523,
+      "合作的": 41524,
+      "adal": 41525,
+      "METHO": 41526,
+      "▁uterine": 41527,
+      "▁endoscopic": 41528,
+      "▁inactivation": 41529,
+      "对立": 41530,
+      "▁WS": 41531,
+      "recv": 41532,
+      "▁sang": 41533,
+      "ocratic": 41534,
+      "AutoField": 41535,
+      "祝贺": 41536,
+      "疗程": 41537,
+      "odia": 41538,
+      "▁Daw": 41539,
+      "Require": 41540,
+      "angement": 41541,
+      "############": 41542,
+      "剧烈": 41543,
+      "分布式": 41544,
+      "▁XXX": 41545,
+      "▁gluten": 41546,
+      "▁differentiable": 41547,
+      "周五": 41548,
+      "日凌晨": 41549,
+      "})\\\\)": 41550,
+      "主办的": 41551,
+      "▁Respons": 41552,
+      "▁hormonal": 41553,
+      "▁keyboards": 41554,
+      "完工": 41555,
+      "组和": 41556,
+      "uncan": 41557,
+      "▁Numerical": 41558,
+      "▁cutaneous": 41559,
+      "因为它": 41560,
+      "▁wealthy": 41561,
+      "awks": 41562,
+      "antes": 41563,
+      "▁输出样例": 41564,
+      "weighted": 41565,
+      "▁midnight": 41566,
+      "▁等": 41567,
+      "第十二": 41568,
+      "Depth": 41569,
+      "▁unincorporated": 41570,
+      "▁terrorism": 41571,
+      "oS": 41572,
+      "金沙": 41573,
+      "dip": 41574,
+      "国内的": 41575,
+      "▁背景设定": 41576,
+      "radiol": 41577,
+      "▁counterpart": 41578,
+      "▁psychosocial": 41579,
+      "加入了": 41580,
+      "▁Gor": 41581,
+      "瓶颈": 41582,
+      "贵阳市": 41583,
+      "aley": 41584,
+      "▁Bras": 41585,
+      "▁heroes": 41586,
+      "▁constitutes": 41587,
+      "BOOL": 41588,
+      "Liter": 41589,
+      "▁quantification": 41590,
+      "合力": 41591,
+      "更强": 41592,
+      "正版": 41593,
+      "AGES": 41594,
+      "义务教育": 41595,
+      "▁scaled": 41596,
+      "▁wondered": 41597,
+      "▁cytotoxicity": 41598,
+      "年任": 41599,
+      "居委会": 41600,
+      "▁Medicaid": 41601,
+      "改成": 41602,
+      "^^^^": 41603,
+      "被执行人未按": 41604,
+      "为他": 41605,
+      "entric": 41606,
+      "▁Acid": 41607,
+      "▁ester": 41608,
+      "▁mimic": 41609,
+      "prus": 41610,
+      "▁democratic": 41611,
+      "的一致": 41612,
+      "kok": 41613,
+      "▁ate": 41614,
+      "给出解题": 41615,
+      "ridges": 41616,
+      "▁Scotia": 41617,
+      "▁Navigation": 41618,
+      "意向": 41619,
+      "一般都是": 41620,
+      "pressive": 41621,
+      "人力资源和社会保障": 41622,
+      "▁conve": 41623,
+      "▁Tourism": 41624,
+      "RequestMapping": 41625,
+      "▁planting": 41626,
+      "的因素": 41627,
+      "湖北省武汉市": 41628,
+      "▁scandal": 41629,
+      "潮湿": 41630,
+      "meg": 41631,
+      "曹操": 41632,
+      "Works": 41633,
+      "об": 41634,
+      "▁della": 41635,
+      "sale": 41636,
+      "五星": 41637,
+      "avar": 41638,
+      "归还原告": 41639,
+      "▁\"\")\\": 41640,
+      "油漆": 41641,
+      "医务": 41642,
+      "搜集": 41643,
+      "时装": 41644,
+      "nExamples": 41645,
+      "洗手": 41646,
+      ">:\\\\(": 41647,
+      "affeine": 41648,
+      "与中国科学院共同举办": 41649,
+      "落户": 41650,
+      "做不到": 41651,
+      "demand": 41652,
+      "nstart": 41653,
+      "桂花": 41654,
+      "Kernel": 41655,
+      "甲方": 41656,
+      "▁Pb": 41657,
+      "员工的": 41658,
+      "]):\\": 41659,
+      "这样才能": 41660,
+      "scheduler": 41661,
+      "小子": 41662,
+      "自制": 41663,
+      "nLog": 41664,
+      "云中书城": 41665,
+      "AlterField": 41666,
+      "几位": 41667,
+      "电厂": 41668,
+      "一起去": 41669,
+      "Keep": 41670,
+      "▁exogenous": 41671,
+      "数据分析": 41672,
+      "▁cough": 41673,
+      "uscular": 41674,
+      "▁minimizing": 41675,
+      "▁Independence": 41676,
+      "estock": 41677,
+      "▁whisk": 41678,
+      "backends": 41679,
+      "来得": 41680,
+      "Claim": 41681,
+      "rings": 41682,
+      "▁Gent": 41683,
+      "▁Swan": 41684,
+      "▁Ready": 41685,
+      "▁improper": 41686,
+      "▁selenium": 41687,
+      "第二百三": 41688,
+      "nings": 41689,
+      "轻伤": 41690,
+      "Tor": 41691,
+      "iPad": 41692,
+      "forget": 41693,
+      "▁ladder": 41694,
+      "▁sticks": 41695,
+      "十位": 41696,
+      "迭代": 41697,
+      "▁tones": 41698,
+      "▁blogging": 41699,
+      "四舍五入": 41700,
+      "化石": 41701,
+      "泰山": 41702,
+      "质量和": 41703,
+      "Must": 41704,
+      "stitial": 41705,
+      "均未": 41706,
+      "Rub": 41707,
+      "比特币": 41708,
+      "▁Dow": 41709,
+      "passed": 41710,
+      "▁Budget": 41711,
+      "▁concomit": 41712,
+      "esse": 41713,
+      "▁trump": 41714,
+      "社会责任": 41715,
+      "先算": 41716,
+      "递归": 41717,
+      "中含有": 41718,
+      "中兴": 41719,
+      "▁Dress": 41720,
+      "▁mediators": 41721,
+      "在她": 41722,
+      "众生": 41723,
+      "cyte": 41724,
+      "进行了调查": 41725,
+      "▁innocent": 41726,
+      "制动": 41727,
+      "旅程": 41728,
+      "Flex": 41729,
+      "occup": 41730,
+      "nSeason": 41731,
+      "▁embark": 41732,
+      "▁adhesive": 41733,
+      "▁supplying": 41734,
+      "眼球": 41735,
+      "安全感": 41736,
+      "▁complain": 41737,
+      "以期": 41738,
+      "每个人都": 41739,
+      "男方": 41740,
+      "污水处理": 41741,
+      "要求离婚": 41742,
+      "看不": 41743,
+      "现在是": 41744,
+      "预订取消": 41745,
+      "▁overexpression": 41746,
+      "▁hobby": 41747,
+      "▁swift": 41748,
+      "▁Instant": 41749,
+      "预付政策": 41750,
+      ":\\\\\\\\": 41751,
+      "nImage": 41752,
+      "▁moles": 41753,
+      "的可": 41754,
+      "acey": 41755,
+      "▁к": 41756,
+      "▁refere": 41757,
+      "▁Magnetic": 41758,
+      "届时": 41759,
+      "ODY": 41760,
+      "grass": 41761,
+      "indust": 41762,
+      "implicit": 41763,
+      "▁specifies": 41764,
+      "▁γ": 41765,
+      "开朗": 41766,
+      "这是我": 41767,
+      "Assertions": 41768,
+      "西洋": 41769,
+      "Integration": 41770,
+      "▁adequately": 41771,
+      "个大": 41772,
+      "税率": 41773,
+      "inston": 41774,
+      "▁interfere": 41775,
+      "▁下载": 41776,
+      "lodash": 41777,
+      "另外一个": 41778,
+      "易懂": 41779,
+      "▁Evaluate": 41780,
+      "▁dehydrogenase": 41781,
+      "开盘": 41782,
+      "请参阅": 41783,
+      "alia": 41784,
+      "▁Inventory": 41785,
+      "▁dissertation": 41786,
+      "三门": 41787,
+      "两面": 41788,
+      "严厉": 41789,
+      "全球化": 41790,
+      "▁Bach": 41791,
+      "ycin": 41792,
+      "Quick": 41793,
+      "笔记本电脑": 41794,
+      "Expect": 41795,
+      "▁Haven": 41796,
+      "▁illnesses": 41797,
+      "▁asymptomatic": 41798,
+      "头条": 41799,
+      "看成": 41800,
+      "orde": 41801,
+      "东县": 41802,
+      "▁ideally": 41803,
+      "他不": 41804,
+      "▁exceeds": 41805,
+      "延安": 41806,
+      "Inner": 41807,
+      "reports": 41808,
+      "项目建设": 41809,
+      "▁Units": 41810,
+      "readline": 41811,
+      "▁promotions": 41812,
+      "以后的": 41813,
+      "nRet": 41814,
+      "石家庄市": 41815,
+      "▁showc": 41816,
+      "▁polymers": 41817,
+      "听过": 41818,
+      "▁ie": 41819,
+      "▁mammary": 41820,
+      "zd": 41821,
+      "室的": 41822,
+      "捞出": 41823,
+      "▁revolutionary": 41824,
+      "各种各样的": 41825,
+      "ARG": 41826,
+      "\\'},\\": 41827,
+      "photos": 41828,
+      "LIBR": 41829,
+      "ombie": 41830,
+      "Tracker": 41831,
+      "Dispatcher": 41832,
+      "}）": 41833,
+      ";</": 41834,
+      "▁evid": 41835,
+      "▁Happ": 41836,
+      "▁anemia": 41837,
+      "同胞": 41838,
+      "就应该": 41839,
+      "nChild": 41840,
+      "▁firmly": 41841,
+      "未履行生效法律文书确定的义务": 41842,
+      "uawei": 41843,
+      "▁lengthy": 41844,
+      "riot": 41845,
+      "ordering": 41846,
+      "卡通": 41847,
+      "只是一": 41848,
+      "竞争对手": 41849,
+      "▁socks": 41850,
+      "候选人": 41851,
+      "这是因为": 41852,
+      "▁游戏大小": 41853,
+      "Behaviour": 41854,
+      "▁emit": 41855,
+      "儿童和加床": 41856,
+      "镇长": 41857,
+      "trip": 41858,
+      "酒店位于": 41859,
+      "▁Docs": 41860,
+      "不稳定": 41861,
+      "三包服务": 41862,
+      "▁inviting": 41863,
+      "▁RM": 41864,
+      "▁alteration": 41865,
+      "天后": 41866,
+      "▁gum": 41867,
+      "点的坐标": 41868,
+      "▁dash": 41869,
+      "digits": 41870,
+      "Variant": 41871,
+      "有毒": 41872,
+      "salt": 41873,
+      "▁ESPN": 41874,
+      "Alt": 41875,
+      "却没有": 41876,
+      "uchi": 41877,
+      "ilion": 41878,
+      "▁lifes": 41879,
+      "▁polys": 41880,
+      "部门的": 41881,
+      "nCode": 41882,
+      "facing": 41883,
+      "▁athletics": 41884,
+      "▁viscosity": 41885,
+      "nFamily": 41886,
+      "连通": 41887,
+      "▁exported": 41888,
+      "Bucket": 41889,
+      "榆林": 41890,
+      "▁Inject": 41891,
+      "运河": 41892,
+      "的热情": 41893,
+      "\\\\\">": 41894,
+      "ecal": 41895,
+      "gf": 41896,
+      "山河": 41897,
+      "电解": 41898,
+      "▁Ye": 41899,
+      "的安装": 41900,
+      "类似的": 41901,
+      "Lang": 41902,
+      "VELOPM": 41903,
+      "▁phenotypic": 41904,
+      "管教": 41905,
+      "▁sie": 41906,
+      "▁范": 41907,
+      "▁lav": 41908,
+      "▁Bears": 41909,
+      "提起行政诉讼": 41910,
+      "October": 41911,
+      "▁px": 41912,
+      "污染物": 41913,
+      "▁slim": 41914,
+      "▁historians": 41915,
+      "小事": 41916,
+      "rug": 41917,
+      "▁TD": 41918,
+      "elin": 41919,
+      "▁Pir": 41920,
+      "▁Hein": 41921,
+      "比较高": 41922,
+      "端点": 41923,
+      "正方体": 41924,
+      "被申请执行人": 41925,
+      "求值": 41926,
+      "阶层": 41927,
+      "iente": 41928,
+      "onne": 41929,
+      "变态": 41930,
+      "帝王": 41931,
+      "真菌": 41932,
+      "ftp": 41933,
+      "nstd": 41934,
+      "▁Firm": 41935,
+      "OUTPUT": 41936,
+      "▁litter": 41937,
+      "▁passport": 41938,
+      "北京时间": 41939,
+      "finally": 41940,
+      "水产": 41941,
+      "原裁定": 41942,
+      "icidal": 41943,
+      "▁guards": 41944,
+      "loo": 41945,
+      "inventory": 41946,
+      "▁Monitoring": 41947,
+      "预备": 41948,
+      "▁Rams": 41949,
+      "以下有期徒刑": 41950,
+      "▁Territory": 41951,
+      "年了": 41952,
+      "aques": 41953,
+      "条形": 41954,
+      "为广大": 41955,
+      "表面的": 41956,
+      "造就": 41957,
+      "▁erupt": 41958,
+      "无处": 41959,
+      "洁净": 41960,
+      "tet": 41961,
+      "▁beside": 41962,
+      "▁Approved": 41963,
+      "RB": 41964,
+      "告知书": 41965,
+      "▁Wyoming": 41966,
+      "straction": 41967,
+      "地利": 41968,
+      "证号": 41969,
+      "mapsto": 41970,
+      "randint": 41971,
+      "头疼": 41972,
+      "CTV": 41973,
+      "InstanceOf": 41974,
+      "守法": 41975,
+      "寄生": 41976,
+      "nMet": 41977,
+      "计分考核": 41978,
+      "▁pickup": 41979,
+      "不一": 41980,
+      "出轨": 41981,
+      "▁invoice": 41982,
+      "nBur": 41983,
+      "▁prere": 41984,
+      "ancellation": 41985,
+      "上手": 41986,
+      "▁Cand": 41987,
+      "▁copying": 41988,
+      "▁Aviation": 41989,
+      "▁Pressure": 41990,
+      "▁cher": 41991,
+      "为民": 41992,
+      "pac": 41993,
+      "▁eu": 41994,
+      "开始执行": 41995,
+      "▁Cork": 41996,
+      "tracker": 41997,
+      "▁irreducible": 41998,
+      "效力于": 41999,
+      "▁Rib": 42000,
+      "valueOf": 42001,
+      "datasets": 42002,
+      "▁irrigation": 42003,
+      "背叛": 42004,
+      "das": 42005,
+      "▁EB": 42006,
+      "旗下的": 42007,
+      "降临": 42008,
+      "该公司员工": 42009,
+      "上半": 42010,
+      "的女儿": 42011,
+      "好玩": 42012,
+      "退款": 42013,
+      "泉州市": 42014,
+      "进入到": 42015,
+      "布尔": 42016,
+      "▁Pul": 42017,
+      "▁Bryan": 42018,
+      "ortheastern": 42019,
+      "hev": 42020,
+      "他们是": 42021,
+      "▁功能": 42022,
+      "单调递减": 42023,
+      "▁Shen": 42024,
+      "更重要的是": 42025,
+      "▁Located": 42026,
+      "▁imprison": 42027,
+      "万家": 42028,
+      "▁strap": 42029,
+      "又一": 42030,
+      "发烧": 42031,
+      "然后把": 42032,
+      "专业技术": 42033,
+      "景色": 42034,
+      "课时": 42035,
+      "▁contrace": 42036,
+      "▁abolished": 42037,
+      "三项": 42038,
+      "丰收": 42039,
+      "富汗": 42040,
+      "iably": 42041,
+      "ARCHAR": 42042,
+      "\"[": 42043,
+      "翅膀": 42044,
+      "pkgs": 42045,
+      "中华人民共和国行政强制法": 42046,
+      "背光": 42047,
+      "上海科技大学": 42048,
+      "▁Adventure": 42049,
+      "共鸣": 42050,
+      "▁Plaza": 42051,
+      "▁Amateur": 42052,
+      "▁probable": 42053,
+      "严格的": 42054,
+      "lion": 42055,
+      "▁Orthodox": 42056,
+      "▁Measurement": 42057,
+      "▁celebrities": 42058,
+      "▁Assume": 42059,
+      "▁stretching": 42060,
+      "▁packets": 42061,
+      "Bin": 42062,
+      "▁Attempt": 42063,
+      "▁Notification": 42064,
+      "尽力": 42065,
+      "着他": 42066,
+      "▁Loop": 42067,
+      "▁Overview": 42068,
+      "▁communicating": 42069,
+      "▁Cav": 42070,
+      "▁polype": 42071,
+      "oso": 42072,
+      "新西兰": 42073,
+      "recomm": 42074,
+      "Balance": 42075,
+      "▁以": 42076,
+      "彼得": 42077,
+      "大多数人": 42078,
+      "▁Critical": 42079,
+      "复古": 42080,
+      "征求": 42081,
+      "了下来": 42082,
+      "需要在": 42083,
+      "▁Hod": 42084,
+      "▁concise": 42085,
+      "等财产": 42086,
+      "▁tray": 42087,
+      "Without": 42088,
+      "superv": 42089,
+      "▁culmin": 42090,
+      "から": 42091,
+      "统计局": 42092,
+      "anca": 42093,
+      "美术学院": 42094,
+      "▁invoke": 42095,
+      "▁drilling": 42096,
+      "就算是": 42097,
+      "PART": 42098,
+      "副秘书长": 42099,
+      "装有": 42100,
+      "▁baked": 42101,
+      "▁laying": 42102,
+      "▁terrace": 42103,
+      "▁recommends": 42104,
+      "ansible": 42105,
+      "▁roofing": 42106,
+      "也曾": 42107,
+      "恒大": 42108,
+      "这首歌": 42109,
+      "▁Sox": 42110,
+      "phrase": 42111,
+      "▁arthro": 42112,
+      "▁spleen": 42113,
+      "▁acetate": 42114,
+      "▁advancement": 42115,
+      "Learning": 42116,
+      "▁vitamins": 42117,
+      "▁therapist": 42118,
+      "哪家": 42119,
+      "商量": 42120,
+      "arettes": 42121,
+      "ceptors": 42122,
+      "Ha": 42123,
+      "红枣": 42124,
+      "authorization": 42125,
+      "▁POS": 42126,
+      "▁neckl": 42127,
+      "▁eigenvalue": 42128,
+      "南开": 42129,
+      "出色的": 42130,
+      "▁indef": 42131,
+      "则有": 42132,
+      "nAD": 42133,
+      "▁misc": 42134,
+      "▁Zimbabwe": 42135,
+      "▁blockade": 42136,
+      "励志": 42137,
+      "核准": 42138,
+      "invoice": 42139,
+      "▁platelets": 42140,
+      "杨某某": 42141,
+      "ancement": 42142,
+      "▁陆": 42143,
+      "导出": 42144,
+      "饰品": 42145,
+      "▁Prague": 42146,
+      "▁immunodef": 42147,
+      "▁divergence": 42148,
+      "ør": 42149,
+      "车子": 42150,
+      "ROW": 42151,
+      "▁VC": 42152,
+      "一次方程": 42153,
+      "三角函数": 42154,
+      "▁Missing": 42155,
+      "▁epidemiology": 42156,
+      "毛巾": 42157,
+      "▁lipoprotein": 42158,
+      "▁Designs": 42159,
+      "▁Operator": 42160,
+      "▁treasure": 42161,
+      "ographically": 42162,
+      "编曲": 42163,
+      "MES": 42164,
+      "兼任": 42165,
+      "awt": 42166,
+      "申请书": 42167,
+      "进行比较": 42168,
+      "▁Amer": 42169,
+      "并不会": 42170,
+      "▁Bomb": 42171,
+      "▁crisp": 42172,
+      "阳市中级人民法院": 42173,
+      "早点": 42174,
+      "\"}}": 42175,
+      "▁annot": 42176,
+      "▁closes": 42177,
+      "fax": 42178,
+      "▁inconsistent": 42179,
+      "履行能力": 42180,
+      "第五十三条": 42181,
+      "astolic": 42182,
+      "能耗": 42183,
+      "年龄的": 42184,
+      "了呢": 42185,
+      "宽松": 42186,
+      "dic": 42187,
+      "solve": 42188,
+      "▁discovering": 42189,
+      "山上": 42190,
+      "鼓楼": 42191,
+      "detect": 42192,
+      "二〇一六年十二月": 42193,
+      "在那里": 42194,
+      "Ship": 42195,
+      "telif": 42196,
+      "运会": 42197,
+      "nee": 42198,
+      "▁GPL": 42199,
+      "Twitter": 42200,
+      "EW": 42201,
+      "首位": 42202,
+      "基本信息": 42203,
+      "Foo": 42204,
+      "▁nurt": 42205,
+      "▁urge": 42206,
+      "▁worries": 42207,
+      "快餐": 42208,
+      "谈到": 42209,
+      "Logo": 42210,
+      "issan": 42211,
+      "▁robustness": 42212,
+      "imus": 42213,
+      "▁tribe": 42214,
+      "▁nominee": 42215,
+      "属于自己的": 42216,
+      "▁lending": 42217,
+      "nOper": 42218,
+      "符合法律规定的": 42219,
+      "▁ListNode": 42220,
+      "转入": 42221,
+      "udi": 42222,
+      "INPUT": 42223,
+      "▁Ruth": 42224,
+      "海滩": 42225,
+      "pmod": 42226,
+      "把自己的": 42227,
+      "中东": 42228,
+      "PLICATION": 42229,
+      "必将": 42230,
+      "站的": 42231,
+      "illar": 42232,
+      "▁Graphics": 42233,
+      "置于": 42234,
+      "议会": 42235,
+      "▁Jenkins": 42236,
+      "plicative": 42237,
+      "▁steroids": 42238,
+      "TOR": 42239,
+      "jay": 42240,
+      "绝句": 42241,
+      "中国科学院研究生院": 42242,
+      "环比": 42243,
+      "▁Julie": 42244,
+      "▁chord": 42245,
+      "▁phosphatase": 42246,
+      "▁Gmina": 42247,
+      "Forward": 42248,
+      "凭什么": 42249,
+      "ENTER": 42250,
+      "tfunction": 42251,
+      "REAM": 42252,
+      "alion": 42253,
+      "otent": 42254,
+      "他自己": 42255,
+      "oxin": 42256,
+      "▁raid": 42257,
+      "归属": 42258,
+      "钱包": 42259,
+      "assis": 42260,
+      "表白": 42261,
+      "代表作": 42262,
+      "omycin": 42263,
+      "dictionary": 42264,
+      "本题的关键": 42265,
+      "▁bride": 42266,
+      "▁glands": 42267,
+      "最高人民法院关于审理": 42268,
+      "我发现": 42269,
+      "▁Productions": 42270,
+      "xz": 42271,
+      "▁OnInit": 42272,
+      "▁resolutions": 42273,
+      "再现": 42274,
+      "▁你的": 42275,
+      "▁veins": 42276,
+      "▁kinases": 42277,
+      "▁hooks": 42278,
+      "▁allies": 42279,
+      "金银": 42280,
+      "生产力": 42281,
+      "▁rug": 42282,
+      "恰恰": 42283,
+      "有能力": 42284,
+      "▁Winn": 42285,
+      "▁Gly": 42286,
+      "seeing": 42287,
+      ")_{": 42288,
+      "shi": 42289,
+      "idin": 42290,
+      "▁bending": 42291,
+      "▁alveolar": 42292,
+      "进士": 42293,
+      "})^": 42294,
+      "enry": 42295,
+      "▁Adjust": 42296,
+      "jour": 42297,
+      "▁brightness": 42298,
+      "▁questioned": 42299,
+      "|>": 42300,
+      "标志着": 42301,
+      "\":\"\",\"": 42302,
+      "▁retire": 42303,
+      "显现": 42304,
+      "sembling": 42305,
+      "▁_\\": 42306,
+      "使之": 42307,
+      "烟花": 42308,
+      "了一份": 42309,
+      "胰岛素": 42310,
+      "▁Jeffrey": 42311,
+      "▁isomorphic": 42312,
+      "▁七": 42313,
+      "徐某": 42314,
+      "红尘": 42315,
+      "绽放": 42316,
+      "▁pant": 42317,
+      "▁radicals": 42318,
+      "▁Southwest": 42319,
+      "Serialization": 42320,
+      "这本": 42321,
+      "Vel": 42322,
+      "▁calibr": 42323,
+      "problems": 42324,
+      "第三个": 42325,
+      "▁Leo": 42326,
+      "精通": 42327,
+      "ovic": 42328,
+      "▁Dipl": 42329,
+      "ACCESS": 42330,
+      "▁dealers": 42331,
+      "哺乳": 42332,
+      "阿富汗": 42333,
+      "minor": 42334,
+      "▁intestine": 42335,
+      "▁formations": 42336,
+      "技法": 42337,
+      "白发": 42338,
+      "是怎样的": 42339,
+      "Apply": 42340,
+      "usable": 42341,
+      "学长": 42342,
+      "—\\": 42343,
+      "红柿": 42344,
+      "的可能": 42345,
+      "运动队": 42346,
+      "tout": 42347,
+      "▁Twin": 42348,
+      "▁Certain": 42349,
+      "MIC": 42350,
+      "Bill": 42351,
+      "▁nap": 42352,
+      "▁lasts": 42353,
+      "▁webpage": 42354,
+      "▁partnered": 42355,
+      "克隆": 42356,
+      "拌匀": 42357,
+      "油腻": 42358,
+      "GPU": 42359,
+      "nlogger": 42360,
+      "创新创业": 42361,
+      "ighthouse": 42362,
+      "哭了": 42363,
+      "▁lub": 42364,
+      "▁yii": 42365,
+      "抓获经过": 42366,
+      "cause": 42367,
+      "▁lightly": 42368,
+      "仅为": 42369,
+      "▁perfor": 42370,
+      "▁cess": 42371,
+      "▁Canon": 42372,
+      "▁endors": 42373,
+      "▁Investig": 42374,
+      "管理者": 42375,
+      "经营的": 42376,
+      "两份": 42377,
+      "父母的": 42378,
+      "▁endocrine": 42379,
+      "年夜": 42380,
+      "搬迁": 42381,
+      "Fun": 42382,
+      "▁Contr": 42383,
+      "国学": 42384,
+      "▁nails": 42385,
+      "▁initialization": 42386,
+      "$:": 42387,
+      "GNU": 42388,
+      "nProperty": 42389,
+      "▁DEALINGS": 42390,
+      "尾巴": 42391,
+      "宣告缓刑": 42392,
+      "▁horiz": 42393,
+      "问答": 42394,
+      "School": 42395,
+      "Singleton": 42396,
+      "不佳": 42397,
+      "壁纸": 42398,
+      "外来": 42399,
+      "acetyl": 42400,
+      "Currency": 42401,
+      "及照片": 42402,
+      "▁wit": 42403,
+      "opensource": 42404,
+      "应注意": 42405,
+      "Alias": 42406,
+      "▁但是": 42407,
+      "▁Manit": 42408,
+      "▁thrive": 42409,
+      "▁pathophysiology": 42410,
+      "星辰": 42411,
+      "触犯": 42412,
+      "华盛顿": 42413,
+      "VELOPMENT": 42414,
+      "合并同类项": 42415,
+      "ninput": 42416,
+      "油烟": 42417,
+      "不限": 42418,
+      "微型": 42419,
+      "量大": 42420,
+      "在面对": 42421,
+      "▁推荐菜": 42422,
+      "wat": 42423,
+      "短视频": 42424,
+      "董事会": 42425,
+      "Hyper": 42426,
+      "并提出": 42427,
+      "avia": 42428,
+      "▁oun": 42429,
+      "▁eukary": 42430,
+      "▁agonists": 42431,
+      "▁standings": 42432,
+      "殴打": 42433,
+      "流动性": 42434,
+      "民事权利和": 42435,
+      "▁Compl": 42436,
+      "焦作": 42437,
+      "老师们": 42438,
+      "射手": 42439,
+      "即可解答": 42440,
+      "班长": 42441,
+      "科教": 42442,
+      "酸奶": 42443,
+      "▁melting": 42444,
+      "▁Romanian": 42445,
+      "ellites": 42446,
+      "▁guiding": 42447,
+      "维尔": 42448,
+      "悬挂": 42449,
+      "similar": 42450,
+      "▁trajectories": 42451,
+      "▁unsuccessful": 42452,
+      "桌子": 42453,
+      "访谈": 42454,
+      "因原告": 42455,
+      "开始的": 42456,
+      "身份信息": 42457,
+      "▁stairs": 42458,
+      "在这一": 42459,
+      "Repos": 42460,
+      "▁compliant": 42461,
+      "▁imperative": 42462,
+      "INSERT": 42463,
+      "位移": 42464,
+      "嘴巴": 42465,
+      "于此": 42466,
+      "矿物": 42467,
+      "太原市": 42468,
+      "无视": 42469,
+      "男人的": 42470,
+      "▁nat": 42471,
+      "技术咨询": 42472,
+      "四海": 42473,
+      "家务": 42474,
+      "报销": 42475,
+      "▁Wide": 42476,
+      "▁Shopping": 42477,
+      ")])\\": 42478,
+      "arashtra": 42479,
+      "nSO": 42480,
+      "干涉": 42481,
+      "▁scrut": 42482,
+      "可以考虑": 42483,
+      "后记": 42484,
+      "魔术": 42485,
+      "avian": 42486,
+      "中海": 42487,
+      "黄昏": 42488,
+      "如果想": 42489,
+      "▁酒店星级": 42490,
+      "▁amend": 42491,
+      "ListItem": 42492,
+      "做到了": 42493,
+      "的好处": 42494,
+      "erent": 42495,
+      "tR": 42496,
+      "用车": 42497,
+      "refn": 42498,
+      "运算法则": 42499,
+      "During": 42500,
+      "ottle": 42501,
+      "颠覆": 42502,
+      "charts": 42503,
+      "▁Concert": 42504,
+      "编织": 42505,
+      "数量的": 42506,
+      "ToProps": 42507,
+      "机能": 42508,
+      "这么说": 42509,
+      "等一系列": 42510,
+      "动能": 42511,
+      "责令其": 42512,
+      "不得超过": 42513,
+      "▁Bundes": 42514,
+      "jac": 42515,
+      "kub": 42516,
+      "此题主要": 42517,
+      "▁serotonin": 42518,
+      "▁про": 42519,
+      "▁trio": 42520,
+      "▁Petersburg": 42521,
+      "的程度": 42522,
+      "▁Fem": 42523,
+      "▁nickel": 42524,
+      "approved": 42525,
+      "中南": 42526,
+      "▁Rails": 42527,
+      "remarks": 42528,
+      "▁stderr": 42529,
+      "游玩": 42530,
+      "overview": 42531,
+      "nhe": 42532,
+      "▁shorts": 42533,
+      "产出": 42534,
+      "afa": 42535,
+      "nlike": 42536,
+      "▁hurd": 42537,
+      "▁丹": 42538,
+      "擅自": 42539,
+      "▁bipolar": 42540,
+      "回落": 42541,
+      "报复": 42542,
+      "(\\'<": 42543,
+      "常务委员会": 42544,
+      "▁observing": 42545,
+      "▁Distributed": 42546,
+      "rw": 42547,
+      "▁属性": 42548,
+      "的一本": 42549,
+      "▁Mechanical": 42550,
+      "iento": 42551,
+      "Archive": 42552,
+      "▁somatic": 42553,
+      "ublicense": 42554,
+      "▁conformation": 42555,
+      "▁崔": 42556,
+      "ropic": 42557,
+      "additional": 42558,
+      "ORDER": 42559,
+      "▁predis": 42560,
+      "支配": 42561,
+      "FIELD": 42562,
+      "▁mais": 42563,
+      "▁neurom": 42564,
+      "▁Potential": 42565,
+      "▁porous": 42566,
+      "气血": 42567,
+      "fee": 42568,
+      "/\\'\\": 42569,
+      "▁Canal": 42570,
+      "nCompet": 42571,
+      "▁upward": 42572,
+      "▁Brunswick": 42573,
+      "机动车交通事故责任纠纷一案": 42574,
+      "普陀": 42575,
+      "▁shirts": 42576,
+      "反转": 42577,
+      "省钱": 42578,
+      "}}_{": 42579,
+      "威海": 42580,
+      "▁flee": 42581,
+      "▁lifted": 42582,
+      "▁lowered": 42583,
+      "partition": 42584,
+      "萎缩": 42585,
+      "要多": 42586,
+      "托管": 42587,
+      "pypi": 42588,
+      "告诉记者": 42589,
+      "其前身为中国科学院研究生院": 42590,
+      "毛重": 42591,
+      "Even": 42592,
+      "书面申请": 42593,
+      "▁ventric": 42594,
+      "▁suspicious": 42595,
+      "Sy": 42596,
+      "的请求": 42597,
+      "▁scaffold": 42598,
+      "King": 42599,
+      "MenuItem": 42600,
+      "▁División": 42601,
+      "▁securities": 42602,
+      "▁cylindrical": 42603,
+      "];//": 42604,
+      "▁Items": 42605,
+      "该书": 42606,
+      "SCs": 42607,
+      "▁een": 42608,
+      "饮酒": 42609,
+      "▁quantified": 42610,
+      "▁Interactive": 42611,
+      "从头": 42612,
+      "团长": 42613,
+      "ulae": 42614,
+      "▁bee": 42615,
+      "Ing": 42616,
+      "▁MED": 42617,
+      "▁bored": 42618,
+      "十二条第一款": 42619,
+      "▁billing": 42620,
+      "▁UT": 42621,
+      "不足以": 42622,
+      "seven": 42623,
+      "▁Stir": 42624,
+      "seo": 42625,
+      "▁prayers": 42626,
+      "数百": 42627,
+      "owl": 42628,
+      "astore": 42629,
+      "洪水": 42630,
+      "百花": 42631,
+      "▁robotic": 42632,
+      "YI": 42633,
+      "侦探": 42634,
+      "▁blah": 42635,
+      "▁literal": 42636,
+      "两只": 42637,
+      "▁Db": 42638,
+      "phis": 42639,
+      "aines": 42640,
+      "三亚": 42641,
+      "▁`\\": 42642,
+      "▁chromosomes": 42643,
+      "人次": 42644,
+      "助攻": 42645,
+      "▁Pom": 42646,
+      "游击": 42647,
+      "▁projective": 42648,
+      "▁Topics": 42649,
+      "住所地上海市": 42650,
+      "从容": 42651,
+      "odor": 42652,
+      "▁posters": 42653,
+      "))$": 42654,
+      "Four": 42655,
+      "十堰": 42656,
+      "生抽": 42657,
+      "在家里": 42658,
+      "afood": 42659,
+      "▁tits": 42660,
+      "▁Tickets": 42661,
+      "▁OA": 42662,
+      "nSch": 42663,
+      "▁indexed": 42664,
+      "既往": 42665,
+      "流入": 42666,
+      "学生会": 42667,
+      "▁anomalies": 42668,
+      "▁interim": 42669,
+      "面临着": 42670,
+      "▁Coloring": 42671,
+      "▁weaken": 42672,
+      "受人": 42673,
+      "最喜欢": 42674,
+      "▁flowering": 42675,
+      "▁constituencies": 42676,
+      "stimulated": 42677,
+      "haul": 42678,
+      "以非法占有": 42679,
+      "▁Pakistani": 42680,
+      "▁Temperature": 42681,
+      "▁facilitated": 42682,
+      "▁Euclidean": 42683,
+      "帮我做": 42684,
+      "▁Sul": 42685,
+      "▁poles": 42686,
+      "▁patron": 42687,
+      "第二百四十四条": 42688,
+      "▁RGB": 42689,
+      "▁habitats": 42690,
+      "无烟": 42691,
+      "cob": 42692,
+      "学到": 42693,
+      "▁中文学名": 42694,
+      "▁Plasma": 42695,
+      "制剂": 42696,
+      "日本人": 42697,
+      "istro": 42698,
+      "ierarchy": 42699,
+      "常熟": 42700,
+      "多样性": 42701,
+      "adin": 42702,
+      "▁Sensor": 42703,
+      "▁micrograms": 42704,
+      "拉伸": 42705,
+      "灰尘": 42706,
+      "实用性": 42707,
+      "相互作用": 42708,
+      "▁sublicense": 42709,
+      "是把": 42710,
+      "▁Venice": 42711,
+      "▁Interview": 42712,
+      "sof": 42713,
+      "高性能": 42714,
+      "nbody": 42715,
+      "▁goodness": 42716,
+      "医院的": 42717,
+      "VERY": 42718,
+      "▁sponsorship": 42719,
+      "▁Oz": 42720,
+      "ercises": 42721,
+      "▁NONINFRINGEMENT": 42722,
+      "▁Lynn": 42723,
+      "▁Articles": 42724,
+      "▁encounters": 42725,
+      "急诊": 42726,
+      "鸿蒙": 42727,
+      "uga": 42728,
+      "▁Crew": 42729,
+      "Observable": 42730,
+      "河道": 42731,
+      "代表的": 42732,
+      "virus": 42733,
+      "▁Assert": 42734,
+      "吃到": 42735,
+      "ydia": 42736,
+      "TITLE": 42737,
+      "▁attachments": 42738,
+      "修饰": 42739,
+      "抗战": 42740,
+      "kat": 42741,
+      "athi": 42742,
+      "蔓延": 42743,
+      "黎明": 42744,
+      "Chunk": 42745,
+      "▁GOOD": 42746,
+      "冻结被执行人": 42747,
+      "▁kotlin": 42748,
+      "进的": 42749,
+      "▁Authentication": 42750,
+      "▁花": 42751,
+      "▁Aid": 42752,
+      "杂质": 42753,
+      "在本案": 42754,
+      "▁apprent": 42755,
+      "检疫": 42756,
+      "非常高兴": 42757,
+      "):\\\\(": 42758,
+      "▁affine": 42759,
+      "nRichard": 42760,
+      "重审": 42761,
+      "]:\\\\(": 42762,
+      "nvector": 42763,
+      "▁Integrated": 42764,
+      "▁scared": 42765,
+      "▁invented": 42766,
+      "▁transmitter": 42767,
+      "▁CG": 42768,
+      "solid": 42769,
+      "可以自收到": 42770,
+      "报告财产令": 42771,
+      "或有": 42772,
+      "管家": 42773,
+      "nPrior": 42774,
+      "▁Muhammad": 42775,
+      "▁fileName": 42776,
+      "finder": 42777,
+      "说服": 42778,
+      "▁ONLY": 42779,
+      "▁javascript": 42780,
+      "▁carbohydrate": 42781,
+      "宝鸡": 42782,
+      "条约": 42783,
+      "▁regimens": 42784,
+      "▁beverage": 42785,
+      "▁香": 42786,
+      "▁chond": 42787,
+      "tooltip": 42788,
+      "▁Lieutenant": 42789,
+      "丝绸": 42790,
+      "主干": 42791,
+      "会所": 42792,
+      "其所": 42793,
+      "提供在线": 42794,
+      "nSpeaking": 42795,
+      "▁uncertainties": 42796,
+      "决字": 42797,
+      "影院": 42798,
+      "umbs": 42799,
+      "Storm": 42800,
+      "▁Chev": 42801,
+      "定义域": 42802,
+      "由题意得": 42803,
+      "OnClick": 42804,
+      "▁ideals": 42805,
+      "▁quicker": 42806,
+      "ERO": 42807,
+      "控制的": 42808,
+      "chard": 42809,
+      "ersen": 42810,
+      "mongo": 42811,
+      "▁rationale": 42812,
+      "后又": 42813,
+      "PES": 42814,
+      "Emitter": 42815,
+      "▁生长": 42816,
+      "]](/": 42817,
+      "ento": 42818,
+      "weed": 42819,
+      "名叫": 42820,
+      "奥迪": 42821,
+      "昌县": 42822,
+      "内角和": 42823,
+      "▁angiogenesis": 42824,
+      "二分": 42825,
+      "sheets": 42826,
+      "▁interven": 42827,
+      "levard": 42828,
+      "了你": 42829,
+      "nRef": 42830,
+      "▁Bert": 42831,
+      "▁crowds": 42832,
+      "领导下": 42833,
+      "▁anaest": 42834,
+      "▁ventil": 42835,
+      "做为": 42836,
+      "以非法占有为目的": 42837,
+      "nCompanies": 42838,
+      "▁姚": 42839,
+      "裤子": 42840,
+      "nMass": 42841,
+      "nsome": 42842,
+      "▁splic": 42843,
+      "こと": 42844,
+      "说一下": 42845,
+      "平方差": 42846,
+      "等功能": 42847,
+      "代理检察员": 42848,
+      "▁bou": 42849,
+      "行业协会": 42850,
+      "phalt": 42851,
+      "▁Ralph": 42852,
+      "collections": 42853,
+      "系统和": 42854,
+      "多方": 42855,
+      "弱势": 42856,
+      "▁这个": 42857,
+      "▁Asc": 42858,
+      "EVENT": 42859,
+      "▁incentive": 42860,
+      "▁inversion": 42861,
+      "日因": 42862,
+      "魔龙": 42863,
+      "bw": 42864,
+      "课外": 42865,
+      "yna": 42866,
+      "▁[^](/": 42867,
+      "nThomas": 42868,
+      "REC": 42869,
+      "bright": 42870,
+      "▁disagree": 42871,
+      ">﹣": 42872,
+      "▁现在": 42873,
+      "促进了": 42874,
+      "面积公式": 42875,
+      "never": 42876,
+      "▁genotypes": 42877,
+      "参谋": 42878,
+      "头皮": 42879,
+      "$(\"": 42880,
+      "ographed": 42881,
+      "▁overweight": 42882,
+      "Bitmap": 42883,
+      "▁alleles": 42884,
+      "分解因式": 42885,
+      "▁Electronics": 42886,
+      "▁dps": 42887,
+      "民营企业": 42888,
+      "nPlay": 42889,
+      "▁Wiki": 42890,
+      "▁Cards": 42891,
+      "▁overlay": 42892,
+      "▁activating": 42893,
+      "▁admissions": 42894,
+      "十个": 42895,
+      "页数": 42896,
+      "▁fax": 42897,
+      "在刑罚执行期间": 42898,
+      "相声": 42899,
+      "nDO": 42900,
+      "婴幼儿": 42901,
+      "icia": 42902,
+      "agland": 42903,
+      "▁timezone": 42904,
+      "▁Retro": 42905,
+      "绑架": 42906,
+      "和企业": 42907,
+      "是美国": 42908,
+      "ENCES": 42909,
+      "▁Subjects": 42910,
+      "|^": 42911,
+      "临近": 42912,
+      "癫痫病": 42913,
+      "antib": 42914,
+      "文体": 42915,
+      "▁rider": 42916,
+      "很明显": 42917,
+      "nters": 42918,
+      "▁Suff": 42919,
+      "▁outlines": 42920,
+      "实效": 42921,
+      "auf": 42922,
+      "▁Brent": 42923,
+      "强力": 42924,
+      "邪恶": 42925,
+      "▁yr": 42926,
+      "▁dst": 42927,
+      "Tuple": 42928,
+      "ustin": 42929,
+      "▁scalable": 42930,
+      "▁PlayStation": 42931,
+      "▁mins": 42932,
+      "解题的关键": 42933,
+      "复试": 42934,
+      "就可": 42935,
+      "赤峰市": 42936,
+      "孙子": 42937,
+      "ija": 42938,
+      "▁Organic": 42939,
+      "▁methodological": 42940,
+      "nbecause": 42941,
+      "nSelected": 42942,
+      "▁outdated": 42943,
+      "adb": 42944,
+      "急需": 42945,
+      "▁grill": 42946,
+      "酌情从轻处罚": 42947,
+      "▁anonym": 42948,
+      "迷失": 42949,
+      "nJun": 42950,
+      "未成年人": 42951,
+      "偏偏": 42952,
+      "的非": 42953,
+      "总结了": 42954,
+      ")</": 42955,
+      "▁Elite": 42956,
+      "(|": 42957,
+      "▁EST": 42958,
+      "▁futures": 42959,
+      "山地": 42960,
+      "goal": 42961,
+      "oit": 42962,
+      "年轻的": 42963,
+      "的含量": 42964,
+      "葡萄牙": 42965,
+      "期末": 42966,
+      "Zip": 42967,
+      "Decor": 42968,
+      "Bas": 42969,
+      "nSk": 42970,
+      "▁gloves": 42971,
+      "▁theoretically": 42972,
+      "家庭的": 42973,
+      "语气": 42974,
+      "类小说": 42975,
+      "民事权利和诉讼权利": 42976,
+      "传来": 42977,
+      "nDE": 42978,
+      "]])\\": 42979,
+      "prep": 42980,
+      "nJack": 42981,
+      "ypical": 42982,
+      "▁Colin": 42983,
+      "▁frontal": 42984,
+      "normalize": 42985,
+      "悟空": 42986,
+      "▁Portal": 42987,
+      "▁attacking": 42988,
+      "▁irrational": 42989,
+      ":</": 42990,
+      "▁Kyle": 42991,
+      "PUB": 42992,
+      "大家在": 42993,
+      "▁economies": 42994,
+      "早晚": 42995,
+      "▁glyph": 42996,
+      "小结": 42997,
+      "所未": 42998,
+      "nfn": 42999,
+      "aturated": 43000,
+      "加西亚": 43001,
+      "融券": 43002,
+      "pref": 43003,
+      "单项式": 43004,
+      "▁awkward": 43005,
+      "低位": 43006,
+      "ERP": 43007,
+      "第一批": 43008,
+      "reads": 43009,
+      "▁damp": 43010,
+      "▁watches": 43011,
+      "三款": 43012,
+      "adh": 43013,
+      "工艺品": 43014,
+      "()),": 43015,
+      "子弹": 43016,
+      "宝贵": 43017,
+      "strom": 43018,
+      "工学": 43019,
+      "恭喜": 43020,
+      "ми": 43021,
+      "阵地": 43022,
+      "utc": 43023,
+      "小时内": 43024,
+      "Ur": 43025,
+      "第十四": 43026,
+      "ASC": 43027,
+      "ibo": 43028,
+      "▁terminated": 43029,
+      "prog": 43030,
+      "tracking": 43031,
+      "▁screenshots": 43032,
+      "增高": 43033,
+      "我用": 43034,
+      "杜绝": 43035,
+      "西风": 43036,
+      "来看看": 43037,
+      "asso": 43038,
+      "▁tid": 43039,
+      "大兴": 43040,
+      "mutex": 43041,
+      "▁collar": 43042,
+      "▁pairing": 43043,
+      "网络的": 43044,
+      "▁Symphony": 43045,
+      "不定": 43046,
+      "请在": 43047,
+      "金色": 43048,
+      "▁fog": 43049,
+      "及其委托代理人": 43050,
+      "▁hypertensive": 43051,
+      "已是": 43052,
+      "而被": 43053,
+      "▁sarc": 43054,
+      "▁psychiat": 43055,
+      "高潮": 43056,
+      "East": 43057,
+      "科学研究": 43058,
+      "▁deny": 43059,
+      "▁Derby": 43060,
+      "yclerView": 43061,
+      "▁Ambassador": 43062,
+      "blr": 43063,
+      "▁chi": 43064,
+      "先进集体": 43065,
+      "医疗机构": 43066,
+      "nStudents": 43067,
+      "prehensive": 43068,
+      "相爱": 43069,
+      "的人都": 43070,
+      "▁hospitalization": 43071,
+      "日用": 43072,
+      "棋牌": 43073,
+      "▁Mü": 43074,
+      "unless": 43075,
+      "▁pioneer": 43076,
+      "rpm": 43077,
+      "pies": 43078,
+      "也越来越": 43079,
+      "ZZ": 43080,
+      "副总裁": 43081,
+      "▁argc": 43082,
+      "科技有限公司于": 43083,
+      "Connector": 43084,
+      "（）": 43085,
+      "edi": 43086,
+      "▁polished": 43087,
+      "▁invisible": 43088,
+      "Maps": 43089,
+      "phon": 43090,
+      "nCharts": 43091,
+      "▁weaker": 43092,
+      "▁adenosine": 43093,
+      "再由": 43094,
+      "糖果": 43095,
+      "威尼斯": 43096,
+      "▁Scheme": 43097,
+      "ndist": 43098,
+      "适用法律正确": 43099,
+      "creation": 43100,
+      "威廉": 43101,
+      "▁ak": 43102,
+      "网络上": 43103,
+      "▁ses": 43104,
+      "▁因为": 43105,
+      "▁fortune": 43106,
+      "tj": 43107,
+      "ULD": 43108,
+      "二十年": 43109,
+      "解决办法": 43110,
+      "azione": 43111,
+      "▁Produ": 43112,
+      "▁comprise": 43113,
+      "nine": 43114,
+      "particle": 43115,
+      "时钟": 43116,
+      "▁toe": 43117,
+      "▁pencil": 43118,
+      "打败": 43119,
+      "裂缝": 43120,
+      "量刑建议": 43121,
+      "▁yacht": 43122,
+      "餐桌": 43123,
+      "acia": 43124,
+      "atra": 43125,
+      "有关单位": 43126,
+      "tborder": 43127,
+      "元年": 43128,
+      "nAut": 43129,
+      "ulner": 43130,
+      "usually": 43131,
+      "▁provinces": 43132,
+      "▁thriller": 43133,
+      "▁biomarker": 43134,
+      "Req": 43135,
+      "▁redundant": 43136,
+      "▁terminology": 43137,
+      "锅内": 43138,
+      "▁NSW": 43139,
+      "微观": 43140,
+      "登上": 43141,
+      "LOB": 43142,
+      "提交了": 43143,
+      "▁Mis": 43144,
+      "▁Bobby": 43145,
+      "▁intriguing": 43146,
+      "千秋": 43147,
+      "赌场": 43148,
+      "空间的": 43149,
+      "ixes": 43150,
+      "▁bush": 43151,
+      "▁巴": 43152,
+      "LEX": 43153,
+      "起到了": 43154,
+      "queryset": 43155,
+      "▁motility": 43156,
+      "遍布": 43157,
+      "Hop": 43158,
+      "▁instrumentation": 43159,
+      "拖拉": 43160,
+      "Buf": 43161,
+      "试剂": 43162,
+      "用户名": 43163,
+      "▁pid": 43164,
+      "复杂选择题": 43165,
+      "商户": 43166,
+      "Widgets": 43167,
+      "▁missile": 43168,
+      "本院依法组成合议庭对本案": 43169,
+      "新村": 43170,
+      "青铜": 43171,
+      "发生率": 43172,
+      "▁academics": 43173,
+      "说完": 43174,
+      "ews": 43175,
+      "原有的": 43176,
+      "orta": 43177,
+      "oints": 43178,
+      "也不错": 43179,
+      "nwould": 43180,
+      "typedef": 43181,
+      "ularity": 43182,
+      "nPlayers": 43183,
+      "表述": 43184,
+      "Geo": 43185,
+      "的土地": 43186,
+      "丰县": 43187,
+      "果断": 43188,
+      "那我": 43189,
+      "可得答案": 43190,
+      "▁Johns": 43191,
+      "一世": 43192,
+      "▁cosine": 43193,
+      "▁undertake": 43194,
+      "▁epigenetic": 43195,
+      "cg": 43196,
+      "水资源": 43197,
+      "▁aids": 43198,
+      "为客户提供": 43199,
+      "▁cached": 43200,
+      "出生地": 43201,
+      "发改委": 43202,
+      "nTitle": 43203,
+      "DOC": 43204,
+      "▁Hunting": 43205,
+      "Neg": 43206,
+      "oge": 43207,
+      "领导的": 43208,
+      "abal": 43209,
+      "选项中挑选出": 43210,
+      "▁λ": 43211,
+      "之王": 43212,
+      "后经": 43213,
+      "▁pir": 43214,
+      "▁confer": 43215,
+      "▁Raymond": 43216,
+      "▁DH": 43217,
+      "tprintf": 43218,
+      "▁ud": 43219,
+      "椭圆形": 43220,
+      "▁tattoo": 43221,
+      "▁replied": 43222,
+      "▁sensation": 43223,
+      "▁recognizes": 43224,
+      "IMG": 43225,
+      "喜爱的": 43226,
+      "公民身份号码": 43227,
+      "▁marathon": 43228,
+      ":,": 43229,
+      "vp": 43230,
+      "一笔": 43231,
+      "▁编辑": 43232,
+      "January": 43233,
+      "▁catering": 43234,
+      "StatusCode": 43235,
+      "▁sap": 43236,
+      "▁factorization": 43237,
+      "全会": 43238,
+      "板上": 43239,
+      "▁realise": 43240,
+      "团的": 43241,
+      "找我": 43242,
+      ")\\\\)\\": 43243,
+      "▁Keys": 43244,
+      "encoder": 43245,
+      "visibility": 43246,
+      "▁Publisher": 43247,
+      "▁Collections": 43248,
+      "吉利": 43249,
+      "性状": 43250,
+      "▁phag": 43251,
+      "▁fragr": 43252,
+      "▁supervised": 43253,
+      "肢体": 43254,
+      "nPeter": 43255,
+      "▁financially": 43256,
+      "衬衫": 43257,
+      "▁Sugg": 43258,
+      "▁Basin": 43259,
+      "▁flora": 43260,
+      "道题": 43261,
+      "分钟左右": 43262,
+      "tabular": 43263,
+      "▁Rav": 43264,
+      "▁Lights": 43265,
+      "玛丽": 43266,
+      "▁FX": 43267,
+      "▁irres": 43268,
+      "September": 43269,
+      "特长": 43270,
+      "坚持以": 43271,
+      "mort": 43272,
+      "Autom": 43273,
+      "▁Jamie": 43274,
+      "nBrazil": 43275,
+      "▁supers": 43276,
+      "通车": 43277,
+      "成年人": 43278,
+      "TABLE": 43279,
+      "绘本": 43280,
+      "落败": 43281,
+      "大城市": 43282,
+      "的文件": 43283,
+      "治疗方法": 43284,
+      "references": 43285,
+      "万多": 43286,
+      "做起": 43287,
+      "尼尔": 43288,
+      "behavior": 43289,
+      "▁flatten": 43290,
+      "▁Templates": 43291,
+      "▁Calculator": 43292,
+      "▁Wellington": 43293,
+      "▁alternating": 43294,
+      "的比赛": 43295,
+      "并处": 43296,
+      "拿来": 43297,
+      "的就": 43298,
+      "▁remin": 43299,
+      "学士学位": 43300,
+      "Break": 43301,
+      "▁skirt": 43302,
+      "fab": 43303,
+      "▁Diocese": 43304,
+      "#>": 43305,
+      "seys": 43306,
+      "发力": 43307,
+      "volumes": 43308,
+      "▁Shadow": 43309,
+      "cott": 43310,
+      "orphism": 43311,
+      "▁testimony": 43312,
+      "固化": 43313,
+      "▁Clo": 43314,
+      "本院经审理": 43315,
+      "▁Chase": 43316,
+      "AddField": 43317,
+      "民用": 43318,
+      "Arc": 43319,
+      "itos": 43320,
+      "VB": 43321,
+      "膜炎": 43322,
+      "的国际": 43323,
+      "▁ribbon": 43324,
+      "filtered": 43325,
+      "▁Automatic": 43326,
+      "▁valuation": 43327,
+      "新一轮": 43328,
+      "注册资本": 43329,
+      "▁rapper": 43330,
+      "▁biography": 43331,
+      "一贯": 43332,
+      "同情": 43333,
+      "评委": 43334,
+      "过关": 43335,
+      "创业者": 43336,
+      "并结合": 43337,
+      "有意义": 43338,
+      "▁jumps": 43339,
+      "在法律规定的范围内处分自己的": 43340,
+      "▁Jake": 43341,
+      "▁handlers": 43342,
+      "仙女": 43343,
+      "各省": 43344,
+      "救灾": 43345,
+      "本事": 43346,
+      "▁routines": 43347,
+      "脑海": 43348,
+      "现阶段": 43349,
+      "Clip": 43350,
+      "▁Legisl": 43351,
+      "二楼": 43352,
+      "iculty": 43353,
+      "▁Durham": 43354,
+      "环绕": 43355,
+      "ovsk": 43356,
+      "myanswer": 43357,
+      "达成和解协议": 43358,
+      "▁Preview": 43359,
+      "▁Saskatchewan": 43360,
+      "等各种": 43361,
+      "重庆市渝": 43362,
+      "受案": 43363,
+      "而我": 43364,
+      "草案": 43365,
+      "▁DP": 43366,
+      "/\",\\": 43367,
+      "asured": 43368,
+      "▁Conduct": 43369,
+      "▁nowadays": 43370,
+      "backgroundColor": 43371,
+      "内衣": 43372,
+      "uxe": 43373,
+      "银川市": 43374,
+      "upal": 43375,
+      "▁Ari": 43376,
+      "▁Estab": 43377,
+      "credited": 43378,
+      "复活": 43379,
+      "slash": 43380,
+      "▁Superv": 43381,
+      "▁helicopter": 43382,
+      "ICO": 43383,
+      "nPass": 43384,
+      "第三十二条": 43385,
+      "▁sb": 43386,
+      "什么样": 43387,
+      "▁Empty": 43388,
+      "▁ü": 43389,
+      "景象": 43390,
+      "水和": 43391,
+      "装扮": 43392,
+      "解锁": 43393,
+      "▁TT": 43394,
+      "▁quer": 43395,
+      "上诉于浙江省": 43396,
+      "低头": 43397,
+      "的面": 43398,
+      "畜牧业": 43399,
+      "▁Lil": 43400,
+      "▁pore": 43401,
+      "browse": 43402,
+      "blankInput": 43403,
+      "人说": 43404,
+      "解压": 43405,
+      "nob": 43406,
+      "第一百九": 43407,
+      "▁administer": 43408,
+      "驾车": 43409,
+      "结构和": 43410,
+      "▁darkness": 43411,
+      "arker": 43412,
+      "▁场上位置": 43413,
+      "ENTIAL": 43414,
+      "▁mening": 43415,
+      "informatics": 43416,
+      "templateUrl": 43417,
+      "▁ACE": 43418,
+      "▁Wer": 43419,
+      "VICES": 43420,
+      "\\\\}\\\\)": 43421,
+      "optera": 43422,
+      "▁Marcus": 43423,
+      "举个": 43424,
+      "贫血": 43425,
+      "▁',": 43426,
+      "会自动": 43427,
+      "工作和": 43428,
+      "并提供": 43429,
+      "发声": 43430,
+      "▁Opportun": 43431,
+      "\"|": 43432,
+      "LU": 43433,
+      "发还": 43434,
+      "sinx": 43435,
+      "系列产品": 43436,
+      "▁mmol": 43437,
+      "nAnyone": 43438,
+      "▁î": 43439,
+      "罚字": 43440,
+      "葡萄糖": 43441,
+      "▁Tip": 43442,
+      "downloads": 43443,
+      "▁Designer": 43444,
+      "CHO": 43445,
+      "LAS": 43446,
+      "adjusted": 43447,
+      "天线": 43448,
+      "的投资": 43449,
+      "Pick": 43450,
+      "▁Wrest": 43451,
+      "▁DIRECT": 43452,
+      "▁playoff": 43453,
+      "处处长": 43454,
+      "tpadding": 43455,
+      "本刊": 43456,
+      "鸡汤": 43457,
+      "nCc": 43458,
+      "INCLUDING": 43459,
+      "源码": 43460,
+      "目前在": 43461,
+      "verte": 43462,
+      "cation": 43463,
+      "▁patrol": 43464,
+      "片区": 43465,
+      "人民网": 43466,
+      "完全平方公式": 43467,
+      "▁Heaven": 43468,
+      "▁epidemiological": 43469,
+      "▁divides": 43470,
+      "▁harassment": 43471,
+      "▁realization": 43472,
+      "▁sag": 43473,
+      "▁Revival": 43474,
+      "十里": 43475,
+      "查阅": 43476,
+      "本科学": 43477,
+      "Clock": 43478,
+      "▁Contents": 43479,
+      "▁traction": 43480,
+      "公摊": 43481,
+      "卡拉": 43482,
+      "着眼": 43483,
+      "tweet": 43484,
+      "mbalib": 43485,
+      "▁Checklist": 43486,
+      "油气": 43487,
+      "命名为": 43488,
+      "吸血": 43489,
+      "视线": 43490,
+      "▁&#": 43491,
+      "▁tuition": 43492,
+      "\\'}": 43493,
+      "nSpecies": 43494,
+      "巧妙": 43495,
+      "租车": 43496,
+      "Btn": 43497,
+      "▁MG": 43498,
+      "那样的": 43499,
+      "wild": 43500,
+      "producer": 43501,
+      "解剖": 43502,
+      "那年": 43503,
+      "jon": 43504,
+      "月中旬": 43505,
+      "▁合作人物": 43506,
+      "稳健": 43507,
+      "atem": 43508,
+      "Decoder": 43509,
+      "urnames": 43510,
+      "▁Blake": 43511,
+      "第一中级人民法院": 43512,
+      "带的": 43513,
+      "tls": 43514,
+      "更多的是": 43515,
+      "▁cite": 43516,
+      "nAthletes": 43517,
+      "▁Significant": 43518,
+      "▁chic": 43519,
+      "▁reluct": 43520,
+      "财保": 43521,
+      "总决赛": 43522,
+      "拿出来": 43523,
+      "教师的": 43524,
+      "就是因为": 43525,
+      "▁Bedroom": 43526,
+      "▁inevitable": 43527,
+      "▁reinforcement": 43528,
+      "后在": 43529,
+      "满族": 43530,
+      "享有的": 43531,
+      "▁Hass": 43532,
+      "进行了分析": 43533,
+      "▁Ether": 43534,
+      "▁Rapid": 43535,
+      "▁safegu": 43536,
+      "国资": 43537,
+      "▁HM": 43538,
+      "中西医": 43539,
+      "蒙古族": 43540,
+      "▁booth": 43541,
+      "馆藏": 43542,
+      "血液中": 43543,
+      "nSpecial": 43544,
+      "▁dessert": 43545,
+      "nRecently": 43546,
+      "可口": 43547,
+      "著有": 43548,
+      "▁zb": 43549,
+      "AW": 43550,
+      "nMale": 43551,
+      "▁grasp": 43552,
+      "不支持": 43553,
+      "__':\\": 43554,
+      "▁tandem": 43555,
+      "▁conformational": 43556,
+      "估值": 43557,
+      "挤压": 43558,
+      "流的": 43559,
+      "以为是": 43560,
+      "Added": 43561,
+      "▁upgrading": 43562,
+      "妥善": 43563,
+      "REST": 43564,
+      "toolbar": 43565,
+      "天花": 43566,
+      "自称": 43567,
+      "——“": 43568,
+      "Anim": 43569,
+      "Interfaces": 43570,
+      "品类": 43571,
+      "首先是": 43572,
+      "▁Parser": 43573,
+      "▁herein": 43574,
+      "房源": 43575,
+      "BER": 43576,
+      "▁violin": 43577,
+      "▁academy": 43578,
+      "填词": 43579,
+      "麻辣": 43580,
+      "推荐的": 43581,
+      "同时也是": 43582,
+      "▁cerebro": 43583,
+      "getItem": 43584,
+      "department": 43585,
+      "▁defeating": 43586,
+      "滤波": 43587,
+      "龙泉": 43588,
+      "了这个": 43589,
+      "Todo": 43590,
+      "▁transformer": 43591,
+      "重合": 43592,
+      "antiago": 43593,
+      "▁packaged": 43594,
+      "ISC": 43595,
+      "▁设置": 43596,
+      "nEntering": 43597,
+      "▁fp": 43598,
+      "'):\\": 43599,
+      "Phase": 43600,
+      "品尝": 43601,
+      "承接": 43602,
+      "emo": 43603,
+      "▁creature": 43604,
+      "将对": 43605,
+      "开阔": 43606,
+      "orf": 43607,
+      "ать": 43608,
+      "▁Spin": 43609,
+      "▁owing": 43610,
+      "▁tenants": 43611,
+      "▁eligibility": 43612,
+      "bigg": 43613,
+      "}\\\\}\\\\)": 43614,
+      "▁civilian": 43615,
+      "▁Copenhagen": 43616,
+      "▁frustrating": 43617,
+      "嘴里": 43618,
+      "脾胃": 43619,
+      "结合起来": 43620,
+      "▁ruby": 43621,
+      "Market": 43622,
+      "domains": 43623,
+      "membrane": 43624,
+      "▁Implement": 43625,
+      "物管": 43626,
+      "//#": 43627,
+      "rtype": 43628,
+      "▁Teachers": 43629,
+      "▁vertically": 43630,
+      "逐年": 43631,
+      "链表": 43632,
+      "ndouble": 43633,
+      "▁diagnoses": 43634,
+      "又不是": 43635,
+      "agra": 43636,
+      "▁Fol": 43637,
+      "anmar": 43638,
+      "▁boiler": 43639,
+      "assistant": 43640,
+      "isis": 43641,
+      "▁curt": 43642,
+      "enerated": 43643,
+      "牛皮": 43644,
+      "小姐姐": 43645,
+      "银行的": 43646,
+      "▁lbs": 43647,
+      "制作人": 43648,
+      "DUCT": 43649,
+      "▁pads": 43650,
+      "▁ranch": 43651,
+      "▁inaccur": 43652,
+      "会造成": 43653,
+      "healthy": 43654,
+      "了她": 43655,
+      "涨价": 43656,
+      "不代表": 43657,
+      "▁dorm": 43658,
+      "▁stitch": 43659,
+      "眼科": 43660,
+      "满意的": 43661,
+      "()]\\": 43662,
+      "▁sab": 43663,
+      "绝大部分": 43664,
+      "会更": 43665,
+      "的特征": 43666,
+      "))**": 43667,
+      "▁\\'$": 43668,
+      "▁Rent": 43669,
+      "▁redox": 43670,
+      "Scripts": 43671,
+      "▁encrypt": 43672,
+      "短板": 43673,
+      "afety": 43674,
+      "Bounds": 43675,
+      "▁belonged": 43676,
+      "每件": 43677,
+      "▁RI": 43678,
+      "提起的": 43679,
+      "的一大": 43680,
+      "▁McL": 43681,
+      "有下列情形": 43682,
+      "▁readable": 43683,
+      "极度": 43684,
+      "lies": 43685,
+      "总部位于": 43686,
+      "▁hull": 43687,
+      "▁Warriors": 43688,
+      "▁Ultimate": 43689,
+      "速度快": 43690,
+      "berger": 43691,
+      "▁Spencer": 43692,
+      "▁punishment": 43693,
+      "地方政府": 43694,
+      "SG": 43695,
+      "志刚": 43696,
+      "梦里": 43697,
+      "当年的": 43698,
+      "▁Failed": 43699,
+      "▁cyclists": 43700,
+      "Ba": 43701,
+      "申请费": 43702,
+      "TAIN": 43703,
+      "igion": 43704,
+      "retch": 43705,
+      "XL": 43706,
+      "**,": 43707,
+      "提出异议": 43708,
+      "pagination": 43709,
+      "&\\\\": 43710,
+      "养老金": 43711,
+      "▁Comics": 43712,
+      "白色的": 43713,
+      "inin": 43714,
+      "视图": 43715,
+      "首要": 43716,
+      "nRob": 43717,
+      "arange": 43718,
+      "▁beaten": 43719,
+      "▁economical": 43720,
+      "▁designation": 43721,
+      "改装": 43722,
+      "无毛": 43723,
+      "楼下": 43724,
+      "绿地": 43725,
+      "SEM": 43726,
+      "的办法": 43727,
+      "▁Facility": 43728,
+      "▁Going": 43729,
+      "▁Jamaica": 43730,
+      "产于": 43731,
+      "黄花": 43732,
+      "Mad": 43733,
+      "rina": 43734,
+      "▁parenting": 43735,
+      "并能": 43736,
+      "法官助理": 43737,
+      "特别是在": 43738,
+      "illon": 43739,
+      "▁suppressor": 43740,
+      "在小": 43741,
+      "河口": 43742,
+      "Pars": 43743,
+      "▁relying": 43744,
+      "误区": 43745,
+      "Bre": 43746,
+      "损耗": 43747,
+      "课后": 43748,
+      "吸入": 43749,
+      "规避": 43750,
+      "Apache": 43751,
+      "商的": 43752,
+      "很低": 43753,
+      "电极": 43754,
+      "APH": 43755,
+      "所以我们": 43756,
+      "▁EF": 43757,
+      "产业园": 43758,
+      "curve": 43759,
+      "请仔细研究": 43760,
+      "▁corrections": 43761,
+      "▁constituents": 43762,
+      "这只": 43763,
+      "▁tended": 43764,
+      "楼上": 43765,
+      "'])": 43766,
+      "启蒙": 43767,
+      "收据": 43768,
+      "uates": 43769,
+      "▁Hybrid": 43770,
+      "国民经济": 43771,
+      "▁Corner": 43772,
+      "▁Mirror": 43773,
+      "▁refine": 43774,
+      "物业费": 43775,
+      "昨晚": 43776,
+      "magic": 43777,
+      "▁onions": 43778,
+      "嫌弃": 43779,
+      "▁nn": 43780,
+      "本文从": 43781,
+      "今年以来": 43782,
+      "▁Thought": 43783,
+      "elsen": 43784,
+      "volution": 43785,
+      "stant": 43786,
+      "▁criticized": 43787,
+      "国旗": 43788,
+      "▁Triple": 43789,
+      "未发现被执行人": 43790,
+      "▁unfair": 43791,
+      "夺冠": 43792,
+      "认真的": 43793,
+      "TU": 43794,
+      "和不": 43795,
+      "榜样": 43796,
+      "agul": 43797,
+      "重要意义": 43798,
+      "hydrox": 43799,
+      "▁hometown": 43800,
+      "经典的": 43801,
+      "巡逻": 43802,
+      "朝着": 43803,
+      "▁socially": 43804,
+      "▁saturated": 43805,
+      "次会议": 43806,
+      "▁Skills": 43807,
+      "执行案件立案": 43808,
+      "▁Cannot": 43809,
+      "▁vertebr": 43810,
+      ")\\\\;": 43811,
+      "▁Nokia": 43812,
+      "▁Shakespeare": 43813,
+      "专卖": 43814,
+      "八卦": 43815,
+      "OPEN": 43816,
+      "Additional": 43817,
+      "▁hippocampal": 43818,
+      "▁Pretty": 43819,
+      "第二中级人民法院": 43820,
+      "▁repaired": 43821,
+      "▁malignancy": 43822,
+      "变了": 43823,
+      "死后": 43824,
+      ")}$\\": 43825,
+      "拒不履行": 43826,
+      "▁GPIO": 43827,
+      "▁Artists": 43828,
+      "▁holistic": 43829,
+      "▁Din": 43830,
+      "▁NSString": 43831,
+      "▁architectures": 43832,
+      "也挺": 43833,
+      "应变": 43834,
+      "黏膜": 43835,
+      "ryl": 43836,
+      "压制": 43837,
+      "大面积": 43838,
+      "脑血管": 43839,
+      "Used": 43840,
+      "burse": 43841,
+      "▁graded": 43842,
+      "▁Psychology": 43843,
+      "拓宽": 43844,
+      "itr": 43845,
+      "yon": 43846,
+      "▁Fork": 43847,
+      "▁thankful": 43848,
+      "偏向": 43849,
+      "rals": 43850,
+      "▁Bros": 43851,
+      "▁Norfolk": 43852,
+      "▁infiltr": 43853,
+      "▁headlines": 43854,
+      "▁学校": 43855,
+      "ellery": 43856,
+      "nAnyway": 43857,
+      "▁Paralympics": 43858,
+      "高分": 43859,
+      "jsp": 43860,
+      "SERVICE": 43861,
+      "▁Potter": 43862,
+      "生机": 43863,
+      "▁详细": 43864,
+      "▁voy": 43865,
+      "▁rook": 43866,
+      "▁reinforced": 43867,
+      "相见": 43868,
+      "划算": 43869,
+      "火花": 43870,
+      "绍兴市": 43871,
+      "omeric": 43872,
+      "▁skeleton": 43873,
+      "赞同": 43874,
+      "Stub": 43875,
+      "benz": 43876,
+      "▁guit": 43877,
+      "▁Surre": 43878,
+      "▁reciprocal": 43879,
+      "▁gh": 43880,
+      "nbuild": 43881,
+      "observer": 43882,
+      "ifty": 43883,
+      "unused": 43884,
+      "▁underwater": 43885,
+      "无职业": 43886,
+      "fair": 43887,
+      "严格按照": 43888,
+      "▁adulthood": 43889,
+      ",’": 43890,
+      "回了": 43891,
+      "轮廓": 43892,
+      "WiFi": 43893,
+      "lived": 43894,
+      "videos": 43895,
+      "▁pursued": 43896,
+      "▁alkaline": 43897,
+      "▁conveniently": 43898,
+      "▁contradiction": 43899,
+      "不由": 43900,
+      "施肥": 43901,
+      "吃过": 43902,
+      "氯化": 43903,
+      "anie": 43904,
+      "▁Brandon": 43905,
+      "有很大的": 43906,
+      "▁acquiring": 43907,
+      "北市": 43908,
+      "迫切": 43909,
+      "▁edu": 43910,
+      "▁Teams": 43911,
+      "▁embol": 43912,
+      "▁Origin": 43913,
+      "淋漓": 43914,
+      "如果我们": 43915,
+      "minded": 43916,
+      "可以使": 43917,
+      "适配器": 43918,
+      "▁gameplay": 43919,
+      "▁extinction": 43920,
+      "国会": 43921,
+      "▁ADD": 43922,
+      "▁heater": 43923,
+      "enia": 43924,
+      "iblical": 43925,
+      "▁modulo": 43926,
+      "▁impulse": 43927,
+      "nfe": 43928,
+      "工程的": 43929,
+      "▁Discord": 43930,
+      "传染病": 43931,
+      "内分泌": 43932,
+      "广泛应用于": 43933,
+      "▁Topic": 43934,
+      "分娩": 43935,
+      "ncl": 43936,
+      "nCRE": 43937,
+      "▁marry": 43938,
+      "openapi": 43939,
+      "neapolis": 43940,
+      "funded": 43941,
+      "gitcommit": 43942,
+      "▁ра": 43943,
+      "▁adjuvant": 43944,
+      "▁translocation": 43945,
+      "Friday": 43946,
+      "▁cakes": 43947,
+      "▁helm": 43948,
+      "▁sqlite": 43949,
+      "▁squamous": 43950,
+      "angled": 43951,
+      "▁Vladimir": 43952,
+      "▁qualifications": 43953,
+      "量和": 43954,
+      "olan": 43955,
+      "▁Apps": 43956,
+      "南非": 43957,
+      "▁浙江": 43958,
+      "入党": 43959,
+      "听见": 43960,
+      "等信息": 43961,
+      "羽毛球": 43962,
+      "warf": 43963,
+      "Processing": 43964,
+      "台词": 43965,
+      "不低于": 43966,
+      "对我们": 43967,
+      "▁Dent": 43968,
+      "Filters": 43969,
+      "监区": 43970,
+      "土地承包": 43971,
+      "城县人民法院": 43972,
+      "######\\": 43973,
+      "▁Traffic": 43974,
+      "▁chromosomal": 43975,
+      "▁Rush": 43976,
+      "▁Million": 43977,
+      "▁prenatal": 43978,
+      "杨柳": 43979,
+      "▁acclaimed": 43980,
+      "高一": 43981,
+      "从一个": 43982,
+      "理论上": 43983,
+      "的水平": 43984,
+      "uning": 43985,
+      "▁preclinical": 43986,
+      "▁rentals": 43987,
+      "▁teenager": 43988,
+      "translationtype": 43989,
+      "隐形": 43990,
+      "▁}}/": 43991,
+      "▁Rica": 43992,
+      "篇章": 43993,
+      "除去": 43994,
+      "▁玄幻": 43995,
+      "zhen": 43996,
+      "▁kom": 43997,
+      "yclic": 43998,
+      "▁flank": 43999,
+      "({\"": 44000,
+      "没关系": 44001,
+      "dash": 44002,
+      "▁Aub": 44003,
+      "▁Manufacturing": 44004,
+      "参会": 44005,
+      "nAND": 44006,
+      "optimal": 44007,
+      "▁freight": 44008,
+      "▁optimizing": 44009,
+      "nTim": 44010,
+      "locfilehash": 44011,
+      "openlocfilehash": 44012,
+      "sourcegitcommit": 44013,
+      "▁Tunis": 44014,
+      "▁Gabriel": 44015,
+      "▁Desert": 44016,
+      "▁Maintenance": 44017,
+      "出差": 44018,
+      "有不少": 44019,
+      "▁dilig": 44020,
+      "▁Browse": 44021,
+      "▁immunore": 44022,
+      "Dam": 44023,
+      "lickr": 44024,
+      "▁Bailey": 44025,
+      "展出": 44026,
+      "buk": 44027,
+      "▁Pump": 44028,
+      "▁bang": 44029,
+      "▁backpack": 44030,
+      "光的": 44031,
+      "Hit": 44032,
+      "双曲线": 44033,
+      "日报道": 44034,
+      "本公司": 44035,
+      "▁chor": 44036,
+      "pretty": 44037,
+      "▁knees": 44038,
+      "▁versa": 44039,
+      "nEarlier": 44040,
+      "TRIBUTORS": 44041,
+      "nWeek": 44042,
+      "▁angiography": 44043,
+      "▁\">": 44044,
+      "有自己的": 44045,
+      "法律依据": 44046,
+      "资格考试": 44047,
+      "软件开发": 44048,
+      "▁Sheet": 44049,
+      "▁fungi": 44050,
+      "verages": 44051,
+      "▁detectors": 44052,
+      "欢喜": 44053,
+      "潇湘": 44054,
+      "hee": 44055,
+      "驾驶的": 44056,
+      "自然科学基金": 44057,
+      "谈谈": 44058,
+      "非凡": 44059,
+      "▁TLR": 44060,
+      "▁Kiss": 44061,
+      "▁Pearl": 44062,
+      "nstring": 44063,
+      "▁unsafe": 44064,
+      "cobacter": 44065,
+      "▁Philippine": 44066,
+      "▁grandmother": 44067,
+      "疲惫": 44068,
+      "这里是": 44069,
+      "igroup": 44070,
+      "施行": 44071,
+      "越是": 44072,
+      "不断提升": 44073,
+      "▁ceil": 44074,
+      "otives": 44075,
+      "pq": 44076,
+      "引人": 44077,
+      "STER": 44078,
+      "tails": 44079,
+      "uliar": 44080,
+      "▁bitter": 44081,
+      "▁Syndrome": 44082,
+      "统战": 44083,
+      "▁SUV": 44084,
+      "ntrans": 44085,
+      "▁Tommy": 44086,
+      "onuclear": 44087,
+      "下行": 44088,
+      "多月": 44089,
+      "收纳": 44090,
+      "木兰": 44091,
+      "]}\\": 44092,
+      "▁HREF": 44093,
+      "Looking": 44094,
+      "▁likewise": 44095,
+      "▁米": 44096,
+      "口径": 44097,
+      "接头": 44098,
+      "zed": 44099,
+      "更何况": 44100,
+      "▁Exc": 44101,
+      "icorn": 44102,
+      "▁Avoid": 44103,
+      "▁Devon": 44104,
+      "▁migrants": 44105,
+      "Department": 44106,
+      "之日起五日内向本院申请复议一次": 44107,
+      "听着": 44108,
+      "奔跑": 44109,
+      "崇拜": 44110,
+      "RPC": 44111,
+      "▁MV": 44112,
+      "allenges": 44113,
+      "▁WARNING": 44114,
+      "水源": 44115,
+      "透气": 44116,
+      "▁Pam": 44117,
+      "idos": 44118,
+      "▁heels": 44119,
+      "▁plasmid": 44120,
+      "different": 44121,
+      "综述": 44122,
+      "形式为": 44123,
+      "▁identifiers": 44124,
+      "动了": 44125,
+      "油的": 44126,
+      "spr": 44127,
+      "cass": 44128,
+      "namon": 44129,
+      "aucoma": 44130,
+      "WARNING": 44131,
+      "evaluate": 44132,
+      "ieg": 44133,
+      "▁Voc": 44134,
+      "ippers": 44135,
+      "恐龙": 44136,
+      "▁seizure": 44137,
+      "▁butterfly": 44138,
+      "▁continent": 44139,
+      "Certificate": 44140,
+      "上报": 44141,
+      "机车": 44142,
+      "就觉得": 44143,
+      "nels": 44144,
+      "▁Submit": 44145,
+      "▁diagnostics": 44146,
+      "▁贾": 44147,
+      "克的": 44148,
+      "秦皇": 44149,
+      "▁LT": 44150,
+      "生产和": 44151,
+      "国际贸易": 44152,
+      "你这": 44153,
+      "/((": 44154,
+      "留下了": 44155,
+      "市场需求": 44156,
+      "groupId": 44157,
+      "平淡": 44158,
+      "uples": 44159,
+      "▁bins": 44160,
+      "▁cath": 44161,
+      "▁scrub": 44162,
+      "单纯的": 44163,
+      "bands": 44164,
+      "▁Maced": 44165,
+      "改正": 44166,
+      "植入": 44167,
+      "▁demanded": 44168,
+      "▁Wend": 44169,
+      "Utility": 44170,
+      "▁interchange": 44171,
+      "Comb": 44172,
+      "ONLINE": 44173,
+      "alamic": 44174,
+      "selenium": 44175,
+      "▁horrible": 44176,
+      "好听": 44177,
+      "经纬": 44178,
+      "冲刺": 44179,
+      "发达国家": 44180,
+      "▁chunks": 44181,
+      "restrial": 44182,
+      "▁escaped": 44183,
+      "Talk": 44184,
+      "ptime": 44185,
+      "nWater": 44186,
+      "▁Ticket": 44187,
+      "▁DOM": 44188,
+      "说自己": 44189,
+      "nOkay": 44190,
+      "台式": 44191,
+      "建构": 44192,
+      "▁内容": 44193,
+      "▁Cars": 44194,
+      "▁Manuel": 44195,
+      "▁Previously": 44196,
+      "尿酸": 44197,
+      "LIN": 44198,
+      "▁cli": 44199,
+      "▁angiotensin": 44200,
+      "血样": 44201,
+      "atos": 44202,
+      "▁Serbian": 44203,
+      "航母": 44204,
+      "pem": 44205,
+      "确定了": 44206,
+      "▁Nazi": 44207,
+      "ergarten": 44208,
+      "▁transporter": 44209,
+      "细心": 44210,
+      "Received": 44211,
+      "▁presently": 44212,
+      "▁ambassador": 44213,
+      "再加": 44214,
+      "▁macOS": 44215,
+      "▁fasting": 44216,
+      "协定": 44217,
+      "嵌入式": 44218,
+      "▁aboard": 44219,
+      "CreateModel": 44220,
+      "▁德": 44221,
+      "斯塔": 44222,
+      "突出的": 44223,
+      "Cannot": 44224,
+      "▁abbrev": 44225,
+      "released": 44226,
+      "▁Correct": 44227,
+      "▁Failure": 44228,
+      "▁carcinomas": 44229,
+      "▁wt": 44230,
+      "imits": 44231,
+      "指甲": 44232,
+      "MSG": 44233,
+      "快速的": 44234,
+      "外出务工": 44235,
+      "▁varchar": 44236,
+      "函数的单调性": 44237,
+      "▁Decision": 44238,
+      "▁citizenship": 44239,
+      "IFY": 44240,
+      "pix": 44241,
+      "▁DD": 44242,
+      "整体的": 44243,
+      "要从": 44244,
+      "nEND": 44245,
+      "plastic": 44246,
+      "▁wardrobe": 44247,
+      "过来的": 44248,
+      "shall": 44249,
+      "colour": 44250,
+      "申请执行人可以": 44251,
+      "instances": 44252,
+      "QS": 44253,
+      "之情": 44254,
+      "商是": 44255,
+      "将被": 44256,
+      "▁Cod": 44257,
+      "管理制度": 44258,
+      "▁congrat": 44259,
+      "▁preschool": 44260,
+      "grey": 44261,
+      "▁Retail": 44262,
+      "达人": 44263,
+      "监委": 44264,
+      "▁lb": 44265,
+      "enna": 44266,
+      "▁flagship": 44267,
+      "▁categorized": 44268,
+      "公认": 44269,
+      "▁fingerprint": 44270,
+      "二的": 44271,
+      "馒头": 44272,
+      "orange": 44273,
+      "不等式组": 44274,
+      "▁Cort": 44275,
+      "▁wand": 44276,
+      "formerly": 44277,
+      "stricted": 44278,
+      "nScottish": 44279,
+      "))]\\": 44280,
+      "▁Zoo": 44281,
+      "▁Reid": 44282,
+      "抓紧": 44283,
+      "sugg": 44284,
+      "▁Capitol": 44285,
+      "尊严": 44286,
+      "滋润": 44287,
+      "mis": 44288,
+      "之以": 44289,
+      "反比例": 44290,
+      "不断发展": 44291,
+      "\\'$": 44292,
+      "▁committees": 44293,
+      "无正当理由未到庭参加诉讼": 44294,
+      "▁perturbation": 44295,
+      "往前": 44296,
+      "▁我的": 44297,
+      "创新的": 44298,
+      "lifting": 44299,
+      "AO": 44300,
+      "倡议": 44301,
+      "avir": 44302,
+      "▁tailor": 44303,
+      "particip": 44304,
+      "▁milestone": 44305,
+      "术中": 44306,
+      "▁duck": 44307,
+      "paration": 44308,
+      "▁hippocampus": 44309,
+      "▁practitioner": 44310,
+      "idia": 44311,
+      "▁FUN": 44312,
+      "inspired": 44313,
+      "thandoff": 44314,
+      "深情": 44315,
+      "特质": 44316,
+      "路与": 44317,
+      "▁sleeve": 44318,
+      "集的": 44319,
+      "}%\\": 44320,
+      "idian": 44321,
+      "▁finances": 44322,
+      "▁frustrated": 44323,
+      "风貌": 44324,
+      "▁confirming": 44325,
+      "包围": 44326,
+      "建筑材料": 44327,
+      "rocyte": 44328,
+      "▁Objects": 44329,
+      "是企业": 44330,
+      "apur": 44331,
+      "▁aluminium": 44332,
+      "▁retaining": 44333,
+      "▁北": 44334,
+      "▁OC": 44335,
+      "是不会": 44336,
+      "Bridge": 44337,
+      "▁downt": 44338,
+      "▁suspects": 44339,
+      "▁infiltration": 44340,
+      "▁@@": 44341,
+      "lasthandoff": 44342,
+      "能使": 44343,
+      "采访时": 44344,
+      "建筑工程有限公司": 44345,
+      "与否": 44346,
+      "右脚": 44347,
+      "英文名": 44348,
+      "TRAN": 44349,
+      "toire": 44350,
+      "urate": 44351,
+      "ecting": 44352,
+      "nUpdate": 44353,
+      "nProduct": 44354,
+      "赡养": 44355,
+      "▁whis": 44356,
+      "可能有": 44357,
+      "balanced": 44358,
+      "▁residing": 44359,
+      "第二百六十四条": 44360,
+      "▁preterm": 44361,
+      "浓缩": 44362,
+      "国际机场": 44363,
+      "gevity": 44364,
+      "sharing": 44365,
+      "▁indirectly": 44366,
+      "世家": 44367,
+      "红色的": 44368,
+      "Branch": 44369,
+      "▁Bil": 44370,
+      "▁Eddie": 44371,
+      "▁spirits": 44372,
+      "▁judicial": 44373,
+      "有线电视": 44374,
+      "isers": 44375,
+      "contentlocale": 44376,
+      "拍的": 44377,
+      "above": 44378,
+      "中性": 44379,
+      "kal": 44380,
+      "Lookup": 44381,
+      "▁thats": 44382,
+      "▁rename": 44383,
+      "▁sticky": 44384,
+      "ectl": 44385,
+      "▁Guitar": 44386,
+      "▁Downtown": 44387,
+      "疑难": 44388,
+      "省略": 44389,
+      "▁differed": 44390,
+      "▁EPA": 44391,
+      "的最小值": 44392,
+      "被执行人的": 44393,
+      "rename": 44394,
+      "▁metre": 44395,
+      "▁Socket": 44396,
+      "▁warmth": 44397,
+      "December": 44398,
+      "课本": 44399,
+      "PVC": 44400,
+      "美国人": 44401,
+      "horn": 44402,
+      "从小到大": 44403,
+      "ataka": 44404,
+      "▁Considering": 44405,
+      "穿上": 44406,
+      "自己去": 44407,
+      "余名": 44408,
+      "大局": 44409,
+      "capture": 44410,
+      "Commerce": 44411,
+      "增生": 44412,
+      "▁政治面貌": 44413,
+      "RESULT": 44414,
+      "导游": 44415,
+      "CLE": 44416,
+      "▁wars": 44417,
+      "▁Leicester": 44418,
+      "爱人": 44419,
+      "oretic": 44420,
+      "▁Brooks": 44421,
+      "▁genres": 44422,
+      "▁Silicon": 44423,
+      "Tax": 44424,
+      "交通警察": 44425,
+      "▁quiz": 44426,
+      "交纳诉讼费用": 44427,
+      "telling": 44428,
+      "▁Julian": 44429,
+      "▁sulfur": 44430,
+      "案件受理费减半收取": 44431,
+      "诸如": 44432,
+      "生产企业": 44433,
+      "爱尔": 44434,
+      "珊瑚": 44435,
+      "urdy": 44436,
+      "ichel": 44437,
+      "▁个人简介": 44438,
+      "▁semiconductor": 44439,
+      "ylated": 44440,
+      "▁malware": 44441,
+      "▁integrates": 44442,
+      "won": 44443,
+      "▁hey": 44444,
+      "▁backs": 44445,
+      "▁gradients": 44446,
+      "得名": 44447,
+      "▁Alert": 44448,
+      "▁tornado": 44449,
+      "▁latitude": 44450,
+      "大臣": 44451,
+      "就用": 44452,
+      "秋冬": 44453,
+      "frared": 44454,
+      "ncategories": 44455,
+      "jer": 44456,
+      "▁Okay": 44457,
+      "nLater": 44458,
+      "▁colleague": 44459,
+      "茄子": 44460,
+      "追偿": 44461,
+      "期间内": 44462,
+      "forcing": 44463,
+      "▁dealership": 44464,
+      "张力": 44465,
+      "二〇一七年八月": 44466,
+      "▁subtype": 44467,
+      "停留": 44468,
+      "▁Ya": 44469,
+      "esar": 44470,
+      "住房公积金": 44471,
+      "Finder": 44472,
+      "从重": 44473,
+      "切口": 44474,
+      "突变": 44475,
+      "solete": 44476,
+      "jetbrains": 44477,
+      "▁eventual": 44478,
+      "闻名": 44479,
+      "开玩笑": 44480,
+      "ović": 44481,
+      "adors": 44482,
+      "▁Hood": 44483,
+      "▁bloom": 44484,
+      "▁spinning": 44485,
+      "enaissance": 44486,
+      "▁Conditions": 44487,
+      "PID": 44488,
+      "▁springs": 44489,
+      "职业技术学院": 44490,
+      "patients": 44491,
+      "migration": 44492,
+      "峰会": 44493,
+      "bil": 44494,
+      "节假日": 44495,
+      "▁cemetery": 44496,
+      "和尚": 44497,
+      "heit": 44498,
+      "从业人员": 44499,
+      "nCreated": 44500,
+      "▁hydrolysis": 44501,
+      "无理": 44502,
+      "PII": 44503,
+      "游泳池": 44504,
+      "▁Bib": 44505,
+      "▁Lay": 44506,
+      "▁ale": 44507,
+      "Invoke": 44508,
+      "getAll": 44509,
+      "▁Lesson": 44510,
+      "为真": 44511,
+      "▁Heath": 44512,
+      "preventDefault": 44513,
+      "▁sz": 44514,
+      "队伍建设": 44515,
+      "▁Truth": 44516,
+      "▁nickname": 44517,
+      "▁contextual": 44518,
+      "ви": 44519,
+      "民航": 44520,
+      "远方": 44521,
+      "精神的": 44522,
+      "▁pertaining": 44523,
+      "ller": 44524,
+      "Regular": 44525,
+      "pure": 44526,
+      "工程项目": 44527,
+      "▁angel": 44528,
+      "▁Integral": 44529,
+      "债务人": 44530,
+      "以下内容": 44531,
+      "第一百零一条": 44532,
+      "▁LinkedIn": 44533,
+      "全长": 44534,
+      "▁brom": 44535,
+      "▁Anyway": 44536,
+      "▁biopsies": 44537,
+      "营养价值": 44538,
+      "isodes": 44539,
+      "▁specialize": 44540,
+      "ifa": 44541,
+      "Brand": 44542,
+      "履行法律文书确定的义务": 44543,
+      "要求其": 44544,
+      "namese": 44545,
+      "Journal": 44546,
+      "getUser": 44547,
+      "▁Managing": 44548,
+      "loe": 44549,
+      "用户提供": 44550,
+      "sprite": 44551,
+      "两项": 44552,
+      "猴子": 44553,
+      "ulla": 44554,
+      "▁prey": 44555,
+      "▁overlooked": 44556,
+      "▁repetitive": 44557,
+      "▁completeness": 44558,
+      "WW": 44559,
+      "日讯": 44560,
+      "用餐": 44561,
+      ")`\\": 44562,
+      "▁shy": 44563,
+      "▁guitars": 44564,
+      "IBLE": 44565,
+      "就必须": 44566,
+      "Regex": 44567,
+      "▁discour": 44568,
+      "落下": 44569,
+      "在我的": 44570,
+      "科技成果": 44571,
+      "参考书": 44572,
+      "▁Rip": 44573,
+      "美化": 44574,
+      "seat": 44575,
+      "▁CRC": 44576,
+      "▁Rut": 44577,
+      "悔罪表现": 44578,
+      "imeric": 44579,
+      "nestjs": 44580,
+      "ertools": 44581,
+      "▁Hav": 44582,
+      "League": 44583,
+      "▁Pearson": 44584,
+      "▁Encyclopedia": 44585,
+      "越野": 44586,
+      "外交部": 44587,
+      "产品类型": 44588,
+      "检验报告": 44589,
+      "▁Garage": 44590,
+      "▁registr": 44591,
+      "▁retrospectively": 44592,
+      "减刑一年": 44593,
+      "overlay": 44594,
+      "textrm": 44595,
+      "▁shocked": 44596,
+      "都比": 44597,
+      "ioni": 44598,
+      "的评价": 44599,
+      "▁Anth": 44600,
+      "立即开始执行": 44601,
+      "▁stance": 44602,
+      "▁washed": 44603,
+      "▁Pere": 44604,
+      "(\\',": 44605,
+      "非常重要": 44606,
+      "▁antique": 44607,
+      "inescence": 44608,
+      "▁practiced": 44609,
+      "eryl": 44610,
+      "Alignment": 44611,
+      "▁residency": 44612,
+      "ENAB": 44613,
+      "ivier": 44614,
+      "▁Banach": 44615,
+      "▁streak": 44616,
+      "▁Sheffield": 44617,
+      "带走": 44618,
+      "电工": 44619,
+      "由被告负担": 44620,
+      "Transition": 44621,
+      "法的意义": 44622,
+      "letes": 44623,
+      "▁Wake": 44624,
+      "▁personalities": 44625,
+      "正方体的": 44626,
+      "▁reim": 44627,
+      "束缚": 44628,
+      "fork": 44629,
+      "rance": 44630,
+      "▁cozy": 44631,
+      "Licensed": 44632,
+      "nGallery": 44633,
+      "未婚": 44634,
+      "nbr": 44635,
+      "人死亡": 44636,
+      "保护区": 44637,
+      "ergus": 44638,
+      "means": 44639,
+      "▁weddings": 44640,
+      "▁Á": 44641,
+      "▁适用": 44642,
+      "ninline": 44643,
+      "▁directing": 44644,
+      "和大家": 44645,
+      "帅哥": 44646,
+      "前几天": 44647,
+      "▁yog": 44648,
+      "swing": 44649,
+      "▁typo": 44650,
+      "价值人民币": 44651,
+      "资产管理": 44652,
+      "▁Rein": 44653,
+      "▁entre": 44654,
+      "arag": 44655,
+      "monton": 44656,
+      "▁osteopor": 44657,
+      "{[": 44658,
+      "因果": 44659,
+      "▁hospitalized": 44660,
+      "整个人": 44661,
+      "asan": 44662,
+      "▁hollow": 44663,
+      "工程款": 44664,
+      "人均价格": 44665,
+      "▁pave": 44666,
+      "▁Mentor": 44667,
+      "▁CONT": 44668,
+      "▁Serum": 44669,
+      "▁Andrea": 44670,
+      "争论": 44671,
+      "庞大的": 44672,
+      "▁preceded": 44673,
+      "▁vicinity": 44674,
+      "▁SEM": 44675,
+      "▁punk": 44676,
+      "emaker": 44677,
+      "▁Gilbert": 44678,
+      "德华": 44679,
+      "的开发": 44680,
+      "[][]": 44681,
+      "Toast": 44682,
+      "▁cate": 44683,
+      "otoxin": 44684,
+      "▁annotations": 44685,
+      "YL": 44686,
+      "▁spider": 44687,
+      "一幅": 44688,
+      "悬浮": 44689,
+      "千万别": 44690,
+      "媒体报道": 44691,
+      "▁launches": 44692,
+      "▁formulations": 44693,
+      "杠杆": 44694,
+      "纺织品": 44695,
+      "nGame": 44696,
+      "▁Baronet": 44697,
+      "▁行政区": 44698,
+      "nTaxonomy": 44699,
+      "International": 44700,
+      "▁WI": 44701,
+      "avier": 44702,
+      "初次": 44703,
+      "酵母": 44704,
+      "都不知道": 44705,
+      "▁adap": 44706,
+      "▁reboot": 44707,
+      "电容": 44708,
+      "载荷": 44709,
+      "▁Gregory": 44710,
+      "谋划": 44711,
+      "和国际": 44712,
+      "Slice": 44713,
+      "▁analogous": 44714,
+      "▁contaminated": 44715,
+      "均在": 44716,
+      "▁handmade": 44717,
+      "天真": 44718,
+      "▁bore": 44719,
+      "olutely": 44720,
+      "▁Keyboard": 44721,
+      "树林": 44722,
+      "▁pilots": 44723,
+      "第一百三十三条之一第一款": 44724,
+      "Imm": 44725,
+      "▁allergy": 44726,
+      "▁colonies": 44727,
+      "实名": 44728,
+      "农业大学": 44729,
+      "药物治疗": 44730,
+      "ucing": 44731,
+      "Follow": 44732,
+      "折腾": 44733,
+      "Nil": 44734,
+      "▁kay": 44735,
+      "nhelp": 44736,
+      "SDL": 44737,
+      "▁SPDX": 44738,
+      "▁Honey": 44739,
+      "▁Fantasy": 44740,
+      "工商银行股份有限公司": 44741,
+      "representation": 44742,
+      "andel": 44743,
+      "▁rivals": 44744,
+      "商人": 44745,
+      "就跟": 44746,
+      "▁答案": 44747,
+      "▁Pedro": 44748,
+      "▁malicious": 44749,
+      "▁adopting": 44750,
+      "vl": 44751,
+      "而生": 44752,
+      "▁cessation": 44753,
+      "三轮": 44754,
+      "主页": 44755,
+      "第二百二十": 44756,
+      "▁edema": 44757,
+      "▁SIM": 44758,
+      "▁Chan": 44759,
+      "biased": 44760,
+      "nReturns": 44761,
+      "ASF": 44762,
+      "James": 44763,
+      "▁competitor": 44764,
+      "▁discoveries": 44765,
+      "▁植物界": 44766,
+      "needed": 44767,
+      "▁Clone": 44768,
+      "同日": 44769,
+      "某某于": 44770,
+      "Priority": 44771,
+      "性感": 44772,
+      "人体的": 44773,
+      "成长的": 44774,
+      "▁Blo": 44775,
+      "Entries": 44776,
+      "ogenetic": 44777,
+      "尘埃": 44778,
+      "▁Inj": 44779,
+      "▁yang": 44780,
+      "omyces": 44781,
+      "listing": 44782,
+      "OU": 44783,
+      "举动": 44784,
+      "cong": 44785,
+      "▁Inner": 44786,
+      "▁Scene": 44787,
+      "motor": 44788,
+      "purpose": 44789,
+      "大肠": 44790,
+      "的分": 44791,
+      "spons": 44792,
+      "▁learnt": 44793,
+      "▁erosion": 44794,
+      "▁Harper": 44795,
+      "▁enthusiasts": 44796,
+      "一楼": 44797,
+      "总有": 44798,
+      "ров": 44799,
+      "在进行": 44800,
+      "irie": 44801,
+      "▁cirrh": 44802,
+      "▁obstacle": 44803,
+      "▁auxiliary": 44804,
+      "▁Alternative": 44805,
+      "膀胱": 44806,
+      "路过": 44807,
+      "办理减刑": 44808,
+      "▁modulated": 44809,
+      "泄露": 44810,
+      "▁atten": 44811,
+      "答复": 44812,
+      "▁TeV": 44813,
+      "科学发展观": 44814,
+      "▁averaging": 44815,
+      ")$$,": 44816,
+      "▁sci": 44817,
+      "nEDIT": 44818,
+      "▁sacred": 44819,
+      "race": 44820,
+      "▁Hyd": 44821,
+      "▁sensible": 44822,
+      "oelect": 44823,
+      "▁VEGF": 44824,
+      "▁chees": 44825,
+      "▁Jump": 44826,
+      "移动互联网": 44827,
+      "▁Words": 44828,
+      "▁coded": 44829,
+      "▁themed": 44830,
+      "▁weakly": 44831,
+      "▁tracing": 44832,
+      "drug": 44833,
+      "▁Bend": 44834,
+      "Numbers": 44835,
+      "▁quarterly": 44836,
+      "▁precursors": 44837,
+      "再看": 44838,
+      "我为": 44839,
+      "▁Bald": 44840,
+      "signed": 44841,
+      "▁Includes": 44842,
+      "footballer": 44843,
+      "外套": 44844,
+      "度数": 44845,
+      "廉洁": 44846,
+      "boxed": 44847,
+      "asmine": 44848,
+      "nBeing": 44849,
+      "▁shells": 44850,
+      "classifiers": 44851,
+      "▁cot": 44852,
+      "▁digestion": 44853,
+      "▁rheumatoid": 44854,
+      "utz": 44855,
+      "slack": 44856,
+      "▁Amendment": 44857,
+      "▁kilometres": 44858,
+      "▁mentally": 44859,
+      "▁standalone": 44860,
+      "价为": 44861,
+      "贴吧": 44862,
+      "逆转": 44863,
+      "▁Chip": 44864,
+      "▁damn": 44865,
+      "Compare": 44866,
+      "干货": 44867,
+      "的值是": 44868,
+      "Patch": 44869,
+      "▁amel": 44870,
+      "▁Truck": 44871,
+      "他又": 44872,
+      "针对性的": 44873,
+      "▁customs": 44874,
+      "▁Kor": 44875,
+      "▁leaks": 44876,
+      "提供在线菜单": 44877,
+      "▁杰": 44878,
+      "还不是": 44879,
+      "师范学院": 44880,
+      "▁costumes": 44881,
+      "地段": 44882,
+      "适合自己的": 44883,
+      "▁Retrieved": 44884,
+      "▁gestation": 44885,
+      "满分": 44886,
+      "▁Indonesian": 44887,
+      "涂层": 44888,
+      "aside": 44889,
+      "▁penetr": 44890,
+      "▁hardwood": 44891,
+      "干脆": 44892,
+      "中国建设": 44893,
+      "▁stepping": 44894,
+      "▁immunological": 44895,
+      "也得": 44896,
+      "发展趋势": 44897,
+      "尚有": 44898,
+      "榜单": 44899,
+      "怎么看": 44900,
+      "HttpClient": 44901,
+      "大蒜": 44902,
+      "视听": 44903,
+      "▁穿越": 44904,
+      "ListNode": 44905,
+      "▁inception": 44906,
+      "▁梅": 44907,
+      "长度为": 44908,
+      "▁Fleet": 44909,
+      "▁eaten": 44910,
+      "双床": 44911,
+      "▁Triangle": 44912,
+      "▁therapeutics": 44913,
+      "是基于": 44914,
+      "▁Jur": 44915,
+      "▁Bullet": 44916,
+      "我才": 44917,
+      "▁RR": 44918,
+      "▁dynasty": 44919,
+      "▁Findings": 44920,
+      "▁Modeling": 44921,
+      "▁Panama": 44922,
+      "星空": 44923,
+      "要以": 44924,
+      "拥有的": 44925,
+      "阴性": 44926,
+      "▁solves": 44927,
+      "卵形": 44928,
+      "冠心病": 44929,
+      "armac": 44930,
+      "村村民委员会": 44931,
+      "▁cyan": 44932,
+      "▁enrichment": 44933,
+      "山大": 44934,
+      "胶原": 44935,
+      "▁特点": 44936,
+      "arxiv": 44937,
+      "Building": 44938,
+      "非常有": 44939,
+      "是这样的": 44940,
+      "进行调查": 44941,
+      "▁tetra": 44942,
+      "平整": 44943,
+      "▁内置": 44944,
+      "屋顶": 44945,
+      "nres": 44946,
+      "▁Hip": 44947,
+      "一审法院": 44948,
+      "owing": 44949,
+      "▁spans": 44950,
+      "displayName": 44951,
+      "和信息化": 44952,
+      "▁第一": 44953,
+      "哪一个": 44954,
+      "▁teenage": 44955,
+      "Variables": 44956,
+      "▁Antarctic": 44957,
+      "▁ted": 44958,
+      "本案按撤诉处理": 44959,
+      "▁spermat": 44960,
+      "出货": 44961,
+      "明清": 44962,
+      "以外的": 44963,
+      "▁rim": 44964,
+      "铜陵": 44965,
+      "ellen": 44966,
+      "▁swimmers": 44967,
+      "▁飞": 44968,
+      "在与": 44969,
+      "▁网站": 44970,
+      "新材料": 44971,
+      "▁tors": 44972,
+      "CHANGE": 44973,
+      "▁darker": 44974,
+      "nGovernment": 44975,
+      "▁Continuous": 44976,
+      "累了": 44977,
+      "umption": 44978,
+      "压抑": 44979,
+      "尚书": 44980,
+      "新世纪": 44981,
+      "umph": 44982,
+      "▁onCreate": 44983,
+      "雨水": 44984,
+      "▁Combined": 44985,
+      "一并": 44986,
+      "的运动": 44987,
+      "▁bend": 44988,
+      "▁intuition": 44989,
+      "去皮": 44990,
+      "原告要求被告": 44991,
+      "assetid": 44992,
+      "▁labelled": 44993,
+      "再一次": 44994,
+      "▁Grass": 44995,
+      "逐一": 44996,
+      "的形成": 44997,
+      "olecules": 44998,
+      "再次提出执行申请": 44999,
+      "ivot": 45000,
+      "▁Sue": 45001,
+      "ochem": 45002,
+      "KK": 45003,
+      "刘海": 45004,
+      "会很": 45005,
+      "一个大": 45006,
+      "还记得": 45007,
+      "▁Ore": 45008,
+      "ultan": 45009,
+      "优惠券": 45010,
+      "nTEST": 45011,
+      "▁fauna": 45012,
+      "restart": 45013,
+      "Interceptor": 45014,
+      "一所": 45015,
+      "吃苦": 45016,
+      "呵护": 45017,
+      "起始": 45018,
+      "rimp": 45019,
+      "nEnjoy": 45020,
+      "▁Barack": 45021,
+      "▁所属运动队": 45022,
+      "Permissions": 45023,
+      "▁ridiculous": 45024,
+      "亚军": 45025,
+      "▁umbre": 45026,
+      "报请": 45027,
+      "▁MBA": 45028,
+      "▁\\\\|": 45029,
+      "你会发现": 45030,
+      "放进": 45031,
+      "');": 45032,
+      "▁Bennett": 45033,
+      "诉前财产保全": 45034,
+      "▁oxidase": 45035,
+      "▁Translation": 45036,
+      "小孩子": 45037,
+      "的思考": 45038,
+      "▁Taken": 45039,
+      "nbefore": 45040,
+      ":\\',": 45041,
+      "REQUI": 45042,
+      "▁Diabetes": 45043,
+      "安康": 45044,
+      "▁);": 45045,
+      "到这里": 45046,
+      "azar": 45047,
+      "▁tamb": 45048,
+      "▁Essex": 45049,
+      "城中": 45050,
+      "处女": 45051,
+      "chnical": 45052,
+      "coholic": 45053,
+      "放过": 45054,
+      "▁harmony": 45055,
+      "▁kan": 45056,
+      "▁Creating": 45057,
+      "▁emphasized": 45058,
+      "▁毛": 45059,
+      "tops": 45060,
+      "▁\"\"\"": 45061,
+      "小型轿车": 45062,
+      "▁combo": 45063,
+      "歌声": 45064,
+      "的手机": 45065,
+      "omore": 45066,
+      "进一步加强": 45067,
+      "identified": 45068,
+      "▁HCC": 45069,
+      "▁WAY": 45070,
+      "▁disclosed": 45071,
+      "一刻": 45072,
+      "稀释": 45073,
+      "好朋友": 45074,
+      "nton": 45075,
+      "▁cuff": 45076,
+      "▁virulence": 45077,
+      "的第": 45078,
+      "酸性": 45079,
+      "aje": 45080,
+      "只剩下": 45081,
+      "相关性": 45082,
+      "泄漏": 45083,
+      "IVES": 45084,
+      "缅甸": 45085,
+      "dap": 45086,
+      "▁这": 45087,
+      "▁deduct": 45088,
+      "▁folded": 45089,
+      "导读": 45090,
+      "汽车销售": 45091,
+      "▁glam": 45092,
+      "olition": 45093,
+      "▁Confed": 45094,
+      "Related": 45095,
+      "也只是": 45096,
+      "apest": 45097,
+      "▁Dame": 45098,
+      "的办学": 45099,
+      "西红柿": 45100,
+      "rape": 45101,
+      "体制改革": 45102,
+      ")\\\\\\\\": 45103,
+      "一如": 45104,
+      "▁RN": 45105,
+      "梗死": 45106,
+      "没能": 45107,
+      "nAltern": 45108,
+      "nSingles": 45109,
+      "会给": 45110,
+      "▁Š": 45111,
+      "手持": 45112,
+      "▁Ri": 45113,
+      "一步步": 45114,
+      "▁holy": 45115,
+      "zk": 45116,
+      "就更": 45117,
+      "减刑后": 45118,
+      "应有的": 45119,
+      "Manag": 45120,
+      "unordered": 45121,
+      "appendChild": 45122,
+      "发生变化": 45123,
+      "▁thresholds": 45124,
+      "说你": 45125,
+      "▁Sussex": 45126,
+      "印尼": 45127,
+      "矿产": 45128,
+      "▁firef": 45129,
+      "validators": 45130,
+      "▁Gates": 45131,
+      "▁Ensure": 45132,
+      "老爷": 45133,
+      "展现了": 45134,
+      "平方差公式": 45135,
+      "▁teenagers": 45136,
+      "涌现": 45137,
+      "Mir": 45138,
+      "plants": 45139,
+      "jenkins": 45140,
+      "▁Reason": 45141,
+      "▁Unknown": 45142,
+      "▁checklist": 45143,
+      "可以看": 45144,
+      "▁GRE": 45145,
+      "▁mistaken": 45146,
+      "▁curiosity": 45147,
+      "alex": 45148,
+      "教学质量": 45149,
+      "▁thoracic": 45150,
+      "饼干": 45151,
+      "▁discord": 45152,
+      "带宽": 45153,
+      "摇滚": 45154,
+      "花期": 45155,
+      "配偶": 45156,
+      "可以的": 45157,
+      "▁cohomology": 45158,
+      "endswith": 45159,
+      "Way": 45160,
+      "ozygous": 45161,
+      "Urls": 45162,
+      "▁fres": 45163,
+      "▁Clarke": 45164,
+      "▁Criminal": 45165,
+      "三日内": 45166,
+      "ovol": 45167,
+      "号行政处罚决定": 45168,
+      "▁Gust": 45169,
+      "▁histone": 45170,
+      "classList": 45171,
+      "▁countable": 45172,
+      "▁capacities": 45173,
+      "经验和": 45174,
+      "证明了": 45175,
+      "▁ARM": 45176,
+      "▁misleading": 45177,
+      "九州": 45178,
+      "元二次方程": 45179,
+      "▁Guang": 45180,
+      "做完": 45181,
+      "爆料": 45182,
+      "bear": 45183,
+      "pause": 45184,
+      "▁webin": 45185,
+      "▁Libraries": 45186,
+      "▁ecc": 45187,
+      "最喜欢的": 45188,
+      "oubtedly": 45189,
+      "二〇一七年十一月": 45190,
+      "名录": 45191,
+      "▁Knights": 45192,
+      "▁perfection": 45193,
+      "散步": 45194,
+      "举办了": 45195,
+      "代理权限": 45196,
+      "▁Ashley": 45197,
+      "▁getName": 45198,
+      "▁trailing": 45199,
+      "外伤": 45200,
+      "eco": 45201,
+      "▁Kend": 45202,
+      "▁Chuck": 45203,
+      "▁fuels": 45204,
+      "▁eBay": 45205,
+      "▁supervisor": 45206,
+      "nBas": 45207,
+      "▁eine": 45208,
+      "给我的": 45209,
+      "▁ELISA": 45210,
+      "ymmetric": 45211,
+      "自私": 45212,
+      "紫外线": 45213,
+      "▁Tips": 45214,
+      "武昌": 45215,
+      "▁案": 45216,
+      "容积率": 45217,
+      "icides": 45218,
+      "之称": 45219,
+      "▁bs": 45220,
+      "▁bark": 45221,
+      "▁crashed": 45222,
+      "普洱": 45223,
+      "药材": 45224,
+      "淮安市": 45225,
+      "Days": 45226,
+      "▁antiviral": 45227,
+      "▁Cyprus": 45228,
+      "▁fighters": 45229,
+      "▁isoforms": 45230,
+      "Story": 45231,
+      "愿景": 45232,
+      "月饼": 45233,
+      "▁ao": 45234,
+      "已缴纳": 45235,
+      "▁neo": 45236,
+      "Lists": 45237,
+      "ewidth": 45238,
+      "皮革": 45239,
+      "oner": 45240,
+      "nSuper": 45241,
+      "▁Cauchy": 45242,
+      "▁glance": 45243,
+      "▁deviations": 45244,
+      "▁ND": 45245,
+      "oderm": 45246,
+      "▁Kashmir": 45247,
+      "▁securing": 45248,
+      "▁inheritance": 45249,
+      "绝对是": 45250,
+      "公诉机关指控被告人": 45251,
+      "▁Movies": 45252,
+      "低压": 45253,
+      "near": 45254,
+      "依法适用": 45255,
+      "▁adipose": 45256,
+      "预定": 45257,
+      "▁Politics": 45258,
+      "nOverall": 45259,
+      "▁Instance": 45260,
+      "▁Probably": 45261,
+      "▁directive": 45262,
+      "我之前": 45263,
+      "isma": 45264,
+      "▁combustion": 45265,
+      "判别": 45266,
+      "增殖": 45267,
+      "Maybe": 45268,
+      "▁Reddit": 45269,
+      "▁evoked": 45270,
+      "▁crossover": 45271,
+      "每组": 45272,
+      "沙滩": 45273,
+      "nMax": 45274,
+      "nMel": 45275,
+      "他认为": 45276,
+      "▁foil": 45277,
+      "▁prod": 45278,
+      "▁esophageal": 45279,
+      "黄油": 45280,
+      "我记得": 45281,
+      "▁译者": 45282,
+      "▁Requirements": 45283,
+      "山林": 45284,
+      "镇江": 45285,
+      "aires": 45286,
+      "巨人": 45287,
+      "而定": 45288,
+      "▁bp": 45289,
+      "我坚信": 45290,
+      "▁Uttar": 45291,
+      "传导": 45292,
+      "原著": 45293,
+      "直线与": 45294,
+      "▁TEXT": 45295,
+      "▁Replace": 45296,
+      "▁invites": 45297,
+      "企图": 45298,
+      "只在": 45299,
+      "▁Hours": 45300,
+      "上海浦东发展": 45301,
+      "▁Nottingham": 45302,
+      "直达": 45303,
+      "乳腺癌": 45304,
+      "也是一种": 45305,
+      "病原": 45306,
+      "▁knot": 45307,
+      "▁checkpoint": 45308,
+      "龙门": 45309,
+      "▁REG": 45310,
+      "完全没有": 45311,
+      "▁amendment": 45312,
+      "▁cong": 45313,
+      "▁loses": 45314,
+      "原神": 45315,
+      "OTA": 45316,
+      "▁RB": 45317,
+      "ifinals": 45318,
+      "▁Animation": 45319,
+      "×(": 45320,
+      "前缀": 45321,
+      "▁PROC": 45322,
+      "▁VO": 45323,
+      "▁Sac": 45324,
+      "Unable": 45325,
+      "smooth": 45326,
+      "源市": 45327,
+      "窃取": 45328,
+      "车祸": 45329,
+      "ICU": 45330,
+      "▁attenuation": 45331,
+      "oos": 45332,
+      "▁Hb": 45333,
+      "每年的": 45334,
+      "环境变量": 45335,
+      "irectional": 45336,
+      "▁discretion": 45337,
+      "▁С": 45338,
+      "安德": 45339,
+      "dater": 45340,
+      ";<": 45341,
+      "▁사": 45342,
+      "光芒": 45343,
+      "▁verse": 45344,
+      "▁Infantry": 45345,
+      "▁Variables": 45346,
+      "手套": 45347,
+      "挫折": 45348,
+      "通电": 45349,
+      "God": 45350,
+      "▁Og": 45351,
+      "▁Kra": 45352,
+      "oscope": 45353,
+      "催讨": 45354,
+      "转型升级": 45355,
+      "▁opposing": 45356,
+      "HM": 45357,
+      "▁depths": 45358,
+      "Joh": 45359,
+      "tsx": 45360,
+      "▁twins": 45361,
+      "▁firmware": 45362,
+      "阻塞": 45363,
+      "不予执行": 45364,
+      "▁intram": 45365,
+      "▁rou": 45366,
+      "▁Wool": 45367,
+      "Rock": 45368,
+      "值得一提": 45369,
+      "▁instructors": 45370,
+      "王爷": 45371,
+      "nFl": 45372,
+      "▁lemma": 45373,
+      "分校": 45374,
+      "开办": 45375,
+      "测绘": 45376,
+      "的单": 45377,
+      "helf": 45378,
+      "Paper": 45379,
+      "▁Chad": 45380,
+      "▁Nest": 45381,
+      "nAnn": 45382,
+      "Ubuntu": 45383,
+      "promise": 45384,
+      "kom": 45385,
+      "alan": 45386,
+      "icularly": 45387,
+      "▁deaf": 45388,
+      "▁glutathione": 45389,
+      "开出": 45390,
+      "▁Cay": 45391,
+      "▁Johannes": 45392,
+      "金融借款合同纠纷一案中": 45393,
+      "▁amin": 45394,
+      "▁Hardy": 45395,
+      "▁clues": 45396,
+      "▁sculptures": 45397,
+      "积的": 45398,
+      "是这样": 45399,
+      "ylinder": 45400,
+      "▁filming": 45401,
+      "▁Fabric": 45402,
+      "▁Secure": 45403,
+      "户数": 45404,
+      "萨克": 45405,
+      "fts": 45406,
+      "printed": 45407,
+      "终结本次执行程序后": 45408,
+      "▁continuum": 45409,
+      "遇到的": 45410,
+      "陌生人": 45411,
+      "▁}$\\": 45412,
+      "▁jQuery": 45413,
+      "▁lumbar": 45414,
+      "ecc": 45415,
+      "calls": 45416,
+      "外国人": 45417,
+      "▁Ble": 45418,
+      "爸爸妈妈": 45419,
+      "▁她": 45420,
+      "追溯": 45421,
+      "大力发展": 45422,
+      "xlabel": 45423,
+      "November": 45424,
+      "))**(": 45425,
+      "▁'';\\": 45426,
+      "▁assass": 45427,
+      "▁Gateway": 45428,
+      "▁技术": 45429,
+      "▁monot": 45430,
+      "▁Started": 45431,
+      "▁creators": 45432,
+      "▁commutative": 45433,
+      "孙某": 45434,
+      "无形": 45435,
+      "开启了": 45436,
+      "crop": 45437,
+      "不仅可以": 45438,
+      "基准利率": 45439,
+      "guild": 45440,
+      "比重": 45441,
+      "CURRENT": 45442,
+      "bounded": 45443,
+      "▁Generally": 45444,
+      "▁deliber": 45445,
+      "▁incumbent": 45446,
+      "天内": 45447,
+      "([[": 45448,
+      "和文化": 45449,
+      "请求依法": 45450,
+      "agged": 45451,
+      "的抗": 45452,
+      "不合理": 45453,
+      "▁ART": 45454,
+      "▁frustration": 45455,
+      "▁sheer": 45456,
+      "▁functor": 45457,
+      "试着": 45458,
+      "▁olf": 45459,
+      "奥林匹": 45460,
+      "的事业": 45461,
+      "移动电话": 45462,
+      "▁Maxwell": 45463,
+      "iones": 45464,
+      "▁Sage": 45465,
+      "atis": 45466,
+      "nInstall": 45467,
+      "管辖权的": 45468,
+      "▁wraps": 45469,
+      "▁secreted": 45470,
+      "▁fellowship": 45471,
+      "▁奥": 45472,
+      "弹出": 45473,
+      "开幕式": 45474,
+      "是一项": 45475,
+      "declar": 45476,
+      "▁indie": 45477,
+      "▁�": 45478,
+      "规程": 45479,
+      "▁cohes": 45480,
+      "▁performer": 45481,
+      "走廊": 45482,
+      "▁pigment": 45483,
+      "ttype": 45484,
+      "▁apologize": 45485,
+      "津贴": 45486,
+      "代表着": 45487,
+      "etype": 45488,
+      "假释案件具体应用法律": 45489,
+      "assertFalse": 45490,
+      "~~~~~~~~~~~~~~~~": 45491,
+      "一书": 45492,
+      "都比较": 45493,
+      "ceil": 45494,
+      "eted": 45495,
+      "thed": 45496,
+      "▁Pilot": 45497,
+      "▁Enable": 45498,
+      "事了": 45499,
+      "aviolet": 45500,
+      "▁Cinema": 45501,
+      "成份": 45502,
+      "▁RD": 45503,
+      "prec": 45504,
+      "▁happily": 45505,
+      "▁lenders": 45506,
+      "平安银行股份有限公司": 45507,
+      "晓东": 45508,
+      "路灯": 45509,
+      "购买的": 45510,
+      "nCor": 45511,
+      "Tables": 45512,
+      "ographers": 45513,
+      "▁cortisol": 45514,
+      "工业区": 45515,
+      "▁Sounds": 45516,
+      "▁usable": 45517,
+      "收益率": 45518,
+      "现实中": 45519,
+      "▁conventions": 45520,
+      "guest": 45521,
+      "▁Leonard": 45522,
+      "详解": 45523,
+      "▁surfact": 45524,
+      "rile": 45525,
+      "nCome": 45526,
+      "五花": 45527,
+      "MBA": 45528,
+      "保全费": 45529,
+      "兰州市": 45530,
+      "请登录": 45531,
+      "▁\\'\\',": 45532,
+      "▁adher": 45533,
+      "国家标准": 45534,
+      "哮喘": 45535,
+      "CNN": 45536,
+      "▁剧情": 45537,
+      "▁indul": 45538,
+      "closing": 45539,
+      "eleration": 45540,
+      "陵区": 45541,
+      "然后就": 45542,
+      "认定事实清楚": 45543,
+      "▁assemble": 45544,
+      "▁desperate": 45545,
+      "▁grandchildren": 45546,
+      "地毯": 45547,
+      "▁SIG": 45548,
+      "▁cav": 45549,
+      "founded": 45550,
+      "FU": 45551,
+      "hz": 45552,
+      "Pipe": 45553,
+      "ambia": 45554,
+      "▁platinum": 45555,
+      "人设": 45556,
+      "prem": 45557,
+      "▁Hir": 45558,
+      "变种": 45559,
+      "换货": 45560,
+      "Food": 45561,
+      "skin": 45562,
+      "▁aka": 45563,
+      "住广东省": 45564,
+      "ocyan": 45565,
+      "一二": 45566,
+      "致敬": 45567,
+      "▁建筑": 45568,
+      "和支持": 45569,
+      "IMARY": 45570,
+      "能有": 45571,
+      "gcc": 45572,
+      "总工会": 45573,
+      "▁Nike": 45574,
+      "▁Mixed": 45575,
+      "▁Identification": 45576,
+      "nContin": 45577,
+      "connections": 45578,
+      "源自": 45579,
+      "upe": 45580,
+      "了解了": 45581,
+      "received": 45582,
+      "Supported": 45583,
+      "▁iterator": 45584,
+      "Jun": 45585,
+      "两位数": 45586,
+      "▁redd": 45587,
+      "practice": 45588,
+      "能把": 45589,
+      "ISTS": 45590,
+      "▁cens": 45591,
+      "mozilla": 45592,
+      "想做": 45593,
+      "铺设": 45594,
+      "间距": 45595,
+      "Dan": 45596,
+      "只有在": 45597,
+      "▁言情小说": 45598,
+      "条第一款的规定": 45599,
+      "▁reviewer": 45600,
+      "不看": 45601,
+      "▁Dart": 45602,
+      "▁tasting": 45603,
+      "arum": 45604,
+      "主任医师": 45605,
+      "▁biod": 45606,
+      "▁pylint": 45607,
+      "▁accountability": 45608,
+      "澎湃": 45609,
+      "钢丝": 45610,
+      "▁casp": 45611,
+      "并按对方当事人的人数": 45612,
+      "孢子": 45613,
+      "局的": 45614,
+      "本判决生效后": 45615,
+      "火腿": 45616,
+      "见识": 45617,
+      "▁TP": 45618,
+      "▁stab": 45619,
+      "属于基础题": 45620,
+      "charset": 45621,
+      "长征": 45622,
+      "▁NU": 45623,
+      "hart": 45624,
+      "▁Biom": 45625,
+      "ademic": 45626,
+      "▁Hamburg": 45627,
+      "▁artifact": 45628,
+      "▁hardcore": 45629,
+      "官兵": 45630,
+      "戒指": 45631,
+      "以及对": 45632,
+      "nConn": 45633,
+      "▁naive": 45634,
+      "并无不当": 45635,
+      "retry": 45636,
+      "背部": 45637,
+      "交通便利": 45638,
+      "借款期限": 45639,
+      "容留他人": 45640,
+      "▁recurring": 45641,
+      "精湛": 45642,
+      "维奇": 45643,
+      "▁mont": 45644,
+      "▁钟": 45645,
+      "抵制": 45646,
+      "zzle": 45647,
+      "▁Detailed": 45648,
+      "深厚": 45649,
+      "▁ADHD": 45650,
+      "▁Pair": 45651,
+      "olecule": 45652,
+      "▁exempt": 45653,
+      "砂糖": 45654,
+      "没那么": 45655,
+      "tfrac": 45656,
+      "▁overlooking": 45657,
+      "万像素": 45658,
+      "excel": 45659,
+      "aterials": 45660,
+      "▁SSH": 45661,
+      "()){\\": 45662,
+      "▁expr": 45663,
+      "▁Strip": 45664,
+      "Digital": 45665,
+      "▁countryside": 45666,
+      "NONE": 45667,
+      "▁Whats": 45668,
+      "LDL": 45669,
+      "精彩的": 45670,
+      "行政村": 45671,
+      "中日": 45672,
+      "cia": 45673,
+      "▁Liv": 45674,
+      "▁shooter": 45675,
+      "▁interrupt": 45676,
+      "右下": 45677,
+      "AME": 45678,
+      "看不见": 45679,
+      "yers": 45680,
+      "▁Robot": 45681,
+      "▁recovering": 45682,
+      "▁把": 45683,
+      "你是否": 45684,
+      "tti": 45685,
+      "nCharles": 45686,
+      "▁convinc": 45687,
+      "nObjective": 45688,
+      "▁关于": 45689,
+      "不成立": 45690,
+      "的网络小说": 45691,
+      "noutput": 45692,
+      "▁trache": 45693,
+      "成正": 45694,
+      "▁Fest": 45695,
+      "Checked": 45696,
+      "▁arrhythm": 45697,
+      "将该": 45698,
+      "▁Foo": 45699,
+      "ughty": 45700,
+      "本院依职权": 45701,
+      "amming": 45702,
+      "金龙": 45703,
+      "集聚": 45704,
+      "wyn": 45705,
+      "▁}}\">\\": 45706,
+      "utility": 45707,
+      "▁cytometry": 45708,
+      "蛋黄": 45709,
+      "最好是": 45710,
+      "来实现": 45711,
+      "▁lime": 45712,
+      "ernels": 45713,
+      "openia": 45714,
+      "▁incremental": 45715,
+      "ön": 45716,
+      "爱我": 45717,
+      "▁Lemma": 45718,
+      "▁toxins": 45719,
+      "办案": 45720,
+      "jdbc": 45721,
+      "ermany": 45722,
+      "▁localhost": 45723,
+      "外星": 45724,
+      "nHot": 45725,
+      "▁dissociation": 45726,
+      "HOT": 45727,
+      "▁inconven": 45728,
+      "{{'": 45729,
+      "寝室": 45730,
+      "收敛": 45731,
+      "艺术的": 45732,
+      "beforeEach": 45733,
+      "▁Henderson": 45734,
+      "▁shareholders": 45735,
+      "▁socioeconomic": 45736,
+      "▁electrochemical": 45737,
+      "用自己的": 45738,
+      "重度": 45739,
+      "▁电源": 45740,
+      "▁merit": 45741,
+      "kB": 45742,
+      "反诉": 45743,
+      "节的": 45744,
+      "香油": 45745,
+      "Tour": 45746,
+      "所以我很": 45747,
+      "▁reconstructed": 45748,
+      "香水": 45749,
+      "方法是": 45750,
+      "UIKit": 45751,
+      "▁CMOS": 45752,
+      "elected": 45753,
+      "▁δ": 45754,
+      "对象的": 45755,
+      "▁lent": 45756,
+      "▁moderately": 45757,
+      "三层": 45758,
+      "腰椎": 45759,
+      "▁GS": 45760,
+      "医疗保险": 45761,
+      "研究表明": 45762,
+      "▁neglected": 45763,
+      "革新": 45764,
+      "中国特色社会主义": 45765,
+      "并且在": 45766,
+      "Completed": 45767,
+      "▁Depression": 45768,
+      "知晓": 45769,
+      "外国语": 45770,
+      "tion": 45771,
+      "ysql": 45772,
+      "证据证明": 45773,
+      "▁nevertheless": 45774,
+      "urnal": 45775,
+      "▁Henri": 45776,
+      "汇集": 45777,
+      "苏宁": 45778,
+      "麻将": 45779,
+      "/**/*": 45780,
+      "nrun": 45781,
+      "研究人员": 45782,
+      "▁bookmark": 45783,
+      "▁initialized": 45784,
+      "GY": 45785,
+      "LEFT": 45786,
+      "▁diplomatic": 45787,
+      "干了": 45788,
+      "潜水": 45789,
+      "lis": 45790,
+      "trade": 45791,
+      "▁caut": 45792,
+      "nodejs": 45793,
+      "▁refurb": 45794,
+      "的压力": 45795,
+      "Ratio": 45796,
+      "flare": 45797,
+      "工程技术": 45798,
+      "职业技能": 45799,
+      "▁小说进度": 45800,
+      "▁adhere": 45801,
+      "▁Herbert": 45802,
+      "vw": 45803,
+      "晚报": 45804,
+      "{})": 45805,
+      "ubation": 45806,
+      "第二百四十三条": 45807,
+      "▁malignancies": 45808,
+      "手臂": 45809,
+      "\"])": 45810,
+      "▁liner": 45811,
+      "OPTIONS": 45812,
+      "矿物质": 45813,
+      "用脚": 45814,
+      "(\"{": 45815,
+      "sgi": 45816,
+      "nevery": 45817,
+      "▁towers": 45818,
+      "Inventory": 45819,
+      "耽误": 45820,
+      "快乐的": 45821,
+      "▁Tum": 45822,
+      "▁WR": 45823,
+      "市场监管": 45824,
+      "nChris": 45825,
+      "▁Basically": 45826,
+      "强奸": 45827,
+      "抽烟": 45828,
+      "山县人民法院": 45829,
+      "▁zbMATH": 45830,
+      "uca": 45831,
+      "▁其五": 45832,
+      "ertz": 45833,
+      "▁Morocco": 45834,
+      "PPT": 45835,
+      "届人民代表大会": 45836,
+      "硅胶": 45837,
+      "▁面积": 45838,
+      "▁citation": 45839,
+      "ITHOUT": 45840,
+      "▁differing": 45841,
+      "ahr": 45842,
+      "▁MHC": 45843,
+      "▁Cable": 45844,
+      "▁Wallace": 45845,
+      "▁clearer": 45846,
+      "▁progesterone": 45847,
+      "职员表": 45848,
+      "▁Jin": 45849,
+      "ensit": 45850,
+      "为证": 45851,
+      "格里": 45852,
+      "▁cage": 45853,
+      "locfileid": 45854,
+      "▁Regarding": 45855,
+      "一举": 45856,
+      ")}}": 45857,
+      "gran": 45858,
+      "镇人民政府": 45859,
+      "▁Areas": 45860,
+      "群里": 45861,
+      "肩膀": 45862,
+      "申请执行人亦": 45863,
+      "好在": 45864,
+      "配电": 45865,
+      "∴\\\\": 45866,
+      "▁UUID": 45867,
+      "▁boxing": 45868,
+      "全员": 45869,
+      "问他": 45870,
+      "FIL": 45871,
+      "▁eosin": 45872,
+      "▁pivotal": 45873,
+      "▁Concrete": 45874,
+      "作案": 45875,
+      "▁weed": 45876,
+      "县人民检察院以": 45877,
+      "attering": 45878,
+      "结案若干问题的意见": 45879,
+      "XC": 45880,
+      "▁histories": 45881,
+      "上世纪": 45882,
+      "当地的": 45883,
+      "曾经的": 45884,
+      "▁Dell": 45885,
+      "schemas": 45886,
+      "清凉": 45887,
+      "金水": 45888,
+      "台区": 45889,
+      "▁flew": 45890,
+      "▁ny": 45891,
+      "等症状": 45892,
+      "讨论了": 45893,
+      "tswitch": 45894,
+      "▁Chicken": 45895,
+      "▁Hait": 45896,
+      "▁Genetic": 45897,
+      "▁orchestra": 45898,
+      "IDD": 45899,
+      "说明了": 45900,
+      "▁Nutrition": 45901,
+      "更新的": 45902,
+      "▁主要作品": 45903,
+      "nassert": 45904,
+      "设想": 45905,
+      "aea": 45906,
+      "iya": 45907,
+      "分析师": 45908,
+      "全国联保": 45909,
+      "现在已经": 45910,
+      "▁餐馆类型": 45911,
+      "▁faults": 45912,
+      "}<": 45913,
+      "带你": 45914,
+      "胃肠": 45915,
+      "农民人均纯收入": 45916,
+      "▁Agricultural": 45917,
+      "矿石": 45918,
+      "不正确": 45919,
+      "水平和": 45920,
+      "yses": 45921,
+      "▁Nathan": 45922,
+      "▁hazardous": 45923,
+      "回执": 45924,
+      "有小": 45925,
+      "首饰": 45926,
+      "预交上诉案件受理费": 45927,
+      "▁ст": 45928,
+      "Tick": 45929,
+      "nENV": 45930,
+      "nNever": 45931,
+      "restore": 45932,
+      "▁Louise": 45933,
+      "▁pollut": 45934,
+      "ievement": 45935,
+      "活血": 45936,
+      "起着": 45937,
+      "esch": 45938,
+      "nplot": 45939,
+      "▁illeg": 45940,
+      "地在": 45941,
+      "发展规划": 45942,
+      "atype": 45943,
+      "▁grafts": 45944,
+      "Typed": 45945,
+      "▁Urugu": 45946,
+      "▁euros": 45947,
+      "▁Factors": 45948,
+      "共同努力": 45949,
+      "rians": 45950,
+      "ylate": 45951,
+      "▁Krish": 45952,
+      "▁composites": 45953,
+      "与我": 45954,
+      "增强了": 45955,
+      "无机": 45956,
+      "致辞": 45957,
+      "ASP": 45958,
+      "二进制": 45959,
+      "/\\',": 45960,
+      "丰富多彩": 45961,
+      "▁cohorts": 45962,
+      "documents": 45963,
+      "▁celebrations": 45964,
+      "▁reagents": 45965,
+      "恶化": 45966,
+      "还行": 45967,
+      "aaa": 45968,
+      "增长率": 45969,
+      "fest": 45970,
+      "ridor": 45971,
+      "▁blonde": 45972,
+      "社科": 45973,
+      "留下的": 45974,
+      "▁Outlook": 45975,
+      "巴士": 45976,
+      "教育和": 45977,
+      "Viewer": 45978,
+      "nRecent": 45979,
+      "▁faithful": 45980,
+      "▁summarize": 45981,
+      "Vo": 45982,
+      "▁수": 45983,
+      "不完": 45984,
+      "勾选": 45985,
+      "▁人均价格": 45986,
+      "nChinese": 45987,
+      "帷幕": 45988,
+      "nMart": 45989,
+      "roots": 45990,
+      "organisms": 45991,
+      "伪造": 45992,
+      "向阳": 45993,
+      "eeper": 45994,
+      "▁Yugoslav": 45995,
+      "▁headline": 45996,
+      "妇联": 45997,
+      "Roles": 45998,
+      "nNice": 45999,
+      "attack": 46000,
+      "filePath": 46001,
+      "▁Treasury": 46002,
+      "▁realised": 46003,
+      "▁Rochester": 46004,
+      "▁detailing": 46005,
+      "▁老": 46006,
+      "又被": 46007,
+      "字的": 46008,
+      "自考": 46009,
+      "▁multiplex": 46010,
+      "▁汪": 46011,
+      "一说": 46012,
+      "议题": 46013,
+      "青云": 46014,
+      "Bob": 46015,
+      "GMP": 46016,
+      "法定代表人刘": 46017,
+      "▁canine": 46018,
+      "▁implanted": 46019,
+      "leston": 46020,
+      "新娘": 46021,
+      "▁§": 46022,
+      "半径为": 46023,
+      "建立起": 46024,
+      "▁polish": 46025,
+      "Connected": 46026,
+      "党风": 46027,
+      "反向": 46028,
+      "okemon": 46029,
+      "▁Geneva": 46030,
+      "▁recalled": 46031,
+      "少年儿童": 46032,
+      "▁Rating": 46033,
+      "冰糖": 46034,
+      "/__": 46035,
+      "Bag": 46036,
+      "审定委员会": 46037,
+      "▁tragedy": 46038,
+      "]$$": 46039,
+      "但是他": 46040,
+      ")],": 46041,
+      "▁Atlas": 46042,
+      "▁Tibet": 46043,
+      "落到": 46044,
+      "雇佣": 46045,
+      "▁NV": 46046,
+      "▁bees": 46047,
+      "专业委员会": 46048,
+      "amiliar": 46049,
+      "永不": 46050,
+      "▁Warsaw": 46051,
+      "nmetadata": 46052,
+      "▁reliably": 46053,
+      "▁WHAT": 46054,
+      "patterns": 46055,
+      "位是": 46056,
+      "式中": 46057,
+      "立即发生法律效力": 46058,
+      "languages": 46059,
+      "Transformer": 46060,
+      "▁[#": 46061,
+      "▁analyzes": 46062,
+      "▁mismatch": 46063,
+      "▁Shape": 46064,
+      "▁spacecraft": 46065,
+      "英格兰": 46066,
+      "原告主张": 46067,
+      "▁jets": 46068,
+      "▁Plastic": 46069,
+      "▁pancreas": 46070,
+      "乔治": 46071,
+      "峡谷": 46072,
+      "纸质": 46073,
+      "▁awful": 46074,
+      "交代": 46075,
+      "……”": 46076,
+      "▁楼盘": 46077,
+      "thag": 46078,
+      "bolds": 46079,
+      "nLooks": 46080,
+      "▁encephal": 46081,
+      "为首": 46082,
+      "乐团": 46083,
+      "遇上": 46084,
+      "不能成立": 46085,
+      "orman": 46086,
+      "nEvaluate": 46087,
+      "黄山": 46088,
+      "是一样的": 46089,
+      "nIntroduction": 46090,
+      "▁pare": 46091,
+      "混沌": 46092,
+      "lican": 46093,
+      "Wallet": 46094,
+      "▁\\'\\',\\": 46095,
+      "aternion": 46096,
+      "▁CONTRIBUTORS": 46097,
+      "Codes": 46098,
+      "▁Andrews": 46099,
+      "▁Bradley": 46100,
+      "▁repeating": 46101,
+      "三明": 46102,
+      "nTur": 46103,
+      "ynec": 46104,
+      "▁uniqueness": 46105,
+      "劳动关系": 46106,
+      "交纳案件受理费": 46107,
+      "▁Execute": 46108,
+      "▁renovated": 46109,
+      "光照": 46110,
+      "购置": 46111,
+      "nPrim": 46112,
+      "▁Hide": 46113,
+      "▁specializing": 46114,
+      "▁adenocarcinoma": 46115,
+      "imento": 46116,
+      "对付": 46117,
+      "洛克": 46118,
+      "▁史": 46119,
+      "互通": 46120,
+      "探险": 46121,
+      "比利时": 46122,
+      "tour": 46123,
+      "▁kills": 46124,
+      "reviewed": 46125,
+      "学姐": 46126,
+      "▁Beer": 46127,
+      "珠海市": 46128,
+      "▁Cho": 46129,
+      "▁Bone": 46130,
+      "cribing": 46131,
+      "▁Marathon": 46132,
+      "旺盛": 46133,
+      "讽刺": 46134,
+      "Invest": 46135,
+      "▁Indoor": 46136,
+      "▁П": 46137,
+      "潇洒": 46138,
+      "▁[…": 46139,
+      "▁Cum": 46140,
+      "anson": 46141,
+      "▁cloning": 46142,
+      "▁motions": 46143,
+      "▁presenter": 46144,
+      "▁Partnership": 46145,
+      "▁Indianapolis": 46146,
+      "是一些": 46147,
+      "nCam": 46148,
+      "▁parab": 46149,
+      "Checker": 46150,
+      "不肯": 46151,
+      "其后": 46152,
+      "微量元素": 46153,
+      "▁lobe": 46154,
+      "▁maths": 46155,
+      "构筑": 46156,
+      "▁candle": 46157,
+      "▁可以": 46158,
+      "nsupported": 46159,
+      "点是": 46160,
+      "还想": 46161,
+      "▁chol": 46162,
+      "▁remarkably": 46163,
+      "目前已": 46164,
+      "虽然是": 46165,
+      "▁[])\\": 46166,
+      "▁odor": 46167,
+      ",$$": 46168,
+      "Ber": 46169,
+      "▁exponents": 46170,
+      "手机的": 46171,
+      "▁UFC": 46172,
+      "▁boom": 46173,
+      "endforeach": 46174,
+      "不全": 46175,
+      "▁freestyle": 46176,
+      "大发": 46177,
+      "桥区": 46178,
+      "发生后": 46179,
+      "▁TRA": 46180,
+      "▁Gross": 46181,
+      "▁elast": 46182,
+      "一门": 46183,
+      "大叔": 46184,
+      "环境和": 46185,
+      "介绍一下": 46186,
+      "对其进行": 46187,
+      "charAt": 46188,
+      "打游戏": 46189,
+      "房屋租赁": 46190,
+      "搜狐": 46191,
+      "的证言": 46192,
+      "Keyboard": 46193,
+      "减排": 46194,
+      "较差": 46195,
+      "每天都": 46196,
+      "oris": 46197,
+      "计算能力": 46198,
+      "msgs": 46199,
+      "▁Cock": 46200,
+      "不等式的解": 46201,
+      "▁sourced": 46202,
+      "而去": 46203,
+      "命题的": 46204,
+      "学会了": 46205,
+      "▁reef": 46206,
+      "xff": 46207,
+      "▁tumours": 46208,
+      "mbH": 46209,
+      "tek": 46210,
+      "图形的": 46211,
+      "▁coloured": 46212,
+      "好几": 46213,
+      "是一所": 46214,
+      "▁acrylic": 46215,
+      "物业服务合同纠纷一案中": 46216,
+      "重力": 46217,
+      "Peter": 46218,
+      "▁towel": 46219,
+      "▁martial": 46220,
+      "将为": 46221,
+      "oyl": 46222,
+      "等离子": 46223,
+      "▁Sett": 46224,
+      "▁Sandy": 46225,
+      "强国": 46226,
+      "海燕": 46227,
+      "颈部": 46228,
+      ")\\'": 46229,
+      "acqu": 46230,
+      "din": 46231,
+      "nPurpose": 46232,
+      "▁Parallel": 46233,
+      "共用": 46234,
+      "一种是": 46235,
+      "不用担心": 46236,
+      "▁cares": 46237,
+      "▁Belt": 46238,
+      "▁oversight": 46239,
+      "头晕": 46240,
+      "清香": 46241,
+      "纵坐标": 46242,
+      "风景区": 46243,
+      "nStat": 46244,
+      "cultural": 46245,
+      "派的": 46246,
+      "ninter": 46247,
+      "▁photons": 46248,
+      "工商行政管理局登记成立": 46249,
+      "Tokens": 46250,
+      "Performance": 46251,
+      "以致": 46252,
+      "必须是": 46253,
+      "imedia": 46254,
+      "▁hatch": 46255,
+      "▁Stress": 46256,
+      "丝毫": 46257,
+      "两张": 46258,
+      "▁Porter": 46259,
+      "师大": 46260,
+      "cible": 46261,
+      "▁menus": 46262,
+      "▁differentially": 46263,
+      "提及": 46264,
+      "▁成语": 46265,
+      "线城市": 46266,
+      "▁Joel": 46267,
+      "▁IEnumerable": 46268,
+      "devDependencies": 46269,
+      "入境": 46270,
+      "山庄": 46271,
+      "▁Marco": 46272,
+      "othermal": 46273,
+      "号执行": 46274,
+      "▁HEAD": 46275,
+      "Daniel": 46276,
+      "▁Manitoba": 46277,
+      "▁Sullivan": 46278,
+      "会把": 46279,
+      "理科": 46280,
+      "心理咨询": 46281,
+      "法定代表人陈": 46282,
+      "▁successes": 46283,
+      "▁登场": 46284,
+      "项目中": 46285,
+      "ullah": 46286,
+      "房贷": 46287,
+      "虽说": 46288,
+      "会觉得": 46289,
+      "▁youtube": 46290,
+      "NON": 46291,
+      "▁Simpson": 46292,
+      "▁normalization": 46293,
+      "加分": 46294,
+      "因家庭": 46295,
+      "方式的": 46296,
+      "antha": 46297,
+      "nReal": 46298,
+      "▁Hoff": 46299,
+      "▁Abbey": 46300,
+      "▁dolor": 46301,
+      "▁hedge": 46302,
+      "Contracts": 46303,
+      "ocellular": 46304,
+      "▁councill": 46305,
+      "nSir": 46306,
+      "▁CAT": 46307,
+      "▁familial": 46308,
+      "boldsymbol": 46309,
+      "它可以": 46310,
+      "▁北京市": 46311,
+      "而来的": 46312,
+      "constraints": 46313,
+      "▁Fellowship": 46314,
+      "技术与": 46315,
+      "炸弹": 46316,
+      "不得不说": 46317,
+      "中国银行": 46318,
+      "commercial": 46319,
+      "▁interacts": 46320,
+      "一到": 46321,
+      "马云": 46322,
+      "acid": 46323,
+      "晋级": 46324,
+      "▁lifelong": 46325,
+      "▁beverages": 46326,
+      "安装的": 46327,
+      "amines": 46328,
+      "▁judged": 46329,
+      "中了": 46330,
+      "▁Whole": 46331,
+      "▁velocities": 46332,
+      "BooleanField": 46333,
+      "吸食": 46334,
+      "多重": 46335,
+      "靠着": 46336,
+      "这几年": 46337,
+      "▁Natal": 46338,
+      "▁Mé": 46339,
+      "elize": 46340,
+      "总建筑面积": 46341,
+      "nMethod": 46342,
+      "▁altering": 46343,
+      "▁unavailable": 46344,
+      "甜甜": 46345,
+      "▁syll": 46346,
+      "党支部书记": 46347,
+      "ophagy": 46348,
+      "▁Warning": 46349,
+      "▁spreadsheet": 46350,
+      "▁collectively": 46351,
+      "mw": 46352,
+      "很是": 46353,
+      "nPan": 46354,
+      "▁Mang": 46355,
+      "specified": 46356,
+      "nag": 46357,
+      "▁Might": 46358,
+      "▁∼": 46359,
+      "尔特": 46360,
+      "更具": 46361,
+      "Figure": 46362,
+      "晚餐": 46363,
+      "骗子": 46364,
+      "中国古代": 46365,
+      "affle": 46366,
+      "stuff": 46367,
+      "▁balcony": 46368,
+      "标配": 46369,
+      "▁免费": 46370,
+      "泡泡": 46371,
+      "看他": 46372,
+      "中关村在线": 46373,
+      "Scheme": 46374,
+      "妥协": 46375,
+      "转到": 46376,
+      "未约定": 46377,
+      "▁Superior": 46378,
+      "▁Redistributions": 46379,
+      "加水": 46380,
+      "懒得": 46381,
+      "比你": 46382,
+      "chef": 46383,
+      "▁spaw": 46384,
+      "幼稚": 46385,
+      "Sets": 46386,
+      "▁scent": 46387,
+      "▁insured": 46388,
+      "海绵": 46389,
+      "hra": 46390,
+      "旗人民法院": 46391,
+      "▁sovere": 46392,
+      "Ta": 46393,
+      "人有": 46394,
+      "贴心": 46395,
+      "履行了": 46396,
+      "不是一个": 46397,
+      "得益": 46398,
+      "自媒体": 46399,
+      "usan": 46400,
+      "nMath": 46401,
+      "▁Battery": 46402,
+      "Fat": 46403,
+      "▁Paint": 46404,
+      "▁nginx": 46405,
+      "▁breathe": 46406,
+      "鱼的": 46407,
+      "oak": 46408,
+      "来到了一": 46409,
+      "以上事实有": 46410,
+      "ienced": 46411,
+      "logged": 46412,
+      "安防": 46413,
+      "提案": 46414,
+      "充分的": 46415,
+      "wings": 46416,
+      "ampagne": 46417,
+      "tdefault": 46418,
+      "registered": 46419,
+      "存活": 46420,
+      "艾滋": 46421,
+      "万元及利息": 46422,
+      "▁Actual": 46423,
+      "▁casinos": 46424,
+      "可以自收到裁定书": 46425,
+      "▁witnesses": 46426,
+      "栖息": 46427,
+      "former": 46428,
+      "Creation": 46429,
+      "▁exposing": 46430,
+      "▁Bulgarian": 46431,
+      "病虫": 46432,
+      "ucumber": 46433,
+      "▁Expect": 46434,
+      "zA": 46435,
+      "Separ": 46436,
+      "rg": 46437,
+      "再遭": 46438,
+      "梯形": 46439,
+      "长辈": 46440,
+      "▁sich": 46441,
+      "▁sulph": 46442,
+      "劳动人事": 46443,
+      "浦东新区": 46444,
+      "▁dancers": 46445,
+      "▁wrapping": 46446,
+      "▁dissection": 46447,
+      "世代": 46448,
+      "增收": 46449,
+      "红利": 46450,
+      "Smith": 46451,
+      "刑满": 46452,
+      "▁sore": 46453,
+      "yk": 46454,
+      "统一的": 46455,
+      "▁silica": 46456,
+      "▁EE": 46457,
+      "四条的规定": 46458,
+      "onomical": 46459,
+      "▁Frequency": 46460,
+      "视网": 46461,
+      "male": 46462,
+      "轨道交通": 46463,
+      "▁catalogue": 46464,
+      "汇款": 46465,
+      "▁MLS": 46466,
+      "Chapter": 46467,
+      "▁Einstein": 46468,
+      "{|": 46469,
+      "写法": 46470,
+      "困难的": 46471,
+      "▁premise": 46472,
+      "▁Istanbul": 46473,
+      "▁immunization": 46474,
+      "出炉": 46475,
+      "每当": 46476,
+      "这个是": 46477,
+      "▁Kazakh": 46478,
+      "▁thrust": 46479,
+      "灯具": 46480,
+      "加之": 46481,
+      "平滑": 46482,
+      "血浆": 46483,
+      "▁crore": 46484,
+      "魔兽": 46485,
+      "工作日": 46486,
+      "ovsky": 46487,
+      "适时": 46488,
+      "Quality": 46489,
+      "▁Bootstrap": 46490,
+      "Org": 46491,
+      "有多大": 46492,
+      "▁leap": 46493,
+      "qp": 46494,
+      "几年来": 46495,
+      "▁\\'*": 46496,
+      "▁masters": 46497,
+      "▁logarithm": 46498,
+      "开支": 46499,
+      "▁Craw": 46500,
+      "idepress": 46501,
+      "↓↓": 46502,
+      "▁Cherry": 46503,
+      "如果要": 46504,
+      "护肤品": 46505,
+      "管理法": 46506,
+      "▁locking": 46507,
+      "掀起": 46508,
+      "IEEE": 46509,
+      "▁cystic": 46510,
+      "应准予": 46511,
+      "▁caller": 46512,
+      "电量": 46513,
+      "▁ginger": 46514,
+      "▁ceremon": 46515,
+      "▁negotiate": 46516,
+      "▁programmed": 46517,
+      "源县": 46518,
+      "准备工作": 46519,
+      "尤其是在": 46520,
+      "saving": 46521,
+      "▁Stats": 46522,
+      "▁superficial": 46523,
+      "不利于": 46524,
+      "▁microg": 46525,
+      "所用": 46526,
+      "▁phi": 46527,
+      "▁Walt": 46528,
+      "们都": 46529,
+      "大爷": 46530,
+      "犹太": 46531,
+      "cycles": 46532,
+      "rientation": 46533,
+      "▁classrooms": 46534,
+      "▁dwell": 46535,
+      "Far": 46536,
+      "对这个": 46537,
+      "redients": 46538,
+      "▁cognition": 46539,
+      "的身份": 46540,
+      "amen": 46541,
+      "主教": 46542,
+      "之余": 46543,
+      "总的来说": 46544,
+      "经济管理": 46545,
+      "\\\\{\\\\": 46546,
+      "ayette": 46547,
+      "fixtures": 46548,
+      "atha": 46549,
+      "uw": 46550,
+      "]$\\": 46551,
+      "nsim": 46552,
+      "▁lute": 46553,
+      "▁portraits": 46554,
+      "communication": 46555,
+      ",\\\\,": 46556,
+      "nlong": 46557,
+      "油菜": 46558,
+      "aude": 46559,
+      "yahoo": 46560,
+      "二〇一七年九月": 46561,
+      "▁Edmonton": 46562,
+      "▁Arguments": 46563,
+      "不绝": 46564,
+      "SRC": 46565,
+      "国家和地区": 46566,
+      "ressing": 46567,
+      "▁rebuilt": 46568,
+      "▁finitely": 46569,
+      "▁locomotives": 46570,
+      "切入": 46571,
+      "那位": 46572,
+      "▁DL": 46573,
+      "▁角色": 46574,
+      "是否存在": 46575,
+      "ttest": 46576,
+      "leader": 46577,
+      "increment": 46578,
+      "▁disclose": 46579,
+      "哥伦": 46580,
+      "秒钟": 46581,
+      "微商": 46582,
+      "打着": 46583,
+      "snap": 46584,
+      "▁Hugh": 46585,
+      "▁musc": 46586,
+      "▁shiny": 46587,
+      "▁Waters": 46588,
+      "threatening": 46589,
+      "▁willingness": 46590,
+      "furt": 46591,
+      "Mongo": 46592,
+      "quito": 46593,
+      "▁Constructor": 46594,
+      "▁characterizing": 46595,
+      "uche": 46596,
+      "婚姻关系": 46597,
+      "▁homologous": 46598,
+      "草地": 46599,
+      "对企业": 46600,
+      "横坐标": 46601,
+      "▁potency": 46602,
+      "▁decentral": 46603,
+      "▁researching": 46604,
+      "▁Abr": 46605,
+      "▁UTF": 46606,
+      "▁UEFA": 46607,
+      "▁Atomic": 46608,
+      "▁reside": 46609,
+      "▁backdrop": 46610,
+      "▁云": 46611,
+      "▁reun": 46612,
+      "▁accus": 46613,
+      "▁inbox": 46614,
+      "▁concomitant": 46615,
+      "▁intermittent": 46616,
+      "▁iff": 46617,
+      "raises": 46618,
+      "▁EA": 46619,
+      "enchymal": 46620,
+      "▁Alcohol": 46621,
+      "～～": 46622,
+      "叙利亚": 46623,
+      "▁compiling": 46624,
+      "往后": 46625,
+      "ricia": 46626,
+      "不受申请执行时效": 46627,
+      "nAdditional": 46628,
+      "localStorage": 46629,
+      "遗忘": 46630,
+      "etz": 46631,
+      "▁山东": 46632,
+      "▁MATLAB": 46633,
+      "▁laparoscopic": 46634,
+      "审慎": 46635,
+      "uthors": 46636,
+      "texcept": 46637,
+      "▁attent": 46638,
+      "下达": 46639,
+      "四处": 46640,
+      "opp": 46641,
+      "受不了": 46642,
+      "Frank": 46643,
+      "▁Tube": 46644,
+      "ynaptic": 46645,
+      "▁Powell": 46646,
+      "fontSize": 46647,
+      "▁applicability": 46648,
+      "Au": 46649,
+      "盘中": 46650,
+      "结实": 46651,
+      "▁ABO": 46652,
+      "▁Junction": 46653,
+      "人社": 46654,
+      "催收": 46655,
+      "隐瞒": 46656,
+      "直升": 46657,
+      "▁fried": 46658,
+      "是想": 46659,
+      "齐齐": 46660,
+      "UNK": 46661,
+      "olation": 46662,
+      "▁Artificial": 46663,
+      "▁Specialist": 46664,
+      "▁disputes": 46665,
+      "按中国人民银行同期": 46666,
+      "口水": 46667,
+      "打个": 46668,
+      "大幅度": 46669,
+      "▁mmHg": 46670,
+      "▁bulbs": 46671,
+      "upytext": 46672,
+      "图纸": 46673,
+      "大自然": 46674,
+      "未答辩": 46675,
+      "▁valor": 46676,
+      "面条": 46677,
+      "hua": 46678,
+      "市场监督管理局登记成立": 46679,
+      "▁cracks": 46680,
+      "回购": 46681,
+      "城南": 46682,
+      "Cos": 46683,
+      "▁fortunate": 46684,
+      "干活": 46685,
+      "flo": 46686,
+      "的个人": 46687,
+      "intendent": 46688,
+      "可为": 46689,
+      "比上": 46690,
+      "▁restoring": 46691,
+      "拇指": 46692,
+      "东亚": 46693,
+      "梅西": 46694,
+      "窗帘": 46695,
+      "DDR": 46696,
+      "▁选择": 46697,
+      "马拉松": 46698,
+      "▁lender": 46699,
+      "▁exploited": 46700,
+      "▁Abraham": 46701,
+      "师父": 46702,
+      "日军": 46703,
+      "▁»\\": 46704,
+      "事实与": 46705,
+      "降水量": 46706,
+      "▁Lum": 46707,
+      "amycin": 46708,
+      "▁Names": 46709,
+      "Statistics": 46710,
+      "▁installer": 46711,
+      "▁э": 46712,
+      "强者": 46713,
+      "▁ze": 46714,
+      "音乐学院": 46715,
+      "yards": 46716,
+      "▁Somerset": 46717,
+      "亦有": 46718,
+      "依法予以": 46719,
+      "▁apoptotic": 46720,
+      "▁discharged": 46721,
+      "▁ب": 46722,
+      "知道的": 46723,
+      "atsu": 46724,
+      "常住人口": 46725,
+      "▁agrees": 46726,
+      "▁MX": 46727,
+      "几十年": 46728,
+      "ickness": 46729,
+      "发文": 46730,
+      "奇的": 46731,
+      "蓬莱": 46732,
+      "下一代": 46733,
+      "作他用": 46734,
+      "所说": 46735,
+      "百度地图": 46736,
+      "这意味着": 46737,
+      "atterns": 46738,
+      "附相关法律条文": 46739,
+      "▁epsilon": 46740,
+      "▁Classical": 46741,
+      "再将": 46742,
+      "和应用": 46743,
+      "已全部": 46744,
+      "计划的": 46745,
+      "跟我说": 46746,
+      "本院立案": 46747,
+      "▁\\'\\';\\": 46748,
+      "▁freshly": 46749,
+      "之意": 46750,
+      "安定": 46751,
+      "谁知": 46752,
+      "▁pillow": 46753,
+      "kW": 46754,
+      "бо": 46755,
+      "igated": 46756,
+      "▁mobil": 46757,
+      "▁Methodist": 46758,
+      "JD": 46759,
+      "人选": 46760,
+      "眼部": 46761,
+      "雄厚": 46762,
+      "leys": 46763,
+      "▁Rus": 46764,
+      "存在于": 46765,
+      "正能量": 46766,
+      "ariat": 46767,
+      "建设集团有限公司": 46768,
+      "▁emigrants": 46769,
+      "nHint": 46770,
+      "成一": 46771,
+      "襄阳": 46772,
+      "▁hyster": 46773,
+      "▁spheres": 46774,
+      "amazonaws": 46775,
+      "等量": 46776,
+      "领土": 46777,
+      "做生意": 46778,
+      "历史文化": 46779,
+      "endering": 46780,
+      "▁Species": 46781,
+      "QC": 46782,
+      "伊斯兰": 46783,
+      "▁Obs": 46784,
+      "Complex": 46785,
+      "▁united": 46786,
+      "▁grounded": 46787,
+      "小说阅读网": 46788,
+      "▁pitcher": 46789,
+      "▁progenitor": 46790,
+      "▁rifle": 46791,
+      "▁helpers": 46792,
+      "▁wellbeing": 46793,
+      "奥会": 46794,
+      "DEL": 46795,
+      "nNA": 46796,
+      "▁Belarus": 46797,
+      "bh": 46798,
+      "▁LIKE": 46799,
+      "▁Mate": 46800,
+      "▁recognise": 46801,
+      "四肢": 46802,
+      "湿润": 46803,
+      "▁MET": 46804,
+      "▁Tub": 46805,
+      "▁budgets": 46806,
+      "▁Boulevard": 46807,
+      "▁勇": 46808,
+      "viol": 46809,
+      "枪支": 46810,
+      "注意我": 46811,
+      "函数的性质": 46812,
+      "hooting": 46813,
+      "▁philanth": 46814,
+      "供销": 46815,
+      "身的": 46816,
+      "yse": 46817,
+      "纳税人": 46818,
+      "芯片组": 46819,
+      "▁dile": 46820,
+      "▁reload": 46821,
+      "▁possesses": 46822,
+      "▁transfusion": 46823,
+      "金币": 46824,
+      "第一部": 46825,
+      "instrument": 46826,
+      "骨髓": 46827,
+      "回到了": 46828,
+      "▁Cotton": 46829,
+      "▁Fitness": 46830,
+      "▁substituted": 46831,
+      "降解": 46832,
+      "Sen": 46833,
+      "▁Imagine": 46834,
+      "▁descriptor": 46835,
+      "固然": 46836,
+      "▁系统": 46837,
+      "生成的": 46838,
+      "▁Difference": 46839,
+      "正经": 46840,
+      "▁Teh": 46841,
+      "Marker": 46842,
+      "professional": 46843,
+      "食谱": 46844,
+      "农业生产": 46845,
+      "挽回": 46846,
+      "▁Ellis": 46847,
+      "▁noisy": 46848,
+      "终结本次执行": 46849,
+      "前置": 46850,
+      "受体": 46851,
+      "玉林": 46852,
+      "排版": 46853,
+      "热力": 46854,
+      "续行查封": 46855,
+      "不错的选择": 46856,
+      "xp": 46857,
+      "分之": 46858,
+      "后代": 46859,
+      "航空公司": 46860,
+      "瞄准": 46861,
+      "选择上述": 46862,
+      "党工委": 46863,
+      "的一个重要": 46864,
+      "▁maneu": 46865,
+      "SSH": 46866,
+      "也表示": 46867,
+      "会发生": 46868,
+      "nBig": 46869,
+      "itian": 46870,
+      "▁Shift": 46871,
+      "▁Investigation": 46872,
+      "▁ile": 46873,
+      "证明被告": 46874,
+      "Reason": 46875,
+      "nRecord": 46876,
+      "刘备": 46877,
+      "代表性": 46878,
+      "▁circadian": 46879,
+      "▁Modified": 46880,
+      "▁risky": 46881,
+      "▁vascul": 46882,
+      "▁forecasts": 46883,
+      "学前": 46884,
+      "订购": 46885,
+      "不要太": 46886,
+      "什么呢": 46887,
+      "hmad": 46888,
+      "isto": 46889,
+      "nother": 46890,
+      "▁Parents": 46891,
+      "▁oscillations": 46892,
+      "坐堂": 46893,
+      "武林": 46894,
+      "户居住": 46895,
+      "找一个": 46896,
+      "▁blanket": 46897,
+      "▁conditioned": 46898,
+      "商圈": 46899,
+      "驾驶人": 46900,
+      "▁BIG": 46901,
+      "最大限度": 46902,
+      "packed": 46903,
+      "绿茶": 46904,
+      "▁FB": 46905,
+      "PIPE": 46906,
+      "arat": 46907,
+      "answers": 46908,
+      "▁uncovered": 46909,
+      "出走": 46910,
+      "▁Tank": 46911,
+      "tikzpicture": 46912,
+      "▁recognizing": 46913,
+      "他人财物": 46914,
+      "▁Architect": 46915,
+      "私吞": 46916,
+      "etles": 46917,
+      "Quantity": 46918,
+      "▁Salvador": 46919,
+      "▁activator": 46920,
+      "▁Polar": 46921,
+      "▁Props": 46922,
+      "perfect": 46923,
+      "▁分": 46924,
+      "国库": 46925,
+      "画画": 46926,
+      "过点": 46927,
+      "特点是": 46928,
+      "我们应该": 46929,
+      "bounds": 46930,
+      "contacts": 46931,
+      "▁Examination": 46932,
+      "江阴": 46933,
+      "apid": 46934,
+      "购物中心": 46935,
+      "▁predecessor": 46936,
+      "红豆": 46937,
+      "YES": 46938,
+      "置换": 46939,
+      "\">[": 46940,
+      "thouse": 46941,
+      "▁zoo": 46942,
+      "nyear": 46943,
+      "OBJECT": 46944,
+      "▁amplifier": 46945,
+      "▁protesters": 46946,
+      "公布了": 46947,
+      "▁Angels": 46948,
+      "▁dragon": 46949,
+      "治疗效果": 46950,
+      "▁discounted": 46951,
+      "FG": 46952,
+      "海口": 46953,
+      "▁sham": 46954,
+      "借款到期后": 46955,
+      "connector": 46956,
+      "救治": 46957,
+      "作用的": 46958,
+      "▁ACL": 46959,
+      "▁resultant": 46960,
+      ">()\\": 46961,
+      "一个不断": 46962,
+      "delivery": 46963,
+      "▁branching": 46964,
+      "ivel": 46965,
+      "▁Seoul": 46966,
+      "udos": 46967,
+      "gitlab": 46968,
+      "quarter": 46969,
+      "这种方法": 46970,
+      "▁Hebrew": 46971,
+      "温水": 46972,
+      "版次": 46973,
+      "豆浆": 46974,
+      "论述了": 46975,
+      "nGreen": 46976,
+      "▁Arsenal": 46977,
+      "▁Unicode": 46978,
+      "▁myeloid": 46979,
+      "ál": 46980,
+      "但其实": 46981,
+      "▁pedag": 46982,
+      "▁framed": 46983,
+      "CAM": 46984,
+      "第十一章": 46985,
+      "▁Loan": 46986,
+      "真爱": 46987,
+      "opoly": 46988,
+      "争执": 46989,
+      "掉的": 46990,
+      "玩家们": 46991,
+      "▁Cafe": 46992,
+      "▁Intent": 46993,
+      "Reflection": 46994,
+      "qrt": 46995,
+      "着一个": 46996,
+      "▁Ware": 46997,
+      "▁spice": 46998,
+      "铃声": 46999,
+      "leave": 47000,
+      "不计": 47001,
+      "}})": 47002,
+      "imetry": 47003,
+      "不远": 47004,
+      "群众的": 47005,
+      "个人觉得": 47006,
+      "▁pivot": 47007,
+      "▁abnorm": 47008,
+      "纪念馆": 47009,
+      "里面有": 47010,
+      "nOption": 47011,
+      "▁Composite": 47012,
+      "ajo": 47013,
+      "组织和": 47014,
+      "Bear": 47015,
+      "▁Glad": 47016,
+      "resolver": 47017,
+      "fruit": 47018,
+      "▁jupytext": 47019,
+      "▁discontinued": 47020,
+      "▁periodically": 47021,
+      "控制在": 47022,
+      "已经开始": 47023,
+      "xb": 47024,
+      "自贡": 47025,
+      "误会": 47026,
+      "god": 47027,
+      "Pull": 47028,
+      "rone": 47029,
+      "▁Aires": 47030,
+      "▁minimally": 47031,
+      "▁tolerated": 47032,
+      "你可能": 47033,
+      "同时还": 47034,
+      "▁Forty": 47035,
+      "▁SLE": 47036,
+      "▁solv": 47037,
+      "粉尘": 47038,
+      "▁pts": 47039,
+      "Dim": 47040,
+      "废弃了": 47041,
+      "acycl": 47042,
+      "▁mandate": 47043,
+      "科尔": 47044,
+      "▁Philosophy": 47045,
+      "五千": 47046,
+      "出水": 47047,
+      "nAh": 47048,
+      "Soup": 47049,
+      "archives": 47050,
+      "进食": 47051,
+      "适量的": 47052,
+      "abic": 47053,
+      "imaging": 47054,
+      "gia": 47055,
+      "nIT": 47056,
+      "welcome": 47057,
+      "▁turkey": 47058,
+      "▁conspiracy": 47059,
+      "▁repertoire": 47060,
+      "▁quantitatively": 47061,
+      "点和": 47062,
+      "viv": 47063,
+      "Although": 47064,
+      "天子": 47065,
+      "培养学生": 47066,
+      "▁nominal": 47067,
+      "求助": 47068,
+      "▁$\\": 47069,
+      "▁）\\": 47070,
+      "nfun": 47071,
+      "一审判决": 47072,
+      "▁dipole": 47073,
+      "者在": 47074,
+      "▁Ded": 47075,
+      "▁sock": 47076,
+      "▁Embed": 47077,
+      "▁trillion": 47078,
+      "镇政府": 47079,
+      "▁Minneapolis": 47080,
+      "学问": 47081,
+      "伊拉克": 47082,
+      "反映给": 47083,
+      "和质量": 47084,
+      "这样的人": 47085,
+      "▁Audi": 47086,
+      "▁Bent": 47087,
+      "弱的": 47088,
+      "至尊": 47089,
+      "▁guinea": 47090,
+      "▁BBQ": 47091,
+      "第二百零七条": 47092,
+      "▁Height": 47093,
+      "HEADER": 47094,
+      "挪作他用": 47095,
+      "▁factories": 47096,
+      "▁Recognition": 47097,
+      "咽喉": 47098,
+      "肥料": 47099,
+      "去括号": 47100,
+      "非线性": 47101,
+      "imize": 47102,
+      "ischen": 47103,
+      "叫我": 47104,
+      "棱长": 47105,
+      "我们对": 47106,
+      "Hint": 47107,
+      "一审原告": 47108,
+      "decorators": 47109,
+      "▁其它": 47110,
+      "tright": 47111,
+      "▁senses": 47112,
+      "coordinate": 47113,
+      "▁epidermal": 47114,
+      "问候": 47115,
+      "▁KO": 47116,
+      "utta": 47117,
+      "▁appetite": 47118,
+      "▁fulfilled": 47119,
+      "回避": 47120,
+      "到案后": 47121,
+      "▁布": 47122,
+      "分歧": 47123,
+      "火力": 47124,
+      "点燃": 47125,
+      "风口": 47126,
+      "▁waterproof": 47127,
+      "▁proposition": 47128,
+      "UCT": 47129,
+      "的最大值": 47130,
+      "nDear": 47131,
+      "▁androgen": 47132,
+      "notifications": 47133,
+      "下调": 47134,
+      "不经": 47135,
+      "定为": 47136,
+      "crum": 47137,
+      "Posts": 47138,
+      "}),": 47139,
+      "oyle": 47140,
+      "▁WOR": 47141,
+      "与众": 47142,
+      "机票": 47143,
+      "考场": 47144,
+      "热水器": 47145,
+      "目标是": 47146,
+      "asar": 47147,
+      "ToMany": 47148,
+      "entieth": 47149,
+      "▁cipher": 47150,
+      "▁footer": 47151,
+      "▁periodontal": 47152,
+      "QP": 47153,
+      "九章": 47154,
+      "睡着": 47155,
+      "CMD": 47156,
+      "贸市场": 47157,
+      "calculate": 47158,
+      "针织": 47159,
+      "ympt": 47160,
+      "从重处罚": 47161,
+      "leetcode": 47162,
+      "四面": 47163,
+      "大片": 47164,
+      "起草": 47165,
+      "怎么会": 47166,
+      "▁retry": 47167,
+      "▁tread": 47168,
+      "▁freshman": 47169,
+      "▁Obviously": 47170,
+      "▁unilateral": 47171,
+      "WL": 47172,
+      "低下": 47173,
+      "的要": 47174,
+      "延长赛": 47175,
+      "误解": 47176,
+      "agrang": 47177,
+      "▁evaluates": 47178,
+      "▁disestablish": 47179,
+      "新高": 47180,
+      "sev": 47181,
+      "▁Hang": 47182,
+      "▁suites": 47183,
+      "政协委员": 47184,
+      "本院受理": 47185,
+      "▁bishops": 47186,
+      "▁digestive": 47187,
+      "苗族": 47188,
+      "Prim": 47189,
+      "coal": 47190,
+      "icos": 47191,
+      "libraries": 47192,
+      "奥特": 47193,
+      "akk": 47194,
+      "于一体的": 47195,
+      "以及其他": 47196,
+      "▁lupus": 47197,
+      "▁parade": 47198,
+      "该卡": 47199,
+      "GEN": 47200,
+      "住址同上": 47201,
+      "GROUP": 47202,
+      "isive": 47203,
+      "▁Worcester": 47204,
+      "读物": 47205,
+      "▁Isle": 47206,
+      "庭院": 47207,
+      "neur": 47208,
+      "追踪报道": 47209,
+      "▁Gear": 47210,
+      "德尔": 47211,
+      "田间": 47212,
+      "网吧": 47213,
+      "Kel": 47214,
+      "很不错": 47215,
+      "artan": 47216,
+      "ighbors": 47217,
+      "nCREATE": 47218,
+      "▁WIN": 47219,
+      "Central": 47220,
+      "▁Christianity": 47221,
+      "搞定": 47222,
+      "在他的": 47223,
+      "CKET": 47224,
+      "esses": 47225,
+      "二〇一六年三月": 47226,
+      "海区": 47227,
+      "▁ridge": 47228,
+      "experiment": 47229,
+      "nUS": 47230,
+      "▁FS": 47231,
+      "▁xi": 47232,
+      "▁Income": 47233,
+      "▁Regulation": 47234,
+      "▁humanitarian": 47235,
+      "筷子": 47236,
+      "认定为": 47237,
+      "anck": 47238,
+      "比较方便": 47239,
+      "国强": 47240,
+      "星光": 47241,
+      "极坐标": 47242,
+      "▁Offer": 47243,
+      "认定事实如下": 47244,
+      "讨论版": 47245,
+      "visited": 47246,
+      "▁Broadcasting": 47247,
+      "测验": 47248,
+      "驻马": 47249,
+      "ienne": 47250,
+      "▁ipsum": 47251,
+      "▁privileged": 47252,
+      "▁enthusiastic": 47253,
+      "▁reservations": 47254,
+      "thel": 47255,
+      "有限公司负担": 47256,
+      "Screenshot": 47257,
+      "核对": 47258,
+      "请勿": 47259,
+      "以下是": 47260,
+      "莆田市": 47261,
+      "这方面": 47262,
+      "nUSE": 47263,
+      "nval": 47264,
+      "marshal": 47265,
+      "一两": 47266,
+      "▁Forms": 47267,
+      "▁homotopy": 47268,
+      "代表人的人数提出副本": 47269,
+      "yun": 47270,
+      "安阳市": 47271,
+      "tera": 47272,
+      "▁branded": 47273,
+      "President": 47274,
+      "廊坊": 47275,
+      "▁kDa": 47276,
+      "▁tee": 47277,
+      "在高": 47278,
+      "ennium": 47279,
+      "▁elder": 47280,
+      "Serialized": 47281,
+      "之初": 47282,
+      "号征收社会抚养费": 47283,
+      "▁educator": 47284,
+      "伤亡": 47285,
+      "支柱": 47286,
+      "estones": 47287,
+      "▁undertaking": 47288,
+      "▁pretreatment": 47289,
+      "野外": 47290,
+      "ISP": 47291,
+      "nMac": 47292,
+      "▁aquatic": 47293,
+      "邮票": 47294,
+      "全网": 47295,
+      "历年": 47296,
+      "战的": 47297,
+      "烘干": 47298,
+      "医学会": 47299,
+      "受害者": 47300,
+      "▁营业时间": 47301,
+      "扭曲": 47302,
+      "找回": 47303,
+      "潜能": 47304,
+      "▁lace": 47305,
+      "▁人口数量": 47306,
+      "▁mentors": 47307,
+      "▁Tanzania": 47308,
+      "▁mutually": 47309,
+      "多样化": 47310,
+      "voice": 47311,
+      "▁Clif": 47312,
+      "▁brave": 47313,
+      "偿还借款本金": 47314,
+      "▁biased": 47315,
+      "▁paradox": 47316,
+      "Implementation": 47317,
+      "▁mercury": 47318,
+      "ighborhood": 47319,
+      "▁Computational": 47320,
+      "吃吃": 47321,
+      "炒作": 47322,
+      "西侧": 47323,
+      "▁bent": 47324,
+      "▁oval": 47325,
+      "店里": 47326,
+      "绿豆": 47327,
+      "ICH": 47328,
+      "也开始": 47329,
+      "作品的": 47330,
+      "popup": 47331,
+      "▁Fact": 47332,
+      "▁Holmes": 47333,
+      "▁Drawing": 47334,
+      "▁TestCase": 47335,
+      "▁immigrant": 47336,
+      "变价": 47337,
+      "▁[<": 47338,
+      "着名的": 47339,
+      "lng": 47340,
+      "工商户": 47341,
+      "世界上最": 47342,
+      "申请仲裁": 47343,
+      "experimental": 47344,
+      "themes": 47345,
+      "chief": 47346,
+      "courses": 47347,
+      "▁deterioration": 47348,
+      "学生们": 47349,
+      "cock": 47350,
+      "nProof": 47351,
+      "▁Eclipse": 47352,
+      "plantation": 47353,
+      "▁Infrastructure": 47354,
+      "卡车": 47355,
+      "向南": 47356,
+      "tbl": 47357,
+      "下面是": 47358,
+      "▁Trek": 47359,
+      "▁sword": 47360,
+      "▁retard": 47361,
+      "▁stereo": 47362,
+      "▁resignation": 47363,
+      "其一": 47364,
+      "有不": 47365,
+      "/\">": 47366,
+      "方程的解": 47367,
+      "getting": 47368,
+      "精益": 47369,
+      "▁Ci": 47370,
+      "▁MK": 47371,
+      "▁`@": 47372,
+      "履行上述": 47373,
+      "▁cada": 47374,
+      "Warnings": 47375,
+      "包子": 47376,
+      "小伙": 47377,
+      "▁\"\")": 47378,
+      "科学出版社": 47379,
+      "▁牛": 47380,
+      "这片": 47381,
+      "▁fights": 47382,
+      "▁peculiar": 47383,
+      "▁underest": 47384,
+      "放电": 47385,
+      "anj": 47386,
+      "入围": 47387,
+      "放开": 47388,
+      "万人次": 47389,
+      "▁syndromes": 47390,
+      "饮品": 47391,
+      "▁HDL": 47392,
+      "▁computations": 47393,
+      "秦皇岛": 47394,
+      "而且在": 47395,
+      "EDIT": 47396,
+      "]){\\": 47397,
+      "adia": 47398,
+      "避开": 47399,
+      "的成本": 47400,
+      "还是挺": 47401,
+      "erville": 47402,
+      "▁proudly": 47403,
+      "▁Provincial": 47404,
+      "替补": 47405,
+      "▁最后": 47406,
+      "副县长": 47407,
+      "▁Boss": 47408,
+      "▁神": 47409,
+      "▁nonzero": 47410,
+      "和计划生育委员会": 47411,
+      "upuncture": 47412,
+      "▁incorrectly": 47413,
+      "辆车": 47414,
+      "因为是": 47415,
+      "▁GmbH": 47416,
+      "nEuropean": 47417,
+      "▁ignoring": 47418,
+      "下手": 47419,
+      "ucked": 47420,
+      "USERNAME": 47421,
+      "ес": 47422,
+      "tada": 47423,
+      "tiny": 47424,
+      "▁Nom": 47425,
+      "还有什么": 47426,
+      "ometers": 47427,
+      "元为": 47428,
+      "西祠": 47429,
+      "角膜": 47430,
+      "hon": 47431,
+      "emory": 47432,
+      "二〇一七年五月": 47433,
+      "年会": 47434,
+      "纠缠": 47435,
+      "呼吸道": 47436,
+      "Money": 47437,
+      "▁tubular": 47438,
+      "▁coherence": 47439,
+      "的政治": 47440,
+      "▁Maths": 47441,
+      "▁Dining": 47442,
+      "\\\\\\\\\\\\\\\\": 47443,
+      "▁nursery": 47444,
+      "▁alleviate": 47445,
+      "一女": 47446,
+      "ORG": 47447,
+      "gson": 47448,
+      "情况进行": 47449,
+      "▁spill": 47450,
+      "▁programmer": 47451,
+      "————————————————": 47452,
+      "中之": 47453,
+      "抗议": 47454,
+      "andi": 47455,
+      "campaign": 47456,
+      "lj": 47457,
+      "磁盘": 47458,
+      "▁inhal": 47459,
+      "▁heparin": 47460,
+      "后人": 47461,
+      "巴拉": 47462,
+      "帮助下": 47463,
+      "Avatar": 47464,
+      "▁herpes": 47465,
+      "▁SAT": 47466,
+      "短时间内": 47467,
+      "脚下": 47468,
+      "Occ": 47469,
+      "双方在": 47470,
+      "业主收取": 47471,
+      "▁liable": 47472,
+      "▁expedition": 47473,
+      "侮辱": 47474,
+      "MAIL": 47475,
+      "▁cpu": 47476,
+      "▁lou": 47477,
+      "非常高兴能": 47478,
+      "▁Syntax": 47479,
+      "▁longevity": 47480,
+      "▁Ö": 47481,
+      "事的": 47482,
+      "和人": 47483,
+      "盐城市": 47484,
+      "nhas": 47485,
+      "nPower": 47486,
+      "▁Qatar": 47487,
+      "▁lightning": 47488,
+      "Aff": 47489,
+      "▁Ancient": 47490,
+      "▁Chennai": 47491,
+      "▁Vitamin": 47492,
+      "▁workload": 47493,
+      "FFER": 47494,
+      "号案件的执行": 47495,
+      "updates": 47496,
+      "▁Healthy": 47497,
+      "ERENCES": 47498,
+      "▁procurement": 47499,
+      "但你": 47500,
+      "鉴赏": 47501,
+      "两部分": 47502,
+      "▁numerically": 47503,
+      "▁speculation": 47504,
+      "刻画": 47505,
+      "陆游": 47506,
+      "仅仅是": 47507,
+      "opencv": 47508,
+      "▁Former": 47509,
+      "▁Minimum": 47510,
+      "▁postgres": 47511,
+      "向被执行人发出执行通知书": 47512,
+      "实木": 47513,
+      "食管": 47514,
+      "▁sunset": 47515,
+      "▁文学": 47516,
+      "有时间": 47517,
+      "▁Bav": 47518,
+      "enburg": 47519,
+      "▁Patch": 47520,
+      "decimal": 47521,
+      "patibility": 47522,
+      "耐药": 47523,
+      "DisplayName": 47524,
+      "Cy": 47525,
+      "各主要": 47526,
+      "ifug": 47527,
+      "nkind": 47528,
+      "▁Prote": 47529,
+      "represented": 47530,
+      "黑的": 47531,
+      "门户网站": 47532,
+      "▁naval": 47533,
+      "第十三条第二款": 47534,
+      "不断累": 47535,
+      "▁Apollo": 47536,
+      "▁ASF": 47537,
+      "▁turbine": 47538,
+      "▁technician": 47539,
+      "住建": 47540,
+      "招待": 47541,
+      "退赔": 47542,
+      "stones": 47543,
+      "▁Crist": 47544,
+      "▁Kubernetes": 47545,
+      "olith": 47546,
+      "unts": 47547,
+      "ighed": 47548,
+      "aciones": 47549,
+      "▁exceeding": 47550,
+      "上车": 47551,
+      "我的经纪人": 47552,
+      "nTrump": 47553,
+      "▁Miles": 47554,
+      "configs": 47555,
+      "▁linguistic": 47556,
+      "USED": 47557,
+      "▁herbal": 47558,
+      "▁Ottoman": 47559,
+      "producing": 47560,
+      "隔音": 47561,
+      "IMO": 47562,
+      "日美股": 47563,
+      "pairs": 47564,
+      "▁Raid": 47565,
+      "我非常感激": 47566,
+      "▁ferry": 47567,
+      "▁rupture": 47568,
+      "工业化": 47569,
+      "应用的": 47570,
+      "chemical": 47571,
+      "棵树": 47572,
+      "而上": 47573,
+      "lld": 47574,
+      "与本案": 47575,
+      "adren": 47576,
+      "▁Perc": 47577,
+      "▁impose": 47578,
+      "defineProperty": 47579,
+      "对我国": 47580,
+      "mant": 47581,
+      "这么多年": 47582,
+      "asper": 47583,
+      "▁Belf": 47584,
+      "itchens": 47585,
+      "▁Moment": 47586,
+      "▁Crit": 47587,
+      "▁Sites": 47588,
+      "▁frank": 47589,
+      "和知识": 47590,
+      "状态下": 47591,
+      "心底": 47592,
+      "▁AE": 47593,
+      "违反了": 47594,
+      "我们将予以": 47595,
+      "▁Lynch": 47596,
+      "TN": 47597,
+      "▁Joan": 47598,
+      "▁avail": 47599,
+      "▁Initially": 47600,
+      "hint": 47601,
+      "你家": 47602,
+      "更要": 47603,
+      "的犯罪": 47604,
+      "Prep": 47605,
+      "▁PTSD": 47606,
+      "二〇一七年三月": 47607,
+      "nFootballers": 47608,
+      "一向": 47609,
+      "妨碍": 47610,
+      "Calc": 47611,
+      "▁cruel": 47612,
+      "止痛": 47613,
+      "的案件": 47614,
+      "quez": 47615,
+      "listdir": 47616,
+      "dimension": 47617,
+      "调制": 47618,
+      "iguez": 47619,
+      "nRelease": 47620,
+      "引发的": 47621,
+      "赢得胜利的": 47622,
+      "▁Economy": 47623,
+      "威力": 47624,
+      "枝花": 47625,
+      "是由于": 47626,
+      "的控制": 47627,
+      "多边形的": 47628,
+      "▁localities": 47629,
+      "▁neuropsych": 47630,
+      "浓厚": 47631,
+      "读了": 47632,
+      "研究室": 47633,
+      "这部分": 47634,
+      "motion": 47635,
+      "▁Myanmar": 47636,
+      "▁responds": 47637,
+      "POR": 47638,
+      "防火墙": 47639,
+      "Inflater": 47640,
+      "▁广州": 47641,
+      "得更全面": 47642,
+      "Blueprint": 47643,
+      "Operations": 47644,
+      "酰胺": 47645,
+      "Band": 47646,
+      "人民调解协议": 47647,
+      "▁Coalition": 47648,
+      "▁wastewater": 47649,
+      "责任心": 47650,
+      "重组案": 47651,
+      "Theta": 47652,
+      "▁Stefan": 47653,
+      "▁influx": 47654,
+      "▁retiring": 47655,
+      "▁Strategic": 47656,
+      "产科": 47657,
+      "偶函数": 47658,
+      "就直接": 47659,
+      "缺席审理": 47660,
+      "演唱的歌曲": 47661,
+      "信道": 47662,
+      "向西": 47663,
+      "▁Few": 47664,
+      "明确的被告": 47665,
+      "London": 47666,
+      "全都": 47667,
+      "锅里": 47668,
+      "专家坐堂": 47669,
+      "南京还有": 47670,
+      "antage": 47671,
+      "▁compensate": 47672,
+      "▁PU": 47673,
+      "nBuy": 47674,
+      "▁Labs": 47675,
+      "所有的位于": 47676,
+      "▁cubes": 47677,
+      "广东省深圳市": 47678,
+      "provided": 47679,
+      "}'": 47680,
+      "先说": 47681,
+      "吻合": 47682,
+      "怀特说": 47683,
+      "ipeg": 47684,
+      "支出色的": 47685,
+      "▁Sharp": 47686,
+      "nImport": 47687,
+      "会签下": 47688,
+      "nnow": 47689,
+      "▁Brief": 47690,
+      "▁Hurricane": 47691,
+      "五行": 47692,
+      "嘲讽": 47693,
+      "农业农村": 47694,
+      "▁congressional": 47695,
+      "图表": 47696,
+      "巴基": 47697,
+      "生产劳动": 47698,
+      "十三条规定": 47699,
+      "▁midst": 47700,
+      "[#": 47701,
+      "进军": 47702,
+      "了一批": 47703,
+      "你以为": 47704,
+      "Spot": 47705,
+      "ypress": 47706,
+      "号行政": 47707,
+      "公摊水费": 47708,
+      "延长赛落败": 47709,
+      "证监会否决": 47710,
+      "反映给相关部门": 47711,
+      "把她": 47712,
+      "白玉": 47713,
+      "加深天马": 47714,
+      "融券筹码": 47715,
+      "实录加西亚": 47716,
+      "重组案再遭": 47717,
+      "▁Ratio": 47718,
+      "不断累加深天马": 47719,
+      "日美股专家坐堂": 47720,
+      "实录加西亚延长赛落败": 47721,
+      "融券筹码不断累加深天马": 47722,
+      "融券筹码不断累加深天马重组案再遭": 47723,
+      "羊毛": 47724,
+      "学术论文": 47725,
+      "求和": 47726,
+      "lnx": 47727,
+      "▁英语": 47728,
+      "indy": 47729,
+      "与被告王": 47730,
+      "platin": 47731,
+      "rss": 47732,
+      "▁NZ": 47733,
+      "不了解": 47734,
+      "▁Dod": 47735,
+      "▁Clin": 47736,
+      "▁Dale": 47737,
+      "西祠讨论版": 47738,
+      "支出色的球队": 47739,
+      "▁Breast": 47740,
+      "我的经纪人告诉我": 47741,
+      "万能": 47742,
+      "推介": 47743,
+      "特此通知": 47744,
+      "签约火箭": 47745,
+      "▁fier": 47746,
+      "将自己磨练": 47747,
+      "有几支球队": 47748,
+      "benchmark": 47749,
+      "▁Correction": 47750,
+      "第一百五十四条裁定适用于下列范围": 47751,
+      "ilk": 47752,
+      "晚饭团": 47753,
+      "月获得": 47754,
+      "会签下我": 47755,
+      "却被物管": 47756,
+      "请求判令": 47757,
+      "你的遭遇吧": 47758,
+      "却依然在向": 47759,
+      "扬子晚饭团": 47760,
+      "明明废弃了": 47761,
+      "一直在关注我": 47762,
+      "却被物管私吞": 47763,
+      "并将你的遭遇": 47764,
+      "我坚信有球队": 47765,
+      "我知道有球队": 47766,
+      "所以我很刻苦": 47767,
+      "而且这个城市": 47768,
+      "怀特非常高兴能": 47769,
+      "我现在来到了一": 47770,
+      "说出你的遭遇吧": 47771,
+      "赢得胜利的球队": 47772,
+      "业主收取公摊水费": 47773,
+      "我非常感激这一切": 47774,
+      "租金却被物管私吞": 47775,
+      "请登录西祠讨论版": 47776,
+      "又有多少小区的水景": 47777,
+      "将自己磨练得更全面": 47778,
+      "我们将予以追踪报道": 47779,
+      "我知道有球队注意我": 47780,
+      "而且这个城市也很好": 47781,
+      "我坚信有球队会签下我": 47782,
+      "一个不断赢得胜利的球队": 47783,
+      "南京还有多少小区的水景": 47784,
+      "怀特非常高兴能签约火箭": 47785,
+      "有几支球队一直在关注我": 47786,
+      "却依然在向业主收取公摊水费": 47787,
+      "又有多少小区的水景挪作他用": 47788,
+      "并将你的遭遇反映给相关部门": 47789,
+      "我现在来到了一支出色的球队": 47790,
+      "南京还有多少小区的水景明明废弃了": 47791,
+      "tG": 47792,
+      "会谈": 47793,
+      "血流": 47794,
+      "▁最大": 47795,
+      "▁Thing": 47796,
+      "▁macros": 47797,
+      "▁cottage": 47798,
+      "▁Buddhist": 47799,
+      "▁Chocolate": 47800,
+      "bz": 47801,
+      "九十": 47802,
+      "的粉丝": 47803,
+      "nEqu": 47804,
+      "▁belly": 47805,
+      "慎重": 47806,
+      "氧气": 47807,
+      "而过": 47808,
+      "▁Fra": 47809,
+      "essive": 47810,
+      "▁Rocky": 47811,
+      "centric": 47812,
+      "穿刺": 47813,
+      "Lif": 47814,
+      "sentence": 47815,
+      "房的": 47816,
+      "ermine": 47817,
+      "nRugby": 47818,
+      "▁Someone": 47819,
+      "▁hydrophobic": 47820,
+      "收割": 47821,
+      "艰辛": 47822,
+      "nJim": 47823,
+      "▁登场作品": 47824,
+      "▁screws": 47825,
+      "关节炎": 47826,
+      "▁resistor": 47827,
+      "Sa": 47828,
+      "收盘": 47829,
+      "做出的": 47830,
+      "pieces": 47831,
+      "▁Rescue": 47832,
+      "▁sel": 47833,
+      "▁Pag": 47834,
+      "Catalog": 47835,
+      "▁analogs": 47836,
+      "是人": 47837,
+      "不明白": 47838,
+      "也是很": 47839,
+      "激烈的": 47840,
+      "搅拌均匀": 47841,
+      "▁toolbar": 47842,
+      "▁Maharashtra": 47843,
+      "Nor": 47844,
+      "irical": 47845,
+      "implementation": 47846,
+      "即可求得": 47847,
+      "▁Sync": 47848,
+      "向我": 47849,
+      "年少": 47850,
+      "雪花": 47851,
+      "ceipt": 47852,
+      "ку": 47853,
+      "▁Echo": 47854,
+      "▁TYPE": 47855,
+      "▁soda": 47856,
+      "▁advertisement": 47857,
+      "plural": 47858,
+      "▁reviewers": 47859,
+      "马某": 47860,
+      "eres": 47861,
+      "▁Roh": 47862,
+      "▁peel": 47863,
+      "▁invaluable": 47864,
+      "▁pregnancies": 47865,
+      "▁Cities": 47866,
+      "共振": 47867,
+      "芜湖市": 47868,
+      "设计了": 47869,
+      "住山东省": 47870,
+      "balls": 47871,
+      "nBook": 47872,
+      "lantic": 47873,
+      "▁validator": 47874,
+      "▁conviction": 47875,
+      "TK": 47876,
+      "断了": 47877,
+      "济宁": 47878,
+      "联手": 47879,
+      "请用": 47880,
+      "INESS": 47881,
+      "nHomework": 47882,
+      "▁deleting": 47883,
+      "分布在": 47884,
+      "Specific": 47885,
+      "▁Symposium": 47886,
+      "ifax": 47887,
+      "individual": 47888,
+      "▁goalkeeper": 47889,
+      "▁Legislature": 47890,
+      "整顿": 47891,
+      "各自的": 47892,
+      "ivary": 47893,
+      "derive": 47894,
+      "▁Abdul": 47895,
+      "夫妻感情破裂": 47896,
+      "原型": 47897,
+      "CMS": 47898,
+      "▁Shin": 47899,
+      "▁kindly": 47900,
+      "▁ionization": 47901,
+      "该局": 47902,
+      "▁drummer": 47903,
+      "运势": 47904,
+      "各单位": 47905,
+      "▁lanes": 47906,
+      "搞好": 47907,
+      "构件": 47908,
+      "reviews": 47909,
+      "▁Prepare": 47910,
+      "▁pedestrian": 47911,
+      "祝愿": 47912,
+      "过期": 47913,
+      "chai": 47914,
+      "stry": 47915,
+      "▁sights": 47916,
+      "▁separating": 47917,
+      "率达": 47918,
+      "父亲的": 47919,
+      "市值": 47920,
+      "更适合": 47921,
+      "电动机": 47922,
+      "神秘的": 47923,
+      "\\\\;\\\\": 47924,
+      "/'": 47925,
+      "明知": 47926,
+      "意识形态": 47927,
+      "科技股份有限公司": 47928,
+      "VGA": 47929,
+      "责任的": 47930,
+      ">(\\": 47931,
+      "专业课": 47932,
+      "Accessor": 47933,
+      "血症": 47934,
+      "iec": 47935,
+      "~~~": 47936,
+      "脂肪酸": 47937,
+      "▁Nadu": 47938,
+      "ichlet": 47939,
+      "▁deterministic": 47940,
+      "ANY": 47941,
+      "ulance": 47942,
+      "▁Chess": 47943,
+      "▁admits": 47944,
+      "▁racism": 47945,
+      "等证据予以证实": 47946,
+      "子孙": 47947,
+      "▁fd": 47948,
+      "生前为": 47949,
+      "▁rebounds": 47950,
+      "▁postnatal": 47951,
+      "▁Pt": 47952,
+      "心脏病": 47953,
+      "tName": 47954,
+      "▁peek": 47955,
+      "nAdditionally": 47956,
+      "执行至": 47957,
+      "trad": 47958,
+      "一个新的": 47959,
+      "学习教育": 47960,
+      "▁showers": 47961,
+      "limitations": 47962,
+      "深远": 47963,
+      "▁——": 47964,
+      "▁nont": 47965,
+      "▁PayPal": 47966,
+      "▁pumping": 47967,
+      "▁commodity": 47968,
+      "▁repositories": 47969,
+      "▁transferring": 47970,
+      "INDEX": 47971,
+      "nSystem": 47972,
+      "▁Northeast": 47973,
+      "略有": 47974,
+      "创造性": 47975,
+      "半个月": 47976,
+      "▁Herr": 47977,
+      "▁diarrhea": 47978,
+      "和他的": 47979,
+      "▁rehe": 47980,
+      "DATABASE": 47981,
+      "聘请": 47982,
+      "Jen": 47983,
+      "不知不觉": 47984,
+      "▁frog": 47985,
+      "▁中共党员": 47986,
+      "▁limbs": 47987,
+      "主线": 47988,
+      "nKnow": 47989,
+      "制成的": 47990,
+      "Verify": 47991,
+      "raisal": 47992,
+      "日开始": 47993,
+      "营养成分": 47994,
+      "/,": 47995,
+      "今生": 47996,
+      "是怎样": 47997,
+      "▁Franç": 47998,
+      "▁weaknesses": 47999,
+      "关系到": 48000,
+      "nsage": 48001,
+      "▁builders": 48002,
+      "减弱": 48003,
+      "Vers": 48004,
+      "cone": 48005,
+      "euclid": 48006,
+      "nSocial": 48007,
+      "nParameters": 48008,
+      "场馆": 48009,
+      "Thursday": 48010,
+      "Jul": 48011,
+      "人意": 48012,
+      "设在": 48013,
+      "beck": 48014,
+      "多点": 48015,
+      "nFin": 48016,
+      "▁Oral": 48017,
+      "▁Listen": 48018,
+      "对被执行人的财产": 48019,
+      "▁metabolite": 48020,
+      "少于": 48021,
+      "提到的": 48022,
+      "符合下列": 48023,
+      "▁warmer": 48024,
+      "▁combinatorial": 48025,
+      "合影": 48026,
+      "中止或者": 48027,
+      "▁elem": 48028,
+      "▁consolidation": 48029,
+      "▁Cec": 48030,
+      "▁VII": 48031,
+      "erobic": 48032,
+      "▁excluding": 48033,
+      "乘车": 48034,
+      "这三": 48035,
+      "▁abd": 48036,
+      "什么东西": 48037,
+      "networks": 48038,
+      "▁inclined": 48039,
+      "▁collapsed": 48040,
+      "physics": 48041,
+      "▁尺寸": 48042,
+      "astype": 48043,
+      "receive": 48044,
+      "▁beginner": 48045,
+      "维持原裁定": 48046,
+      "ttle": 48047,
+      "ynth": 48048,
+      "▁Async": 48049,
+      "directories": 48050,
+      "不符": 48051,
+      "宽容": 48052,
+      "arah": 48053,
+      "unity": 48054,
+      "anions": 48055,
+      "敬老": 48056,
+      "压缩机": 48057,
+      "ncheck": 48058,
+      "▁行政区类别": 48059,
+      "osexual": 48060,
+      "students": 48061,
+      "▁arguing": 48062,
+      "建的": 48063,
+      "akan": 48064,
+      "▁modality": 48065,
+      "▁charitable": 48066,
+      "纵向": 48067,
+      "整数的": 48068,
+      "治疗的": 48069,
+      "brown": 48070,
+      "▁couns": 48071,
+      "▁probabilistic": 48072,
+      "就越": 48073,
+      "▁MU": 48074,
+      "因为我们": 48075,
+      "▁Trace": 48076,
+      "▁sleek": 48077,
+      "▁carved": 48078,
+      "▁referendum": 48079,
+      "应以": 48080,
+      "网球": 48081,
+      "道教": 48082,
+      "骚扰": 48083,
+      "MER": 48084,
+      "nDeath": 48085,
+      "illiams": 48086,
+      "nol": 48087,
+      "▁RED": 48088,
+      "▁geo": 48089,
+      "▁aggregates": 48090,
+      "峰值": 48091,
+      "▁Viv": 48092,
+      "▁helmet": 48093,
+      "▁resides": 48094,
+      "四边形的": 48095,
+      "▁bake": 48096,
+      "▁gestational": 48097,
+      "花瓣": 48098,
+      "评级": 48099,
+      "SCI": 48100,
+      "▁GI": 48101,
+      "本质上": 48102,
+      "罪犯刘": 48103,
+      "States": 48104,
+      "OX": 48105,
+      "▁noticeable": 48106,
+      "关键字": 48107,
+      "otti": 48108,
+      "▁scenery": 48109,
+      "converter": 48110,
+      "▁separator": 48111,
+      "山下": 48112,
+      "占用的": 48113,
+      "▁Sql": 48114,
+      "二〇一七年六月": 48115,
+      "▁Reverse": 48116,
+      "▁translates": 48117,
+      "赞助": 48118,
+      "系数的": 48119,
+      "▁está": 48120,
+      "提交的证据": 48121,
+      "Capture": 48122,
+      "古老的": 48123,
+      "很好地": 48124,
+      "aney": 48125,
+      "statistics": 48126,
+      "▁Bod": 48127,
+      "不只是": 48128,
+      "HN": 48129,
+      "growing": 48130,
+      "▁murdered": 48131,
+      "▁Louisville": 48132,
+      "▁afore": 48133,
+      "osus": 48134,
+      "▁Vern": 48135,
+      "▁游戏介绍": 48136,
+      "Decode": 48137,
+      "▁comma": 48138,
+      "▁LS": 48139,
+      "▁Ful": 48140,
+      "云龙": 48141,
+      "风波": 48142,
+      "教学的": 48143,
+      "但是如果": 48144,
+      "▁avoidance": 48145,
+      "▁松": 48146,
+      "沿革": 48147,
+      "perf": 48148,
+      "▁overt": 48149,
+      "▁valves": 48150,
+      "mongoose": 48151,
+      "ynthetic": 48152,
+      "▁Squadron": 48153,
+      "statusCode": 48154,
+      "做工": 48155,
+      "招牌": 48156,
+      "▁Duncan": 48157,
+      "ermanent": 48158,
+      "相近": 48159,
+      "▁HL": 48160,
+      "obia": 48161,
+      "▁photoc": 48162,
+      "曲折": 48163,
+      "电能": 48164,
+      "进出": 48165,
+      "\"<<": 48166,
+      "▁[{": 48167,
+      "mente": 48168,
+      "是什么意思": 48169,
+      "▁refractory": 48170,
+      "写着": 48171,
+      "模拟器": 48172,
+      "shirts": 48173,
+      "回车": 48174,
+      "民宿": 48175,
+      "elic": 48176,
+      "▁bibli": 48177,
+      "▁Ladies": 48178,
+      "▁Purchase": 48179,
+      "▁collateral": 48180,
+      "oucester": 48181,
+      "icultural": 48182,
+      "iJ": 48183,
+      "出路": 48184,
+      "▁feminine": 48185,
+      "▁Consulting": 48186,
+      "琉璃": 48187,
+      "▁Klein": 48188,
+      "unctional": 48189,
+      "自强": 48190,
+      "芹菜": 48191,
+      "Prob": 48192,
+      "▁als": 48193,
+      "nProt": 48194,
+      "German": 48195,
+      "▁Components": 48196,
+      "中式": 48197,
+      "喷雾": 48198,
+      "黄石": 48199,
+      "apo": 48200,
+      "的问题和": 48201,
+      "could": 48202,
+      "ovina": 48203,
+      "玩耍": 48204,
+      "▁pb": 48205,
+      "unal": 48206,
+      "▁Garcia": 48207,
+      "▁Leather": 48208,
+      "▁reversal": 48209,
+      "▁exploratory": 48210,
+      "▁Ober": 48211,
+      "元整": 48212,
+      "去过": 48213,
+      "月上": 48214,
+      "▁inherit": 48215,
+      "争霸": 48216,
+      "olla": 48217,
+      "▁Rew": 48218,
+      "▁Ethiopia": 48219,
+      "评测": 48220,
+      "大学学历": 48221,
+      "数额较大": 48222,
+      "▁stove": 48223,
+      "athyroid": 48224,
+      "▁renderer": 48225,
+      "▁(“": 48226,
+      "不论是": 48227,
+      "▁bacon": 48228,
+      "公因式": 48229,
+      "FileSync": 48230,
+      "▁aerobic": 48231,
+      "▁kidneys": 48232,
+      "离散": 48233,
+      "▁RUN": 48234,
+      "MEM": 48235,
+      "▁cf": 48236,
+      "文化传媒": 48237,
+      "nFrank": 48238,
+      "▁Revenue": 48239,
+      "▁deceased": 48240,
+      "tRE": 48241,
+      "▁broch": 48242,
+      "▁navbar": 48243,
+      "▁Oslo": 48244,
+      "▁commitments": 48245,
+      "淡水": 48246,
+      "工作时间": 48247,
+      "▁HVAC": 48248,
+      "▁Romanized": 48249,
+      "▁synchronization": 48250,
+      "相较": 48251,
+      "categor": 48252,
+      "▁picnic": 48253,
+      "executor": 48254,
+      "nDefinition": 48255,
+      "▁deliveries": 48256,
+      "一子": 48257,
+      "第二款": 48258,
+      "企事业单位": 48259,
+      "建设有限公司": 48260,
+      "▁Dashboard": 48261,
+      "▁exceptionally": 48262,
+      "$$(": 48263,
+      "otta": 48264,
+      "▁endorsed": 48265,
+      "男士": 48266,
+      "送货": 48267,
+      "对人体": 48268,
+      "美联储": 48269,
+      "▁boiling": 48270,
+      "wm": 48271,
+      "可将": 48272,
+      "uku": 48273,
+      "▁arrows": 48274,
+      "▁weighing": 48275,
+      "▁Resistance": 48276,
+      "食盐": 48277,
+      "接受了": 48278,
+      "imited": 48279,
+      "▁Forex": 48280,
+      "Another": 48281,
+      "▁pylori": 48282,
+      "▁hostname": 48283,
+      "东山": 48284,
+      "发售": 48285,
+      "魔鬼": 48286,
+      "▁BM": 48287,
+      "arte": 48288,
+      "带入": 48289,
+      "VK": 48290,
+      "特产": 48291,
+      "用在": 48292,
+      "然后将": 48293,
+      "alli": 48294,
+      "称赞": 48295,
+      "▁Pray": 48296,
+      "▁Thu": 48297,
+      "教育培训": 48298,
+      "▁dissect": 48299,
+      "员的": 48300,
+      "鲜艳": 48301,
+      "那个人": 48302,
+      "公用": 48303,
+      "lfloor": 48304,
+      "tclass": 48305,
+      "Records": 48306,
+      "▁upside": 48307,
+      "▁english": 48308,
+      "亩产": 48309,
+      "nice": 48310,
+      "xd": 48311,
+      "谁的": 48312,
+      "发展到": 48313,
+      "发病率": 48314,
+      "可行的": 48315,
+      "武功": 48316,
+      "ninstall": 48317,
+      "余款": 48318,
+      "都想": 48319,
+      "keepers": 48320,
+      "▁Miguel": 48321,
+      "法定代理人": 48322,
+      "▁quilt": 48323,
+      "}\",\\": 48324,
+      "成立以来": 48325,
+      "LOCAL": 48326,
+      "▁Isaac": 48327,
+      "▁pulls": 48328,
+      "▁justified": 48329,
+      "DEVELOPMENT": 48330,
+      "sett": 48331,
+      "nus": 48332,
+      "itee": 48333,
+      "▁Fas": 48334,
+      "Party": 48335,
+      "▁lamps": 48336,
+      "▁Hotels": 48337,
+      "八章": 48338,
+      "私立": 48339,
+      "cosx": 48340,
+      "nameof": 48341,
+      "▁extinct": 48342,
+      ">）<": 48343,
+      "厘米的": 48344,
+      "有限的": 48345,
+      "▁Mood": 48346,
+      "▁wells": 48347,
+      "assadors": 48348,
+      "istorical": 48349,
+      "裙子": 48350,
+      "▁QR": 48351,
+      "▁bounce": 48352,
+      "▁irrelevant": 48353,
+      "▁(~": 48354,
+      "Bench": 48355,
+      "▁Prices": 48356,
+      "▁Improvement": 48357,
+      "首家": 48358,
+      "我爱你": 48359,
+      "▁omitted": 48360,
+      "▁Seed": 48361,
+      "西山": 48362,
+      ">\\'\\": 48363,
+      "────": 48364,
+      "申请复议": 48365,
+      "这是一种": 48366,
+      "proving": 48367,
+      "▁Administrator": 48368,
+      "供货": 48369,
+      "闵行": 48370,
+      "毛主席": 48371,
+      "的坐标": 48372,
+      "▁输出格式": 48373,
+      "三条之规定": 48374,
+      "foreign": 48375,
+      "▁Fiction": 48376,
+      "执勤": 48377,
+      "洽谈": 48378,
+      "鸡精": 48379,
+      "dong": 48380,
+      "实践活动": 48381,
+      "▁erad": 48382,
+      "powers": 48383,
+      "terror": 48384,
+      "starting": 48385,
+      "▁shuttle": 48386,
+      "Something": 48387,
+      "▁downward": 48388,
+      "▁negligible": 48389,
+      "▁permutations": 48390,
+      "惩处": 48391,
+      "}$)": 48392,
+      "▁pens": 48393,
+      "物流有限公司": 48394,
+      "▁Laplace": 48395,
+      "▁Dimension": 48396,
+      "▁unfortunate": 48397,
+      "干旱": 48398,
+      "protobuf": 48399,
+      "WIDTH": 48400,
+      "nCommon": 48401,
+      "▁Batman": 48402,
+      "▁brokers": 48403,
+      "征字": 48404,
+      "\\\\\\'": 48405,
+      "▁delegates": 48406,
+      "▁revisions": 48407,
+      "传授": 48408,
+      "坦白": 48409,
+      "egan": 48410,
+      "Steps": 48411,
+      "▁发行日期": 48412,
+      "▁glucocortic": 48413,
+      "以防": 48414,
+      "Fre": 48415,
+      "▁Dob": 48416,
+      "▁PRE": 48417,
+      "▁Proxy": 48418,
+      "▁lubric": 48419,
+      "右上": 48420,
+      "穿搭": 48421,
+      "本案移送": 48422,
+      "▁输入格式": 48423,
+      "▁debuts": 48424,
+      "▁Carroll": 48425,
+      "▁Purpose": 48426,
+      "▁activates": 48427,
+      "▁semantics": 48428,
+      "nTw": 48429,
+      "如果一个": 48430,
+      "成为中国": 48431,
+      "▁theor": 48432,
+      "品位": 48433,
+      "存续": 48434,
+      "柔毛": 48435,
+      "肉质": 48436,
+      "这部剧": 48437,
+      "▁Wear": 48438,
+      "▁pastor": 48439,
+      "▁Balance": 48440,
+      "▁exhibiting": 48441,
+      "alsy": 48442,
+      "▁crowded": 48443,
+      "▁morphine": 48444,
+      "▁medicinal": 48445,
+      "忽悠": 48446,
+      "消极": 48447,
+      "高大": 48448,
+      "人才的": 48449,
+      "▁Accounting": 48450,
+      "核查": 48451,
+      "点半": 48452,
+      "民事执行中": 48453,
+      "nFigure": 48454,
+      "▁grouping": 48455,
+      "▁congestion": 48456,
+      "▁动物": 48457,
+      "ocin": 48458,
+      "▁Renaissance": 48459,
+      "Pan": 48460,
+      "文明的": 48461,
+      "nMiss": 48462,
+      "▁retic": 48463,
+      "encrypt": 48464,
+      "▁deprec": 48465,
+      "城关": 48466,
+      "blade": 48467,
+      "党和": 48468,
+      "二氧化碳": 48469,
+      "大学英语": 48470,
+      "资本市场": 48471,
+      "Polygon": 48472,
+      "eteenth": 48473,
+      "日向本院提出申请": 48474,
+      "赶到": 48475,
+      "节点的": 48476,
+      "▁UAE": 48477,
+      "▁messy": 48478,
+      "厨师": 48479,
+      "同仁": 48480,
+      "学院的": 48481,
+      "工商局": 48482,
+      "软硬件": 48483,
+      "安全管理": 48484,
+      "▁immunoglobulin": 48485,
+      "单品": 48486,
+      "arde": 48487,
+      "▁heel": 48488,
+      "nAsian": 48489,
+      "▁tributary": 48490,
+      "▁BUS": 48491,
+      "nTags": 48492,
+      "▁dumb": 48493,
+      "自诉": 48494,
+      "精神和": 48495,
+      "个体工商户": 48496,
+      "▁hourly": 48497,
+      "Utilities": 48498,
+      "看我": 48499,
+      "Cop": 48500,
+      "就算了": 48501,
+      "金华市": 48502,
+      "skill": 48503,
+      "Pressed": 48504,
+      "▁layered": 48505,
+      "▁accumulate": 48506,
+      "的超": 48507,
+      "纠错": 48508,
+      "nAnal": 48509,
+      "▁flesh": 48510,
+      "▁Speech": 48511,
+      "主料": 48512,
+      "efe": 48513,
+      "hg": 48514,
+      "民字第": 48515,
+      "▁debates": 48516,
+      "▁Inspector": 48517,
+      "打磨": 48518,
+      "元减半收取": 48519,
+      "▁Monster": 48520,
+      "四人": 48521,
+      "闹钟": 48522,
+      "过滤器": 48523,
+      "写在": 48524,
+      "所有人都": 48525,
+      "ITHUB": 48526,
+      "▁dick": 48527,
+      "February": 48528,
+      "▁enjoyment": 48529,
+      "▁explosive": 48530,
+      "碱性": 48531,
+      "▁Liquid": 48532,
+      "▁enamel": 48533,
+      "▁Elements": 48534,
+      "编排": 48535,
+      "ERSION": 48536,
+      "▁Lecture": 48537,
+      "临界": 48538,
+      "流氓": 48539,
+      "可视化": 48540,
+      "的感情": 48541,
+      "▁Copper": 48542,
+      "精装": 48543,
+      "lys": 48544,
+      "}/{": 48545,
+      "Hour": 48546,
+      "核酸检测": 48547,
+      "▁Drew": 48548,
+      "nVisit": 48549,
+      "值得一提的是": 48550,
+      "后被": 48551,
+      "日趋": 48552,
+      "▁tendon": 48553,
+      "Presenter": 48554,
+      "Annotations": 48555,
+      "大腿": 48556,
+      "荔枝": 48557,
+      "土地上": 48558,
+      "的习惯": 48559,
+      "▁depicted": 48560,
+      "▁livestock": 48561,
+      "▁Sacramento": 48562,
+      "★★": 48563,
+      "平均数": 48564,
+      "▁Bulld": 48565,
+      "▁Gujar": 48566,
+      "infection": 48567,
+      "合伙人": 48568,
+      "▁microtub": 48569,
+      "气球": 48570,
+      "缓期": 48571,
+      "▁Xen": 48572,
+      "Distribution": 48573,
+      "义词": 48574,
+      "字号": 48575,
+      "拿了": 48576,
+      "一切都": 48577,
+      "大家好": 48578,
+      "住湖南省": 48579,
+      "▁EDIT": 48580,
+      "Nothing": 48581,
+      "端午": 48582,
+      "ILS": 48583,
+      "inho": 48584,
+      "blogs": 48585,
+      "ycles": 48586,
+      "Ox": 48587,
+      "赶快": 48588,
+      "文明建设": 48589,
+      "itches": 48590,
+      "levels": 48591,
+      "imetric": 48592,
+      "典雅": 48593,
+      "采取了": 48594,
+      "episode": 48595,
+      "▁protagon": 48596,
+      "古老": 48597,
+      "石墨": 48598,
+      "使用者": 48599,
+      "▁apo": 48600,
+      "▁Berry": 48601,
+      "flatten": 48602,
+      "▁listened": 48603,
+      "▁securely": 48604,
+      "理学": 48605,
+      "离合": 48606,
+      "紊乱": 48607,
+      "顺应": 48608,
+      "他表示": 48609,
+      "可以做": 48610,
+      "可直接": 48611,
+      ")\")\\": 48612,
+      "edic": 48613,
+      "▁MON": 48614,
+      "▁voter": 48615,
+      "▁Important": 48616,
+      "▁resolving": 48617,
+      "兴隆": 48618,
+      "Mean": 48619,
+      "▁AST": 48620,
+      "FALSE": 48621,
+      "布斯": 48622,
+      "膝盖": 48623,
+      "自营": 48624,
+      "▁Pizza": 48625,
+      "▁synchronized": 48626,
+      "▁法": 48627,
+      "接纳": 48628,
+      "▁};": 48629,
+      "Been": 48630,
+      "▁rhs": 48631,
+      "共产党员": 48632,
+      "developers": 48633,
+      "石灰": 48634,
+      "以上就是": 48635,
+      "步步": 48636,
+      "这才": 48637,
+      "tfinal": 48638,
+      "第一百五十四条第一款第十一项": 48639,
+      "底蕴": 48640,
+      "林果": 48641,
+      "▁Westminster": 48642,
+      "埔寨": 48643,
+      "nPhot": 48644,
+      "▁Mesh": 48645,
+      "▁Reset": 48646,
+      "▁Indigenous": 48647,
+      "江北": 48648,
+      "知名的": 48649,
+      "▁SSD": 48650,
+      "cleaned": 48651,
+      "字节": 48652,
+      "美德": 48653,
+      "Wave": 48654,
+      "subsection": 48655,
+      "一言": 48656,
+      "敬请": 48657,
+      "扣押其": 48658,
+      "Hide": 48659,
+      "ilog": 48660,
+      "LEMENT": 48661,
+      "idable": 48662,
+      "▁civic": 48663,
+      "ococcal": 48664,
+      "制定了": 48665,
+      "短期内": 48666,
+      "▁Mueller": 48667,
+      "楼市": 48668,
+      "被骗": 48669,
+      "ono": 48670,
+      "lact": 48671,
+      "▁Yemen": 48672,
+      "段的": 48673,
+      "viz": 48674,
+      "bola": 48675,
+      "▁replaces": 48676,
+      "Robert": 48677,
+      "▁Quest": 48678,
+      "▁Caucas": 48679,
+      "Starting": 48680,
+      "▁endurance": 48681,
+      "海事": 48682,
+      "modify": 48683,
+      "▁sewing": 48684,
+      "мо": 48685,
+      "突击": 48686,
+      "罕见": 48687,
+      "宝宝的": 48688,
+      "本院依照": 48689,
+      "硬盘容量": 48690,
+      "▁Guarant": 48691,
+      "▁billions": 48692,
+      "洗浴": 48693,
+      "疏松": 48694,
+      "政策和": 48695,
+      "如果你是": 48696,
+      "强制措施": 48697,
+      "本月": 48698,
+      "花样": 48699,
+      "公元前": 48700,
+      "nEst": 48701,
+      "坐着": 48702,
+      "这个词": 48703,
+      "▁连载网站": 48704,
+      "▁Claire": 48705,
+      "▁remedy": 48706,
+      "▁Kenneth": 48707,
+      "▁plotting": 48708,
+      "▁reproduced": 48709,
+      "解说": 48710,
+      "▁();\\": 48711,
+      "歷史": 48712,
+      "白头": 48713,
+      "防汛": 48714,
+      "的人来说": 48715,
+      "直辖": 48716,
+      "▁Modi": 48717,
+      "Computer": 48718,
+      "▁terminus": 48719,
+      "▁Configure": 48720,
+      "▁根据": 48721,
+      "nPRO": 48722,
+      "formats": 48723,
+      "Rh": 48724,
+      "的供述": 48725,
+      "▁advertisements": 48726,
+      "▁mr": 48727,
+      "及应用": 48728,
+      "nQue": 48729,
+      "startup": 48730,
+      "PNG": 48731,
+      "dpi": 48732,
+      "nLand": 48733,
+      "▁Peer": 48734,
+      "setTimeout": 48735,
+      "王的": 48736,
+      "▁ppm": 48737,
+      "Operating": 48738,
+      "药用": 48739,
+      "在未来": 48740,
+      "▁fragile": 48741,
+      "区人民政府": 48742,
+      "ortical": 48743,
+      "ég": 48744,
+      "思想的": 48745,
+      "▁disjoint": 48746,
+      "两端": 48747,
+      "▁ICT": 48748,
+      "纪检监察": 48749,
+      "▁emitted": 48750,
+      "umberland": 48751,
+      "影响力的": 48752,
+      "nDutch": 48753,
+      "▁vanity": 48754,
+      "▁backbone": 48755,
+      "▁Typically": 48756,
+      "▁summation": 48757,
+      "能看到": 48758,
+      "▁mappings": 48759,
+      "▁reflective": 48760,
+      "▁到": 48761,
+      "财产刑": 48762,
+      "▁Dani": 48763,
+      "▁coch": 48764,
+      "合击": 48765,
+      "cpy": 48766,
+      "第二种": 48767,
+      "▁hats": 48768,
+      "probably": 48769,
+      "碳酸": 48770,
+      "保修卡": 48771,
+      "ewish": 48772,
+      "▁songwriter": 48773,
+      "邂逅": 48774,
+      "身份证号": 48775,
+      "askets": 48776,
+      "teacher": 48777,
+      "▁memoir": 48778,
+      "▁writings": 48779,
+      "▁CHAR": 48780,
+      "closures": 48781,
+      "KM": 48782,
+      "▁Arr": 48783,
+      "▁rhet": 48784,
+      "▁艾": 48785,
+      "体型": 48786,
+      "萃取": 48787,
+      "骨质": 48788,
+      "▁Drama": 48789,
+      "▁panor": 48790,
+      "▁repeats": 48791,
+      "▁secretory": 48792,
+      "▁antidepress": 48793,
+      "细化": 48794,
+      "追缴": 48795,
+      "▁NW": 48796,
+      "▁地点": 48797,
+      "partum": 48798,
+      "▁Meyer": 48799,
+      "但对": 48800,
+      "收支": 48801,
+      "ntz": 48802,
+      "解题思路": 48803,
+      "想必": 48804,
+      "BOT": 48805,
+      "}))\\": 48806,
+      "cis": 48807,
+      "Gate": 48808,
+      "ophen": 48809,
+      "▁Hern": 48810,
+      "▁Owner": 48811,
+      "▁alcoholic": 48812,
+      "▁olfactory": 48813,
+      "▁originating": 48814,
+      "*{": 48815,
+      "参数的": 48816,
+      "输入法": 48817,
+      "ayed": 48818,
+      "▁Phen": 48819,
+      "父子": 48820,
+      "Fund": 48821,
+      "信息咨询": 48822,
+      "或直接向": 48823,
+      "Indicator": 48824,
+      "▁defended": 48825,
+      "▁eliminates": 48826,
+      "indle": 48827,
+      "teams": 48828,
+      "ainless": 48829,
+      "▁imagined": 48830,
+      "▁playlist": 48831,
+      "▁charities": 48832,
+      "▁clarification": 48833,
+      "的房屋": 48834,
+      "▁Coc": 48835,
+      "▁loos": 48836,
+      "minton": 48837,
+      "▁glaucoma": 48838,
+      "保洁": 48839,
+      "allic": 48840,
+      "▁Richardson": 48841,
+      "嫉妒": 48842,
+      "前所未": 48843,
+      "收到了": 48844,
+      "日因涉嫌": 48845,
+      "▁chase": 48846,
+      "Through": 48847,
+      "▁creamy": 48848,
+      "气流": 48849,
+      "▁Iraqi": 48850,
+      "▁villa": 48851,
+      "利害": 48852,
+      "▁glimp": 48853,
+      "劳人": 48854,
+      "和县": 48855,
+      "意义上": 48856,
+      "新民": 48857,
+      "难的": 48858,
+      "stmt": 48859,
+      "倾听": 48860,
+      "家装": 48861,
+      "撞击": 48862,
+      "重叠": 48863,
+      "顺畅": 48864,
+      "iffe": 48865,
+      "Slider": 48866,
+      "▁coached": 48867,
+      "杀害": 48868,
+      "▁analogues": 48869,
+      "亲身": 48870,
+      "随手": 48871,
+      "本次活动": 48872,
+      "pliers": 48873,
+      "▁lively": 48874,
+      "▁knocked": 48875,
+      "▁diamonds": 48876,
+      "奴隶": 48877,
+      "▁Kaw": 48878,
+      "▁Fergus": 48879,
+      "▁Chronic": 48880,
+      "$$)": 48881,
+      "▁/\\": 48882,
+      "为导向": 48883,
+      "▁Tong": 48884,
+      "nJournal": 48885,
+      "screenshot": 48886,
+      "▁neutrophils": 48887,
+      "少先": 48888,
+      "而有": 48889,
+      "的新型": 48890,
+      "红楼梦": 48891,
+      "舒适的": 48892,
+      "enzie": 48893,
+      "▁Actions": 48894,
+      "▁ASP": 48895,
+      "咨询医生": 48896,
+      "合同的": 48897,
+      "柬埔寨": 48898,
+      "▁borough": 48899,
+      "看看吧": 48900,
+      "nspec": 48901,
+      "低级": 48902,
+      "米尔": 48903,
+      "京剧": 48904,
+      "护栏": 48905,
+      "ATIVE": 48906,
+      "andez": 48907,
+      "▁cafe": 48908,
+      "▁Amanda": 48909,
+      "▁glimpse": 48910,
+      "向其": 48911,
+      "故依照": 48912,
+      "▁Odd": 48913,
+      "▁[{\\": 48914,
+      "不予支持": 48915,
+      "blood": 48916,
+      "▁choir": 48917,
+      "▁Buenos": 48918,
+      "▁dancer": 48919,
+      "掌声": 48920,
+      "概论": 48921,
+      "防盗": 48922,
+      "在没有": 48923,
+      "}/${": 48924,
+      "二次根式": 48925,
+      "Rotation": 48926,
+      "▁messenger": 48927,
+      "iste": 48928,
+      "▁spoon": 48929,
+      "▁wreck": 48930,
+      "▁appeals": 48931,
+      "▁beginners": 48932,
+      "的努力": 48933,
+      "Skip": 48934,
+      "申请执行人刘": 48935,
+      "▁immuno": 48936,
+      "▁caffeine": 48937,
+      "▁chickens": 48938,
+      "▁stimulates": 48939,
+      "溶性": 48940,
+      "▁wip": 48941,
+      "Friend": 48942,
+      "▁depot": 48943,
+      "▁drawer": 48944,
+      "According": 48945,
+      "界限": 48946,
+      "nOK": 48947,
+      "Square": 48948,
+      "enhanced": 48949,
+      "意为": 48950,
+      "预热": 48951,
+      "带领下": 48952,
+      "日期间": 48953,
+      "rounded": 48954,
+      "readlines": 48955,
+      "山顶": 48956,
+      "Tur": 48957,
+      "▁音乐": 48958,
+      "每一天": 48959,
+      "▁Estonia": 48960,
+      "力气": 48961,
+      "不具有": 48962,
+      "Blob": 48963,
+      "▁mould": 48964,
+      "南充": 48965,
+      "坚果": 48966,
+      "▁Ek": 48967,
+      "多少人": 48968,
+      "颁发的": 48969,
+      "remember": 48970,
+      "▁predictable": 48971,
+      "火热": 48972,
+      "的工资": 48973,
+      "▁mascul": 48974,
+      "firstName": 48975,
+      "在我看来": 48976,
+      "opent": 48977,
+      "coffee": 48978,
+      "▁citations": 48979,
+      "北极": 48980,
+      "洛杉": 48981,
+      "都得": 48982,
+      "▁dd": 48983,
+      "但还是": 48984,
+      "告诉我们": 48985,
+      "▁textile": 48986,
+      "岛屿": 48987,
+      "痘痘": 48988,
+      "▁Lav": 48989,
+      "▁Sig": 48990,
+      "▁zebra": 48991,
+      "▁Hospit": 48992,
+      "▁spices": 48993,
+      "▁obstructive": 48994,
+      "不受申请执行时效期间的限制": 48995,
+      "得太": 48996,
+      "高管": 48997,
+      "▁frost": 48998,
+      "▁Fortunately": 48999,
+      "状元": 49000,
+      "不平衡": 49001,
+      "绿化率": 49002,
+      "▁Pharmac": 49003,
+      "ributable": 49004,
+      "红花": 49005,
+      "oren": 49006,
+      "▁insult": 49007,
+      "▁partitions": 49008,
+      "itzer": 49009,
+      "▁WORK": 49010,
+      "uti": 49011,
+      "▁Occup": 49012,
+      "covered": 49013,
+      "尼克": 49014,
+      "山脉": 49015,
+      "应是": 49016,
+      "清淡": 49017,
+      "▁NET": 49018,
+      "nLine": 49019,
+      "▁hasht": 49020,
+      "designated": 49021,
+      "▁disasters": 49022,
+      "了我的": 49023,
+      "▁genital": 49024,
+      "▁BJP": 49025,
+      "▁Vil": 49026,
+      "Assign": 49027,
+      "iry": 49028,
+      "停留在": 49029,
+      "('\\\\": 49030,
+      "▁optimistic": 49031,
+      "我不想": 49032,
+      "flip": 49033,
+      "▁Morr": 49034,
+      "▁pend": 49035,
+      "nadmin": 49036,
+      "▁Utility": 49037,
+      "声誉": 49038,
+      "车内": 49039,
+      "▁存储": 49040,
+      "▁dull": 49041,
+      "▁fing": 49042,
+      "_)": 49043,
+      "屏障": 49044,
+      "往上": 49045,
+      "高产": 49046,
+      "二〇一六年八月": 49047,
+      "私营": 49048,
+      "INAL": 49049,
+      "▁Ski": 49050,
+      "▁Classes": 49051,
+      "▁polling": 49052,
+      "▁Discrete": 49053,
+      "▁canceled": 49054,
+      "▁neutrophil": 49055,
+      "#:": 49056,
+      "JR": 49057,
+      "▁圣": 49058,
+      "出版时间": 49059,
+      "▁polo": 49060,
+      "中国人民解放军": 49061,
+      "国有企业": 49062,
+      "▁Persian": 49063,
+      "▁welding": 49064,
+      "谁是": 49065,
+      "▁*,": 49066,
+      "中的一": 49067,
+      "▁eyel": 49068,
+      "oultry": 49069,
+      "▁Barnes": 49070,
+      "▁blended": 49071,
+      "orientation": 49072,
+      "调剂": 49073,
+      "鼓舞": 49074,
+      "精确到": 49075,
+      "▁Americas": 49076,
+      "▁aperture": 49077,
+      "▁ε": 49078,
+      "房东": 49079,
+      "让它": 49080,
+      "信息服务": 49081,
+      "▁ubiquitous": 49082,
+      "和建议": 49083,
+      "故本案": 49084,
+      "高效的": 49085,
+      "意义上的": 49086,
+      "成形": 49087,
+      "PAT": 49088,
+      "因此在": 49089,
+      "daemon": 49090,
+      "战绩": 49091,
+      "▁Gender": 49092,
+      "▁Slovenia": 49093,
+      "原地": 49094,
+      "创新型": 49095,
+      "由已知": 49096,
+      "hbar": 49097,
+      "rfloor": 49098,
+      "▁Memphis": 49099,
+      "▁attained": 49100,
+      "微波炉": 49101,
+      ">',\\": 49102,
+      "esty": 49103,
+      "▁Rhe": 49104,
+      "▁Away": 49105,
+      "节课": 49106,
+      "马里": 49107,
+      "和完善": 49108,
+      "nParticip": 49109,
+      "▁observers": 49110,
+      "arb": 49111,
+      "你现在": 49112,
+      "注射液": 49113,
+      "▁Neo": 49114,
+      "cleanup": 49115,
+      "▁Canvas": 49116,
+      "▁Exhibition": 49117,
+      "▁hypertrophy": 49118,
+      "求法": 49119,
+      "▁Cycle": 49120,
+      "▁opted": 49121,
+      "属于中档题": 49122,
+      "▁marsh": 49123,
+      "▁forecasting": 49124,
+      "主义者": 49125,
+      "的整体": 49126,
+      "nMary": 49127,
+      "▁hardest": 49128,
+      "▁ecosystems": 49129,
+      "年终": 49130,
+      "rax": 49131,
+      "命令行": 49132,
+      "▁Dawn": 49133,
+      "\":[],\"": 49134,
+      "▁Routes": 49135,
+      "浑身": 49136,
+      "▁Sau": 49137,
+      "▁Swe": 49138,
+      "野生动物": 49139,
+      "ceans": 49140,
+      "ensed": 49141,
+      "划拨被执行人": 49142,
+      "▁commence": 49143,
+      "差分": 49144,
+      "▁об": 49145,
+      "不忘初心": 49146,
+      "Manifest": 49147,
+      "棕色": 49148,
+      "双人间": 49149,
+      "CASE": 49150,
+      "▁Cul": 49151,
+      "▁vow": 49152,
+      "▁catar": 49153,
+      "strained": 49154,
+      "repreneur": 49155,
+      "▁typename": 49156,
+      "▁displaced": 49157,
+      "agrangian": 49158,
+      "nOriginal": 49159,
+      "金的": 49160,
+      "元素的": 49161,
+      "公安部": 49162,
+      "包含一个": 49163,
+      "▁Ellen": 49164,
+      "▁garner": 49165,
+      "有效期": 49166,
+      "▁carn": 49167,
+      "▁autoc": 49168,
+      "▁primer": 49169,
+      "Kg": 49170,
+      "几句": 49171,
+      "▁KeyError": 49172,
+      "变现": 49173,
+      "庞大": 49174,
+      "给被告": 49175,
+      "▁beers": 49176,
+      "▁detox": 49177,
+      "▁StringBuilder": 49178,
+      "▁NSA": 49179,
+      "Motor": 49180,
+      "embedded": 49181,
+      "proportion": 49182,
+      "间断": 49183,
+      "作出裁定": 49184,
+      "本身就是": 49185,
+      "血液循环": 49186,
+      "executable": 49187,
+      "领先的": 49188,
+      "apro": 49189,
+      "Magic": 49190,
+      "▁scanf": 49191,
+      "▁locksmith": 49192,
+      "▁registering": 49193,
+      "jd": 49194,
+      "上部": 49195,
+      "真情": 49196,
+      "Sleep": 49197,
+      "▁multis": 49198,
+      "▁chopped": 49199,
+      "号称": 49200,
+      "ilis": 49201,
+      "明显高于": 49202,
+      "Literal": 49203,
+      "architecture": 49204,
+      "▁下": 49205,
+      "不一致": 49206,
+      "vens": 49207,
+      "北斗": 49208,
+      "婷婷": 49209,
+      "conc": 49210,
+      "axial": 49211,
+      "exter": 49212,
+      "▁audi": 49213,
+      "▁Shield": 49214,
+      "用书": 49215,
+      "geries": 49216,
+      "▁Highland": 49217,
+      "▁В": 49218,
+      "交集": 49219,
+      "Fig": 49220,
+      "tur": 49221,
+      "rosse": 49222,
+      "▁Pixel": 49223,
+      "iropractic": 49224,
+      "▁reproducible": 49225,
+      "年末": 49226,
+      "提要": 49227,
+      "都是有": 49228,
+      "进一步提高": 49229,
+      "▁longtime": 49230,
+      "嘲笑": 49231,
+      "纸上": 49232,
+      "椭圆的": 49233,
+      "等行业": 49234,
+      "Shadow": 49235,
+      "▁collaborated": 49236,
+      "lane": 49237,
+      "olone": 49238,
+      "riages": 49239,
+      "▁blogger": 49240,
+      "ufficiency": 49241,
+      "或缺": 49242,
+      "沿线": 49243,
+      "需在": 49244,
+      "PIN": 49245,
+      "▁Cs": 49246,
+      "流产": 49247,
+      "一些人": 49248,
+      "倾城": 49249,
+      "▁Onc": 49250,
+      "停车服务": 49251,
+      "nArgent": 49252,
+      "▁Markets": 49253,
+      "rav": 49254,
+      "抗氧化": 49255,
+      "(\"\");\\": 49256,
+      "▁hunger": 49257,
+      "▁posture": 49258,
+      "▁Pipeline": 49259,
+      "出锅": 49260,
+      "友人": 49261,
+      "▁Santiago": 49262,
+      "▁wheelchair": 49263,
+      "oye": 49264,
+      "上发表": 49265,
+      "不是说": 49266,
+      "须经": 49267,
+      "ystone": 49268,
+      "▁comfortably": 49269,
+      "典范": 49270,
+      "晓明": 49271,
+      "渊博": 49272,
+      "Wire": 49273,
+      "nTeam": 49274,
+      "▁ignor": 49275,
+      "Positive": 49276,
+      "▁terminate": 49277,
+      "vels": 49278,
+      "比较好的": 49279,
+      "中西医结合": 49280,
+      "▁Claim": 49281,
+      "▁legislature": 49282,
+      "修补": 49283,
+      "须知": 49284,
+      "▁Boat": 49285,
+      "▁Glenn": 49286,
+      "▁grief": 49287,
+      "▁Instrument": 49288,
+      "]^": 49289,
+      "磁场": 49290,
+      "结果是": 49291,
+      "ctomy": 49292,
+      "▁nause": 49293,
+      "▁surrender": 49294,
+      "▁delightful": 49295,
+      "狠狠": 49296,
+      "讲的": 49297,
+      "▁Seq": 49298,
+      "nDeaths": 49299,
+      "▁Cornwall": 49300,
+      "▁blown": 49301,
+      "constraint": 49302,
+      "上个": 49303,
+      "天数": 49304,
+      "率领": 49305,
+      "▁declining": 49306,
+      "ISBN": 49307,
+      "nStarting": 49308,
+      "等价": 49309,
+      "▁tutor": 49310,
+      "▁ligament": 49311,
+      "也无": 49312,
+      "fluor": 49313,
+      "▁Structural": 49314,
+      "现实的": 49315,
+      "▁经营状态": 49316,
+      "profiles": 49317,
+      "东侧": 49318,
+      "帅气": 49319,
+      "想起来": 49320,
+      "▁CUR": 49321,
+      "京都": 49322,
+      "mmm": 49323,
+      "抗击": 49324,
+      "▁fucked": 49325,
+      "nBeetles": 49326,
+      "符合法定减刑条件": 49327,
+      "宽敞": 49328,
+      "perl": 49329,
+      "horse": 49330,
+      "DataType": 49331,
+      "▁戴": 49332,
+      "来访": 49333,
+      "等候": 49334,
+      "铅笔": 49335,
+      "我以为": 49336,
+      "▁Joshua": 49337,
+      "▁Mapping": 49338,
+      "交友": 49339,
+      "▁四川": 49340,
+      "▁PKC": 49341,
+      "▁Cisco": 49342,
+      "/`": 49343,
+      "开局": 49344,
+      "血栓": 49345,
+      "biz": 49346,
+      "中的一个": 49347,
+      "▁spends": 49348,
+      "postgresql": 49349,
+      "凡是": 49350,
+      "tum": 49351,
+      "▁RP": 49352,
+      "reed": 49353,
+      "▁Leslie": 49354,
+      "▁waited": 49355,
+      "▁Editorial": 49356,
+      "温室": 49357,
+      "了一点": 49358,
+      "科科长": 49359,
+      "墙壁": 49360,
+      "时应": 49361,
+      "就在于": 49362,
+      "nLive": 49363,
+      "▁Desk": 49364,
+      "better": 49365,
+      "▁Fusion": 49366,
+      "上人": 49367,
+      "▁rc": 49368,
+      "▁Boost": 49369,
+      "gettext": 49370,
+      "▁princess": 49371,
+      "▁斯": 49372,
+      "卫东": 49373,
+      "的她": 49374,
+      "▁Fake": 49375,
+      "▁Emirates": 49376,
+      "的车辆": 49377,
+      "建筑设计": 49378,
+      "nQual": 49379,
+      "填报": 49380,
+      "尚可": 49381,
+      "度高": 49382,
+      "日被告": 49383,
+      "本裁定书送达后立即执行": 49384,
+      "▁comprehension": 49385,
+      "北上": 49386,
+      "的重量": 49387,
+      "转换为": 49388,
+      "工程设计": 49389,
+      "▁Shower": 49390,
+      "▁Hardware": 49391,
+      "▁titanium": 49392,
+      "än": 49393,
+      "▁\"#{": 49394,
+      "不能提供": 49395,
+      "▁Para": 49396,
+      "Configure": 49397,
+      "▁attracting": 49398,
+      "▁peritoneal": 49399,
+      "围棋": 49400,
+      "推测": 49401,
+      "▁Ey": 49402,
+      "colored": 49403,
+      "molecular": 49404,
+      "为国家": 49405,
+      "ceral": 49406,
+      "▁argu": 49407,
+      "相关法律规定": 49408,
+      "▁rooted": 49409,
+      "高水平": 49410,
+      "▁Seminar": 49411,
+      "▁coatings": 49412,
+      "快要": 49413,
+      "杰克": 49414,
+      "金奖": 49415,
+      "ijn": 49416,
+      "计算方法": 49417,
+      "izarre": 49418,
+      "红酒": 49419,
+      "霸王": 49420,
+      "quisites": 49421,
+      "位居": 49422,
+      "lez": 49423,
+      "iani": 49424,
+      "▁COPD": 49425,
+      "LOGGER": 49426,
+      "Destination": 49427,
+      "▁announcing": 49428,
+      "▁异": 49429,
+      "披针": 49430,
+      "ikk": 49431,
+      "}`\\": 49432,
+      "▁[:": 49433,
+      "anan": 49434,
+      "▁repay": 49435,
+      "▁tensions": 49436,
+      "擦拭": 49437,
+      "antis": 49438,
+      "nfind": 49439,
+      "山人": 49440,
+      "巴克": 49441,
+      "项的": 49442,
+      "Union": 49443,
+      "▁kicks": 49444,
+      "时会": 49445,
+      "根源": 49446,
+      "生生": 49447,
+      "advert": 49448,
+      "linking": 49449,
+      "▁relieve": 49450,
+      "▁recession": 49451,
+      "MH": 49452,
+      "末端": 49453,
+      "必要性": 49454,
+      "第一种": 49455,
+      "inki": 49456,
+      "▁homolog": 49457,
+      "存的": 49458,
+      "etches": 49459,
+      "findOne": 49460,
+      "nChoose": 49461,
+      "▁dialect": 49462,
+      "▁ophthal": 49463,
+      "▁Bangalore": 49464,
+      "格拉": 49465,
+      "男孩子": 49466,
+      "iris": 49467,
+      "▁Foods": 49468,
+      "评判": 49469,
+      "Trip": 49470,
+      "▁Cake": 49471,
+      "▁Gerald": 49472,
+      "lpatterns": 49473,
+      "▁overload": 49474,
+      "螺纹": 49475,
+      "科研成果": 49476,
+      "nRecipients": 49477,
+      "得意": 49478,
+      "村内": 49479,
+      "特区": 49480,
+      "成功率": 49481,
+      "▁pulp": 49482,
+      "Receiver": 49483,
+      "▁pointers": 49484,
+      "往返": 49485,
+      "网盘": 49486,
+      "iesel": 49487,
+      "kinson": 49488,
+      "▁Cuban": 49489,
+      "号刑事附带民事": 49490,
+      "▁monocytes": 49491,
+      "按下": 49492,
+      "ahi": 49493,
+      "oors": 49494,
+      "食品药品": 49495,
+      "▁ging": 49496,
+      "voltage": 49497,
+      "台中": 49498,
+      "了一套": 49499,
+      "umbia": 49500,
+      "生了": 49501,
+      "▁Gren": 49502,
+      "▁数据范围": 49503,
+      "▁pitched": 49504,
+      "▁thoughtful": 49505,
+      "主频": 49506,
+      "朝廷": 49507,
+      "▁Mist": 49508,
+      "Clause": 49509,
+      "局域": 49510,
+      "Know": 49511,
+      "和我们": 49512,
+      "可以申请": 49513,
+      "qualified": 49514,
+      "就是你": 49515,
+      "ierre": 49516,
+      "海棠": 49517,
+      "菠萝": 49518,
+      "研究对象": 49519,
+      "church": 49520,
+      "▁colspan": 49521,
+      "▁Karnataka": 49522,
+      "[_": 49523,
+      "Easy": 49524,
+      "astom": 49525,
+      "pletely": 49526,
+      "prisingly": 49527,
+      "画质": 49528,
+      "软骨": 49529,
+      "我一个": 49530,
+      "▁Flying": 49531,
+      "▁commerce": 49532,
+      "有感": 49533,
+      "沙拉": 49534,
+      "几分钟": 49535,
+      "▁Nissan": 49536,
+      "▁interstitial": 49537,
+      "▁nanot": 49538,
+      "▁undoubtedly": 49539,
+      "上千": 49540,
+      "偿付": 49541,
+      "四是": 49542,
+      "回升": 49543,
+      "通关": 49544,
+      "无明显": 49545,
+      "▁handic": 49546,
+      "KY": 49547,
+      "trait": 49548,
+      "propTypes": 49549,
+      "interactive": 49550,
+      "▁disappointing": 49551,
+      "的性能": 49552,
+      "▁splash": 49553,
+      "▁comorbid": 49554,
+      "▁spotlight": 49555,
+      "tList": 49556,
+      "中国邮政储蓄银行股份有限公司": 49557,
+      "店面": 49558,
+      "曾用": 49559,
+      "rats": 49560,
+      "总和": 49561,
+      "过硬": 49562,
+      "(((": 49563,
+      "mAh": 49564,
+      "▁Xia": 49565,
+      "▁Dong": 49566,
+      "财产的规定": 49567,
+      "▁Writer": 49568,
+      "heduling": 49569,
+      "▁Bloomberg": 49570,
+      "扩建": 49571,
+      "西县": 49572,
+      "同底数": 49573,
+      "▁Luxem": 49574,
+      "nMusical": 49575,
+      "wc": 49576,
+      "风俗": 49577,
+      "怎么能": 49578,
+      "tcontinue": 49579,
+      "写入": 49580,
+      "数千": 49581,
+      "闪烁": 49582,
+      "时间是": 49583,
+      "洛杉矶": 49584,
+      "ereg": 49585,
+      "colum": 49586,
+      "cephal": 49587,
+      "全能": 49588,
+      "旗帜": 49589,
+      "<>(": 49590,
+      "在同一": 49591,
+      "▁aforementioned": 49592,
+      "喇叭": 49593,
+      "录用": 49594,
+      "惊人": 49595,
+      "鉴定意见": 49596,
+      "▁parentheses": 49597,
+      "王世": 49598,
+      "支付货款": 49599,
+      "nDay": 49600,
+      "上诉人的": 49601,
+      "住所地北京市": 49602,
+      "▁endfor": 49603,
+      "internet": 49604,
+      "排气": 49605,
+      "毛孔": 49606,
+      "▁nb": 49607,
+      "▁со": 49608,
+      "▁flick": 49609,
+      "初恋": 49610,
+      "王者荣耀": 49611,
+      "orate": 49612,
+      "ourmet": 49613,
+      "势头": 49614,
+      "cellent": 49615,
+      "强制性": 49616,
+      "▁IDs": 49617,
+      "▁outper": 49618,
+      "Introduction": 49619,
+      "▁bb": 49620,
+      "在里面": 49621,
+      "我不会": 49622,
+      "旅游业": 49623,
+      "toxic": 49624,
+      "▁corticoster": 49625,
+      "交替": 49626,
+      "南阳": 49627,
+      "点缀": 49628,
+      "opo": 49629,
+      "slow": 49630,
+      "▁Muk": 49631,
+      "Robot": 49632,
+      "ramer": 49633,
+      "Millis": 49634,
+      "蕴含": 49635,
+      "大家都知道": 49636,
+      "▁capita": 49637,
+      "石膏": 49638,
+      "nJeff": 49639,
+      "▁Kurt": 49640,
+      "平装": 49641,
+      "田径": 49642,
+      "喜欢你": 49643,
+      "▁suspicion": 49644,
+      "JA": 49645,
+      "不再是": 49646,
+      "oload": 49647,
+      "▁Clock": 49648,
+      "▁undes": 49649,
+      "▁brakes": 49650,
+      "排在": 49651,
+      "降水": 49652,
+      "▁fox": 49653,
+      "万个": 49654,
+      "物证": 49655,
+      "▁Lauren": 49656,
+      "▁Mechanics": 49657,
+      "Ng": 49658,
+      "文集": 49659,
+      "Ball": 49660,
+      "认真学习": 49661,
+      "bestos": 49662,
+      "▁Dating": 49663,
+      "arthritis": 49664,
+      "脸色": 49665,
+      "ndefine": 49666,
+      "本草": 49667,
+      "hui": 49668,
+      "▁已注销": 49669,
+      "ADMIN": 49670,
+      "▁Poetry": 49671,
+      "LCD": 49672,
+      "OVA": 49673,
+      "充电器": 49674,
+      "▁lottery": 49675,
+      "丹麦": 49676,
+      "五官": 49677,
+      "利用率": 49678,
+      "业内人士": 49679,
+      "▁Putin": 49680,
+      "▁Armstrong": 49681,
+      "宝藏": 49682,
+      "意识的": 49683,
+      "Templates": 49684,
+      "▁veterinary": 49685,
+      "▁Observatory": 49686,
+      "▁PY": 49687,
+      "掌握了": 49688,
+      "▁**[": 49689,
+      "产业结构": 49690,
+      "▁amazed": 49691,
+      "▁sticking": 49692,
+      "difference": 49693,
+      "·\\": 49694,
+      "为己": 49695,
+      "医用": 49696,
+      "yson": 49697,
+      "列出方程": 49698,
+      "okine": 49699,
+      "化管理": 49700,
+      "▁chalk": 49701,
+      "▁encodes": 49702,
+      "▁unpredict": 49703,
+      "STA": 49704,
+      "最适合": 49705,
+      "itals": 49706,
+      "cookies": 49707,
+      "文具": 49708,
+      "执民字第": 49709,
+      "exact": 49710,
+      "oracle": 49711,
+      "货架": 49712,
+      "▁apical": 49713,
+      "▁relocated": 49714,
+      "ABS": 49715,
+      "罪犯改造": 49716,
+      "nucle": 49717,
+      "hadoop": 49718,
+      "理论的": 49719,
+      "enne": 49720,
+      "这部电影": 49721,
+      "nAccess": 49722,
+      "▁unaffected": 49723,
+      "单片": 49724,
+      "但是你": 49725,
+      "▁Tet": 49726,
+      "个人信息": 49727,
+      "▁staged": 49728,
+      "▁radioactive": 49729,
+      "缘分": 49730,
+      "nLaw": 49731,
+      "市场竞争": 49732,
+      "▁registers": 49733,
+      "▁satellites": 49734,
+      "ís": 49735,
+      "wav": 49736,
+      "sects": 49737,
+      "EXPORT": 49738,
+      "▁onclick": 49739,
+      "外婆": 49740,
+      "和自己": 49741,
+      "▁Idea": 49742,
+      "igmoid": 49743,
+      "▁dialing": 49744,
+      "▁emotionally": 49745,
+      "▁spontaneously": 49746,
+      "采摘": 49747,
+      "进行判断": 49748,
+      "▁公司简介": 49749,
+      "▁triglycer": 49750,
+      "▁Profession": 49751,
+      "但她": 49752,
+      "光大": 49753,
+      "厂的": 49754,
+      "fessor": 49755,
+      "otions": 49756,
+      "▁attacker": 49757,
+      "Raises": 49758,
+      "challenge": 49759,
+      "拨打": 49760,
+      "输了": 49761,
+      "这首": 49762,
+      "想办法": 49763,
+      "▁labs": 49764,
+      "aptops": 49765,
+      "inside": 49766,
+      "▁chill": 49767,
+      "REN": 49768,
+      "年以上": 49769,
+      "INSTALL": 49770,
+      "outines": 49771,
+      "第七十二条第一款": 49772,
+      "第一百四十五条宣判前": 49773,
+      "▁后": 49774,
+      "esp": 49775,
+      "▁Ferm": 49776,
+      "动词": 49777,
+      "新款": 49778,
+      "月球": 49779,
+      "chaft": 49780,
+      "▁Surrey": 49781,
+      "▁Advisor": 49782,
+      "▁Regression": 49783,
+      "▁signalling": 49784,
+      "刊登": 49785,
+      "哔哩": 49786,
+      "▁LHC": 49787,
+      "取消预订": 49788,
+      "sandbox": 49789,
+      "保姆": 49790,
+      "拼接": 49791,
+      "玩了": 49792,
+      "聆听": 49793,
+      "▁dryer": 49794,
+      "想像": 49795,
+      "ogon": 49796,
+      "▁readme": 49797,
+      "▁洪": 49798,
+      "全额": 49799,
+      "写字楼": 49800,
+      "▁{})": 49801,
+      "项目管理": 49802,
+      "▁freed": 49803,
+      "▁Ethics": 49804,
+      "太湖": 49805,
+      "有据": 49806,
+      "可以参考": 49807,
+      "becca": 49808,
+      "▁Employment": 49809,
+      "▁Silva": 49810,
+      "Drawing": 49811,
+      "▁asleep": 49812,
+      "▁Flowers": 49813,
+      "▁abandon": 49814,
+      "▁hilarious": 49815,
+      "FK": 49816,
+      "广义": 49817,
+      "的的": 49818,
+      "要加强": 49819,
+      "配置文件": 49820,
+      "▁daytime": 49821,
+      "▁Activities": 49822,
+      "游乐": 49823,
+      "▁Extended": 49824,
+      "nPublications": 49825,
+      "宜兴": 49826,
+      "登山": 49827,
+      "粤语": 49828,
+      "haz": 49829,
+      "Hidden": 49830,
+      "▁parity": 49831,
+      "▁proactive": 49832,
+      "nUK": 49833,
+      "和研究": 49834,
+      "▁Legacy": 49835,
+      "Scheduler": 49836,
+      "喂养": 49837,
+      "护照": 49838,
+      "邮电": 49839,
+      "基本都是": 49840,
+      "▁excision": 49841,
+      "▁Gö": 49842,
+      "acos": 49843,
+      "上诉理由": 49844,
+      "撤诉条件": 49845,
+      "▁acne": 49846,
+      "Examples": 49847,
+      "春晚": 49848,
+      "予以查封": 49849,
+      "SciNet": 49850,
+      "早年": 49851,
+      "设备和": 49852,
+      "▁PRs": 49853,
+      "ffffff": 49854,
+      "▁tales": 49855,
+      "抉择": 49856,
+      "招募": 49857,
+      "肉类": 49858,
+      "▁Ronald": 49859,
+      "▁scholarships": 49860,
+      "MLE": 49861,
+      "本院将依法": 49862,
+      "判处有期徒刑三年": 49863,
+      "不惜": 49864,
+      "开后": 49865,
+      "▁Rid": 49866,
+      "用户体验": 49867,
+      "▁burns": 49868,
+      "▁diode": 49869,
+      "Workflow": 49870,
+      "throughput": 49871,
+      "并由": 49872,
+      "房租": 49873,
+      "螃蟹": 49874,
+      "({\\\\": 49875,
+      "deck": 49876,
+      "▁gri": 49877,
+      "pherd": 49878,
+      "▁Geoff": 49879,
+      "▁surnames": 49880,
+      "▁Outstanding": 49881,
+      "▁biologically": 49882,
+      "皮质": 49883,
+      "预后": 49884,
+      "城市建设": 49885,
+      "金融服务": 49886,
+      "▁rede": 49887,
+      "ACTION": 49888,
+      "▁Flower": 49889,
+      "▁swallow": 49890,
+      "▁Armed": 49891,
+      "▁Handler": 49892,
+      "▁expands": 49893,
+      "▁dominance": 49894,
+      "mA": 49895,
+      "一瓶": 49896,
+      "还剩": 49897,
+      "▁HH": 49898,
+      "▁lamb": 49899,
+      "▁relied": 49900,
+      "▁盐": 49901,
+      "或多": 49902,
+      "盛宴": 49903,
+      "和预先": 49904,
+      "▁vivid": 49905,
+      "▁Eventually": 49906,
+      "苯丙": 49907,
+      "istar": 49908,
+      "的名": 49909,
+      "▁inh": 49910,
+      "▁Owen": 49911,
+      "▁\\'\\')\\": 49912,
+      "迹象": 49913,
+      "TING": 49914,
+      "▁prospectively": 49915,
+      "四次": 49916,
+      "Jet": 49917,
+      "下来了": 49918,
+      "arance": 49919,
+      "cash": 49920,
+      "客户提供": 49921,
+      "范围内的": 49922,
+      "routing": 49923,
+      "▁tumorigen": 49924,
+      "球菌": 49925,
+      "mak": 49926,
+      "知道自己": 49927,
+      "组织开展": 49928,
+      "▁stromal": 49929,
+      "刀具": 49930,
+      "▁适宜环境": 49931,
+      "一圈": 49932,
+      "▁/\\\\": 49933,
+      "▁anomaly": 49934,
+      "▁microphone": 49935,
+      "▁cultivation": 49936,
+      "买车": 49937,
+      "▁Lots": 49938,
+      "graphs": 49939,
+      "▁immense": 49940,
+      "▁nx": 49941,
+      "adoc": 49942,
+      "arctica": 49943,
+      "conscious": 49944,
+      "gis": 49945,
+      "andin": 49946,
+      "▁THEY": 49947,
+      "金黄": 49948,
+      "IPO": 49949,
+      "Jax": 49950,
+      "实际行动": 49951,
+      "联系电话": 49952,
+      "▁seized": 49953,
+      "▁playwright": 49954,
+      "HOW": 49955,
+      "如果能": 49956,
+      "明确了": 49957,
+      "的时候就": 49958,
+      "nCase": 49959,
+      "toHave": 49960,
+      "▁Bened": 49961,
+      "▁omega": 49962,
+      "▁endothelium": 49963,
+      "▁Subscription": 49964,
+      "印花": 49965,
+      "▁Peg": 49966,
+      "▁strept": 49967,
+      "▁wishing": 49968,
+      "不打": 49969,
+      "本科学历": 49970,
+      "trust": 49971,
+      "cessing": 49972,
+      "不同类型的客房": 49973,
+      "▁atypical": 49974,
+      "▁longitude": 49975,
+      "▁pathologic": 49976,
+      "nmsg": 49977,
+      "umped": 49978,
+      "reating": 49979,
+      "nSwedish": 49980,
+      "▁Cardinals": 49981,
+      "但又": 49982,
+      "向来": 49983,
+      "刑事案件": 49984,
+      "Chris": 49985,
+      "▁replica": 49986,
+      "显微镜": 49987,
+      "巴基斯坦": 49988,
+      "songwriter": 49989,
+      "机和": 49990,
+      "nTO": 49991,
+      "医疗费": 49992,
+      "▁卫生间": 49993,
+      "▁Brigade": 49994,
+      "Az": 49995,
+      "巡查": 49996,
+      "definitions": 49997,
+      "亲近": 49998,
+      "非物质": 49999,
+      "历时": 50000,
+      "限量": 50001,
+      "Mix": 50002,
+      "Qual": 50003,
+      "的一系列": 50004,
+      "▁puzzles": 50005,
+      "▁Lighting": 50006,
+      "omorphisms": 50007,
+      "EMP": 50008,
+      "当事人陈述": 50009,
+      "SEARCH": 50010,
+      "▁thigh": 50011,
+      "余姚": 50012,
+      "遗迹": 50013,
+      "其主要": 50014,
+      "▁OEM": 50015,
+      "animate": 50016,
+      "capital": 50017,
+      "▁culinary": 50018,
+      "▁predicate": 50019,
+      "可信": 50020,
+      "IVATE": 50021,
+      "ovanni": 50022,
+      "到我": 50023,
+      "直肠": 50024,
+      "是不能": 50025,
+      "▁ABS": 50026,
+      "▁relational": 50027,
+      "将有": 50028,
+      "细则": 50029,
+      "长生": 50030,
+      "▁peg": 50031,
+      "rites": 50032,
+      "▁Arduino": 50033,
+      "▁curated": 50034,
+      "▁gesture": 50035,
+      "mbic": 50036,
+      "▁\"/\"": 50037,
+      "▁Tir": 50038,
+      "▁Illustr": 50039,
+      "forgettable": 50040,
+      "图画": 50041,
+      "Nov": 50042,
+      "在工作": 50043,
+      "Advanced": 50044,
+      "otoxicity": 50045,
+      "▁Customers": 50046,
+      "▁山": 50047,
+      "向上的": 50048,
+      "▁Quadr": 50049,
+      "▁anxious": 50050,
+      "李白": 50051,
+      "为一个": 50052,
+      "▁aws": 50053,
+      "uggage": 50054,
+      "内饰": 50055,
+      "无私": 50056,
+      "确切": 50057,
+      "nhere": 50058,
+      "ucceed": 50059,
+      "▁patents": 50060,
+      "▁resembles": 50061,
+      "周长是": 50062,
+      "nMike": 50063,
+      "▁Ernest": 50064,
+      "▁Spotify": 50065,
+      "▁lecturer": 50066,
+      "IQ": 50067,
+      "中队": 50068,
+      "双语": 50069,
+      "有你": 50070,
+      "▁Ace": 50071,
+      "当地时间": 50072,
+      "▁gratitude": 50073,
+      "植树": 50074,
+      "Mah": 50075,
+      "riad": 50076,
+      "付费政策": 50077,
+      "▁antico": 50078,
+      "addClass": 50079,
+      "▁aneurysm": 50080,
+      "▁交通信息": 50081,
+      "劳动人事争议": 50082,
+      "▁osteoporosis": 50083,
+      "失调": 50084,
+      "随地": 50085,
+      "aning": 50086,
+      "▁Newport": 50087,
+      "▁prostagland": 50088,
+      "七个": 50089,
+      "同样是": 50090,
+      "Song": 50091,
+      "刑一": 50092,
+      "穴位": 50093,
+      "形势下": 50094,
+      "附带不同的": 50095,
+      "▁Designed": 50096,
+      "附带不同的取消预订": 50097,
+      "不同类型的客房附带不同的取消预订": 50098,
+      "假命题": 50099,
+      "的不是": 50100,
+      "劣势": 50101,
+      "悬疑": 50102,
+      "桥镇": 50103,
+      "▁dehyd": 50104,
+      "▁multin": 50105,
+      "和预先付费政策": 50106,
+      "便可": 50107,
+      "农牧": 50108,
+      "对任意": 50109,
+      "▁blades": 50110,
+      "▁decorate": 50111,
+      "口岸": 50112,
+      "yrus": 50113,
+      "止血": 50114,
+      "福特": 50115,
+      "shan": 50116,
+      "iku": 50117,
+      "力资源": 50118,
+      "ortal": 50119,
+      "▁所获荣誉": 50120,
+      "nBrazilian": 50121,
+      "三代": 50122,
+      "睡前": 50123,
+      "AMI": 50124,
+      "`:\\": 50125,
+      "视为放弃": 50126,
+      "▁racist": 50127,
+      "▁adviser": 50128,
+      "nArticles": 50129,
+      "为宜": 50130,
+      "平平": 50131,
+      "此文": 50132,
+      "▁XI": 50133,
+      "nJoh": 50134,
+      "death": 50135,
+      "信用卡纠纷一案": 50136,
+      "造纸": 50137,
+      "▁Hels": 50138,
+      "within": 50139,
+      "可求": 50140,
+      "过错": 50141,
+      "▁COR": 50142,
+      "▁continental": 50143,
+      "岁了": 50144,
+      "等职": 50145,
+      "▁AU": 50146,
+      "▁江苏": 50147,
+      "公共服务": 50148,
+      "nGive": 50149,
+      "NullOr": 50150,
+      "相伴": 50151,
+      "TERN": 50152,
+      "以上资料": 50153,
+      "▁menstr": 50154,
+      "jc": 50155,
+      "iele": 50156,
+      "▁striker": 50157,
+      "▁eccentric": 50158,
+      "打好": 50159,
+      "众多的": 50160,
+      "有所帮助": 50161,
+      "▁mkdir": 50162,
+      "jm": 50163,
+      "军团": 50164,
+      "▁dol": 50165,
+      "建筑装饰": 50166,
+      "olvers": 50167,
+      "ytical": 50168,
+      "▁Comedy": 50169,
+      "▁BI": 50170,
+      "▁dic": 50171,
+      "损害赔偿": 50172,
+      "款人民币": 50173,
+      "▁bonuses": 50174,
+      "▁assertion": 50175,
+      "轻度": 50176,
+      "辩证": 50177,
+      "▁IQ": 50178,
+      "▁accountable": 50179,
+      "prox": 50180,
+      "国度": 50181,
+      "▁方法": 50182,
+      "的保护": 50183,
+      "周围景观": 50184,
+      "▁amet": 50185,
+      "Manage": 50186,
+      "▁clamp": 50187,
+      "▁Tutorial": 50188,
+      "setAttribute": 50189,
+      "而行": 50190,
+      "赢了": 50191,
+      "罪犯陈": 50192,
+      "▁ECG": 50193,
+      "深层": 50194,
+      "清市": 50195,
+      "演化": 50196,
+      "BOX": 50197,
+      "陌生的": 50198,
+      "农业人口": 50199,
+      "regions": 50200,
+      "脱水": 50201,
+      "腌制": 50202,
+      "nBooks": 50203,
+      "▁moistur": 50204,
+      "办公楼": 50205,
+      "ndev": 50206,
+      "peated": 50207,
+      "成功了": 50208,
+      "▁ole": 50209,
+      "▁spp": 50210,
+      "▁phyt": 50211,
+      "▁allograft": 50212,
+      ">*": 50213,
+      "日夜": 50214,
+      "获胜": 50215,
+      "ravel": 50216,
+      "▁\\')\\": 50217,
+      "▁crap": 50218,
+      "二〇一六年五月": 50219,
+      "庄园": 50220,
+      "暧昧": 50221,
+      "amon": 50222,
+      "enin": 50223,
+      "ário": 50224,
+      "▁dislike": 50225,
+      "▁competit": 50226,
+      "上岸": 50227,
+      "formatted": 50228,
+      "＊＊": 50229,
+      "一站式": 50230,
+      "基督教": 50231,
+      "扬州市": 50232,
+      "quarters": 50233,
+      "▁parasites": 50234,
+      "▁强": 50235,
+      "配上": 50236,
+      "的战略": 50237,
+      "osyl": 50238,
+      "tdouble": 50239,
+      "▁clutter": 50240,
+      "下巴": 50241,
+      "儒家": 50242,
+      "抗辩": 50243,
+      "方块": 50244,
+      "临床上": 50245,
+      "ATES": 50246,
+      "mathsf": 50247,
+      "receiver": 50248,
+      "▁synchronous": 50249,
+      "▁esp": 50250,
+      "游戏加载": 50251,
+      "paced": 50252,
+      "▁stamps": 50253,
+      "▁observable": 50254,
+      "听听": 50255,
+      "纵横中文网": 50256,
+      "metics": 50257,
+      "▁Jesse": 50258,
+      "▁Logan": 50259,
+      "▁steadily": 50260,
+      "各组": 50261,
+      "几个人": 50262,
+      "▁Into": 50263,
+      "iscopal": 50264,
+      "▁subclass": 50265,
+      "▁phosphorus": 50266,
+      "果汁": 50267,
+      "枣庄": 50268,
+      "限额": 50269,
+      "马桶": 50270,
+      "此题考查了": 50271,
+      "delegate": 50272,
+      "▁sleeves": 50273,
+      "加了": 50274,
+      "的动作": 50275,
+      "nVer": 50276,
+      "改造期间": 50277,
+      "wares": 50278,
+      "▁trainers": 50279,
+      "▁hepatocytes": 50280,
+      "草坪": 50281,
+      "compet": 50282,
+      "▁Gibson": 50283,
+      "是单": 50284,
+      "极值": 50285,
+      "在今年": 50286,
+      "▁Sob": 50287,
+      "brush": 50288,
+      "alignment": 50289,
+      "▁Armenian": 50290,
+      "送达之日起发生法律效力": 50291,
+      "供暖": 50292,
+      "变色": 50293,
+      "古镇": 50294,
+      "yyyy": 50295,
+      "▁endings": 50296,
+      "▁laughing": 50297,
+      "觉得很": 50298,
+      "▁realizing": 50299,
+      "▁privileges": 50300,
+      "▁拉": 50301,
+      "住房和": 50302,
+      "Hero": 50303,
+      "alamus": 50304,
+      "▁heavier": 50305,
+      "nActually": 50306,
+      "特大": 50307,
+      "queries": 50308,
+      "▁bubbles": 50309,
+      "responses": 50310,
+      "▁Dominican": 50311,
+      "phen": 50312,
+      "耕地面积": 50313,
+      "nuser": 50314,
+      "washer": 50315,
+      "▁Harold": 50316,
+      "图为": 50317,
+      "得住": 50318,
+      "蔷薇": 50319,
+      "}},\\": 50320,
+      "但它": 50321,
+      "什么叫": 50322,
+      "接受的": 50323,
+      "▁COM": 50324,
+      "▁Sib": 50325,
+      "▁Zip": 50326,
+      "返还借款": 50327,
+      "▁Strat": 50328,
+      "▁spins": 50329,
+      "ouncing": 50330,
+      "▁tragic": 50331,
+      "▁illumination": 50332,
+      "读卡": 50333,
+      "vtk": 50334,
+      "聚合物": 50335,
+      "Blocks": 50336,
+      "▁rails": 50337,
+      "珠江": 50338,
+      "▁ye": 50339,
+      "▁cirrhosis": 50340,
+      "友们": 50341,
+      "有以下": 50342,
+      "arrays": 50343,
+      "取向": 50344,
+      "流星": 50345,
+      "ndb": 50346,
+      "▁overly": 50347,
+      "textless": 50348,
+      "▁supporter": 50349,
+      "NOTE": 50350,
+      "按年利率": 50351,
+      "India": 50352,
+      "▁Ideal": 50353,
+      "▁filament": 50354,
+      "▁diastolic": 50355,
+      "倾向于": 50356,
+      "分数线": 50357,
+      "nWeb": 50358,
+      "期盼": 50359,
+      "母婴": 50360,
+      "自习": 50361,
+      "角和": 50362,
+      "▁chapel": 50363,
+      "▁truncated": 50364,
+      "}%": 50365,
+      "htaking": 50366,
+      "二〇一五年六月": 50367,
+      "▁Citizens": 50368,
+      "år": 50369,
+      "元钱": 50370,
+      "投身": 50371,
+      "起飞": 50372,
+      "打造的": 50373,
+      "十一个月": 50374,
+      "▁nasty": 50375,
+      "recipes": 50376,
+      "▁pyramid": 50377,
+      "agin": 50378,
+      "orel": 50379,
+      "乌鲁木齐市": 50380,
+      "▁Intermediate": 50381,
+      "四位": 50382,
+      "批次": 50383,
+      "证据材料": 50384,
+      "分析法": 50385,
+      "圆柱的": 50386,
+      "idis": 50387,
+      "共和": 50388,
+      "IEnum": 50389,
+      "textgreater": 50390,
+      "JC": 50391,
+      "路易": 50392,
+      "Nom": 50393,
+      "▁Hex": 50394,
+      "Weather": 50395,
+      "▁quartz": 50396,
+      "▁frontend": 50397,
+      "▁senators": 50398,
+      "缺口": 50399,
+      "列方程": 50400,
+      "是一场": 50401,
+      "chester": 50402,
+      "▁keratin": 50403,
+      "coins": 50404,
+      "hurst": 50405,
+      "的最佳选择": 50406,
+      "考量": 50407,
+      "酒后驾驶": 50408,
+      "▁Mint": 50409,
+      "前任": 50410,
+      "综合素质": 50411,
+      "BEGIN": 50412,
+      "Aggreg": 50413,
+      "▁scenic": 50414,
+      "▁craftsm": 50415,
+      "真假": 50416,
+      "的一位": 50417,
+      "奥林匹克": 50418,
+      "▁puls": 50419,
+      "arying": 50420,
+      "▁furnish": 50421,
+      "▁deprivation": 50422,
+      "售出": 50423,
+      "hedron": 50424,
+      "龙岗": 50425,
+      "免费的停车服务": 50426,
+      "acceptable": 50427,
+      "▁inpatient": 50428,
+      "回路": 50429,
+      "ONY": 50430,
+      "Wil": 50431,
+      "Insp": 50432,
+      "artifactId": 50433,
+      "▁therapists": 50434,
+      "▁intracranial": 50435,
+      "▁CDC": 50436,
+      "▁OSI": 50437,
+      "执行和解协议": 50438,
+      "▁tackles": 50439,
+      "▁comedian": 50440,
+      "▁dropdown": 50441,
+      "▁freelance": 50442,
+      "▁Algorithms": 50443,
+      "▁dissolution": 50444,
+      "▁drastically": 50445,
+      "▁CW": 50446,
+      "▁flor": 50447,
+      "委员会主任": 50448,
+      "celain": 50449,
+      "▁孟": 50450,
+      "征信": 50451,
+      "就能够": 50452,
+      "Bits": 50453,
+      "▁Bangkok": 50454,
+      "Detection": 50455,
+      "▁…\\": 50456,
+      "▁第五十": 50457,
+      "▁PyQt": 50458,
+      "▁blob": 50459,
+      "▁jersey": 50460,
+      "▁atrophy": 50461,
+      "Australia": 50462,
+      "▁endangered": 50463,
+      "定额": 50464,
+      "内部的": 50465,
+      "▁sanit": 50466,
+      "America": 50467,
+      "nUnless": 50468,
+      "▁Legion": 50469,
+      "▁inertia": 50470,
+      "▁invention": 50471,
+      "tpar": 50472,
+      "▁txt": 50473,
+      "▁amelior": 50474,
+      "àn": 50475,
+      "电影节": 50476,
+      "澄清": 50477,
+      "萨尔": 50478,
+      "独一无": 50479,
+      "keyboard": 50480,
+      "▁parseInt": 50481,
+      "不息": 50482,
+      "菱形": 50483,
+      "▁通过": 50484,
+      "▁RSA": 50485,
+      "▁Malta": 50486,
+      "▁Scroll": 50487,
+      "Typography": 50488,
+      "hematically": 50489,
+      "的去": 50490,
+      "高出": 50491,
+      "▁ric": 50492,
+      "▁archaeological": 50493,
+      "收治": 50494,
+      "而对于": 50495,
+      "ATIC": 50496,
+      "ilinear": 50497,
+      "▁Electro": 50498,
+      "ativistic": 50499,
+      "不平": 50500,
+      "ungs": 50501,
+      "▁Exit": 50502,
+      "▁withstand": 50503,
+      "ít": 50504,
+      "石榴": 50505,
+      "▁coral": 50506,
+      "▁jokes": 50507,
+      "灌注": 50508,
+      "osta": 50509,
+      "相关负责人": 50510,
+      "▁fines": 50511,
+      "身份证复印件": 50512,
+      "▁Binding": 50513,
+      "好好的": 50514,
+      "▁timeless": 50515,
+      "▁criminals": 50516,
+      "上层": 50517,
+      "变质": 50518,
+      "▁浴室": 50519,
+      "按规定": 50520,
+      "Coin": 50521,
+      "\\')}}": 50522,
+      "tensorflow": 50523,
+      "前台": 50524,
+      "金额为": 50525,
+      "opez": 50526,
+      "nCollege": 50527,
+      "同济": 50528,
+      "沙特": 50529,
+      "▁Ji": 50530,
+      "wife": 50531,
+      "▁Koch": 50532,
+      "▁encompasses": 50533,
+      "惠民": 50534,
+      "黑人": 50535,
+      "obo": 50536,
+      "同时在": 50537,
+      "是否能": 50538,
+      "▁Dogs": 50539,
+      "嘿嘿": 50540,
+      "意识和": 50541,
+      "issu": 50542,
+      "▁Closed": 50543,
+      "了一句": 50544,
+      "调研员": 50545,
+      "摇头": 50546,
+      "oki": 50547,
+      "▁thi": 50548,
+      "anian": 50549,
+      "▁Atom": 50550,
+      "tenham": 50551,
+      "matplotlib": 50552,
+      "上百": 50553,
+      "唐朝": 50554,
+      "社会学": 50555,
+      "alin": 50556,
+      "长期以来": 50557,
+      "Arrow": 50558,
+      "有管辖权的": 50559,
+      "descriptor": 50560,
+      "思议": 50561,
+      "的模式": 50562,
+      "Logs": 50563,
+      "ByteArray": 50564,
+      "少爷": 50565,
+      "真的太": 50566,
+      "这种人": 50567,
+      "驻马店": 50568,
+      "▁septic": 50569,
+      "震动": 50570,
+      "首相": 50571,
+      "即便是": 50572,
+      "▁Troy": 50573,
+      "Producer": 50574,
+      "▁stretched": 50575,
+      "▁Apparently": 50576,
+      "▁aggression": 50577,
+      "插座": 50578,
+      "沸腾": 50579,
+      "装潢": 50580,
+      "但我们": 50581,
+      "▁Excellent": 50582,
+      "有三": 50583,
+      "穷尽": 50584,
+      "opping": 50585,
+      "大成": 50586,
+      "委书记": 50587,
+      "vertices": 50588,
+      "像我": 50589,
+      "海涛": 50590,
+      "utilities": 50591,
+      "为正": 50592,
+      "卫浴": 50593,
+      "的再审申请": 50594,
+      "derabad": 50595,
+      "bbox": 50596,
+      "odon": 50597,
+      "Receive": 50598,
+      "合规": 50599,
+      "optic": 50600,
+      "▁boarding": 50601,
+      "于法": 50602,
+      "好久": 50603,
+      "ест": 50604,
+      "▁Lap": 50605,
+      "▁reductase": 50606,
+      "nPoliticians": 50607,
+      "天鹅": 50608,
+      "斯顿": 50609,
+      "超声波": 50610,
+      "▁Mile": 50611,
+      "athlon": 50612,
+      "▁Sector": 50613,
+      "Disabled": 50614,
+      "▁elicited": 50615,
+      "aphylococcus": 50616,
+      "与时": 50617,
+      "沐浴": 50618,
+      "▁refinement": 50619,
+      "▁nonetheless": 50620,
+      "六大": 50621,
+      "指点": 50622,
+      "carousel": 50623,
+      "借钱": 50624,
+      "截面": 50625,
+      "尚欠原告": 50626,
+      "▁EGFR": 50627,
+      "olding": 50628,
+      "▁regimes": 50629,
+      "▁usability": 50630,
+      "真的有": 50631,
+      "▁novo": 50632,
+      "▁motifs": 50633,
+      "▁Musical": 50634,
+      "不改": 50635,
+      "恋人": 50636,
+      "oulder": 50637,
+      "▁economically": 50638,
+      "脸部": 50639,
+      "的趋势": 50640,
+      "▁Crypto": 50641,
+      "沧海": 50642,
+      "白雪": 50643,
+      "制定的": 50644,
+      "▁Chin": 50645,
+      "ListView": 50646,
+      "physical": 50647,
+      "再算": 50648,
+      "异味": 50649,
+      "水性": 50650,
+      "Sat": 50651,
+      "▁Trou": 50652,
+      "▁aggress": 50653,
+      "▁inverted": 50654,
+      "佩服": 50655,
+      "而知": 50656,
+      "指挥部": 50657,
+      "▁Rig": 50658,
+      "直接利用": 50659,
+      "nAvailable": 50660,
+      "精髓": 50661,
+      "草木": 50662,
+      "三位数": 50663,
+      "紧接着": 50664,
+      "▁liquids": 50665,
+      "Specification": 50666,
+      "想了": 50667,
+      "请注意": 50668,
+      "达不到": 50669,
+      "Fall": 50670,
+      "Plugins": 50671,
+      "有异议": 50672,
+      "比亚迪": 50673,
+      "▁HEL": 50674,
+      "Relative": 50675,
+      "esModule": 50676,
+      "produced": 50677,
+      "主要由": 50678,
+      "arga": 50679,
+      "<>\\": 50680,
+      "aways": 50681,
+      "nPrint": 50682,
+      "……………………": 50683,
+      "但从": 50684,
+      "tgl": 50685,
+      "filer": 50686,
+      "oprop": 50687,
+      "▁phylogenetic": 50688,
+      "理化": 50689,
+      "▁Recruit": 50690,
+      "老头": 50691,
+      "脱颖": 50692,
+      "CPP": 50693,
+      "cake": 50694,
+      "的方程为": 50695,
+      "cemic": 50696,
+      "▁getId": 50697,
+      "Blank": 50698,
+      "▁sher": 50699,
+      "another": 50700,
+      "▁mindset": 50701,
+      "pv": 50702,
+      "VersionUID": 50703,
+      "▁polarized": 50704,
+      "华山": 50705,
+      "坚固": 50706,
+      "多篇": 50707,
+      "富豪": 50708,
+      "花香": 50709,
+      "▁rodent": 50710,
+      "▁trigonometric": 50711,
+      "PQ": 50712,
+      "wu": 50713,
+      "си": 50714,
+      "fade": 50715,
+      "二〇一七年七月": 50716,
+      "当他": 50717,
+      "折射": 50718,
+      "vern": 50719,
+      "所需要的": 50720,
+      "mirror": 50721,
+      "plicate": 50722,
+      "▁Mercury": 50723,
+      "▁transgender": 50724,
+      "八大": 50725,
+      "时而": 50726,
+      "manship": 50727,
+      "▁Monica": 50728,
+      "▁invoked": 50729,
+      "冷的": 50730,
+      "昆山": 50731,
+      "REM": 50732,
+      "▁Gel": 50733,
+      "Because": 50734,
+      "Routing": 50735,
+      "oplasty": 50736,
+      "▁Transit": 50737,
+      "[^": 50738,
+      "写明": 50739,
+      "双方约定": 50740,
+      "▁Cult": 50741,
+      "▁urls": 50742,
+      "relations": 50743,
+      "▁shoppers": 50744,
+      "▁submarine": 50745,
+      "做一些": 50746,
+      "▁Bes": 50747,
+      "▁multitude": 50748,
+      "原先": 50749,
+      "器和": 50750,
+      "张国": 50751,
+      "曙光": 50752,
+      "ecd": 50753,
+      "ения": 50754,
+      "▁Bean": 50755,
+      "▁Animals": 50756,
+      "等腰三角形": 50757,
+      "▁shortcut": 50758,
+      "▁Regardless": 50759,
+      "药剂": 50760,
+      "▁Dj": 50761,
+      "▁Sessions": 50762,
+      "▁morphism": 50763,
+      "▁章": 50764,
+      "声称": 50765,
+      "逃脱": 50766,
+      "ticks": 50767,
+      "▁salon": 50768,
+      "▁alpine": 50769,
+      "▁Robertson": 50770,
+      "ollar": 50771,
+      "腹痛": 50772,
+      "飞速": 50773,
+      "ADA": 50774,
+      "ensors": 50775,
+      "▁renders": 50776,
+      "▁announcements": 50777,
+      "Ren": 50778,
+      "事件的": 50779,
+      ")**(": 50780,
+      "书面答辩": 50781,
+      "▁plaus": 50782,
+      "speaking": 50783,
+      "定居": 50784,
+      "开着": 50785,
+      "杰伦": 50786,
+      "必要时": 50787,
+      "signup": 50788,
+      "▁affirm": 50789,
+      "▁Cognitive": 50790,
+      "▁forthcoming": 50791,
+      "▁physiologic": 50792,
+      "Ans": 50793,
+      "也会有": 50794,
+      "新华网": 50795,
+      "▁ATM": 50796,
+      "没有约定": 50797,
+      "▁hect": 50798,
+      "▁doctoral": 50799,
+      "▁demolished": 50800,
+      "不离": 50801,
+      "直立": 50802,
+      "名词审定委员会": 50803,
+      "记作": 50804,
+      "党总支": 50805,
+      "让更多": 50806,
+      "nInvest": 50807,
+      "mV": 50808,
+      "埃尔": 50809,
+      "}\\\\)（": 50810,
+      "同伴": 50811,
+      "露天": 50812,
+      "多少个": 50813,
+      "▁sturdy": 50814,
+      "什么事": 50815,
+      "的工具": 50816,
+      "▁Lon": 50817,
+      "生活质量": 50818,
+      "▁plag": 50819,
+      "peptide": 50820,
+      "accuracy": 50821,
+      "禁用": 50822,
+      "\"})\\": 50823,
+      "▁jewellery": 50824,
+      "▁neuropath": 50825,
+      "审阅": 50826,
+      "甲乙": 50827,
+      "羽绒": 50828,
+      "▁DT": 50829,
+      "nHappy": 50830,
+      "▁Marin": 50831,
+      "萨斯": 50832,
+      "pgf": 50833,
+      "glyph": 50834,
+      "▁humble": 50835,
+      "Declaration": 50836,
+      "}`;\\": 50837,
+      "▁dividend": 50838,
+      "人造": 50839,
+      "巧合": 50840,
+      "]));\\": 50841,
+      "▁Pret": 50842,
+      "▁职业生涯": 50843,
+      "ogenes": 50844,
+      "advanced": 50845,
+      "▁internship": 50846,
+      "▁introductory": 50847,
+      "▁Stephanie": 50848,
+      "▁gardening": 50849,
+      "开店": 50850,
+      "isValid": 50851,
+      "▁geography": 50852,
+      "▁reinforce": 50853,
+      "星际": 50854,
+      "见于": 50855,
+      "系列的": 50856,
+      "▁fulfilling": 50857,
+      "qr": 50858,
+      "城管": 50859,
+      "▁mixer": 50860,
+      "▁curtain": 50861,
+      "▁opacity": 50862,
+      "jee": 50863,
+      "▁Cry": 50864,
+      "直接影响": 50865,
+      "▁spectroscopic": 50866,
+      "龙的": 50867,
+      "Advisor": 50868,
+      "▁bespoke": 50869,
+      "▁gradual": 50870,
+      "回调": 50871,
+      "大三": 50872,
+      "春日": 50873,
+      "nMag": 50874,
+      "变得更加": 50875,
+      "encers": 50876,
+      "党内": 50877,
+      "达尔": 50878,
+      "你喜欢": 50879,
+      "▁Greens": 50880,
+      "▁hackers": 50881,
+      "▁Accordingly": 50882,
+      "张艺": 50883,
+      "historic": 50884,
+      "▁ventral": 50885,
+      "凋亡": 50886,
+      "\"][": 50887,
+      "也都是": 50888,
+      "艺术品": 50889,
+      "▁McD": 50890,
+      "可以随时": 50891,
+      "rolet": 50892,
+      "Publisher": 50893,
+      "上台": 50894,
+      "思明": 50895,
+      "▁GU": 50896,
+      "laus": 50897,
+      "▁MUST": 50898,
+      "限制高消费": 50899,
+      "▁Angela": 50900,
+      "天台": 50901,
+      "干警": 50902,
+      "朱某": 50903,
+      "泌尿": 50904,
+      "闪电": 50905,
+      "▁Slack": 50906,
+      "▁pumpkin": 50907,
+      "nDistrict": 50908,
+      "打扫": 50909,
+      "浦区": 50910,
+      "Bug": 50911,
+      "nps": 50912,
+      "▁td": 50913,
+      "})$$": 50914,
+      "Average": 50915,
+      "▁coarse": 50916,
+      "nAgain": 50917,
+      "▁Walmart": 50918,
+      "勘探": 50919,
+      "环氧": 50920,
+      "的吧": 50921,
+      ">\";\\": 50922,
+      "▁syrup": 50923,
+      "夫妻感情确已破裂": 50924,
+      "▁meanings": 50925,
+      "▁philosophical": 50926,
+      "永康": 50927,
+      "gio": 50928,
+      "说白了": 50929,
+      "devel": 50930,
+      "社会抚养费征收": 50931,
+      "就是把": 50932,
+      "的法定": 50933,
+      "tiff": 50934,
+      "▁currencies": 50935,
+      "老了": 50936,
+      "谱曲": 50937,
+      "sap": 50938,
+      "农业部": 50939,
+      "电影院": 50940,
+      "等材料": 50941,
+      "▁gait": 50942,
+      "▁Breakfast": 50943,
+      "▁landscaping": 50944,
+      "交所": 50945,
+      "看重": 50946,
+      "拖拉机": 50947,
+      "柳州市": 50948,
+      "▁Journey": 50949,
+      "▁interpreter": 50950,
+      "量身": 50951,
+      "▁注意": 50952,
+      "▁Ply": 50953,
+      "目前为止": 50954,
+      "▁acre": 50955,
+      "▁参考资料": 50956,
+      "▁quark": 50957,
+      "▁happier": 50958,
+      "救护": 50959,
+      "习惯了": 50960,
+      "nDen": 50961,
+      "▁Progressive": 50962,
+      "给付义务": 50963,
+      "▁Lect": 50964,
+      "▁optics": 50965,
+      "▁festive": 50966,
+      "▁autonomy": 50967,
+      "▁foremost": 50968,
+      "▁Gospel": 50969,
+      "▁trophy": 50970,
+      "▁harvested": 50971,
+      "▁美": 50972,
+      "与他": 50973,
+      "已达": 50974,
+      "ahren": 50975,
+      "▁breadth": 50976,
+      "OV": 50977,
+      "Mus": 50978,
+      "保持着": 50979,
+      "epid": 50980,
+      "▁chemically": 50981,
+      "头顶": 50982,
+      "快三": 50983,
+      "县国土资源局": 50984,
+      "▁Modules": 50985,
+      "可以为": 50986,
+      "▁crews": 50987,
+      "奇特": 50988,
+      "请选择": 50989,
+      "Drag": 50990,
+      "uces": 50991,
+      "nowledg": 50992,
+      "theight": 50993,
+      "▁curtains": 50994,
+      "工件": 50995,
+      "Bed": 50996,
+      "▁bc": 50997,
+      "统计图": 50998,
+      "nLin": 50999,
+      "▁dangers": 51000,
+      "湖人": 51001,
+      "在学习": 51002,
+      "▁sandw": 51003,
+      "助推": 51004,
+      "苗木": 51005,
+      "以提高": 51006,
+      "满足了": 51007,
+      "Little": 51008,
+      "▁temples": 51009,
+      "后会": 51010,
+      "▁Eco": 51011,
+      "__))\\": 51012,
+      "METHOD": 51013,
+      "CCC": 51014,
+      "从而使": 51015,
+      "现代化的": 51016,
+      "▁Whereas": 51017,
+      "飞船": 51018,
+      "▁Elli": 51019,
+      "▁Walking": 51020,
+      "十几年": 51021,
+      "女主角": 51022,
+      "meal": 51023,
+      "文物保护": 51024,
+      "▁demonstrations": 51025,
+      "▁ф": 51026,
+      "契合": 51027,
+      "▁POL": 51028,
+      "▁歌曲语言": 51029,
+      "▁Coming": 51030,
+      "▁Davies": 51031,
+      "▁forensic": 51032,
+      "现今": 51033,
+      "脚踏": 51034,
+      "与方法": 51035,
+      "无统计学意义": 51036,
+      "▁Spider": 51037,
+      "hetamine": 51038,
+      "▁Reality": 51039,
+      "▁isotope": 51040,
+      "罚金限": 51041,
+      "Fake": 51042,
+      "▁libr": 51043,
+      "SPI": 51044,
+      "分子量": 51045,
+      "nEndemic": 51046,
+      "▁poisoning": 51047,
+      "▁organizers": 51048,
+      "主的": 51049,
+      "地为": 51050,
+      "极高": 51051,
+      "▁banana": 51052,
+      "不复": 51053,
+      "纳斯": 51054,
+      "▁palace": 51055,
+      "▁seafood": 51056,
+      "纹理": 51057,
+      "并同意": 51058,
+      "我看到": 51059,
+      "▁ICU": 51060,
+      "etting": 51061,
+      "▁Signature": 51062,
+      "祖先": 51063,
+      "钢笔": 51064,
+      "Merge": 51065,
+      "▁wasted": 51066,
+      "▁decoding": 51067,
+      "三边": 51068,
+      "宫廷": 51069,
+      "况且": 51070,
+      "▁普通": 51071,
+      "积累了": 51072,
+      "extern": 51073,
+      "nInteresting": 51074,
+      "的呢": 51075,
+      "▁AW": 51076,
+      "▁Ack": 51077,
+      "▁Fuel": 51078,
+      "▁vegg": 51079,
+      "vehicle": 51080,
+      "▁skating": 51081,
+      "▁mutagenesis": 51082,
+      "再次申请": 51083,
+      "▁datas": 51084,
+      "installer": 51085,
+      "上路": 51086,
+      "league": 51087,
+      "▁lactate": 51088,
+      "▁percentages": 51089,
+      "适用的": 51090,
+      "团员": 51091,
+      "既能": 51092,
+      "全村有": 51093,
+      "▁necklace": 51094,
+      "▁splicing": 51095,
+      "下面就": 51096,
+      "大有": 51097,
+      "一生的": 51098,
+      "nSport": 51099,
+      "nafter": 51100,
+      "▁Herald": 51101,
+      "です": 51102,
+      "eti": 51103,
+      "vee": 51104,
+      "券商": 51105,
+      "的量": 51106,
+      "香菜": 51107,
+      "Recipe": 51108,
+      "▁Vince": 51109,
+      "民进": 51110,
+      "STE": 51111,
+      "ntext": 51112,
+      "▁Deck": 51113,
+      "Maker": 51114,
+      "▁wasting": 51115,
+      "交谈": 51116,
+      "家家": 51117,
+      "用的是": 51118,
+      "ogenicity": 51119,
+      "响起": 51120,
+      "textit": 51121,
+      "被执行人应当": 51122,
+      "▁senator": 51123,
+      "▁imbalance": 51124,
+      "vv": 51125,
+      "会议上": 51126,
+      "诸葛亮": 51127,
+      "nSal": 51128,
+      "实施方案": 51129,
+      "Rating": 51130,
+      "harmonic": 51131,
+      "▁modulating": 51132,
+      "分院": 51133,
+      "(\":": 51134,
+      "容留他人吸毒": 51135,
+      "Intended": 51136,
+      "▁freshwater": 51137,
+      "▁正": 51138,
+      "打通": 51139,
+      "诊所": 51140,
+      "更喜欢": 51141,
+      "▁contracted": 51142,
+      "死者": 51143,
+      "犯非法": 51144,
+      "▁seismic": 51145,
+      "▁Photoshop": 51146,
+      "▁Fc": 51147,
+      "在整个": 51148,
+      "uracy": 51149,
+      "专业合作社": 51150,
+      "▁Brighton": 51151,
+      "周刊": 51152,
+      "序号": 51153,
+      "这种方式": 51154,
+      "▁Zoom": 51155,
+      "▁postal": 51156,
+      "▁雪": 51157,
+      "不完全": 51158,
+      "他们会": 51159,
+      "▁REL": 51160,
+      "▁Tah": 51161,
+      "第七十九": 51162,
+      "vergence": 51163,
+      "俊杰": 51164,
+      "我跟": 51165,
+      "种的": 51166,
+      "纯净": 51167,
+      "降至": 51168,
+      "colspan": 51169,
+      "▁genomes": 51170,
+      "▁counters": 51171,
+      "轰炸": 51172,
+      "aye": 51173,
+      "Disk": 51174,
+      "headed": 51175,
+      "EqualTo": 51176,
+      "nBetween": 51177,
+      "shutdown": 51178,
+      "白日": 51179,
+      "也是非常": 51180,
+      "高等院校": 51181,
+      "▁thru": 51182,
+      "适度": 51183,
+      "GRAM": 51184,
+      "▁portrayed": 51185,
+      "书名": 51186,
+      "排斥": 51187,
+      "相距": 51188,
+      "神州": 51189,
+      "▁XL": 51190,
+      "第三届": 51191,
+      "▁Quote": 51192,
+      "▁worrying": 51193,
+      "献血": 51194,
+      "下部": 51195,
+      "庭的": 51196,
+      "贫穷": 51197,
+      "esi": 51198,
+      "▁RFC": 51199,
+      "illet": 51200,
+      "和非": 51201,
+      "）,": 51202,
+      "不就是": 51203,
+      "据介绍": 51204,
+      "Latest": 51205,
+      "enguin": 51206,
+      "▁ether": 51207,
+      "▁oscillator": 51208,
+      "CX": 51209,
+      "上饶": 51210,
+      "许昌": 51211,
+      "ampire": 51212,
+      "委托代理人刘": 51213,
+      "▁Revision": 51214,
+      "时被": 51215,
+      "ifs": 51216,
+      "▁MIC": 51217,
+      "orthy": 51218,
+      "▁Rebecca": 51219,
+      "▁behaviours": 51220,
+      "方向的": 51221,
+      "好不容易": 51222,
+      "Happy": 51223,
+      "nEcon": 51224,
+      "▁Effective": 51225,
+      "▁synthesize": 51226,
+      "舍不得": 51227,
+      "▁Batch": 51228,
+      "getTime": 51229,
+      "respect": 51230,
+      "位上的": 51231,
+      "}{{{": 51232,
+      "▁discs": 51233,
+      "▁supplemented": 51234,
+      "的组织": 51235,
+      "▁avec": 51236,
+      "▁Assuming": 51237,
+      "热心": 51238,
+      "通路": 51239,
+      "物理学": 51240,
+      "DIRS": 51241,
+      "Defined": 51242,
+      "▁neighbouring": 51243,
+      "▁冰": 51244,
+      "比是": 51245,
+      "有理数的": 51246,
+      "▁Pros": 51247,
+      "▁mediator": 51248,
+      "▁Wrestling": 51249,
+      "▁reflections": 51250,
+      "打野": 51251,
+      "火影": 51252,
+      "主要为": 51253,
+      "在回答": 51254,
+      "我的心": 51255,
+      "nNone": 51256,
+      "▁glycoprotein": 51257,
+      "十五年": 51258,
+      "回馈": 51259,
+      "手中的": 51260,
+      "的制作": 51261,
+      "Mont": 51262,
+      "▁Feder": 51263,
+      "▁learns": 51264,
+      "▁cigarettes": 51265,
+      "心力": 51266,
+      "rack": 51267,
+      "rens": 51268,
+      "▁有限公司": 51269,
+      "Respon": 51270,
+      "▁printers": 51271,
+      "▁professors": 51272,
+      "哈佛": 51273,
+      "指望": 51274,
+      "Collector": 51275,
+      "rika": 51276,
+      "terbury": 51277,
+      "平顶": 51278,
+      "得不": 51279,
+      "PEC": 51280,
+      "Than": 51281,
+      "ucking": 51282,
+      "▁rescued": 51283,
+      "切断": 51284,
+      "▁abc": 51285,
+      "enues": 51286,
+      "▁Ride": 51287,
+      "▁Attack": 51288,
+      "波特": 51289,
+      "▁för": 51290,
+      "coded": 51291,
+      "▁Lucy": 51292,
+      "或财产线索": 51293,
+      "orphic": 51294,
+      "▁Marina": 51295,
+      "▁chooses": 51296,
+      "▁inquiries": 51297,
+      "房权证": 51298,
+      "贫困户": 51299,
+      "nSil": 51300,
+      "征程": 51301,
+      "技术在": 51302,
+      "crow": 51303,
+      "fordable": 51304,
+      "▁terrific": 51305,
+      "桃园": 51306,
+      "▁exon": 51307,
+      "▁skate": 51308,
+      "nOnline": 51309,
+      "zegovina": 51310,
+      "▁payroll": 51311,
+      "▁animations": 51312,
+      "年到": 51313,
+      "▁($\\\\": 51314,
+      "French": 51315,
+      "CSV": 51316,
+      "SEQU": 51317,
+      "▁mediates": 51318,
+      "追寻": 51319,
+      "nWE": 51320,
+      "▁Eyes": 51321,
+      "日被取保候审": 51322,
+      "english": 51323,
+      "穿梭": 51324,
+      "OPLE": 51325,
+      "nDaniel": 51326,
+      "▁shoots": 51327,
+      "▁Fernando": 51328,
+      "ма": 51329,
+      "军训": 51330,
+      "ijay": 51331,
+      "slides": 51332,
+      "CAA": 51333,
+      "▁SAR": 51334,
+      "科研项目": 51335,
+      "▁scam": 51336,
+      "nRoyal": 51337,
+      "▁reconst": 51338,
+      "▁resemble": 51339,
+      "▁ascertain": 51340,
+      "▁crystalline": 51341,
+      "▁subcutaneous": 51342,
+      "文人": 51343,
+      "aryng": 51344,
+      "是一次": 51345,
+      "nant": 51346,
+      "文化建设": 51347,
+      "nFore": 51348,
+      "▁Trin": 51349,
+      "▁exploitation": 51350,
+      "吸取": 51351,
+      "Bud": 51352,
+      "天文学": 51353,
+      "暴力枚举": 51354,
+      "mable": 51355,
+      "另案": 51356,
+      "部级": 51357,
+      "全面发展": 51358,
+      "▁Harbour": 51359,
+      "▁vulnerabilities": 51360,
+      "完备": 51361,
+      "回来的": 51362,
+      "真实性": 51363,
+      "▁iPod": 51364,
+      "▁rotor": 51365,
+      "▁ticks": 51366,
+      "犯交通肇事罪": 51367,
+      "▁ecology": 51368,
+      "百里": 51369,
+      "而这": 51370,
+      "ATS": 51371,
+      "关于人民法院": 51372,
+      "▁envision": 51373,
+      "朴素": 51374,
+      "ande": 51375,
+      "▁Liz": 51376,
+      "outhern": 51377,
+      "Mor": 51378,
+      "展览会": 51379,
+      "长期的": 51380,
+      "Pacific": 51381,
+      "▁condom": 51382,
+      "▁intends": 51383,
+      "名列": 51384,
+      "vivo": 51385,
+      "▁umbrella": 51386,
+      "尽早": 51387,
+      "期满后": 51388,
+      "解除保全": 51389,
+      "']]],\\": 51390,
+      "做大": 51391,
+      "批复": 51392,
+      "算术": 51393,
+      "▁YES": 51394,
+      "▁Moss": 51395,
+      "桌上": 51396,
+      "开心的": 51397,
+      "集团的": 51398,
+      "刑满释放": 51399,
+      "投资基金": 51400,
+      "生产基地": 51401,
+      "Heart": 51402,
+      "mixed": 51403,
+      "▁Fraser": 51404,
+      "▁centralized": 51405,
+      "OnClickListener": 51406,
+      "费的": 51407,
+      "ozo": 51408,
+      "▁masc": 51409,
+      "nTable": 51410,
+      "▁braces": 51411,
+      "汇编": 51412,
+      "readLine": 51413,
+      "▁elegance": 51414,
+      "搜寻": 51415,
+      "生孩子": 51416,
+      "中对": 51417,
+      "尽情": 51418,
+      "nInc": 51419,
+      "▁dubbed": 51420,
+      "▁mathematician": 51421,
+      "结构性": 51422,
+      "drag": 51423,
+      "▁sunshine": 51424,
+      "口区": 51425,
+      "草药": 51426,
+      "enary": 51427,
+      "▁rustic": 51428,
+      "Creating": 51429,
+      "▁Windsor": 51430,
+      "bk": 51431,
+      "志明": 51432,
+      "秉持": 51433,
+      "教育工作": 51434,
+      "ediatric": 51435,
+      "中使用": 51436,
+      "机构和": 51437,
+      "athy": 51438,
+      "maint": 51439,
+      "▁canopy": 51440,
+      "▁oncology": 51441,
+      "松弛": 51442,
+      "淮南": 51443,
+      "Its": 51444,
+      "精神病": 51445,
+      "▁制作人": 51446,
+      "Buttons": 51447,
+      "好后": 51448,
+      "新春": 51449,
+      "Sol": 51450,
+      "投资的": 51451,
+      "▁ISSN": 51452,
+      "第一行包含": 51453,
+      "roducing": 51454,
+      "外围": 51455,
+      "抽奖": 51456,
+      "ruce": 51457,
+      "▁Lic": 51458,
+      "▁cosm": 51459,
+      "▁Makes": 51460,
+      "Working": 51461,
+      "▁genius": 51462,
+      "▁rainbow": 51463,
+      "花朵": 51464,
+      "▁Arrow": 51465,
+      "反应的": 51466,
+      "果园": 51467,
+      "中新网": 51468,
+      "社会各界": 51469,
+      "▁stains": 51470,
+      "heels": 51471,
+      "idual": 51472,
+      "▁cols": 51473,
+      "▁Adults": 51474,
+      "▁crying": 51475,
+      "▁Successful": 51476,
+      "柔性": 51477,
+      "毗邻": 51478,
+      "最大化": 51479,
+      "AAAAAAAA": 51480,
+      "▁exquisite": 51481,
+      "伸缩": 51482,
+      "迟迟": 51483,
+      "郊区": 51484,
+      "aiden": 51485,
+      "▁Mell": 51486,
+      "▁shadows": 51487,
+      "厂长": 51488,
+      "阻挡": 51489,
+      "▁rushed": 51490,
+      "▁worthwhile": 51491,
+      "的材料": 51492,
+      "\\\\({{": 51493,
+      "archy": 51494,
+      "▁Expo": 51495,
+      "▁caspase": 51496,
+      "QT": 51497,
+      "…(": 51498,
+      "投产": 51499,
+      "落日": 51500,
+      "DEV": 51501,
+      "中出现": 51502,
+      "并取得": 51503,
+      "esare": 51504,
+      "▁Cheap": 51505,
+      "▁divine": 51506,
+      "▁convincing": 51507,
+      "彝族": 51508,
+      "水土": 51509,
+      "领军": 51510,
+      "▁Kab": 51511,
+      "信息进行": 51512,
+      "认真分析": 51513,
+      "▁Reserv": 51514,
+      "▁Gur": 51515,
+      "▁Cornell": 51516,
+      "ение": 51517,
+      "由申请人": 51518,
+      "▁chiral": 51519,
+      "▁forefront": 51520,
+      "▁summarized": 51521,
+      "▁涛": 51522,
+      "哥们": 51523,
+      "Mel": 51524,
+      "ilda": 51525,
+      "▁Carp": 51526,
+      "▁yielding": 51527,
+      "▁Securities": 51528,
+      "rn": 51529,
+      "年薪": 51530,
+      "领导者": 51531,
+      "xing": 51532,
+      "▁Sterling": 51533,
+      "sx": 51534,
+      "极品": 51535,
+      "▁PID": 51536,
+      "atars": 51537,
+      "Notify": 51538,
+      "sizing": 51539,
+      "findAll": 51540,
+      "useEffect": 51541,
+      "盆地": 51542,
+      "mad": 51543,
+      "▁接口": 51544,
+      "▁unanim": 51545,
+      "▁nitrate": 51546,
+      "▁stacked": 51547,
+      "▁scholarly": 51548,
+      "▁Submission": 51549,
+      "▁interrupted": 51550,
+      "始于": 51551,
+      "影响因素": 51552,
+      "princ": 51553,
+      "▁USER": 51554,
+      "Amazon": 51555,
+      "▁notch": 51556,
+      "科学记数法的": 51557,
+      "▁Optical": 51558,
+      "▁boutique": 51559,
+      "店内": 51560,
+      "集资": 51561,
+      "ensation": 51562,
+      "nEducational": 51563,
+      "两类": 51564,
+      "被查封": 51565,
+      "nTurn": 51566,
+      "omnia": 51567,
+      "▁Pride": 51568,
+      "▁Protest": 51569,
+      "印证": 51570,
+      "了出来": 51571,
+      "安县人民法院": 51572,
+      "▁stranger": 51573,
+      "趋于": 51574,
+      "区域内": 51575,
+      "润滑油": 51576,
+      "unix": 51577,
+      "来往": 51578,
+      "挑战性的": 51579,
+      "提供一个": 51580,
+      "orers": 51581,
+      "NotExist": 51582,
+      "区人民检察院指控": 51583,
+      "▁questioning": 51584,
+      "奢华": 51585,
+      "小板": 51586,
+      "西医": 51587,
+      "instead": 51588,
+      "▁indexing": 51589,
+      "▁加": 51590,
+      "万辆": 51591,
+      "InM": 51592,
+      "neo": 51593,
+      "Chinese": 51594,
+      "valence": 51595,
+      "formations": 51596,
+      "▁triggering": 51597,
+      "__)": 51598,
+      "apk": 51599,
+      "egal": 51600,
+      "TYPES": 51601,
+      "▁Malay": 51602,
+      "▁gotta": 51603,
+      "运输有限公司": 51604,
+      "▁Finite": 51605,
+      "Finished": 51606,
+      "▁importing": 51607,
+      "▁abstraction": 51608,
+      "应试": 51609,
+      "的山": 51610,
+      "博士后": 51611,
+      "就会有": 51612,
+      "的事物": 51613,
+      "▁Globe": 51614,
+      "▁SELECT": 51615,
+      "业态": 51616,
+      "为整数": 51617,
+      "市场化": 51618,
+      "能有效": 51619,
+      "就是这个": 51620,
+      "▁dominate": 51621,
+      "▁Testament": 51622,
+      "▁Certification": 51623,
+      "商机": 51624,
+      "工信": 51625,
+      "▁belie": 51626,
+      "整车": 51627,
+      "晚年": 51628,
+      "道中": 51629,
+      "日正式": 51630,
+      "与被告张": 51631,
+      "政府采购": 51632,
+      "酒店提供免费": 51633,
+      "batis": 51634,
+      "separator": 51635,
+      "classification": 51636,
+      "人参": 51637,
+      "迪士": 51638,
+      "只能是": 51639,
+      "nour": 51640,
+      "▁blends": 51641,
+      "确认人民调解协议": 51642,
+      "oL": 51643,
+      "保修政策": 51644,
+      "▁slider": 51645,
+      "▁Updates": 51646,
+      "apat": 51647,
+      "出席会议": 51648,
+      "指派代理检察员": 51649,
+      "按自动撤回上诉处理": 51650,
+      "鸳鸯": 51651,
+      "SNA": 51652,
+      "▁Wr": 51653,
+      "▁BASE": 51654,
+      "▁Poor": 51655,
+      "▁Recon": 51656,
+      "▁Publication": 51657,
+      "▁hone": 51658,
+      "▁digging": 51659,
+      "▁伊": 51660,
+      "挂在": 51661,
+      "▁Url": 51662,
+      "▁wonders": 51663,
+      "壮族": 51664,
+      "大庆": 51665,
+      "的学校": 51666,
+      "▁Hull": 51667,
+      "Principal": 51668,
+      "▁Vietnamese": 51669,
+      "希尔": 51670,
+      "▁VIII": 51671,
+      "航线": 51672,
+      "证明材料": 51673,
+      "inker": 51674,
+      "根据题意得": 51675,
+      "nChildren": 51676,
+      "▁junctions": 51677,
+      "▁Salmonella": 51678,
+      "浓浓": 51679,
+      "高尚": 51680,
+      "▁Pig": 51681,
+      "weekly": 51682,
+      "每分钟": 51683,
+      "运行的": 51684,
+      "▁fauc": 51685,
+      "▁Gloucester": 51686,
+      "▁Subsequently": 51687,
+      "立功": 51688,
+      "怎么了": 51689,
+      "▁stip": 51690,
+      "丢了": 51691,
+      "分量": 51692,
+      "古籍": 51693,
+      "换届": 51694,
+      "的回答": 51695,
+      "组委会": 51696,
+      "PROC": 51697,
+      "nOverview": 51698,
+      "_\"": 51699,
+      "qli": 51700,
+      "你应该": 51701,
+      "创建于": 51702,
+      "tlocal": 51703,
+      "▁amput": 51704,
+      "otherapeutic": 51705,
+      "字幕": 51706,
+      "篮板": 51707,
+      "holst": 51708,
+      "假装": 51709,
+      "开除": 51710,
+      "掩盖": 51711,
+      "FRA": 51712,
+      "世界大战": 51713,
+      "coated": 51714,
+      "theory": 51715,
+      "Horizontal": 51716,
+      "▁textbooks": 51717,
+      "二十一": 51718,
+      "acha": 51719,
+      "菠菜": 51720,
+      "说要": 51721,
+      "小时的": 51722,
+      "women": 51723,
+      "▁toast": 51724,
+      "全景": 51725,
+      "地基": 51726,
+      "的老师": 51727,
+      "iban": 51728,
+      "▁MPI": 51729,
+      "▁antitumor": 51730,
+      "▁guideline": 51731,
+      "▁coagulation": 51732,
+      ",[": 51733,
+      "KF": 51734,
+      "一直到": 51735,
+      "▁complained": 51736,
+      "中标": 51737,
+      "增大而": 51738,
+      "总觉得": 51739,
+      "淡淡的": 51740,
+      "真的好": 51741,
+      "二审上诉人": 51742,
+      "▁assort": 51743,
+      "烦躁": 51744,
+      "Proc": 51745,
+      "平方千米": 51746,
+      "frontal": 51747,
+      "公有": 51748,
+      "细细": 51749,
+      "莱坞": 51750,
+      "正弦定理": 51751,
+      "认罪态度": 51752,
+      "graphic": 51753,
+      "▁peanut": 51754,
+      "▁credibility": 51755,
+      "预言": 51756,
+      "▁SG": 51757,
+      "发挥了": 51758,
+      "▁pes": 51759,
+      "usuario": 51760,
+      "二〇一六年十一": 51761,
+      "nmanager": 51762,
+      "▁Spectrum": 51763,
+      "吃什么": 51764,
+      "烟台市": 51765,
+      "▁Dil": 51766,
+      "▁Heather": 51767,
+      "填补": 51768,
+      "kick": 51769,
+      "▁Coordinator": 51770,
+      "夺得": 51771,
+      "德镇": 51772,
+      "永安": 51773,
+      "sus": 51774,
+      "干细胞": 51775,
+      "管理与": 51776,
+      "▁{})\\": 51777,
+      "ashire": 51778,
+      "alsa": 51779,
+      "rxjs": 51780,
+      "mur": 51781,
+      "▁DEL": 51782,
+      "▁Roth": 51783,
+      "▁silicone": 51784,
+      "▁minorities": 51785,
+      "一般为": 51786,
+      "▁albeit": 51787,
+      "Extractor": 51788,
+      "冰冰": 51789,
+      "港区": 51790,
+      "基础的": 51791,
+      "卫生健康": 51792,
+      "ascar": 51793,
+      "▁Grab": 51794,
+      "▁concaten": 51795,
+      "▁elevator": 51796,
+      "▁noexcept": 51797,
+      "枯燥": 51798,
+      "rael": 51799,
+      "几乎没有": 51800,
+      "注销信息": 51801,
+      "齐齐哈尔": 51802,
+      "▁Transl": 51803,
+      "▁ineffective": 51804,
+      "上调": 51805,
+      "不可或缺": 51806,
+      "()):\\": 51807,
+      "▁clan": 51808,
+      "::~": 51809,
+      "人均耕地": 51810,
+      "▁Authent": 51811,
+      "Threshold": 51812,
+      "▁affection": 51813,
+      "念头": 51814,
+      "瘦肉": 51815,
+      "bab": 51816,
+      "wagen": 51817,
+      "▁Rolling": 51818,
+      "安利": 51819,
+      "抗拒": 51820,
+      "Jim": 51821,
+      "有理数": 51822,
+      "克里": 51823,
+      "页无": 51824,
+      "在一次": 51825,
+      "Clone": 51826,
+      "▁fills": 51827,
+      "西装": 51828,
+      "TEM": 51829,
+      "将面临": 51830,
+      "nWORK": 51831,
+      "▁anne": 51832,
+      "▁arXiv": 51833,
+      "▁openly": 51834,
+      "▁squash": 51835,
+      "第五百一十九条": 51836,
+      "RX": 51837,
+      "tF": 51838,
+      "带到": 51839,
+      "渐进": 51840,
+      "空的": 51841,
+      "第七十三": 51842,
+      "▁gonad": 51843,
+      "checker": 51844,
+      "▁reminis": 51845,
+      "煤气": 51846,
+      "实力的": 51847,
+      "▁pam": 51848,
+      "▁SPECIAL": 51849,
+      "▁Selected": 51850,
+      "最高人民法院关于办理减刑": 51851,
+      "▁Sev": 51852,
+      "nThink": 51853,
+      "词条注销信息": 51854,
+      "utation": 51855,
+      "Mn": 51856,
+      "都觉得": 51857,
+      "工程专业": 51858,
+      "elsius": 51859,
+      "tH": 51860,
+      "▁首": 51861,
+      "双击": 51862,
+      "▁AES": 51863,
+      "▁CAL": 51864,
+      "▁Laur": 51865,
+      "▁nour": 51866,
+      "travel": 51867,
+      "教务": 51868,
+      "也没什么": 51869,
+      "▁leuc": 51870,
+      "应及时": 51871,
+      "的房子": 51872,
+      "▁absc": 51873,
+      "▁blessing": 51874,
+      "▁lifespan": 51875,
+      "xm": 51876,
+      "SDN": 51877,
+      "邯郸市": 51878,
+      "▁HAS": 51879,
+      "考前": 51880,
+      "▁DAY": 51881,
+      "这件事情": 51882,
+      "OAuth": 51883,
+      "▁arsen": 51884,
+      "crafted": 51885,
+      "▁programmers": 51886,
+      "每种": 51887,
+      "管理学": 51888,
+      "不可避免": 51889,
+      "▁rumors": 51890,
+      "凝胶": 51891,
+      "arna": 51892,
+      "被执行人名下": 51893,
+      "▁Gaming": 51894,
+      "▁vinegar": 51895,
+      "nPresident": 51896,
+      "▁histogram": 51897,
+      "两个数": 51898,
+      "你就是": 51899,
+      "刑期执行至": 51900,
+      "▁expire": 51901,
+      "▁Charter": 51902,
+      "▁hopeful": 51903,
+      "八个": 51904,
+      "的合理": 51905,
+      "的形象": 51906,
+      "▁IgE": 51907,
+      "narray": 51908,
+      "▁Verify": 51909,
+      "▁alters": 51910,
+      "方正": 51911,
+      "点为": 51912,
+      "PCA": 51913,
+      "应该如何": 51914,
+      "▁promoters": 51915,
+      "看了一": 51916,
+      "▁Belle": 51917,
+      "▁postpon": 51918,
+      "▁evidenced": 51919,
+      "vu": 51920,
+      "arton": 51921,
+      "▁Manila": 51922,
+      "▁creatinine": 51923,
+      "就拿": 51924,
+      "在世界": 51925,
+      "▁obscure": 51926,
+      "DV": 51927,
+      "一曲": 51928,
+      "且不": 51929,
+      "▁RNAs": 51930,
+      "marked": 51931,
+      "进货": 51932,
+      "的未来": 51933,
+      "数据中心": 51934,
+      "病房": 51935,
+      "轻薄": 51936,
+      "IDI": 51937,
+      "Such": 51938,
+      "nlibrary": 51939,
+      "▁tetr": 51940,
+      "▁Nicole": 51941,
+      "▁notions": 51942,
+      "一朝": 51943,
+      "基建": 51944,
+      "好感": 51945,
+      "玩家的": 51946,
+      "grown": 51947,
+      "▁emphasizes": 51948,
+      "大体": 51949,
+      "\"/\"": 51950,
+      "/\",": 51951,
+      "▁feminist": 51952,
+      "▁Cyt": 51953,
+      "公共交通": 51954,
+      "France": 51955,
+      "▁Croatian": 51956,
+      "▁encouragement": 51957,
+      "展厅": 51958,
+      "装机": 51959,
+      "林业局": 51960,
+      "());": 51961,
+      "党建工作": 51962,
+      "持股": 51963,
+      "提炼": 51964,
+      "是世界": 51965,
+      "大部分人": 51966,
+      "Guide": 51967,
+      "▁OAuth": 51968,
+      "▁Solomon": 51969,
+      "▁encoder": 51970,
+      "乖乖": 51971,
+      "标杆": 51972,
+      "留守": 51973,
+      "预案": 51974,
+      "tadd": 51975,
+      "Trait": 51976,
+      "▁Unix": 51977,
+      "▁Hidden": 51978,
+      "▁Slovakia": 51979,
+      "自适应": 51980,
+      "将自己的": 51981,
+      "▁产品类型": 51982,
+      "▁Surve": 51983,
+      "▁Tables": 51984,
+      "▁genuinely": 51985,
+      "▁superconduct": 51986,
+      "一见": 51987,
+      "圣母": 51988,
+      "不少人": 51989,
+      "如今的": 51990,
+      "▁\",\"": 51991,
+      "trivial": 51992,
+      "▁meanwhile": 51993,
+      "二手车": 51994,
+      "Iiw": 51995,
+      "▁\",\\": 51996,
+      "党风廉政": 51997,
+      "▁Kingston": 51998,
+      "▁seamlessly": 51999,
+      "用工": 52000,
+      "迪士尼": 52001,
+      "▁shred": 52002,
+      "linkedin": 52003,
+      "以上资料仅供参考": 52004,
+      "申请人于": 52005,
+      "百度小说": 52006,
+      "▁公司性质": 52007,
+      "▁snail": 52008,
+      "▁percutaneous": 52009,
+      "光绪": 52010,
+      "能用": 52011,
+      "那一刻": 52012,
+      "▁lasers": 52013,
+      "金陵": 52014,
+      "tone": 52015,
+      "具体情况还请": 52016,
+      "▁directional": 52017,
+      "化了": 52018,
+      "复旦大学": 52019,
+      "WINDO": 52020,
+      "Started": 52021,
+      "儿科": 52022,
+      "班车": 52023,
+      "rar": 52024,
+      "元左右": 52025,
+      "▁NATO": 52026,
+      "ún": 52027,
+      "没用": 52028,
+      "中国国际": 52029,
+      "Games": 52030,
+      "Picture": 52031,
+      "nProject": 52032,
+      "具体情况还请咨询医生": 52033,
+      "▁薛": 52034,
+      "反感": 52035,
+      "向北": 52036,
+      "catal": 52037,
+      "exican": 52038,
+      "▁Marks": 52039,
+      "▁Vintage": 52040,
+      "▁amplified": 52041,
+      "▁paperwork": 52042,
+      "第十二章": 52043,
+      "▁orch": 52044,
+      "nEmail": 52045,
+      "▁elbow": 52046,
+      "贷款基准利率": 52047,
+      "爆破": 52048,
+      "红薯": 52049,
+      "dow": 52050,
+      "变速箱": 52051,
+      "timed": 52052,
+      "▁Rosa": 52053,
+      "confidence": 52054,
+      "汉堡": 52055,
+      "oho": 52056,
+      "也是有": 52057,
+      "▁erect": 52058,
+      "Sources": 52059,
+      "▁Widget": 52060,
+      "▁refractive": 52061,
+      "台州": 52062,
+      "回味": 52063,
+      "▁Pipe": 52064,
+      "▁fellows": 52065,
+      "Organization": 52066,
+      "▁Um": 52067,
+      "▁imperfect": 52068,
+      "▁succession": 52069,
+      "▁Fo": 52070,
+      "南通市": 52071,
+      "▁Seems": 52072,
+      "▁disks": 52073,
+      "culture": 52074,
+      "▁glomerular": 52075,
+      "RJ": 52076,
+      "讲授": 52077,
+      "经常会": 52078,
+      "inkles": 52079,
+      "筛查": 52080,
+      "谈论": 52081,
+      "▁más": 52082,
+      "▁Kang": 52083,
+      "▁nore": 52084,
+      "Overlay": 52085,
+      "▁allocate": 52086,
+      "accessible": 52087,
+      "Zn": 52088,
+      "票价": 52089,
+      "ugh": 52090,
+      "Proto": 52091,
+      "▁汤": 52092,
+      "不光": 52093,
+      "▁tram": 52094,
+      "frag": 52095,
+      "▁Polynomial": 52096,
+      "▁pesticides": 52097,
+      "的风格": 52098,
+      "▁IPv": 52099,
+      "▁Laws": 52100,
+      "▁libert": 52101,
+      "▁creations": 52102,
+      "参与者": 52103,
+      "对学生": 52104,
+      "▁Bike": 52105,
+      "▁Sally": 52106,
+      "▁Darwin": 52107,
+      "▁RBI": 52108,
+      "ostruct": 52109,
+      "▁Anglican": 52110,
+      "文库": 52111,
+      "离去": 52112,
+      "的良好": 52113,
+      "在自己的": 52114,
+      "每一个人": 52115,
+      "almost": 52116,
+      "▁Acute": 52117,
+      "▁Perform": 52118,
+      "滋事": 52119,
+      "etur": 52120,
+      "▁AAA": 52121,
+      "景点的距离": 52122,
+      "▁mainland": 52123,
+      "三家": 52124,
+      "新房": 52125,
+      "离谱": 52126,
+      "较轻": 52127,
+      "对我来说": 52128,
+      "builds": 52129,
+      "▁Merit": 52130,
+      "circuit": 52131,
+      "tlogger": 52132,
+      "vetica": 52133,
+      "号行政处罚决定书": 52134,
+      "就别": 52135,
+      "本赛季": 52136,
+      "赵某某": 52137,
+      "veal": 52138,
+      "asone": 52139,
+      "▁Monthly": 52140,
+      "▁Paradise": 52141,
+      "▁affective": 52142,
+      "Rx": 52143,
+      "复读": 52144,
+      "评比": 52145,
+      "▁Zag": 52146,
+      "ilogy": 52147,
+      "景点的距离如下": 52148,
+      "安宁": 52149,
+      "datab": 52150,
+      "▁DEBUG": 52151,
+      "拥有移动电话": 52152,
+      "▁Albany": 52153,
+      "▁Solving": 52154,
+      "▁averages": 52155,
+      "▁国际": 52156,
+      "多样的": 52157,
+      "▁Worker": 52158,
+      "▁professionalism": 52159,
+      "KV": 52160,
+      "opies": 52161,
+      "othesis": 52162,
+      "▁stickers": 52163,
+      "草本": 52164,
+      "链条": 52165,
+      "Styled": 52166,
+      "▁sandy": 52167,
+      "▁spicy": 52168,
+      "▁psycho": 52169,
+      "presentation": 52170,
+      "之谜": 52171,
+      "▁GW": 52172,
+      "中央电视台": 52173,
+      "▁Fifty": 52174,
+      "撤回起诉处理": 52175,
+      "▁Bosnia": 52176,
+      "warnings": 52177,
+      "▁lymphoid": 52178,
+      "▁Anniversary": 52179,
+      "▁intersections": 52180,
+      "等设施": 52181,
+      "商业模式": 52182,
+      "▁distributor": 52183,
+      "Park": 52184,
+      "▁Nacional": 52185,
+      "故本院": 52186,
+      "的比较": 52187,
+      "▁Dum": 52188,
+      "▁washer": 52189,
+      "▁chaotic": 52190,
+      "▁notices": 52191,
+      "红星": 52192,
+      "个数是": 52193,
+      "▁Haus": 52194,
+      "▁Arrays": 52195,
+      "匈牙": 52196,
+      "示范区": 52197,
+      "▁Cedar": 52198,
+      "onge": 52199,
+      "▁iterate": 52200,
+      "▁fragmentation": 52201,
+      "办的": 52202,
+      "大棚": 52203,
+      "看病": 52204,
+      "byter": 52205,
+      "▁Diss": 52206,
+      "▁aorta": 52207,
+      "▁Shannon": 52208,
+      "▁upwards": 52209,
+      "▁付": 52210,
+      "知己": 52211,
+      "evo": 52212,
+      "akov": 52213,
+      "LIBRARY": 52214,
+      "▁Providence": 52215,
+      "升本": 52216,
+      "块的": 52217,
+      "扭转": 52218,
+      "爱情的": 52219,
+      "nTown": 52220,
+      "▁Hugo": 52221,
+      "▁erected": 52222,
+      "▁theaters": 52223,
+      "转折": 52224,
+      "和经济": 52225,
+      "LAST": 52226,
+      "icans": 52227,
+      "imshow": 52228,
+      "isfile": 52229,
+      "▁Anything": 52230,
+      "不放": 52231,
+      "树叶": 52232,
+      "浪潮": 52233,
+      "配色": 52234,
+      "的免费": 52235,
+      "这是个": 52236,
+      "exual": 52237,
+      "▁Pf": 52238,
+      "▁Sic": 52239,
+      "这是什么": 52240,
+      "▁UCLA": 52241,
+      "dismiss": 52242,
+      "▁pianist": 52243,
+      "▁translator": 52244,
+      "EI": 52245,
+      "▁SAS": 52246,
+      "和谐社会": 52247,
+      "号民事裁定书": 52248,
+      "▁crunch": 52249,
+      "▁destructive": 52250,
+      "分段": 52251,
+      "来袭": 52252,
+      "燕子": 52253,
+      "qing": 52254,
+      "Diagnostics": 52255,
+      "分销": 52256,
+      "正常使用": 52257,
+      "Credit": 52258,
+      "finger": 52259,
+      "▁progressively": 52260,
+      "是她": 52261,
+      "车管": 52262,
+      "防爆": 52263,
+      "()}": 52264,
+      "的版本": 52265,
+      "IFIC": 52266,
+      "mgmt": 52267,
+      "▁wherein": 52268,
+      "declarations": 52269,
+      "Brien": 52270,
+      "▁expiration": 52271,
+      "严密": 52272,
+      "各行": 52273,
+      "蜜蜂": 52274,
+      "oby": 52275,
+      "▁用料": 52276,
+      "▁GRO": 52277,
+      "专业领域": 52278,
+      "▁Bhar": 52279,
+      "你所": 52280,
+      "后发现": 52281,
+      "▁Dim": 52282,
+      "▁Wish": 52283,
+      "▁subs": 52284,
+      "iforms": 52285,
+      "▁insisted": 52286,
+      "▁cytoplasm": 52287,
+      "▁defendant": 52288,
+      "桃源": 52289,
+      "欠据": 52290,
+      "COD": 52291,
+      "另一半": 52292,
+      "cerol": 52293,
+      "▁Jacques": 52294,
+      "▁stripped": 52295,
+      "适用缓刑": 52296,
+      "genre": 52297,
+      "▁Pare": 52298,
+      "▁corrupt": 52299,
+      "▁fibrillation": 52300,
+      "nUl": 52301,
+      "申请执行人陈": 52302,
+      "▁formulate": 52303,
+      "▁headaches": 52304,
+      "▁%\\": 52305,
+      "▁Won": 52306,
+      "vault": 52307,
+      "▁cutoff": 52308,
+      "商务部": 52309,
+      "nExt": 52310,
+      "▁Gan": 52311,
+      "odied": 52312,
+      "▁Identifier": 52313,
+      "▁carcinogenesis": 52314,
+      "▁Nordic": 52315,
+      "Evaluate": 52316,
+      "▁overtime": 52317,
+      "accharides": 52318,
+      "全社会": 52319,
+      "rots": 52320,
+      "的是什么": 52321,
+      "由被执行人": 52322,
+      "▁subdiv": 52323,
+      "下设": 52324,
+      "海滨": 52325,
+      "▁Chern": 52326,
+      "nAfrican": 52327,
+      "族自治县人民法院": 52328,
+      "一定能": 52329,
+      "双十一": 52330,
+      "的调查": 52331,
+      "bang": 52332,
+      "▁Ish": 52333,
+      "▁Dayton": 52334,
+      "▁Morrison": 52335,
+      "位列": 52336,
+      "厅长": 52337,
+      "打得": 52338,
+      "sport": 52339,
+      "▁Buzz": 52340,
+      "▁Milk": 52341,
+      "分枝": 52342,
+      "恋情": 52343,
+      "整整": 52344,
+      "▁制片": 52345,
+      "国土资": 52346,
+      "tight": 52347,
+      "▁Neck": 52348,
+      "微信小程序": 52349,
+      "表的": 52350,
+      "在这些": 52351,
+      "▁AMS": 52352,
+      "ncategory": 52353,
+      "▁preserves": 52354,
+      "污泥": 52355,
+      "减函数": 52356,
+      "披针形": 52357,
+      "▁EPS": 52358,
+      "receptor": 52359,
+      "▁photore": 52360,
+      "▁Caroline": 52361,
+      "硫化": 52362,
+      "eden": 52363,
+      "个人认为": 52364,
+      "inatal": 52365,
+      "▁pedest": 52366,
+      "温岭": 52367,
+      "环路": 52368,
+      "votes": 52369,
+      "▁slowing": 52370,
+      "傍晚": 52371,
+      "大风": 52372,
+      "常务理事": 52373,
+      "▁refugee": 52374,
+      "移项": 52375,
+      "▁BG": 52376,
+      "黑色的": 52377,
+      "flux": 52378,
+      "▁repetition": 52379,
+      "十首": 52380,
+      "总计": 52381,
+      "▁HSV": 52382,
+      "navigate": 52383,
+      "▁cushion": 52384,
+      "预处理": 52385,
+      ">\\',": 52386,
+      "ouns": 52387,
+      "▁Rabbit": 52388,
+      "Modifier": 52389,
+      "▁crushed": 52390,
+      "维斯": 52391,
+      "而且是": 52392,
+      "Mess": 52393,
+      "▁anion": 52394,
+      "报到": 52395,
+      "colm": 52396,
+      "edes": 52397,
+      "ikan": 52398,
+      "提出了一种": 52399,
+      "itably": 52400,
+      "orbent": 52401,
+      "▁Allan": 52402,
+      "▁impart": 52403,
+      "▁grabbed": 52404,
+      "▁fermentation": 52405,
+      "提琴": 52406,
+      "雷锋": 52407,
+      "COR": 52408,
+      "▁fright": 52409,
+      "挺好": 52410,
+      "农户数": 52411,
+      "▁SVG": 52412,
+      "crumbs": 52413,
+      "▁salaries": 52414,
+      "▁deduction": 52415,
+      "▁constructive": 52416,
+      "Cd": 52417,
+      "时分": 52418,
+      "hoot": 52419,
+      "ialis": 52420,
+      "▁urea": 52421,
+      "写下": 52422,
+      "读作": 52423,
+      "转运": 52424,
+      "民主党": 52425,
+      "NOWN": 52426,
+      "oque": 52427,
+      "arnish": 52428,
+      "积极参加政治": 52429,
+      "▁Visa": 52430,
+      "nRegular": 52431,
+      "不带": 52432,
+      "潜艇": 52433,
+      "将上述": 52434,
+      "道路为": 52435,
+      "nVir": 52436,
+      "▁Jets": 52437,
+      "▁Elliott": 52438,
+      "▁Transmission": 52439,
+      "乔木": 52440,
+      "现身": 52441,
+      "侵犯了": 52442,
+      "hetical": 52443,
+      "▁asphalt": 52444,
+      "甘草": 52445,
+      "▁ws": 52446,
+      "的大力": 52447,
+      "nmax": 52448,
+      "被害人的": 52449,
+      "nExpl": 52450,
+      "QtCore": 52451,
+      "▁tunes": 52452,
+      "▁Explore": 52453,
+      "▁authenticated": 52454,
+      "UMN": 52455,
+      "dou": 52456,
+      "克里斯": 52457,
+      "旗舰店": 52458,
+      "liness": 52459,
+      "▁puppy": 52460,
+      "QUAL": 52461,
+      "▁Buc": 52462,
+      "cancer": 52463,
+      "Defaults": 52464,
+      "mutation": 52465,
+      "留给": 52466,
+      "诺贝尔": 52467,
+      "inqu": 52468,
+      "umbers": 52469,
+      "▁Luxury": 52470,
+      "二〇一七年十月": 52471,
+      "▁Cardiff": 52472,
+      "▁Recreation": 52473,
+      "▁Fi": 52474,
+      "则实数": 52475,
+      "▁hairst": 52476,
+      "▁relegated": 52477,
+      "在实际": 52478,
+      "bags": 52479,
+      "ovel": 52480,
+      "产品重量": 52481,
+      "▁surgeries": 52482,
+      "ек": 52483,
+      "不忍": 52484,
+      "幕后": 52485,
+      "打下": 52486,
+      "▁lend": 52487,
+      "▁Codes": 52488,
+      "Relation": 52489,
+      "踏上": 52490,
+      "不要再": 52491,
+      "较多的": 52492,
+      "ndel": 52493,
+      "▁Bradford": 52494,
+      "▁shortened": 52495,
+      "根本没有": 52496,
+      "美国总统": 52497,
+      "行政拘留": 52498,
+      "▁impairments": 52499,
+      "Girl": 52500,
+      "▁Trav": 52501,
+      "▁wont": 52502,
+      "管理委员会": 52503,
+      "太小": 52504,
+      "约有": 52505,
+      "不少于": 52506,
+      "就成了": 52507,
+      "这几天": 52508,
+      "越来越大": 52509,
+      "品德": 52510,
+      "拦截": 52511,
+      "的经历": 52512,
+      "STOR": 52513,
+      "Heading": 52514,
+      "陈代谢": 52515,
+      "Prime": 52516,
+      "nshow": 52517,
+      "家协会会员": 52518,
+      "乙方": 52519,
+      "挪威": 52520,
+      "香的": 52521,
+      "我这个": 52522,
+      "效果好": 52523,
+      "特异性": 52524,
+      "看电影": 52525,
+      "▁taxa": 52526,
+      "▁Categories": 52527,
+      "IBM": 52528,
+      "USH": 52529,
+      "wrong": 52530,
+      "▁macrom": 52531,
+      "残忍": 52532,
+      "英烈": 52533,
+      "Miss": 52534,
+      "▁|\\\\": 52535,
+      "▁orbits": 52536,
+      "▁Nicolas": 52537,
+      "▁asymmetry": 52538,
+      "▁proposing": 52539,
+      "dL": 52540,
+      "世俗": 52541,
+      "侧重": 52542,
+      "认识的": 52543,
+      "ryan": 52544,
+      "所选定的": 52545,
+      "你我": 52546,
+      "计画": 52547,
+      "▁от": 52548,
+      "▁Gauss": 52549,
+      "Minimum": 52550,
+      "▁rookie": 52551,
+      "▁Whenever": 52552,
+      "wl": 52553,
+      "胚胎": 52554,
+      "kie": 52555,
+      "nur": 52556,
+      "腹腔镜": 52557,
+      "▁hrs": 52558,
+      "前段时间": 52559,
+      "武侯": 52560,
+      "nSl": 52561,
+      "先根据": 52562,
+      "Apps": 52563,
+      "▁cep": 52564,
+      "guides": 52565,
+      "▁ceilings": 52566,
+      "▁evenings": 52567,
+      "wY": 52568,
+      "南宋": 52569,
+      "神圣": 52570,
+      "跌幅": 52571,
+      "躲避": 52572,
+      "该村到": 52573,
+      "页无正文": 52574,
+      "nHand": 52575,
+      "AreEqual": 52576,
+      "▁radians": 52577,
+      "foundation": 52578,
+      "责任感": 52579,
+      "朋友聚餐": 52580,
+      "成名": 52581,
+      "老爸": 52582,
+      "事业的": 52583,
+      "iade": 52584,
+      "ificance": 52585,
+      "▁spatially": 52586,
+      "厌恶": 52587,
+      "cyst": 52588,
+      "eches": 52589,
+      "都有自己的": 52590,
+      "延吉": 52591,
+      "鼠标点击": 52592,
+      "▁deng": 52593,
+      "nAlong": 52594,
+      "surname": 52595,
+      "▁Sprint": 52596,
+      "▁gastrop": 52597,
+      "▁被": 52598,
+      "私有": 52599,
+      "阳市人民法院": 52600,
+      "▁acquis": 52601,
+      "Frequency": 52602,
+      "送上": 52603,
+      "最后一次": 52604,
+      "金融危机": 52605,
+      "ornings": 52606,
+      "▁Martha": 52607,
+      "▁solids": 52608,
+      "水管": 52609,
+      "沧桑": 52610,
+      "装卸": 52611,
+      "辨别": 52612,
+      "abh": 52613,
+      "分割线": 52614,
+      "iona": 52615,
+      "▁Ole": 52616,
+      "▁repr": 52617,
+      "▁vague": 52618,
+      "▁ansible": 52619,
+      "发芽": 52620,
+      "地层": 52621,
+      "比的": 52622,
+      "otoxic": 52623,
+      "ourced": 52624,
+      "乌拉": 52625,
+      "机油": 52626,
+      "ncia": 52627,
+      "▁GTP": 52628,
+      "tfile": 52629,
+      "▁Youtube": 52630,
+      "心目": 52631,
+      "▁工作分工": 52632,
+      "pheres": 52633,
+      "▁deter": 52634,
+      "nsystem": 52635,
+      "万亿": 52636,
+      "参与了": 52637,
+      "洗干净": 52638,
+      "▁Aus": 52639,
+      "tcout": 52640,
+      "▁Validate": 52641,
+      "城里": 52642,
+      "verters": 52643,
+      "▁Outcomes": 52644,
+      "个赞": 52645,
+      "教养": 52646,
+      "tret": 52647,
+      "getData": 52648,
+      "安娜": 52649,
+      "法语": 52650,
+      "研究与": 52651,
+      "鲜明的": 52652,
+      "与被告李": 52653,
+      "▁rods": 52654,
+      "ealous": 52655,
+      "▁Reproduce": 52656,
+      "▁sf": 52657,
+      "oley": 52658,
+      "无烟客房": 52659,
+      "spread": 52660,
+      "▁Olive": 52661,
+      "nVolume": 52662,
+      "▁suburban": 52663,
+      "▁reputable": 52664,
+      "▁Pant": 52665,
+      "inative": 52666,
+      "▁查看": 52667,
+      "不相同": 52668,
+      "▁ODE": 52669,
+      "如果你的": 52670,
+      "plica": 52671,
+      "survey": 52672,
+      "▁sexuality": 52673,
+      "▁unforgettable": 52674,
+      "部份": 52675,
+      "▁aa": 52676,
+      "▁kh": 52677,
+      "ermo": 52678,
+      "▁Spaces": 52679,
+      "▁toured": 52680,
+      "bj": 52681,
+      "(\\\\)": 52682,
+      "▁),\\": 52683,
+      "创建一个": 52684,
+      "Sorry": 52685,
+      "▁Angle": 52686,
+      "摸索": 52687,
+      "▁ruler": 52688,
+      "不允许携带宠物": 52689,
+      "出错": 52690,
+      "最简": 52691,
+      "芬兰": 52692,
+      "rizz": 52693,
+      "彻底破裂": 52694,
+      "▁敏": 52695,
+      "切尔": 52696,
+      "同盟": 52697,
+      "thr": 52698,
+      "▁ét": 52699,
+      "读卡器": 52700,
+      "▁mens": 52701,
+      "rabbit": 52702,
+      "兴建": 52703,
+      "周六": 52704,
+      "anus": 52705,
+      "▁biome": 52706,
+      "通往": 52707,
+      "卫生部": 52708,
+      "▁Prison": 52709,
+      "往年": 52710,
+      "了一张": 52711,
+      "创新能力": 52712,
+      "▁Fragment": 52713,
+      "▁enlarged": 52714,
+      "nInformation": 52715,
+      "▁Continental": 52716,
+      "▁conflicting": 52717,
+      "双曲线的": 52718,
+      "年的时间": 52719,
+      "Units": 52720,
+      "能够认罪悔罪": 52721,
+      "石英": 52722,
+      "▁MVP": 52723,
+      "元或查封": 52724,
+      "▁Cann": 52725,
+      "Construct": 52726,
+      "Repositories": 52727,
+      "个个": 52728,
+      "植被": 52729,
+      "Cut": 52730,
+      "Popup": 52731,
+      "edged": 52732,
+      "上诉于山东省": 52733,
+      "建设工程施工": 52734,
+      "▁outfits": 52735,
+      "MSC": 52736,
+      "分析与": 52737,
+      "已完成": 52738,
+      "nEvents": 52739,
+      "ontally": 52740,
+      "▁palate": 52741,
+      "▁tensile": 52742,
+      "engineering": 52743,
+      "nPopulation": 52744,
+      "在下": 52745,
+      "水域": 52746,
+      "海量": 52747,
+      "锦绣": 52748,
+      "价格的": 52749,
+      "高度的": 52750,
+      "ière": 52751,
+      "▁JOIN": 52752,
+      "互为相反数": 52753,
+      "outfile": 52754,
+      "▁christ": 52755,
+      "晶晶": 52756,
+      "▁其六": 52757,
+      "后来的": 52758,
+      "▁Bottom": 52759,
+      "▁Santos": 52760,
+      "个子": 52761,
+      "nHealth": 52762,
+      "▁extracting": 52763,
+      "果树": 52764,
+      "构成的": 52765,
+      "▁Pie": 52766,
+      "idium": 52767,
+      "▁Stick": 52768,
+      "▁ruins": 52769,
+      "▁Strept": 52770,
+      "内膜": 52771,
+      "本章": 52772,
+      "▁`$": 52773,
+      "ynes": 52774,
+      "▁lan": 52775,
+      "▁astro": 52776,
+      "张文": 52777,
+      "\">{": 52778,
+      "isher": 52779,
+      "▁Notre": 52780,
+      ">{{": 52781,
+      "▁tm": 52782,
+      "妈妈的": 52783,
+      "nLess": 52784,
+      "▁fluoride": 52785,
+      "▁conceived": 52786,
+      "▁Replacement": 52787,
+      "庆典": 52788,
+      "开荒": 52789,
+      "▁tk": 52790,
+      "▁Coul": 52791,
+      "administration": 52792,
+      ")∵": 52793,
+      "学了": 52794,
+      "风电": 52795,
+      "▁upright": 52796,
+      "本裁定送达后立即执行": 52797,
+      "女装": 52798,
+      "师兄": 52799,
+      "猎人": 52800,
+      "lake": 52801,
+      "▁Jal": 52802,
+      "琢磨": 52803,
+      "表皮": 52804,
+      "身处": 52805,
+      "型企业": 52806,
+      "带头人": 52807,
+      "参加会议": 52808,
+      "是什么样的": 52809,
+      "nWindows": 52810,
+      "developed": 52811,
+      "▁endometrial": 52812,
+      "▁钱": 52813,
+      "大王": 52814,
+      "heroku": 52815,
+      "▁红袖添香网": 52816,
+      "Resolved": 52817,
+      "▁Colonel": 52818,
+      "▁Ultimately": 52819,
+      "主讲": 52820,
+      "年由": 52821,
+      "Zoom": 52822,
+      "want": 52823,
+      "▁Ads": 52824,
+      "很大程度上": 52825,
+      "Connell": 52826,
+      "nNorwegian": 52827,
+      "_]": 52828,
+      "不留": 52829,
+      "马达": 52830,
+      "nNO": 52831,
+      "但对于": 52832,
+      "feat": 52833,
+      "▁Burns": 52834,
+      "▁Herzegovina": 52835,
+      "uen": 52836,
+      "▁Xi": 52837,
+      "和学习": 52838,
+      "或者代表人的人数提出副本": 52839,
+      "常态化": 52840,
+      "采用的": 52841,
+      "▁Zach": 52842,
+      "▁receivers": 52843,
+      "▁莫": 52844,
+      "出不": 52845,
+      "山川": 52846,
+      "称呼": 52847,
+      "证人证言": 52848,
+      "▁rectal": 52849,
+      "▁adjunct": 52850,
+      "五首": 52851,
+      "贯通": 52852,
+      "报道称": 52853,
+      "ODAY": 52854,
+      "▁PAC": 52855,
+      "odegrad": 52856,
+      "▁Writers": 52857,
+      "▁exponentially": 52858,
+      "▁immunotherapy": 52859,
+      "宴会": 52860,
+      "oise": 52861,
+      "可以作为": 52862,
+      "▁avatar": 52863,
+      "▁harvesting": 52864,
+      "网贷": 52865,
+      "jit": 52866,
+      "电子书": 52867,
+      "▁pom": 52868,
+      "家教": 52869,
+      "搞得": 52870,
+      "爱护": 52871,
+      "虐待": 52872,
+      "▁NB": 52873,
+      "▁salts": 52874,
+      "▁stirring": 52875,
+      "▁compartments": 52876,
+      "太难": 52877,
+      "长治": 52878,
+      "akis": 52879,
+      "▁TCR": 52880,
+      "avors": 52881,
+      "仍不": 52882,
+      "油耗": 52883,
+      "重阳节": 52884,
+      "懂的": 52885,
+      "片子": 52886,
+      "督察": 52887,
+      "收费是": 52888,
+      "政府部门": 52889,
+      "Reuters": 52890,
+      "findall": 52891,
+      "集团股份有限公司": 52892,
+      "▁inorganic": 52893,
+      "▁structurally": 52894,
+      "番禺": 52895,
+      "大家一起": 52896,
+      "电动汽车": 52897,
+      "onset": 52898,
+      "▁Transition": 52899,
+      "东盟": 52900,
+      "PCs": 52901,
+      "关系是": 52902,
+      "理性的": 52903,
+      "▁slog": 52904,
+      "LON": 52905,
+      "aty": 52906,
+      "爱尔兰": 52907,
+      "▁编辑推荐": 52908,
+      "德州": 52909,
+      "又一个": 52910,
+      "证据的": 52911,
+      "▁ASD": 52912,
+      "synthes": 52913,
+      "▁virtue": 52914,
+      "▁antibacterial": 52915,
+      "nDES": 52916,
+      "▁Append": 52917,
+      "▁RESULTS": 52918,
+      "▁Secondly": 52919,
+      ",%": 52920,
+      "二项": 52921,
+      "日出": 52922,
+      "牵头": 52923,
+      "blo": 52924,
+      "产品在": 52925,
+      "体内的": 52926,
+      "可靠的": 52927,
+      "))))\\": 52928,
+      "▁restor": 52929,
+      "▁breathtaking": 52930,
+      "两句": 52931,
+      "之事": 52932,
+      "他能": 52933,
+      "秸秆": 52934,
+      "▁ald": 52935,
+      "在卷佐证": 52936,
+      "▁Mayo": 52937,
+      "▁ammonia": 52938,
+      "assignment": 52939,
+      "▁elasticity": 52940,
+      "EventHandler": 52941,
+      "名胜": 52942,
+      "▁purity": 52943,
+      "▁resize": 52944,
+      "nContent": 52945,
+      "inguished": 52946,
+      "传闻": 52947,
+      "reon": 52948,
+      "ronym": 52949,
+      "nested": 52950,
+      "▁Dollar": 52951,
+      "▁丽": 52952,
+      "切开": 52953,
+      "拖延": 52954,
+      "多少元": 52955,
+      "防腐剂": 52956,
+      "▁vibe": 52957,
+      "nSteve": 52958,
+      "也对": 52959,
+      "至少有": 52960,
+      "认定书": 52961,
+      "utory": 52962,
+      "nSix": 52963,
+      "nfloat": 52964,
+      "▁Monroe": 52965,
+      "▁Sharma": 52966,
+      "▁Javascript": 52967,
+      "球形": 52968,
+      "直角坐标系": 52969,
+      "broken": 52970,
+      "versed": 52971,
+      "▁storms": 52972,
+      "▁seasoned": 52973,
+      "夜里": 52974,
+      "马力": 52975,
+      "oza": 52976,
+      "yum": 52977,
+      "一下吧": 52978,
+      "▁bure": 52979,
+      "Compile": 52980,
+      "▁telescope": 52981,
+      "冬瓜": 52982,
+      "我再": 52983,
+      "大学毕业": 52984,
+      "▁Randy": 52985,
+      "河镇": 52986,
+      "然的": 52987,
+      "▁microarray": 52988,
+      "托福": 52989,
+      "nCH": 52990,
+      "▁填词": 52991,
+      "ecko": 52992,
+      "extended": 52993,
+      "▁Outside": 52994,
+      "▁accomplishments": 52995,
+      "tuint": 52996,
+      "▁idiopathic": 52997,
+      "长效": 52998,
+      "▁tonnes": 52999,
+      "二〇一六年九月": 53000,
+      "昌平": 53001,
+      "有啥": 53002,
+      "落花": 53003,
+      "▁orally": 53004,
+      "本裁定立即执行": 53005,
+      "▁constructions": 53006,
+      "qc": 53007,
+      "心事": 53008,
+      "然后从": 53009,
+      "▁hav": 53010,
+      "输出格式": 53011,
+      "▁\\'\\')": 53012,
+      "教学模式": 53013,
+      "▁Particle": 53014,
+      "皱纹": 53015,
+      "upo": 53016,
+      "爱国主义": 53017,
+      "或不": 53018,
+      "▁runway": 53019,
+      "Clinical": 53020,
+      "ServiceProvider": 53021,
+      "errno": 53022,
+      "actice": 53023,
+      "▁Depot": 53024,
+      "▁wavelengths": 53025,
+      "▁interpretations": 53026,
+      "便民": 53027,
+      "奇妙": 53028,
+      "改制": 53029,
+      "患者在": 53030,
+      "▁leaked": 53031,
+      "▁ancestors": 53032,
+      "▁Serializable": 53033,
+      "deen": 53034,
+      "▁Ordered": 53035,
+      "RN": 53036,
+      "兵器": 53037,
+      "ego": 53038,
+      "Icons": 53039,
+      "eenth": 53040,
+      "trunc": 53041,
+      "certificate": 53042,
+      "▁coordinating": 53043,
+      "胆囊": 53044,
+      "可以先": 53045,
+      "的业务": 53046,
+      "计算公式": 53047,
+      "▁nome": 53048,
+      "county": 53049,
+      "▁Screens": 53050,
+      "渤海": 53051,
+      "结肠": 53052,
+      "对你的": 53053,
+      "海淀区": 53054,
+      "▁brutal": 53055,
+      "▁Topology": 53056,
+      "气泡": 53057,
+      "知情": 53058,
+      "借款人民币": 53059,
+      "太好": 53060,
+      "自负": 53061,
+      "表现的": 53062,
+      "体系建设": 53063,
+      "RH": 53064,
+      "ulk": 53065,
+      "nChina": 53066,
+      "nWorking": 53067,
+      "▁enlight": 53068,
+      "史料": 53069,
+      "抗诉": 53070,
+      "画风": 53071,
+      "其次是": 53072,
+      "rogens": 53073,
+      "Pending": 53074,
+      "▁hydroxy": 53075,
+      "所给": 53076,
+      "新月": 53077,
+      "缤纷": 53078,
+      "黑客": 53079,
+      "第二步": 53080,
+      "noop": 53081,
+      "▁multil": 53082,
+      "▁limestone": 53083,
+      "▁collectors": 53084,
+      "▁hemodynamic": 53085,
+      "▁谭": 53086,
+      "开会": 53087,
+      ">(\"": 53088,
+      "▁mc": 53089,
+      "大辞典": 53090,
+      "essing": 53091,
+      "所选定的客房": 53092,
+      "holding": 53093,
+      "iatrics": 53094,
+      "DY": 53095,
+      "不可思议": 53096,
+      "中国社会": 53097,
+      "▁doubts": 53098,
+      "购车": 53099,
+      "博士生导师": 53100,
+      "nRadio": 53101,
+      "ntable": 53102,
+      "▁unexpl": 53103,
+      "rimental": 53104,
+      "▁sidebar": 53105,
+      "portfolio": 53106,
+      "▁Malaysian": 53107,
+      "耦合": 53108,
+      "还说": 53109,
+      "法律允许": 53110,
+      "▁stall": 53111,
+      "nNatural": 53112,
+      "▁expires": 53113,
+      "jam": 53114,
+      "地进行": 53115,
+      "教研室": 53116,
+      "危害程度": 53117,
+      "▁Budd": 53118,
+      "▁remix": 53119,
+      "单击": 53120,
+      "第六节": 53121,
+      "网站上": 53122,
+      "▁Aer": 53123,
+      "▁violated": 53124,
+      "浮云": 53125,
+      "▁Fischer": 53126,
+      "随笔": 53127,
+      "▁Cir": 53128,
+      "新闻出版": 53129,
+      "adapt": 53130,
+      "▁Slow": 53131,
+      "▁abused": 53132,
+      "房管": 53133,
+      "糊涂": 53134,
+      "原审判决": 53135,
+      "▁Charleston": 53136,
+      "▁fundamentally": 53137,
+      "平日": 53138,
+      "霸道": 53139,
+      "有其他": 53140,
+      "shaw": 53141,
+      "▁Sebastian": 53142,
+      "▁overwhelmed": 53143,
+      "▁politically": 53144,
+      "曾因": 53145,
+      "淇淋": 53146,
+      "较快": 53147,
+      "auses": 53148,
+      "pickle": 53149,
+      "▁gospel": 53150,
+      "▁Midwest": 53151,
+      "文笔": 53152,
+      "odil": 53153,
+      "▁Bea": 53154,
+      "▁Devices": 53155,
+      "在社会": 53156,
+      "▁Syr": 53157,
+      "▁bib": 53158,
+      "二审被上诉人": 53159,
+      "化工有限公司": 53160,
+      "▁Samples": 53161,
+      "▁supposedly": 53162,
+      "▁و": 53163,
+      "▁👍": 53164,
+      "了啊": 53165,
+      "孕育": 53166,
+      "那段": 53167,
+      "▁WP": 53168,
+      "都无法": 53169,
+      "_[": 53170,
+      "惠州": 53171,
+      "▁Sap": 53172,
+      "▁Devil": 53173,
+      "COMMAND": 53174,
+      "二〇一六年十二": 53175,
+      "Mid": 53176,
+      "好像是": 53177,
+      "服务员": 53178,
+      "urai": 53179,
+      "农村合作": 53180,
+      "shield": 53181,
+      "▁greens": 53182,
+      "招收": 53183,
+      "▁IX": 53184,
+      "数据范围": 53185,
+      "▁jupyter": 53186,
+      "▁superoxide": 53187,
+      "ück": 53188,
+      "▁文章": 53189,
+      "▁Frankfurt": 53190,
+      "虚构": 53191,
+      "▁人物生平": 53192,
+      "▁Serie": 53193,
+      "▁Confirm": 53194,
+      "▁intricate": 53195,
+      "▁microorganisms": 53196,
+      "健脾": 53197,
+      "惠普": 53198,
+      "麻痹": 53199,
+      "病人的": 53200,
+      "nSET": 53201,
+      "▁ROM": 53202,
+      "BLOCK": 53203,
+      "▁furnace": 53204,
+      "operators": 53205,
+      "▁Recommend": 53206,
+      "▁HU": 53207,
+      "注意的是": 53208,
+      "表现在": 53209,
+      "▁hij": 53210,
+      "ndarray": 53211,
+      "二〇一六年六月": 53212,
+      "nComplete": 53213,
+      "▁Cardinal": 53214,
+      "▁settling": 53215,
+      "▁ah": 53216,
+      "世纪的": 53217,
+      "功能性": 53218,
+      "utsche": 53219,
+      "stackrel": 53220,
+      "▁halfway": 53221,
+      "西斯": 53222,
+      "▁Intr": 53223,
+      "由于被执行人": 53224,
+      "请求判令被告": 53225,
+      "hardware": 53226,
+      "▁caching": 53227,
+      "审视": 53228,
+      "火烧": 53229,
+      "}{|": 53230,
+      "科学记数法的表示": 53231,
+      "第一百七十条第一款第": 53232,
+      "℃\\": 53233,
+      "元未": 53234,
+      "谁知道": 53235,
+      "spectrum": 53236,
+      "▁compost": 53237,
+      "▁celebrates": 53238,
+      "▁lieutenant": 53239,
+      "nWH": 53240,
+      "▁barr": 53241,
+      "法律的规定": 53242,
+      "expanded": 53243,
+      "一碗": 53244,
+      "效能": 53245,
+      "jah": 53246,
+      "对他们": 53247,
+      "截至目前": 53248,
+      "▁Goals": 53249,
+      "▁miniature": 53250,
+      "增幅": 53251,
+      "族的": 53252,
+      "皮带": 53253,
+      "内容和": 53254,
+      "攀枝花": 53255,
+      "orget": 53256,
+      "▁Piano": 53257,
+      "▁orphan": 53258,
+      "▁vacant": 53259,
+      "▁moderator": 53260,
+      "出任": 53261,
+      "投递": 53262,
+      "胃癌": 53263,
+      "重伤": 53264,
+      "\"\">\\": 53265,
+      "为空": 53266,
+      "与他人": 53267,
+      "理解的": 53268,
+      "assa": 53269,
+      "卫星频道": 53270,
+      "文化交流": 53271,
+      "pless": 53272,
+      "▁Lyon": 53273,
+      "JK": 53274,
+      "听课": 53275,
+      "▁现代": 53276,
+      "余弦定理": 53277,
+      "抗日战争": 53278,
+      "▁pills": 53279,
+      "日公开开庭审理了本案": 53280,
+      "三份": 53281,
+      "目录下": 53282,
+      "送达地址": 53283,
+      "WRITE": 53284,
+      "nDigital": 53285,
+      "▁commute": 53286,
+      "上图": 53287,
+      "之大": 53288,
+      "总队": 53289,
+      "ritz": 53290,
+      "▁Vine": 53291,
+      "▁Doors": 53292,
+      "上列": 53293,
+      "▁IU": 53294,
+      "stitu": 53295,
+      "建材有限公司": 53296,
+      "istency": 53297,
+      "▁尹": 53298,
+      "学好": 53299,
+      "重来": 53300,
+      "Bur": 53301,
+      "xFF": 53302,
+      "和朋友": 53303,
+      "▁Meth": 53304,
+      "editary": 53305,
+      "▁fibres": 53306,
+      "Tc": 53307,
+      "京津": 53308,
+      "累累": 53309,
+      "▁Env": 53310,
+      "三个方面": 53311,
+      "▁Goal": 53312,
+      "ertext": 53313,
+      "▁fused": 53314,
+      "recognized": 53315,
+      "前世": 53316,
+      "高素质": 53317,
+      "tpass": 53318,
+      "▁outrage": 53319,
+      "▁Producer": 53320,
+      "制止": 53321,
+      "高通": 53322,
+      "而言之": 53323,
+      "npermalink": 53324,
+      "不作": 53325,
+      "居室": 53326,
+      "汉服": 53327,
+      "▁Mats": 53328,
+      "▁Explain": 53329,
+      "▁Predict": 53330,
+      "nabstract": 53331,
+      "连环": 53332,
+      "医护人员": 53333,
+      "▁Dana": 53334,
+      "Encode": 53335,
+      "Published": 53336,
+      "▁thermodynamic": 53337,
+      "▁Å": 53338,
+      "摧毁": 53339,
+      "收音": 53340,
+      "盖上": 53341,
+      "adores": 53342,
+      "第二百四十二": 53343,
+      "▁filepath": 53344,
+      "▁immature": 53345,
+      "侵蚀": 53346,
+      "专业人士": 53347,
+      "▁werden": 53348,
+      "shake": 53349,
+      "▁Gamma": 53350,
+      "▁forex": 53351,
+      "▁decays": 53352,
+      "▁refres": 53353,
+      "▁Winnipeg": 53354,
+      "立志": 53355,
+      "宝贵的": 53356,
+      "由上诉人": 53357,
+      "的主要原因": 53358,
+      "▁conda": 53359,
+      "▁tribal": 53360,
+      "小巧": 53361,
+      "收视": 53362,
+      "高楼": 53363,
+      "▁tribes": 53364,
+      "▁arguably": 53365,
+      "Sdk": 53366,
+      "cyt": 53367,
+      "orno": 53368,
+      "▁Dro": 53369,
+      "Codec": 53370,
+      "二〇一七年十一": 53371,
+      "▁Feedback": 53372,
+      ">]": 53373,
+      "东坡": 53374,
+      "向社会": 53375,
+      "nnum": 53376,
+      "▁antif": 53377,
+      "▁cyclo": 53378,
+      "在服刑期间的": 53379,
+      "nSample": 53380,
+      "▁prosecutors": 53381,
+      "一辈": 53382,
+      "打折": 53383,
+      "颜色的": 53384,
+      "liquid": 53385,
+      "您所选定的客房": 53386,
+      "▁salivary": 53387,
+      "区和": 53388,
+      "赛道": 53389,
+      "mgr": 53390,
+      "▁HBO": 53391,
+      "▁intentionally": 53392,
+      "▁解": 53393,
+      "行列": 53394,
+      "{{\\": 53395,
+      "▁skiers": 53396,
+      "双层": 53397,
+      "对各": 53398,
+      "沉迷": 53399,
+      "车厢": 53400,
+      "国庆节": 53401,
+      "教学方法": 53402,
+      "▁Barb": 53403,
+      "##############\\": 53404,
+      "specs": 53405,
+      "▁progressed": 53406,
+      "也叫": 53407,
+      "何以": 53408,
+      "护航": 53409,
+      "立项": 53410,
+      "▁kon": 53411,
+      "▁就": 53412,
+      "孤单": 53413,
+      "CAN": 53414,
+      "也有很多": 53415,
+      "Allowed": 53416,
+      "▁anastom": 53417,
+      "复查": 53418,
+      "小人": 53419,
+      "报案": 53420,
+      "▁ko": 53421,
+      "inis": 53422,
+      "nHas": 53423,
+      "nNov": 53424,
+      "之所": 53425,
+      "唿吁": 53426,
+      "赞美": 53427,
+      "个体户": 53428,
+      "érie": 53429,
+      "中心主任": 53430,
+      "▁Rates": 53431,
+      "▁Sustainable": 53432,
+      "得利": 53433,
+      "对管辖权": 53434,
+      "nfield": 53435,
+      "生子": 53436,
+      "建筑安装": 53437,
+      "▁dosing": 53438,
+      "▁Comprehensive": 53439,
+      "▁electrostatic": 53440,
+      "坏死": 53441,
+      "buck": 53442,
+      "amics": 53443,
+      "▁underlie": 53444,
+      "▁首先": 53445,
+      "参考答案": 53446,
+      "crest": 53447,
+      "▁STEM": 53448,
+      "▁openings": 53449,
+      "nTransportation": 53450,
+      "定性": 53451,
+      "为实现": 53452,
+      "inez": 53453,
+      "▁taper": 53454,
+      "号民事判决书的执行": 53455,
+      "摄像机": 53456,
+      "▁scanned": 53457,
+      "▁civilians": 53458,
+      "敦煌": 53459,
+      "Gly": 53460,
+      "ndi": 53461,
+      "国家重点": 53462,
+      "享受三包服务": 53463,
+      "罪犯奖惩审批表": 53464,
+      "▁suburbs": 53465,
+      "centration": 53466,
+      "▁interplay": 53467,
+      "向外": 53468,
+      "安居": 53469,
+      "底盘": 53470,
+      "极了": 53471,
+      "是完全": 53472,
+      "grep": 53473,
+      "symbols": 53474,
+      "建设工程有限公司": 53475,
+      "▁TL": 53476,
+      "审理中": 53477,
+      "月下旬": 53478,
+      "购买了": 53479,
+      "nKe": 53480,
+      "于一体": 53481,
+      "产品和": 53482,
+      "的了解": 53483,
+      "past": 53484,
+      "▁GST": 53485,
+      "国家安全": 53486,
+      "amarin": 53487,
+      "▁recurs": 53488,
+      "▁delegation": 53489,
+      "款的": 53490,
+      "families": 53491,
+      "三课": 53492,
+      "外墙": 53493,
+      "正值": 53494,
+      "章程": 53495,
+      "BUT": 53496,
+      "▁nig": 53497,
+      "ilers": 53498,
+      "paren": 53499,
+      "preset": 53500,
+      "据此解答即可": 53501,
+      "▁ceased": 53502,
+      "▁culturally": 53503,
+      "数和": 53504,
+      "会将": 53505,
+      "点多": 53506,
+      "▁适量": 53507,
+      "▁Kot": 53508,
+      "几个方面": 53509,
+      "▁gulp": 53510,
+      "▁squeeze": 53511,
+      "▁unaware": 53512,
+      "ár": 53513,
+      "nConvert": 53514,
+      "▁fathers": 53515,
+      "▁oscillation": 53516,
+      "亦可": 53517,
+      "池塘": 53518,
+      "CLI": 53519,
+      "▁authenticate": 53520,
+      "六合": 53521,
+      "称作": 53522,
+      "▁命令": 53523,
+      "▁unpack": 53524,
+      "进阶": 53525,
+      "这对于": 53526,
+      "tips": 53527,
+      "▁{$": 53528,
+      "▁Pall": 53529,
+      "▁Maker": 53530,
+      "减为有期徒刑": 53531,
+      "非物质文化遗产": 53532,
+      "▁ellipse": 53533,
+      "CALL": 53534,
+      "▁发行时间": 53535,
+      "▁freezer": 53536,
+      "▁侯": 53537,
+      "祖国的": 53538,
+      "/{{": 53539,
+      "icl": 53540,
+      ",“": 53541,
+      "构思": 53542,
+      "禁毒": 53543,
+      "GPL": 53544,
+      "神奇的": 53545,
+      "规模的": 53546,
+      "nPolish": 53547,
+      "addresses": 53548,
+      "嘻嘻": 53549,
+      "安装在": 53550,
+      "▁clash": 53551,
+      "▁Forbes": 53552,
+      "▁Holdings": 53553,
+      "fk": 53554,
+      "▁带": 53555,
+      "布朗": 53556,
+      "故此": 53557,
+      "身世": 53558,
+      "降噪": 53559,
+      "也算是": 53560,
+      "区域的": 53561,
+      "往往是": 53562,
+      "的一份": 53563,
+      "__),": 53564,
+      "tstruct": 53565,
+      "▁mosaic": 53566,
+      "▁Fortune": 53567,
+      "好运": 53568,
+      "RAM": 53569,
+      "▁Brom": 53570,
+      "皇宫": 53571,
+      "▁worm": 53572,
+      "▁Doing": 53573,
+      "▁motivate": 53574,
+      "tdb": 53575,
+      "anza": 53576,
+      "nBill": 53577,
+      "subtitle": 53578,
+      "▁ultrasonic": 53579,
+      "死去": 53580,
+      "证明其": 53581,
+      "▁Apost": 53582,
+      "favicon": 53583,
+      "跟进": 53584,
+      "召开了": 53585,
+      "▁condo": 53586,
+      "Community": 53587,
+      "▁withdrawn": 53588,
+      "▁disadvantages": 53589,
+      "学霸": 53590,
+      "推迟": 53591,
+      "FTER": 53592,
+      "工艺美术": 53593,
+      "▁breeze": 53594,
+      "▁Satellite": 53595,
+      "高分子": 53596,
+      "tall": 53597,
+      "生命周期": 53598,
+      "资金周转": 53599,
+      "Street": 53600,
+      "▁Strategies": 53601,
+      "▁Kad": 53602,
+      "ovolta": 53603,
+      "▁Scala": 53604,
+      "▁mushrooms": 53605,
+      "国安": 53606,
+      "距离最近的": 53607,
+      "文化传播有限公司": 53608,
+      "稀土": 53609,
+      "不能再": 53610,
+      "▁ling": 53611,
+      "Partial": 53612,
+      "▁Budapest": 53613,
+      "积水": 53614,
+      "等优点": 53615,
+      "nSep": 53616,
+      "▁cogn": 53617,
+      "olphins": 53618,
+      "不解": 53619,
+      "稳固": 53620,
+      "还被": 53621,
+      "会比较": 53622,
+      "作用下": 53623,
+      "群体的": 53624,
+      "▁Tyr": 53625,
+      "▁kar": 53626,
+      "▁ethnicity": 53627,
+      "antlr": 53628,
+      "▁teammates": 53629,
+      "中区": 53630,
+      "日元": 53631,
+      "运送": 53632,
+      "tras": 53633,
+      "éric": 53634,
+      "▁Shim": 53635,
+      "Correct": 53636,
+      "▁Identify": 53637,
+      "公差": 53638,
+      "寻衅": 53639,
+      "为原告": 53640,
+      "编号为": 53641,
+      "pulse": 53642,
+      "▁plural": 53643,
+      "▁Christine": 53644,
+      "二手房": 53645,
+      "优缺点": 53646,
+      "依然是": 53647,
+      "fuel": 53648,
+      "▁bik": 53649,
+      "▁Above": 53650,
+      "▁Katie": 53651,
+      "▁logos": 53652,
+      "▁intermediates": 53653,
+      "Sales": 53654,
+      "▁Diana": 53655,
+      "本裁定立即开始执行": 53656,
+      "未见": 53657,
+      "查到": 53658,
+      "胰腺": 53659,
+      "用一个": 53660,
+      "▁autologous": 53661,
+      "▁KS": 53662,
+      "体育馆": 53663,
+      "hler": 53664,
+      "▁diary": 53665,
+      "ittings": 53666,
+      "ordinate": 53667,
+      "题文": 53668,
+      "ław": 53669,
+      "Mini": 53670,
+      "▁Mask": 53671,
+      "▁启动方式": 53672,
+      "快点": 53673,
+      "车载": 53674,
+      "酷睿": 53675,
+      "IFE": 53676,
+      "生活费": 53677,
+      "**/\\": 53678,
+      "occer": 53679,
+      "nfinal": 53680,
+      "▁Derek": 53681,
+      "扭矩": 53682,
+      "行业中": 53683,
+      "mach": 53684,
+      "nfig": 53685,
+      "你是不是": 53686,
+      "这批": 53687,
+      "▁ik": 53688,
+      "月开始": 53689,
+      "▁proliferative": 53690,
+      "胡同": 53691,
+      "(${": 53692,
+      "这些东西": 53693,
+      "之日起十五日内": 53694,
+      "▁catches": 53695,
+      "wk": 53696,
+      "她就": 53697,
+      "价格为": 53698,
+      "abus": 53699,
+      "nserver": 53700,
+      "ntarget": 53701,
+      "HttpRequest": 53702,
+      "最美的": 53703,
+      "一直没有": 53704,
+      "Parts": 53705,
+      "之年": 53706,
+      "ánd": 53707,
+      "rano": 53708,
+      "第六十条": 53709,
+      "meaning": 53710,
+      "凝固": 53711,
+      "建新": 53712,
+      "江汉": 53713,
+      "尼西亚": 53714,
+      "的产生": 53715,
+      "重金属": 53716,
+      "acuse": 53717,
+      "应准予离婚": 53718,
+      "olinergic": 53719,
+      "洗碗": 53720,
+      "产品有": 53721,
+      "日消息": 53722,
+      "::__": 53723,
+      "哥伦比亚": 53724,
+      "清偿责任": 53725,
+      "athing": 53726,
+      "▁Drake": 53727,
+      "mounted": 53728,
+      "▁postdoctoral": 53729,
+      "亲手": 53730,
+      "无双": 53731,
+      "birds": 53732,
+      "nHelp": 53733,
+      "二〇一七年九": 53734,
+      "Ws": 53735,
+      "考入": 53736,
+      "建议你": 53737,
+      "Roll": 53738,
+      "▁fname": 53739,
+      "▁breeds": 53740,
+      "氧化物": 53741,
+      "nGen": 53742,
+      "消防安全": 53743,
+      "▁tempfile": 53744,
+      "短缺": 53745,
+      "スト": 53746,
+      "等我": 53747,
+      "香烟": 53748,
+      "▁CVD": 53749,
+      "▁erotic": 53750,
+      "nBesides": 53751,
+      "▁Belfast": 53752,
+      "▁planner": 53753,
+      "▁benchmarks": 53754,
+      "▁凯": 53755,
+      "弥漫": 53756,
+      "情愿": 53757,
+      "拓扑": 53758,
+      "纲要": 53759,
+      "损害国家": 53760,
+      "▁accommodations": 53761,
+      "破损": 53762,
+      "镀锌": 53763,
+      "nNOTE": 53764,
+      "▁Musk": 53765,
+      "有什么区别": 53766,
+      "▁parcel": 53767,
+      "recipitation": 53768,
+      "椅子": 53769,
+      "段子": 53770,
+      "中文版": 53771,
+      "只是在": 53772,
+      "setter": 53773,
+      "nMunicip": 53774,
+      "身子": 53775,
+      "如实供述自己": 53776,
+      "nComment": 53777,
+      "短线": 53778,
+      "▁Ib": 53779,
+      "aminergic": 53780,
+      "▁perceive": 53781,
+      "▁implication": 53782,
+      "▁demographics": 53783,
+      "清真": 53784,
+      "芳草": 53785,
+      "pas": 53786,
+      "Flat": 53787,
+      "▁sir": 53788,
+      "▁shrimp": 53789,
+      "区人民检察院指派检察员": 53790,
+      "tN": 53791,
+      "治病": 53792,
+      "Pot": 53793,
+      "全过程": 53794,
+      "避免了": 53795,
+      "▁jur": 53796,
+      "ENTRY": 53797,
+      "▁Mant": 53798,
+      "Letter": 53799,
+      "▁RIGHT": 53800,
+      "ropolis": 53801,
+      "▁cinnamon": 53802,
+      "▁Pediatric": 53803,
+      "黄豆": 53804,
+      "cum": 53805,
+      "宣判后": 53806,
+      "输出样例": 53807,
+      "▁Lank": 53808,
+      "olkata": 53809,
+      "soever": 53810,
+      "nLocated": 53811,
+      "redient": 53812,
+      "expressing": 53813,
+      "半月": 53814,
+      "的本质": 53815,
+      "bull": 53816,
+      "▁republic": 53817,
+      "▁Bool": 53818,
+      "▁Lives": 53819,
+      "二〇一四年十二月": 53820,
+      "会见": 53821,
+      "▁abst": 53822,
+      "▁staple": 53823,
+      "ReadOnly": 53824,
+      "奔腾": 53825,
+      "清澈": 53826,
+      "音质": 53827,
+      "IVER": 53828,
+      "inen": 53829,
+      "▁ENV": 53830,
+      "consuming": 53831,
+      "▁fragrance": 53832,
+      "了下": 53833,
+      "BODY": 53834,
+      "▁guild": 53835,
+      "杭州市中级人民法院": 53836,
+      "社群": 53837,
+      "淮南市": 53838,
+      "▁Vid": 53839,
+      "▁finale": 53840,
+      "▁decorator": 53841,
+      "▁capacitance": 53842,
+      "▁synergistic": 53843,
+      "汤匙": 53844,
+      "胡某": 53845,
+      "▁专业": 53846,
+      "mocha": 53847,
+      "objective": 53848,
+      "▁factorial": 53849,
+      "▁headquartered": 53850,
+      "一件事": 53851,
+      "舞台上": 53852,
+      "▁Tik": 53853,
+      "nPage": 53854,
+      "nApple": 53855,
+      "▁squir": 53856,
+      "▁onwards": 53857,
+      "▁delimiter": 53858,
+      "▁Importantly": 53859,
+      "YX": 53860,
+      "人品": 53861,
+      "德拉": 53862,
+      "总值": 53863,
+      "格子": 53864,
+      "鸡腿": 53865,
+      "으로": 53866,
+      "发展中": 53867,
+      "court": 53868,
+      "▁Were": 53869,
+      "▁lush": 53870,
+      "Scalar": 53871,
+      "▁catechol": 53872,
+      "▁conception": 53873,
+      "十佳": 53874,
+      "忍者": 53875,
+      "监局": 53876,
+      "防空": 53877,
+      "▁DN": 53878,
+      "rotic": 53879,
+      "DOMAIN": 53880,
+      "可控": 53881,
+      "faker": 53882,
+      "trail": 53883,
+      "Setter": 53884,
+      "▁START": 53885,
+      "▁candles": 53886,
+      "▁aspiration": 53887,
+      "▁WT": 53888,
+      "完全是": 53889,
+      "thirds": 53890,
+      "authenticated": 53891,
+      "▁非": 53892,
+      "格斗": 53893,
+      "▁Dear": 53894,
+      "Closed": 53895,
+      "igraphy": 53896,
+      "北侧": 53897,
+      "▁ji": 53898,
+      "hton": 53899,
+      "▁virgin": 53900,
+      "下厨房": 53901,
+      "▁occupy": 53902,
+      "巴尔": 53903,
+      "合格的": 53904,
+      "严重影响": 53905,
+      "▁traps": 53906,
+      "▁grading": 53907,
+      "▁enhancements": 53908,
+      "新版": 53909,
+      "昆仑": 53910,
+      "健身房": 53911,
+      "lastName": 53912,
+      "▁ultraviolet": 53913,
+      "▁康": 53914,
+      "他家": 53915,
+      "에서": 53916,
+      "▁湖南": 53917,
+      "也应该": 53918,
+      "nExpress": 53919,
+      "各主要景点的距离如下": 53920,
+      "先端": 53921,
+      "喷涂": 53922,
+      "平和": 53923,
+      "日渐": 53924,
+      "院内": 53925,
+      "Handlers": 53926,
+      "explicit": 53927,
+      "我有一": 53928,
+      "▁RSV": 53929,
+      "可以减刑": 53930,
+      "开端": 53931,
+      "便宜的": 53932,
+      "Learn": 53933,
+      "接连": 53934,
+      "曼联": 53935,
+      "猜想": 53936,
+      "▁mk": 53937,
+      "▁(((": 53938,
+      "▁上海市": 53939,
+      "▁Cave": 53940,
+      "▁Named": 53941,
+      "住所地重庆市": 53942,
+      "director": 53943,
+      "▁Antarctica": 53944,
+      "频频": 53945,
+      "CXX": 53946,
+      "▁Dw": 53947,
+      "第二季": 53948,
+      "不同的是": 53949,
+      "levation": 53950,
+      "▁约": 53951,
+      "吃饱": 53952,
+      "大笑": 53953,
+      "kees": 53954,
+      "▁Duty": 53955,
+      "inement": 53956,
+      "APPLICATION": 53957,
+      "|:": 53958,
+      "▁вы": 53959,
+      "▁HBV": 53960,
+      "▁osteo": 53961,
+      "▁penis": 53962,
+      "Tooltip": 53963,
+      "▁abrupt": 53964,
+      "Vertical": 53965,
+      "▁congregation": 53966,
+      "并被": 53967,
+      "匈牙利": 53968,
+      "英特尔": 53969,
+      "}\\',": 53970,
+      "▁TLS": 53971,
+      "▁ain": 53972,
+      "工程机械": 53973,
+      "容留他人吸毒罪": 53974,
+      "▁nucleic": 53975,
+      "▁Patriots": 53976,
+      "repositories": 53977,
+      "苏轼": 53978,
+      "rox": 53979,
+      "不怎么": 53980,
+      "没有异议": 53981,
+      "输入样例": 53982,
+      "被执行人的财产": 53983,
+      "nUnivers": 53984,
+      "▁rotated": 53985,
+      "organized": 53986,
+      "亚科": 53987,
+      "引资": 53988,
+      "星的": 53989,
+      "谎言": 53990,
+      "并根据": 53991,
+      "显卡芯片": 53992,
+      "这段时间": 53993,
+      "▁tide": 53994,
+      "▁polyg": 53995,
+      "ABSTRACT": 53996,
+      "▁顾": 53997,
+      "也比": 53998,
+      "底座": 53999,
+      "还需": 54000,
+      "rsa": 54001,
+      "HTTPS": 54002,
+      "▁Sexy": 54003,
+      "nfirst": 54004,
+      "▁saxophone": 54005,
+      "▁calculates": 54006,
+      "npip": 54007,
+      "▁Melissa": 54008,
+      ":])\\": 54009,
+      "ambo": 54010,
+      "zier": 54011,
+      "第六十七": 54012,
+      "▁fuse": 54013,
+      "thedocs": 54014,
+      "二〇一五年七月": 54015,
+      "▁figuring": 54016,
+      "之城": 54017,
+      "书桌": 54018,
+      "大到": 54019,
+      "署名": 54020,
+      "进行治疗": 54021,
+      "▁Wong": 54022,
+      "texpect": 54023,
+      "▁envelop": 54024,
+      "▁upholst": 54025,
+      "▁Socialist": 54026,
+      "▁photographic": 54027,
+      "位上": 54028,
+      "墙体": 54029,
+      "红梅": 54030,
+      "▁FK": 54031,
+      "mins": 54032,
+      "outil": 54033,
+      "nUnlike": 54034,
+      "▁Marion": 54035,
+      ";;\\": 54036,
+      "Jobs": 54037,
+      "▁Via": 54038,
+      "优秀教师": 54039,
+      "▁unbiased": 54040,
+      "素描": 54041,
+      "遵纪": 54042,
+      "elm": 54043,
+      "可能在": 54044,
+      "马某某": 54045,
+      "裁判文书": 54046,
+      "pitch": 54047,
+      "▁hose": 54048,
+      "overed": 54049,
+      "▁shark": 54050,
+      "▁javafx": 54051,
+      "▁Evening": 54052,
+      "▁ambiguous": 54053,
+      "德斯": 54054,
+      "CRP": 54055,
+      "▁的值": 54056,
+      "还存在": 54057,
+      "始终坚持": 54058,
+      "enums": 54059,
+      "▁Railways": 54060,
+      "▁Luxembourg": 54061,
+      "花花": 54062,
+      "监督检查": 54063,
+      "不知道怎么": 54064,
+      "female": 54065,
+      "ptiles": 54066,
+      "actoring": 54067,
+      "▁但": 54068,
+      "少吃": 54069,
+      "王小": 54070,
+      "的在": 54071,
+      "容易被": 54072,
+      "▁bak": 54073,
+      "▁disrupted": 54074,
+      "呼吁": 54075,
+      "通辽": 54076,
+      "青蛙": 54077,
+      "大理石": 54078,
+      "据统计": 54079,
+      "▁GeV": 54080,
+      "▁Pon": 54081,
+      "▁ally": 54082,
+      "▁sophomore": 54083,
+      "喜欢吃": 54084,
+      "epit": 54085,
+      "rique": 54086,
+      "▁Mond": 54087,
+      "nModel": 54088,
+      "íc": 54089,
+      "TRY": 54090,
+      "▁联赛": 54091,
+      "在本院": 54092,
+      "配套设施": 54093,
+      "了大": 54094,
+      "人来": 54095,
+      "的金": 54096,
+      "SIM": 54097,
+      "中国家": 54098,
+      "Law": 54099,
+      "加油站": 54100,
+      "jump": 54101,
+      "globals": 54102,
+      "拼音是": 54103,
+      "标的为": 54104,
+      "still": 54105,
+      "▁Alumin": 54106,
+      "▁Bru": 54107,
+      "nmessage": 54108,
+      "符合法律规定的撤诉条件": 54109,
+      "steam": 54110,
+      "增减": 54111,
+      "结节": 54112,
+      "风力": 54113,
+      "vez": 54114,
+      "liner": 54115,
+      "▁suffers": 54116,
+      "有可供执行的财产": 54117,
+      "Mu": 54118,
+      "无辜": 54119,
+      "爱豆": 54120,
+      "▁Canad": 54121,
+      "▁vigil": 54122,
+      "▁diplom": 54123,
+      "▁detrimental": 54124,
+      "豪门": 54125,
+      "本人的": 54126,
+      "▁Shi": 54127,
+      "attempt": 54128,
+      "投机": 54129,
+      "雷斯": 54130,
+      "的观点": 54131,
+      "▁wow": 54132,
+      "顺着": 54133,
+      "fft": 54134,
+      "▁运行": 54135,
+      "潍坊市": 54136,
+      "帝国主义": 54137,
+      "▁Skip": 54138,
+      "▁abdomen": 54139,
+      "水池": 54140,
+      "流出": 54141,
+      "▁INFO": 54142,
+      "planes": 54143,
+      "ricing": 54144,
+      "osopher": 54145,
+      "▁bipart": 54146,
+      "填空": 54147,
+      "近来": 54148,
+      "decl": 54149,
+      "dens": 54150,
+      "▁\"<<": 54151,
+      "▁Ort": 54152,
+      "▁intox": 54153,
+      "▁Rhodes": 54154,
+      "▁resorts": 54155,
+      "▁migrated": 54156,
+      "遏制": 54157,
+      "▁Epic": 54158,
+      "▁ruin": 54159,
+      "▁pupil": 54160,
+      "▁retains": 54161,
+      "▁Differences": 54162,
+      "奇艺": 54163,
+      "荷花": 54164,
+      "ibi": 54165,
+      "Heap": 54166,
+      "▁unm": 54167,
+      "开展经营活动": 54168,
+      "普京": 54169,
+      "深渊": 54170,
+      "自理": 54171,
+      "]},": 54172,
+      "两人的": 54173,
+      "Peer": 54174,
+      "▁resonant": 54175,
+      ")&": 54176,
+      "绝不": 54177,
+      "谣言": 54178,
+      "($\"": 54179,
+      "▁常": 54180,
+      "▁sv": 54181,
+      "({},": 54182,
+      "jian": 54183,
+      "ToOne": 54184,
+      "▁Celtic": 54185,
+      "相应价值的财产": 54186,
+      "▁modifier": 54187,
+      "▁辅料": 54188,
+      "养殖业": 54189,
+      "egro": 54190,
+      "nFre": 54191,
+      "▁CIA": 54192,
+      "▁debit": 54193,
+      "reporter": 54194,
+      "▁gasoline": 54195,
+      "sci": 54196,
+      "ska": 54197,
+      "▁GF": 54198,
+      "▁牺牲": 54199,
+      "幸福感": 54200,
+      "教学楼": 54201,
+      "看出来": 54202,
+      "国债": 54203,
+      "的头": 54204,
+      "管理处": 54205,
+      "▁Ion": 54206,
+      "▁Hannah": 54207,
+      "Visibility": 54208,
+      "rz": 54209,
+      "▁nM": 54210,
+      "向量的": 54211,
+      "▁SOL": 54212,
+      "▁turbulent": 54213,
+      "图解": 54214,
+      "重要性": 54215,
+      "大家可以": 54216,
+      "▁crest": 54217,
+      "▁Boeing": 54218,
+      "▁aggrav": 54219,
+      "▁paddle": 54220,
+      "▁VARCHAR": 54221,
+      "▁computes": 54222,
+      "▁neighbours": 54223,
+      "单曲": 54224,
+      "Jeff": 54225,
+      "overn": 54226,
+      "志愿服务队": 54227,
+      "▁nausea": 54228,
+      "风气": 54229,
+      "subplot": 54230,
+      "▁Battalion": 54231,
+      "inp": 54232,
+      "▁sph": 54233,
+      "Quote": 54234,
+      "▁nood": 54235,
+      "▁assim": 54236,
+      "▁dimer": 54237,
+      "▁Thermal": 54238,
+      "SOL": 54239,
+      "graf": 54240,
+      "国华": 54241,
+      "抗震": 54242,
+      "自己也": 54243,
+      "Checks": 54244,
+      "▁strokes": 54245,
+      "的刑罚": 54246,
+      "都不敢": 54247,
+      "META": 54248,
+      "▁BAL": 54249,
+      "▁Pione": 54250,
+      "▁troph": 54251,
+      "mockito": 54252,
+      "▁getAll": 54253,
+      "微创": 54254,
+      "无为": 54255,
+      "情况说明": 54256,
+      "ndone": 54257,
+      "owitz": 54258,
+      "▁Firstly": 54259,
+      "海波": 54260,
+      "ubin": 54261,
+      "▁osm": 54262,
+      "▁blowing": 54263,
+      "▁Rational": 54264,
+      "▁grandson": 54265,
+      "▁hardness": 54266,
+      "conditional": 54267,
+      "▁hemisphere": 54268,
+      "途经": 54269,
+      "▁NN": 54270,
+      "▁tomb": 54271,
+      "党的十九大": 54272,
+      "执行机关提供的": 54273,
+      "是人类": 54274,
+      "办公室副主任": 54275,
+      "▁manifestation": 54276,
+      "无心": 54277,
+      "▁tb": 54278,
+      "nlist": 54279,
+      "▁Lamp": 54280,
+      "▁narc": 54281,
+      "Hs": 54282,
+      "因为这个": 54283,
+      "▁Staf": 54284,
+      "Decorator": 54285,
+      "▁Kazakhstan": 54286,
+      "查证": 54287,
+      "▁hob": 54288,
+      "PERTY": 54289,
+      "▁Logo": 54290,
+      "itures": 54291,
+      "▁mRNAs": 54292,
+      "照样": 54293,
+      "私募": 54294,
+      "产阶级": 54295,
+      "这一天": 54296,
+      "专业学习": 54297,
+      "完成任务": 54298,
+      "Western": 54299,
+      "drivers": 54300,
+      "▁antire": 54301,
+      "▁SUMMARY": 54302,
+      "第一百四十五条的规定": 54303,
+      "▁perturbations": 54304,
+      "▁Written": 54305,
+      "Come": 54306,
+      "两个整数": 54307,
+      "▁Exped": 54308,
+      "joining": 54309,
+      "经庭审": 54310,
+      "▁consortium": 54311,
+      "▁Scholarship": 54312,
+      "▁Staphylococcus": 54313,
+      "墙上": 54314,
+      "▁版本": 54315,
+      "▁Percent": 54316,
+      "▁servicing": 54317,
+      "▁neuroscience": 54318,
+      "河西": 54319,
+      "社长": 54320,
+      "▁HI": 54321,
+      "选择出": 54322,
+      "asaki": 54323,
+      "▁antennas": 54324,
+      "JB": 54325,
+      "奠定": 54326,
+      "打动": 54327,
+      "法医": 54328,
+      "thought": 54329,
+      "▁eldest": 54330,
+      "高级人民法院于": 54331,
+      "名气": 54332,
+      "英镑": 54333,
+      "▁PSA": 54334,
+      "▁bamboo": 54335,
+      "nUpdated": 54336,
+      "人称": 54337,
+      "水泵": 54338,
+      "alks": 54339,
+      "taking": 54340,
+      "▁Dodge": 54341,
+      "▁Refer": 54342,
+      "新冠肺炎疫情": 54343,
+      "▁Surgical": 54344,
+      "▁disturbing": 54345,
+      "之夜": 54346,
+      "老龄": 54347,
+      "淋巴结": 54348,
+      "awan": 54349,
+      "▁Bom": 54350,
+      "▁Hammer": 54351,
+      "▁Palestine": 54352,
+      "物权": 54353,
+      "设置了": 54354,
+      "\"#\">": 54355,
+      "]{\\\\": 54356,
+      "▁gamers": 54357,
+      "▁segregation": 54358,
+      "独有的": 54359,
+      "的资金": 54360,
+      "重大的": 54361,
+      "国家图书馆": 54362,
+      "gather": 54363,
+      "有下列情形之一的": 54364,
+      "▁automate": 54365,
+      "nIndividual": 54366,
+      "无缘": 54367,
+      "Luc": 54368,
+      "学习了": 54369,
+      "Altern": 54370,
+      "▁Presidential": 54371,
+      "▁MonoBehaviour": 54372,
+      "才行": 54373,
+      "nJo": 54374,
+      "满满的": 54375,
+      "的性格": 54376,
+      "▁文学体裁": 54377,
+      "essential": 54378,
+      "专心": 54379,
+      "▁casc": 54380,
+      "Ticket": 54381,
+      "▁Sultan": 54382,
+      "vx": 54383,
+      "两周": 54384,
+      "改性": 54385,
+      "▁周围景观": 54386,
+      "nSELECT": 54387,
+      "打字": 54388,
+      "鱼肉": 54389,
+      "的身体": 54390,
+      "可以得到": 54391,
+      "▁Clara": 54392,
+      "ategorized": 54393,
+      "宜春": 54394,
+      "淮安": 54395,
+      "依法可": 54396,
+      "故本选项错误": 54397,
+      "▁terrorists": 54398,
+      "▁biodiversity": 54399,
+      "超高": 54400,
+      "首付": 54401,
+      "膝关节": 54402,
+      "mRNA": 54403,
+      "grant": 54404,
+      "▁navy": 54405,
+      "UTION": 54406,
+      "▁Heroes": 54407,
+      "▁conversions": 54408,
+      "玲珑": 54409,
+      "最主要": 54410,
+      "▁Orders": 54411,
+      "▁bedding": 54412,
+      "调和": 54413,
+      "baum": 54414,
+      "▁Powers": 54415,
+      "双色": 54416,
+      "不仅能": 54417,
+      "Nick": 54418,
+      "农业银行": 54419,
+      "assic": 54420,
+      "▁Catch": 54421,
+      "本院予以采纳": 54422,
+      "Toolbar": 54423,
+      "公开发": 54424,
+      "卖的": 54425,
+      "固态": 54426,
+      "▁没有": 54427,
+      "第一届": 54428,
+      "nWar": 54429,
+      "▁trab": 54430,
+      "arcoma": 54431,
+      "Enh": 54432,
+      "最开始": 54433,
+      "▁Holly": 54434,
+      "akedirs": 54435,
+      "会同": 54436,
+      "地主": 54437,
+      "大便": 54438,
+      "恩施": 54439,
+      "车展": 54440,
+      "国资委": 54441,
+      "ARGS": 54442,
+      "▁awake": 54443,
+      "blogspot": 54444,
+      "江上": 54445,
+      "特约": 54446,
+      "鲜美": 54447,
+      "因为她": 54448,
+      "$;": 54449,
+      "月薪": 54450,
+      "Too": 54451,
+      "本来就是": 54452,
+      "ComponentModel": 54453,
+      "莫名其妙": 54454,
+      "least": 54455,
+      "是怎么回事": 54456,
+      "▁Highly": 54457,
+      "▁skipped": 54458,
+      "▁Membership": 54459,
+      "体能": 54460,
+      "题解": 54461,
+      "所以要": 54462,
+      "selective": 54463,
+      "▁thirteen": 54464,
+      "イン": 54465,
+      "招呼": 54466,
+      "eln": 54467,
+      "子序列": 54468,
+      "中的作用": 54469,
+      "生长发育": 54470,
+      "thttp": 54471,
+      "nNotice": 54472,
+      "▁profitability": 54473,
+      "先是": 54474,
+      "基亚": 54475,
+      "总分": 54476,
+      "浮动": 54477,
+      "爵士": 54478,
+      "▁Fuj": 54479,
+      "▁OVER": 54480,
+      "向本院申请再审": 54481,
+      "持卡": 54482,
+      "科级": 54483,
+      "但没有": 54484,
+      "orously": 54485,
+      "分隔": 54486,
+      "去买": 54487,
+      "片刻": 54488,
+      "家企业": 54489,
+      "的参数": 54490,
+      "rant": 54491,
+      "公告送达": 54492,
+      "▁indoors": 54493,
+      "Spl": 54494,
+      "你就会": 54495,
+      "健康成长": 54496,
+      "的工作人员": 54497,
+      "pectral": 54498,
+      "▁bloggers": 54499,
+      "▁behavioural": 54500,
+      "▁Measurements": 54501,
+      "说得": 54502,
+      "量表": 54503,
+      "▁'%": 54504,
+      "的自然": 54505,
+      "还是一": 54506,
+      "▁trek": 54507,
+      "▁分布范围": 54508,
+      "回顾性分析": 54509,
+      "初衷": 54510,
+      "孤儿": 54511,
+      "葱花": 54512,
+      "▁Intro": 54513,
+      "▁chars": 54514,
+      "IBILITY": 54515,
+      "二中": 54516,
+      "透视": 54517,
+      "▁FD": 54518,
+      "忘记了": 54519,
+      "▁ICC": 54520,
+      "二〇一五年八月": 54521,
+      "则为": 54522,
+      "▁法国": 54523,
+      "▁Lenn": 54524,
+      "nWithout": 54525,
+      "▁winding": 54526,
+      "▁companions": 54527,
+      "▁Presentation": 54528,
+      "域为": 54529,
+      "领悟": 54530,
+      "Cases": 54531,
+      "scalar": 54532,
+      "▁gifted": 54533,
+      "▁assemblies": 54534,
+      "hk": 54535,
+      "到家": 54536,
+      "北宋": 54537,
+      "▁hp": 54538,
+      "▁sobre": 54539,
+      "▁Reduction": 54540,
+      "火了": 54541,
+      "寻衅滋事": 54542,
+      "▁fairy": 54543,
+      "ishable": 54544,
+      "▁starch": 54545,
+      "▁utmost": 54546,
+      "▁vocalist": 54547,
+      "天长": 54548,
+      "是老": 54549,
+      "ugo": 54550,
+      "▁ptr": 54551,
+      "▁spreads": 54552,
+      "▁southeastern": 54553,
+      "专人": 54554,
+      "药师": 54555,
+      "青青": 54556,
+      "yet": 54557,
+      "aday": 54558,
+      "▁Skype": 54559,
+      "取决于被执行人": 54560,
+      "国家自然科学基金": 54561,
+      "▁Lithuania": 54562,
+      "温暖的": 54563,
+      "方面的知识": 54564,
+      "Martin": 54565,
+      "▁acquisitions": 54566,
+      "чи": 54567,
+      "并行": 54568,
+      "煮熟": 54569,
+      "▁Kw": 54570,
+      "只能在": 54571,
+      "xhtml": 54572,
+      "orsche": 54573,
+      "▁kicking": 54574,
+      "▁Ferguson": 54575,
+      "▁researched": 54576,
+      "▁mathematicians": 54577,
+      "抛光": 54578,
+      "▁Uri": 54579,
+      "竞争力的": 54580,
+      "▁Charge": 54581,
+      "▁Purple": 54582,
+      "▁augmentation": 54583,
+      "苯丙胺": 54584,
+      "▁Wet": 54585,
+      "RV": 54586,
+      "五四": 54587,
+      "着自己": 54588,
+      "CKER": 54589,
+      "▁Semi": 54590,
+      "▁toes": 54591,
+      "aliases": 54592,
+      "▁synonym": 54593,
+      "乱世": 54594,
+      "RAW": 54595,
+      "▁CRM": 54596,
+      "▁subfamily": 54597,
+      "▁Jacksonville": 54598,
+      "三元": 54599,
+      "承揽": 54600,
+      "还包括": 54601,
+      "Mike": 54602,
+      "违约责任": 54603,
+      "tuser": 54604,
+      "Unicode": 54605,
+      "Overflow": 54606,
+      "丰满": 54607,
+      "在第": 54608,
+      "▁rode": 54609,
+      "legacy": 54610,
+      "▁gravel": 54611,
+      "Packages": 54612,
+      "年中": 54613,
+      "窗外": 54614,
+      "到案经过": 54615,
+      "▁Curtis": 54616,
+      "▁Panthers": 54617,
+      "▁Constants": 54618,
+      "▁alternatively": 54619,
+      "扫地": 54620,
+      "准确性": 54621,
+      "▁cosmic": 54622,
+      "▁fading": 54623,
+      "tdisplay": 54624,
+      "▁esophag": 54625,
+      "▁layouts": 54626,
+      "▁collects": 54627,
+      "不掉": 54628,
+      "科的": 54629,
+      "Sin": 54630,
+      "fif": 54631,
+      "债权债务": 54632,
+      "▁PROF": 54633,
+      "▁Funeral": 54634,
+      "▁vomiting": 54635,
+      "外在": 54636,
+      "nOld": 54637,
+      "npost": 54638,
+      "enzyme": 54639,
+      "▁owning": 54640,
+      "主权": 54641,
+      "传入": 54642,
+      "轻重": 54643,
+      "▁规格参数": 54644,
+      "candidate": 54645,
+      "▁advisors": 54646,
+      "狂欢": 54647,
+      "edd": 54648,
+      "▁slack": 54649,
+      "会说": 54650,
+      "供需": 54651,
+      "吉尔": 54652,
+      "期为": 54653,
+      "构成了": 54654,
+      "magn": 54655,
+      "ometh": 54656,
+      "▁Blair": 54657,
+      "Subscriber": 54658,
+      "▁ω": 54659,
+      "国外的": 54660,
+      "flake": 54661,
+      "serializers": 54662,
+      "哈利": 54663,
+      "白马": 54664,
+      "}))": 54665,
+      "▁配置": 54666,
+      "专科学": 54667,
+      "类游戏": 54668,
+      "通过网络": 54669,
+      "不听": 54670,
+      "水文": 54671,
+      "副处长": 54672,
+      "网站建设": 54673,
+      "▁Tail": 54674,
+      "被执行人杨": 54675,
+      "nEvent": 54676,
+      "变身": 54677,
+      "高空": 54678,
+      "仔细地": 54679,
+      "告诉他": 54680,
+      "概率为": 54681,
+      "爱的人": 54682,
+      "由原告承担": 54683,
+      "▁Casey": 54684,
+      "二〇一七年四月": 54685,
+      "心跳": 54686,
+      "这名": 54687,
+      "TLS": 54688,
+      "insk": 54689,
+      "▁daemon": 54690,
+      "服务商": 54691,
+      "▁Dew": 54692,
+      "其行为已构成危险驾驶罪": 54693,
+      "Norm": 54694,
+      "▁*)\\": 54695,
+      "baidu": 54696,
+      "▁enumer": 54697,
+      "expressed": 54698,
+      "primes": 54699,
+      "proper": 54700,
+      "▁homomorphism": 54701,
+      "街上": 54702,
+      "在外面": 54703,
+      "▁PCA": 54704,
+      "plets": 54705,
+      "sizes": 54706,
+      "▁Runner": 54707,
+      "▁Tehran": 54708,
+      "angements": 54709,
+      "entricular": 54710,
+      "私下": 54711,
+      "▁HPLC": 54712,
+      "▁indicative": 54713,
+      "收录于": 54714,
+      "考试的": 54715,
+      "asin": 54716,
+      "因本案于": 54717,
+      "复合材料": 54718,
+      "▁leptin": 54719,
+      "▁Hyderabad": 54720,
+      "名著": 54721,
+      "辛勤": 54722,
+      "ovi": 54723,
+      "▁NGO": 54724,
+      "修真": 54725,
+      "Stamp": 54726,
+      "laravel": 54727,
+      "▁methane": 54728,
+      "nWhite": 54729,
+      "▁Sigma": 54730,
+      "▁Increased": 54731,
+      "▁Approximately": 54732,
+      "@\\": 54733,
+      "气压": 54734,
+      "藏在": 54735,
+      "专门的": 54736,
+      "▁SEE": 54737,
+      "经济林果": 54738,
+      "上前": 54739,
+      "升温": 54740,
+      "变焦": 54741,
+      "均已": 54742,
+      "umming": 54743,
+      "▁fetus": 54744,
+      "▁humid": 54745,
+      "▁lockdown": 54746,
+      "▁Biological": 54747,
+      "▁groundwater": 54748,
+      "▁intentional": 54749,
+      "元和": 54750,
+      "对人": 54751,
+      "课文": 54752,
+      "Fly": 54753,
+      "romb": 54754,
+      "感受到了": 54755,
+      "▁incubated": 54756,
+      "加点": 54757,
+      "\"}]": 54758,
+      "▁eV": 54759,
+      "这么大": 54760,
+      "▁Renew": 54761,
+      "acement": 54762,
+      "biology": 54763,
+      "▁bibliography": 54764,
+      "▁filmmaker": 54765,
+      "莱斯": 54766,
+      "司法局": 54767,
+      "一个数的": 54768,
+      "北京赛车": 54769,
+      "质量管理": 54770,
+      "副主任医师": 54771,
+      "▁halls": 54772,
+      "Compute": 54773,
+      "rescent": 54774,
+      "FileSystem": 54775,
+      "▁allergies": 54776,
+      "▁Increasing": 54777,
+      "下肢": 54778,
+      "但凡": 54779,
+      "后将": 54780,
+      "烘焙": 54781,
+      "Sugg": 54782,
+      "年上半年": 54783,
+      "sequences": 54784,
+      "过头": 54785,
+      "bias": 54786,
+      "在案佐证": 54787,
+      "法律援助": 54788,
+      "▁Duck": 54789,
+      "aclass": 54790,
+      "▁Kenny": 54791,
+      "项链": 54792,
+      "▁{}\\'": 54793,
+      "债权的实现": 54794,
+      "\\')),\\": 54795,
+      "养猪": 54796,
+      "嘉定": 54797,
+      "欧冠": 54798,
+      "fro": 54799,
+      "▁ка": 54800,
+      "compressed": 54801,
+      "▁号": 54802,
+      "他妈": 54803,
+      "器具": 54804,
+      "▁OD": 54805,
+      "动不动": 54806,
+      "海内外": 54807,
+      "第十三条": 54808,
+      "BUFFER": 54809,
+      "▁Adrian": 54810,
+      "▁storytelling": 54811,
+      "▁instructional": 54812,
+      "偏好": 54813,
+      "波澜": 54814,
+      "炒股": 54815,
+      ")}{(": 54816,
+      "Else": 54817,
+      "▁widgets": 54818,
+      "▁mythology": 54819,
+      "意境": 54820,
+      "noh": 54821,
+      "needs": 54822,
+      "▁Diseases": 54823,
+      "▁Rodriguez": 54824,
+      "▁transmembrane": 54825,
+      "▁г": 54826,
+      "———": 54827,
+      "一个问题": 54828,
+      "▁hooked": 54829,
+      "▁Podcast": 54830,
+      "▁manuscripts": 54831,
+      "十余": 54832,
+      "她也": 54833,
+      "阶梯": 54834,
+      "/\\\\/": 54835,
+      "▁DON": 54836,
+      "▁Marx": 54837,
+      "▁immers": 54838,
+      "▁reproduc": 54839,
+      "▁Raspberry": 54840,
+      "表态": 54841,
+      "可不是": 54842,
+      "\\'\"\\": 54843,
+      "htub": 54844,
+      "defer": 54845,
+      "▁Noise": 54846,
+      "WV": 54847,
+      "盘点": 54848,
+      "]\",": 54849,
+      "zon": 54850,
+      "还不够": 54851,
+      "nRest": 54852,
+      "▁EVEN": 54853,
+      "▁dignity": 54854,
+      "▁Likewise": 54855,
+      "▁highways": 54856,
+      "当晚": 54857,
+      "这话": 54858,
+      "间盘": 54859,
+      "apper": 54860,
+      "▁Brady": 54861,
+      "所欠": 54862,
+      "ROL": 54863,
+      "中国农业银行": 54864,
+      "▁markup": 54865,
+      "Exceptions": 54866,
+      "苦难": 54867,
+      "wald": 54868,
+      "estim": 54869,
+      "加拉": 54870,
+      "后就": 54871,
+      "辨识": 54872,
+      "遥远": 54873,
+      "第九十七条": 54874,
+      "exposed": 54875,
+      "paralle": 54876,
+      "prettier": 54877,
+      "▁ö": 54878,
+      "元旦": 54879,
+      "较短": 54880,
+      "CRC": 54881,
+      "rdf": 54882,
+      "normalized": 54883,
+      "Ka": 54884,
+      "异能": 54885,
+      "面具": 54886,
+      "的元素": 54887,
+      "艺术学院": 54888,
+      "USTOM": 54889,
+      "migrate": 54890,
+      "▁uninstall": 54891,
+      "▁Archbishop": 54892,
+      "▁Springfield": 54893,
+      "赞赏": 54894,
+      "异议人": 54895,
+      "的广告": 54896,
+      "▁ballet": 54897,
+      "▁Ethereum": 54898,
+      "filesystem": 54899,
+      "▁negotiation": 54900,
+      "spi": 54901,
+      "高中生": 54902,
+      "▁metam": 54903,
+      "▁GraphQL": 54904,
+      "▁milliseconds": 54905,
+      "Css": 54906,
+      "脱颖而出": 54907,
+      "avings": 54908,
+      "Between": 54909,
+      "死于": 54910,
+      "脱发": 54911,
+      "在接受": 54912,
+      "在银行": 54913,
+      "▁apex": 54914,
+      "FN": 54915,
+      "您选择": 54916,
+      "▁###\\": 54917,
+      "asthan": 54918,
+      "▁bloss": 54919,
+      "▁patri": 54920,
+      "▁rodents": 54921,
+      "生猪": 54922,
+      "▁VAT": 54923,
+      "juven": 54924,
+      "▁Fight": 54925,
+      "▁irreversible": 54926,
+      "▁za": 54927,
+      "动物门": 54928,
+      "nLead": 54929,
+      "▁autophagy": 54930,
+      "七星": 54931,
+      "体会到": 54932,
+      "▁elit": 54933,
+      "会比": 54934,
+      "溶于": 54935,
+      "不了的": 54936,
+      "mime": 54937,
+      "nSyn": 54938,
+      "做的事情": 54939,
+      "patched": 54940,
+      "▁EXISTS": 54941,
+      "▁cultivated": 54942,
+      "牢牢": 54943,
+      "种质": 54944,
+      "Ctx": 54945,
+      "无线电": 54946,
+      "艾滋病": 54947,
+      "adam": 54948,
+      "eeds": 54949,
+      "▁thy": 54950,
+      "▁rocky": 54951,
+      "▁victories": 54952,
+      "巴斯": 54953,
+      "油价": 54954,
+      "nip": 54955,
+      "求实数": 54956,
+      "纤维素": 54957,
+      "自动驾驶": 54958,
+      "diffusion": 54959,
+      "像个": 54960,
+      "INC": 54961,
+      "``\\": 54962,
+      "▁WC": 54963,
+      "营业部": 54964,
+      "inky": 54965,
+      "质量问题": 54966,
+      "monitoring": 54967,
+      "nDifferent": 54968,
+      "▁reperfusion": 54969,
+      "▁特": 54970,
+      "不仅要": 54971,
+      "的函数": 54972,
+      "的道路": 54973,
+      "▁burial": 54974,
+      "打死": 54975,
+      "漫长": 54976,
+      "▁WWE": 54977,
+      "详细介绍": 54978,
+      "▁Scout": 54979,
+      "▁sixteen": 54980,
+      "大厂": 54981,
+      "拉着": 54982,
+      "立新": 54983,
+      "老是": 54984,
+      "我们一起": 54985,
+      "提出减刑建议书": 54986,
+      "setValue": 54987,
+      "点数": 54988,
+      "谴责": 54989,
+      "韧性": 54990,
+      "家长们": 54991,
+      "▁Fear": 54992,
+      "Finish": 54993,
+      "movies": 54994,
+      "▁autor": 54995,
+      "transparent": 54996,
+      "一字": 54997,
+      "nUser": 54998,
+      "ycling": 54999,
+      "可笑": 55000,
+      "调侃": 55001,
+      "万元整": 55002,
+      "亚热带": 55003,
+      "二次根式的": 55004,
+      "iciary": 55005,
+      "▁itertools": 55006,
+      "州的": 55007,
+      "湖泊": 55008,
+      "▁Eq": 55009,
+      "元器件": 55010,
+      "▁Males": 55011,
+      "宁愿": 55012,
+      "读题": 55013,
+      "▁iz": 55014,
+      "一般人": 55015,
+      "洗衣间": 55016,
+      "esso": 55017,
+      "mpeg": 55018,
+      "▁suprem": 55019,
+      ")'": 55020,
+      "购进": 55021,
+      "企业发展": 55022,
+      "查控系统": 55023,
+      "getText": 55024,
+      "▁adduct": 55025,
+      "连连": 55026,
+      "都为": 55027,
+      "nBi": 55028,
+      "三十年": 55029,
+      ")\\\\,": 55030,
+      "▁Schmidt": 55031,
+      "illas": 55032,
+      "▁Comic": 55033,
+      "Generation": 55034,
+      "▁colonization": 55035,
+      "入场": 55036,
+      "向法庭": 55037,
+      "nEff": 55038,
+      "纪委书记": 55039,
+      "ressor": 55040,
+      "JM": 55041,
+      "andr": 55042,
+      "▁Gos": 55043,
+      "▁Gut": 55044,
+      "政府信息公开": 55045,
+      "Instant": 55046,
+      "▁Lancaster": 55047,
+      "卡的": 55048,
+      "开户行": 55049,
+      "结果的": 55050,
+      "igible": 55051,
+      "▁FIRST": 55052,
+      "▁Volks": 55053,
+      "uitable": 55054,
+      "地将": 55055,
+      "灾区": 55056,
+      "▁注解": 55057,
+      "nJon": 55058,
+      "神经网络": 55059,
+      "provide": 55060,
+      "▁reflux": 55061,
+      "中枢": 55062,
+      "保险箱": 55063,
+      "几年的": 55064,
+      "▁Grammy": 55065,
+      "▁interpersonal": 55066,
+      "无事": 55067,
+      "▁Lastly": 55068,
+      "▁bizarre": 55069,
+      "▁Discount": 55070,
+      "▁asteroid": 55071,
+      "塑性": 55072,
+      "▁Solo": 55073,
+      "indsay": 55074,
+      "oslovak": 55075,
+      "mediately": 55076,
+      "▁merchants": 55077,
+      "getAttribute": 55078,
+      "孕期": 55079,
+      "钻研": 55080,
+      "劳人仲": 55081,
+      "daughter": 55082,
+      "▁dentistry": 55083,
+      "Dead": 55084,
+      "▁Yii": 55085,
+      "▁asylum": 55086,
+      "▁rugged": 55087,
+      "nBaseball": 55088,
+      "▁prophylaxis": 55089,
+      "▁antioxidants": 55090,
+      "摄氏": 55091,
+      "srv": 55092,
+      "usb": 55093,
+      "nPal": 55094,
+      "▁Gul": 55095,
+      "分析方法": 55096,
+      "如何选择": 55097,
+      "eware": 55098,
+      "▁Pale": 55099,
+      "▁grac": 55100,
+      "▁infused": 55101,
+      "地表": 55102,
+      "先后在": 55103,
+      "和美国": 55104,
+      "▁rer": 55105,
+      "因与被申请人": 55106,
+      "▁awaiting": 55107,
+      "▁subtraction": 55108,
+      "在地上": 55109,
+      "语言的": 55110,
+      "▁Filipino": 55111,
+      "▁detached": 55112,
+      "▁advertised": 55113,
+      ",…": 55114,
+      "其对": 55115,
+      "给别人": 55116,
+      "inia": 55117,
+      "▁Brick": 55118,
+      "aurants": 55119,
+      "交叉口": 55120,
+      "nMarch": 55121,
+      "▁Removal": 55122,
+      "他要": 55123,
+      "环卫": 55124,
+      "采暖": 55125,
+      "FLAG": 55126,
+      "duty": 55127,
+      "▁$$|": 55128,
+      "据新华社": 55129,
+      "chemy": 55130,
+      "tvalue": 55131,
+      "branches": 55132,
+      "▁laptops": 55133,
+      ")”": 55134,
+      "以罪犯": 55135,
+      "相对来说": 55136,
+      "structures": 55137,
+      "每隔": 55138,
+      "穿的": 55139,
+      "▁kb": 55140,
+      "内存类型": 55141,
+      "土地管理法": 55142,
+      "▁Hands": 55143,
+      "▁scint": 55144,
+      "▁unitary": 55145,
+      "discovery": 55146,
+      "农林": 55147,
+      "宣言": 55148,
+      "人民卫生": 55149,
+      "▁protons": 55150,
+      "▁handball": 55151,
+      "▁patented": 55152,
+      "▁Lancashire": 55153,
+      "庄镇": 55154,
+      "行了": 55155,
+      "turl": 55156,
+      "罪犯奖励审批表": 55157,
+      "▁reimburse": 55158,
+      "根的": 55159,
+      "ekt": 55160,
+      "▁OT": 55161,
+      "选择性": 55162,
+      "研究发现": 55163,
+      "▁Rocket": 55164,
+      "坐落在": 55165,
+      "这次的": 55166,
+      "arine": 55167,
+      "游戏大小为": 55168,
+      "▁outgoing": 55169,
+      "武帝": 55170,
+      "cox": 55171,
+      "▁Warm": 55172,
+      "▁Higgs": 55173,
+      "▁Kerry": 55174,
+      "不值得": 55175,
+      "周围的": 55176,
+      "▁Laf": 55177,
+      "ivatives": 55178,
+      "▁Changing": 55179,
+      "广泛应用": 55180,
+      "olics": 55181,
+      "连载的小说": 55182,
+      "▁Charl": 55183,
+      "roduced": 55184,
+      "intensity": 55185,
+      "软件的": 55186,
+      "▁Morph": 55187,
+      "▁Wow": 55188,
+      "▁ISIS": 55189,
+      "▁instinct": 55190,
+      "▁assortment": 55191,
+      "ofluorescence": 55192,
+      "写真": 55193,
+      "nSmall": 55194,
+      "▁Fixes": 55195,
+      "onCreate": 55196,
+      "奋进": 55197,
+      "巴萨": 55198,
+      "bnb": 55199,
+      "▁bob": 55200,
+      "Et": 55201,
+      "受贿": 55202,
+      "和睦": 55203,
+      "肺部": 55204,
+      "▁Dock": 55205,
+      "▁weighs": 55206,
+      "nFunction": 55207,
+      "无知": 55208,
+      "▁ay": 55209,
+      "broker": 55210,
+      "ospice": 55211,
+      "▁rhythms": 55212,
+      "▁Organisation": 55213,
+      "▁Wnt": 55214,
+      "▁ace": 55215,
+      "▁Knox": 55216,
+      "▁bullying": 55217,
+      "transactions": 55218,
+      "Cou": 55219,
+      "nWithin": 55220,
+      "playlist": 55221,
+      "▁cyclist": 55222,
+      "▁trademarks": 55223,
+      "开来": 55224,
+      "▁wavelet": 55225,
+      "хо": 55226,
+      "洋洋": 55227,
+      "Flu": 55228,
+      "生效的": 55229,
+      "还提供": 55230,
+      "▁赛季": 55231,
+      "比赛的": 55232,
+      "可以分为": 55233,
+      "混合运算": 55234,
+      "▁divergent": 55235,
+      "归去": 55236,
+      "日光": 55237,
+      "▁口味": 55238,
+      "▁电话": 55239,
+      "inse": 55240,
+      "环境污染": 55241,
+      "至关重要": 55242,
+      "▁Gandhi": 55243,
+      "▁Ethernet": 55244,
+      "{<": 55245,
+      "再多": 55246,
+      "MMP": 55247,
+      "教育学院": 55248,
+      "▁myosin": 55249,
+      "▁stretches": 55250,
+      "天道": 55251,
+      "夯实": 55252,
+      "次日": 55253,
+      "画像": 55254,
+      "▁Zur": 55255,
+      "为主题的": 55256,
+      "clair": 55257,
+      "委员会副主任": 55258,
+      "中国工商银行股份有限公司": 55259,
+      "在某": 55260,
+      "用具": 55261,
+      "只要是": 55262,
+      "视网膜": 55263,
+      "actually": 55264,
+      "▁Redirect": 55265,
+      "子树": 55266,
+      "学说": 55267,
+      "菜肴": 55268,
+      "观的": 55269,
+      "这群": 55270,
+      "毕竟是": 55271,
+      "hardt": 55272,
+      "istem": 55273,
+      "由被告住所地": 55274,
+      "Gi": 55275,
+      "余杭": 55276,
+      "ARE": 55277,
+      "▁Yun": 55278,
+      "外观设计": 55279,
+      "生活习惯": 55280,
+      "明智的选择": 55281,
+      "甲基苯丙胺": 55282,
+      "▁thunder": 55283,
+      "▁Essential": 55284,
+      "▁completes": 55285,
+      "适中": 55286,
+      "一千元": 55287,
+      "rupting": 55288,
+      "研发的": 55289,
+      "要及时": 55290,
+      "其他需要": 55291,
+      "▁revert": 55292,
+      "Detector": 55293,
+      "▁priests": 55294,
+      "活塞": 55295,
+      "ajs": 55296,
+      "▁IG": 55297,
+      "▁德国": 55298,
+      "▁reck": 55299,
+      "REGIST": 55300,
+      "▁Consortium": 55301,
+      "▁approximations": 55302,
+      "▁名": 55303,
+      "成因": 55304,
+      "▁HK": 55305,
+      "立方分米": 55306,
+      "ilian": 55307,
+      "tiles": 55308,
+      "▁fungus": 55309,
+      "▁multiples": 55310,
+      "村落": 55311,
+      "▁anticancer": 55312,
+      "倒霉": 55313,
+      "看清": 55314,
+      "陵县": 55315,
+      "▁bog": 55316,
+      "CACHE": 55317,
+      "tchar": 55318,
+      "▁imperial": 55319,
+      "都可": 55320,
+      "ADC": 55321,
+      "同学的": 55322,
+      "CEPT": 55323,
+      "▁SPI": 55324,
+      "tested": 55325,
+      "▁Bonus": 55326,
+      "omerase": 55327,
+      "劳动人事争议仲裁委员会": 55328,
+      "Authenticated": 55329,
+      "募集": 55330,
+      "大军": 55331,
+      "ylp": 55332,
+      "予以采纳": 55333,
+      "▁penn": 55334,
+      "▁bombs": 55335,
+      "missible": 55336,
+      "界定": 55337,
+      "BMI": 55338,
+      "切换到": 55339,
+      "方程的": 55340,
+      "简直是": 55341,
+      "▁Bryant": 55342,
+      "ImageView": 55343,
+      "▁supermarket": 55344,
+      "年和": 55345,
+      "高地": 55346,
+      "这样可以": 55347,
+      "酒店交通": 55348,
+      "▁Spar": 55349,
+      "▁Monten": 55350,
+      "▁那": 55351,
+      "学分": 55352,
+      "构图": 55353,
+      "祭祀": 55354,
+      "是全国": 55355,
+      "十三条的规定": 55356,
+      "▁hereditary": 55357,
+      "以各主要景点的距离如下": 55358,
+      "踊跃": 55359,
+      "▁动物界": 55360,
+      "▁Cruise": 55361,
+      "▁complaining": 55362,
+      "楼房": 55363,
+      "上大学": 55364,
+      "景德镇": 55365,
+      "更重要": 55366,
+      "▁bonded": 55367,
+      "▁landmarks": 55368,
+      "▁arbitrarily": 55369,
+      "▁breastfeeding": 55370,
+      "能效": 55371,
+      "elog": 55372,
+      "▁Clearly": 55373,
+      "▁Lagrangian": 55374,
+      "应邀": 55375,
+      "新鲜的": 55376,
+      "问题进行分析": 55377,
+      "readable": 55378,
+      "▁Armenia": 55379,
+      "▁Converter": 55380,
+      "Pt": 55381,
+      "魔王": 55382,
+      "▁Sofia": 55383,
+      "▁apparel": 55384,
+      "玉环": 55385,
+      "理赔": 55386,
+      "芝士": 55387,
+      "血小板": 55388,
+      "▁Giant": 55389,
+      "▁itiner": 55390,
+      "Partition": 55391,
+      "碳水": 55392,
+      "首尔": 55393,
+      "Vars": 55394,
+      "asser": 55395,
+      "▁Sche": 55396,
+      "▁Tier": 55397,
+      "▁Rally": 55398,
+      "笼罩": 55399,
+      "车库": 55400,
+      "高级工程师": 55401,
+      "搭档": 55402,
+      "少年的": 55403,
+      "lash": 55404,
+      "▁GIS": 55405,
+      "系统地": 55406,
+      "▁Burton": 55407,
+      "南侧": 55408,
+      "操作的": 55409,
+      "应予终结": 55410,
+      "nHuman": 55411,
+      "uginosa": 55412,
+      "▁balances": 55413,
+      "▁palliative": 55414,
+      "sms": 55415,
+      "▁Bj": 55416,
+      "▁位置": 55417,
+      "情况和": 55418,
+      "根本上": 55419,
+      "ENGINE": 55420,
+      "▁Eugene": 55421,
+      "研磨": 55422,
+      "▁丛书": 55423,
+      "类专业": 55424,
+      "inverse": 55425,
+      "▁responders": 55426,
+      "出游": 55427,
+      "海边": 55428,
+      "酿酒": 55429,
+      "里克": 55430,
+      "▁WO": 55431,
+      "比较有": 55432,
+      "的疗效": 55433,
+      "otherm": 55434,
+      "▁insign": 55435,
+      "人也": 55436,
+      "便会": 55437,
+      "每行": 55438,
+      "诸暨": 55439,
+      "小店": 55440,
+      "依法应予准许": 55441,
+      "▁cardio": 55442,
+      "▁quietly": 55443,
+      "▁eukaryotic": 55444,
+      "▁mentioning": 55445,
+      "路店": 55446,
+      "等方法": 55447,
+      "esque": 55448,
+      "▁weighed": 55449,
+      "▁mitigation": 55450,
+      "▁collaborating": 55451,
+      "中铁": 55452,
+      "足迹": 55453,
+      "▁hu": 55454,
+      "▁до": 55455,
+      "为学生": 55456,
+      "南北朝": 55457,
+      "▁Spy": 55458,
+      "▁umb": 55459,
+      "都有一个": 55460,
+      "backed": 55461,
+      "▁Kindle": 55462,
+      "水深": 55463,
+      "个月的": 55464,
+      "给予了": 55465,
+      "详细信息": 55466,
+      "▁Schwar": 55467,
+      "▁luggage": 55468,
+      "Dimension": 55469,
+      "▁asynchronous": 55470,
+      "几家": 55471,
+      "联谊": 55472,
+      "韶关": 55473,
+      "实际的": 55474,
+      "触摸屏": 55475,
+      "提供参考": 55476,
+      "▁Lund": 55477,
+      "律诗": 55478,
+      "目标的": 55479,
+      "}^{{": 55480,
+      "▁Scan": 55481,
+      "纷纷表示": 55482,
+      "thanks": 55483,
+      "▁Aurora": 55484,
+      "conference": 55485,
+      "▁inspirational": 55486,
+      "入院": 55487,
+      "▁CAM": 55488,
+      "nUsage": 55489,
+      "Whether": 55490,
+      "▁Repeat": 55491,
+      "▁redesign": 55492,
+      "▁Quantitative": 55493,
+      "多半": 55494,
+      "quil": 55495,
+      "reli": 55496,
+      "一个月内": 55497,
+      "家用电器": 55498,
+      "▁servant": 55499,
+      "ContentType": 55500,
+      "出卖": 55501,
+      "孤立": 55502,
+      "education": 55503,
+      "两会": 55504,
+      "灌木": 55505,
+      "▁Pour": 55506,
+      "▁decoder": 55507,
+      "▁exagger": 55508,
+      "▁multiplier": 55509,
+      "代号": 55510,
+      "铃薯": 55511,
+      "办主任": 55512,
+      "直播间": 55513,
+      "▁при": 55514,
+      "Science": 55515,
+      "▁Milton": 55516,
+      "▁Peterson": 55517,
+      "▁subscriber": 55518,
+      "守门": 55519,
+      "柔和": 55520,
+      "补水": 55521,
+      "超大": 55522,
+      "阐释": 55523,
+      "有很大": 55524,
+      "caps": 55525,
+      "alking": 55526,
+      "cerely": 55527,
+      "仇恨": 55528,
+      "复位": 55529,
+      "fps": 55530,
+      "减轻处罚": 55531,
+      "below": 55532,
+      "GITHUB": 55533,
+      "投资管理有限公司": 55534,
+      "就不用": 55535,
+      "▁ERA": 55536,
+      "Fa": 55537,
+      "岭南": 55538,
+      "排球": 55539,
+      "放学": 55540,
+      "▁Glob": 55541,
+      "▁Annie": 55542,
+      "▁mills": 55543,
+      "icillin": 55544,
+      "▁Divide": 55545,
+      "▁chambers": 55546,
+      "罚息": 55547,
+      "▁VT": 55548,
+      "和治疗": 55549,
+      "▁wished": 55550,
+      "encrypted": 55551,
+      "▁Infinity": 55552,
+      "云端": 55553,
+      "它能": 55554,
+      "nez": 55555,
+      "消化道": 55556,
+      "tang": 55557,
+      "exico": 55558,
+      "▁Called": 55559,
+      "二〇一八年五月": 55560,
+      "▁Dirichlet": 55561,
+      "姓氏": 55562,
+      "皮的": 55563,
+      "园林绿化": 55564,
+      "inges": 55565,
+      "▁playback": 55566,
+      "▁Interaction": 55567,
+      "\\\\'": 55568,
+      "▁%%": 55569,
+      "机械化": 55570,
+      "解决的": 55571,
+      "取暖": 55572,
+      "后备": 55573,
+      "本体": 55574,
+      "局域网": 55575,
+      "ebug": 55576,
+      "▁Wrong": 55577,
+      "▁Breaking": 55578,
+      "▁presidency": 55579,
+      "YA": 55580,
+      "怀抱": 55581,
+      "本能": 55582,
+      "当然了": 55583,
+      "的人类": 55584,
+      "▁secular": 55585,
+      "▁terrestrial": 55586,
+      "▁interpreting": 55587,
+      "佣金": 55588,
+      "佳人": 55589,
+      "法案": 55590,
+      "立方厘米": 55591,
+      "▁turbulence": 55592,
+      "交界": 55593,
+      "资金的": 55594,
+      "▁HAL": 55595,
+      "▁Bing": 55596,
+      "▁spokeswoman": 55597,
+      "巴马": 55598,
+      "市东": 55599,
+      "anyl": 55600,
+      "ovir": 55601,
+      "文化产业": 55602,
+      "▁Janet": 55603,
+      "▁Banking": 55604,
+      "Predicate": 55605,
+      "不需": 55606,
+      "出借": 55607,
+      "后端": 55608,
+      "知乎上": 55609,
+      "dial": 55610,
+      "根据自己的": 55611,
+      "unsafe": 55612,
+      "▁Nobel": 55613,
+      "▁telev": 55614,
+      "nutrition": 55615,
+      "大唐": 55616,
+      "▁pores": 55617,
+      "在服刑改造期间": 55618,
+      "▁histologic": 55619,
+      "▁Theoretical": 55620,
+      "后方": 55621,
+      "情趣": 55622,
+      "血的": 55623,
+      "讯息": 55624,
+      "▁cared": 55625,
+      "▁UL": 55626,
+      "过多的": 55627,
+      "TINGS": 55628,
+      "INVALID": 55629,
+      "▁Comfort": 55630,
+      "▁Conversely": 55631,
+      "▁unpredictable": 55632,
+      "皇上": 55633,
+      "OTS": 55634,
+      "▁第六": 55635,
+      "信息安全": 55636,
+      "▁bolt": 55637,
+      "▁graz": 55638,
+      "▁Older": 55639,
+      "上缴": 55640,
+      "如其": 55641,
+      "支持下": 55642,
+      "▁trop": 55643,
+      "itsu": 55644,
+      "▁Rag": 55645,
+      "▁subscriptions": 55646,
+      "某某诉被告": 55647,
+      "nHy": 55648,
+      "nnode": 55649,
+      "▁clinician": 55650,
+      "▁scrolling": 55651,
+      "▁deficiencies": 55652,
+      "固件": 55653,
+      "eka": 55654,
+      "在过去": 55655,
+      "问题了": 55656,
+      "STIT": 55657,
+      "olum": 55658,
+      "npeople": 55659,
+      "▁Wagner": 55660,
+      "▁french": 55661,
+      "piration": 55662,
+      "▁Friedrich": 55663,
+      "▁commenced": 55664,
+      "远县": 55665,
+      "阵营": 55666,
+      "TEX": 55667,
+      "领导和": 55668,
+      "▁jwt": 55669,
+      "刑终字第": 55670,
+      "司法解释": 55671,
+      "▁northeastern": 55672,
+      "江东": 55673,
+      "而易": 55674,
+      "银杏": 55675,
+      "▁Evan": 55676,
+      "▁Romney": 55677,
+      "▁Families": 55678,
+      "气动": 55679,
+      "飞的": 55680,
+      "▁pued": 55681,
+      "▁archived": 55682,
+      "单人": 55683,
+      "空心": 55684,
+      "nrm": 55685,
+      "发表了": 55686,
+      "▁Lah": 55687,
+      "Russian": 55688,
+      "TEMPLATE": 55689,
+      "▁thriving": 55690,
+      "兑现": 55691,
+      "废话": 55692,
+      "网易云": 55693,
+      "envs": 55694,
+      "formal": 55695,
+      "neutral": 55696,
+      "▁chassis": 55697,
+      "就大": 55698,
+      "垃圾分类": 55699,
+      "urses": 55700,
+      "greSQL": 55701,
+      "▁theatrical": 55702,
+      "改动": 55703,
+      "花落": 55704,
+      "oqu": 55705,
+      "▁CDP": 55706,
+      "avery": 55707,
+      "▁eryt": 55708,
+      "▁extras": 55709,
+      "▁pendant": 55710,
+      "▁parabola": 55711,
+      "▁neuropathy": 55712,
+      "与研究": 55713,
+      "emos": 55714,
+      "本案申请执行人": 55715,
+      "跑道": 55716,
+      "FOUND": 55717,
+      "▁knit": 55718,
+      "▁Aware": 55719,
+      "将从": 55720,
+      "推向": 55721,
+      "省长": 55722,
+      "自如": 55723,
+      "\\',$": 55724,
+      "upta": 55725,
+      "burst": 55726,
+      "▁Lambda": 55727,
+      "今朝": 55728,
+      "ECD": 55729,
+      "惯用脚": 55730,
+      "</>\\": 55731,
+      "高职院校": 55732,
+      "Recent": 55733,
+      "▁Blind": 55734,
+      "短路": 55735,
+      "'>\\": 55736,
+      "[]>": 55737,
+      "管理协会": 55738,
+      "▁horr": 55739,
+      "▁colitis": 55740,
+      "享受到": 55741,
+      "总人口": 55742,
+      "法律责任": 55743,
+      "Death": 55744,
+      "estly": 55745,
+      "二〇一五年九月": 55746,
+      "▁inversely": 55747,
+      "地亚": 55748,
+      "外接": 55749,
+      "安装包": 55750,
+      "有的是": 55751,
+      "JAVA": 55752,
+      "nyes": 55753,
+      "rgba": 55754,
+      "eters": 55755,
+      "▁Ping": 55756,
+      "nThird": 55757,
+      "ncreate": 55758,
+      "otherap": 55759,
+      "功课": 55760,
+      "岁以上": 55761,
+      "来不及": 55762,
+      "interop": 55763,
+      "▁solubility": 55764,
+      "部和": 55765,
+      "ESCO": 55766,
+      "所提出的": 55767,
+      "的时间内": 55768,
+      "▁traced": 55769,
+      "▁procedural": 55770,
+      "红灯": 55771,
+      "本报讯": 55772,
+      "输入格式": 55773,
+      "事实与理由": 55774,
+      "危险化学品": 55775,
+      "Cook": 55776,
+      "比率": 55777,
+      "APS": 55778,
+      "二千元": 55779,
+      "▁PLC": 55780,
+      "▁SDL": 55781,
+      "unpack": 55782,
+      "▁pollen": 55783,
+      "▁pollutants": 55784,
+      "Trim": 55785,
+      "dinger": 55786,
+      "▁categorical": 55787,
+      "nCross": 55788,
+      "▁Dylan": 55789,
+      "▁breasts": 55790,
+      "▁scheduler": 55791,
+      "乏力": 55792,
+      "密钥": 55793,
+      "还有人": 55794,
+      "hashCode": 55795,
+      "▁irrespective": 55796,
+      "军官": 55797,
+      "专业从事": 55798,
+      "erground": 55799,
+      "▁elephant": 55800,
+      "responding": 55801,
+      "applications": 55802,
+      "监听": 55803,
+      "▁广西": 55804,
+      "▁MTV": 55805,
+      "一点都不": 55806,
+      "行业发展": 55807,
+      "▁Camero": 55808,
+      "的人口": 55809,
+      "▁ENG": 55810,
+      "▁Jaw": 55811,
+      "这个东西": 55812,
+      "源性": 55813,
+      "预览": 55814,
+      "Ast": 55815,
+      "文史": 55816,
+      "往期": 55817,
+      "我已": 55818,
+      "Emp": 55819,
+      "死亡率": 55820,
+      "的动力": 55821,
+      "john": 55822,
+      "▁Ink": 55823,
+      "▁Mega": 55824,
+      "iaries": 55825,
+      "▁elicit": 55826,
+      "▁authored": 55827,
+      "▁monuments": 55828,
+      "▁Aboriginal": 55829,
+      "诚意": 55830,
+      "▁ros": 55831,
+      "怎么可能": 55832,
+      "神经系统": 55833,
+      "▁goat": 55834,
+      "umbnails": 55835,
+      "▁barrels": 55836,
+      "迷宫": 55837,
+      "dag": 55838,
+      "水处理": 55839,
+      "▁installment": 55840,
+      "ryn": 55841,
+      "▁Sach": 55842,
+      "补正": 55843,
+      "▁LM": 55844,
+      "一小时": 55845,
+      "在他们": 55846,
+      "▁****": 55847,
+      "▁pharmacokinetic": 55848,
+      "也称": 55849,
+      "子叶": 55850,
+      "iad": 55851,
+      "▁lign": 55852,
+      "province": 55853,
+      "▁biliary": 55854,
+      "修仙": 55855,
+      "四百": 55856,
+      "定于": 55857,
+      "徒弟": 55858,
+      "Mate": 55859,
+      "合同签订": 55860,
+      "科学技术出版社": 55861,
+      "优美的": 55862,
+      "机器翻译": 55863,
+      "oprot": 55864,
+      "▁rushing": 55865,
+      "▁removable": 55866,
+      "▁prosecution": 55867,
+      "中被": 55868,
+      "李文": 55869,
+      "盛大": 55870,
+      "▁意大利": 55871,
+      "故诉至法院": 55872,
+      "▁Vander": 55873,
+      "European": 55874,
+      "▁ultrason": 55875,
+      "伸出": 55876,
+      "年是": 55877,
+      "这天": 55878,
+      "▁pk": 55879,
+      "最合适的": 55880,
+      "但未": 55881,
+      "声道": 55882,
+      "并获得": 55883,
+      "的儿子": 55884,
+      "Cong": 55885,
+      "保修信息": 55886,
+      "▁classics": 55887,
+      "取决于被执行人是否有": 55888,
+      "▁confronted": 55889,
+      "春光": 55890,
+      "\\\\(|": 55891,
+      "▁NaCl": 55892,
+      "▁neoplastic": 55893,
+      "大妈": 55894,
+      "母乳": 55895,
+      "肾功能": 55896,
+      "▁whatsoever": 55897,
+      "单机": 55898,
+      "{}\"": 55899,
+      "思想和": 55900,
+      "nCzech": 55901,
+      "▁Allah": 55902,
+      "世事": 55903,
+      "的最后": 55904,
+      "之名": 55905,
+      "电图": 55906,
+      "经由": 55907,
+      "冰淇淋": 55908,
+      "辖区内": 55909,
+      "Uint": 55910,
+      "ndan": 55911,
+      "ounty": 55912,
+      "INGTON": 55913,
+      "辛辣": 55914,
+      "高了": 55915,
+      "▁谱曲": 55916,
+      "阿里云": 55917,
+      "ITEM": 55918,
+      "就会出现": 55919,
+      "▁Reporting": 55920,
+      "自尊": 55921,
+      "▁Lov": 55922,
+      "依据问题": 55923,
+      "▁Sources": 55924,
+      "▁detention": 55925,
+      "IU": 55926,
+      "洋溢": 55927,
+      "爱吃": 55928,
+      "驱逐": 55929,
+      "公司和": 55930,
+      "品牌策划": 55931,
+      "申请行政复议": 55932,
+      "▁micron": 55933,
+      "▁shader": 55934,
+      "郭某": 55935,
+      "Bon": 55936,
+      "Ivy": 55937,
+      "gam": 55938,
+      "▁OUR": 55939,
+      "不会出现": 55940,
+      "遵纪守法": 55941,
+      "nEnron": 55942,
+      "▁Antib": 55943,
+      "streams": 55944,
+      ")×": 55945,
+      "挂钩": 55946,
+      "此案": 55947,
+      "要将": 55948,
+      "有这么": 55949,
+      "干部职工": 55950,
+      "▁Cove": 55951,
+      "oxygen": 55952,
+      "clicked": 55953,
+      "opening": 55954,
+      "▁Janeiro": 55955,
+      "病菌": 55956,
+      "不在乎": 55957,
+      "为原料": 55958,
+      "Sequ": 55959,
+      "▁IND": 55960,
+      "▁referrals": 55961,
+      "一词": 55962,
+      "螺栓": 55963,
+      "就像是": 55964,
+      "▁Laz": 55965,
+      "rador": 55966,
+      "}\");\\": 55967,
+      "▁CERN": 55968,
+      "▁Pill": 55969,
+      "▁bearings": 55970,
+      "课件": 55971,
+      "▁Fres": 55972,
+      "▁Reply": 55973,
+      "▁Bronze": 55974,
+      "拆卸": 55975,
+      "未央": 55976,
+      "运维": 55977,
+      "VAR": 55978,
+      "部主任": 55979,
+      "gist": 55980,
+      "nsize": 55981,
+      "Providers": 55982,
+      "陪你": 55983,
+      "体现出": 55984,
+      "plen": 55985,
+      "▁`\\\\": 55986,
+      "非常重要的": 55987,
+      "日提出减刑建议": 55988,
+      "answered": 55989,
+      "函授": 55990,
+      "囊肿": 55991,
+      "导管": 55992,
+      "巡察": 55993,
+      "火炬": 55994,
+      "Bio": 55995,
+      "的合作": 55996,
+      "▁PAT": 55997,
+      "血量": 55998,
+      "在香港": 55999,
+      "nDevelop": 56000,
+      "▁Cowboys": 56001,
+      "▁Timothy": 56002,
+      "▁airplane": 56003,
+      "▁Questionnaire": 56004,
+      "不理": 56005,
+      "何人": 56006,
+      "家村": 56007,
+      "球衣": 56008,
+      "奥地利": 56009,
+      "oriasis": 56010,
+      "▁Sharon": 56011,
+      "▁Griffin": 56012,
+      "汗水": 56013,
+      "骁龙": 56014,
+      "▁cue": 56015,
+      "Deg": 56016,
+      "gust": 56017,
+      "▁slash": 56018,
+      "▁Bulldogs": 56019,
+      "adrenergic": 56020,
+      "会场": 56021,
+      "很棒": 56022,
+      "IFN": 56023,
+      "安庆市": 56024,
+      "Mime": 56025,
+      "Pods": 56026,
+      "代表性的": 56027,
+      "nSure": 56028,
+      "▁Adventures": 56029,
+      "Ho": 56030,
+      "上衣": 56031,
+      "口红": 56032,
+      "线和": 56033,
+      "支持和": 56034,
+      "而这些": 56035,
+      "agascar": 56036,
+      "▁Perl": 56037,
+      "▁disproportion": 56038,
+      "烟火": 56039,
+      "Jac": 56040,
+      "共同体": 56041,
+      "▁pots": 56042,
+      "公诉机关指控的罪名成立": 56043,
+      "建立在": 56044,
+      "联系人": 56045,
+      "riven": 56046,
+      "HZ": 56047,
+      "ettle": 56048,
+      "伪装": 56049,
+      "肝癌": 56050,
+      "重点是": 56051,
+      "akin": 56052,
+      "原审裁定": 56053,
+      "▁Bren": 56054,
+      "▁spawn": 56055,
+      "▁sparked": 56056,
+      "▁kindness": 56057,
+      "一想": 56058,
+      "舍得": 56059,
+      "邻近": 56060,
+      "给你们": 56061,
+      "▁Isn": 56062,
+      "▁Relative": 56063,
+      "长假": 56064,
+      "▁DU": 56065,
+      "的高级": 56066,
+      "蚌埠市": 56067,
+      "anon": 56068,
+      "▁gra": 56069,
+      "占了": 56070,
+      "可比": 56071,
+      "逛街": 56072,
+      "高贵": 56073,
+      "nRO": 56074,
+      "动物园": 56075,
+      "当选为": 56076,
+      "试题分析": 56077,
+      "越来越好": 56078,
+      "▁Bruno": 56079,
+      "▁mapper": 56080,
+      "▁rotations": 56081,
+      "六级": 56082,
+      "高点": 56083,
+      "▁clerk": 56084,
+      "市人民法院于": 56085,
+      "▁spells": 56086,
+      "bitration": 56087,
+      "群的": 56088,
+      "预示": 56089,
+      "太极拳": 56090,
+      "HDMI": 56091,
+      "罪犯考核": 56092,
+      "▁Hilton": 56093,
+      "▁accelerating": 56094,
+      "WO": 56095,
+      "做梦": 56096,
+      "很正常": 56097,
+      "arty": 56098,
+      "nLib": 56099,
+      "▁sadly": 56100,
+      "▁contempl": 56101,
+      "▁Chevrolet": 56102,
+      "轴的": 56103,
+      "分别于": 56104,
+      "设置为": 56105,
+      "TOOL": 56106,
+      "▁Jiang": 56107,
+      "▁dread": 56108,
+      "创业板": 56109,
+      "▁tempo": 56110,
+      "就该": 56111,
+      "这笔": 56112,
+      "bard": 56113,
+      "onate": 56114,
+      "neider": 56115,
+      "Numeric": 56116,
+      "仍旧": 56117,
+      "在场": 56118,
+      "▁sd": 56119,
+      "举个例子": 56120,
+      "住江苏省": 56121,
+      "Rever": 56122,
+      "▁Dataset": 56123,
+      "lifecycle": 56124,
+      "▁shocking": 56125,
+      "▁believing": 56126,
+      "校正": 56127,
+      "箱子": 56128,
+      "通透": 56129,
+      "Django": 56130,
+      "either": 56131,
+      "ان": 56132,
+      "眼界": 56133,
+      "碾压": 56134,
+      "把一个": 56135,
+      "▁femin": 56136,
+      "▁groove": 56137,
+      "matching": 56138,
+      "clamation": 56139,
+      "Compatible": 56140,
+      "何种": 56141,
+      "提议": 56142,
+      "\\'_": 56143,
+      "()}\\": 56144,
+      "ndiv": 56145,
+      "Ce": 56146,
+      "不会再": 56147,
+      "会引起": 56148,
+      "▁Gaza": 56149,
+      "对申请执行人": 56150,
+      "育儿": 56151,
+      "健康教育": 56152,
+      "在天": 56153,
+      "驻地": 56154,
+      "高某": 56155,
+      "zan": 56156,
+      "因为有": 56157,
+      "nAdv": 56158,
+      "▁melted": 56159,
+      "曾获": 56160,
+      "abo": 56161,
+      "▁Edu": 56162,
+      "中央空调": 56163,
+      "▁Till": 56164,
+      "okines": 56165,
+      "▁axons": 56166,
+      "fitting": 56167,
+      "▁Handbook": 56168,
+      "▁disconnect": 56169,
+      "Documentation": 56170,
+      "上为": 56171,
+      "包包": 56172,
+      "无不": 56173,
+      "竞选": 56174,
+      "▁&\\": 56175,
+      "与企业": 56176,
+      "ikit": 56177,
+      "reward": 56178,
+      "▁buddy": 56179,
+      "▁maple": 56180,
+      "转弯": 56181,
+      "直升机": 56182,
+      "政协副主席": 56183,
+      "ordion": 56184,
+      "Temperature": 56185,
+      "Io": 56186,
+      "北约": 56187,
+      "杀伤": 56188,
+      "正宗": 56189,
+      "nJud": 56190,
+      "Roman": 56191,
+      "undai": 56192,
+      "▁lion": 56193,
+      "▁Manor": 56194,
+      "▁traceback": 56195,
+      "▁hyperplasia": 56196,
+      "▁Implications": 56197,
+      "要对": 56198,
+      "钝角": 56199,
+      "第六十四条": 56200,
+      "SYSTEM": 56201,
+      "existent": 56202,
+      "Association": 56203,
+      "▁BeautifulSoup": 56204,
+      "我问": 56205,
+      "曲目": 56206,
+      "织物": 56207,
+      "▁Boh": 56208,
+      ">\\',\\": 56209,
+      "▁scalp": 56210,
+      "声乐": 56211,
+      "绳子": 56212,
+      "曾用名": 56213,
+      "thews": 56214,
+      "uloskeletal": 56215,
+      "耶稣": 56216,
+      "不会被": 56217,
+      "执行异议": 56218,
+      "rstrip": 56219,
+      "▁Caval": 56220,
+      "▁armor": 56221,
+      "▁knots": 56222,
+      "▁stuffed": 56223,
+      "出境": 56224,
+      "asn": 56225,
+      "Wiki": 56226,
+      "▁diaph": 56227,
+      "▁elective": 56228,
+      "▁shipment": 56229,
+      "ValueError": 56230,
+      "▁collaborators": 56231,
+      "免交": 56232,
+      "圈的": 56233,
+      "▁Egg": 56234,
+      "ribly": 56235,
+      "准许上诉人": 56236,
+      "byterian": 56237,
+      "他们对": 56238,
+      "边长为": 56239,
+      "uten": 56240,
+      "▁wag": 56241,
+      "第七十九条和": 56242,
+      "houette": 56243,
+      "▁Smooth": 56244,
+      "▁trader": 56245,
+      "▁statewide": 56246,
+      "中段": 56247,
+      "打到": 56248,
+      "财经大学": 56249,
+      "ighbour": 56250,
+      "thermal": 56251,
+      "nRegards": 56252,
+      "▁Gujarat": 56253,
+      "odds": 56254,
+      "▁roses": 56255,
+      "▁respects": 56256,
+      "TiO": 56257,
+      "▁Beg": 56258,
+      "CMake": 56259,
+      "ossible": 56260,
+      "recursive": 56261,
+      "医治": 56262,
+      "触及": 56263,
+      "thor": 56264,
+      "SECON": 56265,
+      "▁pleas": 56266,
+      "▁bowling": 56267,
+      "▁cocktails": 56268,
+      "▁Achievement": 56269,
+      "▁commissioner": 56270,
+      "发现被执行人有财产可供执行": 56271,
+      "Die": 56272,
+      "等活动": 56273,
+      "中国企业": 56274,
+      "税务总局": 56275,
+      "▁scav": 56276,
+      "▁Ahmad": 56277,
+      "▁fibroblast": 56278,
+      "世界里": 56279,
+      "utan": 56280,
+      "▁Modal": 56281,
+      "二〇一六年十月": 56282,
+      "findViewById": 56283,
+      "不自": 56284,
+      "在被": 56285,
+      "红十字": 56286,
+      "atham": 56287,
+      "iazep": 56288,
+      "▁lone": 56289,
+      "Resolve": 56290,
+      "▁Phillip": 56291,
+      "▁hacking": 56292,
+      "▁electrolyte": 56293,
+      "ancellationToken": 56294,
+      "一扫": 56295,
+      "纯属": 56296,
+      "negie": 56297,
+      "▁Nonlinear": 56298,
+      "▁smo": 56299,
+      "rawling": 56300,
+      "included": 56301,
+      "▁entrepreneurial": 56302,
+      "大都": 56303,
+      "越小": 56304,
+      "atonin": 56305,
+      "istine": 56306,
+      "习近平总书记": 56307,
+      "Getting": 56308,
+      "éc": 56309,
+      "ód": 56310,
+      "EOF": 56311,
+      "Lex": 56312,
+      "Cube": 56313,
+      "wsgi": 56314,
+      "▁Rup": 56315,
+      "向本院提供被执行人": 56316,
+      "▁Techniques": 56317,
+      "梦境": 56318,
+      "考评": 56319,
+      "牛仔裤": 56320,
+      "▁Rough": 56321,
+      "▁surfing": 56322,
+      "▁twentieth": 56323,
+      "同名": 56324,
+      "▁MF": 56325,
+      "▁scrutiny": 56326,
+      "乌兰": 56327,
+      "人多": 56328,
+      "宾客": 56329,
+      "弊端": 56330,
+      "usty": 56331,
+      "第一部分": 56332,
+      "nMountains": 56333,
+      "▁viewpoint": 56334,
+      "李小": 56335,
+      "皇马": 56336,
+      "当前的": 56337,
+      "nPay": 56338,
+      "Cycle": 56339,
+      "▁hairy": 56340,
+      "oblasts": 56341,
+      "▁harbor": 56342,
+      "ej": 56343,
+      "嘉奖": 56344,
+      "如期": 56345,
+      "蝙蝠": 56346,
+      "▁俱乐部": 56347,
+      "所产生的": 56348,
+      "dagger": 56349,
+      "▁Canterbury": 56350,
+      "_*": 56351,
+      "过量": 56352,
+      "▁slate": 56353,
+      "二〇一七年六": 56354,
+      "本领": 56355,
+      "ipv": 56356,
+      "条件和": 56357,
+      "所带来的": 56358,
+      "▁peric": 56359,
+      "▁Median": 56360,
+      "Importer": 56361,
+      "ndefault": 56362,
+      "menopausal": 56363,
+      "偏见": 56364,
+      "赋值": 56365,
+      "nGl": 56366,
+      "审核表": 56367,
+      "▁PST": 56368,
+      "知识渊博": 56369,
+      "▁Eigen": 56370,
+      "TIMEOUT": 56371,
+      "▁pulsed": 56372,
+      "nBasketball": 56373,
+      "▁accidental": 56374,
+      "▁differentiating": 56375,
+      "DUCTION": 56376,
+      "经原告多次催要": 56377,
+      "HU": 56378,
+      "乙肝": 56379,
+      "分别在": 56380,
+      "▁awa": 56381,
+      "Think": 56382,
+      "Until": 56383,
+      "nItem": 56384,
+      "而有所不同": 56385,
+      "附带民事诉讼": 56386,
+      "temporal": 56387,
+      "Interaction": 56388,
+      "▁风": 56389,
+      "车轮": 56390,
+      "▁都市言情": 56391,
+      "presso": 56392,
+      "判决生效后十日内": 56393,
+      "介意": 56394,
+      "提供的证据": 56395,
+      "第一百四十四条的规定": 56396,
+      "▁Contributors": 56397,
+      "一枝": 56398,
+      "较小的": 56399,
+      "oppy": 56400,
+      "▁Gian": 56401,
+      "▁teamed": 56402,
+      "▁fourteen": 56403,
+      "与国际": 56404,
+      "\":[\\": 56405,
+      "债权依法": 56406,
+      "在人民法院": 56407,
+      "批准的项目": 56408,
+      "具有挑战性的": 56409,
+      "该犯在服刑期间": 56410,
+      "readthedocs": 56411,
+      "副总": 56412,
+      "lisher": 56413,
+      "▁Wings": 56414,
+      "▁wears": 56415,
+      "eliness": 56416,
+      "▁torrent": 56417,
+      "压根": 56418,
+      "床房": 56419,
+      "的技能": 56420,
+      "英雄联盟": 56421,
+      "▁peroxide": 56422,
+      "▁蓝": 56423,
+      "花纹": 56424,
+      "$),": 56425,
+      "原来是": 56426,
+      "很快就": 56427,
+      "▁SCI": 56428,
+      "arger": 56429,
+      "nHeat": 56430,
+      "符合下列条件": 56431,
+      "matcher": 56432,
+      "▁Finals": 56433,
+      "▁Lutheran": 56434,
+      "getProperty": 56435,
+      "下场": 56436,
+      "宁区": 56437,
+      "Cas": 56438,
+      "也许是": 56439,
+      "的种类": 56440,
+      "nYet": 56441,
+      "liche": 56442,
+      "▁malt": 56443,
+      "具备执行条件": 56444,
+      "makedirs": 56445,
+      "▁Preston": 56446,
+      "▁Swimming": 56447,
+      "▁Utilities": 56448,
+      "▁ActiveRecord": 56449,
+      "▁hepatocellular": 56450,
+      "军民": 56451,
+      "押金": 56452,
+      "creative": 56453,
+      "▁Rainbow": 56454,
+      "tgz": 56455,
+      "惠州市": 56456,
+      "iola": 56457,
+      "请告诉我": 56458,
+      "骑行": 56459,
+      "aes": 56460,
+      "也希望": 56461,
+      "非常适合": 56462,
+      "但债权的实现": 56463,
+      "minimal": 56464,
+      "▁Execution": 56465,
+      "▁aeruginosa": 56466,
+      "周日": 56467,
+      "投降": 56468,
+      "盐酸": 56469,
+      "支行的": 56470,
+      "的一句话": 56471,
+      "abbit": 56472,
+      "理论与实践": 56473,
+      "确定的内容": 56474,
+      "▁Fetch": 56475,
+      "▁microgl": 56476,
+      "▁testimon": 56477,
+      "穿透": 56478,
+      "▁TG": 56479,
+      "nBro": 56480,
+      "Spark": 56481,
+      "Thomas": 56482,
+      "▁Tasmania": 56483,
+      "▁aberrant": 56484,
+      "Pic": 56485,
+      "▁化学": 56486,
+      "规章制度": 56487,
+      "recent": 56488,
+      "▁attracts": 56489,
+      "▁multiplicative": 56490,
+      "勘察": 56491,
+      "FAI": 56492,
+      "三年级": 56493,
+      "借款利息": 56494,
+      "▁myriad": 56495,
+      "clerotic": 56496,
+      "KI": 56497,
+      "played": 56498,
+      "平行四边形的": 56499,
+      "ousands": 56500,
+      "▁Sister": 56501,
+      "▁empire": 56502,
+      "▁towels": 56503,
+      "APE": 56504,
+      "yel": 56505,
+      "连云港": 56506,
+      "▁curb": 56507,
+      ",\\\\,\\\\": 56508,
+      "ntheir": 56509,
+      "▁prism": 56510,
+      "anganese": 56511,
+      "▁aerosol": 56512,
+      "▁Plymouth": 56513,
+      "▁Investigator": 56514,
+      "的理想": 56515,
+      "▁tant": 56516,
+      "▁Grammar": 56517,
+      "▁adolescence": 56518,
+      "只剩": 56519,
+      "望远": 56520,
+      "波罗": 56521,
+      "对身体": 56522,
+      "▁tachy": 56523,
+      "▁neonates": 56524,
+      "▁surfactant": 56525,
+      "每晚": 56526,
+      "inz": 56527,
+      "▁có": 56528,
+      "▁SME": 56529,
+      "㈰㈰㈰㈰": 56530,
+      "怎么回事": 56531,
+      "▁Amber": 56532,
+      "▁algae": 56533,
+      "▁verdict": 56534,
+      "内侧": 56535,
+      "奇偶": 56536,
+      "收条": 56537,
+      "灯泡": 56538,
+      "ERA": 56539,
+      "局限性": 56540,
+      "男主角": 56541,
+      "错过了": 56542,
+      "UEST": 56543,
+      "ieri": 56544,
+      "Motion": 56545,
+      "▁infertility": 56546,
+      "KW": 56547,
+      "npl": 56548,
+      "也只有": 56549,
+      "南极": 56550,
+      "颇有": 56551,
+      "太大的": 56552,
+      "ichever": 56553,
+      "requent": 56554,
+      "▁leukocyte": 56555,
+      "的女性": 56556,
+      "这样的话": 56557,
+      "oulli": 56558,
+      "金石": 56559,
+      "Rich": 56560,
+      "▁PCB": 56561,
+      "▁图书信息": 56562,
+      "ocaust": 56563,
+      "▁pertinent": 56564,
+      "万户": 56565,
+      "暑期": 56566,
+      "nMor": 56567,
+      "brids": 56568,
+      "▁scrat": 56569,
+      "nConcept": 56570,
+      "人行": 56571,
+      "从我": 56572,
+      "大雨": 56573,
+      "▁BMP": 56574,
+      "▁Anglo": 56575,
+      "mutable": 56576,
+      "▁judgments": 56577,
+      "丛林": 56578,
+      "瑕疵": 56579,
+      "襄阳市": 56580,
+      "soup": 56581,
+      "▁Poker": 56582,
+      "有明确的被告": 56583,
+      "▁fencing": 56584,
+      "▁initiating": 56585,
+      "▁Instruction": 56586,
+      "上行": 56587,
+      "有其": 56588,
+      "▁bt": 56589,
+      "cant": 56590,
+      "住江西省": 56591,
+      "▁Boris": 56592,
+      "commits": 56593,
+      "▁stripes": 56594,
+      "子弟": 56595,
+      "揭露": 56596,
+      "珍贵的": 56597,
+      "▁Known": 56598,
+      "percentage": 56599,
+      "rk": 56600,
+      "代会": 56601,
+      "\"#\"": 56602,
+      "能力与": 56603,
+      "WHERE": 56604,
+      "▁Piece": 56605,
+      "▁Weekend": 56606,
+      "messaging": 56607,
+      "platforms": 56608,
+      "▁mediation": 56609,
+      "反派": 56610,
+      "ecomment": 56611,
+      "▁autonomic": 56612,
+      "▁heuristic": 56613,
+      "▁得": 56614,
+      "之乡": 56615,
+      "奉贤": 56616,
+      "振华": 56617,
+      "提现": 56618,
+      "橱柜": 56619,
+      "▁Franz": 56620,
+      "▁Suffolk": 56621,
+      "▁synapses": 56622,
+      "乐于": 56623,
+      "公分": 56624,
+      "含量的": 56625,
+      "完毕后": 56626,
+      "旅游局": 56627,
+      "病虫害": 56628,
+      "speech": 56629,
+      "延缓": 56630,
+      "货源": 56631,
+      "▁OB": 56632,
+      "个国家": 56633,
+      "特别授权代理": 56634,
+      "nArgentine": 56635,
+      "▁manipulated": 56636,
+      "拉萨": 56637,
+      "的正确": 56638,
+      "▁Fruit": 56639,
+      "bellion": 56640,
+      "▁Tracking": 56641,
+      "▁Cartesian": 56642,
+      "兽医": 56643,
+      "飞翔": 56644,
+      "）”": 56645,
+      "准备了": 56646,
+      "就比较": 56647,
+      "方向键": 56648,
+      "▁agar": 56649,
+      "▁tryp": 56650,
+      "▁grind": 56651,
+      "两千": 56652,
+      "无故": 56653,
+      "作家协会": 56654,
+      "民事行为": 56655,
+      "连云港市": 56656,
+      "▁persisted": 56657,
+      "天王": 56658,
+      "转眼": 56659,
+      "▁*_": 56660,
+      "nSave": 56661,
+      "▁Mohammed": 56662,
+      "Completion": 56663,
+      "PSC": 56664,
+      "wart": 56665,
+      "▁seated": 56666,
+      "linspace": 56667,
+      "▁LANGUAGE": 56668,
+      "体操": 56669,
+      "在读": 56670,
+      "待机": 56671,
+      "攻关": 56672,
+      "样板": 56673,
+      "一流的": 56674,
+      "就会被": 56675,
+      "饮用水": 56676,
+      "▁coff": 56677,
+      "具有重要的": 56678,
+      "技术转让": 56679,
+      "▁Burke": 56680,
+      "▁Marcel": 56681,
+      "对所": 56682,
+      "掩饰": 56683,
+      "缺损": 56684,
+      "hdl": 56685,
+      "放射性": 56686,
+      "有任何": 56687,
+      "Atom": 56688,
+      "行政法规": 56689,
+      "▁Brett": 56690,
+      "▁stacks": 56691,
+      "▁amended": 56692,
+      "无疑是": 56693,
+      "长大了": 56694,
+      "ATEG": 56695,
+      "rels": 56696,
+      "▁ADC": 56697,
+      "的绝对值": 56698,
+      "▁debts": 56699,
+      "使者": 56700,
+      "年产": 56701,
+      "ussed": 56702,
+      "▁vous": 56703,
+      "ematics": 56704,
+      "共创": 56705,
+      "河东": 56706,
+      "ани": 56707,
+      "▁marketed": 56708,
+      "王一": 56709,
+      "精盐": 56710,
+      "紧迫": 56711,
+      "是一条": 56712,
+      "▁mitral": 56713,
+      ";/": 56714,
+      "抗原": 56715,
+      "逃跑": 56716,
+      "和相关": 56717,
+      "客房时": 56718,
+      "清晰的": 56719,
+      "个村民小组": 56720,
+      "小数点移动": 56721,
+      "▁cliff": 56722,
+      "网课": 56723,
+      "▁原创": 56724,
+      "不明显": 56725,
+      "该问题": 56726,
+      "销售的": 56727,
+      "▁Nurse": 56728,
+      "nIntegr": 56729,
+      "tstring": 56730,
+      "师的": 56731,
+      "▁Bast": 56732,
+      "izards": 56733,
+      "▁scoop": 56734,
+      "▁avoids": 56735,
+      "Classifier": 56736,
+      "升学": 56737,
+      "noc": 56738,
+      "▁TOP": 56739,
+      "非法占用的": 56740,
+      "electro": 56741,
+      "同性": 56742,
+      "萧萧": 56743,
+      "角相等": 56744,
+      "ATING": 56745,
+      "哭泣": 56746,
+      "对症": 56747,
+      "霸气": 56748,
+      "自由的": 56749,
+      "nMont": 56750,
+      "endars": 56751,
+      "▁Assistance": 56752,
+      "牛排": 56753,
+      "▁AJ": 56754,
+      "万亿元": 56755,
+      "容易出现": 56756,
+      "tyard": 56757,
+      "▁SOME": 56758,
+      "▁Dinner": 56759,
+      "▁populate": 56760,
+      "▁deploying": 56761,
+      "▁К": 56762,
+      "马丁": 56763,
+      "就好像": 56764,
+      "深厚的": 56765,
+      "▁pou": 56766,
+      "▁Jewel": 56767,
+      "▁splits": 56768,
+      "也想": 56769,
+      "装甲": 56770,
+      "IRA": 56771,
+      "今天是": 56772,
+      "wana": 56773,
+      "▁Nil": 56774,
+      "tumor": 56775,
+      "utsch": 56776,
+      "▁Bless": 56777,
+      "▁Spons": 56778,
+      "▁invertible": 56779,
+      "SuppressWarnings": 56780,
+      "保健品": 56781,
+      "▁triv": 56782,
+      "▁colum": 56783,
+      "▁Kimber": 56784,
+      "全称": 56785,
+      "分针": 56786,
+      "胳膊": 56787,
+      "自驾": 56788,
+      "透析": 56789,
+      "ADDR": 56790,
+      "交通管理": 56791,
+      "开奖结果": 56792,
+      "nSuch": 56793,
+      "roads": 56794,
+      "酒店交通比较方便": 56795,
+      "Resolution": 56796,
+      "ETS": 56797,
+      "▁DF": 56798,
+      "保定市": 56799,
+      "▁GCC": 56800,
+      "学生学习": 56801,
+      "abort": 56802,
+      "ornado": 56803,
+      "升华": 56804,
+      "Glu": 56805,
+      "同时向": 56806,
+      "Hist": 56807,
+      "±\\": 56808,
+      "坎坷": 56809,
+      "车队": 56810,
+      "BRE": 56811,
+      "投资或": 56812,
+      "▁gems": 56813,
+      "酒店为您提供": 56814,
+      "uggling": 56815,
+      "协助执行通知书": 56816,
+      "整式": 56817,
+      "客房条款": 56818,
+      "前所未有的": 56819,
+      "Transl": 56820,
+      "选择上述客房时": 56821,
+      "▁sweeping": 56822,
+      "Particip": 56823,
+      "傻子": 56824,
+      "hay": 56825,
+      "pdb": 56826,
+      "的陈述": 56827,
+      "pins": 56828,
+      "iggins": 56829,
+      "setattr": 56830,
+      "削弱": 56831,
+      "川普": 56832,
+      "DIY": 56833,
+      "的温度": 56834,
+      "▁Opening": 56835,
+      "▁rehears": 56836,
+      "hya": 56837,
+      "准许或者": 56838,
+      "▁Quinn": 56839,
+      "▁nephew": 56840,
+      "▁catalysts": 56841,
+      "▁cooperate": 56842,
+      "▁logarithmic": 56843,
+      "的条件下": 56844,
+      "▁\\'/\\'": 56845,
+      "▁webinar": 56846,
+      "阅览": 56847,
+      "▁guilt": 56848,
+      "请参阅客房条款": 56849,
+      "nChristian": 56850,
+      "▁publicity": 56851,
+      "找了": 56852,
+      "滥用": 56853,
+      "我也不知道": 56854,
+      "提起的诉讼": 56855,
+      "你来": 56856,
+      "佳节": 56857,
+      "灯火": 56858,
+      "老妈": 56859,
+      "的特殊": 56860,
+      "这么做": 56861,
+      "马铃薯": 56862,
+      "▁Php": 56863,
+      "两种情况": 56864,
+      "质保时间": 56865,
+      "anking": 56866,
+      "arctan": 56867,
+      "上诉于河南省": 56868,
+      "元宵": 56869,
+      "唐诗": 56870,
+      "尽头": 56871,
+      "承德": 56872,
+      "▁rt": 56873,
+      "好用的": 56874,
+      "Fort": 56875,
+      "保证责任": 56876,
+      "着自己的": 56877,
+      "stash": 56878,
+      "▁Horizon": 56879,
+      "作息": 56880,
+      "过剩": 56881,
+      "omew": 56882,
+      "▁risen": 56883,
+      "参保": 56884,
+      "奶酪": 56885,
+      "cession": 56886,
+      "Markdown": 56887,
+      "▁HB": 56888,
+      "为我国": 56889,
+      "的实力": 56890,
+      "设备及": 56891,
+      "BOSS": 56892,
+      "土地使用权": 56893,
+      "Inputs": 56894,
+      "Logged": 56895,
+      "▁Griffith": 56896,
+      "附有": 56897,
+      "旁边的": 56898,
+      "节目的": 56899,
+      "毫无疑问": 56900,
+      "货币政策": 56901,
+      "ncase": 56902,
+      "▁Spread": 56903,
+      "水箱": 56904,
+      "eas": 56905,
+      "▁FCC": 56906,
+      "查询回执": 56907,
+      "▁Lucky": 56908,
+      "▁unbel": 56909,
+      "▁phantom": 56910,
+      "▁earrings": 56911,
+      "免去": 56912,
+      "广元": 56913,
+      "水温": 56914,
+      "珍藏": 56915,
+      "战略合作": 56916,
+      "无论如何": 56917,
+      "表扬奖励": 56918,
+      "avoid": 56919,
+      "▁lonely": 56920,
+      "▁Texture": 56921,
+      "有管辖权的人民法院": 56922,
+      "生态系统": 56923,
+      "▁inspections": 56924,
+      "qua": 56925,
+      "aber": 56926,
+      "Quest": 56927,
+      "▁facets": 56928,
+      "Revision": 56929,
+      "▁elucidated": 56930,
+      "酒店为您提供免费的停车服务": 56931,
+      "虚弱": 56932,
+      "非负": 56933,
+      "仍然是": 56934,
+      "市场中": 56935,
+      "Curve": 56936,
+      "alive": 56937,
+      "▁小说阅读网": 56938,
+      "的逆": 56939,
+      "象棋": 56940,
+      "趁着": 56941,
+      "▁Luckily": 56942,
+      "▁detective": 56943,
+      "♀♀": 56944,
+      "万名": 56945,
+      "既要": 56946,
+      "分析这些": 56947,
+      "▁Lips": 56948,
+      "Seeder": 56949,
+      "oscale": 56950,
+      "nCritical": 56951,
+      "无果": 56952,
+      "让你的": 56953,
+      "▁Scr": 56954,
+      "▁Lists": 56955,
+      "nscript": 56956,
+      "▁chorus": 56957,
+      "citation": 56958,
+      "ndocument": 56959,
+      "唯独": 56960,
+      "小三": 56961,
+      "WAY": 56962,
+      "firm": 56963,
+      "▁polarity": 56964,
+      "▁attributable": 56965,
+      "三部": 56966,
+      "农行": 56967,
+      "中科院": 56968,
+      "有人在": 56969,
+      "CDMA": 56970,
+      "▁LEG": 56971,
+      "▁grit": 56972,
+      "▁Uruguay": 56973,
+      "▁pictured": 56974,
+      "蒙特": 56975,
+      "▁Crawford": 56976,
+      "▁versatility": 56977,
+      "换热": 56978,
+      "璀璨": 56979,
+      "融化": 56980,
+      "较量": 56981,
+      "附着": 56982,
+      "bukkit": 56983,
+      "ncolor": 56984,
+      "factors": 56985,
+      "开场": 56986,
+      "航运": 56987,
+      "nuid": 56988,
+      "就是这么": 56989,
+      "▁woven": 56990,
+      "▁Georgian": 56991,
+      "充要": 56992,
+      "抽屉": 56993,
+      "▁lg": 56994,
+      "听说过": 56995,
+      "hana": 56996,
+      "onyms": 56997,
+      "▁Merr": 56998,
+      "TZ": 56999,
+      "走来": 57000,
+      "CHE": 57001,
+      "这种事情": 57002,
+      "iphertext": 57003,
+      "企查": 57004,
+      "超标": 57005,
+      "查笔录": 57006,
+      "当庭陈述": 57007,
+      "▁Kann": 57008,
+      "ustering": 57009,
+      "▁favored": 57010,
+      "▁aftermath": 57011,
+      "肿胀": 57012,
+      "追随": 57013,
+      "镇痛": 57014,
+      "时之前": 57015,
+      "生产建设": 57016,
+      "科技公司": 57017,
+      "▁Experts": 57018,
+      "▁honorary": 57019,
+      "麻木": 57020,
+      "极高的": 57021,
+      "电子有限公司": 57022,
+      "琵琶": 57023,
+      "青龙": 57024,
+      "发出的": 57025,
+      "ismo": 57026,
+      "▁Sew": 57027,
+      "▁gods": 57028,
+      "如下图所示": 57029,
+      "▁rever": 57030,
+      "inosaur": 57031,
+      "Rectangle": 57032,
+      "MU": 57033,
+      "tA": 57034,
+      "лу": 57035,
+      "美团": 57036,
+      "SAT": 57037,
+      "理解为": 57038,
+      "▁Winner": 57039,
+      "▁Contains": 57040,
+      "ttf": 57041,
+      "ysc": 57042,
+      "▁广东": 57043,
+      "服务和": 57044,
+      "ousand": 57045,
+      "▁stool": 57046,
+      "erebral": 57047,
+      "UserName": 57048,
+      "▁oligonucle": 57049,
+      "上也": 57050,
+      "滤镜": 57051,
+      "▁bowls": 57052,
+      "venture": 57053,
+      "小块": 57054,
+      "昼夜": 57055,
+      "并将其": 57056,
+      "的顺序": 57057,
+      "▁CBC": 57058,
+      "▁Tuc": 57059,
+      "规划设计": 57060,
+      "▁Hann": 57061,
+      "Orders": 57062,
+      "nCities": 57063,
+      "▁Except": 57064,
+      "▁mutated": 57065,
+      "▁BUSINESS": 57066,
+      "▁monsters": 57067,
+      "▁conductance": 57068,
+      "▁southwestern": 57069,
+      "云霄": 57070,
+      "▁rugs": 57071,
+      "palette": 57072,
+      "▁plateau": 57073,
+      "▁Negative": 57074,
+      "确认人民调解协议效力的": 57075,
+      "▁κ": 57076,
+      "涨停": 57077,
+      "协助执行": 57078,
+      "并提出了": 57079,
+      "nAugust": 57080,
+      "一朵": 57081,
+      "点了": 57082,
+      "poons": 57083,
+      "ün": 57084,
+      "天河": 57085,
+      "本国": 57086,
+      "▁Cher": 57087,
+      "Actual": 57088,
+      "▁binaries": 57089,
+      "一年级": 57090,
+      "方位的": 57091,
+      "第四届": 57092,
+      "工作报告": 57093,
+      "▁Birds": 57094,
+      "Natural": 57095,
+      "/[": 57096,
+      "时髦": 57097,
+      "真伪": 57098,
+      "STREAM": 57099,
+      "▁codecs": 57100,
+      "▁deterg": 57101,
+      "小花": 57102,
+      "民事判决书": 57103,
+      "▁pretend": 57104,
+      "oblastoma": 57105,
+      "▁nineteenth": 57106,
+      "住的": 57107,
+      "awai": 57108,
+      "▁Matlab": 57109,
+      "▁transfected": 57110,
+      "tq": 57111,
+      "少林": 57112,
+      "烧热": 57113,
+      "支持的": 57114,
+      "去年同期": 57115,
+      "getElements": 57116,
+      "是与": 57117,
+      "该项": 57118,
+      "支付了": 57119,
+      "政府和": 57120,
+      "组组长": 57121,
+      "绵阳市": 57122,
+      "odule": 57123,
+      "oodle": 57124,
+      "▁Lung": 57125,
+      "Immutable": 57126,
+      "breadcrumb": 57127,
+      "缝隙": 57128,
+      "中国传统": 57129,
+      "▁glitter": 57130,
+      "大明": 57131,
+      "ogh": 57132,
+      "进行计算": 57133,
+      "所欲": 57134,
+      "处三年": 57135,
+      "▁wipe": 57136,
+      "▁exploiting": 57137,
+      "▁instructed": 57138,
+      "以南": 57139,
+      "作弊": 57140,
+      "司职": 57141,
+      "角逐": 57142,
+      "人性化": 57143,
+      "▁MLB": 57144,
+      ";\\\\;\\\\": 57145,
+      "Maximum": 57146,
+      "requently": 57147,
+      "就和": 57148,
+      "点在": 57149,
+      "nCentral": 57150,
+      "▁Cambodia": 57151,
+      "▁SERVICES": 57152,
+      "▁societal": 57153,
+      "扁平": 57154,
+      "有事": 57155,
+      "观影": 57156,
+      "在一些": 57157,
+      "csdn": 57158,
+      "nINS": 57159,
+      "▁ERP": 57160,
+      "应当承担": 57161,
+      "一定要注意": 57162,
+      "▁Remodel": 57163,
+      "不值": 57164,
+      "之分": 57165,
+      "听证": 57166,
+      "辜负": 57167,
+      "Jon": 57168,
+      "是属于": 57169,
+      "此次活动": 57170,
+      "olate": 57171,
+      "▁Waste": 57172,
+      "▁hemod": 57173,
+      "nChange": 57174,
+      "不起来": 57175,
+      "▁RET": 57176,
+      "招商引资": 57177,
+      "istors": 57178,
+      "▁emergencies": 57179,
+      "抽查": 57180,
+      "没有一个": 57181,
+      "▁Schr": 57182,
+      "▁Rovers": 57183,
+      "亲友": 57184,
+      "保利": 57185,
+      "联立": 57186,
+      "nose": 57187,
+      "▁IgA": 57188,
+      "市场份额": 57189,
+      "raeli": 57190,
+      "其他需要裁定": 57191,
+      "▁Hawaiian": 57192,
+      "emm": 57193,
+      "撤回执行申请": 57194,
+      "系该": 57195,
+      "^{*": 57196,
+      "新时期": 57197,
+      "Berry": 57198,
+      "▁quint": 57199,
+      "nTaking": 57200,
+      "▁establishes": 57201,
+      "includegraphics": 57202,
+      "防线": 57203,
+      "成功后": 57204,
+      "有那么": 57205,
+      "nProb": 57206,
+      "等腰三角形的": 57207,
+      "Criteria": 57208,
+      "▁Actress": 57209,
+      "▁solitary": 57210,
+      "中也": 57211,
+      "松原": 57212,
+      "nMs": 57213,
+      "igar": 57214,
+      "除以进率": 57215,
+      "▁Estim": 57216,
+      "▁tails": 57217,
+      "▁uploading": 57218,
+      "乡县": 57219,
+      "几张": 57220,
+      "推崇": 57221,
+      "油炸": 57222,
+      "Tap": 57223,
+      "▁Hib": 57224,
+      "▁PVC": 57225,
+      "▁han": 57226,
+      "▁Hiro": 57227,
+      "之间的距离": 57228,
+      "▁Rapids": 57229,
+      "住房和城乡建设": 57230,
+      "nArticle": 57231,
+      "Bootstrap": 57232,
+      "publication": 57233,
+      "封锁": 57234,
+      "比对照": 57235,
+      "nDan": 57236,
+      "恶性肿瘤": 57237,
+      "borns": 57238,
+      "▁noqa": 57239,
+      "wrapped": 57240,
+      "▁customization": 57241,
+      "杏仁": 57242,
+      "nSong": 57243,
+      "▁Tracy": 57244,
+      "▁serine": 57245,
+      "▁Freeman": 57246,
+      "traumatic": 57247,
+      "determined": 57248,
+      "▁placental": 57249,
+      "泪水": 57250,
+      "▁FIX": 57251,
+      "▁Uns": 57252,
+      "▁ich": 57253,
+      "▁Lomb": 57254,
+      "nselect": 57255,
+      "▁Compact": 57256,
+      "▁endeavor": 57257,
+      "▁Throwable": 57258,
+      "暂行": 57259,
+      "眼看": 57260,
+      "骨科": 57261,
+      "rng": 57262,
+      "买受人": 57263,
+      "化学物质": 57264,
+      "fault": 57265,
+      "▁Chrys": 57266,
+      "entropy": 57267,
+      "金秋": 57268,
+      "上一层": 57269,
+      "nJoe": 57270,
+      "默契": 57271,
+      "家庭成员": 57272,
+      "piratory": 57273,
+      "▁revenge": 57274,
+      "▁ventricle": 57275,
+      "好坏": 57276,
+      "梧桐": 57277,
+      "uds": 57278,
+      "▁Wo": 57279,
+      "弹幕": 57280,
+      "nets": 57281,
+      "管理模式": 57282,
+      "▁Cases": 57283,
+      "axel": 57284,
+      "▁harb": 57285,
+      "getenv": 57286,
+      "▁coats": 57287,
+      "▁flaws": 57288,
+      "▁divorced": 57289,
+      "中位数": 57290,
+      "中发现": 57291,
+      "websocket": 57292,
+      "▁driveway": 57293,
+      "▁timedelta": 57294,
+      "vk": 57295,
+      "继而": 57296,
+      "▁Iz": 57297,
+      "女生的": 57298,
+      "▁Needs": 57299,
+      "orporate": 57300,
+      "▁detained": 57301,
+      "虾仁": 57302,
+      "faq": 57303,
+      "▁Fen": 57304,
+      "且在": 57305,
+      "治国": 57306,
+      "薄荷": 57307,
+      "cron": 57308,
+      "▁[{\"": 57309,
+      "▁pym": 57310,
+      "arris": 57311,
+      "▁Carson": 57312,
+      "▁perceptual": 57313,
+      "▁hapl": 57314,
+      "▁vegetarian": 57315,
+      "一轮": 57316,
+      "要好": 57317,
+      "inde": 57318,
+      "▁Lagos": 57319,
+      "▁phenyl": 57320,
+      "MERCHANTABILITY": 57321,
+      "▁yo": 57322,
+      "等进行": 57323,
+      "ndec": 57324,
+      "tape": 57325,
+      "▁subdivision": 57326,
+      "串联": 57327,
+      "崇尚": 57328,
+      "观摩": 57329,
+      "的情绪": 57330,
+      "nRoad": 57331,
+      "anched": 57332,
+      "▁degraded": 57333,
+      "▁serializers": 57334,
+      "校验": 57335,
+      "累犯": 57336,
+      "教育出版社": 57337,
+      "Carthy": 57338,
+      "ToList": 57339,
+      "▁Epidem": 57340,
+      "▁Lightning": 57341,
+      "滋养": 57342,
+      "镶嵌": 57343,
+      "Mil": 57344,
+      "低成本": 57345,
+      "要坚持": 57346,
+      "▁社会": 57347,
+      "力量的": 57348,
+      "公共卫生": 57349,
+      "govuk": 57350,
+      "▁Midlands": 57351,
+      "▁deliberately": 57352,
+      "脊柱": 57353,
+      "首创": 57354,
+      "该系统": 57355,
+      "▁Fans": 57356,
+      "低保": 57357,
+      "先予": 57358,
+      "▁ig": 57359,
+      "到直线": 57360,
+      "直辖市": 57361,
+      "nTer": 57362,
+      "bucks": 57363,
+      "chdir": 57364,
+      "▁Folk": 57365,
+      "▁idle": 57366,
+      "▁linen": 57367,
+      "▁Bernie": 57368,
+      "▁Mozilla": 57369,
+      "▁devised": 57370,
+      "▁recharge": 57371,
+      "卖了": 57372,
+      "拉动": 57373,
+      "接线": 57374,
+      "效率高": 57375,
+      "Trump": 57376,
+      "▁kidn": 57377,
+      "nPhoto": 57378,
+      "▁todos": 57379,
+      "光华": 57380,
+      "陆地": 57381,
+      "bac": 57382,
+      "▁Evil": 57383,
+      "ры": 57384,
+      "道路交通安全": 57385,
+      "▁boosting": 57386,
+      "▁fallback": 57387,
+      "▁IllegalArgument": 57388,
+      "伸手": 57389,
+      "初学者": 57390,
+      "▁Injury": 57391,
+      ">@": 57392,
+      "屠杀": 57393,
+      "流泪": 57394,
+      "划分为": 57395,
+      "可随时": 57396,
+      "的科学": 57397,
+      "nKim": 57398,
+      "不必要的": 57399,
+      "对外开放": 57400,
+      "Padding": 57401,
+      "▁Priority": 57402,
+      "Credential": 57403,
+      "利好": 57404,
+      "后天": 57405,
+      "了他的": 57406,
+      "rios": 57407,
+      "▁Caf": 57408,
+      "affin": 57409,
+      "ogeneic": 57410,
+      "本次执行程序中": 57411,
+      "▁intraoperative": 57412,
+      "\"^": 57413,
+      "光彩": 57414,
+      "安东": 57415,
+      "漫步": 57416,
+      "dims": 57417,
+      "▁pinch": 57418,
+      "arrison": 57419,
+      "▁autore": 57420,
+      "▁Verizon": 57421,
+      "一刀": 57422,
+      "与王": 57423,
+      "史诗": 57424,
+      "使用时": 57425,
+      "多年前": 57426,
+      "prof": 57427,
+      "但是对于": 57428,
+      "▁Employ": 57429,
+      "▁alloys": 57430,
+      "▁GameObject": 57431,
+      "挥发": 57432,
+      "舍友": 57433,
+      "起身": 57434,
+      "学研究": 57435,
+      "▁Madd": 57436,
+      "things": 57437,
+      "▁sanctuary": 57438,
+      "村人": 57439,
+      "lio": 57440,
+      "▁Pole": 57441,
+      "SError": 57442,
+      "中天": 57443,
+      "不太好": 57444,
+      "请不要": 57445,
+      "▁Luk": 57446,
+      "pytorch": 57447,
+      "▁Patricia": 57448,
+      "德的": 57449,
+      "甘油": 57450,
+      "bly": 57451,
+      "据报道": 57452,
+      "issuecomment": 57453,
+      "侵入": 57454,
+      "地里": 57455,
+      "▁UR": 57456,
+      "▁这里": 57457,
+      "Lead": 57458,
+      "▁山东省": 57459,
+      "计数单位": 57460,
+      "ır": 57461,
+      "加持": 57462,
+      "店家": 57463,
+      "茂名": 57464,
+      "▁nip": 57465,
+      "trend": 57466,
+      "▁包装清单": 57467,
+      "委托代理人陈": 57468,
+      "▁ingestion": 57469,
+      "高高": 57470,
+      "STD": 57471,
+      "经营项目": 57472,
+      "ampoo": 57473,
+      "ophilia": 57474,
+      "▁Laurent": 57475,
+      "▁Survival": 57476,
+      "▁biologic": 57477,
+      "社员": 57478,
+      "召开的": 57479,
+      "▁ferr": 57480,
+      "▁funct": 57481,
+      "atemala": 57482,
+      "Workspace": 57483,
+      "SerializedName": 57484,
+      "绘图": 57485,
+      "充足的": 57486,
+      "各行各": 57487,
+      "效果图": 57488,
+      "thus": 57489,
+      "uren": 57490,
+      "▁luminosity": 57491,
+      "花鸟": 57492,
+      "衰退": 57493,
+      "学专业": 57494,
+      "urus": 57495,
+      "▁Chest": 57496,
+      "▁nostalg": 57497,
+      "unciation": 57498,
+      "Relationship": 57499,
+      "单单": 57500,
+      "复工": 57501,
+      "结果为": 57502,
+      "▁Walsh": 57503,
+      "▁Massage": 57504,
+      "▁welcomes": 57505,
+      "▁请你": 57506,
+      "等原因": 57507,
+      "转移到": 57508,
+      "Train": 57509,
+      "inalg": 57510,
+      "解决的事项": 57511,
+      "atured": 57512,
+      "▁delib": 57513,
+      "▁bacterium": 57514,
+      "▁Southampton": 57515,
+      "代的": 57516,
+      "剑桥": 57517,
+      "大二": 57518,
+      "▁nh": 57519,
+      "adir": 57520,
+      "aspect": 57521,
+      "ionine": 57522,
+      "▁organizer": 57523,
+      "克制": 57524,
+      "无证": 57525,
+      "胖子": 57526,
+      "集结": 57527,
+      "一家人": 57528,
+      "都应该": 57529,
+      "年被评为": 57530,
+      "▁Chop": 57531,
+      "odontic": 57532,
+      "▁negotiating": 57533,
+      "ADS": 57534,
+      "但他们": 57535,
+      "struments": 57536,
+      "▁点": 57537,
+      "也别": 57538,
+      "战线": 57539,
+      "李明": 57540,
+      "肛门": 57541,
+      "鹦鹉": 57542,
+      "▁(:": 57543,
+      "solver": 57544,
+      "'}": 57545,
+      "▁重": 57546,
+      "Ring": 57547,
+      "▁pipelines": 57548,
+      "暗恋": 57549,
+      "accum": 57550,
+      "▁stabilized": 57551,
+      "跟她": 57552,
+      "▁Eb": 57553,
+      "但现在": 57554,
+      "证件号码": 57555,
+      "罚金人民币": 57556,
+      "▁Libert": 57557,
+      "▁pledge": 57558,
+      "getCurrent": 57559,
+      "切削": 57560,
+      "吃喝": 57561,
+      "眼底": 57562,
+      "你这个": 57563,
+      "oten": 57564,
+      "▁nach": 57565,
+      "▁aston": 57566,
+      "▁swept": 57567,
+      "▁poultry": 57568,
+      "▁Officers": 57569,
+      "病害": 57570,
+      "cir": 57571,
+      "pole": 57572,
+      "▁taxpayers": 57573,
+      "不曾": 57574,
+      "睫毛": 57575,
+      "语义": 57576,
+      "nRH": 57577,
+      "慢慢地": 57578,
+      "爱奇艺": 57579,
+      "中国平安": 57580,
+      "ernal": 57581,
+      "actical": 57582,
+      "ου": 57583,
+      "用什么": 57584,
+      "nSun": 57585,
+      "etable": 57586,
+      "▁inning": 57587,
+      "▁puppet": 57588,
+      "▁depicts": 57589,
+      "*>": 57590,
+      "▁Zh": 57591,
+      "看了看": 57592,
+      "起诉的": 57593,
+      "sulf": 57594,
+      "▁novelists": 57595,
+      "UU": 57596,
+      "民三": 57597,
+      "粉色": 57598,
+      "PUBLIC": 57599,
+      "▁efficiencies": 57600,
+      "娘家": 57601,
+      "科协": 57602,
+      "})}{": 57603,
+      "▁Admiral": 57604,
+      "▁Raiders": 57605,
+      "\"*": 57606,
+      "早早": 57607,
+      "死神": 57608,
+      "句话说": 57609,
+      "打开了": 57610,
+      "▁Jupiter": 57611,
+      "Attachment": 57612,
+      "▁stratified": 57613,
+      "中立": 57614,
+      "省会": 57615,
+      "矢量": 57616,
+      "├──": 57617,
+      "▁排名": 57618,
+      "只要有": 57619,
+      "ricks": 57620,
+      "▁Cron": 57621,
+      "▁dude": 57622,
+      "▁Hayes": 57623,
+      "popular": 57624,
+      "▁obsolete": 57625,
+      "一味": 57626,
+      "八角": 57627,
+      "二〇一五年十月": 57628,
+      "平安银行股份有限公司上海": 57629,
+      "的时刻": 57630,
+      "▁ZIP": 57631,
+      "违法犯罪": 57632,
+      "nover": 57633,
+      "ollen": 57634,
+      "▁Jung": 57635,
+      "nshould": 57636,
+      "▁insane": 57637,
+      "▁discret": 57638,
+      "予以减刑": 57639,
+      "keras": 57640,
+      "▁STUD": 57641,
+      "▁sixty": 57642,
+      "▁glorious": 57643,
+      "themed": 57644,
+      "▁Frost": 57645,
+      "nFactor": 57646,
+      "▁partnering": 57647,
+      "长是": 57648,
+      "pez": 57649,
+      "▁Rw": 57650,
+      "▁职称": 57651,
+      "Physics": 57652,
+      "▁monastery": 57653,
+      "繁琐": 57654,
+      "执行措施": 57655,
+      "金属材料": 57656,
+      "orbit": 57657,
+      "▁Sisters": 57658,
+      "代为": 57659,
+      "向后": 57660,
+      "尼奥": 57661,
+      "想看": 57662,
+      "▁resh": 57663,
+      "ivating": 57664,
+      "▁Patrol": 57665,
+      "▁issuing": 57666,
+      "前科": 57667,
+      "PRS": 57668,
+      "▁Rif": 57669,
+      "▁Riv": 57670,
+      "▁mug": 57671,
+      "函数的图象": 57672,
+      "存入": 57673,
+      "请假": 57674,
+      "雨中": 57675,
+      "范围的": 57676,
+      "▁exh": 57677,
+      "nUpon": 57678,
+      "遵守监规纪律": 57679,
+      "Kn": 57680,
+      "继承人": 57681,
+      "▁cathe": 57682,
+      "nSupport": 57683,
+      "indicator": 57684,
+      "揭开": 57685,
+      "贵人": 57686,
+      "cts": 57687,
+      "arya": 57688,
+      "▁Dund": 57689,
+      "▁batches": 57690,
+      "▁outflow": 57691,
+      "▁mosquito": 57692,
+      "▁cisplatin": 57693,
+      "万象": 57694,
+      "重工": 57695,
+      "义乌市": 57696,
+      "的秘密": 57697,
+      "▁grape": 57698,
+      "nLeague": 57699,
+      "Velocity": 57700,
+      "nWORKDIR": 57701,
+      "▁campuses": 57702,
+      "IEnumerable": 57703,
+      "和他们": 57704,
+      "的父亲": 57705,
+      "皮肤病": 57706,
+      "资源和": 57707,
+      "邓小平": 57708,
+      "▁DEF": 57709,
+      "lective": 57710,
+      "▁宁": 57711,
+      "▁hh": 57712,
+      "极大地": 57713,
+      "pora": 57714,
+      "▁Mud": 57715,
+      "Leave": 57716,
+      "▁aesthetics": 57717,
+      "▁endorsement": 57718,
+      "米粉": 57719,
+      "创新性": 57720,
+      "须经批准的项目": 57721,
+      "安保": 57722,
+      "ELS": 57723,
+      "bam": 57724,
+      "你没有": 57725,
+      "遇到过": 57726,
+      "osest": 57727,
+      "▁heme": 57728,
+      "▁muff": 57729,
+      "sector": 57730,
+      "Locator": 57731,
+      "▁acuity": 57732,
+      "▁bounding": 57733,
+      "contributors": 57734,
+      "监狱认为": 57735,
+      "nApril": 57736,
+      "▁costing": 57737,
+      "该车": 57738,
+      "vik": 57739,
+      "对社会": 57740,
+      "科研工作": 57741,
+      "trial": 57742,
+      "nColor": 57743,
+      "located": 57744,
+      "▁turtle": 57745,
+      "▁bindings": 57746,
+      "conversion": 57747,
+      "▁exchanged": 57748,
+      "橘子": 57749,
+      "lol": 57750,
+      "日受理": 57751,
+      "▁_(\"": 57752,
+      "▁Guatemala": 57753,
+      "之光": 57754,
+      "atoms": 57755,
+      "同底数幂的": 57756,
+      "▁CHECK": 57757,
+      "▁Quart": 57758,
+      "▁Earlier": 57759,
+      "▁elapsed": 57760,
+      "打赢": 57761,
+      "骨架": 57762,
+      "打破了": 57763,
+      "Cards": 57764,
+      "千瓦": 57765,
+      "药店": 57766,
+      "阳春": 57767,
+      "\\'{": 57768,
+      "Both": 57769,
+      "▁Rao": 57770,
+      "▁SARS": 57771,
+      "▁asbestos": 57772,
+      "微信号": 57773,
+      "此题考查": 57774,
+      "▁Hond": 57775,
+      "▁mesenchymal": 57776,
+      "技术推广": 57777,
+      "cet": 57778,
+      "把这些": 57779,
+      "根据不同": 57780,
+      "eastern": 57781,
+      "▁liking": 57782,
+      "nWriters": 57783,
+      "iteritems": 57784,
+      "▁maritime": 57785,
+      "completion": 57786,
+      "▁characterised": 57787,
+      "在国家": 57788,
+      "émon": 57789,
+      "我们必须": 57790,
+      "probe": 57791,
+      "▁Meteor": 57792,
+      "到场": 57793,
+      "藏书": 57794,
+      "也只能": 57795,
+      "blic": 57796,
+      "▁JUST": 57797,
+      "▁baths": 57798,
+      "polymer": 57799,
+      "▁stealing": 57800,
+      "bos": 57801,
+      "不满意": 57802,
+      "eous": 57803,
+      "rink": 57804,
+      "▁McN": 57805,
+      "▁gcd": 57806,
+      "负责同志": 57807,
+      "ocado": 57808,
+      "二〇一七年十二": 57809,
+      "▁sharply": 57810,
+      "敏锐": 57811,
+      "▁\"$(": 57812,
+      "明确表示": 57813,
+      "是一家集": 57814,
+      "OLDER": 57815,
+      "ATFORM": 57816,
+      "eprint": 57817,
+      "toArray": 57818,
+      "▁contracting": 57819,
+      "龙华": 57820,
+      "]\"\\": 57821,
+      "▁bac": 57822,
+      "nPlaying": 57823,
+      "▁Winners": 57824,
+      "▁polypeptide": 57825,
+      "indi": 57826,
+      "onance": 57827,
+      "Feedback": 57828,
+      "▁oocytes": 57829,
+      "▁estradiol": 57830,
+      "经商": 57831,
+      "Lite": 57832,
+      "UInt": 57833,
+      "▁HOME": 57834,
+      "工匠": 57835,
+      "从这个": 57836,
+      "周杰伦": 57837,
+      "oxel": 57838,
+      "▁herm": 57839,
+      "HW": 57840,
+      "kbd": 57841,
+      "▁Wid": 57842,
+      "不同程度的": 57843,
+      "▁Algebraic": 57844,
+      "将要": 57845,
+      "直销": 57846,
+      "▁perm": 57847,
+      "▁Cheng": 57848,
+      "▁effic": 57849,
+      "▁Vertical": 57850,
+      "想念": 57851,
+      "kid": 57852,
+      "百分比": 57853,
+      "▁cmp": 57854,
+      "感觉自己": 57855,
+      "folios": 57856,
+      "杜鹃": 57857,
+      "meet": 57858,
+      "▁Emmy": 57859,
+      "▁Serve": 57860,
+      "Keyword": 57861,
+      "▁ulcers": 57862,
+      "▁Maurice": 57863,
+      "▁connective": 57864,
+      "最初的": 57865,
+      "还会有": 57866,
+      "emis": 57867,
+      "确诊病例": 57868,
+      "nDetails": 57869,
+      "▁disconnected": 57870,
+      "专为": 57871,
+      "冷暖": 57872,
+      "地去": 57873,
+      "通讯员": 57874,
+      "}}}}": 57875,
+      "GRAPH": 57876,
+      "enden": 57877,
+      "▁Haskell": 57878,
+      "át": 57879,
+      "▁Ages": 57880,
+      "ncount": 57881,
+      "▁Carpet": 57882,
+      "▁firewall": 57883,
+      "continuous": 57884,
+      "百强": 57885,
+      "翻身": 57886,
+      "喜欢的人": 57887,
+      "▁Huff": 57888,
+      "▁生长环境": 57889,
+      "▁sequenced": 57890,
+      "EXP": 57891,
+      "第一家": 57892,
+      "▁Styl": 57893,
+      "edited": 57894,
+      "▁injective": 57895,
+      "▁概述": 57896,
+      "小女孩": 57897,
+      "自我的": 57898,
+      "EMPL": 57899,
+      "▁zur": 57900,
+      "AMPLE": 57901,
+      "plicant": 57902,
+      "▁Measures": 57903,
+      "▁Alexandria": 57904,
+      "▁Ł": 57905,
+      "他就是": 57906,
+      "上诉请求": 57907,
+      "▁contests": 57908,
+      "整套": 57909,
+      "velt": 57910,
+      "九江": 57911,
+      "我好": 57912,
+      "经络": 57913,
+      "与时俱": 57914,
+      "营业额": 57915,
+      "透明的": 57916,
+      "unov": 57917,
+      "▁[…]\\": 57918,
+      "stress": 57919,
+      "▁learner": 57920,
+      "▁adapting": 57921,
+      "▁dwelling": 57922,
+      "▁liquidity": 57923,
+      "▁reminiscent": 57924,
+      "罗拉": 57925,
+      "酸钠": 57926,
+      "可谓是": 57927,
+      "together": 57928,
+      "▁biochemistry": 57929,
+      "多糖": 57930,
+      "Truth": 57931,
+      "▁Dial": 57932,
+      "npr": 57933,
+      "的品质": 57934,
+      "ribe": 57935,
+      "RELEASE": 57936,
+      "preferred": 57937,
+      "^\\": 57938,
+      "yii": 57939,
+      "有三个": 57940,
+      "的含义": 57941,
+      "▁}$$": 57942,
+      "Visit": 57943,
+      "▁Otto": 57944,
+      "▁committing": 57945,
+      "对数": 57946,
+      "秘诀": 57947,
+      "移除": 57948,
+      "随心": 57949,
+      "变量的": 57950,
+      "▁Pip": 57951,
+      "▁Cham": 57952,
+      "opened": 57953,
+      "▁steak": 57954,
+      "▁tracer": 57955,
+      "atization": 57956,
+      "▁Mortgage": 57957,
+      "指数幂": 57958,
+      "这些年": 57959,
+      "值得关注": 57960,
+      "▁damping": 57961,
+      "▁Subscribe": 57962,
+      "▁∈": 57963,
+      "平行的": 57964,
+      "▁doesnt": 57965,
+      "准许或者不准许撤诉": 57966,
+      "通货": 57967,
+      "audit": 57968,
+      "ermis": 57969,
+      "quiet": 57970,
+      "本田": 57971,
+      "Sex": 57972,
+      "音乐会": 57973,
+      "问题如下": 57974,
+      "acute": 57975,
+      "▁chefs": 57976,
+      "▁grapes": 57977,
+      "optimizer": 57978,
+      "▁cryptography": 57979,
+      "▁informational": 57980,
+      "不孕": 57981,
+      "处级": 57982,
+      "富强": 57983,
+      "抵御": 57984,
+      "第二百零五条": 57985,
+      "▁Stainless": 57986,
+      "银耳": 57987,
+      "CRM": 57988,
+      "不规则": 57989,
+      "染色体": 57990,
+      "等服务": 57991,
+      "第二部分": 57992,
+      "▁leng": 57993,
+      "▁über": 57994,
+      "▁poorer": 57995,
+      "Personal": 57996,
+      "▁inducted": 57997,
+      "打扰": 57998,
+      "SOD": 57999,
+      "icher": 58000,
+      "▁feather": 58001,
+      "▁roasted": 58002,
+      "外加": 58003,
+      "Euro": 58004,
+      "squared": 58005,
+      "▁Archae": 58006,
+      "▁torsion": 58007,
+      "IRS": 58008,
+      "▁мо": 58009,
+      "clusters": 58010,
+      "▁marketers": 58011,
+      "自拍": 58012,
+      "正常人": 58013,
+      "理解和": 58014,
+      "相比较": 58015,
+      "▁Func": 58016,
+      "▁fundamentals": 58017,
+      "缓交": 58018,
+      "Svg": 58019,
+      "第十八": 58020,
+      "课程的": 58021,
+      "ansen": 58022,
+      "ttime": 58023,
+      "encycl": 58024,
+      "iguous": 58025,
+      "▁Raven": 58026,
+      "▁citrus": 58027,
+      "呜呜": 58028,
+      "lew": 58029,
+      "事物的": 58030,
+      "好吃的": 58031,
+      "百分百": 58032,
+      "这辈子": 58033,
+      "▁Bis": 58034,
+      "对该领域的": 58035,
+      "▁Pirates": 58036,
+      "▁prepares": 58037,
+      "▁petroleum": 58038,
+      "国情": 58039,
+      "彼岸": 58040,
+      "我真": 58041,
+      "esteem": 58042,
+      "十二五": 58043,
+      "▁doubling": 58044,
+      "均需": 58045,
+      "关联性": 58046,
+      "志愿军": 58047,
+      "执行款": 58048,
+      "面上的": 58049,
+      "▁Jill": 58050,
+      "末日": 58051,
+      "十二年": 58052,
+      "snow": 58053,
+      "▁vor": 58054,
+      "Absolute": 58055,
+      "orgetown": 58056,
+      "▁Helsinki": 58057,
+      "▁Automation": 58058,
+      "机房": 58059,
+      "▁chin": 58060,
+      "locals": 58061,
+      "▁widow": 58062,
+      "▁medall": 58063,
+      "trequire": 58064,
+      "▁kitchens": 58065,
+      "从左": 58066,
+      "也一样": 58067,
+      "涵盖了": 58068,
+      "通过分析": 58069,
+      "Album": 58070,
+      "fullName": 58071,
+      "▁Benefits": 58072,
+      "trightarrow": 58073,
+      "RW": 58074,
+      "缘故": 58075,
+      "离家出走": 58076,
+      "Instruction": 58077,
+      "分流": 58078,
+      "迷惑": 58079,
+      "不至于": 58080,
+      "出了一": 58081,
+      "拿到了": 58082,
+      "),\\\\": 58083,
+      "▁coup": 58084,
+      "▁Aggreg": 58085,
+      "▁hashlib": 58086,
+      "▁explorer": 58087,
+      "▁Individuals": 58088,
+      "千载": 58089,
+      "汇票": 58090,
+      "desk": 58091,
+      "管理学院": 58092,
+      "▁infl": 58093,
+      "▁slab": 58094,
+      "legant": 58095,
+      "spinal": 58096,
+      "▁Miche": 58097,
+      "▁scarce": 58098,
+      "▁trendy": 58099,
+      "我县": 58100,
+      "CRR": 58101,
+      "活性炭": 58102,
+      "的根本": 58103,
+      "不断完善": 58104,
+      "Trade": 58105,
+      "▁Cube": 58106,
+      "▁Joined": 58107,
+      "冻结财产的规定": 58108,
+      "▁analogy": 58109,
+      "detection": 58110,
+      "▁offenders": 58111,
+      "ну": 58112,
+      "降到": 58113,
+      "也未能": 58114,
+      "月份的": 58115,
+      "▁whistle": 58116,
+      "▁histology": 58117,
+      "▁resilient": 58118,
+      "▁irradiated": 58119,
+      "也太": 58120,
+      "ilight": 58121,
+      "尖锐": 58122,
+      "抗疫": 58123,
+      "在上诉": 58124,
+      "精美的": 58125,
+      "▁SNP": 58126,
+      "▁SUP": 58127,
+      "▁owe": 58128,
+      "相关规定": 58129,
+      "▁Paso": 58130,
+      "▁Rising": 58131,
+      "依法由代理审判员": 58132,
+      "▁destroying": 58133,
+      "XA": 58134,
+      "万科": 58135,
+      "公司与": 58136,
+      "聊城市": 58137,
+      "▁cleft": 58138,
+      "injection": 58139,
+      "iographic": 58140,
+      "▁exhausted": 58141,
+      "其他需要裁定解决的事项": 58142,
+      "案件适用法律若干问题的规定": 58143,
+      "悬崖": 58144,
+      "自己喜欢的": 58145,
+      "▁Plants": 58146,
+      "▁ageing": 58147,
+      "▁papill": 58148,
+      "nservices": 58149,
+      "剂的": 58150,
+      "重磅": 58151,
+      "老师和": 58152,
+      "▁ALS": 58153,
+      "▁sto": 58154,
+      "▁svg": 58155,
+      "一条直线": 58156,
+      "可以实现": 58157,
+      "▁brig": 58158,
+      "▁kerat": 58159,
+      "事发": 58160,
+      "年均": 58161,
+      "点上": 58162,
+      "电影的": 58163,
+      "uder": 58164,
+      "Oracle": 58165,
+      "nGrand": 58166,
+      "选择正确答案": 58167,
+      "▁expresses": 58168,
+      "▁commenting": 58169,
+      "▁constituent": 58170,
+      "翻转": 58171,
+      "▁南京": 58172,
+      "请大家": 58173,
+      "这都是": 58174,
+      "AccessToken": 58175,
+      "nWashington": 58176,
+      "▁discriminate": 58177,
+      "®,": 58178,
+      "▁Acqu": 58179,
+      "▁Adds": 58180,
+      "▁derives": 58181,
+      "introduction": 58182,
+      "RequestHandler": 58183,
+      "JJ": 58184,
+      "借用": 58185,
+      "势必": 58186,
+      "umbo": 58187,
+      "是最好的": 58188,
+      "Deleted": 58189,
+      "timestamps": 58190,
+      "为她": 58191,
+      "住户": 58192,
+      "尿道": 58193,
+      "摩尔": 58194,
+      "第三条": 58195,
+      "罪行的": 58196,
+      "签字确认": 58197,
+      "nLoad": 58198,
+      "▁根据您的": 58199,
+      "▁horizontally": 58200,
+      "VL": 58201,
+      "且仅": 58202,
+      "REL": 58203,
+      "所依据": 58204,
+      "▁Fur": 58205,
+      "▁Tiny": 58206,
+      "pherical": 58207,
+      "▁Uniform": 58208,
+      "▁computerized": 58209,
+      "怀着": 58210,
+      "教科书": 58211,
+      "▁Gothic": 58212,
+      "▁plausible": 58213,
+      "大树": 58214,
+      "在收到": 58215,
+      "▁Chiefs": 58216,
+      "▁Monter": 58217,
+      "▁Leaders": 58218,
+      "▁strangers": 58219,
+      "适用简易程序公开开庭进行了审理": 58220,
+      "春雨": 58221,
+      "成本低": 58222,
+      "某的证言": 58223,
+      "▁Vale": 58224,
+      "▁barbec": 58225,
+      "在道路上醉酒驾驶机动车": 58226,
+      "▁discrepancy": 58227,
+      "传销": 58228,
+      "翩翩": 58229,
+      "聘用": 58230,
+      "通常是": 58231,
+      "traff": 58232,
+      "▁Erik": 58233,
+      "▁bets": 58234,
+      "单体": 58235,
+      "esk": 58236,
+      "▁重庆": 58237,
+      "长三角": 58238,
+      "▁AFC": 58239,
+      "▁Pert": 58240,
+      "▁Myers": 58241,
+      "▁Kuwait": 58242,
+      "▁Embassy": 58243,
+      "新旧": 58244,
+      "触控": 58245,
+      "▁KY": 58246,
+      "Greg": 58247,
+      "amoto": 58248,
+      "向本院提供": 58249,
+      "▁consect": 58250,
+      "▁Diagnostic": 58251,
+      "▁unfamiliar": 58252,
+      "gtk": 58253,
+      "Mutable": 58254,
+      "▁Candid": 58255,
+      "▁marital": 58256,
+      "▁reacted": 58257,
+      "Kab": 58258,
+      "ansk": 58259,
+      "moving": 58260,
+      "王晓": 58261,
+      "母亲的": 58262,
+      "表达了": 58263,
+      "MARK": 58264,
+      "被他": 58265,
+      "\"{%": 58266,
+      "与直线": 58267,
+      "itin": 58268,
+      "▁maneuver": 58269,
+      "▁â": 58270,
+      "校内": 58271,
+      "芳香": 58272,
+      "故事的": 58273,
+      "istas": 58274,
+      "rival": 58275,
+      "uguese": 58276,
+      "▁postpartum": 58277,
+      "外侧": 58278,
+      "理所": 58279,
+      "cxx": 58280,
+      "漫长的": 58281,
+      "ansom": 58282,
+      "▁Agents": 58283,
+      "▁cracking": 58284,
+      "▁metaphor": 58285,
+      "浙江大学": 58286,
+      "otropy": 58287,
+      "可可": 58288,
+      "岳阳": 58289,
+      "益气": 58290,
+      "媒体的": 58291,
+      "farm": 58292,
+      "▁Agn": 58293,
+      "▁Lily": 58294,
+      "parsers": 58295,
+      "河水": 58296,
+      "▁Dee": 58297,
+      "▁Webb": 58298,
+      "nSaint": 58299,
+      "nUntil": 58300,
+      "▁giveaway": 58301,
+      "再也不": 58302,
+      "STAMP": 58303,
+      "▁Greene": 58304,
+      "勋章": 58305,
+      "代表了": 58306,
+      "iked": 58307,
+      "▁sinc": 58308,
+      "Mockito": 58309,
+      "riterion": 58310,
+      "▁flavours": 58311,
+      "分红": 58312,
+      "长了": 58313,
+      "隔热": 58314,
+      "opolys": 58315,
+      "▁Virus": 58316,
+      "blocking": 58317,
+      "由题意可得": 58318,
+      "▁JD": 58319,
+      "ATIONAL": 58320,
+      "▁Galois": 58321,
+      "▁Diagram": 58322,
+      "一出": 58323,
+      "上古": 58324,
+      "上都": 58325,
+      "之妻": 58326,
+      "农商": 58327,
+      "▁Suk": 58328,
+      "▁Omega": 58329,
+      "▁gallon": 58330,
+      "对该领域的了解": 58331,
+      "随风": 58332,
+      "齐鲁": 58333,
+      "有限元": 58334,
+      "八条之规定": 58335,
+      "请仔细阅读选项": 58336,
+      "scenario": 58337,
+      "您对该领域的了解": 58338,
+      "▁declines": 58339,
+      "这是一道关于多学科知识的选择题": 58340,
+      "中去": 58341,
+      "居多": 58342,
+      "牌照": 58343,
+      "xor": 58344,
+      "Crypt": 58345,
+      "Browse": 58346,
+      "▁Norton": 58347,
+      "传出": 58348,
+      "依题意": 58349,
+      "oyer": 58350,
+      "vine": 58351,
+      "▁AFP": 58352,
+      "一如既往": 58353,
+      "审慎考虑": 58354,
+      "▁Cairo": 58355,
+      "▁Civic": 58356,
+      "Injection": 58357,
+      "停在": 58358,
+      "现存": 58359,
+      "▁用法": 58360,
+      "因生活": 58361,
+      "nLes": 58362,
+      "poon": 58363,
+      "▁Bax": 58364,
+      "uttle": 58365,
+      "证据确实充分": 58366,
+      "▁overwrite": 58367,
+      "Applications": 58368,
+      "▁hypothalamic": 58369,
+      "失常": 58370,
+      "美方": 58371,
+      "致的": 58372,
+      "Aug": 58373,
+      "▁//\\": 58374,
+      "nModern": 58375,
+      "▁Forums": 58376,
+      "二〇一八年八月": 58377,
+      "boundary": 58378,
+      "purchase": 58379,
+      "自贸": 58380,
+      "lea": 58381,
+      "为宗旨": 58382,
+      "▁calf": 58383,
+      "▁Richards": 58384,
+      "/…": 58385,
+      "Spy": 58386,
+      "uled": 58387,
+      "都有哪些": 58388,
+      "是一道选择题": 58389,
+      "▁analyzer": 58390,
+      "▁proceeded": 58391,
+      "▁duplication": 58392,
+      "名声": 58393,
+      "咫尺": 58394,
+      "霹雳": 58395,
+      "ALK": 58396,
+      "ITA": 58397,
+      "▁Brock": 58398,
+      "SpringBoot": 58399,
+      "月明": 58400,
+      "交汇处": 58401,
+      "出席了": 58402,
+      "▁ebook": 58403,
+      "▁selects": 58404,
+      "白石": 58405,
+      "▁keras": 58406,
+      "办理结婚登记": 58407,
+      "activities": 58408,
+      "▁燕": 58409,
+      "向着": 58410,
+      "秀丽": 58411,
+      "质检": 58412,
+      "实实在": 58413,
+      "geons": 58414,
+      "▁pedal": 58415,
+      "createdAt": 58416,
+      "▁Featuring": 58417,
+      "责编": 58418,
+      "高二": 58419,
+      "hog": 58420,
+      "微博上": 58421,
+      "nresult": 58422,
+      "▁Bolivia": 58423,
+      "▁Volunteer": 58424,
+      "▁subscribed": 58425,
+      "|<": 58426,
+      "▁🙂": 58427,
+      "服药": 58428,
+      "ibm": 58429,
+      "мен": 58430,
+      "听起来": 58431,
+      "规划和": 58432,
+      "▁CHO": 58433,
+      "▁\\')": 58434,
+      "issued": 58435,
+      "isement": 58436,
+      "▁Hansen": 58437,
+      "▁patrons": 58438,
+      "王文": 58439,
+      "肤色": 58440,
+      "驾驭": 58441,
+      "$^{": 58442,
+      "买了一": 58443,
+      "二十五": 58444,
+      "会发现": 58445,
+      "orbed": 58446,
+      "▁wird": 58447,
+      "▁peppers": 58448,
+      "▁numbering": 58449,
+      "你得": 58450,
+      "进驻": 58451,
+      "ESP": 58452,
+      "nEast": 58453,
+      "hosted": 58454,
+      "proton": 58455,
+      "▁Sympt": 58456,
+      "▁Supplement": 58457,
+      "察觉": 58458,
+      "接力": 58459,
+      "日均": 58460,
+      "▁APC": 58461,
+      "▁DSL": 58462,
+      "大学生的": 58463,
+      "请你仔细读题": 58464,
+      "从选项中选择出": 58465,
+      "▁sailors": 58466,
+      "如上是一道选择题": 58467,
+      "ivariable": 58468,
+      "lef": 58469,
+      "▁`[": 58470,
+      "并确定": 58471,
+      "的路程": 58472,
+      "可以达到": 58473,
+      "判处有期徒刑六个月": 58474,
+      "MPI": 58475,
+      "由本院": 58476,
+      "SIGN": 58477,
+      "贯彻执行": 58478,
+      "▁Platinum": 58479,
+      "▁dispersed": 58480,
+      "▁inspector": 58481,
+      "▁transplanted": 58482,
+      "通告": 58483,
+      "BAR": 58484,
+      "读者的": 58485,
+      "atore": 58486,
+      "▁Latex": 58487,
+      "▁glial": 58488,
+      "▁mongo": 58489,
+      "具有重要意义": 58490,
+      "为社会": 58491,
+      "是指在": 58492,
+      "▁Ost": 58493,
+      "▁Tue": 58494,
+      "▁Marl": 58495,
+      "isArray": 58496,
+      "uclease": 58497,
+      "▁Huawei": 58498,
+      "▁nightmare": 58499,
+      "壁画": 58500,
+      "桃李": 58501,
+      "航海": 58502,
+      "和孩子": 58503,
+      "中心小学": 58504,
+      "tlist": 58505,
+      "▁THEN": 58506,
+      "▁popup": 58507,
+      "▁torture": 58508,
+      "particles": 58509,
+      "▁filenames": 58510,
+      "▁watershed": 58511,
+      "BW": 58512,
+      "免除": 58513,
+      "嫁给": 58514,
+      "窒息": 58515,
+      "血脉": 58516,
+      "relu": 58517,
+      "▁Todo": 58518,
+      "▁wedge": 58519,
+      "内容如下所示": 58520,
+      "民事执行中查封": 58521,
+      "ReadLine": 58522,
+      "▁bracelet": 58523,
+      "kea": 58524,
+      "▁Hew": 58525,
+      "nwork": 58526,
+      "▁cones": 58527,
+      "▁stagger": 58528,
+      "▁Northwestern": 58529,
+      "他有": 58530,
+      "日到": 58531,
+      "▁Gig": 58532,
+      "▁Eden": 58533,
+      "IsNullOr": 58534,
+      "Particle": 58535,
+      "Hb": 58536,
+      "庄严": 58537,
+      "政党": 58538,
+      "青梅": 58539,
+      "EUR": 58540,
+      "外面的": 58541,
+      "▁lifts": 58542,
+      "▁anaerobic": 58543,
+      "Lazy": 58544,
+      "▁для": 58545,
+      "▁dwarf": 58546,
+      "letcher": 58547,
+      "exercise": 58548,
+      "▁comeback": 58549,
+      "必有": 58550,
+      "等同": 58551,
+      "醒了": 58552,
+      "中途退": 58553,
+      "的女子": 58554,
+      "▁Equity": 58555,
+      "ressions": 58556,
+      "▁Improved": 58557,
+      "▁visceral": 58558,
+      "一连": 58559,
+      "免税": 58560,
+      "冲锋": 58561,
+      "妨害": 58562,
+      "涡轮": 58563,
+      "DAT": 58564,
+      "^*$": 58565,
+      "想象力": 58566,
+      "acian": 58567,
+      "▁Adaptive": 58568,
+      "▁Painting": 58569,
+      "▁Pharmaceutical": 58570,
+      "永生": 58571,
+      "二次元": 58572,
+      "带孩子": 58573,
+      "plug": 58574,
+      "▁Nin": 58575,
+      "truth": 58576,
+      "▁kernels": 58577,
+      "heon": 58578,
+      "ensory": 58579,
+      "▁Riley": 58580,
+      "▁drones": 58581,
+      "本次执行程序应予终结": 58582,
+      "▁pancreatitis": 58583,
+      "eba": 58584,
+      "ynch": 58585,
+      "进行研究": 58586,
+      "▁witch": 58587,
+      "Players": 58588,
+      "badges": 58589,
+      "▁Calif": 58590,
+      "▁conditioner": 58591,
+      "就让": 58592,
+      "身旁": 58593,
+      "MCs": 58594,
+      "俗话说": 58595,
+      "memo": 58596,
+      "nProm": 58597,
+      "经济发展的": 58598,
+      "▁clutch": 58599,
+      "▁onboard": 58600,
+      "或在": 58601,
+      "看向": 58602,
+      "raj": 58603,
+      "好莱坞": 58604,
+      "▁PII": 58605,
+      "nwell": 58606,
+      "troviral": 58607,
+      "▁Hampton": 58608,
+      "▁specials": 58609,
+      "圣经": 58610,
+      "太后": 58611,
+      "幼虫": 58612,
+      "数据线": 58613,
+      "显示出": 58614,
+      "\\\\}$$": 58615,
+      "freeze": 58616,
+      "▁oxidized": 58617,
+      "瘙痒": 58618,
+      "追回": 58619,
+      "飞行员": 58620,
+      "▁}}{": 58621,
+      "由审判员": 58622,
+      "liers": 58623,
+      "族自治州中级人民法院": 58624,
+      "哈萨克": 58625,
+      "空气中": 58626,
+      "▁webcam": 58627,
+      "▁diploma": 58628,
+      "▁试": 58629,
+      "俄国": 58630,
+      "的组合": 58631,
+      "主要内容": 58632,
+      "问卷调查": 58633,
+      "plasty": 58634,
+      "▁Merge": 58635,
+      "ennials": 58636,
+      "▁detects": 58637,
+      "劳务合同纠纷一案": 58638,
+      "人脸": 58639,
+      "soon": 58640,
+      "itating": 58641,
+      "礼拜": 58642,
+      "Mur": 58643,
+      "就够了": 58644,
+      "▁prince": 58645,
+      "▁critique": 58646,
+      "监制": 58647,
+      "做好了": 58648,
+      "▁extrav": 58649,
+      "▁Reporter": 58650,
+      "▁contractile": 58651,
+      "▁unexpectedly": 58652,
+      "鼻炎": 58653,
+      "我们有": 58654,
+      "▁dei": 58655,
+      "scanner": 58656,
+      "▁absurd": 58657,
+      "▁inserts": 58658,
+      "鸟类": 58659,
+      "bulk": 58660,
+      "▁Slot": 58661,
+      "▁heur": 58662,
+      "可供您选择": 58663,
+      "ouched": 58664,
+      "secrets": 58665,
+      "Preference": 58666,
+      "CompatActivity": 58667,
+      "红茶": 58668,
+      "青椒": 58669,
+      "▁{%": 58670,
+      "中国画": 58671,
+      "是这个": 58672,
+      "的优点": 58673,
+      "stay": 58674,
+      "荆州": 58675,
+      "nUN": 58676,
+      "桂林市": 58677,
+      "▁Cly": 58678,
+      "▁Funk": 58679,
+      "▁radios": 58680,
+      "▁tractor": 58681,
+      "startsWith": 58682,
+      "▁forbidden": 58683,
+      "香料": 58684,
+      "医疗卫生": 58685,
+      "然后告诉我": 58686,
+      "请告诉我您的选择": 58687,
+      "confirmed": 58688,
+      "日向本院提出财产保全": 58689,
+      "▁narratives": 58690,
+      "请仔细阅读下面这道选择题": 58691,
+      "文旅": 58692,
+      "中全会": 58693,
+      "的课程": 58694,
+      "某某负担": 58695,
+      "▁Salem": 58696,
+      "▁stigma": 58697,
+      "▁phosphorylated": 58698,
+      "瓷器": 58699,
+      "长子": 58700,
+      "Sky": 58701,
+      "高精度": 58702,
+      "▁aur": 58703,
+      "▁lord": 58704,
+      "Deploy": 58705,
+      "cardia": 58706,
+      "古装": 58707,
+      "备战": 58708,
+      "武警": 58709,
+      "金字": 58710,
+      "▁号码": 58711,
+      "去年的": 58712,
+      "▁Twelve": 58713,
+      "▁sprite": 58714,
+      "中华人民共和国土地管理法": 58715,
+      "追捧": 58716,
+      "速度和": 58717,
+      "▁Jag": 58718,
+      "Backup": 58719,
+      "▁Betty": 58720,
+      "知识渊博的人": 58721,
+      "▁Ballet": 58722,
+      "等诸多": 58723,
+      "组织中": 58724,
+      "ilin": 58725,
+      "更多信息": 58726,
+      "注册资金": 58727,
+      "▁heir": 58728,
+      "▁perv": 58729,
+      "▁payday": 58730,
+      "园艺": 58731,
+      "izi": 58732,
+      "多万元": 58733,
+      "添加了": 58734,
+      "您认为的": 58735,
+      "▁占地面积": 58736,
+      "人民卫生出版社": 58737,
+      "▁anticipation": 58738,
+      "NX": 58739,
+      "读写": 58740,
+      "EPS": 58741,
+      "应依法": 58742,
+      "Loss": 58743,
+      "继续进行": 58744,
+      "ACTER": 58745,
+      "▁plur": 58746,
+      "▁Lopez": 58747,
+      "relevant": 58748,
+      "▁honesty": 58749,
+      "▁individualized": 58750,
+      "白领": 58751,
+      "nKen": 58752,
+      "请用您的": 58753,
+      "arbox": 58754,
+      "▁Mens": 58755,
+      "▁Equal": 58756,
+      "▁维": 58757,
+      "Never": 58758,
+      "请您务必注意": 58759,
+      "在回答以下这道": 58760,
+      "问题和选项的内容": 58761,
+      "▁worksheet": 58762,
+      "从选项中选定正确答案": 58763,
+      "具有挑战性的选择题时": 58764,
+      "▁astrocytes": 58765,
+      "请您务必注意问题和选项的内容": 58766,
+      "COPY": 58767,
+      "ponge": 58768,
+      "Secure": 58769,
+      "▁ozone": 58770,
+      "▁Lounge": 58771,
+      "%%%%%%%%": 58772,
+      "▁Davidson": 58773,
+      "▁inefficient": 58774,
+      "交织": 58775,
+      "相比于": 58776,
+      "醉酒后": 58777,
+      "▁Bali": 58778,
+      "brahim": 58779,
+      "工贸有限公司": 58780,
+      "▁Debian": 58781,
+      "▁batting": 58782,
+      "审慎考虑您的选择": 58783,
+      "并告诉我您认为的": 58784,
+      "请用您的专业知识": 58785,
+      "nEveryone": 58786,
+      "▁workouts": 58787,
+      "▁summarizes": 58788,
+      "以下备选答案可供您选择": 58789,
+      "请用您的专业知识回答以下问题": 58790,
+      "南阳市": 58791,
+      "以下命令": 58792,
+      "▁laughter": 58793,
+      "▁impressions": 58794,
+      "约占": 58795,
+      "理由正当": 58796,
+      "dorff": 58797,
+      "guess": 58798,
+      "▁avid": 58799,
+      "▁Wesley": 58800,
+      "currentTime": 58801,
+      "混淆": 58802,
+      "迷人": 58803,
+      "▁XY": 58804,
+      "开放式": 58805,
+      "指出了": 58806,
+      "oxid": 58807,
+      "likes": 58808,
+      "▁Chal": 58809,
+      "▁VERY": 58810,
+      "▁ferm": 58811,
+      "nAdmin": 58812,
+      "▁sarcoma": 58813,
+      "▁advertise": 58814,
+      "长短": 58815,
+      "领略": 58816,
+      "灵敏度": 58817,
+      "▁Seth": 58818,
+      "openssl": 58819,
+      "XXXXXXXX": 58820,
+      "ellaneous": 58821,
+      "季后": 58822,
+      "解散": 58823,
+      "▁Lt": 58824,
+      "的一段": 58825,
+      "起诉必须": 58826,
+      "请选择你认为": 58827,
+      "foundland": 58828,
+      "ду": 58829,
+      "不归": 58830,
+      "此地": 58831,
+      "专升本": 58832,
+      "如果你想": 58833,
+      "请选择你认为合适的答案": 58834,
+      "人权": 58835,
+      "丫头": 58836,
+      "丽莎": 58837,
+      "营利": 58838,
+      "转子": 58839,
+      "都喜欢": 58840,
+      "森林公园": 58841,
+      "▁Jagu": 58842,
+      "puppet": 58843,
+      "|[": 58844,
+      "LEN": 58845,
+      "fluent": 58846,
+      "▁pioneering": 58847,
+      "心率": 58848,
+      "OTH": 58849,
+      "▁{/": 58850,
+      "吸血鬼": 58851,
+      "请审查": 58852,
+      "近距离": 58853,
+      "严格执行": 58854,
+      "▁Thomson": 58855,
+      "astructural": 58856,
+      "▁filesystem": 58857,
+      "▁centimeters": 58858,
+      "准时": 58859,
+      "坏人": 58860,
+      "移交": 58861,
+      "……[": 58862,
+      "fine": 58863,
+      "▁hoc": 58864,
+      "nenum": 58865,
+      "urtles": 58866,
+      "实际应用领域": 58867,
+      "▁rented": 58868,
+      "他俩": 58869,
+      "录入": 58870,
+      "面目": 58871,
+      "cors": 58872,
+      "▁Byz": 58873,
+      "▁blu": 58874,
+      "发展方向": 58875,
+      "人民检察院以": 58876,
+      "▁doping": 58877,
+      "▁petrol": 58878,
+      "最高人民法院关于执行案件立案": 58879,
+      "发给": 58880,
+      "失控": 58881,
+      "次于": 58882,
+      "起初": 58883,
+      "▁开发商": 58884,
+      "第十三章": 58885,
+      "选择是什么": 58886,
+      "▁Floyd": 58887,
+      "getType": 58888,
+      "▁cracked": 58889,
+      "▁contrasting": 58890,
+      "参见": 58891,
+      "博物院": 58892,
+      "在生活中": 58893,
+      "▁Joyce": 58894,
+      "▁straps": 58895,
+      "▁disadvantage": 58896,
+      "刷牙": 58897,
+      "在同": 58898,
+      "提取物": 58899,
+      "nIce": 58900,
+      "随着我国": 58901,
+      "▁tidal": 58902,
+      "Dropdown": 58903,
+      "Extended": 58904,
+      "▁obsessed": 58905,
+      "▁AttributeError": 58906,
+      "随处": 58907,
+      "非常好的": 58908,
+      "▁stabilize": 58909,
+      "凌云": 58910,
+      "环形": 58911,
+      "竹林": 58912,
+      "乐意": 58913,
+      "Own": 58914,
+      "Drawer": 58915,
+      "stance": 58916,
+      "{{'}}(": 58917,
+      "▁cheek": 58918,
+      "contained": 58919,
+      "▁sandwiches": 58920,
+      "他将": 58921,
+      "口中": 58922,
+      "引言": 58923,
+      "水务": 58924,
+      "MPa": 58925,
+      "乘进率": 58926,
+      "▁Rise": 58927,
+      "itorium": 58928,
+      "请审查以下问题": 58929,
+      "▁elusive": 58930,
+      "▁ensured": 58931,
+      "▁Stability": 58932,
+      "作为一名知识渊博的人": 58933,
+      "OwnProperty": 58934,
+      "南岸": 58935,
+      "垫付": 58936,
+      "烟雾": 58937,
+      "智能家居": 58938,
+      "andan": 58939,
+      "effer": 58940,
+      "▁hips": 58941,
+      "OnInit": 58942,
+      "工业和信息化": 58943,
+      "▁amazon": 58944,
+      "人流": 58945,
+      "ebo": 58946,
+      "▁Ada": 58947,
+      "▁Sham": 58948,
+      "Parallel": 58949,
+      "▁mornings": 58950,
+      "恐慌": 58951,
+      "nAg": 58952,
+      "▁ct": 58953,
+      "有一条": 58954,
+      "cels": 58955,
+      "只是为了": 58956,
+      "▁Alleg": 58957,
+      "▁Scandin": 58958,
+      "前瞻": 58959,
+      "房地": 58960,
+      "雨后": 58961,
+      "tto": 58962,
+      "nHol": 58963,
+      "得到结果": 58964,
+      "选项涉及到": 58965,
+      "▁radiographic": 58966,
+      "推翻": 58967,
+      "人口与": 58968,
+      "和个人": 58969,
+      "eppe": 58970,
+      "与对照组": 58971,
+      "▁readiness": 58972,
+      "延边": 58973,
+      "疤痕": 58974,
+      "以下的问题和": 58975,
+      "审阅这道选择题": 58976,
+      "terraform": 58977,
+      "以下的问题和选项涉及到": 58978,
+      "并从给出的选项中选定正确答案": 58979,
+      "交响": 58980,
+      "▁pu": 58981,
+      "的最高": 58982,
+      "rollup": 58983,
+      "▁Fluid": 58984,
+      "mathscr": 58985,
+      "▁Singer": 58986,
+      "'_": 58987,
+      "▁廖": 58988,
+      "日为": 58989,
+      "瓦斯": 58990,
+      "跑去": 58991,
+      "迷信": 58992,
+      "特别好": 58993,
+      "▁accents": 58994,
+      "▁intimid": 58995,
+      "共分": 58996,
+      "有空": 58997,
+      "诅咒": 58998,
+      "你不能": 58999,
+      "不良影响": 59000,
+      "Strict": 59001,
+      "▁centrifug": 59002,
+      "▁digitally": 59003,
+      "征地": 59004,
+      "飞扬": 59005,
+      "在有限": 59006,
+      "▁Gap": 59007,
+      "希望通过": 59008,
+      "▁nombre": 59009,
+      "▁coincide": 59010,
+      "塔尔": 59011,
+      "事业发展": 59012,
+      "krainian": 59013,
+      "▁noninvasive": 59014,
+      "切的": 59015,
+      "dream": 59016,
+      "▁headphones": 59017,
+      "▁renin": 59018,
+      "GameObject": 59019,
+      "▁advertisers": 59020,
+      "国有资产": 59021,
+      "▁evangel": 59022,
+      "▁Madagascar": 59023,
+      "▁authenticity": 59024,
+      "库里": 59025,
+      "角质": 59026,
+      "饭后": 59027,
+      "atu": 59028,
+      "这玩意": 59029,
+      "▁Arms": 59030,
+      "▁Shan": 59031,
+      "▁neurologic": 59032,
+      "ErrorMessage": 59033,
+      "Son": 59034,
+      "吃起来": 59035,
+      "植物门": 59036,
+      "的现状": 59037,
+      "smtp": 59038,
+      "积极开展": 59039,
+      "associ": 59040,
+      "mental": 59041,
+      "▁alkyl": 59042,
+      "▁brides": 59043,
+      "▁surplus": 59044,
+      "排污": 59045,
+      "松下": 59046,
+      "隐蔽": 59047,
+      "lips": 59048,
+      "▁justification": 59049,
+      "出众": 59050,
+      "液化": 59051,
+      "nGO": 59052,
+      "圆的半径": 59053,
+      "学生的学习": 59054,
+      "▁annex": 59055,
+      "▁geared": 59056,
+      "zs": 59057,
+      "▁邹": 59058,
+      "几分": 59059,
+      "鸡翅": 59060,
+      "CIS": 59061,
+      "Uns": 59062,
+      "tgo": 59063,
+      "问题上": 59064,
+      "▁Mys": 59065,
+      "推广应用": 59066,
+      "ossip": 59067,
+      "▁Educ": 59068,
+      "▁Johann": 59069,
+      "iteration": 59070,
+      "wf": 59071,
+      "中层": 59072,
+      "打压": 59073,
+      "大大的": 59074,
+      "enko": 59075,
+      "asInstanceOf": 59076,
+      "▁∠": 59077,
+      "tmax": 59078,
+      "▁Mam": 59079,
+      "▁CODE": 59080,
+      "▁tweak": 59081,
+      "Samples": 59082,
+      "inspect": 59083,
+      "▁Browns": 59084,
+      "▁facade": 59085,
+      "▁perennial": 59086,
+      "▁unconscious": 59087,
+      "张伟": 59088,
+      "民意": 59089,
+      "它本身": 59090,
+      "{}\\'": 59091,
+      "▁第一章": 59092,
+      ")^{\\\\": 59093,
+      "stadt": 59094,
+      "▁Recommended": 59095,
+      "履职": 59096,
+      "拓片": 59097,
+      "活在": 59098,
+      "Squ": 59099,
+      "Prod": 59100,
+      "▁Aval": 59101,
+      "你都": 59102,
+      "BLE": 59103,
+      "等式的": 59104,
+      "angs": 59105,
+      "enet": 59106,
+      "▁avg": 59107,
+      "印度尼西亚": 59108,
+      "▁simplex": 59109,
+      "调配": 59110,
+      "最好不要": 59111,
+      "▁Metric": 59112,
+      "▁Naples": 59113,
+      "▁disturbed": 59114,
+      "▁touchdowns": 59115,
+      "远处": 59116,
+      "雨花": 59117,
+      "HDL": 59118,
+      "Kin": 59119,
+      "平均每天": 59120,
+      "builtin": 59121,
+      "▁rewarded": 59122,
+      "$:\\": 59123,
+      "的记忆": 59124,
+      "的诊断": 59125,
+      "▁Ley": 59126,
+      "▁Jenny": 59127,
+      "其中农业人口": 59128,
+      "▁uneven": 59129,
+      "▁Prevent": 59130,
+      "▁nonsense": 59131,
+      "▁partitioning": 59132,
+      "▁Transformation": 59133,
+      "不买": 59134,
+      "留住": 59135,
+      "▁CONF": 59136,
+      "▁esta": 59137,
+      "▁mods": 59138,
+      "▁Cardi": 59139,
+      "▁proceeding": 59140,
+      "▁disappearance": 59141,
+      "嘴角": 59142,
+      "王建": 59143,
+      "迎合": 59144,
+      "sov": 59145,
+      "可以进行": 59146,
+      "▁lista": 59147,
+      "海贼": 59148,
+      "lbl": 59149,
+      "▁Ard": 59150,
+      "城市管理": 59151,
+      "是错误的": 59152,
+      "越来越高": 59153,
+      "且申请执行人": 59154,
+      "▁oracle": 59155,
+      "▁Activation": 59156,
+      "下山": 59157,
+      "人居": 59158,
+      "穷人": 59159,
+      "George": 59160,
+      "▁Noble": 59161,
+      "▁Hollow": 59162,
+      "▁reticulum": 59163,
+      "白斑": 59164,
+      "▁Sag": 59165,
+      "他把": 59166,
+      "啦啦": 59167,
+      "木马": 59168,
+      "案款": 59169,
+      "试管": 59170,
+      "驾照": 59171,
+      "转换成": 59172,
+      "esus": 59173,
+      "structuring": 59174,
+      "入坑": 59175,
+      "演义": 59176,
+      "的类型": 59177,
+      "根据原告": 59178,
+      "研究报告": 59179,
+      "asuring": 59180,
+      "NotEmpty": 59181,
+      "▁migraine": 59182,
+      "依法须经批准的项目": 59183,
+      "|}": 59184,
+      "创造出": 59185,
+      "病患者": 59186,
+      "nche": 59187,
+      "▁ECM": 59188,
+      "urous": 59189,
+      "nunder": 59190,
+      "计算器": 59191,
+      "中国文化": 59192,
+      "通知如下": 59193,
+      "fiber": 59194,
+      "在本次执行程序中": 59195,
+      "▁Accounts": 59196,
+      "▁ganglion": 59197,
+      "▁Interstate": 59198,
+      "两款": 59199,
+      "这组": 59200,
+      "▁代表": 59201,
+      "依赖于": 59202,
+      "▁Colour": 59203,
+      "▁Startup": 59204,
+      "生效之日起十日内": 59205,
+      "▁Chromium": 59206,
+      "interaction": 59207,
+      "围观": 59208,
+      "得罪": 59209,
+      "omod": 59210,
+      "▁Hers": 59211,
+      "ifolia": 59212,
+      "dimensions": 59213,
+      "spawn": 59214,
+      "不甘": 59215,
+      "汉书": 59216,
+      "语境": 59217,
+      "InB": 59218,
+      "了一年": 59219,
+      "小数的": 59220,
+      "▁iris": 59221,
+      "▁游戏平台": 59222,
+      "司法鉴定中心": 59223,
+      "uncredited": 59224,
+      "东湖": 59225,
+      "玄武": 59226,
+      "这一步": 59227,
+      "投资或控股": 59228,
+      "申请执行的": 59229,
+      "nconfig": 59230,
+      "▁borrowed": 59231,
+      "▁trustworthy": 59232,
+      "精油": 59233,
+      "▁Glacier": 59234,
+      "▁firstly": 59235,
+      "回转": 59236,
+      "稿件": 59237,
+      "切除术": 59238,
+      "aryl": 59239,
+      "▁buds": 59240,
+      "您的最佳选择": 59241,
+      "送达了执行通知书": 59242,
+      "▁interiors": 59243,
+      "他没有": 59244,
+      "又包含了": 59245,
+      "nalso": 59246,
+      "▁sucks": 59247,
+      "自然人投资或控股": 59248,
+      "工况": 59249,
+      "nLow": 59250,
+      "clang": 59251,
+      "▁pops": 59252,
+      "▁bricks": 59253,
+      "▁dilution": 59254,
+      "Vi": 59255,
+      "сь": 59256,
+      "一章": 59257,
+      "粮油": 59258,
+      "限时": 59259,
+      "全明星": 59260,
+      "成功地": 59261,
+      "社会组织": 59262,
+      "Skill": 59263,
+      "loxacin": 59264,
+      "▁inevitably": 59265,
+      "▁serialized": 59266,
+      "小微": 59267,
+      "水电站": 59268,
+      "之间存在": 59269,
+      "▁Lands": 59270,
+      "▁Rooms": 59271,
+      "▁Tuple": 59272,
+      "▁Jacobs": 59273,
+      "▁gallons": 59274,
+      "▁warranted": 59275,
+      "话剧": 59276,
+      "看不懂": 59277,
+      "nCity": 59278,
+      "科学研究院": 59279,
+      "▁Gibbs": 59280,
+      "knowledge": 59281,
+      "低端": 59282,
+      "▁LR": 59283,
+      "江门市": 59284,
+      "一次函数": 59285,
+      "▁Roma": 59286,
+      "▁quir": 59287,
+      "▁liquor": 59288,
+      "ethasone": 59289,
+      "verified": 59290,
+      "▁gadgets": 59291,
+      "▁translating": 59292,
+      "戴着": 59293,
+      "细微": 59294,
+      "noid": 59295,
+      "▁COX": 59296,
+      "既包含了": 59297,
+      "▁Ling": 59298,
+      "▁edits": 59299,
+      "parametric": 59300,
+      "失落": 59301,
+      "点头": 59302,
+      "准确地": 59303,
+      "要根据": 59304,
+      "发布会上": 59305,
+      "anchez": 59306,
+      "▁Pythag": 59307,
+      "▁evapor": 59308,
+      "▁Coleman": 59309,
+      "▁positivity": 59310,
+      "隐隐": 59311,
+      "再进行": 59312,
+      "政策根据": 59313,
+      "extras": 59314,
+      "▁replicated": 59315,
+      "不还": 59316,
+      "保质": 59317,
+      "辨证": 59318,
+      "▁调料": 59319,
+      "iast": 59320,
+      "▁Lua": 59321,
+      "行政管理": 59322,
+      "请仔细浏览": 59323,
+      "并给出正确答案": 59324,
+      "古巴": 59325,
+      "缠绕": 59326,
+      "位数是": 59327,
+      "县委常委": 59328,
+      "在面对以下": 59329,
+      "请您认真分析": 59330,
+      "▁Decimal": 59331,
+      "并告诉我您的选择": 59332,
+      "在面对以下选择题时": 59333,
+      "▁economist": 59334,
+      "请仔细浏览以下备选答案": 59335,
+      "请您认真分析并给出正确答案": 59336,
+      "假冒": 59337,
+      "全班": 59338,
+      "来院": 59339,
+      "金黄色": 59340,
+      "主要种植": 59341,
+      "住上海市": 59342,
+      "flask": 59343,
+      "Richard": 59344,
+      "▁Beverly": 59345,
+      "installation": 59346,
+      "Notifications": 59347,
+      "此生": 59348,
+      "翼翼": 59349,
+      "CoV": 59350,
+      "成交量": 59351,
+      "▁Sta": 59352,
+      "如何使用": 59353,
+      "▁DOWN": 59354,
+      "▁Tested": 59355,
+      "Drawable": 59356,
+      "ovoltaic": 59357,
+      "渣男": 59358,
+      "英勇": 59359,
+      "物质的": 59360,
+      "Aware": 59361,
+      "▁duplex": 59362,
+      "▁firearms": 59363,
+      "▁Landscape": 59364,
+      "▁disappointment": 59365,
+      "他却": 59366,
+      "汲取": 59367,
+      "bob": 59368,
+      "▁Nec": 59369,
+      "后可": 59370,
+      "奢侈品": 59371,
+      "▁eps": 59372,
+      "不断创新": 59373,
+      "▁Blank": 59374,
+      "▁smarter": 59375,
+      "劳务费": 59376,
+      "▁Borg": 59377,
+      "▁Univ": 59378,
+      "pering": 59379,
+      "①②": 59380,
+      "区位": 59381,
+      "纯洁": 59382,
+      "诗集": 59383,
+      "ACA": 59384,
+      "worm": 59385,
+      "▁PCs": 59386,
+      "可以裁定": 59387,
+      "▁cysts": 59388,
+      "▁spoil": 59389,
+      "点亮": 59390,
+      "保持了": 59391,
+      "Ijoi": 59392,
+      "adan": 59393,
+      "urger": 59394,
+      "▁Mare": 59395,
+      "法民初字第": 59396,
+      "currently": 59397,
+      "▁customizable": 59398,
+      "▁AUTO": 59399,
+      "▁plat": 59400,
+      "▁EVERY": 59401,
+      "房地产开发有限责任公司": 59402,
+      "但债权的实现取决于被执行人是否有": 59403,
+      "▁'_": 59404,
+      "就是我": 59405,
+      "并表示": 59406,
+      "在校学生": 59407,
+      "otron": 59408,
+      "aleigh": 59409,
+      "▁conserve": 59410,
+      "▁overhaul": 59411,
+      "paralleled": 59412,
+      "最先": 59413,
+      "%);": 59414,
+      "nFeb": 59415,
+      "使用的是": 59416,
+      "▁mega": 59417,
+      "nAndrew": 59418,
+      "▁Dreams": 59419,
+      "ある": 59420,
+      "盛会": 59421,
+      "nKn": 59422,
+      "不方便": 59423,
+      "obox": 59424,
+      "▁playful": 59425,
+      "prediction": 59426,
+      "▁duplicates": 59427,
+      "之声": 59428,
+      "京城": 59429,
+      "插图": 59430,
+      "Ack": 59431,
+      "眼中的": 59432,
+      "nAust": 59433,
+      "nCamp": 59434,
+      "Stable": 59435,
+      "propag": 59436,
+      "▁shuffle": 59437,
+      "▁smiling": 59438,
+      "▁respiration": 59439,
+      "▁unsubscribe": 59440,
+      "上进": 59441,
+      "很有可能": 59442,
+      "我是一个": 59443,
+      "ispens": 59444,
+      "▁fulfil": 59445,
+      "号牌": 59446,
+      "诗文": 59447,
+      "_{(": 59448,
+      "▁Lig": 59449,
+      "企业形象": 59450,
+      "执行程序": 59451,
+      "▁sued": 59452,
+      "生产建设兵团": 59453,
+      "nGetting": 59454,
+      "▁Halifax": 59455,
+      "中进行": 59456,
+      "民事责任": 59457,
+      "▁auch": 59458,
+      "▁Demand": 59459,
+      "▁HashSet": 59460,
+      "包厢": 59461,
+      "方式和": 59462,
+      "看不起": 59463,
+      "▁Neut": 59464,
+      "▁blur": 59465,
+      "▁Kernel": 59466,
+      "▁要": 59467,
+      "手足": 59468,
+      "跳过": 59469,
+      "进场": 59470,
+      "whit": 59471,
+      "intel": 59472,
+      "著作权集体": 59473,
+      "▁Ferrari": 59474,
+      "▁thereof": 59475,
+      "▁perinatal": 59476,
+      "五六": 59477,
+      "大象": 59478,
+      "选址": 59479,
+      "▁ops": 59480,
+      "nSimple": 59481,
+      "▁Minecraft": 59482,
+      "<>": 59483,
+      "族群": 59484,
+      "ymn": 59485,
+      "为重点": 59486,
+      "动画片": 59487,
+      "安装了": 59488,
+      "现代农业": 59489,
+      "▁Pitch": 59490,
+      "▁pilgr": 59491,
+      "▁Edmund": 59492,
+      "▁comprehend": 59493,
+      "可由": 59494,
+      "小哥": 59495,
+      "废气": 59496,
+      "想吃": 59497,
+      "扶养": 59498,
+      "▁Kum": 59499,
+      "drawable": 59500,
+      "canonical": 59501,
+      "▁incision": 59502,
+      "科学记数法的表示形式为": 59503,
+      "与李": 59504,
+      "在各": 59505,
+      "担负": 59506,
+      "为主料的": 59507,
+      "上面这个问题": 59508,
+      "财运": 59509,
+      "Joe": 59510,
+      "gkin": 59511,
+      "若干选项": 59512,
+      "主要负责人": 59513,
+      "的其他财产": 59514,
+      "▁polymerization": 59515,
+      "宋朝": 59516,
+      "巡回": 59517,
+      "捆绑": 59518,
+      "邮局": 59519,
+      "osm": 59520,
+      "坚定的": 59521,
+      "▁LAN": 59522,
+      "开户银行": 59523,
+      "职业道德": 59524,
+      "hesus": 59525,
+      "ptical": 59526,
+      "shader": 59527,
+      "terase": 59528,
+      "▁Aeros": 59529,
+      "attices": 59530,
+      "kip": 59531,
+      "一直都是": 59532,
+      "然后点击": 59533,
+      "计算机网络": 59534,
+      "这是一道需": 59535,
+      "分析这些选项": 59536,
+      "综合运用多方面": 59537,
+      "▁Parking": 59538,
+      "▁plotted": 59539,
+      "▁brighter": 59540,
+      "▁diplomat": 59541,
+      "请仔细分析这些选项": 59542,
+      "然后告诉我您的最佳选择": 59543,
+      "这是一道需综合运用多方面": 59544,
+      "thi": 59545,
+      "保持在": 59546,
+      "SATA": 59547,
+      "Daily": 59548,
+      "内容判断出": 59549,
+      "包含一个问题和": 59550,
+      "内容判断出您认为": 59551,
+      "以下内容包含一个问题和": 59552,
+      "▁Pseudomonas": 59553,
+      "依据问题内容判断出您认为": 59554,
+      "以下内容包含一个问题和若干选项": 59555,
+      "前夕": 59556,
+      "感性": 59557,
+      "这儿": 59558,
+      "Ell": 59559,
+      "法律的": 59560,
+      "魔法师": 59561,
+      "orax": 59562,
+      "▁AFL": 59563,
+      "▁Agu": 59564,
+      "▁Voy": 59565,
+      "fills": 59566,
+      "▁Maint": 59567,
+      "shortcuts": 59568,
+      "▁Wik": 59569,
+      "UMENT": 59570,
+      "initis": 59571,
+      "lectic": 59572,
+      "▁Alexa": 59573,
+      "▁empathy": 59574,
+      "▁Estonian": 59575,
+      ",{": 59576,
+      "▁AML": 59577,
+      "周边配套": 59578,
+      "▁Dash": 59579,
+      "turned": 59580,
+      "▁fb": 59581,
+      "erializer": 59582,
+      "▁recreate": 59583,
+      "毛细": 59584,
+      "短片": 59585,
+      "csc": 59586,
+      "▁Hutch": 59587,
+      "▁Chapman": 59588,
+      "▁Nonetheless": 59589,
+      "县的": 59590,
+      "拳头": 59591,
+      "JDK": 59592,
+      "交流与": 59593,
+      "这个地图": 59594,
+      "▁Helm": 59595,
+      "sphere": 59596,
+      "▁presumed": 59597,
+      "依依": 59598,
+      "沙龙": 59599,
+      "营商": 59600,
+      "仍不能": 59601,
+      "所用的": 59602,
+      "TEMP": 59603,
+      "▁ABOUT": 59604,
+      "utively": 59605,
+      "▁Bollywood": 59606,
+      "▁Settlement": 59607,
+      "▁documenting": 59608,
+      "▁fashionable": 59609,
+      "二层": 59610,
+      "疏散": 59611,
+      "的还是": 59612,
+      "nAmb": 59613,
+      "oblastic": 59614,
+      "▁vaccinated": 59615,
+      "烤肉": 59616,
+      "热衷": 59617,
+      "为他们": 59618,
+      "审判庭": 59619,
+      "与被上诉人": 59620,
+      "accessor": 59621,
+      "▁Standing": 59622,
+      "▁compassionate": 59623,
+      "机动车交通事故责任纠纷一案中": 59624,
+      "无言": 59625,
+      "猕猴": 59626,
+      "着作": 59627,
+      "得出答案": 59628,
+      "龙头企业": 59629,
+      "isdir": 59630,
+      "onium": 59631,
+      "给定的问题": 59632,
+      "▁superc": 59633,
+      "▁nonatomic": 59634,
+      "毛衣": 59635,
+      "的快": 59636,
+      "皮下": 59637,
+      "ILY": 59638,
+      "▁Quad": 59639,
+      "▁Cookie": 59640,
+      "▁climbed": 59641,
+      "领会": 59642,
+      "tctx": 59643,
+      "njobs": 59644,
+      "ehicles": 59645,
+      "largest": 59646,
+      "▁paints": 59647,
+      "|_": 59648,
+      "超额": 59649,
+      "在国外": 59650,
+      "给定问题": 59651,
+      "▁Hundred": 59652,
+      "▁jsonify": 59653,
+      "▁Apartment": 59654,
+      "Dy": 59655,
+      "东市": 59656,
+      "孔雀": 59657,
+      "无毒": 59658,
+      "Pag": 59659,
+      "otto": 59660,
+      "这个行业": 59661,
+      "sequent": 59662,
+      "Ly": 59663,
+      "嘴唇": 59664,
+      "GRE": 59665,
+      "付出的": 59666,
+      "▁Plat": 59667,
+      "现在我会给你": 59668,
+      "依据给定的问题": 59669,
+      "ApiModel": 59670,
+      "仔细阅读上面这个问题": 59671,
+      "现在我会给你几个选项": 59672,
+      "请你根据问题进行分析": 59673,
+      "从选项中选出合适的答案": 59674,
+      "揭晓": 59675,
+      "新书": 59676,
+      "比方": 59677,
+      "▁Comm": 59678,
+      "上面这段话": 59679,
+      "▁theology": 59680,
+      "Collectors": 59681,
+      "nSimilarly": 59682,
+      "attachments": 59683,
+      "▁credential": 59684,
+      "▁φ": 59685,
+      "双面": 59686,
+      "eus": 59687,
+      "assuming": 59688,
+      "atterson": 59689,
+      "'}\\": 59690,
+      "emap": 59691,
+      "monds": 59692,
+      "▁obey": 59693,
+      "DEVICE": 59694,
+      "Ireland": 59695,
+      "▁Finish": 59696,
+      "▁PEOPLE": 59697,
+      "▁saliva": 59698,
+      "▁Dragons": 59699,
+      "已发生法律效力的": 59700,
+      "▁insufficiency": 59701,
+      "与传统": 59702,
+      "▁cataract": 59703,
+      "▁Distinguished": 59704,
+      "Nb": 59705,
+      "▁星": 59706,
+      "深海": 59707,
+      "使用方法": 59708,
+      "Versions": 59709,
+      "▁immobil": 59710,
+      "依法适用普通程序": 59711,
+      "操场": 59712,
+      "有声": 59713,
+      "一分钟": 59714,
+      "至少要": 59715,
+      "是否可以": 59716,
+      "inium": 59717,
+      "▁pleasing": 59718,
+      "▁assistants": 59719,
+      "二三": 59720,
+      "奋力": 59721,
+      "jQuery": 59722,
+      "▁pests": 59723,
+      "\\\\\":\\\\\"": 59724,
+      "▁depicting": 59725,
+      "▁来": 59726,
+      "近些": 59727,
+      "逃生": 59728,
+      "feng": 59729,
+      "▁caud": 59730,
+      "opathology": 59731,
+      "▁Recording": 59732,
+      "▁Geographic": 59733,
+      "▁interfering": 59734,
+      "刻录": 59735,
+      "kas": 59736,
+      "nApr": 59737,
+      "▁Disk": 59738,
+      "icators": 59739,
+      "supervised": 59740,
+      "▁Riverside": 59741,
+      "建工": 59742,
+      "显存容量": 59743,
+      "removed": 59744,
+      "▁Efficient": 59745,
+      "波长": 59746,
+      "科创": 59747,
+      "红颜": 59748,
+      "贮藏": 59749,
+      "Major": 59750,
+      "▁maze": 59751,
+      "▁prosecutor": 59752,
+      "[]{": 59753,
+      "▁moi": 59754,
+      "期限届满前": 59755,
+      "nViews": 59756,
+      "esarean": 59757,
+      "▁Deploy": 59758,
+      "possibly": 59759,
+      "亦是": 59760,
+      "低迷": 59761,
+      "写道": 59762,
+      "加尔": 59763,
+      "筹划": 59764,
+      "▁创办": 59765,
+      "alist": 59766,
+      "▁Donna": 59767,
+      "handled": 59768,
+      "▁smoked": 59769,
+      "▁granules": 59770,
+      "▁integrin": 59771,
+      "oooo": 59772,
+      "Retry": 59773,
+      "▁Nath": 59774,
+      "申请执行费": 59775,
+      "出具借条一份": 59776,
+      "▁revival": 59777,
+      "▁tc": 59778,
+      "ngoing": 59779,
+      "▁Residential": 59780,
+      "小憩": 59781,
+      "都用": 59782,
+      "第一季": 59783,
+      "ixir": 59784,
+      "Scott": 59785,
+      "▁transitional": 59786,
+      "九天": 59787,
+      "欺诈": 59788,
+      "账单": 59789,
+      "这样就": 59790,
+      "roat": 59791,
+      "olphin": 59792,
+      "webapp": 59793,
+      "▁Complexity": 59794,
+      "▁inherently": 59795,
+      "▁carbohydrates": 59796,
+      "kV": 59797,
+      "校外": 59798,
+      "牙膏": 59799,
+      "配角": 59800,
+      "决策和": 59801,
+      "cars": 59802,
+      "从而达到": 59803,
+      "▁newcom": 59804,
+      "▁generalize": 59805,
+      "二胎": 59806,
+      "对战": 59807,
+      "数的定义": 59808,
+      "社区矫正": 59809,
+      "*\\\\*\\\\": 59810,
+      "resume": 59811,
+      "shutil": 59812,
+      "▁HTTPS": 59813,
+      "▁blending": 59814,
+      "wed": 59815,
+      "以达到": 59816,
+      "Slug": 59817,
+      "▁acupuncture": 59818,
+      "生产能力": 59819,
+      "▁Vall": 59820,
+      "▁Calvin": 59821,
+      "▁honours": 59822,
+      "▁referee": 59823,
+      "▁tactical": 59824,
+      "丽江": 59825,
+      "变得更": 59826,
+      "在现场": 59827,
+      "老干部": 59828,
+      "Ruby": 59829,
+      "新冠病毒": 59830,
+      "cargo": 59831,
+      "▁Giovanni": 59832,
+      "▁incompatible": 59833,
+      "upp": 59834,
+      "加起来": 59835,
+      "时间段": 59836,
+      "高级的": 59837,
+      "ucci": 59838,
+      "▁Rex": 59839,
+      "▁Yuk": 59840,
+      "执行到位": 59841,
+      "▁quas": 59842,
+      "\\']))\\": 59843,
+      "▁extrapol": 59844,
+      "evaluation": 59845,
+      "太守": 59846,
+      "慷慨": 59847,
+      "▁Reb": 59848,
+      "orset": 59849,
+      "▁Pond": 59850,
+      "▁Treaty": 59851,
+      "▁pushes": 59852,
+      "munition": 59853,
+      "▁Fibonacci": 59854,
+      "新风": 59855,
+      "航空航天": 59856,
+      "▁plex": 59857,
+      "吃得": 59858,
+      "成套": 59859,
+      "泥土": 59860,
+      "号房屋": 59861,
+      "anno": 59862,
+      "imab": 59863,
+      "有的时候": 59864,
+      "正是因为": 59865,
+      "nVari": 59866,
+      "▁Ming": 59867,
+      "orphous": 59868,
+      "oxygenase": 59869,
+      "▁solvents": 59870,
+      "▁Automotive": 59871,
+      "总能": 59872,
+      "cus": 59873,
+      "Leaf": 59874,
+      "▁EGF": 59875,
+      "▁Ctrl": 59876,
+      "▁Franco": 59877,
+      "Physical": 59878,
+      "▁sickness": 59879,
+      "▁volcanic": 59880,
+      "▁hydration": 59881,
+      "▁zebrafish": 59882,
+      "减压": 59883,
+      "橡皮": 59884,
+      "认清": 59885,
+      "贴合": 59886,
+      "uxt": 59887,
+      "但是这": 59888,
+      "▁]];": 59889,
+      "什么都不": 59890,
+      "民终字第": 59891,
+      "刑二": 59892,
+      "售卖": 59893,
+      "洗脑": 59894,
+      "问了": 59895,
+      "tty": 59896,
+      "有效性": 59897,
+      "▁EUR": 59898,
+      "▁Density": 59899,
+      "▁aspirin": 59900,
+      "斑点": 59901,
+      "也能够": 59902,
+      "奥斯卡": 59903,
+      "▁babe": 59904,
+      "Impact": 59905,
+      "blockquote": 59906,
+      "(\"@": 59907,
+      "zoo": 59908,
+      "观众的": 59909,
+      "▁职务任免": 59910,
+      "受案登记表": 59911,
+      "在一定程度上": 59912,
+      "▁Barrett": 59913,
+      "▁porcine": 59914,
+      "▁clarified": 59915,
+      "洞察": 59916,
+      "肉眼": 59917,
+      "闭幕": 59918,
+      "哪怕是": 59919,
+      "▁synov": 59920,
+      "请你根据题意": 59921,
+      "stories": 59922,
+      "▁giants": 59923,
+      "▁Extreme": 59924,
+      "▁immersion": 59925,
+      "皆是": 59926,
+      "▁Tin": 59927,
+      "▁blamed": 59928,
+      "▁Creation": 59929,
+      "▁inferred": 59930,
+      "nPublication": 59931,
+      "层面的": 59932,
+      "gens": 59933,
+      "分类讨论": 59934,
+      "ndoes": 59935,
+      "tAssert": 59936,
+      "▁Kolkata": 59937,
+      "nFacebook": 59938,
+      "周三": 59939,
+      "贪污": 59940,
+      "▁TI": 59941,
+      "bike": 59942,
+      "▁warfare": 59943,
+      "不休": 59944,
+      "为大": 59945,
+      "奖章": 59946,
+      "小球": 59947,
+      "frm": 59948,
+      "层次的": 59949,
+      "消失了": 59950,
+      "设置的": 59951,
+      "▁foi": 59952,
+      "▁Tours": 59953,
+      "又包含了选项": 59954,
+      "既包含了问题": 59955,
+      "Usuario": 59956,
+      "上面这段话既包含了问题": 59957,
+      "请你根据题意选出正确的答案": 59958,
+      "书本": 59959,
+      "喝茶": 59960,
+      "患上": 59961,
+      "枕头": 59962,
+      "IZE": 59963,
+      "的危险": 59964,
+      "▁\\'[": 59965,
+      "nthet": 59966,
+      "▁Trials": 59967,
+      "▁bureauc": 59968,
+      "▁InputStream": 59969,
+      "▁photographed": 59970,
+      "ей": 59971,
+      "▁紫": 59972,
+      "西市": 59973,
+      "贯注": 59974,
+      "▁AH": 59975,
+      "中文名": 59976,
+      "▁Erd": 59977,
+      "glucose": 59978,
+      "此题主要考查了": 59979,
+      "▁handsome": 59980,
+      "▁consultations": 59981,
+      "▁instantaneous": 59982,
+      "倍数的": 59983,
+      "临床资料": 59984,
+      "职业学院": 59985,
+      "aille": 59986,
+      "▁Pitt": 59987,
+      "▁slavery": 59988,
+      "李世": 59989,
+      "漫漫": 59990,
+      "eyond": 59991,
+      "▁chel": 59992,
+      "格斯": 59993,
+      "分子的": 59994,
+      "zbek": 59995,
+      "Directive": 59996,
+      "▁grandparents": 59997,
+      "nFire": 59998,
+      "Choose": 59999,
+      "hammer": 60000,
+      "▁eclipse": 60001,
+      "sometimes": 60002,
+      "未发现可供执行的财产": 60003,
+      "▁Universities": 60004,
+      "不长": 60005,
+      "你说的": 60006,
+      "办理了": 60007,
+      "在各种": 60008,
+      "没见过": 60009,
+      "统计分析": 60010,
+      "Appro": 60011,
+      "odend": 60012,
+      "红袖添香网": 60013,
+      "▁doctorate": 60014,
+      "▁amplitudes": 60015,
+      "上司": 60016,
+      "的差": 60017,
+      "表征": 60018,
+      "CMV": 60019,
+      "功能和": 60020,
+      "ncom": 60021,
+      "▁Sys": 60022,
+      "▁apk": 60023,
+      "▁Hess": 60024,
+      "limate": 60025,
+      "闪耀": 60026,
+      "urf": 60027,
+      "一口气": 60028,
+      "行业内": 60029,
+      "thin": 60030,
+      "▁Tul": 60031,
+      "全方位的": 60032,
+      "▁slaves": 60033,
+      "▁pleaded": 60034,
+      "▁cheating": 60035,
+      "Assignment": 60036,
+      "重返": 60037,
+      "▁中国大陆": 60038,
+      "▁embassy": 60039,
+      "▁uppercase": 60040,
+      "之父": 60041,
+      "倒闭": 60042,
+      "寺庙": 60043,
+      "放的": 60044,
+      "手机上": 60045,
+      "▁/></": 60046,
+      "▁Bold": 60047,
+      "系夫妻关系": 60048,
+      "▁avenues": 60049,
+      "▁outward": 60050,
+      "▁sibling": 60051,
+      "▁novelist": 60052,
+      "▁parabolic": 60053,
+      ":)": 60054,
+      "一脚": 60055,
+      "何如": 60056,
+      "去向": 60057,
+      "▁河南": 60058,
+      "阜阳市": 60059,
+      "stores": 60060,
+      "多是": 60061,
+      "漏水": 60062,
+      "药业": 60063,
+      "dyn": 60064,
+      "在公司": 60065,
+      "▁Trees": 60066,
+      "小额贷款有限公司": 60067,
+      "▁histamine": 60068,
+      "黄帝": 60069,
+      "evt": 60070,
+      "igms": 60071,
+      "符合减刑": 60072,
+      "▁Flip": 60073,
+      "▁Canadians": 60074,
+      "体长": 60075,
+      "_(\"": 60076,
+      "经过了": 60077,
+      "▁Tonight": 60078,
+      "▁consectetur": 60079,
+      "专线": 60080,
+      "天龙": 60081,
+      "空腹": 60082,
+      "靠谱的": 60083,
+      "▁SAM": 60084,
+      "posites": 60085,
+      "▁rebound": 60086,
+      "zl": 60087,
+      "助学": 60088,
+      "实操": 60089,
+      "导电": 60090,
+      "\"]:\\": 60091,
+      "nTor": 60092,
+      "的最小值为": 60093,
+      "▁Benedict": 60094,
+      "▁colourful": 60095,
+      "大词典": 60096,
+      "▁trout": 60097,
+      "nScient": 60098,
+      "▁Pokemon": 60099,
+      "collector": 60100,
+      "▁Behavioral": 60101,
+      "▁invitations": 60102,
+      "丸子": 60103,
+      "兼具": 60104,
+      "如是": 60105,
+      "时就": 60106,
+      "请教": 60107,
+      "nper": 60108,
+      "▁Bid": 60109,
+      "▁accr": 60110,
+      "iplier": 60111,
+      "▁Founder": 60112,
+      "▁Exposure": 60113,
+      "▁deadlines": 60114,
+      "▁judgement": 60115,
+      "▁experimentation": 60116,
+      "以求": 60117,
+      "有心": 60118,
+      "JPG": 60119,
+      "Obs": 60120,
+      "办实事": 60121,
+      "▁logout": 60122,
+      "▁showcasing": 60123,
+      "关机": 60124,
+      "带队": 60125,
+      "ncode": 60126,
+      "▁smiles": 60127,
+      "setLevel": 60128,
+      "公章": 60129,
+      "周四": 60130,
+      "胡子": 60131,
+      "}${": 60132,
+      "nist": 60133,
+      "单位出具": 60134,
+      "没有办法": 60135,
+      "犯罪情节": 60136,
+      "▁Mick": 60137,
+      "ophore": 60138,
+      "scopes": 60139,
+      "▁facet": 60140,
+      "▁Broken": 60141,
+      "内脏": 60142,
+      "格力": 60143,
+      "站起来": 60144,
+      "▁gcc": 60145,
+      "arcer": 60146,
+      "Matchers": 60147,
+      "舒缓": 60148,
+      "Hours": 60149,
+      "benef": 60150,
+      "▁flock": 60151,
+      "作出明智的选择": 60152,
+      "lessness": 60153,
+      "belongsTo": 60154,
+      "▁decentralized": 60155,
+      "▁fertilization": 60156,
+      "哈希": 60157,
+      "民申": 60158,
+      "文化旅游": 60159,
+      "▁Noah": 60160,
+      "人力资源管理": 60161,
+      "▁infamous": 60162,
+      "▁quotation": 60163,
+      "▁Collaboration": 60164,
+      "再也": 60165,
+      "喝了": 60166,
+      "座的": 60167,
+      "胜负": 60168,
+      "诸侯": 60169,
+      "▁LIB": 60170,
+      "ewood": 60171,
+      "ureka": 60172,
+      "▁commissions": 60173,
+      "Beta": 60174,
+      "}`,\\": 60175,
+      "▁bunk": 60176,
+      "▁{}\",": 60177,
+      "高新技术企业": 60178,
+      "pathname": 60179,
+      "▁Courses": 60180,
+      "此项": 60181,
+      "王妃": 60182,
+      "astric": 60183,
+      "中医药学名词审定委员会": 60184,
+      "也更": 60185,
+      "听的": 60186,
+      "条纹": 60187,
+      "餐具": 60188,
+      "▁STATE": 60189,
+      "▁Hearts": 60190,
+      "▁Absolute": 60191,
+      "著作权集体管理协会": 60192,
+      "凶手": 60193,
+      "天平": 60194,
+      "计较": 60195,
+      "▁из": 60196,
+      "▁Dispatch": 60197,
+      "▁Thinking": 60198,
+      "▁understandable": 60199,
+      "▁DG": 60200,
+      "▁kHz": 60201,
+      "recht": 60202,
+      "▁fats": 60203,
+      "riates": 60204,
+      "人力资源和社会保障局": 60205,
+      "海盗": 60206,
+      "▁东方": 60207,
+      "深入开展": 60208,
+      "alcoh": 60209,
+      "收费是每人": 60210,
+      "abbage": 60211,
+      "▁trape": 60212,
+      "attered": 60213,
+      "▁Andreas": 60214,
+      "其他财产可供执行": 60215,
+      "hibernate": 60216,
+      "ouncement": 60217,
+      "▁mononuclear": 60218,
+      "求是": 60219,
+      "而对": 60220,
+      "armed": 60221,
+      "nFound": 60222,
+      "请您根据您的": 60223,
+      "vertebr": 60224,
+      "nlanguage": 60225,
+      "▁淋浴": 60226,
+      "可以刷卡": 60227,
+      "判断和知识": 60228,
+      "浏览备选答案": 60229,
+      "▁CONFIG": 60230,
+      "▁Baldwin": 60231,
+      "▁publishes": 60232,
+      "▁surrogate": 60233,
+      "判断和知识回答以下问题": 60234,
+      "StateToProps": 60235,
+      "authenticate": 60236,
+      "读本": 60237,
+      "▁Styles": 60238,
+      "▁resumed": 60239,
+      "workflows": 60240,
+      "▁myocardium": 60241,
+      "抗癌": 60242,
+      "热处理": 60243,
+      "acea": 60244,
+      "转为普通程序": 60245,
+      "▁Diversity": 60246,
+      "▁Specification": 60247,
+      "子集": 60248,
+      "开元": 60249,
+      "梯度": 60250,
+      "竞价": 60251,
+      "ADI": 60252,
+      "大教堂": 60253,
+      "资格证": 60254,
+      "▁cysteine": 60255,
+      "protection": 60256,
+      "各家": 60257,
+      "宣称": 60258,
+      "深思": 60259,
+      "赛场": 60260,
+      "毕业的": 60261,
+      "Training": 60262,
+      "格的": 60263,
+      "Rpc": 60264,
+      "relax": 60265,
+      "年平均气温": 60266,
+      "经济的发展": 60267,
+      "HashCode": 60268,
+      "ingement": 60269,
+      "▁certainty": 60270,
+      "怀里": 60271,
+      "摆在": 60272,
+      "裂纹": 60273,
+      "prefer": 60274,
+      "▁skies": 60275,
+      "▁Levels": 60276,
+      "提出的撤诉申请": 60277,
+      "▁Calling": 60278,
+      "▁plaques": 60279,
+      "▁Metadata": 60280,
+      "痊愈": 60281,
+      "风险管理": 60282,
+      "erated": 60283,
+      "▁nodules": 60284,
+      "▁Messages": 60285,
+      "DateFormat": 60286,
+      "文联": 60287,
+      "潜伏": 60288,
+      "选了": 60289,
+      "Lorem": 60290,
+      "emoji": 60291,
+      "leter": 60292,
+      "mybatis": 60293,
+      "ostatin": 60294,
+      "▁occurrences": 60295,
+      "dynamics": 60296,
+      "Ak": 60297,
+      "拍了": 60298,
+      "igt": 60299,
+      "▁rx": 60300,
+      "共同的": 60301,
+      "▁Somal": 60302,
+      "▁Weber": 60303,
+      "▁admire": 60304,
+      "▁astronaut": 60305,
+      "▁remembering": 60306,
+      "恒温": 60307,
+      "quist": 60308,
+      "▁cans": 60309,
+      "▁sage": 60310,
+      "遥感": 60311,
+      "emi": 60312,
+      "FACE": 60313,
+      "ammers": 60314,
+      "▁roughness": 60315,
+      "▁replacements": 60316,
+      "藏着": 60317,
+      "▁Wei": 60318,
+      "▁Holl": 60319,
+      "Canada": 60320,
+      "opause": 60321,
+      "▁Algeria": 60322,
+      "音像著作权集体管理协会": 60323,
+      "的电": 60324,
+      "自发": 60325,
+      "言之": 60326,
+      "通胀": 60327,
+      "DOT": 60328,
+      "nFort": 60329,
+      "▁alkal": 60330,
+      "▁neutrino": 60331,
+      "▁leveraging": 60332,
+      "他想": 60333,
+      "叹息": 60334,
+      "并就": 60335,
+      "救济": 60336,
+      "药房": 60337,
+      ")(\"": 60338,
+      "创建的": 60339,
+      "▁Hole": 60340,
+      "▁voor": 60341,
+      "ontology": 60342,
+      "▁BET": 60343,
+      "elman": 60344,
+      "▁wholly": 60345,
+      "▁Currency": 60346,
+      "Calculator": 60347,
+      "又将": 60348,
+      "越秀": 60349,
+      "软化": 60350,
+      "名下位于": 60351,
+      "基础设施建设": 60352,
+      "addWidget": 60353,
+      "Migrations": 60354,
+      "成龙": 60355,
+      "水体": 60356,
+      "王家": 60357,
+      "正处于": 60358,
+      "▁coils": 60359,
+      "▁awhile": 60360,
+      "▁staffing": 60361,
+      "▁legislators": 60362,
+      "Sr": 60363,
+      "只看": 60364,
+      "常有": 60365,
+      "棋子": 60366,
+      "站立": 60367,
+      "聊聊": 60368,
+      "作品中": 60369,
+      "条形码": 60370,
+      "那么在": 60371,
+      "深入推进": 60372,
+      "▁omit": 60373,
+      "▁islet": 60374,
+      "nEverything": 60375,
+      "亲和": 60376,
+      "近百": 60377,
+      "音乐的": 60378,
+      "▁genomics": 60379,
+      "▁în": 60380,
+      "曾多次": 60381,
+      "▁Jong": 60382,
+      "▁\"\");\\": 60383,
+      "▁unint": 60384,
+      "▁caries": 60385,
+      "▁nucleotides": 60386,
+      "三农": 60387,
+      "nBob": 60388,
+      "nJune": 60389,
+      "▁André": 60390,
+      "▁greeting": 60391,
+      "惆怅": 60392,
+      "惩戒": 60393,
+      "文凭": 60394,
+      "皇冠": 60395,
+      "济宁市": 60396,
+      "▁dare": 60397,
+      "▁dyes": 60398,
+      "▁nets": 60399,
+      "▁smells": 60400,
+      "▁stripe": 60401,
+      "出击": 60402,
+      "微分": 60403,
+      "高中的": 60404,
+      "ipel": 60405,
+      "▁Slim": 60406,
+      "naming": 60407,
+      "▁Lorentz": 60408,
+      "▁NG": 60409,
+      "▁Ps": 60410,
+      "不包括": 60411,
+      "公里的": 60412,
+      "刷题": 60413,
+      "拥挤": 60414,
+      "木瓜": 60415,
+      "**（": 60416,
+      "工作时": 60417,
+      "nCle": 60418,
+      "asList": 60419,
+      "▁Libya": 60420,
+      "Supplier": 60421,
+      "▁recount": 60422,
+      "▁correspondent": 60423,
+      "两车": 60424,
+      "出让": 60425,
+      "答疑": 60426,
+      "装载": 60427,
+      "豌豆": 60428,
+      "遗失": 60429,
+      "WOR": 60430,
+      "人数的": 60431,
+      "remark": 60432,
+      "▁heroin": 60433,
+      "▁eternal": 60434,
+      "▁workflows": 60435,
+      "DoesNotExist": 60436,
+      "▁volunteering": 60437,
+      "▁先": 60438,
+      "记账": 60439,
+      "贴士": 60440,
+      "的几个": 60441,
+      "otin": 60442,
+      "资格证书": 60443,
+      "rored": 60444,
+      "▁Lars": 60445,
+      "▁atop": 60446,
+      "Sphere": 60447,
+      "ullary": 60448,
+      "二〇一六年六": 60449,
+      "KD": 60450,
+      "全神": 60451,
+      "天府": 60452,
+      "造影": 60453,
+      "(\"\",": 60454,
+      "▁MIL": 60455,
+      "刑事判决书": 60456,
+      "▁Shack": 60457,
+      "CONTENT": 60458,
+      "nStatistics": 60459,
+      "冷凝": 60460,
+      "出汗": 60461,
+      "教的": 60462,
+      "UNIT": 60463,
+      "▁Taj": 60464,
+      "那么这个": 60465,
+      "declare": 60466,
+      "nothing": 60467,
+      "WebSocket": 60468,
+      "王牌": 60469,
+      "编者": 60470,
+      "平均每": 60471,
+      "依据是什么": 60472,
+      "impact": 60473,
+      "▁<%": 60474,
+      "▁PGA": 60475,
+      "▁pods": 60476,
+      "通过本院或直接向": 60477,
+      "▁slaughter": 60478,
+      "▁prototypes": 60479,
+      "她和": 60480,
+      "悉尼": 60481,
+      "清扫": 60482,
+      "私家": 60483,
+      "远古": 60484,
+      "hid": 60485,
+      "onio": 60486,
+      "违法所得": 60487,
+      "amble": 60488,
+      "ール": 60489,
+      "化身": 60490,
+      "岛上": 60491,
+      "▁言情": 60492,
+      "这东西": 60493,
+      "环境卫生": 60494,
+      "公诉机关认为": 60495,
+      "▁Changed": 60496,
+      "▁Coastal": 60497,
+      "▁Machines": 60498,
+      "deprecated": 60499,
+      "nMicrosoft": 60500,
+      "有一位": 60501,
+      "▁TBI": 60502,
+      "未经法庭": 60503,
+      "二〇一六年四月": 60504,
+      "▁judging": 60505,
+      "▁carbonate": 60506,
+      "和控制": 60507,
+      "Produ": 60508,
+      "▁Turing": 60509,
+      "▁т": 60510,
+      "的手段": 60511,
+      "▁prag": 60512,
+      "▁unle": 60513,
+      "nAccept": 60514,
+      "▁interrog": 60515,
+      "AUTO": 60516,
+      "nProve": 60517,
+      "▁arose": 60518,
+      "依据已经发生法律效力的": 60519,
+      "互换": 60520,
+      "野心": 60521,
+      "▁Notch": 60522,
+      "▁feast": 60523,
+      "▁replay": 60524,
+      "▁Immigration": 60525,
+      "交接": 60526,
+      "必修": 60527,
+      "直言": 60528,
+      "看懂": 60529,
+      "操作简单": 60530,
+      "nColl": 60531,
+      "▁Gael": 60532,
+      "▁Quiz": 60533,
+      "▁Knock": 60534,
+      "▁epitopes": 60535,
+      "庆幸": 60536,
+      "日历": 60537,
+      "稳产": 60538,
+      "送餐": 60539,
+      "zek": 60540,
+      "的收入": 60541,
+      "实习律师": 60542,
+      "\\'})\\": 60543,
+      "▁**/\\": 60544,
+      "▁Powder": 60545,
+      "plusplus": 60546,
+      "用科学记数法表示为": 60547,
+      "▁cathedral": 60548,
+      "▁RuntimeError": 60549,
+      "则在": 60550,
+      "朗诵": 60551,
+      "Dog": 60552,
+      "命运的": 60553,
+      "拼多多": 60554,
+      "nwww": 60555,
+      "ritt": 60556,
+      "▁fus": 60557,
+      "市委常委": 60558,
+      "▁cron": 60559,
+      "▁dell": 60560,
+      "rystalline": 60561,
+      "▁cartridge": 60562,
+      "▁adaptations": 60563,
+      "编纂": 60564,
+      "赶上": 60565,
+      "QString": 60566,
+      "本院审理过程中": 60567,
+      "▁cerebellar": 60568,
+      "▁leukocytes": 60569,
+      "完好": 60570,
+      "新老": 60571,
+      "诺基亚": 60572,
+      "arbon": 60573,
+      "rived": 60574,
+      "为终审判决": 60575,
+      "▁Europa": 60576,
+      "妇幼": 60577,
+      "消炎": 60578,
+      "等边": 60579,
+      "ulg": 60580,
+      "自然灾害": 60581,
+      "pandas": 60582,
+      "▁crawl": 60583,
+      "▁Latvia": 60584,
+      "▁starters": 60585,
+      "多达": 60586,
+      "旅途": 60587,
+      "BAD": 60588,
+      "llvm": 60589,
+      "FAILED": 60590,
+      "Indexed": 60591,
+      "▁maximizing": 60592,
+      "▁multiplicity": 60593,
+      "客家": 60594,
+      "神木": 60595,
+      "选的": 60596,
+      "郑某": 60597,
+      "nPe": 60598,
+      "▁Wii": 60599,
+      "ooters": 60600,
+      "▁biases": 60601,
+      "▁framing": 60602,
+      "卡斯": 60603,
+      "日程": 60604,
+      "是好": 60605,
+      "读完": 60606,
+      "nCap": 60607,
+      "上市时间": 60608,
+      "nTowns": 60609,
+      "rollment": 60610,
+      "▁participates": 60611,
+      "警务": 60612,
+      "遗漏": 60613,
+      "小时后": 60614,
+      "最早的": 60615,
+      "等量关系": 60616,
+      "nmkdir": 60617,
+      "Terminal": 60618,
+      "istrator": 60619,
+      "给你的": 60620,
+      "astings": 60621,
+      "▁simmer": 60622,
+      "▁treaty": 60623,
+      "timeline": 60624,
+      "我被": 60625,
+      "藏族": 60626,
+      "远程仓库": 60627,
+      "heavy": 60628,
+      "lipid": 60629,
+      "▁Maxim": 60630,
+      "▁recap": 60631,
+      "▁defaultdict": 60632,
+      "详尽": 60633,
+      "跨界": 60634,
+      "雪山": 60635,
+      "opedic": 60636,
+      "▁Chall": 60637,
+      "nJanuary": 60638,
+      "▁Creator": 60639,
+      "以待": 60640,
+      "养分": 60641,
+      "venv": 60642,
+      "▁Rim": 60643,
+      "▁Marian": 60644,
+      "▁bottlen": 60645,
+      "号民事调解书已经发生法律效力": 60646,
+      "ós": 60647,
+      "游离": 60648,
+      "rike": 60649,
+      "ademy": 60650,
+      "▁DELETE": 60651,
+      "▁inaccurate": 60652,
+      "正品": 60653,
+      "nyt": 60654,
+      "会提供": 60655,
+      "安全和": 60656,
+      "加快推进": 60657,
+      "进行调解": 60658,
+      "▁ASAP": 60659,
+      "▁elegans": 60660,
+      "▁wearable": 60661,
+      "债权依法受法律保护": 60662,
+      "号民事调解书的本次执行程序": 60663,
+      "▁乔": 60664,
+      "乡村人口": 60665,
+      "QtGui": 60666,
+      "▁Chic": 60667,
+      "addons": 60668,
+      "▁Alter": 60669,
+      "▁sprink": 60670,
+      "▁cardboard": 60671,
+      "▁identifiable": 60672,
+      "▁邵": 60673,
+      "的行业": 60674,
+      "教学工作": 60675,
+      "▁NEWS": 60676,
+      "▁mating": 60677,
+      "assertRaises": 60678,
+      "CHA": 60679,
+      "PLA": 60680,
+      "oks": 60681,
+      "技术的发展": 60682,
+      "▁Sense": 60683,
+      "Delivery": 60684,
+      "▁tolerate": 60685,
+      "ObjectName": 60686,
+      "▁dysplasia": 60687,
+      "以保证": 60688,
+      "认证的": 60689,
+      "充分肯定": 60690,
+      "特色产业": 60691,
+      "\\'\\',": 60692,
+      "ioids": 60693,
+      "▁souls": 60694,
+      "continence": 60695,
+      "光阴": 60696,
+      "接触到": 60697,
+      "管理器": 60698,
+      "nmay": 60699,
+      "▁FPGA": 60700,
+      "▁lunar": 60701,
+      "是在法律允许": 60702,
+      "▁Chilean": 60703,
+      "simulation": 60704,
+      "以北": 60705,
+      "底的": 60706,
+      "涉外": 60707,
+      "高雅": 60708,
+      "HPV": 60709,
+      "GENER": 60710,
+      "▁suspend": 60711,
+      "restricted": 60712,
+      "▁lu": 60713,
+      "大家对": 60714,
+      "nTrad": 60715,
+      "▁Aviv": 60716,
+      "▁achievable": 60717,
+      "erb": 60718,
+      "的特性": 60719,
+      "的经典": 60720,
+      "estro": 60721,
+      "▁aroma": 60722,
+      "▁affiliation": 60723,
+      "开具": 60724,
+      "赔付": 60725,
+      "阵阵": 60726,
+      "▁orn": 60727,
+      "被执行人无": 60728,
+      "▁Venus": 60729,
+      "▁lodge": 60730,
+      "现申请执行人": 60731,
+      "▁Checks": 60732,
+      "▁repairing": 60733,
+      "城东": 60734,
+      "城镇化": 60735,
+      "是无法": 60736,
+      "高强度": 60737,
+      "数据处理": 60738,
+      "British": 60739,
+      "umetric": 60740,
+      "dispatcher": 60741,
+      "▁欧": 60742,
+      "说一句": 60743,
+      "▁PIN": 60744,
+      "▁Pharmacy": 60745,
+      "中含": 60746,
+      "繁忙": 60747,
+      "有特殊": 60748,
+      "耐腐蚀": 60749,
+      "本院查明": 60750,
+      "otation": 60751,
+      "▁subway": 60752,
+      "▁Biography": 60753,
+      "春夏": 60754,
+      "体育场": 60755,
+      "▁bapt": 60756,
+      "chains": 60757,
+      "▁Jackie": 60758,
+      "二本": 60759,
+      "树种": 60760,
+      "止损": 60761,
+      "因数的": 60762,
+      "近似数": 60763,
+      "希望能够": 60764,
+      "考试成绩": 60765,
+      "Carbon": 60766,
+      "medical": 60767,
+      "▁Formal": 60768,
+      "▁Trigon": 60769,
+      "两根": 60770,
+      "小华": 60771,
+      "铁皮": 60772,
+      "WITH": 60773,
+      "rail": 60774,
+      "品牌形象": 60775,
+      "▁impe": 60776,
+      "许可中途退": 60777,
+      "lander": 60778,
+      "▁Brass": 60779,
+      "iliated": 60780,
+      "nSchools": 60781,
+      "▁Colonial": 60782,
+      "▁exploits": 60783,
+      "HQ": 60784,
+      "▁Aj": 60785,
+      "▁安徽": 60786,
+      "社区居民": 60787,
+      "emann": 60788,
+      "▁empirically": 60789,
+      "Ol": 60790,
+      "何为": 60791,
+      "剪切": 60792,
+      "安稳": 60793,
+      "RNG": 60794,
+      "▁Oc": 60795,
+      "属植物": 60796,
+      "高标准": 60797,
+      "可以上诉": 60798,
+      "▁radiol": 60799,
+      "Interrupt": 60800,
+      "instagram": 60801,
+      "▁robotics": 60802,
+      "▁separable": 60803,
+      "xu": 60804,
+      "绝世": 60805,
+      "出现问题": 60806,
+      "▁Hire": 60807,
+      "▁crab": 60808,
+      "如发现被执行人": 60809,
+      "▁sandbox": 60810,
+      "不时": 60811,
+      "宫殿": 60812,
+      "给药": 60813,
+      "这与": 60814,
+      "一直未": 60815,
+      "宜昌市": 60816,
+      "走出去": 60817,
+      "laws": 60818,
+      "Grand": 60819,
+      "▁congruent": 60820,
+      "zc": 60821,
+      "nBo": 60822,
+      "abolism": 60823,
+      "同龄": 60824,
+      "炫耀": 60825,
+      "播放器": 60826,
+      "baseUrl": 60827,
+      "NumberOf": 60828,
+      "▁Primera": 60829,
+      "序幕": 60830,
+      "nsl": 60831,
+      "▁Priv": 60832,
+      "一定程度上": 60833,
+      "umping": 60834,
+      "县人民法院作出的": 60835,
+      "▁methanol": 60836,
+      "廉价": 60837,
+      "此书": 60838,
+      "nimg": 60839,
+      "进行计算即可": 60840,
+      "它会": 60841,
+      "室温": 60842,
+      "杯子": 60843,
+      "会展中心": 60844,
+      "▁Maritime": 60845,
+      "▁newsletters": 60846,
+      "依法应": 60847,
+      "Uuid": 60848,
+      "▁tedious": 60849,
+      "▁六": 60850,
+      "三方": 60851,
+      "借到": 60852,
+      "国画": 60853,
+      "小狗": 60854,
+      "机等": 60855,
+      "白宫": 60856,
+      "预留": 60857,
+      "urm": 60858,
+      "人物的": 60859,
+      "结婚了": 60860,
+      "设计中": 60861,
+      "公共场所": 60862,
+      "▁POSS": 60863,
+      "▁Utils": 60864,
+      "▁rainy": 60865,
+      "Concept": 60866,
+      "▁capped": 60867,
+      "▁porcelain": 60868,
+      "IoT": 60869,
+      "九龙坡": 60870,
+      "编辑器": 60871,
+      "CLAI": 60872,
+      "详细内容": 60873,
+      "▁wizard": 60874,
+      "submission": 60875,
+      "▁Advertising": 60876,
+      "个案": 60877,
+      "放手": 60878,
+      "▁relocation": 60879,
+      "性情": 60880,
+      "金星": 60881,
+      "治疗前": 60882,
+      "npath": 60883,
+      "▁з": 60884,
+      "五彩": 60885,
+      "抬起": 60886,
+      "腰部": 60887,
+      "Lee": 60888,
+      "]];\\": 60889,
+      "idan": 60890,
+      "车牌号为": 60891,
+      "PYTHON": 60892,
+      "ToArray": 60893,
+      "▁Pocket": 60894,
+      "▁waveform": 60895,
+      "类比": 60896,
+      "繁多": 60897,
+      "不能用": 60898,
+      "'},\\": 60899,
+      "ilee": 60900,
+      "一步一步": 60901,
+      "▁extrac": 60902,
+      "▁crafting": 60903,
+      "▁phosphatidyl": 60904,
+      "▁干": 60905,
+      "编写的": 60906,
+      "iceps": 60907,
+      "bigcup": 60908,
+      "nShips": 60909,
+      "▁desperately": 60910,
+      "倍的": 60911,
+      "太低": 60912,
+      "说她": 60913,
+      "▁plea": 60914,
+      "▁Tales": 60915,
+      "▁broadcasts": 60916,
+      ">×": 60917,
+      "王府": 60918,
+      "双床间": 60919,
+      "来解决": 60920,
+      "要做好": 60921,
+      "nFac": 60922,
+      "▁PTH": 60923,
+      "▁spaced": 60924,
+      "▁throne": 60925,
+      "▁Johnston": 60926,
+      "▁FOX": 60927,
+      "没有那么": 60928,
+      "研究领域": 60929,
+      "▁tipo": 60930,
+      "italic": 60931,
+      "expires": 60932,
+      "▁yogurt": 60933,
+      "Subscribe": 60934,
+      "▁doctrine": 60935,
+      "▁М": 60936,
+      "为先": 60937,
+      "地把": 60938,
+      "MIS": 60939,
+      "服务质量": 60940,
+      "综合考虑": 60941,
+      "prior": 60942,
+      "▁Moses": 60943,
+      "▁clocks": 60944,
+      "▁slated": 60945,
+      "Primitive": 60946,
+      "▁footwear": 60947,
+      "▁multipro": 60948,
+      "▁masculine": 60949,
+      "布莱": 60950,
+      "闲置": 60951,
+      "消费品": 60952,
+      "等形式": 60953,
+      "azol": 60954,
+      "▁PNG": 60955,
+      "▁Bread": 60956,
+      "经人介绍相识": 60957,
+      "违章": 60958,
+      "逆天": 60959,
+      "▁教师": 60960,
+      "对这些": 60961,
+      "汕头市": 60962,
+      "ankton": 60963,
+      "▁shutil": 60964,
+      "▁discern": 60965,
+      "▁Concepts": 60966,
+      "利斯": 60967,
+      "即成": 60968,
+      "奇异": 60969,
+      "所见": 60970,
+      "置身": 60971,
+      "▁moms": 60972,
+      "nIrish": 60973,
+      "omized": 60974,
+      "▁charger": 60975,
+      "probability": 60976,
+      "▁distinguishing": 60977,
+      "有待": 60978,
+      "jie": 60979,
+      "参观了": 60980,
+      "情人节": 60981,
+      "▁GSM": 60982,
+      "▁Newman": 60983,
+      "▁Appeals": 60984,
+      "某某到庭参加诉讼": 60985,
+      "▁ordained": 60986,
+      "▁Architects": 60987,
+      "▁Structures": 60988,
+      "UGH": 60989,
+      "Opacity": 60990,
+      "▁Sophie": 60991,
+      "▁parked": 60992,
+      "▁Upgrade": 60993,
+      "▁Portfolio": 60994,
+      "▁inserting": 60995,
+      "攀升": 60996,
+      "iOiJ": 60997,
+      "疫情防控工作": 60998,
+      "Clicked": 60999,
+      "本判决为终审判决": 61000,
+      "Deployment": 61001,
+      "内向": 61002,
+      "进门": 61003,
+      "GAN": 61004,
+      "Php": 61005,
+      "传说中的": 61006,
+      "▁wifi": 61007,
+      "inarily": 61008,
+      "▁Escort": 61009,
+      "▁Christie": 61010,
+      "_)\\": 61011,
+      "能达到": 61012,
+      "▁Dup": 61013,
+      "asome": 61014,
+      "▁burger": 61015,
+      "INSTANCE": 61016,
+      "厉害的": 61017,
+      "nWood": 61018,
+      "▁Crazy": 61019,
+      "▁Commod": 61020,
+      "committee": 61021,
+      "▁Conversion": 61022,
+      "ча": 61023,
+      "华尔": 61024,
+      "真好": 61025,
+      "饥饿": 61026,
+      "VEGF": 61027,
+      "ibia": 61028,
+      "▁corridor": 61029,
+      "Injectable": 61030,
+      "▁unparalleled": 61031,
+      "中卫": 61032,
+      "譬如": 61033,
+      "目的观察": 61034,
+      "▁Kerr": 61035,
+      "▁Phosph": 61036,
+      "▁expectancy": 61037,
+      "对焦": 61038,
+      "有各": 61039,
+      "甘蔗": 61040,
+      "从来不": 61041,
+      "自己做": 61042,
+      "nKing": 61043,
+      "pound": 61044,
+      "riding": 61045,
+      "▁explode": 61046,
+      "UF": 61047,
+      "凝结": 61048,
+      "力强": 61049,
+      "KTV": 61050,
+      "想象的": 61051,
+      "没有再": 61052,
+      "alez": 61053,
+      "Young": 61054,
+      "做题的思路": 61055,
+      "▁Arlington": 61056,
+      "吸尘": 61057,
+      "ngr": 61058,
+      "反过来": 61059,
+      "感觉很": 61060,
+      "ères": 61061,
+      "▁decks": 61062,
+      "▁devote": 61063,
+      "▁Customs": 61064,
+      "沙坪": 61065,
+      "mixin": 61066,
+      "▁woke": 61067,
+      "分析和依据": 61068,
+      "▁Haiti": 61069,
+      "通气": 61070,
+      "ENU": 61071,
+      "▁hd": 61072,
+      "主干道": 61073,
+      "他们都": 61074,
+      "关心的": 61075,
+      "创作者": 61076,
+      "nDer": 61077,
+      "质量控制": 61078,
+      "lover": 61079,
+      "▁resusc": 61080,
+      "并给出你的解题": 61081,
+      "anaconda": 61082,
+      "▁Landing": 61083,
+      "从选项中选择出合适的答案": 61084,
+      "并给出你的解题分析和依据": 61085,
+      "拖鞋": 61086,
+      "超人": 61087,
+      ",,,": 61088,
+      "stor": 61089,
+      "到底是什么": 61090,
+      "▁naughty": 61091,
+      "红球": 61092,
+      "街区": 61093,
+      "hoe": 61094,
+      "sty": 61095,
+      "优异的": 61096,
+      "宅基地": 61097,
+      "▁flown": 61098,
+      "▁treas": 61099,
+      "surgical": 61100,
+      "帮他": 61101,
+      "战机": 61102,
+      "无偿": 61103,
+      "求精": 61104,
+      "着她": 61105,
+      "▁Concer": 61106,
+      "▁Falcon": 61107,
+      "▁showcases": 61108,
+      "以赴": 61109,
+      "Mind": 61110,
+      "Built": 61111,
+      "并做出解释": 61112,
+      "▁sneak": 61113,
+      "极少": 61114,
+      "LPL": 61115,
+      "性能和": 61116,
+      "nDou": 61117,
+      "缓刑一年": 61118,
+      "▁Fors": 61119,
+      "▁rehab": 61120,
+      "booking": 61121,
+      "选出正确的答案并做出解释": 61122,
+      "告白": 61123,
+      "重构": 61124,
+      "IDA": 61125,
+      "会选择": 61126,
+      "的直线": 61127,
+      "▁USS": 61128,
+      "数的绝对值": 61129,
+      "ATPase": 61130,
+      "上诉于江苏省": 61131,
+      "▁credible": 61132,
+      "华语": 61133,
+      "沉重": 61134,
+      "chy": 61135,
+      "马鞍山市": 61136,
+      "plings": 61137,
+      "Tracking": 61138,
+      "▁Practices": 61139,
+      "pyl": 61140,
+      "ofib": 61141,
+      "Travel": 61142,
+      "▁Behind": 61143,
+      "位和": 61144,
+      "议论": 61145,
+      "onet": 61146,
+      "nFish": 61147,
+      "申请执行人亦未": 61148,
+      "GetMapping": 61149,
+      "hq": 61150,
+      "▁韦": 61151,
+      "久久": 61152,
+      "你和": 61153,
+      "全等": 61154,
+      "勤劳": 61155,
+      "怪的": 61156,
+      "点心": 61157,
+      "才开始": 61158,
+      "现场照片": 61159,
+      "▁Dirac": 61160,
+      "▁babel": 61161,
+      "▁Wizard": 61162,
+      "kube": 61163,
+      "▁MOS": 61164,
+      "融资租赁": 61165,
+      "通俗易懂": 61166,
+      "aught": 61167,
+      "OrDefault": 61168,
+      "▁Rajasthan": 61169,
+      "▁propaganda": 61170,
+      "万字": 61171,
+      "普惠": 61172,
+      "这在": 61173,
+      "自诉人": 61174,
+      "▁pleth": 61175,
+      "坚硬": 61176,
+      "家政": 61177,
+      "缺氧": 61178,
+      "先天性": 61179,
+      "否命题": 61180,
+      "Formula": 61181,
+      "▁recursion": 61182,
+      "▁territorial": 61183,
+      "▁Revolutionary": 61184,
+      "动人": 61185,
+      "悲哀": 61186,
+      "武士": 61187,
+      "状的": 61188,
+      "要先": 61189,
+      "为己任": 61190,
+      "有网友": 61191,
+      "CLAS": 61192,
+      "phony": 61193,
+      "▁Alps": 61194,
+      "Marshal": 61195,
+      "▁Lankan": 61196,
+      "ungarian": 61197,
+      "正确答案和做题的思路": 61198,
+      "不加": 61199,
+      "寄托": 61200,
+      "▁武汉": 61201,
+      "理论研究": 61202,
+      "graded": 61203,
+      "▁Worldwide": 61204,
+      "年获得": 61205,
+      "▁hap": 61206,
+      "coeff": 61207,
+      "LOAT": 61208,
+      "也有一些": 61209,
+      "▁Cran": 61210,
+      "▁REAL": 61211,
+      "间谍": 61212,
+      "▁'',": 61213,
+      "▁maize": 61214,
+      "申请撤回上诉": 61215,
+      "fashion": 61216,
+      "recipient": 61217,
+      "▁abnormality": 61218,
+      "Wr": 61219,
+      "ZA": 61220,
+      "大会上": 61221,
+      "面对面": 61222,
+      "▁SKF": 61223,
+      "berra": 61224,
+      "ermen": 61225,
+      "二〇一五年三月": 61226,
+      "▁singleton": 61227,
+      "▁shortening": 61228,
+      "何事": 61229,
+      "听众": 61230,
+      "建明": 61231,
+      "倍数是": 61232,
+      "性问题": 61233,
+      "此时的": 61234,
+      "ucus": 61235,
+      "也请给出你的": 61236,
+      "▁chasing": 61237,
+      "解题过程和分析思路": 61238,
+      "也请给出你的解题过程和分析思路": 61239,
+      "finance": 61240,
+      "nLittle": 61241,
+      "▁Principles": 61242,
+      "▁contradict": 61243,
+      "▁即": 61244,
+      "个好": 61245,
+      "病学": 61246,
+      "为契机": 61247,
+      "为了避免": 61248,
+      "▁Rare": 61249,
+      "▁saddle": 61250,
+      "▁berries": 61251,
+      "▁emailed": 61252,
+      "LoggerFactory": 61253,
+      "本是": 61254,
+      "建立的": 61255,
+      "Opts": 61256,
+      "▁YAML": 61257,
+      "LH": 61258,
+      "动静": 61259,
+      "学生对": 61260,
+      "的地位": 61261,
+      "▁Tac": 61262,
+      "decoder": 61263,
+      "▁MongoDB": 61264,
+      "MLElement": 61265,
+      "getElementsBy": 61266,
+      "忙着": 61267,
+      "自此": 61268,
+      "▁ORDER": 61269,
+      "条第一款规定": 61270,
+      "未经法庭许可中途退": 61271,
+      "▁polyester": 61272,
+      "一侧": 61273,
+      "建房": 61274,
+      "浓度为": 61275,
+      "▁Coh": 61276,
+      "lymph": 61277,
+      "HEIGHT": 61278,
+      "▁dissip": 61279,
+      "serialization": 61280,
+      "泰安": 61281,
+      "实验组": 61282,
+      "你也可以": 61283,
+      "getContext": 61284,
+      "▁Employees": 61285,
+      "捷克": 61286,
+      "▁外观": 61287,
+      "收到的": 61288,
+      "第四款": 61289,
+      "中国当代": 61290,
+      "数据传输": 61291,
+      "▁Quint": 61292,
+      "▁spindle": 61293,
+      "▁Oncology": 61294,
+      "▁daunting": 61295,
+      "请告诉我您的选择和解题的思路": 61296,
+      "文华": 61297,
+      "更深": 61298,
+      "友好的": 61299,
+      ":\\\\/\\\\/": 61300,
+      "▁fierce": 61301,
+      "endocrine": 61302,
+      "▁WhatsApp": 61303,
+      "▁温": 61304,
+      "他与": 61305,
+      "/',": 61306,
+      "SVG": 61307,
+      "详情请": 61308,
+      "atro": 61309,
+      "仪器仪表": 61310,
+      "▁turf": 61311,
+      "▁folds": 61312,
+      "▁faulty": 61313,
+      "▁needles": 61314,
+      "▁specialised": 61315,
+      "丽丽": 61316,
+      "斜阳": 61317,
+      "昏迷": 61318,
+      "了一段": 61319,
+      "永远是": 61320,
+      "▁Bun": 61321,
+      "▁Wins": 61322,
+      "attrib": 61323,
+      "roidery": 61324,
+      "▁斌": 61325,
+      "一等": 61326,
+      "中国市场": 61327,
+      "▁Burk": 61328,
+      "▁Tort": 61329,
+      "必不可少的": 61330,
+      "tclient": 61331,
+      "佛罗": 61332,
+      "心想": 61333,
+      "感官": 61334,
+      "甜美": 61335,
+      "鄙视": 61336,
+      "하는": 61337,
+      "nwho": 61338,
+      "Disable": 61339,
+      "▁Prediction": 61340,
+      "UNC": 61341,
+      "gae": 61342,
+      "▁Rings": 61343,
+      "▁Aberdeen": 61344,
+      "上皮": 61345,
+      "关门": 61346,
+      "签收": 61347,
+      "需要进行": 61348,
+      "▁LIVE": 61349,
+      "astrophic": 61350,
+      "opulmonary": 61351,
+      "玉兰": 61352,
+      "石子": 61353,
+      "算出": 61354,
+      "原则上": 61355,
+      "▁marvel": 61356,
+      "▁Lambert": 61357,
+      "▁renting": 61358,
+      "▁sounded": 61359,
+      "▁quadrant": 61360,
+      "▁fertilizer": 61361,
+      "城建": 61362,
+      "很差": 61363,
+      "科夫": 61364,
+      "孙悟空": 61365,
+      "▁Rum": 61366,
+      "主要作品有": 61367,
+      "▁exits": 61368,
+      "stations": 61369,
+      "▁paragraphs": 61370,
+      "▁unacceptable": 61371,
+      "并做出分析": 61372,
+      "rectomy": 61373,
+      "▁Sketch": 61374,
+      "▁ambition": 61375,
+      "▁acetylcholine": 61376,
+      "初三": 61377,
+      "快手": 61378,
+      "mium": 61379,
+      "具有较强的": 61380,
+      "▁Packers": 61381,
+      "▁lifecycle": 61382,
+      "选择正确答案并做出分析": 61383,
+      "enol": 61384,
+      "身边的人": 61385,
+      "▁mour": 61386,
+      "▁Plains": 61387,
+      "▁hostile": 61388,
+      "▁cleaners": 61389,
+      "二甲": 61390,
+      "工农": 61391,
+      "握手": 61392,
+      "特邀": 61393,
+      "nits": 61394,
+      "▁Glor": 61395,
+      "tpanic": 61396,
+      "▁behaves": 61397,
+      "▁ambulatory": 61398,
+      "第二百六十二条第二款和": 61399,
+      "水下": 61400,
+      ":${": 61401,
+      "Fox": 61402,
+      "zie": 61403,
+      "迈克尔": 61404,
+      "▁osteoc": 61405,
+      "▁restrictive": 61406,
+      "▁Transactions": 61407,
+      "他被": 61408,
+      "并给出依据": 61409,
+      "▁consp": 61410,
+      "月子": 61411,
+      "ynb": 61412,
+      "服务费": 61413,
+      "服刑期间": 61414,
+      "emerg": 61415,
+      "nwindow": 61416,
+      "▁remedies": 61417,
+      "▁discovers": 61418,
+      "并给出依据和解题过程": 61419,
+      "军工": 61420,
+      "慕容": 61421,
+      "格雷": 61422,
+      "优化的": 61423,
+      "$\\',": 61424,
+      "▁Eat": 61425,
+      "atched": 61426,
+      "▁bombing": 61427,
+      "▁strengthened": 61428,
+      "▁UX": 61429,
+      "▁hairs": 61430,
+      "▁regularity": 61431,
+      "hY": 61432,
+      "登高": 61433,
+      "站着": 61434,
+      "说出来": 61435,
+      "builders": 61436,
+      "▁menstrual": 61437,
+      "▁disposable": 61438,
+      "高龄": 61439,
+      "加工厂": 61440,
+      "hass": 61441,
+      "hews": 61442,
+      "给人一种": 61443,
+      "▁Marqu": 61444,
+      "▁Inverse": 61445,
+      "▁제": 61446,
+      "不移": 61447,
+      "昔日": 61448,
+      "赃物": 61449,
+      "我们家": 61450,
+      "的水果": 61451,
+      "全神贯注": 61452,
+      "技术研究": 61453,
+      "haven": 61454,
+      "▁Piet": 61455,
+      "▁Talent": 61456,
+      "▁Monument": 61457,
+      "▁fidelity": 61458,
+      "▁fireworks": 61459,
+      "晓得": 61460,
+      "表示出": 61461,
+      "▁Isabel": 61462,
+      "▁Vertex": 61463,
+      "nAuthors": 61464,
+      "▁prosperity": 61465,
+      "▁reproducibility": 61466,
+      "史记": 61467,
+      "ARP": 61468,
+      "▁RL": 61469,
+      "时不时": 61470,
+      "Wild": 61471,
+      "▁Pastor": 61472,
+      "▁Valencia": 61473,
+      "后世": 61474,
+      "管理层": 61475,
+      "▁ICE": 61476,
+      "alese": 61477,
+      "▁Berks": 61478,
+      "▁Papua": 61479,
+      "nSingle": 61480,
+      "▁isoform": 61481,
+      "此外请给出你做出选择的依据和分析": 61482,
+      "一堂": 61483,
+      "为零": 61484,
+      "受力": 61485,
+      "手脚": 61486,
+      "地下水": 61487,
+      "▁Ade": 61488,
+      "▁LTE": 61489,
+      "chang": 61490,
+      "itemap": 61491,
+      "单位出具的财产": 61492,
+      "武夷": 61493,
+      "素有": 61494,
+      "听取了": 61495,
+      "ureth": 61496,
+      "▁Sail": 61497,
+      "▁paren": 61498,
+      "synthesize": 61499,
+      "▁Registered": 61500,
+      "▁distributing": 61501,
+      "よう": 61502,
+      "瘦身": 61503,
+      "经开": 61504,
+      "budget": 61505,
+      "csharp": 61506,
+      "ocative": 61507,
+      "生物科技有限公司": 61508,
+      "▁amendments": 61509,
+      "▁hypothetical": 61510,
+      "相望": 61511,
+      "pig": 61512,
+      "提供商": 61513,
+      "ymes": 61514,
+      "▁联赛等级": 61515,
+      "▁sheriff": 61516,
+      "prometheus": 61517,
+      "▁consultancy": 61518,
+      "不强": 61519,
+      "加成": 61520,
+      "okk": 61521,
+      "上诉称": 61522,
+      "惊人的": 61523,
+      "batim": 61524,
+      "eworthy": 61525,
+      "▁toddler": 61526,
+      "耗费": 61527,
+      "自知": 61528,
+      "十一条的规定": 61529,
+      "▁shocks": 61530,
+      "▁exerted": 61531,
+      "▁dishwasher": 61532,
+      "死人": 61533,
+      "高价": 61534,
+      "▁wk": 61535,
+      "nMer": 61536,
+      "▁ENA": 61537,
+      "nDepend": 61538,
+      "otically": 61539,
+      "对孩子": 61540,
+      "然后我": 61541,
+      "▁wre": 61542,
+      "▁Compet": 61543,
+      "▁Yankees": 61544,
+      "▁Restrict": 61545,
+      "▁editable": 61546,
+      "熟知": 61547,
+      "瓶子": 61548,
+      "动物的": 61549,
+      "的人工": 61550,
+      "社会化": 61551,
+      "grow": 61552,
+      "▁Lal": 61553,
+      "小伙伴们": 61554,
+      "▁Maya": 61555,
+      "nPlaces": 61556,
+      "▁mansion": 61557,
+      "拜访": 61558,
+      "▁nylon": 61559,
+      "▁Establish": 61560,
+      "全心": 61561,
+      "资本家": 61562,
+      "▁gastropod": 61563,
+      "▁institutes": 61564,
+      "▁manipulating": 61565,
+      "倒了": 61566,
+      "急剧": 61567,
+      "无脑": 61568,
+      "里亚": 61569,
+      "浓厚的": 61570,
+      "nMusicians": 61571,
+      "草堂": 61572,
+      "nPen": 61573,
+      "▁fals": 61574,
+      "代表作品有": 61575,
+      "▁Roberto": 61576,
+      "▁newborns": 61577,
+      "▁onChange": 61578,
+      "跳出": 61579,
+      "ctime": 61580,
+      "▁Cats": 61581,
+      "cognitive": 61582,
+      "▁pendulum": 61583,
+      "▁occupancy": 61584,
+      "上架": 61585,
+      "买单": 61586,
+      "svn": 61587,
+      "但不能": 61588,
+      "季后赛": 61589,
+      "无法送达": 61590,
+      "▁Edgar": 61591,
+      "▁wired": 61592,
+      "▁retract": 61593,
+      "▁ideology": 61594,
+      "▁eigenvectors": 61595,
+      "胶带": 61596,
+      "解密": 61597,
+      "项为": 61598,
+      "▁球队": 61599,
+      "无任何": 61600,
+      "聪明的": 61601,
+      "▁ovary": 61602,
+      "▁Fighting": 61603,
+      "修剪": 61604,
+      "小鱼": 61605,
+      "尾部": 61606,
+      "撒上": 61607,
+      "渐近": 61608,
+      "阻燃": 61609,
+      "建设中": 61610,
+      "管理所": 61611,
+      "acio": 61612,
+      "ServiceImpl": 61613,
+      "▁REFERENCES": 61614,
+      "吊顶": 61615,
+      "CAC": 61616,
+      "▁teas": 61617,
+      "▁apnea": 61618,
+      "▁merits": 61619,
+      "Analyzer": 61620,
+      "▁Capacity": 61621,
+      "▁Macedonia": 61622,
+      "石门": 61623,
+      "▁更新": 61624,
+      "单片机": 61625,
+      "▁grav": 61626,
+      "▁Evangel": 61627,
+      "万古": 61628,
+      "外商": 61629,
+      "早起": 61630,
+      "ONS": 61631,
+      "的人文": 61632,
+      "电池类型": 61633,
+      "cmath": 61634,
+      "▁Observer": 61635,
+      "▁hurricane": 61636,
+      "▁integrable": 61637,
+      "▁multiplayer": 61638,
+      "nLo": 61639,
+      "▁青春": 61640,
+      "▁Across": 61641,
+      "▁Autumn": 61642,
+      "GetValue": 61643,
+      "▁specifics": 61644,
+      "林区": 61645,
+      "stm": 61646,
+      "本研究": 61647,
+      "表面上": 61648,
+      "也可能是": 61649,
+      "▁KNOW": 61650,
+      "▁Crisis": 61651,
+      "▁stakes": 61652,
+      "增量": 61653,
+      "工期": 61654,
+      "nty": 61655,
+      "}>{": 61656,
+      "圣诞节": 61657,
+      "的特色": 61658,
+      "▁Wade": 61659,
+      "解题过程中的": 61660,
+      "▁follicular": 61661,
+      "台县": 61662,
+      "复方": 61663,
+      "异物": 61664,
+      "避孕": 61665,
+      "人可以": 61666,
+      "泰州市": 61667,
+      "▁ROI": 61668,
+      "▁Rho": 61669,
+      "京东图书": 61670,
+      "▁incurred": 61671,
+      "循序": 61672,
+      "收货": 61673,
+      "笑脸": 61674,
+      "申请撤回": 61675,
+      "inence": 61676,
+      "▁concurrently": 61677,
+      "无愧": 61678,
+      "脊椎": 61679,
+      "跳转": 61680,
+      "▁Yuan": 61681,
+      "RestController": 61682,
+      "NZ": 61683,
+      "▁ре": 61684,
+      "恢复正常": 61685,
+      "日向原告": 61686,
+      "nINSERT": 61687,
+      "▁Herman": 61688,
+      "Accounts": 61689,
+      "▁Broncos": 61690,
+      "▁Portable": 61691,
+      "五代": 61692,
+      "是没": 61693,
+      "点火": 61694,
+      "鱿鱼": 61695,
+      "ngx": 61696,
+      "bases": 61697,
+      "nLight": 61698,
+      "▁arcade": 61699,
+      "▁outbreaks": 61700,
+      "大使馆": 61701,
+      "FONT": 61702,
+      "▁Mitt": 61703,
+      "Banner": 61704,
+      "布袋": 61705,
+      "发展成为": 61706,
+      "▁Winston": 61707,
+      "▁Chancellor": 61708,
+      "▁geological": 61709,
+      "▁subcellular": 61710,
+      "议院": 61711,
+      "▁RH": 61712,
+      "的各项": 61713,
+      "APPS": 61714,
+      "Conv": 61715,
+      "▁Copa": 61716,
+      "▁enchant": 61717,
+      "▁salvage": 61718,
+      "▁Newfoundland": 61719,
+      "偏离": 61720,
+      "树枝": 61721,
+      "Elem": 61722,
+      "mens": 61723,
+      "▁PGE": 61724,
+      "电子信息": 61725,
+      "▁Wife": 61726,
+      "catenin": 61727,
+      "▁illusion": 61728,
+      "月第": 61729,
+      "美金": 61730,
+      "赃款": 61731,
+      ">](": 61732,
+      "▁Treg": 61733,
+      "pleasant": 61734,
+      "▁podcasts": 61735,
+      "解题过程中的依据是什么": 61736,
+      "▁descendants": 61737,
+      "性及": 61738,
+      "特地": 61739,
+      "UPS": 61740,
+      "是否会": 61741,
+      "osse": 61742,
+      "otopic": 61743,
+      "▁Himal": 61744,
+      "▁Apartments": 61745,
+      "ffi": 61746,
+      "ogi": 61747,
+      "可获得": 61748,
+      "的最好": 61749,
+      "了解更多": 61750,
+      "Ability": 61751,
+      "▁Farmer": 61752,
+      "▁Gardner": 61753,
+      "德惠": 61754,
+      "▁系列": 61755,
+      "十一五": 61756,
+      "和生产": 61757,
+      "罪犯杨": 61758,
+      "▁Grow": 61759,
+      "▁$${\\\\": 61760,
+      "stituted": 61761,
+      "separated": 61762,
+      "跟上": 61763,
+      "骗取": 61764,
+      "▁其七": 61765,
+      "事业部": 61766,
+      "winner": 61767,
+      "▁Performing": 61768,
+      "Plane": 61769,
+      "▁Nano": 61770,
+      "和分析的过程": 61771,
+      "chromium": 61772,
+      "▁Tribune": 61773,
+      "▁permitting": 61774,
+      "特许": 61775,
+      "讲课": 61776,
+      "▁Tian": 61777,
+      "▁Diane": 61778,
+      "▁Relay": 61779,
+      "nJoseph": 61780,
+      "不耐": 61781,
+      "尿素": 61782,
+      "为单位": 61783,
+      "表情包": 61784,
+      "abis": 61785,
+      "▁Guests": 61786,
+      "▁triumph": 61787,
+      "▁熊": 61788,
+      "女排": 61789,
+      "龙潭": 61790,
+      "住天津市": 61791,
+      "▁Fully": 61792,
+      "Awesome": 61793,
+      "▁Addition": 61794,
+      "奖品": 61795,
+      "Utf": 61796,
+      "的知识和": 61797,
+      "部副部长": 61798,
+      "▁Kard": 61799,
+      "Attach": 61800,
+      "▁Legends": 61801,
+      "▁tangible": 61802,
+      "▁computationally": 61803,
+      "书包": 61804,
+      "芝加": 61805,
+      "团队的": 61806,
+      "展现出": 61807,
+      "dock": 61808,
+      "▁homogen": 61809,
+      "互利": 61810,
+      "床垫": 61811,
+      "更低": 61812,
+      "砂浆": 61813,
+      "考的": 61814,
+      "舰队": 61815,
+      "不是我": 61816,
+      "深深的": 61817,
+      "heus": 61818,
+      "sell": 61819,
+      "spatial": 61820,
+      "▁cooled": 61821,
+      "▁rejuven": 61822,
+      "▁serializer": 61823,
+      "ック": 61824,
+      "插画": 61825,
+      "江水": 61826,
+      "nFive": 61827,
+      "▁ASCII": 61828,
+      "Elastic": 61829,
+      "nVersion": 61830,
+      "不育": 61831,
+      "病程": 61832,
+      "研究生学历": 61833,
+      "protective": 61834,
+      "▁proficiency": 61835,
+      "▁가": 61836,
+      "年头": 61837,
+      "是新": 61838,
+      "波浪": 61839,
+      "rak": 61840,
+      "是两个": 61841,
+      "行政区域": 61842,
+      "aspoon": 61843,
+      "▁Fiber": 61844,
+      "enstein": 61845,
+      "▁Detail": 61846,
+      "▁economists": 61847,
+      "▁antiretroviral": 61848,
+      "已成": 61849,
+      "闭合": 61850,
+      "▁Vote": 61851,
+      "▁usize": 61852,
+      "▁troubled": 61853,
+      "改判": 61854,
+      "春色": 61855,
+      "智库": 61856,
+      "歉意": 61857,
+      "保字第": 61858,
+      "▁buys": 61859,
+      "Ord": 61860,
+      "开庭传票": 61861,
+      "aiser": 61862,
+      "▁Trigger": 61863,
+      "▁decrypt": 61864,
+      "▁wilderness": 61865,
+      "KR": 61866,
+      "▁Р": 61867,
+      "全线": 61868,
+      "已然": 61869,
+      "月被": 61870,
+      "清远": 61871,
+      "▁paved": 61872,
+      "otrophic": 61873,
+      "▁getUser": 61874,
+      "▁Counties": 61875,
+      "▁smoothing": 61876,
+      "并告诉我您的选择和分析的过程": 61877,
+      "纽带": 61878,
+      "Arm": 61879,
+      "中医院": 61880,
+      "MPLE": 61881,
+      "全力以赴": 61882,
+      "牢记使命": 61883,
+      "wizard": 61884,
+      "▁Concord": 61885,
+      "▁Meaning": 61886,
+      "▁airflow": 61887,
+      "▁craving": 61888,
+      "▁flooded": 61889,
+      "▁bookings": 61890,
+      "▁caregiver": 61891,
+      "光度": 61892,
+      "染料": 61893,
+      "树上": 61894,
+      "ASA": 61895,
+      "tas": 61896,
+      "和解为由": 61897,
+      "otech": 61898,
+      "▁china": 61899,
+      "affinity": 61900,
+      "▁ceremonies": 61901,
+      "▁左": 61902,
+      "眉毛": 61903,
+      "北京的": 61904,
+      "就有了": 61905,
+      "最快的": 61906,
+      "要怎么": 61907,
+      "被执行人黄": 61908,
+      "▁Basil": 61909,
+      "▁Porsche": 61910,
+      "▁Exercises": 61911,
+      "▁travellers": 61912,
+      "其为": 61913,
+      "▁开始": 61914,
+      "的看法": 61915,
+      "并给出合理的解释": 61916,
+      "▁stringent": 61917,
+      "▁motivational": 61918,
+      "▁邱": 61919,
+      "▁Eh": 61920,
+      "▁主体": 61921,
+      "这里有": 61922,
+      "赔偿被害人": 61923,
+      "genesis": 61924,
+      "▁Tropical": 61925,
+      "业者": 61926,
+      "对前款": 61927,
+      "得益于": 61928,
+      "看电视": 61929,
+      "emma": 61930,
+      "nonce": 61931,
+      "▁Booth": 61932,
+      "▁Malcolm": 61933,
+      "并给出你的解题过程": 61934,
+      "▁coincidence": 61935,
+      "秋日": 61936,
+      "were": 61937,
+      "▁jelly": 61938,
+      "▁UPDATE": 61939,
+      "Capacity": 61940,
+      "Retention": 61941,
+      "▁landlord": 61942,
+      "Instrument": 61943,
+      "WG": 61944,
+      "节水": 61945,
+      "已交纳": 61946,
+      "谅解书": 61947,
+      "EXEC": 61948,
+      "vester": 61949,
+      "▁Herein": 61950,
+      "▁sporadic": 61951,
+      "惯性": 61952,
+      "打入": 61953,
+      "信息网": 61954,
+      "▁Communities": 61955,
+      "Pref": 61956,
+      "可以提高": 61957,
+      "▁oppose": 61958,
+      "▁sediments": 61959,
+      "刺客": 61960,
+      "沉降": 61961,
+      "▁综合": 61962,
+      "一季度": 61963,
+      "▁vib": 61964,
+      "中国政府": 61965,
+      "但是我们": 61966,
+      "nHouse": 61967,
+      "▁Driving": 61968,
+      "▁chlorine": 61969,
+      "▁inducible": 61970,
+      "▁Entrepreneur": 61971,
+      "型和": 61972,
+      "总站": 61973,
+      "洞庭": 61974,
+      "矿区": 61975,
+      "造林": 61976,
+      "ivered": 61977,
+      "▁Aging": 61978,
+      "imester": 61979,
+      "次要": 61980,
+      "茶匙": 61981,
+      "另一方": 61982,
+      "投入到": 61983,
+      "持卡人": 61984,
+      "芝加哥": 61985,
+      "]\\\\)": 61986,
+      "Undefined": 61987,
+      "▁silencing": 61988,
+      "独有": 61989,
+      "处理好": 61990,
+      "老人的": 61991,
+      "▁arrests": 61992,
+      "▁bronchial": 61993,
+      "▁preferably": 61994,
+      "然后告诉我您的最佳选择和分析思路": 61995,
+      "ldap": 61996,
+      "▁rall": 61997,
+      "emptyset": 61998,
+      "PrimaryKey": 61999,
+      "▁Wholesale": 62000,
+      "▁scalability": 62001,
+      "▁包": 62002,
+      "赛区": 62003,
+      "冻结的": 62004,
+      "irds": 62005,
+      "▁iodine": 62006,
+      "亿万": 62007,
+      "成败": 62008,
+      "*((": 62009,
+      "▁广州市": 62010,
+      "endra": 62011,
+      "▁Boyd": 62012,
+      "▁Escape": 62013,
+      "▁edible": 62014,
+      "▁cytosolic": 62015,
+      "白血": 62016,
+      "都属于": 62017,
+      "lista": 62018,
+      "phenyl": 62019,
+      "▁granular": 62020,
+      "签到": 62021,
+      "脱硫": 62022,
+      "问责": 62023,
+      "优良的": 62024,
+      "对他的": 62025,
+      "intern": 62026,
+      "purple": 62027,
+      "▁Nexus": 62028,
+      "▁Castro": 62029,
+      "▁Nearly": 62030,
+      "▁Welfare": 62031,
+      "nBasically": 62032,
+      "▁impacting": 62033,
+      "▁nc": 62034,
+      "十分钟": 62035,
+      "有需要": 62036,
+      "▁intrav": 62037,
+      "诉刑诉": 62038,
+      "二〇一七年三": 62039,
+      "escence": 62040,
+      "▁Screening": 62041,
+      "林场": 62042,
+      "比拟": 62043,
+      "航行": 62044,
+      "路桥": 62045,
+      "AES": 62046,
+      "努力的": 62047,
+      "这种事": 62048,
+      "目前正在": 62049,
+      "▁Huang": 62050,
+      "nString": 62051,
+      "mz": 62052,
+      "大卫": 62053,
+      "忠实": 62054,
+      "颜料": 62055,
+      "▁CMV": 62056,
+      "▁Extr": 62057,
+      "▁Qing": 62058,
+      "translated": 62059,
+      "GetComponent": 62060,
+      "元气": 62061,
+      "荷载": 62062,
+      "▁Fiji": 62063,
+      "▁Weak": 62064,
+      "abella": 62065,
+      "ilical": 62066,
+      "▁Released": 62067,
+      "effectiveness": 62068,
+      "▁transporters": 62069,
+      "kd": 62070,
+      "▁则": 62071,
+      "▼\\": 62072,
+      "双核": 62073,
+      "山坡": 62074,
+      "洗脸": 62075,
+      "读懂": 62076,
+      "增长的": 62077,
+      "建筑师": 62078,
+      "Plain": 62079,
+      "▁Indies": 62080,
+      "▁Vernon": 62081,
+      "未经法庭许可中途退庭的": 62082,
+      "转变为": 62083,
+      "▁Akt": 62084,
+      "▁主要参数": 62085,
+      "existence": 62086,
+      "president": 62087,
+      "SEQUENTIAL": 62088,
+      "▁Episcopal": 62089,
+      "冷酷": 62090,
+      "华润": 62091,
+      "hhh": 62092,
+      "以确保": 62093,
+      "和安全": 62094,
+      "更方便": 62095,
+      "的主人": 62096,
+      "▁sewer": 62097,
+      "▁Abbott": 62098,
+      "▁Packet": 62099,
+      "▁blender": 62100,
+      "本案在审理过程中": 62101,
+      "曾被": 62102,
+      "]<<": 62103,
+      "}`)\\": 62104,
+      "中国移动": 62105,
+      "但是现在": 62106,
+      "ivable": 62107,
+      "▁Saturn": 62108,
+      "regnancy": 62109,
+      "remaining": 62110,
+      "▁formulae": 62111,
+      "为凭": 62112,
+      "拷贝": 62113,
+      "甚么": 62114,
+      "Gas": 62115,
+      "时免费": 62116,
+      "imeo": 62117,
+      "undy": 62118,
+      "▁seper": 62119,
+      "差异无统计学意义": 62120,
+      "▁Tomorrow": 62121,
+      "亲朋": 62122,
+      "到此": 62123,
+      "定金": 62124,
+      "不等于": 62125,
+      "睡不着": 62126,
+      "密切相关": 62127,
+      "estry": 62128,
+      "▁Mast": 62129,
+      "▁Britt": 62130,
+      "▁steer": 62131,
+      "▁philosopher": 62132,
+      "@{": 62133,
+      "凹陷": 62134,
+      "▁Been": 62135,
+      "▁TODAY": 62136,
+      "▁keynote": 62137,
+      "▁valence": 62138,
+      "在西": 62139,
+      "鹏飞": 62140,
+      "▁HTC": 62141,
+      "制度改革": 62142,
+      "▁ribs": 62143,
+      "pecific": 62144,
+      "toutput": 62145,
+      "▁Wallet": 62146,
+      "nWinners": 62147,
+      "iculously": 62148,
+      "起诉必须符合下列条件": 62149,
+      "▁substitutions": 62150,
+      "Ty": 62151,
+      "并茂": 62152,
+      "解码": 62153,
+      "ctr": 62154,
+      "的这个": 62155,
+      "maxn": 62156,
+      "▁FSH": 62157,
+      "▁副标题": 62158,
+      "可能导致": 62159,
+      "核心价值观": 62160,
+      "▁Lessons": 62161,
+      "▁unhappy": 62162,
+      "following": 62163,
+      "其行为已构成盗窃罪": 62164,
+      "▁statutory": 62165,
+      "凄凉": 62166,
+      "方形": 62167,
+      "权属": 62168,
+      "离别": 62169,
+      "EEK": 62170,
+      "▁杭州": 62171,
+      "▁比赛": 62172,
+      "综合治理": 62173,
+      "设计方案": 62174,
+      "ORMAL": 62175,
+      "Parcel": 62176,
+      "▁holog": 62177,
+      "▁antigenic": 62178,
+      "花草": 62179,
+      "还很": 62180,
+      "cep": 62181,
+      "▁}:": 62182,
+      "gems": 62183,
+      "一点也不": 62184,
+      "分别求出": 62185,
+      "你选择的依据": 62186,
+      "affolds": 62187,
+      "▁Presbyterian": 62188,
+      "对口": 62189,
+      "山楂": 62190,
+      "饭菜": 62191,
+      "nMa": 62192,
+      "▁rv": 62193,
+      "▁Mey": 62194,
+      "儿童文学": 62195,
+      "共产主义": 62196,
+      "第二十五": 62197,
+      "交通警察大队": 62198,
+      "仔细地阅读选项": 62199,
+      "▁Strength": 62200,
+      "并给出你选择的依据": 62201,
+      "▁indispens": 62202,
+      "请您务必仔细地阅读选项": 62203,
+      "并给出你选择的依据和分析思路": 62204,
+      "命的": 62205,
+      "肯定的": 62206,
+      "节目中": 62207,
+      "要不然": 62208,
+      "▁pon": 62209,
+      "Gradient": 62210,
+      "八字": 62211,
+      "小幅": 62212,
+      "neck": 62213,
+      "rupts": 62214,
+      "▁stdin": 62215,
+      "▁tasked": 62216,
+      "scribers": 62217,
+      "▁Spatial": 62218,
+      "▁propensity": 62219,
+      "▁apprentices": 62220,
+      "作法": 62221,
+      "(\"_": 62222,
+      "充要条件": 62223,
+      "冠军联赛": 62224,
+      "stamp": 62225,
+      "▁distributors": 62226,
+      "▁血": 62227,
+      "alm": 62228,
+      "▁异界": 62229,
+      "没有被": 62230,
+      "NNNN": 62231,
+      "▁sparkling": 62232,
+      "师德": 62233,
+      "湖州": 62234,
+      "自动履行": 62235,
+      "▁broth": 62236,
+      "▁spelled": 62237,
+      "内径": 62238,
+      "和李": 62239,
+      "宇航": 62240,
+      "сти": 62241,
+      "经济学家": 62242,
+      "hybrid": 62243,
+      "nerror": 62244,
+      "▁bites": 62245,
+      "signals": 62246,
+      "一听": 62247,
+      "抗美": 62248,
+      "推拿": 62249,
+      "抵抗力": 62250,
+      "▁MAG": 62251,
+      "公司员工": 62252,
+      "uvian": 62253,
+      "▁Included": 62254,
+      "▁businesspeople": 62255,
+      "一角": 62256,
+      "封印": 62257,
+      "osx": 62258,
+      "人们在": 62259,
+      "科技的": 62260,
+      "等人的": 62261,
+      "故对原告": 62262,
+      "▁Wrap": 62263,
+      "进一步提升": 62264,
+      "之风": 62265,
+      "终生": 62266,
+      "者为": 62267,
+      "背影": 62268,
+      "ANA": 62269,
+      "Crypto": 62270,
+      "▁reluctant": 62271,
+      "ITT": 62272,
+      "glm": 62273,
+      "预交的": 62274,
+      "Void": 62275,
+      "placements": 62276,
+      "任性": 62277,
+      "台的": 62278,
+      "职能部门": 62279,
+      "▁gaze": 62280,
+      "解决的问题": 62281,
+      "Determ": 62282,
+      "lorida": 62283,
+      "▁lacked": 62284,
+      "▁sculptor": 62285,
+      "▁furnishings": 62286,
+      "kil": 62287,
+      "▁yy": 62288,
+      "在什么": 62289,
+      "能量的": 62290,
+      "镇江市": 62291,
+      "abit": 62292,
+      "▁WHEN": 62293,
+      "▁paradise": 62294,
+      "▁transistor": 62295,
+      "▁whitespace": 62296,
+      "海参": 62297,
+      "着实": 62298,
+      "额为": 62299,
+      "高明": 62300,
+      "▁Chow": 62301,
+      "▁Keeping": 62302,
+      "地貌": 62303,
+      "巨星": 62304,
+      "雄性": 62305,
+      "▁状态": 62306,
+      "出发点": 62307,
+      "线上的": 62308,
+      "▁$${": 62309,
+      "诗词正文": 62310,
+      "▁BACK": 62311,
+      "▁Straight": 62312,
+      "pinephrine": 62313,
+      "▁Practical": 62314,
+      "longrightarrow": 62315,
+      "AJ": 62316,
+      "亚纲": 62317,
+      "小品": 62318,
+      "曲面": 62319,
+      "贵宾": 62320,
+      "蓄电池": 62321,
+      "SCHE": 62322,
+      "TERS": 62323,
+      "otropin": 62324,
+      "▁dentists": 62325,
+      "▁neoplasms": 62326,
+      "七夕": 62327,
+      "使我": 62328,
+      "重现": 62329,
+      "arf": 62330,
+      "玩手机": 62331,
+      "非常大": 62332,
+      "nwin": 62333,
+      "▁Fry": 62334,
+      "Damage": 62335,
+      "userName": 62336,
+      "▁biofilm": 62337,
+      "俱全": 62338,
+      "则会": 62339,
+      "温差": 62340,
+      "耗材": 62341,
+      "身穿": 62342,
+      "阶段性": 62343,
+      "▁Sind": 62344,
+      "环境保护局": 62345,
+      "▁Cooperative": 62346,
+      "上岗": 62347,
+      "友善": 62348,
+      "▁Stern": 62349,
+      "▁mosque": 62350,
+      "▁sampler": 62351,
+      "authorize": 62352,
+      "setContent": 62353,
+      "▁lowercase": 62354,
+      "▁beforehand": 62355,
+      "getter": 62356,
+      "nSound": 62357,
+      "浙江省杭州市": 62358,
+      "UpperCase": 62359,
+      "▁pneumoniae": 62360,
+      "▁picturesque": 62361,
+      "Reb": 62362,
+      "可实现": 62363,
+      "文件中": 62364,
+      "quark": 62365,
+      "invariant": 62366,
+      "▁Awareness": 62367,
+      "▁protagonist": 62368,
+      "平面内": 62369,
+      "▁pys": 62370,
+      "由原告抚养": 62371,
+      "需要注意的是": 62372,
+      "▁confocal": 62373,
+      "▁passages": 62374,
+      "专员": 62375,
+      "也成为": 62376,
+      "言律诗": 62377,
+      "或者未经法庭许可中途退庭的": 62378,
+      "出栏": 62379,
+      "礼包": 62380,
+      "位置的": 62381,
+      "Fold": 62382,
+      "▁Strange": 62383,
+      "▁Deployment": 62384,
+      "▁governmental": 62385,
+      "可予": 62386,
+      "已将": 62387,
+      "老区": 62388,
+      "被我": 62389,
+      "Jav": 62390,
+      "▁Kö": 62391,
+      ">\\';\\": 62392,
+      "第七十二条": 62393,
+      "越来越多的人": 62394,
+      "Patient": 62395,
+      "▁Magnet": 62396,
+      "▁trailers": 62397,
+      "▁exhaustive": 62398,
+      "余家": 62399,
+      "双赢": 62400,
+      "太仓": 62401,
+      "补肾": 62402,
+      "过着": 62403,
+      "ingen": 62404,
+      "reements": 62405,
+      "想来": 62406,
+      "huang": 62407,
+      "▁Earn": 62408,
+      "arching": 62409,
+      "▁Terror": 62410,
+      "bindings": 62411,
+      "▁Objectives": 62412,
+      "想买": 62413,
+      "方法将": 62414,
+      "中小学生": 62415,
+      "Alloc": 62416,
+      "agnet": 62417,
+      "iffany": 62418,
+      "▁Pablo": 62419,
+      "新疆生产建设兵团": 62420,
+      "blueprint": 62421,
+      "▁Anonymous": 62422,
+      "▁unhealthy": 62423,
+      "功名": 62424,
+      "勾股": 62425,
+      "的发": 62426,
+      "是日本": 62427,
+      "比较适合": 62428,
+      "▁ureth": 62429,
+      "▁brewery": 62430,
+      "安检": 62431,
+      "肩负": 62432,
+      "轻型": 62433,
+      "截止到": 62434,
+      "Into": 62435,
+      "▁hemodialysis": 62436,
+      "用料": 62437,
+      "组分": 62438,
+      "家人的": 62439,
+      "让别人": 62440,
+      "▁Sard": 62441,
+      "履行债务的能力": 62442,
+      "▁Bedford": 62443,
+      "▁colonic": 62444,
+      "▁reliance": 62445,
+      "Functional": 62446,
+      "NullException": 62447,
+      "esm": 62448,
+      "iage": 62449,
+      "▁Kub": 62450,
+      "▁RBC": 62451,
+      "青年教师": 62452,
+      "lename": 62453,
+      "▁Osaka": 62454,
+      "▁educating": 62455,
+      "▁乌": 62456,
+      "在建": 62457,
+      "心头": 62458,
+      "ITER": 62459,
+      "▁sunk": 62460,
+      "▁Jorge": 62461,
+      "▁stride": 62462,
+      "忧伤": 62463,
+      "提拔": 62464,
+      "▁ff": 62465,
+      "垃圾桶": 62466,
+      "不等式的解集": 62467,
+      "▁crater": 62468,
+      "▁Interpret": 62469,
+      "接送": 62470,
+      "的情": 62471,
+      "中央党校": 62472,
+      "但实际上": 62473,
+      "▁Veget": 62474,
+      ",)": 62475,
+      "地道": 62476,
+      "影音": 62477,
+      "▁Ens": 62478,
+      "URCES": 62479,
+      "cobra": 62480,
+      "ieder": 62481,
+      "▁tidy": 62482,
+      "alculus": 62483,
+      "▁undergoes": 62484,
+      "滞后": 62485,
+      "▁本名": 62486,
+      "VENT": 62487,
+      "hits": 62488,
+      "▁LOT": 62489,
+      "▁Deer": 62490,
+      "的免费政策": 62491,
+      "lamide": 62492,
+      "追索劳动报酬": 62493,
+      "▁Andhra": 62494,
+      "▁cardiomyopathy": 62495,
+      "痉挛": 62496,
+      "ouve": 62497,
+      "一项住宿": 62498,
+      "▁Ways": 62499,
+      "forces": 62500,
+      "quotes": 62501,
+      "▁Pomer": 62502,
+      "▁Recipe": 62503,
+      "你最": 62504,
+      "品格": 62505,
+      "对新": 62506,
+      "世界观": 62507,
+      "的医疗": 62508,
+      "的成长": 62509,
+      "的高速": 62510,
+      "▁Forever": 62511,
+      "▁primate": 62512,
+      "▁Chambers": 62513,
+      "▁regiment": 62514,
+      "简明": 62515,
+      "远了": 62516,
+      "LOY": 62517,
+      "▁SAY": 62518,
+      "重点项目": 62519,
+      "▁Pierce": 62520,
+      "interpol": 62521,
+      "▁fountain": 62522,
+      "▁expatriates": 62523,
+      "书房": 62524,
+      "卖给": 62525,
+      "吃亏": 62526,
+      "无解": 62527,
+      "Cov": 62528,
+      "yam": 62529,
+      "木地板": 62530,
+      "本案系": 62531,
+      "正文语种": 62532,
+      "▁bifur": 62533,
+      "▁shining": 62534,
+      "▁fairness": 62535,
+      "一项住宿的免费政策": 62536,
+      "▁sectional": 62537,
+      "享有的债权依法受法律保护": 62538,
+      "三类": 62539,
+      "二字": 62540,
+      "工科": 62541,
+      "牙刷": 62542,
+      "]$,": 62543,
+      "▁ERK": 62544,
+      "▁jug": 62545,
+      "▁omn": 62546,
+      "提供被告": 62547,
+      "终结执行后": 62548,
+      "insulin": 62549,
+      "featured": 62550,
+      "卧龙": 62551,
+      "双边": 62552,
+      "菌株": 62553,
+      "飞跃": 62554,
+      "的文字": 62555,
+      "sink": 62556,
+      "empor": 62557,
+      "▁Beau": 62558,
+      "▁Stim": 62559,
+      "shapes": 62560,
+      "▁hurdles": 62561,
+      "▁Supporting": 62562,
+      "▁cartridges": 62563,
+      "单位出具的财产查询回执": 62564,
+      "拟合": 62565,
+      "Van": 62566,
+      "不停地": 62567,
+      "都能够": 62568,
+      "\":\"/": 62569,
+      "▁第二章": 62570,
+      "提请减刑": 62571,
+      "▁Trap": 62572,
+      "uretic": 62573,
+      "autical": 62574,
+      "▁inmates": 62575,
+      "▁jurisdict": 62576,
+      "并给出相应的解题过程": 62577,
+      "▁cardinality": 62578,
+      "▁commercials": 62579,
+      "咖喱": 62580,
+      "日日": 62581,
+      "赵丽": 62582,
+      "igu": 62583,
+      "五千元": 62584,
+      "知名品牌": 62585,
+      "chnic": 62586,
+      "nHenry": 62587,
+      "▁questionable": 62588,
+      "侦察": 62589,
+      "▁FTP": 62590,
+      "▁pyt": 62591,
+      "▁LOSS": 62592,
+      "sphinx": 62593,
+      "▁purse": 62594,
+      "artifacts": 62595,
+      "▁expansive": 62596,
+      "Nu": 62597,
+      "RAN": 62598,
+      "▁sugars": 62599,
+      "▁Advance": 62600,
+      "▁Matthews": 62601,
+      "▁empowerment": 62602,
+      ">`\\": 62603,
+      "结果单": 62604,
+      "请求撤销": 62605,
+      "▁SUCH": 62606,
+      "▁Demonstr": 62607,
+      "▁reusable": 62608,
+      "▁simplifies": 62609,
+      "恒星": 62610,
+      "Gene": 62611,
+      "baby": 62612,
+      "▁Ank": 62613,
+      "▁IgM": 62614,
+      "▁BEST": 62615,
+      "FAILURE": 62616,
+      "被执行人应当履行义务": 62617,
+      "▁Estimation": 62618,
+      "第二百五十七条第一款第": 62619,
+      "后因": 62620,
+      "志勇": 62621,
+      "或将": 62622,
+      "未曾": 62623,
+      "进村": 62624,
+      "造假": 62625,
+      "▁Amp": 62626,
+      "▁neurot": 62627,
+      "▁urging": 62628,
+      "computed": 62629,
+      "▁enduring": 62630,
+      "▁entirety": 62631,
+      "▁Geometric": 62632,
+      "atie": 62633,
+      "▁SHE": 62634,
+      "一线城市": 62635,
+      "▁Axis": 62636,
+      "▁Tumor": 62637,
+      "nclient": 62638,
+      "▁Manage": 62639,
+      "▁extremity": 62640,
+      "▁associative": 62641,
+      "▁好": 62642,
+      "上加": 62643,
+      "在后": 62644,
+      "▁GIF": 62645,
+      "管理体制": 62646,
+      "这个数是": 62647,
+      "nIndex": 62648,
+      "▁Middles": 62649,
+      "▁separates": 62650,
+      "▁施": 62651,
+      "演习": 62652,
+      "青菜": 62653,
+      "nll": 62654,
+      "▁Serg": 62655,
+      "▁牺牲情况": 62656,
+      "▁psoriasis": 62657,
+      "StringBuilder": 62658,
+      "填上": 62659,
+      "准备的": 62660,
+      "爱上了": 62661,
+      "Steve": 62662,
+      "omegal": 62663,
+      "▁hypoxic": 62664,
+      "▁Professionals": 62665,
+      "朦胧": 62666,
+      "画出": 62667,
+      "订立": 62668,
+      "早期的": 62669,
+      "▁Infl": 62670,
+      "▁rRNA": 62671,
+      "▁boson": 62672,
+      "本院依法受理": 62673,
+      "XM": 62674,
+      "fq": 62675,
+      "いた": 62676,
+      "飘飘": 62677,
+      "儿子的": 62678,
+      "▁UDP": 62679,
+      "市场上的": 62680,
+      "Trust": 62681,
+      "▁Andr": 62682,
+      "▁occupies": 62683,
+      "nFurthermore": 62684,
+      "▁confidentiality": 62685,
+      "国军": 62686,
+      "在手": 62687,
+      "篇幅": 62688,
+      "苦恼": 62689,
+      "▁`\"": 62690,
+      "imple": 62691,
+      "▁HDMI": 62692,
+      "先进工作者": 62693,
+      "▁Grill": 62694,
+      "▁agile": 62695,
+      "服饰有限公司": 62696,
+      "▁NASCAR": 62697,
+      "▁Sundays": 62698,
+      "又叫": 62699,
+      "平湖": 62700,
+      "有道": 62701,
+      "计算的": 62702,
+      "大大提高": 62703,
+      "VERTI": 62704,
+      "▁oder": 62705,
+      "smouth": 62706,
+      "▁Errors": 62707,
+      "▁asyncio": 62708,
+      "下发": 62709,
+      "乳胶": 62710,
+      "可调": 62711,
+      "放映": 62712,
+      "立式": 62713,
+      "的房产": 62714,
+      "的看着": 62715,
+      "Harm": 62716,
+      "▁Lad": 62717,
+      "▁Chance": 62718,
+      "▁contral": 62719,
+      "▁Martinez": 62720,
+      "Ale": 62721,
+      "分式的": 62722,
+      "建设银行": 62723,
+      "紧紧围绕": 62724,
+      "PROTO": 62725,
+      "▁Kafka": 62726,
+      "▁geometrical": 62727,
+      "ши": 62728,
+      "由其": 62729,
+      "▁spikes": 62730,
+      "月山": 62731,
+      "秒杀": 62732,
+      "tte": 62733,
+      "完之后": 62734,
+      "eren": 62735,
+      "▁Buk": 62736,
+      "经营理念": 62737,
+      "▁Mohammad": 62738,
+      "▁elongation": 62739,
+      "不入": 62740,
+      "谁说": 62741,
+      "已经很": 62742,
+      "OLED": 62743,
+      "ictionaries": 62744,
+      "华硕": 62745,
+      "矿井": 62746,
+      "网店": 62747,
+      "鲤鱼": 62748,
+      "▁mv": 62749,
+      "corp": 62750,
+      "故意杀人": 62751,
+      "程序设计": 62752,
+      "自我介绍": 62753,
+      "▁floats": 62754,
+      "▁unauthorized": 62755,
+      "八百": 62756,
+      "张玉": 62757,
+      "意料": 62758,
+      "nDev": 62759,
+      "virt": 62760,
+      "相关信息": 62761,
+      "▁lawsuits": 62762,
+      "▁博": 62763,
+      "容忍": 62764,
+      "就将": 62765,
+      "满天": 62766,
+      "该剧": 62767,
+      "黑夜": 62768,
+      "SUS": 62769,
+      "nfl": 62770,
+      "▁то": 62771,
+      "可能需要": 62772,
+      "家庭聚会": 62773,
+      "\\\\_\\\\_": 62774,
+      "▁popped": 62775,
+      "nPrimary": 62776,
+      "▁graphql": 62777,
+      "▁violate": 62778,
+      "▁servants": 62779,
+      "▁Democracy": 62780,
+      "▁disabling": 62781,
+      "▁vertebrate": 62782,
+      "▁accreditation": 62783,
+      "▁艳": 62784,
+      "你把": 62785,
+      "实景": 62786,
+      "]])": 62787,
+      "▁平面": 62788,
+      "Shot": 62789,
+      "oble": 62790,
+      "atase": 62791,
+      "manufact": 62792,
+      "▁Underground": 62793,
+      "裂片": 62794,
+      "Tabs": 62795,
+      "idades": 62796,
+      "缩写": 62797,
+      "▁PBS": 62798,
+      "集成电路": 62799,
+      "IGENCE": 62800,
+      "▁jealous": 62801,
+      "▁McCarthy": 62802,
+      "kj": 62803,
+      "▁ssl": 62804,
+      "uggle": 62805,
+      "▁Yard": 62806,
+      "床头": 62807,
+      "晚了": 62808,
+      "某人": 62809,
+      "选秀": 62810,
+      "城市化": 62811,
+      "守门员": 62812,
+      "过渡期": 62813,
+      "goods": 62814,
+      "▁NEED": 62815,
+      "Developer": 62816,
+      "志军": 62817,
+      "Feb": 62818,
+      "描述和": 62819,
+      "最佳的": 62820,
+      "▁Wen": 62821,
+      "fasta": 62822,
+      "▁pooled": 62823,
+      "thinking": 62824,
+      "▁hybrids": 62825,
+      "团伙": 62826,
+      "手下": 62827,
+      "杂物": 62828,
+      "爱与": 62829,
+      "无限的": 62830,
+      "是利用": 62831,
+      "分享一下": 62832,
+      "ologne": 62833,
+      "▁stole": 62834,
+      "orderBy": 62835,
+      "▁shaded": 62836,
+      "▁slowed": 62837,
+      "▁ambulance": 62838,
+      "他从": 62839,
+      "实施的": 62840,
+      "已超过": 62841,
+      "所学校": 62842,
+      "竞争优势": 62843,
+      "stripe": 62844,
+      "▁Shows": 62845,
+      "▁poured": 62846,
+      "▁Prairie": 62847,
+      "▁goats": 62848,
+      "▁novelty": 62849,
+      "▁spectrometer": 62850,
+      "▁ا": 62851,
+      "日式": 62852,
+      "甲级": 62853,
+      "阿克": 62854,
+      "的大量": 62855,
+      "chunks": 62856,
+      "nOrder": 62857,
+      "给出解题步骤": 62858,
+      "压实": 62859,
+      "年鉴": 62860,
+      "你要是": 62861,
+      "地面积": 62862,
+      ";;;;": 62863,
+      "▁Kane": 62864,
+      "ledger": 62865,
+      "▁liberty": 62866,
+      "▁Yugoslavia": 62867,
+      "未发现被执行人有可供执行的财产": 62868,
+      "↓\\": 62869,
+      "▁陶": 62870,
+      "节气": 62871,
+      "质押": 62872,
+      "仅次于": 62873,
+      "Bearer": 62874,
+      "▁九": 62875,
+      "勃勃": 62876,
+      "开裂": 62877,
+      "交换机": 62878,
+      "▁ASS": 62879,
+      "▁Rack": 62880,
+      "▁peroxidase": 62881,
+      "效果的": 62882,
+      "abine": 62883,
+      "nbegin": 62884,
+      "nFellows": 62885,
+      "▁legends": 62886,
+      "合数": 62887,
+      "村位于": 62888,
+      "重要讲话": 62889,
+      "Devices": 62890,
+      "▁deferred": 62891,
+      "▁withdrew": 62892,
+      "适应性": 62893,
+      ":\")\\": 62894,
+      "nTeams": 62895,
+      "▁Beast": 62896,
+      "olerant": 62897,
+      "plified": 62898,
+      "剪刀": 62899,
+      "寿司": 62900,
+      "且原告": 62901,
+      "也因此": 62902,
+      "分别占": 62903,
+      "Sink": 62904,
+      "eced": 62905,
+      "▁CDs": 62906,
+      "怎么说呢": 62907,
+      "ARRAY": 62908,
+      "david": 62909,
+      "jected": 62910,
+      "getattr": 62911,
+      "▁swimmer": 62912,
+      "Unix": 62913,
+      "Website": 62914,
+      "▁ration": 62915,
+      "preserve": 62916,
+      "▁homeowner": 62917,
+      "是出于": 62918,
+      "▁吹风机": 62919,
+      "▁Doesn": 62920,
+      "DataSet": 62921,
+      "usement": 62922,
+      "交于": 62923,
+      "茉莉": 62924,
+      "nStatus": 62925,
+      "▁premiums": 62926,
+      "政委": 62927,
+      "比对": 62928,
+      "hir": 62929,
+      "▁世界": 62930,
+      "离婚后": 62931,
+      "jord": 62932,
+      "NAMESPACE": 62933,
+      "▁Disorder": 62934,
+      "▁illegally": 62935,
+      "YC": 62936,
+      "风尘": 62937,
+      "SURE": 62938,
+      "etus": 62939,
+      "严重后果": 62940,
+      "stalk": 62941,
+      "young": 62942,
+      "▁Felix": 62943,
+      "▁downside": 62944,
+      "▁grafting": 62945,
+      "▁violating": 62946,
+      "没过": 62947,
+      "▁VL": 62948,
+      "中输入": 62949,
+      "就是说": 62950,
+      "是一座": 62951,
+      "能做到": 62952,
+      "黑色素": 62953,
+      "▁jog": 62954,
+      "投资管理": 62955,
+      "▁harbour": 62956,
+      "▁prolific": 62957,
+      "保税": 62958,
+      "支付的": 62959,
+      "Made": 62960,
+      "nroot": 62961,
+      "vocab": 62962,
+      "▁柳": 62963,
+      "受过": 62964,
+      "小腿": 62965,
+      "成才": 62966,
+      "手写": 62967,
+      "攻读": 62968,
+      "野蛮": 62969,
+      "是全球": 62970,
+      "第五届": 62971,
+      "▁PIL": 62972,
+      "优惠政策": 62973,
+      "调解无效": 62974,
+      "▁Pune": 62975,
+      "▁tart": 62976,
+      "nnumber": 62977,
+      "▁Baltic": 62978,
+      "一模": 62979,
+      "只用": 62980,
+      "战神": 62981,
+      "马蹄": 62982,
+      "的话语": 62983,
+      "cult": 62984,
+      "dorf": 62985,
+      "ncomp": 62986,
+      "▁axon": 62987,
+      "▁}}>\\": 62988,
+      "二〇一八年四月": 62989,
+      "▁quarant": 62990,
+      "▁Carnegie": 62991,
+      "产区": 62992,
+      "山谷": 62993,
+      "只知道": 62994,
+      "nWas": 62995,
+      "▁pari": 62996,
+      "▁tally": 62997,
+      "▁uterus": 62998,
+      "discount": 62999,
+      "solutions": 63000,
+      "▁carriage": 63001,
+      "▁commanded": 63002,
+      "▁testicular": 63003,
+      "做强": 63004,
+      "出新": 63005,
+      "同桌": 63006,
+      "张小": 63007,
+      "▁标准": 63008,
+      "如果我": 63009,
+      "蓝色的": 63010,
+      "得出结论": 63011,
+      "nFriday": 63012,
+      "yb": 63013,
+      "协力": 63014,
+      "叛逆": 63015,
+      "把控": 63016,
+      "贮存": 63017,
+      "马的": 63018,
+      "驾校": 63019,
+      "垂直于": 63020,
+      "Want": 63021,
+      "▁bead": 63022,
+      "▁Tissue": 63023,
+      "有限责任公司于": 63024,
+      "▁missiles": 63025,
+      "大白": 63026,
+      "捍卫": 63027,
+      "教育学": 63028,
+      "亿元人民币": 63029,
+      "局党组成员": 63030,
+      "facial": 63031,
+      "▁antiqu": 63032,
+      ">;": 63033,
+      "低音": 63034,
+      "求证": 63035,
+      "rito": 63036,
+      "缓期二年": 63037,
+      "nRural": 63038,
+      "▁embro": 63039,
+      "▁Offset": 63040,
+      "▁Cameroon": 63041,
+      "中度": 63042,
+      "或少": 63043,
+      "考了": 63044,
+      "()]": 63045,
+      "ratic": 63046,
+      "▁Sharing": 63047,
+      "▁PowerShell": 63048,
+      "▁정": 63049,
+      "多年生": 63050,
+      "nfil": 63051,
+      "附属医院": 63052,
+      "▁dams": 63053,
+      "atosis": 63054,
+      "▁Basel": 63055,
+      "▁axonal": 63056,
+      "▁hypogly": 63057,
+      "▁inductive": 63058,
+      "▁xenograft": 63059,
+      "▁restructuring": 63060,
+      "事事": 63061,
+      "兰特": 63062,
+      "脉络": 63063,
+      "rek": 63064,
+      "为保证": 63065,
+      "泸州市": 63066,
+      "这四个": 63067,
+      "rein": 63068,
+      "内容丰富": 63069,
+      "法刑初字第": 63070,
+      "▁denial": 63071,
+      "▁suitability": 63072,
+      "上班族": 63073,
+      "张家界": 63074,
+      "拥有了": 63075,
+      "rello": 63076,
+      "▁LEDs": 63077,
+      "[<": 63078,
+      "分配律": 63079,
+      "生产工艺": 63080,
+      "▁Bark": 63081,
+      "时间复杂度": 63082,
+      "▁Doctrine": 63083,
+      "▁Sunshine": 63084,
+      "▁Restoration": 63085,
+      "▁Instructions": 63086,
+      "归于": 63087,
+      "毛发": 63088,
+      "WHO": 63089,
+      "的季节": 63090,
+      "请写出": 63091,
+      "人际关系": 63092,
+      "▁Audit": 63093,
+      "▁grids": 63094,
+      "▁shout": 63095,
+      "▁Neumann": 63096,
+      "itutional": 63097,
+      "▁assigning": 63098,
+      "cobacterium": 63099,
+      "倘若": 63100,
+      "天门": 63101,
+      "太少": 63102,
+      "悄然": 63103,
+      "海峰": 63104,
+      "ncp": 63105,
+      "uant": 63106,
+      "ylic": 63107,
+      "orgia": 63108,
+      "Vehicle": 63109,
+      "iership": 63110,
+      "▁checkbox": 63111,
+      "▁sketches": 63112,
+      "▁schooling": 63113,
+      "测算": 63114,
+      "相册": 63115,
+      "记为": 63116,
+      "\"\">": 63117,
+      "▁如何": 63118,
+      "在台湾": 63119,
+      "iera": 63120,
+      "两边同时": 63121,
+      "本院审查": 63122,
+      "debian": 63123,
+      "owards": 63124,
+      "农村经济总收入": 63125,
+      "decorator": 63126,
+      "▁Synthesis": 63127,
+      "▁exercising": 63128,
+      "▁cx": 63129,
+      "展开了": 63130,
+      "(\\'_": 63131,
+      "提供不出": 63132,
+      "英语学习": 63133,
+      "▁cellar": 63134,
+      "了么": 63135,
+      "性质的": 63136,
+      "▁Wan": 63137,
+      "▁ants": 63138,
+      "经常居住地": 63139,
+      "上万": 63140,
+      "年期": 63141,
+      "每条": 63142,
+      "▁Yo": 63143,
+      "ecker": 63144,
+      "userData": 63145,
+      "实测": 63146,
+      "译文": 63147,
+      ")\\',": 63148,
+      "▁[\\\\": 63149,
+      "▁rpm": 63150,
+      "planet": 63151,
+      "toList": 63152,
+      "endance": 63153,
+      "二〇一八年三月": 63154,
+      "▁Č": 63155,
+      "元宝": 63156,
+      "婚生子": 63157,
+      "眼前的": 63158,
+      "▁CCR": 63159,
+      "▁Autonomous": 63160,
+      "▁worksheets": 63161,
+      "英超": 63162,
+      "主办方": 63163,
+      "give": 63164,
+      "▁cured": 63165,
+      "夫妻共同财产": 63166,
+      "nNothing": 63167,
+      "▁HttpClient": 63168,
+      "ocompatibility": 63169,
+      "NW": 63170,
+      "▁初": 63171,
+      "边际": 63172,
+      "橄榄油": 63173,
+      "▁Berm": 63174,
+      "linalg": 63175,
+      "▁flats": 63176,
+      "▁myeloma": 63177,
+      "▁cholinergic": 63178,
+      "停电": 63179,
+      "纯正": 63180,
+      "针刺": 63181,
+      "▁Param": 63182,
+      "▁Pascal": 63183,
+      "▁invariants": 63184,
+      "理发": 63185,
+      "腹黑": 63186,
+      "颇为": 63187,
+      "任务的": 63188,
+      "▁Hyundai": 63189,
+      "tassertEquals": 63190,
+      "反响": 63191,
+      "扇形": 63192,
+      "汽车的": 63193,
+      "Invoice": 63194,
+      "debugger": 63195,
+      "可选择": 63196,
+      "Road": 63197,
+      "我们知道": 63198,
+      "执行法院": 63199,
+      "lords": 63200,
+      "培养学生的": 63201,
+      "指定期限内": 63202,
+      "▁rhetoric": 63203,
+      "引力": 63204,
+      "白金": 63205,
+      "黄冈": 63206,
+      "ithe": 63207,
+      "与时俱进": 63208,
+      "▁sticker": 63209,
+      "}÷": 63210,
+      "出头": 63211,
+      "苦瓜": 63212,
+      "边上": 63213,
+      "年代初": 63214,
+      "采用的是": 63215,
+      "▁quin": 63216,
+      "▁hugely": 63217,
+      "滋补": 63218,
+      "相撞": 63219,
+      "NLP": 63220,
+      "▁空调": 63221,
+      "发出了": 63222,
+      "我怎么": 63223,
+      "nAug": 63224,
+      "▁TiO": 63225,
+      "专家学者": 63226,
+      "emorrh": 63227,
+      "osaurs": 63228,
+      "▁useEffect": 63229,
+      "想在": 63230,
+      "▁gy": 63231,
+      "最容易": 63232,
+      "货物运输": 63233,
+      "相等的实数根": 63234,
+      "▁Pepper": 63235,
+      "▁thicker": 63236,
+      "provision": 63237,
+      "▁Beginning": 63238,
+      "申请执行人向本院申请执行": 63239,
+      "不小": 63240,
+      "视察": 63241,
+      "第三百零": 63242,
+      "▁Judy": 63243,
+      "▁melody": 63244,
+      "immers": 63245,
+      "▁Publishers": 63246,
+      "拳击": 63247,
+      "管线": 63248,
+      "赞誉": 63249,
+      "▁城市": 63250,
+      "便利店": 63251,
+      "▁possessed": 63252,
+      "和无": 63253,
+      "密的": 63254,
+      "ecan": 63255,
+      "▁undo": 63256,
+      "▁VERSION": 63257,
+      "▁goodbye": 63258,
+      "援朝": 63259,
+      "送去": 63260,
+      "为提高": 63261,
+      "▁Tou": 63262,
+      "利害关系": 63263,
+      "总有效率": 63264,
+      "iscus": 63265,
+      "▁Mystery": 63266,
+      "教委": 63267,
+      "衣橱": 63268,
+      "赞扬": 63269,
+      "Combo": 63270,
+      "▁Omaha": 63271,
+      "multipart": 63272,
+      "供求": 63273,
+      "前身": 63274,
+      "均无": 63275,
+      "有明显": 63276,
+      "比较少": 63277,
+      "验证码": 63278,
+      "nuts": 63279,
+      "tuce": 63280,
+      "▁dug": 63281,
+      "▁CHANGE": 63282,
+      "▁awaken": 63283,
+      "▁siding": 63284,
+      "urrection": 63285,
+      "▁subspecies": 63286,
+      "吃药": 63287,
+      "征税": 63288,
+      "最受": 63289,
+      "▁Adapt": 63290,
+      "民事行为能力": 63291,
+      "▁Alembic": 63292,
+      "▁INDIRECT": 63293,
+      "同时对": 63294,
+      "但是由于": 63295,
+      "nIdent": 63296,
+      "▁kings": 63297,
+      "▁inverter": 63298,
+      "▁Huntington": 63299,
+      "*\"": 63300,
+      "岁以下": 63301,
+      "试验区": 63302,
+      "▁RPG": 63303,
+      "nService": 63304,
+      "Definitions": 63305,
+      "她会": 63306,
+      "流速": 63307,
+      "Tel": 63308,
+      "oran": 63309,
+      "hemer": 63310,
+      "人口与计划生育": 63311,
+      "▁mosquit": 63312,
+      "$\"": 63313,
+      "之歌": 63314,
+      "写得": 63315,
+      "此举": 63316,
+      "衡水": 63317,
+      "进水": 63318,
+      "全覆盖": 63319,
+      "而导致": 63320,
+      "umers": 63321,
+      "南县人民法院": 63322,
+      "rystals": 63323,
+      "人头": 63324,
+      "旧的": 63325,
+      "破了": 63326,
+      "中国国家": 63327,
+      "为自己的": 63328,
+      "Pager": 63329,
+      "erdin": 63330,
+      "privacy": 63331,
+      "▁simplifying": 63332,
+      "景气": 63333,
+      "甲醇": 63334,
+      "EEP": 63335,
+      "湖州市": 63336,
+      "生产厂家": 63337,
+      "dragon": 63338,
+      "▁mater": 63339,
+      "▁wholes": 63340,
+      "%^": 63341,
+      "预估": 63342,
+      "是否是": 63343,
+      "wolf": 63344,
+      "高级教师": 63345,
+      "▁nanoc": 63346,
+      "▁oversee": 63347,
+      "大展": 63348,
+      "ankar": 63349,
+      "regexp": 63350,
+      "▁ATLAS": 63351,
+      "nQuestions": 63352,
+      "▁appraisal": 63353,
+      "▁kindergarten": 63354,
+      "探寻": 63355,
+      "极易": 63356,
+      "格栅": 63357,
+      "特么": 63358,
+      "诗句": 63359,
+      "BOOK": 63360,
+      "▁boxer": 63361,
+      "nRemove": 63362,
+      "▁bundled": 63363,
+      "滚滚": 63364,
+      "趋向": 63365,
+      "MCA": 63366,
+      "vian": 63367,
+      "▁Lima": 63368,
+      "▁contestants": 63369,
+      "卫计": 63370,
+      "台南": 63371,
+      "就从": 63372,
+      "浸润": 63373,
+      "腿部": 63374,
+      "的母亲": 63375,
+      "▁что": 63376,
+      "即可开始游戏": 63377,
+      "▁blinds": 63378,
+      "云山": 63379,
+      "协和": 63380,
+      "的圆": 63381,
+      "von": 63382,
+      "的商业": 63383,
+      "车服务": 63384,
+      "非常高": 63385,
+      "▁RPM": 63386,
+      "▁bids": 63387,
+      "iliate": 63388,
+      "▁beast": 63389,
+      "ximation": 63390,
+      "nHistoric": 63391,
+      "BufferedReader": 63392,
+      "之久": 63393,
+      "一次次": 63394,
+      "却发现": 63395,
+      "商丘市": 63396,
+      "▁Hed": 63397,
+      "犯诈骗罪": 63398,
+      "Earth": 63399,
+      "toast": 63400,
+      "uding": 63401,
+      "Discovery": 63402,
+      "▁annealing": 63403,
+      "何某": 63404,
+      "先前": 63405,
+      "▁<>": 63406,
+      "nGod": 63407,
+      "sand": 63408,
+      "tpath": 63409,
+      "▁evap": 63410,
+      "Anchor": 63411,
+      "就餐": 63412,
+      "雷霆": 63413,
+      "*)\\": 63414,
+      "oti": 63415,
+      "为指导": 63416,
+      "在企业": 63417,
+      "证明原": 63418,
+      "slots": 63419,
+      "nPrivate": 63420,
+      "第七十八条之规定": 63421,
+      "▁Provides": 63422,
+      "▁Blockchain": 63423,
+      "▁percentile": 63424,
+      "和为": 63425,
+      "灵芝": 63426,
+      "魅族": 63427,
+      "也有人": 63428,
+      "转化成": 63429,
+      "nDetermine": 63430,
+      "vf": 63431,
+      "春华": 63432,
+      "罐头": 63433,
+      "在现代": 63434,
+      "▁sdk": 63435,
+      "▁vid": 63436,
+      "▁biking": 63437,
+      "等号": 63438,
+      "闪闪": 63439,
+      "SPA": 63440,
+      "的原理": 63441,
+      "过一个": 63442,
+      "教学改革": 63443,
+      "dison": 63444,
+      "▁Mits": 63445,
+      "▁actu": 63446,
+      "并支付利息": 63447,
+      "轴对称图形": 63448,
+      "corpus": 63449,
+      "▁hydrochlor": 63450,
+      "重温": 63451,
+      "本文以": 63452,
+      "不同程度": 63453,
+      "imuth": 63454,
+      "▁Rafael": 63455,
+      "▁Heating": 63456,
+      "▁orbitals": 63457,
+      "standalone": 63458,
+      "▁intensities": 63459,
+      "育种": 63460,
+      "该产品": 63461,
+      "那么大": 63462,
+      "INET": 63463,
+      "psin": 63464,
+      "Paint": 63465,
+      "▁Tweet": 63466,
+      "▁voucher": 63467,
+      "YLE": 63468,
+      "exas": 63469,
+      "实实在在": 63470,
+      "▁Stuff": 63471,
+      "nIndeed": 63472,
+      "▁Timber": 63473,
+      "▁onsite": 63474,
+      "▁Printing": 63475,
+      "主意": 63476,
+      "你别": 63477,
+      "▁需要": 63478,
+      "工作量": 63479,
+      "Club": 63480,
+      "上述证据": 63481,
+      "▁Hawks": 63482,
+      "▁Macro": 63483,
+      "▁Graphic": 63484,
+      "▁eighteen": 63485,
+      "申请执行人可向本院申请恢复执行": 63486,
+      "当了": 63487,
+      "不重要": 63488,
+      "推动了": 63489,
+      "的情感": 63490,
+      "达成的": 63491,
+      "aban": 63492,
+      "URITY": 63493,
+      "▁Loans": 63494,
+      "gravity": 63495,
+      "▁imprint": 63496,
+      "▁Precision": 63497,
+      "▁biodegrad": 63498,
+      "▁Dependency": 63499,
+      "天安": 63500,
+      "应收": 63501,
+      "Ten": 63502,
+      "经原告": 63503,
+      "iren": 63504,
+      "临床表现": 63505,
+      "Popen": 63506,
+      "nPain": 63507,
+      "▁cheat": 63508,
+      "autions": 63509,
+      "▁sludge": 63510,
+      "▁mustard": 63511,
+      "▁Oriental": 63512,
+      "▁electrom": 63513,
+      "IPP": 63514,
+      "新陈代谢": 63515,
+      "方法采用": 63516,
+      "weeks": 63517,
+      "▁Oman": 63518,
+      "▁drains": 63519,
+      "projection": 63520,
+      "一旁": 63521,
+      "所含": 63522,
+      "生源": 63523,
+      "退化": 63524,
+      "不接受": 63525,
+      "手机号": 63526,
+      "的可以": 63527,
+      "hoff": 63528,
+      "▁cease": 63529,
+      "▁boxers": 63530,
+      "▁bidding": 63531,
+      "potential": 63532,
+      "减退": 63533,
+      "靠自己": 63534,
+      "Wind": 63535,
+      "alus": 63536,
+      "▁演员表": 63537,
+      "THING": 63538,
+      "ético": 63539,
+      "▁Rear": 63540,
+      "▁dirname": 63541,
+      "Analytics": 63542,
+      "▁broadcaster": 63543,
+      "灭亡": 63544,
+      "让其": 63545,
+      "yla": 63546,
+      "还有个": 63547,
+      "perg": 63548,
+      "▁Alf": 63549,
+      "itespace": 63550,
+      "二〇一四年十一月": 63551,
+      "yma": 63552,
+      "▁bisc": 63553,
+      "▁tint": 63554,
+      "▁Sexual": 63555,
+      "ivy": 63556,
+      "多领域": 63557,
+      "▁Dip": 63558,
+      "orian": 63559,
+      "▁Balk": 63560,
+      "▁sind": 63561,
+      "Combin": 63562,
+      "STATIC": 63563,
+      "▁Playing": 63564,
+      "▁Marriage": 63565,
+      "▁implying": 63566,
+      "▁screaming": 63567,
+      "Persistence": 63568,
+      "保费": 63569,
+      "升值": 63570,
+      "晴天": 63571,
+      "集镇": 63572,
+      "与环境": 63573,
+      "发明专利": 63574,
+      "年轻人的": 63575,
+      "oulos": 63576,
+      "▁reps": 63577,
+      "xin": 63578,
+      "不考虑": 63579,
+      "并处没收": 63580,
+      "技术研发": 63581,
+      "短篇小说": 63582,
+      "Posted": 63583,
+      "estation": 63584,
+      "▁plastics": 63585,
+      "▁filaments": 63586,
+      "▁postponed": 63587,
+      "▁sunglasses": 63588,
+      "学堂": 63589,
+      "emu": 63590,
+      "的画面": 63591,
+      "▁dit": 63592,
+      "▁wolf": 63593,
+      "▁Vijay": 63594,
+      "江河": 63595,
+      "虚无": 63596,
+      "▁Rey": 63597,
+      "小心翼翼": 63598,
+      "有停车位": 63599,
+      "basis": 63600,
+      "Medium": 63601,
+      "PACKAGE": 63602,
+      "增效": 63603,
+      "魔幻": 63604,
+      "使学生": 63605,
+      "akar": 63606,
+      "nExec": 63607,
+      "shuffle": 63608,
+      "▁linger": 63609,
+      "案件受理费人民币": 63610,
+      "DIRECTORY": 63611,
+      "▁anatomic": 63612,
+      "在京": 63613,
+      "该笔": 63614,
+      "▁\"&": 63615,
+      "ANTS": 63616,
+      "▁fry": 63617,
+      "集体经济": 63618,
+      "alpine": 63619,
+      "nHouston": 63620,
+      "▁farther": 63621,
+      "▁prioritize": 63622,
+      "▁repression": 63623,
+      "▁vibrational": 63624,
+      "▁erythrocytes": 63625,
+      "致癌": 63626,
+      "行之": 63627,
+      "出不穷": 63628,
+      "更加的": 63629,
+      "死亡的": 63630,
+      "oreal": 63631,
+      "依法可以从轻处罚": 63632,
+      "▁autoantib": 63633,
+      "红线": 63634,
+      "jpa": 63635,
+      "diss": 63636,
+      "uric": 63637,
+      "▁hue": 63638,
+      "将本案移送": 63639,
+      "databases": 63640,
+      "▁Lebanese": 63641,
+      "▁vacancies": 63642,
+      "中国队": 63643,
+      "新型的": 63644,
+      "yrin": 63645,
+      "基础教育": 63646,
+      "▁mileage": 63647,
+      "▁StringIO": 63648,
+      "Identification": 63649,
+      "发觉": 63650,
+      "客栈": 63651,
+      "票务": 63652,
+      "▁Gle": 63653,
+      "▁Ler": 63654,
+      "中国电信": 63655,
+      "差不多的": 63656,
+      "换句话说": 63657,
+      "▁heck": 63658,
+      "STRUCT": 63659,
+      "Production": 63660,
+      "外径": 63661,
+      "DFS": 63662,
+      "制造的": 63663,
+      "))))": 63664,
+      "▁Victory": 63665,
+      "HG": 63666,
+      "偏低": 63667,
+      "扩充": 63668,
+      "派对": 63669,
+      "gos": 63670,
+      "▁NI": 63671,
+      "法律顾问": 63672,
+      "随便吃吃": 63673,
+      "二〇一六年九": 63674,
+      "纪念碑": 63675,
+      "ifen": 63676,
+      "欢迎大家": 63677,
+      "atorics": 63678,
+      "▁bargain": 63679,
+      "▁macular": 63680,
+      "correlation": 63681,
+      "▁overcoming": 63682,
+      "examethasone": 63683,
+      "大山": 63684,
+      "组团": 63685,
+      "和专业": 63686,
+      "年北京": 63687,
+      "Very": 63688,
+      "▁MMA": 63689,
+      "Partai": 63690,
+      "Interop": 63691,
+      "▁Proper": 63692,
+      "▁Knowing": 63693,
+      "凡事": 63694,
+      "并可": 63695,
+      "身亡": 63696,
+      "cci": 63697,
+      "会让你": 63698,
+      "项目部": 63699,
+      "Boost": 63700,
+      "EFORE": 63701,
+      "▁Hank": 63702,
+      "中选择一个": 63703,
+      "unding": 63704,
+      "opically": 63705,
+      "而异": 63706,
+      "钻井": 63707,
+      "你已经": 63708,
+      "添加到": 63709,
+      "▁alot": 63710,
+      "▁prejud": 63711,
+      "istrative": 63712,
+      "则该": 63713,
+      "胜任": 63714,
+      "永远不会": 63715,
+      "▁foci": 63716,
+      "▁云中书城": 63717,
+      "▁人物简介": 63718,
+      "▁Leone": 63719,
+      "中国人民大学": 63720,
+      "editable": 63721,
+      "▁Optimal": 63722,
+      "▁dilemma": 63723,
+      "电子科技有限公司": 63724,
+      "contracts": 63725,
+      "▁Radiation": 63726,
+      "执行情况告知申请执行人": 63727,
+      "Su": 63728,
+      "姜片": 63729,
+      "雷电": 63730,
+      "\"\",": 63731,
+      "▁修改": 63732,
+      "Slide": 63733,
+      "ircon": 63734,
+      "sparse": 63735,
+      "▁felony": 63736,
+      "盗版": 63737,
+      "到哪里": 63738,
+      "我曾经": 63739,
+      "农村居民": 63740,
+      "ipzig": 63741,
+      "▁Loren": 63742,
+      "▁Genesis": 63743,
+      "▁Assignment": 63744,
+      "三期": 63745,
+      "代办": 63746,
+      "新乡": 63747,
+      "江津": 63748,
+      "除湿": 63749,
+      "和创新": 63750,
+      "新建的": 63751,
+      "的好坏": 63752,
+      "Dear": 63753,
+      "▁HMS": 63754,
+      "▁TAG": 63755,
+      "▁处理器": 63756,
+      "ulfide": 63757,
+      "▁Chronicle": 63758,
+      "保值": 63759,
+      "得过": 63760,
+      "误导": 63761,
+      "née": 63762,
+      "预期的": 63763,
+      "经济贸易": 63764,
+      "omies": 63765,
+      "▁Alto": 63766,
+      "▁Plas": 63767,
+      "Checkbox": 63768,
+      "▁checker": 63769,
+      "▁soybean": 63770,
+      "▁autosomal": 63771,
+      "伺服": 63772,
+      "你没": 63773,
+      "服役": 63774,
+      "lost": 63775,
+      "不确定性": 63776,
+      "检诉刑诉": 63777,
+      "身心健康": 63778,
+      "▁Curve": 63779,
+      "▁Devils": 63780,
+      "二〇一六年七月": 63781,
+      "JSONObject": 63782,
+      "▁mathematically": 63783,
+      "后半": 63784,
+      "年版": 63785,
+      "toml": 63786,
+      "▁Kod": 63787,
+      "▁Von": 63788,
+      "采取保全措施后": 63789,
+      "▁cylinders": 63790,
+      "▁thumbnail": 63791,
+      "▁experimenting": 63792,
+      "烟雨": 63793,
+      "粉的": 63794,
+      "镜片": 63795,
+      "回家的": 63796,
+      "▁Mai": 63797,
+      "▁scare": 63798,
+      "fillable": 63799,
+      "▁academia": 63800,
+      "本校": 63801,
+      "))}\\": 63802,
+      "bath": 63803,
+      "▁MSN": 63804,
+      "▁SDS": 63805,
+      "nFood": 63806,
+      "ophan": 63807,
+      "▁Thin": 63808,
+      "▁Wick": 63809,
+      "住所地福建省": 63810,
+      "▁cyclase": 63811,
+      "制服": 63812,
+      "是原": 63813,
+      "汉化": 63814,
+      "精子": 63815,
+      "▁大小": 63816,
+      "npre": 63817,
+      "▁PORT": 63818,
+      "人去": 63819,
+      "但因": 63820,
+      "号之二": 63821,
+      "我来说": 63822,
+      "unner": 63823,
+      "▁Parade": 63824,
+      "▁Wallpaper": 63825,
+      "▁contaminants": 63826,
+      "乳酸": 63827,
+      "王俊": 63828,
+      "ddd": 63829,
+      "▁%)": 63830,
+      "到医院": 63831,
+      "▁Eva": 63832,
+      "下列说法": 63833,
+      "年降水量": 63834,
+      "codep": 63835,
+      "ipynb": 63836,
+      "▁expenditures": 63837,
+      "▁沙": 63838,
+      "对阵": 63839,
+      "汤汁": 63840,
+      "upa": 63841,
+      "ATTR": 63842,
+      "分式方程": 63843,
+      "▁peac": 63844,
+      "▁商品毛重": 63845,
+      "进行了深入": 63846,
+      "Following": 63847,
+      "▁次": 63848,
+      "交办": 63849,
+      "Cho": 63850,
+      "SAR": 63851,
+      "iesis": 63852,
+      "▁surpass": 63853,
+      "中线": 63854,
+      "先看": 63855,
+      "和弦": 63856,
+      "而无": 63857,
+      "公司于": 63858,
+      "esto": 63859,
+      "Margin": 63860,
+      "patches": 63861,
+      "spinner": 63862,
+      "▁fal": 63863,
+      "▁Clip": 63864,
+      "▁emerges": 63865,
+      "点儿": 63866,
+      "▁对于": 63867,
+      "省部级": 63868,
+      "nPak": 63869,
+      "这个游戏": 63870,
+      "coupled": 63871,
+      "▁tariffs": 63872,
+      "disconnect": 63873,
+      "▁Disorders": 63874,
+      "▁protections": 63875,
+      "他了": 63876,
+      "日落": 63877,
+      "活化": 63878,
+      "保险人": 63879,
+      "公里处": 63880,
+      "各领域": 63881,
+      "位置关系": 63882,
+      "和相应的": 63883,
+      "neuro": 63884,
+      "▁NVIDIA": 63885,
+      "▁fistula": 63886,
+      "▁declaring": 63887,
+      "分別": 63888,
+      "获得者": 63889,
+      "有效数字": 63890,
+      "自己喜欢": 63891,
+      "FOLDER": 63892,
+      "Gallery": 63893,
+      "Reading": 63894,
+      "▁Sandra": 63895,
+      ">|": 63896,
+      "休假": 63897,
+      "文昌": 63898,
+      "波斯": 63899,
+      "▁LU": 63900,
+      "周某某": 63901,
+      "在生产": 63902,
+      "方向盘": 63903,
+      "}\\\\;": 63904,
+      "产品设计": 63905,
+      "寻衅滋事罪": 63906,
+      "efficacy": 63907,
+      "改建": 63908,
+      "渊源": 63909,
+      "▁Languages": 63910,
+      "当归": 63911,
+      "放宽": 63912,
+      "溪镇": 63913,
+      ":',": 63914,
+      "创造力": 63915,
+      "达成了": 63916,
+      "destruct": 63917,
+      "▁cellulose": 63918,
+      "approximately": 63919,
+      "是集": 63920,
+      "焦作市": 63921,
+      "予以证明": 63922,
+      "公报": 63923,
+      "升起": 63924,
+      "几乎是": 63925,
+      "cous": 63926,
+      "waiting": 63927,
+      "▁quarry": 63928,
+      "shipping": 63929,
+      "▁Lindsay": 63930,
+      "▁annotated": 63931,
+      "加息": 63932,
+      "宣扬": 63933,
+      "语的": 63934,
+      "浓度的": 63935,
+      ")(\\\\": 63936,
+      "embr": 63937,
+      "summ": 63938,
+      "getKey": 63939,
+      "▁THREE": 63940,
+      "安得": 63941,
+      "haw": 63942,
+      "物业类别": 63943,
+      "▁herd": 63944,
+      "Separator": 63945,
+      "nElections": 63946,
+      "▁inhalation": 63947,
+      "二轮": 63948,
+      "寒假": 63949,
+      "打法": 63950,
+      "断面": 63951,
+      "▁FW": 63952,
+      "isin": 63953,
+      "nMuch": 63954,
+      "▁Uncle": 63955,
+      "▁pairwise": 63956,
+      "▁Dimensions": 63957,
+      "▁correctness": 63958,
+      "方舟": 63959,
+      "来个": 63960,
+      "jvm": 63961,
+      "从严治": 63962,
+      "▁Xml": 63963,
+      "自由贸易": 63964,
+      "nPrem": 63965,
+      "▁dome": 63966,
+      "▁bucks": 63967,
+      "▁ejection": 63968,
+      "▁prescriptions": 63969,
+      "▁夜": 63970,
+      "充斥": 63971,
+      "同位": 63972,
+      "巴彦": 63973,
+      "幸好": 63974,
+      "我与": 63975,
+      "胞胎": 63976,
+      "也已经": 63977,
+      "▁cocoa": 63978,
+      "第二百四十条": 63979,
+      "INTEGER": 63980,
+      "Christian": 63981,
+      "▁gatherings": 63982,
+      "▁configurable": 63983,
+      "体贴": 63984,
+      "吞噬": 63985,
+      "意外的": 63986,
+      "▁CKD": 63987,
+      "inform": 63988,
+      "▁ordinal": 63989,
+      "入睡": 63990,
+      "中国人的": 63991,
+      "可能会有": 63992,
+      "▁disreg": 63993,
+      "▁favourites": 63994,
+      "推断": 63995,
+      "秋水": 63996,
+      "aan": 63997,
+      "这只是": 63998,
+      "▁Increase": 63999,
+      "发生法律效力之日起": 64000,
+      "▁discriminant": 64001,
+      "一式": 64002,
+      "和多": 64003,
+      "教育活动": 64004,
+      "被告辩称": 64005,
+      "达成协议": 64006,
+      "netty": 64007,
+      "▁商品产地": 64008,
+      "曾是": 64009,
+      "黄土": 64010,
+      "ста": 64011,
+      "▁WB": 64012,
+      "又因为": 64013,
+      "家常菜": 64014,
+      "IALS": 64015,
+      "产品尺寸": 64016,
+      "zilla": 64017,
+      "kinase": 64018,
+      "▁seals": 64019,
+      "▁refusing": 64020,
+      "太白": 64021,
+      "很像": 64022,
+      "COME": 64023,
+      "{})\\": 64024,
+      "▁contractions": 64025,
+      "▁osteoarthritis": 64026,
+      "别克": 64027,
+      "战友": 64028,
+      "抚顺": 64029,
+      "背诵": 64030,
+      "在经济": 64031,
+      "obac": 64032,
+      "oders": 64033,
+      "Vision": 64034,
+      "buster": 64035,
+      "▁bloody": 64036,
+      "▁payable": 64037,
+      "▁holomorphic": 64038,
+      "很想": 64039,
+      "言行": 64040,
+      "nHop": 64041,
+      "▁Bic": 64042,
+      "数据进行": 64043,
+      "之神": 64044,
+      "公会": 64045,
+      "利物": 64046,
+      "危险的": 64047,
+      "Would": 64048,
+      "nLink": 64049,
+      "▁roast": 64050,
+      "rectangle": 64051,
+      "▁tackling": 64052,
+      "马尾": 64053,
+      "的研发": 64054,
+      "nMap": 64055,
+      "oger": 64056,
+      "征求意见": 64057,
+      "ETHOD": 64058,
+      "efile": 64059,
+      "▁sinks": 64060,
+      "▁betray": 64061,
+      "reporting": 64062,
+      "▁regenerative": 64063,
+      "架空历史": 64064,
+      "▁loft": 64065,
+      "查封被申请人": 64066,
+      "folders": 64067,
+      "▁charcoal": 64068,
+      "▁primates": 64069,
+      "▁delineate": 64070,
+      "丞相": 64071,
+      "五种": 64072,
+      "开销": 64073,
+      "杂志社": 64074,
+      "矿泉水": 64075,
+      "▁conced": 64076,
+      "endpoints": 64077,
+      "▁Disability": 64078,
+      "▁northwestern": 64079,
+      "传记": 64080,
+      "DIV": 64081,
+      "inster": 64082,
+      "joined": 64083,
+      "leting": 64084,
+      "▁seminal": 64085,
+      "nNovember": 64086,
+      "IAN": 64087,
+      "▁Bcl": 64088,
+      "二审法院": 64089,
+      "社交媒体": 64090,
+      "Portal": 64091,
+      "ridden": 64092,
+      "▁postoperatively": 64093,
+      "一发": 64094,
+      "入库": 64095,
+      "家有": 64096,
+      "你必须": 64097,
+      "教学班": 64098,
+      "forth": 64099,
+      "▁Rover": 64100,
+      "▁canon": 64101,
+      "▁autopsy": 64102,
+      "功底": 64103,
+      "歌剧": 64104,
+      "就一直": 64105,
+      "newline": 64106,
+      "▁Strike": 64107,
+      "▁formatter": 64108,
+      "▁implicitly": 64109,
+      "▁Researchers": 64110,
+      "══": 64111,
+      "热议": 64112,
+      "不到位": 64113,
+      "▁gaz": 64114,
+      "项目经理": 64115,
+      "nJuly": 64116,
+      "XB": 64117,
+      "决战": 64118,
+      "扁舟": 64119,
+      "Ham": 64120,
+      "Pur": 64121,
+      "▁DEC": 64122,
+      "此页无正文": 64123,
+      "土木结构住房": 64124,
+      "▁Uncategorized": 64125,
+      "取胜": 64126,
+      "阻抗": 64127,
+      "高专": 64128,
+      "ivi": 64129,
+      "征决字": 64130,
+      "起源于": 64131,
+      "WEVER": 64132,
+      "RunWith": 64133,
+      "▁whoever": 64134,
+      "有为": 64135,
+      "罪恶": 64136,
+      "耐火": 64137,
+      "的约定": 64138,
+      "劳务派遣": 64139,
+      "nHard": 64140,
+      "▁wards": 64141,
+      "▁Bolton": 64142,
+      "不许": 64143,
+      "过长": 64144,
+      ")},": 64145,
+      "▁Rican": 64146,
+      "▁axiom": 64147,
+      "▁Bulletin": 64148,
+      "▁airborne": 64149,
+      "感人": 64150,
+      "我叫": 64151,
+      "设法": 64152,
+      "麦当": 64153,
+      "vip": 64154,
+      "▁ki": 64155,
+      "建筑的": 64156,
+      "▁rud": 64157,
+      "mtime": 64158,
+      "▁Moor": 64159,
+      "▁quil": 64160,
+      "urator": 64161,
+      "▁yummy": 64162,
+      "nprintf": 64163,
+      "▁cautious": 64164,
+      "乳房": 64165,
+      "劳工": 64166,
+      "率是": 64167,
+      "orh": 64168,
+      "▁та": 64169,
+      "了一大": 64170,
+      "人与人": 64171,
+      "验证了": 64172,
+      "建设工作": 64173,
+      "userid": 64174,
+      "▁belts": 64175,
+      "判处无期徒刑": 64176,
+      "▁Detective": 64177,
+      "绿灯": 64178,
+      "itu": 64179,
+      "▁sr": 64180,
+      "▁Til": 64181,
+      "Jones": 64182,
+      "photon": 64183,
+      "charged": 64184,
+      "▁Alumni": 64185,
+      "▁Literary": 64186,
+      "▁Holocaust": 64187,
+      "第二百五十三条的规定": 64188,
+      "曾于": 64189,
+      "难怪": 64190,
+      "我们来": 64191,
+      "营业收入": 64192,
+      "营销策划": 64193,
+      "dense": 64194,
+      "these": 64195,
+      "▁dawn": 64196,
+      "犯盗窃罪被": 64197,
+      "▁比": 64198,
+      "推导": 64199,
+      "atura": 64200,
+      "▁Stores": 64201,
+      "▁firearm": 64202,
+      "▁Elasticsearch": 64203,
+      "三楼": 64204,
+      "内阁": 64205,
+      "aol": 64206,
+      "冷空气": 64207,
+      "nmin": 64208,
+      "好自己的": 64209,
+      "▁litre": 64210,
+      "▁midpoint": 64211,
+      "克罗": 64212,
+      "官僚": 64213,
+      "了大量": 64214,
+      "}}{{\\\\": 64215,
+      "Conversion": 64216,
+      "▁neuromuscular": 64217,
+      "▁成": 64218,
+      "口镇": 64219,
+      "]):": 64220,
+      "十三年": 64221,
+      "投入使用": 64222,
+      "▁Hits": 64223,
+      "▁podium": 64224,
+      "▁tumorigenesis": 64225,
+      "三者": 64226,
+      "临汾": 64227,
+      "墓志": 64228,
+      "洗发": 64229,
+      "爱民": 64230,
+      "脊髓": 64231,
+      "▁jungle": 64232,
+      "▁Vikings": 64233,
+      "subscriptions": 64234,
+      "ня": 64235,
+      "他为": 64236,
+      "甜品": 64237,
+      "空气质量": 64238,
+      "vidia": 64239,
+      "▁faux": 64240,
+      "▁paradigms": 64241,
+      "该犯在服刑期间确有悔改表现": 64242,
+      "妖怪": 64243,
+      "绵阳": 64244,
+      "正规的": 64245,
+      "DESC": 64246,
+      "水利水电": 64247,
+      "研究机构": 64248,
+      "部分地区": 64249,
+      "▁assayed": 64250,
+      "▁qualifier": 64251,
+      "中说": 64252,
+      "流向": 64253,
+      "祖宗": 64254,
+      "通的": 64255,
+      "和一些": 64256,
+      "LOWED": 64257,
+      "SEMENT": 64258,
+      "infile": 64259,
+      "▁buffet": 64260,
+      "▁geodes": 64261,
+      "▁vortex": 64262,
+      "▁unusually": 64263,
+      "利民": 64264,
+      "念念": 64265,
+      "美感": 64266,
+      "前缀和": 64267,
+      "iken": 64268,
+      "▁Bergen": 64269,
+      "▁refuses": 64270,
+      "▁callable": 64271,
+      "▁diameters": 64272,
+      "例行": 64273,
+      "含金": 64274,
+      "应激": 64275,
+      "粪便": 64276,
+      "静止": 64277,
+      "层出不穷": 64278,
+      "铁路运输": 64279,
+      "LIMIT": 64280,
+      "sofIvy": 64281,
+      "▁Charity": 64282,
+      "▁Supported": 64283,
+      "▁paramount": 64284,
+      "赣州市": 64285,
+      "imid": 64286,
+      "pray": 64287,
+      "▁Era": 64288,
+      "▁padd": 64289,
+      "▁主要原料": 64290,
+      "▁Suzuki": 64291,
+      "▁humour": 64292,
+      "DbContext": 64293,
+      "▁lineback": 64294,
+      "▁stumbled": 64295,
+      "▁ambiguity": 64296,
+      "▁Volkswagen": 64297,
+      "Fx": 64298,
+      "将与": 64299,
+      "qty": 64300,
+      "在其中": 64301,
+      "某负担": 64302,
+      "的营养": 64303,
+      "这两年": 64304,
+      "erta": 64305,
+      "Grade": 64306,
+      "nTHIS": 64307,
+      "ypass": 64308,
+      "▁glioma": 64309,
+      "▁hunter": 64310,
+      "酿造": 64311,
+      "阳明": 64312,
+      "▁CTL": 64313,
+      "整个过程": 64314,
+      "不准许撤诉的": 64315,
+      "nheader": 64316,
+      "▁Euclid": 64317,
+      "▁droplets": 64318,
+      "憧憬": 64319,
+      "演的": 64320,
+      "工作了": 64321,
+      "方方面": 64322,
+      "次日起": 64323,
+      "全面提升": 64324,
+      "▁制片地区": 64325,
+      "getcwd": 64326,
+      "如果把": 64327,
+      "想到的": 64328,
+      "手术后": 64329,
+      "半个小时": 64330,
+      "odenal": 64331,
+      "▁ganglia": 64332,
+      "▁condemned": 64333,
+      "▁externally": 64334,
+      "灭绝": 64335,
+      "铁丝": 64336,
+      "专卖店": 64337,
+      "代理商": 64338,
+      "的中心": 64339,
+      "语言学": 64340,
+      "Benz": 64341,
+      "解析式为": 64342,
+      "nMATER": 64343,
+      "▁richness": 64344,
+      "▁validators": 64345,
+      "一览": 64346,
+      "bbc": 64347,
+      "jia": 64348,
+      "我就不": 64349,
+      "来表示": 64350,
+      "第二审": 64351,
+      "alon": 64352,
+      "培训学校": 64353,
+      "▁griev": 64354,
+      "nSenior": 64355,
+      "▁Lisbon": 64356,
+      "▁floods": 64357,
+      "▁mA": 64358,
+      "是什么呢": 64359,
+      "被认为是": 64360,
+      "ELISA": 64361,
+      "frica": 64362,
+      "changer": 64363,
+      "▁Farmers": 64364,
+      "就读于": 64365,
+      "▁FIN": 64366,
+      "▁sont": 64367,
+      "urlopen": 64368,
+      "▁toolkit": 64369,
+      "technical": 64370,
+      "oC": 64371,
+      "伙同": 64372,
+      "王玉": 64373,
+      "PRI": 64374,
+      "ème": 64375,
+      "▁HQ": 64376,
+      "▁Witch": 64377,
+      "▁Albums": 64378,
+      "▁retardation": 64379,
+      "御史": 64380,
+      "热潮": 64381,
+      "时需要": 64382,
+      "Ware": 64383,
+      "acyl": 64384,
+      "etin": 64385,
+      "值得注意的是": 64386,
+      "广东省佛山市": 64387,
+      "乘除": 64388,
+      "余个": 64389,
+      "分包": 64390,
+      "就职": 64391,
+      ":\"\\": 64392,
+      "世界各国": 64393,
+      "公司总经理": 64394,
+      "已减半收取": 64395,
+      "▁Controls": 64396,
+      "▁prisoner": 64397,
+      "▁constitutive": 64398,
+      "nCongratulations": 64399,
+      "▁贺": 64400,
+      "佳作": 64401,
+      "TIM": 64402,
+      "▁RAF": 64403,
+      "▁vue": 64404,
+      "尤为重要": 64405,
+      "fried": 64406,
+      "nDivide": 64407,
+      "▁Inform": 64408,
+      "▁reagent": 64409,
+      "切记": 64410,
+      "轻盈": 64411,
+      "法测定": 64412,
+      "pard": 64413,
+      "魔龙令牌": 64414,
+      "▁Blow": 64415,
+      "▁mixes": 64416,
+      "▁runoff": 64417,
+      "▁strives": 64418,
+      "▁modulates": 64419,
+      "▁radiological": 64420,
+      "极管": 64421,
+      "相符": 64422,
+      "UCN": 64423,
+      "的有关": 64424,
+      "nSoviet": 64425,
+      "▁Trevor": 64426,
+      "▁Ministers": 64427,
+      "▁aggregated": 64428,
+      "▁negotiated": 64429,
+      "多情": 64430,
+      "而今": 64431,
+      "表决": 64432,
+      "DTD": 64433,
+      "HAN": 64434,
+      "一本书": 64435,
+      "在诉讼": 64436,
+      "专科学校": 64437,
+      "是一名演员": 64438,
+      "ulsory": 64439,
+      "▁Alpine": 64440,
+      "▁Webster": 64441,
+      "▁vacancy": 64442,
+      "IPv": 64443,
+      "很适合": 64444,
+      "冠状动脉": 64445,
+      "acons": 64446,
+      "▁Omar": 64447,
+      "\\\\\\\\\\\\": 64448,
+      "▁sponge": 64449,
+      "▁Syracuse": 64450,
+      "坏事": 64451,
+      "杏花": 64452,
+      "村干部": 64453,
+      "第三边": 64454,
+      "▁chr": 64455,
+      "爷爷奶奶": 64456,
+      "相关证据": 64457,
+      "▁mong": 64458,
+      "公办": 64459,
+      "切勿": 64460,
+      "Got": 64461,
+      "ibu": 64462,
+      "▁KC": 64463,
+      "▁eh": 64464,
+      "工商联": 64465,
+      "由此可见": 64466,
+      "inged": 64467,
+      "stret": 64468,
+      "糖尿病患者": 64469,
+      "nprocess": 64470,
+      "▁substituting": 64471,
+      "漂流": 64472,
+      "argc": 64473,
+      "nova": 64474,
+      "ombo": 64475,
+      "▁Roland": 64476,
+      "▁advising": 64477,
+      "中国银行股份有限公司": 64478,
+      "双双": 64479,
+      "小猫": 64480,
+      "拉黑": 64481,
+      "福德": 64482,
+      "挖掘机": 64483,
+      "非常多": 64484,
+      "araj": 64485,
+      "{\\\\\"": 64486,
+      "▁kub": 64487,
+      "▁Shane": 64488,
+      "小林": 64489,
+      "我以前": 64490,
+      "apsing": 64491,
+      "retrieve": 64492,
+      "▁shaking": 64493,
+      "▁thrombin": 64494,
+      "▁Comparative": 64495,
+      "Ki": 64496,
+      "也行": 64497,
+      "青天": 64498,
+      "TES": 64499,
+      "可以有": 64500,
+      "持有的": 64501,
+      "nRen": 64502,
+      "Navigator": 64503,
+      "▁pitching": 64504,
+      "▁analgesia": 64505,
+      "抢购": 64506,
+      "而下": 64507,
+      "本项目": 64508,
+      "系列活动": 64509,
+      "Voice": 64510,
+      "▁Claus": 64511,
+      "二〇一六年十": 64512,
+      "icating": 64513,
+      "▁arranging": 64514,
+      "SZ": 64515,
+      "状态的": 64516,
+      "Mary": 64517,
+      "icki": 64518,
+      "第十四章": 64519,
+      "mortem": 64520,
+      "▁horny": 64521,
+      "▁diaphrag": 64522,
+      "▁distraction": 64523,
+      "州市中级人民法院提出上诉": 64524,
+      "焚烧": 64525,
+      "▁Ves": 64526,
+      "下面是一个": 64527,
+      "indexes": 64528,
+      "Instances": 64529,
+      "▁Interval": 64530,
+      "▁microbiota": 64531,
+      "azo": 64532,
+      "销售等": 64533,
+      "▁Roots": 64534,
+      "▁amalg": 64535,
+      "▁arche": 64536,
+      "▁wives": 64537,
+      "ный": 64538,
+      "结果与": 64539,
+      "裁决书": 64540,
+      "fatal": 64541,
+      "▁sealing": 64542,
+      "described": 64543,
+      "▁embraced": 64544,
+      "▁circuitry": 64545,
+      "▁panoramic": 64546,
+      "▁Engagement": 64547,
+      "两三": 64548,
+      "你不是": 64549,
+      "▁MAR": 64550,
+      "vdots": 64551,
+      "▁drawers": 64552,
+      "▁toilets": 64553,
+      "二号": 64554,
+      "对决": 64555,
+      "铸铁": 64556,
+      "][:": 64557,
+      "svc": 64558,
+      "区间为": 64559,
+      "的一样": 64560,
+      "的数是": 64561,
+      "污染防治": 64562,
+      "简直就是": 64563,
+      "doing": 64564,
+      "▁hype": 64565,
+      "▁congru": 64566,
+      "▁trapez": 64567,
+      "prepared": 64568,
+      "本案诉讼费由被告承担": 64569,
+      "中风": 64570,
+      "拟定": 64571,
+      "捕获": 64572,
+      "是会": 64573,
+      "皮书": 64574,
+      "福音": 64575,
+      "荣华": 64576,
+      "EAA": 64577,
+      "UNG": 64578,
+      "▁dm": 64579,
+      "加大对": 64580,
+      "▁Dul": 64581,
+      "PARAM": 64582,
+      "▁functools": 64583,
+      "拟任": 64584,
+      "是种": 64585,
+      "gren": 64586,
+      "▁AIM": 64587,
+      "mless": 64588,
+      "住所地江苏省": 64589,
+      "▁hacker": 64590,
+      "Overview": 64591,
+      "▁forgive": 64592,
+      "▁veterin": 64593,
+      "▁reservoirs": 64594,
+      "▁civilization": 64595,
+      "专访": 64596,
+      "养的": 64597,
+      "泊尔": 64598,
+      "圆柱形": 64599,
+      "随时随地": 64600,
+      "厦门市思明": 64601,
+      "nIncre": 64602,
+      "上诉于上海市": 64603,
+      "▁briefing": 64604,
+      "▁divisors": 64605,
+      "▁disposition": 64606,
+      "とい": 64607,
+      "创设": 64608,
+      "特拉": 64609,
+      "泰安市": 64610,
+      "▁Vaugh": 64611,
+      "▁threaten": 64612,
+      "▁simulating": 64613,
+      "一汽": 64614,
+      "基石": 64615,
+      "大牌": 64616,
+      "食疗": 64617,
+      "ousse": 64618,
+      "▁mans": 64619,
+      "etched": 64620,
+      "▁原料": 64621,
+      "皮肤的": 64622,
+      "▁ego": 64623,
+      "行政审批": 64624,
+      "▁Gain": 64625,
+      "izumab": 64626,
+      "▁pyram": 64627,
+      "sampling": 64628,
+      "▁snippets": 64629,
+      "homogeneous": 64630,
+      "乳化": 64631,
+      "保研": 64632,
+      "减灾": 64633,
+      "执教": 64634,
+      "站位": 64635,
+      "这支": 64636,
+      "NAS": 64637,
+      "▁STAR": 64638,
+      "▁Kathy": 64639,
+      "二〇一四年九月": 64640,
+      "▁Krishna": 64641,
+      "▁musculoskeletal": 64642,
+      "抽出": 64643,
+      "时可": 64644,
+      "和时间": 64645,
+      "nasium": 64646,
+      "geometric": 64647,
+      "并入": 64648,
+      "永嘉": 64649,
+      "会议由": 64650,
+      "长大的": 64651,
+      "▁Bri": 64652,
+      "市场价格": 64653,
+      "▁Mineral": 64654,
+      "▁prompts": 64655,
+      "▁trapping": 64656,
+      "∈（": 64657,
+      "十堰市": 64658,
+      "红细胞": 64659,
+      "bish": 64660,
+      "placing": 64661,
+      "GLIGENCE": 64662,
+      "▁Pokémon": 64663,
+      "▁placenta": 64664,
+      "上火": 64665,
+      "超强": 64666,
+      "blatt": 64667,
+      "grave": 64668,
+      "npass": 64669,
+      "▁Bind": 64670,
+      "劳动争议纠纷一案": 64671,
+      "▁daylight": 64672,
+      "▁announces": 64673,
+      "▁waveguide": 64674,
+      "▁ş": 64675,
+      "cdf": 64676,
+      "▁NL": 64677,
+      "别忘了": 64678,
+      "这个小": 64679,
+      "▁Comes": 64680,
+      "▁Banner": 64681,
+      "▁CONSEQUENTIAL": 64682,
+      "öm": 64683,
+      "欧式": 64684,
+      "▁QA": 64685,
+      "工资款": 64686,
+      "需付费": 64687,
+      "nWil": 64688,
+      "▁Satisf": 64689,
+      "▁outright": 64690,
+      "この": 64691,
+      "不正": 64692,
+      "克林": 64693,
+      "老天": 64694,
+      "输液": 64695,
+      "CoA": 64696,
+      "占有率": 64697,
+      "第一行": 64698,
+      "nVue": 64699,
+      "apopt": 64700,
+      "怎样的体验": 64701,
+      "管辖权异议": 64702,
+      "opters": 64703,
+      "explorer": 64704,
+      "RequestBody": 64705,
+      "排泄": 64706,
+      "机上": 64707,
+      "极小": 64708,
+      "软体": 64709,
+      "包括了": 64710,
+      "等产品": 64711,
+      "与被害人": 64712,
+      "感情基础": 64713,
+      "▁swagger": 64714,
+      "▁executor": 64715,
+      "千人": 64716,
+      "换乘": 64717,
+      "\"//": 64718,
+      "下辖村": 64719,
+      "是用来": 64720,
+      "Marc": 64721,
+      "orce": 64722,
+      "形成一个": 64723,
+      "Logout": 64724,
+      "河南省郑州市": 64725,
+      "▁chemot": 64726,
+      "rollable": 64727,
+      "openstack": 64728,
+      "▁parasitic": 64729,
+      "张晓": 64730,
+      "杜甫": 64731,
+      "在选择": 64732,
+      "ubicin": 64733,
+      "▁Remix": 64734,
+      "▁corps": 64735,
+      "ensible": 64736,
+      "traffic": 64737,
+      "▁Clement": 64738,
+      "▁hectares": 64739,
+      "ード": 64740,
+      "东汉": 64741,
+      "坚韧": 64742,
+      "第一项至": 64743,
+      "morph": 64744,
+      "omotor": 64745,
+      "▁Roose": 64746,
+      "tutorials": 64747,
+      "California": 64748,
+      "祖父": 64749,
+      "誓言": 64750,
+      "风筝": 64751,
+      "️\\": 64752,
+      "ipei": 64753,
+      "苏州市吴": 64754,
+      "would": 64755,
+      "被执行人吴": 64756,
+      "little": 64757,
+      "pygame": 64758,
+      "▁prosthesis": 64759,
+      "▁investigative": 64760,
+      "三章": 64761,
+      "龙山": 64762,
+      "可以对": 64763,
+      "计算在": 64764,
+      "▁gol": 64765,
+      "jq": 64766,
+      "▁段": 64767,
+      "影人": 64768,
+      "tsc": 64769,
+      "也存在": 64770,
+      "但不是": 64771,
+      "▁Kai": 64772,
+      "怎样才能": 64773,
+      "英语教学": 64774,
+      "执行机关认为": 64775,
+      "被执行人暂无": 64776,
+      "plitude": 64777,
+      "experience": 64778,
+      "各样": 64779,
+      "思索": 64780,
+      "icum": 64781,
+      "broadcast": 64782,
+      "▁persists": 64783,
+      "▁ambitions": 64784,
+      "平坦": 64785,
+      "来宾": 64786,
+      "▁gui": 64787,
+      "好好学习": 64788,
+      "tagged": 64789,
+      "▁Wendy": 64790,
+      "Nonnull": 64791,
+      "STITUTE": 64792,
+      "▁verses": 64793,
+      "▁singularity": 64794,
+      "lw": 64795,
+      "与原": 64796,
+      "处境": 64797,
+      "▁är": 64798,
+      "敏感性": 64799,
+      "完全破裂": 64800,
+      "▁Oste": 64801,
+      "▁regain": 64802,
+      "▁ubiquitin": 64803,
+      "正月": 64804,
+      "▁wp": 64805,
+      "▁Pix": 64806,
+      "▁psi": 64807,
+      "县委书记": 64808,
+      "▁Sixty": 64809,
+      "▁Tucker": 64810,
+      "zf": 64811,
+      "▁建": 64812,
+      "▁闫": 64813,
+      "田县": 64814,
+      "花岗": 64815,
+      "装入": 64816,
+      "对本院": 64817,
+      "的小孩": 64818,
+      "Night": 64819,
+      "▁camel": 64820,
+      "ranging": 64821,
+      "▁crashing": 64822,
+      "▁deserved": 64823,
+      "▁Inspection": 64824,
+      "一滴": 64825,
+      "六年级": 64826,
+      "mans": 64827,
+      "▁Gad": 64828,
+      "气候变化": 64829,
+      "▁Candy": 64830,
+      "▁SPECT": 64831,
+      "creating": 64832,
+      "▁buffers": 64833,
+      "游行": 64834,
+      "Ele": 64835,
+      "好消息": 64836,
+      "汽车站": 64837,
+      "▁Helic": 64838,
+      "▁Lawyers": 64839,
+      "FORMATION": 64840,
+      "arra": 64841,
+      "tapp": 64842,
+      "▁PPAR": 64843,
+      "▁blunt": 64844,
+      "▁burnt": 64845,
+      "uttering": 64846,
+      "宴请": 64847,
+      "sbt": 64848,
+      "十一年": 64849,
+      "和要求": 64850,
+      "适合于": 64851,
+      "都具有": 64852,
+      "ZONE": 64853,
+      "ohan": 64854,
+      "▁phage": 64855,
+      "▁Willie": 64856,
+      "▁enanti": 64857,
+      "initialized": 64858,
+      "前卫": 64859,
+      "无色": 64860,
+      "Ban": 64861,
+      "崭新的": 64862,
+      "胶原蛋白": 64863,
+      "▁rage": 64864,
+      "▁innoc": 64865,
+      "LOCATION": 64866,
+      "▁Patterns": 64867,
+      "▁craftsmanship": 64868,
+      "异步": 64869,
+      "我是个": 64870,
+      "是关于": 64871,
+      "进程中": 64872,
+      "ONLY": 64873,
+      "Risk": 64874,
+      "fang": 64875,
+      "doped": 64876,
+      "ubert": 64877,
+      "hetized": 64878,
+      "nMartin": 64879,
+      "瓦尔": 64880,
+      "福彩": 64881,
+      "PCB": 64882,
+      "也是在": 64883,
+      "右下角": 64884,
+      "这一年": 64885,
+      "ogel": 64886,
+      "TAINER": 64887,
+      "letion": 64888,
+      "▁Bulls": 64889,
+      "~(": 64890,
+      "田野": 64891,
+      "程式": 64892,
+      "自然而": 64893,
+      "纪委监委": 64894,
+      "paired": 64895,
+      "▁ethyl": 64896,
+      "▁excitatory": 64897,
+      "赏析": 64898,
+      "奥特曼": 64899,
+      "__(*": 64900,
+      "▁OWN": 64901,
+      "理论知识": 64902,
+      "aunch": 64903,
+      "ieren": 64904,
+      "▁Reds": 64905,
+      "▁posing": 64906,
+      "之于": 64907,
+      "抚育": 64908,
+      "有关系": 64909,
+      "让自己的": 64910,
+      "▁Narr": 64911,
+      "▁quota": 64912,
+      "FUNCTION": 64913,
+      "▁Reaction": 64914,
+      "▁multicenter": 64915,
+      "唯物": 64916,
+      "村级": 64917,
+      "▁函数": 64918,
+      "物体的": 64919,
+      "呼和浩特市": 64920,
+      "indrome": 64921,
+      "removeClass": 64922,
+      "创刊": 64923,
+      "差价": 64924,
+      "手绘": 64925,
+      "nHE": 64926,
+      "owo": 64927,
+      "之后再": 64928,
+      "eldom": 64929,
+      "▁amyg": 64930,
+      "visiae": 64931,
+      "▁shunt": 64932,
+      "依法提起诉讼": 64933,
+      "nVictor": 64934,
+      "▁veggies": 64935,
+      "文化传媒有限公司": 64936,
+      "出名": 64937,
+      "处以": 64938,
+      "家住": 64939,
+      "异侠": 64940,
+      "▁Elm": 64941,
+      "cakes": 64942,
+      "setItem": 64943,
+      "▁sucrose": 64944,
+      "▁favourable": 64945,
+      "克莱": 64946,
+      "挂靠": 64947,
+      "可以被": 64948,
+      "西游记": 64949,
+      "▁Nem": 64950,
+      "损伤程度": 64951,
+      "▁Beam": 64952,
+      "▁Cater": 64953,
+      "▁Expedition": 64954,
+      "▁expansions": 64955,
+      "▁吉": 64956,
+      "按期": 64957,
+      "自救": 64958,
+      "dup": 64959,
+      "都一样": 64960,
+      "▁UPS": 64961,
+      "▁gir": 64962,
+      "三大阶段": 64963,
+      "第三项裁定": 64964,
+      "▁Torres": 64965,
+      "uitively": 64966,
+      "▁kotlinx": 64967,
+      "marketing": 64968,
+      "▁teammate": 64969,
+      "traditional": 64970,
+      "永定": 64971,
+      "盗走": 64972,
+      "赋能": 64973,
+      "dea": 64974,
+      "▁天津": 64975,
+      "海口市": 64976,
+      "nVan": 64977,
+      "nstr": 64978,
+      "该局局长": 64979,
+      "▁gels": 64980,
+      "paring": 64981,
+      "▁docking": 64982,
+      "Infrastructure": 64983,
+      "▁Parliamentary": 64984,
+      "忧郁": 64985,
+      "肾病": 64986,
+      "这将": 64987,
+      "连结": 64988,
+      "▁显示": 64989,
+      "绿色的": 64990,
+      "▁Zurich": 64991,
+      "▁filler": 64992,
+      "▁prevailing": 64993,
+      "▁兰": 64994,
+      "幕墙": 64995,
+      "老挝": 64996,
+      "平板电脑": 64997,
+      "▁Width": 64998,
+      "▁Belgrade": 64999,
+      "▁melatonin": 65000,
+      "aui": 65001,
+      "不公平": 65002,
+      "人员在": 65003,
+      "卫生所": 65004,
+      "ivos": 65005,
+      "urin": 65006,
+      "Xgate": 65007,
+      "ipient": 65008,
+      "▁Dixon": 65009,
+      "▁stint": 65010,
+      "▁Whilst": 65011,
+      "segments": 65012,
+      "▁battling": 65013,
+      "▁anesthetic": 65014,
+      "ço": 65015,
+      "▁光": 65016,
+      "到头": 65017,
+      "太行": 65018,
+      "相切": 65019,
+      "nAS": 65020,
+      "veyor": 65021,
+      "民间借贷纠纷": 65022,
+      "Managed": 65023,
+      "▁Czechoslovak": 65024,
+      "财税": 65025,
+      "长发": 65026,
+      "▁faint": 65027,
+      "▁答": 65028,
+      "ット": 65029,
+      "丽娜": 65030,
+      "名城": 65031,
+      "换装": 65032,
+      "最值": 65033,
+      "烈的": 65034,
+      "等为": 65035,
+      "足额": 65036,
+      "ATM": 65037,
+      "atas": 65038,
+      "tipo": 65039,
+      "▁Arbor": 65040,
+      "IZATION": 65041,
+      "casters": 65042,
+      "▁caramel": 65043,
+      "Experimental": 65044,
+      "蓝光": 65045,
+      "入住时": 65046,
+      "有用的": 65047,
+      "因为没有": 65048,
+      "差不多了": 65049,
+      "decre": 65050,
+      "howto": 65051,
+      "▁Sell": 65052,
+      "osecond": 65053,
+      "▁locker": 65054,
+      "科技发展有限公司": 65055,
+      "玉石": 65056,
+      "▁深圳": 65057,
+      "去看看": 65058,
+      "网友们": 65059,
+      "骨干教师": 65060,
+      "▁Fold": 65061,
+      "▁seas": 65062,
+      "▁Flood": 65063,
+      "▁flames": 65064,
+      "XD": 65065,
+      "▁磊": 65066,
+      "因病": 65067,
+      "艳丽": 65068,
+      "nak": 65069,
+      "中没有": 65070,
+      "引发了": 65071,
+      "滁州市": 65072,
+      "Decl": 65073,
+      "imap": 65074,
+      "astatin": 65075,
+      "▁Fighter": 65076,
+      "Five": 65077,
+      "▁jan": 65078,
+      "icola": 65079,
+      "▁Whitney": 65080,
+      "准入": 65081,
+      "贬值": 65082,
+      "的客户": 65083,
+      "enas": 65084,
+      "wort": 65085,
+      "各个方面": 65086,
+      "▁Stroke": 65087,
+      "nAlternatively": 65088,
+      "查验": 65089,
+      "每秒": 65090,
+      "气场": 65091,
+      "疏通": 65092,
+      "秋色": 65093,
+      "失败的": 65094,
+      "ROLL": 65095,
+      "eday": 65096,
+      "还会提供": 65097,
+      "▁Chak": 65098,
+      "▁lumen": 65099,
+      "instant": 65100,
+      "analyzer": 65101,
+      "baseline": 65102,
+      "structural": 65103,
+      "等单位": 65104,
+      "▁Dys": 65105,
+      "公交线路": 65106,
+      "限制消费": 65107,
+      "Estim": 65108,
+      "▁Jeep": 65109,
+      "ellate": 65110,
+      "▁radii": 65111,
+      "▁Newark": 65112,
+      "▁incontinence": 65113,
+      "▁İ": 65114,
+      "心意": 65115,
+      "手艺": 65116,
+      "Ken": 65117,
+      "就只能": 65118,
+      "toos": 65119,
+      "▁Raz": 65120,
+      "正确判断": 65121,
+      "法律后果": 65122,
+      "autoload": 65123,
+      "combined": 65124,
+      "▁donating": 65125,
+      "ModelAdmin": 65126,
+      "▁geometries": 65127,
+      "▁degenerative": 65128,
+      "农夫": 65129,
+      "博彩": 65130,
+      "故居": 65131,
+      "本的": 65132,
+      "alom": 65133,
+      "varn": 65134,
+      "▁bfs": 65135,
+      "henko": 65136,
+      "atcher": 65137,
+      "▁Combat": 65138,
+      "▁voyage": 65139,
+      "亚历": 65140,
+      "动感": 65141,
+      "北欧": 65142,
+      "销售量": 65143,
+      "Bold": 65144,
+      "pink": 65145,
+      "▁ACM": 65146,
+      "▁cochlear": 65147,
+      "nBeautiful": 65148,
+      "▁Religious": 65149,
+      "户名": 65150,
+      "祝你": 65151,
+      "idf": 65152,
+      "的魅力": 65153,
+      "▁AMA": 65154,
+      "认定被告人": 65155,
+      "TagName": 65156,
+      "▁Refuge": 65157,
+      "▁Diffusion": 65158,
+      "▁enthusiast": 65159,
+      "性命": 65160,
+      "希望你": 65161,
+      "都必须": 65162,
+      "子叶植物": 65163,
+      "▁Lens": 65164,
+      "权纠纷一案中": 65165,
+      "▁Emmanuel": 65166,
+      "▁councils": 65167,
+      "▁Benchmark": 65168,
+      "▁empowering": 65169,
+      "司长": 65170,
+      "选修": 65171,
+      "bilt": 65172,
+      "▁ini": 65173,
+      "▁rash": 65174,
+      "▁Thames": 65175,
+      "PublicKey": 65176,
+      "▁compensated": 65177,
+      "▁indispensable": 65178,
+      "各县": 65179,
+      "补办": 65180,
+      "反映出": 65181,
+      "▁Barton": 65182,
+      "▁Damage": 65183,
+      "▁brochure": 65184,
+      "▁半": 65185,
+      "与张": 65186,
+      "人寿": 65187,
+      "本组": 65188,
+      "相色谱": 65189,
+      "allo": 65190,
+      "nFoot": 65191,
+      "▁Lazy": 65192,
+      "omatous": 65193,
+      "▁strands": 65194,
+      "▁Aircraft": 65195,
+      "Orientation": 65196,
+      "▁electronically": 65197,
+      "我做": 65198,
+      "Mas": 65199,
+      "▁房价": 65200,
+      "uala": 65201,
+      "Suffix": 65202,
+      "nright": 65203,
+      "▁Claud": 65204,
+      "第一项至第三项裁定": 65205,
+      "▁千": 65206,
+      "小鸟": 65207,
+      "铁矿": 65208,
+      "在任何": 65209,
+      "▁Cot": 65210,
+      "▁Eas": 65211,
+      "emporal": 65212,
+      "▁complemented": 65213,
+      "了然": 65214,
+      "发泄": 65215,
+      "物件": 65216,
+      "bfs": 65217,
+      "stk": 65218,
+      "vill": 65219,
+      "▁Uni": 65220,
+      "也是如此": 65221,
+      "▁unto": 65222,
+      "▁palsy": 65223,
+      "▁tallest": 65224,
+      "▁Religion": 65225,
+      "感触": 65226,
+      "采光": 65227,
+      "黄芪": 65228,
+      "未知数": 65229,
+      "求函数": 65230,
+      "▁hurry": 65231,
+      "▁tunnels": 65232,
+      "丽萍": 65233,
+      "的设备": 65234,
+      "ginia": 65235,
+      "▁baskets": 65236,
+      "▁mastering": 65237,
+      "▁navigating": 65238,
+      "自幼": 65239,
+      "宿迁市": 65240,
+      ")}$$": 65241,
+      "重点工作": 65242,
+      "erring": 65243,
+      "estring": 65244,
+      "▁insist": 65245,
+      "▁Unified": 65246,
+      "▁Enforcement": 65247,
+      "▁complexities": 65248,
+      "安吉": 65249,
+      "找准": 65250,
+      "前公示": 65251,
+      "实在太": 65252,
+      "第一百二十": 65253,
+      "▁enthal": 65254,
+      "nPURPOSE": 65255,
+      "▁Compiler": 65256,
+      "▁discarded": 65257,
+      "▁anisotropy": 65258,
+      "可疑": 65259,
+      "的优秀": 65260,
+      "heid": 65261,
+      "LOBAL": 65262,
+      "▁Lemon": 65263,
+      "Similar": 65264,
+      "synchron": 65265,
+      "▁cranial": 65266,
+      "▁depolar": 65267,
+      "▁loosely": 65268,
+      "▁biblical": 65269,
+      "▁minimized": 65270,
+      "▁symmetrical": 65271,
+      "Swift": 65272,
+      "▁aunt": 65273,
+      "Middle": 65274,
+      "▁Shock": 65275,
+      "二〇一四年七月": 65276,
+      "▁Regulatory": 65277,
+      "粘膜": 65278,
+      "跨度": 65279,
+      "principal": 65280,
+      "一文": 65281,
+      "pto": 65282,
+      "encils": 65283,
+      "ilevel": 65284,
+      "conditioning": 65285,
+      "Rew": 65286,
+      "最强的": 65287,
+      "遥控器": 65288,
+      "iten": 65289,
+      "任免信息": 65290,
+      "urface": 65291,
+      "▁Toast": 65292,
+      "步兵": 65293,
+      "等值": 65294,
+      "这让": 65295,
+      "▁Dud": 65296,
+      "▁SCC": 65297,
+      "住吉林省": 65298,
+      "nEasy": 65299,
+      "向本院起诉": 65300,
+      "▁downs": 65301,
+      "▁kappa": 65302,
+      "▁Polbot": 65303,
+      "▁wording": 65304,
+      "▁Christina": 65305,
+      "▁Scientist": 65306,
+      "▁Affordable": 65307,
+      "▁accusations": 65308,
+      "▁personalised": 65309,
+      "坏的": 65310,
+      "Say": 65311,
+      "所以他": 65312,
+      "请分析": 65313,
+      "ansi": 65314,
+      "市委书记": 65315,
+      "zhihu": 65316,
+      "▁vene": 65317,
+      "consult": 65318,
+      "SETTINGS": 65319,
+      "optimize": 65320,
+      "▁trumpet": 65321,
+      "▁speeches": 65322,
+      "SerializeField": 65323,
+      "▁preferentially": 65324,
+      "独生": 65325,
+      "在其他": 65326,
+      "athe": 65327,
+      "经相关部门": 65328,
+      "nPlants": 65329,
+      "▁Paragu": 65330,
+      "▁Trends": 65331,
+      "▁Messenger": 65332,
+      "▁sensitization": 65333,
+      "榆树": 65334,
+      "追星": 65335,
+      "将持续": 65336,
+      "试用期": 65337,
+      "配备了": 65338,
+      "▁CLE": 65339,
+      "comings": 65340,
+      "▁robbery": 65341,
+      "▁incarcer": 65342,
+      "nSomething": 65343,
+      "VERTISEMENT": 65344,
+      "喷射": 65345,
+      "山羊": 65346,
+      "BRO": 65347,
+      "Eth": 65348,
+      "XYZ": 65349,
+      "dad": 65350,
+      "无数的": 65351,
+      "▁myc": 65352,
+      "我们已经": 65353,
+      "▁soak": 65354,
+      "▁Vinyl": 65355,
+      "▁Firebase": 65356,
+      "▁Colombian": 65357,
+      "▁anomalous": 65358,
+      "声卡": 65359,
+      "正直": 65360,
+      "了一起": 65361,
+      "▁Aleks": 65362,
+      "▁PROFITS": 65363,
+      "▁grilled": 65364,
+      "印记": 65365,
+      "夫子": 65366,
+      "沥干": 65367,
+      "莱芜": 65368,
+      "▁GV": 65369,
+      "atican": 65370,
+      "Uniform": 65371,
+      "gameObject": 65372,
+      "▁predefined": 65373,
+      "夺取": 65374,
+      "小众": 65375,
+      "果蔬": 65376,
+      "▁用量": 65377,
+      "▁TWO": 65378,
+      "你知道吗": 65379,
+      "animal": 65380,
+      "文物保护单位": 65381,
+      "▁cowork": 65382,
+      "▁Pleasant": 65383,
+      "▁foreground": 65384,
+      "▁bioinformatics": 65385,
+      "滨江": 65386,
+      "能手": 65387,
+      "idl": 65388,
+      "和精神": 65389,
+      "的措施": 65390,
+      "Adam": 65391,
+      "Give": 65392,
+      "▁ANA": 65393,
+      "基本情况": 65394,
+      "apiro": 65395,
+      "€™": 65396,
+      "三五": 65397,
+      "中超": 65398,
+      "别离": 65399,
+      "核定": 65400,
+      "孙某某": 65401,
+      "工作站": 65402,
+      "的资源": 65403,
+      "▁Dag": 65404,
+      "本科毕业": 65405,
+      "▁thrill": 65406,
+      "积极参加生产劳动": 65407,
+      "Composite": 65408,
+      "上新": 65409,
+      "杨幂": 65410,
+      "流派": 65411,
+      "突起": 65412,
+      "nAw": 65413,
+      "▁Pup": 65414,
+      "节能减排": 65415,
+      "enronXgate": 65416,
+      "▁bioavailability": 65417,
+      "穿戴": 65418,
+      "这道题目": 65419,
+      "▁sweater": 65420,
+      "▁Speedway": 65421,
+      "▁journeys": 65422,
+      "MappingURL": 65423,
+      "▁transfection": 65424,
+      "nBer": 65425,
+      "▁Temp": 65426,
+      "emails": 65427,
+      "webhook": 65428,
+      "▁urlparse": 65429,
+      "▁addictive": 65430,
+      "和外": 65431,
+      "WNr": 65432,
+      "与美国": 65433,
+      "罗马尼亚": 65434,
+      "▁Lite": 65435,
+      "nCommand": 65436,
+      "ometrics": 65437,
+      "nMultiply": 65438,
+      "▁pharmacologic": 65439,
+      "而未": 65440,
+      "莲子": 65441,
+      "从来没": 65442,
+      "利物浦": 65443,
+      "nStay": 65444,
+      "▁_(\\'": 65445,
+      "▁Jakarta": 65446,
+      "conditioned": 65447,
+      "▁essentials": 65448,
+      "再者": 65449,
+      "托盘": 65450,
+      "民进党": 65451,
+      "opress": 65452,
+      "\"__": 65453,
+      "直接把": 65454,
+      "创新发展": 65455,
+      "进行审查": 65456,
+      "▁Infect": 65457,
+      "nickname": 65458,
+      "▁irritation": 65459,
+      "▁prescribing": 65460,
+      "对前款第一项至第三项裁定": 65461,
+      "▁firefighters": 65462,
+      "后宫": 65463,
+      "谁来": 65464,
+      "发生过": 65465,
+      "是基础题": 65466,
+      "▁deriving": 65467,
+      "▁Volleyball": 65468,
+      "▁Congressional": 65469,
+      "周二": 65470,
+      "芬芳": 65471,
+      "长兴": 65472,
+      "风向": 65473,
+      "SYS": 65474,
+      "可以不": 65475,
+      "的建筑": 65476,
+      "lein": 65477,
+      "协商解决": 65478,
+      "解不等式": 65479,
+      "国家税务总局": 65480,
+      "▁miners": 65481,
+      "▁Albania": 65482,
+      "▁interconnected": 65483,
+      "ай": 65484,
+      "息息": 65485,
+      "挤出": 65486,
+      "仙侠异侠": 65487,
+      "▁Ensemble": 65488,
+      "受偿": 65489,
+      "▁во": 65490,
+      "▁发布": 65491,
+      "和各种": 65492,
+      "的氛围": 65493,
+      "社会发展": 65494,
+      "Attempt": 65495,
+      "criteria": 65496,
+      "▁enclosure": 65497,
+      "此种": 65498,
+      "的年龄": 65499,
+      "nQuick": 65500,
+      "▁THEORY": 65501,
+      "▁UNESCO": 65502,
+      "characters": 65503,
+      "▁deletions": 65504,
+      "sourceforge": 65505,
+      "▁advantageous": 65506,
+      "无忌": 65507,
+      "oha": 65508,
+      "嘉兴市": 65509,
+      "依法应予": 65510,
+      "nMicro": 65511,
+      "申请执行人申请": 65512,
+      "▁aneurysms": 65513,
+      "▁vertebral": 65514,
+      "减免": 65515,
+      "左上": 65516,
+      "烟气": 65517,
+      "rfc": 65518,
+      "的日常": 65519,
+      "USIC": 65520,
+      "alty": 65521,
+      "决定如下": 65522,
+      "nthan": 65523,
+      "▁Kaplan": 65524,
+      "▁parench": 65525,
+      "▁teaspoon": 65526,
+      "▁mentorship": 65527,
+      "ZY": 65528,
+      "你又": 65529,
+      "削减": 65530,
+      "奋战": 65531,
+      "每亩": 65532,
+      "理会": 65533,
+      "wap": 65534,
+      "他已经": 65535,
+      "Chip": 65536,
+      "空气净化": 65537,
+      "\\\\}\\\\": 65538,
+      "现实生活中": 65539,
+      "nYoung": 65540,
+      "▁Stable": 65541,
+      "▁chicks": 65542,
+      "系其真实意思表示": 65543,
+      "nRepublic": 65544,
+      "▁skipping": 65545,
+      "▁courtyard": 65546,
+      "▁supplementary": 65547,
+      "内含": 65548,
+      ",)\\": 65549,
+      "▁{|": 65550,
+      "cele": 65551,
+      "可以获得": 65552,
+      "外形尺寸": 65553,
+      "kp": 65554,
+      "多地": 65555,
+      "手腕": 65556,
+      "报废": 65557,
+      "张艺兴": 65558,
+      "这一个": 65559,
+      "食用油": 65560,
+      "交易中心": 65561,
+      "unset": 65562,
+      "▁玄幻小说": 65563,
+      "信息技术有限公司": 65564,
+      "singleton": 65565,
+      "▁Existing": 65566,
+      "▁analgesic": 65567,
+      "▁辉": 65568,
+      "叶柄": 65569,
+      "源泉": 65570,
+      "整数部分": 65571,
+      "▁creek": 65572,
+      "▁deduce": 65573,
+      "▁miracle": 65574,
+      "▁smoother": 65575,
+      "薄的": 65576,
+      "与现代": 65577,
+      "▁Valle": 65578,
+      "▁axioms": 65579,
+      "▁majors": 65580,
+      "MetaData": 65581,
+      "tposition": 65582,
+      "▁competencies": 65583,
+      "是经": 65584,
+      "笑声": 65585,
+      "红了": 65586,
+      "Wood": 65587,
+      "adeon": 65588,
+      "driving": 65589,
+      "▁enforced": 65590,
+      "▁fundraiser": 65591,
+      "松动": 65592,
+      "病灶": 65593,
+      "元人民币": 65594,
+      "减刑幅度": 65595,
+      "教育资源": 65596,
+      "scrap": 65597,
+      "▁helix": 65598,
+      "▁Gather": 65599,
+      "Craft": 65600,
+      "▁SNPs": 65601,
+      "Resize": 65602,
+      "▁devil": 65603,
+      "▁Senators": 65604,
+      "%（": 65605,
+      "示威": 65606,
+      "▁经济": 65607,
+      "机场等": 65608,
+      "rove": 65609,
+      "出租车约": 65610,
+      "是不可能": 65611,
+      "▁corro": 65612,
+      "并按照对方当事人": 65613,
+      "Highlight": 65614,
+      "▁Northampton": 65615,
+      "▁beneficiaries": 65616,
+      "滚筒": 65617,
+      "慈溪市": 65618,
+      "▁Rental": 65619,
+      "二〇一八年一月": 65620,
+      "东升": 65621,
+      "eee": 65622,
+      "▁用途": 65623,
+      "本来就": 65624,
+      "woff": 65625,
+      "▁NMDA": 65626,
+      "vation": 65627,
+      "▁Rectangle": 65628,
+      "▁allowance": 65629,
+      "上场": 65630,
+      "厚重": 65631,
+      "的美国": 65632,
+      "Baby": 65633,
+      "nIss": 65634,
+      "inhib": 65635,
+      "▁tense": 65636,
+      "▁结": 65637,
+      "这台": 65638,
+      "不合适": 65639,
+      "买东西": 65640,
+      "出现过": 65641,
+      "和别人": 65642,
+      "的反应": 65643,
+      "重新启动": 65644,
+      "▁Rouge": 65645,
+      "ancelled": 65646,
+      "▁Institutes": 65647,
+      "好歹": 65648,
+      "增加到": 65649,
+      "战斗机": 65650,
+      "情侣约会": 65651,
+      "▁Lovely": 65652,
+      "nproject": 65653,
+      "▁Katherine": 65654,
+      "臭氧": 65655,
+      "的分类": 65656,
+      "▁miRNA": 65657,
+      "▁spars": 65658,
+      "▁Footer": 65659,
+      "▁randomised": 65660,
+      "堡垒": 65661,
+      "无常": 65662,
+      "是天": 65663,
+      "短篇": 65664,
+      "阳泉": 65665,
+      "探测器": 65666,
+      "深度学习": 65667,
+      "▁Tate": 65668,
+      "BUTTON": 65669,
+      "Throws": 65670,
+      "▁stroll": 65671,
+      "动荡": 65672,
+      "宁海": 65673,
+      "无害": 65674,
+      "imar": 65675,
+      "nenv": 65676,
+      "▁Fuk": 65677,
+      "与被告刘": 65678,
+      "varez": 65679,
+      "olymer": 65680,
+      "ouncill": 65681,
+      "不求": 65682,
+      "后排": 65683,
+      "MOS": 65684,
+      "▁Isa": 65685,
+      "phire": 65686,
+      "可能是因为": 65687,
+      "therapy": 65688,
+      "▁migrant": 65689,
+      "▁progresses": 65690,
+      "杨紫": 65691,
+      "病历": 65692,
+      "落幕": 65693,
+      "你真的": 65694,
+      "催告书": 65695,
+      "新品种": 65696,
+      "▁Trinidad": 65697,
+      "datasource": 65698,
+      "▁contralateral": 65699,
+      "▁₹": 65700,
+      "两层": 65701,
+      "卖场": 65702,
+      "▁SNA": 65703,
+      "gloss": 65704,
+      "▁zeta": 65705,
+      "号执行案件": 65706,
+      "▁pprint": 65707,
+      "▁DISCLAI": 65708,
+      "CW": 65709,
+      "借此": 65710,
+      "成活": 65711,
+      "hare": 65712,
+      ")\\')\\": 65713,
+      "bower": 65714,
+      "umina": 65715,
+      "显着性差异": 65716,
+      "给付之日止": 65717,
+      "▁PROJECT": 65718,
+      "▁minimization": 65719,
+      "不须": 65720,
+      "居士": 65721,
+      "浇水": 65722,
+      "音色": 65723,
+      "GIT": 65724,
+      "和维护": 65725,
+      "▁tcp": 65726,
+      "实践经验": 65727,
+      "▁Casa": 65728,
+      "▁小说类型": 65729,
+      "▁upregulation": 65730,
+      "下沉": 65731,
+      "烧伤": 65732,
+      "▁fir": 65733,
+      "从根本上": 65734,
+      "toolkit": 65735,
+      "▁traverse": 65736,
+      "▁½": 65737,
+      "urdish": 65738,
+      "▁Nicol": 65739,
+      "▁aided": 65740,
+      "▁comfy": 65741,
+      "▁snails": 65742,
+      "▁distracted": 65743,
+      "编委": 65744,
+      "SMS": 65745,
+      "可以向": 65746,
+      "编辑部": 65747,
+      "sbin": 65748,
+      "xlsx": 65749,
+      "▁Correlation": 65750,
+      "但有": 65751,
+      "末世": 65752,
+      "汕头": 65753,
+      "磁性": 65754,
+      "袜子": 65755,
+      "nLOG": 65756,
+      "cedures": 65757,
+      "▁efflux": 65758,
+      "▁sounding": 65759,
+      "▁accelerator": 65760,
+      "▁microvascular": 65761,
+      "的眼睛": 65762,
+      "大学教授": 65763,
+      "▁telecom": 65764,
+      "▁aspirations": 65765,
+      "▁乐": 65766,
+      "附上": 65767,
+      "停下来": 65768,
+      "▁soo": 65769,
+      "抗美援朝": 65770,
+      "▁adorn": 65771,
+      "nAssume": 65772,
+      "第六十七条第一款": 65773,
+      "▁Inspired": 65774,
+      "▁vibrations": 65775,
+      "受邀": 65776,
+      "多变": 65777,
+      "逆袭": 65778,
+      "好几个": 65779,
+      "skins": 65780,
+      "▁Auburn": 65781,
+      "滤芯": 65782,
+      "研修": 65783,
+      "雇主": 65784,
+      "cea": 65785,
+      "双方已": 65786,
+      "予以冻结": 65787,
+      "然后利用": 65788,
+      "ahead": 65789,
+      "nelif": 65790,
+      "二〇一五年四月": 65791,
+      "▁Minutes": 65792,
+      "最高人民法院关于审理民间借贷": 65793,
+      "ор": 65794,
+      "一试": 65795,
+      "燕窝": 65796,
+      "跳槽": 65797,
+      "Lar": 65798,
+      "jas": 65799,
+      "立足于": 65800,
+      "YWdl": 65801,
+      "▁foul": 65802,
+      "nApply": 65803,
+      "在这种情况下": 65804,
+      "▁hamlet": 65805,
+      "▁notorious": 65806,
+      "体征": 65807,
+      "的女孩子": 65808,
+      "uities": 65809,
+      "▁nitro": 65810,
+      "▁mechanic": 65811,
+      "基质": 65812,
+      "领事": 65813,
+      "寄生虫": 65814,
+      "培训中心": 65815,
+      "工作原理": 65816,
+      "承担责任": 65817,
+      "▁godd": 65818,
+      "▁infancy": 65819,
+      "ASHINGTON": 65820,
+      "▁Electoral": 65821,
+      "卫国": 65822,
+      "在南": 65823,
+      "并把": 65824,
+      "相约": 65825,
+      "ICC": 65826,
+      "不停的": 65827,
+      "二年级": 65828,
+      "uchs": 65829,
+      "债权文书": 65830,
+      "学术交流": 65831,
+      "Dummy": 65832,
+      "Early": 65833,
+      "上单调递增": 65834,
+      "▁prose": 65835,
+      "nCategory": 65836,
+      "esophageal": 65837,
+      "两级": 65838,
+      "最难": 65839,
+      "副产品": 65840,
+      "的那些": 65841,
+      "）×（": 65842,
+      "▁Seat": 65843,
+      "▁whale": 65844,
+      "利浦": 65845,
+      "weep": 65846,
+      "reaction": 65847,
+      "▁inspected": 65848,
+      "▁ammunition": 65849,
+      "安抚": 65850,
+      "惬意": 65851,
+      "cub": 65852,
+      "▁BF": 65853,
+      "灵活性": 65854,
+      "duit": 65855,
+      "实验小学": 65856,
+      "虽然没有": 65857,
+      "LOGIN": 65858,
+      "▁Soil": 65859,
+      "▁lakh": 65860,
+      "licted": 65861,
+      "▁gmail": 65862,
+      "nNovels": 65863,
+      "Simulation": 65864,
+      "\")（": 65865,
+      "是关键": 65866,
+      "presence": 65867,
+      "ág": 65868,
+      "永乐": 65869,
+      "给它": 65870,
+      "辱骂": 65871,
+      "沙坪坝": 65872,
+      "▁Obt": 65873,
+      "▁Claude": 65874,
+      "▁gourmet": 65875,
+      "▁reunion": 65876,
+      "▁strategically": 65877,
+      "七八糟": 65878,
+      "十三五": 65879,
+      "在路上": 65880,
+      "▁bei": 65881,
+      "跨境电商": 65882,
+      "Neill": 65883,
+      "▁mang": 65884,
+      "▁motto": 65885,
+      "▁inconvenience": 65886,
+      "不言": 65887,
+      "不走": 65888,
+      "吓得": 65889,
+      "非遗": 65890,
+      "qtt": 65891,
+      "ipay": 65892,
+      "▁presidents": 65893,
+      "封信": 65894,
+      "江岸": 65895,
+      "降落": 65896,
+      "百科全书": 65897,
+      "nDiff": 65898,
+      "剧组": 65899,
+      "战后": 65900,
+      "送往": 65901,
+      "问世": 65902,
+      "鲜血": 65903,
+      "因此而": 65904,
+      "责任制": 65905,
+      "ández": 65906,
+      "criptive": 65907,
+      "▁Diploma": 65908,
+      "愚蠢": 65909,
+      "滋生": 65910,
+      "找到一个": 65911,
+      "VISED": 65912,
+      "wives": 65913,
+      "▁Caps": 65914,
+      "nChart": 65915,
+      "二〇一七年一月": 65916,
+      "家境": 65917,
+      "降压": 65918,
+      "做单位": 65919,
+      "▁Skill": 65920,
+      "▁Amount": 65921,
+      "▁probation": 65922,
+      "▁topography": 65923,
+      "烧开": 65924,
+      "超时": 65925,
+      "威海市": 65926,
+      "úmer": 65927,
+      "家庭暴力": 65928,
+      "▁Veterinary": 65929,
+      "▁而": 65930,
+      "你可": 65931,
+      "牙龈": 65932,
+      "笑笑": 65933,
+      "都好": 65934,
+      "在规定": 65935,
+      "方便的": 65936,
+      "▁Ris": 65937,
+      "▁McKin": 65938,
+      "▁highs": 65939,
+      "Culture": 65940,
+      "▁covalent": 65941,
+      "邑县": 65942,
+      "顺风": 65943,
+      "Pix": 65944,
+      "是你的": 65945,
+      "▁Lob": 65946,
+      "tline": 65947,
+      "▁OPEN": 65948,
+      "nFiles": 65949,
+      "savefig": 65950,
+      "▁MathJax": 65951,
+      "▁ammonium": 65952,
+      "▁portfolios": 65953,
+      "下垂": 65954,
+      "地方的": 65955,
+      "奇怪的": 65956,
+      "dies": 65957,
+      "▁Ivy": 65958,
+      "▁Yak": 65959,
+      "▁lia": 65960,
+      "存储容量": 65961,
+      "▁pane": 65962,
+      "▁Assign": 65963,
+      "▁cerevisiae": 65964,
+      "线圈": 65965,
+      "'][": 65966,
+      "▁Tay": 65967,
+      "具有一定": 65968,
+      "分支机构": 65969,
+      "本文主要": 65970,
+      "datat": 65971,
+      "entin": 65972,
+      "▁predators": 65973,
+      "▁prostaglandin": 65974,
+      "现代人": 65975,
+      "▁PEG": 65976,
+      "molecule": 65977,
+      "▁plaster": 65978,
+      "▁soothing": 65979,
+      "▁connectors": 65980,
+      "不受申请执行期间的限制": 65981,
+      "设立了": 65982,
+      "bolt": 65983,
+      "▁glm": 65984,
+      "▁fMRI": 65985,
+      "▁disgu": 65986,
+      "nPython": 65987,
+      "▁remake": 65988,
+      "痴呆": 65989,
+      "骗局": 65990,
+      "所以就": 65991,
+      "我们现在": 65992,
+      "▁剧情简介": 65993,
+      "▁misuse": 65994,
+      "▁constituted": 65995,
+      "向你": 65996,
+      "旺季": 65997,
+      "对我的": 65998,
+      "精致的": 65999,
+      "▁EBV": 66000,
+      "销售收入": 66001,
+      "ALLED": 66002,
+      "ophon": 66003,
+      "chanics": 66004,
+      "Continue": 66005,
+      "▁centrally": 66006,
+      "StringUtils": 66007,
+      "在全": 66008,
+      "她还": 66009,
+      "远东": 66010,
+      "下午茶": 66011,
+      "两个方面": 66012,
+      "Notice": 66013,
+      "traits": 66014,
+      "▁rooft": 66015,
+      "▁shelters": 66016,
+      "▁consolidated": 66017,
+      "XS": 66018,
+      "耳边": 66019,
+      "该片": 66020,
+      "sess": 66021,
+      "▁Marshal": 66022,
+      "▁cations": 66023,
+      "▁signage": 66024,
+      "长老": 66025,
+      "收音机": 66026,
+      "甚至连": 66027,
+      "▁CSR": 66028,
+      "▁Ips": 66029,
+      "ombies": 66030,
+      "▁Elder": 66031,
+      "efficiency": 66032,
+      "挑衅": 66033,
+      "牵手": 66034,
+      "nEr": 66035,
+      "在执行": 66036,
+      "直接在": 66037,
+      "ricanes": 66038,
+      "买来": 66039,
+      "改名": 66040,
+      "薪水": 66041,
+      "CPA": 66042,
+      "OND": 66043,
+      "也是个": 66044,
+      "帮助你": 66045,
+      "近几年来": 66046,
+      "▁Bhat": 66047,
+      "▁Godd": 66048,
+      "▁Sacred": 66049,
+      "家和": 66050,
+      "▁kr": 66051,
+      "你想要": 66052,
+      "nVal": 66053,
+      "▁第三章": 66054,
+      "教学内容": 66055,
+      "ignon": 66056,
+      "被子植物门": 66057,
+      "▁Salon": 66058,
+      "References": 66059,
+      "东南部": 66060,
+      "巡视员": 66061,
+      "很好看": 66062,
+      "goog": 66063,
+      "正常运行": 66064,
+      "▁Canberra": 66065,
+      "▁retrofit": 66066,
+      "▁upstairs": 66067,
+      "▁Montenegro": 66068,
+      "gid": 66069,
+      "结果和": 66070,
+      "▁创办时间": 66071,
+      "phants": 66072,
+      "▁voltages": 66073,
+      "▁Consultant": 66074,
+      "存货": 66075,
+      "心仪": 66076,
+      "消肿": 66077,
+      "▁FY": 66078,
+      "下一次": 66079,
+      "\"\")\\": 66080,
+      "rews": 66081,
+      "行业标准": 66082,
+      "nGreek": 66083,
+      "▁Fuzzy": 66084,
+      "本院依法送达": 66085,
+      "湖南省长沙市": 66086,
+      "▁nominees": 66087,
+      "▁physicist": 66088,
+      "▁mobilization": 66089,
+      "姨妈": 66090,
+      "女演员": 66091,
+      "不会自动": 66092,
+      "系统集成": 66093,
+      "买卖合同纠纷": 66094,
+      "privile": 66095,
+      "▁Conway": 66096,
+      "▁mighty": 66097,
+      "nFebruary": 66098,
+      "交错": 66099,
+      "壁垒": 66100,
+      "既不": 66101,
+      "端子": 66102,
+      "活下去": 66103,
+      "是我们的": 66104,
+      "党史学习教育": 66105,
+      "▁Recruitment": 66106,
+      "▁BufferedReader": 66107,
+      "丽华": 66108,
+      "凯旋": 66109,
+      "绿水": 66110,
+      "▁sliced": 66111,
+      "第一百一十九条第": 66112,
+      "▁ventures": 66113,
+      "hf": 66114,
+      "召集": 66115,
+      "推移": 66116,
+      "校企": 66117,
+      "甄嬛": 66118,
+      "轮椅": 66119,
+      "工信部": 66120,
+      "的状况": 66121,
+      "mlin": 66122,
+      "lipsum": 66123,
+      "nullptr": 66124,
+      "▁DAMAGE": 66125,
+      "▁indist": 66126,
+      "菩提": 66127,
+      "zew": 66128,
+      "▁Kem": 66129,
+      ")}$$\\": 66130,
+      "moves": 66131,
+      "kappaB": 66132,
+      "CONNECT": 66133,
+      "aggregate": 66134,
+      "▁householder": 66135,
+      "▁nud": 66136,
+      "xmlns": 66137,
+      "仲裁裁决书": 66138,
+      "nOctober": 66139,
+      "▁capsules": 66140,
+      "nprotected": 66141,
+      "抓手": 66142,
+      "王永": 66143,
+      "缝合": 66144,
+      "题为": 66145,
+      "oxia": 66146,
+      "▁Muse": 66147,
+      "▁Crack": 66148,
+      "▁masterpiece": 66149,
+      "如梦": 66150,
+      "还应": 66151,
+      "ibal": 66152,
+      "公共安全": 66153,
+      "PLACE": 66154,
+      "actus": 66155,
+      "periodic": 66156,
+      "nStandard": 66157,
+      "▁Infinite": 66158,
+      "enterprise": 66159,
+      "▁Protestant": 66160,
+      "小雨": 66161,
+      "工夫": 66162,
+      "燕麦": 66163,
+      "绒毛": 66164,
+      "saf": 66165,
+      "价值和": 66166,
+      "转基因": 66167,
+      "spam": 66168,
+      "▁decon": 66169,
+      "▁Minimal": 66170,
+      "▁EventArgs": 66171,
+      "▁expressive": 66172,
+      "▁postmenopausal": 66173,
+      "带了": 66174,
+      "演戏": 66175,
+      "Sale": 66176,
+      "▁Planck": 66177,
+      "Japanese": 66178,
+      "▁disband": 66179,
+      "▁woodland": 66180,
+      "compliance": 66181,
+      "▁Reservoir": 66182,
+      "洗礼": 66183,
+      "EOS": 66184,
+      "中文系": 66185,
+      "规模化": 66186,
+      "ninto": 66187,
+      "▁freq": 66188,
+      "▁Verde": 66189,
+      "▁Hitler": 66190,
+      "▁Streets": 66191,
+      "▁visuals": 66192,
+      "populations": 66193,
+      "defaultValue": 66194,
+      "中国建设银行股份有限公司": 66195,
+      "会用": 66196,
+      "未还": 66197,
+      "问你": 66198,
+      "eele": 66199,
+      "gear": 66200,
+      "▁Miy": 66201,
+      "aceted": 66202,
+      "olateral": 66203,
+      "▁Dynasty": 66204,
+      "Throwable": 66205,
+      "放了": 66206,
+      "更需要": 66207,
+      "ndra": 66208,
+      "trig": 66209,
+      "都是一个": 66210,
+      "Kafka": 66211,
+      "▁counselling": 66212,
+      "伟业": 66213,
+      "居易": 66214,
+      "chel": 66215,
+      "证据不足": 66216,
+      "nAdvert": 66217,
+      "▁Jensen": 66218,
+      "▁calibrated": 66219,
+      "曲靖": 66220,
+      "近乎": 66221,
+      "▁enlisted": 66222,
+      "▁NEGLIGENCE": 66223,
+      "小刚": 66224,
+      "小号": 66225,
+      "满脸": 66226,
+      "自立": 66227,
+      "职业学校": 66228,
+      "银行贷款": 66229,
+      "▁Burl": 66230,
+      "sexual": 66231,
+      "▁Activ": 66232,
+      "二〇一四年六月": 66233,
+      "compiled": 66234,
+      "▁opioids": 66235,
+      "▁Documents": 66236,
+      "白细胞": 66237,
+      "负事故": 66238,
+      "过了一": 66239,
+      "▁Telugu": 66240,
+      "▁fluent": 66241,
+      "▁mushroom": 66242,
+      "▁pristine": 66243,
+      "▁Diagnosis": 66244,
+      "POP": 66245,
+      "以及在": 66246,
+      "diag": 66247,
+      "疫情期间": 66248,
+      "ccoli": 66249,
+      "▁Appeal": 66250,
+      "Pagination": 66251,
+      "还会提供一项住宿的免费政策": 66252,
+      "▁Rehabilitation": 66253,
+      "执保": 66254,
+      "母校": 66255,
+      "TGF": 66256,
+      "uml": 66257,
+      "充分不": 66258,
+      "现如今": 66259,
+      "的事儿": 66260,
+      "peon": 66261,
+      "prit": 66262,
+      "undef": 66263,
+      "▁Reduce": 66264,
+      "▁Manning": 66265,
+      "nMunicipalities": 66266,
+      "中大": 66267,
+      "畜禽": 66268,
+      "蜗牛": 66269,
+      "题干": 66270,
+      "发现问题": 66271,
+      "▁Shoes": 66272,
+      "▁Coverage": 66273,
+      "Kh": 66274,
+      "专场": 66275,
+      "徒步": 66276,
+      "我先": 66277,
+      "cyclic": 66278,
+      "nSimply": 66279,
+      "▁contours": 66280,
+      "Bt": 66281,
+      "仓位": 66282,
+      "超长": 66283,
+      "作者的": 66284,
+      "几年前": 66285,
+      "因数是": 66286,
+      "等主编": 66287,
+      "顺利的": 66288,
+      "▁Tox": 66289,
+      "功德": 66290,
+      "增设": 66291,
+      "头盔": 66292,
+      "地球上": 66293,
+      "地铁站": 66294,
+      "▁Psy": 66295,
+      "严格落实": 66296,
+      "存续期间": 66297,
+      "金融市场": 66298,
+      "▁Trying": 66299,
+      "抛开": 66300,
+      "VOL": 66301,
+      "信息和": 66302,
+      "凝聚力": 66303,
+      "pont": 66304,
+      "洗浴用品": 66305,
+      "▁Porto": 66306,
+      "▁lyric": 66307,
+      "▁scaffolds": 66308,
+      "县人民检察院指派检察员": 66309,
+      "(#": 66310,
+      "微风": 66311,
+      "这下": 66312,
+      "建筑业": 66313,
+      "要不是": 66314,
+      "▁ITS": 66315,
+      "司法救助": 66316,
+      "更多的人": 66317,
+      "▁文章目录": 66318,
+      "▁cyclin": 66319,
+      "▁borrowing": 66320,
+      "Ir": 66321,
+      "双子": 66322,
+      "一大堆": 66323,
+      "妇产科": 66324,
+      "玻璃钢": 66325,
+      "ENSION": 66326,
+      "Unlock": 66327,
+      "▁demos": 66328,
+      "期限不得超过": 66329,
+      "pressor": 66330,
+      "道上": 66331,
+      "пис": 66332,
+      "Goal": 66333,
+      "Poll": 66334,
+      "undi": 66335,
+      "并提供了": 66336,
+      "▁Colony": 66337,
+      "虽有": 66338,
+      "▁nr": 66339,
+      "付出了": 66340,
+      "disp": 66341,
+      "solo": 66342,
+      "ionate": 66343,
+      "iterate": 66344,
+      "serving": 66345,
+      "▁Cheese": 66346,
+      "▁ounces": 66347,
+      "▁dissipation": 66348,
+      "拥堵": 66349,
+      "成本的": 66350,
+      "时间里": 66351,
+      "}/>\\": 66352,
+      "特别喜欢": 66353,
+      "ocracy": 66354,
+      "▁Weiss": 66355,
+      "Explorer": 66356,
+      "几度": 66357,
+      "重置": 66358,
+      "ová": 66359,
+      "uzi": 66360,
+      "vil": 66361,
+      "▁BTC": 66362,
+      "▁gynec": 66363,
+      "▁Laravel": 66364,
+      "▁chronically": 66365,
+      "无损": 66366,
+      "沼气": 66367,
+      "调入": 66368,
+      "__,\\": 66369,
+      "个人感觉": 66370,
+      "agging": 66371,
+      "ystick": 66372,
+      "rophysics": 66373,
+      "▁Efficiency": 66374,
+      "及其它": 66375,
+      "才会有": 66376,
+      "来说是": 66377,
+      "第二批": 66378,
+      "nHel": 66379,
+      "▁Irr": 66380,
+      "▁Falk": 66381,
+      "ecolor": 66382,
+      "seller": 66383,
+      "▁Nielsen": 66384,
+      "▁distort": 66385,
+      "▁dopaminergic": 66386,
+      "成大": 66387,
+      "水壶": 66388,
+      "干净的": 66389,
+      "风险的": 66390,
+      "的是一个": 66391,
+      "MENTS": 66392,
+      "nendif": 66393,
+      "欧洲冠军联赛": 66394,
+      "Upgrade": 66395,
+      "▁persec": 66396,
+      "▁compromising": 66397,
+      "BH": 66398,
+      "反面": 66399,
+      "正向": 66400,
+      "缩减": 66401,
+      "▁dz": 66402,
+      "意想不到": 66403,
+      "alach": 66404,
+      "▁Bare": 66405,
+      "▁sexes": 66406,
+      "▁Relief": 66407,
+      "▁turbines": 66408,
+      "较好地": 66409,
+      "broad": 66410,
+      "Define": 66411,
+      "▁communal": 66412,
+      "▁advancements": 66413,
+      "出一": 66414,
+      "庄子": 66415,
+      "新安": 66416,
+      "脱口": 66417,
+      "乡政府": 66418,
+      "idle": 66419,
+      "nmap": 66420,
+      "知识分子": 66421,
+      "饮食习惯": 66422,
+      "▁Seal": 66423,
+      "nAward": 66424,
+      "▁finely": 66425,
+      "▁ontology": 66426,
+      "▁dermatitis": 66427,
+      "FH": 66428,
+      "ää": 66429,
+      "德语": 66430,
+      "üss": 66431,
+      "能满足": 66432,
+      "可能就是": 66433,
+      "dynam": 66434,
+      "▁Danger": 66435,
+      "▁Headers": 66436,
+      "▁Shepherd": 66437,
+      "▁ceramics": 66438,
+      "▁misconduct": 66439,
+      "之门": 66440,
+      "估价": 66441,
+      "所占": 66442,
+      "深造": 66443,
+      "稀缺": 66444,
+      "购销": 66445,
+      "辗转": 66446,
+      "OME": 66447,
+      "nRock": 66448,
+      "Attack": 66449,
+      "nResult": 66450,
+      "▁sausage": 66451,
+      "满怀": 66452,
+      "quiz": 66453,
+      "▁EMP": 66454,
+      "Compact": 66455,
+      "存量": 66456,
+      "鲜肉": 66457,
+      "▁mV": 66458,
+      "▁IBD": 66459,
+      "▁Kras": 66460,
+      "▁Trim": 66461,
+      "▁cephal": 66462,
+      "▁Already": 66463,
+      "▁Cheshire": 66464,
+      "▁Lifetime": 66465,
+      "宜居": 66466,
+      "那份": 66467,
+      "最准确": 66468,
+      "追究其": 66469,
+      "····": 66470,
+      "elier": 66471,
+      "passwd": 66472,
+      "并列": 66473,
+      "是该": 66474,
+      "锦赛": 66475,
+      "在设计": 66476,
+      "emas": 66477,
+      "迅速发展": 66478,
+      "▁Said": 66479,
+      "▁Allied": 66480,
+      "▁Wolver": 66481,
+      "▁shortcuts": 66482,
+      "▁Confederate": 66483,
+      "字样": 66484,
+      "民辖": 66485,
+      "给力": 66486,
+      "阴凉": 66487,
+      "本文对": 66488,
+      "▁'@/": 66489,
+      "_____": 66490,
+      "▁Gmail": 66491,
+      "▁amphib": 66492,
+      "今夜": 66493,
+      "鲜活": 66494,
+      "▁Feld": 66495,
+      "▁frontier": 66496,
+      "▁advisable": 66497,
+      "东昌": 66498,
+      "停产": 66499,
+      "红米": 66500,
+      "造谣": 66501,
+      "）;": 66502,
+      "体系和": 66503,
+      "最关键": 66504,
+      "tbody": 66505,
+      "提供了一个": 66506,
+      "具备执行条件时": 66507,
+      "▁decoded": 66508,
+      "内障": 66509,
+      "峨眉": 66510,
+      "shu": 66511,
+      ")**\\": 66512,
+      "nEver": 66513,
+      "ncould": 66514,
+      "tensive": 66515,
+      "▁preferable": 66516,
+      "侵袭": 66517,
+      "图上": 66518,
+      "型材": 66519,
+      "大头": 66520,
+      "形体": 66521,
+      "葱姜": 66522,
+      "血腥": 66523,
+      "UAL": 66524,
+      "▁HOW": 66525,
+      "QUERY": 66526,
+      "▁spear": 66527,
+      "styleUrls": 66528,
+      "号刑事附带民事判决": 66529,
+      "▁diagnosing": 66530,
+      "大招": 66531,
+      "小黑": 66532,
+      "摄取": 66533,
+      "纸箱": 66534,
+      "非得": 66535,
+      "$,\\": 66536,
+      "▁POP": 66537,
+      "▁STD": 66538,
+      "▁Clients": 66539,
+      "▁slipped": 66540,
+      "▁elevations": 66541,
+      "监事": 66542,
+      "酸甜": 66543,
+      "支持其": 66544,
+      "})$\\": 66545,
+      "自首情节": 66546,
+      "cosmos": 66547,
+      "▁pouch": 66548,
+      "▁Reagan": 66549,
+      "▁hypergly": 66550,
+      "▁sourcing": 66551,
+      "伯格": 66552,
+      "侵占": 66553,
+      "根治": 66554,
+      "遗弃": 66555,
+      "▁PN": 66556,
+      "pron": 66557,
+      "infer": 66558,
+      "年龄较大的": 66559,
+      "日刑满释放": 66560,
+      "▁heightened": 66561,
+      "▁residences": 66562,
+      "反省": 66563,
+      "液的": 66564,
+      "物语": 66565,
+      "',$": 66566,
+      "▁办学": 66567,
+      "▁距离": 66568,
+      "审理期间": 66569,
+      "积极探索": 66570,
+      "▁Shak": 66571,
+      "▁Vinc": 66572,
+      "的法定条件": 66573,
+      "osystem": 66574,
+      "findById": 66575,
+      "▁Enhanced": 66576,
+      "协助执行单位出具的财产查询回执": 66577,
+      "变大": 66578,
+      "多一点": 66579,
+      "▁Lys": 66580,
+      "▁HIGH": 66581,
+      "▁aval": 66582,
+      "时不需付费": 66583,
+      "pixels": 66584,
+      "▁Denis": 66585,
+      "城市人民法院": 66586,
+      "strength": 66587,
+      "▁concave": 66588,
+      "▁unpleasant": 66589,
+      "▁supplemental": 66590,
+      "姥姥": 66591,
+      "苍穹": 66592,
+      "创业的": 66593,
+      "开封市": 66594,
+      "点个赞": 66595,
+      "狐月山": 66596,
+      "nargs": 66597,
+      "Monday": 66598,
+      "nGraph": 66599,
+      "约是": 66600,
+      "罗汉": 66601,
+      "而至": 66602,
+      "订货": 66603,
+      "nBox": 66604,
+      "witz": 66605,
+      "或者约定": 66606,
+      "ovable": 66607,
+      "▁mercy": 66608,
+      "▁mantle": 66609,
+      "▁textured": 66610,
+      "几人": 66611,
+      "地势": 66612,
+      "智利": 66613,
+      "穿衣": 66614,
+      "愉快的": 66615,
+      "的表情": 66616,
+      "eeze": 66617,
+      "makeat": 66618,
+      "idental": 66619,
+      "▁Coupon": 66620,
+      "▁arsenic": 66621,
+      "▁Criteria": 66622,
+      "▁intraper": 66623,
+      "该文": 66624,
+      "▁颜色": 66625,
+      "heed": 66626,
+      "ooke": 66627,
+      "家庭教育": 66628,
+      "▁Levi": 66629,
+      "▁Kyoto": 66630,
+      "▁Stokes": 66631,
+      "作为被执行人的": 66632,
+      "▁desserts": 66633,
+      "▁renovations": 66634,
+      "大意": 66635,
+      "昆山市": 66636,
+      "骨质疏松": 66637,
+      "ocice": 66638,
+      "▁quench": 66639,
+      "▁composing": 66640,
+      "五次": 66641,
+      "音量": 66642,
+      "也不想": 66643,
+      "auff": 66644,
+      "fresh": 66645,
+      "▁MAPK": 66646,
+      "▁cereal": 66647,
+      "掉落": 66648,
+      "深入浅": 66649,
+      "受欢迎的": 66650,
+      "othelial": 66651,
+      "▁perioperative": 66652,
+      "Gy": 66653,
+      "不只": 66654,
+      "回想": 66655,
+      "的联系": 66656,
+      "一段文字": 66657,
+      "总体规划": 66658,
+      "▁Lact": 66659,
+      "▁Scientists": 66660,
+      "预装": 66661,
+      "最小公": 66662,
+      "的关注": 66663,
+      "问一下": 66664,
+      "mqtt": 66665,
+      "▁gust": 66666,
+      "住黑龙江省": 66667,
+      "▁Monkey": 66668,
+      "▁Zambia": 66669,
+      "▁praying": 66670,
+      "nChurches": 66671,
+      "▁Promotion": 66672,
+      "一再": 66673,
+      "学区": 66674,
+      "的商": 66675,
+      "▁CPR": 66676,
+      "▁Auction": 66677,
+      "▁Wilhelm": 66678,
+      "▁dilation": 66679,
+      "是连载于晋江文学城": 66680,
+      "▁conjugated": 66681,
+      "▁manifested": 66682,
+      "▁interviewing": 66683,
+      "剥削": 66684,
+      "县域": 66685,
+      "见习": 66686,
+      "▁EG": 66687,
+      "原发性": 66688,
+      "▁admir": 66689,
+      "blastoma": 66690,
+      "▁Harvest": 66691,
+      "▁conducts": 66692,
+      "时隔": 66693,
+      "Kar": 66694,
+      "生动的": 66695,
+      "的危害": 66696,
+      "的大部分": 66697,
+      "▁Roosevelt": 66698,
+      "光环": 66699,
+      "多岁": 66700,
+      "特工": 66701,
+      "Kim": 66702,
+      "作品集": 66703,
+      "▁HIS": 66704,
+      "满足条件": 66705,
+      "nLists": 66706,
+      "medicine": 66707,
+      "▁Tracker": 66708,
+      "▁magnets": 66709,
+      "▁additives": 66710,
+      "▁Generalized": 66711,
+      "▁А": 66712,
+      "可与": 66713,
+      "解脱": 66714,
+      "瑞安市": 66715,
+      "英烈网": 66716,
+      "▁ADP": 66717,
+      "chrane": 66718,
+      "▁Aston": 66719,
+      "▁Mitch": 66720,
+      "▁Stark": 66721,
+      "经过财产调查": 66722,
+      "estimate": 66723,
+      "▁discard": 66724,
+      "Xi": 66725,
+      "功用": 66726,
+      "圣人": 66727,
+      "FDA": 66728,
+      "和学生": 66729,
+      "ecca": 66730,
+      "给定一个": 66731,
+      "▁bait": 66732,
+      "房地产市场": 66733,
+      "really": 66734,
+      "ndepend": 66735,
+      "▁Limits": 66736,
+      "▁brilli": 66737,
+      "▁Arabian": 66738,
+      "▁filePath": 66739,
+      "zx": 66740,
+      "▁MH": 66741,
+      "▁来源": 66742,
+      ">\">\\": 66743,
+      "olev": 66744,
+      "peek": 66745,
+      "▁sop": 66746,
+      "▁Candida": 66747,
+      "▁Procedure": 66748,
+      "▁Georgetown": 66749,
+      "▁redundancy": 66750,
+      "地坪": 66751,
+      "外貌": 66752,
+      "和提高": 66753,
+      "这个名字": 66754,
+      "国家统计局": 66755,
+      "ermann": 66756,
+      "▁brewing": 66757,
+      "nBuilding": 66758,
+      "▁Normally": 66759,
+      "催款": 66760,
+      "绝佳": 66761,
+      "采收": 66762,
+      "很开心": 66763,
+      "eston": 66764,
+      "▁Lunch": 66765,
+      "assigned": 66766,
+      "Coordinate": 66767,
+      "所学": 66768,
+      "更像": 66769,
+      "碳水化合物": 66770,
+      "WITHOUT": 66771,
+      "▁Nobody": 66772,
+      "▁asymmet": 66773,
+      "でき": 66774,
+      "守望": 66775,
+      "就近": 66776,
+      "GTM": 66777,
+      "等多项": 66778,
+      "公司拥有": 66779,
+      "可能出现": 66780,
+      "▁closures": 66781,
+      "台币": 66782,
+      "大涨": 66783,
+      "年金": 66784,
+      "戒毒": 66785,
+      "是一门": 66786,
+      "第一期": 66787,
+      "chalk": 66788,
+      "nfull": 66789,
+      "severity": 66790,
+      "▁Tunisia": 66791,
+      "`<": 66792,
+      "▁词": 66793,
+      "害羞": 66794,
+      "院的": 66795,
+      "лов": 66796,
+      "uddle": 66797,
+      "▁Elim": 66798,
+      "▁scler": 66799,
+      "Coverage": 66800,
+      "▁jackets": 66801,
+      "▁Managers": 66802,
+      "撤离": 66803,
+      "杂草": 66804,
+      "也不敢": 66805,
+      "营商环境": 66806,
+      "their": 66807,
+      "▁vacu": 66808,
+      "采取保全措施后三十日内": 66809,
+      "▁particulate": 66810,
+      "标价": 66811,
+      "范例": 66812,
+      "霉菌": 66813,
+      "▁Pé": 66814,
+      "优秀奖": 66815,
+      "要考虑": 66816,
+      "nSay": 66817,
+      "国土面积": 66818,
+      "▁Penny": 66819,
+      "romycin": 66820,
+      "▁pacing": 66821,
+      "▁wikipedia": 66822,
+      "天成": 66823,
+      "纷争": 66824,
+      "_);\\": 66825,
+      "劳动模范": 66826,
+      "圆的周长": 66827,
+      "十分重要的": 66828,
+      "dollar": 66829,
+      "▁Bella": 66830,
+      "elapsed": 66831,
+      "▁superhero": 66832,
+      "救命": 66833,
+      "无须": 66834,
+      "荷叶": 66835,
+      "天花板": 66836,
+      "▁revisit": 66837,
+      "▁upregulated": 66838,
+      "扎根": 66839,
+      "能做": 66840,
+      "beg": 66841,
+      "为确保": 66842,
+      "的生长": 66843,
+      "开发利用": 66844,
+      "nCheers": 66845,
+      "plectic": 66846,
+      "▁notebooks": 66847,
+      "刚性": 66848,
+      "▁重量": 66849,
+      "Ping": 66850,
+      "师资力量": 66851,
+      "▁nicer": 66852,
+      "▁Boulder": 66853,
+      "▁monarch": 66854,
+      "transpose": 66855,
+      "聚氨": 66856,
+      "的少年": 66857,
+      "eson": 66858,
+      "▁UCI": 66859,
+      "企业提供": 66860,
+      "各行各业": 66861,
+      "▁worms": 66862,
+      "▁Airways": 66863,
+      "lı": 66864,
+      "干啥": 66865,
+      "orio": 66866,
+      "nWeekly": 66867,
+      "▁\\'\\'\\'\\": 66868,
+      "▁textual": 66869,
+      "▁follicles": 66870,
+      "▁prerequisite": 66871,
+      "交货": 66872,
+      "木头": 66873,
+      "祈祷": 66874,
+      "请看": 66875,
+      "酱香": 66876,
+      "▁Ré": 66877,
+      "不那么": 66878,
+      "出去玩": 66879,
+      "刺激性": 66880,
+      "当一个": 66881,
+      "▁Ath": 66882,
+      "▁HIF": 66883,
+      "侵权行为": 66884,
+      "▁Responses": 66885,
+      "▁⟨": 66886,
+      "药学": 66887,
+      "grab": 66888,
+      "临床症状": 66889,
+      "浪费时间": 66890,
+      "精益求精": 66891,
+      "▁circa": 66892,
+      "▁HOWEVER": 66893,
+      "scheduled": 66894,
+      "▁Fletcher": 66895,
+      "▁民": 66896,
+      "八一": 66897,
+      "虹桥": 66898,
+      "遵义市": 66899,
+      "新开传奇": 66900,
+      "▁roofs": 66901,
+      "▁condensation": 66902,
+      "dala": 66903,
+      "的人力资源": 66904,
+      "▁Megan": 66905,
+      "小猪": 66906,
+      "毕节": 66907,
+      "的实施": 66908,
+      "▁Hou": 66909,
+      "cipher": 66910,
+      "offline": 66911,
+      "不问": 66912,
+      "射频": 66913,
+      "打球": 66914,
+      "跟鞋": 66915,
+      "CDC": 66916,
+      "rets": 66917,
+      "umpy": 66918,
+      "决定执行": 66919,
+      "第二十九": 66920,
+      "nBlue": 66921,
+      "Algebra": 66922,
+      "▁helium": 66923,
+      "▁carrots": 66924,
+      "▁Enterprises": 66925,
+      "▁variational": 66926,
+      "▁成都": 66927,
+      "会因为": 66928,
+      "在本次": 66929,
+      "▁BRCA": 66930,
+      "tforeach": 66931,
+      "callbacks": 66932,
+      "▁amorphous": 66933,
+      "是在法律允许的范围内": 66934,
+      "中是": 66935,
+      "地审查": 66936,
+      "日用品": 66937,
+      "课堂上": 66938,
+      "ogeny": 66939,
+      "▁deps": 66940,
+      "▁François": 66941,
+      "▁reopened": 66942,
+      "对边": 66943,
+      "新星": 66944,
+      "求生": 66945,
+      "额头": 66946,
+      "家公司": 66947,
+      "欧洲三大": 66948,
+      "年参演电影": 66949,
+      "\\\\\",\\\\\"": 66950,
+      "▁bullets": 66951,
+      "听得": 66952,
+      "颠倒": 66953,
+      "Ped": 66954,
+      "与实践": 66955,
+      "则需要": 66956,
+      "心灵的": 66957,
+      "Being": 66958,
+      "▁tenor": 66959,
+      "nRunning": 66960,
+      "▁manuals": 66961,
+      "PropTypes": 66962,
+      "▁vocational": 66963,
+      "一族": 66964,
+      "企鹅": 66965,
+      "资源共享": 66966,
+      "▁penny": 66967,
+      "▁Telegraph": 66968,
+      "▁allogeneic": 66969,
+      "▁prosthetic": 66970,
+      "▁Opportunity": 66971,
+      "一般都": 66972,
+      "交警大队": 66973,
+      "▁QtCore": 66974,
+      "▁Titans": 66975,
+      "sendMessage": 66976,
+      "▁commodities": 66977,
+      "充沛": 66978,
+      "丝绸之路": 66979,
+      "二〇一七年八": 66980,
+      "终结本案的本次执行程序": 66981,
+      "▁catastrophic": 66982,
+      "▁atherosclerotic": 66983,
+      "沿途": 66984,
+      "版纸": 66985,
+      "电费": 66986,
+      "他指出": 66987,
+      "Eric": 66988,
+      "▁ACS": 66989,
+      "▁OCT": 66990,
+      "适用范围": 66991,
+      "ocode": 66992,
+      "ynthia": 66993,
+      "▁Murder": 66994,
+      "▁immunologic": 66995,
+      "▁Conditioning": 66996,
+      "桂园": 66997,
+      "盐城": 66998,
+      "简算": 66999,
+      "▁Lé": 67000,
+      "hetti": 67001,
+      "enches": 67002,
+      "▁Ferry": 67003,
+      "开个": 67004,
+      "结识": 67005,
+      "绞痛": 67006,
+      "nya": 67007,
+      "▁BK": 67008,
+      "都有着": 67009,
+      "habi": 67010,
+      "还款期限": 67011,
+      "Artist": 67012,
+      "▁miRNAs": 67013,
+      "variants": 67014,
+      "律师事务所实习律师": 67015,
+      "▁Unlimited": 67016,
+      "确有悔改表现的事实清楚": 67017,
+      "能给": 67018,
+      "要让": 67019,
+      "▁DSM": 67020,
+      "在一起的": 67021,
+      "▁软件名称": 67022,
+      "▁pitches": 67023,
+      "▁immersive": 67024,
+      "ruptedException": 67025,
+      "云飞": 67026,
+      "大宗": 67027,
+      "太差": 67028,
+      "狠抓": 67029,
+      "鸽子": 67030,
+      "cab": 67031,
+      "引导学生": 67032,
+      "忘记密码": 67033,
+      "▁gown": 67034,
+      "mington": 67035,
+      "▁pantry": 67036,
+      "承担连带清偿责任": 67037,
+      "▁reciproc": 67038,
+      "cloudflare": 67039,
+      "宝典": 67040,
+      "[])\\": 67041,
+      "nNews": 67042,
+      "▁McCl": 67043,
+      "对该犯减刑": 67044,
+      "的经营理念": 67045,
+      "▁ureter": 67046,
+      "nProgram": 67047,
+      "▁revital": 67048,
+      "MDA": 67049,
+      "casts": 67050,
+      "Europe": 67051,
+      "▁homic": 67052,
+      "analytic": 67053,
+      "▁Sherman": 67054,
+      "▁popping": 67055,
+      "▁Philippe": 67056,
+      "同月": 67057,
+      "石狮": 67058,
+      "防卫": 67059,
+      "的宗旨": 67060,
+      "▁cops": 67061,
+      "▁gzip": 67062,
+      "estead": 67063,
+      "▁Wander": 67064,
+      "▁Domestic": 67065,
+      "出动": 67066,
+      "呼气": 67067,
+      "巴勒": 67068,
+      "幽灵": 67069,
+      "思义": 67070,
+      "钱了": 67071,
+      "moon": 67072,
+      "enity": 67073,
+      "ipine": 67074,
+      "▁Yaml": 67075,
+      "ート": 67076,
+      "导热": 67077,
+      "昨夜": 67078,
+      "是中": 67079,
+      "了一位": 67080,
+      "武汉大学": 67081,
+      "omencl": 67082,
+      "iferous": 67083,
+      "▁badges": 67084,
+      "▁pleural": 67085,
+      "▁semigroup": 67086,
+      "与前": 67087,
+      "修身": 67088,
+      "都快": 67089,
+      "门面": 67090,
+      "opre": 67091,
+      "odore": 67092,
+      "waves": 67093,
+      "▁Mickey": 67094,
+      "▁troubleshooting": 67095,
+      "一趟": 67096,
+      "连续的": 67097,
+      "▁Recogn": 67098,
+      "▁Taliban": 67099,
+      "长和": 67100,
+      "）∵": 67101,
+      "▁wo": 67102,
+      "记忆力": 67103,
+      "IFIED": 67104,
+      "agara": 67105,
+      "ghost": 67106,
+      "texit": 67107,
+      "▁Quantity": 67108,
+      "下限": 67109,
+      "黄岛": 67110,
+      "ере": 67111,
+      "▁爱情": 67112,
+      "绝对的": 67113,
+      "突出贡献": 67114,
+      "▁alph": 67115,
+      "▁Chlor": 67116,
+      "{,": 67117,
+      "了其": 67118,
+      "%)\\": 67119,
+      "联系的": 67120,
+      "iran": 67121,
+      "可以做到": 67122,
+      "ograf": 67123,
+      "▁mason": 67124,
+      "Ci": 67125,
+      "三江": 67126,
+      "常温": 67127,
+      "弱点": 67128,
+      "水墨": 67129,
+      "而为": 67130,
+      "与被告陈": 67131,
+      "▁dass": 67132,
+      "▁dumps": 67133,
+      "credible": 67134,
+      "▁seekers": 67135,
+      "▁Processes": 67136,
+      "放缓": 67137,
+      "LABEL": 67138,
+      "ricula": 67139,
+      "▁fluxes": 67140,
+      ":*": 67141,
+      "介于": 67142,
+      "每到": 67143,
+      "表层": 67144,
+      "▁第七": 67145,
+      "不仅有": 67146,
+      "不好看": 67147,
+      "双方均": 67148,
+      "正式版": 67149,
+      "belt": 67150,
+      "▁laps": 67151,
+      "ungeon": 67152,
+      "(\\'/\\',": 67153,
+      "Matches": 67154,
+      "Startup": 67155,
+      "▁Dorothy": 67156,
+      "▁ignition": 67157,
+      "лю": 67158,
+      "未有": 67159,
+      "顽强": 67160,
+      "odus": 67161,
+      "▁MESS": 67162,
+      "▁Takes": 67163,
+      "申请撤回对被告": 67164,
+      "令我": 67165,
+      "省心": 67166,
+      "▁OPT": 67167,
+      "在执行中": 67168,
+      "长期从事": 67169,
+      "▁nmol": 67170,
+      "▁sess": 67171,
+      "▁arach": 67172,
+      "▁whats": 67173,
+      "arrants": 67174,
+      "▁Attention": 67175,
+      "[,": 67176,
+      "ADO": 67177,
+      "模式下": 67178,
+      "cccc": 67179,
+      "可以有效": 67180,
+      "经相关部门批准后": 67181,
+      "下乡": 67182,
+      "佳能": 67183,
+      "loy": 67184,
+      "分子式": 67185,
+      "台湾的": 67186,
+      "对于一个": 67187,
+      "nFeel": 67188,
+      "issors": 67189,
+      "▁Graves": 67190,
+      "▁Stakes": 67191,
+      "▁litres": 67192,
+      "▁contingent": 67193,
+      "▁microstructure": 67194,
+      "堕落": 67195,
+      "▁湖北": 67196,
+      "开发了": 67197,
+      "日期为": 67198,
+      "▁Fee": 67199,
+      "乱七八糟": 67200,
+      "有异议的": 67201,
+      "▁INST": 67202,
+      "▁plasm": 67203,
+      "▁originate": 67204,
+      "▁disseminated": 67205,
+      "丁香": 67206,
+      "之争": 67207,
+      "▁ost": 67208,
+      "以上学历": 67209,
+      "文明单位": 67210,
+      "patrick": 67211,
+      "▁Bavaria": 67212,
+      "▁Cochrane": 67213,
+      "▁Operators": 67214,
+      "▁Collaborative": 67215,
+      "我司": 67216,
+      "真挚": 67217,
+      "mnt": 67218,
+      "即可得解": 67219,
+      "环境影响": 67220,
+      "Dream": 67221,
+      "▁英文参考": 67222,
+      "▁prefrontal": 67223,
+      "一新": 67224,
+      "谁家": 67225,
+      "收费标准": 67226,
+      "Connor": 67227,
+      "▁Assets": 67228,
+      "▁STRICT": 67229,
+      "nScience": 67230,
+      "▁wrinkles": 67231,
+      "▁易": 67232,
+      "入味": 67233,
+      "后有": 67234,
+      "试探": 67235,
+      "赢家": 67236,
+      "那么的": 67237,
+      "深入了解": 67238,
+      "网络游戏": 67239,
+      "该行行长": 67240,
+      "Reverse": 67241,
+      "▁prefers": 67242,
+      "▁sterile": 67243,
+      "▁villain": 67244,
+      "▁benefited": 67245,
+      "▁forwarded": 67246,
+      "列宁": 67247,
+      "纬度": 67248,
+      "希望对": 67249,
+      "时间短": 67250,
+      "Matt": 67251,
+      "cole": 67252,
+      "▁Hipp": 67253,
+      "安市人民法院": 67254,
+      "itively": 67255,
+      "▁Giving": 67256,
+      "nDiscover": 67257,
+      "napiVersion": 67258,
+      "▁illustrator": 67259,
+      "xn": 67260,
+      "南湖": 67261,
+      "▁用户": 67262,
+      "情况及": 67263,
+      "ateurs": 67264,
+      "▁cmake": 67265,
+      "atholic": 67266,
+      "▁Napole": 67267,
+      "▁Unable": 67268,
+      "▁PRIMARY": 67269,
+      "translational": 67270,
+      "个点": 67271,
+      "优酷": 67272,
+      "则可": 67273,
+      "拼图": 67274,
+      "玩过": 67275,
+      "白沙": 67276,
+      "ppt": 67277,
+      "去哪里": 67278,
+      "法学院": 67279,
+      "经过点": 67280,
+      "(/\\\\": 67281,
+      "Vault": 67282,
+      "Summer": 67283,
+      "speaker": 67284,
+      "在服刑期间的表现": 67285,
+      "侍郎": 67286,
+      "每股": 67287,
+      "走私": 67288,
+      "}`,": 67289,
+      "是正确的": 67290,
+      "▁você": 67291,
+      "linger": 67292,
+      "nDecember": 67293,
+      "▁compressor": 67294,
+      "до": 67295,
+      "▁字": 67296,
+      "即事": 67297,
+      "受限": 67298,
+      "巨额": 67299,
+      "得失": 67300,
+      "蚊子": 67301,
+      "bsd": 67302,
+      "ocide": 67303,
+      "▁Elev": 67304,
+      "▁Bihar": 67305,
+      "▁emoji": 67306,
+      "compatibility": 67307,
+      "观望": 67308,
+      "punk": 67309,
+      "有一部分": 67310,
+      "▁Yacht": 67311,
+      "ceiving": 67312,
+      "▁Mohamed": 67313,
+      "▁suggestive": 67314,
+      "供给侧": 67315,
+      "受害人": 67316,
+      "bart": 67317,
+      "styl": 67318,
+      "不是什么": 67319,
+      "paginate": 67320,
+      "▁audition": 67321,
+      "▁prolactin": 67322,
+      "▁葛": 67323,
+      "去何": 67324,
+      "用语": 67325,
+      "来得及": 67326,
+      "的衣服": 67327,
+      "pond": 67328,
+      "目前已经": 67329,
+      "▁Greatest": 67330,
+      "▁medically": 67331,
+      "变性": 67332,
+      "无能": 67333,
+      "笔误": 67334,
+      "党组副书记": 67335,
+      "▁Spectro": 67336,
+      "▁planners": 67337,
+      "▁tolerant": 67338,
+      "不败": 67339,
+      "多用": 67340,
+      "有情": 67341,
+      "要走": 67342,
+      "▁wi": 67343,
+      "大多数的": 67344,
+      "成为一名": 67345,
+      "▁PLAY": 67346,
+      "▁Wellness": 67347,
+      "▁Newtonsoft": 67348,
+      "▁会": 67349,
+      "地对": 67350,
+      "说的话": 67351,
+      "IDEO": 67352,
+      "都是为了": 67353,
+      "▁Locks": 67354,
+      "PROCESS": 67355,
+      "enumerate": 67356,
+      "Invocation": 67357,
+      "ве": 67358,
+      "开播": 67359,
+      "认购": 67360,
+      "的不断": 67361,
+      "被授予": 67362,
+      "combe": 67363,
+      "nthem": 67364,
+      "▁Polyte": 67365,
+      "▁widths": 67366,
+      "▁garment": 67367,
+      "▁jurisdictions": 67368,
+      "卡牌": 67369,
+      "东营市": 67370,
+      "▁gtk": 67371,
+      "儿童或者": 67372,
+      "▁sailed": 67373,
+      "▁Bacillus": 67374,
+      "▁surgically": 67375,
+      "酒店还会提供一项住宿的免费政策": 67376,
+      "人脉": 67377,
+      "无用": 67378,
+      "玩乐": 67379,
+      "重回": 67380,
+      "▁|>": 67381,
+      "bots": 67382,
+      "libc": 67383,
+      "▁Aster": 67384,
+      "▁telesc": 67385,
+      "▁Circular": 67386,
+      "▁settlers": 67387,
+      "Jean": 67388,
+      "▁UAV": 67389,
+      "尽量不要": 67390,
+      "ughed": 67391,
+      "评审委员会": 67392,
+      "FFFFFF": 67393,
+      "inians": 67394,
+      "▁Rocks": 67395,
+      "或者申请仲裁": 67396,
+      "Printer": 67397,
+      "ranking": 67398,
+      "▁Declaration": 67399,
+      "哎呀": 67400,
+      "售楼": 67401,
+      "案情": 67402,
+      "步枪": 67403,
+      "阳镇": 67404,
+      "的场景": 67405,
+      "genes": 67406,
+      "might": 67407,
+      "进一步完善": 67408,
+      "ostomy": 67409,
+      "▁rounding": 67410,
+      "nCitations": 67411,
+      "Constraints": 67412,
+      "学风": 67413,
+      "树下": 67414,
+      "ppm": 67415,
+      "发挥着": 67416,
+      "系统性": 67417,
+      "每平方米": 67418,
+      "病毒感染": 67419,
+      "nhtml": 67420,
+      "spots": 67421,
+      "▁Blvd": 67422,
+      "▁dispose": 67423,
+      "directive": 67424,
+      "▁Workflow": 67425,
+      "▁rectangles": 67426,
+      "边框": 67427,
+      "连线": 67428,
+      "长篇": 67429,
+      "摄氏度": 67430,
+      "的次数": 67431,
+      "“\\\\(": 67432,
+      "率分别为": 67433,
+      "含有丰富的": 67434,
+      "nApart": 67435,
+      "▁CLASS": 67436,
+      "▁staircase": 67437,
+      "では": 67438,
+      "一动": 67439,
+      "开方": 67440,
+      "银色": 67441,
+      "而是在": 67442,
+      "hive": 67443,
+      "宣传活动": 67444,
+      "工商管理": 67445,
+      "裁定驳回": 67446,
+      "Techn": 67447,
+      "▁LGBTQ": 67448,
+      "▁caves": 67449,
+      "▁IOError": 67450,
+      "▁Waterloo": 67451,
+      "Ru": 67452,
+      "不愧": 67453,
+      "卒中": 67454,
+      "增压": 67455,
+      "菜鸟": 67456,
+      "Dar": 67457,
+      "义务的": 67458,
+      "esser": 67459,
+      "urved": 67460,
+      "▁guessed": 67461,
+      "▁简": 67462,
+      "之恋": 67463,
+      "瘫痪": 67464,
+      "Eye": 67465,
+      "与客户": 67466,
+      "我开始": 67467,
+      "▁/><": 67468,
+      "facts": 67469,
+      "▁calor": 67470,
+      "Germany": 67471,
+      "▁estimators": 67472,
+      "万方": 67473,
+      "SCA": 67474,
+      "监护人": 67475,
+      "Cash": 67476,
+      "OVER": 67477,
+      "▁英格兰": 67478,
+      "areth": 67479,
+      "▁improv": 67480,
+      "▁finalists": 67481,
+      "▁upholstery": 67482,
+      "刘邦": 67483,
+      "nSU": 67484,
+      "▁或者": 67485,
+      "率达到": 67486,
+      "nADD": 67487,
+      "▁plasmin": 67488,
+      "▁probing": 67489,
+      "▁SECURITY": 67490,
+      "▁fittings": 67491,
+      "▁peninsula": 67492,
+      "Af": 67493,
+      "不就": 67494,
+      "诡异": 67495,
+      "高中文化": 67496,
+      "clave": 67497,
+      "ngroup": 67498,
+      "▁rebel": 67499,
+      "▁pathologies": 67500,
+      "对被执行人的财产进行了调查": 67501,
+      "密闭": 67502,
+      "敬畏": 67503,
+      "红糖": 67504,
+      "成立后": 67505,
+      "本决定": 67506,
+      "Birth": 67507,
+      "tconfig": 67508,
+      "adaptive": 67509,
+      "▁duplicated": 67510,
+      "▁烤": 67511,
+      "山海": 67512,
+      "王志": 67513,
+      "造船": 67514,
+      "ных": 67515,
+      "上进行": 67516,
+      "venth": 67517,
+      "电源适配器": 67518,
+      "ernary": 67519,
+      "▁Schwartz": 67520,
+      "▁immersed": 67521,
+      "▁organise": 67522,
+      "第二百五十七条第六": 67523,
+      "亚种": 67524,
+      "过节": 67525,
+      "阻滞": 67526,
+      "开展的": 67527,
+      ")\\'\\": 67528,
+      "▁];\\": 67529,
+      "nSide": 67530,
+      "▁Panc": 67531,
+      "▁slick": 67532,
+      "ronectin": 67533,
+      "▁Starbucks": 67534,
+      "▁℃": 67535,
+      "SiO": 67536,
+      "izh": 67537,
+      "丰富了": 67538,
+      "出版物": 67539,
+      "ovis": 67540,
+      "▁LTD": 67541,
+      "情况属实": 67542,
+      "执行一案": 67543,
+      "中医药大学": 67544,
+      "乙醇含量为": 67545,
+      "▁leasing": 67546,
+      "气和": 67547,
+      "黄埔": 67548,
+      "牛皮癣": 67549,
+      "urllib": 67550,
+      "whatwg": 67551,
+      "▁einer": 67552,
+      "小说网的一部": 67553,
+      "▁rebels": 67554,
+      "Listeners": 67555,
+      "号之": 67556,
+      "âte": 67557,
+      "▁返回": 67558,
+      "认识了": 67559,
+      "nLee": 67560,
+      "▁RMS": 67561,
+      "有不同的": 67562,
+      "nJean": 67563,
+      "webui": 67564,
+      "让我们一起": 67565,
+      "ocusing": 67566,
+      "paramet": 67567,
+      "nPolitics": 67568,
+      "▁Collective": 67569,
+      "▁wonderfully": 67570,
+      "▁向": 67571,
+      "君主": 67572,
+      "如水": 67573,
+      "▁暖气": 67574,
+      "指导和": 67575,
+      "▁sue": 67576,
+      "infect": 67577,
+      "▁Turbo": 67578,
+      "Translate": 67579,
+      "▁Callback": 67580,
+      "▁Frontier": 67581,
+      "▁archival": 67582,
+      "&:": 67583,
+      "电镀": 67584,
+      "\">&": 67585,
+      "签署了": 67586,
+      "rums": 67587,
+      "美国政府": 67588,
+      "eming": 67589,
+      "iphery": 67590,
+      "▁GENER": 67591,
+      "▁opaque": 67592,
+      "▁Ideally": 67593,
+      "▁softball": 67594,
+      "gw": 67595,
+      "想把": 67596,
+      "胜地": 67597,
+      "在你的": 67598,
+      "宜宾市": 67599,
+      "▁THERE": 67600,
+      "第一百零三条": 67601,
+      "bitcoin": 67602,
+      "▁Folder": 67603,
+      "▁hashCode": 67604,
+      "但仍": 67605,
+      "刘晓": 67606,
+      "大洋": 67607,
+      "大类": 67608,
+      "水浒": 67609,
+      "点滴": 67610,
+      "idor": 67611,
+      "▁spur": 67612,
+      "年出版的图书": 67613,
+      "可以在判决书送达之日起十五日内": 67614,
+      "owe": 67615,
+      "收录了": 67616,
+      "的焦点": 67617,
+      "HPLC": 67618,
+      "历史悠久": 67619,
+      "学习能力": 67620,
+      "现在很多": 67621,
+      "camel": 67622,
+      "▁PASS": 67623,
+      "Animal": 67624,
+      "如未按本判决": 67625,
+      "▁barbecue": 67626,
+      "▁convection": 67627,
+      "按月": 67628,
+      "游人": 67629,
+      "跳绳": 67630,
+      "gic": 67631,
+      "rei": 67632,
+      "要求和": 67633,
+      "▁Leh": 67634,
+      "▁Meh": 67635,
+      "nComb": 67636,
+      "▁Gone": 67637,
+      "▁scorer": 67638,
+      "nInsects": 67639,
+      "▁pledged": 67640,
+      "▁Including": 67641,
+      "小波": 67642,
+      "层级": 67643,
+      "抢夺": 67644,
+      "漳州": 67645,
+      "在当地": 67646,
+      "说到底": 67647,
+      "主体责任": 67648,
+      "深入研究": 67649,
+      "思想政治教育": 67650,
+      "▁urgency": 67651,
+      "县人民检察院指控": 67652,
+      "年龄较大的儿童或者": 67653,
+      "▁abstracts": 67654,
+      "fem": 67655,
+      "双眼皮": 67656,
+      "Spin": 67657,
+      "▁ESR": 67658,
+      "IONAL": 67659,
+      "▁clog": 67660,
+      "ickers": 67661,
+      "nation": 67662,
+      "▁wickets": 67663,
+      "▁Williamson": 67664,
+      "流年": 67665,
+      "陪我": 67666,
+      "魔力": 67667,
+      "inders": 67668,
+      "latable": 67669,
+      "nThroughout": 67670,
+      "报仇": 67671,
+      "ynn": 67672,
+      "不能说": 67673,
+      "其实我": 67674,
+      "担保法": 67675,
+      "istema": 67676,
+      "widehat": 67677,
+      "▁Pavilion": 67678,
+      "▁Multivariate": 67679,
+      "无异": 67680,
+      "注塑": 67681,
+      "背负": 67682,
+      "нов": 67683,
+      "副区长": 67684,
+      "▁HELP": 67685,
+      "nCorrect": 67686,
+      "直觉": 67687,
+      "稀有": 67688,
+      "解开": 67689,
+      "asu": 67690,
+      "shit": 67691,
+      "▁Goa": 67692,
+      "▁dal": 67693,
+      "小朋友们": 67694,
+      "我们的生活": 67695,
+      "▁Shirley": 67696,
+      "countries": 67697,
+      "▁aspiring": 67698,
+      "▁기": 67699,
+      "按揭": 67700,
+      "骆驼": 67701,
+      "十八大": 67702,
+      "的唯一": 67703,
+      "离退休": 67704,
+      "能求出": 67705,
+      "tcmd": 67706,
+      "▁Teresa": 67707,
+      "▁divert": 67708,
+      "▁wander": 67709,
+      "recorded": 67710,
+      "▁utterly": 67711,
+      "▁possessing": 67712,
+      "施加": 67713,
+      "篆刻": 67714,
+      "解法": 67715,
+      "分数的": 67716,
+      "MANA": 67717,
+      "lé": 67718,
+      "土豪": 67719,
+      "看点": 67720,
+      "都对": 67721,
+      "魏晋": 67722,
+      "Apr": 67723,
+      "▁特性": 67724,
+      "nsee": 67725,
+      "Fatalf": 67726,
+      "planned": 67727,
+      "yroidism": 67728,
+      "nConstruction": 67729,
+      "▁relativistic": 67730,
+      "前十": 67731,
+      "求婚": 67732,
+      "深耕": 67733,
+      "crt": 67734,
+      "来找出": 67735,
+      "的网站": 67736,
+      "igor": 67737,
+      "细致分析": 67738,
+      "sponsored": 67739,
+      "▁decisive": 67740,
+      "▁prostatic": 67741,
+      "{/": 67742,
+      "天际": 67743,
+      "就当": 67744,
+      "片长": 67745,
+      "完全不": 67746,
+      "主要还是": 67747,
+      "bedroom": 67748,
+      "▁estates": 67749,
+      "▁markings": 67750,
+      "▁converters": 67751,
+      "nDevelopment": 67752,
+      "▁shareholder": 67753,
+      "MMMMMMMMMMMMMMMM": 67754,
+      "се": 67755,
+      "山寨": 67756,
+      "狙击": 67757,
+      "解出": 67758,
+      "赛后": 67759,
+      "ONSE": 67760,
+      "▁Tus": 67761,
+      "▁出版日期": 67762,
+      "illian": 67763,
+      "nAlways": 67764,
+      "▁Tunnel": 67765,
+      "▁diagon": 67766,
+      "▁phospholipid": 67767,
+      "一的": 67768,
+      "北仑": 67769,
+      "柯南": 67770,
+      "第三步": 67771,
+      "车辆的": 67772,
+      "另行支付": 67773,
+      "潇湘书院": 67774,
+      "▁Gore": 67775,
+      "tailed": 67776,
+      "▁semic": 67777,
+      "用房": 67778,
+      "开放的": 67779,
+      "Bern": 67780,
+      "FACT": 67781,
+      "两个人的": 67782,
+      "mails": 67783,
+      "nArea": 67784,
+      "▁distorted": 67785,
+      "天山": 67786,
+      "漫游": 67787,
+      "牧场": 67788,
+      "\\\\@": 67789,
+      "这里面": 67790,
+      "不仅如此": 67791,
+      "赔偿责任": 67792,
+      "▁Bret": 67793,
+      "欧洲三大杯": 67794,
+      "▁Ludwig": 67795,
+      "▁Mobility": 67796,
+      "选项中挑选出正确答案": 67797,
+      "confirmation": 67798,
+      "生效法律文书所确定的义务": 67799,
+      "ValidationError": 67800,
+      "盛开": 67801,
+      "zap": 67802,
+      "英文名称": 67803,
+      "▁Bard": 67804,
+      "▁jars": 67805,
+      "ichigan": 67806,
+      "▁electr": 67807,
+      "归根": 67808,
+      "育苗": 67809,
+      "▁NOR": 67810,
+      "回答一个": 67811,
+      "▁trypt": 67812,
+      "HashSet": 67813,
+      "ên": 67814,
+      "乱的": 67815,
+      "作答": 67816,
+      "城墙": 67817,
+      "藏品": 67818,
+      "Lean": 67819,
+      "▁NGC": 67820,
+      "Solid": 67821,
+      "alarm": 67822,
+      "profits": 67823,
+      "visitor": 67824,
+      "▁oceans": 67825,
+      "variance": 67826,
+      "Different": 67827,
+      "▁imposing": 67828,
+      "xr": 67829,
+      "一战": 67830,
+      "大跌": 67831,
+      "abr": 67832,
+      "工作岗位": 67833,
+      "比较容易": 67834,
+      "customers": 67835,
+      "一目": 67836,
+      "变小": 67837,
+      "撕裂": 67838,
+      "淡化": 67839,
+      "粘度": 67840,
+      "用手机": 67841,
+      "MEDIA": 67842,
+      "nField": 67843,
+      "▁cloudy": 67844,
+      "satellite": 67845,
+      "▁NSObject": 67846,
+      "▁Regulations": 67847,
+      "▁earthquakes": 67848,
+      "内裤": 67849,
+      "大衣": 67850,
+      "实习生": 67851,
+      "▁Org": 67852,
+      "▁quo": 67853,
+      "第六十四条之规定": 67854,
+      "▁Portsmouth": 67855,
+      "▁negligence": 67856,
+      "兴旺": 67857,
+      "静音": 67858,
+      "都不一样": 67859,
+      "STAND": 67860,
+      "▁typh": 67861,
+      "William": 67862,
+      "▁policym": 67863,
+      "▁outperforms": 67864,
+      "▁referencing": 67865,
+      "中将": 67866,
+      "nPo": 67867,
+      "我只能": 67868,
+      "清楚了": 67869,
+      "▁Machinery": 67870,
+      "▁forwarding": 67871,
+      "▁Plane": 67872,
+      "▁waking": 67873,
+      "最高人民检察院": 67874,
+      "如皋": 67875,
+      "宣传片": 67876,
+      "要求您": 67877,
+      "宏观经济": 67878,
+      "▁Winc": 67879,
+      "alanine": 67880,
+      "interopRequire": 67881,
+      "普法": 67882,
+      "眩晕": 67883,
+      "成员国": 67884,
+      "Prem": 67885,
+      "WARD": 67886,
+      "nyder": 67887,
+      "▁Oaks": 67888,
+      "合法的借贷关系": 67889,
+      "▁Received": 67890,
+      "▁别": 67891,
+      "▁Moto": 67892,
+      "▁ellipt": 67893,
+      "▁Goldman": 67894,
+      "▁Riemannian": 67895,
+      "Bu": 67896,
+      "不舍": 67897,
+      "系原告": 67898,
+      "▁JEE": 67899,
+      "可以看看": 67900,
+      "lasty": 67901,
+      "▁anthem": 67902,
+      "ilitating": 67903,
+      "▁Byzantine": 67904,
+      "▁thrilling": 67905,
+      "▁fulfillment": 67906,
+      "学府": 67907,
+      "理应": 67908,
+      "翻新": 67909,
+      "Ess": 67910,
+      "`${": 67911,
+      "▁bru": 67912,
+      "▁gin": 67913,
+      "shock": 67914,
+      "assertIs": 67915,
+      "▁Germans": 67916,
+      "▁ancestor": 67917,
+      "▁oversized": 67918,
+      "▁Liberation": 67919,
+      "Sport": 67920,
+      "idered": 67921,
+      "▁Hartford": 67922,
+      "▁deliberate": 67923,
+      "▁retinopathy": 67924,
+      "▁pharmacology": 67925,
+      "IDDLE": 67926,
+      "atches": 67927,
+      "Reporter": 67928,
+      "▁grazing": 67929,
+      "▁cavities": 67930,
+      "▁bilingual": 67931,
+      "塞罗": 67932,
+      "极佳": 67933,
+      "屏幕比例": 67934,
+      "Queen": 67935,
+      "▁fringe": 67936,
+      "Articles": 67937,
+      "▁Celsius": 67938,
+      "可有": 67939,
+      "城北": 67940,
+      "的现实": 67941,
+      "ITCH": 67942,
+      "ixOS": 67943,
+      "avorites": 67944,
+      "▁carpets": 67945,
+      "nAssuming": 67946,
+      "外媒": 67947,
+      "林某": 67948,
+      "饭圈": 67949,
+      "）÷（": 67950,
+      "总体来说": 67951,
+      "umbling": 67952,
+      "▁relativity": 67953,
+      "中药材": 67954,
+      "底面积": 67955,
+      "判处死刑": 67956,
+      "poster": 67957,
+      "▁hurts": 67958,
+      "▁deepest": 67959,
+      "作为被执行人的公民": 67960,
+      "士顿": 67961,
+      "走去": 67962,
+      "AAF": 67963,
+      "Jar": 67964,
+      "goto": 67965,
+      "▁Zion": 67966,
+      "▁petals": 67967,
+      "交房": 67968,
+      "动摇": 67969,
+      "同为": 67970,
+      "造福": 67971,
+      "pom": 67972,
+      "uso": 67973,
+      "相较于": 67974,
+      "部位的": 67975,
+      "nTri": 67976,
+      "▁PDT": 67977,
+      "服务机构": 67978,
+      "tauto": 67979,
+      "▁ctrl": 67980,
+      "▁sincere": 67981,
+      "changelog": 67982,
+      "nCalifornia": 67983,
+      "▁PowerPoint": 67984,
+      "▁evacuation": 67985,
+      "instructions": 67986,
+      "伤残": 67987,
+      "蟑螂": 67988,
+      "Les": 67989,
+      "▁OM": 67990,
+      "的增大而": 67991,
+      "▁peas": 67992,
+      "nReally": 67993,
+      "nDefault": 67994,
+      "QM": 67995,
+      "帐篷": 67996,
+      "西汉": 67997,
+      "scm": 67998,
+      "了我们": 67999,
+      "儿童的": 68000,
+      "多边形": 68001,
+      "宿州市": 68002,
+      "思考题": 68003,
+      "iesta": 68004,
+      "▁Cone": 68005,
+      "▁furthermore": 68006,
+      "▁unidentified": 68007,
+      "异世": 68008,
+      "弗兰": 68009,
+      "袋子": 68010,
+      "躺着": 68011,
+      "afi": 68012,
+      "且仅当": 68013,
+      "某一个": 68014,
+      "的普通": 68015,
+      "▁bridal": 68016,
+      "▁contrasts": 68017,
+      "▁玉": 68018,
+      "一班": 68019,
+      "发帖": 68020,
+      "湿热": 68021,
+      "Zbl": 68022,
+      "sav": 68023,
+      "▁Spart": 68024,
+      "▁tester": 68025,
+      "Mutation": 68026,
+      "▁Carlisle": 68027,
+      "▁serviced": 68028,
+      "▁survivor": 68029,
+      "▁coastline": 68030,
+      "▁Physicians": 68031,
+      "▁ResponseEntity": 68032,
+      "JU": 68033,
+      "他所": 68034,
+      "白衣": 68035,
+      "nMil": 68036,
+      "ulsa": 68037,
+      "▁Conv": 68038,
+      "▁MATH": 68039,
+      "Updates": 68040,
+      "▁pumped": 68041,
+      "▁seldom": 68042,
+      "▁Colleges": 68043,
+      "▁aerospace": 68044,
+      "▁Authorization": 68045,
+      "中国平安财产保险股份有限公司": 68046,
+      "▁生": 68047,
+      "舌头": 68048,
+      "hoc": 68049,
+      "▁LB": 68050,
+      "▁查询": 68051,
+      "ogie": 68052,
+      "▁Siem": 68053,
+      "numberOf": 68054,
+      "nsomething": 68055,
+      "omposition": 68056,
+      "nADVERTISEMENT": 68057,
+      "自由基": 68058,
+      "那么就": 68059,
+      "damage": 68060,
+      "▁Curry": 68061,
+      "nSpring": 68062,
+      "▁Dynamo": 68063,
+      "关羽": 68064,
+      "启迪": 68065,
+      "李玉": 68066,
+      "还给": 68067,
+      "三门峡": 68068,
+      "应该在": 68069,
+      "淮北市": 68070,
+      "::::": 68071,
+      "prote": 68072,
+      "tsize": 68073,
+      "CLUDED": 68074,
+      "▁Trent": 68075,
+      "▁scarc": 68076,
+      "▁enacted": 68077,
+      "ochemically": 68078,
+      "▁complements": 68079,
+      "讨好": 68080,
+      "参加过": 68081,
+      "开发和": 68082,
+      "Piece": 68083,
+      "Frames": 68084,
+      "解本题的关键": 68085,
+      "actively": 68086,
+      "reetings": 68087,
+      "▁causative": 68088,
+      "▁unofficial": 68089,
+      "几款": 68090,
+      "远近": 68091,
+      "▁Xu": 68092,
+      "发展与": 68093,
+      "可有效": 68094,
+      "正因为": 68095,
+      "聚乙烯": 68096,
+      "里程碑": 68097,
+      "需要您": 68098,
+      "▁wie": 68099,
+      "还有一种": 68100,
+      "▁noun": 68101,
+      "中华英烈网": 68102,
+      "cannot": 68103,
+      "▁blink": 68104,
+      "classic": 68105,
+      "solving": 68106,
+      "▁streamline": 68107,
+      "▁rearrangement": 68108,
+      "受访": 68109,
+      "挽救": 68110,
+      "被困": 68111,
+      "送的": 68112,
+      "▁dl": 68113,
+      "五花肉": 68114,
+      "房产证": 68115,
+      "▁FFT": 68116,
+      "服务体系": 68117,
+      "nIran": 68118,
+      "▁Leading": 68119,
+      "▁missionary": 68120,
+      "▁峰": 68121,
+      "▁龚": 68122,
+      "安庆": 68123,
+      "那只": 68124,
+      "主演的": 68125,
+      "以满足": 68126,
+      "附加费用": 68127,
+      "andex": 68128,
+      "▁INCIDENT": 68129,
+      "▁teachings": 68130,
+      "多说": 68131,
+      "酸碱": 68132,
+      ":&#": 68133,
+      "▁Bayes": 68134,
+      "▁Pulse": 68135,
+      "二〇一六年一": 68136,
+      "▁Ability": 68137,
+      "▁comparatively": 68138,
+      "本裁定送达后立即发生法律效力": 68139,
+      "中获得": 68140,
+      "冬奥会": 68141,
+      "▁bor": 68142,
+      "很重要的": 68143,
+      "▁triang": 68144,
+      "▁Logistic": 68145,
+      "凭着": 68146,
+      "献给": 68147,
+      "盖章": 68148,
+      "aned": 68149,
+      "▁NPC": 68150,
+      "突发事件": 68151,
+      "finding": 68152,
+      "Handling": 68153,
+      "▁Brownian": 68154,
+      "专区": 68155,
+      "换取": 68156,
+      "格调": 68157,
+      "讲堂": 68158,
+      "会成为": 68159,
+      "银行业": 68160,
+      "hark": 68161,
+      "jeto": 68162,
+      "nPet": 68163,
+      "mostly": 68164,
+      "▁symposium": 68165,
+      "召回": 68166,
+      "学和": 68167,
+      "ETF": 68168,
+      "nHa": 68169,
+      "也不好": 68170,
+      "执行官": 68171,
+      "民族的": 68172,
+      "estag": 68173,
+      "▁deduced": 68174,
+      "▁uniforms": 68175,
+      "▁inhabited": 68176,
+      "▁migrating": 68177,
+      "提早": 68178,
+      "ARA": 68179,
+      "NPs": 68180,
+      "▁HW": 68181,
+      "▁教育": 68182,
+      "小姑娘": 68183,
+      "的心灵": 68184,
+      "nRay": 68185,
+      "▁QCD": 68186,
+      "四边形是": 68187,
+      "期限为一年": 68188,
+      "Sports": 68189,
+      "scenes": 68190,
+      "▁Capture": 68191,
+      "▁sterling": 68192,
+      "▁Alexandra": 68193,
+      "▁lymphatic": 68194,
+      "活得": 68195,
+      "▁uv": 68196,
+      "多么的": 68197,
+      "nreg": 68198,
+      "▁rests": 68199,
+      "▁Zimmer": 68200,
+      "根基": 68201,
+      "波形": 68202,
+      "IGF": 68203,
+      "▁kJ": 68204,
+      "自己对": 68205,
+      "varying": 68206,
+      "▁livelihood": 68207,
+      "▁arrhythmias": 68208,
+      "▁transcribed": 68209,
+      "▁LoggerFactory": 68210,
+      "指着": 68211,
+      "相减": 68212,
+      "迟到": 68213,
+      "{%\\": 68214,
+      "ência": 68215,
+      "▁brun": 68216,
+      "的一元二次方程": 68217,
+      "Patterns": 68218,
+      "▁Bundesliga": 68219,
+      "▁respectful": 68220,
+      "▁surjective": 68221,
+      "▁anisotropic": 68222,
+      "也应": 68223,
+      "盛行": 68224,
+      "功能障碍": 68225,
+      "▁Mesa": 68226,
+      "两江": 68227,
+      "打过": 68228,
+      "时速": 68229,
+      "聊城": 68230,
+      "读研": 68231,
+      "的斜率": 68232,
+      "Whit": 68233,
+      "▁relieved": 68234,
+      "▁nl": 68235,
+      "天左右": 68236,
+      "大概就是": 68237,
+      "▁Mice": 68238,
+      "基层党组织": 68239,
+      "▁monos": 68240,
+      "avorable": 68241,
+      "大巴": 68242,
+      "拜登": 68243,
+      "新课": 68244,
+      "自已": 68245,
+      "连忙": 68246,
+      "金凤": 68247,
+      "又开始": 68248,
+      "履行的": 68249,
+      "猕猴桃": 68250,
+      "为了提高": 68251,
+      "ircle": 68252,
+      "▁房价信息": 68253,
+      "二〇一七年十": 68254,
+      "untarily": 68255,
+      "▁duality": 68256,
+      "▁murders": 68257,
+      "▁Emerging": 68258,
+      "▁polymorphic": 68259,
+      "因地": 68260,
+      "多做": 68261,
+      "nah": 68262,
+      "自然是": 68263,
+      "起诉状": 68264,
+      "(\\'[": 68265,
+      "▁ecl": 68266,
+      "栽培技术": 68267,
+      "▁puff": 68268,
+      "本院已依法": 68269,
+      "▁mural": 68270,
+      "▁novice": 68271,
+      "IFICATION": 68272,
+      "▁condensed": 68273,
+      "相依": 68274,
+      "这门": 68275,
+      "公司以": 68276,
+      "麦当劳": 68277,
+      "▁Kob": 68278,
+      "文化程度": 68279,
+      "ampsia": 68280,
+      "▁心": 68281,
+      "午后": 68282,
+      "迈进": 68283,
+      "auk": 68284,
+      "cdc": 68285,
+      "战争的": 68286,
+      "▁Deals": 68287,
+      "▁Essay": 68288,
+      "的委托代理人": 68289,
+      "▁McCain": 68290,
+      "▁Xavier": 68291,
+      "▁grease": 68292,
+      "tresponse": 68293,
+      "▁efficacious": 68294,
+      "▁Dependencies": 68295,
+      "特级": 68296,
+      "遍及": 68297,
+      "kHz": 68298,
+      "上来说": 68299,
+      "提供给": 68300,
+      "esan": 68301,
+      "先予执行": 68302,
+      "apons": 68303,
+      "nTell": 68304,
+      "▁Joey": 68305,
+      "toxicity": 68306,
+      "▁Clayton": 68307,
+      "▁enquiry": 68308,
+      "诉讼费由被告承担": 68309,
+      "▁optimism": 68310,
+      "checkpoint": 68311,
+      "▁waterfall": 68312,
+      "把关": 68313,
+      "adc": 68314,
+      "▁\\\\&": 68315,
+      "学习成绩": 68316,
+      "▁Broker": 68317,
+      "▁piston": 68318,
+      "▁stamped": 68319,
+      "▁ligation": 68320,
+      "▁Challenges": 68321,
+      "▁insightful": 68322,
+      "痛点": 68323,
+      "控制和": 68324,
+      "植物油": 68325,
+      "klass": 68326,
+      "ngOnInit": 68327,
+      "▁Theresa": 68328,
+      "▁crowned": 68329,
+      "▁foliage": 68330,
+      "▁trending": 68331,
+      "▁clustered": 68332,
+      "▁maternity": 68333,
+      "▁normative": 68334,
+      "▁summaries": 68335,
+      "▁Developing": 68336,
+      "团圆": 68337,
+      "均以": 68338,
+      "法令": 68339,
+      "交易的": 68340,
+      "大约是": 68341,
+      "uttg": 68342,
+      "pable": 68343,
+      "▁Goods": 68344,
+      "认罪态度较好": 68345,
+      "jectory": 68346,
+      "▁Anders": 68347,
+      "抢险": 68348,
+      "觉得他": 68349,
+      "▁DLL": 68350,
+      "监控系统": 68351,
+      "发展和改革": 68352,
+      "Broker": 68353,
+      "▁keeper": 68354,
+      "高新技术产业开发区": 68355,
+      "▁anthology": 68356,
+      "▁conformal": 68357,
+      "▁thrombocyt": 68358,
+      "小车": 68359,
+      "海湾": 68360,
+      "鳞片": 68361,
+      "▁XV": 68362,
+      "世纪初": 68363,
+      "培养基": 68364,
+      "宝鸡市": 68365,
+      "耐高温": 68366,
+      "\\'];": 68367,
+      "jira": 68368,
+      "▁awe": 68369,
+      "nCapt": 68370,
+      "plist": 68371,
+      "▁Drum": 68372,
+      "CREMENT": 68373,
+      "▁Canton": 68374,
+      "▁lumber": 68375,
+      "astropods": 68376,
+      "▁disparity": 68377,
+      "jw": 68378,
+      "但要": 68379,
+      "更长": 68380,
+      "SUM": 68381,
+      "coc": 68382,
+      "做出最佳": 68383,
+      "服从管教": 68384,
+      "remely": 68385,
+      "▁OpenGL": 68386,
+      "▁pistol": 68387,
+      "operated": 68388,
+      "▁INTEGER": 68389,
+      "▁eclectic": 68390,
+      "▁shortcomings": 68391,
+      "多发": 68392,
+      "Comm": 68393,
+      "odian": 68394,
+      "mapped": 68395,
+      "approval": 68396,
+      "erdinand": 68397,
+      "ント": 68398,
+      "可以买": 68399,
+      "注册表": 68400,
+      "COVID": 68401,
+      "webdriver": 68402,
+      "▁solidarity": 68403,
+      "いて": 68404,
+      "发财": 68405,
+      "报关": 68406,
+      "星系": 68407,
+      "的区域": 68408,
+      "医务人员": 68409,
+      "滨海新区": 68410,
+      "▁reacts": 68411,
+      "▁shores": 68412,
+      "▁Replies": 68413,
+      "▁showroom": 68414,
+      "▁thematic": 68415,
+      "▁Patterson": 68416,
+      "也纳": 68417,
+      "抗压": 68418,
+      "红木": 68419,
+      "dub": 68420,
+      "▁]</": 68421,
+      "▁sequentially": 68422,
+      "惯例": 68423,
+      "的二": 68424,
+      "飞来": 68425,
+      "▁VB": 68426,
+      "并要求": 68427,
+      "gard": 68428,
+      "oset": 68429,
+      "plat": 68430,
+      "然后就是": 68431,
+      "herin": 68432,
+      "oters": 68433,
+      "执行标的为": 68434,
+      "▁Doctors": 68435,
+      "contributor": 68436,
+      "承租": 68437,
+      "IRT": 68438,
+      "从业者": 68439,
+      "周长为": 68440,
+      "我们所": 68441,
+      "由于其": 68442,
+      "msdn": 68443,
+      "为了保证": 68444,
+      "ANNEL": 68445,
+      "nCert": 68446,
+      "▁Molly": 68447,
+      "storybook": 68448,
+      "ophosphate": 68449,
+      "°,": 68450,
+      "惊叹": 68451,
+      "mud": 68452,
+      "▁noct": 68453,
+      "FILTER": 68454,
+      "▁relent": 68455,
+      "未发现被执行人有": 68456,
+      "predicate": 68457,
+      "▁Uploaded": 68458,
+      "ار": 68459,
+      "城西": 68460,
+      "礼盒": 68461,
+      "费等": 68462,
+      "become": 68463,
+      "peonato": 68464,
+      "▁Validator": 68465,
+      "乡土": 68466,
+      "法条": 68467,
+      "\\'#": 68468,
+      "lec": 68469,
+      "也随之": 68470,
+      "而现在": 68471,
+      "▁Mash": 68472,
+      "▁保修信息": 68473,
+      "reserved": 68474,
+      "▁Funding": 68475,
+      "Benchmark": 68476,
+      "▁Intervention": 68477,
+      "失衡": 68478,
+      "学派": 68479,
+      "病死": 68480,
+      "借款后": 68481,
+      "株洲市": 68482,
+      "▁Cors": 68483,
+      "Tenant": 68484,
+      "▁Combining": 68485,
+      "▁gg": 68486,
+      "▁RHS": 68487,
+      "istrar": 68488,
+      "▁Blade": 68489,
+      "▁pains": 68490,
+      "degrees": 68491,
+      "staging": 68492,
+      "imentary": 68493,
+      "▁greeted": 68494,
+      "▁shedding": 68495,
+      "mh": 68496,
+      "党务": 68497,
+      "月刊": 68498,
+      "梨花": 68499,
+      "tkey": 68500,
+      "▁vap": 68501,
+      "现场勘验": 68502,
+      "管理条例": 68503,
+      "Alive": 68504,
+      "izzard": 68505,
+      "二〇一八年六月": 68506,
+      "atalytic": 68507,
+      "▁tapping": 68508,
+      "▁depleted": 68509,
+      "▁glycogen": 68510,
+      "▁由": 68511,
+      "张云": 68512,
+      "谁都": 68513,
+      "如果他": 68514,
+      "onica": 68515,
+      "▁Dunn": 68516,
+      "▁被子植物门": 68517,
+      "nTourist": 68518,
+      "passport": 68519,
+      "▁downtime": 68520,
+      "之势": 68521,
+      "盘锦": 68522,
+      "记入": 68523,
+      "ACTIVE": 68524,
+      "▁Brands": 68525,
+      "▁masked": 68526,
+      "▁reconsider": 68527,
+      "想让": 68528,
+      "日和": 68529,
+      "液中": 68530,
+      "STOP": 68531,
+      "wYWNr": 68532,
+      "▁Kamp": 68533,
+      "▁Kara": 68534,
+      "nGrade": 68535,
+      "▁Oxygen": 68536,
+      "▁Raleigh": 68537,
+      "▁pillows": 68538,
+      "▁synthet": 68539,
+      "禁区": 68540,
+      "金庸": 68541,
+      "ETH": 68542,
+      "anders": 68543,
+      "epochs": 68544,
+      "ourage": 68545,
+      "▁Scope": 68546,
+      "▁canoe": 68547,
+      "oconstr": 68548,
+      "▁Candidate": 68549,
+      "toUpperCase": 68550,
+      "作的": 68551,
+      "')[": 68552,
+      "EFT": 68553,
+      "RIA": 68554,
+      "pis": 68555,
+      "zsh": 68556,
+      "提高到": 68557,
+      "ipad": 68558,
+      "▁Seminary": 68559,
+      "▁capitalize": 68560,
+      "▁disappears": 68561,
+      "同城": 68562,
+      "率低": 68563,
+      "wow": 68564,
+      "的不良": 68565,
+      "splash": 68566,
+      "▁Paula": 68567,
+      "opolysaccharide": 68568,
+      "▁ú": 68569,
+      "为国": 68570,
+      "楼的": 68571,
+      "▁Io": 68572,
+      "以内的": 68573,
+      "缴纳罚金": 68574,
+      "eville": 68575,
+      "▁embell": 68576,
+      "▁gingival": 68577,
+      "乌龟": 68578,
+      "安顺": 68579,
+      "胃口": 68580,
+      "▁'*": 68581,
+      "▁calves": 68582,
+      "tTuesday": 68583,
+      "▁infinit": 68584,
+      "戴上": 68585,
+      "手掌": 68586,
+      "▁电视": 68587,
+      "来判断": 68588,
+      "ngulp": 68589,
+      "▁italic": 68590,
+      "targeted": 68591,
+      "▁nightly": 68592,
+      "咀嚼": 68593,
+      "溢出": 68594,
+      "ifter": 68595,
+      "▁Adri": 68596,
+      "▁Mamm": 68597,
+      "▁Scha": 68598,
+      "我们都知道": 68599,
+      "scaled": 68600,
+      "▁dotted": 68601,
+      "accepted": 68602,
+      "▁dismant": 68603,
+      "几万": 68604,
+      "风尚": 68605,
+      "asci": 68606,
+      "▁Surf": 68607,
+      "ILABLE": 68608,
+      "nclean": 68609,
+      "解除财产保全": 68610,
+      "▁mitotic": 68611,
+      "ouncements": 68612,
+      "▁inscribed": 68613,
+      "▁declarations": 68614,
+      "为新": 68615,
+      "精良": 68616,
+      "我们从": 68617,
+      "是他们": 68618,
+      "▁ICD": 68619,
+      "业务范围": 68620,
+      "gages": 68621,
+      "THREAD": 68622,
+      "▁Speaking": 68623,
+      "▁automorphism": 68624,
+      "▁accomplishment": 68625,
+      "▁Н": 68626,
+      "因人": 68627,
+      "张学": 68628,
+      "中找到": 68629,
+      "研究工作": 68630,
+      "通货膨胀": 68631,
+      "▁ripe": 68632,
+      "▁Adapter": 68633,
+      "▁positives": 68634,
+      "▁realities": 68635,
+      "▁originates": 68636,
+      "▁鹏": 68637,
+      "保单": 68638,
+      "盘的": 68639,
+      "石器": 68640,
+      "▁其八": 68641,
+      "▁第四章": 68642,
+      "营养丰富": 68643,
+      "ffiti": 68644,
+      "▁PACK": 68645,
+      "▁famed": 68646,
+      "▁flare": 68647,
+      "igators": 68648,
+      "snippet": 68649,
+      "▁hacked": 68650,
+      "达成执行和解协议": 68651,
+      "###############\\": 68652,
+      "全套": 68653,
+      "南站": 68654,
+      "红烧": 68655,
+      "配以": 68656,
+      "ные": 68657,
+      "Dash": 68658,
+      "▁raft": 68659,
+      "此题的关键": 68660,
+      "▁basil": 68661,
+      "▁Sutton": 68662,
+      "▁localize": 68663,
+      "李晓": 68664,
+      "▁SUN": 68665,
+      "机电设备": 68666,
+      "wYWNrYWdl": 68667,
+      "▁cleansing": 68668,
+      "▁precautions": 68669,
+      "▁specialization": 68670,
+      "戴尔": 68671,
+      "NPC": 68672,
+      "nco": 68673,
+      "但是却": 68674,
+      "的图片": 68675,
+      "有良好的": 68676,
+      "这种现象": 68677,
+      "Third": 68678,
+      "doctype": 68679,
+      "留着": 68680,
+      "穿了": 68681,
+      "道人": 68682,
+      "疯狂的": 68683,
+      "▁bere": 68684,
+      "法人或者其他": 68685,
+      "▁exerts": 68686,
+      "▁Citizen": 68687,
+      "▁aliases": 68688,
+      "▁booklet": 68689,
+      "▁plugged": 68690,
+      "请从给出的选项中": 68691,
+      "nFeatures": 68692,
+      "宁夏回族自治区银川市": 68693,
+      "选出你认为正确的答案": 68694,
+      "▁glucocorticoid": 68695,
+      "救赎": 68696,
+      "电荷": 68697,
+      "之目的": 68698,
+      "Hold": 68699,
+      "nGot": 68700,
+      "▁Sle": 68701,
+      "涉及广泛": 68702,
+      "▁Loch": 68703,
+      "▁laughed": 68704,
+      "áz": 68705,
+      "军医": 68706,
+      "送礼": 68707,
+      "ENE": 68708,
+      "▁NEC": 68709,
+      "人民调解": 68710,
+      "IENTS": 68711,
+      "fires": 68712,
+      "Solver": 68713,
+      "Waiting": 68714,
+      "▁bodily": 68715,
+      "▁greedy": 68716,
+      "MainWindow": 68717,
+      "▁cosmetics": 68718,
+      "жи": 68719,
+      "即有": 68720,
+      "家门": 68721,
+      "泸州": 68722,
+      "（﹣": 68723,
+      "肝硬化": 68724,
+      "pear": 68725,
+      "nMethods": 68726,
+      "▁Representation": 68727,
+      "优选": 68728,
+      "心律": 68729,
+      "部委": 68730,
+      "包头市": 68731,
+      "bass": 68732,
+      "▁Hag": 68733,
+      "临床疗效": 68734,
+      "休克": 68735,
+      "张扬": 68736,
+      "高涨": 68737,
+      "▁机械": 68738,
+      "名下有": 68739,
+      "无固定": 68740,
+      "roma": 68741,
+      "一个挑战": 68742,
+      "ViewSet": 68743,
+      "▁Adoles": 68744,
+      "▁thymus": 68745,
+      "▁gestures": 68746,
+      "▁monocyte": 68747,
+      "▁设": 68748,
+      "汉中": 68749,
+      "爆出": 68750,
+      "返利": 68751,
+      "'',": 68752,
+      "和教育": 68753,
+      "CSDN": 68754,
+      "klore": 68755,
+      "▁Cory": 68756,
+      "▁Sunny": 68757,
+      "▁Profit": 68758,
+      "▁concord": 68759,
+      "resourcemanager": 68760,
+      "众数": 68761,
+      "依约": 68762,
+      "标示": 68763,
+      "海龙": 68764,
+      "莉莉": 68765,
+      "被抓": 68766,
+      "Meet": 68767,
+      "erase": 68768,
+      "▁moss": 68769,
+      "▁分布区域": 68770,
+      "▁poised": 68771,
+      "ocations": 68772,
+      "nPatients": 68773,
+      "Du": 68774,
+      "噩梦": 68775,
+      "学年": 68776,
+      "心愿": 68777,
+      "打完": 68778,
+      "拥护": 68779,
+      "iab": 68780,
+      "正相关": 68781,
+      "uevo": 68782,
+      "chitz": 68783,
+      "质量管理体系": 68784,
+      "▁dependable": 68785,
+      "隔断": 68786,
+      "RSS": 68787,
+      "一万元": 68788,
+      "年龄段": 68789,
+      "构建了": 68790,
+      "aise": 68791,
+      "企业经营": 68792,
+      "有色金属": 68793,
+      "额外费用": 68794,
+      "▁Merg": 68795,
+      "▁Ernst": 68796,
+      "locking": 68797,
+      "nElection": 68798,
+      "兴安": 68799,
+      "无从": 68800,
+      "柳州": 68801,
+      "最起码": 68802,
+      "oras": 68803,
+      "▁Hear": 68804,
+      "▁csrf": 68805,
+      "nGlobal": 68806,
+      "▁unreal": 68807,
+      "napplication": 68808,
+      "▁manipulations": 68809,
+      "QB": 68810,
+      "▁KE": 68811,
+      "▁分钟": 68812,
+      "▁Tek": 68813,
+      "活动现场": 68814,
+      "▁piez": 68815,
+      "一道复杂的": 68816,
+      "▁didnt": 68817,
+      "在有限时间内": 68818,
+      "▁endure": 68819,
+      "nImagine": 68820,
+      "▁Combine": 68821,
+      ",_": 68822,
+      "与大": 68823,
+      "及对": 68824,
+      "咸阳": 68825,
+      "太久": 68826,
+      "Led": 68827,
+      "▁Cubs": 68828,
+      "nMonday": 68829,
+      "▁bioactive": 68830,
+      "▁nanoparticle": 68831,
+      "新编": 68832,
+      "菜的": 68833,
+      "跑路": 68834,
+      "集市": 68835,
+      "ukes": 68836,
+      "我将给您": 68837,
+      "还有其他": 68838,
+      "▁chemo": 68839,
+      "▁Dawson": 68840,
+      "▁pottery": 68841,
+      "▁dragging": 68842,
+      "应交": 68843,
+      "网购": 68844,
+      "主体的": 68845,
+      "之外的": 68846,
+      "TACT": 68847,
+      "crunch": 68848,
+      "请从所提供的": 68849,
+      "选项中做出最佳": 68850,
+      "一道复杂的选择题": 68851,
+      "方面的知识来找出": 68852,
+      "要求您综合运用多": 68853,
+      "▁Sporting": 68854,
+      "我将给您一道复杂的选择题": 68855,
+      "方面的知识来找出正确的答案": 68856,
+      "请从所提供的选项中做出最佳": 68857,
+      "花粉": 68858,
+      "衣食": 68859,
+      "CMC": 68860,
+      "▁*(": 68861,
+      "▁Ui": 68862,
+      "▁RPC": 68863,
+      "机器学习": 68864,
+      "深入学习": 68865,
+      "▁Ches": 68866,
+      "▁LINE": 68867,
+      "▁eats": 68868,
+      "▁protr": 68869,
+      "▁leaking": 68870,
+      "▁mandibular": 68871,
+      "▁viz": 68872,
+      "▁yum": 68873,
+      "▁Hint": 68874,
+      "isational": 68875,
+      "startTime": 68876,
+      "▁endometri": 68877,
+      "▁constellation": 68878,
+      "▁鲁": 68879,
+      "ipa": 68880,
+      "owler": 68881,
+      "Influence": 68882,
+      "▁Ventures": 68883,
+      "▁umbilical": 68884,
+      "▁youngsters": 68885,
+      "○○": 68886,
+      "初审": 68887,
+      "即墨": 68888,
+      "果肉": 68889,
+      "不下去": 68890,
+      "不小的": 68891,
+      "清楚的": 68892,
+      "osarcoma": 68893,
+      "▁belongings": 68894,
+      "▁defaultValue": 68895,
+      "▁hypothalamus": 68896,
+      "卵管": 68897,
+      "浮现": 68898,
+      "Syn": 68899,
+      "▁Drivers": 68900,
+      "本院经审理认定事实如下": 68901,
+      "第十七": 68902,
+      "ción": 68903,
+      "▁Eleven": 68904,
+      "▁Gentle": 68905,
+      "▁pastry": 68906,
+      "▁hashtag": 68907,
+      "▁installs": 68908,
+      "▁analysing": 68909,
+      "徐汇": 68910,
+      "得天": 68911,
+      "疙瘩": 68912,
+      "ISM": 68913,
+      "登记在": 68914,
+      "nSty": 68915,
+      "Audit": 68916,
+      "罗某": 68917,
+      "▁LF": 68918,
+      "[],\\": 68919,
+      "intr": 68920,
+      "nDie": 68921,
+      "绝对不会": 68922,
+      "RESET": 68923,
+      "▁Bronx": 68924,
+      "▁feeder": 68925,
+      "喜欢他": 68926,
+      "多大的": 68927,
+      "下面这道题": 68928,
+      "carbons": 68929,
+      "▁Outcome": 68930,
+      "▁TypeScript": 68931,
+      "午夜": 68932,
+      "外币": 68933,
+      "大帝": 68934,
+      "我俩": 68935,
+      "轮流": 68936,
+      "nih": 68937,
+      "忽略了": 68938,
+      "bbie": 68939,
+      "降低成本": 68940,
+      "▁Schro": 68941,
+      "▁Bahrain": 68942,
+      "streaming": 68943,
+      "▁wrestler": 68944,
+      "▁Voltage": 68945,
+      "▁laminate": 68946,
+      "▁enumeration": 68947,
+      "冶炼": 68948,
+      "湿疹": 68949,
+      "就是不": 68950,
+      "技术及": 68951,
+      "的运用": 68952,
+      "Rain": 68953,
+      "请你帮我": 68954,
+      "▁Funds": 68955,
+      "▁Monaco": 68956,
+      "▁knight": 68957,
+      "▁Vatican": 68958,
+      "买过": 68959,
+      "加权": 68960,
+      "紧邻": 68961,
+      "利益的": 68962,
+      "名同学": 68963,
+      "实体店": 68964,
+      "菏泽市": 68965,
+      "▁Kok": 68966,
+      "▁tak": 68967,
+      "obiles": 68968,
+      "太和": 68969,
+      "撰稿": 68970,
+      "是宋代": 68971,
+      "庭审笔录": 68972,
+      "Cells": 68973,
+      "darwin": 68974,
+      "▁resolves": 68975,
+      "▁presenters": 68976,
+      "▁费": 68977,
+      "人所": 68978,
+      "凝血": 68979,
+      "加装": 68980,
+      "培根": 68981,
+      "海运": 68982,
+      "特勒": 68983,
+      "▁AUC": 68984,
+      "▁lil": 68985,
+      "综艺节目": 68986,
+      "nMatt": 68987,
+      "▁GOODS": 68988,
+      "▁incap": 68989,
+      "▁tapes": 68990,
+      "▁inward": 68991,
+      "▁Simmons": 68992,
+      "▁noodles": 68993,
+      "有名": 68994,
+      "盛典": 68995,
+      "美发": 68996,
+      "原告人": 68997,
+      "和解析": 68998,
+      "王世贞": 68999,
+      "考察了": 69000,
+      "[[\\'": 69001,
+      "▁AIR": 69002,
+      "平均亩产": 69003,
+      "案例分析": 69004,
+      "nDest": 69005,
+      "▁作品体裁": 69006,
+      "routine": 69007,
+      "▁Threat": 69008,
+      "▁airways": 69009,
+      "▁anaesthesia": 69010,
+      "▁epidemiologic": 69011,
+      "万岁": 69012,
+      "听觉": 69013,
+      "讲了": 69014,
+      "顺义": 69015,
+      "Sha": 69016,
+      "▁$$[": 69017,
+      "▁Rak": 69018,
+      "▁mah": 69019,
+      "选项正确": 69020,
+      "nAttached": 69021,
+      "▁garnered": 69022,
+      "上任": 69023,
+      "多处": 69024,
+      "伦贝尔": 69025,
+      "engo": 69026,
+      "minent": 69027,
+      "也给": 69028,
+      "听话": 69029,
+      "现年": 69030,
+      "用处": 69031,
+      "礼服": 69032,
+      "跟风": 69033,
+      "的开始": 69034,
+      "▁Vor": 69035,
+      "▁Morm": 69036,
+      "▁acyl": 69037,
+      "▁Influ": 69038,
+      "查封被执行人": 69039,
+      "▁hunters": 69040,
+      "▁Flooring": 69041,
+      "▁pavement": 69042,
+      "▁tearDown": 69043,
+      "丧尸": 69044,
+      "名思义": 69045,
+      "小学校": 69046,
+      "ocardial": 69047,
+      "▁sadness": 69048,
+      "仅限": 69049,
+      "拜托": 69050,
+      "木斯": 69051,
+      "的人格": 69052,
+      "nInsp": 69053,
+      "▁Bots": 69054,
+      "▁Drain": 69055,
+      "▁Deutsche": 69056,
+      "▁chatting": 69057,
+      "Assessment": 69058,
+      "▁rebuilding": 69059,
+      "古迹": 69060,
+      "泳池": 69061,
+      "香甜": 69062,
+      "bah": 69063,
+      "老太太": 69064,
+      "}}}$\\": 69065,
+      "子叶植物纲": 69066,
+      "▁Camden": 69067,
+      "olitical": 69068,
+      "▁Cellular": 69069,
+      "Activation": 69070,
+      "制图": 69071,
+      "uggy": 69072,
+      "机场等接": 69073,
+      "提取公因式": 69074,
+      "Sunday": 69075,
+      "▁moduli": 69076,
+      "▁suicidal": 69077,
+      "accharomyces": 69078,
+      "▁aggressively": 69079,
+      "▁psychologist": 69080,
+      "宝玉": 69081,
+      "度为": 69082,
+      "硝酸": 69083,
+      "的创新": 69084,
+      "▁Glu": 69085,
+      "进行检查": 69086,
+      "feeds": 69087,
+      "▁brushes": 69088,
+      "assessment": 69089,
+      "▁religions": 69090,
+      "▁increments": 69091,
+      "▁paediatric": 69092,
+      "▁warranties": 69093,
+      "浮生": 69094,
+      "连同": 69095,
+      "不得已": 69096,
+      "往往会": 69097,
+      "PLUS": 69098,
+      "世界第一": 69099,
+      "为代表的": 69100,
+      "▁TRAN": 69101,
+      "▁fecal": 69102,
+      "并给出正确的答案": 69103,
+      "▁adventurous": 69104,
+      "不振": 69105,
+      "成就感": 69106,
+      "网络通信": 69107,
+      "atant": 69108,
+      "ledon": 69109,
+      "▁mats": 69110,
+      "olfram": 69111,
+      "▁boobs": 69112,
+      "请分析下面这道题": 69113,
+      "▁sedation": 69114,
+      "文创": 69115,
+      "yar": 69116,
+      "大队长": 69117,
+      "henol": 69118,
+      "ndown": 69119,
+      "▁Vest": 69120,
+      "▁eyeb": 69121,
+      "一道选择题": 69122,
+      "▁parishes": 69123,
+      "▁absorbing": 69124,
+      "分队": 69125,
+      "流逝": 69126,
+      "穿插": 69127,
+      "行销": 69128,
+      "赛的": 69129,
+      "Boss": 69130,
+      "▁Cros": 69131,
+      "▁toug": 69132,
+      "Recomm": 69133,
+      "▁decou": 69134,
+      "▁Carmen": 69135,
+      "cylinder": 69136,
+      "▁buckets": 69137,
+      "▁recursively": 69138,
+      "停机": 69139,
+      "而得名": 69140,
+      "还不能": 69141,
+      "所用到的": 69142,
+      "(\"\")\\": 69143,
+      "isease": 69144,
+      "▁Lymph": 69145,
+      "▁Nikol": 69146,
+      "▁Beatles": 69147,
+      "▁Falcons": 69148,
+      "▁foreach": 69149,
+      "▁informing": 69150,
+      "▁treadmill": 69151,
+      "▁milestones": 69152,
+      "实干": 69153,
+      "萨拉": 69154,
+      "被除": 69155,
+      "AVI": 69156,
+      "及技术": 69157,
+      "手机版": 69158,
+      "nPDF": 69159,
+      "效力球队": 69160,
+      "的影响力": 69161,
+      "chosen": 69162,
+      "spaper": 69163,
+      "▁Divid": 69164,
+      "▁Norwich": 69165,
+      "▁Boundary": 69166,
+      "▁vestibular": 69167,
+      "优劣": 69168,
+      "目睹": 69169,
+      "相聚": 69170,
+      "keh": 69171,
+      "▁dv": 69172,
+      "土地利用": 69173,
+      "聊天记录": 69174,
+      "paint": 69175,
+      "▁Leigh": 69176,
+      "idopsis": 69177,
+      "▁realizes": 69178,
+      "栗子": 69179,
+      "次年": 69180,
+      "玩玩": 69181,
+      "花果": 69182,
+      "长白": 69183,
+      "阅历": 69184,
+      "首款": 69185,
+      "请立即": 69186,
+      "队员们": 69187,
+      "UGIN": 69188,
+      "国际标准": 69189,
+      "endid": 69190,
+      "iggle": 69191,
+      "代表国家队": 69192,
+      "iuseppe": 69193,
+      "▁fractal": 69194,
+      "公布失信被执行人名单": 69195,
+      "管网": 69196,
+      "多利亚": 69197,
+      "政治局": 69198,
+      "方式为": 69199,
+      "等技术": 69200,
+      "▁hen": 69201,
+      "scratch": 69202,
+      "可酌情从轻处罚": 69203,
+      "▁insomnia": 69204,
+      "▁norepinephrine": 69205,
+      "之气": 69206,
+      "楼宇": 69207,
+      "bak": 69208,
+      "▁河北": 69209,
+      "行驶证": 69210,
+      "▁Gott": 69211,
+      "未到庭应诉": 69212,
+      "estyles": 69213,
+      "idences": 69214,
+      "一斤": 69215,
+      "冰冷": 69216,
+      "吕布": 69217,
+      "尿管": 69218,
+      "教书": 69219,
+      "纳德": 69220,
+      "Rom": 69221,
+      "的眼神": 69222,
+      "MSCs": 69223,
+      "生产总值": 69224,
+      "Broad": 69225,
+      "▁disruptive": 69226,
+      "全职": 69227,
+      "受灾": 69228,
+      "基底": 69229,
+      "建伟": 69230,
+      "水解": 69231,
+      "sts": 69232,
+      "瞩目的": 69233,
+      "过敏性": 69234,
+      "▁Neon": 69235,
+      "方可开展经营活动": 69236,
+      "剧团": 69237,
+      "置疑": 69238,
+      "对该案": 69239,
+      "一个小时": 69240,
+      "auter": 69241,
+      "othes": 69242,
+      "▁Lans": 69243,
+      "▁Vish": 69244,
+      "▁Nagar": 69245,
+      "RUPTION": 69246,
+      "欠缺": 69247,
+      "▁'\\": 69248,
+      "东北部": 69249,
+      "▁Sonic": 69250,
+      "▁Uzbek": 69251,
+      "ainment": 69252,
+      "打成": 69253,
+      "脑残": 69254,
+      "自成": 69255,
+      "▁主料": 69256,
+      "发表的": 69257,
+      "交易平台": 69258,
+      "Offer": 69259,
+      "▁Aside": 69260,
+      "▁Muscle": 69261,
+      "▁hamster": 69262,
+      "▁weekday": 69263,
+      "吸水": 69264,
+      "海的": 69265,
+      "益生": 69266,
+      "顽固": 69267,
+      "鼻腔": 69268,
+      "Asc": 69269,
+      "在上面": 69270,
+      "公益事业": 69271,
+      "\\'\\'\\": 69272,
+      "ittal": 69273,
+      "▁staring": 69274,
+      "▁ت": 69275,
+      "▁Nd": 69276,
+      "ncat": 69277,
+      "nQuant": 69278,
+      "缓期二年执行": 69279,
+      "▁Cookies": 69280,
+      "▁Insights": 69281,
+      "▁wrestlers": 69282,
+      "nExperience": 69283,
+      "不比": 69284,
+      "写上": 69285,
+      "是否为": 69286,
+      "住河北省": 69287,
+      "进行处理": 69288,
+      "▁sway": 69289,
+      "经本院依法": 69290,
+      "▁Abuse": 69291,
+      "itemize": 69292,
+      "银联": 69293,
+      "ORA": 69294,
+      "▁视频": 69295,
+      "某某诉称": 69296,
+      "比较大的": 69297,
+      "监狱表扬": 69298,
+      "▁Nile": 69299,
+      "besgue": 69300,
+      "Currently": 69301,
+      "▁competitiveness": 69302,
+      "夜市": 69303,
+      "批表": 69304,
+      "时事": 69305,
+      "<\\'": 69306,
+      "nano": 69307,
+      "unky": 69308,
+      "▁gov": 69309,
+      "inches": 69310,
+      "▁tagging": 69311,
+      "▁tempted": 69312,
+      "▁outsourcing": 69313,
+      "▁Mycobacterium": 69314,
+      "DRS": 69315,
+      "▁HSP": 69316,
+      "共同债务": 69317,
+      "鼠标操作": 69318,
+      "skills": 69319,
+      "▁hydroxyl": 69320,
+      "二章": 69321,
+      "来之": 69322,
+      "这人": 69323,
+      ">',": 69324,
+      "nnp": 69325,
+      "(\\':": 69326,
+      "▁Mae": 69327,
+      "元为基数": 69328,
+      "▁Café": 69329,
+      "对应的选项": 69330,
+      "▁milling": 69331,
+      "▁universally": 69332,
+      "▁neutralizing": 69333,
+      "兴国": 69334,
+      "再有": 69335,
+      "定罪": 69336,
+      "nEv": 69337,
+      "第八条": 69338,
+      "▁CCD": 69339,
+      "claims": 69340,
+      "▁grating": 69341,
+      "▁synapse": 69342,
+      "▁handheld": 69343,
+      "▁polyethylene": 69344,
+      "偏高": 69345,
+      "转行": 69346,
+      "PHA": 69347,
+      "OTAL": 69348,
+      "▁Dept": 69349,
+      "自然保护区": 69350,
+      "重庆时时彩": 69351,
+      "▁chords": 69352,
+      "▁Catholics": 69353,
+      "之手": 69354,
+      "美式": 69355,
+      "▁CY": 69356,
+      "六安市": 69357,
+      "方式是": 69358,
+      "羽绒服": 69359,
+      "▁Lean": 69360,
+      "可以向人民法院": 69361,
+      "▁semifinals": 69362,
+      "▁certifications": 69363,
+      "二个": 69364,
+      "MOVE": 69365,
+      "印象深刻": 69366,
+      "▁conqu": 69367,
+      "▁supreme": 69368,
+      "▁blindness": 69369,
+      "之首": 69370,
+      "正气": 69371,
+      "迪亚": 69372,
+      "kot": 69373,
+      "判别式": 69374,
+      "大小的": 69375,
+      "Season": 69376,
+      "▁Beaver": 69377,
+      "▁feared": 69378,
+      "片中": 69379,
+      "觉悟": 69380,
+      "顺手": 69381,
+      "))^": 69382,
+      "▁[%": 69383,
+      "一起吃": 69384,
+      "当天的": 69385,
+      "▁STO": 69386,
+      "ichen": 69387,
+      "▁scars": 69388,
+      "combine": 69389,
+      "▁Assess": 69390,
+      "▁Quotes": 69391,
+      "▁lineages": 69392,
+      "▁dataframe": 69393,
+      "▁ribosomal": 69394,
+      "上过": 69395,
+      "嘉善": 69396,
+      "暴躁": 69397,
+      "警报": 69398,
+      "的天空": 69399,
+      "的路上": 69400,
+      "▁dow": 69401,
+      "暂不具备": 69402,
+      "nAction": 69403,
+      "▁motives": 69404,
+      "▁newline": 69405,
+      "▁stranded": 69406,
+      "▁modifiers": 69407,
+      "Deserialize": 69408,
+      "▁prominence": 69409,
+      "▁confinement": 69410,
+      "三世": 69411,
+      "下周": 69412,
+      "害虫": 69413,
+      "建民": 69414,
+      "放着": 69415,
+      "很高兴": 69416,
+      "金字塔": 69417,
+      "Known": 69418,
+      "▁Toll": 69419,
+      "Victor": 69420,
+      "作出正确判断": 69421,
+      "▁shooters": 69422,
+      "▁Principle": 69423,
+      "▁Extensions": 69424,
+      "洛夫": 69425,
+      "与文化": 69426,
+      "总价中": 69427,
+      "paul": 69428,
+      "中心对称": 69429,
+      "负面影响": 69430,
+      "刑罚执行机关": 69431,
+      "▁snakes": 69432,
+      "TypeName": 69433,
+      "没有约定或者约定": 69434,
+      "▁Genetics": 69435,
+      "▁Greenland": 69436,
+      "▁Networking": 69437,
+      "▁uncontrolled": 69438,
+      "张志": 69439,
+      "雾霾": 69440,
+      "第六届": 69441,
+      "HOLD": 69442,
+      "▁FGF": 69443,
+      "所确定的": 69444,
+      "▁buoy": 69445,
+      "▁juicy": 69446,
+      "▁manufactures": 69447,
+      "HJ": 69448,
+      "▁利": 69449,
+      "玩转": 69450,
+      "轮候": 69451,
+      "顶着": 69452,
+      "FTP": 69453,
+      "▁Mp": 69454,
+      "在网络": 69455,
+      "望远镜": 69456,
+      "不是因为": 69457,
+      "▁Founded": 69458,
+      "▁Technique": 69459,
+      "东城": 69460,
+      "充当": 69461,
+      "带回": 69462,
+      "德基": 69463,
+      "捷径": 69464,
+      "省的": 69465,
+      "的身影": 69466,
+      "聚氨酯": 69467,
+      "让用户": 69468,
+      "yout": 69469,
+      "▁DFT": 69470,
+      "您将面临": 69471,
+      "文学作品": 69472,
+      "检验检疫": 69473,
+      "淋浴设施": 69474,
+      "ensing": 69475,
+      "BK": 69476,
+      "三生": 69477,
+      "不乏": 69478,
+      "尽责": 69479,
+      "大多是": 69480,
+      "anye": 69481,
+      "nBre": 69482,
+      "▁IOS": 69483,
+      "详细评估": 69484,
+      "两次以上的": 69485,
+      "escent": 69486,
+      "incare": 69487,
+      "nSeeds": 69488,
+      "▁unbeat": 69489,
+      "▁arginine": 69490,
+      "沦为": 69491,
+      "骑兵": 69492,
+      "在英国": 69493,
+      "葫芦岛": 69494,
+      "amus": 69495,
+      "▁Zar": 69496,
+      "▁nth": 69497,
+      "ignty": 69498,
+      "▁合作关系": 69499,
+      "▁mailbox": 69500,
+      "▁Supplier": 69501,
+      "epithelial": 69502,
+      "哈密": 69503,
+      "憔悴": 69504,
+      "熟的": 69505,
+      "而他": 69506,
+      "到自己": 69507,
+      "未提出": 69508,
+      "▁Cra": 69509,
+      "不断扩大": 69510,
+      "给你一段": 69511,
+      "最佳的答案": 69512,
+      "▁sucking": 69513,
+      "typeparam": 69514,
+      "▁stressors": 69515,
+      "直接用": 69516,
+      "bron": 69517,
+      "回答一道": 69518,
+      "大家都是": 69519,
+      "这个事情": 69520,
+      "这方面的": 69521,
+      "Japan": 69522,
+      "Terms": 69523,
+      "▁bred": 69524,
+      "▁Cancel": 69525,
+      "平面直角坐标系": 69526,
+      "机场等接车服务": 69527,
+      "▁Daniels": 69528,
+      "cplusplus": 69529,
+      "▁departed": 69530,
+      "写过": 69531,
+      "西欧": 69532,
+      "尽全力": 69533,
+      "▁Sey": 69534,
+      "这个过程": 69535,
+      "Brown": 69536,
+      "ncall": 69537,
+      "▁mage": 69538,
+      "电动自行车": 69539,
+      "▁exposes": 69540,
+      "▁Requests": 69541,
+      "▁Telegram": 69542,
+      "将它": 69543,
+      "数中": 69544,
+      "海带": 69545,
+      "羟基": 69546,
+      "陵园": 69547,
+      "evil": 69548,
+      "▁DUI": 69549,
+      "元的事实": 69550,
+      "▁geom": 69551,
+      "▁proj": 69552,
+      "需在入住时": 69553,
+      "计算在总价中": 69554,
+      "AppData": 69555,
+      "▁faucet": 69556,
+      "附加费用不会自动": 69557,
+      "需在入住时另行支付": 69558,
+      "setObjectName": 69559,
+      "附加费用不会自动计算在总价中": 69560,
+      "▁郝": 69561,
+      "不识": 69562,
+      "张开": 69563,
+      "五年级": 69564,
+      "Half": 69565,
+      "▁cra": 69566,
+      "▁Reich": 69567,
+      "▁tiger": 69568,
+      "scalajs": 69569,
+      "▁skillet": 69570,
+      "▁Highlands": 69571,
+      "RecyclerView": 69572,
+      "物业管理有限公司撤回起诉": 69573,
+      "班组": 69574,
+      "▁eb": 69575,
+      "三千元": 69576,
+      "这条路": 69577,
+      "直线距离": 69578,
+      "▁CUDA": 69579,
+      "RETURN": 69580,
+      "regnant": 69581,
+      "▁geneal": 69582,
+      "▁masturb": 69583,
+      "nSaturday": 69584,
+      "reduction": 69585,
+      "▁阳": 69586,
+      "内镜": 69587,
+      "断电": 69588,
+      "而你": 69589,
+      "肉牛": 69590,
+      "风月": 69591,
+      "华尔街": 69592,
+      "▁MPa": 69593,
+      "如果真的": 69594,
+      "主要是因为": 69595,
+      "然后从提供的": 69596,
+      "请您细致分析": 69597,
+      "ointment": 69598,
+      "▁ADVISED": 69599,
+      "您将面临一个挑战": 69600,
+      "回答一道复杂选择题": 69601,
+      "▁photovoltaic": 69602,
+      "在有限时间内回答一道复杂选择题": 69603,
+      "人会": 69604,
+      "夜深": 69605,
+      "张大": 69606,
+      "某某在": 69607,
+      "aure": 69608,
+      "acteria": 69609,
+      "该国": 69610,
+      "赛中": 69611,
+      "敌人的": 69612,
+      "发展前景": 69613,
+      "歌曲歌词": 69614,
+      "找出您认为": 69615,
+      "请您全神贯注": 69616,
+      "地审查以下这道": 69617,
+      "Students": 69618,
+      "▁blocker": 69619,
+      "找出您认为最佳的答案": 69620,
+      "涉及广泛知识的选择题": 69621,
+      "▁correcting": 69622,
+      "▁redirected": 69623,
+      "请您全神贯注地审查以下这道": 69624,
+      "▁微": 69625,
+      "宝塔": 69626,
+      "▁;)": 69627,
+      "专门为": 69628,
+      "在深圳": 69629,
+      "必须有": 69630,
+      "是现代": 69631,
+      "crate": 69632,
+      "nSend": 69633,
+      "▁Flint": 69634,
+      "▁StObject": 69635,
+      "公尺": 69636,
+      "废旧": 69637,
+      "活了": 69638,
+      "%}\\": 69639,
+      "eat": 69640,
+      "上升到": 69641,
+      "立方根": 69642,
+      "请思考": 69643,
+      "▁Gat": 69644,
+      "anford": 69645,
+      "uously": 69646,
+      "▁nemat": 69647,
+      "atation": 69648,
+      "童装": 69649,
+      "面面": 69650,
+      "▁BW": 69651,
+      "▁Fault": 69652,
+      "▁bells": 69653,
+      "▁shampoo": 69654,
+      "▁inventor": 69655,
+      "▁exporting": 69656,
+      "CCA": 69657,
+      "GFP": 69658,
+      "▁DX": 69659,
+      "可能与": 69660,
+      "eles": 69661,
+      "hbox": 69662,
+      "▁Kah": 69663,
+      "▁GnRH": 69664,
+      "不动产登记": 69665,
+      "公司董事长": 69666,
+      "证券交易所": 69667,
+      "gerald": 69668,
+      "▁configs": 69669,
+      "人大常委会副主任": 69670,
+      "巴掌": 69671,
+      "货品": 69672,
+      "也并不": 69673,
+      "恨不得": 69674,
+      "所以你": 69675,
+      "▁IMO": 69676,
+      "以下有关": 69677,
+      "数学模型": 69678,
+      "ecode": 69679,
+      "▁Hyde": 69680,
+      "▁Tale": 69681,
+      "▁domest": 69682,
+      "amentals": 69683,
+      "▁Intelligent": 69684,
+      "▁screenwriter": 69685,
+      "之差": 69686,
+      "照亮": 69687,
+      "迫使": 69688,
+      "bmp": 69689,
+      "上诉费": 69690,
+      "流量的": 69691,
+      "\\\\)”": 69692,
+      "igne": 69693,
+      "不可替代": 69694,
+      "年下半年": 69695,
+      "批发市场": 69696,
+      "▁Ajax": 69697,
+      "▁Silk": 69698,
+      "▁miscon": 69699,
+      "powershell": 69700,
+      "LESS": 69701,
+      "neut": 69702,
+      "▁Exhib": 69703,
+      "▁unemployed": 69704,
+      "储能": 69705,
+      "铭记": 69706,
+      "快捷键": 69707,
+      "简单来说": 69708,
+      "这个角色": 69709,
+      "▁Getty": 69710,
+      "▁heats": 69711,
+      "Canadian": 69712,
+      "▁Sanchez": 69713,
+      "帮我做一道选择题": 69714,
+      "nMoreover": 69715,
+      "▁attaching": 69716,
+      "一回": 69717,
+      "按压": 69718,
+      "接管": 69719,
+      "整容": 69720,
+      "筹资": 69721,
+      "白内障": 69722,
+      "hara": 69723,
+      "nEnt": 69724,
+      "▁Mour": 69725,
+      "▁Yosh": 69726,
+      "▁regularization": 69727,
+      "▁Yi": 69728,
+      "权利的": 69729,
+      "twig": 69730,
+      "▁suture": 69731,
+      "信息的若干规定": 69732,
+      "给出正确的选项": 69733,
+      "Gaussian": 69734,
+      "aturdays": 69735,
+      "▁fossils": 69736,
+      "▁spinach": 69737,
+      "▁filmmakers": 69738,
+      "仔细阅读下面这道选择题": 69739,
+      "▁Accessories": 69740,
+      "振荡": 69741,
+      "▁dns": 69742,
+      "进行了研究": 69743,
+      "ichlor": 69744,
+      "racial": 69745,
+      "是一部连载于": 69746,
+      "identally": 69747,
+      "▁electroc": 69748,
+      "▁sociology": 69749,
+      "▁abbreviated": 69750,
+      "照相": 69751,
+      "了解的": 69752,
+      "▁Duc": 69753,
+      "不当得利": 69754,
+      "aston": 69755,
+      "▁bots": 69756,
+      "▁pyro": 69757,
+      "▁inlet": 69758,
+      "▁undet": 69759,
+      "请您仔细阅读": 69760,
+      "▁celery": 69761,
+      "占总": 69762,
+      "记事": 69763,
+      "wor": 69764,
+      "呈批表": 69765,
+      "YYYY": 69766,
+      "▁人物名称": 69767,
+      "agency": 69768,
+      "▁MacOS": 69769,
+      "第一百四十三条": 69770,
+      "▁roadmap": 69771,
+      "▁entrants": 69772,
+      "您需在入住时另行支付": 69773,
+      "▁degenerate": 69774,
+      "▁resembling": 69775,
+      "天一": 69776,
+      "托运": 69777,
+      "登临": 69778,
+      "红叶": 69779,
+      "谁会": 69780,
+      "vre": 69781,
+      "分享了": 69782,
+      "nMad": 69783,
+      "到了一个": 69784,
+      "我有一道题": 69785,
+      "▁inland": 69786,
+      "▁Schneider": 69787,
+      "两委": 69788,
+      "侥幸": 69789,
+      "想你": 69790,
+      "旭东": 69791,
+      "热电": 69792,
+      "闭环": 69793,
+      "顾虑": 69794,
+      "SAP": 69795,
+      "虽然在": 69796,
+      "面积约": 69797,
+      "受到影响": 69798,
+      "▁Bates": 69799,
+      "十一条之规定": 69800,
+      "▁Hacker": 69801,
+      "▁CONCLUS": 69802,
+      "▁rhythmic": 69803,
+      "▁Connected": 69804,
+      "最深": 69805,
+      "洞口": 69806,
+      "粗暴": 69807,
+      "衣裙": 69808,
+      "十八年": 69809,
+      "我老公": 69810,
+      "Sever": 69811,
+      "isans": 69812,
+      "▁Else": 69813,
+      "▁Erin": 69814,
+      "cliffe": 69815,
+      "intosh": 69816,
+      "▁nutrit": 69817,
+      "ncontent": 69818,
+      "▁resolver": 69819,
+      "相位": 69820,
+      "石桥": 69821,
+      "▁书评": 69822,
+      "尤其在": 69823,
+      "月上旬": 69824,
+      "Spider": 69825,
+      "rative": 69826,
+      "及早": 69827,
+      "各异": 69828,
+      "晶莹": 69829,
+      "退伍": 69830,
+      "EFF": 69831,
+      "发现有": 69832,
+      "详细介绍了": 69833,
+      "nSteps": 69834,
+      "▁Heter": 69835,
+      "▁Nucle": 69836,
+      "▁Wolfram": 69837,
+      "▁competency": 69838,
+      "▁Fundamental": 69839,
+      "扑克": 69840,
+      "ammu": 69841,
+      "结构设计": 69842,
+      "nline": 69843,
+      "▁Crus": 69844,
+      "二〇一七年七": 69845,
+      "各种信息进行": 69846,
+      "REQUIRE": 69847,
+      "Remember": 69848,
+      "▁statues": 69849,
+      "三联": 69850,
+      "大庆市": 69851,
+      "检察官": 69852,
+      "的感受": 69853,
+      "OPER": 69854,
+      "▁Ist": 69855,
+      "相关专业": 69856,
+      "多领域知识": 69857,
+      "▁Buddy": 69858,
+      "▁Holder": 69859,
+      "infinity": 69860,
+      "▁Brewing": 69861,
+      "anu": 69862,
+      "的软件": 69863,
+      "Wing": 69864,
+      "▁vtk": 69865,
+      "强有力的": 69866,
+      "最常见的": 69867,
+      "Women": 69868,
+      "▁商品编号": 69869,
+      "并在选项中": 69870,
+      "进行解答即可": 69871,
+      "exporter": 69872,
+      "▁Refresh": 69873,
+      "Percentage": 69874,
+      "报记者": 69875,
+      "unde": 69876,
+      "▁Aux": 69877,
+      "▁pitchers": 69878,
+      "▁elemental": 69879,
+      "▁manganese": 69880,
+      "▁Compliance": 69881,
+      "请你帮我选出正确的选项": 69882,
+      "renewcommand": 69883,
+      "去哪": 69884,
+      "泛滥": 69885,
+      "(\"\"": 69886,
+      "JWT": 69887,
+      "但随着": 69888,
+      "这种东西": 69889,
+      "ngood": 69890,
+      "综合运用各种": 69891,
+      "LayoutInflater": 69892,
+      "不失": 69893,
+      "入伍": 69894,
+      "户通": 69895,
+      "源源": 69896,
+      "补习": 69897,
+      "))(": 69898,
+      "可怕的": 69899,
+      "hair": 69900,
+      "lsen": 69901,
+      "Louis": 69902,
+      "Torch": 69903,
+      "dotenv": 69904,
+      "markup": 69905,
+      "最准确的答案": 69906,
+      "▁sitcom": 69907,
+      "二〇一五年二月": 69908,
+      "▁Metrics": 69909,
+      "perfusion": 69910,
+      "一米": 69911,
+      "兰花": 69912,
+      "新诗": 69913,
+      "普拉": 69914,
+      "溯源": 69915,
+      "虚伪": 69916,
+      "▁`_": 69917,
+      "天然的": 69918,
+      "数据和": 69919,
+      "子宫内膜": 69920,
+      "Traits": 69921,
+      "▁Mature": 69922,
+      "▁dermal": 69923,
+      "▁Hausdorff": 69924,
+      "▁callbacks": 69925,
+      "剥离": 69926,
+      "君王": 69927,
+      "土路": 69928,
+      "排毒": 69929,
+      "复制到": 69930,
+      "ANCH": 69931,
+      "FPGA": 69932,
+      "下面是一道": 69933,
+      "nAdded": 69934,
+      "▁curry": 69935,
+      "hearted": 69936,
+      "ostatic": 69937,
+      "reeting": 69938,
+      "并在选项中确定": 69939,
+      "涉及多领域知识": 69940,
+      "Beautiful": 69941,
+      "▁plethora": 69942,
+      "▁stacking": 69943,
+      "▁Permanent": 69944,
+      "▁attackers": 69945,
+      "▁nanotubes": 69946,
+      "下面是一道复杂选择题": 69947,
+      "请仔细阅读问题和选项": 69948,
+      "translations": 69949,
+      "▁Organizations": 69950,
+      "Lu": 69951,
+      "水印": 69952,
+      "法宝": 69953,
+      "独创": 69954,
+      "otr": 69955,
+      "体育运动": 69956,
+      "nlast": 69957,
+      "美术出版社": 69958,
+      "apollo": 69959,
+      "贷款利率计算": 69960,
+      "▁submerged": 69961,
+      "▁Buckingham": 69962,
+      "劳累": 69963,
+      "善意": 69964,
+      "实处": 69965,
+      "时下": 69966,
+      "▁添加": 69967,
+      "lide": 69968,
+      "▁RFID": 69969,
+      "whatever": 69970,
+      "nLicensed": 69971,
+      "▁plantation": 69972,
+      "▁Australians": 69973,
+      "▁齐": 69974,
+      "南沙": 69975,
+      "多于": 69976,
+      "纹身": 69977,
+      "gil": 69978,
+      "出入境": 69979,
+      "▁Gin": 69980,
+      "OTHER": 69981,
+      "▁serm": 69982,
+      "提出撤诉申请": 69983,
+      "最合适的选项": 69984,
+      "▁myelin": 69985,
+      "这道题目是一道": 69986,
+      "信息科技有限公司": 69987,
+      "Collision": 69988,
+      "▁Applying": 69989,
+      "▁unbounded": 69990,
+      "▁neuroimaging": 69991,
+      "VPN": 69992,
+      "在广州": 69993,
+      "▁Rats": 69994,
+      "▁funnel": 69995,
+      "▁noises": 69996,
+      "algebras": 69997,
+      "▁militants": 69998,
+      "第六十七条第三款之规定": 69999,
+      "▁explanatory": 70000,
+      "主食": 70001,
+      "滚子": 70002,
+      "箭头": 70003,
+      "逼近": 70004,
+      "INO": 70005,
+      "▁SOD": 70006,
+      "广播电台": 70007,
+      "首先根据": 70008,
+      "▁funk": 70009,
+      "然后会提供": 70010,
+      "▁saint": 70011,
+      "并确定您认为": 70012,
+      "接下来这道题目是一道": 70013,
+      "▁Hypothesis": 70014,
+      "请您仔细阅读问题和选项": 70015,
+      "▁Perspective": 70016,
+      "并确定您认为最准确的答案": 70017,
+      "买方": 70018,
+      "棒的": 70019,
+      "每夜": 70020,
+      "肯干": 70021,
+      "nCD": 70022,
+      "▁oss": 70023,
+      "建立健全": 70024,
+      "液晶电视": 70025,
+      "膳食纤维": 70026,
+      "日交付执行": 70027,
+      "▁María": 70028,
+      "第二百零八条": 70029,
+      "给你一段描述": 70030,
+      "请你根据这段话": 70031,
+      "▁flipped": 70032,
+      "然后会提供几个选项": 70033,
+      "▁Logistics": 70034,
+      "从选项中选出最合适的选项": 70035,
+      "执行通知履行法律文书确定的义务": 70036,
+      "ác": 70037,
+      "三相": 70038,
+      "国营": 70039,
+      "烧结": 70040,
+      "腾飞": 70041,
+      "国家对": 70042,
+      "特性的": 70043,
+      "JPEG": 70044,
+      "▁tad": 70045,
+      "Serve": 70046,
+      "▁uninter": 70047,
+      "▁Friendly": 70048,
+      "▁Countries": 70049,
+      "迅猛": 70050,
+      "▁删除": 70051,
+      "项和为": 70052,
+      "日起至实际": 70053,
+      "▁garments": 70054,
+      "▁Candidates": 70055,
+      "和王": 70056,
+      "安阳": 70057,
+      "药理": 70058,
+      "Amb": 70059,
+      "了一家": 70060,
+      "eeks": 70061,
+      "的朋友们": 70062,
+      "仔细的思考": 70063,
+      "二〇一五年五月": 70064,
+      "tK": 70065,
+      "题库": 70066,
+      "的新闻": 70067,
+      "],[\"": 70068,
+      "▁Paw": 70069,
+      "相比之下": 70070,
+      "▁nanow": 70071,
+      "▁insider": 70072,
+      "ahrenheit": 70073,
+      "nThursday": 70074,
+      "Everything": 70075,
+      "参与到": 70076,
+      "Dial": 70077,
+      "▁ETH": 70078,
+      "nbase": 70079,
+      "请在给定的": 70080,
+      "接下来您需要": 70081,
+      "该问题需要对": 70082,
+      "回答一个选择题": 70083,
+      "继续执行的条件": 70084,
+      "各种信息进行详细评估": 70085,
+      "▁streamlined": 70086,
+      "接下来您需要回答一个选择题": 70087,
+      "请在给定的选项中挑选出正确答案": 70088,
+      "该问题需要对各种信息进行详细评估": 70089,
+      "改版": 70090,
+      "硕果": 70091,
+      "体育局": 70092,
+      "有原告": 70093,
+      "步行街": 70094,
+      "视为其": 70095,
+      "大户": 70096,
+      "的规律": 70097,
+      "▁liters": 70098,
+      "▁undergrad": 70099,
+      "transformer": 70100,
+      "▁friendships": 70101,
+      "恩怨": 70102,
+      "常见病": 70103,
+      "ndat": 70104,
+      "▁Dish": 70105,
+      "▁Favorite": 70106,
+      "▁cerebrospinal": 70107,
+      "平价": 70108,
+      "犯了": 70109,
+      "笑语": 70110,
+      "萧条": 70111,
+      "重塑": 70112,
+      "的高端": 70113,
+      "▁sua": 70114,
+      "}}}\\\\": 70115,
+      "▁avian": 70116,
+      "▁esters": 70117,
+      "▁INCIDENTAL": 70118,
+      "房内": 70119,
+      "汉代": 70120,
+      "▁ios": 70121,
+      "在线汉语": 70122,
+      "ograd": 70123,
+      "ipotent": 70124,
+      "▁sewage": 70125,
+      "▁Cycling": 70126,
+      "Americans": 70127,
+      "▁knockdown": 70128,
+      "皇子": 70129,
+      "阴暗": 70130,
+      "雅典": 70131,
+      "{}/": 70132,
+      "Mutex": 70133,
+      "▁Mali": 70134,
+      "▁Thur": 70135,
+      "▁Cause": 70136,
+      "▁paran": 70137,
+      "两次以上的影人": 70138,
+      "预测问题的答案": 70139,
+      "Restrict": 70140,
+      "resistance": 70141,
+      "ostructures": 70142,
+      "▁imprisoned": 70143,
+      "▁图": 70144,
+      "展区": 70145,
+      "越发": 70146,
+      "经营部": 70147,
+      "平方米的": 70148,
+      "资产阶级": 70149,
+      "▁Rash": 70150,
+      "并从提供的": 70151,
+      "▁Saxony": 70152,
+      "一段文字描述和": 70153,
+      "请思考以下有关": 70154,
+      "▁apology": 70155,
+      "下面是一个选择题": 70156,
+      "▁underpin": 70157,
+      "仔细预测问题的答案": 70158,
+      "备选答案中选择一个": 70159,
+      "专业领域知识的选择题": 70160,
+      "▁univariate": 70161,
+      "我会给你一段文字描述和": 70162,
+      "▁mindfulness": 70163,
+      "备选答案中选择一个正确的选项": 70164,
+      "我会给你一段文字描述和对应的选项": 70165,
+      "请你从选项中选出你认为正确的答案": 70166,
+      "万分": 70167,
+      "乔丹": 70168,
+      "住所地在": 70169,
+      "额外的床": 70170,
+      "麻烦帮我做": 70171,
+      "convex": 70172,
+      "▁twists": 70173,
+      "麻烦帮我做一道题": 70174,
+      "ninstance": 70175,
+      "▁wingspan": 70176,
+      "setdefault": 70177,
+      "从选项中选出正确的选项": 70178,
+      "▁imprisonment": 70179,
+      "储藏": 70180,
+      "卑微": 70181,
+      "护卫": 70182,
+      "ABB": 70183,
+      "LTE": 70184,
+      "我不能": 70185,
+      "设立的": 70186,
+      "mani": 70187,
+      "工作温度": 70188,
+      "▁Kell": 70189,
+      "nSeptember": 70190,
+      "会社": 70191,
+      "分发": 70192,
+      "前线": 70193,
+      "包的": 70194,
+      "即在": 70195,
+      "打仗": 70196,
+      "▁gt": 70197,
+      "的解集为": 70198,
+      "这个地方": 70199,
+      "ickson": 70200,
+      "labelled": 70201,
+      "ги": 70202,
+      "▁Ç": 70203,
+      "宽的": 70204,
+      "手柄": 70205,
+      "难民": 70206,
+      "年代的": 70207,
+      "▁个整数": 70208,
+      "张额外的床": 70209,
+      "▁swarm": 70210,
+      "委托代理人杨": 70211,
+      "▁resect": 70212,
+      "launcher": 70213,
+      "▁strawberry": 70214,
+      "也从": 70215,
+      "产学": 70216,
+      "我却": 70217,
+      "玉龙": 70218,
+      "ños": 70219,
+      "otus": 70220,
+      "一个好的": 70221,
+      "英雄合击": 70222,
+      "▁goog": 70223,
+      "▁}),\\": 70224,
+      "申请再审称": 70225,
+      "▁moreover": 70226,
+      "说来": 70227,
+      "izu": 70228,
+      "joh": 70229,
+      "承德市": 70230,
+      "onga": 70231,
+      "▁[])": 70232,
+      "返还原告": 70233,
+      "顾名思义": 70234,
+      "▁Kris": 70235,
+      "▁Crohn": 70236,
+      "▁Graphs": 70237,
+      "▁priori": 70238,
+      "▁Modular": 70239,
+      "▁unlocked": 70240,
+      "▁合作两次以上的影人": 70241,
+      "▁Characteristics": 70242,
+      "▁하": 70243,
+      "也知道": 70244,
+      "有效果": 70245,
+      "网游之": 70246,
+      "▁mars": 70247,
+      "ikhail": 70248,
+      "stellar": 70249,
+      "▁Sunset": 70250,
+      "▁fetuses": 70251,
+      "▁locomotive": 70252,
+      "▁volumetric": 70253,
+      "▁erythrocyte": 70254,
+      "▁Manufacturers": 70255,
+      "nPy": 70256,
+      "输卵管": 70257,
+      "charm": 70258,
+      "eners": 70259,
+      "▁juices": 70260,
+      "▁solute": 70261,
+      "MediaType": 70262,
+      "七日内预交案件受理费": 70263,
+      "备忘": 70264,
+      "多米": 70265,
+      "科比": 70266,
+      "([^": 70267,
+      "三轮车": 70268,
+      "依次为": 70269,
+      "加床或": 70270,
+      "系数为": 70271,
+      "▁NCI": 70272,
+      "正在进行": 70273,
+      "第十五章": 70274,
+      "\\')]\\": 70275,
+      "relude": 70276,
+      "▁Starter": 70277,
+      "▁\\\\\\\\\\\\\\\\": 70278,
+      "▁antidepressant": 70279,
+      "上官": 70280,
+      "人道": 70281,
+      "特权": 70282,
+      "玉树": 70283,
+      "网讯": 70284,
+      "赤壁": 70285,
+      "Due": 70286,
+      "多余的": 70287,
+      "青春期": 70288,
+      "IOUS": 70289,
+      "一定程度": 70290,
+      "新农村建设": 70291,
+      "▁Maven": 70292,
+      "申请执行人同意": 70293,
+      "▁blotting": 70294,
+      "iterations": 70295,
+      "HttpResponse": 70296,
+      "▁transporting": 70297,
+      "▁魔": 70298,
+      "与被": 70299,
+      "冲压": 70300,
+      "国标": 70301,
+      "技师": 70302,
+      "村村": 70303,
+      "颇具": 70304,
+      "Slf": 70305,
+      "以及几个": 70306,
+      "机械工业": 70307,
+      "运用扎实": 70308,
+      "▁Holt": 70309,
+      "Arthur": 70310,
+      "▁cutter": 70311,
+      "▁whichever": 70312,
+      "延时": 70313,
+      "男篮": 70314,
+      "缓和": 70315,
+      "azzo": 70316,
+      "当且仅当": 70317,
+      "igins": 70318,
+      "▁Corb": 70319,
+      "履行生效法律文书所确定的义务": 70320,
+      "宛如": 70321,
+      "月末": 70322,
+      "背着": 70323,
+      "DLL": 70324,
+      "人心的": 70325,
+      "人性的": 70326,
+      "▁Naked": 70327,
+      "▁testis": 70328,
+      "▁Gazette": 70329,
+      "▁modelled": 70330,
+      "全军": 70331,
+      "动向": 70332,
+      "改写成": 70333,
+      "程序的": 70334,
+      "RESP": 70335,
+      "▁fen": 70336,
+      "全体员工": 70337,
+      "至少需要": 70338,
+      "genome": 70339,
+      "▁offence": 70340,
+      "Lj": 70341,
+      "庆区": 70342,
+      "温带": 70343,
+      "睡了": 70344,
+      "铁岭": 70345,
+      "根据其": 70346,
+      "鉴定书": 70347,
+      "▁Koz": 70348,
+      "的真实性": 70349,
+      "PSHOT": 70350,
+      "uiser": 70351,
+      "▁Tina": 70352,
+      "▁seab": 70353,
+      "DOCKER": 70354,
+      "▁Lever": 70355,
+      "所提出的任何": 70356,
+      "AsString": 70357,
+      "symmetric": 70358,
+      "nMur": 70359,
+      "上诉案件": 70360,
+      "均需获得": 70361,
+      "并从所给": 70362,
+      "▁Wheat": 70363,
+      "这道题需要您": 70364,
+      "icially": 70365,
+      "接下来的选择题": 70366,
+      "FILENAME": 70367,
+      "nDisplay": 70368,
+      "以便作出正确判断": 70369,
+      "并从所给选项中挑": 70370,
+      "运用扎实的知识和": 70371,
+      "▁Packages": 70372,
+      "Projection": 70373,
+      "请仔细研究接下来的选择题": 70374,
+      "运用扎实的知识和仔细的思考": 70375,
+      "并从所给选项中挑选出正确的答案": 70376,
+      "久之": 70377,
+      "容貌": 70378,
+      "nLos": 70379,
+      "▁Eis": 70380,
+      "tMonday": 70381,
+      "▁glycoproteins": 70382,
+      "报社": 70383,
+      "禁烟": 70384,
+      "电脑上": 70385,
+      "政法大学": 70386,
+      "最简单的": 70387,
+      "Joint": 70388,
+      "乘法分配律": 70389,
+      "书香": 70390,
+      "冬梅": 70391,
+      "锯齿": 70392,
+      "fur": 70393,
+      "▁$<": 70394,
+      "我想说": 70395,
+      "起重机": 70396,
+      "nica": 70397,
+      "▁Pras": 70398,
+      "酒店的确认": 70399,
+      "apunov": 70400,
+      "macros": 70401,
+      "加床或婴儿床": 70402,
+      "的要求均需获得": 70403,
+      "▁Basement": 70404,
+      "▁bothered": 70405,
+      "▁machining": 70406,
+      "\\\\)\\\\({\\\\,\\\\": 70407,
+      "所提出的任何加床或婴儿床": 70408,
+      "丽娟": 70409,
+      "使君": 70410,
+      "朝天": 70411,
+      "cow": 70412,
+      "vie": 70413,
+      "做的事": 70414,
+      "如下文": 70415,
+      "）**": 70416,
+      "Snap": 70417,
+      "这种问题": 70418,
+      "具有较高的": 70419,
+      "splice": 70420,
+      "▁Divine": 70421,
+      "▁crises": 70422,
+      "人中": 70423,
+      "<%@": 70424,
+      "Rabbit": 70425,
+      "▁favors": 70426,
+      "▁strang": 70427,
+      "programs": 70428,
+      "▁GeForce": 70429,
+      "▁Phantom": 70430,
+      "otransfer": 70431,
+      "▁intraocular": 70432,
+      "不爽": 70433,
+      "两部": 70434,
+      "屋子": 70435,
+      "绪论": 70436,
+      "遍地": 70437,
+      "西城区": 70438,
+      "可以用来": 70439,
+      "是世界上": 70440,
+      "▁USDA": 70441,
+      "月获得表扬": 70442,
+      "galact": 70443,
+      "的要求均需获得酒店的确认": 70444,
+      "欣慰": 70445,
+      "辛亥": 70446,
+      "可以多": 70447,
+      "而努力": 70448,
+      "Vote": 70449,
+      "物品清单": 70450,
+      "系统工程": 70451,
+      "▁salads": 70452,
+      "集团有限责任公司": 70453,
+      "ViewGroup": 70454,
+      "EntityFramework": 70455,
+      "内皮": 70456,
+      "牛津": 70457,
+      "遮挡": 70458,
+      "mpi": 70459,
+      "在很多": 70460,
+      "我们去": 70461,
+      "活力的": 70462,
+      "科尔沁": 70463,
+      "\\\\([": 70464,
+      "slope": 70465,
+      "▁allied": 70466,
+      "▁Pioneer": 70467,
+      "▁recalls": 70468,
+      "eterminate": 70469,
+      "圣地": 70470,
+      "巢湖": 70471,
+      "慈溪": 70472,
+      "细雨": 70473,
+      "迈向": 70474,
+      "齐聚": 70475,
+      "▁Haj": 70476,
+      "互联网金融": 70477,
+      "▁Relax": 70478,
+      "▁Believe": 70479,
+      "▁agreeing": 70480,
+      "美满": 70481,
+      "赣州": 70482,
+      "走着": 70483,
+      "▁Vacc": 70484,
+      "▁Julius": 70485,
+      "▁Growing": 70486,
+      "▁Ibrahim": 70487,
+      "▁clauses": 70488,
+      "▁Facilities": 70489,
+      "宽广": 70490,
+      "odr": 70491,
+      "▁Kul": 70492,
+      "nread": 70493,
+      "▁软件大小": 70494,
+      "Mirror": 70495,
+      "▁mates": 70496,
+      "▁ctypes": 70497,
+      "博大": 70498,
+      "开挖": 70499,
+      "ARN": 70500,
+      "ías": 70501,
+      "▁kl": 70502,
+      "▁사용": 70503,
+      "年前的": 70504,
+      "连载的一部": 70505,
+      "Prompt": 70506,
+      "ologie": 70507,
+      "▁scram": 70508,
+      "Scenario": 70509,
+      "▁entails": 70510,
+      "▁prophylactic": 70511,
+      "敲门": 70512,
+      "正交": 70513,
+      "药水": 70514,
+      "集训": 70515,
+      "但因为": 70516,
+      "房屋的": 70517,
+      "▁MeV": 70518,
+      "PATCH": 70519,
+      "films": 70520,
+      "鄂尔多斯市": 70521,
+      "Illegal": 70522,
+      "PROPERTY": 70523,
+      "stranded": 70524,
+      "ievements": 70525,
+      "serialized": 70526,
+      "PMI": 70527,
+      "拒绝了": 70528,
+      "脑子里": 70529,
+      "LOGO": 70530,
+      "▁hors": 70531,
+      "▁erythemat": 70532,
+      "▁planetary": 70533,
+      "会话": 70534,
+      "寿光": 70535,
+      "知网": 70536,
+      "不开心": 70537,
+      "战略性": 70538,
+      "▁[**": 70539,
+      "认为自己": 70540,
+      "▁JPEG": 70541,
+      "▁appre": 70542,
+      "▁demon": 70543,
+      "▁tuples": 70544,
+      "▁conductors": 70545,
+      "无端": 70546,
+      "欣喜": 70547,
+      "的美好": 70548,
+      "连载的": 70549,
+      "▁Warrior": 70550,
+      "▁distutils": 70551,
+      "▁malformations": 70552,
+      "奠基": 70553,
+      "蠕动": 70554,
+      "光驱类型": 70555,
+      "Brain": 70556,
+      "ranked": 70557,
+      "olecyst": 70558,
+      "▁mortar": 70559,
+      "▁Futures": 70560,
+      "▁overarching": 70561,
+      "喜庆": 70562,
+      "Cli": 70563,
+      "今天就": 70564,
+      "anos": 70565,
+      "tsys": 70566,
+      "迄今为止": 70567,
+      "退役军人": 70568,
+      "▁Vand": 70569,
+      "meeting": 70570,
+      "▁ruined": 70571,
+      "▁defenses": 70572,
+      "▁transcend": 70573,
+      "instruction": 70574,
+      "两头": 70575,
+      "几倍": 70576,
+      "勒索": 70577,
+      "秦淮": 70578,
+      "表单": 70579,
+      "します": 70580,
+      "▁tul": 70581,
+      "Rendering": 70582,
+      "房屋买卖合同纠纷一案": 70583,
+      "身在": 70584,
+      "人家的": 70585,
+      "▁png": 70586,
+      "dirty": 70587,
+      "代理权限为": 70588,
+      "Wizard": 70589,
+      "enteric": 70590,
+      "▁Morton": 70591,
+      "三连": 70592,
+      "推特": 70593,
+      "研判": 70594,
+      "站台": 70595,
+      "蚝油": 70596,
+      "Cod": 70597,
+      "FIX": 70598,
+      "dsl": 70599,
+      "▁电子": 70600,
+      "长时间的": 70601,
+      "▁Vera": 70602,
+      "发展中国家": 70603,
+      "▁oauth": 70604,
+      "▁Stones": 70605,
+      "Channels": 70606,
+      "▁emperor": 70607,
+      "乳头": 70608,
+      "之后就": 70609,
+      "我心里": 70610,
+      "DEST": 70611,
+      "Mono": 70612,
+      "▁([]": 70613,
+      "▁rav": 70614,
+      "将进一步": 70615,
+      "▁datos": 70616,
+      "▁pouring": 70617,
+      "amphetamine": 70618,
+      "]**": 70619,
+      "▁Acad": 70620,
+      "▁blew": 70621,
+      "对其从轻处罚": 70622,
+      "missive": 70623,
+      "▁Backup": 70624,
+      "▁nesting": 70625,
+      "嗓子": 70626,
+      "某于": 70627,
+      "SCR": 70628,
+      "▁lp": 70629,
+      "她们的": 70630,
+      "软组织": 70631,
+      "LONG": 70632,
+      "理论基础": 70633,
+      "Spanish": 70634,
+      "restaurant": 70635,
+      "▁treasures": 70636,
+      "▁mosquitoes": 70637,
+      "▁noteworthy": 70638,
+      "▁严": 70639,
+      "八九": 70640,
+      "毅然": 70641,
+      "钟情": 70642,
+      "':'": 70643,
+      "也很难": 70644,
+      "爱企查": 70645,
+      "Asia": 70646,
+      "▁ACh": 70647,
+      "▁Nina": 70648,
+      "▁owed": 70649,
+      "Gender": 70650,
+      "Nested": 70651,
+      "nLouis": 70652,
+      "▁Botan": 70653,
+      "▁Reach": 70654,
+      "▁Leipzig": 70655,
+      "▁Velocity": 70656,
+      "▁②": 70657,
+      "方剂": 70658,
+      "不相信": 70659,
+      "商业街": 70660,
+      "地下室": 70661,
+      "资料来源": 70662,
+      "▁Weld": 70663,
+      "由题意可知": 70664,
+      "UREMENT": 70665,
+      "quantum": 70666,
+      "infrared": 70667,
+      "短发": 70668,
+      "耐热": 70669,
+      "bir": 70670,
+      "▁Mö": 70671,
+      "▁试读": 70672,
+      "的单位": 70673,
+      "休闲小憩": 70674,
+      "▁Meat": 70675,
+      "中刑执字第": 70676,
+      "▁Ninja": 70677,
+      "notebook": 70678,
+      "▁prompting": 70679,
+      "▁insensitive": 70680,
+      "从上": 70681,
+      "法检测": 70682,
+      "特点和": 70683,
+      "的人是": 70684,
+      "看到这个": 70685,
+      "这种情况下": 70686,
+      "▁Violence": 70687,
+      "▁taxation": 70688,
+      "南宫": 70689,
+      "娄底": 70690,
+      "比尔": 70691,
+      "灵异": 70692,
+      "里奥": 70693,
+      "执行逮捕": 70694,
+      "essie": 70695,
+      "occupied": 70696,
+      "▁NotFound": 70697,
+      "▁Numerous": 70698,
+      "▁groundbreaking": 70699,
+      "平庸": 70700,
+      "林的": 70701,
+      "江海": 70702,
+      "点开": 70703,
+      "球星": 70704,
+      "RIC": 70705,
+      "beh": 70706,
+      "▁Vie": 70707,
+      "行政部门": 70708,
+      "▁Malt": 70709,
+      "nAsked": 70710,
+      "nDepartment": 70711,
+      "▁proteolytic": 70712,
+      "▁approximated": 70713,
+      "更佳": 70714,
+      "西宁": 70715,
+      "aqu": 70716,
+      "▁Cz": 70717,
+      "外星人": 70718,
+      "工作组": 70719,
+      "时间去": 70720,
+      "但是没有": 70721,
+      "在不同的": 70722,
+      "显着高于": 70723,
+      "服务器的": 70724,
+      "这组数据": 70725,
+      "tunsigned": 70726,
+      "▁interoper": 70727,
+      "tE": 70728,
+      "中生": 70729,
+      "我更": 70730,
+      "普罗": 70731,
+      "涂装": 70732,
+      "喜欢上": 70733,
+      "auty": 70734,
+      "mere": 70735,
+      "nDave": 70736,
+      "nEric": 70737,
+      "▁thirst": 70738,
+      "dimethyl": 70739,
+      "又好": 70740,
+      "吃掉": 70741,
+      "过失": 70742,
+      "▁Cp": 70743,
+      "Bron": 70744,
+      "▁Xin": 70745,
+      "▁Kant": 70746,
+      "如下文所示": 70747,
+      "▁Anthrop": 70748,
+      "▁spectators": 70749,
+      "乡长": 70750,
+      "滑坡": 70751,
+      "苍茫": 70752,
+      "▁iç": 70753,
+      "塑料袋": 70754,
+      "定义在": 70755,
+      "管理等": 70756,
+      "nNet": 70757,
+      "trees": 70758,
+      "▁café": 70759,
+      "▁Vault": 70760,
+      "▁rains": 70761,
+      "▁oblique": 70762,
+      "▁HttpServlet": 70763,
+      "不以": 70764,
+      "之都": 70765,
+      "当下的": 70766,
+      "新模式": 70767,
+      "循序渐进": 70768,
+      "nemail": 70769,
+      "Makefile": 70770,
+      "checksum": 70771,
+      "erequisites": 70772,
+      "▁健": 70773,
+      "下颌": 70774,
+      "书生": 70775,
+      "分析仪": 70776,
+      "在欧洲": 70777,
+      "委党校": 70778,
+      "LICK": 70779,
+      "▁IPO": 70780,
+      "▁Zam": 70781,
+      "冻结被告": 70782,
+      "积极作用": 70783,
+      "nBrian": 70784,
+      "▁Choir": 70785,
+      "▁sided": 70786,
+      "ObjectId": 70787,
+      "Tokenizer": 70788,
+      "左键": 70789,
+      "贪婪": 70790,
+      "逆向": 70791,
+      "▁医院": 70792,
+      "不住了": 70793,
+      "乐山市": 70794,
+      "打不过": 70795,
+      "▁Vod": 70796,
+      "▁pav": 70797,
+      "以为自己": 70798,
+      "分析步骤": 70799,
+      "等农作物": 70800,
+      "▁slit": 70801,
+      "做一个选择题": 70802,
+      "printer": 70803,
+      "▁剑": 70804,
+      "上当": 70805,
+      "伴奏": 70806,
+      "抓取": 70807,
+      "十九年": 70808,
+      "byshev": 70809,
+      "ingles": 70810,
+      "▁graphite": 70811,
+      "▁clipboard": 70812,
+      "▁emphasizing": 70813,
+      "≤\\": 70814,
+      "岳麓": 70815,
+      "木工": 70816,
+      "CBA": 70817,
+      "收入的": 70818,
+      "邢台市": 70819,
+      "▁Lor": 70820,
+      "是单选题": 70821,
+      "▁Dust": 70822,
+      "EMPLARY": 70823,
+      "▁feathers": 70824,
+      "ào": 70825,
+      "寄存": 70826,
+      "布料": 70827,
+      "悠然": 70828,
+      "我行": 70829,
+      "河边": 70830,
+      "达拉": 70831,
+      "不对称": 70832,
+      "体积是": 70833,
+      "▁agr": 70834,
+      "▁IDEA": 70835,
+      "nLiter": 70836,
+      "▁gears": 70837,
+      "▁Houses": 70838,
+      "▁trimmed": 70839,
+      "▁standpoint": 70840,
+      "揭牌": 70841,
+      "felt": 70842,
+      "身体素质": 70843,
+      "▁whip": 70844,
+      "verses": 70845,
+      "▁shutter": 70846,
+      "intensive": 70847,
+      "validated": 70848,
+      "▁warrants": 70849,
+      "▁defendants": 70850,
+      "为谁": 70851,
+      "划线": 70852,
+      "妈咪": 70853,
+      "足协": 70854,
+      "题名": 70855,
+      "lei": 70856,
+      "sinA": 70857,
+      "▁${{\\\\": 70858,
+      "悔改表现突出": 70859,
+      "patible": 70860,
+      "▁bassist": 70861,
+      "▁å": 70862,
+      "网易云音乐": 70863,
+      "▁Volvo": 70864,
+      "nMobile": 70865,
+      "▁免费洗浴用品": 70866,
+      "elligence": 70867,
+      "何在": 70868,
+      "毅力": 70869,
+      "畏惧": 70870,
+      "总承包": 70871,
+      "用除法": 70872,
+      "的局面": 70873,
+      "defs": 70874,
+      "icom": 70875,
+      "nOUT": 70876,
+      "nost": 70877,
+      "nthese": 70878,
+      "rights": 70879,
+      "▁Clerk": 70880,
+      "nExercise": 70881,
+      "nSynopsis": 70882,
+      "▁optionally": 70883,
+      "▁silhouette": 70884,
+      "▁recommending": 70885,
+      "不回": 70886,
+      "心血": 70887,
+      "所著": 70888,
+      "是哪": 70889,
+      "神经元": 70890,
+      "Chief": 70891,
+      "centos": 70892,
+      "这道题是单选题": 70893,
+      "请你做一个选择题": 70894,
+      "请你选出正确的选项": 70895,
+      "▁shipments": 70896,
+      "问题和选项如下文所示": 70897,
+      "▁Arithmetic": 70898,
+      "悲惨": 70899,
+      "找寻": 70900,
+      "素的": 70901,
+      "虚空": 70902,
+      "pep": 70903,
+      "tand": 70904,
+      "uckle": 70905,
+      "Despite": 70906,
+      "allocate": 70907,
+      "▁optimally": 70908,
+      "▁verifying": 70909,
+      "史学": 70910,
+      "竹马": 70911,
+      "件事情": 70912,
+      "的热爱": 70913,
+      "\\'\",": 70914,
+      "orna": 70915,
+      "全面推进": 70916,
+      "nWait": 70917,
+      "imilar": 70918,
+      "nMuseum": 70919,
+      "▁disciples": 70920,
+      "▁demolition": 70921,
+      "江门": 70922,
+      "调速": 70923,
+      "▁pits": 70924,
+      "exposure": 70925,
+      "▁reimbursement": 70926,
+      "天蝎": 70927,
+      "扫黑": 70928,
+      "遴选": 70929,
+      "▁rut": 70930,
+      "nsign": 70931,
+      "这是一个涉及": 70932,
+      "▁cassette": 70933,
+      "了很久": 70934,
+      "将达到": 70935,
+      "STAR": 70936,
+      "acom": 70937,
+      "建立一个": 70938,
+      "艺术设计": 70939,
+      "▁vesicle": 70940,
+      "▁POSSIBILITY": 70941,
+      "川市": 70942,
+      "nAbs": 70943,
+      "就是我们": 70944,
+      "▁calendars": 70945,
+      "▁butterflies": 70946,
+      "伤寒": 70947,
+      "特价": 70948,
+      "眉头": 70949,
+      "骑车": 70950,
+      "Esc": 70951,
+      "我都会": 70952,
+      "lain": 70953,
+      "分析并从": 70954,
+      "yzing": 70955,
+      "▁Laud": 70956,
+      "▁fantas": 70957,
+      "userInfo": 70958,
+      "▁locating": 70959,
+      "▁descriptors": 70960,
+      "▁progenitors": 70961,
+      "▁RX": 70962,
+      "赡养费": 70963,
+      "Sense": 70964,
+      "故意杀人罪": 70965,
+      "▁knives": 70966,
+      "biological": 70967,
+      "▁Berkshire": 70968,
+      "▁conjugates": 70969,
+      "▁inclination": 70970,
+      "▁subprojects": 70971,
+      "▁春": 70972,
+      "扰乱": 70973,
+      "睁开": 70974,
+      "虫子": 70975,
+      "BMP": 70976,
+      "MAR": 70977,
+      "人员和": 70978,
+      "文学奖": 70979,
+      "管理水平": 70980,
+      "asers": 70981,
+      "ronics": 70982,
+      "▁CMake": 70983,
+      "▁stark": 70984,
+      "▁Corpus": 70985,
+      "▁Sodium": 70986,
+      "四章": 70987,
+      "泰兴": 70988,
+      "纤细": 70989,
+      ":],": 70990,
+      "年上海": 70991,
+      "示意图": 70992,
+      "等其他": 70993,
+      "结构与": 70994,
+      "信息的选择题": 70995,
+      "给定的选项中": 70996,
+      "▁Tucson": 70997,
+      "您需要分析并从": 70998,
+      "▁QString": 70999,
+      "请仔细研究以下问题": 71000,
+      "综合运用各种信息的选择题": 71001,
+      "您需要分析并从给定的选项中": 71002,
+      "▁'$": 71003,
+      "与我们": 71004,
+      "UBLE": 71005,
+      "▁bye": 71006,
+      "候选答案": 71007,
+      "基本参数": 71008,
+      "说法错误": 71009,
+      "进一步的": 71010,
+      "▁Khal": 71011,
+      "▁Gabri": 71012,
+      "▁tossed": 71013,
+      "▁congest": 71014,
+      "preferences": 71015,
+      "▁cybersecurity": 71016,
+      "鸭子": 71017,
+      "▁[/": 71018,
+      "Proof": 71019,
+      "nMaking": 71020,
+      "setData": 71021,
+      "▁generosity": 71022,
+      "▁integrative": 71023,
+      "▁HY": 71024,
+      "压力大": 71025,
+      "▁Bam": 71026,
+      "伟大复兴": 71027,
+      "感兴趣的": 71028,
+      "抽样调查": 71029,
+      "▁bakery": 71030,
+      "nAlgebra": 71031,
+      "TypeError": 71032,
+      "▁WebSocket": 71033,
+      "▁embracing": 71034,
+      "物业管理有限责任公司": 71035,
+      "▁counterexample": 71036,
+      "苏丹": 71037,
+      "Sig": 71038,
+      "[{\"": 71039,
+      "▁名称": 71040,
+      "▁EMS": 71041,
+      "交通工具": 71042,
+      "目前没有": 71043,
+      "ifera": 71044,
+      "often": 71045,
+      "下面我会给你": 71046,
+      "的撤诉申请符合法律规定": 71047,
+      "▁peroxidation": 71048,
+      "匠心": 71049,
+      "防潮": 71050,
+      "GSM": 71051,
+      "只是个": 71052,
+      "的方案": 71053,
+      "urse": 71054,
+      "中山大学": 71055,
+      "whole": 71056,
+      "请你仔细分析": 71057,
+      "irected": 71058,
+      "▁Icelandic": 71059,
+      "微小": 71060,
+      "筹集": 71061,
+      "▁最新": 71062,
+      "如果想要": 71063,
+      "▁golfers": 71064,
+      "以及几个候选答案": 71065,
+      "▁overdose": 71066,
+      "下面我会给你一道题": 71067,
+      "然后选出正确的选项": 71068,
+      "▁counselor": 71069,
+      "▁marriages": 71070,
+      "打发": 71071,
+      "贪心": 71072,
+      "输血": 71073,
+      "黑板": 71074,
+      "Hyp": 71075,
+      "bla": 71076,
+      "rne": 71077,
+      "可以给": 71078,
+      "衢州市": 71079,
+      "过来了": 71080,
+      "重生之": 71081,
+      "Strip": 71082,
+      "ulton": 71083,
+      "yssey": 71084,
+      "▁Mons": 71085,
+      "▁manic": 71086,
+      "uttgart": 71087,
+      "▁comorbidities": 71088,
+      "jr": 71089,
+      "农资": 71090,
+      "断层": 71091,
+      "▁UM": 71092,
+      "到什么": 71093,
+      "目标和": 71094,
+      "▁SUR": 71095,
+      "城市规划": 71096,
+      "nMade": 71097,
+      "▁Ramsey": 71098,
+      "▁youths": 71099,
+      "▁validating": 71100,
+      "对本": 71101,
+      "放出": 71102,
+      "缔造": 71103,
+      "mvc": 71104,
+      "的地步": 71105,
+      "行为规范": 71106,
+      "银行同期": 71107,
+      "▁dean": 71108,
+      "▁pedig": 71109,
+      "▁aliens": 71110,
+      "▁dances": 71111,
+      "信息查询结果单": 71112,
+      "Negative": 71113,
+      "nuniform": 71114,
+      "unningham": 71115,
+      "▁Warehouse": 71116,
+      "▁periodont": 71117,
+      "▁ultrasonography": 71118,
+      "又说": 71119,
+      "▁gib": 71120,
+      "ubern": 71121,
+      "▁cler": 71122,
+      "#,": 71123,
+      "七天": 71124,
+      "声中": 71125,
+      "大名": 71126,
+      "悲观": 71127,
+      "与世界": 71128,
+      "准考证": 71129,
+      "gpio": 71130,
+      "▁Wit": 71131,
+      "thick": 71132,
+      "申请执行人杨": 71133,
+      "违反法律规定": 71134,
+      "realpath": 71135,
+      "▁royalty": 71136,
+      "▁kinematic": 71137,
+      "栖霞": 71138,
+      "疏导": 71139,
+      "▁cz": 71140,
+      "基本功": 71141,
+      "(\"{}": 71142,
+      "▁Luna": 71143,
+      "▁thrombo": 71144,
+      "textwidth": 71145,
+      "▁disciplinary": 71146,
+      "仙子": 71147,
+      "舅舅": 71148,
+      "颤抖": 71149,
+      "ICP": 71150,
+      "两方面": 71151,
+      "还能够": 71152,
+      "▁wget": 71153,
+      "▁chili": 71154,
+      "oplasma": 71155,
+      "▁Segment": 71156,
+      "▁finalist": 71157,
+      "九九": 71158,
+      "固有": 71159,
+      "李子": 71160,
+      "症的": 71161,
+      "变更为": 71162,
+      "赔偿金": 71163,
+      "Econ": 71164,
+      "▁LOL": 71165,
+      "▁USC": 71166,
+      "高职高专": 71167,
+      "▁Slide": 71168,
+      "▁torus": 71169,
+      "▁Harley": 71170,
+      "▁Willis": 71171,
+      "acoustic": 71172,
+      "▁embroidery": 71173,
+      "北辰": 71174,
+      "各区": 71175,
+      "妆容": 71176,
+      "截然": 71177,
+      "$_{": 71178,
+      "委常委": 71179,
+      "审查核实": 71180,
+      "▁Cliff": 71181,
+      "扣划被执行人": 71182,
+      "▁Buddha": 71183,
+      "▁islets": 71184,
+      "Enumerator": 71185,
+      "▁Recordings": 71186,
+      "墓地": 71187,
+      "官司": 71188,
+      "霍尔": 71189,
+      "▁Ez": 71190,
+      "这两天": 71191,
+      "muted": 71192,
+      "endale": 71193,
+      "▁Realty": 71194,
+      "▁Listing": 71195,
+      "▁Managed": 71196,
+      "不讲": 71197,
+      "光亮": 71198,
+      "大观": 71199,
+      "日产": 71200,
+      "杀虫": 71201,
+      "Cre": 71202,
+      "eque": 71203,
+      "umba": 71204,
+      "▁uni": 71205,
+      "一定要有": 71206,
+      "自由恋爱": 71207,
+      "▁trilogy": 71208,
+      "即得": 71209,
+      "可观": 71210,
+      "大黄": 71211,
+      "轻便": 71212,
+      "迎春": 71213,
+      "udp": 71214,
+      "▁第五章": 71215,
+      "从而导致": 71216,
+      "▁stalk": 71217,
+      "ncurrent": 71218,
+      "▁Allison": 71219,
+      "▁depended": 71220,
+      "▁catalysis": 71221,
+      "▁underline": 71222,
+      "伊始": 71223,
+      "和张": 71224,
+      "白白": 71225,
+      "CSI": 71226,
+      "HAL": 71227,
+      "也同样": 71228,
+      "▁PUT": 71229,
+      "▁dia": 71230,
+      "nKevin": 71231,
+      "▁Nickel": 71232,
+      "Dockerfile": 71233,
+      "▁visualized": 71234,
+      "lx": 71235,
+      "刀片": 71236,
+      "青花": 71237,
+      "调味料": 71238,
+      "mods": 71239,
+      "为真命题": 71240,
+      "质量监督": 71241,
+      "▁siRNA": 71242,
+      "法定代表人杨": 71243,
+      "▁warrior": 71244,
+      "水区": 71245,
+      "肉体": 71246,
+      "了不起": 71247,
+      "永远的": 71248,
+      "Chan": 71249,
+      "nIll": 71250,
+      "优质服务": 71251,
+      "▁Wistar": 71252,
+      "▁invoices": 71253,
+      "▁interruption": 71254,
+      "略微": 71255,
+      "租用": 71256,
+      "änd": 71257,
+      "▁BN": 71258,
+      "intl": 71259,
+      "▁PAN": 71260,
+      "两组患者": 71261,
+      "工业企业": 71262,
+      "最大限度地": 71263,
+      "Clients": 71264,
+      "▁summers": 71265,
+      "▁american": 71266,
+      "▁fostering": 71267,
+      "nGastropods": 71268,
+      "▁councillors": 71269,
+      "养育": 71270,
+      "差额": 71271,
+      "心痛": 71272,
+      "的热点": 71273,
+      "临床应用": 71274,
+      "比较简单": 71275,
+      "规划教材": 71276,
+      "Saved": 71277,
+      "wired": 71278,
+      "▁Crash": 71279,
+      "ennifer": 71280,
+      "▁minors": 71281,
+      "▁pillars": 71282,
+      "▁Childhood": 71283,
+      "一架": 71284,
+      "加斯": 71285,
+      "待定": 71286,
+      "意气": 71287,
+      "中国梦": 71288,
+      "热烈的": 71289,
+      "▁chan": 71290,
+      "udding": 71291,
+      "▁Coral": 71292,
+      "branded": 71293,
+      "vmatrix": 71294,
+      "▁taxonomy": 71295,
+      "Measurement": 71296,
+      "▁transplants": 71297,
+      "小便": 71298,
+      "不过在": 71299,
+      "为人民": 71300,
+      "▁Frem": 71301,
+      "esture": 71302,
+      "▁ischa": 71303,
+      "Minutes": 71304,
+      "otherwise": 71305,
+      "▁Lebesgue": 71306,
+      "Australian": 71307,
+      "opharyngeal": 71308,
+      "transformed": 71309,
+      "▁商": 71310,
+      "制宜": 71311,
+      "备课": 71312,
+      "nMi": 71313,
+      "教学法": 71314,
+      "涉及的": 71315,
+      "arie": 71316,
+      "水平的提高": 71317,
+      "符合条件的": 71318,
+      "▁cobalt": 71319,
+      "▁Hammond": 71320,
+      "▁Powered": 71321,
+      "▁lettuce": 71322,
+      "▁plasmon": 71323,
+      "▁safeguard": 71324,
+      "ík": 71325,
+      "八首": 71326,
+      "axon": 71327,
+      "hael": 71328,
+      "\"]);\\": 71329,
+      "▁gast": 71330,
+      "ologna": 71331,
+      "▁Dancing": 71332,
+      "▁topping": 71333,
+      "astructures": 71334,
+      "亚目": 71335,
+      "会和": 71336,
+      "抗病": 71337,
+      "缅怀": 71338,
+      "英才": 71339,
+      ">}\\": 71340,
+      "单职业": 71341,
+      "喜欢看": 71342,
+      "数字是": 71343,
+      "▁Neb": 71344,
+      "Spect": 71345,
+      "ursed": 71346,
+      "BINARY": 71347,
+      "centering": 71348,
+      "LinkedList": 71349,
+      "墨水": 71350,
+      "抒情": 71351,
+      "省道": 71352,
+      "股指": 71353,
+      "Sah": 71354,
+      "太大了": 71355,
+      "可以利用": 71356,
+      "法律关系": 71357,
+      "▁denim": 71358,
+      "▁ounce": 71359,
+      "▁Visitors": 71360,
+      "▁inoculated": 71361,
+      "▁macroscopic": 71362,
+      "吃不": 71363,
+      "昂贵": 71364,
+      "蒸馏": 71365,
+      "NTs": 71366,
+      "收入来源": 71367,
+      "▁Kens": 71368,
+      "作品展": 71369,
+      "▁parathyroid": 71370,
+      "壮观": 71371,
+      "散户": 71372,
+      "沧州": 71373,
+      "痛风": 71374,
+      "书法家": 71375,
+      "剩余的": 71376,
+      "文化教育": 71377,
+      "strain": 71378,
+      "ENABLED": 71379,
+      "nTurkish": 71380,
+      "▁balloons": 71381,
+      "▁taxpayer": 71382,
+      "遵守法律法规及监规": 71383,
+      "▁diligence": 71384,
+      "归案": 71385,
+      "朝代": 71386,
+      "赫尔": 71387,
+      "韵味": 71388,
+      "竞争中": 71389,
+      "▁Eur": 71390,
+      "Tester": 71391,
+      "nPress": 71392,
+      "▁skaters": 71393,
+      "▁Commands": 71394,
+      "▁proficient": 71395,
+      "文革": 71396,
+      "耐久": 71397,
+      "耗时": 71398,
+      "阻断": 71399,
+      "雄蕊": 71400,
+      "碧桂园": 71401,
+      ">\")\\": 71402,
+      "▁LIN": 71403,
+      "住湖北省": 71404,
+      "看到一个": 71405,
+      "▁Nass": 71406,
+      "annual": 71407,
+      "工作领导小组": 71408,
+      "ilibili": 71409,
+      "▁getaway": 71410,
+      "▁electrocard": 71411,
+      "警官": 71412,
+      "可以和": 71413,
+      "lighter": 71414,
+      "▁Boxing": 71415,
+      "▁rhesus": 71416,
+      "▁Logging": 71417,
+      "声声": 71418,
+      "STS": 71419,
+      "原标题": 71420,
+      "应支付": 71421,
+      "▁Hak": 71422,
+      "slave": 71423,
+      "共计人民币": 71424,
+      "是不一样的": 71425,
+      "▁Cardiac": 71426,
+      "▁checksum": 71427,
+      "▁chromium": 71428,
+      "经相关部门批准后方可开展经营活动": 71429,
+      "中点": 71430,
+      "乙酸": 71431,
+      "台账": 71432,
+      "在有": 71433,
+      "bst": 71434,
+      "▁те": 71435,
+      "Nome": 71436,
+      "▁ETF": 71437,
+      "欢迎所有": 71438,
+      "环境优美": 71439,
+      "稳产农田": 71440,
+      "并给出解析": 71441,
+      "▁Toolkit": 71442,
+      "▁Superman": 71443,
+      "▁oncogenic": 71444,
+      "▁rewritten": 71445,
+      "▁transistors": 71446,
+      "*$": 71447,
+      "Mb": 71448,
+      "▁车": 71449,
+      "上榜": 71450,
+      "武陵": 71451,
+      "连城": 71452,
+      "arynx": 71453,
+      "▁Fuck": 71454,
+      "▁Pied": 71455,
+      "▁BUILD": 71456,
+      "setName": 71457,
+      "▁dumped": 71458,
+      "▁glowing": 71459,
+      "▁humoral": 71460,
+      "toHaveBeen": 71461,
+      "▁compressive": 71462,
+      "äl": 71463,
+      "下有": 71464,
+      "切碎": 71465,
+      "钱财": 71466,
+      "雪梅": 71467,
+      "风起": 71468,
+      "生活和": 71469,
+      "_(\\'": 71470,
+      "▁MVC": 71471,
+      "Marie": 71472,
+      "ERENCE": 71473,
+      "▁Bohem": 71474,
+      "▁CAUSED": 71475,
+      "▁epitope": 71476,
+      "▁goodies": 71477,
+      "▁follicle": 71478,
+      "equivalent": 71479,
+      "▁Composition": 71480,
+      "选出你认为正确的答案并给出解析": 71481,
+      "SX": 71482,
+      "乐曲": 71483,
+      "Fee": 71484,
+      "是负数": 71485,
+      "顺德区": 71486,
+      "ivar": 71487,
+      "其他国家": 71488,
+      "ereal": 71489,
+      "enguins": 71490,
+      "raising": 71491,
+      "▁coursework": 71492,
+      "宝妈": 71493,
+      "应从": 71494,
+      "时限": 71495,
+      "统统": 71496,
+      "黄酮": 71497,
+      "西北部": 71498,
+      "bmod": 71499,
+      "nBal": 71500,
+      "▁Agr": 71501,
+      "Macro": 71502,
+      "agher": 71503,
+      "有限公司与被告": 71504,
+      "▁specialties": 71505,
+      "又无": 71506,
+      "朱元": 71507,
+      "能与": 71508,
+      "锦江": 71509,
+      "的计划": 71510,
+      "个人观点": 71511,
+      "文明城市": 71512,
+      "Ordered": 71513,
+      "▁Alberto": 71514,
+      "▁Defaults": 71515,
+      "▁piecewise": 71516,
+      "拨款": 71517,
+      "版块": 71518,
+      "只是想": 71519,
+      "pinion": 71520,
+      "▁halves": 71521,
+      "犯容留他人吸毒罪": 71522,
+      "▁biophysical": 71523,
+      "后才": 71524,
+      "国泰": 71525,
+      "则这个": 71526,
+      "OPPO": 71527,
+      "▁sut": 71528,
+      "谈话笔录": 71529,
+      "▁raster": 71530,
+      "▁Gravity": 71531,
+      "▁exiting": 71532,
+      "县市": 71533,
+      "可想": 71534,
+      "日将": 71535,
+      "某在": 71536,
+      "肥皂": 71537,
+      "蜂窝": 71538,
+      ")}(": 71539,
+      "社会上": 71540,
+      "财产的": 71541,
+      "estr": 71542,
+      "▁EPs": 71543,
+      "写出你的": 71544,
+      "▁注意事项": 71545,
+      "Rotate": 71546,
+      "opaedic": 71547,
+      "▁changelog": 71548,
+      "▁positional": 71549,
+      "▁initialState": 71550,
+      "拖动": 71551,
+      "hun": 71552,
+      "▁Sv": 71553,
+      "}\";\\": 71554,
+      "应该怎么": 71555,
+      "nSTATIC": 71556,
+      "cw": 71557,
+      "某天": 71558,
+      "及时的": 71559,
+      "都十分": 71560,
+      "▁gan": 71561,
+      "二〇一六年四": 71562,
+      "imiento": 71563,
+      "▁rotary": 71564,
+      "Saturday": 71565,
+      "automatic": 71566,
+      "挂着": 71567,
+      "耕耘": 71568,
+      "锻造": 71569,
+      "ubi": 71570,
+      "зов": 71571,
+      "▁纠错": 71572,
+      "的培养": 71573,
+      "ROLE": 71574,
+      "▁PLEASE": 71575,
+      "充分不必要条件": 71576,
+      "▁Injection": 71577,
+      "自闭": 71578,
+      "诗意": 71579,
+      "差异化": 71580,
+      "系主任": 71581,
+      "图文并茂": 71582,
+      "▁Rubber": 71583,
+      "▁Michele": 71584,
+      "▁denying": 71585,
+      "naissance": 71586,
+      "号征收社会抚养费决定书": 71587,
+      "▁multinational": 71588,
+      "▁functionalities": 71589,
+      "痛经": 71590,
+      "ych": 71591,
+      "自贡市": 71592,
+      "▁rho": 71593,
+      "novel": 71594,
+      "Making": 71595,
+      "liable": 71596,
+      "▁inhaled": 71597,
+      "neighbors": 71598,
+      "▁Wilderness": 71599,
+      "▁suppresses": 71600,
+      "▁同": 71601,
+      "剖宫": 71602,
+      "为公司": 71603,
+      "待今后": 71604,
+      "第四次": 71605,
+      "leon": 71606,
+      "▁CNC": 71607,
+      "blers": 71608,
+      "▁haha": 71609,
+      "▁作品原文": 71610,
+      "▁polyt": 71611,
+      "安徽省合肥市": 71612,
+      "▁booster": 71613,
+      "▁fascinated": 71614,
+      "▁Ω": 71615,
+      "思绪": 71616,
+      "桥头": 71617,
+      "芭蕾": 71618,
+      "三个人": 71619,
+      "被发现": 71620,
+      "▁Pes": 71621,
+      "▁saga": 71622,
+      "orylation": 71623,
+      "▁allergen": 71624,
+      "▁redistribution": 71625,
+      "湛江": 71626,
+      "补给": 71627,
+      "▁rag": 71628,
+      "langs": 71629,
+      "Andrew": 71630,
+      "▁Clair": 71631,
+      "▁ignores": 71632,
+      "▁refactor": 71633,
+      "▁adversely": 71634,
+      "▁endoscopy": 71635,
+      "▁集": 71636,
+      "跳动": 71637,
+      "鹤壁": 71638,
+      "你分析": 71639,
+      "▁orientations": 71640,
+      "▁quantization": 71641,
+      "有点像": 71642,
+      "贵金属": 71643,
+      "습니다": 71644,
+      "▁CRP": 71645,
+      "▁Jah": 71646,
+      "▁Amin": 71647,
+      "主张的事实": 71648,
+      "ouchers": 71649,
+      "▁Harmony": 71650,
+      "▁speeding": 71651,
+      "▁afterward": 71652,
+      "卫生和计划生育委员会": 71653,
+      "是连载于起点中文网的一部": 71654,
+      "第一百五十四条第一款第五": 71655,
+      "前几": 71656,
+      "生怕": 71657,
+      "都与": 71658,
+      "休闲娱乐": 71659,
+      "Hydro": 71660,
+      "Wrong": 71661,
+      "第八十四条": 71662,
+      "▁popul": 71663,
+      "▁Vehicles": 71664,
+      "▁О": 71665,
+      "吞吐": 71666,
+      "在点": 71667,
+      "▁": 71668,
+      "e": 71669,
+      "n": 71670,
+      "t": 71671,
+      "a": 71672,
+      "i": 71673,
+      "o": 71674,
+      "r": 71675,
+      "s": 71676,
+      "l": 71677,
+      "\\": 71678,
+      "h": 71679,
+      "d": 71680,
+      "c": 71681,
+      "u": 71682,
+      "m": 71683,
+      "p": 71684,
+      "f": 71685,
+      "g": 71686,
+      "y": 71687,
+      "b": 71688,
+      "w": 71689,
+      ",": 71690,
+      "v": 71691,
+      "0": 71692,
+      "1": 71693,
+      "k": 71694,
+      ")": 71695,
+      "(": 71696,
+      "2": 71697,
+      "的": 71698,
+      "T": 71699,
+      "S": 71700,
+      "A": 71701,
+      "\"": 71702,
+      "I": 71703,
+      "x": 71704,
+      "C": 71705,
+      ":": 71706,
+      "/": 71707,
+      "_": 71708,
+      "'": 71709,
+      "3": 71710,
+      "P": 71711,
+      "E": 71712,
+      "M": 71713,
+      "R": 71714,
+      "5": 71715,
+      "{": 71716,
+      "}": 71717,
+      "4": 71718,
+      "D": 71719,
+      "|": 71720,
+      "9": 71721,
+      "N": 71722,
+      ";": 71723,
+      "L": 71724,
+      "#": 71725,
+      "B": 71726,
+      "*": 71727,
+      "F": 71728,
+      "6": 71729,
+      "一": 71730,
+      "O": 71731,
+      "8": 71732,
+      "人": 71733,
+      "$": 71734,
+      "7": 71735,
+      ">": 71736,
+      "H": 71737,
+      "是": 71738,
+      "W": 71739,
+      "j": 71740,
+      "q": 71741,
+      "G": 71742,
+      "有": 71743,
+      "<": 71744,
+      "不": 71745,
+      "z": 71746,
+      "[": 71747,
+      "]": 71748,
+      "U": 71749,
+      "中": 71750,
+      "在": 71751,
+      "了": 71752,
+      "行": 71753,
+      "年": 71754,
+      "为": 71755,
+      "V": 71756,
+      "和": 71757,
+      "法": 71758,
+      "个": 71759,
+      "大": 71760,
+      "生": 71761,
+      "以": 71762,
+      "用": 71763,
+      "本": 71764,
+      "国": 71765,
+      "出": 71766,
+      "上": 71767,
+      "日": 71768,
+      "定": 71769,
+      "我": 71770,
+      "民": 71771,
+      "学": 71772,
+      "K": 71773,
+      "J": 71774,
+      "时": 71775,
+      "理": 71776,
+      "可": 71777,
+      "要": 71778,
+      "这": 71779,
+      "作": 71780,
+      "于": 71781,
+      "会": 71782,
+      "公": 71783,
+      "月": 71784,
+      "被": 71785,
+      "第": 71786,
+      "发": 71787,
+      "能": 71788,
+      "告": 71789,
+      "院": 71790,
+      "对": 71791,
+      "）": 71792,
+      "“": 71793,
+      "（": 71794,
+      "Y": 71795,
+      "到": 71796,
+      "”": 71797,
+      "业": 71798,
+      "分": 71799,
+      "成": 71800,
+      "后": 71801,
+      "下": 71802,
+      "来": 71803,
+      "方": 71804,
+      "市": 71805,
+      "%": 71806,
+      "事": 71807,
+      "就": 71808,
+      "地": 71809,
+      "自": 71810,
+      "执": 71811,
+      "多": 71812,
+      "工": 71813,
+      "数": 71814,
+      "家": 71815,
+      "文": 71816,
+      "@": 71817,
+      "现": 71818,
+      "经": 71819,
+      "小": 71820,
+      "原": 71821,
+      "合": 71822,
+      "二": 71823,
+      "过": 71824,
+      "高": 71825,
+      "书": 71826,
+      "如": 71827,
+      "实": 71828,
+      "也": 71829,
+      "十": 71830,
+      "动": 71831,
+      "开": 71832,
+      "之": 71833,
+      "面": 71834,
+      "与": 71835,
+      "员": 71836,
+      "所": 71837,
+      "同": 71838,
+      "他": 71839,
+      "子": 71840,
+      "产": 71841,
+      "进": 71842,
+      "好": 71843,
+      "你": 71844,
+      "期": 71845,
+      "等": 71846,
+      "得": 71847,
+      "^": 71848,
+      "其": 71849,
+      "诉": 71850,
+      "长": 71851,
+      "新": 71852,
+      "关": 71853,
+      "机": 71854,
+      "区": 71855,
+      "间": 71856,
+      "们": 71857,
+      "都": 71858,
+      "力": 71859,
+      "主": 71860,
+      "司": 71861,
+      "三": 71862,
+      "点": 71863,
+      "`": 71864,
+      "体": 71865,
+      "请": 71866,
+      "全": 71867,
+      "内": 71868,
+      "性": 71869,
+      "说": 71870,
+      "提": 71871,
+      "最": 71872,
+      "题": 71873,
+      "还": 71874,
+      "品": 71875,
+      "名": 71876,
+      "化": 71877,
+      "&": 71878,
+      "解": 71879,
+      "者": 71880,
+      "案": 71881,
+      "天": 71882,
+      "X": 71883,
+      "心": 71884,
+      "程": 71885,
+      "审": 71886,
+      "务": 71887,
+      "代": 71888,
+      "部": 71889,
+      "元": 71890,
+      "加": 71891,
+      "Q": 71892,
+      "前": 71893,
+      "起": 71894,
+      "通": 71895,
+      "条": 71896,
+      "明": 71897,
+      "’": 71898,
+      "么": 71899,
+      "应": 71900,
+      "表": 71901,
+      "判": 71902,
+      "当": 71903,
+      "结": 71904,
+      "度": 71905,
+      "建": 71906,
+      "重": 71907,
+      "及": 71908,
+      "电": 71909,
+      "而": 71910,
+      "平": 71911,
+      "情": 71912,
+      "资": 71913,
+      "种": 71914,
+      "看": 71915,
+      "物": 71916,
+      "无": 71917,
+      "水": 71918,
+      "号": 71919,
+      "据": 71920,
+      "因": 71921,
+      "系": 71922,
+      "道": 71923,
+      "设": 71924,
+      "果": 71925,
+      "安": 71926,
+      "量": 71927,
+      "位": 71928,
+      "相": 71929,
+      "次": 71930,
+      "入": 71931,
+      "并": 71932,
+      "式": 71933,
+      "–": 71934,
+      "制": 71935,
+      "金": 71936,
+      "意": 71937,
+      "没": 71938,
+      "很": 71939,
+      "计": 71940,
+      "里": 71941,
+      "规": 71942,
+      "保": 71943,
+      "然": 71944,
+      "向": 71945,
+      "信": 71946,
+      "省": 71947,
+      "件": 71948,
+      "利": 71949,
+      "网": 71950,
+      "问": 71951,
+      "限": 71952,
+      "正": 71953,
+      "选": 71954,
+      "目": 71955,
+      "教": 71956,
+      "外": 71957,
+      "申": 71958,
+      "科": 71959,
+      "管": 71960,
+      "比": 71961,
+      "四": 71962,
+      "华": 71963,
+      "共": 71964,
+      "由": 71965,
+      "知": 71966,
+      "从": 71967,
+      "着": 71968,
+      "考": 71969,
+      "款": 71970,
+      "场": 71971,
+      "已": 71972,
+      "项": 71973,
+      "决": 71974,
+      "展": 71975,
+      "将": 71976,
+      "某": 71977,
+      "去": 71978,
+      "服": 71979,
+      "但": 71980,
+      "车": 71981,
+      "求": 71982,
+      "百": 71983,
+      "记": 71984,
+      "五": 71985,
+      "认": 71986,
+      "美": 71987,
+      "手": 71988,
+      "参": 71989,
+      "那": 71990,
+      "特": 71991,
+      "路": 71992,
+      "东": 71993,
+      "处": 71994,
+      "交": 71995,
+      "效": 71996,
+      "组": 71997,
+      "更": 71998,
+      "—": 71999,
+      "证": 72000,
+      "常": 72001,
+      "或": 72002,
+      "术": 72003,
+      "任": 72004,
+      "想": 72005,
+      "回": 72006,
+      "裁": 72007,
+      "山": 72008,
+      "两": 72009,
+      "依": 72010,
+      "南": 72011,
+      "政": 72012,
+      "西": 72013,
+      "立": 72014,
+      "海": 72015,
+      "技": 72016,
+      "基": 72017,
+      "接": 72018,
+      "样": 72019,
+      "单": 72020,
+      "活": 72021,
+      "此": 72022,
+      "料": 72023,
+      "己": 72024,
+      "律": 72025,
+      "受": 72026,
+      "师": 72027,
+      "Z": 72028,
+      "社": 72029,
+      "些": 72030,
+      "型": 72031,
+      "费": 72032,
+      "查": 72033,
+      "质": 72034,
+      "给": 72035,
+      "使": 72036,
+      "线": 72037,
+      "类": 72038,
+      "感": 72039,
+      "只": 72040,
+      "商": 72041,
+      "女": 72042,
+      "做": 72043,
+      "别": 72044,
+      "级": 72045,
+      "县": 72046,
+      "确": 72047,
+      "身": 72048,
+      "住": 72049,
+      "形": 72050,
+      "收": 72051,
+      "真": 72052,
+      "直": 72053,
+      "价": 72054,
+      "北": 72055,
+      "统": 72056,
+      "该": 72057,
+      "强": 72058,
+      "达": 72059,
+      "字": 72060,
+      "息": 72061,
+      "至": 72062,
+      "张": 72063,
+      "标": 72064,
+      "色": 72065,
+      "口": 72066,
+      "江": 72067,
+      "~": 72068,
+      "少": 72069,
+      "风": 72070,
+      "刑": 72071,
+      "支": 72072,
+      "运": 72073,
+      "清": 72074,
+      "供": 72075,
+      "各": 72076,
+      "照": 72077,
+      "即": 72078,
+      "传": 72079,
+      "总": 72080,
+      "变": 72081,
+      "需": 72082,
+      "调": 72083,
+      "容": 72084,
+      "老": 72085,
+      "万": 72086,
+      "王": 72087,
+      "专": 72088,
+      "讼": 72089,
+      "引": 72090,
+      "具": 72091,
+      "未": 72092,
+      "持": 72093,
+      "研": 72094,
+      "什": 72095,
+      "台": 72096,
+      "流": 72097,
+      "委": 72098,
+      "城": 72099,
+      "门": 72100,
+      "取": 72101,
+      "气": 72102,
+      "指": 72103,
+      "格": 72104,
+      "广": 72105,
+      "装": 72106,
+      "打": 72107,
+      "根": 72108,
+      "再": 72109,
+      "每": 72110,
+      "先": 72111,
+      "导": 72112,
+      "准": 72113,
+      "议": 72114,
+      "算": 72115,
+      "房": 72116,
+      "图": 72117,
+      "男": 72118,
+      "州": 72119,
+      "示": 72120,
+      "节": 72121,
+      "答": 72122,
+      "让": 72123,
+      "治": 72124,
+      "完": 72125,
+      "义": 72126,
+      "影": 72127,
+      "育": 72128,
+      "创": 72129,
+      "×": 72130,
+      "医": 72131,
+      "论": 72132,
+      "报": 72133,
+      "何": 72134,
+      "值": 72135,
+      "族": 72136,
+      "花": 72137,
+      "游": 72138,
+      "放": 72139,
+      "世": 72140,
+      "集": 72141,
+      "改": 72142,
+      "头": 72143,
+      "版": 72144,
+      "爱": 72145,
+      "光": 72146,
+      "…": 72147,
+      "局": 72148,
+      "企": 72149,
+      "校": 72150,
+      "权": 72151,
+      "造": 72152,
+      "初": 72153,
+      "白": 72154,
+      "汉": 72155,
+      "户": 72156,
+      "把": 72157,
+      "简": 72158,
+      "包": 72159,
+      "村": 72160,
+      "较": 72161,
+      "联": 72162,
+      "连": 72163,
+      "器": 72164,
+      "六": 72165,
+      "复": 72166,
+      "米": 72167,
+      "存": 72168,
+      "她": 72169,
+      "带": 72170,
+      "适": 72171,
+      "备": 72172,
+      "源": 72173,
+      "注": 72174,
+      "话": 72175,
+      "章": 72176,
+      "率": 72177,
+      "营": 72178,
+      "视": 72179,
+      "整": 72180,
+      "非": 72181,
+      "积": 72182,
+      "办": 72183,
+      "语": 72184,
+      "则": 72185,
+      "七": 72186,
+      "究": 72187,
+      "农": 72188,
+      "思": 72189,
+      "称": 72190,
+      "见": 72191,
+      "终": 72192,
+      "故": 72193,
+      "功": 72194,
+      "又": 72195,
+      "林": 72196,
+      "罪": 72197,
+      "精": 72198,
+      "几": 72199,
+      "空": 72200,
+      "且": 72201,
+      "财": 72202,
+      "才": 72203,
+      "今": 72204,
+      "环": 72205,
+      "李": 72206,
+      "病": 72207,
+      "周": 72208,
+      "推": 72209,
+      "球": 72210,
+      "职": 72211,
+      "战": 72212,
+      "试": 72213,
+      "•": 72214,
+      "优": 72215,
+      "界": 72216,
+      "习": 72217,
+      "深": 72218,
+      "构": 72219,
+      "离": 72220,
+      "乐": 72221,
+      "阳": 72222,
+      "神": 72223,
+      "觉": 72224,
+      "反": 72225,
+      "庭": 72226,
+      "验": 72227,
+      "观": 72228,
+      "易": 72229,
+      "队": 72230,
+      "转": 72231,
+      "步": 72232,
+      "送": 72233,
+      "近": 72234,
+      "儿": 72235,
+      "角": 72236,
+      "候": 72237,
+      "许": 72238,
+      "置": 72239,
+      "客": 72240,
+      "片": 72241,
+      "减": 72242,
+      "约": 72243,
+      "增": 72244,
+      "太": 72245,
+      "状": 72246,
+      "份": 72247,
+      "模": 72248,
+      "干": 72249,
+      "检": 72250,
+      "介": 72251,
+      "识": 72252,
+      "配": 72253,
+      "边": 72254,
+      "况": 72255,
+      "马": 72256,
+      "团": 72257,
+      "红": 72258,
+      "按": 72259,
+      "友": 72260,
+      "列": 72261,
+      "除": 72262,
+      "布": 72263,
+      "它": 72264,
+      "首": 72265,
+      "责": 72266,
+      "始": 72267,
+      "态": 72268,
+      "热": 72269,
+      "付": 72270,
+      "快": 72271,
+      "养": 72272,
+      "军": 72273,
+      "京": 72274,
+      "属": 72275,
+      "领": 72276,
+      "双": 72277,
+      "克": 72278,
+      "例": 72279,
+      "消": 72280,
+      "斯": 72281,
+      "济": 72282,
+      "越": 72283,
+      "·": 72284,
+      "护": 72285,
+      "控": 72286,
+      "德": 72287,
+      "犯": 72288,
+      "音": 72289,
+      "河": 72290,
+      "英": 72291,
+      "食": 72292,
+      "施": 72293,
+      "际": 72294,
+      "八": 72295,
+      "显": 72296,
+      "载": 72297,
+      "借": 72298,
+      "范": 72299,
+      "投": 72300,
+      "黄": 72301,
+      "速": 72302,
+      "极": 72303,
+      "素": 72304,
+      "难": 72305,
+      "历": 72306,
+      "获": 72307,
+      "负": 72308,
+      "析": 72309,
+      "择": 72310,
+      "低": 72311,
+      "评": 72312,
+      "站": 72313,
+      "断": 72314,
+      "银": 72315,
+      "云": 72316,
+      "防": 72317,
+      "青": 72318,
+      "言": 72319,
+      "测": 72320,
+      "材": 72321,
+      "细": 72322,
+      "序": 72323,
+      "尔": 72324,
+      "维": 72325,
+      "满": 72326,
+      "续": 72327,
+      "担": 72328,
+      "演": 72329,
+      "酒": 72330,
+      "采": 72331,
+      "超": 72332,
+      "监": 72333,
+      "龙": 72334,
+      "足": 72335,
+      "述": 72336,
+      "买": 72337,
+      "命": 72338,
+      "读": 72339,
+      "均": 72340,
+      "店": 72341,
+      "销": 72342,
+      "像": 72343,
+      "党": 72344,
+      "录": 72345,
+      "境": 72346,
+      "九": 72347,
+      "段": 72348,
+      "随": 72349,
+      "必": 72350,
+      "走": 72351,
+      "喜": 72352,
+      "士": 72353,
+      "股": 72354,
+      "陈": 72355,
+      "石": 72356,
+      "助": 72357,
+      "修": 72358,
+      "承": 72359,
+      "予": 72360,
+      "便": 72361,
+      "象": 72362,
+      "星": 72363,
+      "围": 72364,
+      "划": 72365,
+      "赛": 72366,
+      "戏": 72367,
+      "温": 72368,
+      "园": 72369,
+      "预": 72370,
+      "血": 72371,
+      "刘": 72372,
+      "é": 72373,
+      "协": 72374,
+      "写": 72375,
+      "微": 72376,
+      "排": 72377,
+      "油": 72378,
+      "响": 72379,
+      "景": 72380,
+      "拉": 72381,
+      "怎": 72382,
+      "致": 72383,
+      "欢": 72384,
+      "半": 72385,
+      "疗": 72386,
+      "春": 72387,
+      "吃": 72388,
+      "钱": 72389,
+      "药": 72390,
+      "宝": 72391,
+      "副": 72392,
+      "升": 72393,
+      "压": 72394,
+      "孩": 72395,
+      "望": 72396,
+      "土": 72397,
+      "卡": 72398,
+      "层": 72399,
+      "婚": 72400,
+      "失": 72401,
+      "牌": 72402,
+      "室": 72403,
+      "福": 72404,
+      "千": 72405,
+      "编": 72406,
+      "艺": 72407,
+      "险": 72408,
+      "切": 72409,
+      "湖": 72410,
+      "众": 72411,
+      "吧": 72412,
+      "眼": 72413,
+      "找": 72414,
+      "火": 72415,
+      "声": 72416,
+      "古": 72417,
+      "志": 72418,
+      "纷": 72419,
+      "香": 72420,
+      "输": 72421,
+      "差": 72422,
+      "落": 72423,
+      "托": 72424,
+      "课": 72425,
+      "画": 72426,
+      "兴": 72427,
+      "远": 72428,
+      "〇": 72429,
+      "苏": 72430,
+      "轻": 72431,
+      "底": 72432,
+      "亚": 72433,
+      "异": 72434,
+      "府": 72435,
+      "吗": 72436,
+      "镇": 72437,
+      "撤": 72438,
+      "织": 72439,
+      "止": 72440,
+      "域": 72441,
+      "居": 72442,
+      "黑": 72443,
+      "码": 72444,
+      "错": 72445,
+      "史": 72446,
+      "察": 72447,
+      "群": 72448,
+      "普": 72449,
+      "核": 72450,
+      "博": 72451,
+      "害": 72452,
+      "板": 72453,
+      "健": 72454,
+      "纠": 72455,
+      "络": 72456,
+      "够": 72457,
+      "履": 72458,
+      "击": 72459,
+      "势": 72460,
+      "康": 72461,
+      "留": 72462,
+      "策": 72463,
+      "往": 72464,
+      "死": 72465,
+      "充": 72466,
+      "武": 72467,
+      "益": 72468,
+      "宁": 72469,
+      "争": 72470,
+      "含": 72471,
+      "独": 72472,
+      "词": 72473,
+      "奖": 72474,
+      "令": 72475,
+      "若": 72476,
+      "哪": 72477,
+      "售": 72478,
+      "免": 72479,
+      "互": 72480,
+      "纳": 72481,
+      "举": 72482,
+      "购": 72483,
+      "培": 72484,
+      "波": 72485,
+      "换": 72486,
+      "符": 72487,
+      "听": 72488,
+      "富": 72489,
+      "班": 72490,
+      "罗": 72491,
+      "余": 72492,
+      "却": 72493,
+      "临": 72494,
+      "亲": 72495,
+      "念": 72496,
+      "否": 72497,
+      "早": 72498,
+      "货": 72499,
+      "歌": 72500,
+      "伤": 72501,
+      "仅": 72502,
+      "飞": 72503,
+      "登": 72504,
+      "剧": 72505,
+      "庆": 72506,
+      "频": 72507,
+      "智": 72508,
+      "释": 72509,
+      "奇": 72510,
+      "楼": 72511,
+      "良": 72512,
+      "密": 72513,
+      "味": 72514,
+      "劳": 72515,
+      "卫": 72516,
+      "般": 72517,
+      "皮": 72518,
+      "降": 72519,
+      "玩": 72520,
+      "跟": 72521,
+      "严": 72522,
+      "杨": 72523,
+      "待": 72524,
+      "曲": 72525,
+      "融": 72526,
+      "朋": 72527,
+      "钟": 72528,
+      "川": 72529,
+      "粉": 72530,
+      "讲": 72531,
+      "呢": 72532,
+      "纪": 72533,
+      "官": 72534,
+      "毕": 72535,
+      "移": 72536,
+      "索": 72537,
+      "木": 72538,
+      "曾": 72539,
+      "兰": 72540,
+      "岁": 72541,
+      "软": 72542,
+      "破": 72543,
+      "绍": 72544,
+      "端": 72545,
+      "о": 72546,
+      "盘": 72547,
+      "母": 72548,
+      "印": 72549,
+      "玉": 72550,
+      "批": 72551,
+      "乡": 72552,
+      "绝": 72553,
+      "永": 72554,
+      "础": 72555,
+      "补": 72556,
+      "操": 72557,
+      "患": 72558,
+      "罚": 72559,
+      "帮": 72560,
+      "善": 72561,
+      "甲": 72562,
+      "您": 72563,
+      "夫": 72564,
+      "训": 72565,
+      "享": 72566,
+      "括": 72567,
+      "菜": 72568,
+      "树": 72569,
+      "另": 72570,
+      "愿": 72571,
+      "零": 72572,
+      "铁": 72573,
+      "叶": 72574,
+      "归": 72575,
+      "旅": 72576,
+      "卖": 72577,
+      "彩": 72578,
+      "希": 72579,
+      "额": 72580,
+      "征": 72581,
+      "尽": 72582,
+      "继": 72583,
+      "哈": 72584,
+      "票": 72585,
+      "床": 72586,
+      "综": 72587,
+      "啊": 72588,
+      "假": 72589,
+      "松": 72590,
+      "巴": 72591,
+      "左": 72592,
+      "封": 72593,
+      "圆": 72594,
+      "套": 72595,
+      "吸": 72596,
+      "刚": 72597,
+      "坚": 72598,
+      "须": 72599,
+      "笔": 72600,
+      "秀": 72601,
+      "肉": 72602,
+      "函": 72603,
+      "脑": 72604,
+      "馆": 72605,
+      "突": 72606,
+      "阿": 72607,
+      "贵": 72608,
+      "а": 72609,
+      "币": 72610,
+      "宣": 72611,
+      "右": 72612,
+      "块": 72613,
+      "练": 72614,
+      "晚": 72615,
+      "短": 72616,
+      "筑": 72617,
+      "厂": 72618,
+      "略": 72619,
+      "占": 72620,
+      "毒": 72621,
+      "宋": 72622,
+      "夜": 72623,
+      "播": 72624,
+      "坐": 72625,
+      "附": 72626,
+      "丽": 72627,
+      "背": 72628,
+      "妈": 72629,
+      "键": 72630,
+      "丰": 72631,
+      "遇": 72632,
+      "草": 72633,
+      "雨": 72634,
+      "虽": 72635,
+      "е": 72636,
+      "顺": 72637,
+      "谢": 72638,
+      "诗": 72639,
+      "激": 72640,
+      "季": 72641,
+      "概": 72642,
+      "朝": 72643,
+      "灵": 72644,
+      "警": 72645,
+      "典": 72646,
+      "询": 72647,
+      "贷": 72648,
+      "拿": 72649,
+      "针": 72650,
+      "�": 72651,
+      "梦": 72652,
+      "损": 72653,
+      "谁": 72654,
+      "洗": 72655,
+      "页": 72656,
+      "牛": 72657,
+      "似": 72658,
+      "荣": 72659,
+      "鱼": 72660,
+      "偿": 72661,
+      "退": 72662,
+      "田": 72663,
+      "停": 72664,
+      "冷": 72665,
+      "沙": 72666,
+      "丝": 72667,
+      "衣": 72668,
+      "笑": 72669,
+      "抗": 72670,
+      "吉": 72671,
+      "街": 72672,
+      "革": 72673,
+      "守": 72674,
+      "餐": 72675,
+      "库": 72676,
+      "追": 72677,
+      "毛": 72678,
+      "绿": 72679,
+      "缺": 72680,
+      "汽": 72681,
+      "稳": 72682,
+      "久": 72683,
+      "距": 72684,
+      "秋": 72685,
+      "吴": 72686,
+      "植": 72687,
+      "港": 72688,
+      "讨": 72689,
+      "刻": 72690,
+      "探": 72691,
+      "静": 72692,
+      "杂": 72693,
+      "и": 72694,
+      "延": 72695,
+      "座": 72696,
+      "亿": 72697,
+      "浙": 72698,
+      "液": 72699,
+      "酸": 72700,
+      "紧": 72701,
+      "宜": 72702,
+      "启": 72703,
+      "夏": 72704,
+      "亮": 72705,
+      "峰": 72706,
+      "穿": 72707,
+      "父": 72708,
+      "招": 72709,
+      "射": 72710,
+      "签": 72711,
+      "胜": 72712,
+      "童": 72713,
+      "拍": 72714,
+      "‐": 72715,
+      "顾": 72716,
+      "闻": 72717,
+      "藏": 72718,
+      "怀": 72719,
+      "佳": 72720,
+      "尚": 72721,
+      "á": 72722,
+      "固": 72723,
+      "堂": 72724,
+      "税": 72725,
+      "赵": 72726,
+      "钢": 72727,
+      "染": 72728,
+      "句": 72729,
+      "债": 72730,
+      "蛋": 72731,
+      "豆": 72732,
+      "障": 72733,
+      "雪": 72734,
+      "误": 72735,
+      "倍": 72736,
+      "症": 72737,
+      "授": 72738,
+      "危": 72739,
+      "径": 72740,
+      "骨": 72741,
+      "慢": 72742,
+      "‘": 72743,
+      "靠": 72744,
+      "航": 72745,
+      "斗": 72746,
+      "谈": 72747,
+      "址": 72748,
+      "欧": 72749,
+      "唐": 72750,
+      "冲": 72751,
+      "轮": 72752,
+      "届": 72753,
+      "陆": 72754,
+      "杀": 72755,
+      "熟": 72756,
+      "胡": 72757,
+      "烟": 72758,
+      "篇": 72759,
+      "拥": 72760,
+      "н": 72761,
+      "т": 72762,
+      "∴": 72763,
+      "尼": 72764,
+      "乎": 72765,
+      "孙": 72766,
+      "聚": 72767,
+      "散": 72768,
+      "徐": 72769,
+      "叫": 72770,
+      "礼": 72771,
+      "伟": 72772,
+      "鸡": 72773,
+      "架": 72774,
+      "苦": 72775,
+      "痛": 72776,
+      "」": 72777,
+      "卷": 72778,
+      "威": 72779,
+      "甚": 72780,
+      "「": 72781,
+      "茶": 72782,
+      "违": 72783,
+      "阅": 72784,
+      "乘": 72785,
+      "阶": 72786,
+      "岛": 72787,
+      "涉": 72788,
+      "顶": 72789,
+      "桥": 72790,
+      "硬": 72791,
+      "竞": 72792,
+      "°": 72793,
+      "洲": 72794,
+      "圈": 72795,
+      "鲜": 72796,
+      "籍": 72797,
+      "递": 72798,
+      "透": 72799,
+      "急": 72800,
+      "剂": 72801,
+      "驾": 72802,
+      "促": 72803,
+      "君": 72804,
+      "著": 72805,
+      "徒": 72806,
+      "派": 72807,
+      "兵": 72808,
+      "倒": 72809,
+      "屋": 72810,
+      "疫": 72811,
+      "献": 72812,
+      "梅": 72813,
+      "媒": 72814,
+      "雷": 72815,
+      "糖": 72816,
+      "饮": 72817,
+      "哥": 72818,
+      "辆": 72819,
+      "绩": 72820,
+      "饭": 72821,
+      "津": 72822,
+      "驶": 72823,
+      "汇": 72824,
+      "鉴": 72825,
+      "辑": 72826,
+      "席": 72827,
+      "努": 72828,
+      "宽": 72829,
+      "盖": 72830,
+      "ó": 72831,
+      "亦": 72832,
+      "韩": 72833,
+      "休": 72834,
+      "净": 72835,
+      "竟": 72836,
+      "订": 72837,
+      "午": 72838,
+      "册": 72839,
+      "攻": 72840,
+      "仍": 72841,
+      "晓": 72842,
+      "混": 72843,
+      "奥": 72844,
+      "折": 72845,
+      "冰": 72846,
+      "摄": 72847,
+      "野": 72848,
+      "颜": 72849,
+      "箱": 72850,
+      "舞": 72851,
+      "访": 72852,
+      "迎": 72853,
+      "饰": 72854,
+      "厅": 72855,
+      "虑": 72856,
+      "仪": 72857,
+      "救": 72858,
+      "详": 72859,
+      "缓": 72860,
+      "奶": 72861,
+      "翻": 72862,
+      "脱": 72863,
+      "荐": 72864,
+      "镜": 72865,
+      "露": 72866,
+      "纸": 72867,
+      "р": 72868,
+      "郑": 72869,
+      "雄": 72870,
+      "贸": 72871,
+      "扬": 72872,
+      "疑": 72873,
+      "朱": 72874,
+      "租": 72875,
+      "宫": 72876,
+      "盛": 72877,
+      "幸": 72878,
+      "冠": 72879,
+      "徽": 72880,
+      "怕": 72881,
+      "屏": 72882,
+      "牙": 72883,
+      "掉": 72884,
+      "督": 72885,
+      "灯": 72886,
+      "趣": 72887,
+      "池": 72888,
+      "侧": 72889,
+      "困": 72890,
+      "惠": 72891,
+      "乙": 72892,
+      "逐": 72893,
+      "储": 72894,
+      "拟": 72895,
+      "脸": 72896,
+      "鲁": 72897,
+      "跑": 72898,
+      "齐": 72899,
+      "轴": 72900,
+      "私": 72901,
+      "楚": 72902,
+      "虚": 72903,
+      "避": 72904,
+      "阴": 72905,
+      "措": 72906,
+      "旧": 72907,
+      "弹": 72908,
+      "纯": 72909,
+      "с": 72910,
+      "既": 72911,
+      "央": 72912,
+      "湾": 72913,
+      "脚": 72914,
+      "庄": 72915,
+      "陪": 72916,
+      "润": 72917,
+      "董": 72918,
+      "蒙": 72919,
+      "洋": 72920,
+      "肥": 72921,
+      "讯": 72922,
+      "宗": 72923,
+      "累": 72924,
+      "í": 72925,
+      "伴": 72926,
+      "泰": 72927,
+      "唱": 72928,
+      "触": 72929,
+      "蓝": 72930,
+      "佛": 72931,
+      "描": 72932,
+      "魔": 72933,
+      "幕": 72934,
+      "炎": 72935,
+      "扣": 72936,
+      "昌": 72937,
+      "欠": 72938,
+      "妻": 72939,
+      "恶": 72940,
+      "链": 72941,
+      "掌": 72942,
+      "遗": 72943,
+      "疾": 72944,
+      "姐": 72945,
+      "挥": 72946,
+      "仙": 72947,
+      "丁": 72948,
+      "麻": 72949,
+      "÷": 72950,
+      "贴": 72951,
+      "扩": 72952,
+      "肤": 72953,
+      "途": 72954,
+      "寻": 72955,
+      "寒": 72956,
+      "握": 72957,
+      "诊": 72958,
+      "狱": 72959,
+      "α": 72960,
+      "敏": 72961,
+      "π": 72962,
+      "械": 72963,
+      "雅": 72964,
+      "梁": 72965,
+      "押": 72966,
+      "浪": 72967,
+      "漫": 72968,
+      "勇": 72969,
+      "渐": 72970,
+      "盐": 72971,
+      "瑞": 72972,
+      "秘": 72973,
+      "末": 72974,
+      "皇": 72975,
+      "估": 72976,
+      "耳": 72977,
+      "肯": 72978,
+      "菌": 72979,
+      "束": 72980,
+      "抓": 72981,
+      "隐": 72982,
+      "帝": 72983,
+      "爆": 72984,
+      "厚": 72985,
+      "悔": 72986,
+      "坏": 72987,
+      "圣": 72988,
+      "胞": 72989,
+      "挑": 72990,
+      "泉": 72991,
+      "冻": 72992,
+      "胶": 72993,
+      "惊": 72994,
+      "矿": 72995,
+      "档": 72996,
+      "冬": 72997,
+      "偏": 72998,
+      "巨": 72999,
+      "旗": 73000,
+      "杰": 73001,
+      "烈": 73002,
+      "迷": 73003,
+      "塔": 73004,
+      "刷": 73005,
+      "弟": 73006,
+      "幼": 73007,
+      "乱": 73008,
+      "账": 73009,
+      "丹": 73010,
+      "莫": 73011,
+      "隔": 73012,
+      "锅": 73013,
+      "洁": 73014,
+      "洪": 73015,
+      "膜": 73016,
+      "睡": 73017,
+      "烧": 73018,
+      "映": 73019,
+      "缘": 73020,
+      "в": 73021,
+      "甘": 73022,
+      "缩": 73023,
+      "励": 73024,
+      "暖": 73025,
+      "弃": 73026,
+      "遍": 73027,
+      "诚": 73028,
+      "搭": 73029,
+      "汤": 73030,
+      "宇": 73031,
+      "暴": 73032,
+      "尤": 73033,
+      "ü": 73034,
+      "振": 73035,
+      "刺": 73036,
+      "滑": 73037,
+      "寸": 73038,
+      "ä": 73039,
+      "恢": 73040,
+      "搜": 73041,
+      "巧": 73042,
+      "沉": 73043,
+      "±": 73044,
+      "抽": 73045,
+      "盟": 73046,
+      "凤": 73047,
+      "爸": 73048,
+      "览": 73049,
+      "嘉": 73050,
+      "郭": 73051,
+      "尺": 73052,
+      "龄": 73053,
+      "～": 73054,
+      "耐": 73055,
+      "珠": 73056,
+      "沈": 73057,
+      "践": 73058,
+      "恩": 73059,
+      "窗": 73060,
+      "辉": 73061,
+      "л": 73062,
+      "尾": 73063,
+      "咨": 73064,
+      "败": 73065,
+      "船": 73066,
+      "裂": 73067,
+      "脉": 73068,
+      "舒": 73069,
+      "森": 73070,
+      "晋": 73071,
+      "谷": 73072,
+      "恋": 73073,
+      "禁": 73074,
+      "添": 73075,
+      "偶": 73076,
+      "贝": 73077,
+      "圳": 73078,
+      "乌": 73079,
+      "邮": 73080,
+      "洛": 73081,
+      "亡": 73082,
+      "湿": 73083,
+      "污": 73084,
+      "伊": 73085,
+      "毫": 73086,
+      "幅": 73087,
+      "衡": 73088,
+      "暂": 73089,
+      "醒": 73090,
+      "残": 73091,
+      "懂": 73092,
+      "忘": 73093,
+      "跳": 73094,
+      "诺": 73095,
+      "ö": 73096,
+      "返": 73097,
+      "孔": 73098,
+      "焦": 73099,
+      "沟": 73100,
+      "默": 73101,
+      "辩": 73102,
+      "妇": 73103,
+      "杯": 73104,
+      "插": 73105,
+      "宿": 73106,
+      "横": 73107,
+      "弱": 73108,
+      "狗": 73109,
+      "溪": 73110,
+      "泽": 73111,
+      "辅": 73112,
+      "怪": 73113,
+      "岩": 73114,
+      "刀": 73115,
+      "缴": 73116,
+      "译": 73117,
+      "拼": 73118,
+      "忆": 73119,
+      "欲": 73120,
+      "辖": 73121,
+      "呈": 73122,
+      "岗": 73123,
+      "妹": 73124,
+      "暗": 73125,
+      "摩": 73126,
+      "塞": 73127,
+      "截": 73128,
+      "敢": 73129,
+      "喝": 73130,
+      "阵": 73131,
+      "挂": 73132,
+      "祖": 73133,
+      "仔": 73134,
+      "к": 73135,
+      "忠": 73136,
+      "紫": 73137,
+      "唯": 73138,
+      "剑": 73139,
+      "塑": 73140,
+      "伦": 73141,
+      "贫": 73142,
+      "猫": 73143,
+      "泛": 73144,
+      "肌": 73145,
+      "闭": 73146,
+      "呼": 73147,
+      "脂": 73148,
+      "淡": 73149,
+      "潮": 73150,
+      "繁": 73151,
+      "磨": 73152,
+      "恒": 73153,
+      "泡": 73154,
+      "迟": 73155,
+      "浓": 73156,
+      "舍": 73157,
+      "仁": 73158,
+      "猪": 73159,
+      "拒": 73160,
+      "震": 73161,
+      "芳": 73162,
+      "尘": 73163,
+      "氧": 73164,
+      "厘": 73165,
+      "闲": 73166,
+      "麦": 73167,
+      "谓": 73168,
+      "尊": 73169,
+      "抚": 73170,
+      "墙": 73171,
+      "殊": 73172,
+      "盗": 73173,
+      "伙": 73174,
+      "鼓": 73175,
+      "寿": 73176,
+      "兼": 73177,
+      "阻": 73178,
+      "涨": 73179,
+      "竹": 73180,
+      "亩": 73181,
+      "珍": 73182,
+      "杭": 73183,
+      "宏": 73184,
+      "虫": 73185,
+      "幻": 73186,
+      "挺": 73187,
+      "桃": 73188,
+      "搞": 73189,
+      "填": 73190,
+      "桂": 73191,
+      "遵": 73192,
+      "凭": 73193,
+      "赞": 73194,
+      "慧": 73195,
+      "赔": 73196,
+      "凝": 73197,
+      "惯": 73198,
+      "券": 73199,
+      "誉": 73200,
+      "苗": 73201,
+      "迪": 73202,
+      "聊": 73203,
+      "凡": 73204,
+      "辽": 73205,
+      "锁": 73206,
+      "侵": 73207,
+      "抵": 73208,
+      "隆": 73209,
+      "燕": 73210,
+      "陕": 73211,
+      "虎": 73212,
+      "啥": 73213,
+      "凉": 73214,
+      "肠": 73215,
+      "晶": 73216,
+      "刊": 73217,
+      "粒": 73218,
+      "杜": 73219,
+      "贯": 73220,
+      "迹": 73221,
+      "宾": 73222,
+      "秦": 73223,
+      "豪": 73224,
+      "乳": 73225,
+      "循": 73226,
+      "废": 73227,
+      "洞": 73228,
+      "柱": 73229,
+      "诸": 73230,
+      "祥": 73231,
+      "锦": 73232,
+      "羊": 73233,
+      "彻": 73234,
+      "鼠": 73235,
+      "召": 73236,
+      "≤": 73237,
+      "抱": 73238,
+      "薄": 73239,
+      "壁": 73240,
+      "顿": 73241,
+      "戴": 73242,
+      "腾": 73243,
+      "岸": 73244,
+      "泥": 73245,
+      "甜": 73246,
+      "浮": 73247,
+      "腹": 73248,
+      "铺": 73249,
+      "谱": 73250,
+      "仓": 73251,
+      "仲": 73252,
+      "剩": 73253,
+      "潜": 73254,
+      "抢": 73255,
+      "艳": 73256,
+      "锋": 73257,
+      "趋": 73258,
+      "莱": 73259,
+      "尿": 73260,
+      "瓜": 73261,
+      "涂": 73262,
+      "д": 73263,
+      "勤": 73264,
+      "柔": 73265,
+      "绕": 73266,
+      "沿": 73267,
+      "伯": 73268,
+      "腐": 73269,
+      "催": 73270,
+      "炒": 73271,
+      "ç": 73272,
+      "柳": 73273,
+      "啦": 73274,
+      "墨": 73275,
+      "唤": 73276,
+      "煤": 73277,
+      "赢": 73278,
+      "の": 73279,
+      "琴": 73280,
+      "俄": 73281,
+      "扫": 73282,
+      "吹": 73283,
+      "耗": 73284,
+      "替": 73285,
+      "姜": 73286,
+      "枝": 73287,
+      "赏": 73288,
+      "朗": 73289,
+      "燃": 73290,
+      "娱": 73291,
+      "井": 73292,
+      "伍": 73293,
+      "夺": 73294,
+      "灰": 73295,
+      "俗": 73296,
+      "旋": 73297,
+      "萨": 73298,
+      "遭": 73299,
+      "扶": 73300,
+      "鸟": 73301,
+      "寄": 73302,
+      "肝": 73303,
+      "滨": 73304,
+      "纹": 73305,
+      "壮": 73306,
+      "氏": 73307,
+      "−": 73308,
+      "爷": 73309,
+      "祝": 73310,
+      "м": 73311,
+      "∵": 73312,
+      "姓": 73313,
+      "灾": 73314,
+      "肿": 73315,
+      "辛": 73316,
+      "灭": 73317,
+      "瓦": 73318,
+      "驱": 73319,
+      "醉": 73320,
+      "敬": 73321,
+      "俊": 73322,
+      "凯": 73323,
+      "瓣": 73324,
+      "恐": 73325,
+      "悉": 73326,
+      "曹": 73327,
+      "孤": 73328,
+      "п": 73329,
+      "陵": 73330,
+      "署": 73331,
+      "碎": 73332,
+      "癌": 73333,
+      "炼": 73334,
+      "捷": 73335,
+      "秒": 73336,
+      "β": 73337,
+      "荷": 73338,
+      "敌": 73339,
+      "跨": 73340,
+      "烦": 73341,
+      "喷": 73342,
+      "胃": 73343,
+      "狂": 73344,
+      "迁": 73345,
+      "腰": 73346,
+      "倾": 73347,
+      "绪": 73348,
+      "贡": 73349,
+      "陶": 73350,
+      "奋": 73351,
+      "驳": 73352,
+      "绘": 73353,
+      "澳": 73354,
+      "魂": 73355,
+      "忙": 73356,
+      "坛": 73357,
+      "迅": 73358,
+      "邀": 73359,
+      "覆": 73360,
+      "厦": 73361,
+      "孕": 73362,
+      "忽": 73363,
+      "瓶": 73364,
+      "浦": 73365,
+      "拓": 73366,
+      "骗": 73367,
+      "嫌": 73368,
+      "硕": 73369,
+      "晨": 73370,
+      "肺": 73371,
+      "袋": 73372,
+      "颗": 73373,
+      "殖": 73374,
+      "垂": 73375,
+      "腿": 73376,
+      "莲": 73377,
+      "赖": 73378,
+      "芯": 73379,
+      "鬼": 73380,
+      "纵": 73381,
+      "赋": 73382,
+      "铜": 73383,
+      "携": 73384,
+      "窃": 73385,
+      "у": 73386,
+      "桌": 73387,
+      "扎": 73388,
+      "娘": 73389,
+      "ã": 73390,
+      "鞋": 73391,
+      "赶": 73392,
+      "霞": 73393,
+      "嘴": 73394,
+      "韵": 73395,
+      "吨": 73396,
+      "婆": 73397,
+      "拔": 73398,
+      "坡": 73399,
+      "欣": 73400,
+      "勒": 73401,
+      "仿": 73402,
+      "摆": 73403,
+      "肖": 73404,
+      "邻": 73405,
+      "涛": 73406,
+      "辞": 73407,
+      "忍": 73408,
+      "纤": 73409,
+      "脏": 73410,
+      "串": 73411,
+      "粮": 73412,
+      "坦": 73413,
+      "驻": 73414,
+      "逃": 73415,
+      "尖": 73416,
+      "宅": 73417,
+      "胸": 73418,
+      "炸": 73419,
+      "堆": 73420,
+      "旁": 73421,
+      "逆": 73422,
+      "哦": 73423,
+      "胎": 73424,
+      "曼": 73425,
+      "悲": 73426,
+      "凌": 73427,
+      "è": 73428,
+      "赚": 73429,
+      "跌": 73430,
+      "姆": 73431,
+      "栏": 73432,
+      "→": 73433,
+      "艾": 73434,
+      "犹": 73435,
+      "疼": 73436,
+      "尝": 73437,
+      "枪": 73438,
+      "苹": 73439,
+      "貌": 73440,
+      "陷": 73441,
+      "盾": 73442,
+      "à": 73443,
+      "寺": 73444,
+      "辨": 73445,
+      "浩": 73446,
+      "拖": 73447,
+      "妙": 73448,
+      "魏": 73449,
+      "泪": 73450,
+      "旦": 73451,
+      "姻": 73452,
+      "炉": 73453,
+      "湘": 73454,
+      "斜": 73455,
+      "鼻": 73456,
+      "穷": 73457,
+      "肃": 73458,
+      "拘": 73459,
+      "漏": 73460,
+      "夹": 73461,
+      "滴": 73462,
+      "筹": 73463,
+      "宠": 73464,
+      "浴": 73465,
+      "磁": 73466,
+      "妆": 73467,
+      "援": 73468,
+      "£": 73469,
+      "∈": 73470,
+      "胆": 73471,
+      "昆": 73472,
+      "抑": 73473,
+      "拆": 73474,
+      "椒": 73475,
+      "岭": 73476,
+      "辈": 73477,
+      "跃": 73478,
+      "璃": 73479,
+      "ı": 73480,
+      "拜": 73481,
+      "辣": 73482,
+      "腔": 73483,
+      "轨": 73484,
+      "惜": 73485,
+      "俱": 73486,
+      "兄": 73487,
+      "溶": 73488,
+      "浅": 73489,
+      "玻": 73490,
+      "割": 73491,
+      "齿": 73492,
+      "骑": 73493,
+      "卢": 73494,
+      "役": 73495,
+      "哲": 73496,
+      "畅": 73497,
+      "羽": 73498,
+      "蔡": 73499,
+      "框": 73500,
+      "幽": 73501,
+      "谋": 73502,
+      "奉": 73503,
+      "黎": 73504,
+      "耀": 73505,
+      "奏": 73506,
+      "玲": 73507,
+      "稿": 73508,
+      "聘": 73509,
+      "疏": 73510,
+      "冒": 73511,
+      "疆": 73512,
+      "抛": 73513,
+      "摸": 73514,
+      "瘤": 73515,
+      "梯": 73516,
+      "闪": 73517,
+      "迫": 73518,
+      "渠": 73519,
+      "逼": 73520,
+      "贤": 73521,
+      "菲": 73522,
+      "剪": 73523,
+      "杆": 73524,
+      "鹏": 73525,
+      "鸣": 73526,
+      "壳": 73527,
+      "伸": 73528,
+      "滤": 73529,
+      "涵": 73530,
+      "允": 73531,
+      "①": 73532,
+      "蒸": 73533,
+      "乏": 73534,
+      "逾": 73535,
+      "纲": 73536,
+      "卓": 73537,
+      "睛": 73538,
+      "玄": 73539,
+      "邦": 73540,
+      "吐": 73541,
+      "骤": 73542,
+      "粗": 73543,
+      "熊": 73544,
+      "亭": 73545,
+      "匀": 73546,
+      "删": 73547,
+      "煮": 73548,
+      "鸿": 73549,
+      "碳": 73550,
+      "彭": 73551,
+      "哭": 73552,
+      "昨": 73553,
+      "烤": 73554,
+      "兽": 73555,
+      "μ": 73556,
+      "摇": 73557,
+      "偷": 73558,
+      "蜜": 73559,
+      "斤": 73560,
+      "郁": 73561,
+      "盈": 73562,
+      "番": 73563,
+      "岳": 73564,
+      "赠": 73565,
+      "ا": 73566,
+      "я": 73567,
+      "②": 73568,
+      "愈": 73569,
+      "捕": 73570,
+      "舟": 73571,
+      "翠": 73572,
+      "赤": 73573,
+      "肾": 73574,
+      "谐": 73575,
+      "稍": 73576,
+      "孟": 73577,
+      "碰": 73578,
+      "弄": 73579,
+      "乔": 73580,
+      "娜": 73581,
+      "盒": 73582,
+      "锡": 73583,
+      "呀": 73584,
+      "碧": 73585,
+      "佩": 73586,
+      "邓": 73587,
+      "淮": 73588,
+      "吕": 73589,
+      "扰": 73590,
+      "遥": 73591,
+      "滚": 73592,
+      "斑": 73593,
+      "郎": 73594,
+      "伏": 73595,
+      "冯": 73596,
+      "侯": 73597,
+      "荒": 73598,
+      "漂": 73599,
+      "耕": 73600,
+      "帅": 73601,
+      "槽": 73602,
+      "厨": 73603,
+      "丛": 73604,
+      "′": 73605,
+      "粘": 73606,
+      "旨": 73607,
+      "®": 73608,
+      "皆": 73609,
+      "螺": 73610,
+      "篮": 73611,
+      "薪": 73612,
+      "柜": 73613,
+      "碍": 73614,
+      "爽": 73615,
+      "酱": 73616,
+      "苍": 73617,
+      "撑": 73618,
+      "霸": 73619,
+      "囊": 73620,
+      "腺": 73621,
+      "眠": 73622,
+      "株": 73623,
+      "御": 73624,
+      "崇": 73625,
+      "灌": 73626,
+      "豫": 73627,
+      "踪": 73628,
+      "矛": 73629,
+      "巡": 73630,
+      "彼": 73631,
+      "乃": 73632,
+      "忧": 73633,
+      "斌": 73634,
+      "潘": 73635,
+      "弦": 73636,
+      "芝": 73637,
+      "卧": 73638,
+      "闹": 73639,
+      "牧": 73640,
+      "℃": 73641,
+      "袁": 73642,
+      "瓷": 73643,
+      "劲": 73644,
+      "奔": 73645,
+      "ー": 73646,
+      "渡": 73647,
+      "霜": 73648,
+      "氛": 73649,
+      "榜": 73650,
+      "寓": 73651,
+      "臣": 73652,
+      "衰": 73653,
+      "逻": 73654,
+      "淋": 73655,
+      "撞": 73656,
+      "з": 73657,
+      "桑": 73658,
+      "埃": 73659,
+      "鑫": 73660,
+      "醇": 73661,
+      "稀": 73662,
+      "砂": 73663,
+      "咖": 73664,
+      "恨": 73665,
+      "萧": 73666,
+      "沒": 73667,
+      "汁": 73668,
+      "烂": 73669,
+      "浏": 73670,
+      "雕": 73671,
+      "擦": 73672,
+      "挖": 73673,
+      "棒": 73674,
+      "骂": 73675,
+      "婴": 73676,
+      "贺": 73677,
+      "剥": 73678,
+      "砖": 73679,
+      "踏": 73680,
+      "矩": 73681,
+      "姑": 73682,
+      "坑": 73683,
+      "玛": 73684,
+      "滋": 73685,
+      "酷": 73686,
+      "吾": 73687,
+      "い": 73688,
+      "姚": 73689,
+      "悬": 73690,
+      "垃": 73691,
+      "雾": 73692,
+      "丙": 73693,
+      "圾": 73694,
+      "葡": 73695,
+      "钻": 73696,
+      "悠": 73697,
+      "侠": 73698,
+      "奈": 73699,
+      "廉": 73700,
+      "愁": 73701,
+      "─": 73702,
+      "羁": 73703,
+      "袭": 73704,
+      "ы": 73705,
+      "贾": 73706,
+      "莉": 73707,
+      "淘": 73708,
+      "嘛": 73709,
+      "堡": 73710,
+      "扮": 73711,
+      "葱": 73712,
+      "枚": 73713,
+      "摘": 73714,
+      "拨": 73715,
+      "匹": 73716,
+      "ь": 73717,
+      "缝": 73718,
+      "碑": 73719,
+      "蒋": 73720,
+      "ā": 73721,
+      "恰": 73722,
+      "眉": 73723,
+      "涯": 73724,
+      "に": 73725,
+      "≥": 73726,
+      "颈": 73727,
+      "拌": 73728,
+      "〉": 73729,
+      "塘": 73730,
+      "å": 73731,
+      "叔": 73732,
+      "毁": 73733,
+      "坊": 73734,
+      "夕": 73735,
+      "旺": 73736,
+      "飘": 73737,
+      "饱": 73738,
+      "芬": 73739,
+      "赁": 73740,
+      "诞": 73741,
+      "〈": 73742,
+      "萍": 73743,
+      "柏": 73744,
+      "饼": 73745,
+      "荡": 73746,
+      "汗": 73747,
+      "瘦": 73748,
+      "汪": 73749,
+      "栋": 73750,
+      "萝": 73751,
+      "る": 73752,
+      "蔬": 73753,
+      "б": 73754,
+      "棋": 73755,
+      "棉": 73756,
+      "悦": 73757,
+      "姿": 73758,
+      "浆": 73759,
+      "揭": 73760,
+      "朵": 73761,
+      "牵": 73762,
+      "丘": 73763,
+      "牢": 73764,
+      "を": 73765,
+      "厉": 73766,
+      "孝": 73767,
+      "卜": 73768,
+      "肩": 73769,
+      "丢": 73770,
+      "逢": 73771,
+      "锐": 73772,
+      "阁": 73773,
+      "鄂": 73774,
+      "萄": 73775,
+      "翼": 73776,
+      "诱": 73777,
+      "鹤": 73778,
+      "弯": 73779,
+      "し": 73780,
+      "堪": 73781,
+      "慎": 73782,
+      "∞": 73783,
+      "厌": 73784,
+      "搬": 73785,
+      "»": 73786,
+      "翔": 73787,
+      "θ": 73788,
+      "卵": 73789,
+      "盆": 73790,
+      "虾": 73791,
+      "颖": 73792,
+      "帧": 73793,
+      "阔": 73794,
+      "菊": 73795,
+      "悟": 73796,
+      "舰": 73797,
+      "傅": 73798,
+      "霍": 73799,
+      "慈": 73800,
+      "戒": 73801,
+      "拳": 73802,
+      "叠": 73803,
+      "茂": 73804,
+      "仰": 73805,
+      "糊": 73806,
+      "辰": 73807,
+      "怒": 73808,
+      "娃": 73809,
+      "淀": 73810,
+      "蒂": 73811,
+      "阀": 73812,
+      "啡": 73813,
+      "炮": 73814,
+      "殿": 73815,
+      "箭": 73816,
+      "鹿": 73817,
+      "稻": 73818,
+      "沃": 73819,
+      "虹": 73820,
+      "叉": 73821,
+      "兔": 73822,
+      "瞬": 73823,
+      "狼": 73824,
+      "扇": 73825,
+      "聪": 73826,
+      "た": 73827,
+      "晴": 73828,
+      "浸": 73829,
+      "鼎": 73830,
+      "壤": 73831,
+      "赫": 73832,
+      "廷": 73833,
+      "Ｘ": 73834,
+      "ú": 73835,
+      "妖": 73836,
+      "佐": 73837,
+      "绵": 73838,
+      "屈": 73839,
+      "糕": 73840,
+      "蓄": 73841,
+      "碗": 73842,
+      "崔": 73843,
+      "铝": 73844,
+      "亏": 73845,
+      "迈": 73846,
+      "庙": 73847,
+      "乾": 73848,
+      "墓": 73849,
+      "纽": 73850,
+      "韦": 73851,
+      "魅": 73852,
+      "吊": 73853,
+      "燥": 73854,
+      "泵": 73855,
+      "蜂": 73856,
+      "疯": 73857,
+      "渔": 73858,
+      "慕": 73859,
+      "葛": 73860,
+      "萌": 73861,
+      "﹣": 73862,
+      "柴": 73863,
+      "伪": 73864,
+      "赴": 73865,
+      "胖": 73866,
+      "猜": 73867,
+      "丈": 73868,
+      "酬": 73869,
+      "遂": 73870,
+      "怨": 73871,
+      "粤": 73872,
+      "袖": 73873,
+      "г": 73874,
+      "娟": 73875,
+      "坤": 73876,
+      "弘": 73877,
+      "と": 73878,
+      "痕": 73879,
+      "锻": 73880,
+      "晰": 73881,
+      "鸭": 73882,
+      "漆": 73883,
+      "晒": 73884,
+      "筋": 73885,
+      "纺": 73886,
+      "て": 73887,
+      "焊": 73888,
+      "挡": 73889,
+      "轩": 73890,
+      "й": 73891,
+      "栽": 73892,
+      "阐": 73893,
+      "で": 73894,
+      "帐": 73895,
+      "氨": 73896,
+      "馨": 73897,
+      "掘": 73898,
+      "垫": 73899,
+      "酶": 73900,
+      "ł": 73901,
+      "弗": 73902,
+      "咸": 73903,
+      "怜": 73904,
+      "叹": 73905,
+      "脾": 73906,
+      "③": 73907,
+      "逝": 73908,
+      "勾": 73909,
+      "裤": 73910,
+      "惑": 73911,
+      "柯": 73912,
+      "披": 73913,
+      "旭": 73914,
+      "©": 73915,
+      "呵": 73916,
+      "ل": 73917,
+      "â": 73918,
+      "谨": 73919,
+      "捐": 73920,
+      "泳": 73921,
+      "契": 73922,
+      "邪": 73923,
+      "俩": 73924,
+      "艰": 73925,
+      "翁": 73926,
+      "は": 73927,
+      "诈": 73928,
+      "罢": 73929,
+      "慰": 73930,
+      "撒": 73931,
+      "仇": 73932,
+      "ч": 73933,
+      "赌": 73934,
+      "惩": 73935,
+      "ン": 73936,
+      "滞": 73937,
+      "锈": 73938,
+      "宴": 73939,
+      "傻": 73940,
+      "筛": 73941,
+      "炭": 73942,
+      "劫": 73943,
+      "蛇": 73944,
+      "峡": 73945,
+      "な": 73946,
+      "ñ": 73947,
+      "琳": 73948,
+      "肪": 73949,
+      "吵": 73950,
+      "苑": 73951,
+      "桶": 73952,
+      "馈": 73953,
+      "兹": 73954,
+      "傲": 73955,
+      "彦": 73956,
+      "渗": 73957,
+      "薛": 73958,
+      "颁": 73959,
+      "谅": 73960,
+      "铭": 73961,
+      "渝": 73962,
+      "寂": 73963,
+      "抹": 73964,
+      "腻": 73965,
+      "琼": 73966,
+      "㈰": 73967,
+      "蚀": 73968,
+      "噪": 73969,
+      "胀": 73970,
+      "脆": 73971,
+      "饲": 73972,
+      "敦": 73973,
+      "吟": 73974,
+      "欺": 73975,
+      "挤": 73976,
+      "ø": 73977,
+      "磊": 73978,
+      "す": 73979,
+      "毅": 73980,
+      "逸": 73981,
+      "兆": 73982,
+      "帕": 73983,
+      "牲": 73984,
+      "咱": 73985,
+      "雁": 73986,
+      "爬": 73987,
+      "滩": 73988,
+      "∠": 73989,
+      "罩": 73990,
+      "―": 73991,
+      "尸": 73992,
+      "ć": 73993,
+      "š": 73994,
+      "懒": 73995,
+      "猎": 73996,
+      "卸": 73997,
+      "嫩": 73998,
+      "弥": 73999,
+      "●": 74000,
+      "陌": 74001,
+      "唿": 74002,
+      "穴": 74003,
+      "♪": 74004,
+      "が": 74005,
+      "丧": 74006,
+      "嫁": 74007,
+      "彰": 74008,
+      "铸": 74009,
+      "朴": 74010,
+      "惨": 74011,
+      "蝶": 74012,
+      "狐": 74013,
+      "忌": 74014,
+      "劣": 74015,
+      "擅": 74016,
+      "帽": 74017,
+      "搅": 74018,
+      "蓬": 74019,
+      "菇": 74020,
+      "丑": 74021,
+      "闽": 74022,
+      "丸": 74023,
+      "茫": 74024,
+      "娇": 74025,
+      "暑": 74026,
+      "冀": 74027,
+      "喊": 74028,
+      "椅": 74029,
+      "泄": 74030,
+      "潭": 74031,
+      "梨": 74032,
+      "埋": 74033,
+      "硅": 74034,
+      "邵": 74035,
+      "沧": 74036,
+      "蹈": 74037,
+      "伐": 74038,
+      "ă": 74039,
+      "寨": 74040,
+      "別": 74041,
+      "劝": 74042,
+      "이": 74043,
+      "耶": 74044,
+      "裕": 74045,
+      "渣": 74046,
+      "ê": 74047,
+      "谭": 74048,
+      "叙": 74049,
+      "宙": 74050,
+      "昂": 74051,
+      "靖": 74052,
+      "蒜": 74053,
+      "膏": 74054,
+      "疲": 74055,
+      "巾": 74056,
+      "婷": 74057,
+      "驰": 74058,
+      "坪": 74059,
+      "泌": 74060,
+      "芦": 74061,
+      "λ": 74062,
+      "纱": 74063,
+      "芽": 74064,
+      "彬": 74065,
+      "漠": 74066,
+      "扑": 74067,
+      "梗": 74068,
+      "翅": 74069,
+      "茅": 74070,
+      "瑶": 74071,
+      "庞": 74072,
+      "宪": 74073,
+      "暮": 74074,
+      "碱": 74075,
+      "岂": 74076,
+      "洒": 74077,
+      "贩": 74078,
+      "盲": 74079,
+      "淑": 74080,
+      "枣": 74081,
+      "筒": 74082,
+      "夸": 74083,
+      "扭": 74084,
+      "堵": 74085,
+      "抖": 74086,
+      "攀": 74087,
+      "坝": 74088,
+      "牺": 74089,
+      "桐": 74090,
+      "妥": 74091,
+      "倡": 74092,
+      "颇": 74093,
+      "硫": 74094,
+      "蛮": 74095,
+      "胁": 74096,
+      "摊": 74097,
+      "畜": 74098,
+      "铃": 74099,
+      "猛": 74100,
+      "臭": 74101,
+      "窝": 74102,
+      "馀": 74103,
+      "臂": 74104,
+      "č": 74105,
+      "帆": 74106,
+      "腊": 74107,
+      "渊": 74108,
+      "吞": 74109,
+      "熙": 74110,
+      "涌": 74111,
+      "卿": 74112,
+      "舌": 74113,
+      "穆": 74114,
+      "霉": 74115,
+      "肚": 74116,
+      "荆": 74117,
+      "莎": 74118,
+      "★": 74119,
+      "妨": 74120,
+      "钙": 74121,
+      "削": 74122,
+      "磷": 74123,
+      "贪": 74124,
+      "戈": 74125,
+      "擎": 74126,
+      "肢": 74127,
+      "帖": 74128,
+      "昏": 74129,
+      "狮": 74130,
+      "煎": 74131,
+      "侦": 74132,
+      "翰": 74133,
+      "崩": 74134,
+      "尹": 74135,
+      "掩": 74136,
+      "钓": 74137,
+      "蓉": 74138,
+      "橡": 74139,
+      "拾": 74140,
+      "椭": 74141,
+      "藤": 74142,
+      "ş": 74143,
+      "僧": 74144,
+      "巢": 74145,
+      "€": 74146,
+      "邹": 74147,
+      "枯": 74148,
+      "苯": 74149,
+      "沪": 74150,
+      "犬": 74151,
+      "昭": 74152,
+      "谊": 74153,
+      "泊": 74154,
+      "罐": 74155,
+      "郊": 74156,
+      "笼": 74157,
+      "琐": 74158,
+      "裙": 74159,
+      "儒": 74160,
+      "绑": 74161,
+      "呆": 74162,
+      "剖": 74163,
+      "绳": 74164,
+      "莞": 74165,
+      "椎": 74166,
+      "匙": 74167,
+      "廊": 74168,
+      "昔": 74169,
+      "辐": 74170,
+      "○": 74171,
+      "壶": 74172,
+      "衍": 74173,
+      "・": 74174,
+      "禅": 74175,
+      "抄": 74176,
+      "咬": 74177,
+      "巷": 74178,
+      "胺": 74179,
+      "兑": 74180,
+      "狠": 74181,
+      "™": 74182,
+      "芒": 74183,
+      "侣": 74184,
+      "唇": 74185,
+      "隶": 74186,
+      "赣": 74187,
+      "缐": 74188,
+      "栖": 74189,
+      "沂": 74190,
+      "ō": 74191,
+      "勘": 74192,
+      "咏": 74193,
+      "帘": 74194,
+      "妃": 74195,
+      "曝": 74196,
+      "△": 74197,
+      "挽": 74198,
+      "煌": 74199,
+      "邱": 74200,
+      "逊": 74201,
+      "哀": 74202,
+      "轿": 74203,
+      "抬": 74204,
+      "鹅": 74205,
+      "喂": 74206,
+      "皖": 74207,
+      "杏": 74208,
+      "歉": 74209,
+      "溢": 74210,
+      "樱": 74211,
+      "挣": 74212,
+      "秉": 74213,
+      "蟹": 74214,
+      "兮": 74215,
+      "敲": 74216,
+      "墅": 74217,
+      "喻": 74218,
+      "烯": 74219,
+      "曰": 74220,
+      "癫": 74221,
+      "鹰": 74222,
+      "遮": 74223,
+      "凶": 74224,
+      "다": 74225,
+      "暨": 74226,
+      "茄": 74227,
+      "秩": 74228,
+      "棵": 74229,
+      "恼": 74230,
+      "酵": 74231,
+      "幂": 74232,
+      "勿": 74233,
+      "缸": 74234,
+      "浑": 74235,
+      "榆": 74236,
+      "熬": 74237,
+      "薇": 74238,
+      "م": 74239,
+      "悄": 74240,
+      "旬": 74241,
+      "凰": 74242,
+      "舆": 74243,
+      "褐": 74244,
+      "冶": 74245,
+      "穗": 74246,
+      "④": 74247,
+      "勃": 74248,
+      "ま": 74249,
+      "撰": 74250,
+      "ω": 74251,
+      "缆": 74252,
+      "渴": 74253,
+      "ρ": 74254,
+      "ス": 74255,
+      "脊": 74256,
+      "猴": 74257,
+      "肇": 74258,
+      "惟": 74259,
+      "咳": 74260,
+      "祭": 74261,
+      "惧": 74262,
+      "溃": 74263,
+      "氯": 74264,
+      "绣": 74265,
+      "酌": 74266,
+      "氢": 74267,
+      "怡": 74268,
+      "锥": 74269,
+      "咽": 74270,
+      "汝": 74271,
+      "衔": 74272,
+      "郡": 74273,
+      "枕": 74274,
+      "痫": 74275,
+      "棱": 74276,
+      "巩": 74277,
+      "矫": 74278,
+      "醋": 74279,
+      "ト": 74280,
+      "廖": 74281,
+      "钮": 74282,
+      "祸": 74283,
+      "逮": 74284,
+      "吓": 74285,
+      "缠": 74286,
+      "贞": 74287,
+      "饶": 74288,
+      "隙": 74289,
+      "γ": 74290,
+      "枫": 74291,
+      "晕": 74292,
+      "妮": 74293,
+      "扁": 74294,
+      "冈": 74295,
+      "奴": 74296,
+      "か": 74297,
+      "ж": 74298,
+      "襄": 74299,
+      "²": 74300,
+      "趁": 74301,
+      "⋅": 74302,
+      "怖": 74303,
+      "撕": 74304,
+      "殷": 74305,
+      "戚": 74306,
+      "葬": 74307,
+      "贼": 74308,
+      "酯": 74309,
+      "ε": 74310,
+      "ر": 74311,
+      "ن": 74312,
+      "幢": 74313,
+      "绒": 74314,
+      "凑": 74315,
+      "踩": 74316,
+      "薯": 74317,
+      "吻": 74318,
+      "愉": 74319,
+      "瑜": 74320,
+      "棚": 74321,
+      "阜": 74322,
+      "歧": 74323,
+      "芜": 74324,
+      "凸": 74325,
+      "詹": 74326,
+      "羡": 74327,
+      "斋": 74328,
+      "痴": 74329,
+      "雀": 74330,
+      "沫": 74331,
+      "粥": 74332,
+      "躺": 74333,
+      "麟": 74334,
+      "瑟": 74335,
+      "辱": 74336,
+      "衬": 74337,
+      "捞": 74338,
+      "れ": 74339,
+      "淄": 74340,
+      "姨": 74341,
+      "茎": 74342,
+      "闷": 74343,
+      "ي": 74344,
+      "黏": 74345,
+      "柄": 74346,
+      "遣": 74347,
+      "«": 74348,
+      "↓": 74349,
+      "️": 74350,
+      "躲": 74351,
+      "搏": 74352,
+      "髓": 74353,
+      "夷": 74354,
+      "捧": 74355,
+      "庐": 74356,
+      "蕾": 74357,
+      "泻": 74358,
+      "扯": 74359,
+      "钩": 74360,
+      "钥": 74361,
+      "钦": 74362,
+      "蒲": 74363,
+      "缀": 74364,
+      "窄": 74365,
+      "х": 74366,
+      "嵌": 74367,
+      "τ": 74368,
+      "і": 74369,
+      "仕": 74370,
+      "莹": 74371,
+      "冥": 74372,
+      "澄": 74373,
+      "钉": 74374,
+      "尬": 74375,
+      "イ": 74376,
+      "珊": 74377,
+      "倚": 74378,
+      "刮": 74379,
+      "氮": 74380,
+      "愤": 74381,
+      "畏": 74382,
+      "侨": 74383,
+      "□": 74384,
+      "瑰": 74385,
+      "咪": 74386,
+      "و": 74387,
+      "屁": 74388,
+      "闯": 74389,
+      "奎": 74390,
+      "酿": 74391,
+      "龟": 74392,
+      "蕴": 74393,
+      "沸": 74394,
+      "琪": 74395,
+      "蜀": 74396,
+      "募": 74397,
+      "潇": 74398,
+      "睿": 74399,
+      "勋": 74400,
+      "膝": 74401,
+      "颠": 74402,
+      "ô": 74403,
+      "禽": 74404,
+      "ц": 74405,
+      "轰": 74406,
+      "狭": 74407,
+      "沾": 74408,
+      "σ": 74409,
+      "屯": 74410,
+      "甸": 74411,
+      "愧": 74412,
+      "敷": 74413,
+      "梳": 74414,
+      "厕": 74415,
+      "痘": 74416,
+      "喉": 74417,
+      "辟": 74418,
+      "ο": 74419,
+      "谦": 74420,
+      "锤": 74421,
+      "ī": 74422,
+      "烫": 74423,
+      "膀": 74424,
+      "玫": 74425,
+      "盼": 74426,
+      "湛": 74427,
+      "恭": 74428,
+      "羞": 74429,
+      "旱": 74430,
+      "竭": 74431,
+      "灿": 74432,
+      "庸": 74433,
+      "纬": 74434,
+      "菱": 74435,
+      "É": 74436,
+      "鲍": 74437,
+      "邢": 74438,
+      "侍": 74439,
+      "衷": 74440,
+      "卑": 74441,
+      "浇": 74442,
+      "颂": 74443,
+      "顷": 74444,
+      "宵": 74445,
+      "芹": 74446,
+      "茨": 74447,
+      "√": 74448,
+      "ž": 74449,
+      "는": 74450,
+      "虞": 74451,
+      "艇": 74452,
+      "烹": 74453,
+      "奢": 74454,
+      "粹": 74455,
+      "骚": 74456,
+      "糟": 74457,
+      "逛": 74458,
+      "ル": 74459,
+      "衫": 74460,
+      "泼": 74461,
+      "弧": 74462,
+      "崖": 74463,
+      "鳞": 74464,
+      "叛": 74465,
+      "铅": 74466,
+      "歇": 74467,
+      "¶": 74468,
+      "谣": 74469,
+      "邑": 74470,
+      "憾": 74471,
+      "杠": 74472,
+      "↑": 74473,
+      "烘": 74474,
+      "栓": 74475,
+      "勉": 74476,
+      "膨": 74477,
+      "裹": 74478,
+      "骄": 74479,
+      "屠": 74480,
+      "龚": 74481,
+      "栈": 74482,
+      "嗯": 74483,
+      "饿": 74484,
+      "斥": 74485,
+      "伞": 74486,
+      "爵": 74487,
+      "っ": 74488,
+      "芙": 74489,
+      "斩": 74490,
+      "杉": 74491,
+      "惹": 74492,
+      "俞": 74493,
+      "挪": 74494,
+      "虐": 74495,
+      "陀": 74496,
+      "祠": 74497,
+      "捉": 74498,
+      "姬": 74499,
+      "闫": 74500,
+      "하": 74501,
+      "佬": 74502,
+      "≠": 74503,
+      "枢": 74504,
+      "Δ": 74505,
+      "う": 74506,
+      "甫": 74507,
+      "乖": 74508,
+      "屡": 74509,
+      "ら": 74510,
+      "熔": 74511,
+      "弓": 74512,
+      "踢": 74513,
+      "誓": 74514,
+      "尴": 74515,
+      "ت": 74516,
+      "镀": 74517,
+      "匠": 74518,
+      "樊": 74519,
+      "楠": 74520,
+      "橙": 74521,
+      "鞍": 74522,
+      "躁": 74523,
+      "坠": 74524,
+      "扔": 74525,
+      "ν": 74526,
+      "藻": 74527,
+      "奠": 74528,
+      "黔": 74529,
+      "雇": 74530,
+      "笋": 74531,
+      "钠": 74532,
+      "皱": 74533,
+      "慌": 74534,
+      "郝": 74535,
+      "酮": 74536,
+      "匆": 74537,
+      "こ": 74538,
+      "牡": 74539,
+      "耿": 74540,
+      "摔": 74541,
+      "⊥": 74542,
+      "巫": 74543,
+      "裸": 74544,
+      "弊": 74545,
+      "苔": 74546,
+      "坎": 74547,
+      "灶": 74548,
+      "禹": 74549,
+      "咋": 74550,
+      "僵": 74551,
+      "诵": 74552,
+      "隧": 74553,
+      "에": 74554,
+      "桩": 74555,
+      "厢": 74556,
+      "谜": 74557,
+      "锌": 74558,
+      "爹": 74559,
+      "宰": 74560,
+      "砍": 74561,
+      "ę": 74562,
+      "勺": 74563,
+      "赐": 74564,
+      "汰": 74565,
+      "ğ": 74566,
+      "嘲": 74567,
+      "蔽": 74568,
+      "晖": 74569,
+      "蕉": 74570,
+      "砸": 74571,
+      "癜": 74572,
+      "韧": 74573,
+      "蚁": 74574,
+      "闸": 74575,
+      "澜": 74576,
+      "焕": 74577,
+      "沥": 74578,
+      "醛": 74579,
+      "의": 74580,
+      "沐": 74581,
+      "█": 74582,
+      "も": 74583,
+      "罕": 74584,
+      "栗": 74585,
+      "炫": 74586,
+      "ì": 74587,
+      "峻": 74588,
+      "壹": 74589,
+      "り": 74590,
+      "倪": 74591,
+      "奸": 74592,
+      "橘": 74593,
+      "炖": 74594,
+      "堤": 74595,
+      "锂": 74596,
+      "喘": 74597,
+      "卒": 74598,
+      "ю": 74599,
+      "寞": 74600,
+      "倩": 74601,
+      "蜡": 74602,
+      "禄": 74603,
+      "潍": 74604,
+      "祁": 74605,
+      "ū": 74606,
+      "匿": 74607,
+      "尧": 74608,
+      "刹": 74609,
+      "仗": 74610,
+      "闺": 74611,
+      "κ": 74612,
+      "ク": 74613,
+      "屑": 74614,
+      "로": 74615,
+      "雌": 74616,
+      "棕": 74617,
+      "φ": 74618,
+      "æ": 74619,
+      "巅": 74620,
+      "禾": 74621,
+      "을": 74622,
+      "槛": 74623,
+      "さ": 74624,
+      "肆": 74625,
+      "傍": 74626,
+      "د": 74627,
+      "ι": 74628,
+      "´": 74629,
+      "焰": 74630,
+      "ш": 74631,
+      "窑": 74632,
+      "啤": 74633,
+      "凹": 74634,
+      "莓": 74635,
+      "媚": 74636,
+      "≈": 74637,
+      "婉": 74638,
+      "霄": 74639,
+      "伽": 74640,
+      "昊": 74641,
+      "痒": 74642,
+      "沁": 74643,
+      "琦": 74644,
+      "氟": 74645,
+      "塌": 74646,
+      "舶": 74647,
+      "慨": 74648,
+      "绎": 74649,
+      "仑": 74650,
+      "溯": 74651,
+      "ب": 74652,
+      "盯": 74653,
+      "韶": 74654,
+      "荔": 74655,
+      "澡": 74656,
+      "拦": 74657,
+      "畔": 74658,
+      "愚": 74659,
+      "疹": 74660,
+      "亨": 74661,
+      "媛": 74662,
+      "倦": 74663,
+      "ß": 74664,
+      "胰": 74665,
+      "矣": 74666,
+      "涩": 74667,
+      "硝": 74668,
+      "リ": 74669,
+      "揽": 74670,
+      "讽": 74671,
+      "疮": 74672,
+      "ラ": 74673,
+      "堰": 74674,
+      "〔": 74675,
+      "ッ": 74676,
+      "佣": 74677,
+      "渭": 74678,
+      "〕": 74679,
+      "∩": 74680,
+      "ą": 74681,
+      "가": 74682,
+      "痰": 74683,
+      "垒": 74684,
+      "鞭": 74685,
+      "娥": 74686,
+      "媳": 74687,
+      "涡": 74688,
+      "匪": 74689,
+      "閒": 74690,
+      "嘿": 74691,
+      "″": 74692,
+      "挫": 74693,
+      "竖": 74694,
+      "贮": 74695,
+      "淳": 74696,
+      "漳": 74697,
+      "阮": 74698,
+      "顽": 74699,
+      "酥": 74700,
+      "畴": 74701,
+      "腕": 74702,
+      "岚": 74703,
+      "钝": 74704,
+      "烛": 74705,
+      "＼": 74706,
+      "拱": 74707,
+      "С": 74708,
+      "爪": 74709,
+      "õ": 74710,
+      "耍": 74711,
+      "呜": 74712,
+      "凄": 74713,
+      "舱": 74714,
+      "揉": 74715,
+      "矮": 74716,
+      "柠": 74717,
+      "Ⅱ": 74718,
+      "＊": 74719,
+      "炳": 74720,
+      "啼": 74721,
+      "荧": 74722,
+      "蛛": 74723,
+      "魄": 74724,
+      "蚕": 74725,
+      "骏": 74726,
+      "诀": 74727,
+      "笛": 74728,
+      "卉": 74729,
+      "晃": 74730,
+      "汀": 74731,
+      "勐": 74732,
+      "한": 74733,
+      "捡": 74734,
+      "捏": 74735,
+      "豹": 74736,
+      "灸": 74737,
+      "刃": 74738,
+      "饥": 74739,
+      "く": 74740,
+      "耻": 74741,
+      "逗": 74742,
+      "蔓": 74743,
+      "鸦": 74744,
+      "汕": 74745,
+      "驿": 74746,
+      "祛": 74747,
+      "拯": 74748,
+      "冉": 74749,
+      "俯": 74750,
+      "δ": 74751,
+      "孢": 74752,
+      "懈": 74753,
+      "梧": 74754,
+      "籽": 74755,
+      "き": 74756,
+      "妊": 74757,
+      "宛": 74758,
+      "崛": 74759,
+      "杖": 74760,
+      "滥": 74761,
+      "П": 74762,
+      "雯": 74763,
+      "ë": 74764,
+      "蕊": 74765,
+      "咒": 74766,
+      "畸": 74767,
+      "棘": 74768,
+      "昼": 74769,
+      "窦": 74770,
+      "洽": 74771,
+      "庵": 74772,
+      "垄": 74773,
+      "迭": 74774,
+      "棠": 74775,
+      "霖": 74776,
+      "吁": 74777,
+      "溜": 74778,
+      "菩": 74779,
+      "哎": 74780,
+      "埔": 74781,
+      "涤": 74782,
+      "敞": 74783,
+      "瞎": 74784,
+      "瞻": 74785,
+      "钾": 74786,
+      "ù": 74787,
+      "기": 74788,
+      "敛": 74789,
+      "寝": 74790,
+      "馅": 74791,
+      "埠": 74792,
+      "甩": 74793,
+      "ア": 74794,
+      "지": 74795,
+      "柿": 74796,
+      "袍": 74797,
+      "撼": 74798,
+      "ś": 74799,
+      "µ": 74800,
+      "梭": 74801,
+      "蹄": 74802,
+      "娠": 74803,
+      "缅": 74804,
+      "歪": 74805,
+      "寡": 74806,
+      "乒": 74807,
+      "妄": 74808,
+      "沛": 74809,
+      "蛙": 74810,
+      "檬": 74811,
+      "를": 74812,
+      "娶": 74813,
+      "泣": 74814,
+      "葵": 74815,
+      "歷": 74816,
+      "啸": 74817,
+      "邯": 74818,
+      "鸽": 74819,
+      "糯": 74820,
+      "拂": 74821,
+      "棍": 74822,
+      "妍": 74823,
+      "矢": 74824,
+      "呗": 74825,
+      "復": 74826,
+      "膳": 74827,
+      "腌": 74828,
+      "∥": 74829,
+      "嘻": 74830,
+      "碌": 74831,
+      "Ⅰ": 74832,
+      "觅": 74833,
+      "吏": 74834,
+      "挨": 74835,
+      "稚": 74836,
+      "巍": 74837,
+      "萃": 74838,
+      "﹥": 74839,
+      "杞": 74840,
+      "ě": 74841,
+      "﹤": 74842,
+      "▲": 74843,
+      "絮": 74844,
+      "蔚": 74845,
+      "牟": 74846,
+      "酰": 74847,
+      "ż": 74848,
+      "㎡": 74849,
+      "芭": 74850,
+      "丫": 74851,
+      "祈": 74852,
+      "ه": 74853,
+      "莺": 74854,
+      "锯": 74855,
+      "熏": 74856,
+      "서": 74857,
+      "哉": 74858,
+      "◆": 74859,
+      "掀": 74860,
+      "聂": 74861,
+      "赃": 74862,
+      "ș": 74863,
+      "ф": 74864,
+      "萎": 74865,
+      "喀": 74866,
+      "î": 74867,
+      "甄": 74868,
+      "滕": 74869,
+      "哇": 74870,
+      "タ": 74871,
+      "ん": 74872,
+      "朔": 74873,
+      "谎": 74874,
+      "饺": 74875,
+      "榴": 74876,
+      "沦": 74877,
+      "蠢": 74878,
+      "艘": 74879,
+      "ń": 74880,
+      "郸": 74881,
+      "彝": 74882,
+      "蝉": 74883,
+      "奕": 74884,
+      "拐": 74885,
+      "渺": 74886,
+      "裴": 74887,
+      "嘱": 74888,
+      "煲": 74889,
+      "梓": 74890,
+      "堕": 74891,
+      "제": 74892,
+      "颅": 74893,
+      "だ": 74894,
+      "贱": 74895,
+      "⇒": 74896,
+      "あ": 74897,
+      "黛": 74898,
+      "粪": 74899,
+      "屿": 74900,
+      "ř": 74901,
+      "檐": 74902,
+      "於": 74903,
+      "裔": 74904,
+      "蹲": 74905,
+      "§": 74906,
+      "阑": 74907,
+      "В": 74908,
+      "泸": 74909,
+      "闵": 74910,
+      "鬓": 74911,
+      "蚊": 74912,
+      "葫": 74913,
+      "贬": 74914,
+      "毯": 74915,
+      "η": 74916,
+      "수": 74917,
+      "雍": 74918,
+      "س": 74919,
+      "涧": 74920,
+      "糙": 74921,
+      "耽": 74922,
+      "呕": 74923,
+      "颤": 74924,
+      "腥": 74925,
+      "烷": 74926,
+      "隋": 74927,
+      "靶": 74928,
+      "嗽": 74929,
+      "翟": 74930,
+      "蝴": 74931,
+      "衢": 74932,
+      "睁": 74933,
+      "稽": 74934,
+      "垦": 74935,
+      "佑": 74936,
+      "藉": 74937,
+      "ド": 74938,
+      "窥": 74939,
+      "ی": 74940,
+      "脖": 74941,
+      "彤": 74942,
+      "đ": 74943,
+      "浊": 74944,
+      "∪": 74945,
+      "跪": 74946,
+      "陇": 74947,
+      "喔": 74948,
+      "莆": 74949,
+      "辜": 74950,
+      "戳": 74951,
+      "冤": 74952,
+      "卦": 74953,
+      "昧": 74954,
+      "茜": 74955,
+      "胚": 74956,
+      "焚": 74957,
+      "眸": 74958,
+      "ý": 74959,
+      "정": 74960,
+      "皂": 74961,
+      "舅": 74962,
+      "彪": 74963,
+      "翩": 74964,
+      "摧": 74965,
+      "磅": 74966,
+      "骆": 74967,
+      "缕": 74968,
+      "诠": 74969,
+      "仆": 74970,
+      "芸": 74971,
+      "凋": 74972,
+      "♀": 74973,
+      "고": 74974,
+      "轧": 74975,
+      "丞": 74976,
+      "乞": 74977,
+      "砌": 74978,
+      "瘾": 74979,
+      "绥": 74980,
+      "栅": 74981,
+      "寥": 74982,
+      "蚌": 74983,
+      "喧": 74984,
+      "琅": 74985,
+      "肛": 74986,
+      "ع": 74987,
+      "淹": 74988,
+      "酚": 74989,
+      "魁": 74990,
+      "笨": 74991,
+      "疡": 74992,
+      "镁": 74993,
+      "稠": 74994,
+      "掏": 74995,
+      "溉": 74996,
+      "荫": 74997,
+      "榨": 74998,
+      "曦": 74999,
+      "碟": 75000,
+      "丨": 75001,
+      "殴": 75002,
+      "阙": 75003,
+      "钧": 75004,
+      "榄": 75005,
+      "щ": 75006,
+      "娄": 75007,
+      "旷": 75008,
+      "瀑": 75009,
+      "卤": 75010,
+      "曙": 75011,
+      "ㄧ": 75012,
+      "翘": 75013,
+      "淫": 75014,
+      "狸": 75015,
+      "니": 75016,
+      "睹": 75017,
+      "菠": 75018,
+      "廓": 75019,
+      "袜": 75020,
+      "橱": 75021,
+      "挚": 75022,
+      "兜": 75023,
+      "肋": 75024,
+      "绞": 75025,
+      "よ": 75026,
+      "圭": 75027,
+      "椰": 75028,
+      "プ": 75029,
+      "皓": 75030,
+      "璐": 75031,
+      "诡": 75032,
+      "饪": 75033,
+      "扒": 75034,
+      "屎": 75035,
+      "泗": 75036,
+      "庚": 75037,
+      "蛾": 75038,
+      "리": 75039,
+      "鲤": 75040,
+      "睐": 75041,
+      "璧": 75042,
+      "ò": 75043,
+      "猿": 75044,
+      "瞧": 75045,
+      "槐": 75046,
+      "淇": 75047,
+      "姥": 75048,
+      "☆": 75049,
+      "ा": 75050,
+      "躯": 75051,
+      "←": 75052,
+      "苟": 75053,
+      "沼": 75054,
+      "焉": 75055,
+      "鸥": 75056,
+      "사": 75057,
+      "绮": 75058,
+      "韭": 75059,
+      "鳍": 75060,
+      "缔": 75061,
+      "咯": 75062,
+      "垢": 75063,
+      "弈": 75064,
+      "拭": 75065,
+      "簧": 75066,
+      "徵": 75067,
+      "讶": 75068,
+      "绽": 75069,
+      "皋": 75070,
+      "崎": 75071,
+      "孵": 75072,
+      "孜": 75073,
+      "哺": 75074,
+      "麒": 75075,
+      "苷": 75076,
+      "フ": 75077,
+      "芋": 75078,
+      "嗟": 75079,
+      "乓": 75080,
+      "绸": 75081,
+      "灼": 75082,
+      "Р": 75083,
+      "■": 75084,
+      "瞒": 75085,
+      "∼": 75086,
+      "犀": 75087,
+      "俭": 75088,
+      "驴": 75089,
+      "靓": 75090,
+      "尉": 75091,
+      "哑": 75092,
+      "攒": 75093,
+      "竣": 75094,
+      "趟": 75095,
+      "峨": 75096,
+      "渲": 75097,
+      "ロ": 75098,
+      "晏": 75099,
+      "瞳": 75100,
+      "贿": 75101,
+      "渍": 75102,
+      "吼": 75103,
+      "逍": 75104,
+      "蚂": 75105,
+      "哮": 75106,
+      "▼": 75107,
+      "Á": 75108,
+      "蘑": 75109,
+      "枸": 75110,
+      "剔": 75111,
+      "Д": 75112,
+      "‰": 75113,
+      "簿": 75114,
+      "铲": 75115,
+      "镶": 75116,
+      "搓": 75117,
+      "梵": 75118,
+      "스": 75119,
+      "ˋ": 75120,
+      "坞": 75121,
+      "嵩": 75122,
+      "哄": 75123,
+      "乍": 75124,
+      "茹": 75125,
+      "惕": 75126,
+      "叮": 75127,
+      "鲸": 75128,
+      "ς": 75129,
+      "汾": 75130,
+      "凿": 75131,
+      "娅": 75132,
+      "飙": 75133,
+      "盏": 75134,
+      "竿": 75135,
+      "嗜": 75136,
+      "К": 75137,
+      "麓": 75138,
+      "帷": 75139,
+      "け": 75140,
+      "어": 75141,
+      "喇": 75142,
+      "汛": 75143,
+      "诏": 75144,
+      "鄙": 75145,
+      "坟": 75146,
+      "│": 75147,
+      "⩽": 75148,
+      "瘫": 75149,
+      "楷": 75150,
+      "岐": 75151,
+      "Ω": 75152,
+      "暇": 75153,
+      "襟": 75154,
+      "蹭": 75155,
+      "煳": 75156,
+      "篱": 75157,
+      "О": 75158,
+      "Ö": 75159,
+      "驼": 75160,
+      "窟": 75161,
+      "자": 75162,
+      "═": 75163,
+      "捣": 75164,
+      "磕": 75165,
+      "岑": 75166,
+      "缪": 75167,
+      "桦": 75168,
+      "脓": 75169,
+      "嚼": 75170,
+      "掷": 75171,
+      "Н": 75172,
+      "祀": 75173,
+      "ف": 75174,
+      "钛": 75175,
+      "„": 75176,
+      "昕": 75177,
+      "哟": 75178,
+      "橄": 75179,
+      "缚": 75180,
+      "霆": 75181,
+      "寇": 75182,
+      "琢": 75183,
+      "シ": 75184,
+      "澎": 75185,
+      "匈": 75186,
+      "ة": 75187,
+      "涅": 75188,
+      "藕": 75189,
+      "臀": 75190,
+      "嫂": 75191,
+      "釉": 75192,
+      "颌": 75193,
+      "笙": 75194,
+      "妒": 75195,
+      "懵": 75196,
+      "厄": 75197,
+      "甬": 75198,
+      "墩": 75199,
+      "哩": 75200,
+      "扛": 75201,
+      "篆": 75202,
+      "菏": 75203,
+      "侈": 75204,
+      "怅": 75205,
+      "酪": 75206,
+      "侃": 75207,
+      "後": 75208,
+      "寅": 75209,
+      "葩": 75210,
+      "覃": 75211,
+      "晾": 75212,
+      "阎": 75213,
+      "桔": 75214,
+      "眷": 75215,
+      "嚣": 75216,
+      "裳": 75217,
+      "舜": 75218,
+      "噬": 75219,
+      "採": 75220,
+      "시": 75221,
+      "衅": 75222,
+      "해": 75223,
+      "斐": 75224,
+      "靡": 75225,
+      "鸾": 75226,
+      "呐": 75227,
+      "抒": 75228,
+      "粟": 75229,
+      "つ": 75230,
+      "纶": 75231,
+      "瑕": 75232,
+      "翡": 75233,
+      "ジ": 75234,
+      "ț": 75235,
+      "瘀": 75236,
+      "筝": 75237,
+      "狄": 75238,
+      "漓": 75239,
+      "戎": 75240,
+      "壑": 75241,
+      "蜘": 75242,
+      "喵": 75243,
+      "哨": 75244,
+      "コ": 75245,
+      "ǎ": 75246,
+      "唉": 75247,
+      "钰": 75248,
+      "斧": 75249,
+      "雏": 75250,
+      "睦": 75251,
+      "垣": 75252,
+      "י": 75253,
+      "佰": 75254,
+      "瀚": 75255,
+      "А": 75256,
+      "盡": 75257,
+      "汶": 75258,
+      "琉": 75259,
+      "矶": 75260,
+      "レ": 75261,
+      "叭": 75262,
+      "掺": 75263,
+      "钞": 75264,
+      "碾": 75265,
+      "ē": 75266,
+      "肽": 75267,
+      "颐": 75268,
+      "柬": 75269,
+      "茵": 75270,
+      "υ": 75271,
+      "篷": 75272,
+      "桓": 75273,
+      "⊙": 75274,
+      "½": 75275,
+      "薰": 75276,
+      "胱": 75277,
+      "挠": 75278,
+      "伺": 75279,
+      "蹦": 75280,
+      "窍": 75281,
+      "哼": 75282,
+      "掠": 75283,
+      "玺": 75284,
+      "梢": 75285,
+      "炜": 75286,
+      "澈": 75287,
+      "玮": 75288,
+      "Š": 75289,
+      "撩": 75290,
+      "蝇": 75291,
+      "唑": 75292,
+      "捆": 75293,
+      "인": 75294,
+      "ㄨ": 75295,
+      "赎": 75296,
+      "ï": 75297,
+      "苞": 75298,
+      "陋": 75299,
+      "嘎": 75300,
+      "濮": 75301,
+      "拙": 75302,
+      "榕": 75303,
+      "э": 75304,
+      "⑤": 75305,
+      "Â": 75306,
+      "蛟": 75307,
+      "簇": 75308,
+      "्": 75309,
+      "憩": 75310,
+      "χ": 75311,
+      "朽": 75312,
+      "惭": 75313,
+      "嗓": 75314,
+      "坯": 75315,
+      "疤": 75316,
+      "俏": 75317,
+      "藩": 75318,
+      "筷": 75319,
+      "秤": 75320,
+      "은": 75321,
+      "瑾": 75322,
+      "徊": 75323,
+      "恳": 75324,
+      "恍": 75325,
+      "毗": 75326,
+      "瞄": 75327,
+      "帜": 75328,
+      "蝎": 75329,
+      "洱": 75330,
+      "め": 75331,
+      "悍": 75332,
+      "樟": 75333,
+      "ˊ": 75334,
+      "迄": 75335,
+      "∀": 75336,
+      "旌": 75337,
+      "र": 75338,
+      "テ": 75339,
+      "萤": 75340,
+      "阱": 75341,
+      "溺": 75342,
+      "萼": 75343,
+      "そ": 75344,
+      "晟": 75345,
+      "穹": 75346,
+      "颍": 75347,
+      "恤": 75348,
+      "嗨": 75349,
+      "柚": 75350,
+      "檀": 75351,
+      "※": 75352,
+      "있": 75353,
+      "愣": 75354,
+      "鹭": 75355,
+      "蜗": 75356,
+      "蔗": 75357,
+      "趴": 75358,
+      "榻": 75359,
+      "肴": 75360,
+      "锚": 75361,
+      "怼": 75362,
+      "秆": 75363,
+      "泾": 75364,
+      "枉": 75365,
+      "莽": 75366,
+      "汲": 75367,
+      "痹": 75368,
+      "烁": 75369,
+      "洼": 75370,
+      "哒": 75371,
+      "窜": 75372,
+      "舔": 75373,
+      "弛": 75374,
+      "ق": 75375,
+      "ו": 75376,
+      "嗣": 75377,
+      "え": 75378,
+      "涕": 75379,
+      "碘": 75380,
+      "敖": 75381,
+      "용": 75382,
+      "紊": 75383,
+      "豁": 75384,
+      "徘": 75385,
+      "涟": 75386,
+      "亥": 75387,
+      "圃": 75388,
+      "炬": 75389,
+      "デ": 75390,
+      "瞩": 75391,
+      "肘": 75392,
+      "煞": 75393,
+      "М": 75394,
+      "聆": 75395,
+      "荟": 75396,
+      "焙": 75397,
+      "º": 75398,
+      "裘": 75399,
+      "メ": 75400,
+      "渤": 75401,
+      "脐": 75402,
+      "焖": 75403,
+      "厥": 75404,
+      "珀": 75405,
+      "宸": 75406,
+      "骼": 75407,
+      "绚": 75408,
+      "ð": 75409,
+      "靳": 75410,
+      "砚": 75411,
+      "夯": 75412,
+      "迦": 75413,
+      "妾": 75414,
+      "町": 75415,
+      "週": 75416,
+      "坂": 75417,
+      "า": 75418,
+      "啪": 75419,
+      "İ": 75420,
+      "贰": 75421,
+      "睫": 75422,
+      "茸": 75423,
+      "み": 75424,
+      "일": 75425,
+      "哔": 75426,
+      "庶": 75427,
+      "眩": 75428,
+      "鞘": 75429,
+      "亢": 75430,
+      "羹": 75431,
+      "ك": 75432,
+      "拢": 75433,
+      "芥": 75434,
+      "驭": 75435,
+      "骁": 75436,
+      "Т": 75437,
+      "娩": 75438,
+      "殇": 75439,
+      "도": 75440,
+      "箫": 75441,
+      "熨": 75442,
+      "谍": 75443,
+      "Å": 75444,
+      "叨": 75445,
+      "ど": 75446,
+      "戊": 75447,
+      "耦": 75448,
+      "跋": 75449,
+      "マ": 75450,
+      "搁": 75451,
+      "沭": 75452,
+      "樵": 75453,
+      "臻": 75454,
+      "瓯": 75455,
+      "赡": 75456,
+      "崭": 75457,
+      "淤": 75458,
+      "苛": 75459,
+      "◎": 75460,
+      "噩": 75461,
+      "お": 75462,
+      "潢": 75463,
+      "滁": 75464,
+      "萱": 75465,
+      "拽": 75466,
+      "보": 75467,
+      "滔": 75468,
+      "瘟": 75469,
+      "鲨": 75470,
+      "ح": 75471,
+      "咕": 75472,
+      "霾": 75473,
+      "И": 75474,
+      "匡": 75475,
+      "囚": 75476,
+      "铬": 75477,
+      "馒": 75478,
+      "秃": 75479,
+      "鹊": 75480,
+      "瑚": 75481,
+      "饵": 75482,
+      "惫": 75483,
+      "歹": 75484,
+      "盔": 75485,
+      "豚": 75486,
+      "グ": 75487,
+      "嫉": 75488,
+      "鸳": 75489,
+      "孰": 75490,
+      "韬": 75491,
+      "颊": 75492,
+      "菁": 75493,
+      "鞠": 75494,
+      "辊": 75495,
+      "아": 75496,
+      "侮": 75497,
+      "憋": 75498,
+      "拇": 75499,
+      "僚": 75500,
+      "沽": 75501,
+      "沌": 75502,
+      "扉": 75503,
+      "禧": 75504,
+      "क": 75505,
+      "ư": 75506,
+      "孽": 75507,
+      "僻": 75508,
+      "樽": 75509,
+      "烽": 75510,
+      "柑": 75511,
+      "祺": 75512,
+      "犁": 75513,
+      "冕": 75514,
+      "茗": 75515,
+      "カ": 75516,
+      "绅": 75517,
+      "髮": 75518,
+      "绛": 75519,
+      "湃": 75520,
+      "그": 75521,
+      "蒿": 75522,
+      "绊": 75523,
+      "炙": 75524,
+      "濒": 75525,
+      "트": 75526,
+      "遏": 75527,
+      "⭐": 75528,
+      "漱": 75529,
+      "❤": 75530,
+      "熄": 75531,
+      "時": 75532,
+      "蛤": 75533,
+      "鹃": 75534,
+      "墟": 75535,
+      "毋": 75536,
+      "茧": 75537,
+      "玖": 75538,
+      "觞": 75539,
+      "倘": 75540,
+      "👍": 75541,
+      "氓": 75542,
+      "秧": 75543,
+      "や": 75544,
+      "歼": 75545,
+      "문": 75546,
+      "渚": 75547,
+      "怯": 75548,
+      "珑": 75549,
+      "砺": 75550,
+      "诛": 75551,
+      "峪": 75552,
+      "น": 75553,
+      "祷": 75554,
+      "褪": 75555,
+      "颓": 75556,
+      "耸": 75557,
+      "¥": 75558,
+      "懿": 75559,
+      "浣": 75560,
+      "대": 75561,
+      "刁": 75562,
+      "靴": 75563,
+      "나": 75564,
+      "Ü": 75565,
+      "啰": 75566,
+      "磺": 75567,
+      "辙": 75568,
+      "瞭": 75569,
+      "晦": 75570,
+      "腋": 75571,
+      "揣": 75572,
+      "宦": 75573,
+      "晗": 75574,
+      "莘": 75575,
+      "钳": 75576,
+      "煜": 75577,
+      "贻": 75578,
+      "溅": 75579,
+      "ő": 75580,
+      "讳": 75581,
+      "烨": 75582,
+      "嘟": 75583,
+      "凳": 75584,
+      "¿": 75585,
+      "睢": 75586,
+      "滇": 75587,
+      "黯": 75588,
+      "羟": 75589,
+      "镍": 75590,
+      "苓": 75591,
+      "聋": 75592,
+      "郴": 75593,
+      "撇": 75594,
+      "趾": 75595,
+      "岱": 75596,
+      "ə": 75597,
+      "サ": 75598,
+      "ち": 75599,
+      "으": 75600,
+      "蕃": 75601,
+      "拧": 75602,
+      "闰": 75603,
+      "邂": 75604,
+      "陂": 75605,
+      "窖": 75606,
+      "褚": 75607,
+      "暧": 75608,
+      "ィ": 75609,
+      "潼": 75610,
+      "铂": 75611,
+      "赘": 75612,
+      "迢": 75613,
+      "バ": 75614,
+      "仟": 75615,
+      "蔷": 75616,
+      "婧": 75617,
+      "珂": 75618,
+      "眺": 75619,
+      "哗": 75620,
+      "癣": 75621,
+      "璋": 75622,
+      "筐": 75623,
+      "焯": 75624,
+      "奚": 75625,
+      "漾": 75626,
+      "鄞": 75627,
+      "怠": 75628,
+      "礁": 75629,
+      "요": 75630,
+      "兀": 75631,
+      "捂": 75632,
+      "恕": 75633,
+      "ج": 75634,
+      "侄": 75635,
+      "躬": 75636,
+      "ش": 75637,
+      "‑": 75638,
+      "凛": 75639,
+      "ה": 75640,
+      "抉": 75641,
+      "垮": 75642,
+      "袅": 75643,
+      "蠕": 75644,
+      "ί": 75645,
+      "陨": 75646,
+      "†": 75647,
+      "羲": 75648,
+      "筱": 75649,
+      "⩾": 75650,
+      "俘": 75651,
+      "绰": 75652,
+      "迥": 75653,
+      "盎": 75654,
+      "酣": 75655,
+      "扳": 75656,
+      "挟": 75657,
+      "棺": 75658,
+      "蕨": 75659,
+      "禺": 75660,
+      "绷": 75661,
+      "涪": 75662,
+      "Ã": 75663,
+      "拷": 75664,
+      "蓟": 75665,
+      "咀": 75666,
+      "├": 75667,
+      "蜕": 75668,
+      "驯": 75669,
+      "￣": 75670,
+      "Ç": 75671,
+      "과": 75672,
+      "锰": 75673,
+      "剁": 75674,
+      "陡": 75675,
+      "纂": 75676,
+      "わ": 75677,
+      "缤": 75678,
+      "寐": 75679,
+      "姊": 75680,
+      "›": 75681,
+      "蚝": 75682,
+      "憨": 75683,
+      "秸": 75684,
+      "姗": 75685,
+      "े": 75686,
+      "蝠": 75687,
+      "悼": 75688,
+      "嗅": 75689,
+      "ξ": 75690,
+      "兢": 75691,
+      "轼": 75692,
+      "ム": 75693,
+      "磐": 75694,
+      "嘘": 75695,
+      "粽": 75696,
+      "キ": 75697,
+      "栾": 75698,
+      "璇": 75699,
+      "昱": 75700,
+      "逅": 75701,
+      "ร": 75702,
+      "fi": 75703,
+      "劈": 75704,
+      "胥": 75705,
+      "嬉": 75706,
+      "드": 75707,
+      "嫣": 75708,
+      "镑": 75709,
+      "簪": 75710,
+      "惬": 75711,
+      "ţ": 75712,
+      "岔": 75713,
+      "藓": 75714,
+      "ˇ": 75715,
+      "琛": 75716,
+      "亳": 75717,
+      "孚": 75718,
+      "俺": 75719,
+      "霓": 75720,
+      "寰": 75721,
+      "遐": 75722,
+      "飒": 75723,
+      "冢": 75724,
+      "芷": 75725,
+      "疵": 75726,
+      "螃": 75727,
+      "ブ": 75728,
+      "면": 75729,
+      "霏": 75730,
+      "鳄": 75731,
+      "琶": 75732,
+      "棹": 75733,
+      "猩": 75734,
+      "淼": 75735,
+      "烙": 75736,
+      "▽": 75737,
+      "唾": 75738,
+      "筠": 75739,
+      "钗": 75740,
+      "霁": 75741,
+      "泓": 75742,
+      "羌": 75743,
+      "せ": 75744,
+      "瀛": 75745,
+      "라": 75746,
+      "镖": 75747,
+      "⊂": 75748,
+      "萦": 75749,
+      "氪": 75750,
+      "笃": 75751,
+      "妞": 75752,
+      "笺": 75753,
+      "恺": 75754,
+      "唔": 75755,
+      "瞪": 75756,
+      "성": 75757,
+      "鸯": 75758,
+      "隅": 75759,
+      "筵": 75760,
+      "峥": 75761,
+      "谴": 75762,
+      "［": 75763,
+      "₹": 75764,
+      "忻": 75765,
+      "痔": 75766,
+      "釜": 75767,
+      "佼": 75768,
+      "栩": 75769,
+      "입": 75770,
+      "∫": 75771,
+      "伶": 75772,
+      "辄": 75773,
+      "瑛": 75774,
+      "汐": 75775,
+      "］": 75776,
+      "挛": 75777,
+      "祯": 75778,
+      "杳": 75779,
+      "أ": 75780,
+      "螨": 75781,
+      "ǐ": 75782,
+      "惰": 75783,
+      "璨": 75784,
+      "蟾": 75785,
+      "鹦": 75786,
+      "褶": 75787,
+      "ạ": 75788,
+      "戟": 75789,
+      "婺": 75790,
+      "¬": 75791,
+      "佟": 75792,
+      "啃": 75793,
+      "Č": 75794,
+      "狙": 75795,
+      "屉": 75796,
+      "苇": 75797,
+      "◇": 75798,
+      "槟": 75799,
+      "拣": 75800,
+      "踊": 75801,
+      "炊": 75802,
+      "晤": 75803,
+      "徙": 75804,
+      "螂": 75805,
+      "庇": 75806,
+      "锣": 75807,
+      "糜": 75808,
+      "Ł": 75809,
+      "咧": 75810,
+      "粱": 75811,
+      "硒": 75812,
+      "鳌": 75813,
+      "貂": 75814,
+      "曳": 75815,
+      "疱": 75816,
+      "绯": 75817,
+      "溧": 75818,
+      "缉": 75819,
+      "醚": 75820,
+      "髦": 75821,
+      "력": 75822,
+      "茬": 75823,
+      "剃": 75824,
+      "峭": 75825,
+      "札": 75826,
+      "眶": 75827,
+      "譬": 75828,
+      "睑": 75829,
+      "谏": 75830,
+      "悖": 75831,
+      "遁": 75832,
+      "瞿": 75833,
+      "チ": 75834,
+      "娴": 75835,
+      "만": 75836,
+      "浒": 75837,
+      "게": 75838,
+      "峙": 75839,
+      "慷": 75840,
+      "叁": 75841,
+      "滦": 75842,
+      "皎": 75843,
+      "亟": 75844,
+      "อ": 75845,
+      "膊": 75846,
+      "ウ": 75847,
+      "佈": 75848,
+      "辕": 75849,
+      "茉": 75850,
+      "न": 75851,
+      "∁": 75852,
+      "泠": 75853,
+      "馏": 75854,
+      "澹": 75855,
+      "주": 75856,
+      "楔": 75857,
+      "莒": 75858,
+      "铣": 75859,
+      "動": 75860,
+      "∃": 75861,
+      "鸠": 75862,
+      "蘸": 75863,
+      "桨": 75864,
+      "屹": 75865,
+      "驹": 75866,
+      "眯": 75867,
+      "喃": 75868,
+      "⊆": 75869,
+      "钵": 75870,
+      "腑": 75871,
+      "個": 75872,
+      "坷": 75873,
+      "烃": 75874,
+      "钊": 75875,
+      "赊": 75876,
+      "夭": 75877,
+      "ל": 75878,
+      "상": 75879,
+      "掰": 75880,
+      "회": 75881,
+      "オ": 75882,
+      "掖": 75883,
+      "摹": 75884,
+      "耘": 75885,
+      "浚": 75886,
+      "沅": 75887,
+      "柒": 75888,
+      "婿": 75889,
+      "ि": 75890,
+      "谒": 75891,
+      "剿": 75892,
+      "撸": 75893,
+      "恬": 75894,
+      "酋": 75895,
+      "パ": 75896,
+      "ά": 75897,
+      "炽": 75898,
+      "璀": 75899,
+      "Ø": 75900,
+      "胧": 75901,
+      "掐": 75902,
+      "侬": 75903,
+      "嵴": 75904,
+      "胳": 75905,
+      "汹": 75906,
+      "虏": 75907,
+      "眨": 75908,
+      "痉": 75909,
+      "琵": 75910,
+      "瓢": 75911,
+      "朦": 75912,
+      "स": 75913,
+      "龈": 75914,
+      "刨": 75915,
+      "冗": 75916,
+      "浔": 75917,
+      "巳": 75918,
+      "่": 75919,
+      "소": 75920,
+      "偈": 75921,
+      "⟩": 75922,
+      "嫖": 75923,
+      "嘶": 75924,
+      "Ó": 75925,
+      "毙": 75926,
+      "전": 75927,
+      "涝": 75928,
+      "诫": 75929,
+      "ό": 75930,
+      "ュ": 75931,
+      "窒": 75932,
+      "구": 75933,
+      "稣": 75934,
+      "抠": 75935,
+      "侗": 75936,
+      "들": 75937,
+      "戮": 75938,
+      "瓮": 75939,
+      "峦": 75940,
+      "Б": 75941,
+      "怂": 75942,
+      "藜": 75943,
+      "위": 75944,
+      "蝙": 75945,
+      "妳": 75946,
+      "ง": 75947,
+      "厮": 75948,
+      "芪": 75949,
+      "谬": 75950,
+      "砥": 75951,
+      "부": 75952,
+      "開": 75953,
+      "叟": 75954,
+      "咫": 75955,
+      "慑": 75956,
+      "毓": 75957,
+      "У": 75958,
+      "楞": 75959,
+      "熠": 75960,
+      "噢": 75961,
+      "腮": 75962,
+      "ả": 75963,
+      "エ": 75964,
+      "偃": 75965,
+      "谩": 75966,
+      "稼": 75967,
+      "🙂": 75968,
+      "氰": 75969,
+      "म": 75970,
+      "Đ": 75971,
+      "匮": 75972,
+      "捍": 75973,
+      "踝": 75974,
+      "癖": 75975,
+      "睾": 75976,
+      "褒": 75977,
+      "터": 75978,
+      "臧": 75979,
+      "楂": 75980,
+      "Σ": 75981,
+      "邸": 75982,
+      "诅": 75983,
+      "芍": 75984,
+      "酉": 75985,
+      "玥": 75986,
+      "霹": 75987,
+      "骇": 75988,
+      "仄": 75989,
+      "笠": 75990,
+      "雳": 75991,
+      "彷": 75992,
+      "ก": 75993,
+      "痊": 75994,
+      "ר": 75995,
+      "蟠": 75996,
+      "鹉": 75997,
+      "磋": 75998,
+      "锄": 75999,
+      "廿": 76000,
+      "淆": 76001,
+      "卯": 76002,
+      "鱿": 76003,
+      "³": 76004,
+      "할": 76005,
+      "挝": 76006,
+      "铵": 76007,
+      "⟨": 76008,
+      "椿": 76009,
+      "¡": 76010,
+      "喱": 76011,
+      "煸": 76012,
+      "咚": 76013,
+      "祇": 76014,
+      "吡": 76015,
+      "狩": 76016,
+      "त": 76017,
+      "洮": 76018,
+      "阈": 76019,
+      "א": 76020,
+      "斛": 76021,
+      "缨": 76022,
+      "Ž": 76023,
+      "瘙": 76024,
+      "辗": 76025,
+      "원": 76026,
+      "Г": 76027,
+      "خ": 76028,
+      "扼": 76029,
+      "棣": 76030,
+      "腱": 76031,
+      "淌": 76032,
+      "ी": 76033,
+      "圩": 76034,
+      "啧": 76035,
+      "籁": 76036,
+      "ǔ": 76037,
+      "屌": 76038,
+      "袂": 76039,
+      "谯": 76040,
+      "偕": 76041,
+      "ї": 76042,
+      "崽": 76043,
+      "轶": 76044,
+      "여": 76045,
+      "猕": 76046,
+      "诶": 76047,
+      "呦": 76048,
+      "場": 76049,
+      "炯": 76050,
+      "髋": 76051,
+      "間": 76052,
+      "เ": 76053,
+      "註": 76054,
+      "芮": 76055,
+      "豌": 76056,
+      "芡": 76057,
+      "谛": 76058,
+      "娣": 76059,
+      "惆": 76060,
+      "拎": 76061,
+      "俨": 76062,
+      "卞": 76063,
+      "ョ": 76064,
+      "恪": 76065,
+      "昵": 76066,
+      "출": 76067,
+      "漯": 76068,
+      "慵": 76069,
+      "潺": 76070,
+      "叩": 76071,
+      "セ": 76072,
+      "囤": 76073,
+      "溴": 76074,
+      "裏": 76075,
+      "丐": 76076,
+      "猥": 76077,
+      "憧": 76078,
+      "啶": 76079,
+      "疚": 76080,
+      "箔": 76081,
+      "懦": 76082,
+      "宕": 76083,
+      "匯": 76084,
+      "唧": 76085,
+      "噜": 76086,
+      "憬": 76087,
+      "毡": 76088,
+      "硼": 76089,
+      "铎": 76090,
+      "ѕ": 76091,
+      "έ": 76092,
+      "漪": 76093,
+      "痣": 76094,
+      "跻": 76095,
+      "じ": 76096,
+      "羯": 76097,
+      "안": 76098,
+      "Φ": 76099,
+      "嬛": 76100,
+      "狡": 76101,
+      "荚": 76102,
+      "้": 76103,
+      "悚": 76104,
+      "煽": 76105,
+      "ナ": 76106,
+      "淬": 76107,
+      "斟": 76108,
+      "唠": 76109,
+      "湄": 76110,
+      "秽": 76111,
+      "璞": 76112,
+      "鲈": 76113,
+      "ת": 76114,
+      "姣": 76115,
+      "诣": 76116,
+      "猝": 76117,
+      "発": 76118,
+      "虔": 76119,
+      "匣": 76120,
+      "З": 76121,
+      "鲫": 76122,
+      "汞": 76123,
+      "荼": 76124,
+      "篡": 76125,
+      "ャ": 76126,
+      "缭": 76127,
+      "揪": 76128,
+      "우": 76129,
+      "擒": 76130,
+      "ニ": 76131,
+      "铠": 76132,
+      "邳": 76133,
+      "Е": 76134,
+      "경": 76135,
+      "忱": 76136,
+      "悸": 76137,
+      "铮": 76138,
+      "悴": 76139,
+      "霭": 76140,
+      "骸": 76141,
+      "攸": 76142,
+      "弼": 76143,
+      "沖": 76144,
+      "모": 76145,
+      "啄": 76146,
+      "丶": 76147,
+      "∧": 76148,
+      "遴": 76149,
+      "ㄢ": 76150,
+      "∆": 76151,
+      "麽": 76152,
+      "습": 76153,
+      "荀": 76154,
+      "惶": 76155,
+      "ั": 76156,
+      "衙": 76157,
+      "장": 76158,
+      "Ⅲ": 76159,
+      "ز": 76160,
+      "━": 76161,
+      "́": 76162,
+      "蓓": 76163,
+      "稷": 76164,
+      "擀": 76165,
+      "胫": 76166,
+      "揍": 76167,
+      "夙": 76168,
+      "燮": 76169,
+      "曜": 76170,
+      "嗷": 76171,
+      "骥": 76172,
+      "濛": 76173,
+      "ė": 76174,
+      "濯": 76175,
+      "鬟": 76176,
+      "榛": 76177,
+      "∂": 76178,
+      "ह": 76179,
+      "柘": 76180,
+      "오": 76181,
+      "窘": 76182,
+      "ź": 76183,
+      "ص": 76184,
+      "阪": 76185,
+      "ъ": 76186,
+      "琥": 76187,
+      "黍": 76188,
+      "捅": 76189,
+      "ダ": 76190,
+      "ế": 76191,
+      "拈": 76192,
+      "ビ": 76193,
+      "俐": 76194,
+      "妓": 76195,
+      "枭": 76196,
+      "捲": 76197,
+      "ک": 76198,
+      "ば": 76199,
+      "瞰": 76200,
+      "내": 76201,
+      "辍": 76202,
+      "幹": 76203,
+      "莅": 76204,
+      "憎": 76205,
+      "ệ": 76206,
+      "壬": 76207,
+      "⚠": 76208,
+      "斓": 76209,
+      "嫦": 76210,
+      "岫": 76211,
+      "ม": 76212,
+      "嵘": 76213,
+      "蔑": 76214,
+      "陛": 76215,
+      "⁄": 76216,
+      "학": 76217,
+      "ェ": 76218,
+      "蕙": 76219,
+      "戍": 76220,
+      "熹": 76221,
+      "薏": 76222,
+      "茯": 76223,
+      "적": 76224,
+      "♦": 76225,
+      "¯": 76226,
+      "蟑": 76227,
+      "ב": 76228,
+      "ψ": 76229,
+      "缥": 76230,
+      "ط": 76231,
+      "喽": 76232,
+      "∶": 76233,
+      "這": 76234,
+      "朕": 76235,
+      "郯": 76236,
+      "馋": 76237,
+      "熘": 76238,
+      "嘅": 76239,
+      "讷": 76240,
+      "嚷": 76241,
+      "榭": 76242,
+      "מ": 76243,
+      "侥": 76244,
+      "비": 76245,
+      "隘": 76246,
+      "ミ": 76247,
+      "ů": 76248,
+      "設": 76249,
+      "되": 76250,
+      "공": 76251,
+      "隽": 76252,
+      "Ä": 76253,
+      "▄": 76254,
+      "脯": 76255,
+      "疣": 76256,
+      "弋": 76257,
+      "溟": 76258,
+      "﹐": 76259,
+      "邃": 76260,
+      "晁": 76261,
+      "婪": 76262,
+      "婵": 76263,
+      "痪": 76264,
+      "殡": 76265,
+      "夔": 76266,
+      "昙": 76267,
+      "粳": 76268,
+      "嫡": 76269,
+      "禀": 76270,
+      "矜": 76271,
+      "祐": 76272,
+      "沮": 76273,
+      "実": 76274,
+      "綦": 76275,
+      "會": 76276,
+      "翌": 76277,
+      "悯": 76278,
+      "膛": 76279,
+      "砾": 76280,
+      "咐": 76281,
+      "業": 76282,
+      "₂": 76283,
+      "瘘": 76284,
+      "嵊": 76285,
+      "竺": 76286,
+      "槎": 76287,
+      "疙": 76288,
+      "撬": 76289,
+      "瘩": 76290,
+      "Î": 76291,
+      "闾": 76292,
+      "崧": 76293,
+      "淅": 76294,
+      "豉": 76295,
+      "痞": 76296,
+      "痢": 76297,
+      "骅": 76298,
+      "惮": 76299,
+      "蔺": 76300,
+      "骋": 76301,
+      "⑥": 76302,
+      "婕": 76303,
+      "疝": 76304,
+      "見": 76305,
+      "娓": 76306,
+      "데": 76307,
+      "傣": 76308,
+      "毂": 76309,
+      "‚": 76310,
+      "동": 76311,
+      "憔": 76312,
+      "प": 76313,
+      "합": 76314,
+      "荞": 76315,
+      "葆": 76316,
+      "ש": 76317,
+      "프": 76318,
+      "개": 76319,
+      "犊": 76320,
+      "颚": 76321,
+      "생": 76322,
+      "ल": 76323,
+      "涿": 76324,
+      "楹": 76325,
+      "ं": 76326,
+      "缈": 76327,
+      "湍": 76328,
+      "恣": 76329,
+      "寮": 76330,
+      "粑": 76331,
+      "镰": 76332,
+      "嘤": 76333,
+      "濑": 76334,
+      "榔": 76335,
+      "咔": 76336,
+      "酝": 76337,
+      "伫": 76338,
+      "戌": 76339,
+      "モ": 76340,
+      "記": 76341,
+      "镯": 76342,
+      "Ś": 76343,
+      "撮": 76344,
+      "ァ": 76345,
+      "👇": 76346,
+      "蛊": 76347,
+      "🎦": 76348,
+      "珡": 76349,
+      "👳‍♀️": 76350,
+      "掲": 76351,
+      "綰": 76352,
+      "礭": 76353,
+      "愓": 76354,
+      "掅": 76355,
+      "炟": 76356,
+      "毆": 76357,
+      "肸": 76358,
+      "蚎": 76359,
+      "霣": 76360,
+      "豤": 76361,
+      "👨‍⚖️": 76362,
+      "睺": 76363,
+      "窷": 76364,
+      "嫮": 76365,
+      "萕": 76366,
+      "嶛": 76367,
+      "ˉ": 76368,
+      "鲷": 76369,
+      "渫": 76370,
+      "😑": 76371,
+      "΢": 76372,
+      "鈰": 76373,
+      "腉": 76374,
+      "👩🏼‍❤️‍👩🏾": 76375,
+      "⋪": 76376,
+      "悽": 76377,
+      "Ƚ": 76378,
+      "脴": 76379,
+      "韼": 76380,
+      "薻": 76381,
+      "訮": 76382,
+      "毊": 76383,
+      "笥": 76384,
+      "🫅🏾": 76385,
+      "蜵": 76386,
+      "枆": 76387,
+      "珴": 76388,
+      "肮": 76389,
+      "騜": 76390,
+      "狖": 76391,
+      "碽": 76392,
+      "璂": 76393,
+      "襁": 76394,
+      "⇞": 76395,
+      "🧜🏽‍♂️": 76396,
+      "俼": 76397,
+      "絇": 76398,
+      "輖": 76399,
+      "⛗": 76400,
+      "碠": 76401,
+      "⍻": 76402,
+      "翄": 76403,
+      "牉": 76404,
+      "🚵🏻": 76405,
+      "塎": 76406,
+      "狃": 76407,
+      "栺": 76408,
+      "剡": 76409,
+      "叵": 76410,
+      "💢": 76411,
+      "槴": 76412,
+      "劮": 76413,
+      "珦": 76414,
+      "⓶": 76415,
+      "🫲": 76416,
+      "乴": 76417,
+      "戯": 76418,
+      "ϟ": 76419,
+      "瀩": 76420,
+      "❄️": 76421,
+      "⋐": 76422,
+      "廈": 76423,
+      "哬": 76424,
+      "Ϯ": 76425,
+      "啯": 76426,
+      "👂🏽": 76427,
+      "頝": 76428,
+      "磴": 76429,
+      "貗": 76430,
+      "橨": 76431,
+      "罓": 76432,
+      "🇰🇳": 76433,
+      "ₐ": 76434,
+      "廧": 76435,
+      "龑": 76436,
+      "Ⓐ": 76437,
+      "鴂": 76438,
+      "軾": 76439,
+      "嫠": 76440,
+      "潃": 76441,
+      "礏": 76442,
+      "崕": 76443,
+      "菺": 76444,
+      "🙌🏼": 76445,
+      "虊": 76446,
+      "👨🏾‍❤️‍👨🏽": 76447,
+      "騕": 76448,
+      "縭": 76449,
+      "弅": 76450,
+      "犔": 76451,
+      "鼣": 76452,
+      "鋨": 76453,
+      "🥝": 76454,
+      "腸": 76455,
+      "穞": 76456,
+      "ΐ": 76457,
+      "怄": 76458,
+      "虨": 76459,
+      "铼": 76460,
+      "呷": 76461,
+      "撙": 76462,
+      "羅": 76463,
+      "嵹": 76464,
+      "篝": 76465,
+      "蹓": 76466,
+      "🇧🇹": 76467,
+      "寴": 76468,
+      "谕": 76469,
+      "摭": 76470,
+      "酂": 76471,
+      "芞": 76472,
+      "兾": 76473,
+      "陬": 76474,
+      "硻": 76475,
+      "遶": 76476,
+      "烆": 76477,
+      "褛": 76478,
+      "嶖": 76479,
+      "拞": 76480,
+      "🚱": 76481,
+      "🦯": 76482,
+      "🤘🏾": 76483,
+      "👩🏾‍🦰": 76484,
+      "韡": 76485,
+      "読": 76486,
+      "偗": 76487,
+      "⓼": 76488,
+      "耟": 76489,
+      "☩": 76490,
+      "遖": 76491,
+      "寜": 76492,
+      "婤": 76493,
+      "搤": 76494,
+      "鼫": 76495,
+      "礛": 76496,
+      "苊": 76497,
+      "旓": 76498,
+      "橏": 76499,
+      "薡": 76500,
+      "灛": 76501,
+      "圗": 76502,
+      "揗": 76503,
+      "翐": 76504,
+      "郅": 76505,
+      "🇧🇭": 76506,
+      "␑": 76507,
+      "氕": 76508,
+      "␼": 76509,
+      "ŋ": 76510,
+      "啩": 76511,
+      "玬": 76512,
+      "瓓": 76513,
+      "苜": 76514,
+      "魭": 76515,
+      "頶": 76516,
+      "鐱": 76517,
+      "嘙": 76518,
+      "餀": 76519,
+      "🦶🏿": 76520,
+      "耚": 76521,
+      "驀": 76522,
+      "凅": 76523,
+      "郒": 76524,
+      "🇸🇱": 76525,
+      "埵": 76526,
+      "阨": 76527,
+      "樾": 76528,
+      "芤": 76529,
+      "💮": 76530,
+      "竲": 76531,
+      "淙": 76532,
+      "🫅🏿": 76533,
+      "虜": 76534,
+      "蝣": 76535,
+      "👨🏼‍✈️": 76536,
+      "訃": 76537,
+      "霕": 76538,
+      "啎": 76539,
+      "絓": 76540,
+      "🍊": 76541,
+      "掫": 76542,
+      "鱾": 76543,
+      "寙": 76544,
+      "簎": 76545,
+      "⁾": 76546,
+      "麵": 76547,
+      "⋤": 76548,
+      "襡": 76549,
+      "🦶🏽": 76550,
+      "剏": 76551,
+      "芾": 76552,
+      "🤌🏻": 76553,
+      "梎": 76554,
+      "毞": 76555,
+      "ͬ": 76556,
+      "筊": 76557,
+      "猆": 76558,
+      "💭": 76559,
+      "镒": 76560,
+      "菧": 76561,
+      "岢": 76562,
+      "獗": 76563,
+      "橁": 76564,
+      "乻": 76565,
+      "寁": 76566,
+      "刕": 76567,
+      "🦹🏾": 76568,
+      "鸷": 76569,
+      "媅": 76570,
+      "蒦": 76571,
+      "斷": 76572,
+      "⋳": 76573,
+      "榼": 76574,
+      "偢": 76575,
+      "☺": 76576,
+      "😒": 76577,
+      "🧚🏽‍♂️": 76578,
+      "螠": 76579,
+      "蹣": 76580,
+      "⃷": 76581,
+      "⛻": 76582,
+      "┕": 76583,
+      "Ɔ": 76584,
+      "齖": 76585,
+      "👩‍💼": 76586,
+      "🩹": 76587,
+      "≖": 76588,
+      "鼴": 76589,
+      "隈": 76590,
+      "🧟": 76591,
+      "≝": 76592,
+      "寕": 76593,
+      "鑝": 76594,
+      "棥": 76595,
+      "掇": 76596,
+      "❗️": 76597,
+      "☙": 76598,
+      "柲": 76599,
+      "鉭": 76600,
+      "蛔": 76601,
+      "膷": 76602,
+      "🇻🇦": 76603,
+      "ʕ": 76604,
+      "翀": 76605,
+      "胝": 76606,
+      "👶🏼": 76607,
+      "罅": 76608,
+      "跛": 76609,
+      "⁂": 76610,
+      "勮": 76611,
+      "錃": 76612,
+      "厠": 76613,
+      "嵑": 76614,
+      "👨🏼‍🔬": 76615,
+      "鷉": 76616,
+      "顗": 76617,
+      "塗": 76618,
+      "郔": 76619,
+      "敩": 76620,
+      "⌶": 76621,
+      "颋": 76622,
+      "鍫": 76623,
+      "蒗": 76624,
+      "👩🏼‍🦽": 76625,
+      "踂": 76626,
+      "蜹": 76627,
+      "鏉": 76628,
+      "嫄": 76629,
+      "踖": 76630,
+      "堢": 76631,
+      "镌": 76632,
+      "庘": 76633,
+      "⁗": 76634,
+      "⑼": 76635,
+      "搻": 76636,
+      "🐞": 76637,
+      "樃": 76638,
+      "簗": 76639,
+      "庌": 76640,
+      "睳": 76641,
+      "噡": 76642,
+      "🇰🇷": 76643,
+      "駴": 76644,
+      "⌐": 76645,
+      "▣": 76646,
+      "煁": 76647,
+      "Ŵ": 76648,
+      "苴": 76649,
+      "閱": 76650,
+      "仱": 76651,
+      "🛑": 76652,
+      "殶": 76653,
+      "⌦": 76654,
+      "硐": 76655,
+      "鯯": 76656,
+      "🇮🇱": 76657,
+      "鷗": 76658,
+      "噮": 76659,
+      "峁": 76660,
+      "缾": 76661,
+      "🪞": 76662,
+      "⍪": 76663,
+      "Ɩ": 76664,
+      "◍": 76665,
+      "🤸‍♀️": 76666,
+      "荁": 76667,
+      "籴": 76668,
+      "誊": 76669,
+      "巖": 76670,
+      "镚": 76671,
+      "艛": 76672,
+      "帩": 76673,
+      "弞": 76674,
+      "嘥": 76675,
+      "📬": 76676,
+      "馍": 76677,
+      "媱": 76678,
+      "🧳": 76679,
+      "羴": 76680,
+      "8️⃣": 76681,
+      "鄡": 76682,
+      "👩🏼‍🎤": 76683,
+      "蝅": 76684,
+      "狣": 76685,
+      "鄃": 76686,
+      "晙": 76687,
+      "欹": 76688,
+      "飡": 76689,
+      "乕": 76690,
+      "🏥": 76691,
+      "靹": 76692,
+      "氭": 76693,
+      "鳾": 76694,
+      "虝": 76695,
+      "諆": 76696,
+      "軮": 76697,
+      "孮": 76698,
+      "薸": 76699,
+      "🦹🏻": 76700,
+      "瀆": 76701,
+      "鴎": 76702,
+      "訝": 76703,
+      "鎿": 76704,
+      "窚": 76705,
+      "舡": 76706,
+      "禣": 76707,
+      "瞃": 76708,
+      "噈": 76709,
+      "蜱": 76710,
+      "娲": 76711,
+      "毠": 76712,
+      "浽": 76713,
+      "顴": 76714,
+      "欰": 76715,
+      "呲": 76716,
+      "涮": 76717,
+      "ʸ": 76718,
+      "繇": 76719,
+      "褻": 76720,
+      "⁣": 76721,
+      "訜": 76722,
+      "汒": 76723,
+      "🫚": 76724,
+      "硹": 76725,
+      "壻": 76726,
+      "莕": 76727,
+      "殢": 76728,
+      "👩🏾‍⚕️": 76729,
+      "👥": 76730,
+      "🇬🇱": 76731,
+      "⁌": 76732,
+      "雗": 76733,
+      "狞": 76734,
+      "冎": 76735,
+      "欥": 76736,
+      "溄": 76737,
+      "譌": 76738,
+      "伉": 76739,
+      "烀": 76740,
+      "媿": 76741,
+      "佝": 76742,
+      "🧀": 76743,
+      "嶉": 76744,
+      "瓄": 76745,
+      "镩": 76746,
+      "⃇": 76747,
+      "攁": 76748,
+      "笰": 76749,
+      "🇱🇮": 76750,
+      "豸": 76751,
+      "駿": 76752,
+      "頩": 76753,
+      "🍲": 76754,
+      "璩": 76755,
+      "👩🏿‍❤️‍👩🏾": 76756,
+      "👩🏼‍❤️‍👨🏿": 76757,
+      "牳": 76758,
+      "颏": 76759,
+      "婗": 76760,
+      "噧": 76761,
+      "迺": 76762,
+      "嶾": 76763,
+      "🎠": 76764,
+      "轊": 76765,
+      "☱": 76766,
+      "婹": 76767,
+      "韪": 76768,
+      "埧": 76769,
+      "榙": 76770,
+      "襯": 76771,
+      "筶": 76772,
+      "爧": 76773,
+      "畨": 76774,
+      "稝": 76775,
+      "ⓝ": 76776,
+      "淭": 76777,
+      "Η": 76778,
+      "🐔": 76779,
+      "褌": 76780,
+      "蕆": 76781,
+      "鋦": 76782,
+      "ⅈ": 76783,
+      "🔊": 76784,
+      "┲": 76785,
+      "▜": 76786,
+      "羜": 76787,
+      "🇦🇪": 76788,
+      "℺": 76789,
+      "唒": 76790,
+      "勔": 76791,
+      "飜": 76792,
+      "焅": 76793,
+      "軎": 76794,
+      "⁴": 76795,
+      "謸": 76796,
+      "憸": 76797,
+      "🧑‍🍳": 76798,
+      "姏": 76799,
+      "🫴🏿": 76800,
+      "👳🏽": 76801,
+      "鋴": 76802,
+      "償": 76803,
+      "⅐": 76804,
+      "猟": 76805,
+      "韖": 76806,
+      "筢": 76807,
+      "🤙🏼": 76808,
+      "儊": 76809,
+      "🧗🏽‍♂️": 76810,
+      "驽": 76811,
+      "␷": 76812,
+      "痀": 76813,
+      "犮": 76814,
+      "芵": 76815,
+      "襨": 76816,
+      "嘠": 76817,
+      "婎": 76818,
+      "聟": 76819,
+      "睻": 76820,
+      "呌": 76821,
+      "嬿": 76822,
+      "昁": 76823,
+      "骬": 76824,
+      "廁": 76825,
+      "骩": 76826,
+      "": 76827,
+      "凪": 76828,
+      "杸": 76829,
+      "奅": 76830,
+      "嚉": 76831,
+      "荩": 76832,
+      "鄿": 76833,
+      "̋": 76834,
+      "妧": 76835,
+      "檺": 76836,
+      "瀙": 76837,
+      "⋣": 76838,
+      "∻": 76839,
+      "▙": 76840,
+      "Ŭ": 76841,
+      "┪": 76842,
+      "🫡": 76843,
+      "輎": 76844,
+      "潯": 76845,
+      "枷": 76846,
+      "🇺🇳": 76847,
+      "澞": 76848,
+      "👩🏼‍🦼": 76849,
+      "黻": 76850,
+      "懢": 76851,
+      "鄀": 76852,
+      "崯": 76853,
+      "彿": 76854,
+      "📝": 76855,
+      "ℜ": 76856,
+      "鷋": 76857,
+      "踅": 76858,
+      "摣": 76859,
+      "菽": 76860,
+      "鄅": 76861,
+      "🧃": 76862,
+      "🧨": 76863,
+      "闚": 76864,
+      "绨": 76865,
+      "荋": 76866,
+      "逶": 76867,
+      "顸": 76868,
+      "Ϡ": 76869,
+      "钭": 76870,
+      "譫": 76871,
+      "拵": 76872,
+      "眃": 76873,
+      "臉": 76874,
+      "蹘": 76875,
+      "贗": 76876,
+      "椫": 76877,
+      "拗": 76878,
+      "👩🏿‍🎨": 76879,
+      "筬": 76880,
+      "Ɉ": 76881,
+      "嗙": 76882,
+      "┑": 76883,
+      "峕": 76884,
+      "廟": 76885,
+      "箁": 76886,
+      "熒": 76887,
+      "弬": 76888,
+      "陘": 76889,
+      "飲": 76890,
+      "潅": 76891,
+      "襺": 76892,
+      "塂": 76893,
+      "齦": 76894,
+      "唃": 76895,
+      "Ϣ": 76896,
+      "煘": 76897,
+      "┐": 76898,
+      "騊": 76899,
+      "鷟": 76900,
+      "瘻": 76901,
+      "儍": 76902,
+      "🇲🇩": 76903,
+      "窋": 76904,
+      "嵕": 76905,
+      "鹝": 76906,
+      "潣": 76907,
+      "鲦": 76908,
+      "⇚": 76909,
+      "巁": 76910,
+      "悡": 76911,
+      "劻": 76912,
+      "🙎‍♂️": 76913,
+      "肔": 76914,
+      "侚": 76915,
+      "綏": 76916,
+      "眢": 76917,
+      "🫴🏾": 76918,
+      "龣": 76919,
+      "謚": 76920,
+      "₮": 76921,
+      "🚨": 76922,
+      "贜": 76923,
+      "瘑": 76924,
+      "鱋": 76925,
+      "鰬": 76926,
+      "謅": 76927,
+      "🤌": 76928,
+      "馩": 76929,
+      "騄": 76930,
+      "🤦🏾‍♀️": 76931,
+      "┨": 76932,
+      "従": 76933,
+      "🪭": 76934,
+      "琩": 76935,
+      "羰": 76936,
+      "翮": 76937,
+      "鶐": 76938,
+      "瓱": 76939,
+      "肷": 76940,
+      "🦮": 76941,
+      "滓": 76942,
+      "👨🏾‍⚕️": 76943,
+      "暿": 76944,
+      "鱹": 76945,
+      "蛑": 76946,
+      "戬": 76947,
+      "撔": 76948,
+      "簸": 76949,
+      "栛": 76950,
+      "聎": 76951,
+      "愃": 76952,
+      "豳": 76953,
+      "菞": 76954,
+      "脝": 76955,
+      "嗐": 76956,
+      "ɺ": 76957,
+      "覶": 76958,
+      "岎": 76959,
+      "緓": 76960,
+      "😸": 76961,
+      "💇": 76962,
+      "楅": 76963,
+      "鹘": 76964,
+      "箧": 76965,
+      "閝": 76966,
+      "憏": 76967,
+      "↋": 76968,
+      "寶": 76969,
+      "🏊🏾‍♂️": 76970,
+      "撣": 76971,
+      "伷": 76972,
+      "蚆": 76973,
+      "🤴🏿": 76974,
+      "磝": 76975,
+      "酾": 76976,
+      "≄": 76977,
+      "擿": 76978,
+      "鮗": 76979,
+      "踷": 76980,
+      "℄": 76981,
+      "˧": 76982,
+      "👨🏻‍❤️‍👨🏾": 76983,
+      "俑": 76984,
+      "紃": 76985,
+      "鋸": 76986,
+      "蔱": 76987,
+      "⏾": 76988,
+      "🎨": 76989,
+      "覉": 76990,
+      "禰": 76991,
+      "麉": 76992,
+      "⃖": 76993,
+      "歅": 76994,
+      "亀": 76995,
+      "讵": 76996,
+      "椂": 76997,
+      "钌": 76998,
+      "濃": 76999,
+      "彚": 77000,
+      "鵸": 77001,
+      "吺": 77002,
+      "⅂": 77003,
+      "🌴": 77004,
+      "倝": 77005,
+      "鳠": 77006,
+      "颔": 77007,
+      "簃": 77008,
+      "慄": 77009,
+      "萋": 77010,
+      "焇": 77011,
+      "觝": 77012,
+      "啾": 77013,
+      "👨🏻‍🏫": 77014,
+      "鳵": 77015,
+      "怛": 77016,
+      "軅": 77017,
+      "🧘🏼‍♀️": 77018,
+      "Ǻ": 77019,
+      "Ǽ": 77020,
+      "缲": 77021,
+      "蔥": 77022,
+      "衠": 77023,
+      "氣": 77024,
+      "躞": 77025,
+      "蛕": 77026,
+      "煕": 77027,
+      "袃": 77028,
+      "驂": 77029,
+      "黨": 77030,
+      "裱": 77031,
+      "⎲": 77032,
+      "抙": 77033,
+      "厈": 77034,
+      "烏": 77035,
+      "鮪": 77036,
+      "黳": 77037,
+      "黉": 77038,
+      "鑀": 77039,
+      "賽": 77040,
+      "懍": 77041,
+      "📶": 77042,
+      "🇷🇺": 77043,
+      "垘": 77044,
+      "揤": 77045,
+      "璆": 77046,
+      "👩🏾‍❤️‍👨🏻": 77047,
+      "趚": 77048,
+      "ⓩ": 77049,
+      "褈": 77050,
+      "嶤": 77051,
+      "蛄": 77052,
+      "膎": 77053,
+      "姡": 77054,
+      "ͅ": 77055,
+      "Ȼ": 77056,
+      "隖": 77057,
+      "稙": 77058,
+      "歄": 77059,
+      "鍐": 77060,
+      "璫": 77061,
+      "➕": 77062,
+      "👩🏻‍💼": 77063,
+      "芅": 77064,
+      "爎": 77065,
+      "膪": 77066,
+      "銊": 77067,
+      "萯": 77068,
+      "恧": 77069,
+      "Ì": 77070,
+      "顼": 77071,
+      "熝": 77072,
+      "弽": 77073,
+      "釟": 77074,
+      "餋": 77075,
+      "藙": 77076,
+      "👩🏻‍❤️‍👩🏿": 77077,
+      "蕭": 77078,
+      "蝦": 77079,
+      "🇦🇶": 77080,
+      "譆": 77081,
+      "🚶🏿‍♂️": 77082,
+      "搧": 77083,
+      "氷": 77084,
+      "隄": 77085,
+      "鎪": 77086,
+      "🥃": 77087,
+      "耥": 77088,
+      "綋": 77089,
+      "矲": 77090,
+      "鸜": 77091,
+      "👨‍👦": 77092,
+      "溁": 77093,
+      "雠": 77094,
+      "剓": 77095,
+      "裲": 77096,
+      "橢": 77097,
+      "鰩": 77098,
+      "覑": 77099,
+      "蔖": 77100,
+      "僛": 77101,
+      "蝧": 77102,
+      "窐": 77103,
+      "ƥ": 77104,
+      "⑲": 77105,
+      "咎": 77106,
+      "⛃": 77107,
+      "褦": 77108,
+      "閪": 77109,
+      "隦": 77110,
+      "⑔": 77111,
+      "釬": 77112,
+      "⒚": 77113,
+      "骵": 77114,
+      "🤴🏼": 77115,
+      "♋": 77116,
+      "貼": 77117,
+      "👨": 77118,
+      "偾": 77119,
+      "驋": 77120,
+      "騫": 77121,
+      "熇": 77122,
+      "煂": 77123,
+      "釅": 77124,
+      "邍": 77125,
+      "⛲": 77126,
+      "Ϥ": 77127,
+      "贶": 77128,
+      "楆": 77129,
+      "⌖": 77130,
+      "甊": 77131,
+      "鹥": 77132,
+      "敘": 77133,
+      "🚬": 77134,
+      "皅": 77135,
+      "赺": 77136,
+      "駧": 77137,
+      "珚": 77138,
+      "獪": 77139,
+      "箊": 77140,
+      "鎞": 77141,
+      "Ϗ": 77142,
+      "薉": 77143,
+      "🙋🏼‍♀️": 77144,
+      "唢": 77145,
+      "⊐": 77146,
+      "绹": 77147,
+      "璦": 77148,
+      "翲": 77149,
+      "ℌ": 77150,
+      "唽": 77151,
+      "僶": 77152,
+      "稞": 77153,
+      "⛂": 77154,
+      "槌": 77155,
+      "梏": 77156,
+      "挸": 77157,
+      "攥": 77158,
+      "🌃": 77159,
+      "鬹": 77160,
+      "🇬🇭": 77161,
+      "葎": 77162,
+      "瓻": 77163,
+      "蟒": 77164,
+      "灂": 77165,
+      "緥": 77166,
+      "🦹🏽‍♂️": 77167,
+      "雱": 77168,
+      "眖": 77169,
+      "湺": 77170,
+      "🦹🏾‍♂️": 77171,
+      "僾": 77172,
+      "趱": 77173,
+      "摲": 77174,
+      "鏱": 77175,
+      "牭": 77176,
+      "🥷🏻": 77177,
+      "獔": 77178,
+      "眴": 77179,
+      "獹": 77180,
+      "鵛": 77181,
+      "⓹": 77182,
+      "Ｒ": 77183,
+      "嬝": 77184,
+      "⋢": 77185,
+      "茽": 77186,
+      "⛸": 77187,
+      "氝": 77188,
+      "姍": 77189,
+      "犵": 77190,
+      "勊": 77191,
+      "🕵🏿‍♂️": 77192,
+      "🇸🇮": 77193,
+      "騶": 77194,
+      "纊": 77195,
+      "溰": 77196,
+      "蓀": 77197,
+      "🛌🏾": 77198,
+      "帇": 77199,
+      "讒": 77200,
+      "⒋": 77201,
+      "惻": 77202,
+      "礊": 77203,
+      "揟": 77204,
+      "🔤": 77205,
+      "惈": 77206,
+      "🙎🏾": 77207,
+      "⎤": 77208,
+      "🧚🏿‍♂️": 77209,
+      "👄": 77210,
+      "剼": 77211,
+      "邩": 77212,
+      "ƭ": 77213,
+      "瓅": 77214,
+      "鲮": 77215,
+      "塃": 77216,
+      "甽": 77217,
+      "🙇🏾‍♀️": 77218,
+      "梔": 77219,
+      "囫": 77220,
+      "髌": 77221,
+      "伿": 77222,
+      "硔": 77223,
+      "⒢": 77224,
+      "ⅆ": 77225,
+      "👩🏻‍🏭": 77226,
+      "聐": 77227,
+      "跖": 77228,
+      "扟": 77229,
+      "僭": 77230,
+      "噵": 77231,
+      "鵠": 77232,
+      "倧": 77233,
+      "鋰": 77234,
+      "⎦": 77235,
+      "呁": 77236,
+      "▂": 77237,
+      "⋡": 77238,
+      "芊": 77239,
+      "礗": 77240,
+      "≎": 77241,
+      "╯": 77242,
+      "✥": 77243,
+      "隨": 77244,
+      "慞": 77245,
+      "昘": 77246,
+      "蘛": 77247,
+      "◤": 77248,
+      "徆": 77249,
+      "⛒": 77250,
+      "溂": 77251,
+      "裈": 77252,
+      "洐": 77253,
+      "琀": 77254,
+      "贁": 77255,
+      "朒": 77256,
+      "懭": 77257,
+      "浜": 77258,
+      "亘": 77259,
+      "虣": 77260,
+      "鋺": 77261,
+      "譍": 77262,
+      "繕": 77263,
+      "渾": 77264,
+      "聒": 77265,
+      "⋦": 77266,
+      "🤚🏿": 77267,
+      "豶": 77268,
+      "踮": 77269,
+      "漗": 77270,
+      "遄": 77271,
+      "苬": 77272,
+      "靚": 77273,
+      "鈈": 77274,
+      "📲": 77275,
+      "癟": 77276,
+      "⏿": 77277,
+      "🤝🏻": 77278,
+      "鲊": 77279,
+      "氐": 77280,
+      "棔": 77281,
+      "🏋🏾‍♀️": 77282,
+      "犇": 77283,
+      "👨🏼‍🏭": 77284,
+      "哘": 77285,
+      "酫": 77286,
+      "稔": 77287,
+      "鹂": 77288,
+      "✯": 77289,
+      "塪": 77290,
+      "☥": 77291,
+      "🚷": 77292,
+      "钐": 77293,
+      "🚖": 77294,
+      "嬁": 77295,
+      "🌈": 77296,
+      "洤": 77297,
+      "Ő": 77298,
+      "♥︎": 77299,
+      "⚬": 77300,
+      "鶸": 77301,
+      "⎴": 77302,
+      "茕": 77303,
+      "拪": 77304,
+      "␧": 77305,
+      "莂": 77306,
+      "峂": 77307,
+      "謰": 77308,
+      "锇": 77309,
+      "鵨": 77310,
+      "暛": 77311,
+      "湜": 77312,
+      "👭🏿": 77313,
+      "斉": 77314,
+      "熆": 77315,
+      "秳": 77316,
+      "竇": 77317,
+      "鏬": 77318,
+      "姌": 77319,
+      "仭": 77320,
+      "鑓": 77321,
+      "⍢": 77322,
+      "莄": 77323,
+      "蝈": 77324,
+      "亸": 77325,
+      "宥": 77326,
+      "👋🏼": 77327,
+      "👩‍👩‍👧‍👦": 77328,
+      "梮": 77329,
+      "碂": 77330,
+      "疭": 77331,
+      "⑄": 77332,
+      "禬": 77333,
+      "⒅": 77334,
+      "阬": 77335,
+      "侏": 77336,
+      "牐": 77337,
+      "碊": 77338,
+      "鞌": 77339,
+      "癦": 77340,
+      "疰": 77341,
+      "獰": 77342,
+      "侼": 77343,
+      "👨🏽‍💻": 77344,
+      "🌛": 77345,
+      "飸": 77346,
+      "渀": 77347,
+      "鹀": 77348,
+      "閷": 77349,
+      "檊": 77350,
+      "馎": 77351,
+      "ⓘ": 77352,
+      "闡": 77353,
+      "扌": 77354,
+      "蒑": 77355,
+      "🧔🏻‍♂️": 77356,
+      "脃": 77357,
+      "ϧ": 77358,
+      "⑎": 77359,
+      "毥": 77360,
+      "鞃": 77361,
+      "粏": 77362,
+      "͕": 77363,
+      "钏": 77364,
+      "拻": 77365,
+      "😫": 77366,
+      "吲": 77367,
+      "茁": 77368,
+      "🇹🇴": 77369,
+      "⏌": 77370,
+      "犈": 77371,
+      "鹜": 77372,
+      "💁": 77373,
+      "謈": 77374,
+      "栂": 77375,
+      "碻": 77376,
+      "鞝": 77377,
+      "嵞": 77378,
+      "廮": 77379,
+      "苁": 77380,
+      "韒": 77381,
+      "屼": 77382,
+      "蛅": 77383,
+      "虓": 77384,
+      "綅": 77385,
+      "　": 77386,
+      "👨🏿‍⚕️": 77387,
+      "阄": 77388,
+      "🧏‍♀️": 77389,
+      "苽": 77390,
+      "蔩": 77391,
+      "摰": 77392,
+      "螜": 77393,
+      "髈": 77394,
+      "👧🏽": 77395,
+      "呸": 77396,
+      "蚖": 77397,
+      "🖖": 77398,
+      "牚": 77399,
+      "驓": 77400,
+      "脁": 77401,
+      "絡": 77402,
+      "詶": 77403,
+      "恘": 77404,
+      "颃": 77405,
+      "🤝🏿": 77406,
+      "苿": 77407,
+      "姫": 77408,
+      "髂": 77409,
+      "ⓨ": 77410,
+      "🤷🏾": 77411,
+      "榟": 77412,
+      "砗": 77413,
+      "鬯": 77414,
+      "墮": 77415,
+      "磳": 77416,
+      "💳": 77417,
+      "闉": 77418,
+      "😡": 77419,
+      "🤴🏾": 77420,
+      "帬": 77421,
+      "薲": 77422,
+      "🆓": 77423,
+      "髭": 77424,
+      "👩🏼‍⚕️": 77425,
+      "🧑‍⚕️": 77426,
+      "👱🏾": 77427,
+      "畍": 77428,
+      "瞯": 77429,
+      "葠": 77430,
+      "🥑": 77431,
+      "縊": 77432,
+      "颙": 77433,
+      "點": 77434,
+      "🧑🏿‍❤️‍💋‍🧑🏻": 77435,
+      "ϩ": 77436,
+      "茇": 77437,
+      "涍": 77438,
+      "洷": 77439,
+      "獯": 77440,
+      "🎻": 77441,
+      "麑": 77442,
+      "蚫": 77443,
+      "嗴": 77444,
+      "Ń": 77445,
+      "骰": 77446,
+      "🧑🏼‍✈️": 77447,
+      "倅": 77448,
+      "贌": 77449,
+      "痂": 77450,
+      "⒞": 77451,
+      "♩": 77452,
+      "泝": 77453,
+      "莭": 77454,
+      "氂": 77455,
+      "🧙🏼‍♂️": 77456,
+      "齰": 77457,
+      "儱": 77458,
+      "鎉": 77459,
+      "鳹": 77460,
+      "蓱": 77461,
+      "盃": 77462,
+      "蟩": 77463,
+      "铧": 77464,
+      "炞": 77465,
+      "墢": 77466,
+      "🧑🏻‍🏭": 77467,
+      "🧅": 77468,
+      "🇧🇯": 77469,
+      "仢": 77470,
+      "🫵🏻": 77471,
+      "荓": 77472,
+      "̤": 77473,
+      "殅": 77474,
+      "鮨": 77475,
+      "驑": 77476,
+      "翋": 77477,
+      "⚓": 77478,
+      "辭": 77479,
+      "馚": 77480,
+      "茱": 77481,
+      "硳": 77482,
+      "📖": 77483,
+      "戢": 77484,
+      "漄": 77485,
+      "⊾": 77486,
+      "褜": 77487,
+      "🐋": 77488,
+      "👨🏻‍🚀": 77489,
+      "爢": 77490,
+      "Ⓥ": 77491,
+      "墔": 77492,
+      "dz": 77493,
+      "緜": 77494,
+      "嚫": 77495,
+      "🦹🏾‍♀️": 77496,
+      "鏈": 77497,
+      "ɦ": 77498,
+      "¤": 77499,
+      "龌": 77500,
+      "匝": 77501,
+      "˙": 77502,
+      "涷": 77503,
+      "攏": 77504,
+      "縑": 77505,
+      "🙇🏼‍♂️": 77506,
+      "籕": 77507,
+      "粺": 77508,
+      "⓿": 77509,
+      "鐋": 77510,
+      "親": 77511,
+      "溸": 77512,
+      "媁": 77513,
+      "嚜": 77514,
+      "┒": 77515,
+      "撂": 77516,
+      "鐏": 77517,
+      "讅": 77518,
+      "🩰": 77519,
+      "椛": 77520,
+      "👹": 77521,
+      "蠔": 77522,
+      "⇌": 77523,
+      "澌": 77524,
+      "藵": 77525,
+      "👈🏿": 77526,
+      "麩": 77527,
+      "勲": 77528,
+      "囿": 77529,
+      "俜": 77530,
+      "╻": 77531,
+      "鮐": 77532,
+      "淜": 77533,
+      "⇯": 77534,
+      "ₖ": 77535,
+      "鱕": 77536,
+      "蟡": 77537,
+      "₡": 77538,
+      "👩🏿‍🦽": 77539,
+      "🛝": 77540,
+      "🧑🏼‍❤️‍💋‍🧑🏽": 77541,
+      "🏬": 77542,
+      "拃": 77543,
+      "釶": 77544,
+      "抡": 77545,
+      "靂": 77546,
+      "趽": 77547,
+      "欵": 77548,
+      "堗": 77549,
+      "紷": 77550,
+      "槠": 77551,
+      "巓": 77552,
+      "硉": 77553,
+      "螹": 77554,
+      "̊": 77555,
+      "笈": 77556,
+      "🥭": 77557,
+      "龊": 77558,
+      "頵": 77559,
+      "坜": 77560,
+      "薑": 77561,
+      "裣": 77562,
+      "刏": 77563,
+      "梣": 77564,
+      "謲": 77565,
+      "幓": 77566,
+      "筸": 77567,
+      "┧": 77568,
+      "谀": 77569,
+      "覮": 77570,
+      "觍": 77571,
+      "璗": 77572,
+      "桎": 77573,
+      "墾": 77574,
+      "鯼": 77575,
+      "À": 77576,
+      "冸": 77577,
+      "嶐": 77578,
+      "逑": 77579,
+      "👧🏾": 77580,
+      "✨": 77581,
+      "🌊": 77582,
+      "乷": 77583,
+      "孛": 77584,
+      "劯": 77585,
+      "涻": 77586,
+      "鋂": 77587,
+      "鍘": 77588,
+      "璝": 77589,
+      "⇵": 77590,
+      "瓆": 77591,
+      "鲝": 77592,
+      "墘": 77593,
+      "哝": 77594,
+      "兘": 77595,
+      "煥": 77596,
+      "👨🏽‍❤️‍👨🏾": 77597,
+      "傤": 77598,
+      "捃": 77599,
+      "條": 77600,
+      "梼": 77601,
+      "忔": 77602,
+      "缶": 77603,
+      "偖": 77604,
+      "稃": 77605,
+      "皘": 77606,
+      "辦": 77607,
+      "溞": 77608,
+      "苳": 77609,
+      "胴": 77610,
+      "孂": 77611,
+      "鉚": 77612,
+      "暁": 77613,
+      "餴": 77614,
+      "⚄": 77615,
+      "磂": 77616,
+      "縜": 77617,
+      "Ȧ": 77618,
+      "弡": 77619,
+      "淽": 77620,
+      "賎": 77621,
+      "嬢": 77622,
+      "餲": 77623,
+      "🙆🏼‍♀️": 77624,
+      "🤝🏽": 77625,
+      "翥": 77626,
+      "箘": 77627,
+      "鰠": 77628,
+      "鷻": 77629,
+      "躆": 77630,
+      "篘": 77631,
+      "頗": 77632,
+      "輿": 77633,
+      "🤝🏾": 77634,
+      "摐": 77635,
+      "🧮": 77636,
+      "👩🏿‍❤️‍💋‍👨🏿": 77637,
+      "🧚🏾": 77638,
+      "遝": 77639,
+      "嘵": 77640,
+      "Ÿ": 77641,
+      "濾": 77642,
+      "輘": 77643,
+      "虆": 77644,
+      "噱": 77645,
+      "劖": 77646,
+      "籤": 77647,
+      "🐘": 77648,
+      "籏": 77649,
+      "棩": 77650,
+      "嚪": 77651,
+      "鴙": 77652,
+      "🧖🏽‍♂️": 77653,
+      "筎": 77654,
+      "餬": 77655,
+      "Ϳ": 77656,
+      "蚬": 77657,
+      "鈚": 77658,
+      "鏃": 77659,
+      "髖": 77660,
+      "蜙": 77661,
+      "騾": 77662,
+      "揬": 77663,
+      "韱": 77664,
+      "憥": 77665,
+      "殰": 77666,
+      "瀫": 77667,
+      "婄": 77668,
+      "̼": 77669,
+      "👩🏿‍❤️‍👩🏻": 77670,
+      "⅃": 77671,
+      "鐸": 77672,
+      "◨": 77673,
+      "併": 77674,
+      "珕": 77675,
+      "🙇🏾": 77676,
+      "繦": 77677,
+      "🇻🇨": 77678,
+      "🚃": 77679,
+      "俿": 77680,
+      "氳": 77681,
+      "郗": 77682,
+      "呹": 77683,
+      "籭": 77684,
+      "🤽🏼‍♂️": 77685,
+      "鸌": 77686,
+      "崆": 77687,
+      "弐": 77688,
+      "縟": 77689,
+      "徯": 77690,
+      "豲": 77691,
+      "糁": 77692,
+      "🚵🏾‍♂️": 77693,
+      "🌎": 77694,
+      "👊🏽": 77695,
+      "🧔🏿‍♂️": 77696,
+      "褺": 77697,
+      "👩🏽‍❤️‍💋‍👩🏿": 77698,
+      "欀": 77699,
+      "🐊": 77700,
+      "貢": 77701,
+      "妗": 77702,
+      "瘣": 77703,
+      "俫": 77704,
+      "鸆": 77705,
+      "飑": 77706,
+      "倶": 77707,
+      "瀦": 77708,
+      "⚱": 77709,
+      "樋": 77710,
+      "葦": 77711,
+      "羕": 77712,
+      "懙": 77713,
+      "抪": 77714,
+      "倸": 77715,
+      "∍": 77716,
+      "鵖": 77717,
+      "疿": 77718,
+      "梛": 77719,
+      "祙": 77720,
+      "氹": 77721,
+      "◷": 77722,
+      "湝": 77723,
+      "渢": 77724,
+      "栐": 77725,
+      "≚": 77726,
+      "祱": 77727,
+      "🙇🏽": 77728,
+      "倂": 77729,
+      "馉": 77730,
+      "🧖🏾": 77731,
+      "箦": 77732,
+      "蹳": 77733,
+      "飈": 77734,
+      "蒨": 77735,
+      "杈": 77736,
+      "柪": 77737,
+      "🫛": 77738,
+      "贓": 77739,
+      "帀": 77740,
+      "粍": 77741,
+      "譝": 77742,
+      "篦": 77743,
+      "旯": 77744,
+      "🤹🏿‍♀️": 77745,
+      "頽": 77746,
+      "褓": 77747,
+      "轑": 77748,
+      "🕙": 77749,
+      "玆": 77750,
+      "頏": 77751,
+      "蚺": 77752,
+      "⛨": 77753,
+      "˵": 77754,
+      "👩🏻‍🍼": 77755,
+      "宬": 77756,
+      "🪈": 77757,
+      "∤": 77758,
+      "∌": 77759,
+      "檄": 77760,
+      "欌": 77761,
+      "桷": 77762,
+      "鏵": 77763,
+      "锲": 77764,
+      "Æ": 77765,
+      "経": 77766,
+      "诇": 77767,
+      "咝": 77768,
+      "鯟": 77769,
+      "遧": 77770,
+      "崲": 77771,
+      "⃟": 77772,
+      "╍": 77773,
+      "懮": 77774,
+      "嘚": 77775,
+      "彄": 77776,
+      "🧇": 77777,
+      "賅": 77778,
+      "沢": 77779,
+      "哱": 77780,
+      "糧": 77781,
+      "凔": 77782,
+      "膚": 77783,
+      "🟫": 77784,
+      "👵🏼": 77785,
+      "猒": 77786,
+      "弣": 77787,
+      "🇧🇷": 77788,
+      "毼": 77789,
+      "鵔": 77790,
+      "ʉ": 77791,
+      "垔": 77792,
+      "Ƃ": 77793,
+      "硲": 77794,
+      "🥎": 77795,
+      "幠": 77796,
+      "𓀝": 77797,
+      "洯": 77798,
+      "顠": 77799,
+      "磸": 77800,
+      "窹": 77801,
+      "🧑🏿": 77802,
+      "嚽": 77803,
+      "攐": 77804,
+      "轘": 77805,
+      "Ħ": 77806,
+      "璪": 77807,
+      "齋": 77808,
+      "茙": 77809,
+      "侊": 77810,
+      "🇲🇭": 77811,
+      "肊": 77812,
+      "黌": 77813,
+      "☽": 77814,
+      "瑺": 77815,
+      "滂": 77816,
+      "尜": 77817,
+      "硜": 77818,
+      "瞟": 77819,
+      "🧑🏼‍💻": 77820,
+      "⏼": 77821,
+      "Ȃ": 77822,
+      "蟵": 77823,
+      "℟": 77824,
+      "釹": 77825,
+      "ⅷ": 77826,
+      "ⓥ": 77827,
+      "蒪": 77828,
+      "檎": 77829,
+      "莯": 77830,
+      "湕": 77831,
+      "閽": 77832,
+      "珱": 77833,
+      "🧘🏼": 77834,
+      "渥": 77835,
+      "塀": 77836,
+      "斅": 77837,
+      "艥": 77838,
+      "⏦": 77839,
+      "澝": 77840,
+      "嫚": 77841,
+      "忉": 77842,
+      "閭": 77843,
+      "鲏": 77844,
+      "Ʉ": 77845,
+      "徇": 77846,
+      "⎰": 77847,
+      "🙆🏼": 77848,
+      "鳀": 77849,
+      "苘": 77850,
+      "羬": 77851,
+      "醥": 77852,
+      "匓": 77853,
+      "蚮": 77854,
+      "🏴‍☠️": 77855,
+      "岪": 77856,
+      "箒": 77857,
+      "谞": 77858,
+      "們": 77859,
+      "渞": 77860,
+      "裇": 77861,
+      "敀": 77862,
+      "蔳": 77863,
+      "櫛": 77864,
+      "綗": 77865,
+      "秹": 77866,
+      "溤": 77867,
+      "乜": 77868,
+      "瓛": 77869,
+      "儮": 77870,
+      "⏪": 77871,
+      "哴": 77872,
+      "⃉": 77873,
+      "Λ": 77874,
+      "烴": 77875,
+      "嬓": 77876,
+      "銩": 77877,
+      "烕": 77878,
+      "🩷": 77879,
+      "隡": 77880,
+      "鐼": 77881,
+      "🔨": 77882,
+      "逯": 77883,
+      "屶": 77884,
+      "洦": 77885,
+      "莾": 77886,
+      "👏🏻": 77887,
+      "茡": 77888,
+      "嵧": 77889,
+      "膰": 77890,
+      "🇧🇳": 77891,
+      "塴": 77892,
+      "弜": 77893,
+      "磉": 77894,
+      "轵": 77895,
+      "桮": 77896,
+      "鯴": 77897,
+      "🧎🏼": 77898,
+      "媀": 77899,
+      "屵": 77900,
+      "👩🏼‍❤️‍💋‍👩🏿": 77901,
+      "伡": 77902,
+      "🪰": 77903,
+      "秪": 77904,
+      "鰲": 77905,
+      "瞖": 77906,
+      "🌹": 77907,
+      "椱": 77908,
+      "裃": 77909,
+      "氲": 77910,
+      "∊": 77911,
+      "📜": 77912,
+      "說": 77913,
+      "嬩": 77914,
+      "濵": 77915,
+      "⍋": 77916,
+      "矋": 77917,
+      "杺": 77918,
+      "𓀌": 77919,
+      "堭": 77920,
+      "縮": 77921,
+      "慫": 77922,
+      "⌊": 77923,
+      "嫷": 77924,
+      "哌": 77925,
+      "ɶ": 77926,
+      "🤸🏾": 77927,
+      "跇": 77928,
+      "竔": 77929,
+      "΀": 77930,
+      "訨": 77931,
+      "🎑": 77932,
+      "鷩": 77933,
+      "‷": 77934,
+      "咓": 77935,
+      "荾": 77936,
+      "釡": 77937,
+      "馐": 77938,
+      "蹱": 77939,
+      "珘": 77940,
+      "洶": 77941,
+      "輦": 77942,
+      "疕": 77943,
+      "叚": 77944,
+      "吤": 77945,
+      "龝": 77946,
+      "抂": 77947,
+      "萡": 77948,
+      "梲": 77949,
+      "轂": 77950,
+      "倲": 77951,
+      "搂": 77952,
+      "兖": 77953,
+      "🕚": 77954,
+      "⛤": 77955,
+      "牯": 77956,
+      "凙": 77957,
+      "鉃": 77958,
+      "🦾": 77959,
+      "吥": 77960,
+      "鈼": 77961,
+      "💑🏽": 77962,
+      "潩": 77963,
+      "鼀": 77964,
+      "楳": 77965,
+      "塨": 77966,
+      "谶": 77967,
+      "帛": 77968,
+      "錌": 77969,
+      "Ŗ": 77970,
+      "爣": 77971,
+      "趂": 77972,
+      "狢": 77973,
+      "悱": 77974,
+      "欙": 77975,
+      "笐": 77976,
+      "鹧": 77977,
+      "忾": 77978,
+      "晻": 77979,
+      "譠": 77980,
+      "〰️": 77981,
+      "渖": 77982,
+      "嘖": 77983,
+      "咭": 77984,
+      "丄": 77985,
+      "觡": 77986,
+      "罼": 77987,
+      "繚": 77988,
+      "荗": 77989,
+      "搆": 77990,
+      "詫": 77991,
+      "栍": 77992,
+      "頨": 77993,
+      "邥": 77994,
+      "🩲": 77995,
+      "🧑‍🚒": 77996,
+      "缫": 77997,
+      "绁": 77998,
+      "晩": 77999,
+      "眒": 78000,
+      "≨": 78001,
+      "堒": 78002,
+      "垜": 78003,
+      "戫": 78004,
+      "艆": 78005,
+      "遲": 78006,
+      "窂": 78007,
+      "貆": 78008,
+      "覢": 78009,
+      "喌": 78010,
+      "抰": 78011,
+      "龁": 78012,
+      "櫷": 78013,
+      "箥": 78014,
+      "籎": 78015,
+      "琟": 78016,
+      "肰": 78017,
+      "˰": 78018,
+      "笮": 78019,
+      "簊": 78020,
+      "Ⓟ": 78021,
+      "揲": 78022,
+      "🍄": 78023,
+      "Ǐ": 78024,
+      "🈺": 78025,
+      "譃": 78026,
+      "鬽": 78027,
+      "趄": 78028,
+      "♢": 78029,
+      "巋": 78030,
+      "⒳": 78031,
+      "醞": 78032,
+      "雡": 78033,
+      "Ǵ": 78034,
+      "涎": 78035,
+      "溚": 78036,
+      "肎": 78037,
+      "👳🏾": 78038,
+      "🎆": 78039,
+      "昦": 78040,
+      "🧜🏾‍♂️": 78041,
+      "暠": 78042,
+      "魚": 78043,
+      "⋭": 78044,
+      "癱": 78045,
+      "ʂ": 78046,
+      "趀": 78047,
+      "歘": 78048,
+      "糀": 78049,
+      "麛": 78050,
+      "鸚": 78051,
+      "掆": 78052,
+      "醴": 78053,
+      "夘": 78054,
+      "鑾": 78055,
+      "夂": 78056,
+      "蟃": 78057,
+      "👰‍♂️": 78058,
+      "葓": 78059,
+      "懖": 78060,
+      "鳐": 78061,
+      "欁": 78062,
+      "👨🏿‍🚀": 78063,
+      "來": 78064,
+      "⁋": 78065,
+      "犖": 78066,
+      "嗫": 78067,
+      "祘": 78068,
+      "簷": 78069,
+      "飯": 78070,
+      "⎀": 78071,
+      "躶": 78072,
+      "尒": 78073,
+      "⌭": 78074,
+      "΂": 78075,
+      "萙": 78076,
+      "閾": 78077,
+      "瞅": 78078,
+      "睪": 78079,
+      "🧑‍🦱": 78080,
+      "嗕": 78081,
+      "咙": 78082,
+      "扥": 78083,
+      "👴🏻": 78084,
+      "🤾🏽": 78085,
+      "輠": 78086,
+      "ℇ": 78087,
+      "🥥": 78088,
+      "扦": 78089,
+      "📰": 78090,
+      "🫴": 78091,
+      "妪": 78092,
+      "👩🏼‍✈️": 78093,
+      "☾": 78094,
+      "惍": 78095,
+      "👱🏻‍♂️": 78096,
+      "濦": 78097,
+      "騡": 78098,
+      "鴝": 78099,
+      "紐": 78100,
+      "儷": 78101,
+      "👨🏿‍🦲": 78102,
+      "♔": 78103,
+      "梐": 78104,
+      "⋺": 78105,
+      "柕": 78106,
+      "馤": 78107,
+      "孊": 78108,
+      "瑪": 78109,
+      "┽": 78110,
+      "墡": 78111,
+      "貰": 78112,
+      "鈽": 78113,
+      "襗": 78114,
+      "秷": 78115,
+      "👨🏿‍🦳": 78116,
+      "塕": 78117,
+      "👉🏿": 78118,
+      "禜": 78119,
+      "襳": 78120,
+      "🌖": 78121,
+      "╊": 78122,
+      "鬎": 78123,
+      "硁": 78124,
+      "羉": 78125,
+      "俉": 78126,
+      "撳": 78127,
+      "歫": 78128,
+      "轍": 78129,
+      "牴": 78130,
+      "讀": 78131,
+      "胣": 78132,
+      "璈": 78133,
+      "₝": 78134,
+      "妎": 78135,
+      "洳": 78136,
+      "👨‍🦼": 78137,
+      "谰": 78138,
+      "錤": 78139,
+      "蠒": 78140,
+      "👵🏿": 78141,
+      "缢": 78142,
+      "坮": 78143,
+      "諾": 78144,
+      "骊": 78145,
+      "燾": 78146,
+      "墝": 78147,
+      "崿": 78148,
+      "縹": 78149,
+      "🆕": 78150,
+      "␨": 78151,
+      "鲐": 78152,
+      "蒰": 78153,
+      "纍": 78154,
+      "唫": 78155,
+      "萜": 78156,
+      "婡": 78157,
+      "̟": 78158,
+      "繱": 78159,
+      "葚": 78160,
+      "缽": 78161,
+      "👨‍🦱": 78162,
+      "璘": 78163,
+      "悊": 78164,
+      "懸": 78165,
+      "屽": 78166,
+      "厰": 78167,
+      "浬": 78168,
+      "ℝ": 78169,
+      "桼": 78170,
+      "禱": 78171,
+      "駥": 78172,
+      "ƽ": 78173,
+      "ċ": 78174,
+      "⍘": 78175,
+      "🏋🏽‍♂️": 78176,
+      "縞": 78177,
+      "仒": 78178,
+      "寲": 78179,
+      "🤹🏿‍♂️": 78180,
+      "嗳": 78181,
+      "禿": 78182,
+      "齷": 78183,
+      "訧": 78184,
+      "◉": 78185,
+      "騴": 78186,
+      "🪶": 78187,
+      "蓙": 78188,
+      "扲": 78189,
+      "綜": 78190,
+      "Ί": 78191,
+      "阽": 78192,
+      "😉": 78193,
+      "靻": 78194,
+      "斞": 78195,
+      "⓯": 78196,
+      "釫": 78197,
+      "鱢": 78198,
+      "晣": 78199,
+      "酡": 78200,
+      "俽": 78201,
+      "🧚🏿‍♀️": 78202,
+      "脋": 78203,
+      "♃": 78204,
+      "👨🏽‍🎨": 78205,
+      "擟": 78206,
+      "Ă": 78207,
+      "徑": 78208,
+      "鷢": 78209,
+      "鋁": 78210,
+      "👻": 78211,
+      "廾": 78212,
+      "孈": 78213,
+      "笣": 78214,
+      "叽": 78215,
+      "👨‍👩‍👦‍👦": 78216,
+      "衼": 78217,
+      "媥": 78218,
+      "菐": 78219,
+      "杮": 78220,
+      "郍": 78221,
+      "🪲": 78222,
+      "竼": 78223,
+      "觌": 78224,
+      "翜": 78225,
+      "鵼": 78226,
+      "蕄": 78227,
+      "稴": 78228,
+      "💁🏽": 78229,
+      "✩": 78230,
+      "鱦": 78231,
+      "覿": 78232,
+      "綒": 78233,
+      "怴": 78234,
+      "晄": 78235,
+      "幚": 78236,
+      "⛹🏾‍♀️": 78237,
+      "椻": 78238,
+      "杲": 78239,
+      "栲": 78240,
+      "👩🏿‍⚖️": 78241,
+      "倯": 78242,
+      "墚": 78243,
+      "罦": 78244,
+      "💁🏾": 78245,
+      "趘": 78246,
+      "蹕": 78247,
+      "蝖": 78248,
+      "紻": 78249,
+      "銕": 78250,
+      "訵": 78251,
+      "鱻": 78252,
+      "懟": 78253,
+      "榩": 78254,
+      "轗": 78255,
+      "🎇": 78256,
+      "Ú": 78257,
+      "噖": 78258,
+      "熉": 78259,
+      "懷": 78260,
+      "💏🏻": 78261,
+      "暀": 78262,
+      "噙": 78263,
+      "膦": 78264,
+      "僿": 78265,
+      "搀": 78266,
+      "莃": 78267,
+      "釳": 78268,
+      "礌": 78269,
+      "煟": 78270,
+      "劦": 78271,
+      "磮": 78272,
+      "刔": 78273,
+      "⊲": 78274,
+      "婜": 78275,
+      "ʭ": 78276,
+      "ʙ": 78277,
+      "皢": 78278,
+      "颕": 78279,
+      "娡": 78280,
+      "鉒": 78281,
+      "潵": 78282,
+      "ℓ": 78283,
+      "捵": 78284,
+      "懣": 78285,
+      "🤾🏾": 78286,
+      "琬": 78287,
+      "💂🏿": 78288,
+      "暣": 78289,
+      "🐸": 78290,
+      "␟": 78291,
+      "赬": 78292,
+      "蜆": 78293,
+      "╠": 78294,
+      "徴": 78295,
+      "膄": 78296,
+      "💙": 78297,
+      "鞾": 78298,
+      "內": 78299,
+      "昶": 78300,
+      "爞": 78301,
+      "敵": 78302,
+      "↺": 78303,
+      "纫": 78304,
+      "湤": 78305,
+      "🫷": 78306,
+      "瑱": 78307,
+      "堉": 78308,
+      "🧜‍♀️": 78309,
+      "🧑🏾‍❤️‍💋‍🧑🏻": 78310,
+      "瓞": 78311,
+      "隤": 78312,
+      "穙": 78313,
+      "💄": 78314,
+      "繘": 78315,
+      "揨": 78316,
+      "櫲": 78317,
+      "覧": 78318,
+      "鎫": 78319,
+      "谑": 78320,
+      "脅": 78321,
+      "冓": 78322,
+      "̉": 78323,
+      "蟦": 78324,
+      "滣": 78325,
+      "鞪": 78326,
+      "🚢": 78327,
+      "濣": 78328,
+      "傛": 78329,
+      "衮": 78330,
+      "譺": 78331,
+      "攦": 78332,
+      "厖": 78333,
+      "髀": 78334,
+      "縯": 78335,
+      "茦": 78336,
+      "🦸🏽‍♂️": 78337,
+      "怓": 78338,
+      "🤷🏻‍♀️": 78339,
+      "奀": 78340,
+      "坶": 78341,
+      "窀": 78342,
+      "陳": 78343,
+      "屫": 78344,
+      "🥀": 78345,
+      "🤵🏿‍♂️": 78346,
+      "禯": 78347,
+      "詮": 78348,
+      "楴": 78349,
+      "ʆ": 78350,
+      "鞎": 78351,
+      "夌": 78352,
+      "駮": 78353,
+      "荑": 78354,
+      "肑": 78355,
+      "夐": 78356,
+      "剹": 78357,
+      "紱": 78358,
+      "攛": 78359,
+      "忓": 78360,
+      "蘘": 78361,
+      "囡": 78362,
+      "铀": 78363,
+      "阫": 78364,
+      "🇬🇶": 78365,
+      "⒔": 78366,
+      "箍": 78367,
+      "尐": 78368,
+      "醼": 78369,
+      "禲": 78370,
+      "🙇‍♀️": 78371,
+      "溣": 78372,
+      "鲄": 78373,
+      "磩": 78374,
+      "殏": 78375,
+      "╙": 78376,
+      "鰑": 78377,
+      "薴": 78378,
+      "飻": 78379,
+      "🏊‍♂️": 78380,
+      "Ɍ": 78381,
+      "♘": 78382,
+      "昈": 78383,
+      "🧑🏻‍🌾": 78384,
+      "橵": 78385,
+      "睃": 78386,
+      "˃": 78387,
+      "堲": 78388,
+      "乵": 78389,
+      "據": 78390,
+      "🚵🏻‍♀️": 78391,
+      "Ǘ": 78392,
+      "槲": 78393,
+      "鮞": 78394,
+      "鈗": 78395,
+      "姺": 78396,
+      "煐": 78397,
+      "栰": 78398,
+      "糫": 78399,
+      "垍": 78400,
+      "聹": 78401,
+      "岟": 78402,
+      "☟": 78403,
+      "叼": 78404,
+      "瞙": 78405,
+      "玵": 78406,
+      "啕": 78407,
+      "樥": 78408,
+      "俒": 78409,
+      "躢": 78410,
+      "礷": 78411,
+      "睚": 78412,
+      "銽": 78413,
+      "柷": 78414,
+      "櫀": 78415,
+      "篰": 78416,
+      "籣": 78417,
+      "爋": 78418,
+      "霎": 78419,
+      "耮": 78420,
+      "諝": 78421,
+      "💤": 78422,
+      "蝀": 78423,
+      "氆": 78424,
+      "阝": 78425,
+      "衻": 78426,
+      "觑": 78427,
+      "👨‍👩‍👧‍👦": 78428,
+      "🚴🏻": 78429,
+      "裚": 78430,
+      "傈": 78431,
+      "粖": 78432,
+      "蚲": 78433,
+      "🧐": 78434,
+      "獌": 78435,
+      "瘥": 78436,
+      "瓠": 78437,
+      "ˏ": 78438,
+      "ɍ": 78439,
+      "⊌": 78440,
+      "🎵": 78441,
+      "髊": 78442,
+      "畩": 78443,
+      "趝": 78444,
+      "噍": 78445,
+      "藛": 78446,
+      "獼": 78447,
+      "轓": 78448,
+      "⑀": 78449,
+      "軋": 78450,
+      "唣": 78451,
+      "⏃": 78452,
+      "孌": 78453,
+      "硚": 78454,
+      "胉": 78455,
+      "袕": 78456,
+      "˪": 78457,
+      "銐": 78458,
+      "👊": 78459,
+      "嚍": 78460,
+      "🦸🏼": 78461,
+      "ɯ": 78462,
+      "棐": 78463,
+      "榑": 78464,
+      "犜": 78465,
+      "偍": 78466,
+      "蠓": 78467,
+      "乁": 78468,
+      "皹": 78469,
+      "🏌🏻": 78470,
+      "∦": 78471,
+      "捳": 78472,
+      "騳": 78473,
+      "號": 78474,
+      "耑": 78475,
+      "窓": 78476,
+      "呻": 78477,
+      "諰": 78478,
+      "🤣": 78479,
+      "🎅🏽": 78480,
+      "鰳": 78481,
+      "硗": 78482,
+      "🫶🏽": 78483,
+      "搼": 78484,
+      "覀": 78485,
+      "☿": 78486,
+      "孿": 78487,
+      "沋": 78488,
+      "🚣🏼‍♂️": 78489,
+      "蝛": 78490,
+      "軓": 78491,
+      "勗": 78492,
+      "遅": 78493,
+      "🧝🏼‍♀️": 78494,
+      "猖": 78495,
+      "珙": 78496,
+      "饉": 78497,
+      "瀝": 78498,
+      "僎": 78499,
+      "髃": 78500,
+      "⒟": 78501,
+      "烗": 78502,
+      "枳": 78503,
+      "🤹🏻‍♀️": 78504,
+      "┝": 78505,
+      "愶": 78506,
+      "👩🏼‍❤️‍👩🏻": 78507,
+      "偯": 78508,
+      "脢": 78509,
+      "竌": 78510,
+      "藭": 78511,
+      "蜬": 78512,
+      "樮": 78513,
+      "贛": 78514,
+      "竒": 78515,
+      "嬘": 78516,
+      "繹": 78517,
+      "鈐": 78518,
+      "伋": 78519,
+      "岶": 78520,
+      "剙": 78521,
+      "⍭": 78522,
+      "兝": 78523,
+      "邷": 78524,
+      "≜": 78525,
+      "鼆": 78526,
+      "壧": 78527,
+      "🤥": 78528,
+      "镤": 78529,
+      "褕": 78530,
+      "😔": 78531,
+      "責": 78532,
+      "篎": 78533,
+      "蛩": 78534,
+      "丬": 78535,
+      "🧿": 78536,
+      "棻": 78537,
+      "湳": 78538,
+      "焌": 78539,
+      "唓": 78540,
+      "曶": 78541,
+      "🇸🇽": 78542,
+      "衿": 78543,
+      "顜": 78544,
+      "國": 78545,
+      "陉": 78546,
+      "袘": 78547,
+      "佋": 78548,
+      "欍": 78549,
+      "珲": 78550,
+      "👩🏽‍🔧": 78551,
+      "輸": 78552,
+      "┴": 78553,
+      "🚣🏾": 78554,
+      "顝": 78555,
+      "≮": 78556,
+      "🖖🏼": 78557,
+      "鼌": 78558,
+      "蚭": 78559,
+      "Ↄ": 78560,
+      "辤": 78561,
+      "峽": 78562,
+      "僦": 78563,
+      "╽": 78564,
+      "🌑": 78565,
+      "鉰": 78566,
+      "₩": 78567,
+      "蚗": 78568,
+      "锓": 78569,
+      "謙": 78570,
+      "銰": 78571,
+      "👋🏻": 78572,
+      "垝": 78573,
+      "灇": 78574,
+      "🏄🏼‍♀️": 78575,
+      "嫟": 78576,
+      "壿": 78577,
+      "摡": 78578,
+      "遚": 78579,
+      "⒕": 78580,
+      "輜": 78581,
+      "橧": 78582,
+      "猺": 78583,
+      "婬": 78584,
+      "⌢": 78585,
+      "忯": 78586,
+      "寢": 78587,
+      "柟": 78588,
+      "驸": 78589,
+      "💑": 78590,
+      "涗": 78591,
+      "⊤": 78592,
+      "啲": 78593,
+      "🇵🇳": 78594,
+      "🕎": 78595,
+      "錨": 78596,
+      "祦": 78597,
+      "蘤": 78598,
+      "冿": 78599,
+      "≷": 78600,
+      "鉏": 78601,
+      "紭": 78602,
+      "讉": 78603,
+      "睠": 78604,
+      "媝": 78605,
+      "ʯ": 78606,
+      "狎": 78607,
+      "嬥": 78608,
+      "🧖🏿‍♂️": 78609,
+      "跺": 78610,
+      "🧑🏾‍🔬": 78611,
+      "枦": 78612,
+      "灟": 78613,
+      "鈳": 78614,
+      "┟": 78615,
+      "熵": 78616,
+      "覅": 78617,
+      "汄": 78618,
+      "嬮": 78619,
+      "悝": 78620,
+      "腪": 78621,
+      "屦": 78622,
+      "姟": 78623,
+      "🤤": 78624,
+      "袠": 78625,
+      "鸲": 78626,
+      "熶": 78627,
+      "韝": 78628,
+      "🕢": 78629,
+      "◌": 78630,
+      "🧚": 78631,
+      "玂": 78632,
+      "👧🏼": 78633,
+      "顩": 78634,
+      "餚": 78635,
+      "謆": 78636,
+      "趲": 78637,
+      "驵": 78638,
+      "怣": 78639,
+      "囸": 78640,
+      "暄": 78641,
+      "梕": 78642,
+      "羥": 78643,
+      "🤟": 78644,
+      "賊": 78645,
+      "🚮": 78646,
+      "萩": 78647,
+      "Μ": 78648,
+      "澘": 78649,
+      "👨🏾‍⚖️": 78650,
+      "訯": 78651,
+      "檟": 78652,
+      "峴": 78653,
+      "榒": 78654,
+      "鷪": 78655,
+      "🤛🏼": 78656,
+      "蟇": 78657,
+      "瑥": 78658,
+      "📇": 78659,
+      "弒": 78660,
+      "紟": 78661,
+      "鍞": 78662,
+      "诒": 78663,
+      "朚": 78664,
+      "🖐🏿": 78665,
+      "⃀": 78666,
+      "🥮": 78667,
+      "⎆": 78668,
+      "孨": 78669,
+      "漀": 78670,
+      "淺": 78671,
+      "🚊": 78672,
+      "櫯": 78673,
+      "夤": 78674,
+      "垸": 78675,
+      "鳶": 78676,
+      "鈫": 78677,
+      "鹚": 78678,
+      "訟": 78679,
+      "嚅": 78680,
+      "謼": 78681,
+      "疻": 78682,
+      "麾": 78683,
+      "殭": 78684,
+      "刬": 78685,
+      "啐": 78686,
+      "昪": 78687,
+      "稈": 78688,
+      "🏔": 78689,
+      "🤹🏻": 78690,
+      "膺": 78691,
+      "🛳": 78692,
+      "噏": 78693,
+      "妸": 78694,
+      "🐥": 78695,
+      "🫱🏻‍🫲🏾": 78696,
+      "飤": 78697,
+      "🧝🏽‍♂️": 78698,
+      "餃": 78699,
+      "歒": 78700,
+      "螷": 78701,
+      "蜌": 78702,
+      "⇻": 78703,
+      "轔": 78704,
+      "鷦": 78705,
+      "踒": 78706,
+      "袌": 78707,
+      "🫨": 78708,
+      "硰": 78709,
+      "锷": 78710,
+      "蚜": 78711,
+      "磡": 78712,
+      "県": 78713,
+      "嬐": 78714,
+      "錓": 78715,
+      "娔": 78716,
+      "↱": 78717,
+      "儌": 78718,
+      "嗑": 78719,
+      "戄": 78720,
+      "絖": 78721,
+      "ⅉ": 78722,
+      "岃": 78723,
+      "澔": 78724,
+      "⇨": 78725,
+      "鏢": 78726,
+      "穇": 78727,
+      "⁘": 78728,
+      "櫉": 78729,
+      "Ý": 78730,
+      "洓": 78731,
+      "徂": 78732,
+      "榱": 78733,
+      "鼧": 78734,
+      "卅": 78735,
+      "␙": 78736,
+      "僝": 78737,
+      "膫": 78738,
+      "🧍": 78739,
+      "👩‍🎤": 78740,
+      "🫠": 78741,
+      "📩": 78742,
+      "鎙": 78743,
+      "徕": 78744,
+      "亵": 78745,
+      "鶮": 78746,
+      "騿": 78747,
+      "獠": 78748,
+      "鯌": 78749,
+      "咄": 78750,
+      "遫": 78751,
+      "🔜": 78752,
+      "⛐": 78753,
+      "綈": 78754,
+      "徶": 78755,
+      "📃": 78756,
+      "濆": 78757,
+      "蹪": 78758,
+      "瑂": 78759,
+      "緉": 78760,
+      "噚": 78761,
+      "鐤": 78762,
+      "🧚‍♂️": 78763,
+      "呙": 78764,
+      "斮": 78765,
+      "🧋": 78766,
+      "ώ": 78767,
+      "⇇": 78768,
+      "懥": 78769,
+      "錝": 78770,
+      "琔": 78771,
+      "耓": 78772,
+      "👩🏽‍❤️‍👩🏿": 78773,
+      "瓡": 78774,
+      "貪": 78775,
+      "僢": 78776,
+      "汊": 78777,
+      "憿": 78778,
+      "揧": 78779,
+      "梆": 78780,
+      "⛘": 78781,
+      "虥": 78782,
+      "繜": 78783,
+      "麁": 78784,
+      "磛": 78785,
+      "罁": 78786,
+      "廛": 78787,
+      "燲": 78788,
+      "崗": 78789,
+      "擁": 78790,
+      "獀": 78791,
+      "伌": 78792,
+      "睔": 78793,
+      "練": 78794,
+      "穱": 78795,
+      "⌷": 78796,
+      "🛀🏽": 78797,
+      "唹": 78798,
+      "棷": 78799,
+      "瑒": 78800,
+      "僗": 78801,
+      "鋱": 78802,
+      "鼥": 78803,
+      "奁": 78804,
+      "獿": 78805,
+      "◒": 78806,
+      "鱟": 78807,
+      "胦": 78808,
+      "🇨🇰": 78809,
+      "Ċ": 78810,
+      "🚤": 78811,
+      "碲": 78812,
+      "濧": 78813,
+      "酓": 78814,
+      "💆🏿": 78815,
+      "稭": 78816,
+      "詆": 78817,
+      "塲": 78818,
+      "羖": 78819,
+      "畓": 78820,
+      "妿": 78821,
+      "僴": 78822,
+      "巜": 78823,
+      "湽": 78824,
+      "幈": 78825,
+      "◱": 78826,
+      "🕘": 78827,
+      "袏": 78828,
+      "嗤": 78829,
+      "🤴": 78830,
+      "‎": 78831,
+      "頣": 78832,
+      "鑟": 78833,
+      "毃": 78834,
+      "瞼": 78835,
+      "黂": 78836,
+      "盝": 78837,
+      "🧸": 78838,
+      "窫": 78839,
+      "璶": 78840,
+      "稓": 78841,
+      "糱": 78842,
+      "絚": 78843,
+      "淦": 78844,
+      "綠": 78845,
+      "⍬": 78846,
+      "蔮": 78847,
+      "🍇": 78848,
+      "衾": 78849,
+      "毐": 78850,
+      "閧": 78851,
+      "環": 78852,
+      "⇐": 78853,
+      "盥": 78854,
+      "🪳": 78855,
+      "奂": 78856,
+      "姖": 78857,
+      "眞": 78858,
+      "鏅": 78859,
+      "飳": 78860,
+      "秬": 78861,
+      "涘": 78862,
+      "沣": 78863,
+      "輡": 78864,
+      "眪": 78865,
+      "🈯️": 78866,
+      "璙": 78867,
+      "溘": 78868,
+      "厃": 78869,
+      "悹": 78870,
+      "👱🏾‍♂️": 78871,
+      "Ｙ": 78872,
+      "樯": 78873,
+      "🍿": 78874,
+      "癔": 78875,
+      "跴": 78876,
+      "🕵🏽‍♀️": 78877,
+      "袙": 78878,
+      "葍": 78879,
+      "鍌": 78880,
+      "鐢": 78881,
+      "洸": 78882,
+      "⏰": 78883,
+      "餼": 78884,
+      "碛": 78885,
+      "𓀘": 78886,
+      "🇲🇨": 78887,
+      "⋝": 78888,
+      "覫": 78889,
+      "狾": 78890,
+      "蜑": 78891,
+      "麸": 78892,
+      "Ν": 78893,
+      "颷": 78894,
+      "璏": 78895,
+      "鞵": 78896,
+      "貺": 78897,
+      "郀": 78898,
+      "硄": 78899,
+      "囀": 78900,
+      "峖": 78901,
+      "儜": 78902,
+      "璽": 78903,
+      "籔": 78904,
+      "焩": 78905,
+      "∭": 78906,
+      "囗": 78907,
+      "嘪": 78908,
+      "獅": 78909,
+      "捼": 78910,
+      "☘": 78911,
+      "瘎": 78912,
+      "🧑‍🦲": 78913,
+      "宮": 78914,
+      "荖": 78915,
+      "堠": 78916,
+      "💂🏾‍♂️": 78917,
+      "鯅": 78918,
+      "覨": 78919,
+      "豾": 78920,
+      "覘": 78921,
+      "囙": 78922,
+      "♧": 78923,
+      "蒫": 78924,
+      "浟": 78925,
+      "盁": 78926,
+      "疃": 78927,
+      "炨": 78928,
+      "嫫": 78929,
+      "匫": 78930,
+      "蔍": 78931,
+      "杇": 78932,
+      "霛": 78933,
+      "滎": 78934,
+      "稦": 78935,
+      "赝": 78936,
+      "譤": 78937,
+      "🤽🏾": 78938,
+      "🍁": 78939,
+      "🖕🏿": 78940,
+      "汿": 78941,
+      "臢": 78942,
+      "堝": 78943,
+      "縉": 78944,
+      "͝": 78945,
+      "🥴": 78946,
+      "鵂": 78947,
+      "🚵🏽‍♂️": 78948,
+      "剢": 78949,
+      "媌": 78950,
+      "籮": 78951,
+      "錎": 78952,
+      "汻": 78953,
+      "筞": 78954,
+      "咲": 78955,
+      "🏃🏼‍♂️": 78956,
+      "焪": 78957,
+      "✤": 78958,
+      "衟": 78959,
+      "躥": 78960,
+      "螑": 78961,
+      "⅊": 78962,
+      "摨": 78963,
+      "鏣": 78964,
+      "璿": 78965,
+      "佥": 78966,
+      "魕": 78967,
+      "篁": 78968,
+      "撟": 78969,
+      "⌯": 78970,
+      "塓": 78971,
+      "侜": 78972,
+      "爒": 78973,
+      "榵": 78974,
+      "鼞": 78975,
+      "浯": 78976,
+      "紣": 78977,
+      "痖": 78978,
+      "姶": 78979,
+      "枿": 78980,
+      "溡": 78981,
+      "豟": 78982,
+      "🧔🏾": 78983,
+      "⇛": 78984,
+      "狍": 78985,
+      "〈": 78986,
+      "椸": 78987,
+      "椠": 78988,
+      "煵": 78989,
+      "蓘": 78990,
+      "🤛🏾": 78991,
+      "鱮": 78992,
+      "唞": 78993,
+      "臚": 78994,
+      "␖": 78995,
+      "🔛": 78996,
+      "餡": 78997,
+      "Ƥ": 78998,
+      "⛕": 78999,
+      "ʩ": 79000,
+      "鶦": 79001,
+      "👮🏾‍♂️": 79002,
+      "焟": 79003,
+      "約": 79004,
+      "曠": 79005,
+      "锭": 79006,
+      "羪": 79007,
+      "▵": 79008,
+      "硑": 79009,
+      "🐴": 79010,
+      "∮": 79011,
+      "藫": 79012,
+      "輟": 79013,
+      "叢": 79014,
+      "涺": 79015,
+      "鹛": 79016,
+      "摳": 79017,
+      "澅": 79018,
+      "🏃🏽‍♂️": 79019,
+      "滱": 79020,
+      "₟": 79021,
+      "髍": 79022,
+      "殆": 79023,
+      "欨": 79024,
+      "炵": 79025,
+      "舗": 79026,
+      "👩🏽‍🚒": 79027,
+      "鳡": 79028,
+      "矺": 79029,
+      "鞨": 79030,
+      "粀": 79031,
+      "晠": 79032,
+      "舷": 79033,
+      "⛫": 79034,
+      "滺": 79035,
+      "灻": 79036,
+      "灱": 79037,
+      "銒": 79038,
+      "試": 79039,
+      "鲺": 79040,
+      "ȫ": 79041,
+      "鸧": 79042,
+      "🏊🏼‍♀️": 79043,
+      "剨": 79044,
+      "狷": 79045,
+      "柝": 79046,
+      "藣": 79047,
+      "彛": 79048,
+      "飺": 79049,
+      "礑": 79050,
+      "蓑": 79051,
+      "缷": 79052,
+      "⎒": 79053,
+      "皪": 79054,
+      "┳": 79055,
+      "耤": 79056,
+      "悐": 79057,
+      "旿": 79058,
+      "綝": 79059,
+      "潴": 79060,
+      "弑": 79061,
+      "毉": 79062,
+      "🏃🏽": 79063,
+      "斿": 79064,
+      "˅": 79065,
+      "╒": 79066,
+      "搯": 79067,
+      "茰": 79068,
+      "猧": 79069,
+      "愠": 79070,
+      "倗": 79071,
+      "頁": 79072,
+      "👨🏽‍🏭": 79073,
+      "躵": 79074,
+      "⎩": 79075,
+      "🌙": 79076,
+      "鲠": 79077,
+      "鮼": 79078,
+      "藧": 79079,
+      "軚": 79080,
+      "櫁": 79081,
+      "梽": 79082,
+      "旤": 79083,
+      "⌌": 79084,
+      "餈": 79085,
+      "胹": 79086,
+      "⍍": 79087,
+      "鱝": 79088,
+      "🏃🏻": 79089,
+      "讻": 79090,
+      "🧑🏾‍❤️‍🧑🏻": 79091,
+      "🧓🏿": 79092,
+      "鯜": 79093,
+      "旜": 79094,
+      "盳": 79095,
+      "🛷": 79096,
+      "袟": 79097,
+      "楙": 79098,
+      "狕": 79099,
+      "蜉": 79100,
+      "幌": 79101,
+      "瓙": 79102,
+      "巗": 79103,
+      "📓": 79104,
+      "緒": 79105,
+      "閮": 79106,
+      "⑝": 79107,
+      "慊": 79108,
+      "⋂": 79109,
+      "營": 79110,
+      "⅑": 79111,
+      "锢": 79112,
+      "🫃🏾": 79113,
+      "聭": 79114,
+      "踲": 79115,
+      "丅": 79116,
+      "嫰": 79117,
+      "揿": 79118,
+      "騃": 79119,
+      "⚡": 79120,
+      "纈": 79121,
+      "鈍": 79122,
+      "浖": 79123,
+      "┷": 79124,
+      "孞": 79125,
+      "嵙": 79126,
+      "灁": 79127,
+      "矆": 79128,
+      "藠": 79129,
+      "椥": 79130,
+      "觕": 79131,
+      "ⅿ": 79132,
+      "🧫": 79133,
+      "亐": 79134,
+      "縛": 79135,
+      "炠": 79136,
+      "烰": 79137,
+      "宊": 79138,
+      "⇉": 79139,
+      "曷": 79140,
+      "⑟": 79141,
+      "鯩": 79142,
+      "嬦": 79143,
+      "篖": 79144,
+      "🧑🏼‍🔬": 79145,
+      "閵": 79146,
+      "👐": 79147,
+      "菸": 79148,
+      "錭": 79149,
+      "嵤": 79150,
+      "囒": 79151,
+      "賘": 79152,
+      "猙": 79153,
+      "LJ": 79154,
+      "褞": 79155,
+      "臩": 79156,
+      "鐔": 79157,
+      "蹴": 79158,
+      "塋": 79159,
+      "脹": 79160,
+      "犣": 79161,
+      "給": 79162,
+      "🫷🏻": 79163,
+      "👾": 79164,
+      "窌": 79165,
+      "嚺": 79166,
+      "舻": 79167,
+      "鐀": 79168,
+      "嫻": 79169,
+      "紞": 79170,
+      "倥": 79171,
+      "🧆": 79172,
+      "ͨ": 79173,
+      "厺": 79174,
+      "🤵🏽‍♂️": 79175,
+      "衚": 79176,
+      "奓": 79177,
+      "鴃": 79178,
+      "⚌": 79179,
+      "👨🏼‍🍼": 79180,
+      "趮": 79181,
+      "碝": 79182,
+      "筀": 79183,
+      "詝": 79184,
+      "郮": 79185,
+      "虤": 79186,
+      "🙆🏽‍♂️": 79187,
+      "👩‍🦽": 79188,
+      "餷": 79189,
+      "掣": 79190,
+      "癛": 79191,
+      "ŵ": 79192,
+      "啬": 79193,
+      "縆": 79194,
+      "贕": 79195,
+      "罆": 79196,
+      "⊷": 79197,
+      "樺": 79198,
+      "嗹": 79199,
+      "挆": 79200,
+      "鑨": 79201,
+      "釿": 79202,
+      "🇪🇨": 79203,
+      "👨🏽‍❤️‍👨🏼": 79204,
+      "鎳": 79205,
+      "蜟": 79206,
+      "鴮": 79207,
+      "鲡": 79208,
+      "屇": 79209,
+      "壝": 79210,
+      "铯": 79211,
+      "嵜": 79212,
+      "蹩": 79213,
+      "🧑🏿‍🎤": 79214,
+      "岒": 79215,
+      "🎖": 79216,
+      "玊": 79217,
+      "轜": 79218,
+      "偄": 79219,
+      "🩵": 79220,
+      "矅": 79221,
+      "沵": 79222,
+      "侩": 79223,
+      "窙": 79224,
+      "鋒": 79225,
+      "👰🏿‍♂️": 79226,
+      "麣": 79227,
+      "匸": 79228,
+      "傘": 79229,
+      "烄": 79230,
+      "砒": 79231,
+      "籸": 79232,
+      "儡": 79233,
+      "涓": 79234,
+      "⁵": 79235,
+      "藒": 79236,
+      "兦": 79237,
+      "誾": 79238,
+      "⏇": 79239,
+      "🦪": 79240,
+      "坈": 79241,
+      "庁": 79242,
+      "儎": 79243,
+      "錱": 79244,
+      "鐭": 79245,
+      "Ｚ": 79246,
+      "鵳": 79247,
+      "₾": 79248,
+      "寛": 79249,
+      "庎": 79250,
+      "蓳": 79251,
+      "糦": 79252,
+      "愯": 79253,
+      "🧔🏿‍♀️": 79254,
+      "ⅼ": 79255,
+      "舄": 79256,
+      "👌🏻": 79257,
+      "🤸🏼": 79258,
+      "汭": 79259,
+      "▒": 79260,
+      "皉": 79261,
+      "峠": 79262,
+      "踜": 79263,
+      "🧩": 79264,
+      "磎": 79265,
+      "劸": 79266,
+      "麮": 79267,
+      "腡": 79268,
+      "儢": 79269,
+      "裨": 79270,
+      "🫱🏾‍🫲🏽": 79271,
+      "̢": 79272,
+      "🥇": 79273,
+      "盨": 79274,
+      "┱": 79275,
+      "翿": 79276,
+      "😃": 79277,
+      "👩🏾‍❤️‍👨🏼": 79278,
+      "臋": 79279,
+      "薥": 79280,
+      "ǟ": 79281,
+      "咂": 79282,
+      "嶹": 79283,
+      "鴅": 79284,
+      "柽": 79285,
+      "釖": 79286,
+      "挞": 79287,
+      "嚋": 79288,
+      "🏂🏽": 79289,
+      "💲": 79290,
+      "┛": 79291,
+      "蘡": 79292,
+      "腬": 79293,
+      "👷🏽‍♀️": 79294,
+      "🧑🏻‍🎄": 79295,
+      "抟": 79296,
+      "諩": 79297,
+      "搪": 79298,
+      "瘽": 79299,
+      "玶": 79300,
+      "🧑🏿‍🦳": 79301,
+      "矡": 79302,
+      "👮🏿‍♂️": 79303,
+      "蟕": 79304,
+      "塉": 79305,
+      "💂🏻": 79306,
+      "爖": 79307,
+      "碶": 79308,
+      "噯": 79309,
+      "芧": 79310,
+      "🥠": 79311,
+      "犭": 79312,
+      "誒": 79313,
+      "嵋": 79314,
+      "憄": 79315,
+      "🇦🇺": 79316,
+      "☭": 79317,
+      "嗂": 79318,
+      "壢": 79319,
+      "薾": 79320,
+      "🌨": 79321,
+      "蝌": 79322,
+      "苅": 79323,
+      "玚": 79324,
+      "逕": 79325,
+      "伳": 79326,
+      "羷": 79327,
+      "酀": 79328,
+      "晆": 79329,
+      "🚴‍♀️": 79330,
+      "唆": 79331,
+      "謖": 79332,
+      "酢": 79333,
+      "⁡": 79334,
+      "宲": 79335,
+      "窊": 79336,
+      "🧑🏽‍🍳": 79337,
+      "👨🏿‍🔬": 79338,
+      "齌": 79339,
+      "鎇": 79340,
+      "迓": 79341,
+      "👍🏽": 79342,
+      "͏": 79343,
+      "疊": 79344,
+      "鯬": 79345,
+      "労": 79346,
+      "梄": 79347,
+      "鍸": 79348,
+      "搣": 79349,
+      "晑": 79350,
+      "頭": 79351,
+      "挹": 79352,
+      "⍥": 79353,
+      "嬭": 79354,
+      "擵": 79355,
+      "糚": 79356,
+      "铑": 79357,
+      "🇧🇲": 79358,
+      "🧑🏾‍💼": 79359,
+      "躽": 79360,
+      "Ñ": 79361,
+      "╮": 79362,
+      "塍": 79363,
+      "羙": 79364,
+      "铍": 79365,
+      "₳": 79366,
+      "锆": 79367,
+      "塁": 79368,
+      "愹": 79369,
+      "⛹🏼": 79370,
+      "儖": 79371,
+      "🦴": 79372,
+      "ℊ": 79373,
+      "👨‍❤️‍💋‍👨": 79374,
+      "髷": 79375,
+      "🫴🏼": 79376,
+      "🎧": 79377,
+      "≗": 79378,
+      "劾": 79379,
+      "薂": 79380,
+      "Ơ": 79381,
+      "睕": 79382,
+      "🙍‍♀️": 79383,
+      "趰": 79384,
+      "螎": 79385,
+      "絨": 79386,
+      "靔": 79387,
+      "猗": 79388,
+      "屜": 79389,
+      "觷": 79390,
+      "ʟ": 79391,
+      "摈": 79392,
+      "攟": 79393,
+      "␛": 79394,
+      "闧": 79395,
+      "貨": 79396,
+      "🤽": 79397,
+      "🔑": 79398,
+      "協": 79399,
+      "伥": 79400,
+      "壋": 79401,
+      "蠶": 79402,
+      "螦": 79403,
+      "诂": 79404,
+      "楖": 79405,
+      "逷": 79406,
+      "🔢": 79407,
+      "Ƈ": 79408,
+      "禤": 79409,
+      "屰": 79410,
+      "吝": 79411,
+      "菥": 79412,
+      "捬": 79413,
+      "朏": 79414,
+      "诪": 79415,
+      "嶙": 79416,
+      "👮🏼‍♂️": 79417,
+      "陻": 79418,
+      "澢": 79419,
+      "儽": 79420,
+      "🏫": 79421,
+      "笇": 79422,
+      "💂🏽": 79423,
+      "桗": 79424,
+      "☇": 79425,
+      "瘄": 79426,
+      "摌": 79427,
+      "攜": 79428,
+      "💁🏻‍♂️": 79429,
+      "牀": 79430,
+      "嫢": 79431,
+      "⁹": 79432,
+      "♾": 79433,
+      "👨‍💼": 79434,
+      "Ȁ": 79435,
+      "銿": 79436,
+      "剤": 79437,
+      "⅗": 79438,
+      "鯔": 79439,
+      "鍓": 79440,
+      "寀": 79441,
+      "汌": 79442,
+      "脿": 79443,
+      "釩": 79444,
+      "愵": 79445,
+      "靸": 79446,
+      "誏": 79447,
+      "蜊": 79448,
+      "粈": 79449,
+      "鯶": 79450,
+      "͉": 79451,
+      "梇": 79452,
+      "钼": 79453,
+      "Ɵ": 79454,
+      "挎": 79455,
+      "⑛": 79456,
+      "🫸🏽": 79457,
+      "髨": 79458,
+      "藢": 79459,
+      "稛": 79460,
+      "燳": 79461,
+      "潲": 79462,
+      "☸": 79463,
+      "🏌🏾‍♂️": 79464,
+      "鬀": 79465,
+      "訙": 79466,
+      "鉘": 79467,
+      "昖": 79468,
+      "邧": 79469,
+      "禠": 79470,
+      "嫊": 79471,
+      "🏌🏿‍♀️": 79472,
+      "蕅": 79473,
+      "堥": 79474,
+      "緯": 79475,
+      "蛚": 79476,
+      "🇸🇭": 79477,
+      "⚤": 79478,
+      "谲": 79479,
+      "媎": 79480,
+      "╄": 79481,
+      "皍": 79482,
+      "ȇ": 79483,
+      "🤛": 79484,
+      "👨🏼‍⚖️": 79485,
+      "猳": 79486,
+      "噌": 79487,
+      "喺": 79488,
+      "牮": 79489,
+      "麳": 79490,
+      "漢": 79491,
+      "駄": 79492,
+      "絟": 79493,
+      "ℹ": 79494,
+      "🪢": 79495,
+      "🧙🏻": 79496,
+      "🫲🏿": 79497,
+      "ϑ": 79498,
+      "嵛": 79499,
+      "瓖": 79500,
+      "辺": 79501,
+      "蚻": 79502,
+      "⇲": 79503,
+      "鸢": 79504,
+      "⇳": 79505,
+      "侇": 79506,
+      "俓": 79507,
+      "褠": 79508,
+      "捶": 79509,
+      "鶭": 79510,
+      "遠": 79511,
+      "⊃": 79512,
+      "靱": 79513,
+      "绂": 79514,
+      "脽": 79515,
+      "碐": 79516,
+      "🧑🏿‍❤️‍🧑🏽": 79517,
+      "邁": 79518,
+      "嫸": 79519,
+      "詿": 79520,
+      "譞": 79521,
+      "絴": 79522,
+      "捠": 79523,
+      "滸": 79524,
+      "颽": 79525,
+      "⛧": 79526,
+      "癗": 79527,
+      "劇": 79528,
+      "烾": 79529,
+      "尋": 79530,
+      "漥": 79531,
+      "誨": 79532,
+      "👰🏼‍♂️": 79533,
+      "灀": 79534,
+      "ǥ": 79535,
+      "⇱": 79536,
+      "醝": 79537,
+      "氾": 79538,
+      "藆": 79539,
+      "誛": 79540,
+      "齓": 79541,
+      "🌮": 79542,
+      "ƾ": 79543,
+      "👨🏾‍🎤": 79544,
+      "瘂": 79545,
+      "莐": 79546,
+      "杗": 79547,
+      "鷌": 79548,
+      "ɱ": 79549,
+      "佖": 79550,
+      "柮": 79551,
+      "爚": 79552,
+      "⍲": 79553,
+      "搥": 79554,
+      "擆": 79555,
+      "⒁": 79556,
+      "沔": 79557,
+      "蓫": 79558,
+      "對": 79559,
+      "妌": 79560,
+      "绉": 79561,
+      "笒": 79562,
+      "捻": 79563,
+      "▌": 79564,
+      "殛": 79565,
+      "軪": 79566,
+      "幡": 79567,
+      "痧": 79568,
+      "繬": 79569,
+      "焬": 79570,
+      "聥": 79571,
+      "崼": 79572,
+      "媜": 79573,
+      "🤦🏿": 79574,
+      "頯": 79575,
+      "蝺": 79576,
+      "鹯": 79577,
+      "賴": 79578,
+      "🥄": 79579,
+      "慳": 79580,
+      "殎": 79581,
+      "艂": 79582,
+      "鸁": 79583,
+      "餞": 79584,
+      "🇬🇧": 79585,
+      "💁🏻‍♀️": 79586,
+      "伎": 79587,
+      "欷": 79588,
+      "ͮ": 79589,
+      "🚭": 79590,
+      "贚": 79591,
+      "尃": 79592,
+      "砰": 79593,
+      "漎": 79594,
+      "皡": 79595,
+      "💆🏾": 79596,
+      "捘": 79597,
+      "🦢": 79598,
+      "👩‍🦲": 79599,
+      "壌": 79600,
+      "宧": 79601,
+      "呎": 79602,
+      "沨": 79603,
+      "綳": 79604,
+      "↖": 79605,
+      "竸": 79606,
+      "梶": 79607,
+      "違": 79608,
+      "赧": 79609,
+      "酦": 79610,
+      "📒": 79611,
+      "👩🏾‍❤️‍👩🏿": 79612,
+      "勡": 79613,
+      "讂": 79614,
+      "糌": 79615,
+      "咴": 79616,
+      "⌂": 79617,
+      "廀": 79618,
+      "聴": 79619,
+      "麏": 79620,
+      "欻": 79621,
+      "佷": 79622,
+      "聏": 79623,
+      "臸": 79624,
+      "⍃": 79625,
+      "👱🏼‍♂️": 79626,
+      "嬒": 79627,
+      "🤰🏽": 79628,
+      "⃓": 79629,
+      "嶥": 79630,
+      "敋": 79631,
+      "⓮": 79632,
+      "歾": 79633,
+      "錶": 79634,
+      "Ɨ": 79635,
+      "勎": 79636,
+      "💕": 79637,
+      "縂": 79638,
+      "簵": 79639,
+      "糈": 79640,
+      "👷‍♂️": 79641,
+      "⚕": 79642,
+      "孍": 79643,
+      "霢": 79644,
+      "🤵🏾": 79645,
+      "匢": 79646,
+      "紴": 79647,
+      "戞": 79648,
+      "🧑🏿‍🔬": 79649,
+      "😮": 79650,
+      "褉": 79651,
+      "犋": 79652,
+      "礂": 79653,
+      "┭": 79654,
+      "坒": 79655,
+      "袪": 79656,
+      "堽": 79657,
+      "顥": 79658,
+      "蟜": 79659,
+      "輶": 79660,
+      "歊": 79661,
+      "墻": 79662,
+      "疎": 79663,
+      "躨": 79664,
+      "椩": 79665,
+      "臟": 79666,
+      "萁": 79667,
+      "簾": 79668,
+      "肅": 79669,
+      "⏹": 79670,
+      "⓬": 79671,
+      "♠︎": 79672,
+      "媹": 79673,
+      "摱": 79674,
+      "⃔": 79675,
+      "🫸🏼": 79676,
+      "≊": 79677,
+      "骶": 79678,
+      "縲": 79679,
+      "燼": 79680,
+      "邘": 79681,
+      "埤": 79682,
+      "蕌": 79683,
+      "痙": 79684,
+      "🧑🏿‍❤️‍🧑🏻": 79685,
+      "◑": 79686,
+      "懾": 79687,
+      "錞": 79688,
+      "籡": 79689,
+      "捀": 79690,
+      "遜": 79691,
+      "睽": 79692,
+      "貀": 79693,
+      "≿": 79694,
+      "葘": 79695,
+      "🫃": 79696,
+      "匲": 79697,
+      "琺": 79698,
+      "梥": 79699,
+      "薠": 79700,
+      "🗻": 79701,
+      "⑬": 79702,
+      "憰": 79703,
+      "🥺": 79704,
+      "獁": 79705,
+      "澇": 79706,
+      "砯": 79707,
+      "垳": 79708,
+      "輁": 79709,
+      "囝": 79710,
+      "嶝": 79711,
+      "酏": 79712,
+      "苌": 79713,
+      "⋔": 79714,
+      "塼": 79715,
+      "燋": 79716,
+      "溈": 79717,
+      "滪": 79718,
+      "纑": 79719,
+      "癅": 79720,
+      "ǂ": 79721,
+      "轥": 79722,
+      "骣": 79723,
+      "👩‍🎓": 79724,
+      "蘩": 79725,
+      "🙍🏼‍♀️": 79726,
+      "鶃": 79727,
+      "💖": 79728,
+      "瀾": 79729,
+      "蟍": 79730,
+      "🤵🏼‍♀️": 79731,
+      "鶳": 79732,
+      "🏙": 79733,
+      "芣": 79734,
+      "獃": 79735,
+      "囪": 79736,
+      "酲": 79737,
+      "樎": 79738,
+      "⒧": 79739,
+      "隟": 79740,
+      "⃕": 79741,
+      "鋣": 79742,
+      "鬋": 79743,
+      "鑻": 79744,
+      "璥": 79745,
+      "進": 79746,
+      "乊": 79747,
+      "齘": 79748,
+      "涆": 79749,
+      "蚽": 79750,
+      "菖": 79751,
+      "甎": 79752,
+      "💪🏻": 79753,
+      "釸": 79754,
+      "Ȏ": 79755,
+      "櫙": 79756,
+      "🧑🏽‍💼": 79757,
+      "颞": 79758,
+      "憳": 79759,
+      "绺": 79760,
+      "誠": 79761,
+      "梂": 79762,
+      "黊": 79763,
+      "僋": 79764,
+      "賻": 79765,
+      "吷": 79766,
+      "絢": 79767,
+      "騎": 79768,
+      "椁": 79769,
+      "瓺": 79770,
+      "🤵🏿": 79771,
+      "🌆": 79772,
+      "穰": 79773,
+      "瑫": 79774,
+      "啹": 79775,
+      "蕞": 79776,
+      "誕": 79777,
+      "卋": 79778,
+      "箴": 79779,
+      "鞐": 79780,
+      "霌": 79781,
+      "🧒🏽": 79782,
+      "屙": 79783,
+      "͇": 79784,
+      "≓": 79785,
+      "🧕🏻": 79786,
+      "霝": 79787,
+      "蜣": 79788,
+      "艋": 79789,
+      "馢": 79790,
+      "芿": 79791,
+      "疂": 79792,
+      "峩": 79793,
+      "灍": 79794,
+      "嬪": 79795,
+      "鍨": 79796,
+      "鈥": 79797,
+      "蓮": 79798,
+      "澵": 79799,
+      "燶": 79800,
+      "缹": 79801,
+      "滬": 79802,
+      "🧑🏾‍🍳": 79803,
+      "娍": 79804,
+      "鄼": 79805,
+      "俥": 79806,
+      "汚": 79807,
+      "婩": 79808,
+      "👩‍🔧": 79809,
+      "🤞🏻": 79810,
+      "璬": 79811,
+      "⛙": 79812,
+      "昣": 79813,
+      "詭": 79814,
+      "禗": 79815,
+      "熲": 79816,
+      "🚇": 79817,
+      "囹": 79818,
+      "曘": 79819,
+      "▪": 79820,
+      "帣": 79821,
+      "🧑‍✈️": 79822,
+      "👩🏼‍🔧": 79823,
+      "魉": 79824,
+      "萰": 79825,
+      "˘": 79826,
+      "蜪": 79827,
+      "櫟": 79828,
+      "鮬": 79829,
+      "鱃": 79830,
+      "侕": 79831,
+      "🛌": 79832,
+      "👳🏽‍♂️": 79833,
+      "鯏": 79834,
+      "🪂": 79835,
+      "⒥": 79836,
+      "笗": 79837,
+      "鴐": 79838,
+      "慴": 79839,
+      "": 79840,
+      "槪": 79841,
+      "┦": 79842,
+      "搘": 79843,
+      "謢": 79844,
+      "🇨🇱": 79845,
+      "🤾🏾‍♀️": 79846,
+      "🤷🏼‍♂️": 79847,
+      "↕": 79848,
+      "濺": 79849,
+      "笯": 79850,
+      "杧": 79851,
+      "鞓": 79852,
+      "綫": 79853,
+      "⌟": 79854,
+      "褵": 79855,
+      "澓": 79856,
+      "🤹🏽‍♀️": 79857,
+      "襭": 79858,
+      "👨🏿‍🚒": 79859,
+      "憀": 79860,
+      "簌": 79861,
+      "👶🏻": 79862,
+      "⅚": 79863,
+      "獝": 79864,
+      "嫑": 79865,
+      "摪": 79866,
+      "👨🏾‍💻": 79867,
+      "払": 79868,
+      "鄳": 79869,
+      "藥": 79870,
+      "蟏": 79871,
+      "軔": 79872,
+      "讈": 79873,
+      "佻": 79874,
+      "懌": 79875,
+      "🤦‍♂️": 79876,
+      "紪": 79877,
+      "🫷🏿": 79878,
+      "萖": 79879,
+      "秥": 79880,
+      "👩‍🦱": 79881,
+      "辧": 79882,
+      "疌": 79883,
+      "☹": 79884,
+      "␋": 79885,
+      "≵": 79886,
+      "翱": 79887,
+      "炋": 79888,
+      "👩🏿‍❤️‍💋‍👨🏼": 79889,
+      "箋": 79890,
+      "篠": 79891,
+      "⑴": 79892,
+      "垙": 79893,
+      "⒀": 79894,
+      "螫": 79895,
+      "鳛": 79896,
+      "嫘": 79897,
+      "毭": 79898,
+      "🚴🏼‍♀️": 79899,
+      "矒": 79900,
+      "👇🏽": 79901,
+      "櫫": 79902,
+      "欸": 79903,
+      "絏": 79904,
+      "⃧": 79905,
+      "曌": 79906,
+      "🚣🏿‍♀️": 79907,
+      "鷼": 79908,
+      "椼": 79909,
+      "匼": 79910,
+      "灅": 79911,
+      "橺": 79912,
+      "齭": 79913,
+      "袻": 79914,
+      "👩‍⚖️": 79915,
+      "😢": 79916,
+      "壯": 79917,
+      "橓": 79918,
+      "黿": 79919,
+      "頇": 79920,
+      "碁": 79921,
+      "蝍": 79922,
+      "尪": 79923,
+      "岵": 79924,
+      "酘": 79925,
+      "晜": 79926,
+      "œ": 79927,
+      "閦": 79928,
+      "缒": 79929,
+      "皕": 79930,
+      "碹": 79931,
+      "↔": 79932,
+      "覂": 79933,
+      "髇": 79934,
+      "灦": 79935,
+      "╟": 79936,
+      "曑": 79937,
+      "₷": 79938,
+      "袸": 79939,
+      "诩": 79940,
+      "☠": 79941,
+      "🍋": 79942,
+      "姛": 79943,
+      "蓖": 79944,
+      "圻": 79945,
+      "彟": 79946,
+      "👮🏻‍♀️": 79947,
+      "🧑🏿‍🦰": 79948,
+      "🪼": 79949,
+      "🧔🏼‍♂️": 79950,
+      "▅": 79951,
+      "枧": 79952,
+      "艅": 79953,
+      "̀": 79954,
+      "詜": 79955,
+      "◔": 79956,
+      "陗": 79957,
+      "彃": 79958,
+      "牖": 79959,
+      "竱": 79960,
+      "鱇": 79961,
+      "嶡": 79962,
+      "渜": 79963,
+      "🌗": 79964,
+      "Ć": 79965,
+      "檿": 79966,
+      "塦": 79967,
+      "娿": 79968,
+      "餙": 79969,
+      "🕺🏿": 79970,
+      "😍": 79971,
+      "澮": 79972,
+      "⒄": 79973,
+      "鄲": 79974,
+      "劤": 79975,
+      "⌉": 79976,
+      "撅": 79977,
+      "钿": 79978,
+      "譥": 79979,
+      "鴹": 79980,
+      "訹": 79981,
+      "盌": 79982,
+      "燂": 79983,
+      "矉": 79984,
+      "⍱": 79985,
+      "熭": 79986,
+      "Ǝ": 79987,
+      "蹖": 79988,
+      "妏": 79989,
+      "黡": 79990,
+      "瀔": 79991,
+      "剎": 79992,
+      "₢": 79993,
+      "镧": 79994,
+      "夡": 79995,
+      "飗": 79996,
+      "髏": 79997,
+      "踰": 79998,
+      "怏": 79999,
+      "棫": 80000,
+      "铰": 80001,
+      "絽": 80002,
+      "🤽‍♂️": 80003,
+      "翬": 80004,
+      "圸": 80005,
+      "欇": 80006,
+      "瀡": 80007,
+      "🏄": 80008,
+      "侓": 80009,
+      "冝": 80010,
+      "儠": 80011,
+      "嘆": 80012,
+      "憡": 80013,
+      "劶": 80014,
+      "Ë": 80015,
+      "盕": 80016,
+      "黽": 80017,
+      "👩‍👦‍👦": 80018,
+      "揔": 80019,
+      "僫": 80020,
+      "泐": 80021,
+      "蝄": 80022,
+      "鴏": 80023,
+      "珨": 80024,
+      "襋": 80025,
+      "笞": 80026,
+      "夑": 80027,
+      "塙": 80028,
+      "🖖🏽": 80029,
+      "冽": 80030,
+      "椢": 80031,
+      "訌": 80032,
+      "⌙": 80033,
+      "脫": 80034,
+      "俻": 80035,
+      "輰": 80036,
+      "🦤": 80037,
+      "頺": 80038,
+      "骔": 80039,
+      "谂": 80040,
+      "ǝ": 80041,
+      "櫌": 80042,
+      "镔": 80043,
+      "ɏ": 80044,
+      "虒": 80045,
+      "˞": 80046,
+      "梷": 80047,
+      "枴": 80048,
+      "ⅽ": 80049,
+      "韺": 80050,
+      "簢": 80051,
+      "礱": 80052,
+      "霯": 80053,
+      "椔": 80054,
+      "桖": 80055,
+      "濶": 80056,
+      "檭": 80057,
+      "聙": 80058,
+      "娽": 80059,
+      "∇": 80060,
+      "曤": 80061,
+      "筥": 80062,
+      "⌇": 80063,
+      "縈": 80064,
+      "餯": 80065,
+      "擈": 80066,
+      "籨": 80067,
+      "殞": 80068,
+      "騲": 80069,
+      "🌱": 80070,
+      "🧑🏼": 80071,
+      "🌼": 80072,
+      "粎": 80073,
+      "醙": 80074,
+      "壏": 80075,
+      "筃": 80076,
+      "魒": 80077,
+      "鮟": 80078,
+      "祆": 80079,
+      "佔": 80080,
+      "🙎🏿‍♀️": 80081,
+      "阗": 80082,
+      "🗄": 80083,
+      "侪": 80084,
+      "譽": 80085,
+      "Ð": 80086,
+      "泯": 80087,
+      "啅": 80088,
+      "楯": 80089,
+      "簅": 80090,
+      "驧": 80091,
+      "躟": 80092,
+      "🧔🏼‍♀️": 80093,
+      "💂🏽‍♀️": 80094,
+      "⎹": 80095,
+      "跊": 80096,
+      "🧍🏿": 80097,
+      "蛪": 80098,
+      "␫": 80099,
+      "瞺": 80100,
+      "猍": 80101,
+      "婂": 80102,
+      "榚": 80103,
+      "汃": 80104,
+      "偮": 80105,
+      "滧": 80106,
+      "铞": 80107,
+      "扗": 80108,
+      "壘": 80109,
+      "薱": 80110,
+      "輞": 80111,
+      "爼": 80112,
+      "橸": 80113,
+      "巑": 80114,
+      "骯": 80115,
+      "🧜🏿": 80116,
+      "藯": 80117,
+      "撴": 80118,
+      "": 80119,
+      "虱": 80120,
+      "譚": 80121,
+      "蠏": 80122,
+      "👵🏽": 80123,
+      "👨‍⚕️": 80124,
+      "羸": 80125,
+      "筕": 80126,
+      "嵁": 80127,
+      "栄": 80128,
+      "👱🏾‍♀️": 80129,
+      "👨🏾‍🦱": 80130,
+      "💆🏻‍♀️": 80131,
+      "⋴": 80132,
+      "钷": 80133,
+      "昩": 80134,
+      "蕍": 80135,
+      "胗": 80136,
+      "桅": 80137,
+      "╈": 80138,
+      "憷": 80139,
+      "🚶🏼‍♂️": 80140,
+      "攪": 80141,
+      "蚧": 80142,
+      "鋮": 80143,
+      "偻": 80144,
+      "Ⓜ": 80145,
+      "🧍🏻‍♀️": 80146,
+      "👩‍🦼": 80147,
+      "傑": 80148,
+      "參": 80149,
+      "⎘": 80150,
+      "✊": 80151,
+      "🌘": 80152,
+      "沺": 80153,
+      "熰": 80154,
+      "👩‍❤️‍💋‍👨": 80155,
+      "耆": 80156,
+      "録": 80157,
+      "⊍": 80158,
+      "觀": 80159,
+      "👩‍🚒": 80160,
+      "⇧": 80161,
+      "熦": 80162,
+      "😠": 80163,
+      "鋽": 80164,
+      "📻": 80165,
+      "皽": 80166,
+      "👨🏼‍❤️‍💋‍👨🏼": 80167,
+      "͖": 80168,
+      "🙏🏾": 80169,
+      "瓨": 80170,
+      "摑": 80171,
+      "💆🏿‍♀️": 80172,
+      "緽": 80173,
+      "衹": 80174,
+      "虇": 80175,
+      "🌄": 80176,
+      "枟": 80177,
+      "潞": 80178,
+      "秚": 80179,
+      "誃": 80180,
+      "蹍": 80181,
+      "餅": 80182,
+      "⇶": 80183,
+      "慙": 80184,
+      "␪": 80185,
+      "瓭": 80186,
+      "菼": 80187,
+      "⏠": 80188,
+      "诃": 80189,
+      "蟴": 80190,
+      "炏": 80191,
+      "蟎": 80192,
+      "👩🏽‍✈️": 80193,
+      "阂": 80194,
+      "韏": 80195,
+      "ȯ": 80196,
+      "錼": 80197,
+      "⊳": 80198,
+      "鬦": 80199,
+      "👷🏻‍♀️": 80200,
+      "筚": 80201,
+      "涹": 80202,
+      "🤧": 80203,
+      "↾": 80204,
+      "颲": 80205,
+      "慗": 80206,
+      "棡": 80207,
+      "╆": 80208,
+      "帙": 80209,
+      "執": 80210,
+      "☏": 80211,
+      "惀": 80212,
+      "胏": 80213,
+      "巽": 80214,
+      "議": 80215,
+      "珶": 80216,
+      "秐": 80217,
+      "糲": 80218,
+      "🇦🇸": 80219,
+      "🧑🏾‍🎨": 80220,
+      "🚎": 80221,
+      "繴": 80222,
+      "団": 80223,
+      "蚨": 80224,
+      "咰": 80225,
+      "˿": 80226,
+      "醎": 80227,
+      "唥": 80228,
+      "╬": 80229,
+      "揫": 80230,
+      "🚞": 80231,
+      "偐": 80232,
+      "阊": 80233,
+      "鰀": 80234,
+      "馔": 80235,
+      "🇲🇱": 80236,
+      "晵": 80237,
+      "薹": 80238,
+      "潷": 80239,
+      "掂": 80240,
+      "⋚": 80241,
+      "綇": 80242,
+      "偳": 80243,
+      "摓": 80244,
+      "🔋": 80245,
+      "🦊": 80246,
+      "擳": 80247,
+      "鵽": 80248,
+      "␔": 80249,
+      "🛀🏾": 80250,
+      "貲": 80251,
+      "憲": 80252,
+      "👮": 80253,
+      "鷅": 80254,
+      "嫛": 80255,
+      "篴": 80256,
+      "甾": 80257,
+      "珵": 80258,
+      "🪱": 80259,
+      "💂🏻‍♀️": 80260,
+      "癵": 80261,
+      "ₜ": 80262,
+      "睂": 80263,
+      "躋": 80264,
+      "🫱🏿‍🫲🏾": 80265,
+      "墿": 80266,
+      "帪": 80267,
+      "冃": 80268,
+      "轱": 80269,
+      "谾": 80270,
+      "佱": 80271,
+      "淐": 80272,
+      "鲹": 80273,
+      "済": 80274,
+      "摽": 80275,
+      "廙": 80276,
+      "柦": 80277,
+      "🍳": 80278,
+      "🏰": 80279,
+      "Ǜ": 80280,
+      "堾": 80281,
+      "篅": 80282,
+      "趸": 80283,
+      "ₑ": 80284,
+      "涇": 80285,
+      "桧": 80286,
+      "樏": 80287,
+      "閬": 80288,
+      "💬": 80289,
+      "⑕": 80290,
+      "🫱🏿‍🫲🏽": 80291,
+      "焝": 80292,
+      "🐩": 80293,
+      "卐": 80294,
+      "🚝": 80295,
+      "灞": 80296,
+      "硋": 80297,
+      "浐": 80298,
+      "烵": 80299,
+      "碷": 80300,
+      "Ⱦ": 80301,
+      "巕": 80302,
+      "笁": 80303,
+      "🤜": 80304,
+      "⎍": 80305,
+      "👩‍👧‍👧": 80306,
+      "繮": 80307,
+      "蝯": 80308,
+      "儏": 80309,
+      "蓾": 80310,
+      "⌳": 80311,
+      "👩🏾‍❤️‍👩🏽": 80312,
+      "鬕": 80313,
+      "⒣": 80314,
+      "湭": 80315,
+      "韇": 80316,
+      "紕": 80317,
+      "銥": 80318,
+      "錇": 80319,
+      "壞": 80320,
+      "♑": 80321,
+      "煓": 80322,
+      "👩🏾‍❤️‍👨🏽": 80323,
+      "攝": 80324,
+      "Ƅ": 80325,
+      "👩🏽‍🦯": 80326,
+      "鬧": 80327,
+      "℣": 80328,
+      "♤": 80329,
+      "髲": 80330,
+      "🦵🏻": 80331,
+      "獴": 80332,
+      "ʄ": 80333,
+      "埘": 80334,
+      "🚵🏼‍♂️": 80335,
+      "耒": 80336,
+      "🫅🏻": 80337,
+      "曡": 80338,
+      "齬": 80339,
+      "薨": 80340,
+      "恖": 80341,
+      "魌": 80342,
+      "滅": 80343,
+      "酔": 80344,
+      "癙": 80345,
+      "桯": 80346,
+      "脔": 80347,
+      "逄": 80348,
+      "燉": 80349,
+      "慆": 80350,
+      "Ǯ": 80351,
+      "👨🏿‍❤️‍👨🏾": 80352,
+      "黭": 80353,
+      "衞": 80354,
+      "娭": 80355,
+      "💂‍♀️": 80356,
+      "脣": 80357,
+      "櫖": 80358,
+      "奯": 80359,
+      "矃": 80360,
+      "煀": 80361,
+      "🇳🇷": 80362,
+      "烌": 80363,
+      "撈": 80364,
+      "尫": 80365,
+      "⍗": 80366,
+      "隱": 80367,
+      "≛": 80368,
+      "⬆️": 80369,
+      "刟": 80370,
+      "優": 80371,
+      "剰": 80372,
+      "斃": 80373,
+      "鯃": 80374,
+      "媸": 80375,
+      "葌": 80376,
+      "蕶": 80377,
+      "帯": 80378,
+      "掿": 80379,
+      "氀": 80380,
+      "甓": 80381,
+      "誢": 80382,
+      "粴": 80383,
+      "藂": 80384,
+      "壄": 80385,
+      "ĭ": 80386,
+      "崘": 80387,
+      "奲": 80388,
+      "濚": 80389,
+      "痃": 80390,
+      "喢": 80391,
+      "麭": 80392,
+      "莔": 80393,
+      "滳": 80394,
+      "颰": 80395,
+      "贅": 80396,
+      "🟪": 80397,
+      "謇": 80398,
+      "鱆": 80399,
+      "⃺": 80400,
+      "駭": 80401,
+      "侽": 80402,
+      "礢": 80403,
+      "␣": 80404,
+      "ɭ": 80405,
+      "🥬": 80406,
+      "痭": 80407,
+      "鰰": 80408,
+      "蘙": 80409,
+      "苉": 80410,
+      "🧍🏼‍♂️": 80411,
+      "Ⓓ": 80412,
+      "喑": 80413,
+      "💽": 80414,
+      "懺": 80415,
+      "馼": 80416,
+      "祪": 80417,
+      "铳": 80418,
+      "挢": 80419,
+      "牏": 80420,
+      "🧕": 80421,
+      "靏": 80422,
+      "緲": 80423,
+      "🚶🏻‍♀️": 80424,
+      "緹": 80425,
+      "˾": 80426,
+      "偨": 80427,
+      "翸": 80428,
+      "苨": 80429,
+      "#️⃣": 80430,
+      "👩🏾‍🚒": 80431,
+      "屣": 80432,
+      "駣": 80433,
+      "夆": 80434,
+      "沬": 80435,
+      "緆": 80436,
+      "匨": 80437,
+      "溓": 80438,
+      "∋": 80439,
+      "艜": 80440,
+      "隿": 80441,
+      "鮣": 80442,
+      "🦻🏽": 80443,
+      "駟": 80444,
+      "吅": 80445,
+      "觱": 80446,
+      "韻": 80447,
+      "顄": 80448,
+      "饆": 80449,
+      "瞐": 80450,
+      "洉": 80451,
+      "覝": 80452,
+      "扷": 80453,
+      "逹": 80454,
+      "審": 80455,
+      "锗": 80456,
+      "鯦": 80457,
+      "俈": 80458,
+      "瘸": 80459,
+      "炆": 80460,
+      "⤴️": 80461,
+      "嫜": 80462,
+      "⒌": 80463,
+      "␭": 80464,
+      "冴": 80465,
+      "岬": 80466,
+      "栟": 80467,
+      "鎥": 80468,
+      "乆": 80469,
+      "觶": 80470,
+      "戤": 80471,
+      "噷": 80472,
+      "膶": 80473,
+      "匑": 80474,
+      "≦": 80475,
+      "▸": 80476,
+      "🧑🏿‍🌾": 80477,
+      "潀": 80478,
+      "Ǩ": 80479,
+      "榇": 80480,
+      "朜": 80481,
+      "襓": 80482,
+      "鉡": 80483,
+      "炣": 80484,
+      "鯿": 80485,
+      "敽": 80486,
+      "🥘": 80487,
+      "鍮": 80488,
+      "暦": 80489,
+      "瀽": 80490,
+      "衊": 80491,
+      "⇃": 80492,
+      "🇬🇫": 80493,
+      "娹": 80494,
+      "🫷🏾": 80495,
+      "塮": 80496,
+      "🏃🏾‍♀️": 80497,
+      "孎": 80498,
+      "👋🏾": 80499,
+      "埞": 80500,
+      "涱": 80501,
+      "晊": 80502,
+      "廤": 80503,
+      "僩": 80504,
+      "剋": 80505,
+      "薣": 80506,
+      "渰": 80507,
+      "葞": 80508,
+      "▇": 80509,
+      "蹯": 80510,
+      "刼": 80511,
+      "閘": 80512,
+      "🌦": 80513,
+      "蜮": 80514,
+      "衶": 80515,
+      "軼": 80516,
+      "資": 80517,
+      "🐳": 80518,
+      "🦹🏻‍♀️": 80519,
+      "屪": 80520,
+      "籟": 80521,
+      "⇙": 80522,
+      "嚈": 80523,
+      "荦": 80524,
+      "🧘": 80525,
+      "⛹🏽‍♀️": 80526,
+      "捌": 80527,
+      "泧": 80528,
+      "崻": 80529,
+      "暊": 80530,
+      "鴘": 80531,
+      "觮": 80532,
+      "蚛": 80533,
+      "鐂": 80534,
+      "醽": 80535,
+      "忒": 80536,
+      "🤹🏽‍♂️": 80537,
+      "˶": 80538,
+      "蠘": 80539,
+      "👮🏽‍♂️": 80540,
+      "膍": 80541,
+      "🤽🏼‍♀️": 80542,
+      "椌": 80543,
+      "👰🏾‍♂️": 80544,
+      "畀": 80545,
+      "泃": 80546,
+      "暅": 80547,
+      "ȉ": 80548,
+      "幬": 80549,
+      "竨": 80550,
+      "🎬": 80551,
+      "↷": 80552,
+      "廰": 80553,
+      "棓": 80554,
+      "殦": 80555,
+      "婲": 80556,
+      "魞": 80557,
+      "芖": 80558,
+      "👐🏽": 80559,
+      "鉉": 80560,
+      "錙": 80561,
+      "蓈": 80562,
+      "⎛": 80563,
+      "◴": 80564,
+      "葰": 80565,
+      "狏": 80566,
+      "ɡ": 80567,
+      "嵃": 80568,
+      "橜": 80569,
+      "酐": 80570,
+      "掞": 80571,
+      "👱🏼‍♀️": 80572,
+      "槜": 80573,
+      "憵": 80574,
+      "藇": 80575,
+      "燸": 80576,
+      "聇": 80577,
+      "裊": 80578,
+      "湋": 80579,
+      "覐": 80580,
+      "羄": 80581,
+      "🧖🏼‍♂️": 80582,
+      "虩": 80583,
+      "℧": 80584,
+      "📉": 80585,
+      "♙": 80586,
+      "◾": 80587,
+      "緊": 80588,
+      "筈": 80589,
+      "🧝‍♂️": 80590,
+      "筦": 80591,
+      "頰": 80592,
+      "磑": 80593,
+      "嵍": 80594,
+      "🏊🏼‍♂️": 80595,
+      "隳": 80596,
+      "杔": 80597,
+      "穟": 80598,
+      "ͻ": 80599,
+      "鱵": 80600,
+      "璻": 80601,
+      "討": 80602,
+      "窉": 80603,
+      "槬": 80604,
+      "犒": 80605,
+      "狊": 80606,
+      "儉": 80607,
+      "釢": 80608,
+      "碕": 80609,
+      "♭": 80610,
+      "✌🏼": 80611,
+      "鑄": 80612,
+      "儤": 80613,
+      "螈": 80614,
+      "籞": 80615,
+      "🦑": 80616,
+      "豓": 80617,
+      "饨": 80618,
+      "摼": 80619,
+      "裶": 80620,
+      "⓭": 80621,
+      "渱": 80622,
+      "蝂": 80623,
+      "乤": 80624,
+      "🫄🏻": 80625,
+      "灄": 80626,
+      "🪣": 80627,
+      "笵": 80628,
+      "佚": 80629,
+      "🙆‍♂️": 80630,
+      "苂": 80631,
+      "粦": 80632,
+      "櫋": 80633,
+      "嘔": 80634,
+      "⇀": 80635,
+      "蹗": 80636,
+      "跒": 80637,
+      "铫": 80638,
+      "嶓": 80639,
+      "擴": 80640,
+      "✔️": 80641,
+      "✌🏻": 80642,
+      "弚": 80643,
+      "⃥": 80644,
+      "獏": 80645,
+      "͎": 80646,
+      "涐": 80647,
+      "✋🏼": 80648,
+      "鹐": 80649,
+      "🚣🏽‍♂️": 80650,
+      "桇": 80651,
+      "揸": 80652,
+      "⛹🏼‍♂️": 80653,
+      "👩🏻": 80654,
+      "💷": 80655,
+      "嚻": 80656,
+      "洿": 80657,
+      "骳": 80658,
+      "腲": 80659,
+      "襌": 80660,
+      "虳": 80661,
+      "⒏": 80662,
+      "竕": 80663,
+      "🔎": 80664,
+      "牦": 80665,
+      "儘": 80666,
+      "柅": 80667,
+      "鰞": 80668,
+      "∕": 80669,
+      "醫": 80670,
+      "簒": 80671,
+      "鬜": 80672,
+      "苾": 80673,
+      "峛": 80674,
+      "喦": 80675,
+      "騍": 80676,
+      "🧑🏼‍🍼": 80677,
+      "唂": 80678,
+      "🩱": 80679,
+      "鄎": 80680,
+      "椾": 80681,
+      "誖": 80682,
+      "🧟‍♀️": 80683,
+      "⁍": 80684,
+      "蕧": 80685,
+      "瀎": 80686,
+      "🇸🇪": 80687,
+      "👍🏾": 80688,
+      "💟": 80689,
+      "咅": 80690,
+      "堄": 80691,
+      "訲": 80692,
+      "🤦🏼‍♂️": 80693,
+      "🍂": 80694,
+      "垹": 80695,
+      "穤": 80696,
+      "儦": 80697,
+      "蟟": 80698,
+      "澃": 80699,
+      "焭": 80700,
+      "嫲": 80701,
+      "罇": 80702,
+      "🙎🏽": 80703,
+      "璁": 80704,
+      "🎸": 80705,
+      "藌": 80706,
+      "勦": 80707,
+      "膸": 80708,
+      "禕": 80709,
+      "Ō": 80710,
+      "̬": 80711,
+      "課": 80712,
+      "槿": 80713,
+      "皛": 80714,
+      "榋": 80715,
+      "🚪": 80716,
+      "🏹": 80717,
+      "👰🏼": 80718,
+      "👩🏾‍🍼": 80719,
+      "闑": 80720,
+      "👩🏿‍🦰": 80721,
+      "👨🏼‍🚀": 80722,
+      "躿": 80723,
+      "⑉": 80724,
+      "🤍": 80725,
+      "楕": 80726,
+      "˗": 80727,
+      "簚": 80728,
+      "暉": 80729,
+      "蝘": 80730,
+      "鰶": 80731,
+      "歵": 80732,
+      "蘟": 80733,
+      "瀢": 80734,
+      "溏": 80735,
+      "🇵🇾": 80736,
+      "螳": 80737,
+      "耂": 80738,
+      "👨🏻‍🔬": 80739,
+      "讱": 80740,
+      "⑍": 80741,
+      "悩": 80742,
+      "鰔": 80743,
+      "畿": 80744,
+      "垈": 80745,
+      "翑": 80746,
+      "🏴": 80747,
+      "ſ": 80748,
+      "🧑🏾‍🔧": 80749,
+      "Ⓝ": 80750,
+      "┗": 80751,
+      "肼": 80752,
+      "掜": 80753,
+      "貊": 80754,
+      "🚘": 80755,
+      "🤸🏼‍♂️": 80756,
+      "👨🏻‍🦳": 80757,
+      "🇩🇲": 80758,
+      "襬": 80759,
+      "踭": 80760,
+      "糾": 80761,
+      "牬": 80762,
+      "毻": 80763,
+      "倫": 80764,
+      "嫯": 80765,
+      "Ƶ": 80766,
+      "╓": 80767,
+      "⒱": 80768,
+      "掎": 80769,
+      "☴": 80770,
+      "剕": 80771,
+      "淓": 80772,
+      "灑": 80773,
+      "纟": 80774,
+      "🎋": 80775,
+      "鴽": 80776,
+      "愴": 80777,
+      "∑": 80778,
+      "魓": 80779,
+      "樌": 80780,
+      "菭": 80781,
+      "ġ": 80782,
+      "🤐": 80783,
+      "🤌🏽": 80784,
+      "✦": 80785,
+      "遟": 80786,
+      "峏": 80787,
+      "ˑ": 80788,
+      "罀": 80789,
+      "揅": 80790,
+      "😪": 80791,
+      "鑘": 80792,
+      "郙": 80793,
+      "鉟": 80794,
+      "🚴‍♂️": 80795,
+      "韮": 80796,
+      "曥": 80797,
+      "洆": 80798,
+      "🫱🏻‍🫲🏽": 80799,
+      "😤": 80800,
+      "鏁": 80801,
+      "╜": 80802,
+      "矟": 80803,
+      "Ⅵ": 80804,
+      "贍": 80805,
+      "斒": 80806,
+      "獜": 80807,
+      "爈": 80808,
+      "⑱": 80809,
+      "霐": 80810,
+      "畑": 80811,
+      "躡": 80812,
+      "柇": 80813,
+      "闣": 80814,
+      "決": 80815,
+      "哿": 80816,
+      "袨": 80817,
+      "蒩": 80818,
+      "纴": 80819,
+      "綀": 80820,
+      "峸": 80821,
+      "挕": 80822,
+      "◶": 80823,
+      "🚵‍♀️": 80824,
+      "跜": 80825,
+      "蔙": 80826,
+      "✋🏿": 80827,
+      "乯": 80828,
+      "扊": 80829,
+      "犼": 80830,
+      "仂": 80831,
+      "榮": 80832,
+      "繶": 80833,
+      "🀄️": 80834,
+      "揜": 80835,
+      "吽": 80836,
+      "呄": 80837,
+      "肒": 80838,
+      "刵": 80839,
+      "銖": 80840,
+      "⇑": 80841,
+      "爄": 80842,
+      "吳": 80843,
+      "枠": 80844,
+      "邎": 80845,
+      "🌫": 80846,
+      "‥": 80847,
+      "舖": 80848,
+      "ƀ": 80849,
+      "🇨🇳": 80850,
+      "盻": 80851,
+      "蕈": 80852,
+      "ȅ": 80853,
+      "籛": 80854,
+      "郓": 80855,
+      "麠": 80856,
+      "憃": 80857,
+      "宩": 80858,
+      "觻": 80859,
+      "赅": 80860,
+      "穸": 80861,
+      "妼": 80862,
+      "铐": 80863,
+      "浥": 80864,
+      "旽": 80865,
+      "歡": 80866,
+      "齤": 80867,
+      "呇": 80868,
+      "喳": 80869,
+      "巣": 80870,
+      "蠅": 80871,
+      "饞": 80872,
+      "銷": 80873,
+      "滮": 80874,
+      "🦃": 80875,
+      "惴": 80876,
+      "視": 80877,
+      "耩": 80878,
+      "🔀": 80879,
+      "傢": 80880,
+      "眧": 80881,
+      "柛": 80882,
+      "懋": 80883,
+      "陜": 80884,
+      "⇖": 80885,
+      "齊": 80886,
+      "婼": 80887,
+      "際": 80888,
+      "噴": 80889,
+      "倔": 80890,
+      "₴": 80891,
+      "蚿": 80892,
+      "🧙": 80893,
+      "邈": 80894,
+      "煋": 80895,
+      "糛": 80896,
+      "⋾": 80897,
+      "👰🏽": 80898,
+      "✋": 80899,
+      "纾": 80900,
+      "荨": 80901,
+      "霅": 80902,
+      "訩": 80903,
+      "Ⅽ": 80904,
+      "鉷": 80905,
+      "魩": 80906,
+      "⍒": 80907,
+      "珌": 80908,
+      "箓": 80909,
+      "埸": 80910,
+      "欚": 80911,
+      "啘": 80912,
+      "秊": 80913,
+      "₫": 80914,
+      "灺": 80915,
+      "💇🏿‍♂️": 80916,
+      "轙": 80917,
+      "Ι": 80918,
+      "狯": 80919,
+      "🅱️": 80920,
+      "恊": 80921,
+      "恱": 80922,
+      "瑌": 80923,
+      "莦": 80924,
+      "湠": 80925,
+      "菣": 80926,
+      "蓌": 80927,
+      "痶": 80928,
+      "畤": 80929,
+      "啞": 80930,
+      "🫸🏾": 80931,
+      "泘": 80932,
+      "躠": 80933,
+      "厒": 80934,
+      "巌": 80935,
+      "爇": 80936,
+      "墁": 80937,
+      "魈": 80938,
+      "諣": 80939,
+      "⍺": 80940,
+      "鏄": 80941,
+      "蓏": 80942,
+      "湼": 80943,
+      "嚆": 80944,
+      "鉔": 80945,
+      "傯": 80946,
+      "壗": 80947,
+      "榅": 80948,
+      "ʴ": 80949,
+      "琇": 80950,
+      "讃": 80951,
+      "鰡": 80952,
+      "⎺": 80953,
+      "̻": 80954,
+      "迚": 80955,
+      "渨": 80956,
+      "颴": 80957,
+      "糼": 80958,
+      "🛡": 80959,
+      "倓": 80960,
+      "躊": 80961,
+      "順": 80962,
+      "🔖": 80963,
+      "⛹🏻‍♂️": 80964,
+      "躄": 80965,
+      "≍": 80966,
+      "婙": 80967,
+      "駱": 80968,
+      "羝": 80969,
+      "喍": 80970,
+      "漬": 80971,
+      "夝": 80972,
+      "🇲🇺": 80973,
+      "矙": 80974,
+      "鶁": 80975,
+      "忕": 80976,
+      "觔": 80977,
+      "嗺": 80978,
+      "噤": 80979,
+      "堚": 80980,
+      "嗱": 80981,
+      "鳅": 80982,
+      "鮻": 80983,
+      "淃": 80984,
+      "甌": 80985,
+      "顙": 80986,
+      "♻": 80987,
+      "婘": 80988,
+      "⛔": 80989,
+      "👨‍🎓": 80990,
+      "巬": 80991,
+      "🤸🏻‍♂️": 80992,
+      "塽": 80993,
+      "🧑🏻": 80994,
+      "壠": 80995,
+      "垱": 80996,
+      "鑚": 80997,
+      "菟": 80998,
+      "瑑": 80999,
+      "👨🏿‍🦯": 81000,
+      "煃": 81001,
+      "⌎": 81002,
+      "贆": 81003,
+      "慜": 81004,
+      "赭": 81005,
+      "Ô": 81006,
+      "侾": 81007,
+      "鄬": 81008,
+      "鞷": 81009,
+      "膕": 81010,
+      "畠": 81011,
+      "褱": 81012,
+      "檴": 81013,
+      "欉": 81014,
+      "衯": 81015,
+      "🏊🏻": 81016,
+      "賜": 81017,
+      "␬": 81018,
+      "閹": 81019,
+      "珢": 81020,
+      "嗩": 81021,
+      "殁": 81022,
+      "鳼": 81023,
+      "檇": 81024,
+      "佲": 81025,
+      "🤚🏾": 81026,
+      "👨🏿‍🦰": 81027,
+      "綖": 81028,
+      "臼": 81029,
+      "艬": 81030,
+      "鯮": 81031,
+      "硩": 81032,
+      "蕎": 81033,
+      "蒥": 81034,
+      "帰": 81035,
+      "齒": 81036,
+      "孬": 81037,
+      "🇺🇸": 81038,
+      "╩": 81039,
+      "閁": 81040,
+      "🚅": 81041,
+      "嶣": 81042,
+      "銞": 81043,
+      "唜": 81044,
+      "臍": 81045,
+      "犦": 81046,
+      "钍": 81047,
+      "茠": 81048,
+      "蝹": 81049,
+      "暈": 81050,
+      "癮": 81051,
+      "👷🏽": 81052,
+      "蔊": 81053,
+      "旒": 81054,
+      "🤸🏻": 81055,
+      "傁": 81056,
+      "蓗": 81057,
+      "秭": 81058,
+      "犐": 81059,
+      "遾": 81060,
+      "歲": 81061,
+      "🤌🏼": 81062,
+      "👨🏻‍🦯": 81063,
+      "鄆": 81064,
+      "鰉": 81065,
+      "翇": 81066,
+      "噽": 81067,
+      "ʳ": 81068,
+      "鉾": 81069,
+      "↌": 81070,
+      "鶜": 81071,
+      "汢": 81072,
+      "謑": 81073,
+      "詨": 81074,
+      "翭": 81075,
+      "詼": 81076,
+      "芇": 81077,
+      "蛨": 81078,
+      "ɪ": 81079,
+      "🇭🇰": 81080,
+      "📄": 81081,
+      "锪": 81082,
+      "👨🏻‍🦽": 81083,
+      "🧑🏽‍❤️‍🧑🏼": 81084,
+      "🚣🏽‍♀️": 81085,
+      "▋": 81086,
+      "₣": 81087,
+      "鬫": 81088,
+      "麐": 81089,
+      "慃": 81090,
+      "锱": 81091,
+      "戛": 81092,
+      "赪": 81093,
+      "🚓": 81094,
+      "僙": 81095,
+      "柖": 81096,
+      "籘": 81097,
+      "ª": 81098,
+      "訐": 81099,
+      "墴": 81100,
+      "ϔ": 81101,
+      "捗": 81102,
+      "👨🏾‍❤️‍👨🏻": 81103,
+      "煏": 81104,
+      "坉": 81105,
+      "̆": 81106,
+      "鉗": 81107,
+      "襏": 81108,
+      "稊": 81109,
+      "睄": 81110,
+      "扽": 81111,
+      "🎪": 81112,
+      "癏": 81113,
+      "🎅🏼": 81114,
+      "冾": 81115,
+      "🇪🇬": 81116,
+      "梜": 81117,
+      "蠾": 81118,
+      "怹": 81119,
+      "鵿": 81120,
+      "兎": 81121,
+      "塳": 81122,
+      "轢": 81123,
+      "🚻": 81124,
+      "嶸": 81125,
+      "鞩": 81126,
+      "⁃": 81127,
+      "離": 81128,
+      "漲": 81129,
+      "鶇": 81130,
+      "敐": 81131,
+      "漇": 81132,
+      "恠": 81133,
+      "鎻": 81134,
+      "兌": 81135,
+      "ʻ": 81136,
+      "咮": 81137,
+      "唼": 81138,
+      "襖": 81139,
+      "誜": 81140,
+      "獙": 81141,
+      "駏": 81142,
+      "閅": 81143,
+      "鵬": 81144,
+      "槧": 81145,
+      "弝": 81146,
+      "铓": 81147,
+      "Ĵ": 81148,
+      "鑐": 81149,
+      "𓀈": 81150,
+      "譅": 81151,
+      "👩🏾‍🏫": 81152,
+      "Ȣ": 81153,
+      "趡": 81154,
+      "歍": 81155,
+      "🦹": 81156,
+      "債": 81157,
+      "ʗ": 81158,
+      "壃": 81159,
+      "餳": 81160,
+      "夅": 81161,
+      "怤": 81162,
+      "襼": 81163,
+      "鈙": 81164,
+      "孠": 81165,
+      "袈": 81166,
+      "👨🏻‍❤️‍💋‍👨🏼": 81167,
+      "從": 81168,
+      "◙": 81169,
+      "丗": 81170,
+      "釣": 81171,
+      "縧": 81172,
+      "蚍": 81173,
+      "攭": 81174,
+      "匂": 81175,
+      "茍": 81176,
+      "🇸🇰": 81177,
+      "殩": 81178,
+      "湞": 81179,
+      "鋆": 81180,
+      "🧙🏼‍♀️": 81181,
+      "眂": 81182,
+      "鸻": 81183,
+      "✍🏿": 81184,
+      "鬣": 81185,
+      "鄭": 81186,
+      "楤": 81187,
+      "稹": 81188,
+      "🕵🏾": 81189,
+      "柉": 81190,
+      "婖": 81191,
+      "🇬🇪": 81192,
+      "🐵": 81193,
+      "爯": 81194,
+      "Ʃ": 81195,
+      "👩🏼‍❤️‍💋‍👩🏼": 81196,
+      "頡": 81197,
+      "臫": 81198,
+      "姰": 81199,
+      "煊": 81200,
+      "缄": 81201,
+      "韣": 81202,
+      "↙": 81203,
+      "筮": 81204,
+      "ˈ": 81205,
+      "Ř": 81206,
+      "👩🏽‍❤️‍👩🏻": 81207,
+      "欩": 81208,
+      "菍": 81209,
+      "鵟": 81210,
+      "℗": 81211,
+      "藅": 81212,
+      "塷": 81213,
+      "軠": 81214,
+      "踁": 81215,
+      "篑": 81216,
+      "硦": 81217,
+      "ℯ": 81218,
+      "峓": 81219,
+      "鵹": 81220,
+      "雹": 81221,
+      "榌": 81222,
+      "▬": 81223,
+      "👨‍🍼": 81224,
+      "塧": 81225,
+      "躔": 81226,
+      "ͥ": 81227,
+      "猌": 81228,
+      "罷": 81229,
+      "冣": 81230,
+      "黚": 81231,
+      "芘": 81232,
+      "膂": 81233,
+      "👨🏾‍🚒": 81234,
+      "抔": 81235,
+      "👩🏿‍✈️": 81236,
+      "萓": 81237,
+      "媔": 81238,
+      "佴": 81239,
+      "蒄": 81240,
+      "儁": 81241,
+      "袮": 81242,
+      "缜": 81243,
+      "亝": 81244,
+      "豙": 81245,
+      "👜": 81246,
+      "‮": 81247,
+      "紖": 81248,
+      "纆": 81249,
+      "唗": 81250,
+      "Ɗ": 81251,
+      "楦": 81252,
+      "鏍": 81253,
+      "⌱": 81254,
+      "屆": 81255,
+      "簳": 81256,
+      "ǯ": 81257,
+      "侷": 81258,
+      "砃": 81259,
+      "鋎": 81260,
+      "潔": 81261,
+      "╘": 81262,
+      "儞": 81263,
+      "櫒": 81264,
+      "艨": 81265,
+      "屺": 81266,
+      "🍒": 81267,
+      "炦": 81268,
+      "櫮": 81269,
+      "秝": 81270,
+      "檼": 81271,
+      "茼": 81272,
+      "̯": 81273,
+      "荻": 81274,
+      "嵎": 81275,
+      "瑷": 81276,
+      "⋍": 81277,
+      "傖": 81278,
+      "🙌🏿": 81279,
+      "菗": 81280,
+      "彐": 81281,
+      "璸": 81282,
+      "鼄": 81283,
+      "襒": 81284,
+      "鑡": 81285,
+      "🤦‍♀️": 81286,
+      "侀": 81287,
+      "蟋": 81288,
+      "Ⓦ": 81289,
+      "蹥": 81290,
+      "愜": 81291,
+      "訋": 81292,
+      "绱": 81293,
+      "鵭": 81294,
+      "ͱ": 81295,
+      "骖": 81296,
+      "菘": 81297,
+      "杁": 81298,
+      "緪": 81299,
+      "鬉": 81300,
+      "瞂": 81301,
+      "⋼": 81302,
+      "繍": 81303,
+      "許": 81304,
+      "藎": 81305,
+      "🧏🏻": 81306,
+      "歂": 81307,
+      "驞": 81308,
+      "艹": 81309,
+      "終": 81310,
+      "🫶🏼": 81311,
+      "⒩": 81312,
+      "₤": 81313,
+      "譹": 81314,
+      "龠": 81315,
+      "🕉": 81316,
+      "鹑": 81317,
+      "👨🏼‍🦯": 81318,
+      "諙": 81319,
+      "褫": 81320,
+      "觵": 81321,
+      "蓕": 81322,
+      "墶": 81323,
+      "Ƀ": 81324,
+      "ή": 81325,
+      "♉": 81326,
+      "⁛": 81327,
+      "駖": 81328,
+      "俦": 81329,
+      "⑃": 81330,
+      "🧑🏽‍🏫": 81331,
+      "🇦🇼": 81332,
+      "⚽": 81333,
+      "🧍🏾‍♀️": 81334,
+      "櫇": 81335,
+      "諮": 81336,
+      "縏": 81337,
+      "脞": 81338,
+      "繿": 81339,
+      "氶": 81340,
+      "🇮🇳": 81341,
+      "丼": 81342,
+      "藞": 81343,
+      "醨": 81344,
+      "廄": 81345,
+      "嶯": 81346,
+      "趢": 81347,
+      "🧙🏾‍♀️": 81348,
+      "🧑‍⚖️": 81349,
+      "🧑🏿‍🏭": 81350,
+      "媋": 81351,
+      "忺": 81352,
+      "🛁": 81353,
+      "⁼": 81354,
+      "諠": 81355,
+      "🧝🏽‍♀️": 81356,
+      "͒": 81357,
+      "↞": 81358,
+      "媘": 81359,
+      "鸙": 81360,
+      "埍": 81361,
+      "🧜🏼‍♂️": 81362,
+      "嵉": 81363,
+      "橉": 81364,
+      "🧚🏼": 81365,
+      "曮": 81366,
+      "辡": 81367,
+      "螔": 81368,
+      "犉": 81369,
+      "稨": 81370,
+      "崂": 81371,
+      "鍬": 81372,
+      "疀": 81373,
+      "👌": 81374,
+      "腨": 81375,
+      "鶼": 81376,
+      "壍": 81377,
+      "麚": 81378,
+      "屴": 81379,
+      "👩‍🦳": 81380,
+      "夣": 81381,
+      "娰": 81382,
+      "貶": 81383,
+      "翺": 81384,
+      "癐": 81385,
+      "⇈": 81386,
+      "👩🏾‍❤️‍👩🏻": 81387,
+      "⚰": 81388,
+      "👩🏻‍🏫": 81389,
+      "褹": 81390,
+      "熺": 81391,
+      "夗": 81392,
+      "檨": 81393,
+      "怐": 81394,
+      "槚": 81395,
+      "▴": 81396,
+      "圝": 81397,
+      "ℬ": 81398,
+      "竦": 81399,
+      "🙇🏻": 81400,
+      "疳": 81401,
+      "🐕": 81402,
+      "🇮🇷": 81403,
+      "養": 81404,
+      "👨🏻‍❤️‍👨🏽": 81405,
+      "丣": 81406,
+      "莡": 81407,
+      "譳": 81408,
+      "餤": 81409,
+      "蠦": 81410,
+      "⁤": 81411,
+      "咵": 81412,
+      "荊": 81413,
+      "媾": 81414,
+      "聶": 81415,
+      "↏": 81416,
+      "毿": 81417,
+      "铥": 81418,
+      "橹": 81419,
+      "鼼": 81420,
+      "骦": 81421,
+      "嘺": 81422,
+      "暫": 81423,
+      "譨": 81424,
+      "詚": 81425,
+      "踹": 81426,
+      "⃭": 81427,
+      "鬑": 81428,
+      "図": 81429,
+      "🔏": 81430,
+      "賐": 81431,
+      "惎": 81432,
+      "漽": 81433,
+      "齆": 81434,
+      "璅": 81435,
+      "湡": 81436,
+      "褊": 81437,
+      "穡": 81438,
+      "贑": 81439,
+      "彳": 81440,
+      "🚵🏼": 81441,
+      "蜞": 81442,
+      "😋": 81443,
+      "⛹🏿‍♀️": 81444,
+      "̥": 81445,
+      "🤲": 81446,
+      "伻": 81447,
+      "洑": 81448,
+      "牠": 81449,
+      "總": 81450,
+      "🤘🏿": 81451,
+      "厸": 81452,
+      "睏": 81453,
+      "͸": 81454,
+      "奟": 81455,
+      "婥": 81456,
+      "挏": 81457,
+      "挅": 81458,
+      "👩🏼‍❤️‍💋‍👨🏾": 81459,
+      "洺": 81460,
+      "⃾": 81461,
+      "😚": 81462,
+      "鍒": 81463,
+      "␌": 81464,
+      "術": 81465,
+      "価": 81466,
+      "怶": 81467,
+      "藖": 81468,
+      "楬": 81469,
+      "🏨": 81470,
+      "攢": 81471,
+      "Ĭ": 81472,
+      "俕": 81473,
+      "螓": 81474,
+      "浵": 81475,
+      "轕": 81476,
+      "甛": 81477,
+      "晍": 81478,
+      "▿": 81479,
+      "槨": 81480,
+      "膉": 81481,
+      "雺": 81482,
+      "諥": 81483,
+      "恿": 81484,
+      "🇴🇲": 81485,
+      "窠": 81486,
+      "ₔ": 81487,
+      "揢": 81488,
+      "齟": 81489,
+      "篃": 81490,
+      "搋": 81491,
+      "毢": 81492,
+      "謳": 81493,
+      "🥜": 81494,
+      "⃎": 81495,
+      "ʹ": 81496,
+      "誱": 81497,
+      "獇": 81498,
+      "抅": 81499,
+      "拤": 81500,
+      "◂": 81501,
+      "⑅": 81502,
+      "魴": 81503,
+      "筽": 81504,
+      "獍": 81505,
+      "葪": 81506,
+      "👨🏻‍🍼": 81507,
+      "玁": 81508,
+      "😲": 81509,
+      "🎂": 81510,
+      "駤": 81511,
+      "圹": 81512,
+      "謹": 81513,
+      "㊗️": 81514,
+      "棙": 81515,
+      "瓃": 81516,
+      "嚦": 81517,
+      "憊": 81518,
+      "楻": 81519,
+      "覛": 81520,
+      "黱": 81521,
+      "爳": 81522,
+      "🚵🏽": 81523,
+      "螻": 81524,
+      "ŀ": 81525,
+      "⅋": 81526,
+      "帼": 81527,
+      "👳🏿": 81528,
+      "蜶": 81529,
+      "壨": 81530,
+      "鞺": 81531,
+      "廅": 81532,
+      "🙇🏿‍♀️": 81533,
+      "騗": 81534,
+      "妚": 81535,
+      "矁": 81536,
+      "矞": 81537,
+      "鐣": 81538,
+      "鯇": 81539,
+      "℩": 81540,
+      "榀": 81541,
+      "狻": 81542,
+      "蹼": 81543,
+      "霼": 81544,
+      "聈": 81545,
+      "穽": 81546,
+      "擗": 81547,
+      "譏": 81548,
+      "🗞": 81549,
+      "啜": 81550,
+      "紹": 81551,
+      "浨": 81552,
+      "龉": 81553,
+      "螟": 81554,
+      "浻": 81555,
+      "👩🏽‍❤️‍💋‍👨🏻": 81556,
+      "🧑🏻‍🤝‍🧑🏻": 81557,
+      "廨": 81558,
+      "🎉": 81559,
+      "犿": 81560,
+      "緂": 81561,
+      "虮": 81562,
+      "遦": 81563,
+      "費": 81564,
+      "🔌": 81565,
+      "梺": 81566,
+      "伱": 81567,
+      "鞉": 81568,
+      "悅": 81569,
+      "蕪": 81570,
+      "訍": 81571,
+      "髽": 81572,
+      "禵": 81573,
+      "咃": 81574,
+      "ȭ": 81575,
+      "👬🏾": 81576,
+      "噐": 81577,
+      "💂🏾": 81578,
+      "僲": 81579,
+      "輏": 81580,
+      "虖": 81581,
+      "庾": 81582,
+      "瀃": 81583,
+      "鐡": 81584,
+      "龐": 81585,
+      "盪": 81586,
+      "₸": 81587,
+      "愷": 81588,
+      "🧑🏼‍🎓": 81589,
+      "懘": 81590,
+      "鑔": 81591,
+      "岮": 81592,
+      "鐫": 81593,
+      "覟": 81594,
+      "雽": 81595,
+      "齞": 81596,
+      "覔": 81597,
+      "🙆🏾‍♂️": 81598,
+      "嗘": 81599,
+      "俶": 81600,
+      "觴": 81601,
+      "樒": 81602,
+      "騉": 81603,
+      "◁": 81604,
+      "皫": 81605,
+      "閻": 81606,
+      "塤": 81607,
+      "絶": 81608,
+      "夿": 81609,
+      "繫": 81610,
+      "秇": 81611,
+      "覄": 81612,
+      "捨": 81613,
+      "🤦🏿‍♂️": 81614,
+      "驁": 81615,
+      "妘": 81616,
+      "🙆": 81617,
+      "悳": 81618,
+      "酖": 81619,
+      "觐": 81620,
+      "鎣": 81621,
+      "鵎": 81622,
+      "莍": 81623,
+      "甋": 81624,
+      "稂": 81625,
+      "疺": 81626,
+      "鬲": 81627,
+      "🧽": 81628,
+      "碯": 81629,
+      "亼": 81630,
+      "啠": 81631,
+      "👷🏿‍♂️": 81632,
+      "蛗": 81633,
+      "鄉": 81634,
+      "🧑‍🔧": 81635,
+      "愨": 81636,
+      "瀴": 81637,
+      "毜": 81638,
+      "衛": 81639,
+      "👩🏿‍🦯": 81640,
+      "嶁": 81641,
+      "焳": 81642,
+      "斬": 81643,
+      "寈": 81644,
+      "岻": 81645,
+      "垛": 81646,
+      "🇧🇱": 81647,
+      "囅": 81648,
+      "╴": 81649,
+      "妯": 81650,
+      "軷": 81651,
+      "🧏🏾‍♀️": 81652,
+      "🤷🏾‍♂️": 81653,
+      "🏸": 81654,
+      "販": 81655,
+      "崀": 81656,
+      "頋": 81657,
+      "籥": 81658,
+      "硟": 81659,
+      "螮": 81660,
+      "毶": 81661,
+      "龇": 81662,
+      "侶": 81663,
+      "≰": 81664,
+      "塿": 81665,
+      "崄": 81666,
+      "泀": 81667,
+      "掱": 81668,
+      "亪": 81669,
+      "緩": 81670,
+      "🧑🏼‍🔧": 81671,
+      "渂": 81672,
+      "瓑": 81673,
+      "藊": 81674,
+      "⑹": 81675,
+      "傔": 81676,
+      "塄": 81677,
+      "媺": 81678,
+      "蛍": 81679,
+      "阣": 81680,
+      "钬": 81681,
+      "💿": 81682,
+      "達": 81683,
+      "🚴🏾": 81684,
+      "泇": 81685,
+      "誟": 81686,
+      "锃": 81687,
+      "🏃‍♀️": 81688,
+      "紾": 81689,
+      "矷": 81690,
+      "恓": 81691,
+      "🈁": 81692,
+      "鑇": 81693,
+      "🙅🏾‍♂️": 81694,
+      "腃": 81695,
+      "7️⃣": 81696,
+      "🏡": 81697,
+      "亣": 81698,
+      "萗": 81699,
+      "驇": 81700,
+      "🎣": 81701,
+      "鯞": 81702,
+      "哕": 81703,
+      "曞": 81704,
+      "Ϟ": 81705,
+      "🕥": 81706,
+      "慠": 81707,
+      "灷": 81708,
+      "侂": 81709,
+      "🧑🏻‍🎨": 81710,
+      "祴": 81711,
+      "钕": 81712,
+      "‭": 81713,
+      "鸨": 81714,
+      "曀": 81715,
+      "噗": 81716,
+      "🚴🏼": 81717,
+      "忞": 81718,
+      "霫": 81719,
+      "鑩": 81720,
+      "🧜🏽‍♀️": 81721,
+      "Ɓ": 81722,
+      "鐊": 81723,
+      "岡": 81724,
+      "🦡": 81725,
+      "亽": 81726,
+      "蘿": 81727,
+      "堌": 81728,
+      "銟": 81729,
+      "⍚": 81730,
+      "🫱🏽‍🫲🏿": 81731,
+      "礝": 81732,
+      "穎": 81733,
+      "🔹": 81734,
+      "犟": 81735,
+      "阧": 81736,
+      "秨": 81737,
+      "螱": 81738,
+      "譈": 81739,
+      "姀": 81740,
+      "🏌🏿": 81741,
+      "凢": 81742,
+      "楲": 81743,
+      "驒": 81744,
+      "怩": 81745,
+      "迲": 81746,
+      "👨🏾‍🏭": 81747,
+      "唺": 81748,
+      "⑆": 81749,
+      "瑝": 81750,
+      "隂": 81751,
+      "Ｌ": 81752,
+      "逵": 81753,
+      "旾": 81754,
+      "縄": 81755,
+      "罥": 81756,
+      "⌃": 81757,
+      "滽": 81758,
+      "莸": 81759,
+      "⒂": 81760,
+      "濈": 81761,
+      "瑋": 81762,
+      "鳜": 81763,
+      "藰": 81764,
+      "錬": 81765,
+      "獛": 81766,
+      "ℐ": 81767,
+      "ͪ": 81768,
+      "恁": 81769,
+      "鵶": 81770,
+      "澂": 81771,
+      "敫": 81772,
+      "芉": 81773,
+      "焐": 81774,
+      "櫞": 81775,
+      "ℵ": 81776,
+      "鈞": 81777,
+      "噰": 81778,
+      "鬾": 81779,
+      "峊": 81780,
+      "∿": 81781,
+      "瑹": 81782,
+      "鍔": 81783,
+      "🫶🏾": 81784,
+      "⋄": 81785,
+      "勣": 81786,
+      "🤟🏾": 81787,
+      "蠡": 81788,
+      "嘫": 81789,
+      "锫": 81790,
+      "鱖": 81791,
+      "Ƹ": 81792,
+      "邉": 81793,
+      "鵇": 81794,
+      "陞": 81795,
+      "絔": 81796,
+      "圔": 81797,
+      "锸": 81798,
+      "炈": 81799,
+      "烜": 81800,
+      "阃": 81801,
+      "篶": 81802,
+      "🦿": 81803,
+      "閍": 81804,
+      "👶🏽": 81805,
+      "玪": 81806,
+      "墖": 81807,
+      "⏚": 81808,
+      "攼": 81809,
+      "峔": 81810,
+      "棪": 81811,
+      "🩳": 81812,
+      "🧚🏻": 81813,
+      "🧑🏿‍⚖️": 81814,
+      "悇": 81815,
+      "🙆🏿": 81816,
+      "┆": 81817,
+      "蠤": 81818,
+      "蜒": 81819,
+      "烪": 81820,
+      "曇": 81821,
+      "蘳": 81822,
+      "戜": 81823,
+      "躮": 81824,
+      "較": 81825,
+      "宷": 81826,
+      "胩": 81827,
+      "婟": 81828,
+      "👩‍❤️‍👩": 81829,
+      "婏": 81830,
+      "🤽🏽": 81831,
+      "歛": 81832,
+      "烞": 81833,
+      "禒": 81834,
+      "蟊": 81835,
+      "楶": 81836,
+      "👨🏼‍❤️‍👨🏿": 81837,
+      "黕": 81838,
+      "哸": 81839,
+      "嚲": 81840,
+      "椶": 81841,
+      "🤽‍♀️": 81842,
+      "℀": 81843,
+      "暝": 81844,
+      "塱": 81845,
+      "卶": 81846,
+      "💒": 81847,
+      "橑": 81848,
+      "虃": 81849,
+      "皬": 81850,
+      "棅": 81851,
+      "褀": 81852,
+      "瓔": 81853,
+      "⚐": 81854,
+      "偒": 81855,
+      "🇰🇭": 81856,
+      "🙆🏻‍♀️": 81857,
+      "蒯": 81858,
+      "妫": 81859,
+      "偲": 81860,
+      "🙎🏼‍♂️": 81861,
+      "螿": 81862,
+      "襴": 81863,
+      "🧔🏽‍♀️": 81864,
+      "鏒": 81865,
+      "簫": 81866,
+      "🧝🏼‍♂️": 81867,
+      "樆": 81868,
+      "薵": 81869,
+      "騰": 81870,
+      "怇": 81871,
+      "闖": 81872,
+      "拊": 81873,
+      "靺": 81874,
+      "凲": 81875,
+      "羀": 81876,
+      "😹": 81877,
+      "膲": 81878,
+      "侻": 81879,
+      "浫": 81880,
+      "Ͱ": 81881,
+      "櫧": 81882,
+      "靲": 81883,
+      "🤹🏿": 81884,
+      "斔": 81885,
+      "茛": 81886,
+      "⛟": 81887,
+      "羾": 81888,
+      "诎": 81889,
+      "鑁": 81890,
+      "塠": 81891,
+      "🤗": 81892,
+      "钚": 81893,
+      "埌": 81894,
+      "嬟": 81895,
+      "皰": 81896,
+      "炿": 81897,
+      "👩🏻‍🌾": 81898,
+      "竪": 81899,
+      "袿": 81900,
+      "胻": 81901,
+      "嬗": 81902,
+      "舊": 81903,
+      "ǃ": 81904,
+      "🏄🏻‍♀️": 81905,
+      "帋": 81906,
+      "Ｆ": 81907,
+      "侔": 81908,
+      "踼": 81909,
+      "娎": 81910,
+      "馵": 81911,
+      "憮": 81912,
+      "鯡": 81913,
+      "儹": 81914,
+      "悮": 81915,
+      "腆": 81916,
+      "桻": 81917,
+      "悥": 81918,
+      "儥": 81919,
+      "💪": 81920,
+      "🇺🇬": 81921,
+      "⬛️": 81922,
+      "溇": 81923,
+      "莋": 81924,
+      "嘦": 81925,
+      "翵": 81926,
+      "黟": 81927,
+      "🎺": 81928,
+      "諧": 81929,
+      "怋": 81930,
+      "隃": 81931,
+      "鹱": 81932,
+      "♆": 81933,
+      "鉬": 81934,
+      "醓": 81935,
+      "詾": 81936,
+      "藾": 81937,
+      "🧍🏾": 81938,
+      "嚗": 81939,
+      "奊": 81940,
+      "邶": 81941,
+      "🧖🏼": 81942,
+      "桒": 81943,
+      "檥": 81944,
+      "嚓": 81945,
+      "氘": 81946,
+      "虍": 81947,
+      "悙": 81948,
+      "抷": 81949,
+      "🙎🏻‍♂️": 81950,
+      "⒦": 81951,
+      "陏": 81952,
+      "輇": 81953,
+      "髒": 81954,
+      "🇶🇦": 81955,
+      "谸": 81956,
+      "茻": 81957,
+      "攫": 81958,
+      "鏓": 81959,
+      "蔯": 81960,
+      "▧": 81961,
+      "霑": 81962,
+      "⅍": 81963,
+      "崈": 81964,
+      "諨": 81965,
+      "👵🏾": 81966,
+      "嗈": 81967,
+      "廆": 81968,
+      "褼": 81969,
+      "庩": 81970,
+      "嘳": 81971,
+      "笾": 81972,
+      "霃": 81973,
+      "鼬": 81974,
+      "幋": 81975,
+      "渏": 81976,
+      "狺": 81977,
+      "侌": 81978,
+      "漤": 81979,
+      "ű": 81980,
+      "🪮": 81981,
+      "崦": 81982,
+      "尥": 81983,
+      "Ⓔ": 81984,
+      "檮": 81985,
+      "礮": 81986,
+      "傡": 81987,
+      "梞": 81988,
+      "曗": 81989,
+      "釈": 81990,
+      "巂": 81991,
+      "🛖": 81992,
+      "恉": 81993,
+      "豼": 81994,
+      "廽": 81995,
+      "辢": 81996,
+      "诳": 81997,
+      "虺": 81998,
+      "蝐": 81999,
+      "飪": 82000,
+      "迯": 82001,
+      "僀": 82002,
+      "鉱": 82003,
+      "鬒": 82004,
+      "戕": 82005,
+      "Ϋ": 82006,
+      "話": 82007,
+      "蔈": 82008,
+      "掸": 82009,
+      "偟": 82010,
+      "騭": 82011,
+      "🇦🇱": 82012,
+      "鍤": 82013,
+      "₁": 82014,
+      "傟": 82015,
+      "潌": 82016,
+      "🥷🏾": 82017,
+      "👩‍👩‍👦‍👦": 82018,
+      "🏊🏿‍♂️": 82019,
+      "燡": 82020,
+      "艢": 82021,
+      "髢": 82022,
+      "湒": 82023,
+      "穓": 82024,
+      "蘌": 82025,
+      "謡": 82026,
+      "🤜🏽": 82027,
+      "脪": 82028,
+      "橪": 82029,
+      "漝": 82030,
+      "搉": 82031,
+      "鏮": 82032,
+      "🫰🏾": 82033,
+      "睨": 82034,
+      "蟈": 82035,
+      "😱": 82036,
+      "穅": 82037,
+      "稧": 82038,
+      "🏄‍♂️": 82039,
+      "肄": 82040,
+      "護": 82041,
+      "罟": 82042,
+      "罭": 82043,
+      "樹": 82044,
+      "👨🏾‍🎨": 82045,
+      "👩🏽‍⚕️": 82046,
+      "₥": 82047,
+      "Ｐ": 82048,
+      "赆": 82049,
+      "祢": 82050,
+      "🧘🏾‍♀️": 82051,
+      "📨": 82052,
+      "裻": 82053,
+      "擧": 82054,
+      "ˍ": 82055,
+      "骙": 82056,
+      "🤾🏾‍♂️": 82057,
+      "釷": 82058,
+      "蘗": 82059,
+      "蜴": 82060,
+      "🥁": 82061,
+      "坣": 82062,
+      "崠": 82063,
+      "氋": 82064,
+      "⏝": 82065,
+      "驃": 82066,
+      "涋": 82067,
+      "̫": 82068,
+      "樗": 82069,
+      "敎": 82070,
+      "濠": 82071,
+      "饃": 82072,
+      "𓀛": 82073,
+      "🧯": 82074,
+      "鐯": 82075,
+      "憛": 82076,
+      "嫽": 82077,
+      "🦨": 82078,
+      "╖": 82079,
+      "͂": 82080,
+      "饏": 82081,
+      "◪": 82082,
+      "穩": 82083,
+      "ⅺ": 82084,
+      "檾": 82085,
+      "👵🏻": 82086,
+      "跐": 82087,
+      "龆": 82088,
+      "👦": 82089,
+      "鋊": 82090,
+      "⒘": 82091,
+      "蛻": 82092,
+      "裖": 82093,
+      "麌": 82094,
+      "莝": 82095,
+      "洂": 82096,
+      "鵰": 82097,
+      "蹺": 82098,
+      "輻": 82099,
+      "蚑": 82100,
+      "燜": 82101,
+      "徟": 82102,
+      "枹": 82103,
+      "喿": 82104,
+      "👇🏻": 82105,
+      "髉": 82106,
+      "纔": 82107,
+      "囑": 82108,
+      "睭": 82109,
+      "飋": 82110,
+      "👨🏻‍🎤": 82111,
+      "鋤": 82112,
+      "鈹": 82113,
+      "柫": 82114,
+      "桋": 82115,
+      "濢": 82116,
+      "歚": 82117,
+      "↡": 82118,
+      "箖": 82119,
+      "涰": 82120,
+      "≌": 82121,
+      "👩‍✈️": 82122,
+      "蔣": 82123,
+      "怬": 82124,
+      "🖖🏾": 82125,
+      "貮": 82126,
+      "鉕": 82127,
+      "凐": 82128,
+      "騦": 82129,
+      "釀": 82130,
+      "庨": 82131,
+      "慬": 82132,
+      "裉": 82133,
+      "秖": 82134,
+      "鵧": 82135,
+      "⛹🏻": 82136,
+      "ϋ": 82137,
+      "垑": 82138,
+      "ɮ": 82139,
+      "梟": 82140,
+      "鯉": 82141,
+      "霘": 82142,
+      "👩🏿‍❤️‍💋‍👩🏼": 82143,
+      "🏋️‍♂️": 82144,
+      "炚": 82145,
+      "👨🏾‍🦼": 82146,
+      "釴": 82147,
+      "竈": 82148,
+      "🧶": 82149,
+      "狫": 82150,
+      "蛽": 82151,
+      "🫰🏼": 82152,
+      "鞖": 82153,
+      "👃🏼": 82154,
+      "👩🏼‍❤️‍💋‍👩🏻": 82155,
+      "蟚": 82156,
+      "⛹🏽‍♂️": 82157,
+      "賑": 82158,
+      "🧘‍♀️": 82159,
+      "🤵‍♂️": 82160,
+      "碅": 82161,
+      "̽": 82162,
+      "鲯": 82163,
+      "頑": 82164,
+      "窛": 82165,
+      "聞": 82166,
+      "↧": 82167,
+      "忤": 82168,
+      "哚": 82169,
+      "簼": 82170,
+      "👨‍🔬": 82171,
+      "揰": 82172,
+      "瞕": 82173,
+      "廔": 82174,
+      "🎞": 82175,
+      "轅": 82176,
+      "楺": 82177,
+      "媄": 82178,
+      "🙅🏽‍♂️": 82179,
+      "▀": 82180,
+      "🧎🏻‍♀️": 82181,
+      "鼗": 82182,
+      "旉": 82183,
+      "Ż": 82184,
+      "鶨": 82185,
+      "玜": 82186,
+      "翳": 82187,
+      "鍼": 82188,
+      "⊼": 82189,
+      "鯘": 82190,
+      "徚": 82191,
+      "傶": 82192,
+      "睬": 82193,
+      "餮": 82194,
+      "漑": 82195,
+      "釆": 82196,
+      "♲": 82197,
+      "鮴": 82198,
+      "襢": 82199,
+      "鎯": 82200,
+      "⅄": 82201,
+      "灳": 82202,
+      "🦒": 82203,
+      "⒨": 82204,
+      "亻": 82205,
+      "稕": 82206,
+      "汦": 82207,
+      "膌": 82208,
+      "🧑🏽‍🎨": 82209,
+      "嵽": 82210,
+      "煗": 82211,
+      "韞": 82212,
+      "拝": 82213,
+      "誫": 82214,
+      "頳": 82215,
+      "掃": 82216,
+      "👃🏻": 82217,
+      "癁": 82218,
+      "閯": 82219,
+      "逥": 82220,
+      "⚻": 82221,
+      "Ĥ": 82222,
+      "Ⓕ": 82223,
+      "棝": 82224,
+      "雴": 82225,
+      "迾": 82226,
+      "⌲": 82227,
+      "穻": 82228,
+      "👩🏿‍🔧": 82229,
+      "鈿": 82230,
+      "皐": 82231,
+      "淗": 82232,
+      "鰙": 82233,
+      "🚦": 82234,
+      "⓻": 82235,
+      "廑": 82236,
+      "爅": 82237,
+      "ˠ": 82238,
+      "飖": 82239,
+      "蔠": 82240,
+      "🍃": 82241,
+      "沏": 82242,
+      "♰": 82243,
+      "胯": 82244,
+      "晬": 82245,
+      "🚶‍♂️": 82246,
+      "愩": 82247,
+      "🤘🏻": 82248,
+      "⍎": 82249,
+      "": 82250,
+      "畮": 82251,
+      "菳": 82252,
+      "阒": 82253,
+      "蹟": 82254,
+      "禷": 82255,
+      "呿": 82256,
+      "瑁": 82257,
+      "瀷": 82258,
+      "厫": 82259,
+      "檦": 82260,
+      "🧑🏿‍🦼": 82261,
+      "痾": 82262,
+      "✌🏾": 82263,
+      "衖": 82264,
+      "赑": 82265,
+      "👩🏽‍🏭": 82266,
+      "诼": 82267,
+      "籙": 82268,
+      "埱": 82269,
+      "冔": 82270,
+      "铘": 82271,
+      "蜼": 82272,
+      "懹": 82273,
+      "鷚": 82274,
+      "緻": 82275,
+      "疓": 82276,
+      "🫳🏻": 82277,
+      "🏄🏼": 82278,
+      "⋟": 82279,
+      "趥": 82280,
+      "鑅": 82281,
+      "珋": 82282,
+      "茝": 82283,
+      "巒": 82284,
+      "皻": 82285,
+      "⎅": 82286,
+      "⃗": 82287,
+      "🐗": 82288,
+      "弢": 82289,
+      "╁": 82290,
+      "飴": 82291,
+      "⍛": 82292,
+      "驔": 82293,
+      "🧔🏻‍♀️": 82294,
+      "凫": 82295,
+      "閃": 82296,
+      "罖": 82297,
+      "铔": 82298,
+      "饛": 82299,
+      "橻": 82300,
+      "癆": 82301,
+      "楮": 82302,
+      "👩🏽‍⚖️": 82303,
+      "濘": 82304,
+      "檉": 82305,
+      "絎": 82306,
+      "霰": 82307,
+      "強": 82308,
+      "桠": 82309,
+      "⊿": 82310,
+      "鴺": 82311,
+      "怰": 82312,
+      "閿": 82313,
+      "仯": 82314,
+      "镴": 82315,
+      "🙆🏾": 82316,
+      "苮": 82317,
+      "蕟": 82318,
+      "鋬": 82319,
+      "翍": 82320,
+      "阤": 82321,
+      "🚥": 82322,
+      "劥": 82323,
+      "聵": 82324,
+      "紵": 82325,
+      "👨🏻‍🎨": 82326,
+      "💣": 82327,
+      "螣": 82328,
+      "苀": 82329,
+      "偩": 82330,
+      "🤦🏻‍♀️": 82331,
+      "🦀": 82332,
+      "斆": 82333,
+      "帥": 82334,
+      "眱": 82335,
+      "韓": 82336,
+      "迡": 82337,
+      "鐕": 82338,
+      "罝": 82339,
+      "ͼ": 82340,
+      "攗": 82341,
+      "伝": 82342,
+      "诨": 82343,
+      "垌": 82344,
+      "☍": 82345,
+      "晎": 82346,
+      "鄐": 82347,
+      "鱌": 82348,
+      "綮": 82349,
+      "媑": 82350,
+      "崉": 82351,
+      "✅": 82352,
+      "💁🏼‍♀️": 82353,
+      "⁈": 82354,
+      "ĵ": 82355,
+      "購": 82356,
+      "慯": 82357,
+      "鵲": 82358,
+      "☮": 82359,
+      "闀": 82360,
+      "蹃": 82361,
+      "⎸": 82362,
+      "襇": 82363,
+      "į": 82364,
+      "該": 82365,
+      "旡": 82366,
+      "啇": 82367,
+      "戅": 82368,
+      "輮": 82369,
+      "⓫": 82370,
+      "伮": 82371,
+      "皷": 82372,
+      "豰": 82373,
+      "窪": 82374,
+      "怌": 82375,
+      "褯": 82376,
+      "愾": 82377,
+      "🖐🏾": 82378,
+      "錦": 82379,
+      "瑼": 82380,
+      "鼪": 82381,
+      "興": 82382,
+      "娀": 82383,
+      "袆": 82384,
+      "瞞": 82385,
+      "餥": 82386,
+      "鲽": 82387,
+      "甪": 82388,
+      "❣️": 82389,
+      "帴": 82390,
+      "🤽🏽‍♀️": 82391,
+      "⒐": 82392,
+      "醌": 82393,
+      "⍳": 82394,
+      "嫐": 82395,
+      "👩🏼‍🦳": 82396,
+      "眥": 82397,
+      "蟳": 82398,
+      "鐝": 82399,
+      "🖨": 82400,
+      "漟": 82401,
+      "殝": 82402,
+      "閠": 82403,
+      "℡": 82404,
+      "怚": 82405,
+      "濬": 82406,
+      "悀": 82407,
+      "豍": 82408,
+      "砦": 82409,
+      "凚": 82410,
+      "霶": 82411,
+      "阺": 82412,
+      "孇": 82413,
+      "⃴": 82414,
+      "騪": 82415,
+      "宀": 82416,
+      "葼": 82417,
+      "谥": 82418,
+      "趵": 82419,
+      "🚶🏾": 82420,
+      "啍": 82421,
+      "愼": 82422,
+      "迧": 82423,
+      "⏟": 82424,
+      "揇": 82425,
+      "🧙🏽": 82426,
+      "🛫": 82427,
+      "厀": 82428,
+      "軀": 82429,
+      "聜": 82430,
+      "撻": 82431,
+      "淎": 82432,
+      "垠": 82433,
+      "Ⓗ": 82434,
+      "⚛": 82435,
+      "∓": 82436,
+      "ƈ": 82437,
+      "耛": 82438,
+      "擨": 82439,
+      "鍍": 82440,
+      "砲": 82441,
+      "ʱ": 82442,
+      "蝤": 82443,
+      "璷": 82444,
+      "鵈": 82445,
+      "袉": 82446,
+      "⇴": 82447,
+      "輫": 82448,
+      "駓": 82449,
+      "🤔": 82450,
+      "潦": 82451,
+      "👨🏻‍🌾": 82452,
+      "🥱": 82453,
+      "戋": 82454,
+      "👰🏾": 82455,
+      "腣": 82456,
+      "崙": 82457,
+      "僸": 82458,
+      "豦": 82459,
+      "箸": 82460,
+      "铡": 82461,
+      "棼": 82462,
+      "笤": 82463,
+      "↬": 82464,
+      "甧": 82465,
+      "讓": 82466,
+      "扡": 82467,
+      "逡": 82468,
+      "蹙": 82469,
+      "💌": 82470,
+      "錷": 82471,
+      "蟔": 82472,
+      "̀": 82473,
+      "🚵🏾": 82474,
+      "麅": 82475,
+      "💇🏾‍♂️": 82476,
+      "歝": 82477,
+      "饝": 82478,
+      "轚": 82479,
+      "👳🏾‍♀️": 82480,
+      "ʶ": 82481,
+      "鮏": 82482,
+      "✡️": 82483,
+      "泦": 82484,
+      "縡": 82485,
+      "👨🏼‍🦼": 82486,
+      "摁": 82487,
+      "🏃🏿‍♀️": 82488,
+      "摋": 82489,
+      "🧚🏾‍♂️": 82490,
+      "訚": 82491,
+      "磆": 82492,
+      "铢": 82493,
+      "葜": 82494,
+      "篣": 82495,
+      "掽": 82496,
+      "辫": 82497,
+      "💅": 82498,
+      "糃": 82499,
+      "薤": 82500,
+      "跿": 82501,
+      "🇯🇵": 82502,
+      "慀": 82503,
+      "摅": 82504,
+      "晘": 82505,
+      "溿": 82506,
+      "義": 82507,
+      "👨🏿‍🏭": 82508,
+      "莇": 82509,
+      "🇲🇻": 82510,
+      "韃": 82511,
+      "ǵ": 82512,
+      "踔": 82513,
+      "浿": 82514,
+      "貔": 82515,
+      "嶵": 82516,
+      "莨": 82517,
+      "裫": 82518,
+      "≧": 82519,
+      "✸": 82520,
+      "聑": 82521,
+      "峺": 82522,
+      "劌": 82523,
+      "谌": 82524,
+      "鰽": 82525,
+      "😿": 82526,
+      "ϊ": 82527,
+      "啖": 82528,
+      "漞": 82529,
+      "趆": 82530,
+      "紤": 82531,
+      "🏋🏿‍♀️": 82532,
+      "ʁ": 82533,
+      "葅": 82534,
+      "懴": 82535,
+      "稑": 82536,
+      "旴": 82537,
+      "≾": 82538,
+      "幣": 82539,
+      "餟": 82540,
+      "⃫": 82541,
+      "🧾": 82542,
+      "狚": 82543,
+      "嶗": 82544,
+      "🐨": 82545,
+      "钆": 82546,
+      "崓": 82547,
+      "🥾": 82548,
+      "鳪": 82549,
+      "謫": 82550,
+      "胑": 82551,
+      "搕": 82552,
+      "啌": 82553,
+      "耹": 82554,
+      "笉": 82555,
+      "惗": 82556,
+      "嶠": 82557,
+      "蔸": 82558,
+      "櫣": 82559,
+      "鍎": 82560,
+      "滃": 82561,
+      "籫": 82562,
+      "聛": 82563,
+      "懡": 82564,
+      "褽": 82565,
+      "敺": 82566,
+      "虿": 82567,
+      "🧛🏿‍♀️": 82568,
+      "🏋️": 82569,
+      "桱": 82570,
+      "彅": 82571,
+      "垉": 82572,
+      "跎": 82573,
+      "贀": 82574,
+      "篋": 82575,
+      "攠": 82576,
+      "Ⓒ": 82577,
+      "擾": 82578,
+      "啻": 82579,
+      "楜": 82580,
+      "鱈": 82581,
+      "圀": 82582,
+      "橃": 82583,
+      "鲌": 82584,
+      "髕": 82585,
+      "璄": 82586,
+      "扆": 82587,
+      "楩": 82588,
+      "飽": 82589,
+      "蘭": 82590,
+      "嫪": 82591,
+      "鑴": 82592,
+      "👨🏽‍❤️‍💋‍👨🏾": 82593,
+      "稲": 82594,
+      "橲": 82595,
+      "鬨": 82596,
+      "聨": 82597,
+      "汳": 82598,
+      "鎟": 82599,
+      "煒": 82600,
+      "🏢": 82601,
+      "勂": 82602,
+      "苎": 82603,
+      "凒": 82604,
+      "螢": 82605,
+      "麂": 82606,
+      "鳫": 82607,
+      "庫": 82608,
+      "韥": 82609,
+      "緇": 82610,
+      "✒️": 82611,
+      "葮": 82612,
+      "🇹🇹": 82613,
+      "👮🏽‍♀️": 82614,
+      "趍": 82615,
+      "┮": 82616,
+      "譧": 82617,
+      "錍": 82618,
+      "瀹": 82619,
+      "蝳": 82620,
+      "橣": 82621,
+      "👩🏾‍✈️": 82622,
+      "粕": 82623,
+      "鴁": 82624,
+      "燎": 82625,
+      "掹": 82626,
+      "墱": 82627,
+      "緃": 82628,
+      "搸": 82629,
+      "玦": 82630,
+      "🙋🏾": 82631,
+      "僷": 82632,
+      "🫸": 82633,
+      "驤": 82634,
+      "僉": 82635,
+      "頜": 82636,
+      "摢": 82637,
+      "👩🏽‍🍳": 82638,
+      "🦻🏼": 82639,
+      "邺": 82640,
+      "Ț": 82641,
+      "🤰🏻": 82642,
+      "櫸": 82643,
+      "缦": 82644,
+      "邊": 82645,
+      "🌡": 82646,
+      "盙": 82647,
+      "錮": 82648,
+      "⌒": 82649,
+      "嬅": 82650,
+      "萀": 82651,
+      "␀": 82652,
+      "🤾🏼‍♀️": 82653,
+      "⍑": 82654,
+      "蝭": 82655,
+      "豊": 82656,
+      "迶": 82657,
+      "桟": 82658,
+      "禈": 82659,
+      "湊": 82660,
+      "瑇": 82661,
+      "⚯": 82662,
+      "堛": 82663,
+      "筧": 82664,
+      "🧗🏾": 82665,
+      "釂": 82666,
+      "⒑": 82667,
+      "竢": 82668,
+      "渃": 82669,
+      "赨": 82670,
+      "悂": 82671,
+      "涀": 82672,
+      "⃌": 82673,
+      "讁": 82674,
+      "⛝": 82675,
+      "⋫": 82676,
+      "楇": 82677,
+      "秮": 82678,
+      "蒳": 82679,
+      "栻": 82680,
+      "🙇🏽‍♀️": 82681,
+      "轣": 82682,
+      "🦸‍♀️": 82683,
+      "聱": 82684,
+      "鲀": 82685,
+      "纩": 82686,
+      "飇": 82687,
+      "┤": 82688,
+      "禙": 82689,
+      "🙅🏾‍♀️": 82690,
+      "🎅🏾": 82691,
+      "襦": 82692,
+      "漸": 82693,
+      "鈬": 82694,
+      "呮": 82695,
+      "🧑🏾‍❤️‍🧑🏼": 82696,
+      "ⓗ": 82697,
+      "嚁": 82698,
+      "♷": 82699,
+      "缿": 82700,
+      "膻": 82701,
+      "袵": 82702,
+      "🧏🏻‍♀️": 82703,
+      "晀": 82704,
+      "裾": 82705,
+      "恎": 82706,
+      "燁": 82707,
+      "巪": 82708,
+      "蚓": 82709,
+      "🌟": 82710,
+      "鞥": 82711,
+      "秼": 82712,
+      "蒉": 82713,
+      "兡": 82714,
+      "袶": 82715,
+      "摀": 82716,
+      "👩🏼‍🏭": 82717,
+      "🏄🏿": 82718,
+      "溽": 82719,
+      "擏": 82720,
+      "駃": 82721,
+      "趨": 82722,
+      "柞": 82723,
+      "遛": 82724,
+      "垬": 82725,
+      "僘": 82726,
+      "昗": 82727,
+      "₧": 82728,
+      "贳": 82729,
+      "鶝": 82730,
+      "磰": 82731,
+      "廗": 82732,
+      "螭": 82733,
+      "蕀": 82734,
+      "坥": 82735,
+      "濩": 82736,
+      "沲": 82737,
+      "😙": 82738,
+      "閸": 82739,
+      "捹": 82740,
+      "凗": 82741,
+      "贄": 82742,
+      "👩🏼‍🎨": 82743,
+      "亴": 82744,
+      "Θ": 82745,
+      "爕": 82746,
+      "炲": 82747,
+      "掳": 82748,
+      "👩🏿‍❤️‍👨🏾": 82749,
+      "淈": 82750,
+      "剟": 82751,
+      "烣": 82752,
+      "┢": 82753,
+      "🧗🏻": 82754,
+      "毨": 82755,
+      "涒": 82756,
+      "🇰🇾": 82757,
+      "棳": 82758,
+      "眚": 82759,
+      "🦆": 82760,
+      "佧": 82761,
+      "洹": 82762,
+      "緌": 82763,
+      "◖": 82764,
+      "鴆": 82765,
+      "鈅": 82766,
+      "汏": 82767,
+      "體": 82768,
+      "👩🏽‍❤️‍👨🏿": 82769,
+      "徔": 82770,
+      "👨🏻‍🔧": 82771,
+      "脟": 82772,
+      "鄕": 82773,
+      "鑏": 82774,
+      "🪷": 82775,
+      "🧑🏽‍❤️‍💋‍🧑🏾": 82776,
+      "🫳🏽": 82777,
+      "🧛🏿‍♂️": 82778,
+      "琾": 82779,
+      "懇": 82780,
+      "偽": 82781,
+      "喏": 82782,
+      "堜": 82783,
+      "饕": 82784,
+      "ɝ": 82785,
+      "ȁ": 82786,
+      "墼": 82787,
+      "擌": 82788,
+      "魟": 82789,
+      "👰‍♀️": 82790,
+      "◚": 82791,
+      "奰": 82792,
+      "🙋🏼": 82793,
+      "轷": 82794,
+      "吜": 82795,
+      "鲉": 82796,
+      "🙏🏻": 82797,
+      "◥": 82798,
+      "⎚": 82799,
+      "蘂": 82800,
+      "🙉": 82801,
+      "⋩": 82802,
+      "祰": 82803,
+      "碡": 82804,
+      "◻": 82805,
+      "饍": 82806,
+      "␡": 82807,
+      "籧": 82808,
+      "挬": 82809,
+      "🤦🏾‍♂️": 82810,
+      "💰": 82811,
+      "敗": 82812,
+      "玡": 82813,
+      "敶": 82814,
+      "懼": 82815,
+      "勖": 82816,
+      "⏏": 82817,
+      "塬": 82818,
+      "廴": 82819,
+      "💆🏼": 82820,
+      "囆": 82821,
+      "埇": 82822,
+      "犱": 82823,
+      "渁": 82824,
+      "譪": 82825,
+      "芻": 82826,
+      "⒜": 82827,
+      "🛥": 82828,
+      "陎": 82829,
+      "禼": 82830,
+      "墌": 82831,
+      "🚣🏼‍♀️": 82832,
+      "枒": 82833,
+      "譖": 82834,
+      "玤": 82835,
+      "鈄": 82836,
+      "🇲🇿": 82837,
+      "詊": 82838,
+      "圕": 82839,
+      "澁": 82840,
+      "愥": 82841,
+      "胨": 82842,
+      "丯": 82843,
+      "宆": 82844,
+      "疋": 82845,
+      "郢": 82846,
+      "絀": 82847,
+      "痈": 82848,
+      "訞": 82849,
+      "斝": 82850,
+      "莚": 82851,
+      "鱅": 82852,
+      "嵮": 82853,
+      "湐": 82854,
+      "襻": 82855,
+      "虀": 82856,
+      "篒": 82857,
+      "嘁": 82858,
+      "💶": 82859,
+      "烝": 82860,
+      "卺": 82861,
+      "覒": 82862,
+      "夦": 82863,
+      "⇸": 82864,
+      "↍": 82865,
+      "闘": 82866,
+      "惣": 82867,
+      "Ē": 82868,
+      "ͩ": 82869,
+      "╹": 82870,
+      "⎉": 82871,
+      "🤶🏻": 82872,
+      "驪": 82873,
+      "鸵": 82874,
+      "腗": 82875,
+      "孲": 82876,
+      "愳": 82877,
+      "😁": 82878,
+      "鸎": 82879,
+      "韂": 82880,
+      "👨🏽‍❤️‍💋‍👨🏼": 82881,
+      "囈": 82882,
+      "痷": 82883,
+      "嶒": 82884,
+      "骺": 82885,
+      "乩": 82886,
+      "⁷": 82887,
+      "疧": 82888,
+      "鄹": 82889,
+      "隲": 82890,
+      "梊": 82891,
+      "⇰": 82892,
+      "😐": 82893,
+      "🧹": 82894,
+      "囩": 82895,
+      "垏": 82896,
+      "👨🏻‍🏭": 82897,
+      "飐": 82898,
+      "👩🏽‍🎨": 82899,
+      "鷣": 82900,
+      "瞱": 82901,
+      "⑷": 82902,
+      "氉": 82903,
+      "枮": 82904,
+      "槦": 82905,
+      "嬙": 82906,
+      "翨": 82907,
+      "蠮": 82908,
+      "垟": 82909,
+      "鬤": 82910,
+      "疬": 82911,
+      "💏🏼": 82912,
+      "🤒": 82913,
+      "燓": 82914,
+      "⅓": 82915,
+      "枓": 82916,
+      "圓": 82917,
+      "挰": 82918,
+      "檂": 82919,
+      "🦝": 82920,
+      "斂": 82921,
+      "蠙": 82922,
+      "盿": 82923,
+      "攑": 82924,
+      "鞗": 82925,
+      "蠞": 82926,
+      "賫": 82927,
+      "玣": 82928,
+      "霒": 82929,
+      "靰": 82930,
+      "🧝🏾‍♀️": 82931,
+      "腘": 82932,
+      "🙍🏽": 82933,
+      "跓": 82934,
+      "騚": 82935,
+      "甈": 82936,
+      "襥": 82937,
+      "葾": 82938,
+      "蘅": 82939,
+      "濊": 82940,
+      "搫": 82941,
+      "Υ": 82942,
+      "ℍ": 82943,
+      "矝": 82944,
+      "嬚": 82945,
+      "🧑🏻‍🚒": 82946,
+      "脧": 82947,
+      "鉅": 82948,
+      "絞": 82949,
+      "⏫": 82950,
+      "謧": 82951,
+      "摒": 82952,
+      "傽": 82953,
+      "🦹‍♂️": 82954,
+      "◺": 82955,
+      "驷": 82956,
+      "⛵": 82957,
+      "⃍": 82958,
+      "儫": 82959,
+      "蘈": 82960,
+      "Ě": 82961,
+      "屖": 82962,
+      "蝡": 82963,
+      "湓": 82964,
+      "釞": 82965,
+      "辮": 82966,
+      "瓀": 82967,
+      "⊞": 82968,
+      "趒": 82969,
+      "濇": 82970,
+      "郃": 82971,
+      "撠": 82972,
+      "儼": 82973,
+      "ǻ": 82974,
+      "潓": 82975,
+      "⍈": 82976,
+      "⍖": 82977,
+      "讇": 82978,
+      "繻": 82979,
+      "墒": 82980,
+      "灃": 82981,
+      "ϗ": 82982,
+      "雲": 82983,
+      "🤙🏻": 82984,
+      "搩": 82985,
+      "鰸": 82986,
+      "恹": 82987,
+      "瑩": 82988,
+      "埆": 82989,
+      "嚶": 82990,
+      "餠": 82991,
+      "醖": 82992,
+      "晲": 82993,
+      "DZ": 82994,
+      "稶": 82995,
+      "皒": 82996,
+      "🤢": 82997,
+      "旑": 82998,
+      "漍": 82999,
+      "嗒": 83000,
+      "⁇": 83001,
+      "ↇ": 83002,
+      "顳": 83003,
+      "鉇": 83004,
+      "犧": 83005,
+      "🐉": 83006,
+      "👲🏾": 83007,
+      "̍": 83008,
+      "🗑": 83009,
+      "墯": 83010,
+      "歟": 83011,
+      "觭": 83012,
+      "🥷🏿": 83013,
+      "攣": 83014,
+      "痐": 83015,
+      "厙": 83016,
+      "跲": 83017,
+      "♺": 83018,
+      "飫": 83019,
+      "🏴󠁧󠁢󠁷󠁬󠁳󠁿": 83020,
+      "珷": 83021,
+      "搌": 83022,
+      "⊺": 83023,
+      "⃯": 83024,
+      "隚": 83025,
+      "橞": 83026,
+      "蹠": 83027,
+      "殽": 83028,
+      "蜄": 83029,
+      "┯": 83030,
+      "緺": 83031,
+      "罫": 83032,
+      "鏷": 83033,
+      "埒": 83034,
+      "歕": 83035,
+      "錣": 83036,
+      "👩🏼‍❤️‍💋‍👨🏻": 83037,
+      "🦸🏻": 83038,
+      "♯": 83039,
+      "昤": 83040,
+      "⌑": 83041,
+      "🙏🏿": 83042,
+      "嬜": 83043,
+      "🎊": 83044,
+      "踐": 83045,
+      "攉": 83046,
+      "鎰": 83047,
+      "欆": 83048,
+      "惽": 83049,
+      "粂": 83050,
+      "滊": 83051,
+      "睮": 83052,
+      "灈": 83053,
+      "緳": 83054,
+      "珬": 83055,
+      "殲": 83056,
+      "🚛": 83057,
+      "🏟": 83058,
+      "璢": 83059,
+      "晭": 83060,
+      "砏": 83061,
+      "逩": 83062,
+      "锟": 83063,
+      "値": 83064,
+      "罛": 83065,
+      "❤️‍🩹": 83066,
+      "🤹🏼": 83067,
+      "鹺": 83068,
+      "垭": 83069,
+      "🕵🏿‍♀️": 83070,
+      "麨": 83071,
+      "鏂": 83072,
+      "瘺": 83073,
+      "嚞": 83074,
+      "鑪": 83075,
+      "拏": 83076,
+      "灝": 83077,
+      "畭": 83078,
+      "駢": 83079,
+      "辷": 83080,
+      "🕦": 83081,
+      "滗": 83082,
+      "診": 83083,
+      "妋": 83084,
+      "℻": 83085,
+      "祌": 83086,
+      "帤": 83087,
+      "烢": 83088,
+      "旎": 83089,
+      "穌": 83090,
+      "蜤": 83091,
+      "嗶": 83092,
+      "🎗": 83093,
+      "糉": 83094,
+      "魷": 83095,
+      "鬩": 83096,
+      "蘕": 83097,
+      "💎": 83098,
+      "茢": 83099,
+      "頧": 83100,
+      "阸": 83101,
+      "楿": 83102,
+      "腯": 83103,
+      "镃": 83104,
+      "颮": 83105,
+      "⛖": 83106,
+      "🪘": 83107,
+      "妬": 83108,
+      "螲": 83109,
+      "秎": 83110,
+      "刦": 83111,
+      "皈": 83112,
+      "酳": 83113,
+      "🚣🏽": 83114,
+      "🥐": 83115,
+      "呬": 83116,
+      "澷": 83117,
+      "鬐": 83118,
+      "匶": 83119,
+      "蔏": 83120,
+      "餸": 83121,
+      "👩‍👩‍👧‍👧": 83122,
+      "👨🏽‍⚕️": 83123,
+      "👲": 83124,
+      "鄨": 83125,
+      "臊": 83126,
+      "⑮": 83127,
+      "辿": 83128,
+      "₯": 83129,
+      "糅": 83130,
+      "暹": 83131,
+      "畈": 83132,
+      "濿": 83133,
+      "鶉": 83134,
+      "庒": 83135,
+      "撖": 83136,
+      "🇬🇾": 83137,
+      "陥": 83138,
+      "勏": 83139,
+      "耧": 83140,
+      "椘": 83141,
+      "⑰": 83142,
+      "礩": 83143,
+      "攆": 83144,
+      "🇨🇼": 83145,
+      "谘": 83146,
+      "耲": 83147,
+      "钲": 83148,
+      "埉": 83149,
+      "浺": 83150,
+      "乗": 83151,
+      "駈": 83152,
+      "⑾": 83153,
+      "矬": 83154,
+      "鼔": 83155,
+      "û": 83156,
+      "🏷": 83157,
+      "琹": 83158,
+      "椮": 83159,
+      "嗃": 83160,
+      "澒": 83161,
+      "簭": 83162,
+      "斚": 83163,
+      "罃": 83164,
+      "⊋": 83165,
+      "唨": 83166,
+      "鷔": 83167,
+      "鋑": 83168,
+      "驟": 83169,
+      "桫": 83170,
+      "眀": 83171,
+      "蕤": 83172,
+      "頓": 83173,
+      "👁‍🗨": 83174,
+      "🇱🇸": 83175,
+      "恄": 83176,
+      "絘": 83177,
+      "籺": 83178,
+      "👩‍🍳": 83179,
+      "🧑‍💻": 83180,
+      "鍥": 83181,
+      "俬": 83182,
+      "⏗": 83183,
+      "沤": 83184,
+      "鍕": 83185,
+      "蛝": 83186,
+      "潏": 83187,
+      "⚮": 83188,
+      "靤": 83189,
+      "壉": 83190,
+      "眿": 83191,
+      "鞆": 83192,
+      "觇": 83193,
+      "帢": 83194,
+      "⑜": 83195,
+      "蹝": 83196,
+      "葂": 83197,
+      "欿": 83198,
+      "焀": 83199,
+      "斍": 83200,
+      "咷": 83201,
+      "荍": 83202,
+      "柣": 83203,
+      "岷": 83204,
+      "🍧": 83205,
+      "珈": 83206,
+      "祵": 83207,
+      "棰": 83208,
+      "嚟": 83209,
+      "祜": 83210,
+      "斖": 83211,
+      "朠": 83212,
+      "劷": 83213,
+      "綘": 83214,
+      "涊": 83215,
+      "栴": 83216,
+      "彞": 83217,
+      "🥲": 83218,
+      "窧": 83219,
+      "浲": 83220,
+      "應": 83221,
+      "捊": 83222,
+      "ℾ": 83223,
+      "炗": 83224,
+      "ϛ": 83225,
+      "闃": 83226,
+      "牽": 83227,
+      "崏": 83228,
+      "竁": 83229,
+      "缧": 83230,
+      "侑": 83231,
+      "釙": 83232,
+      "🩴": 83233,
+      "覩": 83234,
+      "鑢": 83235,
+      "厷": 83236,
+      "旂": 83237,
+      "😼": 83238,
+      "鈀": 83239,
+      "摺": 83240,
+      "癑": 83241,
+      "簓": 83242,
+      "瞍": 83243,
+      "萒": 83244,
+      "獫": 83245,
+      "窳": 83246,
+      "嚠": 83247,
+      "煪": 83248,
+      "紂": 83249,
+      "涭": 83250,
+      "癕": 83251,
+      "箇": 83252,
+      "棲": 83253,
+      "櫢": 83254,
+      "孑": 83255,
+      "Ɛ": 83256,
+      "齑": 83257,
+      "唵": 83258,
+      "拄": 83259,
+      "覬": 83260,
+      "垇": 83261,
+      "笫": 83262,
+      "繼": 83263,
+      "皺": 83264,
+      "訿": 83265,
+      "鑖": 83266,
+      "戸": 83267,
+      "冏": 83268,
+      "岌": 83269,
+      "飃": 83270,
+      "嵯": 83271,
+      "錔": 83272,
+      "🤹🏻‍♂️": 83273,
+      "剌": 83274,
+      "汩": 83275,
+      "珗": 83276,
+      "犪": 83277,
+      "邭": 83278,
+      "宯": 83279,
+      "痓": 83280,
+      "糑": 83281,
+      "續": 83282,
+      "∽": 83283,
+      "婒": 83284,
+      "⇩": 83285,
+      "眅": 83286,
+      "溾": 83287,
+      "椲": 83288,
+      "嵅": 83289,
+      "籊": 83290,
+      "瘠": 83291,
+      "睙": 83292,
+      "🧪": 83293,
+      "绋": 83294,
+      "矊": 83295,
+      "靨": 83296,
+      "煉": 83297,
+      "💀": 83298,
+      "⛡": 83299,
+      "爰": 83300,
+      "魱": 83301,
+      "鸈": 83302,
+      "琠": 83303,
+      "榹": 83304,
+      "墋": 83305,
+      "壽": 83306,
+      "愭": 83307,
+      "👩🏻‍❤️‍👨🏼": 83308,
+      "胲": 83309,
+      "垩": 83310,
+      "椴": 83311,
+      "鵆": 83312,
+      "絝": 83313,
+      "劒": 83314,
+      "橾": 83315,
+      "🧑🏻‍🍼": 83316,
+      "堣": 83317,
+      "祹": 83318,
+      "抣": 83319,
+      "琁": 83320,
+      "癉": 83321,
+      "🏐": 83322,
+      "⎏": 83323,
+      "✊🏽": 83324,
+      "缵": 83325,
+      "滟": 83326,
+      "渌": 83327,
+      "驐": 83328,
+      "猶": 83329,
+      "謯": 83330,
+      "衪": 83331,
+      "趦": 83332,
+      "🥷": 83333,
+      "🗝": 83334,
+      "鰵": 83335,
+      "諎": 83336,
+      "滿": 83337,
+      "憇": 83338,
+      "哆": 83339,
+      "僅": 83340,
+      "👱🏿": 83341,
+      "🧝": 83342,
+      "鵊": 83343,
+      "铷": 83344,
+      "竏": 83345,
+      "幨": 83346,
+      "閎": 83347,
+      "煨": 83348,
+      "🗯": 83349,
+      "蝰": 83350,
+      "螏": 83351,
+      "羺": 83352,
+      "饓": 83353,
+      "帲": 83354,
+      "骃": 83355,
+      "祃": 83356,
+      "屾": 83357,
+      "怳": 83358,
+      "郳": 83359,
+      "鴶": 83360,
+      "皵": 83361,
+      "仏": 83362,
+      "👱🏿‍♀️": 83363,
+      "Ά": 83364,
+      "🦄": 83365,
+      "跞": 83366,
+      "🦬": 83367,
+      "🇧🇫": 83368,
+      "💚": 83369,
+      "ȟ": 83370,
+      "脥": 83371,
+      "梾": 83372,
+      "鮮": 83373,
+      "攕": 83374,
+      "🧎🏿‍♀️": 83375,
+      "┡": 83376,
+      "鍽": 83377,
+      "👩🏾‍❤️‍💋‍👨🏾": 83378,
+      "嬃": 83379,
+      "˂": 83380,
+      "閕": 83381,
+      "欑": 83382,
+      "濼": 83383,
+      "補": 83384,
+      "Ƿ": 83385,
+      "悁": 83386,
+      "獕": 83387,
+      "酁": 83388,
+      "瘁": 83389,
+      "🧗🏻‍♀️": 83390,
+      "₎": 83391,
+      "濽": 83392,
+      "🕊": 83393,
+      "鮂": 83394,
+      "⌔": 83395,
+      "秔": 83396,
+      "刄": 83397,
+      "柈": 83398,
+      "̭": 83399,
+      "槞": 83400,
+      "夋": 83401,
+      "焾": 83402,
+      "🦸🏾‍♀️": 83403,
+      "ȼ": 83404,
+      "悃": 83405,
+      "🧑🏼‍⚕️": 83406,
+      "泬": 83407,
+      "磗": 83408,
+      "🧑‍🎄": 83409,
+      "機": 83410,
+      "👷‍♀️": 83411,
+      "馪": 83412,
+      "橊": 83413,
+      "媏": 83414,
+      "綑": 83415,
+      "耈": 83416,
+      "幉": 83417,
+      "敻": 83418,
+      "👩🏽‍🚀": 83419,
+      "藳": 83420,
+      "̩": 83421,
+      "⚪": 83422,
+      "卂": 83423,
+      "棎": 83424,
+      "斸": 83425,
+      "瓘": 83426,
+      "恆": 83427,
+      "▾": 83428,
+      "幔": 83429,
+      "厗": 83430,
+      "🇬🇹": 83431,
+      "欘": 83432,
+      "燝": 83433,
+      "獊": 83434,
+      "奷": 83435,
+      "仦": 83436,
+      "櫦": 83437,
+      "鐇": 83438,
+      "賺": 83439,
+      "駍": 83440,
+      "誀": 83441,
+      "镘": 83442,
+      "奱": 83443,
+      "͹": 83444,
+      "擙": 83445,
+      "愖": 83446,
+      "瓤": 83447,
+      "皸": 83448,
+      "獸": 83449,
+      "🖼": 83450,
+      "砋": 83451,
+      "🕵🏻‍♀️": 83452,
+      "🏊🏼": 83453,
+      "籵": 83454,
+      "藿": 83455,
+      "鱜": 83456,
+      "🧝🏾": 83457,
+      "谙": 83458,
+      "⛷": 83459,
+      "匬": 83460,
+      "蝟": 83461,
+      "灬": 83462,
+      "🧜🏾‍♀️": 83463,
+      "": 83464,
+      "幏": 83465,
+      "溒": 83466,
+      "讖": 83467,
+      "嗄": 83468,
+      "槊": 83469,
+      "硸": 83470,
+      "蓣": 83471,
+      "蕵": 83472,
+      "☁": 83473,
+      "櫃": 83474,
+      "潠": 83475,
+      "鶗": 83476,
+      "◘": 83477,
+      "伃": 83478,
+      "攔": 83479,
+      "缳": 83480,
+      "镏": 83481,
+      "🕋": 83482,
+      "👨🏿‍❤️‍💋‍👨🏼": 83483,
+      "頥": 83484,
+      "💼": 83485,
+      "倇": 83486,
+      "🌜": 83487,
+      "💜": 83488,
+      "鏼": 83489,
+      "ͣ": 83490,
+      "曺": 83491,
+      "蚷": 83492,
+      "咛": 83493,
+      "綉": 83494,
+      "⛬": 83495,
+      "鷎": 83496,
+      "鋖": 83497,
+      "踸": 83498,
+      "贽": 83499,
+      "鳑": 83500,
+      "斦": 83501,
+      "矪": 83502,
+      "濰": 83503,
+      "▘": 83504,
+      "彘": 83505,
+      "辏": 83506,
+      "斴": 83507,
+      "逴": 83508,
+      "漴": 83509,
+      "🤽🏻‍♂️": 83510,
+      "⎾": 83511,
+      "🦼": 83512,
+      "纘": 83513,
+      "⒈": 83514,
+      "🧏🏾‍♂️": 83515,
+      "籶": 83516,
+      "侐": 83517,
+      "緤": 83518,
+      "殔": 83519,
+      "🔬": 83520,
+      "蓹": 83521,
+      "鲣": 83522,
+      "莬": 83523,
+      "叄": 83524,
+      "抴": 83525,
+      "叕": 83526,
+      "綿": 83527,
+      "膙": 83528,
+      "淕": 83529,
+      "秠": 83530,
+      "巘": 83531,
+      "粨": 83532,
+      "🏇": 83533,
+      "◯": 83534,
+      "秈": 83535,
+      "髚": 83536,
+      "赳": 83537,
+      "匊": 83538,
+      "晐": 83539,
+      "災": 83540,
+      "廹": 83541,
+      "🤾🏻‍♂️": 83542,
+      "鄷": 83543,
+      "♁": 83544,
+      "呡": 83545,
+      "汸": 83546,
+      "諱": 83547,
+      "蓿": 83548,
+      "峒": 83549,
+      "ʼ": 83550,
+      "崥": 83551,
+      "訰": 83552,
+      "玼": 83553,
+      "忖": 83554,
+      "峀": 83555,
+      "嵄": 83556,
+      "忲": 83557,
+      "🍫": 83558,
+      "偊": 83559,
+      "烱": 83560,
+      "🤶🏼": 83561,
+      "👨🏽‍🎤": 83562,
+      "ϕ": 83563,
+      "🧎🏾": 83564,
+      "榢": 83565,
+      "貙": 83566,
+      "🧑🏼‍🦲": 83567,
+      "駎": 83568,
+      "🙇🏾‍♂️": 83569,
+      "鞴": 83570,
+      "丠": 83571,
+      "毳": 83572,
+      "湰": 83573,
+      "摂": 83574,
+      "慤": 83575,
+      "袧": 83576,
+      "🧑🏼‍❤️‍💋‍🧑🏿": 83577,
+      "馘": 83578,
+      "粛": 83579,
+      "廏": 83580,
+      "瘜": 83581,
+      "幗": 83582,
+      "😴": 83583,
+      "♣︎": 83584,
+      "缯": 83585,
+      "隊": 83586,
+      "鳧": 83587,
+      "Ⓑ": 83588,
+      "鵑": 83589,
+      "顦": 83590,
+      "☔": 83591,
+      "冑": 83592,
+      "闊": 83593,
+      "醬": 83594,
+      "💂‍♂️": 83595,
+      "👦🏽": 83596,
+      "嘃": 83597,
+      "煹": 83598,
+      "丿": 83599,
+      "∰": 83600,
+      "櫓": 83601,
+      "癬": 83602,
+      "囔": 83603,
+      "熼": 83604,
+      "糴": 83605,
+      "菕": 83606,
+      "悈": 83607,
+      "👀": 83608,
+      "🐪": 83609,
+      "🫄🏽": 83610,
+      "🧏🏼‍♀️": 83611,
+      "琧": 83612,
+      "ʐ": 83613,
+      "⏍": 83614,
+      "薩": 83615,
+      "疒": 83616,
+      "羓": 83617,
+      "軬": 83618,
+      "喆": 83619,
+      "👎🏽": 83620,
+      "讆": 83621,
+      "堔": 83622,
+      "🧑🏻‍✈️": 83623,
+      "厩": 83624,
+      "萾": 83625,
+      "悜": 83626,
+      "諗": 83627,
+      "爉": 83628,
+      "铄": 83629,
+      "Ͳ": 83630,
+      "炡": 83631,
+      "搹": 83632,
+      "犛": 83633,
+      "🫲🏻": 83634,
+      "🇸🇬": 83635,
+      "斀": 83636,
+      "蜫": 83637,
+      "惚": 83638,
+      "🌅": 83639,
+      "愒": 83640,
+      "槾": 83641,
+      "璠": 83642,
+      "ƒ": 83643,
+      "準": 83644,
+      "👩🏾‍❤️‍👨🏾": 83645,
+      "竳": 83646,
+      "秵": 83647,
+      "🍏": 83648,
+      "🥖": 83649,
+      "賆": 83650,
+      "🍞": 83651,
+      "瀣": 83652,
+      "鳁": 83653,
+      "👩🏿‍❤️‍💋‍👨🏽": 83654,
+      "🤽🏻": 83655,
+      "欅": 83656,
+      "👷🏽‍♂️": 83657,
+      "忦": 83658,
+      "偛": 83659,
+      "籅": 83660,
+      "ʋ": 83661,
+      "👩🏼‍🎓": 83662,
+      "獟": 83663,
+      "🍬": 83664,
+      "駑": 83665,
+      "惁": 83666,
+      "鏻": 83667,
+      "⎋": 83668,
+      "單": 83669,
+      "窻": 83670,
+      "曓": 83671,
+      "蚇": 83672,
+      "鯾": 83673,
+      "鋹": 83674,
+      "蝼": 83675,
+      "∯": 83676,
+      "🦶🏼": 83677,
+      "踵": 83678,
+      "╞": 83679,
+      "螩": 83680,
+      "鴫": 83681,
+      "洄": 83682,
+      "嵏": 83683,
+      "✫": 83684,
+      "Ⓧ": 83685,
+      "鞽": 83686,
+      "鬬": 83687,
+      "忪": 83688,
+      "貘": 83689,
+      "蟤": 83690,
+      "⇷": 83691,
+      "渋": 83692,
+      "贠": 83693,
+      "🇮🇹": 83694,
+      "钣": 83695,
+      "👩‍🦰": 83696,
+      "捾": 83697,
+      "👩🏾‍❤️‍💋‍👨🏼": 83698,
+      "鬥": 83699,
+      "煱": 83700,
+      "鴑": 83701,
+      "艞": 83702,
+      "⋿": 83703,
+      "貚": 83704,
+      "莊": 83705,
+      "楏": 83706,
+      "菴": 83707,
+      "菿": 83708,
+      "🇦🇮": 83709,
+      "🆎": 83710,
+      "🚣🏾‍♀️": 83711,
+      "🏊🏽‍♀️": 83712,
+      "礵": 83713,
+      "🇸🇸": 83714,
+      "🤏🏾": 83715,
+      "躣": 83716,
+      "扤": 83717,
+      "蟙": 83718,
+      "昛": 83719,
+      "熎": 83720,
+      "樤": 83721,
+      "🈹": 83722,
+      "👙": 83723,
+      "祩": 83724,
+      "爮": 83725,
+      "豈": 83726,
+      "袗": 83727,
+      "澾": 83728,
+      "錗": 83729,
+      "鞮": 83730,
+      "楰": 83731,
+      "鞕": 83732,
+      "喖": 83733,
+      "刯": 83734,
+      "祚": 83735,
+      "🇯🇪": 83736,
+      "̔": 83737,
+      "蝓": 83738,
+      "輂": 83739,
+      "⚜": 83740,
+      "👮‍♀️": 83741,
+      "礆": 83742,
+      "訸": 83743,
+      "鐚": 83744,
+      "絵": 83745,
+      "綯": 83746,
+      "🤸🏽‍♀️": 83747,
+      "鐠": 83748,
+      "泺": 83749,
+      "龥": 83750,
+      "🚜": 83751,
+      "👩🏾‍❤️‍💋‍👩🏾": 83752,
+      "朇": 83753,
+      "🇬🇷": 83754,
+      "🔓": 83755,
+      "湟": 83756,
+      "僇": 83757,
+      "懚": 83758,
+      "緍": 83759,
+      "暭": 83760,
+      "🧑🏾‍🦰": 83761,
+      "🇬🇵": 83762,
+      "燈": 83763,
+      "⊔": 83764,
+      "⟹": 83765,
+      "縳": 83766,
+      "薟": 83767,
+      "鱉": 83768,
+      "🧑🏾‍❤️‍🧑🏽": 83769,
+      "鞚": 83770,
+      "罎": 83771,
+      "葉": 83772,
+      "翗": 83773,
+      "鷹": 83774,
+      "砡": 83775,
+      "蒾": 83776,
+      "靊": 83777,
+      "🐂": 83778,
+      "捤": 83779,
+      "褥": 83780,
+      "袱": 83781,
+      "栜": 83782,
+      "蘆": 83783,
+      "魶": 83784,
+      "ʬ": 83785,
+      "荿": 83786,
+      "醂": 83787,
+      "磄": 83788,
+      "🫵🏾": 83789,
+      "咁": 83790,
+      "沄": 83791,
+      "🦹🏿‍♂️": 83792,
+      "憐": 83793,
+      "靄": 83794,
+      "湪": 83795,
+      "鈎": 83796,
+      "⁙": 83797,
+      "標": 83798,
+      "魃": 83799,
+      "瑨": 83800,
+      "榧": 83801,
+      "祤": 83802,
+      "響": 83803,
+      "毽": 83804,
+      "筟": 83805,
+      "🧑🏼‍❤️‍🧑🏻": 83806,
+      "幒": 83807,
+      "💇🏿‍♀️": 83808,
+      "蕺": 83809,
+      "澣": 83810,
+      "👳🏻": 83811,
+      "嬻": 83812,
+      "⚎": 83813,
+      "👲🏻": 83814,
+      "🐭": 83815,
+      "嬽": 83816,
+      "🙆🏾‍♀️": 83817,
+      "緦": 83818,
+      "掙": 83819,
+      "殹": 83820,
+      "鎝": 83821,
+      "詯": 83822,
+      "阓": 83823,
+      "夢": 83824,
+      "圂": 83825,
+      "驆": 83826,
+      "濭": 83827,
+      "🚴🏿‍♀️": 83828,
+      "╵": 83829,
+      "詥": 83830,
+      "🌀": 83831,
+      "旚": 83832,
+      "🏇🏼": 83833,
+      "鵁": 83834,
+      "₽": 83835,
+      "隞": 83836,
+      "␹": 83837,
+      "␻": 83838,
+      "纡": 83839,
+      "̣": 83840,
+      "👲🏼": 83841,
+      "🤾🏽‍♂️": 83842,
+      "幺": 83843,
+      "⃁": 83844,
+      "脒": 83845,
+      "覻": 83846,
+      "墷": 83847,
+      "搛": 83848,
+      "妔": 83849,
+      "剸": 83850,
+      "揎": 83851,
+      "儸": 83852,
+      "🧑🏻‍🎤": 83853,
+      "赩": 83854,
+      "巎": 83855,
+      "齵": 83856,
+      "⊧": 83857,
+      "窼": 83858,
+      "鹇": 83859,
+      "鍯": 83860,
+      "梀": 83861,
+      "匰": 83862,
+      "掤": 83863,
+      "朑": 83864,
+      "訂": 83865,
+      "稅": 83866,
+      "晸": 83867,
+      "馗": 83868,
+      "₇": 83869,
+      "⃮": 83870,
+      "⇝": 83871,
+      "抌": 83872,
+      "烼": 83873,
+      "桪": 83874,
+      "摤": 83875,
+      "舙": 83876,
+      "嶫": 83877,
+      "薝": 83878,
+      "🤞🏿": 83879,
+      "⎧": 83880,
+      "琻": 83881,
+      "懲": 83882,
+      "🧑🏾": 83883,
+      "沑": 83884,
+      "紿": 83885,
+      "⃆": 83886,
+      "铤": 83887,
+      "ℨ": 83888,
+      "👲🏽": 83889,
+      "諳": 83890,
+      "耠": 83891,
+      "👩🏾‍❤️‍💋‍👨🏻": 83892,
+      "芟": 83893,
+      "嬄": 83894,
+      "👩🏽‍🎓": 83895,
+      "🪃": 83896,
+      "辝": 83897,
+      "矔": 83898,
+      "矏": 83899,
+      "📂": 83900,
+      "懳": 83901,
+      "仺": 83902,
+      "◟": 83903,
+      "黦": 83904,
+      "豝": 83905,
+      "蘷": 83906,
+      "蚼": 83907,
+      "婮": 83908,
+      "韜": 83909,
+      "躙": 83910,
+      "驎": 83911,
+      "🍑": 83912,
+      "獧": 83913,
+      "🪗": 83914,
+      "撏": 83915,
+      "躀": 83916,
+      "👨🏻‍🍳": 83917,
+      "🧑🏽‍💻": 83918,
+      "鏟": 83919,
+      "殐": 83920,
+      "碓": 83921,
+      "箯": 83922,
+      "👨🏾‍🍳": 83923,
+      "樲": 83924,
+      "🙎🏾‍♀️": 83925,
+      "⏞": 83926,
+      "阆": 83927,
+      "匷": 83928,
+      "🤞🏾": 83929,
+      "☻": 83930,
+      "炄": 83931,
+      "🙎🏽‍♀️": 83932,
+      "挴": 83933,
+      "👩🏿‍❤️‍💋‍👩🏿": 83934,
+      "齨": 83935,
+      "蠵": 83936,
+      "慂": 83937,
+      "巺": 83938,
+      "戾": 83939,
+      "诔": 83940,
+      "痱": 83941,
+      "糋": 83942,
+      "🙎🏼": 83943,
+      "佶": 83944,
+      "桛": 83945,
+      "琽": 83946,
+      "豎": 83947,
+      "嚥": 83948,
+      "镐": 83949,
+      "塏": 83950,
+      "擥": 83951,
+      "倐": 83952,
+      "椣": 83953,
+      "👷": 83954,
+      "忇": 83955,
+      "🥩": 83956,
+      "娑": 83957,
+      "⚊": 83958,
+      "籒": 83959,
+      "棗": 83960,
+      "΅": 83961,
+      "🧞‍♀️": 83962,
+      "🦵🏾": 83963,
+      "螀": 83964,
+      "醷": 83965,
+      "筂": 83966,
+      "¼": 83967,
+      "轳": 83968,
+      "ǘ": 83969,
+      "黒": 83970,
+      "镪": 83971,
+      "砘": 83972,
+      "骷": 83973,
+      "豂": 83974,
+      "👨‍🔧": 83975,
+      "👈🏻": 83976,
+      "銸": 83977,
+      "ʫ": 83978,
+      "✂": 83979,
+      "奍": 83980,
+      "祳": 83981,
+      "骭": 83982,
+      "蒹": 83983,
+      "🧏": 83984,
+      "揺": 83985,
+      "沓": 83986,
+      "🧑🏽‍❤️‍💋‍🧑🏼": 83987,
+      "⒵": 83988,
+      "壂": 83989,
+      "Ϧ": 83990,
+      "宻": 83991,
+      "🌂": 83992,
+      "盵": 83993,
+      "🤵🏽‍♀️": 83994,
+      "🫵": 83995,
+      "檆": 83996,
+      "癝": 83997,
+      "紗": 83998,
+      "潡": 83999,
+      "唏": 84000,
+      "岦": 84001,
+      "臤": 84002,
+      "蚈": 84003,
+      "👨🏽‍🍼": 84004,
+      "砨": 84005,
+      "勥": 84006,
+      "徱": 84007,
+      "螛": 84008,
+      "‒": 84009,
+      "懏": 84010,
+      "篲": 84011,
+      "👝": 84012,
+      "繑": 84013,
+      "堦": 84014,
+      "獋": 84015,
+      "👚": 84016,
+      "倠": 84017,
+      "⇭": 84018,
+      "衃": 84019,
+      "鱔": 84020,
+      "👨‍👩‍👧‍👧": 84021,
+      "渵": 84022,
+      "🧛🏼": 84023,
+      "刖": 84024,
+      "桚": 84025,
+      "⑁": 84026,
+      "鵄": 84027,
+      "湌": 84028,
+      "蟐": 84029,
+      "🧗🏼‍♂️": 84030,
+      "ʼn": 84031,
+      "殮": 84032,
+      "ɟ": 84033,
+      "💃": 84034,
+      "馾": 84035,
+      "鞶": 84036,
+      "齺": 84037,
+      "嘑": 84038,
+      "媶": 84039,
+      "膵": 84040,
+      "洨": 84041,
+      "櫱": 84042,
+      "羔": 84043,
+      "🇵🇭": 84044,
+      "🧑🏾‍🦱": 84045,
+      "莧": 84046,
+      "宎": 84047,
+      "蒁": 84048,
+      "牆": 84049,
+      "搨": 84050,
+      "鶔": 84051,
+      "胮": 84052,
+      "蘋": 84053,
+      "穄": 84054,
+      "鎓": 84055,
+      "巊": 84056,
+      "袳": 84057,
+      "黇": 84058,
+      "剄": 84059,
+      "孷": 84060,
+      "貯": 84061,
+      "髤": 84062,
+      "嗞": 84063,
+      "鬃": 84064,
+      "👰🏻‍♀️": 84065,
+      "颛": 84066,
+      "ɔ": 84067,
+      "耢": 84068,
+      "竓": 84069,
+      "脈": 84070,
+      "紼": 84071,
+      "谡": 84072,
+      "芫": 84073,
+      "鄻": 84074,
+      "🕺🏽": 84075,
+      "雰": 84076,
+      "賀": 84077,
+      "≽": 84078,
+      "嫴": 84079,
+      "駕": 84080,
+      "饘": 84081,
+      "👩‍💻": 84082,
+      "鹋": 84083,
+      "␍": 84084,
+      "鶓": 84085,
+      "婃": 84086,
+      "孺": 84087,
+      "伛": 84088,
+      "兩": 84089,
+      "遆": 84090,
+      "ℭ": 84091,
+      "眳": 84092,
+      "孉": 84093,
+      "賯": 84094,
+      "柺": 84095,
+      "癤": 84096,
+      "舺": 84097,
+      "餹": 84098,
+      "唝": 84099,
+      "悰": 84100,
+      "鞸": 84101,
+      "醐": 84102,
+      "庮": 84103,
+      "肧": 84104,
+      "⏴": 84105,
+      "瀳": 84106,
+      "墣": 84107,
+      "鑧": 84108,
+      "仾": 84109,
+      "蓼": 84110,
+      "擬": 84111,
+      "爘": 84112,
+      "猓": 84113,
+      "瀀": 84114,
+      "唈": 84115,
+      "嶩": 84116,
+      "👈🏼": 84117,
+      "簥": 84118,
+      "崣": 84119,
+      "鷬": 84120,
+      "臜": 84121,
+      "摾": 84122,
+      "軦": 84123,
+      "🍙": 84124,
+      "獤": 84125,
+      "🦸🏻‍♀️": 84126,
+      "掯": 84127,
+      "🐻‍❄️": 84128,
+      "傓": 84129,
+      "癰": 84130,
+      "🦛": 84131,
+      "彋": 84132,
+      "Ȭ": 84133,
+      "鍂": 84134,
+      "祉": 84135,
+      "聓": 84136,
+      "🫱🏼‍🫲🏻": 84137,
+      "単": 84138,
+      "瀊": 84139,
+      "鞹": 84140,
+      "◈": 84141,
+      "櫰": 84142,
+      "蜧": 84143,
+      "🇸🇲": 84144,
+      "绤": 84145,
+      "◊": 84146,
+      "ʎ": 84147,
+      "🧘🏿": 84148,
+      "碞": 84149,
+      "悾": 84150,
+      "🤸🏾‍♂️": 84151,
+      "鼂": 84152,
+      "焞": 84153,
+      "蕢": 84154,
+      "計": 84155,
+      "虂": 84156,
+      "稁": 84157,
+      "銫": 84158,
+      "茘": 84159,
+      "👶": 84160,
+      "ƕ": 84161,
+      "♄": 84162,
+      "瓊": 84163,
+      "怾": 84164,
+      "忥": 84165,
+      "腇": 84166,
+      "醏": 84167,
+      "铉": 84168,
+      "怗": 84169,
+      "揚": 84170,
+      "🙌🏽": 84171,
+      "贂": 84172,
+      "鰱": 84173,
+      "髿": 84174,
+      "幎": 84175,
+      "嫵": 84176,
+      "娆": 84177,
+      "轒": 84178,
+      "忹": 84179,
+      "岘": 84180,
+      "↠": 84181,
+      "👨🏼": 84182,
+      "偌": 84183,
+      "‌": 84184,
+      "奻": 84185,
+      "匦": 84186,
+      "🤦": 84187,
+      "穊": 84188,
+      "獭": 84189,
+      "酗": 84190,
+      "怙": 84191,
+      "漖": 84192,
+      "鉋": 84193,
+      "矸": 84194,
+      "骢": 84195,
+      "椏": 84196,
+      "鐷": 84197,
+      "傒": 84198,
+      "驛": 84199,
+      "縓": 84200,
+      "飰": 84201,
+      "Ŀ": 84202,
+      "敓": 84203,
+      "鑰": 84204,
+      "鉄": 84205,
+      "🥷🏽": 84206,
+      "衘": 84207,
+      "颼": 84208,
+      "🖖🏻": 84209,
+      "篼": 84210,
+      "侘": 84211,
+      "傄": 84212,
+      "詖": 84213,
+      "🍎": 84214,
+      "縻": 84215,
+      "皩": 84216,
+      "喣": 84217,
+      "昜": 84218,
+      "🧙🏿": 84219,
+      "覞": 84220,
+      "✮": 84221,
+      "瘝": 84222,
+      "覰": 84223,
+      "✵": 84224,
+      "🏺": 84225,
+      "魜": 84226,
+      "鯭": 84227,
+      "薢": 84228,
+      "遹": 84229,
+      "冦": 84230,
+      "🥻": 84231,
+      "慭": 84232,
+      "輭": 84233,
+      "鋏": 84234,
+      "挾": 84235,
+      "擐": 84236,
+      "🧑🏿‍🔧": 84237,
+      "🙅🏿": 84238,
+      "電": 84239,
+      "蹌": 84240,
+      "⊉": 84241,
+      "¾": 84242,
+      "熮": 84243,
+      "🥹": 84244,
+      "臁": 84245,
+      "🦟": 84246,
+      "隉": 84247,
+      "劓": 84248,
+      "🙍🏽‍♂️": 84249,
+      "鲗": 84250,
+      "ʅ": 84251,
+      "🎹": 84252,
+      "👨🏼‍❤️‍💋‍👨🏽": 84253,
+      "☞": 84254,
+      "頒": 84255,
+      "嚵": 84256,
+      "杼": 84257,
+      "瀮": 84258,
+      "🇮🇸": 84259,
+      "礓": 84260,
+      "桾": 84261,
+      "謜": 84262,
+      "👩🏻‍🎓": 84263,
+      "鼲": 84264,
+      "醄": 84265,
+      "嫹": 84266,
+      "👨🏽‍🔧": 84267,
+      "嵱": 84268,
+      "咶": 84269,
+      "峷": 84270,
+      "ℴ": 84271,
+      "赸": 84272,
+      "⃻": 84273,
+      "闗": 84274,
+      "踎": 84275,
+      "駙": 84276,
+      "堍": 84277,
+      "崵": 84278,
+      "⍰": 84279,
+      "˲": 84280,
+      "諿": 84281,
+      "賰": 84282,
+      "蓢": 84283,
+      "襱": 84284,
+      "齛": 84285,
+      "銃": 84286,
+      "輣": 84287,
+      "囜": 84288,
+      "蒧": 84289,
+      "蝩": 84290,
+      "盉": 84291,
+      "🤟🏼": 84292,
+      "🇱🇰": 84293,
+      "檌": 84294,
+      "椝": 84295,
+      "👩🏼‍🔬": 84296,
+      "隻": 84297,
+      "迬": 84298,
+      "̴": 84299,
+      "✏": 84300,
+      "擩": 84301,
+      "柎": 84302,
+      "庢": 84303,
+      "棑": 84304,
+      "聸": 84305,
+      "輌": 84306,
+      "徨": 84307,
+      "荠": 84308,
+      "🧑🏿‍🚀": 84309,
+      "欝": 84310,
+      "玅": 84311,
+      "旖": 84312,
+      "偓": 84313,
+      "喋": 84314,
+      "⋨": 84315,
+      "倵": 84316,
+      "曯": 84317,
+      "Ļ": 84318,
+      "ɚ": 84319,
+      "👸🏽": 84320,
+      "鼒": 84321,
+      "泟": 84322,
+      "迂": 84323,
+      "⑐": 84324,
+      "嚹": 84325,
+      "噭": 84326,
+      "⃏": 84327,
+      "騖": 84328,
+      "蝜": 84329,
+      "鹹": 84330,
+      "⓾": 84331,
+      "摷": 84332,
+      "迉": 84333,
+      "鼯": 84334,
+      "傾": 84335,
+      "🕖": 84336,
+      "狜": 84337,
+      "⛴": 84338,
+      "🖤": 84339,
+      "🇩🇯": 84340,
+      "喬": 84341,
+      "烔": 84342,
+      "騱": 84343,
+      "盍": 84344,
+      "偂": 84345,
+      "酭": 84346,
+      "🇮🇶": 84347,
+      "峢": 84348,
+      "🧞": 84349,
+      "賃": 84350,
+      "↤": 84351,
+      "📟": 84352,
+      "␂": 84353,
+      "篂": 84354,
+      "璍": 84355,
+      "琱": 84356,
+      "夼": 84357,
+      "須": 84358,
+      "癧": 84359,
+      "鞱": 84360,
+      "塶": 84361,
+      "鵙": 84362,
+      "鯐": 84363,
+      "唙": 84364,
+      "崫": 84365,
+      "舥": 84366,
+      "⒆": 84367,
+      "譐": 84368,
+      "嵨": 84369,
+      "褭": 84370,
+      "ħ": 84371,
+      "仃": 84372,
+      "滻": 84373,
+      "猹": 84374,
+      "皯": 84375,
+      "佤": 84376,
+      "🐯": 84377,
+      "鮀": 84378,
+      "篺": 84379,
+      "👨🏾‍🌾": 84380,
+      "齸": 84381,
+      "鐰": 84382,
+      "薍": 84383,
+      "蕜": 84384,
+      "🐛": 84385,
+      "軩": 84386,
+      "園": 84387,
+      "漕": 84388,
+      "耉": 84389,
+      "鈶": 84390,
+      "摞": 84391,
+      "觨": 84392,
+      "鮯": 84393,
+      "鲕": 84394,
+      "ʮ": 84395,
+      "辚": 84396,
+      "獓": 84397,
+      "蛧": 84398,
+      "💩": 84399,
+      "◛": 84400,
+      "眘": 84401,
+      "鸬": 84402,
+      "箼": 84403,
+      "祣": 84404,
+      "🧙🏾": 84405,
+      "詤": 84406,
+      "錆": 84407,
+      "貟": 84408,
+      "": 84409,
+      "誙": 84410,
+      "蹾": 84411,
+      "鎎": 84412,
+      "甖": 84413,
+      "礋": 84414,
+      "🧜🏻": 84415,
+      "嚨": 84416,
+      "👮🏽": 84417,
+      "尡": 84418,
+      "👂🏿": 84419,
+      "👩🏾‍❤️‍👨🏿": 84420,
+      "🪫": 84421,
+      "鳈": 84422,
+      "詈": 84423,
+      "倬": 84424,
+      "🧑🏼‍🦱": 84425,
+      "浀": 84426,
+      "鈦": 84427,
+      "皤": 84428,
+      "蜰": 84429,
+      "💁🏿": 84430,
+      "鑒": 84431,
+      "琰": 84432,
+      "績": 84433,
+      "叒": 84434,
+      "勌": 84435,
+      "緸": 84436,
+      "鏸": 84437,
+      "摶": 84438,
+      "🧑‍🎓": 84439,
+      "鎕": 84440,
+      "龀": 84441,
+      "鶆": 84442,
+      "ɵ": 84443,
+      "墉": 84444,
+      "钺": 84445,
+      "眓": 84446,
+      "🙎🏿‍♂️": 84447,
+      "腖": 84448,
+      "嵀": 84449,
+      "倌": 84450,
+      "勝": 84451,
+      "灙": 84452,
+      "崒": 84453,
+      "謾": 84454,
+      "珏": 84455,
+      "帍": 84456,
+      "麢": 84457,
+      "梿": 84458,
+      "硯": 84459,
+      "Õ": 84460,
+      "螆": 84461,
+      "陚": 84462,
+      "氙": 84463,
+      "⏑": 84464,
+      "柸": 84465,
+      "˻": 84466,
+      "衑": 84467,
+      "岲": 84468,
+      "👨🏾‍❤️‍💋‍👨🏻": 84469,
+      "哻": 84470,
+      "螄": 84471,
+      "👨🏻‍🦲": 84472,
+      "郱": 84473,
+      "鳙": 84474,
+      "積": 84475,
+      "昰": 84476,
+      "牱": 84477,
+      "濁": 84478,
+      "虉": 84479,
+      "餝": 84480,
+      "寎": 84481,
+      "恃": 84482,
+      "💱": 84483,
+      "煬": 84484,
+      "儭": 84485,
+      "樶": 84486,
+      "赗": 84487,
+      "🌁": 84488,
+      "覯": 84489,
+      "🤷🏿‍♂️": 84490,
+      "鑷": 84491,
+      "蓵": 84492,
+      "師": 84493,
+      "掬": 84494,
+      "ℶ": 84495,
+      "槭": 84496,
+      "扴": 84497,
+      "🍼": 84498,
+      "袽": 84499,
+      "渷": 84500,
+      "驚": 84501,
+      "🧑🏾‍🎤": 84502,
+      "侴": 84503,
+      "∺": 84504,
+      "芴": 84505,
+      "衋": 84506,
+      "儐": 84507,
+      "✹": 84508,
+      "睍": 84509,
+      "庠": 84510,
+      "縺": 84511,
+      "🗽": 84512,
+      "鈩": 84513,
+      "椐": 84514,
+      "満": 84515,
+      "鑜": 84516,
+      "劄": 84517,
+      "統": 84518,
+      "ɢ": 84519,
+      "⑸": 84520,
+      "🎰": 84521,
+      "癋": 84522,
+      "嗭": 84523,
+      "🫱🏾‍🫲🏼": 84524,
+      "抃": 84525,
+      "魠": 84526,
+      "惔": 84527,
+      "蝒": 84528,
+      "嶊": 84529,
+      "礲": 84530,
+      "👦🏾": 84531,
+      "喹": 84532,
+      "墵": 84533,
+      "閩": 84534,
+      "浢": 84535,
+      "喒": 84536,
+      "惘": 84537,
+      "🫰": 84538,
+      "鏝": 84539,
+      "猈": 84540,
+      "屒": 84541,
+      "竰": 84542,
+      "藨": 84543,
+      "╗": 84544,
+      "甍": 84545,
+      "纖": 84546,
+      "滉": 84547,
+      "獉": 84548,
+      "鬪": 84549,
+      "蹏": 84550,
+      "👩‍🏫": 84551,
+      "斏": 84552,
+      "泿": 84553,
+      "諼": 84554,
+      "侎": 84555,
+      "堮": 84556,
+      "🐜": 84557,
+      "🏊🏻‍♂️": 84558,
+      "👨🏽‍🚒": 84559,
+      "⋻": 84560,
+      "緙": 84561,
+      "髝": 84562,
+      "🚂": 84563,
+      "Ⅺ": 84564,
+      "怆": 84565,
+      "👩🏾‍💼": 84566,
+      "餵": 84567,
+      "塛": 84568,
+      "Ｃ": 84569,
+      "廕": 84570,
+      "潖": 84571,
+      "ď": 84572,
+      "暼": 84573,
+      "愰": 84574,
+      "👼🏽": 84575,
+      "嗍": 84576,
+      "梌": 84577,
+      "祟": 84578,
+      "葿": 84579,
+      "➙": 84580,
+      "裢": 84581,
+      "铚": 84582,
+      "淁": 84583,
+      "錄": 84584,
+      "鐁": 84585,
+      "嚝": 84586,
+      "🕞": 84587,
+      "娖": 84588,
+      "阏": 84589,
+      "鲵": 84590,
+      "悶": 84591,
+      "吔": 84592,
+      "🔁": 84593,
+      "肕": 84594,
+      "鰐": 84595,
+      "蓋": 84596,
+      "黧": 84597,
+      "戶": 84598,
+      "🧗🏼‍♀️": 84599,
+      "✬": 84600,
+      "🏊🏻‍♀️": 84601,
+      "軽": 84602,
+      "叞": 84603,
+      "纻": 84604,
+      "🚣": 84605,
+      "廯": 84606,
+      "侳": 84607,
+      "訅": 84608,
+      "癲": 84609,
+      "躛": 84610,
+      "璕": 84611,
+      "ij": 84612,
+      "曱": 84613,
+      "✣": 84614,
+      "🕴🏼": 84615,
+      "譕": 84616,
+      "姙": 84617,
+      "耔": 84618,
+      "∄": 84619,
+      "ϥ": 84620,
+      "📡": 84621,
+      "ɋ": 84622,
+      "鸂": 84623,
+      "鼵": 84624,
+      "矹": 84625,
+      "鹳": 84626,
+      "徥": 84627,
+      "▶": 84628,
+      "椤": 84629,
+      "钖": 84630,
+      "隍": 84631,
+      "🪸": 84632,
+      "杚": 84633,
+      "攊": 84634,
+      "筡": 84635,
+      "笸": 84636,
+      "🇾🇹": 84637,
+      "尛": 84638,
+      "瑿": 84639,
+      "葈": 84640,
+      "紡": 84641,
+      "騹": 84642,
+      "塩": 84643,
+      "🤭": 84644,
+      "犏": 84645,
+      "杬": 84646,
+      "粵": 84647,
+      "🪴": 84648,
+      "⇟": 84649,
+      "魤": 84650,
+      "硶": 84651,
+      "📗": 84652,
+      "笡": 84653,
+      "屔": 84654,
+      "嵣": 84655,
+      "🦻": 84656,
+      "傜": 84657,
+      "皳": 84658,
+      "瞓": 84659,
+      "茴": 84660,
+      "✌🏽": 84661,
+      "ʺ": 84662,
+      "🧔🏻": 84663,
+      "邒": 84664,
+      "🥡": 84665,
+      "媡": 84666,
+      "Ψ": 84667,
+      "◀": 84668,
+      "崱": 84669,
+      "⁑": 84670,
+      "🕴": 84671,
+      "檅": 84672,
+      "跅": 84673,
+      "👨🏿‍🎓": 84674,
+      "慐": 84675,
+      "⇡": 84676,
+      "轌": 84677,
+      "拑": 84678,
+      "輚": 84679,
+      "碙": 84680,
+      "軭": 84681,
+      "黆": 84682,
+      "躇": 84683,
+      "啽": 84684,
+      "🏋🏽‍♀️": 84685,
+      "嵼": 84686,
+      "鬈": 84687,
+      "肶": 84688,
+      "𓀄": 84689,
+      "泫": 84690,
+      "貦": 84691,
+      "鈔": 84692,
+      "̲": 84693,
+      "蛌": 84694,
+      "Ｅ": 84695,
+      "👮🏼‍♀️": 84696,
+      "熍": 84697,
+      "旃": 84698,
+      "襲": 84699,
+      "🧑🏻‍🦽": 84700,
+      "📮": 84701,
+      "俙": 84702,
+      "乲": 84703,
+      "🌾": 84704,
+      "胘": 84705,
+      "紑": 84706,
+      "蕳": 84707,
+      "膢": 84708,
+      "詩": 84709,
+      "鵱": 84710,
+      "Ů": 84711,
+      "戙": 84712,
+      "↜": 84713,
+      "葀": 84714,
+      "嬼": 84715,
+      "漈": 84716,
+      "掼": 84717,
+      "禢": 84718,
+      "ƹ": 84719,
+      "绢": 84720,
+      "齳": 84721,
+      "硎": 84722,
+      "🇭🇺": 84723,
+      "靈": 84724,
+      "嵝": 84725,
+      "箬": 84726,
+      "沀": 84727,
+      "籓": 84728,
+      "罤": 84729,
+      "檕": 84730,
+      "昬": 84731,
+      "暆": 84732,
+      "⋋": 84733,
+      "蟌": 84734,
+      "瀥": 84735,
+      "傚": 84736,
+      "淍": 84737,
+      "藬": 84738,
+      "姤": 84739,
+      "霗": 84740,
+      "篛": 84741,
+      "ʵ": 84742,
+      "渎": 84743,
+      "🧤": 84744,
+      "溷": 84745,
+      "揊": 84746,
+      "藀": 84747,
+      "🚁": 84748,
+      "🫴🏻": 84749,
+      "戀": 84750,
+      "钅": 84751,
+      "📚": 84752,
+      "Ｑ": 84753,
+      "誔": 84754,
+      "🤕": 84755,
+      "麰": 84756,
+      "🪨": 84757,
+      "🙍🏽‍♀️": 84758,
+      "鳭": 84759,
+      "🤘": 84760,
+      "🐿": 84761,
+      "俇": 84762,
+      "慾": 84763,
+      "缼": 84764,
+      "翊": 84765,
+      "娊": 84766,
+      "峫": 84767,
+      "孯": 84768,
+      "┘": 84769,
+      "荶": 84770,
+      "攅": 84771,
+      "繗": 84772,
+      "畚": 84773,
+      "缣": 84774,
+      "🫐": 84775,
+      "喨": 84776,
+      "愑": 84777,
+      "芶": 84778,
+      "焑": 84779,
+      "凍": 84780,
+      "🇸🇨": 84781,
+      "猭": 84782,
+      "輓": 84783,
+      "颹": 84784,
+      "鮅": 84785,
+      "ͳ": 84786,
+      "🦥": 84787,
+      "🧑🏿‍🎓": 84788,
+      "糿": 84789,
+      "跙": 84790,
+      "騅": 84791,
+      "熢": 84792,
+      "擱": 84793,
+      "擇": 84794,
+      "搲": 84795,
+      "夻": 84796,
+      "眐": 84797,
+      "畁": 84798,
+      "Ｊ": 84799,
+      "闁": 84800,
+      "堩": 84801,
+      "₄": 84802,
+      "鑑": 84803,
+      "抳": 84804,
+      "嚊": 84805,
+      "╷": 84806,
+      "掝": 84807,
+      "潙": 84808,
+      "穾": 84809,
+      "👩🏾‍🦽": 84810,
+      "遈": 84811,
+      "敍": 84812,
+      "洎": 84813,
+      "虠": 84814,
+      "␦": 84815,
+      "諚": 84816,
+      "♐": 84817,
+      "涜": 84818,
+      "烅": 84819,
+      "👨‍🎤": 84820,
+      "祎": 84821,
+      "賏": 84822,
+      "👩🏽‍❤️‍💋‍👨🏽": 84823,
+      "🇳🇱": 84824,
+      "👲🏿": 84825,
+      "蛯": 84826,
+      "簖": 84827,
+      "邜": 84828,
+      "燗": 84829,
+      "钯": 84830,
+      "烚": 84831,
+      "熱": 84832,
+      "塸": 84833,
+      "猲": 84834,
+      "Ή": 84835,
+      "巻": 84836,
+      "瑖": 84837,
+      "鶴": 84838,
+      "売": 84839,
+      "榶": 84840,
+      "妤": 84841,
+      "👨🏻‍🦰": 84842,
+      "蟸": 84843,
+      "硾": 84844,
+      "楌": 84845,
+      "⅀": 84846,
+      "甙": 84847,
+      "🧄": 84848,
+      "騸": 84849,
+      "剾": 84850,
+      "囉": 84851,
+      "詳": 84852,
+      "圏": 84853,
+      "隰": 84854,
+      "嬲": 84855,
+      "🎙": 84856,
+      "篐": 84857,
+      "磾": 84858,
+      "軰": 84859,
+      "👩🏻‍⚕️": 84860,
+      "箲": 84861,
+      "梬": 84862,
+      "銆": 84863,
+      "騑": 84864,
+      "諟": 84865,
+      "🧑🏻‍❤️‍💋‍🧑🏿": 84866,
+      "鮲": 84867,
+      "膡": 84868,
+      "牰": 84869,
+      "楃": 84870,
+      "銙": 84871,
+      "膒": 84872,
+      "纛": 84873,
+      "釦": 84874,
+      "淛": 84875,
+      "咆": 84876,
+      "纗": 84877,
+      "儻": 84878,
+      "冩": 84879,
+      "擣": 84880,
+      "拲": 84881,
+      "竴": 84882,
+      "钔": 84883,
+      "鳘": 84884,
+      "0️⃣": 84885,
+      "壈": 84886,
+      "黙": 84887,
+      "🏊🏿": 84888,
+      "餩": 84889,
+      "煍": 84890,
+      "阞": 84891,
+      "垖": 84892,
+      "鄈": 84893,
+      "辴": 84894,
+      "郫": 84895,
+      "⚋": 84896,
+      "殕": 84897,
+      "蕗": 84898,
+      "貎": 84899,
+      "汮": 84900,
+      "陁": 84901,
+      "🤶🏾": 84902,
+      "駦": 84903,
+      "缞": 84904,
+      "踋": 84905,
+      "👘": 84906,
+      "禫": 84907,
+      "🍺": 84908,
+      "罞": 84909,
+      "👨🏼‍❤️‍💋‍👨🏾": 84910,
+      "畾": 84911,
+      "邰": 84912,
+      "⇄": 84913,
+      "笊": 84914,
+      "苪": 84915,
+      "靯": 84916,
+      "凬": 84917,
+      "皃": 84918,
+      "傭": 84919,
+      "Ƞ": 84920,
+      "颦": 84921,
+      "壇": 84922,
+      "犍": 84923,
+      "痿": 84924,
+      "湇": 84925,
+      "蛏": 84926,
+      "梹": 84927,
+      "隢": 84928,
+      "🏌🏼‍♂️": 84929,
+      "⍀": 84930,
+      "🙍🏿‍♀️": 84931,
+      "萅": 84932,
+      "桺": 84933,
+      "⑒": 84934,
+      "🏴󠁧󠁢󠁥󠁮󠁧󠁿": 84935,
+      "⒗": 84936,
+      "岨": 84937,
+      "⇏": 84938,
+      "旰": 84939,
+      "🫱🏽‍🫲🏼": 84940,
+      "嬏": 84941,
+      "蠁": 84942,
+      "🏜": 84943,
+      "⛽": 84944,
+      "噣": 84945,
+      "蝆": 84946,
+      "窡": 84947,
+      "翫": 84948,
+      "ɾ": 84949,
+      "煴": 84950,
+      "攱": 84951,
+      "臺": 84952,
+      "韙": 84953,
+      "🇨🇻": 84954,
+      "徃": 84955,
+      "盩": 84956,
+      "東": 84957,
+      "馻": 84958,
+      "🙁": 84959,
+      "偎": 84960,
+      "⃪": 84961,
+      "笖": 84962,
+      "認": 84963,
+      "瞥": 84964,
+      "吮": 84965,
+      "莤": 84966,
+      "躴": 84967,
+      "剒": 84968,
+      "鎘": 84969,
+      "礨": 84970,
+      "罌": 84971,
+      "脘": 84972,
+      "麜": 84973,
+      "睎": 84974,
+      "𓀓": 84975,
+      "陙": 84976,
+      "⛣": 84977,
+      "峉": 84978,
+      "喠": 84979,
+      "圉": 84980,
+      "È": 84981,
+      "臛": 84982,
+      "瑙": 84983,
+      "輢": 84984,
+      "溲": 84985,
+      "窸": 84986,
+      "齅": 84987,
+      "澏": 84988,
+      "捈": 84989,
+      "鳩": 84990,
+      "緈": 84991,
+      "🤺": 84992,
+      "眏": 84993,
+      "嗬": 84994,
+      "懶": 84995,
+      "撉": 84996,
+      "愪": 84997,
+      "🇲🇬": 84998,
+      "鲓": 84999,
+      "熿": 85000,
+      "杄": 85001,
+      "蓷": 85002,
+      "◓": 85003,
+      "亱": 85004,
+      "圥": 85005,
+      "谪": 85006,
+      "軱": 85007,
+      "鲙": 85008,
+      "庥": 85009,
+      "鴪": 85010,
+      "🥕": 85011,
+      "嚢": 85012,
+      "🟣": 85013,
+      "杢": 85014,
+      "溕": 85015,
+      "🧑🏾‍🌾": 85016,
+      "擼": 85017,
+      "裎": 85018,
+      "␮": 85019,
+      "🧑🏿‍🚒": 85020,
+      "徤": 85021,
+      "彨": 85022,
+      "嘞": 85023,
+      "🦹🏽‍♀️": 85024,
+      "愗": 85025,
+      "芑": 85026,
+      "😵‍💫": 85027,
+      "鱺": 85028,
+      "獽": 85029,
+      "誗": 85030,
+      "茥": 85031,
+      "矚": 85032,
+      "虼": 85033,
+      "瑍": 85034,
+      "👴🏼": 85035,
+      "寉": 85036,
+      "🥞": 85037,
+      "瀏": 85038,
+      "珞": 85039,
+      "骐": 85040,
+      "鄛": 85041,
+      "峝": 85042,
+      "楑": 85043,
+      "翞": 85044,
+      "灓": 85045,
+      "莜": 85046,
+      "卟": 85047,
+      "顫": 85048,
+      "烥": 85049,
+      "徸": 85050,
+      "鐞": 85051,
+      "犫": 85052,
+      "🤵🏻‍♂️": 85053,
+      "👱🏽‍♂️": 85054,
+      "蝔": 85055,
+      "😣": 85056,
+      "鱍": 85057,
+      "糷": 85058,
+      "⑦": 85059,
+      "竃": 85060,
+      "⌄": 85061,
+      "粃": 85062,
+      "灪": 85063,
+      "綄": 85064,
+      "🙋🏿‍♂️": 85065,
+      "鼝": 85066,
+      "姂": 85067,
+      "👩‍👧‍👦": 85068,
+      "賈": 85069,
+      "誌": 85070,
+      "🇻🇮": 85071,
+      "縝": 85072,
+      "🚋": 85073,
+      "匎": 85074,
+      "鼭": 85075,
+      "貈": 85076,
+      "🧎‍♂️": 85077,
+      "暵": 85078,
+      "囶": 85079,
+      "趼": 85080,
+      "瘊": 85081,
+      "抆": 85082,
+      "╼": 85083,
+      "🤲🏼": 85084,
+      "誵": 85085,
+      "👍🏿": 85086,
+      "腭": 85087,
+      "酛": 85088,
+      "穧": 85089,
+      "兞": 85090,
+      "潝": 85091,
+      "👨🏼‍🎨": 85092,
+      "鴔": 85093,
+      "Ƌ": 85094,
+      "塜": 85095,
+      "諭": 85096,
+      "␕": 85097,
+      "簨": 85098,
+      "讪": 85099,
+      "殬": 85100,
+      "躕": 85101,
+      "稫": 85102,
+      "魾": 85103,
+      "ɴ": 85104,
+      "犲": 85105,
+      "谳": 85106,
+      "儳": 85107,
+      "紁": 85108,
+      "啷": 85109,
+      "🖐🏼": 85110,
+      "蓲": 85111,
+      "颱": 85112,
+      "賵": 85113,
+      "≭": 85114,
+      "⏩": 85115,
+      "龛": 85116,
+      "轡": 85117,
+      "艑": 85118,
+      "颳": 85119,
+      "軶": 85120,
+      "👩🏿‍🦲": 85121,
+      "煚": 85122,
+      "靎": 85123,
+      "廎": 85124,
+      "蘴": 85125,
+      "秿": 85126,
+      "讐": 85127,
+      "雖": 85128,
+      "霿": 85129,
+      "🧝🏿": 85130,
+      "⎎": 85131,
+      "陸": 85132,
+      "浳": 85133,
+      "🎏": 85134,
+      "🛶": 85135,
+      "齃": 85136,
+      "蔲": 85137,
+      "◼": 85138,
+      "趏": 85139,
+      "馧": 85140,
+      "論": 85141,
+      "蘖": 85142,
+      "鄓": 85143,
+      "宖": 85144,
+      "喼": 85145,
+      "🫄🏾": 85146,
+      "瓰": 85147,
+      "槀": 85148,
+      "睓": 85149,
+      "͚": 85150,
+      "蕁": 85151,
+      "簩": 85152,
+      "🛞": 85153,
+      "碚": 85154,
+      "鹽": 85155,
+      "鲾": 85156,
+      "蟫": 85157,
+      "遡": 85158,
+      "∨": 85159,
+      "厳": 85160,
+      "╌": 85161,
+      "犗": 85162,
+      "彇": 85163,
+      "▊": 85164,
+      "氤": 85165,
+      "貒": 85166,
+      "筴": 85167,
+      "牊": 85168,
+      "": 85169,
+      "鮩": 85170,
+      "誹": 85171,
+      "飩": 85172,
+      "吰": 85173,
+      "邡": 85174,
+      "孥": 85175,
+      "圤": 85176,
+      "涠": 85177,
+      "撀": 85178,
+      "恥": 85179,
+      "藘": 85180,
+      "竗": 85181,
+      "糪": 85182,
+      "晫": 85183,
+      "撋": 85184,
+      "蠐": 85185,
+      "塚": 85186,
+      "筁": 85187,
+      "🫃🏿": 85188,
+      "朲": 85189,
+      "饴": 85190,
+      "璜": 85191,
+      "┍": 85192,
+      "邇": 85193,
+      "郻": 85194,
+      "␶": 85195,
+      "琏": 85196,
+      "侁": 85197,
+      "锔": 85198,
+      "昐": 85199,
+      "曻": 85200,
+      "褔": 85201,
+      "🏌🏻‍♂️": 85202,
+      "🤸🏿‍♂️": 85203,
+      "矴": 85204,
+      "箏": 85205,
+      "簍": 85206,
+      "贏": 85207,
+      "篙": 85208,
+      "颧": 85209,
+      "阼": 85210,
+      "哧": 85211,
+      "挀": 85212,
+      "繖": 85213,
+      "栒": 85214,
+      "觫": 85215,
+      "▦": 85216,
+      "埝": 85217,
+      "剗": 85218,
+      "亾": 85219,
+      "♫": 85220,
+      "鮭": 85221,
+      "編": 85222,
+      "湩": 85223,
+      "🐶": 85224,
+      "揾": 85225,
+      "隒": 85226,
+      "褘": 85227,
+      "🙍🏿‍♂️": 85228,
+      "唰": 85229,
+      "🗓": 85230,
+      "👩‍🔬": 85231,
+      "嚩": 85232,
+      "遃": 85233,
+      "▃": 85234,
+      "羮": 85235,
+      "勸": 85236,
+      "檹": 85237,
+      "匴": 85238,
+      "彊": 85239,
+      "孡": 85240,
+      "₨": 85241,
+      "玈": 85242,
+      "葄": 85243,
+      "滾": 85244,
+      "饔": 85245,
+      "蛲": 85246,
+      "蚯": 85247,
+      "🙅🏾": 85248,
+      "臥": 85249,
+      "ȥ": 85250,
+      "鮌": 85251,
+      "ƌ": 85252,
+      "齲": 85253,
+      "⌤": 85254,
+      "礅": 85255,
+      "◕": 85256,
+      "唖": 85257,
+      "蘍": 85258,
+      "慥": 85259,
+      "臵": 85260,
+      "幯": 85261,
+      "穫": 85262,
+      "🇬🇳": 85263,
+      "鶑": 85264,
+      "Ȱ": 85265,
+      "嘐": 85266,
+      "🇳🇦": 85267,
+      "👩🏽‍🦲": 85268,
+      "舋": 85269,
+      "阉": 85270,
+      "铈": 85271,
+      "ↄ": 85272,
+      "倹": 85273,
+      "嶦": 85274,
+      "饬": 85275,
+      "暾": 85276,
+      "惿": 85277,
+      "雛": 85278,
+      "➾": 85279,
+      "鯖": 85280,
+      "耜": 85281,
+      "撢": 85282,
+      "🧑🏻‍🎓": 85283,
+      "戧": 85284,
+      "焼": 85285,
+      "👆": 85286,
+      "🌉": 85287,
+      "⑓": 85288,
+      "亙": 85289,
+      "魧": 85290,
+      "洝": 85291,
+      "廬": 85292,
+      "絫": 85293,
+      "🫱🏻": 85294,
+      "菾": 85295,
+      "酇": 85296,
+      "埪": 85297,
+      "墹": 85298,
+      "稆": 85299,
+      "篻": 85300,
+      "畂": 85301,
+      "鑳": 85302,
+      "⊨": 85303,
+      "覸": 85304,
+      "韁": 85305,
+      "鈤": 85306,
+      "👩🏽‍❤️‍💋‍👩🏼": 85307,
+      "鯳": 85308,
+      "瓗": 85309,
+      "馌": 85310,
+      "鴠": 85311,
+      "蝾": 85312,
+      "竻": 85313,
+      "潥": 85314,
+      "👨🏼‍🦰": 85315,
+      "訇": 85316,
+      "騞": 85317,
+      "📱": 85318,
+      "氱": 85319,
+      "🙅🏼‍♂️": 85320,
+      "髯": 85321,
+      "赍": 85322,
+      "玾": 85323,
+      "✍🏼": 85324,
+      "掵": 85325,
+      "橈": 85326,
+      "鞯": 85327,
+      "␩": 85328,
+      "稒": 85329,
+      "覍": 85330,
+      "嘡": 85331,
+      "∅": 85332,
+      "畯": 85333,
+      "昺": 85334,
+      "誈": 85335,
+      "烠": 85336,
+      "蚾": 85337,
+      "弰": 85338,
+      "ɧ": 85339,
+      "麥": 85340,
+      "靵": 85341,
+      "♮": 85342,
+      "⋖": 85343,
+      "彥": 85344,
+      "碫": 85345,
+      "娗": 85346,
+      "煝": 85347,
+      "睴": 85348,
+      "鶂": 85349,
+      "貉": 85350,
+      "愘": 85351,
+      "⒒": 85352,
+      "翙": 85353,
+      "🕟": 85354,
+      "⋃": 85355,
+      "奺": 85356,
+      "馷": 85357,
+      "👨🏻‍🚒": 85358,
+      "笕": 85359,
+      "蘢": 85360,
+      "👨🏽‍❤️‍👨🏽": 85361,
+      "櫍": 85362,
+      "聼": 85363,
+      "驙": 85364,
+      "臄": 85365,
+      "⓷": 85366,
+      "饢": 85367,
+      "報": 85368,
+      "恂": 85369,
+      "蕩": 85370,
+      "蕝": 85371,
+      "∐": 85372,
+      "啈": 85373,
+      "垧": 85374,
+      "╇": 85375,
+      "⛪": 85376,
+      "▟": 85377,
+      "仞": 85378,
+      "珖": 85379,
+      "嗲": 85380,
+      "鬅": 85381,
+      "蘓": 85382,
+      "贊": 85383,
+      "👨🏿‍🏫": 85384,
+      "雦": 85385,
+      "鮖": 85386,
+      "笻": 85387,
+      "鴡": 85388,
+      "Ȕ": 85389,
+      "疷": 85390,
+      "🔸": 85391,
+      "曅": 85392,
+      "瘮": 85393,
+      "宔": 85394,
+      "醺": 85395,
+      "勄": 85396,
+      "🏃🏾": 85397,
+      "縼": 85398,
+      "Ƙ": 85399,
+      "🩶": 85400,
+      "🚌": 85401,
+      "閏": 85402,
+      "塆": 85403,
+      "耝": 85404,
+      "碭": 85405,
+      "類": 85406,
+      "杊": 85407,
+      "⋵": 85408,
+      "鶧": 85409,
+      "👩🏻‍🦯": 85410,
+      "🇹🇭": 85411,
+      "啔": 85412,
+      "搷": 85413,
+      "瘇": 85414,
+      "囌": 85415,
+      "懅": 85416,
+      "⒬": 85417,
+      "🏖": 85418,
+      "槂": 85419,
+      "湷": 85420,
+      "Ⅼ": 85421,
+      "錯": 85422,
+      "撺": 85423,
+      "鉐": 85424,
+      "夒": 85425,
+      "⚆": 85426,
+      "癎": 85427,
+      "🇿🇼": 85428,
+      "荄": 85429,
+      "誄": 85430,
+      "嘮": 85431,
+      "獐": 85432,
+      "摛": 85433,
+      "旆": 85434,
+      "鏊": 85435,
+      "🧘🏿‍♂️": 85436,
+      "鉊": 85437,
+      "潁": 85438,
+      "搊": 85439,
+      "☰": 85440,
+      "✍🏻": 85441,
+      "😅": 85442,
+      "鄒": 85443,
+      "🇪🇹": 85444,
+      "籈": 85445,
+      "鈪": 85446,
+      "躗": 85447,
+      "璡": 85448,
+      "踉": 85449,
+      "΄": 85450,
+      "駽": 85451,
+      "🎅": 85452,
+      "墏": 85453,
+      "嗡": 85454,
+      "縪": 85455,
+      "🏋🏿‍♂️": 85456,
+      "👌🏿": 85457,
+      "偬": 85458,
+      "蹷": 85459,
+      "┚": 85460,
+      "覲": 85461,
+      "糣": 85462,
+      "垺": 85463,
+      "暏": 85464,
+      "腴": 85465,
+      "🆒": 85466,
+      "鏳": 85467,
+      "璓": 85468,
+      "懝": 85469,
+      "鑤": 85470,
+      "👼🏻": 85471,
+      "叿": 85472,
+      "瓋": 85473,
+      "檈": 85474,
+      "👨🏽‍🏫": 85475,
+      "Ⅸ": 85476,
+      "⏖": 85477,
+      "霮": 85478,
+      "熐": 85479,
+      "℘": 85480,
+      "櫠": 85481,
+      "煆": 85482,
+      "蔋": 85483,
+      "‣": 85484,
+      "蝫": 85485,
+      "🙍🏾‍♀️": 85486,
+      "🏊🏽": 85487,
+      "乺": 85488,
+      "洍": 85489,
+      "瞮": 85490,
+      "墊": 85491,
+      "💏🏿": 85492,
+      "Ȗ": 85493,
+      "幑": 85494,
+      "迣": 85495,
+      "🤹‍♀️": 85496,
+      "庍": 85497,
+      "忰": 85498,
+      "🤵🏼‍♂️": 85499,
+      "🕵🏾‍♂️": 85500,
+      "髑": 85501,
+      "焢": 85502,
+      "鯓": 85503,
+      "襑": 85504,
+      "鑽": 85505,
+      "塇": 85506,
+      "💇‍♀️": 85507,
+      "耬": 85508,
+      "🤟🏽": 85509,
+      "佪": 85510,
+      "🖍": 85511,
+      "虯": 85512,
+      "枂": 85513,
+      "Ｈ": 85514,
+      "␰": 85515,
+      "ↅ": 85516,
+      "傪": 85517,
+      "罳": 85518,
+      "閟": 85519,
+      "🧛🏻‍♂️": 85520,
+      "袰": 85521,
+      "犺": 85522,
+      "陔": 85523,
+      "願": 85524,
+      "茿": 85525,
+      "␈": 85526,
+      "蝁": 85527,
+      "缃": 85528,
+      "蕖": 85529,
+      "扸": 85530,
+      "徳": 85531,
+      "🧖🏻": 85532,
+      "💘": 85533,
+      "釁": 85534,
+      "兊": 85535,
+      "脜": 85536,
+      "怭": 85537,
+      "疶": 85538,
+      "匵": 85539,
+      "揖": 85540,
+      "🍷": 85541,
+      "⌛": 85542,
+      "僌": 85543,
+      "😻": 85544,
+      "趩": 85545,
+      "⑳": 85546,
+      "焱": 85547,
+      "̷": 85548,
+      "覊": 85549,
+      "潐": 85550,
+      "螕": 85551,
+      "竑": 85552,
+      "倽": 85553,
+      "韕": 85554,
+      "醀": 85555,
+      "珅": 85556,
+      "嗮": 85557,
+      "恻": 85558,
+      "蹀": 85559,
+      "翣": 85560,
+      "籬": 85561,
+      "襂": 85562,
+      "撆": 85563,
+      "繳": 85564,
+      "💾": 85565,
+      "忛": 85566,
+      "溦": 85567,
+      "ʚ": 85568,
+      "碈": 85569,
+      "🏌🏽": 85570,
+      "膟": 85571,
+      "🍜": 85572,
+      "賡": 85573,
+      "雸": 85574,
+      "👨🏿‍🦱": 85575,
+      "鐹": 85576,
+      "罿": 85577,
+      "痡": 85578,
+      "扜": 85579,
+      "沜": 85580,
+      "渳": 85581,
+      "鴄": 85582,
+      "鵯": 85583,
+      "婋": 85584,
+      "榥": 85585,
+      "袲": 85586,
+      "豷": 85587,
+      "⒰": 85588,
+      "┏": 85589,
+      "👬🏼": 85590,
+      "賚": 85591,
+      "靾": 85592,
+      "梤": 85593,
+      "祶": 85594,
+      "ǀ": 85595,
+      "🧑": 85596,
+      "瑢": 85597,
+      "💁🏽‍♂️": 85598,
+      "👩": 85599,
+      "縀": 85600,
+      "呏": 85601,
+      "掛": 85602,
+      "💊": 85603,
+      "🇬🇩": 85604,
+      "蘄": 85605,
+      "朧": 85606,
+      "薭": 85607,
+      "滏": 85608,
+      "圌": 85609,
+      "鷠": 85610,
+      "焺": 85611,
+      "耨": 85612,
+      "雮": 85613,
+      "矾": 85614,
+      "兺": 85615,
+      "撛": 85616,
+      "艒": 85617,
+      "佌": 85618,
+      "靐": 85619,
+      "👮🏿": 85620,
+      "笳": 85621,
+      "燿": 85622,
+      "陽": 85623,
+      "磵": 85624,
+      "侺": 85625,
+      "櫔": 85626,
+      "屢": 85627,
+      "橮": 85628,
+      "旪": 85629,
+      "👨🏽‍❤️‍👨🏻": 85630,
+      "僈": 85631,
+      "蒢": 85632,
+      "謩": 85633,
+      "嵚": 85634,
+      "歬": 85635,
+      "濉": 85636,
+      "呃": 85637,
+      "₍": 85638,
+      "劋": 85639,
+      "丟": 85640,
+      "榪": 85641,
+      "Ò": 85642,
+      "嗏": 85643,
+      "啙": 85644,
+      "🐦": 85645,
+      "錘": 85646,
+      "郥": 85647,
+      "澴": 85648,
+      "亞": 85649,
+      "愡": 85650,
+      "˨": 85651,
+      "锨": 85652,
+      "脛": 85653,
+      "粙": 85654,
+      "👨🏽": 85655,
+      "鴾": 85656,
+      "謕": 85657,
+      "緿": 85658,
+      "≬": 85659,
+      "🧖‍♀️": 85660,
+      "喎": 85661,
+      "讬": 85662,
+      "務": 85663,
+      "誧": 85664,
+      "祒": 85665,
+      "湁": 85666,
+      "🧑🏽‍🔧": 85667,
+      "軖": 85668,
+      "ˢ": 85669,
+      "🌵": 85670,
+      "🇫🇷": 85671,
+      "⍯": 85672,
+      "苄": 85673,
+      "誩": 85674,
+      "婅": 85675,
+      "叧": 85676,
+      "̌": 85677,
+      "膞": 85678,
+      "軏": 85679,
+      "纨": 85680,
+      "訖": 85681,
+      "̏": 85682,
+      "鷶": 85683,
+      "靪": 85684,
+      "🦵": 85685,
+      "窤": 85686,
+      "戠": 85687,
+      "鑈": 85688,
+      "曍": 85689,
+      "👨🏻‍✈️": 85690,
+      "塝": 85691,
+      "👎🏼": 85692,
+      "館": 85693,
+      "⚶": 85694,
+      "渑": 85695,
+      "臈": 85696,
+      "鳳": 85697,
+      "緝": 85698,
+      "昑": 85699,
+      "繥": 85700,
+      "踱": 85701,
+      "🪡": 85702,
+      "👩‍🏭": 85703,
+      "崅": 85704,
+      "🪒": 85705,
+      "漒": 85706,
+      "棭": 85707,
+      "坕": 85708,
+      "␄": 85709,
+      "舁": 85710,
+      "💑🏾": 85711,
+      "蹑": 85712,
+      "倰": 85713,
+      "💇🏾‍♀️": 85714,
+      "狋": 85715,
+      "殨": 85716,
+      "夥": 85717,
+      "坰": 85718,
+      "╤": 85719,
+      "🙅🏻": 85720,
+      "蘠": 85721,
+      "∔": 85722,
+      "犽": 85723,
+      "儔": 85724,
+      "鑎": 85725,
+      "Ȍ": 85726,
+      "斠": 85727,
+      "厧": 85728,
+      "躚": 85729,
+      "鍉": 85730,
+      "📺": 85731,
+      "縤": 85732,
+      "錚": 85733,
+      "軐": 85734,
+      "毌": 85735,
+      "跤": 85736,
+      "🦹🏻‍♂️": 85737,
+      "鹣": 85738,
+      "驢": 85739,
+      "紲": 85740,
+      "畇": 85741,
+      "齯": 85742,
+      "鰇": 85743,
+      "勰": 85744,
+      "僒": 85745,
+      "硠": 85746,
+      "薽": 85747,
+      "ƫ": 85748,
+      "郿": 85749,
+      "踥": 85750,
+      "呑": 85751,
+      "≟": 85752,
+      "🕓": 85753,
+      "盹": 85754,
+      "袒": 85755,
+      "聝": 85756,
+      "⋇": 85757,
+      "眦": 85758,
+      "鵘": 85759,
+      "⛑": 85760,
+      "🤹🏾": 85761,
+      "鲑": 85762,
+      "澙": 85763,
+      "昿": 85764,
+      "胅": 85765,
+      "鯚": 85766,
+      "襤": 85767,
+      "娛": 85768,
+      "滵": 85769,
+      "臮": 85770,
+      "榎": 85771,
+      "鉥": 85772,
+      "讔": 85773,
+      "跘": 85774,
+      "剶": 85775,
+      "雉": 85776,
+      "∏": 85777,
+      "幞": 85778,
+      "ϖ": 85779,
+      "驘": 85780,
+      "劉": 85781,
+      "玝": 85782,
+      "🍨": 85783,
+      "Ｋ": 85784,
+      "鴯": 85785,
+      "瀜": 85786,
+      "騈": 85787,
+      "頹": 85788,
+      "嵂": 85789,
+      "潽": 85790,
+      "厪": 85791,
+      "≘": 85792,
+      "鹫": 85793,
+      "韈": 85794,
+      "涬": 85795,
+      "竂": 85796,
+      "蒼": 85797,
+      "🏋🏼‍♀️": 85798,
+      "啒": 85799,
+      "卥": 85800,
+      "ⓛ": 85801,
+      "🤷": 85802,
+      "翚": 85803,
+      "閶": 85804,
+      "盇": 85805,
+      "咾": 85806,
+      "镠": 85807,
+      "跶": 85808,
+      "🦶🏾": 85809,
+      "殂": 85810,
+      "👩🏽‍❤️‍👨🏻": 85811,
+      "灚": 85812,
+      "耾": 85813,
+      "⚀": 85814,
+      "謿": 85815,
+      "郉": 85816,
+      "蠃": 85817,
+      "🦸🏽": 85818,
+      "☜": 85819,
+      "🧏🏿‍♂️": 85820,
+      "桞": 85821,
+      "懐": 85822,
+      "譴": 85823,
+      "烶": 85824,
+      "酙": 85825,
+      "璑": 85826,
+      "〉": 85827,
+      "髱": 85828,
+      "隝": 85829,
+      "鶵": 85830,
+      "鮵": 85831,
+      "畊": 85832,
+      "頦": 85833,
+      "紅": 85834,
+      "⛋": 85835,
+      "餾": 85836,
+      "卆": 85837,
+      "甝": 85838,
+      "🆑": 85839,
+      "厣": 85840,
+      "魫": 85841,
+      "踺": 85842,
+      "靇": 85843,
+      "🐄": 85844,
+      "瓎": 85845,
+      "爨": 85846,
+      "麗": 85847,
+      "👩🏾‍🦱": 85848,
+      "⎇": 85849,
+      "荃": 85850,
+      "馺": 85851,
+      "虁": 85852,
+      "樕": 85853,
+      "挍": 85854,
+      "弙": 85855,
+      "掴": 85856,
+      "樧": 85857,
+      "🤷🏿": 85858,
+      "虴": 85859,
+      "縸": 85860,
+      "秡": 85861,
+      "蟀": 85862,
+      "⚧": 85863,
+      "Ə": 85864,
+      "⇤": 85865,
+      "遞": 85866,
+      "鍡": 85867,
+      "郼": 85868,
+      "₉": 85869,
+      "鴟": 85870,
+      "鷽": 85871,
+      "🔚": 85872,
+      "崟": 85873,
+      "瞣": 85874,
+      "橭": 85875,
+      "魵": 85876,
+      "糠": 85877,
+      "羢": 85878,
+      "猚": 85879,
+      "🧛‍♀️": 85880,
+      "眝": 85881,
+      "鋳": 85882,
+      "嗀": 85883,
+      "厍": 85884,
+      "🫁": 85885,
+      "垤": 85886,
+      "侰": 85887,
+      "💁🏾‍♀️": 85888,
+      "輾": 85889,
+      "飷": 85890,
+      "孁": 85891,
+      "歜": 85892,
+      "裬": 85893,
+      "萆": 85894,
+      "銑": 85895,
+      "椨": 85896,
+      "砼": 85897,
+      "Ď": 85898,
+      "暱": 85899,
+      "堊": 85900,
+      "Ŋ": 85901,
+      "♸": 85902,
+      "👴🏿": 85903,
+      "辶": 85904,
+      "萪": 85905,
+      "钁": 85906,
+      "掗": 85907,
+      "樻": 85908,
+      "譾": 85909,
+      "Ⅻ": 85910,
+      "魍": 85911,
+      "酽": 85912,
+      "鱙": 85913,
+      "鋘": 85914,
+      "轤": 85915,
+      "淴": 85916,
+      "🧬": 85917,
+      "萛": 85918,
+      "遺": 85919,
+      "薞": 85920,
+      "螤": 85921,
+      "Ϙ": 85922,
+      "姱": 85923,
+      "昚": 85924,
+      "🐺": 85925,
+      "捚": 85926,
+      "袡": 85927,
+      "讎": 85928,
+      "睘": 85929,
+      "篹": 85930,
+      "💸": 85931,
+      "薐": 85932,
+      "👨🏼‍🦱": 85933,
+      "湻": 85934,
+      "槉": 85935,
+      "菒": 85936,
+      "鐜": 85937,
+      "葁": 85938,
+      "厱": 85939,
+      "粷": 85940,
+      "玌": 85941,
+      "潎": 85942,
+      "🧜🏽": 85943,
+      "瘿": 85944,
+      "仸": 85945,
+      "🔘": 85946,
+      "騷": 85947,
+      "閲": 85948,
+      "鳤": 85949,
+      "坌": 85950,
+      "铽": 85951,
+      "⋯": 85952,
+      "妢": 85953,
+      "嗼": 85954,
+      "暽": 85955,
+      "👦🏼": 85956,
+      "1️⃣": 85957,
+      "繯": 85958,
+      "醊": 85959,
+      "ₒ": 85960,
+      "趷": 85961,
+      "捺": 85962,
+      "羿": 85963,
+      "匥": 85964,
+      "眾": 85965,
+      "鏶": 85966,
+      "歙": 85967,
+      "嶜": 85968,
+      "柧": 85969,
+      "煯": 85970,
+      "牝": 85971,
+      "👨🏽‍🦲": 85972,
+      "隹": 85973,
+      "朙": 85974,
+      "啫": 85975,
+      "繾": 85976,
+      "👳🏽‍♀️": 85977,
+      "甿": 85978,
+      "🧑🏾‍✈️": 85979,
+      "燬": 85980,
+      "蓥": 85981,
+      "竘": 85982,
+      "嵸": 85983,
+      "笝": 85984,
+      "蕻": 85985,
+      "糐": 85986,
+      "邌": 85987,
+      "⏥": 85988,
+      "蓞": 85989,
+      "🧑🏻‍❤️‍🧑🏾": 85990,
+      "🏳️‍⚧️": 85991,
+      "饟": 85992,
+      "婁": 85993,
+      "灆": 85994,
+      "簉": 85995,
+      "褲": 85996,
+      "👨🏻‍💼": 85997,
+      "🤹‍♂️": 85998,
+      "菂": 85999,
+      "瞴": 86000,
+      "紓": 86001,
+      "麖": 86002,
+      "珟": 86003,
+      "🥚": 86004,
+      "Ƒ": 86005,
+      "崝": 86006,
+      "辒": 86007,
+      "靀": 86008,
+      "牄": 86009,
+      "癒": 86010,
+      "溑": 86011,
+      "臠": 86012,
+      "鶄": 86013,
+      "煻": 86014,
+      "舳": 86015,
+      "翉": 86016,
+      "👩🏾‍🔬": 86017,
+      "湢": 86018,
+      "姹": 86019,
+      "樅": 86020,
+      "⅛": 86021,
+      "齮": 86022,
+      "鮒": 86023,
+      "鰣": 86024,
+      "賙": 86025,
+      "玑": 86026,
+      "鏚": 86027,
+      "傥": 86028,
+      "Ώ": 86029,
+      "砝": 86030,
+      "増": 86031,
+      "噸": 86032,
+      "楊": 86033,
+      "ℋ": 86034,
+      "麺": 86035,
+      "鴊": 86036,
+      "栁": 86037,
+      "餑": 86038,
+      "恑": 86039,
+      "裀": 86040,
+      "膱": 86041,
+      "Ｔ": 86042,
+      "鷮": 86043,
+      "🩻": 86044,
+      "🐮": 86045,
+      "瀱": 86046,
+      "麹": 86047,
+      "繆": 86048,
+      "ʲ": 86049,
+      "珳": 86050,
+      "🦹🏿": 86051,
+      "ͤ": 86052,
+      "艎": 86053,
+      "👩🏽‍❤️‍💋‍👨🏾": 86054,
+      "🌞": 86055,
+      "藦": 86056,
+      "跄": 86057,
+      "鎨": 86058,
+      "皴": 86059,
+      "熂": 86060,
+      "艀": 86061,
+      "鉸": 86062,
+      "灐": 86063,
+      "襈": 86064,
+      "軘": 86065,
+      "菙": 86066,
+      "郪": 86067,
+      "艌": 86068,
+      "陠": 86069,
+      "袺": 86070,
+      "鉪": 86071,
+      "鞛": 86072,
+      "◠": 86073,
+      "魣": 86074,
+      "衦": 86075,
+      "谖": 86076,
+      "儙": 86077,
+      "繟": 86078,
+      "疨": 86079,
+      "👩🏽‍🏫": 86080,
+      "憝": 86081,
+      "嚇": 86082,
+      "睜": 86083,
+      "睤": 86084,
+      "耏": 86085,
+      "髐": 86086,
+      "鋜": 86087,
+      "ₓ": 86088,
+      "賿": 86089,
+      "凃": 86090,
+      "🙅‍♂️": 86091,
+      "雝": 86092,
+      "⇪": 86093,
+      "🧏🏽‍♀️": 86094,
+      "⑩": 86095,
+      "🛄": 86096,
+      "彏": 86097,
+      "🏂🏿": 86098,
+      "棨": 86099,
+      "膅": 86100,
+      "瘢": 86101,
+      "嬬": 86102,
+      "🖌": 86103,
+      "⚅": 86104,
+      "🧕🏼": 86105,
+      "诐": 86106,
+      "蚴": 86107,
+      "淵": 86108,
+      "爾": 86109,
+      "蘵": 86110,
+      "佃": 86111,
+      "嵾": 86112,
+      "箝": 86113,
+      "錥": 86114,
+      "蛈": 86115,
+      "犓": 86116,
+      "賞": 86117,
+      "匏": 86118,
+      "掦": 86119,
+      "谿": 86120,
+      "廘": 86121,
+      "鄠": 86122,
+      "酧": 86123,
+      "剳": 86124,
+      "朎": 86125,
+      "逦": 86126,
+      "👨‍👨‍👦‍👦": 86127,
+      "庲": 86128,
+      "嫅": 86129,
+      "緷": 86130,
+      "≆": 86131,
+      "癥": 86132,
+      "🇲🇴": 86133,
+      "仨": 86134,
+      "馁": 86135,
+      "忷": 86136,
+      "梍": 86137,
+      "ʘ": 86138,
+      "翝": 86139,
+      "輩": 86140,
+      "珮": 86141,
+      "铖": 86142,
+      "驈": 86143,
+      "邔": 86144,
+      "駠": 86145,
+      "粐": 86146,
+      "貖": 86147,
+      "籠": 86148,
+      "餂": 86149,
+      "蠄": 86150,
+      "缬": 86151,
+      "緫": 86152,
+      "牎": 86153,
+      "疴": 86154,
+      "覇": 86155,
+      "⇠": 86156,
+      "藷": 86157,
+      "👨🏻": 86158,
+      "蠼": 86159,
+      "簑": 86160,
+      "苆": 86161,
+      "僃": 86162,
+      "湴": 86163,
+      "ǩ": 86164,
+      "🤟🏿": 86165,
+      "繊": 86166,
+      "烮": 86167,
+      "袹": 86168,
+      "厬": 86169,
+      "訆": 86170,
+      "寘": 86171,
+      "忶": 86172,
+      "🪄": 86173,
+      "ƙ": 86174,
+      "🇨🇭": 86175,
+      "僞": 86176,
+      "🦹🏼‍♂️": 86177,
+      "懃": 86178,
+      "罬": 86179,
+      "ŧ": 86180,
+      "澋": 86181,
+      "蚳": 86182,
+      "忚": 86183,
+      "靼": 86184,
+      "芎": 86185,
+      "專": 86186,
+      "噊": 86187,
+      "吿": 86188,
+      "婻": 86189,
+      "舘": 86190,
+      "⏁": 86191,
+      "虷": 86192,
+      "╀": 86193,
+      "唚": 86194,
+      "頕": 86195,
+      "衱": 86196,
+      "谤": 86197,
+      "🧑🏾‍⚕️": 86198,
+      "胒": 86199,
+      "鞦": 86200,
+      "祏": 86201,
+      "浼": 86202,
+      "呥": 86203,
+      "豵": 86204,
+      "菷": 86205,
+      "蝥": 86206,
+      "琚": 86207,
+      "🕣": 86208,
+      "⃐": 86209,
+      "⎓": 86210,
+      "赀": 86211,
+      "婞": 86212,
+      "🤾🏻": 86213,
+      "鳖": 86214,
+      "璴": 86215,
+      "脭": 86216,
+      "筏": 86217,
+      "🚶🏿‍♀️": 86218,
+      "ⓞ": 86219,
+      "👉": 86220,
+      "旘": 86221,
+      "戡": 86222,
+      "牪": 86223,
+      "♽": 86224,
+      "酩": 86225,
+      "仐": 86226,
+      "⏮": 86227,
+      "俆": 86228,
+      "荛": 86229,
+      "逋": 86230,
+      "怱": 86231,
+      "査": 86232,
+      "🪹": 86233,
+      "琓": 86234,
+      "矂": 86235,
+      "衜": 86236,
+      "膩": 86237,
+      "🐷": 86238,
+      "燄": 86239,
+      "慡": 86240,
+      "濔": 86241,
+      "☫": 86242,
+      "厾": 86243,
+      "氽": 86244,
+      "邛": 86245,
+      "雔": 86246,
+      "ť": 86247,
+      "燽": 86248,
+      "軕": 86249,
+      "鷵": 86250,
+      "繤": 86251,
+      "巟": 86252,
+      "Ŧ": 86253,
+      "": 86254,
+      "碆": 86255,
+      "焆": 86256,
+      "⃊": 86257,
+      "👨🏾‍🦯": 86258,
+      "趕": 86259,
+      "钴": 86260,
+      "齁": 86261,
+      "韽": 86262,
+      "颬": 86263,
+      "💇🏽‍♀️": 86264,
+      "鎢": 86265,
+      "🐾": 86266,
+      "🇹🇯": 86267,
+      "賳": 86268,
+      "↼": 86269,
+      "鍵": 86270,
+      "晥": 86271,
+      "倆": 86272,
+      "餣": 86273,
+      "谼": 86274,
+      "偱": 86275,
+      "👎": 86276,
+      "渼": 86277,
+      "̐": 86278,
+      "諫": 86279,
+      "饡": 86280,
+      "囁": 86281,
+      "忀": 86282,
+      "姝": 86283,
+      "☣": 86284,
+      "笪": 86285,
+      "鳝": 86286,
+      "險": 86287,
+      "鑿": 86288,
+      "♳": 86289,
+      "涸": 86290,
+      "恚": 86291,
+      "鎆": 86292,
+      "⚙": 86293,
+      "🌐": 86294,
+      "喁": 86295,
+      "諬": 86296,
+      "峾": 86297,
+      "娙": 86298,
+      "渽": 86299,
+      "蜓": 86300,
+      "🔰": 86301,
+      "挳": 86302,
+      "閑": 86303,
+      "坙": 86304,
+      "峑": 86305,
+      "錋": 86306,
+      "😊": 86307,
+      "輥": 86308,
+      "蔜": 86309,
+      "┥": 86310,
+      "峐": 86311,
+      "櫭": 86312,
+      "潆": 86313,
+      "葏": 86314,
+      "艖": 86315,
+      "⌜": 86316,
+      "瑐": 86317,
+      "泱": 86318,
+      "玓": 86319,
+      "鋫": 86320,
+      "彔": 86321,
+      "椊": 86322,
+      "竀": 86323,
+      "桏": 86324,
+      "鋈": 86325,
+      "🌌": 86326,
+      "柤": 86327,
+      "備": 86328,
+      "樖": 86329,
+      "靋": 86330,
+      "錵": 86331,
+      "🤏": 86332,
+      "熛": 86333,
+      "鹶": 86334,
+      "🧣": 86335,
+      "椳": 86336,
+      "杓": 86337,
+      "⒠": 86338,
+      "慮": 86339,
+      "👱🏻‍♀️": 86340,
+      "撨": 86341,
+      "綺": 86342,
+      "🧚🏽": 86343,
+      "摠": 86344,
+      "耄": 86345,
+      "僼": 86346,
+      "👱": 86347,
+      "糢": 86348,
+      "簂": 86349,
+      "霱": 86350,
+      "妝": 86351,
+      "🧑🏿‍❤️‍💋‍🧑🏽": 86352,
+      "軲": 86353,
+      "🇫🇰": 86354,
+      "讕": 86355,
+      "ⓠ": 86356,
+      "蠻": 86357,
+      "圛": 86358,
+      "叆": 86359,
+      "🎲": 86360,
+      "褎": 86361,
+      "焤": 86362,
+      "🪁": 86363,
+      "黈": 86364,
+      "鱑": 86365,
+      "枅": 86366,
+      "ŕ": 86367,
+      "遬": 86368,
+      "蛣": 86369,
+      "睼": 86370,
+      "璮": 86371,
+      "潗": 86372,
+      "锘": 86373,
+      "垥": 86374,
+      "姽": 86375,
+      "毖": 86376,
+      "≢": 86377,
+      "磌": 86378,
+      "🏌🏻‍♀️": 86379,
+      "幫": 86380,
+      "豠": 86381,
+      "睝": 86382,
+      "霬": 86383,
+      "␤": 86384,
+      "🧖🏼‍♀️": 86385,
+      "輋": 86386,
+      "峘": 86387,
+      "崸": 86388,
+      "镢": 86389,
+      "嬌": 86390,
+      "👨🏻‍❤️‍👨🏼": 86391,
+      "🧑‍🏭": 86392,
+      "🇷🇴": 86393,
+      "莥": 86394,
+      "嚕": 86395,
+      "慣": 86396,
+      "┠": 86397,
+      "蒞": 86398,
+      "隓": 86399,
+      "傉": 86400,
+      "🧓🏻": 86401,
+      "憉": 86402,
+      "谵": 86403,
+      "匒": 86404,
+      "溍": 86405,
+      "痜": 86406,
+      "焴": 86407,
+      "❓": 86408,
+      "🎓": 86409,
+      "🪜": 86410,
+      "丕": 86411,
+      "葹": 86412,
+      "員": 86413,
+      "🤲🏽": 86414,
+      "🤦🏻": 86415,
+      "岇": 86416,
+      "艄": 86417,
+      "槕": 86418,
+      "極": 86419,
+      "🫳": 86420,
+      "漣": 86421,
+      "👨🏿‍❤️‍👨🏽": 86422,
+      "🦠": 86423,
+      "狧": 86424,
+      "緔": 86425,
+      "ʦ": 86426,
+      "🙄": 86427,
+      "🌺": 86428,
+      "垀": 86429,
+      "鯁": 86430,
+      "檫": 86431,
+      "💂🏼": 86432,
+      "钋": 86433,
+      "✊🏻": 86434,
+      "贈": 86435,
+      "裒": 86436,
+      "嶌": 86437,
+      "👨🏽‍❤️‍💋‍👨🏽": 86438,
+      "槵": 86439,
+      "🧖🏻‍♀️": 86440,
+      "熸": 86441,
+      "馇": 86442,
+      "💆‍♂️": 86443,
+      "輐": 86444,
+      "顅": 86445,
+      "嘈": 86446,
+      "総": 86447,
+      "恇": 86448,
+      "礸": 86449,
+      "贙": 86450,
+      "鯻": 86451,
+      "疽": 86452,
+      "秢": 86453,
+      "🇪🇭": 86454,
+      "瞽": 86455,
+      "🇪🇪": 86456,
+      "鐵": 86457,
+      "脨": 86458,
+      "遽": 86459,
+      "桸": 86460,
+      "戥": 86461,
+      "噼": 86462,
+      "缻": 86463,
+      "謏": 86464,
+      "枍": 86465,
+      "蔟": 86466,
+      "簺": 86467,
+      "碤": 86468,
+      "╲": 86469,
+      "髸": 86470,
+      "孫": 86471,
+      "⋎": 86472,
+      "👩🏿‍❤️‍💋‍👨🏻": 86473,
+      "韌": 86474,
+      "硓": 86475,
+      "蹐": 86476,
+      "唘": 86477,
+      "🧖🏽": 86478,
+      "媣": 86479,
+      "Ⓚ": 86480,
+      "兲": 86481,
+      "₶": 86482,
+      "櫅": 86483,
+      "🇭🇳": 86484,
+      "🪟": 86485,
+      "楗": 86486,
+      "睊": 86487,
+      "👩‍👧": 86488,
+      "讜": 86489,
+      "∗": 86490,
+      "敯": 86491,
+      "榡": 86492,
+      "詍": 86493,
+      "😨": 86494,
+      "衭": 86495,
+      "🧖🏾‍♂️": 86496,
+      "̎": 86497,
+      "歑": 86498,
+      "⎞": 86499,
+      "跗": 86500,
+      "薮": 86501,
+      "鄘": 86502,
+      "轇": 86503,
+      "ⅳ": 86504,
+      "窮": 86505,
+      "聲": 86506,
+      "詀": 86507,
+      "🏌🏼": 86508,
+      "♗": 86509,
+      "椋": 86510,
+      "ₙ": 86511,
+      "😂": 86512,
+      "継": 86513,
+      "ϡ": 86514,
+      "⊁": 86515,
+      "👨🏽‍❤️‍💋‍👨🏻": 86516,
+      "顯": 86517,
+      "鳽": 86518,
+      "踾": 86519,
+      "朿": 86520,
+      "🙋🏿": 86521,
+      "埰": 86522,
+      "摕": 86523,
+      "艭": 86524,
+      "⚢": 86525,
+      "嚎": 86526,
+      "篏": 86527,
+      "朰": 86528,
+      "鯋": 86529,
+      "鴳": 86530,
+      "⚖": 86531,
+      "晧": 86532,
+      "渹": 86533,
+      "🤱🏾": 86534,
+      "髛": 86535,
+      "悫": 86536,
+      "碥": 86537,
+      "炾": 86538,
+      "眈": 86539,
+      "⃨": 86540,
+      "髳": 86541,
+      "緡": 86542,
+      "🦸🏿": 86543,
+      "汱": 86544,
+      "饹": 86545,
+      "檛": 86546,
+      "玹": 86547,
+      "飾": 86548,
+      "埳": 86549,
+      "虲": 86550,
+      "👳🏾‍♂️": 86551,
+      "聻": 86552,
+      "鎷": 86553,
+      "竛": 86554,
+      "鍑": 86555,
+      "尗": 86556,
+      "沎": 86557,
+      "寳": 86558,
+      "撌": 86559,
+      "蜚": 86560,
+      "🙊": 86561,
+      "栵": 86562,
+      "👧🏻": 86563,
+      "顰": 86564,
+      "殱": 86565,
+      "痵": 86566,
+      "閫": 86567,
+      "濤": 86568,
+      "貫": 86569,
+      "伭": 86570,
+      "鞂": 86571,
+      "緰": 86572,
+      "門": 86573,
+      "枌": 86574,
+      "圠": 86575,
+      "踠": 86576,
+      "妀": 86577,
+      "墪": 86578,
+      "👺": 86579,
+      "🏳️‍🌈": 86580,
+      "吀": 86581,
+      "呋": 86582,
+      "₪": 86583,
+      "⁰": 86584,
+      "澐": 86585,
+      "ȍ": 86586,
+      "苶": 86587,
+      "ɩ": 86588,
+      "🥼": 86589,
+      "蓂": 86590,
+      "黸": 86591,
+      "蘶": 86592,
+      "駉": 86593,
+      "喩": 86594,
+      "氒": 86595,
+      "🤸🏿‍♀️": 86596,
+      "兂": 86597,
+      "👩🏿‍⚕️": 86598,
+      "篩": 86599,
+      "紇": 86600,
+      "魙": 86601,
+      "梪": 86602,
+      "篤": 86603,
+      "亹": 86604,
+      "躌": 86605,
+      "皝": 86606,
+      "鄦": 86607,
+      "雂": 86608,
+      "沊": 86609,
+      "籃": 86610,
+      "説": 86611,
+      "鑠": 86612,
+      "🇮🇩": 86613,
+      "挃": 86614,
+      "ķ": 86615,
+      "懩": 86616,
+      "帳": 86617,
+      "👤": 86618,
+      "⊊": 86619,
+      "鄴": 86620,
+      "殄": 86621,
+      "旍": 86622,
+      "嬊": 86623,
+      "墲": 86624,
+      "👧🏿": 86625,
+      "🗳": 86626,
+      "瑣": 86627,
+      "🥊": 86628,
+      "🇲🇲": 86629,
+      "⓸": 86630,
+      "⋬": 86631,
+      "鐲": 86632,
+      "Ȫ": 86633,
+      "🇵🇫": 86634,
+      "巃": 86635,
+      "鎭": 86636,
+      "💍": 86637,
+      "憂": 86638,
+      "顭": 86639,
+      "麞": 86640,
+      "Ľ": 86641,
+      "橖": 86642,
+      "鬇": 86643,
+      "惛": 86644,
+      "戂": 86645,
+      "緑": 86646,
+      "燨": 86647,
+      "👩🏼‍❤️‍👨🏻": 86648,
+      "斨": 86649,
+      "曕": 86650,
+      "滛": 86651,
+      "螶": 86652,
+      "🇦🇽": 86653,
+      "鳨": 86654,
+      "鱽": 86655,
+      "膬": 86656,
+      "傩": 86657,
+      "刌": 86658,
+      "態": 86659,
+      "✋🏽": 86660,
+      "顕": 86661,
+      "摝": 86662,
+      "熧": 86663,
+      "膃": 86664,
+      "胐": 86665,
+      "鰤": 86666,
+      "鷂": 86667,
+      "🧘‍♂️": 86668,
+      "眤": 86669,
+      "⍂": 86670,
+      "鈠": 86671,
+      "铒": 86672,
+      "鷭": 86673,
+      "宱": 86674,
+      "👩🏾‍🎓": 86675,
+      "呺": 86676,
+      "🥰": 86677,
+      "禆": 86678,
+      "棌": 86679,
+      "浃": 86680,
+      "圁": 86681,
+      "踻": 86682,
+      "嘼": 86683,
+      "毰": 86684,
+      "渻": 86685,
+      "🧑🏼‍❤️‍💋‍🧑🏾": 86686,
+      "譀": 86687,
+      "摻": 86688,
+      "慻": 86689,
+      "🏃🏽‍♀️": 86690,
+      "₠": 86691,
+      "✀": 86692,
+      "霈": 86693,
+      "郤": 86694,
+      "唟": 86695,
+      "羇": 86696,
+      "頿": 86697,
+      "彽": 86698,
+      "噅": 86699,
+      "驺": 86700,
+      "腄": 86701,
+      "🤷🏿‍♀️": 86702,
+      "茐": 86703,
+      "篜": 86704,
+      "🐓": 86705,
+      "皞": 86706,
+      "蟝": 86707,
+      "讋": 86708,
+      "穮": 86709,
+      "軵": 86710,
+      "牜": 86711,
+      "🫖": 86712,
+      "龢": 86713,
+      "甒": 86714,
+      "偵": 86715,
+      "⃛": 86716,
+      "咘": 86717,
+      "榁": 86718,
+      "蠗": 86719,
+      "↲": 86720,
+      "鄮": 86721,
+      "🙎": 86722,
+      "鋇": 86723,
+      "昮": 86724,
+      "😰": 86725,
+      "嬺": 86726,
+      "諘": 86727,
+      "⑿": 86728,
+      "檪": 86729,
+      "旲": 86730,
+      "🤾🏿‍♂️": 86731,
+      "餧": 86732,
+      "釱": 86733,
+      "譟": 86734,
+      "袊": 86735,
+      "孴": 86736,
+      "ʿ": 86737,
+      "嵆": 86738,
+      "甔": 86739,
+      "🉑": 86740,
+      "啋": 86741,
+      "詋": 86742,
+      "🧑🏾‍⚖️": 86743,
+      "🐚": 86744,
+      "綤": 86745,
+      "鯫": 86746,
+      "槷": 86747,
+      "跣": 86748,
+      "🧎🏿‍♂️": 86749,
+      "轆": 86750,
+      "暜": 86751,
+      "択": 86752,
+      "莻": 86753,
+      "蕛": 86754,
+      "👨🏼‍❤️‍💋‍👨🏿": 86755,
+      "腏": 86756,
+      "亃": 86757,
+      "嵭": 86758,
+      "💠": 86759,
+      "踘": 86760,
+      "躸": 86761,
+      "馦": 86762,
+      "窺": 86763,
+      "馶": 86764,
+      "◩": 86765,
+      "衉": 86766,
+      "👨🏿‍❤️‍💋‍👨🏾": 86767,
+      "鶈": 86768,
+      "琌": 86769,
+      "靁": 86770,
+      "穁": 86771,
+      "˱": 86772,
+      "🦣": 86773,
+      "珼": 86774,
+      "✄": 86775,
+      "🇧🇶": 86776,
+      "ĕ": 86777,
+      "寱": 86778,
+      "┊": 86779,
+      "燢": 86780,
+      "鱏": 86781,
+      "ℕ": 86782,
+      "拡": 86783,
+      "⎳": 86784,
+      "蚐": 86785,
+      "🧑🏼‍🎄": 86786,
+      "剀": 86787,
+      "抻": 86788,
+      "銼": 86789,
+      "榖": 86790,
+      "皙": 86791,
+      "迩": 86792,
+      "桉": 86793,
+      "徖": 86794,
+      "🦙": 86795,
+      "黗": 86796,
+      "⇘": 86797,
+      "⅟": 86798,
+      "庛": 86799,
+      "枖": 86800,
+      "蛷": 86801,
+      "蠽": 86802,
+      "綻": 86803,
+      "徛": 86804,
+      "礪": 86805,
+      "膗": 86806,
+      "杴": 86807,
+      "闩": 86808,
+      "鏾": 86809,
+      "蒤": 86810,
+      "鱄": 86811,
+      "镫": 86812,
+      "⁩": 86813,
+      "♨": 86814,
+      "畋": 86815,
+      "跩": 86816,
+      "寵": 86817,
+      "痠": 86818,
+      "黫": 86819,
+      "盧": 86820,
+      "嫇": 86821,
+      "壜": 86822,
+      "🌒": 86823,
+      "捒": 86824,
+      "诲": 86825,
+      "荜": 86826,
+      "燘": 86827,
+      "🐰": 86828,
+      "嘧": 86829,
+      "锍": 86830,
+      "摥": 86831,
+      "⋶": 86832,
+      "晈": 86833,
+      "𓀚": 86834,
+      "霴": 86835,
+      "褋": 86836,
+      "鰦": 86837,
+      "發": 86838,
+      "尙": 86839,
+      "∟": 86840,
+      "媬": 86841,
+      "軉": 86842,
+      "歮": 86843,
+      "俴": 86844,
+      "垎": 86845,
+      "嘂": 86846,
+      "🥉": 86847,
+      "邨": 86848,
+      "廩": 86849,
+      "⁉": 86850,
+      "琂": 86851,
+      "泍": 86852,
+      "․": 86853,
+      "霵": 86854,
+      "̓": 86855,
+      "娬": 86856,
+      "墠": 86857,
+      "飁": 86858,
+      "ϭ": 86859,
+      "畢": 86860,
+      "憪": 86861,
+      "🤶🏽": 86862,
+      "鈯": 86863,
+      "猅": 86864,
+      "淩": 86865,
+      "凊": 86866,
+      "礀": 86867,
+      "轹": 86868,
+      "␅": 86869,
+      "圴": 86870,
+      "屮": 86871,
+      "菝": 86872,
+      "劂": 86873,
+      "芰": 86874,
+      "鮤": 86875,
+      "蕒": 86876,
+      "🚡": 86877,
+      "斊": 86878,
+      "肙": 86879,
+      "腦": 86880,
+      "🙇🏼‍♀️": 86881,
+      "郵": 86882,
+      "薳": 86883,
+      "鈵": 86884,
+      "🧜🏼‍♀️": 86885,
+      "⁖": 86886,
+      "蜝": 86887,
+      "🏋🏼‍♂️": 86888,
+      "誋": 86889,
+      "鲛": 86890,
+      "👩🏽‍❤️‍👩🏾": 86891,
+      "魎": 86892,
+      "窔": 86893,
+      "澺": 86894,
+      "懓": 86895,
+      "🧍‍♀️": 86896,
+      "坢": 86897,
+      "倿": 86898,
+      "🤹🏾‍♀️": 86899,
+      "👂🏻": 86900,
+      "繠": 86901,
+      "ℸ": 86902,
+      "🧑🏿‍🦯": 86903,
+      "🥫": 86904,
+      "蝑": 86905,
+      "🤹🏼‍♀️": 86906,
+      "菀": 86907,
+      "誯": 86908,
+      "缊": 86909,
+      "眮": 86910,
+      "槏": 86911,
+      "繺": 86912,
+      "👩🏿‍❤️‍👨🏼": 86913,
+      "蔢": 86914,
+      "檝": 86915,
+      "缡": 86916,
+      "鼜": 86917,
+      "🧑🏽‍🎤": 86918,
+      "霙": 86919,
+      "淔": 86920,
+      "瞲": 86921,
+      "詞": 86922,
+      "蘝": 86923,
+      "鬢": 86924,
+      "˯": 86925,
+      "羋": 86926,
+      "項": 86927,
+      "👨🏽‍✈️": 86928,
+      "奒": 86929,
+      "👩🏼": 86930,
+      "🤵‍♀️": 86931,
+      "苕": 86932,
+      "♥": 86933,
+      "穭": 86934,
+      "鱸": 86935,
+      "🍍": 86936,
+      "袬": 86937,
+      "臨": 86938,
+      "👆🏾": 86939,
+      "🦸🏿‍♀️": 86940,
+      "駅": 86941,
+      "Dz": 86942,
+      "羆": 86943,
+      "抦": 86944,
+      "礖": 86945,
+      "瞚": 86946,
+      "簮": 86947,
+      "踕": 86948,
+      "曋": 86949,
+      "豻": 86950,
+      "媉": 86951,
+      "🧴": 86952,
+      "嚂": 86953,
+      "讞": 86954,
+      "偧": 86955,
+      "👆🏻": 86956,
+      "築": 86957,
+      "爥": 86958,
+      "儇": 86959,
+      "斲": 86960,
+      "酊": 86961,
+      "‫": 86962,
+      "趖": 86963,
+      "納": 86964,
+      "伣": 86965,
+      "鰄": 86966,
+      "蟁": 86967,
+      "旔": 86968,
+      "倀": 86969,
+      "虙": 86970,
+      "鈊": 86971,
+      "劺": 86972,
+      "鋢": 86973,
+      "濌": 86974,
+      "嬂": 86975,
+      "ǫ": 86976,
+      "鏘": 86977,
+      "畃": 86978,
+      "遷": 86979,
+      "欎": 86980,
+      "駛": 86981,
+      "獣": 86982,
+      "郺": 86983,
+      "阯": 86984,
+      "͍": 86985,
+      "🫦": 86986,
+      "甗": 86987,
+      "嵵": 86988,
+      "馬": 86989,
+      "萫": 86990,
+      "ǹ": 86991,
+      "忋": 86992,
+      "篊": 86993,
+      "🏄🏽‍♂️": 86994,
+      "柆": 86995,
+      "柊": 86996,
+      "🇰🇪": 86997,
+      "😮‍💨": 86998,
+      "蕘": 86999,
+      "⛎": 87000,
+      "淧": 87001,
+      "悗": 87002,
+      "獻": 87003,
+      "踽": 87004,
+      "✋🏾": 87005,
+      "褖": 87006,
+      "蠺": 87007,
+      "贖": 87008,
+      "🤽🏿‍♂️": 87009,
+      "🌭": 87010,
+      "匛": 87011,
+      "涄": 87012,
+      "咈": 87013,
+      "珥": 87014,
+      "𓀇": 87015,
+      "繙": 87016,
+      "ʹ": 87017,
+      "懻": 87018,
+      "鼢": 87019,
+      "嵢": 87020,
+      "婰": 87021,
+      "熞": 87022,
+      "鍁": 87023,
+      "硌": 87024,
+      "Ｎ": 87025,
+      "狁": 87026,
+      "穢": 87027,
+      "彺": 87028,
+      "莵": 87029,
+      "笱": 87030,
+      "谫": 87031,
+      "沇": 87032,
+      "峟": 87033,
+      "嬫": 87034,
+      "濷": 87035,
+      "┬": 87036,
+      "鱓": 87037,
+      "戵": 87038,
+      "👩🏽‍❤️‍💋‍👨🏼": 87039,
+      "宨": 87040,
+      "蘽": 87041,
+      "🇰🇼": 87042,
+      "侒": 87043,
+      "杌": 87044,
+      "敹": 87045,
+      "僤": 87046,
+      "曨": 87047,
+      "燛": 87048,
+      "鷱": 87049,
+      "🫵🏿": 87050,
+      "纚": 87051,
+      "殪": 87052,
+      "莏": 87053,
+      "淝": 87054,
+      "褢": 87055,
+      "┞": 87056,
+      "棯": 87057,
+      "嬷": 87058,
+      "鞇": 87059,
+      "欽": 87060,
+      "🔍": 87061,
+      "凴": 87062,
+      "澆": 87063,
+      "塵": 87064,
+      "蒔": 87065,
+      "訬": 87066,
+      "棇": 87067,
+      "濸": 87068,
+      "隫": 87069,
+      "鯝": 87070,
+      "珰": 87071,
+      "鸩": 87072,
+      "晪": 87073,
+      "🈲": 87074,
+      "鋃": 87075,
+      "螘": 87076,
+      "喓": 87077,
+      "ɐ": 87078,
+      "獺": 87079,
+      "Τ": 87080,
+      "趑": 87081,
+      "圞": 87082,
+      "壒": 87083,
+      "鮄": 87084,
+      "贉": 87085,
+      "惤": 87086,
+      "👨🏻‍❤️‍💋‍👨🏽": 87087,
+      "稏": 87088,
+      "⃹": 87089,
+      "🧑🏽‍⚖️": 87090,
+      "撄": 87091,
+      "鐳": 87092,
+      "貹": 87093,
+      "恴": 87094,
+      "謓": 87095,
+      "✝️": 87096,
+      "劭": 87097,
+      "竄": 87098,
+      "臱": 87099,
+      "👏🏽": 87100,
+      "⊅": 87101,
+      "濳": 87102,
+      "嘸": 87103,
+      "蚰": 87104,
+      "営": 87105,
+      "誁": 87106,
+      "酺": 87107,
+      "櫬": 87108,
+      "趹": 87109,
+      "┄": 87110,
+      "娫": 87111,
+      "汼": 87112,
+      "笧": 87113,
+      "艃": 87114,
+      "衈": 87115,
+      "魪": 87116,
+      "⑪": 87117,
+      "敨": 87118,
+      "👐🏿": 87119,
+      "钡": 87120,
+      "鎁": 87121,
+      "裮": 87122,
+      "糔": 87123,
+      "蝗": 87124,
+      "💯": 87125,
+      "櫵": 87126,
+      "渦": 87127,
+      "鉌": 87128,
+      "晷": 87129,
+      "⒊": 87130,
+      "忐": 87131,
+      "髜": 87132,
+      "篪": 87133,
+      "羒": 87134,
+      "妩": 87135,
+      "黜": 87136,
+      "璉": 87137,
+      "😜": 87138,
+      "挐": 87139,
+      "梑": 87140,
+      "渆": 87141,
+      "夓": 87142,
+      "蘒": 87143,
+      "⇬": 87144,
+      "蕿": 87145,
+      "柡": 87146,
+      "麫": 87147,
+      "杷": 87148,
+      "ȱ": 87149,
+      "疠": 87150,
+      "蹵": 87151,
+      "豬": 87152,
+      "❎": 87153,
+      "窎": 87154,
+      "樨": 87155,
+      "🚯": 87156,
+      "枬": 87157,
+      "礜": 87158,
+      "⏲": 87159,
+      "箽": 87160,
+      "扱": 87161,
+      "🖕🏽": 87162,
+      "🚶🏾‍♂️": 87163,
+      "篫": 87164,
+      "🈵": 87165,
+      "ϐ": 87166,
+      "鰊": 87167,
+      "煿": 87168,
+      "燩": 87169,
+      "🍕": 87170,
+      "蹆": 87171,
+      "‿": 87172,
+      "螐": 87173,
+      "蚋": 87174,
+      "槩": 87175,
+      "⍨": 87176,
+      "蛴": 87177,
+      "叻": 87178,
+      "碨": 87179,
+      "駔": 87180,
+      "酠": 87181,
+      "抾": 87182,
+      "🖕🏾": 87183,
+      "矈": 87184,
+      "癷": 87185,
+      "鄁": 87186,
+      "隬": 87187,
+      "👉🏻": 87188,
+      "稗": 87189,
+      "Β": 87190,
+      "袝": 87191,
+      "阠": 87192,
+      "膖": 87193,
+      "Ε": 87194,
+      "🖲": 87195,
+      "嗝": 87196,
+      "愐": 87197,
+      "厭": 87198,
+      "🙆🏿‍♀️": 87199,
+      "👱🏼": 87200,
+      "瞔": 87201,
+      "簰": 87202,
+      "銅": 87203,
+      "郧": 87204,
+      "🫲🏾": 87205,
+      "槻": 87206,
+      "聯": 87207,
+      "🈴": 87208,
+      "🇦🇲": 87209,
+      "秄": 87210,
+      "🤵🏻‍♀️": 87211,
+      "亷": 87212,
+      "☵": 87213,
+      "窢": 87214,
+      "輯": 87215,
+      "🕕": 87216,
+      "⍏": 87217,
+      "灡": 87218,
+      "咩": 87219,
+      "減": 87220,
+      "粁": 87221,
+      "Ȉ": 87222,
+      "歿": 87223,
+      "🧜🏼": 87224,
+      "⇋": 87225,
+      "鞞": 87226,
+      "饤": 87227,
+      "⛁": 87228,
+      "櫴": 87229,
+      "閜": 87230,
+      "掄": 87231,
+      "喲": 87232,
+      "觙": 87233,
+      "躩": 87234,
+      "拕": 87235,
+      "鷘": 87236,
+      "潿": 87237,
+      "黲": 87238,
+      "勨": 87239,
+      "呒": 87240,
+      "狆": 87241,
+      "斁": 87242,
+      "絻": 87243,
+      "⌬": 87244,
+      "搵": 87245,
+      "箾": 87246,
+      "🏊": 87247,
+      "惓": 87248,
+      "觗": 87249,
+      "👩🏿": 87250,
+      "屲": 87251,
+      "↢": 87252,
+      "疦": 87253,
+      "👑": 87254,
+      "抿": 87255,
+      "娯": 87256,
+      "筺": 87257,
+      "瓽": 87258,
+      "嘌": 87259,
+      "🚴🏾‍♂️": 87260,
+      "‽": 87261,
+      "🤿": 87262,
+      "慲": 87263,
+      "榯": 87264,
+      "澬": 87265,
+      "釵": 87266,
+      "苰": 87267,
+      "🤰🏾": 87268,
+      "⏺": 87269,
+      "雎": 87270,
+      "溗": 87271,
+      "氵": 87272,
+      "🛌🏼": 87273,
+      "掍": 87274,
+      "🚴🏽": 87275,
+      "🤾🏼": 87276,
+      "蹿": 87277,
+      "坽": 87278,
+      "鸰": 87279,
+      "甏": 87280,
+      "蟖": 87281,
+      "💂🏼‍♂️": 87282,
+      "濙": 87283,
+      "痝": 87284,
+      "👩🏼‍🚀": 87285,
+      "銵": 87286,
+      "纼": 87287,
+      "驡": 87288,
+      "蚶": 87289,
+      "🪅": 87290,
+      "蕦": 87291,
+      "鲔": 87292,
+      "🙅🏻‍♀️": 87293,
+      "🦸🏼‍♀️": 87294,
+      "颺": 87295,
+      "鏔": 87296,
+      "麬": 87297,
+      "豞": 87298,
+      "鹞": 87299,
+      "冂": 87300,
+      "婐": 87301,
+      "衩": 87302,
+      "渒": 87303,
+      "乢": 87304,
+      "樜": 87305,
+      "槫": 87306,
+      "🦽": 87307,
+      "庣": 87308,
+      "幛": 87309,
+      "菦": 87310,
+      "ₘ": 87311,
+      "悎": 87312,
+      "邞": 87313,
+      "🚶🏽": 87314,
+      "🧑‍🔬": 87315,
+      "埮": 87316,
+      "烻": 87317,
+      "駷": 87318,
+      "Ⓡ": 87319,
+      "聄": 87320,
+      "緵": 87321,
+      "🦐": 87322,
+      "葃": 87323,
+      "饦": 87324,
+      "伇": 87325,
+      "👩🏽‍❤️‍👨🏼": 87326,
+      "贔": 87327,
+      "鄱": 87328,
+      "👁": 87329,
+      "葢": 87330,
+      "講": 87331,
+      "🫂": 87332,
+      "憅": 87333,
+      "👩🏿‍🍼": 87334,
+      "侸": 87335,
+      "艩": 87336,
+      "蕫": 87337,
+      "⬇️": 87338,
+      "畖": 87339,
+      "慺": 87340,
+      "🇹🇷": 87341,
+      "👏🏾": 87342,
+      "濕": 87343,
+      "呚": 87344,
+      "蟼": 87345,
+      "🚶‍♀️": 87346,
+      "邅": 87347,
+      "粰": 87348,
+      "憺": 87349,
+      "楋": 87350,
+      "偫": 87351,
+      "匍": 87352,
+      "阘": 87353,
+      "🙋": 87354,
+      "🙌🏾": 87355,
+      "瑉": 87356,
+      "慚": 87357,
+      "🧛🏽‍♀️": 87358,
+      "骡": 87359,
+      "🔯": 87360,
+      "蠫": 87361,
+      "禨": 87362,
+      "魦": 87363,
+      "敥": 87364,
+      "腷": 87365,
+      "韊": 87366,
+      "甑": 87367,
+      "眜": 87368,
+      "徼": 87369,
+      "垾": 87370,
+      "盄": 87371,
+      "脼": 87372,
+      "厵": 87373,
+      "鉦": 87374,
+      "˛": 87375,
+      "盬": 87376,
+      "撽": 87377,
+      "凇": 87378,
+      "劰": 87379,
+      "鐦": 87380,
+      "慩": 87381,
+      "朅": 87382,
+      "襛": 87383,
+      "╛": 87384,
+      "氿": 87385,
+      "獄": 87386,
+      "隁": 87387,
+      "枾": 87388,
+      "鍟": 87389,
+      "蔴": 87390,
+      "💂": 87391,
+      "匐": 87392,
+      "韆": 87393,
+      "戓": 87394,
+      "⊄": 87395,
+      "湵": 87396,
+      "刂": 87397,
+      "碒": 87398,
+      "⛱": 87399,
+      "盅": 87400,
+      "🚴🏽‍♀️": 87401,
+      "🚣‍♀️": 87402,
+      "朞": 87403,
+      "💂🏼‍♀️": 87404,
+      "Ⓤ": 87405,
+      "鐥": 87406,
+      "〽️": 87407,
+      "鈾": 87408,
+      "🫱🏽‍🫲🏾": 87409,
+      "鰢": 87410,
+      "⒯": 87411,
+      "駡": 87412,
+      "抐": 87413,
+      "栯": 87414,
+      "ℽ": 87415,
+      "摿": 87416,
+      "捰": 87417,
+      "畻": 87418,
+      "彵": 87419,
+      "轈": 87420,
+      "⒴": 87421,
+      "繡": 87422,
+      "🚒": 87423,
+      "👩🏽‍❤️‍💋‍👩🏽": 87424,
+      "鮾": 87425,
+      "幱": 87426,
+      "辔": 87427,
+      "鳓": 87428,
+      "鴀": 87429,
+      "😟": 87430,
+      "囍": 87431,
+      "蕹": 87432,
+      "蒵": 87433,
+      "🩼": 87434,
+      "🇪🇺": 87435,
+      "狲": 87436,
+      "潶": 87437,
+      "⒝": 87438,
+      "贋": 87439,
+      "曐": 87440,
+      "碣": 87441,
+      "醡": 87442,
+      "垻": 87443,
+      "👩🏼‍❤️‍👩🏼": 87444,
+      "嚴": 87445,
+      "嬠": 87446,
+      "詣": 87447,
+      "暌": 87448,
+      "堙": 87449,
+      "蘫": 87450,
+      "鸶": 87451,
+      "顒": 87452,
+      "⏶": 87453,
+      "饩": 87454,
+      "攘": 87455,
+      "˓": 87456,
+      "閌": 87457,
+      "裆": 87458,
+      "💇🏻‍♂️": 87459,
+      "耙": 87460,
+      "庖": 87461,
+      "壷": 87462,
+      "哓": 87463,
+      "🇧🇴": 87464,
+      "嗉": 87465,
+      "匜": 87466,
+      "灢": 87467,
+      "峄": 87468,
+      "🏕": 87469,
+      "觚": 87470,
+      "垨": 87471,
+      "鎗": 87472,
+      "幰": 87473,
+      "瑄": 87474,
+      "硡": 87475,
+      "閞": 87476,
+      "塻": 87477,
+      "⎖": 87478,
+      "簏": 87479,
+      "婾": 87480,
+      "彲": 87481,
+      "懯": 87482,
+      "連": 87483,
+      "悛": 87484,
+      "臓": 87485,
+      "⎿": 87486,
+      "薀": 87487,
+      "🙋🏽‍♂️": 87488,
+      "🍤": 87489,
+      "鄺": 87490,
+      "擂": 87491,
+      "╧": 87492,
+      "澠": 87493,
+      "乑": 87494,
+      "⊴": 87495,
+      "🪵": 87496,
+      "骪": 87497,
+      "氬": 87498,
+      "槆": 87499,
+      "📅": 87500,
+      "蝨": 87501,
+      "儺": 87502,
+      "糬": 87503,
+      "轋": 87504,
+      "ƨ": 87505,
+      "柰": 87506,
+      "⎂": 87507,
+      "陮": 87508,
+      "畆": 87509,
+      "嶻": 87510,
+      "捦": 87511,
+      "溊": 87512,
+      "縁": 87513,
+      "👳🏻‍♂️": 87514,
+      "藲": 87515,
+      "荲": 87516,
+      "枰": 87517,
+      "秙": 87518,
+      "袞": 87519,
+      "徭": 87520,
+      "顉": 87521,
+      "羍": 87522,
+      "嚸": 87523,
+      "逫": 87524,
+      "逭": 87525,
+      "巉": 87526,
+      "囯": 87527,
+      "齻": 87528,
+      "誂": 87529,
+      "漧": 87530,
+      "廚": 87531,
+      "慸": 87532,
+      "揂": 87533,
+      "俀": 87534,
+      "櫘": 87535,
+      "佘": 87536,
+      "鸼": 87537,
+      "巸": 87538,
+      "橳": 87539,
+      "耇": 87540,
+      "🇦🇷": 87541,
+      "🧛🏼‍♂️": 87542,
+      "飦": 87543,
+      "諔": 87544,
+      "➡️": 87545,
+      "莣": 87546,
+      "瀺": 87547,
+      "餇": 87548,
+      "👩🏿‍🔬": 87549,
+      "茊": 87550,
+      "躦": 87551,
+      "箚": 87552,
+      "🕸": 87553,
+      "鰷": 87554,
+      "纃": 87555,
+      "崁": 87556,
+      "魑": 87557,
+      "⓺": 87558,
+      "👩🏾‍🦲": 87559,
+      "觲": 87560,
+      "袚": 87561,
+      "僓": 87562,
+      "諞": 87563,
+      "🩺": 87564,
+      "淲": 87565,
+      "縔": 87566,
+      "⑑": 87567,
+      "縚": 87568,
+      "♓": 87569,
+      "匚": 87570,
+      "墆": 87571,
+      "‱": 87572,
+      "💔": 87573,
+      "🐡": 87574,
+      "毣": 87575,
+      "膭": 87576,
+      "🚣🏻‍♂️": 87577,
+      "窽": 87578,
+      "榏": 87579,
+      "🎭": 87580,
+      "趌": 87581,
+      "猇": 87582,
+      "徣": 87583,
+      "擯": 87584,
+      "̠": 87585,
+      "嚛": 87586,
+      "鹎": 87587,
+      "嫿": 87588,
+      "⊑": 87589,
+      "芄": 87590,
+      "獬": 87591,
+      "転": 87592,
+      "🇨🇬": 87593,
+      "ŭ": 87594,
+      "👨🏾‍🦲": 87595,
+      "蠳": 87596,
+      "娚": 87597,
+      "凎": 87598,
+      "託": 87599,
+      "羫": 87600,
+      "悿": 87601,
+      "芀": 87602,
+      "耴": 87603,
+      "祋": 87604,
+      "賮": 87605,
+      "ʈ": 87606,
+      "桜": 87607,
+      "勯": 87608,
+      "錀": 87609,
+      "怑": 87610,
+      "頸": 87611,
+      "噆": 87612,
+      "🕵️": 87613,
+      "擰": 87614,
+      "🕯": 87615,
+      "荭": 87616,
+      "滀": 87617,
+      "華": 87618,
+      "茤": 87619,
+      "埁": 87620,
+      "駐": 87621,
+      "鋀": 87622,
+      "啢": 87623,
+      "◵": 87624,
+      "⚴": 87625,
+      "琗": 87626,
+      "貄": 87627,
+      "庱": 87628,
+      "綾": 87629,
+      "櫎": 87630,
+      "🇹🇼": 87631,
+      "坫": 87632,
+      "👂🏼": 87633,
+      "瞜": 87634,
+      "⚺": 87635,
+      "醲": 87636,
+      "🔱": 87637,
+      "蔞": 87638,
+      "✳️": 87639,
+      "犝": 87640,
+      "莶": 87641,
+      "磚": 87642,
+      "攈": 87643,
+      "攎": 87644,
+      "扻": 87645,
+      "诮": 87646,
+      "蒈": 87647,
+      "禟": 87648,
+      "蘁": 87649,
+      "鬗": 87650,
+      "豇": 87651,
+      "導": 87652,
+      "‏": 87653,
+      "💅🏿": 87654,
+      "襷": 87655,
+      "Ȓ": 87656,
+      "辘": 87657,
+      "齴": 87658,
+      "🫑": 87659,
+      "̦": 87660,
+      "抭": 87661,
+      "匞": 87662,
+      "坓": 87663,
+      "ℂ": 87664,
+      "畘": 87665,
+      "诋": 87666,
+      "轸": 87667,
+      "產": 87668,
+      "桹": 87669,
+      "偺": 87670,
+      "嚌": 87671,
+      "妑": 87672,
+      "蟆": 87673,
+      "鉙": 87674,
+      "盞": 87675,
+      "絜": 87676,
+      "蔁": 87677,
+      "掶": 87678,
+      "藱": 87679,
+      "蜃": 87680,
+      "靕": 87681,
+      "琨": 87682,
+      "悤": 87683,
+      "瀬": 87684,
+      "裓": 87685,
+      "猑": 87686,
+      "荌": 87687,
+      "鍭": 87688,
+      "🇵🇹": 87689,
+      "艧": 87690,
+      "棜": 87691,
+      "翯": 87692,
+      "枑": 87693,
+      "皁": 87694,
+      "🧙🏾‍♂️": 87695,
+      "搗": 87696,
+      "钜": 87697,
+      "雙": 87698,
+      "┵": 87699,
+      "艕": 87700,
+      "薊": 87701,
+      "蛳": 87702,
+      "浧": 87703,
+      "焈": 87704,
+      "擃": 87705,
+      "豘": 87706,
+      "弨": 87707,
+      "笀": 87708,
+      "觯": 87709,
+      "📋": 87710,
+      "豑": 87711,
+      "樢": 87712,
+      "礕": 87713,
+      "撁": 87714,
+      "鼨": 87715,
+      "👩🏼‍❤️‍👨🏾": 87716,
+      "赕": 87717,
+      "鍜": 87718,
+      "蓩": 87719,
+      "⃦": 87720,
+      "睞": 87721,
+      "⃄": 87722,
+      "呞": 87723,
+      "🇱🇦": 87724,
+      "呾": 87725,
+      "鶌": 87726,
+      "萳": 87727,
+      "轄": 87728,
+      "⅁": 87729,
+      "🪬": 87730,
+      "劜": 87731,
+      "锩": 87732,
+      "埈": 87733,
+      "鷞": 87734,
+      "鶢": 87735,
+      "攲": 87736,
+      "揮": 87737,
+      "駲": 87738,
+      "淪": 87739,
+      "瘰": 87740,
+      "袥": 87741,
+      "諃": 87742,
+      "騟": 87743,
+      "哙": 87744,
+      "🦧": 87745,
+      "眫": 87746,
+      "彮": 87747,
+      "幐": 87748,
+      "夲": 87749,
+      "儛": 87750,
+      "螖": 87751,
+      "摉": 87752,
+      "竫": 87753,
+      "鷥": 87754,
+      "锏": 87755,
+      "🔣": 87756,
+      "🏌🏼‍♀️": 87757,
+      "菔": 87758,
+      "靧": 87759,
+      "戣": 87760,
+      "冋": 87761,
+      "孱": 87762,
+      "塟": 87763,
+      "⌾": 87764,
+      "綱": 87765,
+      "衲": 87766,
+      "莿": 87767,
+      "🤎": 87768,
+      "蕇": 87769,
+      "踍": 87770,
+      "愂": 87771,
+      "戃": 87772,
+      "岧": 87773,
+      "🚺": 87774,
+      "樉": 87775,
+      "筍": 87776,
+      "🇲🇾": 87777,
+      "🏎": 87778,
+      "🙅🏽": 87779,
+      "欜": 87780,
+      "鵥": 87781,
+      "頻": 87782,
+      "垓": 87783,
+      "˖": 87784,
+      "🔞": 87785,
+      "惉": 87786,
+      "₦": 87787,
+      "鈴": 87788,
+      "萺": 87789,
+      "媤": 87790,
+      "蓨": 87791,
+      "讗": 87792,
+      "禔": 87793,
+      "愽": 87794,
+      "痑": 87795,
+      "茚": 87796,
+      "踃": 87797,
+      "🧑🏿‍✈️": 87798,
+      "Ⅳ": 87799,
+      "∡": 87800,
+      "鍹": 87801,
+      "🤡": 87802,
+      "壊": 87803,
+      "孏": 87804,
+      "陭": 87805,
+      "辎": 87806,
+      "勱": 87807,
+      "尓": 87808,
+      "麿": 87809,
+      "雞": 87810,
+      "肐": 87811,
+      "鋉": 87812,
+      "🇳🇨": 87813,
+      "杍": 87814,
+      "萉": 87815,
+      "庡": 87816,
+      "👩🏼‍❤️‍👨🏽": 87817,
+      "咉": 87818,
+      "紀": 87819,
+      "倨": 87820,
+      "鵚": 87821,
+      "橎": 87822,
+      "⍷": 87823,
+      "諅": 87824,
+      "濱": 87825,
+      "舫": 87826,
+      "╿": 87827,
+      "👔": 87828,
+      "杻": 87829,
+      "⏽": 87830,
+      "螴": 87831,
+      "綩": 87832,
+      "荥": 87833,
+      "蓸": 87834,
+      "👩🏽‍❤️‍👨🏾": 87835,
+      "樸": 87836,
+      "襸": 87837,
+      "嗗": 87838,
+      "鱘": 87839,
+      "鱱": 87840,
+      "勩": 87841,
+      "🚴🏽‍♂️": 87842,
+      "祓": 87843,
+      "芼": 87844,
+      "裍": 87845,
+      "閚": 87846,
+      "豿": 87847,
+      "胙": 87848,
+      "臽": 87849,
+      "稪": 87850,
+      "◸": 87851,
+      "鉖": 87852,
+      "ˡ": 87853,
+      "␗": 87854,
+      "⚹": 87855,
+      "綷": 87856,
+      "☃": 87857,
+      "Ɠ": 87858,
+      "🧑🏽‍⚕️": 87859,
+      "鍛": 87860,
+      "⅌": 87861,
+      "鸫": 87862,
+      "莈": 87863,
+      "暕": 87864,
+      "🇲🇼": 87865,
+      "⚦": 87866,
+      "適": 87867,
+      "🇧🇦": 87868,
+      "妰": 87869,
+      "韄": 87870,
+      "畷": 87871,
+      "涥": 87872,
+      "鳋": 87873,
+      "怟": 87874,
+      "顲": 87875,
+      "殥": 87876,
+      "鼮": 87877,
+      "莙": 87878,
+      "🚩": 87879,
+      "愬": 87880,
+      "闐": 87881,
+      "咿": 87882,
+      "銺": 87883,
+      "鴱": 87884,
+      "覴": 87885,
+      "鶶": 87886,
+      "Ρ": 87887,
+      "↨": 87888,
+      "洕": 87889,
+      "悑": 87890,
+      "瑠": 87891,
+      "穦": 87892,
+      "偠": 87893,
+      "啣": 87894,
+      "覥": 87895,
+      "虋": 87896,
+      "顃": 87897,
+      "騙": 87898,
+      "尞": 87899,
+      "⁀": 87900,
+      "缱": 87901,
+      "颶": 87902,
+      "咻": 87903,
+      "旮": 87904,
+      "👩🏼‍❤️‍💋‍👩🏾": 87905,
+      "鏗": 87906,
+      "卮": 87907,
+      "繈": 87908,
+      "蚩": 87909,
+      "翆": 87910,
+      "🍟": 87911,
+      "扄": 87912,
+      "🧑🏾‍❤️‍💋‍🧑🏿": 87913,
+      "梙": 87914,
+      "骓": 87915,
+      "绬": 87916,
+      "壾": 87917,
+      "杋": 87918,
+      "🛴": 87919,
+      "揻": 87920,
+      "💇‍♂️": 87921,
+      "˒": 87922,
+      "橩": 87923,
+      "玞": 87924,
+      "瀟": 87925,
+      "峵": 87926,
+      "栙": 87927,
+      "觰": 87928,
+      "腼": 87929,
+      "韢": 87930,
+      "蟛": 87931,
+      "瘹": 87932,
+      "歠": 87933,
+      "🇦🇴": 87934,
+      "韷": 87935,
+      "👨🏽‍🦯": 87936,
+      "謥": 87937,
+      "鯰": 87938,
+      "牿": 87939,
+      "镥": 87940,
+      "農": 87941,
+      "嘕": 87942,
+      "蛖": 87943,
+      "➿": 87944,
+      "釧": 87945,
+      "👫🏿": 87946,
+      "婭": 87947,
+      "͗": 87948,
+      "靜": 87949,
+      "圯": 87950,
+      "🍵": 87951,
+      "踚": 87952,
+      "☗": 87953,
+      "嶴": 87954,
+      "赮": 87955,
+      "旝": 87956,
+      "楫": 87957,
+      "ǜ": 87958,
+      "埐": 87959,
+      "🚵🏿‍♀️": 87960,
+      "皠": 87961,
+      "蟻": 87962,
+      "亗": 87963,
+      "藑": 87964,
+      "顈": 87965,
+      "🧏🏼": 87966,
+      "牃": 87967,
+      "🎼": 87968,
+      "熈": 87969,
+      "柃": 87970,
+      "幄": 87971,
+      "戺": 87972,
+      "💪🏽": 87973,
+      "軻": 87974,
+      "氞": 87975,
+      "癚": 87976,
+      "歋": 87977,
+      "鈧": 87978,
+      "愲": 87979,
+      "鶾": 87980,
+      "🍶": 87981,
+      "鳥": 87982,
+      "虛": 87983,
+      "辥": 87984,
+      "鵜": 87985,
+      "ύ": 87986,
+      "🧑🏽‍🦳": 87987,
+      "劗": 87988,
+      "🤷‍♀️": 87989,
+      "▭": 87990,
+      "煛": 87991,
+      "頞": 87992,
+      "慁": 87993,
+      "黺": 87994,
+      "⃤": 87995,
+      "̱": 87996,
+      "罉": 87997,
+      "烇": 87998,
+      "焄": 87999,
+      "傐": 88000,
+      "鍻": 88001,
+      "鍦": 88002,
+      "質": 88003,
+      "欪": 88004,
+      "👨🏽‍❤️‍💋‍👨🏿": 88005,
+      "庴": 88006,
+      "疢": 88007,
+      "蘐": 88008,
+      "🇵🇷": 88009,
+      "🤲🏻": 88010,
+      "鉫": 88011,
+      "雧": 88012,
+      "銪": 88013,
+      "詑": 88014,
+      "墐": 88015,
+      "裺": 88016,
+      "讥": 88017,
+      "尨": 88018,
+      "滶": 88019,
+      "匃": 88020,
+      "蹎": 88021,
+      "🦸": 88022,
+      "⎄": 88023,
+      "鄩": 88024,
+      "尕": 88025,
+      "👇🏿": 88026,
+      "旐": 88027,
+      "氇": 88028,
+      "Ȯ": 88029,
+      "褡": 88030,
+      "焣": 88031,
+      "籌": 88032,
+      "👩🏼‍🏫": 88033,
+      "笟": 88034,
+      "浭": 88035,
+      "曢": 88036,
+      "⚼": 88037,
+      "坋": 88038,
+      "矦": 88039,
+      "鍆": 88040,
+      "謺": 88041,
+      "豀": 88042,
+      "裑": 88043,
+      "餛": 88044,
+      "絺": 88045,
+      "潉": 88046,
+      "🫅🏽": 88047,
+      "♕": 88048,
+      "蔄": 88049,
+      "礽": 88050,
+      "笴": 88051,
+      "藚": 88052,
+      "锾": 88053,
+      "磖": 88054,
+      "浱": 88055,
+      "🕹": 88056,
+      "🇸🇩": 88057,
+      "賭": 88058,
+      "玴": 88059,
+      "🤙🏽": 88060,
+      "嗇": 88061,
+      "紽": 88062,
+      "👰🏽‍♀️": 88063,
+      "薎": 88064,
+      "硏": 88065,
+      "躐": 88066,
+      "⃋": 88067,
+      "腓": 88068,
+      "儧": 88069,
+      "👋🏽": 88070,
+      "誦": 88071,
+      "硣": 88072,
+      "👱🏽‍♀️": 88073,
+      "📌": 88074,
+      "灤": 88075,
+      "緋": 88076,
+      "怈": 88077,
+      "葕": 88078,
+      "鄍": 88079,
+      "攓": 88080,
+      "柀": 88081,
+      "🧙🏻‍♂️": 88082,
+      "撾": 88083,
+      "撜": 88084,
+      "屬": 88085,
+      "🦔": 88086,
+      "🫱🏼‍🫲🏾": 88087,
+      "⒍": 88088,
+      "◞": 88089,
+      "隑": 88090,
+      "🧒🏻": 88091,
+      "幇": 88092,
+      "壙": 88093,
+      "龡": 88094,
+      "饽": 88095,
+      "獲": 88096,
+      "哞": 88097,
+      "朂": 88098,
+      "欏": 88099,
+      "👩🏿‍🚒": 88100,
+      "粄": 88101,
+      "榊": 88102,
+      "塢": 88103,
+      "蕑": 88104,
+      "瞾": 88105,
+      "̳": 88106,
+      "▥": 88107,
+      "🧑🏻‍🦱": 88108,
+      "👨🏿‍🔧": 88109,
+      "圵": 88110,
+      "騏": 88111,
+      "鲿": 88112,
+      "囥": 88113,
+      "濗": 88114,
+      "跈": 88115,
+      "⏙": 88116,
+      "烊": 88117,
+      "✁": 88118,
+      "擻": 88119,
+      "✡︎": 88120,
+      "🐏": 88121,
+      "鸪": 88122,
+      "圄": 88123,
+      "籀": 88124,
+      "蟰": 88125,
+      "🧑‍🦼": 88126,
+      "🇹🇨": 88127,
+      "🧑🏾‍🦯": 88128,
+      "♵": 88129,
+      "藄": 88130,
+      "婌": 88131,
+      "頃": 88132,
+      "腂": 88133,
+      "溔": 88134,
+      "磽": 88135,
+      "悢": 88136,
+      "👩🏻‍❤️‍👩🏾": 88137,
+      "瑴": 88138,
+      "羣": 88139,
+      "灲": 88140,
+      "幤": 88141,
+      "楓": 88142,
+      "鍣": 88143,
+      "👨🏿‍💻": 88144,
+      "焹": 88145,
+      "劊": 88146,
+      "Ⓘ": 88147,
+      "怿": 88148,
+      "莀": 88149,
+      "臖": 88150,
+      "俪": 88151,
+      "呉": 88152,
+      "驅": 88153,
+      "礞": 88154,
+      "畎": 88155,
+      "燚": 88156,
+      "諻": 88157,
+      "勚": 88158,
+      "蹸": 88159,
+      "哊": 88160,
+      "崺": 88161,
+      "Ǧ": 88162,
+      "泭": 88163,
+      "齉": 88164,
+      "鐒": 88165,
+      "💂🏿‍♀️": 88166,
+      "⍴": 88167,
+      "骎": 88168,
+      "┋": 88169,
+      "柼": 88170,
+      "濏": 88171,
+      "嵡": 88172,
+      "翾": 88173,
+      "蛫": 88174,
+      "鹙": 88175,
+      "跾": 88176,
+      "🫙": 88177,
+      "誘": 88178,
+      "覦": 88179,
+      "鐌": 88180,
+      "搔": 88181,
+      "鷐": 88182,
+      "冺": 88183,
+      "🧑‍🍼": 88184,
+      "媯": 88185,
+      "葖": 88186,
+      "🚴🏿": 88187,
+      "ʧ": 88188,
+      "瘛": 88189,
+      "歺": 88190,
+      "蝬": 88191,
+      "乽": 88192,
+      "鰌": 88193,
+      "曫": 88194,
+      "祑": 88195,
+      "鋕": 88196,
+      "⃩": 88197,
+      "佒": 88198,
+      "哳": 88199,
+      "隩": 88200,
+      "琒": 88201,
+      "窏": 88202,
+      "🤽🏾‍♂️": 88203,
+      "粓": 88204,
+      "晿": 88205,
+      "朳": 88206,
+      "鄜": 88207,
+      "跹": 88208,
+      "鞈": 88209,
+      "豒": 88210,
+      "橯": 88211,
+      "襎": 88212,
+      "歃": 88213,
+      "搽": 88214,
+      "▤": 88215,
+      "蚪": 88216,
+      "珉": 88217,
+      "巇": 88218,
+      "簔": 88219,
+      "覱": 88220,
+      "🫱": 88221,
+      "灗": 88222,
+      "冱": 88223,
+      "邲": 88224,
+      "瞇": 88225,
+      "誆": 88226,
+      "洌": 88227,
+      "倮": 88228,
+      "🧑‍💼": 88229,
+      "疉": 88230,
+      "║": 88231,
+      "麪": 88232,
+      "悓": 88233,
+      "惵": 88234,
+      "😦": 88235,
+      "煷": 88236,
+      "歎": 88237,
+      "譶": 88238,
+      "韐": 88239,
+      "🔵": 88240,
+      "訕": 88241,
+      "ₕ": 88242,
+      "歭": 88243,
+      "ζ": 88244,
+      "鳕": 88245,
+      "芛": 88246,
+      "瘗": 88247,
+      "苖": 88248,
+      "阥": 88249,
+      "濓": 88250,
+      "銯": 88251,
+      "Ⓖ": 88252,
+      "👨🏼‍🎓": 88253,
+      "鳆": 88254,
+      "輽": 88255,
+      "帿": 88256,
+      "載": 88257,
+      "镋": 88258,
+      "嚒": 88259,
+      "綁": 88260,
+      "圱": 88261,
+      "絰": 88262,
+      "🚴🏼‍♂️": 88263,
+      "镡": 88264,
+      "⍶": 88265,
+      "鱡": 88266,
+      "蕠": 88267,
+      "ͺ": 88268,
+      "珩": 88269,
+      "舏": 88270,
+      "禦": 88271,
+      "觸": 88272,
+      "邐": 88273,
+      "🪽": 88274,
+      "🕵🏼": 88275,
+      "搖": 88276,
+      "🦫": 88277,
+      "萭": 88278,
+      "🇸🇾": 88279,
+      "💂🏻‍♂️": 88280,
+      "馭": 88281,
+      "糓": 88282,
+      "媃": 88283,
+      "🧚🏻‍♂️": 88284,
+      "⚂": 88285,
+      "蹰": 88286,
+      "炩": 88287,
+      "踛": 88288,
+      "␁": 88289,
+      "鸐": 88290,
+      "轞": 88291,
+      "棴": 88292,
+      "郦": 88293,
+      "趤": 88294,
+      "牾": 88295,
+      "👎🏿": 88296,
+      "⁦": 88297,
+      "📯": 88298,
+      "鰹": 88299,
+      "箷": 88300,
+      "搃": 88301,
+      "誷": 88302,
+      "俵": 88303,
+      "櫾": 88304,
+      "🍩": 88305,
+      "欛": 88306,
+      "稥": 88307,
+      "觩": 88308,
+      "凵": 88309,
+      "͋": 88310,
+      "⑶": 88311,
+      "欳": 88312,
+      "怫": 88313,
+      "艈": 88314,
+      "摟": 88315,
+      "伀": 88316,
+      "騠": 88317,
+      "踳": 88318,
+      "翃": 88319,
+      "Ū": 88320,
+      "愸": 88321,
+      "鴭": 88322,
+      "👩‍👩‍👦": 88323,
+      "轟": 88324,
+      "垚": 88325,
+      "🧑🏽‍❤️‍🧑🏿": 88326,
+      "枔": 88327,
+      "歩": 88328,
+      "跦": 88329,
+      "瞉": 88330,
+      "𓀒": 88331,
+      "🧑🏽‍🔬": 88332,
+      "繌": 88333,
+      "埓": 88334,
+      "怸": 88335,
+      "碉": 88336,
+      "♎": 88337,
+      "🧑🏼‍⚖️": 88338,
+      "💆🏾‍♀️": 88339,
+      "膴": 88340,
+      "聳": 88341,
+      "殚": 88342,
+      "敧": 88343,
+      "擸": 88344,
+      "泔": 88345,
+      "⃅": 88346,
+      "┶": 88347,
+      "👨🏿‍❤️‍💋‍👨🏻": 88348,
+      "冪": 88349,
+      "磯": 88350,
+      "蔫": 88351,
+      "慏": 88352,
+      "橽": 88353,
+      "瀭": 88354,
+      "🪯": 88355,
+      "鐓": 88356,
+      "飄": 88357,
+      "🤹🏼‍♂️": 88358,
+      "痚": 88359,
+      "˥": 88360,
+      "蜇": 88361,
+      "🧒": 88362,
+      "𓀋": 88363,
+      "誰": 88364,
+      "鍲": 88365,
+      "🚉": 88366,
+      "ĸ": 88367,
+      "🕤": 88368,
+      "牞": 88369,
+      "磻": 88370,
+      "嘣": 88371,
+      "誥": 88372,
+      "🛸": 88373,
+      "熜": 88374,
+      "鰍": 88375,
+      "揳": 88376,
+      "廥": 88377,
+      "̕": 88378,
+      "鲢": 88379,
+      "僖": 88380,
+      "儬": 88381,
+      "鬡": 88382,
+      "魹": 88383,
+      "樠": 88384,
+      "鰆": 88385,
+      "鸹": 88386,
+      "💆🏽‍♂️": 88387,
+      "畽": 88388,
+      "絧": 88389,
+      "擺": 88390,
+      "謗": 88391,
+      "禸": 88392,
+      "讴": 88393,
+      "餍": 88394,
+      "⒇": 88395,
+      "Ｍ": 88396,
+      "虧": 88397,
+      "栘": 88398,
+      "僠": 88399,
+      "欈": 88400,
+      "玔": 88401,
+      "🧑🏻‍❤️‍💋‍🧑🏼": 88402,
+      "ˌ": 88403,
+      "嗌": 88404,
+      "栌": 88405,
+      "⁔": 88406,
+      "甡": 88407,
+      "挿": 88408,
+      "🌔": 88409,
+      "墦": 88410,
+      "棊": 88411,
+      "惒": 88412,
+      "🇲🇳": 88413,
+      "虘": 88414,
+      "顟": 88415,
+      "毘": 88416,
+      "🧑🏽‍🌾": 88417,
+      "仚": 88418,
+      "Α": 88419,
+      "🧘🏻‍♂️": 88420,
+      "鬔": 88421,
+      "赲": 88422,
+      "Ǥ": 88423,
+      "✉": 88424,
+      "梒": 88425,
+      "燖": 88426,
+      "抎": 88427,
+      "⊽": 88428,
+      "鯹": 88429,
+      "笷": 88430,
+      "噾": 88431,
+      "薘": 88432,
+      "沘": 88433,
+      "吢": 88434,
+      "洚": 88435,
+      "倖": 88436,
+      "鴚": 88437,
+      "廦": 88438,
+      "艚": 88439,
+      "🦵🏿": 88440,
+      "͠": 88441,
+      "崑": 88442,
+      "⛰": 88443,
+      "屨": 88444,
+      "Ű": 88445,
+      "朥": 88446,
+      "␏": 88447,
+      "槔": 88448,
+      "硷": 88449,
+      "矽": 88450,
+      "鬂": 88451,
+      "葋": 88452,
+      "璟": 88453,
+      "拺": 88454,
+      "礄": 88455,
+      "捭": 88456,
+      "娕": 88457,
+      "遱": 88458,
+      "👩‍👦": 88459,
+      "勴": 88460,
+      "峌": 88461,
+      "勹": 88462,
+      "秞": 88463,
+      "獨": 88464,
+      "唪": 88465,
+      "嗋": 88466,
+      "👨‍👦‍👦": 88467,
+      "閙": 88468,
+      "擜": 88469,
+      "鳮": 88470,
+      "赟": 88471,
+      "挵": 88472,
+      "ʠ": 88473,
+      "亂": 88474,
+      "閐": 88475,
+      "縒": 88476,
+      "柂": 88477,
+      "橆": 88478,
+      "👋": 88479,
+      "琷": 88480,
+      "劆": 88481,
+      "泏": 88482,
+      "秗": 88483,
+      "坱": 88484,
+      "诘": 88485,
+      "詠": 88486,
+      "⌕": 88487,
+      "踨": 88488,
+      "謟": 88489,
+      "": 88490,
+      "耣": 88491,
+      "🇳🇺": 88492,
+      "芩": 88493,
+      "鄋": 88494,
+      "菢": 88495,
+      "歏": 88496,
+      "珹": 88497,
+      "黢": 88498,
+      "乮": 88499,
+      "圑": 88500,
+      "誽": 88501,
+      "憟": 88502,
+      "👇🏼": 88503,
+      "攨": 88504,
+      "́": 88505,
+      "🔡": 88506,
+      "怘": 88507,
+      "則": 88508,
+      "🇾🇪": 88509,
+      "捽": 88510,
+      "荴": 88511,
+      "稺": 88512,
+      "薌": 88513,
+      "仴": 88514,
+      "雃": 88515,
+      "鐐": 88516,
+      "鼟": 88517,
+      "傝": 88518,
+      "炇": 88519,
+      "訢": 88520,
+      "閼": 88521,
+      "踇": 88522,
+      "鮉": 88523,
+      "櫜": 88524,
+      "韴": 88525,
+      "袓": 88526,
+      "伢": 88527,
+      "扂": 88528,
+      "🕵🏻‍♂️": 88529,
+      "昝": 88530,
+      "渄": 88531,
+      "缌": 88532,
+      "晹": 88533,
+      "胬": 88534,
+      "擤": 88535,
+      "👩🏻‍💻": 88536,
+      "栦": 88537,
+      "ⅻ": 88538,
+      "噠": 88539,
+      "豏": 88540,
+      "╉": 88541,
+      "雚": 88542,
+      "💆🏾‍♂️": 88543,
+      "噒": 88544,
+      "瘬": 88545,
+      "鮿": 88546,
+      "銧": 88547,
+      "奦": 88548,
+      "鋛": 88549,
+      "隼": 88550,
+      "艪": 88551,
+      "剐": 88552,
+      "≒": 88553,
+      "": 88554,
+      "棽": 88555,
+      "₺": 88556,
+      "彸": 88557,
+      "鷰": 88558,
+      "锵": 88559,
+      "勆": 88560,
+      "聦": 88561,
+      "鬸": 88562,
+      "堖": 88563,
+      "瀲": 88564,
+      "覷": 88565,
+      "🧍🏼‍♀️": 88566,
+      "4️⃣": 88567,
+      "鞲": 88568,
+      "岺": 88569,
+      "Ⓢ": 88570,
+      "👨🏽‍⚖️": 88571,
+      "礉": 88572,
+      "扃": 88573,
+      "娒": 88574,
+      "鼚": 88575,
+      "麈": 88576,
+      "妴": 88577,
+      "奧": 88578,
+      "蹨": 88579,
+      "邝": 88580,
+      "⍸": 88581,
+      "艝": 88582,
+      "耰": 88583,
+      "峋": 88584,
+      "熗": 88585,
+      "裄": 88586,
+      "鲂": 88587,
+      "🏑": 88588,
+      "帻": 88589,
+      "🛬": 88590,
+      "裌": 88591,
+      "蹢": 88592,
+      "Ī": 88593,
+      "🐈‍⬛": 88594,
+      "鱊": 88595,
+      "縎": 88596,
+      "螞": 88597,
+      "纉": 88598,
+      "🚕": 88599,
+      "銠": 88600,
+      "📿": 88601,
+      "鮥": 88602,
+      "揦": 88603,
+      "👨🏼‍❤️‍💋‍👨🏻": 88604,
+      "矠": 88605,
+      "👩‍🍼": 88606,
+      "鐾": 88607,
+      "鰭": 88608,
+      "欶": 88609,
+      "噃": 88610,
+      "籾": 88611,
+      "嘏": 88612,
+      "🇺🇦": 88613,
+      "茈": 88614,
+      "⍦": 88615,
+      "🚧": 88616,
+      "侲": 88617,
+      "噺": 88618,
+      "込": 88619,
+      "傊": 88620,
+      "鼍": 88621,
+      "巀": 88622,
+      "戼": 88623,
+      "謴": 88624,
+      "糎": 88625,
+      "訦": 88626,
+      "鷆": 88627,
+      "厏": 88628,
+      "蹛": 88629,
+      "🎃": 88630,
+      "屩": 88631,
+      "涫": 88632,
+      "忬": 88633,
+      "悕": 88634,
+      "癪": 88635,
+      "¸": 88636,
+      "郷": 88637,
+      "佗": 88638,
+      "戱": 88639,
+      "鮢": 88640,
+      "娾": 88641,
+      "鶩": 88642,
+      "硺": 88643,
+      "頱": 88644,
+      "鈮": 88645,
+      "瑘": 88646,
+      "⛄": 88647,
+      "礃": 88648,
+      "孳": 88649,
+      "汓": 88650,
+      "禥": 88651,
+      "彶": 88652,
+      "閉": 88653,
+      "燇": 88654,
+      "壓": 88655,
+      "┼": 88656,
+      "杒": 88657,
+      "捯": 88658,
+      "烩": 88659,
+      "乪": 88660,
+      "悒": 88661,
+      "諄": 88662,
+      "諡": 88663,
+      "╡": 88664,
+      "弌": 88665,
+      "喫": 88666,
+      "膑": 88667,
+      "彆": 88668,
+      "炷": 88669,
+      "諽": 88670,
+      "罡": 88671,
+      "💆‍♀️": 88672,
+      "淉": 88673,
+      "蜭": 88674,
+      "🙍🏼‍♂️": 88675,
+      "濲": 88676,
+      "瘧": 88677,
+      "臑": 88678,
+      "酤": 88679,
+      "譲": 88680,
+      "桀": 88681,
+      "幥": 88682,
+      "∳": 88683,
+      "迻": 88684,
+      "💹": 88685,
+      "貧": 88686,
+      "鎅": 88687,
+      "絸": 88688,
+      "蠯": 88689,
+      "🚠": 88690,
+      "桭": 88691,
+      "繂": 88692,
+      "偹": 88693,
+      "猠": 88694,
+      "鳚": 88695,
+      "胊": 88696,
+      "忊": 88697,
+      "迗": 88698,
+      "撪": 88699,
+      "ʽ": 88700,
+      "蓐": 88701,
+      "谝": 88702,
+      "蘇": 88703,
+      "☦": 88704,
+      "匳": 88705,
+      "馜": 88706,
+      "羂": 88707,
+      "伈": 88708,
+      "憱": 88709,
+      "邗": 88710,
+      "捿": 88711,
+      "嬇": 88712,
+      "猷": 88713,
+      "旀": 88714,
+      "烡": 88715,
+      "虰": 88716,
+      "🎅🏿": 88717,
+      "卼": 88718,
+      "癸": 88719,
+      "≡": 88720,
+      "饯": 88721,
+      "幖": 88722,
+      "蜏": 88723,
+      "祄": 88724,
+      "漮": 88725,
+      "圫": 88726,
+      "朁": 88727,
+      "匌": 88728,
+      "🕗": 88729,
+      "🏉": 88730,
+      "埗": 88731,
+      "⚗": 88732,
+      "曏": 88733,
+      "挶": 88734,
+      "岙": 88735,
+      "铟": 88736,
+      "鬏": 88737,
+      "櫚": 88738,
+      "檋": 88739,
+      "剞": 88740,
+      "ȧ": 88741,
+      "爜": 88742,
+      "汆": 88743,
+      "谠": 88744,
+      "🧑🏻‍🚀": 88745,
+      "卩": 88746,
+      "◣": 88747,
+      "拶": 88748,
+      "頊": 88749,
+      "桘": 88750,
+      "🍚": 88751,
+      "栤": 88752,
+      "🇦🇫": 88753,
+      "薿": 88754,
+      "錴": 88755,
+      "盚": 88756,
+      "梋": 88757,
+      "藹": 88758,
+      "髧": 88759,
+      "舕": 88760,
+      "縕": 88761,
+      "豥": 88762,
+      "纐": 88763,
+      "庝": 88764,
+      "抶": 88765,
+      "哣": 88766,
+      "忑": 88767,
+      "璚": 88768,
+      "瘋": 88769,
+      "燏": 88770,
+      "斈": 88771,
+      "🧎🏽": 88772,
+      "⛾": 88773,
+      "眕": 88774,
+      "堟": 88775,
+      "⊬": 88776,
+      "島": 88777,
+      "剽": 88778,
+      "郩": 88779,
+      "箃": 88780,
+      "👓": 88781,
+      "⇊": 88782,
+      "椉": 88783,
+      "蠂": 88784,
+      "癈": 88785,
+      "鉻": 88786,
+      "頾": 88787,
+      "壺": 88788,
+      "🤦🏾": 88789,
+      "🧑🏿‍❤️‍🧑🏼": 88790,
+      "紈": 88791,
+      "躾": 88792,
+      "濐": 88793,
+      "👨🏾‍🔧": 88794,
+      "韑": 88795,
+      "數": 88796,
+      "蒭": 88797,
+      "♊": 88798,
+      "煫": 88799,
+      "鷯": 88800,
+      "壟": 88801,
+      "紘": 88802,
+      "餢": 88803,
+      "猬": 88804,
+      "🤓": 88805,
+      "🇹🇰": 88806,
+      "🐤": 88807,
+      "唻": 88808,
+      "吙": 88809,
+      "嵖": 88810,
+      "🇵🇸": 88811,
+      "酅": 88812,
+      "皑": 88813,
+      "🌷": 88814,
+      "淠": 88815,
+      "畼": 88816,
+      "蒶": 88817,
+      "襐": 88818,
+      "鉠": 88819,
+      "Ģ": 88820,
+      "鼐": 88821,
+      "篿": 88822,
+      "鹸": 88823,
+      "趙": 88824,
+      "𓀑": 88825,
+      "繧": 88826,
+      "👩🏻‍⚖️": 88827,
+      "弴": 88828,
+      "✌": 88829,
+      "墥": 88830,
+      "罰": 88831,
+      "絣": 88832,
+      "緭": 88833,
+      "ǭ": 88834,
+      "☢": 88835,
+      "⛹🏿‍♂️": 88836,
+      "≁": 88837,
+      "鵒": 88838,
+      "奞": 88839,
+      "挦": 88840,
+      "暯": 88841,
+      "⊵": 88842,
+      "蟷": 88843,
+      "棸": 88844,
+      "💅🏾": 88845,
+      "軸": 88846,
+      "鵗": 88847,
+      "鐿": 88848,
+      "梻": 88849,
+      "怃": 88850,
+      "🧔‍♂️": 88851,
+      "🤷🏻‍♂️": 88852,
+      "琄": 88853,
+      "乼": 88854,
+      "ɀ": 88855,
+      "蟞": 88856,
+      "摵": 88857,
+      "漘": 88858,
+      "攡": 88859,
+      "畲": 88860,
+      "蕸": 88861,
+      "朸": 88862,
+      "🙋🏻‍♂️": 88863,
+      "蠭": 88864,
+      "瀐": 88865,
+      "炢": 88866,
+      "躅": 88867,
+      "⑞": 88868,
+      "嚑": 88869,
+      "🤷🏽": 88870,
+      "⓱": 88871,
+      "倊": 88872,
+      "杣": 88873,
+      "鴖": 88874,
+      "笿": 88875,
+      "⬅️": 88876,
+      "乹": 88877,
+      "🚆": 88878,
+      "矍": 88879,
+      "肈": 88880,
+      "庪": 88881,
+      "鏞": 88882,
+      "⒓": 88883,
+      "萇": 88884,
+      "笲": 88885,
+      "🧑🏿‍🦽": 88886,
+      "祻": 88887,
+      "襝": 88888,
+      "*️⃣": 88889,
+      "櫄": 88890,
+      "🧑🏽‍🚀": 88891,
+      "簞": 88892,
+      "🛩": 88893,
+      "雑": 88894,
+      "蟿": 88895,
+      "堀": 88896,
+      "摗": 88897,
+      "瘏": 88898,
+      "襜": 88899,
+      "奮": 88900,
+      "缇": 88901,
+      "庈": 88902,
+      "爊": 88903,
+      "級": 88904,
+      "嵪": 88905,
+      "纋": 88906,
+      "😕": 88907,
+      "孒": 88908,
+      "ƻ": 88909,
+      "鳂": 88910,
+      "💂🏽‍♂️": 88911,
+      "🐠": 88912,
+      "紮": 88913,
+      "͞": 88914,
+      "惦": 88915,
+      "蹁": 88916,
+      "燍": 88917,
+      "🤱🏻": 88918,
+      "🫒": 88919,
+      "暩": 88920,
+      "炛": 88921,
+      "🧍🏿‍♀️": 88922,
+      "鑙": 88923,
+      "🧑🏻‍❤️‍🧑🏽": 88924,
+      "婳": 88925,
+      "臷": 88926,
+      "💇🏿": 88927,
+      "溝": 88928,
+      "眊": 88929,
+      "␞": 88930,
+      "🍢": 88931,
+      "翎": 88932,
+      "梴": 88933,
+      "👷🏿‍♀️": 88934,
+      "峹": 88935,
+      "譱": 88936,
+      "鴌": 88937,
+      "渪": 88938,
+      "🧑🏽‍❤️‍🧑🏻": 88939,
+      "睗": 88940,
+      "迠": 88941,
+      "鰪": 88942,
+      "🤸🏽": 88943,
+      "珐": 88944,
+      "孭": 88945,
+      "黹": 88946,
+      "徲": 88947,
+      "搟": 88948,
+      "😧": 88949,
+      "👩🏻‍🦽": 88950,
+      "檗": 88951,
+      "钸": 88952,
+      "艠": 88953,
+      "菚": 88954,
+      "翢": 88955,
+      "氎": 88956,
+      "旳": 88957,
+      "乣": 88958,
+      "抋": 88959,
+      "凘": 88960,
+      "🏋🏾‍♂️": 88961,
+      "˼": 88962,
+      "儃": 88963,
+      "摏": 88964,
+      "👱‍♂️": 88965,
+      "靬": 88966,
+      "蝮": 88967,
+      "摍": 88968,
+      "沰": 88969,
+      "🧕🏾": 88970,
+      "韫": 88971,
+      "ɜ": 88972,
+      "瓫": 88973,
+      "栫": 88974,
+      "衄": 88975,
+      "殉": 88976,
+      "嗠": 88977,
+      "诜": 88978,
+      "🐼": 88979,
+      "騢": 88980,
+      "萻": 88981,
+      "黋": 88982,
+      "収": 88983,
+      "湮": 88984,
+      "螥": 88985,
+      "礯": 88986,
+      "鲪": 88987,
+      "哯": 88988,
+      "痟": 88989,
+      "🥅": 88990,
+      "鶙": 88991,
+      "籿": 88992,
+      "娸": 88993,
+      "蒣": 88994,
+      "乫": 88995,
+      "📞": 88996,
+      "⏎": 88997,
+      "鰯": 88998,
+      "漐": 88999,
+      "萘": 89000,
+      "♅": 89001,
+      "繃": 89002,
+      "椕": 89003,
+      "🧘🏽‍♀️": 89004,
+      "绾": 89005,
+      "熤": 89006,
+      "苐": 89007,
+      "衺": 89008,
+      "扢": 89009,
+      "闌": 89010,
+      "赂": 89011,
+      "擋": 89012,
+      "鱭": 89013,
+      "蓆": 89014,
+      "頟": 89015,
+      "瀶": 89016,
+      "髗": 89017,
+      "🦇": 89018,
+      "蜦": 89019,
+      "垐": 89020,
+      "🌸": 89021,
+      "螵": 89022,
+      "釰": 89023,
+      "⋙": 89024,
+      "夛": 89025,
+      "維": 89026,
+      "魮": 89027,
+      "瘶": 89028,
+      "鈋": 89029,
+      "⇺": 89030,
+      "錫": 89031,
+      "👨🏼‍🎤": 89032,
+      "埲": 89033,
+      "舎": 89034,
+      "翈": 89035,
+      "鮃": 89036,
+      "⬜️": 89037,
+      "璲": 89038,
+      "鱶": 89039,
+      "卄": 89040,
+      "惲": 89041,
+      "杵": 89042,
+      "📙": 89043,
+      "涖": 89044,
+      "⏅": 89045,
+      "🧑🏽‍✈️": 89046,
+      "✴️": 89047,
+      "锒": 89048,
+      "🎽": 89049,
+      "臶": 89050,
+      "🤽🏾‍♀️": 89051,
+      "踴": 89052,
+      "🔃": 89053,
+      "鯄": 89054,
+      "硙": 89055,
+      "闟": 89056,
+      "🧑🏽‍🦲": 89057,
+      "臦": 89058,
+      "壎": 89059,
+      "🍰": 89060,
+      "蚱": 89061,
+      "厔": 89062,
+      "簛": 89063,
+      "👩🏽‍💼": 89064,
+      "ⓡ": 89065,
+      "🔴": 89066,
+      "🤶🏿": 89067,
+      "🏝": 89068,
+      "僬": 89069,
+      "殑": 89070,
+      "🤾": 89071,
+      "⑗": 89072,
+      "軧": 89073,
+      "鳴": 89074,
+      "厡": 89075,
+      "炓": 89076,
+      "槁": 89077,
+      "篥": 89078,
+      "🛎": 89079,
+      "鲆": 89080,
+      "哢": 89081,
+      "孧": 89082,
+      "鲬": 89083,
+      "欂": 89084,
+      "鶿": 89085,
+      "竐": 89086,
+      "韲": 89087,
+      "🏮": 89088,
+      "禞": 89089,
+      "哖": 89090,
+      "莛": 89091,
+      "Ũ": 89092,
+      "釲": 89093,
+      "鬿": 89094,
+      "👨‍🎨": 89095,
+      "廱": 89096,
+      "鰎": 89097,
+      "楝": 89098,
+      "闂": 89099,
+      "馮": 89100,
+      "輕": 89101,
+      "喛": 89102,
+      "聧": 89103,
+      "矱": 89104,
+      "簝": 89105,
+      "🈶": 89106,
+      "🫱🏿‍🫲🏼": 89107,
+      "蓊": 89108,
+      "詟": 89109,
+      "蜩": 89110,
+      "💑🏼": 89111,
+      "⏈": 89112,
+      "睅": 89113,
+      "弉": 89114,
+      "惡": 89115,
+      "💁🏻": 89116,
+      "齫": 89117,
+      "鮑": 89118,
+      "🙅🏿‍♀️": 89119,
+      "ˮ": 89120,
+      "黖": 89121,
+      "📥": 89122,
+      "婛": 89123,
+      "💅🏼": 89124,
+      "獞": 89125,
+      "輙": 89126,
+      "⌠": 89127,
+      "鳉": 89128,
+      "穜": 89129,
+      "峆": 89130,
+      "闎": 89131,
+      "✢": 89132,
+      "⎐": 89133,
+      "筨": 89134,
+      "豴": 89135,
+      "ɿ": 89136,
+      "🙋🏾‍♂️": 89137,
+      "搮": 89138,
+      "櫆": 89139,
+      "👐🏼": 89140,
+      "珧": 89141,
+      "鐎": 89142,
+      "🫱🏿‍🫲🏻": 89143,
+      "馥": 89144,
+      "⍤": 89145,
+      "枺": 89146,
+      "🍖": 89147,
+      "訾": 89148,
+      "趛": 89149,
+      "蔔": 89150,
+      "鰨": 89151,
+      "敃": 89152,
+      "圜": 89153,
+      "姮": 89154,
+      "釘": 89155,
+      "☌": 89156,
+      "輄": 89157,
+      "書": 89158,
+      "憢": 89159,
+      "👨‍👧‍👦": 89160,
+      "曩": 89161,
+      "₅": 89162,
+      "♇": 89163,
+      "羠": 89164,
+      "戝": 89165,
+      "刪": 89166,
+      "🧗‍♀️": 89167,
+      "坭": 89168,
+      "棃": 89169,
+      "📛": 89170,
+      "鼙": 89171,
+      "嬞": 89172,
+      "讛": 89173,
+      "🏃🏼‍♀️": 89174,
+      "铏": 89175,
+      "纭": 89176,
+      "🧑🏼‍🦼": 89177,
+      "🧓": 89178,
+      "跡": 89179,
+      "瑎": 89180,
+      "騨": 89181,
+      "😏": 89182,
+      "蒡": 89183,
+      "🧭": 89184,
+      "鏹": 89185,
+      "逌": 89186,
+      "倭": 89187,
+      "禊": 89188,
+      "璺": 89189,
+      "🚣🏻": 89190,
+      "蚞": 89191,
+      "🧑‍🦰": 89192,
+      "瓁": 89193,
+      "🪐": 89194,
+      "🙍": 89195,
+      "譮": 89196,
+      "禌": 89197,
+      "⒡": 89198,
+      "舵": 89199,
+      "釐": 89200,
+      "⌀": 89201,
+      "‬": 89202,
+      "⅞": 89203,
+      "Ⅾ": 89204,
+      "虈": 89205,
+      "鴛": 89206,
+      "檁": 89207,
+      "劔": 89208,
+      "烉": 89209,
+      "鎸": 89210,
+      "蔦": 89211,
+      "👮🏾‍♀️": 89212,
+      "窣": 89213,
+      "👏🏿": 89214,
+      "": 89215,
+      "🦕": 89216,
+      "鎦": 89217,
+      "🫱🏿": 89218,
+      "樿": 89219,
+      "鎩": 89220,
+      "╝": 89221,
+      "玕": 89222,
+      "嫧": 89223,
+      "鐆": 89224,
+      "汔": 89225,
+      "瘱": 89226,
+      "鲥": 89227,
+      "勛": 89228,
+      "🧥": 89229,
+      "邫": 89230,
+      "蓺": 89231,
+      "銇": 89232,
+      "🥍": 89233,
+      "琡": 89234,
+      "👩🏿‍❤️‍💋‍👩🏽": 89235,
+      "⛠": 89236,
+      "馞": 89237,
+      "粌": 89238,
+      "鯍": 89239,
+      "嚰": 89240,
+      "腒": 89241,
+      "蚒": 89242,
+      "鷍": 89243,
+      "🕔": 89244,
+      "🧖🏿‍♀️": 89245,
+      "誤": 89246,
+      "酎": 89247,
+      "🧚🏿": 89248,
+      "🛹": 89249,
+      "5️⃣": 89250,
+      "🇹🇳": 89251,
+      "珪": 89252,
+      "钨": 89253,
+      "迋": 89254,
+      "圎": 89255,
+      "薧": 89256,
+      "🚗": 89257,
+      "渇": 89258,
+      "玧": 89259,
+      "刣": 89260,
+      "锕": 89261,
+      "琑": 89262,
+      "蛶": 89263,
+      "猡": 89264,
+      "縣": 89265,
+      "忈": 89266,
+      "鑊": 89267,
+      "璼": 89268,
+      "🤑": 89269,
+      "🇭🇷": 89270,
+      "ϓ": 89271,
+      "🫱🏽": 89272,
+      "鳔": 89273,
+      "俲": 89274,
+      "弾": 89275,
+      "雓": 89276,
+      "絼": 89277,
+      "賨": 89278,
+      "⚿": 89279,
+      "偪": 89280,
+      "縖": 89281,
+      "堃": 89282,
+      "罵": 89283,
+      "滝": 89284,
+      "🃏": 89285,
+      "浌": 89286,
+      "螅": 89287,
+      "籢": 89288,
+      "踙": 89289,
+      "峚": 89290,
+      "鱳": 89291,
+      "篚": 89292,
+      "惥": 89293,
+      "敼": 89294,
+      "秜": 89295,
+      "⊇": 89296,
+      "鯙": 89297,
+      "🧗🏿‍♀️": 89298,
+      "绔": 89299,
+      "姇": 89300,
+      "黰": 89301,
+      "妶": 89302,
+      "Ï": 89303,
+      "萲": 89304,
+      "餱": 89305,
+      "梩": 89306,
+      "🫱🏼‍🫲🏽": 89307,
+      "🪙": 89308,
+      "铪": 89309,
+      "勈": 89310,
+      "谗": 89311,
+      "枥": 89312,
+      "👨🏿‍⚖️": 89313,
+      "痯": 89314,
+      "🦋": 89315,
+      "芠": 89316,
+      "侫": 89317,
+      "骲": 89318,
+      "螸": 89319,
+      "👨‍🏭": 89320,
+      "👨🏾‍🦳": 89321,
+      "趜": 89322,
+      "鳱": 89323,
+      "颩": 89324,
+      "栞": 89325,
+      "駒": 89326,
+      "嶚": 89327,
+      "貴": 89328,
+      "鰫": 89329,
+      "埯": 89330,
+      "鏏": 89331,
+      "竝": 89332,
+      "寯": 89333,
+      "莮": 89334,
+      "倞": 89335,
+      "墇": 89336,
+      "蟧": 89337,
+      "🛂": 89338,
+      "偰": 89339,
+      "簠": 89340,
+      "攳": 89341,
+      "聃": 89342,
+      "磏": 89343,
+      "⓴": 89344,
+      "賛": 89345,
+      "珛": 89346,
+      "篬": 89347,
+      "憣": 89348,
+      "⎭": 89349,
+      "癶": 89350,
+      "倕": 89351,
+      "騽": 89352,
+      "鶠": 89353,
+      "捩": 89354,
+      "🇨🇿": 89355,
+      "饐": 89356,
+      "♣": 89357,
+      "艴": 89358,
+      "蹚": 89359,
+      "鹡": 89360,
+      "⚣": 89361,
+      "畹": 89362,
+      "𓀅": 89363,
+      "妦": 89364,
+      "逈": 89365,
+      "燔": 89366,
+      "薬": 89367,
+      "艮": 89368,
+      "竆": 89369,
+      "庿": 89370,
+      "晳": 89371,
+      "湣": 89372,
+      "闕": 89373,
+      "≹": 89374,
+      "唳": 89375,
+      "👨🏽‍🦼": 89376,
+      "仌": 89377,
+      "鞬": 89378,
+      "鹗": 89379,
+      "暙": 89380,
+      "🥒": 89381,
+      "鵕": 89382,
+      "藽": 89383,
+      "🤰": 89384,
+      "磟": 89385,
+      "琍": 89386,
+      "𓀂": 89387,
+      "🧑‍🎨": 89388,
+      "賶": 89389,
+      "⌿": 89390,
+      "⚒": 89391,
+      "蠴": 89392,
+      "褰": 89393,
+      "啀": 89394,
+      "撃": 89395,
+      "爿": 89396,
+      "⁺": 89397,
+      "禚": 89398,
+      "Ύ": 89399,
+      "谽": 89400,
+      "ļ": 89401,
+      "猼": 89402,
+      "棧": 89403,
+      "🦹🏼": 89404,
+      "秅": 89405,
+      "薋": 89406,
+      "🇩🇪": 89407,
+      "旟": 89408,
+      "👍🏼": 89409,
+      "💁🏿‍♂️": 89410,
+      "纀": 89411,
+      "锊": 89412,
+      "⇜": 89413,
+      "🇸🇦": 89414,
+      "姎": 89415,
+      "爺": 89416,
+      "濫": 89417,
+      "🧧": 89418,
+      "♶": 89419,
+      "訳": 89420,
+      "坳": 89421,
+      "🫎": 89422,
+      "扚": 89423,
+      "堷": 89424,
+      "谹": 89425,
+      "岹": 89426,
+      "層": 89427,
+      "洩": 89428,
+      "涏": 89429,
+      "玱": 89430,
+      "阩": 89431,
+      "桊": 89432,
+      "唊": 89433,
+      "🙇🏻‍♀️": 89434,
+      "🦹‍♀️": 89435,
+      "舂": 89436,
+      "絯": 89437,
+      "艟": 89438,
+      "棤": 89439,
+      "嬶": 89440,
+      "🧚🏾‍♀️": 89441,
+      "👩🏻‍✈️": 89442,
+      "⚥": 89443,
+      "餌": 89444,
+      "敇": 89445,
+      "姾": 89446,
+      "鞄": 89447,
+      "⚔": 89448,
+      "↽": 89449,
+      "癃": 89450,
+      "👩‍❤️‍💋‍👩": 89451,
+      "歁": 89452,
+      "💇🏼‍♀️": 89453,
+      "ⁿ": 89454,
+      "膋": 89455,
+      "Ê": 89456,
+      "🙍🏼": 89457,
+      "瀉": 89458,
+      "茆": 89459,
+      "🧝🏾‍♂️": 89460,
+      "镄": 89461,
+      "墄": 89462,
+      "冫": 89463,
+      "錟": 89464,
+      "脳": 89465,
+      "貳": 89466,
+      "鹕": 89467,
+      "噋": 89468,
+      "👇🏾": 89469,
+      "喙": 89470,
+      "辯": 89471,
+      "鐨": 89472,
+      "婑": 89473,
+      "錈": 89474,
+      "镻": 89475,
+      "瘅": 89476,
+      "郂": 89477,
+      "⒫": 89478,
+      "Ǚ": 89479,
+      "阦": 89480,
+      "躪": 89481,
+      "🏏": 89482,
+      "妭": 89483,
+      "珎": 89484,
+      "箩": 89485,
+      "弻": 89486,
+      "🇨🇦": 89487,
+      "邕": 89488,
+      "暲": 89489,
+      "闏": 89490,
+      "黷": 89491,
+      "👆🏼": 89492,
+      "熕": 89493,
+      "蒮": 89494,
+      "⑌": 89495,
+      "咟": 89496,
+      "膣": 89497,
+      "鱎": 89498,
+      "彌": 89499,
+      "弪": 89500,
+      "貏": 89501,
+      "脀": 89502,
+      "尌": 89503,
+      "箄": 89504,
+      "寑": 89505,
+      "彑": 89506,
+      "鞁": 89507,
+      "⌓": 89508,
+      "🏒": 89509,
+      "奿": 89510,
+      "⛦": 89511,
+      "婊": 89512,
+      "誳": 89513,
+      "歰": 89514,
+      "葻": 89515,
+      "嵒": 89516,
+      "鄶": 89517,
+      "雬": 89518,
+      "‴": 89519,
+      "颾": 89520,
+      "菨": 89521,
+      "伕": 89522,
+      "攖": 89523,
+      "褸": 89524,
+      "🙋🏿‍♀️": 89525,
+      "羗": 89526,
+      "邴": 89527,
+      "🥯": 89528,
+      "旞": 89529,
+      "铕": 89530,
+      "赼": 89531,
+      "👩🏾": 89532,
+      "憍": 89533,
+      "鍃": 89534,
+      "🙇🏼": 89535,
+      "朶": 89536,
+      "嘊": 89537,
+      "縵": 89538,
+      "楾": 89539,
+      "疞": 89540,
+      "巿": 89541,
+      "👩🏽‍🦽": 89542,
+      "ɥ": 89543,
+      "钽": 89544,
+      "尠": 89545,
+      "躭": 89546,
+      "鎈": 89547,
+      "惌": 89548,
+      "嬆": 89549,
+      "証": 89550,
+      "🌚": 89551,
+      "溮": 89552,
+      "🏄🏿‍♀️": 89553,
+      "岖": 89554,
+      "長": 89555,
+      "遙": 89556,
+      "稌": 89557,
+      "⍾": 89558,
+      "琞": 89559,
+      "♿": 89560,
+      "𓀆": 89561,
+      "鋄": 89562,
+      "⊭": 89563,
+      "歈": 89564,
+      "壭": 89565,
+      "埊": 89566,
+      "诿": 89567,
+      "绫": 89568,
+      "讘": 89569,
+      "ƞ": 89570,
+      "鉜": 89571,
+      "🤲🏿": 89572,
+      "蓒": 89573,
+      "殈": 89574,
+      "檶": 89575,
+      "忸": 89576,
+      "尣": 89577,
+      "犂": 89578,
+      "偋": 89579,
+      "泆": 89580,
+      "薓": 89581,
+      "悭": 89582,
+      "輱": 89583,
+      "📸": 89584,
+      "帔": 89585,
+      "↰": 89586,
+      "‗": 89587,
+      "尮": 89588,
+      "喪": 89589,
+      "🙆🏿‍♂️": 89590,
+      "Œ": 89591,
+      "榘": 89592,
+      "📐": 89593,
+      "矇": 89594,
+      "傗": 89595,
+      "軗": 89596,
+      "黐": 89597,
+      "嫗": 89598,
+      "🦓": 89599,
+      "⌺": 89600,
+      "諐": 89601,
+      "鸱": 89602,
+      "癄": 89603,
+      "栭": 89604,
+      "媕": 89605,
+      "峅": 89606,
+      "鷄": 89607,
+      "愌": 89608,
+      "吇": 89609,
+      "謠": 89610,
+      "籄": 89611,
+      "驏": 89612,
+      "額": 89613,
+      "🧛": 89614,
+      "🧙🏼": 89615,
+      "麤": 89616,
+      "傸": 89617,
+      "菪": 89618,
+      "窈": 89619,
+      "籇": 89620,
+      "馳": 89621,
+      "🍔": 89622,
+      "🧝‍♀️": 89623,
+      "鶞": 89624,
+      "怔": 89625,
+      "弆": 89626,
+      "𓀀": 89627,
+      "枃": 89628,
+      "鏠": 89629,
+      "΁": 89630,
+      "砶": 89631,
+      "甯": 89632,
+      "➛": 89633,
+      "貕": 89634,
+      "臅": 89635,
+      "🤷🏽‍♀️": 89636,
+      "ͽ": 89637,
+      "帹": 89638,
+      "耼": 89639,
+      "夞": 89640,
+      "🧻": 89641,
+      "砩": 89642,
+      "殃": 89643,
+      "舸": 89644,
+      "⅙": 89645,
+      "曉": 89646,
+      "₭": 89647,
+      "錰": 89648,
+      "汋": 89649,
+      "橦": 89650,
+      "ɠ": 89651,
+      "罂": 89652,
+      "🙅🏿‍♂️": 89653,
+      "犴": 89654,
+      "懜": 89655,
+      "葽": 89656,
+      "鎑": 89657,
+      "揯": 89658,
+      "趠": 89659,
+      "塹": 89660,
+      "垅": 89661,
+      "˦": 89662,
+      "猊": 89663,
+      "藝": 89664,
+      "🚴🏿‍♂️": 89665,
+      "暬": 89666,
+      "鮙": 89667,
+      "騝": 89668,
+      "嚃": 89669,
+      "蹡": 89670,
+      "瞋": 89671,
+      "嗆": 89672,
+      "🇳🇫": 89673,
+      "⛹🏽": 89674,
+      "脤": 89675,
+      "Ͷ": 89676,
+      "梈": 89677,
+      "蕲": 89678,
+      "👨‍🍳": 89679,
+      "👨🏻‍🦼": 89680,
+      "甦": 89681,
+      "嫒": 89682,
+      "👩🏼‍🌾": 89683,
+      "锖": 89684,
+      "釪": 89685,
+      "🏄🏾‍♂️": 89686,
+      "⚷": 89687,
+      "酞": 89688,
+      "艊": 89689,
+      "樄": 89690,
+      "躎": 89691,
+      "焜": 89692,
+      "👨🏿‍❤️‍💋‍👨🏿": 89693,
+      "朢": 89694,
+      "骒": 89695,
+      "⛹🏼‍♀️": 89696,
+      "呅": 89697,
+      "榣": 89698,
+      "焘": 89699,
+      "犠": 89700,
+      "睩": 89701,
+      "祾": 89702,
+      "👨🏼‍🌾": 89703,
+      "👎🏻": 89704,
+      "⓽": 89705,
+      "笂": 89706,
+      "鞿": 89707,
+      "堎": 89708,
+      "赾": 89709,
+      "ƚ": 89710,
+      "奤": 89711,
+      "臯": 89712,
+      "啗": 89713,
+      "鴨": 89714,
+      "焏": 89715,
+      "寋": 89716,
+      "譜": 89717,
+      "揙": 89718,
+      "粣": 89719,
+      "繋": 89720,
+      "緁": 89721,
+      "🙆🏻‍♂️": 89722,
+      "➖": 89723,
+      "叐": 89724,
+      "🧑🏻‍🦲": 89725,
+      "囎": 89726,
+      "觽": 89727,
+      "硘": 89728,
+      "筆": 89729,
+      "褅": 89730,
+      "🙏🏽": 89731,
+      "垯": 89732,
+      "乬": 89733,
+      "綽": 89734,
+      "豣": 89735,
+      "媰": 89736,
+      "🤦🏼‍♀️": 89737,
+      "巵": 89738,
+      "珓": 89739,
+      "溬": 89740,
+      "穨": 89741,
+      "🚶": 89742,
+      "轺": 89743,
+      "芈": 89744,
+      "迿": 89745,
+      "": 89746,
+      "菄": 89747,
+      "氁": 89748,
+      "學": 89749,
+      "甁": 89750,
+      "跸": 89751,
+      "駂": 89752,
+      "儩": 89753,
+      "铹": 89754,
+      "🦈": 89755,
+      "幾": 89756,
+      "🛗": 89757,
+      "顡": 89758,
+      "犰": 89759,
+      "瀯": 89760,
+      "溙": 89761,
+      "🛕": 89762,
+      "襮": 89763,
+      "絒": 89764,
+      "馯": 89765,
+      "肱": 89766,
+      "菮": 89767,
+      "搝": 89768,
+      "🟠": 89769,
+      "諲": 89770,
+      "菻": 89771,
+      "仜": 89772,
+      "椀": 89773,
+      "骠": 89774,
+      "細": 89775,
+      "褿": 89776,
+      "迕": 89777,
+      "窰": 89778,
+      "瞶": 89779,
+      "餁": 89780,
+      "捓": 89781,
+      "栮": 89782,
+      "樰": 89783,
+      "佯": 89784,
+      "斎": 89785,
+      "菎": 89786,
+      "瘵": 89787,
+      "樳": 89788,
+      "绀": 89789,
+      "┫": 89790,
+      "耵": 89791,
+      "蟶": 89792,
+      "圡": 89793,
+      "≺": 89794,
+      "煇": 89795,
+      "僰": 89796,
+      "呤": 89797,
+      "💆🏻": 89798,
+      "塐": 89799,
+      "ĥ": 89800,
+      "綡": 89801,
+      "監": 89802,
+      "🧑🏼‍❤️‍🧑🏾": 89803,
+      "婀": 89804,
+      "␥": 89805,
+      "錹": 89806,
+      "椓": 89807,
+      "噄": 89808,
+      "🧑🏼‍💼": 89809,
+      "␓": 89810,
+      "濍": 89811,
+      "鲱": 89812,
+      "廞": 89813,
+      "褙": 89814,
+      "帉": 89815,
+      "圇": 89816,
+      "屚": 89817,
+      "☯": 89818,
+      "撱": 89819,
+      "奣": 89820,
+      "🇻🇺": 89821,
+      "ɼ": 89822,
+      "祕": 89823,
+      "͆": 89824,
+      "濂": 89825,
+      "轲": 89826,
+      "鐖": 89827,
+      "😩": 89828,
+      "踀": 89829,
+      "嫭": 89830,
+      "罴": 89831,
+      "崋": 89832,
+      "稱": 89833,
+      "忏": 89834,
+      "睌": 89835,
+      "⃚": 89836,
+      "翖": 89837,
+      "儋": 89838,
+      "⋞": 89839,
+      "隗": 89840,
+      "蘺": 89841,
+      "諕": 89842,
+      "埄": 89843,
+      "‵": 89844,
+      "晅": 89845,
+      "惄": 89846,
+      "謣": 89847,
+      "絩": 89848,
+      "鲴": 89849,
+      "诹": 89850,
+      "湱": 89851,
+      "儆": 89852,
+      "鏴": 89853,
+      "炕": 89854,
+      "胵": 89855,
+      "鴣": 89856,
+      "秲": 89857,
+      "隌": 89858,
+      "賔": 89859,
+      "👨🏾‍🔬": 89860,
+      "🚶🏽‍♂️": 89861,
+      "戨": 89862,
+      "異": 89863,
+      "梱": 89864,
+      "脵": 89865,
+      "臎": 89866,
+      "譁": 89867,
+      "傎": 89868,
+      "昷": 89869,
+      "珔": 89870,
+      "現": 89871,
+      "畣": 89872,
+      "愕": 89873,
+      "墎": 89874,
+      "娈": 89875,
+      "礶": 89876,
+      "挧": 89877,
+      "⊏": 89878,
+      "Ɲ": 89879,
+      "镼": 89880,
+      "阭": 89881,
+      "奃": 89882,
+      "⌧": 89883,
+      "篈": 89884,
+      "憈": 89885,
+      "闳": 89886,
+      "艸": 89887,
+      "恵": 89888,
+      "甕": 89889,
+      "钃": 89890,
+      "苫": 89891,
+      "盰": 89892,
+      "鶻": 89893,
+      "⃝": 89894,
+      "̈́": 89895,
+      "碖": 89896,
+      "🇸🇷": 89897,
+      "💴": 89898,
+      "鯎": 89899,
+      "蕥": 89900,
+      "饚": 89901,
+      "眑": 89902,
+      "📍": 89903,
+      "墧": 89904,
+      "枲": 89905,
+      "▍": 89906,
+      "競": 89907,
+      "🧁": 89908,
+      "磢": 89909,
+      "儂": 89910,
+      "萐": 89911,
+      "馰": 89912,
+      "峯": 89913,
+      "🧑🏽‍🤝‍🧑🏽": 89914,
+      "針": 89915,
+      "龏": 89916,
+      "琎": 89917,
+      "屘": 89918,
+      "鉮": 89919,
+      "痍": 89920,
+      "賣": 89921,
+      "🧘🏾‍♂️": 89922,
+      "脗": 89923,
+      "熖": 89924,
+      "🕛": 89925,
+      "姭": 89926,
+      "柹": 89927,
+      "🟨": 89928,
+      "牁": 89929,
+      "卪": 89930,
+      "逨": 89931,
+      "茞": 89932,
+      "瞢": 89933,
+      "盷": 89934,
+      "🏄‍♀️": 89935,
+      "▕": 89936,
+      "蠪": 89937,
+      "乨": 89938,
+      "咑": 89939,
+      "佦": 89940,
+      "鵪": 89941,
+      "詺": 89942,
+      "钀": 89943,
+      "ÿ": 89944,
+      "蓭": 89945,
+      "橌": 89946,
+      "諯": 89947,
+      "Ĉ": 89948,
+      "鴸": 89949,
+      "厯": 89950,
+      "檜": 89951,
+      "汷": 89952,
+      "騯": 89953,
+      "滜": 89954,
+      "玟": 89955,
+      "肦": 89956,
+      "矐": 89957,
+      "篵": 89958,
+      "綔": 89959,
+      "蘮": 89960,
+      "💺": 89961,
+      "⊯": 89962,
+      "暟": 89963,
+      "砤": 89964,
+      "皊": 89965,
+      "澉": 89966,
+      "羛": 89967,
+      "癯": 89968,
+      "啺": 89969,
+      "👨🏿‍❤️‍👨🏿": 89970,
+      "貍": 89971,
+      "耯": 89972,
+      "🤾🏿‍♀️": 89973,
+      "鋔": 89974,
+      "怺": 89975,
+      "硆": 89976,
+      "缂": 89977,
+      "鞰": 89978,
+      "🐁": 89979,
+      "鞜": 89980,
+      "謔": 89981,
+      "荈": 89982,
+      "蕏": 89983,
+      "檤": 89984,
+      "襞": 89985,
+      "⎶": 89986,
+      "壚": 89987,
+      "ǰ": 89988,
+      "蛎": 89989,
+      "🙆🏻": 89990,
+      "厛": 89991,
+      "蓽": 89992,
+      "👨🏿‍💼": 89993,
+      "瘳": 89994,
+      "箰": 89995,
+      "駩": 89996,
+      "鴼": 89997,
+      "▎": 89998,
+      "🧝🏽": 89999,
+      "瑡": 90000,
+      "🇫🇯": 90001,
+      "🧗‍♂️": 90002,
+      "騘": 90003,
+      "🖐🏽": 90004,
+      "嗰": 90005,
+      "擊": 90006,
+      "ⓐ": 90007,
+      "蜋": 90008,
+      "髬": 90009,
+      "蒇": 90010,
+      "碩": 90011,
+      "麆": 90012,
+      "毧": 90013,
+      "🇬🇦": 90014,
+      "椯": 90015,
+      "鵢": 90016,
+      "珿": 90017,
+      "穒": 90018,
+      "鎶": 90019,
+      "👩🏾‍🚀": 90020,
+      "👨‍👨‍👦": 90021,
+      "⛮": 90022,
+      "🏌🏿‍♂️": 90023,
+      "錛": 90024,
+      "畉": 90025,
+      "踓": 90026,
+      "戲": 90027,
+      "壴": 90028,
+      "蚔": 90029,
+      "鼖": 90030,
+      "🫵🏽": 90031,
+      "軹": 90032,
+      "戁": 90033,
+      "◬": 90034,
+      "🫱🏼": 90035,
+      "雋": 90036,
+      "詎": 90037,
+      "🧞‍♂️": 90038,
+      "⇕": 90039,
+      "鞟": 90040,
+      "踄": 90041,
+      "盂": 90042,
+      "🕺🏾": 90043,
+      "匄": 90044,
+      "滍": 90045,
+      "攃": 90046,
+      "胂": 90047,
+      "↵": 90048,
+      "錁": 90049,
+      "⎨": 90050,
+      "叴": 90051,
+      "酈": 90052,
+      "🏃🏾‍♂️": 90053,
+      "蒌": 90054,
+      "凣": 90055,
+      "瘼": 90056,
+      "👸🏾": 90057,
+      "💏🏽": 90058,
+      "ʖ": 90059,
+      "觧": 90060,
+      "濅": 90061,
+      "⊓": 90062,
+      "廻": 90063,
+      "庹": 90064,
+      "噇": 90065,
+      "棄": 90066,
+      "🐢": 90067,
+      "枛": 90068,
+      "銈": 90069,
+      "蚡": 90070,
+      "夠": 90071,
+      "瞑": 90072,
+      "敜": 90073,
+      "慼": 90074,
+      "蓍": 90075,
+      "菶": 90076,
+      "簋": 90077,
+      "垆": 90078,
+      "荙": 90079,
+      "鄸": 90080,
+      "縅": 90081,
+      "談": 90082,
+      "潟": 90083,
+      "訑": 90084,
+      "枘": 90085,
+      "滒": 90086,
+      "縌": 90087,
+      "萮": 90088,
+      "🏊🏽‍♂️": 90089,
+      "🧜🏻‍♂️": 90090,
+      "黾": 90091,
+      "岓": 90092,
+      "皶": 90093,
+      "舑": 90094,
+      "弇": 90095,
+      "畵": 90096,
+      "💛": 90097,
+      "濋": 90098,
+      "醾": 90099,
+      "竎": 90100,
+      "裯": 90101,
+      "谻": 90102,
+      "🧑🏼‍🏫": 90103,
+      "鈑": 90104,
+      "朾": 90105,
+      "🦸🏽‍♀️": 90106,
+      "糞": 90107,
+      "Ｄ": 90108,
+      "買": 90109,
+      "ȩ": 90110,
+      "ň": 90111,
+      "枀": 90112,
+      "愞": 90113,
+      "🧑🏽": 90114,
+      "恀": 90115,
+      "躓": 90116,
+      "嚖": 90117,
+      "鈣": 90118,
+      "疄": 90119,
+      "媇": 90120,
+      "墸": 90121,
+      "檩": 90122,
+      "🤱🏼": 90123,
+      "棦": 90124,
+      "騬": 90125,
+      "撚": 90126,
+      "滯": 90127,
+      "嬡": 90128,
+      "艔": 90129,
+      "唁": 90130,
+      "琈": 90131,
+      "檞": 90132,
+      "嶪": 90133,
+      "弁": 90134,
+      "蜳": 90135,
+      "鬷": 90136,
+      "⇅": 90137,
+      "獮": 90138,
+      "👼🏾": 90139,
+      "瀵": 90140,
+      "習": 90141,
+      "嶱": 90142,
+      "窇": 90143,
+      "慅": 90144,
+      "錑": 90145,
+      "澶": 90146,
+      "橕": 90147,
+      "Ɋ": 90148,
+      "潸": 90149,
+      "侱": 90150,
+      "襽": 90151,
+      "閨": 90152,
+      "訥": 90153,
+      "茒": 90154,
+      "乥": 90155,
+      "🕵🏻": 90156,
+      "琙": 90157,
+      "🇱🇹": 90158,
+      "🧑🏾‍🤝‍🧑🏾": 90159,
+      "蓎": 90160,
+      "朹": 90161,
+      "咗": 90162,
+      "🤦🏻‍♂️": 90163,
+      "棂": 90164,
+      "闋": 90165,
+      "偦": 90166,
+      "箛": 90167,
+      "銜": 90168,
+      "粚": 90169,
+      "⎃": 90170,
+      "荎": 90171,
+      "偼": 90172,
+      "鵀": 90173,
+      "🏋🏽": 90174,
+      "珝": 90175,
+      "骞": 90176,
+      "郇": 90177,
+      "豕": 90178,
+      "腫": 90179,
+      "賥": 90180,
+      "厓": 90181,
+      "Ǹ": 90182,
+      "胠": 90183,
+      "倏": 90184,
+      "粠": 90185,
+      "赦": 90186,
+      "蔿": 90187,
+      "胭": 90188,
+      "閔": 90189,
+      "餭": 90190,
+      "曚": 90191,
+      "缰": 90192,
+      "歞": 90193,
+      "₰": 90194,
+      "狇": 90195,
+      "👩🏿‍🌾": 90196,
+      "🛀🏼": 90197,
+      "覜": 90198,
+      "顧": 90199,
+      "鶪": 90200,
+      "敊": 90201,
+      "🎌": 90202,
+      "曭": 90203,
+      "胾": 90204,
+      "Ａ": 90205,
+      "劍": 90206,
+      "🧗🏾‍♀️": 90207,
+      "雒": 90208,
+      "隭": 90209,
+      "餦": 90210,
+      "吋": 90211,
+      "🇨🇷": 90212,
+      "👷🏼‍♀️": 90213,
+      "瞝": 90214,
+      "遀": 90215,
+      "🈳": 90216,
+      "筪": 90217,
+      "釯": 90218,
+      "撓": 90219,
+      "暳": 90220,
+      "桬": 90221,
+      "̛": 90222,
+      "刋": 90223,
+      "🖐🏻": 90224,
+      "👷🏼‍♂️": 90225,
+      "刳": 90226,
+      "捖": 90227,
+      "Ʒ": 90228,
+      "廌": 90229,
+      "👸🏻": 90230,
+      "裭": 90231,
+      "嬔": 90232,
+      "鏕": 90233,
+      "懎": 90234,
+      "ģ": 90235,
+      "箵": 90236,
+      "穣": 90237,
+      "馛": 90238,
+      "🤝🏼": 90239,
+      "挘": 90240,
+      "圪": 90241,
+      "觿": 90242,
+      "🐧": 90243,
+      "芏": 90244,
+      "💪🏼": 90245,
+      "焠": 90246,
+      "🙇‍♂️": 90247,
+      "🤚": 90248,
+      "🫶🏻": 90249,
+      "忁": 90250,
+      "窭": 90251,
+      "耺": 90252,
+      "蚥": 90253,
+      "⎜": 90254,
+      "殗": 90255,
+      "抨": 90256,
+      "⎝": 90257,
+      "効": 90258,
+      "唷": 90259,
+      "鹾": 90260,
+      "磱": 90261,
+      "絛": 90262,
+      "俷": 90263,
+      "坔": 90264,
+      "痽": 90265,
+      "骧": 90266,
+      "筌": 90267,
+      "蟲": 90268,
+      "煦": 90269,
+      "攧": 90270,
+      "糤": 90271,
+      "虵": 90272,
+      "滄": 90273,
+      "⅕": 90274,
+      "瀤": 90275,
+      "検": 90276,
+      "👩🏿‍❤️‍💋‍👩🏻": 90277,
+      "灠": 90278,
+      "靌": 90279,
+      "嫆": 90280,
+      "⁊": 90281,
+      "🐦‍⬛": 90282,
+      "⌰": 90283,
+      "鵣": 90284,
+      "淣": 90285,
+      "褃": 90286,
+      "氜": 90287,
+      "瀓": 90288,
+      "💃🏿": 90289,
+      "徹": 90290,
+      "🧑🏽‍🚒": 90291,
+      "俠": 90292,
+      "劀": 90293,
+      "洡": 90294,
+      "🇪🇷": 90295,
+      "娉": 90296,
+      "斘": 90297,
+      "🇨🇽": 90298,
+      "詷": 90299,
+      "埨": 90300,
+      "ǿ": 90301,
+      "⑺": 90302,
+      "踞": 90303,
+      "👨🏾‍🦽": 90304,
+      "垕": 90305,
+      "鍀": 90306,
+      "🧚🏼‍♀️": 90307,
+      "蟄": 90308,
+      "⁜": 90309,
+      "眎": 90310,
+      "瀕": 90311,
+      "暓": 90312,
+      "痺": 90313,
+      "⁻": 90314,
+      "禍": 90315,
+      "谟": 90316,
+      "嶢": 90317,
+      "忨": 90318,
+      "評": 90319,
+      "詙": 90320,
+      "镾": 90321,
+      "證": 90322,
+      "敳": 90323,
+      "殯": 90324,
+      "🚵🏽‍♀️": 90325,
+      "唴": 90326,
+      "灴": 90327,
+      "萷": 90328,
+      "╃": 90329,
+      "💁‍♂️": 90330,
+      "頚": 90331,
+      "🧑🏼‍🎨": 90332,
+      "ɬ": 90333,
+      "👏": 90334,
+      "🦺": 90335,
+      "姁": 90336,
+      "鎋": 90337,
+      "ϒ": 90338,
+      "槗": 90339,
+      "苼": 90340,
+      "鱥": 90341,
+      "韸": 90342,
+      "👩‍🚀": 90343,
+      "竬": 90344,
+      "檙": 90345,
+      "祲": 90346,
+      "碋": 90347,
+      "鎠": 90348,
+      "⍩": 90349,
+      "褑": 90350,
+      "衳": 90351,
+      "👭": 90352,
+      "🖕🏻": 90353,
+      "疇": 90354,
+      "焮": 90355,
+      "丌": 90356,
+      "🧕🏽": 90357,
+      "‶": 90358,
+      "⁓": 90359,
+      "剦": 90360,
+      "獩": 90361,
+      "🏃‍♂️": 90362,
+      "≲": 90363,
+      "蒷": 90364,
+      "穲": 90365,
+      "🎫": 90366,
+      "💇🏻‍♀️": 90367,
+      "啛": 90368,
+      "🏃🏿‍♂️": 90369,
+      "徝": 90370,
+      "広": 90371,
+      "☝🏽": 90372,
+      "権": 90373,
+      "枡": 90374,
+      "麲": 90375,
+      "猯": 90376,
+      "縩": 90377,
+      "鲚": 90378,
+      "⇣": 90379,
+      "Ζ": 90380,
+      "🧖": 90381,
+      "萵": 90382,
+      "⇾": 90383,
+      "鍅": 90384,
+      "瓼": 90385,
+      "麝": 90386,
+      "🧂": 90387,
+      "唕": 90388,
+      "巚": 90389,
+      "蛿": 90390,
+      "麍": 90391,
+      "🦎": 90392,
+      "ⅅ": 90393,
+      "炥": 90394,
+      "瘷": 90395,
+      "🏋️‍♀️": 90396,
+      "徬": 90397,
+      "ɳ": 90398,
+      "噻": 90399,
+      "賍": 90400,
+      "軒": 90401,
+      "☛": 90402,
+      "偸": 90403,
+      "諂": 90404,
+      "軯": 90405,
+      "姴": 90406,
+      "浈": 90407,
+      "蚵": 90408,
+      "🕵🏽": 90409,
+      "嶬": 90410,
+      "詔": 90411,
+      "牅": 90412,
+      "玒": 90413,
+      "扞": 90414,
+      "毪": 90415,
+      "🧎🏾‍♂️": 90416,
+      "Ș": 90417,
+      "⊕": 90418,
+      "罱": 90419,
+      "魇": 90420,
+      "柋": 90421,
+      "橴": 90422,
+      "揌": 90423,
+      "噿": 90424,
+      "⋛": 90425,
+      "🛻": 90426,
+      "宑": 90427,
+      "屓": 90428,
+      "榉": 90429,
+      "鯈": 90430,
+      "苧": 90431,
+      "ⅴ": 90432,
+      "蕂": 90433,
+      "蹽": 90434,
+      "糂": 90435,
+      "🧗🏼": 90436,
+      "ℱ": 90437,
+      "翽": 90438,
+      "☊": 90439,
+      "萿": 90440,
+      "猨": 90441,
+      "👰🏿‍♀️": 90442,
+      "喐": 90443,
+      "🕑": 90444,
+      "鉑": 90445,
+      "苚": 90446,
+      "豽": 90447,
+      "昀": 90448,
+      "👴🏽": 90449,
+      "珜": 90450,
+      "Ɣ": 90451,
+      "搈": 90452,
+      "🏗": 90453,
+      "罊": 90454,
+      "栕": 90455,
+      "鴦": 90456,
+      "硃": 90457,
+      "滙": 90458,
+      "爀": 90459,
+      "垞": 90460,
+      "🚏": 90461,
+      "榗": 90462,
+      "钒": 90463,
+      "勢": 90464,
+      "🪇": 90465,
+      "坹": 90466,
+      "稬": 90467,
+      "📷": 90468,
+      "繄": 90469,
+      "噘": 90470,
+      "👩🏻‍❤️‍💋‍👩🏿": 90471,
+      "鱁": 90472,
+      "碇": 90473,
+      "郶": 90474,
+      "💃🏻": 90475,
+      "Ŕ": 90476,
+      "矯": 90477,
+      "🇦🇩": 90478,
+      "犑": 90479,
+      "洰": 90480,
+      "鄏": 90481,
+      "冊": 90482,
+      "禮": 90483,
+      "咥": 90484,
+      "🏩": 90485,
+      "蟪": 90486,
+      "驉": 90487,
+      "鉿": 90488,
+      "呭": 90489,
+      "掁": 90490,
+      "迮": 90491,
+      "珆": 90492,
+      "傰": 90493,
+      "酹": 90494,
+      "抍": 90495,
+      "锜": 90496,
+      "栱": 90497,
+      "氚": 90498,
+      "豨": 90499,
+      "癴": 90500,
+      "👨🏾‍🎓": 90501,
+      "🏊‍♀️": 90502,
+      "螧": 90503,
+      "ȶ": 90504,
+      "镱": 90505,
+      "👨🏼‍🦲": 90506,
+      "榸": 90507,
+      "蠈": 90508,
+      "圿": 90509,
+      "🙍🏻‍♂️": 90510,
+      "尻": 90511,
+      "瀗": 90512,
+      "蘯": 90513,
+      "鬚": 90514,
+      "哹": 90515,
+      "🔕": 90516,
+      "槶": 90517,
+      "鐘": 90518,
+      "鼺": 90519,
+      "髻": 90520,
+      "糆": 90521,
+      "筅": 90522,
+      "賾": 90523,
+      "嵔": 90524,
+      "👩‍❤️‍👨": 90525,
+      "況": 90526,
+      "慉": 90527,
+      "肓": 90528,
+      "🗼": 90529,
+      "‧": 90530,
+      "鞑": 90531,
+      "徿": 90532,
+      "胛": 90533,
+      "🚵🏾‍♀️": 90534,
+      "穈": 90535,
+      "鷀": 90536,
+      "姯": 90537,
+      "糒": 90538,
+      "秾": 90539,
+      "🛜": 90540,
+      "袣": 90541,
+      "霟": 90542,
+      "椧": 90543,
+      "鮈": 90544,
+      "痻": 90545,
+      "🧛🏼‍♀️": 90546,
+      "縠": 90547,
+      "⎬": 90548,
+      "襠": 90549,
+      "魡": 90550,
+      "崪": 90551,
+      "筼": 90552,
+      "🔅": 90553,
+      "🕰": 90554,
+      "🧑🏽‍❤️‍💋‍🧑🏿": 90555,
+      "🫱🏼‍🫲🏿": 90556,
+      "⃘": 90557,
+      "祔": 90558,
+      "👬🏽": 90559,
+      "懕": 90560,
+      "顀": 90561,
+      "🧷": 90562,
+      "馿": 90563,
+      "劏": 90564,
+      "ⓢ": 90565,
+      "諀": 90566,
+      "🧑🏼‍🚒": 90567,
+      "⭕️": 90568,
+      "窬": 90569,
+      "偣": 90570,
+      "俌": 90571,
+      "蒎": 90572,
+      "縬": 90573,
+      "咠": 90574,
+      "蓯": 90575,
+      "謋": 90576,
+      "臝": 90577,
+      "稯": 90578,
+      "敆": 90579,
+      "熫": 90580,
+      "熻": 90581,
+      "💑🏿": 90582,
+      "巈": 90583,
+      "酆": 90584,
+      "🧑🏼‍🦽": 90585,
+      "臲": 90586,
+      "媭": 90587,
+      "♠": 90588,
+      "圦": 90589,
+      "𓀖": 90590,
+      "懨": 90591,
+      "🥓": 90592,
+      "ȓ": 90593,
+      "🧑🏻‍❤️‍🧑🏼": 90594,
+      "縱": 90595,
+      "魋": 90596,
+      "渮": 90597,
+      "趫": 90598,
+      "嗢": 90599,
+      "搶": 90600,
+      "👨🏽‍🦱": 90601,
+      "👩🏾‍❤️‍💋‍👩🏻": 90602,
+      "瓩": 90603,
+      "滭": 90604,
+      "ⓤ": 90605,
+      "欼": 90606,
+      "鷃": 90607,
+      "毹": 90608,
+      "🧜🏻‍♀️": 90609,
+      "鱚": 90610,
+      "狴": 90611,
+      "嫬": 90612,
+      "創": 90613,
+      "鑺": 90614,
+      "͊": 90615,
+      "笅": 90616,
+      "🟰": 90617,
+      "蟬": 90618,
+      "釽": 90619,
+      "Ş": 90620,
+      "鵍": 90621,
+      "垶": 90622,
+      "🧎🏿": 90623,
+      "帱": 90624,
+      "👩🏼‍🦲": 90625,
+      "🤦🏿‍♀️": 90626,
+      "👩🏿‍❤️‍👨🏽": 90627,
+      "窩": 90628,
+      "👴": 90629,
+      "Ţ": 90630,
+      "呔": 90631,
+      "澱": 90632,
+      "艗": 90633,
+      "蝷": 90634,
+      "鲳": 90635,
+      "➞": 90636,
+      "篸": 90637,
+      "媓": 90638,
+      "☐": 90639,
+      "藼": 90640,
+      "鴲": 90641,
+      "鳒": 90642,
+      "🧘🏻‍♀️": 90643,
+      "🏣": 90644,
+      "疍": 90645,
+      "踿": 90646,
+      "捁": 90647,
+      "衴": 90648,
+      "醯": 90649,
+      "鳣": 90650,
+      "穳": 90651,
+      "‡": 90652,
+      "ⓜ": 90653,
+      "鲻": 90654,
+      "Ⅿ": 90655,
+      "詅": 90656,
+      "槮": 90657,
+      "罧": 90658,
+      "Ư": 90659,
+      "耱": 90660,
+      "嬳": 90661,
+      "👨🏿‍🍳": 90662,
+      "呖": 90663,
+      "忟": 90664,
+      "鼩": 90665,
+      "骽": 90666,
+      "🧑‍🏫": 90667,
+      "💇🏽": 90668,
+      "顪": 90669,
+      "絈": 90670,
+      "诤": 90671,
+      "鋵": 90672,
+      "璭": 90673,
+      "煅": 90674,
+      "侢": 90675,
+      "軤": 90676,
+      "剭": 90677,
+      "🇧🇩": 90678,
+      "髆": 90679,
+      "籪": 90680,
+      "𓀏": 90681,
+      "峿": 90682,
+      "👸🏿": 90683,
+      "🔦": 90684,
+      "鏨": 90685,
+      "🏃🏼": 90686,
+      "␸": 90687,
+      "冭": 90688,
+      "蛡": 90689,
+      "誴": 90690,
+      "📠": 90691,
+      "袾": 90692,
+      "罒": 90693,
+      "霳": 90694,
+      "蠇": 90695,
+      "峤": 90696,
+      "髥": 90697,
+      "珯": 90698,
+      "葧": 90699,
+      "🕌": 90700,
+      "嚮": 90701,
+      "憑": 90702,
+      "跚": 90703,
+      "恔": 90704,
+      "匋": 90705,
+      "鉢": 90706,
+      "杦": 90707,
+      "🦻🏻": 90708,
+      "🧑🏽‍🦼": 90709,
+      "鰚": 90710,
+      "👩🏼‍❤️‍👩🏿": 90711,
+      "⏄": 90712,
+      "篞": 90713,
+      "夁": 90714,
+      "焲": 90715,
+      "溱": 90716,
+      "炐": 90717,
+      "⏣": 90718,
+      "裞": 90719,
+      "妉": 90720,
+      "髰": 90721,
+      "🏊🏾‍♀️": 90722,
+      "枈": 90723,
+      "躷": 90724,
+      "侹": 90725,
+      "╏": 90726,
+      "憚": 90727,
+      "焍": 90728,
+      "禎": 90729,
+      "詪": 90730,
+      "輧": 90731,
+      "郄": 90732,
+      "⌼": 90733,
+      "穚": 90734,
+      "俳": 90735,
+      "稉": 90736,
+      "桴": 90737,
+      "🔆": 90738,
+      "癠": 90739,
+      "龕": 90740,
+      "棾": 90741,
+      "梃": 90742,
+      "哏": 90743,
+      "垊": 90744,
+      "滖": 90745,
+      "鈻": 90746,
+      "🧒🏼": 90747,
+      "虪": 90748,
+      "ʔ": 90749,
+      "帞": 90750,
+      "欦": 90751,
+      "寽": 90752,
+      "ℰ": 90753,
+      "埕": 90754,
+      "翧": 90755,
+      "伓": 90756,
+      "笜": 90757,
+      "欫": 90758,
+      "椈": 90759,
+      "犢": 90760,
+      "↯": 90761,
+      "鸄": 90762,
+      "絕": 90763,
+      "碢": 90764,
+      "遌": 90765,
+      "鹈": 90766,
+      "鲅": 90767,
+      "搎": 90768,
+      "▚": 90769,
+      "誡": 90770,
+      "⋒": 90771,
+      "∣": 90772,
+      "靣": 90773,
+      "苲": 90774,
+      "烖": 90775,
+      "≱": 90776,
+      "埫": 90777,
+      "揈": 90778,
+      "漼": 90779,
+      "廡": 90780,
+      "漛": 90781,
+      "籝": 90782,
+      "˽": 90783,
+      "魻": 90784,
+      "渉": 90785,
+      "↫": 90786,
+      "浞": 90787,
+      "礇": 90788,
+      "櫩": 90789,
+      "涑": 90790,
+      "庼": 90791,
+      "儾": 90792,
+      "揘": 90793,
+      "攌": 90794,
+      "闝": 90795,
+      "妟": 90796,
+      "敠": 90797,
+      "洬": 90798,
+      "鈢": 90799,
+      "牑": 90800,
+      "漡": 90801,
+      "👩🏻‍🍳": 90802,
+      "⛺": 90803,
+      "跂": 90804,
+      "👮🏼": 90805,
+      "隷": 90806,
+      "廫": 90807,
+      "嗥": 90808,
+      "墛": 90809,
+      "☲": 90810,
+      "筭": 90811,
+      "🔪": 90812,
+      "祗": 90813,
+      "畬": 90814,
+      "🟤": 90815,
+      "⚸": 90816,
+      "鯗": 90817,
+      "焿": 90818,
+      "鉁": 90819,
+      "🍝": 90820,
+      "轪": 90821,
+      "騼": 90822,
+      "萸": 90823,
+      "鎮": 90824,
+      "譓": 90825,
+      "🔩": 90826,
+      "礴": 90827,
+      "頪": 90828,
+      "蚚": 90829,
+      "◮": 90830,
+      "彉": 90831,
+      "楡": 90832,
+      "纕": 90833,
+      "≀": 90834,
+      "換": 90835,
+      "眹": 90836,
+      "厐": 90837,
+      "🌠": 90838,
+      "倢": 90839,
+      "裪": 90840,
+      "彂": 90841,
+      "盫": 90842,
+      "💅🏻": 90843,
+      "🙌": 90844,
+      "鸸": 90845,
+      "ʌ": 90846,
+      "譢": 90847,
+      "🎟": 90848,
+      "▩": 90849,
+      "枵": 90850,
+      "👩🏼‍🍳": 90851,
+      "🇧🇪": 90852,
+      "🚣🏿‍♂️": 90853,
+      "嘝": 90854,
+      "🤦🏽‍♀️": 90855,
+      "≑": 90856,
+      "孃": 90857,
+      "┖": 90858,
+      "猏": 90859,
+      "洔": 90860,
+      "👰🏿": 90861,
+      "嚄": 90862,
+      "磭": 90863,
+      "紒": 90864,
+      "栀": 90865,
+      "ŏ": 90866,
+      "≇": 90867,
+      "🤷🏼": 90868,
+      "🎱": 90869,
+      "湧": 90870,
+      "羚": 90871,
+      "傀": 90872,
+      "磇": 90873,
+      "瑀": 90874,
+      "眗": 90875,
+      "🍻": 90876,
+      "茮": 90877,
+      "僄": 90878,
+      "礹": 90879,
+      "慶": 90880,
+      "龍": 90881,
+      "牤": 90882,
+      "👩🏼‍❤️‍👩🏽": 90883,
+      "俔": 90884,
+      "旻": 90885,
+      "📀": 90886,
+      "掑": 90887,
+      "╳": 90888,
+      "凖": 90889,
+      "嘄": 90890,
+      "肹": 90891,
+      "襚": 90892,
+      "蠆": 90893,
+      "鎚": 90894,
+      "愀": 90895,
+      "粅": 90896,
+      "霧": 90897,
+      "🤷🏻": 90898,
+      "廳": 90899,
+      "🫗": 90900,
+      "燧": 90901,
+      "幊": 90902,
+      "🟡": 90903,
+      "潂": 90904,
+      "脲": 90905,
+      "🛺": 90906,
+      "燅": 90907,
+      "孀": 90908,
+      "無": 90909,
+      "↴": 90910,
+      "🧙🏿‍♂️": 90911,
+      "🐇": 90912,
+      "榓": 90913,
+      "谆": 90914,
+      "窱": 90915,
+      "镊": 90916,
+      "鰘": 90917,
+      "媆": 90918,
+      "璱": 90919,
+      "儨": 90920,
+      "🍐": 90921,
+      "缎": 90922,
+      "澦": 90923,
+      "瞫": 90924,
+      "顐": 90925,
+      "蔝": 90926,
+      "聡": 90927,
+      "🐲": 90928,
+      "馣": 90929,
+      "⚈": 90930,
+      "鼤": 90931,
+      "遳": 90932,
+      "跫": 90933,
+      "🌥": 90934,
+      "銳": 90935,
+      "蕮": 90936,
+      "験": 90937,
+      "擞": 90938,
+      "酄": 90939,
+      "漚": 90940,
+      "逇": 90941,
+      "圶": 90942,
+      "風": 90943,
+      "檖": 90944,
+      "栣": 90945,
+      "勻": 90946,
+      "弖": 90947,
+      "崌": 90948,
+      "🍘": 90949,
+      "晼": 90950,
+      "領": 90951,
+      "獢": 90952,
+      "Ⓠ": 90953,
+      "ʢ": 90954,
+      "🧘🏿‍♀️": 90955,
+      "穐": 90956,
+      "鴕": 90957,
+      "ɑ": 90958,
+      "ǚ": 90959,
+      "枩": 90960,
+      "瞨": 90961,
+      "圢": 90962,
+      "啱": 90963,
+      "👶🏿": 90964,
+      "溩": 90965,
+      "⍁": 90966,
+      "鐮": 90967,
+      "儀": 90968,
+      "顊": 90969,
+      "✈": 90970,
+      "徢": 90971,
+      "膽": 90972,
+      "🥂": 90973,
+      "黥": 90974,
+      "醘": 90975,
+      "貭": 90976,
+      "↛": 90977,
+      "犤": 90978,
+      "榈": 90979,
+      "詌": 90980,
+      "闛": 90981,
+      "⚾": 90982,
+      "輅": 90983,
+      "娏": 90984,
+      "懀": 90985,
+      "儶": 90986,
+      "抩": 90987,
+      "厑": 90988,
+      "Ƣ": 90989,
+      "Ƴ": 90990,
+      "👩🏾‍🦳": 90991,
+      "闈": 90992,
+      "藁": 90993,
+      "遑": 90994,
+      "溭": 90995,
+      "朌": 90996,
+      "楟": 90997,
+      "ǁ": 90998,
+      "◜": 90999,
+      "鷙": 91000,
+      "艁": 91001,
+      "矕": 91002,
+      "漰": 91003,
+      "ⓙ": 91004,
+      "仛": 91005,
+      "🚰": 91006,
+      "⛭": 91007,
+      "鄔": 91008,
+      "霤": 91009,
+      "緞": 91010,
+      "栚": 91011,
+      "爴": 91012,
+      "🇷🇸": 91013,
+      "虦": 91014,
+      "🚾": 91015,
+      "🇧🇮": 91016,
+      "␉": 91017,
+      "燷": 91018,
+      "疟": 91019,
+      "👨🏼‍🔧": 91020,
+      "嶔": 91021,
+      "瞏": 91022,
+      "⊠": 91023,
+      "擲": 91024,
+      "🤹🏾‍♂️": 91025,
+      "鶣": 91026,
+      "濥": 91027,
+      "鰅": 91028,
+      "潫": 91029,
+      "徎": 91030,
+      "朖": 91031,
+      "⛹": 91032,
+      "渧": 91033,
+      "暒": 91034,
+      "蠝": 91035,
+      "櫕": 91036,
+      "霷": 91037,
+      "秕": 91038,
+      "💉": 91039,
+      "嶭": 91040,
+      "鱷": 91041,
+      "鉞": 91042,
+      "傷": 91043,
+      "⊎": 91044,
+      "狿": 91045,
+      "媻": 91046,
+      "薅": 91047,
+      "鶚": 91048,
+      "氖": 91049,
+      "甞": 91050,
+      "貞": 91051,
+      "燠": 91052,
+      "絁": 91053,
+      "詬": 91054,
+      "👨🏿": 91055,
+      "簄": 91056,
+      "駊": 91057,
+      "娼": 91058,
+      "🎮": 91059,
+      "虗": 91060,
+      "≻": 91061,
+      "柶": 91062,
+      "🧑🏼‍🍳": 91063,
+      "诌": 91064,
+      "⇦": 91065,
+      "呓": 91066,
+      "嵥": 91067,
+      "佹": 91068,
+      "湙": 91069,
+      "‪": 91070,
+      "螽": 91071,
+      "桙": 91072,
+      "🧲": 91073,
+      "˴": 91074,
+      "豭": 91075,
+      "鯺": 91076,
+      "戹": 91077,
+      "琕": 91078,
+      "灕": 91079,
+      "劑": 91080,
+      "ɤ": 91081,
+      "褏": 91082,
+      "陿": 91083,
+      "饧": 91084,
+      "嬈": 91085,
+      "⋮": 91086,
+      "腍": 91087,
+      "叺": 91088,
+      "̨": 91089,
+      "🫃🏻": 91090,
+      "譯": 91091,
+      "豋": 91092,
+      "痁": 91093,
+      "毷": 91094,
+      "骉": 91095,
+      "穃": 91096,
+      "◢": 91097,
+      "篗": 91098,
+      "🧑🏽‍🏭": 91099,
+      "Ų": 91100,
+      "鶏": 91101,
+      "𓀕": 91102,
+      "軴": 91103,
+      "鄰": 91104,
+      "軈": 91105,
+      "🍥": 91106,
+      "撗": 91107,
+      "𓀃": 91108,
+      "👨🏿‍🎨": 91109,
+      "鸕": 91110,
+      "貋": 91111,
+      "佇": 91112,
+      "辳": 91113,
+      "椖": 91114,
+      "閂": 91115,
+      "☖": 91116,
+      "兠": 91117,
+      "旫": 91118,
+      "嵰": 91119,
+      "饒": 91120,
+      "蕯": 91121,
+      "͟": 91122,
+      "艡": 91123,
+      "阷": 91124,
+      "趎": 91125,
+      "磍": 91126,
+      "弎": 91127,
+      "蘹": 91128,
+      "⌮": 91129,
+      "娂": 91130,
+      "澻": 91131,
+      "稇": 91132,
+      "檡": 91133,
+      "葙": 91134,
+      "剠": 91135,
+      "坅": 91136,
+      "顆": 91137,
+      "鉆": 91138,
+      "⇼": 91139,
+      "預": 91140,
+      "綨": 91141,
+      "滫": 91142,
+      "逜": 91143,
+      "ɘ": 91144,
+      "頖": 91145,
+      "錸": 91146,
+      "鏿": 91147,
+      "暎": 91148,
+      "👨🏼‍🏫": 91149,
+      "屝": 91150,
+      "蟽": 91151,
+      "帄": 91152,
+      "🔠": 91153,
+      "顱": 91154,
+      "🧑🏿‍🦲": 91155,
+      "跱": 91156,
+      "痏": 91157,
+      "鱩": 91158,
+      "堁": 91159,
+      "朮": 91160,
+      "仫": 91161,
+      "蘜": 91162,
+      "涣": 91163,
+      "鑋": 91164,
+      "絪": 91165,
+      "┉": 91166,
+      "撵": 91167,
+      "乭": 91168,
+      "👩🏽‍🎤": 91169,
+      "🏌️": 91170,
+      "菆": 91171,
+      "嘇": 91172,
+      "😗": 91173,
+      "丏": 91174,
+      "眣": 91175,
+      "🏄🏾‍♀️": 91176,
+      "👷🏾‍♂️": 91177,
+      "齎": 91178,
+      "馃": 91179,
+      "雈": 91180,
+      "凟": 91181,
+      "緖": 91182,
+      "⎙": 91183,
+      "鹬": 91184,
+      "麋": 91185,
+      "Ⓣ": 91186,
+      "斳": 91187,
+      "穕": 91188,
+      "橅": 91189,
+      "銓": 91190,
+      "烬": 91191,
+      "龤": 91192,
+      "ƣ": 91193,
+      "咼": 91194,
+      "顔": 91195,
+      "鑌": 91196,
+      "癹": 91197,
+      "籐": 91198,
+      "哽": 91199,
+      "製": 91200,
+      "燣": 91201,
+      "⛹🏾‍♂️": 91202,
+      "镺": 91203,
+      "鏧": 91204,
+      "穂": 91205,
+      "↮": 91206,
+      "🔂": 91207,
+      "👈": 91208,
+      "🏚": 91209,
+      "懰": 91210,
+      "戭": 91211,
+      "岝": 91212,
+      "🚑": 91213,
+      "毩": 91214,
+      "暔": 91215,
+      "✭": 91216,
+      "👨🏼‍🦽": 91217,
+      "丮": 91218,
+      "嶘": 91219,
+      "湉": 91220,
+      "屃": 91221,
+      "鵡": 91222,
+      "撡": 91223,
+      "斪": 91224,
+      "吶": 91225,
+      "🥳": 91226,
+      "鵤": 91227,
+      "釨": 91228,
+      "⏆": 91229,
+      "萈": 91230,
+      "髣": 91231,
+      "鈲": 91232,
+      "˹": 91233,
+      "辀": 91234,
+      "蛼": 91235,
+      "ȡ": 91236,
+      "澼": 91237,
+      "賂": 91238,
+      "╺": 91239,
+      "蒃": 91240,
+      "鄪": 91241,
+      "顬": 91242,
+      "👼": 91243,
+      "硵": 91244,
+      "⊦": 91245,
+      "渘": 91246,
+      "缮": 91247,
+      "袩": 91248,
+      "⎯": 91249,
+      "鄯": 91250,
+      "螝": 91251,
+      "纜": 91252,
+      "搦": 91253,
+      "樀": 91254,
+      "枇": 91255,
+      "⍙": 91256,
+      "🙋‍♂️": 91257,
+      "唲": 91258,
+      "箅": 91259,
+      "ɉ": 91260,
+      "💁🏾‍♂️": 91261,
+      "Ƽ": 91262,
+      "訄": 91263,
+      "🚈": 91264,
+      "卣": 91265,
+      "迃": 91266,
+      "蘾": 91267,
+      "╚": 91268,
+      "哤": 91269,
+      "壥": 91270,
+      "袔": 91271,
+      "赹": 91272,
+      "媠": 91273,
+      "👨🏽‍🍳": 91274,
+      "皭": 91275,
+      "鋷": 91276,
+      "嘯": 91277,
+      "錢": 91278,
+      "碿": 91279,
+      "🧑🏼‍🌾": 91280,
+      "遤": 91281,
+      "🚣🏻‍♀️": 91282,
+      "闦": 91283,
+      "蚅": 91284,
+      "聿": 91285,
+      "哛": 91286,
+      "刍": 91287,
+      "␐": 91288,
+      "⌅": 91289,
+      "🇲🇪": 91290,
+      "荕": 91291,
+      "鏌": 91292,
+      "餻": 91293,
+      "佡": 91294,
+      "🤷🏽‍♂️": 91295,
+      "怉": 91296,
+      "琊": 91297,
+      "珺": 91298,
+      "🐍": 91299,
+      "鎬": 91300,
+      "恷": 91301,
+      "魝": 91302,
+      "⍿": 91303,
+      "🫳🏿": 91304,
+      "秺": 91305,
+      "👨🏽‍❤️‍👨🏿": 91306,
+      "≃": 91307,
+      "巠": 91308,
+      "🦶🏻": 91309,
+      "籚": 91310,
+      "餽": 91311,
+      "倃": 91312,
+      "襙": 91313,
+      "俾": 91314,
+      "僪": 91315,
+      "殸": 91316,
+      "揹": 91317,
+      "眇": 91318,
+      "侙": 91319,
+      "槓": 91320,
+      "淢": 91321,
+      "爤": 91322,
+      "遰": 91323,
+      "岰": 91324,
+      "漻": 91325,
+      "ȵ": 91326,
+      "谧": 91327,
+      "瑯": 91328,
+      "聗": 91329,
+      "妡": 91330,
+      "詂": 91331,
+      "🇫🇴": 91332,
+      "👩🏻‍❤️‍👩🏻": 91333,
+      "繀": 91334,
+      "嗪": 91335,
+      "🧑🏼‍🎤": 91336,
+      "軞": 91337,
+      "∷": 91338,
+      "痮": 91339,
+      "觬": 91340,
+      "炪": 91341,
+      "唩": 91342,
+      "蒝": 91343,
+      "ˬ": 91344,
+      "😷": 91345,
+      "骫": 91346,
+      "颣": 91347,
+      "拁": 91348,
+      "👩🏿‍🦼": 91349,
+      "蛜": 91350,
+      "驗": 91351,
+      "抯": 91352,
+      "爃": 91353,
+      "柗": 91354,
+      "赿": 91355,
+      "渕": 91356,
+      "鄤": 91357,
+      "⁞": 91358,
+      "霋": 91359,
+      "🧙🏻‍♀️": 91360,
+      "芕": 91361,
+      "🧑🏾‍🍼": 91362,
+      "匽": 91363,
+      "癊": 91364,
+      "猽": 91365,
+      "俛": 91366,
+      "鬛": 91367,
+      "⑊": 91368,
+      "懔": 91369,
+      "稸": 91370,
+      "圮": 91371,
+      "🚶🏼‍♀️": 91372,
+      "👩🏾‍❤️‍💋‍👨🏿": 91373,
+      "疐": 91374,
+      "齄": 91375,
+      "譼": 91376,
+      "鈷": 91377,
+      "🧏🏻‍♂️": 91378,
+      "淊": 91379,
+      "壕": 91380,
+      "₌": 91381,
+      "📳": 91382,
+      "鷇": 91383,
+      "磪": 91384,
+      "谇": 91385,
+      "🇧🇬": 91386,
+      "礥": 91387,
+      "寍": 91388,
+      "👩🏼‍❤️‍💋‍👨🏼": 91389,
+      "⑫": 91390,
+      "扨": 91391,
+      "緼": 91392,
+      "硍": 91393,
+      "摴": 91394,
+      "🧑🏼‍❤️‍💋‍🧑🏻": 91395,
+      "祼": 91396,
+      "廋": 91397,
+      "橗": 91398,
+      "焫": 91399,
+      "竍": 91400,
+      "3️⃣": 91401,
+      "鬰": 91402,
+      "郾": 91403,
+      "齠": 91404,
+      "鐧": 91405,
+      "ⓧ": 91406,
+      "杅": 91407,
+      "蓠": 91408,
+      "戗": 91409,
+      "泲": 91410,
+      "🎍": 91411,
+      "夎": 91412,
+      "觾": 91413,
+      "鼹": 91414,
+      "🧑🏾‍🎄": 91415,
+      "髪": 91416,
+      "鵅": 91417,
+      "镲": 91418,
+      "卝": 91419,
+      "詸": 91420,
+      "聣": 91421,
+      "廼": 91422,
+      "誼": 91423,
+      "棁": 91424,
+      "🈚️": 91425,
+      "箕": 91426,
+      "籆": 91427,
+      "⇁": 91428,
+      "捫": 91429,
+      "🇲🇵": 91430,
+      "賟": 91431,
+      "揝": 91432,
+      "ȕ": 91433,
+      "諊": 91434,
+      "龎": 91435,
+      "駗": 91436,
+      "̺": 91437,
+      "🤼‍♀️": 91438,
+      "蓛": 91439,
+      "🇧🇼": 91440,
+      "🔒": 91441,
+      "⏡": 91442,
+      "愱": 91443,
+      "鄽": 91444,
+      "劚": 91445,
+      "稜": 91446,
+      "洭": 91447,
+      "犄": 91448,
+      "芁": 91449,
+      "漿": 91450,
+      "雐": 91451,
+      "黀": 91452,
+      "賠": 91453,
+      "簦": 91454,
+      "檏": 91455,
+      "箨": 91456,
+      "🏋🏻": 91457,
+      "墕": 91458,
+      "剉": 91459,
+      "😬": 91460,
+      "╨": 91461,
+      "崹": 91462,
+      "蜷": 91463,
+      "↸": 91464,
+      "仈": 91465,
+      "鵵": 91466,
+      "👩🏼‍🦯": 91467,
+      "臆": 91468,
+      "酑": 91469,
+      "洃": 91470,
+      "蓻": 91471,
+      "濻": 91472,
+      "飧": 91473,
+      "⃠": 91474,
+      "🇰🇲": 91475,
+      "皗": 91476,
+      "嗾": 91477,
+      "爏": 91478,
+      "砞": 91479,
+      "刡": 91480,
+      "🏂🏼": 91481,
+      "蛂": 91482,
+      "👩🏿‍❤️‍💋‍👩🏾": 91483,
+      "裡": 91484,
+      "禑": 91485,
+      "🫵🏼": 91486,
+      "廪": 91487,
+      "塯": 91488,
+      "獱": 91489,
+      "嬑": 91490,
+      "⅜": 91491,
+      "镦": 91492,
+      "朘": 91493,
+      "晔": 91494,
+      "淒": 91495,
+      "槹": 91496,
+      "鍶": 91497,
+      "壛": 91498,
+      "厁": 91499,
+      "狪": 91500,
+      "⎫": 91501,
+      "漨": 91502,
+      "蕱": 91503,
+      "鐛": 91504,
+      "竽": 91505,
+      "罘": 91506,
+      "嗛": 91507,
+      "惾": 91508,
+      "🤱": 91509,
+      "✌🏿": 91510,
+      "≳": 91511,
+      "🪪": 91512,
+      "囄": 91513,
+      "謘": 91514,
+      "荬": 91515,
+      "桿": 91516,
+      "鹌": 91517,
+      "巙": 91518,
+      "🇧🇾": 91519,
+      "▝": 91520,
+      "廣": 91521,
+      "👩🏾‍⚖️": 91522,
+      "̚": 91523,
+      "⚵": 91524,
+      "蛘": 91525,
+      "謂": 91526,
+      "舐": 91527,
+      "🕡": 91528,
+      "␊": 91529,
+      "濄": 91530,
+      "貁": 91531,
+      "垰": 91532,
+      "🤌🏿": 91533,
+      "礙": 91534,
+      "鷤": 91535,
+      "錺": 91536,
+      "蠹": 91537,
+      "曔": 91538,
+      "🚵🏼‍♀️": 91539,
+      "丳": 91540,
+      "烸": 91541,
+      "🧑🏿‍❤️‍💋‍🧑🏼": 91542,
+      "👩‍🦯": 91543,
+      "👩🏻‍🦲": 91544,
+      "漌": 91545,
+      "攇": 91546,
+      "瞵": 91547,
+      "躻": 91548,
+      "抏": 91549,
+      "駀": 91550,
+      "┻": 91551,
+      "⁐": 91552,
+      "⃶": 91553,
+      "🕐": 91554,
+      "繣": 91555,
+      "媊": 91556,
+      "🫄🏿": 91557,
+      "歸": 91558,
+      "脦": 91559,
+      "溠": 91560,
+      "鍢": 91561,
+      "Ù": 91562,
+      "弸": 91563,
+      "簽": 91564,
+      "👩🏽‍🦼": 91565,
+      "⃲": 91566,
+      "⃬": 91567,
+      "🧱": 91568,
+      "縍": 91569,
+      "鹓": 91570,
+      "␽": 91571,
+      "遢": 91572,
+      "ȴ": 91573,
+      "徻": 91574,
+      "纞": 91575,
+      "綪": 91576,
+      "氈": 91577,
+      "瘍": 91578,
+      "烲": 91579,
+      "隠": 91580,
+      "奬": 91581,
+      "請": 91582,
+      "稡": 91583,
+      "喥": 91584,
+      "毬": 91585,
+      "鬵": 91586,
+      "▨": 91587,
+      "莳": 91588,
+      "紳": 91589,
+      "酟": 91590,
+      "👷🏾‍♀️": 91591,
+      "萢": 91592,
+      "窴": 91593,
+      "🪠": 91594,
+      "♏": 91595,
+      "凱": 91596,
+      "猻": 91597,
+      "🍹": 91598,
+      "楽": 91599,
+      "疈": 91600,
+      "🇦🇹": 91601,
+      "👩🏾‍❤️‍👩🏼": 91602,
+      "謦": 91603,
+      "葝": 91604,
+      "↎": 91605,
+      "噝": 91606,
+      "🆙": 91607,
+      "☉": 91608,
+      "爛": 91609,
+      "飓": 91610,
+      "礍": 91611,
+      "謞": 91612,
+      "岽": 91613,
+      "⋉": 91614,
+      "🤞🏽": 91615,
+      "荽": 91616,
+      "睷": 91617,
+      "掓": 91618,
+      "🍮": 91619,
+      "猢": 91620,
+      "閛": 91621,
+      "濖": 91622,
+      "👬🏿": 91623,
+      "傼": 91624,
+      "舾": 91625,
+      "薖": 91626,
+      "餘": 91627,
+      "鐈": 91628,
+      "砆": 91629,
+      "😵": 91630,
+      "粋": 91631,
+      "掭": 91632,
+      "靽": 91633,
+      "傃": 91634,
+      "窅": 91635,
+      "螋": 91636,
+      "皣": 91637,
+      "🦷": 91638,
+      "産": 91639,
+      "蝢": 91640,
+      "洠": 91641,
+      "摎": 91642,
+      "⑻": 91643,
+      "̙": 91644,
+      "糽": 91645,
+      "👩🏻‍❤️‍💋‍👨🏼": 91646,
+      "嘾": 91647,
+      "咦": 91648,
+      "絤": 91649,
+      "Έ": 91650,
+      "琮": 91651,
+      "🦸🏼‍♂️": 91652,
+      "攚": 91653,
+      "撷": 91654,
+      "👸🏼": 91655,
+      "馠": 91656,
+      "驝": 91657,
+      "彙": 91658,
+      "菋": 91659,
+      "搄": 91660,
+      "⏘": 91661,
+      "跼": 91662,
+      "萟": 91663,
+      "聅": 91664,
+      "叱": 91665,
+      "躏": 91666,
+      "羈": 91667,
+      "惱": 91668,
+      "躖": 91669,
+      "🇳🇮": 91670,
+      "泙": 91671,
+      "爌": 91672,
+      "趶": 91673,
+      "鹟": 91674,
+      "℥": 91675,
+      "⊀": 91676,
+      "髩": 91677,
+      "偔": 91678,
+      "襣": 91679,
+      "畡": 91680,
+      "靘": 91681,
+      "謱": 91682,
+      "熷": 91683,
+      "薆": 91684,
+      "̹": 91685,
+      "剛": 91686,
+      "涾": 91687,
+      "👩🏻‍❤️‍💋‍👩🏻": 91688,
+      "轛": 91689,
+      "奭": 91690,
+      "餓": 91691,
+      "榲": 91692,
+      "诧": 91693,
+      "廸": 91694,
+      "👨🏽‍🚀": 91695,
+      "憒": 91696,
+      "笭": 91697,
+      "瞹": 91698,
+      "梉": 91699,
+      "◄": 91700,
+      "͓": 91701,
+      "🦹🏿‍♀️": 91702,
+      "婝": 91703,
+      "苵": 91704,
+      "琭": 91705,
+      "姒": 91706,
+      "🏊🏿‍♀️": 91707,
+      "⎟": 91708,
+      "砪": 91709,
+      "⊹": 91710,
+      "滈": 91711,
+      "┃": 91712,
+      "恅": 91713,
+      "獑": 91714,
+      "🇯🇴": 91715,
+      "負": 91716,
+      "甀": 91717,
+      "雊": 91718,
+      "鬍": 91719,
+      "弍": 91720,
+      "娐": 91721,
+      "潤": 91722,
+      "臌": 91723,
+      "詧": 91724,
+      "荏": 91725,
+      "罜": 91726,
+      "坃": 91727,
+      "🧎🏽‍♀️": 91728,
+      "鉛": 91729,
+      "蹹": 91730,
+      "🌧": 91731,
+      "係": 91732,
+      "╰": 91733,
+      "榷": 91734,
+      "猐": 91735,
+      "📤": 91736,
+      "♜": 91737,
+      "徜": 91738,
+      "緐": 91739,
+      "羵": 91740,
+      "✇": 91741,
+      "🧛🏾‍♀️": 91742,
+      "淚": 91743,
+      "⇥": 91744,
+      "聤": 91745,
+      "釛": 91746,
+      "¢": 91747,
+      "潑": 91748,
+      "挔": 91749,
+      "狨": 91750,
+      "吖": 91751,
+      "👩🏽‍🦰": 91752,
+      "錜": 91753,
+      "昋": 91754,
+      "豛": 91755,
+      "犌": 91756,
+      "郈": 91757,
+      "⏓": 91758,
+      "🛀🏿": 91759,
+      "葴": 91760,
+      "👊🏻": 91761,
+      "藟": 91762,
+      "襃": 91763,
+      "襶": 91764,
+      "∙": 91765,
+      "圬": 91766,
+      "觃": 91767,
+      "陊": 91768,
+      "銻": 91769,
+      "痩": 91770,
+      "衸": 91771,
+      "ℎ": 91772,
+      "💏🏾": 91773,
+      "苩": 91774,
+      "🧑🏻‍🦰": 91775,
+      "囨": 91776,
+      "⁆": 91777,
+      "馂": 91778,
+      "仩": 91779,
+      "恦": 91780,
+      "顖": 91781,
+      "愻": 91782,
+      "跍": 91783,
+      "瓒": 91784,
+      "鄚": 91785,
+      "瀨": 91786,
+      "蜾": 91787,
+      "💆": 91788,
+      "🧗🏾‍♂️": 91789,
+      "腢": 91790,
+      "焨": 91791,
+      "🖥": 91792,
+      "": 91793,
+      "縶": 91794,
+      "🍸": 91795,
+      "📈": 91796,
+      "靥": 91797,
+      "鴈": 91798,
+      "薙": 91799,
+      "擮": 91800,
+      "笶": 91801,
+      "🚍": 91802,
+      "蝋": 91803,
+      "🚳": 91804,
+      "撧": 91805,
+      "氩": 91806,
+      "鈛": 91807,
+      "桍": 91808,
+      "蛦": 91809,
+      "盀": 91810,
+      "🫘": 91811,
+      "駹": 91812,
+      "箣": 91813,
+      "🇽🇰": 91814,
+      "绌": 91815,
+      "絲": 91816,
+      "鹵": 91817,
+      "◰": 91818,
+      "隇": 91819,
+      "⚑": 91820,
+      "篍": 91821,
+      "👋🏿": 91822,
+      "🤱🏽": 91823,
+      "襰": 91824,
+      "🇰🇮": 91825,
+      "麼": 91826,
+      "嘢": 91827,
+      "坻": 91828,
+      "蚦": 91829,
+      "玭": 91830,
+      "閣": 91831,
+      "煾": 91832,
+      "蒒": 91833,
+      "🕜": 91834,
+      "勑": 91835,
+      "曧": 91836,
+      "矀": 91837,
+      "蒓": 91838,
+      "濜": 91839,
+      "🇮🇨": 91840,
+      "鑸": 91841,
+      "婇": 91842,
+      "潨": 91843,
+      "佀": 91844,
+      "宺": 91845,
+      "埚": 91846,
+      "轭": 91847,
+      "悪": 91848,
+      "🤏🏽": 91849,
+      "嘹": 91850,
+      "ⅹ": 91851,
+      "≼": 91852,
+      "噥": 91853,
+      "🚸": 91854,
+      "汙": 91855,
+      "訴": 91856,
+      "騣": 91857,
+      "裋": 91858,
+      "镬": 91859,
+      "偅": 91860,
+      "👨🏻‍💻": 91861,
+      "鴷": 91862,
+      "奆": 91863,
+      "躝": 91864,
+      "棈": 91865,
+      "猾": 91866,
+      "芔": 91867,
+      "薶": 91868,
+      "ͯ": 91869,
+      "驥": 91870,
+      "👍🏻": 91871,
+      "鬮": 91872,
+      "繭": 91873,
+      "丆": 91874,
+      "婍": 91875,
+      "匱": 91876,
+      "橋": 91877,
+      "葷": 91878,
+      "👃🏿": 91879,
+      "蟓": 91880,
+      "瑻": 91881,
+      "茔": 91882,
+      "⛹🏻‍♀️": 91883,
+      "毵": 91884,
+      "饀": 91885,
+      "嘰": 91886,
+      "蒆": 91887,
+      "搙": 91888,
+      "🧝🏻": 91889,
+      "澽": 91890,
+      "藸": 91891,
+      "諤": 91892,
+      "儑": 91893,
+      "橀": 91894,
+      "鐪": 91895,
+      "ɣ": 91896,
+      "訠": 91897,
+      "🛠": 91898,
+      "纝": 91899,
+      "💇🏾": 91900,
+      "缑": 91901,
+      "諵": 91902,
+      "爝": 91903,
+      "趓": 91904,
+      "純": 91905,
+      "醰": 91906,
+      "洧": 91907,
+      "啟": 91908,
+      "Ϩ": 91909,
+      "⛞": 91910,
+      "劧": 91911,
+      "🏦": 91912,
+      "↿": 91913,
+      "螒": 91914,
+      "醳": 91915,
+      "📁": 91916,
+      "凨": 91917,
+      "隮": 91918,
+      "驄": 91919,
+      "刞": 91920,
+      "癨": 91921,
+      "黮": 91922,
+      "鎤": 91923,
+      "​": 91924,
+      "⛈": 91925,
+      "誸": 91926,
+      "⌈": 91927,
+      "兪": 91928,
+      "🛰": 91929,
+      "龂": 91930,
+      "拠": 91931,
+      "瀧": 91932,
+      "脰": 91933,
+      "殤": 91934,
+      "齽": 91935,
+      "藡": 91936,
+      "媗": 91937,
+      "⑋": 91938,
+      "🤜🏾": 91939,
+      "盠": 91940,
+      "苻": 91941,
+      "🧔🏼": 91942,
+      "⚩": 91943,
+      "踌": 91944,
+      "簯": 91945,
+      "狦": 91946,
+      "獎": 91947,
+      "蒖": 91948,
+      "🚚": 91949,
+      "🍅": 91950,
+      "🇵🇪": 91951,
+      "怦": 91952,
+      "葲": 91953,
+      "ℒ": 91954,
+      "迀": 91955,
+      "蒊": 91956,
+      "魥": 91957,
+      "鐑": 91958,
+      "鉽": 91959,
+      "镽": 91960,
+      "鳿": 91961,
+      "璛": 91962,
+      "靑": 91963,
+      "亄": 91964,
+      "🐣": 91965,
+      "墺": 91966,
+      "釓": 91967,
+      "擽": 91968,
+      "懑": 91969,
+      "楧": 91970,
+      "輛": 91971,
+      "縘": 91972,
+      "柢": 91973,
+      "躜": 91974,
+      "砊": 91975,
+      "腽": 91976,
+      "賢": 91977,
+      "鏜": 91978,
+      "🏃🏿": 91979,
+      "兕": 91980,
+      "穼": 91981,
+      "嶕": 91982,
+      "🎒": 91983,
+      "烿": 91984,
+      "獵": 91985,
+      "鎵": 91986,
+      "̜": 91987,
+      "☑": 91988,
+      "網": 91989,
+      "👩🏽‍🍼": 91990,
+      "⎔": 91991,
+      "🍱": 91992,
+      "镹": 91993,
+      "灜": 91994,
+      "俁": 91995,
+      "溋": 91996,
+      "咣": 91997,
+      "涚": 91998,
+      "榦": 91999,
+      "鷧": 92000,
+      "钇": 92001,
+      "篔": 92002,
+      "寧": 92003,
+      "⛹️‍♂️": 92004,
+      "礟": 92005,
+      "曈": 92006,
+      "⃑": 92007,
+      "榿": 92008,
+      "馑": 92009,
+      "歽": 92010,
+      "₿": 92011,
+      "謝": 92012,
+      "∖": 92013,
+      "醵": 92014,
+      "偁": 92015,
+      "⏤": 92016,
+      "勭": 92017,
+      "🤞": 92018,
+      "萏": 92019,
+      "飉": 92020,
+      "誺": 92021,
+      "迱": 92022,
+      "襕": 92023,
+      "仧": 92024,
+      "⇿": 92025,
+      "藔": 92026,
+      "弿": 92027,
+      "🖊": 92028,
+      "嗁": 92029,
+      "拫": 92030,
+      "坲": 92031,
+      "澤": 92032,
+      "👳🏿‍♀️": 92033,
+      "疩": 92034,
+      "櫐": 92035,
+      "秫": 92036,
+      "扪": 92037,
+      "茌": 92038,
+      "🏇🏿": 92039,
+      "愍": 92040,
+      "猤": 92041,
+      "觓": 92042,
+      "砜": 92043,
+      "囱": 92044,
+      "鈘": 92045,
+      "🤵🏻": 92046,
+      "⎮": 92047,
+      "⎪": 92048,
+      "扖": 92049,
+      "殙": 92050,
+      "毑": 92051,
+      "豄": 92052,
+      "蛁": 92053,
+      "覓": 92054,
+      "櫡": 92055,
+      "髙": 92056,
+      "竊": 92057,
+      "⊘": 92058,
+      "岀": 92059,
+      "淰": 92060,
+      "箺": 92061,
+      "問": 92062,
+      "鑗": 92063,
+      "衵": 92064,
+      "誑": 92065,
+      "猱": 92066,
+      "憴": 92067,
+      "┙": 92068,
+      "呧": 92069,
+      "恽": 92070,
+      "襘": 92071,
+      "镈": 92072,
+      "蹇": 92073,
+      "葇": 92074,
+      "砉": 92075,
+      "簙": 92076,
+      "懪": 92077,
+      "胇": 92078,
+      "辻": 92079,
+      "鴜": 92080,
+      "浾": 92081,
+      "🗾": 92082,
+      "牼": 92083,
+      "缍": 92084,
+      "⊢": 92085,
+      "鎂": 92086,
+      "礠": 92087,
+      "椺": 92088,
+      "缋": 92089,
+      "溛": 92090,
+      "⎑": 92091,
+      "傕": 92092,
+      "➔": 92093,
+      "刜": 92094,
+      "膾": 92095,
+      "姄": 92096,
+      "髹": 92097,
+      "ↀ": 92098,
+      "🧒🏿": 92099,
+      "瀈": 92100,
+      "擘": 92101,
+      "訷": 92102,
+      "🐫": 92103,
+      "🇲🇰": 92104,
+      "慘": 92105,
+      "娧": 92106,
+      "💪🏿": 92107,
+      "鞢": 92108,
+      "罻": 92109,
+      "Ȥ": 92110,
+      "😌": 92111,
+      "🍣": 92112,
+      "樂": 92113,
+      "皀": 92114,
+      "妐": 92115,
+      "涔": 92116,
+      "牥": 92117,
+      "谔": 92118,
+      "🤹": 92119,
+      "⊣": 92120,
+      "謃": 92121,
+      "冞": 92122,
+      "鹢": 92123,
+      "掾": 92124,
+      "攙": 92125,
+      "堞": 92126,
+      "瓧": 92127,
+      "塊": 92128,
+      "📎": 92129,
+      "鉺": 92130,
+      "颪": 92131,
+      "🎤": 92132,
+      "檧": 92133,
+      "⇮": 92134,
+      "🧛🏾‍♂️": 92135,
+      "Ｗ": 92136,
+      "褤": 92137,
+      "糹": 92138,
+      "╸": 92139,
+      "聩": 92140,
+      "顚": 92141,
+      "疁": 92142,
+      "驨": 92143,
+      "崮": 92144,
+      "軟": 92145,
+      "剫": 92146,
+      "憭": 92147,
+      "跏": 92148,
+      "湬": 92149,
+      "蠀": 92150,
+      "籯": 92151,
+      "褴": 92152,
+      "衇": 92153,
+      "铋": 92154,
+      "📫": 92155,
+      "魳": 92156,
+      "🧛🏽": 92157,
+      "🔗": 92158,
+      "祡": 92159,
+      "猁": 92160,
+      "鎼": 92161,
+      "綴": 92162,
+      "荅": 92163,
+      "魗": 92164,
+      "冄": 92165,
+      "緄": 92166,
+      "þ": 92167,
+      "藮": 92168,
+      "魯": 92169,
+      "璔": 92170,
+      "縰": 92171,
+      "聀": 92172,
+      "楁": 92173,
+      "🫱🏻‍🫲🏿": 92174,
+      "陒": 92175,
+      "Ȳ": 92176,
+      "👩🏿‍🚀": 92177,
+      "₈": 92178,
+      "廐": 92179,
+      "愎": 92180,
+      "彎": 92181,
+      "篽": 92182,
+      "酜": 92183,
+      "瞗": 92184,
+      "挜": 92185,
+      "🧏🏼‍♂️": 92186,
+      "👩🏽": 92187,
+      "嵗": 92188,
+      "👱🏽": 92189,
+      "Ϝ": 92190,
+      "匉": 92191,
+      "≉": 92192,
+      "粭": 92193,
+      "奨": 92194,
+      "̖": 92195,
+      "漺": 92196,
+      "綼": 92197,
+      "綊": 92198,
+      "夽": 92199,
+      "暢": 92200,
+      "🐙": 92201,
+      "圼": 92202,
+      "謶": 92203,
+      "闱": 92204,
+      "塥": 92205,
+      "👩🏼‍🍼": 92206,
+      "嶆": 92207,
+      "倈": 92208,
+      "Ɏ": 92209,
+      "⋰": 92210,
+      "焗": 92211,
+      "莟": 92212,
+      "跽": 92213,
+      "囖": 92214,
+      "坵": 92215,
+      "☶": 92216,
+      "锼": 92217,
+      "👩🏽‍❤️‍💋‍👩🏾": 92218,
+      "嗿": 92219,
+      "乇": 92220,
+      "勍": 92221,
+      "🥸": 92222,
+      "卾": 92223,
+      "蠎": 92224,
+      "嚏": 92225,
+      "亁": 92226,
+      "赈": 92227,
+      "🔇": 92228,
+      "槥": 92229,
+      "牣": 92230,
+      "醜": 92231,
+      "氠": 92232,
+      "挊": 92233,
+      "嶎": 92234,
+      "倴": 92235,
+      "寚": 92236,
+      "魼": 92237,
+      "馽": 92238,
+      "Í": 92239,
+      "崨": 92240,
+      "朣": 92241,
+      "兤": 92242,
+      "🥏": 92243,
+      "栔": 92244,
+      "釥": 92245,
+      "🚹": 92246,
+      "🤰🏿": 92247,
+      "敝": 92248,
+      "鵏": 92249,
+      "皥": 92250,
+      "綹": 92251,
+      "謍": 92252,
+      "蝲": 92253,
+      "牍": 92254,
+      "銱": 92255,
+      "裧": 92256,
+      "髠": 92257,
+      "狶": 92258,
+      "萂": 92259,
+      "敟": 92260,
+      "曂": 92261,
+      "愮": 92262,
+      "駌": 92263,
+      "艤": 92264,
+      "⇹": 92265,
+      "嫎": 92266,
+      "💵": 92267,
+      "蠚": 92268,
+      "嚭": 92269,
+      "訽": 92270,
+      "儰": 92271,
+      "礈": 92272,
+      "🌿": 92273,
+      "瞆": 92274,
+      "嬴": 92275,
+      "嫨": 92276,
+      "糰": 92277,
+      "嫤": 92278,
+      "暷": 92279,
+      "鍴": 92280,
+      "燵": 92281,
+      "狽": 92282,
+      "當": 92283,
+      "骜": 92284,
+      "绠": 92285,
+      "🅿️": 92286,
+      "闙": 92287,
+      "倛": 92288,
+      "粜": 92289,
+      "冡": 92290,
+      "櫤": 92291,
+      "葺": 92292,
+      "觺": 92293,
+      "蜯": 92294,
+      "瑏": 92295,
+      "🐝": 92296,
+      "🛢": 92297,
+      "Ɯ": 92298,
+      "讚": 92299,
+      "恡": 92300,
+      "鄟": 92301,
+      "檚": 92302,
+      "曵": 92303,
+      "👫": 92304,
+      "擭": 92305,
+      "👨🏽‍🦽": 92306,
+      "纄": 92307,
+      "箿": 92308,
+      "鏭": 92309,
+      "鑯": 92310,
+      "╂": 92311,
+      "╔": 92312,
+      "畫": 92313,
+      "猉": 92314,
+      "👨‍🦰": 92315,
+      "蘪": 92316,
+      "汅": 92317,
+      "粞": 92318,
+      "鹲": 92319,
+      "淿": 92320,
+      "⋓": 92321,
+      "韨": 92322,
+      "蛱": 92323,
+      "泎": 92324,
+      "硴": 92325,
+      "漭": 92326,
+      "鴢": 92327,
+      "🚣🏾‍♂️": 92328,
+      "賷": 92329,
+      "笽": 92330,
+      "褾": 92331,
+      "🇵🇰": 92332,
+      "絿": 92333,
+      "俖": 92334,
+      "頠": 92335,
+      "縴": 92336,
+      "鴋": 92337,
+      "夶": 92338,
+      "ⓟ": 92339,
+      "谮": 92340,
+      "碴": 92341,
+      "讹": 92342,
+      "👨‍❤️‍👨": 92343,
+      "穘": 92344,
+      "苡": 92345,
+      "趬": 92346,
+      "菑": 92347,
+      "ⓒ": 92348,
+      "❔": 92349,
+      "彫": 92350,
+      "蛢": 92351,
+      "噁": 92352,
+      "⊚": 92353,
+      "灹": 92354,
+      "Ǡ": 92355,
+      "槤": 92356,
+      "鷒": 92357,
+      "箆": 92358,
+      "趺": 92359,
+      "帓": 92360,
+      "懤": 92361,
+      "詏": 92362,
+      "邋": 92363,
+      "埩": 92364,
+      "摚": 92365,
+      "🫱🏻‍🫲🏼": 92366,
+      "👨🏻‍🎓": 92367,
+      "⋁": 92368,
+      "憌": 92369,
+      "芲": 92370,
+      "镵": 92371,
+      "榰": 92372,
+      "焻": 92373,
+      "澲": 92374,
+      "̸": 92375,
+      "貇": 92376,
+      "🌰": 92377,
+      "熡": 92378,
+      "乚": 92379,
+      "炝": 92380,
+      "♛": 92381,
+      "敕": 92382,
+      "鸋": 92383,
+      "苠": 92384,
+      "畦": 92385,
+      "摮": 92386,
+      "鸔": 92387,
+      "嶶": 92388,
+      "蟺": 92389,
+      "🦵🏼": 92390,
+      "鍱": 92391,
+      "雘": 92392,
+      "👨‍✈️": 92393,
+      "薚": 92394,
+      "龅": 92395,
+      "‍": 92396,
+      "阌": 92397,
+      "ʇ": 92398,
+      "鞻": 92399,
+      "🧑🏾‍🦳": 92400,
+      "⋌": 92401,
+      "靃": 92402,
+      "厹": 92403,
+      "≴": 92404,
+      "臰": 92405,
+      "麄": 92406,
+      "鶷": 92407,
+      "ƶ": 92408,
+      "腚": 92409,
+      "ͧ": 92410,
+      "ℤ": 92411,
+      "盦": 92412,
+      "˸": 92413,
+      "觏": 92414,
+      "👱🏿‍♂️": 92415,
+      "圍": 92416,
+      "蟱": 92417,
+      "蜻": 92418,
+      "🫢": 92419,
+      "魬": 92420,
+      "🪥": 92421,
+      "睧": 92422,
+      "凧": 92423,
+      "垲": 92424,
+      "👩🏾‍💻": 92425,
+      "鄥": 92426,
+      "帎": 92427,
+      "峗": 92428,
+      "🎯": 92429,
+      "篯": 92430,
+      "幀": 92431,
+      "🇨🇨": 92432,
+      "曊": 92433,
+      "粶": 92434,
+      "斕": 92435,
+      "堨": 92436,
+      "◧": 92437,
+      "處": 92438,
+      "胢": 92439,
+      "ȏ": 92440,
+      "🚵": 92441,
+      "奡": 92442,
+      "鴍": 92443,
+      "繲": 92444,
+      "粻": 92445,
+      "弔": 92446,
+      "⑖": 92447,
+      "歓": 92448,
+      "˔": 92449,
+      "🎐": 92450,
+      "碬": 92451,
+      "脻": 92452,
+      "🧗🏻‍♂️": 92453,
+      "鄵": 92454,
+      "僊": 92455,
+      "🫧": 92456,
+      "👷🏻‍♂️": 92457,
+      "暶": 92458,
+      "嗻": 92459,
+      "隴": 92460,
+      "渓": 92461,
+      "蓦": 92462,
+      "襾": 92463,
+      "🫰🏿": 92464,
+      "紬": 92465,
+      "🧑🏾‍💻": 92466,
+      "褧": 92467,
+      "斡": 92468,
+      "杹": 92469,
+      "捄": 92470,
+      "🤾‍♂️": 92471,
+      "­": 92472,
+      "🤚🏼": 92473,
+      "⍟": 92474,
+      "侅": 92475,
+      "ȷ": 92476,
+      "焎": 92477,
+      "煣": 92478,
+      "焷": 92479,
+      "湫": 92480,
+      "尅": 92481,
+      "唅": 92482,
+      "汘": 92483,
+      "峎": 92484,
+      "铛": 92485,
+      "瑃": 92486,
+      "脄": 92487,
+      "颵": 92488,
+      "🇳🇵": 92489,
+      "譭": 92490,
+      "ˤ": 92491,
+      "詘": 92492,
+      "脶": 92493,
+      "盱": 92494,
+      "柍": 92495,
+      "蓰": 92496,
+      "樁": 92497,
+      "📽": 92498,
+      "镆": 92499,
+      "炍": 92500,
+      "灒": 92501,
+      "鋪": 92502,
+      "闄": 92503,
+      "飶": 92504,
+      "撯": 92505,
+      "邤": 92506,
+      "⌞": 92507,
+      "澖": 92508,
+      "😓": 92509,
+      "👨‍💻": 92510,
+      "澭": 92511,
+      "🚶🏾‍♀️": 92512,
+      "厼": 92513,
+      "斵": 92514,
+      "銚": 92515,
+      "噛": 92516,
+      "嶄": 92517,
+      "沠": 92518,
+      "稾": 92519,
+      "勜": 92520,
+      "侞": 92521,
+      "酨": 92522,
+      "敱": 92523,
+      "銗": 92524,
+      "惂": 92525,
+      "蛺": 92526,
+      "睱": 92527,
+      "樛": 92528,
+      "🌶": 92529,
+      "舛": 92530,
+      "Ė": 92531,
+      "剻": 92532,
+      "鍪": 92533,
+      "睖": 92534,
+      "荘": 92535,
+      "蘚": 92536,
+      "饠": 92537,
+      "筄": 92538,
+      "穔": 92539,
+      "嵈": 92540,
+      "煰": 92541,
+      "🦩": 92542,
+      "濴": 92543,
+      "幧": 92544,
+      "磞": 92545,
+      "≋": 92546,
+      "啉": 92547,
+      "ͭ": 92548,
+      "鎾": 92549,
+      "堸": 92550,
+      "鱀": 92551,
+      "ℛ": 92552,
+      "俣": 92553,
+      "険": 92554,
+      "闿": 92555,
+      "🧎‍♀️": 92556,
+      "豩": 92557,
+      "🏂🏾": 92558,
+      "⛅": 92559,
+      "🧘🏽‍♂️": 92560,
+      "詛": 92561,
+      "爁": 92562,
+      "绲": 92563,
+      "刾": 92564,
+      "薦": 92565,
+      "顨": 92566,
+      "暰": 92567,
+      "␾": 92568,
+      "ʀ": 92569,
+      "燻": 92570,
+      "銉": 92571,
+      "綵": 92572,
+      "ȃ": 92573,
+      "郆": 92574,
+      "彁": 92575,
+      "鵩": 92576,
+      "錿": 92577,
+      "΋": 92578,
+      "嶽": 92579,
+      "𓀁": 92580,
+      "🧓🏽": 92581,
+      "🇷🇼": 92582,
+      "嫝": 92583,
+      "伅": 92584,
+      "並": 92585,
+      "繽": 92586,
+      "癩": 92587,
+      "処": 92588,
+      "岊": 92589,
+      "醈": 92590,
+      "繝": 92591,
+      "縢": 92592,
+      "榐": 92593,
+      "斻": 92594,
+      "👨🏽‍🦰": 92595,
+      "🧔🏿": 92596,
+      "⍼": 92597,
+      "琖": 92598,
+      "诟": 92599,
+      "熋": 92600,
+      "脩": 92601,
+      "牒": 92602,
+      "鞏": 92603,
+      "葊": 92604,
+      "瘌": 92605,
+      "👩🏾‍❤️‍💋‍👩🏿": 92606,
+      "◹": 92607,
+      "鎹": 92608,
+      "蜛": 92609,
+      "鐉": 92610,
+      "ʛ": 92611,
+      "⎌": 92612,
+      "蜁": 92613,
+      "銴": 92614,
+      "攺": 92615,
+      "亅": 92616,
+      "酕": 92617,
+      "銁": 92618,
+      "輍": 92619,
+      "嘭": 92620,
+      "槝": 92621,
+      "🤽🏼": 92622,
+      "剺": 92623,
+      "裰": 92624,
+      "賖": 92625,
+      "痦": 92626,
+      "朼": 92627,
+      "圲": 92628,
+      "̇": 92629,
+      "箂": 92630,
+      "黤": 92631,
+      "蹧": 92632,
+      "奩": 92633,
+      "👩🏼‍❤️‍💋‍👩🏽": 92634,
+      "煠": 92635,
+      "🧑🏾‍🏭": 92636,
+      "斺": 92637,
+      "嚘": 92638,
+      "玗": 92639,
+      "🧏🏿": 92640,
+      "蹫": 92641,
+      "狤": 92642,
+      "⌹": 92643,
+      "👨🏿‍🎤": 92644,
+      "⃸": 92645,
+      "␳": 92646,
+      "🧑🏼‍🤝‍🧑🏼": 92647,
+      "姢": 92648,
+      "🦸🏿‍♂️": 92649,
+      "⎊": 92650,
+      "燞": 92651,
+      "🧑🏿‍🎄": 92652,
+      "簲": 92653,
+      "⚞": 92654,
+      "鯥": 92655,
+      "緢": 92656,
+      "👩🏿‍❤️‍👩🏽": 92657,
+      "🌋": 92658,
+      "砠": 92659,
+      "浝": 92660,
+      "Ǒ": 92661,
+      "牷": 92662,
+      "媫": 92663,
+      "庂": 92664,
+      "働": 92665,
+      "毀": 92666,
+      "慛": 92667,
+      "╾": 92668,
+      "⒛": 92669,
+      "鸊": 92670,
+      "罸": 92671,
+      "沱": 92672,
+      "艱": 92673,
+      "辪": 92674,
+      "寤": 92675,
+      "郣": 92676,
+      "ⓣ": 92677,
+      "Ⅹ": 92678,
+      "筓": 92679,
+      "尳": 92680,
+      "砣": 92681,
+      "剱": 92682,
+      "咡": 92683,
+      "嘜": 92684,
+      "靫": 92685,
+      "氥": 92686,
+      "礒": 92687,
+      "嗸": 92688,
+      "💇🏻": 92689,
+      "椪": 92690,
+      "鶱": 92691,
+      "飚": 92692,
+      "🚵🏿": 92693,
+      "鬘": 92694,
+      "👨🏼‍💼": 92695,
+      "🎅🏻": 92696,
+      "⛢": 92697,
+      "霡": 92698,
+      "ɒ": 92699,
+      "霦": 92700,
+      "桲": 92701,
+      "蠌": 92702,
+      "窶": 92703,
+      "髡": 92704,
+      "鼦": 92705,
+      "🦖": 92706,
+      "⌗": 92707,
+      "⓰": 92708,
+      "湲": 92709,
+      "佮": 92710,
+      "鲲": 92711,
+      "埡": 92712,
+      "聍": 92713,
+      "🧑🏻‍🦼": 92714,
+      "碵": 92715,
+      "🇲🇷": 92716,
+      "旹": 92717,
+      "蹮": 92718,
+      "苺": 92719,
+      "圧": 92720,
+      "✋🏻": 92721,
+      "讍": 92722,
+      "璵": 92723,
+      "甃": 92724,
+      "辵": 92725,
+      "偑": 92726,
+      "鉝": 92727,
+      "劎": 92728,
+      "庀": 92729,
+      "諁": 92730,
+      "镗": 92731,
+      "葨": 92732,
+      "搱": 92733,
+      "Ȩ": 92734,
+      "壩": 92735,
+      "堅": 92736,
+      "⏔": 92737,
+      "🧑🏾‍🏫": 92738,
+      "瓵": 92739,
+      "覽": 92740,
+      "椑": 92741,
+      "鼾": 92742,
+      "揕": 92743,
+      "喭": 92744,
+      "瓍": 92745,
+      "蓃": 92746,
+      "躼": 92747,
+      "譗": 92748,
+      "妺": 92749,
+      "➰": 92750,
+      "欤": 92751,
+      "絍": 92752,
+      "≶": 92753,
+      "涽": 92754,
+      "鳯": 92755,
+      "鹮": 92756,
+      "鷺": 92757,
+      "銀": 92758,
+      "矎": 92759,
+      "蝿": 92760,
+      "寭": 92761,
+      "ⅲ": 92762,
+      "˄": 92763,
+      "揥": 92764,
+      "岅": 92765,
+      "曒": 92766,
+      "剝": 92767,
+      "昍": 92768,
+      "👩🏻‍❤️‍💋‍👨🏽": 92769,
+      "禳": 92770,
+      "湸": 92771,
+      "衆": 92772,
+      "庅": 92773,
+      "鶛": 92774,
+      "燌": 92775,
+      "雟": 92776,
+      "🇵🇼": 92777,
+      "℈": 92778,
+      "ũ": 92779,
+      "枤": 92780,
+      "ɛ": 92781,
+      "椵": 92782,
+      "腀": 92783,
+      "猣": 92784,
+      "腵": 92785,
+      "∹": 92786,
+      "╭": 92787,
+      "雼": 92788,
+      "舢": 92789,
+      "坬": 92790,
+      "鋟": 92791,
+      "複": 92792,
+      "℠": 92793,
+      "玃": 92794,
+      "🧍‍♂️": 92795,
+      "👨‍👧‍👧": 92796,
+      "枏": 92797,
+      "囵": 92798,
+      "🥗": 92799,
+      "鳰": 92800,
+      "昽": 92801,
+      "眰": 92802,
+      "秛": 92803,
+      "🇳🇪": 92804,
+      "藗": 92805,
+      "錳": 92806,
+      "繢": 92807,
+      "👩🏿‍🦳": 92808,
+      "銎": 92809,
+      "🎛": 92810,
+      "頢": 92811,
+      "駻": 92812,
+      "👩🏻‍🔧": 92813,
+      "脙": 92814,
+      "溵": 92815,
+      "韚": 92816,
+      "雿": 92817,
+      "砐": 92818,
+      "圊": 92819,
+      "韟": 92820,
+      "劃": 92821,
+      "亍": 92822,
+      "齀": 92823,
+      "└": 92824,
+      "頔": 92825,
+      "鏋": 92826,
+      "倣": 92827,
+      "焛": 92828,
+      "廃": 92829,
+      "鷏": 92830,
+      "薒": 92831,
+      "聺": 92832,
+      "笘": 92833,
+      "氼": 92834,
+      "膼": 92835,
+      "緮": 92836,
+      "⏕": 92837,
+      "狘": 92838,
+      "鈕": 92839,
+      "疘": 92840,
+      "幦": 92841,
+      "👩🏾‍🎤": 92842,
+      "嫍": 92843,
+      "⤵️": 92844,
+      "🫤": 92845,
+      "🦚": 92846,
+      "歶": 92847,
+      "陑": 92848,
+      "玎": 92849,
+      "2️⃣": 92850,
+      "🕳": 92851,
+      "嶨": 92852,
+      "氌": 92853,
+      "⍡": 92854,
+      "瓲": 92855,
+      "Ⅴ": 92856,
+      "櫽": 92857,
+      "枊": 92858,
+      "憠": 92859,
+      "矰": 92860,
+      "╅": 92861,
+      "⁥": 92862,
+      "鐄": 92863,
+      "娺": 92864,
+      "⎱": 92865,
+      "貱": 92866,
+      "🚶🏻": 92867,
+      "Ǭ": 92868,
+      "恛": 92869,
+      "裐": 92870,
+      "媴": 92871,
+      "馟": 92872,
+      "洫": 92873,
+      "垴": 92874,
+      "轎": 92875,
+      "⚏": 92876,
+      "諓": 92877,
+      "頎": 92878,
+      "⊈": 92879,
+      "尟": 92880,
+      "悷": 92881,
+      "ȣ": 92882,
+      "柙": 92883,
+      "讠": 92884,
+      "ↆ": 92885,
+      "挈": 92886,
+      "驣": 92887,
+      "嫋": 92888,
+      "Ǣ": 92889,
+      "範": 92890,
+      "Ｇ": 92891,
+      "🐽": 92892,
+      "🏆": 92893,
+      "碪": 92894,
+      "茭": 92895,
+      "躃": 92896,
+      "鞫": 92897,
+      "鮸": 92898,
+      "隀": 92899,
+      "稐": 92900,
+      "怲": 92901,
+      "😎": 92902,
+      "簁": 92903,
+      "紙": 92904,
+      "楸": 92905,
+      "鏯": 92906,
+      "👩🏻‍🦱": 92907,
+      "譇": 92908,
+      "漋": 92909,
+      "穀": 92910,
+      "Į": 92911,
+      "⏵": 92912,
+      "狟": 92913,
+      "枎": 92914,
+      "湗": 92915,
+      "ɂ": 92916,
+      "♈": 92917,
+      "🚴🏻‍♀️": 92918,
+      "🤵🏽": 92919,
+      "焂": 92920,
+      "惺": 92921,
+      "椙": 92922,
+      "籷": 92923,
+      "緕": 92924,
+      "樦": 92925,
+      "⍮": 92926,
+      "🥣": 92927,
+      "🏋🏻‍♀️": 92928,
+      "镉": 92929,
+      "嘬": 92930,
+      "諉": 92931,
+      "🙅": 92932,
+      "橫": 92933,
+      "鎌": 92934,
+      "彜": 92935,
+      "🔭": 92936,
+      "纒": 92937,
+      "冖": 92938,
+      "囓": 92939,
+      "🤶": 92940,
+      "葟": 92941,
+      "🤵🏾‍♂️": 92942,
+      "砬": 92943,
+      "瞁": 92944,
+      "🇨🇲": 92945,
+      "趇": 92946,
+      "軛": 92947,
+      "👩🏼‍💻": 92948,
+      "檰": 92949,
+      "😄": 92950,
+      "⑏": 92951,
+      "勪": 92952,
+      "鋥": 92953,
+      "帗": 92954,
+      "黝": 92955,
+      "罔": 92956,
+      "儣": 92957,
+      "💃🏽": 92958,
+      "邟": 92959,
+      "牂": 92960,
+      "秣": 92961,
+      "鯢": 92962,
+      "桁": 92963,
+      "ǧ": 92964,
+      "╶": 92965,
+      "饙": 92966,
+      "埙": 92967,
+      "🎄": 92968,
+      "逪": 92969,
+      "罍": 92970,
+      "瘔": 92971,
+      "奪": 92972,
+      "🖐": 92973,
+      "🖕": 92974,
+      "澯": 92975,
+      "蓴": 92976,
+      "韉": 92977,
+      "輷": 92978,
+      "佢": 92979,
+      "挭": 92980,
+      "縋": 92981,
+      "鶋": 92982,
+      "揶": 92983,
+      "⎽": 92984,
+      "膁": 92985,
+      "丂": 92986,
+      "癡": 92987,
+      "胷": 92988,
+      "観": 92989,
+      "抺": 92990,
+      "瘃": 92991,
+      "卍": 92992,
+      "稰": 92993,
+      "🤵🏼": 92994,
+      "敤": 92995,
+      "蕽": 92996,
+      "綎": 92997,
+      "伔": 92998,
+      "訣": 92999,
+      "筩": 93000,
+      "捥": 93001,
+      "磓": 93002,
+      "篓": 93003,
+      "轏": 93004,
+      "顤": 93005,
+      "捛": 93006,
+      "櫊": 93007,
+      "⍓": 93008,
+      "脬": 93009,
+      "锴": 93010,
+      "専": 93011,
+      "揃": 93012,
+      "疛": 93013,
+      "袀": 93014,
+      "∸": 93015,
+      "₱": 93016,
+      "稤": 93017,
+      "熴": 93018,
+      "┣": 93019,
+      "🥢": 93020,
+      "泷": 93021,
+      "⏸": 93022,
+      "Ǟ": 93023,
+      "艣": 93024,
+      "飌": 93025,
+      "🐃": 93026,
+      "Ą": 93027,
+      "诖": 93028,
+      "🇳🇿": 93029,
+      "賬": 93030,
+      "┈": 93031,
+      "👨‍🦯": 93032,
+      "箻": 93033,
+      "玷": 93034,
+      "∲": 93035,
+      "槰": 93036,
+      "膹": 93037,
+      "讄": 93038,
+      "ǣ": 93039,
+      "帚": 93040,
+      "蒠": 93041,
+      "˩": 93042,
+      "焋": 93043,
+      "🔐": 93044,
+      "譎": 93045,
+      "🇹🇱": 93046,
+      "杶": 93047,
+      "燹": 93048,
+      "苒": 93049,
+      "Ɇ": 93050,
+      "🤳🏾": 93051,
+      "鼕": 93052,
+      "價": 93053,
+      "睉": 93054,
+      "僡": 93055,
+      "謀": 93056,
+      "艷": 93057,
+      "🧏🏿‍♀️": 93058,
+      "芗": 93059,
+      "贲": 93060,
+      "螡": 93061,
+      "蕼": 93062,
+      "甥": 93063,
+      "😯": 93064,
+      "≪": 93065,
+      "膮": 93066,
+      "🛀": 93067,
+      "嶍": 93068,
+      "戩": 93069,
+      "鼅": 93070,
+      "蚘": 93071,
+      "趐": 93072,
+      "歨": 93073,
+      "鎏": 93074,
+      "篳": 93075,
+      "🪛": 93076,
+      "佫": 93077,
+      "凈": 93078,
+      "肻": 93079,
+      "🇬🇮": 93080,
+      "∜": 93081,
+      "璯": 93082,
+      "瞌": 93083,
+      "垷": 93084,
+      "♹": 93085,
+      "K": 93086,
+      "翕": 93087,
+      "孅": 93088,
+      "硱": 93089,
+      "👩🏻‍❤️‍👨🏽": 93090,
+      "肵": 93091,
+      "錊": 93092,
+      "👳🏼‍♂️": 93093,
+      "蘥": 93094,
+      "⛛": 93095,
+      "🦶": 93096,
+      "矵": 93097,
+      "蛓": 93098,
+      "栠": 93099,
+      "砎": 93100,
+      "醭": 93101,
+      "嵇": 93102,
+      "瓕": 93103,
+      "汧": 93104,
+      "魸": 93105,
+      "儓": 93106,
+      "尢": 93107,
+      "鍰": 93108,
+      "̃": 93109,
+      "汯": 93110,
+      "讫": 93111,
+      "荇": 93112,
+      "茺": 93113,
+      "躤": 93114,
+      "鮇": 93115,
+      "欞": 93116,
+      "厽": 93117,
+      "🧛🏻‍♀️": 93118,
+      "嚱": 93119,
+      "晇": 93120,
+      "鷜": 93121,
+      "壸": 93122,
+      "鰋": 93123,
+      "彧": 93124,
+      "蜎": 93125,
+      "實": 93126,
+      "鮚": 93127,
+      "Ʊ": 93128,
+      "伖": 93129,
+      "🇧🇿": 93130,
+      "巼": 93131,
+      "🏄🏼‍♂️": 93132,
+      "✧": 93133,
+      "抮": 93134,
+      "瓐": 93135,
+      "箎": 93136,
+      "👰🏽‍♂️": 93137,
+      "蠩": 93138,
+      "諦": 93139,
+      "昹": 93140,
+      "⋹": 93141,
+      "曎": 93142,
+      "哂": 93143,
+      "絆": 93144,
+      "粬": 93145,
+      "鸺": 93146,
+      "繸": 93147,
+      "蕡": 93148,
+      "擕": 93149,
+      "榾": 93150,
+      "👱‍♀️": 93151,
+      "豗": 93152,
+      "👨🏾‍❤️‍👨🏼": 93153,
+      "慹": 93154,
+      " ": 93155,
+      "匁": 93156,
+      "疖": 93157,
+      "谉": 93158,
+      "∬": 93159,
+      "臃": 93160,
+      "芃": 93161,
+      "歗": 93162,
+      "壅": 93163,
+      "寃": 93164,
+      "馱": 93165,
+      "馸": 93166,
+      "瞡": 93167,
+      "蹔": 93168,
+      "耊": 93169,
+      "衁": 93170,
+      "膓": 93171,
+      "🧎🏼‍♂️": 93172,
+      "📊": 93173,
+      "曪": 93174,
+      "鎍": 93175,
+      "🐒": 93176,
+      "🫳🏾": 93177,
+      "鯪": 93178,
+      "┿": 93179,
+      "暺": 93180,
+      "👩🏿‍🎤": 93181,
+      "壆": 93182,
+      "婯": 93183,
+      "ĩ": 93184,
+      "␎": 93185,
+      "側": 93186,
+      "👩🏻‍🎤": 93187,
+      "嘀": 93188,
+      "歐": 93189,
+      "痲": 93190,
+      "洈": 93191,
+      "😥": 93192,
+      "皌": 93193,
+      "🏭": 93194,
+      "隣": 93195,
+      "ⓔ": 93196,
+      "憁": 93197,
+      "僨": 93198,
+      "敄": 93199,
+      "觛": 93200,
+      "忢": 93201,
+      "🤛🏿": 93202,
+      "戽": 93203,
+      "🏇🏽": 93204,
+      "螬": 93205,
+      "冇": 93206,
+      "鏎": 93207,
+      "腤": 93208,
+      "餒": 93209,
+      "👨🏼‍❤️‍👨🏾": 93210,
+      "桄": 93211,
+      "騮": 93212,
+      "榬": 93213,
+      "韠": 93214,
+      "🛌🏽": 93215,
+      "驜": 93216,
+      "耪": 93217,
+      "趉": 93218,
+      "🫄🏼": 93219,
+      "鮠": 93220,
+      "鈇": 93221,
+      "經": 93222,
+      "℅": 93223,
+      "⊟": 93224,
+      "⇫": 93225,
+      "̧": 93226,
+      "袯": 93227,
+      "敾": 93228,
+      "⇢": 93229,
+      "湔": 93230,
+      "箮": 93231,
+      "🤸‍♂️": 93232,
+      "粊": 93233,
+      "選": 93234,
+      "滌": 93235,
+      "纎": 93236,
+      "埖": 93237,
+      "🇬🇺": 93238,
+      "梚": 93239,
+      "咺": 93240,
+      "咜": 93241,
+      "箹": 93242,
+      "⎷": 93243,
+      "🎈": 93244,
+      "夀": 93245,
+      "紦": 93246,
+      "栳": 93247,
+      "🔄": 93248,
+      "厶": 93249,
+      "蝻": 93250,
+      "佭": 93251,
+      "嫕": 93252,
+      "岕": 93253,
+      "唎": 93254,
+      "垗": 93255,
+      "髶": 93256,
+      "🧑🏼‍🦰": 93257,
+      "鬴": 93258,
+      "譛": 93259,
+      "👨‍🌾": 93260,
+      "鱰": 93261,
+      "鈉": 93262,
+      "楐": 93263,
+      "磿": 93264,
+      "坖": 93265,
+      "曄": 93266,
+      "椡": 93267,
+      "鰻": 93268,
+      "裛": 93269,
+      "趿": 93270,
+      "舭": 93271,
+      "⚟": 93272,
+      "瓝": 93273,
+      "贐": 93274,
+      "槳": 93275,
+      "🖇": 93276,
+      "蠑": 93277,
+      "韎": 93278,
+      "嵲": 93279,
+      "繨": 93280,
+      "燑": 93281,
+      "↘": 93282,
+      "鵴": 93283,
+      "橔": 93284,
+      "舧": 93285,
+      "瀅": 93286,
+      "遯": 93287,
+      "👩🏽‍❤️‍👨🏽": 93288,
+      "⎈": 93289,
+      "菛": 93290,
+      "鳗": 93291,
+      "泩": 93292,
+      "慍": 93293,
+      "🇨🇺": 93294,
+      "👮🏾": 93295,
+      "🔺": 93296,
+      "舀": 93297,
+      "浘": 93298,
+      "ʏ": 93299,
+      "叏": 93300,
+      "絠": 93301,
+      "巭": 93302,
+      "⋘": 93303,
+      "啁": 93304,
+      "💁🏽‍♀️": 93305,
+      "崾": 93306,
+      "埢": 93307,
+      "歱": 93308,
+      "藪": 93309,
+      "ℑ": 93310,
+      "堋": 93311,
+      "炁": 93312,
+      "愔": 93313,
+      "Þ": 93314,
+      "氦": 93315,
+      "蹉": 93316,
+      "趃": 93317,
+      "鸛": 93318,
+      "鷓": 93319,
+      "鵌": 93320,
+      "嬎": 93321,
+      "貅": 93322,
+      "妅": 93323,
+      "蚸": 93324,
+      "呠": 93325,
+      "👩🏻‍🔬": 93326,
+      "⛿": 93327,
+      "鶯": 93328,
+      "繞": 93329,
+      "袦": 93330,
+      "🍽": 93331,
+      "澪": 93332,
+      "濹": 93333,
+      "鱧": 93334,
+      "ⓚ": 93335,
+      "脎": 93336,
+      "螰": 93337,
+      "杩": 93338,
+      "毈": 93339,
+      "轾": 93340,
+      "剮": 93341,
+      "👶🏾": 93342,
+      "娪": 93343,
+      "▖": 93344,
+      "題": 93345,
+      "痗": 93346,
+      "種": 93347,
+      "⅝": 93348,
+      "勳": 93349,
+      "躉": 93350,
+      "慟": 93351,
+      "嬵": 93352,
+      "🛏": 93353,
+      "😾": 93354,
+      "∉": 93355,
+      "🧙🏽‍♂️": 93356,
+      "廍": 93357,
+      "🏌️‍♀️": 93358,
+      "兛": 93359,
+      "⊛": 93360,
+      "橥": 93361,
+      "麇": 93362,
+      "飢": 93363,
+      "鄊": 93364,
+      "🤦🏽‍♂️": 93365,
+      "螗": 93366,
+      "🤏🏻": 93367,
+      "🕶": 93368,
+      "榤": 93369,
+      "鼊": 93370,
+      "͙": 93371,
+      "栢": 93372,
+      "腶": 93373,
+      "鱐": 93374,
+      "⏨": 93375,
+      "毸": 93376,
+      "詗": 93377,
+      "🐑": 93378,
+      "🧑🏽‍🦱": 93379,
+      "簘": 93380,
+      "犥": 93381,
+      "⁠": 93382,
+      "┺": 93383,
+      "駫": 93384,
+      "⎣": 93385,
+      "茑": 93386,
+      "👨‍👨‍👧": 93387,
+      "扵": 93388,
+      "👨🏿‍❤️‍👨🏻": 93389,
+      "崡": 93390,
+      "湥": 93391,
+      "竡": 93392,
+      "鋠": 93393,
+      "殺": 93394,
+      "尲": 93395,
+      "踧": 93396,
+      "輆": 93397,
+      "💏": 93398,
+      "醩": 93399,
+      "奝": 93400,
+      "笌": 93401,
+      "菤": 93402,
+      "🧘🏻": 93403,
+      "刱": 93404,
+      "恮": 93405,
+      "芢": 93406,
+      "👨‍🚒": 93407,
+      "峃": 93408,
+      "潱": 93409,
+      "␿": 93410,
+      "Ǎ": 93411,
+      "ϫ": 93412,
+      "俍": 93413,
+      "忄": 93414,
+      "罋": 93415,
+      "澧": 93416,
+      "灊": 93417,
+      "趈": 93418,
+      "襆": 93419,
+      "闆": 93420,
+      "✖️": 93421,
+      "玢": 93422,
+      "ʊ": 93423,
+      "¦": 93424,
+      "揱": 93425,
+      "汈": 93426,
+      "碦": 93427,
+      "鶘": 93428,
+      "勽": 93429,
+      "鍾": 93430,
+      "瑊": 93431,
+      "炔": 93432,
+      "姃": 93433,
+      "宄": 93434,
+      "傺": 93435,
+      "荹": 93436,
+      "镭": 93437,
+      "苝": 93438,
+      "襪": 93439,
+      "潪": 93440,
+      "⛯": 93441,
+      "觼": 93442,
+      "饜": 93443,
+      "訫": 93444,
+      "鵫": 93445,
+      "贘": 93446,
+      "🔔": 93447,
+      "💁🏼‍♂️": 93448,
+      "牻": 93449,
+      "簟": 93450,
+      "塰": 93451,
+      "诬": 93452,
+      "檳": 93453,
+      "鹷": 93454,
+      "耎": 93455,
+      "😘": 93456,
+      "聕": 93457,
+      "儈": 93458,
+      "儲": 93459,
+      "颡": 93460,
+      "饇": 93461,
+      "墍": 93462,
+      "譒": 93463,
+      "腛": 93464,
+      "擪": 93465,
+      "燰": 93466,
+      "🇧🇸": 93467,
+      "爓": 93468,
+      "佨": 93469,
+      "╋": 93470,
+      "锽": 93471,
+      "👩🏽‍❤️‍💋‍👩🏻": 93472,
+      "甂": 93473,
+      "👆🏿": 93474,
+      "🧑🏻‍🦯": 93475,
+      "痼": 93476,
+      "耋": 93477,
+      "劙": 93478,
+      "嘗": 93479,
+      "糥": 93480,
+      "⍔": 93481,
+      "厴": 93482,
+      "桳": 93483,
+      "蜲": 93484,
+      "𓀗": 93485,
+      "溌": 93486,
+      "辇": 93487,
+      "扝": 93488,
+      "爠": 93489,
+      "獥": 93490,
+      "饊": 93491,
+      "媢": 93492,
+      "峍": 93493,
+      "珄": 93494,
+      "╎": 93495,
+      "潈": 93496,
+      "邙": 93497,
+      "梖": 93498,
+      "襔": 93499,
+      "🎴": 93500,
+      "硂": 93501,
+      "鏦": 93502,
+      "詁": 93503,
+      "瑗": 93504,
+      "恏": 93505,
+      "嶃": 93506,
+      "莪": 93507,
+      "₊": 93508,
+      "ₚ": 93509,
+      "坼": 93510,
+      "櫨": 93511,
+      "犃": 93512,
+      "杙": 93513,
+      "兟": 93514,
+      "刈": 93515,
+      "鋚": 93516,
+      "麙": 93517,
+      "🖱": 93518,
+      "伂": 93519,
+      "🏋🏾": 93520,
+      "砻": 93521,
+      "蒅": 93522,
+      "唸": 93523,
+      "妠": 93524,
+      "俅": 93525,
+      "莩": 93526,
+      "珇": 93527,
+      "堇": 93528,
+      "迏": 93529,
+      "蠷": 93530,
+      "毱": 93531,
+      "諍": 93532,
+      "仝": 93533,
+      "哋": 93534,
+      "蒕": 93535,
+      "溨": 93536,
+      "📦": 93537,
+      "💈": 93538,
+      "葥": 93539,
+      "ⅇ": 93540,
+      "◿": 93541,
+      "🤱🏿": 93542,
+      "炃": 93543,
+      "🇰🇿": 93544,
+      "⛜": 93545,
+      "饂": 93546,
+      "▓": 93547,
+      "挋": 93548,
+      "↚": 93549,
+      "👨🏿‍❤️‍👨🏼": 93550,
+      "鯂": 93551,
+      "昄": 93552,
+      "蹻": 93553,
+      "饷": 93554,
+      "呟": 93555,
+      "厎": 93556,
+      "様": 93557,
+      "狳": 93558,
+      "諸": 93559,
+      "涙": 93560,
+      "鴇": 93561,
+      "譿": 93562,
+      "锶": 93563,
+      "🤾🏻‍♀️": 93564,
+      "Ⓨ": 93565,
+      "̑": 93566,
+      "釾": 93567,
+      "搴": 93568,
+      "⃜": 93569,
+      "≫": 93570,
+      "縷": 93571,
+      "蚣": 93572,
+      "肗": 93573,
+      "瞊": 93574,
+      "勼": 93575,
+      "颟": 93576,
+      "締": 93577,
+      "纰": 93578,
+      "弶": 93579,
+      "蹂": 93580,
+      "殧": 93581,
+      "🏅": 93582,
+      "龗": 93583,
+      "棆": 93584,
+      "酴": 93585,
+      "蒏": 93586,
+      "🎾": 93587,
+      "攍": 93588,
+      "⒤": 93589,
+      "楉": 93590,
+      "Ⅎ": 93591,
+      "栊": 93592,
+      "詒": 93593,
+      "鵝": 93594,
+      "凷": 93595,
+      "宐": 93596,
+      "嬰": 93597,
+      "筙": 93598,
+      "🧑🏾‍❤️‍💋‍🧑🏼": 93599,
+      "🙋🏼‍♂️": 93600,
+      "磘": 93601,
+      "赜": 93602,
+      "匇": 93603,
+      "鴰": 93604,
+      "綐": 93605,
+      "偞": 93606,
+      "🧑🏿‍⚕️": 93607,
+      "棏": 93608,
+      "鷕": 93609,
+      "頛": 93610,
+      "倄": 93611,
+      "訶": 93612,
+      "岆": 93613,
+      "悵": 93614,
+      "🤳🏻": 93615,
+      "餫": 93616,
+      "廵": 93617,
+      "惢": 93618,
+      "陓": 93619,
+      "荺": 93620,
+      "🏃🏻‍♀️": 93621,
+      "關": 93622,
+      "毴": 93623,
+      "褷": 93624,
+      "🤛🏽": 93625,
+      "搚": 93626,
+      "甆": 93627,
+      "👫🏼": 93628,
+      "嫙": 93629,
+      "🥨": 93630,
+      "睈": 93631,
+      "婸": 93632,
+      "勅": 93633,
+      "埿": 93634,
+      "妁": 93635,
+      "冹": 93636,
+      "👂": 93637,
+      "麀": 93638,
+      "啂": 93639,
+      "頂": 93640,
+      "绖": 93641,
+      "烒": 93642,
+      "🎀": 93643,
+      "鉹": 93644,
+      "浶": 93645,
+      "覾": 93646,
+      "鈱": 93647,
+      "齹": 93648,
+      "埥": 93649,
+      "🇼🇸": 93650,
+      "🤙🏿": 93651,
+      "啑": 93652,
+      "羃": 93653,
+      "👨🏻‍⚕️": 93654,
+      "鉧": 93655,
+      "訪": 93656,
+      "♍": 93657,
+      "🖋": 93658,
+      "慝": 93659,
+      "💧": 93660,
+      "羳": 93661,
+      "铊": 93662,
+      "≔": 93663,
+      "斱": 93664,
+      "篾": 93665,
+      "逞": 93666,
+      "兣": 93667,
+      "🚣‍♂️": 93668,
+      "蠜": 93669,
+      "圷": 93670,
+      "甤": 93671,
+      "龞": 93672,
+      "⃂": 93673,
+      "吩": 93674,
+      "檲": 93675,
+      "陣": 93676,
+      "珃": 93677,
+      "僥": 93678,
+      "👨🏽‍🦳": 93679,
+      "赓": 93680,
+      "🇮🇲": 93681,
+      "蒍": 93682,
+      "嶅": 93683,
+      "ˁ": 93684,
+      "噓": 93685,
+      "賤": 93686,
+      "🧑🏻‍💻": 93687,
+      "輀": 93688,
+      "🎥": 93689,
+      "峲": 93690,
+      "輳": 93691,
+      "🚵🏻‍♂️": 93692,
+      "傋": 93693,
+      "矌": 93694,
+      "̵": 93695,
+      "錕": 93696,
+      "👮‍♂️": 93697,
+      "颫": 93698,
+      "⚲": 93699,
+      "橐": 93700,
+      "戔": 93701,
+      "栶": 93702,
+      "俹": 93703,
+      "舉": 93704,
+      "昲": 93705,
+      "ℷ": 93706,
+      "婔": 93707,
+      "濟": 93708,
+      "鶅": 93709,
+      "👨‍👨‍👧‍👦": 93710,
+      "脺": 93711,
+      "忿": 93712,
+      "撊": 93713,
+      "诙": 93714,
+      "🧖🏻‍♂️": 93715,
+      "妱": 93716,
+      "👩🏿‍🎓": 93717,
+      "✶": 93718,
+      "👩🏾‍🍳": 93719,
+      "⍅": 93720,
+      "杕": 93721,
+      "岋": 93722,
+      "✎": 93723,
+      "枽": 93724,
+      "dž": 93725,
+      "🧑‍🤝‍🧑": 93726,
+      "嘨": 93727,
+      "怢": 93728,
+      "傆": 93729,
+      "狅": 93730,
+      "輴": 93731,
+      "纙": 93732,
+      "抇": 93733,
+      "劐": 93734,
+      "訤": 93735,
+      "蕬": 93736,
+      "🐆": 93737,
+      "噑": 93738,
+      "玙": 93739,
+      "瞷": 93740,
+      "灣": 93741,
+      "👷🏾": 93742,
+      "槈": 93743,
+      "泒": 93744,
+      "嗚": 93745,
+      "崢": 93746,
+      "秓": 93747,
+      "絹": 93748,
+      "湏": 93749,
+      "΍": 93750,
+      "🥙": 93751,
+      "👯‍♂️": 93752,
+      "▞": 93753,
+      "朷": 93754,
+      "汬": 93755,
+      "壐": 93756,
+      "苣": 93757,
+      "銨": 93758,
+      "╪": 93759,
+      "愛": 93760,
+      "缏": 93761,
+      "焽": 93762,
+      "🤛🏻": 93763,
+      "ʤ": 93764,
+      "仳": 93765,
+      "燐": 93766,
+      "🦞": 93767,
+      "埴": 93768,
+      "繅": 93769,
+      "軳": 93770,
+      "👖": 93771,
+      "ɇ": 93772,
+      "禃": 93773,
+      "̅": 93774,
+      "锝": 93775,
+      "燊": 93776,
+      "舣": 93777,
+      "硪": 93778,
+      "鑃": 93779,
+      "刭": 93780,
+      "舼": 93781,
+      "齙": 93782,
+      "弳": 93783,
+      "渿": 93784,
+      "箉": 93785,
+      "✃": 93786,
+      "揵": 93787,
+      "凾": 93788,
+      "鋯": 93789,
+      "嬕": 93790,
+      "壪": 93791,
+      "灥": 93792,
+      "緣": 93793,
+      "💑🏻": 93794,
+      "箤": 93795,
+      "璣": 93796,
+      "ͦ": 93797,
+      "⁧": 93798,
+      "囲": 93799,
+      "幍": 93800,
+      "₋": 93801,
+      "敮": 93802,
+      "錪": 93803,
+      "熚": 93804,
+      "👵": 93805,
+      "ƛ": 93806,
+      "氫": 93807,
+      "擝": 93808,
+      "⍜": 93809,
+      "撶": 93810,
+      "娷": 93811,
+      "🔷": 93812,
+      "滲": 93813,
+      "搳": 93814,
+      "蠉": 93815,
+      "汣": 93816,
+      "婠": 93817,
+      "💆🏼‍♀️": 93818,
+      "檯": 93819,
+      "鞣": 93820,
+      "絅": 93821,
+      "鎐": 93822,
+      "嫾": 93823,
+      "鋗": 93824,
+      "琸": 93825,
+      "禛": 93826,
+      "幮": 93827,
+      "楄": 93828,
+      "硢": 93829,
+      "觘": 93830,
+      "卭": 93831,
+      "噟": 93832,
+      "緛": 93833,
+      "🈂️": 93834,
+      "虭": 93835,
+      "⅒": 93836,
+      "👅": 93837,
+      "🟢": 93838,
+      "測": 93839,
+      "翦": 93840,
+      "畞": 93841,
+      "🏪": 93842,
+      "蒚": 93843,
+      "🥿": 93844,
+      "🐐": 93845,
+      "瘲": 93846,
+      "扈": 93847,
+      "坸": 93848,
+      "痤": 93849,
+      "礿": 93850,
+      "階": 93851,
+      "聁": 93852,
+      "豐": 93853,
+      "繉": 93854,
+      "羘": 93855,
+      "澀": 93856,
+      "銝": 93857,
+      "扐": 93858,
+      "Ϊ": 93859,
+      "犨": 93860,
+      "洘": 93861,
+      "💐": 93862,
+      "峳": 93863,
+      "櫻": 93864,
+      "枼": 93865,
+      "庳": 93866,
+      "僕": 93867,
+      "🥔": 93868,
+      "勶": 93869,
+      "⚨": 93870,
+      "馊": 93871,
+      "暪": 93872,
+      "瘓": 93873,
+      "▐": 93874,
+      "溥": 93875,
+      "邠": 93876,
+      "媐": 93877,
+      "Ĺ": 93878,
+      "誅": 93879,
+      "铌": 93880,
+      "💆🏻‍♂️": 93881,
+      "👨‍🏫": 93882,
+      "簶": 93883,
+      "傮": 93884,
+      "👨🏾‍✈️": 93885,
+      "💃🏾": 93886,
+      "捔": 93887,
+      "銹": 93888,
+      "👫🏻": 93889,
+      "👊🏾": 93890,
+      "囮": 93891,
+      "🍆": 93892,
+      "糨": 93893,
+      "鼶": 93894,
+      "鴥": 93895,
+      "剘": 93896,
+      "語": 93897,
+      "鹨": 93898,
+      "蓔": 93899,
+      "͵": 93900,
+      "舴": 93901,
+      "🕵🏽‍♂️": 93902,
+      "鼳": 93903,
+      "ɸ": 93904,
+      "戉": 93905,
+      "禡": 93906,
+      "⚘": 93907,
+      "彣": 93908,
+      "竮": 93909,
+      "☄": 93910,
+      "镸": 93911,
+      "釔": 93912,
+      "飔": 93913,
+      "臞": 93914,
+      "蚹": 93915,
+      "蘧": 93916,
+      "⇂": 93917,
+      "尩": 93918,
+      "👟": 93919,
+      "👨‍👩‍👧": 93920,
+      "暸": 93921,
+      "揞": 93922,
+      "🧍🏻": 93923,
+      "鶕": 93924,
+      "👿": 93925,
+      "枋": 93926,
+      "🤚🏽": 93927,
+      "澗": 93928,
+      "鵉": 93929,
+      "髅": 93930,
+      "賋": 93931,
+      "沶": 93932,
+      "徧": 93933,
+      "齧": 93934,
+      "菰": 93935,
+      "埣": 93936,
+      "🚲": 93937,
+      "旵": 93938,
+      "🦸🏾": 93939,
+      "捙": 93940,
+      "謬": 93941,
+      "檻": 93942,
+      "ʍ": 93943,
+      "埶": 93944,
+      "蒴": 93945,
+      "␆": 93946,
+      "💫": 93947,
+      "韅": 93948,
+      "扙": 93949,
+      "砿": 93950,
+      "宍": 93951,
+      "ɻ": 93952,
+      "哵": 93953,
+      "臬": 93954,
+      "軍": 93955,
+      "囐": 93956,
+      "磙": 93957,
+      "⃈": 93958,
+      "👩🏼‍🚒": 93959,
+      "砈": 93960,
+      "⏧": 93961,
+      "鞤": 93962,
+      "惖": 93963,
+      "為": 93964,
+      "媼": 93965,
+      "粢": 93966,
+      "腟": 93967,
+      "蠰": 93968,
+      "🪩": 93969,
+      "粩": 93970,
+      "绦": 93971,
+      "🍪": 93972,
+      "轐": 93973,
+      "緗": 93974,
+      "🇵🇲": 93975,
+      "⌽": 93976,
+      "罾": 93977,
+      "闞": 93978,
+      "阍": 93979,
+      "☪": 93980,
+      "鹼": 93981,
+      "👩🏻‍❤️‍👩🏽": 93982,
+      "鐃": 93983,
+      "劁": 93984,
+      "⅘": 93985,
+      "𓀍": 93986,
+      "💡": 93987,
+      "🇩🇴": 93988,
+      "澫": 93989,
+      "僯": 93990,
+      "軌": 93991,
+      "筲": 93992,
+      "💇🏽‍♂️": 93993,
+      "Ę": 93994,
+      "刲": 93995,
+      "🙇🏽‍♂️": 93996,
+      "👱🏻": 93997,
+      "🙀": 93998,
+      "匕": 93999,
+      "淂": 94000,
+      "🙍🏾‍♂️": 94001,
+      "騛": 94002,
+      "両": 94003,
+      "逿": 94004,
+      "灧": 94005,
+      "🧑🏻‍💼": 94006,
+      "🇵🇦": 94007,
+      "🧦": 94008,
+      "呯": 94009,
+      "瑓": 94010,
+      "婱": 94011,
+      "🫸🏿": 94012,
+      "鱲": 94013,
+      "倷": 94014,
+      "觖": 94015,
+      "堈": 94016,
+      "鲇": 94017,
+      "逬": 94018,
+      "⌁": 94019,
+      "觢": 94020,
+      "蝞": 94021,
+      "🇱🇻": 94022,
+      "媍": 94023,
+      "叀": 94024,
+      "ˎ": 94025,
+      "樍": 94026,
+      "👨🏽‍🌾": 94027,
+      "湂": 94028,
+      "鮧": 94029,
+      "箳": 94030,
+      "匭": 94031,
+      "薃": 94032,
+      "懗": 94033,
+      "槙": 94034,
+      "艍": 94035,
+      "媞": 94036,
+      "🇬🇬": 94037,
+      "🌏": 94038,
+      "傫": 94039,
+      "劕": 94040,
+      "鲟": 94041,
+      "🪦": 94042,
+      "蒬": 94043,
+      "毺": 94044,
+      "硖": 94045,
+      "瑆": 94046,
+      "👨🏻‍❤️‍💋‍👨🏻": 94047,
+      "硈": 94048,
+      "↳": 94049,
+      "嬀": 94050,
+      "🥋": 94051,
+      "ɲ": 94052,
+      "賁": 94053,
+      "巆": 94054,
+      "譵": 94055,
+      "黵": 94056,
+      "桝": 94057,
+      "蘲": 94058,
+      "喞": 94059,
+      "綕": 94060,
+      "驮": 94061,
+      "簴": 94062,
+      "乶": 94063,
+      "熁": 94064,
+      "熃": 94065,
+      "🇫🇲": 94066,
+      "虢": 94067,
+      "禐": 94068,
+      "顏": 94069,
+      "冘": 94070,
+      "ʓ": 94071,
+      "飨": 94072,
+      "🤦🏽": 94073,
+      "馕": 94074,
+      "礎": 94075,
+      "跉": 94076,
+      "鯛": 94077,
+      "甠": 94078,
+      "膯": 94079,
+      "氅": 94080,
+      "埽": 94081,
+      "爙": 94082,
+      "櫺": 94083,
+      "⋊": 94084,
+      "颀": 94085,
+      "␱": 94086,
+      "怞": 94087,
+      "⎢": 94088,
+      "觳": 94089,
+      "菓": 94090,
+      "吣": 94091,
+      "饎": 94092,
+      "讏": 94093,
+      "嚙": 94094,
+      "饑": 94095,
+      "鹖": 94096,
+      "褩": 94097,
+      "🙅🏽‍♀️": 94098,
+      "🖖🏿": 94099,
+      "憽": 94100,
+      "蒟": 94101,
+      "烎": 94102,
+      "亇": 94103,
+      "垼": 94104,
+      "瓪": 94105,
+      "睯": 94106,
+      "哅": 94107,
+      "娦": 94108,
+      "瓈": 94109,
+      "諒": 94110,
+      "🇩🇰": 94111,
+      "矓": 94112,
+      "寖": 94113,
+      "烑": 94114,
+      "莠": 94115,
+      "菹": 94116,
+      "蜺": 94117,
+      "隺": 94118,
+      "矗": 94119,
+      "🏄🏾": 94120,
+      "児": 94121,
+      "砮": 94122,
+      "嬨": 94123,
+      "捎": 94124,
+      "😀": 94125,
+      "礔": 94126,
+      "侟": 94127,
+      "廜": 94128,
+      "🔶": 94129,
+      "⊜": 94130,
+      "Ʀ": 94131,
+      "釄": 94132,
+      "讌": 94133,
+      "愺": 94134,
+      "滐": 94135,
+      "栬": 94136,
+      "胼": 94137,
+      "楣": 94138,
+      "🐹": 94139,
+      "鎱": 94140,
+      "獦": 94141,
+      "蜢": 94142,
+      "砳": 94143,
+      "漩": 94144,
+      "樬": 94145,
+      "糩": 94146,
+      "軜": 94147,
+      "軆": 94148,
+      "惪": 94149,
+      "⏻": 94150,
+      "頙": 94151,
+      "👨🏿‍🌾": 94152,
+      "忮": 94153,
+      "騋": 94154,
+      "懆": 94155,
+      "👩🏾‍🔧": 94156,
+      "犙": 94157,
+      "樇": 94158,
+      "盢": 94159,
+      "迵": 94160,
+      "邚": 94161,
+      "孓": 94162,
+      "ⅵ": 94163,
+      "👎🏾": 94164,
+      "卲": 94165,
+      "緅": 94166,
+      "肭": 94167,
+      "醧": 94168,
+      "躱": 94169,
+      "☝🏼": 94170,
+      "¹": 94171,
+      "肀": 94172,
+      "凥": 94173,
+      "欱": 94174,
+      "👩🏻‍❤️‍💋‍👩🏾": 94175,
+      "勠": 94176,
+      "楀": 94177,
+      "鍄": 94178,
+      "荳": 94179,
+      "糮": 94180,
+      "镮": 94181,
+      "旇": 94182,
+      "ĉ": 94183,
+      "凼": 94184,
+      "爦": 94185,
+      "🔙": 94186,
+      "鄾": 94187,
+      "🈷️": 94188,
+      "筳": 94189,
+      "屷": 94190,
+      "髼": 94191,
+      "郛": 94192,
+      "🏌🏾": 94193,
+      "ɷ": 94194,
+      "騁": 94195,
+      "Ķ": 94196,
+      "淖": 94197,
+      "膧": 94198,
+      "郏": 94199,
+      "唛": 94200,
+      "滠": 94201,
+      "敒": 94202,
+      "🙅🏻‍♂️": 94203,
+      "奌": 94204,
+      "烍": 94205,
+      "箜": 94206,
+      "┅": 94207,
+      "💁🏿‍♀️": 94208,
+      "聢": 94209,
+      "忣": 94210,
+      "≯": 94211,
+      "暍": 94212,
+      "坴": 94213,
+      "圖": 94214,
+      "旄": 94215,
+      "敭": 94216,
+      "袢": 94217,
+      "🧑🏿‍🦱": 94218,
+      "燙": 94219,
+      "蒛": 94220,
+      "缁": 94221,
+      "魆": 94222,
+      "🧑🏾‍🦽": 94223,
+      "薕": 94224,
+      "膔": 94225,
+      "鈖": 94226,
+      "齼": 94227,
+      "漉": 94228,
+      "鈓": 94229,
+      "👨🏾‍💼": 94230,
+      "鬳": 94231,
+      "葸": 94232,
+      "铨": 94233,
+      "姷": 94234,
+      "結": 94235,
+      "👩🏾‍🦯": 94236,
+      "🍌": 94237,
+      "👨🏿‍🦼": 94238,
+      "🚐": 94239,
+      "⑇": 94240,
+      "⁢": 94241,
+      "旼": 94242,
+      "縙": 94243,
+      "拀": 94244,
+      "蠿": 94245,
+      "🙆🏼‍♂️": 94246,
+      "醶": 94247,
+      "禂": 94248,
+      "👩🏻‍❤️‍👩🏼": 94249,
+      "🧚🏽‍♀️": 94250,
+      "壖": 94251,
+      "🍛": 94252,
+      "觊": 94253,
+      "蘎": 94254,
+      "₏": 94255,
+      "覤": 94256,
+      "荪": 94257,
+      "梡": 94258,
+      "🧡": 94259,
+      "焸": 94260,
+      "燒": 94261,
+      "阚": 94262,
+      "掋": 94263,
+      "絑": 94264,
+      "鐺": 94265,
+      "🧙🏽‍♀️": 94266,
+      "詢": 94267,
+      "轉": 94268,
+      "▯": 94269,
+      "岏": 94270,
+      "蘨": 94271,
+      "牗": 94272,
+      "悺": 94273,
+      "瓚": 94274,
+      "欐": 94275,
+      "齍": 94276,
+      "脍": 94277,
+      "腳": 94278,
+      "簕": 94279,
+      "祂": 94280,
+      "⓵": 94281,
+      "灎": 94282,
+      "鶍": 94283,
+      "忩": 94284,
+      "隯": 94285,
+      "禉": 94286,
+      "詴": 94287,
+      "🏋🏿": 94288,
+      "ℳ": 94289,
+      "🇻🇪": 94290,
+      "缛": 94291,
+      "澟": 94292,
+      "庬": 94293,
+      "₲": 94294,
+      "🇬🇸": 94295,
+      "蕷": 94296,
+      "筗": 94297,
+      "⑂": 94298,
+      "癓": 94299,
+      "讝": 94300,
+      "鯨": 94301,
+      "⍕": 94302,
+      "🫓": 94303,
+      "🦗": 94304,
+      "顣": 94305,
+      "🐀": 94306,
+      "嫞": 94307,
+      "瀖": 94308,
+      "愦": 94309,
+      "Ʈ": 94310,
+      "錻": 94311,
+      "邏": 94312,
+      "曛": 94313,
+      "豺": 94314,
+      "騐": 94315,
+      "覼": 94316,
+      "臏": 94317,
+      "棖": 94318,
+      "痌": 94319,
+      "嬖": 94320,
+      "桵": 94321,
+      "磔": 94322,
+      "砀": 94323,
+      "焵": 94324,
+      "鴿": 94325,
+      "陾": 94326,
+      "犞": 94327,
+      "ϙ": 94328,
+      "錽": 94329,
+      "遪": 94330,
+      "烳": 94331,
+      "👠": 94332,
+      "甇": 94333,
+      "🦹🏼‍♀️": 94334,
+      "⃽": 94335,
+      "虬": 94336,
+      "銶": 94337,
+      "碜": 94338,
+      "🪺": 94339,
+      "渙": 94340,
+      "彖": 94341,
+      "🇲🇹": 94342,
+      "鑣": 94343,
+      "😆": 94344,
+      "瓴": 94345,
+      "👩🏾‍🏭": 94346,
+      "👩🏾‍❤️‍👩🏾": 94347,
+      "🧏‍♂️": 94348,
+      "掔": 94349,
+      "🧛🏾": 94350,
+      "蹒": 94351,
+      "毟": 94352,
+      "瘈": 94353,
+      "诰": 94354,
+      "◡": 94355,
+      "椷": 94356,
+      "獷": 94357,
+      "🙅🏼": 94358,
+      "穵": 94359,
+      "鍷": 94360,
+      "崜": 94361,
+      "顮": 94362,
+      "簡": 94363,
+      "毚": 94364,
+      "↊": 94365,
+      "鋐": 94366,
+      "漃": 94367,
+      "砹": 94368,
+      "炴": 94369,
+      "湹": 94370,
+      "詄": 94371,
+      "👭🏽": 94372,
+      "锎": 94373,
+      "👩🏻‍🦳": 94374,
+      "臡": 94375,
+      "紩": 94376,
+      "墭": 94377,
+      "欃": 94378,
+      "蠟": 94379,
+      "黩": 94380,
+      "洙": 94381,
+      "瘨": 94382,
+      "傳": 94383,
+      "櫼": 94384,
+      "竵": 94385,
+      "灖": 94386,
+      "歳": 94387,
+      "瘪": 94388,
+      "∛": 94389,
+      "翓": 94390,
+      "嫈": 94391,
+      "忡": 94392,
+      "盺": 94393,
+      "壔": 94394,
+      "℉": 94395,
+      "✊🏾": 94396,
+      "攰": 94397,
+      "👨🏼‍🚒": 94398,
+      "嚾": 94399,
+      "沝": 94400,
+      "跔": 94401,
+      "👨🏽‍🎓": 94402,
+      "觠": 94403,
+      "媽": 94404,
+      "⌡": 94405,
+      "囂": 94406,
+      "毝": 94407,
+      "衽": 94408,
+      "锬": 94409,
+      "琝": 94410,
+      "鋼": 94411,
+      "訛": 94412,
+      "痸": 94413,
+      "👩🏿‍💻": 94414,
+      "🇹🇿": 94415,
+      "泈": 94416,
+      "佾": 94417,
+      "駘": 94418,
+      "🏌🏽‍♀️": 94419,
+      "霺": 94420,
+      "櫥": 94421,
+      "筣": 94422,
+      "対": 94423,
+      "彾": 94424,
+      "淯": 94425,
+      "暤": 94426,
+      "卻": 94427,
+      "˚": 94428,
+      "悏": 94429,
+      "輊": 94430,
+      "┾": 94431,
+      "烓": 94432,
+      "Ȅ": 94433,
+      "͘": 94434,
+      "閥": 94435,
+      "椗": 94436,
+      "絋": 94437,
+      "崃": 94438,
+      "ȿ": 94439,
+      "☷": 94440,
+      "🪆": 94441,
+      "賉": 94442,
+      "␯": 94443,
+      "泞": 94444,
+      "鬊": 94445,
+      "磲": 94446,
+      "🙅🏼‍♀️": 94447,
+      "幜": 94448,
+      "✍🏽": 94449,
+      "巹": 94450,
+      "骻": 94451,
+      "ʒ": 94452,
+      "糍": 94453,
+      "嘍": 94454,
+      "泹": 94455,
+      "掻": 94456,
+      "舩": 94457,
+      "悋": 94458,
+      "聰": 94459,
+      "檘": 94460,
+      "胤": 94461,
+      "裩": 94462,
+      "🧛‍♂️": 94463,
+      "緶": 94464,
+      "橇": 94465,
+      "鬁": 94466,
+      "箑": 94467,
+      "変": 94468,
+      "蔛": 94469,
+      "˺": 94470,
+      "嶿": 94471,
+      "匘": 94472,
+      "🧑🏼‍❤️‍🧑🏽": 94473,
+      "愫": 94474,
+      "獒": 94475,
+      "顢": 94476,
+      "旸": 94477,
+      "織": 94478,
+      "␃": 94479,
+      "👩🏽‍❤️‍👩🏼": 94480,
+      "ⅎ": 94481,
+      "夾": 94482,
+      "嬤": 94483,
+      "⚫": 94484,
+      "禘": 94485,
+      "犡": 94486,
+      "繰": 94487,
+      "螊": 94488,
+      "舽": 94489,
+      "🎶": 94490,
+      "抝": 94491,
+      "㊙️": 94492,
+      "⊝": 94493,
+      "娮": 94494,
+      "伧": 94495,
+      "🏄🏽": 94496,
+      "郟": 94497,
+      "🔳": 94498,
+      "謒": 94499,
+      "峧": 94500,
+      "砢": 94501,
+      "👬🏻": 94502,
+      "溎": 94503,
+      "饖": 94504,
+      "🙇🏿‍♂️": 94505,
+      "橷": 94506,
+      "🕺🏻": 94507,
+      "绻": 94508,
+      "⒮": 94509,
+      "惐": 94510,
+      "鷈": 94511,
+      "嘷": 94512,
+      "鑹": 94513,
+      "璖": 94514,
+      "鈝": 94515,
+      "搿": 94516,
+      "釤": 94517,
+      "🎿": 94518,
+      "ↁ": 94519,
+      "鷑": 94520,
+      "佺": 94521,
+      "毤": 94522,
+      "迸": 94523,
+      "硿": 94524,
+      "齂": 94525,
+      "琣": 94526,
+      "埭": 94527,
+      "媦": 94528,
+      "蔶": 94529,
+      "噳": 94530,
+      "湀": 94531,
+      "衕": 94532,
+      "►": 94533,
+      "蹤": 94534,
+      "竅": 94535,
+      "搰": 94536,
+      "栥": 94537,
+      "傦": 94538,
+      "釮": 94539,
+      "珽": 94540,
+      "℁": 94541,
+      "畝": 94542,
+      "🍗": 94543,
+      "絳": 94544,
+      "亶": 94545,
+      "🫃🏽": 94546,
+      "恜": 94547,
+      "靷": 94548,
+      "齣": 94549,
+      "铦": 94550,
+      "蟣": 94551,
+      "兇": 94552,
+      "礤": 94553,
+      "穥": 94554,
+      "▱": 94555,
+      "靦": 94556,
+      "👨‍👧": 94557,
+      "貿": 94558,
+      "錉": 94559,
+      "砵": 94560,
+      "蜐": 94561,
+      "🤸🏼‍♀️": 94562,
+      "闔": 94563,
+      "帨": 94564,
+      "槇": 94565,
+      "宭": 94566,
+      "萣": 94567,
+      "茋": 94568,
+      "镎": 94569,
+      "🤸🏾‍♀️": 94570,
+      "嫀": 94571,
+      "勀": 94572,
+      "❇️": 94573,
+      "👏🏼": 94574,
+      "ϝ": 94575,
+      "牘": 94576,
+      "🤽🏿‍♀️": 94577,
+      "灏": 94578,
+      "🗃": 94579,
+      "瞛": 94580,
+      "喾": 94581,
+      "炤": 94582,
+      "撍": 94583,
+      "👨🏾‍❤️‍💋‍👨🏿": 94584,
+      "楢": 94585,
+      "鞼": 94586,
+      "緬": 94587,
+      "ɓ": 94588,
+      "珣": 94589,
+      "🛌🏻": 94590,
+      "👳🏿‍♂️": 94591,
+      "⍧": 94592,
+      "谄": 94593,
+      "洖": 94594,
+      "裠": 94595,
+      "鑶": 94596,
+      "幟": 94597,
+      "罨": 94598,
+      "芚": 94599,
+      "⁶": 94600,
+      "˜": 94601,
+      "龃": 94602,
+      "鑮": 94603,
+      "藴": 94604,
+      "👩🏽‍❤️‍💋‍👨🏿": 94605,
+      "冼": 94606,
+      "碸": 94607,
+      "捪": 94608,
+      "蕐": 94609,
+      "薗": 94610,
+      "砓": 94611,
+      "⏱": 94612,
+      "💝": 94613,
+      "🏯": 94614,
+      "埦": 94615,
+      "粫": 94616,
+      "🫳🏼": 94617,
+      "騆": 94618,
+      "袼": 94619,
+      "賕": 94620,
+      "傏": 94621,
+      "艶": 94622,
+      "鑂": 94623,
+      "髎": 94624,
+      "眆": 94625,
+      "葶": 94626,
+      "挩": 94627,
+      "鴻": 94628,
+      "狥": 94629,
+      "篧": 94630,
+      "鹻": 94631,
+      "籜": 94632,
+      "薼": 94633,
+      "臿": 94634,
+      "靟": 94635,
+      "♚": 94636,
+      "噞": 94637,
+      "覕": 94638,
+      "鸗": 94639,
+      "📵": 94640,
+      "嗦": 94641,
+      "鰧": 94642,
+      "愄": 94643,
+      "蔒": 94644,
+      "絷": 94645,
+      "铩": 94646,
+      "嬣": 94647,
+      "ɫ": 94648,
+      "甴": 94649,
+      "舲": 94650,
+      "髵": 94651,
+      "罺": 94652,
+      "˭": 94653,
+      "🚵‍♂️": 94654,
+      "蝏": 94655,
+      "殀": 94656,
+      "磣": 94657,
+      "舠": 94658,
+      "鶽": 94659,
+      "鍳": 94660,
+      "👣": 94661,
+      "湯": 94662,
+      "肨": 94663,
+      "🫅🏼": 94664,
+      "奛": 94665,
+      "勧": 94666,
+      "嫺": 94667,
+      "🫲🏽": 94668,
+      "飆": 94669,
+      "虄": 94670,
+      "韀": 94671,
+      "嬋": 94672,
+      "碄": 94673,
+      "畺": 94674,
+      "🧑🏼‍🦳": 94675,
+      "砱": 94676,
+      "🧜🏿‍♂️": 94677,
+      "肏": 94678,
+      "徰": 94679,
+      "🕵🏼‍♀️": 94680,
+      "沆": 94681,
+      "恸": 94682,
+      "🦻🏿": 94683,
+      "鐅": 94684,
+      "鮎": 94685,
+      "騔": 94686,
+      "豯": 94687,
+      "犕": 94688,
+      "竉": 94689,
+      "⏂": 94690,
+      "秱": 94691,
+      "🤵": 94692,
+      "鸍": 94693,
+      "🧺": 94694,
+      "撿": 94695,
+      "螁": 94696,
+      "鯑": 94697,
+      "蚟": 94698,
+      "絊": 94699,
+      "睋": 94700,
+      "箶": 94701,
+      "厲": 94702,
+      "骕": 94703,
+      "蠧": 94704,
+      "诽": 94705,
+      "妕": 94706,
+      "蘼": 94707,
+      "⑘": 94708,
+      "顁": 94709,
+      "胕": 94710,
+      "荰": 94711,
+      "🤙": 94712,
+      "齪": 94713,
+      "檸": 94714,
+      "摖": 94715,
+      "扠": 94716,
+      "昇": 94717,
+      "翷": 94718,
+      "拮": 94719,
+      "🇺🇾": 94720,
+      "婦": 94721,
+      "😽": 94722,
+      "迆": 94723,
+      "僟": 94724,
+      "琘": 94725,
+      "藐": 94726,
+      "阖": 94727,
+      "阋": 94728,
+      "庑": 94729,
+      "👩🏻‍❤️‍👨🏿": 94730,
+      "臕": 94731,
+      "ℿ": 94732,
+      "疔": 94733,
+      "堘": 94734,
+      "磥": 94735,
+      "Γ": 94736,
+      "謤": 94737,
+      "⍆": 94738,
+      "銬": 94739,
+      "⏛": 94740,
+      "餆": 94741,
+      "煩": 94742,
+      "艏": 94743,
+      "蝵": 94744,
+      "鈏": 94745,
+      "玽": 94746,
+      "巔": 94747,
+      "噔": 94748,
+      "🙈": 94749,
+      "襵": 94750,
+      "嶼": 94751,
+      "騵": 94752,
+      "傿": 94753,
+      "僜": 94754,
+      "🧑🏿‍💻": 94755,
+      "泂": 94756,
+      "茷": 94757,
+      "泚": 94758,
+      "齩": 94759,
+      "黅": 94760,
+      "詓": 94761,
+      "☡": 94762,
+      "⃞": 94763,
+      "雜": 94764,
+      "釺": 94765,
+      "螚": 94766,
+      "劵": 94767,
+      "哜": 94768,
+      "樔": 94769,
+      "圽": 94770,
+      "🙍🏻": 94771,
+      "惝": 94772,
+      "繏": 94773,
+      "擑": 94774,
+      "稩": 94775,
+      "鰿": 94776,
+      "徍": 94777,
+      "潊": 94778,
+      "跠": 94779,
+      "🇱🇧": 94780,
+      "鲰": 94781,
+      "牶": 94782,
+      "膇": 94783,
+      "賗": 94784,
+      "褳": 94785,
+      "‹": 94786,
+      "⇎": 94787,
+      "⛉": 94788,
+      "関": 94789,
+      "髺": 94790,
+      "👂🏾": 94791,
+      "馴": 94792,
+      "镙": 94793,
+      "儴": 94794,
+      "剈": 94795,
+      "👨🏿‍🦽": 94796,
+      "ϣ": 94797,
+      "錧": 94798,
+      "嫶": 94799,
+      "僂": 94800,
+      "☋": 94801,
+      "喸": 94802,
+      "鰥": 94803,
+      "僔": 94804,
+      "徉": 94805,
+      "釻": 94806,
+      "職": 94807,
+      "笄": 94808,
+      "绗": 94809,
+      "傱": 94810,
+      "🏌️‍♂️": 94811,
+      "耞": 94812,
+      "ℼ": 94813,
+      "葭": 94814,
+      "痎": 94815,
+      "墫": 94816,
+      "竩": 94817,
+      "礬": 94818,
+      "蟨": 94819,
+      "犩": 94820,
+      "覹": 94821,
+      "◅": 94822,
+      "裷": 94823,
+      "🫰🏽": 94824,
+      "婶": 94825,
+      "歀": 94826,
+      "鯽": 94827,
+      "糸": 94828,
+      "鈟": 94829,
+      "帾": 94830,
+      "浄": 94831,
+      "🤾🏽‍♀️": 94832,
+      "塈": 94833,
+      "崶": 94834,
+      "勵": 94835,
+      "楱": 94836,
+      "隵": 94837,
+      "轖": 94838,
+      "搡": 94839,
+      "🍦": 94840,
+      "贃": 94841,
+      "厝": 94842,
+      "亠": 94843,
+      "艉": 94844,
+      "瀠": 94845,
+      "笹": 94846,
+      "殣": 94847,
+      "岈": 94848,
+      "誮": 94849,
+      "ⓓ": 94850,
+      "♼": 94851,
+      "慦": 94852,
+      "鯊": 94853,
+      "🧙‍♀️": 94854,
+      "將": 94855,
+      "曁": 94856,
+      "馝": 94857,
+      "鏤": 94858,
+      "洣": 94859,
+      "泋": 94860,
+      "瀄": 94861,
+      "鹁": 94862,
+      "齏": 94863,
+      "爗": 94864,
+      "晡": 94865,
+      "🧝🏿‍♂️": 94866,
+      "貤": 94867,
+      "聠": 94868,
+      "胈": 94869,
+      "🇼🇫": 94870,
+      "👩🏻‍🚒": 94871,
+      "洴": 94872,
+      "醢": 94873,
+      "璊": 94874,
+      "⒲": 94875,
+      "̶": 94876,
+      "鵦": 94877,
+      "餿": 94878,
+      "鳇": 94879,
+      "榃": 94880,
+      "飵": 94881,
+      "怽": 94882,
+      "🧑🏾‍❤️‍🧑🏿": 94883,
+      "🤖": 94884,
+      "饸": 94885,
+      "樈": 94886,
+      "娢": 94887,
+      "瓇": 94888,
+      "纁": 94889,
+      "臾": 94890,
+      "🏠": 94891,
+      "兙": 94892,
+      "瘒": 94893,
+      "赙": 94894,
+      "潚": 94895,
+      "炻": 94896,
+      "涞": 94897,
+      "菡": 94898,
+      "翹": 94899,
+      "焓": 94900,
+      "淾": 94901,
+      "莴": 94902,
+      "寫": 94903,
+      "🛅": 94904,
+      "殳": 94905,
+      "啴": 94906,
+      "🇿🇦": 94907,
+      "🗜": 94908,
+      "猃": 94909,
+      "🧜🏿‍♀️": 94910,
+      "⎻": 94911,
+      "夵": 94912,
+      "鐍": 94913,
+      "鲶": 94914,
+      "🛒": 94915,
+      "鼘": 94916,
+      "埬": 94917,
+      "塺": 94918,
+      "龘": 94919,
+      "斫": 94920,
+      "辠": 94921,
+      "⍞": 94922,
+      "旈": 94923,
+      "🤴🏽": 94924,
+      "蘀": 94925,
+      "猂": 94926,
+      "櫗": 94927,
+      "貑": 94928,
+      "棞": 94929,
+      "桕": 94930,
+      "孼": 94931,
+      "鍧": 94932,
+      "猰": 94933,
+      "瓟": 94934,
+      "⎠": 94935,
+      "‾": 94936,
+      "寊": 94937,
+      "👨🏾‍🍼": 94938,
+      "捇": 94939,
+      "笍": 94940,
+      "⚇": 94941,
+      "乧": 94942,
+      "櫹": 94943,
+      "╣": 94944,
+      "粆": 94945,
+      "ↂ": 94946,
+      "🟩": 94947,
+      "̿": 94948,
+      "陹": 94949,
+      "狔": 94950,
+      "荸": 94951,
+      "🧝🏻‍♂️": 94952,
+      "🧓🏾": 94953,
+      "琜": 94954,
+      "屸": 94955,
+      "倎": 94956,
+      "🚙": 94957,
+      "🌯": 94958,
+      "伒": 94959,
+      "郰": 94960,
+      "甹": 94961,
+      "讦": 94962,
+      "掮": 94963,
+      "⒖": 94964,
+      "槣": 94965,
+      "趭": 94966,
+      "儗": 94967,
+      "攵": 94968,
+      "🦘": 94969,
+      "⁨": 94970,
+      "寣": 94971,
+      "乿": 94972,
+      "摫": 94973,
+      "憘": 94974,
+      "抲": 94975,
+      "偡": 94976,
+      "嫼": 94977,
+      "畟": 94978,
+      "☳": 94979,
+      "跰": 94980,
+      "僳": 94981,
+      "簆": 94982,
+      "坆": 94983,
+      "暻": 94984,
+      "尶": 94985,
+      "躒": 94986,
+      "痋": 94987,
+      "澩": 94988,
+      "襀": 94989,
+      "咤": 94990,
+      "坺": 94991,
+      "鰕": 94992,
+      "👨🏿‍❤️‍💋‍👨🏽": 94993,
+      "霂": 94994,
+      "狒": 94995,
+      "鉂": 94996,
+      "滆": 94997,
+      "🉐": 94998,
+      "鶒": 94999,
+      "堶": 95000,
+      "麔": 95001,
+      "蠋": 95002,
+      "🤘🏽": 95003,
+      "凓": 95004,
+      "墰": 95005,
+      "🧛🏽‍♂️": 95006,
+      "🧠": 95007,
+      "獾": 95008,
+      "畐": 95009,
+      "謌": 95010,
+      "舤": 95011,
+      "磼": 95012,
+      "鞊": 95013,
+      "劽": 95014,
+      "鸴": 95015,
+      "袛": 95016,
+      "軄": 95017,
+      "颎": 95018,
+      "唶": 95019,
+      "菉": 95020,
+      "噎": 95021,
+      "饾": 95022,
+      "餗": 95023,
+      "湶": 95024,
+      "羭": 95025,
+      "倱": 95026,
+      "塭": 95027,
+      "Ŏ": 95028,
+      "埼": 95029,
+      "迌": 95030,
+      "👨🏻‍❤️‍👨🏻": 95031,
+      "柾": 95032,
+      "擢": 95033,
+      "镨": 95034,
+      "楘": 95035,
+      "張": 95036,
+      "鎀": 95037,
+      "🧎🏽‍♂️": 95038,
+      "爔": 95039,
+      "閄": 95040,
+      "閖": 95041,
+      "欋": 95042,
+      "🤞🏼": 95043,
+      "👩🏼‍🦰": 95044,
+      "畄": 95045,
+      "岴": 95046,
+      "ƃ": 95047,
+      "Ɂ": 95048,
+      "爐": 95049,
+      "鼏": 95050,
+      "🤼": 95051,
+      "𓀔": 95052,
+      "堓": 95053,
+      "駬": 95054,
+      "豖": 95055,
+      "🥟": 95056,
+      "縇": 95057,
+      "撘": 95058,
+      "拹": 95059,
+      "🐱": 95060,
+      "👩🏼‍🦱": 95061,
+      "🧑🏽‍🍼": 95062,
+      "喗": 95063,
+      "椇": 95064,
+      "茾": 95065,
+      "👮🏻": 95066,
+      "⛹🏿": 95067,
+      "📧": 95068,
+      "賓": 95069,
+      "呍": 95070,
+      "₀": 95071,
+      "矧": 95072,
+      "🛐": 95073,
+      "墳": 95074,
+      "陟": 95075,
+      "🧑🏻‍❤️‍💋‍🧑🏾": 95076,
+      "锞": 95077,
+      "↶": 95078,
+      "麶": 95079,
+      "👩🏿‍❤️‍👩🏼": 95080,
+      "苤": 95081,
+      "🥌": 95082,
+      "̞": 95083,
+      "嫱": 95084,
+      "晱": 95085,
+      "⑨": 95086,
+      "ʰ": 95087,
+      "弭": 95088,
+      "礘": 95089,
+      "℞": 95090,
+      "砕": 95091,
+      "΃": 95092,
+      "⅖": 95093,
+      "鸞": 95094,
+      "翪": 95095,
+      "遊": 95096,
+      "釠": 95097,
+      "帺": 95098,
+      "棬": 95099,
+      "茲": 95100,
+      "🕺🏼": 95101,
+      "棟": 95102,
+      "蘔": 95103,
+      "刢": 95104,
+      "鎴": 95105,
+      "🐟": 95106,
+      "◦": 95107,
+      "姅": 95108,
+      "踡": 95109,
+      "痥": 95110,
+      "∢": 95111,
+      "🤜🏼": 95112,
+      "钄": 95113,
+      "☚": 95114,
+      "": 95115,
+      "蕓": 95116,
+      "🥷🏼": 95117,
+      "🙋🏻‍♀️": 95118,
+      "尵": 95119,
+      "噶": 95120,
+      "輪": 95121,
+      "褄": 95122,
+      "衂": 95123,
+      "糘": 95124,
+      "綟": 95125,
+      "👰🏼‍♀️": 95126,
+      "遼": 95127,
+      "鲃": 95128,
+      "祽": 95129,
+      "禓": 95130,
+      "慇": 95131,
+      "譩": 95132,
+      "🧑🏽‍🎓": 95133,
+      "鷿": 95134,
+      "睒": 95135,
+      "🤬": 95136,
+      "漁": 95137,
+      "冟": 95138,
+      "謉": 95139,
+      "🍀": 95140,
+      "眻": 95141,
+      "伵": 95142,
+      "閗": 95143,
+      "🤟🏻": 95144,
+      "̂": 95145,
+      "浠": 95146,
+      "啭": 95147,
+      "蘏": 95148,
+      "憼": 95149,
+      "伨": 95150,
+      "銘": 95151,
+      "⏀": 95152,
+      "線": 95153,
+      "鞧": 95154,
+      "🤠": 95155,
+      "氻": 95156,
+      "ȗ": 95157,
+      "竤": 95158,
+      "纣": 95159,
+      "弤": 95160,
+      "冚": 95161,
+      "噉": 95162,
+      "縫": 95163,
+      "🥪": 95164,
+      "👩🏾‍🌾": 95165,
+      "儝": 95166,
+      "⋏": 95167,
+      "⒭": 95168,
+      "👮🏻‍♂️": 95169,
+      "丱": 95170,
+      "🤜🏿": 95171,
+      "茟": 95172,
+      "眙": 95173,
+      "謎": 95174,
+      "鬞": 95175,
+      "続": 95176,
+      "駶": 95177,
+      "纇": 95178,
+      "轃": 95179,
+      "垡": 95180,
+      "俟": 95181,
+      "泜": 95182,
+      "👆🏽": 95183,
+      "虸": 95184,
+      "🇻🇬": 95185,
+      "躑": 95186,
+      "裼": 95187,
+      "傧": 95188,
+      "偉": 95189,
+      "⎡": 95190,
+      "堐": 95191,
+      "籱": 95192,
+      "覣": 95193,
+      "炅": 95194,
+      "😶": 95195,
+      "汥": 95196,
+      "軇": 95197,
+      "👨🏾‍❤️‍💋‍👨🏼": 95198,
+      "⏯": 95199,
+      "ƴ": 95200,
+      "帵": 95201,
+      "惇": 95202,
+      "敡": 95203,
+      "櫏": 95204,
+      "暞": 95205,
+      "鲖": 95206,
+      "鱪": 95207,
+      "🍯": 95208,
+      "輑": 95209,
+      "Ο": 95210,
+      "憻": 95211,
+      "嶳": 95212,
+      "疅": 95213,
+      "袤": 95214,
+      "垵": 95215,
+      "恙": 95216,
+      "⒪": 95217,
+      "誶": 95218,
+      "鏑": 95219,
+      "嶑": 95220,
+      "痬": 95221,
+      "鳻": 95222,
+      "≸": 95223,
+      "坧": 95224,
+      "撦": 95225,
+      "漅": 95226,
+      "🧍🏿‍♂️": 95227,
+      "誝": 95228,
+      "槒": 95229,
+      "錠": 95230,
+      "磹": 95231,
+      "櫈": 95232,
+      "👉🏽": 95233,
+      "仼": 95234,
+      "葳": 95235,
+      "匟": 95236,
+      "髟": 95237,
+      "籂": 95238,
+      "🇫🇮": 95239,
+      "孖": 95240,
+      "欟": 95241,
+      "禶": 95242,
+      "繵": 95243,
+      "擄": 95244,
+      "綛": 95245,
+      "鈁": 95246,
+      "湈": 95247,
+      "恫": 95248,
+      "尀": 95249,
+      "詇": 95250,
+      "🇩🇿": 95251,
+      "👫🏽": 95252,
+      "洟": 95253,
+      "譂": 95254,
+      "挓": 95255,
+      "庤": 95256,
+      "牕": 95257,
+      "沗": 95258,
+      "乂": 95259,
+      "涼": 95260,
+      "矼": 95261,
+      "郚": 95262,
+      "╕": 95263,
+      "尯": 95264,
+      "讟": 95265,
+      "糝": 95266,
+      "桢": 95267,
+      "璾": 95268,
+      "淨": 95269,
+      "礻": 95270,
+      "袑": 95271,
+      "瞘": 95272,
+      "廭": 95273,
+      "🏋🏻‍♂️": 95274,
+      "ĝ": 95275,
+      "圅": 95276,
+      "鉓": 95277,
+      "珁": 95278,
+      "伆": 95279,
+      "雫": 95280,
+      "倉": 95281,
+      "⅏": 95282,
+      "貽": 95283,
+      "襍": 95284,
+      "佞": 95285,
+      "俎": 95286,
+      "🤽🏻‍♀️": 95287,
+      "饋": 95288,
+      "柭": 95289,
+      "靮": 95290,
+      "調": 95291,
+      "ₛ": 95292,
+      "憯": 95293,
+      "👨🏾‍🚀": 95294,
+      "🧍🏾‍♂️": 95295,
+      "腙": 95296,
+      "鞒": 95297,
+      "萔": 95298,
+      "醔": 95299,
+      "敂": 95300,
+      "漷": 95301,
+      "蛉": 95302,
+      "彴": 95303,
+      "脇": 95304,
+      "獡": 95305,
+      "‖": 95306,
+      "艻": 95307,
+      "🏀": 95308,
+      "👨🏻‍❤️‍👨🏿": 95309,
+      "▆": 95310,
+      "円": 95311,
+      "磃": 95312,
+      "垽": 95313,
+      "贒": 95314,
+      "黃": 95315,
+      "徺": 95316,
+      "汨": 95317,
+      "徦": 95318,
+      "醪": 95319,
+      "銤": 95320,
+      "垁": 95321,
+      "唬": 95322,
+      "👩🏻‍❤️‍💋‍👨🏿": 95323,
+      "綥": 95324,
+      "乄": 95325,
+      "猔": 95326,
+      "荵": 95327,
+      "狌": 95328,
+      "皨": 95329,
+      "貥": 95330,
+      "嵻": 95331,
+      "🧖🏽‍♀️": 95332,
+      "🕷": 95333,
+      "貐": 95334,
+      "凞": 95335,
+      "韔": 95336,
+      "汖": 95337,
+      "飠": 95338,
+      "俤": 95339,
+      "劅": 95340,
+      "🧍🏽‍♀️": 95341,
+      "🪿": 95342,
+      "胋": 95343,
+      "忂": 95344,
+      "澰": 95345,
+      "鯣": 95346,
+      "≞": 95347,
+      "噂": 95348,
+      "濎": 95349,
+      "瑦": 95350,
+      "輨": 95351,
+      "劢": 95352,
+      "峣": 95353,
+      "␴": 95354,
+      "兏": 95355,
+      "恲": 95356,
+      "煺": 95357,
+      "毲": 95358,
+      "鴤": 95359,
+      "捜": 95360,
+      "👩🏿‍❤️‍👨🏻": 95361,
+      "籑": 95362,
+      "🔈": 95363,
+      "◽": 95364,
+      "鉯": 95365,
+      "鄝": 95366,
+      "👷🏿": 95367,
+      "飂": 95368,
+      "🧗": 95369,
+      "貃": 95370,
+      "泶": 95371,
+      "凜": 95372,
+      "🟦": 95373,
+      "遨": 95374,
+      "🔥": 95375,
+      "嬧": 95376,
+      "熾": 95377,
+      "赯": 95378,
+      "⛹🏾": 95379,
+      "锳": 95380,
+      "↹": 95381,
+      "黓": 95382,
+      "懧": 95383,
+      "蕔": 95384,
+      "睶": 95385,
+      "嵠": 95386,
+      "薷": 95387,
+      "庉": 95388,
+      "諛": 95389,
+      "喕": 95390,
+      "͌": 95391,
+      "·": 95392,
+      "褬": 95393,
+      "⒙": 95394,
+      "🧑🏿‍❤️‍💋‍🧑🏾": 95395,
+      "殻": 95396,
+      "侉": 95397,
+      "娌": 95398,
+      "焃": 95399,
+      "譻": 95400,
+      "狵": 95401,
+      "鲋": 95402,
+      "肂": 95403,
+      "🇹🇩": 95404,
+      "🅾️": 95405,
+      "🦵🏽": 95406,
+      "鷾": 95407,
+      "顎": 95408,
+      "聮": 95409,
+      "褮": 95410,
+      "溳": 95411,
+      "迨": 95412,
+      "椞": 95413,
+      "🦌": 95414,
+      "妜": 95415,
+      "跷": 95416,
+      "📹": 95417,
+      "⍣": 95418,
+      "🧑🏿‍💼": 95419,
+      "鏩": 95420,
+      "鮰": 95421,
+      "汴": 95422,
+      "蹶": 95423,
+      "燺": 95424,
+      "Ω": 95425,
+      "蠛": 95426,
+      "☂": 95427,
+      "🧑🏻‍🍳": 95428,
+      "🇸🇿": 95429,
+      "竾": 95430,
+      "湆": 95431,
+      "綣": 95432,
+      "運": 95433,
+      "🇸🇴": 95434,
+      "蛆": 95435,
+      "👩🏽‍🦱": 95436,
+      "扺": 95437,
+      "⍫": 95438,
+      "暥": 95439,
+      "⛆": 95440,
+      "黬": 95441,
+      "骿": 95442,
+      "👐🏻": 95443,
+      "🧎🏻": 95444,
+      "玘": 95445,
+      "尷": 95446,
+      "蓅": 95447,
+      "🇲🇸": 95448,
+      "皜": 95449,
+      "🫲🏼": 95450,
+      "侤": 95451,
+      "嶮": 95452,
+      "樞": 95453,
+      "怊": 95454,
+      "⑚": 95455,
+      "沕": 95456,
+      "譋": 95457,
+      "嚀": 95458,
+      "攤": 95459,
+      "榽": 95460,
+      "昅": 95461,
+      "瑵": 95462,
+      "恌": 95463,
+      "🧘🏾": 95464,
+      "💞": 95465,
+      "覎": 95466,
+      "👸": 95467,
+      "␠": 95468,
+      "趗": 95469,
+      "🇱🇺": 95470,
+      "🍓": 95471,
+      "🙋🏽": 95472,
+      "衏": 95473,
+      "脠": 95474,
+      "隥": 95475,
+      "🪝": 95476,
+      "髄": 95477,
+      "ɹ": 95478,
+      "🙎‍♀️": 95479,
+      "魀": 95480,
+      "ͫ": 95481,
+      "啓": 95482,
+      "◐": 95483,
+      "梘": 95484,
+      "侖": 95485,
+      "浰": 95486,
+      "ⅾ": 95487,
+      "媟": 95488,
+      "瓾": 95489,
+      "璹": 95490,
+      "咢": 95491,
+      "👨🏼‍❤️‍👨🏽": 95492,
+      "俋": 95493,
+      "楥": 95494,
+      "鰝": 95495,
+      "汎": 95496,
+      "鑲": 95497,
+      "🇷🇪": 95498,
+      "戆": 95499,
+      "謪": 95500,
+      "拋": 95501,
+      "硞": 95502,
+      "鱬": 95503,
+      "🗣": 95504,
+      "簐": 95505,
+      "🕵🏾‍♀️": 95506,
+      "瘆": 95507,
+      "ǖ": 95508,
+      "蜸": 95509,
+      "🍠": 95510,
+      "鄇": 95511,
+      "瓿": 95512,
+      "🤼‍♂️": 95513,
+      "箌": 95514,
+      "钪": 95515,
+      "槍": 95516,
+      "磀": 95517,
+      "昞": 95518,
+      "狛": 95519,
+      "☝🏻": 95520,
+      "⋽": 95521,
+      "🇦🇿": 95522,
+      "嚤": 95523,
+      "Ǔ": 95524,
+      "稢": 95525,
+      "龒": 95526,
+      "秌": 95527,
+      "舿": 95528,
+      "≐": 95529,
+      "崤": 95530,
+      "🧓🏼": 95531,
+      "翤": 95532,
+      "崐": 95533,
+      "熓": 95534,
+      "胟": 95535,
+      "姩": 95536,
+      "兯": 95537,
+      "籼": 95538,
+      "銲": 95539,
+      "詵": 95540,
+      "趔": 95541,
+      "瀿": 95542,
+      "縃": 95543,
+      "藋": 95544,
+      "⌻": 95545,
+      "🧜": 95546,
+      "拚": 95547,
+      "鬠": 95548,
+      "Ƕ": 95549,
+      "🗡": 95550,
+      "賝": 95551,
+      "浂": 95552,
+      "👩🏻‍🚀": 95553,
+      "熅": 95554,
+      "≕": 95555,
+      "齡": 95556,
+      "漊": 95557,
+      "甅": 95558,
+      "膠": 95559,
+      "怮": 95560,
+      "ŷ": 95561,
+      "粿": 95562,
+      "睇": 95563,
+      "眄": 95564,
+      "覚": 95565,
+      "🧑🏻‍🦳": 95566,
+      "倻": 95567,
+      "胔": 95568,
+      "緀": 95569,
+      "蛒": 95570,
+      "銋": 95571,
+      "熑": 95572,
+      "嵶": 95573,
+      "鷛": 95574,
+      "钑": 95575,
+      "粼": 95576,
+      "蟮": 95577,
+      "誣": 95578,
+      "倁": 95579,
+      "Ⓛ": 95580,
+      "鲼": 95581,
+      "躘": 95582,
+      "疜": 95583,
+      "癳": 95584,
+      "櫝": 95585,
+      "🥵": 95586,
+      "👩🏽‍🌾": 95587,
+      "": 95588,
+      "": 95589,
+      "齱": 95590,
+      "妲": 95591,
+      "💂🏿‍♂️": 95592,
+      "檱": 95593,
+      "瘉": 95594,
+      "ⓕ": 95595,
+      "皚": 95596,
+      "👩🏽‍💻": 95597,
+      "肣": 95598,
+      "銾": 95599,
+      "癀": 95600,
+      "絬": 95601,
+      "挗": 95602,
+      "靗": 95603,
+      "濝": 95604,
+      "褨": 95605,
+      "臗": 95606,
+      "柁": 95607,
+      "璒": 95608,
+      "👩🏿‍❤️‍👨🏿": 95609,
+      "瓳": 95610,
+      "卽": 95611,
+      "佄": 95612,
+      "👨🏼‍🦳": 95613,
+      "鐙": 95614,
+      "坍": 95615,
+      "🤩": 95616,
+      "樘": 95617,
+      "亯": 95618,
+      "👩🏼‍❤️‍💋‍👨🏿": 95619,
+      "寷": 95620,
+      "铱": 95621,
+      "刧": 95622,
+      "¨": 95623,
+      "躧": 95624,
+      "⋜": 95625,
+      "🛣": 95626,
+      "🍈": 95627,
+      "赒": 95628,
+      "∾": 95629,
+      "祫": 95630,
+      "訁": 95631,
+      "謊": 95632,
+      "斣": 95633,
+      "篌": 95634,
+      "‛": 95635,
+      "⒎": 95636,
+      "伜": 95637,
+      "琤": 95638,
+      "齇": 95639,
+      "朩": 95640,
+      "醸": 95641,
+      "応": 95642,
+      "櫳": 95643,
+      "紌": 95644,
+      "臘": 95645,
+      "◫": 95646,
+      "噫": 95647,
+      "伬": 95648,
+      "▷": 95649,
+      "👩🏻‍🦰": 95650,
+      "⑵": 95651,
+      "傞": 95652,
+      "芐": 95653,
+      "顑": 95654,
+      "慪": 95655,
+      "堿": 95656,
+      "👨🏼‍💻": 95657,
+      "吂": 95658,
+      "釒": 95659,
+      "幙": 95660,
+      "麷": 95661,
+      "👩🏻‍❤️‍👨🏾": 95662,
+      "堏": 95663,
+      "巤": 95664,
+      "👒": 95665,
+      "啝": 95666,
+      "珻": 95667,
+      "岼": 95668,
+      "敉": 95669,
+      "曃": 95670,
+      "醗": 95671,
+      "衎": 95672,
+      "鹒": 95673,
+      "烋": 95674,
+      "螇": 95675,
+      "麃": 95676,
+      "舨": 95677,
+      "🇹🇬": 95678,
+      "鍋": 95679,
+      "韗": 95680,
+      "溆": 95681,
+      "⛩": 95682,
+      "籦": 95683,
+      "🎳": 95684,
+      "灉": 95685,
+      "爩": 95686,
+      "🧖‍♂️": 95687,
+      "汵": 95688,
+      "俧": 95689,
+      "⃒": 95690,
+      "鲭": 95691,
+      "乸": 95692,
+      "蟉": 95693,
+      "錐": 95694,
+      "🐈": 95695,
+      "冁": 95696,
+      "艙": 95697,
+      "獚": 95698,
+      "嫳": 95699,
+      "甐": 95700,
+      "駪": 95701,
+      "鲒": 95702,
+      "唀": 95703,
+      "餉": 95704,
+      "嵓": 95705,
+      "👩🏼‍💼": 95706,
+      "揓": 95707,
+      "🙎🏼‍♀️": 95708,
+      "虅": 95709,
+      "ɖ": 95710,
+      "嶷": 95711,
+      "灮": 95712,
+      "坁": 95713,
+      "圐": 95714,
+      "岠": 95715,
+      "慿": 95716,
+      "欕": 95717,
+      "😭": 95718,
+      "挌": 95719,
+      "雩": 95720,
+      "💂🏾‍♀️": 95721,
+      "踤": 95722,
+      "夬": 95723,
+      "啨": 95724,
+      "鄌": 95725,
+      "飛": 95726,
+      "玨": 95727,
+      "邿": 95728,
+      "膈": 95729,
+      "賲": 95730,
+      "褁": 95731,
+      "鷲": 95732,
+      "茓": 95733,
+      "熥": 95734,
+      "墤": 95735,
+      "稟": 95736,
+      "呫": 95737,
+      "ˀ": 95738,
+      "筻": 95739,
+      "麱": 95740,
+      "👩🏾‍🎨": 95741,
+      "脕": 95742,
+      "靛": 95743,
+      "苢": 95744,
+      "尭": 95745,
+      "阕": 95746,
+      "🙅‍♀️": 95747,
+      "鋻": 95748,
+      "轝": 95749,
+      "菫": 95750,
+      "⏐": 95751,
+      "蔪": 95752,
+      "鄧": 95753,
+      "鞔": 95754,
+      "🫀": 95755,
+      "👨🏽‍🔬": 95756,
+      "詻": 95757,
+      "鍺": 95758,
+      "𓀙": 95759,
+      "鬝": 95760,
+      "皟": 95761,
+      "膿": 95762,
+      "佅": 95763,
+      "懊": 95764,
+      "黼": 95765,
+      "🚄": 95766,
+      "錾": 95767,
+      "損": 95768,
+      "沴": 95769,
+      "⋲": 95770,
+      "頤": 95771,
+      "▏": 95772,
+      "齾": 95773,
+      "頷": 95774,
+      "窯": 95775,
+      "🫴🏽": 95776,
+      "謻": 95777,
+      "🤨": 95778,
+      "焁": 95779,
+      "🇿🇲": 95780,
+      "ΰ": 95781,
+      "⌆": 95782,
+      "舓": 95783,
+      "锧": 95784,
+      "肁": 95785,
+      "悞": 95786,
+      "晉": 95787,
+      "鋾": 95788,
+      "皔": 95789,
+      "🏇🏾": 95790,
+      "燯": 95791,
+      "樓": 95792,
+      "⚉": 95793,
+      "毾": 95794,
+      "盶": 95795,
+      "髁": 95796,
+      "堯": 95797,
+      "ȸ": 95798,
+      "鎧": 95799,
+      "ƅ": 95800,
+      "囷": 95801,
+      "籉": 95802,
+      "鬶": 95803,
+      "Ġ": 95804,
+      "懬": 95805,
+      "陲": 95806,
+      "醅": 95807,
+      "🦸‍♂️": 95808,
+      "🧔": 95809,
+      "憆": 95810,
+      "😺": 95811,
+      "虌": 95812,
+      "黪": 95813,
+      "Ⓞ": 95814,
+      "➜": 95815,
+      "騧": 95816,
+      "鎽": 95817,
+      "穬": 95818,
+      "鋲": 95819,
+      "螼": 95820,
+      "蚉": 95821,
+      "擷": 95822,
+      "擫": 95823,
+      "岜": 95824,
+      "蚏": 95825,
+      "🏄🏽‍♀️": 95826,
+      "獈": 95827,
+      "笢": 95828,
+      "✍🏾": 95829,
+      "僐": 95830,
+      "洅": 95831,
+      "濨": 95832,
+      "Ŷ": 95833,
+      "縦": 95834,
+      "肍": 95835,
+      "醃": 95836,
+      "軣": 95837,
+      "暘": 95838,
+      "秶": 95839,
+      "氡": 95840,
+      "🇹🇻": 95841,
+      "豅": 95842,
+      "邼": 95843,
+      "蔐": 95844,
+      "蘬": 95845,
+      "怵": 95846,
+      "👈🏾": 95847,
+      "箐": 95848,
+      "ɕ": 95849,
+      "鰈": 95850,
+      "饫": 95851,
+      "𓀎": 95852,
+      "嚡": 95853,
+      "轁": 95854,
+      "̝": 95855,
+      "璌": 95856,
+      "腧": 95857,
+      "澍": 95858,
+      "囟": 95859,
+      "熀": 95860,
+      "🫏": 95861,
+      "蝊": 95862,
+      "紺": 95863,
+      "唦": 95864,
+      "🤳🏽": 95865,
+      "鉣": 95866,
+      "砛": 95867,
+      "🚵🏿‍♂️": 95868,
+      "獶": 95869,
+      "筜": 95870,
+      "軡": 95871,
+      "囻": 95872,
+      "弫": 95873,
+      "綬": 95874,
+      "醁": 95875,
+      "輒": 95876,
+      "攽": 95877,
+      "骾": 95878,
+      "扏": 95879,
+      "艫": 95880,
+      "擖": 95881,
+      "魽": 95882,
+      "誎": 95883,
+      "仹": 95884,
+      "璤": 95885,
+      "燪": 95886,
+      "梫": 95887,
+      "雤": 95888,
+      "槑": 95889,
+      "丒": 95890,
+      "⃰": 95891,
+      "芆": 95892,
+      "俚": 95893,
+      "億": 95894,
+      "鞅": 95895,
+      "蒽": 95896,
+      "🗒": 95897,
+      "拸": 95898,
+      "頼": 95899,
+      "唭": 95900,
+      "蛃": 95901,
+      "🏤": 95902,
+      "蕰": 95903,
+      "紶": 95904,
+      "峱": 95905,
+      "鼱": 95906,
+      "榫": 95907,
+      "淏": 95908,
+      "⇆": 95909,
+      "綞": 95910,
+      "✷": 95911,
+      "聖": 95912,
+      "瑈": 95913,
+      "弩": 95914,
+      "磜": 95915,
+      "🤴🏻": 95916,
+      "瓬": 95917,
+      "杛": 95918,
+      "🧑🏼‍🚀": 95919,
+      "眵": 95920,
+      "宒": 95921,
+      "廒": 95922,
+      "駯": 95923,
+      "⏳": 95924,
+      "抜": 95925,
+      "🦜": 95926,
+      "舦": 95927,
+      "🧒🏾": 95928,
+      "蔃": 95929,
+      "lj": 95930,
+      "蔎": 95931,
+      "澥": 95932,
+      "阰": 95933,
+      "偙": 95934,
+      "🇨🇾": 95935,
+      "昃": 95936,
+      "韾": 95937,
+      "氄": 95938,
+      "裦": 95939,
+      "榝": 95940,
+      "蠍": 95941,
+      "▗": 95942,
+      "ǡ": 95943,
+      "譸": 95944,
+      "樚": 95945,
+      "嗵": 95946,
+      "銭": 95947,
+      "⌍": 95948,
+      "鵋": 95949,
+      "Nj": 95950,
+      "桽": 95951,
+      "矄": 95952,
+      "絗": 95953,
+      "饈": 95954,
+      "鐩": 95955,
+      "陖": 95956,
+      "∱": 95957,
+      "👨‍🚀": 95958,
+      "勓": 95959,
+      "👰🏻‍♂️": 95960,
+      "嵳": 95961,
+      "鑛": 95962,
+      "炧": 95963,
+      "晢": 95964,
+      "崚": 95965,
+      "狝": 95966,
+      "伹": 95967,
+      "煔": 95968,
+      "🤌🏾": 95969,
+      "諌": 95970,
+      "👨🏾‍❤️‍👨🏾": 95971,
+      "趯": 95972,
+      "劬": 95973,
+      "襫": 95974,
+      "◝": 95975,
+      "🧑🏻‍🏫": 95976,
+      "苭": 95977,
+      "刿": 95978,
+      "吭": 95979,
+      "奜": 95980,
+      "蠥": 95981,
+      "跭": 95982,
+      "潕": 95983,
+      "撹": 95984,
+      "漙": 95985,
+      "⃙": 95986,
+      "🧢": 95987,
+      "枞": 95988,
+      "潹": 95989,
+      "袎": 95990,
+      "🦦": 95991,
+      "沍": 95992,
+      "獂": 95993,
+      "靿": 95994,
+      "玀": 95995,
+      "鶟": 95996,
+      "₃": 95997,
+      "宼": 95998,
+      "Ƨ": 95999,
+      "靍": 96000,
+      "鋡": 96001,
+      "蹋": 96002,
+      "▹": 96003,
+      "铴": 96004,
+      "樴": 96005,
+      "鲞": 96006,
+      "綶": 96007,
+      "憓": 96008,
+      "韰": 96009,
+      "溹": 96010,
+      "婢": 96011,
+      "顂": 96012,
+      "鶫": 96013,
+      "仵": 96014,
+      "迤": 96015,
+      "犎": 96016,
+      "⚃": 96017,
+      "↻": 96018,
+      "🏂🏻": 96019,
+      "琲": 96020,
+      "踆": 96021,
+      "邬": 96022,
+      "浍": 96023,
+      "飹": 96024,
+      "倳": 96025,
+      "⏭": 96026,
+      "💆🏼‍♂️": 96027,
+      "蘦": 96028,
+      "▢": 96029,
+      "迍": 96030,
+      "廝": 96031,
+      "龔": 96032,
+      "ʑ": 96033,
+      "摃": 96034,
+      "鸅": 96035,
+      "🦻🏾": 96036,
+      "馫": 96037,
+      "硊": 96038,
+      "駺": 96039,
+      "埑": 96040,
+      "🚼": 96041,
+      "絥": 96042,
+      "巏": 96043,
+      "哐": 96044,
+      "鐗": 96045,
+      "氍": 96046,
+      "🦂": 96047,
+      "🤷‍♂️": 96048,
+      "澑": 96049,
+      "瓏": 96050,
+      "🫸🏻": 96051,
+      "襊": 96052,
+      "📆": 96053,
+      "⌸": 96054,
+      "毄": 96055,
+      "匔": 96056,
+      "鬺": 96057,
+      "煢": 96058,
+      "燭": 96059,
+      "厜": 96060,
+      "萞": 96061,
+      "鱛": 96062,
+      "辸": 96063,
+      "毎": 96064,
+      "爟": 96065,
+      "⁏": 96066,
+      "耭": 96067,
+      "偝": 96068,
+      "堼": 96069,
+      "欮": 96070,
+      "澕": 96071,
+      "罶": 96072,
+      "Ȑ": 96073,
+      "敔": 96074,
+      "覡": 96075,
+      "絙": 96076,
+      "苋": 96077,
+      "聉": 96078,
+      "🇸🇻": 96079,
+      "觥": 96080,
+      "闢": 96081,
+      "徾": 96082,
+      "👡": 96083,
+      "樣": 96084,
+      "🥧": 96085,
+      "觋": 96086,
+      "鴗": 96087,
+      "撎": 96088,
+      "箪": 96089,
+      "磦": 96090,
+      "筿": 96091,
+      "媷": 96092,
+      "逤": 96093,
+      "牓": 96094,
+      "爍": 96095,
+      "鶀": 96096,
+      "≂": 96097,
+      "蜖": 96098,
+      "栧": 96099,
+      "戦": 96100,
+      "🧑‍🦽": 96101,
+      "姞": 96102,
+      "脮": 96103,
+      "壦": 96104,
+      "🧑🏾‍🚀": 96105,
+      "⌨": 96106,
+      "Ȟ": 96107,
+      "🧼": 96108,
+      "吘": 96109,
+      "🧰": 96110,
+      "蕣": 96111,
+      "ƍ": 96112,
+      "Ť": 96113,
+      "瀒": 96114,
+      "Ȝ": 96115,
+      "搠": 96116,
+      "杘": 96117,
+      "镳": 96118,
+      "騻": 96119,
+      "̪": 96120,
+      "猸": 96121,
+      "鏆": 96122,
+      "🧑‍🦯": 96123,
+      "曣": 96124,
+      "⒉": 96125,
+      "釋": 96126,
+      "過": 96127,
+      "蔾": 96128,
+      "暋": 96129,
+      "橤": 96130,
+      "囏": 96131,
+      "锹": 96132,
+      "昸": 96133,
+      "昻": 96134,
+      "輝": 96135,
+      "胿": 96136,
+      "橠": 96137,
+      "珫": 96138,
+      "悆": 96139,
+      "柌": 96140,
+      "⓪": 96141,
+      "玸": 96142,
+      "灩": 96143,
+      "👳🏼‍♀️": 96144,
+      "🧑🏽‍❤️‍💋‍🧑🏻": 96145,
+      "玳": 96146,
+      "髫": 96147,
+      "眬": 96148,
+      "頀": 96149,
+      "邽": 96150,
+      "槄": 96151,
+      "囘": 96152,
+      "辂": 96153,
+      "捑": 96154,
+      "屗": 96155,
+      "⎥": 96156,
+      "缗": 96157,
+      "讙": 96158,
+      "筘": 96159,
+      "橝": 96160,
+      "貵": 96161,
+      "◳": 96162,
+      "👭🏻": 96163,
+      "確": 96164,
+      "蓤": 96165,
+      "惙": 96166,
+      "磬": 96167,
+      "謷": 96168,
+      "歴": 96169,
+      "👩🏿‍❤️‍👩🏿": 96170,
+      "鍏": 96171,
+      "鵓": 96172,
+      "糏": 96173,
+      "曖": 96174,
+      "🧔‍♀️": 96175,
+      "袄": 96176,
+      "␜": 96177,
+      "鐬": 96178,
+      "趻": 96179,
+      "驊": 96180,
+      "夃": 96181,
+      "孻": 96182,
+      "嚚": 96183,
+      "熩": 96184,
+      "祧": 96185,
+      "贎": 96186,
+      "憜": 96187,
+      "斢": 96188,
+      "郕": 96189,
+      "鸀": 96190,
+      "🧍🏼": 96191,
+      "⚭": 96192,
+      "誐": 96193,
+      "碏": 96194,
+      "慓": 96195,
+      "💆🏽‍♀️": 96196,
+      "铿": 96197,
+      "🔉": 96198,
+      "踦": 96199,
+      "🫅": 96200,
+      "玐": 96201,
+      "📘": 96202,
+      "焒": 96203,
+      "͢": 96204,
+      "屟": 96205,
+      "涁": 96206,
+      "🙍🏿": 96207,
+      "ɗ": 96208,
+      "硛": 96209,
+      "鵻": 96210,
+      "🤸🏻‍♀️": 96211,
+      "␝": 96212,
+      "📏": 96213,
+      "貾": 96214,
+      "瑭": 96215,
+      "詡": 96216,
+      "👰": 96217,
+      "冨": 96218,
+      "镅": 96219,
+      "褆": 96220,
+      "Ⅶ": 96221,
+      "↭": 96222,
+      "👴🏾": 96223,
+      "₼": 96224,
+      "泮": 96225,
+      "娵": 96226,
+      "缟": 96227,
+      "乀": 96228,
+      "緱": 96229,
+      "Å": 96230,
+      "IJ": 96231,
+      "悘": 96232,
+      "📪": 96233,
+      "姵": 96234,
+      "🚶🏻‍♂️": 96235,
+      "胪": 96236,
+      "ϯ": 96237,
+      "欬": 96238,
+      "剷": 96239,
+      "縨": 96240,
+      "⌚": 96241,
+      "匤": 96242,
+      "繷": 96243,
+      "褗": 96244,
+      "🌓": 96245,
+      "瞠": 96246,
+      "棿": 96247,
+      "粇": 96248,
+      "馓": 96249,
+      "糳": 96250,
+      "檃": 96251,
+      "⍠": 96252,
+      "̄": 96253,
+      "擛": 96254,
+      "叅": 96255,
+      "嶇": 96256,
+      "🤫": 96257,
+      "伄": 96258,
+      "鍩": 96259,
+      "笚": 96260,
+      "⍝": 96261,
+      "🐕‍🦺": 96262,
+      "狰": 96263,
+      "漶": 96264,
+      "礐": 96265,
+      "🗺": 96266,
+      "駝": 96267,
+      "訡": 96268,
+      "⋑": 96269,
+      "蔌": 96270,
+      "袴": 96271,
+      "🧑‍🌾": 96272,
+      "炶": 96273,
+      "宂": 96274,
+      "潻": 96275,
+      "🇸🇳": 96276,
+      "與": 96277,
+      "撫": 96278,
+      "👨🏽‍💼": 96279,
+      "紥": 96280,
+      "稄": 96281,
+      "Ⱥ": 96282,
+      "🤳": 96283,
+      "🏴󠁧󠁢󠁳󠁣󠁴󠁿": 96284,
+      "͈": 96285,
+      "嶰": 96286,
+      "丩": 96287,
+      "熣": 96288,
+      "埜": 96289,
+      "俰": 96290,
+      "⋷": 96291,
+      "鈡": 96292,
+      "醮": 96293,
+      "闼": 96294,
+      "╦": 96295,
+      "🐖": 96296,
+      "ȝ": 96297,
+      "眭": 96298,
+      "ĺ": 96299,
+      "🧑🏻‍⚕️": 96300,
+      "🕒": 96301,
+      "栉": 96302,
+      "岯": 96303,
+      "蜿": 96304,
+      "叜": 96305,
+      "鏡": 96306,
+      "睰": 96307,
+      "愊": 96308,
+      "蓪": 96309,
+      "鉵": 96310,
+      "6️⃣": 96311,
+      "瀸": 96312,
+      "簤": 96313,
+      "ⓦ": 96314,
+      "甮": 96315,
+      "釃": 96316,
+      "🎩": 96317,
+      "┰": 96318,
+      "莼": 96319,
+      "鴉": 96320,
+      "犅": 96321,
+      "郲": 96322,
+      "⏬": 96323,
+      "˕": 96324,
+      "揩": 96325,
+      "耡": 96326,
+      "惸": 96327,
+      "鉨": 96328,
+      "朄": 96329,
+      "埅": 96330,
+      "晞": 96331,
+      "疥": 96332,
+      "礦": 96333,
+      "Ϫ": 96334,
+      "埻": 96335,
+      "☨": 96336,
+      "苈": 96337,
+      "勫": 96338,
+      "鲘": 96339,
+      "庽": 96340,
+      "眍": 96341,
+      "憹": 96342,
+      "👩🏽‍🦳": 96343,
+      "╱": 96344,
+      "諹": 96345,
+      "呪": 96346,
+      "緟": 96347,
+      "嚯": 96348,
+      "蒐": 96349,
+      "🫕": 96350,
+      "蠸": 96351,
+      "Ｂ": 96352,
+      "閴": 96353,
+      "閰": 96354,
+      "╢": 96355,
+      "玍": 96356,
+      "繩": 96357,
+      "💓": 96358,
+      "巰": 96359,
+      "🛍": 96360,
+      "烐": 96361,
+      "萶": 96362,
+      "犚": 96363,
+      "拰": 96364,
+      "剴": 96365,
+      "訊": 96366,
+      "摜": 96367,
+      "🕧": 96368,
+      "攷": 96369,
+      "釭": 96370,
+      "曬": 96371,
+      "蔼": 96372,
+      "祍": 96373,
+      "鸉": 96374,
+      "镛": 96375,
+      "🙃": 96376,
+      "罣": 96377,
+      "荱": 96378,
+      "͔": 96379,
+      "屛": 96380,
+      "🇱🇾": 96381,
+      "謵": 96382,
+      "琯": 96383,
+      "奾": 96384,
+      "楈": 96385,
+      "夈": 96386,
+      "鑍": 96387,
+      "🫷🏽": 96388,
+      "鄢": 96389,
+      "鑆": 96390,
+      "搢": 96391,
+      "睥": 96392,
+      "岉": 96393,
+      "⁽": 96394,
+      "🇨🇮": 96395,
+      "簣": 96396,
+      "🤝": 96397,
+      "娨": 96398,
+      "禝": 96399,
+      "麧": 96400,
+      "閆": 96401,
+      "辋": 96402,
+      "賩": 96403,
+      "皧": 96404,
+      "咞": 96405,
+      "嗎": 96406,
+      "攮": 96407,
+      "軿": 96408,
+      "🧎🏼‍♀️": 96409,
+      "壵": 96410,
+      "盭": 96411,
+      "Ā": 96412,
+      "勬": 96413,
+      "槃": 96414,
+      "鳎": 96415,
+      "😞": 96416,
+      "岄": 96417,
+      "駵": 96418,
+      "羼": 96419,
+      "構": 96420,
+      "欄": 96421,
+      "箢": 96422,
+      "圙": 96423,
+      "帡": 96424,
+      "鏽": 96425,
+      "🏓": 96426,
+      "🏄🏻": 96427,
+      "❌": 96428,
+      "祮": 96429,
+      "🧗🏿": 96430,
+      "锠": 96431,
+      "燫": 96432,
+      "鮁": 96433,
+      "耷": 96434,
+      "👗": 96435,
+      "🙇🏿": 96436,
+      "臐": 96437,
+      "🧑‍🦳": 96438,
+      "沚": 96439,
+      "炱": 96440,
+      "鉩": 96441,
+      "穝": 96442,
+      "🍭": 96443,
+      "↝": 96444,
+      "👨🏻‍❤️‍💋‍👨🏾": 96445,
+      "℮": 96446,
+      "鏲": 96447,
+      "薈": 96448,
+      "ˣ": 96449,
+      "蔤": 96450,
+      "🧔🏾‍♀️": 96451,
+      "泅": 96452,
+      "鄄": 96453,
+      "鸮": 96454,
+      "抁": 96455,
+      "攄": 96456,
+      "羧": 96457,
+      "╥": 96458,
+      "憦": 96459,
+      "朆": 96460,
+      "🧜🏾": 96461,
+      "ˆ": 96462,
+      "蝚": 96463,
+      "縿": 96464,
+      "霊": 96465,
+      "🤯": 96466,
+      "滰": 96467,
+      "凕": 96468,
+      "骍": 96469,
+      "綂": 96470,
+      "鳢": 96471,
+      "🂠": 96472,
+      "鴧": 96473,
+      "奙": 96474,
+      "虻": 96475,
+      "崳": 96476,
+      "🍡": 96477,
+      "逎": 96478,
+      "媪": 96479,
+      "閡": 96480,
+      "🫄": 96481,
+      "襧": 96482,
+      "崞": 96483,
+      "⎵": 96484,
+      "氺": 96485,
+      "鷸": 96486,
+      "🤽🏿": 96487,
+      "🙏🏼": 96488,
+      "👐🏾": 96489,
+      "橰": 96490,
+      "亊": 96491,
+      "鼽": 96492,
+      "🧑🏼‍🦯": 96493,
+      "⊩": 96494,
+      "詰": 96495,
+      "馡": 96496,
+      "👨🏾‍🦰": 96497,
+      "👩🏼‍❤️‍👨🏼": 96498,
+      "櫶": 96499,
+      "臙": 96500,
+      "顇": 96501,
+      "郋": 96502,
+      "翛": 96503,
+      "🫱🏽‍🫲🏻": 96504,
+      "岿": 96505,
+      "⌋": 96506,
+      "👳🏼": 96507,
+      "🏌🏾‍♀️": 96508,
+      "🏈": 96509,
+      "℔": 96510,
+      "皏": 96511,
+      "嶏": 96512,
+      "憫": 96513,
+      "縗": 96514,
+      "⊶": 96515,
+      "霪": 96516,
+      "堬": 96517,
+      "殠": 96518,
+      "繎": 96519,
+      "囬": 96520,
+      "驠": 96521,
+      "纅": 96522,
+      "韹": 96523,
+      "庯": 96524,
+      "椹": 96525,
+      "訔": 96526,
+      "弲": 96527,
+      "🎢": 96528,
+      "垪": 96529,
+      "轠": 96530,
+      "殫": 96531,
+      "🛌🏿": 96532,
+      "孪": 96533,
+      "🕺": 96534,
+      "🧚‍♀️": 96535,
+      "鎺": 96536,
+      "⁕": 96537,
+      "窲": 96538,
+      "鰖": 96539,
+      "🧗🏽‍♀️": 96540,
+      "誻": 96541,
+      "🕵🏼‍♂️": 96542,
+      "蟢": 96543,
+      "鼷": 96544,
+      "Ό": 96545,
+      "捸": 96546,
+      "👃🏽": 96547,
+      "憙": 96548,
+      "掊": 96549,
+      "蘣": 96550,
+      "輵": 96551,
+      "殌": 96552,
+      "螪": 96553,
+      "蓇": 96554,
+      "扅": 96555,
+      "劼": 96556,
+      "鈜": 96557,
+      "諶": 96558,
+      "🙋🏾‍♀️": 96559,
+      "儵": 96560,
+      "菵": 96561,
+      "🤪": 96562,
+      "憞": 96563,
+      "鏖": 96564,
+      "💇🏼": 96565,
+      "覵": 96566,
+      "⁳": 96567,
+      "畗": 96568,
+      "偭": 96569,
+      "箟": 96570,
+      "绶": 96571,
+      "踗": 96572,
+      "耖": 96573,
+      "🧟‍♂️": 96574,
+      "篢": 96575,
+      "🚴🏻‍♂️": 96576,
+      "瘭": 96577,
+      "桡": 96578,
+      "壱": 96579,
+      "鎖": 96580,
+      "₆": 96581,
+      "猞": 96582,
+      "🔟": 96583,
+      "糵": 96584,
+      "幷": 96585,
+      "铇": 96586,
+      "榺": 96587,
+      "苃": 96588,
+      "┩": 96589,
+      "捝": 96590,
+      "👨‍🦽": 96591,
+      "茳": 96592,
+      "僑": 96593,
+      "螉": 96594,
+      "岤": 96595,
+      "塒": 96596,
+      "🙌🏻": 96597,
+      "咊": 96598,
+      "薜": 96599,
+      "躰": 96600,
+      "靅": 96601,
+      "槯": 96602,
+      "廠": 96603,
+      "🔼": 96604,
+      "̈": 96605,
+      "郐": 96606,
+      "粡": 96607,
+      "鬌": 96608,
+      "昡": 96609,
+      "👨🏻‍🦱": 96610,
+      "魲": 96611,
+      "罙": 96612,
+      "👩🏿‍🦱": 96613,
+      "涶": 96614,
+      "茪": 96615,
+      "姪": 96616,
+      "ƿ": 96617,
+      "🇪🇸": 96618,
+      "怷": 96619,
+      "蔨": 96620,
+      "浛": 96621,
+      "͐": 96622,
+      "鳏": 96623,
+      "🧘🏽": 96624,
+      "賌": 96625,
+      "豮": 96626,
+      "膆": 96627,
+      "🇲🇶": 96628,
+      "檣": 96629,
+      "齔": 96630,
+      "鐟": 96631,
+      "壡": 96632,
+      "≣": 96633,
+      "⍽": 96634,
+      "⌘": 96635,
+      "铻": 96636,
+      "衝": 96637,
+      "🚔": 96638,
+      "顋": 96639,
+      "嬱": 96640,
+      "輺": 96641,
+      "繐": 96642,
+      "龋": 96643,
+      "🦏": 96644,
+      "鋓": 96645,
+      "ʃ": 96646,
+      "🦉": 96647,
+      "豢": 96648,
+      "佂": 96649,
+      "罈": 96650,
+      "睵": 96651,
+      "♂": 96652,
+      "臹": 96653,
+      "蟯": 96654,
+      "鋿": 96655,
+      "揀": 96656,
+      "瀇": 96657,
+      "⌣": 96658,
+      "🤽🏽‍♂️": 96659,
+      "🧑🏽‍❤️‍🧑🏾": 96660,
+      "蔧": 96661,
+      "裥": 96662,
+      "🇳🇴": 96663,
+      "酼": 96664,
+      "⊗": 96665,
+      "樐": 96666,
+      "凁": 96667,
+      "訏": 96668,
+      "擹": 96669,
+      "懞": 96670,
+      "檢": 96671,
+      "鱫": 96672,
+      "≅": 96673,
+      "∝": 96674,
+      "隸": 96675,
+      "雵": 96676,
+      "👩🏽‍🔬": 96677,
+      "蘱": 96678,
+      "謭": 96679,
+      "墂": 96680,
+      "颉": 96681,
+      "⁸": 96682,
+      "堺": 96683,
+      "蔘": 96684,
+      "⛀": 96685,
+      "橛": 96686,
+      "♖": 96687,
+      "骴": 96688,
+      "👨‍👨‍👧‍👧": 96689,
+      "␵": 96690,
+      "庋": 96691,
+      "🫷🏼": 96692,
+      "涢": 96693,
+      "屻": 96694,
+      "⑭": 96695,
+      "蟅": 96696,
+      "👨‍🦲": 96697,
+      "嶋": 96698,
+      "罯": 96699,
+      "誚": 96700,
+      "彀": 96701,
+      "♱": 96702,
+      "醠": 96703,
+      "廇": 96704,
+      "坘": 96705,
+      "👪": 96706,
+      "🕵️‍♀️": 96707,
+      "揼": 96708,
+      "嫃": 96709,
+      "␺": 96710,
+      "穋": 96711,
+      "🏃": 96712,
+      "巯": 96713,
+      "绐": 96714,
+      "⁅": 96715,
+      "獘": 96716,
+      "鋞": 96717,
+      "憖": 96718,
+      "⃳": 96719,
+      "赥": 96720,
+      "戻": 96721,
+      "齢": 96722,
+      "曆": 96723,
+      "骮": 96724,
+      "懱": 96725,
+      "厊": 96726,
+      "鎊": 96727,
+      "嶞": 96728,
+      "💪🏾": 96729,
+      "瞀": 96730,
+      "堑": 96731,
+      "慒": 96732,
+      "颥": 96733,
+      "浗": 96734,
+      "譊": 96735,
+      "巐": 96736,
+      "吱": 96737,
+      "🦁": 96738,
+      "🥛": 96739,
+      "⇽": 96740,
+      "🚀": 96741,
+      "🇯🇲": 96742,
+      "̰": 96743,
+      "⍹": 96744,
+      "鴓": 96745,
+      "騩": 96746,
+      "儯": 96747,
+      "翶": 96748,
+      "纏": 96749,
+      "🙋‍♀️": 96750,
+      "鬄": 96751,
+      "🧏🏽‍♂️": 96752,
+      "☬": 96753,
+      "滢": 96754,
+      "巄": 96755,
+      "℆": 96756,
+      "👳‍♂️": 96757,
+      "樝": 96758,
+      "⍉": 96759,
+      "蜅": 96760,
+      "鉎": 96761,
+      "ȑ": 96762,
+      "谚": 96763,
+      "菈": 96764,
+      "😶‍🌫️": 96765,
+      "穯": 96766,
+      "🌕": 96767,
+      "🚿": 96768,
+      "鮹": 96769,
+      "鏥": 96770,
+      "僣": 96771,
+      "娋": 96772,
+      "蓜": 96773,
+      "🧑🏽‍🦽": 96774,
+      "叇": 96775,
+      "呩": 96776,
+      "🕴🏽": 96777,
+      "♒": 96778,
+      "頌": 96779,
+      "厇": 96780,
+      "🏋🏼": 96781,
+      "🥶": 96782,
+      "揆": 96783,
+      "杝": 96784,
+      "逽": 96785,
+      "🇺🇿": 96786,
+      "盽": 96787,
+      "☒": 96788,
+      "🫃🏼": 96789,
+      "砽": 96790,
+      "冐": 96791,
+      "湦": 96792,
+      "龓": 96793,
+      "瘯": 96794,
+      "坾": 96795,
+      "呰": 96796,
+      "稵": 96797,
+      "🫱🏾‍🫲🏻": 96798,
+      "槢": 96799,
+      "ƺ": 96800,
+      "腠": 96801,
+      "🧑🏽‍🦰": 96802,
+      "豱": 96803,
+      "帏": 96804,
+      "粲": 96805,
+      "勁": 96806,
+      "潧": 96807,
+      "🧑🏾‍🦲": 96808,
+      "Ⅷ": 96809,
+      "齈": 96810,
+      "ⓑ": 96811,
+      "泖": 96812,
+      "裵": 96813,
+      "媲": 96814,
+      "💃🏼": 96815,
+      "艵": 96816,
+      "谺": 96817,
+      "⇓": 96818,
+      "窨": 96819,
+      "珤": 96820,
+      "蹊": 96821,
+      "刅": 96822,
+      "⒃": 96823,
+      "🕍": 96824,
+      "腅": 96825,
+      "刓": 96826,
+      "嫔": 96827,
+      "⏒": 96828,
+      "畧": 96829,
+      "燀": 96830,
+      "騒": 96831,
+      "踟": 96832,
+      "卌": 96833,
+      "蒱": 96834,
+      "厞": 96835,
+      "犾": 96836,
+      "槱": 96837,
+      "鈸": 96838,
+      "鳃": 96839,
+      "🧎": 96840,
+      "👩🏻‍❤️‍💋‍👩🏼": 96841,
+      "婨": 96842,
+      "蔀": 96843,
+      "柵": 96844,
+      "👩‍⚕️": 96845,
+      "叓": 96846,
+      "囧": 96847,
+      "毇": 96848,
+      "陫": 96849,
+      "茩": 96850,
+      "苸": 96851,
+      "🪑": 96852,
+      "婣": 96853,
+      "辁": 96854,
+      "灋": 96855,
+      "嬸": 96856,
+      "啵": 96857,
+      "鮆": 96858,
+      "椚": 96859,
+      "嵷": 96860,
+      "軝": 96861,
+      "擉": 96862,
+      "犳": 96863,
+      "🐬": 96864,
+      "賄": 96865,
+      "✊🏼": 96866,
+      "⑧": 96867,
+      "🇦🇬": 96868,
+      "👦🏿": 96869,
+      "萠": 96870,
+      "獆": 96871,
+      "偆": 96872,
+      "📔": 96873,
+      "饅": 96874,
+      "👩‍🎨": 96875,
+      "👊🏿": 96876,
+      "ʾ": 96877,
+      "😖": 96878,
+      "皲": 96879,
+      "🥈": 96880,
+      "┇": 96881,
+      "🤷🏾‍♀️": 96882,
+      "惼": 96883,
+      "🍉": 96884,
+      "硨": 96885,
+      "鱴": 96886,
+      "洜": 96887,
+      "踈": 96888,
+      "謽": 96889,
+      "劘": 96890,
+      "髴": 96891,
+      "諢": 96892,
+      "⎼": 96893,
+      "傇": 96894,
+      "腎": 96895,
+      "砇": 96896,
+      "🪖": 96897,
+      "逘": 96898,
+      "銦": 96899,
+      "⍄": 96900,
+      "鷴": 96901,
+      "腈": 96902,
+      "伾": 96903,
+      "陯": 96904,
+      "痆": 96905,
+      "寔": 96906,
+      "聽": 96907,
+      "譙": 96908,
+      "魊": 96909,
+      "唍": 96910,
+      "甼": 96911,
+      "齥": 96912,
+      "🚶🏽‍♀️": 96913,
+      "⌏": 96914,
+      "焧": 96915,
+      "🫥": 96916,
+      "欒": 96917,
+      "趪": 96918,
+      "哰": 96919,
+      "讑": 96920,
+      "誇": 96921,
+      "畱": 96922,
+      "🖕🏼": 96923,
+      "𓀐": 96924,
+      "礡": 96925,
+      "❕": 96926,
+      "嵟": 96927,
+      "🧛🏿": 96928,
+      "羦": 96929,
+      "嚿": 96930,
+      "軫": 96931,
+      "鉍": 96932,
+      "鍿": 96933,
+      "ɰ": 96934,
+      "榍": 96935,
+      "⚍": 96936,
+      "👨🏾‍❤️‍💋‍👨🏾": 96937,
+      "陼": 96938,
+      "🤮": 96939,
+      "👩‍🌾": 96940,
+      "哃": 96941,
+      "駾": 96942,
+      "⚳": 96943,
+      "畳": 96944,
+      "榳": 96945,
+      "酃": 96946,
+      "遉": 96947,
+      "鑵": 96948,
+      "祿": 96949,
+      "稖": 96950,
+      "僮": 96951,
+      "蘰": 96952,
+      "咇": 96953,
+      "蔵": 96954,
+      "賦": 96955,
+      "溼": 96956,
+      "ɽ": 96957,
+      "聫": 96958,
+      "👨🏾‍🏫": 96959,
+      "潬": 96960,
+      "鄣": 96961,
+      "🤸🏽‍♂️": 96962,
+      "奵": 96963,
+      "鲧": 96964,
+      "糄": 96965,
+      "楒": 96966,
+      "鷝": 96967,
+      "滷": 96968,
+      "畛": 96969,
+      "鸒": 96970,
+      "鲎": 96971,
+      "郘": 96972,
+      "羶": 96973,
+      "媩": 96974,
+      "匩": 96975,
+      "🌩": 96976,
+      "捱": 96977,
+      "🇵🇱": 96978,
+      "卹": 96979,
+      "😳": 96980,
+      "暡": 96981,
+      "淶": 96982,
+      "黣": 96983,
+      "穛": 96984,
+      "蠨": 96985,
+      "鳊": 96986,
+      "🦸🏾‍♂️": 96987,
+      "塖": 96988,
+      "癿": 96989,
+      "灨": 96990,
+      "欖": 96991,
+      "🙏": 96992,
+      "👨🏿‍🍼": 96993,
+      "弮": 96994,
+      "碃": 96995,
+      "埏": 96996,
+      "俢": 96997,
+      "Lj": 96998,
+      "詽": 96999,
+      "Ĕ": 97000,
+      "‟": 97001,
+      "籖": 97002,
+      "潛": 97003,
+      "🪓": 97004,
+      "💻": 97005,
+      "銌": 97006,
+      "裟": 97007,
+      "瑲": 97008,
+      "糗": 97009,
+      "銮": 97010,
+      "潾": 97011,
+      "篟": 97012,
+      "👫🏾": 97013,
+      "渟": 97014,
+      "墽": 97015,
+      "姼": 97016,
+      "婓": 97017,
+      "玏": 97018,
+      "撲": 97019,
+      "💋": 97020,
+      "Ⓙ": 97021,
+      "熌": 97022,
+      "槖": 97023,
+      "👛": 97024,
+      "📕": 97025,
+      "躹": 97026,
+      "撝": 97027,
+      "🙎🏻": 97028,
+      "⃣": 97029,
+      "🧍🏽": 97030,
+      "蛰": 97031,
+      "裿": 97032,
+      "🤳🏼": 97033,
+      "爡": 97034,
+      "喰": 97035,
+      "卛": 97036,
+      "Û": 97037,
+      "莌": 97038,
+      "霠": 97039,
+      "鸃": 97040,
+      "跆": 97041,
+      "齜": 97042,
+      "骟": 97043,
+      "喅": 97044,
+      "駇": 97045,
+      "🪤": 97046,
+      "擠": 97047,
+      "橂": 97048,
+      "畕": 97049,
+      "筤": 97050,
+      "🏞": 97051,
+      "慱": 97052,
+      "ơ": 97053,
+      "🤦🏼": 97054,
+      "岥": 97055,
+      "狑": 97056,
+      "🤵🏿‍♀️": 97057,
+      "˫": 97058,
+      "筰": 97059,
+      "旣": 97060,
+      "阡": 97061,
+      "隕": 97062,
+      "⁒": 97063,
+      "襅": 97064,
+      "輹": 97065,
+      "👳🏻‍♀️": 97066,
+      "搑": 97067,
+      "褣": 97068,
+      "妣": 97069,
+      "⋱": 97070,
+      "醹": 97071,
+      "🤏🏿": 97072,
+      "鋧": 97073,
+      "蔻": 97074,
+      "焥": 97075,
+      "🫱🏾‍🫲🏿": 97076,
+      "讣": 97077,
+      "💁🏼": 97078,
+      "悌": 97079,
+      "☎": 97080,
+      "璳": 97081,
+      "椃": 97082,
+      "Ƭ": 97083,
+      "👩🏽‍❤️‍👩🏽": 97084,
+      "紝": 97085,
+      "塡": 97086,
+      "粸": 97087,
+      "唡": 97088,
+      "ⓖ": 97089,
+      "🟧": 97090,
+      "胓": 97091,
+      "矤": 97092,
+      "煖": 97093,
+      "浕": 97094,
+      "鋶": 97095,
+      "燱": 97096,
+      "蟭": 97097,
+      "楛": 97098,
+      "👩🏻‍❤️‍💋‍👨🏾": 97099,
+      "👢": 97100,
+      "⑯": 97101,
+      "訓": 97102,
+      "寠": 97103,
+      "姸": 97104,
+      "狈": 97105,
+      "佊": 97106,
+      "🤵🏾‍♀️": 97107,
+      "耫": 97108,
+      "紆": 97109,
+      "檽": 97110,
+      "蛬": 97111,
+      "冧": 97112,
+      "👉🏾": 97113,
+      "饣": 97114,
+      "貛": 97115,
+      "茣": 97116,
+      "娤": 97117,
+      "栝": 97118,
+      "鬙": 97119,
+      "🚶🏼": 97120,
+      "⌝": 97121,
+      "⊫": 97122,
+      "蜠": 97123,
+      "桤": 97124,
+      "炘": 97125,
+      "🧑🏿‍🏫": 97126,
+      "磠": 97127,
+      "▮": 97128,
+      "ȳ": 97129,
+      "🧗🏿‍♂️": 97130,
+      "啚": 97131,
+      "⁲": 97132,
+      "澛": 97133,
+      "槸": 97134,
+      "🧑🏻‍⚖️": 97135,
+      "眛": 97136,
+      "峮": 97137,
+      "坨": 97138,
+      "🌝": 97139,
+      "胍": 97140,
+      "9️⃣": 97141,
+      "砟": 97142,
+      "鯲": 97143,
+      "佸": 97144,
+      "鮕": 97145,
+      "堧": 97146,
+      "⃿": 97147,
+      "哾": 97148,
+      "叾": 97149,
+      "鳦": 97150,
+      "飬": 97151,
+      "戇": 97152,
+      "鈨": 97153,
+      "襉": 97154,
+      "槅": 97155,
+      "🧑🏻‍❤️‍🧑🏿": 97156,
+      "鼉": 97157,
+      "🇨🇩": 97158,
+      "鹄": 97159,
+      "鹍": 97160,
+      "襩": 97161,
+      "盋": 97162,
+      "⍌": 97163,
+      "ʜ": 97164,
+      "🏄🏿‍♂️": 97165,
+      "禩": 97166,
+      "庺": 97167,
+      "⁚": 97168,
+      "⛏": 97169,
+      "竧": 97170,
+      "蛭": 97171,
+      "鉤": 97172,
+      "鱗": 97173,
+      "諑": 97174,
+      "湅": 97175,
+      "澊": 97176,
+      "👬": 97177,
+      "葯": 97178,
+      "閳": 97179,
+      "嗖": 97180,
+      "🧑🏾‍🚒": 97181,
+      "嵫": 97182,
+      "沩": 97183,
+      "钎": 97184,
+      "糡": 97185,
+      "Ϭ": 97186,
+      "🗿": 97187,
+      "👩🏻‍❤️‍💋‍👨🏻": 97188,
+      "瓸": 97189,
+      "🚴": 97190,
+      "歔": 97191,
+      "遻": 97192,
+      "👩🏿‍💼": 97193,
+      "箙": 97194,
+      "蠠": 97195,
+      "炌": 97196,
+      "傴": 97197,
+      "礚": 97198,
+      "💨": 97199,
+      "🪕": 97200,
+      "鉼": 97201,
+      "枻": 97202,
+      "菬": 97203,
+      "腞": 97204,
+      "闬": 97205,
+      "🎁": 97206,
+      "袷": 97207,
+      "蝸": 97208,
+      "❤️‍🔥": 97209,
+      "窿": 97210,
+      "娞": 97211,
+      "奄": 97212,
+      "纮": 97213,
+      "🇹🇲": 97214,
+      "難": 97215,
+      "墬": 97216,
+      "嬾": 97217,
+      "雭": 97218,
+      "┹": 97219,
+      "醿": 97220,
+      "組": 97221,
+      "廲": 97222,
+      "🤷🏼‍♀️": 97223,
+      "襹": 97224,
+      "紛": 97225,
+      "匾": 97226,
+      "ų": 97227,
+      "盜": 97228,
+      "琿": 97229,
+      "ℚ": 97230,
+      "缙": 97231,
+      "萬": 97232,
+      "鷡": 97233,
+      "漦": 97234,
+      "🫱🏾": 97235,
+      "幪": 97236,
+      "擔": 97237,
+      "熳": 97238,
+      "熯": 97239,
+      "阾": 97240,
+      "嵐": 97241,
+      "鍙": 97242,
+      "穉": 97243,
+      "瘖": 97244,
+      "殒": 97245,
+      "˝": 97246,
+      "軺": 97247,
+      "懫": 97248,
+      "硤": 97249,
+      "👯": 97250,
+      "噲": 97251,
+      "锿": 97252,
+      "💥": 97253,
+      "☤": 97254,
+      "蛠": 97255,
+      "笏": 97256,
+      "鸝": 97257,
+      "賧": 97258,
+      "豜": 97259,
+      "癭": 97260,
+      "蚠": 97261,
+      "̮": 97262,
+      "ⅸ": 97263,
+      "👼🏿": 97264,
+      "橬": 97265,
+      "墀": 97266,
+      "杫": 97267,
+      "蜽": 97268,
+      "倜": 97269,
+      "↣": 97270,
+      "狹": 97271,
+      "訒": 97272,
+      "ƪ": 97273,
+      "醕": 97274,
+      "莑": 97275,
+      "韘": 97276,
+      "懛": 97277,
+      "綧": 97278,
+      "蔂": 97279,
+      "🧕🏿": 97280,
+      "Ğ": 97281,
+      "橼": 97282,
+      "財": 97283,
+      "鈺": 97284,
+      "↉": 97285,
+      "迼": 97286,
+      "骹": 97287,
+      "齝": 97288,
+      "螾": 97289,
+      "炑": 97290,
+      "瀘": 97291,
+      "粔": 97292,
+      "🤰🏼": 97293,
+      "朡": 97294,
+      "欗": 97295,
+      "幵": 97296,
+      "恈": 97297,
+      "⋗": 97298,
+      "🔝": 97299,
+      "儅": 97300,
+      "卨": 97301,
+      "⚚": 97302,
+      "醍": 97303,
+      "葑": 97304,
+      "♴": 97305,
+      "煈": 97306,
+      "闇": 97307,
+      "祊": 97308,
+      "镝": 97309,
+      "棢": 97310,
+      "皿": 97311,
+      "😛": 97312,
+      "🧑🏾‍❤️‍💋‍🧑🏽": 97313,
+      "欔": 97314,
+      "№": 97315,
+      "殟": 97316,
+      "荢": 97317,
+      "砅": 97318,
+      "搇": 97319,
+      "👩🏿‍🍳": 97320,
+      "🧑🏽‍🎄": 97321,
+      "̓": 97322,
+      "魘": 97323,
+      "͡": 97324,
+      "👨🏾": 97325,
+      "鯠": 97326,
+      "崍": 97327,
+      "蔹": 97328,
+      "⁝": 97329,
+      "♞": 97330,
+      "🧔🏽‍♂️": 97331,
+      "夨": 97332,
+      "檷": 97333,
+      "竚": 97334,
+      "憶": 97335,
+      "抈": 97336,
+      "喚": 97337,
+      "賹": 97338,
+      "殜": 97339,
+      "⋠": 97340,
+      "钫": 97341,
+      "淸": 97342,
+      "蕕": 97343,
+      "綆": 97344,
+      "✪": 97345,
+      "鏙": 97346,
+      "晌": 97347,
+      "姘": 97348,
+      "沷": 97349,
+      "跢": 97350,
+      "🇨🇫": 97351,
+      "𓀜": 97352,
+      "瀞": 97353,
+      "犸": 97354,
+      "痳": 97355,
+      "佁": 97356,
+      "🏊🏾": 97357,
+      "莁": 97358,
+      "姧": 97359,
+      "儚": 97360,
+      "ŗ": 97361,
+      "妷": 97362,
+      "伲": 97363,
+      "龖": 97364,
+      "嶀": 97365,
+      "⊱": 97366,
+      "淥": 97367,
+      "兒": 97368,
+      "🇮🇴": 97369,
+      "渶": 97370,
+      "薫": 97371,
+      "↗": 97372,
+      "▉": 97373,
+      "孋": 97374,
+      "囋": 97375,
+      "亰": 97376,
+      "☕": 97377,
+      "譄": 97378,
+      "堹": 97379,
+      "🦹🏽": 97380,
+      "窞": 97381,
+      "刽": 97382,
+      "苙": 97383,
+      "鍊": 97384,
+      "泴": 97385,
+      "⏊": 97386,
+      "鉳": 97387,
+      "鑬": 97388,
+      "🎎": 97389,
+      "鼰": 97390,
+      "頫": 97391,
+      "迊": 97392,
+      "🕵️‍♂️": 97393,
+      "骱": 97394,
+      "煑": 97395,
+      "🌬": 97396,
+      "樷": 97397,
+      "褝": 97398,
+      "愢": 97399,
+      "楎": 97400,
+      "羨": 97401,
+      "揷": 97402,
+      "彈": 97403,
+      "霨": 97404,
+      "凩": 97405,
+      "挼": 97406,
+      "镓": 97407,
+      "🤸🏿": 97408,
+      "跮": 97409,
+      "樭": 97410,
+      "皦": 97411,
+      "鉈": 97412,
+      "鸑": 97413,
+      "揁": 97414,
+      "␚": 97415,
+      "🤏🏼": 97416,
+      "瘐": 97417,
+      "禋": 97418,
+      "瞤": 97419,
+      "飱": 97420,
+      "🙆🏽‍♀️": 97421,
+      "禭": 97422,
+      "Ǫ": 97423,
+      "艓": 97424,
+      "⊰": 97425,
+      "貝": 97426,
+      "🔻": 97427,
+      "鯤": 97428,
+      "飊": 97429,
+      "麊": 97430,
+      "⏜": 97431,
+      "邖": 97432,
+      "▻": 97433,
+      "笩": 97434,
+      "🧑‍🚀": 97435,
+      "嘩": 97436,
+      "嚐": 97437,
+      "🇱🇨": 97438,
+      "驩": 97439,
+      "楍": 97440,
+      "厤": 97441,
+      "犷": 97442,
+      "傠": 97443,
+      "碼": 97444,
+      "驕": 97445,
+      "茏": 97446,
+      "▛": 97447,
+      "🧔🏽": 97448,
+      "鹩": 97449,
+      "愅": 97450,
+      "畒": 97451,
+      "蒘": 97452,
+      "🙋🏻": 97453,
+      "鍖": 97454,
+      "穑": 97455,
+      "曽": 97456,
+      "挱": 97457,
+      "尦": 97458,
+      "銢": 97459,
+      "晝": 97460,
+      "🙆🏽": 97461,
+      "👨🏾‍❤️‍👨🏿": 97462,
+      "骘": 97463,
+      "暐": 97464,
+      "朊": 97465,
+      "⎕": 97466,
+      "唄": 97467,
+      "Ǿ": 97468,
+      "舮": 97469,
+      "⌵": 97470,
+      "瀼": 97471,
+      "鳷": 97472,
+      "憤": 97473,
+      "罽": 97474,
+      "仠": 97475,
+      "🙎🏿": 97476,
+      "夊": 97477,
+      "敪": 97478,
+      "銏": 97479,
+      "👷🏻": 97480,
+      "裗": 97481,
+      "嘋": 97482,
+      "↪": 97483,
+      "娳": 97484,
+      "骀": 97485,
+      "敁": 97486,
+      "🫶": 97487,
+      "囦": 97488,
+      "褟": 97489,
+      "呛": 97490,
+      "巛": 97491,
+      "ⅱ": 97492,
+      "鋝": 97493,
+      "摙": 97494,
+      "駚": 97495,
+      "砑": 97496,
+      "澚": 97497,
+      "頄": 97498,
+      "粧": 97499,
+      "瘞": 97500,
+      "蘃": 97501,
+      "碀": 97502,
+      "镣": 97503,
+      "禇": 97504,
+      "罠": 97505,
+      "踣": 97506,
+      "膘": 97507,
+      "锛": 97508,
+      "幩": 97509,
+      "磧": 97510,
+      "腝": 97511,
+      "臇": 97512,
+      "覈": 97513,
+      "貓": 97514,
+      "啿": 97515,
+      "霽": 97516,
+      "槼": 97517,
+      "👨‍👩‍👦": 97518,
+      "荝": 97519,
+      "礳": 97520,
+      "🎚": 97521,
+      "🇱🇷": 97522,
+      "遒": 97523,
+      "蝽": 97524,
+      "🇳🇬": 97525,
+      "栆": 97526,
+      "蔉": 97527,
+      "幝": 97528,
+      "忼": 97529,
+      "砙": 97530,
+      "眽": 97531,
+      "艐": 97532,
+      "汜": 97533,
+      "橍": 97534,
+      "傌": 97535,
+      "巶": 97536,
+      "峞": 97537,
+      "🇧🇧": 97538,
+      "牋": 97539,
+      "笓": 97540,
+      "岣": 97541,
+      "熽": 97542,
+      "兗": 97543,
+      "刴": 97544,
+      "燆": 97545,
+      "娝": 97546,
+      "👃🏾": 97547,
+      "鷳": 97548,
+      "擓": 97549,
+      "抸": 97550,
+      "揄": 97551,
+      "蓶": 97552,
+      "🔽": 97553,
+      "療": 97554,
+      "癢": 97555,
+      "磒": 97556,
+      "嶺": 97557,
+      "霔": 97558,
+      "砄": 97559,
+      "飕": 97560,
+      "罄": 97561,
+      "諈": 97562,
+      "喟": 97563,
+      "鍇": 97564,
+      "椬": 97565,
+      "👩🏻‍🦼": 97566,
+      "◭": 97567,
+      "廂": 97568,
+      "車": 97569,
+      "⛚": 97570,
+      "晽": 97571,
+      "浡": 97572,
+      "礣": 97573,
+      "諜": 97574,
+      "擡": 97575,
+      "嶂": 97576,
+      "彡": 97577,
+      "燦": 97578,
+      "鰴": 97579,
+      "皼": 97580,
+      "灘": 97581,
+      "͜": 97582,
+      "Χ": 97583,
+      "🤜🏻": 97584,
+      "┌": 97585,
+      "嵿": 97586,
+      "邆": 97587,
+      "羏": 97588,
+      "⃵": 97589,
+      "昳": 97590,
+      "郖": 97591,
+      "屄": 97592,
+      "珸": 97593,
+      "␘": 97594,
+      "枨": 97595,
+      "✊🏿": 97596,
+      "髞": 97597,
+      "蚙": 97598,
+      "忳": 97599,
+      "搒": 97600,
+      "⛌": 97601,
+      "Ȇ": 97602,
+      "🇲🇦": 97603,
+      "⃃": 97604,
+      "🪔": 97605,
+      "🧏🏾": 97606,
+      "鹔": 97607,
+      "ņ": 97608,
+      "颢": 97609,
+      "👩🏿‍🏭": 97610,
+      "🇹🇫": 97611,
+      "Κ": 97612,
+      "孄": 97613,
+      "輔": 97614,
+      "剆": 97615,
+      "肳": 97616,
+      "樼": 97617,
+      "剚": 97618,
+      "駁": 97619,
+      "眔": 97620,
+      "喤": 97621,
+      "黴": 97622,
+      "👩🏿‍🏫": 97623,
+      "🧎🏾‍♀️": 97624,
+      "啮": 97625,
+      "堳": 97626,
+      "🚶🏿": 97627,
+      "殍": 97628,
+      "🕝": 97629,
+      "晛": 97630,
+      "駳": 97631,
+      "芓": 97632,
+      "镕": 97633,
+      "🇨🇴": 97634,
+      "脡": 97635,
+      "饄": 97636,
+      "灔": 97637,
+      "͑": 97638,
+      "橒": 97639,
+      "鼇": 97640,
+      "詐": 97641,
+      "氊": 97642,
+      "磈": 97643,
+      "湚": 97644,
+      "芌": 97645,
+      "嫏": 97646,
+      "綌": 97647,
+      "闅": 97648,
+      "螙": 97649,
+      "菃": 97650,
+      "跥": 97651,
+      "˳": 97652,
+      "⌫": 97653,
+      "嚳": 97654,
+      "儕": 97655,
+      "團": 97656,
+      "👌🏾": 97657,
+      "🧑🏿‍🍼": 97658,
+      "豃": 97659,
+      "濞": 97660,
+      "🤹🏽": 97661,
+      "🌲": 97662,
+      "艦": 97663,
+      "噦": 97664,
+      "騇": 97665,
+      "迖": 97666,
+      "椄": 97667,
+      "愋": 97668,
+      "黁": 97669,
+      "恗": 97670,
+      "惋": 97671,
+      "昴": 97672,
+      "蘊": 97673,
+      "霥": 97674,
+      "徫": 97675,
+      "钹": 97676,
+      "莗": 97677,
+      "篭": 97678,
+      "飍": 97679,
+      "砷": 97680,
+      "呱": 97681,
+      "觤": 97682,
+      "趧": 97683,
+      "恾": 97684,
+      "🧜‍♂️": 97685,
+      "钶": 97686,
+      "⏋": 97687,
+      "🕴🏿": 97688,
+      "🐅": 97689,
+      "篨": 97690,
+      "牸": 97691,
+      "嵺": 97692,
+      "孆": 97693,
+      "屭": 97694,
+      "奫": 97695,
+      "臒": 97696,
+      "椆": 97697,
+      "躳": 97698,
+      "堫": 97699,
+      "呂": 97700,
+      "🐌": 97701,
+      "剅": 97702,
+      "蚄": 97703,
+      "硥": 97704,
+      "鶎": 97705,
+      "磶": 97706,
+      "瘚": 97707,
+      "翏": 97708,
+      "珒": 97709,
+      "˟": 97710,
+      "👩🏻‍❤️‍💋‍👩🏽": 97711,
+      "郜": 97712,
+      "圚": 97713,
+      "迴": 97714,
+      "鍝": 97715,
+      "␇": 97716,
+      "閤": 97717,
+      "軥": 97718,
+      "毮": 97719,
+      "鵺": 97720,
+      "銡": 97721,
+      "變": 97722,
+      "硭": 97723,
+      "🙎🏻‍♀️": 97724,
+      "🏁": 97725,
+      "鞙": 97726,
+      "崬": 97727,
+      "嶲": 97728,
+      "悧": 97729,
+      "忴": 97730,
+      "簻": 97731,
+      "羱": 97732,
+      "瀰": 97733,
+      "蘉": 97734,
+      "⓲": 97735,
+      "⓳": 97736,
+      "欯": 97737,
+      "瑽": 97738,
+      "盤": 97739,
+      "奐": 97740,
+      "辬": 97741,
+      "孾": 97742,
+      "‸": 97743,
+      "卙": 97744,
+      "區": 97745,
+      "ː": 97746,
+      "ℏ": 97747,
+      "阢": 97748,
+      "碮": 97749,
+      "💆🏽": 97750,
+      "噀": 97751,
+      "⊻": 97752,
+      "ʥ": 97753,
+      "艽": 97754,
+      "粝": 97755,
+      "̗": 97756,
+      "嫥": 97757,
+      "攞": 97758,
+      "庻": 97759,
+      "猋": 97760,
+      "☓": 97761,
+      "鼁": 97762,
+      "猀": 97763,
+      "⛶": 97764,
+      "🇵🇬": 97765,
+      "斄": 97766,
+      "▰": 97767,
+      "燤": 97768,
+      ";": 97769,
+      "饗": 97770,
+      "␒": 97771,
+      "湎": 97772,
+      "瞦": 97773,
+      "醟": 97774,
+      "駆": 97775,
+      "蓧": 97776,
+      "🧝🏿‍♀️": 97777,
+      "呴": 97778,
+      "鲩": 97779,
+      "譔": 97780,
+      "獖": 97781,
+      "癇": 97782,
+      "媈": 97783,
+      "👈🏽": 97784,
+      "窾": 97785,
+      "Ｓ": 97786,
+      "🈸": 97787,
+      "⎁": 97788,
+      "ľ": 97789,
+      "₞": 97790,
+      "陝": 97791,
+      "🧗🏽": 97792,
+      "蝱": 97793,
+      "肟": 97794,
+      "ↈ": 97795,
+      "柩": 97796,
+      "薔": 97797,
+      "̾": 97798,
+      "嬯": 97799,
+      "鬻": 97800,
+      "🦅": 97801,
+      "熪": 97802,
+      "廢": 97803,
+      "蛞": 97804,
+      "鏛": 97805,
+      "鑥": 97806,
+      "鞡": 97807,
+      "🇬🇲": 97808,
+      "詉": 97809,
+      "鋍": 97810,
+      "郌": 97811,
+      "藈": 97812,
+      "吚": 97813,
+      "旙": 97814,
+      "跁": 97815,
+      "Ĩ": 97816,
+      "🧑🏾‍🎓": 97817,
+      "Ň": 97818,
+      "₻": 97819,
+      "埛": 97820,
+      "貣": 97821,
+      "仡": 97822,
+      "👼🏼": 97823,
+      "⏷": 97824,
+      "孹": 97825,
+      "籲": 97826,
+      "朻": 97827,
+      "鼑": 97828,
+      "詃": 97829,
+      "踯": 97830,
+      "闥": 97831,
+      "葔": 97832,
+      "🎷": 97833,
+      "喯": 97834,
+      "👽": 97835,
+      "🙋🏽‍♀️": 97836,
+      "坄": 97837,
+      "萹": 97838,
+      "嬹": 97839,
+      "錂": 97840,
+      "🕴🏻": 97841,
+      "逺": 97842,
+      "囕": 97843,
+      "涃": 97844,
+      "彗": 97845,
+      "忝": 97846,
+      "鶰": 97847,
+      "醻": 97848,
+      "DŽ": 97849,
+      "屐": 97850,
+      "蔭": 97851,
+      "煙": 97852,
+      "衒": 97853,
+      "垿": 97854,
+      "▫": 97855,
+      "👨🏻‍❤️‍💋‍👨🏿": 97856,
+      "铗": 97857,
+      "👧": 97858,
+      "絭": 97859,
+      "醤": 97860,
+      "鋅": 97861,
+      "矻": 97862,
+      "埂": 97863,
+      "👰🏾‍♀️": 97864,
+      "脷": 97865,
+      "咍": 97866,
+      "⍵": 97867,
+      "辌": 97868,
+      "◗": 97869,
+      "赇": 97870,
+      "姳": 97871,
+      "踬": 97872,
+      "🚣🏼": 97873,
+      "礧": 97874,
+      "礫": 97875,
+      "蜈": 97876,
+      "秂": 97877,
+      "朤": 97878,
+      "駨": 97879,
+      "炰": 97880,
+      "霚": 97881,
+      "🎡": 97882,
+      "旕": 97883,
+      "紏": 97884,
+      "賱": 97885,
+      "婈": 97886,
+      "浹": 97887,
+      "ǽ": 97888,
+      "筇": 97889,
+      "🪧": 97890,
+      "醣": 97891,
+      "綢": 97892,
+      "砧": 97893,
+      "妛": 97894,
+      "囇": 97895,
+      "蜍": 97896,
+      "漵": 97897,
+      "👨🏼‍❤️‍👨🏼": 97898,
+      "鯀": 97899,
+      "偀": 97900,
+      "┎": 97901,
+      "睟": 97902,
+      "♡": 97903,
+      "胄": 97904,
+      "暃": 97905,
+      "籩": 97906,
+      "➗": 97907,
+      "挙": 97908,
+      "甭": 97909,
+      "黠": 97910,
+      "陰": 97911,
+      "覌": 97912,
+      "澿": 97913,
+      "朓": 97914,
+      "觹": 97915,
+      "褍": 97916,
+      "赉": 97917,
+      "🧍🏽‍♂️": 97918,
+      "槡": 97919,
+      "簀": 97920,
+      "冮": 97921,
+      "✍": 97922,
+      "崊": 97923,
+      "🤳🏿": 97924,
+      "瀋": 97925,
+      "Ʌ": 97926,
+      "⎗": 97927,
+      "👩🏼‍⚖️": 97928,
+      "☝🏾": 97929,
+      "鐻": 97930,
+      "龜": 97931,
+      "玠": 97932,
+      "👩🏿‍❤️‍💋‍👨🏾": 97933,
+      "👷🏼": 97934,
+      "悻": 97935,
+      "頬": 97936,
+      "🧑🏻‍❤️‍💋‍🧑🏽": 97937,
+      "蝕": 97938,
+      "覺": 97939,
+      "🤾🏼‍♂️": 97940,
+      "檍": 97941,
+      "鍗": 97942,
+      "漜": 97943,
+      "姠": 97944,
+      "衐": 97945,
+      "韛": 97946,
+      "炂": 97947,
+      "🏧": 97948,
+      "👩🏼‍❤️‍💋‍👨🏽": 97949,
+      "💦": 97950,
+      "⇔": 97951,
+      "蟘": 97952,
+      "驲": 97953,
+      "籹": 97954,
+      "🙇🏻‍♂️": 97955,
+      "忎": 97956,
+      "懠": 97957,
+      "⌥": 97958,
+      "💗": 97959,
+      "饻": 97960,
+      "齶": 97961,
+      "☀": 97962,
+      "媮": 97963,
+      "崰": 97964,
+      "膤": 97965,
+      "躈": 97966,
+      "捋": 97967,
+      "鮽": 97968,
+      "嗊": 97969,
+      "逓": 97970,
+      "慔": 97971,
+      "哷": 97972,
+      "鏺": 97973,
+      "鳟": 97974,
+      "節": 97975,
+      "⇗": 97976,
+      "🧑🏿‍🎨": 97977,
+      "麕": 97978,
+      "葐": 97979,
+      "樫": 97980,
+      "⁎": 97981,
+      "👮🏿‍♀️": 97982,
+      "ͷ": 97983,
+      "跀": 97984,
+      "簹": 97985,
+      "滹": 97986,
+      "🏘": 97987,
+      "輲": 97988,
+      "僺": 97989,
+      "媨": 97990,
+      "賒": 97991,
+      "鬆": 97992,
+      "🇰🇬": 97993,
+      "敿": 97994,
+      "挒": 97995,
+      "膐": 97996,
+      "⚝": 97997,
+      "🩸": 97998,
+      "蕋": 97999,
+      "篕": 98000,
+      "疪": 98001,
+      "舝": 98002,
+      "ⅰ": 98003,
+      "鬱": 98004,
+      "刐": 98005,
+      "潰": 98006,
+      "懄": 98007,
+      "甉": 98008,
+      "蠲": 98009,
+      "⑙": 98010,
+      "檒": 98011,
+      "🛀🏻": 98012,
+      "饌": 98013,
+      "🇭🇹": 98014,
+      "瘴": 98015,
+      "紉": 98016,
+      "陦": 98017,
+      "敚": 98018,
+      "鏀": 98019,
+      "銂": 98020,
+      "昫": 98021,
+      "钤": 98022,
+      "骛": 98023,
+      "莰": 98024,
+      "紎": 98025,
+      "驖": 98026,
+      "朐": 98027,
+      "腁": 98028,
+      "玿": 98029,
+      "刉": 98030,
+      "鬖": 98031,
+      "蒻": 98032,
+      "気": 98033,
+      "瑳": 98034,
+      "蠱": 98035,
+      "🐻": 98036,
+      "袇": 98037,
+      "͛": 98038,
+      "🧑🏿‍❤️‍🧑🏾": 98039,
+      "幆": 98040,
+      "跧": 98041,
+      "楪": 98042,
+      "仮": 98043,
+      "萊": 98044,
+      "逰": 98045,
+      "殼": 98046,
+      "Ɖ": 98047,
+      "哶": 98048,
+      "⊡": 98049,
+      "🍾": 98050,
+      "飅": 98051,
+      "🧌": 98052,
+      "🏍": 98053,
+      "🇻🇳": 98054,
+      "毦": 98055,
+      "👨🏿‍✈️": 98056,
+      "瀻": 98057,
+      "荂": 98058,
+      "🫶🏿": 98059,
+      "蟥": 98060,
+      "ʡ": 98061,
+      "瑸": 98062,
+      "徏": 98063,
+      "🇰🇵": 98064,
+      "嫓": 98065,
+      "淞": 98066,
+      "棛": 98067,
+      "🌍": 98068,
+      "陱": 98069,
+      "皾": 98070,
+      "跬": 98071,
+      "⛥": 98072,
+      "鐽": 98073,
+      "🧵": 98074,
+      "薁": 98075,
+      "鈌": 98076,
+      "犻": 98077,
+      "餪": 98078,
+      "橶": 98079,
+      "⛳": 98080,
+      "🧉": 98081,
+      "聬": 98082,
+      "穪": 98083,
+      "靝": 98084,
+      "濡": 98085,
+      "狉": 98086,
+      "禖": 98087,
+      "⋥": 98088,
+      "祅": 98089,
+      "挻": 98090,
+      "搾": 98091,
+      "瑔": 98092,
+      "纥": 98093,
+      "瀂": 98094,
+      "瀍": 98095,
+      "嚧": 98096,
+      "幁": 98097,
+      "鴵": 98098,
+      "🛼": 98099,
+      "魰": 98100,
+      "🥽": 98101,
+      "🛃": 98102,
+      "👨🏾‍❤️‍💋‍👨🏽": 98103,
+      "芨": 98104,
+      "滘": 98105,
+      "棶": 98106,
+      "牔": 98107,
+      "勷": 98108,
+      "鈂": 98109,
+      "駞": 98110,
+      "⛹️‍♀️": 98111,
+      "緎": 98112,
+      "👞": 98113,
+      "蟗": 98114,
+      "⛊": 98115,
+      "裝": 98116,
+      "嶧": 98117,
+      "粯": 98118,
+      "🍴": 98119,
+      "ₗ": 98120,
+      "勞": 98121,
+      "噕": 98122,
+      "鮡": 98123,
+      "眡": 98124,
+      "愝": 98125,
+      "愆": 98126,
+      "檵": 98127,
+      "竜": 98128,
+      "👭🏼": 98129,
+      "歆": 98130,
+      "💇🏼‍♂️": 98131,
+      "衤": 98132,
+      "橿": 98133,
+      "⍇": 98134,
+      "剣": 98135,
+      "慽": 98136,
+      "陃": 98137,
+      "🕵🏿": 98138,
+      "🕠": 98139,
+      "謨": 98140,
+      "🔫": 98141,
+      "紋": 98142,
+      "踫": 98143,
+      "揋": 98144,
+      "𓀉": 98145,
+      "🆘": 98146,
+      "萚": 98147,
+      "🏃🏻‍♂️": 98148,
+      "羻": 98149,
+      "癍": 98150,
+      "̒": 98151,
+      "謮": 98152,
+      "埾": 98153,
+      "🏳️": 98154,
+      "Ŝ": 98155,
+      "謁": 98156,
+      "眲": 98157,
+      "喡": 98158,
+      "👃": 98159,
+      "塅": 98160,
+      "岞": 98161,
+      "鸤": 98162,
+      "軙": 98163,
+      "鹴": 98164,
+      "峇": 98165,
+      "🚫": 98166,
+      "蕚": 98167,
+      "壀": 98168,
+      "笎": 98169,
+      "鞳": 98170,
+      "膥": 98171,
+      "杪": 98172,
+      "🥤": 98173,
+      "嚔": 98174,
+      "鞀": 98175,
+      "👰🏻": 98176,
+      "掚": 98177,
+      "柨": 98178,
+      "氃": 98179,
+      "Π": 98180,
+      "檠": 98181,
+      "軁": 98182,
+      "吪": 98183,
+      "鹠": 98184,
+      "湨": 98185,
+      "欭": 98186,
+      "仉": 98187,
+      "秏": 98188,
+      "鑦": 98189,
+      "⍊": 98190,
+      "鎃": 98191,
+      "🌻": 98192,
+      "喴": 98193,
+      "瓂": 98194,
+      "僆": 98195,
+      "瘡": 98196,
+      "孶": 98197,
+      "铆": 98198,
+      "菅": 98199,
+      "洇": 98200,
+      "🤸": 98201,
+      "釗": 98202,
+      "簬": 98203,
+      "盓": 98204,
+      "潋": 98205,
+      "🧑🏻‍🔧": 98206,
+      "塾": 98207,
+      "鼿": 98208,
+      "↦": 98209,
+      "峼": 98210,
+      "饳": 98211,
+      "綃": 98212,
+      "醑": 98213,
+      "🐎": 98214,
+      "屧": 98215,
+      "箞": 98216,
+      "⌴": 98217,
+      "痄": 98218,
+      "謄": 98219,
+      "腜": 98220,
+      "ȋ": 98221,
+      "桰": 98222,
+      "彯": 98223,
+      "蔇": 98224,
+      "🧝🏻‍♀️": 98225,
+      "賸": 98226,
+      "瑧": 98227,
+      "褂": 98228,
+      "咹": 98229,
+      "🪚": 98230,
+      "鮱": 98231,
+      "逳": 98232,
+      "沯": 98233,
+      "縥": 98234,
+      "搐": 98235,
+      "␢": 98236,
+      "坿": 98237,
+      "蚃": 98238,
+      "姕": 98239,
+      "佉": 98240,
+      "縐": 98241,
+      "ʝ": 98242,
+      "頍": 98243,
+      "瞈": 98244,
+      "堻": 98245,
+      "痨": 98246,
+      "煡": 98247,
+      "⅔": 98248,
+      "涳": 98249,
+      "嶟": 98250,
+      "莢": 98251,
+      "洏": 98252,
+      "Ξ": 98253,
+      "稘": 98254,
+      "Ⓩ": 98255,
+      "櫂": 98256,
+      "🌇": 98257,
+      "錖": 98258,
+      "穠": 98259,
+      "🦍": 98260,
+      "嘓": 98261,
+      "鴩": 98262,
+      "峈": 98263,
+      "擶": 98264,
+      "铙": 98265,
+      "鉲": 98266,
+      "蓁": 98267,
+      "潒": 98268,
+      "絃": 98269,
+      "覠": 98270,
+      "煶": 98271,
+      "👨🏼‍⚕️": 98272,
+      "赱": 98273,
+      "👭🏾": 98274,
+      "瀌": 98275,
+      "ɨ": 98276,
+      "汑": 98277,
+      "罏": 98278,
+      "🆖": 98279,
+      "Ź": 98280,
+      "蔰": 98281,
+      "挮": 98282,
+      "镞": 98283,
+      "🚽": 98284,
+      "甶": 98285,
+      "峜": 98286,
+      "🅰️": 98287,
+      "醦": 98288,
+      "臔": 98289,
+      "拴": 98290,
+      "沞": 98291,
+      "鄗": 98292,
+      "╫": 98293,
+      "姈": 98294,
+      "🪀": 98295,
+      "憕": 98296,
+      "鑼": 98297,
+      "籋": 98298,
+      "♟": 98299,
+      "♝": 98300,
+      "🇮🇪": 98301,
+      "鄖": 98302,
+      "箠": 98303,
+      "🚴🏾‍♀️": 98304,
+      "煄": 98305,
+      "藃": 98306,
+      "箈": 98307,
+      "讧": 98308,
+      "駼": 98309,
+      "怍": 98310,
+      "嚬": 98311,
+      "癞": 98312,
+      "螯": 98313,
+      "🏌🏽‍♂️": 98314,
+      "坩": 98315,
+      "獳": 98316,
+      "枙": 98317,
+      "櫿": 98318,
+      "帊": 98319,
+      "🙎🏽‍♂️": 98320,
+      "🔲": 98321,
+      "ʨ": 98322,
+      "枱": 98323,
+      "NJ": 98324,
+      "窕": 98325,
+      "藍": 98326,
+      "哠": 98327,
+      "∎": 98328,
+      "┓": 98329,
+      "寪": 98330,
+      "攩": 98331,
+      "牫": 98332,
+      "🤙🏾": 98333,
+      "嘒": 98334,
+      "覭": 98335,
+      "吒": 98336,
+      "Ϛ": 98337,
+      "瓌": 98338,
+      "筫": 98339,
+      "彍": 98340,
+      "釼": 98341,
+      "藺": 98342,
+      "邾": 98343,
+      "蠬": 98344,
+      "🤾🏿": 98345,
+      "繛": 98346,
+      "喈": 98347,
+      "猘": 98348,
+      "蘞": 98349,
+      "潄": 98350,
+      "媙": 98351,
+      "蘻": 98352,
+      "恟": 98353,
+      "媧": 98354,
+      "蝃": 98355,
+      "🛵": 98356,
+      "阇": 98357,
+      "˷": 98358,
+      "寬": 98359,
+      "🇲🇽": 98360,
+      "哫": 98361,
+      "🔮": 98362,
+      "◲": 98363,
+      "🆚": 98364,
+      "兿": 98365,
+      "癘": 98366,
+      "蠊": 98367,
+      "荮": 98368,
+      "蛀": 98369,
+      "媵": 98370,
+      "佽": 98371,
+      "侭": 98372,
+      "禴": 98373,
+      "夰": 98374,
+      "🗂": 98375,
+      "👨🏼‍❤️‍👨🏻": 98376,
+      "瑅": 98377,
+      "肫": 98378,
+      "蓚": 98379,
+      "麘": 98380,
+      "吆": 98381,
+      "囼": 98382,
+      "⊒": 98383,
+      "镂": 98384,
+      "庰": 98385,
+      "諺": 98386,
+      "狓": 98387,
+      "👯‍♀️": 98388,
+      "🆗": 98389,
+      "斶": 98390,
+      "規": 98391,
+      "🙆‍♀️": 98392,
+      "鰓": 98393,
+      "僱": 98394,
+      "繪": 98395,
+      "攂": 98396,
+      "榞": 98397,
+      "磫": 98398,
+      "鉀": 98399,
+      "蛥": 98400,
+      "碔": 98401,
+      "🧊": 98402,
+      "瑬": 98403,
+      "尰": 98404,
+      "⛼": 98405,
+      "羑": 98406,
+      "錅": 98407,
+      "肬": 98408,
+      "靭": 98409,
+      "💁‍♀️": 98410,
+      "₵": 98411,
+      "霻": 98412,
+      "🏂": 98413,
+      "爻": 98414,
+      "➝": 98415,
+      "愙": 98416,
+      "杤": 98417,
+      "鬭": 98418,
+      "剬": 98419,
+      "🧑🏼‍❤️‍🧑🏿": 98420,
+      "犆": 98421,
+      "帒": 98422,
+      "紸": 98423,
+      "魿": 98424,
+      "↥": 98425,
+      "蠖": 98426,
+      "⏉": 98427,
+      "妵": 98428,
+      "昒": 98429,
+      "綍": 98430,
+      "🧑🏻‍🔬": 98431,
+      "掟": 98432,
+      "譣": 98433,
+      "璎": 98434,
+      "酻": 98435,
+      "銔": 98436,
+      "ʪ": 98437,
+      "翴": 98438,
+      "⃢": 98439,
+      "欓": 98440,
+      "傹": 98441,
+      "🚣🏿": 98442,
+      "魐": 98443,
+      "郹": 98444,
+      "啳": 98445,
+      "妽": 98446,
+      "🦸🏻‍♂️": 98447,
+      "Dž": 98448,
+      "麡": 98449,
+      "婫": 98450,
+      "伩": 98451,
+      "蝝": 98452,
+      "🏇🏻": 98453,
+      "橚": 98454,
+      "虶": 98455,
+      "埀": 98456,
+      "螌": 98457,
+      "🤚🏻": 98458,
+      "壼": 98459,
+      "鮫": 98460,
+      "骝": 98461,
+      "婽": 98462,
+      "愇": 98463,
+      "麎": 98464,
+      "蓡": 98465,
+      "蒺": 98466,
+      "踶": 98467,
+      "🤲🏾": 98468,
+      "袐": 98469,
+      "惃": 98470,
+      "屍": 98471,
+      "瀪": 98472,
+      "矨": 98473,
+      "殘": 98474,
+      "馲": 98475,
+      "⋧": 98476,
+      "銣": 98477,
+      "岍": 98478,
+      "📭": 98479,
+      "恞": 98480,
+      "📣": 98481,
+      "唌": 98482,
+      "緾": 98483,
+      "⊮": 98484,
+      "𓀊": 98485,
+      "鹆": 98486,
+      "👌🏽": 98487,
+      "蛹": 98488,
+      "📼": 98489,
+      "琫": 98490,
+      "👕": 98491,
+      "澨": 98492,
+      "👩🏻‍❤️‍👨🏻": 98493,
+      "嵬": 98494,
+      "髾": 98495,
+      "祬": 98496,
+      "郬": 98497,
+      "帶": 98498,
+      "📑": 98499,
+      "凮": 98500,
+      "惷": 98501,
+      "贇": 98502,
+      "昉": 98503,
+      "☝🏿": 98504,
+      "ȹ": 98505,
+      "墜": 98506,
+      "汫": 98507,
+      "🏛": 98508,
+      "僽": 98509,
+      "懁": 98510,
+      "緧": 98511,
+      "跕": 98512,
+      "ʷ": 98513,
+      "礰": 98514,
+      "💆🏿‍♂️": 98515,
+      "椽": 98516,
+      "轫": 98517,
+      "殾": 98518,
+      "壣": 98519,
+      "竷": 98520,
+      "鎜": 98521,
+      "👨‍🦳": 98522,
+      "諏": 98523,
+      "蹜": 98524,
+      "曟": 98525,
+      "🛟": 98526,
+      "恝": 98527,
+      "摦": 98528,
+      "諷": 98529,
+      "貜": 98530,
+      "鏫": 98531,
+      "糇": 98532,
+      "軨": 98533,
+      "馄": 98534,
+      "陧": 98535,
+      "諴": 98536,
+      "鼋": 98537,
+      "≙": 98538,
+      "爫": 98539,
+      "舯": 98540,
+      "亓": 98541,
+      "👨🏻‍⚖️": 98542,
+      "👩‍👩‍👧": 98543,
+      "躍": 98544,
+      "鼈": 98545,
+      "屳": 98546,
+      "墑": 98547,
+      "ŝ": 98548,
+      "笆": 98549,
+      "雥": 98550,
+      "檓": 98551,
+      "鯆": 98552,
+      "鷨": 98553,
+      "⑽": 98554,
+      "莖": 98555,
+      "🧑🏽‍🦯": 98556,
+      "穖": 98557,
+      "撐": 98558,
+      "郠": 98559,
+      "👨🏼‍🍳": 98560,
+      "遘": 98561,
+      "鎒": 98562,
+      "👩🏾‍❤️‍💋‍👩🏼": 98563,
+      "炀": 98564,
+      "栃": 98565,
+      "Ȋ": 98566,
+      "糶": 98567,
+      "騺": 98568,
+      "疸": 98569,
+      "梸": 98570,
+      "舚": 98571,
+      "✆": 98572,
+      "薺": 98573,
+      "浉": 98574,
+      "昢": 98575,
+      "阹": 98576,
+      "⋈": 98577,
+      "瑤": 98578,
+      "亜": 98579,
+      "懽": 98580,
+      "👉🏼": 98581,
+      "亖": 98582,
+      "蛸": 98583,
+      "鳺": 98584,
+      "丷": 98585,
+      "🧏🏽": 98586,
+      "輗": 98587,
+      "瑮": 98588,
+      "洀": 98589,
+      "觜": 98590,
+      "徠": 98591,
+      "偤": 98592,
+      "窵": 98593,
+      "诓": 98594,
+      "寏": 98595,
+      "🧑‍🎤": 98596,
+      "煭": 98597,
+      "銄": 98598,
+      "娻": 98599,
+      "牨": 98600,
+      "鼛": 98601,
+      "糭": 98602,
+      "墈": 98603,
+      "🇸🇧": 98604,
+      "⛓": 98605,
+      "☈": 98606,
+      "呣": 98607,
+      "🧍🏻‍♂️": 98608,
+      "泑": 98609,
+      "ʣ": 98610,
+      "蚢": 98611,
+      "◃": 98612,
+      "虡": 98613,
+      "捴": 98614,
+      "卬": 98615,
+      "硽": 98616,
+      "飼": 98617,
+      "慖": 98618,
+      "⊪": 98619,
+      "呝": 98620,
+      "挲": 98621,
+      "茀": 98622,
+      "鼡": 98623,
+      "覙": 98624,
+      "艼": 98625,
+      "颯": 98626,
+      "帑": 98627,
+      "萑": 98628,
+      "♦︎": 98629,
+      "歖": 98630,
+      "📢": 98631,
+      "鱞": 98632,
+      "韋": 98633,
+      "🧈": 98634,
+      "簈": 98635,
+      "釚": 98636,
+      "砫": 98637,
+      "🧑🏿‍🍳": 98638,
+      "乛": 98639,
+      "⏢": 98640,
+      "潳": 98641,
+      "禪": 98642,
+      "鈆": 98643,
+      "譑": 98644,
+      "🙍🏻‍♀️": 98645,
+      "撥": 98646,
+      "呶": 98647,
+      "穏": 98648,
+      "豔": 98649,
+      "刎": 98650,
+      "斾": 98651,
+      "纓": 98652,
+      "羐": 98653,
+      "閊": 98654,
+      "闶": 98655,
+      "艿": 98656,
+      "栎": 98657,
+      "蜔": 98658,
+      "猵": 98659,
+      "彠": 98660,
+      "≏": 98661,
+      "迒": 98662,
+      "媖": 98663,
+      "蒀": 98664,
+      "筯": 98665,
+      "爂": 98666,
+      "葤": 98667,
+      "👩🏾‍🦼": 98668,
+      "閺": 98669,
+      "攴": 98670,
+      "櫪": 98671,
+      "鮛": 98672,
+      "秴": 98673,
+      "紨": 98674,
+      "葒": 98675,
+      "旛": 98676,
+      "Ǖ": 98677,
+      "寗": 98678,
+      "淟": 98679,
+      "窆": 98680,
+      "🙍🏾": 98681,
+      "栨": 98682,
+      "籗": 98683,
+      "諪": 98684,
+      "蝪": 98685,
+      "🤾‍♀️": 98686,
+      "椟": 98687,
+      "肞": 98688,
+      "↩": 98689,
+      "🙇": 98690,
+      "🫣": 98691,
+      "奘": 98692,
+      "逖": 98693,
+      "蹅": 98694,
+      "脌": 98695,
+      "🧑🏾‍🦼": 98696,
+      "🧘🏼‍♂️": 98697,
+      "叝": 98698,
+      "飮": 98699,
+      "瓥": 98700,
+      "ɞ": 98701,
+      "̡": 98702,
+      "摯": 98703,
+      "譡": 98704,
+      "👩🏻‍🎨": 98705,
+      "👩🏾‍❤️‍💋‍👨🏽": 98706,
+      "⛇": 98707,
+      "檔": 98708,
+      "矑": 98709,
+      "濪": 98710,
+      "🆔": 98711,
+      "⇍": 98712,
+      "🔧": 98713,
+      "🌤": 98714,
+      "溻": 98715,
+      "🪻": 98716,
+      "ǒ": 98717,
+      "鸓": 98718,
+      "⊸": 98719,
+      "∘": 98720,
+      "蚤": 98721,
+      "舃": 98722,
+      "觟": 98723,
+      "👊🏼": 98724,
+      "旊": 98725,
+      "姉": 98726,
+      "櫑": 98727,
+      "涴": 98728,
+      "杽": 98729,
+      "琋": 98730,
+      "崷": 98731,
+      "Ʋ": 98732,
+      "🇬🇼": 98733,
+      "瞸": 98734,
+      "蔅": 98735,
+      "韯": 98736,
+      "攬": 98737,
+      "飭": 98738,
+      "麯": 98739,
+      "焔": 98740,
+      "釕": 98741,
+      "岾": 98742,
+      "↟": 98743,
+      "嶈": 98744,
+      "♬": 98745,
+      "溫": 98746,
+      "钘": 98747,
+      "̘": 98748,
+      "👦🏻": 98749,
+      "鰾": 98750,
+      "踑": 98751,
+      "ⁱ": 98752,
+      "洵": 98753,
+      "鶥": 98754,
+      "🤘🏼": 98755,
+      "繓": 98756,
+      "駜": 98757,
+      "揑": 98758,
+      "歯": 98759,
+      "💅🏽": 98760,
+      "鐴": 98761,
+      "叡": 98762,
+      "颿": 98763,
+      "😈": 98764,
+      "🧖🏾‍♀️": 98765,
+      "▔": 98766,
+      "幘": 98767,
+      "覗": 98768,
+      "囃": 98769,
+      "觪": 98770,
+      "洢": 98771,
+      "洊": 98772,
+      "⑈": 98773,
+      "崴": 98774,
+      "驍": 98775,
+      "鳸": 98776,
+      "揠": 98777,
+      "☝": 98778,
+      "🧙🏿‍♀️": 98779,
+      "⊖": 98780,
+      "⃡": 98781,
+      "⛍": 98782,
+      "嵦": 98783,
+      "峬": 98784,
+      "還": 98785,
+      "蜨": 98786,
+      "🧑🏿‍🤝‍🧑🏿": 98787,
+      "蟂": 98788,
+      "硧": 98789,
+      "鄫": 98790,
+      "颒": 98791,
+      "繒": 98792,
+      "鯸": 98793,
+      "🥦": 98794,
+      "肜": 98795,
+      "┸": 98796,
+      "鼃": 98797,
+      "🛋": 98798,
+      "謐": 98799,
+      "🧑🏼‍🏭": 98800,
+      "頉": 98801,
+      "戰": 98802,
+      "匧": 98803,
+      "陴": 98804,
+      "🧚🏼‍♂️": 98805,
+      "幭": 98806,
+      "睆": 98807,
+      "🧝🏼": 98808,
+      "😝": 98809,
+      "␲": 98810,
+      "汍": 98811,
+      "丵": 98812,
+      "鑭": 98813,
+      "籰": 98814,
+      "顓": 98815,
+      "權": 98816,
+      "蔕": 98817,
+      "⋸": 98818,
+      "腩": 98819,
+      "👩🏾‍❤️‍💋‍👩🏽": 98820,
+      "紜": 98821,
+      "锉": 98822,
+      "貸": 98823,
+      "⃱": 98824,
+      "🧔🏾‍♂️": 98825,
+      "燴": 98826,
+      "抧": 98827,
+      "偘": 98828,
+      "傂": 98829,
+      "👳": 98830,
+      "瘕": 98831,
+      "⋕": 98832,
+      "睲": 98833,
+      "兓": 98834,
+      "鑉": 98835,
+      "🧎🏻‍♂️": 98836,
+      "ⅶ": 98837,
+      "僁": 98838,
+      "⃼": 98839,
+      "挄": 98840,
+      "绡": 98841,
+      "🟥": 98842,
+      "摬": 98843,
+      "赻": 98844,
+      "🫰🏻": 98845,
+      "鸖": 98846,
+      "🙎🏾‍♂️": 98847,
+      "宓": 98848,
+      "夳": 98849,
+      "鋋": 98850,
+      "🏵": 98851,
+      "滼": 98852,
+      "砭": 98853,
+      "糺": 98854,
+      "荤": 98855,
+      "锑": 98856,
+      "豧": 98857,
+      "☼": 98858,
+      "矘": 98859,
+      "⋆": 98860,
+      "🧛🏻": 98861,
+      "阛": 98862,
+      "⋀": 98863,
+      "🫔": 98864,
+      "鰒": 98865,
+      "🛤": 98866,
+      "絾": 98867,
+      "鋌": 98868,
+      "鏇": 98869,
+      "鏪": 98870,
+      "♌": 98871,
+      "嘽": 98872,
+      "幃": 98873,
+      "瓉": 98874,
+      "樑": 98875,
+      "蜥": 98876,
+      "觎": 98877,
+      "蛵": 98878,
+      "姲": 98879,
+      "衧": 98880,
+      "聾": 98881,
+      "澸": 98882,
+      "逧": 98883,
+      "狀": 98884,
+      "👌🏼": 98885,
+      "📴": 98886,
+      "憗": 98887,
+      "殓": 98888,
+      "漹": 98889,
+      "🚟": 98890,
+      "瀁": 98891,
+      "簱": 98892,
+      "🏄🏻‍♂️": 98893,
+      "惞": 98894,
+      "蛐": 98895,
+      "凂": 98896,
+      "萴": 98897,
+      "煼": 98898,
+      "駋": 98899,
+      "顛": 98900,
+      "欴": 98901,
+      "猦": 98902,
+      "騥": 98903,
+      "犘": 98904,
+      "鮔": 98905,
+      "爑": 98906,
+      "帟": 98907,
+      "齚": 98908,
+      "鄑": 98909,
+      "奼": 98910,
+      "🧚🏻‍♀️": 98911,
+      "矖": 98912,
+      "挷": 98913,
+      "訉": 98914,
+      "鰗": 98915,
+      "茖": 98916,
+      "蠣": 98917,
+      "餰": 98918,
+      "榠": 98919,
+      "栿": 98920,
+      "🧙‍♂️": 98921,
+      "罹": 98922,
+      "鹪": 98923,
+      "椦": 98924,
+      "唋": 98925,
+      "🌪": 98926,
+      "鯱": 98927,
+      "湑": 98928,
+      "骈": 98929,
+      "🌳": 98930,
+      "忭": 98931,
+      "赽": 98932,
+      "瓹": 98933,
+      "趞": 98934,
+      "≩": 98935,
+      "⍐": 98936,
+      "硇": 98937,
+      "霩": 98938,
+      "緘": 98939,
+      "鷷": 98940,
+      "輤": 98941,
+      "攋": 98942,
+      "楨": 98943,
+      "猄": 98944,
+      "░": 98945,
+      "槺": 98946,
+      "帠": 98947,
+      "姦": 98948,
+      "烺": 98949,
+      "睸": 98950,
+      "綸": 98951,
+      "塣": 98952,
+      "蹞": 98953,
+      "☧": 98954,
+      "嬍": 98955,
+      "鯧": 98956,
+      "噹": 98957,
+      "⚁": 98958,
+      "浤": 98959,
+      "藶": 98960,
+      "🌽": 98961,
+      "Ĝ": 98962,
+      "⁁": 98963,
+      "蔆": 98964,
+      "剜": 98965,
+      "nj": 98966,
+      "泤": 98967,
+      "🙍‍♂️": 98968,
+      "飝": 98969,
+      "軂": 98970,
+      "吠": 98971,
+      "爲": 98972,
+      "暚": 98973,
+      "簜": 98974,
+      "😇": 98975,
+      "嗔": 98976,
+      "ℙ": 98977,
+      "媂": 98978,
+      "絉": 98979,
+      "惏": 98980,
+      "檑": 98981,
+      "磤": 98982,
+      "🧖🏿": 98983,
+      "翂": 98984,
+      "芺": 98985,
+      "掕": 98986,
+      "鮘": 98987,
+      "Ņ": 98988,
+      "嗧": 98989,
+      "識": 98990,
+      "ʞ": 98991,
+      "俸": 98992,
+      "梠": 98993,
+      "魖": 98994,
+      "韤": 98995,
+      "爭": 98996,
+      "鵞": 98997,
+      "騀": 98998,
+      "閇": 98999
+    },
+    "merges": [
+      "。。。 。。。",
+      "`` `",
+      "` ``",
+      "▁ ▁",
+      "▁ t",
+      "i n",
+      "▁ a",
+      "e r",
+      "o n",
+      "h e",
+      "r e",
+      "a t",
+      "▁▁ ▁▁",
+      "▁▁▁ ▁",
+      "▁ ▁▁▁",
+      "o r",
+      "e n",
+      "▁ s",
+      "▁t he",
+      "▁th e",
+      "▁ the",
+      "e s",
+      "i s",
+      "▁ c",
+      "i t",
+      "a n",
+      "a l",
+      "▁ o",
+      "▁ w",
+      "e d",
+      "▁ f",
+      "▁ p",
+      "a r",
+      "in g",
+      "i ng",
+      "n d",
+      "io n",
+      "i on",
+      "i c",
+      "o u",
+      "▁ b",
+      "▁i n",
+      "▁ in",
+      "l e",
+      "▁o f",
+      "▁ of",
+      "▁ m",
+      "r o",
+      "a s",
+      "▁ d",
+      "▁a nd",
+      "▁an d",
+      "▁ and",
+      "▁t o",
+      "▁ to",
+      "en t",
+      "e nt",
+      "c t",
+      "e t",
+      "s t",
+      "o m",
+      "e l",
+      "▁ h",
+      "▁t h",
+      "▁ th",
+      "i l",
+      "▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁ ▁▁",
+      "▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁▁ ▁",
+      "▁ ▁▁▁▁▁▁▁",
+      "▁r e",
+      "▁ re",
+      "▁ n",
+      "a m",
+      "▁ l",
+      "i m",
+      "▁ e",
+      "i d",
+      "u t",
+      "o l",
+      "a c",
+      "u s",
+      "u r",
+      "▁ T",
+      "at ion",
+      "ati on",
+      "atio n",
+      "a tion",
+      "▁i s",
+      "▁ is",
+      "o t",
+      "v e",
+      "▁ S",
+      "▁ g",
+      "\\ \\",
+      "i g",
+      "▁ I",
+      "\\ '",
+      "▁ C",
+      "▁ A",
+      "▁f or",
+      "▁fo r",
+      "▁ for",
+      "a d",
+      "u l",
+      "o w",
+      "▁ |",
+      "▁ (",
+      "▁ y",
+      "; \\",
+      "it h",
+      "i th",
+      "ve r",
+      "v er",
+      "a y",
+      "▁o n",
+      "▁ on",
+      "▁s t",
+      "▁ st",
+      "l y",
+      "s e",
+      "i f",
+      "c e",
+      "o d",
+      "c h",
+      "t h",
+      "▁b e",
+      "▁ be",
+      "a g",
+      "te r",
+      "t er",
+      "u n",
+      "▁ M",
+      "at e",
+      "a te",
+      "es t",
+      "e st",
+      "i r",
+      "▁y ou",
+      "▁yo u",
+      "▁ you",
+      "▁w ith",
+      "▁wit h",
+      "▁wi th",
+      "▁ with",
+      "u m",
+      "▁ P",
+      "▁p ro",
+      "▁pr o",
+      "▁ pro",
+      "e m",
+      ") \\",
+      "a b",
+      "▁t hat",
+      "▁th at",
+      "▁ that",
+      "▁a n",
+      "▁ an",
+      "or t",
+      "o rt",
+      "r i",
+      "a p",
+      "q u",
+      "▁c on",
+      "▁co n",
+      "▁ con",
+      "o p",
+      "er s",
+      "e rs",
+      "is t",
+      "i st",
+      "▁a s",
+      "▁ as",
+      "ec t",
+      "e ct",
+      "▁ B",
+      "re s",
+      "r es",
+      "▁ v",
+      "o s",
+      "o c",
+      "▁a l",
+      "▁ al",
+      "ro m",
+      "r om",
+      "▁i t",
+      "▁ it",
+      "▁w h",
+      "▁ wh",
+      "h t",
+      "▁ D",
+      "▁c om",
+      "▁co m",
+      "▁ com",
+      "es s",
+      "e ss",
+      "me nt",
+      "men t",
+      "m ent",
+      "an d",
+      "a nd",
+      "u b",
+      "ai n",
+      "a in",
+      "er e",
+      "e re",
+      "il l",
+      "i ll",
+      "▁ r",
+      "am e",
+      "a me",
+      "▁ R",
+      "▁ H",
+      "▁e x",
+      "▁ ex",
+      "▁ F",
+      "▁ L",
+      "▁ W",
+      "▁ \"",
+      "u p",
+      "i v",
+      "it y",
+      "i ty",
+      "por t",
+      "po rt",
+      "p ort",
+      "e c",
+      "▁a re",
+      "▁ar e",
+      "▁ are",
+      "p l",
+      "▁ {",
+      "e w",
+      "e x",
+      "▁T he",
+      "▁Th e",
+      "▁ The",
+      "▁o r",
+      "▁ or",
+      "k e",
+      "▁a t",
+      "▁ at",
+      "u e",
+      "or e",
+      "o re",
+      "n t",
+      "o g",
+      "▁ N",
+      "an t",
+      "a nt",
+      "en d",
+      "e nd",
+      "ac t",
+      "a ct",
+      "iv e",
+      "i ve",
+      "▁b y",
+      "▁ by",
+      "e f",
+      "ar t",
+      "a rt",
+      "▁w as",
+      "▁wa s",
+      "▁ was",
+      "p t",
+      "▁f rom",
+      "▁fr om",
+      "▁fro m",
+      "▁ from",
+      ", \\",
+      "▁ G",
+      "u d",
+      "▁ E",
+      ": \\",
+      "▁u s",
+      "▁ us",
+      "ie s",
+      "i es",
+      "ag e",
+      "a ge",
+      "in e",
+      "i ne",
+      "▁▁ ▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁ ▁▁",
+      "▁▁▁ ▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁ ▁",
+      "▁ ▁▁▁▁▁▁▁▁▁▁▁",
+      "▁t his",
+      "▁th is",
+      "▁thi s",
+      "▁ this",
+      "as s",
+      "a ss",
+      "▁ $",
+      "ur e",
+      "u re",
+      "l d",
+      "al l",
+      "a ll",
+      "} \\",
+      "av e",
+      "a ve",
+      "▁h e",
+      "▁ he",
+      "); \\",
+      ") ;\\",
+      "ia l",
+      "i al",
+      "co m",
+      "c om",
+      "▁c h",
+      "▁ ch",
+      "▁ O",
+      "ig ht",
+      "igh t",
+      "i ght",
+      "c l",
+      "ct ion",
+      "c tion",
+      "n T",
+      "▁w e",
+      "▁ we",
+      "# #",
+      "at ed",
+      "ate d",
+      "a ted",
+      "i z",
+      "ac k",
+      "a ck",
+      "ul t",
+      "u lt",
+      "▁t r",
+      "▁ tr",
+      "g e",
+      "i p",
+      "ou r",
+      "o ur",
+      "▁d e",
+      "▁ de",
+      "in t",
+      "i nt",
+      "/ /",
+      "fo r",
+      "f or",
+      "el l",
+      "e ll",
+      "us t",
+      "u st",
+      "▁s h",
+      "▁ sh",
+      "▁c an",
+      "▁ca n",
+      "▁ can",
+      "os t",
+      "o st",
+      "▁n ot",
+      "▁no t",
+      "▁ not",
+      "ou t",
+      "o ut",
+      "▁\\ \\",
+      "▁ \\\\",
+      "pe r",
+      "p er",
+      "▁s e",
+      "▁ se",
+      "ar d",
+      "a rd",
+      "ar e",
+      "a re",
+      "▁h ave",
+      "▁ha ve",
+      "▁hav e",
+      "▁ have",
+      "▁i m",
+      "▁ im",
+      "as e",
+      "a se",
+      "ic al",
+      "ica l",
+      "i cal",
+      "▁c l",
+      "▁ cl",
+      "in d",
+      "i nd",
+      "t r",
+      "▁{ \\",
+      "▁ {\\",
+      "* *",
+      "ro u",
+      "r ou",
+      "p e",
+      "a v",
+      "ion s",
+      "io ns",
+      "i ons",
+      "▁l e",
+      "▁ le",
+      "▁p l",
+      "▁ pl",
+      "f f",
+      "co n",
+      "c on",
+      "ab le",
+      "abl e",
+      "a ble",
+      "at h",
+      "a th",
+      "i b",
+      "id e",
+      "i de",
+      "▁w ill",
+      "▁wi ll",
+      "▁ will",
+      "▁c omp",
+      "▁com p",
+      "▁co mp",
+      "▁ comp",
+      "il e",
+      "i le",
+      "ir e",
+      "i re",
+      "ur n",
+      "u rn",
+      "as t",
+      "a st",
+      "o k",
+      "▁re s",
+      "▁r es",
+      "▁ res",
+      "▁ J",
+      "or k",
+      "o rk",
+      "an s",
+      "a ns",
+      "▁ U",
+      "ic e",
+      "i ce",
+      "▁ *",
+      "▁y our",
+      "▁you r",
+      "▁yo ur",
+      "▁ your",
+      "▁ j",
+      "▁a d",
+      "▁ ad",
+      "▁s p",
+      "▁ sp",
+      "ou ld",
+      "oul d",
+      "ic h",
+      "i ch",
+      "▁ k",
+      "a k",
+      "ig n",
+      "i gn",
+      "im e",
+      "i me",
+      "d u",
+      "re t",
+      "r et",
+      "i a",
+      "▁a b",
+      "▁ ab",
+      "ar y",
+      "a ry",
+      "ai l",
+      "a il",
+      "▁ \\",
+      "▁T h",
+      "▁ Th",
+      "ac e",
+      "a ce",
+      "执 行",
+      "on g",
+      "o ng",
+      "e p",
+      "el f",
+      "e lf",
+      "en ts",
+      "ent s",
+      "on e",
+      "o ne",
+      "be r",
+      "b er",
+      "at ions",
+      "ation s",
+      "ati ons",
+      "atio ns",
+      "ea r",
+      "e ar",
+      "en s",
+      "e ns",
+      "pr o",
+      "p ro",
+      "▁a ll",
+      "▁al l",
+      "▁ all",
+      "▁I n",
+      "▁ In",
+      "en ce",
+      "enc e",
+      "e nce",
+      "▁a p",
+      "▁ ap",
+      "▁h as",
+      "▁ha s",
+      "▁ has",
+      "▁in t",
+      "▁i nt",
+      "▁ int",
+      "ex t",
+      "e xt",
+      "> \\",
+      "om e",
+      "o me",
+      "n I",
+      "or d",
+      "o rd",
+      "s o",
+      "an g",
+      "a ng",
+      "for m",
+      "fo rm",
+      "f orm",
+      "an ce",
+      "anc e",
+      "a nce",
+      "un d",
+      "u nd",
+      "li c",
+      "l ic",
+      "▁c ont",
+      "▁con t",
+      "▁co nt",
+      "▁ cont",
+      "at a",
+      "a ta",
+      "▁m e",
+      "▁ me",
+      "c c",
+      "▁| \\",
+      "▁ |\\",
+      "fe r",
+      "f er",
+      "he n",
+      "h en",
+      "( )",
+      "▁▁ ▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁ ▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁",
+      "▁▁▁ ▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁",
+      "▁ ▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "it e",
+      "i te",
+      "人 民",
+      "en c",
+      "e nc",
+      "re ss",
+      "res s",
+      "r ess",
+      "▁d o",
+      "▁ do",
+      "n A",
+      "is s",
+      "i ss",
+      "je ct",
+      "j ect",
+      "ou s",
+      "o us",
+      "er v",
+      "e rv",
+      "▁ K",
+      "th er",
+      "the r",
+      "t her",
+      "st r",
+      "s tr",
+      "ac h",
+      "a ch",
+      "l o",
+      "ge t",
+      "g et",
+      "r y",
+      "na me",
+      "nam e",
+      "n ame",
+      "t p",
+      "▁ V",
+      "▁n e",
+      "▁ ne",
+      "ve l",
+      "v el",
+      "or y",
+      "o ry",
+      "▁u n",
+      "▁ un",
+      "g h",
+      "▁b ut",
+      "▁bu t",
+      "▁ but",
+      "▁wh ich",
+      "▁ which",
+      "or s",
+      "o rs",
+      "l l",
+      "nT he",
+      "nTh e",
+      "n The",
+      "it ion",
+      "iti on",
+      "i tion",
+      "n S",
+      "▁p er",
+      "▁pe r",
+      "▁ per",
+      "▁▁ ▁▁▁▁",
+      "▁▁▁▁ ▁▁",
+      "▁▁▁ ▁▁▁",
+      "▁▁▁▁▁ ▁",
+      "▁ ▁▁▁▁▁",
+      "▁e n",
+      "▁ en",
+      "公 司",
+      "▁o ut",
+      "▁ou t",
+      "▁ out",
+      "f r",
+      "▁ [",
+      "ak e",
+      "a ke",
+      "▁\\ '",
+      "▁ \\'",
+      "▁u p",
+      "▁ up",
+      "▁q u",
+      "▁ qu",
+      "ou nt",
+      "oun t",
+      "o unt",
+      "▁n ew",
+      "▁ne w",
+      "▁ new",
+      "▁S t",
+      "▁ St",
+      "▁g o",
+      "▁ go",
+      "oo k",
+      "o ok",
+      "u g",
+      "re turn",
+      "ret urn",
+      "r eturn",
+      "os e",
+      "o se",
+      "▁a r",
+      "▁ ar",
+      "ub lic",
+      "u blic",
+      "im port",
+      "imp ort",
+      "f t",
+      "▁d is",
+      "▁di s",
+      "▁ dis",
+      "▁w ere",
+      "▁we re",
+      "▁wer e",
+      "▁ were",
+      "pl e",
+      "p le",
+      "y s",
+      "ve rs",
+      "ver s",
+      "v ers",
+      "< /",
+      "\" \\",
+      "▁s o",
+      "▁ so",
+      "od e",
+      "o de",
+      "il d",
+      "i ld",
+      "al ly",
+      "all y",
+      "yp e",
+      "y pe",
+      "p h",
+      "t e",
+      "▁s tr",
+      "▁st r",
+      "▁ str",
+      "re e",
+      "r ee",
+      "k s",
+      ") ,",
+      "一 个",
+      "▁o ne",
+      "▁on e",
+      "▁ one",
+      "\" :",
+      "ou nd",
+      "oun d",
+      "o und",
+      "▁g et",
+      "▁ge t",
+      "▁ get",
+      "po n",
+      "p on",
+      "ow n",
+      "o wn",
+      "▁al so",
+      "▁als o",
+      "▁ also",
+      "▁in ter",
+      "▁int er",
+      "▁i nter",
+      "▁ inter",
+      "▁m ore",
+      "▁mor e",
+      "▁mo re",
+      "▁ more",
+      "se lf",
+      "sel f",
+      "s elf",
+      "g r",
+      "{ \\",
+      "▁s c",
+      "▁ sc",
+      "▁w ork",
+      "▁wor k",
+      "▁wo rk",
+      "▁ work",
+      "▁p re",
+      "▁pr e",
+      "▁ pre",
+      "ot h",
+      "o th",
+      "cl ud",
+      "被 告",
+      "on t",
+      "o nt",
+      "] \\",
+      "r a",
+      "▁the ir",
+      "▁ their",
+      "re at",
+      "rea t",
+      "r eat",
+      "un ction",
+      "unc tion",
+      "unct ion",
+      "el y",
+      "e ly",
+      "ia n",
+      "i an",
+      "et h",
+      "e th",
+      "▁ '",
+      "ro w",
+      "r ow",
+      "ve ry",
+      "ver y",
+      "v ery",
+      "ys t",
+      "y st",
+      "▁e v",
+      "▁ ev",
+      "nim port",
+      "n import",
+      "i e",
+      "_ _",
+      "n R",
+      "\" ,",
+      "▁m an",
+      "▁ma n",
+      "▁ man",
+      "is h",
+      "i sh",
+      "if ic",
+      "ifi c",
+      "ri b",
+      "r ib",
+      "▁ <",
+      "w e",
+      "▁ Y",
+      "▁p art",
+      "▁par t",
+      "▁pa rt",
+      "▁ part",
+      "o b",
+      "m s",
+      "▁t e",
+      "▁ te",
+      "▁C h",
+      "▁ Ch",
+      "ig h",
+      "i gh",
+      "本 院",
+      "▁i f",
+      "▁ if",
+      "lo w",
+      "l ow",
+      "in gs",
+      "ing s",
+      "od el",
+      "ode l",
+      "o del",
+      "ie w",
+      "i ew",
+      "} {",
+      "可 以",
+      "le ct",
+      "lec t",
+      "l ect",
+      "o v",
+      "me r",
+      "m er",
+      "un der",
+      "und er",
+      "unde r",
+      "u nder",
+      "▁o ur",
+      "▁ou r",
+      "▁ our",
+      "ar k",
+      "a rk",
+      "ic k",
+      "i ck",
+      "on d",
+      "o nd",
+      "原 告",
+      ":/ /",
+      ": //",
+      "▁l i",
+      "▁ li",
+      "p r",
+      "ir st",
+      "irs t",
+      "i rst",
+      "▁s u",
+      "▁ su",
+      "▁t ime",
+      "▁tim e",
+      "▁ti me",
+      "▁ time",
+      "ys tem",
+      "yst em",
+      "y stem",
+      "▁st ud",
+      "▁ stud",
+      "▁p r",
+      "▁ pr",
+      "ic t",
+      "i ct",
+      "er t",
+      "e rt",
+      "at es",
+      "ate s",
+      "a tes",
+      "w o",
+      "▁r o",
+      "▁ ro",
+      "▁I t",
+      "▁ It",
+      "oc k",
+      "o ck",
+      "▁t hey",
+      "▁the y",
+      "▁th ey",
+      "▁ they",
+      "ay s",
+      "a ys",
+      "▁d es",
+      "▁de s",
+      "▁ des",
+      "申 请",
+      "( \"",
+      "n C",
+      "en er",
+      "ene r",
+      "e ner",
+      "▁ab out",
+      "▁ about",
+      "us er",
+      "use r",
+      "u ser",
+      "in k",
+      "i nk",
+      "▁c lass",
+      "▁cl ass",
+      "▁ class",
+      "i o",
+      "pt ion",
+      "p tion",
+      "a x",
+      "▁a ny",
+      "▁an y",
+      "▁ any",
+      "▁b een",
+      "▁be en",
+      "▁bee n",
+      "bl e",
+      "b le",
+      "▁a g",
+      "▁ ag",
+      "n W",
+      "ie nt",
+      "ien t",
+      "i ent",
+      "us e",
+      "u se",
+      "ol d",
+      "o ld",
+      "oo l",
+      "o ol",
+      "iz e",
+      "i ze",
+      "pu t",
+      "p ut",
+      "n g",
+      "fr ac",
+      "fra c",
+      "f rac",
+      "u c",
+      "▁y ear",
+      "▁ye ar",
+      "▁ year",
+      "ol og",
+      "olo g",
+      "o log",
+      "ul ar",
+      "ula r",
+      "u lar",
+      "▁h is",
+      "▁hi s",
+      "▁ his",
+      "ft er",
+      "f ter",
+      "n e",
+      "at er",
+      "ate r",
+      "a ter",
+      "▁m y",
+      "▁ my",
+      "ua l",
+      "u al",
+      "▁o ver",
+      "▁ov er",
+      "▁ over",
+      "ov e",
+      "o ve",
+      "al ue",
+      "alu e",
+      "▁ i",
+      "er m",
+      "e rm",
+      "ic s",
+      "i cs",
+      "as ed",
+      "ase d",
+      "a sed",
+      "▁w ho",
+      "▁wh o",
+      "▁ who",
+      "ul l",
+      "u ll",
+      "▁W e",
+      "▁ We",
+      "▁o ther",
+      "▁ot her",
+      "▁ other",
+      "▁a dd",
+      "▁ad d",
+      "▁ add",
+      "▁un der",
+      "▁u nder",
+      "▁und er",
+      "▁ under",
+      "▁k n",
+      "▁ kn",
+      "▁in d",
+      "▁i nd",
+      "▁ ind",
+      "ter n",
+      "te rn",
+      "t ern",
+      "▁b et",
+      "▁be t",
+      "▁ bet",
+      "▁w hen",
+      "▁wh en",
+      "▁whe n",
+      "▁ when",
+      "▁c omm",
+      "▁com m",
+      "▁co mm",
+      "▁ comm",
+      "▁c ol",
+      "▁co l",
+      "▁ col",
+      "s s",
+      "i x",
+      "▁t here",
+      "▁the re",
+      "▁th ere",
+      "▁ther e",
+      "▁ there",
+      "le s",
+      "l es",
+      "he d",
+      "h ed",
+      "执行 人",
+      "执 行人",
+      "▁a m",
+      "▁ am",
+      "el d",
+      "e ld",
+      "s c",
+      "an c",
+      "a nc",
+      "▁s ome",
+      "▁so me",
+      "▁som e",
+      "▁ some",
+      "▁us e",
+      "▁u se",
+      "▁ use",
+      "ri t",
+      "r it",
+      "ap p",
+      "a pp",
+      "▁s et",
+      "▁se t",
+      "▁ set",
+      "▁T his",
+      "▁Th is",
+      "▁ This",
+      "▁re c",
+      "▁r ec",
+      "▁ rec",
+      "): \\",
+      ") :\\",
+      "▁d if",
+      "▁di f",
+      "e g",
+      "at ive",
+      "ati ve",
+      "ativ e",
+      "▁a ss",
+      "▁as s",
+      "▁ ass",
+      "ie d",
+      "i ed",
+      "▁a ct",
+      "▁ac t",
+      "▁ act",
+      "s h",
+      "▁t han",
+      "▁th an",
+      "▁ than",
+      "中 国",
+      "▁e xp",
+      "▁ex p",
+      "▁ exp",
+      "an ge",
+      "ang e",
+      "a nge",
+      "▁f unction",
+      "▁fun ction",
+      "▁func tion",
+      "▁funct ion",
+      "▁ function",
+      "▁n um",
+      "▁nu m",
+      "▁ num",
+      "▁o b",
+      "▁ ob",
+      "ro up",
+      "rou p",
+      "r oup",
+      "oin t",
+      "oi nt",
+      "o int",
+      "▁n o",
+      "▁ no",
+      "ar i",
+      "a ri",
+      "▁ :",
+      "x a",
+      "n P",
+      "▁s up",
+      "▁su p",
+      "▁ sup",
+      "ht tp",
+      "h ttp",
+      "▁s ub",
+      "▁su b",
+      "▁ sub",
+      "a w",
+      "法 院",
+      "▁n eed",
+      "▁ne ed",
+      "▁ need",
+      "▁ &",
+      "▁im port",
+      "▁imp ort",
+      "▁ import",
+      "▁it s",
+      "▁i ts",
+      "▁ its",
+      "▁f irst",
+      "▁fi rst",
+      "▁fir st",
+      "▁ first",
+      "▁p rov",
+      "▁pro v",
+      "▁pr ov",
+      "▁ prov",
+      "s p",
+      "▁re g",
+      "▁r eg",
+      "▁ reg",
+      "ie ld",
+      "iel d",
+      "i eld",
+      "n f",
+      "▁t wo",
+      "▁tw o",
+      "▁ two",
+      "▁p h",
+      "▁ ph",
+      "自 己",
+      "ut ion",
+      "uti on",
+      "u tion",
+      "▁p e",
+      "▁ pe",
+      "ment s",
+      "men ts",
+      "m ents",
+      "▁a cc",
+      "▁ac c",
+      "▁ acc",
+      "ar ch",
+      "arc h",
+      "cr e",
+      "c re",
+      "▁e ff",
+      "▁ eff",
+      "▁ x",
+      "al s",
+      "a ls",
+      "▁in to",
+      "▁int o",
+      "▁i nto",
+      "▁ into",
+      "▁w ould",
+      "▁ would",
+      "▁f e",
+      "▁ fe",
+      "il ity",
+      "ilit y",
+      "ili ty",
+      "▁c o",
+      "▁ co",
+      "^ {",
+      "第 一",
+      "ad e",
+      "a de",
+      "at ing",
+      "atin g",
+      "ati ng",
+      "a ting",
+      "▁s pec",
+      "▁sp ec",
+      "▁spe c",
+      "▁ spec",
+      "t y",
+      "de f",
+      "d ef",
+      "iss ue",
+      "issu e",
+      "vel op",
+      "ar g",
+      "a rg",
+      "ni ng",
+      "nin g",
+      "n ing",
+      "le t",
+      "l et",
+      "pub lic",
+      "pu blic",
+      "p ublic",
+      "er r",
+      "e rr",
+      "▁in clud",
+      "▁incl ud",
+      "▁t rans",
+      "▁tr ans",
+      "▁tra ns",
+      "▁tran s",
+      "▁ trans",
+      "▁l ike",
+      "▁li ke",
+      "▁lik e",
+      "▁ like",
+      "▁re l",
+      "▁r el",
+      "▁ rel",
+      "op le",
+      "opl e",
+      "o ple",
+      "工 作",
+      "▁p at",
+      "▁pa t",
+      "▁ pat",
+      "▁a c",
+      "▁ ac",
+      "有 限",
+      "po s",
+      "p os",
+      "▁h ow",
+      "▁ho w",
+      "▁ how",
+      "▁h igh",
+      "▁hi gh",
+      "▁ high",
+      "▁m ay",
+      "▁ma y",
+      "▁ may",
+      "▁d ef",
+      "▁de f",
+      "▁ def",
+      "pa r",
+      "p ar",
+      "▁re qu",
+      "▁r equ",
+      "▁req u",
+      "▁ requ",
+      "n s",
+      "n M",
+      "at ional",
+      "ation al",
+      "ati onal",
+      "atio nal",
+      "问 题",
+      "▁p rodu",
+      "▁pro du",
+      "▁prod u",
+      "▁ produ",
+      "▁ –",
+      "ra p",
+      "r ap",
+      "人民 法院",
+      "or m",
+      "o rm",
+      "▁in v",
+      "▁i nv",
+      "▁ inv",
+      "li ne",
+      "lin e",
+      "l ine",
+      "ur re",
+      "is e",
+      "i se",
+      "我 们",
+      "E R",
+      "▁s a",
+      "▁ sa",
+      "▁o ff",
+      "▁of f",
+      "▁ off",
+      "▁h ad",
+      "▁ha d",
+      "▁ had",
+      "o y",
+      "n F",
+      "进 行",
+      "n c",
+      "▁d iffer",
+      "▁dif fer",
+      "▁diff er",
+      "▁the se",
+      "▁th ese",
+      "▁ these",
+      "E x",
+      "$ $",
+      "pl ic",
+      "p lic",
+      "ce ss",
+      "ces s",
+      "c ess",
+      "l p",
+      "uc t",
+      "u ct",
+      "t o",
+      "▁p lay",
+      "▁pl ay",
+      "▁ play",
+      "▁d ata",
+      "▁dat a",
+      "▁da ta",
+      "▁ data",
+      "O N",
+      "▁p os",
+      "▁po s",
+      "▁ pos",
+      "ur es",
+      "ure s",
+      "u res",
+      "in al",
+      "ina l",
+      "i nal",
+      "▁U n",
+      "▁ Un",
+      "▁w hat",
+      "▁wh at",
+      "▁ what",
+      "▁on ly",
+      "▁ only",
+      "() \\",
+      "( )\\",
+      "en ces",
+      "ence s",
+      "enc es",
+      "se t",
+      "s et",
+      "_ {",
+      "it ies",
+      "iti es",
+      "i ties",
+      "ha t",
+      "h at",
+      "▁d et",
+      "▁de t",
+      "▁ det",
+      ": :",
+      "n B",
+      "▁res ult",
+      "▁ result",
+      "e v",
+      "ue s",
+      "u es",
+      "ro und",
+      "rou nd",
+      "r ound",
+      "ui ld",
+      "u ild",
+      "n H",
+      "没 有",
+      "▁j ust",
+      "▁ju st",
+      "▁ just",
+      "▁c ons",
+      "▁con s",
+      "▁co ns",
+      "▁ cons",
+      "he s",
+      "h es",
+      "▁o p",
+      "▁ op",
+      "诉 讼",
+      "Pr o",
+      "P ro",
+      "ri v",
+      "r iv",
+      "y m",
+      "S t",
+      "▁e m",
+      "▁ em",
+      "▁d ec",
+      "▁de c",
+      "▁ dec",
+      "ic es",
+      "ice s",
+      "i ces",
+      "▁t hem",
+      "▁the m",
+      "▁th em",
+      "▁ them",
+      "▁a fter",
+      "▁af ter",
+      "▁ after",
+      "(\\ '",
+      "( \\'",
+      "ag es",
+      "age s",
+      "a ges",
+      "lo g",
+      "l og",
+      "id ent",
+      "ide nt",
+      "iden t",
+      "有限 公司",
+      "ro ugh",
+      "rou gh",
+      "r ough",
+      "什 么",
+      "▁H e",
+      "▁ He",
+      "oi d",
+      "o id",
+      "ib le",
+      "ibl e",
+      "i ble",
+      "se r",
+      "s er",
+      "▁us ing",
+      "▁u sing",
+      "▁ using",
+      "▁s ign",
+      "▁si gn",
+      "▁sig n",
+      "▁ sign",
+      "en se",
+      "ens e",
+      "e nse",
+      "▁s ystem",
+      "▁sy stem",
+      "▁sys tem",
+      "▁syst em",
+      "▁ system",
+      "▁f orm",
+      "▁for m",
+      "▁fo rm",
+      "▁ form",
+      "on s",
+      "o ns",
+      "at or",
+      "ato r",
+      "a tor",
+      "\", \\",
+      "\" ,\\",
+      "am es",
+      "ame s",
+      "a mes",
+      "▁t est",
+      "▁te st",
+      "▁ test",
+      "▁l oc",
+      "▁lo c",
+      "▁ loc",
+      "▁P ro",
+      "▁Pr o",
+      "▁ Pro",
+      "it s",
+      "i ts",
+      "时 间",
+      "ur ing",
+      "uri ng",
+      "urin g",
+      "u ring",
+      "ri ght",
+      "rig ht",
+      "r ight",
+      "▁de velop",
+      "▁ develop",
+      "eth od",
+      "▁s erv",
+      "▁se rv",
+      "▁ser v",
+      "▁ serv",
+      "▁us ed",
+      "▁use d",
+      "▁u sed",
+      "▁ used",
+      "▁s how",
+      "▁sh ow",
+      "▁ show",
+      "ke y",
+      "k ey",
+      "n L",
+      "user name",
+      "ol low",
+      "oll ow",
+      "ollo w",
+      "o x",
+      "nT h",
+      "n Th",
+      "▁s ec",
+      "▁se c",
+      "▁ sec",
+      "we en",
+      "w een",
+      "ient s",
+      "ien ts",
+      "i ents",
+      "I N",
+      "ut h",
+      "u th",
+      "▁ }",
+      "▁h er",
+      "▁he r",
+      "▁ her",
+      "规 定",
+      "at ch",
+      "con t",
+      "co nt",
+      "c ont",
+      "▁p eople",
+      "▁pe ople",
+      "▁ people",
+      "▁p res",
+      "▁pre s",
+      "▁pr es",
+      "▁ pres",
+      "▁d ist",
+      "▁dis t",
+      "▁di st",
+      "▁ dist",
+      "n D",
+      "c k",
+      "就 是",
+      "▁t hen",
+      "▁the n",
+      "▁th en",
+      "▁ then",
+      "▁h elp",
+      "▁he lp",
+      "▁hel p",
+      "▁ help",
+      "he m",
+      "h em",
+      "th is",
+      "thi s",
+      "t his",
+      "re ad",
+      "rea d",
+      "r ead",
+      "au se",
+      "aus e",
+      "a use",
+      "as h",
+      "a sh",
+      "ce pt",
+      "cep t",
+      "en g",
+      "e ng",
+      "▁w ell",
+      "▁we ll",
+      "▁wel l",
+      "▁ well",
+      "ne r",
+      "n er",
+      "▁m ost",
+      "▁mo st",
+      "▁mos t",
+      "▁ most",
+      "▁s im",
+      "▁si m",
+      "▁ sim",
+      "▁b r",
+      "▁ br",
+      "at ic",
+      "ati c",
+      "a tic",
+      "ti ng",
+      "tin g",
+      "t ing",
+      "y p",
+      "▁k now",
+      "▁kn ow",
+      "▁ know",
+      "▁th rough",
+      "▁thr ough",
+      "▁thro ugh",
+      "▁ through",
+      "mi n",
+      "m in",
+      "tr ing",
+      "tri ng",
+      "t ring",
+      "▁b etween",
+      "▁bet ween",
+      "▁ between",
+      "▁w here",
+      "▁wh ere",
+      "▁whe re",
+      "▁ where",
+      "▁c ell",
+      "▁ce ll",
+      "▁cel l",
+      "▁ cell",
+      "de r",
+      "d er",
+      "▁re t",
+      "▁r et",
+      "▁ ret",
+      "▁p ol",
+      "▁po l",
+      "▁ pol",
+      "▁ @",
+      "ir ect",
+      "ire ct",
+      "i rect",
+      "it ed",
+      "ite d",
+      "i ted",
+      "▁re m",
+      "▁r em",
+      "▁ rem",
+      "() ;\\",
+      "(); \\",
+      "( );\\",
+      "▁C om",
+      "▁Co m",
+      "▁ Com",
+      "▁c re",
+      "▁cr e",
+      "▁ cre",
+      "▁m ake",
+      "▁ma ke",
+      "▁ make",
+      "\\\\ (",
+      "ai r",
+      "a ir",
+      "▁sh ould",
+      "▁ should",
+      "▁ `",
+      "wo rk",
+      "wor k",
+      "w ork",
+      "这 个",
+      "▁ex per",
+      "▁exp er",
+      "▁ exper",
+      "fi g",
+      "f ig",
+      "I n",
+      "▁s ur",
+      "▁su r",
+      "▁ sur",
+      "被 执行人",
+      "管 理",
+      "oo d",
+      "o od",
+      "▁m odel",
+      "▁mod el",
+      "▁mode l",
+      "▁mo del",
+      "▁ model",
+      "ur rent",
+      "urre nt",
+      "al e",
+      "a le",
+      "ut e",
+      "u te",
+      "so n",
+      "s on",
+      "▁f ind",
+      "▁fin d",
+      "▁fi nd",
+      "▁ find",
+      "▁e nd",
+      "▁en d",
+      "▁ end",
+      "\\' ,",
+      "\\ ',",
+      "r t",
+      "▁s ol",
+      "▁so l",
+      "▁ sol",
+      "il y",
+      "i ly",
+      "裁 定",
+      "oo t",
+      "o ot",
+      "am p",
+      "a mp",
+      "发 展",
+      "R E",
+      "ar s",
+      "a rs",
+      "war d",
+      "wa rd",
+      "w ard",
+      "ow er",
+      "owe r",
+      "o wer",
+      "}\\ \\",
+      "} \\\\",
+      "io us",
+      "i ous",
+      "▁s er",
+      "▁se r",
+      "▁ ser",
+      "▁f il",
+      "▁fi l",
+      "▁ fil",
+      "ov er",
+      "ove r",
+      "o ver",
+      "cl ass",
+      "c lass",
+      "▁a tt",
+      "▁at t",
+      "▁ att",
+      "for e",
+      "fo re",
+      "f ore",
+      "▁f ollow",
+      "▁fol low",
+      "▁ follow",
+      "an y",
+      "a ny",
+      "in es",
+      "ine s",
+      "i nes",
+      "ra ph",
+      "rap h",
+      "r aph",
+      "▁n umber",
+      "▁num ber",
+      "▁nu mber",
+      "▁ number",
+      "ou gh",
+      "oug h",
+      "o ugh",
+      "m l",
+      "▁s m",
+      "▁ sm",
+      "▁l ook",
+      "▁lo ok",
+      "▁ look",
+      "▁b l",
+      "▁ bl",
+      "n O",
+      "▁in st",
+      "▁i nst",
+      "▁ins t",
+      "▁ inst",
+      "▁g r",
+      "▁ gr",
+      "▁s ee",
+      "▁se e",
+      "▁ see",
+      "e e",
+      "at ure",
+      "atur e",
+      "atu re",
+      "a ture",
+      "it le",
+      "i tle",
+      "iz ed",
+      "ize d",
+      "i zed",
+      "ve nt",
+      "ven t",
+      "v ent",
+      "▁m in",
+      "▁mi n",
+      "▁ min",
+      "服 务",
+      "gr am",
+      "gra m",
+      "g ram",
+      "our ce",
+      "Q u",
+      "da y",
+      "d ay",
+      "▁v ari",
+      "▁var i",
+      "▁va ri",
+      "▁ vari",
+      "b r",
+      "er y",
+      "e ry",
+      "▁s uch",
+      "▁su ch",
+      "▁suc h",
+      "▁ such",
+      "A T",
+      "a z",
+      "法 律",
+      "企 业",
+      "▁eff ect",
+      "▁ effect",
+      "Th e",
+      "T he",
+      "▁g ener",
+      "▁gen er",
+      "▁gene r",
+      "▁ge ner",
+      "▁ gener",
+      "中 华",
+      "ce nt",
+      "c ent",
+      "▁d ifferent",
+      "▁differ ent",
+      "▁diff erent",
+      "▁ different",
+      "form ation",
+      "format ion",
+      "t s",
+      "or ld",
+      "▁e nt",
+      "▁en t",
+      "▁ ent",
+      "es tion",
+      "est ion",
+      "vi ew",
+      "vie w",
+      "v iew",
+      "ie r",
+      "i er",
+      "an k",
+      "a nk",
+      "to n",
+      "t on",
+      "in s",
+      "i ns",
+      "an n",
+      "a nn",
+      "▁w ant",
+      "▁wa nt",
+      "▁ want",
+      "▁g roup",
+      "▁gr oup",
+      "▁gro up",
+      "▁ group",
+      "▁b ack",
+      "▁ba ck",
+      "▁bac k",
+      "▁ back",
+      "▁ /",
+      "oc i",
+      "o ci",
+      "第 二",
+      "## ##",
+      "### #",
+      "# ###",
+      "▁c or",
+      "▁co r",
+      "▁ cor",
+      "▁y ears",
+      "▁year s",
+      "▁ye ars",
+      "▁ years",
+      "ble m",
+      "bl em",
+      "b lem",
+      "▁p ar",
+      "▁pa r",
+      "▁ par",
+      "▁in f",
+      "▁i nf",
+      "▁ inf",
+      "ect ion",
+      "ec tion",
+      "e ction",
+      "con st",
+      "co nst",
+      "cons t",
+      "( '",
+      "ch n",
+      "c hn",
+      "cr i",
+      "c ri",
+      "） \\",
+      "pe ct",
+      "pec t",
+      "p ect",
+      "技 术",
+      "id er",
+      "ide r",
+      "i der",
+      "▁d i",
+      "▁ di",
+      "和 国",
+      "▁n ame",
+      "▁na me",
+      "▁nam e",
+      "▁ name",
+      "com ment",
+      "comm ent",
+      "va l",
+      "v al",
+      "re n",
+      "r en",
+      "▁m on",
+      "▁mo n",
+      "▁ mon",
+      "/ \\",
+      "▁s aid",
+      "▁sa id",
+      "le ase",
+      "lea se",
+      "in ed",
+      "ine d",
+      "i ned",
+      "S U",
+      "I T",
+      "as es",
+      "ase s",
+      "a ses",
+      "▁A l",
+      "▁ Al",
+      "▁s he",
+      "▁sh e",
+      "▁ she",
+      "la nd",
+      "lan d",
+      "l and",
+      "如 果",
+      "▁b ec",
+      "▁be c",
+      "▁ bec",
+      "N A",
+      "▁s elf",
+      "▁se lf",
+      "▁sel f",
+      "▁ self",
+      "ir ed",
+      "ire d",
+      "i red",
+      "▁T r",
+      "▁ Tr",
+      "▁st art",
+      "▁star t",
+      "▁ start",
+      "http s",
+      "▁m ed",
+      "▁me d",
+      "▁ med",
+      "da ta",
+      "dat a",
+      "d ata",
+      "nI n",
+      "n In",
+      "di v",
+      "d iv",
+      "fu l",
+      "f ul",
+      "▁c ould",
+      "▁cou ld",
+      "▁ could",
+      "▁f r",
+      "▁ fr",
+      "▁l ong",
+      "▁lo ng",
+      "▁lon g",
+      "▁ long",
+      "ad d",
+      "a dd",
+      "▁e qu",
+      "▁eq u",
+      "▁ equ",
+      "▁e very",
+      "▁ev ery",
+      "▁ever y",
+      "▁ every",
+      "▁Y ou",
+      "▁Yo u",
+      "▁ You",
+      "共和 国",
+      "共 和国",
+      "if y",
+      "i fy",
+      "▁f l",
+      "▁ fl",
+      "ri c",
+      "r ic",
+      "em ent",
+      "emen t",
+      "eme nt",
+      "e ment",
+      "▁in cre",
+      "▁inc re",
+      "▁ incre",
+      "am ple",
+      "amp le",
+      "al ity",
+      "ali ty",
+      "in ce",
+      "inc e",
+      "i nce",
+      "▁m any",
+      "▁man y",
+      "▁ma ny",
+      "▁ many",
+      "▁f ound",
+      "▁fo und",
+      "▁ found",
+      "if e",
+      "i fe",
+      "ga n",
+      "g an",
+      "hi p",
+      "h ip",
+      "n N",
+      "is ts",
+      "ist s",
+      "i sts",
+      "al th",
+      "alt h",
+      "ev er",
+      "eve r",
+      "e ver",
+      "人民 共和国",
+      "S T",
+      "au lt",
+      "aul t",
+      "a ult",
+      "▁c all",
+      "▁ca ll",
+      "▁cal l",
+      "▁ call",
+      "中华 人民共和国",
+      "▁m od",
+      "▁mo d",
+      "▁ mod",
+      "w n",
+      "ce d",
+      "c ed",
+      "▁b oth",
+      "▁bo th",
+      "▁bot h",
+      "▁ both",
+      "um ent",
+      "ume nt",
+      "umen t",
+      "u ment",
+      "▁m em",
+      "▁me m",
+      "▁ mem",
+      "ra y",
+      "r ay",
+      "ic ally",
+      "ical ly",
+      "A R",
+      "ol e",
+      "o le",
+      "ol ogy",
+      "olog y",
+      "olo gy",
+      "o logy",
+      "re d",
+      "r ed",
+      ") )",
+      "od y",
+      "o dy",
+      "民 事",
+      "fil e",
+      "fi le",
+      "f ile",
+      "er son",
+      "ers on",
+      "▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁",
+      "▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁",
+      "▁ ▁▁▁▁▁▁▁▁▁",
+      "he ck",
+      "h eck",
+      "▁a nal",
+      "▁an al",
+      "▁ana l",
+      "▁ anal",
+      "▁n ow",
+      "▁no w",
+      "▁ now",
+      "uc h",
+      "u ch",
+      "im es",
+      "ime s",
+      "i mes",
+      "▁a pp",
+      "▁ap p",
+      "▁ app",
+      "▁c har",
+      "▁ch ar",
+      "▁cha r",
+      "▁ char",
+      "ro l",
+      "r ol",
+      "' ,",
+      "E N",
+      "le ction",
+      "lect ion",
+      "lec tion",
+      "l ection",
+      "▁v alue",
+      "▁val ue",
+      "▁valu e",
+      "▁ value",
+      "com m",
+      "co mm",
+      "c omm",
+      "判 决",
+      "ea d",
+      "e ad",
+      "it t",
+      "i tt",
+      "le d",
+      "l ed",
+      "▁e ach",
+      "▁ each",
+      "▁r ight",
+      "▁rig ht",
+      "▁ri ght",
+      "▁ right",
+      "研 究",
+      "▁f act",
+      "▁fac t",
+      "▁fa ct",
+      "▁ fact",
+      "or n",
+      "o rn",
+      "iz ation",
+      "iza tion",
+      "▁m ain",
+      "▁ma in",
+      "▁ main",
+      "▁ap pro",
+      "▁app ro",
+      "▁ appro",
+      "ro n",
+      "r on",
+      "▁de sign",
+      "▁des ign",
+      "▁ design",
+      "▁pro cess",
+      "▁proc ess",
+      "▁ process",
+      "▁v ery",
+      "▁ve ry",
+      "▁ver y",
+      "▁ very",
+      "re qu",
+      "req u",
+      "r equ",
+      "▁e d",
+      "▁ ed",
+      "iv ers",
+      "ive rs",
+      "iver s",
+      "i vers",
+      "▁$ $",
+      "▁ $$",
+      "▁stud y",
+      "▁ study",
+      "▁c ar",
+      "▁ca r",
+      "▁ car",
+      "▁w ay",
+      "▁wa y",
+      "▁ way",
+      "▁a v",
+      "▁ av",
+      "i k",
+      "▁m ethod",
+      "▁meth od",
+      "▁ method",
+      "▁d oes",
+      "▁do es",
+      "▁ does",
+      "▁v al",
+      "▁va l",
+      "▁ val",
+      "su p",
+      "s up",
+      "▁re ad",
+      "▁r ead",
+      "▁ read",
+      "ie nce",
+      "ien ce",
+      "i ence",
+      "▁res pon",
+      "▁resp on",
+      "▁ respon",
+      "et s",
+      "e ts",
+      "] (",
+      "▁l ist",
+      "▁li st",
+      "▁ list",
+      "o f",
+      "▁g u",
+      "▁ gu",
+      "te xt",
+      "tex t",
+      "t ext",
+      "▁p ost",
+      "▁pos t",
+      "▁po st",
+      "▁ post",
+      "lo ad",
+      "l oad",
+      "▁E x",
+      "▁ Ex",
+      "▁pat ients",
+      "▁patient s",
+      "▁ patients",
+      "al se",
+      "als e",
+      "Na me",
+      "N ame",
+      "co l",
+      "c ol",
+      "is ion",
+      "isi on",
+      "ul e",
+      "u le",
+      "ch ool",
+      "cho ol",
+      "ak ing",
+      "aki ng",
+      "akin g",
+      "a king",
+      "su b",
+      "s ub",
+      "vers ion",
+      "v ersion",
+      "需 要",
+      "▁r un",
+      "▁ru n",
+      "▁ run",
+      "▁e xt",
+      "▁ex t",
+      "▁ ext",
+      "war e",
+      "wa re",
+      "w are",
+      "▁e st",
+      "▁es t",
+      "▁ est",
+      "tt p",
+      "t tp",
+      "▁N ew",
+      "▁Ne w",
+      "▁ New",
+      "te n",
+      "t en",
+      "ar ed",
+      "are d",
+      "a red",
+      "▁g reat",
+      "▁gr eat",
+      "▁gre at",
+      "▁ great",
+      "的 人",
+      "ul ation",
+      "ula tion",
+      "u lation",
+      "O R",
+      "ot her",
+      "oth er",
+      "othe r",
+      "o ther",
+      "▁p oint",
+      "▁po int",
+      "▁ point",
+      "wa y",
+      "w ay",
+      "mer ic",
+      "me ric",
+      "m eric",
+      "▁d isc",
+      "▁dis c",
+      "▁di sc",
+      "▁ disc",
+      "se arch",
+      "sear ch",
+      "s earch",
+      "an ts",
+      "ant s",
+      "ect ed",
+      "ec ted",
+      "▁S h",
+      "▁ Sh",
+      "a h",
+      "因 为",
+      "as on",
+      "a son",
+      "不 是",
+      "▁d on",
+      "▁do n",
+      "▁ don",
+      "iq u",
+      "i qu",
+      "▁A n",
+      "▁ An",
+      "信 息",
+      "ef er",
+      "efe r",
+      "e fer",
+      "d d",
+      "ess age",
+      "essa ge",
+      "Con t",
+      "Co nt",
+      "C ont",
+      "使 用",
+      "▁$ \\\\",
+      "▁$\\ \\",
+      "▁ $\\\\",
+      "▁in formation",
+      "▁inform ation",
+      "▁ information",
+      "fi l",
+      "f il",
+      "ion al",
+      "io nal",
+      "iona l",
+      "i onal",
+      "▁re p",
+      "▁r ep",
+      "▁ rep",
+      "▁h ere",
+      "▁he re",
+      "▁her e",
+      "▁ here",
+      "▁f am",
+      "▁fa m",
+      "in ess",
+      "ine ss",
+      "ines s",
+      "i ness",
+      "ou se",
+      "ous e",
+      "o use",
+      "▁I f",
+      "▁ If",
+      "ent ial",
+      "enti al",
+      "pat h",
+      "pa th",
+      "p ath",
+      "th e",
+      "t he",
+      "▁a rt",
+      "▁ar t",
+      "▁ art",
+      "y n",
+      "** \\",
+      "* *\\",
+      "根 据",
+      "ma n",
+      "m an",
+      "\" \"",
+      "▁act iv",
+      "▁ activ",
+      "p s",
+      "系 统",
+      "产 品",
+      "▁sup port",
+      "▁ support",
+      "日 期",
+      "▁g iv",
+      "▁gi v",
+      "▁d uring",
+      "▁du ring",
+      "▁dur ing",
+      "op y",
+      "o py",
+      "▁s ame",
+      "▁sa me",
+      "▁sam e",
+      "▁ same",
+      "at eg",
+      "ate g",
+      "▁f ile",
+      "▁fil e",
+      "▁fi le",
+      "▁ file",
+      "ty pe",
+      "typ e",
+      "t ype",
+      "通 过",
+      "pr int",
+      "pri nt",
+      "p rint",
+      "is hed",
+      "ish ed",
+      "ra l",
+      "r al",
+      "e k",
+      "审 判",
+      "▁g ood",
+      "▁go od",
+      "▁ good",
+      "I G",
+      "▁l ine",
+      "▁li ne",
+      "▁lin e",
+      "▁ line",
+      "所 以",
+      "da te",
+      "dat e",
+      "d ate",
+      "ur al",
+      "ura l",
+      "u ral",
+      "ic ense",
+      "ice nse",
+      "icens e",
+      "▁e ven",
+      "▁ev en",
+      "▁ even",
+      "S E",
+      "## #",
+      "# ##",
+      "中 文",
+      "el se",
+      "els e",
+      "n E",
+      "\" )",
+      "us s",
+      "u ss",
+      "} $",
+      "▁t ype",
+      "▁typ e",
+      "▁ty pe",
+      "▁ type",
+      "en cy",
+      "enc y",
+      "e ncy",
+      "▁le vel",
+      "▁lev el",
+      "▁ level",
+      "d e",
+      "he r",
+      "h er",
+      "认 为",
+      "▁b ecause",
+      "▁be cause",
+      "▁bec ause",
+      "▁ because",
+      "y l",
+      "▁b el",
+      "▁be l",
+      "▁ bel",
+      "▁ >",
+      "ce r",
+      "c er",
+      "▁ %",
+      "ain ing",
+      "ai ning",
+      "a ining",
+      "ur l",
+      "u rl",
+      "M E",
+      "主 要",
+      "▁te chn",
+      "▁tech n",
+      "▁ techn",
+      "nf rom",
+      "n from",
+      "▁per form",
+      "▁perf orm",
+      "▁perfor m",
+      "▁ perform",
+      "▁st ate",
+      "▁stat e",
+      "▁ state",
+      "▁c ent",
+      "▁ce nt",
+      "▁ cent",
+      "iu m",
+      "i um",
+      "o h",
+      "▁m et",
+      "▁me t",
+      "▁ met",
+      "riv ate",
+      "r ivate",
+      "ber s",
+      "be rs",
+      "b ers",
+      "▁pro gram",
+      "▁pr ogram",
+      "▁ program",
+      "if ied",
+      "ifi ed",
+      "▁pro ject",
+      "▁proj ect",
+      "▁ project",
+      "提 供",
+      "▁a gain",
+      "▁ag ain",
+      "▁ again",
+      "▁R e",
+      "▁ Re",
+      "as k",
+      "a sk",
+      "ac tion",
+      "act ion",
+      "a ction",
+      "ain s",
+      "ai ns",
+      "a ins",
+      "▁F or",
+      "▁Fo r",
+      "▁ For",
+      "w h",
+      "▁d own",
+      "▁do wn",
+      "▁dow n",
+      "▁ down",
+      "时 候",
+      "教 育",
+      "▁m at",
+      "▁ma t",
+      "▁ mat",
+      "ce s",
+      "c es",
+      "ro ll",
+      "rol l",
+      "r oll",
+      "▁c urrent",
+      "▁cur rent",
+      "▁curr ent",
+      "▁ current",
+      "资 料",
+      "pl ay",
+      "p lay",
+      "▁b efore",
+      "▁be fore",
+      "▁ before",
+      "▁C l",
+      "▁ Cl",
+      "▁e rr",
+      "▁er r",
+      "▁ err",
+      "le x",
+      "l ex",
+      "I d",
+      "b s",
+      "选 择",
+      "▁ z",
+      "A N",
+      "▁part ic",
+      "▁par tic",
+      "如 下",
+      "us tom",
+      "ust om",
+      "ter s",
+      "te rs",
+      "t ers",
+      "p p",
+      "▁g en",
+      "▁ge n",
+      "▁ gen",
+      "▁A r",
+      "▁ Ar",
+      "a j",
+      "ve d",
+      "v ed",
+      "▁m ult",
+      "▁mu lt",
+      "▁mul t",
+      "▁ mult",
+      "am s",
+      "a ms",
+      "ba ck",
+      "bac k",
+      "b ack",
+      "▁se cond",
+      "▁sec ond",
+      "▁ second",
+      "▁d ep",
+      "▁de p",
+      "▁ dep",
+      "] ,",
+      "ual ly",
+      "u ally",
+      "▁b est",
+      "▁be st",
+      "▁bes t",
+      "▁ best",
+      "Co n",
+      "C on",
+      "▁be ing",
+      "▁bei ng",
+      "▁ being",
+      "te st",
+      "tes t",
+      "t est",
+      "efer ences",
+      "eference s",
+      "e ferences",
+      "▁m ade",
+      "▁ma de",
+      "▁mad e",
+      "▁ made",
+      "b l",
+      "us iness",
+      "▁th ree",
+      "▁thr ee",
+      "▁ three",
+      "参 考",
+      "▁c are",
+      "▁car e",
+      "▁ca re",
+      "▁ care",
+      "▁pro vid",
+      "▁prov id",
+      "大 学",
+      "Co m",
+      "C om",
+      "▁ Z",
+      "▁ “",
+      "** **",
+      "*** *",
+      "* ***",
+      "St ring",
+      "Str ing",
+      "S tring",
+      "io r",
+      "i or",
+      "I D",
+      "ot e",
+      "o te",
+      "▁d id",
+      "▁di d",
+      "▁ did",
+      "ress ion",
+      "r ession",
+      "▁m ark",
+      "▁mar k",
+      "▁ma rk",
+      "▁ mark",
+      "可 能",
+      "is m",
+      "i sm",
+      "▁C on",
+      "▁Co n",
+      "▁ Con",
+      "df rac",
+      "d frac",
+      "C T",
+      "▁h and",
+      "▁ha nd",
+      "▁han d",
+      "▁ hand",
+      "yl e",
+      "y le",
+      "▁d irect",
+      "▁di rect",
+      "▁dir ect",
+      "▁dire ct",
+      "▁ direct",
+      "ri but",
+      "rib ut",
+      "it al",
+      "ita l",
+      "i tal",
+      "社 会",
+      "ab ility",
+      "abil ity",
+      "cc ess",
+      "c cess",
+      "▁pro duct",
+      "▁produ ct",
+      "▁prod uct",
+      "▁ product",
+      "▁wh ile",
+      "▁ while",
+      "s q",
+      "we r",
+      "w er",
+      "▁c ount",
+      "▁co unt",
+      "▁cou nt",
+      "▁ count",
+      "他 们",
+      "发 生",
+      "nR eferences",
+      "nRe ferences",
+      "n References",
+      "▁d r",
+      "▁ dr",
+      "国 家",
+      "it hub",
+      "ith ub",
+      "ack age",
+      "▁l ead",
+      "▁le ad",
+      "▁ lead",
+      "▁th ose",
+      "生 活",
+      "▁b uild",
+      "▁bu ild",
+      "▁ build",
+      "an e",
+      "a ne",
+      "方 法",
+      "▁f in",
+      "▁fi n",
+      "▁ fin",
+      "L E",
+      "ac ter",
+      "act er",
+      "cl ude",
+      "clud e",
+      "▁o pen",
+      "▁op en",
+      "▁ open",
+      "▁* /\\",
+      "▁*/ \\",
+      "▁ */\\",
+      "ar n",
+      "a rn",
+      ", \"",
+      "审 理",
+      "情 况",
+      "en ds",
+      "end s",
+      "▁res ults",
+      "▁result s",
+      "▁ results",
+      "ab el",
+      "abe l",
+      "a bel",
+      "ba ll",
+      "bal l",
+      "b all",
+      "但 是",
+      "n G",
+      "▁th ink",
+      "▁thin k",
+      "▁thi nk",
+      "▁ think",
+      "p y",
+      "us ing",
+      "u sing",
+      "nT his",
+      "nTh is",
+      "n This",
+      "诉讼 法",
+      "现 在",
+      "▁n on",
+      "▁no n",
+      "▁ non",
+      "▁m uch",
+      "▁muc h",
+      "▁mu ch",
+      "▁in s",
+      "▁i ns",
+      "▁ ins",
+      "ro ss",
+      "ros s",
+      "r oss",
+      "经 济",
+      "的 一",
+      "ar m",
+      "a rm",
+      "b e",
+      "va r",
+      "v ar",
+      "▁ch ild",
+      "▁chi ld",
+      "▁ child",
+      "b y",
+      "ne t",
+      "n et",
+      "or g",
+      "o rg",
+      "fa ce",
+      "fac e",
+      "f ace",
+      "是 一",
+      "▁re port",
+      "▁rep ort",
+      "▁repo rt",
+      "▁ report",
+      "al c",
+      "a lc",
+      "▁l ast",
+      "▁la st",
+      "▁las t",
+      "▁ last",
+      "▁ #",
+      "提 出",
+      "▁H ow",
+      "▁Ho w",
+      "▁ How",
+      "▁com mun",
+      "▁comm un",
+      "▁ commun",
+      "ic ro",
+      "i cro",
+      "it ive",
+      "iti ve",
+      "▁c ap",
+      "▁ca p",
+      "▁ cap",
+      "eng th",
+      "e ngth",
+      "▁t ake",
+      "▁ta ke",
+      "▁tak e",
+      "▁ take",
+      "ain ed",
+      "ai ned",
+      "aine d",
+      "a ined",
+      "li sh",
+      "lis h",
+      "l ish",
+      "▁d el",
+      "▁de l",
+      "▁ del",
+      "\") \\",
+      "\" )\\",
+      "at us",
+      "atu s",
+      "og raph",
+      "ogr aph",
+      "ograp h",
+      "o graph",
+      "已 经",
+      "设 计",
+      "▁p oss",
+      "▁pos s",
+      "▁po ss",
+      "▁ poss",
+      "▁p resent",
+      "▁pre sent",
+      "▁pres ent",
+      "▁ present",
+      "t a",
+      "▁in vest",
+      "▁inv est",
+      "▁ invest",
+      "com e",
+      "co me",
+      "c ome",
+      "▁or der",
+      "▁ord er",
+      "▁ order",
+      "▁d ay",
+      "▁da y",
+      "▁ day",
+      "I C",
+      "▁ 书",
+      "▁o per",
+      "▁op er",
+      "▁ oper",
+      "上 诉",
+      "ow s",
+      "o ws",
+      "eq u",
+      "e qu",
+      "ge r",
+      "g er",
+      "▁s om",
+      "▁so m",
+      "▁A s",
+      "▁ As",
+      "▁re search",
+      "▁res earch",
+      "▁resear ch",
+      "▁ research",
+      "分 析",
+      "出 生",
+      "st and",
+      "stan d",
+      "sta nd",
+      "s tand",
+      "▁o pt",
+      "▁op t",
+      "▁ opt",
+      "erv ice",
+      "\\\\ )",
+      "tim e",
+      "ti me",
+      "t ime",
+      "ar get",
+      "arg et",
+      "arge t",
+      "▁is s",
+      "▁i ss",
+      "▁ iss",
+      "vi r",
+      "v ir",
+      "▁h ome",
+      "▁hom e",
+      "▁ho me",
+      "▁ home",
+      "还 是",
+      "▁v is",
+      "▁vi s",
+      "▁ vis",
+      "at her",
+      "ath er",
+      "athe r",
+      "a ther",
+      "▁o wn",
+      "▁ow n",
+      "▁ own",
+      "▁re turn",
+      "▁r eturn",
+      "▁ret urn",
+      "▁ return",
+      "▁a nt",
+      "▁an t",
+      "▁ ant",
+      "git hub",
+      "g ithub",
+      "▁t op",
+      "▁to p",
+      "▁ top",
+      "代 表",
+      "▁A meric",
+      "▁Am eric",
+      "▁Amer ic",
+      "▁ Americ",
+      "ut hor",
+      "uth or",
+      "u thor",
+      "▁s k",
+      "▁ sk",
+      "ail able",
+      "ht ml",
+      "htm l",
+      "h tml",
+      "▁h ttp",
+      "▁ http",
+      "ma th",
+      "mat h",
+      "m ath",
+      "ne ss",
+      "nes s",
+      "n ess",
+      "▁t reat",
+      "▁tr eat",
+      "▁tre at",
+      "T r",
+      "▁pro per",
+      "▁pr oper",
+      "▁prop er",
+      "▁ proper",
+      "▁s mall",
+      "▁sm all",
+      "▁ small",
+      "引 用",
+      "▁c r",
+      "▁ cr",
+      ")) \\",
+      ") )\\",
+      "▁ap plic",
+      "▁app lic",
+      "▁appl ic",
+      "E F",
+      "汉 族",
+      "内 容",
+      "le ft",
+      "lef t",
+      "l eft",
+      "r r",
+      "m a",
+      "ug g",
+      "u gg",
+      "▁c hang",
+      "▁ch ang",
+      "▁cha ng",
+      "▁chan g",
+      "▁ chang",
+      "iv ely",
+      "ive ly",
+      "ivel y",
+      "i vely",
+      "▁S c",
+      "▁ Sc",
+      "du ct",
+      "duc t",
+      "d uct",
+      "oa d",
+      "o ad",
+      "ut il",
+      "uti l",
+      "u til",
+      "ce ption",
+      "cept ion",
+      "cep tion",
+      "ivers ity",
+      "▁A nd",
+      "▁An d",
+      "▁ And",
+      "T h",
+      "要 求",
+      "ne w",
+      "n ew",
+      "re am",
+      "rea m",
+      "市 场",
+      "▁p op",
+      "▁po p",
+      "▁ pop",
+      "▁s ing",
+      "▁sin g",
+      "▁si ng",
+      "▁ sing",
+      "or th",
+      "ort h",
+      "C l",
+      "▁D e",
+      "▁ De",
+      "E S",
+      "▁with out",
+      "▁ without",
+      "ak es",
+      "ake s",
+      "a kes",
+      "▁ide nt",
+      "▁id ent",
+      "▁ ident",
+      "出 版",
+      "▁in cluding",
+      "▁includ ing",
+      "▁incl uding",
+      "▁ including",
+      "A S",
+      "em s",
+      "e ms",
+      "ys is",
+      "y sis",
+      "▁C ol",
+      "▁Co l",
+      "▁ Col",
+      "▁tr ue",
+      "▁ true",
+      "▁he alth",
+      "▁heal th",
+      "▁ health",
+      "过 程",
+      "其 他",
+      "re g",
+      "r eg",
+      "▁I nd",
+      "▁In d",
+      "▁ Ind",
+      "▁ _",
+      "me d",
+      "m ed",
+      "st ring",
+      "str ing",
+      "stri ng",
+      "s tring",
+      "bi n",
+      "b in",
+      "申请 执行人",
+      "申请执行 人",
+      "is ter",
+      "ist er",
+      "iste r",
+      "i ster",
+      "} }",
+      "▁s tring",
+      "▁st ring",
+      "▁str ing",
+      "▁stri ng",
+      "▁ string",
+      "引用 日期",
+      "di t",
+      "d it",
+      "st art",
+      "star t",
+      "sta rt",
+      "ur o",
+      "u ro",
+      "le ment",
+      "lem ent",
+      "l ement",
+      "C h",
+      "▁p ath",
+      "▁pat h",
+      "▁pa th",
+      "▁ path",
+      "▁e lect",
+      "▁el ect",
+      "▁ele ct",
+      "▁ elect",
+      "Li st",
+      "L ist",
+      "▁t ri",
+      "▁tr i",
+      "▁ tri",
+      "▁c le",
+      "▁cl e",
+      "▁ cle",
+      "代 理",
+      "▁ X",
+      "▁b ased",
+      "▁bas ed",
+      "▁base d",
+      "▁ba sed",
+      "▁ based",
+      "▁p erson",
+      "▁per son",
+      "▁pers on",
+      "▁ person",
+      "A L",
+      "al k",
+      "at s",
+      "a ts",
+      "cri ption",
+      "cript ion",
+      "\" >",
+      "in clude",
+      "\"> \\",
+      "\" >\\",
+      "民事 诉讼法",
+      "民事诉讼 法",
+      "er g",
+      "e rg",
+      "av a",
+      "a va",
+      "tr act",
+      "tra ct",
+      "” \\",
+      "▁f ollowing",
+      "▁follow ing",
+      "▁ following",
+      "y t",
+      "iv es",
+      "ive s",
+      "i ves",
+      "Typ e",
+      "Ty pe",
+      "T ype",
+      "Re s",
+      "R es",
+      "审判 员",
+      "re en",
+      "ree n",
+      "r een",
+      "th s",
+      "t hs",
+      "oc ument",
+      "▁w orld",
+      "▁wor ld",
+      "▁ world",
+      "▁l ife",
+      "▁li fe",
+      "▁lif e",
+      "▁ life",
+      "ar ge",
+      "arg e",
+      "oi n",
+      "o in",
+      "中华人民共和国 民事诉讼法",
+      "▁W h",
+      "▁ Wh",
+      "iv id",
+      "ivi d",
+      "i vid",
+      "l i",
+      "wo rd",
+      "wor d",
+      "w ord",
+      "▁sign ific",
+      "et y",
+      "e ty",
+      "▁k e",
+      "▁ ke",
+      "om en",
+      "ome n",
+      "o men",
+      "le ss",
+      "les s",
+      "l ess",
+      "▁中 文",
+      "▁ 中文",
+      "sq rt",
+      "s qrt",
+      "e y",
+      "▁m eas",
+      "▁me as",
+      "pos e",
+      "po se",
+      "p ose",
+      "▁con st",
+      "▁co nst",
+      "▁cons t",
+      "▁ const",
+      "ir c",
+      "i rc",
+      "▁R es",
+      "▁Re s",
+      "▁ Res",
+      "ol ution",
+      "▁st ruct",
+      "▁str uct",
+      "▁ struct",
+      "am et",
+      "ame t",
+      "a met",
+      "▁b usiness",
+      "▁bus iness",
+      "▁ business",
+      "▁a ut",
+      "▁au t",
+      "▁ aut",
+      "▁P l",
+      "▁ Pl",
+      "▁e as",
+      "▁ eas",
+      "▁e ar",
+      "▁ ear",
+      "都 是",
+      "▁te am",
+      "▁tea m",
+      "▁ team",
+      "re f",
+      "r ef",
+      "po r",
+      "p or",
+      "▁p ass",
+      "▁pa ss",
+      "▁pas s",
+      "▁ pass",
+      "▁c ase",
+      "▁ca se",
+      "▁cas e",
+      "▁ case",
+      "文 化",
+      "ve n",
+      "v en",
+      "ur y",
+      "u ry",
+      "▁b u",
+      "▁ bu",
+      "▁f ree",
+      "▁fr ee",
+      "▁fre e",
+      "▁ free",
+      "ra tion",
+      "rat ion",
+      "ratio n",
+      "r ation",
+      "ac es",
+      "ace s",
+      "a ces",
+      "▁w rit",
+      "▁wr it",
+      "▁ writ",
+      "很 多",
+      "中 的",
+      "财 产",
+      "oa rd",
+      "o ard",
+      "way s",
+      "wa ys",
+      "w ays",
+      "▁a dv",
+      "▁ad v",
+      "▁ adv",
+      "项 目",
+      "人 的",
+      "▁F r",
+      "▁ Fr",
+      "▁L icense",
+      "▁Lic ense",
+      "▁ License",
+      "▁中文 名",
+      "▁中 文名",
+      "▁ 中文名",
+      "▁f ield",
+      "▁fi eld",
+      "▁ field",
+      "▁re ce",
+      "▁rec e",
+      "iss ion",
+      "学 习",
+      "二 〇",
+      "}, \\",
+      "} ,\\",
+      "▁S p",
+      "▁ Sp",
+      "au t",
+      "a ut",
+      "▁av ailable",
+      "▁avail able",
+      "▁ available",
+      "▁cont rol",
+      "▁contr ol",
+      "▁ control",
+      "a u",
+      "ur s",
+      "u rs",
+      "ap er",
+      "ape r",
+      "a per",
+      "re m",
+      "r em",
+      "SU P",
+      "S UP",
+      "▁con tin",
+      "▁cont in",
+      "▁ contin",
+      "▁c ells",
+      "▁cell s",
+      "▁cel ls",
+      "▁ cells",
+      "n p",
+      "学 生",
+      "二〇 一",
+      "数 据",
+      "▁pl ace",
+      "▁plac e",
+      "▁ place",
+      "n Y",
+      "或 者",
+      "in a",
+      "i na",
+      "▁prov ide",
+      "▁provid e",
+      "▁ provide",
+      "依 照",
+      "di s",
+      "d is",
+      "ac he",
+      "ach e",
+      "a che",
+      "建 设",
+      "专 业",
+      "mo d",
+      "m od",
+      "ap e",
+      "a pe",
+      "不 同",
+      "ter nal",
+      "tern al",
+      "t ernal",
+      "▁I s",
+      "▁ Is",
+      "lo b",
+      "l ob",
+      "开 始",
+      "▁re al",
+      "▁r eal",
+      "▁ real",
+      "us h",
+      "u sh",
+      "] [",
+      "▁pro blem",
+      "▁prob lem",
+      "▁ problem",
+      "▁d em",
+      "▁de m",
+      "▁ dem",
+      "向 本院",
+      "ind ex",
+      "inde x",
+      "ic le",
+      "icl e",
+      "i cle",
+      "r c",
+      "▁st ill",
+      "▁ still",
+      "iv ed",
+      "ive d",
+      "i ved",
+      "作 者",
+      "▁a round",
+      "▁ar ound",
+      "▁ around",
+      "▁t em",
+      "▁te m",
+      "▁ tem",
+      "j s",
+      "FI G",
+      "F IG",
+      "le te",
+      "let e",
+      "l ete",
+      "▁c our",
+      "▁co ur",
+      "▁cou r",
+      "▁s tand",
+      "▁st and",
+      "▁ stand",
+      "op e",
+      "o pe",
+      "书 记",
+      "\\' \\",
+      "\\ '\\",
+      "vir on",
+      "vi ron",
+      "v iron",
+      "世 界",
+      "as ter",
+      "ast er",
+      "aste r",
+      "a ster",
+      "▁h um",
+      "▁hu m",
+      "▁ hum",
+      "no t",
+      "n ot",
+      "▁or gan",
+      "▁org an",
+      "▁ organ",
+      "▁ex pl",
+      "▁exp l",
+      "▁ expl",
+      "ve rt",
+      "ver t",
+      "v ert",
+      "这 样",
+      "ou th",
+      "out h",
+      "o uth",
+      "c m",
+      "pl oy",
+      "p loy",
+      "in ks",
+      "ink s",
+      "ti l",
+      "t il",
+      "活 动",
+      "ri te",
+      "rit e",
+      "r ite",
+      "▁k ey",
+      "▁ke y",
+      "▁ key",
+      "知 道",
+      "参考 资料",
+      "te m",
+      "t em",
+      "▁P h",
+      "▁ Ph",
+      "▁l og",
+      "▁lo g",
+      "▁ log",
+      "w w",
+      "▁sign ificant",
+      "▁signific ant",
+      "re sent",
+      "res ent",
+      "工 程",
+      "in it",
+      "ini t",
+      "i nit",
+      "ail s",
+      "ai ls",
+      "a ils",
+      "银 行",
+      "{\\ \\",
+      "{ \\\\",
+      "▁b ook",
+      "▁bo ok",
+      "▁ book",
+      "▁g row",
+      "▁gr ow",
+      "▁gro w",
+      "▁ grow",
+      "▁de velopment",
+      "▁develop ment",
+      "▁ development",
+      "▁a rea",
+      "▁are a",
+      "▁ar ea",
+      "▁ area",
+      "关 于",
+      "▁im pro",
+      "▁imp ro",
+      "▁v e",
+      "▁ ve",
+      "▁d eterm",
+      "▁de term",
+      "▁det erm",
+      "▁deter m",
+      "▁ determ",
+      "重 要",
+      "▁l im",
+      "▁li m",
+      "▁ lim",
+      "iv ing",
+      "ivi ng",
+      "i ving",
+      "▁L et",
+      "▁Le t",
+      "▁ Let",
+      "at t",
+      "a tt",
+      "av ing",
+      "avi ng",
+      "avin g",
+      "a ving",
+      "el s",
+      "e ls",
+      "li st",
+      "lis t",
+      "l ist",
+      "以 及",
+      "ar ent",
+      "are nt",
+      "aren t",
+      "a rent",
+      "▁T here",
+      "▁The re",
+      "▁Th ere",
+      "▁Ther e",
+      "▁ There",
+      "的 是",
+      "ot s",
+      "o ts",
+      "nd ef",
+      "nde f",
+      "n def",
+      "if ication",
+      "ific ation",
+      "ifi cation",
+      "▁s ever",
+      "▁se ver",
+      "▁ sever",
+      "a f",
+      "安 全",
+      "▁import ant",
+      "▁ important",
+      "re l",
+      "r el",
+      "an gu",
+      "ang u",
+      "关 系",
+      "os s",
+      "o ss",
+      "ww w",
+      "w ww",
+      "方 式",
+      "好 的",
+      "ai t",
+      "a it",
+      "▁S e",
+      "▁ Se",
+      "本 案",
+      "▁m us",
+      "▁mu s",
+      "▁ mus",
+      "游 戏",
+      "▁re du",
+      "▁r edu",
+      "▁red u",
+      "▁ redu",
+      "▁f amily",
+      "▁fam ily",
+      "▁famil y",
+      "▁ family",
+      "▁h im",
+      "▁hi m",
+      "▁ him",
+      "cri b",
+      "cr ib",
+      "c rib",
+      "▁con f",
+      "▁co nf",
+      "▁ conf",
+      "▁b i",
+      "▁ bi",
+      "中 心",
+      "mu n",
+      "m un",
+      "部 分",
+      "▁ad dition",
+      "▁add ition",
+      "n a",
+      "val ue",
+      "v alue",
+      "▁h ead",
+      "▁he ad",
+      "▁ head",
+      "▁ 定",
+      "op t",
+      "o pt",
+      "na l",
+      "n al",
+      "), \\",
+      ") ,\\",
+      "依 法",
+      "▁l ocal",
+      "▁loc al",
+      "▁lo cal",
+      "▁ local",
+      "比 较",
+      "▁L e",
+      "▁ Le",
+      "l u",
+      "▁T o",
+      "▁ To",
+      "ot t",
+      "o tt",
+      "st ruct",
+      "str uct",
+      "影 响",
+      "▁with in",
+      "▁wit hin",
+      "▁wi thin",
+      "▁ within",
+      "▁ch ange",
+      "▁chang e",
+      "▁cha nge",
+      "▁chan ge",
+      "▁ change",
+      "▁s ince",
+      "▁sin ce",
+      "▁si nce",
+      "▁sinc e",
+      "▁ since",
+      "uc k",
+      "u ck",
+      "— —",
+      "iqu e",
+      "iq ue",
+      "i que",
+      "自己 的",
+      "lish ed",
+      "lis hed",
+      "l ished",
+      "at ely",
+      "ate ly",
+      "不 能",
+      "gh t",
+      "g ht",
+      "▁A ll",
+      "▁Al l",
+      "▁ All",
+      "▁o bject",
+      "▁ob ject",
+      "▁obj ect",
+      "▁ object",
+      "发 现",
+      "▁m anag",
+      "▁man ag",
+      "▁ma nag",
+      "同 时",
+      "▁im p",
+      "▁i mp",
+      "▁ imp",
+      "▁cons ider",
+      "▁ consider",
+      "怎 么",
+      "vi ous",
+      "v ious",
+      "mod el",
+      "mode l",
+      "mo del",
+      "m odel",
+      "( \\",
+      "▁g iven",
+      "▁giv en",
+      "▁give n",
+      "▁gi ven",
+      "▁ given",
+      "rr or",
+      "r ror",
+      "▁ass oci",
+      "▁ associ",
+      "IO N",
+      "I ON",
+      "相 关",
+      "▁ ,",
+      "I S",
+      "生 产",
+      "▁p ublic",
+      "▁pub lic",
+      "▁pu blic",
+      "▁ public",
+      "▁b eh",
+      "▁be h",
+      "▁ beh",
+      "et work",
+      "▁ Q",
+      "▁p ot",
+      "▁po t",
+      "▁ pot",
+      "▁b as",
+      "▁ba s",
+      "▁ bas",
+      "▁S o",
+      "▁ So",
+      "eg r",
+      "e gr",
+      "Dat a",
+      "Da ta",
+      "D ata",
+      "▁c ustom",
+      "▁cust om",
+      "▁ custom",
+      "▁s ervice",
+      "▁serv ice",
+      "▁ service",
+      "▁us er",
+      "▁use r",
+      "▁u ser",
+      "▁ user",
+      "le r",
+      "l er",
+      "nE x",
+      "n Ex",
+      "▁Un iversity",
+      "▁Univers ity",
+      "▁ University",
+      "', \\",
+      "' ,\\",
+      "▁v oid",
+      "▁vo id",
+      "▁ void",
+      "U T",
+      "▁W hat",
+      "▁Wh at",
+      "▁ What",
+      "计 算",
+      "▁re quire",
+      "▁requ ire",
+      "▁ require",
+      "▁n ext",
+      "▁ne xt",
+      "▁ next",
+      "nu m",
+      "n um",
+      "▁s it",
+      "▁si t",
+      "▁ sit",
+      "▁b re",
+      "▁br e",
+      "▁ bre",
+      "第一 百",
+      "第 一百",
+      "em ber",
+      "emb er",
+      "e mber",
+      "fe ss",
+      "f ess",
+      "▁s pecific",
+      "▁spec ific",
+      "▁ specific",
+      "▁c ode",
+      "▁co de",
+      "▁cod e",
+      "▁ code",
+      "▁t ry",
+      "▁tr y",
+      "▁ try",
+      "ent ly",
+      "▁c ou",
+      "▁co u",
+      "▁ cou",
+      "A C",
+      "北 京",
+      "▁p ower",
+      "▁po wer",
+      "▁pow er",
+      "▁ power",
+      "ir d",
+      "i rd",
+      "▁i de",
+      "▁id e",
+      "▁ ide",
+      "\"\" \"\\",
+      "\"\"\" \\",
+      "\" \"\"\\",
+      "m p",
+      "▁v iew",
+      "▁vi ew",
+      "▁ view",
+      "能 力",
+      "方 面",
+      "um m",
+      "u mm",
+      "Qu estion",
+      "Quest ion",
+      "▁m ust",
+      "▁mus t",
+      "▁mu st",
+      "▁ must",
+      "字 第",
+      "ft ware",
+      "er n",
+      "e rn",
+      "▁re ally",
+      "▁real ly",
+      "▁ really",
+      "被告 人",
+      "en ame",
+      "ena me",
+      "e name",
+      "/* *\\",
+      "/** \\",
+      "/ **\\",
+      "le v",
+      "l ev",
+      "▁inter est",
+      "▁ interest",
+      "也 是",
+      "▁s chool",
+      "▁sch ool",
+      "▁ school",
+      "he ad",
+      "hea d",
+      "h ead",
+      "然 后",
+      "一 定",
+      "▁s l",
+      "▁ sl",
+      "▁h ist",
+      "▁his t",
+      "▁hi st",
+      "▁ hist",
+      "nW e",
+      "n We",
+      "ex port",
+      "exp ort",
+      "结 果",
+      "▁l et",
+      "▁le t",
+      "▁ let",
+      "os ed",
+      "ose d",
+      "o sed",
+      "c o",
+      "▁ap pe",
+      "▁app e",
+      "▁ appe",
+      "参 加",
+      "me t",
+      "m et",
+      "如 何",
+      "ep end",
+      "e pend",
+      "▁under stand",
+      "▁unders tand",
+      "网 络",
+      "作 为",
+      "▁p ur",
+      "▁pu r",
+      "▁ pur",
+      "▁w ater",
+      "▁wa ter",
+      "▁wat er",
+      "▁ water",
+      "pl es",
+      "ple s",
+      "p les",
+      "g o",
+      "▁w eek",
+      "▁we ek",
+      "▁ week",
+      "A P",
+      "R O",
+      "履 行",
+      "▁v ersion",
+      "▁vers ion",
+      "▁ version",
+      "环 境",
+      "的 时候",
+      "it or",
+      "ito r",
+      "i tor",
+      "纠 纷",
+      "▁r ange",
+      "▁ra nge",
+      "▁ran ge",
+      "▁ range",
+      "\\' )\\",
+      "\\') \\",
+      "\\ ')\\",
+      "ol ogical",
+      "olog ical",
+      "ologic al",
+      "o logical",
+      "▁A t",
+      "▁ At",
+      "uf f",
+      "u ff",
+      "or mal",
+      "orm al",
+      "\\' ,\\",
+      "\\', \\",
+      "\\ ',\\",
+      "▁c reate",
+      "▁cre ate",
+      "▁creat e",
+      "▁ create",
+      "pa ce",
+      "pac e",
+      "p ace",
+      "nc lass",
+      "ncl ass",
+      "n class",
+      "un c",
+      "u nc",
+      "期 间",
+      "支 持",
+      "▁f our",
+      "▁fo ur",
+      "▁ four",
+      "af f",
+      "a ff",
+      "or der",
+      "ord er",
+      "orde r",
+      "▁O n",
+      "▁ On",
+      "▁f ew",
+      "▁fe w",
+      "\") ;\\",
+      "\"); \\",
+      "\" );\\",
+      "id th",
+      "nR EF",
+      "nRE F",
+      "n REF",
+      "ur ed",
+      "ure d",
+      "u red",
+      "iv er",
+      "ive r",
+      "i ver",
+      "th on",
+      "t hon",
+      "▁b etter",
+      "▁bet ter",
+      "▁ better",
+      "▁B ut",
+      "▁Bu t",
+      "▁ But",
+      "程 序",
+      "ch an",
+      "cha n",
+      "c han",
+      "]; \\",
+      "] ;\\",
+      "▁p rote",
+      "▁pro te",
+      "▁pr ote",
+      "▁prot e",
+      "▁ prote",
+      "ul es",
+      "ule s",
+      "u les",
+      "ial ly",
+      "i ally",
+      "Co l",
+      "C ol",
+      "▁p rot",
+      "▁pro t",
+      "▁pr ot",
+      "▁ prot",
+      "▁in it",
+      "▁i nit",
+      "▁ini t",
+      "▁ init",
+      "ra w",
+      "r aw",
+      "viron ment",
+      "▁re cord",
+      "▁rec ord",
+      "▁ record",
+      "▁of fer",
+      "▁off er",
+      "▁ offer",
+      "E T",
+      "▁comp any",
+      "▁ company",
+      "表 示",
+      "▁t ext",
+      "▁te xt",
+      "▁tex t",
+      "▁ text",
+      "▁a rg",
+      "▁ar g",
+      "▁ arg",
+      "大 家",
+      "it her",
+      "ith er",
+      "ithe r",
+      "i ther",
+      "▁e vent",
+      "▁ev ent",
+      "▁even t",
+      "▁ event",
+      "组 织",
+      "▁d ise",
+      "▁dis e",
+      "▁di se",
+      "▁char acter",
+      "▁ character",
+      "这 些",
+      "了 一",
+      "ro p",
+      "r op",
+      "id d",
+      "i dd",
+      "孩 子",
+      "▁c heck",
+      "▁ch eck",
+      "▁che ck",
+      "▁ check",
+      "合 同",
+      "▁c onn",
+      "▁con n",
+      "▁co nn",
+      "▁ conn",
+      "fer ence",
+      "fe rence",
+      "et hing",
+      "eth ing",
+      "e thing",
+      "y d",
+      "大 的",
+      "▁d ev",
+      "▁de v",
+      "▁ dev",
+      "类 型",
+      "政 府",
+      "▁re f",
+      "▁r ef",
+      "▁ ref",
+      "it tle",
+      "itt le",
+      "i ttle",
+      "▁o cc",
+      "▁oc c",
+      "▁ occ",
+      "alc ul",
+      "en e",
+      "e ne",
+      "那 么",
+      "▁▁ ▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁",
+      "▁▁▁▁▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁ ▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁",
+      "▁ ▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "is k",
+      "i sk",
+      "▁f oot",
+      "▁fo ot",
+      "▁foo t",
+      "▁ foot",
+      "个 人",
+      "id eo",
+      "ide o",
+      "科 学",
+      "e b",
+      "所 有",
+      "▁t reatment",
+      "▁treat ment",
+      "▁ treatment",
+      "答 案",
+      "第 三",
+      "▁go ing",
+      "▁ going",
+      "▁s ure",
+      "▁su re",
+      "▁sur e",
+      ") $",
+      "at erial",
+      "ate rial",
+      "ater ial",
+      "▁A p",
+      "▁ Ap",
+      "开 发",
+      "选 项",
+      "▁A d",
+      "▁ Ad",
+      "hi b",
+      "h ib",
+      "at ures",
+      "ature s",
+      "atur es",
+      "atu res",
+      "a tures",
+      "lo c",
+      "l oc",
+      "aj or",
+      "ajo r",
+      "u f",
+      "▁o ld",
+      "▁ol d",
+      "▁ old",
+      "n U",
+      "▁a ccess",
+      "▁acc ess",
+      "▁ac cess",
+      "▁ access",
+      "▁n ull",
+      "▁nu ll",
+      "▁ null",
+      "▁c ost",
+      "▁co st",
+      "▁cos t",
+      "▁ cost",
+      "▁ 裁",
+      "id es",
+      "ide s",
+      "i des",
+      "一 案",
+      "裁定 如下",
+      "学 校",
+      "a i",
+      "基 本",
+      "▁t yp",
+      "▁ty p",
+      "▁ typ",
+      "A D",
+      "ur ther",
+      "urt her",
+      "▁o rig",
+      "▁or ig",
+      "▁ orig",
+      "▁l ess",
+      "▁le ss",
+      "▁les s",
+      "▁ less",
+      "▁l ow",
+      "▁lo w",
+      "▁ low",
+      "受 理",
+      "n J",
+      "对 于",
+      "angu age",
+      "▁s ay",
+      "▁sa y",
+      "▁ say",
+      "▁anal ysis",
+      "▁analy sis",
+      "▁analys is",
+      "▁ analysis",
+      "▁h app",
+      "▁ha pp",
+      "▁hap p",
+      "▁p rom",
+      "▁pro m",
+      "▁pr om",
+      "▁ prom",
+      "▁p ossible",
+      "▁poss ible",
+      "▁ possible",
+      "ib r",
+      "i br",
+      "▁_ _",
+      "▁ __",
+      "bj ect",
+      "b ject",
+      "▁k nown",
+      "▁kn own",
+      "▁know n",
+      "▁ known",
+      "作 品",
+      "▁t oo",
+      "▁to o",
+      "▁ too",
+      "▁s ize",
+      "▁si ze",
+      "▁ size",
+      "ac y",
+      "a cy",
+      "▁ex ample",
+      "▁exam ple",
+      "▁ example",
+      "ve s",
+      "v es",
+      "▁c ho",
+      "▁ch o",
+      "▁ cho",
+      "▁O r",
+      "▁ Or",
+      "p x",
+      "pl ement",
+      "ple ment",
+      "p lement",
+      "平 台",
+      "▁s ugg",
+      "▁su gg",
+      "▁sug g",
+      "▁ sugg",
+      "ou rs",
+      "our s",
+      "o urs",
+      "案 件",
+      "某 某",
+      "tr eturn",
+      "tre turn",
+      "tret urn",
+      "t return",
+      "应 用",
+      "[\\ '",
+      "[ \\'",
+      "▁l arge",
+      "▁larg e",
+      "▁lar ge",
+      "▁ large",
+      "▁S tring",
+      "▁St ring",
+      "▁Str ing",
+      "▁Stri ng",
+      "▁ String",
+      "▁p layers",
+      "▁pl ayers",
+      "▁play ers",
+      "▁player s",
+      "▁ players",
+      "▁l aw",
+      "▁la w",
+      "▁ law",
+      "▁t ot",
+      "▁to t",
+      "▁ tot",
+      "in ter",
+      "int er",
+      "i nter",
+      "▁T e",
+      "▁ Te",
+      "▁m ov",
+      "▁mo v",
+      "▁ mov",
+      "oc ial",
+      "oci al",
+      "▁a ns",
+      "▁an s",
+      "▁ ans",
+      "▁again st",
+      "um p",
+      "u mp",
+      "se rv",
+      "ser v",
+      "s erv",
+      "▁A merican",
+      "▁Americ an",
+      "▁America n",
+      "▁Amer ican",
+      "▁ American",
+      "▁invest ig",
+      "si ze",
+      "s ize",
+      "▁com mon",
+      "▁comm on",
+      "▁ common",
+      "▁g ame",
+      "▁gam e",
+      "▁ga me",
+      "▁ game",
+      "un g",
+      "u ng",
+      "cl e",
+      "c le",
+      "日出 生",
+      "日 出生",
+      "▁s ite",
+      "▁sit e",
+      "▁si te",
+      "▁ site",
+      "▁f ull",
+      "▁ful l",
+      "▁fu ll",
+      "▁ full",
+      "um e",
+      "u me",
+      "io d",
+      "i od",
+      ") :",
+      "▁S he",
+      "▁Sh e",
+      "▁ She",
+      "▁S ystem",
+      "▁Sy stem",
+      "▁Sys tem",
+      "▁ System",
+      "▁e s",
+      "▁ es",
+      "▁ex am",
+      "▁ exam",
+      "处 理",
+      "ass ert",
+      "asse rt",
+      "asser t",
+      "非 常",
+      "is ed",
+      "ise d",
+      "i sed",
+      "al u",
+      "a lu",
+      "▁B r",
+      "▁ Br",
+      "▁M ar",
+      "▁Ma r",
+      "▁ Mar",
+      "▁f ac",
+      "▁fa c",
+      "▁ fac",
+      "单 位",
+      "im al",
+      "ima l",
+      "i mal",
+      "▁child ren",
+      "▁ children",
+      "科 技",
+      "I L",
+      "er ing",
+      "eri ng",
+      "e ring",
+      "▁g l",
+      "▁ gl",
+      "▁P r",
+      "▁ Pr",
+      "▁s uccess",
+      "▁su ccess",
+      "▁suc cess",
+      "▁succ ess",
+      "▁ success",
+      "▁v ol",
+      "▁vo l",
+      "▁ vol",
+      "▁T hey",
+      "▁The y",
+      "▁Th ey",
+      "▁ They",
+      "▁s ingle",
+      "▁sing le",
+      "▁ single",
+      "▁s pe",
+      "▁sp e",
+      "▁ spe",
+      "uc tion",
+      "uct ion",
+      "u ction",
+      "▁re quest",
+      "▁requ est",
+      "▁ request",
+      "our ces",
+      "ource s",
+      "具 有",
+      "al es",
+      "ale s",
+      "a les",
+      "ct or",
+      "c tor",
+      "th ing",
+      "thi ng",
+      "thin g",
+      "t hing",
+      "en tal",
+      "ent al",
+      "enta l",
+      "▁m ean",
+      "▁me an",
+      "▁ mean",
+      "这 种",
+      "▁p lan",
+      "▁pl an",
+      "▁ plan",
+      "▁e ng",
+      "▁en g",
+      "▁ eng",
+      "喜 欢",
+      "▁E ng",
+      "▁En g",
+      "▁ Eng",
+      "▁pro pos",
+      "▁prop os",
+      "▁The se",
+      "▁Th ese",
+      "▁ These",
+      "de s",
+      "d es",
+      "ua ry",
+      "u ary",
+      "▁a nother",
+      "▁an other",
+      "▁ another",
+      "at ors",
+      "ator s",
+      "ato rs",
+      "pon se",
+      "po nse",
+      "pons e",
+      "上 海",
+      "in ation",
+      "ina tion",
+      "i nation",
+      "E D",
+      "in c",
+      "i nc",
+      ") ;",
+      "\", \"",
+      "\" ,\"",
+      "wi t",
+      "w it",
+      "▁qu estion",
+      "▁quest ion",
+      "▁ question",
+      "n y",
+      "▁r ad",
+      "▁ra d",
+      "▁ rad",
+      "▁l inks",
+      "▁link s",
+      "▁lin ks",
+      "▁ links",
+      "p i",
+      "上 的",
+      "还 有",
+      "mi t",
+      "m it",
+      "Vi ew",
+      "V iew",
+      "▁p ut",
+      "▁pu t",
+      "▁ put",
+      "at ory",
+      "ator y",
+      "ato ry",
+      "到 了",
+      "朋 友",
+      "▁a ble",
+      "▁ab le",
+      "▁ able",
+      "▁s omething",
+      "▁some thing",
+      "▁som ething",
+      "▁ something",
+      "▁sugg est",
+      "▁sug gest",
+      "成 为",
+      "▁l ight",
+      "▁li ght",
+      "▁lig ht",
+      "▁ light",
+      "▁p hot",
+      "▁ph ot",
+      "▁ phot",
+      "▁a ff",
+      "▁af f",
+      "▁ aff",
+      "▁mark et",
+      "▁mar ket",
+      "▁ market",
+      "▁in clude",
+      "▁includ e",
+      "▁incl ude",
+      "▁ include",
+      "nI t",
+      "n It",
+      "▁ u",
+      "▁exper ience",
+      "▁experi ence",
+      "▁ experience",
+      "▁k eep",
+      "▁ke ep",
+      "▁ keep",
+      "▁pr act",
+      "▁pra ct",
+      "▁ pract",
+      "en ded",
+      "end ed",
+      "ende d",
+      "le y",
+      "l ey",
+      "Err or",
+      "Er ror",
+      "E rror",
+      "mb er",
+      "m ber",
+      "▁s ervices",
+      "▁serv ices",
+      "▁service s",
+      "▁ services",
+      "▁g over",
+      "▁go ver",
+      "▁gov er",
+      "A r",
+      "> <",
+      "ag n",
+      "a gn",
+      "功 能",
+      "ab les",
+      "able s",
+      "abl es",
+      "a bles",
+      "▁p age",
+      "▁pa ge",
+      "▁pag e",
+      "▁ page",
+      "州 市",
+      "\"; \\",
+      "\" ;\\",
+      "▁g ive",
+      "▁giv e",
+      "▁gi ve",
+      "▁ give",
+      "不 会",
+      "pr ivate",
+      "priv ate",
+      "p rivate",
+      "即 可",
+      "ri x",
+      "r ix",
+      "Te st",
+      "T est",
+      "基 础",
+      "A B",
+      "▁al ways",
+      "▁ always",
+      "an ks",
+      "ank s",
+      "▁stud ies",
+      "▁e t",
+      "▁ et",
+      "觉 得",
+      "or ies",
+      "ori es",
+      "orie s",
+      "o ries",
+      "▁w he",
+      "▁wh e",
+      "ri es",
+      "rie s",
+      "r ies",
+      "ar r",
+      "a rr",
+      "出 现",
+      "it er",
+      "ite r",
+      "i ter",
+      "iv ity",
+      "ivi ty",
+      "▁al low",
+      "▁all ow",
+      "▁ allow",
+      "il ar",
+      "ila r",
+      "i lar",
+      "▁d ays",
+      "▁day s",
+      "▁da ys",
+      "▁ days",
+      "M A",
+      "▁C an",
+      "▁Ca n",
+      "▁ Can",
+      "▁m ight",
+      "▁mi ght",
+      "▁ might",
+      "▁l eft",
+      "▁le ft",
+      "▁ left",
+      "▁work ing",
+      "▁wor king",
+      "▁ working",
+      "▁in vol",
+      "▁inv ol",
+      "责 任",
+      "▁inst all",
+      "▁ins tall",
+      "▁ install",
+      "an ds",
+      "and s",
+      "om et",
+      "ome t",
+      "o met",
+      "ti es",
+      "tie s",
+      "t ies",
+      "nI f",
+      "n If",
+      "nu sing",
+      "nus ing",
+      "n using",
+      "终 结",
+      "ou ght",
+      "ough t",
+      "oug ht",
+      "o ught",
+      "▁e l",
+      "▁ el",
+      "▁h uman",
+      "▁hum an",
+      "▁hu man",
+      "▁ human",
+      "us ed",
+      "use d",
+      "u sed",
+      "bu t",
+      "b ut",
+      "▁C ent",
+      "▁Ce nt",
+      "▁ Cent",
+      "di ct",
+      "dic t",
+      "d ict",
+      "▁f alse",
+      "▁fals e",
+      "▁fal se",
+      "▁ false",
+      "▁p rob",
+      "▁pro b",
+      "▁pr ob",
+      "▁ prob",
+      "ou ble",
+      "o uble",
+      "▁C ount",
+      "▁Co unt",
+      "▁Cou nt",
+      "▁ Count",
+      "▁ch anges",
+      "▁chang es",
+      "▁change s",
+      "▁chan ges",
+      "▁ changes",
+      "an dom",
+      "and om",
+      "ando m",
+      "▁l a",
+      "▁ la",
+      "▁i d",
+      "▁ id",
+      "H E",
+      "▁res ponse",
+      "▁respon se",
+      "▁respons e",
+      "▁ response",
+      "gr ound",
+      "gro und",
+      "g round",
+      "T o",
+      "ge n",
+      "g en",
+      "U R",
+      "( $",
+      "' \\",
+      "▁e rror",
+      "▁err or",
+      "▁er ror",
+      "▁ error",
+      "成 立",
+      "iv idual",
+      "ivid ual",
+      "美 国",
+      "▁qu ality",
+      "▁q uality",
+      "▁qual ity",
+      "▁ quality",
+      "ame space",
+      "ames pace",
+      "} ,",
+      "m e",
+      "▁r ole",
+      "▁ro le",
+      "▁ role",
+      "人 员",
+      "两 个",
+      "ri pt",
+      "rip t",
+      "r ipt",
+      "▁B l",
+      "▁ Bl",
+      "▁spec ial",
+      "▁ special",
+      "直 接",
+      "▁b ro",
+      "▁br o",
+      "▁ bro",
+      "re c",
+      "r ec",
+      "其 中",
+      "国 际",
+      "], \\",
+      "] ,\\",
+      "▁f inal",
+      "▁fin al",
+      "▁fi nal",
+      "▁ final",
+      "con fig",
+      "co nfig",
+      "conf ig",
+      "ic i",
+      "i ci",
+      "or ed",
+      "ore d",
+      "o red",
+      "纠纷 一案",
+      "mo n",
+      "m on",
+      "▁* \\",
+      "▁ *\\",
+      "ur ation",
+      "ura tion",
+      "u ration",
+      "申请 人",
+      "▁M ay",
+      "▁Ma y",
+      "▁ May",
+      "ef f",
+      "e ff",
+      "li ng",
+      "lin g",
+      "l ing",
+      "er tain",
+      "ert ain",
+      "erta in",
+      "▁a nn",
+      "▁an n",
+      "▁ ann",
+      "条 件",
+      "▁me chan",
+      "in f",
+      "i nf",
+      "▁t arget",
+      "▁tar get",
+      "▁ target",
+      "▁de scrib",
+      "▁des crib",
+      "▁desc rib",
+      "ip le",
+      "ipl e",
+      "i ple",
+      "am b",
+      "a mb",
+      "一 些",
+      "wit h",
+      "wi th",
+      "w ith",
+      "▁s everal",
+      "▁sever al",
+      "送 达",
+      "▁p aper",
+      "▁pa per",
+      "▁pap er",
+      "▁ paper",
+      "d b",
+      "▁G r",
+      "▁ Gr",
+      "▁b ig",
+      "▁bi g",
+      "▁ big",
+      "▁c lin",
+      "▁cl in",
+      "▁cli n",
+      "▁s eries",
+      "▁se ries",
+      "▁ser ies",
+      "▁ series",
+      "你 的",
+      "um n",
+      "u mn",
+      "▁l ittle",
+      "▁li ttle",
+      "▁lit tle",
+      "▁ little",
+      "▁a ction",
+      "▁act ion",
+      "▁ac tion",
+      "▁ action",
+      "▁o bt",
+      "▁ob t",
+      "日 起",
+      "▁How ever",
+      "▁ However",
+      "ap t",
+      "a pt",
+      "简 介",
+      "城 市",
+      "合 作",
+      "ut ure",
+      "u ture",
+      "事 实",
+      "▁t er",
+      "▁te r",
+      "▁ ter",
+      "▁re view",
+      "▁rev iew",
+      "▁ review",
+      "H O",
+      "re ct",
+      "rec t",
+      "r ect",
+      "▁model s",
+      "▁mod els",
+      "▁mode ls",
+      "▁ models",
+      "▁l ove",
+      "▁lo ve",
+      "▁ love",
+      "▁Un ited",
+      "▁Unit ed",
+      "▁Uni ted",
+      "▁ United",
+      "提 高",
+      "id s",
+      "i ds",
+      "▁m ax",
+      "▁ma x",
+      "▁ max",
+      "ap s",
+      "a ps",
+      "▁in c",
+      "▁i nc",
+      "▁ inc",
+      "▁h ard",
+      "▁har d",
+      "▁ha rd",
+      "▁ hard",
+      "'; \\",
+      "' ;\\",
+      "id ence",
+      "ide nce",
+      "iden ce",
+      "or ding",
+      "ord ing",
+      "▁U S",
+      "▁ US",
+      "▁pot ential",
+      "▁potent ial",
+      "▁ potential",
+      "ar gs",
+      "arg s",
+      "▁w omen",
+      "▁wo men",
+      "▁ women",
+      "{ {",
+      "▁re st",
+      "▁r est",
+      "▁res t",
+      "▁ rest",
+      "fi eld",
+      "f ield",
+      "我 的",
+      "知 识",
+      "en ding",
+      "end ing",
+      "\": \"",
+      "\" :\"",
+      "op h",
+      "o ph",
+      "▁sh ort",
+      "▁ short",
+      "f e",
+      "ab ly",
+      "abl y",
+      "a bly",
+      "▁est ab",
+      "▁es tab",
+      "▁esta b",
+      "▁ estab",
+      "M P",
+      "范 围",
+      "m y",
+      "投 资",
+      "▁f oc",
+      "▁fo c",
+      "▁S up",
+      "▁Su p",
+      "▁ Sup",
+      "▁pro fess",
+      "▁prof ess",
+      "an ces",
+      "ance s",
+      "anc es",
+      "▁* *",
+      "▁ **",
+      "}\\\\ )",
+      "} \\\\)",
+      "▁m ajor",
+      "▁maj or",
+      "▁ major",
+      "▁e ither",
+      "▁ either",
+      "▁c a",
+      "▁ ca",
+      "▁int egr",
+      "▁ integr",
+      "ra ft",
+      "raf t",
+      "法 定",
+      "▁b o",
+      "▁ bo",
+      "▁p ay",
+      "▁pa y",
+      "▁ pay",
+      "ap i",
+      "a pi",
+      "▁requ ired",
+      "▁require d",
+      "▁ required",
+      "s y",
+      "▁b ir",
+      "▁bi r",
+      "▁ bir",
+      "▁e xc",
+      "▁ex c",
+      "▁ exc",
+      "▁s elect",
+      "▁se lect",
+      "▁sel ect",
+      "▁ select",
+      "n n",
+      "▁se ason",
+      "▁sea son",
+      "▁seas on",
+      "▁ season",
+      "if t",
+      "i ft",
+      "t t",
+      "▁c ard",
+      "▁car d",
+      "▁ca rd",
+      "▁ card",
+      "s l",
+      "▁a long",
+      "▁al ong",
+      "▁ along",
+      "en tion",
+      "ent ion",
+      "enti on",
+      "一 般",
+      "ti f",
+      "t if",
+      "▁ev alu",
+      "▁eval u",
+      "▁ evalu",
+      "▁re present",
+      "▁rep resent",
+      "▁ represent",
+      "▁B e",
+      "▁ Be",
+      "ri st",
+      "ris t",
+      "r ist",
+      "▁g ot",
+      "▁go t",
+      "▁ got",
+      "▁s pace",
+      "▁sp ace",
+      "▁spac e",
+      "▁spa ce",
+      "▁ space",
+      "// /",
+      "/ //",
+      "来 的",
+      "u x",
+      "pl ate",
+      "plat e",
+      "p late",
+      "目 前",
+      "▁ 判",
+      "最 高",
+      "适 用",
+      "nt er",
+      "nte r",
+      "n ter",
+      "▁de fault",
+      "▁def ault",
+      "▁ default",
+      "治 疗",
+      "▁} \\",
+      "▁ }\\",
+      "借 款",
+      "▁c ome",
+      "▁com e",
+      "▁co me",
+      "▁ come",
+      "▁a ge",
+      "▁ag e",
+      "▁ age",
+      "▁ph ys",
+      "▁phy s",
+      "▁ phys",
+      "▁st ep",
+      "▁ste p",
+      "▁ step",
+      "ens ion",
+      "▁t otal",
+      "▁to tal",
+      "▁tot al",
+      "▁ total",
+      "小 说",
+      "利 用",
+      "iv al",
+      "iva l",
+      "i val",
+      "▁f ri",
+      "▁fr i",
+      "▁n etwork",
+      "▁net work",
+      "▁ network",
+      "▁b en",
+      "▁be n",
+      "▁ ben",
+      "▁ 事",
+      "an ag",
+      "ana g",
+      "a nag",
+      "质 量",
+      "书记 员",
+      "为 了",
+      "▁th ings",
+      "▁thing s",
+      "▁thin gs",
+      "▁ things",
+      "bo ok",
+      "b ook",
+      "} )",
+      "最 后",
+      "▁o rg",
+      "▁or g",
+      "▁ org",
+      "委 托",
+      "▁we bs",
+      "▁web s",
+      "▁ webs",
+      "标 准",
+      "▁ind ividual",
+      "▁ individual",
+      "pon ent",
+      "▁c alled",
+      "▁call ed",
+      "▁cal led",
+      "▁ called",
+      "▁un til",
+      "▁u ntil",
+      "▁unt il",
+      "▁ until",
+      "cri pt",
+      "cr ipt",
+      "c ript",
+      "ri ct",
+      "ric t",
+      "r ict",
+      "▁d ue",
+      "▁du e",
+      "▁ due",
+      "▁st atic",
+      "▁stat ic",
+      "▁ static",
+      "re ady",
+      "read y",
+      "rea dy",
+      "os is",
+      "osi s",
+      "o sis",
+      "le g",
+      "l eg",
+      "设 备",
+      "}; \\",
+      "} ;\\",
+      "▁d er",
+      "▁de r",
+      "▁ der",
+      "▁stud ents",
+      "▁student s",
+      "▁ students",
+      "▁t imes",
+      "▁time s",
+      "▁tim es",
+      "▁ti mes",
+      "▁ times",
+      "介 绍",
+      "co unt",
+      "cou nt",
+      "c ount",
+      "二 十",
+      "st ate",
+      "stat e",
+      "sta te",
+      "O S",
+      "机 关",
+      "O T",
+      "▁t urn",
+      "▁tur n",
+      "▁tu rn",
+      "▁ turn",
+      "▁cont ent",
+      "▁ content",
+      "▁l ater",
+      "▁la ter",
+      "▁late r",
+      "▁lat er",
+      "▁ later",
+      "pl y",
+      "p ly",
+      "作 用",
+      "资 源",
+      "il lion",
+      "ill ion",
+      "illi on",
+      "实 现",
+      "}) ;\\",
+      "}); \\",
+      "} );\\",
+      "of f",
+      "o ff",
+      "}^ {",
+      "} ^{",
+      "▁l eg",
+      "▁le g",
+      "▁ leg",
+      "▁fe el",
+      "▁fee l",
+      "nex port",
+      "n export",
+      "\\' )",
+      "\\ ')",
+      "\\' :",
+      "\\ ':",
+      "ie ve",
+      "iev e",
+      "i eve",
+      "cu lar",
+      "c ular",
+      "律 师",
+      "定 的",
+      "▁n ear",
+      "▁ne ar",
+      "▁ near",
+      "og n",
+      "o gn",
+      "file name",
+      "fil ename",
+      "fi lename",
+      "▁s y",
+      "▁ sy",
+      "▁p ub",
+      "▁pu b",
+      "▁ pub",
+      "▁C ar",
+      "▁Ca r",
+      "▁ Car",
+      "包 括",
+      "▁el se",
+      "▁ else",
+      "▁com plet",
+      "▁comp let",
+      "▁compl et",
+      "存 在",
+      "par am",
+      "pa ram",
+      "para m",
+      "p aram",
+      "ab or",
+      "abo r",
+      "a bor",
+      "co de",
+      "cod e",
+      "c ode",
+      "gi n",
+      "g in",
+      "▁com put",
+      "▁comp ut",
+      "▁ comput",
+      "Pro blem",
+      "Prob lem",
+      "js on",
+      "j son",
+      "▁r ate",
+      "▁rat e",
+      "▁ra te",
+      "▁ rate",
+      "▁d isease",
+      "▁dise ase",
+      "▁http s",
+      "▁ https",
+      "经 营",
+      "▁s ide",
+      "▁si de",
+      "▁sid e",
+      "▁ side",
+      "行 政",
+      "IN G",
+      "I NG",
+      "而 且",
+      "Fil e",
+      "Fi le",
+      "F ile",
+      "ro id",
+      "r oid",
+      "si n",
+      "s in",
+      "NA ME",
+      "N AME",
+      "▁e nerg",
+      "▁en erg",
+      "Mod el",
+      "Mode l",
+      "Mo del",
+      "M odel",
+      "est ions",
+      "estion s",
+      "▁C ont",
+      "▁Con t",
+      "▁Co nt",
+      "▁ Cont",
+      "起 诉",
+      "连 载",
+      "▁per iod",
+      "▁peri od",
+      "▁ period",
+      "ui t",
+      "u it",
+      "▁a sk",
+      "▁as k",
+      "▁ ask",
+      "▁t erm",
+      "▁te rm",
+      "▁ter m",
+      "▁ term",
+      "▁l ot",
+      "▁lo t",
+      "▁ lot",
+      "ve rage",
+      "ver age",
+      "… …",
+      "▁N o",
+      "▁ No",
+      "m b",
+      "Ti tle",
+      "T itle",
+      "▁m icro",
+      "▁mic ro",
+      "▁mi cro",
+      "▁ micro",
+      "$ \\",
+      "lo bal",
+      "lob al",
+      "ess ion",
+      "rib ution",
+      "ribut ion",
+      "▁t ool",
+      "▁to ol",
+      "▁too l",
+      "▁ tool",
+      "in put",
+      "inp ut",
+      "c d",
+      "▁in dic",
+      "▁ind ic",
+      "解 决",
+      "▁o nline",
+      "▁on line",
+      "▁ online",
+      "▁me et",
+      "▁ meet",
+      "▁e nc",
+      "▁en c",
+      "▁ enc",
+      "str uction",
+      "struct ion",
+      "五 十",
+      "▁gener al",
+      "▁gen eral",
+      "▁gene ral",
+      "▁genera l",
+      "▁ general",
+      "m m",
+      "符 合",
+      "on ey",
+      "one y",
+      "o ney",
+      "一 次",
+      "func tion",
+      "fun ction",
+      "f unction",
+      "ward s",
+      "war ds",
+      "w ards",
+      "效 力",
+      "S c",
+      "col or",
+      "co lor",
+      "c olor",
+      "vo id",
+      "v oid",
+      "▁p ain",
+      "▁pa in",
+      "▁ pain",
+      "▁stand ard",
+      "▁ standard",
+      "应 该",
+      "ay er",
+      "aye r",
+      "a yer",
+      "og le",
+      "教 学",
+      "▁value s",
+      "▁val ues",
+      "▁valu es",
+      "▁ values",
+      "▁com plex",
+      "▁comp lex",
+      "▁compl ex",
+      "▁ complex",
+      "ar ning",
+      "arn ing",
+      "ti tle",
+      "tit le",
+      "t itle",
+      "之 规定",
+      "he l",
+      "h el",
+      "cont ent",
+      "ur ity",
+      "uri ty",
+      "bu m",
+      "b um",
+      "ma x",
+      "m ax",
+      "▁im age",
+      "▁imag e",
+      "▁ image",
+      "本院 认为",
+      "n o",
+      "控 制",
+      "a e",
+      "▁ex ist",
+      "▁ exist",
+      "ting s",
+      "tin gs",
+      "t ings",
+      "ograph y",
+      "ograp hy",
+      "pe c",
+      "p ec",
+      "▁acc ount",
+      "▁ac count",
+      "▁ account",
+      "u k",
+      "ncon st",
+      "nco nst",
+      "n const",
+      "▁comm unity",
+      "▁commun ity",
+      "▁ community",
+      "m d",
+      "▁disc uss",
+      "本 裁定",
+      "面 积",
+      "on es",
+      "one s",
+      "o nes",
+      "▁com pet",
+      "▁comp et",
+      "▁ compet",
+      "行 业",
+      "nd er",
+      "nde r",
+      "n der",
+      "▁en vironment",
+      "▁ environment",
+      "ma p",
+      "m ap",
+      "代理 人",
+      "第二 百",
+      "第 二百",
+      "▁re spect",
+      "▁res pect",
+      "▁resp ect",
+      "▁ respect",
+      "民 初",
+      "▁p rim",
+      "▁pr im",
+      "▁pri m",
+      "▁ prim",
+      "▁f ootball",
+      "▁foot ball",
+      "▁ football",
+      "▁b ody",
+      "▁bo dy",
+      "▁bod y",
+      "▁ body",
+      "▁ass ociated",
+      "▁associ ated",
+      "▁associate d",
+      "▁ associated",
+      "▁m en",
+      "▁me n",
+      "▁ men",
+      "is ing",
+      "isi ng",
+      "isin g",
+      "i sing",
+      "ob er",
+      "obe r",
+      "o ber",
+      "之 后",
+      "U S",
+      "材 料",
+      "▁A N",
+      "▁ AN",
+      "▁O ur",
+      "▁ Our",
+      "▁t re",
+      "▁tr e",
+      "▁ tre",
+      "▁ab ove",
+      "▁ above",
+      "函 数",
+      "]) \\",
+      "] )\\",
+      "▁eff ects",
+      "▁effect s",
+      "▁ effects",
+      "▁s ym",
+      "▁sy m",
+      "▁ sym",
+      "sid e",
+      "si de",
+      "s ide",
+      "能 够",
+      "▁act ivity",
+      "▁activ ity",
+      "▁ activity",
+      "ne y",
+      "n ey",
+      "▁e p",
+      "▁ ep",
+      "▁r isk",
+      "▁ris k",
+      "▁ri sk",
+      "▁ risk",
+      "ar ing",
+      "ari ng",
+      "arin g",
+      "a ring",
+      "▁m ar",
+      "▁ma r",
+      "▁ mar",
+      "▁s olution",
+      "▁sol ution",
+      "▁ solution",
+      ") /",
+      "▁m aterial",
+      "▁mat erial",
+      "▁mate rial",
+      "▁mater ial",
+      "▁ material",
+      "▁s ens",
+      "▁se ns",
+      "▁sen s",
+      "▁ sens",
+      "▁I S",
+      "▁ IS",
+      "us ion",
+      "▁part icular",
+      "▁partic ular",
+      "er ror",
+      "err or",
+      "e rror",
+      "▁a ir",
+      "▁ai r",
+      "▁ air",
+      "品 牌",
+      "▁e arly",
+      "▁ear ly",
+      "▁ early",
+      "nm ent",
+      "nme nt",
+      "n ment",
+      "▁per formance",
+      "▁perform ance",
+      "▁ performance",
+      "li b",
+      "l ib",
+      "真 的",
+      "ma t",
+      "m at",
+      "▁l ink",
+      "▁li nk",
+      "▁lin k",
+      "▁ link",
+      "▁a mong",
+      "▁am ong",
+      "Pa r",
+      "P ar",
+      "全 国",
+      "om s",
+      "o ms",
+      "fr om",
+      "fro m",
+      "f rom",
+      "ib ility",
+      "出版 社",
+      "▁iss ue",
+      "▁ issue",
+      "▁c rit",
+      "▁cr it",
+      "bo x",
+      "b ox",
+      "z e",
+      "的 规定",
+      "▁w ord",
+      "▁wor d",
+      "▁wo rd",
+      "▁ word",
+      "完 成",
+      "ol ic",
+      "oli c",
+      "o lic",
+      "▁d at",
+      "▁da t",
+      "▁ dat",
+      "**** ****",
+      "in ary",
+      "ina ry",
+      "inar y",
+      "▁e nergy",
+      "▁en ergy",
+      "▁energ y",
+      "▁ energy",
+      "正 确",
+      "er o",
+      "e ro",
+      "▁b ase",
+      "▁bas e",
+      "▁ba se",
+      "▁ base",
+      "▁p ort",
+      "▁po rt",
+      "▁por t",
+      "▁ port",
+      "▁m ove",
+      "▁mov e",
+      "▁mo ve",
+      "▁ move",
+      "il t",
+      "i lt",
+      "un e",
+      "u ne",
+      "▁pre vious",
+      "▁prev ious",
+      "▁ previous",
+      "▁l ooking",
+      "▁look ing",
+      "▁lo oking",
+      "▁ looking",
+      "as ing",
+      "asi ng",
+      "asin g",
+      "a sing",
+      "en der",
+      "end er",
+      "ende r",
+      "e nder",
+      "od es",
+      "ode s",
+      "o des",
+      "▁W ith",
+      "▁Wi th",
+      "▁Wit h",
+      "▁ With",
+      "▁J an",
+      "▁Ja n",
+      "▁ Jan",
+      "▁e duc",
+      "▁ed uc",
+      "▁edu c",
+      "▁ educ",
+      "▁s imilar",
+      "▁sim ilar",
+      "▁ similar",
+      "▁be low",
+      "▁bel ow",
+      "▁ below",
+      "Se t",
+      "S et",
+      "Val ue",
+      "V alue",
+      "(\\ \\",
+      "( \\\\",
+      "▁rel ations",
+      "▁relation s",
+      "▁ relations",
+      "pl ied",
+      "▁re spons",
+      "▁res pons",
+      "▁respon s",
+      "▁resp ons",
+      "▁ respons",
+      "一 起",
+      "▁m ot",
+      "▁mo t",
+      "▁ mot",
+      "in st",
+      "ins t",
+      "i nst",
+      "▁w hy",
+      "▁wh y",
+      "▁ why",
+      "其 实",
+      "▁c ourse",
+      "▁co urse",
+      "▁cour se",
+      "▁ course",
+      "ic ient",
+      "ici ent",
+      "ro ller",
+      "rol ler",
+      "roll er",
+      "▁f ix",
+      "▁fi x",
+      "▁ fix",
+      "or es",
+      "ore s",
+      "o res",
+      "▁C ounty",
+      "▁Count y",
+      "▁Cou nty",
+      "co s",
+      "c os",
+      "miss ion",
+      "m ission",
+      "▁c reat",
+      "▁cre at",
+      "▁cr eat",
+      "▁ creat",
+      "价 格",
+      "ip p",
+      "i pp",
+      "▁appro ach",
+      "▁produ cts",
+      "▁product s",
+      "▁ products",
+      "▁m aking",
+      "▁ma king",
+      "▁ making",
+      "Serv ice",
+      "S ervice",
+      "to p",
+      "t op",
+      "代表 人",
+      "四 十",
+      "I m",
+      "d a",
+      "nEx ternal",
+      "nExt ernal",
+      "n External",
+      "不 要",
+      "委 员",
+      "f l",
+      "ight s",
+      "igh ts",
+      "▁system s",
+      "▁syst ems",
+      "▁ systems",
+      "支 付",
+      "▁t oday",
+      "▁to day",
+      "▁ today",
+      "▁f lo",
+      "▁fl o",
+      "▁ flo",
+      "▁so ftware",
+      "▁soft ware",
+      "▁ software",
+      "▁me ans",
+      "▁mean s",
+      "▁ means",
+      "ro du",
+      "rod u",
+      "▁ind ust",
+      "▁indu st",
+      "▁ indust",
+      "建 议",
+      "ma in",
+      "m ain",
+      "▁con fig",
+      "▁co nfig",
+      "▁conf ig",
+      "▁ config",
+      "▁d om",
+      "▁do m",
+      "▁ dom",
+      "▁o bserv",
+      "▁ob serv",
+      "▁obs erv",
+      "▁ observ",
+      "C O",
+      "▁n eeds",
+      "▁ne eds",
+      "▁need s",
+      "▁ needs",
+      "▁s ource",
+      "▁sour ce",
+      "▁ source",
+      "Fi eld",
+      "F ield",
+      "▁im prove",
+      "▁impro ve",
+      "▁imp rove",
+      "▁improv e",
+      "tr a",
+      "t ra",
+      "有 效",
+      "Re qu",
+      "Req u",
+      "R equ",
+      "地 区",
+      "im um",
+      "▁h aving",
+      "▁ha ving",
+      "▁hav ing",
+      "▁l o",
+      "▁ lo",
+      "▁ —",
+      "go r",
+      "g or",
+      "过程 中",
+      "建 筑",
+      "pa ss",
+      "pas s",
+      "p ass",
+      "ok en",
+      "oke n",
+      "o ken",
+      "S h",
+      "ph a",
+      "p ha",
+      "la bel",
+      "lab el",
+      "l abel",
+      "▁al ready",
+      "▁t her",
+      "▁the r",
+      "▁th er",
+      "▁ ther",
+      "ic a",
+      "i ca",
+      "▁n ever",
+      "▁ne ver",
+      "▁ never",
+      "▁add ress",
+      "▁addr ess",
+      "▁ address",
+      "tr ans",
+      "tra ns",
+      "t rans",
+      "证 据",
+      "I P",
+      "pr e",
+      "p re",
+      "▁d one",
+      "▁do ne",
+      "▁don e",
+      "▁ done",
+      "▁le vels",
+      "▁level s",
+      "▁lev els",
+      "▁ levels",
+      "op er",
+      "ope r",
+      "o per",
+      "n b",
+      "is tic",
+      "ist ic",
+      "ut ions",
+      "ution s",
+      "uti ons",
+      "ind ows",
+      "indow s",
+      "▁j ob",
+      "▁jo b",
+      "▁ job",
+      "▁m embers",
+      "▁mem bers",
+      "▁member s",
+      "▁ members",
+      "▁o ffic",
+      "▁off ic",
+      "In t",
+      "I nt",
+      "▁n ec",
+      "▁ne c",
+      "▁ nec",
+      "职 业",
+      "▁w hether",
+      "▁wh ether",
+      "▁whe ther",
+      "ag ue",
+      "agu e",
+      "a gue",
+      "res ult",
+      "rib ute",
+      "ribut e",
+      "▁J oh",
+      "▁Jo h",
+      "▁ Joh",
+      "ho ld",
+      "hol d",
+      "h old",
+      "di r",
+      "d ir",
+      "in o",
+      "i no",
+      "nY ou",
+      "n You",
+      "▁A ss",
+      "▁As s",
+      "▁ Ass",
+      "一 下",
+      "采 用",
+      "一 直",
+      "▁a nswer",
+      "▁ans wer",
+      "▁ answer",
+      "▁c irc",
+      "▁cir c",
+      "▁ci rc",
+      "▁ circ",
+      "▁be hav",
+      "▁beh av",
+      "rib uted",
+      "ribut ed",
+      "ribute d",
+      "▁to get",
+      "ph p",
+      "p hp",
+      "ia tion",
+      "iat ion",
+      "i ation",
+      "▁t ogether",
+      "▁toget her",
+      "▁ together",
+      "▁d ocument",
+      "▁doc ument",
+      "▁ document",
+      "▁de ath",
+      "▁ death",
+      "Ex ception",
+      "一 样",
+      "port s",
+      "por ts",
+      "p orts",
+      "表 现",
+      "▁ex pression",
+      "▁exp ression",
+      "▁express ion",
+      "▁expr ession",
+      "▁ expression",
+      "▁em ploy",
+      "▁emp loy",
+      "操 作",
+      "▁l ik",
+      "▁li k",
+      "▁ lik",
+      "▁group s",
+      "▁gro ups",
+      "▁ groups",
+      "Te xt",
+      "Tex t",
+      "T ext",
+      "行 为",
+      "▁l east",
+      "▁le ast",
+      "▁ least",
+      "▁p ress",
+      "▁pre ss",
+      "▁pr ess",
+      "▁pres s",
+      "▁ press",
+      "▁v ideo",
+      "▁vide o",
+      "▁vid eo",
+      "▁ video",
+      "▁P ar",
+      "▁Pa r",
+      "▁ Par",
+      "▁w in",
+      "▁wi n",
+      "▁ win",
+      "al f",
+      "a lf",
+      "D E",
+      "il ls",
+      "ill s",
+      "▁– \\",
+      "▁ –\\",
+      "▁ap plication",
+      "▁applic ation",
+      "▁appl ication",
+      "▁ application",
+      "art ment",
+      "结 构",
+      "进行 了",
+      "进 行了",
+      "▁N ational",
+      "▁Nation al",
+      "▁Nat ional",
+      "▁ National",
+      "base d",
+      "ba sed",
+      "bas ed",
+      "b ased",
+      "th ers",
+      "ther s",
+      "the rs",
+      "▁c opy",
+      "▁co py",
+      "▁cop y",
+      "▁ copy",
+      "▁I m",
+      "▁ Im",
+      "▁v arious",
+      "▁vari ous",
+      "▁var ious",
+      "ear ch",
+      "e arch",
+      "▁exper im",
+      "▁experi m",
+      "▁fe atures",
+      "▁feature s",
+      "▁feat ures",
+      "▁ features",
+      "at ter",
+      "att er",
+      "▁o ften",
+      "▁of ten",
+      "▁ often",
+      "各 种",
+      "▁we bsite",
+      "▁webs ite",
+      "▁web site",
+      "▁ website",
+      "▁St ates",
+      "▁State s",
+      "▁Stat es",
+      "▁Sta tes",
+      "▁ States",
+      "ain er",
+      "ai ner",
+      "aine r",
+      "a iner",
+      "▁f ail",
+      "▁fa il",
+      "▁ fail",
+      "pl ace",
+      "plac e",
+      "Us er",
+      "Use r",
+      "U ser",
+      "▁par amet",
+      "▁para met",
+      "▁param et",
+      "▁ paramet",
+      "▁high er",
+      "▁s equ",
+      "▁se qu",
+      "▁seq u",
+      "▁ sequ",
+      "ar row",
+      "arr ow",
+      "▁s ocial",
+      "▁soc ial",
+      "▁soci al",
+      "▁ social",
+      "之 间",
+      "▁v ir",
+      "▁vi r",
+      "▁ vir",
+      "Ob ject",
+      "Obj ect",
+      "O bject",
+      "▁/ /",
+      "▁ //",
+      "oo r",
+      "o or",
+      "了 解",
+      "获 得",
+      "lo r",
+      "l or",
+      "/ *",
+      "决 定",
+      "ou nc",
+      "oun c",
+      "o unc",
+      "Ke y",
+      "K ey",
+      "▁t rad",
+      "▁tr ad",
+      "▁tra d",
+      "▁ trad",
+      "▁c ur",
+      "▁cu r",
+      "▁ cur",
+      "▁re lated",
+      "▁rel ated",
+      "▁relate d",
+      "▁ related",
+      "o o",
+      "罪 犯",
+      "umm ary",
+      "no w",
+      "n ow",
+      "it em",
+      "ite m",
+      "i tem",
+      "▁c ases",
+      "▁case s",
+      "▁ca ses",
+      "▁cas es",
+      "▁ cases",
+      "▁im plement",
+      "▁imp lement",
+      "▁impl ement",
+      "b b",
+      "os p",
+      "o sp",
+      "以 下",
+      "▁A ust",
+      "▁Au st",
+      "▁Aus t",
+      "健 康",
+      "ou d",
+      "o ud",
+      "▁F e",
+      "▁ Fe",
+      "▁d ig",
+      "▁di g",
+      "▁ dig",
+      "▁s imple",
+      "▁sim ple",
+      "▁simpl e",
+      "▁ simple",
+      "ut es",
+      "ute s",
+      "u tes",
+      "自 然",
+      "▁M ed",
+      "▁Me d",
+      "▁ Med",
+      "gin g",
+      "gi ng",
+      "g ing",
+      "第一 款",
+      "第 一款",
+      "交 通",
+      "og en",
+      "oge n",
+      "o gen",
+      "▁a cross",
+      "▁ac ross",
+      "以 上",
+      "法定 代表人",
+      "积 极",
+      "法律 效力",
+      "nW hat",
+      "nWh at",
+      "n What",
+      "▁w on",
+      "▁wo n",
+      "▁ won",
+      "eg in",
+      "e gin",
+      "ru n",
+      "r un",
+      "▁ 员",
+      "Tim e",
+      "Ti me",
+      "T ime",
+      "注 意",
+      "dit ions",
+      "dition s",
+      "d itions",
+      "由 于",
+      "▁s um",
+      "▁su m",
+      "▁ sum",
+      "▁D es",
+      "▁De s",
+      "▁ Des",
+      "ag s",
+      "a gs",
+      "▁N one",
+      "▁No ne",
+      "▁Non e",
+      "▁ None",
+      "gor ith",
+      "▁f low",
+      "▁fl ow",
+      "▁flo w",
+      "▁ flow",
+      "er c",
+      "e rc",
+      "am ed",
+      "ame d",
+      "a med",
+      "am m",
+      "a mm",
+      "om ic",
+      "omi c",
+      "o mic",
+      "li m",
+      "l im",
+      "▁R eg",
+      "▁Re g",
+      "▁ Reg",
+      "n V",
+      "re quest",
+      "requ est",
+      "▁fil m",
+      "▁fi lm",
+      "▁ film",
+      "ba r",
+      "b ar",
+      "▁I nt",
+      "▁In t",
+      "▁ Int",
+      "▁c lear",
+      "▁cl ear",
+      "▁cle ar",
+      "▁ clear",
+      "▁in put",
+      "▁ input",
+      "P l",
+      "ble ms",
+      "blem s",
+      "bl ems",
+      "ch es",
+      "che s",
+      "c hes",
+      "▁p ast",
+      "▁pa st",
+      "▁pas t",
+      "▁ past",
+      "▁( )",
+      "▁ ()",
+      "▁c ult",
+      "▁cul t",
+      "▁cu lt",
+      "▁ cult",
+      "▁f und",
+      "▁fun d",
+      "▁fu nd",
+      "▁ fund",
+      "学 院",
+      "▁h ouse",
+      "▁hous e",
+      "▁ho use",
+      "▁ house",
+      "ye ar",
+      "y ear",
+      "st it",
+      "s tit",
+      "pe n",
+      "p en",
+      "S S",
+      "▁M in",
+      "▁Mi n",
+      "▁ Min",
+      "作 出",
+      "▁com pon",
+      "▁comp on",
+      "po int",
+      "p oint",
+      "▁f urther",
+      "▁fur ther",
+      "▁h ost",
+      "▁ho st",
+      "▁ host",
+      "is on",
+      "iso n",
+      "i son",
+      "ud e",
+      "u de",
+      "~ ~",
+      "▁p lease",
+      "▁ple ase",
+      "▁pleas e",
+      "▁plea se",
+      "▁ please",
+      "发生 法律效力",
+      "par t",
+      "pa rt",
+      "p art",
+      "pl it",
+      "p lit",
+      "st yle",
+      "sty le",
+      "styl e",
+      "▁n ormal",
+      "▁nor mal",
+      "▁norm al",
+      "▁ normal",
+      "▁incre ase",
+      "分 钟",
+      "▁f un",
+      "▁fu n",
+      "▁ fun",
+      "ai m",
+      "a im",
+      "运 动",
+      "▁W hen",
+      "▁Wh en",
+      "▁ When",
+      "li n",
+      "l in",
+      "万 元",
+      "ip s",
+      "i ps",
+      "ave l",
+      "av el",
+      "a vel",
+      "x y",
+      "com p",
+      "co mp",
+      "c omp",
+      "bo dy",
+      "b ody",
+      "pl ication",
+      "plic ation",
+      "plica tion",
+      "▁p oints",
+      "▁point s",
+      "▁po ints",
+      "▁ points",
+      "do m",
+      "d om",
+      "个 月",
+      "▁cont act",
+      "▁ contact",
+      "▁a bs",
+      "▁ab s",
+      "▁ abs",
+      "Tr ue",
+      "▁C o",
+      "▁ Co",
+      "▁c he",
+      "▁ch e",
+      "▁ che",
+      "▁J ul",
+      "▁Ju l",
+      "▁ Jul",
+      "▁work s",
+      "▁wor ks",
+      "▁ works",
+      "ic ult",
+      "icul t",
+      "i cult",
+      "▁d ri",
+      "▁dr i",
+      "( (",
+      "dit ion",
+      "di tion",
+      "d ition",
+      "▁M y",
+      "▁ My",
+      "产 业",
+      "▁f actors",
+      "▁fact ors",
+      "▁factor s",
+      "▁fa ctors",
+      "▁ factors",
+      "▁as sess",
+      "▁ass ess",
+      "ri de",
+      "rid e",
+      "r ide",
+      "ir m",
+      "i rm",
+      "▁S outh",
+      "▁So uth",
+      "▁Sou th",
+      "▁ South",
+      "am ework",
+      "ame work",
+      "# \\",
+      "价 值",
+      "vin g",
+      "vi ng",
+      "v ing",
+      "▁mon ths",
+      "▁month s",
+      "▁mont hs",
+      "▁ months",
+      "▁su bject",
+      "▁sub ject",
+      "▁ subject",
+      "义 务",
+      "in dow",
+      "ind ow",
+      "▁mult iple",
+      "▁multi ple",
+      "▁multip le",
+      "▁ multiple",
+      "▁W orld",
+      "▁Wor ld",
+      "▁ World",
+      "▁cl inical",
+      "▁clin ical",
+      "▁clinic al",
+      "▁ clinical",
+      "ci n",
+      "c in",
+      "▁I N",
+      "▁ IN",
+      "▁d eg",
+      "▁de g",
+      "▁ deg",
+      "le ge",
+      "leg e",
+      "这 是",
+      "▁c ancer",
+      "▁can cer",
+      "▁ cancer",
+      "▁f ood",
+      "▁fo od",
+      "▁foo d",
+      "▁ food",
+      "希 望",
+      "不 过",
+      "▁ter ms",
+      "▁term s",
+      "▁ terms",
+      "特 别",
+      "at form",
+      "do wn",
+      "dow n",
+      "d own",
+      "▁st yle",
+      "▁styl e",
+      "▁ style",
+      "用 户",
+      "▁b it",
+      "▁bi t",
+      "▁ bit",
+      "患 者",
+      "条第 一款",
+      "条 第一款",
+      "▁ex press",
+      "▁exp ress",
+      "▁expr ess",
+      "▁ express",
+      "▁s earch",
+      "▁se arch",
+      "▁sear ch",
+      "▁ search",
+      "▁St ate",
+      "▁Stat e",
+      "▁Sta te",
+      "▁ State",
+      "▁st rong",
+      "▁str ong",
+      "▁stro ng",
+      "▁ strong",
+      "ag ing",
+      "agi ng",
+      "agin g",
+      "a ging",
+      "组 成",
+      "▁me asure",
+      "▁meas ure",
+      "▁ measure",
+      "予 以",
+      "I V",
+      "## ######",
+      "#### ####",
+      "### #####",
+      "##### ###",
+      "###### ##",
+      ")) ;\\",
+      ")); \\",
+      ") );\\",
+      "▁f ore",
+      "▁for e",
+      "▁fo re",
+      "▁ fore",
+      "机 构",
+      "st d",
+      "s td",
+      "gr oup",
+      "gro up",
+      "g roup",
+      "起 来",
+      "as tic",
+      "ast ic",
+      "sr c",
+      "s rc",
+      "▁l arg",
+      "▁la rg",
+      "▁lar g",
+      "▁ larg",
+      "im ate",
+      "ima te",
+      "i mate",
+      "act ions",
+      "action s",
+      "ver se",
+      "vers e",
+      "▁A m",
+      "▁ Am",
+      "W e",
+      "fe ct",
+      "f ect",
+      "on str",
+      "ons tr",
+      "o nstr",
+      "d f",
+      "▁d iff",
+      "▁dif f",
+      "▁di ff",
+      "▁ diff",
+      "▁con ditions",
+      "▁condition s",
+      "▁cond itions",
+      "▁ conditions",
+      "这 一",
+      "实 际",
+      "il it",
+      "ili t",
+      "i lit",
+      "效 果",
+      "ect or",
+      "ec tor",
+      "e ctor",
+      "▁G ener",
+      "▁Ge ner",
+      "▁Gen er",
+      "▁Gene r",
+      "▁ Gener",
+      "ir l",
+      "i rl",
+      "▁S chool",
+      "▁Sch ool",
+      "▁ School",
+      "ni nt",
+      "nin t",
+      "n int",
+      "gr ess",
+      "gres s",
+      "gre ss",
+      "g ress",
+      "per ty",
+      "pert y",
+      "p erty",
+      "for d",
+      "fo rd",
+      "f ord",
+      "▁E uro",
+      "▁Eur o",
+      "▁ Euro",
+      "▁m ass",
+      "▁ma ss",
+      "▁mas s",
+      "▁ mass",
+      "yt hon",
+      "y thon",
+      "历 史",
+      "手 机",
+      "▁f uture",
+      "▁fu ture",
+      "▁fut ure",
+      "▁ future",
+      "\\' ]",
+      "\\ ']",
+      "▁method s",
+      "▁meth ods",
+      "▁ methods",
+      "综 合",
+      "ef ul",
+      "e ful",
+      "Lo g",
+      "L og",
+      "是 否",
+      "am a",
+      "a ma",
+      "▁part icip",
+      "▁partic ip",
+      "▁ particip",
+      "St ate",
+      "Stat e",
+      "C H",
+      "▁G u",
+      "▁ Gu",
+      "为 什么",
+      "P I",
+      "▁f ar",
+      "▁fa r",
+      "▁ far",
+      "保 护",
+      "jo y",
+      "j oy",
+      "▁< <",
+      "▁ <<",
+      "ap y",
+      "a py",
+      "ad a",
+      "a da",
+      "▁m usic",
+      "▁mus ic",
+      "▁ music",
+      "住 所",
+      "酒 店",
+      "级 人民法院",
+      "Th is",
+      "T his",
+      "Ge t",
+      "G et",
+      "an a",
+      "a na",
+      "em pt",
+      "emp t",
+      "▁pro tein",
+      "▁prote in",
+      "▁prot ein",
+      "▁ protein",
+      "ia te",
+      "iat e",
+      "i ate",
+      "ba se",
+      "bas e",
+      "b ase",
+      "多 少",
+      "▁& &",
+      "▁ &&",
+      "▁does n",
+      "▁s tre",
+      "▁st re",
+      "▁str e",
+      "▁ stre",
+      "▁k ind",
+      "▁kin d",
+      "▁ki nd",
+      "▁ kind",
+      "▁l oad",
+      "▁lo ad",
+      "▁ load",
+      "▁incre ased",
+      "▁increase d",
+      "nF or",
+      "n For",
+      "老 师",
+      "ab ase",
+      "aba se",
+      "a base",
+      "U L",
+      "▁Q u",
+      "▁ Qu",
+      "yn am",
+      "yna m",
+      "y nam",
+      "▁comp ared",
+      "▁compar ed",
+      "▁compare d",
+      "状 态",
+      "▁m essage",
+      "▁mess age",
+      "▁ message",
+      "▁m iss",
+      "▁mis s",
+      "▁mi ss",
+      "▁ miss",
+      "▁contin u",
+      "只 有",
+      "▁p rint",
+      "▁pr int",
+      "▁pri nt",
+      "▁ print",
+      "得 到",
+      "le ngth",
+      "l ength",
+      "th ough",
+      "集 团",
+      "ch o",
+      "c ho",
+      "ou te",
+      "out e",
+      "o ute",
+      "n K",
+      "▁e asy",
+      "▁eas y",
+      "▁ easy",
+      "▁t able",
+      "▁tab le",
+      "▁ta ble",
+      "▁ table",
+      "▁O ct",
+      "▁Oc t",
+      "▁ Oct",
+      "is c",
+      "i sc",
+      "▁al bum",
+      "▁ album",
+      "bo ard",
+      "b oard",
+      "因 此",
+      "▁a im",
+      "▁ai m",
+      "▁ aim",
+      "▁p osition",
+      "▁pos ition",
+      "▁posit ion",
+      "▁ position",
+      "▁str ateg",
+      "▁strat eg",
+      "ob ject",
+      "obj ect",
+      "o bject",
+      "▁m akes",
+      "▁make s",
+      "▁ma kes",
+      "im age",
+      "ima ge",
+      "imag e",
+      "le ased",
+      "lease d",
+      "lea sed",
+      "at ural",
+      "atur al",
+      "atu ral",
+      "atura l",
+      "cre ate",
+      "creat e",
+      "c reate",
+      "▁vis it",
+      "▁vi sit",
+      "▁ visit",
+      "▁ben ef",
+      "▁ benef",
+      "ri or",
+      "rio r",
+      "r ior",
+      "at al",
+      "ata l",
+      "a tal",
+      "li nk",
+      "lin k",
+      "l ink",
+      "▁A pr",
+      "▁Ap r",
+      "▁ Apr",
+      "et a",
+      "e ta",
+      "ic ation",
+      "ica tion",
+      "i cation",
+      "se c",
+      "s ec",
+      "nuser name",
+      "n username",
+      "销 售",
+      "▁tr aining",
+      "▁train ing",
+      "▁tra ining",
+      "▁ training",
+      "看 到",
+      "▁l ower",
+      "▁low er",
+      "▁lo wer",
+      "▁ lower",
+      "ep endent",
+      "epend ent",
+      "▁c oll",
+      "▁col l",
+      "▁co ll",
+      "▁ coll",
+      "C L",
+      "▁c y",
+      "▁ cy",
+      "感 觉",
+      "▁o s",
+      "▁ os",
+      "▁l earn",
+      "▁le arn",
+      "▁lear n",
+      "▁ learn",
+      "▁a ch",
+      "▁ac h",
+      "▁ ach",
+      "▁p olit",
+      "▁pol it",
+      "▁po lit",
+      "▁ polit",
+      "eg er",
+      "e ger",
+      "▁c ity",
+      "▁cit y",
+      "▁ci ty",
+      "▁ city",
+      "er a",
+      "e ra",
+      "lo od",
+      "loo d",
+      "l ood",
+      "an ced",
+      "ance d",
+      "anc ed",
+      "不 断",
+      "bu ild",
+      "b uild",
+      "il ities",
+      "ilit ies",
+      "ili ties",
+      "> >",
+      "▁l imit",
+      "▁li mit",
+      "▁lim it",
+      "▁ limit",
+      "▁c olor",
+      "▁col or",
+      "▁co lor",
+      "▁ color",
+      "年 的",
+      "日 内",
+      "依 据",
+      "ak en",
+      "ake n",
+      "a ken",
+      "▁be come",
+      "▁bec ome",
+      "▁ become",
+      "ibr ary",
+      "bi t",
+      "b it",
+      "利 息",
+      "et ic",
+      "eti c",
+      "e tic",
+      "股 份",
+      "原 因",
+      "十 二",
+      "▁p roduction",
+      "▁pro duction",
+      "▁produ ction",
+      "▁product ion",
+      "▁prod uction",
+      "▁ production",
+      "ers ion",
+      "应 当",
+      "劳 动",
+      "om an",
+      "oma n",
+      "o man",
+      "▁v ia",
+      "▁vi a",
+      "▁ via",
+      "▁t rack",
+      "▁tr ack",
+      "▁tra ck",
+      "▁ track",
+      "部 门",
+      "ateg ory",
+      "ategor y",
+      "ound s",
+      "oun ds",
+      "面 的",
+      "$$ \\",
+      "$ $\\",
+      "en n",
+      "e nn",
+      "an nel",
+      "ann el",
+      "anne l",
+      "co re",
+      "cor e",
+      "c ore",
+      "pl ac",
+      "负 责",
+      "){ \\",
+      ") {\\",
+      "br e",
+      "b re",
+      "▁rel ationship",
+      "▁relations hip",
+      "▁relation ship",
+      "▁ relationship",
+      "nc e",
+      "n ce",
+      "创 新",
+      "▁s qu",
+      "▁sq u",
+      "▁ squ",
+      "▁In tern",
+      "▁Int ern",
+      "▁Inter n",
+      "▁str ucture",
+      "▁struct ure",
+      "▁ structure",
+      "or age",
+      "ora ge",
+      "o rage",
+      "A M",
+      "▁M e",
+      "▁ Me",
+      "▁f requ",
+      "▁fr equ",
+      "▁fre qu",
+      "▁freq u",
+      "ir es",
+      "ire s",
+      "i res",
+      "ol l",
+      "o ll",
+      "▁G o",
+      "▁ Go",
+      "▁G e",
+      "▁ Ge",
+      "南 省",
+      "计 划",
+      "Con fig",
+      "Co nfig",
+      "Conf ig",
+      "▁ad ded",
+      "▁add ed",
+      "▁ added",
+      "▁I I",
+      "▁ II",
+      "▁c orrect",
+      "▁cor rect",
+      "▁ correct",
+      "虽 然",
+      "▁ 作者",
+      "▁c over",
+      "▁co ver",
+      "▁cov er",
+      "▁ cover",
+      "▁A dd",
+      "▁Ad d",
+      "▁ Add",
+      "▁R ec",
+      "▁Re c",
+      "▁ Rec",
+      "解 释",
+      "led ge",
+      "l edge",
+      "st ars",
+      "star s",
+      "sta rs",
+      "er al",
+      "era l",
+      "e ral",
+      "住所 地",
+      "▁e vents",
+      "▁ev ents",
+      "▁even ts",
+      "▁event s",
+      "▁ events",
+      "ns wer",
+      "wid th",
+      "w idth",
+      "▁im mun",
+      "▁imm un",
+      "▁ immun",
+      "op s",
+      "o ps",
+      "def ault",
+      "de fault",
+      "$ ,",
+      "▁dem onstr",
+      "▁demo nstr",
+      "▁demon str",
+      "▁h y",
+      "▁ hy",
+      "▁pro blems",
+      "▁problem s",
+      "▁ problems",
+      "om y",
+      "o my",
+      "▁g rowth",
+      "▁grow th",
+      "▁ growth",
+      "有 关",
+      "▁ 连载",
+      "▁re d",
+      "▁r ed",
+      "▁ red",
+      "地 方",
+      "▁R ep",
+      "▁Re p",
+      "▁ Rep",
+      "▁re ceived",
+      "▁rece ived",
+      "▁receive d",
+      "▁ received",
+      "di ng",
+      "din g",
+      "d ing",
+      "am ent",
+      "ame nt",
+      "amen t",
+      "a ment",
+      "▁pos itive",
+      "▁posit ive",
+      "▁ positive",
+      "的 话",
+      "be gin",
+      "beg in",
+      "b egin",
+      "▁Y ork",
+      "▁Yo rk",
+      "tp ublic",
+      "t public",
+      "R e",
+      "安 装",
+      "▁j ava",
+      "▁jav a",
+      "▁ja va",
+      "▁ java",
+      "▁T hat",
+      "▁Th at",
+      "▁ That",
+      "成 功",
+      "il es",
+      "ile s",
+      "i les",
+      "sp ec",
+      "spe c",
+      "s pec",
+      "▁fe ature",
+      "▁feat ure",
+      "▁ feature",
+      "▁l anguage",
+      "▁ language",
+      "I R",
+      "回 答",
+      "受理 费",
+      "▁D r",
+      "▁ Dr",
+      "▁count ry",
+      "▁cou ntry",
+      "▁ country",
+      "▁l ength",
+      "▁le ngth",
+      "▁leng th",
+      "▁ length",
+      "I t",
+      "idd le",
+      "▁th ought",
+      "▁though t",
+      "▁ thought",
+      "▁f actor",
+      "▁fact or",
+      "▁fac tor",
+      "▁fa ctor",
+      "▁ factor",
+      "▁h om",
+      "▁ho m",
+      "▁ hom",
+      "pack age",
+      "p ackage",
+      "EN T",
+      "E NT",
+      "医 院",
+      "▁ad min",
+      "▁adm in",
+      "▁ admin",
+      "▁a way",
+      "▁aw ay",
+      "▁awa y",
+      "▁ away",
+      "水 平",
+      "也 不",
+      "▁h yp",
+      "▁hy p",
+      "▁ hyp",
+      "本 次",
+      "\\\\ (\\\\",
+      "\\\\( \\\\",
+      "pos t",
+      "po st",
+      "p ost",
+      "U n",
+      "▁P art",
+      "▁Par t",
+      "▁Pa rt",
+      "▁ Part",
+      "▁pop ulation",
+      "▁popul ation",
+      "▁ population",
+      "nT o",
+      "n To",
+      "▁re fer",
+      "▁r efer",
+      "▁ref er",
+      "▁ refer",
+      "▁n eg",
+      "▁ne g",
+      "▁ neg",
+      "▁b lock",
+      "▁bl ock",
+      "▁blo ck",
+      "▁ block",
+      "▁out put",
+      "▁ output",
+      "ist ics",
+      "istic s",
+      "▁pro posed",
+      "▁propos ed",
+      "▁prop osed",
+      "▁propose d",
+      "▁m anagement",
+      "▁man agement",
+      "▁manag ement",
+      "▁manage ment",
+      "▁ management",
+      "mo st",
+      "mos t",
+      "m ost",
+      "qu e",
+      "q ue",
+      "消 费",
+      "▁qu estions",
+      "▁question s",
+      "▁quest ions",
+      "▁ questions",
+      "医 学",
+      "▁d rug",
+      "▁dr ug",
+      "▁ drug",
+      "Requ est",
+      "Re quest",
+      "▁th ird",
+      "▁thi rd",
+      "▁ third",
+      "▁f ive",
+      "▁fi ve",
+      "▁ five",
+      "艺 术",
+      "必 须",
+      "ar ray",
+      "arr ay",
+      "arra y",
+      "um ents",
+      "ument s",
+      "umen ts",
+      "u ments",
+      "et er",
+      "ete r",
+      "e ter",
+      "▁w eb",
+      "▁we b",
+      "▁ web",
+      "国 内",
+      "分 别",
+      "▁o ption",
+      "▁op tion",
+      "▁opt ion",
+      "▁ option",
+      "▁D o",
+      "▁ Do",
+      "s k",
+      "so le",
+      "sol e",
+      "s ole",
+      "之 一",
+      "▁O ne",
+      "▁On e",
+      "▁ One",
+      "▁a mount",
+      "▁am ount",
+      "▁ amount",
+      "来 说",
+      "▁c ertain",
+      "▁cert ain",
+      "▁cer tain",
+      "it es",
+      "ite s",
+      "i tes",
+      "只 是",
+      "方 程",
+      "十 一",
+      "op en",
+      "ope n",
+      "o pen",
+      "▁prov ides",
+      "▁provid es",
+      "▁provide s",
+      "▁c reated",
+      "▁cre ated",
+      "▁create d",
+      "▁creat ed",
+      "▁ created",
+      "▁A ug",
+      "▁Au g",
+      "▁ Aug",
+      "▁N orth",
+      "▁Nor th",
+      "▁ North",
+      "▁h old",
+      "▁hol d",
+      "▁ho ld",
+      "▁ hold",
+      "▁Ch rist",
+      "▁Chris t",
+      "▁ Christ",
+      "▁s ong",
+      "▁so ng",
+      "▁son g",
+      "▁ song",
+      "请 求",
+      "▁L ist",
+      "▁Li st",
+      "▁Lis t",
+      "▁ List",
+      "▁l ocated",
+      "▁loc ated",
+      "▁locate d",
+      "▁ located",
+      "re pon",
+      "rep on",
+      "repo n",
+      "▁det ermine",
+      "▁determ ine",
+      "▁determin e",
+      "▁deter mine",
+      "▁m or",
+      "▁mo r",
+      "▁ mor",
+      "▁in cluded",
+      "▁includ ed",
+      "▁include d",
+      "▁incl uded",
+      "▁ included",
+      "A G",
+      "▁develop ed",
+      "▁ developed",
+      "电 子",
+      "▁a uthor",
+      "▁aut hor",
+      "▁auth or",
+      "▁au thor",
+      "▁ author",
+      "▁l and",
+      "▁la nd",
+      "▁lan d",
+      "▁ land",
+      "for mat",
+      "form at",
+      "str act",
+      "stra ct",
+      "s tract",
+      "▁ )",
+      "▁a ng",
+      "▁an g",
+      "▁ ang",
+      "port un",
+      "por tun",
+      "repon ame",
+      "repo name",
+      "ne rs",
+      "ner s",
+      "n ers",
+      "▁h istory",
+      "▁hist ory",
+      "▁histor y",
+      "▁hi story",
+      "▁ history",
+      "el ling",
+      "ell ing",
+      "elli ng",
+      "▁you ng",
+      "▁yo ung",
+      "▁ young",
+      "▁r at",
+      "▁ra t",
+      "▁ rat",
+      "▁c hall",
+      "▁ch all",
+      "▁cha ll",
+      "Pat h",
+      "Pa th",
+      "P ath",
+      "us es",
+      "use s",
+      "u ses",
+      "▁C al",
+      "▁Ca l",
+      "▁ Cal",
+      "▁com plete",
+      "▁comp lete",
+      "▁complet e",
+      "▁compl ete",
+      "▁ complete",
+      "他 的",
+      "▁m ole",
+      "▁mo le",
+      "▁mol e",
+      "▁mon th",
+      "▁mo nth",
+      "▁mont h",
+      "▁ month",
+      "pr ess",
+      "pre ss",
+      "pres s",
+      "p ress",
+      "it ory",
+      "itor y",
+      "ito ry",
+      "▁pre dict",
+      "▁pred ict",
+      "▁ predict",
+      "um b",
+      "u mb",
+      "iz ing",
+      "izi ng",
+      "i zing",
+      "ma il",
+      "m ail",
+      "▁c alcul",
+      "▁calc ul",
+      "▁ calcul",
+      "np ublic",
+      "npub lic",
+      "n public",
+      "do c",
+      "d oc",
+      "ri ed",
+      "rie d",
+      "r ied",
+      "撤 诉",
+      "生 效",
+      "un ch",
+      "unc h",
+      "u nch",
+      "▁t hing",
+      "▁th ing",
+      "▁thin g",
+      "▁thi ng",
+      "▁ thing",
+      "精 神",
+      "al ing",
+      "ali ng",
+      "alin g",
+      "a ling",
+      "no de",
+      "nod e",
+      "n ode",
+      "▁e ns",
+      "▁en s",
+      "▁ ens",
+      "▁m oney",
+      "▁mon ey",
+      "▁mo ney",
+      "▁ money",
+      "U N",
+      "it ions",
+      "ition s",
+      "iti ons",
+      "▁ind ex",
+      "▁ index",
+      "O r",
+      "▁o ptions",
+      "▁opt ions",
+      "▁option s",
+      "▁ options",
+      "P S",
+      "结 合",
+      "▁s at",
+      "▁sa t",
+      "▁ sat",
+      "一 步",
+      "L O",
+      "▁v alid",
+      "▁val id",
+      "▁ valid",
+      "中级 人民法院",
+      "中 级人民法院",
+      "▁prov ided",
+      "▁provid ed",
+      "▁provide d",
+      "▁ provided",
+      "se qu",
+      "seq u",
+      "s equ",
+      "▁iss ues",
+      "▁issue s",
+      "▁ issues",
+      "汽 车",
+      "se ction",
+      "sec tion",
+      "sect ion",
+      "s ection",
+      "无 法",
+      "▁o nce",
+      "▁on ce",
+      "▁onc e",
+      "▁ once",
+      "山 东",
+      "ul a",
+      "u la",
+      "▁A b",
+      "▁ Ab",
+      "▁O F",
+      "▁ OF",
+      "ic les",
+      "icle s",
+      "icl es",
+      "i cles",
+      "▁P ol",
+      "▁Po l",
+      "▁ Pol",
+      "M S",
+      "▁[ \\",
+      "▁ [\\",
+      "ist ance",
+      "istan ce",
+      "ista nce",
+      "i stance",
+      "B y",
+      "▁$ $\\\\",
+      "▁$$ \\\\",
+      "▁ $$\\\\",
+      "▁c amp",
+      "▁ca mp",
+      "▁cam p",
+      "▁ camp",
+      "▁C ity",
+      "▁Cit y",
+      "▁Ci ty",
+      "▁ City",
+      "▁pro perty",
+      "▁proper ty",
+      "▁prop erty",
+      "▁ property",
+      "ik e",
+      "i ke",
+      "▁imp act",
+      "▁ impact",
+      "▁start ed",
+      "▁star ted",
+      "▁ started",
+      "Ad d",
+      "A dd",
+      "al t",
+      "a lt",
+      "▁build ing",
+      "▁ building",
+      "it ional",
+      "ition al",
+      "iti onal",
+      "增 加",
+      "he st",
+      "hes t",
+      "h est",
+      "[ ]",
+      "▁or iginal",
+      "▁orig inal",
+      "▁origin al",
+      "▁ original",
+      "▁N ot",
+      "▁No t",
+      "▁ Not",
+      "ur ch",
+      "in fo",
+      "inf o",
+      "这 里",
+      "ang le",
+      "ir ection",
+      "ire ction",
+      "irect ion",
+      "i rection",
+      "u i",
+      "双 方",
+      "am ples",
+      "amp les",
+      "ample s",
+      "it ted",
+      "itt ed",
+      "▁E urope",
+      "▁Euro pe",
+      "▁Eur ope",
+      "▁ Europe",
+      "er ed",
+      "ere d",
+      "e red",
+      "ul ts",
+      "ult s",
+      "sc ript",
+      "scri pt",
+      "scr ipt",
+      "s cript",
+      "▁p arent",
+      "▁par ent",
+      "▁pa rent",
+      "▁pare nt",
+      "▁paren t",
+      "▁ parent",
+      "▁v er",
+      "▁ve r",
+      "▁ ver",
+      "模 式",
+      "ra m",
+      "r am",
+      "▁per cent",
+      "▁perc ent",
+      "▁ percent",
+      "▁con text",
+      "▁cont ext",
+      "▁co ntext",
+      "▁ context",
+      "判决 书",
+      "▁m ag",
+      "▁ma g",
+      "▁ mag",
+      "__ __",
+      "___ _",
+      "_ ___",
+      "mess age",
+      "m essage",
+      "ic tion",
+      "ict ion",
+      "i ction",
+      "▁t each",
+      "▁te ach",
+      "▁tea ch",
+      "▁S er",
+      "▁Se r",
+      "▁ Ser",
+      "▁ q",
+      "an ies",
+      "ani es",
+      "anie s",
+      "▁s uper",
+      "▁su per",
+      "▁sup er",
+      "▁ super",
+      "▁f iles",
+      "▁fil es",
+      "▁file s",
+      "▁fi les",
+      "▁ files",
+      "an go",
+      "ang o",
+      "a ngo",
+      "act ive",
+      "activ e",
+      "目 标",
+      "政 策",
+      "dis play",
+      "disp lay",
+      "▁o thers",
+      "▁other s",
+      "▁ others",
+      "▁s ex",
+      "▁se x",
+      "▁ sex",
+      "▁te chnology",
+      "▁techn ology",
+      "▁ technology",
+      "od ing",
+      "odi ng",
+      "o ding",
+      "▁A rt",
+      "▁Ar t",
+      "▁ Art",
+      "文 件",
+      "ic on",
+      "ico n",
+      "i con",
+      "` \\",
+      "▁re lease",
+      "▁ release",
+      "▁m ix",
+      "▁mi x",
+      "▁ mix",
+      "▁m atch",
+      "▁mat ch",
+      "▁ match",
+      "$\\ \\",
+      "$ \\\\",
+      "▁a reas",
+      "▁are as",
+      "▁ar eas",
+      "▁area s",
+      "可 供",
+      "at est",
+      "ate st",
+      "ates t",
+      "a test",
+      "▁r andom",
+      "▁ran dom",
+      "▁rand om",
+      "▁ random",
+      "▁h ot",
+      "▁ho t",
+      "▁ hot",
+      "ol ve",
+      "olv e",
+      "系 列",
+      "[ \"",
+      "▁C O",
+      "▁ CO",
+      "客 户",
+      "wh ile",
+      "软 件",
+      "▁d iv",
+      "▁di v",
+      "▁ div",
+      "我 国",
+      "▁op portun",
+      "这 么",
+      "继 续",
+      "le ments",
+      "lement s",
+      "lem ents",
+      "l ements",
+      "▁function s",
+      "▁funct ions",
+      "▁ functions",
+      "| \\",
+      "▁h ours",
+      "▁hour s",
+      "▁ho urs",
+      "▁ hours",
+      "▁y et",
+      "▁ye t",
+      "▁ yet",
+      "ua tion",
+      "u ation",
+      "ul ated",
+      "ula ted",
+      "ulate d",
+      "u lated",
+      "▁g rad",
+      "▁gr ad",
+      "▁gra d",
+      "▁ grad",
+      "For m",
+      "F orm",
+      "▁current ly",
+      "▁curr ently",
+      "▁ currently",
+      "an ch",
+      "anc h",
+      "a nch",
+      "os ph",
+      "osp h",
+      "su m",
+      "s um",
+      "ens ive",
+      "▁spec ies",
+      "▁ species",
+      "▁eng ine",
+      "▁ engine",
+      "im g",
+      "i mg",
+      "▁g overnment",
+      "▁gover nment",
+      "▁govern ment",
+      "▁ government",
+      "▁art icle",
+      "▁artic le",
+      "▁ article",
+      "if f",
+      "i ff",
+      "▁l ive",
+      "▁li ve",
+      "▁liv e",
+      "▁ live",
+      "▁t ypes",
+      "▁type s",
+      "▁typ es",
+      "▁ty pes",
+      "▁ types",
+      "工 业",
+      "av or",
+      "avo r",
+      "a vor",
+      "位 置",
+      "▁re ason",
+      "▁ reason",
+      "▁m ember",
+      "▁me mber",
+      "▁mem ber",
+      "▁ member",
+      "▁p ick",
+      "▁pi ck",
+      "▁pic k",
+      "▁ pick",
+      "Ma p",
+      "M ap",
+      "进 入",
+      "▁St ud",
+      "▁ Stud",
+      "▁en ough",
+      "ay out",
+      "a yout",
+      "▁re gard",
+      "▁reg ard",
+      "ma rk",
+      "mar k",
+      "m ark",
+      "▁d ate",
+      "▁dat e",
+      "▁da te",
+      "▁ date",
+      "Se r",
+      "S er",
+      "▁c los",
+      "▁cl os",
+      "▁ex ec",
+      "▁ exec",
+      "}} {",
+      "} }{",
+      "▁A ustral",
+      "▁Aust ral",
+      "▁g etting",
+      "▁get ting",
+      "▁ getting",
+      "▁th ough",
+      "▁ though",
+      "▁A fter",
+      "▁Af ter",
+      "▁ After",
+      "con n",
+      "co nn",
+      "c onn",
+      "') \\",
+      "' )\\",
+      "农 村",
+      "▁p rice",
+      "▁pr ice",
+      "▁pri ce",
+      "▁ price",
+      "in ding",
+      "ind ing",
+      "indi ng",
+      "ex pect",
+      "exp ect",
+      "ro om",
+      "r oom",
+      "ur g",
+      "u rg",
+      "▁c omes",
+      "▁com es",
+      "▁co mes",
+      "▁come s",
+      "▁ comes",
+      "pa t",
+      "p at",
+      "}) \\",
+      "} )\\",
+      "ur b",
+      "u rb",
+      "▁n umbers",
+      "▁num bers",
+      "▁number s",
+      "▁ numbers",
+      "r m",
+      "出 来",
+      "案件 受理费",
+      "opy right",
+      "▁r oom",
+      "▁ro om",
+      "▁ room",
+      "M L",
+      "▁en joy",
+      "ob ile",
+      "obi le",
+      "obil e",
+      "简 单",
+      "▁de signed",
+      "▁des igned",
+      "▁design ed",
+      "▁ designed",
+      "▁in hib",
+      "▁inh ib",
+      "▁ inhib",
+      "证 明",
+      "▁s ix",
+      "▁si x",
+      "▁ six",
+      "股份 有限公司",
+      "▁qu ant",
+      "▁q uant",
+      "▁ quant",
+      "自 动",
+      "') ;\\",
+      "'); \\",
+      "' );\\",
+      "nL et",
+      "nLe t",
+      "n Let",
+      "电 影",
+      "~ \\",
+      "准 许",
+      "▁s een",
+      "▁se en",
+      "▁see n",
+      "▁ seen",
+      "▁p ersonal",
+      "▁person al",
+      "▁pers onal",
+      "▁ personal",
+      "▁do ing",
+      "▁doi ng",
+      "▁ doing",
+      "▁S ec",
+      "▁Se c",
+      "▁ Sec",
+      "▁s w",
+      "▁ sw",
+      "▁min utes",
+      "▁minute s",
+      "▁ minutes",
+      "▁aut om",
+      "▁auto m",
+      "▁au tom",
+      "▁ autom",
+      "wh ere",
+      "w here",
+      "▁pat tern",
+      "▁ pattern",
+      "min g",
+      "mi ng",
+      "m ing",
+      "um ber",
+      "umb er",
+      "u mber",
+      "▁qu ick",
+      "▁qui ck",
+      "▁ quick",
+      "▁ne cess",
+      "▁nec ess",
+      "es e",
+      "e se",
+      "▁O R",
+      "▁ OR",
+      "$ {",
+      "按 照",
+      "In f",
+      "I nf",
+      "▁pro perties",
+      "▁proper ties",
+      "▁ properties",
+      "on y",
+      "o ny",
+      "re ate",
+      "reat e",
+      "rea te",
+      "产 生",
+      "更 多",
+      "▁g raph",
+      "▁gr aph",
+      "▁grap h",
+      "▁gra ph",
+      "▁ graph",
+      "de v",
+      "d ev",
+      "▁act ually",
+      "▁actual ly",
+      "▁actu ally",
+      "▁ actually",
+      "▁c lose",
+      "▁cl ose",
+      "▁clos e",
+      "▁ close",
+      "\\' ;\\",
+      "\\'; \\",
+      "\\ ';\\",
+      "完 全",
+      "通 知",
+      "▁in flu",
+      "▁inf lu",
+      "▁infl u",
+      "▁un ique",
+      "▁uniqu e",
+      "▁uni que",
+      "▁ unique",
+      "浙 江",
+      "{ \"",
+      "▁m illion",
+      "▁mill ion",
+      "▁mil lion",
+      "▁ million",
+      "▁b ar",
+      "▁ba r",
+      "▁ bar",
+      "▁init ial",
+      "▁initi al",
+      "▁ initial",
+      "so ft",
+      "sof t",
+      "s oft",
+      "▁s urface",
+      "▁sur face",
+      "▁surf ace",
+      "▁ surface",
+      "▁m om",
+      "▁mo m",
+      "▁per fect",
+      "▁perf ect",
+      "▁ perfect",
+      "is try",
+      "ist ry",
+      "istr y",
+      "i stry",
+      "ge s",
+      "g es",
+      "sin g",
+      "si ng",
+      "s ing",
+      "▁int rodu",
+      "▁intro du",
+      "▁di agn",
+      "▁dia gn",
+      "▁ diagn",
+      "▁J ohn",
+      "▁Joh n",
+      "▁Jo hn",
+      "▁ John",
+      "▁c lean",
+      "▁cl ean",
+      "▁cle an",
+      "▁ clean",
+      "C C",
+      "▁c ut",
+      "▁cu t",
+      "▁ cut",
+      "▁s td",
+      "▁st d",
+      "▁ std",
+      "g n",
+      "▁he art",
+      "▁hear t",
+      "▁ heart",
+      "▁us ers",
+      "▁use rs",
+      "▁user s",
+      "▁ users",
+      "新 的",
+      "之 前",
+      "▁de cre",
+      "▁dec re",
+      "▁ decre",
+      "▁de tails",
+      "▁det ails",
+      "▁detail s",
+      "▁ details",
+      "nA s",
+      "n As",
+      "▁P er",
+      "▁Pe r",
+      "▁ Per",
+      "ho st",
+      "h ost",
+      "▁b uy",
+      "▁bu y",
+      "▁ buy",
+      "▁H igh",
+      "▁Hi gh",
+      "▁ High",
+      "P O",
+      "▁se par",
+      "▁sep ar",
+      "▁ separ",
+      "n m",
+      "领 导",
+      "位 于",
+      "A l",
+      "上 述",
+      "in ally",
+      "inal ly",
+      "om a",
+      "o ma",
+      "东 西",
+      "we b",
+      "w eb",
+      "容 易",
+      "iu s",
+      "i us",
+      "▁e con",
+      "▁ec on",
+      "▁eco n",
+      "▁ econ",
+      "经 过",
+      "▁sh ows",
+      "▁show s",
+      "共 同",
+      "平 方",
+      "re sh",
+      "res h",
+      "r esh",
+      "▁c omb",
+      "▁com b",
+      "▁co mb",
+      "▁ comb",
+      "业 务",
+      "日 本",
+      "▁ar ch",
+      "▁arc h",
+      "▁ arch",
+      "▁n ovel",
+      "▁no vel",
+      "▁nov el",
+      "▁ novel",
+      "P E",
+      "nS t",
+      "n St",
+      "is es",
+      "ise s",
+      "i ses",
+      "▁be aut",
+      "▁t um",
+      "▁tu m",
+      "▁ tum",
+      "▁in tern",
+      "▁int ern",
+      "▁inter n",
+      "▁ intern",
+      "▁b egin",
+      "▁be gin",
+      "▁beg in",
+      "▁ begin",
+      "理 论",
+      "开 展",
+      "信 用",
+      "It em",
+      "I tem",
+      "▁t itle",
+      "▁tit le",
+      "▁ti tle",
+      "▁ title",
+      "委托 代理人",
+      "▁b reak",
+      "▁bre ak",
+      "▁ break",
+      "ta ble",
+      "tab le",
+      "t able",
+      "uc le",
+      "ucl e",
+      "u cle",
+      "▁b lood",
+      "▁bl ood",
+      "▁blo od",
+      "▁ blood",
+      "▁con cent",
+      "▁conc ent",
+      "▁conce nt",
+      "传 统",
+      "ha nd",
+      "han d",
+      "h and",
+      "之日 起",
+      "之 日起",
+      "最高 人民法院",
+      "网 站",
+      "en v",
+      "e nv",
+      "处 罚",
+      "Mo d",
+      "M od",
+      "▁g ames",
+      "▁game s",
+      "▁gam es",
+      "▁ga mes",
+      "▁ games",
+      "ca st",
+      "cas t",
+      "c ast",
+      "▁t ook",
+      "▁to ok",
+      "▁too k",
+      "on al",
+      "ona l",
+      "o nal",
+      "▁con duct",
+      "▁cond uct",
+      "▁ conduct",
+      "▁l oss",
+      "▁lo ss",
+      "▁los s",
+      "▁ loss",
+      "的 大",
+      "val id",
+      "v alid",
+      "家 庭",
+      "▁re ported",
+      "▁rep orted",
+      "▁report ed",
+      "▁ reported",
+      "月 二十",
+      "具 体",
+      "ug in",
+      "u gin",
+      "is tration",
+      "ist ration",
+      "istr ation",
+      "▁p olic",
+      "▁pol ic",
+      "▁po lic",
+      "委员 会",
+      "yd ro",
+      "y dro",
+      "ac hed",
+      "ach ed",
+      "ache d",
+      "a ched",
+      "nW h",
+      "n Wh",
+      "领 域",
+      "▁re comm",
+      "▁rec omm",
+      "▁ recomm",
+      "▁dist ributed",
+      "▁distribut ed",
+      "▁distribute d",
+      "▁ distributed",
+      "ap pend",
+      "app end",
+      "appe nd",
+      "li ght",
+      "lig ht",
+      "l ight",
+      "意 义",
+      "Cont ext",
+      "Con text",
+      "Co ntext",
+      "B C",
+      "▁success ful",
+      "enc ies",
+      "县 人民法院",
+      "is ms",
+      "ism s",
+      "i sms",
+      "▁ext ends",
+      "▁extend s",
+      "▁ extends",
+      "ub e",
+      "u be",
+      "id ge",
+      "检 察",
+      "cl ient",
+      "cli ent",
+      "有 一",
+      "撤 回",
+      "▁pre vent",
+      "▁pr event",
+      "▁prev ent",
+      "▁ prevent",
+      "▁need ed",
+      "▁ needed",
+      "▁C or",
+      "▁Co r",
+      "▁ Cor",
+      "▁signific antly",
+      "▁significant ly",
+      "▁sh owed",
+      "▁show ed",
+      "政 治",
+      "in ate",
+      "ina te",
+      "▁tem per",
+      "▁temp er",
+      "▁re leased",
+      "▁release d",
+      "▁ released",
+      "▁E n",
+      "▁ En",
+      "形 成",
+      "▁reg ion",
+      "▁ region",
+      "ge d",
+      "g ed",
+      "目 的",
+      "ie f",
+      "i ef",
+      "的 事",
+      "mod ule",
+      "mo dule",
+      "m odule",
+      "no wn",
+      "now n",
+      "n own",
+      "▁m edia",
+      "▁med ia",
+      "▁medi a",
+      "▁ media",
+      "▁s creen",
+      "▁sc reen",
+      "▁scr een",
+      "▁ screen",
+      "or ing",
+      "ori ng",
+      "o ring",
+      "ja ngo",
+      "jan go",
+      "j ango",
+      "yn c",
+      "y nc",
+      "▁equ ation",
+      "▁eq uation",
+      "▁ equation",
+      "ai d",
+      "a id",
+      "▁e mail",
+      "▁em ail",
+      "▁ email",
+      "▁acc ording",
+      "▁accord ing",
+      "主 任",
+      "▁s ays",
+      "▁sa ys",
+      "▁say s",
+      "pro t",
+      "pr ot",
+      "p rot",
+      "▁indust ry",
+      "▁indu stry",
+      "au r",
+      "a ur",
+      "▁col lection",
+      "▁coll ection",
+      "▁collect ion",
+      "▁colle ction",
+      "▁ collection",
+      "re st",
+      "res t",
+      "r est",
+      "ro s",
+      "r os",
+      "ad em",
+      "ade m",
+      "a dem",
+      "现 代",
+      "b t",
+      "▁how ever",
+      "▁know ledge",
+      "▁ knowledge",
+      "▁e ss",
+      "▁es s",
+      "▁ ess",
+      "nT here",
+      "nThe re",
+      "nTh ere",
+      "n There",
+      "ac hes",
+      "ach es",
+      "ache s",
+      "a ches",
+      "▁for mer",
+      "▁form er",
+      "▁ former",
+      "图 书",
+      "lect ed",
+      "lec ted",
+      "l ected",
+      "ri ng",
+      "rin g",
+      "r ing",
+      "gorith m",
+      "▁s hare",
+      "▁sh are",
+      "▁sha re",
+      "▁shar e",
+      "▁ share",
+      "il led",
+      "ill ed",
+      "ille d",
+      "▁pro fessional",
+      "▁profess ional",
+      "▁profession al",
+      "▁ professional",
+      "▁pl ayed",
+      "▁play ed",
+      "▁ played",
+      "▁c hem",
+      "▁ch em",
+      "▁che m",
+      "▁ chem",
+      "da t",
+      "d at",
+      "▁l earning",
+      "▁le arning",
+      "▁learn ing",
+      "▁lear ning",
+      "▁ learning",
+      "act ory",
+      "actor y",
+      "承 担",
+      "需 求",
+      "▁e ver",
+      "▁ev er",
+      "▁ ever",
+      "▁re cent",
+      "▁rec ent",
+      "▁rece nt",
+      "▁ recent",
+      "电 视",
+      "amp ions",
+      "ampion s",
+      "st atus",
+      "stat us",
+      "当 事",
+      "负 担",
+      "% )",
+      "显 示",
+      "▁| |",
+      "▁ ||",
+      "▁D ist",
+      "▁Dis t",
+      "▁Di st",
+      "▁ Dist",
+      "▁n ight",
+      "▁ni ght",
+      "▁nig ht",
+      "▁ night",
+      "▁s top",
+      "▁st op",
+      "▁sto p",
+      "▁ stop",
+      "ig ned",
+      "ign ed",
+      "igne d",
+      "re w",
+      "r ew",
+      "▁G et",
+      "▁Ge t",
+      "▁ Get",
+      "an ager",
+      "anag er",
+      "ana ger",
+      "努 力",
+      "▁T HE",
+      "▁TH E",
+      "▁ THE",
+      "属 于",
+      "▁t alk",
+      "▁tal k",
+      "▁ talk",
+      "under line",
+      "可供 执行",
+      "▁M arch",
+      "▁Mar ch",
+      "▁Marc h",
+      "▁ March",
+      "会 议",
+      "▁car eer",
+      "▁care er",
+      "P h",
+      "nH e",
+      "n He",
+      "教 师",
+      "今 天",
+      "不 到",
+      "ro ws",
+      "row s",
+      "r ows",
+      "接 受",
+      "Ev ent",
+      "Even t",
+      "E vent",
+      "tr y",
+      "t ry",
+      "▁pat ient",
+      "▁ patient",
+      "▁a gre",
+      "▁ag re",
+      "▁agr e",
+      "nu se",
+      "nus e",
+      "n use",
+      "▁m ut",
+      "▁mu t",
+      "▁ mut",
+      "数 的",
+      "ch ange",
+      "chan ge",
+      "cha nge",
+      "chang e",
+      "任 何",
+      "小 时",
+      "io l",
+      "i ol",
+      "▁b uilt",
+      "▁bu ilt",
+      "▁ built",
+      "pr ing",
+      "pri ng",
+      "p ring",
+      "▁i dea",
+      "▁ide a",
+      "▁id ea",
+      "▁ idea",
+      "公 里",
+      "▁writ ing",
+      "▁wr iting",
+      "▁ writing",
+      "人民 币",
+      "fl ow",
+      "flo w",
+      "f low",
+      "▁pub lished",
+      "▁publish ed",
+      "▁ published",
+      "帮 助",
+      ", ”",
+      "[ '",
+      "提 升",
+      "收 入",
+      "导 致",
+      "最 大",
+      "▁sh own",
+      "▁show n",
+      "▁st ory",
+      "▁sto ry",
+      "▁ story",
+      "登 记",
+      "ia ted",
+      "iate d",
+      "iat ed",
+      "i ated",
+      "ect s",
+      "ec ts",
+      "e cts",
+      "T E",
+      "确 定",
+      "pos ition",
+      "p osition",
+      "h a",
+      "▁go al",
+      "▁ goal",
+      "ch ed",
+      "che d",
+      "c hed",
+      "ol ean",
+      "ole an",
+      "o lean",
+      "ro nic",
+      "ron ic",
+      "r onic",
+      "有 的",
+      "C E",
+      "▁ar ray",
+      "▁arr ay",
+      "▁ array",
+      "毕 业",
+      "▁c ross",
+      "▁cr oss",
+      "▁cro ss",
+      "▁ cross",
+      "im ens",
+      "ime ns",
+      "imen s",
+      "i mens",
+      "ly ing",
+      "l ying",
+      "▁p ack",
+      "▁pa ck",
+      "▁pac k",
+      "▁ pack",
+      "故 事",
+      "▁B rit",
+      "▁Br it",
+      "▁Bri t",
+      "ap an",
+      "apa n",
+      "a pan",
+      "▁h alf",
+      "▁ha lf",
+      "▁hal f",
+      "▁ half",
+      "▁up date",
+      "▁upd ate",
+      "▁ update",
+      "] )",
+      "▁i r",
+      "▁ ir",
+      "▁ex pected",
+      "▁exp ected",
+      "▁expect ed",
+      "▁ expected",
+      "ee d",
+      "e ed",
+      "普 通",
+      "H T",
+      "he sis",
+      "hes is",
+      "h esis",
+      "▁p ict",
+      "▁pi ct",
+      "▁pic t",
+      "bo rn",
+      "bor n",
+      "b orn",
+      "qu ery",
+      "que ry",
+      "quer y",
+      "▁w hole",
+      "▁wh ole",
+      "▁who le",
+      "▁ whole",
+      "▁comp anies",
+      "▁in j",
+      "▁i nj",
+      "▁ inj",
+      "com es",
+      "come s",
+      "co mes",
+      "c omes",
+      "A A",
+      "st atic",
+      "stat ic",
+      "sta tic",
+      "▁pr imary",
+      "▁prim ary",
+      "▁ primary",
+      "▁w rite",
+      "▁writ e",
+      "▁wr ite",
+      "▁ write",
+      "重 庆",
+      "id ay",
+      "ida y",
+      "i day",
+      "▁n atural",
+      "▁nat ural",
+      "▁ natural",
+      "C R",
+      "Cl ass",
+      "C lass",
+      "任 务",
+      "▁s pect",
+      "▁sp ect",
+      "▁spec t",
+      "▁spe ct",
+      "▁ spect",
+      "el e",
+      "e le",
+      "▁rec ogn",
+      "▁ recogn",
+      "▁eff ective",
+      "▁effect ive",
+      "▁ effective",
+      "▁m ap",
+      "▁ma p",
+      "▁ map",
+      "\\\\ \\\\",
+      "\\\\\\ \\",
+      "\\ \\\\\\",
+      "il li",
+      "ill i",
+      "i lli",
+      "甚 至",
+      "oc ol",
+      "oco l",
+      "o col",
+      "▁B y",
+      "▁ By",
+      "▁de scribed",
+      "▁des cribed",
+      "▁describ ed",
+      "▁describe d",
+      "▁ described",
+      "送达 后",
+      "SU B",
+      "S UB",
+      "▁h ref",
+      "▁hr ef",
+      "▁ href",
+      "▁st atus",
+      "▁stat us",
+      "▁ status",
+      "th ern",
+      "ther n",
+      "the rn",
+      "▁C omp",
+      "▁Com p",
+      "▁Co mp",
+      "▁ Comp",
+      "▁a verage",
+      "▁aver age",
+      "▁ average",
+      "▁A pril",
+      "▁Ap ril",
+      "▁Apr il",
+      "▁ April",
+      "▁t w",
+      "▁ tw",
+      "bo l",
+      "b ol",
+      "空 间",
+      "d o",
+      "ter m",
+      "te rm",
+      "t erm",
+      "与被 告",
+      "与 被告",
+      "▁S im",
+      "▁Si m",
+      "▁ Sim",
+      "AN T",
+      "A NT",
+      "ser ver",
+      "serv er",
+      "serve r",
+      "s erver",
+      "▁O ff",
+      "▁Of f",
+      "▁ Off",
+      "▁G erm",
+      "▁Ge rm",
+      "▁Ger m",
+      "aut hor",
+      "au thor",
+      "auth or",
+      "a uthor",
+      "▁▁ ▁",
+      "▁ ▁▁",
+      "iv en",
+      "ive n",
+      "i ven",
+      ") (",
+      "▁e y",
+      "▁ ey",
+      "nt he",
+      "nth e",
+      "n the",
+      "▁s tat",
+      "▁st at",
+      "▁ stat",
+      "▁T est",
+      "▁Te st",
+      "▁Tes t",
+      "▁ Test",
+      "__ (",
+      "_ _(",
+      "▁add itional",
+      "▁addition al",
+      "▁ additional",
+      "本 判决",
+      "if ul",
+      "i ful",
+      "▁ap pear",
+      "▁app ear",
+      "▁appe ar",
+      "l n",
+      "man d",
+      "ma nd",
+      "m and",
+      "▁bl ack",
+      "▁ black",
+      "▁in du",
+      "▁ind u",
+      "▁con dition",
+      "▁cond ition",
+      "▁ condition",
+      "▁bir ths",
+      "▁birth s",
+      "刑 事",
+      "▁a go",
+      "▁ag o",
+      "▁ ago",
+      "百 科",
+      "▁c enter",
+      "▁cent er",
+      "▁ce nter",
+      "▁ center",
+      "ad es",
+      "ade s",
+      "a des",
+      "() );\\",
+      "()) ;\\",
+      "()); \\",
+      "( ));\\",
+      "▁se ems",
+      "▁see ms",
+      "▁seem s",
+      "▁D ep",
+      "▁De p",
+      "▁ Dep",
+      "▁d am",
+      "▁da m",
+      "▁ dam",
+      "u z",
+      "lic k",
+      "li ck",
+      "l ick",
+      "t x",
+      "金 融",
+      "sp an",
+      "s pan",
+      "旅 游",
+      "c s",
+      "I M",
+      "思 想",
+      "\"\" \"",
+      "\" \"\"",
+      "▁f em",
+      "▁fe m",
+      "▁ fem",
+      "id ed",
+      "ide d",
+      "i ded",
+      "un t",
+      "u nt",
+      "▁Jan uary",
+      "▁ January",
+      "▁t our",
+      "▁to ur",
+      "▁tou r",
+      "▁ tour",
+      "▁con cer",
+      "▁conc er",
+      "▁conce r",
+      "推 荐",
+      "▁r unning",
+      "▁run ning",
+      "▁ running",
+      "上诉 人",
+      "变 化",
+      "文 章",
+      "▁m ach",
+      "▁mac h",
+      "▁ma ch",
+      "▁ mach",
+      "s m",
+      "ar ies",
+      "ari es",
+      "arie s",
+      "a ries",
+      "pa ck",
+      "pac k",
+      "p ack",
+      "▁like ly",
+      "▁lik ely",
+      "▁ likely",
+      "▁w eight",
+      "▁we ight",
+      "▁weigh t",
+      "▁ weight",
+      "▁p ers",
+      "▁per s",
+      "▁pe rs",
+      "▁ pers",
+      "▁s ent",
+      "▁se nt",
+      "▁sen t",
+      "▁ sent",
+      "▁Se pt",
+      "▁Sep t",
+      "▁d epend",
+      "▁de pend",
+      "▁dep end",
+      "▁ depend",
+      "▁B ar",
+      "▁Ba r",
+      "▁ Bar",
+      "O ver",
+      "培 训",
+      "No de",
+      "N ode",
+      "保 险",
+      "▁com mand",
+      "▁comm and",
+      "▁comma nd",
+      "▁ command",
+      "日向 本院",
+      "日 向本院",
+      "四 川",
+      "农 业",
+      "ro wn",
+      "row n",
+      "r own",
+      "time s",
+      "tim es",
+      "ti mes",
+      "t imes",
+      "out put",
+      "▁M an",
+      "▁Ma n",
+      "▁ Man",
+      "▁di dn",
+      "▁did n",
+      "nS o",
+      "n So",
+      "▁dis play",
+      "▁disp lay",
+      "▁ display",
+      "▁estab lished",
+      "▁establish ed",
+      "▁ established",
+      "▁dist ribution",
+      "▁distribut ion",
+      "▁ distribution",
+      "nH ow",
+      "n How",
+      "关 键",
+      "osp ital",
+      "ospit al",
+      "▁g ene",
+      "▁gen e",
+      "▁ge ne",
+      "▁ gene",
+      "ens ity",
+      "ensit y",
+      "ju st",
+      "j ust",
+      "nS ee",
+      "nSe e",
+      "n See",
+      "F or",
+      "ca se",
+      "cas e",
+      "c ase",
+      "nA l",
+      "n Al",
+      "▁f ig",
+      "▁fi g",
+      "▁ fig",
+      "▁be havior",
+      "▁behav ior",
+      "▁behavi or",
+      "▁ behavior",
+      "× ×",
+      "pec ially",
+      "ek s",
+      "e ks",
+      "▁J une",
+      "▁Jun e",
+      "▁Ju ne",
+      "▁ June",
+      "li ke",
+      "lik e",
+      "l ike",
+      "lo cal",
+      "loc al",
+      "l ocal",
+      "ca l",
+      "c al",
+      "时 代",
+      "▁e tc",
+      "▁et c",
+      "▁ etc",
+      "In d",
+      "I nd",
+      "▁g e",
+      "▁ ge",
+      "il le",
+      "ill e",
+      "▁cons ist",
+      "our nal",
+      "o urnal",
+      "明 显",
+      "▁in cludes",
+      "▁includ es",
+      "▁include s",
+      "▁incl udes",
+      "▁ includes",
+      "▁ident ify",
+      "随 着",
+      "是一 个",
+      "是 一个",
+      "▁re ve",
+      "▁r eve",
+      "▁rev e",
+      "ap h",
+      "a ph",
+      "O D",
+      "不 服",
+      "视 频",
+      "▁A pp",
+      "▁Ap p",
+      "▁ App",
+      "Mess age",
+      "M essage",
+      "for t",
+      "fo rt",
+      "f ort",
+      "▁d eal",
+      "▁de al",
+      "▁ deal",
+      "▁t rying",
+      "▁tr ying",
+      "▁try ing",
+      "xi m",
+      "x im",
+      "▁e nh",
+      "▁en h",
+      "▁ enh",
+      "的 问题",
+      "pn g",
+      "p ng",
+      "▁fr ont",
+      "▁fro nt",
+      "▁ front",
+      "k g",
+      "En t",
+      "E nt",
+      "▁in stead",
+      "▁inst ead",
+      "▁ instead",
+      "经 验",
+      "▁h eld",
+      "▁he ld",
+      "▁hel d",
+      "▁ held",
+      "▁W ar",
+      "▁Wa r",
+      "▁ War",
+      "vi s",
+      "v is",
+      "▁inv olved",
+      "▁invol ved",
+      "▁involve d",
+      "事 务",
+      "▁ex act",
+      "▁ exact",
+      "icro soft",
+      "江 苏",
+      "nC h",
+      "n Ch",
+      "▁it em",
+      "▁i tem",
+      "▁ item",
+      "ak er",
+      "ake r",
+      "a ker",
+      "de scription",
+      "des cription",
+      "▁w r",
+      "▁ wr",
+      "音 乐",
+      "▁b ring",
+      "▁br ing",
+      "当事 人",
+      "▁M on",
+      "▁Mo n",
+      "▁ Mon",
+      "g y",
+      "▁c ame",
+      "▁ca me",
+      "▁cam e",
+      "▁ac cept",
+      "▁ accept",
+      "A d",
+      "▁a ud",
+      "▁au d",
+      "▁ aud",
+      "ty p",
+      "t yp",
+      "▁u til",
+      "▁ut il",
+      "▁ util",
+      "▁t aking",
+      "▁ta king",
+      "▁tak ing",
+      "▁ taking",
+      "重 点",
+      "关 注",
+      "▁a w",
+      "▁ aw",
+      "in ing",
+      "ini ng",
+      "inin g",
+      "i ning",
+      "ist ory",
+      "istor y",
+      "isto ry",
+      "i story",
+      "性 的",
+      "eo ple",
+      "e ople",
+      "l a",
+      "EC T",
+      "E CT",
+      "条 之规定",
+      "summ ary",
+      "s ummary",
+      "▁i ll",
+      "▁il l",
+      "▁ ill",
+      "费 用",
+      "一 点",
+      "ag ed",
+      "age d",
+      "a ged",
+      "res pon",
+      "resp on",
+      "▁M at",
+      "▁Ma t",
+      "▁ Mat",
+      "▁eff ort",
+      "F r",
+      "▁of fers",
+      "▁off ers",
+      "▁offer s",
+      "距 离",
+      "▁happ en",
+      "▁hap pen",
+      "cur rent",
+      "curr ent",
+      "c urrent",
+      "▁t aken",
+      "▁take n",
+      "▁ta ken",
+      "▁tak en",
+      "▁E l",
+      "▁ El",
+      "ap ter",
+      "apt er",
+      "▁M c",
+      "▁ Mc",
+      "▁f eed",
+      "▁fe ed",
+      "▁fee d",
+      "▁ feed",
+      "S e",
+      "cin g",
+      "ci ng",
+      "c ing",
+      "▁T rans",
+      "▁Tr ans",
+      "▁Tra ns",
+      "▁ Trans",
+      "ov ember",
+      "ove mber",
+      "ig r",
+      "i gr",
+      "正 常",
+      "pro ject",
+      "proj ect",
+      "达 到",
+      "In fo",
+      "Inf o",
+      "]( /",
+      "] (/",
+      "wit ch",
+      "w itch",
+      "C D",
+      "▁mult i",
+      "▁mul ti",
+      "▁ multi",
+      "的 小",
+      "▁Gener al",
+      "▁Gen eral",
+      "▁Gene ral",
+      "▁ General",
+      "▁com bin",
+      "▁comb in",
+      "▁ combin",
+      "▁r a",
+      "▁ ra",
+      "▁l ines",
+      "▁li nes",
+      "▁line s",
+      "▁lin es",
+      "▁ lines",
+      "ra d",
+      "r ad",
+      "con text",
+      "cont ext",
+      "co ntext",
+      "un k",
+      "u nk",
+      "air s",
+      "ai rs",
+      "a irs",
+      "ol es",
+      "ole s",
+      "o les",
+      "() ,",
+      "( ),",
+      "▁some one",
+      "▁W est",
+      "▁We st",
+      "▁Wes t",
+      "▁ West",
+      "制 作",
+      "▁us eful",
+      "▁use ful",
+      "T R",
+      "▁S m",
+      "▁ Sm",
+      "no v",
+      "n ov",
+      "pr i",
+      "p ri",
+      "Al l",
+      "A ll",
+      "}_ {",
+      "} _{",
+      "io s",
+      "i os",
+      "▁ 记",
+      "() :\\",
+      "(): \\",
+      "( ):\\",
+      "▁t ell",
+      "▁te ll",
+      "▁tel l",
+      "商 品",
+      "▁diff icult",
+      "o e",
+      "▁under standing",
+      "▁understand ing",
+      "广 东",
+      "Ha nd",
+      "H and",
+      "调 查",
+      "法律 规定",
+      "oc h",
+      "o ch",
+      "▁b ox",
+      "▁bo x",
+      "▁ box",
+      "▁e lement",
+      "▁el ement",
+      "▁ele ment",
+      "▁elem ent",
+      "▁ element",
+      "▁l abel",
+      "▁la bel",
+      "▁lab el",
+      "▁ label",
+      "fin d",
+      "fi nd",
+      "f ind",
+      "▁bec ame",
+      "▁A fr",
+      "▁Af r",
+      "▁writ ten",
+      "▁wr itten",
+      "▁ written",
+      "号 民事",
+      "al pha",
+      "alph a",
+      "▁con sum",
+      "▁cons um",
+      "▁ consum",
+      "▁com pl",
+      "▁comp l",
+      "▁ compl",
+      "ic ip",
+      "ici p",
+      "}$ \\",
+      "} $\\",
+      "Eq u",
+      "E qu",
+      "ot es",
+      "ote s",
+      "o tes",
+      "ui d",
+      "u id",
+      "▁exp eri",
+      "▁exper i",
+      ")\\ \\",
+      ") \\\\",
+      "▁ent ire",
+      "▁wh ite",
+      "▁whit e",
+      "▁ white",
+      "权 利",
+      "方 案",
+      "ou ch",
+      "o uch",
+      "主 义",
+      "▁n ational",
+      "▁nation al",
+      "▁nat ional",
+      "▁ national",
+      "la tion",
+      "lat ion",
+      "l ation",
+      "不 仅",
+      "pa ge",
+      "pag e",
+      "p age",
+      "▁I nter",
+      "▁In ter",
+      "▁Int er",
+      "▁ Inter",
+      "fo nt",
+      "fon t",
+      "f ont",
+      "it ing",
+      "iti ng",
+      "itin g",
+      "i ting",
+      "▁per formed",
+      "▁perform ed",
+      "▁perfor med",
+      "ain t",
+      "ai nt",
+      "a int",
+      "▁n p",
+      "▁ np",
+      "▁up d",
+      "▁u pd",
+      "▁m ind",
+      "▁min d",
+      "▁mi nd",
+      "▁ mind",
+      "你 们",
+      "商 业",
+      "公 开",
+      "理 由",
+      "▁in sp",
+      "▁i nsp",
+      "▁ins p",
+      "▁ insp",
+      "▁d ouble",
+      "▁do uble",
+      "▁dou ble",
+      "▁doub le",
+      "▁ double",
+      "立 案",
+      "▁p rior",
+      "▁pr ior",
+      "▁pri or",
+      "▁ prior",
+      "以 后",
+      "▁ 其",
+      "▁e stim",
+      "▁est im",
+      "▁es tim",
+      "▁ estim",
+      "al ls",
+      "all s",
+      "▁M ark",
+      "▁Mar k",
+      "▁Ma rk",
+      "▁ Mark",
+      "▁\\\\ \\\\",
+      "▁\\ \\\\\\",
+      "▁\\\\\\ \\",
+      "▁ \\\\\\\\",
+      "ch eck",
+      "che ck",
+      "c heck",
+      "b o",
+      "▁f amil",
+      "▁fam il",
+      "▁fa mil",
+      "ov ed",
+      "ove d",
+      "o ved",
+      "ce mber",
+      "c ember",
+      "安 徽",
+      "特 点",
+      "▁J uly",
+      "▁Jul y",
+      "▁Ju ly",
+      "▁ July",
+      "发 布",
+      "Dat e",
+      "Da te",
+      "D ate",
+      "▁conn ect",
+      "▁ connect",
+      "ri p",
+      "r ip",
+      "一 种",
+      "代 码",
+      "检 查",
+      "▁p urch",
+      "▁pur ch",
+      "文 学",
+      "en ced",
+      "ence d",
+      "enc ed",
+      "ave s",
+      "av es",
+      "a ves",
+      "ha m",
+      "h am",
+      "U p",
+      "re lated",
+      "rel ated",
+      "▁act ive",
+      "▁activ e",
+      "▁ active",
+      "ue l",
+      "u el",
+      "离 婚",
+      "▁Oct ober",
+      "▁ October",
+      "▁e ducation",
+      "▁educ ation",
+      "▁edu cation",
+      "▁ education",
+      "造 成",
+      "▁ 行",
+      "Ap p",
+      "A pp",
+      "▁e specially",
+      "▁es pecially",
+      "▁ especially",
+      "at ives",
+      "ative s",
+      "ati ves",
+      "ativ es",
+      "▁un d",
+      "▁u nd",
+      "▁ und",
+      "uf act",
+      "u fact",
+      "实 验",
+      "I s",
+      "Co de",
+      "Cod e",
+      "C ode",
+      "▁s uff",
+      "▁su ff",
+      "▁n ews",
+      "▁ne ws",
+      "▁new s",
+      "▁ news",
+      "an i",
+      "a ni",
+      "T P",
+      "减 刑",
+      "▁d en",
+      "▁de n",
+      "▁ den",
+      "ot tom",
+      "ott om",
+      "otto m",
+      "拥 有",
+      "▁cl ub",
+      "▁ club",
+      "▁s em",
+      "▁se m",
+      "▁ sem",
+      "▁re plac",
+      "名 称",
+      "▁m edical",
+      "▁med ical",
+      "▁medi cal",
+      "▁medic al",
+      "▁ medical",
+      "▁R ed",
+      "▁Re d",
+      "▁ Red",
+      "ail y",
+      "ai ly",
+      "a ily",
+      "▁c annot",
+      "▁can not",
+      "▁ca nnot",
+      "▁cann ot",
+      "▁ cannot",
+      "▁P ark",
+      "▁Par k",
+      "▁Pa rk",
+      "▁ Park",
+      "▁A c",
+      "▁ Ac",
+      "up t",
+      "u pt",
+      "▁A ct",
+      "▁Ac t",
+      "▁ Act",
+      "▁cl aim",
+      "▁ claim",
+      "语 言",
+      "现 场",
+      "▁t ax",
+      "▁ta x",
+      "▁ tax",
+      "卫 生",
+      "mi ss",
+      "mis s",
+      "m iss",
+      "▁s af",
+      "▁sa f",
+      "▁ saf",
+      "▁ex pect",
+      "▁exp ect",
+      "▁ expect",
+      "▁A ugust",
+      "▁Aug ust",
+      "▁Au gust",
+      "▁ August",
+      "ry pt",
+      "▁f ocus",
+      "▁foc us",
+      "▁fo cus",
+      "▁ focus",
+      "le n",
+      "l en",
+      "▁p ret",
+      "▁pre t",
+      "▁pr et",
+      "▁ pret",
+      "▁d est",
+      "▁de st",
+      "▁des t",
+      "▁ dest",
+      "al id",
+      "ali d",
+      "▁reg ular",
+      "▁regul ar",
+      "▁ regular",
+      "▁f ast",
+      "▁fa st",
+      "▁ fast",
+      "al ed",
+      "ale d",
+      "a led",
+      "▁C enter",
+      "▁Cent er",
+      "▁Ce nter",
+      "▁ Center",
+      "▁fil ms",
+      "▁film s",
+      "▁ films",
+      "O L",
+      "▁Se ptember",
+      "▁Sept ember",
+      "▁ September",
+      "ee s",
+      "e es",
+      "writ e",
+      "wr ite",
+      "w rite",
+      "▁l ocation",
+      "▁loc ation",
+      "▁lo cation",
+      "▁ location",
+      "▁( \\",
+      "▁ (\\",
+      "▁re cept",
+      "▁rece pt",
+      "▁r ound",
+      "▁ro und",
+      "▁rou nd",
+      "▁ round",
+      "▁d ifference",
+      "▁dif ference",
+      "▁differ ence",
+      "▁ difference",
+      "pro cess",
+      "proc ess",
+      "▁c ause",
+      "▁ca use",
+      "▁caus e",
+      "▁ cause",
+      "最高人民法院 关于",
+      "file s",
+      "fil es",
+      "fi les",
+      "f iles",
+      "▁s peed",
+      "▁sp eed",
+      "▁spe ed",
+      "▁ speed",
+      "▁p rivate",
+      "▁pr ivate",
+      "▁priv ate",
+      "▁ private",
+      "经 理",
+      "jo in",
+      "j oin",
+      "anc ial",
+      "物 业",
+      "ave n",
+      "av en",
+      "a ven",
+      "▁t ests",
+      "▁te sts",
+      "▁test s",
+      "▁ tests",
+      "he mat",
+      "hem at",
+      "hema t",
+      "h emat",
+      "形 式",
+      "al ign",
+      "ali gn",
+      "ce ed",
+      "c eed",
+      "▁pl atform",
+      "▁plat form",
+      "▁ platform",
+      "bo r",
+      "b or",
+      "ar ds",
+      "ard s",
+      "▁pr actice",
+      "▁pract ice",
+      "▁ practice",
+      "▁b ound",
+      "▁bo und",
+      "▁bou nd",
+      "▁ bound",
+      "▁cl ient",
+      "▁cli ent",
+      "▁ client",
+      "▁observ ed",
+      "▁observe d",
+      "▁obs erved",
+      "比 如",
+      "▁prob ably",
+      "▁ probably",
+      "十 五",
+      "保 证",
+      "对 方",
+      "ro ph",
+      "rop h",
+      "r oph",
+      "▁im ages",
+      "▁image s",
+      "▁imag es",
+      "▁ images",
+      "▁off ice",
+      "▁offic e",
+      "▁ office",
+      "an did",
+      "and id",
+      "andi d",
+      "▁con tract",
+      "▁contr act",
+      "▁ contract",
+      "▁p layer",
+      "▁pl ayer",
+      "▁play er",
+      "▁ player",
+      "▁r ather",
+      "▁rat her",
+      "▁ra ther",
+      "▁cons idered",
+      "▁consider ed",
+      "ri al",
+      "ria l",
+      "r ial",
+      "▁l iving",
+      "▁li ving",
+      "▁liv ing",
+      "▁ living",
+      "ef ore",
+      "e fore",
+      "ot al",
+      "ota l",
+      "o tal",
+      "▁r oot",
+      "▁ro ot",
+      "▁ root",
+      "▁w eeks",
+      "▁we eks",
+      "▁week s",
+      "▁ weeks",
+      "▁v ill",
+      "▁vi ll",
+      "▁ vill",
+      "方 向",
+      "▁un it",
+      "▁u nit",
+      "▁uni t",
+      "▁ unit",
+      "▁b and",
+      "▁ban d",
+      "▁ba nd",
+      "▁ band",
+      "今 年",
+      "▁act ivities",
+      "▁activ ities",
+      "▁ activities",
+      "▁h ope",
+      "▁hop e",
+      "▁ho pe",
+      "本院 于",
+      "全 部",
+      "ie t",
+      "i et",
+      "ve y",
+      "v ey",
+      "▁Eng lish",
+      "▁ English",
+      "ail ed",
+      "ai led",
+      "a iled",
+      "le ep",
+      "lee p",
+      "l eep",
+      "ni f",
+      "n if",
+      "cal l",
+      "ca ll",
+      "c all",
+      "ne xt",
+      "nex t",
+      "n ext",
+      "E n",
+      "▁f it",
+      "▁fi t",
+      "▁ fit",
+      "Bu ild",
+      "B uild",
+      "▁d efined",
+      "▁def ined",
+      "▁define d",
+      "▁defin ed",
+      "▁ defined",
+      "B A",
+      "▁w ent",
+      "▁we nt",
+      "ev ent",
+      "even t",
+      "eve nt",
+      "e vent",
+      "pi ng",
+      "pin g",
+      "p ing",
+      "i j",
+      "▁f riend",
+      "▁fri end",
+      "▁ friend",
+      "全 面",
+      "oc al",
+      "oca l",
+      "o cal",
+      "▁ab ility",
+      "▁ ability",
+      "▁s olutions",
+      "▁sol utions",
+      "▁solution s",
+      "▁ solutions",
+      "数 字",
+      "并 且",
+      "▁S ub",
+      "▁Su b",
+      "▁ Sub",
+      "iz er",
+      "ize r",
+      "i zer",
+      "ia nce",
+      "ian ce",
+      "i ance",
+      "才 能",
+      "nn amespace",
+      "nname space",
+      "n namespace",
+      "ke n",
+      "k en",
+      "▁M ore",
+      "▁Mor e",
+      "▁Mo re",
+      "▁ More",
+      "▁G roup",
+      "▁Gr oup",
+      "▁Gro up",
+      "▁ Group",
+      "▁for mat",
+      "▁form at",
+      "▁ format",
+      "首 先",
+      "左 右",
+      "▁acc ur",
+      "▁ac cur",
+      "▁ accur",
+      "RE F",
+      "R EF",
+      "lo ck",
+      "loc k",
+      "l ock",
+      "au l",
+      "a ul",
+      "▁obt ained",
+      "▁obtain ed",
+      "ch ar",
+      "cha r",
+      "c har",
+      "▁f ace",
+      "▁fac e",
+      "▁fa ce",
+      "▁ face",
+      "农 民",
+      "ne s",
+      "n es",
+      "De s",
+      "D es",
+      "▁H ouse",
+      "▁Ho use",
+      "▁Hou se",
+      "▁ House",
+      "▁P res",
+      "▁Pr es",
+      "▁Pre s",
+      "▁ Pres",
+      "be d",
+      "b ed",
+      "at ur",
+      "atu r",
+      "a tur",
+      "nu ll",
+      "n ull",
+      "▁m id",
+      "▁mi d",
+      "▁ mid",
+      "H e",
+      "ect ive",
+      "res ponse",
+      "respon se",
+      "respons e",
+      "身 体",
+      "▁N ovember",
+      "▁Nov ember",
+      "▁ November",
+      "▁st ress",
+      "▁str ess",
+      "▁stre ss",
+      "▁ stress",
+      "▁s il",
+      "▁si l",
+      "▁ sil",
+      "iqu es",
+      "ique s",
+      "iq ues",
+      "i ques",
+      "▁m aint",
+      "▁main t",
+      "▁ma int",
+      "▁ maint",
+      "▁every thing",
+      "▁P ublic",
+      "▁Pub lic",
+      "▁Pu blic",
+      "▁ Public",
+      "oc ket",
+      "ock et",
+      "当 然",
+      "i i",
+      "▁p rec",
+      "▁pre c",
+      "▁pr ec",
+      "▁ prec",
+      "▁P at",
+      "▁Pa t",
+      "▁ Pat",
+      "▁an aly",
+      "▁anal y",
+      "▁ana ly",
+      "q l",
+      "一 年",
+      "都 有",
+      "Cl ient",
+      "Cli ent",
+      "' )",
+      "au gh",
+      "aug h",
+      "a ugh",
+      "▁com pos",
+      "▁comp os",
+      "No t",
+      "N ot",
+      "现 已",
+      "▁par ameters",
+      "▁paramet ers",
+      "▁parameter s",
+      "▁param eters",
+      "▁ parameters",
+      "lo at",
+      "ool s",
+      "oo ls",
+      "o ols",
+      "( [",
+      "准 备",
+      "▁reg ul",
+      "▁st aff",
+      "▁ staff",
+      "▁n ode",
+      "▁no de",
+      "▁nod e",
+      "▁ node",
+      "▁pop ular",
+      "▁popul ar",
+      "▁ popular",
+      "in ition",
+      "init ion",
+      "ini tion",
+      "E L",
+      "▁u t",
+      "▁ ut",
+      "nO n",
+      "n On",
+      "▁A g",
+      "▁ Ag",
+      "▁all ows",
+      "▁allow s",
+      "▁f ul",
+      "▁fu l",
+      "▁ ful",
+      "▁press ure",
+      "▁ pressure",
+      "▁S ee",
+      "▁Se e",
+      "▁ See",
+      "T H",
+      "ap ping",
+      "app ing",
+      "a pping",
+      "儿 童",
+      "ac c",
+      "a cc",
+      "▁L a",
+      "▁ La",
+      "ar gin",
+      "arg in",
+      "▁rec ently",
+      "▁recent ly",
+      "▁Tr ue",
+      "▁ True",
+      "ad o",
+      "a do",
+      "▁ ]",
+      "说 明",
+      "▁high ly",
+      "▁st ore",
+      "▁sto re",
+      "▁ store",
+      "en u",
+      "e nu",
+      "▁g lobal",
+      "▁glob al",
+      "▁glo bal",
+      "▁ global",
+      "▁d ynam",
+      "▁dy nam",
+      "▁ dynam",
+      "S D",
+      "阅 读",
+      "la st",
+      "las t",
+      "l ast",
+      "参 数",
+      "▁ap plications",
+      "▁applic ations",
+      "▁application s",
+      "▁appl ications",
+      "▁ applications",
+      "Res ponse",
+      "Respon se",
+      "he me",
+      "hem e",
+      "h eme",
+      "就 会",
+      "ch e",
+      "c he",
+      "▁n amed",
+      "▁name d",
+      "▁na med",
+      "▁nam ed",
+      "▁ named",
+      "P L",
+      "▁l en",
+      "▁le n",
+      "▁ len",
+      "▁s us",
+      "▁su s",
+      "▁ sus",
+      "\\' );\\",
+      "\\') ;\\",
+      "\\'); \\",
+      "\\ ');\\",
+      "s ource",
+      "de c",
+      "d ec",
+      "协 议",
+      "▁l ate",
+      "▁la te",
+      "▁lat e",
+      "▁ late",
+      "g s",
+      "Com ponent",
+      "智 能",
+      "▁for ward",
+      "▁ forward",
+      "▁s urg",
+      "▁su rg",
+      "▁sur g",
+      "en ses",
+      "ens es",
+      "ense s",
+      "也 有",
+      "bu g",
+      "b ug",
+      "十 九",
+      "本次 执行",
+      "充 分",
+      "▁S w",
+      "▁ Sw",
+      "▁De cember",
+      "▁Dec ember",
+      "▁ December",
+      "▁custom ers",
+      "▁customer s",
+      "▁cust omers",
+      "▁ customers",
+      "实 施",
+      "▁mat rix",
+      "▁ matrix",
+      "Cont roller",
+      "Control ler",
+      "ic o",
+      "i co",
+      "▁in tr",
+      "▁int r",
+      "▁i ntr",
+      "▁ intr",
+      "进 一步",
+      "▁in side",
+      "▁ins ide",
+      "▁ inside",
+      "▁direct ly",
+      "▁Le ague",
+      "▁ League",
+      "( {",
+      "In ter",
+      "Int er",
+      "I nter",
+      "▁st ay",
+      "▁ stay",
+      "▁s erver",
+      "▁serv er",
+      "▁ser ver",
+      "▁serve r",
+      "▁ server",
+      "▁p ackage",
+      "▁pack age",
+      "▁ package",
+      "// //",
+      "/// /",
+      "/ ///",
+      "▁c lick",
+      "▁cl ick",
+      "▁cli ck",
+      "▁ click",
+      "com mit",
+      "comm it",
+      "创 建",
+      "▁a cid",
+      "▁ac id",
+      "▁ acid",
+      "ampions hip",
+      "ampion ship",
+      "加 强",
+      "▁v ers",
+      "▁ve rs",
+      "▁ver s",
+      "▁ vers",
+      "▁off icial",
+      "▁offic ial",
+      "▁ official",
+      "but ton",
+      "b utton",
+      "超 过",
+      "ho od",
+      "h ood",
+      "十 三",
+      "▁con cept",
+      "▁conce pt",
+      "▁ concept",
+      "▁t own",
+      "▁to wn",
+      "▁tow n",
+      "▁ town",
+      "▁s ection",
+      "▁se ction",
+      "▁sec tion",
+      "▁ section",
+      "ce ll",
+      "cel l",
+      "c ell",
+      "人 们",
+      "mod els",
+      "model s",
+      "mode ls",
+      "民 族",
+      "▁w alk",
+      "▁wal k",
+      "▁ walk",
+      "▁g irl",
+      "▁gi rl",
+      "▁gir l",
+      "▁ girl",
+      "r l",
+      "re nch",
+      "ren ch",
+      "r ench",
+      "▁in ternational",
+      "▁inter national",
+      "▁intern ational",
+      "▁b ed",
+      "▁be d",
+      "▁ bed",
+      "el ine",
+      "eli ne",
+      "elin e",
+      "e line",
+      "▁J apan",
+      "▁Ja pan",
+      "▁ Japan",
+      "规 划",
+      "le vel",
+      "lev el",
+      "' :",
+      "建 立",
+      "▁contin ue",
+      "▁continu e",
+      "▁ continue",
+      "是 在",
+      "审 查",
+      "下 的",
+      "同 学",
+      "ip e",
+      "i pe",
+      "办 公",
+      "tr ue",
+      "I F",
+      "满 足",
+      "▁at tempt",
+      "▁att empt",
+      "▁ attempt",
+      "公 安",
+      "优 秀",
+      "ro me",
+      "rom e",
+      "r ome",
+      "ol der",
+      "old er",
+      "Sys tem",
+      "Sy stem",
+      "S ystem",
+      "ype s",
+      "yp es",
+      "y pes",
+      "▁w ays",
+      "▁way s",
+      "▁wa ys",
+      "▁ ways",
+      "on don",
+      "ond on",
+      "ondo n",
+      "o ndon",
+      "ro te",
+      "rot e",
+      "r ote",
+      "▁us es",
+      "▁use s",
+      "▁u ses",
+      "▁ uses",
+      "De f",
+      "D ef",
+      "▁ 长",
+      "▁any thing",
+      "g l",
+      "▁sec urity",
+      "▁ security",
+      "▁N e",
+      "▁ Ne",
+      "is f",
+      "i sf",
+      "几 个",
+      "▁incre asing",
+      "em ail",
+      "ema il",
+      "e mail",
+      "▁e nsure",
+      "▁ens ure",
+      "▁ ensure",
+      "▁al most",
+      "▁ almost",
+      "cl osed",
+      "close d",
+      "▁d ownload",
+      "▁down load",
+      "▁ download",
+      "评 价",
+      "O P",
+      "▁l iter",
+      "▁li ter",
+      "▁lit er",
+      "▁ liter",
+      "▁P ost",
+      "▁Po st",
+      "▁Pos t",
+      "▁ Post",
+      "d s",
+      "措 施",
+      "if ier",
+      "ifi er",
+      "▁hy per",
+      "▁hyp er",
+      "▁hype r",
+      "▁ hyper",
+      "判 断",
+      "lo ng",
+      "lon g",
+      "l ong",
+      "▁s ites",
+      "▁sit es",
+      "▁site s",
+      "▁si tes",
+      "▁ sites",
+      "只 要",
+      "未 来",
+      "com ing",
+      "co ming",
+      "c oming",
+      "re et",
+      "ree t",
+      "M M",
+      "期 限",
+      "▁C r",
+      "▁ Cr",
+      "七 年",
+      "▁ ‘",
+      "▁com ment",
+      "▁comm ent",
+      "▁ comment",
+      "P r",
+      "ac ity",
+      "aci ty",
+      "a city",
+      "ab led",
+      "able d",
+      "abl ed",
+      "a bled",
+      "问题 的",
+      "考 虑",
+      "体 系",
+      "驾 驶",
+      "▁su rv",
+      "▁sur v",
+      "每 天",
+      "br uary",
+      "▁cont ains",
+      "▁contain s",
+      "▁ contains",
+      "er ation",
+      "era tion",
+      "e ration",
+      "▁d raw",
+      "▁dr aw",
+      "▁ draw",
+      "▁le ading",
+      "▁lead ing",
+      "▁ leading",
+      "D e",
+      "第 四",
+      "报 告",
+      "Ar ray",
+      "Arr ay",
+      "第一百 四十",
+      "Wi th",
+      "W ith",
+      "▁s te",
+      "▁st e",
+      "▁ ste",
+      "bo n",
+      "b on",
+      "每 个",
+      "性 能",
+      "▁con stant",
+      "▁const ant",
+      "▁ constant",
+      "▁qu ite",
+      "▁q uite",
+      "▁quit e",
+      "▁qui te",
+      "▁v eh",
+      "▁ve h",
+      "▁ veh",
+      "地 址",
+      "▁pro ced",
+      "▁proc ed",
+      "▁w ond",
+      "▁won d",
+      "▁wo nd",
+      "▁H ere",
+      "▁He re",
+      "▁Her e",
+      "▁ Here",
+      "▁t old",
+      "▁to ld",
+      "平 均",
+      "re act",
+      "rea ct",
+      "iz es",
+      "ize s",
+      "i zes",
+      "调 解",
+      "nA nd",
+      "nAn d",
+      "n And",
+      "▁ev idence",
+      "▁evid ence",
+      "▁r oad",
+      "▁ro ad",
+      "▁ road",
+      "除 了",
+      "▁t akes",
+      "▁take s",
+      "▁ta kes",
+      "▁tak es",
+      "▁se em",
+      "▁see m",
+      "给 出",
+      "▁st ation",
+      "▁stat ion",
+      "▁ station",
+      "公 诉",
+      "输 入",
+      "▁a rr",
+      "▁ar r",
+      "▁ arr",
+      "▁sim ply",
+      "▁simpl y",
+      "▁dev ice",
+      "▁ device",
+      "fi t",
+      "f it",
+      "amb da",
+      "▁g round",
+      "▁gr ound",
+      "▁gro und",
+      "▁ ground",
+      "oc ker",
+      "ock er",
+      "know n",
+      "kn own",
+      "k nown",
+      "▁U p",
+      "▁ Up",
+      "▁S et",
+      "▁Se t",
+      "▁ Set",
+      "ar a",
+      "a ra",
+      ") *",
+      "▁inter face",
+      "▁interf ace",
+      "▁ interface",
+      "b f",
+      "▁C D",
+      "▁ CD",
+      "▁ne uro",
+      "▁neur o",
+      "▁ neuro",
+      "▁P R",
+      "▁ PR",
+      "S C",
+      "oo n",
+      "o on",
+      "检 测",
+      "▁s equence",
+      "▁sequ ence",
+      "▁ sequence",
+      "▁w ait",
+      "▁wa it",
+      "▁ wait",
+      "制 度",
+      "▁D ec",
+      "▁De c",
+      "▁ Dec",
+      "▁it ems",
+      "▁item s",
+      "▁ items",
+      "pt y",
+      "p ty",
+      "▁nec essary",
+      "▁necess ary",
+      "交 流",
+      "right arrow",
+      "sy ch",
+      "s ych",
+      "▁con tr",
+      "▁cont r",
+      "▁co ntr",
+      "▁ contr",
+      "oc iation",
+      "oci ation",
+      "不知 道",
+      "不 知道",
+      "oc r",
+      "o cr",
+      "▁bel ieve",
+      "▁belie ve",
+      "培 养",
+      "def ine",
+      "de fine",
+      "▁g as",
+      "▁ga s",
+      "▁ gas",
+      "▁Col lege",
+      "有 些",
+      "▁t ravel",
+      "▁tr avel",
+      "▁tra vel",
+      "▁trav el",
+      "▁ travel",
+      "徒 刑",
+      "立 即",
+      "▁b rain",
+      "▁br ain",
+      "▁bra in",
+      "▁ brain",
+      "▁ther apy",
+      "▁therap y",
+      "▁ therapy",
+      "▁D ata",
+      "▁Dat a",
+      "▁Da ta",
+      "▁ Data",
+      "参 与",
+      "考 试",
+      "Wh at",
+      "W hat",
+      "Bu t",
+      "B ut",
+      "A s",
+      "▁R uss",
+      "▁Ru ss",
+      "▁Rus s",
+      "▁e lements",
+      "▁el ements",
+      "▁element s",
+      "▁ele ments",
+      "▁elem ents",
+      "▁ elements",
+      "期 徒刑",
+      "▁qu e",
+      "▁q ue",
+      "▁ que",
+      "五 年",
+      "那 个",
+      "▁c apt",
+      "▁cap t",
+      "▁ca pt",
+      "▁ capt",
+      "B e",
+      "▁project s",
+      "▁proj ects",
+      "▁ projects",
+      "ok e",
+      "o ke",
+      "本院 在",
+      "不 可",
+      "▁s ample",
+      "▁sam ple",
+      "▁ sample",
+      "ar ter",
+      "art er",
+      "arte r",
+      "合同 纠纷一案",
+      "合同纠纷 一案",
+      "▁respect ively",
+      "▁respective ly",
+      "ot ic",
+      "oti c",
+      "o tic",
+      "本裁定 送达后",
+      "cs s",
+      "c ss",
+      "▁T hen",
+      "▁The n",
+      "▁Th en",
+      "▁ Then",
+      "联 网",
+      "人 物",
+      "只 能",
+      "% ,",
+      "▁f riends",
+      "▁fri ends",
+      "▁friend s",
+      "▁ friends",
+      "▁H is",
+      "▁Hi s",
+      "▁ His",
+      "mer c",
+      "me rc",
+      "m erc",
+      "程 度",
+      "ic an",
+      "ica n",
+      "i can",
+      "▁d irection",
+      "▁di rection",
+      "▁direct ion",
+      "▁dir ection",
+      "▁dire ction",
+      "▁ direction",
+      "he ma",
+      "hem a",
+      "h ema",
+      "() )\\",
+      "()) \\",
+      "( ))\\",
+      "▁A ng",
+      "▁An g",
+      "▁ Ang",
+      "▁a nim",
+      "▁an im",
+      "▁ anim",
+      "▁ch oose",
+      "▁cho ose",
+      "▁ choose",
+      "Ne w",
+      "N ew",
+      "▁com mit",
+      "▁comm it",
+      "▁ commit",
+      "当 时",
+      "事务 所",
+      "的 财产",
+      "机 械",
+      "六 年",
+      "▁work ed",
+      "ly mp",
+      "ro ot",
+      "r oot",
+      "P C",
+      "速 度",
+      "数 量",
+      "I I",
+      "▁p roduced",
+      "▁produ ced",
+      "▁produce d",
+      "▁prod uced",
+      "▁ produced",
+      "at ively",
+      "ative ly",
+      "ati vely",
+      "ativ ely",
+      "改 革",
+      "it ude",
+      "itud e",
+      "itu de",
+      "ra z",
+      "r az",
+      "▁In ternational",
+      "▁Intern ational",
+      "▁Inter national",
+      "▁ International",
+      "市中 级人民法院",
+      "市 中级人民法院",
+      "] :",
+      "▁b oard",
+      "▁bo ard",
+      "▁ board",
+      "▁l atest",
+      "▁la test",
+      "▁late st",
+      "▁lat est",
+      "▁ latest",
+      "O f",
+      "▁be autiful",
+      "▁beaut iful",
+      "if orn",
+      "av id",
+      "avi d",
+      "a vid",
+      "st er",
+      "ste r",
+      "s ter",
+      "nW hen",
+      "nWh en",
+      "n When",
+      "tu al",
+      "t ual",
+      "es tern",
+      "est ern",
+      "ester n",
+      "este rn",
+      "El ement",
+      "Elem ent",
+      "Ele ment",
+      "E lement",
+      "ig en",
+      "ige n",
+      "i gen",
+      "本次 执行程序",
+      "本次执行 程序",
+      "▁E d",
+      "▁ Ed",
+      "math bb",
+      "ur ance",
+      "ura nce",
+      "uran ce",
+      "u rance",
+      "▁de riv",
+      "▁der iv",
+      "cl us",
+      "ol ar",
+      "ola r",
+      "o lar",
+      "▁in stance",
+      "▁inst ance",
+      "▁ instance",
+      "▁l ooks",
+      "▁look s",
+      "▁lo oks",
+      ",\\ \\",
+      ", \\\\",
+      "mor e",
+      "mo re",
+      "m ore",
+      "ing ton",
+      "an cy",
+      "anc y",
+      "a ncy",
+      "▁e mer",
+      "▁em er",
+      "认 定",
+      "课 程",
+      "▁estab lish",
+      "▁ establish",
+      "▁tem perature",
+      "▁temper ature",
+      "▁ temperature",
+      "电 话",
+      "ue d",
+      "u ed",
+      "ia ns",
+      "ian s",
+      "i ans",
+      "自 治",
+      "▁s oon",
+      "▁so on",
+      "▁soo n",
+      "▁ soon",
+      "▁C opyright",
+      "▁Copy right",
+      "▁ Copyright",
+      "制 造",
+      "gr ad",
+      "gra d",
+      "g rad",
+      "理 解",
+      "av y",
+      "a vy",
+      "}{ \\\\",
+      "} {\\\\",
+      "▁is n",
+      "▁i sn",
+      "ex p",
+      "e xp",
+      "▁p rep",
+      "▁pre p",
+      "▁pr ep",
+      "▁ prep",
+      "心 理",
+      "媒 体",
+      "英 语",
+      "▁b ad",
+      "▁ba d",
+      "▁ bad",
+      "▁P lease",
+      "▁Ple ase",
+      "▁ Please",
+      "成 本",
+      "▁al tern",
+      "▁alter n",
+      "▁alt ern",
+      "▁ altern",
+      "▁S ch",
+      "▁Sc h",
+      "▁ Sch",
+      "nT r",
+      "n Tr",
+      "▁pl ac",
+      "▁ plac",
+      "▁> >",
+      "▁ >>",
+      "保 持",
+      "▁b asis",
+      "▁bas is",
+      "▁ba sis",
+      "▁ basis",
+      "oci ety",
+      "坚 持",
+      "lect ions",
+      "lection s",
+      "l ections",
+      "分 享",
+      "▁b log",
+      "▁bl og",
+      "▁blo g",
+      "▁ blog",
+      "▁b road",
+      "▁br oad",
+      "▁bro ad",
+      "▁ broad",
+      "id get",
+      "idge t",
+      "oc ks",
+      "ock s",
+      "o cks",
+      "下 来",
+      "适 合",
+      "▁t im",
+      "▁ti m",
+      "▁ tim",
+      "▁f avor",
+      "▁fa vor",
+      "▁fav or",
+      "oe s",
+      "o es",
+      "▁re ceive",
+      "▁rece ive",
+      "▁ receive",
+      "▁Dist rict",
+      "▁Di strict",
+      "el ta",
+      "elt a",
+      "▁com par",
+      "▁comp ar",
+      "▁ compar",
+      "▁ob tain",
+      "▁obt ain",
+      "针 对",
+      "哪 些",
+      "▁vari ety",
+      "el low",
+      "ell ow",
+      "ello w",
+      "▁Ind ia",
+      "▁ India",
+      "up date",
+      "▁w ide",
+      "▁wid e",
+      "▁wi de",
+      "▁ wide",
+      "▁f ire",
+      "▁fi re",
+      "▁fir e",
+      "▁ fire",
+      "serv ice",
+      "s ervice",
+      "社 区",
+      "▁s end",
+      "▁se nd",
+      "▁sen d",
+      "▁ send",
+      "A n",
+      "第 六",
+      "▁pres ented",
+      "▁present ed",
+      "se mb",
+      "sem b",
+      "s emb",
+      "} (",
+      "▁deg ree",
+      "▁ degree",
+      "▁trans fer",
+      "▁transf er",
+      "▁ transfer",
+      "▁ident ified",
+      "▁ identified",
+      "的情 况",
+      "的 情况",
+      "iforn ia",
+      "▁p arts",
+      "▁part s",
+      "▁par ts",
+      "▁ parts",
+      "wi se",
+      "w ise",
+      "S P",
+      "风 险",
+      "Man ager",
+      "Manag er",
+      "Manage r",
+      "M anager",
+      "ic ians",
+      "ici ans",
+      "ician s",
+      "icia ns",
+      "ce l",
+      "c el",
+      "许 多",
+      "▁p arty",
+      "▁part y",
+      "▁par ty",
+      "▁ party",
+      "▁ 王",
+      "hi ng",
+      "hin g",
+      "h ing",
+      "中 央",
+      "ead er",
+      "ea der",
+      "e ader",
+      "联 合",
+      "持 续",
+      "In put",
+      "▁re sid",
+      "▁res id",
+      "▁att ack",
+      "▁ attack",
+      "et ime",
+      "eti me",
+      "e time",
+      "▁s ort",
+      "▁so rt",
+      "▁sor t",
+      "▁ sort",
+      "▁u lt",
+      "▁ul t",
+      "▁ ult",
+      "严 重",
+      "资 金",
+      "▁comp uter",
+      "▁comput er",
+      "▁compute r",
+      "▁ computer",
+      "or ph",
+      "orp h",
+      "aut h",
+      "au th",
+      "a uth",
+      "交 易",
+      "▁great er",
+      "▁gre ater",
+      "▁ greater",
+      "▁g ives",
+      "▁giv es",
+      "▁give s",
+      "▁gi ves",
+      "\" ]",
+      "▁s tage",
+      "▁st age",
+      "▁ stage",
+      "加 入",
+      "Bas e",
+      "Ba se",
+      "B ase",
+      "▁n umer",
+      "▁num er",
+      "▁nu mer",
+      "k a",
+      "▁r ates",
+      "▁rate s",
+      "▁rat es",
+      "▁ra tes",
+      "▁ rates",
+      "丰 富",
+      "▁qu al",
+      "▁q ual",
+      "▁ qual",
+      "ie rs",
+      "ier s",
+      "i ers",
+      "▁F rom",
+      "▁Fr om",
+      "▁Fro m",
+      "▁ From",
+      "▁b orn",
+      "▁bo rn",
+      "▁bor n",
+      "▁ born",
+      "▁us ually",
+      "▁usual ly",
+      "▁ usually",
+      "▁phys ical",
+      "▁physic al",
+      "▁physi cal",
+      "▁ physical",
+      "深 圳",
+      "意 见",
+      "▁through out",
+      "it ect",
+      "ite ct",
+      "指 导",
+      "▁M us",
+      "▁Mu s",
+      "▁ Mus",
+      "C P",
+      "内容 简介",
+      "比 赛",
+      "▁a x",
+      "▁ ax",
+      "par se",
+      "pars e",
+      "p arse",
+      "▁S tr",
+      "▁St r",
+      "▁ Str",
+      "▁D is",
+      "▁Di s",
+      "▁ Dis",
+      "▁m oment",
+      "▁mom ent",
+      "▁mo ment",
+      "▁ moment",
+      "du c",
+      "d uc",
+      "咨 询",
+      "▁p oly",
+      "▁pol y",
+      "▁po ly",
+      "▁ poly",
+      "初 字第",
+      "sh ip",
+      "shi p",
+      "s hip",
+      "合 法",
+      "▁Fe bruary",
+      "▁ February",
+      "▁hy pot",
+      "▁hyp ot",
+      "ici ency",
+      "▁J ust",
+      "▁Ju st",
+      "▁ Just",
+      "全 球",
+      "▁t ools",
+      "▁to ols",
+      "▁too ls",
+      "▁tool s",
+      "▁ tools",
+      "▁sec ret",
+      "▁ secret",
+      "▁de liver",
+      "▁del iver",
+      "No ne",
+      "Non e",
+      "N one",
+      "▁stud ied",
+      "增 长",
+      "收 取",
+      "▁the ory",
+      "▁theor y",
+      "▁ theory",
+      "▁m ale",
+      "▁mal e",
+      "▁ma le",
+      "▁ male",
+      "▁long er",
+      "▁lon ger",
+      "o z",
+      "▁o il",
+      "▁ oil",
+      "▁w atch",
+      "▁wat ch",
+      "▁ watch",
+      "▁dist rict",
+      "▁di strict",
+      "▁distr ict",
+      "▁ district",
+      "▁m atter",
+      "▁mat ter",
+      "▁matt er",
+      "▁ matter",
+      "设 置",
+      "Tr ans",
+      "Tra ns",
+      "T rans",
+      "▁R E",
+      "▁ RE",
+      "▁L ondon",
+      "▁Lo ndon",
+      "▁Lon don",
+      "▁ London",
+      "▁ex isting",
+      "▁exist ing",
+      "▁ existing",
+      "不 得",
+      "▁w or",
+      "▁wo r",
+      "▁ wor",
+      "as ons",
+      "ason s",
+      "▁s kin",
+      "▁sk in",
+      "▁ski n",
+      "▁ skin",
+      "it ation",
+      "ita tion",
+      "itat ion",
+      "感 情",
+      "er ve",
+      "erv e",
+      "▁I nc",
+      "▁In c",
+      "▁ Inc",
+      "ja va",
+      "jav a",
+      "j ava",
+      "▁C ON",
+      "▁CO N",
+      "▁ CON",
+      "de n",
+      "d en",
+      "ti c",
+      "t ic",
+      "▁man ufact",
+      "▁ manufact",
+      "北 省",
+      "doc ument",
+      "d ocument",
+      "▁R ev",
+      "▁Re v",
+      "▁ Rev",
+      "ab s",
+      "a bs",
+      "Over ride",
+      "▁C har",
+      "▁Ch ar",
+      "▁Cha r",
+      "▁ Char",
+      "h y",
+      "meth od",
+      "m ethod",
+      "n w",
+      "▁con struct",
+      "▁co nstruct",
+      "▁const ruct",
+      "▁constr uct",
+      "▁ construct",
+      "tar get",
+      "t arget",
+      "Re p",
+      "R ep",
+      "▁re ading",
+      "▁read ing",
+      "▁ reading",
+      "ms g",
+      "m sg",
+      "nB ut",
+      "n But",
+      "an el",
+      "ane l",
+      "a nel",
+      "st ream",
+      "stre am",
+      "细 胞",
+      "▁re sources",
+      "▁res ources",
+      "▁resource s",
+      "▁ resources",
+      "▁K ing",
+      "▁Kin g",
+      "▁Ki ng",
+      "▁ King",
+      "▁an ti",
+      "▁ant i",
+      "▁ anti",
+      "▁st ates",
+      "▁state s",
+      "▁stat es",
+      "▁ states",
+      "▁S pec",
+      "▁Sp ec",
+      "▁Spe c",
+      "▁ Spec",
+      "D S",
+      "ri ver",
+      "riv er",
+      "rive r",
+      "r iver",
+      "▁up on",
+      "▁u pon",
+      "医学 百科",
+      "▁Y our",
+      "▁You r",
+      "▁Yo ur",
+      "▁ Your",
+      "▁w ords",
+      "▁word s",
+      "▁wor ds",
+      "▁ words",
+      "更 加",
+      "it c",
+      "i tc",
+      "▁r ights",
+      "▁right s",
+      "▁ rights",
+      "生 物",
+      "\\\\ _",
+      "▁A ir",
+      "▁ Air",
+      "正 在",
+      "ex cept",
+      "稳 定",
+      "检察 院",
+      "▁P M",
+      "▁ PM",
+      "T S",
+      "三 角",
+      "查 明",
+      "特 色",
+      "律师 事务所",
+      "ia s",
+      "i as",
+      "这样 的",
+      "这 样的",
+      "越 来",
+      "▁I nst",
+      "▁In st",
+      "▁Ins t",
+      "▁ Inst",
+      "▁n ucle",
+      "▁nu cle",
+      "▁ nucle",
+      "取 得",
+      "~~ ~\\",
+      "~~~ \\",
+      "~ ~~\\",
+      "西 省",
+      "有 期徒刑",
+      "ge ner",
+      "gen er",
+      "gene r",
+      "g ener",
+      "I f",
+      "oi ce",
+      "o ice",
+      "as ks",
+      "ask s",
+      "A p",
+      "▁pl ant",
+      "▁plan t",
+      "▁ plant",
+      "▁appro xim",
+      "▁approx im",
+      "pt ions",
+      "ption s",
+      "监 狱",
+      "▁$ {",
+      "▁ ${",
+      "▁techn iques",
+      "▁technique s",
+      "▁s ave",
+      "▁sa ve",
+      "▁sav e",
+      "▁ save",
+      "▁e qual",
+      "▁equ al",
+      "▁eq ual",
+      "▁ equal",
+      "▁n il",
+      "▁ni l",
+      "▁ nil",
+      "n h",
+      "▁: :",
+      "▁ ::",
+      "▁N ow",
+      "▁No w",
+      "▁ Now",
+      "保 障",
+      "Ge ner",
+      "Gen er",
+      "Gene r",
+      "G ener",
+      "▁A M",
+      "▁ AM",
+      "▁f orce",
+      "▁for ce",
+      "▁ force",
+      "▁ex cept",
+      "▁ except",
+      "p m",
+      "▁m aterials",
+      "▁material s",
+      "▁mater ials",
+      "越来 越",
+      "购 买",
+      "区 域",
+      "样 的",
+      "▁li near",
+      "▁line ar",
+      "▁lin ear",
+      "▁ linear",
+      "t d",
+      "目 录",
+      "▁ess ential",
+      "▁ essential",
+      "体 验",
+      "▁op tim",
+      "▁opt im",
+      "▁ optim",
+      "用 于",
+      "专 家",
+      "▁He alth",
+      "▁ Health",
+      "gi t",
+      "g it",
+      "ig hest",
+      "igh est",
+      "提 交",
+      "代理 审判员",
+      "▁out side",
+      "▁outs ide",
+      "点 击",
+      "运 行",
+      "▁ch alleng",
+      "▁chall eng",
+      "▁F irst",
+      "▁Fi rst",
+      "▁ First",
+      "▁Ch ina",
+      "▁Chi na",
+      "▁Chin a",
+      "▁ China",
+      "新 闻",
+      "阶 段",
+      "▁f all",
+      "▁fa ll",
+      "▁fal l",
+      "▁ fall",
+      "▁al gorithm",
+      "▁ algorithm",
+      "AT E",
+      "A TE",
+      "▁ex tra",
+      "▁ext ra",
+      "▁extr a",
+      "▁ extra",
+      "因 素",
+      "那 些",
+      "▁A cc",
+      "▁Ac c",
+      "▁ Acc",
+      "求 出",
+      "▁w rong",
+      "▁wr ong",
+      "▁ wrong",
+      "后 的",
+      "为 主",
+      "▁P re",
+      "▁Pr e",
+      "▁ Pre",
+      "输 出",
+      "▁D NA",
+      "▁DN A",
+      "▁ DNA",
+      "广 告",
+      "▁B ritish",
+      "▁Brit ish",
+      "▁ British",
+      "▁p retty",
+      "▁pre tty",
+      "▁pret ty",
+      "▁ pretty",
+      "cd ot",
+      "c dot",
+      "el ves",
+      "elve s",
+      "改 造",
+      "▁ph one",
+      "▁phon e",
+      "▁ phone",
+      "▁Th anks",
+      "▁Thank s",
+      "▁Than ks",
+      "▁ Thanks",
+      "▁t esting",
+      "▁test ing",
+      "▁ testing",
+      "▁\\\\ \\",
+      "▁\\ \\\\",
+      "▁ \\\\\\",
+      "ie ce",
+      "iec e",
+      "▁ 李",
+      "al le",
+      "all e",
+      "▁mechan isms",
+      "▁mechanism s",
+      "合 理",
+      "第 七",
+      "▁re main",
+      "▁rem ain",
+      "▁b asic",
+      "▁bas ic",
+      "▁ basic",
+      "▁H ar",
+      "▁Ha r",
+      "▁ Har",
+      "▁O ut",
+      "▁ Out",
+      "促 进",
+      "▁s erved",
+      "▁serv ed",
+      "▁ser ved",
+      "▁serve d",
+      "出 了",
+      "C A",
+      "▁d istance",
+      "▁dist ance",
+      "▁di stance",
+      "▁ distance",
+      "▁t ask",
+      "▁ta sk",
+      "▁ task",
+      "▁s elected",
+      "▁se lected",
+      "▁select ed",
+      "▁sel ected",
+      "▁ selected",
+      "ai gn",
+      "a ign",
+      "er ved",
+      "erv ed",
+      "erve d",
+      "pl ications",
+      "plic ations",
+      "plication s",
+      "第 五",
+      "人 才",
+      "R A",
+      "▁Americ a",
+      "▁Amer ica",
+      "▁ America",
+      "▁S ome",
+      "▁So me",
+      "▁Som e",
+      "▁ Some",
+      "▁b irth",
+      "▁bir th",
+      "▁ birth",
+      "▁pro mot",
+      "▁prom ot",
+      "▁promo t",
+      "▁re ach",
+      "▁r each",
+      "▁ reach",
+      "食 品",
+      "受 到",
+      "uc cess",
+      "ucc ess",
+      "u ccess",
+      "▁pro grams",
+      "▁pr ograms",
+      "▁program s",
+      "▁ programs",
+      "p g",
+      "▁s n",
+      "▁ sn",
+      "▁G od",
+      "▁Go d",
+      "▁ God",
+      "ra n",
+      "r an",
+      "▁ap ply",
+      "▁app ly",
+      "▁appl y",
+      "▁ apply",
+      "改 变",
+      "▁u rl",
+      "▁ur l",
+      "▁ url",
+      "▁D ef",
+      "▁De f",
+      "▁ Def",
+      "▁over all",
+      "▁$ (",
+      "▁ $(",
+      "vi der",
+      "vid er",
+      "v ider",
+      "▁b al",
+      "▁ba l",
+      "▁ bal",
+      "有 人",
+      "▁c entral",
+      "▁cent ral",
+      "▁centr al",
+      "▁ central",
+      "mer ican",
+      "meric an",
+      "]) ;\\",
+      "]); \\",
+      "] );\\",
+      "▁it self",
+      "▁i tself",
+      "▁its elf",
+      "测 试",
+      "▁prov iding",
+      "▁provid ing",
+      "不同 的",
+      "stit ute",
+      "stitu te",
+      "▁Cal ifornia",
+      "▁ California",
+      "opt ions",
+      "option s",
+      "o ptions",
+      "▁M ich",
+      "▁Mi ch",
+      "▁Mic h",
+      "▁c ritical",
+      "▁crit ical",
+      "▁critic al",
+      "▁ critical",
+      "▁h it",
+      "▁hi t",
+      "▁ hit",
+      "员 工",
+      "▁f ib",
+      "▁fi b",
+      "▁ fib",
+      "T O",
+      "cent ury",
+      "G B",
+      "▁ph ase",
+      "▁ phase",
+      "K E",
+      "▁p olicy",
+      "▁pol icy",
+      "▁polic y",
+      "▁ policy",
+      "▁trans form",
+      "▁transf orm",
+      "▁ transform",
+      "解 答",
+      "都 会",
+      "oot h",
+      "oo th",
+      "o oth",
+      "lev ant",
+      "▁dec ision",
+      "▁S an",
+      "▁Sa n",
+      "▁ San",
+      "宣 传",
+      "In st",
+      "Ins t",
+      "I nst",
+      "▁d oc",
+      "▁do c",
+      "▁ doc",
+      "犯 罪",
+      "yl es",
+      "yle s",
+      "y les",
+      "▁neg ative",
+      "▁ negative",
+      "统 计",
+      "以 来",
+      "设 施",
+      "▁e ntry",
+      "▁en try",
+      "▁ent ry",
+      "▁entr y",
+      "▁ entry",
+      "▁C lass",
+      "▁Cl ass",
+      "▁ Class",
+      "pon ents",
+      "ponent s",
+      "▁pres ence",
+      "▁ presence",
+      "我们 的",
+      "我 们的",
+      "ep s",
+      "e ps",
+      "▁com ponent",
+      "▁compon ent",
+      "▁ component",
+      "wi n",
+      "w in",
+      "量 的",
+      "An d",
+      "A nd",
+      "▁individual s",
+      "ad min",
+      "▁m er",
+      "▁me r",
+      "▁ mer",
+      "merc ial",
+      "▁h eight",
+      "▁he ight",
+      "▁ height",
+      "优 势",
+      "ug e",
+      "u ge",
+      "() )",
+      "( ))",
+      "▁ 张",
+      "民 间",
+      "pl ies",
+      "p lies",
+      "c a",
+      "▁ch oice",
+      "▁cho ice",
+      "▁ choice",
+      "人民 检察院",
+      "▁v it",
+      "▁vi t",
+      "▁ vit",
+      "的发 展",
+      "的 发展",
+      "▁vari ables",
+      "▁variable s",
+      "▁ variables",
+      "bl ock",
+      "blo ck",
+      "b lock",
+      "直 线",
+      "▁c al",
+      "▁ca l",
+      "▁ cal",
+      "▁a void",
+      "▁av oid",
+      "▁ avoid",
+      "ut ils",
+      "util s",
+      "uti ls",
+      "▁Go ogle",
+      "▁ Google",
+      "快 速",
+      "恢 复",
+      "se e",
+      "s ee",
+      "le vision",
+      "lev ision",
+      "▁ap plied",
+      "▁appl ied",
+      "et urn",
+      "etur n",
+      "e turn",
+      "▁c andid",
+      "▁can did",
+      "▁cand id",
+      "身 份",
+      "▁F l",
+      "▁ Fl",
+      "Si ze",
+      "S ize",
+      "性 质",
+      "or ds",
+      "ord s",
+      "是一 种",
+      "是 一种",
+      "记 者",
+      "▁every one",
+      "ul ations",
+      "ulation s",
+      "im ages",
+      "image s",
+      "ima ges",
+      "imag es",
+      "▁s al",
+      "▁sa l",
+      "▁ sal",
+      "num ber",
+      "nu mber",
+      "n umber",
+      "▁s rc",
+      "▁sr c",
+      "▁ src",
+      "nC om",
+      "nCo m",
+      "n Com",
+      "▁s atisf",
+      "▁sat isf",
+      "▁eas ily",
+      "▁s ynt",
+      "▁sy nt",
+      "▁syn t",
+      "▁ synt",
+      "he nt",
+      "hen t",
+      "h ent",
+      "▁s ound",
+      "▁so und",
+      "▁sou nd",
+      "▁ sound",
+      "ounc ed",
+      "ounce d",
+      "oun ced",
+      "成 绩",
+      "土 地",
+      "▁B ro",
+      "▁Br o",
+      "▁ Bro",
+      "▁F ound",
+      "▁Fo und",
+      "▁ Found",
+      "协 会",
+      "体 育",
+      "VE R",
+      "V ER",
+      "▁l imited",
+      "▁lim ited",
+      "▁limit ed",
+      "▁ limited",
+      "en ge",
+      "eng e",
+      "e nge",
+      "▁redu ce",
+      "▁red uce",
+      "▁ reduce",
+      "▁s tim",
+      "▁st im",
+      "▁ stim",
+      "免 费",
+      "▁W ork",
+      "▁Wor k",
+      "▁Wo rk",
+      "▁ Work",
+      "▁pro gress",
+      "▁ progress",
+      "▁d eep",
+      "▁de ep",
+      "▁ deep",
+      "na nd",
+      "nan d",
+      "n and",
+      "sh ow",
+      "s how",
+      "yt es",
+      "yte s",
+      "y tes",
+      "▁pl aces",
+      "▁place s",
+      "▁plac es",
+      "▁ places",
+      "办 法",
+      "的 高",
+      "申请 执行",
+      "▁O pen",
+      "▁Op en",
+      "▁ Open",
+      "▁D et",
+      "▁De t",
+      "▁ Det",
+      "告 诉",
+      "▁trans port",
+      "▁tran sport",
+      "▁ transport",
+      "别 人",
+      "▁fin ancial",
+      "▁financ ial",
+      "▁s on",
+      "▁so n",
+      "▁ son",
+      "he ight",
+      "h eight",
+      "▁s hop",
+      "▁sh op",
+      "▁ shop",
+      "▁c am",
+      "▁ca m",
+      "▁ cam",
+      "再 次",
+      "▁d ivid",
+      "▁di vid",
+      "▁div id",
+      "\"> </",
+      "\">< /",
+      "\" ></",
+      "ut ive",
+      "uti ve",
+      "py thon",
+      "p ython",
+      "服 刑",
+      "ar es",
+      "are s",
+      "a res",
+      "▁c oming",
+      "▁com ing",
+      "▁co ming",
+      "▁ coming",
+      "▁h ighest",
+      "▁high est",
+      "▁s ense",
+      "▁se nse",
+      "▁sens e",
+      "▁sen se",
+      "▁M ath",
+      "▁Mat h",
+      "▁Ma th",
+      "▁ Math",
+      "实 践",
+      "vent ion",
+      "ven tion",
+      "v ention",
+      "As s",
+      "A ss",
+      "el t",
+      "e lt",
+      "▁vol ume",
+      "▁ volume",
+      "▁mod ern",
+      "▁mode rn",
+      "▁moder n",
+      "▁ modern",
+      "▁vari able",
+      "▁var iable",
+      "▁ variable",
+      "th eta",
+      "the ta",
+      "we et",
+      "找 到",
+      "O ut",
+      "▁re ady",
+      "▁read y",
+      "▁ ready",
+      "/ (",
+      "▁Ind ian",
+      "▁India n",
+      "nA n",
+      "n An",
+      "▁in dependent",
+      "▁ind ependent",
+      "▁independ ent",
+      "▁ independent",
+      "▁U K",
+      "▁ UK",
+      "▁j oin",
+      "▁jo in",
+      "▁ join",
+      "ic ket",
+      "ick et",
+      "y r",
+      "经 常",
+      "▁c ook",
+      "▁co ok",
+      "▁ cook",
+      "医 疗",
+      "▁com ponents",
+      "▁compon ents",
+      "▁component s",
+      "▁ components",
+      "用 的",
+      "ess ages",
+      "essage s",
+      "essa ges",
+      "工 具",
+      "意 思",
+      "▁l ed",
+      "▁le d",
+      "▁ led",
+      "D F",
+      "or ge",
+      "org e",
+      "地 产",
+      "br a",
+      "b ra",
+      "查 封",
+      "**** ************",
+      "******** ********",
+      "************ ****",
+      "N o",
+      "▁any one",
+      "}$ $",
+      "} $$",
+      "Q L",
+      "X X",
+      "▁cor respon",
+      "ae l",
+      "a el",
+      "np ackage",
+      "n package",
+      "▁n amespace",
+      "▁name space",
+      "▁names pace",
+      "▁ namespace",
+      "// \\",
+      "/ /\\",
+      "▁s afety",
+      "▁saf ety",
+      "▁safe ty",
+      "联 系",
+      "Su b",
+      "S ub",
+      "二〇一 七年",
+      "▁m ode",
+      "▁mod e",
+      "▁mo de",
+      "▁ mode",
+      "▁ach ieve",
+      "创 作",
+      "< <",
+      "▁m olecular",
+      "▁mole cular",
+      "▁molec ular",
+      "▁ molecular",
+      "是 什么",
+      "施 工",
+      "ty pes",
+      "type s",
+      "typ es",
+      "t ypes",
+      "Im age",
+      "▁S k",
+      "▁ Sk",
+      "强 制",
+      "▁N or",
+      "▁No r",
+      "▁ Nor",
+      "错 误",
+      "am in",
+      "ami n",
+      "a min",
+      "▁m issing",
+      "▁miss ing",
+      "▁mis sing",
+      "▁ missing",
+      "▁v ector",
+      "▁ve ctor",
+      "▁vec tor",
+      "▁ vector",
+      "互联 网",
+      "互 联网",
+      "▁be hind",
+      "▁beh ind",
+      "b a",
+      "oc ation",
+      "oca tion",
+      "o cation",
+      "sp lit",
+      "spl it",
+      "s plit",
+      "et ch",
+      "etc h",
+      "D A",
+      "十 八",
+      "真 正",
+      "am eter",
+      "ame ter",
+      "amet er",
+      "a meter",
+      "▁f unctional",
+      "▁function al",
+      "▁funct ional",
+      "▁ functional",
+      "▁count ries",
+      "▁ countries",
+      "到 庭",
+      "▁E m",
+      "▁ Em",
+      "▁I tal",
+      "▁It al",
+      "▁h on",
+      "▁ho n",
+      "▁ hon",
+      "▁as sum",
+      "▁ass um",
+      "ed ia",
+      "edi a",
+      "ous ly",
+      "N U",
+      "({ \\",
+      "( {\\",
+      "al a",
+      "a la",
+      "▁A nt",
+      "▁An t",
+      "▁ Ant",
+      "fi rst",
+      "f irst",
+      "ic ens",
+      "ice ns",
+      "▁\\\\ )",
+      "▁ \\\\)",
+      "▁h or",
+      "▁ho r",
+      "▁ hor",
+      "机 会",
+      "sp ace",
+      "s pace",
+      "ip ment",
+      "▁} }",
+      "▁ }}",
+      "Ind ex",
+      "▁r ap",
+      "▁ra p",
+      "▁ rap",
+      "▁d riv",
+      "▁dr iv",
+      "▁dri v",
+      "▁ driv",
+      "ir ing",
+      "iri ng",
+      "i ring",
+      "▁A PI",
+      "▁AP I",
+      "▁ API",
+      "贷 款",
+      "SO N",
+      "S ON",
+      "st e",
+      "s te",
+      "生 态",
+      "▁ 类型",
+      "▁c onc",
+      "▁con c",
+      "▁co nc",
+      "▁ conc",
+      "▁e nter",
+      "▁en ter",
+      "▁ent er",
+      "▁ enter",
+      "us ter",
+      "ust er",
+      "u ster",
+      "▁re v",
+      "▁r ev",
+      "▁ rev",
+      "都 不",
+      "os ure",
+      "▁C ons",
+      "▁Con s",
+      "▁Co ns",
+      "▁ Cons",
+      "审理 终结",
+      "F alse",
+      "▁sk ills",
+      "▁skill s",
+      "▁ skills",
+      "S A",
+      "▁str eng",
+      "▁stre ng",
+      "整 个",
+      "▁stud ent",
+      "▁ student",
+      "is tr",
+      "ist r",
+      "i str",
+      "or ial",
+      "ori al",
+      "oria l",
+      "o rial",
+      "开 庭",
+      "▁prote ins",
+      "▁protein s",
+      "认 识",
+      "▁p ull",
+      "▁pul l",
+      "▁pu ll",
+      "▁ pull",
+      "▁P ort",
+      "▁Po rt",
+      "▁Por t",
+      "▁ Port",
+      "从 而",
+      "▁s cript",
+      "▁sc ript",
+      "▁scr ipt",
+      "▁ script",
+      "▁ide as",
+      "▁id eas",
+      "▁idea s",
+      "▁H er",
+      "▁He r",
+      "▁ Her",
+      "▁pr inc",
+      "▁pri nc",
+      "▁ princ",
+      "pro duct",
+      "produ ct",
+      "prod uct",
+      "▁sc ient",
+      "▁sci ent",
+      "▁ scient",
+      "▁ar gs",
+      "▁arg s",
+      "▁ args",
+      "减 少",
+      "▁Dep artment",
+      "▁ Department",
+      "年 来",
+      "te s",
+      "t es",
+      "con sole",
+      "cons ole",
+      "m t",
+      "▁im med",
+      "▁imm ed",
+      "模 型",
+      "确 认",
+      "小 学",
+      "Co unt",
+      "Cou nt",
+      "C ount",
+      "▁appro pri",
+      "▁D em",
+      "▁De m",
+      "▁ Dem",
+      "福 建",
+      "nP ro",
+      "nPr o",
+      "n Pro",
+      "▁ask ed",
+      "ar ily",
+      "ari ly",
+      "Y ou",
+      "Res ult",
+      "生 命",
+      "经 历",
+      "临 床",
+      "Pa ge",
+      "Pag e",
+      "P age",
+      "▁com pleted",
+      "▁comp leted",
+      "▁complet ed",
+      "▁complete d",
+      "▁compl eted",
+      "▁ completed",
+      "▁co ord",
+      "▁ coord",
+      "律师事务所 律师",
+      "wit ter",
+      "w itter",
+      "pi te",
+      "pit e",
+      "p ite",
+      "le q",
+      "l eq",
+      "核 心",
+      "B N",
+      "id a",
+      "i da",
+      "ec ause",
+      "eca use",
+      "e cause",
+      "wo od",
+      "w ood",
+      "▁is ol",
+      "▁i sol",
+      "▁iso l",
+      "▁ isol",
+      "▁d og",
+      "▁do g",
+      "▁ dog",
+      "se lect",
+      "sel ect",
+      "s elect",
+      "▁Pres ident",
+      "▁ President",
+      "de t",
+      "d et",
+      "ect ions",
+      "ection s",
+      "▁opportun ity",
+      "zy m",
+      "z ym",
+      "如 不服",
+      "▁follow ed",
+      "▁reg ister",
+      "▁ register",
+      "velop ment",
+      "iv ery",
+      "ive ry",
+      "iver y",
+      "i very",
+      "Act ion",
+      "Ac tion",
+      "A ction",
+      "人 口",
+      "▁process es",
+      "▁proc esses",
+      "▁d igital",
+      "▁dig ital",
+      "▁digit al",
+      "▁ digital",
+      "最 终",
+      "en ing",
+      "eni ng",
+      "enin g",
+      "e ning",
+      "▁recomm end",
+      "▁s olve",
+      "▁sol ve",
+      "▁solv e",
+      "▁ solve",
+      "▁redu ced",
+      "▁red uced",
+      "▁reduce d",
+      "▁re ference",
+      "▁r eference",
+      "▁refer ence",
+      "▁refere nce",
+      "▁ reference",
+      "he re",
+      "her e",
+      "h ere",
+      "阳 市",
+      "A t",
+      "▁V al",
+      "▁Va l",
+      "▁ Val",
+      "由 原告",
+      "移 动",
+      "pos ed",
+      "pose d",
+      "po sed",
+      "p osed",
+      "调 整",
+      "▁vis ual",
+      "▁ visual",
+      "Re g",
+      "R eg",
+      "mit ted",
+      "mitt ed",
+      "m itted",
+      "降 低",
+      "记 录",
+      "s v",
+      "▁cont ribut",
+      "com mon",
+      "comm on",
+      "▁h ydro",
+      "▁hy dro",
+      "▁hyd ro",
+      "▁ hydro",
+      "tem plate",
+      "temp late",
+      "t emplate",
+      "olog ies",
+      "ologie s",
+      "想 要",
+      "▁A NY",
+      "▁AN Y",
+      "▁ ANY",
+      "山东 省",
+      "at tle",
+      "att le",
+      "a ttle",
+      "▁O f",
+      "▁ Of",
+      "ri ter",
+      "rit er",
+      "rite r",
+      "r iter",
+      "ION S",
+      "IO NS",
+      "I ONS",
+      "▁cl ients",
+      "▁client s",
+      "▁cli ents",
+      "▁ clients",
+      "体 的",
+      "ul ate",
+      "ula te",
+      "u late",
+      "▁custom er",
+      "▁cust omer",
+      "▁ customer",
+      "▁b rand",
+      "▁br and",
+      "▁bra nd",
+      "▁ brand",
+      "▁ext rem",
+      "▁extr em",
+      "yc l",
+      "y cl",
+      "▁ant ib",
+      "▁anti b",
+      "▁ antib",
+      "▁b egan",
+      "▁be gan",
+      "▁beg an",
+      "land s",
+      "lan ds",
+      "l ands",
+      "ru pt",
+      "rup t",
+      "r upt",
+      "本院 依法",
+      "I B",
+      "dis t",
+      "di st",
+      "d ist",
+      "现已 审理终结",
+      "及 其",
+      "pass word",
+      "Hand ler",
+      "Handle r",
+      "▁f ields",
+      "▁field s",
+      "▁ fields",
+      "▁L og",
+      "▁Lo g",
+      "▁ Log",
+      "alc ulate",
+      "alcul ate",
+      "▁w anted",
+      "▁want ed",
+      "▁t ried",
+      "▁tr ied",
+      "▁tri ed",
+      "▁D ay",
+      "▁Da y",
+      "▁ Day",
+      "版 本",
+      "▁max imum",
+      "▁maxim um",
+      "▁ maximum",
+      "L e",
+      "车 辆",
+      "ba nd",
+      "ban d",
+      "b and",
+      "事 件",
+      "ag er",
+      "age r",
+      "a ger",
+      "▁att ention",
+      "▁atten tion",
+      "▁attent ion",
+      "▁ac qu",
+      "▁ acqu",
+      "里 的",
+      "及 时",
+      "压 力",
+      "▁ass ist",
+      "▁s amples",
+      "▁sample s",
+      "▁sam ples",
+      "▁ samples",
+      "▁un its",
+      "▁u nits",
+      "▁unit s",
+      "▁uni ts",
+      "▁ units",
+      "▁me eting",
+      "▁meet ing",
+      "▁ meeting",
+      "▁ex hib",
+      "▁exh ib",
+      "to ken",
+      "tok en",
+      "t oken",
+      "ri s",
+      "r is",
+      "认 真",
+      "就可 以",
+      "就 可以",
+      "▁h istor",
+      "▁his tor",
+      "▁hist or",
+      "▁hi stor",
+      "运 用",
+      "▁A ustralia",
+      "▁Austral ia",
+      "▁ Australia",
+      "▁con struction",
+      "▁construct ion",
+      "▁constr uction",
+      "am ily",
+      "amil y",
+      "ami ly",
+      "par ent",
+      "pa rent",
+      "pare nt",
+      "paren t",
+      "p arent",
+      "▁t umor",
+      "▁tum or",
+      "▁tu mor",
+      "▁ tumor",
+      "文 书",
+      "\" /",
+      "▁pl aying",
+      "▁play ing",
+      "▁ playing",
+      "Re c",
+      "R ec",
+      "Col or",
+      "Co lor",
+      "C olor",
+      "▁l ost",
+      "▁lo st",
+      "▁los t",
+      "▁ lost",
+      "即 发生法律效力",
+      "▁dis cover",
+      "▁disc over",
+      "▁t issue",
+      "至 今",
+      "il ling",
+      "ill ing",
+      "illi ng",
+      "illin g",
+      "▁r ow",
+      "▁ro w",
+      "▁ row",
+      "math rm",
+      "出 的",
+      "另 外",
+      "不 好",
+      "▁T V",
+      "▁ TV",
+      "▁ 第",
+      "▁s low",
+      "▁sl ow",
+      "▁ slow",
+      "om b",
+      "o mb",
+      "nf unc",
+      "nfun c",
+      "n func",
+      "ar io",
+      "ari o",
+      "a rio",
+      "▁F ree",
+      "▁Fr ee",
+      "▁Fre e",
+      "▁ Free",
+      "g u",
+      "erv ices",
+      "ervice s",
+      "▁U ser",
+      "▁Use r",
+      "▁Us er",
+      "▁ User",
+      "方 便",
+      "清 楚",
+      "▁s quare",
+      "▁squ are",
+      "▁ square",
+      "天 津",
+      "▁d omain",
+      "▁do main",
+      "▁dom ain",
+      "▁ domain",
+      "▁A re",
+      "▁Ar e",
+      "▁ Are",
+      "▁F in",
+      "▁Fi n",
+      "▁ Fin",
+      "▁f orms",
+      "▁for ms",
+      "▁form s",
+      "▁ forms",
+      "二〇一 六年",
+      "bo t",
+      "b ot",
+      "电 脑",
+      "▁M od",
+      "▁Mo d",
+      "▁ Mod",
+      "bre ak",
+      "b reak",
+      "定 义",
+      "被 申请人",
+      "约 定",
+      "t f",
+      "工 艺",
+      "▁Wh ile",
+      "▁ While",
+      "▁vill age",
+      "▁villa ge",
+      "z a",
+      "▁eff icient",
+      "▁effic ient",
+      "▁ efficient",
+      "▁in stit",
+      "▁inst it",
+      "▁ins tit",
+      "St atus",
+      "Stat us",
+      "y a",
+      "as cular",
+      "asc ular",
+      "ascul ar",
+      "做 好",
+      "el come",
+      "ot ype",
+      "otyp e",
+      "o type",
+      "nL iving",
+      "ft y",
+      "f ty",
+      "三 个",
+      "pl ot",
+      "p lot",
+      "li er",
+      "lie r",
+      "l ier",
+      "T e",
+      "应 的",
+      "债 务",
+      "疾 病",
+      "nS olution",
+      "n Solution",
+      "可 在",
+      "▁p urpose",
+      "▁pur pose",
+      "▁ purpose",
+      "ma ster",
+      "mas ter",
+      "m aster",
+      "om as",
+      "oma s",
+      "o mas",
+      "▁start ing",
+      "▁star ting",
+      "▁ starting",
+      "▁determ ined",
+      "▁determine d",
+      "▁determin ed",
+      "▁ determined",
+      "▁the ore",
+      "▁theor e",
+      "ic ial",
+      "ici al",
+      "icia l",
+      "▁Re search",
+      "▁Res earch",
+      "▁ Research",
+      "监 督",
+      "▁m emory",
+      "▁mem ory",
+      "▁memo ry",
+      "▁memor y",
+      "▁ memory",
+      "▁f requency",
+      "▁frequ ency",
+      "▁ frequency",
+      "竞 争",
+      "加 工",
+      "演 员",
+      "长 期",
+      "back ground",
+      "G F",
+      "ta g",
+      "t ag",
+      "▁b ooks",
+      "▁book s",
+      "▁bo oks",
+      "▁ books",
+      "▁A ut",
+      "▁Au t",
+      "▁ Aut",
+      "C S",
+      "▁Bl ack",
+      "▁ Black",
+      "表 达",
+      "注 册",
+      "n l",
+      "▁D irect",
+      "▁Di rect",
+      "▁Dir ect",
+      "▁ Direct",
+      "ac ing",
+      "acin g",
+      "aci ng",
+      "a cing",
+      "▁c op",
+      "▁co p",
+      "▁ cop",
+      "fil ter",
+      "▁su bst",
+      "▁sub st",
+      "▁subs t",
+      "结 论",
+      "七 条",
+      "▁n ice",
+      "▁nic e",
+      "▁ni ce",
+      "▁ nice",
+      "日 生",
+      "▁w all",
+      "▁wal l",
+      "▁wa ll",
+      "▁ wall",
+      "▁Te chn",
+      "▁Tech n",
+      "▁ Techn",
+      "▁W AR",
+      "▁WA R",
+      "▁ WAR",
+      "pu sh",
+      "pus h",
+      "p ush",
+      "▁d ifferences",
+      "▁dif ferences",
+      "▁differ ences",
+      "▁difference s",
+      "▁s etting",
+      "▁set ting",
+      "▁sett ing",
+      "▁ setting",
+      "▁d aily",
+      "▁da ily",
+      "▁ daily",
+      "▁cl osed",
+      "▁clos ed",
+      "▁close d",
+      "▁ closed",
+      "ig ation",
+      "iga tion",
+      "▁le ave",
+      "▁ leave",
+      "生 的",
+      "▁m achine",
+      "▁mach ine",
+      "▁ machine",
+      "ele br",
+      "下 面",
+      "▁pre viously",
+      "▁previous ly",
+      "▁prev iously",
+      "配 置",
+      "▁n othing",
+      "▁not hing",
+      "▁no thing",
+      "▁ nothing",
+      "as c",
+      "a sc",
+      "om in",
+      "omi n",
+      "o min",
+      "▁ch arg",
+      "▁char g",
+      "▁cha rg",
+      "▁ charg",
+      "▁< /",
+      "▁ </",
+      "过 去",
+      "ch ild",
+      "chi ld",
+      "第二百 五十",
+      "▁f ully",
+      "▁full y",
+      "▁ful ly",
+      "▁ fully",
+      "▁a ward",
+      "▁aw ard",
+      "▁awa rd",
+      "更 新",
+      "▁state ment",
+      "▁stat ement",
+      "▁ statement",
+      "▁E uropean",
+      "▁Europe an",
+      "▁ European",
+      "▁ 外",
+      "Cont ent",
+      "▁p o",
+      "▁ po",
+      "ot o",
+      "o to",
+      "厘 米",
+      "营 销",
+      "▁p refer",
+      "▁pre fer",
+      "▁pr efer",
+      "▁ prefer",
+      "查 询",
+      "row ser",
+      "rows er",
+      "rowse r",
+      "ie ved",
+      "ieve d",
+      "iev ed",
+      "副 本",
+      "▁[ ]\\",
+      "▁[] \\",
+      "▁ []\\",
+      "基 金",
+      "▁c ulture",
+      "▁cult ure",
+      "▁cul ture",
+      "▁ culture",
+      "思 路",
+      "cl usion",
+      "clus ion",
+      "H ttp",
+      "事 业",
+      "银行 股份有限公司",
+      "群 众",
+      "▁Com mun",
+      "▁Comm un",
+      "▁ Commun",
+      "▁m ice",
+      "▁mic e",
+      "▁mi ce",
+      "真 实",
+      "▁ch anged",
+      "▁chang ed",
+      "▁change d",
+      "▁chan ged",
+      "▁ changed",
+      "▁t hus",
+      "▁th us",
+      "▁ thus",
+      "na m",
+      "n am",
+      "▁r atio",
+      "▁rat io",
+      "▁ ratio",
+      "提 起",
+      "D B",
+      "它 的",
+      "▁A nswer",
+      "▁Ans wer",
+      "▁ Answer",
+      "▁S en",
+      "▁Se n",
+      "▁ Sen",
+      "▁re move",
+      "▁rem ove",
+      "▁remov e",
+      "▁ remove",
+      "▁Ch ampionship",
+      "▁Champions hip",
+      "▁Champion ship",
+      "un it",
+      "uni t",
+      "u nit",
+      "▁M r",
+      "▁ Mr",
+      "正 式",
+      "▁C our",
+      "▁Co ur",
+      "▁Cou r",
+      "▁p sych",
+      "▁ps ych",
+      "▁psy ch",
+      "▁ psych",
+      "看 看",
+      "▁p hen",
+      "▁ph en",
+      "▁ phen",
+      "M B",
+      "▁m obile",
+      "▁mob ile",
+      "▁mobil e",
+      "▁ mobile",
+      "▁d atabase",
+      "▁data base",
+      "▁dat abase",
+      "▁ database",
+      "ia ble",
+      "iab le",
+      "i able",
+      "re p",
+      "r ep",
+      "▁c ool",
+      "▁co ol",
+      "▁ cool",
+      "rol led",
+      "roll ed",
+      "作者 是",
+      "撤诉 申请",
+      "mission s",
+      "miss ions",
+      "m issions",
+      "▁H ol",
+      "▁Ho l",
+      "▁ Hol",
+      "十 四",
+      "也可 以",
+      "也 可以",
+      "os ing",
+      "osin g",
+      "osi ng",
+      "o sing",
+      "也 会",
+      "▁co uple",
+      "▁cou ple",
+      "▁coup le",
+      "植 物",
+      "▁back ground",
+      "▁ background",
+      "房 屋",
+      "▁mechan ism",
+      "化 的",
+      "▁A ND",
+      "▁AN D",
+      "▁ AND",
+      "意 识",
+      "民初 字第",
+      "民 初字第",
+      "团 队",
+      "▁l ack",
+      "▁la ck",
+      "▁lac k",
+      "▁le aders",
+      "▁lead ers",
+      "▁leader s",
+      "▁st orage",
+      "▁sto rage",
+      "▁ storage",
+      "后 来",
+      "f s",
+      "an ger",
+      "ang er",
+      "ange r",
+      "▁ch arge",
+      "▁char ge",
+      "▁charg e",
+      "▁ charge",
+      "ne d",
+      "n ed",
+      "▁s afe",
+      "▁sa fe",
+      "▁saf e",
+      "▁ safe",
+      "iz ations",
+      "ization s",
+      "▁Pro ject",
+      "▁ Project",
+      "明 确",
+      "ot a",
+      "o ta",
+      "来 自",
+      "at in",
+      "ati n",
+      "a tin",
+      "▁M ac",
+      "▁Ma c",
+      "▁ Mac",
+      "▁g enerated",
+      "▁gener ated",
+      "▁gen erated",
+      "▁gene rated",
+      "▁generate d",
+      "▁genera ted",
+      "▁ generated",
+      "▁your self",
+      "▁yours elf",
+      "▁O lymp",
+      "▁m odule",
+      "▁mod ule",
+      "▁mo dule",
+      "▁ module",
+      "▁S ervice",
+      "▁Serv ice",
+      "▁ Service",
+      "考 查",
+      "▁F alse",
+      "▁Fal se",
+      "▁ False",
+      "az ing",
+      "azi ng",
+      "a zing",
+      "▁n ote",
+      "▁not e",
+      "▁no te",
+      "▁ note",
+      "▁o nes",
+      "▁on es",
+      "▁one s",
+      "▁ ones",
+      "▁ex pert",
+      "▁exp ert",
+      "▁exper t",
+      "存 款",
+      "▁) \\",
+      "▁ )\\",
+      "▁t en",
+      "▁te n",
+      "▁ ten",
+      "h i",
+      "给 付",
+      "oin ed",
+      "oi ned",
+      "o ined",
+      "先 进",
+      "word s",
+      "wor ds",
+      "w ords",
+      ") ^",
+      "ze r",
+      "z er",
+      "ov ery",
+      "ove ry",
+      "over y",
+      "o very",
+      "原 审",
+      "▁ex cell",
+      "▁exc ell",
+      "▁excel l",
+      "▁find ings",
+      "▁fin dings",
+      "▁finding s",
+      "ma r",
+      "m ar",
+      "g a",
+      "Ac t",
+      "A ct",
+      "借 贷",
+      "▁v s",
+      "▁ vs",
+      "▁me tab",
+      "▁met ab",
+      "▁meta b",
+      "▁fil ter",
+      "▁ filter",
+      "ob j",
+      "o bj",
+      "▁f ixed",
+      "▁fix ed",
+      "▁ fixed",
+      "▁f ine",
+      "▁fin e",
+      "▁fi ne",
+      "▁ fine",
+      "多 的",
+      "▁C up",
+      "▁Cu p",
+      "▁m ention",
+      "▁men tion",
+      "▁ment ion",
+      "浙江 省",
+      "▁l ibrary",
+      "▁libr ary",
+      "▁ library",
+      "D O",
+      "tx t",
+      "t xt",
+      "▁trad itional",
+      "▁tradition al",
+      "▁ traditional",
+      "▁s uit",
+      "▁su it",
+      "▁n ature",
+      "▁na ture",
+      "▁nat ure",
+      "on ly",
+      "in fty",
+      "inf ty",
+      "▁E ast",
+      "▁Eas t",
+      "▁ East",
+      "正确 的",
+      "I O",
+      "on ic",
+      "oni c",
+      "o nic",
+      "re es",
+      "ree s",
+      "r ees",
+      "is ions",
+      "ision s",
+      "isi ons",
+      "ial og",
+      "ia log",
+      "结 婚",
+      "减 半",
+      "he ns",
+      "hen s",
+      "h ens",
+      "Lin e",
+      "Li ne",
+      "L ine",
+      "ca n",
+      "c an",
+      "下 载",
+      "om etry",
+      "ome try",
+      "omet ry",
+      "o metry",
+      "▁coll abor",
+      "Gr oup",
+      "G roup",
+      "Ur l",
+      "U rl",
+      "题 目",
+      "an ged",
+      "ang ed",
+      "ange d",
+      "▁co sts",
+      "▁cost s",
+      "▁cos ts",
+      "▁W IT",
+      "▁WI T",
+      "战 略",
+      "▁M et",
+      "▁Me t",
+      "▁ Met",
+      "▁F rench",
+      "▁Fr ench",
+      "▁Fre nch",
+      "▁ French",
+      "com mand",
+      "comm and",
+      "fr amework",
+      "frame work",
+      "fram ework",
+      "E X",
+      "▁p rem",
+      "▁pre m",
+      "▁pr em",
+      "▁ prem",
+      "F T",
+      "▁d ied",
+      "▁di ed",
+      "▁die d",
+      "▁ died",
+      "ra te",
+      "rat e",
+      "r ate",
+      "es is",
+      "esi s",
+      "e sis",
+      "自 由",
+      "▁r ot",
+      "▁ro t",
+      "▁ rot",
+      "▁W eb",
+      "▁We b",
+      "▁ Web",
+      "已 知",
+      "LI C",
+      "L IC",
+      "▁h appy",
+      "▁happ y",
+      "▁hap py",
+      "▁ happy",
+      "▁Can ada",
+      "▁Canad a",
+      "▁ Canada",
+      "▁v ert",
+      "▁ve rt",
+      "▁ver t",
+      "▁ vert",
+      "▁m ac",
+      "▁ma c",
+      "▁ mac",
+      "UR L",
+      "U RL",
+      "▁op eration",
+      "▁oper ation",
+      "▁opera tion",
+      "▁ operation",
+      "at ically",
+      "atic ally",
+      "a tically",
+      "N S",
+      "But ton",
+      "B utton",
+      "pi r",
+      "p ir",
+      "▁a uth",
+      "▁aut h",
+      "▁au th",
+      "▁ auth",
+      "▁a spect",
+      "▁as pect",
+      "▁asp ect",
+      "▁ aspect",
+      "#### ############",
+      "######## ########",
+      "############ ####",
+      "iv o",
+      "i vo",
+      "ac tor",
+      "act or",
+      "a ctor",
+      "若 干",
+      "doc s",
+      "do cs",
+      "d ocs",
+      "r d",
+      "me m",
+      "m em",
+      "vill e",
+      "vil le",
+      "v ille",
+      "▁c yt",
+      "▁cy t",
+      "▁ cyt",
+      "EN SE",
+      "ENS E",
+      "不 错",
+      "semb ly",
+      "sem bly",
+      "sembl y",
+      "▁w idth",
+      "▁wid th",
+      "▁ width",
+      "am ma",
+      "amm a",
+      "a mma",
+      "▁Al so",
+      "▁ Also",
+      "ic ks",
+      "ick s",
+      "i cks",
+      "or por",
+      "orp or",
+      "us r",
+      "u sr",
+      "x x",
+      "董 事",
+      "a o",
+      "反 应",
+      "it ors",
+      "itor s",
+      "ito rs",
+      "▁set tings",
+      "▁setting s",
+      "▁sett ings",
+      "▁ settings",
+      "▁W indows",
+      "▁Wind ows",
+      "▁Window s",
+      "▁ Windows",
+      "▁e th",
+      "▁et h",
+      "▁ eth",
+      "统 一",
+      "in ct",
+      "inc t",
+      "▁c ore",
+      "▁co re",
+      "▁cor e",
+      "▁ core",
+      "事 故",
+      "▁de scription",
+      "▁des cription",
+      "▁descript ion",
+      "▁ description",
+      "▁pres ident",
+      "▁ president",
+      "▁pr ime",
+      "▁prim e",
+      "▁pri me",
+      "▁ prime",
+      "▁v iol",
+      "▁vi ol",
+      "▁ viol",
+      "▁ne igh",
+      "▁m ount",
+      "▁mo unt",
+      "▁ mount",
+      "it is",
+      "iti s",
+      "i tis",
+      "▁z ero",
+      "▁ze ro",
+      "▁ zero",
+      "最 近",
+      "鉴 定",
+      "干 部",
+      "▁s ummer",
+      "▁sum mer",
+      "▁summ er",
+      "con f",
+      "co nf",
+      "▁requ irements",
+      "▁require ments",
+      "▁requirement s",
+      "▁ requirements",
+      "▁R em",
+      "▁Re m",
+      "▁ Rem",
+      "▁w indow",
+      "▁win dow",
+      "▁wind ow",
+      "▁ window",
+      "动 物",
+      "▁inter esting",
+      "▁interest ing",
+      "▁o w",
+      "▁ ow",
+      "▁wh ose",
+      "▁who se",
+      "肯 定",
+      "l s",
+      "▁s cale",
+      "▁sc ale",
+      "▁scal e",
+      "▁ scale",
+      "安徽 省",
+      "prot ected",
+      "▁d ed",
+      "▁de d",
+      "▁ ded",
+      "▁f amilies",
+      "▁famil ies",
+      "▁ families",
+      "il itary",
+      "ilit ary",
+      "▁s ets",
+      "▁se ts",
+      "▁set s",
+      "▁ sets",
+      "▁ex actly",
+      "▁exact ly",
+      "re quire",
+      "requ ire",
+      "▁D on",
+      "▁Do n",
+      "▁ Don",
+      "发 表",
+      "▁h eat",
+      "▁he at",
+      "▁ heat",
+      "▁te ams",
+      "▁team s",
+      "▁tea ms",
+      "▁ teams",
+      "来 了",
+      "▁com fort",
+      "▁ comfort",
+      "▁I D",
+      "▁ ID",
+      "du le",
+      "d ule",
+      "ap ache",
+      "apa che",
+      "Con n",
+      "Co nn",
+      "C onn",
+      "▁s eg",
+      "▁se g",
+      "▁ seg",
+      "父 母",
+      "复 杂",
+      "▁me asured",
+      "▁meas ured",
+      "▁measure d",
+      "▁comb ination",
+      "▁combin ation",
+      "▁Ph il",
+      "▁ Phil",
+      "id den",
+      "idd en",
+      "▁ab sol",
+      "▁abs ol",
+      "▁ex erc",
+      "▁d rive",
+      "▁dr ive",
+      "▁dri ve",
+      "▁driv e",
+      "▁ drive",
+      "▁h op",
+      "▁ho p",
+      "▁ hop",
+      "大 小",
+      "十 分",
+      "▁S ince",
+      "▁Si nce",
+      "▁Sin ce",
+      "▁ Since",
+      "▁col lect",
+      "▁coll ect",
+      "▁colle ct",
+      "▁ collect",
+      "▁mon itor",
+      "▁ monitor",
+      "▁com pletely",
+      "▁complet ely",
+      "▁complete ly",
+      "Pro perty",
+      "Prop erty",
+      "cl ose",
+      "c lose",
+      "▁w ar",
+      "▁wa r",
+      "▁ war",
+      "▁c ycl",
+      "▁cy cl",
+      "▁ cycl",
+      "▁f re",
+      "▁fr e",
+      "▁ fre",
+      "▁s old",
+      "▁so ld",
+      "▁sol d",
+      "的解 释",
+      "的 解释",
+      "wi de",
+      "wid e",
+      "w ide",
+      "▁s olid",
+      "▁sol id",
+      "▁ solid",
+      "▁/ >\\",
+      "▁/> \\",
+      "▁ />\\",
+      "▁c ardi",
+      "▁car di",
+      "▁card i",
+      "▁n et",
+      "▁ne t",
+      "▁ net",
+      "RA NT",
+      "RAN T",
+      "R ANT",
+      "▁p roduce",
+      "▁produ ce",
+      "▁prod uce",
+      "▁reg arding",
+      "▁regard ing",
+      "道 路",
+      "an not",
+      "ann ot",
+      "anno t",
+      "a nnot",
+      "▁t ag",
+      "▁ta g",
+      "▁ tag",
+      "的重 要",
+      "的 重要",
+      "P A",
+      "ve st",
+      "ves t",
+      "v est",
+      "ou nter",
+      "ount er",
+      "oun ter",
+      "o unter",
+      "▁t ree",
+      "▁tr ee",
+      "▁tre e",
+      "▁ tree",
+      "打 开",
+      "形 的",
+      "gr aph",
+      "gra ph",
+      "g raph",
+      "ug ht",
+      "ugh t",
+      "u ght",
+      "pl ant",
+      "plan t",
+      "引 起",
+      "te e",
+      "t ee",
+      "观 察",
+      "▁Re public",
+      "▁Rep ublic",
+      "uff er",
+      "uf fer",
+      "▁s ympt",
+      "▁sym pt",
+      "mat rix",
+      "▁w ood",
+      "▁wo od",
+      "▁ wood",
+      "▁j son",
+      "▁js on",
+      "▁ json",
+      "文 名",
+      "▁V ersion",
+      "▁Vers ion",
+      "▁ Version",
+      "▁hand le",
+      "▁ handle",
+      "▁t owards",
+      "▁to wards",
+      "▁toward s",
+      "▁tow ards",
+      "特 征",
+      "描 述",
+      "se m",
+      "s em",
+      "▁de aths",
+      "▁death s",
+      "▁organ ization",
+      "▁ organization",
+      "冻 结",
+      "必 要",
+      "▁p ark",
+      "▁par k",
+      "▁pa rk",
+      "▁ park",
+      "我 是",
+      "温 度",
+      "▁n ames",
+      "▁name s",
+      "▁na mes",
+      "▁nam es",
+      "▁ names",
+      "▁in nov",
+      "▁inn ov",
+      "技 能",
+      "▁ex pand",
+      "▁exp and",
+      "▁ expand",
+      "▁o x",
+      "▁ ox",
+      "▁ch annel",
+      "▁chan nel",
+      "▁ channel",
+      "ri er",
+      "rie r",
+      "r ier",
+      "▁h ospital",
+      "▁hospit al",
+      "▁ hospital",
+      "AR T",
+      "A RT",
+      "▁sign al",
+      "▁sig nal",
+      "▁ signal",
+      "高 的",
+      "▁in form",
+      "▁inf orm",
+      "▁info rm",
+      "▁ inform",
+      "▁s ales",
+      "▁sa les",
+      "▁sal es",
+      "▁sale s",
+      "▁ sales",
+      "▁s election",
+      "▁se lection",
+      "▁select ion",
+      "▁sel ection",
+      "▁ selection",
+      "ens or",
+      "ik i",
+      "i ki",
+      "▁W illi",
+      "▁Will i",
+      "▁Wil li",
+      "▁Wi lli",
+      "U s",
+      "▁techn ique",
+      "▁b all",
+      "▁bal l",
+      "▁ba ll",
+      "▁ ball",
+      "▁dam age",
+      "▁ damage",
+      "▁re source",
+      "▁res ource",
+      "▁ resource",
+      "化 学",
+      "▁H el",
+      "▁He l",
+      "▁ Hel",
+      "par ams",
+      "param s",
+      "para ms",
+      "▁Pro gram",
+      "▁Pr ogram",
+      "▁ Program",
+      "ic ated",
+      "ica ted",
+      "icate d",
+      "▁adv ant",
+      "▁ advant",
+      "he ader",
+      "head er",
+      "hea der",
+      "h eader",
+      "Fr om",
+      "F rom",
+      "per ties",
+      "pert ies",
+      "▁n or",
+      "▁no r",
+      "▁ nor",
+      "▁p olitical",
+      "▁polit ical",
+      "E M",
+      "▁v ar",
+      "▁va r",
+      "▁ var",
+      "in ner",
+      "inn er",
+      "▁b ank",
+      "▁ban k",
+      "▁ba nk",
+      "▁ bank",
+      "日 立案",
+      "▁w orth",
+      "▁wor th",
+      "▁ worth",
+      "▁inter view",
+      "▁u r",
+      "▁ ur",
+      "▁aff ect",
+      "▁af fect",
+      "ic ian",
+      "ici an",
+      "icia n",
+      "nhttp s",
+      "n https",
+      "规 模",
+      "▁l if",
+      "▁li f",
+      "▁ lif",
+      "▁f lex",
+      "▁fl ex",
+      "▁fle x",
+      "▁ flex",
+      "里 面",
+      "St r",
+      "S tr",
+      "▁F orm",
+      "▁For m",
+      "▁Fo rm",
+      "▁ Form",
+      "▁So ftware",
+      "▁Soft ware",
+      "▁ Software",
+      "▁g enes",
+      "▁gen es",
+      "▁gene s",
+      "▁ge nes",
+      "▁ genes",
+      "ri d",
+      "r id",
+      "▁struct ures",
+      "▁structure s",
+      "▁ structures",
+      "nv ar",
+      "n var",
+      "▁D i",
+      "▁ Di",
+      "ha s",
+      "h as",
+      "成 了",
+      "世 纪",
+      "ic ensed",
+      "icense d",
+      "icens ed",
+      "相 对",
+      "▁camp aign",
+      "▁ campaign",
+      "Com mand",
+      "Comm and",
+      "并 不",
+      "▁pol ice",
+      "▁polic e",
+      "▁ 页",
+      "之间 的",
+      "之 间的",
+      "F L",
+      "▁R iver",
+      "▁Ri ver",
+      "▁Riv er",
+      "in ese",
+      "ine se",
+      "ines e",
+      "im ent",
+      "ime nt",
+      "imen t",
+      "i ment",
+      "it ch",
+      "itc h",
+      "▁M or",
+      "▁Mo r",
+      "▁ Mor",
+      "h o",
+      "▁s core",
+      "▁sc ore",
+      "▁ score",
+      "Fact ory",
+      "Factor y",
+      "F actory",
+      "从 事",
+      "▁re str",
+      "▁r estr",
+      "▁res tr",
+      "▁rest r",
+      "ym ph",
+      "y mph",
+      "Mod ule",
+      "Mo dule",
+      "M odule",
+      "Le t",
+      "L et",
+      "nf or",
+      "n for",
+      "一 份",
+      "pr et",
+      "pre t",
+      "p ret",
+      "be ta",
+      "bet a",
+      "b eta",
+      "fr ee",
+      "fre e",
+      "f ree",
+      "ig ration",
+      "igr ation",
+      "ua te",
+      "u ate",
+      "▁a uto",
+      "▁aut o",
+      "▁au to",
+      "▁ auto",
+      "▁go als",
+      "▁goal s",
+      "st ack",
+      "sta ck",
+      "撤回 起诉",
+      "At t",
+      "A tt",
+      "▁g old",
+      "▁go ld",
+      "▁gol d",
+      "▁ gold",
+      "▁dev ices",
+      "▁device s",
+      "▁ devices",
+      "is hing",
+      "ish ing",
+      "ishi ng",
+      "▁L aw",
+      "▁La w",
+      "▁ Law",
+      "分 为",
+      "ri ve",
+      "riv e",
+      "r ive",
+      "la s",
+      "l as",
+      "we ight",
+      "wei ght",
+      "w eight",
+      "▁s ell",
+      "▁se ll",
+      "▁sel l",
+      "▁ sell",
+      "▁g oes",
+      "▁go es",
+      "▁str ategy",
+      "▁strateg y",
+      "▁ strategy",
+      "减半 收取",
+      "判 处",
+      "▁c reating",
+      "▁cre ating",
+      "▁creat ing",
+      "▁ creating",
+      "▁cor rel",
+      "平方 米",
+      "▁ann ounced",
+      "▁announce d",
+      "▁announc ed",
+      "起 点",
+      "成 长",
+      "夫 妻",
+      "香 港",
+      "台 湾",
+      "ace book",
+      "he lp",
+      "hel p",
+      "h elp",
+      "mar gin",
+      "m argin",
+      "人 类",
+      "nA nswer",
+      "n Answer",
+      "确定 的",
+      "确 定的",
+      "▁g reen",
+      "▁gr een",
+      "▁gre en",
+      "▁ green",
+      "nA fter",
+      "n After",
+      "时 期",
+      "ho me",
+      "hom e",
+      "h ome",
+      "▁appro priate",
+      "▁appropri ate",
+      "▁ appropriate",
+      "ed eral",
+      "eder al",
+      "ede ral",
+      "▁c atch",
+      "▁cat ch",
+      "▁ catch",
+      "工 资",
+      "sp ring",
+      "spr ing",
+      "s pring",
+      "▁Sup pose",
+      "▁ Suppose",
+      "ad ding",
+      "add ing",
+      "use um",
+      "房地 产",
+      "房 地产",
+      "is tor",
+      "ist or",
+      "isto r",
+      "i stor",
+      "▁B el",
+      "▁Be l",
+      "▁ Bel",
+      "如 此",
+      "▁quick ly",
+      "ne l",
+      "n el",
+      "To ken",
+      "Tok en",
+      "T oken",
+      "子 的",
+      "C I",
+      "有 点",
+      "采 取",
+      "O n",
+      "▁l abor",
+      "▁la bor",
+      "▁lab or",
+      "相 信",
+      "不 少",
+      "办 理",
+      "ain e",
+      "ai ne",
+      "a ine",
+      "▁V ir",
+      "▁Vi r",
+      "▁ Vir",
+      "▁r out",
+      "▁ro ut",
+      "▁rou t",
+      "Su p",
+      "S up",
+      "▁ad apt",
+      "▁adap t",
+      "▁ adapt",
+      "▁H T",
+      "▁ HT",
+      "▁g ets",
+      "▁get s",
+      "▁ge ts",
+      "▁ gets",
+      "有限 责任",
+      "▁inter ested",
+      "▁interest ed",
+      "▁re levant",
+      "▁ relevant",
+      "RN A",
+      "R NA",
+      "商 务",
+      "▁B o",
+      "▁ Bo",
+      "▁St reet",
+      "▁Stre et",
+      "▁ Street",
+      "ro y",
+      "r oy",
+      "nU n",
+      "n Un",
+      "ad ian",
+      "adi an",
+      "adia n",
+      "R I",
+      "la mm",
+      "lam m",
+      "l amm",
+      "▁con tain",
+      "▁cont ain",
+      "um mer",
+      "umm er",
+      "es ts",
+      "est s",
+      "e sts",
+      "第一百四十 五",
+      "sup er",
+      "su per",
+      "s uper",
+      "im er",
+      "ime r",
+      "i mer",
+      "Qu ery",
+      "Que ry",
+      "CO N",
+      "C ON",
+      "Au t",
+      "A ut",
+      "ad ata",
+      "ada ta",
+      "a data",
+      "▁E ar",
+      "推 进",
+      "解 题",
+      "居 民",
+      "in i",
+      "i ni",
+      "▁conn ection",
+      "▁connect ion",
+      "▁ connection",
+      "R L",
+      "重要 的",
+      "ex ec",
+      "exe c",
+      "有限责任 公司",
+      "▁inf ection",
+      "▁infect ion",
+      "▁ infection",
+      "教 授",
+      "对 象",
+      "妈 妈",
+      "hed ul",
+      "▁view s",
+      "▁vi ews",
+      "▁ views",
+      "▁S ing",
+      "▁Si ng",
+      "▁Sin g",
+      "▁ Sing",
+      "▁w onder",
+      "▁won der",
+      "▁wond er",
+      "▁wo nder",
+      "ograph ic",
+      "o graphic",
+      "接 口",
+      "▁st eps",
+      "▁step s",
+      "▁ste ps",
+      "▁ steps",
+      "▁D iv",
+      "▁Di v",
+      "▁ Div",
+      "▁M al",
+      "▁Ma l",
+      "▁ Mal",
+      "op ic",
+      "o pic",
+      "▁l es",
+      "▁le s",
+      "▁ les",
+      "▁l ives",
+      "▁li ves",
+      "▁live s",
+      "▁liv es",
+      "▁requ ires",
+      "▁require s",
+      "▁ requires",
+      "▁t reated",
+      "▁treat ed",
+      "▁tre ated",
+      "▁ treated",
+      "确 实",
+      "手 术",
+      "▁rel ative",
+      "▁ relative",
+      "ge bra",
+      "g ebra",
+      "▁O ver",
+      "▁ Over",
+      "▁WAR RANT",
+      "ag o",
+      "a go",
+      "陪 审",
+      "▁N OT",
+      "▁NO T",
+      "▁ NOT",
+      "par ser",
+      "parse r",
+      "pars er",
+      "▁F il",
+      "▁Fi l",
+      "▁ Fil",
+      "Build er",
+      "▁ap point",
+      "▁app oint",
+      "▁ appoint",
+      "▁inf luence",
+      "▁influ ence",
+      "结 束",
+      "ur t",
+      "u rt",
+      "▁evalu ate",
+      "▁eval uate",
+      "▁ evaluate",
+      "条 的规定",
+      "▁l arger",
+      "▁large r",
+      "▁larg er",
+      "▁lar ger",
+      "▁m oved",
+      "▁mov ed",
+      "▁move d",
+      "▁mo ved",
+      "iss ues",
+      "issue s",
+      "issu es",
+      "详 细",
+      "ab y",
+      "a by",
+      "▁B en",
+      "▁Be n",
+      "▁ Ben",
+      "F F",
+      "l ambda",
+      "训 练",
+      "k i",
+      "到庭 参加",
+      "n i",
+      "▁health y",
+      "▁heal thy",
+      "▁ healthy",
+      "Di s",
+      "D is",
+      "en ty",
+      "ent y",
+      "e nty",
+      "an o",
+      "a no",
+      "U I",
+      "ig ma",
+      "igm a",
+      "nh ttp",
+      "n http",
+      "Ent ity",
+      "规 范",
+      "▁ca used",
+      "▁cause d",
+      "▁caus ed",
+      "当 前",
+      "▁part icularly",
+      "▁particular ly",
+      "陪审 员",
+      "掌 握",
+      "▁im plementation",
+      "▁implement ation",
+      "▁ implementation",
+      "np rint",
+      "npr int",
+      "n print",
+      "▁p las",
+      "▁pl as",
+      "▁com mercial",
+      "▁ commercial",
+      "▁h ol",
+      "▁ho l",
+      "▁ hol",
+      "re ated",
+      "reat ed",
+      "reate d",
+      "rea ted",
+      "▁H ome",
+      "▁Hom e",
+      "▁Ho me",
+      "▁ Home",
+      "涉 及",
+      "er ate",
+      "era te",
+      "e rate",
+      "▁s ession",
+      "▁sess ion",
+      "▁ session",
+      "▁p ul",
+      "▁pu l",
+      "nt h",
+      "n th",
+      "od ies",
+      "odi es",
+      "o dies",
+      "报 道",
+      "ne tic",
+      "net ic",
+      "n etic",
+      "公 共",
+      "nv oid",
+      "n void",
+      "的 最",
+      "▁e r",
+      "▁ er",
+      "k w",
+      "尤 其",
+      "Equ al",
+      "Eq ual",
+      "E qual",
+      "ay ers",
+      "ayer s",
+      "aye rs",
+      "a yers",
+      "以 前",
+      "▁arch itect",
+      "Ru n",
+      "R un",
+      "▁re ports",
+      "▁rep orts",
+      "▁report s",
+      "▁ reports",
+      "ri a",
+      "r ia",
+      "日 止",
+      "危 险",
+      "▁con ver",
+      "▁conv er",
+      "▁conve r",
+      "▁p air",
+      "▁pa ir",
+      "▁ pair",
+      "本 文",
+      "▁inj ury",
+      "公 式",
+      "ia t",
+      "i at",
+      "in stance",
+      "inst ance",
+      "▁t ested",
+      "▁te sted",
+      "▁test ed",
+      "▁ tested",
+      "上诉 状",
+      "U E",
+      "na v",
+      "n av",
+      "她 的",
+      "nP eople",
+      "nPe ople",
+      "n People",
+      "es ter",
+      "est er",
+      "este r",
+      "e ster",
+      "U til",
+      "▁pl ans",
+      "▁plan s",
+      "on day",
+      "ond ay",
+      "onda y",
+      "▁F ind",
+      "▁Fin d",
+      "▁Fi nd",
+      "▁ Find",
+      "避 免",
+      "▁l egal",
+      "▁le gal",
+      "▁leg al",
+      "▁ legal",
+      "的 值",
+      "ci rc",
+      "cir c",
+      "c irc",
+      "ag on",
+      "ago n",
+      "a gon",
+      "本裁定送达后 即发生法律效力",
+      "▁f emale",
+      "▁fe male",
+      "▁fem ale",
+      "▁ female",
+      "▁reve aled",
+      "▁reveal ed",
+      "▁S olution",
+      "▁Sol ution",
+      "▁ Solution",
+      "▁c elebr",
+      "▁cele br",
+      "行 的",
+      "E O",
+      "▁p oor",
+      "▁po or",
+      "偿 还",
+      "ounc il",
+      "▁der ivative",
+      "▁deriv ative",
+      "人民 陪审员",
+      "▁S te",
+      "▁St e",
+      "▁ Ste",
+      "微 信",
+      "id ing",
+      "idi ng",
+      "idin g",
+      "i ding",
+      "as tr",
+      "ast r",
+      "a str",
+      "河南 省",
+      "河 南省",
+      "ma s",
+      "m as",
+      "▁cap acity",
+      "▁capac ity",
+      "▁ capacity",
+      "▁te levision",
+      "▁tele vision",
+      "▁telev ision",
+      "▁c lasses",
+      "▁class es",
+      "▁ classes",
+      "四川 省",
+      "▁ser ious",
+      "▁p ython",
+      "▁py thon",
+      "▁pyt hon",
+      "▁ python",
+      "IT Y",
+      "I TY",
+      "运 输",
+      "▁equ ipment",
+      "▁equip ment",
+      "▁V ol",
+      "▁Vo l",
+      "▁ Vol",
+      "Creat e",
+      "Cre ate",
+      "C reate",
+      "本 书",
+      "▁pl anning",
+      "▁plan ning",
+      "独 立",
+      "nW ith",
+      "n With",
+      "pe d",
+      "p ed",
+      "的 新",
+      "四 年",
+      "si m",
+      "s im",
+      "b i",
+      "审判 长",
+      "▁t rain",
+      "▁tr ain",
+      "▁tra in",
+      "▁ train",
+      "▁U sing",
+      "▁Us ing",
+      "▁ Using",
+      "St ream",
+      "▁f a",
+      "▁ fa",
+      "绝 对",
+      "▁r ules",
+      "▁rule s",
+      "▁rul es",
+      "▁ru les",
+      "▁ rules",
+      "uc ed",
+      "uce d",
+      "u ced",
+      "xm l",
+      "x ml",
+      "▁Ass ociation",
+      "▁Associ ation",
+      "▁ Association",
+      "▁I ll",
+      "▁Il l",
+      "▁ Ill",
+      "▁develop ing",
+      "▁re ceptor",
+      "▁recept or",
+      "▁ receptor",
+      "创 造",
+      "面 对",
+      "▁form ation",
+      "▁format ion",
+      "▁ formation",
+      "pro du",
+      "prod u",
+      "p rodu",
+      "机 制",
+      "▁s en",
+      "▁se n",
+      "▁ sen",
+      "{ %",
+      "▁Fr iday",
+      "▁Fri day",
+      "▁ Friday",
+      "for ce",
+      "f orce",
+      "力 量",
+      "▁en zym",
+      "ib ly",
+      "ibl y",
+      "i bly",
+      "or ders",
+      "ord ers",
+      "order s",
+      "orde rs",
+      "维 护",
+      "▁d ensity",
+      "▁dens ity",
+      "▁ density",
+      "▁ 公司",
+      "深 入",
+      "文 明",
+      "c i",
+      "▁g enerate",
+      "▁gener ate",
+      "▁gen erate",
+      "▁gene rate",
+      "▁genera te",
+      "▁ generate",
+      "▁sh ape",
+      "▁sha pe",
+      "▁ shape",
+      "nm s",
+      "n ms",
+      "▁sc ience",
+      "▁sci ence",
+      "▁ science",
+      "fi x",
+      "f ix",
+      "▁Found ation",
+      "▁ Foundation",
+      "Vers ion",
+      "V ersion",
+      "▁m oving",
+      "▁mov ing",
+      "▁mo ving",
+      "▁ moving",
+      "IT E",
+      "I TE",
+      "整 体",
+      "▁D avid",
+      "▁Da vid",
+      "▁Dav id",
+      "▁ David",
+      "▁( $",
+      "▁ ($",
+      "▁s yn",
+      "▁sy n",
+      "▁ syn",
+      "if ically",
+      "ific ally",
+      "▁h air",
+      "▁ha ir",
+      "▁ hair",
+      "▁benef its",
+      "▁benefit s",
+      "We b",
+      "W eb",
+      "消费 者",
+      "de lete",
+      "del ete",
+      "delet e",
+      "ca rd",
+      "car d",
+      "c ard",
+      "se s",
+      "s es",
+      "N O",
+      "▁m iddle",
+      "▁ middle",
+      "▁p icture",
+      "▁pict ure",
+      "▁pic ture",
+      "▁ picture",
+      "Lo ad",
+      "L oad",
+      "皮 肤",
+      "理 念",
+      "▁par ents",
+      "▁parent s",
+      "▁paren ts",
+      "▁ parents",
+      "Tab le",
+      "Ta ble",
+      "T able",
+      "▁T O",
+      "▁ TO",
+      "指 控",
+      "und le",
+      "▁st ream",
+      "▁stre am",
+      "▁ stream",
+      "E C",
+      "$$ \\\\",
+      "$$\\ \\",
+      "$ $\\\\",
+      "▁c ert",
+      "▁cer t",
+      "▁ce rt",
+      "▁ cert",
+      "s g",
+      "▁large st",
+      "▁larg est",
+      "▁lar gest",
+      "▁ largest",
+      "er ry",
+      "err y",
+      "u a",
+      "贸 易",
+      "▁T ime",
+      "▁Tim e",
+      "▁Ti me",
+      "▁ Time",
+      "nI s",
+      "n Is",
+      "ru m",
+      "r um",
+      "s w",
+      "自 身",
+      "he dule",
+      "hed ule",
+      "hedul e",
+      "▁re duction",
+      "▁redu ction",
+      "▁red uction",
+      "▁reduct ion",
+      "▁ reduction",
+      "▁Tr ump",
+      "▁ Trump",
+      "it chen",
+      "itc hen",
+      "itch en",
+      "分 布",
+      "讨 论",
+      "▁G reat",
+      "▁Gr eat",
+      "▁Gre at",
+      "▁ Great",
+      "▁A nal",
+      "▁An al",
+      "▁Ana l",
+      "▁ Anal",
+      "▁e lectric",
+      "▁elect ric",
+      "▁electr ic",
+      "▁ electric",
+      "特 殊",
+      "ry st",
+      "rys t",
+      "r yst",
+      "ac ters",
+      "act ers",
+      "acter s",
+      "▁M odel",
+      "▁Mod el",
+      "▁Mo del",
+      "▁Mode l",
+      "▁ Model",
+      "ve ctor",
+      "vec tor",
+      "v ector",
+      "▁c at",
+      "▁ca t",
+      "▁ cat",
+      "▁f resh",
+      "▁fr esh",
+      "▁fre sh",
+      "▁fres h",
+      "▁ fresh",
+      "一定 要",
+      "re q",
+      "r eq",
+      "▁act ual",
+      "▁ac tual",
+      "▁actu al",
+      "▁ actual",
+      "▁respon ses",
+      "▁response s",
+      "▁respons es",
+      "▁ responses",
+      "P D",
+      "▁re pe",
+      "▁rep e",
+      "flo at",
+      "f loat",
+      "▁al though",
+      "▁ although",
+      "▁j ud",
+      "▁ju d",
+      "▁ jud",
+      "nA ll",
+      "nAl l",
+      "n All",
+      "▁W ash",
+      "▁Was h",
+      "▁Wa sh",
+      "u y",
+      "包 含",
+      "▁f igure",
+      "▁fig ure",
+      "▁ figure",
+      "▁form ula",
+      "▁ formula",
+      "▁U se",
+      "▁Us e",
+      "▁ Use",
+      "角 色",
+      "▁b us",
+      "▁bu s",
+      "▁ bus",
+      "▁b ud",
+      "▁bu d",
+      "▁ bud",
+      "▁s erve",
+      "▁serv e",
+      "▁ser ve",
+      "▁ serve",
+      "多 次",
+      "南 京",
+      "合 议",
+      "el lo",
+      "ell o",
+      "▁w rote",
+      "▁wr ote",
+      "重 新",
+      "line s",
+      "li nes",
+      "lin es",
+      "l ines",
+      "f alse",
+      "er ia",
+      "eri a",
+      "e ria",
+      "利 率",
+      "地 点",
+      "ie n",
+      "i en",
+      "go ogle",
+      "goog le",
+      "了一 个",
+      "了 一个",
+      "▁Eng land",
+      "▁f at",
+      "▁fa t",
+      "▁ fat",
+      "▁p y",
+      "▁ py",
+      "▁b inding",
+      "▁bin ding",
+      "▁bind ing",
+      "▁ binding",
+      "推 动",
+      "▁all owed",
+      "▁allow ed",
+      "▁ allowed",
+      "它 们",
+      "sy s",
+      "s ys",
+      "M O",
+      "oo m",
+      "o om",
+      "▁s ources",
+      "▁source s",
+      "▁sour ces",
+      "▁ sources",
+      "▁re member",
+      "▁rem ember",
+      "▁ remember",
+      "发 挥",
+      "▁ch ance",
+      "▁cha nce",
+      "▁chan ce",
+      "case s",
+      "ca ses",
+      "cas es",
+      "c ases",
+      "k y",
+      "pro perty",
+      "prop erty",
+      "proper ty",
+      "计算 机",
+      "▁c m",
+      "▁ cm",
+      "风 格",
+      "ide os",
+      "ideo s",
+      "▁B ook",
+      "▁Bo ok",
+      "▁ Book",
+      "先 生",
+      "▁me asures",
+      "▁meas ures",
+      "▁measure s",
+      "display style",
+      "em plate",
+      "emp late",
+      "empl ate",
+      "限 制",
+      "ol f",
+      "o lf",
+      "os c",
+      "o sc",
+      "▁pro cessing",
+      "▁process ing",
+      "▁proc essing",
+      "▁ processing",
+      "▁s aw",
+      "▁sa w",
+      "十日 内",
+      "十 日内",
+      "武 汉",
+      "相 同",
+      "▁s urgery",
+      "▁surg ery",
+      "▁surge ry",
+      "▁dec ided",
+      "▁decide d",
+      "▁s ou",
+      "▁so u",
+      "IE S",
+      "I ES",
+      "▁b utton",
+      "▁but ton",
+      "▁butt on",
+      "▁ button",
+      "st ore",
+      "sto re",
+      "stor e",
+      "负责 人",
+      "sl ant",
+      "ce nter",
+      "cent er",
+      "c enter",
+      "▁r ock",
+      "▁ro ck",
+      "▁ rock",
+      "▁T ex",
+      "▁Te x",
+      "▁ Tex",
+      "无 论",
+      "▁C ouncil",
+      "资 产",
+      "L L",
+      "▁ex tract",
+      "▁extra ct",
+      "▁extr act",
+      "▁extrac t",
+      "▁ extract",
+      "▁im proved",
+      "▁impro ved",
+      "▁improve d",
+      "▁improv ed",
+      "ki ng",
+      "kin g",
+      "k ing",
+      "ber g",
+      "be rg",
+      "b erg",
+      "▁exper imental",
+      "▁experim ental",
+      "▁experi mental",
+      "▁experiment al",
+      "▁ experimental",
+      "▁w oman",
+      "▁wo man",
+      "▁ woman",
+      "三 年",
+      "nA t",
+      "n At",
+      "▁m aster",
+      "▁ma ster",
+      "▁mast er",
+      "▁mas ter",
+      "▁ master",
+      "iv il",
+      "ivi l",
+      "i vil",
+      "合议 庭",
+      "▁S T",
+      "▁ ST",
+      "▁there fore",
+      "▁ther efore",
+      "▁ therefore",
+      "▁D el",
+      "▁De l",
+      "▁ Del",
+      "▁del ivery",
+      "▁deliver y",
+      "▁ delivery",
+      "▁m ys",
+      "▁my s",
+      "▁ mys",
+      "▁t ouch",
+      "▁to uch",
+      "▁tou ch",
+      "▁ touch",
+      "▁part ners",
+      "▁partner s",
+      "数 是",
+      "▁l isted",
+      "▁li sted",
+      "▁list ed",
+      "▁ listed",
+      "▁div ided",
+      "▁divid ed",
+      "▁divide d",
+      "nTh anks",
+      "nThank s",
+      "n Thanks",
+      "es h",
+      "e sh",
+      "营 养",
+      "▁trans l",
+      "▁tra nsl",
+      "▁tran sl",
+      "▁ transl",
+      "消 息",
+      "▁W ed",
+      "▁We d",
+      "▁help s",
+      "▁hel ps",
+      "▁g iving",
+      "▁giv ing",
+      "▁gi ving",
+      "▁ giving",
+      "nor mal",
+      "norm al",
+      "n ormal",
+      "if ying",
+      "ify ing",
+      "▁In stitute",
+      "▁Instit ute",
+      "pat ch",
+      "p atch",
+      "▁enc our",
+      "Num ber",
+      "Nu mber",
+      "N umber",
+      "▁v ict",
+      "▁vi ct",
+      "▁vic t",
+      "ri ef",
+      "rie f",
+      "r ief",
+      "年 轻",
+      "▁def init",
+      "▁defin it",
+      "现 象",
+      "可 是",
+      "Fi l",
+      "F il",
+      "co ut",
+      "cou t",
+      "c out",
+      "图 片",
+      "异 议",
+      "M y",
+      "开 放",
+      "坐 标",
+      "▁at tract",
+      "▁attr act",
+      "二〇一 五年",
+      "pro ps",
+      "pr ops",
+      "prop s",
+      "▁O bject",
+      "▁Ob ject",
+      "▁ Object",
+      "让 人",
+      "\\\\ \"",
+      "th en",
+      "the n",
+      "t hen",
+      "▁r oll",
+      "▁ro ll",
+      "▁ roll",
+      "ar ant",
+      "ara nt",
+      "aran t",
+      "a rant",
+      "发展 的",
+      "▁N ov",
+      "▁No v",
+      "▁ Nov",
+      "▁h ous",
+      "▁ho us",
+      "▁F urther",
+      "▁Fur ther",
+      "▁外 文名",
+      "st on",
+      "sto n",
+      "s ton",
+      "▁character istics",
+      "▁characteristic s",
+      "▁c alc",
+      "▁ca lc",
+      "▁cal c",
+      "▁ calc",
+      "▁m ort",
+      "▁mor t",
+      "▁mo rt",
+      "▁ mort",
+      "照 片",
+      "AT H",
+      "A TH",
+      "▁s leep",
+      "▁sl eep",
+      "▁sle ep",
+      "▁ sleep",
+      "ph one",
+      "phon e",
+      "oy al",
+      "oya l",
+      "o yal",
+      "P R",
+      "▁school s",
+      "▁sch ools",
+      "创 业",
+      "▁连载 平台",
+      "亿 元",
+      "▁ad just",
+      "▁adj ust",
+      "▁ adjust",
+      "ha ps",
+      "h aps",
+      "▁中 国",
+      "▁ 中国",
+      "an tic",
+      "ant ic",
+      "anti c",
+      "▁s un",
+      "▁su n",
+      "▁ sun",
+      "an ges",
+      "ang es",
+      "ange s",
+      "as ket",
+      "ask et",
+      "▁th anks",
+      "▁than ks",
+      "▁thank s",
+      "▁ thanks",
+      "少 年",
+      "基 于",
+      "▁respons ible",
+      "▁S und",
+      "▁Sun d",
+      "▁Su nd",
+      "▁r apid",
+      "▁ra pid",
+      "▁rap id",
+      "pa n",
+      "p an",
+      "op tion",
+      "opt ion",
+      "o ption",
+      "▁de fine",
+      "▁def ine",
+      "▁defin e",
+      "▁ define",
+      "▁C am",
+      "▁Ca m",
+      "▁ Cam",
+      "▁s ys",
+      "▁sy s",
+      "▁ sys",
+      "▁T wo",
+      "▁Tw o",
+      "▁ Two",
+      "Ch ar",
+      "C har",
+      "▁in ternal",
+      "▁int ernal",
+      "▁inter nal",
+      "▁intern al",
+      "▁ internal",
+      "江苏 省",
+      "Ser ver",
+      "Serv er",
+      "Serve r",
+      "S erver",
+      "▁g rowing",
+      "▁gr owing",
+      "▁grow ing",
+      "▁gro wing",
+      "▁ growing",
+      "▁P rov",
+      "▁Pro v",
+      "▁Pr ov",
+      "▁ Prov",
+      "▁T H",
+      "▁ TH",
+      "ra nge",
+      "ran ge",
+      "r ange",
+      "log ger",
+      "▁res istance",
+      "▁resist ance",
+      "▁ resistance",
+      "ins tall",
+      "inst all",
+      "分 类",
+      "该 公司",
+      "et imes",
+      "etime s",
+      "eti mes",
+      "e times",
+      "is ation",
+      "isa tion",
+      "▁equ ations",
+      "▁equation s",
+      "▁eq uations",
+      "成 都",
+      "▁M ag",
+      "▁Ma g",
+      "▁ Mag",
+      "▁s ports",
+      "▁sp orts",
+      "▁sport s",
+      "▁spor ts",
+      "会 有",
+      "▁n eur",
+      "▁ne ur",
+      "▁ neur",
+      "ES S",
+      "E SS",
+      "而 是",
+      "ri m",
+      "r im",
+      "元 素",
+      "我 觉得",
+      "▁anal yz",
+      "▁analy z",
+      "▁ex cellent",
+      "▁excell ent",
+      "▁occ ur",
+      "▁oc cur",
+      "效 率",
+      "仔 细",
+      "▁ph osph",
+      "▁ phosph",
+      "▁ 刘",
+      "▁cent ury",
+      "▁ century",
+      "连 续",
+      "遇 到",
+      "▁A ny",
+      "▁An y",
+      "▁ Any",
+      "等 等",
+      "▁and roid",
+      "▁ android",
+      "ab et",
+      "abe t",
+      "a bet",
+      "▁pro file",
+      "▁prof ile",
+      "▁profil e",
+      "▁ profile",
+      "at tr",
+      "att r",
+      "行 动",
+      "▁e ight",
+      "▁ eight",
+      "▁l ib",
+      "▁li b",
+      "▁ lib",
+      "▁P al",
+      "▁Pa l",
+      "▁ Pal",
+      "数 学",
+      "ca r",
+      "c ar",
+      "▁c ounter",
+      "▁co unter",
+      "▁count er",
+      "▁cou nter",
+      "▁ counter",
+      "ig ger",
+      "igg er",
+      "不 了",
+      "值 得",
+      "比 例",
+      "ms elves",
+      "安 排",
+      "P T",
+      "最高人民法院 关于适用",
+      "最高人民法院关于 适用",
+      "▁oper ations",
+      "▁operation s",
+      "▁ operations",
+      "重庆 市",
+      "重 庆市",
+      "re te",
+      "ret e",
+      "r ete",
+      ") }",
+      "▁s tar",
+      "▁st ar",
+      "▁ star",
+      "▁v ent",
+      "▁ve nt",
+      "▁ven t",
+      "▁ vent",
+      "▁re action",
+      "▁react ion",
+      "▁ reaction",
+      "▁Re act",
+      "▁ React",
+      "▁p iece",
+      "▁pie ce",
+      "▁ piece",
+      "青 年",
+      "▁P aul",
+      "▁Pa ul",
+      "▁ Paul",
+      "▁t rust",
+      "▁tr ust",
+      "▁ trust",
+      "▁k in",
+      "▁ki n",
+      "▁ kin",
+      "▁st ock",
+      "▁sto ck",
+      "▁ stock",
+      "rib utes",
+      "ribut es",
+      "ribute s",
+      "▁sit uation",
+      "▁situ ation",
+      "it ems",
+      "ite ms",
+      "item s",
+      "构 成",
+      "▁m embr",
+      "▁mem br",
+      "nN ow",
+      "nNo w",
+      "n Now",
+      "力 的",
+      "ri cal",
+      "ric al",
+      "r ical",
+      "ic ine",
+      "ici ne",
+      "icin e",
+      "i cine",
+      "▁D ate",
+      "▁Dat e",
+      "▁Da te",
+      "▁ Date",
+      "广东 省",
+      "▁Q ue",
+      "▁Qu e",
+      "▁ Que",
+      "符合 法律规定",
+      "▁al ter",
+      "▁alt er",
+      "▁ alter",
+      "au d",
+      "a ud",
+      "▁p en",
+      "▁pe n",
+      "▁ pen",
+      "可 得",
+      "▁\\\\ (",
+      "▁ \\\\(",
+      "Js on",
+      "J son",
+      "▁f irm",
+      "▁fi rm",
+      "▁fir m",
+      "▁ firm",
+      "ar den",
+      "ard en",
+      "arde n",
+      "▁K e",
+      "▁ Ke",
+      "▁pos ted",
+      "▁post ed",
+      "▁po sted",
+      "augh ter",
+      "aught er",
+      "根 本",
+      "ra ise",
+      "rais e",
+      "r aise",
+      "进行了 审理",
+      "▁ad ding",
+      "▁add ing",
+      "▁ adding",
+      "保 全",
+      "北京 市",
+      "▁appro ximately",
+      "▁approxim ately",
+      "▁approx imately",
+      "▁approximate ly",
+      "▁ approximately",
+      "no r",
+      "n or",
+      "as er",
+      "ase r",
+      "a ser",
+      "uf eff",
+      "▁cont aining",
+      "▁contain ing",
+      "▁ containing",
+      "委 会",
+      "od ay",
+      "oda y",
+      "o day",
+      "▁comb ined",
+      "▁combin ed",
+      "▁combine d",
+      "▁ combined",
+      "▁old er",
+      "▁ol der",
+      "▁ older",
+      "▁R un",
+      "▁Ru n",
+      "▁ Run",
+      "幸 福",
+      "工 商",
+      "▁e arlier",
+      "▁ear lier",
+      "▁G reen",
+      "▁Gr een",
+      "▁Gree n",
+      "▁Gre en",
+      "▁ Green",
+      "主 题",
+      "▁rem ains",
+      "▁remain s",
+      "va n",
+      "v an",
+      "▁mor ning",
+      "ak ers",
+      "ake rs",
+      "aker s",
+      "a kers",
+      "▁the mselves",
+      "ec h",
+      "e ch",
+      "f a",
+      "on a",
+      "o na",
+      "val ues",
+      "value s",
+      "▁c arry",
+      "▁car ry",
+      "▁G erman",
+      "▁Germ an",
+      "▁Ger man",
+      "▁ German",
+      "的一 个",
+      "的 一个",
+      "os itory",
+      "osit ory",
+      "▁col umn",
+      "▁colum n",
+      "▁ column",
+      "re y",
+      "r ey",
+      "▁strateg ies",
+      "吉 林",
+      "▁d im",
+      "▁di m",
+      "▁ dim",
+      "▁Cl ub",
+      "▁ Club",
+      "的 影响",
+      "ic ate",
+      "ica te",
+      "▁l icense",
+      "▁lic ense",
+      "▁ license",
+      "成 果",
+      "球 队",
+      "▁n orth",
+      "▁nor th",
+      "▁ north",
+      "▁commun ication",
+      "▁communic ation",
+      "▁ communication",
+      "法律 文书",
+      "试 验",
+      "▁r ace",
+      "▁ra ce",
+      "▁rac e",
+      "▁ race",
+      "▁sup ply",
+      "▁suppl y",
+      "ape ut",
+      "de d",
+      "d ed",
+      "uc ture",
+      "uct ure",
+      "▁com ments",
+      "▁comm ents",
+      "▁comment s",
+      "▁ comments",
+      "▁conduct ed",
+      "ho use",
+      "hou se",
+      "h ouse",
+      "ri an",
+      "ria n",
+      "r ian",
+      "▁Al though",
+      "▁ Although",
+      "大 多",
+      "▁c redit",
+      "▁cre dit",
+      "▁cr edit",
+      "▁cred it",
+      "▁ credit",
+      "现 实",
+      "^{ \\\\",
+      "^ {\\\\",
+      "▁me aning",
+      "▁mean ing",
+      "▁ meaning",
+      "▁a ware",
+      "▁aw are",
+      "▁awa re",
+      "▁ aware",
+      "▁act ivation",
+      "▁activ ation",
+      "▁ activation",
+      "EM A",
+      "E MA",
+      "▁v ehicle",
+      "▁veh icle",
+      "▁ vehicle",
+      "▁err ors",
+      "▁error s",
+      "▁er rors",
+      "▁ errors",
+      "本 身",
+      "ap ped",
+      "app ed",
+      "appe d",
+      "a pped",
+      "▁app ears",
+      "▁appe ars",
+      "▁appear s",
+      "连 接",
+      "▁phot os",
+      "▁photo s",
+      "▁ photos",
+      "▁l ip",
+      "▁li p",
+      "▁ lip",
+      "om ial",
+      "omi al",
+      "urre nc",
+      "▁det ection",
+      "▁detect ion",
+      "▁ detection",
+      "▁H all",
+      "▁Ha ll",
+      "▁Hal l",
+      "▁d jango",
+      "▁ django",
+      "▁bo olean",
+      "▁bool ean",
+      "▁ boolean",
+      "por ary",
+      "pora ry",
+      "d i",
+      "▁k ids",
+      "▁kid s",
+      "▁ki ds",
+      "il s",
+      "i ls",
+      "▁an imals",
+      "▁anim als",
+      "▁animal s",
+      "▁M icrosoft",
+      "▁Micro soft",
+      "▁ Microsoft",
+      "▁subject s",
+      "▁G over",
+      "▁Go ver",
+      "▁Gov er",
+      "蛋 白",
+      "差 异",
+      "▁ins urance",
+      "▁s outh",
+      "▁so uth",
+      "▁sou th",
+      "▁ south",
+      "原 则",
+      ")/ (",
+      ") /(",
+      ": \"",
+      "no n",
+      "n on",
+      "▁h our",
+      "▁ho ur",
+      "▁ hour",
+      "算 法",
+      "所 在",
+      "F I",
+      "iv eness",
+      "ive ness",
+      "iven ess",
+      "的 工作",
+      "▁con vert",
+      "▁conver t",
+      "▁conv ert",
+      "▁conve rt",
+      "▁ convert",
+      "完 善",
+      "▁C heck",
+      "▁Ch eck",
+      "▁Che ck",
+      "▁ Check",
+      "▁gen etic",
+      "▁gene tic",
+      "▁ge netic",
+      "民间 借贷",
+      "▁c hemical",
+      "▁chem ical",
+      "▁ chemical",
+      "▁par ameter",
+      "▁paramet er",
+      "▁para meter",
+      "▁param eter",
+      "▁ parameter",
+      "se nd",
+      "sen d",
+      "s end",
+      "▁invest igate",
+      "▁investig ate",
+      "证 实",
+      "und red",
+      "▁t al",
+      "▁ta l",
+      "▁ tal",
+      "自治 区",
+      "' ]",
+      "▁direct ory",
+      "▁director y",
+      "▁ directory",
+      "A V",
+      "ke r",
+      "k er",
+      "机 器",
+      "UL L",
+      "U LL",
+      "▁pl us",
+      "▁ plus",
+      "背 景",
+      "▁de mand",
+      "▁dem and",
+      "▁ demand",
+      "改 善",
+      "年 代",
+      "▁s ometimes",
+      "▁some times",
+      "▁som etimes",
+      "▁sometime s",
+      "▁ sometimes",
+      "ateg ories",
+      "ategor ies",
+      "▁d ark",
+      "▁da rk",
+      "▁dar k",
+      "▁ dark",
+      "Cal l",
+      "Ca ll",
+      "C all",
+      "百 度",
+      "ro ught",
+      "rou ght",
+      "rough t",
+      "r ought",
+      "um an",
+      "uma n",
+      "u man",
+      "▁econ omic",
+      "▁ec onomic",
+      "▁econom ic",
+      "▁ economic",
+      "本 题",
+      "C K",
+      "nP l",
+      "n Pl",
+      "▁con centration",
+      "▁concent ration",
+      "▁concentr ation",
+      "组成 合议庭",
+      "▁A D",
+      "▁ AD",
+      "▁R el",
+      "▁Re l",
+      "▁ Rel",
+      "wh ich",
+      "▁S ur",
+      "▁Su r",
+      "▁ Sur",
+      "▁r ank",
+      "▁ra nk",
+      "▁ran k",
+      "▁ rank",
+      "over rightarrow",
+      "▁ap prec",
+      "▁app rec",
+      "▁appre c",
+      "以 为",
+      "▁s tri",
+      "▁st ri",
+      "▁str i",
+      "▁ stri",
+      "运 营",
+      "作出 的",
+      "作 出的",
+      "▁o rd",
+      "▁or d",
+      "▁ ord",
+      "▁in fo",
+      "▁inf o",
+      "▁ info",
+      "▁co ach",
+      "女 性",
+      "c p",
+      "欢 迎",
+      "推 广",
+      "▁t ra",
+      "▁tr a",
+      "▁ tra",
+      "▁e pis",
+      "▁ep is",
+      "事 情",
+      "来 源",
+      "▁p i",
+      "▁ pi",
+      "▁demonstr ated",
+      "▁demonstrate d",
+      "▁M ost",
+      "▁Mo st",
+      "▁Mos t",
+      "▁ Most",
+      "Lin k",
+      "Li nk",
+      "L ink",
+      "ol ved",
+      "olve d",
+      "olv ed",
+      "▁De sign",
+      "▁Des ign",
+      "▁ Design",
+      "▁m m",
+      "▁ mm",
+      "in ity",
+      "init y",
+      "ini ty",
+      "上海 市",
+      "上 海市",
+      "ar ts",
+      "art s",
+      "po l",
+      "p ol",
+      "幼 儿",
+      "▁g ave",
+      "▁ga ve",
+      "▁p ages",
+      "▁page s",
+      "▁pa ges",
+      "▁pag es",
+      "▁ pages",
+      "▁{ \\\\",
+      "▁{\\ \\",
+      "▁ {\\\\",
+      "ea l",
+      "e al",
+      "▁not ice",
+      "▁ notice",
+      "的 时间",
+      "▁small er",
+      "▁r ule",
+      "▁rul e",
+      "▁ru le",
+      "▁ rule",
+      "▁B ay",
+      "▁Ba y",
+      "▁ Bay",
+      "爱 情",
+      "S I",
+      "il la",
+      "ill a",
+      "i lla",
+      "▁fail ure",
+      "▁ failure",
+      "▁f loat",
+      "▁flo at",
+      "▁ float",
+      "H z",
+      "我 就",
+      "▁m g",
+      "▁ mg",
+      "▁sup ported",
+      "▁support ed",
+      "▁ supported",
+      "不 用",
+      "de p",
+      "d ep",
+      "▁s itu",
+      "▁sit u",
+      "▁si tu",
+      "▁cons ult",
+      "▁ consult",
+      "} /",
+      "▁b ur",
+      "▁bu r",
+      "▁ bur",
+      "fr ont",
+      "fro nt",
+      "} ^",
+      "▁Ar ch",
+      "▁Arc h",
+      "▁ Arch",
+      "ab ilities",
+      "abil ities",
+      "hem ical",
+      ">< /",
+      "> </",
+      "g g",
+      "第一百 五十",
+      "▁bl ue",
+      "▁blu e",
+      "▁ blue",
+      "十 六",
+      "et te",
+      "ett e",
+      "e tte",
+      "▁class Name",
+      "▁ className",
+      "m u",
+      "▁qu ery",
+      "▁que ry",
+      "▁quer y",
+      "▁ query",
+      "▁l ayer",
+      "▁la yer",
+      "▁lay er",
+      "▁ layer",
+      "毕 竟",
+      "Po s",
+      "P os",
+      "▁s even",
+      "▁se ven",
+      "▁ seven",
+      "▁g uid",
+      "▁gu id",
+      "▁gui d",
+      "▁ guid",
+      "娱 乐",
+      "内 的",
+      "Option s",
+      "Opt ions",
+      "O ptions",
+      "医 生",
+      "d x",
+      "▁ag reed",
+      "▁agre ed",
+      "▁agree d",
+      "▁agr eed",
+      "▁object s",
+      "▁obj ects",
+      "▁ objects",
+      "▁fr amework",
+      "▁frame work",
+      "▁ framework",
+      "▁sur round",
+      "多 种",
+      "很 好",
+      "el la",
+      "ell a",
+      "e lla",
+      "▁am azing",
+      "▁sur pr",
+      "▁Wh ite",
+      "▁Whit e",
+      "▁ White",
+      "不 足",
+      "po p",
+      "p op",
+      "▁d ecl",
+      "▁de cl",
+      "▁dec l",
+      "▁ decl",
+      "▁direct or",
+      "▁dir ector",
+      "▁dire ctor",
+      "▁ director",
+      "ig inal",
+      "igin al",
+      "igi nal",
+      "打 造",
+      "▁ref lect",
+      "▁ reflect",
+      "▁ex tr",
+      "▁ext r",
+      "▁ extr",
+      "▁up dated",
+      "▁update d",
+      "▁upd ated",
+      "▁ updated",
+      "c f",
+      "TE R",
+      "T ER",
+      "▁ 决",
+      "▁f loor",
+      "▁fl oor",
+      "▁flo or",
+      "▁ floor",
+      "疫 情",
+      "HO UT",
+      "HOU T",
+      "H OUT",
+      "L A",
+      "到 底",
+      "Type s",
+      "Typ es",
+      "Ty pes",
+      "T ypes",
+      "▁F OR",
+      "▁FO R",
+      "▁ FOR",
+      "人民 政府",
+      "颜 色",
+      "▁concent r",
+      "▁conce ntr",
+      "atur day",
+      "广 泛",
+      "on ym",
+      "ony m",
+      "cont ainer",
+      "▁cap ital",
+      "▁capita l",
+      "▁ capital",
+      "▁re turned",
+      "▁return ed",
+      "ip ped",
+      "ipp ed",
+      "i pped",
+      "▁inter action",
+      "▁interact ion",
+      "▁ interaction",
+      "日向本院 提出",
+      "日 向本院提出",
+      "jp g",
+      "j pg",
+      "党 员",
+      "▁U s",
+      "▁ Us",
+      "ca p",
+      "c ap",
+      "年 龄",
+      "一 切",
+      "imens ional",
+      "imension al",
+      "gr ade",
+      "grad e",
+      "gra de",
+      "C M",
+      "▁y ield",
+      "▁ yield",
+      "一个 人",
+      "一 个人",
+      "c y",
+      "sh e",
+      "s he",
+      "物 质",
+      "分 子",
+      "sc rib",
+      "scri b",
+      "scr ib",
+      "s crib",
+      "选择 题",
+      "eth yl",
+      "▁diagn osis",
+      "家 的",
+      "主 动",
+      "AS S",
+      "A SS",
+      "▁Sc ience",
+      "▁Sci ence",
+      "▁ Science",
+      "▁im m",
+      "▁i mm",
+      "▁ imm",
+      "▁adv anced",
+      "▁advance d",
+      "▁ advanced",
+      "nM y",
+      "n My",
+      "▁def inition",
+      "▁definit ion",
+      "▁defin ition",
+      "▁ definition",
+      "face s",
+      "fa ces",
+      "fac es",
+      "f aces",
+      "▁T ype",
+      "▁Ty pe",
+      "▁Typ e",
+      "▁ Type",
+      "▁e lection",
+      "▁elect ion",
+      "▁el ection",
+      "▁ele ction",
+      "▁ election",
+      "▁f inding",
+      "▁find ing",
+      "▁fin ding",
+      "▁ finding",
+      "il i",
+      "i li",
+      "▁invest igated",
+      "▁investig ated",
+      "▁investigate d",
+      "▁f air",
+      "▁fa ir",
+      "▁ fair",
+      "ct ors",
+      "ctor s",
+      "nF rom",
+      "nFr om",
+      "n From",
+      "▁f ailed",
+      "▁fail ed",
+      "▁fa iled",
+      "▁ failed",
+      "▁F acebook",
+      "▁Face book",
+      "▁ Facebook",
+      "Val id",
+      "V alid",
+      "▁t rend",
+      "▁tr end",
+      "▁tre nd",
+      "▁ trend",
+      "Pro vider",
+      "Prov ider",
+      "小 区",
+      "an cing",
+      "anc ing",
+      "▁dise ases",
+      "▁disease s",
+      "他们 的",
+      "他 们的",
+      "▁rad io",
+      "▁radi o",
+      "▁ radio",
+      "ful ly",
+      "full y",
+      "f ully",
+      "▁e ye",
+      "▁ey e",
+      "▁ eye",
+      "▁\" /",
+      "▁ \"/",
+      "▁m otor",
+      "▁mot or",
+      "▁mo tor",
+      "▁ motor",
+      "S ource",
+      "ri cs",
+      "ric s",
+      "r ics",
+      "概 念",
+      "▁evalu ation",
+      "▁eval uation",
+      "▁ evaluation",
+      "(\" %",
+      "( \"%",
+      "▁R ead",
+      "▁Re ad",
+      "▁ Read",
+      "per s",
+      "pe rs",
+      "p ers",
+      "aut o",
+      "au to",
+      "a uto",
+      "学 会",
+      "优 化",
+      "▁n ut",
+      "▁nu t",
+      "▁ nut",
+      "让 我",
+      "add ress",
+      "addr ess",
+      "Bo x",
+      "B ox",
+      "▁G l",
+      "▁ Gl",
+      "▁f lu",
+      "▁fl u",
+      "▁ flu",
+      "▁al one",
+      "▁ alone",
+      "▁g eneration",
+      "▁gener ation",
+      "▁gen eration",
+      "▁gene ration",
+      "▁genera tion",
+      "▁ generation",
+      "表 明",
+      "▁s pread",
+      "▁sp read",
+      "▁spr ead",
+      "▁ spread",
+      "包 装",
+      "re nder",
+      "ren der",
+      "rend er",
+      "r ender",
+      "▁m other",
+      "▁mot her",
+      "▁mo ther",
+      "▁moth er",
+      "▁ mother",
+      "号民事 判决书",
+      "号民事判决 书",
+      "号 民事判决书",
+      "▁p atterns",
+      "▁pat terns",
+      "▁pattern s",
+      "▁ patterns",
+      "▁e arn",
+      "▁ear n",
+      "▁ earn",
+      "▁re mov",
+      "▁rem ov",
+      "bs tract",
+      "b stract",
+      "as et",
+      "ase t",
+      "a set",
+      "▁ex posure",
+      "▁exp osure",
+      "▁ exposure",
+      "struct or",
+      "Po int",
+      "P oint",
+      "we rs",
+      "wer s",
+      "w ers",
+      "az on",
+      "azo n",
+      "a zon",
+      "▁B uild",
+      "▁Bu ild",
+      "▁ Build",
+      "r s",
+      "内 部",
+      "▁applic able",
+      "pos itory",
+      "p ository",
+      "▁C he",
+      "▁Ch e",
+      "▁ Che",
+      "▁record ed",
+      "▁ recorded",
+      "经 审理",
+      "pu b",
+      "p ub",
+      "▁return s",
+      "▁ returns",
+      "▁出 生",
+      "▁ 出生",
+      "fin al",
+      "fi nal",
+      "f inal",
+      "bor der",
+      "b order",
+      "▁or igin",
+      "▁orig in",
+      "▁ origin",
+      "ho w",
+      "h ow",
+      "▁m olec",
+      "▁mole c",
+      "▁mo lec",
+      "▁mol ec",
+      "if ications",
+      "ific ations",
+      "ification s",
+      "▁R ich",
+      "▁Ric h",
+      "▁Ri ch",
+      "▁ Rich",
+      "▁benef it",
+      "▁e mb",
+      "▁em b",
+      "▁ emb",
+      "G E",
+      "▁out comes",
+      "▁outcome s",
+      "an agement",
+      "anag ement",
+      "总 经理",
+      "ag ram",
+      "agr am",
+      "agra m",
+      "a gram",
+      "ai ght",
+      "a ight",
+      "研 发",
+      "ne nd",
+      "nen d",
+      "n end",
+      "▁M onday",
+      "▁Mon day",
+      "▁Mond ay",
+      "▁ Monday",
+      "▁W ashington",
+      "▁Wash ington",
+      "伤 害",
+      "ph i",
+      "p hi",
+      "\\' \\'",
+      "\\'\\ '",
+      "if orm",
+      "i form",
+      "re ek",
+      "ree k",
+      "r eek",
+      "总 结",
+      "▁th inking",
+      "▁think ing",
+      "▁thin king",
+      "▁ thinking",
+      "习 惯",
+      "▁ch ronic",
+      "▁chron ic",
+      "▁chr onic",
+      "d y",
+      "iate ly",
+      "iat ely",
+      "i ately",
+      "▁st uff",
+      "▁ stuff",
+      "▁Ph ys",
+      "▁ Phys",
+      "P U",
+      "▁ ;",
+      "▁m it",
+      "▁mi t",
+      "▁ mit",
+      "▁col lege",
+      "▁colle ge",
+      "▁w ild",
+      "▁wi ld",
+      "▁ wild",
+      "启 动",
+      "▁b ool",
+      "▁bo ol",
+      "▁ bool",
+      "点 的",
+      "据 此",
+      "rap per",
+      "r apper",
+      "De t",
+      "D et",
+      "▁G old",
+      "▁Go ld",
+      "▁Gol d",
+      "▁ Gold",
+      "sy stem",
+      "sys tem",
+      "s ystem",
+      "▁m etal",
+      "▁me tal",
+      "▁met al",
+      "▁meta l",
+      "▁ metal",
+      "▁T exas",
+      "▁Tex as",
+      "▁mark eting",
+      "▁market ing",
+      "▁ marketing",
+      "公开 开庭",
+      "执 字第",
+      "▁begin ning",
+      "▁enh ance",
+      "▁ch ain",
+      "▁cha in",
+      "▁ chain",
+      "▁prote ct",
+      "▁prot ect",
+      "ni n",
+      "n in",
+      "em ents",
+      "ement s",
+      "emen ts",
+      "e ments",
+      "ic ity",
+      "ici ty",
+      "icit y",
+      "i city",
+      "un icip",
+      "▁exper iments",
+      "▁experim ents",
+      "▁experi ments",
+      "▁experiment s",
+      "▁J ames",
+      "▁Jam es",
+      "▁Ja mes",
+      "▁ James",
+      "题 意",
+      "▁c ategory",
+      "▁categor y",
+      "▁ category",
+      "▁in ject",
+      "▁inj ect",
+      "▁ inject",
+      "pl icit",
+      "plic it",
+      "▁B est",
+      "▁Be st",
+      "▁Bes t",
+      "▁ Best",
+      "▁p rices",
+      "▁pr ices",
+      "▁price s",
+      "▁pri ces",
+      "* \\",
+      "nO ur",
+      "n Our",
+      "▁R ad",
+      "▁Ra d",
+      "▁ Rad",
+      "▁char acters",
+      "▁character s",
+      "▁ characters",
+      "加 上",
+      "▁int eger",
+      "▁ integer",
+      "▁f ather",
+      "▁fa ther",
+      "▁fat her",
+      "▁ father",
+      "▁con stit",
+      "▁cons tit",
+      "▁const it",
+      "si te",
+      "sit e",
+      "s ite",
+      "▁appro aches",
+      "▁approach es",
+      "an ging",
+      "ang ing",
+      "angi ng",
+      "▁im plied",
+      "▁impl ied",
+      "IN D",
+      "I ND",
+      "as is",
+      "asi s",
+      "a sis",
+      "▁f avorite",
+      "▁favor ite",
+      "ave d",
+      "av ed",
+      "a ved",
+      "▁R oad",
+      "▁Ro ad",
+      "▁ Road",
+      "▁P ython",
+      "▁Py thon",
+      "▁ Python",
+      "all el",
+      "alle l",
+      "这 次",
+      "In stance",
+      "Inst ance",
+      "资 本",
+      "▁m ath",
+      "▁mat h",
+      "▁ma th",
+      "▁ math",
+      "▁Ch inese",
+      "▁Chin ese",
+      "▁ Chinese",
+      "ul in",
+      "uli n",
+      "u lin",
+      "nA merican",
+      "n American",
+      "如果 你",
+      "▁p urchase",
+      "▁purch ase",
+      "▁ purchase",
+      "▁P arty",
+      "▁Par ty",
+      "▁Part y",
+      "▁ Party",
+      "\\\\ ((",
+      "\\\\( (",
+      "ge nt",
+      "gen t",
+      "g ent",
+      "▁b ottom",
+      "▁bott om",
+      "▁bot tom",
+      "▁ bottom",
+      "▁f rame",
+      "▁fr ame",
+      "▁fra me",
+      "▁ frame",
+      "▁re nder",
+      "▁r ender",
+      "▁rend er",
+      "▁ render",
+      "有 什么",
+      "▁c ourt",
+      "▁co urt",
+      "▁cour t",
+      "▁cou rt",
+      "▁ court",
+      "▁b acter",
+      "▁bac ter",
+      "▁G ames",
+      "▁Game s",
+      "▁Ga mes",
+      "▁Gam es",
+      "▁ Games",
+      "I A",
+      "H P",
+      "▁S ol",
+      "▁So l",
+      "▁ Sol",
+      "co h",
+      "c oh",
+      "▁ ~",
+      "相 当",
+      "▁tr ade",
+      "▁trad e",
+      "▁tra de",
+      "▁ trade",
+      "英 国",
+      "▁follow s",
+      "sess ion",
+      "s ession",
+      "读 书",
+      "▁altern ative",
+      "▁alter native",
+      "ues day",
+      "▁c ast",
+      "▁ca st",
+      "▁cas t",
+      "▁ cast",
+      "[ :",
+      "pe t",
+      "p et",
+      "▁H IV",
+      "▁HI V",
+      "▁ HIV",
+      "ve c",
+      "v ec",
+      "ge st",
+      "ges t",
+      "g est",
+      "▁F ile",
+      "▁Fil e",
+      "▁Fi le",
+      "▁ File",
+      "▁separ ate",
+      "▁M em",
+      "▁Me m",
+      "▁ Mem",
+      "▁beh avi",
+      "▁behav i",
+      "▁ph oto",
+      "▁phot o",
+      "▁ photo",
+      "ind uced",
+      "lic enses",
+      "license s",
+      "的 方法",
+      "ob ert",
+      "ober t",
+      "obe rt",
+      "o bert",
+      "▁( (",
+      "▁ ((",
+      "pl ete",
+      "ple te",
+      "plet e",
+      "p lete",
+      "DI R",
+      "D IR",
+      "▁v el",
+      "▁ve l",
+      "▁ vel",
+      "an ned",
+      "ann ed",
+      "anne d",
+      "ent ity",
+      "enti ty",
+      "日 常",
+      "▁pub lish",
+      "▁ publish",
+      "高 级",
+      "山 市",
+      "所有 的",
+      "所 有的",
+      "▁d ro",
+      "▁dr o",
+      "▁ dro",
+      "▁L in",
+      "▁Li n",
+      "▁ Lin",
+      "th let",
+      "最大 的",
+      "最 大的",
+      "▁op ened",
+      "▁open ed",
+      "▁ opened",
+      "▁N ews",
+      "▁New s",
+      "▁Ne ws",
+      "▁ News",
+      "人 生",
+      "in der",
+      "ind er",
+      "inde r",
+      "i nder",
+      "B S",
+      "▁compl iance",
+      "▁ compliance",
+      "▁F rance",
+      "▁Fr ance",
+      "▁Franc e",
+      "▁Fra nce",
+      "▁ France",
+      "▁ch rom",
+      "▁chr om",
+      "▁ chrom",
+      "▁ch allenge",
+      "▁chall enge",
+      "▁challeng e",
+      "▁ challenge",
+      "▁profess ion",
+      "▁prof ession",
+      "tf or",
+      "t for",
+      "▁min imum",
+      "▁minim um",
+      "▁ minimum",
+      "af ter",
+      "a fter",
+      "ri end",
+      "rie nd",
+      "rien d",
+      "▁B ase",
+      "▁Bas e",
+      "▁Ba se",
+      "▁ Base",
+      "在 线",
+      "尺 寸",
+      "cl ick",
+      "cli ck",
+      "c lick",
+      "r u",
+      "EMA IL",
+      "E MAIL",
+      "▁f ab",
+      "▁fa b",
+      "▁ fab",
+      "▁M ethod",
+      "▁Meth od",
+      "▁ Method",
+      "▁p rop",
+      "▁pro p",
+      "▁pr op",
+      "▁ prop",
+      "▁h uge",
+      "▁hug e",
+      "▁hu ge",
+      "▁up per",
+      "▁ upper",
+      "nf unction",
+      "nfunc tion",
+      "nfun ction",
+      "n function",
+      "oot ball",
+      "▁ex amples",
+      "▁example s",
+      "▁exam ples",
+      "▁ examples",
+      "▁prob ability",
+      "▁ probability",
+      "▁m ovie",
+      "▁mov ie",
+      "▁mo vie",
+      "▁ movie",
+      "两 种",
+      "刑 期",
+      "▁b one",
+      "▁bo ne",
+      "▁bon e",
+      "▁ bone",
+      "▁ex ists",
+      "▁exist s",
+      "▁ exists",
+      "整 数",
+      "▁re pair",
+      "▁rep air",
+      "or a",
+      "o ra",
+      "an ner",
+      "ann er",
+      "anne r",
+      "至 少",
+      "▁s ch",
+      "▁sc h",
+      "▁ sch",
+      "▁gu ide",
+      "▁guid e",
+      "▁gui de",
+      "▁ guide",
+      "二 次",
+      "▁in str",
+      "▁i nstr",
+      "▁inst r",
+      "▁ins tr",
+      "的地 方",
+      "的 地方",
+      "tu re",
+      "tur e",
+      "t ure",
+      "or ation",
+      "ora tion",
+      "o ration",
+      "com ponent",
+      "角 度",
+      "▁im ag",
+      "▁i mag",
+      "▁ imag",
+      "sa ve",
+      "sav e",
+      "s ave",
+      "▁sm ooth",
+      "▁smo oth",
+      "▁ smooth",
+      "▁B re",
+      "▁Br e",
+      "▁ Bre",
+      "et her",
+      "eth er",
+      "ethe r",
+      "e ther",
+      "ca re",
+      "car e",
+      "c are",
+      "▁s ale",
+      "▁sa le",
+      "▁sal e",
+      "▁ sale",
+      "能 源",
+      "con tin",
+      "cont in",
+      "hem atics",
+      "hemat ics",
+      "hematic s",
+      "h ematics",
+      "给 你",
+      "▁pre par",
+      "▁prep ar",
+      "OR T",
+      "O RT",
+      "▁c oding",
+      "▁co ding",
+      "▁cod ing",
+      "▁ coding",
+      "uz z",
+      "u zz",
+      "▁equ ival",
+      "am eters",
+      "ame ters",
+      "amet ers",
+      "ameter s",
+      "下 列",
+      "▁H ot",
+      "▁Ho t",
+      "▁ Hot",
+      "▁l ymph",
+      "▁ly mph",
+      "▁ lymph",
+      "足 球",
+      "ha n",
+      "h an",
+      "oc ity",
+      "oci ty",
+      "o city",
+      "▁a f",
+      "▁ af",
+      "▁sc en",
+      "er vation",
+      "erv ation",
+      "▁s ust",
+      "▁su st",
+      "▁sus t",
+      "se y",
+      "s ey",
+      "▁admin istration",
+      "▁administr ation",
+      "▁ administration",
+      "▁tr aff",
+      "▁tra ff",
+      "▁ traff",
+      "▁g lass",
+      "▁gl ass",
+      "▁ glass",
+      "spec ific",
+      "s pecific",
+      "Co re",
+      "Cor e",
+      "C ore",
+      "ar p",
+      "a rp",
+      "是不 是",
+      "是 不是",
+      "探 讨",
+      "落 实",
+      "▁L ou",
+      "▁Lo u",
+      "困 难",
+      "▁m ill",
+      "▁mil l",
+      "▁mi ll",
+      "▁ mill",
+      "ti m",
+      "t im",
+      "广 西",
+      "▁ann ual",
+      "▁ annual",
+      "▁s oft",
+      "▁so ft",
+      "▁ soft",
+      "▁S O",
+      "▁ SO",
+      "▁rest aur",
+      "as ure",
+      "asu re",
+      "lo cation",
+      "loc ation",
+      "l ocation",
+      "中 学",
+      "est ival",
+      "AT ION",
+      "▁ang le",
+      "▁ angle",
+      "▁M art",
+      "▁Mar t",
+      "▁Ma rt",
+      "▁ Mart",
+      "go ing",
+      "▁re placement",
+      "▁replac ement",
+      "▁replace ment",
+      "损 失",
+      "▁Ch urch",
+      "n x",
+      "E m",
+      "▁eas ier",
+      "▁sur vey",
+      "▁surv ey",
+      "▁surve y",
+      "▁ survey",
+      "中 医",
+      "▁p rove",
+      "▁pro ve",
+      "▁pr ove",
+      "▁prov e",
+      "▁ prove",
+      "▁Comp any",
+      "▁ Company",
+      "Ca se",
+      "Cas e",
+      "C ase",
+      "▁P en",
+      "▁Pe n",
+      "▁ Pen",
+      "刑 法",
+      "判决 如下",
+      "▁A frica",
+      "▁Afr ica",
+      "/* \\",
+      "/ *\\",
+      "▁str aight",
+      "urs day",
+      "ir med",
+      "irm ed",
+      "ic ations",
+      "ication s",
+      "▁occ urre",
+      "▁occur re",
+      "ic ed",
+      "ice d",
+      "i ced",
+      "▁t rial",
+      "▁tr ial",
+      "▁tri al",
+      "▁ trial",
+      "的事 情",
+      "的 事情",
+      "金 属",
+      "发现 被执行人",
+      "一 条",
+      "▁d ry",
+      "▁dr y",
+      "▁ dry",
+      "▁sympt oms",
+      "▁symptom s",
+      "▁ac adem",
+      "▁S al",
+      "▁Sa l",
+      "▁ Sal",
+      "▁d oor",
+      "▁do or",
+      "▁ door",
+      "▁t emplate",
+      "▁tem plate",
+      "▁temp late",
+      "▁ template",
+      "然 而",
+      "▁IS BN",
+      "▁ ISBN",
+      "▁fin ished",
+      "▁finish ed",
+      "▁ finished",
+      "Nu ll",
+      "N ull",
+      "va s",
+      "v as",
+      "L C",
+      "ex ample",
+      "exam ple",
+      "M ethod",
+      "▁pl ays",
+      "▁play s",
+      "▁ plays",
+      "一 日",
+      "▁elect ron",
+      "▁electro n",
+      "▁electr on",
+      "▁ electron",
+      "▁s aying",
+      "▁sa ying",
+      "▁say ing",
+      "▁j oint",
+      "▁join t",
+      "▁jo int",
+      "▁ joint",
+      "▁m is",
+      "▁mi s",
+      "▁ mis",
+      "in y",
+      "i ny",
+      "带 来",
+      "增 强",
+      "▁th ous",
+      "▁r are",
+      "▁ra re",
+      "▁C a",
+      "▁ Ca",
+      "▁em ployees",
+      "▁employ ees",
+      "▁employee s",
+      "公诉 机关",
+      "ap ers",
+      "aper s",
+      "ape rs",
+      "a pers",
+      "▁h ands",
+      "▁hand s",
+      "▁han ds",
+      "▁T hus",
+      "▁Th us",
+      "▁Thu s",
+      "ke ys",
+      "key s",
+      "▁conn ected",
+      "▁connect ed",
+      "▁ connected",
+      "▁N etwork",
+      "▁Net work",
+      "▁ Network",
+      "$ (",
+      "it ivity",
+      "▁S unday",
+      "▁Sund ay",
+      "▁Sun day",
+      "▁ Sunday",
+      "粉 丝",
+      "ca t",
+      "c at",
+      "▁gener ally",
+      "▁general ly",
+      "都 在",
+      "状 况",
+      "▁pro of",
+      "▁ proof",
+      "D R",
+      "▁eff iciency",
+      "▁ efficiency",
+      "▁ind icate",
+      "▁indic ate",
+      "j a",
+      "评 论",
+      "第一百五十 四",
+      "▁M ult",
+      "▁Mu lt",
+      "▁Mul t",
+      "▁ Mult",
+      "▁n an",
+      "▁na n",
+      "▁ nan",
+      "com ponents",
+      "component s",
+      "▁contr ast",
+      "担 保",
+      "met a",
+      "me ta",
+      "m eta",
+      "▁{ }\\",
+      "▁{} \\",
+      "▁ {}\\",
+      "于 是",
+      "▁de but",
+      "▁deb ut",
+      "还 要",
+      "▁b ath",
+      "▁bat h",
+      "▁ba th",
+      "▁ bath",
+      "um ns",
+      "umn s",
+      "年 度",
+      "si gn",
+      "sig n",
+      "s ign",
+      "▁w ave",
+      "▁wa ve",
+      "▁ wave",
+      "cont rol",
+      "contr ol",
+      "▁S ummer",
+      "▁Summ er",
+      "▁Sum mer",
+      "▁ Summer",
+      "▁pl aced",
+      "▁place d",
+      "▁plac ed",
+      "▁ placed",
+      "▁w ish",
+      "▁wis h",
+      "▁wi sh",
+      "ash ion",
+      "ashi on",
+      "VI D",
+      "V ID",
+      "▁s mart",
+      "▁sm art",
+      "▁ smart",
+      "▁d ie",
+      "▁di e",
+      "▁ die",
+      "set tings",
+      "setting s",
+      "sett ings",
+      "nes day",
+      "af e",
+      "a fe",
+      "▁M ex",
+      "▁Me x",
+      "▁d iet",
+      "▁di et",
+      "▁die t",
+      "it ar",
+      "ita r",
+      "i tar",
+      "学 术",
+      "展 示",
+      "▁sh ared",
+      "▁share d",
+      "▁sha red",
+      "▁shar ed",
+      "▁ shared",
+      "法 规",
+      "▁import ance",
+      "▁ass ociation",
+      "▁associ ation",
+      "ten ance",
+      "▁mus cle",
+      "▁musc le",
+      "os es",
+      "ose s",
+      "o ses",
+      "� �",
+      "▁s pent",
+      "▁sp ent",
+      "▁spe nt",
+      "品 质",
+      "ki n",
+      "k in",
+      "准许 原告",
+      "▁ad ult",
+      "▁adul t",
+      "ga me",
+      "gam e",
+      "g ame",
+      "▁s witch",
+      "▁sw itch",
+      "▁swit ch",
+      "▁ switch",
+      "由 被告",
+      "p c",
+      "▁g rant",
+      "▁gr ant",
+      "▁gran t",
+      "▁gra nt",
+      "▁ grant",
+      "送达 之日起",
+      "a a",
+      "ist ant",
+      "istan t",
+      "ista nt",
+      "i stant",
+      "▁C entral",
+      "▁Cent ral",
+      "▁ Central",
+      "就 能",
+      "▁sub sequ",
+      "▁subs equ",
+      "S H",
+      "出 口",
+      "▁liter ature",
+      "▁ch urch",
+      "▁ church",
+      "k m",
+      "▁opportun ities",
+      "公 布",
+      "▁environment al",
+      "▁i deal",
+      "▁ide al",
+      "▁id eal",
+      "▁idea l",
+      "论 文",
+      "当 地",
+      "▁f ight",
+      "▁fig ht",
+      "▁fi ght",
+      "▁B er",
+      "▁Be r",
+      "▁ Ber",
+      ":\\ \\",
+      ": \\\\",
+      "ã o",
+      ")^ {",
+      ") ^{",
+      "▁d ead",
+      "▁de ad",
+      "▁ dead",
+      "▁j oined",
+      "▁join ed",
+      "▁jo ined",
+      "▁ joined",
+      "环 保",
+      "诉讼 代理人",
+      "ain ts",
+      "aint s",
+      "a ints",
+      "▁ex ception",
+      "▁except ion",
+      "▁ exception",
+      "▁config uration",
+      "▁ configuration",
+      "▁M iss",
+      "▁Mi ss",
+      "▁Mis s",
+      "▁ Miss",
+      "▁b ul",
+      "▁bu l",
+      "▁ bul",
+      "快 乐",
+      "pl ing",
+      "p ling",
+      "▁h undred",
+      "一 天",
+      "举 办",
+      "诉 被告",
+      "▁W ITHOUT",
+      "▁WIT HOUT",
+      "▁WITH OUT",
+      "▁ WITHOUT",
+      "▁exper ienced",
+      "▁experience d",
+      "▁experi enced",
+      "▁Qu estion",
+      "▁Quest ion",
+      "▁ Question",
+      "▁I T",
+      "▁ IT",
+      "网 上",
+      "▁evalu ated",
+      "▁evaluate d",
+      "▁eval uated",
+      "nT hat",
+      "nTh at",
+      "n That",
+      "nH ere",
+      "nHe re",
+      "nHer e",
+      "n Here",
+      "▁result ing",
+      "▁pass ed",
+      "▁pas sed",
+      "▁ passed",
+      "▁may be",
+      "▁ maybe",
+      "▁Ap ache",
+      "▁ Apache",
+      "He l",
+      "H el",
+      "▁Y ear",
+      "▁Ye ar",
+      "▁ Year",
+      "很多 人",
+      "很 多人",
+      "本院在 审理",
+      "tm l",
+      "t ml",
+      "▁ar riv",
+      "▁arr iv",
+      "确 保",
+      "en cing",
+      "enc ing",
+      "nR e",
+      "n Re",
+      "L icense",
+      "▁E lect",
+      "▁El ect",
+      "▁Ele ct",
+      "▁ Elect",
+      "▁im mediately",
+      "▁immed iately",
+      "▁immediate ly",
+      "bo ur",
+      "b our",
+      "la ng",
+      "lan g",
+      "l ang",
+      "▁th read",
+      "▁thr ead",
+      "▁ thread",
+      "▁C ong",
+      "▁Con g",
+      "▁Co ng",
+      "▁ Cong",
+      "▁p ush",
+      "▁pu sh",
+      "▁ push",
+      "▁O ther",
+      "▁Ot her",
+      "▁ Other",
+      "▁E ach",
+      "▁ Each",
+      "▁f ill",
+      "▁fil l",
+      "▁fi ll",
+      "▁ fill",
+      "in ger",
+      "ing er",
+      "inge r",
+      "▁int roduced",
+      "▁introdu ced",
+      "▁introduce d",
+      "F A",
+      "可供执行 的财产",
+      "▁L eg",
+      "▁Le g",
+      "▁ Leg",
+      "hold er",
+      "hol der",
+      "h older",
+      "P M",
+      "思 考",
+      "办公 室",
+      "▁act ions",
+      "▁action s",
+      "▁ actions",
+      "▁br east",
+      "▁bre ast",
+      "ar est",
+      "are st",
+      "ares t",
+      "a rest",
+      "▁J un",
+      "▁Ju n",
+      "▁ Jun",
+      "at abase",
+      "ata base",
+      "Ap i",
+      "A pi",
+      "` ,",
+      "▁de scribe",
+      "▁des cribe",
+      "▁describ e",
+      "▁desc ribe",
+      "▁ describe",
+      "Se c",
+      "S ec",
+      "▁reg ions",
+      "▁region s",
+      "▁ regions",
+      "▁s igned",
+      "▁sign ed",
+      "▁sig ned",
+      "▁ signed",
+      "S O",
+      "▁dr ugs",
+      "▁drug s",
+      "▁d rop",
+      "▁dr op",
+      "▁dro p",
+      "▁ drop",
+      "ck s",
+      "c ks",
+      "reg ister",
+      "▁S aturday",
+      "▁ Saturday",
+      "▁pro posal",
+      "▁propos al",
+      "▁prop osal",
+      "De c",
+      "D ec",
+      "at o",
+      "a to",
+      "党 委",
+      "ua n",
+      "u an",
+      "▁gu ess",
+      "▁ guess",
+      "是 指",
+      "ha b",
+      "h ab",
+      "▁ex tension",
+      "▁ext ension",
+      "▁ extension",
+      "大 会",
+      "抚 养",
+      "说 的",
+      "同 意",
+      "▁comp etition",
+      "▁compet ition",
+      "▁competit ion",
+      "▁F ore",
+      "▁For e",
+      "▁Fo re",
+      "▁ Fore",
+      "▁s impl",
+      "▁sim pl",
+      "到庭 参加诉讼",
+      "到庭参加 诉讼",
+      "Re pository",
+      "Rep ository",
+      "Repos itory",
+      "ial ize",
+      "▁n orm",
+      "▁no rm",
+      "▁nor m",
+      "▁ norm",
+      "os en",
+      "ose n",
+      "o sen",
+      "▁j our",
+      "▁jo ur",
+      "▁ jour",
+      "▁出 版",
+      "▁ 出版",
+      "兴 趣",
+      "出 具",
+      "▁b ug",
+      "▁bu g",
+      "▁ bug",
+      "▁the rm",
+      "▁th erm",
+      "▁ther m",
+      "ens ions",
+      "ension s",
+      "du cation",
+      "duc ation",
+      "▁F lor",
+      "▁Fl or",
+      "▁Flo r",
+      "▁Eng ine",
+      "▁ Engine",
+      "▁s ongs",
+      "▁song s",
+      "▁son gs",
+      "式 的",
+      "im ary",
+      "ima ry",
+      "imar y",
+      "▁J ack",
+      "▁Jac k",
+      "▁Ja ck",
+      "▁ Jack",
+      "as te",
+      "ast e",
+      "a ste",
+      "▁h ar",
+      "▁ha r",
+      "▁ har",
+      "▁w ife",
+      "▁wi fe",
+      "▁ wife",
+      "动 力",
+      "▁c os",
+      "▁co s",
+      "▁ cos",
+      "利 益",
+      "AP I",
+      "A PI",
+      "突 然",
+      "街 道",
+      "▁p le",
+      "▁pl e",
+      "▁ ple",
+      "▁v ac",
+      "▁va c",
+      "▁ vac",
+      "Re f",
+      "R ef",
+      "美 元",
+      "▁r ob",
+      "▁ro b",
+      "▁ rob",
+      "▁set up",
+      "▁ setup",
+      "Def ault",
+      "De fault",
+      "▁w id",
+      "▁wi d",
+      "▁ wid",
+      "委托 诉讼代理人",
+      "▁V is",
+      "▁Vi s",
+      "▁ Vis",
+      "c r",
+      "悔 改",
+      "▁s ett",
+      "▁se tt",
+      "▁set t",
+      "▁ sett",
+      "if fer",
+      "iff er",
+      "iffe r",
+      "成 员",
+      "▁prot ocol",
+      "▁proto col",
+      "▁ protocol",
+      "▁eff orts",
+      "▁effort s",
+      "ail way",
+      "▁det ect",
+      "▁ detect",
+      "某 甲",
+      "Re ad",
+      "R ead",
+      "放 在",
+      "▁n early",
+      "▁ne arly",
+      "▁near ly",
+      "▁d ynamic",
+      "▁dynam ic",
+      "▁ dynamic",
+      "▁part ner",
+      "▁ partner",
+      "▁im aging",
+      "▁imag ing",
+      "▁ imaging",
+      "合 适",
+      "part ial",
+      "▁pro pose",
+      "▁propos e",
+      "▁prop ose",
+      "▁L ong",
+      "▁Lo ng",
+      "▁Lon g",
+      "▁ Long",
+      "不 等",
+      "ha sh",
+      "has h",
+      "h ash",
+      "到 的",
+      "▁exper iment",
+      "▁experim ent",
+      "▁experi ment",
+      "▁ experiment",
+      "大 部分",
+      "执行 机关",
+      "yo nd",
+      "yon d",
+      "y ond",
+      "it a",
+      "i ta",
+      "en sus",
+      "ens us",
+      "Se lect",
+      "S elect",
+      "升 级",
+      "▁ 开",
+      "食 物",
+      "un s",
+      "u ns",
+      "来 看",
+      "▁d b",
+      "▁ db",
+      "own load",
+      "▁f inally",
+      "▁fin ally",
+      "▁final ly",
+      "▁ finally",
+      "▁m ouse",
+      "▁mo use",
+      "▁ mouse",
+      "▁\" ^",
+      "▁ \"^",
+      "▁j ournal",
+      "▁jour nal",
+      "▁jo urnal",
+      "▁ journal",
+      "apeut ic",
+      "ex pected",
+      "expect ed",
+      "exp ected",
+      "▁C reate",
+      "▁Creat e",
+      "▁Cre ate",
+      "▁ Create",
+      "常 见",
+      "lo gin",
+      "log in",
+      "添 加",
+      "lo ok",
+      "loo k",
+      "l ook",
+      "▁ass essment",
+      "▁assess ment",
+      "▁ assessment",
+      "▁d irected",
+      "▁direct ed",
+      "▁dir ected",
+      "▁ directed",
+      "sc ale",
+      "scal e",
+      "s cale",
+      "董事 长",
+      "▁pro cedure",
+      "▁proced ure",
+      "G A",
+      "绿 色",
+      "死 亡",
+      "co r",
+      "c or",
+      "重 大",
+      "▁pass word",
+      "▁ password",
+      "kw args",
+      "ul y",
+      "u ly",
+      "field s",
+      "f ields",
+      "买 卖",
+      "▁ed ge",
+      "▁ edge",
+      "▁r uns",
+      "▁run s",
+      "▁ru ns",
+      "▁ runs",
+      "an ish",
+      "ani sh",
+      "▁B i",
+      "▁ Bi",
+      "▁dist inct",
+      "an eous",
+      "ane ous",
+      "Ph one",
+      "▁ex ercise",
+      "▁exerc ise",
+      "▁ exercise",
+      "stand ing",
+      "stan ding",
+      "递 交",
+      "▁particip ants",
+      "▁participant s",
+      "表 面",
+      "▁inf lamm",
+      "▁infl amm",
+      "▁ inflamm",
+      "科技 有限公司",
+      "本院 予以",
+      "▁mov ement",
+      "▁move ment",
+      "▁ 陈",
+      "▁G ermany",
+      "▁Germ any",
+      "▁German y",
+      "▁Ger many",
+      "▁ Germany",
+      "▁disc ussed",
+      "▁discuss ed",
+      "▁V iew",
+      "▁Vi ew",
+      "▁Vie w",
+      "▁ View",
+      "ol icy",
+      "olic y",
+      "oli cy",
+      "第一 次",
+      "第 一次",
+      "ec es",
+      "e ces",
+      "an nels",
+      "ann els",
+      "annel s",
+      "anne ls",
+      "▁anal ys",
+      "▁analy s",
+      "▁ana lys",
+      "▁v irus",
+      "▁vir us",
+      "▁vi rus",
+      "▁ virus",
+      "▁Ac adem",
+      "▁Acad em",
+      "ang ular",
+      "angu lar",
+      "▁own ers",
+      "▁ow ners",
+      "▁owner s",
+      "▁ owners",
+      "▁A frican",
+      "▁Afr ican",
+      "▁Africa n",
+      "公安 局",
+      "li f",
+      "l if",
+      "▁s light",
+      "▁sl ight",
+      "通 常",
+      "集 中",
+      "▁p ool",
+      "▁po ol",
+      "▁ pool",
+      "▁a ims",
+      "▁aim s",
+      "▁ai ms",
+      "▁d ou",
+      "▁do u",
+      "▁ dou",
+      "如不服 本判决",
+      "▁S ociety",
+      "▁Soci ety",
+      "▁expl ore",
+      "▁explo re",
+      "▁explor e",
+      "广 州",
+      "玩 家",
+      "sq l",
+      "s ql",
+      "jo b",
+      "j ob",
+      "▁my self",
+      "▁mys elf",
+      "ea n",
+      "e an",
+      "te d",
+      "t ed",
+      "▁med ium",
+      "▁medi um",
+      "▁ medium",
+      "it ative",
+      "itat ive",
+      "▁ag ents",
+      "▁agent s",
+      "▁ agents",
+      "li g",
+      "l ig",
+      "▁C ourt",
+      "▁Co urt",
+      "▁Cour t",
+      "▁Cou rt",
+      "we st",
+      "w est",
+      "▁p riv",
+      "▁pr iv",
+      "▁pri v",
+      "▁ priv",
+      "▁I d",
+      "▁ Id",
+      "▁m ilitary",
+      "▁mil itary",
+      "▁milit ary",
+      "em ic",
+      "emi c",
+      "e mic",
+      "pe ed",
+      "p eed",
+      "的 基础",
+      "▁A ustralian",
+      "▁Austral ian",
+      "▁Australia n",
+      "▁ Australian",
+      "苹 果",
+      "对 照",
+      "P s",
+      "是 个",
+      "me n",
+      "m en",
+      "nS ub",
+      "n Sub",
+      "▁E v",
+      "▁ Ev",
+      "▁h ear",
+      "▁he ar",
+      "▁A ward",
+      "▁Aw ard",
+      "▁t it",
+      "▁ti t",
+      "▁ tit",
+      "▁surv ival",
+      "▁surviv al",
+      "▁N O",
+      "▁ NO",
+      "所 谓",
+      "so rt",
+      "s ort",
+      "▁D uring",
+      "▁Du ring",
+      "▁Dur ing",
+      "▁ During",
+      "▁m anager",
+      "▁man ager",
+      "▁manag er",
+      "▁manage r",
+      "▁ manager",
+      "▁col lected",
+      "▁coll ected",
+      "▁collect ed",
+      "▁c ycle",
+      "▁cy cle",
+      "▁cycl e",
+      "▁ cycle",
+      "ul f",
+      "u lf",
+      "有 了",
+      "的主 要",
+      "的 主要",
+      "▁th reat",
+      "▁thr eat",
+      "▁ threat",
+      "职 工",
+      "赔 偿",
+      "▁P ower",
+      "▁Po wer",
+      "▁Pow er",
+      "▁ Power",
+      "基 地",
+      "as ic",
+      "asi c",
+      "St y",
+      "S ty",
+      "简 称",
+      "▁S uper",
+      "▁Sup er",
+      "▁Su per",
+      "▁ Super",
+      "L I",
+      "gr eg",
+      "gre g",
+      "g reg",
+      "福建 省",
+      "▁document ation",
+      "▁ documentation",
+      "指 标",
+      "科 研",
+      "▁inter actions",
+      "▁interaction s",
+      "▁interact ions",
+      "pr ession",
+      "press ion",
+      "p ression",
+      "▁M ount",
+      "▁Mo unt",
+      "▁ Mount",
+      "强制 执行",
+      "▁commun ities",
+      "曾 经",
+      "re al",
+      "rea l",
+      "r eal",
+      "▁g ain",
+      "▁ga in",
+      "▁ gain",
+      "▁ar ter",
+      "▁art er",
+      "▁ arter",
+      "心 的",
+      "▁ind uced",
+      "▁indu ced",
+      "▁induce d",
+      "▁ induced",
+      "▁s pir",
+      "▁sp ir",
+      "▁ spir",
+      "On e",
+      "O ne",
+      "发 出",
+      "确 有",
+      "书 面",
+      "▁incre ases",
+      "▁increase s",
+      "形 象",
+      "诉讼 请求",
+      "Ch eck",
+      "Che ck",
+      "C heck",
+      "▁a uthors",
+      "▁author s",
+      "▁auth ors",
+      "▁ authors",
+      "条第一款 第",
+      "ber t",
+      "be rt",
+      "b ert",
+      "as y",
+      "a sy",
+      "▁Dem ocr",
+      "▁Demo cr",
+      "▁T op",
+      "▁To p",
+      "▁ Top",
+      "▁vit ro",
+      "▁vi tro",
+      "家 长",
+      "re move",
+      "rem ove",
+      "*/ \\",
+      "* /\\",
+      "感 染",
+      "▁prote ction",
+      "▁prot ection",
+      "▁protect ion",
+      "▁ protection",
+      "▁s paces",
+      "▁sp aces",
+      "▁space s",
+      "▁spac es",
+      "▁spa ces",
+      "▁ spaces",
+      "市 人民法院",
+      "T A",
+      "症 状",
+      "▁sh owing",
+      "▁show ing",
+      "▁f arm",
+      "▁far m",
+      "▁fa rm",
+      "▁ farm",
+      "or ry",
+      "orr y",
+      "每 次",
+      "mat ch",
+      "m atch",
+      "▁M ain",
+      "▁Ma in",
+      "▁Mai n",
+      "▁ Main",
+      "也就 是",
+      "也 就是",
+      "yt ic",
+      "y tic",
+      "▁pl asma",
+      "▁plas ma",
+      "▁plasm a",
+      "() ;",
+      "( );",
+      "sy nc",
+      "syn c",
+      "s ync",
+      "▁car ried",
+      "▁l ung",
+      "▁lu ng",
+      "有一 个",
+      "有 一个",
+      "大 概",
+      "是 不",
+      "严 格",
+      "空 气",
+      "▁s hoot",
+      "▁sh oot",
+      "▁contin uous",
+      "▁continu ous",
+      "▁ continuous",
+      "\\' ]\\",
+      "\\'] \\",
+      "\\ ']\\",
+      "▁b eyond",
+      "▁be yond",
+      "的 女",
+      "▁In ternet",
+      "▁Intern et",
+      "▁Inter net",
+      "▁ Internet",
+      "▁There fore",
+      "▁Ther efore",
+      "▁t ox",
+      "▁to x",
+      "▁ tox",
+      "就 业",
+      "博 士",
+      "▁exam ined",
+      "▁examine d",
+      "ope rative",
+      "oper ative",
+      "者 的",
+      "局 长",
+      "U D",
+      "▁M ary",
+      "▁Mar y",
+      "▁Ma ry",
+      "▁ Mary",
+      "x i",
+      "tr ack",
+      "tra ck",
+      "t rack",
+      "▁T own",
+      "▁To wn",
+      "▁Tow n",
+      "all ery",
+      "alle ry",
+      "d k",
+      "▁rec ords",
+      "▁record s",
+      "▁ records",
+      "▁v ideos",
+      "▁video s",
+      "▁vide os",
+      "▁ videos",
+      "蒙 古",
+      "▁an imal",
+      "▁anim al",
+      "▁ animal",
+      "bo ol",
+      "b ool",
+      "os ter",
+      "ost er",
+      "o ster",
+      "se con",
+      "sec on",
+      "s econ",
+      "最 好",
+      "el if",
+      "eli f",
+      "e lif",
+      "▁s eed",
+      "▁se ed",
+      "▁see d",
+      "▁ seed",
+      "AL L",
+      "A LL",
+      "▁n ation",
+      "▁na tion",
+      "▁nat ion",
+      "▁ nation",
+      "▁d ose",
+      "▁do se",
+      "▁dos e",
+      "▁ dose",
+      "▁K ore",
+      "▁Ko re",
+      "▁Kor e",
+      "学 科",
+      "▁v ivo",
+      "▁vi vo",
+      "▁ vivo",
+      "▁r oute",
+      "▁ro ute",
+      "▁rout e",
+      "▁rou te",
+      "▁ route",
+      "沟 通",
+      "ou nce",
+      "ounc e",
+      "oun ce",
+      "▁inc orpor",
+      "药 物",
+      "om etric",
+      "omet ric",
+      "o metric",
+      "宝 宝",
+      "ot ed",
+      "ote d",
+      "o ted",
+      "▁a wait",
+      "▁aw ait",
+      "▁awa it",
+      "▁ await",
+      "▁B ig",
+      "▁Bi g",
+      "▁ Big",
+      "▁Is land",
+      "os h",
+      "o sh",
+      "▁hyp othesis",
+      "▁hypot hesis",
+      "▁hypothes is",
+      "▁hypoth esis",
+      "▁p harm",
+      "▁ph arm",
+      "nP lease",
+      "n Please",
+      "▁P et",
+      "▁Pe t",
+      "▁ Pet",
+      "去 了",
+      "命 题",
+      "悔改 表现",
+      "St art",
+      "Star t",
+      "o j",
+      "▁n av",
+      "▁na v",
+      "▁ nav",
+      "ta n",
+      "t an",
+      "um ni",
+      "umn i",
+      "▁deg rees",
+      "▁degree s",
+      "▁ degrees",
+      "▁ch allenges",
+      "▁challeng es",
+      "▁challenge s",
+      "qu ad",
+      "qua d",
+      "看 着",
+      "▁ex pressed",
+      "▁exp ressed",
+      "▁express ed",
+      "▁expr essed",
+      "▁ expressed",
+      "▁P C",
+      "▁ PC",
+      "驳 回",
+      "▁contin ued",
+      "▁continu ed",
+      "▁continue d",
+      "▁str ength",
+      "▁stre ngth",
+      "▁streng th",
+      "▁ strength",
+      "▁P e",
+      "▁ Pe",
+      "▁T ext",
+      "▁Te xt",
+      "▁Tex t",
+      "▁ Text",
+      "r b",
+      "支 行",
+      "▁e mot",
+      "▁em ot",
+      "mod ules",
+      "module s",
+      "ca che",
+      "c ache",
+      "享 受",
+      "传 播",
+      "▁lim itations",
+      "▁limit ations",
+      "▁limitation s",
+      "▁ limitations",
+      "▁w arm",
+      "▁war m",
+      "▁wa rm",
+      "En g",
+      "E ng",
+      "ol a",
+      "o la",
+      "▁R ob",
+      "▁Ro b",
+      "▁ Rob",
+      "▁n one",
+      "▁no ne",
+      "▁non e",
+      "▁ none",
+      "▁b ond",
+      "▁bo nd",
+      "▁bon d",
+      "▁ bond",
+      "提供 的",
+      "▁re solution",
+      "▁res olution",
+      "▁ resolution",
+      "() {\\",
+      "(){ \\",
+      "( ){\\",
+      "▁Un ion",
+      "▁Uni on",
+      "▁ Union",
+      "T Y",
+      "编 辑",
+      "▁for med",
+      "▁form ed",
+      "▁ formed",
+      "▁un s",
+      "▁u ns",
+      "▁ uns",
+      "il er",
+      "ile r",
+      "i ler",
+      "name nt",
+      "na ment",
+      "nam ent",
+      "n ament",
+      "nO ne",
+      "nOn e",
+      "n One",
+      "管理 局",
+      "ó n",
+      "En d",
+      "E nd",
+      "▁in g",
+      "▁i ng",
+      "▁ ing",
+      "▁Res ults",
+      "▁Result s",
+      "▁ Results",
+      "▁r ats",
+      "▁rat s",
+      "▁ra ts",
+      "▁ rats",
+      "经审理 查明",
+      "de bug",
+      "deb ug",
+      "d ebug",
+      "▁mar ried",
+      "var i",
+      "va ri",
+      "v ari",
+      "res ents",
+      "resent s",
+      "▁re asons",
+      "▁reason s",
+      "ro r",
+      "r or",
+      "不 管",
+      "tt his",
+      "t this",
+      "in ated",
+      "ina ted",
+      "inate d",
+      "城 区",
+      "we ll",
+      "w ell",
+      "os tic",
+      "ost ic",
+      "▁l aunch",
+      "▁la unch",
+      "▁ launch",
+      "▁a gent",
+      "▁ag ent",
+      "▁age nt",
+      "▁ agent",
+      "el lig",
+      "ell ig",
+      "elli g",
+      "co hol",
+      "coh ol",
+      "▁in fr",
+      "▁inf r",
+      "▁rec ip",
+      "▁ recip",
+      "F C",
+      "▁inst alled",
+      "▁install ed",
+      "▁ installed",
+      "度 的",
+      "举 行",
+      "▁O h",
+      "▁ Oh",
+      "an cer",
+      "ance r",
+      "anc er",
+      "▁l etter",
+      "▁let ter",
+      "▁ letter",
+      "Sc ript",
+      "S cript",
+      "mu t",
+      "m ut",
+      "▁te chnical",
+      "▁techn ical",
+      "▁ technical",
+      "ia r",
+      "i ar",
+      "▁em pty",
+      "▁emp ty",
+      "▁ empty",
+      "> (",
+      "▁con v",
+      "▁co nv",
+      "▁ conv",
+      "an ta",
+      "ant a",
+      "▁i gn",
+      "▁ig n",
+      "▁ ign",
+      "另 一",
+      "▁ex plain",
+      "▁exp lain",
+      "▁expl ain",
+      "▁com pre",
+      "▁comp re",
+      "▁compr e",
+      "W A",
+      "期间 的",
+      "期 间的",
+      "▁p regn",
+      "▁pre gn",
+      "此 外",
+      "DI T",
+      "D IT",
+      "▁S ervices",
+      "▁Service s",
+      "▁Serv ices",
+      "▁ Services",
+      "▁op erator",
+      "▁oper ator",
+      "▁opera tor",
+      "▁ operator",
+      "▁w ed",
+      "▁we d",
+      "▁ wed",
+      "▁s ymbol",
+      "▁sym bol",
+      "▁ symbol",
+      "ab in",
+      "abi n",
+      "a bin",
+      "▁gu arant",
+      "▁min or",
+      "▁mi nor",
+      "▁ minor",
+      "▁S how",
+      "▁Sh ow",
+      "▁ Show",
+      "▁bud get",
+      "▁ budget",
+      "婚 姻",
+      "动 车",
+      "革 命",
+      "▁B usiness",
+      "▁Bus iness",
+      "▁ Business",
+      "歌 曲",
+      "▁I ts",
+      "▁It s",
+      "▁ Its",
+      "dr aw",
+      "d raw",
+      "mi d",
+      "m id",
+      "▁K ey",
+      "▁Ke y",
+      "▁ Key",
+      "▁un known",
+      "▁ unknown",
+      "▁com pare",
+      "▁comp are",
+      "▁compar e",
+      "▁ compare",
+      "公 示",
+      "▁l iver",
+      "▁li ver",
+      "▁live r",
+      "▁liv er",
+      "▁ liver",
+      "▁cor responding",
+      "▁correspon ding",
+      "▁correspond ing",
+      "▁de lay",
+      "▁del ay",
+      "▁ delay",
+      "▁M ichael",
+      "▁Mich ael",
+      "▁Mic hael",
+      "▁ Michael",
+      "▁s in",
+      "▁si n",
+      "▁ sin",
+      "广 大",
+      "el ines",
+      "eline s",
+      "eli nes",
+      "elin es",
+      "e lines",
+      "装 修",
+      "in ite",
+      "init e",
+      "ini te",
+      "民间借贷 纠纷一案",
+      "民间借贷纠纷 一案",
+      "Si m",
+      "S im",
+      "的人 数",
+      "的 人数",
+      "▁be comes",
+      "▁bec omes",
+      "▁become s",
+      "ur a",
+      "u ra",
+      "▁aff ected",
+      "▁affect ed",
+      "▁ affected",
+      "RO M",
+      "R OM",
+      "循 环",
+      "多 个",
+      "下 降",
+      "八 年",
+      "▁br anch",
+      "▁bra nch",
+      "▁ branch",
+      "el i",
+      "e li",
+      "▁A R",
+      "▁ AR",
+      "中华人民共和国 刑法",
+      "▁in struction",
+      "▁instr uction",
+      "▁instruct ion",
+      "▁ instruction",
+      "pl ements",
+      "ple ments",
+      "plement s",
+      "p lements",
+      "▁S am",
+      "▁Sa m",
+      "▁ Sam",
+      "▁De velopment",
+      "▁Develop ment",
+      "▁ Development",
+      "Pos t",
+      "Po st",
+      "P ost",
+      "文 字",
+      "## ###",
+      "#### #",
+      "### ##",
+      "# ####",
+      "微 博",
+      "▁business es",
+      "requ ired",
+      "require d",
+      "生 育",
+      "▁p ictures",
+      "▁pict ures",
+      "▁picture s",
+      "▁pic tures",
+      "几 乎",
+      "ro ad",
+      "r oad",
+      "ou ter",
+      "out er",
+      "oute r",
+      "o uter",
+      "▁d eb",
+      "▁de b",
+      "▁ deb",
+      "评 估",
+      "选 出",
+      "dir ect",
+      "di rect",
+      "d irect",
+      "F l",
+      "英 雄",
+      "▁L ine",
+      "▁Lin e",
+      "▁Li ne",
+      "▁ Line",
+      "▁decre ased",
+      "▁decrease d",
+      "▁G ood",
+      "▁Go od",
+      "▁ Good",
+      "OR D",
+      "O RD",
+      "搜 索",
+      "er man",
+      "erm an",
+      "pl an",
+      "p lan",
+      "▁stat ist",
+      "l anguage",
+      "是 因为",
+      "az ine",
+      "azi ne",
+      "▁p et",
+      "▁pe t",
+      "▁ pet",
+      "▁ne ut",
+      "▁ neut",
+      "▁st ories",
+      "▁sto ries",
+      "▁ stories",
+      "us tr",
+      "ust r",
+      "u str",
+      "pa y",
+      "p ay",
+      ")\\\\ )",
+      ") \\\\)",
+      "▁off ered",
+      "▁offer ed",
+      "▁[ \"",
+      "▁ [\"",
+      "▁b order",
+      "▁bor der",
+      "▁ border",
+      "sh ire",
+      "shi re",
+      "s hire",
+      "▁G NU",
+      "▁ GNU",
+      "▁network s",
+      "▁net works",
+      "▁ networks",
+      "良好 的",
+      "良 好的",
+      "应 予",
+      "bl og",
+      "blo g",
+      "b log",
+      "Re d",
+      "R ed",
+      "▁control s",
+      "▁contr ols",
+      "▁ controls",
+      "▁appe ared",
+      "▁appear ed",
+      "大 量",
+      "的 好",
+      "上 了",
+      "to tal",
+      "tot al",
+      "t otal",
+      "高 校",
+      "具 备",
+      "▁B oard",
+      "▁Bo ard",
+      "▁ Board",
+      "▁in ternet",
+      "▁inter net",
+      "▁intern et",
+      "▁ internet",
+      "▁invest ment",
+      "贵 州",
+      "向本院 申请",
+      "▁J er",
+      "▁Je r",
+      "▁d ress",
+      "▁dr ess",
+      "▁fin ish",
+      "▁ finish",
+      "de l",
+      "d el",
+      "主 席",
+      "▁feed back",
+      "▁ feedback",
+      "bl ob",
+      "blo b",
+      "b lob",
+      "pl ugin",
+      "plug in",
+      "▁Fr anc",
+      "▁Fra nc",
+      "▁ex ternal",
+      "▁ext ernal",
+      "▁extern al",
+      "▁ external",
+      "▁f ish",
+      "▁fi sh",
+      "▁ fish",
+      "接 触",
+      "ience s",
+      "ien ces",
+      "i ences",
+      "▁ad ults",
+      "▁adult s",
+      "▁adul ts",
+      "自 我",
+      "▁work ers",
+      "▁wor kers",
+      "▁worker s",
+      "▁ workers",
+      "生 长",
+      "bo ot",
+      "b oot",
+      "ag en",
+      "age n",
+      "a gen",
+      "▁f urn",
+      "▁fur n",
+      "▁fu rn",
+      "▁ 选项",
+      "▁[ \\'",
+      "▁[\\ '",
+      "▁ [\\'",
+      "▁ex amine",
+      "▁exam ine",
+      "在 服刑",
+      "神 经",
+      "▁e lev",
+      "▁el ev",
+      "▁ele v",
+      "B O",
+      "▁arg ument",
+      "▁argu ment",
+      "▁ argument",
+      "▁An gel",
+      "▁Ang el",
+      "导 演",
+      "唯 一",
+      "对 其",
+      "▁( \"",
+      "▁ (\"",
+      "▁op ening",
+      "▁open ing",
+      "▁ opening",
+      "▁St ep",
+      "▁Ste p",
+      "▁ Step",
+      "▁T uesday",
+      "W S",
+      "ro ke",
+      "r oke",
+      "▁invest igation",
+      "▁investig ation",
+      "▁dep artment",
+      "▁depart ment",
+      "▁ department",
+      "B l",
+      "▁de tail",
+      "▁det ail",
+      "▁ detail",
+      "re fix",
+      "ref ix",
+      "up s",
+      "u ps",
+      "\\' :\\",
+      "\\': \\",
+      "\\ ':\\",
+      "▁A tt",
+      "▁At t",
+      "▁ Att",
+      "挑 战",
+      "▁S y",
+      "▁ Sy",
+      "er ies",
+      "eri es",
+      "erie s",
+      "e ries",
+      "it ure",
+      "itu re",
+      "i ture",
+      "▁s hot",
+      "▁sh ot",
+      "▁ shot",
+      "in ates",
+      "ina tes",
+      "inate s",
+      "▁s ides",
+      "▁side s",
+      "▁si des",
+      "▁sid es",
+      "▁d u",
+      "▁ du",
+      "▁m il",
+      "▁mi l",
+      "▁ mil",
+      "通知 书",
+      "男 人",
+      "▁rem oved",
+      "▁remove d",
+      "▁remov ed",
+      "▁ removed",
+      "Op en",
+      "O pen",
+      "us ers",
+      "user s",
+      "use rs",
+      "nS h",
+      "n Sh",
+      "▁demonstr ate",
+      "会 计",
+      "conn ect",
+      "ig nment",
+      "ign ment",
+      "▁w ind",
+      "▁win d",
+      "▁wi nd",
+      "▁ wind",
+      "ir t",
+      "i rt",
+      "am il",
+      "ami l",
+      "a mil",
+      "▁co eff",
+      "▁ coeff",
+      "cm d",
+      "c md",
+      "an der",
+      "and er",
+      "ande r",
+      "a nder",
+      "ro c",
+      "r oc",
+      "▁ins ert",
+      "▁ insert",
+      "▁immun e",
+      "▁imm une",
+      "▁ immune",
+      "▁came ra",
+      "▁cam era",
+      "▁ camera",
+      "线 索",
+      "▁Se cond",
+      "▁Sec ond",
+      "▁ Second",
+      "▁S l",
+      "▁ Sl",
+      "ul ty",
+      "ult y",
+      "▁p an",
+      "▁pa n",
+      "▁ pan",
+      "公 路",
+      "ak s",
+      "a ks",
+      "il on",
+      "ilo n",
+      "i lon",
+      "西 安",
+      "一定 的",
+      "一 定的",
+      "▁f ol",
+      "▁fo l",
+      "▁ fol",
+      "▁A nn",
+      "▁An n",
+      "▁ Ann",
+      "d r",
+      "▁Car ol",
+      "▁Ca rol",
+      "es ome",
+      "e some",
+      "下 午",
+      "给 我",
+      "▁con firm",
+      "▁conf irm",
+      "▁ confirm",
+      "▁ob jective",
+      "▁object ive",
+      "▁obj ective",
+      "▁ objective",
+      "sh ould",
+      "▁inter pret",
+      "▁ interpret",
+      "数 列",
+      "▁b rought",
+      "▁br ought",
+      "▁bro ught",
+      "um s",
+      "u ms",
+      "▁Th ank",
+      "▁Than k",
+      "▁ Thank",
+      "LIC ENSE",
+      "▁s pot",
+      "▁sp ot",
+      "▁spo t",
+      "▁ spot",
+      "▁sec onds",
+      "▁second s",
+      "▁ seconds",
+      "▁reg ulation",
+      "▁regul ation",
+      "▁ regulation",
+      "本院在审理 原告",
+      "ip ping",
+      "ipp ing",
+      "ippi ng",
+      "i pping",
+      "ed y",
+      "e dy",
+      "Res ource",
+      "Re source",
+      "月 份",
+      "▁compar ison",
+      "向 原告",
+      "▁let ters",
+      "▁letter s",
+      "▁ letters",
+      "ba l",
+      "b al",
+      "受理 后",
+      "▁出版 社",
+      "▁ 出版社",
+      "▁ac ute",
+      "▁ acute",
+      "en ed",
+      "ene d",
+      "e ned",
+      "Int eger",
+      "asket ball",
+      "供 应",
+      "终结 本次执行程序",
+      "终结本次执行 程序",
+      "▁Olymp ics",
+      "▁Olympic s",
+      "am ing",
+      "amin g",
+      "ami ng",
+      "a ming",
+      "考 核",
+      "放 弃",
+      "en a",
+      "e na",
+      "含 量",
+      "▁ch anging",
+      "▁chang ing",
+      "▁chan ging",
+      "▁ changing",
+      "▁adv antage",
+      "▁advant age",
+      "▁pl ants",
+      "▁plan ts",
+      "▁plant s",
+      "▁ plants",
+      "▁art icles",
+      "▁article s",
+      "▁artic les",
+      "▁ articles",
+      "math cal",
+      "t n",
+      "▁integr al",
+      "be fore",
+      "b efore",
+      "重 视",
+      "▁T ur",
+      "▁Tu r",
+      "▁ Tur",
+      "AC K",
+      "A CK",
+      "▁c it",
+      "▁ci t",
+      "▁ cit",
+      "act iv",
+      "st ep",
+      "ste p",
+      "▁le ads",
+      "▁lead s",
+      "管 辖",
+      "urs or",
+      "▁g rand",
+      "▁gr and",
+      "▁gran d",
+      "▁gra nd",
+      "▁ grand",
+      "做 出",
+      "al low",
+      "all ow",
+      "allo w",
+      "▁in come",
+      "▁inc ome",
+      "▁ income",
+      "olog ic",
+      "olo gic",
+      "o logic",
+      "愿 意",
+      "更 好",
+      "▁aspect s",
+      "▁asp ects",
+      "▁B ank",
+      "▁Ba nk",
+      "▁Ban k",
+      "▁ Bank",
+      "▁a sync",
+      "▁as ync",
+      "▁ async",
+      "▁he avy",
+      "▁heav y",
+      "▁ heavy",
+      "色 的",
+      "在一 起",
+      "在 一起",
+      "种 植",
+      "感 受",
+      "ir o",
+      "i ro",
+      "▁r ing",
+      "▁ri ng",
+      "▁ ring",
+      "▁off ering",
+      "▁offer ing",
+      "Pro ject",
+      "* (",
+      "第一百四十五 条第一款",
+      "第一百四十五条 第一款",
+      "▁m agn",
+      "▁mag n",
+      "▁ma gn",
+      "▁ magn",
+      "PR O",
+      "P RO",
+      "▁c atal",
+      "▁ca tal",
+      "▁cat al",
+      "▁ catal",
+      "E E",
+      "n Q",
+      "Pe r",
+      "P er",
+      "y e",
+      "ag ment",
+      "a gment",
+      "Ed it",
+      "E dit",
+      "am ine",
+      "amin e",
+      "ami ne",
+      "a mine",
+      "▁sy nd",
+      "▁syn d",
+      "Pa ss",
+      "P ass",
+      "▁t rip",
+      "▁tr ip",
+      "▁tri p",
+      "▁ trip",
+      "fil l",
+      "fi ll",
+      "f ill",
+      "ran dom",
+      "rand om",
+      "r andom",
+      "▁l iqu",
+      "▁li qu",
+      "▁c arbon",
+      "▁car bon",
+      "▁carb on",
+      "▁ carbon",
+      "▁Th ursday",
+      "▁ Thursday",
+      "电 池",
+      "▁conf irmed",
+      "▁confirm ed",
+      "▁ confirmed",
+      "▁C ode",
+      "▁Co de",
+      "▁Cod e",
+      "▁ Code",
+      "▁b eta",
+      "▁be ta",
+      "▁bet a",
+      "▁ beta",
+      "▁A v",
+      "▁ Av",
+      "组 合",
+      "▁con sequ",
+      "▁cons equ",
+      "▁L ake",
+      "▁La ke",
+      "▁Lak e",
+      "▁at tend",
+      "▁att end",
+      "▁atten d",
+      "▁oper ating",
+      "▁opera ting",
+      "哈 哈",
+      "河北 省",
+      "河 北省",
+      "▁v an",
+      "▁va n",
+      "▁ van",
+      "nC on",
+      "nCo n",
+      "n Con",
+      "▁d river",
+      "▁dr iver",
+      "▁dri ver",
+      "▁driv er",
+      "▁drive r",
+      "▁ driver",
+      "▁E qu",
+      "▁Eq u",
+      "▁ Equ",
+      "vo l",
+      "v ol",
+      "方面 的",
+      "方 面的",
+      "节 目",
+      "▁k itchen",
+      "▁kit chen",
+      "▁d rink",
+      "▁dr ink",
+      "▁dri nk",
+      "▁t opic",
+      "▁to pic",
+      "▁top ic",
+      "▁ topic",
+      "做 到",
+      "▁spec ified",
+      "▁ specified",
+      "__ ______",
+      "____ ____",
+      "______ __",
+      "___ _____",
+      "_____ ___",
+      "▁prep ared",
+      "▁prepar ed",
+      "▁prepare d",
+      "▁ prepared",
+      "▁M ad",
+      "▁Ma d",
+      "▁ Mad",
+      "▁W estern",
+      "▁West ern",
+      "▁Wes tern",
+      "▁ Western",
+      "categor y",
+      "c ategory",
+      "▁c onstr",
+      "▁con str",
+      "▁co nstr",
+      "▁cons tr",
+      "▁const r",
+      "cl oud",
+      "趋 势",
+      "▁E very",
+      "▁Ev ery",
+      "▁Ever y",
+      "▁Eve ry",
+      "▁ Every",
+      "▁c auses",
+      "▁ca uses",
+      "▁cause s",
+      "▁caus es",
+      "会 员",
+      "配 合",
+      "cu t",
+      "c ut",
+      "固 定",
+      "故 选",
+      "▁met abol",
+      "▁metab ol",
+      "▁meta bol",
+      "经 典",
+      "▁f ocused",
+      "▁foc used",
+      "▁focus ed",
+      "▁ focused",
+      "▁ 的",
+      "ven ue",
+      "v enue",
+      "▁football ers",
+      "▁footballer s",
+      "yi ng",
+      "y ing",
+      "▁C all",
+      "▁Cal l",
+      "▁Ca ll",
+      "▁ Call",
+      "▁C olor",
+      "▁Col or",
+      "▁Co lor",
+      "▁ Color",
+      "il ing",
+      "ili ng",
+      "ilin g",
+      "i ling",
+      "d t",
+      "▁T or",
+      "▁To r",
+      "▁ Tor",
+      "▁re nt",
+      "▁r ent",
+      "▁ rent",
+      "▁mem brane",
+      "▁membr ane",
+      "▁ membrane",
+      "() ,\\",
+      "(), \\",
+      "( ),\\",
+      "art icle",
+      "素 质",
+      "▁success fully",
+      "▁successful ly",
+      "声 音",
+      "▁M usic",
+      "▁Mus ic",
+      "▁ Music",
+      "本 金",
+      "un te",
+      "unt e",
+      "u nte",
+      "fa st",
+      "fas t",
+      "f ast",
+      "网络 小说",
+      "▁relations hips",
+      "▁relationship s",
+      "▁relation ships",
+      "▁O b",
+      "▁ Ob",
+      "▁ed itor",
+      "▁edit or",
+      "▁ editor",
+      "//// ////",
+      "▁e yes",
+      "▁ey es",
+      "▁eye s",
+      "▁e nable",
+      "▁en able",
+      "▁ enable",
+      "▁week end",
+      "▁out come",
+      "▁de tailed",
+      "▁det ailed",
+      "▁detail ed",
+      "▁w ants",
+      "▁want s",
+      "br id",
+      "b rid",
+      "t ension",
+      "ogen esis",
+      "ogene sis",
+      "ogenes is",
+      "o genesis",
+      "en try",
+      "ent ry",
+      "e ntry",
+      "ze n",
+      "z en",
+      "▁sugg ested",
+      "▁suggest ed",
+      ")) ,\\",
+      ")), \\",
+      ") ),\\",
+      "cri be",
+      "crib e",
+      "cr ibe",
+      "c ribe",
+      "态 度",
+      "投 入",
+      "优 质",
+      "那 种",
+      "吸 引",
+      ")$ \\",
+      ") $\\",
+      "▁a lt",
+      "▁al t",
+      "▁ alt",
+      "▁主 要",
+      "▁ 主要",
+      "Up date",
+      "准 确",
+      "▁c as",
+      "▁ca s",
+      "▁ cas",
+      "▁w asn",
+      "▁was n",
+      "▁wa sn",
+      "▁C onn",
+      "▁Con n",
+      "▁Co nn",
+      "▁ Conn",
+      "经 本院",
+      "▁W ednesday",
+      "▁Wed nesday",
+      "in gu",
+      "ing u",
+      "的 方式",
+      "ri e",
+      "r ie",
+      "怎 样",
+      "est amp",
+      "esta mp",
+      "e stamp",
+      "装 饰",
+      "▁adv ance",
+      "的一 种",
+      "的 一种",
+      "B U",
+      "se ct",
+      "sec t",
+      "s ect",
+      "▁trans ition",
+      "▁transit ion",
+      "▁ transition",
+      "▁R oman",
+      "▁Ro man",
+      "▁Rom an",
+      "▁Roma n",
+      "▁ Roman",
+      "实 力",
+      "or ter",
+      "ort er",
+      "orte r",
+      "▁Off ice",
+      "▁ Office",
+      "▁T our",
+      "▁To ur",
+      "▁Tou r",
+      "▁ Tour",
+      "▁cl oud",
+      "▁ cloud",
+      "研究 生",
+      "rom ise",
+      "▁M anagement",
+      "▁Man agement",
+      "▁Manag ement",
+      "▁Manage ment",
+      "▁ Management",
+      "ot hing",
+      "oth ing",
+      "o thing",
+      "追 求",
+      "dj ango",
+      "djang o",
+      "d jango",
+      "您 的",
+      "▁fe et",
+      "▁fee t",
+      "认 罪",
+      "逐 渐",
+      "▁main tain",
+      "▁maint ain",
+      "例 如",
+      "湖南 省",
+      "湖 南省",
+      "▁b atter",
+      "▁bat ter",
+      "▁batt er",
+      "tp rivate",
+      "t private",
+      "物 流",
+      "ou l",
+      "o ul",
+      "▁requ ests",
+      "▁request s",
+      "▁ requests",
+      "▁P ress",
+      "▁Pr ess",
+      "▁Pres s",
+      "▁Pre ss",
+      "▁ Press",
+      "▁tra ffic",
+      "▁traff ic",
+      "▁ traffic",
+      "▁R equ",
+      "▁Re qu",
+      "▁ Requ",
+      "spring framework",
+      "▁measure ments",
+      "▁measurement s",
+      "an al",
+      "ana l",
+      "a nal",
+      "我 也",
+      "nH istory",
+      "nHi story",
+      "n History",
+      "号 刑事",
+      "▁V er",
+      "▁Ve r",
+      "▁ Ver",
+      "in ant",
+      "ina nt",
+      "i nant",
+      "与被 执行人",
+      "与 被执行人",
+      "▁c overed",
+      "▁cover ed",
+      "▁cov ered",
+      "▁ covered",
+      "ta sk",
+      "tas k",
+      "t ask",
+      "ar c",
+      "a rc",
+      "▁C le",
+      "▁Cl e",
+      "上 市",
+      "资 格",
+      "qu are",
+      "qua re",
+      "高 速",
+      "▁th ick",
+      "▁thi ck",
+      "▁ thick",
+      "让 你",
+      "fo und",
+      "f ound",
+      "▁p aid",
+      "▁pa id",
+      "▁ paid",
+      "▁出版 时间",
+      "▁ 出版时间",
+      "le ts",
+      "let s",
+      "l ets",
+      "▁M ass",
+      "▁Ma ss",
+      "▁Mas s",
+      "▁ Mass",
+      "▁big gest",
+      "▁se vere",
+      "▁sever e",
+      "收 到",
+      "▁Com put",
+      "▁Comp ut",
+      "▁ Comput",
+      "▁other wise",
+      "▁ otherwise",
+      "▁power ful",
+      "最 小",
+      "公司 的",
+      "Ne t",
+      "N et",
+      "t m",
+      "去 年",
+      "% \\",
+      "▁T om",
+      "▁To m",
+      "▁ Tom",
+      "▁art ist",
+      "▁ artist",
+      "永 远",
+      "as ion",
+      "asi on",
+      "▁D ev",
+      "▁De v",
+      "▁ Dev",
+      "de m",
+      "d em",
+      "▁S il",
+      "▁Si l",
+      "▁ Sil",
+      "认 证",
+      "cl usive",
+      "clus ive",
+      "▁U RL",
+      "▁UR L",
+      "▁ URL",
+      "称 为",
+      "▁o nto",
+      "▁on to",
+      "▁ont o",
+      "▁ onto",
+      "ner gy",
+      "nerg y",
+      "n ergy",
+      "▁ £",
+      "指 数",
+      "F e",
+      "污 染",
+      "自 行",
+      "▁p ip",
+      "▁pi p",
+      "▁ pip",
+      "▁un c",
+      "▁u nc",
+      "▁ unc",
+      "传 奇",
+      "▁c ellular",
+      "▁cell ular",
+      "▁ cellular",
+      "美 术",
+      "ge q",
+      "g eq",
+      "il ler",
+      "ill er",
+      "ille r",
+      "i ller",
+      "ook s",
+      "oo ks",
+      "o oks",
+      "原 理",
+      "工作 的",
+      "工 作的",
+      "point s",
+      "po ints",
+      "p oints",
+      "房 产",
+      "▁▁ ▁▁▁",
+      "▁▁▁▁ ▁",
+      "▁▁▁ ▁▁",
+      "▁ ▁▁▁▁",
+      "▁ 游戏",
+      "指 出",
+      "▁k id",
+      "▁ki d",
+      "▁ kid",
+      "math bf",
+      "os a",
+      "o sa",
+      ")$ $",
+      ") $$",
+      "名 单",
+      "检 验",
+      "Ac c",
+      "A cc",
+      "▁spec ifically",
+      "▁specific ally",
+      "▁t oken",
+      "▁to ken",
+      "▁tok en",
+      "▁ token",
+      "O K",
+      "简 易",
+      "m g",
+      "ri z",
+      "r iz",
+      "▁Ge orge",
+      "▁Georg e",
+      "▁ George",
+      "▁de rived",
+      "▁der ived",
+      "▁deriv ed",
+      "▁derive d",
+      "▁ derived",
+      "▁p late",
+      "▁pl ate",
+      "▁plat e",
+      "▁ plate",
+      "cu r",
+      "c ur",
+      "▁c alls",
+      "▁call s",
+      "▁cal ls",
+      "▁ calls",
+      "提供 了",
+      "ya n",
+      "y an",
+      "护 理",
+      "获 取",
+      "▁\" \"",
+      "▁ \"\"",
+      "豆 瓣",
+      "▁Can adian",
+      "▁Canad ian",
+      "▁ Canadian",
+      "▁al pha",
+      "▁alph a",
+      "▁ alpha",
+      "▁W ell",
+      "▁We ll",
+      "▁Wel l",
+      "▁ Well",
+      "▁Com mit",
+      "▁Comm it",
+      "▁ Commit",
+      "▁S cott",
+      "▁Sc ott",
+      "▁Scot t",
+      "▁ Scott",
+      "多 年",
+      "缴 纳",
+      "我 想",
+      "▁i ce",
+      "▁ic e",
+      "▁ ice",
+      "ac cess",
+      "acc ess",
+      "a ccess",
+      "▁major ity",
+      "▁W hy",
+      "▁Wh y",
+      "▁ Why",
+      "▁ext remely",
+      "▁extrem ely",
+      "▁extreme ly",
+      "bi g",
+      "b ig",
+      "▁ex change",
+      "▁ exchange",
+      "▁com prehens",
+      "▁compre hens",
+      "ans ion",
+      "ansi on",
+      "▁g ar",
+      "▁ga r",
+      "▁ gar",
+      "ic tor",
+      "ict or",
+      "i ctor",
+      "▁autom atically",
+      "▁automatic ally",
+      "有 着",
+      "▁B oth",
+      "▁Bo th",
+      "▁Bot h",
+      "▁ Both",
+      "主 持",
+      "bur g",
+      "bu rg",
+      "b urg",
+      "ia nt",
+      "ian t",
+      "i ant",
+      "▁Ex per",
+      "▁Exp er",
+      "▁ Exper",
+      "格 式",
+      "W ork",
+      "po wer",
+      "pow er",
+      "p ower",
+      "V al",
+      "就 不",
+      "广 场",
+      "▁ob vious",
+      "▁d aughter",
+      "▁ daughter",
+      "KE Y",
+      "K EY",
+      "▁WARRANT IES",
+      "一 位",
+      "修 改",
+      "th row",
+      "thro w",
+      "thr ow",
+      "▁t heme",
+      "▁the me",
+      "▁th eme",
+      "▁them e",
+      "▁ theme",
+      "官 方",
+      "▁O per",
+      "▁Op er",
+      "▁ Oper",
+      "校 园",
+      "▁own er",
+      "▁ow ner",
+      "▁ owner",
+      "▁% }\\",
+      "▁%} \\",
+      "▁ %}\\",
+      "> :",
+      "fr ame",
+      "fram e",
+      "fra me",
+      "f rame",
+      "▁c ash",
+      "▁ca sh",
+      "▁cas h",
+      "▁ cash",
+      "▁Di vision",
+      "▁Div ision",
+      "▁Divis ion",
+      "mo r",
+      "m or",
+      "▁con ference",
+      "▁confer ence",
+      "▁ conference",
+      "出版 的",
+      "出 版的",
+      "ag ement",
+      "age ment",
+      "agem ent",
+      "▁c orpor",
+      "▁cor por",
+      "um er",
+      "ume r",
+      "u mer",
+      "| |",
+      "P P",
+      "▁al bums",
+      "▁album s",
+      "am l",
+      "a ml",
+      "urre ncy",
+      "urrenc y",
+      "al i",
+      "a li",
+      "▁fac ilit",
+      "批 准",
+      "他 人",
+      "nt o",
+      "n to",
+      "re place",
+      "名 的",
+      "▁def initely",
+      "▁definit ely",
+      "▁defin itely",
+      "▁definite ly",
+      "有 哪些",
+      "la b",
+      "l ab",
+      "▁U nder",
+      "▁Un der",
+      "▁Und er",
+      "▁ Under",
+      "ad y",
+      "a dy",
+      "uc e",
+      "u ce",
+      "▁v en",
+      "▁ve n",
+      "▁ ven",
+      "▁turn ed",
+      "▁tur ned",
+      "▁ turned",
+      "反 映",
+      "▁acc uracy",
+      "▁accur acy",
+      "▁ accuracy",
+      "fo l",
+      "f ol",
+      "▁ 国",
+      "爱 的",
+      "▁e lim",
+      "▁el im",
+      "▁ach ieved",
+      "▁achieve d",
+      "” “",
+      "▁main tenance",
+      "▁bal ance",
+      "▁ balance",
+      "▁de ploy",
+      "▁dep loy",
+      "▁ deploy",
+      "上 面",
+      "sc ri",
+      "scr i",
+      "s cri",
+      "欧 洲",
+      "im s",
+      "i ms",
+      "▁adv ice",
+      "oc ytes",
+      "ocyte s",
+      "ocyt es",
+      "▁ad vert",
+      "▁adv ert",
+      "▁ advert",
+      "▁integr ation",
+      "▁ integration",
+      "美 的",
+      "od er",
+      "ode r",
+      "o der",
+      "至 于",
+      "nN o",
+      "n No",
+      "▁st rict",
+      "▁str ict",
+      "▁stri ct",
+      "▁ strict",
+      "▁s pend",
+      "▁sp end",
+      "▁spe nd",
+      "被 害",
+      "nC an",
+      "n Can",
+      "av is",
+      "avi s",
+      "a vis",
+      "pd f",
+      "p df",
+      "总 部",
+      "▁h ab",
+      "▁ha b",
+      "▁ hab",
+      "▁he ard",
+      "▁hear d",
+      "ri nt",
+      "rin t",
+      "r int",
+      "正 当",
+      "in n",
+      "i nn",
+      "担 任",
+      "nS up",
+      "n Sup",
+      "晚 上",
+      "▁concentration s",
+      "▁concentr ations",
+      "▁rep resents",
+      "▁represent s",
+      "好 了",
+      "如 今",
+      "am i",
+      "a mi",
+      "▁W omen",
+      "▁Wo men",
+      "▁ Women",
+      "ec k",
+      "e ck",
+      "M C",
+      "▁b illion",
+      "▁bill ion",
+      "▁bil lion",
+      "▁rel ax",
+      "▁ relax",
+      "▁m ixed",
+      "▁mix ed",
+      "▁ mixed",
+      "多 了",
+      "▁re strict",
+      "▁rest rict",
+      "▁restr ict",
+      "▁I L",
+      "▁ IL",
+      "yn omial",
+      "] ]",
+      "▁C ap",
+      "▁Ca p",
+      "▁ Cap",
+      "ro t",
+      "r ot",
+      "确定 的义务",
+      "确定的 义务",
+      "do uble",
+      "dou ble",
+      "d ouble",
+      "▁m orph",
+      "▁mor ph",
+      "▁ morph",
+      "的 不",
+      "▁em erg",
+      "▁emer g",
+      "▁ emerg",
+      "( _",
+      "▁per missions",
+      "▁permission s",
+      "▁perm issions",
+      "▁ permissions",
+      "en ter",
+      "ent er",
+      "ente r",
+      "e nter",
+      "▁F lorida",
+      "▁Flor ida",
+      "职 务",
+      "et t",
+      "e tt",
+      "ess ions",
+      "ession s",
+      "▁organ izations",
+      "▁organization s",
+      "abet es",
+      "abe tes",
+      "▁ther apeutic",
+      "es ted",
+      "est ed",
+      "este d",
+      "e sted",
+      "▁W illiam",
+      "▁Willi am",
+      "▁Will iam",
+      "▁ William",
+      "此 次",
+      "** (",
+      "* *(",
+      "T C",
+      "递交 上诉状",
+      "河 南",
+      "性 和",
+      "▁h arm",
+      "▁har m",
+      "▁ha rm",
+      "▁ harm",
+      "太 阳",
+      "▁s plit",
+      "▁sp lit",
+      "▁spl it",
+      "▁ split",
+      "发 行",
+      "同 样",
+      "▁J apanese",
+      "▁Japan ese",
+      "▁ Japanese",
+      "所 属",
+      "公 园",
+      "▁M any",
+      "▁Man y",
+      "▁Ma ny",
+      "▁ Many",
+      "▁S ep",
+      "▁Se p",
+      "▁ Sep",
+      "都 能",
+      "Ta sk",
+      "T ask",
+      "▁w ire",
+      "▁wi re",
+      "▁ wire",
+      "od ed",
+      "ode d",
+      "o ded",
+      "司 法",
+      "List ener",
+      "Listen er",
+      "▁c andidate",
+      "▁candid ate",
+      "▁ candidate",
+      "德 国",
+      "▁inv olve",
+      "▁invol ve",
+      "D I",
+      "▁we ather",
+      "▁ weather",
+      "id ers",
+      "ide rs",
+      "ider s",
+      "i ders",
+      "▁m otion",
+      "▁mot ion",
+      "▁mo tion",
+      "▁ motion",
+      "av our",
+      "avo ur",
+      "line ar",
+      "li near",
+      "lin ear",
+      "V E",
+      "Pr e",
+      "P re",
+      "适 应",
+      "▁l eader",
+      "▁le ader",
+      "▁lead er",
+      "▁ leader",
+      "S p",
+      "print ln",
+      "各 项",
+      "▁ ^",
+      "at ever",
+      "ate ver",
+      "mi c",
+      "m ic",
+      "即 使",
+      "家 里",
+      "选项 中",
+      "ip l",
+      "i pl",
+      "▁de cor",
+      "▁dec or",
+      "▁ decor",
+      "▁res pond",
+      "▁respon d",
+      "▁resp ond",
+      "▁ respond",
+      "vent ional",
+      "vention al",
+      "高 度",
+      "相 比",
+      "▁anal yzed",
+      "▁analy zed",
+      "▁analyz ed",
+      "▁analyze d",
+      "总 是",
+      "▁ag reement",
+      "▁agre ement",
+      "▁agree ment",
+      "▁acc el",
+      "▁ac cel",
+      "实 行",
+      "▁man age",
+      "▁manag e",
+      "▁ manage",
+      "▁p ed",
+      "▁pe d",
+      "▁ ped",
+      "▁girl s",
+      "▁gir ls",
+      "▁R obert",
+      "▁Rob ert",
+      "▁Ro bert",
+      "▁Rober t",
+      "▁ Robert",
+      "ni ght",
+      "n ight",
+      "▁b elong",
+      "▁be long",
+      "▁bel ong",
+      "Th read",
+      "▁ne arest",
+      "▁near est",
+      "▁p lot",
+      "▁pl ot",
+      "▁ plot",
+      "do or",
+      "d oor",
+      "▁aff ord",
+      "▁af ford",
+      "▁ded icated",
+      "Ho w",
+      "H ow",
+      "想 到",
+      "遵 守",
+      "▁M ont",
+      "▁Mon t",
+      "▁Mo nt",
+      "▁ Mont",
+      "ell ular",
+      "▁S un",
+      "▁Su n",
+      "▁ Sun",
+      "V D",
+      "F O",
+      "贡 献",
+      "▁f ont",
+      "▁fo nt",
+      "▁ font",
+      "我 不",
+      "▁pract ices",
+      "▁practice s",
+      "▁Ch ild",
+      "▁Chi ld",
+      "▁ Child",
+      "▁con ven",
+      "▁conv en",
+      "▁conve n",
+      "ud o",
+      "u do",
+      "\\\\ )\\",
+      "\\\\) \\",
+      "▁Christ mas",
+      "M I",
+      "▁improve ment",
+      "▁improv ement",
+      "▁G en",
+      "▁Ge n",
+      "▁ Gen",
+      "一 段",
+      "{ }",
+      "手 段",
+      "ed it",
+      "edi t",
+      "e dit",
+      "提 示",
+      "▁decre ase",
+      "ma ke",
+      "mak e",
+      "m ake",
+      "▁B ecause",
+      "▁Be cause",
+      "▁Bec ause",
+      "▁ Because",
+      "▁s table",
+      "▁st able",
+      "▁stab le",
+      "▁ stable",
+      "相 互",
+      "村 民",
+      "其 它",
+      "cl ear",
+      "cle ar",
+      "c lear",
+      "战 争",
+      "▁A lex",
+      "▁Al ex",
+      "▁Ale x",
+      "▁ Alex",
+      "▁est imated",
+      "▁estim ated",
+      "▁estimate d",
+      "▁sc heme",
+      "▁sch eme",
+      "▁ scheme",
+      "▁m essages",
+      "▁message s",
+      "▁mess ages",
+      "▁ messages",
+      "玻 璃",
+      "Cl ick",
+      "Cli ck",
+      "C lick",
+      "▁th ank",
+      "▁than k",
+      "& #",
+      "cus tom",
+      "c ustom",
+      "Ba ck",
+      "B ack",
+      "nD es",
+      "nDe s",
+      "n Des",
+      "▁n ative",
+      "▁nat ive",
+      "▁ native",
+      "▁d riving",
+      "▁dr iving",
+      "▁dri ving",
+      "▁driv ing",
+      "▁ driving",
+      "Fil ter",
+      "▁inter val",
+      "▁ interval",
+      "罗 斯",
+      "h r",
+      "▁d oll",
+      "▁do ll",
+      "▁dol l",
+      "▁Pro fess",
+      "▁Prof ess",
+      "▁f inite",
+      "▁fin ite",
+      "▁) ;\\",
+      "▁); \\",
+      "▁ );\\",
+      "确有 悔改表现",
+      "▁A S",
+      "▁ AS",
+      "▁W ill",
+      "▁Wil l",
+      "▁Wi ll",
+      "▁ Will",
+      "▁{ }",
+      "▁ {}",
+      "name s",
+      "na mes",
+      "nam es",
+      "n ames",
+      "▁G rand",
+      "▁Gr and",
+      "▁Gra nd",
+      "▁Gran d",
+      "▁ Grand",
+      "流 程",
+      "▁sc hedul",
+      "gorith ms",
+      "gorithm s",
+      "▁al cohol",
+      "ak ed",
+      "ake d",
+      "end if",
+      "▁re views",
+      "▁review s",
+      "▁ reviews",
+      "ut ed",
+      "ute d",
+      "u ted",
+      "才 是",
+      "Func tion",
+      "Fun ction",
+      "F unction",
+      "上诉 于",
+      "▁R ussian",
+      "▁Russ ian",
+      "▁Russia n",
+      "▁ Russian",
+      "ac ks",
+      "ack s",
+      "a cks",
+      "各 类",
+      "▁D O",
+      "▁ DO",
+      "…… \\",
+      "… …\\",
+      "▁I nf",
+      "▁In f",
+      "▁ Inf",
+      "▁de lete",
+      "▁del ete",
+      "▁delet e",
+      "▁ delete",
+      "Log ger",
+      "就 像",
+      "an ning",
+      "ann ing",
+      "anni ng",
+      "io logical",
+      "iol ogical",
+      "iologic al",
+      "i ological",
+      "最 新",
+      "Ap plication",
+      "▁re ached",
+      "▁reach ed",
+      "▁ident ity",
+      "▁id entity",
+      "▁ identity",
+      "raz il",
+      "系统 的",
+      "il ed",
+      "ile d",
+      "i led",
+      "▁c alculate",
+      "▁calcul ate",
+      "▁calc ulate",
+      "▁ calculate",
+      "很 难",
+      "AG E",
+      "A GE",
+      "管理 有限公司",
+      "过 的",
+      "date d",
+      "da ted",
+      "dat ed",
+      "d ated",
+      "o i",
+      "辽 宁",
+      "▁M ake",
+      "▁Ma ke",
+      "▁Mak e",
+      "▁ Make",
+      "▁t end",
+      "▁te nd",
+      "▁ten d",
+      "▁ tend",
+      "变 得",
+      "志 愿",
+      "信 号",
+      "▁Ap ple",
+      "▁App le",
+      "▁Appl e",
+      "▁ Apple",
+      "对方 当事人",
+      "op ath",
+      "opa th",
+      "o path",
+      "▁M A",
+      "▁ MA",
+      "▁te aching",
+      "▁teach ing",
+      "▁tea ching",
+      "▁vir tual",
+      "▁virt ual",
+      "▁ virtual",
+      "▁him self",
+      "p a",
+      "el ement",
+      "ele ment",
+      "elem ent",
+      "e lement",
+      "st op",
+      "sto p",
+      "s top",
+      "la m",
+      "l am",
+      "▁m ission",
+      "▁miss ion",
+      "▁ mission",
+      "un ately",
+      "unate ly",
+      "变 成",
+      "sh op",
+      "s hop",
+      "▁struct ural",
+      "▁ structural",
+      "▁p al",
+      "▁pa l",
+      "▁ pal",
+      "onym ous",
+      "离 开",
+      "▁O N",
+      "▁ ON",
+      "哪 里",
+      "▁happ ens",
+      "▁happen s",
+      "▁E nd",
+      "▁En d",
+      "▁ End",
+      "▁re lation",
+      "▁rel ation",
+      "▁ relation",
+      "最 佳",
+      "▁cont ribute",
+      "▁contribut e",
+      "▁fac ilities",
+      "▁facilit ies",
+      "▁sec ure",
+      "▁ secure",
+      "▁hot el",
+      "▁ho tel",
+      "L S",
+      "▁se rum",
+      "▁ser um",
+      "▁represent ed",
+      "▁ represented",
+      "大 陆",
+      "IN E",
+      "I NE",
+      "结 案",
+      "ni s",
+      "n is",
+      "扩 大",
+      "▁d iss",
+      "▁dis s",
+      "▁di ss",
+      "▁ diss",
+      "er nel",
+      "ern el",
+      "思 维",
+      "品 种",
+      "▁f lag",
+      "▁fl ag",
+      "▁ flag",
+      "ful l",
+      "fu ll",
+      "f ull",
+      "违 反",
+      "扣 押",
+      "▁build ings",
+      "▁building s",
+      "Sp ec",
+      "Spe c",
+      "S pec",
+      "C o",
+      "other apy",
+      "otherap y",
+      "o therapy",
+      "时 的",
+      "大 型",
+      "hel ial",
+      "▁U SA",
+      "▁US A",
+      "▁ USA",
+      "▁det ected",
+      "▁detect ed",
+      "ta ge",
+      "tag e",
+      "t age",
+      "▁J ournal",
+      "▁Jo urnal",
+      "▁Jour nal",
+      "▁ Journal",
+      "等 于",
+      "▁b ow",
+      "▁bo w",
+      "▁ bow",
+      "▁numer ous",
+      "▁th row",
+      "▁thr ow",
+      "▁thro w",
+      "▁ throw",
+      "本 科",
+      "进 步",
+      "▁rel atively",
+      "▁relative ly",
+      "很 大",
+      "pad ding",
+      "p adding",
+      "▁[ ]",
+      "▁ []",
+      "}$ $\\",
+      "}$$ \\",
+      "} $$\\",
+      "ul ating",
+      "ula ting",
+      "还 能",
+      "二〇一 四年",
+      "▁second ary",
+      "▁ secondary",
+      "▁f lat",
+      "▁fl at",
+      "▁ flat",
+      "经营 范围",
+      "身 边",
+      "策 略",
+      "▁disc ussion",
+      "▁discuss ion",
+      "ip al",
+      "ipa l",
+      "i pal",
+      "mod e",
+      "mo de",
+      "m ode",
+      "ex it",
+      "e xit",
+      "tr ee",
+      "tre e",
+      "t ree",
+      "▁s uitable",
+      "▁su itable",
+      "▁suit able",
+      "ad er",
+      "ade r",
+      "a der",
+      "集 体",
+      "▁to ward",
+      "▁tow ard",
+      "依 然",
+      "▁gover ning",
+      "▁govern ing",
+      "res hold",
+      "resh old",
+      "▁h omes",
+      "▁home s",
+      "▁hom es",
+      "▁ho mes",
+      "山 区",
+      "GE T",
+      "G ET",
+      "盗 窃",
+      "院 校",
+      "往 往",
+      "▁S tand",
+      "▁St and",
+      "▁Stan d",
+      "▁Sta nd",
+      "▁ Stand",
+      "un ched",
+      "unc hed",
+      "unch ed",
+      "▁mod ify",
+      "▁ modify",
+      "▁equ ivalent",
+      "▁equival ent",
+      "▁ equivalent",
+      "财 政",
+      "▁Vir gin",
+      "容 量",
+      "规 律",
+      "T V",
+      "CI TE",
+      "C ITE",
+      "原 来",
+      "io logy",
+      "iol ogy",
+      "i ology",
+      "▁where as",
+      "▁whe reas",
+      "▁p ow",
+      "▁po w",
+      "▁ pow",
+      "g t",
+      "上述 事实",
+      "or al",
+      "ora l",
+      "o ral",
+      "并 按",
+      "is tan",
+      "ist an",
+      "ista n",
+      "i stan",
+      "_ \\",
+      "▁o k",
+      "▁ ok",
+      "▁exper iences",
+      "▁experience s",
+      "▁experi ences",
+      "ra s",
+      "r as",
+      "▁occ up",
+      "▁oc cup",
+      "▁ occup",
+      "名 字",
+      "父 亲",
+      "astr ucture",
+      "astruct ure",
+      "a structure",
+      "▁s chedule",
+      "▁sc hedule",
+      "▁schedul e",
+      "▁ schedule",
+      "的 前",
+      "▁E ven",
+      "▁Ev en",
+      "▁Eve n",
+      "▁ Even",
+      "u v",
+      "pe ople",
+      "p eople",
+      "port ed",
+      "por ted",
+      "p orted",
+      "▁part ial",
+      "▁ partial",
+      "眼 睛",
+      "屏 幕",
+      "▁g ift",
+      "▁gi ft",
+      "物 理",
+      "▁art ists",
+      "▁artist s",
+      "▁Te am",
+      "▁Tea m",
+      "▁ Team",
+      "▁col ors",
+      "▁color s",
+      "▁ colors",
+      "▁S tar",
+      "▁St ar",
+      "▁Sta r",
+      "▁ Star",
+      "Ser ial",
+      "Se rial",
+      "S erial",
+      "减 去",
+      "per son",
+      "pers on",
+      "p erson",
+      "id ae",
+      "ida e",
+      "▁E rror",
+      "▁Er ror",
+      "▁Err or",
+      "▁ Error",
+      "ic ing",
+      "ici ng",
+      "icin g",
+      "i cing",
+      "st yles",
+      "style s",
+      "sty les",
+      "styl es",
+      "> ,",
+      "▁inter vention",
+      "▁interven tion",
+      "天 下",
+      "ac ts",
+      "act s",
+      "a cts",
+      "W h",
+      "EN D",
+      "E ND",
+      "病 毒",
+      "▁M el",
+      "▁Me l",
+      "▁ Mel",
+      "▁sugg ests",
+      "▁suggest s",
+      "电 源",
+      "▁a head",
+      "▁ah ead",
+      "▁ ahead",
+      "▁qu arter",
+      "▁quart er",
+      "▁ quarter",
+      "▁L ear",
+      "▁Le ar",
+      "▁s hift",
+      "▁sh ift",
+      "▁ shift",
+      "ef ined",
+      "k l",
+      "▁H ill",
+      "▁Hi ll",
+      "▁Hil l",
+      "日向本院 提出撤诉申请",
+      "日向本院提出 撤诉申请",
+      "日 向本院提出撤诉申请",
+      "▁help ed",
+      "▁hel ped",
+      "▁m al",
+      "▁ma l",
+      "▁ mal",
+      "好 像",
+      "▁C ard",
+      "▁Car d",
+      "▁Ca rd",
+      "▁ Card",
+      "先 后",
+      "地 理",
+      "中 共",
+      "Window s",
+      "Wind ows",
+      "W indows",
+      "处 于",
+      "的 生活",
+      "▁admin istr",
+      "作 业",
+      "▁d ys",
+      "▁dy s",
+      "▁mention ed",
+      "▁ment ioned",
+      "▁ mentioned",
+      "未 按",
+      "▁monitor ing",
+      "▁ monitoring",
+      "终 于",
+      "不 知",
+      "▁C amp",
+      "▁Cam p",
+      "▁Ca mp",
+      "▁ Camp",
+      "模 拟",
+      "ud io",
+      "udi o",
+      "\": \\",
+      "\" :\\",
+      "▁w ear",
+      "▁we ar",
+      "▁ wear",
+      "zi p",
+      "z ip",
+      "▁stand ards",
+      "▁standard s",
+      "▁sen ior",
+      "▁Z eal",
+      "▁Ze al",
+      "og ene",
+      "ogen e",
+      "oge ne",
+      "o gene",
+      "T I",
+      "id ential",
+      "ident ial",
+      "ap plication",
+      "▁ext ended",
+      "▁extend ed",
+      "▁ extended",
+      "te sts",
+      "test s",
+      "tes ts",
+      "t ests",
+      "是 由",
+      "or gan",
+      "org an",
+      "▁sub stant",
+      "▁subst ant",
+      "▁m ental",
+      "▁men tal",
+      "▁ment al",
+      "▁ mental",
+      "▁连载 状态",
+      "U B",
+      "处罚 金",
+      "处 罚金",
+      "规 则",
+      "▁help ing",
+      "▁hel ping",
+      "▁h aven",
+      "▁have n",
+      "▁ha ven",
+      "▁hav en",
+      "▁ haven",
+      "女 生",
+      "gr id",
+      "g rid",
+      "在 这",
+      "▁cent re",
+      "▁centr e",
+      "ur ban",
+      "urb an",
+      "u rban",
+      "▁m agnetic",
+      "▁mag netic",
+      "▁magn etic",
+      "▁magnet ic",
+      "Of f",
+      "O ff",
+      "ur i",
+      "u ri",
+      "sc ope",
+      ") ]",
+      "▁H istor",
+      "▁His tor",
+      "▁Hi stor",
+      "▁Hist or",
+      "▁o bj",
+      "▁ob j",
+      "▁ obj",
+      "▁b eg",
+      "▁be g",
+      "▁ beg",
+      "▁r oles",
+      "▁ro les",
+      "▁role s",
+      "▁ roles",
+      "的 心",
+      "nA cc",
+      "nAc c",
+      "n Acc",
+      "高 中",
+      "et ry",
+      "etr y",
+      "e try",
+      "下面 的",
+      "下 面的",
+      "Ba r",
+      "B ar",
+      "▁abs olute",
+      "▁absol ute",
+      "▁ absolute",
+      "▁C alculate",
+      "▁Calcul ate",
+      "▁Calc ulate",
+      "▁ Calculate",
+      "e q",
+      "存 储",
+      "he t",
+      "h et",
+      "▁St art",
+      "▁Star t",
+      "▁Sta rt",
+      "▁ Start",
+      "可能 会",
+      "情 绪",
+      "is l",
+      "i sl",
+      "▁It alian",
+      "▁Ital ian",
+      "定 位",
+      "re source",
+      "res ource",
+      "） <",
+      "xy gen",
+      "▁t asks",
+      "▁task s",
+      "▁ tasks",
+      "ick ets",
+      "icket s",
+      "z h",
+      "▁G ame",
+      "▁Ga me",
+      "▁Gam e",
+      "▁ Game",
+      "dep endent",
+      "depend ent",
+      "d ependent",
+      "aw s",
+      "a ws",
+      "Out put",
+      "ic ago",
+      "ica go",
+      "▁l oop",
+      "▁lo op",
+      "▁ loop",
+      "ri ch",
+      "ric h",
+      "r ich",
+      "Me t",
+      "M et",
+      "看 了",
+      "nO r",
+      "n Or",
+      "▁Re view",
+      "▁Rev iew",
+      "▁ Review",
+      "▁n otes",
+      "▁not es",
+      "▁no tes",
+      "▁note s",
+      "▁ notes",
+      "igr ations",
+      "igration s",
+      "▁ra ised",
+      "▁raise d",
+      "▁And roid",
+      "▁Andr oid",
+      "▁ Android",
+      "在 于",
+      "动 作",
+      "▁main ly",
+      "ac count",
+      "acc ount",
+      "acco unt",
+      "▁K IND",
+      "S QL",
+      "美 丽",
+      "▁E ducation",
+      "▁Edu cation",
+      "▁Educ ation",
+      "▁di vision",
+      "▁div ision",
+      "▁ division",
+      "有 所",
+      "▁f ans",
+      "▁fa ns",
+      "▁fan s",
+      "▁re place",
+      "▁replac e",
+      "▁ replace",
+      "就 要",
+      "▁F ire",
+      "▁Fi re",
+      "▁ Fire",
+      "um in",
+      "umi n",
+      "u min",
+      "▁I P",
+      "▁ IP",
+      "▁e dit",
+      "▁ed it",
+      "▁ edit",
+      "print f",
+      "tm p",
+      "t mp",
+      "com pl",
+      "comp l",
+      "Pro ps",
+      "Pr ops",
+      "Prop s",
+      "▁res idents",
+      "▁resid ents",
+      "▁resident s",
+      "Par ser",
+      "Parse r",
+      "Pars er",
+      "公 告",
+      "▁B ack",
+      "▁Ba ck",
+      "▁Bac k",
+      "▁ Back",
+      "▁Zeal and",
+      "R C",
+      "▁vel ocity",
+      "▁veloc ity",
+      "▁ velocity",
+      "▁s pring",
+      "▁sp ring",
+      "▁spr ing",
+      "▁ spring",
+      "长 度",
+      "az z",
+      "a zz",
+      "步 骤",
+      "财 务",
+      "▁l aws",
+      "▁law s",
+      "▁la ws",
+      "▁ laws",
+      "探 索",
+      "▁Ex ception",
+      "▁Except ion",
+      "▁ Exception",
+      "▁成 立",
+      "▁ 成立",
+      "▁T witter",
+      "▁Tw itter",
+      "▁ Twitter",
+      "▁a ren",
+      "▁are n",
+      "▁ar en",
+      "▁ aren",
+      "业 有限公司",
+      "传 唤",
+      "pl ates",
+      "plate s",
+      "plat es",
+      "ur ies",
+      "uri es",
+      "u ries",
+      "p o",
+      "明 白",
+      "儿 子",
+      "人 家",
+      "▁dis orders",
+      "▁disorder s",
+      "▁Ac ademy",
+      "▁Academ y",
+      "▁Acad emy",
+      "hand le",
+      "win dow",
+      "wind ow",
+      "w indow",
+      "k r",
+      "▁b uff",
+      "▁bu ff",
+      "▁buf f",
+      "▁ buff",
+      "\\' ,\\'",
+      "\\', \\'",
+      "\\',\\ '",
+      "湖北 省",
+      "湖 北省",
+      "▁h el",
+      "▁he l",
+      "▁ hel",
+      "▁c old",
+      "▁col d",
+      "▁co ld",
+      "▁sign aling",
+      "▁signal ing",
+      "龙 江",
+      "迅 速",
+      "▁four th",
+      "日 被",
+      "▁s and",
+      "▁sa nd",
+      "▁san d",
+      "▁ sand",
+      "de st",
+      "des t",
+      "d est",
+      "▁per spect",
+      "▁pers pect",
+      "ogn itive",
+      "ug ins",
+      "ugin s",
+      "▁s exual",
+      "▁sex ual",
+      "▁ sexual",
+      "ar rant",
+      "arr ant",
+      "arra nt",
+      "Re m",
+      "R em",
+      "▁ox id",
+      "▁ oxid",
+      "▁meas urement",
+      "▁measure ment",
+      "' '",
+      "十 年",
+      "ig an",
+      "iga n",
+      "i gan",
+      "成 就",
+      "▁I r",
+      "▁ Ir",
+      "is hes",
+      "ish es",
+      "M D",
+      "ym bol",
+      "▁L im",
+      "▁Li m",
+      "_{ \\\\",
+      "_ {\\\\",
+      "特 性",
+      "深圳 市",
+      "re ens",
+      "ree ns",
+      "reen s",
+      "线 的",
+      "ff ee",
+      "f fee",
+      "▁c ryst",
+      "▁cr yst",
+      "▁cry st",
+      "Bl ock",
+      "B lock",
+      "li fe",
+      "lif e",
+      "l ife",
+      "nS he",
+      "nSh e",
+      "n She",
+      "Cont ainer",
+      "▁c ards",
+      "▁car ds",
+      "▁card s",
+      "▁ cards",
+      "轻 松",
+      "完 毕",
+      "▁prom ote",
+      "▁promot e",
+      "▁promo te",
+      "▁ag ree",
+      "▁agre e",
+      "▁agr ee",
+      "本 人",
+      "▁M ax",
+      "▁Ma x",
+      "▁ Max",
+      "▁most ly",
+      "▁ mostly",
+      "▁r ich",
+      "▁ri ch",
+      "▁ric h",
+      "▁ rich",
+      "▁isol ated",
+      "▁iso lated",
+      "▁isolate d",
+      "▁t un",
+      "▁tu n",
+      "▁ tun",
+      "在中 国",
+      "在 中国",
+      "ost ream",
+      "o stream",
+      "▁L ife",
+      "▁Li fe",
+      "▁Lif e",
+      "▁ Life",
+      "一 家",
+      "ER S",
+      "E RS",
+      "io t",
+      "i ot",
+      "LE D",
+      "L ED",
+      "体 现",
+      "日 作出",
+      "奖 励",
+      "om er",
+      "ome r",
+      "o mer",
+      "▁b aby",
+      "▁bab y",
+      "▁ba by",
+      "▁ baby",
+      "▁f ashion",
+      "▁ fashion",
+      "判处 有期徒刑",
+      "▁st arts",
+      "▁start s",
+      "▁star ts",
+      "▁ starts",
+      "▁R oyal",
+      "▁Ro yal",
+      "▁Roy al",
+      "每 年",
+      "▁we ak",
+      "▁ weak",
+      "原告 的",
+      "▁techn ologies",
+      "积极 参加",
+      "▁i on",
+      "▁io n",
+      "▁ ion",
+      "▁F unction",
+      "▁Fun ction",
+      "▁Func tion",
+      "▁ Function",
+      "▁C A",
+      "▁ CA",
+      "强 的",
+      "▁h us",
+      "▁hu s",
+      "▁ hus",
+      "▁g ly",
+      "▁gl y",
+      "▁ gly",
+      "▁c art",
+      "▁car t",
+      "▁ca rt",
+      "▁ cart",
+      "om es",
+      "ome s",
+      "o mes",
+      "▁s ea",
+      "▁se a",
+      "▁ sea",
+      "▁scient ific",
+      "客 房",
+      "▁engine ering",
+      "▁engineer ing",
+      "▁ engineering",
+      "向本院 递交上诉状",
+      "▁m el",
+      "▁me l",
+      "▁ mel",
+      "▁Direct or",
+      "▁Dir ector",
+      "], [",
+      "] ,[",
+      "ip her",
+      "iph er",
+      "i pher",
+      "om al",
+      "oma l",
+      "o mal",
+      "▁j obs",
+      "▁job s",
+      "▁jo bs",
+      "▁ jobs",
+      "▁mod ified",
+      "▁ modified",
+      "em pty",
+      "empt y",
+      "emp ty",
+      "▁t ub",
+      "▁tu b",
+      "▁ tub",
+      "▁high light",
+      "▁ highlight",
+      "H S",
+      "▁l ocations",
+      "▁loc ations",
+      "▁location s",
+      "▁ locations",
+      "Re turn",
+      "Ret urn",
+      "R eturn",
+      "▁m ist",
+      "▁mis t",
+      "▁mi st",
+      "命 令",
+      "as ync",
+      "asy nc",
+      "a sync",
+      "句 话",
+      "og enic",
+      "ogen ic",
+      "oge nic",
+      "o genic",
+      "▁re covery",
+      "▁rec overy",
+      "▁recover y",
+      "▁c red",
+      "▁cre d",
+      "▁cr ed",
+      "▁ cred",
+      "网 友",
+      "bu r",
+      "b ur",
+      "▁b rief",
+      "▁br ief",
+      "▁ brief",
+      "for ms",
+      "form s",
+      "f orms",
+      "▁M ap",
+      "▁Ma p",
+      "▁ Map",
+      "Op er",
+      "O per",
+      "▁p icked",
+      "▁pick ed",
+      "ar ia",
+      "ari a",
+      "a ria",
+      "Add ress",
+      "Addr ess",
+      "bin d",
+      "bi nd",
+      "b ind",
+      "▁sc ene",
+      "▁scen e",
+      "▁ scene",
+      "▁S ign",
+      "▁Si gn",
+      "▁Sig n",
+      "▁ Sign",
+      "是 我",
+      "▁exp erts",
+      "▁exper ts",
+      "▁expert s",
+      "各 个",
+      "▁书 名",
+      "▁ 书名",
+      "▁cons istent",
+      "▁consist ent",
+      "▁pr actical",
+      "▁pract ical",
+      "▁{ \"",
+      "▁ {\"",
+      "pp ed",
+      "p ped",
+      "文 献",
+      "ad s",
+      "a ds",
+      "▁In formation",
+      "▁Inform ation",
+      "▁ Information",
+      "▁C I",
+      "▁ CI",
+      "em ia",
+      "emi a",
+      "e mia",
+      "an notation",
+      "ann otation",
+      "annot ation",
+      "执行 过程中",
+      "▁Wh ich",
+      "▁ Which",
+      "▁o pin",
+      "▁op in",
+      "第一百 零",
+      "▁Mar ket",
+      "▁Mark et",
+      "▁ Market",
+      "▁cons ists",
+      "▁consist s",
+      "con s",
+      "co ns",
+      "c ons",
+      "dom ain",
+      "do main",
+      "d omain",
+      "▁s hip",
+      "▁sh ip",
+      "▁ ship",
+      "▁l ock",
+      "▁loc k",
+      "▁lo ck",
+      "▁ lock",
+      "ce an",
+      "cea n",
+      "c ean",
+      "▁J ava",
+      "▁Ja va",
+      "▁Jav a",
+      "▁ Java",
+      "理 想",
+      "▁d a",
+      "▁ da",
+      "inter face",
+      ")* (",
+      ") *(",
+      "请 你",
+      "▁M ot",
+      "▁Mo t",
+      "韩 国",
+      "记 得",
+      "▁st ations",
+      "▁stat ions",
+      "▁station s",
+      "▁ stations",
+      "▁document s",
+      "▁doc uments",
+      "▁ documents",
+      "三角 形",
+      "um a",
+      "u ma",
+      "也没 有",
+      "也 没有",
+      "li ament",
+      "lia ment",
+      "▁C are",
+      "▁Car e",
+      "▁Ca re",
+      "▁ Care",
+      "▁Ch ristian",
+      "▁Christ ian",
+      "▁ Christian",
+      "▁m iles",
+      "▁mil es",
+      "▁mi les",
+      "▁mile s",
+      "▁t or",
+      "▁to r",
+      "▁ tor",
+      "几 年",
+      "▁re lig",
+      "▁rel ig",
+      "Ch ange",
+      "Chan ge",
+      "▁Im age",
+      "▁ Image",
+      "Ma t",
+      "M at",
+      "▁l ab",
+      "▁la b",
+      "▁ lab",
+      "▁o dd",
+      "▁od d",
+      "▁ odd",
+      "▁M o",
+      "▁ Mo",
+      "▁s usp",
+      "▁su sp",
+      "▁sus p",
+      "▁pro cedures",
+      "▁proced ures",
+      "▁procedure s",
+      "队 伍",
+      "en dar",
+      "end ar",
+      "enda r",
+      "不 予",
+      "▁acc urate",
+      "▁accur ate",
+      "▁M essage",
+      "▁Mess age",
+      "▁ Message",
+      "保 存",
+      "居 住",
+      "▁par allel",
+      "▁paralle l",
+      "▁ parallel",
+      "aw ait",
+      "awa it",
+      "awai t",
+      "a wait",
+      "ho t",
+      "h ot",
+      "▁co verage",
+      "▁cover age",
+      "▁ coverage",
+      "▁cor relation",
+      "▁correl ation",
+      "▁ correlation",
+      "满 意",
+      "pl ayer",
+      "play er",
+      "p layer",
+      "D P",
+      "en ted",
+      "ent ed",
+      "ente d",
+      "su ccess",
+      "s uccess",
+      "共 有",
+      "mu lt",
+      "mul t",
+      "m ult",
+      "▁rem aining",
+      "▁remain ing",
+      "▁ remaining",
+      "▁certain ly",
+      "▁Ad min",
+      "▁ Admin",
+      "中 间",
+      "▁Min ister",
+      "▁Mini ster",
+      "的 水",
+      "▁p lane",
+      "▁pl ane",
+      "▁plan e",
+      "▁ plane",
+      "▁s umm",
+      "▁su mm",
+      "▁sum m",
+      "▁ summ",
+      "▁s omew",
+      "▁some w",
+      "▁som ew",
+      "▁phot ograph",
+      "▁photo graph",
+      "ev en",
+      "eve n",
+      "e ven",
+      "大 利",
+      "id x",
+      "i dx",
+      "▁P lay",
+      "▁Pl ay",
+      "▁ Play",
+      "▁F eb",
+      "▁Fe b",
+      "▁ Feb",
+      "foo t",
+      "fo ot",
+      "f oot",
+      "ad r",
+      "a dr",
+      "公 众",
+      "oh n",
+      "o hn",
+      "nR es",
+      "nRe s",
+      "n Res",
+      "▁s ynthesis",
+      "▁synt hesis",
+      "▁synthes is",
+      "▁synth esis",
+      "召 开",
+      "▁sim ulation",
+      "▁ simulation",
+      "▁be coming",
+      "▁bec oming",
+      "es tic",
+      "est ic",
+      "▁m enu",
+      "▁me nu",
+      "▁men u",
+      "▁ menu",
+      "ig ital",
+      "igi tal",
+      "机 的",
+      "▁c ort",
+      "▁co rt",
+      "▁cor t",
+      "▁piece s",
+      "▁pi eces",
+      "▁pie ces",
+      "▁ pieces",
+      "▁b iological",
+      "▁bi ological",
+      "▁bio logical",
+      "▁biologic al",
+      "▁ biological",
+      "▁w elcome",
+      "▁wel come",
+      "▁ welcome",
+      "▁des pite",
+      "he ther",
+      "het her",
+      "h ether",
+      "Pro cess",
+      "Proc ess",
+      "简易 程序",
+      "▁F re",
+      "▁Fr e",
+      "▁ Fre",
+      "▁P RO",
+      "▁PR O",
+      "▁ PRO",
+      "▁S mith",
+      "▁Sm ith",
+      "▁ Smith",
+      "] ;",
+      "▁p or",
+      "▁po r",
+      "▁ por",
+      "▁v oice",
+      "▁vo ice",
+      "▁ voice",
+      "诊 断",
+      "▁e sc",
+      "▁es c",
+      "▁ esc",
+      "情 形",
+      "维 持",
+      "下 去",
+      "▁p anel",
+      "▁pan el",
+      "▁pa nel",
+      "▁pane l",
+      "▁ panel",
+      "▁N ote",
+      "▁No te",
+      "▁Not e",
+      "▁ Note",
+      "La yout",
+      "L ayout",
+      "▁integr ated",
+      "▁integrate d",
+      "▁l ots",
+      "▁lot s",
+      "▁lo ts",
+      "▁L ib",
+      "▁Li b",
+      "▁ Lib",
+      "▁pro port",
+      "▁prop ort",
+      "▁for ces",
+      "▁force s",
+      "▁ forces",
+      "母 亲",
+      "[ \\",
+      "▁im plemented",
+      "▁implement ed",
+      "ap pro",
+      "app ro",
+      "▁link ed",
+      "▁ linked",
+      "▁occ urs",
+      "▁occur s",
+      "ach ine",
+      "achi ne",
+      "慢 慢",
+      "尤其 是",
+      "on ald",
+      "onal d",
+      "ona ld",
+      "▁cont rolled",
+      "▁control led",
+      "▁ controlled",
+      "▁v ary",
+      "▁var y",
+      "▁va ry",
+      "ur er",
+      "ure r",
+      "u rer",
+      "G L",
+      "相应 的",
+      "相 应的",
+      "▁: \\",
+      "▁ :\\",
+      "▁f elt",
+      "▁fe lt",
+      "▁fel t",
+      "▁ felt",
+      "wa ter",
+      "wat er",
+      "w ater",
+      "▁F inally",
+      "▁Fin ally",
+      "▁Final ly",
+      "hemat ical",
+      "hema tical",
+      "hematic al",
+      "st rap",
+      "str ap",
+      "stra p",
+      "S o",
+      "▁great est",
+      "▁gre atest",
+      "pl icated",
+      "plic ated",
+      "plicate d",
+      "plica ted",
+      "Col umn",
+      "Colum n",
+      "ic acy",
+      "ica cy",
+      "智 慧",
+      "时 尚",
+      "自 愿",
+      "▁Reg ister",
+      "▁ Register",
+      "主 张",
+      "▁version s",
+      "▁vers ions",
+      "▁ versions",
+      "▁m atches",
+      "▁mat ches",
+      "▁match es",
+      "▁ matches",
+      "▁sh aring",
+      "▁sha ring",
+      "▁shar ing",
+      "▁ sharing",
+      "▁L ove",
+      "▁Lo ve",
+      "▁Lov e",
+      "▁ Love",
+      "青 春",
+      "IL ITY",
+      "心 里",
+      "ki t",
+      "k it",
+      "▁D en",
+      "▁De n",
+      "▁ Den",
+      "技 巧",
+      "▁man if",
+      "▁ma nif",
+      "▁ manif",
+      "▁B ill",
+      "▁Bi ll",
+      "▁Bil l",
+      "▁ Bill",
+      "▁ar rest",
+      "▁arr est",
+      "The re",
+      "Th ere",
+      "T here",
+      "四 个",
+      "▁abs or",
+      "▁min im",
+      "▁mini m",
+      "▁mi nim",
+      "an da",
+      "and a",
+      "a nda",
+      "Vi s",
+      "V is",
+      "▁f ederal",
+      "▁fed eral",
+      "▁under lying",
+      "的 执行",
+      "▁Com mission",
+      "▁Comm ission",
+      "sc opy",
+      "s copy",
+      "Aut h",
+      "Au th",
+      "A uth",
+      "arg v",
+      "▁opt imal",
+      "▁optim al",
+      "▁ optimal",
+      "▁fore ign",
+      "▁ foreign",
+      "HE R",
+      "H ER",
+      "▁Is ra",
+      "▁b inary",
+      "▁bin ary",
+      "▁ binary",
+      "而 言",
+      "▁le ague",
+      "▁ league",
+      "签 订",
+      "▁M er",
+      "▁Me r",
+      "▁ Mer",
+      "的情况 下",
+      "的 情况下",
+      "突 出",
+      "em porary",
+      "empor ary",
+      "▁occ urred",
+      "▁occur red",
+      "▁occurre d",
+      "女 人",
+      "利 润",
+      "停 止",
+      "类 似",
+      "▁man aged",
+      "▁manag ed",
+      "▁manage d",
+      "▁ managed",
+      "制 定",
+      "▁al umni",
+      "▁con vers",
+      "▁conver s",
+      "▁conv ers",
+      "▁conve rs",
+      "f riend",
+      "▁Commit tee",
+      "未 履行",
+      "v m",
+      "法 国",
+      "▁me ant",
+      "▁mean t",
+      "Or der",
+      "Ord er",
+      "岗 位",
+      "▁Or der",
+      "▁Ord er",
+      "▁ Order",
+      "监 管",
+      "▁M il",
+      "▁Mi l",
+      "▁ Mil",
+      "ha v",
+      "h av",
+      "更 是",
+      "解 得",
+      "▁arch itecture",
+      "▁architect ure",
+      "▁ architecture",
+      "▁se gment",
+      "▁seg ment",
+      "▁ segment",
+      "iat ric",
+      "的事 实",
+      "的 事实",
+      "G o",
+      "▁st ability",
+      "▁stabil ity",
+      "▁stab ility",
+      "▁i con",
+      "▁ic on",
+      "▁ icon",
+      "cl uded",
+      "clud ed",
+      "clude d",
+      "▁G al",
+      "▁Ga l",
+      "▁ Gal",
+      "▁all owing",
+      "▁allow ing",
+      "I E",
+      "放 入",
+      "▁pro sp",
+      "▁pr osp",
+      "▁pros p",
+      "bo und",
+      "b ound",
+      "企业 的",
+      "企 业的",
+      "等 证据",
+      "sh ape",
+      "sha pe",
+      "H ub",
+      "科学 院",
+      "科 学院",
+      "▁成立 时间",
+      "地 位",
+      "dimension al",
+      "d imensional",
+      ": (",
+      "Form at",
+      "For mat",
+      "▁cl early",
+      "▁cle arly",
+      "▁clear ly",
+      "M e",
+      "▁di git",
+      "▁dig it",
+      "▁ digit",
+      "do t",
+      "d ot",
+      "np m",
+      "n pm",
+      "▁fe eling",
+      "▁feel ing",
+      "▁fee ling",
+      "流 量",
+      "查 看",
+      "空 调",
+      "Se arch",
+      "S earch",
+      "Ar g",
+      "A rg",
+      "等 级",
+      "▁e state",
+      "▁est ate",
+      "▁esta te",
+      "▁ ·",
+      "被 上诉人",
+      "▁conf idence",
+      "▁ confidence",
+      "给 予",
+      "突 破",
+      "N T",
+      "还 款",
+      "R ight",
+      "▁W ater",
+      "▁Wat er",
+      "▁Wa ter",
+      "▁ Water",
+      "得 出",
+      "记 忆",
+      "B r",
+      "▁gu ys",
+      "▁guy s",
+      "AB LE",
+      "A BLE",
+      "强 调",
+      "▁e mph",
+      "▁em ph",
+      "▁emp h",
+      "▁ emph",
+      "ut ter",
+      "utt er",
+      "ax is",
+      "▁C ore",
+      "▁Co re",
+      "▁Cor e",
+      "▁ Core",
+      "▁f ant",
+      "▁fa nt",
+      "▁fan t",
+      "▁g luc",
+      "▁gl uc",
+      "▁glu c",
+      "▁ gluc",
+      "▁v acc",
+      "▁vac c",
+      "▁va cc",
+      "▁f unding",
+      "▁fund ing",
+      "▁fun ding",
+      "ct x",
+      "c tx",
+      "▁est imate",
+      "▁estim ate",
+      "▁ estimate",
+      "H A",
+      "▁slight ly",
+      "▁p arse",
+      "▁par se",
+      "▁pars e",
+      "▁ parse",
+      "▁th rows",
+      "▁throw s",
+      "▁thr ows",
+      "▁thro ws",
+      "▁ throws",
+      "日 在",
+      "D L",
+      "▁per haps",
+      "显 着",
+      "人 数",
+      "复 议",
+      "pos ite",
+      "po site",
+      "▁pos ts",
+      "▁post s",
+      "▁po sts",
+      "▁ posts",
+      "▁program ming",
+      "▁ programming",
+      "Writ e",
+      "Wr ite",
+      "W rite",
+      "pro file",
+      "prof ile",
+      "ig er",
+      "ige r",
+      "i ger",
+      "ic ious",
+      "ici ous",
+      "icio us",
+      "St ore",
+      "成 熟",
+      "▁typ ically",
+      "▁typical ly",
+      "▁b ill",
+      "▁bi ll",
+      "▁bil l",
+      "▁ bill",
+      "▁log ging",
+      "▁ logging",
+      "la yout",
+      "lay out",
+      "l ayout",
+      "垃 圾",
+      "▁reg ional",
+      "▁region al",
+      "▁s ector",
+      "▁se ctor",
+      "▁sec tor",
+      "▁ sector",
+      "分 配",
+      "ion e",
+      "io ne",
+      "i one",
+      "词 条",
+      "装 备",
+      "学 历",
+      "cre ated",
+      "create d",
+      "creat ed",
+      "c reated",
+      "并处 罚金",
+      "并 处罚金",
+      "矛 盾",
+      ")] \\",
+      ") ]\\",
+      "▁W il",
+      "▁Wi l",
+      "▁ Wil",
+      "Inter face",
+      "▁elect ro",
+      "▁electr o",
+      "▁ electro",
+      "▁res erved",
+      "▁reserve d",
+      "▁reserv ed",
+      "▁ reserved",
+      "la te",
+      "lat e",
+      "l ate",
+      "▁t ele",
+      "▁te le",
+      "▁tel e",
+      "▁ tele",
+      "摄 影",
+      "▁s ensitivity",
+      "▁sens itivity",
+      "▁sensit ivity",
+      "观 众",
+      "▁L ouis",
+      "▁Lou is",
+      "▁Lo uis",
+      "▁ Louis",
+      "预 防",
+      "▁ac ademic",
+      "▁academ ic",
+      "▁F rank",
+      "▁Fr ank",
+      "▁Fra nk",
+      "▁ Frank",
+      "▁professional s",
+      "▁profession als",
+      "l t",
+      "完 整",
+      "▁C ustom",
+      "▁ Custom",
+      "v a",
+      "访 问",
+      "wr it",
+      "w rit",
+      "区 别",
+      "感 谢",
+      "ui nt",
+      "u int",
+      "view s",
+      "vi ews",
+      "vie ws",
+      "De v",
+      "D ev",
+      "▁u tf",
+      "▁ut f",
+      "▁ utf",
+      "▁re tail",
+      "▁ret ail",
+      "at ial",
+      "ati al",
+      "▁f ounded",
+      "▁found ed",
+      "▁ founded",
+      "▁a st",
+      "▁as t",
+      "▁ ast",
+      "Le ngth",
+      "L ength",
+      "uc ky",
+      "uck y",
+      "\"] \\",
+      "\" ]\\",
+      "十 七",
+      "衣 服",
+      "▁pay ment",
+      "▁ payment",
+      "▁wonder ful",
+      "on tal",
+      "ont al",
+      "▁R ock",
+      "▁Ro ck",
+      "▁Roc k",
+      "▁ Rock",
+      "La bel",
+      "Lab el",
+      "L abel",
+      "eg a",
+      "e ga",
+      "青 岛",
+      "做 的",
+      "▁synd rome",
+      "▁Franc is",
+      "使 得",
+      "▁mort ality",
+      "▁th ousands",
+      "▁thous ands",
+      "▁thousand s",
+      "angu ages",
+      "anguage s",
+      "▁W ood",
+      "▁Wo od",
+      "▁ Wood",
+      "▁m ur",
+      "▁mu r",
+      "▁ mur",
+      "AT A",
+      "A TA",
+      "长 的",
+      "▁d uration",
+      "▁du ration",
+      "▁dur ation",
+      "▁ duration",
+      "at ers",
+      "ate rs",
+      "ater s",
+      "a ters",
+      "ts elf",
+      "t self",
+      "Par t",
+      "Pa rt",
+      "P art",
+      "本题 考查",
+      "▁Further more",
+      "一 张",
+      "as tern",
+      "ast ern",
+      "aster n",
+      "aste rn",
+      "▁L and",
+      "▁La nd",
+      "▁Lan d",
+      "▁ Land",
+      "▁g uitar",
+      "▁gu itar",
+      "▁guit ar",
+      "▁gui tar",
+      "测 量",
+      "om orph",
+      "o morph",
+      "▁P oint",
+      "▁Po int",
+      "▁ Point",
+      "▁I ns",
+      "▁In s",
+      "▁ Ins",
+      "▁L icensed",
+      "▁License d",
+      "▁Lic ensed",
+      "▁ Licensed",
+      "▁n oise",
+      "▁no ise",
+      "▁ noise",
+      "pl t",
+      "p lt",
+      "▁Te chnology",
+      "▁Techn ology",
+      "充 满",
+      "ar ian",
+      "ari an",
+      "aria n",
+      "a rian",
+      "bo se",
+      "bos e",
+      "b ose",
+      "ad ow",
+      "ado w",
+      "a dow",
+      "▁v ote",
+      "▁vot e",
+      "▁vo te",
+      "▁ vote",
+      "ex ports",
+      "export s",
+      "exp orts",
+      "▁B AS",
+      "▁BA S",
+      "黑 龙江",
+      "ne ws",
+      "new s",
+      "n ews",
+      "▁d ifferential",
+      "▁differ ential",
+      "▁different ial",
+      "▁le aves",
+      "▁leave s",
+      "也 没",
+      "ug ar",
+      "uga r",
+      "u gar",
+      "um inate",
+      "umin ate",
+      "umina te",
+      "AP P",
+      "A PP",
+      "bit s",
+      "bi ts",
+      "b its",
+      "湖 南",
+      "M T",
+      "阳 光",
+      "▁cur ve",
+      "▁curv e",
+      "▁ curve",
+      "可能 是",
+      "▁c ars",
+      "▁car s",
+      "▁ca rs",
+      "▁ cars",
+      "出版的 图书",
+      "/* *",
+      "/ **",
+      "▁n oted",
+      "▁not ed",
+      "▁no ted",
+      "▁note d",
+      "el ess",
+      "ele ss",
+      "eles s",
+      "e less",
+      "▁str ugg",
+      "发现 被执行人有",
+      "发现被执行人 有",
+      "ition ally",
+      "itional ly",
+      "在服刑 期间",
+      "在 服刑期间",
+      "铁 路",
+      "▁concer ns",
+      "▁concern s",
+      "t w",
+      "不 再",
+      "sig ma",
+      "s igma",
+      "教 材",
+      "成 分",
+      "▁D E",
+      "▁ DE",
+      "nB y",
+      "n By",
+      "▁F ootball",
+      "▁Foot ball",
+      "展 开",
+      "Ma x",
+      "M ax",
+      "营 业",
+      "ch ron",
+      "chr on",
+      "出 去",
+      "周 期",
+      "▁com prehensive",
+      "▁comprehens ive",
+      "▁E nt",
+      "▁En t",
+      "▁ Ent",
+      "Se e",
+      "S ee",
+      "实 在",
+      "可供执行 财产",
+      "证 书",
+      "对 比",
+      "un ct",
+      "unc t",
+      "▁c ivil",
+      "▁ci vil",
+      "▁calcul ated",
+      "▁calc ulated",
+      "▁calculate d",
+      "▁cre ative",
+      "▁creat ive",
+      "▁ creative",
+      "R S",
+      "之 外",
+      "med ia",
+      "m edia",
+      "▁appe arance",
+      "▁appear ance",
+      "res sed",
+      "ress ed",
+      "r essed",
+      "引 导",
+      "▁Pro duct",
+      "▁Produ ct",
+      "▁ Product",
+      "B I",
+      "gr icult",
+      "对 称",
+      "A Y",
+      "▁Th omas",
+      "▁Thom as",
+      "▁ Thomas",
+      "▁w hom",
+      "▁wh om",
+      "▁who m",
+      "ad ing",
+      "adi ng",
+      "adin g",
+      "a ding",
+      "▁j av",
+      "▁ja v",
+      "▁ jav",
+      "▁s q",
+      "▁ sq",
+      "▁trans mission",
+      "成 的",
+      "来 到",
+      "du ction",
+      "duct ion",
+      "duc tion",
+      "d uction",
+      "小 组",
+      "▁re ct",
+      "▁r ect",
+      "▁rec t",
+      "▁ rect",
+      "▁s ociety",
+      "▁soci ety",
+      "▁l iquid",
+      "▁liqu id",
+      "▁ liquid",
+      "nWh ile",
+      "n While",
+      "▁int ended",
+      "▁intend ed",
+      "▁# #",
+      "▁ ##",
+      "He ader",
+      "Head er",
+      "H eader",
+      "▁O S",
+      "▁ OS",
+      "▁R eal",
+      "▁Re al",
+      "▁ Real",
+      "▁E nter",
+      "▁En ter",
+      "▁Ent er",
+      "▁ Enter",
+      "up lic",
+      "u plic",
+      "▁a dj",
+      "▁ad j",
+      "▁ adj",
+      "大 于",
+      "裁定 书",
+      "▁ex tensive",
+      "▁ext ensive",
+      "英 文",
+      "平 面",
+      "▁research ers",
+      "▁researcher s",
+      "已经 发生法律效力",
+      "oc key",
+      "ock ey",
+      "it ud",
+      "itu d",
+      "▁ind icated",
+      "▁indic ated",
+      "▁indicate d",
+      "的原 因",
+      "的 原因",
+      "的 作用",
+      "ab il",
+      "abi l",
+      "a bil",
+      "▁rem ote",
+      "▁remot e",
+      "▁ remote",
+      "▁w riter",
+      "▁writ er",
+      "▁write r",
+      "▁wr iter",
+      "▁ writer",
+      "逾 期",
+      "分 公司",
+      "黄 金",
+      "ut y",
+      "u ty",
+      "构 建",
+      "he et",
+      "hee t",
+      "▁S ocial",
+      "▁Soc ial",
+      "▁Soci al",
+      "▁ Social",
+      "▁m o",
+      "▁ mo",
+      "▁al gebra",
+      "▁alg ebra",
+      "▁ algebra",
+      "A I",
+      "Di r",
+      "D ir",
+      "做 法",
+      "并按 对方当事人",
+      "并不 是",
+      "并 不是",
+      "▁A T",
+      "▁ AT",
+      "at eral",
+      "ate ral",
+      "ater al",
+      "▁g ather",
+      "▁ga ther",
+      "▁ gather",
+      "人 士",
+      "▁character ized",
+      "▁characterize d",
+      "▁ …",
+      "▁contin ues",
+      "▁continu es",
+      "▁continue s",
+      "不 如",
+      "十五 日内",
+      "十五日 内",
+      "十 五日内",
+      "▁ans wers",
+      "▁answer s",
+      "▁ answers",
+      "重 量",
+      "一 致",
+      "人 工",
+      "▁V alue",
+      "▁Val ue",
+      "▁ Value",
+      "我 在",
+      "be n",
+      "b en",
+      "▁t alking",
+      "▁talk ing",
+      "▁tal king",
+      "▁Cont rol",
+      "▁Contr ol",
+      "▁ Control",
+      "▁car cin",
+      "▁fr action",
+      "▁fract ion",
+      "▁fra ction",
+      "▁ fraction",
+      "▁H and",
+      "▁Ha nd",
+      "▁Han d",
+      "▁ Hand",
+      "▁mark ed",
+      "▁ marked",
+      "d c",
+      "▁c ircle",
+      "▁circ le",
+      "▁cir cle",
+      "▁ circle",
+      "▁R NA",
+      "▁RN A",
+      "▁ RNA",
+      "trans form",
+      "▁{ {",
+      "▁ {{",
+      "▁see ing",
+      "▁ seeing",
+      "▁s am",
+      "▁sa m",
+      "▁ sam",
+      "▁N umber",
+      "▁Num ber",
+      "▁Nu mber",
+      "▁ Number",
+      "▁dis covered",
+      "▁disc overed",
+      "▁discover ed",
+      "排 名",
+      "违 法",
+      "▁N ame",
+      "▁Na me",
+      "▁Nam e",
+      "▁ Name",
+      "▁n it",
+      "▁ni t",
+      "▁ nit",
+      "▁recomm ended",
+      "▁recommend ed",
+      "ch ain",
+      "cha in",
+      "chai n",
+      "▁up load",
+      "▁u pload",
+      "▁ upload",
+      "y g",
+      "am ental",
+      "ament al",
+      "amen tal",
+      "a mental",
+      "▁e pit",
+      "▁ep it",
+      "▁ epit",
+      "动 态",
+      "▁m anner",
+      "▁man ner",
+      "集 合",
+      "our se",
+      "ours e",
+      "o urse",
+      "经 审查",
+      "ol it",
+      "oli t",
+      "o lit",
+      "一 十九",
+      "是一 款",
+      "是 一款",
+      "提出 了",
+      "提 出了",
+      "还可 以",
+      "还 可以",
+      "▁内容 简介",
+      "▁ 内容简介",
+      "kt op",
+      "k top",
+      "防 止",
+      "▁H en",
+      "▁He n",
+      "▁A uthor",
+      "▁Aut hor",
+      "▁Auth or",
+      "▁Au thor",
+      "▁ Author",
+      "就 算",
+      "▁proper ly",
+      "▁prop erly",
+      "产品 的",
+      "产 品的",
+      "▁r ain",
+      "▁ra in",
+      "▁ rain",
+      "提出 副本",
+      "io stream",
+      "i ostream",
+      "Mi n",
+      "M in",
+      "▁Ar ray",
+      "▁Arr ay",
+      "▁ Array",
+      "此 时",
+      "▁di abetes",
+      "▁E X",
+      "▁ EX",
+      "▁b urn",
+      "▁bu rn",
+      "▁bur n",
+      "▁ burn",
+      "▁pol ynomial",
+      "杭 州",
+      "wi ki",
+      "wik i",
+      "w iki",
+      "强 化",
+      "▁pe pt",
+      "▁pep t",
+      "子 女",
+      "▁ste el",
+      "re land",
+      "rel and",
+      "r eland",
+      "▁v oc",
+      "▁vo c",
+      "▁ voc",
+      "Le vel",
+      "▁M useum",
+      "▁Muse um",
+      "▁fac ility",
+      "▁facilit y",
+      "fi n",
+      "f in",
+      "▁e volution",
+      "▁ev olution",
+      "生效 法律文书",
+      "▁wed ding",
+      "leq slant",
+      "▁f amous",
+      "▁fam ous",
+      "Config uration",
+      "▁国 籍",
+      "▁ 国籍",
+      "▁F ield",
+      "▁Fi eld",
+      "▁ Field",
+      "▁install ation",
+      "▁ installation",
+      "im ensions",
+      "imens ions",
+      "imension s",
+      "被害 人",
+      "都 要",
+      "通 道",
+      "ut ch",
+      "utc h",
+      "hent ication",
+      "单 元",
+      "▁sub mit",
+      "▁ submit",
+      "▁e at",
+      "▁ eat",
+      "▁s erving",
+      "▁serv ing",
+      "▁ser ving",
+      "▁ serving",
+      "ac le",
+      "acl e",
+      "a cle",
+      "▁en zyme",
+      "▁enzym e",
+      "▁ enzyme",
+      "适用 简易程序",
+      "▁} ,\\",
+      "▁}, \\",
+      "▁ },\\",
+      "Ass ert",
+      "▁n odes",
+      "▁no des",
+      "▁node s",
+      "▁nod es",
+      "▁ nodes",
+      "S L",
+      "出来 的",
+      "出 来的",
+      "▁hus band",
+      "允 许",
+      "▁S eries",
+      "▁Se ries",
+      "▁Ser ies",
+      "▁Serie s",
+      "▁ Series",
+      "▁Commun ity",
+      "▁Comm unity",
+      "▁ Community",
+      "▁W al",
+      "▁Wa l",
+      "▁ Wal",
+      "nC l",
+      "n Cl",
+      "带 着",
+      "b c",
+      "og g",
+      "o gg",
+      "equ al",
+      "eq ual",
+      "e qual",
+      "met ric",
+      "m etric",
+      "氧 化",
+      "解 析",
+      "ta b",
+      "t ab",
+      "攻 击",
+      "▁s ustain",
+      "▁sus tain",
+      "▁sust ain",
+      "hi gh",
+      "h igh",
+      "se cond",
+      "sec ond",
+      "secon d",
+      "PA TH",
+      "PAT H",
+      "P ATH",
+      "▁ 合作",
+      "相 等",
+      "补 充",
+      "独 特",
+      "on go",
+      "ong o",
+      "o ngo",
+      "还 在",
+      "不 多",
+      "类 别",
+      "▁elect ronic",
+      "▁electron ic",
+      "▁electro nic",
+      "▁electr onic",
+      "▁dep th",
+      "▁ depth",
+      "俄 罗斯",
+      "▁ （",
+      "▁de pending",
+      "▁dep ending",
+      "▁depend ing",
+      "各 地",
+      "fun c",
+      "fu nc",
+      "f unc",
+      "na p",
+      "n ap",
+      "▁P an",
+      "▁Pa n",
+      "▁ Pan",
+      "\"> <",
+      "\" ><",
+      "pl ed",
+      "ple d",
+      "p led",
+      "▁ap parent",
+      "▁app arent",
+      "▁appar ent",
+      "▁n uclear",
+      "▁nucle ar",
+      "▁nu clear",
+      "▁Bl ue",
+      "▁Blu e",
+      "▁ Blue",
+      "伙 伴",
+      "▁b rowser",
+      "▁brow ser",
+      "▁browse r",
+      "▁ browser",
+      "▁vol unte",
+      "\\\\ {",
+      "副 主任",
+      "▁t ail",
+      "▁ta il",
+      "▁ tail",
+      "ir a",
+      "i ra",
+      "案 例",
+      "▁ch osen",
+      "▁cho sen",
+      "▁chose n",
+      "▁ chosen",
+      "N N",
+      "主 体",
+      "通 信",
+      "▁wide ly",
+      "▁wid ely",
+      "sc ape",
+      "ro gen",
+      "rog en",
+      "r ogen",
+      "▁p resents",
+      "▁pres ents",
+      "▁present s",
+      "▁P ub",
+      "▁Pu b",
+      "▁ Pub",
+      "il ation",
+      "ila tion",
+      "i lation",
+      "▁T em",
+      "▁Te m",
+      "▁ Tem",
+      "pl ex",
+      "ple x",
+      "p lex",
+      "天 然",
+      "}} \\",
+      "} }\\",
+      "package s",
+      "pack ages",
+      "近 日",
+      "并 给出",
+      "▁trans cription",
+      "▁transcript ion",
+      "▁tran scription",
+      "▁ne ighbor",
+      "▁neigh bor",
+      "▁neighb or",
+      "战 斗",
+      "bt n",
+      "b tn",
+      "▁T im",
+      "▁Ti m",
+      "▁ Tim",
+      "身 上",
+      "条第 二",
+      "条 第二",
+      "债 权",
+      "▁i l",
+      "▁ il",
+      "▁P eople",
+      "▁Pe ople",
+      "▁ People",
+      "▁Con gress",
+      "▁Cong ress",
+      "de pend",
+      "dep end",
+      "d epend",
+      "Des cription",
+      "De scription",
+      "▁f if",
+      "▁fi f",
+      "▁ fif",
+      "▁F ort",
+      "▁For t",
+      "▁Fo rt",
+      "▁ Fort",
+      "Cont rol",
+      "▁y es",
+      "▁ye s",
+      "▁ yes",
+      "In it",
+      "I nit",
+      "fol d",
+      "fo ld",
+      "f old",
+      "te mp",
+      "tem p",
+      "t emp",
+      "um es",
+      "ume s",
+      "u mes",
+      "不 变",
+      "▁h eter",
+      "▁he ter",
+      "▁het er",
+      "▁ heter",
+      "D C",
+      "▁w arrant",
+      "▁war rant",
+      "il ot",
+      "ilo t",
+      "i lot",
+      "▁p urs",
+      "▁pur s",
+      "▁pu rs",
+      "平 衡",
+      "▁st ack",
+      "▁ stack",
+      "▁f abric",
+      "▁fab ric",
+      "▁ fabric",
+      "基 因",
+      "▁s ufficient",
+      "▁suff icient",
+      "▁help ful",
+      "▁E vent",
+      "▁Ev ent",
+      "▁Even t",
+      "▁Eve nt",
+      "▁ Event",
+      "sv g",
+      "s vg",
+      "▁econ omy",
+      "▁ec onomy",
+      "▁econom y",
+      "▁pro vider",
+      "▁prov ider",
+      "▁provid er",
+      "▁provide r",
+      "▁ provider",
+      "芯 片",
+      "▁f older",
+      "▁fol der",
+      "▁fold er",
+      "▁ folder",
+      "争 议",
+      "▁event ually",
+      "▁eventual ly",
+      "ad or",
+      "ado r",
+      "a dor",
+      "特别 是",
+      "▁wait ing",
+      "▁wa iting",
+      "▁ waiting",
+      "nN ot",
+      "nNo t",
+      "n Not",
+      "▁house hold",
+      "ut f",
+      "u tf",
+      "▁N ULL",
+      "▁NU LL",
+      "▁ NULL",
+      "▁v ision",
+      "▁vis ion",
+      "▁ vision",
+      "▁la unched",
+      "▁launch ed",
+      "▁y outh",
+      "▁you th",
+      "▁yo uth",
+      "第一 个",
+      "第 一个",
+      "非 法",
+      "▁i o",
+      "▁ io",
+      "搭 配",
+      "▁ch annels",
+      "▁channel s",
+      "▁chan nels",
+      "▁ channels",
+      "▁ 在",
+      "▁up dates",
+      "▁update s",
+      "▁upd ates",
+      "▁ updates",
+      "▁m unicip",
+      "▁ass ume",
+      "▁assum e",
+      "二〇一 八年",
+      "▁f aster",
+      "▁fast er",
+      "▁fa ster",
+      "▁ faster",
+      "▁re con",
+      "▁r econ",
+      "▁rec on",
+      "▁ass essed",
+      "▁assess ed",
+      "que ue",
+      "▁cont ainer",
+      "▁contain er",
+      "▁ container",
+      "t c",
+      "ro le",
+      "rol e",
+      "r ole",
+      "▁part ies",
+      "▁par ties",
+      "▁s hall",
+      "▁sh all",
+      "▁sha ll",
+      "▁ shall",
+      "▁m ent",
+      "▁me nt",
+      "▁men t",
+      "▁ ment",
+      "▁grad u",
+      "▁gra du",
+      "▁im plements",
+      "▁imp lements",
+      "▁implement s",
+      "▁impl ements",
+      "n v",
+      "ba d",
+      "b ad",
+      "▁R ef",
+      "▁Re f",
+      "▁ Ref",
+      "k o",
+      "期 待",
+      "▁o xygen",
+      "▁ oxygen",
+      "▁learn ed",
+      "▁lear ned",
+      "glob al",
+      "g lobal",
+      "学生 的",
+      "学 生的",
+      "ef t",
+      "e ft",
+      "pt r",
+      "p tr",
+      "▁h orm",
+      "▁hor m",
+      "▁ho rm",
+      "▁R ussia",
+      "▁Russ ia",
+      "▁a nc",
+      "▁an c",
+      "▁ anc",
+      "▁tr ials",
+      "▁tri als",
+      "▁trial s",
+      "LO G",
+      "L OG",
+      "▁b oy",
+      "▁bo y",
+      "▁ boy",
+      "mon th",
+      "mo nth",
+      "mont h",
+      "E ff",
+      "( :",
+      "当 代",
+      "E A",
+      "天 的",
+      "▁c riteria",
+      "▁crit eria",
+      "▁criter ia",
+      "▁ criteria",
+      "▁win ning",
+      "▁ winning",
+      "cs v",
+      "c sv",
+      "▁arg uments",
+      "▁argument s",
+      "▁argu ments",
+      "▁ arguments",
+      "Mod e",
+      "Mo de",
+      "M ode",
+      "ki ns",
+      "kin s",
+      "k ins",
+      "物 品",
+      "直 播",
+      "都可 以",
+      "都 可以",
+      "br idge",
+      "brid ge",
+      "b ridge",
+      "号 码",
+      "as sets",
+      "ass ets",
+      "asset s",
+      "asse ts",
+      "ins ert",
+      "inse rt",
+      "▁M edical",
+      "▁Med ical",
+      "▁Medic al",
+      "▁f an",
+      "▁fa n",
+      "▁ fan",
+      "j i",
+      "▁D isc",
+      "▁Dis c",
+      "▁Di sc",
+      "▁ Disc",
+      "▁B ased",
+      "▁Base d",
+      "▁Bas ed",
+      "▁Ba sed",
+      "▁ Based",
+      "▁sup ports",
+      "▁support s",
+      "▁me dal",
+      "▁med al",
+      "▁ax is",
+      "▁axi s",
+      "▁ axis",
+      "w s",
+      "ear s",
+      "ea rs",
+      "e ars",
+      "▁rec ognition",
+      "▁recogn ition",
+      "▁mot iv",
+      "Util s",
+      "交 纳",
+      "▁O nce",
+      "▁On ce",
+      "▁Onc e",
+      "▁ Once",
+      "▁path way",
+      "▁tr uly",
+      "( *",
+      "ne g",
+      "n eg",
+      "▁ 杨",
+      "▁min ute",
+      "▁ minute",
+      "man ager",
+      "manage r",
+      "m anager",
+      "▁look ed",
+      "▁lo oked",
+      "') ,\\",
+      "'), \\",
+      "' ),\\",
+      "是 有",
+      "哪 个",
+      "▁te acher",
+      "▁teach er",
+      "▁tea cher",
+      "▁ teacher",
+      "▁re presentation",
+      "▁rep resentation",
+      "▁represent ation",
+      "▁ representation",
+      "▁r ise",
+      "▁ris e",
+      "▁ri se",
+      "▁ rise",
+      "▁s port",
+      "▁sp ort",
+      "▁spo rt",
+      "▁spor t",
+      "▁ sport",
+      "user content",
+      "▁出生 日期",
+      "人 体",
+      "▁tw ice",
+      "▁Acc ording",
+      "▁ According",
+      "at re",
+      "a tre",
+      "▁P age",
+      "▁Pa ge",
+      "▁Pag e",
+      "▁ Page",
+      "nm odule",
+      "nmod ule",
+      "n module",
+      "浏 览",
+      "FI LE",
+      "FIL E",
+      "F ILE",
+      "性 格",
+      "▁c hair",
+      "▁ch air",
+      "▁cha ir",
+      "▁ chair",
+      "▁fl uid",
+      "▁flu id",
+      "▁ fluid",
+      "▁Ge org",
+      "▁Geo rg",
+      "pos es",
+      "pose s",
+      "po ses",
+      "p oses",
+      "最 多",
+      "iction ary",
+      "练 习",
+      "ch annel",
+      "chan nel",
+      "ok es",
+      "oke s",
+      "o kes",
+      "▁over ride",
+      "▁ override",
+      "▁comfort able",
+      "▁ comfortable",
+      "▁C H",
+      "▁ CH",
+      "dev ice",
+      "▁l oved",
+      "▁love d",
+      "▁lo ved",
+      "▁labor atory",
+      "del ta",
+      "d elta",
+      "▁CON DIT",
+      "Ta g",
+      "T ag",
+      "硕 士",
+      "▁M exico",
+      "▁Mex ico",
+      "▁w indows",
+      "▁window s",
+      "▁wind ows",
+      "▁ windows",
+      "权 益",
+      "he ster",
+      "hes ter",
+      "hest er",
+      "h ester",
+      "的 东西",
+      "gr ession",
+      "gress ion",
+      "g ression",
+      "ec ho",
+      "ech o",
+      "e cho",
+      "▁s weet",
+      "▁swe et",
+      "▁or iginally",
+      "▁orig inally",
+      "▁original ly",
+      "▁origin ally",
+      "审 批",
+      "金 钱",
+      "▁L os",
+      "▁Lo s",
+      "▁bi om",
+      "▁bio m",
+      "顺 利",
+      "oc hemical",
+      "ochem ical",
+      "o chemical",
+      "▁c offee",
+      "▁co ffee",
+      "▁coff ee",
+      "▁ coffee",
+      "的 基本",
+      "▁C ath",
+      "▁Ca th",
+      "▁Cat h",
+      "租 赁",
+      "后 面",
+      "人 在",
+      "覆 盖",
+      "提 取",
+      "初 始",
+      "▁happ ened",
+      "▁happen ed",
+      "▁I II",
+      "▁II I",
+      "▁ III",
+      "nThe se",
+      "nTh ese",
+      "n These",
+      "▁P aris",
+      "▁Par is",
+      "▁Pa ris",
+      "本案 现已审理终结",
+      "▁design s",
+      "nS ince",
+      "n Since",
+      "大多 数",
+      "大 多数",
+      "还 会",
+      "▁le aving",
+      "▁re liable",
+      "▁rel iable",
+      "▁anal yses",
+      "▁analy ses",
+      "▁analys es",
+      "▁analyse s",
+      "nN ew",
+      "nNe w",
+      "n New",
+      "推 出",
+      "nAcc ording",
+      "n According",
+      "请 回答",
+      "▁P eter",
+      "▁Pet er",
+      "▁Pe ter",
+      "▁Pete r",
+      "▁ Peter",
+      "▁m olecules",
+      "▁molec ules",
+      "▁molecule s",
+      "me nu",
+      "men u",
+      "m enu",
+      "▁Ch icago",
+      "▁Chic ago",
+      "▁s tick",
+      "▁st ick",
+      "▁ stick",
+      "▁e cho",
+      "▁ec ho",
+      "▁ech o",
+      "▁ echo",
+      "▁mult ip",
+      "▁multi p",
+      "▁mul tip",
+      "▁V ictor",
+      "▁Vi ctor",
+      "▁Vic tor",
+      "▁ Victor",
+      "东 方",
+      "▁b io",
+      "▁bi o",
+      "▁ bio",
+      "\") ,",
+      "\" ),",
+      "in ds",
+      "ind s",
+      "人 大",
+      "▁B et",
+      "▁Be t",
+      "▁Val ley",
+      "▁Vall ey",
+      "▁Valle y",
+      "▁the orem",
+      "▁theore m",
+      "▁theor em",
+      "bot tom",
+      "b ottom",
+      "▁The ir",
+      "re lease",
+      "的人数 提出副本",
+      "日 前",
+      "AC T",
+      "A CT",
+      "▁S ou",
+      "▁So u",
+      "▁h a",
+      "▁ ha",
+      "wor ld",
+      "w orld",
+      "成 年",
+      "ni t",
+      "n it",
+      "▁m ail",
+      "▁ma il",
+      "▁ mail",
+      "报 名",
+      "▁E arth",
+      "▁Ear th",
+      "▁ Earth",
+      "更多 的",
+      "更 多的",
+      "object s",
+      "obj ects",
+      "▁spir it",
+      "有 没有",
+      "▁al leg",
+      "▁all eg",
+      "▁alle g",
+      "ro oms",
+      "room s",
+      "第二百 四十",
+      "机 场",
+      "证 券",
+      "回 家",
+      "维 修",
+      "le rs",
+      "ler s",
+      "l ers",
+      "▁effect ively",
+      "▁effective ly",
+      "Att ribute",
+      "iddle ware",
+      "uit s",
+      "ui ts",
+      "u its",
+      "el led",
+      "ell ed",
+      "elle d",
+      "Ar gs",
+      "Arg s",
+      "▁BAS IS",
+      "il a",
+      "i la",
+      "efer ence",
+      "efe rence",
+      "e ference",
+      "▁in iti",
+      "▁init i",
+      "▁ini ti",
+      "▁Ar my",
+      "▁Arm y",
+      "▁CONDIT IONS",
+      "形 态",
+      "协 调",
+      "日起 至",
+      "▁pre val",
+      "▁pr eval",
+      "▁prev al",
+      "▁n ic",
+      "▁ni c",
+      "▁ nic",
+      "▁conf lic",
+      "▁pl astic",
+      "▁plas tic",
+      "▁ plastic",
+      "ia na",
+      "ian a",
+      "i ana",
+      "晋 江",
+      "col umn",
+      "colum n",
+      "▁c opyright",
+      "▁copy right",
+      "▁ copyright",
+      "▁un ivers",
+      "▁uni vers",
+      "也 许",
+      "新 疆",
+      "St ack",
+      "ble s",
+      "bl es",
+      "b les",
+      "百 分",
+      "项 之规定",
+      "ra t",
+      "r at",
+      "▁inv olving",
+      "▁invol ving",
+      "▁indust rial",
+      "if ies",
+      "ifi es",
+      "自 主",
+      "▁r aw",
+      "▁ra w",
+      "▁ raw",
+      "良 好",
+      "▁j ump",
+      "▁ju mp",
+      "▁ jump",
+      "▁H y",
+      "▁ Hy",
+      "合 并",
+      "▁un less",
+      "▁unle ss",
+      "▁ unless",
+      "▁B razil",
+      "股 票",
+      "▁c alling",
+      "▁call ing",
+      "▁cal ling",
+      "▁e ducational",
+      "▁educ ational",
+      "▁education al",
+      "▁cl assic",
+      "▁class ic",
+      "▁ classic",
+      "▁连载 中",
+      "和 谐",
+      "Conn ection",
+      "Connect ion",
+      "▁L ight",
+      "▁Li ght",
+      "▁Lig ht",
+      "▁ Light",
+      "▁Spec ial",
+      "▁ Special",
+      "▁G overnment",
+      "▁Gover nment",
+      "▁Govern ment",
+      "▁lo an",
+      "S l",
+      ">( );\\",
+      ">() ;\\",
+      "> ();\\",
+      "em a",
+      "e ma",
+      "提 前",
+      "yt e",
+      "y te",
+      "n j",
+      "Me nu",
+      "Men u",
+      "M enu",
+      "也 能",
+      "▁n ine",
+      "▁nin e",
+      "▁ni ne",
+      "▁ nine",
+      "(_ _",
+      "( __",
+      "▁ke pt",
+      "▁in strument",
+      "▁instr ument",
+      "▁ instrument",
+      "R T",
+      "▁C urrent",
+      "▁Cur rent",
+      "▁ Current",
+      "ok ing",
+      "okin g",
+      "oki ng",
+      "o king",
+      "ce ived",
+      "ceive d",
+      "ut orial",
+      "utor ial",
+      "uto rial",
+      "仲 裁",
+      "▁H ospital",
+      "▁Hospit al",
+      "平 时",
+      "▁comp ounds",
+      "▁compound s",
+      "▁e arth",
+      "▁ear th",
+      "▁ earth",
+      "▁dec isions",
+      "▁decision s",
+      "nHow ever",
+      "n However",
+      "res ults",
+      "result s",
+      "面 临",
+      "Acc ess",
+      "Ac cess",
+      "A ccess",
+      "ed om",
+      "edo m",
+      "e dom",
+      "AN D",
+      "A ND",
+      "▁On ly",
+      "▁ Only",
+      "读 者",
+      "▁s pin",
+      "▁sp in",
+      "▁ spin",
+      "▁Vir ginia",
+      "▁Virgin ia",
+      "nL ist",
+      "n List",
+      "▁N a",
+      "▁ Na",
+      "co py",
+      "cop y",
+      "c opy",
+      "▁c limate",
+      "▁cl imate",
+      "▁clim ate",
+      "▁cli mate",
+      "ov ers",
+      "ove rs",
+      "over s",
+      "o vers",
+      "le ton",
+      "let on",
+      "l eton",
+      "▁s eat",
+      "▁se at",
+      "▁sea t",
+      "▁ seat",
+      "▁f requently",
+      "▁frequ ently",
+      "▁frequent ly",
+      "▁M a",
+      "▁ Ma",
+      "Mo n",
+      "M on",
+      "并 没有",
+      "( &",
+      "Pro duct",
+      "Prod uct",
+      "Produ ct",
+      "▁b ird",
+      "▁bi rd",
+      "▁bir d",
+      "▁ bird",
+      ";\\ \\",
+      "; \\\\",
+      "▁could n",
+      "大学 生",
+      "大 学生",
+      "▁Carol ina",
+      "▁o ral",
+      "▁or al",
+      "▁ oral",
+      "红 色",
+      "▁al gorithms",
+      "▁algorithm s",
+      "不 够",
+      "▁S and",
+      "▁San d",
+      "▁Sa nd",
+      "ail ability",
+      "Com p",
+      "Co mp",
+      "C omp",
+      "▁t ab",
+      "▁ta b",
+      "▁ tab",
+      "cl es",
+      "cle s",
+      "c les",
+      "▁st reet",
+      "▁stre et",
+      "▁ street",
+      "com mod",
+      "comm od",
+      "or ough",
+      "oro ugh",
+      "o rough",
+      "eng ine",
+      "▁w inter",
+      "▁win ter",
+      "▁wi nter",
+      "sc ribe",
+      "scrib e",
+      "scri be",
+      "scr ibe",
+      "s cribe",
+      "▁' @",
+      "▁p ure",
+      "▁pur e",
+      "▁pu re",
+      "▁ pure",
+      "Col lections",
+      "Collection s",
+      "Collect ions",
+      "Coll ections",
+      "等 奖",
+      "整 理",
+      "vent s",
+      "ven ts",
+      "v ents",
+      "▁as king",
+      "▁ask ing",
+      "AB ILITY",
+      "▁dep ends",
+      "▁depend s",
+      "▁ depends",
+      "▁J ose",
+      "▁Jo se",
+      "▁Jos e",
+      "区 间",
+      "ix ed",
+      "sc reen",
+      "scr een",
+      "s creen",
+      "▁s ensitive",
+      "▁sens itive",
+      "▁sensit ive",
+      "▁ sensitive",
+      "策 划",
+      "as p",
+      "a sp",
+      "配 套",
+      "原 料",
+      "▁for est",
+      "▁fore st",
+      "▁fo rest",
+      "▁ forest",
+      "授 权",
+      "ro ute",
+      "rou te",
+      "r oute",
+      "IG HT",
+      "IGH T",
+      "ru it",
+      "r uit",
+      "▁pos itions",
+      "▁position s",
+      "▁posit ions",
+      "▁ positions",
+      "▁S wed",
+      "▁Sw ed",
+      "▁Swe d",
+      "▁\" <",
+      "▁ \"<",
+      "也 很",
+      "▁e lected",
+      "▁elect ed",
+      "▁el ected",
+      "▁ elected",
+      "逻 辑",
+      "▁s hell",
+      "▁sh ell",
+      "▁she ll",
+      "▁shel l",
+      "▁ shell",
+      "▁target s",
+      "▁tar gets",
+      "▁ targets",
+      "cl uding",
+      "clud ing",
+      "an es",
+      "ane s",
+      "a nes",
+      "故 答案",
+      "有 个",
+      "▁comm only",
+      "▁common ly",
+      "Del ta",
+      "D elta",
+      "▁c ities",
+      "▁cit ies",
+      "▁ci ties",
+      "ic ular",
+      "icul ar",
+      "i cular",
+      "—— ——",
+      "——— —",
+      "— ———",
+      "JS ON",
+      "J SON",
+      "() :",
+      "( ):",
+      "▁de let",
+      "▁del et",
+      "▁ delet",
+      "同 志",
+      "▁re ceiving",
+      "▁rece iving",
+      "▁hum ans",
+      "▁human s",
+      "▁hu mans",
+      "Ad min",
+      "st itution",
+      "stit ution",
+      "stitu tion",
+      "法 的",
+      "吉林 省",
+      "天津 市",
+      "公开开庭 进行了审理",
+      "Op tion",
+      "Opt ion",
+      "O ption",
+      "▁c omposition",
+      "▁com position",
+      "▁comp osition",
+      "▁compos ition",
+      "住 房",
+      "融 合",
+      "可以 在",
+      "令 人",
+      "利 于",
+      "ha r",
+      "h ar",
+      "nt ype",
+      "nty pe",
+      "n type",
+      "▁o rient",
+      "▁or ient",
+      "▁ orient",
+      "向本院 提出",
+      "一个 月",
+      "一 个月",
+      "▁a bstract",
+      "▁ab stract",
+      "▁abs tract",
+      "▁ abstract",
+      "▁t rees",
+      "▁tr ees",
+      "▁tre es",
+      "▁tree s",
+      "▁ trees",
+      "▁[ '",
+      "▁ ['",
+      "ollow ing",
+      "oll owing",
+      "ollo wing",
+      "ha l",
+      "h al",
+      "▁Y es",
+      "▁Ye s",
+      "▁ Yes",
+      "优 惠",
+      "▁circ um",
+      "▁cir cum",
+      "▁keep ing",
+      "▁ keeping",
+      "arg ument",
+      "▁h olds",
+      "▁hold s",
+      "▁hol ds",
+      "▁re ceptors",
+      "▁recept ors",
+      "▁receptor s",
+      "▁innov ative",
+      "服 装",
+      "tv ar",
+      "t var",
+      "始 终",
+      "F ound",
+      "制 品",
+      "两 年",
+      "论 坛",
+      "form ance",
+      "▁ ±",
+      "也 要",
+      "▁re gression",
+      "▁reg ression",
+      "} _",
+      "类 的",
+      "拍 摄",
+      "▁cor ner",
+      "▁corn er",
+      "W idget",
+      "▁J es",
+      "▁Je s",
+      "▁i ron",
+      "▁ir on",
+      "▁ iron",
+      "im b",
+      "i mb",
+      "周 围",
+      "▁circ uit",
+      "▁ circuit",
+      "▁J SON",
+      "▁JS ON",
+      "▁ JSON",
+      "老 年",
+      "ep er",
+      "e per",
+      "na b",
+      "n ab",
+      "▁c ere",
+      "▁cer e",
+      "▁ce re",
+      "w d",
+      "▁n umpy",
+      "▁num py",
+      "▁ numpy",
+      "Re ader",
+      "Read er",
+      "R eader",
+      "Po rt",
+      "P ort",
+      "▁I nit",
+      "▁In it",
+      "▁ Init",
+      "nD o",
+      "n Do",
+      "上 午",
+      "情 节",
+      "lo se",
+      "los e",
+      "l ose",
+      "▁c raft",
+      "▁cra ft",
+      "▁ craft",
+      "▁jour ney",
+      "▁页 数",
+      "▁ 页数",
+      "Has h",
+      "Ha sh",
+      "H ash",
+      "女 儿",
+      "▁g rid",
+      "▁gr id",
+      "▁gri d",
+      "▁ grid",
+      "pos s",
+      "po ss",
+      "p oss",
+      "朋友 圈",
+      "▁re ality",
+      "▁real ity",
+      "不 想",
+      "ist ent",
+      "iste nt",
+      "▁v ehicles",
+      "▁veh icles",
+      "▁vehicle s",
+      "▁fac ulty",
+      "▁O ld",
+      "▁Ol d",
+      "▁ Old",
+      ")) )\\",
+      "))) \\",
+      ") ))\\",
+      "▁k m",
+      "▁ km",
+      "千 万",
+      "寻 找",
+      "房 间",
+      "ut ely",
+      "ute ly",
+      "▁exp ansion",
+      "▁expans ion",
+      "纤 维",
+      "内 蒙古",
+      "▁st udio",
+      "▁stud io",
+      "▁ studio",
+      "血 管",
+      "act ivity",
+      "activ ity",
+      "ce rs",
+      "cer s",
+      "c ers",
+      "dr op",
+      "dro p",
+      "d rop",
+      "▁A im",
+      "▁ Aim",
+      "▁Rec ords",
+      "▁Record s",
+      "▁ Records",
+      "▁Or gan",
+      "▁Org an",
+      "▁ Organ",
+      "▁treat ments",
+      "▁treatment s",
+      "她 们",
+      "预 测",
+      "▁d ict",
+      "▁di ct",
+      "▁dic t",
+      "▁ dict",
+      "f amily",
+      "ro g",
+      "r og",
+      "产 权",
+      "范围 内",
+      "ik ip",
+      "iki p",
+      "i kip",
+      "Win dow",
+      "Wind ow",
+      "W indow",
+      "中文 网",
+      "装 置",
+      "in ical",
+      "ini cal",
+      "inic al",
+      "Tem plate",
+      "Temp late",
+      "T emplate",
+      "随 机",
+      "ed ing",
+      "edi ng",
+      "e ding",
+      "接 到",
+      "po ol",
+      "p ool",
+      "▁Anal ysis",
+      "▁ Analysis",
+      "at i",
+      "a ti",
+      "▁con ventional",
+      "▁convention al",
+      "陈 述",
+      "Ex ec",
+      "ap es",
+      "ape s",
+      "a pes",
+      "▁e dition",
+      "▁ed ition",
+      "▁edit ion",
+      "本 来",
+      "S ession",
+      "pr ise",
+      "pri se",
+      "p rise",
+      "▁re produ",
+      "▁rep rodu",
+      "▁repro du",
+      "Un it",
+      "U nit",
+      "ri age",
+      "ria ge",
+      "r iage",
+      "▁cont ribution",
+      "▁contribut ion",
+      "▁t rail",
+      "▁tr ail",
+      "▁tra il",
+      "▁ trail",
+      "▁poss ibly",
+      "▁ possibly",
+      "▁correct ly",
+      "小 的",
+      "Tr ee",
+      "T ree",
+      "it z",
+      "i tz",
+      "机动 车",
+      "机 动车",
+      "ar ks",
+      "ark s",
+      "if icate",
+      "ific ate",
+      "r f",
+      "▁h eader",
+      "▁he ader",
+      "▁head er",
+      "▁ header",
+      "ed ge",
+      "的内 容",
+      "的 内容",
+      "Equ als",
+      "Equal s",
+      "对照 组",
+      "半 年",
+      "也 就",
+      "▁ п",
+      "▁col our",
+      "▁ colour",
+      "▁us age",
+      "▁ usage",
+      "▁b ag",
+      "▁ba g",
+      "▁ bag",
+      "函数 的",
+      "函 数的",
+      "nN ote",
+      "nNo te",
+      "nNot e",
+      "n Note",
+      "附 近",
+      "N F",
+      "电 力",
+      "▁f orget",
+      "▁for get",
+      "▁forg et",
+      "▁ forget",
+      "▁s cores",
+      "▁sc ores",
+      "▁score s",
+      "▁ scores",
+      "Li n",
+      "L in",
+      "新 区",
+      "▁re placed",
+      "▁replac ed",
+      "▁replace d",
+      "▁k illed",
+      "▁kil led",
+      "▁kill ed",
+      "I con",
+      "▁E nergy",
+      "▁En ergy",
+      "▁ Energy",
+      "高 等",
+      "ufeff using",
+      "th read",
+      "thr ead",
+      "▁ac compl",
+      "▁accomp l",
+      "yo u",
+      "y ou",
+      "近年 来",
+      "近 年来",
+      "im pl",
+      "imp l",
+      "超 级",
+      "天 气",
+      "Dir ect",
+      "Di rect",
+      "D irect",
+      "▁fr act",
+      "▁fra ct",
+      "▁pl ugin",
+      "▁plug in",
+      "▁ plugin",
+      "▁c ourses",
+      "▁co urses",
+      "▁cour ses",
+      "▁course s",
+      "▁ courses",
+      "高 效",
+      "▁T er",
+      "▁Te r",
+      "▁ Ter",
+      "▁ident ification",
+      "地 图",
+      "By Id",
+      "吸 收",
+      "▁c ultural",
+      "▁cult ural",
+      "▁ cultural",
+      "▁l aser",
+      "▁la ser",
+      "▁las er",
+      "做 了",
+      "ap se",
+      "aps e",
+      "账 户",
+      "IV E",
+      "I VE",
+      "▁g arden",
+      "▁gar den",
+      "▁gard en",
+      "▁w est",
+      "▁we st",
+      "▁ west",
+      "翻 译",
+      "▁co ast",
+      "▁n urs",
+      "▁nur s",
+      "▁nu rs",
+      "▁c er",
+      "▁ce r",
+      "▁ cer",
+      "ri ze",
+      "riz e",
+      "r ize",
+      "id ity",
+      "idi ty",
+      "▁Con ference",
+      "▁g one",
+      "▁go ne",
+      "▁gon e",
+      "▁ gone",
+      "▁exc ited",
+      "▁excit ed",
+      "千 克",
+      "条 规定",
+      "▁R ichard",
+      "▁Rich ard",
+      "▁Ric hard",
+      "▁Ri chard",
+      "▁ Richard",
+      "▁s erial",
+      "▁se rial",
+      "▁ser ial",
+      "▁ serial",
+      "中国 的",
+      "中 国的",
+      "▁D oes",
+      "▁Do es",
+      "▁ Does",
+      "出 发",
+      "ce nce",
+      "c ence",
+      "周 边",
+      "▁a gency",
+      "▁ag ency",
+      "▁age ncy",
+      "▁ agency",
+      "ac hing",
+      "ach ing",
+      "achi ng",
+      "a ching",
+      "it o",
+      "i to",
+      "▁s urgical",
+      "▁surg ical",
+      "▁ surgical",
+      "an iel",
+      "ani el",
+      "anie l",
+      "▁More over",
+      "▁col on",
+      "▁co lon",
+      "▁ colon",
+      "ul s",
+      "u ls",
+      "oint er",
+      "oin ter",
+      "oi nter",
+      "o inter",
+      "房 子",
+      "▁J ew",
+      "▁Je w",
+      "介绍 了",
+      "J ECT",
+      "ac ket",
+      "ack et",
+      "▁law y",
+      "▁la wy",
+      "tr ain",
+      "tra in",
+      "t rain",
+      "$$ ,",
+      "$ $,",
+      "▁B ur",
+      "▁Bu r",
+      "▁ Bur",
+      "▁cons umption",
+      "▁consum ption",
+      "▁a rm",
+      "▁ar m",
+      "▁ arm",
+      "S M",
+      "st rip",
+      "str ip",
+      "stri p",
+      "s trip",
+      "H R",
+      "▁B rown",
+      "▁Br own",
+      "▁Bro wn",
+      "▁Brow n",
+      "▁ Brown",
+      "葡 萄",
+      "明 星",
+      "z z",
+      "银行 存款",
+      "▁Sec ret",
+      "▁ Secret",
+      "三 十",
+      "基 层",
+      "环 节",
+      "▁br other",
+      "▁bro ther",
+      "▁broth er",
+      "中国 人民",
+      "中国人 民",
+      "os ton",
+      "ost on",
+      "o ston",
+      "net work",
+      "n etwork",
+      "▁S pring",
+      "▁Sp ring",
+      "▁Spr ing",
+      "▁ Spring",
+      "▁King dom",
+      "▁c oron",
+      "▁co ron",
+      "▁cor on",
+      "PO ST",
+      "POS T",
+      "P OST",
+      "渠 道",
+      "nI nd",
+      "nIn d",
+      "n Ind",
+      "A F",
+      "达 成",
+      "现 任",
+      "▁S tat",
+      "▁St at",
+      "▁Sta t",
+      "▁ Stat",
+      "女 孩",
+      "用 品",
+      "▁C orpor",
+      "▁Cor por",
+      "▁Corp or",
+      "▁friend ly",
+      "▁ friendly",
+      "开 心",
+      "一 名",
+      "script ion",
+      "scri ption",
+      "s cription",
+      "转 化",
+      "▁s oc",
+      "▁so c",
+      "交 付",
+      "▁re ject",
+      "▁ reject",
+      "休 闲",
+      "mos ph",
+      "m osph",
+      "en ger",
+      "eng er",
+      "enge r",
+      "ikip edia",
+      "▁claim s",
+      "▁ claims",
+      "▁Cent re",
+      "▁eff icacy",
+      "▁effic acy",
+      "▁ efficacy",
+      "▁un iversity",
+      "▁univers ity",
+      "St yle",
+      "Sty le",
+      "▁p resentation",
+      "▁present ation",
+      "▁ presentation",
+      "▁enh anced",
+      "▁enhance d",
+      "▁ enhanced",
+      "▁d ivers",
+      "▁di vers",
+      "▁div ers",
+      "▁diver s",
+      "▁dive rs",
+      "▁sign s",
+      "▁sig ns",
+      "f n",
+      "国 土",
+      "▁support ing",
+      "U P",
+      "▁tr igger",
+      "▁trig ger",
+      "▁ trigger",
+      "▁c hart",
+      "▁ch art",
+      "▁char t",
+      "▁cha rt",
+      "▁ chart",
+      "▁offic ials",
+      "▁official s",
+      "] {",
+      "h u",
+      "om ics",
+      "omic s",
+      "omi cs",
+      "▁is land",
+      "nt itle",
+      "n title",
+      "▁fund s",
+      "▁fun ds",
+      "__ ____",
+      "____ __",
+      "___ ___",
+      "_____ _",
+      "_ _____",
+      "▁rad ius",
+      "▁radi us",
+      "▁ radius",
+      "▁re tro",
+      "▁ret ro",
+      "ual ity",
+      "u ality",
+      "nA bstract",
+      "nAb stract",
+      "nAbs tract",
+      "n Abstract",
+      "n k",
+      "治 理",
+      "Car d",
+      "Ca rd",
+      "C ard",
+      "▁Com mand",
+      "▁Comm and",
+      "▁ Command",
+      "▁att ached",
+      "▁attach ed",
+      "wid get",
+      "w idget",
+      "▁sub str",
+      "▁subst r",
+      "▁subs tr",
+      "▁ substr",
+      "▁T re",
+      "▁Tr e",
+      "难 以",
+      "调 节",
+      "▁P ass",
+      "▁Pa ss",
+      "▁Pas s",
+      "▁ Pass",
+      "剧 情",
+      "刺 激",
+      "▁neuro ns",
+      "▁neur ons",
+      "▁neuron s",
+      "▁p olicies",
+      "▁polic ies",
+      "供 述",
+      "E G",
+      "▁en ded",
+      "▁end ed",
+      "▁ende d",
+      "▁ ended",
+      "▁U N",
+      "▁ UN",
+      "▁e ast",
+      "▁eas t",
+      "▁ east",
+      "▁S earch",
+      "▁Se arch",
+      "▁ Search",
+      "▁path ways",
+      "▁pathway s",
+      "fo o",
+      "f oo",
+      "is or",
+      "iso r",
+      "O M",
+      "▁L o",
+      "▁ Lo",
+      "n u",
+      "专 门",
+      "nT hey",
+      "nThe y",
+      "nTh ey",
+      "n They",
+      "▁v alu",
+      "▁val u",
+      "▁va lu",
+      "▁s ounds",
+      "▁sound s",
+      "▁emer gency",
+      "▁emerg ency",
+      "▁emerge ncy",
+      "▁b right",
+      "▁br ight",
+      "▁brig ht",
+      "▁ bright",
+      "由 此",
+      "ic ode",
+      "ico de",
+      "i code",
+      "这个 问题",
+      "失 败",
+      "ac ellular",
+      "a cellular",
+      "▁c reation",
+      "▁cre ation",
+      "▁creat ion",
+      "▁ creation",
+      "sc an",
+      "s can",
+      "Fr ame",
+      "F rame",
+      "air ed",
+      "ai red",
+      "aire d",
+      "a ired",
+      "Wh en",
+      "W hen",
+      "lic ense",
+      "l icense",
+      "公 交",
+      "m i",
+      "▁stat istics",
+      "▁statist ics",
+      "▁statistic s",
+      "▁ statistics",
+      "▁expert ise",
+      "▁percent age",
+      "▁ percentage",
+      "什么 是",
+      "货 币",
+      "▁acc omp",
+      "▁ac comp",
+      "(\\' /",
+      "(\\ '/",
+      "( \\'/",
+      "饮 食",
+      "本 地",
+      "▁W all",
+      "▁Wal l",
+      "▁Wa ll",
+      "▁ Wall",
+      "Set tings",
+      "Setting s",
+      "▁d ynamics",
+      "▁dynam ics",
+      "▁dynamic s",
+      "▁ dynamics",
+      "▁reg istered",
+      "▁register ed",
+      "▁ registered",
+      "强 度",
+      "iab ility",
+      "i ability",
+      "▁e mbed",
+      "▁em bed",
+      "▁emb ed",
+      "▁ embed",
+      "▁I reland",
+      "▁Ir eland",
+      "▁ Ireland",
+      "转 移",
+      "好 好",
+      "▁in ner",
+      "▁inn er",
+      "▁ inner",
+      "▁s ensor",
+      "▁sens or",
+      "▁ sensor",
+      "合同 纠纷一案中",
+      "合同纠纷一案 中",
+      "合同纠纷 一案中",
+      "▁ap ps",
+      "▁app s",
+      "▁ apps",
+      "混 合",
+      "ER R",
+      "E RR",
+      "▁fe atured",
+      "▁feature d",
+      "▁feat ured",
+      "▁ featured",
+      "▁vol tage",
+      "▁volt age",
+      "▁ voltage",
+      "n Z",
+      "现 金",
+      "▁ep isode",
+      "▁epis ode",
+      "▁ episode",
+      "▁B al",
+      "▁Ba l",
+      "▁ Bal",
+      "▁min imal",
+      "▁minim al",
+      "▁mini mal",
+      "▁ minimal",
+      "▁f old",
+      "▁fol d",
+      "▁fo ld",
+      "▁ fold",
+      "D o",
+      "ro ud",
+      "rou d",
+      "r oud",
+      "▁Championship s",
+      "▁Champions hips",
+      "▁Champion ships",
+      "外 的",
+      "▁M edia",
+      "▁Med ia",
+      "▁ Media",
+      "Ar e",
+      "A re",
+      "▁d ependent",
+      "▁dep endent",
+      "▁depend ent",
+      "▁ dependent",
+      "▁cit iz",
+      "▁gr ade",
+      "▁grad e",
+      "▁gra de",
+      "▁ grade",
+      "▁in cred",
+      "▁incre d",
+      "▁inc red",
+      "▁consider ing",
+      "运 算",
+      "n r",
+      "Sc hema",
+      "Sch ema",
+      "▁lead ership",
+      "▁leaders hip",
+      "▁leader ship",
+      "怎么 办",
+      "说 是",
+      "体 积",
+      "第一百五十四 条第一款第",
+      "第一百五十四条第一款 第",
+      "荣 誉",
+      "▁It aly",
+      "▁Ital y",
+      "ma l",
+      "m al",
+      "ga mma",
+      "gam ma",
+      "g amma",
+      "决 策",
+      "▁su spect",
+      "▁sus pect",
+      "▁susp ect",
+      "高 考",
+      "IS T",
+      "I ST",
+      "▁con j",
+      "▁co nj",
+      "▁c lar",
+      "▁cl ar",
+      "▁H istory",
+      "▁Histor y",
+      "▁Hi story",
+      "▁Hist ory",
+      "▁ History",
+      "▁S cot",
+      "▁Sc ot",
+      "▁A tl",
+      "▁At l",
+      "pro p",
+      "pr op",
+      "p rop",
+      "wh ite",
+      "whit e",
+      "▁aud ience",
+      "▁audi ence",
+      "nTh ank",
+      "n Thank",
+      "▁Th ree",
+      "▁Thr ee",
+      "▁ Three",
+      "to ols",
+      "tool s",
+      "too ls",
+      "t ools",
+      "▁Am azon",
+      "▁ Amazon",
+      "child ren",
+      "冠 军",
+      "为 由",
+      "收 益",
+      "▁R et",
+      "▁Re t",
+      "▁ Ret",
+      "主要 是",
+      "主 要是",
+      "▁f uel",
+      "▁fu el",
+      "▁ fuel",
+      "内 存",
+      "器 的",
+      "▁fund amental",
+      "▁fundament al",
+      "▁H D",
+      "▁ HD",
+      "实 用",
+      "率 为",
+      "▁L ibrary",
+      "▁ Library",
+      "否 则",
+      "▁r is",
+      "▁ri s",
+      "▁ ris",
+      "ak a",
+      "a ka",
+      "av ity",
+      "avit y",
+      "avi ty",
+      "▁I ran",
+      "▁Ir an",
+      "事 项",
+      "▁sh opping",
+      "▁shop ping",
+      "▁Sec urity",
+      "▁ Security",
+      "at uring",
+      "atur ing",
+      "atu ring",
+      "▁ver sus",
+      "▁vers us",
+      "▁k new",
+      "▁kn ew",
+      "▁kne w",
+      "▁in stitution",
+      "▁inst itution",
+      "▁instit ution",
+      "▁sub set",
+      "▁subs et",
+      "▁ subset",
+      "▁s tem",
+      "▁st em",
+      "▁ste m",
+      "▁ stem",
+      "路 径",
+      "est yle",
+      "esty le",
+      "e style",
+      "▁ad vent",
+      "▁adv ent",
+      "▁screen ing",
+      "▁P ay",
+      "▁Pa y",
+      "▁ Pay",
+      "\") ,\\",
+      "\"), \\",
+      "\" ),\\",
+      "Ga me",
+      "G ame",
+      "▁e pid",
+      "▁ep id",
+      "▁ epid",
+      "▁do ctor",
+      "▁doc tor",
+      "▁ doctor",
+      "呈 现",
+      "▁sw im",
+      "com bin",
+      "comb in",
+      "ma c",
+      "m ac",
+      "代 入",
+      "el ect",
+      "ele ct",
+      "e lect",
+      "▁ne ural",
+      "▁neur al",
+      "加 快",
+      "室 内",
+      "最好 的",
+      "最 好的",
+      "▁sequ ences",
+      "▁sequence s",
+      "▁ sequences",
+      "有 多",
+      "▁Un less",
+      "▁ Unless",
+      "vent ions",
+      "vention s",
+      "再 审",
+      "▁N on",
+      "▁No n",
+      "▁ Non",
+      "S V",
+      "um ps",
+      "ump s",
+      "▁te achers",
+      "▁teach ers",
+      "▁teacher s",
+      "im a",
+      "i ma",
+      "工作 人员",
+      "一 体",
+      "迟 延",
+      "OD O",
+      "O DO",
+      "回 到",
+      "指定 的",
+      "指 定的",
+      "仍 然",
+      "▁st ars",
+      "▁star s",
+      "▁ stars",
+      "爸 爸",
+      "▁Ar ts",
+      "▁Art s",
+      "过 来",
+      "上 下",
+      "在 此",
+      "\\' ),\\",
+      "\\') ,\\",
+      "\\'), \\",
+      "\\ '),\\",
+      "▁r aise",
+      "▁ra ise",
+      "▁ raise",
+      "▁ke ys",
+      "▁key s",
+      "▁ keys",
+      "▁wh atever",
+      "▁what ever",
+      "▁ whatever",
+      "ta gs",
+      "tag s",
+      "t ags",
+      "金 额",
+      "都没 有",
+      "都 没有",
+      "ed s",
+      "e ds",
+      "▁Sp anish",
+      "▁ Spanish",
+      "▁人 物",
+      "▁ 人物",
+      "▁card iac",
+      "▁cardi ac",
+      "SU L",
+      "S UL",
+      "ibr aries",
+      "▁d raft",
+      "▁ draft",
+      "▁T R",
+      "▁ TR",
+      "▁d efe",
+      "▁de fe",
+      "▁def e",
+      "om ega",
+      "ome ga",
+      "o mega",
+      "un tu",
+      "unt u",
+      "▁would n",
+      "cre en",
+      "cr een",
+      "c reen",
+      "\"{ {",
+      "\" {{",
+      "▁s oil",
+      "▁so il",
+      "色 彩",
+      "iz ontal",
+      "izon tal",
+      "on ia",
+      "oni a",
+      "o nia",
+      "▁S erver",
+      "▁Ser ver",
+      "▁Serv er",
+      "▁Serve r",
+      "▁ Server",
+      "by te",
+      "b yte",
+      "wa n",
+      "w an",
+      "} ;",
+      "▁pro ceed",
+      "▁proc eed",
+      "很 有",
+      "nA ss",
+      "nAs s",
+      "n Ass",
+      "细 节",
+      "发 动",
+      "Em pty",
+      "Emp ty",
+      "br o",
+      "b ro",
+      "▁A ff",
+      "▁Af f",
+      "▁ Aff",
+      "博 物",
+      "Buff er",
+      "Buf fer",
+      "B uffer",
+      "▁t ips",
+      "▁tip s",
+      "▁ti ps",
+      "▁ tips",
+      "开发 区",
+      "▁ev ening",
+      "▁even ing",
+      "up le",
+      "u ple",
+      "▁v ot",
+      "▁vo t",
+      "▁c row",
+      "▁cr ow",
+      "▁cro w",
+      "▁ crow",
+      "▁dis count",
+      "▁disc ount",
+      "▁ discount",
+      "有 很多",
+      "i y",
+      "▁pass ion",
+      "}$ ,",
+      "} $,",
+      "甘 肃",
+      "▁appoint ed",
+      "ip t",
+      "i pt",
+      "cont roller",
+      "control ler",
+      "知 名",
+      "om atic",
+      "oma tic",
+      "omat ic",
+      "o matic",
+      "}} \\\\",
+      "}}\\ \\",
+      "} }\\\\",
+      "▁ext ent",
+      "cl aim",
+      "▁H a",
+      "▁ Ha",
+      "▁st ores",
+      "▁store s",
+      "▁sto res",
+      "▁ stores",
+      "▁ph il",
+      "▁phi l",
+      "▁ phil",
+      "me an",
+      "m ean",
+      "▁ph armac",
+      "▁pharm ac",
+      "信用 卡",
+      "▁s alt",
+      "▁sa lt",
+      "▁sal t",
+      "▁ salt",
+      "in ternal",
+      "int ernal",
+      "inter nal",
+      "intern al",
+      "的 同时",
+      "▁c ognitive",
+      "▁cogn itive",
+      "▁ cognitive",
+      "故答案 为",
+      "登 录",
+      "▁sign als",
+      "▁signal s",
+      "▁sig nals",
+      "▁ signals",
+      "down load",
+      "d ownload",
+      "学 位",
+      "回 来",
+      "性 别",
+      "}, {\"",
+      "},{ \"",
+      "▁pot entially",
+      "▁potential ly",
+      "▁potent ially",
+      "正当 理由",
+      "▁n a",
+      "▁ na",
+      "精 彩",
+      "om ain",
+      "oma in",
+      "o main",
+      "▁pack ages",
+      "▁package s",
+      "▁ packages",
+      "▁b asketball",
+      "▁basket ball",
+      "数据 库",
+      "▁C R",
+      "▁ CR",
+      "▁As ian",
+      "▁Asia n",
+      "▁t ypical",
+      "▁typ ical",
+      "▁c andidates",
+      "▁candid ates",
+      "▁candidate s",
+      "▁h em",
+      "▁he m",
+      "▁ hem",
+      "▁C ross",
+      "▁Cr oss",
+      "▁Cro ss",
+      "▁Cros s",
+      "▁ Cross",
+      "Gener ic",
+      "Gen eric",
+      "Gene ric",
+      "▁h earing",
+      "▁he aring",
+      "▁hear ing",
+      "印 度",
+      "ov es",
+      "ove s",
+      "o ves",
+      "ent ials",
+      "ential s",
+      "enti als",
+      "Te X",
+      "tic al",
+      "ti cal",
+      "t ical",
+      "宁 波",
+      "O b",
+      "▁h istorical",
+      "▁histor ical",
+      "▁historic al",
+      "▁o ptical",
+      "▁op tical",
+      "▁opt ical",
+      "▁optic al",
+      "的 关键",
+      "Ca n",
+      "C an",
+      "▁Con fig",
+      "▁Co nfig",
+      "▁Conf ig",
+      "▁ Config",
+      "刑初 字第",
+      "刑 初字第",
+      "丰富 的",
+      "▁recogn ized",
+      "▁recognize d",
+      "▁ recognized",
+      "专业 的",
+      "专 业的",
+      "预 交",
+      "▁f ormal",
+      "▁for mal",
+      "▁form al",
+      "▁ formal",
+      "la r",
+      "l ar",
+      "EF A",
+      "E FA",
+      "▁A bstract",
+      "▁Ab stract",
+      "▁Abs tract",
+      "▁ Abstract",
+      "▁c e",
+      "▁ ce",
+      "情况 下",
+      "▁acc um",
+      "▁ac cum",
+      "▁ accum",
+      "▁pop ulations",
+      "▁population s",
+      "▁popul ations",
+      "▁ populations",
+      "▁m ir",
+      "▁mi r",
+      "▁ mir",
+      "上 升",
+      "di g",
+      "d ig",
+      "▁vari ation",
+      "▁var iation",
+      "▁P rem",
+      "▁Pr em",
+      "▁Pre m",
+      "▁ Prem",
+      "▁un iform",
+      "▁uni form",
+      "▁ uniform",
+      "if ts",
+      "ift s",
+      "i fts",
+      "li nt",
+      "lin t",
+      "l int",
+      "▁Up date",
+      "▁ Update",
+      "师 范",
+      "▁m ic",
+      "▁mi c",
+      "▁ mic",
+      "老 人",
+      "房地产 开发",
+      "▁b oot",
+      "▁bo ot",
+      "▁ boot",
+      "解题 过程",
+      "▁d iverse",
+      "▁di verse",
+      "▁divers e",
+      "▁diver se",
+      "▁d ream",
+      "▁ dream",
+      "▁tum ors",
+      "▁tumor s",
+      "▁K ar",
+      "▁Ka r",
+      "▁ Kar",
+      "梦 想",
+      "▁spe ech",
+      "▁ speech",
+      "▁par ams",
+      "▁para ms",
+      "▁param s",
+      "▁ params",
+      "未 能",
+      "不 起",
+      "共 享",
+      "Bo dy",
+      "B ody",
+      "▁ve get",
+      "el ler",
+      "ell er",
+      "elle r",
+      "e ller",
+      "化 工",
+      "▁ac commod",
+      "取 消",
+      "型 的",
+      "G R",
+      "▁hol iday",
+      "id ents",
+      "ident s",
+      "iden ts",
+      "▁h ash",
+      "▁has h",
+      "▁ha sh",
+      "▁ hash",
+      "tr i",
+      "t ri",
+      "▁un ion",
+      "▁uni on",
+      "▁ union",
+      "迟延 履行",
+      "WO RD",
+      "WOR D",
+      "W ORD",
+      "▁cle aning",
+      "▁clean ing",
+      "▁dis order",
+      "▁Ang eles",
+      "▁Angel es",
+      "答 辩",
+      "▁d ir",
+      "▁di r",
+      "▁ dir",
+      "地 球",
+      "▁O nline",
+      "▁On line",
+      "▁ Online",
+      "▁b ott",
+      "▁bo tt",
+      "▁bot t",
+      "▁G lobal",
+      "▁Glo bal",
+      "▁Glob al",
+      "▁ Global",
+      "▁u rban",
+      "▁ur ban",
+      "▁ urban",
+      "cont rib",
+      "contr ib",
+      "conn ection",
+      "connect ion",
+      "▁clos ely",
+      "▁close ly",
+      "▁w w",
+      "▁ ww",
+      "hy dro",
+      "hyd ro",
+      "h ydro",
+      "▁rem ained",
+      "▁remain ed",
+      "▁surround ing",
+      "▁l ig",
+      "▁li g",
+      "▁ lig",
+      "en nis",
+      "enn is",
+      "B D",
+      "SI ON",
+      "S ION",
+      "▁ro oms",
+      "▁room s",
+      "▁ rooms",
+      "▁comp act",
+      "▁ compact",
+      "互 动",
+      "的 研究",
+      "HT ML",
+      "d l",
+      "ag r",
+      "a gr",
+      "uo us",
+      "u ous",
+      "em on",
+      "emo n",
+      "e mon",
+      "▁anal og",
+      "▁ana log",
+      "生活 中",
+      "第 八",
+      "uk e",
+      "u ke",
+      "ov ing",
+      "ovi ng",
+      "o ving",
+      "▁N E",
+      "▁ NE",
+      "▁t oler",
+      "▁to ler",
+      "▁dec ide",
+      "art ments",
+      "artment s",
+      "▁E ff",
+      "▁ Eff",
+      "▁r ub",
+      "▁ru b",
+      "▁ rub",
+      "IN T",
+      "I NT",
+      "▁g ender",
+      "▁gen der",
+      "▁ge nder",
+      "▁ gender",
+      "▁ro of",
+      "▁e nv",
+      "▁en v",
+      "▁ env",
+      "I X",
+      "▁a udio",
+      "▁aud io",
+      "▁audi o",
+      "▁ audio",
+      "新 华",
+      "▁E p",
+      "▁ Ep",
+      "审 核",
+      "▁ch ief",
+      "▁chi ef",
+      "▁ chief",
+      "▁w ine",
+      "▁win e",
+      "▁wi ne",
+      "^\\ \\",
+      "^ \\\\",
+      "▁a mb",
+      "▁am b",
+      "▁ amb",
+      "▁ac counts",
+      "▁account s",
+      "▁ accounts",
+      "ia no",
+      "ian o",
+      "i ano",
+      "免 疫",
+      "▁organ ic",
+      "▁org anic",
+      "▁ organic",
+      "▁p od",
+      "▁po d",
+      "▁ pod",
+      "and roid",
+      "andro id",
+      "andr oid",
+      "▁A dv",
+      "▁Ad v",
+      "▁ Adv",
+      "道 德",
+      "▁new ly",
+      "▁C ast",
+      "▁Ca st",
+      "▁Cas t",
+      "▁ Cast",
+      "属 性",
+      "uc ket",
+      "uck et",
+      "so l",
+      "s ol",
+      "现在 的",
+      "doc ker",
+      "dock er",
+      "d ocker",
+      "ch er",
+      "che r",
+      "c her",
+      "县 人",
+      "无 正当理由",
+      "▁rad iation",
+      "▁radi ation",
+      "▁As ia",
+      "▁ Asia",
+      "▁re verse",
+      "▁revers e",
+      "▁rever se",
+      "▁ reverse",
+      "nF il",
+      "n Fil",
+      "▁F estival",
+      "▁Fest ival",
+      "mo ck",
+      "m ock",
+      "图书 馆",
+      "东 北",
+      "▁st ages",
+      "▁stage s",
+      "直 到",
+      "is ters",
+      "ist ers",
+      "ister s",
+      "iste rs",
+      "i sters",
+      "许 可",
+      "nT hen",
+      "nThe n",
+      "nTh en",
+      "n Then",
+      "Ch ild",
+      "▁r iver",
+      "▁ri ver",
+      "▁riv er",
+      "▁ river",
+      "▁in stant",
+      "▁inst ant",
+      "▁ instant",
+      "as cript",
+      "asc ript",
+      "a script",
+      "▁int ellig",
+      "示 范",
+      "H D",
+      "▁f illed",
+      "▁fil led",
+      "▁fill ed",
+      "▁ filled",
+      "▁M en",
+      "▁Me n",
+      "▁ Men",
+      "收 费",
+      "Col lection",
+      "Collect ion",
+      "Coll ection",
+      "的 其他",
+      "▁ ×",
+      "归 还",
+      "▁r ac",
+      "▁ra c",
+      "▁ rac",
+      "▁class ification",
+      "▁ classification",
+      "M R",
+      "看 待",
+      "▁trans formation",
+      "▁transform ation",
+      "工作 者",
+      "工 作者",
+      "▁s now",
+      "▁sn ow",
+      "▁ snow",
+      "nR ec",
+      "nRe c",
+      "n Rec",
+      "▁A ction",
+      "▁Ac tion",
+      "▁Act ion",
+      "▁ Action",
+      "每一 个",
+      "每 一个",
+      "ix el",
+      "▁c overs",
+      "▁co vers",
+      "▁cover s",
+      "▁cov ers",
+      "▁t issues",
+      "▁tissue s",
+      "▁se asons",
+      "▁season s",
+      "▁seas ons",
+      "▁ad opt",
+      "▁adop t",
+      ")} {",
+      ") }{",
+      "▁Y O",
+      "▁ YO",
+      "▁com pat",
+      "▁comp at",
+      "▁ compat",
+      "每 一",
+      "Pos ition",
+      "P osition",
+      "▁pre pare",
+      "▁prep are",
+      "▁prepar e",
+      "▁ prepare",
+      "▁sc ope",
+      "▁ scope",
+      "ne x",
+      "n ex",
+      "▁B as",
+      "▁Ba s",
+      "▁ Bas",
+      "故 意",
+      "▁result ed",
+      "▁p oll",
+      "▁pol l",
+      "▁po ll",
+      "▁ poll",
+      "进 口",
+      "幼儿 园",
+      "▁c ris",
+      "▁cr is",
+      "▁ex plan",
+      "▁exp lan",
+      "▁expl an",
+      "mon ary",
+      "河 北",
+      "▁t on",
+      "▁to n",
+      "▁ ton",
+      "▁com pr",
+      "▁comp r",
+      "责 令",
+      "▁pe ak",
+      "▁ peak",
+      "在 我",
+      "▁R eturn",
+      "▁Re turn",
+      "▁Ret urn",
+      "▁ Return",
+      "感 到",
+      "不 良",
+      "Pro gram",
+      "Pr ogram",
+      "ash board",
+      "▁T imes",
+      "▁Time s",
+      "▁Tim es",
+      "▁Ti mes",
+      "▁ Times",
+      "ai s",
+      "a is",
+      "▁re nal",
+      "User s",
+      "Us ers",
+      "Use rs",
+      "pr ed",
+      "pre d",
+      "p red",
+      "▁own ed",
+      "▁ow ned",
+      "▁ owned",
+      "▁c rypt",
+      "▁cry pt",
+      "▁ crypt",
+      "该 犯",
+      "j o",
+      "▁mark ets",
+      "▁market s",
+      "▁T w",
+      "▁ Tw",
+      "S R",
+      "ug by",
+      "ro duction",
+      "rodu ction",
+      "rod uction",
+      "▁d imension",
+      "▁dim ension",
+      "▁ dimension",
+      "lo or",
+      "loo r",
+      "l oor",
+      "估 计",
+      "场 景",
+      "▁med icine",
+      "▁medi cine",
+      "▁medic ine",
+      "▁ medicine",
+      "▁com plement",
+      "▁comp lement",
+      "▁compl ement",
+      "▁m ess",
+      "▁me ss",
+      "▁mes s",
+      "▁ mess",
+      "nL ast",
+      "nLa st",
+      "n Last",
+      "ch a",
+      "c ha",
+      "羁 押",
+      "▁C T",
+      "▁ CT",
+      "▁l imits",
+      "▁li mits",
+      "▁lim its",
+      "▁limit s",
+      "▁ limits",
+      "ul um",
+      "ulu m",
+      "▁Prov ince",
+      "or ney",
+      "orn ey",
+      "orne y",
+      "本院 在执行",
+      "本院在 执行",
+      "大 道",
+      "Data base",
+      "Dat abase",
+      "D atabase",
+      "能 量",
+      "En um",
+      "E num",
+      "B P",
+      "ip les",
+      "iple s",
+      "ipl es",
+      "i ples",
+      "干 净",
+      "真 是",
+      "nA dd",
+      "nAd d",
+      "n Add",
+      "▁P rot",
+      "▁Pro t",
+      "▁Pr ot",
+      "▁ Prot",
+      "▁cl uster",
+      "▁clust er",
+      "▁ cluster",
+      "▁dest roy",
+      "▁destr oy",
+      "▁ destroy",
+      "▁sil ver",
+      "执行 通知书",
+      "执行通知 书",
+      "等 方面",
+      "时 刻",
+      "▁poss ibility",
+      "▁S olve",
+      "▁Sol ve",
+      "▁ Solve",
+      "和 平",
+      "表 演",
+      "la t",
+      "l at",
+      "▁remain der",
+      "W E",
+      "监 测",
+      "Pr int",
+      "P rint",
+      "ui s",
+      "u is",
+      "解 放",
+      "▁be haviour",
+      "▁behav iour",
+      "▁behavi our",
+      "购 物",
+      "(\" /",
+      "( \"/",
+      "▁T able",
+      "▁Ta ble",
+      "▁Tab le",
+      "▁ Table",
+      "▁L O",
+      "▁ LO",
+      "Ro ute",
+      "R oute",
+      "▁re cover",
+      "▁rec over",
+      "▁con sole",
+      "▁cons ole",
+      "▁ console",
+      "判决书 送达之日起",
+      "be l",
+      "b el",
+      "ap py",
+      "app y",
+      "test ing",
+      "tes ting",
+      "t esting",
+      "缺 乏",
+      "ns on",
+      "nso n",
+      "n son",
+      "ter ior",
+      "te rior",
+      "t erior",
+      "reat ment",
+      "轴 承",
+      "IC E",
+      "I CE",
+      "▁rest aurant",
+      "▁restaur ant",
+      "▁ restaurant",
+      "江西 省",
+      "江 西省",
+      "U M",
+      "%; \\",
+      "% ;\\",
+      "RE AD",
+      "食 用",
+      "▁to pics",
+      "▁top ics",
+      "▁topic s",
+      "▁ topics",
+      "的 产品",
+      "条第二 款",
+      "条 第二款",
+      "ar ea",
+      "are a",
+      "a rea",
+      "联 盟",
+      "▁prim arily",
+      "▁v ul",
+      "▁acc ident",
+      "适 当",
+      "▁hy brid",
+      "▁ hybrid",
+      "似 乎",
+      "餐 厅",
+      "er rors",
+      "err ors",
+      "error s",
+      "C F",
+      "▁G it",
+      "▁Gi t",
+      "▁ Git",
+      "宠 物",
+      "男 生",
+      "ap ed",
+      "ape d",
+      "a ped",
+      "▁t a",
+      "▁ ta",
+      "▁A rab",
+      "▁Ar ab",
+      "▁st ated",
+      "▁state d",
+      "▁stat ed",
+      "▁l isting",
+      "▁list ing",
+      "▁ listing",
+      "过 了",
+      "▁Ex ec",
+      "▁ Exec",
+      "二 条",
+      "me s",
+      "m es",
+      "汉 语",
+      "咖 啡",
+      "▁pro lif",
+      "▁prol if",
+      "▁R am",
+      "▁Ra m",
+      "UL T",
+      "U LT",
+      "担 心",
+      "▁L ast",
+      "▁La st",
+      "▁Las t",
+      "▁ Last",
+      "▁redu cing",
+      "▁red ucing",
+      "▁K en",
+      "▁Ke n",
+      "▁ Ken",
+      "▁E R",
+      "▁ ER",
+      "r h",
+      "▁\" \\\\",
+      "▁\"\\ \\",
+      "▁ \"\\\\",
+      "航 空",
+      "▁describ es",
+      "▁describe s",
+      "社会 主义",
+      "ra wn",
+      "raw n",
+      "r awn",
+      "▁elect rical",
+      "▁electric al",
+      "▁electr ical",
+      "▁per spective",
+      "▁pers pective",
+      "▁perspect ive",
+      "▁mer ge",
+      "▁merg e",
+      "▁ merge",
+      "求 解",
+      "sh ot",
+      "s hot",
+      "ut or",
+      "uto r",
+      "u tor",
+      "▁p olar",
+      "▁pol ar",
+      "▁po lar",
+      "▁ polar",
+      "▁E dit",
+      "▁Ed it",
+      "▁ Edit",
+      "oc ard",
+      "oca rd",
+      "ocar d",
+      "o card",
+      "at ar",
+      "ata r",
+      "a tar",
+      "共 产",
+      "▁aw arded",
+      "▁award ed",
+      "▁L anguage",
+      "▁ Language",
+      "nE ng",
+      "nEn g",
+      "n Eng",
+      "▁w ww",
+      "▁ww w",
+      "▁ www",
+      "St orage",
+      "ycl e",
+      "yc le",
+      "y cle",
+      "▁r an",
+      "▁ra n",
+      "▁ ran",
+      "前 的",
+      "穿 越",
+      "nF irst",
+      "n First",
+      "部 长",
+      "▁cr uc",
+      "▁cru c",
+      "ov a",
+      "o va",
+      "现 状",
+      "▁s peak",
+      "▁spe ak",
+      "大 众",
+      "er ts",
+      "ert s",
+      "▁in depend",
+      "▁i ndepend",
+      "▁ind epend",
+      "某 些",
+      "期 的",
+      "含 有",
+      "▁inter act",
+      "▁im proving",
+      "▁impro ving",
+      "▁improv ing",
+      "с т",
+      "E ach",
+      "▁s izes",
+      "▁size s",
+      "▁si zes",
+      "▁ sizes",
+      "▁所 属",
+      "▁ 所属",
+      "辅 助",
+      "合适 的",
+      "舒 适",
+      "▁p p",
+      "▁ pp",
+      "serv ices",
+      "service s",
+      "s ervices",
+      "▁P op",
+      "▁Po p",
+      "▁ Pop",
+      "pre fix",
+      "pref ix",
+      "p refix",
+      "et ics",
+      "etic s",
+      "eti cs",
+      "较 大",
+      "▁P ac",
+      "▁Pa c",
+      "▁ Pac",
+      "td ef",
+      "t def",
+      "t i",
+      "P G",
+      "▁java x",
+      "▁jav ax",
+      "下面 的问题",
+      "下面的 问题",
+      "st ar",
+      "sta r",
+      "s tar",
+      "un a",
+      "u na",
+      "厦 门",
+      "im ation",
+      "ima tion",
+      "C ustom",
+      "nC ont",
+      "nCon t",
+      "nCo nt",
+      "n Cont",
+      "不 在",
+      "ir craft",
+      "irc raft",
+      "这 款",
+      "▁med ian",
+      "▁media n",
+      "▁medi an",
+      "▁ median",
+      "sc hema",
+      "sch ema",
+      "sche ma",
+      "湖 北",
+      "美 食",
+      "l f",
+      "▁M aster",
+      "▁Ma ster",
+      "▁Mas ter",
+      "▁Mast er",
+      "▁ Master",
+      "位 数",
+      "▁f light",
+      "▁fl ight",
+      "▁ flight",
+      "协 助",
+      "老 板",
+      "▁spect rum",
+      "▁ spectrum",
+      "i ography",
+      "zo ne",
+      "zon e",
+      "z one",
+      "平 行",
+      "▁C lick",
+      "▁Cl ick",
+      "▁ Click",
+      "▁phen omen",
+      "生 存",
+      "异 常",
+      "▁p ara",
+      "▁par a",
+      "▁pa ra",
+      "▁ para",
+      "▁strong ly",
+      "生效 后",
+      "▁accept ed",
+      "▁ accepted",
+      "单 调",
+      "想 法",
+      "请回答 下面的问题",
+      "▁L ink",
+      "▁Lin k",
+      "▁Li nk",
+      "▁ Link",
+      "分别 为",
+      "第 九",
+      "pl us",
+      "ON E",
+      "O NE",
+      "lo wer",
+      "low er",
+      "l ower",
+      "yn amic",
+      "ynam ic",
+      "yna mic",
+      "▁f amiliar",
+      "▁fam iliar",
+      "▁famil iar",
+      "Qu e",
+      "Q ue",
+      "bl ing",
+      "b ling",
+      "兄 弟",
+      "▁kn ows",
+      "▁know s",
+      "一 旦",
+      "N G",
+      "nB e",
+      "n Be",
+      "Pl ayer",
+      "Play er",
+      "P layer",
+      "z y",
+      "的要 求",
+      "的 要求",
+      "\\\\ ()",
+      "\\\\( )",
+      "▁effect iveness",
+      "▁effective ness",
+      "▁ effectiveness",
+      "▁f avour",
+      "▁fav our",
+      "密 码",
+      "em bers",
+      "ember s",
+      "emb ers",
+      "▁b locks",
+      "▁bl ocks",
+      "▁block s",
+      "▁blo cks",
+      "▁ blocks",
+      "▁a pi",
+      "▁ap i",
+      "▁ api",
+      "这就 是",
+      "这 就是",
+      "ta ke",
+      "t ake",
+      "相 应",
+      "就 在",
+      "os pit",
+      "osp it",
+      "指 定",
+      "▁g uy",
+      "▁gu y",
+      "▁ guy",
+      "▁E st",
+      "▁Es t",
+      "▁ Est",
+      "▁exec utive",
+      "▁execut ive",
+      "▁B E",
+      "▁ BE",
+      "▁r ough",
+      "▁ro ugh",
+      "▁rou gh",
+      "▁ rough",
+      "▁ann ounce",
+      "▁announc e",
+      "陕西 省",
+      "陕 西省",
+      "太 多",
+      "有 机",
+      "ri ev",
+      "rie v",
+      "r iev",
+      "▁del ivered",
+      "▁deliver ed",
+      "债务 利息",
+      "▁comp etitive",
+      "▁compet itive",
+      "▁competit ive",
+      "认 可",
+      "白 色",
+      "▁pre vention",
+      "▁prevent ion",
+      "▁prev ention",
+      "▁inc idence",
+      "▁prom pt",
+      "▁ prompt",
+      "▁in frastructure",
+      "▁infr astructure",
+      "▁Mar tin",
+      "▁Mart in",
+      "▁ Martin",
+      "▁s olar",
+      "▁so lar",
+      "▁sol ar",
+      "链 接",
+      "LE CT",
+      "L ECT",
+      "▁t ables",
+      "▁table s",
+      "▁tab les",
+      "▁ta bles",
+      "▁ tables",
+      "问题 的答案",
+      "问题的 答案",
+      "ru b",
+      "r ub",
+      "▁ins ulin",
+      "▁ insulin",
+      "课 堂",
+      "W idth",
+      "适用 于",
+      "适 用于",
+      "A b",
+      "部 署",
+      "方式 结案",
+      "尝 试",
+      "▁V ari",
+      "▁Var i",
+      "▁Va ri",
+      "▁ Vari",
+      "▁me chanical",
+      "▁mechan ical",
+      "▁mechanic al",
+      "用 来",
+      "集 成",
+      "▁b attery",
+      "▁batter y",
+      "▁bat tery",
+      "▁batt ery",
+      "规 格",
+      "▁ing red",
+      "项的 规定",
+      "项 的规定",
+      "▁c up",
+      "▁cu p",
+      "▁ cup",
+      "党 的",
+      "采 购",
+      "d p",
+      "Ca r",
+      "C ar",
+      "▁st yles",
+      "▁style s",
+      "▁styl es",
+      "▁ styles",
+      "ir us",
+      "i rus",
+      "纪 念",
+      "不 住",
+      "围 绕",
+      "▁char ges",
+      "▁charg es",
+      "▁charge s",
+      "▁instit utions",
+      "▁institution s",
+      "ro uter",
+      "rou ter",
+      "route r",
+      "r outer",
+      "▁H uman",
+      "▁Hu man",
+      "▁Hum an",
+      "▁ Human",
+      "元 及",
+      "▁tr ained",
+      "▁train ed",
+      "▁tra ined",
+      "▁ trained",
+      "nS ome",
+      "nSo me",
+      "n Some",
+      "全 体",
+      "▁l atter",
+      "▁lat ter",
+      "u h",
+      "▁B each",
+      "▁Be ach",
+      "▁Bea ch",
+      "解 除",
+      "或 许",
+      "yl van",
+      "or rect",
+      "orr ect",
+      "拒 绝",
+      "cp p",
+      "c pp",
+      "le ases",
+      "lease s",
+      "lea ses",
+      "▁c ategories",
+      "▁categor ies",
+      "▁ categories",
+      "▁be auty",
+      "▁beaut y",
+      "本裁定 书",
+      "本 裁定书",
+      "s a",
+      "都 市",
+      "Hom e",
+      "Ho me",
+      "H ome",
+      "▁s elling",
+      "▁sell ing",
+      "▁sel ling",
+      "▁ selling",
+      "▁d anger",
+      "▁dan ger",
+      "▁ danger",
+      "Gr aph",
+      "G raph",
+      "▁e num",
+      "▁en um",
+      "▁ enum",
+      "nAl so",
+      "n Also",
+      "清 洁",
+      "ans as",
+      "出版社 出版的图书",
+      "nA ny",
+      "nAn y",
+      "n Any",
+      "an th",
+      "ant h",
+      "a nth",
+      "人 力",
+      "▁ex posed",
+      "▁exp osed",
+      "▁expose d",
+      "▁ exposed",
+      "▁ 出",
+      "▁in hibition",
+      "▁inhib ition",
+      "▁inhibit ion",
+      "▁dec ades",
+      "▁decade s",
+      "身份 证",
+      "味 道",
+      "bs ite",
+      "b site",
+      "说 话",
+      "我 会",
+      "Ca che",
+      "C ache",
+      "旅 行",
+      "观 点",
+      "▁l at",
+      "▁la t",
+      "▁ lat",
+      "▁ex cess",
+      "▁exc ess",
+      "▁tour nament",
+      "等 待",
+      "▁D C",
+      "▁ DC",
+      "▁stim ul",
+      "▁d ating",
+      "▁dat ing",
+      "▁da ting",
+      "▁ dating",
+      "▁furn iture",
+      "Var i",
+      "V ari",
+      "不一 样",
+      "不 一样",
+      "收 藏",
+      "▁sub sequent",
+      "▁subsequ ent",
+      "rol lers",
+      "roll ers",
+      "roller s",
+      "▁h tml",
+      "▁ html",
+      "▁Char les",
+      "▁Charl es",
+      "nD e",
+      "n De",
+      "▁com mitted",
+      "▁comm itted",
+      "▁commit ted",
+      "▁m sg",
+      "▁ms g",
+      "▁ msg",
+      "Pass word",
+      "晋江 文学",
+      "▁l it",
+      "▁li t",
+      "▁ lit",
+      "ed u",
+      "e du",
+      "▁ag greg",
+      "▁ aggreg",
+      "▁sim ult",
+      "▁si mult",
+      "▁b eat",
+      "▁be at",
+      "▁ beat",
+      "▁de als",
+      "▁deal s",
+      "▁p erman",
+      "▁per man",
+      "▁perm an",
+      "服务 器",
+      "ni c",
+      "n ic",
+      "宣 布",
+      "pl et",
+      "ple t",
+      "p let",
+      "▁ch oices",
+      "▁cho ices",
+      "▁choice s",
+      "▁ choices",
+      "up y",
+      "u py",
+      "▁H ead",
+      "▁He ad",
+      "▁ Head",
+      "邀 请",
+      "×× ×",
+      "× ××",
+      "to ol",
+      "too l",
+      "t ool",
+      "nSup pose",
+      "n Suppose",
+      "▁M ur",
+      "▁Mu r",
+      "▁ Mur",
+      "▁v aluable",
+      "▁valu able",
+      "te x",
+      "t ex",
+      "▁a thlet",
+      "▁b ought",
+      "▁bo ught",
+      "▁bou ght",
+      "两 人",
+      "应该 是",
+      "可 靠",
+      "家 人",
+      "华 为",
+      "ers onal",
+      "erson al",
+      "▁en ds",
+      "▁end s",
+      "▁ ends",
+      "▁sc ored",
+      "▁score d",
+      "▁a sym",
+      "▁as ym",
+      "not e",
+      "no te",
+      "n ote",
+      "▁hard ware",
+      "▁ hardware",
+      "it ness",
+      "▁Fil m",
+      "▁Fi lm",
+      "开 启",
+      "生 素",
+      "ach ment",
+      "log ging",
+      "val s",
+      "va ls",
+      "v als",
+      "识 别",
+      "ress ive",
+      "r essive",
+      "▁p ig",
+      "▁pi g",
+      "▁ pig",
+      "） （",
+      "生 于",
+      "bl ue",
+      "▁p articles",
+      "▁part icles",
+      "▁partic les",
+      "▁particle s",
+      "▁ particles",
+      "uss ian",
+      "ussia n",
+      "说 了",
+      "▁v ess",
+      "▁ve ss",
+      "▁ves s",
+      "A m",
+      "▁re son",
+      "▁r eson",
+      "▁res on",
+      "撤 销",
+      "▁t empor",
+      "▁tem por",
+      "▁temp or",
+      "▁tempo r",
+      "▁collabor ation",
+      "总 体",
+      "nSub ject",
+      "n Subject",
+      "第五 十",
+      "第 五十",
+      "准 予",
+      "▁t hr",
+      "▁th r",
+      "▁ thr",
+      "on ds",
+      "ond s",
+      "cl udes",
+      "clud es",
+      "clude s",
+      "write r",
+      "writ er",
+      "wr iter",
+      "w riter",
+      "乡 村",
+      "▁s pons",
+      "▁sp ons",
+      "▁spo ns",
+      "▁ spons",
+      "▁sp atial",
+      "▁spat ial",
+      "▁ spatial",
+      "▁s eek",
+      "▁se ek",
+      "▁see k",
+      "▁ seek",
+      "ay a",
+      "a ya",
+      "hand ler",
+      "handle r",
+      "第二百五十 七条",
+      "第二百五十七 条",
+      "▁Stud y",
+      "▁ Study",
+      "le gal",
+      "leg al",
+      "l egal",
+      "ur able",
+      "ura ble",
+      "▁ass istance",
+      "▁assist ance",
+      "融 资",
+      "he ast",
+      "hea st",
+      "h east",
+      "P H",
+      "手 续",
+      "iff erent",
+      "iffer ent",
+      "iffe rent",
+      "▁rob ust",
+      "决定 书",
+      "当 年",
+      "更好 的",
+      "更 好的",
+      "物业 服务",
+      "▁exc iting",
+      "▁excit ing",
+      "中 小",
+      "▁c oh",
+      "▁co h",
+      "▁ coh",
+      "os oph",
+      "oso ph",
+      "e a",
+      "pu ll",
+      "p ull",
+      "的 关系",
+      "足 够",
+      "▁N S",
+      "▁ NS",
+      "▁dat etime",
+      "▁date time",
+      "▁ datetime",
+      "▁r ural",
+      "▁ru ral",
+      "这 道",
+      "▁p roud",
+      "▁pro ud",
+      "▁pr oud",
+      "▁m aps",
+      "▁map s",
+      "▁ma ps",
+      "▁ maps",
+      "立即 生效",
+      "评 审",
+      "▁l in",
+      "▁li n",
+      "▁ lin",
+      "笔 记",
+      "Mar k",
+      "Ma rk",
+      "M ark",
+      "▁in her",
+      "▁inh er",
+      "▁You ng",
+      "▁Yo ung",
+      "▁ Young",
+      "Ar t",
+      "A rt",
+      "ui ng",
+      "u ing",
+      "Hel per",
+      "Help er",
+      "▁trans plant",
+      "ic ken",
+      "ick en",
+      "以 调解",
+      "▁hyp othes",
+      "▁hypot hes",
+      "▁hypoth es",
+      "公 民",
+      "转 换",
+      "▁In vest",
+      "▁Inv est",
+      "▁ Invest",
+      "正确 的答案",
+      "正确的 答案",
+      "ol ds",
+      "old s",
+      "▁pur poses",
+      "▁purpose s",
+      "▁Ind ust",
+      "▁H istoric",
+      "▁Histor ic",
+      "▁Hist oric",
+      "各 位",
+      "wa r",
+      "w ar",
+      "接 下来",
+      "▁ap proved",
+      "▁appro ved",
+      "▁approve d",
+      "▁ approved",
+      "ob s",
+      "o bs",
+      "i ón",
+      "以调解 方式结案",
+      "盗窃 罪",
+      "▁stat istical",
+      "▁statist ical",
+      "▁statistic al",
+      "之 中",
+      "▁ 职业",
+      "▁P y",
+      "▁ Py",
+      "▁a ged",
+      "▁ag ed",
+      "▁age d",
+      "▁ aged",
+      "判 令",
+      "▁exp ensive",
+      "▁ expensive",
+      "wat ch",
+      "w atch",
+      "C V",
+      "sh a",
+      "s ha",
+      "▁gu ests",
+      "▁guest s",
+      "元 的",
+      "▁at mosph",
+      "古 代",
+      "是 对",
+      "虚 拟",
+      "HO U",
+      "H OU",
+      "zu re",
+      "z ure",
+      "▁d ogs",
+      "▁do gs",
+      "▁dog s",
+      "▁partners hip",
+      "▁partner ship",
+      "▁s tone",
+      "▁st one",
+      "▁sto ne",
+      "▁ stone",
+      "as ts",
+      "ast s",
+      "a sts",
+      "▁com plications",
+      "▁compl ications",
+      "▁complication s",
+      "▁ex amination",
+      "▁exam ination",
+      "PO SE",
+      "POS E",
+      "Si gn",
+      "Sig n",
+      "S ign",
+      "▁\\ '/",
+      "▁\\' /",
+      "▁ \\'/",
+      "w a",
+      "日 子",
+      "▁De scription",
+      "▁Des cription",
+      "▁ Description",
+      "c b",
+      "停 车",
+      "▁s uc",
+      "▁su c",
+      "同 一",
+      "合 格",
+      "Aut hor",
+      "Auth or",
+      "Au thor",
+      "A uthor",
+      "ook ing",
+      "oo king",
+      "o oking",
+      "怎么 样",
+      "并按对方当事人 的人数提出副本",
+      "并按对方当事人的人数 提出副本",
+      "▁att ended",
+      "▁attend ed",
+      "▁atten ded",
+      "Eng ine",
+      "曲 线",
+      "写 作",
+      "众 多",
+      "▁N et",
+      "▁Ne t",
+      "▁ Net",
+      "▁cap able",
+      "▁ca pable",
+      "} ]",
+      "车 程",
+      "OR M",
+      "O RM",
+      "▁c in",
+      "▁ci n",
+      "▁ cin",
+      "viron ments",
+      "vironment s",
+      "nY our",
+      "nYou r",
+      "n Your",
+      "▁li sten",
+      "▁list en",
+      "▁ listen",
+      "▁us ual",
+      "▁off icer",
+      "▁offic er",
+      "▁office r",
+      "ar ly",
+      "arl y",
+      "ate gor",
+      "ateg or",
+      "▁ 黄",
+      "以调解方式结案 的",
+      "▁establish ments",
+      "▁establishment s",
+      "判决书送达之日起 十五日内",
+      "▁sh ut",
+      "▁bound ary",
+      "▁ boundary",
+      "▁m RNA",
+      "▁ mRNA",
+      "保 留",
+      "he aders",
+      "head ers",
+      "header s",
+      "hea ders",
+      "官 网",
+      "Acc ount",
+      "Ac count",
+      "面 试",
+      "胜 利",
+      "毫 米",
+      "血 液",
+      "▁br oken",
+      "▁bro ken",
+      "▁broke n",
+      "▁ broken",
+      "▁n om",
+      "▁no m",
+      "▁ nom",
+      "ex ists",
+      "exist s",
+      "▁P ak",
+      "▁Pa k",
+      "▁P ath",
+      "▁Pat h",
+      "▁Pa th",
+      "▁ Path",
+      "▁ 主",
+      "T D",
+      "▁M S",
+      "▁ MS",
+      "is ted",
+      "ist ed",
+      "iste d",
+      "i sted",
+      "ylvan ia",
+      "黑 色",
+      "ea h",
+      "e ah",
+      "彻 底",
+      "er ic",
+      "eri c",
+      "e ric",
+      "on gs",
+      "ong s",
+      "标 的",
+      "笔 录",
+      "节 点",
+      "▁t ow",
+      "▁to w",
+      "sam ple",
+      "s ample",
+      "危 害",
+      "ou ts",
+      "out s",
+      "o uts",
+      "& &",
+      "▁s ed",
+      "▁se d",
+      "▁ sed",
+      "恋 爱",
+      "经营范围 包括",
+      "Act ivity",
+      "民 主",
+      "%) ,",
+      "% ),",
+      "re port",
+      "rep ort",
+      "repo rt",
+      "创作 的",
+      "创 作的",
+      "▁L ocal",
+      "▁Lo cal",
+      "▁Loc al",
+      "▁ Local",
+      "▁ex plicit",
+      "▁expl icit",
+      "▁ explicit",
+      "进 程",
+      "▁w aste",
+      "▁was te",
+      "▁wa ste",
+      "▁wast e",
+      "ay lor",
+      "Model s",
+      "Mod els",
+      "Mode ls",
+      "▁b rings",
+      "▁br ings",
+      "▁bring s",
+      "or em",
+      "ore m",
+      "o rem",
+      "▁ne ighborhood",
+      "▁neighbor hood",
+      "et al",
+      "eta l",
+      "e tal",
+      "会 上",
+      "▁HT ML",
+      "▁ HTML",
+      "▁reg ulatory",
+      "▁regul atory",
+      "▁regulator y",
+      "▁world wide",
+      "Cur rent",
+      "C urrent",
+      "▁H aw",
+      "▁Ha w",
+      "征 收",
+      "▁ex tend",
+      "▁ext end",
+      "▁ extend",
+      "pl atform",
+      "plat form",
+      "ra nd",
+      "ran d",
+      "r and",
+      "g i",
+      "在这 里",
+      "在 这里",
+      "▁ent ity",
+      "▁ entity",
+      "B o",
+      "不等 式",
+      "不 等式",
+      "ac ed",
+      "ace d",
+      "a ced",
+      "执行 员",
+      "▁t rav",
+      "▁tr av",
+      "▁tra v",
+      "|\\ \\",
+      "| \\\\",
+      "▁Oh io",
+      "动 画",
+      "▁unc ertain",
+      "N D",
+      "▁as say",
+      "▁ass ay",
+      "▁ assay",
+      "▁in terior",
+      "▁int erior",
+      "▁inter ior",
+      "▁I F",
+      "▁ IF",
+      "\") )\\",
+      "\")) \\",
+      "\" ))\\",
+      "RO N",
+      "R ON",
+      "考 生",
+      "▁pl enty",
+      "▁ple nty",
+      "▁re peated",
+      "▁repe ated",
+      "▁repeat ed",
+      "▁A ccess",
+      "▁Ac cess",
+      "▁Acc ess",
+      "▁ Access",
+      "ir ation",
+      "ira tion",
+      "i ration",
+      "▁v ascular",
+      "▁vas cular",
+      "▁vascul ar",
+      "▁ vascular",
+      "to String",
+      "▁G a",
+      "▁ Ga",
+      "完 美",
+      "▁quant um",
+      "▁ quantum",
+      "sec urity",
+      "▁H i",
+      "▁ Hi",
+      "Pl ugin",
+      "的 故事",
+      "ig in",
+      "igi n",
+      "i gin",
+      "▁S D",
+      "▁ SD",
+      "us ic",
+      "▁expl ained",
+      "▁explain ed",
+      "ov ascular",
+      "o vascular",
+      "▁st ored",
+      "▁store d",
+      "▁sto red",
+      "Doc ument",
+      "D ocument",
+      "普 遍",
+      "夫妻 感情",
+      "情 感",
+      "损 伤",
+      "▁t ight",
+      "▁ti ght",
+      "▁ tight",
+      "注 重",
+      "▁Pro cess",
+      "▁Proc ess",
+      "▁ Process",
+      "▁T ri",
+      "▁Tr i",
+      "▁ Tri",
+      "▁person s",
+      "▁pers ons",
+      "向 被执行人",
+      "车 站",
+      "▁F our",
+      "▁Fo ur",
+      "▁ Four",
+      "加 倍",
+      "▁r ig",
+      "▁ri g",
+      "▁ rig",
+      "ja v",
+      "j av",
+      "▁c hat",
+      "▁ch at",
+      "▁cha t",
+      "▁ chat",
+      "ui lt",
+      "u ilt",
+      "一 场",
+      "型 号",
+      "▁s tret",
+      "▁st ret",
+      "▁str et",
+      "▁stre t",
+      "▁ stret",
+      "▁Inst all",
+      "▁Ins tall",
+      "▁ Install",
+      "▁z one",
+      "▁ zone",
+      "▁in structions",
+      "▁instruction s",
+      "▁instruct ions",
+      "▁ instructions",
+      "▁North ern",
+      "▁Nor thern",
+      "pi n",
+      "p in",
+      "▁writ ers",
+      "▁write rs",
+      "▁writer s",
+      "▁ writers",
+      "st ood",
+      "sto od",
+      "uit e",
+      "ui te",
+      "u ite",
+      "消 防",
+      "二 年",
+      "▁ 作品",
+      "▁L i",
+      "▁ Li",
+      "▁En vironment",
+      "▁ Environment",
+      "page s",
+      "pa ges",
+      "pag es",
+      "p ages",
+      "▁b in",
+      "▁bi n",
+      "▁ bin",
+      "▁inc ident",
+      "fi re",
+      "f ire",
+      "▁L ee",
+      "▁Le e",
+      "▁ Lee",
+      "尽 量",
+      "reg ion",
+      "to r",
+      "t or",
+      "一 边",
+      "if iers",
+      "ifier s",
+      "ifi ers",
+      "▁B oston",
+      "▁Bo ston",
+      "▁Bos ton",
+      "▁particip ate",
+      "区 人民检察院",
+      "an cel",
+      "ance l",
+      "anc el",
+      "有限 公司于",
+      "有限公司 于",
+      "频 率",
+      "▁N av",
+      "▁Na v",
+      "▁ Nav",
+      "熟 悉",
+      "公司 经营范围包括",
+      ") [",
+      "▁aim ed",
+      "▁ai med",
+      "▁ab normal",
+      "▁abnorm al",
+      "▁b each",
+      "▁be ach",
+      "一 路",
+      "Mu lt",
+      "M ult",
+      "户 籍",
+      "发 的",
+      "▁h idden",
+      "▁ hidden",
+      "el a",
+      "e la",
+      "nd es",
+      "nde s",
+      "n des",
+      "▁man ual",
+      "▁ manual",
+      "▁prov iders",
+      "▁provid ers",
+      "▁provide rs",
+      "▁provider s",
+      "▁ providers",
+      "综 上",
+      "并 在",
+      "▁diagn ostic",
+      "▁As sembly",
+      "▁ Assembly",
+      "深 度",
+      "AC E",
+      "A CE",
+      "▁System s",
+      "▁L td",
+      "▁Lt d",
+      "数 组",
+      "▁e ld",
+      "▁el d",
+      "▁ eld",
+      "G e",
+      "▁construct ed",
+      "诉讼 费用",
+      "诉讼费 用",
+      "便 宜",
+      "▁different iation",
+      "▁pro fit",
+      "▁prof it",
+      "▁ profit",
+      "▁d rivers",
+      "▁dr ivers",
+      "▁dri vers",
+      "▁driv ers",
+      "▁drive rs",
+      "▁driver s",
+      "▁ drivers",
+      "uc ks",
+      "uck s",
+      "u cks",
+      "▁aw esome",
+      "▁awe some",
+      "▁ awesome",
+      "另一 个",
+      "另 一个",
+      "co ver",
+      "cov er",
+      "c over",
+      "” ,",
+      "▁H elp",
+      "▁He lp",
+      "▁Hel p",
+      "▁ Help",
+      "一 审",
+      "ad ium",
+      "adi um",
+      "诉 称",
+      "T arget",
+      "▁as signed",
+      "▁ass igned",
+      "▁assign ed",
+      "▁ assigned",
+      "的 设计",
+      "▁sc heduled",
+      "▁schedul ed",
+      "▁schedule d",
+      "▁ scheduled",
+      "nA r",
+      "n Ar",
+      "En c",
+      "E nc",
+      "时 光",
+      "▁C atholic",
+      "▁Cath olic",
+      "DN A",
+      "D NA",
+      "▁J im",
+      "▁Ji m",
+      "▁ Jim",
+      "▁m es",
+      "▁me s",
+      "▁ mes",
+      "SE R",
+      "S ER",
+      "云南 省",
+      "云 南省",
+      "▁s pending",
+      "▁sp ending",
+      "▁spend ing",
+      "复 合",
+      "▁p article",
+      "▁part icle",
+      "▁partic le",
+      "▁ particle",
+      "人 间",
+      "的基础 上",
+      "的 基础上",
+      "▁f unc",
+      "▁fun c",
+      "▁fu nc",
+      "▁ func",
+      "▁p and",
+      "▁pan d",
+      "▁pa nd",
+      "▁ pand",
+      "▁R est",
+      "▁Re st",
+      "▁Res t",
+      "▁ Rest",
+      "驱 动",
+      "▁de pos",
+      "▁dep os",
+      "毒 品",
+      "ge rs",
+      "ger s",
+      "g ers",
+      "▁Col umb",
+      "▁dist ribut",
+      "{ (",
+      "▁ent ered",
+      "▁enter ed",
+      "de g",
+      "d eg",
+      "rol ling",
+      "roll ing",
+      "bl ank",
+      "bla nk",
+      "森 林",
+      "国 外",
+      "▁h ousing",
+      "▁hous ing",
+      "▁ho using",
+      "▁f ruit",
+      "▁fr uit",
+      "▁ fruit",
+      "oid s",
+      "oi ds",
+      "o ids",
+      "安 市",
+      "字 符",
+      "▁ill ustr",
+      "▁illust r",
+      "m o",
+      "支 撑",
+      "而不 是",
+      "而 不是",
+      "med iate",
+      "media te",
+      "不 上",
+      "▁e c",
+      "▁ ec",
+      "▁w alking",
+      "▁walk ing",
+      "▁wal king",
+      "sc ore",
+      "s core",
+      "▁inv olves",
+      "▁invol ves",
+      "▁involve s",
+      "▁s essions",
+      "▁session s",
+      "▁sess ions",
+      "▁ sessions",
+      "▁l ayout",
+      "▁la yout",
+      "▁lay out",
+      "▁ layout",
+      "▁P DF",
+      "▁PD F",
+      "▁ PDF",
+      "▁a id",
+      "▁ai d",
+      "▁ aid",
+      "▁tr ading",
+      "▁trad ing",
+      "▁tra ding",
+      "▁m erg",
+      "▁me rg",
+      "▁mer g",
+      "说 法",
+      "▁re pository",
+      "▁rep ository",
+      "▁repos itory",
+      "▁ repository",
+      "不 太",
+      "ag ger",
+      "agg er",
+      "▁R o",
+      "▁ Ro",
+      "▁p m",
+      "▁ pm",
+      "Ma n",
+      "M an",
+      "op erator",
+      "oper ator",
+      "re me",
+      "rem e",
+      "r eme",
+      "▁des ired",
+      "▁desire d",
+      "▁un defined",
+      "▁u ndefined",
+      "▁und efined",
+      "▁ undefined",
+      "op eration",
+      "ope ration",
+      "oper ation",
+      "▁anal yze",
+      "▁analy ze",
+      "▁analyz e",
+      "▁pl anned",
+      "▁plan ned",
+      "▁ planned",
+      "博物 馆",
+      "▁pro gression",
+      "▁progress ion",
+      "塑 料",
+      "vid eo",
+      "v ideo",
+      "nR E",
+      "n RE",
+      "su r",
+      "s ur",
+      "ul ator",
+      "ula tor",
+      "▁s erves",
+      "▁serv es",
+      "▁ser ves",
+      "▁serve s",
+      "met adata",
+      "meta data",
+      "▁c orporate",
+      "▁corpor ate",
+      "现 住",
+      "▁complex ity",
+      "▁con version",
+      "▁co nversion",
+      "▁conv ersion",
+      "▁convers ion",
+      "▁ conversion",
+      "分 解",
+      "理 事",
+      "▁P lan",
+      "▁Pl an",
+      "▁ Plan",
+      "调解 书",
+      "▁rec ording",
+      "▁record ing",
+      "▁insp ired",
+      "▁inspire d",
+      "▁inspir ed",
+      "▁ inspired",
+      "e z",
+      "轻 处罚",
+      "( `",
+      "▁u tility",
+      "▁util ity",
+      "▁ut ility",
+      "▁ utility",
+      "ge nce",
+      "gen ce",
+      "g ence",
+      "pro v",
+      "pr ov",
+      "p rov",
+      "层 次",
+      "破 坏",
+      "for each",
+      "fore ach",
+      "fo reach",
+      "随 时",
+      "re cord",
+      "rec ord",
+      "dir name",
+      "▁p ump",
+      "▁pu mp",
+      "eps ilon",
+      "e psilon",
+      "顾 客",
+      "oi r",
+      "o ir",
+      "▁it er",
+      "▁i ter",
+      "▁ iter",
+      "mer ce",
+      "merc e",
+      "▁g un",
+      "▁gu n",
+      "▁ gun",
+      "el ls",
+      "ell s",
+      "▁l ies",
+      "▁li es",
+      "▁lie s",
+      "▁ lies",
+      "借 条",
+      "▁E con",
+      "▁Ec on",
+      "▁Eco n",
+      "▁ Econ",
+      "▁N Y",
+      "▁ NY",
+      "Loc ation",
+      "Lo cation",
+      "L ocation",
+      "贯 彻",
+      "▁put ting",
+      "ic ker",
+      "ick er",
+      "fer ences",
+      "ference s",
+      "▁com mission",
+      "▁comm ission",
+      "▁c heap",
+      "▁che ap",
+      "Ma in",
+      "M ain",
+      "▁} }\\",
+      "▁}} \\",
+      "▁ }}\\",
+      "amp ion",
+      "Re cord",
+      "Rec ord",
+      "oc us",
+      "o cus",
+      "▁cap ture",
+      "▁capt ure",
+      "▁ capture",
+      "拼 音",
+      "▁A L",
+      "▁ AL",
+      "重 复",
+      "让我 们",
+      "让 我们",
+      "杂 志",
+      "同 年",
+      "的一 些",
+      "的 一些",
+      "中国 科学院",
+      "▁定 价",
+      "▁ 定价",
+      "人 均",
+      "▁H ave",
+      "▁Ha ve",
+      "▁Hav e",
+      "▁ Have",
+      "Par ams",
+      "Param s",
+      "buf fer",
+      "buff er",
+      "b uffer",
+      "aw a",
+      "a wa",
+      "▁l ose",
+      "▁lo se",
+      "▁los e",
+      "▁ lose",
+      "▁manufact uring",
+      "▁m amm",
+      "▁ma mm",
+      "▁o ngoing",
+      "▁on going",
+      "▁init ially",
+      "▁initial ly",
+      "▁initi ally",
+      "Fi rst",
+      "F irst",
+      "B L",
+      "蔬 菜",
+      "cl i",
+      "c li",
+      "▁anal yt",
+      "▁analy t",
+      "sh ort",
+      "打 车",
+      "对 外",
+      "▁ed ited",
+      "▁edit ed",
+      "▁ edited",
+      "con structor",
+      "construct or",
+      "▁gen us",
+      "▁ge nus",
+      "▁genu s",
+      "▁sup posed",
+      "▁suppose d",
+      "▁其 一",
+      "▁ 其一",
+      "日 至",
+      "▁polit ician",
+      "别 的",
+      "▁log ic",
+      "▁lo gic",
+      "▁ logic",
+      "▁bo ost",
+      "▁ boost",
+      "▁S ar",
+      "▁Sa r",
+      "ight er",
+      "igh ter",
+      "Mat ch",
+      "M atch",
+      "tp rint",
+      "t print",
+      "poss ible",
+      "p ossible",
+      "▁l ists",
+      "▁li sts",
+      "▁list s",
+      "▁ lists",
+      "▁gluc ose",
+      "▁ glucose",
+      "▁out door",
+      "▁t hin",
+      "▁th in",
+      "▁thi n",
+      "▁ thin",
+      "▁optim ization",
+      "▁M i",
+      "▁ Mi",
+      "▁c opies",
+      "▁co pies",
+      "▁cop ies",
+      "▁trans lation",
+      "▁transl ation",
+      "▁ translation",
+      "家 具",
+      "▁s cal",
+      "▁sc al",
+      "▁ scal",
+      "期间的 债务利息",
+      "o a",
+      "е н",
+      "有效 的",
+      "乐 部",
+      "▁char ged",
+      "▁charg ed",
+      "▁charge d",
+      "▁ charged",
+      "图 像",
+      "in ois",
+      "ino is",
+      "Em ail",
+      "E mail",
+      "▁B ul",
+      "▁Bu l",
+      "▁ Bul",
+      "▁tri angle",
+      "▁triang le",
+      "▁ triangle",
+      "▁les ions",
+      "▁lesion s",
+      "▁c rim",
+      "▁cr im",
+      "▁cont ained",
+      "▁contain ed",
+      "▁ contained",
+      "▁rem oval",
+      "▁remov al",
+      "be c",
+      "b ec",
+      "验 证",
+      "线 上",
+      "▁turn s",
+      "▁tur ns",
+      "▁r adi",
+      "▁rad i",
+      "▁ra di",
+      "▁pre paration",
+      "▁prepar ation",
+      "▁bu ying",
+      "▁buy ing",
+      "▁Ap plication",
+      "▁Appl ication",
+      "▁ Application",
+      "ib ilities",
+      "▁M ichigan",
+      "▁Mich igan",
+      "he tic",
+      "het ic",
+      "h etic",
+      "F S",
+      "set up",
+      "ns ylvania",
+      "met ry",
+      "me try",
+      "m etry",
+      "▁ver tical",
+      "▁vert ical",
+      "▁ vertical",
+      "up load",
+      "u pload",
+      "▁guid elines",
+      "▁guide lines",
+      "▁guideline s",
+      "▁ex istence",
+      "▁exist ence",
+      "▁ existence",
+      "▁cont roller",
+      "▁control ler",
+      "▁ controller",
+      "sm all",
+      "s mall",
+      "Th at",
+      "T hat",
+      "样 例",
+      "▁class ical",
+      "▁classic al",
+      "▁B ob",
+      "▁Bo b",
+      "▁ Bob",
+      "你 要",
+      "gi an",
+      "gia n",
+      "g ian",
+      "li k",
+      "l ik",
+      "▁G irl",
+      "▁Gi rl",
+      "▁Gir l",
+      "▁ Girl",
+      "ester day",
+      "▁B ud",
+      "▁Bu d",
+      "▁ Bud",
+      "▁M O",
+      "▁ MO",
+      "▁p orn",
+      "▁po rn",
+      "▁por n",
+      "or ship",
+      "ors hip",
+      "▁tr uth",
+      "▁ truth",
+      "ch em",
+      "che m",
+      "c hem",
+      "li ve",
+      "l ive",
+      "▁or ders",
+      "▁order s",
+      "▁ord ers",
+      "▁ orders",
+      "俱 乐部",
+      "nG e",
+      "n Ge",
+      "▁Ch ief",
+      "▁Chi ef",
+      "▁ Chief",
+      "nTr ue",
+      "n True",
+      "oc yte",
+      "ocyt e",
+      "o cyte",
+      "▁em ployed",
+      "▁employ ed",
+      "足 以",
+      "▁k g",
+      "▁ kg",
+      "▁cont ents",
+      "▁content s",
+      "▁ contents",
+      "P Y",
+      "删 除",
+      "▁thought s",
+      "▁though ts",
+      "ip h",
+      "i ph",
+      "如果 是",
+      "▁al ign",
+      "▁ali gn",
+      "▁ align",
+      "▁ 平",
+      "会 在",
+      "VA L",
+      "V AL",
+      "Loc al",
+      "Lo cal",
+      "L ocal",
+      "▁c ache",
+      "▁ca che",
+      "▁ cache",
+      "pr es",
+      "pre s",
+      "p res",
+      "不 动",
+      "件 事",
+      "Sh ow",
+      "S how",
+      "ist ence",
+      "iste nce",
+      "▁f ear",
+      "▁fe ar",
+      "▁s emi",
+      "▁se mi",
+      "▁sem i",
+      "现 有",
+      "▁mus ical",
+      "▁music al",
+      "▁f ellow",
+      "▁fell ow",
+      "▁fel low",
+      "▁com mands",
+      "▁comm ands",
+      "▁command s",
+      "▁ commands",
+      "▁p airs",
+      "▁pair s",
+      "▁pa irs",
+      "▁ pairs",
+      "均 匀",
+      "首 次",
+      "st ud",
+      "破 裂",
+      "▁r i",
+      "▁ ri",
+      "是一 家",
+      "是 一家",
+      "▁en coding",
+      "▁enc oding",
+      "▁ encoding",
+      "▁m ilk",
+      "▁mil k",
+      "提 醒",
+      "交通 事故",
+      "▁s we",
+      "▁sw e",
+      "▁L ike",
+      "▁Li ke",
+      "▁Lik e",
+      "▁ Like",
+      "▁N ode",
+      "▁No de",
+      "▁ Node",
+      "▁w illing",
+      "▁will ing",
+      "off set",
+      "offs et",
+      "SE T",
+      "S ET",
+      "共 计",
+      "▁D eterm",
+      "▁De term",
+      "▁Det erm",
+      "▁ Determ",
+      "处 分",
+      "▁prot ected",
+      "▁protect ed",
+      "▁ protected",
+      "V A",
+      "起点 中文网",
+      "l g",
+      "▁cons umer",
+      "▁consum er",
+      "▁consume r",
+      "▁ consumer",
+      "no f",
+      "n of",
+      "贵州 省",
+      "ri ers",
+      "rier s",
+      "rie rs",
+      "r iers",
+      "最 低",
+      "本院 依法组成合议庭",
+      "本院依法 组成合议庭",
+      "生 成",
+      "u j",
+      "pon s",
+      "po ns",
+      "p ons",
+      "Ex t",
+      "E xt",
+      "▁d ram",
+      "▁dr am",
+      "▁dis played",
+      "▁display ed",
+      "肿 瘤",
+      "nP ost",
+      "nPo st",
+      "n Post",
+      "▁C irc",
+      "▁Ci rc",
+      "▁Cir c",
+      "▁ Circ",
+      "▁ind icates",
+      "▁indic ates",
+      "▁indicate s",
+      "▁Sup port",
+      "▁ Support",
+      "▁O ption",
+      "▁Opt ion",
+      "▁Op tion",
+      "▁ Option",
+      "pr imary",
+      "prim ary",
+      "eg ers",
+      "eger s",
+      "e gers",
+      "▁公司 名称",
+      "国 民",
+      "货 物",
+      "无 人",
+      "contin ue",
+      "返 回",
+      "▁report ing",
+      "▁ reporting",
+      "wh en",
+      "w hen",
+      "▁observ ations",
+      "▁observation s",
+      "▁th reshold",
+      "▁ threshold",
+      "国 务",
+      "花 园",
+      "零 售",
+      "exec ute",
+      "▁w orry",
+      "▁wor ry",
+      "意 味",
+      "Ro w",
+      "R ow",
+      "Ent ry",
+      "En try",
+      "E ntry",
+      "印 象",
+      "ir ms",
+      "irm s",
+      "ub er",
+      "ube r",
+      "u ber",
+      "▁f ee",
+      "▁fe e",
+      "▁ fee",
+      "fol io",
+      "fo lio",
+      "▁di ameter",
+      "▁dia meter",
+      "▁hundred s",
+      "缓 刑",
+      "▁r ational",
+      "▁rat ional",
+      "▁ratio nal",
+      "▁ration al",
+      "cn t",
+      "c nt",
+      "oc iated",
+      "oci ated",
+      "▁F ed",
+      "▁Fe d",
+      "▁W ord",
+      "▁Wor d",
+      "▁Wo rd",
+      "▁ Word",
+      "哲 学",
+      "深 刻",
+      "group s",
+      "gro ups",
+      "▁antib odies",
+      "涉 嫌",
+      "▁l anguages",
+      "▁language s",
+      "▁ languages",
+      "en dor",
+      "end or",
+      "endo r",
+      "执 法",
+      "▁S C",
+      "▁ SC",
+      "▁s train",
+      "▁st rain",
+      "▁str ain",
+      "▁ strain",
+      "ype d",
+      "yp ed",
+      "y ped",
+      "▁off set",
+      "▁ offset",
+      "院 长",
+      "▁H E",
+      "▁ HE",
+      "▁A rea",
+      "▁Ar ea",
+      "▁Are a",
+      "▁ Area",
+      "▁l ived",
+      "▁li ved",
+      "▁live d",
+      "▁liv ed",
+      "▁ lived",
+      "ox y",
+      "o xy",
+      "▁d imensions",
+      "▁dim ensions",
+      "▁dimension s",
+      "▁ dimensions",
+      "▁s ections",
+      "▁section s",
+      "▁ sections",
+      "▁c ounc",
+      "▁co unc",
+      "▁cou nc",
+      "ba n",
+      "b an",
+      "述 了",
+      "▁b rown",
+      "▁br own",
+      "▁bro wn",
+      "▁brow n",
+      "▁ brown",
+      "par e",
+      "pa re",
+      "p are",
+      "nt hetic",
+      "nthe tic",
+      "nth etic",
+      "nthet ic",
+      "st one",
+      "ston e",
+      "sto ne",
+      "s tone",
+      "Re nder",
+      "Ren der",
+      "R ender",
+      "B undle",
+      "▁l ux",
+      "▁lu x",
+      "▁ lux",
+      "UN D",
+      "U ND",
+      "▁W ild",
+      "▁Wil d",
+      "▁Wi ld",
+      "▁ Wild",
+      "р а",
+      "▁o pinion",
+      "▁opin ion",
+      "高 于",
+      "▁} );\\",
+      "▁}) ;\\",
+      "▁ });\\",
+      "样 子",
+      "▁\" ,",
+      "▁ \",",
+      "▁E r",
+      "▁ Er",
+      "▁Engine ering",
+      "▁Engineer ing",
+      "▁p in",
+      "▁pi n",
+      "▁ pin",
+      "ig ate",
+      "iga te",
+      "i gate",
+      "▁a er",
+      "监 规",
+      "在这 个",
+      "在 这个",
+      "刚 刚",
+      "▁C EO",
+      "▁CE O",
+      "▁ CEO",
+      "▁qu adr",
+      "▁quad r",
+      "鼓 励",
+      "go v",
+      "g ov",
+      "Ne xt",
+      "N ext",
+      "运动 员",
+      "运 动员",
+      "er ves",
+      "erv es",
+      "erve s",
+      "连载 于",
+      "▁W ay",
+      "▁Wa y",
+      "▁ Way",
+      "ap pe",
+      "app e",
+      "▁loc ally",
+      "▁local ly",
+      "的 感觉",
+      "互 相",
+      "▁\" $",
+      "▁ \"$",
+      "▁represent ing",
+      "流 行",
+      "▁d ual",
+      "▁du al",
+      "▁m apping",
+      "▁map ping",
+      "▁ma pping",
+      "▁ mapping",
+      "Sh e",
+      "S he",
+      "En ron",
+      "\": {\"",
+      "▁g it",
+      "▁gi t",
+      "▁ git",
+      "▁abs ence",
+      "▁com posite",
+      "▁compos ite",
+      "▁f ly",
+      "▁fl y",
+      "▁ fly",
+      "To p",
+      "T op",
+      "物业 管理有限公司",
+      "物业管理 有限公司",
+      "▁a ds",
+      "▁ad s",
+      "▁ ads",
+      "D M",
+      "▁L ab",
+      "▁La b",
+      "▁ Lab",
+      "PA SS",
+      "P ASS",
+      "th at",
+      "tha t",
+      "t hat",
+      "即 将",
+      "这个 问题的答案",
+      "这个问题 的答案",
+      "▁met abolism",
+      "▁metabol ism",
+      "▁t aste",
+      "▁ta ste",
+      "▁tast e",
+      "Re l",
+      "R el",
+      "▁V an",
+      "▁Va n",
+      "▁ Van",
+      "ot ox",
+      "oto x",
+      "o tox",
+      "be ll",
+      "bel l",
+      "b ell",
+      "▁G ive",
+      "▁Gi ve",
+      "▁ Give",
+      "▁T oday",
+      "▁To day",
+      "▁ Today",
+      "ca m",
+      "c am",
+      "ane ously",
+      "aneous ly",
+      "就 有",
+      "故 障",
+      "▁col umns",
+      "▁column s",
+      "▁colum ns",
+      "▁ columns",
+      "悔 罪",
+      "si s",
+      "s is",
+      "▁Th rough",
+      "▁Thr ough",
+      "▁ Through",
+      "▁In put",
+      "▁ Input",
+      "▁r om",
+      "▁ro m",
+      "▁ rom",
+      "▁F amily",
+      "▁ Family",
+      "▁concer n",
+      "▁conc ern",
+      "▁conce rn",
+      "▁watch ing",
+      "▁wat ching",
+      "▁c abin",
+      "▁ca bin",
+      "▁cab in",
+      ", ,",
+      "行 驶",
+      "▁re new",
+      "▁ renew",
+      "(( )",
+      "( ()",
+      "e h",
+      "千 米",
+      "▁met ast",
+      "▁meta st",
+      "▁vis itors",
+      "▁visit ors",
+      "▁visitor s",
+      "喜欢 的",
+      "vi n",
+      "v in",
+      "▁F ood",
+      "▁Foo d",
+      "▁Fo od",
+      "▁ Food",
+      "Dial og",
+      "D ialog",
+      "看 见",
+      "▁N ext",
+      "▁Ne xt",
+      "▁ Next",
+      "专 辑",
+      "▁S E",
+      "▁ SE",
+      "进 而",
+      "无 限",
+      "▁bl ank",
+      "▁ blank",
+      "G O",
+      "▁order ed",
+      "▁ord ered",
+      "▁ ordered",
+      "▁de pression",
+      "▁dep ression",
+      "写 的",
+      "▁d ance",
+      "▁da nce",
+      "▁dan ce",
+      "难 度",
+      "▁ 演员",
+      "我 说",
+      "档 案",
+      "黑龙江 省",
+      "调 研",
+      "▁Pol ice",
+      "▁p regnancy",
+      "▁pregn ancy",
+      "ol ver",
+      "olve r",
+      "olv er",
+      "▁ch apter",
+      "▁chap ter",
+      "▁ chapter",
+      "▁function ality",
+      "▁functional ity",
+      "实 数",
+      "▁A C",
+      "▁ AC",
+      "▁as sign",
+      "▁ass ign",
+      "▁ assign",
+      "NE T",
+      "N ET",
+      "▁其 二",
+      "▁a ctor",
+      "▁act or",
+      "▁ac tor",
+      "▁ actor",
+      "ou n",
+      "o un",
+      "▁den omin",
+      "▁small est",
+      "对 手",
+      "▁mat ching",
+      "▁match ing",
+      "▁ matching",
+      "▁M id",
+      "▁Mi d",
+      "▁ Mid",
+      "vir tual",
+      "virt ual",
+      "▁int roduction",
+      "▁introdu ction",
+      "▁intro duction",
+      "▁ introduction",
+      "▁J eff",
+      "▁Je ff",
+      "▁ Jeff",
+      "正 方",
+      "▁b tn",
+      "▁bt n",
+      "▁ btn",
+      "清 晰",
+      "▁I V",
+      "▁ IV",
+      "fl u",
+      "f lu",
+      "▁Is rael",
+      "▁Isra el",
+      "两 次",
+      "第五 百",
+      "第 五百",
+      "▁J ud",
+      "▁Ju d",
+      "这个问题的答案 是",
+      "▁cre ates",
+      "▁create s",
+      "▁creat es",
+      "▁see king",
+      "▁seek ing",
+      "预 计",
+      "▁g ap",
+      "▁ga p",
+      "▁ gap",
+      "画 面",
+      "▁t ickets",
+      "▁ticket s",
+      "▁tick ets",
+      "in v",
+      "i nv",
+      "▁mut ations",
+      "▁mutation s",
+      "UR E",
+      "U RE",
+      "▁in n",
+      "▁i nn",
+      "▁ inn",
+      "▁ 基本",
+      "▁D ie",
+      "▁Di e",
+      "▁ Die",
+      "As ync",
+      "A sync",
+      "王 某",
+      "▁ 元",
+      "▁Com mon",
+      "▁Comm on",
+      "▁ Common",
+      "re ction",
+      "rec tion",
+      "rect ion",
+      "r ection",
+      "住 宅",
+      "▁p apers",
+      "▁paper s",
+      "▁pa pers",
+      "▁pap ers",
+      "▁ papers",
+      "port er",
+      "por ter",
+      "p orter",
+      "可 知",
+      "we ek",
+      "w eek",
+      "▁man ip",
+      "▁ma nip",
+      "pan d",
+      "pa nd",
+      "p and",
+      "op las",
+      "opl as",
+      "步 行",
+      "▁d esc",
+      "▁de sc",
+      "▁des c",
+      "▁ desc",
+      "P V",
+      "go t",
+      "g ot",
+      "▁tot ally",
+      "▁total ly",
+      "名 为",
+      "减刑 建议",
+      "人 群",
+      "实际 上",
+      "od o",
+      "o do",
+      "▁in duction",
+      "▁ind uction",
+      "▁indu ction",
+      "▁induct ion",
+      "W N",
+      "return s",
+      "S m",
+      "风 景",
+      "B M",
+      "ns el",
+      "nse l",
+      "n sel",
+      "migration s",
+      "m igrations",
+      "收 集",
+      "▁abs olutely",
+      "▁absol utely",
+      "▁absolute ly",
+      "▁any where",
+      "us ep",
+      "use p",
+      "u sep",
+      "的 热",
+      "▁G raph",
+      "▁Gr aph",
+      "▁Gra ph",
+      "▁ Graph",
+      "▁m and",
+      "▁man d",
+      "▁ma nd",
+      "▁ mand",
+      "wa it",
+      "w ait",
+      "▁S em",
+      "▁Se m",
+      "▁ Sem",
+      "Nu m",
+      "N um",
+      "功 率",
+      "ig inally",
+      "iginal ly",
+      "igin ally",
+      "▁J ones",
+      "▁Jo nes",
+      "▁Jon es",
+      "▁ Jones",
+      "ag ent",
+      "age nt",
+      "agen t",
+      "a gent",
+      "▁inj uries",
+      "▁Ind ex",
+      "▁ Index",
+      "re nt",
+      "ren t",
+      "r ent",
+      "▁new sp",
+      "▁news p",
+      "▁p ermission",
+      "▁per mission",
+      "▁perm ission",
+      "▁ permission",
+      "▁re ferred",
+      "▁refer red",
+      "的 起诉",
+      "研究 所",
+      "飞 机",
+      "不需 要",
+      "不 需要",
+      "▁M ic",
+      "▁Mi c",
+      "▁webs ites",
+      "▁website s",
+      "▁web sites",
+      "G S",
+      "简单 的",
+      "▁B ang",
+      "▁Ba ng",
+      "▁Ban g",
+      "K e",
+      "地 的",
+      "▁Stud ies",
+      "▁h abit",
+      "▁hab it",
+      "▁ha bit",
+      "迟延履行 期间的债务利息",
+      "展 现",
+      "ew ork",
+      "e work",
+      "AR Y",
+      "A RY",
+      "▁y ellow",
+      "▁ yellow",
+      "▁S outhern",
+      "▁South ern",
+      "▁Sou thern",
+      "▁b attle",
+      "▁bat tle",
+      "▁batt le",
+      "▁ba ttle",
+      "▁per forming",
+      "▁perform ing",
+      "▁perfor ming",
+      "▁v iral",
+      "▁vir al",
+      "▁vi ral",
+      "▁ viral",
+      "▁acqu ired",
+      "▁acquire d",
+      "对 罪犯",
+      "▁W eek",
+      "▁We ek",
+      "▁ Week",
+      "▁comp ute",
+      "▁comput e",
+      "▁ compute",
+      "ort unately",
+      "▁e xit",
+      "▁ex it",
+      "▁ exit",
+      "标 志",
+      "▁ в",
+      "AS E",
+      "A SE",
+      "孩子 的",
+      "孩 子的",
+      "▁cruc ial",
+      "▁the rap",
+      "▁ther ap",
+      "data base",
+      "dat abase",
+      "datab ase",
+      "d atabase",
+      "local host",
+      "st orage",
+      "sto rage",
+      "stor age",
+      "哈 尔",
+      "hp p",
+      "h pp",
+      "on to",
+      "ont o",
+      "o nto",
+      "减去 有期徒刑",
+      "▁s qrt",
+      "▁sq rt",
+      "▁ sqrt",
+      "▁b ridge",
+      "▁br idge",
+      "▁brid ge",
+      "▁ bridge",
+      "小说 网",
+      "fl ex",
+      "f lex",
+      "告诉 我",
+      "▁s inger",
+      "▁sing er",
+      "▁sin ger",
+      "▁act ress",
+      "条 文",
+      "Na v",
+      "N av",
+      "小 于",
+      "尽 管",
+      "ch arge",
+      "char ge",
+      "charg e",
+      "▁as sets",
+      "▁ass ets",
+      "▁asset s",
+      "▁ assets",
+      "清 单",
+      "▁D ownload",
+      "▁Down load",
+      "▁ Download",
+      "电 器",
+      "▁L ive",
+      "▁Li ve",
+      "▁Liv e",
+      "▁ Live",
+      "sec ret",
+      "中 有",
+      "平 安",
+      "TY PE",
+      "T YPE",
+      "流 动",
+      "g b",
+      "no ne",
+      "non e",
+      "n one",
+      "li e",
+      "l ie",
+      "▁F ull",
+      "▁Fu ll",
+      "▁Ful l",
+      "▁ Full",
+      "la test",
+      "late st",
+      "lat est",
+      "l atest",
+      "▁Re port",
+      "▁Rep ort",
+      "▁ Report",
+      "你 是",
+      "到 达",
+      "今 日",
+      "创 意",
+      "kn ow",
+      "k now",
+      "▁rec ruit",
+      "▁recru it",
+      "oc ated",
+      "ocate d",
+      "oca ted",
+      "▁en counter",
+      "▁enc ounter",
+      "云 南",
+      "军 事",
+      "▁mod ules",
+      "▁module s",
+      "▁ modules",
+      "▁g ran",
+      "▁gr an",
+      "▁gra n",
+      "▁ gran",
+      "十七 条",
+      "十 七条",
+      "▁b igger",
+      "▁big ger",
+      "▁stud ying",
+      "▁study ing",
+      "▁low est",
+      "▁lo west",
+      "不 出",
+      "考 察",
+      "▁T ask",
+      "▁Ta sk",
+      "▁Tas k",
+      "▁ Task",
+      "Ho st",
+      "H ost",
+      "▁A wards",
+      "▁Award s",
+      "▁Aw ards",
+      "use package",
+      "usep ackage",
+      "fu ture",
+      "f uture",
+      "个 体",
+      "▁Co ast",
+      "▁b ench",
+      "▁be nch",
+      "▁ben ch",
+      "▁ bench",
+      "▁v isible",
+      "▁vis ible",
+      "▁ visible",
+      "▁O x",
+      "▁ Ox",
+      "pk g",
+      "p kg",
+      "地 下",
+      "身 高",
+      "▁; \\",
+      "▁ ;\\",
+      "▁G ar",
+      "▁Ga r",
+      "▁int roduce",
+      "▁introdu ce",
+      "工 人",
+      "~~ ~~",
+      "~~~ ~",
+      "~ ~~~",
+      "专 利",
+      "损 害",
+      "ea st",
+      "eas t",
+      "e ast",
+      "▁th ermal",
+      "▁ther mal",
+      "▁therm al",
+      "▁ thermal",
+      "岁 的",
+      "bit r",
+      "bi tr",
+      "b itr",
+      "▁s ugar",
+      "▁su gar",
+      "▁sug ar",
+      "▁inf ections",
+      "▁infection s",
+      "▁infect ions",
+      "而 已",
+      "Lo ng",
+      "L ong",
+      "西 方",
+      "▁Res ponse",
+      "▁Respon se",
+      "▁Respons e",
+      "▁ Response",
+      "▁log in",
+      "▁lo gin",
+      "▁ login",
+      "▁c odes",
+      "▁co des",
+      "▁code s",
+      "▁cod es",
+      "▁ codes",
+      "是 以",
+      "济 南",
+      "micro soft",
+      "m icrosoft",
+      "分 局",
+      "ri ck",
+      "ric k",
+      "r ick",
+      "▁F orce",
+      "▁For ce",
+      "▁ Force",
+      "几 天",
+      "▁Lin ux",
+      "▁ Linux",
+      "▁mag azine",
+      "作 家",
+      "or i",
+      "o ri",
+      "▁b on",
+      "▁bo n",
+      "▁ bon",
+      "的 行为",
+      "实 习",
+      "任 意",
+      "▁P o",
+      "▁ Po",
+      "长 沙",
+      "la y",
+      "l ay",
+      "市 民",
+      "▁S ource",
+      "▁ Source",
+      "警 察",
+      "▁ab use",
+      "告 知",
+      "▁m ob",
+      "▁mo b",
+      "▁ mob",
+      "占 地",
+      "B R",
+      "▁F und",
+      "▁Fun d",
+      "▁Fu nd",
+      "▁ Fund",
+      "▁ •",
+      "▁w al",
+      "▁wa l",
+      "▁ wal",
+      "pro perties",
+      "proper ties",
+      "所 得",
+      "ef ully",
+      "eful ly",
+      "e fully",
+      "清 洗",
+      "▁St ation",
+      "▁Stat ion",
+      "▁Sta tion",
+      "▁ Station",
+      "▁sugg esting",
+      "▁suggest ing",
+      "il ib",
+      "ili b",
+      "i lib",
+      "▁st roke",
+      "▁str oke",
+      "▁stro ke",
+      "▁ stroke",
+      "变 量",
+      "as i",
+      "a si",
+      "ec raft",
+      "e craft",
+      "C s",
+      "▁sup erior",
+      "▁super ior",
+      "an do",
+      "and o",
+      "a ndo",
+      "▁apprec iate",
+      "有 时",
+      "▁health care",
+      "脂 肪",
+      "低 于",
+      "nJ ust",
+      "n Just",
+      "▁the oretical",
+      "▁theore tical",
+      "路 上",
+      "▁P ART",
+      "▁PA RT",
+      "▁PAR T",
+      "▁ PART",
+      "石 油",
+      "爱 好",
+      "日 报",
+      "L Y",
+      "ho l",
+      "h ol",
+      "ve ly",
+      "vel y",
+      "v ely",
+      "▁c ritic",
+      "▁cr itic",
+      "▁crit ic",
+      "▁w inner",
+      "▁win ner",
+      "▁ winner",
+      "广州 市",
+      "广 州市",
+      "nC ol",
+      "nCo l",
+      "n Col",
+      "▁C reat",
+      "▁Cr eat",
+      "▁Cre at",
+      "▁ Creat",
+      "长 方",
+      "Q U",
+      "▁cons umers",
+      "▁consum ers",
+      "▁consumer s",
+      "▁consume rs",
+      "设 定",
+      "边 的",
+      "每个 人",
+      "每 个人",
+      "▁s ingles",
+      "▁sing les",
+      "▁single s",
+      "nC ON",
+      "nCO N",
+      "n CON",
+      "▁a ircraft",
+      "▁air craft",
+      "是 最",
+      "▁ro ots",
+      "▁root s",
+      "▁ roots",
+      "ME NT",
+      "M ENT",
+      "▁b roadcast",
+      "▁broad cast",
+      "▁ broadcast",
+      "te en",
+      "tee n",
+      "t een",
+      "ee n",
+      "e en",
+      "▁an onymous",
+      "▁anonym ous",
+      "▁ anonymous",
+      "▁o ste",
+      "▁os te",
+      "▁ost e",
+      "to ns",
+      "ton s",
+      "t ons",
+      "▁j u",
+      "▁ ju",
+      "Mat h",
+      "Ma th",
+      "M ath",
+      "k t",
+      "▁M ember",
+      "▁Me mber",
+      "▁Mem ber",
+      "▁ Member",
+      "▁comp ens",
+      "▁cl ust",
+      "行政 处罚",
+      "▁M ike",
+      "▁Mi ke",
+      "▁Mik e",
+      "▁ Mike",
+      "▁H ttp",
+      "▁ Http",
+      "} `",
+      "办 事",
+      "实验 室",
+      "▁ent ertain",
+      "▁enter tain",
+      "▁entert ain",
+      "▁tr acks",
+      "▁track s",
+      "▁tra cks",
+      "▁ tracks",
+      "De b",
+      "D eb",
+      "na n",
+      "n an",
+      "美 好",
+      "▁Pen nsylvania",
+      "Test s",
+      "Te sts",
+      "T ests",
+      "将 其",
+      "▁s outhern",
+      "▁sou thern",
+      "▁south ern",
+      "效 益",
+      "nC are",
+      "nCar e",
+      "n Care",
+      "一 代",
+      "▁h ack",
+      "▁ha ck",
+      "▁ hack",
+      "▁g overn",
+      "▁go vern",
+      "▁gover n",
+      "▁gov ern",
+      "想 象",
+      "un ning",
+      "unn ing",
+      "re pository",
+      "rep ository",
+      "repos itory",
+      "公 寓",
+      "副 书记",
+      "大 队",
+      "▁r ul",
+      "▁ru l",
+      "家 族",
+      "▁Ch ris",
+      "▁ Chris",
+      "▁b uffer",
+      "▁buff er",
+      "▁buf fer",
+      "▁ buffer",
+      "打 印",
+      "ur ers",
+      "ure rs",
+      "urer s",
+      "ut able",
+      "uta ble",
+      "u table",
+      "适 宜",
+      "Vec tor",
+      "V ector",
+      "L a",
+      "▁v ectors",
+      "▁ve ctors",
+      "▁vector s",
+      "▁ vectors",
+      "call back",
+      "▁Ap pro",
+      "▁App ro",
+      "▁ Appro",
+      "UR POSE",
+      "罚 金",
+      "re du",
+      "red u",
+      "r edu",
+      "WA RE",
+      "WAR E",
+      "W ARE",
+      "▁m igr",
+      "▁mi gr",
+      "▁Jer sey",
+      "编 号",
+      "▁fil ename",
+      "▁file name",
+      "▁fi lename",
+      "▁ filename",
+      "pr ice",
+      "pri ce",
+      "p rice",
+      "\\' \\'\\'\\",
+      "\\'\\' \\'\\",
+      "\\'\\'\\' \\",
+      "\\'\\'\\ '\\",
+      "on om",
+      "ono m",
+      "o nom",
+      "ri um",
+      "r ium",
+      "ut s",
+      "u ts",
+      "▁for ced",
+      "▁force d",
+      "▁ forced",
+      "▁re ader",
+      "▁r eader",
+      "▁read er",
+      "▁ reader",
+      "在执行 过程中",
+      "在 执行过程中",
+      "ivers e",
+      "iver se",
+      "i verse",
+      "▁fig ures",
+      "▁figure s",
+      "▁H ard",
+      "▁Har d",
+      "▁Ha rd",
+      "▁ Hard",
+      "▁Re quest",
+      "▁Requ est",
+      "▁ Request",
+      "Z E",
+      "▁m ovies",
+      "▁mov ies",
+      "▁movie s",
+      "▁ movies",
+      "om p",
+      "o mp",
+      "re ams",
+      "ream s",
+      "rea ms",
+      "▁c ream",
+      "▁cre am",
+      "over ride",
+      "他 说",
+      "▁pre valence",
+      "▁preval ence",
+      "▁fa ith",
+      "课 题",
+      "▁ad ds",
+      "▁add s",
+      "▁R h",
+      "▁ Rh",
+      "an na",
+      "ann a",
+      "模 块",
+      "布 局",
+      "体 重",
+      "新 技术",
+      "▁ex ceed",
+      "▁exc eed",
+      "女 子",
+      "me ga",
+      "meg a",
+      "m ega",
+      "t k",
+      "ro ns",
+      "ron s",
+      "r ons",
+      "▁s ister",
+      "▁si ster",
+      "防 控",
+      "你可 以",
+      "你 可以",
+      "▁C ost",
+      "▁Co st",
+      "▁Cos t",
+      "▁ Cost",
+      "re sources",
+      "res ources",
+      "resource s",
+      "▁cont emporary",
+      "很 快",
+      "鸡 蛋",
+      "▁ 装",
+      "▁L ord",
+      "▁Lo rd",
+      "▁Lor d",
+      "留 下",
+      "▁k inds",
+      "▁kind s",
+      "▁kin ds",
+      "us hed",
+      "ush ed",
+      "▁model ing",
+      "▁mod eling",
+      "▁mode ling",
+      "很好 的",
+      "很 好的",
+      "外 国",
+      "▁S pain",
+      "▁Sp ain",
+      "▁Spa in",
+      "▁pr incipal",
+      "▁princ ipal",
+      "▁princip al",
+      "▁ principal",
+      "▁B ow",
+      "▁Bo w",
+      "▁F C",
+      "▁ FC",
+      "▁p aint",
+      "▁pain t",
+      "▁pa int",
+      "▁ paint",
+      "页 面",
+      "f d",
+      "ri ate",
+      "ria te",
+      "r iate",
+      "真的 是",
+      "真 的是",
+      "re pre",
+      "rep re",
+      "repr e",
+      "本院 受理后",
+      "本院受理 后",
+      "造 型",
+      "监 控",
+      "build er",
+      "▁qual ified",
+      "▁ qualified",
+      "} \"",
+      "▁ill ness",
+      "ba t",
+      "b at",
+      "▁s av",
+      "▁sa v",
+      "▁ sav",
+      "体 制",
+      "▁h ouses",
+      "▁house s",
+      "▁hous es",
+      "▁ho uses",
+      "▁ houses",
+      "▁J esus",
+      "▁Jes us",
+      "▁Je sus",
+      "La st",
+      "L ast",
+      "▁w heel",
+      "▁whe el",
+      "▁ wheel",
+      "基本 上",
+      "▁c ounty",
+      "▁count y",
+      "▁cou nty",
+      "▁ county",
+      "di m",
+      "d im",
+      "on ed",
+      "one d",
+      "o ned",
+      "hy th",
+      "▁enc ourage",
+      "▁encour age",
+      "工 厂",
+      "的 主",
+      "▁E ll",
+      "▁El l",
+      "▁ Ell",
+      "▁Int egr",
+      "▁ Integr",
+      "名 下",
+      "re v",
+      "r ev",
+      "的 应用",
+      "ate gy",
+      "ateg y",
+      "▁f ell",
+      "▁fe ll",
+      "▁fel l",
+      "▁s ucceed",
+      "▁suc ceed",
+      "▁succ eed",
+      "知 乎",
+      "当 中",
+      "Jav a",
+      "J ava",
+      "应予 准许",
+      "电 动",
+      "assert Equals",
+      "assertEqual s",
+      "剥 夺",
+      "ov ie",
+      "ovi e",
+      "o vie",
+      "▁t icket",
+      "▁tick et",
+      "▁ ticket",
+      "顺 序",
+      "天 天",
+      "▁fe aturing",
+      "▁feat uring",
+      "第五百 一十九",
+      "com mun",
+      "comm un",
+      "▁Method s",
+      "▁Meth ods",
+      "▁ Methods",
+      "▁P hot",
+      "▁Ph ot",
+      "▁ Phot",
+      "nF alse",
+      "n False",
+      "pi p",
+      "p ip",
+      ">> >",
+      "> >>",
+      "▁T ake",
+      "▁Ta ke",
+      "▁Tak e",
+      "▁ Take",
+      "▁p atch",
+      "▁pat ch",
+      "▁ patch",
+      "将 会",
+      "nw ith",
+      "n with",
+      "▁cor related",
+      "▁correl ated",
+      "▁correlate d",
+      "ns et",
+      "nse t",
+      "n set",
+      "▁un able",
+      "▁u nable",
+      "▁una ble",
+      "ploy ment",
+      "▁em ployee",
+      "▁employ ee",
+      "▁ employee",
+      "▁h olding",
+      "▁hold ing",
+      "▁hol ding",
+      "▁ holding",
+      "}) ;",
+      "} );",
+      "▁or th",
+      "▁ orth",
+      "计划 生育",
+      "G r",
+      "▁b ron",
+      "▁br on",
+      "▁bro n",
+      "▁ bron",
+      "He ight",
+      "H eight",
+      "pro gram",
+      "pr ogram",
+      "prog ram",
+      "通 讯",
+      "am ber",
+      "amb er",
+      "a mber",
+      "▁concept s",
+      "▁conce pts",
+      "▁W here",
+      "▁Wh ere",
+      "▁ Where",
+      "▁t een",
+      "▁te en",
+      "▁tee n",
+      "▁ teen",
+      "nEng lish",
+      "n English",
+      "az ure",
+      "a zure",
+      "nW hy",
+      "nWh y",
+      "n Why",
+      "▁H om",
+      "▁Ho m",
+      "▁ Hom",
+      "▁r ide",
+      "▁ri de",
+      "▁rid e",
+      "▁ ride",
+      "▁mass ive",
+      "药 品",
+      "改 进",
+      "C G",
+      "olog ist",
+      "olo gist",
+      "▁ins ight",
+      "▁v ulner",
+      "▁vul ner",
+      "无 线",
+      "▁Int eger",
+      "▁ Integer",
+      "no on",
+      "n oon",
+      "▁op posite",
+      "指 南",
+      "分 数",
+      "fu n",
+      "f un",
+      "题 主",
+      "▁data set",
+      "▁dat aset",
+      "▁datas et",
+      "▁ dataset",
+      "op rote",
+      "opro te",
+      "oprot e",
+      "o prote",
+      "is hop",
+      "ish op",
+      "i shop",
+      "▁ear ned",
+      "▁earn ed",
+      "漂 亮",
+      "▁substant ial",
+      "▁ α",
+      "us et",
+      "use t",
+      "u set",
+      "', '",
+      "' ,'",
+      "nF ind",
+      "nFin d",
+      "n Find",
+      "▁off icers",
+      "▁offic ers",
+      "▁office rs",
+      "▁officer s",
+      "▁sit uations",
+      "▁situation s",
+      "▁situ ations",
+      "园 区",
+      "我 要",
+      "▁k ick",
+      "▁ki ck",
+      "▁ kick",
+      "▁m ine",
+      "▁min e",
+      "▁mi ne",
+      "▁ mine",
+      "▁M anager",
+      "▁Man ager",
+      "▁Manag er",
+      "▁Manage r",
+      "▁ Manager",
+      "评 为",
+      "la n",
+      "l an",
+      "lo op",
+      "loo p",
+      "l oop",
+      "man n",
+      "ma nn",
+      "m ann",
+      "▁scient ists",
+      "▁scientist s",
+      "▁large ly",
+      "▁larg ely",
+      "▁con sec",
+      "▁cons ec",
+      "Par ameter",
+      "Param eter",
+      "rep rene",
+      "repre ne",
+      "repr ene",
+      "st at",
+      "sta t",
+      "s tat",
+      "分 离",
+      "day s",
+      "da ys",
+      "d ays",
+      "▁pass ing",
+      "▁pas sing",
+      "▁challeng ing",
+      "▁sup press",
+      "cl usions",
+      "clus ions",
+      "clusion s",
+      "景 观",
+      "对 应",
+      "防 治",
+      "s f",
+      "适 量",
+      "to pic",
+      "top ic",
+      "t opic",
+      "障 碍",
+      "▁l ect",
+      "▁le ct",
+      "▁ lect",
+      "▁P A",
+      "▁ PA",
+      "▁P rop",
+      "▁Pro p",
+      "▁Pr op",
+      "▁ Prop",
+      "▁sh ock",
+      "▁ shock",
+      "▁near by",
+      "▁com posed",
+      "▁comp osed",
+      "▁compos ed",
+      "▁compose d",
+      "▁t emp",
+      "▁te mp",
+      "▁tem p",
+      "▁ temp",
+      "▁F ROM",
+      "▁FR OM",
+      "▁ FROM",
+      "▁fem ales",
+      "▁female s",
+      "▁rapid ly",
+      "也不 是",
+      "也 不是",
+      "▁int ensity",
+      "▁intens ity",
+      "▁ intensity",
+      "▁Sen ate",
+      "iz able",
+      "iza ble",
+      "▁in vent",
+      "▁inv ent",
+      "▁Stand ard",
+      "▁ Standard",
+      "对 此",
+      "um my",
+      "umm y",
+      "▁H or",
+      "▁Ho r",
+      "▁ Hor",
+      "车程 约",
+      "随 后",
+      "三 月",
+      "付 款",
+      "▁A P",
+      "▁ AP",
+      "▁O pt",
+      "▁Op t",
+      "▁ Opt",
+      "好 看",
+      "▁dr ama",
+      "▁dram a",
+      "nY es",
+      "n Yes",
+      "一 级",
+      "▁normal ly",
+      "▁norm ally",
+      "yl ed",
+      "yle d",
+      "y led",
+      "▁P O",
+      "▁ PO",
+      "▁gu est",
+      "▁ guest",
+      "od ium",
+      "odi um",
+      "UL AR",
+      "秘 书",
+      "途 径",
+      "不 见",
+      "海 洋",
+      "致 力",
+      "la ss",
+      "las s",
+      "l ass",
+      "og ether",
+      "oge ther",
+      "第一 节",
+      "▁ с",
+      "男 子",
+      "▁W is",
+      "▁Wi s",
+      "c n",
+      "di ff",
+      "d iff",
+      "▁en vironments",
+      "▁environment s",
+      "▁re solve",
+      "▁res olve",
+      "▁ resolve",
+      "it age",
+      "ita ge",
+      "i tage",
+      "意 外",
+      "加倍 支付",
+      "推 理",
+      "ow nt",
+      "own t",
+      "Bo ok",
+      "B ook",
+      "硬 件",
+      "▁( \\'",
+      "▁(\\ '",
+      "▁ (\\'",
+      "aw n",
+      "a wn",
+      "▁p ortion",
+      "▁port ion",
+      "▁por tion",
+      "▁ portion",
+      "▁Americ ans",
+      "▁American s",
+      "▁America ns",
+      "▁Amer icans",
+      "▁ Americans",
+      "E l",
+      "场 所",
+      "餐 饮",
+      "▁Hot el",
+      "▁Ho tel",
+      "▁act ing",
+      "▁ac ting",
+      "▁actin g",
+      "▁ acting",
+      "▁p relim",
+      "▁pre lim",
+      "书 法",
+      "逐 步",
+      "三 大",
+      "李 某",
+      "看 来",
+      "如不服 本裁定",
+      "Americ an",
+      "America n",
+      "A merican",
+      "▁D ub",
+      "▁Du b",
+      "▁i Phone",
+      "▁ iPhone",
+      "▁coord inates",
+      "▁coordinate s",
+      "▁ coordinates",
+      "ac r",
+      "a cr",
+      "▁iss ued",
+      "▁issue d",
+      "▁ issued",
+      "RE E",
+      "R EE",
+      "养 殖",
+      "▁att acks",
+      "▁attack s",
+      "关 心",
+      "▁pr ison",
+      "▁pri son",
+      "▁ prison",
+      "▁w a",
+      "▁ wa",
+      "▁J ac",
+      "▁Ja c",
+      "▁ Jac",
+      "is a",
+      "i sa",
+      "▁M IT",
+      "▁MI T",
+      "▁ MIT",
+      "群 体",
+      "te am",
+      "小 编",
+      "▁s aved",
+      "▁sa ved",
+      "▁save d",
+      "▁sav ed",
+      "▁ saved",
+      "▁begin s",
+      "▁beg ins",
+      "mon d",
+      "mo nd",
+      "m ond",
+      "从 来",
+      "概 率",
+      "▁nor thern",
+      "▁north ern",
+      "关 节",
+      "玉 米",
+      "接 近",
+      "个 数",
+      "▁ca using",
+      "▁caus ing",
+      "▁{ \\'",
+      "▁{\\ '",
+      "▁ {\\'",
+      "▁under stood",
+      "第二百 零",
+      "ye s",
+      "y es",
+      "修 复",
+      "la yer",
+      "lay er",
+      "l ayer",
+      "预 算",
+      "P lease",
+      "▁e ntr",
+      "▁en tr",
+      "▁ent r",
+      "▁Ill inois",
+      "相关 的",
+      "▁fl owers",
+      "▁flo wers",
+      "▁flow ers",
+      "▁flower s",
+      "个 性",
+      "▁s an",
+      "▁sa n",
+      "▁ san",
+      "那 样",
+      "▁target ed",
+      "▁ targeted",
+      "也不 会",
+      "也 不会",
+      "▁$ {\\\\",
+      "▁${ \\\\",
+      "▁ ${\\\\",
+      "Pa y",
+      "P ay",
+      "an dy",
+      "and y",
+      "▁month ly",
+      "US ER",
+      "USE R",
+      "U SER",
+      "▁person nel",
+      "▁av ailability",
+      "▁avail ability",
+      "▁ availability",
+      "第二 节",
+      "el ly",
+      "ell y",
+      "阳 县",
+      "材 质",
+      "东 南",
+      "水 果",
+      "▁Corpor ation",
+      "▁Corp oration",
+      "▁pro ducer",
+      "▁produ cer",
+      "▁produce r",
+      "▁prod ucer",
+      "▁ producer",
+      "▁n arrow",
+      "▁narr ow",
+      "az y",
+      "a zy",
+      "▁estim ates",
+      "▁estimate s",
+      "▁Cont act",
+      "▁ Contact",
+      "以上 的",
+      "以 上的",
+      "ce rt",
+      "cer t",
+      "c ert",
+      "一 道",
+      "▁b le",
+      "▁bl e",
+      "▁ ble",
+      "▁re ferences",
+      "▁r eferences",
+      "▁refer ences",
+      "▁reference s",
+      "▁ references",
+      "还 没",
+      "▁e ating",
+      "▁eat ing",
+      "医 师",
+      "light s",
+      "l ights",
+      "可 爱",
+      "▁sign ificance",
+      "▁signific ance",
+      "▁Phys ics",
+      "▁ Physics",
+      "Write r",
+      "Writ er",
+      "Wr iter",
+      "W riter",
+      "as m",
+      "a sm",
+      "▁g rown",
+      "▁gr own",
+      "▁grow n",
+      "▁gro wn",
+      "▁ grown",
+      "VER SION",
+      "V ERSION",
+      "辽宁 省",
+      "竟 然",
+      "▁st rings",
+      "▁str ings",
+      "▁string s",
+      "▁ strings",
+      "et c",
+      "e tc",
+      "其 次",
+      "▁hon est",
+      "▁ho nest",
+      "▁hone st",
+      "sp ect",
+      "spec t",
+      "spe ct",
+      "s pect",
+      "UI D",
+      "U ID",
+      "▁dist ingu",
+      "公 益",
+      "nI m",
+      "n Im",
+      "li mit",
+      "lim it",
+      "l imit",
+      "Do c",
+      "D oc",
+      "▁user name",
+      "▁ username",
+      "▁p H",
+      "▁ pH",
+      "必 然",
+      "的 申请",
+      "nA d",
+      "n Ad",
+      "▁oper ators",
+      "▁operator s",
+      "▁ operators",
+      ") \"",
+      "ic ients",
+      "ici ents",
+      "icient s",
+      "/\\ \\",
+      "/ \\\\",
+      "交 换",
+      "ot ion",
+      "oti on",
+      "o tion",
+      "Pro perties",
+      "▁I O",
+      "▁ IO",
+      "陕 西",
+      "▁conven ient",
+      "▁respons ibility",
+      "为 此",
+      "▁found ation",
+      "▁ foundation",
+      "▁Child ren",
+      "▁ Children",
+      "es ting",
+      "est ing",
+      "am ic",
+      "ami c",
+      "a mic",
+      "▁re vers",
+      "▁reve rs",
+      "▁rev ers",
+      "▁rever s",
+      "▁s ummary",
+      "▁summ ary",
+      "▁summar y",
+      "▁ summary",
+      "保 健",
+      "▁s pl",
+      "▁sp l",
+      "▁ spl",
+      "C ategory",
+      "▁Kore a",
+      "▁Ko rea",
+      "▁Kor ea",
+      "▁W ales",
+      "▁Wal es",
+      "▁Wa les",
+      "if icial",
+      "ific ial",
+      "▁M P",
+      "▁ MP",
+      "▁# ##",
+      "▁## #",
+      "▁ ###",
+      "▁int egers",
+      "▁integer s",
+      "UT H",
+      "U TH",
+      "em y",
+      "e my",
+      "al ian",
+      "ali an",
+      "alia n",
+      "超 市",
+      "▁ant igen",
+      "▁anti gen",
+      "▁Rad io",
+      "▁ Radio",
+      "▁p rivacy",
+      "▁priv acy",
+      "▁ privacy",
+      "▁hand ling",
+      "▁stim ulation",
+      "▁stimul ation",
+      "▁pe ace",
+      "▁peac e",
+      "▁Ge orgia",
+      "▁Georg ia",
+      "把 握",
+      "PO RT",
+      "POR T",
+      "P ORT",
+      "▁arr ang",
+      "氛 围",
+      "nf ortunately",
+      "▁platform s",
+      "▁plat forms",
+      "▁ platforms",
+      "呼 吸",
+      "N C",
+      "▁o urs",
+      "▁our s",
+      "▁ou rs",
+      "▁ ours",
+      "▁J o",
+      "▁ Jo",
+      "▁K h",
+      "▁ Kh",
+      "▁quant itative",
+      "▁May be",
+      "▁ Maybe",
+      "的 中",
+      "▁occ asion",
+      "▁occas ion",
+      "H L",
+      "▁concer ned",
+      "▁concern ed",
+      "▁tr acking",
+      "▁track ing",
+      "▁ tracking",
+      "国务 院",
+      "释 放",
+      "\" #",
+      "▁re actions",
+      "▁reaction s",
+      "▁react ions",
+      "Char Field",
+      "H C",
+      "nQ u",
+      "n Qu",
+      "▁p olymer",
+      "▁poly mer",
+      "▁polym er",
+      "▁ polymer",
+      "▁risk s",
+      "▁ris ks",
+      "很 少",
+      "张 某",
+      "ba sh",
+      "bas h",
+      "b ash",
+      "f etch",
+      "你 会",
+      "姐 姐",
+      "▁par king",
+      "▁park ing",
+      "小 米",
+      "▁M icro",
+      "▁Mi cro",
+      "▁Mic ro",
+      "▁ Micro",
+      "et ing",
+      "eti ng",
+      "etin g",
+      "e ting",
+      "re solve",
+      "res olve",
+      "▁St ore",
+      "▁ Store",
+      "ver ter",
+      "vert er",
+      "verte r",
+      "▁review ed",
+      "▁ reviewed",
+      "▁S aint",
+      "▁Sa int",
+      "▁guarant ee",
+      "取 值",
+      "ies t",
+      "ie st",
+      "i est",
+      "▁im plications",
+      "▁impl ications",
+      "▁implication s",
+      "的 能力",
+      "系 数",
+      "结果 表明",
+      "从 轻处罚",
+      "▁purch ased",
+      "▁purchase d",
+      "ERR OR",
+      "er ly",
+      "▁bu sy",
+      "▁bus y",
+      "▁h ence",
+      "▁he nce",
+      "▁hen ce",
+      "▁Franc isco",
+      "▁Francis co",
+      "▁as sembly",
+      "▁assembl y",
+      "▁ assembly",
+      "大 连",
+      "▁总 部",
+      "▁ 总部",
+      "不错 的",
+      "不 错的",
+      "▁/ >",
+      "▁ />",
+      "▁program me",
+      "携 带",
+      "ex e",
+      "e xe",
+      "详 情",
+      "lar ge",
+      "larg e",
+      "l arge",
+      "部 位",
+      "▁W ho",
+      "▁Wh o",
+      "▁ Who",
+      "com put",
+      "comp ut",
+      "im ately",
+      "imate ly",
+      "▁com pound",
+      "▁comp ound",
+      "▁M ah",
+      "▁Ma h",
+      "▁ Mah",
+      "▁S af",
+      "▁Sa f",
+      "▁U l",
+      "▁ Ul",
+      "ti nt",
+      "tin t",
+      "t int",
+      "NE SS",
+      "N ESS",
+      "in als",
+      "inal s",
+      "ina ls",
+      "i nals",
+      "hem istry",
+      "ie ves",
+      "ieve s",
+      "iev es",
+      "i eves",
+      "可 见",
+      "也 在",
+      "都 没",
+      "▁U nit",
+      "▁Un it",
+      "▁Uni t",
+      "▁ Unit",
+      "▁call back",
+      "▁ callback",
+      "▁club s",
+      "真正 的",
+      "EFA ULT",
+      "▁re spir",
+      "▁res pir",
+      "▁resp ir",
+      "▁flex ible",
+      "加 速",
+      "带来 的",
+      "带 来的",
+      "舞 台",
+      "和 解",
+      "▁inter ests",
+      "▁interest s",
+      "▁dest ination",
+      "▁ destination",
+      "累 计",
+      "er ior",
+      "eri or",
+      "e rior",
+      "/> \\",
+      "/ >\\",
+      "▁pro ved",
+      "▁pr oved",
+      "▁prov ed",
+      "▁prove d",
+      "▁ proved",
+      "ac tic",
+      "act ic",
+      "a ctic",
+      "▁bath room",
+      "se x",
+      "s ex",
+      "承 诺",
+      "▁S pace",
+      "▁Sp ace",
+      "▁Spa ce",
+      "▁ Space",
+      "五 月",
+      "▁re act",
+      "▁ react",
+      "劳 务",
+      "播 放",
+      "节 能",
+      "▁cit izens",
+      "▁citiz ens",
+      "▁citizen s",
+      "▁p itch",
+      "▁pit ch",
+      "▁ pitch",
+      "is ta",
+      "ist a",
+      "i sta",
+      "▁man agers",
+      "▁manag ers",
+      "▁manager s",
+      "▁manage rs",
+      "游 客",
+      "尊 重",
+      "▁entire ly",
+      "▁t ip",
+      "▁ti p",
+      "▁ tip",
+      "▁d ates",
+      "▁dat es",
+      "▁date s",
+      "▁da tes",
+      "▁ dates",
+      "▁app lying",
+      "▁apply ing",
+      "▁appl ying",
+      "可在 判决书送达之日起十五日内",
+      "危 机",
+      "▁commit ment",
+      "▁mach ines",
+      "▁machine s",
+      "地 铁",
+      "▁ent ries",
+      "▁entr ies",
+      "▁ entries",
+      "▁sur faces",
+      "▁surface s",
+      "▁surf aces",
+      "展 览",
+      "▁V ideo",
+      "▁Vid eo",
+      "▁ Video",
+      "ipher al",
+      "iph eral",
+      "▁\" \"\\",
+      "▁\"\" \\",
+      "▁ \"\"\\",
+      "▁K ent",
+      "▁Ke nt",
+      "▁Ken t",
+      "na s",
+      "n as",
+      "▁Joh nson",
+      "▁John son",
+      "▁Johns on",
+      "▁Ill uminate",
+      "▁ Illuminate",
+      "十二 月",
+      "十 二月",
+      "▁Pac ific",
+      "▁ Pacific",
+      "为 何",
+      "▁ 周",
+      "间 的",
+      "sh ared",
+      "sha red",
+      "share d",
+      "接受 教育",
+      "加 大",
+      "▁viol ence",
+      "▁m ountain",
+      "▁mount ain",
+      "▁Dem ocratic",
+      "▁Democr atic",
+      "▁Democrat ic",
+      "name space",
+      "names pace",
+      "n amespace",
+      "▁dom estic",
+      "▁domest ic",
+      "小 数",
+      "och ond",
+      "十八 条",
+      "十 八条",
+      "▁c oc",
+      "▁co c",
+      "▁ coc",
+      "第二百五十 三",
+      "▁d ental",
+      "▁den tal",
+      "▁dent al",
+      "▁L ow",
+      "▁Lo w",
+      "▁ Low",
+      "视 觉",
+      "ju nit",
+      "jun it",
+      "j unit",
+      "▁Pro blem",
+      "▁Prob lem",
+      "▁ Problem",
+      "▁en abled",
+      "▁enable d",
+      "▁ enabled",
+      "近 期",
+      "nT e",
+      "n Te",
+      "国 的",
+      "▁e g",
+      "▁ eg",
+      "in er",
+      "ine r",
+      "i ner",
+      "抚养 费",
+      "\") ]\\",
+      "\" )]\\",
+      "站 在",
+      "T ube",
+      "init ial",
+      "om etimes",
+      "ome times",
+      "omet imes",
+      "▁f le",
+      "▁fl e",
+      "战 士",
+      "j u",
+      "el lite",
+      "ell ite",
+      "elli te",
+      "总 统",
+      "▁sur ve",
+      "▁surv e",
+      "的 本次执行程序",
+      "▁ass uming",
+      "▁assum ing",
+      "▁ assuming",
+      "常 规",
+      "b d",
+      "assert Equal",
+      "II I",
+      "I II",
+      "▁b acteria",
+      "▁bacter ia",
+      "▁A BC",
+      "▁AB C",
+      "▁ ABC",
+      "▁we ap",
+      "bi an",
+      "bia n",
+      "b ian",
+      "法律 法规",
+      "▁anti body",
+      "▁antib ody",
+      "F i",
+      "Mo ck",
+      "M ock",
+      "外 观",
+      "电 压",
+      "养 老",
+      "ist ered",
+      "ister ed",
+      "iste red",
+      "设 立",
+      "▁I de",
+      "▁Id e",
+      "补 偿",
+      "▁p ert",
+      "▁per t",
+      "▁pe rt",
+      "▁ pert",
+      "得到 了",
+      "得 到了",
+      "mi r",
+      "m ir",
+      "N e",
+      "▁inflamm atory",
+      "▁ inflammatory",
+      "v c",
+      "▁Mat hematics",
+      "▁Math ematics",
+      "▁Mathematic s",
+      "is ition",
+      "isi tion",
+      "pl ugins",
+      "plugin s",
+      "plug ins",
+      "婚 后",
+      "▁m useum",
+      "混 凝",
+      "明 的",
+      "不 然",
+      "消 失",
+      "▁F ar",
+      "▁Fa r",
+      "▁ Far",
+      "bur y",
+      "bu ry",
+      "b ury",
+      "ib li",
+      "ibl i",
+      "▁ver ify",
+      "▁ verify",
+      "SUL TS",
+      "SULT S",
+      "V M",
+      "▁S U",
+      "▁ SU",
+      "▁L ittle",
+      "▁Li ttle",
+      "▁Lit tle",
+      "▁ Little",
+      "▁ex port",
+      "▁exp ort",
+      "▁ export",
+      "xi ety",
+      "id o",
+      "i do",
+      "联 赛",
+      "bl ack",
+      "bla ck",
+      "长 春",
+      "▁s heet",
+      "▁she et",
+      "▁ sheet",
+      "▁Le vel",
+      "▁Lev el",
+      "▁ Level",
+      "▁f actory",
+      "▁fact ory",
+      "▁factor y",
+      "▁ factory",
+      "▁term inal",
+      "▁ terminal",
+      "申 报",
+      "▁a wards",
+      "▁aw ards",
+      "▁award s",
+      "▁Ab out",
+      "▁ About",
+      "疗 效",
+      "优 点",
+      "▁reg istration",
+      "▁registr ation",
+      "▁ registration",
+      "ic ted",
+      "ict ed",
+      "▁总部 地点",
+      "O ur",
+      "广 播",
+      "io x",
+      "i ox",
+      "cl ean",
+      "cle an",
+      "c lean",
+      "术 后",
+      "S F",
+      "▁Par liament",
+      "▁inflamm ation",
+      "▁ 我",
+      "row se",
+      "rows e",
+      "▁I rish",
+      "▁Ir ish",
+      "ac ent",
+      "ace nt",
+      "a cent",
+      "本院 审理",
+      "▁conc rete",
+      "in ton",
+      "int on",
+      "into n",
+      "i nton",
+      "Date Time",
+      "▁St udio",
+      "▁Stud io",
+      "▁ Studio",
+      "▁re veal",
+      "▁reve al",
+      "▁rev eal",
+      "re view",
+      "rev iew",
+      "▁S ports",
+      "▁Sp orts",
+      "▁Sport s",
+      "▁ Sports",
+      "▁N C",
+      "▁ NC",
+      "▁admin istrative",
+      "▁administr ative",
+      "八 日",
+      "▁a ges",
+      "▁ag es",
+      "▁age s",
+      "▁ ages",
+      "▁im mediate",
+      "▁immed iate",
+      "扩 展",
+      "▁r ailway",
+      "▁rail way",
+      "▁Ex ample",
+      "▁Exam ple",
+      "▁ Example",
+      "▁in jection",
+      "▁inj ection",
+      "▁inject ion",
+      "▁ injection",
+      "in son",
+      "ins on",
+      "i nson",
+      "mat erial",
+      "mate rial",
+      "m aterial",
+      "▁warrant y",
+      "面 前",
+      "圆 的",
+      "▁▁ ▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁ ▁",
+      "▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁ ▁▁",
+      "▁ ▁▁▁▁▁▁",
+      "▁I g",
+      "▁ Ig",
+      "▁h ospit",
+      "账 号",
+      "ri ces",
+      "ric es",
+      "rice s",
+      "r ices",
+      "和 社会",
+      "▁E U",
+      "▁ EU",
+      "江 西",
+      "ogg le",
+      "▁C oll",
+      "▁Col l",
+      "▁Co ll",
+      "▁ Coll",
+      "v o",
+      "▁p reliminary",
+      "▁prelim inary",
+      "▁Ac count",
+      "▁Acc ount",
+      "▁ Account",
+      "承 包",
+      "▁fant astic",
+      "▁fantas tic",
+      "nat ive",
+      "n ative",
+      "▁S P",
+      "▁ SP",
+      "il ly",
+      "ill y",
+      "▁Re present",
+      "▁Rep resent",
+      "au s",
+      "a us",
+      "▁lo ading",
+      "▁load ing",
+      "▁ loading",
+      "那 里",
+      "He ad",
+      "H ead",
+      "▁def ense",
+      "▁defe nse",
+      "应 对",
+      "nCare er",
+      "nCar eer",
+      "▁access ible",
+      "▁ accessible",
+      "拘 留",
+      "的第 二",
+      "的 第二",
+      "市 委",
+      "It ems",
+      "Item s",
+      "工程 师",
+      "▁le uk",
+      "▁ leuk",
+      "期限 内",
+      "休 息",
+      "▁br inging",
+      "▁bring ing",
+      "内 心",
+      "lo ader",
+      "load er",
+      "八 月",
+      "指 挥",
+      "▁s ter",
+      "▁st er",
+      "▁ste r",
+      "▁ ster",
+      "▁em otional",
+      "▁emot ional",
+      "▁emotion al",
+      "▁enjoy ed",
+      "i h",
+      "▁P romise",
+      "▁Prom ise",
+      "▁ Promise",
+      "▁U I",
+      "▁ UI",
+      "指控 被告人",
+      "▁bas eline",
+      "▁base line",
+      "▁ baseline",
+      "图 形",
+      "CL US",
+      "▁M iddle",
+      "▁Midd le",
+      "▁ Middle",
+      "nt hat",
+      "nth at",
+      "n that",
+      "▁character istic",
+      "for ward",
+      "▁bre ath",
+      "▁breat h",
+      "仪 式",
+      "geq slant",
+      "▁p ulmonary",
+      "▁pul monary",
+      "提出 上诉",
+      "о в",
+      "货 款",
+      "孩子 们",
+      "麻 烦",
+      "▁c ensus",
+      "▁cens us",
+      "气 候",
+      "oc olate",
+      "ocol ate",
+      "oco late",
+      "商 标",
+      "▁pro fessor",
+      "▁profess or",
+      "ga r",
+      "g ar",
+      "▁p roducing",
+      "▁produ cing",
+      "▁prod ucing",
+      "▁ producing",
+      "▁u m",
+      "▁ um",
+      "ta r",
+      "t ar",
+      "▁p up",
+      "▁pu p",
+      "著 名",
+      "\\' ],",
+      "\\'] ,",
+      "\\ '],",
+      "▁rec ipe",
+      "▁recip e",
+      "▁ recipe",
+      "远 程",
+      "客 观",
+      "az e",
+      "a ze",
+      "▁cont ributions",
+      "▁contribut ions",
+      "▁contribution s",
+      "▁f ault",
+      "▁fa ult",
+      "▁ fault",
+      "▁am ino",
+      "▁amin o",
+      "▁ amino",
+      "▁valid ate",
+      "▁ validate",
+      "ir th",
+      "irt h",
+      "▁af fordable",
+      "▁afford able",
+      "▁L abor",
+      "▁La bor",
+      "▁Lab or",
+      "▁ é",
+      "海 外",
+      "is er",
+      "ise r",
+      "i ser",
+      "▁coh ort",
+      "垂 直",
+      "合 肥",
+      "if er",
+      "ife r",
+      "i fer",
+      "武 器",
+      "▁met abolic",
+      "▁metab olic",
+      "▁metabol ic",
+      "▁meta bolic",
+      "/ )",
+      "sub mit",
+      "健 身",
+      "近 代",
+      "▁mark ers",
+      "▁mar kers",
+      "▁marker s",
+      "sub ject",
+      "su bject",
+      "学 的",
+      "该 村",
+      "as ters",
+      "ast ers",
+      "aster s",
+      "aste rs",
+      "a sters",
+      "年 底",
+      "如 实",
+      "校 长",
+      "▁L I",
+      "▁ LI",
+      "▁young er",
+      "▁f eas",
+      "▁fe as",
+      "▁inter ventions",
+      "▁intervention s",
+      "an ol",
+      "ano l",
+      "a nol",
+      "Que ue",
+      "▁ac ids",
+      "▁acid s",
+      "▁a gricult",
+      "▁agr icult",
+      "于 本判决",
+      "原审 被告",
+      "L P",
+      "br al",
+      "bra l",
+      "b ral",
+      "ch at",
+      "cha t",
+      "c hat",
+      "二 审",
+      "▁ 赵",
+      "▁l ayers",
+      "▁la yers",
+      "▁layer s",
+      "▁lay ers",
+      "▁ layers",
+      "▁aware ness",
+      "▁calc ium",
+      "\" }",
+      "▁att ribute",
+      "▁ attribute",
+      "维 生素",
+      "产 量",
+      "▁con sequences",
+      "▁consequ ences",
+      "▁consequence s",
+      "A c",
+      "▁v eter",
+      "▁ve ter",
+      "▁vet er",
+      "▁m igration",
+      "▁migr ation",
+      "▁ migration",
+      "_\\ \\",
+      "_ \\\\",
+      "ho m",
+      "h om",
+      "纷 纷",
+      "si d",
+      "s id",
+      "▁L oad",
+      "▁Lo ad",
+      "▁ Load",
+      "▁br ands",
+      "▁brand s",
+      "en ders",
+      "end ers",
+      "ender s",
+      "ende rs",
+      "\" <",
+      "▁cr ime",
+      "▁crim e",
+      "▁flu ores",
+      "▁fluor es",
+      "▁tr ends",
+      "▁trend s",
+      "ic ons",
+      "icon s",
+      "ico ns",
+      "i cons",
+      "▁en dot",
+      "▁end ot",
+      "▁W illiams",
+      "▁Willi ams",
+      "▁William s",
+      "▁he pat",
+      "很大 的",
+      "很 大的",
+      "hu s",
+      "h us",
+      "投资 者",
+      "N E",
+      "▁d oubt",
+      "▁dou bt",
+      "▁doub t",
+      "com plete",
+      "comp lete",
+      "compl ete",
+      "▁sup plement",
+      "▁suppl ement",
+      "▁great ly",
+      "▁T ry",
+      "▁Tr y",
+      "▁ Try",
+      "吃 饭",
+      "图 象",
+      "▁I ss",
+      "▁Is s",
+      "▁ Iss",
+      "后 期",
+      "紧 张",
+      "▁not iced",
+      "▁notice d",
+      "Us e",
+      "U se",
+      "而 不",
+      "▁tr ouble",
+      "▁tro uble",
+      "▁trou ble",
+      "▁regular ly",
+      "▁regul arly",
+      "▁P ut",
+      "▁Pu t",
+      "▁ Put",
+      "▁nut r",
+      "▁nu tr",
+      "PASS WORD",
+      "纠纷 一案中",
+      "纠纷一案 中",
+      "▁R ay",
+      "▁Ra y",
+      "▁ Ray",
+      "olog ists",
+      "ologist s",
+      "合同 法",
+      "▁t ags",
+      "▁tag s",
+      "▁ta gs",
+      "▁ tags",
+      "演 唱",
+      "看 起来",
+      "出现 了",
+      "▁t ells",
+      "▁tell s",
+      "▁tel ls",
+      "▁m ales",
+      "▁male s",
+      "▁mal es",
+      "▁ma les",
+      "▁re fr",
+      "▁ref r",
+      "生活 的",
+      "生 活的",
+      "** *",
+      "* **",
+      "市场 的",
+      "市 场的",
+      "解决 方案",
+      "▁K im",
+      "▁Ki m",
+      "▁ Kim",
+      "▁H on",
+      "▁Ho n",
+      "▁op erated",
+      "▁oper ated",
+      "▁operate d",
+      "▁opera ted",
+      "▁ operated",
+      "▁ins ights",
+      "▁insight s",
+      "条 款",
+      "L D",
+      "▁a cts",
+      "▁act s",
+      "▁ac ts",
+      "▁ acts",
+      "oc c",
+      "o cc",
+      "iz ers",
+      "ize rs",
+      "izer s",
+      "is ting",
+      "ist ing",
+      "▁a ctors",
+      "▁act ors",
+      "▁actor s",
+      "▁ actors",
+      "G u",
+      "es c",
+      "e sc",
+      "od ec",
+      "ode c",
+      "o dec",
+      "▁mix ture",
+      "nre quire",
+      "n require",
+      "车 的",
+      "% ;",
+      "血 压",
+      "▁U kr",
+      "ne q",
+      "n eq",
+      "▁re aders",
+      "▁read ers",
+      "▁reader s",
+      "wa ll",
+      "wal l",
+      "w all",
+      "集团 有限公司",
+      "▁R en",
+      "▁Re n",
+      "▁ Ren",
+      "iss ions",
+      "ission s",
+      "un ting",
+      "unt ing",
+      "▁miss ed",
+      "▁mis sed",
+      "j e",
+      "▁ind eed",
+      "f c",
+      "▁t err",
+      "▁te rr",
+      "▁ter r",
+      "▁ terr",
+      "温 暖",
+      "▁re put",
+      "▁rep ut",
+      "和 分析",
+      "不 宜",
+      "是中 国",
+      "是 中国",
+      "星 级",
+      "as ty",
+      "ast y",
+      "a sty",
+      "强 烈",
+      "▁K n",
+      "▁ Kn",
+      "苏 州",
+      "团 体",
+      "og a",
+      "o ga",
+      "▁conn ections",
+      "▁connect ions",
+      "▁connection s",
+      "▁ connections",
+      "文 艺",
+      "▁str ains",
+      "▁strain s",
+      "食 材",
+      "est inal",
+      "▁e lections",
+      "▁elect ions",
+      "▁el ections",
+      "▁election s",
+      "For e",
+      "F ore",
+      "六 月",
+      "界 面",
+      "De lete",
+      "Del ete",
+      "ze ro",
+      "zer o",
+      "z ero",
+      "▁m s",
+      "▁ ms",
+      "▁vis iting",
+      "▁visit ing",
+      "be t",
+      "b et",
+      "▁cut ting",
+      "▁label s",
+      "▁lab els",
+      "▁ labels",
+      "命 运",
+      "▁W hether",
+      "▁Wh ether",
+      "▁ Whether",
+      "▁f ro",
+      "▁fr o",
+      "▁ fro",
+      "ot ypes",
+      "otype s",
+      "otyp es",
+      "o types",
+      "▁F IT",
+      "▁FI T",
+      "cl s",
+      "c ls",
+      "混凝 土",
+      "▁Ex pl",
+      "▁Exp l",
+      "▁ Expl",
+      "iz z",
+      "i zz",
+      "新 型",
+      "不到 庭",
+      "不 到庭",
+      "有时 候",
+      "有 时候",
+      "▁disc rim",
+      "cat ch",
+      "c atch",
+      "Af ter",
+      "A fter",
+      "▁f acing",
+      "▁fac ing",
+      "▁fa cing",
+      "▁ facing",
+      "ST R",
+      "S TR",
+      "或 是",
+      "File s",
+      "Fil es",
+      "Fi les",
+      "F iles",
+      "▁commit tee",
+      "▁ committee",
+      "Py thon",
+      "P ython",
+      "▁col lections",
+      "▁coll ections",
+      "▁collection s",
+      "▁collect ions",
+      "▁ collections",
+      "▁CO VID",
+      "▁ COVID",
+      "▁d iversity",
+      "▁divers ity",
+      "▁E s",
+      "▁ Es",
+      "山 西",
+      "▁b oys",
+      "▁bo ys",
+      "▁boy s",
+      "▁ boys",
+      "接受教育 改造",
+      "▁( \\\\",
+      "▁(\\ \\",
+      "▁ (\\\\",
+      "意味 着",
+      "危险 驾驶",
+      "▁T ODO",
+      "▁TO DO",
+      "▁ TODO",
+      "re ference",
+      "refer ence",
+      "r eference",
+      "加倍支付 迟延履行期间的债务利息",
+      "ev eral",
+      "ever al",
+      "eve ral",
+      "ul ture",
+      "ult ure",
+      "ev ents",
+      "event s",
+      "even ts",
+      "e vents",
+      "Call back",
+      "ok ed",
+      "oke d",
+      "en code",
+      "enc ode",
+      "e ncode",
+      "是一 道",
+      "是 一道",
+      "▁p ilot",
+      "▁pi lot",
+      "▁pil ot",
+      "▁C ur",
+      "▁Cu r",
+      "▁ Cur",
+      "▁pro ven",
+      "▁pr oven",
+      "▁prov en",
+      "▁prove n",
+      "▁Cl oud",
+      "▁Clo ud",
+      "▁ Cloud",
+      "▁re q",
+      "▁r eq",
+      "▁ req",
+      "\\' ))\\",
+      "\\') )\\",
+      "\\')) \\",
+      "\\ '))\\",
+      "同 步",
+      "诉讼 费",
+      "\") );\\",
+      "\")) ;\\",
+      "\" ));\\",
+      "上 涨",
+      "▁develop ers",
+      "▁developer s",
+      "▁ developers",
+      "No n",
+      "N on",
+      "at eway",
+      "ate way",
+      "re po",
+      "rep o",
+      "▁cent ers",
+      "▁center s",
+      "▁ce nters",
+      "Ro ot",
+      "R oot",
+      "较 为",
+      "法律 条文",
+      "ud den",
+      "udd en",
+      "支 部",
+      "▁V ill",
+      "▁Vi ll",
+      "▁Vil l",
+      "cont act",
+      "走 了",
+      "都 很",
+      "线 路",
+      "世界 上",
+      "▁d ial",
+      "▁di al",
+      "▁dia l",
+      "▁ dial",
+      "ak i",
+      "a ki",
+      "每 月",
+      "▁polit icians",
+      "▁politician s",
+      "▁free dom",
+      "▁fre edom",
+      "▁freed om",
+      "op es",
+      "ope s",
+      "o pes",
+      "厨 房",
+      "▁S ea",
+      "▁Se a",
+      "[ [",
+      "刑事 诉讼法",
+      "大 师",
+      "▁pain ting",
+      "▁paint ing",
+      "使用 的",
+      "使 用的",
+      "▁s olving",
+      "▁sol ving",
+      "▁solv ing",
+      "▁ solving",
+      "▁m oth",
+      "▁mot h",
+      "▁mo th",
+      "ber ry",
+      "b erry",
+      "▁r outine",
+      "▁rout ine",
+      "▁ routine",
+      "▁r id",
+      "▁ri d",
+      "▁ rid",
+      "电 路",
+      "▁L earning",
+      "▁Le arning",
+      "▁Lear ning",
+      "▁Learn ing",
+      "▁ Learning",
+      "▁up coming",
+      "▁A mong",
+      "▁Am ong",
+      "▁ap opt",
+      "▁apo pt",
+      "▁ apopt",
+      "▁w alls",
+      "▁wall s",
+      "▁wal ls",
+      "瞬 间",
+      "处 的",
+      "\\' ])\\",
+      "\\'] )\\",
+      "\\']) \\",
+      "\\ '])\\",
+      "▁H am",
+      "▁Ha m",
+      "▁ Ham",
+      "let ter",
+      "lett er",
+      "lette r",
+      "l etter",
+      "▁m ini",
+      "▁min i",
+      "▁mi ni",
+      "▁ mini",
+      "漫 画",
+      "主 角",
+      "gr een",
+      "gre en",
+      "g reen",
+      "ou ston",
+      "ous ton",
+      "oust on",
+      "▁sh ipping",
+      "▁ship ping",
+      "▁ shipping",
+      "传 票",
+      "Pa d",
+      "P ad",
+      "▁s aving",
+      "▁sa ving",
+      "▁sav ing",
+      "▁ saving",
+      "相 反",
+      "▁A uto",
+      "▁Aut o",
+      "▁Au to",
+      "▁ Auto",
+      "灵 活",
+      "给 了",
+      "▁L uc",
+      "▁Lu c",
+      "▁ Luc",
+      "▁de script",
+      "▁des cript",
+      "▁desc ript",
+      "学 者",
+      "Par am",
+      "Pa ram",
+      "P aram",
+      "▁g lob",
+      "▁gl ob",
+      "▁glo b",
+      "▁ glob",
+      "▁S ym",
+      "▁Sy m",
+      "▁ Sym",
+      "福 利",
+      "▁ed ges",
+      "▁edge s",
+      "▁ edges",
+      "▁real ize",
+      "▁ing redients",
+      "▁ingred ients",
+      "▁ingredient s",
+      "▁te eth",
+      "▁tee th",
+      "月 至",
+      "▁p anc",
+      "▁pan c",
+      "▁pa nc",
+      "ab out",
+      "abo ut",
+      "▁Pat ients",
+      "▁Patient s",
+      "物 的",
+      "▁b ot",
+      "▁bo t",
+      "▁ bot",
+      "优 先",
+      "▁voc als",
+      "▁vocal s",
+      "▁n erve",
+      "▁nerv e",
+      "de tails",
+      "det ails",
+      "detail s",
+      "入 住",
+      "force ment",
+      "审 计",
+      "申请 撤诉",
+      "▁maint ained",
+      "▁maintain ed",
+      "\\\\( (\\\\)",
+      "\\\\(( \\\\)",
+      "认 知",
+      "履行 给付",
+      "▁S uch",
+      "▁Su ch",
+      "▁ Such",
+      "▁denomin ator",
+      "▁g al",
+      "▁ga l",
+      "▁ gal",
+      "水 的",
+      "▁cl oser",
+      "▁clos er",
+      "▁close r",
+      "国内 外",
+      "国 内外",
+      "政治 权利",
+      "▁B ox",
+      "▁Bo x",
+      "▁ Box",
+      "▁M R",
+      "▁ MR",
+      "▁magn itude",
+      "电 机",
+      "lin ux",
+      "▁E mp",
+      "▁Em p",
+      "▁ Emp",
+      "▁increasing ly",
+      "}> \\",
+      "} >\\",
+      "▁装 帧",
+      "E P",
+      "又 是",
+      "▁r ows",
+      "▁ro ws",
+      "▁row s",
+      "▁ rows",
+      "▁p aths",
+      "▁pat hs",
+      "▁path s",
+      "▁pa ths",
+      "▁ paths",
+      "学 员",
+      "script s",
+      "scri pts",
+      "▁stand ing",
+      "▁ standing",
+      "▁b an",
+      "▁ba n",
+      "▁ ban",
+      "mo ve",
+      "mov e",
+      "m ove",
+      "fe ature",
+      "feat ure",
+      "▁c lock",
+      "▁cl ock",
+      "▁ clock",
+      "bu l",
+      "b ul",
+      "更 有",
+      "男 性",
+      "nB ack",
+      "n Back",
+      "nP h",
+      "n Ph",
+      "上 一",
+      "▁F ederal",
+      "▁Fed eral",
+      "▁Feder al",
+      "除 以",
+      "en dl",
+      "end l",
+      "t v",
+      "选 手",
+      "▁b odies",
+      "▁bo dies",
+      "▁bod ies",
+      "▁w arning",
+      "▁war ning",
+      "▁warn ing",
+      "▁ warning",
+      "▁bel ieved",
+      "▁believe d",
+      "▁belie ved",
+      "▁C arl",
+      "▁Car l",
+      "▁Ca rl",
+      "▁behavior al",
+      "▁behavi oral",
+      "( @",
+      "▁d ent",
+      "▁de nt",
+      "▁den t",
+      "分 行",
+      "▁con clusion",
+      "fa ll",
+      "f all",
+      "▁Pl aces",
+      "▁Place s",
+      "▁sup erv",
+      "▁super v",
+      "▁ superv",
+      "民 警",
+      "▁A nother",
+      "▁An other",
+      "▁ Another",
+      "qu ality",
+      "qual ity",
+      "q uality",
+      "O B",
+      "全 市",
+      "tl e",
+      "t le",
+      "▁C ond",
+      "▁Con d",
+      "▁Co nd",
+      "▁ Cond",
+      "他 在",
+      "▁c ro",
+      "▁cr o",
+      "▁ cro",
+      "▁m argin",
+      "▁mar gin",
+      "▁marg in",
+      "▁ margin",
+      "教 程",
+      "根据 题意",
+      "抑 制",
+      "号刑事 判决",
+      "Ma r",
+      "M ar",
+      "西 南",
+      "pa per",
+      "p aper",
+      "▁sh ares",
+      "▁share s",
+      "▁sha res",
+      "▁shar es",
+      "▁S pe",
+      "▁Sp e",
+      "▁ Spe",
+      "▁Sc iences",
+      "▁Science s",
+      "▁Sci ences",
+      "▁e lig",
+      "▁el ig",
+      "▁character ize",
+      "▁f iber",
+      "▁fib er",
+      "▁fi ber",
+      "▁ fiber",
+      "▁O p",
+      "▁ Op",
+      "▁PART IC",
+      "▁C ou",
+      "▁Co u",
+      "▁ Cou",
+      "CH ANT",
+      "CHA NT",
+      "▁pre diction",
+      "▁predict ion",
+      "▁pred iction",
+      "▁ prediction",
+      "▁guid ance",
+      "痛 苦",
+      "知识 的",
+      "▁re po",
+      "▁rep o",
+      "▁ repo",
+      "▁h istoric",
+      "▁hist oric",
+      "▁histor ic",
+      "▁ historic",
+      "了 吗",
+      "he mes",
+      "hem es",
+      "heme s",
+      "boot strap",
+      "▁organ ized",
+      "▁organize d",
+      "▁ organized",
+      "▁s ole",
+      "▁so le",
+      "▁sol e",
+      "▁ sole",
+      "▁bed room",
+      "▁ bedroom",
+      "▁ 高",
+      "百 年",
+      "▁pro min",
+      "▁pr omin",
+      "▁prom in",
+      "upy ter",
+      "art icles",
+      "article s",
+      "D D",
+      "▁m outh",
+      "▁mo uth",
+      "▁ mouth",
+      "uset ts",
+      "id al",
+      "ida l",
+      "i dal",
+      "▁amount s",
+      "▁pol ym",
+      "▁poly m",
+      "ol y",
+      "o ly",
+      "ach usetts",
+      "那 就",
+      "▁H enry",
+      "▁Hen ry",
+      "nD uring",
+      "n During",
+      "▁exec ute",
+      "▁execut e",
+      "▁ execute",
+      "▁go ods",
+      "▁good s",
+      "▁ goods",
+      "土 壤",
+      "▁facilit ate",
+      "nS c",
+      "n Sc",
+      "▁com plicated",
+      "▁compl icated",
+      "▁dom ains",
+      "▁domain s",
+      "▁ domains",
+      "T F",
+      "▁D an",
+      "▁Da n",
+      "▁ Dan",
+      "积 累",
+      "ou st",
+      "ous t",
+      "o ust",
+      "ho ur",
+      "hou r",
+      "h our",
+      "锻 炼",
+      "Ra nge",
+      "R ange",
+      "er as",
+      "era s",
+      "e ras",
+      "nH i",
+      "n Hi",
+      "▁s can",
+      "▁sc an",
+      "▁ scan",
+      "▁public ation",
+      "▁ publication",
+      "剥夺 政治权利",
+      "▁involve ment",
+      "oo p",
+      "o op",
+      "▁att ributes",
+      "▁attribute s",
+      "▁ attributes",
+      "ic ture",
+      "ict ure",
+      "AN CE",
+      "依 赖",
+      "Pa t",
+      "P at",
+      "om ething",
+      "ome thing",
+      "omet hing",
+      "ometh ing",
+      "▁J on",
+      "▁Jo n",
+      "▁ Jon",
+      "▁合作 作品",
+      "专 项",
+      "▁l ay",
+      "▁la y",
+      "▁ lay",
+      "st ein",
+      "ste in",
+      "s tein",
+      "▁W alk",
+      "▁Wal k",
+      "▁ Walk",
+      "▁after noon",
+      "▁rel igious",
+      "▁relig ious",
+      "U C",
+      "Map ping",
+      "Ma pping",
+      "M apping",
+      "AN G",
+      "A NG",
+      "还没 有",
+      "还 没有",
+      "Gi t",
+      "G it",
+      "拒不 到庭",
+      "拒 不到庭",
+      "▁ar bitr",
+      "浓 度",
+      "▁val idation",
+      "▁valid ation",
+      "▁ validation",
+      "面 向",
+      "▁ 中",
+      "il ateral",
+      "▁m i",
+      "▁ mi",
+      "▁D ocument",
+      "▁Doc ument",
+      "▁ Document",
+      "公 主",
+      "ER CHANT",
+      "▁under w",
+      "de sign",
+      "des ign",
+      "五 日",
+      "▁a den",
+      "▁ad en",
+      "▁ aden",
+      "ik a",
+      "i ka",
+      "▁foc uses",
+      "▁focus es",
+      "ch i",
+      "c hi",
+      "▁p it",
+      "▁pi t",
+      "▁ pit",
+      "九 日",
+      "不 受",
+      "从 小",
+      "▁Com ponent",
+      "▁ Component",
+      "ig a",
+      "i ga",
+      "▁predict ed",
+      "▁pred icted",
+      "▁A uth",
+      "▁Aut h",
+      "▁Au th",
+      "▁ Auth",
+      "中华人民共和国 行政",
+      "关 闭",
+      "▁mat hematics",
+      "▁math ematics",
+      "▁mathematic s",
+      "人 为",
+      "无异 议",
+      "无 异议",
+      "谢 谢",
+      "▁H N",
+      "▁ HN",
+      "▁se arching",
+      "▁search ing",
+      "▁sear ching",
+      "▁sign ature",
+      "▁ signature",
+      "▁d inner",
+      "▁din ner",
+      "▁perfect ly",
+      "▁advert ising",
+      "▁advertis ing",
+      "ac er",
+      "ace r",
+      "a cer",
+      "▁en ables",
+      "▁enable s",
+      "言 情",
+      "% %",
+      "nb sp",
+      "ya ml",
+      "yam l",
+      "y aml",
+      "▁max im",
+      "▁ma xim",
+      "aw ay",
+      "awa y",
+      "a way",
+      "▁numer ical",
+      "▁numeric al",
+      "ac ker",
+      "ack er",
+      "▁ 小",
+      "▁F air",
+      "▁Fa ir",
+      "▁ag encies",
+      "Value s",
+      "Val ues",
+      "▁n eck",
+      "▁ne ck",
+      "▁nec k",
+      "▁ neck",
+      "▁P URPOSE",
+      "▁J oe",
+      "▁Jo e",
+      "▁ Joe",
+      "接 着",
+      "农 户",
+      "财产 可供执行",
+      "同 期",
+      "burg h",
+      "bur gh",
+      "se rial",
+      "ser ial",
+      "s erial",
+      "con sin",
+      "cons in",
+      "的 朋友",
+      "对 被告",
+      "sup port",
+      "上 去",
+      "山西 省",
+      "山 西省",
+      "赚 钱",
+      "dr iver",
+      "driv er",
+      "drive r",
+      "d river",
+      "▁And rew",
+      "▁Andre w",
+      "▁Andr ew",
+      "▁ Andrew",
+      "▁dis covery",
+      "▁disc overy",
+      "▁discover y",
+      "▁ discovery",
+      "▁p riority",
+      "▁prior ity",
+      "▁priorit y",
+      "▁priori ty",
+      "▁ priority",
+      "反 正",
+      "▁pro files",
+      "▁profile s",
+      "▁prof iles",
+      "▁profil es",
+      "▁ profiles",
+      "▁ch ampionship",
+      "▁champion ship",
+      "▁champions hip",
+      "▁k il",
+      "▁ki l",
+      "▁ kil",
+      "▁con stitu",
+      "▁const itu",
+      "▁constit u",
+      "ro phy",
+      "rop hy",
+      "roph y",
+      "▁w arn",
+      "▁war n",
+      "▁wa rn",
+      "▁ warn",
+      "▁inhib itors",
+      "▁inhibitor s",
+      "▁inhibit ors",
+      "相 似",
+      "▁G reek",
+      "▁Gr eek",
+      "▁Gree k",
+      "▁Gre ek",
+      "▁prolif eration",
+      "▁prolifer ation",
+      "提出 的",
+      "提 出的",
+      "定 理",
+      "th eme",
+      "the me",
+      "them e",
+      "t heme",
+      "各 自",
+      "▁mar riage",
+      "ho r",
+      "h or",
+      "qu ir",
+      "的 技术",
+      "▁st ands",
+      "▁stand s",
+      "▁re venue",
+      "▁rev enue",
+      "▁reven ue",
+      "▁D igital",
+      "▁Dig ital",
+      "▁ Digital",
+      "\\\\ ,",
+      "第二百 六",
+      "▁ex clusive",
+      "▁exclus ive",
+      "ave r",
+      "av er",
+      "a ver",
+      "ou tes",
+      "out es",
+      "oute s",
+      "o utes",
+      "大 大",
+      "▁de grad",
+      "▁deg rad",
+      "fl ag",
+      "f lag",
+      "de sc",
+      "des c",
+      "d esc",
+      "▁D ou",
+      "▁Do u",
+      "▁J oseph",
+      "▁Jose ph",
+      "▁Jos eph",
+      "▁produ ces",
+      "▁produce s",
+      "▁prod uces",
+      "▁at tr",
+      "▁att r",
+      "▁ attr",
+      "意 大利",
+      "ro py",
+      "rop y",
+      "r opy",
+      "▁m align",
+      "▁mal ign",
+      "辅 导",
+      "框 架",
+      "▁mat hematical",
+      "▁mathematic al",
+      "万 人",
+      "scan f",
+      "危险驾驶 罪",
+      "en th",
+      "ent h",
+      "e nth",
+      "][ \\'",
+      "] [\\'",
+      "ri cular",
+      "ric ular",
+      "ricula r",
+      "r icular",
+      "▁d rawn",
+      "▁dr awn",
+      "▁draw n",
+      "上 班",
+      "大 赛",
+      "\"] ,",
+      "\" ],",
+      "▁d oors",
+      "▁do ors",
+      "▁door s",
+      "▁ doors",
+      "qu estion",
+      "ques tion",
+      "quest ion",
+      "若干 问题",
+      "早 期",
+      "▁concer ning",
+      "▁concern ing",
+      "九 月",
+      "se n",
+      "s en",
+      "切 实",
+      "餐 馆",
+      "▁b ass",
+      "▁bas s",
+      "▁ba ss",
+      "▁ bass",
+      "Log in",
+      "Lo gin",
+      "wort h",
+      "wor th",
+      "w orth",
+      "▁app ointment",
+      "▁appoint ment",
+      "密 度",
+      "▁' /",
+      "▁ '/",
+      "f usion",
+      "ja r",
+      "j ar",
+      "}\\\\ )\\",
+      "}\\\\) \\",
+      "} \\\\)\\",
+      "ERCHANT ABILITY",
+      "med iated",
+      "media ted",
+      "mediate d",
+      "pl ays",
+      "play s",
+      "▁en dl",
+      "▁end l",
+      "▁ endl",
+      "dep th",
+      "化 妆",
+      "▁be am",
+      "▁ beam",
+      "yn t",
+      "y nt",
+      "Id ent",
+      "▁p lain",
+      "▁pl ain",
+      "▁ plain",
+      "de velop",
+      "devel op",
+      "一 是",
+      "ha rd",
+      "har d",
+      "h ard",
+      "name d",
+      "na med",
+      "nam ed",
+      "n amed",
+      "F R",
+      "居 然",
+      "工程 有限公司",
+      "▁B ol",
+      "▁Bo l",
+      "观 念",
+      "od s",
+      "o ds",
+      "▁cr isis",
+      "▁cris is",
+      "▁temper atures",
+      "▁temperature s",
+      "nO ther",
+      "n Other",
+      "纳 入",
+      "▁d x",
+      "▁ dx",
+      "ul pt",
+      "ulp t",
+      "st age",
+      "sta ge",
+      "s tage",
+      "en ess",
+      "ene ss",
+      "enes s",
+      "e ness",
+      "Co r",
+      "C or",
+      "▁Conn ect",
+      "▁ Connect",
+      "н а",
+      "整 合",
+      "▁princ iples",
+      "▁principle s",
+      "▁princip les",
+      "业 主",
+      "京 东",
+      "有 权",
+      "本 质",
+      "▁trans action",
+      "▁ transaction",
+      "ser ve",
+      "serv e",
+      "s erve",
+      "这 两",
+      "▁w et",
+      "▁we t",
+      "l ibrary",
+      "▁co efficient",
+      "▁coeff icient",
+      "中华人民共和国 刑事诉讼法",
+      "\"` \\",
+      "\" `\\",
+      "\") )",
+      "\" ))",
+      "▁doll ars",
+      "▁dollar s",
+      "FT WARE",
+      "iv ari",
+      "iva ri",
+      "ivar i",
+      "i vari",
+      "▁ne got",
+      "▁neg ot",
+      "J S",
+      "▁re in",
+      "▁r ein",
+      "▁ rein",
+      "电视 剧",
+      "欠 款",
+      "▁PARTIC ULAR",
+      "▁op erate",
+      "▁oper ate",
+      "▁opera te",
+      "my s",
+      "m ys",
+      "(' /",
+      "( '/",
+      "便 利",
+      "ir ement",
+      "ire ment",
+      "▁improve ments",
+      "▁improvement s",
+      "▁improv ements",
+      "▁div ide",
+      "▁divid e",
+      "立 方",
+      "pu r",
+      "p ur",
+      "▁sch olar",
+      "▁W in",
+      "▁Wi n",
+      "▁ Win",
+      "pl acement",
+      "place ment",
+      "plac ement",
+      "▁re direct",
+      "▁red irect",
+      "▁ redirect",
+      "▁Im port",
+      "▁Imp ort",
+      "▁ Import",
+      "▁check ed",
+      "▁ checked",
+      "形 势",
+      "▁ad oles",
+      "用 途",
+      "▁draw ing",
+      "▁opt ional",
+      "▁option al",
+      "▁ optional",
+      "▁football er",
+      "▁ footballer",
+      "ag ues",
+      "ague s",
+      "agu es",
+      "起 了",
+      "传 递",
+      "nS e",
+      "n Se",
+      "▁ge ometry",
+      "▁geomet ry",
+      "▁geo metry",
+      "▁geom etry",
+      "▁ geometry",
+      "▁fe els",
+      "▁feel s",
+      "▁fee ls",
+      "count ry",
+      "cou ntry",
+      "es ides",
+      "esi des",
+      "▁observ ation",
+      "▁obs ervation",
+      "联 社",
+      "小 心",
+      "▁Is lam",
+      "ag raph",
+      "agr aph",
+      "agra ph",
+      "a graph",
+      "婴 儿",
+      "fe ed",
+      "fee d",
+      "f eed",
+      "▁coeff icients",
+      "▁coefficient s",
+      "▁f ort",
+      "▁for t",
+      "▁fo rt",
+      "▁ fort",
+      "]: \\",
+      "] :\\",
+      "▁D A",
+      "▁ DA",
+      "称 号",
+      "州市 中级人民法院",
+      "州 市中级人民法院",
+      "▁kin ase",
+      "▁ kinase",
+      "族 自治",
+      "nA re",
+      "nAr e",
+      "n Are",
+      "nN otes",
+      "nNo tes",
+      "nNot es",
+      "nNote s",
+      "n Notes",
+      "gr essive",
+      "gress ive",
+      "g ressive",
+      "▁camp us",
+      "▁cam pus",
+      "` `",
+      "马 上",
+      "▁cap abilities",
+      "七 日",
+      "求 得",
+      "用 了",
+      "▁sup pl",
+      "▁A f",
+      "▁ Af",
+      "Ro le",
+      "R ole",
+      "▁chrom os",
+      "▁dec ade",
+      "恢复 执行",
+      "t b",
+      "▁ 大",
+      "el en",
+      "ele n",
+      "e len",
+      "▁m ock",
+      "▁mo ck",
+      "▁ mock",
+      "▁\" #",
+      "▁ \"#",
+      "ag a",
+      "a ga",
+      "▁ar rived",
+      "▁arr ived",
+      "▁arriv ed",
+      "▁arrive d",
+      "▁T aylor",
+      "▁Tay lor",
+      "Ad apter",
+      "▁sc atter",
+      "▁ scatter",
+      "▁reason able",
+      "▁bo at",
+      "▁ boat",
+      "▁m oves",
+      "▁mov es",
+      "▁move s",
+      "▁mo ves",
+      "▁ moves",
+      "▁F A",
+      "▁ FA",
+      "功 效",
+      "▁p ermanent",
+      "▁perman ent",
+      "V er",
+      "▁N avy",
+      "▁Na vy",
+      "▁Nav y",
+      "▁f ees",
+      "▁fe es",
+      "▁fee s",
+      "消 化",
+      "如何 看待",
+      "助 理",
+      "el le",
+      "ell e",
+      "v s",
+      "致力 于",
+      "▁com pletion",
+      "▁comp letion",
+      "▁complet ion",
+      "▁ completion",
+      "ES T",
+      "E ST",
+      "▁t ough",
+      "▁to ugh",
+      "▁tou gh",
+      "▁toug h",
+      "▁r ib",
+      "▁ri b",
+      "▁ rib",
+      "ex press",
+      "exp ress",
+      "expr ess",
+      "or row",
+      "orr ow",
+      "▁P ot",
+      "▁Po t",
+      "▁ Pot",
+      "▁Brit ain",
+      "▁Bri tain",
+      "不可 能",
+      "不 可能",
+      "回 收",
+      ") {",
+      "同学 们",
+      "▁Ch ange",
+      "▁Cha nge",
+      "▁Chang e",
+      "▁Chan ge",
+      "▁ Change",
+      "il ey",
+      "ile y",
+      "i ley",
+      "▁Scot land",
+      "视 为",
+      "▁b utter",
+      "▁but ter",
+      "▁butt er",
+      "算 是",
+      "CR E",
+      "C RE",
+      "部 队",
+      "▁T rust",
+      "▁Tr ust",
+      "▁ Trust",
+      "▁land scape",
+      "▁landsc ape",
+      "定 期",
+      "▁C ro",
+      "▁Cr o",
+      "将 在",
+      "yr oid",
+      "yro id",
+      "y roid",
+      "▁k ing",
+      "▁kin g",
+      "▁ki ng",
+      "▁ king",
+      "N P",
+      "▁Un ivers",
+      "▁Univ ers",
+      "▁Uni vers",
+      "既 然",
+      "▁e le",
+      "▁el e",
+      "▁ ele",
+      "▁h all",
+      "▁ha ll",
+      "▁hal l",
+      "▁ hall",
+      "factor y",
+      "fact ory",
+      "f actory",
+      "y o",
+      "ail ing",
+      "ai ling",
+      "a iling",
+      "▁S H",
+      "▁ SH",
+      "▁diagn osed",
+      "▁diagnose d",
+      "▁comp uting",
+      "▁comput ing",
+      "▁inf inite",
+      "▁infinit e",
+      "庭 审",
+      "ul ous",
+      "ulo us",
+      "Ab stract",
+      "Abs tract",
+      "A bstract",
+      "Or iginal",
+      "Origin al",
+      "大量 的",
+      "大 量的",
+      "上诉 的",
+      "▁F ac",
+      "▁Fa c",
+      "▁ Fac",
+      "▁grad uate",
+      "▁gradu ate",
+      "▁ graduate",
+      "▁e leg",
+      "▁el eg",
+      "▁ele g",
+      "禁 止",
+      "并处罚金 人民币",
+      "并处 罚金人民币",
+      "力 度",
+      "依法 组成合议庭",
+      "▁b ytes",
+      "▁by tes",
+      "▁byte s",
+      "▁ bytes",
+      "▁D ar",
+      "▁Da r",
+      "▁ Dar",
+      "time stamp",
+      "tim estamp",
+      "十一 月",
+      "十 一月",
+      "不 行",
+      "▁M ean",
+      "▁Me an",
+      "▁ Mean",
+      "▁d ining",
+      "▁di ning",
+      "▁din ing",
+      "▁m edi",
+      "▁me di",
+      "▁med i",
+      "▁O nt",
+      "▁On t",
+      "▁k ill",
+      "▁kil l",
+      "▁ki ll",
+      "▁ kill",
+      "semb le",
+      "sem ble",
+      "sembl e",
+      "▁S anta",
+      "▁San ta",
+      "▁Sant a",
+      "赛 季",
+      "方 的",
+      "人 都",
+      "▁int ent",
+      "▁ intent",
+      "走 向",
+      "or b",
+      "o rb",
+      "▁pro portion",
+      "▁prop ortion",
+      "▁proport ion",
+      "▁ proportion",
+      "ax y",
+      "a xy",
+      "Dis play",
+      "▁ult imately",
+      "▁ul timately",
+      "▁ultimate ly",
+      "▁t aught",
+      "▁ta ught",
+      "▁tau ght",
+      "牺 牲",
+      "借款 本金",
+      "ou red",
+      "our ed",
+      "o ured",
+      "he x",
+      "h ex",
+      "▁l icensed",
+      "▁license d",
+      "▁lic ensed",
+      "ow a",
+      "o wa",
+      "sc ious",
+      "sci ous",
+      "▁Sh ould",
+      "▁ Should",
+      "▁partic ipation",
+      "▁particip ation",
+      "▁up grade",
+      "▁upgrad e",
+      "▁ upgrade",
+      "▁res idential",
+      "▁resid ential",
+      "▁resident ial",
+      "ro vers",
+      "rov ers",
+      "rove rs",
+      "r overs",
+      "▁ind icating",
+      "▁indic ating",
+      "▁prom ising",
+      "回 归",
+      "se lected",
+      "select ed",
+      "sel ected",
+      "s elected",
+      "CT ION",
+      "ook ie",
+      "oo kie",
+      "▁kid ney",
+      "▁kidn ey",
+      "▁Des pite",
+      "▁ Despite",
+      "ra nk",
+      "ran k",
+      "r ank",
+      "▁requ irement",
+      "▁require ment",
+      "p u",
+      "}} $",
+      "} }$",
+      "\" _",
+      "Gr id",
+      "G rid",
+      "▁M ir",
+      "▁Mi r",
+      "▁ Mir",
+      "nw here",
+      "n where",
+      "▁V ector",
+      "▁Ve ctor",
+      "▁Vec tor",
+      "▁ Vector",
+      "Im pl",
+      "Imp l",
+      "▁Med icine",
+      "▁Medic ine",
+      "依 旧",
+      "▁Pol and",
+      "▁Po land",
+      "▁recommend ations",
+      "▁recommendation s",
+      "in ations",
+      "ination s",
+      "的 特点",
+      "▁Air port",
+      "▁ar tery",
+      "▁art ery",
+      "▁arter y",
+      "承 认",
+      "传 输",
+      "教育 部",
+      "秘 密",
+      "▁w estern",
+      "▁west ern",
+      "▁ western",
+      "影 视",
+      "引 发",
+      "▁k ne",
+      "▁kn e",
+      "▁N ight",
+      "▁Ni ght",
+      "▁Nig ht",
+      "▁ Night",
+      "多 项",
+      "▁de bt",
+      "▁deb t",
+      "返 还",
+      "随 便",
+      "al d",
+      "a ld",
+      "季 度",
+      "Select or",
+      "▁carcin oma",
+      "全 省",
+      "ET HO",
+      "ETH O",
+      "▁emb ry",
+      "▁embr y",
+      "▁author ities",
+      "PE R",
+      "P ER",
+      "生 涯",
+      "▁A ud",
+      "▁Au d",
+      "▁ Aud",
+      "▁h ockey",
+      "▁hoc key",
+      "职 责",
+      "区 的",
+      "▁De fault",
+      "▁Def ault",
+      "▁ Default",
+      "▁开 本",
+      "▁ 开本",
+      "au x",
+      "a ux",
+      "Eff ect",
+      "OR S",
+      "O RS",
+      "pp ing",
+      "p ping",
+      "专 题",
+      "▁exp lo",
+      "▁expl o",
+      "Dev ice",
+      "▁e igen",
+      "影 片",
+      "分享 到",
+      "ev in",
+      "e vin",
+      ")$ ,",
+      ") $,",
+      "MA X",
+      "M AX",
+      "▁innov ation",
+      "▁e ase",
+      "▁eas e",
+      "同 比",
+      "▁qu ote",
+      "▁quot e",
+      "▁quo te",
+      "▁ quote",
+      "▁W H",
+      "▁ WH",
+      "▁ass ert",
+      "▁ assert",
+      "st ra",
+      "str a",
+      "s tra",
+      "▁s cr",
+      "▁sc r",
+      "▁ scr",
+      "▁ep isodes",
+      "▁epis odes",
+      "▁episode s",
+      "以下 简称",
+      "一直 在",
+      "w t",
+      "等证据 证实",
+      "新 生",
+      "▁a part",
+      "▁ap art",
+      "▁ad equ",
+      "▁ adequ",
+      "▁C at",
+      "▁Ca t",
+      "▁ Cat",
+      "tw o",
+      "t wo",
+      "ta in",
+      "t ain",
+      "演 出",
+      "▁attempt s",
+      "▁represent ative",
+      "▁S imple",
+      "▁Sim ple",
+      "▁ Simple",
+      "▁m etric",
+      "▁met ric",
+      "▁ metric",
+      "传统 的",
+      "▁as set",
+      "▁ass et",
+      "▁ asset",
+      "房地产 开发有限公司",
+      "房地产开发 有限公司",
+      "… \\",
+      "ib er",
+      "ibe r",
+      "i ber",
+      "十三 条",
+      "十 三条",
+      "ro sis",
+      "ros is",
+      "r osis",
+      "RE SS",
+      "RES S",
+      "R ESS",
+      "▁L IC",
+      "▁LI C",
+      "▁ LIC",
+      "ans wer",
+      "a nswer",
+      "by tes",
+      "byte s",
+      "b ytes",
+      "Act ive",
+      "em p",
+      "e mp",
+      "▁h am",
+      "▁ha m",
+      "▁ ham",
+      "领 先",
+      "初 中",
+      "am big",
+      "amb ig",
+      "▁along side",
+      "uc id",
+      "u cid",
+      "▁sub mitted",
+      "▁submit ted",
+      "▁Ex press",
+      "▁Exp ress",
+      "▁ Express",
+      "▁B ern",
+      "▁Be rn",
+      "▁Ber n",
+      "▁ Bern",
+      "▁st yl",
+      "▁ styl",
+      "进 去",
+      "▁Ob ama",
+      "自 信",
+      "▁con versation",
+      "▁convers ation",
+      "则 是",
+      "政 协",
+      "▁m ask",
+      "▁ma sk",
+      "▁mas k",
+      "▁ mask",
+      "反 而",
+      "显 得",
+      "gy pt",
+      "▁period s",
+      "▁peri ods",
+      "cre d",
+      "cr ed",
+      "c red",
+      "am ount",
+      "amo unt",
+      "a mount",
+      "加 盟",
+      "三 日",
+      "▁T itle",
+      "▁Ti tle",
+      "▁Tit le",
+      "▁ Title",
+      "▁hon or",
+      "▁ho nor",
+      "▁host ed",
+      "▁ho sted",
+      "▁ hosted",
+      "他 是",
+      "▁t ech",
+      "▁te ch",
+      "▁ tech",
+      "赢 得",
+      "▁week ly",
+      "▁ weekly",
+      "] $",
+      "▁s itting",
+      "▁sit ting",
+      "▁t race",
+      "▁tr ace",
+      "▁tra ce",
+      "▁ trace",
+      "分 辨",
+      "▁re mark",
+      "▁rem ark",
+      "▁ remark",
+      "全 村",
+      "▁stop ped",
+      "▁sto pped",
+      "direct ory",
+      "director y",
+      "▁re ward",
+      "▁r eward",
+      "▁ reward",
+      "▁dys function",
+      "▁add resses",
+      "▁address es",
+      "▁addr esses",
+      "▁ addresses",
+      "▁g on",
+      "▁go n",
+      "▁ gon",
+      "▁Reg ion",
+      "▁ Region",
+      "▁E V",
+      "▁ EV",
+      "▁S tan",
+      "▁St an",
+      "▁Sta n",
+      "requ ency",
+      "原 式",
+      "寿 命",
+      "▁B a",
+      "▁ Ba",
+      "book s",
+      "bo oks",
+      "b ooks",
+      "▁meet ings",
+      "▁meeting s",
+      "▁S ir",
+      "▁Si r",
+      "▁R ights",
+      "▁Right s",
+      "听 到",
+      "敏 感",
+      "▁Republic an",
+      "▁ex planation",
+      "▁explan ation",
+      "nP op",
+      "nPo p",
+      "n Pop",
+      "E d",
+      "in line",
+      "i nline",
+      "三 条",
+      "若干问题 的规定",
+      "若干问题的 规定",
+      "ow ner",
+      "own er",
+      "▁m yst",
+      "▁my st",
+      "▁mys t",
+      "O H",
+      "产生 的",
+      "产 生的",
+      "▁S us",
+      "▁Su s",
+      "od ynam",
+      "ody nam",
+      "o dynam",
+      "▁E th",
+      "▁Et h",
+      "▁ Eth",
+      "▁th rom",
+      "▁thr om",
+      "▁thro m",
+      "▁ 马",
+      "▁s kill",
+      "▁sk ill",
+      "▁ski ll",
+      "▁ skill",
+      "缺 点",
+      "met ic",
+      "me tic",
+      "m etic",
+      "违 约",
+      "▁bel ief",
+      "▁belie f",
+      "▁b iology",
+      "▁bi ology",
+      "▁bio logy",
+      "▁ biology",
+      "▁kn ock",
+      "wr ap",
+      "w rap",
+      "ac king",
+      "ack ing",
+      "ad el",
+      "ade l",
+      "a del",
+      "族自治 区",
+      "族 自治区",
+      "▁S pr",
+      "▁Sp r",
+      "▁place holder",
+      "▁ placeholder",
+      "}^{ \\\\",
+      "}^ {\\\\",
+      "} ^{\\\\",
+      "▁T ree",
+      "▁Tr ee",
+      "▁Tre e",
+      "▁ Tree",
+      "了 本案",
+      "▁C as",
+      "▁Ca s",
+      "▁ Cas",
+      "▁satisf action",
+      "▁s creens",
+      "▁sc reens",
+      "▁screen s",
+      "▁ screens",
+      "ri ne",
+      "rin e",
+      "r ine",
+      "rh o",
+      "r ho",
+      "win g",
+      "wi ng",
+      "w ing",
+      "▁ch lor",
+      "▁ chlor",
+      "▁d uplic",
+      "▁du plic",
+      "G T",
+      "at omic",
+      "ato mic",
+      "atom ic",
+      "高 端",
+      "bo olean",
+      "bool ean",
+      "on omy",
+      "onom y",
+      "ono my",
+      "\\' ],\\",
+      "\\'] ,\\",
+      "\\'], \\",
+      "\\ '],\\",
+      "▁C hair",
+      "▁Ch air",
+      "▁Cha ir",
+      "▁comb ine",
+      "▁combin e",
+      "▁ combine",
+      "▁car efully",
+      "▁care fully",
+      "▁careful ly",
+      "li ter",
+      "lit er",
+      "lite r",
+      "l iter",
+      "rupt ion",
+      "ru ption",
+      "rup tion",
+      "ar go",
+      "arg o",
+      "▁per ipheral",
+      "oi s",
+      "o is",
+      "▁r otation",
+      "▁rot ation",
+      "▁ rotation",
+      "▁D er",
+      "▁De r",
+      "▁ Der",
+      "▁c aught",
+      "▁ca ught",
+      "▁m ild",
+      "▁mil d",
+      "▁mi ld",
+      "▁FIT NESS",
+      "▁sus cept",
+      "的 那",
+      "D r",
+      "大 力",
+      "鼠 标",
+      "ier arch",
+      "才 会",
+      "Con st",
+      "Co nst",
+      "Cons t",
+      "plan ation",
+      "的 过程",
+      "cl er",
+      "cle r",
+      "c ler",
+      "▁v on",
+      "▁vo n",
+      "▁ von",
+      "▁phys ics",
+      "▁physic s",
+      "▁physi cs",
+      "▁ physics",
+      "h s",
+      "▁l ights",
+      "▁light s",
+      "▁ lights",
+      "ip se",
+      "ips e",
+      "▁consec utive",
+      "让 他",
+      "延 长",
+      "不 敢",
+      "▁P D",
+      "▁ PD",
+      "仪 器",
+      "fi sh",
+      "f ish",
+      "nes ota",
+      "▁{ @",
+      "规定 的",
+      "规 定的",
+      "▁ 吴",
+      "遭 遇",
+      "▁B ad",
+      "▁Ba d",
+      "▁ Bad",
+      "地 震",
+      "▁C reek",
+      "▁Cr eek",
+      "▁Cre ek",
+      "正 是",
+      "▁Add itionally",
+      "▁Additional ly",
+      "▁Addition ally",
+      "▁pre mium",
+      "▁prem ium",
+      "or er",
+      "ore r",
+      "o rer",
+      "不 懂",
+      "nG et",
+      "nGe t",
+      "n Get",
+      "宇 宙",
+      "无正当理由 拒不到庭",
+      "▁p ars",
+      "▁par s",
+      "▁pa rs",
+      "▁ pars",
+      "人力 资源",
+      "人 力资源",
+      "▁W atch",
+      "▁Wat ch",
+      "▁ Watch",
+      "▁R ub",
+      "▁Ru b",
+      "▁ Rub",
+      "▁d ecom",
+      "▁de com",
+      "▁dec om",
+      "转 变",
+      "BU G",
+      "B UG",
+      "me mber",
+      "mem ber",
+      "m ember",
+      "的 数",
+      "z o",
+      "指定的 期间",
+      "▁synt hes",
+      "▁synth es",
+      "▁ synthes",
+      "▁em ployment",
+      "▁employ ment",
+      "▁anc ient",
+      "表 扬",
+      "▁per mit",
+      "▁perm it",
+      "正确 答案",
+      "olog ically",
+      "ological ly",
+      "ologic ally",
+      "ya rd",
+      "yar d",
+      "y ard",
+      "▁Mass achusetts",
+      "星 期",
+      "很 容易",
+      "▁act ivated",
+      "▁activ ated",
+      "▁activate d",
+      "▁ activated",
+      "▁elev ated",
+      "的 企业",
+      "▁d ocker",
+      "▁doc ker",
+      "▁dock er",
+      "▁ docker",
+      "da rt",
+      "dar t",
+      "d art",
+      "一 只",
+      "起点 网",
+      "▁n m",
+      "▁ nm",
+      "ru le",
+      "r ule",
+      "太 平",
+      "▁aut hent",
+      "▁auth ent",
+      "▁ authent",
+      "▁print ed",
+      "▁ printed",
+      "更 为",
+      "▁expand ed",
+      "▁ expanded",
+      "发 病",
+      "valid ate",
+      "▁S QL",
+      "▁ SQL",
+      "▁inter active",
+      "▁interact ive",
+      "▁ interactive",
+      "iv a",
+      "i va",
+      "é s",
+      "�� �",
+      "� ��",
+      "▁y esterday",
+      "继 承",
+      "▁p romise",
+      "▁prom ise",
+      "▁ promise",
+      "的 结果",
+      "▁cardi ovascular",
+      "▁cardio vascular",
+      "us her",
+      "ush er",
+      "▁R ound",
+      "▁Ro und",
+      "▁Rou nd",
+      "▁ Round",
+      "lo t",
+      "l ot",
+      "▁Pl us",
+      "▁ Plus",
+      "ogen eous",
+      "ogene ous",
+      "ie m",
+      "i em",
+      "直 径",
+      "巨 大",
+      "class es",
+      "c lasses",
+      "▁Jack son",
+      "▁fif th",
+      "▁K ong",
+      "▁Kon g",
+      "▁Ko ng",
+      "▁K orean",
+      "▁Kore an",
+      "▁Korea n",
+      "▁Kor ean",
+      "hav ior",
+      "▁w orse",
+      "▁wor se",
+      "▁wors e",
+      "百 姓",
+      "▁air port",
+      "犯 盗窃罪",
+      "▁p recision",
+      "▁prec ision",
+      "▁ precision",
+      "▁birth day",
+      "▁Br ook",
+      "▁Bro ok",
+      "ny ou",
+      "n you",
+      "▁quant ity",
+      "▁ quantity",
+      "收 获",
+      "▁E mer",
+      "▁Em er",
+      "▁In stead",
+      "▁Inst ead",
+      "企业 信用",
+      "面 上",
+      "舞 蹈",
+      "要求 被告",
+      "▁ter rit",
+      "▁terr it",
+      "▁rest aurants",
+      "▁restaur ants",
+      "▁restaurant s",
+      "▁M ill",
+      "▁Mil l",
+      "▁Mi ll",
+      "▁ Mill",
+      "天 地",
+      "部 的",
+      "▁in verse",
+      "▁ inverse",
+      "传 说",
+      "rad ius",
+      "分析 了",
+      "tel se",
+      "t else",
+      "▁j e",
+      "▁ je",
+      "前 往",
+      "Arg ument",
+      "am ps",
+      "amp s",
+      "发 育",
+      "▁S at",
+      "▁Sa t",
+      "▁ Sat",
+      "ra zy",
+      "raz y",
+      "r azy",
+      "▁plan et",
+      "▁plane t",
+      "▁ planet",
+      "▁lo aded",
+      "▁load ed",
+      "▁ loaded",
+      "An al",
+      "A nal",
+      "写 出",
+      "ole cular",
+      "olec ular",
+      "▁disc ipl",
+      "▁ discipl",
+      "效 应",
+      "▁cr icket",
+      "直接 向",
+      "变 形",
+      "镜 头",
+      "斯 特",
+      "▁t ack",
+      "▁ta ck",
+      "这 位",
+      "▁read only",
+      "▁ readonly",
+      "若干 问题的",
+      "若干问题 的",
+      "▁Pr ime",
+      "▁Prim e",
+      "▁Pri me",
+      "▁ Prime",
+      "▁Pro fessor",
+      "▁Profess or",
+      "著 作",
+      "次 数",
+      "电 商",
+      "▁spec ify",
+      "on omic",
+      "onom ic",
+      "ono mic",
+      "▁invest igators",
+      "▁investig ators",
+      "▁investigator s",
+      "▁par ser",
+      "▁parse r",
+      "▁pars er",
+      "▁ parser",
+      "▁rad ical",
+      "▁radi cal",
+      "ob e",
+      "o be",
+      "▁g lo",
+      "▁gl o",
+      "ly wood",
+      "▁bas ically",
+      "▁basic ally",
+      "▁s ke",
+      "▁sk e",
+      "疼 痛",
+      "▁n g",
+      "▁ ng",
+      "报 送",
+      "▁u int",
+      "▁ui nt",
+      "▁ uint",
+      "oc ate",
+      "oca te",
+      "获 奖",
+      "bs erv",
+      "b serv",
+      "re set",
+      "res et",
+      "▁invest ors",
+      "▁investor s",
+      "▁g rew",
+      "▁gr ew",
+      "▁gre w",
+      "号民事 调解书",
+      "date time",
+      "dat etime",
+      "▁circum st",
+      "▁O K",
+      "▁ OK",
+      "萝 卜",
+      "▁r ating",
+      "▁rat ing",
+      "▁ra ting",
+      "▁ rating",
+      "ad i",
+      "a di",
+      "三 种",
+      "买 了",
+      "十 月",
+      "an ches",
+      "anc hes",
+      "anch es",
+      "anche s",
+      "▁check ing",
+      "导 师",
+      "▁} ;\\",
+      "▁}; \\",
+      "▁ };\\",
+      "at urally",
+      "atural ly",
+      "atur ally",
+      "▁tr uck",
+      "▁phil osoph",
+      "na pp",
+      "nap p",
+      "n app",
+      "转 型",
+      "ogen ous",
+      "▁object ives",
+      "▁objective s",
+      "考 研",
+      "▁mur der",
+      "应 急",
+      "老 公",
+      "七 月",
+      "▁t rick",
+      "▁tr ick",
+      "▁tri ck",
+      "六 日",
+      "Fi nd",
+      "Fin d",
+      "F ind",
+      "▁a bund",
+      "▁ab und",
+      "的 形式",
+      "▁s ending",
+      "▁send ing",
+      "▁sen ding",
+      "pat tern",
+      "公 务",
+      "er mal",
+      "erm al",
+      "▁he av",
+      "Map per",
+      "M apper",
+      "▁an xiety",
+      "▁b atch",
+      "▁bat ch",
+      "▁ batch",
+      "▁v ital",
+      "▁vit al",
+      "▁vi tal",
+      "讲 解",
+      "▁g ained",
+      "▁gain ed",
+      "▁ga ined",
+      "沒 有",
+      "▁im plant",
+      "▁impl ant",
+      "请 您",
+      "v y",
+      "nY ear",
+      "n Year",
+      "▁h yd",
+      "▁hy d",
+      "▁ hyd",
+      "▁love ly",
+      "▁lo vely",
+      "豆瓣 读书",
+      "心 情",
+      "▁d ialog",
+      "▁dial og",
+      "▁dia log",
+      "▁ dialog",
+      "对 被执行人",
+      "▁sup plies",
+      "▁suppl ies",
+      "▁Pl ace",
+      "▁ Place",
+      "en vironment",
+      "environ ment",
+      "管 道",
+      "Pro file",
+      "▁C ancer",
+      "▁Can cer",
+      "▁s le",
+      "▁sl e",
+      "仓 库",
+      "▁princ iple",
+      "▁princip le",
+      "Load er",
+      "Lo ader",
+      "ad v",
+      "a dv",
+      "al o",
+      "a lo",
+      "▁requ ested",
+      "▁request ed",
+      "▁R ight",
+      "▁Ri ght",
+      "▁Rig ht",
+      "▁ Right",
+      "▁St eve",
+      "▁Ste ve",
+      "▁ Steve",
+      "▁crim inal",
+      "▁B all",
+      "▁Bal l",
+      "▁Ba ll",
+      "▁ Ball",
+      "▁kn owing",
+      "▁know ing",
+      "De bug",
+      "Deb ug",
+      "D ebug",
+      "▁L ess",
+      "▁Le ss",
+      "▁Les s",
+      "▁ Less",
+      "一 片",
+      "Micro soft",
+      "M icrosoft",
+      "暴 力",
+      "共产 党",
+      "热 情",
+      "V S",
+      "▁l ic",
+      "▁li c",
+      "▁ lic",
+      "▁des ire",
+      "he el",
+      "hee l",
+      "yd ney",
+      "Bi g",
+      "B ig",
+      "测 定",
+      "取值 范围",
+      "▁l ibraries",
+      "▁libr aries",
+      "▁ libraries",
+      "ç ão",
+      "▁child hood",
+      "Res t",
+      "Re st",
+      "R est",
+      "本裁定书 送达后",
+      "▁cl osest",
+      "▁clos est",
+      "▁close st",
+      "▁closes t",
+      "第一百四十 四",
+      "ig ue",
+      "igu e",
+      "i gue",
+      "ie ving",
+      "iev ing",
+      "TH ER",
+      "THE R",
+      "T HER",
+      "x s",
+      "n X",
+      "大 约",
+      "差不 多",
+      "差 不多",
+      "▁} )",
+      "▁ })",
+      "ad esh",
+      "ade sh",
+      "ades h",
+      "少 女",
+      "bu s",
+      "b us",
+      "财 富",
+      "每 日",
+      "ta il",
+      "t ail",
+      "回 复",
+      "In s",
+      "I ns",
+      "al gorithm",
+      "▁r ab",
+      "▁ra b",
+      "▁ rab",
+      "▁C ert",
+      "▁Cer t",
+      "▁Ce rt",
+      "▁ Cert",
+      "▁a ging",
+      "▁ag ing",
+      "▁ aging",
+      "ba s",
+      "b as",
+      "是 连载于",
+      "▁advant ages",
+      "▁advantage s",
+      "st ats",
+      "stat s",
+      "sta ts",
+      "A K",
+      "[\\ \\",
+      "[ \\\\",
+      "▁f requent",
+      "▁frequ ent",
+      "▁freq uent",
+      "履行给付 金钱",
+      "nM ay",
+      "nMa y",
+      "n May",
+      "▁ 上",
+      "sh ell",
+      "she ll",
+      "s hell",
+      "▁man aging",
+      "▁manag ing",
+      "Ch annel",
+      "Chan nel",
+      "种 类",
+      "西 北",
+      "pon ential",
+      "ponent ial",
+      "▁un us",
+      "▁u nus",
+      "农村 商业",
+      "▁Det ermine",
+      "▁Determ ine",
+      "农村 信用",
+      "erv ative",
+      "原 创",
+      "▁dec imal",
+      "▁ decimal",
+      "in ating",
+      "ina ting",
+      "▁aut omatic",
+      "▁autom atic",
+      "▁auto matic",
+      "▁ automatic",
+      "sub set",
+      "了一 些",
+      "了 一些",
+      "lo go",
+      "log o",
+      "l ogo",
+      "主 办",
+      "▁alleg ed",
+      "▁alle ged",
+      "▁fore cast",
+      "▁mit ochond",
+      "▁happ ening",
+      "▁happen ing",
+      "▁gl ut",
+      "▁glu t",
+      "▁ glut",
+      "P K",
+      "离 子",
+      "▁ro bot",
+      "▁rob ot",
+      "▁ robot",
+      "▁com edy",
+      "▁come dy",
+      "▁comed y",
+      "税 务",
+      "加 床",
+      "X T",
+      "ol i",
+      "o li",
+      "利 亚",
+      "病 人",
+      "▁s pr",
+      "▁sp r",
+      "▁ spr",
+      "Con sole",
+      "Cons ole",
+      "无 数",
+      "▁m eat",
+      "▁me at",
+      "▁m ining",
+      "▁min ing",
+      "▁mini ng",
+      "▁mi ning",
+      "▁necess arily",
+      "▁t ea",
+      "▁te a",
+      "缓 解",
+      "do ne",
+      "don e",
+      "d one",
+      "▁th orough",
+      "▁out standing",
+      "摄 像",
+      "▁ant icip",
+      "▁antic ip",
+      "▁C ase",
+      "▁Ca se",
+      "▁Cas e",
+      "▁ Case",
+      "▁St atus",
+      "▁Stat us",
+      "▁ Status",
+      "ar ry",
+      "arr y",
+      "▁sm oking",
+      "▁smo king",
+      "▁exec ution",
+      "▁execut ion",
+      "▁ execution",
+      "O C",
+      "▁d riven",
+      "▁dr iven",
+      "▁dri ven",
+      "▁driv en",
+      "▁drive n",
+      "▁ driven",
+      "拍 卖",
+      "一 套",
+      "配 件",
+      "舒 服",
+      "中华人民共和国 合同法",
+      "▁ne utral",
+      "▁neut ral",
+      "▁neutr al",
+      "▁ neutral",
+      "▁Z bl",
+      "▁ Zbl",
+      "同 事",
+      "Hel lo",
+      "H ello",
+      "al ert",
+      "ale rt",
+      "aler t",
+      "▁LIC ENSE",
+      "▁ LICENSE",
+      "带 领",
+      "▁E dition",
+      "▁Ed ition",
+      "▁Edit ion",
+      "▁ret ired",
+      "▁retire d",
+      "▁m eta",
+      "▁me ta",
+      "▁met a",
+      "▁ meta",
+      "宿 舍",
+      "审理 过程中",
+      "▁Wis consin",
+      "ap os",
+      "apo s",
+      "a pos",
+      "多 人",
+      "第三 款",
+      "第 三款",
+      "▁Color ado",
+      "▁se emed",
+      "▁see med",
+      "▁seem ed",
+      "as ka",
+      "ask a",
+      "a ska",
+      "监狱 服刑",
+      "▁design er",
+      "▁re dist",
+      "▁red ist",
+      "▁redis t",
+      "▁B efore",
+      "▁Be fore",
+      "▁ Before",
+      "判决 执行",
+      "▁conf lict",
+      "▁conflic t",
+      "uss ion",
+      "窗 口",
+      "Re v",
+      "R ev",
+      "办 学",
+      "退 休",
+      "f b",
+      "一 句",
+      "ust ral",
+      "ustr al",
+      "小 型",
+      "回 忆",
+      "主 管",
+      "▁in formed",
+      "▁inform ed",
+      "科学 技术",
+      "▁C M",
+      "▁ CM",
+      "▁v ictory",
+      "▁vict ory",
+      "ug s",
+      "u gs",
+      "▁f o",
+      "▁ fo",
+      "如实 供述",
+      "▁gr anted",
+      "▁grant ed",
+      "▁gran ted",
+      "场 上",
+      "▁me asuring",
+      "▁meas uring",
+      "▁sam pling",
+      "▁ sampling",
+      "nl et",
+      "n let",
+      "Si ng",
+      "Sin g",
+      "S ing",
+      "▁R oom",
+      "▁Ro om",
+      "▁ Room",
+      "人 事",
+      "ap point",
+      "app oint",
+      "▁w earing",
+      "▁we aring",
+      "▁wear ing",
+      "▁e ra",
+      "▁er a",
+      "▁ era",
+      "nM embers",
+      "n Members",
+      "▁pro gn",
+      "▁pr ogn",
+      "//////// ////////",
+      "▁B C",
+      "▁ BC",
+      "提供 被执行人",
+      "▁l ighting",
+      "▁light ing",
+      "▁r ail",
+      "▁ra il",
+      "▁ rail",
+      "认罪 悔罪",
+      "后 果",
+      "▁inter section",
+      "▁intersect ion",
+      "▁check s",
+      "▁che cks",
+      "▁ checks",
+      "城 镇",
+      "pay load",
+      "Not e",
+      "No te",
+      "N ote",
+      "▁Build ing",
+      "▁ Building",
+      "▁Con tin",
+      "▁Cont in",
+      "▁ Contin",
+      "分享到 朋友圈",
+      "V C",
+      "sd k",
+      "s dk",
+      "nR eg",
+      "nRe g",
+      "n Reg",
+      "跟 着",
+      "▁mat ters",
+      "▁matter s",
+      "▁matt ers",
+      "▁B on",
+      "▁Bo n",
+      "▁ Bon",
+      "▁S hop",
+      "▁Sh op",
+      "▁ Shop",
+      "前 面",
+      "▁st retch",
+      "▁str etch",
+      "▁stret ch",
+      "nM ore",
+      "nMor e",
+      "n More",
+      "小 孩",
+      "▁A sh",
+      "▁As h",
+      "三角 形的",
+      "三角形 的",
+      "▁S mall",
+      "▁Sm all",
+      "▁ Small",
+      "▁Col umbia",
+      "▁Columb ia",
+      "名 词",
+      "To ol",
+      "Too l",
+      "T ool",
+      "▁E lement",
+      "▁El ement",
+      "▁Ele ment",
+      "▁ Element",
+      "对应 的",
+      "对 应的",
+      "▁hydro gen",
+      "▁hyd rogen",
+      "▁T urn",
+      "▁Tur n",
+      "▁Tu rn",
+      "▁ Turn",
+      "▁B ed",
+      "▁Be d",
+      "▁ Bed",
+      "amil ton",
+      "▁l ie",
+      "▁li e",
+      "▁ lie",
+      "长 时间",
+      "新 建",
+      "▁determ ination",
+      "▁determin ation",
+      "新 鲜",
+      "高 兴",
+      "all enge",
+      "alle nge",
+      "alleng e",
+      "▁F ox",
+      "▁Fo x",
+      "▁ Fox",
+      "▁spe aking",
+      "▁speak ing",
+      "▁ speaking",
+      "激 光",
+      "nA m",
+      "n Am",
+      "ap ore",
+      "apor e",
+      "apo re",
+      "▁HT TP",
+      "▁ HTTP",
+      "th al",
+      "tha l",
+      "t hal",
+      "▁Qu een",
+      "▁Que en",
+      "▁ Queen",
+      "▁t rig",
+      "▁tr ig",
+      "▁tri g",
+      "▁ trig",
+      "he matic",
+      "hem atic",
+      "hemat ic",
+      "hema tic",
+      "h ematic",
+      "出 席",
+      "球 员",
+      "A v",
+      "con dition",
+      "cond ition",
+      "qu estions",
+      "question s",
+      "quest ions",
+      "打 击",
+      "▁spec tro",
+      "▁spect ro",
+      "ra se",
+      "ras e",
+      "r ase",
+      "▁state ments",
+      "▁stat ements",
+      "▁statement s",
+      "it us",
+      "itu s",
+      "▁WARRANT Y",
+      "美 容",
+      "▁k ernel",
+      "▁ker nel",
+      "▁ kernel",
+      "et ies",
+      "eti es",
+      "e ties",
+      "nC lick",
+      "nCl ick",
+      "n Click",
+      "▁t ube",
+      "▁tub e",
+      "▁tu be",
+      "▁ tube",
+      "▁enter tainment",
+      "▁entertain ment",
+      "▁entert ainment",
+      "▁expect ations",
+      "▁expectation s",
+      "▁SO FTWARE",
+      "pt on",
+      "pto n",
+      "p ton",
+      "哥 哥",
+      "▁z ip",
+      "▁ zip",
+      "印 刷",
+      "h m",
+      "之日起 计算",
+      "▁D aniel",
+      "▁Dan iel",
+      "▁Dani el",
+      "▁ Daniel",
+      "sp eed",
+      "spe ed",
+      "s peed",
+      "无论 是",
+      "▁im possible",
+      "▁imp ossible",
+      "sock et",
+      "s ocket",
+      "拓 展",
+      "▁f iled",
+      "▁fil ed",
+      "▁file d",
+      "▁fi led",
+      "▁p aying",
+      "▁pay ing",
+      "▁pa ying",
+      "▁t y",
+      "▁ ty",
+      "又 有",
+      "▁res piratory",
+      "▁respir atory",
+      "at om",
+      "ato m",
+      "a tom",
+      "St ud",
+      "cu e",
+      "c ue",
+      "▁U t",
+      "am ond",
+      "amo nd",
+      "amon d",
+      "a mond",
+      "将 文章",
+      "▁h ole",
+      "▁hol e",
+      "▁ho le",
+      "▁ hole",
+      "国 庆",
+      "电子 商务",
+      "前 后",
+      "message s",
+      "mess ages",
+      "m essages",
+      "生 日",
+      "▁con test",
+      "▁cont est",
+      "▁co ntest",
+      "pro fit",
+      "prof it",
+      "▁aff ects",
+      "▁affect s",
+      "Le ft",
+      "L eft",
+      "▁s orry",
+      "▁sor ry",
+      "il st",
+      "ils t",
+      "i lst",
+      "冲 击",
+      "单 纯",
+      "▁B lock",
+      "▁Bl ock",
+      "▁Blo ck",
+      "▁ Block",
+      "on da",
+      "ond a",
+      "o nda",
+      "fm t",
+      "f mt",
+      "▁m eal",
+      "▁me al",
+      "▁ meal",
+      "▁mom ents",
+      "▁moment s",
+      "▁mo ments",
+      "ke s",
+      "k es",
+      "at han",
+      "ath an",
+      "atha n",
+      "a than",
+      "▁H op",
+      "▁Ho p",
+      "▁ Hop",
+      "▁coron ary",
+      "少 数",
+      "照 顾",
+      "\\\\ \\",
+      "\\ \\\\",
+      "生效 法律文书确定的义务",
+      "生效法律文书 确定的义务",
+      "腾 讯",
+      "审理 了本案",
+      "OU T",
+      "O UT",
+      "nP re",
+      "nPr e",
+      "n Pre",
+      "财产 保全",
+      "▁b anks",
+      "▁bank s",
+      "▁ban ks",
+      "▁determ ining",
+      "▁determin ing",
+      "团 结",
+      "ad dr",
+      "add r",
+      "的就 是",
+      "的 就是",
+      "▁di agram",
+      "▁dia gram",
+      "▁j udge",
+      "▁jud ge",
+      "▁judg e",
+      "ep h",
+      "e ph",
+      "▁r outes",
+      "▁ro utes",
+      "▁rout es",
+      "▁route s",
+      "▁rou tes",
+      "▁ routes",
+      "▁O k",
+      "▁ Ok",
+      "机器 人",
+      "的 有",
+      "▁s olved",
+      "▁sol ved",
+      "▁solve d",
+      "▁solv ed",
+      "之 路",
+      "了 吧",
+      "An y",
+      "A ny",
+      "Me m",
+      "M em",
+      "▁I C",
+      "▁ IC",
+      "所 需",
+      "将文章 分享到朋友圈",
+      "▁L atin",
+      "▁La tin",
+      "▁Lat in",
+      "ar ters",
+      "art ers",
+      "arter s",
+      "arte rs",
+      "B B",
+      "st able",
+      "sta ble",
+      "s table",
+      "半 径",
+      "多 元",
+      "col lection",
+      "coll ection",
+      "collect ion",
+      "▁n erv",
+      "▁ne rv",
+      "▁i p",
+      "▁ ip",
+      "▁P rint",
+      "▁Pr int",
+      "▁Pri nt",
+      "▁ Print",
+      "的 效果",
+      "▁简 介",
+      "▁ 简介",
+      "ot yp",
+      "o typ",
+      "▁c ov",
+      "▁co v",
+      "▁ cov",
+      "in x",
+      "i nx",
+      "▁L y",
+      "▁ Ly",
+      "h h",
+      "并 非",
+      "ac on",
+      "aco n",
+      "a con",
+      "sw itch",
+      "s witch",
+      "粮 食",
+      "▁st uck",
+      "匹 配",
+      "第三 节",
+      "▁claim ed",
+      "▁ claimed",
+      "合 金",
+      "▁R od",
+      "▁Ro d",
+      "ps i",
+      "p si",
+      "▁sugg estions",
+      "▁suggest ions",
+      "▁suggestion s",
+      "江 南",
+      "le gate",
+      "leg ate",
+      "▁recogn ize",
+      "ar ios",
+      "ari os",
+      "ario s",
+      "a rios",
+      "window s",
+      "wind ows",
+      "w indows",
+      "▁re peat",
+      "▁rep eat",
+      "▁repe at",
+      "▁ repeat",
+      "Aut o",
+      "Au to",
+      "A uto",
+      "男 女",
+      "▁d ebug",
+      "▁de bug",
+      "▁deb ug",
+      "▁ debug",
+      "哈尔 滨",
+      "ab eth",
+      "abet h",
+      "abe th",
+      "or ous",
+      "oro us",
+      "o rous",
+      "▁co al",
+      "▁ coal",
+      "▁bre aking",
+      "▁break ing",
+      "▁ breaking",
+      "se ud",
+      "▁K now",
+      "▁Kn ow",
+      "▁ Know",
+      "▁Tr aining",
+      "▁Tra ining",
+      "▁Train ing",
+      "▁ Training",
+      "▁le af",
+      "▁ leaf",
+      "和 其他",
+      "六个 月",
+      "六 个月",
+      "▁prov ince",
+      "▁ province",
+      "▁al ert",
+      "▁ale rt",
+      "▁ alert",
+      "姑 娘",
+      "▁E nc",
+      "▁En c",
+      "▁ Enc",
+      "▁de aling",
+      "▁deal ing",
+      "ti p",
+      "t ip",
+      "保 修",
+      "未按 本判决",
+      "▁` `",
+      "▁ ``",
+      "ad der",
+      "add er",
+      "▁ke eps",
+      "▁keep s",
+      "昨 天",
+      "要 是",
+      "wh at",
+      "w hat",
+      "信息 公示",
+      "同 类",
+      "▁s udden",
+      "生 意",
+      "的 解",
+      "下 了",
+      "公 斤",
+      "贫 困",
+      "St ep",
+      "Ste p",
+      "ar et",
+      "are t",
+      "a ret",
+      "履行给付金钱 义务",
+      "▁mod erate",
+      "▁mode rate",
+      "▁moder ate",
+      "ac her",
+      "ach er",
+      "ache r",
+      "a cher",
+      "re gon",
+      "reg on",
+      "刑 罚",
+      "直 角",
+      "▁l unch",
+      "▁lu nch",
+      "ze ch",
+      "z ech",
+      "失 去",
+      "忘 记",
+      "成 人",
+      "当 天",
+      "▁Pak istan",
+      "▁H arr",
+      "▁Har r",
+      "▁Ha rr",
+      "▁inhib itor",
+      "▁inhibit or",
+      "▁E astern",
+      "▁East ern",
+      "▁Easter n",
+      "▁Eas tern",
+      "sc ill",
+      "sci ll",
+      "ul p",
+      "u lp",
+      "▁horm one",
+      "▁Just ice",
+      "'] \\",
+      "' ]\\",
+      "二 日",
+      "▁wire less",
+      "▁est imation",
+      "▁estim ation",
+      "▁D onald",
+      "▁Don ald",
+      "▁ Donald",
+      "▁f lash",
+      "▁fl ash",
+      "▁ flash",
+      "▁p icking",
+      "▁pick ing",
+      "▁pic king",
+      "所在 地",
+      "所 在地",
+      "▁st ead",
+      "▁ste ad",
+      "▁ stead",
+      "这两 个",
+      "这 两个",
+      "▁mod ification",
+      "▁b acterial",
+      "▁bacter ial",
+      "▁bacteria l",
+      "▁ours elves",
+      "bat ch",
+      "b atch",
+      "种 子",
+      "ec l",
+      "e cl",
+      "▁maint aining",
+      "▁maintain ing",
+      "上 有",
+      "Er r",
+      "E rr",
+      "▁H O",
+      "▁ HO",
+      "▁S a",
+      "▁ Sa",
+      "到 期",
+      "发表 于",
+      "的一 部",
+      "的 一部",
+      "▁c rew",
+      "▁cre w",
+      "▁cr ew",
+      "▁ar ts",
+      "▁art s",
+      "▁ arts",
+      "▁li pid",
+      "▁lip id",
+      "▁ lipid",
+      "充 电",
+      "be an",
+      "b ean",
+      "▁w rap",
+      "▁wr ap",
+      "▁ wrap",
+      "▁G i",
+      "▁ Gi",
+      "\") ;",
+      "\" );",
+      "▁St ock",
+      "▁ Stock",
+      "g z",
+      "▁D utch",
+      "nCON CLUS",
+      "医 药",
+      "▁l ogo",
+      "▁log o",
+      "▁lo go",
+      "▁ logo",
+      "▁cover ing",
+      "▁cov ering",
+      "▁W inter",
+      "▁Win ter",
+      "▁Wi nter",
+      "▁join ing",
+      "▁jo ining",
+      "▁ joining",
+      "▁con straints",
+      "▁constr aints",
+      "▁constraint s",
+      "▁ constraints",
+      "int egr",
+      "ol ester",
+      "ole ster",
+      "oles ter",
+      "in cluding",
+      "主 编",
+      "▁p refix",
+      "▁pre fix",
+      "▁ prefix",
+      "四 月",
+      "f i",
+      "▁S elect",
+      "▁Se lect",
+      "▁Sel ect",
+      "▁ Select",
+      "诚 信",
+      "IM E",
+      "I ME",
+      "▁J r",
+      "Com mon",
+      "Comm on",
+      "▁polit ics",
+      "hid den",
+      "h idden",
+      "v i",
+      "vu e",
+      "v ue",
+      "语 文",
+      "周 长",
+      "▁atmosph ere",
+      "G C",
+      "EN RON",
+      "道 理",
+      "▁O s",
+      "▁ Os",
+      "▁inf ected",
+      "▁infect ed",
+      "▁ infected",
+      "反 对",
+      "西 亚",
+      "复 印",
+      "精 准",
+      "▁The ory",
+      "▁m eth",
+      "▁me th",
+      "▁met h",
+      "▁ meth",
+      "奋 斗",
+      "sl eep",
+      "s leep",
+      "常 用",
+      "压 缩",
+      "存在 的",
+      "信息 化",
+      "▁w at",
+      "▁wa t",
+      "▁ wat",
+      "中 药",
+      "水 泥",
+      "原 本",
+      "自动 化",
+      "▁im agine",
+      "▁imag ine",
+      "▁vac cine",
+      "▁vacc ine",
+      "▁vaccin e",
+      "厂 商",
+      "▁G iven",
+      "▁Give n",
+      "▁Gi ven",
+      "▁ Given",
+      "社 交",
+      "防 护",
+      "ture s",
+      "tu res",
+      "tur es",
+      "t ures",
+      "▁Sh ort",
+      "▁ Short",
+      "▁S ection",
+      "▁Se ction",
+      "▁Sec tion",
+      "▁ Section",
+      "▁comp eted",
+      "▁compet ed",
+      "▁compete d",
+      "▁A k",
+      "▁ Ak",
+      "▁equ ipped",
+      "▁equip ped",
+      "▁underw ent",
+      "▁b ind",
+      "▁bi nd",
+      "▁bin d",
+      "▁ bind",
+      "▁l ift",
+      "▁li ft",
+      "▁lif t",
+      "▁ lift",
+      "nP er",
+      "nPe r",
+      "n Per",
+      "▁R oute",
+      "▁Ro ute",
+      "▁Rou te",
+      "▁ Route",
+      "尚 未",
+      "label s",
+      "lab els",
+      "▁t error",
+      "▁te rror",
+      "▁ter ror",
+      "▁terr or",
+      "▁ terror",
+      "▁f ed",
+      "▁fe d",
+      "▁ fed",
+      "ir ty",
+      "irt y",
+      "▁H ong",
+      "▁Hon g",
+      "▁Ho ng",
+      "▁c ock",
+      "▁co ck",
+      "▁coc k",
+      "▁ cock",
+      "▁c ouncil",
+      "▁counc il",
+      "典 型",
+      "▁B road",
+      "▁Br oad",
+      "▁Bro ad",
+      "▁ Broad",
+      "的 相关",
+      "反 复",
+      "ut ing",
+      "uti ng",
+      "u ting",
+      "旋 转",
+      "▁all erg",
+      "▁alle rg",
+      "▁t utorial",
+      "▁tut orial",
+      "▁tutor ial",
+      "▁ tutorial",
+      "一 行",
+      "▁v ast",
+      "▁vas t",
+      "▁va st",
+      "▁grad uated",
+      "▁gradu ated",
+      "▁graduate d",
+      "da m",
+      "d am",
+      "java script",
+      "jav ascript",
+      "神 秘",
+      "其 行为",
+      "▁l ens",
+      "▁le ns",
+      "▁len s",
+      "键 盘",
+      "NO T",
+      "N OT",
+      "▁[ [",
+      "▁ [[",
+      "ke ep",
+      "kee p",
+      "k eep",
+      "▁K r",
+      "READ ME",
+      "if i",
+      "i fi",
+      "Jo b",
+      "J ob",
+      "复 制",
+      "og ram",
+      "ogr am",
+      "o gram",
+      "Re ference",
+      "R eference",
+      "地 质",
+      "成立 于",
+      "es ity",
+      "esi ty",
+      "▁ident ifying",
+      "▁identify ing",
+      "tw itter",
+      "t witter",
+      "▁t ro",
+      "▁tr o",
+      "▁ tro",
+      "xa d",
+      "x ad",
+      "ol id",
+      "oli d",
+      "小 游戏",
+      "▁s trip",
+      "▁st rip",
+      "▁str ip",
+      "▁stri p",
+      "▁ strip",
+      "ee k",
+      "e ek",
+      "▁members hip",
+      "▁member ship",
+      "t const",
+      "一 生",
+      "活 性",
+      "地理 位置",
+      "▁vis ited",
+      "▁visit ed",
+      "▁ visited",
+      "Ru le",
+      "R ule",
+      "一 款",
+      "pro m",
+      "pr om",
+      "p rom",
+      "景 区",
+      "患者 的",
+      "患 者的",
+      "厉 害",
+      "资 讯",
+      "主 人",
+      "▁E xt",
+      "▁Ex t",
+      "▁ Ext",
+      "▁ent reprene",
+      "▁s hel",
+      "▁sh el",
+      "▁she l",
+      "Cl ose",
+      "C lose",
+      "所 示",
+      "公 平",
+      "▁ à",
+      "▁i OS",
+      "▁ iOS",
+      "pl oad",
+      "p load",
+      "ch anged",
+      "chan ged",
+      "change d",
+      "chang ed",
+      "二 级",
+      "▁col le",
+      "▁coll e",
+      "Man y",
+      "Ma ny",
+      "M any",
+      "▁we alth",
+      "▁ wealth",
+      "ep age",
+      "e page",
+      "par ameters",
+      "param eters",
+      "parameter s",
+      "paramet ers",
+      "as set",
+      "ass et",
+      "asse t",
+      "究 竟",
+      "▁f uck",
+      "▁fu ck",
+      "▁pos sess",
+      "▁poss ess",
+      "▁econ om",
+      "▁ec onom",
+      "▁eco nom",
+      "国 有",
+      "出 庭",
+      "男 孩",
+      "AM P",
+      "A MP",
+      "▁t all",
+      "▁tal l",
+      "▁ta ll",
+      "▁ tall",
+      "局 部",
+      "▁random ized",
+      "▁rand omized",
+      "Pr ess",
+      "Pre ss",
+      "Pres s",
+      "P ress",
+      "创作 的网络小说",
+      "创作的 网络小说",
+      "Requ ired",
+      "Require d",
+      "的 两",
+      "妇 女",
+      "ud a",
+      "u da",
+      "▁f aces",
+      "▁fac es",
+      "▁face s",
+      "▁fa ces",
+      "▁ faces",
+      "▁Comp uter",
+      "▁Comput er",
+      "▁Compute r",
+      "▁ Computer",
+      "山 县",
+      "as hed",
+      "ash ed",
+      "▁rel iability",
+      "路 线",
+      "马 克",
+      "专 用",
+      "▁dro pped",
+      "▁drop ped",
+      "▁B oy",
+      "▁Bo y",
+      "▁ass istant",
+      "▁assist ant",
+      "▁ assistant",
+      "▁Pr act",
+      "▁Pra ct",
+      "th rough",
+      "thro ugh",
+      "thr ough",
+      "co st",
+      "cos t",
+      "c ost",
+      "营 造",
+      "截 至",
+      "in ator",
+      "ina tor",
+      "▁Count ry",
+      "▁Cou ntry",
+      "▁ Country",
+      "▁D own",
+      "▁Do wn",
+      "▁Dow n",
+      "▁ Down",
+      "你 说",
+      "▁calcul ation",
+      "▁calc ulation",
+      "▁V alid",
+      "▁Val id",
+      "▁ Valid",
+      "▁life style",
+      "▁lif estyle",
+      "project s",
+      "proj ects",
+      "S N",
+      "▁m olecule",
+      "▁molec ule",
+      "▁ molecule",
+      "hold ers",
+      "holder s",
+      "hol ders",
+      "信息公示 系统",
+      "nO ver",
+      "n Over",
+      "▁million s",
+      "▁mill ions",
+      "▁u rg",
+      "▁ur g",
+      "▁ urg",
+      "▁aff il",
+      "▁af fil",
+      "▁is instance",
+      "▁re ply",
+      "▁rep ly",
+      "▁ reply",
+      "▁Cont ent",
+      "▁ Content",
+      "▁L ED",
+      "▁LE D",
+      "▁ LED",
+      "ea ch",
+      "e ach",
+      "▁s hooting",
+      "▁sh ooting",
+      "▁shoot ing",
+      "对 我",
+      "US T",
+      "U ST",
+      "周 年",
+      "有利 于",
+      "有 利于",
+      "▁se ats",
+      "▁sea ts",
+      "▁seat s",
+      "飞 行",
+      "▁ 产品",
+      "▁h orse",
+      "▁hor se",
+      "▁hors e",
+      "▁ horse",
+      "nWh ich",
+      "n Which",
+      "▁Ont ario",
+      "▁g allery",
+      "▁gall ery",
+      "▁ gallery",
+      "▁s ql",
+      "▁sq l",
+      "▁ sql",
+      "的 条件",
+      "使 命",
+      "加 油",
+      "▁m uc",
+      "▁mu c",
+      "▁eng agement",
+      "▁engage ment",
+      "▁complex es",
+      "▁pre ferred",
+      "▁prefer red",
+      "▁ preferred",
+      "这 段",
+      "at ile",
+      "ati le",
+      "a tile",
+      "se q",
+      "s eq",
+      "at ernal",
+      "ater nal",
+      "atern al",
+      "a ternal",
+      "gm ent",
+      "g ment",
+      "是 要",
+      "研究 院",
+      "加 热",
+      "▁cho osing",
+      "▁Con f",
+      "▁Co nf",
+      "▁ Conf",
+      "▁Tor onto",
+      "▁cont rovers",
+      "▁contr overs",
+      "考 验",
+      "▁t ank",
+      "▁ta nk",
+      "▁tan k",
+      "沈 阳",
+      "企业信用 信息公示系统",
+      "▁w aves",
+      "▁wave s",
+      "▁wa ves",
+      "▁ waves",
+      "级 的",
+      "yi eld",
+      "y ield",
+      "▁N ic",
+      "▁Ni c",
+      "宁 夏",
+      "▁should n",
+      "家 居",
+      "绿 化",
+      "▁cy tok",
+      "▁cyt ok",
+      "▁cyto k",
+      "an te",
+      "ant e",
+      "a nte",
+      "对 该",
+      "▁S elf",
+      "▁Se lf",
+      "▁Sel f",
+      "▁ Self",
+      "B E",
+      "▁par ad",
+      "▁para d",
+      "▁pa rad",
+      "▁g olf",
+      "▁go lf",
+      "▁gol f",
+      "neg ative",
+      "欠 条",
+      "▁emer ging",
+      "▁emerg ing",
+      "▁adopt ed",
+      "▁adop ted",
+      "nD ef",
+      "nDe f",
+      "n Def",
+      "▁b ike",
+      "▁bi ke",
+      "▁bik e",
+      "▁ bike",
+      "转 让",
+      "▁A rg",
+      "▁Ar g",
+      "▁ Arg",
+      "▁fair ly",
+      "的 选择",
+      "▁non linear",
+      "ap ply",
+      "app ly",
+      "▁平 装",
+      "▁ 平装",
+      ", $",
+      "▁f itness",
+      "▁fit ness",
+      "教 练",
+      "取得 了",
+      "取 得了",
+      "eli hood",
+      "▁You Tube",
+      "商 家",
+      "给 大家",
+      "你 就",
+      "▁_ ___",
+      "▁__ __",
+      "▁ ____",
+      "▁p red",
+      "▁pre d",
+      "▁pr ed",
+      "▁ pred",
+      "聊 天",
+      "▁R s",
+      "▁ Rs",
+      "▁add ressed",
+      "▁address ed",
+      "▁addr essed",
+      "▁prec ise",
+      "▁ 和",
+      "传 承",
+      "▁r ounded",
+      "▁round ed",
+      "▁ rounded",
+      "请 仔细",
+      "▁sustain able",
+      "OL D",
+      "O LD",
+      "ib ration",
+      "ibr ation",
+      "文 物",
+      "lo om",
+      "loo m",
+      "l oom",
+      "也 可",
+      "ra it",
+      "r ait",
+      "强 大",
+      "im mun",
+      "imm un",
+      "▁w itness",
+      "▁wit ness",
+      "op l",
+      "o pl",
+      "De tails",
+      "Det ails",
+      "Detail s",
+      "▁Col lection",
+      "▁Coll ection",
+      "▁Collect ion",
+      "▁ Collection",
+      "▁H as",
+      "▁Ha s",
+      "▁ Has",
+      "Test Case",
+      "ki nd",
+      "kin d",
+      "k ind",
+      "▁ch icken",
+      "▁chick en",
+      "▁chic ken",
+      "Par ameters",
+      "Parameter s",
+      "Param eters",
+      "▁s ulf",
+      "▁su lf",
+      "▁sul f",
+      "▁ sulf",
+      "▁f orum",
+      "▁for um",
+      "▁fo rum",
+      "▁ forum",
+      "▁dr inking",
+      "▁drink ing",
+      "▁b read",
+      "▁br ead",
+      "▁bre ad",
+      "▁ bread",
+      "▁min ister",
+      "▁mini ster",
+      "▁hor izontal",
+      "▁horizon tal",
+      "▁horiz ontal",
+      "▁ horizontal",
+      "▁] \\",
+      "▁ ]\\",
+      "ca pt",
+      "cap t",
+      "c apt",
+      "业 绩",
+      "开 了",
+      "独 任",
+      "▁gu ard",
+      "▁ guard",
+      "的 变化",
+      "▁Y OU",
+      "▁YO U",
+      "▁\" \\",
+      "▁ \"\\",
+      "▁l osing",
+      "▁lo sing",
+      "▁los ing",
+      "百 万",
+      "邮 政",
+      "▁b ear",
+      "▁be ar",
+      "▁ bear",
+      "▁del icious",
+      "▁pro f",
+      "▁pr of",
+      "▁ prof",
+      "▁qu iet",
+      "▁qui et",
+      "▁ quiet",
+      "显 卡",
+      "厂 家",
+      "A g",
+      "m k",
+      "▁f etch",
+      "▁fet ch",
+      "▁ fetch",
+      "▁r anging",
+      "▁ran ging",
+      "▁ ranging",
+      "之 下",
+      "▁l os",
+      "▁lo s",
+      "▁ los",
+      "招 生",
+      "大家 都",
+      "Con s",
+      "Co ns",
+      "C ons",
+      "领域 的",
+      "并 将",
+      "Ce ll",
+      "C ell",
+      "pro xy",
+      "pr oxy",
+      "prox y",
+      "oo se",
+      "oos e",
+      "o ose",
+      "Ver t",
+      "V ert",
+      "en da",
+      "end a",
+      "e nda",
+      "ex pression",
+      "exp ression",
+      "express ion",
+      "expr ession",
+      "▁f estival",
+      "▁fest ival",
+      "法 师",
+      "节 奏",
+      "ci ty",
+      "cit y",
+      "c ity",
+      "ld ots",
+      "l dots",
+      "硬 盘",
+      "强大 的",
+      "强 大的",
+      "en eration",
+      "ener ation",
+      "ene ration",
+      "Pro t",
+      "Pr ot",
+      "P rot",
+      "▁apprec iated",
+      "▁appreciate d",
+      "▁c able",
+      "▁ca ble",
+      "▁cab le",
+      "▁ 代码",
+      "▁title s",
+      "▁tit les",
+      "▁ titles",
+      "▁que ue",
+      "▁ queue",
+      "▁ μ",
+      "nW ell",
+      "nWe ll",
+      "n Well",
+      "no rm",
+      "nor m",
+      "n orm",
+      "▁f irms",
+      "▁firm s",
+      "▁fir ms",
+      "国家 企业信用信息公示系统",
+      "▁cr ash",
+      "▁cra sh",
+      "nL e",
+      "n Le",
+      "Ge n",
+      "G en",
+      "▁ad verse",
+      "▁advers e",
+      "\": [",
+      "\" :[",
+      "▁m arks",
+      "▁mark s",
+      "▁mar ks",
+      "▁ marks",
+      "▁M aterial",
+      "▁Mat erial",
+      "▁Mate rial",
+      "▁ Material",
+      "地 上",
+      "▁am pl",
+      "▁amp l",
+      "▁return ing",
+      "pro vider",
+      "prov ider",
+      "provide r",
+      "社会 的",
+      "社 会的",
+      "bin g",
+      "bi ng",
+      "b ing",
+      "ilib rium",
+      "riz ona",
+      "fa il",
+      "f ail",
+      "y z",
+      "▁Post ed",
+      "▁Po sted",
+      "▁Pos ted",
+      "▁ Posted",
+      "▁h int",
+      "▁hi nt",
+      "▁ hint",
+      "正 本",
+      "MI T",
+      "M IT",
+      "▁\" %",
+      "▁ \"%",
+      "Sy nc",
+      "Syn c",
+      "S ync",
+      "▁gr ass",
+      "▁gra ss",
+      "▁ grass",
+      "本裁定送达后 立即生效",
+      "▁ident ical",
+      "ab ling",
+      "abl ing",
+      "a bling",
+      "s n",
+      "建筑 面积",
+      "ex ception",
+      "except ion",
+      "起诉 书",
+      "ga ge",
+      "g age",
+      "▁L ead",
+      "▁Le ad",
+      "▁ Lead",
+      "an ti",
+      "ant i",
+      "▁separ ated",
+      "▁separate d",
+      "▁ separated",
+      "▁method ology",
+      "▁图 书",
+      "▁ 图书",
+      "lo v",
+      "l ov",
+      "▁acqu isition",
+      "▁acquis ition",
+      "▁Min nesota",
+      "▁ext reme",
+      "▁extrem e",
+      "▁extr eme",
+      "▁consider ation",
+      "▁Ch apter",
+      "▁Chap ter",
+      "▁ Chapter",
+      "发 送",
+      "▁when ever",
+      "▁whe never",
+      "De p",
+      "D ep",
+      "▁phys iological",
+      "▁physi ological",
+      "▁physiologic al",
+      "毕业 于",
+      "▁every day",
+      "cal led",
+      "call ed",
+      "c alled",
+      "▁gen ome",
+      "▁ge nome",
+      "▁ genome",
+      "ar en",
+      "are n",
+      "a ren",
+      "▁nurs ing",
+      "▁nur sing",
+      "▁O THER",
+      "▁OT HER",
+      "▁ OTHER",
+      "▁pe ptide",
+      "▁pept ide",
+      "▁ peptide",
+      "自 觉",
+      "▁L ew",
+      "▁Le w",
+      "此 题",
+      "ri n",
+      "r in",
+      "Net work",
+      "N etwork",
+      "默 认",
+      "▁M ountain",
+      "▁Mount ain",
+      "叫 做",
+      "▁pro pag",
+      "▁prop ag",
+      "▁ propag",
+      "消 除",
+      "未 成年",
+      "▁ne ighb",
+      "▁neigh b",
+      "▁Victor ia",
+      "地 面",
+      "建 成",
+      "▁co ordinate",
+      "▁coord inate",
+      "▁ coordinate",
+      "á n",
+      "▁sent ence",
+      "▁ sentence",
+      "F P",
+      "▁inter views",
+      "▁interview s",
+      "▁a cet",
+      "▁ac et",
+      "▁ace t",
+      "▁ acet",
+      "▁c lot",
+      "▁cl ot",
+      "equ ality",
+      "eq uality",
+      "equal ity",
+      "e quality",
+      "hr ef",
+      "h ref",
+      "mod al",
+      "mo dal",
+      "m odal",
+      "▁c yl",
+      "▁cy l",
+      "\" ;",
+      "▁ »",
+      "的 外",
+      "▁some what",
+      "▁somew hat",
+      "▁t ang",
+      "▁ta ng",
+      "▁tan g",
+      "▁ tang",
+      "▁real ized",
+      "▁realize d",
+      "▁simult aneously",
+      "▁simultaneous ly",
+      "▁w ound",
+      "▁wo und",
+      "奥 运",
+      "▁d isk",
+      "▁dis k",
+      "▁di sk",
+      "▁ disk",
+      "过 滤",
+      "▁Min istry",
+      "▁Mini stry",
+      "▁Port ug",
+      "pl er",
+      "ple r",
+      "p ler",
+      "L R",
+      "en nes",
+      "enn es",
+      "enne s",
+      "发 放",
+      "难 道",
+      "ig ure",
+      "igu re",
+      "Name s",
+      "Na mes",
+      "N ames",
+      "▁ap pl",
+      "▁app l",
+      "▁cur ves",
+      "▁curve s",
+      "▁curv es",
+      "en um",
+      "enu m",
+      "e num",
+      "编 制",
+      "own town",
+      "ownt own",
+      "T T",
+      "出 租",
+      "ot ted",
+      "ott ed",
+      "证 人",
+      "ul us",
+      "ulu s",
+      "▁C lean",
+      "▁Cl ean",
+      "▁Cle an",
+      "▁ Clean",
+      "▁pro ps",
+      "▁pr ops",
+      "▁prop s",
+      "▁ props",
+      "▁f ails",
+      "▁fail s",
+      "▁fa ils",
+      "▁mo ist",
+      "▁moi st",
+      "▁con verted",
+      "▁conver ted",
+      "▁convert ed",
+      "nS ports",
+      "nSp orts",
+      "nSport s",
+      "n Sports",
+      "▁cl osing",
+      "▁clos ing",
+      "▁ closing",
+      "▁民 族",
+      "▁ 民族",
+      "ag ic",
+      "agi c",
+      "a gic",
+      "从 选项中",
+      "定 制",
+      "nC ons",
+      "nCon s",
+      "nCo ns",
+      "n Cons",
+      "▁B and",
+      "▁Ba nd",
+      "▁Ban d",
+      "▁ Band",
+      "▁n am",
+      "▁na m",
+      "▁ nam",
+      "▁L at",
+      "▁La t",
+      "▁ Lat",
+      "▁A B",
+      "▁ AB",
+      "A E",
+      "▁E gypt",
+      "▁S ix",
+      "▁Si x",
+      "▁ 是",
+      "be st",
+      "bes t",
+      "b est",
+      "冲 突",
+      "郑 州",
+      "r angle",
+      "an z",
+      "a nz",
+      "协 商",
+      "▁in puts",
+      "▁input s",
+      "▁ inputs",
+      "▁b its",
+      "▁bi ts",
+      "▁bit s",
+      "▁ bits",
+      "▁particip ated",
+      "▁participate d",
+      "▁dist urb",
+      "传 媒",
+      "ou ri",
+      "our i",
+      "o uri",
+      "▁f ocusing",
+      "▁foc using",
+      "▁focus ing",
+      "高 温",
+      "▁m arg",
+      "▁mar g",
+      "▁ma rg",
+      "竞 赛",
+      "▁turn ing",
+      "▁tur ning",
+      "ou ses",
+      "ous es",
+      "ouse s",
+      "o uses",
+      "▁b at",
+      "▁ba t",
+      "▁ bat",
+      "▁cult ures",
+      "▁culture s",
+      "▁cul tures",
+      "的 价格",
+      "or dan",
+      "ord an",
+      "祖 国",
+      "▁P os",
+      "▁Po s",
+      "▁ Pos",
+      "▁redu ces",
+      "▁red uces",
+      "▁reduce s",
+      "▁pe nal",
+      "▁pen al",
+      "▁Cl ient",
+      "▁ Client",
+      "▁behav iors",
+      "▁behavior s",
+      "▁behavi ors",
+      "截 止",
+      "超 越",
+      "找 出",
+      "np lt",
+      "npl t",
+      "n plt",
+      "▁transport ation",
+      ",\\ '",
+      ", \\'",
+      "fol der",
+      "fold er",
+      "f older",
+      "no des",
+      "node s",
+      "nod es",
+      "n odes",
+      "Ra d",
+      "R ad",
+      "▁inv ited",
+      "▁invite d",
+      "od a",
+      "o da",
+      "一 方",
+      "武汉 市",
+      "er mission",
+      "erm ission",
+      "浪 费",
+      "这 部",
+      "▁o uter",
+      "▁out er",
+      "▁ou ter",
+      "▁ outer",
+      "▁mat rices",
+      "ID S",
+      "I DS",
+      "▁M obile",
+      "▁Mob ile",
+      "▁ Mobile",
+      "明 天",
+      "en ron",
+      "▁man ifest",
+      "▁manif est",
+      "▁ manifest",
+      "▁th ickness",
+      "▁thick ness",
+      "▁S pecific",
+      "▁Spec ific",
+      "▁ Specific",
+      "ant ly",
+      "bour ne",
+      "▁h ook",
+      "▁ho ok",
+      "▁ hook",
+      "▁direct ions",
+      "▁direction s",
+      "▁dir ections",
+      "▁enzym es",
+      "▁enzyme s",
+      "的数 据",
+      "的 数据",
+      "信 任",
+      "▁off ices",
+      "▁offic es",
+      "▁office s",
+      "▁P rice",
+      "▁Pr ice",
+      "▁Pri ce",
+      "▁ Price",
+      "▁box es",
+      "▁ boxes",
+      "电 气",
+      "rop ical",
+      "ropic al",
+      "成都 市",
+      "成 都市",
+      "▁to mor",
+      "▁tom or",
+      "剩 余",
+      "s i",
+      "▁n either",
+      "▁ne ither",
+      "▁develop er",
+      "▁ developer",
+      "黄 色",
+      "ma de",
+      "mad e",
+      "m ade",
+      "ü r",
+      "消 耗",
+      "N a",
+      "反 馈",
+      "ac et",
+      "ace t",
+      "a cet",
+      "▁M D",
+      "▁ MD",
+      "▁De velop",
+      "▁ Develop",
+      "if ndef",
+      "顾 问",
+      "nRE SULTS",
+      "▁some where",
+      "▁somew here",
+      "发动 机",
+      "发 动机",
+      "不得 不",
+      "不 得不",
+      "▁Phil ipp",
+      "▁Philip p",
+      "▁arr ested",
+      "▁arrest ed",
+      "т о",
+      "第一 章",
+      "第 一章",
+      "▁ 一",
+      "青 少年",
+      "ph al",
+      "pha l",
+      "p hal",
+      "qu es",
+      "que s",
+      "q ues",
+      "▁pay ments",
+      "▁payment s",
+      "更 换",
+      "▁N eed",
+      "▁Ne ed",
+      "▁ Need",
+      "▁v ibr",
+      "▁vi br",
+      "▁vib r",
+      "▁f er",
+      "▁fe r",
+      "▁ fer",
+      "do n",
+      "d on",
+      "▁m eters",
+      "▁me ters",
+      "▁met ers",
+      "▁meter s",
+      "的 服务",
+      "▁law yer",
+      "▁lawy er",
+      "▁vari ations",
+      "▁var iations",
+      "▁variation s",
+      "ue r",
+      "u er",
+      "就 行",
+      "nO ut",
+      "n Out",
+      "▁contin uing",
+      "▁continu ing",
+      "大 厦",
+      "明 明",
+      "▁vari ant",
+      "▁var iant",
+      "▁ variant",
+      "ar on",
+      "aro n",
+      "a ron",
+      "▁s ocket",
+      "▁soc ket",
+      "▁sock et",
+      "▁ socket",
+      "权 的",
+      "nr e",
+      "n re",
+      "起 到",
+      "肌 肉",
+      "over flow",
+      "▁cor rection",
+      "▁correct ion",
+      "sk y",
+      "s ky",
+      "招 聘",
+      "妻 子",
+      "▁g all",
+      "▁gal l",
+      "▁ga ll",
+      "▁d il",
+      "▁di l",
+      "▁attract ive",
+      "▁attr active",
+      "▁c ub",
+      "▁cu b",
+      "▁ cub",
+      "甘肃 省",
+      "us hing",
+      "ush ing",
+      "ushi ng",
+      "b g",
+      "多 样",
+      "On ly",
+      "ect ors",
+      "ector s",
+      "e ctors",
+      "ag onal",
+      "ago nal",
+      "agon al",
+      "W orld",
+      "技术 的",
+      "▁exhib it",
+      "▁host ing",
+      "▁cons isting",
+      "▁consist ing",
+      "▁ign ore",
+      "▁ignor e",
+      "▁ ignore",
+      "▁dec line",
+      "▁decl ine",
+      "部 件",
+      "pos itive",
+      "▁ 徐",
+      "你 在",
+      "▁ad hes",
+      "▁Pr inc",
+      "▁Pri nc",
+      "HE RE",
+      "HER E",
+      "H ERE",
+      "▁t urb",
+      "▁tur b",
+      "▁tu rb",
+      "la w",
+      "l aw",
+      "ne um",
+      "▁tomor row",
+      "▁tom orrow",
+      "▁b ab",
+      "▁ba b",
+      "▁ bab",
+      "经济 发展",
+      "▁Sub ject",
+      "▁Su bject",
+      "▁ Subject",
+      "公众 号",
+      "nU se",
+      "nUs e",
+      "n Use",
+      "so me",
+      "s ome",
+      "▁sub sequently",
+      "▁subsequ ently",
+      "▁subsequent ly",
+      "▁D B",
+      "▁ DB",
+      "nD on",
+      "nDo n",
+      "n Don",
+      "农村商业 银行股份有限公司",
+      "不 开",
+      "干 燥",
+      "▁b old",
+      "▁bo ld",
+      "▁bol d",
+      "▁ bold",
+      "发 言",
+      "sh are",
+      "sha re",
+      "s hare",
+      "_ ,",
+      "▁base ball",
+      "ill ary",
+      "illa ry",
+      "illar y",
+      "uc l",
+      "u cl",
+      "▁s ear",
+      "▁se ar",
+      "▁sea r",
+      "▁ sear",
+      "ob by",
+      "Do wn",
+      "D own",
+      "歌 手",
+      "引 进",
+      "合法 权益",
+      "▁g raft",
+      "▁gra ft",
+      "▁key board",
+      "▁ keyboard",
+      "I H",
+      "▁in take",
+      "▁int ake",
+      "▁t ensor",
+      "▁tens or",
+      "▁ tensor",
+      "Gener ator",
+      "Gen erator",
+      "彼 此",
+      "ata log",
+      "atal og",
+      "合 成",
+      "这 类",
+      "▁re ar",
+      "▁r ear",
+      "▁strateg ic",
+      "差 距",
+      "en gers",
+      "eng ers",
+      "enge rs",
+      "enger s",
+      "nT wo",
+      "nTw o",
+      "n Two",
+      "▁in fer",
+      "▁inf er",
+      "▁ infer",
+      "▁b yte",
+      "▁by te",
+      "▁ byte",
+      "AR D",
+      "A RD",
+      "lin ks",
+      "link s",
+      "l inks",
+      "or tic",
+      "ort ic",
+      "▁R aj",
+      "▁Ra j",
+      "ap l",
+      "a pl",
+      "HT TP",
+      "▁Vis ual",
+      "▁ Visual",
+      "发 达",
+      "假 设",
+      "城 乡",
+      "▁sub stance",
+      "▁subst ance",
+      "▁S ex",
+      "▁Se x",
+      "▁ Sex",
+      "东 莞",
+      "通过 本院",
+      "ed itor",
+      "edit or",
+      "edi tor",
+      "m n",
+      "角 的",
+      "ot or",
+      "oto r",
+      "o tor",
+      "▁ins piration",
+      "▁insp iration",
+      "▁inspir ation",
+      "▁P ur",
+      "▁Pu r",
+      "▁ Pur",
+      "▁calcul ations",
+      "▁calc ulations",
+      "▁calculation s",
+      "欣 赏",
+      "气 温",
+      "▁food s",
+      "▁fo ods",
+      "▁foo ds",
+      "▁apopt osis",
+      "扮 演",
+      "im it",
+      "imi t",
+      "i mit",
+      "ic er",
+      "ice r",
+      "i cer",
+      "▁f inds",
+      "▁find s",
+      "▁fin ds",
+      "县 人民检察院",
+      "foot er",
+      "foo ter",
+      "fo oter",
+      "f ooter",
+      "▁C reated",
+      "▁Create d",
+      "▁Creat ed",
+      "▁Cre ated",
+      "▁ Created",
+      "▁A venue",
+      "▁Av enue",
+      "au g",
+      "a ug",
+      "E V",
+      "$ )",
+      "▁reg ardless",
+      "▁regard less",
+      "▁ −",
+      "Wrap per",
+      "Wr apper",
+      "W rapper",
+      "li ving",
+      "l iving",
+      "im ental",
+      "iment al",
+      "imen tal",
+      "i mental",
+      "没 什么",
+      "equ iv",
+      "▁s ynthetic",
+      "▁sy nthetic",
+      "▁synt hetic",
+      "▁synth etic",
+      "▁synthet ic",
+      "▁C ivil",
+      "▁Ci vil",
+      "▁S ydney",
+      "我 院",
+      "S k",
+      "▁Cl inical",
+      "▁Clinic al",
+      "▁Clin ical",
+      "▁ Clinical",
+      "▁sever ity",
+      "▁ severity",
+      "▁b eer",
+      "▁be er",
+      "▁bee r",
+      "bin ding",
+      "bind ing",
+      "b inding",
+      "as tro",
+      "ast ro",
+      "astr o",
+      "▁admin istered",
+      "▁administer ed",
+      "吃 的",
+      "实现 了",
+      "as hes",
+      "ash es",
+      "▁he aling",
+      "▁heal ing",
+      "ov iet",
+      "ovie t",
+      "ovi et",
+      "ou tube",
+      "out ube",
+      "给 他",
+      "打 算",
+      "na ire",
+      "nai re",
+      "n aire",
+      "▁D aily",
+      "▁Da ily",
+      "▁ Daily",
+      "▁nerv ous",
+      "耕 地",
+      "▁g lad",
+      "▁gl ad",
+      "职业 技术",
+      "魅 力",
+      "un ique",
+      "uni que",
+      "it led",
+      "itle d",
+      "ig m",
+      "i gm",
+      "▁For est",
+      "▁Fore st",
+      "▁Fo rest",
+      "本院 经审查",
+      "ps y",
+      "p sy",
+      "▁el ucid",
+      "▁moment um",
+      "▁depend encies",
+      "▁ dependencies",
+      "▁L es",
+      "▁Le s",
+      "▁ Les",
+      "条 例",
+      "慢 性",
+      "le an",
+      "lea n",
+      "l ean",
+      "nAl though",
+      "n Although",
+      "▁wond ering",
+      "▁wonder ing",
+      "常 务",
+      "ga l",
+      "g al",
+      "ur ls",
+      "url s",
+      "be ing",
+      "bo w",
+      "b ow",
+      "▁gener ator",
+      "▁gen erator",
+      "▁genera tor",
+      "▁ generator",
+      "▁int elligence",
+      "▁intellig ence",
+      "是 多少",
+      "▁re call",
+      "▁rec all",
+      "ETHO DS",
+      "ETHOD S",
+      "ta u",
+      "t au",
+      "▁c ord",
+      "▁co rd",
+      "▁cor d",
+      "▁ cord",
+      "comp ile",
+      "国土 资源",
+      "国土资 源",
+      "我 和",
+      "但 在",
+      "od d",
+      "o dd",
+      "热 度",
+      "se ason",
+      "sea son",
+      "nO nce",
+      "nOn ce",
+      "n Once",
+      "▁Secret ary",
+      "毫 无",
+      "两 组",
+      "ic ide",
+      "ici de",
+      "各 级",
+      "选择 的",
+      "▁my ocard",
+      "▁subst rate",
+      "▁substr ate",
+      "对 话",
+      "Right arrow",
+      "om eter",
+      "ome ter",
+      "omet er",
+      "o meter",
+      "▁/ *",
+      "▁ /*",
+      "的 历史",
+      "价 比",
+      "▁s oci",
+      "▁so ci",
+      "▁soc i",
+      "▁com patible",
+      "▁compat ible",
+      "▁ compatible",
+      "▁f inger",
+      "▁fin ger",
+      "▁fing er",
+      "▁ finger",
+      "ia ry",
+      "iar y",
+      "i ary",
+      "罪 名",
+      "▁w el",
+      "▁we l",
+      "ear ing",
+      "ea ring",
+      "e aring",
+      "▁benef icial",
+      "▁benefici al",
+      "▁B ridge",
+      "▁Br idge",
+      "▁Brid ge",
+      "▁ Bridge",
+      "专 注",
+      "暂 无",
+      "之日 止",
+      "之 日止",
+      "▁hyper tension",
+      "▁hypert ension",
+      "▁e ll",
+      "▁el l",
+      "▁ ell",
+      "% (",
+      "▁Gu ide",
+      "▁Guid e",
+      "▁ Guide",
+      "▁pro motion",
+      "▁prom otion",
+      "▁promot ion",
+      "▁promo tion",
+      "▁ar ms",
+      "▁arm s",
+      "▁ arms",
+      "nO ff",
+      "nOf f",
+      "n Off",
+      "je t",
+      "j et",
+      "要 么",
+      "批 发",
+      "▁l uck",
+      "▁luc k",
+      "▁lu ck",
+      "较 高",
+      "ar se",
+      "ars e",
+      "▁ex pat",
+      "▁exp at",
+      "(\" \\\\",
+      "( \"\\\\",
+      "▁Be aut",
+      "▁Bea ut",
+      "▁Beau t",
+      "农村信用 合作",
+      "bi al",
+      "bia l",
+      "b ial",
+      "▁pass es",
+      "▁pas ses",
+      "pre sent",
+      "pres ent",
+      "p resent",
+      "▁M ajor",
+      "▁Maj or",
+      "▁ Major",
+      "糖 尿",
+      "一 线",
+      "朝 鲜",
+      "▁com paring",
+      "▁comp aring",
+      "▁compar ing",
+      "▁D am",
+      "▁Da m",
+      "▁ Dam",
+      "▁d op",
+      "▁do p",
+      "▁gener ic",
+      "▁gen eric",
+      "▁gene ric",
+      "▁ generic",
+      "老 婆",
+      "▁adj acent",
+      "mark s",
+      "mar ks",
+      "m arks",
+      "▁op ens",
+      "▁open s",
+      "▁ opens",
+      "pect ed",
+      "pec ted",
+      "p ected",
+      "▁phosph ory",
+      "▁in credible",
+      "▁incred ible",
+      "▁wor st",
+      "▁wors t",
+      "▁wo rst",
+      "了 很多",
+      "c ursor",
+      "注 射",
+      "一 开始",
+      "▁Ar gent",
+      "▁Arg ent",
+      "极 速",
+      "▁text ure",
+      "▁tex ture",
+      "▁ texture",
+      "全 新",
+      "wrap per",
+      "wr apper",
+      "w rapper",
+      "宗 旨",
+      "子 里",
+      "亿 美元",
+      "or ts",
+      "ort s",
+      "的 使用",
+      "▁S imilar",
+      "▁Sim ilar",
+      "▁ Similar",
+      "Ma c",
+      "M ac",
+      "▁I mp",
+      "▁Im p",
+      "▁ Imp",
+      "The y",
+      "Th ey",
+      "T hey",
+      "师范 大学",
+      "圆 形",
+      "不 大",
+      "sh ips",
+      "ship s",
+      "shi ps",
+      "s hips",
+      "nM ost",
+      "n Most",
+      "im et",
+      "ime t",
+      "i met",
+      "▁Ox ford",
+      "ri tis",
+      "rit is",
+      "r itis",
+      "▁Out put",
+      "▁ Output",
+      "▁ ©",
+      "▁T her",
+      "▁The r",
+      "▁Th er",
+      "的 取值范围",
+      "股 权",
+      "▁t iny",
+      "▁ti ny",
+      "▁tin y",
+      "▁ tiny",
+      "▁ass umed",
+      "▁assum ed",
+      "▁assume d",
+      "to Be",
+      "▁Prem ier",
+      "▁H ence",
+      "▁He nce",
+      "▁Hen ce",
+      "▁gener ating",
+      "▁gene rating",
+      "▁genera ting",
+      "进 展",
+      "▁sim ulations",
+      "▁simulation s",
+      "号民事 判决",
+      "常见 的",
+      "常 见的",
+      "▁m utation",
+      "▁mut ation",
+      "▁ mutation",
+      "▁commun ications",
+      "▁communication s",
+      "▁communic ations",
+      "▁ communications",
+      "▁t ill",
+      "▁til l",
+      "▁ti ll",
+      "sim ple",
+      "s imple",
+      "▁sh ower",
+      "▁show er",
+      "▁acc eler",
+      "▁accel er",
+      "边 形",
+      "▁att orney",
+      "▁pro spective",
+      "▁prosp ective",
+      "▁pros pective",
+      "▁prospect ive",
+      "心 中",
+      "电 流",
+      "精 度",
+      "\\' ),",
+      "\\') ,",
+      "\\ '),",
+      "蛋白 质",
+      "▁a sc",
+      "▁as c",
+      "▁ asc",
+      "te in",
+      "t ein",
+      "PU T",
+      "P UT",
+      "▁sp okes",
+      "▁spoke s",
+      "▁spo kes",
+      "基础 上",
+      "▁cor respond",
+      "▁correspon d",
+      "▁mac ro",
+      "▁ma cro",
+      "▁ macro",
+      "▁Git Hub",
+      "▁ GitHub",
+      "un signed",
+      "uns igned",
+      "前 期",
+      "pa l",
+      "p al",
+      "▁sc ulpt",
+      "Su ccess",
+      "S uccess",
+      "实 体",
+      "d m",
+      "re ader",
+      "read er",
+      "rea der",
+      "r eader",
+      "▁c ancel",
+      "▁can cel",
+      "▁ cancel",
+      "de mo",
+      "dem o",
+      "d emo",
+      "后 续",
+      "▁appro val",
+      "▁ approval",
+      "我们 可以",
+      "en able",
+      "ena ble",
+      "e nable",
+      "▁car rying",
+      "▁carry ing",
+      "ot ton",
+      "ott on",
+      "otto n",
+      "的人 生",
+      "的 人生",
+      "▁c um",
+      "▁cu m",
+      "▁ cum",
+      "re dit",
+      "red it",
+      "r edit",
+      "诉 至",
+      "▁k it",
+      "▁ki t",
+      "▁ kit",
+      "ov erset",
+      "over set",
+      "overs et",
+      "▁Th ose",
+      "▁v irt",
+      "▁vir t",
+      "▁vi rt",
+      "▁ virt",
+      "应当 依照",
+      "采 访",
+      "Pl ay",
+      "P lay",
+      "ok er",
+      "oke r",
+      "o ker",
+      "积 分",
+      "大 幅",
+      "▁comput ational",
+      "▁computation al",
+      "纺 织",
+      "而 在",
+      "位 的",
+      "▁He art",
+      "▁Hear t",
+      "▁ Heart",
+      "门 口",
+      "分 支",
+      "▁t oxic",
+      "▁tox ic",
+      "▁ toxic",
+      "ah a",
+      "a ha",
+      "▁emph as",
+      "▁emp has",
+      "发现被执行人有 可供执行财产",
+      "de tail",
+      "det ail",
+      "▁alter ed",
+      "▁alt ered",
+      "” ）",
+      "ve x",
+      "v ex",
+      "▁K eep",
+      "▁Ke ep",
+      "▁ Keep",
+      "he ng",
+      "hen g",
+      "h eng",
+      "▁Ber lin",
+      "二 是",
+      "▁u l",
+      "▁ ul",
+      "某 个",
+      "ER T",
+      "E RT",
+      "▁t alks",
+      "▁talk s",
+      "▁tal ks",
+      "▁danger ous",
+      "▁rep roduce",
+      "▁reprodu ce",
+      "▁reproduc e",
+      "▁cou nsel",
+      "▁couns el",
+      "会 长",
+      "▁b low",
+      "▁bl ow",
+      "▁blo w",
+      "▁pro minent",
+      "▁prom inent",
+      "▁promin ent",
+      "收 录",
+      "▁c rystal",
+      "▁cryst al",
+      "▁circumst ances",
+      "▁Tur key",
+      "op her",
+      "oph er",
+      "o pher",
+      "ic king",
+      "ick ing",
+      "icki ng",
+      "▁T otal",
+      "▁To tal",
+      "▁Tot al",
+      "▁ Total",
+      "▁perform ances",
+      "▁performance s",
+      "▁c ooking",
+      "▁co oking",
+      "▁cook ing",
+      "▁* /",
+      "▁ */",
+      "▁U SB",
+      "▁US B",
+      "▁ USB",
+      "▁D ouble",
+      "▁Do uble",
+      "▁Dou ble",
+      "▁ Double",
+      "长 江",
+      "省 级",
+      "构 造",
+      "出 血",
+      "▁A I",
+      "▁ AI",
+      "te ch",
+      "tec h",
+      "t ech",
+      "的 位置",
+      "▁Off icial",
+      "nA nother",
+      "nAn other",
+      "n Another",
+      "bu f",
+      "b uf",
+      "▁re nd",
+      "▁r end",
+      "▁ rend",
+      "▁s oph",
+      "▁so ph",
+      "▁sop h",
+      "精 品",
+      "的最 大",
+      "的 最大",
+      "nM any",
+      "nMan y",
+      "nMa ny",
+      "n Many",
+      "▁d ish",
+      "▁dis h",
+      "▁di sh",
+      "▁un expected",
+      "▁inst ances",
+      "▁instance s",
+      "▁ instances",
+      "▁system ic",
+      "▁syst emic",
+      "▁micro scopy",
+      "▁micros copy",
+      "支 出",
+      "▁win ners",
+      "▁winner s",
+      "如何 评价",
+      "▁h ers",
+      "▁he rs",
+      "▁her s",
+      "▁d ot",
+      "▁do t",
+      "▁ dot",
+      "▁ap artment",
+      "▁apart ment",
+      "▁con cert",
+      "▁concer t",
+      "▁conc ert",
+      "▁conce rt",
+      "四 日",
+      "▁S ite",
+      "▁Si te",
+      "▁Sit e",
+      "▁ Site",
+      "▁deb ate",
+      "缺 席",
+      "▁P layer",
+      "▁Pl ayer",
+      "▁Play er",
+      "▁ Player",
+      "or ient",
+      "ori ent",
+      "orie nt",
+      "o rient",
+      "▁N i",
+      "▁ Ni",
+      "气 的",
+      "co urse",
+      "c ourse",
+      "▁integr ate",
+      "股 东",
+      "ma king",
+      "mak ing",
+      "m aking",
+      "▁M ERCHANTABILITY",
+      "▁ MERCHANTABILITY",
+      "一 人",
+      "▁s olo",
+      "▁so lo",
+      "▁sol o",
+      "▁ solo",
+      "▁t f",
+      "▁ tf",
+      "or ne",
+      "orn e",
+      "o rne",
+      "威 胁",
+      "▁Ind ones",
+      "林 业",
+      "会 的",
+      "卫 星",
+      "▁b ands",
+      "▁band s",
+      "▁ban ds",
+      "▁ bands",
+      "S ocket",
+      "▁Pub lish",
+      "▁ Publish",
+      "▁just ice",
+      "an gel",
+      "ang el",
+      "ange l",
+      "▁sc enario",
+      "▁scen ario",
+      "▁ scenario",
+      "▁def ects",
+      "▁defe cts",
+      "▁defect s",
+      "这 时",
+      "并 发",
+      "▁h ire",
+      "▁hi re",
+      "▁ hire",
+      "▁manufact urer",
+      "▁manufacture r",
+      "▁h its",
+      "▁hit s",
+      "▁hi ts",
+      "▁ hits",
+      "ex per",
+      "exp er",
+      "nt yped",
+      "ntype d",
+      "nty ped",
+      "n typed",
+      "传票 传唤",
+      "▁im pressive",
+      "▁imp ressive",
+      "▁impress ive",
+      "温 馨",
+      "▁inf ants",
+      "▁infant s",
+      "灵 魂",
+      "简 直",
+      "▁app arently",
+      "▁apparent ly",
+      "▁appar ently",
+      "不 喜欢",
+      "们 的",
+      "▁e scape",
+      "▁esc ape",
+      "▁escap e",
+      "▁ escape",
+      "总 裁",
+      "日向本院 提起",
+      "日 向本院提起",
+      "指定的期间 履行给付金钱义务",
+      "Q t",
+      "Mat rix",
+      "▁dis appe",
+      "丛 书",
+      "▁const antly",
+      "▁constant ly",
+      "▁B Y",
+      "▁ BY",
+      "暂 时",
+      "第二百五十 三条之规定",
+      "第二百五十三 条之规定",
+      "▁volunte ers",
+      "▁volunteer s",
+      "肌 肤",
+      "▁v otes",
+      "▁vote s",
+      "▁vot es",
+      "▁vo tes",
+      "▁ votes",
+      "ur name",
+      "urn ame",
+      "分 米",
+      "▁ad mitted",
+      "▁admit ted",
+      "▁adm itted",
+      "礼 物",
+      "equ als",
+      "equal s",
+      "▁ 三",
+      "▁in tra",
+      "▁int ra",
+      "▁intr a",
+      "ol ine",
+      "oli ne",
+      "olin e",
+      "o line",
+      "ambig uation",
+      "冰 箱",
+      "▁d rag",
+      "▁dr ag",
+      "▁ drag",
+      "depend encies",
+      "▁p otent",
+      "▁pot ent",
+      "▁S on",
+      "▁So n",
+      "▁ Son",
+      "tv oid",
+      "t void",
+      "板 块",
+      "扶 贫",
+      "▁p d",
+      "▁ pd",
+      "C B",
+      "▁B ell",
+      "▁Be ll",
+      "▁Bel l",
+      "▁s ick",
+      "▁si ck",
+      "不 明",
+      "▁F ix",
+      "▁Fi x",
+      "▁ Fix",
+      "正 文",
+      "do ts",
+      "dot s",
+      "d ots",
+      "第二百五十七条 第",
+      "第二百五十七 条第",
+      "▁O regon",
+      "▁Ore gon",
+      "助 于",
+      "ter national",
+      "tern ational",
+      "▁any way",
+      "MI N",
+      "M IN",
+      "ig o",
+      "i go",
+      "▁出生 地",
+      "▁ 出生地",
+      "终 端",
+      "▁host s",
+      "▁ho sts",
+      "▁ hosts",
+      "it ter",
+      "itt er",
+      "排 列",
+      "独任 审判",
+      "化 简",
+      "▁P a",
+      "▁ Pa",
+      "毕业 生",
+      "全 力",
+      "▁n arr",
+      "▁na rr",
+      "Sc reen",
+      "S creen",
+      "水 利",
+      "▁f ert",
+      "▁fe rt",
+      "▁fer t",
+      "死 了",
+      "▁o nset",
+      "▁on set",
+      "▁ onset",
+      "▁hand ler",
+      "▁handle r",
+      "▁ handler",
+      "▁Admin istration",
+      "▁Administr ation",
+      "日向本院 申请",
+      "日 向本院申请",
+      "▁vari ance",
+      "▁var iance",
+      "▁ variance",
+      "▁new spaper",
+      "▁news paper",
+      "▁newsp aper",
+      "▁I MP",
+      "▁IM P",
+      "▁ IMP",
+      "▁im pair",
+      "▁imp air",
+      "ri ends",
+      "riend s",
+      "rien ds",
+      "▁man ually",
+      "▁manual ly",
+      "财产 保险",
+      "Ti m",
+      "T im",
+      "nE arly",
+      "n Early",
+      "▁comp uted",
+      "▁comput ed",
+      "▁compute d",
+      "▁ computed",
+      "▁trans ferred",
+      "▁transfer red",
+      "It er",
+      "I ter",
+      "园 林",
+      "民 一",
+      "ak h",
+      "a kh",
+      "▁t alent",
+      "▁tal ent",
+      "▁tale nt",
+      "日 的",
+      "里面 的",
+      "里 面的",
+      "\\\\( )\\\\)",
+      "\\\\() \\\\)",
+      "oph ys",
+      "o phys",
+      "便 是",
+      "ym l",
+      "y ml",
+      "▁t ur",
+      "▁tu r",
+      "▁ tur",
+      ")] (",
+      ") ](",
+      "有 趣",
+      "减 肥",
+      "季 节",
+      "减刑建议 书",
+      "提交 上诉状",
+      "tri angle",
+      "温 柔",
+      "清 华",
+      "行业 的",
+      "行 业的",
+      "▁p osit",
+      "▁pos it",
+      "▁po sit",
+      "OR K",
+      "▁P S",
+      "▁ PS",
+      "ap ol",
+      "apo l",
+      "a pol",
+      "▁H ouston",
+      "▁Hou ston",
+      "▁grad ient",
+      "▁ gradient",
+      "绝对 值",
+      "over line",
+      "▁graph ics",
+      "▁graphic s",
+      "▁ graphics",
+      "摩 托",
+      "▁en gage",
+      "▁eng age",
+      "Direct ory",
+      "fe atures",
+      "feature s",
+      "feat ures",
+      "标 题",
+      "ns t",
+      "n st",
+      "ch ant",
+      "chan t",
+      "cha nt",
+      "Sim ple",
+      "S imple",
+      "as ets",
+      "ase ts",
+      "aset s",
+      "a sets",
+      "at eur",
+      "ate ur",
+      "不符 合",
+      "不 符合",
+      "▁h ur",
+      "▁hu r",
+      "▁st abil",
+      "▁stab il",
+      "检察 员",
+      "▁R ose",
+      "▁Ro se",
+      "▁Ros e",
+      "▁target ing",
+      "正本 一份",
+      "▁mer ged",
+      "▁merge d",
+      "▁merg ed",
+      "▁ merged",
+      "字 母",
+      "pi d",
+      "p id",
+      "al ities",
+      "ali ties",
+      "def ined",
+      "define d",
+      "d efined",
+      "н о",
+      "千 元",
+      "ab etic",
+      "abet ic",
+      "abe tic",
+      "isl ative",
+      "▁lux ury",
+      "▁Sen ior",
+      "▁view ed",
+      "é e",
+      "▁S i",
+      "▁ Si",
+      "英 寸",
+      "rit ish",
+      "ritis h",
+      "能 否",
+      "▁sur prise",
+      "▁surpr ise",
+      "▁re set",
+      "▁res et",
+      "▁ reset",
+      "ether lands",
+      "序 列",
+      "▁work shop",
+      "▁works hop",
+      "体 会",
+      "cept or",
+      "cep tor",
+      "内 置",
+      "统计 学",
+      "▁g rab",
+      "▁gr ab",
+      "▁gra b",
+      "▁ grab",
+      "nD r",
+      "n Dr",
+      "了一 下",
+      "了 一下",
+      "TE ST",
+      "TES T",
+      "T EST",
+      "▁over view",
+      "▁ overview",
+      "从 此",
+      "Ca l",
+      "C al",
+      "▁hospital s",
+      "▁hospit als",
+      "▁t emporary",
+      "▁tem porary",
+      "▁tempor ary",
+      "▁R os",
+      "▁Ro s",
+      "▁found er",
+      "▁fo under",
+      "▁ founder",
+      "▁j un",
+      "▁ju n",
+      "▁ jun",
+      "IG N",
+      "I GN",
+      "IA L",
+      "I AL",
+      "篮 球",
+      "▁D id",
+      "▁Di d",
+      "▁ Did",
+      "og s",
+      "o gs",
+      "▁目 录",
+      "▁ 目录",
+      "or iginal",
+      "origin al",
+      "orig inal",
+      "Fr amework",
+      "Frame work",
+      "run time",
+      "ru ntime",
+      "常 常",
+      "time out",
+      "nDes cription",
+      "nDe scription",
+      "n Description",
+      "引 擎",
+      "较 好",
+      "▁S quare",
+      "▁Squ are",
+      "▁ Square",
+      "yl l",
+      "y ll",
+      "In valid",
+      "Inv alid",
+      "nAss ociation",
+      "n Association",
+      "糖尿 病",
+      "早 上",
+      "家 庄",
+      "牛 奶",
+      "ro ck",
+      "roc k",
+      "r ock",
+      "nT rans",
+      "nTr ans",
+      "n Trans",
+      "▁h ip",
+      "▁hi p",
+      "▁ hip",
+      "刘 某",
+      "预 期",
+      "▁l ake",
+      "▁la ke",
+      "▁ lake",
+      "农村信用合作 联社",
+      "是为 了",
+      "是 为了",
+      "书面 上诉的",
+      "iz abeth",
+      "▁in correct",
+      "▁inc orrect",
+      "▁port folio",
+      "▁ portfolio",
+      "仅 仅",
+      "司 机",
+      "m c",
+      "刑 诉",
+      "▁accompl ish",
+      "▁accomp lish",
+      "W I",
+      "hel per",
+      "help er",
+      "▁M achine",
+      "▁Mach ine",
+      "▁ Machine",
+      "▁s odium",
+      "co unter",
+      "count er",
+      "cou nter",
+      "c ounter",
+      "▁s pac",
+      "▁sp ac",
+      "▁spa c",
+      "▁sl ope",
+      "▁ slope",
+      "ol ving",
+      "olv ing",
+      "al so",
+      "als o",
+      "ntype def",
+      "ntyped ef",
+      "n typedef",
+      "▁bi ochemical",
+      "▁bio chemical",
+      "西 部",
+      "年轻 人",
+      "▁h az",
+      "▁ha z",
+      "▁ haz",
+      "▁ap pend",
+      "▁app end",
+      "▁appe nd",
+      "▁ append",
+      "V R",
+      "▁Mc C",
+      "▁pow der",
+      "hab ilit",
+      "habi lit",
+      "р о",
+      "an ny",
+      "ann y",
+      "权 限",
+      "▁S ab",
+      "▁Sa b",
+      "密 切",
+      "▁R og",
+      "▁Ro g",
+      "it test",
+      "itt est",
+      "i ttest",
+      "is lation",
+      "isl ation",
+      "ns o",
+      "n so",
+      "▁es say",
+      "▁ess ay",
+      "Po ol",
+      "P ool",
+      "uu id",
+      "u uid",
+      "动 的",
+      "房 价",
+      "▁m ile",
+      "▁mil e",
+      "▁mi le",
+      "▁ mile",
+      "se ns",
+      "sen s",
+      "s ens",
+      "请 问",
+      "▁\\' __",
+      "▁\\'_ _",
+      "\\\\ }",
+      "▁Olymp ic",
+      "更 高",
+      "MA T",
+      "M AT",
+      "Service s",
+      "Serv ices",
+      "S ervices",
+      "▁el igible",
+      "▁elig ible",
+      "▁The atre",
+      "▁municip ality",
+      "▁municipal ity",
+      "例 子",
+      "Med ia",
+      "M edia",
+      "▁j ew",
+      "▁je w",
+      "一 流",
+      "未来 的",
+      "未 来的",
+      "▁c yst",
+      "▁cy st",
+      "▁ cyst",
+      "Met a",
+      "Me ta",
+      "M eta",
+      "▁ch ose",
+      "▁cho se",
+      "接 收",
+      "nSt ep",
+      "nSte p",
+      "n Step",
+      "▁sur ge",
+      "▁surg e",
+      "▁g el",
+      "▁ge l",
+      "▁ gel",
+      "▁} {",
+      "▁ }{",
+      "一 辆",
+      "std out",
+      "▁n omin",
+      "▁no min",
+      "▁nom in",
+      "可在 接到",
+      "受 伤",
+      "▁Scott ish",
+      "> \"",
+      "颗 粒",
+      "▁is chem",
+      "▁s take",
+      "▁st ake",
+      "▁se lective",
+      "▁select ive",
+      "▁sel ective",
+      "▁ selective",
+      "活动 的",
+      "活 动的",
+      "怎样 的",
+      "怎 样的",
+      "斗 争",
+      "per m",
+      "pe rm",
+      "p erm",
+      "ad ed",
+      "ade d",
+      "a ded",
+      "id ine",
+      "idi ne",
+      "idin e",
+      "▁T ok",
+      "▁To k",
+      "▁ Tok",
+      "▁cap ac",
+      "▁ca pac",
+      "up dated",
+      "update d",
+      "▁vari ants",
+      "▁variant s",
+      "▁ variants",
+      "参 观",
+      "响 应",
+      "我 很",
+      "▁co unts",
+      "▁count s",
+      "▁ counts",
+      "▁ess entially",
+      "▁essential ly",
+      "▁s ad",
+      "▁sa d",
+      "} [",
+      "▁N ever",
+      "▁Ne ver",
+      "▁Nev er",
+      "▁ Never",
+      "县 委",
+      "临 时",
+      "▁W i",
+      "▁ Wi",
+      "日起 十日内",
+      "▁f n",
+      "▁ fn",
+      "形 状",
+      "▁第 二",
+      "▁ 第二",
+      "▁V en",
+      "▁Ve n",
+      "容 器",
+      "▁se paration",
+      "▁separ ation",
+      "ra q",
+      "r aq",
+      "▁d ust",
+      "▁du st",
+      "▁P olicy",
+      "▁Pol icy",
+      "▁ Policy",
+      "实 时",
+      "频 道",
+      "▁m arker",
+      "▁mark er",
+      "▁mar ker",
+      "▁ marker",
+      "Vo l",
+      "V ol",
+      "lect ric",
+      "▁G O",
+      "▁ GO",
+      "▁O l",
+      "▁ Ol",
+      "决 赛",
+      "车 型",
+      "': \\",
+      "' :\\",
+      "▁u b",
+      "▁ ub",
+      "Par se",
+      "Pars e",
+      "P arse",
+      "ik es",
+      "ike s",
+      "i kes",
+      "监 察",
+      "抛 物",
+      "宁 市",
+      "▁d oses",
+      "▁do ses",
+      "▁dose s",
+      "▁dos es",
+      "▁ 性别",
+      "但 被执行人",
+      "▁under t",
+      "▁und ert",
+      "▁sh oes",
+      "▁shoe s",
+      "▁u ltimate",
+      "▁ult imate",
+      "▁s ought",
+      "▁so ught",
+      "▁sou ght",
+      "ap ps",
+      "app s",
+      "▁in valid",
+      "▁inv alid",
+      "▁ invalid",
+      "加 拿",
+      "▁F ord",
+      "▁For d",
+      "▁Fo rd",
+      "签 约",
+      "经济 学",
+      "ag netic",
+      "agn etic",
+      "agne tic",
+      "agnet ic",
+      "Copy right",
+      "C opyright",
+      "▁M E",
+      "▁ ME",
+      "▁v ice",
+      "▁vi ce",
+      "▁vic e",
+      "line d",
+      "li ned",
+      "lin ed",
+      "l ined",
+      "nO lymp",
+      "▁suff ering",
+      "▁suffer ing",
+      "▁J ean",
+      "▁Je an",
+      "▁ Jean",
+      "斯 坦",
+      "ph ia",
+      "phi a",
+      "der s",
+      "de rs",
+      "d ers",
+      "▁G ro",
+      "▁Gr o",
+      "▁t emporal",
+      "▁tem poral",
+      "▁tempor al",
+      "▁temp oral",
+      "▁tempo ral",
+      "▁ temporal",
+      "均 为",
+      "ct ic",
+      "c tic",
+      "▁res onance",
+      "▁reson ance",
+      "通过 对",
+      "▁interpret ation",
+      "▁N A",
+      "▁ NA",
+      "De l",
+      "D el",
+      "和 计划生育",
+      "▁in ches",
+      "▁inc hes",
+      "▁inch es",
+      "▁ inches",
+      "P y",
+      "▁Inst agram",
+      "▁break fast",
+      "▁over l",
+      "睡 觉",
+      "ad os",
+      "ado s",
+      "a dos",
+      "▁M att",
+      "▁Mat t",
+      "▁Ma tt",
+      "▁ Matt",
+      "▁n aturally",
+      "▁natural ly",
+      "▁nat urally",
+      "EN V",
+      "E NV",
+      "夏 天",
+      "style sheet",
+      "styles heet",
+      "的 需求",
+      "▁fr agment",
+      "▁frag ment",
+      "▁fra gment",
+      "▁ fragment",
+      "▁m agic",
+      "▁mag ic",
+      "▁ma gic",
+      "▁ magic",
+      "nE very",
+      "nEver y",
+      "nEv ery",
+      "n Every",
+      "Sup port",
+      "▁celebr ate",
+      "U G",
+      "扫 描",
+      "▁tw enty",
+      "▁twe nty",
+      "oprote in",
+      "opro tein",
+      "oprot ein",
+      "o protein",
+      "▁epit helial",
+      "▁ epithelial",
+      "独特 的",
+      "pe nd",
+      "pen d",
+      "p end",
+      "了 我",
+      "it udes",
+      "itude s",
+      "itud es",
+      "itu des",
+      "真的 很",
+      "真 的很",
+      "▁cl one",
+      "▁ clone",
+      "▁g am",
+      "▁ga m",
+      "▁ gam",
+      "巨大 的",
+      "巨 大的",
+      "▁Id ent",
+      "▁Ide nt",
+      "▁ Ident",
+      "▁life time",
+      "▁lif etime",
+      "派 出",
+      "bol ic",
+      "bo lic",
+      "b olic",
+      "cl osure",
+      "key words",
+      "keyword s",
+      "▁f lav",
+      "▁fl av",
+      "▁c ir",
+      "▁ci r",
+      "▁ cir",
+      "▁ang les",
+      "▁angle s",
+      "▁ angles",
+      "▁P rom",
+      "▁Pro m",
+      "▁Pr om",
+      "▁ Prom",
+      "阿 里",
+      "im p",
+      "i mp",
+      "▁eld erly",
+      "▁elder ly",
+      ", '",
+      "ja n",
+      "j an",
+      "▁streng then",
+      "▁strength en",
+      "▁inj ured",
+      "走 进",
+      "标 签",
+      "无 效",
+      "在 家",
+      "ver bose",
+      "verb ose",
+      "处 置",
+      "结 算",
+      "oc om",
+      "oco m",
+      "o com",
+      "▁[ ];\\",
+      "▁[] ;\\",
+      "▁ [];\\",
+      ": /",
+      "er ship",
+      "ers hip",
+      "V P",
+      "dis patch",
+      "disp atch",
+      "▁arbitr ary",
+      "nA p",
+      "n Ap",
+      "fo x",
+      "f ox",
+      "怀 疑",
+      "拿 到",
+      "▁he ating",
+      "▁heat ing",
+      "Bo olean",
+      "Bool ean",
+      "lo n",
+      "l on",
+      "处理 器",
+      "晋江文学 网",
+      "▁print ing",
+      "▁ob viously",
+      "▁obvious ly",
+      "Route r",
+      "Ro uter",
+      "R outer",
+      "sl ug",
+      "由 题意",
+      "▁Exec utive",
+      "▁ch ocolate",
+      "▁C apt",
+      "▁Ca pt",
+      "▁Cap t",
+      "▁ Capt",
+      "建 材",
+      "hel lo",
+      "hell o",
+      "h ello",
+      "▁Ph ase",
+      "▁ Phase",
+      "▁Mus lim",
+      "今 后",
+      "Serial izer",
+      "Serialize r",
+      "S erializer",
+      "书 画",
+      "报 警",
+      "辈 子",
+      "▁ac know",
+      "▁F ish",
+      "▁Fi sh",
+      "id ays",
+      "iday s",
+      "ida ys",
+      "i days",
+      "▁b irds",
+      "▁bir ds",
+      "▁bird s",
+      "▁ birds",
+      "号 起诉书",
+      "class Name",
+      "▁j s",
+      "▁ js",
+      "级 别",
+      "it able",
+      "ita ble",
+      "i table",
+      "▁s orted",
+      "▁sort ed",
+      "▁sor ted",
+      "▁ sorted",
+      "荣 获",
+      "▁cou pled",
+      "▁couple d",
+      "▁coup led",
+      "▁ coupled",
+      "▁he ads",
+      "▁head s",
+      "▁ heads",
+      "现代 化",
+      "不 当",
+      "▁R om",
+      "▁Ro m",
+      "▁ Rom",
+      "足以 认定",
+      "监 护",
+      "▁sit uated",
+      "▁situ ated",
+      "ar ring",
+      "arr ing",
+      "国家 的",
+      "国 家的",
+      "▁efficient ly",
+      "▁abs orption",
+      "▁absor ption",
+      "▁he aders",
+      "▁head ers",
+      "▁header s",
+      "▁ headers",
+      "ir k",
+      "i rk",
+      "▁P ers",
+      "▁Per s",
+      "▁Pe rs",
+      "▁ Pers",
+      "的大 小",
+      "的 大小",
+      "O p",
+      "信 心",
+      "待 遇",
+      "de code",
+      "dec ode",
+      "d ecode",
+      "data set",
+      "dat aset",
+      "datas et",
+      "▁fil ters",
+      "▁filter s",
+      "▁ filters",
+      "olester ol",
+      "▁Re source",
+      "▁Res ource",
+      "▁ Resource",
+      "▁sc hema",
+      "▁sch ema",
+      "▁ schema",
+      "cl uster",
+      "clus ter",
+      "▁en ding",
+      "▁end ing",
+      "▁ ending",
+      "▁L LC",
+      "▁LL C",
+      "▁o scill",
+      "▁L ie",
+      "▁Li e",
+      "ag ma",
+      "ma sk",
+      "mas k",
+      "m ask",
+      "▁ и",
+      "gen erate",
+      "gener ate",
+      "gene rate",
+      "g enerate",
+      "en abled",
+      "ena bled",
+      "enable d",
+      "传 感",
+      "▁TH IS",
+      "电 信",
+      "皇 帝",
+      "Ex pression",
+      "Exp ression",
+      "Express ion",
+      "Expr ession",
+      ")* *",
+      ") **",
+      "▁re ly",
+      "▁r ely",
+      "▁rel y",
+      "透 明",
+      "▁Cam bridge",
+      "▁Camb ridge",
+      "▁b ounded",
+      "▁bound ed",
+      "▁ bounded",
+      "▁as ympt",
+      "▁asym pt",
+      "par ty",
+      "part y",
+      "p arty",
+      "C N",
+      "\" @",
+      "的 定义",
+      "▁T ele",
+      "▁Te le",
+      "▁Tel e",
+      "▁ Tele",
+      "中华人民共和国行政 诉讼法",
+      "留 学",
+      "▁t om",
+      "▁to m",
+      "▁ tom",
+      "st ory",
+      "sto ry",
+      "stor y",
+      "Time out",
+      "▁Sing apore",
+      "▁Q uant",
+      "▁Qu ant",
+      "▁ Quant",
+      "▁re combin",
+      "这也 是",
+      "这 也是",
+      "ab ases",
+      "abase s",
+      "aba ses",
+      "a bases",
+      "的 学生",
+      "▁h at",
+      "▁ha t",
+      "▁ hat",
+      "本院 予以支持",
+      "本院予以 支持",
+      "▁rel ief",
+      "讲 话",
+      "fig ure",
+      "f igure",
+      "ur rence",
+      "urre nce",
+      "urrenc e",
+      "▁\" __",
+      "▁\"_ _",
+      "▁ \"__",
+      "▁ 发",
+      "▁en thus",
+      "▁ent hus",
+      "▁class ified",
+      "Re t",
+      "R et",
+      "adel phia",
+      "已 发生法律效力",
+      "▁Sup reme",
+      "np ub",
+      "n pub",
+      "▁c ond",
+      "▁con d",
+      "▁co nd",
+      "▁ cond",
+      "▁decre asing",
+      "▁Pro perty",
+      "▁Prop erty",
+      "▁Proper ty",
+      "▁ Property",
+      "▁W ITH",
+      "▁WIT H",
+      "▁WI TH",
+      "▁ WITH",
+      "要 注意",
+      "col l",
+      "co ll",
+      "c oll",
+      "笔记 本",
+      "性 价比",
+      "▁A le",
+      "▁Al e",
+      "▁ Ale",
+      "相当 于",
+      "江 湖",
+      "▁d elta",
+      "▁del ta",
+      "▁ delta",
+      "▁inter mediate",
+      "▁intermed iate",
+      "▁Sil ver",
+      "▁ill ust",
+      "▁exp ressions",
+      "▁expression s",
+      "▁express ions",
+      "▁expr essions",
+      "明显 的",
+      "▁less ons",
+      "▁lesson s",
+      "Ta b",
+      "T ab",
+      "一 块",
+      "▁off icially",
+      "▁offic ially",
+      "▁official ly",
+      "▁p aram",
+      "▁par am",
+      "▁para m",
+      "▁pa ram",
+      "▁ param",
+      "城 县",
+      "in os",
+      "ino s",
+      "i nos",
+      "内 涵",
+      "▁s i",
+      "▁ si",
+      "ul atory",
+      "ulator y",
+      "im ents",
+      "iment s",
+      "imen ts",
+      "i ments",
+      "▁f usion",
+      "▁fus ion",
+      "▁ fusion",
+      "▁H amilton",
+      "▁reg ulations",
+      "▁regul ations",
+      "▁regulation s",
+      "browse r",
+      "b rowser",
+      "▁h al",
+      "▁ha l",
+      "▁ hal",
+      "T M",
+      "AB C",
+      "A BC",
+      "▁n ur",
+      "▁nu r",
+      "▁ nur",
+      "OR Y",
+      "O RY",
+      "ix in",
+      "i xin",
+      "▁re aching",
+      "▁reach ing",
+      "▁Cons ider",
+      "清 理",
+      "▁expl ains",
+      "▁explain s",
+      "还 不",
+      "BA SE",
+      "B ASE",
+      "us al",
+      "usa l",
+      "u sal",
+      "▁S oc",
+      "▁So c",
+      "en sis",
+      "ens is",
+      "商 场",
+      "的过程 中",
+      "的 过程中",
+      "not ation",
+      "n otation",
+      "▁art ificial",
+      "ig e",
+      "i ge",
+      "▁A rizona",
+      "Ar r",
+      "A rr",
+      "vious ly",
+      "vi ously",
+      "v iously",
+      "▁therap ies",
+      "▁con clude",
+      "▁de light",
+      "▁del ight",
+      "Pol icy",
+      "P olicy",
+      "耳 机",
+      "▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁",
+      "▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁",
+      "▁ ▁▁▁▁▁▁▁▁",
+      "▁t elling",
+      "▁tell ing",
+      "▁tel ling",
+      "▁ telling",
+      "▁manufact urers",
+      "▁manufacturer s",
+      "▁manufacture rs",
+      "Me d",
+      "M ed",
+      "疫 苗",
+      "Trans form",
+      "▁s avings",
+      "▁sav ings",
+      "▁saving s",
+      "邮 件",
+      "▁cal endar",
+      "▁ calendar",
+      "ST RA",
+      "STR A",
+      "S TRA",
+      "Hi gh",
+      "H igh",
+      "▁der ivatives",
+      "▁deriv atives",
+      "▁derivative s",
+      "▁ver tex",
+      "▁vert ex",
+      "▁verte x",
+      "▁ vertex",
+      "▁in struments",
+      "▁instr uments",
+      "▁instrument s",
+      "不 但",
+      "▁Mult i",
+      "▁Mul ti",
+      "▁ Multi",
+      "▁O cean",
+      "▁Oc ean",
+      "▁g iant",
+      "▁gi ant",
+      "No w",
+      "N ow",
+      "nW ork",
+      "n Work",
+      "au ge",
+      "aug e",
+      "a uge",
+      "nJ ohn",
+      "nJoh n",
+      "nJo hn",
+      "n John",
+      "▁J ewish",
+      "▁Jew ish",
+      "失 信",
+      "▁Ind iana",
+      "▁India na",
+      "▁Indian a",
+      "arch ive",
+      "arc hive",
+      "ave length",
+      "avel ength",
+      "组成 员",
+      "组 成员",
+      "▁end othelial",
+      "▁endot helial",
+      "纪 律",
+      "▁eng aged",
+      "▁engage d",
+      "开 关",
+      "ass ociated",
+      "associ ated",
+      "年 在",
+      "▁A nton",
+      "▁An ton",
+      "▁Ant on",
+      "十二 条第二款",
+      "十二条 第二款",
+      "成 交",
+      "* ;\\",
+      "SE LECT",
+      "QU E",
+      "Q UE",
+      "V I",
+      "í a",
+      "cf g",
+      "c fg",
+      "头 发",
+      "an gers",
+      "ang ers",
+      "ange rs",
+      "anger s",
+      "▁re spective",
+      "▁res pective",
+      "▁respect ive",
+      "▁resp ective",
+      "▁f rag",
+      "▁fr ag",
+      "▁fra g",
+      "▁ frag",
+      "And roid",
+      "线 性",
+      "三个 月",
+      "三 个月",
+      "ri tical",
+      "rit ical",
+      "ritic al",
+      "▁C F",
+      "▁ CF",
+      "▁m odes",
+      "▁mod es",
+      "▁mode s",
+      "▁mo des",
+      "销 量",
+      "n z",
+      "密 封",
+      "es tab",
+      "est ab",
+      "esta b",
+      "val idation",
+      "valid ation",
+      "ous tic",
+      "oust ic",
+      "▁im plies",
+      "▁imp lies",
+      "▁impl ies",
+      "▁ implies",
+      "▁ch ampion",
+      "▁Trans port",
+      "▁ Transport",
+      "▁S mart",
+      "▁Sm art",
+      "▁ Smart",
+      "PA R",
+      "P AR",
+      "▁Qu estions",
+      "▁Question s",
+      "▁Quest ions",
+      "物 资",
+      "nH is",
+      "nHi s",
+      "n His",
+      "▁Stud ents",
+      "▁Student s",
+      "▁ Students",
+      "已 有",
+      "\\' ):\\",
+      "\\') :\\",
+      "\\ '):\\",
+      "▁athlet es",
+      "▁athlete s",
+      "▁H R",
+      "▁ HR",
+      "▁m ature",
+      "▁mat ure",
+      "▁ma ture",
+      "stand ard",
+      "终 审",
+      "et ary",
+      "eta ry",
+      "p k",
+      "▁accomp an",
+      "c v",
+      "回 顾",
+      "▁ar th",
+      "▁art h",
+      "▁ arth",
+      "▁F inal",
+      "▁Fin al",
+      "▁Fi nal",
+      "▁ Final",
+      "Pack age",
+      "P ackage",
+      "hyth m",
+      "ri eve",
+      "rie ve",
+      "riev e",
+      "r ieve",
+      "别 名",
+      "he alth",
+      "squ are",
+      "s quare",
+      "这 篇",
+      "▁p ic",
+      "▁pi c",
+      "▁ pic",
+      "▁indust ries",
+      "终结 执行",
+      "南 方",
+      "▁phot ography",
+      "▁photograph y",
+      "下 一",
+      "Ser v",
+      "Se rv",
+      "S erv",
+      "el lect",
+      "ell ect",
+      "elle ct",
+      "fl y",
+      "f ly",
+      "▁g ang",
+      "▁ga ng",
+      "▁gan g",
+      "▁ gang",
+      "▁strong er",
+      "圆 柱",
+      "nM ed",
+      "nMe d",
+      "n Med",
+      "并 从",
+      "▁w ins",
+      "▁win s",
+      "▁wi ns",
+      "▁intr acellular",
+      "▁intra cellular",
+      "▁intrac ellular",
+      "中国 人",
+      "中 国人",
+      "cri bed",
+      "crib ed",
+      "cribe d",
+      "▁a pt",
+      "▁ap t",
+      "▁ apt",
+      "▁own ership",
+      "▁owners hip",
+      "▁owner ship",
+      "杭州 市",
+      "杭 州市",
+      "dis ambiguation",
+      "Da y",
+      "D ay",
+      "地区 的",
+      "地 区的",
+      "朝 阳",
+      "向 上",
+      "▁ar rive",
+      "▁arr ive",
+      "▁arriv e",
+      "per ature",
+      "▁ 支持",
+      "▁che ese",
+      "▁chees e",
+      "动 脉",
+      "元及 利息",
+      "元 及利息",
+      "▁C op",
+      "▁Co p",
+      "▁ Cop",
+      "江 市",
+      "Do main",
+      "Dom ain",
+      "D omain",
+      "生 理",
+      "tre ated",
+      "t reated",
+      "声 明",
+      "▁W at",
+      "▁Wa t",
+      "在一 个",
+      "在 一个",
+      "▁H ash",
+      "▁Ha sh",
+      "▁Has h",
+      "▁ Hash",
+      "概 述",
+      "▁Bo olean",
+      "▁Bool ean",
+      "▁ Boolean",
+      "er os",
+      "ero s",
+      "e ros",
+      "▁F arm",
+      "▁Far m",
+      "▁Fa rm",
+      "▁C ould",
+      "▁Cou ld",
+      "▁Coul d",
+      "▁ Could",
+      "io rs",
+      "ior s",
+      "i ors",
+      "By tes",
+      "Byte s",
+      "B ytes",
+      "riv ial",
+      "问题 和",
+      "▁g est",
+      "▁ge st",
+      "▁ gest",
+      "设计 师",
+      "▁D atabase",
+      "▁Data base",
+      "▁Dat abase",
+      "▁ Database",
+      "过 于",
+      "▁fl ows",
+      "▁flo ws",
+      "▁flow s",
+      "民 众",
+      "补 贴",
+      "en z",
+      "e nz",
+      "nN ational",
+      "n National",
+      "实 例",
+      "Edit or",
+      "Ed itor",
+      "▁N atural",
+      "▁Nat ural",
+      "▁ Natural",
+      "un es",
+      "une s",
+      "u nes",
+      "▁con cluded",
+      "▁conclude d",
+      "Be an",
+      "B ean",
+      "▁res istant",
+      "▁resist ant",
+      "▁ resistant",
+      "Ob j",
+      "O bj",
+      "▁d ocs",
+      "▁do cs",
+      "▁doc s",
+      "▁ docs",
+      "▁( '",
+      "▁ ('",
+      "▁prom oting",
+      "▁promot ing",
+      "▁promo ting",
+      "work s",
+      "wor ks",
+      "▁E N",
+      "▁ EN",
+      "▁embed ded",
+      "▁ embedded",
+      "▁disc rete",
+      "▁discre te",
+      "▁discret e",
+      "网 页",
+      "log ical",
+      "logic al",
+      "l ogical",
+      "▁dis plays",
+      "▁display s",
+      "et r",
+      "e tr",
+      "▁L A",
+      "▁ LA",
+      "col umns",
+      "column s",
+      "colum ns",
+      "So rt",
+      "S ort",
+      "辐 射",
+      "port ers",
+      "por ters",
+      "porter s",
+      "p orters",
+      "的 世界",
+      "排 除",
+      "________ ________",
+      "▁random ly",
+      "乘 法",
+      "▁s ky",
+      "▁sk y",
+      "▁ sky",
+      "S W",
+      "Oper ation",
+      "Op eration",
+      "▁f inance",
+      "▁fin ance",
+      "▁financ e",
+      "▁ finance",
+      "▁degrad ation",
+      "参考 文献",
+      "'] ,\\",
+      "'], \\",
+      "' ],\\",
+      "▁an ten",
+      "▁ant en",
+      "海 南",
+      "▁w rt",
+      "▁wr t",
+      "in king",
+      "ink ing",
+      "inki ng",
+      "config uration",
+      "Def inition",
+      "▁in equality",
+      "link ed",
+      "▁go ogle",
+      "▁goog le",
+      "▁ google",
+      "▁part ially",
+      "▁partial ly",
+      "▁im g",
+      "▁i mg",
+      "▁ img",
+      "的 知识",
+      "▁eg gs",
+      "▁egg s",
+      "is ations",
+      "isation s",
+      "or ption",
+      "orp tion",
+      "De pend",
+      "Dep end",
+      "D epend",
+      "▁work er",
+      "▁wor ker",
+      "▁ worker",
+      "iz za",
+      "izz a",
+      "cre at",
+      "cr eat",
+      "c reat",
+      "nS ent",
+      "nSe nt",
+      "nSen t",
+      "n Sent",
+      "害 怕",
+      "pl ify",
+      "p lify",
+      "▁specific ity",
+      "as ive",
+      "asi ve",
+      "oe n",
+      "o en",
+      "市 长",
+      "▁h oping",
+      "▁hop ing",
+      "▁ho ping",
+      "▁p eer",
+      "▁pe er",
+      "▁ peer",
+      "▁unus ual",
+      "▁inter vals",
+      "▁interval s",
+      "▁c ancell",
+      "▁can cell",
+      "▁cancel l",
+      "日 上午",
+      "nA rt",
+      "nAr t",
+      "n Art",
+      "nE n",
+      "n En",
+      "▁ о",
+      "▁c arb",
+      "▁car b",
+      "▁ca rb",
+      "▁sw imming",
+      "▁swim ming",
+      "pe at",
+      "p eat",
+      "秩 序",
+      "import ant",
+      "▁ 孙",
+      "作 物",
+      "无 锡",
+      "▁re fresh",
+      "▁ref resh",
+      "▁refr esh",
+      "▁refres h",
+      "▁ refresh",
+      "un ter",
+      "unt er",
+      "unte r",
+      "u nter",
+      "▁dis abled",
+      "▁disable d",
+      "▁ disabled",
+      "▁develop mental",
+      "▁development al",
+      "工作 中",
+      "ivari ate",
+      "iva riate",
+      "ivar iate",
+      "▁W rit",
+      "▁Wr it",
+      "▁ Writ",
+      "▁o rientation",
+      "▁orient ation",
+      "▁ orientation",
+      "* )",
+      "▁e gg",
+      "▁eg g",
+      "▁ egg",
+      "ap pa",
+      "app a",
+      "▁m ethyl",
+      "▁meth yl",
+      "▁ methyl",
+      "▁favour ite",
+      "新 冠",
+      "▁ver bose",
+      "▁verb ose",
+      "▁ verbose",
+      "D i",
+      "▁rep utation",
+      "▁reput ation",
+      "证 言",
+      "提高 了",
+      "提 高了",
+      "彩 票",
+      "att ributes",
+      "attribute s",
+      "attrib utes",
+      "O mega",
+      "副 局长",
+      "撤回 上诉",
+      "音 频",
+      "▁ant agon",
+      "att ribute",
+      "attrib ute",
+      "equ ence",
+      "eque nce",
+      "re ation",
+      "reat ion",
+      "rea tion",
+      "ic hael",
+      "ich ael",
+      "nOlymp ic",
+      "▁ad venture",
+      "▁advent ure",
+      "拘 役",
+      "国 产",
+      "幸 运",
+      "ab a",
+      "a ba",
+      "▁em pir",
+      "▁emp ir",
+      "od ge",
+      "▁r h",
+      "▁ rh",
+      "诈 骗",
+      "▁C ub",
+      "▁Cu b",
+      "见 过",
+      "判决书 的第二",
+      "第一百 七",
+      "单 独",
+      "因 被执行人",
+      "▁re form",
+      "▁ref orm",
+      "gr es",
+      "gre s",
+      "g res",
+      "百度 百科",
+      "▁B rand",
+      "▁Br and",
+      "▁Bra nd",
+      "▁ Brand",
+      "Hand le",
+      "Go ogle",
+      "别人 的",
+      "别 人的",
+      "▁出 处",
+      "▁ 出处",
+      "co ord",
+      "sp ective",
+      "spect ive",
+      "s pective",
+      "▁process ed",
+      "▁proc essed",
+      "▁ processed",
+      "优 良",
+      "它 是",
+      "各 大",
+      ">; \\",
+      "> ;\\",
+      "▁system atic",
+      "▁syst ematic",
+      "能 不能",
+      "将 军",
+      "nU sing",
+      "nUs ing",
+      "n Using",
+      "To tal",
+      "T otal",
+      "一 系列",
+      "IM IT",
+      "I MIT",
+      "▁V iet",
+      "▁Vi et",
+      "▁Vie t",
+      "con struct",
+      "const ruct",
+      "co nstruct",
+      "▁chem otherapy",
+      "▁chemo therapy",
+      "ight h",
+      "igh th",
+      "▁ser iously",
+      "▁serious ly",
+      "编 写",
+      "pat ient",
+      "▁R O",
+      "▁ RO",
+      "收 购",
+      "▁pro state",
+      "▁prost ate",
+      "适用 法律",
+      "ric ulum",
+      "看 的",
+      "美 女",
+      "的 专业",
+      "ivers ary",
+      "▁n ob",
+      "▁no b",
+      "▁ nob",
+      "监督 管理局",
+      "监督管理 局",
+      "w ikipedia",
+      "▁Ed ward",
+      "Do n",
+      "D on",
+      "激 励",
+      "我 知道",
+      "看 法",
+      "▁cert ified",
+      "ne cho",
+      "nec ho",
+      "n echo",
+      "▁author ity",
+      "▁L ook",
+      "▁Lo ok",
+      "▁ Look",
+      "住 安徽省",
+      "▁ob esity",
+      "▁t ens",
+      "▁te ns",
+      "▁ten s",
+      "背 后",
+      "▁ass ays",
+      "▁assay s",
+      "复 习",
+      "判令 被告",
+      "本院经审查 认为",
+      "锈 钢",
+      "不 久",
+      "履行 义务",
+      "口 味",
+      "清 偿",
+      "bu ilt",
+      "b uilt",
+      "nG iven",
+      "nGive n",
+      "n Given",
+      "订 单",
+      "Ex ample",
+      "▁conf ident",
+      "▁list ening",
+      "▁listen ing",
+      "▁up take",
+      "要 素",
+      "dir ection",
+      "di rection",
+      "direct ion",
+      "d irection",
+      "nA b",
+      "n Ab",
+      "lo ss",
+      "los s",
+      "l oss",
+      "是非 常",
+      "是 非常",
+      "▁T ag",
+      "▁Ta g",
+      "▁ Tag",
+      "▁H ung",
+      "▁Hu ng",
+      "▁Hun g",
+      "▁W rite",
+      "▁Writ e",
+      "▁Wr ite",
+      "▁ Write",
+      "d h",
+      "睡 眠",
+      "▁conf used",
+      "YP E",
+      "Y PE",
+      "▁n ums",
+      "▁num s",
+      "▁nu ms",
+      "▁ nums",
+      "rior ity",
+      "细 菌",
+      "▁med ication",
+      "▁medi cation",
+      "▁medic ation",
+      "US E",
+      "U SE",
+      "▁\\' @",
+      "uc lear",
+      "ucle ar",
+      "ucl ear",
+      "u clear",
+      "▁\\\\ \"",
+      "▁ \\\\\"",
+      "▁r ose",
+      "▁ro se",
+      "▁ros e",
+      "▁ rose",
+      "▁S ound",
+      "▁So und",
+      "▁Sou nd",
+      "▁ Sound",
+      "高 质量",
+      "▁r iv",
+      "▁ri v",
+      "▁ riv",
+      "') ,",
+      "' ),",
+      "公开开庭 审理了本案",
+      "▁D avis",
+      "▁Da vis",
+      "▁Dav is",
+      "gl y",
+      "g ly",
+      "oc yt",
+      "o cyt",
+      "▁prom oted",
+      "▁promot ed",
+      "▁promote d",
+      "▁promo ted",
+      "er d",
+      "e rd",
+      "▁C PU",
+      "▁CP U",
+      "▁ CPU",
+      "洗 衣",
+      "▁occ as",
+      "▁oc cas",
+      "wa l",
+      "w al",
+      "▁s pectral",
+      "▁spect ral",
+      "▁spectra l",
+      "▁公司 类型",
+      "▁s pan",
+      "▁sp an",
+      "▁spa n",
+      "▁ span",
+      "▁C y",
+      "▁ Cy",
+      "癫 痫",
+      "plot lib",
+      "新 增",
+      "an ga",
+      "ang a",
+      "▁S A",
+      "▁ SA",
+      "▁A thlet",
+      "▁Ath let",
+      "dis c",
+      "di sc",
+      "d isc",
+      "▁e astern",
+      "▁eas tern",
+      "▁east ern",
+      "▁ eastern",
+      "▁j oy",
+      "▁jo y",
+      "▁ joy",
+      "az a",
+      "a za",
+      "Lo c",
+      "L oc",
+      "▁Q uality",
+      "▁Qu ality",
+      "▁Qual ity",
+      "▁ Quality",
+      "C r",
+      "pa g",
+      "p ag",
+      "▁pos ting",
+      "▁post ing",
+      "工 会",
+      "▁p our",
+      "▁po ur",
+      "▁pou r",
+      "尼 亚",
+      "Ch anged",
+      "Change d",
+      "Chan ged",
+      "im en",
+      "ime n",
+      "i men",
+      "引 领",
+      "▁_ {",
+      "▁ _{",
+      "wi ch",
+      "w ich",
+      "因 涉嫌",
+      "▁sur prising",
+      "▁surpr ising",
+      "交付 执行",
+      "初 步",
+      "买卖 合同纠纷一案",
+      "买卖合同 纠纷一案",
+      "买卖合同纠纷 一案",
+      "试 行",
+      "▁ 拼音",
+      "▁p owers",
+      "▁power s",
+      "▁po wers",
+      "▁pow ers",
+      "▁ powers",
+      "▁R a",
+      "▁ Ra",
+      "▁tox icity",
+      "▁toxic ity",
+      "▁ toxicity",
+      "被 评为",
+      "▁electric ity",
+      "▁electr icity",
+      "▁l ikes",
+      "▁li kes",
+      "▁like s",
+      "▁lik es",
+      "▁ likes",
+      "ut o",
+      "u to",
+      "▁cap ability",
+      "这是 一个",
+      "这是一 个",
+      "这 是一个",
+      "▁\\\\ {",
+      "▁ \\\\{",
+      "项目 的",
+      "项 目的",
+      "Y S",
+      "le arn",
+      "lear n",
+      "lea rn",
+      "l earn",
+      "▁pre ced",
+      "▁pr eced",
+      "▁prec ed",
+      "▁ preced",
+      "em ony",
+      "emon y",
+      "emo ny",
+      "人 文",
+      "▁influ enced",
+      "▁influence d",
+      "▁c ute",
+      "▁cut e",
+      "▁cu te",
+      "▁Ch annel",
+      "▁Chan nel",
+      "▁ Channel",
+      "昆 明",
+      "快 递",
+      "m x",
+      "期 刊",
+      "ry ption",
+      "rypt ion",
+      "Create d",
+      "Creat ed",
+      "Cre ated",
+      "C reated",
+      "质 证",
+      "▁p ep",
+      "▁pe p",
+      "▁ pep",
+      "av ailable",
+      "搅 拌",
+      "▁person ality",
+      "▁personal ity",
+      "is ely",
+      "ise ly",
+      "▁graph s",
+      "▁grap hs",
+      "▁ graphs",
+      "or able",
+      "ora ble",
+      "nb ut",
+      "n but",
+      "世界 的",
+      "世 界的",
+      "凤 凰",
+      "▁Re move",
+      "▁Rem ove",
+      "▁ Remove",
+      "er ty",
+      "ert y",
+      "ha ust",
+      "haus t",
+      "所谓 的",
+      "▁ex traction",
+      "▁extra ction",
+      "▁extract ion",
+      "▁extr action",
+      "▁extrac tion",
+      "my sql",
+      "mys ql",
+      "m ysql",
+      "▁f ighting",
+      "▁fight ing",
+      "没 想到",
+      "em an",
+      "ema n",
+      "e man",
+      "le tal",
+      "let al",
+      "l etal",
+      "▁g ear",
+      "▁ge ar",
+      "▁ gear",
+      "电视 台",
+      "▁C alcul",
+      "▁Calc ul",
+      "▁ Calcul",
+      "报送 本院审理",
+      "▁d os",
+      "▁do s",
+      "▁ dos",
+      "▁a rc",
+      "▁ar c",
+      "▁ arc",
+      "G lobal",
+      "值 为",
+      "宣 告",
+      "编 剧",
+      "一 大",
+      "西 班",
+      "▁tr act",
+      "▁tra ct",
+      "▁ tract",
+      "▁ 用",
+      "▁b id",
+      "▁bi d",
+      "▁ bid",
+      "▁ex hibition",
+      "▁exhib ition",
+      "▁exhibit ion",
+      "anal ysis",
+      "回 应",
+      "▁cl inic",
+      "▁clin ic",
+      "▁cli nic",
+      "▁d eck",
+      "▁de ck",
+      "▁dec k",
+      "▁ deck",
+      "SI ZE",
+      "S IZE",
+      "树 立",
+      "go od",
+      "g ood",
+      "▁b owl",
+      "▁bo wl",
+      "▁bow l",
+      "刑 执字第",
+      "si t",
+      "s it",
+      "TI ME",
+      "TIM E",
+      "T IME",
+      "话 题",
+      "▁sett lement",
+      "▁settle ment",
+      "▁leg islation",
+      "▁legisl ation",
+      "▁Gover nor",
+      "▁Govern or",
+      "一 共",
+      "依 靠",
+      "▁c aps",
+      "▁cap s",
+      "▁ca ps",
+      "▁ caps",
+      "▁autom ated",
+      "▁automate d",
+      "完 了",
+      "▁stre ets",
+      "▁street s",
+      "▁\\ '\\\\",
+      "▁\\' \\\\",
+      "▁\\'\\ \\",
+      "▁ \\'\\\\",
+      "▁m atched",
+      "▁mat ched",
+      "▁match ed",
+      "▁ matched",
+      "信息 的",
+      "▁Gold en",
+      "▁Gol den",
+      "iv ation",
+      "iva tion",
+      "i vation",
+      "} |",
+      "ul o",
+      "u lo",
+      "▁} )\\",
+      "▁}) \\",
+      "▁ })\\",
+      "< %",
+      "GE R",
+      "G ER",
+      "ph ase",
+      "pha se",
+      "▁integr ity",
+      "▁s ynchron",
+      "▁syn chron",
+      "▁ synchron",
+      "se ed",
+      "see d",
+      "s eed",
+      "▁F ollowing",
+      "▁Follow ing",
+      "▁ Following",
+      "▁C ell",
+      "▁Cel l",
+      "▁Ce ll",
+      "▁ Cell",
+      "▁recon struction",
+      "▁reconstruct ion",
+      "深 化",
+      "tr ace",
+      "tra ce",
+      "t race",
+      "▁engine er",
+      "▁car eful",
+      "▁care ful",
+      "▁e mission",
+      "▁em ission",
+      "显 然",
+      "▁l oves",
+      "▁love s",
+      "▁lo ves",
+      "保 养",
+      "西班 牙",
+      "mo l",
+      "m ol",
+      "ap ters",
+      "apt ers",
+      "apter s",
+      "▁G reg",
+      "▁Gr eg",
+      "▁Gre g",
+      "▁ Greg",
+      "进 度",
+      "н и",
+      "par ts",
+      "part s",
+      "p arts",
+      "CL UD",
+      "pub lish",
+      "▁g ym",
+      "▁gy m",
+      "▁r od",
+      "▁ro d",
+      "▁ rod",
+      "普 及",
+      "ic it",
+      "ici t",
+      "i cit",
+      "▁N etherlands",
+      "▁dis appoint",
+      "领 取",
+      "▁al le",
+      "▁all e",
+      "▁ alle",
+      "dis able",
+      "病 例",
+      "关 联",
+      "By te",
+      "B yte",
+      "宗 教",
+      "▁dist ributions",
+      "▁distribution s",
+      "▁distribut ions",
+      "▁ind uce",
+      "▁indu ce",
+      "nM ar",
+      "nMa r",
+      "n Mar",
+      "▁her self",
+      "▁hers elf",
+      "理 工",
+      "边 缘",
+      "通 用",
+      "▁t olerance",
+      "▁toler ance",
+      "▁A gency",
+      "▁Ag ency",
+      "▁Age ncy",
+      "▁appro ximation",
+      "▁approxim ation",
+      "▁approx imation",
+      "Sp ace",
+      "S pace",
+      "ul as",
+      "ula s",
+      "u las",
+      "▁suff ered",
+      "▁suffer ed",
+      "判决书的第二 日起十日内",
+      "nFil ms",
+      "nFilm s",
+      "▁b undle",
+      "▁bund le",
+      "▁ bundle",
+      "qu is",
+      "q uis",
+      "In v",
+      "I nv",
+      "▁s hips",
+      "▁sh ips",
+      "▁ship s",
+      "▁ ships",
+      "An g",
+      "A ng",
+      "▁prot ective",
+      "▁protect ive",
+      "▁ protective",
+      "ar dens",
+      "ard ens",
+      "arden s",
+      "arde ns",
+      "商 贸",
+      "▁f ishing",
+      "▁fish ing",
+      "▁B an",
+      "▁Ba n",
+      "▁ Ban",
+      "▁P ed",
+      "▁Pe d",
+      "▁ Ped",
+      "中医 药",
+      "中 医药",
+      "Cl oud",
+      "消 毒",
+      "短 期",
+      "大 街",
+      "急 性",
+      "疯 狂",
+      "▁C S",
+      "▁ CS",
+      "─ ─",
+      "an chester",
+      "anc hester",
+      "anch ester",
+      "anches ter",
+      "anche ster",
+      "▁D VD",
+      "▁ DVD",
+      "▁exhib ited",
+      "▁exhibit ed",
+      "▁l ets",
+      "▁le ts",
+      "▁let s",
+      "▁ lets",
+      "B T",
+      "you r",
+      "yo ur",
+      "y our",
+      "к о",
+      "▁st unning",
+      "▁r ental",
+      "▁rent al",
+      "代表 作品",
+      "代表作 品",
+      "十四 条",
+      "十 四条",
+      "▁med iated",
+      "▁media ted",
+      "▁medi ated",
+      "▁mediate d",
+      "▁ mediated",
+      "▁seg ments",
+      "▁segment s",
+      "▁ segments",
+      "抓 住",
+      "pro d",
+      "pr od",
+      "p rod",
+      "▁am plit",
+      "▁ampl it",
+      "▁amp lit",
+      "师 生",
+      "▁pand emic",
+      "一 件",
+      "表 情",
+      "听 说",
+      "▁le gs",
+      "▁leg s",
+      "▁consider able",
+      "▁r ice",
+      "▁ri ce",
+      "▁ric e",
+      "▁ rice",
+      "▁name ly",
+      "▁nam ely",
+      "▁sat ellite",
+      "▁ satellite",
+      "Ca m",
+      "C am",
+      "▁f lux",
+      "▁fl ux",
+      "▁flu x",
+      "▁ flux",
+      "▁we ights",
+      "▁weight s",
+      "▁weigh ts",
+      "▁ weights",
+      "com mands",
+      "comm ands",
+      "command s",
+      "not es",
+      "no tes",
+      "note s",
+      "n otes",
+      "心 脏",
+      "日 下午",
+      "文化 的",
+      "文 化的",
+      "Ent er",
+      "En ter",
+      "E nter",
+      "其 余",
+      "▁w er",
+      "▁we r",
+      "▁ wer",
+      "st an",
+      "sta n",
+      "s tan",
+      "大 了",
+      "▁r acing",
+      "▁ra cing",
+      "▁rac ing",
+      "▁c hest",
+      "▁ch est",
+      "▁che st",
+      "F loat",
+      "▁mis take",
+      "▁mist ake",
+      "▁emph asis",
+      "▁emphas is",
+      "▁reg ulated",
+      "▁regul ated",
+      "▁regulate d",
+      "▁ regulated",
+      "▁c md",
+      "▁cm d",
+      "▁ cmd",
+      "pos ts",
+      "post s",
+      "po sts",
+      "▁slow ly",
+      "▁s ensors",
+      "▁sens ors",
+      "▁sensor s",
+      "日 成立",
+      "原 始",
+      "we gian",
+      "▁cont ributed",
+      "▁contribut ed",
+      "▁contribute d",
+      "ou rier",
+      "our ier",
+      "ouri er",
+      "号起诉书 指控被告人",
+      "的 管理",
+      "▁P ack",
+      "▁Pac k",
+      "▁Pa ck",
+      "▁ Pack",
+      "▁I owa",
+      "▁Io wa",
+      "▁sec retion",
+      "▁secret ion",
+      "mit s",
+      "mi ts",
+      "m its",
+      "▁psych ological",
+      "▁psycho logical",
+      "之 处",
+      "的 手",
+      "挖 掘",
+      "RE D",
+      "R ED",
+      "Op t",
+      "O pt",
+      "了 个",
+      "作品 简介",
+      "(\" #",
+      "( \"#",
+      "▁Sw eden",
+      "▁Swed en",
+      "▁Swe den",
+      "▁E val",
+      "▁Ev al",
+      "▁Eva l",
+      "▁ Eval",
+      "▁d ependence",
+      "▁depend ence",
+      "t l",
+      "▁K al",
+      "▁Ka l",
+      ")} ^{",
+      ") }^{",
+      "ce ive",
+      "▁Art icle",
+      "▁ Article",
+      "ple mentation",
+      "plement ation",
+      "sh ift",
+      "shi ft",
+      "s hift",
+      "▁T ra",
+      "▁Tr a",
+      "▁ Tra",
+      "▁serv ers",
+      "▁ser vers",
+      "▁server s",
+      "▁serve rs",
+      "▁ servers",
+      "丈 夫",
+      "▁D I",
+      "▁ DI",
+      "▁Gu ard",
+      "▁ Guard",
+      "理 财",
+      "cu p",
+      "c up",
+      "纪 录",
+      "Pa nel",
+      "Pane l",
+      "Pan el",
+      "P anel",
+      "▁pre dom",
+      "▁pr edom",
+      "▁pred om",
+      "nP ar",
+      "n Par",
+      "▁O UT",
+      "▁ OUT",
+      "队 员",
+      "常 委",
+      "▁cont ributing",
+      "▁contribut ing",
+      "相 结合",
+      "ar ound",
+      "aro und",
+      "a round",
+      "nBack ground",
+      "n Background",
+      "▁util ize",
+      "岁 月",
+      "av irus",
+      "avi rus",
+      "avir us",
+      "a virus",
+      "type of",
+      "是 很",
+      "设计 的",
+      "west ern",
+      "w estern",
+      "y y",
+      "苏 联",
+      "RE S",
+      "R ES",
+      "付 出",
+      "anc ies",
+      "和 我",
+      "/) \\",
+      "/ )\\",
+      "带 动",
+      "lo ading",
+      "load ing",
+      "org eous",
+      "orge ous",
+      "in ch",
+      "inc h",
+      "i nch",
+      "▁s nap",
+      "▁sn ap",
+      "▁ snap",
+      "on i",
+      "o ni",
+      "有 可能",
+      "▁Stud ent",
+      "▁ Student",
+      "▁o cean",
+      "▁oc ean",
+      "中国 农业",
+      "模 板",
+      "▁mus icians",
+      "▁music ians",
+      "▁musician s",
+      "并 对",
+      "▁mac roph",
+      "▁macro ph",
+      "代 谢",
+      "▁Mod ern",
+      "▁Mode rn",
+      "无 需",
+      "ri ent",
+      "rie nt",
+      "rien t",
+      "r ient",
+      "▁protocol s",
+      "▁proto cols",
+      "▁cons ervation",
+      "▁conserv ation",
+      "▁p robe",
+      "▁pro be",
+      "▁pr obe",
+      "▁prob e",
+      "▁ probe",
+      "▁Que ens",
+      "▁Queen s",
+      "nS im",
+      "n Sim",
+      "in tage",
+      "int age",
+      "▁stat istically",
+      "▁statist ically",
+      "▁statistical ly",
+      "▁statistic ally",
+      "ploy ee",
+      "▁sc oring",
+      "▁U nfortunately",
+      "▁B uy",
+      "▁Bu y",
+      "▁ Buy",
+      "na t",
+      "n at",
+      "新 能源",
+      "she et",
+      "s heet",
+      "录 取",
+      "ho ok",
+      "h ook",
+      "▁s urf",
+      "▁su rf",
+      "▁sur f",
+      "ph ere",
+      "pher e",
+      "p here",
+      "D b",
+      "hm a",
+      "h ma",
+      "千 年",
+      "看 出",
+      "▁over come",
+      "▁re stricted",
+      "▁restr icted",
+      "▁restrict ed",
+      "▁ restricted",
+      "nW here",
+      "nWh ere",
+      "n Where",
+      "几 十",
+      "财产保险 股份有限公司",
+      "我们 在",
+      "我 们在",
+      "▁C orp",
+      "▁Co rp",
+      "▁Cor p",
+      "▁ Corp",
+      "刑期 自",
+      "诉讼 权利",
+      "Re port",
+      "Rep ort",
+      "Repo rt",
+      "▁v arying",
+      "▁var ying",
+      "▁vary ing",
+      "▁va rying",
+      "▁ varying",
+      "▁S hare",
+      "▁Sh are",
+      "▁Sha re",
+      "▁Shar e",
+      "▁ Share",
+      "▁crow d",
+      "▁cro wd",
+      "端 口",
+      "▁cop per",
+      "P ub",
+      "豆 腐",
+      "▁app eal",
+      "▁appe al",
+      "想 着",
+      "▁loss es",
+      "▁los ses",
+      "符 号",
+      "▁do ctors",
+      "▁doctor s",
+      "▁comp utation",
+      "▁comput ation",
+      "▁G h",
+      "om at",
+      "oma t",
+      "o mat",
+      "or tion",
+      "ort ion",
+      "ache lor",
+      "achel or",
+      "▁N ig",
+      "▁Ni g",
+      "us c",
+      "u sc",
+      "认真 遵守",
+      "▁ret irement",
+      "▁retire ment",
+      "be s",
+      "b es",
+      "La yer",
+      "L ayer",
+      "▁util ized",
+      "▁utilize d",
+      "en ant",
+      "ena nt",
+      "e nant",
+      "▁s quad",
+      "▁squ ad",
+      "re direct",
+      "red irect",
+      "对本 案",
+      "对 本案",
+      "Middle ware",
+      "M iddleware",
+      "词 汇",
+      "j upyter",
+      "br anch",
+      "bra nch",
+      "▁c v",
+      "▁ cv",
+      "已 构成",
+      "▁\" @",
+      "▁ \"@",
+      "公安 机关",
+      "内 外",
+      "ap tic",
+      "apt ic",
+      "a ptic",
+      "▁vict im",
+      "▁vic tim",
+      "ly n",
+      "l yn",
+      "爱 国",
+      "con vert",
+      "conv ert",
+      "▁special ist",
+      "▁B asic",
+      "▁Bas ic",
+      "▁ Basic",
+      "恐 怖",
+      "▁G ra",
+      "▁Gr a",
+      "▁Th read",
+      "▁Thr ead",
+      "▁ Thread",
+      "▁Mal ays",
+      "▁Malay s",
+      "激 发",
+      "法律 服务",
+      "▁mov ements",
+      "▁move ments",
+      "▁movement s",
+      "UT F",
+      "U TF",
+      "合理 的",
+      "▁mut ant",
+      "▁p aras",
+      "▁par as",
+      "▁para s",
+      "▁pa ras",
+      "▁Al gebra",
+      "▁ Algebra",
+      "▁Is lands",
+      "▁Island s",
+      "全 程",
+      "× （",
+      "ing ham",
+      "Pr ice",
+      "P rice",
+      "dr iv",
+      "d riv",
+      "LI NE",
+      "LIN E",
+      "L INE",
+      "tc ase",
+      "t case",
+      "▁decl ared",
+      "▁declar ed",
+      "▁declare d",
+      "前 景",
+      "女 士",
+      "方程 为",
+      "▁cert ificate",
+      "▁ certificate",
+      "▁R andom",
+      "▁Rand om",
+      "▁Ran dom",
+      "▁ Random",
+      "▁g ate",
+      "▁ga te",
+      "▁ gate",
+      "hi story",
+      "hist ory",
+      "h istory",
+      "F M",
+      "第三 章",
+      "第 三章",
+      "▁t ooth",
+      "▁to oth",
+      "▁too th",
+      "mem ory",
+      "memo ry",
+      "m emory",
+      "▁bo nus",
+      "▁bon us",
+      "▁Sw edish",
+      "▁Swed ish",
+      "▁G arden",
+      "▁Gar den",
+      "▁Gard en",
+      "已经发生法律效力 的",
+      "IL L",
+      "I LL",
+      "ot rop",
+      "otr op",
+      "▁m arine",
+      "▁mar ine",
+      "▁ma rine",
+      "▁ marine",
+      "re nd",
+      "ren d",
+      "r end",
+      "▁d emo",
+      "▁de mo",
+      "▁dem o",
+      "▁ demo",
+      "再次 申请执行",
+      "再次申请 执行",
+      "nG o",
+      "n Go",
+      "澳 门",
+      "▁t rim",
+      "▁tr im",
+      "▁tri m",
+      "▁ trim",
+      "ap ing",
+      "api ng",
+      "a ping",
+      "▁there by",
+      "▁vict ims",
+      "▁victim s",
+      "Return s",
+      "if act",
+      "ifa ct",
+      "i fact",
+      "▁f aced",
+      "▁fac ed",
+      "▁face d",
+      "▁fa ced",
+      "对 你",
+      "ap top",
+      "apt op",
+      "编 程",
+      "> )",
+      "什么 的",
+      "▁Emp ire",
+      "▁ent ering",
+      "▁enter ing",
+      "分 泌",
+      "▁redist ribute",
+      "De m",
+      "D em",
+      "▁diff iculty",
+      "▁difficult y",
+      "▁C L",
+      "▁ CL",
+      "▁A ge",
+      "▁Ag e",
+      "▁ Age",
+      "这 就",
+      "▁N ick",
+      "▁Nic k",
+      "▁Ni ck",
+      "▁ Nick",
+      "可能 性",
+      "▁gar age",
+      "▁ga rage",
+      "pa d",
+      "p ad",
+      "先 行",
+      "AG S",
+      "A GS",
+      "有两 个",
+      "有 两个",
+      "pi c",
+      "p ic",
+      "t z",
+      "▁qu ot",
+      "▁quo t",
+      "▁ quot",
+      "dist ance",
+      "di stance",
+      "d istance",
+      "▁p ricing",
+      "▁pr icing",
+      "▁pri cing",
+      "心 态",
+      "▁public ations",
+      "▁publication s",
+      "▁flex ibility",
+      "一 百",
+      "▁p leased",
+      "▁please d",
+      "▁ple ased",
+      "▁pleas ed",
+      "▁plea sed",
+      "▁trans actions",
+      "▁transaction s",
+      "▁ transactions",
+      "vol ume",
+      "nF urther",
+      "气 体",
+      "优秀 的",
+      "▁any more",
+      "▁I l",
+      "▁ Il",
+      "▁\" \"\"\\",
+      "▁\"\" \"\\",
+      "▁\"\"\" \\",
+      "▁ \"\"\"\\",
+      "po t",
+      "p ot",
+      "一 月",
+      "▁B BC",
+      "▁BB C",
+      "▁b ias",
+      "▁bi as",
+      "▁ bias",
+      "\\\\ ;",
+      "pa ir",
+      "p air",
+      "▁dom inant",
+      "Ha s",
+      "H as",
+      "内 核",
+      "▁vis its",
+      "▁visit s",
+      "/ #",
+      "as ant",
+      "asa nt",
+      "asan t",
+      "st ock",
+      "sto ck",
+      "df s",
+      "d fs",
+      "▁ens uring",
+      "nC alculate",
+      "n Calculate",
+      "公务 员",
+      "需要 的",
+      "Cr eat",
+      "Cre at",
+      "C reat",
+      "yt ics",
+      "ytic s",
+      "friend ly",
+      ": %",
+      "se ts",
+      "set s",
+      "s ets",
+      ";/ /",
+      "; //",
+      "Fr agment",
+      "middle ware",
+      "m iddleware",
+      "▁le ather",
+      "s b",
+      "att ice",
+      "atti ce",
+      "к а",
+      "nT rack",
+      "nTr ack",
+      "n Track",
+      "in ar",
+      "ina r",
+      "i nar",
+      "▁aff ecting",
+      "▁affect ing",
+      "▁ang ular",
+      "▁ angular",
+      "回 去",
+      "引 入",
+      "▁pro vision",
+      "▁prov ision",
+      "▁ provision",
+      "([ \\",
+      "( [\\",
+      "哪 怕",
+      "高 达",
+      "分 割",
+      "en ch",
+      "enc h",
+      "e nch",
+      "▁S ingle",
+      "▁Sing le",
+      "▁ Single",
+      "▁La TeX",
+      "▁ LaTeX",
+      "▁fe wer",
+      "▁few er",
+      "ject ion",
+      "je ction",
+      "j ection",
+      "ru s",
+      "r us",
+      "头 的",
+      "浏览 器",
+      "▁bl ind",
+      "▁ blind",
+      "std io",
+      "▁s uite",
+      "▁su ite",
+      "▁suit e",
+      "▁ suite",
+      "▁s kip",
+      "▁sk ip",
+      "▁ski p",
+      "▁ skip",
+      "▁F all",
+      "▁Fa ll",
+      "▁Fal l",
+      "▁ Fall",
+      "教 室",
+      "▁f alls",
+      "▁fall s",
+      "▁fal ls",
+      "▁ falls",
+      "是 怎么",
+      "bl ast",
+      "bla st",
+      "b last",
+      "激 烈",
+      "AR CH",
+      "ARC H",
+      "撤回 对被告",
+      "▁met rics",
+      "▁metric s",
+      "▁ metrics",
+      "ac ht",
+      "ach t",
+      "a cht",
+      "作为 一个",
+      "作 为一个",
+      "▁ox ide",
+      "▁oxid e",
+      "▁ oxide",
+      "▁fol ks",
+      "▁folk s",
+      "▁d ies",
+      "▁di es",
+      "▁die s",
+      "▁ dies",
+      "▁acc eleration",
+      "▁accel eration",
+      "▁acceler ation",
+      "售 后",
+      "IN K",
+      "I NK",
+      "fl at",
+      "f lat",
+      "Met adata",
+      "Meta data",
+      "变 更",
+      "▁b ulk",
+      "▁bul k",
+      "▁ bulk",
+      "▁spec imens",
+      "▁specimen s",
+      "▁phys ician",
+      "▁physic ian",
+      "▁L ind",
+      "▁Lin d",
+      "▁Li nd",
+      "出现 在",
+      "出 现在",
+      "提交上诉状 正本一份",
+      "▁capt ured",
+      "▁capture d",
+      "▁de leted",
+      "▁del eted",
+      "▁delete d",
+      "▁delet ed",
+      "▁ deleted",
+      "气 象",
+      "op athy",
+      "opath y",
+      "opa thy",
+      "o pathy",
+      "▁Re cord",
+      "▁Rec ord",
+      "▁ Record",
+      "е р",
+      "ol s",
+      "o ls",
+      "对 待",
+      "G M",
+      "▁S av",
+      "▁Sa v",
+      "le ar",
+      "lea r",
+      "l ear",
+      "▁dec ay",
+      "▁\\' \\'\\",
+      "▁\\'\\' \\",
+      "▁\\'\\ '\\",
+      "▁ \\'\\'\\",
+      "▁commun icate",
+      "▁communic ate",
+      "ce ptions",
+      "cept ions",
+      "ception s",
+      "▁c ake",
+      "▁ca ke",
+      "▁ cake",
+      "▁b are",
+      "▁bar e",
+      "▁ba re",
+      "op ts",
+      "opt s",
+      "o pts",
+      "▁B rad",
+      "▁Br ad",
+      "▁Bra d",
+      "▁I R",
+      "▁ IR",
+      "nP art",
+      "nPar t",
+      "n Part",
+      "▁s yntax",
+      "▁synt ax",
+      "▁syn tax",
+      "▁ syntax",
+      "▁re moving",
+      "▁rem oving",
+      "▁remov ing",
+      "浪 漫",
+      "中国 共产党",
+      "row th",
+      "for Each",
+      "外 出",
+      "▁S everal",
+      "▁Sever al",
+      "▁Sev eral",
+      "操作 系统",
+      "D to",
+      "V ar",
+      "var d",
+      "va rd",
+      "v ard",
+      "选 用",
+      "GR O",
+      "G RO",
+      "因 式",
+      "力 和",
+      "uu m",
+      "u um",
+      "▁Miss ouri",
+      "▁medal ists",
+      "▁pay load",
+      "▁ payload",
+      "坐 在",
+      "▁investig ations",
+      "▁investigation s",
+      "> &",
+      "分辨 率",
+      "ya l",
+      "y al",
+      "类型 的",
+      "类 型的",
+      "▁oper ational",
+      "▁operation al",
+      "conn ected",
+      "connect ed",
+      "焦 点",
+      "▁ex ponential",
+      "▁exponent ial",
+      "▁T a",
+      "▁ Ta",
+      "ir al",
+      "ira l",
+      "i ral",
+      "▁restr iction",
+      "▁restrict ion",
+      "i O",
+      "iz on",
+      "i zon",
+      "grad uate",
+      "奶 奶",
+      "sy m",
+      "s ym",
+      "▁Prop Types",
+      "▁ PropTypes",
+      "Da t",
+      "D at",
+      "▁f ra",
+      "▁fr a",
+      "▁ fra",
+      "▁in clusion",
+      "▁incl usion",
+      "安全 生产",
+      "▁ 朱",
+      "▁s park",
+      "▁sp ark",
+      "▁spa rk",
+      "▁ spark",
+      "▁C art",
+      "▁Car t",
+      "▁Ca rt",
+      "▁ Cart",
+      "prot otype",
+      "proto type",
+      "▁d ipl",
+      "▁di pl",
+      "▁dip l",
+      "sc opic",
+      "nQu estion",
+      "n Question",
+      "▁path ogenesis",
+      "▁pathogen esis",
+      "▁ 林",
+      "▁d escent",
+      "▁des cent",
+      "▁desc ent",
+      "▁s wit",
+      "▁sw it",
+      "出 门",
+      "lu x",
+      "l ux",
+      "qu er",
+      "que r",
+      "q uer",
+      "▁Vol ume",
+      "▁ Volume",
+      "疫情 防控",
+      "Sa ve",
+      "S ave",
+      "不 含",
+      "▁trad ition",
+      "▁tra dition",
+      "权 威",
+      "▁I tem",
+      "▁It em",
+      "▁ Item",
+      "▁Dis ney",
+      "我 现在",
+      "▁C orn",
+      "▁Co rn",
+      "▁Cor n",
+      "nb ool",
+      "n bool",
+      "管理 的",
+      "贵 的",
+      "▁re volution",
+      "▁rev olution",
+      "▁v o",
+      "▁ vo",
+      "▁P L",
+      "▁ PL",
+      "▁ 个",
+      "▁cont rolling",
+      "▁control ling",
+      "ac o",
+      "a co",
+      "h d",
+      "mer s",
+      "me rs",
+      "m ers",
+      "ar ma",
+      "arm a",
+      "▁sp inal",
+      "▁spin al",
+      "▁ spinal",
+      "数 为",
+      "Task s",
+      "T asks",
+      "S w",
+      "▁V o",
+      "▁ Vo",
+      "sh ops",
+      "shop s",
+      "▁ap plies",
+      "▁app lies",
+      "▁appl ies",
+      "▁Pr ince",
+      "▁Princ e",
+      "▁Pri nce",
+      "▁M oh",
+      "▁Mo h",
+      "er ving",
+      "erv ing",
+      "不 锈钢",
+      "字符 串",
+      "Dr aw",
+      "D raw",
+      "in ic",
+      "ini c",
+      "i nic",
+      "市 公安局",
+      "职 能",
+      "病 情",
+      "M W",
+      "▁a ug",
+      "▁au g",
+      "▁ aug",
+      "th ere",
+      "ther e",
+      "the re",
+      "t here",
+      "▁particip ating",
+      "▁reg ulate",
+      "▁regul ate",
+      "分别 是",
+      "Ki t",
+      "K it",
+      "附 带",
+      "ab c",
+      "a bc",
+      "du mp",
+      "d ump",
+      "nUn iversity",
+      "nUnivers ity",
+      "n University",
+      "▁phenomen on",
+      "US B",
+      "U SB",
+      "一样 的",
+      "一 样的",
+      "\"] ,\\",
+      "\"], \\",
+      "\" ],\\",
+      "ki sh",
+      "k ish",
+      "▁poly morph",
+      "▁polym orph",
+      "信息 均",
+      "ou ting",
+      "out ing",
+      "o uting",
+      "ne c",
+      "n ec",
+      "▁for got",
+      "▁forg ot",
+      "▁symbol s",
+      "▁ symbols",
+      "南 昌",
+      "back end",
+      "▁util izing",
+      "的 三",
+      "指 派",
+      "▁r ugby",
+      "▁rug by",
+      "志愿 者",
+      "房 型",
+      "日 电",
+      "▁p uts",
+      "▁put s",
+      "▁pu ts",
+      "▁ puts",
+      "▁b ottle",
+      "▁bo ttle",
+      "▁bott le",
+      "▁bot tle",
+      "nt r",
+      "n tr",
+      "F ull",
+      "轿 车",
+      "▁Act ive",
+      "▁Activ e",
+      "▁ Active",
+      "▁independent ly",
+      "▁independ ently",
+      "商 初字第",
+      "▁W ars",
+      "▁War s",
+      "▁Wa rs",
+      "▁dis able",
+      "▁ disable",
+      "一 批",
+      "▁M ove",
+      "▁Mo ve",
+      "▁ Move",
+      "▁P erson",
+      "▁Per son",
+      "▁Pers on",
+      "▁ Person",
+      "妹 妹",
+      "▁D ue",
+      "▁Du e",
+      "▁ Due",
+      "学校 的",
+      "▁ult r",
+      "▁ul tr",
+      "不 清",
+      "▁sc enes",
+      "▁scen es",
+      "▁scene s",
+      "▁ scenes",
+      "▁tim ing",
+      "▁ti ming",
+      "▁P HP",
+      "▁PH P",
+      "▁ PHP",
+      "授 予",
+      "▁s ight",
+      "▁si ght",
+      "▁sig ht",
+      "▁B ath",
+      "▁Ba th",
+      "▁Bat h",
+      "只 会",
+      "▁g astr",
+      "▁gas tr",
+      "▁ga str",
+      "▁gast r",
+      "pro ve",
+      "pr ove",
+      "prov e",
+      "p rove",
+      "▁I raq",
+      "▁Ir aq",
+      "▁t weet",
+      "▁twe et",
+      "▁ tweet",
+      "▁c urrency",
+      "▁curr ency",
+      "▁ currency",
+      "c u",
+      "▁M T",
+      "▁ MT",
+      "因 而",
+      "▁t rib",
+      "▁tr ib",
+      "▁tri b",
+      "▁bar rier",
+      "▁barr ier",
+      "成为 了",
+      "成 为了",
+      "▁obt aining",
+      "▁obtain ing",
+      "▁T al",
+      "▁Ta l",
+      "F ont",
+      "日 益",
+      "▁R ailway",
+      "▁Rail way",
+      "爆 炸",
+      "Pub lic",
+      "P ublic",
+      "洗 净",
+      "out heast",
+      "outh east",
+      "▁d uty",
+      "▁du ty",
+      "▁ duty",
+      "ho ver",
+      "h over",
+      "▁al umin",
+      "绘 画",
+      "采 集",
+      "▁trans cript",
+      "▁tra nscript",
+      "▁tran script",
+      "▁transc ript",
+      "na ls",
+      "nal s",
+      "n als",
+      "▁\\' \\'",
+      "▁\\'\\ '",
+      "▁ \\'\\'",
+      "cy cle",
+      "cycl e",
+      "c ycle",
+      "公诉机关 指控",
+      "▁circ ular",
+      "▁cir cular",
+      "▁trans f",
+      "▁tran sf",
+      "▁pro gressive",
+      "▁progress ive",
+      "▁cl othing",
+      "▁clot hing",
+      "▁cloth ing",
+      "出 行",
+      "当 初",
+      "终 身",
+      "▁t hemes",
+      "▁the mes",
+      "▁them es",
+      "▁theme s",
+      "▁ themes",
+      "nO h",
+      "n Oh",
+      "▁cook ies",
+      "▁cookie s",
+      "▁ cookies",
+      "不 易",
+      "人 和",
+      "本院 依据",
+      "合 计",
+      "▁P u",
+      "Po wer",
+      "P ower",
+      "bur n",
+      "bu rn",
+      "b urn",
+      "分 手",
+      "姓 名",
+      "▁me ets",
+      "▁meet s",
+      "文 本",
+      "▁c erebral",
+      "▁cere bral",
+      "to k",
+      "t ok",
+      "万 里",
+      "▁T el",
+      "▁Te l",
+      "▁ Tel",
+      "ma link",
+      "mal ink",
+      "▁associ ations",
+      "▁association s",
+      "▁b ars",
+      "▁bar s",
+      "▁ba rs",
+      "▁ bars",
+      "控 股",
+      "春 节",
+      "▁H ay",
+      "▁Ha y",
+      "▁satisf y",
+      "这么 多",
+      "d n",
+      "引 自",
+      "▁Ass oci",
+      "es ian",
+      "esi an",
+      "▁de mands",
+      "▁dem ands",
+      "▁demand s",
+      "科 大",
+      "▁R T",
+      "▁ RT",
+      "ip er",
+      "ipe r",
+      "i per",
+      "法 庭",
+      "▁t el",
+      "▁te l",
+      "▁ tel",
+      "▁vari ed",
+      "▁var ied",
+      "▁va ried",
+      "▁al k",
+      "▁ alk",
+      "R andom",
+      "p b",
+      "▁fr ames",
+      "▁frame s",
+      "▁fra mes",
+      "▁ frames",
+      "占地 面积",
+      "占 地面积",
+      "加拿 大",
+      "中文 化",
+      "中 文化",
+      "但 我",
+      "▁t we",
+      "▁tw e",
+      "z i",
+      "▁const ants",
+      "▁constant s",
+      "▁ constants",
+      "▁F L",
+      "▁ FL",
+      "牛 肉",
+      "▁K ansas",
+      "▁P i",
+      "▁ Pi",
+      "▁leg end",
+      "▁ legend",
+      "ill ance",
+      "illa nce",
+      "职 称",
+      "na uthor",
+      "n author",
+      "▁p regnant",
+      "▁pregn ant",
+      "▁part ition",
+      "▁ partition",
+      "▁accum ulation",
+      "▁sa uce",
+      "▁vac ation",
+      "▁va cation",
+      "IS O",
+      "I SO",
+      "at ted",
+      "att ed",
+      "▁g amma",
+      "▁gam ma",
+      "▁ga mma",
+      "▁ gamma",
+      "▁bre aks",
+      "▁break s",
+      "▁Fr ont",
+      "▁Fro nt",
+      "▁ Front",
+      "▁nav igation",
+      "▁ navigation",
+      "化 为",
+      "▁g rap",
+      "▁gr ap",
+      "▁gra p",
+      "}{ {",
+      "} {{",
+      "▁ 导演",
+      ")) :\\",
+      ") ):\\",
+      "▁ret riev",
+      "▁f acts",
+      "▁fact s",
+      "▁fac ts",
+      "▁fa cts",
+      "▁ facts",
+      "▁p ink",
+      "▁pi nk",
+      "▁pin k",
+      "▁ pink",
+      "}\" \\",
+      "} \"\\",
+      "一段 时间",
+      "一 段时间",
+      "no m",
+      "n om",
+      "可以 通过",
+      "ac cept",
+      "ol or",
+      "olo r",
+      "o lor",
+      "▁K at",
+      "▁Ka t",
+      "更好 地",
+      "date s",
+      "da tes",
+      "dat es",
+      "d ates",
+      "▁ap prox",
+      "▁app rox",
+      "▁appro x",
+      "▁ approx",
+      "Fe ature",
+      "▁con vergence",
+      "▁conver gence",
+      "▁converge nce",
+      "pan el",
+      "pa nel",
+      "p anel",
+      "▁con form",
+      "▁conf orm",
+      "▁document ed",
+      "ol ute",
+      "感 动",
+      "Mod al",
+      "Mo dal",
+      "M odal",
+      "▁Wil son",
+      "^ (",
+      "设 有",
+      "▁s ync",
+      "▁sy nc",
+      "▁syn c",
+      "▁ sync",
+      "▁high lights",
+      "▁highlight s",
+      "▁Fin ancial",
+      "▁t rou",
+      "▁tr ou",
+      "▁tro u",
+      "▁p ipeline",
+      "▁pip eline",
+      "▁pipe line",
+      "▁ pipeline",
+      "it coin",
+      "itc oin",
+      "▁des ktop",
+      "▁desk top",
+      "▁ desktop",
+      "em at",
+      "ema t",
+      "e mat",
+      "轨 道",
+      "op ter",
+      "opt er",
+      "ign ore",
+      "的 机会",
+      "看 过",
+      "Con f",
+      "Co nf",
+      "▁g ent",
+      "▁gen t",
+      "▁ge nt",
+      "▁ gent",
+      "询 问",
+      "▁d rain",
+      "▁dr ain",
+      "参加 了",
+      "参 加了",
+      "▁de part",
+      "▁dep art",
+      "旗 下",
+      "▁F uture",
+      "▁Fu ture",
+      "▁ Future",
+      "ex amples",
+      "example s",
+      "exam ples",
+      "地 板",
+      "本题 考查了",
+      "本题考查 了",
+      "办事 处",
+      "▁trans late",
+      "▁transl ate",
+      "▁ translate",
+      "habilit ation",
+      "ix els",
+      "ixel s",
+      "it ting",
+      "itt ing",
+      "FO RM",
+      "FOR M",
+      "F ORM",
+      "生效 之日起",
+      "魔 法",
+      "▁re leases",
+      "▁release s",
+      "▁ releases",
+      "▁H u",
+      "大 数据",
+      "▁C ook",
+      "▁Co ok",
+      "▁ Cook",
+      "▁a nch",
+      "▁an ch",
+      "▁anc h",
+      "▁ anch",
+      "▁ill egal",
+      "▁il legal",
+      "▁illeg al",
+      "▁comb inations",
+      "▁combin ations",
+      "▁combination s",
+      "▁A LL",
+      "▁AL L",
+      "▁ ALL",
+      "ent ral",
+      "▁el astic",
+      "▁elast ic",
+      "▁ elastic",
+      "▁S n",
+      "▁ Sn",
+      "▁p el",
+      "▁pe l",
+      "▁ pel",
+      "ol itan",
+      "olit an",
+      "oli tan",
+      "▁person ally",
+      "▁personal ly",
+      "▁vit amin",
+      "▁ch arts",
+      "▁char ts",
+      "▁chart s",
+      "▁ charts",
+      "到 位",
+      "CO L",
+      "C OL",
+      "▁h asn",
+      "▁has n",
+      "▁ha sn",
+      "市政 府",
+      "市 政府",
+      "LO W",
+      "L OW",
+      "口 腔",
+      "▁ 餐馆",
+      "发 明",
+      "▁E P",
+      "▁ EP",
+      "t g",
+      "▁cl othes",
+      "▁clot hes",
+      "▁cloth es",
+      "▁t ast",
+      "▁ta st",
+      "▁exec uted",
+      "▁execute d",
+      "▁execut ed",
+      "▁C C",
+      "▁ CC",
+      "男 朋友",
+      "防 水",
+      "批 评",
+      "凭 借",
+      "▁vert ices",
+      "▁ vertices",
+      "*\\ \\",
+      "* \\\\",
+      "▁m ood",
+      "▁mo od",
+      "▁enter prise",
+      "▁ enterprise",
+      "对 自己",
+      "上 是",
+      "se d",
+      "s ed",
+      "▁ad aptive",
+      "▁adapt ive",
+      "▁ adaptive",
+      "重 的",
+      "▁P T",
+      "▁ PT",
+      "诞 生",
+      "▁ →",
+      "Sup pose",
+      "ne tes",
+      "net es",
+      "▁ed iting",
+      "▁edit ing",
+      "▁A gain",
+      "▁Ag ain",
+      "▁C P",
+      "▁ CP",
+      "itud inal",
+      "ry pto",
+      "rypt o",
+      "▁w aters",
+      "▁water s",
+      "▁wa ters",
+      "▁wat ers",
+      "▁h ang",
+      "▁ha ng",
+      "▁han g",
+      "▁ hang",
+      "培 育",
+      "放 心",
+      "▁E ric",
+      "▁Er ic",
+      "▁ Eric",
+      "ennes see",
+      "IL D",
+      "I LD",
+      "ial s",
+      "ia ls",
+      "i als",
+      "▁x ml",
+      "▁ xml",
+      "二 首",
+      "in gly",
+      "ing ly",
+      "__ \\':\\",
+      "__\\ ':\\",
+      "▁res olved",
+      "▁resolve d",
+      "▁ resolved",
+      "排 序",
+      "opt ional",
+      "option al",
+      "}) ^{",
+      "})^ {",
+      "} )^{",
+      "ur ches",
+      "urch es",
+      "f x",
+      "cont ains",
+      "co very",
+      "cover y",
+      "cov ery",
+      "c overy",
+      "Ms g",
+      "M sg",
+      "事实 清楚",
+      "ot ive",
+      "oti ve",
+      "ab ad",
+      "aba d",
+      "a bad",
+      "th ur",
+      "可在接到 判决书的第二日起十日内",
+      "精 心",
+      "那就 是",
+      "那 就是",
+      "一 项",
+      "▁T ax",
+      "▁Ta x",
+      "▁ Tax",
+      "pe g",
+      "p eg",
+      "ult ural",
+      "▁m a",
+      "▁ ma",
+      "增加 了",
+      "增 加了",
+      "较大 的",
+      "较 大的",
+      "▁Con st",
+      "▁Co nst",
+      "▁Cons t",
+      "▁ Const",
+      "军 队",
+      "CA D",
+      "C AD",
+      "Tw o",
+      "T wo",
+      "ol o",
+      "o lo",
+      "ch rome",
+      "chr ome",
+      "chrom e",
+      "nw e",
+      "n we",
+      "▁appear ances",
+      "▁appearance s",
+      "eu r",
+      "e ur",
+      "▁C u",
+      "▁ Cu",
+      "入 口",
+      "新华 社",
+      "▁l ateral",
+      "▁later al",
+      "▁late ral",
+      "▁lat eral",
+      "at on",
+      "ato n",
+      "a ton",
+      "▁compar able",
+      "▁S port",
+      "▁Sp ort",
+      "▁ Sport",
+      "▁ 有",
+      "静 脉",
+      "▁N ations",
+      "▁Nation s",
+      "▁Nat ions",
+      "▁: )",
+      "▁ :)",
+      "住 院",
+      "▁neuro nal",
+      "▁neur onal",
+      "▁neuron al",
+      "进行 分析",
+      "▁re fers",
+      "▁ref ers",
+      "▁refer s",
+      "按 摩",
+      "st eps",
+      "ste ps",
+      "step s",
+      "▁c rack",
+      "▁cr ack",
+      "▁cra ck",
+      "带 有",
+      "bas ic",
+      "b asic",
+      "▁equ ally",
+      "▁equal ly",
+      "▁eq ually",
+      "诗 人",
+      "osc opic",
+      "o scopic",
+      "▁ 新",
+      "▁l ambda",
+      "▁lamb da",
+      "▁ lambda",
+      "▁Per formance",
+      "▁Perform ance",
+      "▁ Performance",
+      "▁Town ship",
+      "S B",
+      "一 部",
+      "县 城",
+      "▁c ry",
+      "▁cr y",
+      "du ration",
+      "d uration",
+      "游 泳",
+      "nt his",
+      "nth is",
+      "n this",
+      "文 档",
+      "词 语",
+      "合法 传唤",
+      "北 方",
+      "▁mus cles",
+      "▁muscle s",
+      "▁musc les",
+      "那么 多",
+      "▁s eeds",
+      "▁se eds",
+      "▁see ds",
+      "▁seed s",
+      ")) ,",
+      ") ),",
+      "▁accompan ied",
+      "ps ilon",
+      "psi lon",
+      "多 数",
+      "num s",
+      "nu ms",
+      "n ums",
+      "▁every where",
+      "水 中",
+      "ap sed",
+      "aps ed",
+      "apse d",
+      "▁sh ops",
+      "▁shop s",
+      "▁ shops",
+      "▁comp uters",
+      "▁comput ers",
+      "▁computer s",
+      "▁compute rs",
+      "▁Cap ital",
+      "ER E",
+      "E RE",
+      "▁s pont",
+      "▁sp ont",
+      "▁spo nt",
+      "常委 会",
+      "常 委会",
+      "▁s tain",
+      "▁st ain",
+      "▁dam aged",
+      "▁damage d",
+      "两 边",
+      "gh an",
+      "g han",
+      "发 电",
+      "aw are",
+      "awa re",
+      "a ware",
+      "▁sp ots",
+      "▁spot s",
+      "▁spo ts",
+      "▁ spots",
+      "▁S her",
+      "▁Sh er",
+      "▁She r",
+      "海 拔",
+      "▁c razy",
+      "▁cr azy",
+      "▁cra zy",
+      "▁A h",
+      "▁ Ah",
+      "▁district s",
+      "▁n otation",
+      "▁not ation",
+      "▁ notation",
+      "▁with draw",
+      "▁withd raw",
+      "是 你",
+      "br and",
+      "bra nd",
+      "b rand",
+      "in yl",
+      "iny l",
+      "▁W ould",
+      "▁ Would",
+      "场 的",
+      "am er",
+      "ame r",
+      "a mer",
+      "an ia",
+      "ani a",
+      "a nia",
+      "▁fl avor",
+      "▁flav or",
+      "▁anal ytical",
+      "▁analy tical",
+      "▁analyt ical",
+      "▁analytic al",
+      "获得 了",
+      "获 得了",
+      "经 纪",
+      "ect omy",
+      "e ctomy",
+      "ga te",
+      "g ate",
+      "登记 机关",
+      "▁Ste phen",
+      "▁Step hen",
+      "▁co aches",
+      "▁coach es",
+      "IB UT",
+      "I BUT",
+      "▁L imited",
+      "▁Lim ited",
+      "▁Limit ed",
+      "it ecture",
+      "itect ure",
+      "▁f unny",
+      "▁fun ny",
+      "▁p s",
+      "▁ ps",
+      "ex pand",
+      "exp and",
+      "▁f orth",
+      "▁for th",
+      "▁fort h",
+      "▁ forth",
+      "▁Vill age",
+      "▁Villa ge",
+      "st arts",
+      "start s",
+      "star ts",
+      "▁r aces",
+      "▁ra ces",
+      "▁race s",
+      "▁rac es",
+      "mer ge",
+      "▁cl osure",
+      "▁clos ure",
+      "▁ closure",
+      "qu ant",
+      "qua nt",
+      "q uant",
+      "为 止",
+      "\\' \"",
+      "发生 了",
+      "发 生了",
+      "▁st orm",
+      "▁sto rm",
+      "▁ storm",
+      "cont ents",
+      "content s",
+      "m L",
+      "为 例",
+      "▁I nsp",
+      "▁In sp",
+      "▁Ins p",
+      "▁ Insp",
+      "▁g ithub",
+      "▁git hub",
+      "▁ github",
+      "审批 表",
+      "审 批表",
+      "▁fund ed",
+      "▁fun ded",
+      "▁ funded",
+      "▁r u",
+      "▁ ru",
+      "▁re store",
+      "▁rest ore",
+      "▁restor e",
+      "▁ restore",
+      "▁ 金",
+      "某 种",
+      "}( \\\\",
+      "} (\\\\",
+      "ur ations",
+      "uration s",
+      "喜 爱",
+      "融 入",
+      "▁met adata",
+      "▁meta data",
+      "▁ metadata",
+      "op ed",
+      "ope d",
+      "o ped",
+      "数 额",
+      "增 大",
+      "感 兴趣",
+      "▁a ver",
+      "▁av er",
+      "▁ aver",
+      "▁un clear",
+      "▁unc lear",
+      "▁uncle ar",
+      "二 月",
+      "整 治",
+      "爆 发",
+      "▁D ise",
+      "▁Dis e",
+      "▁Di se",
+      "cd ots",
+      "cdot s",
+      "c dots",
+      "ibli ography",
+      "ibl iography",
+      "Me mber",
+      "Mem ber",
+      "M ember",
+      "用 药",
+      "好 多",
+      "yn chron",
+      "ynch ron",
+      "终 字第",
+      "进出 口",
+      "进 出口",
+      "{\\ '",
+      "{ \\'",
+      "创建 通道",
+      "sym bol",
+      "s ymbol",
+      "词条 初始",
+      "om ing",
+      "omin g",
+      "omi ng",
+      "o ming",
+      "信息均 引自",
+      "词条初始 信息均引自",
+      "ac co",
+      "acc o",
+      "▁b unch",
+      "▁bu nch",
+      "企业 词条",
+      "企业词条 极速",
+      "企业词条极速 创建通道",
+      "百度百科 企业词条极速创建通道",
+      "▁S olutions",
+      "▁Solution s",
+      "▁Sol utions",
+      "▁h ide",
+      "▁hi de",
+      "▁ hide",
+      "▁d estr",
+      "▁de str",
+      "▁des tr",
+      "▁dest r",
+      "▁Current ly",
+      "▁ Currently",
+      "▁fe eding",
+      "▁feed ing",
+      "▁fee ding",
+      "▁ feeding",
+      "▁s lot",
+      "▁sl ot",
+      "▁ slot",
+      "▁n eon",
+      "▁ne on",
+      "▁neo n",
+      "▁ass ignment",
+      "▁assign ment",
+      "▁ assignment",
+      "▁sub mission",
+      "▁ submission",
+      "uber netes",
+      "comp any",
+      "ad ies",
+      "adi es",
+      "a dies",
+      "▁surpr ised",
+      "▁surprise d",
+      "}} {\\\\",
+      "}}{ \\\\",
+      "} }{\\\\",
+      "▁A mb",
+      "▁Am b",
+      "▁ Amb",
+      "▁household s",
+      "两 位",
+      "▁V eg",
+      "▁Ve g",
+      "Ma y",
+      "M ay",
+      "Di ct",
+      "D ict",
+      "的 天",
+      "蛋 糕",
+      "ne rg",
+      "ner g",
+      "n erg",
+      "赶 紧",
+      "nS pec",
+      "nSp ec",
+      "nSpe c",
+      "n Spec",
+      "▁t owns",
+      "▁town s",
+      "▁tow ns",
+      "的 信息",
+      "▁g ay",
+      "▁ga y",
+      "ig nKey",
+      "ign Key",
+      "> ）",
+      "▁Sc ient",
+      "▁Sci ent",
+      "oc l",
+      "o cl",
+      "▁m yel",
+      "▁my el",
+      "is ons",
+      "ison s",
+      "iso ns",
+      "▁a nth",
+      "▁an th",
+      "▁ant h",
+      "▁ anth",
+      "he rence",
+      "her ence",
+      "here nce",
+      "逮 捕",
+      "PC R",
+      "P CR",
+      "▁comp ete",
+      "▁compet e",
+      "工商 行政",
+      "C a",
+      "▁Phil adelphia",
+      "▁F orum",
+      "▁For um",
+      "▁Fo rum",
+      "形成 了",
+      "形 成了",
+      "▁Alex ander",
+      "▁Alexand er",
+      "▁Alexa nder",
+      "_ (",
+      "ace ut",
+      "an ted",
+      "ant ed",
+      "ante d",
+      "▁c ycles",
+      "▁cy cles",
+      "▁cycl es",
+      "▁cycle s",
+      "▁ cycles",
+      "▁w ash",
+      "▁was h",
+      "▁wa sh",
+      "▁ wash",
+      "▁equ ilibrium",
+      "\\\\ /",
+      "▁inhib ited",
+      "▁inhibit ed",
+      "▁accur ately",
+      "▁accurate ly",
+      "▁C os",
+      "▁Co s",
+      "▁ Cos",
+      ":* *",
+      ": **",
+      "▁log ger",
+      "▁ logger",
+      "▁H al",
+      "▁Ha l",
+      "陶 瓷",
+      "ir ts",
+      "irt s",
+      "财 经",
+      "sk ip",
+      "ski p",
+      "s kip",
+      "▁F low",
+      "▁Fl ow",
+      "▁Flo w",
+      "▁ Flow",
+      "▁wh ilst",
+      "病 的",
+      "as ia",
+      "asi a",
+      "厚 的",
+      "社会 抚养费",
+      "▁北 京",
+      "▁ 北京",
+      "▁r ising",
+      "▁ris ing",
+      "▁ri sing",
+      "法 人",
+      "可以 说",
+      "纪 委",
+      "隐 藏",
+      "▁f est",
+      "▁fe st",
+      "▁ fest",
+      "LE V",
+      "L EV",
+      "看到 了",
+      "看 到了",
+      "▁p ulse",
+      "▁pul se",
+      "▁puls e",
+      "▁ pulse",
+      "nt est",
+      "nte st",
+      "n test",
+      "时 许",
+      "▁S tone",
+      "▁St one",
+      "最 为",
+      "▁fr actions",
+      "▁fraction s",
+      "▁fract ions",
+      "爱 你",
+      "在 外",
+      "▁sh arp",
+      "▁sha rp",
+      "▁shar p",
+      "▁ sharp",
+      "nB uild",
+      "n Build",
+      "co d",
+      "c od",
+      "残 疾",
+      "▁Over all",
+      "书 籍",
+      "精 确",
+      "▁p rize",
+      "▁pr ize",
+      "▁pri ze",
+      "▁priz e",
+      "高 压",
+      "▁s oul",
+      "▁so ul",
+      "▁sou l",
+      "▁t ou",
+      "▁to u",
+      "Fl ow",
+      "F low",
+      "宝 贝",
+      "▁rec ipient",
+      "▁recip ient",
+      "▁ recipient",
+      "▁un iversal",
+      "▁univers al",
+      "▁uni versal",
+      "大 气",
+      "十一 日",
+      "十 一日",
+      "▁t agged",
+      "▁tag ged",
+      "▁ tagged",
+      "▁En ron",
+      "▁ Enron",
+      "nD oes",
+      "nDo es",
+      "n Does",
+      "中 山",
+      "as sign",
+      "ass ign",
+      "ni a",
+      "n ia",
+      "▁Res ult",
+      "▁ Result",
+      "▁h opes",
+      "▁hope s",
+      "▁hop es",
+      "▁ho pes",
+      "责令 被执行人",
+      "▁expl ored",
+      "▁explore d",
+      "▁explo red",
+      "▁explor ed",
+      "气 质",
+      "▁a tom",
+      "▁at om",
+      "▁ atom",
+      "▁ 歌曲",
+      "ru les",
+      "rule s",
+      "r ules",
+      "ure au",
+      "Fun c",
+      "F unc",
+      "▁B rian",
+      "▁Br ian",
+      "▁Bri an",
+      "uit y",
+      "ui ty",
+      "u ity",
+      "▁Di ego",
+      "▁Die go",
+      "pr ime",
+      "pri me",
+      "prim e",
+      "安 县",
+      "▁Pro fessional",
+      "▁Profess ional",
+      "▁Profession al",
+      "法 则",
+      "州 区",
+      "替 代",
+      "ik ing",
+      "iki ng",
+      "i king",
+      "▁ 男",
+      "▁comp ile",
+      "▁ compile",
+      "▁int ention",
+      "▁intent ion",
+      "▁lo ans",
+      "▁loan s",
+      "动 漫",
+      "▁satisf ied",
+      "发 射",
+      "被 人",
+      "身 为",
+      "op ro",
+      "o pro",
+      "▁ind ices",
+      "▁indic es",
+      "▁ indices",
+      "▁M iller",
+      "▁Mil ler",
+      "▁Mi ller",
+      "▁Mill er",
+      "]] \\",
+      "] ]\\",
+      "▁B attle",
+      "▁Ba ttle",
+      "▁Bat tle",
+      "▁Batt le",
+      "se a",
+      "s ea",
+      "▁init iative",
+      "▁initi ative",
+      "in ery",
+      "ine ry",
+      "iner y",
+      "▁d iver",
+      "▁di ver",
+      "▁div er",
+      "▁dive r",
+      "▁p neum",
+      "服务 的",
+      "X ml",
+      "造成 的",
+      "造 成的",
+      "▁El izabeth",
+      "task s",
+      "tas ks",
+      "t asks",
+      "▁L iving",
+      "▁Li ving",
+      "▁Liv ing",
+      "入 手",
+      "▁special ized",
+      "▁specialize d",
+      "le m",
+      "l em",
+      "▁de form",
+      "▁def orm",
+      "上 网",
+      "▁Com par",
+      "▁Comp ar",
+      "▁ Compar",
+      "敌 人",
+      "▁G allery",
+      "▁Gall ery",
+      "▁ Gallery",
+      "讨 厌",
+      "▁ir re",
+      "▁irr e",
+      "下来 的",
+      "下 来的",
+      "▁l ogs",
+      "▁log s",
+      "▁lo gs",
+      "▁ logs",
+      "▁F un",
+      "▁Fu n",
+      "▁ Fun",
+      "偶 尔",
+      "ob ic",
+      "obi c",
+      "o bic",
+      "▁Return s",
+      "▁ Returns",
+      "ST AT",
+      "STA T",
+      "ill iant",
+      "illi ant",
+      "illian t",
+      "▁v enue",
+      "▁ven ue",
+      "▁ venue",
+      "▁ 酒店",
+      "▁l ang",
+      "▁la ng",
+      "▁lan g",
+      "▁ lang",
+      "▁H o",
+      "▁ Ho",
+      "re ement",
+      "ree ment",
+      "未 知",
+      "留 言",
+      "▁s hed",
+      "▁sh ed",
+      "▁she d",
+      "ch anges",
+      "chan ges",
+      "change s",
+      "chang es",
+      "De tail",
+      "Det ail",
+      "lim ited",
+      "limit ed",
+      "l imited",
+      "▁C zech",
+      "▁Cz ech",
+      "ov o",
+      "o vo",
+      "相同 的",
+      "不 及",
+      "相关 部门",
+      "▁S ort",
+      "▁So rt",
+      "▁Sor t",
+      "▁ Sort",
+      "康 复",
+      "em bed",
+      "emb ed",
+      "e mbed",
+      "▁implement ing",
+      "的 距离",
+      "第二百六 十二条第二款",
+      "▁Un ity",
+      "▁Unit y",
+      "▁Uni ty",
+      "▁ Unity",
+      "mo unt",
+      "m ount",
+      "▁trans plantation",
+      "▁transplant ation",
+      "io tic",
+      "iot ic",
+      "i otic",
+      "rem ote",
+      "op sy",
+      "ops y",
+      "o psy",
+      "▁* )",
+      "▁ *)",
+      "不 对",
+      "us age",
+      "usa ge",
+      "以 此",
+      "}} }",
+      "} }}",
+      "on ing",
+      "oni ng",
+      "o ning",
+      "而 成",
+      "▁St ory",
+      "▁ Story",
+      "▁k nee",
+      "▁kn ee",
+      "▁kne e",
+      "▁ex patriate",
+      "▁expat riate",
+      "▁ch olesterol",
+      "的第 一",
+      "的 第一",
+      "AS H",
+      "A SH",
+      "一方 面",
+      "一 方面",
+      "▁sh apes",
+      "▁shape s",
+      "▁sha pes",
+      "▁ shapes",
+      "的 目的",
+      "▁att ending",
+      "▁attend ing",
+      "▁atten ding",
+      "假 释",
+      "os ity",
+      "osit y",
+      "osi ty",
+      "de scribe",
+      "des cribe",
+      "desc ribe",
+      "prot ocol",
+      "proto col",
+      "min ute",
+      "▁P roduction",
+      "▁Pro duction",
+      "▁Product ion",
+      "▁Produ ction",
+      "▁ Production",
+      "亚 洲",
+      "奇 怪",
+      "ex tension",
+      "ext ension",
+      "▁consist ently",
+      "▁consistent ly",
+      "▁V ery",
+      "▁Ver y",
+      "▁Ve ry",
+      "▁ Very",
+      "乡 镇",
+      "豪 华",
+      "et e",
+      "e te",
+      "▁t one",
+      "▁to ne",
+      "▁ton e",
+      "▁ tone",
+      "▁s pher",
+      "▁sp her",
+      "▁sph er",
+      "▁adequ ate",
+      "▁P it",
+      "▁Pi t",
+      "co in",
+      "c oin",
+      "▁vol umes",
+      "▁volume s",
+      "▁ volumes",
+      "▁in fect",
+      "▁inf ect",
+      "▁ infect",
+      "▁in vari",
+      "▁inv ari",
+      "Vari able",
+      "Var iable",
+      "▁vess els",
+      "▁vessel s",
+      "想 想",
+      "▁R yan",
+      "▁Ry an",
+      "RE CT",
+      "REC T",
+      "R ECT",
+      "▁c hemistry",
+      "▁chem istry",
+      "▁n ick",
+      "▁nic k",
+      "▁ni ck",
+      "▁ nick",
+      "节 约",
+      "ple tion",
+      "plet ion",
+      "p letion",
+      "▁Pol ish",
+      "▁Po lish",
+      "W ord",
+      "]) )\\",
+      "])) \\",
+      "] ))\\",
+      "▁t ile",
+      "▁til e",
+      "▁ti le",
+      "▁ tile",
+      "ow ers",
+      "ower s",
+      "owe rs",
+      "o wers",
+      "条件 下",
+      "B F",
+      "▁l as",
+      "▁la s",
+      "▁ las",
+      "白 云",
+      "▁design ing",
+      "立 体",
+      "/ {",
+      "ne st",
+      "nes t",
+      "n est",
+      "配 有",
+      "陈 某",
+      "移 民",
+      "内蒙古 自治区",
+      "▁tem plates",
+      "▁template s",
+      "▁ templates",
+      "▁w ing",
+      "▁win g",
+      "▁wi ng",
+      "▁ wing",
+      "▁s poke",
+      "▁sp oke",
+      "▁spo ke",
+      "▁A dam",
+      "▁Ad am",
+      "▁Ada m",
+      "▁ Adam",
+      "▁t ennis",
+      "▁ten nis",
+      "▁B UT",
+      "▁BU T",
+      "▁ BUT",
+      "▁T ool",
+      "▁To ol",
+      "▁Too l",
+      "▁ Tool",
+      "▁ β",
+      "um bn",
+      "umb n",
+      "▁ д",
+      "将 于",
+      "▁equ als",
+      "▁equal s",
+      "▁ equals",
+      "man agement",
+      "manage ment",
+      "m anagement",
+      "[ (",
+      "村 委会",
+      "▁uncertain ty",
+      "of fs",
+      "off s",
+      "IN FO",
+      "INF O",
+      "▁disc ussions",
+      "▁discuss ions",
+      "▁discussion s",
+      "第四 十",
+      "第 四十",
+      "ub ble",
+      "为 一",
+      "进 攻",
+      "▁ex cit",
+      "▁exc it",
+      "▁organ isation",
+      "一次 性",
+      "aq ue",
+      "aqu e",
+      "a que",
+      "ok u",
+      "o ku",
+      ">> \\",
+      "> >\\",
+      "So me",
+      "S ome",
+      "▁Lew is",
+      "i w",
+      "Se ction",
+      "Sec tion",
+      "S ection",
+      "▁st yled",
+      "▁style d",
+      "▁styl ed",
+      "▁ styled",
+      "path s",
+      "pat hs",
+      "pa ths",
+      "p aths",
+      "b matrix",
+      "客 服",
+      "▁product ivity",
+      "坚 定",
+      "▁im per",
+      "▁imp er",
+      "▁impe r",
+      "向原告 借款",
+      "mem bers",
+      "member s",
+      "m embers",
+      "et he",
+      "eth e",
+      "e the",
+      "占 有",
+      "tr l",
+      "t rl",
+      "com pat",
+      "comp at",
+      "Off set",
+      "cy cl",
+      "c ycl",
+      "ul er",
+      "ule r",
+      "u ler",
+      "ar th",
+      "art h",
+      "▁S ong",
+      "▁So ng",
+      "▁Son g",
+      "▁ Song",
+      "二 手",
+      "▁B uff",
+      "▁Bu ff",
+      "▁ Buff",
+      "▁re plic",
+      "▁rep lic",
+      "完成 了",
+      "完 成了",
+      "▁requ iring",
+      "▁car rier",
+      "发出 执行通知书",
+      "▁We bsite",
+      "▁Web site",
+      "▁ Website",
+      "▁bur den",
+      "▁phys icians",
+      "▁physician s",
+      "▁physic ians",
+      "▁r ings",
+      "▁ring s",
+      "▁ rings",
+      "▁del ayed",
+      "▁delay ed",
+      "组 件",
+      "语 音",
+      "随 意",
+      "A x",
+      "的 安全",
+      "▁sc rap",
+      "▁scr ap",
+      "▁ scrap",
+      "▁mir ror",
+      "▁mi rror",
+      "▁ mirror",
+      "▁h ex",
+      "▁he x",
+      "▁ hex",
+      "▁E arly",
+      "▁Ear ly",
+      "▁Earl y",
+      "▁ Early",
+      "he rit",
+      "her it",
+      "in ction",
+      "inc tion",
+      "inct ion",
+      "是 谁",
+      "Trans action",
+      "变成 了",
+      "变 成了",
+      "nH ave",
+      "nHa ve",
+      "n Have",
+      "售 价",
+      "风 雨",
+      "▁c lip",
+      "▁cl ip",
+      "▁cli p",
+      "▁ clip",
+      "快 捷",
+      "chn ology",
+      "▁at oms",
+      "▁atom s",
+      "▁ atoms",
+      "▁T ech",
+      "▁Te ch",
+      "▁ Tech",
+      "的 美",
+      "▁di etary",
+      "▁diet ary",
+      "首 发",
+      "▁D rive",
+      "▁Dr ive",
+      "▁ Drive",
+      "en ergy",
+      "ener gy",
+      "e nergy",
+      "am ination",
+      "amin ation",
+      "ami nation",
+      "ia c",
+      "i ac",
+      "加 载",
+      "颜 值",
+      "▁conn ecting",
+      "▁connect ing",
+      "▁consult ation",
+      "ah oma",
+      "aho ma",
+      "▁in ventory",
+      "▁invent ory",
+      "▁inventor y",
+      "▁ inventory",
+      "nO f",
+      "n Of",
+      "Q Q",
+      "▁develop ments",
+      "▁development s",
+      "列 表",
+      "▁L as",
+      "▁La s",
+      "nr eturn",
+      "nre turn",
+      "n return",
+      "ph ones",
+      "phone s",
+      "phon es",
+      "酒 精",
+      "▁clin ically",
+      "▁clinical ly",
+      "▁clinic ally",
+      "▁ 胡",
+      "▁( {",
+      "▁ ({",
+      "▁t reating",
+      "▁treat ing",
+      "▁tre ating",
+      "ab ama",
+      "aba ma",
+      "em et",
+      "eme t",
+      "e met",
+      "▁st ere",
+      "▁ste re",
+      "▁ster e",
+      "▁ stere",
+      "▁p un",
+      "▁pu n",
+      "ellect ual",
+      "▁t axes",
+      "▁tax es",
+      "可 惜",
+      "罗 马",
+      "犯 危险驾驶罪",
+      "全 身",
+      "然后 再",
+      "然 后再",
+      "NI NG",
+      "N ING",
+      "或者 直接向",
+      "Down load",
+      "D ownload",
+      "ams ung",
+      "党 组",
+      "la d",
+      "l ad",
+      "▁eng ines",
+      "▁engine s",
+      "扎 实",
+      "Ident ifier",
+      "付 费",
+      "▁fa tty",
+      "▁fat ty",
+      "▁G as",
+      "▁Ga s",
+      "▁ Gas",
+      "▁ent rance",
+      "▁entr ance",
+      "▁colle agues",
+      "▁colleague s",
+      "第四 章",
+      "第 四章",
+      "▁co at",
+      "rupt ed",
+      "rup ted",
+      "▁ 注",
+      "aceut ical",
+      "结 局",
+      "争 取",
+      "制 的",
+      "的一 切",
+      "的 一切",
+      "pat ic",
+      "pa tic",
+      "p atic",
+      "主 流",
+      "▁pan creat",
+      "▁panc reat",
+      "ur ope",
+      "uro pe",
+      "▁Ex change",
+      "▁ Exchange",
+      "L ibrary",
+      "Reg ister",
+      "第二 次",
+      "第 二次",
+      "ess ment",
+      "Po p",
+      "P op",
+      "厚 度",
+      "▁C it",
+      "▁Ci t",
+      "Valid ator",
+      "gn u",
+      "g nu",
+      "像 素",
+      "uz zy",
+      "uzz y",
+      "条件 的",
+      "条 件的",
+      "air e",
+      "ai re",
+      "a ire",
+      "▁d isp",
+      "▁dis p",
+      "▁di sp",
+      "▁ disp",
+      "▁o bst",
+      "▁ob st",
+      "▁obs t",
+      "▁U SE",
+      "▁US E",
+      "▁ USE",
+      "▁bel ieves",
+      "▁believe s",
+      "▁belie ves",
+      "盈 利",
+      "Par ent",
+      "Pa rent",
+      "P arent",
+      "缺 陷",
+      "▁tra uma",
+      "nPop ulated",
+      "以 被告人",
+      "Re move",
+      "Rem ove",
+      "创 办",
+      "es sed",
+      "ess ed",
+      "esse d",
+      "▁depend ency",
+      "▁ dependency",
+      "can cel",
+      "c ancel",
+      "▁dig its",
+      "▁digit s",
+      "▁ digits",
+      "ll ib",
+      "lli b",
+      "l lib",
+      "F E",
+      "Dis t",
+      "Di st",
+      "D ist",
+      "▁y ields",
+      "▁yield s",
+      "一 把",
+      "▁or bit",
+      "▁ orbit",
+      "的 精神",
+      "lo cale",
+      "loc ale",
+      "local e",
+      "人 心",
+      "nR em",
+      "nRe m",
+      "n Rem",
+      "Res ults",
+      "Result s",
+      "一 看",
+      "▁s event",
+      "▁se vent",
+      "▁seven t",
+      "抵 押",
+      "日向本院提起 公诉",
+      "▁cl icking",
+      "▁click ing",
+      "▁b roke",
+      "▁br oke",
+      "▁bro ke",
+      "▁me tres",
+      "▁met res",
+      "▁metre s",
+      "{} \\",
+      "{ }\\",
+      "om orphic",
+      "omorph ic",
+      "▁inc orporated",
+      "▁incorpor ated",
+      "▁incorporate d",
+      "▁ incorporated",
+      "▁m esh",
+      "▁me sh",
+      "▁mes h",
+      "▁ mesh",
+      "▁s occer",
+      "▁soc cer",
+      "▁c nt",
+      "▁cn t",
+      "▁ cnt",
+      "劳动 任务",
+      "KE N",
+      "K EN",
+      "▁min eral",
+      "▁mine ral",
+      "es tone",
+      "est one",
+      "esto ne",
+      "eston e",
+      "e stone",
+      "▁M exican",
+      "▁Mex ican",
+      "机 遇",
+      "▁t mp",
+      "▁tm p",
+      "▁ tmp",
+      "pro of",
+      "裁 决",
+      "ka n",
+      "k an",
+      "自行 车",
+      "自 行车",
+      "爷 爷",
+      "▁a qu",
+      "▁ aqu",
+      "oy s",
+      "o ys",
+      "▁p ill",
+      "▁pi ll",
+      "▁pil l",
+      "上 课",
+      "不 论",
+      "▁M u",
+      "▁ Mu",
+      "▁sc hemes",
+      "▁scheme s",
+      "Ex tension",
+      "Ext ension",
+      "cript or",
+      "▁O ak",
+      "▁Com plex",
+      "▁Comp lex",
+      "▁Compl ex",
+      "▁ Complex",
+      "# [",
+      "е т",
+      "}< /",
+      "} </",
+      "▁h ate",
+      "▁ha te",
+      "▁hat e",
+      "ab ric",
+      "abr ic",
+      "▁Mary land",
+      "f g",
+      "▁f ung",
+      "▁fun g",
+      "▁fu ng",
+      "▁re serve",
+      "▁res erve",
+      "▁reserv e",
+      "▁occ urring",
+      "▁occur ring",
+      "▁s uffer",
+      "▁suff er",
+      "别 墅",
+      "nP r",
+      "n Pr",
+      "可 持续",
+      "热 水",
+      "ug ging",
+      "ugg ing",
+      "▁me lan",
+      "▁mel an",
+      "▁f its",
+      "▁fit s",
+      "▁fi ts",
+      "▁ fits",
+      "▁re mind",
+      "▁rem ind",
+      "▁remin d",
+      "发 票",
+      "▁S oviet",
+      "就 好",
+      "无 奈",
+      "分 的",
+      "菜 单",
+      "▁pl ug",
+      "▁ plug",
+      "▁g ifts",
+      "▁gift s",
+      "▁gi fts",
+      "▁my ocardial",
+      "▁myocard ial",
+      "wh o",
+      "w ho",
+      "▁adv ances",
+      "▁advance s",
+      "样 本",
+      "▁ph ones",
+      "▁phone s",
+      "▁phon es",
+      "▁ phones",
+      "提起 诉讼",
+      "▁post operative",
+      "▁dif fusion",
+      "▁diff usion",
+      "▁ diffusion",
+      "能力 的",
+      "能 力的",
+      "product s",
+      "produ cts",
+      "热 爱",
+      "▁con jug",
+      "▁conj ug",
+      "店 铺",
+      "▁comp ensation",
+      "▁compens ation",
+      "是否 有",
+      "职 员",
+      "li mits",
+      "lim its",
+      "limit s",
+      "l imits",
+      "▁r anges",
+      "▁range s",
+      "▁ran ges",
+      "▁d y",
+      "▁ dy",
+      "nC urrent",
+      "n Current",
+      "IP T",
+      "I PT",
+      "Fr ee",
+      "Fre e",
+      "F ree",
+      "知识 产权",
+      "▁Reg ional",
+      "▁Region al",
+      "AA AA",
+      "AAA A",
+      "A AAA",
+      "ex pl",
+      "exp l",
+      "▁account ing",
+      "▁D ark",
+      "▁Dar k",
+      "▁Da rk",
+      "▁ Dark",
+      "ri ble",
+      "rib le",
+      "r ible",
+      "ig ens",
+      "igen s",
+      "ige ns",
+      "i gens",
+      "vent ory",
+      "的 环境",
+      "南 北",
+      "nRes ults",
+      "nResult s",
+      "n Results",
+      "al las",
+      "all as",
+      "alla s",
+      "re sp",
+      "res p",
+      "r esp",
+      "nE ach",
+      "n Each",
+      "赛 事",
+      "一定 是",
+      "sc al",
+      "s cal",
+      "▁imp acts",
+      "▁impact s",
+      "通过本院 或者直接向",
+      "▁ple asure",
+      "▁pleas ure",
+      "▁spectro scopy",
+      "uzz le",
+      "u zzle",
+      "组织 机构",
+      "▁\\\\ ,",
+      "▁ \\\\,",
+      "倍 数",
+      "▁R ail",
+      "▁Ra il",
+      "▁expl oring",
+      "▁explo ring",
+      "▁explor ing",
+      "帝 国",
+      "格 局",
+      "同时 也",
+      "be at",
+      "b eat",
+      "he ll",
+      "hel l",
+      "h ell",
+      "▁sh orter",
+      "▁short er",
+      "nC heck",
+      "nCh eck",
+      "nChe ck",
+      "n Check",
+      "▁extract ed",
+      "▁extr acted",
+      "▁extrac ted",
+      "io p",
+      "i op",
+      "t assert",
+      "▁T erm",
+      "▁Te rm",
+      "▁Ter m",
+      "▁ Term",
+      "为 终审",
+      "en do",
+      "end o",
+      "e ndo",
+      "▁S hip",
+      "▁Sh ip",
+      "▁Shi p",
+      "▁ Ship",
+      "咱 们",
+      "▁re nov",
+      "▁pract ition",
+      "nS ec",
+      "nSe c",
+      "n Sec",
+      "之类 的",
+      "之 类的",
+      "cr oss",
+      "cro ss",
+      "c ross",
+      "减 轻",
+      "▁prem ier",
+      "因 数",
+      "可以 用",
+      "▁acc ord",
+      "▁ac cord",
+      "mon s",
+      "mo ns",
+      "m ons",
+      "依 次",
+      "楼 盘",
+      "教育 的",
+      "▁S ky",
+      "▁Sk y",
+      "▁ Sky",
+      "/ %",
+      "is o",
+      "i so",
+      "▁E ss",
+      "▁Es s",
+      "▁ Ess",
+      "的 老",
+      "▁o vers",
+      "▁over s",
+      "▁ov ers",
+      "▁ overs",
+      "海 军",
+      "记 住",
+      "按 钮",
+      "MB ER",
+      "M BER",
+      "▁ex clus",
+      "▁re cycl",
+      "▁rec ycl",
+      "▁br anches",
+      "▁branch es",
+      "▁ branches",
+      "st ers",
+      "ster s",
+      "ste rs",
+      "s ters",
+      "没 人",
+      "pr agma",
+      "▁heav ily",
+      "ag u",
+      "a gu",
+      "第二 章",
+      "第 二章",
+      "▁res ident",
+      "▁resid ent",
+      "▁reside nt",
+      "肺 炎",
+      "当 你",
+      "/ _",
+      "怀 孕",
+      "活 力",
+      "▁Mar sh",
+      "▁Mars h",
+      "▁i dx",
+      "▁id x",
+      "▁ idx",
+      "作 文",
+      "▁P aper",
+      "▁Pa per",
+      "▁Pap er",
+      "▁ Paper",
+      "▁Mel bourne",
+      "搭 建",
+      "笔 者",
+      "▁poss ibilities",
+      "▁Q t",
+      "▁ Qt",
+      "月二十 日",
+      "第七 十八条",
+      "第七十八 条",
+      "最高人民法院 关于人民法院",
+      "最高人民法院关于 人民法院",
+      "de velopment",
+      "develop ment",
+      "早 餐",
+      "You r",
+      "Y our",
+      "仿 真",
+      "▁ro ads",
+      "▁road s",
+      "▁ roads",
+      "▁present ing",
+      "nst ruct",
+      "nstr uct",
+      "n struct",
+      "mon t",
+      "mo nt",
+      "m ont",
+      "外 交",
+      "▁p a",
+      "▁ pa",
+      "什么 时候",
+      "▁ass umption",
+      "▁assum ption",
+      "连 锁",
+      "ie l",
+      "i el",
+      "要 有",
+      "▁guarant eed",
+      "▁guarantee d",
+      "▁point ed",
+      "ul se",
+      "uls e",
+      "▁c hip",
+      "▁ch ip",
+      "▁chi p",
+      "▁ chip",
+      "En vironment",
+      "发生法律效力 后",
+      "发生 的",
+      "发 生的",
+      "不 停",
+      "旨 在",
+      "reg ular",
+      "t oggle",
+      "火 箭",
+      "gi f",
+      "g if",
+      "这一 点",
+      "这 一点",
+      "▁h emat",
+      "▁he mat",
+      "▁hem at",
+      "▁ hemat",
+      "刑事 拘留",
+      "依法 适用简易程序",
+      "依法适用 简易程序",
+      "▁in line",
+      "▁i nline",
+      "▁ inline",
+      "▁o range",
+      "▁or ange",
+      "▁ orange",
+      "住 宿",
+      "on der",
+      "ond er",
+      "onde r",
+      "o nder",
+      "传感 器",
+      "▁th ousand",
+      "▁thous and",
+      "G P",
+      "▁g rey",
+      "▁gr ey",
+      "▁gre y",
+      "▁ grey",
+      "▁Ins urance",
+      "仿 佛",
+      "▁R oss",
+      "▁Ro ss",
+      "▁Ros s",
+      "sin ce",
+      "si nce",
+      "s ince",
+      "▁alter ations",
+      "▁alteration s",
+      "商 店",
+      "总 之",
+      "ig s",
+      "i gs",
+      "se ries",
+      "ser ies",
+      "s eries",
+      "at ern",
+      "ate rn",
+      "ater n",
+      "a tern",
+      "der ived",
+      "de rived",
+      "derive d",
+      "▁mix ing",
+      "▁mi xing",
+      "稳定 性",
+      "稳 定性",
+      "▁M unicip",
+      "▁Mun icip",
+      "erc ise",
+      "lev ance",
+      "▁F REE",
+      "▁FR EE",
+      "▁decre ases",
+      "▁decrease s",
+      "年 前",
+      "招 商",
+      "un its",
+      "unit s",
+      "uni ts",
+      "u nits",
+      "## ####",
+      "#### ##",
+      "### ###",
+      "##### #",
+      "# #####",
+      "▁c atalog",
+      "▁catal og",
+      "▁ catalog",
+      "一定 会",
+      "hed ral",
+      "引起 的",
+      "引 起的",
+      "抓 获",
+      "▁p ediatric",
+      "▁ped iatric",
+      "▁in sect",
+      "▁ins ect",
+      "An t",
+      "A nt",
+      "▁build er",
+      "▁ builder",
+      "总 收入",
+      "▁M anchester",
+      "▁Man chester",
+      "预 订",
+      "sub scribe",
+      "竞争 力",
+      "mit ter",
+      "mitt er",
+      "m itter",
+      "外 科",
+      "▁\\\\ )\\",
+      "▁\\\\) \\",
+      "▁ \\\\)\\",
+      "▁b ags",
+      "▁bag s",
+      "▁ba gs",
+      "▁ bags",
+      "小 朋友",
+      "▁m orb",
+      "▁mor b",
+      "▁mo rb",
+      "__ \\",
+      "_ _\\",
+      "▁pull ed",
+      "▁pul led",
+      "条 第",
+      "了一 种",
+      "了 一种",
+      "稳定 的",
+      "稳 定的",
+      "出 售",
+      "▁v entricular",
+      "▁vent ricular",
+      "▁ventric ular",
+      "int eger",
+      "EX T",
+      "E XT",
+      "不 存在",
+      "▁J ordan",
+      "am ar",
+      "ama r",
+      "a mar",
+      "▁n pm",
+      "▁np m",
+      "▁ npm",
+      "▁frequ encies",
+      "或者 是",
+      "的 证据",
+      "ch arg",
+      "char g",
+      "cha rg",
+      "work ing",
+      "wor king",
+      "▁c ancers",
+      "▁can cers",
+      "▁cancer s",
+      "▁K evin",
+      "▁Ke vin",
+      "熟 练",
+      "▁P sych",
+      "▁Ps ych",
+      "▁Psy ch",
+      "▁g ro",
+      "▁gr o",
+      "▁ gro",
+      "vi a",
+      "v ia",
+      "c raft",
+      "nI nt",
+      "nIn t",
+      "n Int",
+      "税 收",
+      "▁E mail",
+      "▁Em ail",
+      "▁ Email",
+      "▁s phere",
+      "▁sp here",
+      "▁spher e",
+      "▁sph ere",
+      "▁ sphere",
+      "cent s",
+      "c ents",
+      "▁M arg",
+      "▁Mar g",
+      "▁Ma rg",
+      "党委 书记",
+      "党 委书记",
+      "十 日",
+      "用 水",
+      "AD D",
+      "A DD",
+      "▁he ading",
+      "▁head ing",
+      "▁ heading",
+      "除 数",
+      "▁le ak",
+      "▁t ower",
+      "▁to wer",
+      "▁tow er",
+      "▁cer emony",
+      "▁ceremon y",
+      "sp ot",
+      "s pot",
+      "单 词",
+      "▁N orm",
+      "▁No rm",
+      "▁Nor m",
+      "▁ Norm",
+      "▁U V",
+      "▁ UV",
+      "石 家庄",
+      "▁F ive",
+      "▁Fi ve",
+      "▁ Five",
+      "聪 明",
+      "选 举",
+      "▁Pro perties",
+      "▁Proper ties",
+      "▁ Properties",
+      "承 受",
+      "阿 尔",
+      "▁def initions",
+      "▁definit ions",
+      "▁definition s",
+      "▁defin itions",
+      "▁ definitions",
+      "▁be half",
+      "▁beh alf",
+      "ot lin",
+      "▁access ories",
+      "D T",
+      "▁ste ady",
+      "▁stead y",
+      "▁Jun ior",
+      "▁Y et",
+      "▁Ye t",
+      "▁head qu",
+      "▁difficult ies",
+      "pro b",
+      "pr ob",
+      "p rob",
+      "各 国",
+      "▁P as",
+      "▁Pa s",
+      "▁P rob",
+      "▁Pro b",
+      "▁Pr ob",
+      "▁ Prob",
+      "ot hs",
+      "oth s",
+      "o ths",
+      "感受 到",
+      "感 受到",
+      "什么 样的",
+      "什么样 的",
+      "流 水",
+      "一部 分",
+      "一 部分",
+      "i Phone",
+      "ub le",
+      "u ble",
+      "▁L ady",
+      "▁La dy",
+      "▁Lad y",
+      "ag g",
+      "a gg",
+      ")) )",
+      ") ))",
+      "▁pro long",
+      "▁prol ong",
+      "▁R ow",
+      "▁Ro w",
+      "▁ Row",
+      "W R",
+      "'' '\\",
+      "''' \\",
+      "ograph ical",
+      "ographic al",
+      "▁G L",
+      "▁ GL",
+      "的 学习",
+      "uc c",
+      "u cc",
+      "Tr ace",
+      "Tra ce",
+      "T race",
+      "▁f alling",
+      "▁fall ing",
+      "▁fal ling",
+      "▁f illing",
+      "▁fil ling",
+      "▁fill ing",
+      "▁ filling",
+      "▁influ ences",
+      "▁influence s",
+      "▁st range",
+      "▁str ange",
+      "▁strang e",
+      "合 物",
+      "▁op posed",
+      "▁oppose d",
+      "▁sign ing",
+      "▁sig ning",
+      "出现 的",
+      "▁O w",
+      "▁ Ow",
+      "Pro xy",
+      "Pr oxy",
+      "开 拓",
+      "ian ces",
+      "iance s",
+      "i ances",
+      "▁document ary",
+      "所 述",
+      "ex tra",
+      "ext ra",
+      "extr a",
+      "倒 入",
+      "▁ge ometric",
+      "▁geomet ric",
+      "▁geo metric",
+      "▁geom etric",
+      "▁ geometric",
+      "▁At lantic",
+      "▁Atl antic",
+      "V ideo",
+      "▁P olit",
+      "▁Pol it",
+      "▁Po lit",
+      "第 十",
+      "▁N AS",
+      "▁NA S",
+      "▁ NAS",
+      "▁ 如果",
+      "▁A zure",
+      "▁Az ure",
+      "▁ Azure",
+      "▁evalu ating",
+      "▁eval uating",
+      "的 总",
+      "el lar",
+      "ell ar",
+      "ella r",
+      "Lin q",
+      "Li nq",
+      "ph ys",
+      "phy s",
+      "▁Pr imary",
+      "▁Prim ary",
+      "▁ Primary",
+      "▁p iano",
+      "▁pi ano",
+      "▁pian o",
+      "作 战",
+      "pr ead",
+      "pre ad",
+      "p read",
+      "清 水",
+      "▁re plication",
+      "▁replic ation",
+      "▁replica tion",
+      "▁predict ive",
+      "刑 初",
+      "候 审",
+      "▁p ron",
+      "▁pro n",
+      "▁pr on",
+      "▁ pron",
+      "iation s",
+      "iat ions",
+      "i ations",
+      "隔 离",
+      "五 个",
+      "视 角",
+      "▁cou pling",
+      "▁coup ling",
+      "▁N FL",
+      "▁NF L",
+      "▁G ree",
+      "▁Gr ee",
+      "▁Gre e",
+      "民一 初字第",
+      "▁collabor ative",
+      "al g",
+      "a lg",
+      "▁gl yc",
+      "▁gly c",
+      "▁acc used",
+      "▁accus ed",
+      "钢 铁",
+      "好 友",
+      "辛 苦",
+      "hu b",
+      "h ub",
+      "你 认为",
+      "▁spe aker",
+      "▁speak er",
+      "▁ speaker",
+      "▁a band",
+      "▁ab and",
+      "只需 要",
+      "只 需要",
+      "正确的 选项",
+      "▁chrom at",
+      "▁chr omat",
+      "支持 公诉",
+      "R IGHT",
+      "▁f iction",
+      "▁fi ction",
+      "▁ fiction",
+      "ru st",
+      "rus t",
+      "r ust",
+      "▁l aid",
+      "▁la id",
+      "几 何",
+      "▁educ ated",
+      "▁educate d",
+      "成功 的",
+      "是 可以",
+      "观 看",
+      "ve t",
+      "v et",
+      "HO ME",
+      "H OME",
+      "方 位",
+      "记 载",
+      "我们 要",
+      "iz ard",
+      "iza rd",
+      "izar d",
+      "ang les",
+      "angle s",
+      "上 传",
+      "th ony",
+      "thon y",
+      "取 保",
+      "ield s",
+      "iel ds",
+      "▁S ave",
+      "▁Sa ve",
+      "▁Sav e",
+      "▁ Save",
+      "▁sup pose",
+      "一 周",
+      "lo gic",
+      "log ic",
+      "运 作",
+      "▁demonstr ates",
+      "▁demonstrate s",
+      "▁M eta",
+      "▁Me ta",
+      "▁Met a",
+      "▁ Meta",
+      "w r",
+      "hy d",
+      "h yd",
+      "▁elim inate",
+      "▁v oted",
+      "▁vote d",
+      "▁vot ed",
+      "▁vo ted",
+      "Ｘ Ｘ",
+      "千 里",
+      "的 社会",
+      "▁e ry",
+      "▁er y",
+      "▁ ery",
+      "Con tract",
+      "▁for ever",
+      "▁fore ver",
+      "重要 的是",
+      "重要的 是",
+      "▁view ing",
+      "▁contract s",
+      "▁contr acts",
+      "▁ contracts",
+      "再加 上",
+      "再 加上",
+      "拉 丁",
+      "STR ACT",
+      "STRA CT",
+      "S TRACT",
+      "景 点",
+      "玩 具",
+      "▁rat ios",
+      "▁ratio s",
+      "有 助于",
+      "ol as",
+      "ola s",
+      "o las",
+      "▁nan op",
+      "▁na nop",
+      "▁nano p",
+      "▁Cl inton",
+      "▁Clin ton",
+      "入 门",
+      "er to",
+      "ert o",
+      "▁S ample",
+      "▁Sam ple",
+      "▁ Sample",
+      "▁algebra ic",
+      "▁com pression",
+      "▁comp ression",
+      "▁compr ession",
+      "▁compress ion",
+      "▁\" ${",
+      "▁\"$ {",
+      "▁ \"${",
+      "▁S in",
+      "▁Si n",
+      "▁ Sin",
+      "女 主",
+      "ib l",
+      "i bl",
+      "am an",
+      "ama n",
+      "a man",
+      "ar o",
+      "a ro",
+      "▁sum mar",
+      "▁summ ar",
+      "▁v ocal",
+      "▁voc al",
+      "▁vo cal",
+      "▁G rad",
+      "▁Gr ad",
+      "▁Gra d",
+      "▁ Grad",
+      "nI nter",
+      "nIn ter",
+      "nInt er",
+      "n Inter",
+      "正 好",
+      "▁class room",
+      "▁act ively",
+      "▁activ ely",
+      "▁active ly",
+      "▁ actively",
+      "col lect",
+      "coll ect",
+      "nR UN",
+      "n RUN",
+      "▁f lying",
+      "▁fl ying",
+      "▁fly ing",
+      "▁qu eries",
+      "▁que ries",
+      "▁quer ies",
+      "▁ queries",
+      "▁pre paring",
+      "▁prep aring",
+      "▁prepar ing",
+      "▁M RI",
+      "▁MR I",
+      "▁ MRI",
+      "ox ide",
+      "oxid e",
+      "com ments",
+      "comment s",
+      "comm ents",
+      "多 家",
+      "江 区",
+      "口 感",
+      "辣 椒",
+      "建 于",
+      "▁m agnet",
+      "▁mag net",
+      "▁magn et",
+      "▁re serv",
+      "▁res erv",
+      "▁script s",
+      "▁ scripts",
+      "日作出 的",
+      "日 作出的",
+      "▁per ception",
+      "▁percept ion",
+      "▁Ste el",
+      "act ors",
+      "actor s",
+      "a ctors",
+      "▁y ards",
+      "▁yard s",
+      "▁ yards",
+      "▁M B",
+      "▁ MB",
+      "▁M it",
+      "▁Mi t",
+      "▁ 输入",
+      "▁ex haust",
+      "多年 的",
+      "多 年的",
+      "▁t ension",
+      "▁tens ion",
+      "▁ tension",
+      "▁mitochond rial",
+      "▁mitochondria l",
+      "坚 决",
+      ")} \\",
+      ") }\\",
+      "▁Ukr aine",
+      "▁Mem orial",
+      "if est",
+      "ife st",
+      "i fest",
+      "get Element",
+      "▁in tensive",
+      "▁int ensive",
+      "▁intens ive",
+      "▁ intensive",
+      "▁C SS",
+      "▁CS S",
+      "▁ CSS",
+      "P N",
+      "▁ad vis",
+      "▁adv is",
+      "如果 未按本判决",
+      "橡 胶",
+      "el li",
+      "ell i",
+      "e lli",
+      "nd ata",
+      "nda ta",
+      "ndat a",
+      "n data",
+      "▁n in",
+      "▁ni n",
+      "▁ nin",
+      "▁dem ocr",
+      "▁demo cr",
+      "可 用",
+      "ab e",
+      "a be",
+      "并 未",
+      "▁cl usters",
+      "▁cluster s",
+      "▁clust ers",
+      "▁ clusters",
+      "▁rare ly",
+      "也不 能",
+      "也 不能",
+      "ba nk",
+      "ban k",
+      "b ank",
+      "oot er",
+      "oo ter",
+      "o oter",
+      "出庭 支持公诉",
+      "To String",
+      "▁l ap",
+      "▁la p",
+      "▁ lap",
+      "um or",
+      "u mor",
+      "▁作者 简介",
+      "▁D eep",
+      "▁De ep",
+      "▁Dee p",
+      "▁ Deep",
+      "nu t",
+      "n ut",
+      "两个 人",
+      "两 个人",
+      "▁A rm",
+      "▁Ar m",
+      "▁ Arm",
+      "▁Sur vey",
+      "▁Surv ey",
+      "▁Surve y",
+      "▁design ated",
+      "▁ designated",
+      "▁function ing",
+      "焦 虑",
+      "的 功能",
+      "▁Lin ear",
+      "▁Line ar",
+      "▁Li near",
+      "▁ Linear",
+      "▁reve als",
+      "▁reveal s",
+      "▁T ennessee",
+      "▁adapt ation",
+      "是一 部",
+      "是 一部",
+      "▁S afety",
+      "▁Saf ety",
+      "▁Safe ty",
+      "▁spec tra",
+      "▁spect ra",
+      "▁dis ability",
+      "se min",
+      "sem in",
+      "▁A rk",
+      "▁Ar k",
+      "Fi x",
+      "F ix",
+      "▁sc rew",
+      "▁scr ew",
+      "▁T en",
+      "▁Te n",
+      "▁ Ten",
+      "等 地",
+      "曝 光",
+      "▁local ization",
+      ">\\ \\",
+      "> \\\\",
+      "bo ne",
+      "bon e",
+      "b one",
+      "b untu",
+      "Be fore",
+      "B efore",
+      "▁D un",
+      "▁Du n",
+      "' (",
+      "越 南",
+      "举 报",
+      "为 人",
+      "▁p df",
+      "▁pd f",
+      "▁ pdf",
+      "▁ad mission",
+      "▁adm ission",
+      "贩 卖",
+      "面 包",
+      "▁O il",
+      "▁m ol",
+      "▁mo l",
+      "▁ mol",
+      "▁comb ining",
+      "▁combin ing",
+      "▁switch ing",
+      "▁swit ching",
+      "min ecraft",
+      "mine craft",
+      "ol k",
+      "欠 原告",
+      "以 便",
+      "ho p",
+      "h op",
+      "警 方",
+      "▁: )\\",
+      "▁:) \\",
+      "▁p ipe",
+      "▁pi pe",
+      "▁pip e",
+      "▁ pipe",
+      "H tml",
+      "▁Or iginally",
+      "▁Original ly",
+      "▁Origin ally",
+      "不 去",
+      "伟 大",
+      "▁M embers",
+      "▁Mem bers",
+      "▁Member s",
+      "▁ Members",
+      "om m",
+      "o mm",
+      "▁ext ensions",
+      "▁extension s",
+      "▁ extensions",
+      "说 道",
+      "▁Nor way",
+      "会 不会",
+      "应用 于",
+      "应 用于",
+      "▁Democr ats",
+      "▁Democrat s",
+      "▁J en",
+      "▁Je n",
+      "▁ Jen",
+      "do r",
+      "d or",
+      "▁a live",
+      "▁al ive",
+      "▁ali ve",
+      "▁ alive",
+      "下 落",
+      "荣 耀",
+      "事业 单位",
+      "AR S",
+      "A RS",
+      "几 次",
+      "帮 我",
+      "应当 提交上诉状正本一份",
+      "饭 店",
+      "Event s",
+      "Ev ents",
+      "Even ts",
+      "E vents",
+      "nB ritish",
+      "n British",
+      "遗 产",
+      "▁l iked",
+      "▁like d",
+      "▁lik ed",
+      "▁ph ases",
+      "▁phase s",
+      "时代 的",
+      "时 代的",
+      "弟 弟",
+      "▁f lower",
+      "▁fl ower",
+      "▁flo wer",
+      "▁flow er",
+      "▁ flower",
+      "▁t ap",
+      "▁ta p",
+      "▁ tap",
+      "持 有",
+      "计 量",
+      "▁start up",
+      "▁ startup",
+      "or igin",
+      "ori gin",
+      "orig in",
+      "冬 季",
+      "__ )\\",
+      "__) \\",
+      "_ _)\\",
+      "几 种",
+      "л и",
+      "co system",
+      "cos ystem",
+      "c osystem",
+      "两 条",
+      "ri sk",
+      "ris k",
+      "r isk",
+      "▁E lectric",
+      "▁Elect ric",
+      "好 处",
+      "sing le",
+      "s ingle",
+      "影 像",
+      "是一 名",
+      "是 一名",
+      "的 多",
+      "▁ar my",
+      "▁arm y",
+      "▁ep idem",
+      "▁epid em",
+      "▁jun ior",
+      "十二 条",
+      "十 二条",
+      "编 码",
+      "航 天",
+      "er ver",
+      "erv er",
+      "erve r",
+      "ation ship",
+      "ations hip",
+      "▁pl acement",
+      "▁place ment",
+      "▁plac ement",
+      "▁ placement",
+      "en o",
+      "e no",
+      "没有 任何",
+      "没 有任何",
+      "P F",
+      "腐 蚀",
+      "▁T emplate",
+      "▁Tem plate",
+      "▁Temp late",
+      "▁ Template",
+      "XX X",
+      "X XX",
+      "▁rough ly",
+      "至今 未履行",
+      "至今未 履行",
+      "黑 暗",
+      "服务 中心",
+      "▁we ird",
+      "inf lamm",
+      "有多 少",
+      "有 多少",
+      "▁p ri",
+      "▁pr i",
+      "▁ pri",
+      "月 出生",
+      "el ve",
+      "取保 候审",
+      "民 生",
+      "▁sc anning",
+      "▁scan ning",
+      "gen erated",
+      "gener ated",
+      "generate d",
+      "gene rated",
+      "g enerated",
+      "▁pres idential",
+      "▁president ial",
+      "▁High way",
+      "▁Kent ucky",
+      "gen erator",
+      "gener ator",
+      "▁In tel",
+      "▁Int el",
+      "▁ Intel",
+      "▁lo ads",
+      "▁load s",
+      "▁ loads",
+      "▁a dip",
+      "▁ad ip",
+      "▁w avelength",
+      "▁wave length",
+      "▁R A",
+      "▁ RA",
+      "▁j unction",
+      "▁jun ction",
+      "▁e mails",
+      "▁em ails",
+      "▁email s",
+      "▁ emails",
+      "住 址",
+      "▁c ategor",
+      "▁cate gor",
+      "▁ categor",
+      "▁restr ictions",
+      "▁restrict ions",
+      "▁restriction s",
+      "故意 伤害",
+      "▁in ch",
+      "▁i nch",
+      "▁inc h",
+      "▁ inch",
+      "▁ 起点中文网",
+      "党 支部",
+      "▁att ach",
+      "▁ attach",
+      "▁k illing",
+      "▁kil ling",
+      "▁kill ing",
+      "sc h",
+      "s ch",
+      "因 子",
+      "报 考",
+      "▁f ake",
+      "▁fa ke",
+      "▁ fake",
+      "trans fer",
+      "br ew",
+      "bre w",
+      "b rew",
+      "切 换",
+      "av it",
+      "avi t",
+      "a vit",
+      "原审 原告",
+      "▁opt imize",
+      "▁optim ize",
+      "▁ optimize",
+      "蓝 色",
+      "不 超过",
+      "主 机",
+      "▁R ome",
+      "▁Ro me",
+      "▁Rom e",
+      "▁d ictionary",
+      "▁ dictionary",
+      "▁H ind",
+      "▁Hi nd",
+      "urs ive",
+      "国土资源 局",
+      "▁cl imb",
+      "▁clim b",
+      "▁cli mb",
+      "▁M L",
+      "▁ ML",
+      "▁Con text",
+      "▁Cont ext",
+      "▁Co ntext",
+      "▁ Context",
+      "▁Iss ue",
+      "▁ Issue",
+      "▁quadr atic",
+      "▁quad ratic",
+      "▁F ast",
+      "▁Fa st",
+      "▁Fas t",
+      "▁ Fast",
+      "▁Del hi",
+      "ei n",
+      "e in",
+      "辖 区",
+      "饮 料",
+      "与 其",
+      "的 钱",
+      "▁R ot",
+      "▁Ro t",
+      "▁ Rot",
+      "高新 技术",
+      "高 新技术",
+      "▁P rize",
+      "▁Pr ize",
+      "▁Pri ze",
+      "▁except ional",
+      "▁exception al",
+      "O F",
+      "re nce",
+      "ren ce",
+      "r ence",
+      "LE S",
+      "L ES",
+      "却 是",
+      "▁def end",
+      "▁defe nd",
+      "▁i ons",
+      "▁ion s",
+      "▁io ns",
+      "▁ ions",
+      "▁co uples",
+      "▁cou ples",
+      "▁couple s",
+      "▁coup les",
+      "▁B log",
+      "▁Bl og",
+      "▁Blo g",
+      "▁ Blog",
+      "▁D R",
+      "▁ DR",
+      "ap or",
+      "apo r",
+      "a por",
+      "▁encour aged",
+      "▁encourage d",
+      "照 明",
+      "离婚 诉讼",
+      "▁dis charge",
+      "▁discharg e",
+      "▁st rike",
+      "▁str ike",
+      "▁stri ke",
+      "我 自己",
+      "入 选",
+      "M G",
+      "的 光",
+      "om ent",
+      "ome nt",
+      "omen t",
+      "o ment",
+      "im eter",
+      "ime ter",
+      "imet er",
+      "i meter",
+      "pl eted",
+      "ple ted",
+      "plete d",
+      "plet ed",
+      "p leted",
+      "ig ator",
+      "iga tor",
+      "▁y ours",
+      "▁you rs",
+      "▁your s",
+      "▁yo urs",
+      "期 货",
+      "Li b",
+      "L ib",
+      "依法 由",
+      "▁\\' <",
+      "▁ \\'<",
+      "ex tract",
+      "extra ct",
+      "extr act",
+      "sc roll",
+      "scr oll",
+      "在 使用",
+      "放 松",
+      "DE S",
+      "D ES",
+      "▁def iciency",
+      "▁Y outh",
+      "▁You th",
+      "▁Yo uth",
+      "Ar ea",
+      "Are a",
+      "A rea",
+      "▁mem ories",
+      "▁memo ries",
+      "▁memor ies",
+      "> `",
+      "手 动",
+      "▁In dependent",
+      "▁Ind ependent",
+      "▁conven ience",
+      "var iable",
+      "vari able",
+      "▁occ urrence",
+      "▁occur rence",
+      "▁occurre nce",
+      "解 读",
+      "除 外",
+      "▁c erv",
+      "▁cer v",
+      "▁ce rv",
+      "青 海",
+      "遗 传",
+      "▁rat ings",
+      "▁ra tings",
+      "▁rating s",
+      "er ator",
+      "era tor",
+      "▁g orgeous",
+      "配 备",
+      "pt ic",
+      "p tic",
+      "看 完",
+      "oph il",
+      "o phil",
+      "▁s hots",
+      "▁sh ots",
+      "▁shot s",
+      "▁ shots",
+      "▁per ceived",
+      "▁perceive d",
+      "▁biom ark",
+      "▁bio mark",
+      "Te rm",
+      "Ter m",
+      "T erm",
+      "我 认为",
+      "x t",
+      "号 楼",
+      "提 到",
+      "▁M otor",
+      "▁Mot or",
+      "▁Mo tor",
+      "▁Moto r",
+      "▁ Motor",
+      "▁m aternal",
+      "▁mat ernal",
+      "▁ma ternal",
+      "▁mater nal",
+      "行 长",
+      "祝 福",
+      "东 京",
+      "▁D raw",
+      "▁Dr aw",
+      "▁ Draw",
+      "监狱 于",
+      "▁deliver ing",
+      "▁w idget",
+      "▁wid get",
+      "▁ widget",
+      "神 奇",
+      "干 扰",
+      "▁sm oke",
+      "▁smo ke",
+      "心 灵",
+      "正 规",
+      "th eless",
+      "the less",
+      "thel ess",
+      "▁I nv",
+      "▁In v",
+      "▁ Inv",
+      "▁B at",
+      "▁Ba t",
+      "function s",
+      "▁m l",
+      "▁ ml",
+      "▁B urn",
+      "▁Bur n",
+      "▁Bu rn",
+      "nN ext",
+      "nNe xt",
+      "n Next",
+      "租 金",
+      "▁soph istic",
+      "生产 的",
+      "生 产的",
+      "▁c ortex",
+      "▁cor tex",
+      "▁cort ex",
+      "▁dis put",
+      "▁disp ut",
+      "▁mult iply",
+      "▁multi ply",
+      "▁multip ly",
+      "很 高",
+      "基础 设施",
+      "▁vari ability",
+      "▁var iability",
+      "▁h ierarch",
+      "东 风",
+      "year s",
+      "ye ars",
+      "y ears",
+      "发现被执行人有可供执行财产 的",
+      "资 质",
+      "▁Ph oto",
+      "▁Phot o",
+      "▁ Photo",
+      "▁an imation",
+      "▁anim ation",
+      "▁ animation",
+      "mark et",
+      "mar ket",
+      "per iod",
+      "ca rt",
+      "car t",
+      "c art",
+      "▁O bserv",
+      "▁Ob serv",
+      "▁Obs erv",
+      "▁ Observ",
+      "买 的",
+      "in itely",
+      "init ely",
+      "inite ly",
+      "▁univers ities",
+      "Re ct",
+      "Rec t",
+      "R ect",
+      "▁fr anch",
+      "▁fra nch",
+      "▁mob ility",
+      "▁mobil ity",
+      "nU p",
+      "n Up",
+      "栽 培",
+      "▁an terior",
+      "▁ant erior",
+      "大学 出版社",
+      "▁Re cent",
+      "▁Rec ent",
+      "▁ Recent",
+      "▁Me chan",
+      "re tt",
+      "ret t",
+      "r ett",
+      "nS olve",
+      "n Solve",
+      "▁establish ment",
+      "▁Ch rome",
+      "▁Chrom e",
+      "▁ Chrome",
+      "mo s",
+      "m os",
+      "建设 的",
+      "摄像 头",
+      "ri k",
+      "r ik",
+      "▁un f",
+      "▁u nf",
+      "振 兴",
+      "有一 种",
+      "有 一种",
+      "▁A st",
+      "▁As t",
+      "▁ Ast",
+      "an tom",
+      "ant om",
+      "anto m",
+      "▁f s",
+      "▁ fs",
+      "ys ts",
+      "yst s",
+      "y sts",
+      "▁ch ains",
+      "▁chain s",
+      "▁cha ins",
+      "▁ chains",
+      "▁R ap",
+      "▁Ra p",
+      "供应 商",
+      "ac ial",
+      "aci al",
+      "acia l",
+      "长 安",
+      "ll a",
+      "l la",
+      "▁ 然后",
+      "在 大",
+      "文件 夹",
+      "M F",
+      "带来 了",
+      "带 来了",
+      "▁ast hma",
+      "▁work flow",
+      "▁ workflow",
+      "▁back up",
+      "▁ backup",
+      "不仅 仅",
+      "不 仅仅",
+      "复印 件",
+      "稍 微",
+      ") |",
+      "路 由",
+      "▁des cribing",
+      "▁describ ing",
+      "ul ative",
+      "▁r outer",
+      "▁ro uter",
+      "▁rout er",
+      "▁route r",
+      "▁rou ter",
+      "▁ router",
+      "▁engine ers",
+      "▁engineer s",
+      "架 构",
+      "rec ip",
+      "享 有",
+      "快 的",
+      "Com ment",
+      "Comm ent",
+      "inter est",
+      "in gers",
+      "ing ers",
+      "inger s",
+      "inge rs",
+      "严重 的",
+      "严 重的",
+      "服 饰",
+      "▁F O",
+      "▁ FO",
+      "▁ar rival",
+      "▁arr ival",
+      "▁arriv al",
+      "难 题",
+      "▁st ats",
+      "▁stat s",
+      "▁ stats",
+      "act ivation",
+      "activ ation",
+      "w p",
+      "身 材",
+      "Bi t",
+      "B it",
+      "价值 观",
+      "改 为",
+      "▁S ize",
+      "▁Si ze",
+      "▁ Size",
+      "流 通",
+      "path y",
+      "pat hy",
+      "pa thy",
+      "p athy",
+      "▁ob serve",
+      "▁observ e",
+      "▁obs erve",
+      "emet ery",
+      "eme tery",
+      "ith metic",
+      "ti t",
+      "t it",
+      "sequ ence",
+      "s equence",
+      "> {",
+      "风 机",
+      "▁M un",
+      "▁Mu n",
+      "▁far mers",
+      "▁farm ers",
+      "▁farmer s",
+      "▁T reatment",
+      "▁Treat ment",
+      "某 乙",
+      "罚 款",
+      "高 层",
+      "▁p lt",
+      "▁pl t",
+      "▁ plt",
+      "Eng lish",
+      "▁art erial",
+      "▁arter ial",
+      "▁Pro file",
+      "▁Prof ile",
+      "▁ Profile",
+      "张 家",
+      "广西 壮",
+      "通 报",
+      "▁bron ze",
+      "交 叉",
+      "▁build s",
+      "▁ builds",
+      "模 具",
+      "的发 生",
+      "的 发生",
+      "Sec urity",
+      "▁k w",
+      "▁ kw",
+      "ro us",
+      "rou s",
+      "r ous",
+      "ov erty",
+      "over ty",
+      "▁N ative",
+      "▁Nat ive",
+      "▁ Native",
+      "以 外",
+      "体 内",
+      "▁in direct",
+      "▁ind irect",
+      "▁travel ing",
+      "▁trav eling",
+      "十九 条",
+      "十 九条",
+      "▁s mile",
+      "▁sm ile",
+      "▁design ers",
+      "▁designer s",
+      "▁meaning ful",
+      "博 客",
+      "诸 多",
+      "▁sc ales",
+      "▁scale s",
+      "▁scal es",
+      "隐 患",
+      "水 分",
+      "一 会",
+      "wit zer",
+      "witz er",
+      "w itzer",
+      "yr ics",
+      "y rics",
+      "掉 了",
+      "▁de termin",
+      "▁determ in",
+      "▁deter min",
+      "▁T ai",
+      "▁Ta i",
+      "Loc k",
+      "Lo ck",
+      "L ock",
+      "pl ane",
+      "plan e",
+      "p lane",
+      "挑 选",
+      "透 露",
+      "秘书 长",
+      "br ace",
+      "bra ce",
+      "b race",
+      "nA ct",
+      "nAc t",
+      "n Act",
+      "TR IBUT",
+      "▁Mi ami",
+      "ah oo",
+      "aho o",
+      "▁ult ra",
+      "▁ul tra",
+      "▁ultr a",
+      "nEx ample",
+      "n Example",
+      "向 被告",
+      "▁M aria",
+      "▁Mar ia",
+      "▁Ma ria",
+      "▁Mari a",
+      "▁separ ately",
+      "▁separate ly",
+      "DA TA",
+      "DAT A",
+      "D ATA",
+      "最 早",
+      "▁Re sources",
+      "▁Res ources",
+      "▁Resource s",
+      "▁ Resources",
+      "▁v essel",
+      "▁vess el",
+      "▁ves sel",
+      "研 讨",
+      "管辖 权",
+      "澳 大利",
+      "▁qu otes",
+      "▁quote s",
+      "▁quot es",
+      "▁quo tes",
+      "▁ quotes",
+      "get Logger",
+      "win d",
+      "wi nd",
+      "w ind",
+      "▁D J",
+      "▁ DJ",
+      "▁defe ated",
+      "▁defeat ed",
+      "(\\' #",
+      "( \\'#",
+      "▁top ology",
+      "▁har vest",
+      "拍 照",
+      "nB iography",
+      "nBi ography",
+      "▁h oles",
+      "▁hol es",
+      "▁hole s",
+      "▁ho les",
+      "▁di visor",
+      "▁div isor",
+      "nB efore",
+      "nBe fore",
+      "n Before",
+      "zh e",
+      "z he",
+      "▁t hesis",
+      "▁the sis",
+      "▁th esis",
+      "gm ail",
+      "g mail",
+      "En abled",
+      "Enable d",
+      "ph osph",
+      "▁l augh",
+      "▁la ugh",
+      "▁S R",
+      "▁ SR",
+      "▁A A",
+      "▁ AA",
+      "ro it",
+      "r oit",
+      "st im",
+      "s tim",
+      "山 水",
+      "wal k",
+      "w alk",
+      ")) /",
+      ") )/",
+      "\"/ >\\",
+      "\"/> \\",
+      "\" />\\",
+      "re lation",
+      "rel ation",
+      "广西壮 族自治区",
+      "让 自己",
+      "cle rosis",
+      "cler osis",
+      "▁inter fer",
+      "▁interf er",
+      "澳大利 亚",
+      "▁St ill",
+      "有关 法律规定",
+      "ا ل",
+      "▁answer ed",
+      "▁ answered",
+      "▁Th ird",
+      "▁ Third",
+      "B V",
+      "就没 有",
+      "就 没有",
+      "▁h ub",
+      "▁hu b",
+      "▁ hub",
+      "▁ra ising",
+      "▁ raising",
+      "▁c oli",
+      "▁col i",
+      "▁co li",
+      "▁old est",
+      "▁ol dest",
+      "\\' ))",
+      "\\') )",
+      "\\ '))",
+      "ic ator",
+      "ica tor",
+      "Jo in",
+      "J oin",
+      "较 低",
+      "ot ide",
+      "otid e",
+      "oti de",
+      "nC ar",
+      "n Car",
+      "▁adapt ed",
+      "▁adap ted",
+      "短 信",
+      "开 设",
+      "对 原告",
+      "[: ,",
+      "[ :,",
+      "▁J S",
+      "▁ JS",
+      "擅 长",
+      "多 么",
+      "▁St yle",
+      "▁Styl e",
+      "▁ Style",
+      "▁expand ing",
+      "fl ags",
+      "flag s",
+      "紧 急",
+      "▁short ly",
+      "▁Mark eting",
+      "▁Market ing",
+      "整 改",
+      "社会 信用",
+      "con sum",
+      "cons um",
+      "▁scen arios",
+      "▁scenario s",
+      "的 质量",
+      "本院 不予",
+      "▁spe akers",
+      "▁speak ers",
+      "▁speaker s",
+      "Sc ope",
+      "qu ot",
+      "quo t",
+      "nap shot",
+      "▁sing ular",
+      "▁St aff",
+      "▁Staf f",
+      "▁Sta ff",
+      "tem plates",
+      "template s",
+      "ru p",
+      "r up",
+      "相 识",
+      "au di",
+      "aud i",
+      "a udi",
+      "\\' ][\\'",
+      "\\'] [\\'",
+      "\\'][ \\'",
+      "外 部",
+      "os cow",
+      "osc ow",
+      "dial og",
+      "d ialog",
+      "数 码",
+      "re dis",
+      "red is",
+      "为什么 要",
+      "懂 得",
+      "Ev al",
+      "E val",
+      "b x",
+      "▁f etal",
+      "▁fe tal",
+      "▁fet al",
+      "▁Adv anced",
+      "▁Advance d",
+      "▁ Advanced",
+      "▁ord inary",
+      "▁ ordinary",
+      "Se nd",
+      "Sen d",
+      "S end",
+      "▁( @",
+      "▁ (@",
+      "▁sub sid",
+      "▁subs id",
+      "的最 小",
+      "的 最小",
+      "失信 被执行人",
+      "▁priv ile",
+      "▁ privile",
+      "▁d ump",
+      "▁du mp",
+      "▁ dump",
+      "能 在",
+      "这 条",
+      "一 半",
+      "▁d esk",
+      "▁de sk",
+      "▁des k",
+      "▁ desk",
+      "饲 料",
+      "▁fin ishing",
+      "▁finish ing",
+      "la v",
+      "l av",
+      "▁P I",
+      "▁ PI",
+      "bin ary",
+      "b inary",
+      "; ;",
+      "▁rank ed",
+      "▁ ranked",
+      "▁graph ic",
+      "▁ graphic",
+      "▁c uts",
+      "▁cut s",
+      "▁cu ts",
+      "▁ cuts",
+      "▁phen otype",
+      "id en",
+      "ide n",
+      "i den",
+      "▁fl ags",
+      "▁flag s",
+      "▁ flags",
+      "▁cur riculum",
+      "▁H awai",
+      "▁Haw ai",
+      "阿 拉",
+      "▁H ope",
+      "▁Hop e",
+      "▁Ho pe",
+      "▁ass ault",
+      "附 加",
+      "▁en rolled",
+      "▁enroll ed",
+      "▁A TP",
+      "▁AT P",
+      "▁ ATP",
+      "▁v ec",
+      "▁ve c",
+      "▁ vec",
+      "▁sub stit",
+      "▁subst it",
+      "▁subs tit",
+      "▁H arry",
+      "▁Har ry",
+      "▁Harr y",
+      "跟 我",
+      "▁H art",
+      "▁Har t",
+      "▁Ha rt",
+      "▁in hab",
+      "▁inh ab",
+      "Wi n",
+      "W in",
+      "Con d",
+      "Co nd",
+      "C ond",
+      "▁pan els",
+      "▁panel s",
+      "▁pa nels",
+      "▁pane ls",
+      "bo y",
+      "b oy",
+      "od ynamic",
+      "odynam ic",
+      "o dynamic",
+      "裁 判",
+      "相关 法律条文",
+      "▁ 不",
+      "▁d owntown",
+      "▁down town",
+      "▁downt own",
+      "st dc",
+      "std c",
+      "▁philosoph y",
+      "淘 宝",
+      "ot on",
+      "oto n",
+      "o ton",
+      "储 蓄",
+      "ace ae",
+      "acea e",
+      "越来 越多",
+      "越来越 多",
+      "ie v",
+      "i ev",
+      "▁ 宋",
+      "前 进",
+      "段 时间",
+      "选 答案",
+      "此 前",
+      "Run time",
+      "Ru ntime",
+      "ma y",
+      "m ay",
+      "另一 方面",
+      "另一方 面",
+      "另 一方面",
+      "▁P rivate",
+      "▁Pr ivate",
+      "▁Priv ate",
+      "▁ Private",
+      "▁dev iation",
+      "▁M oon",
+      "▁Mo on",
+      "▁Tr ade",
+      "▁Tra de",
+      "▁Trad e",
+      "▁ Trade",
+      "人大 常委会",
+      "op ens",
+      "ope ns",
+      "open s",
+      "▁a side",
+      "▁as ide",
+      "▁ aside",
+      "▁b ounds",
+      "▁bound s",
+      "▁ bounds",
+      "oc on",
+      "oco n",
+      "o con",
+      "wi ll",
+      "w ill",
+      "or ses",
+      "ors es",
+      "orse s",
+      "▁e legant",
+      "▁eleg ant",
+      "▁re ceiver",
+      "▁rece iver",
+      "▁receive r",
+      "▁ receiver",
+      "▁a fr",
+      "▁af r",
+      "() `",
+      "( )`",
+      "▁res idence",
+      "▁resid ence",
+      "▁reside nce",
+      "de let",
+      "del et",
+      "前 提",
+      "可供执行的财产 线索",
+      "CP U",
+      "C PU",
+      "▁bl end",
+      "▁ble nd",
+      "诗 词",
+      "▁Array List",
+      "▁ ArrayList",
+      "▁ 红",
+      "火 车",
+      "学 名",
+      "组织 的",
+      "▁wood en",
+      "▁wo oden",
+      "摩托 车",
+      "▁H ello",
+      "▁Hel lo",
+      "▁Hell o",
+      "▁ Hello",
+      "脱 贫",
+      "▁un iverse",
+      "▁univers e",
+      "▁uni verse",
+      "非 洲",
+      "供 给",
+      "▁poor ly",
+      "j query",
+      "依 托",
+      "su ch",
+      "s uch",
+      "▁E ver",
+      "▁Ev er",
+      "▁Eve r",
+      "▁under going",
+      "▁undergo ing",
+      "党组 成员",
+      "党 组成员",
+      "▁Com ment",
+      "▁Comm ent",
+      "▁ Comment",
+      "▁end point",
+      "▁ endpoint",
+      "▁we apons",
+      "▁weap ons",
+      "▁weapon s",
+      "▁ann ouncement",
+      "▁announce ment",
+      "▁announc ement",
+      "MH z",
+      "M Hz",
+      "▁Al abama",
+      "附 件",
+      "▁} ^{",
+      "▁ }^{",
+      "出 场",
+      "落 地",
+      "是 从",
+      "witzer land",
+      "Fore ignKey",
+      "▁D at",
+      "▁Da t",
+      "▁ Dat",
+      "▁per f",
+      "▁pe rf",
+      "▁ perf",
+      "▁pro xy",
+      "▁pr oxy",
+      "▁prox y",
+      "▁ proxy",
+      "sp e",
+      "s pe",
+      "J ust",
+      "▁com bat",
+      "▁comb at",
+      "] *",
+      "亦 未",
+      "▁L ev",
+      "▁Le v",
+      "第二 天",
+      "▁Ad vis",
+      "▁Adv is",
+      "sk i",
+      "s ki",
+      "▁Fed eration",
+      "▁Feder ation",
+      "▁St adium",
+      "加 以",
+      "大 地",
+      "理 性",
+      "▁g enu",
+      "▁gen u",
+      "▁ge nu",
+      "▁con scious",
+      "▁ conscious",
+      "▁en forcement",
+      "▁enforce ment",
+      "▁ 登记机关",
+      "rypt ed",
+      "▁sup plied",
+      "▁suppl ied",
+      "▁prot otype",
+      "▁proto type",
+      "▁ prototype",
+      "ol in",
+      "oli n",
+      "o lin",
+      "wa ve",
+      "wav e",
+      "w ave",
+      "见 到",
+      "巴 黎",
+      "▁p review",
+      "▁pre view",
+      "▁prev iew",
+      "▁ preview",
+      "▁general ized",
+      "▁generalize d",
+      "×× ×××",
+      "××× ××",
+      "×××× ×",
+      "× ××××",
+      "净 化",
+      "会 被",
+      "电 梯",
+      "id ad",
+      "ida d",
+      "i dad",
+      "ut ies",
+      "uti es",
+      "u ties",
+      "Dr iver",
+      "Drive r",
+      "D river",
+      "线 下",
+      "Li ght",
+      "L ight",
+      "▁S ant",
+      "▁San t",
+      "▁Sa nt",
+      "我 都",
+      "之所 以",
+      "之 所以",
+      "四 大",
+      "参 赛",
+      "无 关",
+      "▁con figure",
+      "▁conf igure",
+      "▁config ure",
+      "▁ configure",
+      "吃 了",
+      "▁de ployment",
+      "▁deploy ment",
+      "▁ deployment",
+      "se nt",
+      "sen t",
+      "s ent",
+      "的 长",
+      "am ous",
+      "amo us",
+      "kl ahoma",
+      "忘 了",
+      "▁inf ant",
+      "第四 百",
+      "第 四百",
+      "He re",
+      "Her e",
+      "H ere",
+      "在 北京",
+      "▁int ense",
+      "▁intens e",
+      "经 查",
+      "Fa il",
+      "F ail",
+      "CR IPT",
+      "辩 称",
+      "退 出",
+      "Inst all",
+      "Ins tall",
+      "错 了",
+      "Ex p",
+      "E xp",
+      "▁feel ings",
+      "▁feeling s",
+      "▁fee lings",
+      "Po l",
+      "P ol",
+      "▁not able",
+      "▁no table",
+      "▁H an",
+      "▁Ha n",
+      "▁f t",
+      "▁ ft",
+      "括 号",
+      "NU LL",
+      "N ULL",
+      "▁re sort",
+      "▁res ort",
+      "其他 的",
+      "其 他的",
+      "▁h urt",
+      "▁hur t",
+      "▁hu rt",
+      "小 姐",
+      "▁d ip",
+      "▁di p",
+      "▁ dip",
+      "友 好",
+      "om orphism",
+      "omorph ism",
+      "AV E",
+      "A VE",
+      "▁un icode",
+      "▁uni code",
+      "▁ unicode",
+      "▁D u",
+      "▁ Du",
+      "▁ 操作",
+      "▁t ok",
+      "▁to k",
+      "▁ tok",
+      "▁o c",
+      "▁ oc",
+      "借 鉴",
+      "▁re ads",
+      "▁read s",
+      "▁ reads",
+      "eff ective",
+      "effect ive",
+      "Lin ux",
+      "视 野",
+      "pro gress",
+      "prog ress",
+      "活 跃",
+      "RO OT",
+      "ch art",
+      "char t",
+      "cha rt",
+      "c hart",
+      "▁J ay",
+      "▁Ja y",
+      "▁agricult ural",
+      "▁agr icultural",
+      "建 造",
+      "排 放",
+      "Ca p",
+      "C ap",
+      "da rk",
+      "dar k",
+      "d ark",
+      "ö r",
+      "▁P ier",
+      "▁Pi er",
+      "▁Pie r",
+      "界 的",
+      "▁J son",
+      "▁ Json",
+      "AT ED",
+      "ATE D",
+      "十五 日",
+      "十 五日",
+      "▁p ointer",
+      "▁point er",
+      "▁po inter",
+      "▁ pointer",
+      "am mar",
+      "amm ar",
+      "amma r",
+      "▁L earn",
+      "▁Le arn",
+      "▁Lear n",
+      "▁ Learn",
+      "民间借贷 纠纷一案中",
+      "民间借贷纠纷一案 中",
+      "民间借贷纠纷 一案中",
+      "Valid ation",
+      "Val idation",
+      "零 件",
+      "ke eper",
+      "keep er",
+      "kee per",
+      "k eeper",
+      "款 项",
+      "手 册",
+      "▁J ob",
+      "▁Jo b",
+      "▁ Job",
+      "▁prec isely",
+      "▁precise ly",
+      "嫌 疑",
+      "▁Indones ia",
+      "ia z",
+      "i az",
+      "▁C ape",
+      "▁Ca pe",
+      "▁Cap e",
+      "ex change",
+      "▁C hen",
+      "▁Ch en",
+      "▁Che n",
+      "wa s",
+      "w as",
+      "高级 人民法院",
+      "高 级人民法院",
+      "mob ile",
+      "m obile",
+      "▁attempt ed",
+      "党 建",
+      "出具 的",
+      "ex pr",
+      "exp r",
+      "▁Off icer",
+      "▁Office r",
+      "▁bound aries",
+      "打 破",
+      "重 组",
+      "不容 易",
+      "不 容易",
+      "国家 级",
+      "有一 天",
+      "有 一天",
+      "▁nucle us",
+      "▁some how",
+      "攻 坚",
+      "▁coll agen",
+      "lang le",
+      "l angle",
+      "▁hot els",
+      "▁hotel s",
+      "总 局",
+      "▁s pray",
+      "▁sp ray",
+      "▁spr ay",
+      "▁time out",
+      "▁ timeout",
+      "第五 章",
+      "▁w ider",
+      "▁wide r",
+      "▁wid er",
+      "▁wi der",
+      "Des t",
+      "De st",
+      "D est",
+      "▁in door",
+      "▁ind oor",
+      "机 电",
+      "工商行政 管理局",
+      "▁def ining",
+      "▁defin ing",
+      "▁b ases",
+      "▁bas es",
+      "▁base s",
+      "▁ba ses",
+      "▁ bases",
+      "▁un fort",
+      "▁unf ort",
+      "▁impro ves",
+      "▁improve s",
+      "▁improv es",
+      "▁The orem",
+      "▁am plitude",
+      "▁ampl itude",
+      "▁amplit ude",
+      "nM r",
+      "n Mr",
+      "ins ic",
+      "i nsic",
+      "分 成",
+      "▁phosph orylation",
+      "▁phosphory lation",
+      "已成 为",
+      "已 成为",
+      "EE E",
+      "E EE",
+      "▁re tain",
+      "▁ret ain",
+      "如果 有",
+      "弹 性",
+      "hel pers",
+      "help ers",
+      "helper s",
+      "getElement ById",
+      "象 限",
+      "举 证",
+      "▁un a",
+      "▁u na",
+      "▁ una",
+      "▁S pect",
+      "▁Sp ect",
+      "▁Spec t",
+      "▁Spe ct",
+      "▁ Spect",
+      "醉 酒",
+      "▁d f",
+      "▁ df",
+      "▁law yers",
+      "▁lawy ers",
+      "▁lawyer s",
+      "▁d p",
+      "▁ dp",
+      "▁phot ographs",
+      "▁photo graphs",
+      "▁photograph s",
+      "ur is",
+      "uri s",
+      "u ris",
+      "执行 费",
+      "ur ious",
+      "uri ous",
+      "▁d t",
+      "▁ dt",
+      "▁ch ances",
+      "▁chance s",
+      "▁chan ces",
+      "Ag ent",
+      "Age nt",
+      "A gent",
+      "▁str uggle",
+      "▁strugg le",
+      "纽 约",
+      "美好 的",
+      "美 好的",
+      "关系 的",
+      "关 系的",
+      "Cont act",
+      "找 不到",
+      "▁t ie",
+      "▁ti e",
+      "▁ tie",
+      "尽 快",
+      "▁valid ated",
+      "▁validate d",
+      "▁ validated",
+      "▁A z",
+      "▁ Az",
+      "▁me tall",
+      "▁met all",
+      "▁metal l",
+      "▁meta ll",
+      "浦 东",
+      "平方 公里",
+      "▁B ooks",
+      "▁Bo oks",
+      "▁Book s",
+      "▁ Books",
+      "og ens",
+      "ogen s",
+      "oge ns",
+      "o gens",
+      "透 过",
+      "pos ing",
+      "po sing",
+      "p osing",
+      "一 本",
+      "▁l ucky",
+      "▁luck y",
+      "▁luc ky",
+      "打 电话",
+      "Program ming",
+      "▁question naire",
+      "▁over w",
+      "▁s ki",
+      "▁sk i",
+      "▁ ski",
+      "完成 劳动任务",
+      "ma tes",
+      "mat es",
+      "mate s",
+      "m ates",
+      "pi ck",
+      "pic k",
+      "p ick",
+      "nR ead",
+      "nRe ad",
+      "n Read",
+      "▁pe ptides",
+      "▁pept ides",
+      "▁peptide s",
+      "▁L iter",
+      "▁Li ter",
+      "▁Lit er",
+      "▁Lite r",
+      "▁ Liter",
+      "西安 市",
+      "西 安市",
+      "tain ment",
+      "t ainment",
+      "据 悉",
+      "D U",
+      "▁c out",
+      "▁co ut",
+      "▁cou t",
+      "▁ cout",
+      "▁me als",
+      "▁meal s",
+      "col n",
+      "co ln",
+      "养 生",
+      "▁sub scription",
+      "▁subs cription",
+      "▁ subscription",
+      "▁b ic",
+      "▁bi c",
+      "▁ bic",
+      "nB l",
+      "n Bl",
+      "也 未",
+      "这件 事",
+      "这 件事",
+      "光 明",
+      "移 送",
+      "af ka",
+      "走 到",
+      "th ew",
+      "the w",
+      "t hew",
+      "给 自己",
+      "渐 渐",
+      "mic ro",
+      "mi cro",
+      "m icro",
+      "ha i",
+      "h ai",
+      "经本院 合法传唤",
+      "▁re ception",
+      "▁rece ption",
+      "▁recept ion",
+      "▁publish ing",
+      "▁exp enses",
+      "▁expense s",
+      "罪名 成立",
+      "▁o lig",
+      "▁ol ig",
+      "▁G R",
+      "▁ GR",
+      "nB ecause",
+      "nBe cause",
+      "n Because",
+      "iox id",
+      "i oxid",
+      "on ce",
+      "onc e",
+      "o nce",
+      "▁res idual",
+      "▁resid ual",
+      "的 范围",
+      "▁s ymmetric",
+      "▁sym metric",
+      "▁symmet ric",
+      "▁ symmetric",
+      "▁enh ancement",
+      "▁enhance ment",
+      "过 度",
+      "you tube",
+      "yout ube",
+      "y outube",
+      "▁c tx",
+      "▁ct x",
+      "▁ ctx",
+      "师 傅",
+      "抑 郁",
+      "▁b es",
+      "▁be s",
+      "▁ bes",
+      "▁Al tern",
+      "▁Alt ern",
+      "▁Alter n",
+      "▁ Altern",
+      "el astic",
+      "结束 后",
+      "已 被",
+      "LI B",
+      "L IB",
+      "▁co operation",
+      "▁cooper ation",
+      "▁_ ,",
+      "▁ _,",
+      "▁trans ient",
+      "▁trans formed",
+      "▁transform ed",
+      "▁ transformed",
+      "▁B O",
+      "▁ BO",
+      "▁d ont",
+      "▁do nt",
+      "▁don t",
+      "ul ates",
+      "ula tes",
+      "ulate s",
+      "▁D a",
+      "▁ Da",
+      "▁c ant",
+      "▁can t",
+      "▁ca nt",
+      "▁ cant",
+      "▁M ichel",
+      "▁Mich el",
+      "▁Mi chel",
+      "▁Mic hel",
+      "▁Miche l",
+      "事实 上",
+      "text color",
+      "tex tcolor",
+      "到 来",
+      "Pt r",
+      "P tr",
+      "ac ión",
+      "aci ón",
+      "a ción",
+      "Fail ed",
+      "Fa iled",
+      "F ailed",
+      "Ch at",
+      "C hat",
+      "▁Aff airs",
+      "▁Mag azine",
+      "▁lik elihood",
+      "; &",
+      "▁c ul",
+      "▁cu l",
+      "▁a nest",
+      "▁an est",
+      "▁B it",
+      "▁Bi t",
+      "▁ Bit",
+      "▁stre aming",
+      "▁stream ing",
+      "▁ streaming",
+      "代表 大会",
+      "▁P K",
+      "▁ PK",
+      "期限 为",
+      "▁ver ified",
+      "▁ verified",
+      "物业 管理",
+      "or don",
+      "ord on",
+      "尴 尬",
+      "▁l um",
+      "▁lu m",
+      "我 家",
+      "的 项目",
+      "▁C over",
+      "▁Co ver",
+      "▁Cov er",
+      "▁Cove r",
+      "▁ Cover",
+      "解答 即可",
+      "▁pack aging",
+      "▁t orch",
+      "▁tor ch",
+      "▁ torch",
+      "▁arr anged",
+      "▁arrang ed",
+      "▁arrange d",
+      "like ly",
+      "lik ely",
+      "采 纳",
+      "▁Cont roller",
+      "▁Control ler",
+      "▁ Controller",
+      "▁g onna",
+      "▁gon na",
+      "珍 惜",
+      "▁col ours",
+      "▁colour s",
+      "Bl ue",
+      "▁empir ical",
+      "▁emp irical",
+      "h l",
+      "▁d eeper",
+      "▁de eper",
+      "▁deep er",
+      "▁N ev",
+      "▁Ne v",
+      "ide spread",
+      "ides pread",
+      "三 次",
+      "OS S",
+      "O SS",
+      "▁G il",
+      "▁Gi l",
+      "▁Stat istics",
+      "▁ Statistics",
+      "__ \":\\",
+      "▁b alls",
+      "▁bal ls",
+      "▁ball s",
+      "▁ balls",
+      "▁prevent ing",
+      "派出 所",
+      "四 条",
+      "那 是",
+      "▁F actor",
+      "▁Fac tor",
+      "▁Fa ctor",
+      "▁Fact or",
+      "▁ Factor",
+      "Key s",
+      "Ke ys",
+      "▁上 海",
+      "▁ 上海",
+      "输 送",
+      "nT oday",
+      "nTo day",
+      "n Today",
+      "and ra",
+      "andr a",
+      "a ndra",
+      "▁M as",
+      "▁Ma s",
+      "▁ Mas",
+      "从 业",
+      "干 预",
+      "穿 着",
+      "nR el",
+      "nRe l",
+      "n Rel",
+      "人民法院 裁定",
+      "▁S M",
+      "▁ SM",
+      "Hol der",
+      "Hold er",
+      "H older",
+      "MO DE",
+      "MOD E",
+      "th ird",
+      "thi rd",
+      "▁S pir",
+      "▁Sp ir",
+      "▁h ero",
+      "▁he ro",
+      "▁her o",
+      "▁ hero",
+      "▁r ated",
+      "▁rate d",
+      "▁rat ed",
+      "▁ra ted",
+      "▁ rated",
+      "▁six th",
+      "三 星",
+      "hy per",
+      "hyp er",
+      "sc ience",
+      "sci ence",
+      "tu rn",
+      "tur n",
+      "t urn",
+      "▁exclusive ly",
+      "▁exclus ively",
+      "▁п о",
+      "▁ по",
+      "ol ly",
+      "oll y",
+      "▁H il",
+      "▁Hi l",
+      "▁ab ilities",
+      "▁ abilities",
+      "▁comp iled",
+      "▁compile d",
+      "▁ compiled",
+      "▁d reams",
+      "▁dream s",
+      "Do uble",
+      "D ouble",
+      "Gu i",
+      "G ui",
+      "▁pro sec",
+      "▁pros ec",
+      "▁prose c",
+      "▁explicit ly",
+      "e u",
+      "pol y",
+      "po ly",
+      "p oly",
+      "讲 述",
+      "▁E C",
+      "▁ EC",
+      "▁ent ities",
+      "▁ entities",
+      "▁abnormal ities",
+      "▁abnorm alities",
+      "审 美",
+      "不 履行",
+      "▁Hol lywood",
+      "▁Holly wood",
+      "▁d ollar",
+      "▁doll ar",
+      "▁dol lar",
+      "▁ dollar",
+      "本院 已",
+      "我 还",
+      "海 市",
+      "可以 再次申请执行",
+      "上 线",
+      "ex tend",
+      "ext end",
+      "青 山",
+      "Im port",
+      "Imp ort",
+      "方 针",
+      "财产 线索",
+      "ob acter",
+      "obac ter",
+      "▁ent itled",
+      "le ges",
+      "leg es",
+      "lege s",
+      "放 大",
+      "的 意义",
+      "的 市场",
+      "▁t ries",
+      "▁tr ies",
+      "▁tri es",
+      "▁predom inant",
+      "▁trans parent",
+      "▁ transparent",
+      "定 价",
+      "St at",
+      "S tat",
+      "ro se",
+      "ros e",
+      "r ose",
+      "▁St ream",
+      "▁Stre am",
+      "▁ Stream",
+      "▁im pl",
+      "▁imp l",
+      "▁ impl",
+      "▁ach ieving",
+      "s r",
+      "cl ock",
+      "c lock",
+      "▁saf ely",
+      "▁safe ly",
+      "执行 工作",
+      "▁B ody",
+      "▁Bo dy",
+      "▁Bod y",
+      "▁ Body",
+      "▁ag gressive",
+      "▁aggress ive",
+      "▁vulner able",
+      "em pl",
+      "emp l",
+      "nW omen",
+      "n Women",
+      "少 的",
+      "▁ 科",
+      "▁aff inity",
+      "▁ affinity",
+      "▁sub stitute",
+      "▁substit ute",
+      "▁s urname",
+      "▁sur name",
+      "▁ surname",
+      "ia ting",
+      "iat ing",
+      "i ating",
+      "犯罪 事实",
+      "▁liter ally",
+      "▁literal ly",
+      "社会信用 代码",
+      "以 往",
+      "含 义",
+      "il iary",
+      "ili ary",
+      "ilia ry",
+      "iliar y",
+      "Cont rollers",
+      "Controller s",
+      "Control lers",
+      "矩 阵",
+      "都 知道",
+      "试 试",
+      "up per",
+      "upp er",
+      "协 同",
+      "评 分",
+      "ac char",
+      "acc har",
+      "▁mid field",
+      "燃 烧",
+      "▁f ixes",
+      "▁fix es",
+      "精 致",
+      "▁Ch allenge",
+      "▁Chall enge",
+      "▁the atre",
+      "▁theat re",
+      "▁Cast le",
+      "▁Cas tle",
+      "▁accept able",
+      "▁ acceptable",
+      "▁res ume",
+      "▁ resume",
+      "户 外",
+      "▁respon ded",
+      "▁respond ed",
+      "an im",
+      "ani m",
+      "a nim",
+      "天 使",
+      "▁S ales",
+      "▁Sal es",
+      "▁Sa les",
+      "▁Sale s",
+      "▁ Sales",
+      "▁do nor",
+      "▁don or",
+      "▁oxid ative",
+      "爱 心",
+      "▁pos terior",
+      "▁post erior",
+      "▁poster ior",
+      "tl og",
+      "t log",
+      "▁A ustin",
+      "▁Aust in",
+      "▁Aus tin",
+      "▁} ,",
+      "▁ },",
+      "ogn ition",
+      "rop olitan",
+      "▁typ ing",
+      "▁ty ping",
+      "nD isc",
+      "nDis c",
+      "nDi sc",
+      "n Disc",
+      "全 世界",
+      "▁e missions",
+      "▁em issions",
+      "▁emission s",
+      "▁di abetic",
+      "▁flu orescence",
+      "▁fluores cence",
+      "▁fluor escence",
+      "说 不",
+      "▁n as",
+      "▁na s",
+      "▁ nas",
+      "term inal",
+      "termin al",
+      "▁C ategory",
+      "▁ Category",
+      "过 敏",
+      "原 判",
+      "并 有",
+      "▁F a",
+      "▁ Fa",
+      "第十 三",
+      "第 十三",
+      "▁D oc",
+      "▁Do c",
+      "▁ Doc",
+      "▁汉 族",
+      "▁ 汉族",
+      "分 工",
+      "差 别",
+      "▁o nc",
+      "▁on c",
+      "▁ onc",
+      "Not Found",
+      "▁D ifferent",
+      "▁Diff erent",
+      "▁ Different",
+      "HO ST",
+      "H OST",
+      "法律服务 所",
+      "ge nic",
+      "gen ic",
+      "g enic",
+      "ble d",
+      "bl ed",
+      "b led",
+      "O w",
+      "St ar",
+      "S tar",
+      "con tract",
+      "contr act",
+      "淀 粉",
+      "▁elect roph",
+      "▁electro ph",
+      "▁electr oph",
+      "MO D",
+      "M OD",
+      "歌 词",
+      "▁S ri",
+      "▁Sr i",
+      "▁place bo",
+      "▁plac ebo",
+      "▁nut rition",
+      "▁nutr ition",
+      "▁nutrit ion",
+      "▁ nutrition",
+      "到 现在",
+      "牙 齿",
+      "冬 天",
+      "at hon",
+      "ath on",
+      "a thon",
+      "ba bel",
+      "bab el",
+      "b abel",
+      "▁c od",
+      "▁co d",
+      "▁ cod",
+      "▁Se ason",
+      "▁Sea son",
+      "▁ Season",
+      "为 你",
+      ": `",
+      "Su r",
+      "S ur",
+      "马克 思",
+      "▁came ras",
+      "▁cam eras",
+      "▁camera s",
+      "会 出现",
+      "▁c ow",
+      "▁co w",
+      "▁ cow",
+      "积极参加 劳动",
+      "等 问题",
+      "▁S om",
+      "▁So m",
+      "S K",
+      "原来 的",
+      "原 来的",
+      "pi pe",
+      "pip e",
+      "p ipe",
+      "¶ \\",
+      "水 电",
+      "法律 工作者",
+      "▁n otion",
+      "▁not ion",
+      "▁no tion",
+      "CLUD ING",
+      "心理 学",
+      "心 理学",
+      "▁optim ized",
+      "▁optimize d",
+      "二 维",
+      "眼 前",
+      "▁int act",
+      "par ameter",
+      "param eter",
+      "para meter",
+      "paramet er",
+      "ap ple",
+      "app le",
+      "统一 社会信用代码",
+      "▁oper ates",
+      "▁operate s",
+      "▁opera tes",
+      "冒 险",
+      "▁Nig eria",
+      "▁Niger ia",
+      "▁tele phone",
+      "打 工",
+      "nP ol",
+      "nPo l",
+      "n Pol",
+      "ol ves",
+      "olve s",
+      "olv es",
+      "▁L iver",
+      "▁Li ver",
+      "▁Live r",
+      "▁Liv er",
+      "▁y ard",
+      "▁ya rd",
+      "▁ yard",
+      "ma te",
+      "mat e",
+      "m ate",
+      "▁not ification",
+      "▁ notification",
+      "知识的 选择题",
+      "et o",
+      "e to",
+      "Fi n",
+      "F in",
+      "附 录",
+      "▁( *",
+      "▁ (*",
+      "本院在执行 申请执行人",
+      "na ma",
+      "nam a",
+      "n ama",
+      "resent ation",
+      "登记 结婚",
+      "最后 一",
+      "行 情",
+      "月二十 八日",
+      "导 向",
+      "▁c ats",
+      "▁ca ts",
+      "▁cat s",
+      "▁ cats",
+      "sp l",
+      "s pl",
+      "DE BUG",
+      "on ical",
+      "onic al",
+      "oni cal",
+      "onica l",
+      "/ >",
+      "nt il",
+      "n til",
+      "▁fro zen",
+      "oc hemistry",
+      "ochem istry",
+      "有 序",
+      "▁sur viv",
+      "▁surv iv",
+      "▁e levation",
+      "▁elev ation",
+      "▁ele vation",
+      "摘 要",
+      "XM L",
+      "X ML",
+      "▁chair man",
+      "▁recruit ment",
+      "▁T ravel",
+      "▁Tr avel",
+      "▁Tra vel",
+      "▁Trav el",
+      "▁ Travel",
+      "▁V ice",
+      "▁Vi ce",
+      "▁Vic e",
+      "Pa l",
+      "P al",
+      "也就是 说",
+      "也 就是说",
+      "▁le sson",
+      "▁less on",
+      "▁les son",
+      "▁aug ment",
+      "▁au gment",
+      "陆 续",
+      "每 人",
+      "▁predict ions",
+      "▁prediction s",
+      "▁pred ictions",
+      "你 能",
+      "南京 市",
+      "▁invest igating",
+      "▁investig ating",
+      "▁spir itual",
+      "▁spirit ual",
+      "我们 将",
+      "▁t ons",
+      "▁to ns",
+      "▁ton s",
+      "▁ tons",
+      "▁H ur",
+      "▁Hu r",
+      "▁T arget",
+      "▁Tar get",
+      "▁ Target",
+      "en coding",
+      "enc oding",
+      "IN GS",
+      "ING S",
+      "▁int estinal",
+      "▁ intestinal",
+      "孤 独",
+      "ni e",
+      "n ie",
+      "▁mass age",
+      "▁rel ating",
+      "历 程",
+      "执行 完毕",
+      "▁n urses",
+      "▁nurs es",
+      "▁nur ses",
+      "▁nurse s",
+      "绩 效",
+      "数 值",
+      "▁direct ors",
+      "▁director s",
+      "▁dir ectors",
+      "▁dire ctors",
+      "▁Ph D",
+      "▁ 类别",
+      "▁t ar",
+      "▁ta r",
+      "▁ tar",
+      "的通 知",
+      "的 通知",
+      "升 高",
+      "为 准",
+      "来 讲",
+      "的 状态",
+      "日 以",
+      "▁ 地理位置",
+      "谅 解",
+      "▁Represent atives",
+      "▁Representative s",
+      "的 和",
+      "▁for ming",
+      "▁form ing",
+      "▁ forming",
+      "▁under go",
+      "▁J am",
+      "▁Ja m",
+      "特 定",
+      "▁en abling",
+      "▁S ide",
+      "▁Si de",
+      "▁Sid e",
+      "▁ Side",
+      "MM MM",
+      "ma rt",
+      "mar t",
+      "m art",
+      "Re al",
+      "R eal",
+      "▁exp los",
+      "▁expl os",
+      "▁explo s",
+      "oc o",
+      "o co",
+      "留 在",
+      "▁ag reements",
+      "▁agre ements",
+      "▁agreement s",
+      "▁agree ments",
+      "▁ 什么是",
+      "那 一",
+      "▁incred ibly",
+      "er k",
+      "e rk",
+      ")$ $\\",
+      ")$$ \\",
+      ") $$\\",
+      "or se",
+      "ors e",
+      "▁malign ant",
+      "小 程序",
+      "Serv let",
+      "or us",
+      "o rus",
+      "con v",
+      "co nv",
+      "ni k",
+      "n ik",
+      "▁att itude",
+      "块 钱",
+      "dom inal",
+      "夏 季",
+      "▁att achment",
+      "▁attach ment",
+      "▁ attachment",
+      "取 决",
+      "的 生产",
+      "reg istry",
+      "nP ersonal",
+      "nPerson al",
+      "n Personal",
+      "▁co aching",
+      "▁coach ing",
+      "▁re placing",
+      "▁replac ing",
+      "▁Aut om",
+      "▁Auto m",
+      "▁Au tom",
+      "▁ Autom",
+      "ort heast",
+      "orth east",
+      "对 策",
+      "▁S top",
+      "▁St op",
+      "▁ Stop",
+      ") <",
+      "驳回 原告",
+      "关键 词",
+      "ar do",
+      "ard o",
+      "▁New ton",
+      "▁f lights",
+      "▁fl ights",
+      "▁flight s",
+      "▁p ainted",
+      "▁pain ted",
+      "▁paint ed",
+      "nel se",
+      "nels e",
+      "n else",
+      "▁M arc",
+      "▁Mar c",
+      "▁Ma rc",
+      "▁ Marc",
+      "st ation",
+      "stat ion",
+      "sta tion",
+      "———— ————",
+      "UN T",
+      "U NT",
+      "你 想",
+      "▁K enn",
+      "▁Ke nn",
+      "▁Ken n",
+      "期 望",
+      "两 国",
+      "▁pro poses",
+      "▁propos es",
+      "▁prop oses",
+      "▁propose s",
+      "as cin",
+      "asc in",
+      "asci n",
+      "克 斯",
+      "人 人",
+      "科技 大学",
+      "▁p ocket",
+      "tr ation",
+      "tra tion",
+      "t ration",
+      "人工 智能",
+      "的中 国",
+      "的 中国",
+      "▁ad voc",
+      "▁adv oc",
+      "的 所有",
+      "▁b ios",
+      "▁bi os",
+      "▁bio s",
+      "▁b oss",
+      "▁bo ss",
+      "▁bos s",
+      "▁ boss",
+      "周 末",
+      "从选项中 选出",
+      "相 机",
+      "在 新",
+      "▁hard er",
+      "▁har der",
+      "pos al",
+      "po sal",
+      "p osal",
+      "焊 接",
+      "▁fr aud",
+      "▁fra ud",
+      "▁{ };\\",
+      "▁{} ;\\",
+      "/ $",
+      "后 再",
+      "为 您",
+      "or o",
+      "o ro",
+      "▁fl our",
+      "▁flo ur",
+      "or ms",
+      "orm s",
+      "▁work shops",
+      "▁workshop s",
+      "▁data bases",
+      "▁dat abases",
+      "▁database s",
+      "▁ databases",
+      "让 她",
+      "万 吨",
+      "比 亚",
+      "ic ut",
+      "i cut",
+      "▁D ave",
+      "▁Da ve",
+      "▁Dav e",
+      "▁B ry",
+      "▁Br y",
+      "▁b omb",
+      "▁bo mb",
+      "▁bom b",
+      "平 移",
+      "回 事",
+      "▁S ometimes",
+      "▁Some times",
+      "▁Som etimes",
+      "你 不",
+      "▁rec urrence",
+      "ho u",
+      "h ou",
+      "az ines",
+      "azine s",
+      "azi nes",
+      "▁con structor",
+      "▁construct or",
+      "▁ constructor",
+      "▁f mt",
+      "▁ fmt",
+      "▁s ew",
+      "▁se w",
+      "▁p ace",
+      "▁pa ce",
+      "▁pac e",
+      "▁ pace",
+      "▁morb idity",
+      "bi rd",
+      "bir d",
+      "b ird",
+      "The me",
+      "Th eme",
+      "T heme",
+      "Ga mma",
+      "G amma",
+      "多 久",
+      "▁Th ough",
+      "▁S peed",
+      "▁Sp eed",
+      "▁Spe ed",
+      "▁ Speed",
+      "less ly",
+      "齐 全",
+      "▁H aving",
+      "▁Ha ving",
+      "▁Hav ing",
+      "凭 证",
+      "▁H ills",
+      "▁Hill s",
+      "▁Hil ls",
+      "▁e cosystem",
+      "▁ec osystem",
+      "▁eco system",
+      "和 发展",
+      "轻 轻",
+      "▁ne ighbour",
+      "▁neigh bour",
+      "▁neighb our",
+      "下一 步",
+      "下 一步",
+      "sch ool",
+      "s chool",
+      "▁w rites",
+      "▁writ es",
+      "▁write s",
+      "▁wr ites",
+      "防 范",
+      "▁arch ive",
+      "▁arc hive",
+      "▁ archive",
+      "婚 礼",
+      "▁n od",
+      "▁no d",
+      "▁ nod",
+      "nn ame",
+      "n name",
+      "使 其",
+      "▁f ibr",
+      "▁fib r",
+      "▁fi br",
+      "▁mod ifications",
+      "▁modification s",
+      "可 怜",
+      "el son",
+      "els on",
+      "法 官",
+      "实 话",
+      "nF r",
+      "n Fr",
+      "▁elect rons",
+      "▁electron s",
+      "▁electro ns",
+      "▁electr ons",
+      "▁arg parse",
+      "▁b locked",
+      "▁block ed",
+      "玫 瑰",
+      "N R",
+      "▁j et",
+      "▁je t",
+      "▁ jet",
+      "str ucture",
+      "struct ure",
+      "▁D ocker",
+      "▁Doc ker",
+      "▁Dock er",
+      "▁ Docker",
+      "in cludes",
+      "include s",
+      "加 坡",
+      "就 开始",
+      "▁res ist",
+      "▁b ol",
+      "▁bo l",
+      "▁ bol",
+      "ja ck",
+      "jac k",
+      "j ack",
+      "路 口",
+      "波 动",
+      "了 诉讼",
+      "▁g re",
+      "▁gr e",
+      "▁ gre",
+      "▁bar riers",
+      "▁barrier s",
+      "▁barr iers",
+      "当 庭",
+      "很 久",
+      "实 战",
+      "▁pro ducers",
+      "▁produ cers",
+      "▁produce rs",
+      "▁producer s",
+      "一 带",
+      "在 开庭",
+      "初中 文化",
+      "初 中文化",
+      "标准 化",
+      "节 日",
+      "▁min imize",
+      "▁minim ize",
+      "▁imm igr",
+      "由被告 承担",
+      "Th reading",
+      "Thread ing",
+      "▁the ories",
+      "▁theor ies",
+      "▁nit rogen",
+      "▁nitro gen",
+      "fa ctor",
+      "fac tor",
+      "fact or",
+      "f actor",
+      "层 的",
+      "▁com poser",
+      "▁comp oser",
+      "▁compos er",
+      "▁compose r",
+      "▁ composer",
+      "▁destroy ed",
+      "延 伸",
+      "pr ev",
+      "pre v",
+      "p rev",
+      "Y P",
+      "渗 透",
+      "注意 事项",
+      "▁ens ures",
+      "▁ensure s",
+      "ail and",
+      "ai land",
+      "▁organ isations",
+      "▁organisation s",
+      "▁Organ ization",
+      "▁ Organization",
+      "显 著",
+      "经 销",
+      "▁H ost",
+      "▁Ho st",
+      "▁ Host",
+      "▁X ML",
+      "▁ XML",
+      "▁K itchen",
+      "▁Kit chen",
+      "借款 合同纠纷一案",
+      "借款合同 纠纷一案",
+      "orpor ated",
+      "orporate d",
+      "▁is olation",
+      "▁isol ation",
+      "▁iso lation",
+      "Ro b",
+      "R ob",
+      "▁( <",
+      "▁ (<",
+      "▁S imon",
+      "▁Sim on",
+      "▁Si mon",
+      "Not Null",
+      "ig ious",
+      "igi ous",
+      "最 初",
+      "char acter",
+      "▁ref lection",
+      "▁reflect ion",
+      "ou ver",
+      "ouve r",
+      "县 公安局",
+      "点 评",
+      "} :",
+      "桃 花",
+      "▁J ane",
+      "▁Jan e",
+      "▁Ja ne",
+      "ect ure",
+      "ec ture",
+      "dr iven",
+      "driv en",
+      "drive n",
+      "d riven",
+      "的 权利",
+      "nH igh",
+      "nHi gh",
+      "n High",
+      "演 讲",
+      "组成 的",
+      "组 成的",
+      "an ka",
+      "ank a",
+      "▁par liament",
+      "▁ ≤",
+      "▁ 软件",
+      "▁c lim",
+      "▁cl im",
+      "▁cli m",
+      "高速 公路",
+      "向本院 提起",
+      "▁t itled",
+      "▁title d",
+      "▁tit led",
+      "▁pro jection",
+      "▁project ion",
+      "▁proj ection",
+      "▁ projection",
+      "客 人",
+      "▁gr avity",
+      "▁grav ity",
+      "▁ gravity",
+      "▁M and",
+      "▁Man d",
+      "▁Ma nd",
+      "▁s ees",
+      "▁se es",
+      "▁see s",
+      "▁pro hib",
+      "de ep",
+      "d eep",
+      "等 差",
+      "á s",
+      "▁F iles",
+      "▁Fil es",
+      "▁File s",
+      "▁Fi les",
+      "▁ Files",
+      "▁sust ained",
+      "▁sustain ed",
+      "▁ar rays",
+      "▁array s",
+      "▁arr ays",
+      "▁ arrays",
+      "▁ult ras",
+      "▁ul tras",
+      "▁ultr as",
+      "▁ultra s",
+      "照 时间",
+      "▁N otes",
+      "▁No tes",
+      "▁Not es",
+      "▁Note s",
+      "▁ Notes",
+      "is ode",
+      "iso de",
+      "▁U r",
+      "▁ Ur",
+      "bad ge",
+      "▁E ag",
+      "ef fect",
+      "eff ect",
+      "以下 的",
+      "以 下的",
+      "未 经",
+      "▁enjoy ing",
+      "f p",
+      "ess ary",
+      "essa ry",
+      "▁rec ipes",
+      "▁recip es",
+      "▁recipe s",
+      "▁ recipes",
+      "一 对",
+      "le arning",
+      "learn ing",
+      "lear ning",
+      "l earning",
+      "▁B itcoin",
+      "▁Bit coin",
+      "年 纪",
+      "精 力",
+      "л ь",
+      "▁cool ing",
+      "▁def ines",
+      "▁define s",
+      "▁defin es",
+      "平 等",
+      "▁vill ages",
+      "▁village s",
+      "▁villa ges",
+      "校 区",
+      "▁p rev",
+      "▁pre v",
+      "▁pr ev",
+      "▁ prev",
+      "am y",
+      "a my",
+      "插 入",
+      "▁D or",
+      "▁Do r",
+      "▁overw hel",
+      "▁terr itory",
+      "▁territ ory",
+      "cs tring",
+      "c string",
+      "▁p ounds",
+      "▁pound s",
+      "▁L eon",
+      "▁Le on",
+      "▁Leo n",
+      "▁bel ongs",
+      "▁belong s",
+      "▁ belongs",
+      "et ition",
+      "eti tion",
+      "激 素",
+      "点 赞",
+      "Re act",
+      "中共 党员",
+      "登记 成立",
+      "▁ann iversary",
+      "▁odd s",
+      "▁od ds",
+      "▁ odds",
+      "▁Pat rick",
+      "▁Patri ck",
+      "贝 尔",
+      "nt ry",
+      "ntr y",
+      "n try",
+      "nI nst",
+      "nIn st",
+      "nIns t",
+      "n Inst",
+      "▁St orage",
+      "▁ Storage",
+      "aud io",
+      "audi o",
+      "a udio",
+      "▁f ailing",
+      "▁fail ing",
+      "▁fa iling",
+      "▁hol idays",
+      "▁holiday s",
+      "gu ide",
+      "gui de",
+      "guid e",
+      "inc ial",
+      "▁With out",
+      "▁ Without",
+      "nD em",
+      "nDe m",
+      "n Dem",
+      "▁purs ue",
+      "Mo ve",
+      "M ove",
+      "▁pass engers",
+      "▁passenger s",
+      "助 力",
+      "ca mp",
+      "cam p",
+      "c amp",
+      "min us",
+      "mi nus",
+      "m inus",
+      "取决 于",
+      "▁but tons",
+      "▁button s",
+      "▁butt ons",
+      "▁ buttons",
+      "▁Philipp ines",
+      "▁Philippine s",
+      "Il l",
+      "I ll",
+      "for ming",
+      "form ing",
+      "in valid",
+      "inv alid",
+      "损 坏",
+      "gb a",
+      "g ba",
+      "▁o ven",
+      "▁ov en",
+      "▁ oven",
+      "▁w ake",
+      "▁wa ke",
+      "的 性质",
+      "▁L arge",
+      "▁Lar ge",
+      "▁ Large",
+      "▁ir rad",
+      "▁irr ad",
+      "▁b orrow",
+      "▁bor row",
+      "假 如",
+      ")\" \\",
+      ") \"\\",
+      "nl ayout",
+      "n layout",
+      "▁$ $(",
+      "▁$$ (",
+      "▁ $$(",
+      "um atic",
+      "uma tic",
+      "umat ic",
+      "u matic",
+      "▁dist ribute",
+      "▁distribut e",
+      "select or",
+      "sel ector",
+      "▁D ead",
+      "▁De ad",
+      "▁ Dead",
+      "▁sc roll",
+      "▁scr oll",
+      "▁ scroll",
+      "▁Char acter",
+      "▁ Character",
+      "导 航",
+      "pl ain",
+      "p lain",
+      "Field s",
+      "F ields",
+      "▁B us",
+      "▁Bu s",
+      "▁ Bus",
+      "ol ars",
+      "olar s",
+      "ola rs",
+      "▁t ied",
+      "▁tie d",
+      "▁ti ed",
+      "报 酬",
+      "▁E mb",
+      "▁Em b",
+      "程度 上",
+      "▁mort gage",
+      "col ors",
+      "color s",
+      "副 主席",
+      "高 血压",
+      "▁C ome",
+      "▁Com e",
+      "▁Co me",
+      "▁ Come",
+      "供 电",
+      "Wh y",
+      "W hy",
+      "St op",
+      "S top",
+      "hi de",
+      "hid e",
+      "h ide",
+      "▁p as",
+      "▁pa s",
+      "▁ pas",
+      "([ \\'",
+      "([\\ '",
+      "( [\\'",
+      "nSports people",
+      "▁Girl s",
+      "▁Gir ls",
+      "inflamm atory",
+      "ou p",
+      "o up",
+      "▁b onds",
+      "▁bond s",
+      "▁bon ds",
+      "MP L",
+      "M PL",
+      "▁qual itative",
+      "▁ar gue",
+      "▁arg ue",
+      "▁argu e",
+      "ab b",
+      "a bb",
+      "▁rest oration",
+      "▁restor ation",
+      "om on",
+      "omo n",
+      "o mon",
+      "ens ively",
+      "ensive ly",
+      "▁purch asing",
+      "▁Veg as",
+      "▁Ve gas",
+      "时间 内",
+      "nc h",
+      "n ch",
+      "做 一个",
+      "As k",
+      "A sk",
+      "伴随 着",
+      "伴 随着",
+      "f requency",
+      "\"}, {\"",
+      "\" },{\"",
+      "▁P UB",
+      "▁PU B",
+      "▁ PUB",
+      "▁cas ual",
+      "去 除",
+      "sh aped",
+      "shape d",
+      "sha ped",
+      "组织机构 代码",
+      "遇 见",
+      "某 的",
+      "k h",
+      "权 力",
+      "▁ind icator",
+      "▁indic ator",
+      "▁ indicator",
+      "▁L yn",
+      "▁Ly n",
+      "str aint",
+      "stra int",
+      "strain t",
+      "▁Col lect",
+      "▁Coll ect",
+      "▁ Collect",
+      "更 大",
+      "ndes cribe",
+      "nde scribe",
+      "n describe",
+      "小时 候",
+      "小 时候",
+      "真 相",
+      "到庭参加 了诉讼",
+      "▁Fore ign",
+      "▁p rost",
+      "▁pro st",
+      "▁pr ost",
+      "▁pros t",
+      "▁D uke",
+      "▁Du ke",
+      "▁F ollow",
+      "▁Fol low",
+      "▁ Follow",
+      "▁Se attle",
+      "▁Sea ttle",
+      "▁Seat tle",
+      "▁w idespread",
+      "▁wide spread",
+      "发展 有限公司",
+      "并 于",
+      "阶 级",
+      "▁surve illance",
+      "民事 诉讼",
+      "Mo re",
+      "Mor e",
+      "M ore",
+      "看 守",
+      "na dd",
+      "nad d",
+      "n add",
+      "▁mod ulation",
+      "Pat tern",
+      "▁suscept ibility",
+      "▁sim plify",
+      "▁simpl ify",
+      "和计划生育 局",
+      "▁pers istent",
+      "▁persist ent",
+      "有关 部门",
+      "pie ce",
+      "p iece",
+      "document class",
+      "ss h",
+      "s sh",
+      "▁B lu",
+      "▁Bl u",
+      "ay ing",
+      "a ying",
+      "幅 度",
+      "▁R ange",
+      "▁Ra nge",
+      "▁Ran ge",
+      "▁ Range",
+      "▁r iding",
+      "▁ri ding",
+      "▁rid ing",
+      "▁ riding",
+      "▁[ ],\\",
+      "▁[] ,\\",
+      "▁[], \\",
+      "▁ [],\\",
+      "▁initi atives",
+      "▁initiative s",
+      "多少 钱",
+      "war ning",
+      "warn ing",
+      "w arning",
+      "该公司 董事长",
+      "该 公司董事长",
+      "▁par ish",
+      "▁pa rish",
+      "▁pari sh",
+      "▁list ings",
+      "▁listing s",
+      "on de",
+      "ond e",
+      "o nde",
+      "▁im paired",
+      "▁imp aired",
+      "▁impair ed",
+      "▁cons isted",
+      "▁consist ed",
+      "眼 镜",
+      "▁fib ro",
+      "▁fibr o",
+      "▁fi bro",
+      "▁di vor",
+      "▁div or",
+      "▁B oot",
+      "▁Bo ot",
+      "▁ Boot",
+      "旁 边",
+      "▁process or",
+      "▁ processor",
+      "▁extra cellular",
+      "▁extr acellular",
+      "▁extrac ellular",
+      "失信被执行人 名单",
+      "▁ac company",
+      "▁accomp any",
+      "▁accompan y",
+      "Att ributes",
+      "Attribute s",
+      "at io",
+      "ati o",
+      "▁De tails",
+      "▁Det ails",
+      "▁Detail s",
+      "▁ Details",
+      "ab i",
+      "a bi",
+      "最重要 的",
+      "最 重要的",
+      "ip pi",
+      "ipp i",
+      "ow ed",
+      "owe d",
+      "o wed",
+      "热 门",
+      "未到庭 参加诉讼",
+      "未 到庭参加诉讼",
+      "nE ven",
+      "nEv en",
+      "n Even",
+      "奉 献",
+      "防 火",
+      "nP olit",
+      "nPol it",
+      "nPo lit",
+      "ne ed",
+      "nee d",
+      "n eed",
+      "▁E L",
+      "▁ EL",
+      "统计学 意义",
+      "▁F inance",
+      "▁Fin ance",
+      "天 空",
+      "pre dict",
+      "pred ict",
+      "▁else where",
+      "^ ^",
+      "EN CE",
+      "Load ing",
+      "Lo ading",
+      "▁associ ate",
+      "▁e merged",
+      "▁emer ged",
+      "▁emerg ed",
+      "▁emerge d",
+      "相 处",
+      "CO MM",
+      "COM M",
+      "戏 剧",
+      "nT ax",
+      "n Tax",
+      "so ng",
+      "son g",
+      "s ong",
+      "▁kin etic",
+      "▁ki netic",
+      "B JECT",
+      "▁init ialize",
+      "▁initial ize",
+      "▁ initialize",
+      "▁re model",
+      "▁rem odel",
+      "超 声",
+      "▁can vas",
+      "▁ canvas",
+      "▁comp ilation",
+      "退 还",
+      "▁compet itions",
+      "▁competition s",
+      "▁competit ions",
+      "婚姻 法",
+      "tem pt",
+      "temp t",
+      "t empt",
+      "ee p",
+      "e ep",
+      "旅 客",
+      "▁bu yers",
+      "▁buy ers",
+      "▁buyer s",
+      "MA G",
+      "M AG",
+      "处 处",
+      "期间 不",
+      "ha ng",
+      "han g",
+      "h ang",
+      "生态 环境",
+      "初 期",
+      "en ez",
+      "ene z",
+      "e nez",
+      "▁size of",
+      "▁ sizeof",
+      "▁inv asive",
+      "▁ invasive",
+      "宁波 市",
+      "列 出",
+      "▁ 有限责任公司",
+      "Init ial",
+      "▁F ourier",
+      "▁Four ier",
+      "Reg istry",
+      "煤 炭",
+      "▁小 说",
+      "▁ 小说",
+      "in ea",
+      "ine a",
+      "i nea",
+      "tr e",
+      "t re",
+      "▁ox idation",
+      "▁oxid ation",
+      "▁B ible",
+      "▁Bi ble",
+      "▁Bib le",
+      "▁s id",
+      "▁si d",
+      "▁ sid",
+      "▁l aptop",
+      "▁lap top",
+      "保 温",
+      "Y ear",
+      "约 束",
+      "▁B alt",
+      "▁Bal t",
+      "▁Ba lt",
+      "进 了",
+      "elen ium",
+      "▁F I",
+      "▁ FI",
+      "cd n",
+      "c dn",
+      "war n",
+      "wa rn",
+      "w arn",
+      "an ion",
+      "ani on",
+      "附 属",
+      "▁D ur",
+      "▁Du r",
+      "▁represent atives",
+      "▁representative s",
+      "▁g aming",
+      "▁gam ing",
+      "▁ga ming",
+      "产 地",
+      "▁W E",
+      "▁ WE",
+      "▁Tok yo",
+      "女 朋友",
+      "▁p references",
+      "▁pre ferences",
+      "▁pr eferences",
+      "▁prefer ences",
+      "▁preference s",
+      "▁ preferences",
+      "和 你",
+      "交 警",
+      "onom ous",
+      "家 属",
+      "▁G A",
+      "▁ GA",
+      "第六 十七条",
+      "第六十七 条",
+      "守 护",
+      "Col lect",
+      "Coll ect",
+      "▁v oters",
+      "▁vote rs",
+      "▁vot ers",
+      "▁vo ters",
+      "▁voter s",
+      "▁select ing",
+      "▁sel ecting",
+      "▁roll ing",
+      "▁ rolling",
+      "▁P E",
+      "▁ PE",
+      "▁Pl ant",
+      "▁Plan t",
+      "The se",
+      "Th ese",
+      "trans lation",
+      "transl ation",
+      "服 从",
+      "市场 上",
+      "市 场上",
+      "st roke",
+      "str oke",
+      "台 北",
+      "日立案 执行",
+      "日 立案执行",
+      "高 峰",
+      "美丽 的",
+      "美 丽的",
+      "}) $",
+      "} )$",
+      "▁Jac ob",
+      "▁Ja cob",
+      "全 县",
+      "只 不过",
+      "花 费",
+      "本案 在执行过程中",
+      "协 作",
+      "pl aces",
+      "place s",
+      "plac es",
+      "火 锅",
+      "种 种",
+      "家 用",
+      "build ing",
+      "▁ 郑",
+      "▁hand led",
+      "▁handle d",
+      "▁ handled",
+      "问 我",
+      "▁Add itional",
+      "▁Addition al",
+      "▁ Additional",
+      "电 缆",
+      "\"\\ \\",
+      "\" \\\\",
+      "▁N ik",
+      "▁Ni k",
+      "al am",
+      "ala m",
+      "a lam",
+      "▁R oy",
+      "▁Ro y",
+      "ock ets",
+      "ocket s",
+      "▁S uccess",
+      "▁Su ccess",
+      "▁ Success",
+      "▁insp ection",
+      "▁inspect ion",
+      "▁\" \",\\",
+      "▁\"\" ,\\",
+      "▁\"\", \\",
+      "便 捷",
+      "科 普",
+      "▁among st",
+      "查 找",
+      "▁A RE",
+      "▁AR E",
+      "▁ ARE",
+      "▁cont acts",
+      "▁contact s",
+      "▁ contacts",
+      "ar ith",
+      "ari th",
+      "▁spect rom",
+      "▁spectro m",
+      "V O",
+      "▁Part icip",
+      "▁ Particip",
+      "骨 折",
+      "请 给出",
+      "en h",
+      "e nh",
+      "rb an",
+      "r ban",
+      "▁tr iple",
+      "▁tri ple",
+      "▁trip le",
+      "▁G ET",
+      "▁GE T",
+      "▁ GET",
+      "▁H oly",
+      "▁Hol y",
+      "▁Ho ly",
+      "▁D ak",
+      "▁Da k",
+      "en viron",
+      "env iron",
+      "ac les",
+      "acle s",
+      "acl es",
+      "a cles",
+      "▁M odule",
+      "▁Mod ule",
+      "▁Mo dule",
+      "▁ Module",
+      "ab stract",
+      "abs tract",
+      "a bstract",
+      "▁ind icators",
+      "▁indic ators",
+      "▁indicator s",
+      "紧 密",
+      "pol icy",
+      "p olicy",
+      "充 足",
+      "View Model",
+      "▁he patic",
+      "▁hepat ic",
+      "}{ (",
+      "} {(",
+      "是我 们",
+      "是 我们",
+      "▁e q",
+      "▁ eq",
+      "社 保",
+      "鉴 于",
+      "复杂 度",
+      "▁rec ru",
+      "气 氛",
+      "▁H P",
+      "▁ HP",
+      "▁at omic",
+      "▁atom ic",
+      "▁ atomic",
+      "▁th yroid",
+      "▁thy roid",
+      "▁t rem",
+      "▁tr em",
+      "▁tre m",
+      "▁long itudinal",
+      "▁sugg estion",
+      "▁suggest ion",
+      "家 中",
+      "ev al",
+      "eva l",
+      "e val",
+      "▁max imal",
+      "▁maxim al",
+      "被 称为",
+      "▁tra ject",
+      "民 国",
+      "fl ix",
+      "nM ETHODS",
+      "布 置",
+      "第一百四十五 条第一款之规定",
+      "第一百四十五条第一款 之规定",
+      "▁Atl anta",
+      "fo ur",
+      "f our",
+      "大 战",
+      "ack ets",
+      "acket s",
+      "arg uments",
+      "argument s",
+      "下 跌",
+      "幻 想",
+      "PI O",
+      "P IO",
+      "四 边形",
+      "▁ident ifier",
+      "▁ identifier",
+      "* /",
+      "▁ 郭",
+      "▁dep osit",
+      "▁depos it",
+      "▁element ary",
+      "el fare",
+      "elf are",
+      "▁Ex planation",
+      "复杂 的",
+      "考查 了",
+      "本院 予以确认",
+      "本院予以 确认",
+      "填 写",
+      "侵 权",
+      "▁b ull",
+      "▁bu ll",
+      "▁bul l",
+      "▁ bull",
+      "假 期",
+      "艺术 家",
+      "▁ex ercises",
+      "▁exerc ises",
+      "▁exercise s",
+      "还是 要",
+      "还 是要",
+      "Ma ke",
+      "M ake",
+      "′ (",
+      "nF e",
+      "n Fe",
+      "Con verter",
+      "Convert er",
+      "▁volunte er",
+      "少 许",
+      "王 子",
+      "semb l",
+      "sem bl",
+      "re tion",
+      "ret ion",
+      "is p",
+      "i sp",
+      "el and",
+      "ela nd",
+      "e land",
+      "▁T ony",
+      "▁To ny",
+      "▁Ton y",
+      "}} $\\",
+      "}}$ \\",
+      "} }$\\",
+      "给 人",
+      "▁Ass istant",
+      "陷 入",
+      "is cal",
+      "isc al",
+      "i scal",
+      "Per son",
+      "Pers on",
+      "P erson",
+      "首 都",
+      "▁g ut",
+      "▁gu t",
+      "ot ing",
+      "otin g",
+      "oti ng",
+      "o ting",
+      "▁prom oter",
+      "▁promot er",
+      "▁promote r",
+      "▁promo ter",
+      "不 违反",
+      "是真 的",
+      "是 真的",
+      "ter r",
+      "te rr",
+      "t err",
+      "▁c rystall",
+      "▁cryst all",
+      "▁crystal l",
+      "▁G er",
+      "▁Ge r",
+      "qu ote",
+      "quot e",
+      "quo te",
+      "▁rel igion",
+      "▁relig ion",
+      "std err",
+      "建立 了",
+      "i ć",
+      "后 悔",
+      "in ine",
+      "ini ne",
+      "inin e",
+      "i nine",
+      "宣 判",
+      "riv acy",
+      "roph ic",
+      "over nment",
+      "overn ment",
+      "有 钱",
+      "月 底",
+      "▁re vision",
+      "▁rev ision",
+      "▁revis ion",
+      "▁ revision",
+      "▁T y",
+      "▁ Ty",
+      "▁Java Script",
+      "▁ JavaScript",
+      "▁record ings",
+      "▁recording s",
+      "▁B B",
+      "▁ BB",
+      "和解 协议",
+      "▁Tem ple",
+      "▁Temp le",
+      "库 存",
+      "琐 事",
+      "▁ ∴",
+      "▁comp leting",
+      "▁complet ing",
+      "▁compl eting",
+      "▁chrom osome",
+      "▁chromos ome",
+      "▁S tre",
+      "▁St re",
+      "▁Str e",
+      "▁rom antic",
+      "同一 个",
+      "同 一个",
+      "指 令",
+      "微 笑",
+      "er able",
+      "era ble",
+      "▁发 照时间",
+      "▁c otton",
+      "▁cot ton",
+      "▁kil omet",
+      "▁ind ependence",
+      "▁independ ence",
+      "▁g astric",
+      "▁gastr ic",
+      "▁gast ric",
+      "▁de rive",
+      "▁der ive",
+      "▁deriv e",
+      "▁ derive",
+      "▁C redit",
+      "▁Cr edit",
+      "▁Cre dit",
+      "▁Cred it",
+      "▁ Credit",
+      "os in",
+      "osi n",
+      "o sin",
+      "其行为 已构成",
+      "▁o verse",
+      "▁over se",
+      "▁overs e",
+      "le e",
+      "l ee",
+      "▁comp eting",
+      "▁compet ing",
+      "的 教学",
+      "鉴定 表",
+      "▁ac oustic",
+      "▁ acoustic",
+      "m q",
+      "一 遍",
+      "本案 按",
+      "全 书",
+      "惊 喜",
+      "ia ng",
+      "ian g",
+      "i ang",
+      "▁pod cast",
+      "▁aut hentication",
+      "▁authent ication",
+      "▁authentic ation",
+      "▁ authentication",
+      "▁P oly",
+      "▁Pol y",
+      "▁Po ly",
+      "▁ Poly",
+      "og y",
+      "o gy",
+      "op sis",
+      "ops is",
+      "o psis",
+      "▁S witzerland",
+      "插 件",
+      "▁bal anced",
+      "▁balance d",
+      "▁ balanced",
+      "二 份",
+      "试 点",
+      "▁And erson",
+      "▁Anders on",
+      "▁arr angement",
+      "▁arrang ement",
+      "▁arrange ment",
+      "Sc ore",
+      "S core",
+      "▁Product s",
+      "▁Produ cts",
+      "▁ Products",
+      "▁B ureau",
+      "▁act ivate",
+      "▁activ ate",
+      "▁ activate",
+      "▁squ ares",
+      "▁square s",
+      "▁L IMIT",
+      "▁LI MIT",
+      "▁ LIMIT",
+      "▁network ing",
+      "▁net working",
+      "▁C ov",
+      "▁Co v",
+      "▁ Cov",
+      "▁M oscow",
+      "▁Mos cow",
+      "▁de fect",
+      "▁def ect",
+      "▁defe ct",
+      "面 粉",
+      "细 致",
+      "▁M s",
+      "▁ Ms",
+      "二 被告",
+      "指 示",
+      "ac ion",
+      "aci on",
+      "acio n",
+      "▁no vels",
+      "▁novel s",
+      "▁nov els",
+      "说 我",
+      "财 物",
+      "太 原",
+      "▁I ron",
+      "▁Ir on",
+      "若干问题 的解释",
+      "若干问题的 解释",
+      "度 假",
+      "▁ac quire",
+      "▁acqu ire",
+      "脸 上",
+      "be nch",
+      "ben ch",
+      "b ench",
+      "▁remark able",
+      "S ql",
+      "频 繁",
+      "ig g",
+      "i gg",
+      "▁w rapper",
+      "▁wr apper",
+      "▁wrap per",
+      "▁ wrapper",
+      "善 良",
+      "▁b ills",
+      "▁bill s",
+      "▁bil ls",
+      "▁m arrow",
+      "▁mar row",
+      "▁inter ference",
+      "▁interfer ence",
+      "▁interfere nce",
+      "▁F red",
+      "▁Fr ed",
+      "▁Fre d",
+      "▁expl oration",
+      "▁explo ration",
+      "▁explor ation",
+      "▁o v",
+      "▁ ov",
+      "es lint",
+      "一 声",
+      "▁\\' %",
+      "▁ \\'%",
+      "▁jour nals",
+      "▁journal s",
+      "ol ip",
+      "oli p",
+      "o lip",
+      "成 效",
+      "▁c ru",
+      "▁cr u",
+      "assert True",
+      "o T",
+      "域 名",
+      "对 面",
+      "al y",
+      "a ly",
+      "▁A uthors",
+      "▁Author s",
+      "▁Auth ors",
+      "job s",
+      "jo bs",
+      "j obs",
+      "ra ted",
+      "rate d",
+      "rat ed",
+      "r ated",
+      "nS w",
+      "n Sw",
+      "▁re habilitation",
+      "mu s",
+      "m us",
+      "日 提出",
+      "担 当",
+      "第二百四十 二条",
+      "第二百四十二 条",
+      "▁d é",
+      "▁Im ages",
+      "▁Image s",
+      "▁ Images",
+      "▁P red",
+      "▁Pr ed",
+      "▁Pre d",
+      "▁ Pred",
+      "▁A li",
+      "▁Al i",
+      "▁ Ali",
+      "▁d rives",
+      "▁dr ives",
+      "▁dri ves",
+      "▁driv es",
+      "▁drive s",
+      "▁inhib itory",
+      "▁inhibitor y",
+      "▁inhibit ory",
+      "注 销",
+      "第二 个",
+      "第 二个",
+      "LE TE",
+      "LET E",
+      "Com par",
+      "Comp ar",
+      "正常 的",
+      "只 需",
+      "▁op position",
+      "▁K it",
+      "▁Ki t",
+      "▁ Kit",
+      "L anguage",
+      "▁int ellectual",
+      "A X",
+      "▁f acial",
+      "▁fac ial",
+      "▁ facial",
+      "▁accompl ished",
+      "▁accomp lished",
+      "▁accomplish ed",
+      "In ject",
+      "正 义",
+      "▁解 释",
+      "▁ 解释",
+      "判 定",
+      "ur u",
+      "u ru",
+      "忽 略",
+      "必要 的",
+      "▁sk illed",
+      "▁skill ed",
+      "▁feas ibility",
+      "履 历",
+      "接 待",
+      "信息 技术",
+      "▁Gree ce",
+      "微博 热度",
+      "并发 症",
+      "▁cabin et",
+      "▁cab inet",
+      "▁F if",
+      "▁Fi f",
+      "地 说",
+      "nCons ider",
+      "冷 却",
+      "▁M ix",
+      "▁Mi x",
+      "▁ Mix",
+      "▁O P",
+      "▁ OP",
+      "」 \\",
+      "▁C OPY",
+      "▁CO PY",
+      "▁ COPY",
+      "▁r um",
+      "▁ru m",
+      "▁ rum",
+      "▁Ed itor",
+      "▁Edit or",
+      "▁ Editor",
+      "ug u",
+      "u gu",
+      "▁s ensory",
+      "▁sens ory",
+      "▁sensor y",
+      "得到 的",
+      "得 到的",
+      "▁Ap plications",
+      "▁Application s",
+      "▁Appl ications",
+      "▁ Applications",
+      "▁M ade",
+      "▁Mad e",
+      "▁Ma de",
+      "▁ Made",
+      "mit h",
+      "mi th",
+      "m ith",
+      "UR I",
+      "U RI",
+      "▁val ve",
+      "▁inv ite",
+      "al one",
+      "alo ne",
+      "alon e",
+      "的原 则",
+      "的 原则",
+      "▁reg ime",
+      "创 始",
+      "信 仰",
+      "逝 世",
+      "▁c oin",
+      "▁co in",
+      "▁ coin",
+      "it ories",
+      "itor ies",
+      "ito ries",
+      "ation ally",
+      "ational ly",
+      "第三 方",
+      "第 三方",
+      "备 案",
+      "▁up grad",
+      "第四 节",
+      "▁dr inks",
+      "▁drink s",
+      "早 已",
+      "▁char set",
+      "▁chars et",
+      "▁ charset",
+      "/ \"",
+      "Pro v",
+      "Pr ov",
+      "P rov",
+      "▁se xy",
+      "▁sex y",
+      "自身 的",
+      "自 身的",
+      "▁form ats",
+      "▁format s",
+      "▁ formats",
+      "▁B utton",
+      "▁But ton",
+      "▁ Button",
+      "▁E S",
+      "▁ ES",
+      "剩 下",
+      "], \"",
+      "] ,\"",
+      "就 被",
+      "iz ational",
+      "ization al",
+      "和 他",
+      "▁Al len",
+      "▁All en",
+      "var e",
+      "va re",
+      "v are",
+      "▁sh oulder",
+      "▁should er",
+      "的 确",
+      "nB el",
+      "nBe l",
+      "n Bel",
+      "▁ur inary",
+      "▁Dis play",
+      "▁ Display",
+      "d j",
+      "product ive",
+      "prod uctive",
+      "ij ing",
+      "iji ng",
+      "i jing",
+      "la ted",
+      "late d",
+      "lat ed",
+      "l ated",
+      "▁Mo ore",
+      "▁Moor e",
+      "Ev ery",
+      "E very",
+      "▁strict ly",
+      "▁fin est",
+      "▁fine st",
+      "▁fi nest",
+      "▁fines t",
+      "认 同",
+      "转化 为",
+      "转 化为",
+      "▁a est",
+      "均 有",
+      "▁u rine",
+      "▁ur ine",
+      "▁uri ne",
+      "MA P",
+      "M AP",
+      "▁mot ivation",
+      "▁motiv ation",
+      "sample s",
+      "sam ples",
+      "s amples",
+      "边 长",
+      "am ide",
+      "ami de",
+      "am azon",
+      "ama zon",
+      "▁c ater",
+      "▁ca ter",
+      "▁cat er",
+      "▁cate r",
+      "▁cont ributes",
+      "▁contribut es",
+      "▁contribute s",
+      "in cess",
+      "ince ss",
+      "inc ess",
+      "inces s",
+      "恶 心",
+      "▁Cons ult",
+      "▁Part ners",
+      "▁Partner s",
+      "▁val idity",
+      "▁valid ity",
+      "▁g em",
+      "▁ge m",
+      "▁ gem",
+      "提 请",
+      "题 材",
+      "▁t ape",
+      "▁ta pe",
+      "▁tap e",
+      "▁ tape",
+      "市场 监督管理局",
+      "▁ad option",
+      "▁adopt ion",
+      "▁adop tion",
+      "版 权",
+      "投 影",
+      "▁dram atic",
+      "▁drama tic",
+      "▁m others",
+      "▁mother s",
+      "▁mo thers",
+      "▁moth ers",
+      "▁re late",
+      "▁rel ate",
+      "AC H",
+      "A CH",
+      "▁ 海",
+      "ID E",
+      "I DE",
+      "大 脑",
+      "普通 的",
+      "普 通的",
+      "▁C ourse",
+      "▁Co urse",
+      "▁Cour se",
+      "▁ Course",
+      "真 心",
+      "th ree",
+      "thr ee",
+      "pu s",
+      "p us",
+      "is tical",
+      "ist ical",
+      "istic al",
+      "优 雅",
+      "防 御",
+      "▁I SO",
+      "▁IS O",
+      "▁ ISO",
+      "▁a gon",
+      "▁ag on",
+      "▁ago n",
+      "▁ agon",
+      "▁co inc",
+      "▁coin c",
+      "▁al ignment",
+      "▁align ment",
+      "▁ alignment",
+      "▁S W",
+      "▁ SW",
+      "▁sub group",
+      "泰 国",
+      "▁equ ity",
+      "▁eq uity",
+      "符合 有关法律规定",
+      "▁vir uses",
+      "▁virus es",
+      "▁over look",
+      "▁overl ook",
+      "也 将",
+      "绍 兴",
+      "通 风",
+      "▁util ization",
+      "▁ut ilization",
+      "▁contin uously",
+      "▁continu ously",
+      "▁continuous ly",
+      "液 体",
+      "sh adow",
+      "sha dow",
+      "古 典",
+      "ar ding",
+      "ard ing",
+      "ardi ng",
+      "曾 任",
+      "ter y",
+      "te ry",
+      "t ery",
+      "这 场",
+      "▁par agraph",
+      "▁para graph",
+      "▁ paragraph",
+      "nG ener",
+      "nGe ner",
+      "nGen er",
+      "n Gener",
+      "q q",
+      "号 为",
+      "▁surv ive",
+      "▁surviv e",
+      "的 具体",
+      "的 目标",
+      "ustr alian",
+      "ustral ian",
+      "ustralia n",
+      "日 志",
+      "阳 性",
+      "ag an",
+      "aga n",
+      "a gan",
+      "Hi story",
+      "Hist ory",
+      "H istory",
+      "▁dep artments",
+      "▁department s",
+      "▁depart ments",
+      "▁membr anes",
+      "▁membrane s",
+      "sc ene",
+      "▁period ic",
+      "▁peri odic",
+      "▁ periodic",
+      "▁b ooking",
+      "▁book ing",
+      "▁bo oking",
+      "▁ booking",
+      "id i",
+      "i di",
+      "▁Malays ia",
+      "这种 情况",
+      "去 做",
+      "▁Qu ick",
+      "▁ Quick",
+      "除 法",
+      "格 兰",
+      "▁B run",
+      "▁Br un",
+      "▁Bru n",
+      "想 起",
+      "op tim",
+      "opt im",
+      "拿 着",
+      "Sp ring",
+      "S pring",
+      "代 替",
+      "dict ion",
+      "di ction",
+      "dic tion",
+      "d iction",
+      "▁bo ards",
+      "▁board s",
+      "▁ boards",
+      "茶 叶",
+      "的 样子",
+      "但是 我",
+      "但 是我",
+      "▁s ha",
+      "▁sh a",
+      "▁ sha",
+      "两 点",
+      "民 政",
+      "两 天",
+      "^ *",
+      "dash board",
+      "d ashboard",
+      "iv ities",
+      "ivi ties",
+      "▁o ct",
+      "▁oc t",
+      "▁ oct",
+      "▁B arn",
+      "▁Bar n",
+      "▁Ba rn",
+      "categor ies",
+      "c ategories",
+      "▁ent repreneur",
+      "▁entreprene ur",
+      "og r",
+      "o gr",
+      "Base d",
+      "Bas ed",
+      "Ba sed",
+      "B ased",
+      "▁En joy",
+      "定 了",
+      "▁O klahoma",
+      "▁p reference",
+      "▁pre ference",
+      "▁pr eference",
+      "▁prefer ence",
+      "▁smart phone",
+      "M s",
+      "全新 的",
+      "全 新的",
+      "E Y",
+      "▁ant ioxid",
+      "▁anti oxid",
+      "reg ulation",
+      "谈 判",
+      "▁M rs",
+      "▁Mr s",
+      "▁s lide",
+      "▁sl ide",
+      "▁ slide",
+      "▁Id eas",
+      "▁Ide as",
+      "▁Idea s",
+      "▁Ne uro",
+      "nV iew",
+      "n View",
+      "in us",
+      "i nus",
+      "▁mat plotlib",
+      "▁ matplotlib",
+      "nR oute",
+      "n Route",
+      "帮 忙",
+      "▁ret ention",
+      "判决 发生法律效力后",
+      "▁comb ines",
+      "▁combin es",
+      "▁combine s",
+      "王某 某",
+      "王 某某",
+      "▁he aded",
+      "▁head ed",
+      "▁ headed",
+      "一 台",
+      "an notations",
+      "annot ations",
+      "annotation s",
+      "▁> \\",
+      "▁ >\\",
+      "mod ified",
+      "不 说",
+      "▁c urious",
+      "▁cur ious",
+      "▁lim itation",
+      "▁limit ation",
+      "实 地",
+      "▁图书 目录",
+      "Render er",
+      "人 与",
+      "单 一",
+      "sy nt",
+      "syn t",
+      "s ynt",
+      "▁St ack",
+      "▁Sta ck",
+      "▁ Stack",
+      "dis ci",
+      "di sci",
+      "disc i",
+      "料 的",
+      "两 者",
+      "住 了",
+      "▁Gu ys",
+      "▁Guy s",
+      "书 中",
+      "▁to night",
+      "▁ton ight",
+      "Ma il",
+      "M ail",
+      "▁d ense",
+      "▁de nse",
+      "▁den se",
+      "▁dens e",
+      "▁ dense",
+      "▁Spec ifically",
+      "▁Specific ally",
+      "▁ θ",
+      "场 地",
+      "▁encounter ed",
+      "pro c",
+      "pr oc",
+      "p roc",
+      "▁K elly",
+      "▁Kel ly",
+      "▁Kell y",
+      "а т",
+      "▁re active",
+      "▁react ive",
+      "▁ reactive",
+      "增 值",
+      "市 区",
+      "把 自己",
+      "年 初",
+      "执行工作 若干问题的规定",
+      "▁c c",
+      "▁ cc",
+      "re spons",
+      "res pons",
+      "respon s",
+      "resp ons",
+      "▁adjust ed",
+      "▁adj usted",
+      "▁ adjusted",
+      ">\\ '",
+      "> \\'",
+      "▁And re",
+      "▁Andr e",
+      "▁p ushed",
+      "▁push ed",
+      "id ase",
+      "ida se",
+      "idas e",
+      "lo id",
+      "l oid",
+      "Ar ch",
+      "Arc h",
+      "▁Enter tainment",
+      "生 气",
+      "表 彰",
+      "▁p acked",
+      "▁pack ed",
+      "▁ packed",
+      "er mat",
+      "erm at",
+      "▁cap tion",
+      "▁ca ption",
+      "▁capt ion",
+      "▁ caption",
+      "的 声音",
+      "健 全",
+      "等差 数列",
+      "▁relax ation",
+      "六 条",
+      "disc ipl",
+      "disci pl",
+      "凌 晨",
+      "相 遇",
+      "▁b earing",
+      "▁be aring",
+      "▁bear ing",
+      "▁ bearing",
+      "▁ 何",
+      "nc c",
+      "n cc",
+      "该公司 总经理",
+      "该 公司总经理",
+      "▁S ummary",
+      "▁Summ ary",
+      "▁ Summary",
+      "▁别 名",
+      "▁ 别名",
+      "在 美国",
+      "在 互联网",
+      "第三 人",
+      "第 三人",
+      "ro b",
+      "r ob",
+      "▁r unner",
+      "▁run ner",
+      "▁ runner",
+      "nI nit",
+      "nIn it",
+      "n Init",
+      "遗 憾",
+      "▁Nor wegian",
+      "▁les ion",
+      "▁car pet",
+      "▁carp et",
+      "Not ification",
+      "新 浪",
+      "▁( )\\",
+      "▁() \\",
+      "▁ ()\\",
+      "▁enthus i",
+      "发 酵",
+      "将 来",
+      "▁M ust",
+      "▁Mus t",
+      "▁Mu st",
+      "▁ Must",
+      "积极参加 思想",
+      "春 风",
+      "in herit",
+      "的 意思",
+      "▁L aure",
+      "▁La ure",
+      "▁Laur e",
+      "we alth",
+      "▁veget ables",
+      "▁vegetable s",
+      "Fold er",
+      "F older",
+      "nexport s",
+      "nex ports",
+      "n exports",
+      "pro duction",
+      "product ion",
+      "produ ction",
+      "prod uction",
+      "p roduction",
+      "air y",
+      "ai ry",
+      "a iry",
+      "▁D epend",
+      "▁De pend",
+      "▁Dep end",
+      "▁ Depend",
+      "案 后",
+      "custom er",
+      "de lay",
+      "del ay",
+      "长 大",
+      "▁sold iers",
+      "▁soldier s",
+      "uit ion",
+      "ui tion",
+      "u ition",
+      "褐 色",
+      "nT HE",
+      "nTH E",
+      "n THE",
+      "iat ive",
+      "i ative",
+      "web pack",
+      "▁Mat hematical",
+      "▁Mathematic al",
+      "▁Mathematica l",
+      "▁Commun ications",
+      "▁Communication s",
+      "▁Ed ge",
+      "▁ Edge",
+      "便 于",
+      "gu n",
+      "g un",
+      "▁calcul ator",
+      "▁calc ulator",
+      "▁H old",
+      "▁Hol d",
+      "▁Ho ld",
+      "▁ Hold",
+      "iot ics",
+      "iotic s",
+      "i otics",
+      "中国人民 银行",
+      "nC ast",
+      "n Cast",
+      "▁g astro",
+      "▁gas tro",
+      "▁gastr o",
+      "▁gast ro",
+      "▁Al ber",
+      "▁Alb er",
+      "一 元",
+      "月 利率",
+      "ud ge",
+      "mi ll",
+      "mil l",
+      "m ill",
+      "后 台",
+      "al ine",
+      "ali ne",
+      "alin e",
+      "a line",
+      "▁_ _____",
+      "▁__ ____",
+      "▁____ __",
+      "▁ ______",
+      "的 爱",
+      "职业技术 教育",
+      "开 幕",
+      "美 味",
+      "▁o var",
+      "▁ov ar",
+      "Pre fix",
+      "Pref ix",
+      "P refix",
+      "▁B lood",
+      "▁Bl ood",
+      "▁Blo od",
+      "sign al",
+      "sig nal",
+      "疑 问",
+      "▁sophistic ated",
+      "▁con vex",
+      "▁conv ex",
+      "▁conve x",
+      "▁ convex",
+      "ng e",
+      "n ge",
+      "车 间",
+      "纳 税",
+      "med ium",
+      "▁g rain",
+      "▁gr ain",
+      "▁gra in",
+      "Jo hn",
+      "Joh n",
+      "J ohn",
+      "du ctor",
+      "duct or",
+      "duc tor",
+      "▁p ortal",
+      "▁port al",
+      "▁por tal",
+      "▁ portal",
+      "人民法院 认为",
+      "简 历",
+      "back s",
+      "ba cks",
+      "bac ks",
+      "b acks",
+      "▁G all",
+      "▁Gal l",
+      "▁Ga ll",
+      "ac ular",
+      "a cular",
+      "▁inv asion",
+      "f o",
+      "▁v as",
+      "▁va s",
+      "▁ vas",
+      "▁bas al",
+      "▁ba sal",
+      "▁s ender",
+      "▁se nder",
+      "▁send er",
+      "▁sen der",
+      "▁ sender",
+      "W elcome",
+      "li sts",
+      "list s",
+      "lis ts",
+      "l ists",
+      "你 知道",
+      "递 增",
+      "▁H arris",
+      "▁Har ris",
+      "▁Harr is",
+      "▁coll ision",
+      "▁B or",
+      "▁Bo r",
+      "▁impair ment",
+      "▁p ap",
+      "▁pa p",
+      "▁Ex amples",
+      "▁Example s",
+      "▁Exam ples",
+      "▁ Examples",
+      "较高 的",
+      "较 高的",
+      "▁ad mit",
+      "▁adm it",
+      "罪 行",
+      "▁A ns",
+      "▁An s",
+      "▁ Ans",
+      "金 华",
+      "face book",
+      "f acebook",
+      "▁A x",
+      "▁ Ax",
+      "跟 踪",
+      "w x",
+      "of t",
+      "o ft",
+      "add le",
+      "▁B R",
+      "▁ BR",
+      "ar an",
+      "ara n",
+      "a ran",
+      "连 带",
+      "邮 箱",
+      "é t",
+      "]) ,",
+      "] ),",
+      "il ty",
+      "ilt y",
+      "ol ec",
+      "ole c",
+      "o lec",
+      "▁con sent",
+      "▁cons ent",
+      "Dis c",
+      "Di sc",
+      "D isc",
+      "第六 章",
+      "▁c sv",
+      "▁cs v",
+      "▁ csv",
+      "Calcul ate",
+      "Calc ulate",
+      "C alculate",
+      "al ling",
+      "all ing",
+      "alli ng",
+      "er gic",
+      "erg ic",
+      "▁her itage",
+      "bol d",
+      "bo ld",
+      "b old",
+      "▁util s",
+      "▁ut ils",
+      "▁ utils",
+      "▁m u",
+      "▁ mu",
+      "}) ,\\",
+      "}), \\",
+      "} ),\\",
+      "▁Per haps",
+      "us sy",
+      "uss y",
+      "Ext ensions",
+      "Extension s",
+      "以下 问题",
+      "为 基础",
+      "▁Prot ection",
+      "▁Prote ction",
+      "去 看",
+      "有些 人",
+      "▁t et",
+      "▁te t",
+      "▁ tet",
+      "▁O tt",
+      "▁Ot t",
+      "▁ 题目",
+      "▁f itted",
+      "▁fit ted",
+      "nCom mun",
+      "n Commun",
+      "\\' ])",
+      "\\'] )",
+      "\\ '])",
+      "▁m oon",
+      "▁mo on",
+      "▁ moon",
+      "计算 即可",
+      "污 水",
+      "▁Value Error",
+      "▁ ValueError",
+      "平 米",
+      "he im",
+      "▁l on",
+      "▁lo n",
+      "▁ lon",
+      "Bl ack",
+      "的两 个",
+      "的 两个",
+      "▁sim ulated",
+      "▁simulate d",
+      "▁ultra sound",
+      "▁ultras ound",
+      "▁其 三",
+      "▁Me asure",
+      "▁ Measure",
+      "要 点",
+      "▁Islam ic",
+      "▁leg acy",
+      "▁ legacy",
+      "▁v oting",
+      "▁vot ing",
+      "▁vo ting",
+      "借 助",
+      "ng it",
+      "n git",
+      "经济 技术",
+      "签 字",
+      "ryst al",
+      "rys tal",
+      "执 业",
+      "▁l act",
+      "▁la ct",
+      "▁lac t",
+      "▁ lact",
+      "em ale",
+      "ema le",
+      "e male",
+      "淄 博",
+      "讲 座",
+      "if ting",
+      "ift ing",
+      "农 产品",
+      "br as",
+      "bra s",
+      "b ras",
+      "▁ref uge",
+      "▁Viet nam",
+      "nD is",
+      "nDi s",
+      "n Dis",
+      "录 音",
+      "▁import antly",
+      "▁important ly",
+      "nu n",
+      "n un",
+      "三 国",
+      "▁re b",
+      "▁r eb",
+      "▁ reb",
+      "▁conduct ing",
+      "▁sur rounded",
+      "▁surround ed",
+      "聚 集",
+      "救 援",
+      "▁to ile",
+      "不动 产",
+      "数据 的",
+      "▁D allas",
+      "▁Dal las",
+      "研究 中心",
+      "▁K han",
+      "▁Kh an",
+      "债 券",
+      "D EFAULT",
+      "一等 奖",
+      "一 等奖",
+      "▁cons ervative",
+      "▁conserv ative",
+      "▁P ear",
+      "▁Pe ar",
+      "▁代表 作品",
+      "▁ 代表作品",
+      "▁th romb",
+      "▁thr omb",
+      "▁throm b",
+      "▁thro mb",
+      "灾 害",
+      "▁d uct",
+      "▁du ct",
+      "▁ duct",
+      "▁p ros",
+      "▁pro s",
+      "▁pr os",
+      "n which",
+      "月二十 五日",
+      "▁eng aging",
+      "▁时 间",
+      "▁ 时间",
+      "现有 的",
+      "现 有的",
+      "▁ch arity",
+      "▁char ity",
+      "Sh ould",
+      "pl icity",
+      "plic ity",
+      "plicit y",
+      "▁celebr ation",
+      "剩下 的",
+      "剩 下的",
+      "▁R eference",
+      "▁Re ference",
+      "▁Refer ence",
+      "▁ Reference",
+      "ord ered",
+      "order ed",
+      "orde red",
+      "us cript",
+      "usc ript",
+      "u script",
+      "▁pl ates",
+      "▁plate s",
+      "▁plat es",
+      "▁ plates",
+      "▁ 已",
+      "巴 西",
+      "▁S S",
+      "▁ SS",
+      "St and",
+      "S tand",
+      "▁c yclic",
+      "▁cycl ic",
+      "▁ cyclic",
+      "▁For ward",
+      "▁ Forward",
+      "}\\ '",
+      "} \\'",
+      "Aut hentication",
+      "Authent ication",
+      "▁G rant",
+      "▁Gr ant",
+      "▁Gra nt",
+      "▁Gran t",
+      "张某 某",
+      "张 某某",
+      "ok in",
+      "oki n",
+      "o kin",
+      "▁T rail",
+      "▁Tr ail",
+      "▁Tra il",
+      "▁v enues",
+      "▁ven ues",
+      "▁venue s",
+      "倾 向",
+      "试 题",
+      "ni l",
+      "n il",
+      "▁Ann ual",
+      "精 英",
+      "en za",
+      "enz a",
+      "le et",
+      "lee t",
+      "▁res cue",
+      "▁resc ue",
+      "塑 造",
+      "nP ublic",
+      "nPub lic",
+      "n Public",
+      "Ra te",
+      "R ate",
+      "安 卓",
+      "纳 米",
+      "Action s",
+      "Act ions",
+      "三 分",
+      "的 风",
+      "O k",
+      "▁un iqu",
+      "▁uni qu",
+      "▁x x",
+      "▁ xx",
+      "▁sym metry",
+      "▁symmet ry",
+      "▁ 政",
+      "黄 河",
+      "c z",
+      "▁al arm",
+      "▁ alarm",
+      "▁det ector",
+      "▁detect or",
+      "qu iry",
+      "quir y",
+      "▁struct ured",
+      "▁structure d",
+      "▁ structured",
+      "▁trans lated",
+      "▁transl ated",
+      "▁translate d",
+      "▁ translated",
+      "设备 的",
+      "lo ads",
+      "load s",
+      "ut her",
+      "uth er",
+      "u ther",
+      "dl l",
+      "d ll",
+      "▁str uggling",
+      "▁strugg ling",
+      "wi ck",
+      "w ick",
+      "es a",
+      "e sa",
+      "▁br others",
+      "▁bro thers",
+      "▁brother s",
+      "▁broth ers",
+      "em it",
+      "emi t",
+      "e mit",
+      "▁RE AD",
+      "▁ READ",
+      "▁D ream",
+      "▁ Dream",
+      "▁R ol",
+      "▁Ro l",
+      "# {",
+      "▁su icide",
+      "接 种",
+      "Di v",
+      "D iv",
+      "执行 中",
+      "枚 举",
+      "道 的",
+      "ed er",
+      "ede r",
+      "e der",
+      "▁mult id",
+      "▁multi d",
+      "▁mul tid",
+      "▁ 相关",
+      "var s",
+      "va rs",
+      "v ars",
+      "▁In cre",
+      "▁Inc re",
+      "▁ Incre",
+      "局 局长",
+      "▁H ans",
+      "▁Ha ns",
+      "▁Han s",
+      "中 午",
+      "两 岸",
+      "▁ 年",
+      "▁E T",
+      "▁ ET",
+      "tr im",
+      "tri m",
+      "t rim",
+      "新 兴",
+      "▁ 输出",
+      "▁m old",
+      "▁mo ld",
+      "▁mol d",
+      "执行 申请",
+      "▁M ail",
+      "▁Ma il",
+      "▁Mai l",
+      "▁ Mail",
+      "手 指",
+      "这一 切",
+      "这 一切",
+      "PE CT",
+      "PEC T",
+      "P ECT",
+      "oc a",
+      "o ca",
+      "▁Brazil ian",
+      "▁s eeks",
+      "▁se eks",
+      "▁see ks",
+      "▁seek s",
+      "ic tures",
+      "ict ures",
+      "icture s",
+      "开 户",
+      "▁morph ology",
+      "nF rench",
+      "nFr ench",
+      "nFre nch",
+      "n French",
+      "就 这样",
+      "▁length s",
+      "▁leng ths",
+      "och astic",
+      "▁ 天",
+      "玄 幻",
+      "` ;\\",
+      "用 心",
+      "椭 圆",
+      "▁discuss es",
+      "PD ATE",
+      "P DATE",
+      "评审 鉴定表",
+      "▁With in",
+      "▁Wi thin",
+      "▁Wit hin",
+      "血 清",
+      "上 学",
+      "▁K y",
+      "子 宫",
+      "de ploy",
+      "dep loy",
+      "ur ities",
+      "uri ties",
+      "▁Indust ry",
+      "家 园",
+      "▁p roc",
+      "▁pro c",
+      "▁pr oc",
+      "▁ proc",
+      "zhe imer",
+      "▁ant enna",
+      "▁anten na",
+      "▁st aying",
+      "▁stay ing",
+      "平 凡",
+      "▁Log ger",
+      "▁ Logger",
+      "▁s tops",
+      "▁st ops",
+      "▁stop s",
+      "▁sto ps",
+      "ud d",
+      "u dd",
+      "▁resid ues",
+      "▁residue s",
+      "▁gener ates",
+      "▁gene rates",
+      "▁generate s",
+      "▁genera tes",
+      "▁N um",
+      "▁Nu m",
+      "▁ Num",
+      "Gener al",
+      "Gen eral",
+      "Gene ral",
+      "▁S K",
+      "▁ SK",
+      "▁g ray",
+      "▁gr ay",
+      "▁gra y",
+      "▁ gray",
+      "▁a rom",
+      "▁ar om",
+      "um ing",
+      "umin g",
+      "umi ng",
+      "u ming",
+      "trans action",
+      "▁conj unction",
+      "W T",
+      "日出生 于",
+      "日出 生于",
+      "日 出生于",
+      "职 位",
+      "▁altern atives",
+      "▁alternative s",
+      "坐 落",
+      "▁B ishop",
+      "▁Bi shop",
+      "▁Bis hop",
+      "遵 循",
+      "▁Tur kish",
+      "▁ann ually",
+      "▁annual ly",
+      "▁chem icals",
+      "▁chemical s",
+      "▁adoles cents",
+      "▁adolescent s",
+      "五 金",
+      "Se lected",
+      "Select ed",
+      "S elected",
+      "ir i",
+      "i ri",
+      "▁Fe ature",
+      "▁ Feature",
+      "还需 要",
+      "还 需要",
+      "hu man",
+      "hum an",
+      "h uman",
+      "Mem ory",
+      "M emory",
+      "封 闭",
+      "Array List",
+      "告诉 你",
+      "▁p overty",
+      "▁A lb",
+      "▁Al b",
+      "▁lig and",
+      "gi um",
+      "g ium",
+      "sl ice",
+      "时间 的",
+      "时 间的",
+      "杀 人",
+      "▁deep ly",
+      "ri o",
+      "r io",
+      "在开庭 审理过程中",
+      "▁cr usher",
+      "▁crush er",
+      "▁quant ities",
+      "违 规",
+      "法律服务所 法律工作者",
+      "▁t win",
+      "▁tw in",
+      "▁H amp",
+      "▁Ha mp",
+      "▁Ham p",
+      "▁inhib it",
+      "sa fe",
+      "saf e",
+      "s afe",
+      "Sing le",
+      "S ingle",
+      "nA c",
+      "n Ac",
+      "让他 们",
+      "让 他们",
+      "▁Pl atform",
+      "▁Plat form",
+      "▁ Platform",
+      "▁un likely",
+      "▁unlike ly",
+      "for med",
+      "form ed",
+      "ch r",
+      "c hr",
+      "ma ss",
+      "mas s",
+      "m ass",
+      "▁experi encing",
+      "▁r ust",
+      "▁ru st",
+      "▁ rust",
+      "win ning",
+      "小 额",
+      "▁out puts",
+      "▁output s",
+      "▁ outputs",
+      "就 没",
+      "an cell",
+      "ance ll",
+      "anc ell",
+      "ancel l",
+      "汇 报",
+      "pro cessing",
+      "process ing",
+      "proc essing",
+      "▁j azz",
+      "▁ja zz",
+      "代 数",
+      "CO DE",
+      "COD E",
+      "侵 犯",
+      "的 角度",
+      "备 用",
+      "要 想",
+      "的其他 情形",
+      "队 的",
+      "▁Do ctor",
+      "▁Doc tor",
+      "彩 色",
+      "电 磁",
+      "▁run time",
+      "▁ru ntime",
+      "▁ runtime",
+      "裁定书 送达之日起",
+      "CO MP",
+      "COM P",
+      "▁re pairs",
+      "▁rep airs",
+      "▁repair s",
+      "requ ires",
+      "require s",
+      "▁D isease",
+      "▁Dise ase",
+      "复议 一次",
+      "▁ 亚",
+      "writ ing",
+      "wr iting",
+      "av al",
+      "ava l",
+      "a val",
+      "▁Con struction",
+      "▁Construct ion",
+      "品牌 的",
+      "EN G",
+      "E NG",
+      "▁we igh",
+      "S Y",
+      "▁re levance",
+      "▁concept ual",
+      "子女 抚养",
+      "特 朗",
+      "爱好 者",
+      "O G",
+      "ow ned",
+      "own ed",
+      "生 动",
+      "▁s tom",
+      "▁st om",
+      "▁sto m",
+      "▁m inus",
+      "▁min us",
+      "▁mi nus",
+      "▁ minus",
+      "▁fib ers",
+      "▁fiber s",
+      "▁fi bers",
+      "为终审 裁定",
+      "力 学",
+      "▁d ishes",
+      "▁dis hes",
+      "▁dish es",
+      "▁ant igens",
+      "▁anti gens",
+      "▁antigen s",
+      "▁D omin",
+      "▁Do min",
+      "▁Dom in",
+      "▁M ort",
+      "▁Mor t",
+      "▁Mo rt",
+      "▁N F",
+      "▁ NF",
+      "▁un signed",
+      "▁uns igned",
+      "▁ unsigned",
+      "繁 殖",
+      "老师 的",
+      "老 师的",
+      "▁talk ed",
+      "还 原",
+      "ter net",
+      "tern et",
+      "▁S cale",
+      "▁Sc ale",
+      "▁ Scale",
+      "rec ated",
+      "▁journal ist",
+      "老年 人",
+      "reg ulated",
+      "何 处",
+      "M igration",
+      "xx x",
+      "x xx",
+      "▁in cent",
+      "▁inc ent",
+      "▁Sw iss",
+      "inter val",
+      "回 头",
+      "热 点",
+      "▁jew el",
+      "解决 了",
+      "vel ope",
+      "velop e",
+      "O A",
+      "容 积",
+      "放 置",
+      "l c",
+      "▁b last",
+      "▁bl ast",
+      "▁ blast",
+      "os omal",
+      "oso mal",
+      "▁W el",
+      "▁We l",
+      "▁Work ing",
+      "▁Wor king",
+      "▁ Working",
+      "形 容",
+      "it an",
+      "ita n",
+      "i tan",
+      "二 人",
+      "if ference",
+      "iffer ence",
+      "iffe rence",
+      "本裁定 为终审裁定",
+      "加 班",
+      "mult i",
+      "mul ti",
+      "▁att ributed",
+      "▁attribute d",
+      "气 息",
+      "▁col lective",
+      "▁coll ective",
+      "▁collect ive",
+      "▁f ired",
+      "▁fire d",
+      "▁fi red",
+      "▁fir ed",
+      "in ject",
+      "inj ect",
+      "抛物 线",
+      "▁re start",
+      "▁rest art",
+      "▁ restart",
+      "ed ical",
+      "edi cal",
+      "edic al",
+      "\") [",
+      "\" )[",
+      "液 压",
+      "纵 横",
+      "回 报",
+      "ke e",
+      "k ee",
+      "血 糖",
+      "▁C er",
+      "▁Ce r",
+      "国 防",
+      "等 人",
+      "▁sequ encing",
+      "癜 风",
+      "le asing",
+      "lea sing",
+      "发 起",
+      "辩 护",
+      "据 了解",
+      "▁opin ions",
+      "▁opinion s",
+      "▁ok ay",
+      "▁B ah",
+      "▁Ba h",
+      "延 续",
+      "▁c ig",
+      "▁ci g",
+      "writ ten",
+      "wr itten",
+      "un defined",
+      "und efined",
+      "undef ined",
+      "u ndefined",
+      "▁e motion",
+      "▁em otion",
+      "▁emot ion",
+      "li a",
+      "l ia",
+      "▁Ch ile",
+      "▁Chi le",
+      "▁s ectors",
+      "▁se ctors",
+      "▁sector s",
+      "桥 梁",
+      "▁mist akes",
+      "▁mistake s",
+      "▁% }",
+      "因 与",
+      "花 生",
+      "▁pre serve",
+      "▁pres erve",
+      "▁ preserve",
+      "é n",
+      "▁B u",
+      "▁ Bu",
+      "nO pen",
+      "n Open",
+      "▁p ixel",
+      "▁ pixel",
+      "▁arr angements",
+      "▁arrang ements",
+      "▁arrangement s",
+      "▁arrange ments",
+      "is ible",
+      "isi ble",
+      "re solution",
+      "res olution",
+      "Inter val",
+      "▁Add ress",
+      "▁ Address",
+      "am ento",
+      "ame nto",
+      "ament o",
+      "amen to",
+      "为 啥",
+      "的 诉讼请求",
+      "act ivated",
+      "activ ated",
+      "activate d",
+      "十 字",
+      "▁fl oors",
+      "▁flo ors",
+      "▁floor s",
+      "he ter",
+      "het er",
+      "h eter",
+      "投 诉",
+      "in ge",
+      "ing e",
+      "i nge",
+      "和 小",
+      "▁a str",
+      "▁as tr",
+      "▁ast r",
+      "▁ astr",
+      "向 量",
+      "ident ifier",
+      "umbn ail",
+      "班 级",
+      "hel ium",
+      "不 算",
+      "oc ar",
+      "oca r",
+      "o car",
+      "▁cult ured",
+      "▁culture d",
+      "▁w ard",
+      "▁war d",
+      "▁wa rd",
+      "▁ ward",
+      "▁f ascin",
+      "▁fasc in",
+      "出来 了",
+      "出 来了",
+      "▁K am",
+      "▁Ka m",
+      "▁p ride",
+      "▁pr ide",
+      "▁pri de",
+      "▁b rid",
+      "▁br id",
+      "▁ brid",
+      "▁inst alling",
+      "▁install ing",
+      "Dr op",
+      "D rop",
+      "ov en",
+      "ove n",
+      "o ven",
+      "作 了",
+      "Mult i",
+      "▁synthes ized",
+      "▁synthesize d",
+      "D a",
+      "▁s acr",
+      "▁sa cr",
+      "▁sac r",
+      "▁sc alar",
+      "▁scal ar",
+      "▁scala r",
+      "▁ scalar",
+      "▁a broad",
+      "▁ab road",
+      "▁ 罗",
+      "▁Init ial",
+      "▁ Initial",
+      "\"> {{",
+      "\">{ {",
+      "\" >{{",
+      "▁reg ards",
+      "▁regard s",
+      "沈阳 市",
+      "沈 阳市",
+      "▁N ag",
+      "▁Na g",
+      "▁cle arance",
+      "▁clear ance",
+      "gr ay",
+      "gra y",
+      "g ray",
+      "多 媒体",
+      "学习 的",
+      "▁I ce",
+      "如果 没有",
+      "ear n",
+      "ea rn",
+      "e arn",
+      "work er",
+      "wor ker",
+      "▁eth nic",
+      "▁ethn ic",
+      "客户 端",
+      "▁accommod ation",
+      "立即 执行",
+      "我 校",
+      "nJ oined",
+      "nJoin ed",
+      "nJo ined",
+      "HE CK",
+      "新 时代",
+      "陌 生",
+      "▁T E",
+      "▁ TE",
+      "I Z",
+      "om pson",
+      "omp son",
+      "▁g auge",
+      "▁ga uge",
+      "▁work place",
+      "除 非",
+      "up er",
+      "upe r",
+      "u per",
+      "▁po etry",
+      "▁poet ry",
+      "▁B ull",
+      "▁Bul l",
+      "▁Bu ll",
+      "在 一",
+      "当 下",
+      "▁pert urb",
+      "卓 越",
+      "eo r",
+      "e or",
+      "Nor mal",
+      "Norm al",
+      "N ormal",
+      "以 免",
+      "白 癜风",
+      "这个 时候",
+      "▁U til",
+      "▁Ut il",
+      "▁ Util",
+      "土 豆",
+      "李某 某",
+      "李 某某",
+      "en te",
+      "ent e",
+      "e nte",
+      "nD et",
+      "nDe t",
+      "n Det",
+      "主 演",
+      "ar ity",
+      "ari ty",
+      "余 人",
+      "▁mount ains",
+      "▁mountain s",
+      "亦 无异议",
+      "当 日",
+      "▁Con tract",
+      "▁Contr act",
+      "▁ Contract",
+      "▁acknow led",
+      "ement ia",
+      "▁Ant onio",
+      "▁Anton io",
+      "▁con sequence",
+      "▁cons equence",
+      "▁consequ ence",
+      "he art",
+      "hea rt",
+      "开 口",
+      "com fort",
+      "da le",
+      "dal e",
+      "d ale",
+      "护 士",
+      "TE XT",
+      "TEX T",
+      "T EXT",
+      "▁Th ailand",
+      "▁Thai land",
+      "р и",
+      "介 质",
+      "充分 发挥",
+      "▁me chanics",
+      "▁mechan ics",
+      "▁mechanic s",
+      "好 评",
+      "服 用",
+      "sec onds",
+      "secon ds",
+      "second s",
+      "佐 证",
+      "▁represent ations",
+      "▁representation s",
+      "不管 是",
+      "有效 地",
+      "治 愈",
+      "to o",
+      "t oo",
+      "辉 煌",
+      "▁d ismiss",
+      "▁dis miss",
+      "▁ dismiss",
+      "▁comp iler",
+      "▁compile r",
+      "▁ compiler",
+      "▁ref lected",
+      "▁reflect ed",
+      "▁S end",
+      "▁Se nd",
+      "▁Sen d",
+      "▁ Send",
+      "▁pub lisher",
+      "▁publish er",
+      "▁ publisher",
+      "官 员",
+      "▁D og",
+      "▁Do g",
+      "▁ Dog",
+      "▁b att",
+      "▁bat t",
+      "▁ba tt",
+      "▁immun ity",
+      "▁imm unity",
+      "mi a",
+      "m ia",
+      "▁w rest",
+      "▁wr est",
+      "▁wre st",
+      "▁Ch ap",
+      "▁Cha p",
+      "▁inc orporate",
+      "▁incorpor ate",
+      "lo gs",
+      "log s",
+      "l ogs",
+      "传 染",
+      "▁he arts",
+      "▁heart s",
+      "▁hear ts",
+      "sc ala",
+      "scal a",
+      "▁re covered",
+      "▁rec overed",
+      "▁recover ed",
+      "因为 我",
+      "因 为我",
+      "得 以",
+      "ue se",
+      "ues e",
+      "u ese",
+      "之前 的",
+      "之 前的",
+      "） ×",
+      "十六 条",
+      "十 六条",
+      "▁Be autiful",
+      "▁Beaut iful",
+      "▁ Beautiful",
+      "交通 运输",
+      "▁l oving",
+      "▁lo ving",
+      "▁real istic",
+      "ns ub",
+      "n sub",
+      "▁Lin coln",
+      "▁gather ed",
+      "▁Econ omic",
+      "▁Ec onomic",
+      "宾 馆",
+      "层 面",
+      "▁init iated",
+      "▁initi ated",
+      "▁initiate d",
+      "iss ippi",
+      "ie re",
+      "ier e",
+      "i ere",
+      "Object s",
+      "Obj ects",
+      "kl e",
+      "k le",
+      "U V",
+      "水 量",
+      "诗 歌",
+      "▁] ,\\",
+      "▁], \\",
+      "▁ ],\\",
+      "队 友",
+      "▁l attice",
+      "▁b ip",
+      "▁bi p",
+      "ein e",
+      "ei ne",
+      "e ine",
+      "▁J oin",
+      "▁Jo in",
+      "▁ Join",
+      "▁connect ivity",
+      "民 二",
+      "▁s inging",
+      "▁sing ing",
+      "▁sin ging",
+      "到 处",
+      "中 止",
+      "▁N ature",
+      "▁Na ture",
+      "▁Nat ure",
+      "乘 坐",
+      "第一 时间",
+      "at eful",
+      "ate ful",
+      "▁S amsung",
+      "节 省",
+      "予以 证实",
+      "▁st ir",
+      "t u",
+      "实 的",
+      "su n",
+      "s un",
+      "同样 的",
+      "同 样的",
+      "fr ed",
+      "fre d",
+      "f red",
+      "CC ESS",
+      "\\\\ }$",
+      "\\\\} $",
+      "anc ouver",
+      "钢 琴",
+      "▁ex amines",
+      "▁exam ines",
+      "▁examine s",
+      "日 公开开庭进行了审理",
+      "▁bu ried",
+      "▁bur ied",
+      "com a",
+      "co ma",
+      "c oma",
+      "要 在",
+      "▁t ropical",
+      "▁trop ical",
+      "buck et",
+      "b ucket",
+      "▁sc attering",
+      "▁scatter ing",
+      "▁W as",
+      "▁Wa s",
+      "商 城",
+      "nH ello",
+      "nHel lo",
+      "n Hello",
+      "▁gold en",
+      "▁gol den",
+      "友 谊",
+      "法律法规 及",
+      "烦 恼",
+      "ri tic",
+      "rit ic",
+      "r itic",
+      "tr action",
+      "tract ion",
+      "tra ction",
+      "▁invest ments",
+      "▁investment s",
+      "▁special ists",
+      "▁specialist s",
+      "烟 台",
+      "▁Queens land",
+      "ju ana",
+      "▁g rows",
+      "▁gr ows",
+      "▁grow s",
+      "▁gro ws",
+      "提 问",
+      "▁P CR",
+      "▁PC R",
+      "▁ PCR",
+      "▁T ogether",
+      "之 和",
+      "如 有",
+      "▁tr ips",
+      "▁tri ps",
+      "▁trip s",
+      "To ols",
+      "Tool s",
+      "Too ls",
+      "T ools",
+      "在哪 里",
+      "在 哪里",
+      "nGe ography",
+      "Wh ere",
+      "W here",
+      "时 效",
+      "nIn ternational",
+      "nInter national",
+      "n International",
+      "占 用",
+      "做 什么",
+      "▁G ib",
+      "▁Gi b",
+      "▁clar ify",
+      "中 等",
+      "nM at",
+      "nMa t",
+      "n Mat",
+      "的最 佳",
+      "的 最佳",
+      "▁an omal",
+      "▁dec lar",
+      "▁decl ar",
+      "▁ declar",
+      "清 除",
+      "ari juana",
+      "el ements",
+      "ele ments",
+      "element s",
+      "elem ents",
+      "e lements",
+      "RE G",
+      "R EG",
+      "▁t ob",
+      "▁to b",
+      "ro wave",
+      "row ave",
+      "▁hope fully",
+      "▁hop efully",
+      "▁hopeful ly",
+      "人 性",
+      "ious ly",
+      "i ously",
+      "ynt hesis",
+      "ynth esis",
+      "晋江文学 城",
+      "▁em power",
+      "▁emp ower",
+      "▁spec ification",
+      "▁specific ation",
+      "▁gover nor",
+      "▁govern or",
+      "综合 性",
+      "你 好",
+      "ath ers",
+      "ather s",
+      "athe rs",
+      "a thers",
+      "ol er",
+      "ole r",
+      "o ler",
+      "er als",
+      "era ls",
+      "eral s",
+      "e rals",
+      "▁In fo",
+      "▁Inf o",
+      "▁ Info",
+      "▁med ications",
+      "▁medication s",
+      "▁medic ations",
+      "▁Log in",
+      "▁Lo gin",
+      "▁ Login",
+      "str ument",
+      "作 风",
+      "var iant",
+      "vari ant",
+      "×× ××",
+      "××× ×",
+      "× ×××",
+      "▁u re",
+      "▁ur e",
+      "▁ ure",
+      "▁n ose",
+      "▁no se",
+      "▁nos e",
+      "▁ nose",
+      "▁per me",
+      "▁perm e",
+      "的 现象",
+      "▁a spir",
+      "▁as pir",
+      "▁asp ir",
+      "vis ual",
+      "st rings",
+      "str ings",
+      "string s",
+      "▁power ed",
+      "▁pow ered",
+      "▁ powered",
+      "在 了",
+      "la rg",
+      "lar g",
+      "l arg",
+      "fess ional",
+      "的 游戏",
+      "D K",
+      "Dis patch",
+      "ver ify",
+      "yn amics",
+      "ynam ics",
+      "ynamic s",
+      "circ le",
+      "cir cle",
+      "c ircle",
+      "▁fl uct",
+      "▁flu ct",
+      "这 也",
+      "▁Af ghan",
+      "▁sub stitution",
+      "▁subst itution",
+      "▁substit ution",
+      "我们 就",
+      "▁Phys ical",
+      "▁ Physical",
+      "热 带",
+      "相 邻",
+      "其中 的",
+      "其 中的",
+      "▁Mat rix",
+      "▁ Matrix",
+      "ha ve",
+      "hav e",
+      "h ave",
+      "▁s lic",
+      "▁sl ic",
+      "又 称",
+      "▁incre ment",
+      "▁ increment",
+      "还 好",
+      "男 主",
+      "▁differ ently",
+      "▁different ly",
+      "▁Ut ah",
+      "▁pass enger",
+      "民 法",
+      "较 多",
+      "▁v ag",
+      "▁va g",
+      "私 人",
+      "▁assum ptions",
+      "▁assumption s",
+      "每 周",
+      "ik z",
+      "姐 妹",
+      "IS A",
+      "I SA",
+      "尽 可能",
+      "能力 和",
+      "能 力和",
+      "▁report edly",
+      "▁reported ly",
+      "总 理",
+      "▁re named",
+      "▁rename d",
+      "▁a rise",
+      "▁ar ise",
+      "网 易",
+      "nV ill",
+      "ph abet",
+      "pha bet",
+      "▁r anking",
+      "▁rank ing",
+      "▁ran king",
+      "▁ ranking",
+      "陪 伴",
+      "▁ign ored",
+      "▁ignore d",
+      "▁ignor ed",
+      "划 拨",
+      "切 成",
+      "临 沂",
+      "▁P ersonal",
+      "▁Pers onal",
+      "▁Person al",
+      "▁ Personal",
+      "你 看",
+      "空 中",
+      "▁cert ification",
+      "▁m im",
+      "▁mi m",
+      "▁IO Exception",
+      "▁ IOException",
+      "▁ch amber",
+      "▁cha mber",
+      "▁cerv ical",
+      "用户 的",
+      "▁S ession",
+      "▁ Session",
+      "▁R on",
+      "▁Ro n",
+      "期间不 停止",
+      "▁p il",
+      "▁pi l",
+      "境 内",
+      "副本 二份",
+      "can vas",
+      "▁M C",
+      "▁ MC",
+      "st rong",
+      "str ong",
+      "▁p acket",
+      "▁pack et",
+      "▁pac ket",
+      "▁ packet",
+      "▁reflect s",
+      "▁p yl",
+      "▁py l",
+      "▁ pyl",
+      "说 到",
+      "▁H appy",
+      "▁Happ y",
+      "▁ Happy",
+      "实行 独任审判",
+      "不同 意",
+      "不 同意",
+      "十 大",
+      "▁make up",
+      "耐 心",
+      "Da o",
+      "D ao",
+      "▁ 白",
+      "▁A qu",
+      "▁c and",
+      "▁can d",
+      "▁ca nd",
+      "标 记",
+      "饰 演",
+      "▁A w",
+      "▁ Aw",
+      "一 口",
+      "再 生",
+      "▁G T",
+      "▁ GT",
+      "大 专",
+      "投 票",
+      "调解 协议",
+      "▁PUB LIC",
+      "▁ PUBLIC",
+      "▁in far",
+      "▁inf ar",
+      "发 作",
+      "▁L u",
+      "▁ Lu",
+      "▁pancreat ic",
+      "投 标",
+      "▁eth anol",
+      "号 线",
+      "▁M ut",
+      "▁Mu t",
+      "▁ Mut",
+      "赋 予",
+      "▁l ear",
+      "▁le ar",
+      "▁ lear",
+      "M V",
+      "ó w",
+      "nb l",
+      "n bl",
+      "去 世",
+      "有一 些",
+      "有 一些",
+      "走 势",
+      "▁ob lig",
+      "▁obl ig",
+      "▁i Pad",
+      "▁ iPad",
+      "▁t act",
+      "▁ta ct",
+      "▁c ube",
+      "▁cub e",
+      "▁cu be",
+      "▁ cube",
+      "无 疑",
+      "▁port ions",
+      "▁portion s",
+      "▁pos itively",
+      "▁positive ly",
+      "▁posit ively",
+      "届 满",
+      "▁B ron",
+      "▁Br on",
+      "▁Bro n",
+      "▁ Bron",
+      "▁p ione",
+      "▁pi one",
+      "南 部",
+      "Vis ual",
+      "线 段",
+      "st rict",
+      "str ict",
+      "stri ct",
+      "▁cyt otox",
+      "▁cyto tox",
+      "at ories",
+      "ator ies",
+      "ato ries",
+      "<< \"",
+      "那时 候",
+      "那 时候",
+      "nT ake",
+      "n Take",
+      "▁tr ains",
+      "▁train s",
+      "▁tra ins",
+      "▁accommod ate",
+      "▁dis patch",
+      "▁disp atch",
+      "▁ dispatch",
+      "▁N IH",
+      "▁NI H",
+      "▁cy ber",
+      "▁Mon te",
+      "▁Mont e",
+      "▁Mo nte",
+      "务 工",
+      "石 头",
+      "一个 是",
+      "Com put",
+      "Comp ut",
+      "复议 期间不停止",
+      "▁l oyal",
+      "▁lo yal",
+      "▁S ale",
+      "▁Sal e",
+      "▁Sa le",
+      "▁ Sale",
+      "月二十 七日",
+      "un ks",
+      "unk s",
+      "自然 科学",
+      "ph y",
+      "p hy",
+      "▁Me tro",
+      "▁Met ro",
+      "▁Sing h",
+      "▁Sin gh",
+      "igen ous",
+      "龙 头",
+      "均 衡",
+      "▁down loaded",
+      "▁download ed",
+      "▁back end",
+      "▁ backend",
+      "花 钱",
+      "▁br illiant",
+      "▁brilli ant",
+      "▁un like",
+      "▁u nlike",
+      "▁ 无",
+      "▁Republic ans",
+      "▁Republican s",
+      "储 备",
+      "口 罩",
+      "li t",
+      "l it",
+      "▁F lo",
+      "▁Fl o",
+      "功 夫",
+      "化妆 品",
+      "▁T oy",
+      "▁To y",
+      "中华人民共和国 婚姻法",
+      "安 静",
+      "fun d",
+      "fu nd",
+      "f und",
+      "▁pain tings",
+      "▁paint ings",
+      "▁painting s",
+      "asc ade",
+      "if acts",
+      "ifact s",
+      "ifa cts",
+      "i facts",
+      "出 台",
+      "吐 槽",
+      "run ner",
+      "r unner",
+      "▁t raits",
+      "▁tra its",
+      "▁trait s",
+      "▁ traits",
+      "nS impl",
+      "nSim pl",
+      "▁Int ellig",
+      "▁Intel lig",
+      "Des ign",
+      "De sign",
+      "ain ian",
+      "a inian",
+      "▁F ellow",
+      "▁Fel low",
+      "也不 知道",
+      "也 不知道",
+      "▁establish ing",
+      "▁pur ified",
+      "al loc",
+      "all oc",
+      "allo c",
+      "▁Hy per",
+      "▁Hyp er",
+      "▁ Hyper",
+      "m v",
+      "▁B eng",
+      "▁Be ng",
+      "▁Ben g",
+      "▁token s",
+      "▁tok ens",
+      "▁ tokens",
+      "决 心",
+      "量 刑",
+      "这时 候",
+      "这 时候",
+      "gl as",
+      "g las",
+      "本案 受理费",
+      "▁Environment al",
+      "▁inst itutional",
+      "▁institution al",
+      "专 科",
+      "默 默",
+      "的 思路",
+      "▁w atched",
+      "▁watch ed",
+      "▁wat ched",
+      "▁isol ates",
+      "▁isolate s",
+      "小 明",
+      "的 路",
+      "支 公司",
+      "▁circ ulation",
+      "run ning",
+      "r unning",
+      "负 数",
+      "▁re porter",
+      "▁rep orter",
+      "▁report er",
+      "▁ reporter",
+      "ext ensions",
+      "extension s",
+      "Ur i",
+      "U ri",
+      "在 国内",
+      "▁cas ino",
+      "佛山 市",
+      "佛 山市",
+      "sequ ently",
+      "sequent ly",
+      "▁pop ularity",
+      "▁popular ity",
+      "▁popul arity",
+      "诉至 法院",
+      "▁W hit",
+      "▁Wh it",
+      "▁ Whit",
+      "fil m",
+      "fi lm",
+      "新 手",
+      "社 团",
+      "▁ 制作",
+      "Te am",
+      "nE l",
+      "n El",
+      "▁afr aid",
+      "meth yl",
+      "m ethyl",
+      "变 换",
+      "组 的",
+      "试 图",
+      "\") :\\",
+      "\" ):\\",
+      "▁up dating",
+      "▁upd ating",
+      "▁sup pliers",
+      "▁suppl iers",
+      "▁supplier s",
+      "▁d rums",
+      "▁dr ums",
+      "▁drum s",
+      "▁L ater",
+      "▁La ter",
+      "▁Lat er",
+      "▁Late r",
+      "iffer ential",
+      "ifferent ial",
+      "不 成",
+      "▁gen tle",
+      "▁gent le",
+      "he at",
+      "hea t",
+      "h eat",
+      "▁in cub",
+      "▁inc ub",
+      "depend ency",
+      "体 检",
+      "本院 依据已经发生法律效力的",
+      "本院依据 已经发生法律效力的",
+      "携 手",
+      "mo uth",
+      "m outh",
+      "▁vent ure",
+      "▁ven ture",
+      "▁ venture",
+      "s ocial",
+      "Res ources",
+      "Re sources",
+      "Resource s",
+      "▁Connect icut",
+      "ver ted",
+      "vert ed",
+      "verte d",
+      "▁ble eding",
+      "▁gr ounds",
+      "▁ground s",
+      "▁want ing",
+      "▁O ptions",
+      "▁Option s",
+      "▁Opt ions",
+      "▁ Options",
+      "靠 谱",
+      "▁S arah",
+      "▁Sar ah",
+      "▁Sa rah",
+      "▁Sara h",
+      "▁tr usted",
+      "▁trust ed",
+      "▁Opt im",
+      "▁Op tim",
+      "▁ Optim",
+      "▁walk ed",
+      "模 糊",
+      "编 译",
+      "给 原告",
+      "bal ance",
+      "operator name",
+      "▁as ks",
+      "▁ask s",
+      "▁ asks",
+      ": #",
+      "富 有",
+      "▁gen omic",
+      "透 支",
+      "交 往",
+      "十八 日",
+      "十 八日",
+      "高 职",
+      "▁c ha",
+      "▁ch a",
+      "▁ cha",
+      "▁reg istry",
+      "▁registr y",
+      "▁ registry",
+      ">\" ,\\",
+      ">\", \\",
+      "> \",\\",
+      "偶 像",
+      "▁form ulas",
+      "▁formula s",
+      "con figure",
+      "config ure",
+      "conf igure",
+      "▁data sets",
+      "▁dat asets",
+      "▁dataset s",
+      "▁datas ets",
+      "▁ datasets",
+      "▁Labor atory",
+      "ord inary",
+      "ple mented",
+      "plement ed",
+      "企业 管理",
+      "选择 了",
+      "竞 技",
+      "常用 的",
+      "常 用的",
+      "▁add ressing",
+      "▁address ing",
+      "▁addr essing",
+      "nUn ited",
+      "nUnit ed",
+      "n United",
+      "很 简单",
+      "深 的",
+      "R F",
+      "fil led",
+      "fill ed",
+      "f illed",
+      "研 制",
+      "涵 盖",
+      "形成 的",
+      "形 成的",
+      "▁highlight ed",
+      "▁Custom er",
+      "▁ Customer",
+      "▁Fr amework",
+      "▁Frame work",
+      "▁ Framework",
+      "▁t omat",
+      "▁to mat",
+      "▁tom at",
+      "亲 自",
+      "it one",
+      "ito ne",
+      "iton e",
+      "i tone",
+      "▁sal ary",
+      "Sh ort",
+      "看守 所",
+      "▁hab itat",
+      "▁habit at",
+      "▁c one",
+      "▁con e",
+      "▁co ne",
+      "▁ cone",
+      "▁Res erved",
+      "▁Reserve d",
+      "▁Reserv ed",
+      "▁gener ations",
+      "▁generation s",
+      "▁Col umn",
+      "▁ Column",
+      "南 海",
+      "trans late",
+      "transl ate",
+      "故意伤害 罪",
+      "▁rec ipients",
+      "▁recip ients",
+      "▁recipient s",
+      "特朗 普",
+      "ap prox",
+      "app rox",
+      "appro x",
+      "ph oto",
+      "phot o",
+      "▁label ed",
+      "▁ labeled",
+      "民 营",
+      "立 足",
+      "ap is",
+      "api s",
+      "a pis",
+      "▁f itting",
+      "▁fit ting",
+      "▁ fitting",
+      "的 我",
+      "Su m",
+      "S um",
+      "ac tual",
+      "act ual",
+      "▁p seud",
+      "备 选答案",
+      "▁sus pected",
+      "▁susp ected",
+      "▁suspect ed",
+      "▁[ `",
+      "它们 的",
+      "它 们的",
+      "▁T ools",
+      "▁To ols",
+      "▁Tool s",
+      "▁Too ls",
+      "▁ Tools",
+      "▁order ing",
+      "▁ord ering",
+      "▁ ordering",
+      "寻 求",
+      "▁url lib",
+      "▁ur llib",
+      "▁ urllib",
+      "▁Re volution",
+      "▁Rev olution",
+      "▁cred entials",
+      "▁credential s",
+      "▁ credentials",
+      "▁ce iling",
+      "▁ceil ing",
+      "▁ass essing",
+      "▁assess ing",
+      "见 面",
+      "郑州 市",
+      "郑 州市",
+      "月二十 六日",
+      "科 目",
+      "▁M ach",
+      "▁Mac h",
+      "▁Ma ch",
+      "Fi re",
+      "F ire",
+      "联合 国",
+      "dest roy",
+      "▁B os",
+      "▁Bo s",
+      "和 技术",
+      "end ors",
+      "endor s",
+      "endo rs",
+      "年 后",
+      "as a",
+      "a sa",
+      "多 多",
+      "L V",
+      "▁Author ity",
+      "的 表现",
+      "ru by",
+      "rub y",
+      "第一百 三",
+      "▁con vention",
+      "▁conv ention",
+      "▁conven tion",
+      "主要 有",
+      "主 要有",
+      "ori ented",
+      "orient ed",
+      "rom agnetic",
+      "的 孩子",
+      "▁L ower",
+      "▁Lo wer",
+      "▁Low er",
+      "▁ Lower",
+      "▁st ood",
+      "▁sto od",
+      "▁ stood",
+      "ol ph",
+      "nN e",
+      "n Ne",
+      "▁E D",
+      "▁ ED",
+      "仪 表",
+      "十 几",
+      "三十 日",
+      "三 十日",
+      "NE CT",
+      "N ECT",
+      "bar a",
+      "ba ra",
+      "b ara",
+      "Bas ic",
+      "B asic",
+      "新 加坡",
+      "ol utions",
+      "olution s",
+      "QU EST",
+      "QUE ST",
+      "Q UEST",
+      "var phi",
+      "许可 证",
+      "р е",
+      "” （",
+      "之 际",
+      "对 接",
+      "时 空",
+      "人员 的",
+      "人 员的",
+      "▁b ugs",
+      "▁bu gs",
+      "▁bug s",
+      "▁ bugs",
+      "pi ne",
+      "pin e",
+      "p ine",
+      "解决 问题",
+      "b n",
+      "期 内",
+      "▁ju ice",
+      "▁multi plication",
+      "az i",
+      "a zi",
+      "▁S V",
+      "▁ SV",
+      "▁i rr",
+      "▁ir r",
+      "润 滑",
+      "载 体",
+      "十一 章",
+      "十 一章",
+      "中 介",
+      "对 抗",
+      "* }",
+      "把 它",
+      "re leases",
+      "release s",
+      "cp u",
+      "c pu",
+      "建设 工程",
+      "▁R ural",
+      "▁Ru ral",
+      "▁compet itors",
+      "▁competitor s",
+      "▁competit ors",
+      "溶 液",
+      "▁c rop",
+      "▁cr op",
+      "▁cro p",
+      "▁ crop",
+      "▁ev ident",
+      "▁evid ent",
+      "Con structor",
+      "Construct or",
+      "不 必",
+      "ro utes",
+      "rou tes",
+      "route s",
+      "r outes",
+      "▁C ulture",
+      "▁Cul ture",
+      "▁Cult ure",
+      "▁ Culture",
+      "人们 的",
+      "人 们的",
+      "p d",
+      "qu al",
+      "qua l",
+      "q ual",
+      "se udo",
+      "seud o",
+      "graph ics",
+      "graphic s",
+      "▁elucid ate",
+      "▁enum erate",
+      "▁enumer ate",
+      "▁ enumerate",
+      "oe nix",
+      "oen ix",
+      "▁sh adow",
+      "▁sha dow",
+      "▁ shadow",
+      "顶 点",
+      "▁B etween",
+      "▁Bet ween",
+      "▁ Between",
+      "▁time stamp",
+      "▁tim estamp",
+      "▁ timestamp",
+      "▁de put",
+      "▁dep ut",
+      "Const ants",
+      "Constant s",
+      "▁rest ored",
+      "▁restore d",
+      "▁restor ed",
+      "选 取",
+      "Element s",
+      "El ements",
+      "Elem ents",
+      "Ele ments",
+      "E lements",
+      "Sam ple",
+      "S ample",
+      "▁franch ise",
+      "第一百四十四 条之规定",
+      "第一百四十四条 之规定",
+      "▁campaign s",
+      "▁dec lare",
+      "▁decl are",
+      "▁declar e",
+      "▁ declare",
+      "▁government s",
+      "▁govern ments",
+      "▁In nov",
+      "▁Inn ov",
+      "Char acter",
+      "早 就",
+      "re ach",
+      "rea ch",
+      "r each",
+      "▁An thony",
+      "▁Anth ony",
+      "美 人",
+      "res istant",
+      "▁anal yzing",
+      "▁analy zing",
+      "▁analyz ing",
+      "的 点",
+      "ut ers",
+      "ute rs",
+      "uter s",
+      "u ters",
+      "远 远",
+      "目的 是",
+      "目 的是",
+      "▁B ra",
+      "▁Br a",
+      "nav bar",
+      "▁min istry",
+      "▁mini stry",
+      "不 适",
+      "nD ist",
+      "nDis t",
+      "nDi st",
+      "n Dist",
+      "it ored",
+      "itor ed",
+      "ito red",
+      "▁N ormal",
+      "▁Nor mal",
+      "▁Norm al",
+      "▁ Normal",
+      "▁de feat",
+      "▁def eat",
+      "▁defe at",
+      "▁br ush",
+      "▁bru sh",
+      "▁ brush",
+      "▁b anking",
+      "▁bank ing",
+      "▁ban king",
+      "至今 未",
+      "开发 商",
+      "上 来",
+      "▁个 人",
+      "▁ 个人",
+      "▁minor ity",
+      "越 大",
+      "▁den ied",
+      "il de",
+      "ild e",
+      "♪ \\",
+      "New s",
+      "Ne ws",
+      "N ews",
+      "组 建",
+      "nm y",
+      "n my",
+      "▁n u",
+      "▁ nu",
+      "淘 汰",
+      "▁a uction",
+      "▁au ction",
+      "市中级人民法院 提出上诉",
+      "任 免",
+      "年 级",
+      "i u",
+      "AC C",
+      "A CC",
+      "fr action",
+      "frac tion",
+      "fra ction",
+      "h n",
+      "▁c omic",
+      "▁com ic",
+      "▁co mic",
+      "个小 时",
+      "个 小时",
+      "s d",
+      "▁R F",
+      "▁ RF",
+      "Pro gress",
+      "▁ 双",
+      "催 要",
+      "没有 什么",
+      "没 有什么",
+      "立 刻",
+      "gi ving",
+      "g iving",
+      "▁Qu ery",
+      "▁Que ry",
+      "▁ Query",
+      "▁Tour nament",
+      "▁fast est",
+      "▁M atch",
+      "▁Mat ch",
+      "▁ Match",
+      "urg ical",
+      "金融 机构",
+      "在 校",
+      "semb led",
+      "sem bled",
+      "semble d",
+      "sembl ed",
+      "▁poll ution",
+      "▁pollut ion",
+      "\" {",
+      "▁B ird",
+      "▁Bi rd",
+      "▁Bir d",
+      "先进 的",
+      "先 进的",
+      "so uth",
+      "s outh",
+      "[] (",
+      "[ ](",
+      "▁vis itor",
+      "▁visit or",
+      "▁ visitor",
+      "▁L ang",
+      "▁La ng",
+      "▁Lan g",
+      "▁ Lang",
+      "Up load",
+      "U pload",
+      "ch oice",
+      "cho ice",
+      "▁p arks",
+      "▁par ks",
+      "▁park s",
+      "得 很",
+      "…… ……",
+      "区 块",
+      "极 其",
+      "▁S u",
+      "▁ Su",
+      "▁J ason",
+      "▁Ja son",
+      "He alth",
+      "▁s team",
+      "▁ste am",
+      "▁ steam",
+      "激 情",
+      "▁r ounds",
+      "▁round s",
+      "Run ner",
+      "R unner",
+      "的小 说",
+      "的 小说",
+      "价 的",
+      "er on",
+      "ero n",
+      "e ron",
+      "▁se am",
+      "▁sea m",
+      "潜 力",
+      "▁in struct",
+      "▁i nstruct",
+      "▁inst ruct",
+      "▁instr uct",
+      "▁l yrics",
+      "▁ly rics",
+      "▁lyric s",
+      "即 便",
+      "等 到",
+      "Sc ale",
+      "S cale",
+      "▁S orry",
+      "▁Sor ry",
+      "▁ Sorry",
+      "▁Fil ter",
+      "▁ Filter",
+      "is tol",
+      "ist ol",
+      "isto l",
+      "▁cour ts",
+      "▁court s",
+      "Cr ed",
+      "Cre d",
+      "C red",
+      "▁lymph ocytes",
+      "▁lymphocyte s",
+      "ong oose",
+      "ongo ose",
+      "多年 来",
+      "多 年来",
+      "服务 有限公司",
+      "re ak",
+      "rea k",
+      "ri ving",
+      "riv ing",
+      "r iving",
+      "rel ative",
+      "▁Base ball",
+      "re ll",
+      "rel l",
+      "r ell",
+      "额 外",
+      "▁ €",
+      "▁A gricult",
+      "▁Agr icult",
+      "▁con straint",
+      "▁constr aint",
+      "▁ constraint",
+      "▁P ap",
+      "▁Pa p",
+      "rib utions",
+      "ribut ions",
+      "ribution s",
+      "▁dis contin",
+      "▁d uplicate",
+      "▁du plicate",
+      "▁duplic ate",
+      "▁ duplicate",
+      "gu i",
+      "g ui",
+      "inc orporated",
+      "nM e",
+      "n Me",
+      "▁K ind",
+      "▁Kin d",
+      "▁Ki nd",
+      "▁ Kind",
+      "▁h ill",
+      "▁hi ll",
+      "▁hil l",
+      "▁ hill",
+      "▁advert is",
+      "▁distingu ish",
+      "▁dec lined",
+      "▁decl ined",
+      "▁decline d",
+      "▁Argent ina",
+      "家 电",
+      "的 部分",
+      "XX XX",
+      "XXX X",
+      "X XXX",
+      "图 案",
+      "▁User s",
+      "▁Use rs",
+      "▁Us ers",
+      "▁ Users",
+      "个人 的",
+      "个 人的",
+      "▁y east",
+      "▁ye ast",
+      "▁just ify",
+      "▁ justify",
+      "▁scr atch",
+      "▁scrat ch",
+      "▁ scratch",
+      "▁m os",
+      "▁mo s",
+      "▁ mos",
+      "▁L ag",
+      "▁La g",
+      "▁ex terior",
+      "▁ext erior",
+      "置业 有限公司",
+      "置 业有限公司",
+      "ic y",
+      "i cy",
+      "释 义",
+      "Fa c",
+      "F ac",
+      "Conn ect",
+      "▁sub scrib",
+      "▁subs crib",
+      "nL a",
+      "n La",
+      "▁l ady",
+      "▁la dy",
+      "▁s heets",
+      "▁she ets",
+      "▁sheet s",
+      "▁ sheets",
+      "籍 贯",
+      "活动 中",
+      "▁rub ber",
+      "To pic",
+      "Top ic",
+      "T opic",
+      "散 热",
+      "合同 纠纷",
+      "▁p recip",
+      "▁prec ip",
+      "▁dis aster",
+      "▁tom ography",
+      "▁O range",
+      "▁Or ange",
+      "佛 教",
+      "▁Us age",
+      "▁ Usage",
+      "▁\\\\ ;",
+      "▁ \\\\;",
+      "截 图",
+      "▁f et",
+      "▁fe t",
+      "lo qu",
+      "l oqu",
+      "深 受",
+      "高 铁",
+      "ad en",
+      "ade n",
+      "a den",
+      "pl ier",
+      "p lier",
+      "▁sil ent",
+      "▁N ation",
+      "▁Na tion",
+      "▁Nat ion",
+      "▁virtual ly",
+      "▁virt ually",
+      "ker nel",
+      "k ernel",
+      "lo aded",
+      "load ed",
+      "▁art istic",
+      "▁artist ic",
+      "优 越",
+      "dyn amic",
+      "dynam ic",
+      "d ynamic",
+      "发布 的",
+      "像 是",
+      "预 约",
+      "调 用",
+      "市 人民检察院",
+      "un i",
+      "u ni",
+      "it ol",
+      "ito l",
+      "i tol",
+      "▁init iation",
+      "▁initi ation",
+      "X Y",
+      "您 认为",
+      "▁M arine",
+      "▁Mar ine",
+      "▁Ma rine",
+      "▁Mari ne",
+      "▁Marin e",
+      "▁re aches",
+      "▁reach es",
+      "G I",
+      "▁B h",
+      "水平 的",
+      "记 数",
+      "▁accept ance",
+      "不 高",
+      "nS outh",
+      "nSo uth",
+      "n South",
+      "Pl atform",
+      "U K",
+      "▁p ad",
+      "▁pa d",
+      "▁ pad",
+      "▁E state",
+      "▁Est ate",
+      "▁Form ula",
+      "▁ Formula",
+      "▁ 唐",
+      "G D",
+      "▁Spir it",
+      "创 立",
+      "▁con gr",
+      "▁co ngr",
+      "▁cong r",
+      "▁public ly",
+      "▁pen alty",
+      "▁penal ty",
+      "▁ph rase",
+      "▁ phrase",
+      "ma g",
+      "m ag",
+      "这 项",
+      "前 言",
+      "制造 业",
+      "fl ash",
+      "f lash",
+      "in ance",
+      "ina nce",
+      "▁Al zheimer",
+      "ip pet",
+      "ipp et",
+      "西 藏",
+      "pe s",
+      "p es",
+      "▁Is raeli",
+      "▁Isra eli",
+      "▁Israel i",
+      "Sec ret",
+      "没有 人",
+      "没 有人",
+      "▁land ing",
+      "▁lan ding",
+      "的 概念",
+      "招 标",
+      "口 碑",
+      "较好 的",
+      "较 好的",
+      "▁Int roduction",
+      "▁Intro duction",
+      "▁ Introduction",
+      "oj i",
+      "o ji",
+      "▁ne west",
+      "▁new est",
+      "rec ision",
+      "达到 了",
+      "达 到了",
+      "▁E asy",
+      "▁Eas y",
+      "▁ Easy",
+      "▁hy pert",
+      "▁hyp ert",
+      "▁hyper t",
+      "▁hype rt",
+      "住 四川省",
+      "TO KEN",
+      "li est",
+      "lie st",
+      "lies t",
+      "l iest",
+      "vis ible",
+      "v isible",
+      "明 年",
+      "▁G P",
+      "▁ GP",
+      "可 达",
+      "nd ate",
+      "nda te",
+      "ndat e",
+      "n date",
+      "▁discrim ination",
+      "▁F lu",
+      "▁Fl u",
+      "▁ Flu",
+      "▁vac uum",
+      "▁vacu um",
+      "源 于",
+      "▁G lass",
+      "▁Gl ass",
+      "▁Glas s",
+      "▁P ain",
+      "▁Pa in",
+      "▁Ad ded",
+      "▁Add ed",
+      "▁ Added",
+      "▁ 使用",
+      "的一 部分",
+      "的一部 分",
+      "的 一部分",
+      "▁W ikipedia",
+      "▁N ap",
+      "▁Na p",
+      "工作 室",
+      "Te m",
+      "T em",
+      "▁stim uli",
+      "▁stimul i",
+      "Sp an",
+      "S pan",
+      "min d",
+      "mi nd",
+      "m ind",
+      "nC o",
+      "n Co",
+      "市 政",
+      "职 场",
+      "宏 观",
+      "as ures",
+      "asure s",
+      "asu res",
+      "LI ED",
+      "▁C ut",
+      "▁Cu t",
+      "▁ Cut",
+      "Sc ene",
+      "▁s ink",
+      "▁sin k",
+      "▁si nk",
+      "▁ sink",
+      "▁cons istency",
+      "▁consist ency",
+      "从 中",
+      "nF ull",
+      "n Full",
+      "▁T ABLE",
+      "▁TA BLE",
+      "▁ TABLE",
+      "演 技",
+      "▁p ra",
+      "▁pr a",
+      "▁en coded",
+      "▁enc oded",
+      "▁encode d",
+      "▁ encoded",
+      "▁o h",
+      "▁ oh",
+      "▁b acking",
+      "▁back ing",
+      "▁bac king",
+      "▁tack le",
+      "克 服",
+      "▁b is",
+      "▁bi s",
+      "▁ bis",
+      "都不 会",
+      "都 不会",
+      "rid ge",
+      "r idge",
+      "H i",
+      "小 白",
+      "read only",
+      "▁Re serve",
+      "▁Res erve",
+      "▁Reserv e",
+      "▁Bang l",
+      "▁Ban gl",
+      "\"] )\\",
+      "\"]) \\",
+      "\" ])\\",
+      "病 变",
+      "CON FIG",
+      "CONF IG",
+      "ar tz",
+      "art z",
+      "▁dis semin",
+      "S i",
+      "nR ev",
+      "nRe v",
+      "n Rev",
+      "▁fract ure",
+      "完 结",
+      "() [",
+      "( )[",
+      "▁tal ented",
+      "▁talent ed",
+      "伦 敦",
+      "法 制",
+      "当 今",
+      "百分 之",
+      "百 分之",
+      "巩 固",
+      "开 通",
+      "our ses",
+      "ours es",
+      "ourse s",
+      "o urses",
+      "器 材",
+      "trans ition",
+      "▁P in",
+      "▁Pi n",
+      "▁ Pin",
+      "▁s car",
+      "▁sc ar",
+      "▁ scar",
+      "经济 的",
+      "▁kw args",
+      "▁ kwargs",
+      "偿还 原告",
+      "聚 焦",
+      "本院 作出的",
+      "nG erman",
+      "n German",
+      "▁survey s",
+      "▁surve ys",
+      "走 出",
+      "▁F M",
+      "▁ FM",
+      "▁analy sed",
+      "▁analys ed",
+      "▁analyse d",
+      "size d",
+      "si zed",
+      "s ized",
+      "▁b rick",
+      "▁br ick",
+      "▁pro test",
+      "▁prote st",
+      "▁prot est",
+      "▁Option al",
+      "▁Opt ional",
+      "▁ Optional",
+      "缺 少",
+      "特别 授权",
+      "科学 家",
+      "科 学家",
+      "▁is o",
+      "▁i so",
+      "▁ iso",
+      "第六十七条 第三款",
+      "Ig n",
+      "I gn",
+      "ge tName",
+      "get Name",
+      "Col l",
+      "Co ll",
+      "C oll",
+      "▁Vir tual",
+      "▁ Virtual",
+      "Us age",
+      "▁H im",
+      "▁Hi m",
+      "▁G uid",
+      "▁Gu id",
+      "▁ Guid",
+      "▁spont aneous",
+      "ost ics",
+      "ostic s",
+      "eff icient",
+      "ap olis",
+      "apol is",
+      "apo lis",
+      "Text ure",
+      "Tex ture",
+      "概 况",
+      "▁C ow",
+      "▁Co w",
+      "▁path ology",
+      "AR K",
+      "治疗 组",
+      "▁Pal est",
+      "▁Pa lest",
+      "▁Pale st",
+      "客 厅",
+      "▁organ s",
+      "▁org ans",
+      "偿还 借款",
+      "join t",
+      "jo int",
+      "j oint",
+      "nG ood",
+      "nGo od",
+      "n Good",
+      "建 国",
+      "▁may or",
+      "ol en",
+      "ole n",
+      "o len",
+      "同 行",
+      "▁l am",
+      "▁la m",
+      "▁ lam",
+      "` )",
+      "煤 矿",
+      "lo bin",
+      "lob in",
+      "omb ok",
+      "ombo k",
+      "申请 符合法律规定",
+      "划 分",
+      "od b",
+      "o db",
+      "▁M aine",
+      "▁Main e",
+      "▁Ma ine",
+      "▁Mai ne",
+      "Bo t",
+      "B ot",
+      "隧 道",
+      "▁N ob",
+      "▁No b",
+      "fil ters",
+      "filter s",
+      "▁sett led",
+      "▁settle d",
+      "本 报",
+      "ext ends",
+      "extend s",
+      "月二十 三日",
+      "▁un w",
+      "▁u nw",
+      "▁g ains",
+      "▁gain s",
+      "▁ga ins",
+      "nG r",
+      "n Gr",
+      "▁approxim ate",
+      "▁approx imate",
+      "Y es",
+      "▁ver ification",
+      "▁proport ional",
+      "▁proportion al",
+      "▁techn ological",
+      "这 句话",
+      "▁n ights",
+      "▁night s",
+      "▁S EO",
+      "▁SE O",
+      "▁ SEO",
+      "an ean",
+      "ane an",
+      "▁mut ants",
+      "▁mutant s",
+      "▁cre ativity",
+      "▁creat ivity",
+      "的 标准",
+      "aj ax",
+      "aja x",
+      "a jax",
+      "▁r oyal",
+      "▁ro yal",
+      "▁manufact ured",
+      "▁manufacture d",
+      "▁time ly",
+      "▁tim ely",
+      "▁▁ ▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁ ▁",
+      "▁▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁▁",
+      "▁ ▁▁▁▁▁▁▁▁▁▁",
+      "pt ide",
+      "combin ator",
+      "comb inator",
+      "nG reat",
+      "nGr eat",
+      "n Great",
+      "Pro p",
+      "Pr op",
+      "P rop",
+      "判决 生效后",
+      "一 群",
+      "扩 张",
+      "▁ 详情",
+      "▁Pr actice",
+      "▁Pract ice",
+      "地 域",
+      "总 数",
+      "am el",
+      "ame l",
+      "a mel",
+      "确 诊",
+      "振 动",
+      "及其 他",
+      "及 其他",
+      "▁m erc",
+      "▁me rc",
+      "▁mer c",
+      "▁ merc",
+      "um at",
+      "uma t",
+      "u mat",
+      "▁B ag",
+      "▁Ba g",
+      "▁ Bag",
+      "▁Cl ark",
+      "▁Clar k",
+      "▁Th ink",
+      "▁Thin k",
+      "▁ Think",
+      "务 必",
+      "手 工",
+      "▁un nec",
+      "г о",
+      "me l",
+      "m el",
+      "毒品 罪",
+      "▁B aron",
+      "▁Bar on",
+      "▁Ba ron",
+      "▁text s",
+      "▁tex ts",
+      "▁Mult iple",
+      "▁Multi ple",
+      "▁ Multiple",
+      "▁ch urches",
+      "▁church es",
+      "市 人",
+      "nA B",
+      "n AB",
+      "家 协会",
+      "▁G ab",
+      "▁Ga b",
+      "▁S tage",
+      "▁St age",
+      "▁Sta ge",
+      "▁ Stage",
+      "ot ropic",
+      "otrop ic",
+      "otr opic",
+      "把 我",
+      "火车 站",
+      "火 车站",
+      "um ar",
+      "uma r",
+      "u mar",
+      "的高 度",
+      "的 高度",
+      "▁secret ary",
+      "▁coll ecting",
+      "▁collect ing",
+      "球 的",
+      "亮 点",
+      "uff le",
+      "▁ на",
+      "ent ially",
+      "ential ly",
+      "enti ally",
+      "▁feas ible",
+      "in puts",
+      "input s",
+      "inp uts",
+      "哈哈 哈",
+      "哈 哈哈",
+      "▁b elt",
+      "▁be lt",
+      "▁bel t",
+      "▁ belt",
+      "nPost s",
+      "nPo sts",
+      "n Posts",
+      "法律法规及 监规",
+      "列 车",
+      "St yles",
+      "Sty les",
+      "Style s",
+      "▁nom inated",
+      "▁nomin ated",
+      "}\\ \\\\\\",
+      "}\\\\ \\\\",
+      "} \\\\\\\\",
+      "网 址",
+      "圆 心",
+      "意 的",
+      "PD F",
+      "P DF",
+      "▁An ne",
+      "▁Ann e",
+      "Id s",
+      "I ds",
+      "▁compar ative",
+      "▁con vey",
+      "▁conv ey",
+      "▁conve y",
+      "得 了",
+      "▁c ookie",
+      "▁cook ie",
+      "▁ cookie",
+      "追 踪",
+      "▁s ensing",
+      "▁sens ing",
+      "▁sen sing",
+      "▁flo ating",
+      "▁float ing",
+      "ul ent",
+      "ule nt",
+      "项 和",
+      "组成 部分",
+      "▁circ uits",
+      "▁circuit s",
+      "产 物",
+      "有 几",
+      "lov ak",
+      "~~~~ ~~~~",
+      "▁pr ints",
+      "▁print s",
+      "▁ prints",
+      "X R",
+      "孕 妇",
+      "nK ey",
+      "nKe y",
+      "n Key",
+      "▁m edic",
+      "▁me dic",
+      "▁med ic",
+      "▁medi c",
+      "nb y",
+      "n by",
+      "▁bi opsy",
+      "▁bio psy",
+      "▁C onc",
+      "▁Con c",
+      "▁Co nc",
+      "环境 保护",
+      "▁fl ask",
+      "▁ flask",
+      "▁Cl assic",
+      "▁Class ic",
+      "▁le git",
+      "▁leg it",
+      "统 治",
+      "▁h anging",
+      "▁hang ing",
+      "▁han ging",
+      "▁illustr ate",
+      "▁illust rate",
+      "感 冒",
+      "月 参加",
+      "▁tob acco",
+      "▁pass ionate",
+      "▁passion ate",
+      "▁b il",
+      "▁bi l",
+      "▁ bil",
+      "erc ul",
+      "▁Re lease",
+      "▁ Release",
+      "▁em otions",
+      "▁emot ions",
+      "▁emotion s",
+      "En v",
+      "E nv",
+      "▁R ain",
+      "▁Ra in",
+      "▁ Rain",
+      "▁st arring",
+      "▁star ring",
+      "▁threat s",
+      "兴 奋",
+      "质 疑",
+      "sa l",
+      "s al",
+      "安 置",
+      "ee r",
+      "e er",
+      "gr eat",
+      "gre at",
+      "g reat",
+      "▁R ole",
+      "▁Ro le",
+      "▁Rol e",
+      "▁ Role",
+      "▁subst rates",
+      "▁substr ates",
+      "▁substrate s",
+      "外 人",
+      "al ias",
+      "ali as",
+      "alia s",
+      "') )",
+      "' ))",
+      "▁alumin um",
+      "▁confirm ation",
+      "▁ confirmation",
+      "监护 的",
+      "▁p rox",
+      "▁pro x",
+      "▁pr ox",
+      "▁ prox",
+      "Event Listener",
+      "劳动 力",
+      "劳 动力",
+      "尼 斯",
+      "▁adhes ion",
+      "MA GE",
+      "MAG E",
+      "M AGE",
+      "▁S DK",
+      "▁SD K",
+      "▁ SDK",
+      "▁pro spect",
+      "▁prosp ect",
+      "▁pros pect",
+      "g p",
+      "用 地",
+      "都 被",
+      "ul i",
+      "u li",
+      "离婚 纠纷一案",
+      "▁brief ly",
+      "在全 国",
+      "在 全国",
+      "time r",
+      "tim er",
+      "ti mer",
+      "t imer",
+      "▁M other",
+      "▁Mot her",
+      "▁Mo ther",
+      "▁S eg",
+      "▁Se g",
+      "▁ Seg",
+      "Mat erial",
+      "Mate rial",
+      "M aterial",
+      "▁th rown",
+      "▁throw n",
+      "▁thr own",
+      "▁thro wn",
+      "向本院申请 复议一次",
+      "▁B io",
+      "▁Bi o",
+      "▁ Bio",
+      "科学 记数",
+      "▁K a",
+      "▁ Ka",
+      "中小 企业",
+      "ill iam",
+      "illi am",
+      "舆 论",
+      "▁retro spective",
+      "▁retrospect ive",
+      "比较 好",
+      "比 较好",
+      "▁h ipp",
+      "▁hip p",
+      "▁hi pp",
+      "▁is lands",
+      "▁island s",
+      "▁v intage",
+      "取 出",
+      "Sym bol",
+      "S ymbol",
+      "▁Set tings",
+      "▁Setting s",
+      "▁Sett ings",
+      "▁ Settings",
+      "夫 人",
+      "Fe ed",
+      "Fee d",
+      "F eed",
+      "容 纳",
+      "法 治",
+      "话 说",
+      "▁S ER",
+      "▁SE R",
+      "▁ SER",
+      "IO Exception",
+      "▁cont ributor",
+      "▁contribut or",
+      "▁ contributor",
+      "apan ese",
+      "也 好",
+      "的 作品",
+      "涉及 未成年",
+      "▁every body",
+      "对 错",
+      "▁Ad vent",
+      "▁Adv ent",
+      "▁product ive",
+      "▁prod uctive",
+      "▁ productive",
+      "克 兰",
+      "<>( );\\",
+      "<> ();\\",
+      "< >();\\",
+      "cont rolled",
+      "control led",
+      "暴 露",
+      "ip tic",
+      "ipt ic",
+      "i ptic",
+      "rs t",
+      "r st",
+      "得 知",
+      "▁n ations",
+      "▁nation s",
+      "▁nat ions",
+      "▁V in",
+      "▁Vi n",
+      "▁l ogical",
+      "▁log ical",
+      "▁logic al",
+      "▁ logical",
+      "▁T un",
+      "▁Tu n",
+      "W ait",
+      "▁f atal",
+      "▁fa tal",
+      "▁fat al",
+      "▁ fatal",
+      "▁M organ",
+      "▁Mor gan",
+      "v g",
+      "Ha ve",
+      "H ave",
+      "Option al",
+      "Opt ional",
+      "这 点",
+      "sa m",
+      "s am",
+      "▁( ,",
+      "的 经济",
+      "厦门 市",
+      "厦 门市",
+      "▁attempt ing",
+      "分 散",
+      "ov ies",
+      "ovie s",
+      "ovi es",
+      "▁May or",
+      "▁Mayo r",
+      "▁ м",
+      "▁adjust ment",
+      "又 不",
+      "nM on",
+      "n Mon",
+      "你 自己",
+      "ce iver",
+      "ceive r",
+      "的范围 内",
+      "的 范围内",
+      "ivid ually",
+      "ividual ly",
+      "hi re",
+      "hir e",
+      "h ire",
+      "▁di min",
+      "▁dim in",
+      "▁Tai wan",
+      "▁port able",
+      "▁por table",
+      "涉及未成年 子女抚养",
+      "N Y",
+      "可 行",
+      "dis abled",
+      "disable d",
+      "nT ype",
+      "n Type",
+      "为大 家",
+      "为 大家",
+      "离婚诉讼 或者",
+      "▁G olf",
+      "▁Go lf",
+      "▁Gol f",
+      "▁soc io",
+      "▁soci o",
+      "at um",
+      "atu m",
+      "a tum",
+      "nSt art",
+      "nSta rt",
+      "n Start",
+      "离婚诉讼或者 涉及未成年子女抚养",
+      "▁read ily",
+      "▁s our",
+      "▁so ur",
+      "▁sou r",
+      "com pass",
+      "comp ass",
+      "at orial",
+      "ator ial",
+      "ato rial",
+      "这本 书",
+      "这 本书",
+      "▁form ulation",
+      "▁formula tion",
+      "成 品",
+      "is sa",
+      "iss a",
+      "ma ps",
+      "map s",
+      "m aps",
+      "兼 容",
+      "无 力",
+      "没 事",
+      "更大 的",
+      "更 大的",
+      "班 子",
+      "Si te",
+      "S ite",
+      "▁ery th",
+      "▁eryt h",
+      "实 质",
+      "精 选",
+      "LO AD",
+      "▁re fund",
+      "▁ref und",
+      "计算 至",
+      "▁sup pression",
+      "▁suppress ion",
+      "ul u",
+      "u lu",
+      "给出 的",
+      "给 出的",
+      "忽 视",
+      "年 开始",
+      "bund le",
+      "b undle",
+      "利 的",
+      "k x",
+      "就可以 了",
+      "提供 担保",
+      "边 界",
+      "▁exp ressing",
+      "▁express ing",
+      "▁expr essing",
+      "▁ expressing",
+      "本院依法组成合议庭 进行了审理",
+      "ec onomic",
+      "econ omic",
+      "合 伙",
+      "向 前",
+      "oc amp",
+      "oca mp",
+      "o camp",
+      "▁pro ton",
+      "▁pr oton",
+      "▁prot on",
+      "▁proto n",
+      "▁ proton",
+      "精 华",
+      "▁$ [",
+      "an non",
+      "ann on",
+      "anno n",
+      "hes ia",
+      "▁tour ism",
+      "we i",
+      "w ei",
+      "com b",
+      "co mb",
+      "c omb",
+      "南 通",
+      "▁For mat",
+      "▁Form at",
+      "▁ Format",
+      "▁e quality",
+      "▁equ ality",
+      "▁equal ity",
+      "▁eq uality",
+      "▁ equality",
+      "▁null able",
+      "▁ nullable",
+      "Back ground",
+      "▁R ick",
+      "▁Ric k",
+      "▁Ri ck",
+      "▁p ole",
+      "▁pol e",
+      "▁po le",
+      "▁ pole",
+      "叶 片",
+      "包含 了",
+      "同比 增长",
+      "▁high way",
+      "▁s wing",
+      "▁sw ing",
+      "▁ swing",
+      "os terone",
+      "oster one",
+      "▁B SD",
+      "▁BS D",
+      "▁ BSD",
+      "▁S cript",
+      "▁Sc ript",
+      "▁Scr ipt",
+      "▁ Script",
+      "C md",
+      "完整 的",
+      "▁Th ings",
+      "▁Thing s",
+      "▁Thin gs",
+      "▁Re lated",
+      "▁Rel ated",
+      "▁ Related",
+      "') )\\",
+      "')) \\",
+      "' ))\\",
+      "▁W HERE",
+      "▁WH ERE",
+      "▁ WHERE",
+      "▁Dub lin",
+      "凝 聚",
+      "Du ration",
+      "D uration",
+      "▁Indust rial",
+      "▁sub process",
+      "▁ subprocess",
+      "▁Louis iana",
+      "▁macroph ages",
+      "▁macrophage s",
+      "▁S umm",
+      "▁Su mm",
+      "▁Sum m",
+      "nP ub",
+      "n Pub",
+      "▁P harm",
+      "▁Ph arm",
+      "W D",
+      "ow ski",
+      "ows ki",
+      "家 乡",
+      "ip y",
+      "i py",
+      "qu et",
+      "que t",
+      "DT O",
+      "D TO",
+      "▁B ush",
+      "▁Bus h",
+      "▁Bu sh",
+      "urg ery",
+      "urger y",
+      "▁h ired",
+      "▁hire d",
+      "▁hi red",
+      "原 子",
+      "▁W y",
+      "▁per mitted",
+      "▁permit ted",
+      "▁perm itted",
+      "bi e",
+      "b ie",
+      "Message s",
+      "Mess ages",
+      "M essages",
+      "～ \\",
+      "VI CE",
+      "V ICE",
+      "二次 方程",
+      "nL ike",
+      "n Like",
+      "л а",
+      "▁land sc",
+      "▁lands c",
+      "一起 来",
+      "一 起来",
+      "▁P hill",
+      "▁Ph ill",
+      "▁Phil l",
+      "ogene ity",
+      "▁S F",
+      "▁ SF",
+      "▁coord ination",
+      "% ）",
+      "ch unk",
+      "G V",
+      "词 典",
+      "的前 提",
+      "的 前提",
+      "▁hypot heses",
+      "▁hypothes es",
+      "▁N ep",
+      "▁Ne p",
+      "Git Hub",
+      "▁chromat ography",
+      "▁\" \",",
+      "▁\"\" ,",
+      "▁ \"\",",
+      "/$ {",
+      "/ ${",
+      "语 法",
+      "it ations",
+      "itation s",
+      "itat ions",
+      "▁mount ed",
+      "▁mo unted",
+      "▁ mounted",
+      "host s",
+      "ho sts",
+      "▁end ogenous",
+      "十六 日",
+      "十 六日",
+      "可 怕",
+      "签 署",
+      "▁Def inition",
+      "▁ Definition",
+      "▁pep per",
+      "▁de ployed",
+      "▁deploy ed",
+      "▁^ {",
+      "▁ ^{",
+      "不断 的",
+      "▁S ET",
+      "▁SE T",
+      "▁ SET",
+      "▁con trad",
+      "▁cont rad",
+      "▁contr ad",
+      "id el",
+      "ide l",
+      "i del",
+      "oid al",
+      "oi dal",
+      "o idal",
+      "▁im pression",
+      "▁imp ression",
+      "▁impress ion",
+      "▁g ross",
+      "▁gr oss",
+      "▁gro ss",
+      "mon itor",
+      "▁key word",
+      "▁ keyword",
+      "▁c ylinder",
+      "▁cyl inder",
+      "▁cylind er",
+      "▁ cylinder",
+      "迎 来",
+      "再 根据",
+      "在 他",
+      "bus iness",
+      "b usiness",
+      "▁st ainless",
+      "▁stain less",
+      "未 到庭",
+      "▁controvers ial",
+      "ub untu",
+      "u buntu",
+      "▁Work s",
+      "▁Wor ks",
+      "▁ Works",
+      "▁Har vard",
+      "nb e",
+      "n be",
+      "urch ase",
+      "pr ises",
+      "pri ses",
+      "prise s",
+      "▁Med al",
+      "▁Me dal",
+      "H I",
+      "W atch",
+      "ag ers",
+      "age rs",
+      "ager s",
+      "a gers",
+      "\", &",
+      "\" ,&",
+      "下一 个",
+      "下 一个",
+      "组 长",
+      "由原告 负担",
+      "requ ests",
+      "request s",
+      "ir s",
+      "i rs",
+      "的公 司",
+      "的 公司",
+      "al bum",
+      "▁moist ure",
+      "▁moistur e",
+      "剂 量",
+      "服 法",
+      "▁S now",
+      "▁Sn ow",
+      "com mons",
+      "comm ons",
+      "common s",
+      "ograph ics",
+      "ographic s",
+      "o graphics",
+      "小 麦",
+      "uk i",
+      "u ki",
+      "ob last",
+      "obl ast",
+      "o blast",
+      "活 着",
+      "com pose",
+      "comp ose",
+      "▁vari eties",
+      "局 面",
+      "执行 的",
+      "执 行的",
+      "这 家",
+      "▁th inks",
+      "▁think s",
+      "▁thin ks",
+      "ific ant",
+      "ifi cant",
+      "一 一",
+      "ho ts",
+      "hot s",
+      "h ots",
+      "▁l ar",
+      "▁la r",
+      "▁ lar",
+      "长春 市",
+      "队 长",
+      "▁long est",
+      "▁lon gest",
+      "开 车",
+      "▁ful fill",
+      "▁fulfil l",
+      "▁B ey",
+      "▁Be y",
+      "极 限",
+      "双方 当事人",
+      "▁employ er",
+      "▁grad ually",
+      "▁gradu ally",
+      "▁gradual ly",
+      "▁mamm alian",
+      "▁infect ious",
+      "▁R isk",
+      "▁Ri sk",
+      "▁Ris k",
+      "▁ Risk",
+      "▁sl ice",
+      "▁slic e",
+      "▁ slice",
+      "▁polynomial s",
+      "王 者",
+      "'] ,",
+      "' ],",
+      "采用 了",
+      "采 用了",
+      "▁c orn",
+      "▁co rn",
+      "▁cor n",
+      "▁ corn",
+      "▁s lip",
+      "▁sl ip",
+      "验 收",
+      "met rics",
+      "metric s",
+      "▁b und",
+      "▁bu nd",
+      "▁ bund",
+      "▁Port land",
+      "潮 流",
+      "▁ch unk",
+      "▁ chunk",
+      "', ['",
+      "遍 历",
+      "mi x",
+      "m ix",
+      "二 等奖",
+      "Go od",
+      "G ood",
+      "au ra",
+      "aur a",
+      "a ura",
+      "mit t",
+      "mi tt",
+      "m itt",
+      "Sub mit",
+      "东 路",
+      "IE NT",
+      "I ENT",
+      "村民 委员会",
+      "卫生 间",
+      "im ore",
+      "imo re",
+      "i more",
+      "▁l inking",
+      "▁link ing",
+      "▁lin king",
+      "▁ linking",
+      "▁att itudes",
+      "▁attitude s",
+      "日向本院 申请执行",
+      "日向本院申请 执行",
+      "日 向本院申请执行",
+      "▁X X",
+      "▁ XX",
+      "ac a",
+      "a ca",
+      "nO k",
+      "n Ok",
+      "born e",
+      "bo rne",
+      "bor ne",
+      "b orne",
+      "be cause",
+      "bec ause",
+      "b ecause",
+      "承 办",
+      "重 生",
+      "gl ob",
+      "g lob",
+      "si l",
+      "s il",
+      "座 谈",
+      "▁J u",
+      "▁tra iler",
+      "▁trail er",
+      "▁w eren",
+      "▁we ren",
+      "▁were n",
+      "▁wer en",
+      "at ility",
+      "a tility",
+      "死 的",
+      "加 深",
+      "▁gl obe",
+      "▁glob e",
+      "▁glo be",
+      "海 上",
+      "nH er",
+      "nHe r",
+      "n Her",
+      "远 离",
+      "CE S",
+      "C ES",
+      "出 于",
+      "▁\\\\ ((",
+      "▁\\\\( (",
+      "▁ \\\\((",
+      "▁W ind",
+      "▁Win d",
+      "▁Wi nd",
+      "▁ Wind",
+      "r x",
+      "tf mt",
+      "t fmt",
+      "城市 的",
+      "城 市的",
+      "pr ite",
+      "pri te",
+      "prit e",
+      "p rite",
+      "Ma ster",
+      "Mas ter",
+      "M aster",
+      "中 原",
+      "pub lished",
+      "publish ed",
+      "▁d ens",
+      "▁de ns",
+      "▁den s",
+      "▁ dens",
+      "▁haz ard",
+      "am o",
+      "a mo",
+      "Form atter",
+      "For matter",
+      "Format ter",
+      "▁flu or",
+      "▁ fluor",
+      "▁band width",
+      "▁bench mark",
+      "▁ benchmark",
+      "P i",
+      "▁A bs",
+      "▁Ab s",
+      "▁ Abs",
+      "it i",
+      "i ti",
+      "▁J osh",
+      "▁Jo sh",
+      "▁Jos h",
+      "os ine",
+      "osin e",
+      "osi ne",
+      "牡 丹",
+      "▁A WS",
+      "▁AW S",
+      "▁ AWS",
+      "▁All iance",
+      "k u",
+      "ge nder",
+      "gen der",
+      "g ender",
+      "成 语",
+      "ndes cription",
+      "nde scription",
+      "n description",
+      "to wn",
+      "t own",
+      "▁Ga ussian",
+      "▁Gauss ian",
+      "▁ Gaussian",
+      "▁inter faces",
+      "▁interface s",
+      "▁interf aces",
+      "▁ interfaces",
+      "Se cond",
+      "Sec ond",
+      "▁A ch",
+      "▁Ac h",
+      "an ne",
+      "ann e",
+      "ev alu",
+      "eval u",
+      "eva lu",
+      "不 为",
+      "的有 效",
+      "的 有效",
+      "▁exam ining",
+      "不一 定",
+      "不 一定",
+      "▁C ool",
+      "▁Co ol",
+      "▁re current",
+      "▁rec urrent",
+      "Pa ck",
+      "Pac k",
+      "P ack",
+      "日立案 后",
+      "is son",
+      "iss on",
+      "i sson",
+      "▁d urable",
+      "▁dur able",
+      "劳动 争议",
+      "意思 表示",
+      "了 几",
+      "▁got ten",
+      "▁School s",
+      "▁Sch ools",
+      "_ ;\\",
+      "▁题目 描述",
+      "Ch e",
+      "C he",
+      "\\' /",
+      "\\ '/",
+      "解析 式",
+      "十二 日",
+      "十 二日",
+      "索 引",
+      "过 渡",
+      "▁S I",
+      "▁ SI",
+      "Ba d",
+      "B ad",
+      "▁T ab",
+      "▁Ta b",
+      "▁ Tab",
+      "▁n urse",
+      "▁nurs e",
+      "▁nur se",
+      "▁Im mun",
+      "▁Imm un",
+      "nC har",
+      "nCh ar",
+      "n Char",
+      "▁St ew",
+      "▁Ste w",
+      "cl inical",
+      "骑 士",
+      "正 面",
+      "约定 的",
+      "约 定的",
+      "需要 注意",
+      "需 要注意",
+      "▁sim pler",
+      "▁simple r",
+      "▁simpl er",
+      "过 多",
+      "限 于",
+      "per cent",
+      "摩 擦",
+      "is sing",
+      "iss ing",
+      "▁Chair man",
+      "▁ «",
+      "涂 料",
+      "li sten",
+      "list en",
+      "lis ten",
+      "▁Com plete",
+      "▁Comp lete",
+      "▁Compl ete",
+      "▁ Complete",
+      "本 土",
+      "就 已经",
+      "▁g erm",
+      "▁ge rm",
+      "▁ger m",
+      "精 灵",
+      "St ruct",
+      "Str uct",
+      "▁prom otes",
+      "▁promot es",
+      "▁promote s",
+      "▁promo tes",
+      "▁hyp ox",
+      "▁f ork",
+      "▁for k",
+      "▁fo rk",
+      "▁ fork",
+      "▁T oken",
+      "▁To ken",
+      "▁Tok en",
+      "▁ Token",
+      "从 未",
+      "LA B",
+      "L AB",
+      "▁progn osis",
+      "▁phys ically",
+      "▁physical ly",
+      "▁physic ally",
+      "付 清",
+      "欢 乐",
+      "const ants",
+      "constant s",
+      "▁Be ing",
+      "▁ Being",
+      "眼 里",
+      "▁B ot",
+      "▁Bo t",
+      "▁ Bot",
+      "▁Co ord",
+      "▁ Coord",
+      "▁str uck",
+      "猪 肉",
+      "▁p aste",
+      "▁past e",
+      "▁pa ste",
+      "▁pas te",
+      "▁ paste",
+      "nB est",
+      "nBe st",
+      "n Best",
+      "nE nter",
+      "nEn ter",
+      "nEnt er",
+      "n Enter",
+      "Write Line",
+      "▁instance of",
+      "PL C",
+      "P LC",
+      "七日 内",
+      "七 日内",
+      "▁di agonal",
+      "▁diagon al",
+      "▁N an",
+      "▁Na n",
+      "月二十 二日",
+      "▁ref erring",
+      "▁refer ring",
+      "T B",
+      "去 的",
+      "的 图象",
+      "该 如何",
+      "手 法",
+      ">\" \\",
+      "> \"\\",
+      "第七 章",
+      "） ÷",
+      "▁sh aped",
+      "▁shape d",
+      "▁sha ped",
+      "▁ shaped",
+      "▁c ortical",
+      "▁cor tical",
+      "▁cort ical",
+      "▁cortic al",
+      "nM ake",
+      "nMa ke",
+      "n Make",
+      "养 成",
+      "▁reg arded",
+      "▁regard ed",
+      "▁dis placement",
+      "place holder",
+      "In ternal",
+      "Int ernal",
+      "Inter nal",
+      "产生 了",
+      "产 生了",
+      "实际 情况",
+      "▁ °",
+      "说 过",
+      "▁H ub",
+      "▁Hu b",
+      "▁ Hub",
+      "▁LIMIT ED",
+      "▁l oose",
+      "▁lo ose",
+      "▁loos e",
+      "一个 小",
+      "一 个小",
+      "湿 度",
+      "▁Admin istr",
+      "一 层",
+      "武 装",
+      "给 定",
+      "我 有",
+      "nC omp",
+      "nCom p",
+      "nCo mp",
+      "n Comp",
+      "▁cap tain",
+      "▁capt ain",
+      "▁p ixels",
+      "▁pixel s",
+      "▁ pixels",
+      "▁r acial",
+      "▁rac ial",
+      "▁ racial",
+      "一 组",
+      "外 面",
+      "AS T",
+      "A ST",
+      "▁B aby",
+      "▁Ba by",
+      "▁Bab y",
+      "▁ Baby",
+      "▁d imensional",
+      "▁dimension al",
+      "▁ dimensional",
+      "日 晚",
+      "登 场",
+      "▁tun nel",
+      "tr igger",
+      "trig ger",
+      "▁La bour",
+      "▁Lab our",
+      "不 止",
+      "nor th",
+      "n orth",
+      "▁How ard",
+      "▁Ho ward",
+      "工 委",
+      "▁b ash",
+      "▁bas h",
+      "▁ba sh",
+      "▁ bash",
+      "物业 服务有限公司",
+      "物业服务 有限公司",
+      "▁a z",
+      "▁ az",
+      "最后 的",
+      "最 后的",
+      "第三 百",
+      "第 三百",
+      "了 他",
+      "美 观",
+      "▁S audi",
+      "▁Sa udi",
+      "▁Sau di",
+      "▁st ayed",
+      "▁stay ed",
+      "▁P arent",
+      "▁Par ent",
+      "▁Pa rent",
+      "▁Pare nt",
+      "▁ Parent",
+      "▁Te chnical",
+      "▁Techn ical",
+      "意 志",
+      "▁dis par",
+      "▁disp ar",
+      "▁cult iv",
+      "▁[ (",
+      "▁ [(",
+      "折 抵",
+      "ha lf",
+      "hal f",
+      "h alf",
+      "') );\\",
+      "')) ;\\",
+      "' ));\\",
+      "之 旅",
+      "▁L ocation",
+      "▁Lo cation",
+      "▁Loc ation",
+      "▁ Location",
+      "▁ar range",
+      "▁arr ange",
+      "▁arrang e",
+      "ad ded",
+      "add ed",
+      "oph ag",
+      "北 部",
+      "▁out break",
+      "▁ou tbreak",
+      "▁con fusion",
+      "▁conf usion",
+      "sp ir",
+      "spi r",
+      "s pir",
+      "▁A sk",
+      "▁As k",
+      "▁ Ask",
+      "▁present ations",
+      "▁presentation s",
+      "拉 斯",
+      "里 斯",
+      "bp s",
+      "b ps",
+      "受 害",
+      "th an",
+      "tha n",
+      "t han",
+      "io id",
+      "i oid",
+      "▁R oc",
+      "▁Ro c",
+      "d w",
+      "▁el abor",
+      "Com pat",
+      "Comp at",
+      "▁( _",
+      "▁ (_",
+      "Co py",
+      "Cop y",
+      "C opy",
+      "▁cons ensus",
+      "▁t ired",
+      "▁tire d",
+      "▁ti red",
+      "于本判决 生效后",
+      "于 本判决生效后",
+      "乃 至",
+      "宽 度",
+      "刑 更",
+      "▁d istr",
+      "▁dis tr",
+      "▁dist r",
+      "▁di str",
+      "▁M ock",
+      "▁Mo ck",
+      "▁ Mock",
+      "▁credit s",
+      "▁cred its",
+      "▁t imer",
+      "▁time r",
+      "▁tim er",
+      "▁ti mer",
+      "▁ timer",
+      "Re lease",
+      "do i",
+      "d oi",
+      "立 法",
+      "▁ach ievement",
+      "▁achieve ment",
+      "才 有",
+      "al er",
+      "ale r",
+      "a ler",
+      "▁G ot",
+      "▁Go t",
+      "▁ Got",
+      "▁s tiff",
+      "▁st iff",
+      "拿 出",
+      "Th anks",
+      "Thank s",
+      "Than ks",
+      "▁pl acing",
+      "▁plac ing",
+      "▁ placing",
+      "荷 兰",
+      "了解 到",
+      "we ar",
+      "w ear",
+      "nSim plify",
+      "nSimpl ify",
+      "n Simplify",
+      "pan ic",
+      "pa nic",
+      "p anic",
+      "▁h iring",
+      "▁hi ring",
+      "olog ous",
+      "Str ategy",
+      "▁dev oted",
+      "▁devote d",
+      "移 植",
+      "nS et",
+      "nSe t",
+      "n Set",
+      "合作 社",
+      "▁prolong ed",
+      "nc y",
+      "n cy",
+      "▁Gal axy",
+      "ia o",
+      "i ao",
+      "▁a men",
+      "▁am en",
+      "▁ amen",
+      "▁pre sum",
+      "▁pres um",
+      "\\\\( )\\\\)\\",
+      "\\\\() \\\\)\\",
+      "\\\\()\\\\) \\",
+      "改革 开放",
+      "联 想",
+      "claim ed",
+      "l r",
+      "is co",
+      "isc o",
+      "io logic",
+      "i ologic",
+      "▁form erly",
+      "▁former ly",
+      "▁ formerly",
+      "▁l etting",
+      "▁let ting",
+      "重 启",
+      "]\\ \\",
+      "] \\\\",
+      "nCONCLUS IONS",
+      "nCONCLUSION S",
+      "ra ns",
+      "ran s",
+      "r ans",
+      "寂 寞",
+      "▁susp ended",
+      "▁suspend ed",
+      "太 大",
+      "感 恩",
+      "H F",
+      "病 理",
+      "的 态度",
+      "语 种",
+      "▁Miss issippi",
+      "救 助",
+      "▁Rec ently",
+      "▁Recent ly",
+      "放 到",
+      "▁P ackage",
+      "▁Pack age",
+      "▁ Package",
+      "stud ent",
+      "▁\" \";\\",
+      "▁\"\" ;\\",
+      "八 条",
+      "平 常",
+      "选出 正确的答案",
+      "有 理",
+      "▁R ab",
+      "▁Ra b",
+      "统 筹",
+      "ma kers",
+      "make rs",
+      "maker s",
+      "mak ers",
+      "m akers",
+      "梅 花",
+      "缩 小",
+      "▁v aries",
+      "▁vari es",
+      "▁var ies",
+      "▁va ries",
+      "考虑 到",
+      "▁T X",
+      "▁ TX",
+      "th rows",
+      "throw s",
+      "thro ws",
+      "thr ows",
+      "象 征",
+      "欧 盟",
+      "内容 介绍",
+      "▁prot ecting",
+      "▁protect ing",
+      "t ensor",
+      "▁Run time",
+      "▁Ru ntime",
+      "▁ Runtime",
+      "命 名",
+      "▁d in",
+      "▁di n",
+      "▁ din",
+      "iz oph",
+      "▁Date Time",
+      "▁ DateTime",
+      "set ting",
+      "sett ing",
+      "s etting",
+      "占 据",
+      "@ @",
+      "人民 医院",
+      "pro blem",
+      "prob lem",
+      "per formance",
+      "perform ance",
+      "▁V el",
+      "▁Ve l",
+      "▁ Vel",
+      "▁plate let",
+      "目 光",
+      "▁N ut",
+      "▁Nu t",
+      "▁a ux",
+      "▁au x",
+      "▁ aux",
+      "是没 有",
+      "是 没有",
+      "nB oth",
+      "nBo th",
+      "n Both",
+      "Int egr",
+      "nO bject",
+      "nOb ject",
+      "n Object",
+      "spec ial",
+      "▁organ isms",
+      "▁organism s",
+      "▁ organisms",
+      "▁ent ert",
+      "▁enter t",
+      "▁S ud",
+      "▁Su d",
+      "▁abund ance",
+      "排 水",
+      "▁accord ance",
+      "这是 一",
+      "这 是一",
+      "▁R ule",
+      "▁Ru le",
+      "▁ Rule",
+      "▁G un",
+      "▁Gu n",
+      "GB T",
+      "G BT",
+      "▁sudden ly",
+      "HE AD",
+      "的 首",
+      "em ed",
+      "eme d",
+      "e med",
+      "com pare",
+      "comp are",
+      "compar e",
+      "att rs",
+      "attr s",
+      "▁d B",
+      "▁ dB",
+      "面 板",
+      "▁TH AT",
+      "ge bras",
+      "gebra s",
+      "ta x",
+      "t ax",
+      "▁Or th",
+      "▁Ort h",
+      "con firm",
+      "conf irm",
+      "重 建",
+      "▁exc essive",
+      "▁excess ive",
+      "▁ 阿",
+      "▁int roducing",
+      "▁introdu cing",
+      "登 陆",
+      "船 舶",
+      "pro to",
+      "pr oto",
+      "prot o",
+      "La b",
+      "L ab",
+      "映 时间",
+      "nst atic",
+      "n static",
+      "▁co unting",
+      "▁count ing",
+      "我 个人",
+      "而 来",
+      "▁be gun",
+      "▁beg un",
+      "班 主任",
+      "▁c rown",
+      "▁cr own",
+      "▁crow n",
+      "▁cro wn",
+      "书 写",
+      "Bind ing",
+      "Bin ding",
+      "B inding",
+      "▁car ries",
+      "▁head quarters",
+      "▁headqu arters",
+      "自 从",
+      "▁j o",
+      "▁ jo",
+      "▁de cent",
+      "▁dec ent",
+      "▁dece nt",
+      "▁ov arian",
+      "▁ovar ian",
+      "月二十 九日",
+      "▁J os",
+      "▁Jo s",
+      "▁D jango",
+      "▁Dj ango",
+      "▁ Django",
+      "商品 房",
+      "十七 日",
+      "十 七日",
+      "经 费",
+      "▁expl ores",
+      "▁explore s",
+      "▁explo res",
+      "▁explor es",
+      "真 诚",
+      "▁L am",
+      "▁La m",
+      "▁f oss",
+      "▁fo ss",
+      "▁V eter",
+      "▁Ve ter",
+      "新 人",
+      "hed uler",
+      "hedul er",
+      "hedule r",
+      "Si de",
+      "S ide",
+      "觉得 自己",
+      "党 校",
+      "in to",
+      "int o",
+      "i nto",
+      "VAL UE",
+      "is i",
+      "i si",
+      "经济 社会",
+      "au rant",
+      "aur ant",
+      "aura nt",
+      "`, \\",
+      "` ,\\",
+      "的不 同",
+      "的 不同",
+      "▁S L",
+      "▁ SL",
+      "FL AGS",
+      "FLAG S",
+      "▁surv ived",
+      "▁surviv ed",
+      "▁survive d",
+      "▁R ober",
+      "▁Rob er",
+      "▁Ro ber",
+      "如 同",
+      "十二 章",
+      "十 二章",
+      "\\' );",
+      "\\') ;",
+      "\\ ');",
+      "CT YPE",
+      "C TYPE",
+      "▁Phil ip",
+      "公 积",
+      "▁Det roit",
+      "十三 日",
+      "十 三日",
+      "▁e in",
+      "▁ ein",
+      "▁Man ufact",
+      "am ped",
+      "amp ed",
+      "re verse",
+      "▁E vents",
+      "▁Ev ents",
+      "▁Even ts",
+      "▁Event s",
+      "▁ Events",
+      "▁fr ustr",
+      "具体 的",
+      "具 体的",
+      "In formation",
+      "▁do nation",
+      "▁don ation",
+      "mus ic",
+      "m usic",
+      "▁Av ailable",
+      "▁ Available",
+      "学 士",
+      "监 会",
+      "ir able",
+      "ira ble",
+      "时 不",
+      "▁Con clusions",
+      "▁Conclusion s",
+      "▁bed rooms",
+      "▁bedroom s",
+      "ound ing",
+      "oun ding",
+      "o unding",
+      "恐 惧",
+      "ab ul",
+      "a bul",
+      "的 全",
+      "眼 神",
+      "配 方",
+      "在 本",
+      "se par",
+      "sep ar",
+      "▁st ocks",
+      "▁stock s",
+      "▁sto cks",
+      "▁A lan",
+      "▁Al an",
+      "▁Her itage",
+      "disc iplinary",
+      "discipl inary",
+      "d isciplinary",
+      "着 力",
+      "la g",
+      "l ag",
+      "▁mon itored",
+      "▁monitor ed",
+      "][ \"",
+      "] [\"",
+      "▁ 将",
+      "▁b eds",
+      "▁be ds",
+      "▁bed s",
+      "▁occasion ally",
+      "▁occasional ly",
+      "▁H ockey",
+      "▁str iking",
+      "▁stri king",
+      "太平 洋",
+      "▁M ini",
+      "▁Min i",
+      "▁Mi ni",
+      "▁ Mini",
+      "▁spec ifications",
+      "▁specific ations",
+      "▁specification s",
+      "▁T alk",
+      "▁Tal k",
+      "▁ Talk",
+      "▁local ized",
+      "▁localize d",
+      "▁table t",
+      "▁tab let",
+      "▁similar ity",
+      "▁t ear",
+      "▁te ar",
+      "▁tea r",
+      "公 元",
+      "后 卫",
+      "▁st reams",
+      "▁stre ams",
+      "▁stream s",
+      "▁ streams",
+      "▁b ypass",
+      "▁by pass",
+      "▁ab dominal",
+      "▁fragment s",
+      "▁frag ments",
+      "发 性",
+      "▁W ire",
+      "▁Wi re",
+      "▁ Wire",
+      "us k",
+      "u sk",
+      "学 家",
+      "og onal",
+      "ogo nal",
+      "ogon al",
+      "▁gu ilty",
+      "▁guilt y",
+      "▁mer ely",
+      "▁mere ly",
+      "▁art work",
+      "并 无",
+      "▁ad diction",
+      "▁add iction",
+      "▁addict ion",
+      "啤 酒",
+      "ip eline",
+      "ipe line",
+      "ipel ine",
+      "Sc h",
+      "S ch",
+      "on line",
+      "o nline",
+      "uff icient",
+      "▁dist ances",
+      "▁distance s",
+      "双 手",
+      "▁Al bert",
+      "▁Alber t",
+      "▁Alb ert",
+      "ad dy",
+      "add y",
+      "if iable",
+      "ifi able",
+      "p matrix",
+      "细 腻",
+      "调 控",
+      "和 管理",
+      "Mi xin",
+      "Mix in",
+      "M ixin",
+      "ain ers",
+      "ai ners",
+      "ainer s",
+      "aine rs",
+      "▁Col omb",
+      "▁ref used",
+      "▁refuse d",
+      "▁pro pri",
+      "▁prop ri",
+      "▁mus ician",
+      "▁music ian",
+      "oh yd",
+      "o hyd",
+      "▁b urst",
+      "▁bu rst",
+      "▁bur st",
+      "▁ burst",
+      "▁Re member",
+      "▁Rem ember",
+      "▁ Remember",
+      "▁S erial",
+      "▁Se rial",
+      "▁Ser ial",
+      "▁ Serial",
+      "b p",
+      "明确 的",
+      "▁inf inity",
+      "▁infinit y",
+      "▁ infinity",
+      "减刑 条件",
+      "▁$ |",
+      "▁a na",
+      "▁an a",
+      "▁ ana",
+      "immun e",
+      "imm une",
+      "Process or",
+      "▁subst ances",
+      "▁substance s",
+      "外 形",
+      "破 碎",
+      "判处 拘役",
+      "向本院提起 上诉",
+      "▁dec omposition",
+      "▁decom position",
+      "给我 们",
+      "给 我们",
+      "▁phot ographer",
+      "▁photograph er",
+      "▁K in",
+      "▁Ki n",
+      "▁ Kin",
+      "▁Wal ker",
+      "▁Walk er",
+      "▁belong ing",
+      "借款 人",
+      "il lo",
+      "ill o",
+      "▁b locking",
+      "▁block ing",
+      "▁ blocking",
+      "向本院 申请执行",
+      "向本院申请 执行",
+      "就是 一个",
+      "就 是一个",
+      "▁Car ib",
+      "▁Ca rib",
+      "▁sh irt",
+      "▁ shirt",
+      "reg ex",
+      "▁B reak",
+      "▁Bre ak",
+      "▁ Break",
+      "[ $",
+      "计 数",
+      "▁F ixed",
+      "▁Fix ed",
+      "▁ Fixed",
+      "一 度",
+      "繁 荣",
+      "一体 化",
+      "甲 基",
+      "▁p adding",
+      "▁pad ding",
+      "▁padd ing",
+      "▁ padding",
+      "▁Ar thur",
+      "▁ Arthur",
+      "高等 教育",
+      "token s",
+      "tok ens",
+      "标准 的",
+      "Prot ocol",
+      "Proto col",
+      "弘 扬",
+      "▁D ick",
+      "▁Di ck",
+      "▁Ent ity",
+      "▁ Entity",
+      "用 人",
+      "▁h ell",
+      "▁he ll",
+      "▁hel l",
+      "▁ hell",
+      "▁t ours",
+      "▁to urs",
+      "▁tour s",
+      "▁tou rs",
+      "举 措",
+      ")) /(",
+      "))/ (",
+      ") )/(",
+      "▁he sit",
+      "con cept",
+      "▁P ast",
+      "▁Pa st",
+      "▁Pas t",
+      "ow ners",
+      "own ers",
+      "owner s",
+      ":\\ '",
+      ": \\'",
+      "yl ation",
+      "yla tion",
+      "y lation",
+      "il ization",
+      "▁s ons",
+      "▁so ns",
+      "▁son s",
+      "▁Bel gium",
+      "▁p ushing",
+      "▁push ing",
+      "碰 撞",
+      "的 速度",
+      "▁r endering",
+      "▁render ing",
+      "▁rend ering",
+      "▁h orses",
+      "▁hor ses",
+      "▁horse s",
+      "▁hors es",
+      "▁ 运动",
+      "▁M erc",
+      "▁Me rc",
+      "▁Mer c",
+      "变 革",
+      "y combinator",
+      "nVill ages",
+      "▁reg ulating",
+      "▁regul ating",
+      "认罪 服法",
+      "ä r",
+      "Do es",
+      "D oes",
+      "▁M ars",
+      "▁Mar s",
+      "▁Ma rs",
+      "▁recombin ant",
+      "to x",
+      "t ox",
+      "Cent er",
+      "Ce nter",
+      "C enter",
+      "年 间",
+      "日立案 受理",
+      "体 质",
+      "十四 日",
+      "十 四日",
+      "▁K ids",
+      "▁Ki ds",
+      "▁Kid s",
+      "▁T rack",
+      "▁Tr ack",
+      "▁Tra ck",
+      "▁ Track",
+      "有一 定的",
+      "有一定 的",
+      "有 一定的",
+      "▁aggreg ate",
+      "▁ aggregate",
+      "▁occasion s",
+      "▁occas ions",
+      "▁re section",
+      "▁res ection",
+      "▁resect ion",
+      "酒店 的",
+      "酒 店的",
+      "涉 案",
+      "▁Fin land",
+      "▁hypothes ize",
+      "lev eland",
+      "level and",
+      "▁s wap",
+      "▁sw ap",
+      "▁ swap",
+      "▁Pre vious",
+      "▁ Previous",
+      "促 销",
+      "分析 和",
+      "▁Dak ota",
+      "阐 述",
+      "▁g od",
+      "▁go d",
+      "▁ god",
+      "▁re jected",
+      "▁reject ed",
+      "份 额",
+      "▁Ag reement",
+      "两 侧",
+      "▁Ch oose",
+      "▁Cho ose",
+      "▁ Choose",
+      "ho le",
+      "hol e",
+      "h ole",
+      "izoph ren",
+      "Dir ection",
+      "Direct ion",
+      "Di rection",
+      "D irection",
+      "▁De ath",
+      "▁ Death",
+      "ch ell",
+      "che ll",
+      "chel l",
+      "c hell",
+      "VE D",
+      "V ED",
+      "ley ball",
+      "cre ens",
+      "creen s",
+      "c reens",
+      "AN GE",
+      "ANG E",
+      "▁R io",
+      "▁Ri o",
+      "ex ternal",
+      "ext ernal",
+      "exter nal",
+      "extern al",
+      "nThere fore",
+      "气 管",
+      "▁t uple",
+      "▁tu ple",
+      "▁ tuple",
+      "给 她",
+      "put s",
+      "pu ts",
+      "p uts",
+      "oh ist",
+      "o hist",
+      "▁T ests",
+      "▁Te sts",
+      "▁Test s",
+      "▁Tes ts",
+      "▁ Tests",
+      "▁l ombok",
+      "ener ative",
+      "ene rative",
+      "] }",
+      "以上 事实",
+      "网 卡",
+      "行政处罚 决定书",
+      "行政处罚决定 书",
+      "LO C",
+      "L OC",
+      "▁re productive",
+      "B Y",
+      "用 量",
+      "注 入",
+      "▁L ane",
+      "▁La ne",
+      "▁Lan e",
+      "▁基本 信息",
+      "▁ 基本信息",
+      "看 似",
+      "he astern",
+      "heast ern",
+      "h eastern",
+      "属 实",
+      "ph ot",
+      "p hot",
+      "▁Q ual",
+      "▁Qu al",
+      "▁ Qual",
+      "▁license s",
+      "▁lic enses",
+      "▁ licenses",
+      "nU nder",
+      "nUn der",
+      "n Under",
+      "▁en emy",
+      "▁Den mark",
+      "电动 车",
+      "电 动车",
+      "▁we apon",
+      "▁weap on",
+      "an che",
+      "anc he",
+      "anch e",
+      "a nche",
+      "配 送",
+      "GRO UND",
+      "▁B owl",
+      "▁Bo wl",
+      "▁Bow l",
+      "end point",
+      "应当 按照",
+      "▁A G",
+      "▁ AG",
+      "厕 所",
+      "时间 为",
+      "▁se iz",
+      "▁Frank lin",
+      "▁employ ers",
+      "▁employer s",
+      "▁l iv",
+      "▁li v",
+      "人 气",
+      "约 为",
+      "▁N CAA",
+      "▁NC AA",
+      "AT CH",
+      "麦 克",
+      "▁D om",
+      "▁Do m",
+      "▁ Dom",
+      "▁\\\\ $",
+      "▁ \\\\$",
+      "▁N ash",
+      "▁Na sh",
+      "▁Nas h",
+      "▁L ux",
+      "▁Lu x",
+      "н ы",
+      "▁\" [",
+      "▁ \"[",
+      "▁po et",
+      "桂 林",
+      "ob a",
+      "o ba",
+      "▁con stitution",
+      "▁const itution",
+      "▁constit ution",
+      "▁constitu tion",
+      "ii i",
+      "i ii",
+      "home page",
+      "hom epage",
+      "欧 美",
+      "▁rev ised",
+      "▁revis ed",
+      "nd e",
+      "n de",
+      "▁f ocal",
+      "▁foc al",
+      "▁fo cal",
+      "下 方",
+      "▁R oot",
+      "▁Ro ot",
+      "▁ Root",
+      "▁V ancouver",
+      "▁per mut",
+      "▁perm ut",
+      "▁Be havior",
+      "▁ Behavior",
+      "面 貌",
+      "la ter",
+      "late r",
+      "lat er",
+      "l ater",
+      "说 出",
+      "催 化",
+      "百 货",
+      "Li ve",
+      "L ive",
+      "▁T rib",
+      "▁Tr ib",
+      "▁Tri b",
+      "з а",
+      "\\' \\\\",
+      "\\'\\ \\",
+      "\\ '\\\\",
+      "\"} \\",
+      "\" }\\",
+      "▁con front",
+      "▁C B",
+      "▁ CB",
+      "▁c ls",
+      "▁cl s",
+      "▁ cls",
+      "▁max imize",
+      "▁maxim ize",
+      "os ures",
+      "osure s",
+      "解放 军",
+      "▁C V",
+      "▁ CV",
+      "发现 了",
+      "nO BJECT",
+      "n OBJECT",
+      "▁nav igate",
+      "▁ navigate",
+      "▁l oud",
+      "▁lo ud",
+      "▁lou d",
+      "ic he",
+      "ich e",
+      "i che",
+      "▁condition ing",
+      "▁ conditioning",
+      "R a",
+      "▁ 时",
+      "进 球",
+      "▁def ensive",
+      "ts burgh",
+      "火 灾",
+      "▁M ovie",
+      "▁Mo vie",
+      "▁ Movie",
+      "▁a ther",
+      "▁at her",
+      "▁ ather",
+      "▁H z",
+      "▁ Hz",
+      "第一百 一十九",
+      "第一百一 十九",
+      "▁Z one",
+      "▁ Zone",
+      "estab lished",
+      "establish ed",
+      "bit al",
+      "bi tal",
+      "b ital",
+      "la unch",
+      "l aunch",
+      "in burgh",
+      "enez uel",
+      "样 品",
+      "再 说",
+      "法律 保护",
+      "▁M oney",
+      "▁Mon ey",
+      "▁Mo ney",
+      "▁ Money",
+      "nAb out",
+      "n About",
+      "的 教育",
+      "月二十 一日",
+      "上半 年",
+      "上 半年",
+      "nCan adian",
+      "n Canadian",
+      "▁global ly",
+      "▁glob ally",
+      "等 比",
+      "青岛 市",
+      "▁h ung",
+      "▁hun g",
+      "▁hu ng",
+      "▁Ark ansas",
+      "贯彻 落实",
+      "不 下",
+      "▁Que bec",
+      "t M",
+      "区 分",
+      "▁stom ach",
+      "▁sto mach",
+      "▁fr uits",
+      "▁fruit s",
+      "▁log ged",
+      "▁ logged",
+      "圆 锥",
+      "▁S el",
+      "▁Se l",
+      "后面 的",
+      "后 面的",
+      "for ced",
+      "force d",
+      "▁F lash",
+      "▁Fl ash",
+      "▁ Flash",
+      "的 通",
+      "nF ollowing",
+      "nFollow ing",
+      "n Following",
+      "股 市",
+      "▁J enn",
+      "▁Jen n",
+      "▁Je nn",
+      "道路 上",
+      "道 路上",
+      "tt ry",
+      "t try",
+      "▁sc aling",
+      "▁scal ing",
+      "是 大",
+      "存在 的问题",
+      "存在的 问题",
+      "努力 完成劳动任务",
+      "comm unity",
+      "commun ity",
+      "▁M agic",
+      "▁Mag ic",
+      "▁Ma gic",
+      "▁ Magic",
+      "with out",
+      "前 列",
+      "▁C hat",
+      "▁Ch at",
+      "▁Cha t",
+      "▁ Chat",
+      "▁co ins",
+      "▁coin s",
+      "▁ coins",
+      "春 天",
+      "▁f oo",
+      "▁fo o",
+      "▁ foo",
+      "ne m",
+      "n em",
+      "ен и",
+      "е ни",
+      "for ge",
+      "f orge",
+      "nA wards",
+      "nAw ards",
+      "nAward s",
+      "▁mer ch",
+      "▁merc h",
+      "明 月",
+      "必须 要",
+      "pers istence",
+      "依法 由审判员",
+      "依法由 审判员",
+      "路 面",
+      "▁b ucket",
+      "▁buck et",
+      "▁ bucket",
+      "▁D EFAULT",
+      "▁ DEFAULT",
+      "▁narr ative",
+      "发 了",
+      "希 腊",
+      "疲 劳",
+      "观 测",
+      "买 房",
+      "师 资",
+      "▁V erm",
+      "▁Ver m",
+      "▁Ve rm",
+      "oc rine",
+      "ocr ine",
+      "▁ag enda",
+      "▁age nda",
+      "Anal ysis",
+      "▁anc est",
+      "▁y oga",
+      "▁yog a",
+      "▁yo ga",
+      "Am ount",
+      "A mount",
+      "和 服务",
+      "fl oor",
+      "flo or",
+      "f loor",
+      "事实 和",
+      "next ern",
+      "nex tern",
+      "n extern",
+      "▁St ruct",
+      "▁Str uct",
+      "▁ Struct",
+      "▁M ilitary",
+      "▁Mil itary",
+      "物 体",
+      "▁w heels",
+      "▁whe els",
+      "▁wheel s",
+      "▁help er",
+      "▁hel per",
+      "▁ helper",
+      "名 人",
+      "▁sup porters",
+      "▁support ers",
+      "▁supporter s",
+      "err anean",
+      "满 满",
+      "scrib er",
+      "scri ber",
+      "scribe r",
+      "scr iber",
+      "储 存",
+      "▁gather ing",
+      "桌 面",
+      "CO M",
+      "C OM",
+      "fa ther",
+      "fat her",
+      "f ather",
+      "草 原",
+      "de cor",
+      "dec or",
+      "▁L L",
+      "▁ LL",
+      "Sp eed",
+      "Spe ed",
+      "S peed",
+      "▁C lear",
+      "▁Cl ear",
+      "▁Cle ar",
+      "▁ Clear",
+      "F D",
+      "吸 毒",
+      "er ald",
+      "era ld",
+      "eral d",
+      "残疾 人",
+      "▁N umer",
+      "▁Num er",
+      "▁Nu mer",
+      "▁ Numer",
+      ">, \\",
+      "> ,\\",
+      "▁some body",
+      "评 选",
+      "▁P ick",
+      "▁Pi ck",
+      "▁Pic k",
+      "▁ Pick",
+      "U A",
+      "仅 供",
+      "ight ers",
+      "igh ters",
+      "ighter s",
+      "\"\\ '",
+      "\" \\'",
+      "子 上",
+      "白 天",
+      "一 时",
+      "▁A udio",
+      "▁Aud io",
+      "▁Audi o",
+      "▁ Audio",
+      "G a",
+      "▁hand les",
+      "▁handle s",
+      "ion ic",
+      "io nic",
+      "ioni c",
+      "i onic",
+      "tu ple",
+      "t uple",
+      "缩 短",
+      "ol ang",
+      "ola ng",
+      "olan g",
+      "o lang",
+      "▁rece ives",
+      "▁receive s",
+      "以 其",
+      "福 州",
+      "▁f lood",
+      "▁fl ood",
+      "▁flo od",
+      "Res olver",
+      "Re solver",
+      "Resolve r",
+      "▁Aust ria",
+      "勇 敢",
+      "to m",
+      "t om",
+      "▁celebr ated",
+      "▁celebrate d",
+      "月二十 四日",
+      "ON G",
+      "O NG",
+      "单 价",
+      "号刑事 裁定",
+      "oz illa",
+      "o zilla",
+      "▁M ind",
+      "▁Min d",
+      "▁Mi nd",
+      "▁ Mind",
+      "▁tem ple",
+      "▁temp le",
+      "内 在",
+      "中 毒",
+      "▁mut ual",
+      "▁mu tual",
+      "▁met als",
+      "▁metal s",
+      "▁meta ls",
+      "▁nanop articles",
+      "▁nano particles",
+      "▁nanoparticle s",
+      "▁Exper ience",
+      "▁config ured",
+      "▁configure d",
+      "先 锋",
+      "教 研",
+      "otyp ic",
+      "▁sec rets",
+      "▁secret s",
+      "▁ secrets",
+      "就 读",
+      "ex ceptions",
+      "except ions",
+      "exception s",
+      "▁V ent",
+      "▁Ven t",
+      "▁Ve nt",
+      "Ab out",
+      "内 地",
+      "he rent",
+      "her ent",
+      "here nt",
+      "h erent",
+      "▁coast al",
+      "膨 胀",
+      "of fee",
+      "off ee",
+      "o ffee",
+      "en cer",
+      "ence r",
+      "enc er",
+      "to Equal",
+      "▁s keletal",
+      "▁ske letal",
+      "户 型",
+      "▁# {",
+      "▁ #{",
+      "ol ia",
+      "oli a",
+      "o lia",
+      "q a",
+      "▁leuk emia",
+      "▁P od",
+      "▁Po d",
+      "▁ Pod",
+      "▁B art",
+      "▁Bar t",
+      "▁Ba rt",
+      "ti er",
+      "tie r",
+      "t ier",
+      "▁l ean",
+      "▁le an",
+      "▁ lean",
+      "enc il",
+      "▁M emory",
+      "▁Mem ory",
+      "▁ Memory",
+      "▁Trans form",
+      "▁ Transform",
+      "我 对",
+      "red ux",
+      "redu x",
+      "真 空",
+      "T L",
+      "▁d ur",
+      "▁du r",
+      "▁Att orney",
+      "感 的",
+      "Fl ag",
+      "F lag",
+      "▁sup plier",
+      "▁suppl ier",
+      "十一 条",
+      "十 一条",
+      "ye r",
+      "y er",
+      "> }",
+      "与 此",
+      "▁custom ized",
+      "▁cust omized",
+      "▁customize d",
+      "▁发 行",
+      "▁ 发行",
+      "▁gu ided",
+      "▁guid ed",
+      "▁guide d",
+      "▁gui ded",
+      "▁ guided",
+      "发 热",
+      "月 生",
+      "GH z",
+      "G Hz",
+      "▁v i",
+      "▁ vi",
+      "},\\ \\",
+      "}, \\\\",
+      "} ,\\\\",
+      "私 服",
+      "▁succeed ed",
+      "乐 趣",
+      "▁Portug al",
+      "美国 的",
+      "美 国的",
+      "▁lim iting",
+      "▁limit ing",
+      "导 弹",
+      "([ '",
+      "( ['",
+      "co v",
+      "c ov",
+      "nM ark",
+      "nMar k",
+      "nMa rk",
+      "n Mark",
+      "ercul osis",
+      "总 额",
+      "ph er",
+      "p her",
+      "os copy",
+      "osc opy",
+      "o scopy",
+      "ven dor",
+      "v endor",
+      "标 识",
+      "这 份",
+      "▁p ose",
+      "▁pos e",
+      "▁po se",
+      "▁ pose",
+      "左 边",
+      "氨 基",
+      "▁t rivial",
+      "▁triv ial",
+      "▁ trivial",
+      "▁Other wise",
+      "▁susp ension",
+      "ct l",
+      "c tl",
+      "▁liter ary",
+      "ni ck",
+      "nic k",
+      "n ick",
+      "我 只",
+      "▁ind ividually",
+      "▁individual ly",
+      "li ver",
+      "live r",
+      "l iver",
+      "Sequ ence",
+      "S equence",
+      "▁Law rence",
+      "ac eful",
+      "ace ful",
+      "栏 目",
+      "nPl ace",
+      "n Place",
+      "抱 着",
+      "ua rio",
+      "u ario",
+      "▁V S",
+      "▁ VS",
+      "ni m",
+      "n im",
+      "▁o bl",
+      "▁ob l",
+      "▁ obl",
+      "东 部",
+      "十九 日",
+      "十 九日",
+      "UR CE",
+      "姆 斯",
+      "▁gener a",
+      "▁gen era",
+      "▁gene ra",
+      "nM in",
+      "nMi n",
+      "n Min",
+      "嘉 宾",
+      "or ters",
+      "ort ers",
+      "orter s",
+      "orte rs",
+      "pic ture",
+      "p icture",
+      "tS tring",
+      "t String",
+      "总 会",
+      "正确 的是",
+      "正确的 是",
+      "▁bul let",
+      "▁bull et",
+      "▁ bullet",
+      "ch rom",
+      "chr om",
+      "▁m eter",
+      "▁me ter",
+      "▁met er",
+      "▁ meter",
+      "▁personal ized",
+      "综 艺",
+      "hi ll",
+      "h ill",
+      "is an",
+      "isa n",
+      "i san",
+      "s u",
+      "潜 在",
+      "▁` ${",
+      "▁`$ {",
+      "▁ `${",
+      "护 肤",
+      "▁belief s",
+      "▁belie fs",
+      "▁d ementia",
+      "adv ant",
+      "▁sol ely",
+      "▁sole ly",
+      "▁sur ely",
+      "▁sure ly",
+      "她 说",
+      "Fl ags",
+      "Flag s",
+      "公 顷",
+      "▁st adium",
+      "’ ,",
+      "▁m art",
+      "▁mar t",
+      "▁ma rt",
+      "▁ mart",
+      "re peat",
+      "rep eat",
+      "开 奖",
+      "▁C E",
+      "▁ CE",
+      "AU TH",
+      "A UTH",
+      "Input Stream",
+      "J an",
+      "he ne",
+      "hen e",
+      "h ene",
+      "选择 的依据",
+      "选择的 依据",
+      "精 细",
+      "comp ute",
+      "comput e",
+      "▁practition ers",
+      "▁practitioner s",
+      "▁s uck",
+      "▁su ck",
+      "▁suc k",
+      "▁l acking",
+      "▁lack ing",
+      "▁lac king",
+      "我 妈",
+      "nSt ud",
+      "n Stud",
+      "ra ig",
+      "WA R",
+      "W AR",
+      "酱 油",
+      "▁N P",
+      "▁ NP",
+      "augh ters",
+      "aughter s",
+      "aught ers",
+      "▁pr inter",
+      "▁print er",
+      "▁pri nter",
+      "▁ printer",
+      "▁con solid",
+      "▁cons olid",
+      "物 料",
+      "▁wr apped",
+      "▁wrap ped",
+      "▁ wrapped",
+      "全 民",
+      "▁P un",
+      "▁Pu n",
+      "▁ch ron",
+      "▁chr on",
+      "▁ chron",
+      "n while",
+      "约 翰",
+      "▁l ocate",
+      "▁loc ate",
+      "et ically",
+      "etic ally",
+      "e tically",
+      "的 花",
+      "Bu s",
+      "B us",
+      "就是 在",
+      "就 是在",
+      "裁定 的执行",
+      "ra wer",
+      "raw er",
+      "▁I nn",
+      "▁In n",
+      "勇 气",
+      "go n",
+      "g on",
+      "ay an",
+      "aya n",
+      "a yan",
+      "hen yl",
+      "初 级",
+      "▁qual ify",
+      "ys et",
+      "yse t",
+      "y set",
+      "▁Def ense",
+      "▁in cl",
+      "▁i ncl",
+      "▁inc l",
+      "è re",
+      "IT H",
+      "I TH",
+      "Json Property",
+      "Per mission",
+      "Perm ission",
+      "P ermission",
+      "fi ve",
+      "f ive",
+      "▁N ord",
+      "▁No rd",
+      "▁Nor d",
+      "▁T esting",
+      "▁Test ing",
+      "▁Tes ting",
+      "▁ Testing",
+      "攻 略",
+      "▁N ich",
+      "▁Nic h",
+      "▁Ni ch",
+      "ex c",
+      "e xc",
+      "▁B ulg",
+      "▁Bul g",
+      "▁Bu lg",
+      "▁Gr ade",
+      "▁Gra de",
+      "▁Grad e",
+      "▁ Grade",
+      "中国农业 银行股份有限公司",
+      "中国农业银行 股份有限公司",
+      "激 活",
+      "▁l ands",
+      "▁land s",
+      "▁lan ds",
+      "▁ lands",
+      "调 味",
+      "事 物",
+      "▁w orship",
+      "▁wor ship",
+      "▁wors hip",
+      "▁M igration",
+      "▁Mig ration",
+      "▁ Migration",
+      "创始 人",
+      "▁v es",
+      "▁ve s",
+      "▁ ves",
+      "ut ures",
+      "uture s",
+      "u tures",
+      "▁transcription al",
+      "▁transcript ional",
+      "▁V A",
+      "▁ VA",
+      "▁e norm",
+      "▁en orm",
+      "▁M ission",
+      "▁Miss ion",
+      "长 得",
+      "et on",
+      "eto n",
+      "e ton",
+      "差 的",
+      "▁batter ies",
+      "▁batt eries",
+      "li bs",
+      "lib s",
+      "研究 了",
+      "▁T NF",
+      "▁TN F",
+      "▁ TNF",
+      "▁Ch anges",
+      "▁Change s",
+      "▁Chang es",
+      "▁Chan ges",
+      "▁ Changes",
+      "▁gr ateful",
+      "▁grat eful",
+      "问题和 选项",
+      "cl one",
+      "她 是",
+      "出 国",
+      "公 正",
+      "▁en roll",
+      "▁mag azines",
+      "▁magazine s",
+      "联 络",
+      "男 友",
+      "mem set",
+      "▁D anish",
+      "▁Dan ish",
+      "▁Dani sh",
+      "▁t an",
+      "▁ta n",
+      "▁ tan",
+      "DO CTYPE",
+      "DOC TYPE",
+      "▁del ivers",
+      "▁deliver s",
+      "▁I mm",
+      "▁Im m",
+      "▁ Imm",
+      "nCl ub",
+      "n Club",
+      "能 认罪悔罪",
+      "之 地",
+      "ad ic",
+      "adi c",
+      "a dic",
+      "全 年",
+      "在 其",
+      "公积 金",
+      "▁jud ges",
+      "▁judge s",
+      "▁judg es",
+      "▁m arijuana",
+      "▁col leges",
+      "▁college s",
+      "▁colle ges",
+      "▁Link ed",
+      "▁ Linked",
+      "▁cont ainers",
+      "▁contain ers",
+      "▁container s",
+      "▁ containers",
+      "▁not eb",
+      "▁note b",
+      "in ae",
+      "ina e",
+      "▁Mc G",
+      "iv als",
+      "ival s",
+      "iva ls",
+      "i vals",
+      "与其 他",
+      "与 其他",
+      "Des criptor",
+      "不 怕",
+      "还 清",
+      "▁ret inal",
+      "▁retin al",
+      "▁retina l",
+      "情 的",
+      "数字 化",
+      "▁assess ments",
+      "▁assessment s",
+      "▁M os",
+      "▁Mo s",
+      "en cia",
+      "enc ia",
+      "e ncia",
+      "谨 慎",
+      "Aut ow",
+      "Auto w",
+      "▁Mon th",
+      "▁Mont h",
+      "▁Mo nth",
+      "▁ Month",
+      "▁l asting",
+      "▁last ing",
+      "▁las ting",
+      "▁ lasting",
+      "▁w are",
+      "▁war e",
+      "▁wa re",
+      "▁ ware",
+      "▁im ped",
+      "▁imp ed",
+      "▁impe d",
+      "▁o rche",
+      "▁or che",
+      "▁orch e",
+      "就 很",
+      "do se",
+      "dos e",
+      "d ose",
+      "低 温",
+      "南 县",
+      "▁sp oken",
+      "▁spoke n",
+      "▁spo ken",
+      "政 务",
+      "Lo wer",
+      "Low er",
+      "L ower",
+      "▁authent ic",
+      "et ooth",
+      "eto oth",
+      "▁Pl anning",
+      "▁Plan ning",
+      "▁K ath",
+      "▁Kat h",
+      "▁Ka th",
+      "▁ann oy",
+      "ru ct",
+      "r uct",
+      "▁Config uration",
+      "▁ Configuration",
+      "uplic ate",
+      "u plicate",
+      "▁cont ributors",
+      "▁contribut ors",
+      "▁contributor s",
+      "▁ contributors",
+      "材料 的",
+      "材 料的",
+      "的 银行存款",
+      "gen eral",
+      "gener al",
+      "gene ral",
+      "五日 内",
+      "五 日内",
+      "积极 分子",
+      "▁C ole",
+      "▁Col e",
+      "▁Co le",
+      "nMed al",
+      "nMe dal",
+      "▁Network s",
+      "▁Net works",
+      "▁exec ut",
+      "违约 金",
+      "详细 的",
+      "▁season al",
+      "▁seas onal",
+      "vers ely",
+      "verse ly",
+      "就好 了",
+      "就 好了",
+      "▁w ings",
+      "▁win gs",
+      "▁wing s",
+      "▁ wings",
+      "▁R oll",
+      "▁Ro ll",
+      "▁Rol l",
+      "▁ Roll",
+      "▁cyt okines",
+      "▁cytok ines",
+      "▁cytokine s",
+      "ro les",
+      "rol es",
+      "role s",
+      "r oles",
+      "Po d",
+      "P od",
+      "iqu id",
+      "iq uid",
+      "▁Tw enty",
+      "▁L abel",
+      "▁La bel",
+      "▁Lab el",
+      "▁ Label",
+      "v d",
+      "co p",
+      "c op",
+      "▁straight forward",
+      "户 口",
+      "剧 本",
+      "完美 的",
+      "完 美的",
+      "▁G on",
+      "▁Go n",
+      "信 念",
+      "ic ht",
+      "ich t",
+      "i cht",
+      "镜 像",
+      "▁B esides",
+      "▁Bes ides",
+      "н е",
+      "▁per forms",
+      "▁perform s",
+      "▁perf orms",
+      "▁perfor ms",
+      "就不 会",
+      "就 不会",
+      "text bf",
+      "▁Ch ampions",
+      "▁Champion s",
+      "酌 情",
+      "▁A F",
+      "▁ AF",
+      "▁t ort",
+      "▁to rt",
+      "▁tor t",
+      "vers ions",
+      "version s",
+      "▁pl ugins",
+      "▁plugin s",
+      "▁plug ins",
+      "▁ plugins",
+      "fer red",
+      "▁gen re",
+      "▁ge nre",
+      "▁ genre",
+      "身 心",
+      "▁app earing",
+      "▁appe aring",
+      "▁appear ing",
+      "R Y",
+      "▁t oys",
+      "▁to ys",
+      "▁toy s",
+      "▁( [",
+      "▁ ([",
+      "oplas mic",
+      "oplasm ic",
+      "$ _",
+      "▁C in",
+      "▁Ci n",
+      "路 人",
+      "电 网",
+      "▁type of",
+      "▁ typeof",
+      "▁d ad",
+      "▁da d",
+      "▁ dad",
+      "▁substant ially",
+      "▁substantial ly",
+      "交 互",
+      "▁P ages",
+      "▁Page s",
+      "▁Pa ges",
+      "▁Pag es",
+      "▁ Pages",
+      "江 县",
+      "▁de struction",
+      "▁destr uction",
+      "ss l",
+      "s sl",
+      "白 酒",
+      "ja x",
+      "j ax",
+      "re pos",
+      "rep os",
+      "repo s",
+      "▁ 工作",
+      "欲 望",
+      "影响 力",
+      "中华 民族",
+      "情 报",
+      "为 其",
+      "被执行人 王",
+      "ser vers",
+      "serv ers",
+      "server s",
+      "serve rs",
+      "▁Post s",
+      "▁Po sts",
+      "▁Pos ts",
+      "▁ Posts",
+      "的一 款",
+      "的 一款",
+      "世界 杯",
+      "▁Function s",
+      "▁ Functions",
+      "之 上",
+      "▁th reads",
+      "▁thread s",
+      "▁ threads",
+      "▁enh ancing",
+      "Ci ty",
+      "C ity",
+      "在 你",
+      "罢 了",
+      "登记 表",
+      "▁the ater",
+      "▁theat er",
+      "▁point ing",
+      "手 的",
+      "安 区",
+      "第二百 一",
+      "Sh ell",
+      "She ll",
+      "S hell",
+      "一 同",
+      "可以 说是",
+      "可以说 是",
+      "ar in",
+      "ari n",
+      "a rin",
+      "上市 公司",
+      "nW orld",
+      "n World",
+      "斯 拉",
+      "安全 性",
+      "▁im plicit",
+      "▁impl icit",
+      "▁ implicit",
+      "▁j ail",
+      "▁ja il",
+      "地 处",
+      "le en",
+      "lee n",
+      "l een",
+      "一 眼",
+      "加 密",
+      "▁l ith",
+      "▁li th",
+      "▁lit h",
+      "Di d",
+      "D id",
+      "AL TH",
+      "ALT H",
+      "平 稳",
+      "▁N am",
+      "▁Na m",
+      "▁encour aging",
+      "▁light weight",
+      "O ct",
+      "等多 种",
+      "等 多种",
+      "am mad",
+      "amm ad",
+      "amma d",
+      "▁organ izational",
+      "▁organization al",
+      "nOff icial",
+      "nR ep",
+      "nRe p",
+      "n Rep",
+      "▁altern ate",
+      "▁alt ernate",
+      "加 重",
+      "ro rs",
+      "ror s",
+      "r ors",
+      "nS ources",
+      "nSource s",
+      "n Sources",
+      "▁U nc",
+      "▁Un c",
+      "S n",
+      "抖 音",
+      "▁F ab",
+      "▁Fa b",
+      "ge l",
+      "g el",
+      "As set",
+      "Ass et",
+      "沙 发",
+      "▁[ ],",
+      "▁[] ,",
+      "▁ [],",
+      "▁W olf",
+      "▁Wol f",
+      "▁Wo lf",
+      "▁in fusion",
+      "▁inf usion",
+      "▁hab its",
+      "▁ha bits",
+      "▁habit s",
+      "向本院提出 诉讼请求",
+      "▁F ont",
+      "▁Fo nt",
+      "▁ Font",
+      "购 房",
+      "▁dram atically",
+      "▁drama tically",
+      "▁dramatic ally",
+      "▁under graduate",
+      "▁undergrad uate",
+      "历史 上",
+      "历 史上",
+      "ass ador",
+      "assa dor",
+      "开展 了",
+      "iqu it",
+      "iq uit",
+      "i quit",
+      "极 大",
+      "学 期",
+      "业 内",
+      "con cat",
+      "co ncat",
+      "conc at",
+      "▁p rest",
+      "▁pre st",
+      "▁pr est",
+      "▁pres t",
+      "画 家",
+      "任 职",
+      "It erator",
+      "Iter ator",
+      "行 程",
+      "Pa ir",
+      "P air",
+      "User name",
+      "即可 得出",
+      "▁qu it",
+      "▁q uit",
+      "▁qui t",
+      "▁ quit",
+      "法 学",
+      "▁ch ampions",
+      "▁champion s",
+      "ou sing",
+      "ous ing",
+      "o using",
+      "后 者",
+      "转 向",
+      "▁G re",
+      "▁Gr e",
+      "im plies",
+      "impl ies",
+      "imp lies",
+      "的 活动",
+      "; }\\",
+      "Custom er",
+      "string ify",
+      "▁invest igator",
+      "▁investig ator",
+      "▁m igrations",
+      "▁migr ations",
+      "▁migration s",
+      "▁ migrations",
+      "▁m oder",
+      "▁mod er",
+      "▁mode r",
+      "▁mo der",
+      "▁p oster",
+      "▁pos ter",
+      "▁post er",
+      "▁po ster",
+      "▁ poster",
+      "▁ 算法",
+      "▁B L",
+      "▁ BL",
+      "▁s ac",
+      "▁sa c",
+      "▁H ero",
+      "▁He ro",
+      "▁Her o",
+      "▁ Hero",
+      "▁hepat itis",
+      "du cer",
+      "duc er",
+      "d ucer",
+      "Ba l",
+      "B al",
+      "微 软",
+      "St atic",
+      "Stat ic",
+      "▁a ortic",
+      "▁ind uces",
+      "▁indu ces",
+      "▁induce s",
+      "▁Or iginal",
+      "▁Origin al",
+      "▁ Original",
+      "▁intr aven",
+      "▁intra ven",
+      "▁intrav en",
+      "▁r abbit",
+      "▁rab bit",
+      "▁ rabbit",
+      "▁S ens",
+      "▁Se ns",
+      "▁Sen s",
+      "基 准",
+      "▁re per",
+      "▁r eper",
+      "▁rep er",
+      "▁repe r",
+      "▁down stream",
+      "内 角",
+      "的 利息",
+      "▁fr iction",
+      "▁fri ction",
+      "的 日子",
+      "洛 阳",
+      "水 稻",
+      "为 中心",
+      "西 湖",
+      "在 职",
+      "神 话",
+      "意思 是",
+      "▁D elta",
+      "▁Del ta",
+      "▁ Delta",
+      "oc occ",
+      "oco cc",
+      "ococ c",
+      "Sw itch",
+      "S witch",
+      "nF inally",
+      "nFinal ly",
+      "nFin ally",
+      "og ly",
+      "o gly",
+      "') ),\\",
+      "')) ,\\",
+      "' )),\\",
+      "▁Techn ologies",
+      "大 桥",
+      "▁d c",
+      "▁ dc",
+      "▁b aking",
+      "▁ba king",
+      "▁bak ing",
+      "芝 麻",
+      "ninter face",
+      "n interface",
+      "▁intellig ent",
+      "K S",
+      "▁correspon ds",
+      "▁correspond s",
+      "bo th",
+      "bot h",
+      "b oth",
+      "▁im migration",
+      "▁imm igration",
+      "▁immigr ation",
+      "up iter",
+      "Y Y",
+      "se nder",
+      "send er",
+      "sen der",
+      "s ender",
+      "▁D ance",
+      "▁Dan ce",
+      "▁Da nce",
+      "海 关",
+      "bo rough",
+      "bor ough",
+      "boro ugh",
+      "b orough",
+      "长 城",
+      "▁( ;",
+      "― ―",
+      "▁G ordon",
+      "▁Gor don",
+      "▁c url",
+      "▁cur l",
+      "▁cu rl",
+      "▁ curl",
+      "Auto wired",
+      "Autow ired",
+      "▁Mc K",
+      "Ca pt",
+      "Cap t",
+      "C apt",
+      "名 义",
+      "vis it",
+      "vi sit",
+      "遗 址",
+      "太阳 能",
+      "一般 的",
+      "一 般的",
+      "庆 祝",
+      "▁top ological",
+      "▁fat igue",
+      "器 械",
+      "ish ment",
+      "▁bro ader",
+      "▁broad er",
+      "nDes pite",
+      "n Despite",
+      "▁Conn ection",
+      "▁Connect ion",
+      "▁ Connection",
+      "pay ment",
+      "▁pain ter",
+      "▁paint er",
+      "▁pa inter",
+      "一 颗",
+      "再 用",
+      "▁attract ions",
+      "▁attr actions",
+      "▁attraction s",
+      "参 演",
+      "列 入",
+      "▁p ie",
+      "▁pi e",
+      "▁ pie",
+      "ent ries",
+      "Time r",
+      "Tim er",
+      "Ti mer",
+      "T imer",
+      "▁N othing",
+      "▁No thing",
+      "▁Not hing",
+      "▁ Nothing",
+      "nI N",
+      "n IN",
+      "▁ch airs",
+      "▁chair s",
+      "▁cha irs",
+      "▁un lock",
+      "▁s ail",
+      "▁sa il",
+      "▁char ging",
+      "▁charg ing",
+      "pend icular",
+      "▁COPY RIGHT",
+      "Tag s",
+      "Ta gs",
+      "T ags",
+      "著名 的",
+      "著 名的",
+      "▁inhab it",
+      "▁inh abit",
+      "ne n",
+      "n en",
+      "▁M ess",
+      "▁Me ss",
+      "▁Mes s",
+      "▁ Mess",
+      "asc ul",
+      "nInd ian",
+      "▁H um",
+      "▁Hu m",
+      "要 把",
+      "▁c ited",
+      "▁cit ed",
+      "▁ci ted",
+      "▁cite d",
+      "万 平方米",
+      "cel ona",
+      "▁e uro",
+      "▁eu ro",
+      "▁p iv",
+      "▁pi v",
+      "补 助",
+      "▁l ocom",
+      "▁loc om",
+      "▁lo com",
+      "t protected",
+      "切 除",
+      "阵 容",
+      "毕业 院校",
+      "▁G rid",
+      "▁Gr id",
+      "▁ Grid",
+      "向本院 申请恢复执行",
+      "向本院申请 恢复执行",
+      "阳 台",
+      "▁w alks",
+      "▁walk s",
+      "▁wal ks",
+      "不是 很",
+      "不 是很",
+      "▁sub tle",
+      "▁subt le",
+      "nA tt",
+      "nAt t",
+      "n Att",
+      "▁ep ile",
+      "▁sub jected",
+      "▁subject ed",
+      "Sub ject",
+      "Su bject",
+      "ather ine",
+      "athe rine",
+      "自 首",
+      "▁sc iences",
+      "▁science s",
+      "▁sci ences",
+      "▁ا ل",
+      "▁ ال",
+      "▁第 三",
+      "▁ 第三",
+      "折抵 刑期",
+      "▁d uties",
+      "▁du ties",
+      "▁Con clusion",
+      "旗 舰",
+      "错 过",
+      "顶 级",
+      "ia tes",
+      "iate s",
+      "iat es",
+      "i ates",
+      "▁U rban",
+      "▁Ur ban",
+      "す る",
+      "场 合",
+      "▁ne wer",
+      "▁new er",
+      "没 收",
+      "选 购",
+      "pos er",
+      "pose r",
+      "po ser",
+      "p oser",
+      "▁F inn",
+      "▁Fin n",
+      "▁Fi nn",
+      "▁ ≥",
+      "▁$ _",
+      "▁ $_",
+      "▁f av",
+      "▁fa v",
+      "▁ fav",
+      "审 定",
+      "矫 正",
+      "` ](",
+      "te le",
+      "tel e",
+      "t ele",
+      "▁h itting",
+      "▁hit ting",
+      "这 边",
+      "▁col oring",
+      "▁color ing",
+      "▁dis ag",
+      "Ch art",
+      "Char t",
+      "C hart",
+      "▁mult ic",
+      "▁multi c",
+      "▁mul tic",
+      "国家 和",
+      "国 家和",
+      "▁in duct",
+      "▁ind uct",
+      "▁indu ct",
+      "圆 满",
+      "建筑 工程",
+      "▁C hel",
+      "▁Ch el",
+      "▁Che l",
+      "的 上",
+      "▁recommend ation",
+      "▁sc heduling",
+      "▁schedul ing",
+      "▁compl aint",
+      "▁complain t",
+      "▁B ab",
+      "▁Ba b",
+      "\\' \\'\\'",
+      "\\'\\' \\'",
+      "\\'\\'\\ '",
+      "▁fib rosis",
+      "▁fibr osis",
+      "▁fibro sis",
+      "▁ ♪\\",
+      "PR ESS",
+      "PRE SS",
+      "P RESS",
+      "▁tro ops",
+      "入 学",
+      "初 心",
+      "un icode",
+      "uni code",
+      "失 望",
+      "▁e arnings",
+      "▁ear nings",
+      "▁earn ings",
+      "▁earning s",
+      "em en",
+      "eme n",
+      "e men",
+      "▁Vis it",
+      "▁Vi sit",
+      "▁ Visit",
+      "▁cent uries",
+      "▁exhib its",
+      "▁exhibit s",
+      "修 正",
+      "▁undert aken",
+      "▁undertake n",
+      "小 儿",
+      "问题 是",
+      "in vas",
+      "inv as",
+      "香 味",
+      "(\\' \\\\",
+      "(\\ '\\\\",
+      "( \\'\\\\",
+      "触 摸",
+      "gu ard",
+      "女 友",
+      "▁integr ating",
+      "变 动",
+      "fit s",
+      "fi ts",
+      "f its",
+      "▁un cover",
+      "▁unc over",
+      "米 的",
+      "游戏 中",
+      "▁constitu ency",
+      "▁m ast",
+      "▁ma st",
+      "▁mas t",
+      "▁H ug",
+      "▁Hu g",
+      "▁tr adem",
+      "▁trad em",
+      "▁tra dem",
+      "▁trade m",
+      "Ill uminate",
+      "Se g",
+      "S eg",
+      "▁ext ensively",
+      "▁extensive ly",
+      "▁arth ritis",
+      "▁ arthritis",
+      "▁M anag",
+      "▁Man ag",
+      "▁Ma nag",
+      "▁ Manag",
+      "多 名",
+      "le ading",
+      "lead ing",
+      "lea ding",
+      "四 季",
+      "He lp",
+      "Hel p",
+      "H elp",
+      "▁circ les",
+      "▁circle s",
+      "▁cir cles",
+      "mon ths",
+      "month s",
+      "mont hs",
+      "▁O d",
+      "▁S N",
+      "▁ SN",
+      "▁man uscript",
+      "应 力",
+      "nFil m",
+      "撤诉 处理",
+      "暂 停",
+      "负 荷",
+      "送 到",
+      "}^ \\\\",
+      "} ^\\\\",
+      "必 备",
+      "夫妻 关系",
+      "▁ 二",
+      "▁S ie",
+      "▁Si e",
+      "来源 于",
+      "来 源于",
+      "驾驶 机动车",
+      "Ph ys",
+      "ocl onal",
+      "南 路",
+      "食 堂",
+      "end ment",
+      "乐 观",
+      "极 为",
+      "▁tail ored",
+      "▁tailor ed",
+      "同 等",
+      "▁T ro",
+      "▁Tr o",
+      "▁al location",
+      "▁all ocation",
+      "▁alloc ation",
+      "ys ter",
+      "yst er",
+      "y ster",
+      "乙 醇",
+      "▁agricult ure",
+      "▁condition al",
+      "▁cond itional",
+      "▁ conditional",
+      "Re view",
+      "Rev iew",
+      "▁su ited",
+      "▁suit ed",
+      "▁suite d",
+      "nT op",
+      "nTo p",
+      "n Top",
+      "ur ring",
+      "▁in aug",
+      "可 谓",
+      "co des",
+      "code s",
+      "cod es",
+      "c odes",
+      "ug ged",
+      "ugg ed",
+      "up id",
+      "u pid",
+      "先行 羁押",
+      "分 管",
+      "bo ost",
+      "ic its",
+      "ici ts",
+      "icit s",
+      "un ches",
+      "unc hes",
+      "unch es",
+      "▁ 华",
+      "清 新",
+      "▁L ore",
+      "▁Lo re",
+      "▁Lor e",
+      "▁sp eeds",
+      "▁spe eds",
+      "▁speed s",
+      "▁Ph one",
+      "▁ Phone",
+      "▁thorough ly",
+      "uff ix",
+      "uf fix",
+      "安全 的",
+      "▁br oker",
+      "▁bro ker",
+      "▁broke r",
+      "▁ broker",
+      "人 是",
+      "▁dep ict",
+      "na tal",
+      "nat al",
+      "n atal",
+      "的数 量",
+      "的 数量",
+      "▁Con vert",
+      "▁Conv ert",
+      "▁ Convert",
+      "运 转",
+      "我 市",
+      "; \"",
+      "所确定的 义务",
+      "所 确定的义务",
+      "script ions",
+      "scri ptions",
+      "scription s",
+      "s criptions",
+      "复议期间不停止 裁定的执行",
+      "Kin d",
+      "Ki nd",
+      "K ind",
+      "指派 检察员",
+      "tr avis",
+      "tra vis",
+      "▁rect angle",
+      "▁ rectangle",
+      "▁l iability",
+      "▁li ability",
+      "W M",
+      "ACK GROUND",
+      "▁IMP LIED",
+      "灯 光",
+      "▁S olar",
+      "▁So lar",
+      "▁Sol ar",
+      "发生 在",
+      "▁wild life",
+      "Car t",
+      "Ca rt",
+      "C art",
+      "oper ations",
+      "operation s",
+      "og lobin",
+      "oglob in",
+      "卧 室",
+      "▁elect ronics",
+      "▁electron ics",
+      "▁electronic s",
+      "情 侣",
+      "ro bot",
+      "rob ot",
+      "▁ar row",
+      "▁arr ow",
+      "▁ arrow",
+      "mach ine",
+      "m achine",
+      "艰 难",
+      "▁bl adder",
+      "阐述 了",
+      "阐 述了",
+      "▁cross ing",
+      "▁D al",
+      "▁Da l",
+      "▁Sy ria",
+      "▁Syr ia",
+      "av atar",
+      "ava tar",
+      "du cible",
+      "duc ible",
+      "▁g aps",
+      "▁gap s",
+      "▁ga ps",
+      "▁C abin",
+      "▁Ca bin",
+      "▁Cab in",
+      "▁Cap tain",
+      "▁Capt ain",
+      "▁lock ed",
+      "▁ locked",
+      "博 览",
+      "三 是",
+      "奖 金",
+      "▁mass es",
+      "▁mas ses",
+      "▁con current",
+      "▁co ncurrent",
+      "▁conc urrent",
+      "▁ concurrent",
+      "价值 的",
+      "价 值的",
+      "日 语",
+      "首 歌",
+      "▁z ones",
+      "▁zone s",
+      "▁neuro logical",
+      "▁neur ological",
+      "▁neurologic al",
+      "im on",
+      "imo n",
+      "i mon",
+      "▁an imated",
+      "▁anim ated",
+      "▁A pi",
+      "▁Ap i",
+      "▁ Api",
+      "▁un ders",
+      "▁under s",
+      "▁und ers",
+      "▁ unders",
+      "▁o d",
+      "▁ od",
+      "▁r ocks",
+      "▁ro cks",
+      "▁rock s",
+      "▁bab ies",
+      "▁ba bies",
+      "{\\\\ ,\\\\",
+      "{ \\\\,\\\\",
+      "近 似",
+      "out ine",
+      "▁adv ise",
+      "▁advis e",
+      "判决执行 之日起计算",
+      "▁calcul ating",
+      "▁calc ulating",
+      "▁use State",
+      "en ta",
+      "ent a",
+      "pr iate",
+      "pri ate",
+      "p riate",
+      "pat riate",
+      "省 委",
+      "ge ometry",
+      "geo metry",
+      "geom etry",
+      "三 级",
+      "▁di gest",
+      "▁dig est",
+      "▁ digest",
+      "er ial",
+      "eria l",
+      "eri al",
+      "e rial",
+      "State ment",
+      "Stat ement",
+      "联 邦",
+      "风 云",
+      "> $",
+      "一 分",
+      "an try",
+      "ant ry",
+      "a ntry",
+      "▁bur ning",
+      "▁burn ing",
+      "平 板",
+      "at ivity",
+      "ativ ity",
+      "西 路",
+      "▁Al aska",
+      "出 自",
+      "ah o",
+      "a ho",
+      "or ted",
+      "ort ed",
+      "orte d",
+      "▁w orn",
+      "▁wor n",
+      "▁wo rn",
+      "IN ST",
+      "INS T",
+      "▁rul ing",
+      "▁ru ling",
+      "Ra w",
+      "R aw",
+      "y i",
+      "永 久",
+      "▁sp ell",
+      "▁spe ll",
+      "▁w arnings",
+      "▁war nings",
+      "▁warning s",
+      "▁warn ings",
+      "▁ warnings",
+      "TR ACT",
+      "TRA CT",
+      "今年 的",
+      "今 年的",
+      "反 弹",
+      "情 人",
+      "参加 诉讼",
+      "Don ald",
+      "D onald",
+      "潍 坊",
+      "住 重庆市",
+      "▁compl aints",
+      "▁complaint s",
+      "▁complain ts",
+      "▁re construct",
+      "▁recon struct",
+      "▁reconst ruct",
+      "这不 是",
+      "这 不是",
+      "ry ing",
+      "r ying",
+      "▁F ine",
+      "▁Fin e",
+      "▁Fi ne",
+      "大学 的",
+      "大 学的",
+      "▁mod ular",
+      "fr ak",
+      "fra k",
+      "f rak",
+      "d B",
+      "▁Act ivity",
+      "▁Activ ity",
+      "▁ Activity",
+      "刚 开始",
+      "生 前",
+      "im ated",
+      "imate d",
+      "ima ted",
+      "▁Ch amber",
+      "▁Cha mber",
+      "▁Cham ber",
+      "ik o",
+      "i ko",
+      "if icates",
+      "ific ates",
+      "ificate s",
+      "一 支",
+      "ko v",
+      "k ov",
+      "or ac",
+      "ora c",
+      "o rac",
+      "▁genu ine",
+      "落 后",
+      "ch ing",
+      "chi ng",
+      "chin g",
+      "c hing",
+      "抢 劫",
+      "▁ 参考文献",
+      "不 妨",
+      "▁un e",
+      "▁u ne",
+      "▁ une",
+      "ear ly",
+      "e arly",
+      "▁An na",
+      "▁Ann a",
+      "Id entity",
+      "Ident ity",
+      "MR I",
+      "M RI",
+      "Count er",
+      "Co unter",
+      "Cou nter",
+      "C ounter",
+      "轻 易",
+      "▁L ate",
+      "▁La te",
+      "▁Lat e",
+      "▁N eg",
+      "▁Ne g",
+      "▁ Neg",
+      "▁D ean",
+      "▁De an",
+      "单 项",
+      "nS ing",
+      "n Sing",
+      "▁An imal",
+      "▁ Animal",
+      "▁e vil",
+      "▁ev il",
+      "▁ evil",
+      "▁gr ades",
+      "▁grad es",
+      "▁grade s",
+      "▁gra des",
+      "▁ grades",
+      "qu ick",
+      "▁G ulf",
+      "▁Gu lf",
+      "▁Gul f",
+      "在 法定",
+      "▁Up per",
+      "▁ Upper",
+      "nA pp",
+      "nAp p",
+      "n App",
+      "in cre",
+      "inc re",
+      "▁weak ness",
+      "进 来",
+      "找到 了",
+      "找 到了",
+      "▁cricket ers",
+      "▁cricketer s",
+      "即 自",
+      "▁P rior",
+      "▁Pr ior",
+      "▁Pri or",
+      "▁c avity",
+      "▁cav ity",
+      "re tty",
+      "ret ty",
+      "rett y",
+      "变 为",
+      "▁c ycling",
+      "▁cycl ing",
+      "▁cyclin g",
+      "R R",
+      "▁ap ple",
+      "▁app le",
+      "▁appl e",
+      "▁ apple",
+      "素 养",
+      "硬 化",
+      "Ed ge",
+      "▁B eat",
+      "▁Be at",
+      "▁Bea t",
+      "供应 链",
+      "风 吹",
+      "为主 题",
+      "为 主题",
+      "▁car riers",
+      "▁carrier s",
+      "▁de position",
+      "▁dep osition",
+      "▁depos ition",
+      "▁deposit ion",
+      "野 生",
+      "▁dr ops",
+      "▁dro ps",
+      "▁drop s",
+      "▁D M",
+      "▁ DM",
+      "▁h oney",
+      "▁hon ey",
+      "▁ho ney",
+      "▁hone y",
+      "▁exp ense",
+      "fo cus",
+      "f ocus",
+      "ed a",
+      "e da",
+      "Pr ed",
+      "Pre d",
+      "P red",
+      "交通事故 责任",
+      "▁Re ading",
+      "▁Read ing",
+      "▁ Reading",
+      "▁ 为",
+      "web kit",
+      "▁for ests",
+      "▁fore sts",
+      "▁forest s",
+      "▁st opping",
+      "▁stop ping",
+      "▁sto pping",
+      ") `",
+      "gu lp",
+      "g ulp",
+      "我 看",
+      "教 堂",
+      "not ations",
+      "notation s",
+      "首 席",
+      "▁M om",
+      "▁Mo m",
+      "▁cor relations",
+      "▁correl ations",
+      "▁correlation s",
+      "k n",
+      "复 兴",
+      "ge m",
+      "g em",
+      "▁pre vents",
+      "▁pr events",
+      "▁prevent s",
+      "▁prev ents",
+      "▁spons ored",
+      "▁sponsor ed",
+      "▁ sponsored",
+      "lo tte",
+      "lot te",
+      "▁Pu erto",
+      "ex isting",
+      "exist ing",
+      "next Int",
+      "▁hyp oth",
+      "▁hypot h",
+      "相 交",
+      "for all",
+      "性 强",
+      "} ×",
+      "ur d",
+      "u rd",
+      "健康 的",
+      "▁R ing",
+      "▁Ri ng",
+      "▁ Ring",
+      "中 路",
+      "▁pass ive",
+      "Per s",
+      "Pe rs",
+      "P ers",
+      "O O",
+      "▁D ictionary",
+      "▁ Dictionary",
+      "() ))\\",
+      "()) )\\",
+      "())) \\",
+      "( )))\\",
+      "or ect",
+      "ore ct",
+      "o rect",
+      "在开庭审理过程中 亦无异议",
+      "始 建于",
+      "a q",
+      "card s",
+      "car ds",
+      "c ards",
+      "车 主",
+      "ro wing",
+      "row ing",
+      "r owing",
+      "LI ST",
+      "L IST",
+      "▁ext ending",
+      "▁extend ing",
+      "nd o",
+      "n do",
+      "▁extra cts",
+      "▁extract s",
+      "▁extr acts",
+      "▁extrac ts",
+      "K A",
+      "▁respons ive",
+      "▁ responsive",
+      ":\\\\ (",
+      ": \\\\(",
+      "ok a",
+      "o ka",
+      "产品 质量",
+      "ac hel",
+      "ach el",
+      "ache l",
+      "a chel",
+      "▁expl aining",
+      "▁explain ing",
+      "IN TER",
+      "INT ER",
+      "▁co vari",
+      "▁cov ari",
+      "Vi r",
+      "V ir",
+      "疾病 的",
+      "疾 病的",
+      "sc r",
+      "s cr",
+      "▁journal ists",
+      "▁journalist s",
+      "所以 在",
+      "▁De lete",
+      "▁Del ete",
+      "▁ Delete",
+      "▁psych iatric",
+      "▁psychiat ric",
+      "” (",
+      "go w",
+      "g ow",
+      "▁ 地址",
+      "ast rop",
+      "astr op",
+      "astro p",
+      "Pr ivate",
+      "P rivate",
+      "▁E xp",
+      "▁Ex p",
+      "▁ Exp",
+      "fix ed",
+      "f ixed",
+      "▁l ocks",
+      "▁loc ks",
+      "▁lo cks",
+      "▁lock s",
+      "▁ locks",
+      "▁f ever",
+      "▁fe ver",
+      "▁con ject",
+      "▁conj ect",
+      "建筑 物",
+      "切 割",
+      "re ne",
+      "ren e",
+      "r ene",
+      "正 数",
+      "边形 的",
+      "边 形的",
+      "的 义务",
+      "▁J udge",
+      "▁Jud ge",
+      "copy right",
+      "c opyright",
+      "pl aced",
+      "place d",
+      "plac ed",
+      "几 点",
+      "报 价",
+      "is ure",
+      "▁P ad",
+      "▁Pa d",
+      "▁ Pad",
+      "树 脂",
+      "▁android x",
+      "▁D ES",
+      "▁DE S",
+      "▁ DES",
+      "简 体",
+      "胡 萝卜",
+      "起来 的",
+      "起 来的",
+      "Lo ok",
+      "L ook",
+      "▁Y ears",
+      "▁Year s",
+      "▁Ye ars",
+      "nAB STRACT",
+      "n ABSTRACT",
+      "method s",
+      "meth ods",
+      "▁don ors",
+      "▁donor s",
+      "▁an gi",
+      "▁ang i",
+      "▁ angi",
+      "门 诊",
+      "io let",
+      "iol et",
+      "nEx patriate",
+      "Co ord",
+      "归 来",
+      "▁Leg islative",
+      "▁Legisl ative",
+      "▁s hr",
+      "▁sh r",
+      "ie val",
+      "iev al",
+      "i eval",
+      "▁sh ifts",
+      "▁shift s",
+      "▁Pr adesh",
+      "▁L ock",
+      "▁Lo ck",
+      "▁Loc k",
+      "▁ Lock",
+      "做 事",
+      "lo gy",
+      "log y",
+      "l ogy",
+      "▁pre scription",
+      "▁pres cription",
+      "do g",
+      "d og",
+      "▁c d",
+      "▁ cd",
+      "** *\\",
+      "*** \\",
+      "* **\\",
+      "▁B ou",
+      "▁Bo u",
+      "Con vert",
+      "Conv ert",
+      "更 改",
+      "直 至",
+      "▁H ad",
+      "▁Ha d",
+      "新 一",
+      "▁F igure",
+      "▁Fig ure",
+      "▁ Figure",
+      "床 上",
+      "sq u",
+      "s qu",
+      "it arian",
+      "itar ian",
+      "ita rian",
+      "当 当",
+      "贷款 利率",
+      "di git",
+      "dig it",
+      "▁tang ent",
+      "▁tan gent",
+      "▁G ets",
+      "▁Ge ts",
+      "▁Get s",
+      "势 力",
+      "GF R",
+      "G FR",
+      "流 畅",
+      "un ion",
+      "uni on",
+      "▁simult aneous",
+      "up lex",
+      "uple x",
+      "u plex",
+      "ik h",
+      "i kh",
+      "fl utter",
+      "据 说",
+      "li sted",
+      "list ed",
+      "lis ted",
+      "l isted",
+      "去 掉",
+      "nBuild ings",
+      "nBuilding s",
+      "nM ult",
+      "n Mult",
+      "▁architect ural",
+      "ra g",
+      "r ag",
+      "蓝 牙",
+      "▁b ay",
+      "▁ba y",
+      "▁ bay",
+      "Ex per",
+      "Exp er",
+      "筛 选",
+      "▁ab sent",
+      "▁abs ent",
+      "▁Stud ios",
+      "▁Studio s",
+      "▁coron avirus",
+      "说 实话",
+      "▁municip al",
+      "综上 所述",
+      "▁G ran",
+      "▁Gr an",
+      "▁Gra n",
+      "▁t ent",
+      "▁te nt",
+      "▁ten t",
+      "▁Ind eed",
+      "▁unnec essary",
+      "先行羁押 的",
+      "两 大",
+      "液 晶",
+      "▁illustr ated",
+      "▁illust rated",
+      "▁illustrate d",
+      "▁C hart",
+      "▁Ch art",
+      "▁Char t",
+      "▁Cha rt",
+      "▁ Chart",
+      "▁S ony",
+      "▁So ny",
+      "▁Son y",
+      "本院 将",
+      "人民 群众",
+      "im als",
+      "imal s",
+      "ima ls",
+      "ni x",
+      "n ix",
+      "Aud io",
+      "A udio",
+      "▁W indow",
+      "▁Win dow",
+      "▁Wind ow",
+      "▁ Window",
+      "微 生物",
+      "vari ables",
+      "variable s",
+      "N H",
+      "irm ingham",
+      "调 试",
+      "G H",
+      "字 典",
+      "那 天",
+      "▁F ly",
+      "▁Fl y",
+      "▁ Fly",
+      "ri g",
+      "r ig",
+      "▁C ant",
+      "▁Can t",
+      "▁Ca nt",
+      "▁th ym",
+      "▁thy m",
+      "▁s ynaptic",
+      "▁syn aptic",
+      "三 天",
+      "情 怀",
+      "▁in tro",
+      "▁int ro",
+      "▁intr o",
+      "▁ intro",
+      "Un known",
+      "应 在",
+      "温 泉",
+      "支付 利息",
+      "单位 的",
+      "单 位的",
+      "sec ure",
+      "上面 的",
+      "上 面的",
+      "号之 一",
+      "号 之一",
+      "我还 是",
+      "我 还是",
+      ")/ ((",
+      ")/( (",
+      ") /((",
+      "▁gu ides",
+      "▁guid es",
+      "▁guide s",
+      "▁gui des",
+      "▁ guides",
+      "▁sch olars",
+      "▁scholar s",
+      "▁h i",
+      "▁ hi",
+      "错误 的",
+      "▁Mat thew",
+      "▁Matt hew",
+      "▁search ed",
+      "▁sear ched",
+      "nA ustralian",
+      "n Australian",
+      "▁Bangl adesh",
+      "号案 件",
+      "号 案件",
+      "ia m",
+      "i am",
+      "iz a",
+      "i za",
+      "▁Ex cel",
+      "▁Exc el",
+      "▁ Excel",
+      "▁de leg",
+      "▁del eg",
+      "▁leg isl",
+      "na ll",
+      "nal l",
+      "n all",
+      "▁ 苏",
+      "对 了",
+      "▁com patibility",
+      "▁compat ibility",
+      "▁ compatibility",
+      "最 具",
+      "查 获",
+      "▁Dev ice",
+      "▁ Device",
+      "▁T yp",
+      "▁Ty p",
+      "▁ Typ",
+      "假 日",
+      "te nd",
+      "ten d",
+      "t end",
+      "ca ption",
+      "cap tion",
+      "capt ion",
+      "现 原告",
+      "▁% >\\",
+      "▁%> \\",
+      "▁Carib bean",
+      "▁C el",
+      "▁Ce l",
+      "制 备",
+      "▁其 他",
+      "▁ 其他",
+      "十 条",
+      "▁C ab",
+      "▁Ca b",
+      "ier ra",
+      "nSe cond",
+      "nSec ond",
+      "n Second",
+      "process or",
+      "▁in jected",
+      "▁inj ected",
+      "▁inject ed",
+      "▁M edic",
+      "▁Med ic",
+      "▁Me dic",
+      "▁fin ancing",
+      "▁financ ing",
+      "▁ar ithmetic",
+      "P a",
+      "▁trigger ed",
+      "▁mult im",
+      "▁multi m",
+      "▁mul tim",
+      "羁押 一日",
+      "▁U sed",
+      "▁Use d",
+      "▁Us ed",
+      "▁ Used",
+      "ag ine",
+      "agi ne",
+      "agin e",
+      "uate d",
+      "ua ted",
+      "u ated",
+      "分 开",
+      "▁C re",
+      "▁Cr e",
+      "▁ Cre",
+      "▁program mes",
+      "▁programme s",
+      "热 烈",
+      "▁off ensive",
+      "贸易 有限公司",
+      "し て",
+      "产业 发展",
+      "con duct",
+      "cond uct",
+      "▁part ly",
+      "当 场",
+      "的 存在",
+      "头 部",
+      "股 价",
+      "gr e",
+      "g re",
+      "最大 值",
+      "等比 数列",
+      "没 办法",
+      "▁claim ing",
+      "▁t oy",
+      "▁to y",
+      "的 力量",
+      "资 助",
+      "看到 的",
+      "看 到的",
+      "in ely",
+      "ine ly",
+      "inel y",
+      "▁care ers",
+      "▁career s",
+      "上 级",
+      "▁ret rieve",
+      "▁retriev e",
+      "▁ retrieve",
+      "研究 员",
+      "▁un re",
+      "▁u nre",
+      "实业 有限公司",
+      "实 业有限公司",
+      "▁Be ijing",
+      "中 所",
+      "驳回 上诉",
+      "par ency",
+      "pare ncy",
+      "paren cy",
+      "拆 除",
+      "术 语",
+      "dan ger",
+      "d anger",
+      "pre view",
+      "prev iew",
+      "p review",
+      "IC AL",
+      "ICA L",
+      "I CAL",
+      "却 被",
+      "nB orn",
+      "nBo rn",
+      "▁mot ivated",
+      "▁motiv ated",
+      "▁motivate d",
+      "▁Ed inburgh",
+      "分 期",
+      "▁\" {{",
+      "▁\"{ {",
+      "▁ \"{{",
+      "新 城",
+      "激 动",
+      "▁v iable",
+      "▁via ble",
+      "▁vi able",
+      "M o",
+      "▁c alm",
+      "▁ca lm",
+      "▁cal m",
+      "最多 的",
+      "最 多的",
+      "▁f olk",
+      "▁fol k",
+      "▁ folk",
+      "▁ver te",
+      "▁vert e",
+      "▁ verte",
+      "典型 的",
+      "典 型的",
+      "ar ium",
+      "ari um",
+      "a rium",
+      "▁h ind",
+      "▁hi nd",
+      "▁ hind",
+      "th ead",
+      "the ad",
+      "t head",
+      "▁D rug",
+      "▁Dr ug",
+      "▁occup ied",
+      "▁ occupied",
+      "门 户",
+      "第一百 一",
+      "这 张",
+      "▁M I",
+      "▁ MI",
+      "st ates",
+      "state s",
+      "stat es",
+      "sta tes",
+      ")) );\\",
+      "))) ;\\",
+      ") ));\\",
+      "▁path ological",
+      "▁pathologic al",
+      "Se lf",
+      "S elf",
+      "误 差",
+      "▁che aper",
+      "▁cheap er",
+      "▁V alent",
+      "▁Val ent",
+      "▁Vale nt",
+      "除 此",
+      "qu in",
+      "光 学",
+      "▁r he",
+      "▁rh e",
+      "再审 申请人",
+      "再审申请 人",
+      "▁S witch",
+      "▁Sw itch",
+      "▁ Switch",
+      "减去有期徒刑 一年",
+      "ht m",
+      "h tm",
+      "▁pharm aceutical",
+      "要 看",
+      "▁A CT",
+      "▁AC T",
+      "▁ ACT",
+      "▁fant asy",
+      "▁fantas y",
+      "的面 积",
+      "的 面积",
+      "正方 形",
+      "正 方形",
+      "ar as",
+      "ara s",
+      "a ras",
+      "Art icle",
+      "▁m oral",
+      "▁mor al",
+      "▁mo ral",
+      "tr aining",
+      "tra ining",
+      "train ing",
+      "▁ev olved",
+      "▁evolve d",
+      "动 手",
+      "无 可",
+      "模 范",
+      "折抵刑期 一日",
+      "礼 品",
+      "▁L eft",
+      "▁Le ft",
+      "▁ Left",
+      "ra ys",
+      "ray s",
+      "r ays",
+      "re na",
+      "ren a",
+      "r ena",
+      "经历 了",
+      "li ning",
+      "lin ing",
+      "l ining",
+      "▁F ern",
+      "▁Fe rn",
+      "▁Fer n",
+      "▁Mount ains",
+      "▁Mountain s",
+      "▁qu est",
+      "▁que st",
+      "▁ quest",
+      "cal c",
+      "ca lc",
+      "c alc",
+      "▁ 西",
+      "红 旗",
+      "▁sub sets",
+      "▁subset s",
+      "▁subs ets",
+      "▁Hamp shire",
+      "▁c ement",
+      "▁ce ment",
+      "▁m akers",
+      "▁make rs",
+      "▁ma kers",
+      "▁maker s",
+      "▁ makers",
+      "▁v ein",
+      "▁ve in",
+      "靠 近",
+      "▁cur v",
+      "▁cu rv",
+      "▁cons erved",
+      "▁conserv ed",
+      "▁conserve d",
+      "精 密",
+      "off ice",
+      "nS ource",
+      "n Source",
+      "▁M uch",
+      "▁Mu ch",
+      "le ans",
+      "lean s",
+      "lea ns",
+      "▁config urations",
+      "▁configuration s",
+      "ba ch",
+      "bac h",
+      "b ach",
+      "fr astructure",
+      "fra structure",
+      "于本判决 生效之日起",
+      "▁anal ytics",
+      "▁analyt ics",
+      "▁analytic s",
+      "▁ analytics",
+      "t T",
+      "各 界",
+      "TO N",
+      "T ON",
+      "not ification",
+      "an dal",
+      "and al",
+      "anda l",
+      "渴 望",
+      "你 有",
+      "▁act resses",
+      "▁actress es",
+      "物 种",
+      "行政 区",
+      "▁t iles",
+      "▁tile s",
+      "▁til es",
+      "▁ti les",
+      "▁ tiles",
+      "红 包",
+      "We st",
+      "W est",
+      "▁Ex cell",
+      "▁Excel l",
+      "▁Exc ell",
+      "▁P ow",
+      "▁Po w",
+      "▁tour ist",
+      "▁tou rist",
+      "om i",
+      "o mi",
+      "▁V M",
+      "▁ VM",
+      "▁D est",
+      "▁De st",
+      "▁Des t",
+      "▁ Dest",
+      "▁out fit",
+      "▁styl ish",
+      "▁Sp rings",
+      "▁Spring s",
+      "▁Spr ings",
+      "AN S",
+      "A NS",
+      "客户 的",
+      "W F",
+      "▁p izza",
+      "▁conf idential",
+      "▁confident ial",
+      "愿 望",
+      "ch t",
+      "c ht",
+      "显 存",
+      "re fresh",
+      "ref resh",
+      "wi fi",
+      "w ifi",
+      "ft en",
+      "f ten",
+      "▁fig ured",
+      "▁figure d",
+      "{{ \\\\",
+      "{{\\ \\",
+      "{ {\\\\",
+      "▁ne ighbors",
+      "▁neighbor s",
+      "▁neighb ors",
+      "▁ neighbors",
+      "] \"",
+      "WI N",
+      "W IN",
+      "ac cur",
+      "acc ur",
+      "▁B erg",
+      "▁Be rg",
+      "▁Ber g",
+      "管理 工作",
+      "av i",
+      "a vi",
+      "sc ar",
+      "s car",
+      "▁M SC",
+      "▁MS C",
+      "▁ MSC",
+      "▁adv ised",
+      "▁advis ed",
+      "▁advise d",
+      "▁don ations",
+      "▁donation s",
+      "▁( `",
+      "▁ (`",
+      "大 规模",
+      "也是 一",
+      "也 是一",
+      "qu it",
+      "q uit",
+      "~ /",
+      "▁sh eep",
+      "▁she ep",
+      "▁K ings",
+      "▁King s",
+      "▁Kin gs",
+      "下 次",
+      "] /",
+      "大 人",
+      "抵 抗",
+      "man ds",
+      "mand s",
+      "m ands",
+      "oi l",
+      "o il",
+      "rc he",
+      "r che",
+      "志愿 服务",
+      "▁Great er",
+      "▁Gre ater",
+      "is tically",
+      "ist ically",
+      "istic ally",
+      "istical ly",
+      "结构 的",
+      "终结 本院",
+      "▁E VENT",
+      "▁EV ENT",
+      "▁EVEN T",
+      "▁ EVENT",
+      "He aders",
+      "Header s",
+      "Head ers",
+      "版 的",
+      "nE nd",
+      "nEn d",
+      "n End",
+      "ind ent",
+      "inde nt",
+      "av age",
+      "ava ge",
+      "▁weight ed",
+      "▁weigh ted",
+      "▁ weighted",
+      "▁compl iment",
+      "考 点",
+      "fo ny",
+      "fon y",
+      "f ony",
+      "▁B ec",
+      "▁Be c",
+      "▁Brook lyn",
+      "你 怎么",
+      "▁acc essed",
+      "▁access ed",
+      "在 卷",
+      "na bla",
+      "nab la",
+      "▁s pine",
+      "▁sp ine",
+      "▁spin e",
+      "▁sub tract",
+      "开始 了",
+      "So ftware",
+      "Soft ware",
+      "在互联网 公布",
+      "PRO JECT",
+      "▁wor ried",
+      "吸 附",
+      "第二百零 六条",
+      "k appa",
+      "un ate",
+      "una te",
+      "一 座",
+      "天然 气",
+      "do ors",
+      "door s",
+      "d oors",
+      "▁st aining",
+      "▁stain ing",
+      "还 将",
+      "▁pass age",
+      "el lers",
+      "ell ers",
+      "eller s",
+      "elle rs",
+      "隐 私",
+      "▁V it",
+      "▁Vi t",
+      "民二 初字第",
+      "claim er",
+      "我 去",
+      "AL SE",
+      "ALS E",
+      "已经 成为",
+      "▁coll isions",
+      "▁collision s",
+      "及 相关",
+      "St ation",
+      "Stat ion",
+      "é r",
+      "全 方位",
+      "Li mit",
+      "L imit",
+      "ul ence",
+      "ule nce",
+      "▁p orts",
+      "▁port s",
+      "▁por ts",
+      "▁ ports",
+      "▁demonstr ation",
+      "▁c ursor",
+      "▁ cursor",
+      "▁Con dition",
+      "▁Cond ition",
+      "▁ Condition",
+      "毕业 后",
+      "经济 损失",
+      "十 足",
+      "ge ts",
+      "get s",
+      "g ets",
+      "▁M all",
+      "▁Mal l",
+      "▁Ma ll",
+      "如果 不",
+      "in sp",
+      "ins p",
+      "i nsp",
+      "▁p hp",
+      "▁ph p",
+      "▁ php",
+      "制 冷",
+      "口 市",
+      "▁D ays",
+      "▁Day s",
+      "▁Da ys",
+      "▁ Days",
+      "善 于",
+      "▁antib iotics",
+      "▁antibiotic s",
+      "▁consult ing",
+      "▁Arch ive",
+      "▁Arc hive",
+      "▁ Archive",
+      "及 利息",
+      "” )",
+      "就 把",
+      "刑期 从",
+      "▁f uzzy",
+      "▁G B",
+      "▁ GB",
+      "最新 的",
+      "最 新的",
+      "▁f lip",
+      "▁fl ip",
+      "▁ flip",
+      "主 力",
+      "等 腰",
+      "▁stim ulus",
+      "▁stimul us",
+      "一句 话",
+      "一 句话",
+      "▁compr ised",
+      "▁comprise d",
+      "inf ected",
+      "infect ed",
+      "有的 人",
+      "有 的人",
+      "▁margin al",
+      "▁marg inal",
+      "看 得",
+      "▁商 品",
+      "▁ 商品",
+      "▁cent res",
+      "▁centre s",
+      "▁centr es",
+      "nCONCLUS ION",
+      "▁crit ically",
+      "▁critical ly",
+      "▁critic ally",
+      "▁similar ly",
+      "▁Enter prise",
+      "分 居",
+      "参 照",
+      "▁c hips",
+      "▁ch ips",
+      "▁chip s",
+      "▁chi ps",
+      "会 导致",
+      "▁D rop",
+      "▁Dr op",
+      "▁Dro p",
+      "▁ Drop",
+      "妊 娠",
+      "ma ker",
+      "make r",
+      "mak er",
+      "m aker",
+      "bre aking",
+      "break ing",
+      "刚 好",
+      "▁es cap",
+      "▁esc ap",
+      "▁pat ent",
+      "化 合物",
+      "▁arg ued",
+      "▁argue d",
+      "▁argu ed",
+      "▁cent ered",
+      "▁center ed",
+      "▁ centered",
+      "starts with",
+      "说 什么",
+      "duc ational",
+      "ducation al",
+      "射 击",
+      "nF C",
+      "n FC",
+      "it erator",
+      "iter ator",
+      "▁in k",
+      "▁i nk",
+      "▁ ink",
+      "ass ium",
+      "废 弃",
+      "▁ent ropy",
+      "▁entr opy",
+      "▁ entropy",
+      "▁T rad",
+      "▁Tr ad",
+      "▁Tra d",
+      "严 谨",
+      "的 比",
+      "联合 会",
+      "▁cons iders",
+      "▁consider s",
+      "▁intr insic",
+      "ka r",
+      "k ar",
+      "▁satisf ying",
+      "▁satisfy ing",
+      "ff f",
+      "f ff",
+      "▁complement ary",
+      "Qu ant",
+      "Q uant",
+      "H B",
+      "好 奇",
+      "反 射",
+      "骨 干",
+      "bet ween",
+      "b etween",
+      "所有 人",
+      "所 有人",
+      "▁K az",
+      "▁Ka z",
+      "可以 使用",
+      "可以使 用",
+      "▁Every thing",
+      "▁ Everything",
+      "三 维",
+      "▁ 做法",
+      "т е",
+      "▁Mad rid",
+      "受 益",
+      "▁h unting",
+      "▁hun ting",
+      "▁hunt ing",
+      "不宜 在互联网公布",
+      "▁eth ical",
+      "政 法",
+      "nS p",
+      "n Sp",
+      "▁E sc",
+      "▁Es c",
+      "▁ Esc",
+      "▁S lo",
+      "▁Sl o",
+      "之一 的",
+      "之 一的",
+      "▁G ary",
+      "▁Ga ry",
+      "▁Gar y",
+      "▁anal ytic",
+      "▁analy tic",
+      "▁analyt ic",
+      "▁ analytic",
+      "ge ncy",
+      "gen cy",
+      "g ency",
+      "▁c ust",
+      "▁cu st",
+      "▁re sp",
+      "▁r esp",
+      "▁res p",
+      "▁ resp",
+      "在服刑期间 确有悔改表现",
+      "** :",
+      "大 鼠",
+      "nC opyright",
+      "n Copyright",
+      "fl uid",
+      "flu id",
+      "▁Eff ect",
+      "▁ Effect",
+      "ent ities",
+      "enti ties",
+      "G l",
+      "▁b g",
+      "▁ bg",
+      "▁N T",
+      "▁ NT",
+      "nOr iginally",
+      "nOriginal ly",
+      "时 时",
+      "▁C ycl",
+      "▁Cy cl",
+      "立 场",
+      "▁f ür",
+      "淋 浴",
+      "sens itive",
+      "s ensitive",
+      "▁discuss ing",
+      "ac s",
+      "a cs",
+      "MA KE",
+      "▁D eg",
+      "▁De g",
+      "▁ Deg",
+      "▁O cc",
+      "▁Oc c",
+      "▁ Occ",
+      "Ci rc",
+      "C irc",
+      "番 茄",
+      "签 名",
+      "▁B uck",
+      "▁Bu ck",
+      "▁Buc k",
+      "▁W ait",
+      "▁Wa it",
+      "▁ Wait",
+      "nR ailway",
+      "C u",
+      "是 用",
+      "We ight",
+      "W eight",
+      "▁th ro",
+      "▁thr o",
+      "▁ thro",
+      "▁ass isted",
+      "▁assist ed",
+      "▁ assisted",
+      "到 大",
+      "口 服",
+      "区块 链",
+      "说 说",
+      "▁k er",
+      "▁ke r",
+      "▁ ker",
+      "经 传票传唤",
+      "OP T",
+      "O PT",
+      "产业 链",
+      "▁fail ures",
+      "▁failure s",
+      "高 手",
+      "▁Me eting",
+      "▁Meet ing",
+      "th a",
+      "t ha",
+      "▁ 毕业院校",
+      "有一 定",
+      "有 一定",
+      "Be st",
+      "B est",
+      "io ur",
+      "i our",
+      "▁g rat",
+      "▁gr at",
+      "▁gra t",
+      "特 种",
+      "ma pping",
+      "map ping",
+      "m apping",
+      "▁gover ned",
+      "▁govern ed",
+      "▁travel ling",
+      "▁trav elling",
+      "▁stake holders",
+      "或 查封",
+      "▁be ef",
+      "▁bee f",
+      "▁c ure",
+      "▁cur e",
+      "▁cu re",
+      "ut enant",
+      "ute nant",
+      "uten ant",
+      "u tenant",
+      "▁Cor ps",
+      "▁Corp s",
+      "▁inst antly",
+      "▁instant ly",
+      "▁hipp ocamp",
+      "极 端",
+      "▁r ivers",
+      "▁river s",
+      "▁ri vers",
+      "▁riv ers",
+      "▁ rivers",
+      "ar ms",
+      "arm s",
+      "跑 步",
+      "总 监",
+      "G G",
+      "申请 人的",
+      "申请人 的",
+      "▁inter pol",
+      "▁ interpol",
+      "第一百四十五 条第一款的规定",
+      "第一百四十五条第一款 的规定",
+      "he ading",
+      "head ing",
+      "hea ding",
+      "自 学",
+      "脾 气",
+      "▁B P",
+      "▁ BP",
+      "▁t ies",
+      "▁tie s",
+      "▁ti es",
+      "▁ ties",
+      "质 保",
+      "ri el",
+      "rie l",
+      "r iel",
+      "re x",
+      "r ex",
+      "▁com pass",
+      "▁comp ass",
+      "▁ compass",
+      "▁C ircle",
+      "▁Circ le",
+      "▁Cir cle",
+      "▁ Circle",
+      "▁g ardens",
+      "▁gar dens",
+      "▁garden s",
+      "▁gard ens",
+      "▁l adies",
+      "▁la dies",
+      "斯 科",
+      "no s",
+      "n os",
+      "▁ad ren",
+      "▁ adren",
+      "自 来",
+      "唯一 的",
+      "唯 一的",
+      "develop er",
+      "devel oper",
+      "ra me",
+      "ram e",
+      "r ame",
+      "▁K ay",
+      "▁Ka y",
+      "▁propos als",
+      "▁proposal s",
+      "普通 人",
+      "根据 您的",
+      "第七 十九条",
+      "第七十九 条",
+      "有 力",
+      "▁B ruce",
+      "▁Br uce",
+      "▁Bru ce",
+      "Com ponents",
+      "Component s",
+      "防 腐",
+      "▁is ot",
+      "▁iso t",
+      "▁pro found",
+      "▁prof ound",
+      "染 色",
+      "此 类",
+      "职 权",
+      "▁K ir",
+      "▁Ki r",
+      "▁H ERE",
+      "▁HE RE",
+      "▁HER E",
+      "▁ HERE",
+      "东莞 市",
+      "▁f ounding",
+      "▁found ing",
+      "▁fo unding",
+      "▁dead line",
+      "的 各种",
+      "▁b acked",
+      "▁back ed",
+      "▁ backed",
+      "▁stim ulated",
+      "▁stimul ated",
+      "▁stimulate d",
+      "▁ stimulated",
+      "Set ting",
+      "S etting",
+      "h p",
+      "调 料",
+      "bum s",
+      "bu ms",
+      "b ums",
+      "裁定书送达之日起 十日内",
+      "▁horm ones",
+      "▁hormone s",
+      "▁sustain ability",
+      "user Id",
+      "ex tr",
+      "ext r",
+      "▁Th ompson",
+      "▁Thom pson",
+      "帮 你",
+      "ap on",
+      "apo n",
+      "a pon",
+      "▁reserv oir",
+      "res olved",
+      "resolve d",
+      "▁h older",
+      "▁hold er",
+      "▁hol der",
+      "▁ holder",
+      "▁Col on",
+      "▁Co lon",
+      "dump s",
+      "d umps",
+      "很 重要",
+      "这道 题",
+      "这 道题",
+      "le tt",
+      "let t",
+      "l ett",
+      "▁De cor",
+      "▁Dec or",
+      "▁ Decor",
+      "一 身",
+      "▁▁ ▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁ ▁",
+      "▁▁▁▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁ ▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁ ▁▁",
+      "▁ ▁▁▁▁▁▁▁▁▁▁▁▁",
+      "诱 导",
+      "▁n ov",
+      "▁no v",
+      "▁ nov",
+      "▁Per fect",
+      "ia h",
+      "i ah",
+      "IS H",
+      "I SH",
+      "特殊 的",
+      "▁tw elve",
+      "㈰ ㈰",
+      "果 然",
+      "am ping",
+      "amp ing",
+      "res ses",
+      "ress es",
+      "r esses",
+      "▁purch ases",
+      "▁purchase s",
+      "mm a",
+      "m ma",
+      "给定 的",
+      "给 定的",
+      "拯 救",
+      "nW ho",
+      "nWh o",
+      "n Who",
+      "族自治 县",
+      "cor rect",
+      "c orrect",
+      "Bi nd",
+      "Bin d",
+      "B ind",
+      "organ ization",
+      "CS S",
+      "C SS",
+      "nA uthor",
+      "nAut hor",
+      "n Author",
+      "号 的",
+      "▁sch izophren",
+      "演 绎",
+      "▁conf using",
+      "lo cks",
+      "loc ks",
+      "lock s",
+      "l ocks",
+      "▁F riends",
+      "▁Friend s",
+      "▁Fri ends",
+      "▁st ochastic",
+      "起 码",
+      "os omes",
+      "osome s",
+      "oso mes",
+      "} &",
+      "com merce",
+      ": “",
+      "支 队",
+      "nR eturn",
+      "nRe turn",
+      "nRet urn",
+      "n Return",
+      "▁nutr itional",
+      "▁nutrition al",
+      "▁nutrit ional",
+      "春 秋",
+      "▁B ug",
+      "▁Bu g",
+      "▁ Bug",
+      "携带 宠物",
+      "手 中",
+      "▁o pts",
+      "▁op ts",
+      "▁opt s",
+      "▁ opts",
+      "hat tan",
+      "器 件",
+      "O W",
+      "or p",
+      "o rp",
+      "▁p aired",
+      "▁pair ed",
+      "▁pa ired",
+      "▁ paired",
+      "站 点",
+      "被执行人 李",
+      "ke rs",
+      "ker s",
+      "k ers",
+      "ew ise",
+      "e wise",
+      "▁Pub lished",
+      "▁Publish ed",
+      "▁ Published",
+      "▁anticip ated",
+      "▁anticipate d",
+      "客 车",
+      "主持 人",
+      "bl ocks",
+      "block s",
+      "blo cks",
+      "b locks",
+      "▁recruit ed",
+      "▁recru ited",
+      "▁U buntu",
+      "▁ Ubuntu",
+      "we bsite",
+      "web site",
+      "webs ite",
+      "section al",
+      "sect ional",
+      "int estinal",
+      "▁bre athing",
+      "▁breath ing",
+      "▁breat hing",
+      "▁in fil",
+      "▁i nfil",
+      "▁inf il",
+      "▁inv esting",
+      "▁invest ing",
+      "▁G PS",
+      "▁GP S",
+      "▁ GPS",
+      "▁Ass essment",
+      "▁Assess ment",
+      "▁ Assessment",
+      "▁u nfortunately",
+      "▁unf ortunately",
+      "▁unfort unately",
+      "▁unfortunate ly",
+      "都 需要",
+      "阻 止",
+      "|| |",
+      "| ||",
+      "uc er",
+      "uce r",
+      "u cer",
+      "即可 求出",
+      "▁r hythm",
+      "▁rhyth m",
+      "准 则",
+      "示 例",
+      "向 人民法院",
+      "hib ition",
+      "▁p reserved",
+      "▁pres erved",
+      "▁preserve d",
+      "且 有",
+      "恒 成立",
+      "▁Creat ive",
+      "▁Cre ative",
+      "▁rot ate",
+      "▁ rotate",
+      "无 穷",
+      "▁scholar ship",
+      "▁scholars hip",
+      "身体 健康",
+      "▁render ed",
+      "▁rend ered",
+      "▁Fe el",
+      "▁Fee l",
+      "as hi",
+      "ash i",
+      "a shi",
+      "ma tic",
+      "mat ic",
+      "m atic",
+      "▁vers atile",
+      "▁versa tile",
+      "证 件",
+      "▁fund ra",
+      "▁fu ndra",
+      "▁parad igm",
+      "▁Elect ron",
+      "▁Electro n",
+      "▁J ess",
+      "▁Jes s",
+      "▁Je ss",
+      "RE FIX",
+      "REF IX",
+      "▁contract ors",
+      "▁contr actors",
+      "▁contractor s",
+      "oc ent",
+      "o cent",
+      "▁S creen",
+      "▁Sc reen",
+      "▁Scre en",
+      "▁Scr een",
+      "▁ Screen",
+      "还 得",
+      "谷 歌",
+      "se lection",
+      "select ion",
+      "sel ection",
+      "s election",
+      "▁A rc",
+      "▁Ar c",
+      "▁ Arc",
+      "mark down",
+      "可以 选择",
+      "▁B achelor",
+      "Ro t",
+      "R ot",
+      "▁G M",
+      "▁ GM",
+      "ol ished",
+      "olis hed",
+      "o lished",
+      "幽 默",
+      "uit ive",
+      "u itive",
+      "医学 院",
+      "医 学院",
+      "▁v endors",
+      "▁vendor s",
+      "进行 的",
+      "进 行的",
+      "ut ors",
+      "utor s",
+      "uto rs",
+      "研究 的",
+      "▁E RR",
+      "▁ER R",
+      "▁ ERR",
+      "▁A IDS",
+      "▁AI DS",
+      "▁ AIDS",
+      "RE T",
+      "R ET",
+      "be ans",
+      "bean s",
+      "▁Install ation",
+      "▁convers ations",
+      "▁conversation s",
+      "肯定 是",
+      "rist ian",
+      "▁U nlike",
+      "▁Un like",
+      "▁home work",
+      "▁hom ework",
+      "til e",
+      "ti le",
+      "t ile",
+      "▁metast asis",
+      "nL ife",
+      "n Life",
+      "Bo ot",
+      "B oot",
+      "▁phosph at",
+      "▁f ur",
+      "▁fu r",
+      "▁ fur",
+      "▁desc ending",
+      "\" $",
+      "的 存款",
+      "▁flo oring",
+      "▁floor ing",
+      "▁biomark ers",
+      "▁biomarker s",
+      "▁主 演",
+      "▁ 主演",
+      "的人 物",
+      "的 人物",
+      "用 法",
+      "▁a u",
+      "▁ au",
+      "▁Hawai i",
+      "单 身",
+      "王 国",
+      "拆 迁",
+      "▁d ors",
+      "▁do rs",
+      "都 说",
+      "崛 起",
+      "▁cr imes",
+      "▁crim es",
+      "▁crime s",
+      "▁g land",
+      "▁gl and",
+      "取 代",
+      "▁aud it",
+      "▁au dit",
+      "▁audi t",
+      "▁ audit",
+      "▁shel ter",
+      "合 约",
+      "▁t ender",
+      "▁te nder",
+      "▁ten der",
+      "▁tend er",
+      "▁Liver pool",
+      "ding s",
+      "din gs",
+      "d ings",
+      "数 位",
+      "月 经",
+      "浴 室",
+      "免费 的",
+      "免 费的",
+      "一个 个",
+      "一 个个",
+      "con d",
+      "co nd",
+      "c ond",
+      "亲 戚",
+      "的 同学",
+      "▁z inc",
+      "当 选",
+      "sm art",
+      "s mart",
+      "肚 子",
+      "不 允许",
+      "器 官",
+      "▁M Y",
+      "▁ MY",
+      "▁phenomen a",
+      "▁Ex pression",
+      "▁Express ion",
+      "▁Exp ression",
+      "▁ Expression",
+      "▁interpret ed",
+      "立案 执行",
+      "▁dial ogue",
+      "▁dialog ue",
+      "▁t s",
+      "▁ ts",
+      "的重要 性",
+      "的 重要性",
+      "File Name",
+      "世 间",
+      "童 话",
+      "续 航",
+      "才 知道",
+      "nt emplate",
+      "n template",
+      "CU R",
+      "C UR",
+      "ole sale",
+      "oles ale",
+      "▁p g",
+      "▁ pg",
+      "转 载",
+      "被执行人 张",
+      "Sh ared",
+      "Share d",
+      "Sha red",
+      "乘 客",
+      "▁F alls",
+      "▁Fall s",
+      "▁Fal ls",
+      "▁perspect ives",
+      "▁perspective s",
+      "▁Element ary",
+      "Ab s",
+      "A bs",
+      "St ats",
+      "Stat s",
+      "▁W ard",
+      "▁War d",
+      "▁Wa rd",
+      "▁autom ation",
+      "数 十",
+      "秉 承",
+      "花 了",
+      "赛 车",
+      "▁excell ence",
+      "or neys",
+      "orney s",
+      "orne ys",
+      "▁C ass",
+      "▁Ca ss",
+      "▁Cas s",
+      "上 门",
+      "ki ll",
+      "kil l",
+      "k ill",
+      "oph age",
+      "ophag e",
+      "高 清",
+      "▁ch apters",
+      "▁chapter s",
+      "▁chap ters",
+      "芜 湖",
+      "一 笑",
+      "im mer",
+      "imm er",
+      "ie mann",
+      "iem ann",
+      "i emann",
+      "▁m ad",
+      "▁ma d",
+      "▁ mad",
+      "所以 我",
+      "pro tein",
+      "prot ein",
+      "prote in",
+      "大家 的",
+      "大 家的",
+      "▁visual ization",
+      "▁m p",
+      "▁ mp",
+      "▁M Ps",
+      "▁MP s",
+      "us ive",
+      "▁Mon itor",
+      "▁ Monitor",
+      "▁L em",
+      "▁Le m",
+      "▁att enu",
+      "▁atten u",
+      "np rivate",
+      "npr ivate",
+      "n private",
+      "▁respond ents",
+      "cent ral",
+      "c entral",
+      "▁G las",
+      "▁Gl as",
+      "ec ycle",
+      "e cycle",
+      "▁P eters",
+      "▁Pet ers",
+      "▁Pe ters",
+      "▁Peter s",
+      "▁Pete rs",
+      "ER Y",
+      "E RY",
+      "ze ns",
+      "zen s",
+      "z ens",
+      "Da vid",
+      "D avid",
+      "▁C emetery",
+      "图 文",
+      "▁I BM",
+      "▁IB M",
+      "▁ IBM",
+      "▁dr aws",
+      "▁draw s",
+      "对 中国",
+      "▁out line",
+      "▁ou tline",
+      "▁ outline",
+      "Pr es",
+      "Pre s",
+      "P res",
+      "▁P ET",
+      "▁PE T",
+      "▁ PET",
+      "▁s orts",
+      "▁sort s",
+      "▁sor ts",
+      "nP erson",
+      "nPer son",
+      "n Person",
+      "主 观",
+      "P B",
+      "ab olic",
+      "abol ic",
+      "abo lic",
+      "a bolic",
+      "羁押一日 折抵刑期一日",
+      "为主 要",
+      "为 主要",
+      "ro st",
+      "ros t",
+      "r ost",
+      "vi c",
+      "v ic",
+      "▁ex ceptions",
+      "▁except ions",
+      "▁exception s",
+      "▁ exceptions",
+      "> _",
+      "▁air way",
+      "el ection",
+      "ele ction",
+      "elect ion",
+      "e lection",
+      "▁t ends",
+      "▁ten ds",
+      "▁tend s",
+      "▁Te X",
+      "▁ TeX",
+      "修 订",
+      "▁re nown",
+      "目的 探讨",
+      "ze ros",
+      "zer os",
+      "zero s",
+      "z eros",
+      "▁M ode",
+      "▁Mod e",
+      "▁Mo de",
+      "▁ Mode",
+      "▁F ather",
+      "▁Fa ther",
+      "▁Fat her",
+      "default s",
+      "过 后",
+      "▁det ecting",
+      "▁detect ing",
+      "in ian",
+      "ini an",
+      "inia n",
+      "agram s",
+      "agr ams",
+      "agra ms",
+      "a grams",
+      "▁C roat",
+      "▁Cro at",
+      "少 了",
+      "IC S",
+      "I CS",
+      "▁am ateur",
+      "区 委",
+      "▁aband oned",
+      "▁abandon ed",
+      "星 星",
+      "na ments",
+      "nam ents",
+      "nament s",
+      "n aments",
+      "年中 国",
+      "年 中国",
+      "um bai",
+      "umb ai",
+      "umba i",
+      "▁n eph",
+      "▁ne ph",
+      "▁nep h",
+      "ur ally",
+      "ural ly",
+      "▁ 剧",
+      "水 库",
+      "oc ese",
+      "▁作品 目录",
+      "▁conflic ts",
+      "▁conflict s",
+      "预 警",
+      "it ects",
+      "ite cts",
+      "itect s",
+      "Sc roll",
+      "▁st rat",
+      "▁str at",
+      "▁col ored",
+      "▁color ed",
+      "▁ colored",
+      "▁log arith",
+      "ococc us",
+      "ococ cus",
+      "权利 人",
+      "▁网络 小说",
+      "▁ 网络小说",
+      "▁h elic",
+      "▁he lic",
+      "▁hel ic",
+      "Pre sent",
+      "Pres ent",
+      "P resent",
+      "т а",
+      "果 实",
+      "淋 巴",
+      "让 孩子",
+      "▁trem end",
+      "率 先",
+      "▁w elfare",
+      "▁wel fare",
+      "ke ley",
+      "kel ey",
+      "M obile",
+      "▁insp ire",
+      "▁inspir e",
+      "▁attract ed",
+      "▁attr acted",
+      "av an",
+      "ava n",
+      "a van",
+      "on gue",
+      "ong ue",
+      "▁G ram",
+      "▁Gr am",
+      "▁Gra m",
+      "道 士",
+      "Se lection",
+      "Select ion",
+      "S election",
+      "乌 克兰",
+      "▁acc idents",
+      "▁accident s",
+      "▁f leet",
+      "▁fle et",
+      "▁flee t",
+      "▁O t",
+      "IF A",
+      "I FA",
+      "▁K ap",
+      "▁Ka p",
+      "st o",
+      "s to",
+      "hed ules",
+      "hedul es",
+      "hedule s",
+      "▁kin etics",
+      "▁kinetic s",
+      "高新 区",
+      "高 新区",
+      "▁pl ots",
+      "▁plot s",
+      "▁ plots",
+      "时 机",
+      "▁Re pair",
+      "▁Rep air",
+      "间 隔",
+      "▁m d",
+      "▁ md",
+      "▁d as",
+      "▁da s",
+      "▁ das",
+      "▁T ob",
+      "▁To b",
+      "代 价",
+      "▁insert ed",
+      "素 材",
+      "般 的",
+      "▁J uan",
+      "▁Ju an",
+      "nS ongs",
+      "nSong s",
+      "size of",
+      "▁v endor",
+      "▁ven dor",
+      "▁ vendor",
+      "华 人",
+      "▁A bb",
+      "▁Ab b",
+      "新 开",
+      "▁B irth",
+      "▁Bir th",
+      "▁ Birth",
+      "▁gr ants",
+      "▁grant s",
+      "▁gran ts",
+      "畜 牧",
+      "▁im ported",
+      "▁import ed",
+      "▁imp orted",
+      "▁exp ecting",
+      "▁expect ing",
+      "▁o nClick",
+      "▁on Click",
+      "▁ onClick",
+      "电子 产品",
+      "var epsilon",
+      "vare psilon",
+      "原 文",
+      "▁jewel ry",
+      "▁s oup",
+      "▁so up",
+      "▁sou p",
+      "▁ soup",
+      "an ners",
+      "ann ers",
+      "anner s",
+      "anne rs",
+      "▁manip ulation",
+      "奥运 会",
+      "奥 运会",
+      "的 地",
+      "un ittest",
+      "unit test",
+      "uni ttest",
+      "合肥 市",
+      "ats by",
+      "▁M edit",
+      "▁Med it",
+      "▁Me dit",
+      "zi ng",
+      "z ing",
+      "sh ore",
+      "综合 运用",
+      "▁B etter",
+      "▁Bet ter",
+      "▁Fr eder",
+      "▁Fre der",
+      "▁Fred er",
+      "点 点",
+      "▁foot age",
+      "▁foo tage",
+      "gr ades",
+      "grad es",
+      "grade s",
+      "gra des",
+      "▁d fs",
+      "▁df s",
+      "▁ dfs",
+      "▁se venth",
+      "▁seven th",
+      "▁sevent h",
+      "▁enc ryption",
+      "▁encrypt ion",
+      "MAG ES",
+      "MAGE S",
+      "M AGES",
+      "q quad",
+      "na re",
+      "nar e",
+      "n are",
+      "月 的",
+      "等 着",
+      "▁S um",
+      "▁Su m",
+      "▁ Sum",
+      "▁trans itions",
+      "▁transition s",
+      "▁transit ions",
+      "异 性",
+      "enn ial",
+      "▁R y",
+      "应当 在",
+      "nin clude",
+      "n include",
+      "天 涯",
+      "▁D ir",
+      "▁Di r",
+      "▁ Dir",
+      "ab lab",
+      "abl ab",
+      "nM oths",
+      "客 运",
+      "▁t ick",
+      "▁ti ck",
+      "▁ tick",
+      "本院 不予支持",
+      "本院不予 支持",
+      "he ld",
+      "hel d",
+      "h eld",
+      "▁r uled",
+      "▁rule d",
+      "▁rul ed",
+      "▁ru led",
+      "但 他",
+      "▁laws uit",
+      "▁antib iotic",
+      "毫 升",
+      "▁bar rel",
+      "▁barr el",
+      "End point",
+      "副 教授",
+      "案 发",
+      "▁P rim",
+      "▁Pr im",
+      "▁Pri m",
+      "▁ Prim",
+      "何 时",
+      "严 肃",
+      "的女 人",
+      "的 女人",
+      "▁B loom",
+      "▁Bl oom",
+      "▁Blo om",
+      "▁metast atic",
+      "▁meta static",
+      "但是 在",
+      "但 是在",
+      "▁T ower",
+      "▁To wer",
+      "▁Tow er",
+      "涉及 到",
+      "▁over night",
+      "nH on",
+      "温 州",
+      "援 助",
+      "▁C ash",
+      "▁Ca sh",
+      "▁Cas h",
+      "▁ Cash",
+      "倡 导",
+      "对 他",
+      "酒 吧",
+      "os al",
+      "osa l",
+      "o sal",
+      "▁Ad ams",
+      "▁Adam s",
+      "▁Ada ms",
+      "oo ns",
+      "oon s",
+      "o ons",
+      "* ,",
+      "大 佬",
+      "我 没",
+      "board s",
+      "bo ards",
+      "▁com prom",
+      "▁comp rom",
+      "▁compr om",
+      "▁dem ographic",
+      "▁demo graphic",
+      "被告 的",
+      "Wh ile",
+      "() ));\\",
+      "()) );\\",
+      "())) ;\\",
+      "( )));\\",
+      "执行 期间",
+      "▁A y",
+      "▁spect acular",
+      "▁A bd",
+      "▁Ab d",
+      "te l",
+      "t el",
+      "▁P ool",
+      "▁Po ol",
+      "▁ Pool",
+      "▁C lose",
+      "▁Cl ose",
+      "▁Clo se",
+      "▁ Close",
+      "▁intern ationally",
+      "▁international ly",
+      "▁M etal",
+      "▁Me tal",
+      "▁Met al",
+      "▁Meta l",
+      "比赛 中",
+      "比 赛中",
+      "▁consider ably",
+      "今天 的",
+      "今 天的",
+      "▁ex cluded",
+      "▁exclude d",
+      "妹 子",
+      "Sc anner",
+      "Scan ner",
+      "An imation",
+      "Anim ation",
+      "三 人",
+      "暖 气",
+      "▁l ys",
+      "▁ly s",
+      "▁ lys",
+      "▁la bour",
+      "▁lab our",
+      "关 怀",
+      "主 导",
+      "曾 在",
+      "玩 的",
+      "▁it eration",
+      "▁iter ation",
+      "▁ iteration",
+      "at able",
+      "ata ble",
+      "a table",
+      "怀 特",
+      "▁sh ear",
+      "▁she ar",
+      "work flow",
+      "▁S hel",
+      "▁Sh el",
+      "▁She l",
+      "有 害",
+      "IE W",
+      "I EW",
+      "为 中国",
+      "体现 了",
+      "▁L ot",
+      "▁Lo t",
+      "▁repeated ly",
+      "▁repeat edly",
+      "肇 事",
+      "▁K EY",
+      "▁KE Y",
+      "▁ KEY",
+      "▁f ingers",
+      "▁fin gers",
+      "▁finger s",
+      "▁fing ers",
+      "大 使",
+      "▁N elson",
+      "的 文化",
+      "math frak",
+      "▁b ilateral",
+      "▁bil ateral",
+      "▁inc idents",
+      "▁incident s",
+      "见 证",
+      "▁sk etch",
+      "承 载",
+      "广 阔",
+      "下落 不明",
+      "▁introdu ces",
+      "▁introduce s",
+      "ng inx",
+      "n ginx",
+      "足球 运动员",
+      "▁s eal",
+      "▁se al",
+      "▁sea l",
+      "▁prom ises",
+      "▁promise s",
+      "▁v ine",
+      "▁vi ne",
+      "▁ vine",
+      "im ension",
+      "imens ion",
+      "▁sports people",
+      "▁dev ast",
+      "锅 炉",
+      "▁Fac ulty",
+      "趣 味",
+      "▁prevent ed",
+      "▁prev ented",
+      "▁HE ALTH",
+      "您的 选择",
+      "您 的选择",
+      "▁Con stitution",
+      "▁Const itution",
+      "▁my th",
+      "▁c argo",
+      "▁car go",
+      "▁ cargo",
+      "AB A",
+      "A BA",
+      "党组 书记",
+      "备 注",
+      "br ief",
+      "b rief",
+      "nex cept",
+      "n except",
+      "type name",
+      "typ ename",
+      "值 的",
+      "判断 对错",
+      "▁r ival",
+      "▁ri val",
+      "▁riv al",
+      "▁ rival",
+      "▁b row",
+      "▁br ow",
+      "▁bro w",
+      "反 思",
+      "la yers",
+      "lay ers",
+      "layer s",
+      "l ayers",
+      "▁int ens",
+      "▁b er",
+      "▁be r",
+      "▁ ber",
+      "真实 的",
+      "真 实的",
+      "rit eria",
+      "rite ria",
+      "riter ia",
+      "动 员",
+      "▁R eferences",
+      "▁Re ferences",
+      "▁Reference s",
+      "▁Refer ences",
+      "▁ References",
+      "▁veter an",
+      "▁p ets",
+      "▁pe ts",
+      "▁pet s",
+      "大 奖",
+      "nD id",
+      "nDi d",
+      "n Did",
+      "▁Den ver",
+      "um bing",
+      "umb ing",
+      "制作 的",
+      "制 作的",
+      "▁F urn",
+      "▁Fu rn",
+      "▁Fur n",
+      "Reg ion",
+      "ang hai",
+      "▁h andy",
+      "▁hand y",
+      "▁han dy",
+      "▁extra ordinary",
+      "▁ ”",
+      "▁nation wide",
+      "▁T B",
+      "▁ TB",
+      "/ :",
+      "轨 迹",
+      "小 红",
+      "轮 胎",
+      "▁c riter",
+      "▁cr iter",
+      "▁crit er",
+      "▁H A",
+      "▁ HA",
+      "an cers",
+      "ance rs",
+      "anc ers",
+      "ancer s",
+      "▁trans genic",
+      "▁b rowse",
+      "▁brow se",
+      "▁ browse",
+      "杨 某",
+      "▁m elt",
+      "▁me lt",
+      "▁mel t",
+      "en berg",
+      "me sh",
+      "mes h",
+      "m esh",
+      "id entity",
+      "ident ity",
+      "表现 为",
+      "表 现为",
+      "▁ac res",
+      "▁acre s",
+      "▁Dou glas",
+      "▁Doug las",
+      "收 缩",
+      "UL E",
+      "U LE",
+      "Sh op",
+      "S hop",
+      "▁B oys",
+      "▁Bo ys",
+      "▁Boy s",
+      "乌 鲁",
+      "合适 的答案",
+      "合适的 答案",
+      "阅读 选项",
+      "qu eous",
+      "que ous",
+      "▁icon ic",
+      "▁ic onic",
+      "▁reason ing",
+      "▁commission ed",
+      "噪 音",
+      "该 案",
+      "▁de serve",
+      "▁des erve",
+      "W H",
+      "军 人",
+      "在 中",
+      "man y",
+      "ma ny",
+      "m any",
+      "科 长",
+      "这里 的",
+      "这 里的",
+      "▁M AT",
+      "▁MA T",
+      "▁ MAT",
+      "热 的",
+      "年 至",
+      "lo ra",
+      "lor a",
+      "l ora",
+      "▁W elcome",
+      "▁Wel come",
+      "▁ Welcome",
+      "▁propag ation",
+      "pr il",
+      "pri l",
+      "p ril",
+      "ri ents",
+      "rient s",
+      "rien ts",
+      "r ients",
+      "▁author ized",
+      "▁ authorized",
+      "伊 朗",
+      "发展 和",
+      "▁e pic",
+      "▁ep ic",
+      "▁t une",
+      "▁tun e",
+      "▁tu ne",
+      "最高 的",
+      "最 高的",
+      "sh ields",
+      "shield s",
+      "▁s ingers",
+      "▁sing ers",
+      "▁sin gers",
+      "▁singer s",
+      "无 聊",
+      "战 术",
+      "超 出",
+      "北 路",
+      "nc an",
+      "n can",
+      "集团 公司",
+      "▁Scient ific",
+      "ablab la",
+      "▁pr ayer",
+      "▁pra yer",
+      "▁pray er",
+      "▁mys tery",
+      "▁myst ery",
+      "▁myster y",
+      "▁An dy",
+      "▁And y",
+      "cook ie",
+      "c ookie",
+      "▁coll apse",
+      "▁ collapse",
+      "iO S",
+      "i OS",
+      "▁different iate",
+      "Ce rt",
+      "C ert",
+      "py test",
+      "导 入",
+      "▁hor ror",
+      "▁ho rror",
+      "▁horr or",
+      "hu m",
+      "h um",
+      "的一 次",
+      "的 一次",
+      "> /",
+      "De velopment",
+      "Develop ment",
+      "je n",
+      "j en",
+      "nP lot",
+      "nPl ot",
+      "n Plot",
+      "▁D ig",
+      "▁Di g",
+      "▁ Dig",
+      "Col um",
+      "C olum",
+      "▁sl ides",
+      "▁slide s",
+      "▁ slides",
+      "要求 的",
+      "▁tor que",
+      "nM an",
+      "nMa n",
+      "n Man",
+      "SER VER",
+      "ja min",
+      "jam in",
+      "j amin",
+      "姿 态",
+      "▁alleg edly",
+      "▁alleged ly",
+      "大 楼",
+      "ir ates",
+      "ira tes",
+      "i rates",
+      "orect al",
+      "▁mur ine",
+      "▁mu rine",
+      "▁l imb",
+      "▁li mb",
+      "▁lim b",
+      "本 息",
+      "聚 会",
+      "天 赋",
+      "非常 的",
+      "▁ischem ia",
+      "丽 的",
+      "CL A",
+      "C LA",
+      "排 查",
+      "Sig ma",
+      "S igma",
+      "全 是",
+      "▁H eat",
+      "▁He at",
+      "uis ine",
+      "常 识",
+      "▁I M",
+      "▁ IM",
+      "▁R SS",
+      "▁RS S",
+      "▁ RSS",
+      "▁c ss",
+      "▁cs s",
+      "▁ css",
+      "▁High er",
+      "De mo",
+      "Dem o",
+      "D emo",
+      "em or",
+      "emo r",
+      "e mor",
+      "▁y eah",
+      "▁ye ah",
+      "口 头",
+      "面 部",
+      "▁e ncode",
+      "▁en code",
+      "▁enc ode",
+      "▁ encode",
+      "的 核心",
+      "▁N BA",
+      "▁NB A",
+      "▁ NBA",
+      "▁M arie",
+      "▁Mar ie",
+      "▁Ma rie",
+      "▁Mari e",
+      "▁ Marie",
+      "▁s its",
+      "▁sit s",
+      "▁si ts",
+      "▁C leveland",
+      "一辈 子",
+      "一 辈子",
+      "un dry",
+      "und ry",
+      "流 传",
+      "▁a ria",
+      "▁ar ia",
+      "▁ aria",
+      "▁F er",
+      "▁Fe r",
+      "得 分",
+      "言 论",
+      "▁any body",
+      "▁sec ured",
+      "▁secure d",
+      "ot imes",
+      "oti mes",
+      "o times",
+      "os ome",
+      "oso me",
+      "o some",
+      "其 在",
+      "偶 然",
+      "▁ 企业",
+      "强 势",
+      "▁v inyl",
+      "未 发现",
+      "deg ree",
+      "▁Mont real",
+      "螺 旋",
+      "Mi ll",
+      "Mil l",
+      "M ill",
+      "iv an",
+      "iva n",
+      "i van",
+      "▁S ure",
+      "▁Sur e",
+      "▁Su re",
+      "日立案 受理后",
+      "日立案受理 后",
+      "▁toile t",
+      "▁r ay",
+      "▁ra y",
+      "▁ ray",
+      "奖 项",
+      "Ser ies",
+      "Se ries",
+      "S eries",
+      "▁T amil",
+      "▁Ta mil",
+      "▁Tam il",
+      "受 法律保护",
+      "▁Common wealth",
+      "▁H at",
+      "▁Ha t",
+      "[] \\",
+      "[ ]\\",
+      "▁G raham",
+      "▁Gra ham",
+      "▁Nev ada",
+      "▁check out",
+      "▁ checkout",
+      "▁运动 项目",
+      "日 记",
+      "燃 料",
+      "表面 积",
+      "表 面积",
+      "即可 得到",
+      "▁P ref",
+      "▁Pr ef",
+      "▁Pre f",
+      "▁ Pref",
+      "▁up loaded",
+      "▁upload ed",
+      "婚 生",
+      "观 赏",
+      "ev a",
+      "e va",
+      "▁happ iness",
+      "偶 数",
+      "用 电",
+      "▁per pendicular",
+      "▁Christ opher",
+      "R P",
+      "▁ 地区",
+      "读 音",
+      "ment ation",
+      "▁L ar",
+      "▁La r",
+      "▁ Lar",
+      "占 比",
+      "▁c alculus",
+      "▁calcul us",
+      "▁calc ulus",
+      "柔 软",
+      "阴 阳",
+      "人才 培养",
+      "▁C ensus",
+      "▁F actory",
+      "▁Factor y",
+      "▁Fact ory",
+      "▁ Factory",
+      "▁sim plified",
+      "▁simpl ified",
+      "LO CK",
+      "LOC K",
+      "手 里",
+      "马 来",
+      "他们 在",
+      "他 们在",
+      "Sup er",
+      "Su per",
+      "S uper",
+      "不 满",
+      "都不 是",
+      "都 不是",
+      "▁p atio",
+      "▁pat io",
+      "报 错",
+      "▁Man ual",
+      "人民 代表大会",
+      "▁crypt oc",
+      "▁crypto c",
+      "▁inf lation",
+      "▁infl ation",
+      "无 敌",
+      "此 后",
+      "hi st",
+      "his t",
+      "h ist",
+      "▁b uck",
+      "▁bu ck",
+      "▁ buck",
+      "curr ency",
+      "c urrency",
+      "▁sleep ing",
+      "uation s",
+      "u ations",
+      "之 为",
+      "是我 国",
+      "是 我国",
+      "nB ibliography",
+      "▁aggreg ation",
+      "中 最",
+      "an nes",
+      "ann es",
+      "anne s",
+      "oa uth",
+      "o auth",
+      "AT IONS",
+      "ATION S",
+      "空 白",
+      "诉 请",
+      "ch lor",
+      "▁Equ ations",
+      "▁Equation s",
+      "▁Eq uations",
+      "re lim",
+      "rel im",
+      "reli m",
+      "loc file",
+      "▁{ '",
+      "▁ {'",
+      "▁cont amin",
+      "▁Per iod",
+      "▁ Period",
+      "▁att endance",
+      "▁attend ance",
+      "▁distinct ive",
+      "好 吃",
+      "▁m aker",
+      "▁make r",
+      "▁ma ker",
+      "▁ maker",
+      "valid ator",
+      "▁compar isons",
+      "▁comparison s",
+      "伟大 的",
+      "伟 大的",
+      "昨 日",
+      "委员会 委员",
+      "▁val ley",
+      "▁a like",
+      "▁al ike",
+      "▁ali ke",
+      "完成 后",
+      "re ply",
+      "rep ly",
+      "▁t orn",
+      "▁to rn",
+      "▁tor n",
+      "▁L ooking",
+      "▁Lo oking",
+      "▁Look ing",
+      "▁ Looking",
+      "越 多",
+      "本院 予以准许",
+      "本院予以 准许",
+      "柠 檬",
+      "ah l",
+      "a hl",
+      "史 上",
+      "Bin ary",
+      "B inary",
+      "▁sh ade",
+      "▁sha de",
+      "▁Exper imental",
+      "▁Experiment al",
+      "▁ Experimental",
+      "▁上 映时间",
+      "羡 慕",
+      "坐标 为",
+      "大 致",
+      "til ity",
+      "t ility",
+      "▁submit ting",
+      "por al",
+      "po ral",
+      "pora l",
+      "p oral",
+      "▁re lay",
+      "▁rel ay",
+      "▁w ishes",
+      "▁wish es",
+      "▁wis hes",
+      "▁earth qu",
+      "就是 要",
+      "就 是要",
+      "▁in complete",
+      "▁develop s",
+      "▁C alc",
+      "▁Cal c",
+      "▁Ca lc",
+      "▁ Calc",
+      "打 了",
+      "▁f iscal",
+      "▁fi scal",
+      "source s",
+      "s ources",
+      "▁H ydro",
+      "▁Hy dro",
+      "▁Hyd ro",
+      "▁ Hydro",
+      "外 汇",
+      "▁s inus",
+      "▁sin us",
+      "▁si nus",
+      "运 气",
+      "▁L P",
+      "▁ LP",
+      "搭 载",
+      "yl on",
+      "y lon",
+      "sur face",
+      "s urface",
+      "疗 法",
+      "font s",
+      "fon ts",
+      "的 思想",
+      "▁Un iversal",
+      "▁Univers al",
+      "▁Uni versal",
+      "▁ 否",
+      "▁ind ent",
+      "▁ indent",
+      "也 被",
+      "th lete",
+      "thlet e",
+      "▁s ne",
+      "▁sn e",
+      "年 以来",
+      "not ify",
+      "▁particip ant",
+      "▁W ang",
+      "▁Wa ng",
+      "▁Wan g",
+      "▁M aterials",
+      "▁Material s",
+      "▁re ven",
+      "▁r even",
+      "▁reve n",
+      "▁rev en",
+      "前 端",
+      "宁 县",
+      "其实 是",
+      "的 价值",
+      "▁d ans",
+      "▁da ns",
+      "▁dan s",
+      "▁In ternal",
+      "▁Int ernal",
+      "▁Intern al",
+      "▁Inter nal",
+      "▁ Internal",
+      "tS ystem",
+      "t System",
+      "直 观",
+      "一 波",
+      "▁W right",
+      "▁Wr ight",
+      "首 届",
+      "为 客户",
+      "▁up stream",
+      "▁V AL",
+      "▁VA L",
+      "▁ VAL",
+      "▁B und",
+      "▁Bu nd",
+      "▁Bun d",
+      "'] )\\",
+      "']) \\",
+      "' ])\\",
+      "Z one",
+      "▁block chain",
+      "有 线",
+      "梳 理",
+      "▁progn ostic",
+      "奇 数",
+      "▁M AX",
+      "▁MA X",
+      "▁ MAX",
+      "▁\" );\\",
+      "▁\") ;\\",
+      "▁ \");\\",
+      "q s",
+      "传 达",
+      "ME T",
+      "M ET",
+      "▁overs eas",
+      "▁overse as",
+      "之间 的关系",
+      "之间的 关系",
+      "燃 气",
+      "[ {",
+      "H g",
+      "}\" )\\",
+      "} \")\\",
+      "NU MBER",
+      "NUM BER",
+      "追 究",
+      "\\\\ |",
+      "还是 很",
+      "还 是很",
+      "д а",
+      "▁Russ ell",
+      "▁Rus sell",
+      "的 理由",
+      "风 光",
+      "in itions",
+      "init ions",
+      "inition s",
+      "T W",
+      "▁critic ism",
+      "▁\\' #",
+      "▁ \\'#",
+      "▁Balt imore",
+      "Al i",
+      "A li",
+      "的 优势",
+      "▁经营 范围",
+      "▁ 经营范围",
+      "星 座",
+      "ster dam",
+      "us p",
+      "u sp",
+      "▁th irty",
+      "comp iler",
+      "compile r",
+      "尔 多",
+      "纠 结",
+      "的 空间",
+      "ros ine",
+      "r osine",
+      "▁W alter",
+      "▁Wal ter",
+      "▁Walt er",
+      "liter als",
+      "lit erals",
+      "literal s",
+      "lite rals",
+      "缓 存",
+      "ic hi",
+      "ich i",
+      "i chi",
+      "fr ames",
+      "frame s",
+      "fram es",
+      "fra mes",
+      "▁inter f",
+      "▁dist inguished",
+      "▁distingu ished",
+      "▁distinguish ed",
+      "nS tr",
+      "nSt r",
+      "n Str",
+      "不 详",
+      "ah n",
+      "a hn",
+      "受 欢迎",
+      "Vol ume",
+      "警 告",
+      "ff ff",
+      "fff f",
+      "f fff",
+      "cam era",
+      "just ify",
+      "散 文",
+      "▁ar med",
+      "▁arm ed",
+      "▁ armed",
+      "ub a",
+      "u ba",
+      "\\' ];\\",
+      "\\'] ;\\",
+      "\\']; \\",
+      "\\ '];\\",
+      "H y",
+      "x l",
+      "▁w is",
+      "▁wi s",
+      "num py",
+      "n umpy",
+      "▁C NN",
+      "▁CN N",
+      "▁ CNN",
+      "▁micro wave",
+      "▁mic rowave",
+      "UD E",
+      "U DE",
+      "he a",
+      "h ea",
+      "人 介绍",
+      "来自 于",
+      "▁dam ages",
+      "▁damage s",
+      "▁' '",
+      "▁ ''",
+      "st ick",
+      "s tick",
+      "▁b reat",
+      "▁br eat",
+      "▁bre at",
+      "struct ions",
+      "struction s",
+      "nG u",
+      "n Gu",
+      "ch air",
+      "cha ir",
+      "chai r",
+      "c hair",
+      "制 成",
+      "分 会",
+      "([ ]",
+      "( []",
+      "开 业",
+      "不愿 意",
+      "不 愿意",
+      "▁E ight",
+      "▁ 英",
+      "▁re leasing",
+      "▁H allow",
+      "▁Hall ow",
+      "▁Hal low",
+      "爱 上",
+      "国 王",
+      "ac ted",
+      "act ed",
+      "▁predict ing",
+      "▁biom edical",
+      "▁bio medical",
+      "sem antic",
+      "写 了",
+      "so ftware",
+      "soft ware",
+      "▁R ate",
+      "▁Ra te",
+      "▁Rat e",
+      "▁ Rate",
+      "▁N H",
+      "▁ NH",
+      "▁p ending",
+      "▁pen ding",
+      "▁pend ing",
+      "▁ pending",
+      "港 口",
+      "ce ster",
+      "ces ter",
+      "c ester",
+      "农 药",
+      "▁pan das",
+      "▁pand as",
+      "▁ pandas",
+      "▁St even",
+      "▁Ste ven",
+      "▁Steve n",
+      "> （",
+      "▁p ent",
+      "▁pe nt",
+      "▁pen t",
+      "▁ pent",
+      "nD avid",
+      "n David",
+      "抓 好",
+      "▁en compass",
+      "▁mand atory",
+      "门 市",
+      "nTh ree",
+      "n Three",
+      "▁news letter",
+      "▁predominant ly",
+      "▁\\ ',",
+      "▁\\' ,",
+      "▁ \\',",
+      "▁list ener",
+      "▁listen er",
+      "▁ listener",
+      "▁ret ained",
+      "▁retain ed",
+      "乳 腺",
+      "实 业",
+      "嫌疑 人",
+      "nV ol",
+      "n Vol",
+      "O ther",
+      "▁con verter",
+      "▁conver ter",
+      "▁convert er",
+      "▁ converter",
+      "Depend encies",
+      "▁C atal",
+      "▁Ca tal",
+      "▁Cat al",
+      "ro d",
+      "r od",
+      "十四 章",
+      "十 四章",
+      "per mission",
+      "perm ission",
+      "p ermission",
+      "以 内",
+      "▁K on",
+      "▁Ko n",
+      "希望 能",
+      "▁im pressed",
+      "▁imp ressed",
+      "▁impress ed",
+      "做 过",
+      "Point s",
+      "Po ints",
+      "P oints",
+      "▁PRO VID",
+      "▁port ray",
+      "Bo ard",
+      "B oard",
+      "另 行",
+      "核 算",
+      "谢 邀",
+      "个 单位",
+      "Ex pand",
+      "Exp and",
+      "▁F loor",
+      "▁Fl oor",
+      "▁Flo or",
+      "nw hat",
+      "n what",
+      "都是 一",
+      "都 是一",
+      "LEV ANCE",
+      "▁Br anch",
+      "▁Bra nch",
+      "▁ Branch",
+      "的 双",
+      "▁{ (",
+      "▁ {(",
+      "不等 式的",
+      "不等式 的",
+      "不 等式的",
+      "r v",
+      "其他 人",
+      "其 他人",
+      "ne ys",
+      "ney s",
+      "ter ns",
+      "tern s",
+      "nP at",
+      "n Pat",
+      "▁ π",
+      "▁pneum onia",
+      "刘某 某",
+      "刘 某某",
+      "wa sh",
+      "was h",
+      "w ash",
+      "Tr ack",
+      "Tra ck",
+      "T rack",
+      "▁a mid",
+      "▁am id",
+      "简 洁",
+      "▁P H",
+      "▁ PH",
+      "▁L oc",
+      "▁Lo c",
+      "▁ Loc",
+      "▁T i",
+      "▁ Ti",
+      "▁can nab",
+      "▁cann ab",
+      "▁c ab",
+      "▁ca b",
+      "▁ cab",
+      "Me r",
+      "M er",
+      "▁r outing",
+      "▁ro uting",
+      "▁rout ing",
+      "▁rou ting",
+      "▁ routing",
+      "A ir",
+      "db c",
+      "d bc",
+      "si g",
+      "s ig",
+      "▁r gba",
+      "▁rgb a",
+      "▁ rgba",
+      "▁vacc ines",
+      "▁vaccine s",
+      "▁vaccin es",
+      "▁mach inery",
+      "▁machine ry",
+      "小伙 伴",
+      "小 伙伴",
+      "学习 和",
+      "▁} }\"",
+      "▁}} \"",
+      "▁ }}\"",
+      "月 在",
+      "秋 季",
+      "▁Alber ta",
+      "▁Alb erta",
+      "▁Albert a",
+      "sp ath",
+      "s path",
+      "▁P ictures",
+      "▁Pic tures",
+      "▁Picture s",
+      "▁l l",
+      "▁ ll",
+      "搞 笑",
+      "▁r anks",
+      "▁rank s",
+      "▁ran ks",
+      "否 定",
+      "谈 话",
+      "方程 组",
+      "听 力",
+      "证据 确实",
+      "Print ln",
+      "的 车",
+      "破 产",
+      "▁o ils",
+      "▁oil s",
+      "讲 究",
+      "▁value d",
+      "▁val ued",
+      "▁valu ed",
+      "▁ valued",
+      "▁bas ement",
+      "▁base ment",
+      "▁amen ities",
+      "经销 商",
+      "▁b orders",
+      "▁border s",
+      "▁bor ders",
+      "不 像",
+      "▁end if",
+      "▁ endif",
+      "门 槛",
+      "▁filter ing",
+      "原告 诉称",
+      "atin um",
+      "ati num",
+      "▁est rogen",
+      "智 力",
+      "▁co ating",
+      "▁coat ing",
+      "▁standard ized",
+      "▁demonstr ating",
+      "有 问题",
+      "搜索 引擎",
+      "作 曲",
+      "的 说",
+      "as se",
+      "ass e",
+      "球 迷",
+      "Di g",
+      "D ig",
+      "sc ode",
+      "s code",
+      "▁w iring",
+      "▁wi ring",
+      "ng o",
+      "n go",
+      "doc tor",
+      "do ctor",
+      "▁C osta",
+      "▁Co sta",
+      "▁Cost a",
+      "▁Cos ta",
+      "▁search es",
+      "▁sear ches",
+      "host name",
+      "F B",
+      "nW hether",
+      "nWh ether",
+      "n Whether",
+      "non atomic",
+      "人 格",
+      "爸 妈",
+      "合同 履行",
+      "unt ary",
+      "▁f oster",
+      "▁fo ster",
+      "▁fost er",
+      "▁morph ological",
+      "厂 房",
+      "木 齐",
+      "▁l atex",
+      "▁la tex",
+      "▁late x",
+      "▁lat ex",
+      "▁ latex",
+      "生 平",
+      "中 部",
+      "▁over head",
+      "▁div orce",
+      "▁divor ce",
+      "es te",
+      "est e",
+      "e ste",
+      "▁Int elligence",
+      "▁Intellig ence",
+      "com plex",
+      "comp lex",
+      "compl ex",
+      "的 临床",
+      "年 起",
+      "nL ooking",
+      "nLook ing",
+      "nLo oking",
+      "n Looking",
+      "nT hus",
+      "nTh us",
+      "探 测",
+      "▁游戏 类型",
+      "自 带",
+      "▁di amond",
+      "▁dia mond",
+      "的 海",
+      "/\" \\",
+      "/ \"\\",
+      "icro bial",
+      "▁dist ress",
+      "▁di stress",
+      "▁distr ess",
+      "中 西",
+      "\"} ,",
+      "\" },",
+      "▁A V",
+      "▁ AV",
+      "Stand ard",
+      "▁C omb",
+      "▁Com b",
+      "▁Co mb",
+      "▁ Comb",
+      "cont aining",
+      "圣 诞",
+      "Mo use",
+      "M ouse",
+      "▁R AM",
+      "▁RA M",
+      "▁ RAM",
+      "▁st ones",
+      "▁stone s",
+      "▁sto nes",
+      "▁ stones",
+      "进 制",
+      "碰 到",
+      "▁H end",
+      "▁He nd",
+      "▁Hen d",
+      "对 着",
+      "大 批",
+      "▁U P",
+      "▁ UP",
+      "机动车 交通事故责任",
+      "与 服务",
+      "Le g",
+      "L eg",
+      "▁Ke nya",
+      "▁Ken ya",
+      "第十 五",
+      "第 十五",
+      "si x",
+      "s ix",
+      "政府 的",
+      "▁B right",
+      "▁Br ight",
+      "▁Brig ht",
+      "▁Bri ght",
+      "fa c",
+      "f ac",
+      "nV is",
+      "n Vis",
+      "▁ 韩",
+      "gen eric",
+      "gener ic",
+      "gene ric",
+      "意识 到",
+      "足球 俱乐部",
+      "▁Cent ury",
+      "▁k een",
+      "▁ke en",
+      "${ \\\\",
+      "$ {\\\\",
+      "us ions",
+      "usion s",
+      "异 的",
+      "▁Dep uty",
+      "▁sh allow",
+      "▁shall ow",
+      "司 马",
+      "▁cy tos",
+      "▁cyt os",
+      "▁cyto s",
+      "▁organ ize",
+      "C trl",
+      "花 开",
+      "fn ame",
+      "f name",
+      "套 房",
+      "▁L anka",
+      "▁Lan ka",
+      "▁Lank a",
+      "nM od",
+      "n Mod",
+      "▁l ag",
+      "▁la g",
+      "▁ lag",
+      "▁An ti",
+      "▁Ant i",
+      "月 任",
+      "sc hedule",
+      "sche dule",
+      "s chedule",
+      "**** ********",
+      "******** ****",
+      "wh y",
+      "w hy",
+      "▁S yn",
+      "▁Sy n",
+      "▁ Syn",
+      "▁fascin ating",
+      "▁fasc inating",
+      "\\\\) \\\\(",
+      "跟 你",
+      "na r",
+      "n ar",
+      "}} {{",
+      "}}{ {",
+      "} }{{",
+      "▁un em",
+      "▁u nem",
+      "▁une m",
+      "乌鲁 木齐",
+      ")) ;",
+      ") );",
+      "▁Similar ly",
+      "欧 阳",
+      "▁ 本",
+      "社会 科学",
+      "道 歉",
+      "\"$ {",
+      "\" ${",
+      "一 根",
+      "xx xx",
+      "xxx x",
+      "x xxx",
+      "▁Commun ication",
+      "离 心",
+      "SS L",
+      "S SL",
+      "▁Oper ations",
+      "▁Operation s",
+      "▁ Operations",
+      "▁J a",
+      "▁U SD",
+      "▁US D",
+      "▁ USD",
+      "扶 持",
+      "致 使",
+      "TP S",
+      "T PS",
+      "sc reens",
+      "screen s",
+      "s creens",
+      "▁de letion",
+      "▁delet ion",
+      "▁ deletion",
+      "礼 仪",
+      ":\" ,",
+      ": \",",
+      "is ec",
+      "ise c",
+      "i sec",
+      "网络 安全",
+      "即 为",
+      "只有 一个",
+      "只 有一个",
+      "gen eration",
+      "gener ation",
+      "gene ration",
+      "g eneration",
+      "▁C hi",
+      "▁Ch i",
+      "sh al",
+      "sha l",
+      "s hal",
+      "▁C hemistry",
+      "▁Chem istry",
+      "yn tax",
+      "ynt ax",
+      "Cam era",
+      "▁le ase",
+      "▁ lease",
+      "基金 会",
+      "▁s perm",
+      "▁sp erm",
+      "▁spe rm",
+      "▁sper m",
+      "经营 者",
+      "K B",
+      "脱 离",
+      "hydro xy",
+      "hydrox y",
+      "▁switch ed",
+      "▁swit ched",
+      "▁Ap pe",
+      "▁App e",
+      "▁T ags",
+      "▁Tag s",
+      "▁Ta gs",
+      "▁ Tags",
+      "( /",
+      "▁sub scribe",
+      "▁subscrib e",
+      "▁subs cribe",
+      "▁ subscribe",
+      "信 贷",
+      "ien na",
+      "i enna",
+      "▁P orn",
+      "▁Po rn",
+      "▁Por n",
+      "这个 人",
+      "这 个人",
+      "执行 案件",
+      "op a",
+      "o pa",
+      "等 工作",
+      "nFor mer",
+      "nForm er",
+      "▁Or leans",
+      ">( )",
+      "> ()",
+      "▁re pro",
+      "▁rep ro",
+      "▁repr o",
+      "se gment",
+      "seg ment",
+      "ne a",
+      "n ea",
+      "▁f ate",
+      "▁fa te",
+      "▁fat e",
+      "spec ies",
+      "不 忘",
+      "br ain",
+      "bra in",
+      "b rain",
+      "▁d ancing",
+      "▁dan cing",
+      "▁discipl ine",
+      "`) \\",
+      "` )\\",
+      "▁work force",
+      "选 定",
+      "不好 的",
+      "不 好的",
+      "更 容易",
+      "▁M enu",
+      "▁Me nu",
+      "▁Men u",
+      "▁ Menu",
+      "▁h o",
+      "▁ ho",
+      "▁S ay",
+      "▁Sa y",
+      "▁ Say",
+      "Found ation",
+      "▁Medit erranean",
+      "▁ 安",
+      "▁ 对",
+      "管理 和",
+      "可以 看到",
+      "可以看 到",
+      "▁ch ick",
+      "▁chi ck",
+      "▁chic k",
+      "▁under ground",
+      "▁und erground",
+      "癌 症",
+      "三 位",
+      "消 灭",
+      "兰 州",
+      "se g",
+      "s eg",
+      "▁opt imum",
+      "▁optim um",
+      "▁loc ality",
+      "▁local ity",
+      "▁S oft",
+      "▁So ft",
+      "▁Sof t",
+      "▁ Soft",
+      "act ivate",
+      "activ ate",
+      "ncon sole",
+      "n console",
+      "▁sub unit",
+      "▁G ray",
+      "▁Gr ay",
+      "▁Gra y",
+      "Error f",
+      "Err orf",
+      "▁Bar celona",
+      "特色 的",
+      "特 色的",
+      "ch annels",
+      "chan nels",
+      "channel s",
+      "nbl ablabla",
+      "▁C N",
+      "▁ CN",
+      "▁J oy",
+      "▁Jo y",
+      "▁wall et",
+      "▁wal let",
+      "▁ wallet",
+      "iz ar",
+      "iza r",
+      "无 比",
+      "更高 的",
+      "更 高的",
+      "is ol",
+      "iso l",
+      "i sol",
+      "重 阳",
+      "恶 性",
+      "▁qu asi",
+      "▁quas i",
+      "▁con gen",
+      "▁cong en",
+      "▁mon oclonal",
+      "胡 椒",
+      "▁RE LEVANCE",
+      "▁de scriptions",
+      "▁des criptions",
+      "▁description s",
+      "▁descript ions",
+      "抗 日",
+      "电 阻",
+      "纯 粹",
+      "放 下",
+      "tor ch",
+      "t orch",
+      "▁infar ction",
+      "乐 园",
+      "aj a",
+      "a ja",
+      "马克思 主义",
+      "▁R ank",
+      "▁Ra nk",
+      "▁Ran k",
+      "▁ Rank",
+      "喝 酒",
+      "Sh ape",
+      "Sha pe",
+      "公司 在",
+      "▁o m",
+      "▁ om",
+      "那 时",
+      "▁de fin",
+      "▁def in",
+      "出租 车",
+      "出 租车",
+      "UU ID",
+      "U UID",
+      "修 理",
+      "▁dis abilities",
+      "▁be et",
+      "▁bee t",
+      "ogram s",
+      "ogr ams",
+      "o grams",
+      "▁I DE",
+      "▁ID E",
+      "▁ IDE",
+      "党委 副书记",
+      "fol low",
+      "f ollow",
+      "▁re dis",
+      "▁red is",
+      "▁ redis",
+      "is Empty",
+      "x c",
+      "ak u",
+      "a ku",
+      "▁主要 食材",
+      "Is s",
+      "I ss",
+      "尔 斯",
+      "nM ean",
+      "nMe an",
+      "n Mean",
+      "▁ap olog",
+      "▁apo log",
+      "BU ILD",
+      "权 证",
+      "得 更",
+      "扩 散",
+      "应用 程序",
+      "▁fract ures",
+      "▁fracture s",
+      "Te ch",
+      "T ech",
+      "▁Ph oenix",
+      "迎 接",
+      "▁trans mitted",
+      "▁transmit ted",
+      "co u",
+      "c ou",
+      "人生 的",
+      "人 生的",
+      "▁H yp",
+      "▁Hy p",
+      "▁ Hyp",
+      "相等 的",
+      "相 等的",
+      "▁I FN",
+      "▁IF N",
+      "▁ IFN",
+      "▁streng ths",
+      "▁strength s",
+      "к и",
+      "▁follow ers",
+      "县 长",
+      "管理 系统",
+      "钻 石",
+      "等 相关",
+      "▁trans it",
+      "▁tran sit",
+      "风 扇",
+      "nS ta",
+      "nSt a",
+      "▁sm ell",
+      "ib e",
+      "i be",
+      "ra f",
+      "r af",
+      "▁Con vention",
+      "▁Conv ention",
+      "tb reak",
+      "t break",
+      "久 的",
+      "▁st eep",
+      "▁ste ep",
+      "Pl us",
+      "第四百 八",
+      "报 纸",
+      "红 外",
+      "▁V enezuel",
+      "▁c aring",
+      "▁car ing",
+      "▁ca ring",
+      "▁trouble s",
+      "▁trou bles",
+      "ba i",
+      "b ai",
+      "▁M M",
+      "▁ MM",
+      "Sp lit",
+      "Spl it",
+      "S plit",
+      "传统 文化",
+      "▁an tim",
+      "▁ant im",
+      "▁anti m",
+      "wo man",
+      "w oman",
+      "▁S pot",
+      "▁Sp ot",
+      "▁ Spot",
+      "之 家",
+      "sa t",
+      "s at",
+      "DO M",
+      "D OM",
+      "▁m att",
+      "▁mat t",
+      "▁ma tt",
+      "展 会",
+      "▁threat ened",
+      "▁threaten ed",
+      "IF I",
+      "I FI",
+      "设备 有限公司",
+      "▁C opy",
+      "▁Co py",
+      "▁Cop y",
+      "▁ Copy",
+      "▁renown ed",
+      "type script",
+      "types cript",
+      "▁appl iances",
+      "▁appliance s",
+      "▁determ ines",
+      "▁determine s",
+      "▁determin es",
+      "▁Pit tsburgh",
+      "呵 呵",
+      "▁S even",
+      "▁Se ven",
+      "▁Sev en",
+      "恶 意",
+      "fa iled",
+      "fail ed",
+      "f ailed",
+      "不 确定",
+      "Browse r",
+      "B rowser",
+      "和 大",
+      "K T",
+      "归 纳",
+      "为 目的",
+      "向 申请执行人",
+      "这个 世界",
+      "mo use",
+      "m ouse",
+      "R D",
+      "▁Net flix",
+      "▁pro fits",
+      "▁profit s",
+      "▁prof its",
+      "▁ profits",
+      "▁B asketball",
+      "P ut",
+      "▁Comp ared",
+      "▁Compar ed",
+      "▁Compare d",
+      "村 庄",
+      "克 力",
+      "▁N iger",
+      "▁Ni ger",
+      "▁Nig er",
+      "def inition",
+      "▁transl ational",
+      "▁translation al",
+      "▁ translational",
+      "之 内",
+      "讲 师",
+      "▁ischem ic",
+      "情 景",
+      "技术 和",
+      "▁Ste wart",
+      "▁Stew art",
+      "Sol ve",
+      "S olve",
+      "▁所属 地区",
+      "S ummary",
+      "w b",
+      "相 加",
+      "用 到",
+      "Cu r",
+      "C ur",
+      "外 语",
+      "天 堂",
+      "专 属",
+      "ME D",
+      "M ED",
+      "的 治疗",
+      "立方 米",
+      "言情 小说",
+      "▁ 南",
+      "}} ,",
+      "} },",
+      "遭 到",
+      "拉 克",
+      "▁P oss",
+      "▁Po ss",
+      "▁Pos s",
+      "oot ing",
+      "oo ting",
+      "o oting",
+      "For ce",
+      "F orce",
+      "▁Sim ply",
+      "基础 题",
+      "▁neon atal",
+      "▁neo natal",
+      "ed ges",
+      "edge s",
+      "认真 遵守法律法规及监规",
+      "认真遵守 法律法规及监规",
+      "og ue",
+      "o gue",
+      "▁M eg",
+      "▁Me g",
+      "▁p hy",
+      "▁ph y",
+      "▁ phy",
+      "nH ey",
+      "nHe y",
+      "n Hey",
+      "年 利率",
+      "▁游戏 目标",
+      "▁surviv ors",
+      "▁survivor s",
+      "巅 峰",
+      "可以 直接",
+      "▁fluores cent",
+      "▁fluor escent",
+      "防 守",
+      "腹 部",
+      "assert That",
+      "▁contract or",
+      "▁contr actor",
+      "终 点",
+      "▁' #",
+      "▁H unter",
+      "▁Hu nter",
+      "▁Hun ter",
+      "▁Hunt er",
+      ":** \\",
+      ":* *\\",
+      ": **\\",
+      "▁consult ant",
+      "▁in variant",
+      "▁invari ant",
+      "▁ invariant",
+      "签 证",
+      "bo ss",
+      "bos s",
+      "b oss",
+      "困 扰",
+      "增 多",
+      "▁w heat",
+      "▁wh eat",
+      "▁whe at",
+      "Di ff",
+      "D iff",
+      "ra ting",
+      "rat ing",
+      "r ating",
+      "▁mot if",
+      "▁mo tif",
+      "前 来",
+      "▁dis advant",
+      "再 来",
+      "指 针",
+      "已经 是",
+      "an ic",
+      "ani c",
+      "a nic",
+      "▁cl ustering",
+      "▁cluster ing",
+      "▁clust ering",
+      "▁according ly",
+      "▁accord ingly",
+      "roc ytes",
+      "rocyte s",
+      "r ocytes",
+      "出 身",
+      "▁re lies",
+      "▁rel ies",
+      "▁Rub y",
+      "▁Ru by",
+      "▁ Ruby",
+      "▁im plicated",
+      "▁impl icated",
+      "真实 意思表示",
+      "▁stim ulate",
+      "▁stimul ate",
+      "▁f iling",
+      "▁fil ing",
+      "▁fi ling",
+      "▁B ear",
+      "▁Be ar",
+      "▁Bea r",
+      "▁ Bear",
+      "▁s pherical",
+      "▁spher ical",
+      "微 波",
+      "bi a",
+      "b ia",
+      "▁Ind ividual",
+      "简体 中文",
+      "流 浪",
+      "▁B irmingham",
+      "▁metabol ites",
+      "▁metabolite s",
+      "▁n ude",
+      "▁nu de",
+      "▁nud e",
+      "▁sol uble",
+      "▁solub le",
+      "▁ 品牌",
+      "的 建议",
+      "▁re generation",
+      "▁reg eneration",
+      "说明 书",
+      "ip ly",
+      "ipl y",
+      "i ply",
+      "产 能",
+      "治 安",
+      "▁人物 履历",
+      "▁view ers",
+      "▁viewer s",
+      "又 能",
+      "▁break down",
+      "名下 的",
+      "名 下的",
+      "▁j uris",
+      "▁ju ris",
+      "▁jur is",
+      "▁L egal",
+      "▁Le gal",
+      "▁Leg al",
+      "▁l ying",
+      "▁ly ing",
+      "▁ lying",
+      "as ted",
+      "ast ed",
+      "aste d",
+      "a sted",
+      "套 路",
+      "▁Imp act",
+      "▁ Impact",
+      "ut er",
+      "ute r",
+      "u ter",
+      "▁path ogen",
+      "带 头",
+      "▁F ame",
+      "▁Fa me",
+      "nl ocal",
+      "n local",
+      "会 对",
+      "感 激",
+      "他 就",
+      "▁spokes man",
+      "Pl an",
+      "P lan",
+      "▁P el",
+      "▁Pe l",
+      "ev ery",
+      "ever y",
+      "eve ry",
+      "e very",
+      "▁cub ic",
+      "▁cu bic",
+      "cl ip",
+      "cli p",
+      "c lip",
+      "▁c ad",
+      "▁ca d",
+      "▁ cad",
+      "ar ious",
+      "ari ous",
+      "ario us",
+      "宝 石",
+      "pre vious",
+      "prev ious",
+      "pos able",
+      "北 区",
+      "河 流",
+      "▁LI ABILITY",
+      "向 下",
+      "▁S equ",
+      "▁Se qu",
+      "▁Seq u",
+      "▁ Sequ",
+      "▁po ison",
+      "▁l ever",
+      "▁le ver",
+      "▁lev er",
+      "耳 朵",
+      "Wh ich",
+      "Wh ite",
+      "Whit e",
+      "月参加 工作",
+      "忍 不住",
+      "▁Sh ah",
+      "▁Sha h",
+      "L T",
+      "Vis ible",
+      "V isible",
+      "表 格",
+      "▁Mod els",
+      "▁Model s",
+      "▁Mode ls",
+      "▁ Models",
+      "毫 不",
+      "▁G round",
+      "▁Gr ound",
+      "▁Gro und",
+      "ot hy",
+      "oth y",
+      "o thy",
+      "▁H ud",
+      "▁Hu d",
+      "▁Afghan istan",
+      "犹 豫",
+      "UE S",
+      "U ES",
+      "▁main stream",
+      "▁W ant",
+      "▁Wa nt",
+      "▁Wan t",
+      "▁ Want",
+      "金 山",
+      "▁car eg",
+      "▁care g",
+      "▁ca reg",
+      "▁P V",
+      "▁ PV",
+      "ho lm",
+      "hol m",
+      "▁viol ent",
+      "▁com press",
+      "▁comp ress",
+      "▁compre ss",
+      "▁compr ess",
+      "▁ compress",
+      "li p",
+      "l ip",
+      "ic hard",
+      "ich ard",
+      "i chard",
+      "CRIPT ION",
+      "IT S",
+      "I TS",
+      "he nd",
+      "hen d",
+      "h end",
+      "w heel",
+      "▁path ogens",
+      "▁pathogen s",
+      "▁remodel ing",
+      "▁pe ers",
+      "▁peer s",
+      "en cephal",
+      "ence phal",
+      "p H",
+      "ns udo",
+      "n sudo",
+      "sy n",
+      "s yn",
+      "▁Z e",
+      "▁model ling",
+      "▁mod elling",
+      "oc t",
+      "o ct",
+      "id ea",
+      "ide a",
+      "i dea",
+      "▁W ol",
+      "▁Wo l",
+      "▁g reet",
+      "▁gre et",
+      "nM us",
+      "n Mus",
+      "check ed",
+      "▁elim inated",
+      "▁eliminate d",
+      "获 悉",
+      "▁trans mit",
+      "了 好",
+      "Te x",
+      "T ex",
+      "▁} \\\\",
+      "▁}\\ \\",
+      "▁ }\\\\",
+      "哈哈 哈哈",
+      "哈哈哈 哈",
+      "哈 哈哈哈",
+      "▁h emor",
+      "▁he mor",
+      "▁hem or",
+      "生 死",
+      "▁ind ication",
+      "▁indic ation",
+      "▁con ferences",
+      "▁conference s",
+      "▁confer ences",
+      "N ative",
+      "it ance",
+      "ita nce",
+      "itan ce",
+      "▁retail ers",
+      "▁retailer s",
+      "is y",
+      "i sy",
+      "全 局",
+      "eh icle",
+      "▁E aster",
+      "▁East er",
+      "▁Eas ter",
+      "▁orient ed",
+      "▁ oriented",
+      "bin om",
+      "bi nom",
+      "▁w ashing",
+      "▁was hing",
+      "▁wash ing",
+      "导 体",
+      "一 九",
+      "但 不",
+      "or ic",
+      "ori c",
+      "o ric",
+      "▁f avorable",
+      "▁favor able",
+      "▁fav orable",
+      "▁cyt okine",
+      "▁cytok ine",
+      "nPost ed",
+      "nPo sted",
+      "n Posted",
+      "ul ia",
+      "uli a",
+      "u lia",
+      "个 小",
+      "al ym",
+      "aly m",
+      "▁econ omics",
+      "▁economic s",
+      "▁econom ics",
+      "Doc ker",
+      "D ocker",
+      "▁ex ped",
+      "▁exp ed",
+      "\\\\ \":",
+      "\\\\\" :",
+      "▁M aur",
+      "▁Ma ur",
+      "▁dist inction",
+      "▁distinct ion",
+      "▁H ey",
+      "▁He y",
+      "▁ Hey",
+      "进 化",
+      "▁V ert",
+      "▁Ver t",
+      "▁Ve rt",
+      "▁ Vert",
+      "▁Pe ace",
+      "▁R ugby",
+      "发现被执行人 有可供执行的财产",
+      "发现被执行人有 可供执行的财产",
+      "▁commun ic",
+      "▁In valid",
+      "▁Inv alid",
+      "▁ Invalid",
+      "外 界",
+      "失 业",
+      "▁G uy",
+      "▁Gu y",
+      "ns ure",
+      "app ings",
+      "apping s",
+      "▁S usan",
+      "▁Sus an",
+      "▁Su san",
+      "▁ear liest",
+      "mat ches",
+      "match es",
+      "m atches",
+      "▁T ig",
+      "▁Ti g",
+      "队 列",
+      "ad just",
+      "adj ust",
+      "▁K i",
+      "▁ Ki",
+      "慈 善",
+      "阿 姨",
+      "at hs",
+      "ath s",
+      "a ths",
+      "in stein",
+      "ins tein",
+      "inst ein",
+      "环境 下",
+      "理事 长",
+      "建 华",
+      "▁l ecture",
+      "▁lect ure",
+      "这个 数",
+      "这 个数",
+      "存 放",
+      "▁com mented",
+      "▁comm ented",
+      "▁comment ed",
+      "▁dis counts",
+      "▁discount s",
+      "住 福建省",
+      "DI S",
+      "D IS",
+      "▁enorm ous",
+      "▁dom inated",
+      "▁dominate d",
+      "你 觉得",
+      "▁N MR",
+      "▁NM R",
+      "农 场",
+      "抽 取",
+      "nt hen",
+      "nthe n",
+      "nth en",
+      "n then",
+      "模 仿",
+      "▁ex ports",
+      "▁exp orts",
+      "▁export s",
+      "▁ exports",
+      "以前 先行羁押的",
+      "O h",
+      "uc ker",
+      "uck er",
+      "▁vis ibility",
+      "▁ visibility",
+      "▁` /",
+      "▁b ra",
+      "▁br a",
+      "▁ bra",
+      "ug ust",
+      "ugu st",
+      "u gust",
+      "us tain",
+      "ust ain",
+      "▁bu yer",
+      "▁buy er",
+      "多 彩",
+      "▁V R",
+      "▁ VR",
+      "一 万",
+      "fr e",
+      "f re",
+      "执行 情况",
+      "▁B our",
+      "▁Bo ur",
+      "▁Bou r",
+      "▁Hallow een",
+      "▁neg atively",
+      "▁negative ly",
+      "防 疫",
+      "foo d",
+      "fo od",
+      "f ood",
+      "▁st ays",
+      "▁stay s",
+      "常 态",
+      "保 安",
+      "he my",
+      "hem y",
+      "h emy",
+      "少 量",
+      "故 原告",
+      "Rem ote",
+      "物 联网",
+      "▁t rait",
+      "▁tr ait",
+      "▁tra it",
+      "▁ trait",
+      "cut aneous",
+      "多项 式",
+      "▁coc aine",
+      "合作 伙伴",
+      "▁U pon",
+      "▁Up on",
+      "▁Com bin",
+      "▁Comb in",
+      "▁ Combin",
+      "▁c riterion",
+      "▁criter ion",
+      "▁K an",
+      "▁Ka n",
+      "Dec imal",
+      "▁pro fil",
+      "▁prof il",
+      "▁count ies",
+      "通 则",
+      "▁eas iest",
+      "某 一",
+      "汉 字",
+      "TA G",
+      "T AG",
+      "并 与",
+      "Ba y",
+      "B ay",
+      "us hes",
+      "ush es",
+      "马 拉",
+      "▁pur ple",
+      "▁ purple",
+      "ode ship",
+      "odes hip",
+      "query Selector",
+      "有 用",
+      "▁ 数据",
+      "背 包",
+      "了自己 的",
+      "了 自己的",
+      "大 事",
+      "M Q",
+      "▁pr imes",
+      "▁prim es",
+      "▁prime s",
+      "▁pri mes",
+      "▁ primes",
+      "▁lig ands",
+      "▁ligand s",
+      "我 喜欢",
+      "## \\",
+      "# #\\",
+      "ob serv",
+      "obs erv",
+      "o bserv",
+      "null able",
+      "▁H erm",
+      "▁He rm",
+      "▁Her m",
+      "何 况",
+      "▁r ally",
+      "▁rall y",
+      "度 和",
+      "是 他",
+      "▁F riend",
+      "▁Fri end",
+      "▁ Friend",
+      "先 把",
+      "的 国家",
+      "▁A long",
+      "▁Al ong",
+      "▁do zen",
+      "str aints",
+      "stra ints",
+      "straint s",
+      "strain ts",
+      "▁C LI",
+      "▁CL I",
+      "▁ CLI",
+      "听 取",
+      "外 包",
+      "存在 着",
+      "et ween",
+      "跟 随",
+      "Me sh",
+      "M esh",
+      "改 编",
+      "拖 欠",
+      "▁~ /",
+      "▁ ~/",
+      "▁re versed",
+      "▁rev ersed",
+      "▁reverse d",
+      "▁revers ed",
+      "▁rever sed",
+      "烈 士",
+      "▁c rypto",
+      "▁crypt o",
+      "▁cry pto",
+      "▁ crypto",
+      "难 点",
+      "的 答案",
+      "▁ex ert",
+      "po d",
+      "p od",
+      "ov al",
+      "ova l",
+      "o val",
+      "上 帝",
+      "▁f ought",
+      "▁fo ught",
+      "征收 社会抚养费",
+      "▁ter rible",
+      "▁terr ible",
+      "游戏 的",
+      "▁fab ulous",
+      "Ver tex",
+      "Vert ex",
+      "▁P rix",
+      "▁Pr ix",
+      "▁Pri x",
+      "▁S occer",
+      "▁Soc cer",
+      "转 发",
+      "V ec",
+      "第一百四十 五条",
+      "第一百四十五 条",
+      "婴儿 床",
+      "▁b uf",
+      "▁bu f",
+      "▁ buf",
+      "nF ootball",
+      "nFoot ball",
+      "▁mult ivariate",
+      "▁l y",
+      "▁ ly",
+      "就 知道",
+      "珠 宝",
+      "nH ence",
+      "nHe nce",
+      "▁Cont ainer",
+      "▁ Container",
+      "智能 化",
+      "没 啥",
+      "▁b lot",
+      "▁bl ot",
+      "▁blo t",
+      "nW ill",
+      "nWil l",
+      "n Will",
+      "等 多",
+      "▁R ace",
+      "▁Ra ce",
+      "▁Rac e",
+      "▁arch ae",
+      "▁p uzzle",
+      "▁puzz le",
+      "▁pu zzle",
+      "\" %",
+      "日向 被执行人",
+      "日 向被执行人",
+      "▁en rich",
+      "▁harm ful",
+      "AT OR",
+      "A TOR",
+      "▁g lasses",
+      "▁glass es",
+      "刑期从 判决执行之日起计算",
+      "▁air line",
+      "Com plete",
+      "Comp lete",
+      "▁bro nch",
+      "▁bron ch",
+      "朋友 们",
+      "朋 友们",
+      ") >",
+      "泡 沫",
+      "(* *",
+      "( **",
+      "▁s lower",
+      "▁sl ower",
+      "▁slow er",
+      "▁Al ice",
+      "▁Ali ce",
+      "▁Fe atures",
+      "▁Feature s",
+      "▁ Features",
+      "▁don ated",
+      "▁donate d",
+      "蜂 蜜",
+      "rows ers",
+      "rowser s",
+      "rowse rs",
+      "▁withdraw al",
+      "▁Vo iv",
+      "▁s pare",
+      "▁sp are",
+      "▁spa re",
+      "▁S TR",
+      "▁ST R",
+      "▁ STR",
+      "传奇 私服",
+      "▁Com mercial",
+      ", &",
+      "不 准",
+      "an ç",
+      "▁def ence",
+      "▁defe nce",
+      "中小 学",
+      "中 小学",
+      "▁dis tal",
+      "▁dist al",
+      "AR GET",
+      "ARG ET",
+      "▁READ ME",
+      "▁ README",
+      "et itive",
+      "De legate",
+      "▁cannab is",
+      "▁cann abis",
+      "当事 人的",
+      "当事人 的",
+      "▁剧 名",
+      "羽 毛",
+      "还有 一个",
+      "还 有一个",
+      "▁dep uty",
+      "▁deput y",
+      "▁b ust",
+      "▁bu st",
+      "▁bus t",
+      "经济技术 开发区",
+      "ion ed",
+      "io ned",
+      "ione d",
+      "i oned",
+      "老 的",
+      "Sh arp",
+      "Sha rp",
+      "▁N ad",
+      "▁Na d",
+      "yg on",
+      "y gon",
+      "▁Oper ation",
+      "▁Op eration",
+      "▁Opera tion",
+      "▁ Operation",
+      "麻 醉",
+      "az ed",
+      "aze d",
+      "a zed",
+      "开发 的",
+      "开 发的",
+      "教育 局",
+      "ol ytic",
+      "oly tic",
+      "▁al igned",
+      "▁align ed",
+      "▁ aligned",
+      "女 神",
+      "影响 到",
+      "▁c attle",
+      "▁ca ttle",
+      "▁cat tle",
+      "▁Arch itecture",
+      "▁Architect ure",
+      "m r",
+      "思想 政治",
+      "▁s ends",
+      "▁send s",
+      "▁sen ds",
+      "▁ 月",
+      "St e",
+      "S te",
+      "中 美",
+      "不一样 的",
+      "不一 样的",
+      "不 一样的",
+      "▁rad ar",
+      "▁ra dar",
+      "or ia",
+      "ori a",
+      "o ria",
+      "眼 泪",
+      "▁G C",
+      "▁ GC",
+      "ins ula",
+      "唱 歌",
+      "▁T ar",
+      "▁Ta r",
+      "▁g uns",
+      "▁gu ns",
+      "▁gun s",
+      "red uce",
+      "redu ce",
+      "▁esc ort",
+      "▁R ic",
+      "▁Ri c",
+      "▁c is",
+      "▁ci s",
+      "▁ cis",
+      "car bon",
+      "c arbon",
+      "▁D IS",
+      "▁DI S",
+      "▁ DIS",
+      "投 放",
+      "pr act",
+      "战 场",
+      "和分析 思路",
+      "inv asive",
+      "invas ive",
+      "自 杀",
+      "Li fe",
+      "Lif e",
+      "L ife",
+      "▁polymer ase",
+      "▁polym erase",
+      "找 个",
+      "▁R at",
+      "▁Ra t",
+      "▁l azy",
+      "▁la zy",
+      "▁ lazy",
+      "Re direct",
+      "Red irect",
+      "塞 尔",
+      "有关 规定",
+      "理论 与",
+      "一般 来说",
+      "▁YO UR",
+      "▁YOU R",
+      "跨 越",
+      "pers onal",
+      "person al",
+      "p ersonal",
+      "As p",
+      "A sp",
+      "的风 险",
+      "的 风险",
+      "pect ive",
+      "p ective",
+      "▁d aughters",
+      "▁daughter s",
+      "不 堪",
+      "减半收取 计",
+      "▁Al pha",
+      "▁Alp ha",
+      "▁ Alpha",
+      "▁aff iliate",
+      "▁affil iate",
+      "al igned",
+      "align ed",
+      "澳 洲",
+      "▁G hana",
+      "▁Gh ana",
+      "我真 的",
+      "我 真的",
+      "年 毕业于",
+      "瑞 士",
+      "让 学生",
+      "如 图",
+      "▁c ous",
+      "▁co us",
+      "▁cou s",
+      "▁ cous",
+      "▁b less",
+      "▁bl ess",
+      "▁ble ss",
+      "ar ded",
+      "ard ed",
+      "arde d",
+      "ict ional",
+      "iction al",
+      "钥 匙",
+      "ud es",
+      "ude s",
+      "u des",
+      "▁&& \\",
+      "比如 说",
+      "ri tion",
+      "rit ion",
+      "r ition",
+      "▁Jon athan",
+      "▁St ars",
+      "▁Star s",
+      "▁Sta rs",
+      "管 制",
+      "仅供 参考",
+      "▁cast le",
+      "▁cas tle",
+      "▁ castle",
+      "分 裂",
+      "at omy",
+      "ato my",
+      "atom y",
+      "▁State ment",
+      "▁Stat ement",
+      "▁ Statement",
+      "▁p ine",
+      "▁pi ne",
+      "▁pin e",
+      "▁ pine",
+      "Per m",
+      "Pe rm",
+      "P erm",
+      "ac ting",
+      "act ing",
+      "actin g",
+      "起 步",
+      "]) )",
+      "] ))",
+      "sh ine",
+      "shi ne",
+      "▁harm onic",
+      "▁harmon ic",
+      "▁ harmonic",
+      "中 外",
+      "紧 紧",
+      "fe d",
+      "f ed",
+      "to t",
+      "t ot",
+      "出生 于",
+      "出 生于",
+      "深 深",
+      "技术 人员",
+      "▁h emorrh",
+      "▁hemor rh",
+      "▁def icits",
+      "▁deficit s",
+      "▁comm une",
+      "▁commun e",
+      "的 原",
+      "▁pre scribed",
+      "▁pres cribed",
+      "浸 泡",
+      "▁s ized",
+      "▁size d",
+      "▁si zed",
+      "▁ sized",
+      "▁j oints",
+      "▁join ts",
+      "▁joint s",
+      "▁jo ints",
+      "江 山",
+      "lu ence",
+      "知识 和",
+      "vis ion",
+      "v ision",
+      "较强 的",
+      "较 强的",
+      "▁$ \\\\{",
+      "▁$\\\\ {",
+      "▁I con",
+      "▁ Icon",
+      "▁int rac",
+      "▁intr ac",
+      "▁intra c",
+      "心 动",
+      "▁( {\\",
+      "▁({ \\",
+      "▁ ({\\",
+      "t while",
+      "炎 症",
+      "ju n",
+      "j un",
+      "▁cost ly",
+      "th ey",
+      "the y",
+      "t hey",
+      "这道 选择题",
+      "PE G",
+      "P EG",
+      "per malink",
+      "OT E",
+      "O TE",
+      "la ration",
+      "lar ation",
+      "ad apter",
+      "adapt er",
+      "nIt alian",
+      "end ant",
+      "enda nt",
+      "▁ad herence",
+      "▁adher ence",
+      "▁adhere nce",
+      "▁d end",
+      "▁de nd",
+      "▁den d",
+      "IC ATION",
+      "▁T an",
+      "▁Ta n",
+      "▁War ren",
+      "▁am bit",
+      "▁amb it",
+      "不受 申请执行",
+      "feed back",
+      "in dependent",
+      "ind ependent",
+      "准予 原告",
+      "▁Pro blems",
+      "▁Problem s",
+      "度 过",
+      "▁p rogen",
+      "▁pro gen",
+      "▁pr ogen",
+      "▁per mits",
+      "▁permit s",
+      "▁perm its",
+      "▁u ter",
+      "▁ut er",
+      "▁ uter",
+      "du e",
+      "d ue",
+      "nMay be",
+      "n Maybe",
+      "▁m ilit",
+      "▁mil it",
+      "▁mi lit",
+      "le af",
+      "lea f",
+      "▁T GF",
+      "▁TG F",
+      "▁ TGF",
+      "Sh are",
+      "Sha re",
+      "S hare",
+      "▁Pal ace",
+      "亏 损",
+      "▁注 音",
+      "▁F ried",
+      "▁Fr ied",
+      "▁Fri ed",
+      "有 益",
+      "LE ASE",
+      "ok ers",
+      "oke rs",
+      "oker s",
+      "o kers",
+      "▁ass ured",
+      "▁assure d",
+      "制 片",
+      "灰 色",
+      "对自己 的",
+      "对 自己的",
+      "▁Im prove",
+      "▁Imp rove",
+      "ad ays",
+      "ada ys",
+      "aday s",
+      "a days",
+      "▁home ost",
+      "▁S ent",
+      "▁Se nt",
+      "▁Sen t",
+      "▁ Sent",
+      "D ynamic",
+      "▁spectro metry",
+      "▁spectrom etry",
+      "▁relax ing",
+      "仔细 阅读",
+      "▁K le",
+      "▁Kl e",
+      "▁Port uguese",
+      "▁Portug uese",
+      "/\\ '",
+      "/ \\'",
+      "ho oks",
+      "hook s",
+      "h ooks",
+      "着 的",
+      "▁ir rit",
+      "▁irr it",
+      "若 是",
+      "▁dop amine",
+      "na x",
+      "n ax",
+      "▁Camp bell",
+      "fl ush",
+      "flu sh",
+      "▁p enet",
+      "▁pe net",
+      "▁pen et",
+      "nJ an",
+      "n Jan",
+      "▁L C",
+      "▁ LC",
+      "▁n ested",
+      "▁ne sted",
+      "▁nest ed",
+      "▁ nested",
+      "Ex pected",
+      "Exp ected",
+      "Expect ed",
+      "Man agement",
+      "Manag ement",
+      "Manage ment",
+      "M anagement",
+      "▁vent ilation",
+      "▁ventil ation",
+      "▁u g",
+      "▁ ug",
+      "▁b out",
+      "▁bo ut",
+      "▁bou t",
+      "sup ported",
+      "support ed",
+      "▁energ ies",
+      "建 军",
+      "▁a queous",
+      "▁aqu eous",
+      "亲 子",
+      "越 高",
+      "OR E",
+      "O RE",
+      "post gres",
+      "M Y",
+      "核 实",
+      "失去 了",
+      "失 去了",
+      "▁n aked",
+      "][ '",
+      "] ['",
+      "行 包含",
+      "hav iour",
+      "▁C e",
+      "▁ Ce",
+      "的 交",
+      "ar ct",
+      "arc t",
+      "mat ched",
+      "match ed",
+      "m atched",
+      "▁c rystals",
+      "▁cryst als",
+      "▁crystal s",
+      "}} (",
+      "} }(",
+      "un known",
+      "unk nown",
+      "▁W eight",
+      "▁We ight",
+      "▁Wei ght",
+      "▁ Weight",
+      "这 对",
+      "War ning",
+      "W arning",
+      "▁re jection",
+      "▁reject ion",
+      "create Element",
+      "▁re sil",
+      "▁res il",
+      "▁b logs",
+      "▁bl ogs",
+      "▁blog s",
+      "▁blo gs",
+      "▁ blogs",
+      "▁s helf",
+      "▁sh elf",
+      "▁she lf",
+      "▁shel f",
+      "nS T",
+      "n ST",
+      "缓 慢",
+      "xim ity",
+      "▁r anged",
+      "▁range d",
+      "▁ran ged",
+      "佩 戴",
+      "op ol",
+      "opo l",
+      "o pol",
+      "en es",
+      "ene s",
+      "e nes",
+      "▁g aining",
+      "▁gain ing",
+      "▁ga ining",
+      "▁p ics",
+      "▁pi cs",
+      "▁pic s",
+      "▁ pics",
+      "▁predict ors",
+      "▁predictor s",
+      "urope an",
+      "滨 海",
+      "银 川",
+      "独 自",
+      "▁K u",
+      "▁o u",
+      "▁ ou",
+      "▁S TE",
+      "▁ST E",
+      "▁ STE",
+      "▁S alt",
+      "▁Sal t",
+      "▁Sa lt",
+      "ar ians",
+      "ari ans",
+      "aria ns",
+      "arian s",
+      "a rians",
+      "Cl uster",
+      "as sy",
+      "ass y",
+      "摆 脱",
+      "ra in",
+      "r ain",
+      "▁sh ipped",
+      "▁ship ped",
+      "总 量",
+      "ma ll",
+      "mal l",
+      "m all",
+      "▁im ports",
+      "▁import s",
+      "▁imp orts",
+      "▁ imports",
+      "▁an chor",
+      "▁anc hor",
+      "▁anch or",
+      "▁ anchor",
+      "▁Home work",
+      "▁Hom ework",
+      "op edia",
+      "oped ia",
+      "▁hom ogeneous",
+      "▁homogen eous",
+      "▁ homogeneous",
+      "走 的",
+      "yl er",
+      "yle r",
+      "y ler",
+      "ach ers",
+      "ache rs",
+      "acher s",
+      "花 的",
+      "cl ipse",
+      "clip se",
+      "▁D omain",
+      "▁Do main",
+      "▁Dom ain",
+      "▁ Domain",
+      "J P",
+      "本裁定书送达后 立即生效",
+      "▁accompan ying",
+      "▁accompany ing",
+      "di e",
+      "d ie",
+      "判决执行 以前先行羁押的",
+      "负 面",
+      "▁F al",
+      "▁Fa l",
+      "▁E ntry",
+      "▁En try",
+      "▁Ent ry",
+      "▁ Entry",
+      "▁free ly",
+      "▁fre ely",
+      "▁freel y",
+      "∈ [",
+      "递 减",
+      "Ar rays",
+      "Array s",
+      "Arr ays",
+      "img ur",
+      "前 锋",
+      "▁ex pend",
+      "▁exp end",
+      "推 行",
+      "Ex ists",
+      "Exist s",
+      "改造 积极分子",
+      "的 销售",
+      "分 母",
+      "若干问题 的意见",
+      "若干问题的 意见",
+      "▁expect ation",
+      "▁ex isted",
+      "▁exist ed",
+      "▁scient ist",
+      "▁Document ation",
+      "▁ Documentation",
+      "延 迟",
+      "中 秋",
+      "▁key words",
+      "▁keyword s",
+      "▁ keywords",
+      "nS er",
+      "nSe r",
+      "n Ser",
+      "智 商",
+      "▁satisf ies",
+      "relim inary",
+      "▁sil icon",
+      "战 胜",
+      "am ate",
+      "ama te",
+      "a mate",
+      "aut hent",
+      "auth ent",
+      "▁co oked",
+      "▁cook ed",
+      "ty ped",
+      "type d",
+      "typ ed",
+      "t yped",
+      "▁C arbon",
+      "▁Car bon",
+      "▁ Carbon",
+      "▁Voiv odeship",
+      "她 在",
+      "▁S EC",
+      "▁SE C",
+      "▁ SEC",
+      "▁F ant",
+      "▁Fa nt",
+      "▁Fan t",
+      "▁is olate",
+      "▁isol ate",
+      "▁iso late",
+      "各 有",
+      "艺 人",
+      "企业 家",
+      "▁R ud",
+      "▁Ru d",
+      "缓刑 考验",
+      "wi re",
+      "w ire",
+      "fe n",
+      "f en",
+      "电 站",
+      "▁conn ects",
+      "▁connect s",
+      "▁% (",
+      "▁ %(",
+      "的一 项",
+      "的 一项",
+      "▁D av",
+      "▁Da v",
+      "▁T s",
+      "▁ Ts",
+      "功能 的",
+      "▁S qu",
+      "▁ Squ",
+      "att ach",
+      "atta ch",
+      "▁tub erculosis",
+      "蝴 蝶",
+      "Text Field",
+      "▁inter sect",
+      "进行了 审查",
+      "借款 合同纠纷一案中",
+      "借款合同纠纷一案 中",
+      "借款合同 纠纷一案中",
+      "avit ational",
+      "得 多",
+      "Author ization",
+      "刊 物",
+      "所 作",
+      "if fs",
+      "iff s",
+      "奖 惩",
+      "专注 于",
+      "▁sympt om",
+      "rad io",
+      "肠 道",
+      "备 考",
+      "ta t",
+      "t at",
+      "se ver",
+      "sev er",
+      "s ever",
+      "罪犯 评审鉴定表",
+      "▁heter ogeneous",
+      "▁dis rupt",
+      "D H",
+      "cell s",
+      "cel ls",
+      "c ells",
+      "▁D ynamic",
+      "▁Dynam ic",
+      "▁ Dynamic",
+      "▁vibr ant",
+      "▁vib rant",
+      "Error s",
+      "Err ors",
+      "Er rors",
+      "> #",
+      "沉 默",
+      "▁sh ore",
+      "▁ shore",
+      "ast ics",
+      "astic s",
+      "EN S",
+      "E NS",
+      "ba y",
+      "b ay",
+      "dr ive",
+      "driv e",
+      "d rive",
+      "▁W oman",
+      "▁Wo man",
+      "▁Com mons",
+      "▁Common s",
+      "▁Comm ons",
+      "是 为",
+      "▁S ter",
+      "▁St er",
+      "▁Ste r",
+      "IC K",
+      "I CK",
+      "MA C",
+      "M AC",
+      "▁pos session",
+      "▁poss ession",
+      "▁possess ion",
+      "优 于",
+      "不可 以",
+      "不 可以",
+      "▁n ons",
+      "▁no ns",
+      "▁non s",
+      "▁ub iquit",
+      "中 级",
+      "\"} ,\\",
+      "\"}, \\",
+      "\" },\\",
+      "andid ate",
+      "andi date",
+      "om o",
+      "o mo",
+      "ou rt",
+      "our t",
+      "o urt",
+      "ot omy",
+      "oto my",
+      "▁prom ised",
+      "▁promise d",
+      "om ers",
+      "ome rs",
+      "omer s",
+      "o mers",
+      "▁des per",
+      "▁Ig G",
+      "境 界",
+      "rat io",
+      "r atio",
+      "▁wall paper",
+      "不 合",
+      "▁IN CLUDING",
+      "▁ INCLUDING",
+      "▁aff airs",
+      "▁affair s",
+      "▁rect angular",
+      "混 乱",
+      "看 作",
+      "▁com prises",
+      "▁compr ises",
+      "▁comprise s",
+      "▁d ess",
+      "▁de ss",
+      "▁des s",
+      "▁antioxid ant",
+      "▁U krainian",
+      "▁Ukr ainian",
+      "受到 了",
+      "受 到了",
+      "▁sl ots",
+      "▁slot s",
+      "▁ slots",
+      "周 岁",
+      "pi t",
+      "p it",
+      "▁press ures",
+      "▁pressure s",
+      "▁c hop",
+      "▁ch op",
+      "▁cho p",
+      "▁research er",
+      "▁resear cher",
+      "nR ed",
+      "nRe d",
+      "n Red",
+      "gen ase",
+      "▁resid ue",
+      "▁操作 指南",
+      "▁d irty",
+      "▁dir ty",
+      "▁dirt y",
+      "▁ dirty",
+      "∞ )",
+      "侵 害",
+      "lo ve",
+      "lov e",
+      "l ove",
+      "▁pro bes",
+      "▁prob es",
+      "▁probe s",
+      "Y O",
+      "刷 新",
+      "▁S ets",
+      "▁Se ts",
+      "▁Set s",
+      "▁ Sets",
+      "▁end emic",
+      "▁ende mic",
+      "▁F el",
+      "▁Fe l",
+      "平 县",
+      "济南 市",
+      "济 南市",
+      "一 阵",
+      "标 本",
+      "▁c s",
+      "▁ cs",
+      "cut s",
+      "cu ts",
+      "c uts",
+      "▁Athlet ics",
+      "▁Athletic s",
+      "暑 假",
+      "▁N ASA",
+      "▁NA SA",
+      "▁NAS A",
+      "▁L E",
+      "▁ LE",
+      "▁d ashboard",
+      "▁dash board",
+      "▁ dashboard",
+      "支 援",
+      "al ph",
+      "abul ary",
+      "全面 的",
+      "全 面的",
+      "▁re ceipt",
+      "▁rece ipt",
+      "▁Trans fer",
+      "▁ Transfer",
+      "▁elim ination",
+      "j l",
+      "apt ist",
+      "f m",
+      "口 语",
+      "和 高",
+      "▁beautiful ly",
+      "U X",
+      "ic ensing",
+      "icens ing",
+      "▁log istic",
+      "计 生",
+      "各 部门",
+      "我不 知道",
+      "我 不知道",
+      "▁r am",
+      "▁ra m",
+      "▁ ram",
+      "CA S",
+      "C AS",
+      "EM ENT",
+      "E MENT",
+      "木 材",
+      "▁ 清",
+      "▁un ve",
+      "▁w hit",
+      "▁wh it",
+      "▁ whit",
+      "白 菜",
+      "mon g",
+      "mo ng",
+      "m ong",
+      "▁tr ucks",
+      "▁truck s",
+      "第一百三 十三条",
+      "所 长",
+      "辨 认",
+      "▁{ },\\",
+      "▁{} ,\\",
+      "▁{}, \\",
+      "▁lo ops",
+      "▁loop s",
+      "Pr imary",
+      "Prim ary",
+      "在 意",
+      "RE QUEST",
+      "▁relax ed",
+      "▁in clusive",
+      "▁incl usive",
+      "纳入 失信被执行人名单",
+      "赌 博",
+      "▁demand ing",
+      "到期 后",
+      "▁G ene",
+      "▁Ge ne",
+      "▁Gen e",
+      "▁ Gene",
+      "▁sp acious",
+      "▁spac ious",
+      "等 的",
+      "▁E le",
+      "▁El e",
+      "▁ Ele",
+      "可爱 的",
+      "可 爱的",
+      "Point er",
+      "Po inter",
+      "P ointer",
+      "线 程",
+      "▁o re",
+      "▁or e",
+      "▁ ore",
+      "你 还",
+      "婆 婆",
+      "PR E",
+      "P RE",
+      "如实供述 自己的",
+      "如实供述自己 的",
+      "的 方程",
+      "支付 原告",
+      "cast le",
+      "cas tle",
+      "门 店",
+      "一点 点",
+      "一 点点",
+      "▁ 界",
+      "黑 白",
+      "▁$ {{",
+      "▁${ {",
+      "▁M AP",
+      "▁MA P",
+      "▁ MAP",
+      "长沙 市",
+      "Ign ore",
+      "▁irrad iation",
+      "uc lide",
+      "ucl ide",
+      "uclid e",
+      "co ok",
+      "c ook",
+      "▁bank rupt",
+      "un ge",
+      "ung e",
+      "u nge",
+      "开 学",
+      "PO S",
+      "P OS",
+      "hip s",
+      "hi ps",
+      "h ips",
+      "▁p riz",
+      "▁pr iz",
+      "▁pri z",
+      "se en",
+      "see n",
+      "s een",
+      "▁b lo",
+      "▁bl o",
+      "▁ blo",
+      "咨询 服务",
+      "he ap",
+      "hea p",
+      "▁govern ance",
+      "华 夏",
+      "form atics",
+      "format ics",
+      "▁T hai",
+      "▁Th ai",
+      "as sembly",
+      "▁ 样例",
+      "章 节",
+      "ur on",
+      "uro n",
+      "u ron",
+      "▁C ox",
+      "▁Co x",
+      "he ro",
+      "her o",
+      "h ero",
+      "▁Word Press",
+      "弥 补",
+      "▁acc red",
+      "▁ac cred",
+      "▁accr ed",
+      "▁im plants",
+      "▁implant s",
+      "▁impl ants",
+      "每一 位",
+      "每 一位",
+      "▁del ays",
+      "▁delay s",
+      "nN ame",
+      "n Name",
+      "▁determin ants",
+      "▁determinant s",
+      "▁ann ounc",
+      "in da",
+      "ind a",
+      "i nda",
+      "倒 是",
+      "▁need le",
+      "▁th rowing",
+      "▁throw ing",
+      "▁thr owing",
+      "▁thro wing",
+      "i q",
+      "发 光",
+      "的 时代",
+      "▁n eat",
+      "▁ne at",
+      "ot id",
+      "oti d",
+      "o tid",
+      "▁cou pon",
+      "▁coup on",
+      "rp c",
+      "r pc",
+      "▁分 类",
+      "▁ 分类",
+      "过程 中的",
+      "过程中 的",
+      "su ite",
+      "s uite",
+      "Lo op",
+      "L oop",
+      "投资 有限公司",
+      "） [",
+      "re size",
+      "res ize",
+      "▁Every one",
+      "al ib",
+      "ali b",
+      "a lib",
+      "东 区",
+      "AM L",
+      "A ML",
+      "的水 景",
+      "iv ic",
+      "ivi c",
+      "i vic",
+      "▁H it",
+      "▁Hi t",
+      "▁ Hit",
+      "导致 的",
+      "导 致的",
+      "▁A ld",
+      "▁Al d",
+      "sc ss",
+      "s css",
+      "▁s eller",
+      "▁se ller",
+      "▁sell er",
+      "▁sel ler",
+      "▁ seller",
+      "▁F err",
+      "▁Fe rr",
+      "▁Fer r",
+      "▁Be low",
+      "▁Bel ow",
+      "in formation",
+      "inform ation",
+      ") _",
+      "▁Co oper",
+      "均 可",
+      "▁b ones",
+      "▁bo nes",
+      "▁bone s",
+      "▁bon es",
+      "▁J e",
+      "市 级",
+      "▁Publish ing",
+      "\\\\ %",
+      "货 车",
+      "nB ACKGROUND",
+      "し た",
+      "突 发",
+      "提交 的",
+      "观 光",
+      "ene ath",
+      "e neath",
+      "▁re arr",
+      "▁rear r",
+      "nSt aff",
+      "nSta ff",
+      "H H",
+      "一 杯",
+      "既 有",
+      "▁phosph ate",
+      "▁phosphat e",
+      "▁ phosphate",
+      "ou nters",
+      "ount ers",
+      "ounter s",
+      "oun ters",
+      "I K",
+      "有人 说",
+      "有 人说",
+      "▁Car lo",
+      "▁Carl o",
+      "num bers",
+      "number s",
+      "n umbers",
+      "▁de terior",
+      "▁det erior",
+      "▁deter ior",
+      "关 村",
+      "小学 生",
+      "小 学生",
+      "▁o live",
+      "▁ol ive",
+      "▁reg ulates",
+      "▁regul ates",
+      "▁regulate s",
+      "SH A",
+      "S HA",
+      "▁Lib eral",
+      "▁overwhel ming",
+      "▁prec ious",
+      "▁vacc ination",
+      "▁vaccin ation",
+      "安 慰",
+      "水 质",
+      "▁pain ful",
+      "就行 了",
+      "就 行了",
+      "不 应",
+      "扮演 角色",
+      "▁r and",
+      "▁ra nd",
+      "▁ran d",
+      "▁ rand",
+      "之 道",
+      "au er",
+      "a uer",
+      "▁\" \\'",
+      "▁\"\\ '",
+      "▁ \"\\'",
+      "▁depart ure",
+      "▁e arning",
+      "▁ear ning",
+      "▁earn ing",
+      "▁ earning",
+      "▁insert ion",
+      "▁aff iliated",
+      "▁affil iated",
+      "▁affiliate d",
+      "▁is t",
+      "▁i st",
+      "▁ ist",
+      "subset eq",
+      "IS ING",
+      "nG roup",
+      "nGr oup",
+      "n Group",
+      "或 其他",
+      "B i",
+      "su do",
+      "s udo",
+      "缺席 判决",
+      "il les",
+      "ill es",
+      "ille s",
+      "最近 的",
+      "▁inhabit ants",
+      "党组 织",
+      "党 组织",
+      "am pa",
+      "amp a",
+      "▁op era",
+      "▁oper a",
+      "Null able",
+      "▁C ounter",
+      "▁Count er",
+      "▁Co unter",
+      "▁Cou nter",
+      "▁ Counter",
+      "▁Min or",
+      "▁Mi nor",
+      "▁Know ledge",
+      "▁conver ges",
+      "▁converge s",
+      "一个 数",
+      "一 个数",
+      "奇 迹",
+      "有 球队",
+      "▁show case",
+      "▁showc ase",
+      "ur red",
+      "urre d",
+      "init ialize",
+      "initial ize",
+      "▁B A",
+      "▁ BA",
+      "urrenc ies",
+      "ic ul",
+      "at tery",
+      "att ery",
+      "atter y",
+      "Oper ator",
+      "Op erator",
+      "▁Second ary",
+      "食品 安全",
+      "葡萄 酒",
+      "与此 同时",
+      "ia sis",
+      "ias is",
+      "i asis",
+      "▁T am",
+      "▁Ta m",
+      "▁Fire fox",
+      "▁Particip ants",
+      "▁S uite",
+      "▁Su ite",
+      "▁Suit e",
+      "▁ Suite",
+      "▁Municip ality",
+      "▁Municipal ity",
+      "环境 的",
+      "ct rine",
+      "ctr ine",
+      "▁circ ulating",
+      "al is",
+      "ali s",
+      "a lis",
+      "▁manif old",
+      "▁b ridges",
+      "▁bridge s",
+      "▁brid ges",
+      "▁prepar ations",
+      "▁preparation s",
+      "初始 化",
+      "▁ 通",
+      "▁qual ifying",
+      "▁qualify ing",
+      "▁Rob inson",
+      "▁Robin son",
+      "好 事",
+      "▁J ews",
+      "▁Jew s",
+      "▁Je ws",
+      "▁p ale",
+      "▁pal e",
+      "▁pa le",
+      "▁Con stant",
+      "▁Const ant",
+      "▁ Constant",
+      "CO UNT",
+      "▁crit ics",
+      "▁critic s",
+      "—— \\",
+      "— —\\",
+      "信 誉",
+      "▁no body",
+      "▁nob ody",
+      "▁ 职务",
+      "▁b ump",
+      "▁bu mp",
+      "▁H OLD",
+      "▁HO LD",
+      "▁ HOLD",
+      "N B",
+      "减 小",
+      "过去 的",
+      "过 去的",
+      "▁A lt",
+      "▁Al t",
+      "▁ Alt",
+      "▁abund ant",
+      "看 好",
+      "En able",
+      "E nable",
+      "你的 遭遇",
+      "▁IN TO",
+      "▁INT O",
+      "nl og",
+      "n log",
+      "有 过",
+      "Az ure",
+      "A zure",
+      "▁w eld",
+      "▁we ld",
+      "▁wel d",
+      "▁S core",
+      "▁Sc ore",
+      "▁ Score",
+      "ue nt",
+      "uen t",
+      "u ent",
+      "县 级",
+      "阳 区",
+      "判决 确定",
+      "的值 为",
+      "的 值为",
+      "▁Circ uit",
+      "饱 和",
+      "ver tical",
+      "vert ical",
+      "排 行",
+      "▁h un",
+      "▁hu n",
+      "▁ hun",
+      "另一 种",
+      "另 一种",
+      "Int ent",
+      "▁Writ ing",
+      "▁Wr iting",
+      "RE NT",
+      "REN T",
+      "R ENT",
+      "分 式",
+      "▁Att ribute",
+      "▁ Attribute",
+      "▁so ap",
+      "▁E SP",
+      "▁ES P",
+      "▁ ESP",
+      "▁the ta",
+      "▁th eta",
+      "▁ theta",
+      "▁t t",
+      "▁ tt",
+      "▁A UTH",
+      "▁AU TH",
+      "▁ AUTH",
+      "▁F eed",
+      "▁Fe ed",
+      "▁Fee d",
+      "▁ Feed",
+      "▁v ig",
+      "▁vi g",
+      "▁z oom",
+      "▁zoo m",
+      "▁ zoom",
+      "▁m inds",
+      "▁min ds",
+      "▁mind s",
+      "▁H ell",
+      "▁He ll",
+      "▁Hel l",
+      "▁Way ne",
+      "▁c DNA",
+      "She et",
+      "S heet",
+      "▁des ert",
+      "On line",
+      "O nline",
+      "▁c atalytic",
+      "▁catal ytic",
+      "▁pot entials",
+      "▁potential s",
+      "▁potent ials",
+      "· ·",
+      "ST ART",
+      "STA RT",
+      "STAR T",
+      "▁discipl ines",
+      "▁discipline s",
+      "sp aces",
+      "space s",
+      "s paces",
+      "喜 剧",
+      "▁ 梁",
+      "et ric",
+      "etr ic",
+      "▁ex ponent",
+      "老 百姓",
+      "▁ven ous",
+      "F uture",
+      "▁Mean while",
+      "赤 峰",
+      "br aska",
+      "bra ska",
+      "bras ka",
+      "小区 的水景",
+      "届 全国",
+      "Li ke",
+      "L ike",
+      "多少 小区的水景",
+      "酒 后",
+      "是 如何",
+      "▁m L",
+      "▁ mL",
+      "一个 人的",
+      "一个人 的",
+      "一 个人的",
+      "不违反 法律规定",
+      "不 违反法律规定",
+      "▁ather os",
+      "▁E c",
+      "▁orth ogonal",
+      "E v",
+      "WI SE",
+      "▁review ing",
+      "▁s pecially",
+      "▁spec ially",
+      "▁special ly",
+      "▁ specially",
+      "▁K night",
+      "▁Kn ight",
+      "播 出",
+      "▁PROVID ED",
+      "l b",
+      "年 内",
+      "战 役",
+      "速 率",
+      "我已 经",
+      "我 已经",
+      "数据 显示",
+      "▁g er",
+      "▁ge r",
+      "▁ ger",
+      "▁ap artments",
+      "▁apart ments",
+      "▁apartment s",
+      "与 发展",
+      "nF inal",
+      "nFin al",
+      "n Final",
+      "太多 了",
+      "太 多了",
+      "▁l ob",
+      "▁lo b",
+      "▁ lob",
+      "额 定",
+      "an imation",
+      "anim ation",
+      "学 费",
+      "up tools",
+      "upt ools",
+      "A h",
+      "知识 点",
+      "▁sp reading",
+      "▁spread ing",
+      "▁Str ateg",
+      "▁Strat eg",
+      "▁Z ero",
+      "▁Ze ro",
+      "▁ Zero",
+      "去 找",
+      "葫 芦",
+      "因 其",
+      "苏州 市",
+      "苏 州市",
+      "▁dis ruption",
+      "▁disrupt ion",
+      "一 篇",
+      "\"/ >",
+      "\" />",
+      "▁' '\\",
+      "▁'' \\",
+      "nav igation",
+      "nc on",
+      "nco n",
+      "n con",
+      "rib utor",
+      "ribut or",
+      "▁disappe ar",
+      "▁e nde",
+      "▁en de",
+      "▁end e",
+      "▁ ende",
+      "all owed",
+      "allow ed",
+      "allo wed",
+      "真 题",
+      "▁rad ial",
+      "▁ra dial",
+      "▁radi al",
+      "▁l amp",
+      "▁la mp",
+      "▁lam p",
+      "▁in cons",
+      "▁inc ons",
+      "▁incon s",
+      "con stant",
+      "const ant",
+      "▁Pass word",
+      "▁ Password",
+      "t ikz",
+      "副 院长",
+      "履行 完毕",
+      "▁Eval uation",
+      "▁ Evaluation",
+      "▁# \\",
+      "▁ #\\",
+      "▁be aches",
+      "▁beach es",
+      "单调 性",
+      "第五百一十九 条的规定",
+      "第五百一十九条 的规定",
+      "▁Mor ris",
+      "▁Morr is",
+      "▁de note",
+      "▁den ote",
+      "小数 点",
+      "离 不开",
+      "▁P ull",
+      "▁Pu ll",
+      "▁Pul l",
+      "▁ Pull",
+      "▁special ty",
+      "水 晶",
+      "▁micro bi",
+      "an nah",
+      "ann ah",
+      "anna h",
+      "▁preval ent",
+      "▁prev alent",
+      "▁mat hematic",
+      "▁math ematic",
+      "▁d ub",
+      "▁du b",
+      "▁ dub",
+      "▁Never theless",
+      "婚 前",
+      "尔 夫",
+      "▁D raft",
+      "的 食物",
+      "joy s",
+      "jo ys",
+      "j oys",
+      "cur l",
+      "cu rl",
+      "c url",
+      "经纪 人",
+      "自 定义",
+      "汉语 词语",
+      "之 日",
+      "较 小",
+      "贩卖 毒品罪",
+      "终 止",
+      "xy z",
+      "x yz",
+      "appro priate",
+      "侧 面",
+      "建 模",
+      "▁m apped",
+      "▁map ped",
+      "▁ma pped",
+      "▁ mapped",
+      "▁L anc",
+      "▁La nc",
+      "▁Lan c",
+      "id y",
+      "i dy",
+      "▁s ynth",
+      "▁sy nth",
+      "▁synt h",
+      "▁syn th",
+      "放 射",
+      "线 条",
+      "nW ould",
+      "n Would",
+      "ha rm",
+      "har m",
+      "h arm",
+      "uls ion",
+      "SU CCESS",
+      "杆 菌",
+      "M N",
+      "士 兵",
+      "es ium",
+      "esi um",
+      "机 身",
+      "不 爱",
+      "或 打车",
+      "我没 有",
+      "我 没有",
+      "玩 法",
+      "Ph oto",
+      "Phot o",
+      "Mu t",
+      "M ut",
+      "▁ 扮演角色",
+      "▁cru ise",
+      "必 需",
+      "无 所",
+      "长 为",
+      "期 满",
+      "甲 醛",
+      "菜 品",
+      "▁S urface",
+      "▁Sur face",
+      "▁Surf ace",
+      "▁ Surface",
+      "class method",
+      "ts et",
+      "t set",
+      "港 市",
+      "▁l emon",
+      "▁le mon",
+      "▁melan oma",
+      "十 章",
+      "▁e ch",
+      "▁ec h",
+      "▁ ech",
+      "M achine",
+      "To day",
+      "T oday",
+      "▁d irt",
+      "▁di rt",
+      "▁dir t",
+      "下 滑",
+      "prior ity",
+      "p riority",
+      "abs olute",
+      "年 被",
+      "底 部",
+      "路 段",
+      "▁p ussy",
+      "op us",
+      "o pus",
+      "▁ 当",
+      "ba ge",
+      "bag e",
+      "b age",
+      "Si nce",
+      "Sin ce",
+      "S ince",
+      "▁S ynt",
+      "▁Sy nt",
+      "▁Syn t",
+      "▁Dist ribution",
+      "▁ Distribution",
+      "显示 屏",
+      "ac hi",
+      "ach i",
+      "a chi",
+      "▁nice ly",
+      "▁nic ely",
+      "▁ 你",
+      "十 万",
+      "▁psych ology",
+      "▁psycho logy",
+      "▁inst ability",
+      "br ead",
+      "bre ad",
+      "b read",
+      "资源 的",
+      "▁form ally",
+      "▁formal ly",
+      "▁instrument al",
+      "mail to",
+      "犯 故意伤害罪",
+      "正 弦",
+      "▁o uts",
+      "▁out s",
+      "▁ou ts",
+      "▁ outs",
+      "▁trans formations",
+      "▁transform ations",
+      "▁transformation s",
+      "▁a sh",
+      "▁as h",
+      "▁ ash",
+      "▁t rails",
+      "▁tr ails",
+      "▁tra ils",
+      "▁trail s",
+      "▁tun ed",
+      "▁tune d",
+      "▁tu ned",
+      "▁c omed",
+      "▁com ed",
+      "▁co med",
+      "▁come d",
+      "▁ comed",
+      "▁ple asant",
+      "▁pleas ant",
+      "▁ pleasant",
+      "授 课",
+      "漏 洞",
+      "这 块",
+      "途 中",
+      "in et",
+      "ine t",
+      "i net",
+      "忽 然",
+      "密 集",
+      "um i",
+      "u mi",
+      "没 错",
+      "退 房",
+      "▁d type",
+      "▁dt ype",
+      "▁ dtype",
+      "繁 华",
+      "' ;",
+      "▁' <",
+      "▁cont acted",
+      "▁contact ed",
+      "tem perature",
+      "nT ime",
+      "nTim e",
+      "n Time",
+      "▁s parse",
+      "▁sp arse",
+      "▁spars e",
+      "▁ sparse",
+      "▁dis cre",
+      "▁disc re",
+      "门 窗",
+      ")\" ,",
+      ") \",",
+      "t C",
+      "经 贸",
+      "的 表达",
+      "re vision",
+      "rev ision",
+      "L ambda",
+      "▁Char lotte",
+      "不 让",
+      "活 泼",
+      "▁in tend",
+      "▁int end",
+      "ts td",
+      "t std",
+      "ar ness",
+      "arn ess",
+      "▁mod est",
+      "▁mode st",
+      "▁mo dest",
+      "▁modes t",
+      "大 火",
+      "▁N urs",
+      "▁Nur s",
+      "▁Nu rs",
+      "▁g i",
+      "▁ gi",
+      "Pub lish",
+      "▁s ettle",
+      "▁se ttle",
+      "▁set tle",
+      "▁sett le",
+      "ag ar",
+      "aga r",
+      "a gar",
+      "脑 子",
+      "▁P ic",
+      "▁Pi c",
+      "▁ Pic",
+      "▁L etter",
+      "▁Let ter",
+      "▁ Letter",
+      "标 注",
+      "nF ree",
+      "nFr ee",
+      "nFre e",
+      "n Free",
+      "心 肌",
+      "义 乌",
+      "▁S weet",
+      "▁Swe et",
+      "▁run ners",
+      "▁runner s",
+      "太 空",
+      "og raft",
+      "ograf t",
+      "▁Set up",
+      "▁ Setup",
+      "他 还",
+      "虚 假",
+      "总 面积",
+      "▁t oss",
+      "▁to ss",
+      "在 上",
+      "nn ot",
+      "nno t",
+      "n not",
+      "▁首 发",
+      "▁ 首发",
+      "▁Te levision",
+      "▁Tele vision",
+      "没 了",
+      "▁t at",
+      "▁ta t",
+      "▁ tat",
+      "▁pain ters",
+      "▁paint ers",
+      "▁pa inters",
+      "▁painter s",
+      "in ces",
+      "ince s",
+      "inc es",
+      "amil ies",
+      "ami lies",
+      "谁 能",
+      "程度 的",
+      "程 度的",
+      "RNA s",
+      "RN As",
+      "ound ed",
+      "oun ded",
+      "被 动",
+      "除此 之外",
+      "▁sequ el",
+      "▁seq uel",
+      "中华人民共和国 民法",
+      "nE ducation",
+      "te sy",
+      "tes y",
+      "▁aud iences",
+      "▁audience s",
+      "▁audi ences",
+      "NA L",
+      "N AL",
+      "▁L isa",
+      "▁Li sa",
+      "▁Lis a",
+      "▁ 籍贯",
+      "ET HER",
+      "ETH ER",
+      "E THER",
+      "▁suscept ible",
+      "开 工",
+      "▁home less",
+      "▁hom eless",
+      "per or",
+      "pe ror",
+      "ave rage",
+      "aver age",
+      "a verage",
+      "五 条",
+      "抽 象",
+      "本案 中",
+      "理工 大学",
+      "▁re open",
+      "▁war ned",
+      "▁warn ed",
+      "关 爱",
+      "To m",
+      "T om",
+      "per ial",
+      "pe rial",
+      "p erial",
+      "th ink",
+      "thi nk",
+      "thin k",
+      "▁ 原",
+      "着 重",
+      "▁intraven ous",
+      "▁Ther apy",
+      "网 红",
+      "▁l ingu",
+      "▁lin gu",
+      "▁ling u",
+      "▁maint ains",
+      "▁maintain s",
+      "mo ther",
+      "mot her",
+      "m other",
+      "▁le verage",
+      "▁lever age",
+      "万 物",
+      "时期 的",
+      "时 期的",
+      "三 等奖",
+      "女 的",
+      "ng et",
+      "nge t",
+      "n get",
+      "央 视",
+      "▁sent ences",
+      "▁sentence s",
+      "贝 壳",
+      "▁e ager",
+      "fa ct",
+      "fac t",
+      "f act",
+      "Or gan",
+      "Org an",
+      "套 装",
+      "▁u ma",
+      "▁um a",
+      "▁ uma",
+      "不 利",
+      "▁Ev ans",
+      "▁Evan s",
+      "▁Eva ns",
+      "▁Sam uel",
+      "▁S ad",
+      "▁Sa d",
+      "Pl ot",
+      "P lot",
+      "▁F resh",
+      "▁Fr esh",
+      "▁Fre sh",
+      "▁Fres h",
+      "▁bo ats",
+      "▁boat s",
+      "▁curv ature",
+      "背景 下",
+      "▁EX PRESS",
+      "习 题",
+      "▁A gent",
+      "▁Ag ent",
+      "▁Age nt",
+      "▁ Agent",
+      "开 水",
+      "STR ING",
+      "在我 国",
+      "在 我国",
+      "越 好",
+      "at oes",
+      "ato es",
+      "▁E num",
+      "▁En um",
+      "▁ Enum",
+      "▁launch ing",
+      "开 机",
+      "land o",
+      "la ndo",
+      "lan do",
+      "l ando",
+      "▁f ancy",
+      "▁fa ncy",
+      "▁fan cy",
+      "nh ow",
+      "n how",
+      "et ter",
+      "ette r",
+      "ett er",
+      "nM usic",
+      "nMus ic",
+      "n Music",
+      "▁context s",
+      "▁refr iger",
+      "机 床",
+      "ua rt",
+      "u art",
+      "自 助",
+      "▁un t",
+      "▁u nt",
+      "▁ unt",
+      "ob i",
+      "o bi",
+      "▁off line",
+      "▁ offline",
+      "▁p unct",
+      "▁pun ct",
+      "author s",
+      "auth ors",
+      "a uthors",
+      "Exec utor",
+      "▁cor ners",
+      "▁corner s",
+      "▁corn ers",
+      "介 入",
+      "▁hard ly",
+      "咳 嗽",
+      "Y W",
+      "{ '",
+      "沉 淀",
+      "[] ;\\",
+      "[ ];\\",
+      "ns up",
+      "n sup",
+      "一直 是",
+      "ra ul",
+      "r aul",
+      "▁T es",
+      "▁Te s",
+      "▁support ive",
+      "邻 居",
+      "Work er",
+      "pip eline",
+      "pipe line",
+      "p ipeline",
+      "FR OM",
+      "F ROM",
+      "▁in ev",
+      "右 边",
+      "▁L ex",
+      "▁Le x",
+      "▁ Lex",
+      "基 部",
+      "▁neut roph",
+      "▁neutr oph",
+      "Us ing",
+      "U sing",
+      "▁hand ed",
+      "▁han ded",
+      "▁ handed",
+      "Ax is",
+      "en ic",
+      "eni c",
+      "e nic",
+      "crypt o",
+      "c rypto",
+      "▁ б",
+      "全 区",
+      "▁u k",
+      "▁ uk",
+      "▁se aled",
+      "▁sea led",
+      "▁seal ed",
+      ">{ \\",
+      "> {\\",
+      "Pl ace",
+      "▁p sy",
+      "▁ps y",
+      "▁ psy",
+      "▁G rey",
+      "▁Gr ey",
+      "▁Gre y",
+      "dep ends",
+      "depend s",
+      "▁cat heter",
+      "▁cath eter",
+      "▁cathe ter",
+      "▁sequ ential",
+      "得 好",
+      "second ary",
+      "进 取",
+      "女孩 子",
+      "女 孩子",
+      "Mon th",
+      "Mo nth",
+      "Mont h",
+      "integr ation",
+      "我们 会",
+      "▁G enerated",
+      "▁Gener ated",
+      "▁Gen erated",
+      "▁Gene rated",
+      "▁Generate d",
+      "▁ Generated",
+      "▁immun os",
+      "▁immuno s",
+      "▁un ittest",
+      "▁unit test",
+      "▁uni ttest",
+      "▁ unittest",
+      "见 的",
+      "k ubernetes",
+      "▁z u",
+      "▁ zu",
+      "ba g",
+      "b ag",
+      "▁E M",
+      "▁ EM",
+      "Node s",
+      "No des",
+      "N odes",
+      "el ian",
+      "eli an",
+      "elia n",
+      "▁vulner ability",
+      "▁) ,",
+      "▁ ),",
+      "kd ir",
+      "k dir",
+      "í n",
+      "指 向",
+      "▁安 装",
+      "▁ 安装",
+      "在 该",
+      "最 有",
+      "▁ 单",
+      "▁J oint",
+      "▁Jo int",
+      "▁Join t",
+      "▁ Joint",
+      "码 头",
+      "fa re",
+      "far e",
+      "f are",
+      "▁B ond",
+      "▁Bo nd",
+      "▁Bon d",
+      "P ush",
+      "ar us",
+      "aru s",
+      "a rus",
+      "仲裁 委员会",
+      "插 槽",
+      "球 场",
+      "▁O m",
+      "▁not ify",
+      "▁ notify",
+      "In tel",
+      "Int el",
+      "ophag eal",
+      "ophage al",
+      "刚 才",
+      "干 嘛",
+      "Le n",
+      "L en",
+      "il ine",
+      "ili ne",
+      "ilin e",
+      "i line",
+      "论 述",
+      "bu y",
+      "b uy",
+      "现 羁押",
+      "巴 巴",
+      "rib utors",
+      "ribut ors",
+      "ributor s",
+      "Ad v",
+      "A dv",
+      "▁news papers",
+      "▁newsp apers",
+      "▁newspaper s",
+      "▁entertain ing",
+      "▁entert aining",
+      "vi d",
+      "v id",
+      "坚 守",
+      "▁Dr agon",
+      "▁Drag on",
+      "成 型",
+      "▁Sch olar",
+      "生 肖",
+      "▁Program ming",
+      "▁ Programming",
+      "▁h orn",
+      "▁hor n",
+      "▁ho rn",
+      "▁ horn",
+      "▁station ary",
+      "}_{ \\\\",
+      "}_ {\\\\",
+      "} _{\\\\",
+      "▁o bs",
+      "▁ob s",
+      "▁ obs",
+      "▁pull ing",
+      "▁pul ling",
+      "号民事判决书 的本次执行程序",
+      "▁pro jected",
+      "▁project ed",
+      "▁proj ected",
+      "丧 失",
+      "与被告 离婚",
+      "▁Mur ray",
+      "的 无",
+      "不 属于",
+      "ad equ",
+      "ade qu",
+      "小 镇",
+      "▁tw ist",
+      "▁tr unc",
+      "▁ trunc",
+      "▁Am sterdam",
+      "▁D oor",
+      "▁Do or",
+      "技术 服务",
+      "▁prox imal",
+      "▁ 许",
+      "▁s ensit",
+      "▁sens it",
+      "▁sen sit",
+      "▁fo am",
+      "mb ed",
+      "m bed",
+      "pr eter",
+      "pre ter",
+      "pret er",
+      "▁respon ding",
+      "▁respond ing",
+      "▁ responding",
+      "定 量",
+      "▁con clusions",
+      "▁conclusion s",
+      ")} \\\\",
+      ")}\\ \\",
+      ") }\\\\",
+      "▁B rew",
+      "▁Br ew",
+      "▁Bre w",
+      "▁cyl ind",
+      "lo yd",
+      "loy d",
+      "▁inher ited",
+      "▁inherit ed",
+      "行政处罚 决定",
+      "serv let",
+      "▁ 谢",
+      "▁k iller",
+      "▁kil ler",
+      "▁kill er",
+      "▁ki ller",
+      "次 韵",
+      "足够 的",
+      "中的 应用",
+      "中 的应用",
+      "▁L ith",
+      "▁Li th",
+      "▁Lit h",
+      "respons ive",
+      "▁rel atives",
+      "▁relative s",
+      "资 深",
+      "▁N R",
+      "▁ NR",
+      "▁depos its",
+      "▁deposit s",
+      "▁M umbai",
+      "▁Com merce",
+      "▁ Commerce",
+      "▁s erialize",
+      "▁ser ialize",
+      "▁serial ize",
+      "▁ serialize",
+      "▁H em",
+      "▁He m",
+      "em ies",
+      "emi es",
+      "步 伐",
+      "to do",
+      "t odo",
+      "le uk",
+      "▁l ocus",
+      "▁loc us",
+      "▁lo cus",
+      "天 才",
+      "▁O rche",
+      "▁Or che",
+      "Exec ution",
+      "西 兰",
+      "▁Glas gow",
+      "借款 合同",
+      "管理 部门",
+      "你 也",
+      "Co st",
+      "Cos t",
+      "C ost",
+      "▁c ave",
+      "▁ca ve",
+      "▁cav e",
+      "▁qu i",
+      "▁q ui",
+      "又 名",
+      "年 经",
+      "Ins ert",
+      "▁i cons",
+      "▁icon s",
+      "▁ic ons",
+      "▁ icons",
+      "▁he ated",
+      "▁heat ed",
+      "L o",
+      "▁p icks",
+      "▁pick s",
+      "▁pi cks",
+      "▁pic ks",
+      "nS pe",
+      "nSp e",
+      "n Spe",
+      "▁fol lic",
+      "ograph er",
+      "ograp her",
+      "而 出",
+      "▁D iff",
+      "▁Di ff",
+      "▁ Diff",
+      "▁liter acy",
+      "▁\" _",
+      "▁ \"_",
+      "▁ra ises",
+      "▁raise s",
+      "▁ raises",
+      "an ium",
+      "ani um",
+      "▁Ar men",
+      "▁Arm en",
+      "▁in frared",
+      "▁infr ared",
+      "▁ infrared",
+      "EL L",
+      "E LL",
+      "▁DA MAGES",
+      "▁DAMAGE S",
+      "交 给",
+      "Set up",
+      "ed e",
+      "e de",
+      "he ses",
+      "hes es",
+      "N s",
+      "争 吵",
+      "▁M ent",
+      "▁Me nt",
+      "▁Men t",
+      "▁G ay",
+      "▁Ga y",
+      "余 额",
+      "Buf f",
+      "Bu ff",
+      "B uff",
+      "▁ ↑",
+      "][ ]",
+      "] []",
+      "ch en",
+      "che n",
+      "c hen",
+      "行为 的",
+      "科技 创新",
+      "狮 子",
+      "▁cent r",
+      "▁ce ntr",
+      "▁s ums",
+      "▁su ms",
+      "▁sum s",
+      "b m",
+      "no ld",
+      "nol d",
+      "n old",
+      "▁gar lic",
+      "▁over lap",
+      "▁overl ap",
+      "▁d uo",
+      "▁du o",
+      "▁double s",
+      "▁dou bles",
+      "▁doub les",
+      "▁un changed",
+      "▁po em",
+      "▁reg ulator",
+      "▁regul ator",
+      "儿 的",
+      "▁Val idation",
+      "▁Valid ation",
+      "▁ Validation",
+      "行政 诉讼",
+      "▁sufficient ly",
+      "▁f ence",
+      "▁fe nce",
+      "▁fen ce",
+      "锅 中",
+      "司法 鉴定",
+      "▁ar ising",
+      "▁pe aks",
+      "▁peak s",
+      "为 核心",
+      "等方面 的",
+      "等 方面的",
+      "支付 宝",
+      "▁N ova",
+      "▁No va",
+      "▁Nov a",
+      "越来越 多的",
+      "越来越多 的",
+      "卡 片",
+      "间 接",
+      "▁sim ulate",
+      "nCont act",
+      "n Contact",
+      "sp y",
+      "s py",
+      "▁lymph oma",
+      "SE D",
+      "S ED",
+      "bl asts",
+      "blast s",
+      "bla sts",
+      "▁Direct ory",
+      "▁Director y",
+      "▁ Directory",
+      "行 星",
+      "的 城市",
+      "▁en velope",
+      "▁envelop e",
+      "外 卖",
+      "▁ 于",
+      "多 吃",
+      "▁c rate",
+      "▁cr ate",
+      "▁cra te",
+      "▁ crate",
+      "▁Ge ometry",
+      "▁Geo metry",
+      "▁ Geometry",
+      "re z",
+      "r ez",
+      "al chemy",
+      "alc hemy",
+      "▁hydro x",
+      "▁hyd rox",
+      "▁ hydrox",
+      "▁bott les",
+      "▁bottle s",
+      "我 来",
+      "▁bre eding",
+      "▁breed ing",
+      "▁executive s",
+      "▁execut ives",
+      "▁Under standing",
+      "用 手",
+      "是 其",
+      "en ue",
+      "enu e",
+      "am pton",
+      "amp ton",
+      "▁hesit ate",
+      "Form s",
+      "For ms",
+      "F orms",
+      "直 属",
+      ")} $",
+      ") }$",
+      "ic ion",
+      "ici on",
+      "icio n",
+      "▁Econ omics",
+      "▁Economic s",
+      "▁N EW",
+      "▁NE W",
+      "▁ NEW",
+      "期限 届满",
+      "光 滑",
+      "AT AB",
+      "ATA B",
+      "pr ints",
+      "print s",
+      "l m",
+      "▁R and",
+      "▁Ra nd",
+      "▁Ran d",
+      "▁out lined",
+      "▁outline d",
+      "▁comp anion",
+      "紫 外",
+      "▁n ail",
+      "▁na il",
+      "▁avoid ing",
+      "开 门",
+      "▁K rist",
+      "▁Kr ist",
+      "▁Kris t",
+      "▁Con cept",
+      "▁ Concept",
+      "▁L if",
+      "▁Li f",
+      "▁ Lif",
+      "▁Di amond",
+      "Ex press",
+      "Exp ress",
+      "Expr ess",
+      "▁U g",
+      "北 大",
+      "Ac cept",
+      "向本院 申请强制执行",
+      "向本院申请 强制执行",
+      "▁gen etics",
+      "▁genetic s",
+      "▁py test",
+      "▁pyt est",
+      "▁ pytest",
+      "▁im ply",
+      "▁imp ly",
+      "▁impl y",
+      "inter faces",
+      "interface s",
+      "人民 的",
+      "▁N at",
+      "▁Na t",
+      "al ter",
+      "alt er",
+      "path etic",
+      "pat hetic",
+      "Method s",
+      "遭 受",
+      "nM en",
+      "nMe n",
+      "n Men",
+      "▁A lp",
+      "▁Al p",
+      "state ment",
+      "stat ement",
+      "▁kn ife",
+      "考 古",
+      "Ex it",
+      "E xit",
+      "▁C AN",
+      "▁CA N",
+      "▁ CAN",
+      "证 监会",
+      "ss on",
+      "s son",
+      "▁Bl ues",
+      "▁Blue s",
+      "▁Blu es",
+      "nE d",
+      "n Ed",
+      "来 进行",
+      "▁W a",
+      "某某 的",
+      "某 某的",
+      "it ic",
+      "iti c",
+      "i tic",
+      "个 别",
+      "底 层",
+      "sp ark",
+      "s park",
+      "专 栏",
+      "▁m b",
+      "▁ mb",
+      "不断 地",
+      "▁T ed",
+      "▁Te d",
+      "▁con gress",
+      "▁congr ess",
+      "▁cong ress",
+      "吾 尔",
+      "在我 们",
+      "在 我们",
+      "▁cons umed",
+      "▁consum ed",
+      "▁consume d",
+      "首先 要",
+      "Pay ment",
+      "品 的",
+      "推出 的",
+      "推 出的",
+      "ey er",
+      "eye r",
+      "e yer",
+      "▁M ak",
+      "▁Ma k",
+      "▁immun ohist",
+      "▁immuno hist",
+      "替 换",
+      "▁wis dom",
+      "土 木",
+      "过去 了",
+      "过 去了",
+      "▁N ice",
+      "▁Nic e",
+      "▁Ni ce",
+      "▁determ inant",
+      "▁determin ant",
+      "▁rem inder",
+      "▁remind er",
+      "▁remin der",
+      "os er",
+      "ose r",
+      "o ser",
+      "ill us",
+      "▁en joys",
+      "▁enjoy s",
+      "▁auto immune",
+      "▁P P",
+      "▁ PP",
+      "j p",
+      "Exec ute",
+      "绝 望",
+      "感觉 到",
+      "益 智",
+      "▁Ch rom",
+      "▁act in",
+      "▁ac tin",
+      "▁ actin",
+      "仓 储",
+      "▁path ophys",
+      "▁T u",
+      "了 起来",
+      "的分 析",
+      "的 分析",
+      "周 知",
+      "▁I E",
+      "▁ IE",
+      "▁n est",
+      "▁ne st",
+      "▁ nest",
+      "▁Man hattan",
+      "f u",
+      "▁c omo",
+      "▁com o",
+      "▁co mo",
+      "Mi cro",
+      "M icro",
+      "回 国",
+      "▁J O",
+      "▁ JO",
+      "▁con verting",
+      "▁conver ting",
+      "▁convert ing",
+      "上 方",
+      "感 应",
+      "nOut put",
+      "n Output",
+      "每 小时",
+      "▁c anal",
+      "▁can al",
+      "▁ca nal",
+      "▁M aking",
+      "▁Ma king",
+      "▁Mak ing",
+      "▁ Making",
+      "我就 是",
+      "我 就是",
+      "Wi ll",
+      "Wil l",
+      "W ill",
+      "ple ase",
+      "p lease",
+      "Name space",
+      "Names pace",
+      "N amespace",
+      "▁san ction",
+      "皇 家",
+      "Ch ain",
+      "▁cre ator",
+      "▁creat or",
+      "▁ creator",
+      "热 量",
+      "Page s",
+      "Pa ges",
+      "Pag es",
+      "P ages",
+      "▁col ony",
+      "▁colon y",
+      "ca s",
+      "c as",
+      "▁ 种",
+      "▁s ans",
+      "▁sa ns",
+      "▁san s",
+      "▁P G",
+      "▁ PG",
+      "我 又",
+      "tc p",
+      "t cp",
+      "op les",
+      "ople s",
+      "opl es",
+      "o ples",
+      "▁num erator",
+      "▁numer ator",
+      "▁Mor ning",
+      "▁B rig",
+      "▁Br ig",
+      "▁Bri g",
+      "网 游",
+      "有机 会",
+      "有 机会",
+      "极大 的",
+      "极 大的",
+      "gi ven",
+      "give n",
+      "g iven",
+      "▁Y ellow",
+      "▁default s",
+      "▁ defaults",
+      "▁p atches",
+      "▁pat ches",
+      "▁patch es",
+      "▁ patches",
+      "▁Co ach",
+      "▁P OST",
+      "▁PO ST",
+      "▁POS T",
+      "▁ POST",
+      "▁correl ate",
+      ">\" ,",
+      "> \",",
+      "▁Y an",
+      "▁Ya n",
+      "▁Book mark",
+      "▁polar ization",
+      "第一百四十五 条第一款规定",
+      "第一百四十五条第一款 规定",
+      "nc d",
+      "n cd",
+      "ish ops",
+      "ishop s",
+      "i shops",
+      "为 企业",
+      "▁ben ign",
+      "剧 院",
+      "iss ance",
+      "issa nce",
+      "issan ce",
+      "公 证",
+      ":] \\",
+      ": ]\\",
+      "不仅 是",
+      "roc ery",
+      "并给出 你的",
+      "的长 度",
+      "的 长度",
+      "▁not ifications",
+      "▁notification s",
+      "▁ notifications",
+      "it ual",
+      "itu al",
+      "i tual",
+      "▁Hon or",
+      "▁Ho nor",
+      "▁L esser",
+      "▁Less er",
+      "▁Les ser",
+      "▁G etting",
+      "▁Get ting",
+      "▁ Getting",
+      "保 定",
+      "的人 员",
+      "的 人员",
+      "▁Y ang",
+      "▁Yan g",
+      "▁Ya ng",
+      "▁CON TRACT",
+      "▁Char lie",
+      "▁Charl ie",
+      "▁veter ans",
+      "▁veteran s",
+      "▁com plication",
+      "▁compl ication",
+      "人 不",
+      "Mus ic",
+      "M usic",
+      "▁Oper a",
+      "▁Op era",
+      "▁ass ists",
+      "▁assist s",
+      "▁tend ency",
+      "平 静",
+      "it ime",
+      "iti me",
+      "i time",
+      "▁lib eral",
+      "恶 魔",
+      "dens ity",
+      "d ensity",
+      "Z ero",
+      "死 刑",
+      "降 价",
+      "Hash Map",
+      "▁work out",
+      "is ch",
+      "isc h",
+      "i sch",
+      "▁F DA",
+      "▁FD A",
+      "▁ FDA",
+      "号民事判决书 已经发生法律效力",
+      "里 有",
+      "uit ar",
+      "ui tar",
+      "u itar",
+      "噪 声",
+      "方 可",
+      "TO P",
+      "T OP",
+      "借 据",
+      "笑 容",
+      "w y",
+      "控制 器",
+      "加入 中国共产党",
+      "DIR ECT",
+      "DI RECT",
+      "交易 所",
+      "▁autom otive",
+      "▁cal ibration",
+      "▁calibr ation",
+      "ew ay",
+      "e way",
+      "恶 劣",
+      "nU s",
+      "n Us",
+      "汉 典",
+      "注 定",
+      "组织 部",
+      "的 对",
+      "维 吾尔",
+      "安 心",
+      "nCurrent ly",
+      "n Currently",
+      "td ata",
+      "t data",
+      "催 告",
+      "an gan",
+      "ang an",
+      "anga n",
+      "并 通过",
+      "▁experimental ly",
+      "▁experiment ally",
+      "E Q",
+      "q t",
+      "F s",
+      "us a",
+      "u sa",
+      "岁以下 的",
+      "岁 以下的",
+      "nE dit",
+      "nEd it",
+      "n Edit",
+      "申请执行人 于",
+      "PH P",
+      "P HP",
+      "min a",
+      "mi na",
+      "m ina",
+      "困 境",
+      "▁SH ALL",
+      "▁SHA LL",
+      "▁M asters",
+      "▁Ma sters",
+      "▁Master s",
+      "▁Mas ters",
+      "▁Mast ers",
+      "▁prem ature",
+      "浩 特",
+      "ra wl",
+      "raw l",
+      "▁nucle i",
+      "▁redu nd",
+      "▁red und",
+      "▁after wards",
+      "▁afterward s",
+      "▁C old",
+      "▁Col d",
+      "▁Co ld",
+      "▁s ucc",
+      "▁su cc",
+      "▁suc c",
+      "之 时",
+      "▁B row",
+      "▁Br ow",
+      "▁Bro w",
+      "▁ed itors",
+      "▁editor s",
+      "▁edit ors",
+      "л я",
+      "▁B aker",
+      "▁Ba ker",
+      "▁Bak er",
+      "▁B ass",
+      "▁Bas s",
+      "▁Ba ss",
+      "re ason",
+      "reas on",
+      "rea son",
+      "Time stamp",
+      "Tim estamp",
+      "玩 笑",
+      "▁w ise",
+      "▁wis e",
+      "▁wi se",
+      "▁ wise",
+      "业 的",
+      "售后 服务",
+      "▁not ably",
+      "▁K il",
+      "▁Ki l",
+      "▁pharmac ological",
+      "▁pharmacologic al",
+      "和国 家",
+      "和 国家",
+      "PL AY",
+      "PLA Y",
+      "P LAY",
+      "▁nic ot",
+      "▁ni cot",
+      "Di stance",
+      "Dist ance",
+      "D istance",
+      "检 索",
+      "dest ination",
+      "nD ep",
+      "nDe p",
+      "n Dep",
+      "之后 的",
+      "之 后的",
+      "▁in structor",
+      "▁instruct or",
+      "有 无",
+      "▁gener ous",
+      "▁gene rous",
+      "▁d iox",
+      "▁di ox",
+      "月 初",
+      "▁renew able",
+      "▁t rap",
+      "▁tr ap",
+      "▁tra p",
+      "▁re wards",
+      "▁reward s",
+      "调 度",
+      "pr im",
+      "pri m",
+      "p rim",
+      "▁s oy",
+      "▁so y",
+      "PRE FIX",
+      "P REFIX",
+      "D ictionary",
+      "et ta",
+      "ett a",
+      "莫 名",
+      "ot he",
+      "oth e",
+      "o the",
+      "▁T rain",
+      "▁Tr ain",
+      "▁Tra in",
+      "▁ Train",
+      "▁t il",
+      "▁ti l",
+      "▁ til",
+      "af fold",
+      "aff old",
+      "主 板",
+      "W i",
+      "盛 世",
+      "的 方向",
+      "▁M olecular",
+      "▁U nd",
+      "▁Un d",
+      "商贸 有限公司",
+      "index Of",
+      "▁D river",
+      "▁Dr iver",
+      "▁Drive r",
+      "▁ Driver",
+      "▁viol ation",
+      "▁vi olation",
+      "温 和",
+      "产 后",
+      "永 恒",
+      "▁sub project",
+      "ra its",
+      "rait s",
+      "nT H",
+      "n TH",
+      "Off ice",
+      "更 快",
+      "▁per ox",
+      "▁pe rox",
+      "▁qu oted",
+      "▁quote d",
+      "▁quot ed",
+      "▁quo ted",
+      "随 之",
+      "内 科",
+      "zym e",
+      "zy me",
+      "z yme",
+      "▁Project s",
+      "▁ Projects",
+      "恐 怕",
+      "负 债",
+      "奶 茶",
+      "十六 章",
+      "▁P anel",
+      "▁Pan el",
+      "▁Pa nel",
+      "▁ Panel",
+      "给出的 选项中",
+      "▁st rive",
+      "▁str ive",
+      "▁stri ve",
+      "法律文书 确定的义务",
+      "坚 信",
+      "ATA BASE",
+      "ATAB ASE",
+      ":: \\",
+      ": :\\",
+      "化 成",
+      "▁rig id",
+      "▁ri gid",
+      "也很 好",
+      "也 很好",
+      "▁C arter",
+      "▁Car ter",
+      "▁Cart er",
+      "最 少",
+      "if def",
+      "ra b",
+      "r ab",
+      "ount ain",
+      "oun tain",
+      "的人 才",
+      "的 人才",
+      "▁insp iring",
+      "▁inspir ing",
+      "▁format ting",
+      "▁dest inations",
+      "▁destination s",
+      "热 血",
+      "走 在",
+      "en ario",
+      "ena rio",
+      "nN orth",
+      "nNor th",
+      "n North",
+      "东 海",
+      "we ather",
+      "大 胆",
+      "▁arg v",
+      "▁ argv",
+      "▁B rain",
+      "▁Br ain",
+      "▁Bra in",
+      "▁ Brain",
+      "▁neg lig",
+      "脚 步",
+      "▁screen ed",
+      "简 约",
+      "做出 了",
+      "做 出了",
+      "el er",
+      "ele r",
+      "e ler",
+      "▁j er",
+      "▁je r",
+      "▁ jer",
+      "▁own s",
+      "▁ow ns",
+      "▁ owns",
+      "纪 检",
+      "案 外人",
+      "Re po",
+      "Rep o",
+      "感 知",
+      "B atch",
+      "不 容",
+      "▁CO MP",
+      "▁COM P",
+      "▁ COMP",
+      "as ting",
+      "ast ing",
+      "▁Stat istical",
+      "▁carb ohyd",
+      "▁f i",
+      "▁ fi",
+      "ew ater",
+      "e water",
+      "re present",
+      "rep resent",
+      "repre sent",
+      "▁Mar shall",
+      "▁Marsh all",
+      "▁Mars hall",
+      "▁Marshal l",
+      "实 物",
+      "停车 场",
+      "第一百零 二条",
+      "ro ng",
+      "ron g",
+      "r ong",
+      "狗 狗",
+      "▁pot assium",
+      "/ @",
+      "场 比赛",
+      "as ma",
+      "asm a",
+      "nMedal ists",
+      "▁P ER",
+      "▁PE R",
+      "▁ PER",
+      "▁h unt",
+      "▁hun t",
+      "▁hu nt",
+      "▁Ch oice",
+      "▁Cho ice",
+      "▁ Choice",
+      "▁anth rop",
+      "▁att acked",
+      "▁attack ed",
+      "可在 裁定书送达之日起十日内",
+      "力 求",
+      "难 得",
+      "nN or",
+      "nNo r",
+      "n Nor",
+      "银行 卡",
+      "▁H istorical",
+      "▁Histor ical",
+      "▁Historic al",
+      "▁de aler",
+      "▁deal er",
+      "意义 的",
+      "▁R aw",
+      "▁Ra w",
+      "▁ Raw",
+      "▁challeng ed",
+      "▁challenge d",
+      "▁inter acting",
+      "▁interact ing",
+      "X P",
+      "九 年",
+      "缴 费",
+      "互 为",
+      "我会 给你",
+      "▁where ver",
+      "we n",
+      "w en",
+      "▁Ir anian",
+      "▁Iran ian",
+      "ho urs",
+      "hour s",
+      "hou rs",
+      "h ours",
+      "▁sort ing",
+      "▁sor ting",
+      "▁sever ely",
+      "▁severe ly",
+      "鼻 子",
+      "▁prim itive",
+      "因为 他",
+      "因 为他",
+      "▁u id",
+      "▁ui d",
+      "▁ uid",
+      "也 都",
+      "只能 说",
+      "t static",
+      "遇到 了",
+      "遇 到了",
+      "▁h erb",
+      "▁he rb",
+      "▁her b",
+      "▁clin icians",
+      "▁clinic ians",
+      "▁clinician s",
+      "ri se",
+      "ris e",
+      "r ise",
+      "ra h",
+      "r ah",
+      "▁Arab ia",
+      "miss ing",
+      "mis sing",
+      "m issing",
+      "中 考",
+      "所 致",
+      "as onic",
+      "ason ic",
+      "修 建",
+      "file path",
+      "培训 班",
+      "Al ert",
+      "Ale rt",
+      "打 包",
+      "▁Bar bara",
+      "▁Barb ara",
+      "轴 对称",
+      "▁enh ances",
+      "▁enhance s",
+      "历 任",
+      "包 裹",
+      "环 球",
+      "in come",
+      "inc ome",
+      "皇 后",
+      "er set",
+      "ers et",
+      "二〇一 三年",
+      "▁in spect",
+      "▁ins pect",
+      "▁insp ect",
+      "▁ inspect",
+      "很多 的",
+      "很 多的",
+      "ac hy",
+      "ach y",
+      "a chy",
+      "买卖 合同纠纷一案中",
+      "买卖合同纠纷一案 中",
+      "买卖合同 纠纷一案中",
+      "买卖合同纠纷 一案中",
+      "▁Inter esting",
+      "▁Interest ing",
+      "日生 育",
+      "日 生育",
+      "lu a",
+      "l ua",
+      "贵 族",
+      "处分 自己的",
+      "转 为",
+      "▁nec rosis",
+      "要 用",
+      ": $",
+      "抱 怨",
+      "st en",
+      "ste n",
+      "s ten",
+      "▁ro bots",
+      "▁rob ots",
+      "▁robot s",
+      "ri vers",
+      "riv ers",
+      "river s",
+      "rive rs",
+      "r ivers",
+      "▁I ter",
+      "▁It er",
+      "▁ Iter",
+      "真 人",
+      "▁V E",
+      "▁ VE",
+      "交纳 办法",
+      "qu o",
+      "q uo",
+      "科 院",
+      "al led",
+      "all ed",
+      "alle d",
+      "if ice",
+      "ific e",
+      "ifi ce",
+      "▁con dens",
+      "▁cond ens",
+      "▁employ ing",
+      "▁decor ated",
+      "▁decorate d",
+      "最小 的",
+      "最 小的",
+      "▁F at",
+      "▁Fa t",
+      "▁ Fat",
+      "ip es",
+      "ipe s",
+      "i pes",
+      "ft ime",
+      "f time",
+      "▁M agn",
+      "▁Mag n",
+      "▁Ma gn",
+      "▁O H",
+      "▁ OH",
+      "及 以上",
+      "的事实 清楚",
+      "的 事实清楚",
+      "▁Le ban",
+      "▁travel s",
+      "▁tra vels",
+      "▁trav els",
+      "sc he",
+      "sch e",
+      "s che",
+      "L i",
+      "IC ES",
+      "ICE S",
+      "I CES",
+      "▁t ogg",
+      "▁to gg",
+      "▁def inite",
+      "▁definit e",
+      "▁defin ite",
+      "热 闹",
+      "邯 郸",
+      "F etch",
+      "认定 事实",
+      "▁For ces",
+      "▁Force s",
+      "莲 花",
+      "饲 养",
+      "学 金",
+      "实 事",
+      "这 几",
+      "mb ox",
+      "m box",
+      "领导 干部",
+      "ex clude",
+      "芙 蓉",
+      "õ es",
+      "▁accept ing",
+      "你认为 正确的答案",
+      "▁P le",
+      "▁Pl e",
+      "开发 者",
+      "div ision",
+      "di vision",
+      "▁C hemical",
+      "▁Chem ical",
+      "▁T ak",
+      "▁Ta k",
+      "ce land",
+      "cel and",
+      "c eland",
+      "▁tour ists",
+      "▁tourist s",
+      "id ade",
+      "ida de",
+      "idad e",
+      "西 瓜",
+      "us ters",
+      "ust ers",
+      "uster s",
+      "u sters",
+      "▁Unity Engine",
+      "传 动",
+      "mo vie",
+      "mov ie",
+      "m ovie",
+      "▁quant ify",
+      "同类 项",
+      "SO URCE",
+      "par k",
+      "pa rk",
+      "p ark",
+      "▁Roman ia",
+      "▁Ro mania",
+      "▁Rom ania",
+      "▁Roma nia",
+      "▁Arg ument",
+      "▁ Argument",
+      "▁F BI",
+      "▁FB I",
+      "货 运",
+      "ge o",
+      "g eo",
+      "进行 审理",
+      "▁ur gent",
+      "▁urg ent",
+      "▁urge nt",
+      "▁grad uates",
+      "▁gradu ates",
+      "▁graduate s",
+      "红 军",
+      "▁J azz",
+      "▁Ja zz",
+      "▁H S",
+      "▁ HS",
+      "开 创",
+      "显示 器",
+      "▁p ressing",
+      "▁pr essing",
+      "▁pres sing",
+      "▁press ing",
+      "▁qu alities",
+      "▁qual ities",
+      "接下来 的",
+      "接 下来的",
+      "合同履行 地",
+      "▁out let",
+      "▁ou tlet",
+      "▁Pro gress",
+      "▁ Progress",
+      "pt s",
+      "p ts",
+      "os ition",
+      "osit ion",
+      "osi tion",
+      "▁U C",
+      "▁ UC",
+      "▁| ||",
+      "▁|| |",
+      "▁ |||",
+      "一 千",
+      "pro ved",
+      "pr oved",
+      "prov ed",
+      "prove d",
+      "power ed",
+      "pow ered",
+      "ev ens",
+      "even s",
+      "eve ns",
+      "e vens",
+      "da o",
+      "d ao",
+      "Br o",
+      "B ro",
+      "▁ 应用",
+      "loqu ent",
+      "▁prior it",
+      "▁priori t",
+      "▁rig orous",
+      "▁l obby",
+      "▁lob by",
+      "▁P resent",
+      "▁Pres ent",
+      "▁Pre sent",
+      "▁ Present",
+      "▁m icrobial",
+      "▁micro bial",
+      "▁microbi al",
+      "分 区",
+      "▁p ools",
+      "▁po ols",
+      "▁pool s",
+      "▁In terior",
+      "▁Int erior",
+      "▁Inter ior",
+      "长方 形",
+      "长 方形",
+      "出 处",
+      "受 损",
+      "整 形",
+      "经济 开发区",
+      "▁J ustin",
+      "▁Just in",
+      "▁un limited",
+      "add EventListener",
+      "ul ly",
+      "ull y",
+      "ur ia",
+      "uri a",
+      "u ria",
+      "▁g li",
+      "▁gl i",
+      "大 哥",
+      "省 市",
+      "ra c",
+      "r ac",
+      "nPre vious",
+      "n Previous",
+      "▁ É",
+      "FO R",
+      "F OR",
+      "最小 值",
+      "ke l",
+      "k el",
+      "▁evolution ary",
+      "by e",
+      "b ye",
+      "▁L eader",
+      "▁Le ader",
+      "▁Lead er",
+      "▁d istant",
+      "▁dist ant",
+      "▁di stant",
+      "▁manufact ure",
+      "主 播",
+      "gl uc",
+      "g luc",
+      "of fer",
+      "off er",
+      "县 农村信用合作联社",
+      "▁re porters",
+      "▁rep orters",
+      "▁report ers",
+      "▁reporter s",
+      "▁ir regular",
+      "b u",
+      "的 结构",
+      "▁consider ations",
+      "▁consideration s",
+      "▁elect romagnetic",
+      "▁electrom agnetic",
+      "庆 市",
+      "Gr een",
+      "G reen",
+      "▁h ood",
+      "▁ho od",
+      "▁ hood",
+      "▁st raw",
+      "▁str aw",
+      "奶 粉",
+      "ds t",
+      "d st",
+      "阅读 网",
+      "Ro om",
+      "R oom",
+      "▁ 问题",
+      "第一 项",
+      "第 一项",
+      "ito neal",
+      "itone al",
+      "iton eal",
+      "Hi s",
+      "H is",
+      "可持续 发展",
+      "▁ 房型",
+      "▁C arr",
+      "▁Car r",
+      "▁Ca rr",
+      "TH E",
+      "T HE",
+      "▁R V",
+      "▁ RV",
+      "▁ph oton",
+      "▁phot on",
+      "▁photo n",
+      "▁ photon",
+      "elect ron",
+      "electro n",
+      "▁o g",
+      "▁ og",
+      "ts uper",
+      "t super",
+      "DV D",
+      "D VD",
+      "▁Y eah",
+      "▁Ye ah",
+      "体 裁",
+      "an gi",
+      "ang i",
+      "▁cross ed",
+      "▁inv ested",
+      "▁invest ed",
+      "▁amb ient",
+      "J s",
+      "三 百",
+      "▁G eneration",
+      "▁Gener ation",
+      "▁Gen eration",
+      "▁Gene ration",
+      "▁ Generation",
+      "▁N K",
+      "▁ NK",
+      "pt ember",
+      "摄 入",
+      "dj ang",
+      "▁Ter rit",
+      "▁Terr it",
+      "AN K",
+      "A NK",
+      "君 子",
+      "▁v oices",
+      "▁voice s",
+      "▁vo ices",
+      "▁Ex tension",
+      "▁Ext ension",
+      "▁ Extension",
+      "▁aut onomous",
+      "是 的",
+      "cred entials",
+      "静 态",
+      "▁York shire",
+      "Ti le",
+      "T ile",
+      "pos itions",
+      "position s",
+      "姿 势",
+      "th reshold",
+      "nOBJECT IVE",
+      "static method",
+      "Pack et",
+      "Pac ket",
+      "P acket",
+      "NE CTION",
+      "NECT ION",
+      "▁d ock",
+      "▁do ck",
+      "▁doc k",
+      "▁ dock",
+      "倾 斜",
+      "普通 话",
+      "普 通话",
+      "ab spath",
+      "abs path",
+      "▁res erves",
+      "▁reserve s",
+      "▁reserv es",
+      "▁D S",
+      "▁ DS",
+      "公 子",
+      "▁K ate",
+      "▁Kat e",
+      "▁Ka te",
+      "▁Sal v",
+      "▁Sa lv",
+      "金 刚",
+      "um per",
+      "ump er",
+      "▁p ray",
+      "▁pr ay",
+      "▁pra y",
+      "▁ pray",
+      "但 也",
+      "ov irus",
+      "ovi rus",
+      "ovir us",
+      "o virus",
+      "), (",
+      ") ,(",
+      "▁cit izen",
+      "▁citiz en",
+      "▁chlor ide",
+      "该 怎么",
+      "▁sh ield",
+      "▁ shield",
+      "水 市",
+      "equ alities",
+      "equal ities",
+      "▁R iemann",
+      "▁Ri emann",
+      "添 香",
+      "MA IN",
+      "M AIN",
+      "▁Re views",
+      "▁Review s",
+      "▁nucle otide",
+      "look up",
+      "其 间",
+      "▁al ias",
+      "▁ali as",
+      "▁ alias",
+      "ge ne",
+      "gen e",
+      "g ene",
+      "▁T on",
+      "▁To n",
+      "按 时",
+      "▁ 足球",
+      "开 的",
+      "▁r if",
+      "▁ri f",
+      "Pe ople",
+      "P eople",
+      "神 仙",
+      "街道 办事处",
+      "日 照",
+      "To k",
+      "T ok",
+      "▁B egin",
+      "▁Be gin",
+      "▁Beg in",
+      "▁ Begin",
+      "▁P osition",
+      "▁Pos ition",
+      "▁ Position",
+      "▁turn over",
+      "▁tur nover",
+      "▁Str ucture",
+      "▁Struct ure",
+      "▁ Structure",
+      "con current",
+      "co ncurrent",
+      "conc urrent",
+      "▁seem ingly",
+      "王 朝",
+      "ad ows",
+      "ado ws",
+      "adow s",
+      "te chn",
+      "tech n",
+      "tec hn",
+      "co ding",
+      "cod ing",
+      "c oding",
+      "▁c ancelled",
+      "▁cancel led",
+      "▁cancell ed",
+      "nC reate",
+      "n Create",
+      "▁nut rients",
+      "▁nutr ients",
+      "▁nutrient s",
+      "华 东",
+      "保全 措施",
+      "数 控",
+      "千万 不要",
+      "▁S Y",
+      "▁ SY",
+      "▁l enses",
+      "▁len ses",
+      "▁lens es",
+      "出 道",
+      "否 认",
+      "est rian",
+      "estr ian",
+      "▁es lint",
+      "▁ eslint",
+      "童 年",
+      "▁Jim my",
+      "dt o",
+      "d to",
+      "了 不少",
+      "nW rit",
+      "n Writ",
+      "▁o nt",
+      "▁on t",
+      "▁ ont",
+      "▁C K",
+      "▁ CK",
+      "jo urnal",
+      "jour nal",
+      "j ournal",
+      "fa ke",
+      "f ake",
+      "final s",
+      "fin als",
+      "fi nals",
+      "f inals",
+      "差 不",
+      "▁b anned",
+      "▁ban ned",
+      "对 不起",
+      "相对 于",
+      "相 对于",
+      "er ialize",
+      "erial ize",
+      "▁b a",
+      "▁ ba",
+      "垄 断",
+      "公交 车",
+      "▁o ught",
+      "▁ou ght",
+      "▁ ought",
+      "Enc oder",
+      "Encode r",
+      "民 终",
+      "▁R oger",
+      "▁Ro ger",
+      "▁Rog er",
+      "nP res",
+      "nPre s",
+      "nPr es",
+      "n Pres",
+      "▁rot ating",
+      "▁pron ounced",
+      "▁gen etically",
+      "▁gene tically",
+      "▁genetic ally",
+      "不 着",
+      "那 边",
+      "Se m",
+      "S em",
+      "▁ 若",
+      "hi t",
+      "h it",
+      "你们 的",
+      "你 们的",
+      "发布 会",
+      "▁WH ETHER",
+      "Con dition",
+      "Cond ition",
+      "样 式",
+      "▁Al ways",
+      "▁bar ely",
+      "▁bare ly",
+      "▁sport ing",
+      "▁spor ting",
+      "▁s tub",
+      "▁st ub",
+      "▁ stub",
+      "▁be ans",
+      "▁bean s",
+      "▁ beans",
+      "▁d airy",
+      "▁da iry",
+      "▁De fine",
+      "▁Def ine",
+      "▁ Define",
+      "▁co oper",
+      "ut a",
+      "u ta",
+      "▁p ants",
+      "▁pan ts",
+      "▁pant s",
+      "▁div isions",
+      "▁division s",
+      "▁enjoy able",
+      "QU I",
+      "Q UI",
+      "co unts",
+      "count s",
+      "▁incorpor ating",
+      "遵守 监规",
+      "▁Or acle",
+      "▁ Oracle",
+      "是否 准许",
+      "com patible",
+      "compat ible",
+      "log out",
+      "logo ut",
+      "tra nsl",
+      "trans l",
+      "并 承担",
+      "▁R u",
+      "▁ Ru",
+      "未 果",
+      "▁Ch ampion",
+      "经本院 传票传唤",
+      "▁m t",
+      "▁ mt",
+      "▁har mon",
+      "▁harm on",
+      "▁Bill board",
+      "nS te",
+      "nSt e",
+      "n Ste",
+      "湖 区",
+      "os pel",
+      "osp el",
+      "view er",
+      "vie wer",
+      "▁en semble",
+      "范 畴",
+      "le st",
+      "les t",
+      "l est",
+      "普通 程序",
+      "▁pr iced",
+      "▁price d",
+      "▁pri ced",
+      "忠 诚",
+      "▁j aw",
+      "▁ja w",
+      "ric ane",
+      "将 自己",
+      "cl ub",
+      "▁atmosph eric",
+      "os tat",
+      "ost at",
+      "osta t",
+      "o stat",
+      "gr ams",
+      "gram s",
+      "gra ms",
+      "start ed",
+      "star ted",
+      "▁ 都市",
+      "▁st olen",
+      "▁sto len",
+      "▁stole n",
+      "绝 缘",
+      "也比 较",
+      "也 比较",
+      "▁F ace",
+      "▁Fac e",
+      "▁Fa ce",
+      "▁ Face",
+      "▁W heel",
+      "▁s uits",
+      "▁su its",
+      "▁suit s",
+      "nC lass",
+      "nCl ass",
+      "n Class",
+      "pr eced",
+      "pre ced",
+      "prec ed",
+      "珍 珠",
+      "▁op ioid",
+      "柴 油",
+      "▁schizophren ia",
+      "▁T S",
+      "▁ TS",
+      "底 面",
+      "nI S",
+      "n IS",
+      "ri ched",
+      "ric hed",
+      "rich ed",
+      "po w",
+      "p ow",
+      "Cou ld",
+      "C ould",
+      "Stud y",
+      "\"} ,\"",
+      "\"}, \"",
+      "\" },\"",
+      "Can cel",
+      "C ancel",
+      "修 炼",
+      "生命 的",
+      "生 命的",
+      "▁C lar",
+      "▁Cl ar",
+      "F W",
+      "给付 原告",
+      "推出 了",
+      "推 出了",
+      "an ine",
+      "ani ne",
+      "a nine",
+      "▁Con sole",
+      "▁Cons ole",
+      "▁ Console",
+      "ici encies",
+      "的情 形",
+      "的 情形",
+      "▁r ack",
+      "▁ra ck",
+      "▁rac k",
+      "▁ rack",
+      "[] ,",
+      "[ ],",
+      "所以 说",
+      "控制 系统",
+      "Bot tom",
+      "B ottom",
+      "纳 金",
+      "租 房",
+      "讲述 了",
+      "讲 述了",
+      "od ic",
+      "odi c",
+      "o dic",
+      "经济 效益",
+      "▁any time",
+      "÷ （",
+      "头 上",
+      "▁se ating",
+      "▁sea ting",
+      "▁seat ing",
+      "s z",
+      "入 侵",
+      "▁m ud",
+      "▁mu d",
+      "▁ mud",
+      "▁sed iment",
+      "▁def icit",
+      "汇 率",
+      "披 露",
+      "时间 和",
+      "me asure",
+      "▁des irable",
+      "优 美",
+      "提供 证据",
+      "▁add itionally",
+      "▁addition ally",
+      "▁additional ly",
+      "▁e ars",
+      "▁ear s",
+      "▁ ears",
+      "le ad",
+      "lea d",
+      "l ead",
+      "bir th",
+      "b irth",
+      "年 年",
+      "nI ts",
+      "nIt s",
+      "n Its",
+      "几 百",
+      "da l",
+      "d al",
+      "▁C offee",
+      "▁Co ffee",
+      "▁gen otype",
+      "ah an",
+      "aha n",
+      "a han",
+      "▁Kenn edy",
+      "gi rl",
+      "g irl",
+      "▁R ico",
+      "▁Ric o",
+      "▁Ri co",
+      "E B",
+      "原 点",
+      "平台 的",
+      "平 台的",
+      "Res et",
+      "Re set",
+      "眼 光",
+      "月 亮",
+      "发布 了",
+      "eg u",
+      "e gu",
+      "表现 出",
+      "保险 公司",
+      "的那 种",
+      "的 那种",
+      "认为 是",
+      "的名 字",
+      "的 名字",
+      "▁I an",
+      "▁Ne pal",
+      "▁Nep al",
+      "com ed",
+      "come d",
+      "co med",
+      "c omed",
+      "▁b esides",
+      "▁bes ides",
+      "▁beside s",
+      "▁prior ities",
+      "▁priorit ies",
+      "▁priori ties",
+      "p n",
+      "实在 是",
+      "oc used",
+      "ocus ed",
+      "N i",
+      "化 和",
+      "▁ 五",
+      "光 电",
+      "▁s ul",
+      "▁su l",
+      "出具 借条",
+      "▁mem orial",
+      "▁memo rial",
+      "▁memor ial",
+      "Sa m",
+      "S am",
+      "钓 鱼",
+      "▁M g",
+      "▁ Mg",
+      "T unes",
+      "▁C raft",
+      "▁Cra ft",
+      "▁ Craft",
+      "看 上去",
+      "▁O wn",
+      "▁Ow n",
+      "▁ Own",
+      "▁D oll",
+      "▁Do ll",
+      "▁Dol l",
+      "▁sp acing",
+      "▁spac ing",
+      "▁spa cing",
+      "▁ spacing",
+      "▁tra jectory",
+      "▁traject ory",
+      "进 率",
+      "▁pl anes",
+      "▁plan es",
+      "▁plane s",
+      "▁ planes",
+      "▁contin uity",
+      "▁continu ity",
+      "▁y arn",
+      "▁ya rn",
+      "▁ yarn",
+      "▁Car los",
+      "▁Carl os",
+      "▁Carlo s",
+      "求 职",
+      "评 定",
+      "第一百 六",
+      "▁C alculus",
+      "▁Calcul us",
+      "▁Calc ulus",
+      "f w",
+      "\\' ^",
+      "\\ '^",
+      "正 整数",
+      "▁ec ological",
+      "▁eco logical",
+      "▁techn icians",
+      "▁technician s",
+      "歧 视",
+      "▁Back ground",
+      "▁ Background",
+      "ga s",
+      "g as",
+      "▁P ok",
+      "▁Po k",
+      "▁d ated",
+      "▁dat ed",
+      "▁date d",
+      "▁da ted",
+      "▁ dated",
+      "▁ad apter",
+      "▁adapt er",
+      "▁adap ter",
+      "▁ adapter",
+      "丢 失",
+      "左 侧",
+      "各 方",
+      "\": [{\"",
+      "\":[ {\"",
+      "答 应",
+      "Ho t",
+      "H ot",
+      "ps ych",
+      "psy ch",
+      "p sych",
+      "▁Al gorithm",
+      "▁ Algorithm",
+      "pol l",
+      "po ll",
+      "p oll",
+      "▁re flex",
+      "▁ref lex",
+      "日向本院 申请强制执行",
+      "日向本院申请 强制执行",
+      "日 向本院申请强制执行",
+      "比 起",
+      "ia e",
+      "i ae",
+      "leg end",
+      "lege nd",
+      "冷 冻",
+      "行 李",
+      "适当 的",
+      "or ch",
+      "现羁押 于",
+      "▁\" ;\\",
+      "▁ \";\\",
+      "▁fall en",
+      "▁fal len",
+      "选 拔",
+      "难 受",
+      "\": \"<",
+      "\":\" <",
+      "▁C hand",
+      "▁Ch and",
+      "▁Cha nd",
+      "▁Chan d",
+      "一 面",
+      "脱贫 攻坚",
+      "set minus",
+      "次 的",
+      "am d",
+      "a md",
+      "▁S ha",
+      "▁Sh a",
+      "▁ Sha",
+      "co dec",
+      "code c",
+      "cod ec",
+      "c odec",
+      "强 行",
+      "▁Pat ri",
+      "▁Pa tri",
+      "中华人民共和国民法 通则",
+      "料 酒",
+      "▁a uf",
+      "▁au f",
+      "▁ auf",
+      "▁LI ABLE",
+      "per missions",
+      "perm issions",
+      "permission s",
+      "D s",
+      "▁N HL",
+      "▁NH L",
+      "▁correct ed",
+      "▁Cr uz",
+      "▁Cru z",
+      "▁t ongue",
+      "▁ton gue",
+      "自 收到",
+      "制 约",
+      "▁or igins",
+      "▁orig ins",
+      "▁origin s",
+      "日 向",
+      "该公司 经理",
+      "平 原",
+      "▁G H",
+      "▁ GH",
+      "▁Prot ocol",
+      "▁ Protocol",
+      "汇 总",
+      "il lin",
+      "ill in",
+      "illi n",
+      "\" (",
+      "二维 码",
+      "不断 提高",
+      "Al ter",
+      "Alt er",
+      "▁pres ervation",
+      "▁Br others",
+      "▁Bro thers",
+      "▁Brother s",
+      "大 理",
+      "gl ut",
+      "▁ 水",
+      "▁reg ret",
+      "▁b ubble",
+      "▁bub ble",
+      "ME SS",
+      "MES S",
+      "M ESS",
+      "▁T reat",
+      "▁Tr eat",
+      "▁Tre at",
+      "▁V ision",
+      "▁Vis ion",
+      "▁ Vision",
+      "▁custom ize",
+      "ref lect",
+      "▁congen ital",
+      "▁a sp",
+      "▁as p",
+      "▁ asp",
+      "es ters",
+      "est ers",
+      "ester s",
+      "este rs",
+      "e sters",
+      "document ation",
+      "od ont",
+      "odo nt",
+      "odon t",
+      "▁Hung ary",
+      "▁Hun gary",
+      "空 格",
+      "▁pre cursor",
+      "▁prec ursor",
+      "he w",
+      "h ew",
+      "九 龙",
+      "为 自己",
+      "▁Sen ator",
+      "▁j ack",
+      "▁ja ck",
+      "▁ jack",
+      "成一 个",
+      "成 一个",
+      "▁cl ever",
+      "▁cle ver",
+      "▁cin ema",
+      "甲 状",
+      "▁prosp ects",
+      "▁prospect s",
+      "▁T M",
+      "▁ TM",
+      "哈尔滨 市",
+      "又 要",
+      "▁IN TER",
+      "▁INT ER",
+      "▁ INTER",
+      "▁Pro of",
+      "▁ Proof",
+      "其他 可供执行的财产",
+      "▁f u",
+      "▁ fu",
+      "WOR K",
+      "W ORK",
+      "▁in con",
+      "▁i ncon",
+      "▁inc on",
+      "核 桃",
+      "DE X",
+      "D EX",
+      "▁train ees",
+      "▁In stit",
+      "▁Inst it",
+      "▁Ins tit",
+      "一 夜",
+      "▁C arm",
+      "▁Car m",
+      "▁Ca rm",
+      "工 序",
+      "逾期 不",
+      "▁re production",
+      "▁rep roduction",
+      "▁reprodu ction",
+      "▁repro duction",
+      "▁reproduc tion",
+      "应 承担",
+      "目前 的",
+      "目 前的",
+      "STAT US",
+      "▁out doors",
+      "▁outdoor s",
+      "管理 体系",
+      "灌 溉",
+      "风 味",
+      "他 也",
+      "▁p seudo",
+      "▁pseud o",
+      "im o",
+      "i mo",
+      "在 网上",
+      "▁Victor ian",
+      "▁Victoria n",
+      "sw ap",
+      "s wap",
+      "▁dis pers",
+      "▁disp ers",
+      "▁cum ulative",
+      "来 临",
+      "ja b",
+      "j ab",
+      "宁夏 回",
+      "Im plemented",
+      "娃 娃",
+      "蒸 汽",
+      "机 理",
+      "眼 中",
+      "▁w iki",
+      "▁wi ki",
+      "▁ wiki",
+      "▁a thlete",
+      "▁athlet e",
+      "ti n",
+      "t in",
+      "再 度",
+      "▁cop ied",
+      "▁Cam era",
+      "▁ Camera",
+      "科技 进步",
+      "扬 州",
+      "票 据",
+      "ou sel",
+      "ous el",
+      "ouse l",
+      "▁B eta",
+      "▁Be ta",
+      "▁Bet a",
+      "▁ Beta",
+      "▁par sed",
+      "▁parse d",
+      "▁pars ed",
+      "▁ parsed",
+      "е м",
+      "Fac t",
+      "Fa ct",
+      "F act",
+      "▁Mat hematic",
+      "▁Math ematic",
+      "NB A",
+      "N BA",
+      "▁R idge",
+      "▁Rid ge",
+      "▁com ply",
+      "▁comp ly",
+      "▁compl y",
+      "手 表",
+      "SD K",
+      "S DK",
+      "原审 法院",
+      "▁r iders",
+      "▁ri ders",
+      "▁ride rs",
+      "▁rid ers",
+      "▁rider s",
+      "等 式",
+      "▁c ups",
+      "▁cup s",
+      "▁cu ps",
+      "ag ination",
+      "agi nation",
+      "agin ation",
+      "um ble",
+      "umb le",
+      "猫 咪",
+      "▁exp ects",
+      "▁expect s",
+      "弟 子",
+      "Ex port",
+      "Exp ort",
+      "▁int imate",
+      "▁intim ate",
+      "▁electro de",
+      "▁electr ode",
+      "兼 职",
+      "se ek",
+      "see k",
+      "s eek",
+      "▁b rew",
+      "▁br ew",
+      "▁bre w",
+      "▁ brew",
+      "▁V e",
+      "可以 从",
+      "na c",
+      "n ac",
+      "▁Co al",
+      "辞 职",
+      "nt ime",
+      "n time",
+      "th y",
+      "t hy",
+      "ne tt",
+      "net t",
+      "n ett",
+      "▁re sign",
+      "▁res ign",
+      "render er",
+      "is b",
+      "i sb",
+      "▁offer ings",
+      "▁offering s",
+      "前 款",
+      "Bus iness",
+      "B usiness",
+      "充满 了",
+      "充 满了",
+      "Com mun",
+      "Comm un",
+      "ph il",
+      "phi l",
+      "in ja",
+      "inj a",
+      "▁educ ators",
+      "▁educator s",
+      "▁ 逝世",
+      "可靠 性",
+      "nOn ly",
+      "n Only",
+      "▁t oll",
+      "▁to ll",
+      "▁Di agn",
+      "▁ Diagn",
+      "rg b",
+      "r gb",
+      "Cl ear",
+      "C lear",
+      "en sin",
+      "ens in",
+      "共 识",
+      "太 极",
+      "▁T ORT",
+      "▁TO RT",
+      "▁ 门",
+      "el ing",
+      "eli ng",
+      "elin g",
+      "e ling",
+      "▁b reed",
+      "▁br eed",
+      "▁bre ed",
+      "de b",
+      "d eb",
+      "back up",
+      "事 儿",
+      "EL D",
+      "E LD",
+      "项 公式",
+      "▁T ol",
+      "▁To l",
+      "南 市",
+      "比 特",
+      "\": [\"",
+      "\":[ \"",
+      "邮政 储蓄",
+      "▁pro ving",
+      "▁pr oving",
+      "▁prov ing",
+      "▁ proving",
+      "▁Free dom",
+      "▁Fre edom",
+      "▁uniform ly",
+      "nS cott",
+      "nSc ott",
+      "n Scott",
+      "▁fix ing",
+      "▁fi xing",
+      "Ba nk",
+      "Ban k",
+      "B ank",
+      "▁S of",
+      "▁So f",
+      "顺 便",
+      "ri l",
+      "r il",
+      "ol ysis",
+      "oly sis",
+      "▁do zens",
+      "▁dozen s",
+      "▁mark edly",
+      "▁marked ly",
+      "D N",
+      "▁Medic are",
+      "op acity",
+      "opa city",
+      "ER N",
+      "E RN",
+      "▁roll ed",
+      "▁ rolled",
+      "会 让",
+      "党 史",
+      "▁al phabet",
+      "▁alpha bet",
+      "▁alph abet",
+      "▁trans parency",
+      "简 便",
+      "▁occup ational",
+      "▁occupation al",
+      "▁\" *",
+      "▁ \"*",
+      "▁print f",
+      "▁ printf",
+      "签订 的",
+      "el o",
+      "e lo",
+      "ob ar",
+      "oba r",
+      "o bar",
+      "sh irt",
+      "shi rt",
+      "勇 士",
+      "口 号",
+      "▁AR ISING",
+      "妙 的",
+      "▁t ran",
+      "▁tr an",
+      "▁tra n",
+      "▁u uid",
+      "▁ uuid",
+      "▁entreprene urs",
+      "▁entrepreneur s",
+      "连 云",
+      "nn pm",
+      "nnp m",
+      "n npm",
+      "▁sur geon",
+      "▁surg eon",
+      "▁surge on",
+      "▁note book",
+      "▁noteb ook",
+      "▁ notebook",
+      "di ed",
+      "die d",
+      "d ied",
+      "▁d ying",
+      "▁dy ing",
+      "▁adv ocate",
+      "▁advoc ate",
+      "关 税",
+      "风 情",
+      "▁G PU",
+      "▁GP U",
+      "▁ GPU",
+      "宁夏回 族自治区",
+      "work space",
+      "works pace",
+      "▁inv itation",
+      "▁ 龙",
+      "之 人",
+      "▁I GF",
+      "▁IG F",
+      "▁ IGF",
+      "▁tremend ous",
+      "Sc an",
+      "S can",
+      "ipl inary",
+      "众 人",
+      "▁S PE",
+      "▁SP E",
+      "▁ SPE",
+      "骄 傲",
+      "▁C LA",
+      "▁CL A",
+      "▁ CLA",
+      "▁F ig",
+      "▁Fi g",
+      "▁ Fig",
+      "cur r",
+      "cu rr",
+      "▁其 四",
+      "日受理 后",
+      "日 受理后",
+      "Ow ner",
+      "Own er",
+      "▁table ts",
+      "▁tab lets",
+      "▁tablet s",
+      "▁ell iptic",
+      "▁ellipt ic",
+      "▁end orse",
+      "▁endors e",
+      "看 书",
+      "ls t",
+      "l st",
+      "伤 心",
+      "袖 添香",
+      "m iddle",
+      "增 速",
+      "at iv",
+      "ati v",
+      "ul ators",
+      "ulator s",
+      "▁mem orable",
+      "▁memor able",
+      "烹 饪",
+      "有多 种",
+      "有 多种",
+      "▁in ert",
+      "oc s",
+      "o cs",
+      "▁ 材料",
+      "la tex",
+      "late x",
+      "lat ex",
+      "l atex",
+      "实 训",
+      "quis ite",
+      "▁tum our",
+      "本 届",
+      "翻 炒",
+      "nC ould",
+      "n Could",
+      "腐 败",
+      "具体 情况",
+      "光 伏",
+      "媳 妇",
+      "man ifest",
+      "manif est",
+      "mani fest",
+      "院 士",
+      "Child ren",
+      "eth eless",
+      "ethe less",
+      "e theless",
+      "▁for wards",
+      "▁forward s",
+      "▁cle aned",
+      "▁clean ed",
+      "▁ cleaned",
+      "▁st upid",
+      "▁activ ists",
+      "▁activist s",
+      "▁concent rated",
+      "▁concentr ated",
+      "▁concentrate d",
+      "国民 党",
+      "演 奏",
+      "op ing",
+      "o ping",
+      "air o",
+      "ai ro",
+      "a iro",
+      "▁no mine",
+      "▁nom ine",
+      "▁nomin e",
+      "银 河",
+      "cont rollers",
+      "control lers",
+      "controller s",
+      "▁counsel ing",
+      "▁couns eling",
+      "三 千",
+      "synt ax",
+      "syn tax",
+      "s yntax",
+      "Stud ent",
+      "中 关村",
+      "的取值范围 是",
+      "▁end less",
+      "▁endl ess",
+      "款 式",
+      "lo gen",
+      "log en",
+      "l ogen",
+      "C U",
+      "k b",
+      "▁ret reat",
+      "坚 强",
+      "德 市",
+      "辅 料",
+      "前 沿",
+      "▁H orn",
+      "▁Hor n",
+      "▁Ho rn",
+      "ceed ings",
+      "▁att orneys",
+      "▁attorney s",
+      "我 能",
+      "下 属",
+      "廉 政",
+      "对称 轴",
+      "查明 被执行人",
+      "▁gr ammar",
+      "▁gram mar",
+      "Eff ects",
+      "Effect s",
+      "n parser",
+      "▁celebr ating",
+      "科大 学",
+      "科 大学",
+      "f h",
+      "百 家",
+      "▁c u",
+      "▁ cu",
+      "▁Coll ins",
+      "▁alleg ations",
+      "Mon itor",
+      "nRe search",
+      "nRes earch",
+      "n Research",
+      "五 大",
+      "沿 海",
+      "法 和",
+      "▁Ap plied",
+      "▁Appl ied",
+      "▁port rait",
+      "▁por trait",
+      "▁al located",
+      "▁all ocated",
+      "▁alloc ated",
+      "▁allocate d",
+      "各方 面",
+      "各 方面",
+      "▁f arming",
+      "▁far ming",
+      "▁farm ing",
+      "太 子",
+      "把 你",
+      "Bo ol",
+      "B ool",
+      "nY eah",
+      "▁I EEE",
+      "▁IE EE",
+      "▁ IEEE",
+      "当 作",
+      "花 卉",
+      "▁Ver mont",
+      "▁Verm ont",
+      "采 信",
+      "ens itive",
+      "ensit ive",
+      "il lation",
+      "ill ation",
+      "illa tion",
+      "▁V I",
+      "▁ VI",
+      "▁H eader",
+      "▁He ader",
+      "▁Head er",
+      "▁ Header",
+      "联 动",
+      "▁p etition",
+      "▁pet ition",
+      "的 成绩",
+      "香 蕉",
+      "'] ;\\",
+      "' ];\\",
+      "py plot",
+      "pyp lot",
+      "▁Cr icket",
+      "▁excit ement",
+      "吸 烟",
+      "通 话",
+      "▁Gr iff",
+      "▁in coming",
+      "▁inc oming",
+      "nE m",
+      "n Em",
+      "▁organ ised",
+      "▁organise d",
+      "夕 阳",
+      "的话 题",
+      "的 话题",
+      "▁com promise",
+      "▁comp romise",
+      "▁comprom ise",
+      "衡 量",
+      "apt ive",
+      "a ptive",
+      "把 他",
+      "通过 了",
+      "通 过了",
+      "▁or bital",
+      "▁orbit al",
+      "奇 幻",
+      "绘 制",
+      "▁g ig",
+      "▁gi g",
+      "▁E arl",
+      "▁Ear l",
+      "▁get attr",
+      "▁ getattr",
+      "▁delight ed",
+      "per t",
+      "pe rt",
+      "p ert",
+      "或者 其他",
+      "▁util ities",
+      "▁ut ilities",
+      "▁ utilities",
+      "自 有",
+      "两个 月",
+      "两 个月",
+      "▁t ex",
+      "▁te x",
+      "▁ tex",
+      "n q",
+      "▁micro m",
+      "▁mic rom",
+      "▁p igs",
+      "▁pi gs",
+      "▁pig s",
+      "▁o t",
+      "▁ ot",
+      "in go",
+      "ing o",
+      "i ngo",
+      "▁bl ame",
+      "▁art ifacts",
+      "▁artifact s",
+      "▁ artifacts",
+      "▁op ponent",
+      "▁std out",
+      "▁ stdout",
+      "▁b om",
+      "▁bo m",
+      "通 俗",
+      "▁F an",
+      "▁Fa n",
+      "aw esome",
+      "庭审 中",
+      "spec ially",
+      "special ly",
+      "s pecially",
+      "▁highlight ing",
+      "▁im press",
+      "▁imp ress",
+      "▁renew al",
+      "头 脑",
+      "▁T ypes",
+      "▁Type s",
+      "▁Ty pes",
+      "▁Typ es",
+      "▁ Types",
+      "or ious",
+      "ori ous",
+      "orio us",
+      "▁ach ievements",
+      "▁achieve ments",
+      "▁achievement s",
+      "▁C ultural",
+      "▁Cult ural",
+      "周 转",
+      "出一 个",
+      "出 一个",
+      "art ifact",
+      "▁art eries",
+      "▁arter ies",
+      "▁pl aque",
+      "▁m ysql",
+      "▁my sql",
+      "▁mys ql",
+      "▁ mysql",
+      "▁V i",
+      "▁ Vi",
+      "LI NK",
+      "LIN K",
+      "L INK",
+      "re as",
+      "rea s",
+      "r eas",
+      "is ors",
+      "isor s",
+      "iso rs",
+      "▁V oice",
+      "▁Vo ice",
+      "▁ Voice",
+      "nA nt",
+      "nAn t",
+      "n Ant",
+      "nT ry",
+      "nTr y",
+      "n Try",
+      "▁Sc hema",
+      "▁Sch ema",
+      "▁Sche ma",
+      "▁ Schema",
+      "▁festival s",
+      "▁fest ivals",
+      "▁ 创作",
+      "▁z eros",
+      "▁zero s",
+      "▁ze ros",
+      "▁ zeros",
+      "▁re sc",
+      "▁r esc",
+      "▁res c",
+      "▁My SQL",
+      "▁ MySQL",
+      "▁Cert ificate",
+      "▁ Certificate",
+      "▁asc ending",
+      "投 稿",
+      "▁t v",
+      "▁ tv",
+      "▁d ough",
+      "▁do ugh",
+      "▁dou gh",
+      "▁end omet",
+      "▁ 全",
+      "巧 克力",
+      "研究 会",
+      "Un der",
+      "U nder",
+      "nW elcome",
+      "n Welcome",
+      "▁p acking",
+      "▁pack ing",
+      "▁pac king",
+      "▁Integr ation",
+      "▁ Integration",
+      "in ology",
+      "ino logy",
+      "▁current s",
+      "▁curr ents",
+      "ur pose",
+      "▁R ou",
+      "▁Ro u",
+      "必要 条件",
+      "▁set Up",
+      "子 公司",
+      "滞 纳金",
+      "▁re pos",
+      "▁rep os",
+      "▁repo s",
+      "▁ repos",
+      "回 首",
+      "通 行",
+      "▁\\\\ \\\\\\",
+      "▁\\ \\\\\\\\",
+      "▁\\\\\\\\ \\",
+      "▁\\\\\\ \\\\",
+      "▁ \\\\\\\\\\",
+      "sa n",
+      "s an",
+      "工 地",
+      "▁Dub ai",
+      "▁Du bai",
+      "是一 位",
+      "是 一位",
+      "▁C raig",
+      "▁Cra ig",
+      "▁sub jective",
+      "▁subject ive",
+      "▁inform ative",
+      "非常 好",
+      "▁j acket",
+      "▁jack et",
+      "ob il",
+      "obi l",
+      "o bil",
+      "变 的",
+      "右 侧",
+      "▁s ig",
+      "▁si g",
+      "▁ sig",
+      "ful ness",
+      "▁hydro ly",
+      "▁b ite",
+      "▁bi te",
+      "▁bit e",
+      "▁par sing",
+      "▁pars ing",
+      "▁sol vent",
+      "▁solve nt",
+      "▁solv ent",
+      "向 往",
+      "当时 的",
+      "当 时的",
+      "钢 筋",
+      "▁des criptive",
+      "▁descript ive",
+      "ur tle",
+      "urt le",
+      "不 幸",
+      "▁l n",
+      "▁ ln",
+      "EN ER",
+      "ENE R",
+      "E NER",
+      "诉讼费用 交纳办法",
+      "nTe levision",
+      "▁prob abilities",
+      "书 院",
+      "机 型",
+      "▁sc anner",
+      "▁scan ner",
+      "▁ scanner",
+      "伤 口",
+      "候 选",
+      "oz yg",
+      "县 政府",
+      "等 方式",
+      "县 人民政府",
+      "基 督",
+      "本 品",
+      "CHE CK",
+      "C HECK",
+      "▁m issions",
+      "▁miss ions",
+      "▁mission s",
+      "▁ missions",
+      "Test ing",
+      "T esting",
+      "常 数",
+      "钱 的",
+      "▁appro ached",
+      "▁approach ed",
+      "语 句",
+      "▁s aves",
+      "▁sa ves",
+      "▁save s",
+      "▁sav es",
+      "是 通过",
+      "的大 型",
+      "的 大型",
+      "山 路",
+      "}\" ,",
+      "} \",",
+      "ar ten",
+      "art en",
+      "arte n",
+      "found er",
+      "fo under",
+      "▁G ard",
+      "▁Ga rd",
+      "▁Gar d",
+      "▁Stan ley",
+      "I g",
+      "▁L GBT",
+      "▁LG BT",
+      "最 优",
+      "退 役",
+      "十七 章",
+      "▁l id",
+      "▁li d",
+      "stack overflow",
+      "iv alent",
+      "ival ent",
+      "▁leg islative",
+      "▁legisl ative",
+      "一 期",
+      "打 扮",
+      "▁K B",
+      "▁ KB",
+      "▁S r",
+      "▁ Sr",
+      "se ar",
+      "sea r",
+      "s ear",
+      "▁ne eding",
+      "▁need ing",
+      "f y",
+      "▁u ri",
+      "▁ur i",
+      "▁ uri",
+      "▁ch arter",
+      "▁char ter",
+      "▁chart er",
+      "或者 说",
+      "▁col orectal",
+      "og o",
+      "o go",
+      "原 被告",
+      "▁C row",
+      "▁Cr ow",
+      "▁Cro w",
+      "春 季",
+      "US A",
+      "U SA",
+      "▁background s",
+      "ra ses",
+      "ras es",
+      "rase s",
+      "r ases",
+      "▁Ne braska",
+      "帮 扶",
+      "二 百",
+      "St udio",
+      "Stud io",
+      "ib ular",
+      "ibu lar",
+      "是 比较",
+      "▁Colomb ia",
+      "末 尾",
+      "嵌 入",
+      "▁i i",
+      "▁ ii",
+      "▁sc ans",
+      "▁scan s",
+      "__ _",
+      "_ __",
+      "nM c",
+      "n Mc",
+      "▁Pr incess",
+      "▁Princ ess",
+      "▁Prince ss",
+      "▁sem antic",
+      "▁ semantic",
+      "问 卷",
+      "AA A",
+      "A AA",
+      "▁CO MM",
+      "▁COM M",
+      "▁ COMM",
+      "记 功",
+      "这 件",
+      "CL UDE",
+      "CLUD E",
+      "就 需要",
+      "▁K o",
+      "异 界",
+      "▁Ad ult",
+      "在 哪",
+      "▁b ell",
+      "▁be ll",
+      "▁bel l",
+      "▁ bell",
+      "im iter",
+      "imit er",
+      "imi ter",
+      "▁immigr ants",
+      "▁immigrant s",
+      "uk a",
+      "u ka",
+      "ric ally",
+      "rical ly",
+      "r ically",
+      "nC PU",
+      "n CPU",
+      "扣 除",
+      "特 效",
+      "请求 法院",
+      "an ship",
+      "ans hip",
+      "詹 姆斯",
+      "▁ 释义",
+      "▁Rem ote",
+      "▁ Remote",
+      "▁equival ence",
+      "▁ у",
+      "▁invest or",
+      "▁Sim ulation",
+      "▁ Simulation",
+      "原因 是",
+      "▁sp am",
+      "▁spa m",
+      "▁ spam",
+      "▁Mil an",
+      "▁Mi lan",
+      "▁Appro ach",
+      "▁Transport ation",
+      "抢 救",
+      "▁spe aks",
+      "▁speak s",
+      "对 角",
+      "意 愿",
+      "al ar",
+      "ala r",
+      "a lar",
+      "ar ta",
+      "art a",
+      "早 在",
+      "愤 怒",
+      "自然 人",
+      "▁con stitute",
+      "▁constit ute",
+      "▁constitu te",
+      "伦 理",
+      "二次 根",
+      "Ex tract",
+      "Extra ct",
+      "▁market place",
+      "ot ics",
+      "otic s",
+      "oti cs",
+      "Cons umer",
+      "小学 文化",
+      "out puts",
+      "output s",
+      "▁ 从",
+      "乐 队",
+      "▁simple st",
+      "▁simpl est",
+      "▁as sembled",
+      "▁assembl ed",
+      "▁assemble d",
+      "耐 磨",
+      "道 具",
+      "nH aving",
+      "nHa ving",
+      "▁ 叶",
+      "衍 生",
+      "▁Emer gency",
+      "▁逝世 日期",
+      "▁W onder",
+      "▁Won der",
+      "▁Wo nder",
+      "▁mag ical",
+      "▁magic al",
+      "▁reg ulators",
+      "▁regul ators",
+      "▁regulator s",
+      "适 配",
+      "农 田",
+      "ch ol",
+      "cho l",
+      "c hol",
+      "感 和",
+      "▁Associ ate",
+      "▁prov incial",
+      "ro v",
+      "r ov",
+      "nh ave",
+      "n have",
+      "确 立",
+      "▁ 项目",
+      "▁} }\"\\",
+      "▁}} \"\\",
+      "▁}}\" \\",
+      "▁St rong",
+      "▁Str ong",
+      "▁influ ential",
+      "save d",
+      "sa ved",
+      "sav ed",
+      "s aved",
+      "▁m eg",
+      "▁me g",
+      "▁ meg",
+      "元 件",
+      "ov ine",
+      "ovi ne",
+      "o vine",
+      "▁E uler",
+      "xim ately",
+      "x imately",
+      "ig raph",
+      "igr aph",
+      "i graph",
+      "oplas tic",
+      "opl astic",
+      "oplast ic",
+      "o plastic",
+      "▁clin ics",
+      "▁clinic s",
+      "特 別",
+      "fi ction",
+      "f iction",
+      "▁B eyond",
+      "▁Be yond",
+      "▁Bey ond",
+      "▁loc als",
+      "▁local s",
+      "▁ locals",
+      "深 处",
+      "CE O",
+      "C EO",
+      "的 患者",
+      "Se p",
+      "S ep",
+      "代 书记员",
+      "▁P ope",
+      "▁Pop e",
+      "▁Po pe",
+      "nS orry",
+      "n Sorry",
+      "第一百五十四 条第一款",
+      "第一百五十四条 第一款",
+      "关键 是",
+      "地 名",
+      "op last",
+      "oplas t",
+      "opl ast",
+      "set State",
+      "re ational",
+      "reat ional",
+      "reation al",
+      "场上 位置",
+      "▁A my",
+      "▁Am y",
+      "▁d rew",
+      "▁dr ew",
+      "▁cr ops",
+      "▁cro ps",
+      "▁crop s",
+      "▁R unning",
+      "▁Run ning",
+      "▁ Running",
+      "与 中国",
+      "德 里",
+      "没 法",
+      "莆 田",
+      "nJ ames",
+      "n James",
+      "冶 金",
+      "人类 的",
+      "人 类的",
+      "ucl id",
+      "▁Wel sh",
+      "SE S",
+      "S ES",
+      "od i",
+      "o di",
+      "L F",
+      "point er",
+      "po inter",
+      "p ointer",
+      "▁unique ly",
+      "▁uniqu ely",
+      "▁Mont ana",
+      "}, \"",
+      "} ,\"",
+      "指 的是",
+      "同 居",
+      "零 部件",
+      "Fac tor",
+      "Fact or",
+      "Fa ctor",
+      "F actor",
+      "▁can onical",
+      "▁canon ical",
+      "▁ canonical",
+      "立 马",
+      "▁a dm",
+      "▁ad m",
+      "▁et hn",
+      "▁eth n",
+      "▁m ism",
+      "▁mis m",
+      "▁mi sm",
+      "新 颖",
+      "▁h ello",
+      "▁hel lo",
+      "▁hell o",
+      "▁ hello",
+      "主要 的",
+      "an onymous",
+      "方法 的",
+      "方 法的",
+      "on Click",
+      "o nClick",
+      "▁drain age",
+      "At tr",
+      "Att r",
+      "基本 的",
+      "基 本的",
+      "File name",
+      "Fil ename",
+      "Fi lename",
+      "▁travel ed",
+      "打 的",
+      "好 吧",
+      "Co rp",
+      "Cor p",
+      "C orp",
+      "实践 中",
+      "▁b eneath",
+      "▁be neath",
+      "在 不",
+      "cons umer",
+      "consum er",
+      "▁aest hetic",
+      "诠 释",
+      "pe l",
+      "p el",
+      "▁t ire",
+      "▁ti re",
+      "chan ical",
+      "武 术",
+      "已经 被",
+      "▁furn ished",
+      "▁furnish ed",
+      "AS K",
+      "A SK",
+      "▁B iology",
+      "▁Bi ology",
+      "▁Bio logy",
+      "▁position ing",
+      "▁E ye",
+      "▁Ey e",
+      "▁ Eye",
+      "乙 烯",
+      "\\\\ :",
+      "境 外",
+      "异 地",
+      "转 账",
+      "首 批",
+      "综合 征",
+      "lo oking",
+      "look ing",
+      "loo king",
+      "l ooking",
+      "▁gener ators",
+      "▁generator s",
+      "▁rec ycling",
+      "▁recycl ing",
+      "pl asia",
+      "ie u",
+      "i eu",
+      "主管 部门",
+      "▁ident ifies",
+      "并 已",
+      "ta bs",
+      "tab s",
+      "t abs",
+      "▁TR UE",
+      "▁ TRUE",
+      "本 着",
+      "CA P",
+      "C AP",
+      "m ajor",
+      "▁P erm",
+      "▁Per m",
+      "▁Pe rm",
+      "▁ Perm",
+      "▁C hem",
+      "▁Ch em",
+      "▁Che m",
+      "中 期",
+      "袭 击",
+      "党委 委员",
+      "▁trans fers",
+      "▁transfer s",
+      "▁transf ers",
+      "▁at traction",
+      "▁attract ion",
+      "▁attr action",
+      "相 亲",
+      "下 班",
+      "▁at tain",
+      "▁att ain",
+      "▁ 属",
+      "▁p ac",
+      "▁pa c",
+      "▁ pac",
+      "▁c lay",
+      "▁cl ay",
+      "化 解",
+      "有 望",
+      "▁O FF",
+      "▁OF F",
+      "▁ OFF",
+      "▁applic ants",
+      "▁applicant s",
+      "de term",
+      "det erm",
+      "d eterm",
+      "va e",
+      "v ae",
+      "ar char",
+      "arch ar",
+      "arc har",
+      "包 容",
+      "心 疼",
+      "模 特",
+      "▁P oll",
+      "▁Pol l",
+      "▁Po ll",
+      "▁ Poll",
+      "▁OTHER WISE",
+      "味 精",
+      "劳动 者",
+      "▁sign atures",
+      "▁signature s",
+      "▁trad itionally",
+      "▁traditional ly",
+      "▁tradition ally",
+      "愉 快",
+      "▁per malink",
+      "▁ permalink",
+      "Des c",
+      "De sc",
+      "D esc",
+      "远 的",
+      "or r",
+      "o rr",
+      "nD ata",
+      "n Data",
+      "▁com prising",
+      "▁compr ising",
+      "阀 门",
+      "生物 学",
+      "▁Re sort",
+      "▁Res ort",
+      "▁sub groups",
+      "▁subgroup s",
+      "co t",
+      "c ot",
+      "大 门",
+      "▁late ncy",
+      "▁lat ency",
+      "in ist",
+      "ini st",
+      "inis t",
+      "i nist",
+      "▁A rena",
+      "▁Ar ena",
+      "▁Are na",
+      "胶 囊",
+      "da r",
+      "d ar",
+      "本案 诉讼费",
+      "▁c once",
+      "▁con ce",
+      "▁co nce",
+      "▁conc e",
+      "trans port",
+      "可 视",
+      "aw k",
+      "a wk",
+      "▁ 身高",
+      "不适 合",
+      "不 适合",
+      "os hi",
+      "osh i",
+      "o shi",
+      "we ights",
+      "weight s",
+      "▁per ceptions",
+      "▁perception s",
+      "▁percept ions",
+      "我 可以",
+      "ov ich",
+      "ovic h",
+      "ovi ch",
+      "and box",
+      "ch oices",
+      "cho ices",
+      "choice s",
+      "▁N J",
+      "▁ex ams",
+      "▁exam s",
+      "E p",
+      "故 乡",
+      "▁influ enza",
+      "▁L V",
+      "▁ LV",
+      "▁A MP",
+      "▁AM P",
+      "▁ AMP",
+      "▁f law",
+      "▁fl aw",
+      "配 音",
+      "▁Finn ish",
+      "le tte",
+      "let te",
+      "lett e",
+      "l ette",
+      "▁e lite",
+      "▁el ite",
+      "▁elit e",
+      "无 业",
+      "Sim plify",
+      "{ :",
+      "▁h ass",
+      "▁has s",
+      "▁ha ss",
+      "▁ hass",
+      "▁w age",
+      "▁wa ge",
+      "▁wag e",
+      "三 包",
+      "低 的",
+      "▁d ermat",
+      "▁der mat",
+      "路 程",
+      "ta bles",
+      "table s",
+      "tab les",
+      "t ables",
+      "opath ic",
+      "▁pro ximity",
+      "в а",
+      "第八 章",
+      "第 八章",
+      "▁to String",
+      "▁ toString",
+      "ser if",
+      "▁int uitive",
+      "ad as",
+      "ada s",
+      "a das",
+      "角 落",
+      "▁G rowth",
+      "▁Grow th",
+      "▁museum s",
+      "长 达",
+      "▁Eff ects",
+      "▁Effect s",
+      "▁ Effects",
+      "相 思",
+      "▁L iga",
+      "▁Li ga",
+      "▁Lig a",
+      "oc ytic",
+      "ocyt ic",
+      "▁Al low",
+      "▁All ow",
+      "▁ Allow",
+      "Th ree",
+      "三 点",
+      "教学 中",
+      "im ag",
+      "ima g",
+      "i mag",
+      "▁d iesel",
+      "▁die sel",
+      "▁dies el",
+      "云 中",
+      "od ox",
+      "odo x",
+      "▁sl iding",
+      "并 以",
+      "态 势",
+      "如下 图",
+      "如 下图",
+      "阿拉 伯",
+      "▁P arse",
+      "▁Par se",
+      "▁Pars e",
+      "▁ Parse",
+      "才 发现",
+      "▁ob server",
+      "▁observ er",
+      "▁observe r",
+      "▁obs erver",
+      "▁ observer",
+      "众 所",
+      "住 河南省",
+      "at en",
+      "ate n",
+      "a ten",
+      "no ut",
+      "n out",
+      "大 神",
+      "▁here by",
+      "▁Br istol",
+      "抵 达",
+      "▁C OL",
+      "▁CO L",
+      "▁ COL",
+      "▁p ays",
+      "▁pay s",
+      "▁pa ys",
+      "往 来",
+      "▁an ger",
+      "▁ang er",
+      "▁ anger",
+      "▁Per u",
+      "▁Pe ru",
+      "Prom ise",
+      "P romise",
+      "▁V ar",
+      "▁Va r",
+      "▁ Var",
+      "▁s weep",
+      "▁sw eep",
+      "▁swe ep",
+      "杀 手",
+      "front end",
+      "▁t ier",
+      "▁tie r",
+      "▁ti er",
+      "▁ tier",
+      "第一百五十 四条",
+      "第一百五十四 条",
+      "六 个",
+      "旋 律",
+      "雷 达",
+      "▁Mit chell",
+      "▁Mitch ell",
+      "R M",
+      "可 选",
+      "▁assign ments",
+      "▁assignment s",
+      "头 痛",
+      "满 了",
+      "保 守",
+      "am os",
+      "amo s",
+      "a mos",
+      "▁re fin",
+      "▁ref in",
+      "▁pup ils",
+      "▁pupil s",
+      "▁inhib its",
+      "▁inhibit s",
+      "个性 化",
+      "称 之为",
+      "ch apter",
+      "na pi",
+      "nap i",
+      "n api",
+      "▁recogn ised",
+      "▁recognise d",
+      "的 责任",
+      "的 建设",
+      "吵 架",
+      "▁T oo",
+      "▁To o",
+      "▁ Too",
+      "▁P ermission",
+      "▁Per mission",
+      "▁Perm ission",
+      "▁ Permission",
+      "写 字",
+      "的 书",
+      "向本院 提出撤诉申请",
+      "向本院提出 撤诉申请",
+      "▁S omething",
+      "▁Some thing",
+      "▁Som ething",
+      "▁ Something",
+      "▁i ps",
+      "▁ip s",
+      "▁ ips",
+      "了 自己",
+      "site s",
+      "si tes",
+      "sit es",
+      "s ites",
+      "ocar cin",
+      "部 落",
+      "IM P",
+      "I MP",
+      "不出 来",
+      "不 出来",
+      "行 人",
+      "▁C ob",
+      "▁Co b",
+      "ib lings",
+      "ibl ings",
+      "ibling s",
+      "G N",
+      "齿 轮",
+      "▁conj ecture",
+      "▁conject ure",
+      "hi nd",
+      "hin d",
+      "h ind",
+      "▁ob ese",
+      "Th row",
+      "▁H AVE",
+      "▁HA VE",
+      "▁de formation",
+      "▁deform ation",
+      "▁A ve",
+      "▁Av e",
+      "▁P ent",
+      "▁Pen t",
+      "▁Pe nt",
+      "▁eth ics",
+      "如下 所示",
+      "▁bas ics",
+      "▁basic s",
+      "▁op ponents",
+      "▁opponent s",
+      "ab ol",
+      "abo l",
+      "a bol",
+      "il ia",
+      "ili a",
+      "i lia",
+      "巡 视",
+      "担任 职务",
+      "诉 求",
+      "坐落 于",
+      "温州 市",
+      "温 州市",
+      "岁以下的 儿童",
+      "和 对",
+      "▁C rown",
+      "▁Cr own",
+      "▁Cro wn",
+      "▁Crow n",
+      "偷 偷",
+      "▁treat s",
+      "▁tre ats",
+      "▁Pat ient",
+      "▁ Patient",
+      "▁activ ist",
+      "降低 了",
+      "▁H alf",
+      "▁Ha lf",
+      "▁Hal f",
+      "▁ Half",
+      "▁font s",
+      "▁ fonts",
+      "生产 线",
+      "Ex pl",
+      "Exp l",
+      "座 位",
+      "新冠 肺炎",
+      "中 年",
+      "igh bor",
+      "ighb or",
+      "BA CK",
+      "B ACK",
+      "明 亮",
+      "夜 晚",
+      "▁\" )\\",
+      "▁\") \\",
+      "▁ \")\\",
+      "▁up graded",
+      "▁upgrade d",
+      "▁upgrad ed",
+      "▁s igma",
+      "▁sig ma",
+      "▁ sigma",
+      "无 财产可供执行",
+      "▁V ienna",
+      "▁Vi enna",
+      "失 踪",
+      "▁St eps",
+      "▁Ste ps",
+      "▁Step s",
+      "▁ Steps",
+      "上诉 案件受理费",
+      "上诉案件 受理费",
+      "主 治",
+      "会 展",
+      "▁U pload",
+      "▁Up load",
+      "▁ Upload",
+      "的申请 符合法律规定",
+      "的 申请符合法律规定",
+      "无正当理由拒不到庭 参加诉讼",
+      "▁knowledge able",
+      "地 狱",
+      "▁v ic",
+      "▁vi c",
+      "▁ vic",
+      "▁ab ortion",
+      "▁abort ion",
+      "in ent",
+      "ine nt",
+      "inen t",
+      "▁enter prises",
+      "▁enterprise s",
+      "总 的",
+      "▁quot ient",
+      "▁Inter face",
+      "▁ Interface",
+      "刻 苦",
+      "Group s",
+      "悲 剧",
+      "▁Fl ore",
+      "▁Flor e",
+      "▁Flo re",
+      "月 中",
+      "▁elect oral",
+      "试 卷",
+      "内 有",
+      "ed uc",
+      "edu c",
+      "e duc",
+      "▁Wed ding",
+      "▁antagon ist",
+      "能 让",
+      "buf f",
+      "bu ff",
+      "b uff",
+      "▁E sp",
+      "▁Es p",
+      "Stack Trace",
+      "期间 的限制",
+      "期间的 限制",
+      "Vir tual",
+      "CI AL",
+      "C IAL",
+      "Z X",
+      "▁ 曹",
+      "DE F",
+      "D EF",
+      "te es",
+      "tee s",
+      "t ees",
+      "▁d oi",
+      "▁do i",
+      "▁ doi",
+      "▁l esser",
+      "▁less er",
+      "▁les ser",
+      "▁dis charg",
+      "做 个",
+      "明 朝",
+      "▁te ens",
+      "▁teen s",
+      "▁tee ns",
+      "cast ing",
+      "cas ting",
+      "c asting",
+      "▁Innov ation",
+      "un ny",
+      "unn y",
+      "事实和 理由",
+      "实 务",
+      "▁` <",
+      "▁ `<",
+      "还不 错",
+      "还 不错",
+      "最 短",
+      "▁embry onic",
+      "▁embryo nic",
+      "VAL ID",
+      "mk dir",
+      "m kdir",
+      "折 扣",
+      "▁L en",
+      "▁Le n",
+      "▁ Len",
+      "check box",
+      "▁Met ropolitan",
+      "林 地",
+      "大 厅",
+      "生活 在",
+      "生 活在",
+      "体 力",
+      "饮 用",
+      "ur st",
+      "urs t",
+      "u rst",
+      "yz er",
+      "yze r",
+      "y zer",
+      "然后 根据",
+      "▁acet yl",
+      "▁ acetyl",
+      "function al",
+      "f unctional",
+      "批 判",
+      "亲 切",
+      "▁P ure",
+      "▁Pur e",
+      "▁Pu re",
+      "震 撼",
+      "sq lite",
+      "sql ite",
+      "过 年",
+      "E loquent",
+      "nD ec",
+      "nDe c",
+      "n Dec",
+      "nP ort",
+      "nPo rt",
+      "n Port",
+      "avor ite",
+      "avo rite",
+      "R s",
+      "▁f asc",
+      "▁fa sc",
+      "▁T ouch",
+      "▁To uch",
+      "▁Tou ch",
+      "▁ Touch",
+      "四 周",
+      "▁tri angles",
+      "▁triangle s",
+      "▁triang les",
+      "▁p kg",
+      "▁pk g",
+      "▁ pkg",
+      "yn asty",
+      "yna sty",
+      "lo d",
+      "l od",
+      "出 资",
+      "抽 样",
+      "im ir",
+      "imi r",
+      "i mir",
+      "nJ oin",
+      "nJo in",
+      "n Join",
+      "▁R acing",
+      "▁Ra cing",
+      "▁Rac ing",
+      "▁C NS",
+      "▁CN S",
+      "▁bo ots",
+      "▁boot s",
+      "▁cons ume",
+      "▁consum e",
+      "he y",
+      "h ey",
+      "了 中国",
+      "在一 定",
+      "在 一定",
+      "▁X box",
+      "▁ 人",
+      "▁u i",
+      "▁ ui",
+      "▁Ap pl",
+      "▁App l",
+      "v t",
+      "nIn put",
+      "n Input",
+      "▁labor atories",
+      "▁R ever",
+      "▁Re ver",
+      "▁Rev er",
+      "▁ Rever",
+      "br ane",
+      "bra ne",
+      "▁Str ategy",
+      "▁Strateg y",
+      "▁ Strategy",
+      "▁disappoint ed",
+      "We ll",
+      "W ell",
+      ", (",
+      "业 界",
+      "促 使",
+      "ST M",
+      "S TM",
+      "八个 月",
+      "八 个月",
+      "▁a ired",
+      "▁air ed",
+      "▁ai red",
+      "▁ aired",
+      "section s",
+      "sect ions",
+      "s ections",
+      "▁R C",
+      "▁ RC",
+      "▁ch ef",
+      "▁che f",
+      "▁ chef",
+      "同 人",
+      "ar el",
+      "are l",
+      "a rel",
+      "ar l",
+      "a rl",
+      "sp acing",
+      "re ck",
+      "rec k",
+      "r eck",
+      "可 分为",
+      "▁T ell",
+      "▁Te ll",
+      "▁Tel l",
+      "娱乐 圈",
+      "宽 带",
+      "nAr ch",
+      "n Arch",
+      "压 器",
+      "Mo st",
+      "M ost",
+      "温馨 提示",
+      "人 对",
+      "写 成",
+      "▁had n",
+      "▁ha dn",
+      "▁ty rosine",
+      "▁S ean",
+      "▁Se an",
+      "▁Sea n",
+      "▁squ ared",
+      "▁square d",
+      "▁ squared",
+      "▁n ationally",
+      "▁national ly",
+      "▁nation ally",
+      "科学 的",
+      "科 学的",
+      "▁s hit",
+      "▁sh it",
+      "▁ shit",
+      "▁un con",
+      "▁u ncon",
+      "▁unc on",
+      "刑期 至",
+      "atomic al",
+      "atom ical",
+      "nTh rough",
+      "n Through",
+      "\\' )[",
+      "\\') [",
+      "\\ ')[",
+      "▁mat uration",
+      "ur ated",
+      "ura ted",
+      "urate d",
+      "u rated",
+      "▁municip alities",
+      "▁municipal ities",
+      "Ph i",
+      "P hi",
+      "in ts",
+      "int s",
+      "un iform",
+      "uni form",
+      "荧 光",
+      "探 究",
+      "▁P alm",
+      "▁Pal m",
+      "▁Pa lm",
+      "▁W iley",
+      "▁Wil ey",
+      "▁Wi ley",
+      "美 洲",
+      "金 牌",
+      "ev es",
+      "eve s",
+      "e ves",
+      "增 产",
+      "widget s",
+      "wid gets",
+      "由 中国",
+      "GD P",
+      "G DP",
+      "▁S tri",
+      "▁St ri",
+      "▁Str i",
+      "科学记数 法",
+      "nd f",
+      "n df",
+      "▁B eck",
+      "▁Be ck",
+      "▁Bec k",
+      "▁W ave",
+      "▁Wa ve",
+      "▁ Wave",
+      "▁ill um",
+      "▁shift ed",
+      "▁A sp",
+      "▁As p",
+      "▁ Asp",
+      "() ),\\",
+      "()) ,\\",
+      "()), \\",
+      "( )),\\",
+      "▁H ist",
+      "▁His t",
+      "▁Hi st",
+      "▁ Hist",
+      "流 失",
+      "作为 一名",
+      "Th ank",
+      "Than k",
+      "▁ste roid",
+      "▁ster oid",
+      "▁国 家",
+      "▁ 国家",
+      "▁s isters",
+      "▁sister s",
+      "▁si sters",
+      "▁ 请",
+      "海 鲜",
+      "▁f arms",
+      "▁far ms",
+      "▁farm s",
+      "▁Rich mond",
+      "z b",
+      "▁C rim",
+      "▁Cr im",
+      "it ution",
+      "itu tion",
+      "Integer Field",
+      "小 平",
+      "▁m ant",
+      "▁man t",
+      "▁ma nt",
+      "▁ mant",
+      "双 向",
+      "R U",
+      "br is",
+      "b ris",
+      "原判 决",
+      "原 判决",
+      "Ce nt",
+      "C ent",
+      "将被 执行人",
+      "将 被执行人",
+      "▁R outer",
+      "▁Ro uter",
+      "▁Route r",
+      "▁Rou ter",
+      "▁ Router",
+      "Nav igation",
+      "差异 有",
+      "▁S leep",
+      "▁Sl eep",
+      "▁Sle ep",
+      "▁ Sleep",
+      "▁sup erb",
+      "▁super b",
+      "淡 淡",
+      "既 是",
+      "▁super vision",
+      "▁superv ision",
+      "col s",
+      "co ls",
+      "c ols",
+      "▁Bel gian",
+      "ro nes",
+      "ron es",
+      "rone s",
+      "r ones",
+      "nMean while",
+      "历史 的",
+      "B s",
+      "令 牌",
+      "描 写",
+      "▁Y am",
+      "▁Ya m",
+      "常 年",
+      "am med",
+      "amm ed",
+      "和社会 保障",
+      "和 社会保障",
+      "▁sum mit",
+      "▁summ it",
+      "▁null ptr",
+      "▁ nullptr",
+      "ri sh",
+      "ris h",
+      "r ish",
+      "商业 银行",
+      "C Y",
+      "__ ,",
+      "_ _,",
+      "万 美元",
+      "执行 标的",
+      "索 尼",
+      "N K",
+      "▁dis playing",
+      "▁display ing",
+      "副 部长",
+      "jp eg",
+      "j peg",
+      "看不 到",
+      "看 不到",
+      "泽 东",
+      "管 控",
+      "^{ (",
+      "^ {(",
+      "\\' ][",
+      "\\'] [",
+      "\\ '][",
+      "oc ha",
+      "och a",
+      "o cha",
+      "▁rel ates",
+      "▁relate s",
+      "青 睐",
+      "原 有",
+      "pro j",
+      "pr oj",
+      "运营 商",
+      "运 营商",
+      "现已 审查",
+      "▁A CTION",
+      "▁ACT ION",
+      "▁ ACTION",
+      "la zy",
+      "l azy",
+      "qu est",
+      "que st",
+      "ques t",
+      "玩 意",
+      "▁L ayout",
+      "▁La yout",
+      "▁Lay out",
+      "▁ Layout",
+      "opro ject",
+      "o project",
+      "▁M ann",
+      "▁Man n",
+      "▁Ma nn",
+      "▁in adequ",
+      "▁for cing",
+      "▁ forcing",
+      "▁cor ruption",
+      "▁corrupt ion",
+      "增 添",
+      "向本院 提交",
+      "▁sa fer",
+      "▁saf er",
+      "▁safe r",
+      "ir lines",
+      "irl ines",
+      "Cont ents",
+      "Content s",
+      "▁per fusion",
+      "▁perf usion",
+      "▁ perfusion",
+      "出 入",
+      "▁o l",
+      "▁ ol",
+      "▁f iring",
+      "▁fi ring",
+      "▁fir ing",
+      "▁be ating",
+      "▁beat ing",
+      "▁st ating",
+      "▁stat ing",
+      "宣判 前",
+      "or tex",
+      "ort ex",
+      "orte x",
+      "▁F ashion",
+      "Ol d",
+      "O ld",
+      "get Instance",
+      "本案 案件受理费",
+      "▁C urt",
+      "▁Cur t",
+      "▁Cu rt",
+      "u o",
+      "▁S it",
+      "▁Si t",
+      "catal og",
+      "c atalog",
+      "▁hand ful",
+      "高 档",
+      "▁G em",
+      "▁Ge m",
+      "no ise",
+      "n oise",
+      "▁h umor",
+      "▁hum or",
+      "▁hu mor",
+      "An notation",
+      "Ann otation",
+      "▁l akes",
+      "▁la kes",
+      "▁lake s",
+      "▁w ines",
+      "▁win es",
+      "▁wine s",
+      "▁wi nes",
+      "▁K id",
+      "▁Ki d",
+      "长方 形的",
+      "长方形 的",
+      "▁prest igious",
+      "▁Elect ronic",
+      "▁Electron ic",
+      "▁Electro nic",
+      "▁C BS",
+      "▁CB S",
+      "▁s hoe",
+      "▁sh oe",
+      "▁in herent",
+      "▁inher ent",
+      "▁inh erent",
+      "AT P",
+      "A TP",
+      "▁a my",
+      "▁am y",
+      "▁ amy",
+      "▁Sh anghai",
+      "欺 负",
+      "▁p t",
+      "▁ pt",
+      "▁S anders",
+      "▁San ders",
+      "▁Sand ers",
+      "开 源",
+      "旅 馆",
+      "▁trad itions",
+      "▁tra ditions",
+      "▁tradition s",
+      "▁partners hips",
+      "▁partner ships",
+      "▁partnership s",
+      "旅行 社",
+      "买卖 合同",
+      "▁Move ment",
+      "▁Orche stra",
+      "▁深圳 市",
+      "▁ 深圳市",
+      "日常 生活",
+      "▁text book",
+      "子 和",
+      "Se ed",
+      "See d",
+      "S eed",
+      "in ced",
+      "ince d",
+      "inc ed",
+      "p f",
+      "▁r s",
+      "▁ rs",
+      "ed ly",
+      "一 枚",
+      "排 队",
+      "消费 者的",
+      "消费者 的",
+      "App le",
+      "Ap ple",
+      "▁r oster",
+      "▁ro ster",
+      "▁ros ter",
+      "▁dif fraction",
+      "亮 相",
+      "▁normal ized",
+      "▁normalize d",
+      "▁ normalized",
+      "▁P AR",
+      "▁PA R",
+      "▁ PAR",
+      "▁homeost asis",
+      "Wh o",
+      "W ho",
+      "▁n s",
+      "▁ ns",
+      "ch oose",
+      "cho ose",
+      "▁Class ification",
+      "完善 的",
+      "▁e ighth",
+      "▁eight h",
+      "香 菇",
+      "面积 为",
+      "▁M es",
+      "▁Me s",
+      "▁D rag",
+      "▁Dr ag",
+      "▁ Drag",
+      "▁Pl ugin",
+      "▁ Plugin",
+      "re a",
+      "r ea",
+      "不 具备",
+      "▁e ste",
+      "▁est e",
+      "▁es te",
+      "▁ este",
+      "tr m",
+      "t rm",
+      "▁C lay",
+      "▁Cl ay",
+      "▁c ancellation",
+      "▁cancel lation",
+      "▁cancell ation",
+      "\\' ;",
+      "\\ ';",
+      "未 在",
+      "nW illiam",
+      "nWill iam",
+      "n William",
+      "不小 心",
+      "不 小心",
+      "st orm",
+      "sto rm",
+      "stor m",
+      "▁cle ared",
+      "▁clear ed",
+      "RE MO",
+      "REM O",
+      "▁d ried",
+      "▁dr ied",
+      "▁dri ed",
+      "▁G ov",
+      "▁Go v",
+      "pl oys",
+      "ploy s",
+      "有关 的",
+      "▁c ref",
+      "▁cre f",
+      "▁cr ef",
+      "南 区",
+      "▁aud itory",
+      "▁audit ory",
+      "一 中",
+      "乘 积",
+      "的 真",
+      "▁ 起点网",
+      "he mas",
+      "hem as",
+      "hema s",
+      "h emas",
+      "sort ed",
+      "s orted",
+      "uit ary",
+      "uitar y",
+      "u itary",
+      "▁m yster",
+      "▁my ster",
+      "▁mys ter",
+      "▁myst er",
+      "vas cular",
+      "v ascular",
+      "▁d rum",
+      "▁dr um",
+      "▁t witter",
+      "▁tw itter",
+      "▁ twitter",
+      "Tr igger",
+      "之 心",
+      "off icial",
+      "▁ 多",
+      "誉 为",
+      "首 页",
+      "▁per pet",
+      "剪 辑",
+      "职业 教育",
+      "▁fix ture",
+      "▁ fixture",
+      "nm ain",
+      "n main",
+      "▁Athlet ic",
+      "▁perme ability",
+      "嘉 兴",
+      "市中 心",
+      "市 中心",
+      "针对 性",
+      "Mar t",
+      "Ma rt",
+      "M art",
+      "▁J ama",
+      "▁Jam a",
+      "▁Ja ma",
+      "▁end block",
+      "▁K el",
+      "▁Ke l",
+      "▁ Kel",
+      "zer bai",
+      "两 地",
+      "驾驶 员",
+      "专业 知识",
+      "走 过",
+      "ra tes",
+      "rate s",
+      "rat es",
+      "r ates",
+      "▁eigen values",
+      "▁eigenvalue s",
+      "岩 石",
+      "教 训",
+      "ym an",
+      "yma n",
+      "y man",
+      "im ming",
+      "imm ing",
+      "▁climb ing",
+      "▁clim bing",
+      "л о",
+      "阴 谋",
+      "▁con traction",
+      "▁contract ion",
+      "▁contr action",
+      "Im g",
+      "I mg",
+      "▁alt itude",
+      "做 题",
+      "nt ags",
+      "n tags",
+      "z m",
+      "▁com mod",
+      "▁comm od",
+      "▁ commod",
+      "▁add itive",
+      "> '",
+      "新 车",
+      "▁P late",
+      "▁Pl ate",
+      "▁Plat e",
+      "U i",
+      "的人 们",
+      "的 人们",
+      "评论 区",
+      "▁gam bling",
+      "▁к о",
+      "▁ ко",
+      "tal k",
+      "t alk",
+      "▁forgot ten",
+      "排 出",
+      "Tr y",
+      "T ry",
+      "为主 的",
+      "为 主的",
+      "▁Pro f",
+      "▁Pr of",
+      "grad le",
+      "); //",
+      ") ;//",
+      "t utorial",
+      "run s",
+      "ru ns",
+      "r uns",
+      "但 这",
+      "惩 罚",
+      "CO NT",
+      "CON T",
+      "C ONT",
+      "map per",
+      "m apper",
+      "▁w inds",
+      "▁win ds",
+      "▁wind s",
+      "▁f olding",
+      "▁fol ding",
+      "▁fold ing",
+      "高 位",
+      "▁c harm",
+      "▁ch arm",
+      "▁char m",
+      "▁cha rm",
+      "▁ charm",
+      "收 拾",
+      "▁cer amic",
+      "载 明",
+      "侦 查",
+      "▁heter ogeneity",
+      "▁W or",
+      "▁Wo r",
+      "▁F lex",
+      "▁Fl ex",
+      "▁Fle x",
+      "▁ Flex",
+      "▁ge omet",
+      "▁geo met",
+      "▁geom et",
+      "parse Int",
+      "▁ 女",
+      "在 案",
+      "怎么 说",
+      "▁sim plicity",
+      "▁simpl icity",
+      "▁inter cept",
+      "▁ intercept",
+      "甲状 腺",
+      "▁Par ish",
+      "▁Paris h",
+      "▁Pa rish",
+      "▁l ex",
+      "▁le x",
+      "▁ lex",
+      "自 在",
+      "算 式",
+      "▁P and",
+      "▁Pan d",
+      "▁Pa nd",
+      "▁s har",
+      "▁sh ar",
+      "▁sha r",
+      "▁fl avors",
+      "▁flav ors",
+      "▁flavor s",
+      "弯 曲",
+      "▁y a",
+      "▁ ya",
+      "▁organ ism",
+      "CA R",
+      "C AR",
+      "系统 中",
+      "il age",
+      "ila ge",
+      "决定 的",
+      "决 定的",
+      "可以 让",
+      "un da",
+      "und a",
+      "u nda",
+      "nV ideo",
+      "n Video",
+      "▁implement ations",
+      "▁implementation s",
+      "港 澳",
+      "零 食",
+      "老 家",
+      "Non Null",
+      "架 空",
+      "▁T heme",
+      "▁The me",
+      "▁Th eme",
+      "▁ Theme",
+      "▁seg mentation",
+      "▁segment ation",
+      "▁vol can",
+      "企业 在",
+      "内容 的",
+      "价 钱",
+      "执 恢",
+      "崩 溃",
+      "be y",
+      "b ey",
+      "▁legend ary",
+      "相 差",
+      "结 尾",
+      "t S",
+      "▁j ury",
+      "▁ju ry",
+      "▁jur y",
+      "我 相信",
+      "▁Marg aret",
+      "高 原",
+      "不 做",
+      "推 送",
+      "要求 被执行人",
+      "▁Toy ota",
+      "▁r esting",
+      "▁res ting",
+      "▁rest ing",
+      "B bb",
+      "了 多少",
+      "某某 与被告",
+      "某 某与被告",
+      "▁分 布",
+      "▁ 分布",
+      "半 导体",
+      "▁l awn",
+      "▁law n",
+      "▁la wn",
+      "唿 吸",
+      "肝 脏",
+      "▁n d",
+      "▁ nd",
+      "▁expl oit",
+      "▁explo it",
+      "▁sc hedules",
+      "▁schedul es",
+      "▁schedule s",
+      "外 贸",
+      "高度 重视",
+      "an as",
+      "ana s",
+      "a nas",
+      "min i",
+      "mi ni",
+      "m ini",
+      "▁ Δ",
+      "Do ne",
+      "Don e",
+      "D one",
+      "钢 管",
+      "▁sa lad",
+      "▁sal ad",
+      "不 安",
+      "几个 月",
+      "几 个月",
+      "sa me",
+      "sam e",
+      "s ame",
+      "▁L ip",
+      "▁Li p",
+      "Q R",
+      "▁accel erate",
+      "▁acceler ate",
+      "养 护",
+      "▁ax es",
+      "▁ axes",
+      "re ject",
+      "定 点",
+      "醒 来",
+      "to day",
+      "t oday",
+      "ct rl",
+      "ctr l",
+      "c trl",
+      "▁t ale",
+      "▁tal e",
+      "▁ta le",
+      "▁St orm",
+      "▁ Storm",
+      "小 小",
+      "ff ic",
+      "ffi c",
+      "仲裁 裁决",
+      "is set",
+      "iss et",
+      "isse t",
+      "一 堆",
+      "Un iversity",
+      "St d",
+      "S td",
+      "▁N BC",
+      "▁NB C",
+      "年 限",
+      "▁am big",
+      "▁amb ig",
+      "▁ ambig",
+      "悲 伤",
+      "证据 充分",
+      "▁color ful",
+      "有一 次",
+      "有 一次",
+      "inc ipal",
+      "ju d",
+      "j ud",
+      "我一 直",
+      "我 一直",
+      "▁ 交通",
+      "▁s pa",
+      "▁sp a",
+      "en sen",
+      "ens en",
+      "ense n",
+      "os lav",
+      "▁fin ishes",
+      "▁finish es",
+      "带 上",
+      "生产 任务",
+      "T X",
+      "译 者",
+      "北京 大学",
+      "过 得",
+      "▁A nc",
+      "▁An c",
+      "棉 花",
+      "se p",
+      "s ep",
+      "▁P arker",
+      "▁Par ker",
+      "▁Park er",
+      "▁grad uating",
+      "▁gradu ating",
+      "▁c uisine",
+      "沉 积",
+      "▁C amb",
+      "▁Cam b",
+      "▁Ca mb",
+      "你 去",
+      "▁\\ '\\",
+      "▁\\' \\",
+      "▁ \\'\\",
+      "ri ors",
+      "rior s",
+      "rio rs",
+      "r iors",
+      "揭 示",
+      "ma ven",
+      "m aven",
+      "执 照",
+      "yl ase",
+      "yla se",
+      "▁clean er",
+      "▁Pro cessing",
+      "▁Process ing",
+      "▁Proc essing",
+      "▁ Processing",
+      "k k",
+      "美 学",
+      "定 时",
+      "ST EM",
+      "STE M",
+      "S TEM",
+      "还是 有",
+      "还 是有",
+      "▁editor ial",
+      "▁edit orial",
+      "▁F E",
+      "▁ FE",
+      "am as",
+      "ama s",
+      "a mas",
+      "hy de",
+      "hyd e",
+      "网 民",
+      "▁St anford",
+      "▁Stan ford",
+      "十八 章",
+      "十 八章",
+      "▁k its",
+      "▁kit s",
+      "▁ki ts",
+      "key word",
+      "exp ensive",
+      "电 台",
+      "VI P",
+      "V IP",
+      "业务 的",
+      "选项中 挑",
+      "▁n aming",
+      "▁na ming",
+      "▁nam ing",
+      "▁ naming",
+      "ag ne",
+      "agn e",
+      "户籍 证明",
+      "洗 澡",
+      "ga in",
+      "g ain",
+      "司 令",
+      "▁har sh",
+      "nRe lated",
+      "nRel ated",
+      "n Related",
+      "▁antim icrobial",
+      "保 卫",
+      "▁Sw ift",
+      "▁ Swift",
+      "一 番",
+      "▁S B",
+      "▁ SB",
+      "Bl og",
+      "B log",
+      "Arg uments",
+      "Argument s",
+      "人 身",
+      "ll i",
+      "l li",
+      "urs ion",
+      "▁adoles cent",
+      "而 非",
+      "▁s udo",
+      "▁su do",
+      "▁ sudo",
+      "露 出",
+      "nS how",
+      "nSh ow",
+      "n Show",
+      "set Text",
+      "ke leton",
+      "kel eton",
+      "各样 的",
+      "各 样的",
+      "土 耳",
+      "▁w x",
+      "▁ wx",
+      "▁F oot",
+      "▁Foo t",
+      "▁Fo ot",
+      "En coding",
+      "Enc oding",
+      "室 外",
+      "ross o",
+      "ros so",
+      "pe er",
+      "p eer",
+      "re pr",
+      "rep r",
+      "化 市",
+      "起 的",
+      "奶 牛",
+      "nB r",
+      "n Br",
+      "高 品质",
+      "▁sw itches",
+      "▁switch es",
+      "▁swit ches",
+      "▁admin istrator",
+      "▁administr ator",
+      "▁B arry",
+      "▁Bar ry",
+      "▁Barr y",
+      "▁b etting",
+      "▁bet ting",
+      "前 者",
+      "太 太",
+      "as hing",
+      "ash ing",
+      "ashi ng",
+      "▁Mad ison",
+      "▁Ma dison",
+      "▁Commission er",
+      "▁disp ute",
+      "▁disput e",
+      "T G",
+      "uc hy",
+      "uch y",
+      "u chy",
+      "▁S outheast",
+      "▁South east",
+      "▁ 所",
+      "▁p rol",
+      "▁pro l",
+      "▁pr ol",
+      "片 的",
+      "ist ani",
+      "istan i",
+      "ista ni",
+      "ま す",
+      "am ins",
+      "amin s",
+      "ami ns",
+      "a mins",
+      "▁C rypt",
+      "▁Cry pt",
+      "▁ Crypt",
+      "▁J erry",
+      "▁Jer ry",
+      "古 城",
+      "氨 酸",
+      "直 流",
+      "▁S quad",
+      "▁Squ ad",
+      "▁lin ux",
+      "▁ linux",
+      "肥 胖",
+      "在上 海",
+      "在 上海",
+      "▁jud gment",
+      "▁judg ment",
+      "▁Equ ipment",
+      "▁end points",
+      "▁endpoint s",
+      "▁ endpoints",
+      "▁a neur",
+      "▁an eur",
+      "▁D NS",
+      "▁DN S",
+      "▁ DNS",
+      "▁E ston",
+      "▁Est on",
+      "▁Es ton",
+      "▁f inals",
+      "▁fin als",
+      "▁final s",
+      "▁fi nals",
+      "▁ finals",
+      "tu n",
+      "t un",
+      "qu ire",
+      "quir e",
+      "work ers",
+      "worker s",
+      "wor kers",
+      "▁compass ion",
+      "▁comp elling",
+      "ck e",
+      "c ke",
+      "高 科技",
+      "ustr alia",
+      "ustral ia",
+      "pic ker",
+      "pick er",
+      "p icker",
+      "结束 了",
+      "ac ies",
+      "aci es",
+      "st yled",
+      "style d",
+      "sty led",
+      "styl ed",
+      "▁L arry",
+      "▁Lar ry",
+      "乡村 振兴",
+      "▁Art ist",
+      "▁ Artist",
+      "▁grad uation",
+      "▁gradu ation",
+      "试 用",
+      "▁re activity",
+      "▁react ivity",
+      "最 长",
+      "结 石",
+      "ak y",
+      "a ky",
+      "▁} }</",
+      "▁}} </",
+      "▁ }}</",
+      "ba st",
+      "bas t",
+      "b ast",
+      "for um",
+      "fo rum",
+      "f orum",
+      "▁G race",
+      "▁Gr ace",
+      "▁Gra ce",
+      "▁d rove",
+      "▁dr ove",
+      "▁dro ve",
+      "天 生",
+      "一 手",
+      "▁d onate",
+      "▁don ate",
+      "而 又",
+      "还 可",
+      "fr agment",
+      "fra gment",
+      "frag ment",
+      "bu d",
+      "b ud",
+      "▁na sal",
+      "▁nas al",
+      "▁med iate",
+      "▁media te",
+      "▁medi ate",
+      "▁ mediate",
+      "相 连",
+      "一会 儿",
+      "一 会儿",
+      "▁g low",
+      "▁gl ow",
+      "▁glo w",
+      "▁Group s",
+      "▁Gro ups",
+      "▁ Groups",
+      "▁s ar",
+      "▁sa r",
+      "oi etic",
+      "核 酸",
+      "▁L aura",
+      "▁La ura",
+      "▁Laur a",
+      "晚 饭",
+      "▁dr opping",
+      "▁dro pping",
+      "▁drop ping",
+      "第一百三十三条 之一",
+      "有几 个",
+      "有 几个",
+      "▁di oxide",
+      "▁diox ide",
+      "Gr eat",
+      "G reat",
+      "Pro m",
+      "Pr om",
+      "P rom",
+      "DE R",
+      "D ER",
+      "送达 了",
+      "一般 是",
+      "▁Http Response",
+      "▁ HttpResponse",
+      "ich ia",
+      "ichi a",
+      "▁ 担任职务",
+      "起来 了",
+      "起 来了",
+      "is ha",
+      "ish a",
+      "i sha",
+      "ph yl",
+      "phy l",
+      "▁S ame",
+      "▁Sam e",
+      "▁Sa me",
+      "▁ Same",
+      "ag ents",
+      "agen ts",
+      "agent s",
+      "▁F ields",
+      "▁Field s",
+      "▁ Fields",
+      "▁phen otypes",
+      "▁phenotype s",
+      "d z",
+      "▁g ram",
+      "▁gr am",
+      "▁gra m",
+      "▁ gram",
+      "▁r ocket",
+      "▁rock et",
+      "▁ rocket",
+      "▁ins ufficient",
+      "族自治 州",
+      "dis k",
+      "di sk",
+      "d isk",
+      "如果 说",
+      "go ld",
+      "g old",
+      "乒 乓",
+      "▁engine ered",
+      "▁engineer ed",
+      "再审 申请",
+      "▁sup pressed",
+      "▁suppress ed",
+      "▁Cle aning",
+      "▁Clean ing",
+      "sens or",
+      "s ensor",
+      "这样 一个",
+      "nD ate",
+      "n Date",
+      "▁br ackets",
+      "▁bracket s",
+      "主要 以",
+      "主 要以",
+      "▁town ship",
+      "▁towns hip",
+      "▁ 万",
+      "的一 点",
+      "的 一点",
+      "ear ning",
+      "earn ing",
+      "e arning",
+      "tu be",
+      "tub e",
+      "t ube",
+      "▁r ush",
+      "▁ru sh",
+      "preced ented",
+      "▁accel erated",
+      "▁acceler ated",
+      "▁accelerate d",
+      "W C",
+      "卫 视",
+      "光 盘",
+      "Ex cel",
+      "▁Mar kov",
+      "▁Mark ov",
+      "和中 国",
+      "和 中国",
+      "▁cl arity",
+      "▁clar ity",
+      "▁c lips",
+      "▁cl ips",
+      "▁clip s",
+      "▁cli ps",
+      "约 会",
+      "nr eq",
+      "nre q",
+      "n req",
+      "副 会长",
+      "案发 后",
+      "inc inn",
+      "副 作用",
+      "en osis",
+      "eno sis",
+      "enos is",
+      "▁emer gence",
+      "▁emerg ence",
+      "▁emerge nce",
+      "▁negot iations",
+      "▁negotiation s",
+      "▁ident ities",
+      "▁id entities",
+      "即 时",
+      "视 力",
+      "nP ut",
+      "n Put",
+      "了 两",
+      "着 名",
+      "▁hybrid ization",
+      "J O",
+      "想 过",
+      "由 人民法院裁定",
+      "的女 生",
+      "的 女生",
+      "▁P ink",
+      "▁Pi nk",
+      "▁Pin k",
+      "一 小",
+      "Line ar",
+      "Lin ear",
+      "Li near",
+      "▁al lev",
+      "▁all ev",
+      "▁alle v",
+      "风 水",
+      "下半 年",
+      "下 半年",
+      "▁In side",
+      "▁Ins ide",
+      "Ch anges",
+      "Change s",
+      "Chan ges",
+      "赠 送",
+      "甜 蜜",
+      "预 先",
+      "▁conv inced",
+      "▁convince d",
+      "▁convinc ed",
+      "\\' ));\\",
+      "\\') );\\",
+      "\\')) ;\\",
+      "\\ '));\\",
+      "▁ 方",
+      "uc as",
+      "uca s",
+      "u cas",
+      "▁la par",
+      "▁lap ar",
+      "int endo",
+      "▁sl ices",
+      "▁slic es",
+      "▁slice s",
+      "督 促",
+      "as ha",
+      "ash a",
+      "a sha",
+      "re search",
+      "res earch",
+      "▁F ail",
+      "▁Fa il",
+      "▁ Fail",
+      "单调 递增",
+      "职业 生涯",
+      "短 暂",
+      "Pick er",
+      "Pic ker",
+      "P icker",
+      "▁B inary",
+      "▁Bin ary",
+      "▁ Binary",
+      "▁J ennifer",
+      "▁Jenn ifer",
+      "二 者",
+      "毛 泽东",
+      "re gn",
+      "reg n",
+      "局 限",
+      "套 餐",
+      "领 袖",
+      "调 动",
+      "选出 正确的选项",
+      "▁im agination",
+      "▁imag ination",
+      "白 的",
+      "机构 的",
+      "在内 的",
+      "在 内的",
+      "nK eep",
+      "nKe ep",
+      "n Keep",
+      "L M",
+      "固 体",
+      "自然 的",
+      "自 然的",
+      "▁t empt",
+      "▁tem pt",
+      "▁temp t",
+      "▁ tempt",
+      "▁pe aceful",
+      "▁peace ful",
+      "▁peac eful",
+      "价 位",
+      "▁The ater",
+      "▁X P",
+      "▁ XP",
+      "剥夺政治权利 终身",
+      "里 程",
+      "▁p yr",
+      "▁py r",
+      "N I",
+      "冷 静",
+      "or ton",
+      "ort on",
+      "上 看",
+      "NE W",
+      "N EW",
+      "按 键",
+      "av g",
+      "a vg",
+      "▁B U",
+      "▁ BU",
+      "Pay load",
+      "nB ased",
+      "nBase d",
+      "nBas ed",
+      "n Based",
+      "▁ads orption",
+      "文 中",
+      "▁V ers",
+      "▁Ver s",
+      "▁Ve rs",
+      "▁ Vers",
+      "基 数",
+      "已 破裂",
+      "▁for ums",
+      "▁fo rums",
+      "▁forum s",
+      "Depend ency",
+      "raul ic",
+      "坦 克",
+      "▁ass umes",
+      "▁assum es",
+      "▁assume s",
+      "▁Forward ed",
+      "少 儿",
+      "Ag e",
+      "A ge",
+      "▁in ference",
+      "▁infer ence",
+      "为 我",
+      "▁F ib",
+      "▁Fi b",
+      "以 色",
+      "cw d",
+      "c wd",
+      "to c",
+      "t oc",
+      "ng th",
+      "ep och",
+      "正 如",
+      "el ist",
+      "eli st",
+      "e list",
+      "Tr i",
+      "T ri",
+      "xi a",
+      "x ia",
+      "▁Quant um",
+      "溃 疡",
+      "▁D h",
+      "▁man ages",
+      "▁manag es",
+      "▁manage s",
+      "想 去",
+      "都是 在",
+      "都 是在",
+      "ect in",
+      "ec tin",
+      "独立 的",
+      "▁G len",
+      "▁Gl en",
+      "▁Gle n",
+      "展 望",
+      "usal em",
+      "usa lem",
+      "▁g rip",
+      "▁gr ip",
+      "▁gri p",
+      "▁cr ude",
+      "▁cru de",
+      "▁fr actional",
+      "▁fraction al",
+      "▁fract ional",
+      "身 影",
+      "Cont ains",
+      "触 发",
+      "▁E dd",
+      "▁Ed d",
+      "比较 多",
+      "比 较多",
+      "ch anging",
+      "chan ging",
+      "chang ing",
+      "▁d ye",
+      "▁dy e",
+      "aph yl",
+      "a phyl",
+      "\\\\ $",
+      "▁A str",
+      "▁As tr",
+      "▁Ast r",
+      "▁rom ance",
+      "Sa l",
+      "S al",
+      "▁N g",
+      "▁ Ng",
+      "▁J et",
+      "▁Je t",
+      "▁ Jet",
+      "en erate",
+      "ener ate",
+      "ene rate",
+      "▁l overs",
+      "▁love rs",
+      "▁lo vers",
+      "▁lover s",
+      "久 了",
+      "ID ENT",
+      "IDE NT",
+      "▁F lat",
+      "▁Fl at",
+      "▁ Flat",
+      "▁We ather",
+      "▁ Weather",
+      "▁Ben jamin",
+      "▁Re levant",
+      "夜 间",
+      "达 标",
+      "Es t",
+      "E st",
+      "▁E t",
+      "▁ Et",
+      "▁ 杜",
+      "▁dec iding",
+      "▁Pref ecture",
+      "▁aneur ys",
+      "– \\",
+      "复 数",
+      "tle t",
+      "tl et",
+      "t let",
+      "F X",
+      "`) ;\\",
+      "` );\\",
+      "en coded",
+      "enc oded",
+      "encode d",
+      "andid ates",
+      "andidate s",
+      "andi dates",
+      "▁un stable",
+      "▁uns table",
+      "▁un related",
+      "▁unre lated",
+      "事 宜",
+      "版本 的",
+      "版 本的",
+      "他 会",
+      "▁welcome d",
+      "▁wel comed",
+      "it ate",
+      "ita te",
+      "itat e",
+      "$( \"#",
+      "$(\" #",
+      "$ (\"#",
+      "▁epit helium",
+      ">: :",
+      "> ::",
+      "土耳 其",
+      "▁off ense",
+      "▁a trial",
+      "▁at rial",
+      "整 齐",
+      "em b",
+      "e mb",
+      "id ation",
+      "ida tion",
+      "沙 漠",
+      "▁p eri",
+      "▁per i",
+      "▁pe ri",
+      "▁v ibration",
+      "▁vibr ation",
+      "▁vib ration",
+      "分布 于",
+      "nNo table",
+      "nNot able",
+      "▁prom otional",
+      "▁promot ional",
+      "▁promotion al",
+      "django project",
+      "djang oproject",
+      "we bs",
+      "web s",
+      "▁in quiry",
+      "代表 团",
+      "semb les",
+      "sem bles",
+      "semble s",
+      "sembl es",
+      "脉 冲",
+      "酸 盐",
+      "▁S uz",
+      "▁Su z",
+      "富 含",
+      "sa y",
+      "s ay",
+      "最后 一个",
+      "最后一 个",
+      "Class es",
+      "C lasses",
+      "复 发",
+      "底 线",
+      "捐 赠",
+      "ist oric",
+      "istor ic",
+      "isto ric",
+      "▁M otion",
+      "▁Mot ion",
+      "▁Mo tion",
+      "▁ Motion",
+      "相 继",
+      "▁K ur",
+      "▁Ku r",
+      "非 要",
+      "c x",
+      "▁de emed",
+      "nPerson nel",
+      "的 任务",
+      "l v",
+      "据此 解答",
+      "▁P ri",
+      "▁Pr i",
+      "▁b enz",
+      "▁be nz",
+      "▁ben z",
+      "▁ benz",
+      "ort ium",
+      "▁B att",
+      "▁Ba tt",
+      "▁Bat t",
+      "下 车",
+      "只 好",
+      "▁har ass",
+      "ti e",
+      "t ie",
+      "发 扬",
+      "在 对",
+      "大 同",
+      "难 免",
+      "▁re tin",
+      "▁r etin",
+      "▁ret in",
+      "met h",
+      "me th",
+      "m eth",
+      "oph ila",
+      "ophil a",
+      "▁l ighter",
+      "▁light er",
+      "▁ lighter",
+      "黄 瓜",
+      "监督 管理",
+      "▁d ecode",
+      "▁de code",
+      "▁dec ode",
+      "▁ decode",
+      "形式 的",
+      "形 式的",
+      "nPub lished",
+      "n Published",
+      "从 严",
+      "信息 系统",
+      "▁W A",
+      "▁ WA",
+      "Back end",
+      "他 对",
+      "的 后",
+      "ns h",
+      "n sh",
+      "ti d",
+      "t id",
+      "un do",
+      "und o",
+      "u ndo",
+      "大 酒店",
+      "▁l ately",
+      "▁late ly",
+      "▁lat ely",
+      "隶属 于",
+      "隶 属于",
+      "▁mi R",
+      "少数 民族",
+      "st arter",
+      "start er",
+      "star ter",
+      "车 身",
+      "较 少",
+      "▁prov isions",
+      "▁provision s",
+      "面 膜",
+      "Im ages",
+      "Image s",
+      "一 事",
+      "An n",
+      "A nn",
+      "▁prote cts",
+      "▁prot ects",
+      "▁protect s",
+      "保 管",
+      "ou nge",
+      "oun ge",
+      "o unge",
+      "т и",
+      "原 油",
+      "ac in",
+      "aci n",
+      "a cin",
+      "ment ed",
+      "men ted",
+      "mente d",
+      "m ented",
+      "Re ducer",
+      "Red ucer",
+      "▁appro aching",
+      "▁approach ing",
+      "ol lo",
+      "oll o",
+      "▁b anner",
+      "▁ban ner",
+      "▁ banner",
+      "▁st rand",
+      "▁str and",
+      "▁ strand",
+      "兑 换",
+      "Ra y",
+      "R ay",
+      "DE LETE",
+      "磨 损",
+      "▁re ign",
+      "▁label ing",
+      "▁lab eling",
+      "ry s",
+      "r ys",
+      "▁H un",
+      "▁Hu n",
+      "ir ements",
+      "ire ments",
+      "irement s",
+      "一 首",
+      "机械 设备",
+      "▁h oped",
+      "▁hope d",
+      "▁hop ed",
+      "▁ho ped",
+      "” [",
+      "病 因",
+      "V T",
+      "冷 水",
+      "衔 接",
+      "关注 的",
+      "教育 教学",
+      "年 获",
+      "▁concent rate",
+      "▁concentr ate",
+      "▁act ed",
+      "▁ac ted",
+      "▁ acted",
+      "判决确定 之日起计算",
+      "MA N",
+      "M AN",
+      "side bar",
+      "造 价",
+      "▁M aj",
+      "▁Ma j",
+      "第一 步",
+      "第 一步",
+      "da n",
+      "d an",
+      "Rad ius",
+      "art ist",
+      "doctor al",
+      "▁system atically",
+      "▁systematic ally",
+      "▁p hon",
+      "▁ph on",
+      "▁ phon",
+      "分 级",
+      "吉 祥",
+      "四个 月",
+      "四 个月",
+      "代数 式",
+      "的 倍数",
+      "▁di ets",
+      "▁die ts",
+      "▁diet s",
+      "▁le isure",
+      "▁Furn iture",
+      "双 眼",
+      "无 期徒刑",
+      "▁Ap ply",
+      "▁App ly",
+      "▁Appl y",
+      "▁ Apply",
+      "▁controvers y",
+      "无 意",
+      "填 充",
+      "等 你",
+      "▁di agrams",
+      "▁diagram s",
+      "▁dia grams",
+      "可以 去",
+      "▁organ izing",
+      "incinn ati",
+      "风 流",
+      "▁, \\",
+      "▁ ,\\",
+      "nTrans port",
+      "n Transport",
+      "bl ind",
+      "▁Rob in",
+      "▁Ro bin",
+      "Te mp",
+      "Tem p",
+      "T emp",
+      "jack son",
+      "▁avoid ed",
+      "▁v iability",
+      "▁vi ability",
+      "▁trans forming",
+      "▁transform ing",
+      "▁Com ments",
+      "▁Comment s",
+      "▁Comm ents",
+      "▁ Comments",
+      "履行 生效法律文书确定的义务",
+      "履行生效法律文书 确定的义务",
+      "住 在",
+      "部分 的",
+      "部 分的",
+      "▁m M",
+      "▁L uke",
+      "▁Lu ke",
+      "▁Luk e",
+      "▁H omes",
+      "▁Home s",
+      "▁Hom es",
+      "▁Ho mes",
+      "Data Source",
+      "剖 析",
+      "go m",
+      "g om",
+      "▁R ust",
+      "▁Ru st",
+      "▁Rus t",
+      "▁spons or",
+      "原 件",
+      "▁p all",
+      "▁pal l",
+      "▁pa ll",
+      "患 儿",
+      "秋 天",
+      "tn ame",
+      "t name",
+      "▁def ending",
+      "▁defend ing",
+      "NA B",
+      "N AB",
+      "ph ant",
+      "pha nt",
+      "phan t",
+      "## #\\",
+      "### \\",
+      "# ##\\",
+      "EG IN",
+      "在法定 期限内",
+      "▁predict or",
+      "▁pred ictor",
+      "▁Christ ians",
+      "▁Christian s",
+      "▁Invest ment",
+      "不 尽",
+      "cm s",
+      "c ms",
+      "没 必要",
+      "给 孩子",
+      "我不 是",
+      "我 不是",
+      "isc iplinary",
+      "浓 郁",
+      "▁所 以",
+      "▁ 所以",
+      "▁e ngr",
+      "▁en gr",
+      "▁eng r",
+      "▁fun eral",
+      "▁J our",
+      "▁Jo ur",
+      "▁c ows",
+      "▁co ws",
+      "▁cow s",
+      "re ading",
+      "read ing",
+      "rea ding",
+      "▁Mich elle",
+      "▁Michel le",
+      "▁ge ographic",
+      "▁geo graphic",
+      "leuk in",
+      "企业 和",
+      "▁Ber keley",
+      "▁\\' {",
+      "▁ \\'{",
+      "鲜 花",
+      "的 味道",
+      "信 访",
+      "▁h ints",
+      "▁hint s",
+      "tn ew",
+      "t new",
+      "▁nut rient",
+      "▁nutr ient",
+      "读 取",
+      "▁Me et",
+      "▁ Meet",
+      "▁de notes",
+      "▁den otes",
+      "▁denote s",
+      "年 华",
+      "▁设 施",
+      "▁ 设施",
+      "▁par ental",
+      "▁parent al",
+      "▁paren tal",
+      "( <",
+      "大 方",
+      "发表于 起点网",
+      "▁incorpor ation",
+      "应 诉",
+      "薪 资",
+      "▁c ob",
+      "▁co b",
+      "▁ cob",
+      "▁a rena",
+      "▁are na",
+      "▁ar ena",
+      "▁aren a",
+      "车 牌",
+      "辞 典",
+      "的 支持",
+      "so ck",
+      "s ock",
+      "st ri",
+      "str i",
+      "s tri",
+      "▁jud g",
+      "▁ju dg",
+      "商 会",
+      "str ategy",
+      "u u",
+      "产 值",
+      "支 架",
+      "nTh ose",
+      "▁d ressing",
+      "▁dr essing",
+      "▁dress ing",
+      "召 唤",
+      "bi d",
+      "b id",
+      "cm p",
+      "c mp",
+      "分析 过程",
+      "▁un ity",
+      "▁unit y",
+      "▁uni ty",
+      "▁ unity",
+      "▁re med",
+      "▁r emed",
+      "▁rem ed",
+      "▁N ur",
+      "▁Nu r",
+      "▁Re comm",
+      "▁Rec omm",
+      "▁ Recomm",
+      "▁b race",
+      "▁br ace",
+      "▁bra ce",
+      "▁ brace",
+      "vi t",
+      "v it",
+      "iv ia",
+      "ivi a",
+      "i via",
+      "▁v isa",
+      "▁vis a",
+      "▁vi sa",
+      "ya ng",
+      "yan g",
+      "y ang",
+      "就 去",
+      "践 行",
+      "建设 和",
+      "▁im posed",
+      "▁imp osed",
+      "▁impose d",
+      "▁occup ation",
+      "▁ex clude",
+      "▁ exclude",
+      "亲 属",
+      "心中 的",
+      "心 中的",
+      "ar ette",
+      "are tte",
+      "aret te",
+      "▁ang ry",
+      "▁in trig",
+      "▁int rig",
+      "▁intr ig",
+      "▁Ben gal",
+      "▁Beng al",
+      "▁ 明",
+      "纠 正",
+      "▁cons uming",
+      "▁consum ing",
+      "▁ consuming",
+      "ro chemical",
+      "roc hemical",
+      "r ochemical",
+      "性 与",
+      "持 久",
+      "$( \\'#",
+      "$(\\' #",
+      "$ (\\'#",
+      "加 减",
+      "▁la undry",
+      ": ]",
+      "R G",
+      "但 由于",
+      "mu st",
+      "mus t",
+      "m ust",
+      "file Name",
+      "▁Comp uting",
+      "▁Comput ing",
+      "的 品牌",
+      "注 明",
+      "▁bow el",
+      "▁L ength",
+      "▁Le ngth",
+      "▁ Length",
+      "▁Ex ercise",
+      "信 赖",
+      "te cho",
+      "tech o",
+      "tec ho",
+      "t echo",
+      "oph osph",
+      "o phosph",
+      "tic k",
+      "ti ck",
+      "t ick",
+      "就 说",
+      "Con straint",
+      "▁traff icking",
+      "▁traffic king",
+      "落 在",
+      "pos ium",
+      "▁girl friend",
+      "这 套",
+      "AI N",
+      "A IN",
+      "▁\" {",
+      "▁ \"{",
+      "▁Expl orer",
+      "▁Explore r",
+      "▁ Explorer",
+      "金融 借款合同纠纷一案",
+      "▁phys iology",
+      "▁physi ology",
+      "▁pr iest",
+      "▁pri est",
+      "COL OR",
+      "▁c ortic",
+      "▁cor tic",
+      "▁cort ic",
+      "UT O",
+      "U TO",
+      "▁a wk",
+      "▁aw k",
+      "▁ awk",
+      "▁gran ul",
+      "Ca t",
+      "C at",
+      "本案 的",
+      "▁F IFA",
+      "▁FI FA",
+      "▁L uis",
+      "▁Lu is",
+      "il iates",
+      "ili ates",
+      "ilia tes",
+      "iliate s",
+      "Per iod",
+      "nd ecl",
+      "nde cl",
+      "ndec l",
+      "n decl",
+      "Na N",
+      "▁accept s",
+      "领导 小组",
+      "在这 种",
+      "在 这种",
+      "字 体",
+      "当 我",
+      "▁coord inated",
+      "▁coordinate d",
+      "▁g a",
+      "▁ ga",
+      "大概 率",
+      "大 概率",
+      "nf i",
+      "n fi",
+      "eb ra",
+      "e bra",
+      "▁neighbour hood",
+      "($ _",
+      "( $_",
+      "do es",
+      "d oes",
+      "有 利",
+      "饮 水",
+      "▁C ad",
+      "▁Ca d",
+      "▁Hel per",
+      "▁Help er",
+      "▁ Helper",
+      "恰 好",
+      "高 三",
+      "▁l umin",
+      "▁lum in",
+      "▁lu min",
+      "石 化",
+      "制 药",
+      "树 的",
+      "诉 前",
+      "▁arg ues",
+      "▁argue s",
+      "▁argu es",
+      "做 得",
+      "冲 洗",
+      "行 使",
+      "de hyde",
+      "▁r efined",
+      "▁ref ined",
+      "▁refin ed",
+      "▁refine d",
+      "▁nom ination",
+      "▁nomin ation",
+      "履行 债务",
+      "ac ious",
+      "aci ous",
+      "acio us",
+      "▁n iche",
+      "▁nic he",
+      "▁ni che",
+      "▁t ubes",
+      "▁tub es",
+      "▁tube s",
+      "▁tu bes",
+      "对 的",
+      "入住 时间",
+      "入住时 间",
+      "▁m ell",
+      "▁me ll",
+      "▁mel l",
+      "当 成",
+      "{{ {",
+      "{ {{",
+      "▁O NE",
+      "▁ON E",
+      "▁ ONE",
+      "▁a eros",
+      "▁aer os",
+      "uclide an",
+      "uclid ean",
+      "▁dem ocracy",
+      "▁democr acy",
+      "▁T IME",
+      "▁TI ME",
+      "▁ TIME",
+      "▁access ibility",
+      "书 的",
+      "古 人",
+      "向 右",
+      "离 职",
+      "en i",
+      "e ni",
+      "在 前",
+      "原 材料",
+      "▁F ri",
+      "▁Fr i",
+      "▁N ON",
+      "▁NO N",
+      "▁ NON",
+      "▁t ears",
+      "▁te ars",
+      "▁tea rs",
+      "▁tear s",
+      "山 的",
+      "菩 萨",
+      "新生 儿",
+      "ind ices",
+      "indi ces",
+      "▁w arming",
+      "▁war ming",
+      "▁warm ing",
+      "UP DATE",
+      "U PDATE",
+      "▁з а",
+      "▁ за",
+      "▁com mut",
+      "▁comm ut",
+      "nEx planation",
+      "很 小",
+      "雕 塑",
+      "民 俗",
+      "点 到",
+      "▁s urre",
+      "▁sur re",
+      "▁Ac cept",
+      "▁ Accept",
+      "悠 悠",
+      "答 题",
+      "el ong",
+      "elo ng",
+      "e long",
+      "▁Cl inic",
+      "▁Clin ic",
+      "农 历",
+      "oj o",
+      "o jo",
+      "一下 子",
+      "的 认识",
+      "自 家",
+      "可以 从轻处罚",
+      "可以从 轻处罚",
+      "宣传 部",
+      "防 晒",
+      "▁] ;",
+      "▁ ];",
+      "▁I gn",
+      "▁Ig n",
+      "▁ Ign",
+      "dot net",
+      "gu e",
+      "g ue",
+      "nC he",
+      "nCh e",
+      "n Che",
+      "▁作品 名称",
+      "▁n ortheast",
+      "▁north east",
+      "门 票",
+      "申请 撤回起诉",
+      "申请撤回 起诉",
+      "信 托",
+      "▁人物 经历",
+      "▁ball oon",
+      "sol ution",
+      "s olution",
+      "东南 亚",
+      "ne ath",
+      "nea th",
+      "▁phys ic",
+      "▁physi c",
+      "特 尔",
+      "▁B illy",
+      "▁Bill y",
+      "▁Bil ly",
+      "T i",
+      "▁j a",
+      "▁ ja",
+      "met er",
+      "me ter",
+      "m eter",
+      "毛 病",
+      "慰 问",
+      "▁B in",
+      "▁Bi n",
+      "▁ Bin",
+      "wr ight",
+      "w right",
+      "▁相关 条款",
+      "man ual",
+      "检 公诉",
+      "SE O",
+      "S EO",
+      "我国 的",
+      "我 国的",
+      "观察 组",
+      "经济 增长",
+      "▁prem ises",
+      "▁premise s",
+      "aa c",
+      "a ac",
+      "▁so fa",
+      "▁acc ent",
+      "▁ac cent",
+      "扣 划",
+      "▁c ocon",
+      "▁co con",
+      "▁coc on",
+      "清华 大学",
+      "on Change",
+      "o nChange",
+      "▁tr aded",
+      "▁trad ed",
+      "▁tra ded",
+      "▁trade d",
+      "合 一",
+      "▁g loss",
+      "▁gl oss",
+      "▁glo ss",
+      "▁ gloss",
+      "rec ords",
+      "record s",
+      "▁ 东",
+      "其中 有",
+      "其 中有",
+      "▁m ush",
+      "▁mus h",
+      "▁mu sh",
+      "▁s napshot",
+      "▁snap shot",
+      "▁ snapshot",
+      "▁com positions",
+      "▁compos itions",
+      "▁composition s",
+      "Se q",
+      "S eq",
+      "nSt ate",
+      "nSta te",
+      "nStat e",
+      "n State",
+      "▁t ricks",
+      "▁tr icks",
+      "▁tri cks",
+      "▁trick s",
+      "▁sem ester",
+      "当 局",
+      "▁Ev idence",
+      "▁kil ometers",
+      "▁kilomet ers",
+      "第五 十二条",
+      "第五十 二条",
+      "国际 化",
+      "▁bel oved",
+      "ST ATE",
+      "STAT E",
+      "STA TE",
+      "▁R S",
+      "▁ RS",
+      "举行 的",
+      "举 行的",
+      "▁conv icted",
+      "▁integr als",
+      "▁integral s",
+      "分 组",
+      "▁Wat son",
+      "瑞 典",
+      "ight ing",
+      "igh ting",
+      "▁lect ures",
+      "▁lecture s",
+      "▁ 四",
+      "▁v apor",
+      "▁va por",
+      "▁vap or",
+      "q i",
+      "昆 虫",
+      "媒 介",
+      "On ce",
+      "O nce",
+      "了解 一下",
+      "▁all ergic",
+      "▁allerg ic",
+      "▁out lets",
+      "▁outlet s",
+      "走 访",
+      "to pics",
+      "top ics",
+      "topic s",
+      "▁wa ist",
+      "夫 妇",
+      "就 连",
+      "nOr gan",
+      "n Organ",
+      "开 花",
+      "LA Y",
+      "L AY",
+      "Trans fer",
+      "In c",
+      "I nc",
+      "▁m aj",
+      "▁ma j",
+      "▁pr izes",
+      "▁prize s",
+      "▁pri zes",
+      "▁priz es",
+      "是 被",
+      "▁fil mm",
+      "▁film m",
+      "友 情",
+      "我们 需要",
+      "; (",
+      "决定 了",
+      "决 定了",
+      "他们 还",
+      "探讨 了",
+      "lic t",
+      "li ct",
+      "l ict",
+      "日 登记结婚",
+      "仍 有",
+      "向 东",
+      "▁L ions",
+      "▁Li ons",
+      "▁Lion s",
+      "命 中",
+      "▁cont rollers",
+      "▁control lers",
+      "▁controller s",
+      "▁ controllers",
+      "面 料",
+      "宪 法",
+      "▁J ulia",
+      "▁Jul ia",
+      "▁Ju lia",
+      "▁t i",
+      "▁ ti",
+      "▁f avorites",
+      "▁favor ites",
+      "▁favorite s",
+      "太 过",
+      "拒 不",
+      "本 周",
+      "▁h ed",
+      "▁he d",
+      "▁ hed",
+      "大 全",
+      "\\\\ \",",
+      "\\\\\" ,",
+      "og lob",
+      "o glob",
+      "▁al gebras",
+      "▁algebra s",
+      "▁ algebras",
+      "痕 迹",
+      "羊 肉",
+      "也可 能",
+      "也 可能",
+      "▁motor cycle",
+      "▁interview ed",
+      "走 路",
+      "nS ign",
+      "n Sign",
+      "▁Ch ron",
+      "▁ess ays",
+      "▁essay s",
+      "▁r amp",
+      "▁ra mp",
+      "▁ram p",
+      "ocyt osis",
+      "共 建",
+      "近 几年",
+      "it atively",
+      "itative ly",
+      "itat ively",
+      "er mark",
+      "erm ark",
+      "▁position ed",
+      "▁posit ioned",
+      "绝 大多数",
+      "▁time d",
+      "▁tim ed",
+      "▁ti med",
+      "▁ timed",
+      "fe ld",
+      "f eld",
+      "ra ham",
+      "rah am",
+      "每一 次",
+      "每 一次",
+      "d v",
+      "洗衣 机",
+      "▁▁ ▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁ ▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁",
+      "▁▁▁▁▁▁ ▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁",
+      "▁▁▁ ▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁ ▁▁▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁ ▁▁▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁",
+      "▁▁▁▁▁▁▁▁▁▁▁▁▁ ▁▁",
+      "▁ ▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "▁c ele",
+      "▁ce le",
+      "▁cel e",
+      "▁ cele",
+      "众所 周知",
+      "list ener",
+      "listen er",
+      "▁W HO",
+      "▁WH O",
+      "▁ WHO",
+      "▁apprec iation",
+      "▁G Hz",
+      "▁GH z",
+      "▁ GHz",
+      "▁trans duction",
+      "充分 利用",
+      "将 成为",
+      "▁Rob erts",
+      "▁Robert s",
+      "▁Rober ts",
+      "▁Emer itus",
+      "▁gr avitational",
+      "很 喜欢",
+      "▁E nh",
+      "▁En h",
+      "▁ Enh",
+      "▁R ather",
+      "▁Ra ther",
+      "▁Rat her",
+      "▁pur ely",
+      "▁pure ly",
+      "▁m ailing",
+      "▁mail ing",
+      "▁ma iling",
+      "开 封",
+      "必 定",
+      "▁Ch ain",
+      "▁Cha in",
+      "▁ Chain",
+      "成立 了",
+      "强烈 的",
+      "强 烈的",
+      "秋 风",
+      "▁obst acles",
+      "▁obstacle s",
+      "y c",
+      "奥 斯",
+      "▁ap pet",
+      "▁app et",
+      "▁appe t",
+      "▁% >",
+      "en gu",
+      "eng u",
+      "马来 西亚",
+      "▁Ti gers",
+      "▁Tig ers",
+      "▁Tiger s",
+      "大 海",
+      "清 代",
+      "ar ently",
+      "arent ly",
+      "dig ital",
+      "digit al",
+      "d igital",
+      "▁L uck",
+      "▁Luc k",
+      "▁Lu ck",
+      "op oietic",
+      "▁d ressed",
+      "▁dr essed",
+      "▁dress ed",
+      "老 大",
+      "打 卡",
+      "空 军",
+      "老 鼠",
+      "it ary",
+      "itar y",
+      "ita ry",
+      "▁hon our",
+      "▁ho nour",
+      "act erial",
+      "acter ial",
+      "acteria l",
+      "Ty p",
+      "T yp",
+      "总 书记",
+      "对该 犯",
+      "对 该犯",
+      "▁pract icing",
+      "fa r",
+      "f ar",
+      "oc ular",
+      "o cular",
+      "▁R ivers",
+      "▁River s",
+      "▁Ri vers",
+      "▁Riv ers",
+      "男 的",
+      "write rs",
+      "writ ers",
+      "writer s",
+      "man age",
+      "▁b ean",
+      "▁be an",
+      "▁ bean",
+      "fix ture",
+      "▁K irk",
+      "▁Kir k",
+      "▁Ki rk",
+      "▁sup plements",
+      "▁supplement s",
+      "▁suppl ements",
+      "nP S",
+      "n PS",
+      "el les",
+      "ell es",
+      "elle s",
+      "pr essed",
+      "press ed",
+      "pres sed",
+      "p ressed",
+      "年 第",
+      "▁acc essing",
+      "▁ac cessing",
+      "▁access ing",
+      "街 头",
+      "了 被执行人",
+      "▁P B",
+      "▁ PB",
+      "▁Democr at",
+      "< \\",
+      "卖 出",
+      "Com mands",
+      "Command s",
+      "Comm ands",
+      "nV ery",
+      "nVer y",
+      "n Very",
+      "▁fl owing",
+      "▁flo wing",
+      "▁flow ing",
+      "▁friend ship",
+      "▁friends hip",
+      "IP S",
+      "I PS",
+      "}} ^{",
+      "} }^{",
+      "▁r gb",
+      "▁ rgb",
+      "▁Bud dh",
+      "▁Budd h",
+      "研究生 院",
+      "to uch",
+      "t ouch",
+      "▁Ex tract",
+      "▁Extra ct",
+      "▁Extr act",
+      "▁ Extract",
+      "▁I celand",
+      "▁Ice land",
+      "时 长",
+      "论 证",
+      "CL ASS",
+      "CLA SS",
+      "CLAS S",
+      "▁init iate",
+      "▁initi ate",
+      "tu b",
+      "t ub",
+      "▁co operative",
+      "▁cooper ative",
+      "File Path",
+      "老 虎",
+      "给出 了",
+      "给 出了",
+      "▁Chel sea",
+      "▁HOLD ERS",
+      "▁fund ament",
+      "市 中",
+      "监 理",
+      "So und",
+      "S ound",
+      "本院 依法适用简易程序",
+      "本院依法 适用简易程序",
+      "ec om",
+      "eco m",
+      "e com",
+      "▁through put",
+      "▁ throughput",
+      "em ade",
+      "ema de",
+      "e made",
+      "gate way",
+      "g ateway",
+      "瓷 砖",
+      "fa ster",
+      "fast er",
+      "fas ter",
+      "f aster",
+      "▁hon ored",
+      "▁honor ed",
+      "▁Ex pected",
+      "▁Exp ected",
+      "▁Expect ed",
+      "▁ Expected",
+      "▁) :\\",
+      "▁ ):\\",
+      "硕士 学位",
+      "▁拉 丁",
+      "▁ 拉丁",
+      "az ole",
+      "azol e",
+      "azo le",
+      "ecl ipse",
+      "e clipse",
+      "科 室",
+      "一 律",
+      "想 说",
+      "AV A",
+      "A VA",
+      "医 科大学",
+      "可以 根据",
+      "乘 以",
+      "ig ating",
+      "iga ting",
+      "er get",
+      "erg et",
+      "差 点",
+      "贯 穿",
+      "K O",
+      "▁s pat",
+      "▁sp at",
+      "▁spa t",
+      "看 你",
+      "的比 例",
+      "的 比例",
+      "ced ure",
+      "▁F isher",
+      "▁Fish er",
+      "▁wh ites",
+      "▁white s",
+      "▁whit es",
+      "中 途",
+      "我也 是",
+      "我 也是",
+      "▁Not ice",
+      "▁ Notice",
+      "怪 物",
+      "▁St ay",
+      "▁Sta y",
+      "co verage",
+      "cover age",
+      "So uth",
+      "S outh",
+      "▁abs orb",
+      "▁absor b",
+      "ga mes",
+      "game s",
+      "gam es",
+      "g ames",
+      "Lo w",
+      "L ow",
+      "洋 葱",
+      "▁U ntil",
+      "▁Un til",
+      "▁ Until",
+      "▁Bern ard",
+      "▁advent ures",
+      "▁adventure s",
+      "叔 叔",
+      "报 表",
+      "es sa",
+      "ess a",
+      "农村 经济",
+      "Des ktop",
+      "大 米",
+      "du ino",
+      "区人民检察院 以",
+      "区 人民检察院以",
+      "▁V arious",
+      "▁Vari ous",
+      "▁Var ious",
+      "▁appropri ately",
+      "▁appropriate ly",
+      "生 殖",
+      "想 的",
+      "im m",
+      "i mm",
+      "re fs",
+      "ref s",
+      "▁R EST",
+      "▁RE ST",
+      "▁RES T",
+      "▁ REST",
+      "w i",
+      "▁P rep",
+      "▁Pr ep",
+      "▁Pre p",
+      "▁ Prep",
+      "▁G ithub",
+      "▁Git hub",
+      "▁ Github",
+      "▁P icture",
+      "▁Pic ture",
+      "▁ Picture",
+      "DateTime Field",
+      "nP aul",
+      "n Paul",
+      "不 来",
+      "人为 本",
+      "人 为本",
+      "▁R an",
+      "▁Ra n",
+      "▁athlet ic",
+      "价 款",
+      "已经 在",
+      "ph antom",
+      "phant om",
+      "phan tom",
+      "▁B aptist",
+      "▁i ds",
+      "▁id s",
+      "▁ ids",
+      "hand lers",
+      "handle rs",
+      "handler s",
+      "▁Reg ular",
+      "▁ Regular",
+      "▁Ex tra",
+      "▁Ext ra",
+      "▁Extr a",
+      "▁ Extra",
+      "M a",
+      "天 上",
+      "▁Equ ation",
+      "▁Eq uation",
+      "退 货",
+      "NU M",
+      "N UM",
+      "▁c af",
+      "▁ca f",
+      "管理 员",
+      "quant ity",
+      "已经 有",
+      "的 实现",
+      "▁Aust rian",
+      "▁Austria n",
+      "哪 种",
+      "所 求",
+      "用到 的",
+      "用 到的",
+      "view port",
+      "▁hyp oxia",
+      "▁hypox ia",
+      "▁guarant ees",
+      "▁guarantee s",
+      "ew ard",
+      "e ward",
+      "sign ature",
+      "▁P il",
+      "▁Pi l",
+      "还是 比较",
+      "还 是比较",
+      "ick ed",
+      "W P",
+      "▁ 冯",
+      "Line s",
+      "Lin es",
+      "Li nes",
+      "L ines",
+      "▁Agricult ure",
+      "我 最",
+      "缺 失",
+      "以色 列",
+      "xi t",
+      "x it",
+      "结 晶",
+      "副总 经理",
+      "副 总经理",
+      "▁st oring",
+      "▁sto ring",
+      "▁N M",
+      "▁ NM",
+      "Run ning",
+      "R unning",
+      "▁t x",
+      "▁ tx",
+      "▁\\\\ (\\\\",
+      "▁\\\\( \\\\",
+      "▁ \\\\(\\\\",
+      "海 岸",
+      "▁tr ainer",
+      "▁train er",
+      "▁tra iner",
+      "渲 染",
+      "文 案",
+      "这些 都是",
+      "uss els",
+      "他 用",
+      "ro be",
+      "rob e",
+      "r obe",
+      "ca d",
+      "c ad",
+      "再 见",
+      "▁b urg",
+      "▁bu rg",
+      "▁bur g",
+      "▁ burg",
+      "▁D ifferential",
+      "▁Different ial",
+      "▁sur prisingly",
+      "▁surprising ly",
+      "▁un precedented",
+      "Sm all",
+      "S mall",
+      "外 资",
+      "方法 和",
+      "方 法和",
+      "▁H arm",
+      "▁Har m",
+      "▁Ha rm",
+      "▁ Harm",
+      "▁ev olving",
+      "us ted",
+      "ust ed",
+      "u sted",
+      "▁P oints",
+      "▁Point s",
+      "▁Po ints",
+      "▁ Points",
+      "Si l",
+      "S il",
+      "▁qual ification",
+      "外 地",
+      "失 眠",
+      "▁Up dated",
+      "▁Update d",
+      "▁ Updated",
+      "上 映",
+      "NO DE",
+      "▁N ar",
+      "▁Na r",
+      "ac counts",
+      "account s",
+      "acco unts",
+      "▁draw ings",
+      "▁drawing s",
+      "] >",
+      "CO NF",
+      "CON F",
+      "的 艺术",
+      "执行 条件",
+      "% ~",
+      "%% %%",
+      "The n",
+      "Th en",
+      "T hen",
+      "st aff",
+      "sta ff",
+      "轮 回",
+      "uv ant",
+      "▁P lot",
+      "▁Pl ot",
+      "▁ Plot",
+      "ц и",
+      "nJ apanese",
+      "n Japanese",
+      "二个 月",
+      "二 个月",
+      "▁p ressed",
+      "▁pr essed",
+      "▁pres sed",
+      "▁press ed",
+      "▁ pressed",
+      "号 案",
+      "ven ile",
+      "督 导",
+      "注意 力",
+      "▁P olitical",
+      "▁Polit ical",
+      "▁spec imen",
+      "亚 马",
+      "的一 般",
+      "的 一般",
+      "表示 为",
+      "ij i",
+      "i ji",
+      "Fa st",
+      "F ast",
+      "▁b uses",
+      "▁bu ses",
+      "▁bus es",
+      "▁al loy",
+      "▁all oy",
+      "▁exam inations",
+      "▁examination s",
+      "▁refer ral",
+      "▁Hand le",
+      "▁ Handle",
+      "诊 疗",
+      "儿 女",
+      "▁Cons umer",
+      "▁ Consumer",
+      "▁ex citation",
+      "▁exc itation",
+      "▁excit ation",
+      "不 愿",
+      "SU V",
+      "S UV",
+      "ter min",
+      "term in",
+      "IS S",
+      "I SS",
+      "arr ass",
+      "arra ss",
+      "▁S urgery",
+      "零 点",
+      "▁med als",
+      "▁medal s",
+      "▁pro tests",
+      "▁prote sts",
+      "▁prot ests",
+      "▁protest s",
+      "脑 袋",
+      "el p",
+      "e lp",
+      "选项 错误",
+      "本院于 同",
+      "▁fibr oblasts",
+      "▁fibro blasts",
+      "▁fibroblast s",
+      "▁Hol iday",
+      "▁c oil",
+      "▁co il",
+      "▁l ungs",
+      "▁lung s",
+      "aut iful",
+      "科 员",
+      "ra mid",
+      "ram id",
+      "▁S chedule",
+      "▁Sc hedule",
+      "▁Sche dule",
+      "▁ Schedule",
+      "的 意见",
+      "▁tempor arily",
+      "nW atch",
+      "n Watch",
+      "积极 性",
+      "等 因素",
+      "on ucle",
+      "o nucle",
+      "▁Observ able",
+      "▁ Observable",
+      "nA v",
+      "n Av",
+      "▁d ots",
+      "▁do ts",
+      "▁dot s",
+      "▁ dots",
+      "条第一款 之规定",
+      "vis or",
+      "v isor",
+      "ul monary",
+      "▁proceed s",
+      "▁proc eeds",
+      "▁an notation",
+      "▁ann otation",
+      "▁annot ation",
+      "▁ annotation",
+      "七 八",
+      "▁G ate",
+      "▁Ga te",
+      "▁Gat e",
+      "▁ Gate",
+      "▁not ified",
+      "▁Comp etition",
+      "▁Compet ition",
+      "离 家",
+      "办公 厅",
+      "▁the ft",
+      "▁th eft",
+      "y u",
+      "事 先",
+      "简 化",
+      "▁p erc",
+      "▁per c",
+      "▁pe rc",
+      "定 向",
+      "▁ab err",
+      "▁c ores",
+      "▁co res",
+      "▁cor es",
+      "▁core s",
+      "▁ cores",
+      "▁T A",
+      "▁ TA",
+      "▁sy nerg",
+      "▁syn erg",
+      "篇 小说",
+      "▁ob sc",
+      "▁obs c",
+      "estab lish",
+      "ver ts",
+      "vert s",
+      "v erts",
+      "nS chool",
+      "nSch ool",
+      "n School",
+      "▁sound track",
+      "的 体积",
+      "的 理论",
+      "研究 方向",
+      "艰 苦",
+      "▁ √",
+      "id ding",
+      "idd ing",
+      "和 一个",
+      "▁Or lando",
+      "长 久",
+      "▁S olid",
+      "▁Sol id",
+      "▁ Solid",
+      "整 天",
+      "十九 章",
+      "十 九章",
+      "下 游",
+      "味 的",
+      "将 继续",
+      "pos ure",
+      "p osure",
+      "▁c andy",
+      "▁can dy",
+      "▁cand y",
+      "Des cribe",
+      "De scribe",
+      "Desc ribe",
+      "fr eq",
+      "fre q",
+      "f req",
+      "▁sp otted",
+      "▁spot ted",
+      "or um",
+      "o rum",
+      "▁H az",
+      "▁Ha z",
+      "des ktop",
+      "desk top",
+      "信用卡 中心",
+      "pr ising",
+      "pri sing",
+      "▁h ierarchy",
+      "▁hierarch y",
+      "射 线",
+      "ha ll",
+      "hal l",
+      "h all",
+      "共同 生活",
+      "iev able",
+      "分 明",
+      "▁direct ories",
+      "▁director ies",
+      "▁ directories",
+      "使用 了",
+      "使 用了",
+      "nM ichael",
+      "n Michael",
+      "▁over flow",
+      "▁ overflow",
+      "▁cloud s",
+      "▁world s",
+      "▁Tur ner",
+      "▁Turn er",
+      "▁G enerate",
+      "▁Gener ate",
+      "▁Gen erate",
+      "▁Gene rate",
+      "▁ Generate",
+      "bi b",
+      "b ib",
+      "很高 的",
+      "很 高的",
+      "的 用户",
+      "Count ry",
+      "Cou ntry",
+      "rie ved",
+      "riev ed",
+      "rieve d",
+      "r ieved",
+      "买 到",
+      "en ton",
+      "ent on",
+      "ento n",
+      "e nton",
+      "it ten",
+      "itt en",
+      "SU LT",
+      "SUL T",
+      "S ULT",
+      "▁p unch",
+      "▁pun ch",
+      "▁pu nch",
+      "▁en larg",
+      "▁Ac ademic",
+      "▁Academ ic",
+      "▁Acad emic",
+      "M g",
+      "▁S team",
+      "▁Ste am",
+      "高 低",
+      "处 在",
+      "get Value",
+      "可以 将",
+      "▁Cert ified",
+      "所得 税",
+      "▁x en",
+      "报告 财产",
+      "重要 参数",
+      "▁reflect ing",
+      "埃 及",
+      "OF F",
+      "O FF",
+      "检公诉 刑诉",
+      "im i",
+      "i mi",
+      "▁b arg",
+      "▁bar g",
+      "▁ba rg",
+      "▁H ack",
+      "▁Ha ck",
+      "▁corpor ations",
+      "▁corporation s",
+      "ok ia",
+      "oki a",
+      "}} \\\\)",
+      "}}\\\\ )",
+      "} }\\\\)",
+      "UT E",
+      "U TE",
+      "的可能 性",
+      "的 可能性",
+      "ick eter",
+      "icket er",
+      "▁Coll abor",
+      "辩护 人",
+      "Up dated",
+      "Update d",
+      "▁r m",
+      "▁ rm",
+      "▁min erals",
+      "▁mine rals",
+      "▁mineral s",
+      "为 本",
+      "测 评",
+      "筹 码",
+      "ce des",
+      "ced es",
+      "c edes",
+      "▁sw eat",
+      "▁swe at",
+      "尔 德",
+      "徐 州",
+      "▁en erget",
+      "▁energ et",
+      "溶 解",
+      "行 走",
+      "这 座",
+      "min e",
+      "mi ne",
+      "m ine",
+      "▁H ab",
+      "▁Ha b",
+      "正方 形的",
+      "正方形 的",
+      "lish es",
+      "lis hes",
+      "l ishes",
+      "▁in formal",
+      "▁inf ormal",
+      "▁inform al",
+      "夸 张",
+      "LA N",
+      "L AN",
+      "▁Ad obe",
+      "it on",
+      "ito n",
+      "i ton",
+      "▁Eag les",
+      "▁Eagle s",
+      "之 类",
+      "白 糖",
+      "▁S har",
+      "▁Sh ar",
+      "▁Sha r",
+      "org anic",
+      "organ ic",
+      "彰 显",
+      "▁f ires",
+      "▁fire s",
+      "▁fi res",
+      "▁fir es",
+      "▁ fires",
+      "涨 幅",
+      "风 暴",
+      "不 吃",
+      "t B",
+      "不良 反应",
+      "佛 山",
+      "▁J av",
+      "▁Ja v",
+      "▁ Jav",
+      "▁John ny",
+      "sh arp",
+      "sha rp",
+      "▁g rocery",
+      "了 的",
+      "也不 要",
+      "也 不要",
+      "nR ound",
+      "n Round",
+      "EN TS",
+      "ENT S",
+      "▁re signed",
+      "▁res igned",
+      "▁resign ed",
+      "不 该",
+      "ard less",
+      "交通 安全",
+      "▁ter rain",
+      "▁terr ain",
+      "悄 悄",
+      "an co",
+      "anc o",
+      "a nco",
+      "▁V iol",
+      "▁Vi ol",
+      "▁Advis ory",
+      "▁Advisor y",
+      "▁Broad way",
+      "Instance State",
+      "历 经",
+      "▁retriev al",
+      "▁p recipitation",
+      "▁precip itation",
+      "▁V ue",
+      "▁ Vue",
+      "nU nit",
+      "nUn it",
+      "n Unit",
+      "▁Re dist",
+      "▁Red ist",
+      "▁Redis t",
+      "Snap shot",
+      "S napshot",
+      "▁ess ence",
+      "▁cont amination",
+      "▁contamin ation",
+      "layout s",
+      "lay outs",
+      "ym al",
+      "yma l",
+      "y mal",
+      "▁lux urious",
+      "中共 中央",
+      "\\' ]:\\",
+      "\\'] :\\",
+      "▁ap plicant",
+      "▁applic ant",
+      "▁dec laration",
+      "▁declar ation",
+      "▁API s",
+      "▁AP Is",
+      "侵 略",
+      "原 唱",
+      "▁H LA",
+      "▁HL A",
+      "▁S etting",
+      "▁Set ting",
+      "▁Sett ing",
+      "▁ Setting",
+      "▁meth ylation",
+      "▁methyl ation",
+      "的 操作",
+      "ac onda",
+      "acon da",
+      "aco nda",
+      "a conda",
+      "er i",
+      "e ri",
+      "Gr ad",
+      "G rad",
+      "th and",
+      "than d",
+      "tha nd",
+      "t hand",
+      "▁L ooks",
+      "▁Lo oks",
+      "▁Look s",
+      "预 报",
+      "▁st ressed",
+      "▁str essed",
+      "▁stress ed",
+      "▁M AC",
+      "▁MA C",
+      "▁ MAC",
+      "科 幻",
+      "nL eg",
+      "nLe g",
+      "n Leg",
+      "服务 业",
+      "的前提 下",
+      "ADD RESS",
+      "ADDR ESS",
+      "M c",
+      "([ \"",
+      "( [\"",
+      "▁E rr",
+      "▁Er r",
+      "▁ Err",
+      "失 效",
+      "▁S z",
+      "▁i b",
+      "▁ ib",
+      "分之 一",
+      "分 之一",
+      "▁Spe aker",
+      "▁prompt ed",
+      "▁refer enced",
+      "▁reference d",
+      "py p",
+      "p yp",
+      "▁K os",
+      "▁Ko s",
+      "▁evalu ations",
+      "▁evaluation s",
+      "▁eval uations",
+      "ot tage",
+      "ott age",
+      "otta ge",
+      "▁n i",
+      "▁ ni",
+      "in kle",
+      "ink le",
+      "ve olar",
+      "Res earch",
+      "Re search",
+      "ow ns",
+      "own s",
+      "归案 后",
+      "归 案后",
+      "br ev",
+      "bre v",
+      "b rev",
+      "get Id",
+      "wi g",
+      "w ig",
+      "了一 次",
+      "了 一次",
+      "因与 被上诉人",
+      "因 与被上诉人",
+      "Pi n",
+      "P in",
+      "dist rict",
+      "di strict",
+      "▁un fold",
+      "▁unf old",
+      "▁Comp anies",
+      "PA GE",
+      "P AGE",
+      "▁anal yse",
+      "▁analy se",
+      "▁analys e",
+      "▁j umping",
+      "▁jump ing",
+      "灵 敏",
+      "为 进一步",
+      "一 脸",
+      "菊 花",
+      "幂 的",
+      "演 艺",
+      "▁V ec",
+      "▁Ve c",
+      "▁ Vec",
+      "▁pro ves",
+      "▁pr oves",
+      "▁prov es",
+      "▁prove s",
+      "文化 和",
+      "文 化和",
+      "微 微",
+      "▁K arl",
+      "▁Kar l",
+      "▁Ka rl",
+      "点的 距离",
+      "点 的距离",
+      "ap pers",
+      "app ers",
+      "appe rs",
+      "apper s",
+      "M ichael",
+      "nF our",
+      "n Four",
+      "未 作",
+      "聚 合",
+      "多 功能",
+      "快速 发展",
+      "St age",
+      "S tage",
+      "cre ator",
+      "creat or",
+      "▁re moves",
+      "▁rem oves",
+      "▁remove s",
+      "▁remov es",
+      "ex am",
+      "▁l icensing",
+      "▁lic ensing",
+      "清 醒",
+      "▁min isters",
+      "▁mini sters",
+      "▁minister s",
+      "hl ine",
+      "h line",
+      "▁ST AT",
+      "▁ STAT",
+      "▁form ulated",
+      "▁formula ted",
+      "▁formulate d",
+      "D G",
+      "▁n uts",
+      "▁nut s",
+      "▁nu ts",
+      "▁ nuts",
+      "先 天",
+      "南 山",
+      "La t",
+      "L at",
+      "主要 包括",
+      "▁D ental",
+      "▁Den tal",
+      "▁Dent al",
+      "▁foss il",
+      "é m",
+      "gin x",
+      "gi nx",
+      "g inx",
+      "预 付",
+      "be it",
+      "▁Ass ociated",
+      "▁Associ ated",
+      "▁Associate d",
+      "备 受",
+      "Fi nal",
+      "Fin al",
+      "F inal",
+      "▁model ed",
+      "▁mode led",
+      "阻 碍",
+      "第一百四十五 条之规定",
+      "第一百四十五条 之规定",
+      "▁comp artment",
+      "片 段",
+      "nr outer",
+      "n router",
+      "▁M az",
+      "▁Ma z",
+      "制造 有限公司",
+      "出 手",
+      "的 撤诉申请",
+      "▁en emies",
+      "▁ 石",
+      "面积 是",
+      "▁Mor oc",
+      "▁Mo roc",
+      "uck land",
+      "is tle",
+      "ist le",
+      "认真 遵守监规",
+      "认真遵守 监规",
+      "书 店",
+      "▁p est",
+      "▁pe st",
+      "▁pes t",
+      "▁reduction s",
+      "▁reduct ions",
+      "互 联",
+      "]( #",
+      "] (#",
+      "很多 时候",
+      "box es",
+      "▁ar tic",
+      "▁art ic",
+      "床 铺",
+      "▁C ha",
+      "▁Ch a",
+      "Wr it",
+      "W rit",
+      "con da",
+      "co nda",
+      "cond a",
+      "c onda",
+      "e i",
+      "▁B arr",
+      "▁Bar r",
+      "▁Ba rr",
+      "每个 月",
+      "每 个月",
+      "▁conduct ivity",
+      "月 起",
+      "De lay",
+      "Del ay",
+      "▁bl ade",
+      "▁ blade",
+      "▁Sy rian",
+      "▁Syria n",
+      "▁Syr ian",
+      "▁f ucking",
+      "▁fuck ing",
+      "▁epid emic",
+      "▁epidem ic",
+      "光 泽",
+      "▁L ion",
+      "▁Li on",
+      "全 文",
+      "导 数",
+      "调 理",
+      "钢 材",
+      "著作 权",
+      "经原告 多次",
+      "▁veget able",
+      "▁ 知",
+      "pack et",
+      "pac ket",
+      "p acket",
+      "▁chrom atin",
+      "▁chromat in",
+      "学 报",
+      "滴 滴",
+      "nTax a",
+      "市 人民政府",
+      "▁Res ume",
+      "Face book",
+      "F acebook",
+      "Gener ated",
+      "Gen erated",
+      "Generate d",
+      "Gene rated",
+      "G enerated",
+      "▁B eth",
+      "▁Be th",
+      "▁Bet h",
+      "▁Summ it",
+      "▁Sum mit",
+      "另 查明",
+      "若干 规定",
+      "▁B rowser",
+      "▁Brow ser",
+      "▁Browse r",
+      "▁ Browser",
+      "医 保",
+      "是 真",
+      "落 叶",
+      "OP Y",
+      "O PY",
+      "ain ted",
+      "aint ed",
+      "料 理",
+      "实用 的",
+      "实 用的",
+      "▁qu een",
+      "▁que en",
+      "▁Mur phy",
+      "质 数",
+      "▁p ec",
+      "▁pe c",
+      "▁ pec",
+      "结婚 登记",
+      "ot rans",
+      "otr ans",
+      "o trans",
+      "▁d resses",
+      "▁dr esses",
+      "▁dress es",
+      "▁( ),",
+      "▁() ,",
+      "▁ (),",
+      "base name",
+      "bas ename",
+      "} >",
+      "nC all",
+      "nCal l",
+      "n Call",
+      "}} \"",
+      "} }\"",
+      "很强 的",
+      "很 强的",
+      "▁em brace",
+      "▁emb race",
+      "▁embr ace",
+      "却 又",
+      "▁Un iverse",
+      "▁Univers e",
+      "▁Uni verse",
+      "nS ummary",
+      "n Summary",
+      "来到 了",
+      "来 到了",
+      "Pa ul",
+      "P aul",
+      "唤 醒",
+      "em atic",
+      "ema tic",
+      "emat ic",
+      "e matic",
+      "▁His panic",
+      "瑜 伽",
+      "博士 学位",
+      "ac re",
+      "acr e",
+      "a cre",
+      "暗 示",
+      "vi m",
+      "v im",
+      "信用 社",
+      "下列 情形",
+      "res pond",
+      "respon d",
+      "resp ond",
+      "▁t ribute",
+      "▁trib ute",
+      "▁tribut e",
+      "工程 建设",
+      "▁gl omer",
+      "▁glo mer",
+      "▁Par alym",
+      "▁reg imen",
+      "▁regime n",
+      "余 数",
+      "发 货",
+      "代 言",
+      "大 一",
+      "▁se min",
+      "▁sem in",
+      "▁semi n",
+      "▁ semin",
+      "斯 基",
+      "▁S ci",
+      "▁Sc i",
+      "▁ Sci",
+      "Fix ed",
+      "F ixed",
+      "减 速",
+      "仍 未",
+      "假 的",
+      "是 非",
+      "ef an",
+      "e fan",
+      "\\' ]);\\",
+      "\\'] );\\",
+      "\\']) ;\\",
+      "\\ ']);\\",
+      "▁bad ge",
+      "▁ badge",
+      "▁enc rypted",
+      "▁encrypt ed",
+      "▁ encrypted",
+      "Time s",
+      "Tim es",
+      "Ti mes",
+      "T imes",
+      "一体 的",
+      "一 体的",
+      "▁Cr ime",
+      "▁Crim e",
+      "氨基 酸",
+      "al ong",
+      "alo ng",
+      "alon g",
+      "a long",
+      "AT ER",
+      "ATE R",
+      "A TER",
+      "▁fer til",
+      "▁fert il",
+      "举办 的",
+      "举 办的",
+      "▁ad dict",
+      "▁add ict",
+      "技术 开发",
+      "星级 酒店",
+      "br it",
+      "b rit",
+      "▁reve aling",
+      "▁reveal ing",
+      "▁ch eer",
+      "▁che er",
+      "▁Y ahoo",
+      "tic ally",
+      "tical ly",
+      "t ically",
+      "播 种",
+      "nt here",
+      "nthe re",
+      "nth ere",
+      "n there",
+      "▁Pr incipal",
+      "▁Princ ipal",
+      "▁ Principal",
+      "▁Iss ues",
+      "▁Issue s",
+      "▁inn ate",
+      "的 出现",
+      "名 家",
+      "Ma sk",
+      "Mas k",
+      "M ask",
+      "▁R ules",
+      "▁Rule s",
+      "▁Ru les",
+      "▁ Rules",
+      "▁exp lor",
+      "▁expl or",
+      "▁explo r",
+      "▁flu ids",
+      "▁fluid s",
+      "▁div iding",
+      "▁divid ing",
+      "不仅仅 是",
+      "不 仅仅是",
+      "▁hypothes ized",
+      "▁hypothesize d",
+      "▁V ik",
+      "▁Vi k",
+      "▁am mon",
+      "▁an tit",
+      "▁ant it",
+      "▁anti t",
+      "▁join s",
+      "▁jo ins",
+      "农 作物",
+      "的 经营",
+      "▁c rust",
+      "▁cr ust",
+      "▁cru st",
+      "ynchron ous",
+      "造成 了",
+      "造 成了",
+      "IT LE",
+      "nF ollow",
+      "n Follow",
+      "▁Sign al",
+      "▁Sig nal",
+      "▁ Signal",
+      "美 股",
+      "民事 权利",
+      "nU nfortunately",
+      "VI EW",
+      "V IEW",
+      "等 领域",
+      "side d",
+      "sid ed",
+      "si ded",
+      "s ided",
+      "▁P arks",
+      "▁Par ks",
+      "▁Park s",
+      "ed ar",
+      "eda r",
+      "e dar",
+      "▁dif fers",
+      "▁differ s",
+      "▁diff ers",
+      "▁in sects",
+      "▁ins ects",
+      "▁insect s",
+      "川 区",
+      "▁Ke ith",
+      "▁cal ories",
+      "▁calor ies",
+      "转 速",
+      "劳动 合同",
+      "▁fix tures",
+      "▁fixture s",
+      "▁ fixtures",
+      "to s",
+      "t os",
+      "▁annoy ing",
+      "小 龙",
+      "▁Gu inea",
+      "余 篇",
+      "ndecl are",
+      "ndec lare",
+      "n declare",
+      "nC O",
+      "n CO",
+      "▁K l",
+      "▁i v",
+      "▁ iv",
+      "与原 告",
+      "与 原告",
+      "科学 发展",
+      "饱 满",
+      "Re dis",
+      "Red is",
+      "▁Mar vel",
+      "▁H ungarian",
+      "▁Hung arian",
+      "y x",
+      "实数 根",
+      "▁reg ex",
+      "▁ regex",
+      "摆 放",
+      "▁dors al",
+      "好 吗",
+      "am bers",
+      "amb ers",
+      "amber s",
+      "▁met allic",
+      "▁metal lic",
+      "▁metall ic",
+      "一 丝",
+      "中 档",
+      "]) ,\\",
+      "]), \\",
+      "] ),\\",
+      "▁Pre vention",
+      "▁Prevent ion",
+      "答 主",
+      "谈 恋爱",
+      "▁T ow",
+      "▁To w",
+      "▁S erv",
+      "▁Se rv",
+      "▁Ser v",
+      "▁ Serv",
+      "各 式",
+      "keep ing",
+      "kee ping",
+      "法定代表人 王",
+      "ic ulous",
+      "icul ous",
+      "ed nesday",
+      "三十 日内",
+      "三十日 内",
+      "三 十日内",
+      "▁ 规格",
+      "局 副局长",
+      "▁gr ams",
+      "▁gram s",
+      "▁gra ms",
+      "▁ grams",
+      "海南 省",
+      "海 南省",
+      "ur om",
+      "uro m",
+      "u rom",
+      "▁a li",
+      "▁al i",
+      "▁ ali",
+      "▁S park",
+      "▁Sp ark",
+      "▁Spa rk",
+      "▁Spar k",
+      "▁ Spark",
+      "▁美 国",
+      "▁ 美国",
+      "▁h ills",
+      "▁hill s",
+      "▁hil ls",
+      "it ives",
+      "itive s",
+      "iti ves",
+      "Comp any",
+      "/ ](",
+      "区 长",
+      "质 感",
+      "RG B",
+      "R GB",
+      "▁L G",
+      "▁ LG",
+      "M r",
+      "溪 县",
+      "cell ular",
+      "c ellular",
+      "后 勤",
+      "自己 在",
+      "好 人",
+      "新 年",
+      "iz en",
+      "ize n",
+      "i zen",
+      "种 族",
+      "}) \\\\",
+      "})\\ \\",
+      "} )\\\\",
+      "▁h over",
+      "▁ho ver",
+      "▁ hover",
+      "▁i di",
+      "▁id i",
+      "▁ idi",
+      "▁N umbers",
+      "▁Number s",
+      "▁Num bers",
+      "▁ Numbers",
+      "这些 人",
+      "st ad",
+      "sta d",
+      "意 图",
+      "集 群",
+      "▁s eq",
+      "▁se q",
+      "▁ seq",
+      "▁ul cer",
+      "▁Egypt ian",
+      "业 余",
+      "▁Y u",
+      "▁mult if",
+      "▁multi f",
+      "▁mul tif",
+      "名 校",
+      "短 短",
+      "nC al",
+      "n Cal",
+      "od ynamics",
+      "odynam ics",
+      "odynamic s",
+      "o dynamics",
+      "极 致",
+      "nA thlet",
+      "晶 体",
+      "申请执行人 可",
+      "最 强",
+      "的 根",
+      "en de",
+      "end e",
+      "e nde",
+      "做 人",
+      "▁M ol",
+      "▁Mo l",
+      "▁number ed",
+      "降 雨",
+      "▁D ry",
+      "▁Dr y",
+      "Fail ure",
+      "为了 让",
+      "AN GU",
+      "ANG U",
+      "▁O blast",
+      "▁Ob last",
+      "ph an",
+      "pha n",
+      "p han",
+      "fin ished",
+      "finish ed",
+      "国 企",
+      "的 综合",
+      "指 引",
+      "时光 网",
+      "▁North west",
+      "牌 子",
+      "这 不",
+      "判断 出",
+      "zo om",
+      "zoo m",
+      "z oom",
+      "▁un success",
+      "▁uns uccess",
+      "AC HE",
+      "ACH E",
+      "A CHE",
+      "g atsby",
+      "ber ries",
+      "▁bond ing",
+      "▁bon ding",
+      "▁disp ersion",
+      "▁dispers ion",
+      "▁men tor",
+      "▁ment or",
+      "p w",
+      ";\" >",
+      "; \">",
+      "do s",
+      "d os",
+      "等 情况",
+      "水 准",
+      "红 的",
+      "起 源",
+      "▁med ieval",
+      "▁medi eval",
+      "别 说",
+      "Local e",
+      "Loc ale",
+      "Lo cale",
+      "▁purs uing",
+      "自己 是",
+      "工商 银行",
+      "\\' :\\'",
+      "\\': \\'",
+      "\\':\\ '",
+      "自 律",
+      "Fr ont",
+      "▁S hell",
+      "▁Sh ell",
+      "▁She ll",
+      "▁Shel l",
+      "▁ Shell",
+      "▁R ice",
+      "▁Ric e",
+      "▁Ri ce",
+      "▁Cons ervation",
+      "▁Conserv ation",
+      "SE SS",
+      "SES S",
+      "S ESS",
+      "▁midfield er",
+      "' /",
+      "进入 了",
+      "进 入了",
+      "平方 厘米",
+      "re active",
+      "react ive",
+      "oc oc",
+      "oco c",
+      "o coc",
+      "▁ex pose",
+      "▁exp ose",
+      "跟 他",
+      "▁h r",
+      "▁ hr",
+      "总收入 的",
+      "总 收入的",
+      "SE CRE",
+      "SEC RE",
+      "▁cl oned",
+      "▁clone d",
+      "hi n",
+      "h in",
+      "撰 写",
+      "vo te",
+      "v ote",
+      "▁M OD",
+      "▁MO D",
+      "▁ MOD",
+      "il iation",
+      "ili ation",
+      "ilia tion",
+      "▁ 总",
+      "残 留",
+      "▁select ively",
+      "▁selective ly",
+      "市中级人民法院 于",
+      "▁fabric ated",
+      "变 速",
+      "笔 试",
+      "文化 遗产",
+      "▁M ining",
+      "▁Min ing",
+      "▁Mi ning",
+      "▁Mini ng",
+      "process ors",
+      "processor s",
+      "▁ne oplas",
+      "珠 海",
+      "甚至 是",
+      "▁t ert",
+      "▁te rt",
+      "▁ter t",
+      "drop down",
+      "▁Anal ytics",
+      "▁ Analytics",
+      "的最 新",
+      "的 最新",
+      "▁S afe",
+      "▁Saf e",
+      "▁Sa fe",
+      "▁ Safe",
+      "iv ated",
+      "iva ted",
+      "ivate d",
+      "▁estim ating",
+      "长 约",
+      "▁j ar",
+      "▁ja r",
+      "▁ jar",
+      "ic ol",
+      "ico l",
+      "i col",
+      "亮 度",
+      "原 价",
+      "▁N u",
+      "▁ Nu",
+      "▁G UI",
+      "▁GU I",
+      "▁ GUI",
+      "所 提供的",
+      "入 了",
+      "半 小时",
+      "▁E ither",
+      "也 为",
+      "含量 为",
+      "含 量为",
+      "固定 的",
+      "固 定的",
+      "he lm",
+      "hel m",
+      "h elm",
+      "▁D op",
+      "▁Do p",
+      "▁M ack",
+      "▁Mac k",
+      "▁Ma ck",
+      "▁V lad",
+      "人口 和计划生育局",
+      "轻 微",
+      "小额 贷款",
+      "熟练 掌握",
+      "▁Work shop",
+      "▁Works hop",
+      "号 召",
+      "河 区",
+      "▁Bath room",
+      "sl f",
+      "s lf",
+      "顶 尖",
+      "▁f ears",
+      "▁fe ars",
+      "▁fear s",
+      "▁Al fred",
+      "▁Alf red",
+      "▁ 文",
+      "的 经验",
+      "nF lora",
+      "nFl ora",
+      "MO N",
+      "M ON",
+      "▁P erry",
+      "▁Per ry",
+      "▁pharmac okin",
+      "graph ql",
+      "参 阅",
+      "Be l",
+      "B el",
+      "上述事实 有",
+      "▁Con sequently",
+      "▁L uther",
+      "▁Lu ther",
+      "gi e",
+      "g ie",
+      "所在 的",
+      ")\" ,\\",
+      ")\", \\",
+      ") \",\\",
+      "比 分",
+      "▁f reeze",
+      "▁fr eeze",
+      "▁free ze",
+      "▁ freeze",
+      "四 种",
+      "强 迫",
+      "check s",
+      "che cks",
+      "▁Star ting",
+      "▁Start ing",
+      "▁ Starting",
+      "晋 升",
+      "牛 逼",
+      "脚 本",
+      "承担 连带",
+      "es sel",
+      "ess el",
+      "esse l",
+      "▁st arter",
+      "▁start er",
+      "▁star ter",
+      "▁ starter",
+      "ra ge",
+      "rag e",
+      "r age",
+      "▁Grad uate",
+      "第一百 七十",
+      "第一百七 十",
+      "nL ove",
+      "nLo ve",
+      "n Love",
+      "▁m odal",
+      "▁mod al",
+      "▁mo dal",
+      "▁ modal",
+      "▁Con struct",
+      "▁Co nstruct",
+      "▁Const ruct",
+      "▁ Construct",
+      "二 氧化",
+      "机械 有限公司",
+      "▁Lo ading",
+      "▁Load ing",
+      "▁ Loading",
+      "st udio",
+      "stud io",
+      "▁S id",
+      "▁Si d",
+      "▁e rg",
+      "▁er g",
+      "▁ erg",
+      "otyp ing",
+      "和 好",
+      "坏 了",
+      "▁ant iv",
+      "▁anti v",
+      "▁loc ale",
+      "▁local e",
+      "▁lo cale",
+      "▁ locale",
+      "▁Vari able",
+      "▁Var iable",
+      "▁ Variable",
+      "小 火",
+      "布 鲁",
+      "多 学科",
+      "▁rec ursive",
+      "▁recurs ive",
+      "▁ recursive",
+      "▁T W",
+      "▁ TW",
+      "留学 生",
+      "留 学生",
+      "▁re vis",
+      "▁rev is",
+      "的 你",
+      "tf loat",
+      "t float",
+      "申请 强制执行",
+      "ci d",
+      "c id",
+      "▁em igr",
+      "▁g race",
+      "▁gr ace",
+      "▁grac e",
+      "▁gra ce",
+      "▁appe aling",
+      "▁appeal ing",
+      "]] [",
+      "] ][",
+      "▁cy to",
+      "▁cyt o",
+      "nL earn",
+      "nLe arn",
+      "n Learn",
+      "▁embry os",
+      "▁embryo s",
+      "▁anest hesia",
+      "出 色",
+      "vo r",
+      "v or",
+      "会 影响",
+      "nD ownload",
+      "n Download",
+      "▁cabin ets",
+      "▁cabinet s",
+      "▁comm ander",
+      "▁command er",
+      "▁comma nder",
+      "但 却",
+      "再 去",
+      "十三 章",
+      "十 三章",
+      "od al",
+      "oda l",
+      "o dal",
+      "予以 确认",
+      "nF ile",
+      "nFil e",
+      "n File",
+      "}$ $,",
+      "}$$ ,",
+      "} $$,",
+      "▁st amp",
+      "▁ stamp",
+      "宜 昌",
+      "▁reason ably",
+      "形态 特征",
+      "▁new born",
+      "心 思",
+      "▁e scal",
+      "▁es cal",
+      "▁esc al",
+      "▁G lo",
+      "▁Gl o",
+      "▁Cal endar",
+      "▁ Calendar",
+      "▁Rel ations",
+      "易 于",
+      "la re",
+      "lar e",
+      "l are",
+      "▁s ells",
+      "▁sell s",
+      "▁sel ls",
+      "失 误",
+      "▁epile psy",
+      "兔 子",
+      "沿 着",
+      "▁C rystal",
+      "底 数",
+      "{{ $",
+      "{ {$",
+      "上一 个",
+      "上 一个",
+      "教 会",
+      "ne al",
+      "nea l",
+      "n eal",
+      "▁R ender",
+      "▁Re nder",
+      "▁Ren der",
+      "▁ Render",
+      "▁点 击",
+      "▁ 点击",
+      "▁M ono",
+      "▁Mon o",
+      "▁Mo no",
+      "▁ Mono",
+      "ri en",
+      "rie n",
+      "r ien",
+      "▁Ar abic",
+      "▁Arab ic",
+      "▁brow sing",
+      "拿 下",
+      "let ters",
+      "letter s",
+      "lett ers",
+      "lette rs",
+      "▁t uning",
+      "▁tun ing",
+      "▁tu ning",
+      "▁trade mark",
+      "▁tradem ark",
+      "dn s",
+      "d ns",
+      "累 积",
+      "的真 实",
+      "的 真实",
+      "go ne",
+      "gon e",
+      "g one",
+      "Rule s",
+      "Ru les",
+      "R ules",
+      "▁Sl oven",
+      "▁Slo ven",
+      "书 城",
+      "着 我",
+      "mo t",
+      "m ot",
+      "ze l",
+      "z el",
+      "的 区别",
+      "老 实",
+      "▁st aging",
+      "▁ staging",
+      "▁te aches",
+      "▁teach es",
+      "▁tea ches",
+      "▁P reliminary",
+      "的 很",
+      "抄 袭",
+      "跨 境",
+      "低 碳",
+      "墨 西",
+      "够 了",
+      "带 给",
+      "早 晨",
+      "▁R odr",
+      "▁Ro dr",
+      "▁Rod r",
+      "▁sh ame",
+      "▁sha me",
+      "▁sham e",
+      "参 展",
+      "▁all ele",
+      "▁alle le",
+      "▁ 日",
+      "兵 团",
+      "了 许多",
+      "▁m arch",
+      "▁mar ch",
+      "▁V P",
+      "▁ VP",
+      "天 马",
+      "此 刻",
+      "▁C RE",
+      "▁CR E",
+      "▁ CRE",
+      "▁d ive",
+      "▁di ve",
+      "▁div e",
+      "结 核",
+      "bid den",
+      "b idden",
+      "Am eric",
+      "A meric",
+      "▁Lead ership",
+      "▁Leader ship",
+      "▁Leaders hip",
+      "橄 榄",
+      "▁B ir",
+      "▁Bi r",
+      "Sh ader",
+      "Sha der",
+      "▁Bay esian",
+      "▁Bayes ian",
+      "▁Jer usalem",
+      "requ irements",
+      "require ments",
+      "判决发生法律效力后 交付执行",
+      "罪犯 减刑",
+      "▁Leg end",
+      "牢 记",
+      "great er",
+      "gre ater",
+      "封 建",
+      "的 需要",
+      "之 子",
+      "▁merc hand",
+      "▁merch and",
+      "▁elabor ate",
+      "TO DO",
+      "T ODO",
+      "in ging",
+      "ing ing",
+      "▁Ad ding",
+      "▁Add ing",
+      "▁b rowsers",
+      "▁browser s",
+      "▁browse rs",
+      "▁ 姜",
+      "模 样",
+      "为一 体的",
+      "为一体 的",
+      "为 一体的",
+      "▁V illa",
+      "▁Vill a",
+      "▁Vi lla",
+      "▁Vil la",
+      "县人民法院 于",
+      "▁b asket",
+      "▁bas ket",
+      "笑 话",
+      "▁bankrupt cy",
+      "To uch",
+      "T ouch",
+      "▁Par kinson",
+      "▁Park inson",
+      "不 觉",
+      "小 吃",
+      "袖添香 网",
+      "▁B order",
+      "▁Bor der",
+      "▁ Border",
+      "▁earthqu ake",
+      "ill ery",
+      "ille ry",
+      "iller y",
+      "日提出 减刑建议书",
+      "日提出减刑建议 书",
+      "日 提出减刑建议书",
+      "的一 天",
+      "的 一天",
+      "▁H CV",
+      "▁HC V",
+      "的 计算",
+      "结婚 证",
+      "▁E ld",
+      "▁El d",
+      "ad oop",
+      "ado op",
+      "We ek",
+      "W eek",
+      "ros ion",
+      "r osion",
+      "▁sk learn",
+      "ob uf",
+      "o buf",
+      "▁T odd",
+      "▁To dd",
+      "全国 各地",
+      "cher ichia",
+      "央 行",
+      "不应 该",
+      "不 应该",
+      "被 迫",
+      "re fer",
+      "ref er",
+      "r efer",
+      "▁appoint ments",
+      "▁appointment s",
+      "▁i onic",
+      "▁ion ic",
+      "▁io nic",
+      "▁ ionic",
+      "的女 孩",
+      "的 女孩",
+      "op al",
+      "opa l",
+      "o pal",
+      "▁rec reational",
+      "▁recreation al",
+      "▁cocon ut",
+      "▁south west",
+      "▁cl oth",
+      "▁clot h",
+      "sw agger",
+      "ert ation",
+      "erta tion",
+      "▁different iated",
+      "▁differentiate d",
+      "IO S",
+      "I OS",
+      "▁S oul",
+      "▁So ul",
+      "▁Sou l",
+      "Class Name",
+      "依 职权",
+      "▁sh ades",
+      "▁sha des",
+      "▁shade s",
+      "▁Im perial",
+      "▁Imp erial",
+      "物 价",
+      "IP E",
+      "I PE",
+      "▁w ax",
+      "▁wa x",
+      "▁north west",
+      "▁W W",
+      "▁ WW",
+      "aut om",
+      "au tom",
+      "auto m",
+      "光 驱",
+      "cr o",
+      "c ro",
+      "esc ape",
+      "e scape",
+      "▁ad visor",
+      "▁adv isor",
+      "▁advis or",
+      "vir al",
+      "vi ral",
+      "v iral",
+      "è s",
+      "在 学校",
+      "使用 寿命",
+      "▁Part ial",
+      "▁ Partial",
+      "滑 动",
+      "首 选",
+      "钢 板",
+      "用人 单位",
+      "Or igin",
+      "▁dur ability",
+      "▁S ierra",
+      "nS en",
+      "nSe n",
+      "n Sen",
+      "光 谱",
+      "▁K umar",
+      "▁Ku mar",
+      "▁Kum ar",
+      "▁F R",
+      "▁ FR",
+      "我们 都",
+      "我 们都",
+      "▁po ems",
+      "▁poem s",
+      "▁c asting",
+      "▁cast ing",
+      "▁cas ting",
+      "▁ casting",
+      "▁c ourage",
+      "▁cour age",
+      "▁cou rage",
+      "tr es",
+      "tre s",
+      "t res",
+      "▁c ran",
+      "▁cr an",
+      "▁cra n",
+      "▁leg ally",
+      "▁legal ly",
+      "斜 率",
+      "致 富",
+      "▁vol leyball",
+      "克 拉",
+      "Im p",
+      "I mp",
+      "违法 行为",
+      "os ide",
+      "osi de",
+      "o side",
+      "▁G arc",
+      "▁Ga rc",
+      "▁Gar c",
+      "▁he aven",
+      "▁heav en",
+      "▁anal yst",
+      "▁analy st",
+      "▁analys t",
+      "雕 刻",
+      "改变 了",
+      "改 变了",
+      "Graph ics",
+      "额 度",
+      "▁ware house",
+      "t background",
+      "ip per",
+      "ipp er",
+      "▁Tr inity",
+      "▁Trin ity",
+      "▁strong est",
+      "半 岛",
+      "报 刊",
+      "▁D iet",
+      "▁Di et",
+      "▁Die t",
+      "▁cust ody",
+      "枸 杞",
+      "Di agn",
+      "UT C",
+      "U TC",
+      "运动 会",
+      "平 市",
+      "on do",
+      "ond o",
+      "o ndo",
+      "▁O liver",
+      "▁Ol iver",
+      "▁Olive r",
+      "}] \\",
+      "} ]\\",
+      "ograph s",
+      "ograp hs",
+      "o graphs",
+      "▁P enn",
+      "▁Pen n",
+      "▁Pe nn",
+      "▁S oph",
+      "▁So ph",
+      "mi l",
+      "m il",
+      "IN IT",
+      "▁E li",
+      "▁El i",
+      "T ensor",
+      "▁sem icon",
+      "▁semi con",
+      "▁semic on",
+      "▁W ine",
+      "▁Win e",
+      "▁Wi ne",
+      "五 百",
+      "排 骨",
+      "▁de alt",
+      "▁deal t",
+      "nAl umni",
+      "I p",
+      "od ef",
+      "ode f",
+      "o def",
+      "铸 造",
+      "军 区",
+      "▁F u",
+      "au kee",
+      "auk ee",
+      "end section",
+      "ends ection",
+      "nRe ception",
+      "上 次",
+      "管理 人员",
+      "▁del icate",
+      "维 度",
+      "的爱 情",
+      "的 爱情",
+      "的小 伙伴",
+      "的 小伙伴",
+      "ploy ees",
+      "ployee s",
+      "▁margin s",
+      "▁marg ins",
+      "任 命",
+      "多 万",
+      "▁explan ations",
+      "▁explanation s",
+      "国家 队",
+      "▁g ad",
+      "▁ga d",
+      "struct ured",
+      "structure d",
+      "全 家",
+      "Sm art",
+      "S mart",
+      "orac ic",
+      "ra is",
+      "r ais",
+      "▁l ane",
+      "▁la ne",
+      "▁lan e",
+      "▁ lane",
+      "▁ded ication",
+      "iot ensin",
+      "▁contr ary",
+      "的电 影",
+      "的 电影",
+      "自然 村",
+      "变 卖",
+      "or esis",
+      "ore sis",
+      "ores is",
+      "▁Build er",
+      "▁ Builder",
+      "e o",
+      "不 信",
+      "开 头",
+      "▁t oggle",
+      "▁togg le",
+      "▁ toggle",
+      "▁et iology",
+      "与 人",
+      "同 上",
+      "ok o",
+      "o ko",
+      "fol k",
+      "f olk",
+      "薄 膜",
+      "福州 市",
+      "福 州市",
+      "民 办",
+      "ak t",
+      "a kt",
+      "▁Ex pert",
+      "▁Exper t",
+      "▁Exp ert",
+      "中 新",
+      "这是 一道",
+      "这是一 道",
+      "这 是一道",
+      "▁v est",
+      "▁ve st",
+      "▁ves t",
+      "▁ vest",
+      "▁w ires",
+      "▁wire s",
+      "▁wi res",
+      "▁hy pers",
+      "▁hyp ers",
+      "▁hyper s",
+      "▁hype rs",
+      "诸 葛",
+      "▁A aron",
+      "nS imilar",
+      "nSim ilar",
+      "n Similar",
+      "I l",
+      "▁p c",
+      "▁ pc",
+      "成立 的",
+      "nC OPY",
+      "nCO PY",
+      "n COPY",
+      "▁T rip",
+      "▁Tr ip",
+      "▁Tri p",
+      "▁ Trip",
+      "高新技术 产业",
+      "▁Sur v",
+      "▁Su rv",
+      "▁prem iere",
+      "▁premier e",
+      "听 了",
+      "▁G ol",
+      "▁Go l",
+      "交通 肇事",
+      "kot lin",
+      "k otlin",
+      "工业 园区",
+      "工业园 区",
+      "▁I A",
+      "▁ IA",
+      "已 于",
+      "** :\\",
+      "**: \\",
+      "▁squ ee",
+      "▁b oring",
+      "▁bo ring",
+      "▁bor ing",
+      "▁myster ious",
+      "更 能",
+      "▁g ently",
+      "▁gent ly",
+      "▁well ness",
+      "▁tr ash",
+      "▁tra sh",
+      "的 政策",
+      "▁touch down",
+      "其 前",
+      "at z",
+      "a tz",
+      "▁anal ysts",
+      "▁analy sts",
+      "▁analys ts",
+      "▁analyst s",
+      "南 瓜",
+      "富 贵",
+      "抛 弃",
+      "机 组",
+      "牢 固",
+      "近 视",
+      "ge om",
+      "geo m",
+      "▁G rove",
+      "▁Gr ove",
+      "▁Gro ve",
+      "▁colon ial",
+      "台 阶",
+      "▁o ak",
+      "▁ oak",
+      "ly cer",
+      "▁H air",
+      "▁Ha ir",
+      "▁abs orbed",
+      "▁absor bed",
+      "▁absorb ed",
+      "▁adjust able",
+      "▁project ions",
+      "▁projection s",
+      "▁proj ections",
+      "▁g ates",
+      "▁gate s",
+      "▁ga tes",
+      "啊 啊",
+      "水 县",
+      "环境 中",
+      "屏幕 尺寸",
+      "▁M aps",
+      "▁Map s",
+      "▁Ma ps",
+      "▁ Maps",
+      "▁T reas",
+      "▁Tr eas",
+      "▁Tre as",
+      "▁Aff ero",
+      "宝 马",
+      "▁j oke",
+      "▁jo ke",
+      "▁re versible",
+      "▁revers ible",
+      "tf ont",
+      "t font",
+      "▁A verage",
+      "▁Ave rage",
+      "▁ Average",
+      "wo f",
+      "w of",
+      "破 解",
+      "化学 品",
+      "加强 对",
+      "▁re sin",
+      "▁res in",
+      "Be n",
+      "B en",
+      "by s",
+      "b ys",
+      "婚 纱",
+      "水 流",
+      "▁P av",
+      "▁Pa v",
+      "▁C AD",
+      "▁CA D",
+      "▁ CAD",
+      "am ente",
+      "ame nte",
+      "ament e",
+      "amen te",
+      "a mente",
+      "▁a ster",
+      "▁as ter",
+      "▁ast er",
+      "▁ aster",
+      "▁Cat hedral",
+      "共 赢",
+      "AL LY",
+      "ALL Y",
+      "▁young est",
+      "合 资",
+      "惊 艳",
+      "予以 支持",
+      "▁D ATA",
+      "▁DA TA",
+      "▁ DATA",
+      "us tration",
+      "ust ration",
+      "ustr ation",
+      "明 珠",
+      "运 费",
+      "▁{ :",
+      "▁ {:",
+      "▁dis closure",
+      "场 面",
+      "被 子",
+      "▁s ax",
+      "▁sa x",
+      "养老 保险",
+      "▁Del aware",
+      "网 格",
+      "▁f are",
+      "▁far e",
+      "▁fa re",
+      "▁ fare",
+      "▁ste ering",
+      "▁steer ing",
+      "am ents",
+      "ament s",
+      "amen ts",
+      "a ments",
+      "▁cart ilage",
+      "再 利用",
+      "in ters",
+      "int ers",
+      "inter s",
+      "i nters",
+      "指导 下",
+      "▁pers istence",
+      "▁persist ence",
+      "▁ persistence",
+      "领导 人",
+      "▁Sim plify",
+      "▁ Simplify",
+      "▁t ops",
+      "▁to ps",
+      "▁top s",
+      "▁ tops",
+      "▁an kle",
+      "水 面",
+      "脆 弱",
+      "音 响",
+      "▁Key words",
+      "▁h iking",
+      "▁hi king",
+      "▁remember ed",
+      "▁W orth",
+      "▁Wor th",
+      "迁 移",
+      "▁c arp",
+      "▁car p",
+      "▁ca rp",
+      "▁( #",
+      "▁ (#",
+      "有 两",
+      "也 需要",
+      "Al low",
+      "All ow",
+      "▁ob struction",
+      "▁obstruct ion",
+      "列 式",
+      "的全 部",
+      "的 全部",
+      "Trans port",
+      "概 括",
+      "ip art",
+      "ipa rt",
+      "i part",
+      "▁基本 参数",
+      "▁ 基本参数",
+      "▁view er",
+      "▁ viewer",
+      "▁Venezuel a",
+      "▁cl ones",
+      "▁clone s",
+      "▁A PP",
+      "▁AP P",
+      "▁ APP",
+      "ur istic",
+      "uris tic",
+      "par allel",
+      "paralle l",
+      "> [",
+      "ev e",
+      "e ve",
+      "多元 化",
+      "总 公司",
+      "▁g rep",
+      "▁gr ep",
+      "▁gre p",
+      "▁ grep",
+      "▁micro scopic",
+      "▁in equalities",
+      "合同 约定",
+      "量 子",
+      "不清 楚",
+      "不 清楚",
+      "公布 的",
+      "F amily",
+      "熬 夜",
+      "着 手",
+      "霉 素",
+      "▁pre operative",
+      "Add r",
+      "Ad dr",
+      "in th",
+      "int h",
+      "i nth",
+      "▁M ouse",
+      "▁Mo use",
+      "▁ Mouse",
+      "整 除",
+      "香 气",
+      "▁P ierre",
+      "▁Pier re",
+      "as kat",
+      "ask at",
+      "aska t",
+      "el em",
+      "ele m",
+      "e lem",
+      "新 媒体",
+      "yz ed",
+      "yze d",
+      "y zed",
+      "ns c",
+      "n sc",
+      "▁Pe ak",
+      "▁attend ees",
+      "\\\\( \\\\{",
+      "\\\\(\\\\ {",
+      "▁Id aho",
+      "▁p ipes",
+      "▁pi pes",
+      "▁pip es",
+      "▁pipe s",
+      "AN N",
+      "A NN",
+      "{} ,",
+      "{ },",
+      "墨西 哥",
+      "启 发",
+      "户籍 地",
+      "各项 工作",
+      "▁phosph olip",
+      "nC r",
+      "n Cr",
+      "▁ch rome",
+      "▁chrom e",
+      "▁chr ome",
+      "▁ chrome",
+      "▁Camp aign",
+      "M i",
+      "独 家",
+      "ha d",
+      "h ad",
+      "▁\\' \"",
+      "▁ \\'\"",
+      "上 限",
+      "描 绘",
+      "▁filter ed",
+      "▁ filtered",
+      "▁em br",
+      "▁emb r",
+      "▁ embr",
+      "til de",
+      "t ilde",
+      "双 人",
+      "做 成",
+      "▁T erry",
+      "▁Ter ry",
+      "▁Terr y",
+      "▁professional ly",
+      "▁profession ally",
+      "似 的",
+      "便 秘",
+      "ac ci",
+      "acc i",
+      "a cci",
+      "▁C MS",
+      "▁CM S",
+      "▁ CMS",
+      "▁n ão",
+      "之 战",
+      "米 饭",
+      "配 料",
+      "▁L imit",
+      "▁Lim it",
+      "▁Li mit",
+      "▁ Limit",
+      "▁occasion al",
+      "▁occas ional",
+      "陪 同",
+      "Su ite",
+      "S uite",
+      "▁construct s",
+      "no me",
+      "nom e",
+      "n ome",
+      "nB ar",
+      "n Bar",
+      "▁ap par",
+      "▁app ar",
+      "▁knock out",
+      "▁mod ulate",
+      "D X",
+      "接 入",
+      "税务 局",
+      "nPh il",
+      "n Phil",
+      "▁up set",
+      "▁ups et",
+      "事 后",
+      "转 动",
+      "br ook",
+      "bro ok",
+      "▁p ond",
+      "▁po nd",
+      "▁pon d",
+      "▁ pond",
+      "▁M ongo",
+      "▁Mon go",
+      "▁Mo ngo",
+      "▁Mong o",
+      "▁ Mongo",
+      "h b",
+      "那 你",
+      "tr o",
+      "t ro",
+      "ym e",
+      "y me",
+      "卖 家",
+      "▁L ik",
+      "▁Li k",
+      "大 纲",
+      "要 比",
+      "机 体",
+      "树 木",
+      "cm ake",
+      "c make",
+      "▁S tra",
+      "▁St ra",
+      "▁Str a",
+      "查 处",
+      "▁l oci",
+      "▁loc i",
+      "▁lo ci",
+      "▁r ides",
+      "▁ri des",
+      "▁ride s",
+      "▁rid es",
+      "▁ rides",
+      "已知 函数",
+      "▁L oss",
+      "▁Los s",
+      "▁Lo ss",
+      "▁ Loss",
+      "cast er",
+      "ca ster",
+      "cas ter",
+      "c aster",
+      "▁log istics",
+      "▁logistic s",
+      "但 其",
+      "我们 是",
+      "vo c",
+      "v oc",
+      "重要 作用",
+      "Cont in",
+      "Con tin",
+      "ou tline",
+      "out line",
+      "▁sex ually",
+      "▁sexual ly",
+      "▁co ated",
+      "▁coat ed",
+      "▁ coated",
+      "▁correspond ence",
+      "变 频",
+      "网 点",
+      "ET A",
+      "E TA",
+      "▁D ol",
+      "▁Do l",
+      "▁n oting",
+      "▁not ing",
+      "▁no ting",
+      "T x",
+      "要 去",
+      "▁ 经营",
+      "第五 节",
+      "os ion",
+      "osi on",
+      "/ <",
+      "st ere",
+      "ster e",
+      "ste re",
+      "ag rant",
+      "agr ant",
+      "agra nt",
+      "a grant",
+      "nCh apter",
+      "n Chapter",
+      "ke letal",
+      "kel etal",
+      "▁contin ually",
+      "▁continu ally",
+      "将 是",
+      "st reet",
+      "stre et",
+      "▁path ogenic",
+      "▁pathogen ic",
+      "▁V as",
+      "▁Va s",
+      "化 疗",
+      "Ho ok",
+      "H ook",
+      "▁str ips",
+      "▁stri ps",
+      "▁strip s",
+      "左右 的",
+      "▁learn ers",
+      "▁lear ners",
+      "▁learner s",
+      "▁A u",
+      "▁ Au",
+      "▁link age",
+      "雅 思",
+      "▁M ORE",
+      "▁MO RE",
+      "ge tString",
+      "get String",
+      "背后 的",
+      "背 后的",
+      "党 政",
+      "前 行",
+      "拥 抱",
+      "pol it",
+      "po lit",
+      "p olit",
+      "so und",
+      "s ound",
+      "▁anal ges",
+      "六 十",
+      "ar nings",
+      "arn ings",
+      "arning s",
+      "ey e",
+      "e ye",
+      "▁N YC",
+      "▁NY C",
+      "海 淀",
+      "座 椅",
+      "蚂 蚁",
+      "▁F T",
+      "▁ FT",
+      "党员 干部",
+      "nM ajor",
+      "n Major",
+      "直角 坐标",
+      "ir on",
+      "iro n",
+      "i ron",
+      "解决 方法",
+      "B G",
+      "л е",
+      "的 健康",
+      "▁M ans",
+      "▁Man s",
+      "▁Ma ns",
+      "▁s yst",
+      "▁sy st",
+      "▁sys t",
+      "▁surround ings",
+      "▁surrounding s",
+      "Lar ge",
+      "L arge",
+      "▁创作 年代",
+      "Output Stream",
+      "击 败",
+      "映 射",
+      "▁Rest aurant",
+      "▁sa lv",
+      "▁sal v",
+      "CRE ATE",
+      "dec oration",
+      "decor ation",
+      "发 掘",
+      "迷 茫",
+      "灾 难",
+      "回来 了",
+      "回 来了",
+      "hed uled",
+      "hedul ed",
+      "hedule d",
+      "电视 机",
+      "计 分",
+      "Mar ch",
+      "Marc h",
+      "M arch",
+      "Av ailable",
+      "往 事",
+      "切 线",
+      "nF orm",
+      "nFor m",
+      "n Form",
+      "▁de generation",
+      "▁deg eneration",
+      "班 的",
+      "▁H unt",
+      "▁Hu nt",
+      "▁Hun t",
+      "rib uting",
+      "ribut ing",
+      "宋 代",
+      "山 镇",
+      "ga p",
+      "g ap",
+      "tm argin",
+      "t margin",
+      "十九 大",
+      "▁Re form",
+      "▁Ref orm",
+      "▁s ake",
+      "▁sa ke",
+      "▁def ender",
+      "▁defe nder",
+      "▁defend er",
+      "拼 搏",
+      "De n",
+      "D en",
+      "SECRE T",
+      "SEC RET",
+      "▁ 夏",
+      "容 留",
+      "▁ 任",
+      "草 莓",
+      "het ics",
+      "hetic s",
+      "h etics",
+      "▁F inding",
+      "▁Fin ding",
+      "▁Find ing",
+      "▁sha ft",
+      "再 到",
+      "回 族",
+      "ver tex",
+      "vert ex",
+      "verte x",
+      "grad ient",
+      "▁step ped",
+      "▁ste pped",
+      "制造 商",
+      "▁sh ifting",
+      "▁shift ing",
+      "庄 村",
+      "RU N",
+      "R UN",
+      "▁b ail",
+      "▁ba il",
+      "ban ner",
+      "b anner",
+      "Serial izable",
+      "ec a",
+      "e ca",
+      "感 悟",
+      "▁t abs",
+      "▁tab s",
+      "▁ta bs",
+      "▁ tabs",
+      "form ula",
+      "merc ially",
+      "mercial ly",
+      "▁N AT",
+      "▁NA T",
+      "nGener al",
+      "nGen eral",
+      "n General",
+      "被执行人 刘",
+      "与 之",
+      "▁\\\\ :",
+      "▁ \\\\:",
+      "▁met aph",
+      "▁meta ph",
+      "不 顾",
+      "▁S kin",
+      "▁Sk in",
+      "▁Ski n",
+      "▁T een",
+      "▁Te en",
+      "▁tr unk",
+      "▁trans verse",
+      "因 本案",
+      "检 出",
+      "▁W AS",
+      "▁WA S",
+      "地 带",
+      "▁a lg",
+      "▁al g",
+      "▁ alg",
+      "▁c ush",
+      "▁cu sh",
+      "т ь",
+      "▁Cas ino",
+      "▁Cons ervative",
+      "▁Conserv ative",
+      "助 手",
+      "Tree Node",
+      "大 国",
+      "▁Re yn",
+      "▁Rey n",
+      "▁Buff alo",
+      "▁ 汉",
+      "图 标",
+      "希望 大家",
+      "uct ive",
+      "▁prevent ive",
+      "猜 测",
+      "无锡 市",
+      "pix el",
+      "p ixel",
+      "tres ult",
+      "t result",
+      "nt hey",
+      "nthe y",
+      "nth ey",
+      "n they",
+      "▁ 沈",
+      "▁tr icky",
+      "▁trick y",
+      "年 份",
+      "T ARGET",
+      "▁look up",
+      "▁ lookup",
+      "lo x",
+      "l ox",
+      "办公室 主任",
+      "g c",
+      "处 长",
+      "的 家",
+      "AW S",
+      "A WS",
+      "认识 到",
+      "▁p rone",
+      "▁pro ne",
+      "▁pr one",
+      "▁pron e",
+      "笑 着",
+      "▁Hol land",
+      "▁Holl and",
+      "▁ing redient",
+      "▁ingred ient",
+      "音 像",
+      "▁v erb",
+      "▁ve rb",
+      "▁ver b",
+      "▁ verb",
+      "all eng",
+      "alle ng",
+      "遵 义",
+      "\\\\ ,\\\\",
+      "\\\\, \\\\",
+      "▁con ductor",
+      "▁conduct or",
+      "几 率",
+      "的 消息",
+      "▁m ono",
+      "▁mon o",
+      "▁mo no",
+      "▁p anic",
+      "▁pan ic",
+      "▁pa nic",
+      "▁ panic",
+      "▁mon ument",
+      "举 例",
+      "年 出生",
+      "ch ip",
+      "chi p",
+      "c hip",
+      "▁Ind ians",
+      "▁India ns",
+      "▁Indian s",
+      "的 歌",
+      "以 人为本",
+      "il o",
+      "i lo",
+      "lat itude",
+      "▁vol untary",
+      "apeut ics",
+      "apeutic s",
+      "▁b itcoin",
+      "▁bit coin",
+      "▁ bitcoin",
+      "Ga l",
+      "G al",
+      "洗 涤",
+      "腹 泻",
+      "▁D as",
+      "▁Da s",
+      "还有 很多",
+      "还 有很多",
+      "▁d ummy",
+      "▁ dummy",
+      "▁G ardens",
+      "▁Gar dens",
+      "▁Garden s",
+      "▁Gard ens",
+      "如果 您",
+      "▁ 他",
+      "四 首",
+      "固 醇",
+      "ter ms",
+      "term s",
+      "▁Jos é",
+      "▁t ires",
+      "▁tire s",
+      "▁ti res",
+      "出 土",
+      "▁p å",
+      "▁land ed",
+      "▁lan ded",
+      "▁hist orian",
+      "▁histor ian",
+      "乾 坤",
+      "运算 法",
+      "运 算法",
+      "Link s",
+      "Lin ks",
+      "L inks",
+      "▁neut ron",
+      "▁neutr on",
+      "▁Ter ms",
+      "▁Term s",
+      "▁ Terms",
+      "我 爱",
+      "Cook ie",
+      "C ookie",
+      "▁com mits",
+      "▁comm its",
+      "▁commit s",
+      "▁ commits",
+      "at isf",
+      "ati sf",
+      "atis f",
+      "▁rab bits",
+      "▁rabbit s",
+      "▁business man",
+      "折 叠",
+      "合法 的",
+      "合 法的",
+      "zh ou",
+      "z hou",
+      "As sembly",
+      "▁tax i",
+      "▁ta xi",
+      "ju g",
+      "j ug",
+      "▁A K",
+      "▁ AK",
+      "▁Re ally",
+      "▁Real ly",
+      "的 概率",
+      "ry lic",
+      "ryl ic",
+      "r ylic",
+      "esc ope",
+      "e scope",
+      "sub process",
+      "机 动",
+      "▁G ift",
+      "▁Gi ft",
+      "平行 四边形",
+      "▁gr inding",
+      "▁grind ing",
+      "▁cl oset",
+      "▁clos et",
+      "▁close t",
+      "在 乎",
+      "is ton",
+      "ist on",
+      "isto n",
+      "i ston",
+      "▁Log ic",
+      "▁Lo gic",
+      "▁ Logic",
+      "还款 义务",
+      "个 位",
+      "个 百分",
+      "Ser ialize",
+      "Serial ize",
+      "S erialize",
+      "▁count less",
+      "▁Go nz",
+      "▁Gon z",
+      "▁techn ically",
+      "▁technical ly",
+      "nR obert",
+      "nRob ert",
+      "n Robert",
+      "日 月",
+      "nC ount",
+      "nCo unt",
+      "n Count",
+      "月 加入中国共产党",
+      "av o",
+      "a vo",
+      "or on",
+      "oro n",
+      "o ron",
+      "▁dis posal",
+      "▁disp osal",
+      "评 议",
+      "山 中",
+      "性 好",
+      "者 和",
+      "的 做法",
+      "▁dec ides",
+      "▁decide s",
+      "▁mer chant",
+      "▁merch ant",
+      "在 那",
+      "胎 儿",
+      "▁C AS",
+      "▁CA S",
+      "▁ CAS",
+      "假释 案件",
+      "广播 电视",
+      "与 管理",
+      "yel low",
+      "y ellow",
+      "de letion",
+      "delet ion",
+      "性 分析",
+      "减 为",
+      "PO INT",
+      "prov iders",
+      "provider s",
+      "provide rs",
+      "▁T at",
+      "▁Ta t",
+      "liter al",
+      "lit eral",
+      "lite ral",
+      "▁ro ller",
+      "▁roll er",
+      "▁ roller",
+      "是连载于 起点中文网",
+      "赵 某",
+      "满意 度",
+      "nPol ice",
+      "▁polym orphism",
+      "▁polymorph ism",
+      "App end",
+      "Ap pend",
+      "演 练",
+      "ga ry",
+      "gar y",
+      "g ary",
+      "elect ric",
+      "e lectric",
+      "要 知道",
+      ")} {\\\\",
+      ")}{ \\\\",
+      ") }{\\\\",
+      "cal endar",
+      "上 周",
+      "矩 形",
+      "▁Liter ature",
+      ")) {\\",
+      ") ){\\",
+      "ag ain",
+      "aga in",
+      "a gain",
+      "三 首",
+      "生 病",
+      "TRIBUT ING",
+      "▁s outheast",
+      "▁south east",
+      "▁perman ently",
+      "▁permanent ly",
+      "▁b ovine",
+      "▁bo vine",
+      "▁rel apse",
+      "▁gly cos",
+      "▁glyc os",
+      "t virtual",
+      "▁R ect",
+      "▁Re ct",
+      "▁Rec t",
+      "▁ Rect",
+      "▁M ason",
+      "▁Ma son",
+      "▁Mas on",
+      "Lo ve",
+      "L ove",
+      "低 调",
+      "Cl ean",
+      "C lean",
+      "▁micro scope",
+      "逍 遥",
+      "可 根据",
+      "老 子",
+      "af fe",
+      "aff e",
+      "▁H ousing",
+      "▁Ho using",
+      "▁Hou sing",
+      "AD E",
+      "A DE",
+      "ze ta",
+      "z eta",
+      "▁H ep",
+      "▁He p",
+      "一 号",
+      "▁S ymbol",
+      "▁Sym bol",
+      "▁ Symbol",
+      "▁an ime",
+      "▁anim e",
+      "▁Alex and",
+      "▁Alexa nd",
+      "走 上",
+      "供 水",
+      "亚马 逊",
+      "▁e merge",
+      "▁emer ge",
+      "▁emerg e",
+      "负 载",
+      "ta l",
+      "t al",
+      "▁T it",
+      "▁Ti t",
+      "▁split ting",
+      "▁spl itting",
+      "会计 师",
+      "▁A ims",
+      "▁Aim s",
+      "彩 虹",
+      "也 非常",
+      "▁de legate",
+      "▁deleg ate",
+      "▁ delegate",
+      "ha us",
+      "h aus",
+      "au c",
+      "a uc",
+      "灵 感",
+      "nL ocation",
+      "nLo cation",
+      "n Location",
+      "每 位",
+      "老 化",
+      "迷 你",
+      "制度 的",
+      "制 度的",
+      "▁hist ological",
+      "▁histologic al",
+      "▁t u",
+      "▁ tu",
+      "前面 的",
+      "前 面的",
+      "oph yll",
+      "百 合",
+      "() {",
+      "( ){",
+      "抢劫 罪",
+      "▁metast ases",
+      "а н",
+      "应 付",
+      "▁Que ue",
+      "▁ Queue",
+      "端 的",
+      "▁stiff ness",
+      "备 份",
+      "▁i Tunes",
+      "▁W es",
+      "▁We s",
+      "▁coal ition",
+      "村 镇",
+      "ak o",
+      "a ko",
+      "课堂 教学",
+      "一 股",
+      "▁vis ually",
+      "▁visual ly",
+      "’ \\",
+      "心 病",
+      "In cre",
+      "Inc re",
+      "Ex tra",
+      "Ext ra",
+      "设计 和",
+      "▁terror ist",
+      "asc ii",
+      "asci i",
+      "ph thal",
+      "长 寿",
+      "如何 在",
+      "如 何在",
+      "▁mal aria",
+      "应 由",
+      "做 饭",
+      "组织 实施",
+      "▁* **",
+      "▁** *",
+      "▁ ***",
+      "ist ries",
+      "istr ies",
+      "▁fluct uations",
+      "五 一",
+      "注 释",
+      "SP ACE",
+      "SPA CE",
+      "▁s print",
+      "▁sp rint",
+      "▁spr int",
+      "▁trans forms",
+      "▁transform s",
+      "▁transf orms",
+      "该 行",
+      "身 后",
+      "▁f el",
+      "▁fe l",
+      "动 机",
+      "湿 地",
+      "▁S nap",
+      "▁Sn ap",
+      "▁ Snap",
+      "▁p ian",
+      "▁pi an",
+      "nRe view",
+      "nRev iew",
+      "n Review",
+      "▁sulf ate",
+      "续 行",
+      "im ony",
+      "imon y",
+      "imo ny",
+      "c stdio",
+      "s uffix",
+      "▁card inal",
+      "▁cardi nal",
+      "Net Core",
+      "Ex ternal",
+      "Ext ernal",
+      "is en",
+      "ise n",
+      "i sen",
+      "▁line arly",
+      "▁linear ly",
+      "▁lin early",
+      "// /\\",
+      "/// \\",
+      "/ //\\",
+      "在 法律规定",
+      "仅 有",
+      "农 机",
+      "▁T F",
+      "▁ TF",
+      "per form",
+      "perf orm",
+      "▁b ishop",
+      "▁bi shop",
+      "▁bis hop",
+      "▁ bishop",
+      "▁lar vae",
+      "中 专",
+      "an to",
+      "ant o",
+      "a nto",
+      "▁K ash",
+      "▁Ka sh",
+      "▁Kas h",
+      "▁T hor",
+      "▁Th or",
+      "▁T aking",
+      "▁Ta king",
+      "▁Tak ing",
+      "▁c ausal",
+      "▁ca usal",
+      "▁caus al",
+      "▁War ner",
+      "▁C orporate",
+      "▁Corpor ate",
+      "▁Corp orate",
+      "加 法",
+      "act ional",
+      "action al",
+      "▁num eric",
+      "▁numer ic",
+      "▁nu meric",
+      "▁ numeric",
+      "帽 子",
+      "▁B MI",
+      "▁BM I",
+      "▁ BMI",
+      "主要 用于",
+      "▁ex changes",
+      "▁exchange s",
+      "犯 贩卖毒品罪",
+      "hy p",
+      "h yp",
+      "▁a xi",
+      "▁ax i",
+      "nd ocker",
+      "n docker",
+      "一 经",
+      "en ti",
+      "ent i",
+      "▁l over",
+      "▁love r",
+      "▁lo ver",
+      "▁ lover",
+      "▁trigger s",
+      "▁trig gers",
+      "x f",
+      "▁innov ations",
+      "▁innovation s",
+      "这段 话",
+      "pre pare",
+      "prep are",
+      "AL S",
+      "A LS",
+      "中山 市",
+      "中 山市",
+      "▁inter med",
+      "в о",
+      "▁free l",
+      "▁fre el",
+      "▁mot ors",
+      "▁motor s",
+      "▁compos ers",
+      "▁composer s",
+      "▁compose rs",
+      "的 视频",
+      "批 量",
+      "青海 省",
+      "布 拉",
+      "手 游",
+      "sp in",
+      "spi n",
+      "s pin",
+      "油 脂",
+      "熊 猫",
+      "我 感觉",
+      "为什么 会",
+      "nG old",
+      "nGo ld",
+      "n Gold",
+      "▁B orough",
+      "▁Bo rough",
+      "▁Bor ough",
+      "率 高",
+      "ul os",
+      "ulo s",
+      "u los",
+      "智能 手机",
+      "▁account ed",
+      "▁home owners",
+      "▁homeowner s",
+      "DA TE",
+      "DAT E",
+      "D ATE",
+      "Ne ed",
+      "N eed",
+      "▁L ost",
+      "▁Los t",
+      "▁Lo st",
+      "人 了",
+      "foot ball",
+      "f ootball",
+      "▁encour ages",
+      "▁encourage s",
+      "Al pha",
+      "▁L akes",
+      "▁La kes",
+      "▁Lake s",
+      "▁Lak es",
+      "▁veget ation",
+      "▁im plantation",
+      "▁implant ation",
+      "啥 的",
+      "▁r é",
+      "▁ ré",
+      "Spec ial",
+      "▁p ension",
+      "▁pens ion",
+      "▁imped ance",
+      "▁br acket",
+      "▁necess ity",
+      "本 市",
+      "▁F light",
+      "▁Fl ight",
+      "退 回",
+      "text area",
+      "▁ind igenous",
+      "▁Cont emporary",
+      "Path s",
+      "Pat hs",
+      "Pa ths",
+      "P aths",
+      "政策 的",
+      "▁com pressed",
+      "▁comp ressed",
+      "▁compr essed",
+      "▁compress ed",
+      "▁ compressed",
+      "▁construct ing",
+      "神 的",
+      "▁ ）",
+      "而 后",
+      "Aut hent",
+      "Auth ent",
+      "▁nit ric",
+      "过 大",
+      "▁yield ed",
+      "aut hentication",
+      "authent ication",
+      "▁ đ",
+      "니 다",
+      "▁b asin",
+      "▁bas in",
+      "▁ba sin",
+      "▁de hydro",
+      "▁dehyd ro",
+      "] <",
+      "亲 密",
+      "执 政",
+      "演 示",
+      "▁b ub",
+      "▁bu b",
+      "▁sh ine",
+      "▁ shine",
+      "种植 业",
+      "▁ha em",
+      "▁t odo",
+      "▁to do",
+      "▁ todo",
+      "▁pot atoes",
+      "▁potato es",
+      "IN S",
+      "I NS",
+      "ag le",
+      "ag ulation",
+      "agu lation",
+      "agul ation",
+      "▁screen shot",
+      "▁screens hot",
+      "▁ screenshot",
+      "▁unem ployment",
+      "韭 菜",
+      "Sign al",
+      "Sig nal",
+      "▁purs uit",
+      "委 屈",
+      "}} $$",
+      "}}$ $",
+      "} }$$",
+      "在此 基础上",
+      "V s",
+      "诱 惑",
+      "▁P rivacy",
+      "▁Priv acy",
+      "▁Tr ading",
+      "▁Tra ding",
+      "▁Trad ing",
+      "▁Id entity",
+      "▁Ident ity",
+      "▁ Identity",
+      "cont ainers",
+      "container s",
+      "送 给",
+      "nT otal",
+      "nTo tal",
+      "n Total",
+      "▁asympt otic",
+      "研究 和",
+      "▁V IP",
+      "▁VI P",
+      "▁ VIP",
+      "▁with d",
+      "▁wit hd",
+      "nP rov",
+      "nPro v",
+      "nPr ov",
+      "n Prov",
+      "in h",
+      "i nh",
+      "平方 根",
+      "stud y",
+      "Xi v",
+      "X iv",
+      "pre hens",
+      "▁hum idity",
+      "▁humid ity",
+      "▁cost ume",
+      "▁Reg iment",
+      "pt ive",
+      "A Q",
+      "硬 度",
+      "▁inj ections",
+      "▁inject ions",
+      "▁injection s",
+      "有 余",
+      "着 急",
+      "cg i",
+      "c gi",
+      "lo s",
+      "l os",
+      "an dem",
+      "and em",
+      "ande m",
+      "争 夺",
+      "▁boy friend",
+      "鲜 明",
+      "研讨 会",
+      "pe ak",
+      "Log ging",
+      "▁Compar ison",
+      "▁ Comparison",
+      "清 明",
+      "li me",
+      "lim e",
+      "l ime",
+      "▁E ND",
+      "▁EN D",
+      "▁ END",
+      "的 处理",
+      "nN on",
+      "nNo n",
+      "n Non",
+      "▁install ations",
+      "▁installation s",
+      "ask ell",
+      "▁e nters",
+      "▁en ters",
+      "▁ent ers",
+      "▁enter s",
+      "乐 器",
+      "▁image ry",
+      "▁imag ery",
+      "▁cour tesy",
+      "室 友",
+      "隆 重",
+      "▁n ano",
+      "▁nan o",
+      "▁na no",
+      "▁ nano",
+      "和 方法",
+      "▁Spring er",
+      "▁Spr inger",
+      "Mo unt",
+      "M ount",
+      "一 处",
+      "na i",
+      "n ai",
+      "▁Nash ville",
+      "Cred entials",
+      "Credential s",
+      "欧 元",
+      "nCommun es",
+      "em ph",
+      "emp h",
+      "e mph",
+      "me ss",
+      "mes s",
+      "m ess",
+      "▁fin anc",
+      "ov an",
+      "ova n",
+      "o van",
+      "只是 一个",
+      "只是一 个",
+      "只 是一个",
+      "ac erb",
+      "ace rb",
+      "acer b",
+      "Un ique",
+      "十三 条之规定",
+      "十三条 之规定",
+      "十 三条之规定",
+      "▁S aints",
+      "▁Saint s",
+      "▁Sa ints",
+      "▁m ere",
+      "▁me re",
+      "▁mer e",
+      "▁ mere",
+      "横 向",
+      "▁DI Y",
+      "▁ DIY",
+      "华 南",
+      "▁tour naments",
+      "▁tournament s",
+      "了 对",
+      "尤 为",
+      "可以 看出",
+      "可以看 出",
+      "nc har",
+      "nch ar",
+      "n char",
+      "Ro und",
+      "R ound",
+      "er ala",
+      "era la",
+      "eral a",
+      "par ents",
+      "parent s",
+      "paren ts",
+      "实际 应用",
+      "sub str",
+      "模式 的",
+      "模 式的",
+      "▁A ber",
+      "▁Ab er",
+      "▁sat uration",
+      "FU N",
+      "F UN",
+      "fa s",
+      "f as",
+      "我 非常",
+      "nim age",
+      "n image",
+      "▁Vin cent",
+      "▁Vince nt",
+      "▁Vinc ent",
+      "▁F loat",
+      "▁Flo at",
+      "▁ Float",
+      "le ted",
+      "let ed",
+      "lete d",
+      "l eted",
+      "▁C arn",
+      "▁Car n",
+      "▁Ca rn",
+      "▁gran ite",
+      "早 日",
+      "过 高",
+      "▁Sym fony",
+      "▁ 吕",
+      "ps is",
+      "psi s",
+      "p sis",
+      "▁neighbor ing",
+      "▁neighb oring",
+      "▁( >",
+      "▁K ol",
+      "▁Ko l",
+      "规范 化",
+      "▁e mp",
+      "▁em p",
+      "▁ emp",
+      "▁Plan et",
+      "▁Plane t",
+      "▁F P",
+      "▁ FP",
+      "om ed",
+      "ome d",
+      "o med",
+      "档 次",
+      "盲 目",
+      "▁M end",
+      "▁Me nd",
+      "▁Men d",
+      "▁tra umatic",
+      "▁trauma tic",
+      "▁ traumatic",
+      "ic in",
+      "ici n",
+      "i cin",
+      "▁line up",
+      "sh ots",
+      "shot s",
+      "s hots",
+      "▁ar riving",
+      "▁arr iving",
+      "▁arriv ing",
+      "▁embed ding",
+      "▁ 田",
+      "我 省",
+      "震 荡",
+      "心血 管",
+      "心 血管",
+      "Al ex",
+      "Ale x",
+      "A lex",
+      "bo ro",
+      "bor o",
+      "b oro",
+      "▁I oT",
+      "▁Io T",
+      "▁ IoT",
+      "另 有",
+      "规 章",
+      "inner HTML",
+      "ar ial",
+      "ari al",
+      "aria l",
+      "a rial",
+      "特 斯拉",
+      "Com mit",
+      "Comm it",
+      "终 极",
+      "epend ence",
+      "▁Arch ives",
+      "▁Archive s",
+      "ant ics",
+      "antic s",
+      "anti cs",
+      "▁celebr ity",
+      "▁ 波",
+      "鞋 子",
+      "▁Par ad",
+      "▁Pa rad",
+      "▁Para d",
+      "中 场",
+      "锁 定",
+      "拉 开",
+      "随机 分为",
+      "est ine",
+      "▁mess aging",
+      "▁ messaging",
+      "断 裂",
+      "的 反",
+      "ched ule",
+      "che dule",
+      "c hedule",
+      "▁unw anted",
+      "▁non profit",
+      "▁framework s",
+      "▁frame works",
+      "▁hor izon",
+      "▁horiz on",
+      "▁ne uron",
+      "▁neuro n",
+      "▁neur on",
+      "柏 林",
+      "▁re marks",
+      "▁rem arks",
+      "▁remark s",
+      "▁ remarks",
+      "将 以",
+      "▁M ul",
+      "▁Mu l",
+      "ca ched",
+      "cache d",
+      "c ached",
+      "en ames",
+      "ename s",
+      "ena mes",
+      "e names",
+      "中国人民 银行同期",
+      "中国人民银行 同期",
+      "▁roll s",
+      "Init ialize",
+      "Initial ize",
+      "公 关",
+      "体 外",
+      "GP S",
+      "G PS",
+      "类似 于",
+      "▁U RI",
+      "▁UR I",
+      "▁ URI",
+      "In line",
+      "I nline",
+      "Control s",
+      "▁Sher iff",
+      "沃 尔",
+      "dir ected",
+      "direct ed",
+      "d irected",
+      "▁$ (\\\\",
+      "▁$( \\\\",
+      "驾驶 证",
+      "isec onds",
+      "i seconds",
+      "改 良",
+      "道路 交通事故",
+      "战斗 力",
+      "▁ 简体中文",
+      "▁T as",
+      "▁Ta s",
+      "▁pop ulated",
+      "▁populate d",
+      "▁popul ated",
+      "( %",
+      "bo ys",
+      "boy s",
+      "b oys",
+      "▁N HS",
+      "▁NH S",
+      "▁ago nist",
+      "▁agon ist",
+      "uss ia",
+      "mo ment",
+      "m oment",
+      "不 被",
+      "肠 胃",
+      "装 配",
+      "ha ck",
+      "h ack",
+      "▁F unctional",
+      "▁Function al",
+      "▁ Functional",
+      "▁c omics",
+      "▁com ics",
+      "▁comic s",
+      "▁average d",
+      "▁aver aged",
+      "奶 油",
+      "紫 色",
+      "冷 藏",
+      "(\" [",
+      "( \"[",
+      "直角 三角形",
+      "普 查",
+      "▁fl oral",
+      "▁flo ral",
+      "▁flora l",
+      "▁flor al",
+      "reg ation",
+      "par sed",
+      "parse d",
+      "pars ed",
+      "▁Pre mium",
+      "▁Prem ium",
+      "第五百一十九 条规定",
+      "第五百一十九条 规定",
+      "ag h",
+      "a gh",
+      "总 价",
+      "VID EO",
+      "V IDEO",
+      "ra form",
+      "raf orm",
+      "▁un treated",
+      "▁unt reated",
+      "▁set uptools",
+      "▁setup tools",
+      "多 为",
+      "RA Y",
+      "R AY",
+      "内心 的",
+      "内 心的",
+      "▁fabric ation",
+      "盒 子",
+      "经 鉴定",
+      "ling ton",
+      "l ington",
+      "Ans wer",
+      "A nswer",
+      "薪 酬",
+      "有意 思",
+      "有 意思",
+      "▁p ile",
+      "▁pi le",
+      "▁pil e",
+      "ut down",
+      "▁T rophy",
+      "▁Tro phy",
+      "ap ses",
+      "aps es",
+      "apse s",
+      "w aukee",
+      "团 委",
+      "程序 员",
+      "▁S MS",
+      "▁SM S",
+      "▁ SMS",
+      "▁pl umbing",
+      "▁W u",
+      "or ig",
+      "ori g",
+      "o rig",
+      "案件 适用法律",
+      "quir ies",
+      "▁Be auty",
+      "▁Beaut y",
+      "▁Bea uty",
+      "▁Beau ty",
+      "▁S HO",
+      "▁SH O",
+      "▁decor ative",
+      "第九 章",
+      "第 九章",
+      "▁r ip",
+      "▁ri p",
+      "▁ rip",
+      "▁N aval",
+      "▁Na val",
+      "▁Nav al",
+      "一 提",
+      "考 证",
+      "Cal endar",
+      "医疗 器械",
+      "▁C elebr",
+      "▁legit imate",
+      "东 营",
+      "第五百一十九 条之规定",
+      "第五百一十九条 之规定",
+      "▁problem atic",
+      "之 二",
+      "万 一",
+      "Ho r",
+      "H or",
+      "author ized",
+      "authorize d",
+      "挂 牌",
+      "AM S",
+      "A MS",
+      "Sa n",
+      "S an",
+      "▁mon ol",
+      "▁mo nol",
+      "▁mono l",
+      "▁accum ulated",
+      "▁accumulate d",
+      "▁elim inating",
+      "▁L ud",
+      "▁Lu d",
+      "Em bed",
+      "E mbed",
+      "gr unt",
+      "世 的",
+      "Ex ist",
+      "nS pan",
+      "nSp an",
+      "n Span",
+      "手 上",
+      "CA DE",
+      "CAD E",
+      "C ADE",
+      "∈ (",
+      "▁U TC",
+      "▁UT C",
+      "▁ UTC",
+      "所 能",
+      "并不 能",
+      "并 不能",
+      "ac ao",
+      "aca o",
+      "serial izer",
+      "serialize r",
+      "s erializer",
+      "设计 与",
+      "马 尔",
+      "都不 能",
+      "都 不能",
+      "竣 工",
+      "衰 老",
+      "言 语",
+      "gr o",
+      "g ro",
+      "组 数据",
+      "▁concer ts",
+      "▁conc erts",
+      "▁concert s",
+      "▁H F",
+      "▁ HF",
+      "新疆 维吾尔",
+      "▁res pected",
+      "▁respect ed",
+      "▁resp ected",
+      "光 荣",
+      "▁Ro gers",
+      "▁Rog ers",
+      "▁Roger s",
+      "▁A zerbai",
+      "大 豆",
+      "ob l",
+      "o bl",
+      "is se",
+      "iss e",
+      "CS F",
+      "C SF",
+      "nw ill",
+      "n will",
+      "un day",
+      "und ay",
+      "unda y",
+      "▁H g",
+      "▁ Hg",
+      "申请 恢复执行",
+      "P o",
+      "▁Map le",
+      "▁Ma ple",
+      "▁under neath",
+      "je st",
+      "j est",
+      "Ra nk",
+      "R ank",
+      "▁satisf actory",
+      "原审 被告人",
+      "原审被告 人",
+      "未 必",
+      "▁Di vis",
+      "▁Div is",
+      "tr ained",
+      "tra ined",
+      "train ed",
+      "▁similar ities",
+      "腹 腔",
+      "被执行人 陈",
+      "▁adjust ing",
+      "GP IO",
+      "G PIO",
+      "}} $$\\",
+      "}}$ $\\",
+      "}}$$ \\",
+      "} }$$\\",
+      "和 她",
+      "关注 我",
+      "▁Sc hw",
+      "▁Sch w",
+      "sl ot",
+      "s lot",
+      "▁M ario",
+      "▁Mar io",
+      "▁Ma rio",
+      "▁Mari o",
+      "min imum",
+      "▁b ears",
+      "▁be ars",
+      "▁bear s",
+      "coord inates",
+      "coordinate s",
+      "▁m ul",
+      "▁mu l",
+      "▁ mul",
+      "▁S equence",
+      "▁Sequ ence",
+      "▁ Sequence",
+      "nH ar",
+      "nHa r",
+      "n Har",
+      "▁F le",
+      "▁Fl e",
+      "这篇 文章",
+      "这 篇文章",
+      "▁b ile",
+      "▁bi le",
+      "▁bil e",
+      "▁k iss",
+      "▁ki ss",
+      "卫生 和计划生育局",
+      "上 演",
+      "zh ong",
+      "z hong",
+      "▁r ays",
+      "▁ra ys",
+      "▁ray s",
+      "▁ rays",
+      "▁ar ises",
+      "▁arise s",
+      "▁Fif th",
+      "▁sym bolic",
+      "▁symbol ic",
+      "特定 的",
+      "特 定的",
+      "▁down loading",
+      "▁download ing",
+      "的歌 曲",
+      "的 歌曲",
+      "采取 保全措施",
+      "m f",
+      "Ob server",
+      "Observ er",
+      "Obs erver",
+      "契 机",
+      "自然 资源",
+      "▁p alm",
+      "▁pal m",
+      "▁pa lm",
+      "▁Valent ine",
+      "是 无",
+      "推荐 了",
+      "▁Ben ef",
+      "bi c",
+      "b ic",
+      "▁sc ipy",
+      "▁sci py",
+      "▁s illy",
+      "▁sil ly",
+      "▁diss olved",
+      "让 大家",
+      "ta p",
+      "t ap",
+      "的 平均",
+      "ag greg",
+      "agg reg",
+      "伊 斯",
+      "▁clean up",
+      "▁ cleanup",
+      "cri ptions",
+      "cription s",
+      "cript ions",
+      "成 像",
+      "▁de bris",
+      "▁deb ris",
+      "喜 好",
+      "纯 收入",
+      "▁Pay ment",
+      "▁ Payment",
+      "U l",
+      "新一 代",
+      "新 一代",
+      "▁e long",
+      "▁el ong",
+      "▁ elong",
+      "▁fire place",
+      "▁fi replace",
+      "ific ent",
+      "ifi cent",
+      "ifice nt",
+      "日 后",
+      "的事 件",
+      "的 事件",
+      "▁f ool",
+      "▁fo ol",
+      "▁foo l",
+      "▁p ockets",
+      "▁pocket s",
+      "Syn tax",
+      "S yntax",
+      "▁A bu",
+      "▁Ab u",
+      "工作 经验",
+      "▁incorpor ates",
+      "▁incorporate s",
+      "▁index es",
+      "▁ indexes",
+      "不会 有",
+      "不 会有",
+      "os it",
+      "osi t",
+      "o sit",
+      "worth y",
+      "wort hy",
+      "wor thy",
+      "w orthy",
+      "康 熙",
+      "▁en rollment",
+      "▁enroll ment",
+      "汽 油",
+      "▁l ac",
+      "▁la c",
+      "可以 帮助",
+      "▁Ex pand",
+      "▁Exp and",
+      "▁ Expand",
+      "▁ 江",
+      "Can vas",
+      "pl aying",
+      "play ing",
+      "nRe member",
+      "nRem ember",
+      "n Remember",
+      "启 用",
+      "的 四",
+      "的 照片",
+      "▁Act ually",
+      "▁Actual ly",
+      "若 有",
+      "▁已 完结",
+      "▁craft ed",
+      "▁ crafted",
+      "▁Dis covery",
+      "▁Disc overy",
+      "▁Discover y",
+      "▁ Discovery",
+      "共 青",
+      "ex ist",
+      "▁O scar",
+      "▁Os car",
+      "南 宁",
+      "▁hon estly",
+      "▁honest ly",
+      "再 把",
+      "特 此",
+      "保证 金",
+      "与被 申请人",
+      "与 被申请人",
+      "mon ey",
+      "mo ney",
+      "m oney",
+      "▁P rove",
+      "▁Pro ve",
+      "▁Pr ove",
+      "▁Prov e",
+      "固定 资产",
+      "工作 会议",
+      "oprote ins",
+      "oprotein s",
+      "▁inf erior",
+      "▁infer ior",
+      "勇 于",
+      "<\\ \\",
+      "< \\\\",
+      "▁penet ration",
+      "▁penetr ation",
+      "▁sup ern",
+      "▁super n",
+      "nw hen",
+      "n when",
+      "▁ch ore",
+      "▁cho re",
+      "▁chor e",
+      "▁th rew",
+      "▁thr ew",
+      "▁O thers",
+      "▁Other s",
+      "患 有",
+      "▁h iding",
+      "▁hi ding",
+      "▁M ari",
+      "▁Mar i",
+      "▁Ma ri",
+      "▁K aren",
+      "▁Kar en",
+      "▁Ka ren",
+      "在 地",
+      "难 过",
+      "▁Answer s",
+      "▁Ans wers",
+      "疑 惑",
+      "nR ussian",
+      "n Russian",
+      "的人 群",
+      "的 人群",
+      "复 仇",
+      "Bo und",
+      "B ound",
+      "▁diff use",
+      "▁Pop ulation",
+      "锂 电池",
+      "当事人 有权",
+      "▁electro ly",
+      "▁electr oly",
+      "El ect",
+      "Ele ct",
+      "E lect",
+      "▁S ah",
+      "▁Sa h",
+      "▁ Sah",
+      "▁conjug ate",
+      "奖励 审批表",
+      "▁K er",
+      "▁Ke r",
+      "反映 了",
+      "北 海",
+      "▁Es cherichia",
+      "▁l ined",
+      "▁li ned",
+      "▁line d",
+      "▁lin ed",
+      "▁ lined",
+      "▁dec oration",
+      "▁decor ation",
+      "▁ decoration",
+      "蜘 蛛",
+      "fer s",
+      "fe rs",
+      "f ers",
+      "▁ 可",
+      "确 已破裂",
+      "▁human ity",
+      "▁imag inary",
+      "ij k",
+      "i jk",
+      "ios ity",
+      "i osity",
+      "ran ial",
+      "Requ ests",
+      "Request s",
+      "快 来",
+      "▁u buntu",
+      "▁ub untu",
+      "▁ ubuntu",
+      "▁intern ally",
+      "▁internal ly",
+      "也无 法",
+      "也 无法",
+      "创造 了",
+      "nR ivers",
+      "▁cyt otoxic",
+      "▁cytotox ic",
+      "▁cyto toxic",
+      "▁term ination",
+      "创建 了",
+      "▁disappe ared",
+      "▁disappear ed",
+      "▁P ra",
+      "▁Pr a",
+      "▁sample d",
+      "▁sam pled",
+      "微 量",
+      "起 床",
+      "ad j",
+      "a dj",
+      "的 依据",
+      "▁educ ate",
+      "▁method ologies",
+      "戏 曲",
+      "▁ 背景",
+      "Ma l",
+      "M al",
+      "所说 的",
+      "所 说的",
+      "完成 生产任务",
+      "流 域",
+      "is le",
+      "isl e",
+      "▁Hash Map",
+      "▁ HashMap",
+      "▁dent ist",
+      "武 侠",
+      "门 的",
+      "▁d efer",
+      "▁de fer",
+      "▁def er",
+      "▁defe r",
+      "▁ defer",
+      "at an",
+      "ata n",
+      "a tan",
+      "on ian",
+      "onia n",
+      "oni an",
+      "▁c aus",
+      "▁ca us",
+      "Sign ature",
+      "pi cs",
+      "pic s",
+      "p ics",
+      "▁T ensor",
+      "▁ Tensor",
+      "震 惊",
+      "矿 山",
+      "▁w rist",
+      "▁wr ist",
+      "的 理念",
+      "Link ed",
+      "▁b inds",
+      "▁bin ds",
+      "▁bind s",
+      "▁ut ter",
+      "▁ utter",
+      "麦克 风",
+      "外 表",
+      "Ex change",
+      "っ て",
+      "急 救",
+      "门 前",
+      "\\' >",
+      "方 言",
+      "▁Del ivery",
+      "▁ Delivery",
+      "有 意",
+      "已 在",
+      "衣 柜",
+      "OS T",
+      "O ST",
+      "▁M ob",
+      "▁Mo b",
+      "▁prohib ited",
+      "▁C itation",
+      "▁Cit ation",
+      "▁cre atures",
+      "▁creat ures",
+      "▁creature s",
+      "▁R OS",
+      "▁RO S",
+      "▁ ROS",
+      "am ino",
+      "amin o",
+      "ami no",
+      "▁D ynamics",
+      "▁Dynamic s",
+      "▁Dynam ics",
+      "google apis",
+      "至 此",
+      "pa ste",
+      "past e",
+      "pas te",
+      "p aste",
+      "▁ex otic",
+      "▁Th under",
+      "▁Thu nder",
+      "“ \\",
+      "▁G E",
+      "▁ GE",
+      "▁A MD",
+      "▁AM D",
+      "▁ AMD",
+      "▁V ista",
+      "▁Vis ta",
+      "▁Vi sta",
+      "▁comment ary",
+      "▁micro s",
+      "▁mic ros",
+      "j upiter",
+      "▁cert ificates",
+      "▁certificate s",
+      "半 天",
+      "▁per su",
+      "▁pers u",
+      "▁ 目",
+      "有一 点",
+      "有 一点",
+      "闺 蜜",
+      "▁Ne il",
+      "nc ol",
+      "nco l",
+      "n col",
+      "▁P or",
+      "▁Po r",
+      "▁fol ders",
+      "▁folder s",
+      "▁fold ers",
+      "▁ folders",
+      "过 分",
+      "en os",
+      "eno s",
+      "e nos",
+      "▁tr aces",
+      "▁tra ces",
+      "▁trace s",
+      "是 目前",
+      "演唱 会",
+      "x e",
+      "烧 烤",
+      "▁an tis",
+      "▁ant is",
+      "▁anti s",
+      "▁ antis",
+      "He x",
+      "H ex",
+      "▁c i",
+      "▁ ci",
+      "▁Med ium",
+      "▁ Medium",
+      "▁S lovak",
+      "stack exchange",
+      "勉 强",
+      "占 领",
+      "效力 的",
+      "效 力的",
+      "▁n orms",
+      "▁nor ms",
+      "▁norm s",
+      "loc ations",
+      "location s",
+      "l ocations",
+      "▁ 与",
+      "叶 子",
+      "▁s iblings",
+      "▁sibling s",
+      "▁c b",
+      "▁ cb",
+      "se ry",
+      "ser y",
+      "s ery",
+      "se nal",
+      "sen al",
+      "▁H orse",
+      "▁Hor se",
+      "aff ected",
+      "▁Sy nd",
+      "▁Syn d",
+      "▁al ien",
+      "▁ali en",
+      "▁n erves",
+      "▁nerve s",
+      "▁nerv es",
+      "▁Requ ired",
+      "▁ Required",
+      "▁fer tility",
+      "▁fert ility",
+      "▁fertil ity",
+      "▁I NT",
+      "▁IN T",
+      "▁ INT",
+      "▁Reg istration",
+      "▁ Registration",
+      "▁Not Implemented",
+      "只 为",
+      "笑 了",
+      "inv oke",
+      "本裁定书送达后 即发生法律效力",
+      "▁M ETHODS",
+      "日本 的",
+      "日 本的",
+      "▁{ }\"",
+      "▁{} \"",
+      "▁ {}\"",
+      "▁f org",
+      "▁for g",
+      "▁fo rg",
+      "nSh ould",
+      "n Should",
+      "▁s uffix",
+      "▁suff ix",
+      "▁ suffix",
+      "▁ep igen",
+      "之 作",
+      "呈现 出",
+      "行 审",
+      "静 电",
+      "▁N orman",
+      "▁Nor man",
+      "▁Norm an",
+      "转 身",
+      "▁L an",
+      "▁La n",
+      "yl ene",
+      "yle ne",
+      "不 负",
+      "每个 人的",
+      "每个人 的",
+      "每 个人的",
+      "▁C uba",
+      "▁Cub a",
+      "▁Cu ba",
+      "第二 届",
+      "nA ng",
+      "nAn g",
+      "n Ang",
+      "▁A SC",
+      "▁AS C",
+      "▁ ASC",
+      "▁R as",
+      "▁Ra s",
+      "也是 一个",
+      "也是一 个",
+      "也 是一个",
+      "乾 隆",
+      "根 廷",
+      "na sync",
+      "nas ync",
+      "n async",
+      "▁gal axy",
+      "▁put ative",
+      "头 市",
+      "▁S ax",
+      "▁Sa x",
+      "河 市",
+      "” ——",
+      "▁T ib",
+      "▁Ti b",
+      "th reads",
+      "thread s",
+      "▁ 推荐",
+      "▁un common",
+      "权 重",
+      "梦 幻",
+      "▁D one",
+      "▁Do ne",
+      "▁Don e",
+      "▁ Done",
+      "▁dynam ically",
+      "▁dynamic ally",
+      "▁dynamical ly",
+      "公 约",
+      "干 什么",
+      "ax ies",
+      "AR I",
+      "A RI",
+      "t reatment",
+      "▁imp acted",
+      "▁impact ed",
+      "阻 力",
+      "nm odel",
+      "nmod el",
+      "n model",
+      "向 法院",
+      "▁H PV",
+      "▁HP V",
+      "▁ HPV",
+      "reg istration",
+      "半 夜",
+      "atern ity",
+      "w g",
+      "ji ng",
+      "jin g",
+      "j ing",
+      "rec ipe",
+      "recip e",
+      "▁muc osal",
+      "▁mucosa l",
+      "▁week ends",
+      "▁weekend s",
+      "'] );\\",
+      "']) ;\\",
+      "' ]);\\",
+      "显 微",
+      "黄 某",
+      "家里 的",
+      "家 里的",
+      "▁Stand ards",
+      "▁Standard s",
+      "吉林 市",
+      "宜 宾",
+      "牛 仔",
+      "CL IENT",
+      "CLI ENT",
+      "▁c odec",
+      "▁co dec",
+      "▁code c",
+      "▁cod ec",
+      "▁ codec",
+      "即 是",
+      "nS hare",
+      "nSh are",
+      "n Share",
+      "▁El astic",
+      "▁ Elastic",
+      "▁hold ers",
+      "▁hol ders",
+      "▁holder s",
+      "▁ holders",
+      "窗 户",
+      "九 个月",
+      "as px",
+      "asp x",
+      "▁Ex am",
+      "nAl bums",
+      "能 认罪服法",
+      "▁R aman",
+      "▁Ram an",
+      "▁Ra man",
+      "▁as sembl",
+      "▁E specially",
+      "▁Es pecially",
+      "▁M ale",
+      "▁Mal e",
+      "▁Ma le",
+      "bul let",
+      "bull et",
+      "▁mut agen",
+      "▁tr aders",
+      "▁trad ers",
+      "▁tra ders",
+      "▁trade rs",
+      "▁trader s",
+      "▁pro ceedings",
+      "▁proceed ings",
+      "▁proceeding s",
+      "Ent ities",
+      "台州 市",
+      "台 州市",
+      "菲 律",
+      "▁Engine er",
+      "▁exp osures",
+      "▁exposure s",
+      "▁ 曾",
+      "▁l ounge",
+      "▁lo unge",
+      "▁lou nge",
+      "▁F rame",
+      "▁Fr ame",
+      "▁Fra me",
+      "▁ Frame",
+      "Sc reens",
+      "Screen s",
+      "S creens",
+      "▁Sc anner",
+      "▁Scan ner",
+      "▁ Scanner",
+      "SE RT",
+      "SER T",
+      "S ERT",
+      "com it",
+      "co mit",
+      "一 心",
+      "总数 的",
+      "总 数的",
+      "▁u ro",
+      "▁ur o",
+      "▁ uro",
+      "晚 会",
+      "法 兰",
+      "▁ch arming",
+      "▁char ming",
+      "▁charm ing",
+      "Jac k",
+      "J ack",
+      "基础 知识",
+      "ac tin",
+      "act in",
+      "he ets",
+      "heet s",
+      "hee ts",
+      "▁# ###",
+      "▁## ##",
+      "▁### #",
+      "▁ ####",
+      "时 常",
+      "吹 风机",
+      "\\' (",
+      "\\ '(",
+      "继续 执行",
+      "中心 支公司",
+      "▁aim ing",
+      "▁ai ming",
+      "[ @",
+      "之 美",
+      "抗 体",
+      "ela ide",
+      "▁adjust ments",
+      "▁adjustment s",
+      "质 的",
+      "开发 有限公司",
+      "上 游",
+      "合法 有效",
+      "tg et",
+      "t get",
+      "▁mean time",
+      "▁meant ime",
+      "р у",
+      "时 报",
+      "port al",
+      "por tal",
+      "p ortal",
+      "▁Dis cover",
+      "▁Disc over",
+      "▁conver ge",
+      "新疆维吾尔 自治区",
+      "终 究",
+      "US D",
+      "U SD",
+      "di gest",
+      "dig est",
+      "nM ilitary",
+      "nMil itary",
+      "担 忧",
+      "▁Deg ree",
+      "]] ,",
+      "] ],",
+      "请您 务必",
+      "▁Cabin et",
+      "▁Cab inet",
+      "警 惕",
+      "一带 一路",
+      "▁A irlines",
+      "▁Air lines",
+      "▁deb ugging",
+      "▁debug ging",
+      "Gu ard",
+      "▁F itz",
+      "▁Fit z",
+      "▁Fi tz",
+      "▁in expensive",
+      "典 礼",
+      "流行 的",
+      "流 行的",
+      "检 修",
+      "动力 学",
+      "动 力学",
+      "or ie",
+      "ori e",
+      "o rie",
+      "▁smooth ly",
+      "台 上",
+      "▁e ct",
+      "▁ec t",
+      "▁ ect",
+      "math op",
+      "mat hop",
+      "并 告诉我",
+      "不准 许",
+      "不 准许",
+      "在 道路上",
+      "IM AGE",
+      "I MAGE",
+      "省 份",
+      "▁T N",
+      "▁ TN",
+      "▁ar rives",
+      "▁arr ives",
+      "▁arriv es",
+      "▁arrive s",
+      "相 逢",
+      "rat ulations",
+      "▁I van",
+      "▁Iv an",
+      "▁ 卡",
+      "np ython",
+      "npy thon",
+      "n python",
+      "nS an",
+      "n San",
+      "▁歌曲 原唱",
+      "nWork s",
+      "n Works",
+      "im eters",
+      "ime ters",
+      "imet ers",
+      "imeter s",
+      "check out",
+      "é d",
+      "ub es",
+      "ube s",
+      "u bes",
+      "▁P layers",
+      "▁Pl ayers",
+      "▁Play ers",
+      "▁Player s",
+      "▁ Players",
+      "▁mechan istic",
+      "菲律 宾",
+      "营业 时间",
+      "选定 正确答案",
+      "ON T",
+      "O NT",
+      "os ex",
+      "ose x",
+      "o sex",
+      "AT T",
+      "A TT",
+      "▁\\ '_",
+      "▁\\' _",
+      "▁ \\'_",
+      "▁Muslim s",
+      "▁esc orts",
+      "▁escort s",
+      "ci te",
+      "cit e",
+      "c ite",
+      "▁Hil bert",
+      "▁P rest",
+      "▁Pr est",
+      "▁Pres t",
+      "▁Pre st",
+      "▁Eth iop",
+      "Argument Parser",
+      "分 化",
+      "AL T",
+      "A LT",
+      "SP E",
+      "S PE",
+      "Row s",
+      "Ro ws",
+      "R ows",
+      "影 子",
+      "清 热",
+      "都 非常",
+      "▁H amm",
+      "▁Ha mm",
+      "▁Ham m",
+      "路 的",
+      "View Controller",
+      "▁ 丁",
+      "僵 尸",
+      "选择题 时",
+      "▁C limate",
+      "▁Cl imate",
+      "P e",
+      "头 像",
+      "▁Cr usher",
+      "▁Crus her",
+      "▁pers ist",
+      "▁ 爱",
+      "ny our",
+      "nyou r",
+      "n your",
+      "Sy s",
+      "S ys",
+      "五 入",
+      "净 水",
+      "用 以",
+      "▁p uzz",
+      "▁pu zz",
+      "$( \\'",
+      "$ (\\'",
+      "comp uter",
+      "comput er",
+      "compute r",
+      "▁cle avage",
+      "▁radio therapy",
+      "▁radi otherapy",
+      "▁l ips",
+      "▁li ps",
+      "▁lip s",
+      "▁ lips",
+      "占总 收入的",
+      "占 总收入的",
+      "▁Al bum",
+      "▁Alb um",
+      "▁ Album",
+      "▁g ases",
+      "▁gas es",
+      "▁ga ses",
+      "▁c ope",
+      "▁co pe",
+      "▁cop e",
+      "oph one",
+      "ophon e",
+      "o phone",
+      "无 非",
+      "▁B ak",
+      "▁Ba k",
+      "▁O rd",
+      "▁Or d",
+      "▁ Ord",
+      "▁time line",
+      "▁tim eline",
+      "▁ timeline",
+      "W al",
+      "▁mitochond ria",
+      "票 房",
+      "an dro",
+      "and ro",
+      "andr o",
+      "发 音",
+      "副 校长",
+      "展示 了",
+      "▁t weets",
+      "▁tweet s",
+      "▁twe ets",
+      "▁smart phones",
+      "▁smartphone s",
+      "▁de priv",
+      "▁dep riv",
+      "▁ob struct",
+      "▁obst ruct",
+      "nP olitical",
+      "nPolit ical",
+      "xi v",
+      "x iv",
+      "▁c el",
+      "▁ce l",
+      "▁ cel",
+      "bo urg",
+      "bour g",
+      "▁N AME",
+      "▁NA ME",
+      "▁ NAME",
+      "sub string",
+      "substr ing",
+      "nn ew",
+      "n new",
+      "▁N OW",
+      "▁NO W",
+      "▁ NOW",
+      "mill ion",
+      "mil lion",
+      "m illion",
+      "▁l ifting",
+      "▁lif ting",
+      "▁lift ing",
+      "▁ lifting",
+      "忙 碌",
+      "nT IME",
+      "n TIME",
+      "并 做出",
+      "▁S haw",
+      "▁Sh aw",
+      "▁Sha w",
+      "出版社 出版",
+      "▁op ener",
+      "▁open er",
+      "some thing",
+      "s omething",
+      "踏 实",
+      "培训 机构",
+      "▁ ч",
+      "常 州",
+      "nR oman",
+      "n Roman",
+      "▁advis ory",
+      "▁advisor y",
+      "均 价",
+      "▁ca ution",
+      "▁caut ion",
+      "严 峻",
+      "▁F AQ",
+      "▁FA Q",
+      "un wrap",
+      "现 为",
+      "cs rf",
+      "劳动 报酬",
+      "居 家",
+      "信息 公开",
+      "con tr",
+      "cont r",
+      "co ntr",
+      "务 实",
+      "对方 的",
+      "对 方的",
+      "为 期",
+      "率 和",
+      "▁str ikes",
+      "▁stri kes",
+      "▁strike s",
+      "acchar ide",
+      "保 密",
+      "th umbnail",
+      "new command",
+      "nv ersion",
+      "n version",
+      "本院 在执行过程中",
+      "本院在 执行过程中",
+      "本院在执行 过程中",
+      "▁mer ging",
+      "▁merg ing",
+      "▁F ocus",
+      "▁Fo cus",
+      "▁ Focus",
+      "▁时间 复杂度",
+      "▁ 时间复杂度",
+      "▁emb arrass",
+      "▁con volution",
+      "▁conv olution",
+      "品 味",
+      "感 慨",
+      "面 团",
+      "担保 人",
+      "威 尔",
+      "落 的",
+      "▁Blu etooth",
+      "▁P ep",
+      "▁Pe p",
+      "▁t enant",
+      "▁te nant",
+      "▁ten ant",
+      "▁ tenant",
+      "ur ement",
+      "ure ment",
+      "un ar",
+      "una r",
+      "u nar",
+      "▁L apl",
+      "▁La pl",
+      "▁Lap l",
+      "圈 子",
+      "以下 备选答案",
+      "art ered",
+      "arter ed",
+      "arte red",
+      "▁acknowled ged",
+      "▁acknowledge d",
+      "▁d s",
+      "▁ ds",
+      "网络 营销",
+      "▁el even",
+      "▁elev en",
+      "▁ele ven",
+      "▁s aline",
+      "▁sa line",
+      "▁sal ine",
+      "不 通",
+      "想 知道",
+      "ary nge",
+      "aryng e",
+      "之 力",
+      "sel ling",
+      "sell ing",
+      "s elling",
+      "▁Direct ors",
+      "▁Director s",
+      "▁Dir ectors",
+      "nBe low",
+      "nBel ow",
+      "委托代理人 王",
+      "▁hon ors",
+      "▁honor s",
+      "dist ributed",
+      "楼 主",
+      "FE T",
+      "F ET",
+      "执行 通知",
+      "代理 书记员",
+      "▁wel coming",
+      "与 会",
+      "ci t",
+      "c it",
+      "我 当时",
+      "粗 糙",
+      "美 军",
+      "鉴 别",
+      "▁o ch",
+      "▁oc h",
+      "▁ och",
+      "▁C ache",
+      "▁Ca che",
+      "▁ Cache",
+      "磨 练",
+      "▁catal yst",
+      "精 美",
+      "颁 奖",
+      "▁Mathematic a",
+      "▁res ervation",
+      "▁reserv ation",
+      "▁L D",
+      "▁ LD",
+      "诈骗 罪",
+      "▁re ass",
+      "画 的",
+      "作出 了",
+      "作 出了",
+      "cu ador",
+      "▁N uclear",
+      "▁Nu clear",
+      "▁Nucle ar",
+      "▁bios ynthesis",
+      "怎么 做",
+      "▁corpor ation",
+      "▁M N",
+      "▁ MN",
+      "▁Ott awa",
+      "▁Ug anda",
+      "▁mell itus",
+      "▁S or",
+      "▁So r",
+      "▁s ung",
+      "▁su ng",
+      "▁sun g",
+      "操 控",
+      "十个 月",
+      "十 个月",
+      "▁C SV",
+      "▁CS V",
+      "▁ CSV",
+      "Gener ate",
+      "Gen erate",
+      "Gene rate",
+      "G enerate",
+      "▁di electric",
+      "▁die lectric",
+      "文件 的",
+      "文 件的",
+      "▁fix ation",
+      "▁surv iving",
+      "▁surviv ing",
+      "中 法",
+      "cu da",
+      "c uda",
+      "How ever",
+      "▁investig ates",
+      "▁investigate s",
+      "一直 以来",
+      "口 袋",
+      "▁\\\\ *",
+      "▁ \\\\*",
+      "isb ane",
+      "j k",
+      "用 力",
+      "nQu ote",
+      "n Quote",
+      "▁ch ess",
+      "▁che ss",
+      "▁mamm als",
+      "▁Pro vider",
+      "▁Prov ider",
+      "▁Provide r",
+      "▁ Provider",
+      "▁subsid iary",
+      "K C",
+      "▁D eal",
+      "▁De al",
+      "nE rror",
+      "nEr ror",
+      "n Error",
+      "▁N AD",
+      "▁NA D",
+      "▁Elect rical",
+      "▁Electric al",
+      "▁Po isson",
+      "才 可以",
+      "▁re write",
+      "▁ 村",
+      "使用 权",
+      "罪犯 李",
+      "RE EN",
+      "REE N",
+      "先进 单位",
+      "撤回上诉 处理",
+      "▁Hud son",
+      "▁answer ing",
+      "还 未",
+      "看 一下",
+      "市场 经济",
+      "nT om",
+      "nTo m",
+      "n Tom",
+      "这一 次",
+      "这 一次",
+      "神 器",
+      "mun d",
+      "mu nd",
+      "m und",
+      "nm ake",
+      "n make",
+      "▁f acebook",
+      "▁face book",
+      "▁ facebook",
+      "▁fil tration",
+      "▁lit igation",
+      ": {",
+      "IR ST",
+      "IRS T",
+      "bi ll",
+      "bil l",
+      "b ill",
+      "▁ch ap",
+      "▁cha p",
+      "de signed",
+      "des igned",
+      "design ed",
+      "h ospital",
+      "会 使",
+      "BS D",
+      "B SD",
+      "▁r upt",
+      "▁ru pt",
+      "▁ rupt",
+      "质 地",
+      "日被 逮捕",
+      "nH ome",
+      "nHom e",
+      "n Home",
+      "的 重",
+      ";\" >\\",
+      ";\"> \\",
+      "; \">\\",
+      "ri ce",
+      "ric e",
+      "r ice",
+      "车 上",
+      "▁G y",
+      "▁ Gy",
+      "▁correl ates",
+      "▁correlate s",
+      "]] ,\\",
+      "]], \\",
+      "] ],\\",
+      "am onds",
+      "amond s",
+      "amon ds",
+      "a monds",
+      "▁book ed",
+      "▁bo oked",
+      "▁he ap",
+      "▁ heap",
+      "▁van illa",
+      "▁util izes",
+      "▁utilize s",
+      "创 伤",
+      "回答 以下问题",
+      "▁mar ble",
+      "▁Wire less",
+      "晚 期",
+      "服务 平台",
+      "na ires",
+      "naire s",
+      "nai res",
+      "n aires",
+      "dist ribution",
+      "列 为",
+      "Fl ash",
+      "F lash",
+      "st airs",
+      "sta irs",
+      "的 正",
+      "粉 碎",
+      "ol t",
+      "o lt",
+      "Creat or",
+      "Cre ator",
+      "挺好 的",
+      "挺 好的",
+      "▁g host",
+      "▁gh ost",
+      "▁ ghost",
+      "▁Gener ator",
+      "▁Gen erator",
+      "▁ Generator",
+      "中学 生",
+      "中 学生",
+      "Up per",
+      "▁coll ector",
+      "▁collect or",
+      "▁colle ctor",
+      "▁ collector",
+      "▁cryptoc urrency",
+      "▁crypto currency",
+      "鲁 迅",
+      "Str ucture",
+      "Struct ure",
+      "▁R ah",
+      "▁Ra h",
+      "系 其",
+      "▁tim ber",
+      "▁ti mber",
+      "control s",
+      "contr ols",
+      "参演 电影",
+      "▁exp ans",
+      "Act or",
+      "Ac tor",
+      "A ctor",
+      "▁ не",
+      "er ek",
+      "ere k",
+      "e rek",
+      "文化 艺术",
+      "pi res",
+      "pir es",
+      "p ires",
+      "▁m igrate",
+      "▁migr ate",
+      "▁ migrate",
+      "▁In stitution",
+      "▁Inst itution",
+      "▁Instit ution",
+      "临沂 市",
+      "减少 了",
+      "减 少了",
+      "知名 度",
+      "▁N ear",
+      "▁Ne ar",
+      "▁y aml",
+      "▁ya ml",
+      "▁ yaml",
+      "▁pro to",
+      "▁pr oto",
+      "▁prot o",
+      "▁ proto",
+      "为 之",
+      "▁Ah med",
+      "▁com mercially",
+      "▁commercial ly",
+      "见 了",
+      "有 许多",
+      "▁C AR",
+      "▁CA R",
+      "▁ CAR",
+      "up pet",
+      "upp et",
+      "▁Mun ich",
+      "其中 一个",
+      "▁sub units",
+      "▁subunit s",
+      "Te r",
+      "T er",
+      "最 重要",
+      "▁ad dr",
+      "▁add r",
+      "▁ addr",
+      "id ency",
+      "ide ncy",
+      "iden cy",
+      "pr event",
+      "pre vent",
+      "prev ent",
+      "Ag g",
+      "A gg",
+      "▁dos age",
+      "出 演",
+      "nR ad",
+      "n Rad",
+      "▁n ep",
+      "▁ne p",
+      "就 来",
+      "林 木",
+      "▁bl ues",
+      "▁blue s",
+      "▁blu es",
+      "▁ex acerb",
+      "▁s pite",
+      "▁sp ite",
+      "▁refresh ing",
+      "▁refres hing",
+      "▁E ve",
+      "▁Ev e",
+      "PT ION",
+      "一元 二次方程",
+      "一 元二次方程",
+      "▁O ri",
+      "▁Or i",
+      "▁t in",
+      "▁ti n",
+      "▁ tin",
+      "▁ad vers",
+      "▁adv ers",
+      "互 助",
+      "火 爆",
+      "▁S PD",
+      "▁SP D",
+      "nPro blem",
+      "nProb lem",
+      "n Problem",
+      "▁Cam eron",
+      "▁Camero n",
+      "心 得",
+      "内存 容量",
+      "▁J ar",
+      "▁Ja r",
+      "▁ Jar",
+      "荣誉 称号",
+      "tb ool",
+      "t bool",
+      "量 化",
+      "▁S cre",
+      "▁Sc re",
+      "▁Scr e",
+      "umb ent",
+      "Project s",
+      "显着 性",
+      "具体 应用",
+      "▁se pt",
+      "▁sep t",
+      "com par",
+      "comp ar",
+      "两 名",
+      "▁v im",
+      "▁vi m",
+      "▁ vim",
+      "▁sh utdown",
+      "▁shut down",
+      "▁ shutdown",
+      "ñ a",
+      "end ent",
+      "ende nt",
+      "enden t",
+      "O s",
+      "的说 法",
+      "的 说法",
+      "▁J ak",
+      "▁Ja k",
+      "▁tr ades",
+      "▁trad es",
+      "▁tra des",
+      "▁trade s",
+      "nA mong",
+      "nAm ong",
+      "德 育",
+      "火 焰",
+      "▁aut ism",
+      "和 新",
+      "虚拟 机",
+      "▁Inter est",
+      "IL E",
+      "I LE",
+      "с я",
+      "▁Cam pus",
+      "▁Camp us",
+      "{ #",
+      "楼 梯",
+      "▁vac cin",
+      "▁vacc in",
+      "的一 生",
+      "的 一生",
+      "in as",
+      "ina s",
+      "i nas",
+      "执行 董事",
+      "▁back wards",
+      "▁backward s",
+      "` :",
+      "an yon",
+      "any on",
+      "鸡 肉",
+      "ig i",
+      "i gi",
+      "我也 不",
+      "我 也不",
+      "▁viol ations",
+      "▁violation s",
+      "区 政府",
+      "ik er",
+      "ike r",
+      "i ker",
+      "▁w orthy",
+      "▁wor thy",
+      "▁worth y",
+      "▁ worthy",
+      "st ub",
+      "s tub",
+      "小 康",
+      "un n",
+      "u nn",
+      "is cher",
+      "isc her",
+      "isch er",
+      "ische r",
+      "▁so ils",
+      "▁soil s",
+      "os ocial",
+      "o social",
+      "阴 道",
+      "▁some time",
+      "▁som etime",
+      "▁C incinnati",
+      "体 温",
+      "不同 于",
+      "▁c ents",
+      "▁cent s",
+      "▁ cents",
+      "Ob serv",
+      "Obs erv",
+      "O bserv",
+      "仍 在",
+      "▁meas urable",
+      "大 夫",
+      "▁x s",
+      "▁ xs",
+      "▁late nt",
+      "▁lat ent",
+      "因为 在",
+      "微信 公众号",
+      "犯罪 嫌疑人",
+      "▁def enders",
+      "▁defend ers",
+      "▁defender s",
+      "得 起",
+      "▁M IN",
+      "▁MI N",
+      "▁ MIN",
+      "R o",
+      "和 中",
+      "▁D emo",
+      "▁De mo",
+      "▁Dem o",
+      "▁ Demo",
+      "inst alled",
+      "install ed",
+      "半 个",
+      "He y",
+      "H ey",
+      "{( )",
+      "{ ()",
+      "arynge al",
+      "aryng eal",
+      "周 某",
+      "农民 工",
+      "▁S SL",
+      "▁SS L",
+      "▁ SSL",
+      "就不 能",
+      "就 不能",
+      "市场 营销",
+      "▁N ull",
+      "▁Nu ll",
+      "▁ Null",
+      "的家 庭",
+      "的 家庭",
+      "明 代",
+      "▁Det ection",
+      "▁ Detection",
+      "▁Optim ization",
+      "筹 备",
+      "好 不好",
+      "自来 水",
+      "Ch ina",
+      "in el",
+      "ine l",
+      "i nel",
+      "改革 的",
+      "本科 生",
+      "ar di",
+      "ard i",
+      "▁T err",
+      "▁Te rr",
+      "▁Ter r",
+      "但 被告",
+      "▁t anks",
+      "▁tank s",
+      "▁tan ks",
+      "▁Re pository",
+      "▁Rep ository",
+      "▁ Repository",
+      "从 不",
+      "唐 代",
+      "近 平",
+      "拥有 一",
+      "拥 有一",
+      "▁w ounds",
+      "▁wound s",
+      "z u",
+      "is che",
+      "isc he",
+      "isch e",
+      "i sche",
+      "十五 章",
+      "▁H onda",
+      "▁Hon da",
+      "▁Ho nda",
+      "▁Hond a",
+      "协议 书",
+      "所需 的",
+      "bar s",
+      "ba rs",
+      "b ars",
+      "加 盖",
+      "委托代理人 张",
+      "mu l",
+      "m ul",
+      "▁sk ew",
+      "▁ske w",
+      "port ion",
+      "por tion",
+      "p ortion",
+      "屁 股",
+      "▁di arr",
+      "▁dia rr",
+      "Met rics",
+      "Metric s",
+      "em ption",
+      "empt ion",
+      "emp tion",
+      "fail ure",
+      "商 用",
+      "ap a",
+      "a pa",
+      "管理 办法",
+      "因为 你",
+      "因 为你",
+      "pa id",
+      "p aid",
+      "对 被申请人",
+      "现有的 床铺",
+      "▁in tim",
+      "▁int im",
+      "小 鼠",
+      "强 劲",
+      "云 计算",
+      "FORM AT",
+      "FOR MAT",
+      "high light",
+      "ment ioned",
+      "▁b achelor",
+      "接 地",
+      "Rad io",
+      "▁f ifty",
+      "▁fif ty",
+      "▁fi fty",
+      "回 合",
+      "it é",
+      "~~ \\",
+      "~ ~\\",
+      "上 科大",
+      "主义 的",
+      "▁dr ift",
+      "▁dri ft",
+      "九 条",
+      "可能 的",
+      "了一 场",
+      "了 一场",
+      "▁B MW",
+      "▁BM W",
+      "T oggle",
+      "商 铺",
+      "为 维护",
+      "▁Ul tra",
+      "▁Ultr a",
+      "粒 子",
+      "nSp anish",
+      "nSpan ish",
+      "n Spanish",
+      "▁Co hen",
+      "▁Coh en",
+      "四 舍",
+      "发展 战略",
+      "▁bur ned",
+      "▁burn ed",
+      "遥 控",
+      "dt ype",
+      "d type",
+      "▁un ions",
+      "▁union s",
+      "▁uni ons",
+      "hi s",
+      "h is",
+      "铝 合金",
+      "▁V ic",
+      "▁Vi c",
+      "on aut",
+      "ona ut",
+      "}) );\\",
+      "})) ;\\",
+      "} ));\\",
+      "Fix ture",
+      "限 度",
+      "明白 了",
+      "明 白了",
+      "▁in spir",
+      "▁ins pir",
+      "▁insp ir",
+      "▁ampl ification",
+      "▁人 口",
+      "▁ 人口",
+      "ns ys",
+      "n sys",
+      "▁P ush",
+      "▁Pu sh",
+      "▁ Push",
+      "▁po ets",
+      "▁poet s",
+      "nTh ough",
+      "▁N either",
+      "▁Ne ither",
+      "ict ory",
+      "ictor y",
+      "▁touch es",
+      "▁tou ches",
+      "print StackTrace",
+      "欺 骗",
+      "fa t",
+      "f at",
+      "申请执行人 王",
+      "v h",
+      "项 规定",
+      "退房 时间",
+      "▁bad ly",
+      "起诉 人",
+      "壮 大",
+      "|} \\",
+      "| }\\",
+      "▁multi plied",
+      "包装 清单",
+      "sc ores",
+      "score s",
+      "s cores",
+      "▁mon etary",
+      "di agn",
+      "diag n",
+      "mult iple",
+      "multi ple",
+      "▁cos metic",
+      "▁cosm etic",
+      "支 球队",
+      "▁sculpt ure",
+      "多 余",
+      "是 小",
+      "被除 数",
+      "被 除数",
+      "▁Tr avis",
+      "▁Tra vis",
+      "▁Trav is",
+      "▁sil ence",
+      "举行 了",
+      "举 行了",
+      "到 最后",
+      "▁Columb us",
+      "还是 在",
+      "还 是在",
+      "CH AR",
+      "CHA R",
+      "▁br ut",
+      "▁bru t",
+      "世 人",
+      "▁sn ippet",
+      "▁ snippet",
+      "如果 在",
+      "ip ro",
+      "i pro",
+      "▁pre serving",
+      "▁pres erving",
+      "平 生",
+      "▁s s",
+      "▁ ss",
+      "system s",
+      "▁work space",
+      "▁works pace",
+      "▁ workspace",
+      "ere um",
+      "▁down loads",
+      "▁download s",
+      "▁ downloads",
+      "nat ional",
+      "nation al",
+      "n ational",
+      "▁contin uation",
+      "▁continu ation",
+      "就 这么",
+      "▁n os",
+      "▁no s",
+      "▁ nos",
+      "▁start ups",
+      "▁startup s",
+      "教 导",
+      "▁ 获奖",
+      "F ocus",
+      "眼 的",
+      "年 生",
+      "▁Sup ply",
+      "▁Prob ability",
+      "ca ts",
+      "cat s",
+      "c ats",
+      "▁devast ating",
+      "相反 数",
+      "▁do ub",
+      "▁dou b",
+      "本身 的",
+      "本 身的",
+      "说 起",
+      "▁a nis",
+      "▁an is",
+      "下 单",
+      "ig hb",
+      "igh b",
+      "▁Pat tern",
+      "▁ Pattern",
+      "▁propri etary",
+      "绑 定",
+      "▁di visible",
+      "▁div isible",
+      "▁p oker",
+      "▁po ker",
+      "▁cart oon",
+      "▁sp elling",
+      "▁spell ing",
+      "原 谅",
+      "VE LO",
+      "▁Command er",
+      "▁Comm ander",
+      "检 刑诉",
+      "▁Ed wards",
+      "▁Edward s",
+      "很 不",
+      "得 上",
+      "新 品",
+      "ycl er",
+      "ycle r",
+      "yc ler",
+      "y cler",
+      "cos ity",
+      "c osity",
+      "IN F",
+      "I NF",
+      "博览 会",
+      "sl ide",
+      "s lide",
+      "▁em ploys",
+      "▁employ s",
+      "日 制",
+      "▁N ancy",
+      "▁Na ncy",
+      "▁Nan cy",
+      "▁bin omial",
+      "▁t oxin",
+      "▁to xin",
+      "▁tox in",
+      "th ood",
+      "t hood",
+      "清 风",
+      "商品 的",
+      "商 品的",
+      "赢得 了",
+      "赢 得了",
+      "该怎么 办",
+      "该 怎么办",
+      "▁u gly",
+      "▁ug ly",
+      "就 想",
+      "ic us",
+      "i cus",
+      "其中 包括",
+      "▁m all",
+      "▁mal l",
+      "▁ma ll",
+      "▁ mall",
+      "吃 完",
+      "奈 何",
+      "进 修",
+      "med ian",
+      "media n",
+      "换 算",
+      "生 化",
+      "需求 的",
+      "阳县 人民法院",
+      "阳 县人民法院",
+      "和 浩特",
+      "影响 了",
+      "▁ge ographical",
+      "▁geographic al",
+      "▁inadequ ate",
+      "规 矩",
+      "除 尘",
+      "宫 颈",
+      "er sed",
+      "ers ed",
+      "▁d ear",
+      "▁de ar",
+      "▁th umb",
+      "▁ thumb",
+      "博 主",
+      "就 诊",
+      "na ble",
+      "nab le",
+      "n able",
+      "▁M ills",
+      "▁Mil ls",
+      "▁Mill s",
+      "▁Represent ative",
+      "告 别",
+      "▁B rid",
+      "▁Br id",
+      "▁Bri d",
+      "▁H arrison",
+      "▁Harr ison",
+      "▁Harris on",
+      "提 名",
+      "灿 烂",
+      "马 斯",
+      "▁T ang",
+      "▁Ta ng",
+      "▁Tan g",
+      "▁ser oton",
+      "▁T rial",
+      "▁Tr ial",
+      "▁Tri al",
+      "星 球",
+      "不 影响",
+      "is hi",
+      "ish i",
+      "i shi",
+      "get Message",
+      "海 报",
+      "pol ar",
+      "po lar",
+      "p olar",
+      "om ology",
+      "omo logy",
+      "o mology",
+      "家 常",
+      "wi k",
+      "w ik",
+      "▁cou sin",
+      "▁cous in",
+      "▁neuro deg",
+      "帐 号",
+      "▁c urved",
+      "▁cur ved",
+      "▁curve d",
+      "▁curv ed",
+      "矿 业",
+      "or ean",
+      "ore an",
+      "Ta ke",
+      "T ake",
+      "bu nd",
+      "b und",
+      "▁found ations",
+      "▁foundation s",
+      "▁br ass",
+      "▁bra ss",
+      "▁pot ato",
+      "太 高",
+      "nD irect",
+      "nDi rect",
+      "n Direct",
+      "肖 战",
+      "▁neighborhood s",
+      "外 壳",
+      "▁H ob",
+      "▁Ho b",
+      "党 中央",
+      "▁h eal",
+      "▁he al",
+      "math it",
+      "mat hit",
+      "▁re plies",
+      "▁rep lies",
+      "▁t am",
+      "▁ta m",
+      "第一百 条",
+      "cl ients",
+      "client s",
+      "cli ents",
+      "八 十",
+      "ri des",
+      "ride s",
+      "rid es",
+      "r ides",
+      "▁influ encing",
+      "把 这个",
+      "▁Gener ic",
+      "▁Gen eric",
+      "▁Gene ric",
+      "▁ Generic",
+      "▁S ingles",
+      "▁Sing les",
+      "▁Single s",
+      "bo at",
+      "奔 驰",
+      "Nor th",
+      "N orth",
+      "▁w ool",
+      "▁wo ol",
+      "▁c amping",
+      "▁camp ing",
+      "▁cam ping",
+      "句 子",
+      "▁e vac",
+      "▁ev ac",
+      "颁 发",
+      "保 鲜",
+      "量 为",
+      "一 顿",
+      "提 倡",
+      "c rypt",
+      "ib les",
+      "ible s",
+      "ibl es",
+      "i bles",
+      "iro pr",
+      "▁Wild life",
+      "ab we",
+      "▁d rought",
+      "▁dr ought",
+      "▁dro ught",
+      "最 快",
+      "这样 做",
+      "航 班",
+      "Ti p",
+      "T ip",
+      "un gen",
+      "ung en",
+      "unge n",
+      "num eric",
+      "nu meric",
+      "石 材",
+      "▁H ER",
+      "▁HE R",
+      "▁ HER",
+      "nW rite",
+      "nWrit e",
+      "n Write",
+      "▁broad ly",
+      "▁Oper ating",
+      "▁Opera ting",
+      "▁ Operating",
+      "图 的",
+      "富 裕",
+      "注意 到",
+      "▁k et",
+      "▁ke t",
+      "▁ ket",
+      "▁C lare",
+      "▁Cl are",
+      "▁Clar e",
+      "冻结 被申请人",
+      "▁sc attered",
+      "▁scatter ed",
+      "计算 出",
+      "计 算出",
+      "吴 某",
+      "术 前",
+      "HI V",
+      "H IV",
+      "opl us",
+      "o plus",
+      "▁tom ato",
+      "▁tomat o",
+      "de ployment",
+      "deploy ment",
+      "地 块",
+      "on c",
+      "o nc",
+      "au sal",
+      "aus al",
+      "a usal",
+      "▁soci eties",
+      "申请 的",
+      "▁d an",
+      "▁da n",
+      "▁ dan",
+      "▁stat istic",
+      "▁statist ic",
+      "方法 对",
+      "▁A uckland",
+      "▁R ES",
+      "▁RE S",
+      "▁ RES",
+      "em e",
+      "e me",
+      "装 箱",
+      "针 灸",
+      "ji n",
+      "j in",
+      "▁B urg",
+      "▁Bur g",
+      "▁Bu rg",
+      "今 晚",
+      "你 对",
+      "实 录",
+      "烤 箱",
+      "提供被执行人 其他",
+      "▁be ams",
+      "▁beam s",
+      "▁tr apped",
+      "▁tra pped",
+      "▁trap ped",
+      "▁I NC",
+      "▁IN C",
+      "▁ INC",
+      "fire base",
+      "决 议",
+      "▁B ac",
+      "▁Ba c",
+      "▁Ser bia",
+      "▁pro phyl",
+      "人 际",
+      "女 王",
+      "板 的",
+      "xp ath",
+      "x path",
+      "▁to pped",
+      "▁top ped",
+      "▁sub missions",
+      "▁submission s",
+      "和 生活",
+      "▁att rs",
+      "▁attr s",
+      "▁ attrs",
+      "蘑 菇",
+      "on n",
+      "o nn",
+      "▁St evens",
+      "▁Ste vens",
+      "▁Steve ns",
+      "▁Steven s",
+      "都 将",
+      "归还 借款",
+      "Ge ometry",
+      "Geo metry",
+      "枢 纽",
+      "下列 范围",
+      "工程 施工",
+      "▁Z hang",
+      "▁Zh ang",
+      "的 都是",
+      "中医 药学",
+      "中医药 学",
+      "▁re leg",
+      "▁rel eg",
+      "nKey words",
+      "启 示",
+      "▁cou pons",
+      "▁coupon s",
+      "▁coup ons",
+      "▁up grades",
+      "▁upgrade s",
+      "▁upgrad es",
+      "▁Freder ick",
+      "从 判决确定之日起计算",
+      "长 相",
+      "▁ 语种",
+      "▁t ut",
+      "▁tu t",
+      "sw ift",
+      "▁short est",
+      "ole rance",
+      "oler ance",
+      "▁st resses",
+      "▁str esses",
+      "▁stress es",
+      "▁m x",
+      "▁ mx",
+      "H Y",
+      "▁Saf ari",
+      "想 不到",
+      "主 营",
+      "ut ral",
+      "时 要",
+      "▁counter parts",
+      "▁counterpart s",
+      "HP P",
+      "H PP",
+      "bi o",
+      "b io",
+      "▁F lag",
+      "▁Fl ag",
+      "▁ Flag",
+      "禁 忌",
+      "例 患者",
+      "Te le",
+      "Tel e",
+      "T ele",
+      "Ap ril",
+      "Apr il",
+      "A pril",
+      "双 重",
+      "▁Pro tein",
+      "▁Prot ein",
+      "▁Prote in",
+      "TC P",
+      "T CP",
+      "意见 书",
+      "▁cl ue",
+      "▁B uilt",
+      "▁Bu ilt",
+      "▁ Built",
+      "杰 出",
+      "与中国 科学院",
+      "与 中国科学院",
+      "能够 在",
+      "▁g ateway",
+      "▁gate way",
+      "▁ gateway",
+      "右 键",
+      "Sy m",
+      "S ym",
+      "nW ant",
+      "n Want",
+      "▁sold ier",
+      "▁refuge es",
+      "▁refugee s",
+      "な い",
+      "bo ns",
+      "bon s",
+      "b ons",
+      "▁g ard",
+      "▁gar d",
+      "▁ga rd",
+      "▁ gard",
+      "硕士 研究生",
+      "▁tal ents",
+      "▁talent s",
+      "Co urse",
+      "C ourse",
+      "oper atively",
+      "operative ly",
+      "olic ies",
+      "▁L CD",
+      "▁LC D",
+      "▁ LCD",
+      "从来 没有",
+      "从来没 有",
+      "▁S ever",
+      "▁Se ver",
+      "▁Sev er",
+      "▁ Sever",
+      "不 已",
+      "Go ld",
+      "G old",
+      "▁ev olve",
+      "一 无",
+      "升 降",
+      "变 迁",
+      "着 你",
+      "▁rhe umat",
+      "ov y",
+      "o vy",
+      "▁V a",
+      "等 部门",
+      "研究 成果",
+      "NS String",
+      "中 断",
+      "ia tor",
+      "iat or",
+      "i ator",
+      "pl ots",
+      "plot s",
+      "E s",
+      "不 难",
+      "j b",
+      "RO P",
+      "R OP",
+      "Sa fe",
+      "S afe",
+      "▁conscious ness",
+      "▁manifest ations",
+      "▁manifestation s",
+      "比 利",
+      "▁G ill",
+      "▁Gi ll",
+      "▁Gil l",
+      "▁Cent ers",
+      "▁Center s",
+      "▁Ce nters",
+      "堵 塞",
+      "▁ 蔡",
+      "珍 贵",
+      "▁bo asts",
+      "银行股份有限公司 信用卡中心",
+      "▁championship s",
+      "▁champion ships",
+      "▁champions hips",
+      "卸 载",
+      "▁Thanks giving",
+      "吸引 了",
+      "广泛 的",
+      "辅导 员",
+      "内容 包括",
+      "发 改",
+      "nD ue",
+      "n Due",
+      "Fac ades",
+      "par agraph",
+      "para graph",
+      "ma rine",
+      "mar ine",
+      "m arine",
+      "▁sym pathetic",
+      "▁trans ported",
+      "▁transport ed",
+      "一 键",
+      "▁Im plementation",
+      "▁Implement ation",
+      "▁ Implementation",
+      "无 情",
+      "行 车",
+      "WE R",
+      "W ER",
+      "▁F F",
+      "▁ FF",
+      "▁Ul tr",
+      "os hop",
+      "osh op",
+      "o shop",
+      "to l",
+      "t ol",
+      "辨认 笔录",
+      "高等 学校",
+      "lu ent",
+      "l uent",
+      "un til",
+      "unt il",
+      "u ntil",
+      "MA TH",
+      "MAT H",
+      "M ATH",
+      "▁Con serv",
+      "▁Cons erv",
+      "▁bath rooms",
+      "▁bathroom s",
+      "▁D DR",
+      "▁DD R",
+      "▁ DDR",
+      "呕 吐",
+      "开始 游戏",
+      "▁incent ives",
+      "▁incentive s",
+      "▁sub scribers",
+      "▁subscrib ers",
+      "▁subscribe rs",
+      "▁subscriber s",
+      "录 像",
+      "预 售",
+      "的 理解",
+      "▁cr edited",
+      "▁credit ed",
+      "▁cred ited",
+      "▁ credited",
+      "篇 文章",
+      "声 器",
+      "▁t yped",
+      "▁type d",
+      "▁typ ed",
+      "▁ty ped",
+      "▁ typed",
+      "▁p raise",
+      "▁pr aise",
+      "▁pra ise",
+      "吉 他",
+      "城 堡",
+      "Un ity",
+      "Unit y",
+      "不过 是",
+      "nS everal",
+      "▁Reg istry",
+      "▁ Registry",
+      "平 分",
+      "en cv",
+      "enc v",
+      "不准许 撤诉",
+      "it ated",
+      "ita ted",
+      "itate d",
+      "itat ed",
+      "▁re duct",
+      "▁redu ct",
+      "▁red uct",
+      "▁h omology",
+      "▁hom ology",
+      "▁ho mology",
+      "▁homolog y",
+      "实 惠",
+      "▁T C",
+      "▁ TC",
+      "text ure",
+      "tex ture",
+      "买 家",
+      "pr ison",
+      "pri son",
+      "F n",
+      "EE S",
+      "E ES",
+      "心理 健康",
+      "ass isted",
+      "assis ted",
+      "▁ µ",
+      "二审 案件受理费",
+      "花 椒",
+      "inter cept",
+      "▁d yst",
+      "▁dys t",
+      "▁dy st",
+      "▁M uh",
+      "▁Mu h",
+      "▁co unted",
+      "▁count ed",
+      "▁T ea",
+      "▁Te a",
+      "▁Us ually",
+      "▁mon keys",
+      "▁monkey s",
+      "得不 到",
+      "得 不到",
+      "淄博 市",
+      "▁Br isbane",
+      "我 了",
+      "顶 端",
+      "领导 班子",
+      "▁Wood s",
+      "▁Wo ods",
+      "iv ative",
+      "市 面上",
+      "mi sc",
+      "mis c",
+      "m isc",
+      "CON TRIBUTING",
+      "解 方程",
+      "▁E ug",
+      "维 权",
+      "sc ient",
+      "sci ent",
+      "▁pre dec",
+      "▁pred ec",
+      "台 风",
+      "▁a mp",
+      "▁am p",
+      "▁ amp",
+      "mer ged",
+      "merge d",
+      "M X",
+      "公 社",
+      "口 的",
+      "▁sp iral",
+      "▁spir al",
+      "▁in appropriate",
+      "▁T ot",
+      "▁To t",
+      "read me",
+      "▁st air",
+      "▁Associ ates",
+      "▁Associate s",
+      "ia sm",
+      "ias m",
+      "i asm",
+      "成为 一个",
+      "成 为一个",
+      "▁Max imum",
+      "▁Maxim um",
+      "▁ Maximum",
+      "▁amy loid",
+      "▁B S",
+      "▁ BS",
+      "ang elog",
+      "ange log",
+      "angel og",
+      "▁Fr ances",
+      "▁France s",
+      "▁Franc es",
+      "▁un ified",
+      "▁Veter ans",
+      "▁ 晋江文学网",
+      "图 中",
+      "处理 的",
+      "▁{ },",
+      "▁{} ,",
+      "▁ {},",
+      "骨 头",
+      "特色 服务",
+      "▁Import Error",
+      "敬 业",
+      "▁D irection",
+      "▁Direct ion",
+      "▁Di rection",
+      "▁Dir ection",
+      "▁ Direction",
+      "▁special izes",
+      "▁specialize s",
+      "th umb",
+      "▁v olt",
+      "▁vol t",
+      "▁vo lt",
+      "法定代表人 张",
+      "切 片",
+      "逃 离",
+      "▁\" (",
+      "▁ \"(",
+      "▁p estic",
+      "▁pest ic",
+      "▁pes tic",
+      "字 数",
+      "前 三",
+      "▁styl ing",
+      "▁Health care",
+      "bi s",
+      "b is",
+      "st din",
+      "std in",
+      "主 场",
+      "▁c ues",
+      "▁cu es",
+      "▁cue s",
+      "C ursor",
+      "▁back yard",
+      "▁Par ameters",
+      "▁Parameter s",
+      "▁Param eters",
+      "▁ Parameters",
+      "ut m",
+      "u tm",
+      "hor izontal",
+      "书 证",
+      "nE X",
+      "n EX",
+      "▁G uild",
+      "▁Gu ild",
+      "for matter",
+      "form atter",
+      "format ter",
+      "▁' __",
+      "▁'_ _",
+      "pace s",
+      "pa ces",
+      "pac es",
+      "p aces",
+      "▁des erves",
+      "▁deserve s",
+      "▁disturb ance",
+      "剧 中",
+      ")) *",
+      ") )*",
+      "有两 种",
+      "有 两种",
+      "日 经",
+      "▁C BD",
+      "▁CB D",
+      "这些 问题",
+      "▁ass ure",
+      "▁M idd",
+      "▁Mi dd",
+      "▁Mid d",
+      "▁out patient",
+      "就 看",
+      "指 纹",
+      "an ya",
+      "any a",
+      "a nya",
+      "r é",
+      "▁s ug",
+      "▁su g",
+      "▁m ate",
+      "▁mat e",
+      "▁ma te",
+      "▁ mate",
+      "▁S askat",
+      "MMMM MMMM",
+      "要 做",
+      "ns e",
+      "n se",
+      "按月 利率",
+      "按 月利率",
+      "ac ional",
+      "acion al",
+      "aci onal",
+      "acio nal",
+      "▁ac knowledge",
+      "▁acknow ledge",
+      "▁acknowled ge",
+      "▁pract ically",
+      "▁practical ly",
+      "畅 销",
+      "Sr c",
+      "S rc",
+      "说 着",
+      "js x",
+      "j sx",
+      "IN CLUDE",
+      "执 着",
+      "排行 榜",
+      "co ol",
+      "c ool",
+      "iz ens",
+      "ize ns",
+      "izen s",
+      "i zens",
+      "▁ad ul",
+      "▁gall eries",
+      "▁ham mer",
+      "▁ hammer",
+      "▁def ective",
+      "▁defect ive",
+      "r p",
+      "火 山",
+      "ba z",
+      "b az",
+      "亲 爱的",
+      "▁p ear",
+      "▁pe ar",
+      "▁ pear",
+      "▁bundle s",
+      "▁bund les",
+      "▁sem inar",
+      "▁semi nar",
+      "▁semin ar",
+      "mark er",
+      "mar ker",
+      "m arker",
+      "▁rank ings",
+      "▁ranking s",
+      "Gu id",
+      "Gui d",
+      "G uid",
+      "价值 的财产",
+      "价值的 财产",
+      "是 将",
+      "些 什么",
+      "' $",
+      "诊 治",
+      "nC ustom",
+      "n Custom",
+      "▁form atted",
+      "▁format ted",
+      "▁ formatted",
+      "对 上述",
+      "at ics",
+      "atic s",
+      "ati cs",
+      "理想 的",
+      "理 想的",
+      "毁 灭",
+      "Car e",
+      "Ca re",
+      "C are",
+      "FF FF",
+      "ve ctors",
+      "vector s",
+      "v ectors",
+      "▁Work ers",
+      "▁Wor kers",
+      "▁Worker s",
+      "倒 数",
+      "ag i",
+      "a gi",
+      "找 房",
+      "会议 室",
+      "网站 的",
+      "网 站的",
+      "奖惩 审批表",
+      "在 审理过程中",
+      "网络 科技有限公司",
+      "现 将",
+      "理 智",
+      "手术 治疗",
+      "年 平均",
+      "ea sy",
+      "eas y",
+      "e asy",
+      "▁V ac",
+      "▁Va c",
+      "ipp ets",
+      "ippet s",
+      "于本判决 生效之日起十日内",
+      "于本判决生效之日起 十日内",
+      "▁ 潘",
+      "所 选",
+      "二 期",
+      "新 产品",
+      ";& #",
+      "; &#",
+      "▁L aunch",
+      "▁La unch",
+      "▁ Launch",
+      "不 胜",
+      "总 线",
+      "增值 税",
+      "▁Te acher",
+      "▁Tea cher",
+      "保 湿",
+      "Jun e",
+      "J une",
+      "▁st rom",
+      "▁str om",
+      "▁stro m",
+      "▁ strom",
+      "as tically",
+      "ast ically",
+      "astic ally",
+      "南宁 市",
+      "南 宁市",
+      "日常 生活中",
+      "日常生活 中",
+      "▁re mot",
+      "▁rem ot",
+      "硫 酸",
+      "第一 条",
+      "第 一条",
+      "▁A ur",
+      "▁Au r",
+      "应 有",
+      "ou x",
+      "o ux",
+      "国 科大",
+      "扬 子",
+      "no d",
+      "n od",
+      "cal e",
+      "ca le",
+      "c ale",
+      "▁f lies",
+      "▁fl ies",
+      "▁ flies",
+      "这三 个",
+      "这 三个",
+      "▁careg ivers",
+      "▁caregiver s",
+      "与 社会",
+      "▁in oc",
+      "▁i noc",
+      "nR ight",
+      "n Right",
+      "rodu ce",
+      "rod uce",
+      "▁ad ditions",
+      "▁add itions",
+      "▁addition s",
+      "▁co herent",
+      "▁coh erent",
+      "一 来",
+      "故 人",
+      "▁Im aging",
+      "发表于 晋江文学网",
+      "奢 侈",
+      "生 导师",
+      "知道 了",
+      "下 标",
+      "lo rd",
+      "lor d",
+      "l ord",
+      "nW in",
+      "n Win",
+      "▁w anna",
+      "▁Any one",
+      "聚 餐",
+      "▁s ine",
+      "▁sin e",
+      "▁si ne",
+      "执 导",
+      "膳 食",
+      "▁aut umn",
+      "▁polymorph isms",
+      "▁polymorphism s",
+      "匆 匆",
+      "我 爸",
+      "jd k",
+      "j dk",
+      "Sec onds",
+      "Second s",
+      "▁Pen insula",
+      "滋 味",
+      "▁ 重要参数",
+      "▁immunos up",
+      "▁immuno sup",
+      "志 强",
+      "ax es",
+      "ANGU AGE",
+      "▁Ad elaide",
+      "v b",
+      "邮 寄",
+      "玩 游戏",
+      "op or",
+      "opo r",
+      "o por",
+      "▁dr ill",
+      "▁dri ll",
+      "Con firm",
+      "Conf irm",
+      "▁bi cycle",
+      "▁bic ycle",
+      "▁Re covery",
+      "▁Rec overy",
+      "复 旦",
+      "CL O",
+      "C LO",
+      "中档 题",
+      "Mod ules",
+      "Module s",
+      "ivers al",
+      "iver sal",
+      "i versal",
+      "溶 剂",
+      "▁f ist",
+      "▁fi st",
+      "ho uses",
+      "house s",
+      "hou ses",
+      "h ouses",
+      "无人 机",
+      "ig on",
+      "igo n",
+      "i gon",
+      "Be gin",
+      "B egin",
+      "万 事",
+      "财产 调查",
+      "get Class",
+      "惊 讶",
+      "▁w ore",
+      "▁wor e",
+      "▁wo re",
+      "▁A ctor",
+      "▁Ac tor",
+      "▁Act or",
+      "▁ Actor",
+      "▁back ward",
+      "▁a ure",
+      "▁au re",
+      "▁aur e",
+      "▁ aure",
+      "▁st ems",
+      "▁ste ms",
+      "▁stem s",
+      "▁sp anning",
+      "▁span ning",
+      "发 型",
+      "▁New castle",
+      "方 差",
+      "明 细",
+      "Miss ing",
+      "M issing",
+      "极 具",
+      "海 峡",
+      "▁C YP",
+      "▁CY P",
+      "eff ects",
+      "effect s",
+      "▁album in",
+      "end um",
+      "▁be have",
+      "▁beh ave",
+      "▁behav e",
+      "▁d w",
+      "▁ dw",
+      "▁boot strap",
+      "▁ bootstrap",
+      "▁compet ence",
+      "▁compete nce",
+      "UR N",
+      "U RN",
+      "降 温",
+      "先进 个人",
+      "▁qu art",
+      "▁q uart",
+      "变 异",
+      "FA IL",
+      "FAI L",
+      "▁illustr ations",
+      "▁illustration s",
+      "nC ong",
+      "nCon g",
+      "nCo ng",
+      "n Cong",
+      "▁far mer",
+      "▁farm er",
+      "▁recruit ing",
+      "▁recru iting",
+      "时 针",
+      "至 关",
+      "con ditions",
+      "condition s",
+      "cond itions",
+      "溪 市",
+      "▁d rying",
+      "▁dr ying",
+      "▁dry ing",
+      "▁ 魏",
+      "躺 在",
+      "▁t heat",
+      "▁the at",
+      "▁th eat",
+      "nDes ign",
+      "nDe sign",
+      "n Design",
+      "▁tw isted",
+      "▁twist ed",
+      "区 内",
+      "很 强",
+      "▁J obs",
+      "▁Jo bs",
+      "▁Job s",
+      "▁ Jobs",
+      "▁酒店 地址",
+      "▁ver bal",
+      "▁verb al",
+      "▁de pressed",
+      "▁dep ressed",
+      "力 争",
+      "能力 强",
+      "能 力强",
+      "Dr ive",
+      "D rive",
+      "名 师",
+      "th em",
+      "the m",
+      "t hem",
+      "▁L loyd",
+      "▁dead ly",
+      "▁pat ience",
+      "et erm",
+      "eter m",
+      "ete rm",
+      "e term",
+      "开 辟",
+      "先 用",
+      "天 猫",
+      "ocarcin oma",
+      "tc olor",
+      "t color",
+      "从 前",
+      "此 事",
+      "首 个",
+      "现场 勘",
+      "个百分 点",
+      "▁W ILL",
+      "▁WI LL",
+      "but tons",
+      "button s",
+      "▁med itation",
+      "col on",
+      "co lon",
+      "og ical",
+      "ogi cal",
+      "▁S plit",
+      "▁Sp lit",
+      "▁Spl it",
+      "▁ Split",
+      "Re fresh",
+      "Ref resh",
+      "▁con dem",
+      "▁cond em",
+      "▁He ights",
+      "▁Height s",
+      "出 厂",
+      "演 变",
+      "也不 用",
+      "也 不用",
+      "us tion",
+      "ust ion",
+      "▁Ill egal",
+      "▁Il legal",
+      "▁ Illegal",
+      "ar am",
+      "ara m",
+      "a ram",
+      "酒店 提供",
+      "Get ter",
+      "G etter",
+      "▁Per th",
+      "▁Pert h",
+      "ah u",
+      "a hu",
+      "变 压器",
+      "所有 权",
+      "所 有权",
+      "第一百一 十八条",
+      "▁phy logen",
+      "常州 市",
+      "常 州市",
+      "二次 函数",
+      "▁D ennis",
+      "▁Den nis",
+      "公 立",
+      "▁advoc acy",
+      "▁oc clusion",
+      "▁tutorial s",
+      "▁tutor ials",
+      "▁ tutorials",
+      "Su n",
+      "S un",
+      "ko n",
+      "k on",
+      "实现 的",
+      "▁r itual",
+      "▁ri tual",
+      "imb abwe",
+      "▁Ex port",
+      "▁Exp ort",
+      "▁Expo rt",
+      "▁ Export",
+      "▁L inda",
+      "▁Lin da",
+      "▁Li nda",
+      "▁Lind a",
+      "▁predec ess",
+      "罪犯 王",
+      "▁sn ack",
+      "山 村",
+      "合法 性",
+      "ar is",
+      "ari s",
+      "a ris",
+      "▁ 前",
+      "校 友",
+      "fo n",
+      "f on",
+      "\\' }\\",
+      "\\'} \\",
+      "\\ '}\\",
+      "▁l ev",
+      "▁le v",
+      "▁ lev",
+      "中国 特色",
+      "▁Exper iment",
+      "h w",
+      "▁plastic ity",
+      "▁fund raising",
+      "▁fundra ising",
+      "bug s",
+      "bu gs",
+      "b ugs",
+      "nim pl",
+      "n impl",
+      "▁p ork",
+      "▁po rk",
+      "▁por k",
+      "▁ins ulation",
+      "ed o",
+      "e do",
+      "no k",
+      "n ok",
+      "必须 在",
+      "▁r hyth",
+      "式 子",
+      "gu id",
+      "gui d",
+      "g uid",
+      "Pre view",
+      "P review",
+      "号民事判决书 已发生法律效力",
+      "(\" <",
+      "( \"<",
+      "▁P F",
+      "▁ PF",
+      "nW est",
+      "nWe st",
+      "n West",
+      "ull ivan",
+      "教 科",
+      "La nd",
+      "L and",
+      "门 外",
+      "Cr oss",
+      "C ross",
+      "▁N intendo",
+      "di d",
+      "d id",
+      "en ery",
+      "ener y",
+      "ene ry",
+      "▁sm okers",
+      "▁smoke rs",
+      "▁smo kers",
+      "▁t asty",
+      "▁ta sty",
+      "▁tast y",
+      "帖 子",
+      "Do cs",
+      "Doc s",
+      "D ocs",
+      "um en",
+      "ume n",
+      "u men",
+      "可以 缺席判决",
+      "这几 个",
+      "这 几个",
+      "SES SION",
+      "SESS ION",
+      "到 你",
+      "一般 情况下",
+      "}) (",
+      "} )(",
+      "ud ed",
+      "ude d",
+      "u ded",
+      "ast ruct",
+      "astr uct",
+      "a struct",
+      "凸 显",
+      "录 制",
+      "▁Nigeria n",
+      "▁Niger ian",
+      "▁prolif er",
+      "▁prol ifer",
+      "不 给",
+      "团 购",
+      "▁a nx",
+      "▁an x",
+      "细胞 的",
+      "▁ 编剧",
+      "▁pl anted",
+      "▁plan ted",
+      "▁plant ed",
+      "▁E lection",
+      "▁El ection",
+      "▁Elect ion",
+      "▁Ele ction",
+      "▁ 古",
+      "AR M",
+      "A RM",
+      "ix on",
+      "os eph",
+      "ose ph",
+      "▁He avy",
+      "▁sal mon",
+      "▁D uration",
+      "▁Du ration",
+      "▁Dur ation",
+      "▁ Duration",
+      "is Required",
+      "残 酷",
+      "Ex pr",
+      "Exp r",
+      "ject ive",
+      "j ective",
+      "妇 科",
+      "▁test osterone",
+      "对被告 人",
+      "对 被告人",
+      "oice s",
+      "oi ces",
+      "o ices",
+      "▁be ats",
+      "▁beat s",
+      "例 外",
+      "▁on ion",
+      "号刑事 判决书",
+      "号刑事判决 书",
+      "号 刑事判决书",
+      "fall s",
+      "f alls",
+      "▁C oin",
+      "▁Co in",
+      "▁ Coin",
+      "▁av iation",
+      "音 箱",
+      "省 政府",
+      "▁T CP",
+      "▁TC P",
+      "▁ TCP",
+      "Document s",
+      "Doc uments",
+      "以前 的",
+      "以 前的",
+      "然后 在",
+      "然 后在",
+      "然后 用",
+      "等 内容",
+      "充 实",
+      "国 籍",
+      "Re ady",
+      "Read y",
+      "rec ogn",
+      "习 近平",
+      "儿童 和",
+      "一年 的",
+      "一 年的",
+      "▁N OTE",
+      "▁NOT E",
+      "▁NO TE",
+      "▁ NOTE",
+      "西 区",
+      "Num er",
+      "Nu mer",
+      "N umer",
+      "▁' ',\\",
+      "▁'' ,\\",
+      "▁'', \\",
+      "海 底",
+      "客 场",
+      "nH om",
+      "n Hom",
+      "▁mark ing",
+      "▁mar king",
+      "Bu l",
+      "B ul",
+      "要不 要",
+      "要 不要",
+      "▁vol atility",
+      "ve h",
+      "v eh",
+      "My SQL",
+      "▁Lib erty",
+      "▁Libert y",
+      "ti s",
+      "t is",
+      "▁N az",
+      "▁Na z",
+      "人民 日报",
+      "▁en caps",
+      "▁enc aps",
+      "修 养",
+      "▁f eeds",
+      "▁fe eds",
+      "▁feed s",
+      "▁fee ds",
+      "▁ feeds",
+      "Hel pers",
+      "Helper s",
+      "Help ers",
+      "寒 冷",
+      "值 班",
+      "颁 布",
+      "\"] [\"",
+      "\"][ \"",
+      "\" ][\"",
+      "id yl",
+      "idy l",
+      "▁餐馆 名称",
+      "学 业",
+      "的 吗",
+      "xs l",
+      "x sl",
+      "▁plan ar",
+      "DES CRIPTION",
+      "增 进",
+      "只要 你",
+      "▁cre ep",
+      "▁cr eep",
+      "压 迫",
+      "重 型",
+      "中 方",
+      "是一 本",
+      "是 一本",
+      "ENG TH",
+      "乒乓 球",
+      "▁presum ably",
+      "\\' <",
+      "这两 种",
+      "这 两种",
+      "▁H us",
+      "▁Hu s",
+      "ni o",
+      "n io",
+      "新闻 网",
+      "最多 容纳",
+      "▁E agle",
+      "▁Eag le",
+      "▁exec uting",
+      "▁execut ing",
+      "▁L PS",
+      "▁LP S",
+      "高质量 发展",
+      "▁p umps",
+      "▁pump s",
+      "PE T",
+      "P ET",
+      "干 事",
+      "▁select or",
+      "▁sel ector",
+      "▁ selector",
+      "思 念",
+      "S ocial",
+      "▁iter ations",
+      "▁iteration s",
+      "▁ iterations",
+      "op ard",
+      "opa rd",
+      "o pard",
+      "es pecially",
+      "e specially",
+      "M K",
+      "还不 如",
+      "还 不如",
+      "de ps",
+      "dep s",
+      "d eps",
+      "如果 不是",
+      "如果不 是",
+      "▁sim ulator",
+      "坚 实",
+      "契 约",
+      "long itude",
+      "月 下",
+      "说 他",
+      "▁ 植物",
+      "毫 克",
+      "难 忘",
+      "现已审查 终结",
+      "▁x range",
+      "▁c ascade",
+      "▁casc ade",
+      "▁stim ulating",
+      "▁stimul ating",
+      "} ⋅",
+      "朝阳 区",
+      "朝 阳区",
+      "▁S oon",
+      "▁So on",
+      "▁L inks",
+      "▁Lin ks",
+      "▁Link s",
+      "▁ Links",
+      "不予 受理",
+      "油 田",
+      "▁主要 成就",
+      "最重要的 是",
+      "最重要 的是",
+      "最 重要的是",
+      "▁car otid",
+      "二 元",
+      "▁con tour",
+      "▁cont our",
+      "▁under way",
+      "▁underw ay",
+      "选 中",
+      "中 型",
+      "当 做",
+      "海 水",
+      "可行 性",
+      "独 资",
+      "纪录 片",
+      "▁Re ports",
+      "▁Rep orts",
+      "▁Report s",
+      "▁prosp er",
+      "▁pros per",
+      "▁tensor flow",
+      "▁ tensorflow",
+      "废 物",
+      "▁health ier",
+      "▁ 伟",
+      "▁c p",
+      "▁ cp",
+      "▁sql alchemy",
+      "H K",
+      "不 懈",
+      "Ho m",
+      "H om",
+      "▁ch erry",
+      "▁cher ry",
+      "▁profit able",
+      "▁prof itable",
+      "习 俗",
+      "▁ 适宜",
+      "▁gr ave",
+      "▁gra ve",
+      "▁grav e",
+      "▁ grave",
+      "▁nicot ine",
+      "▁ 程",
+      "财政 部",
+      "身上 的",
+      "身 上的",
+      "▁A na",
+      "▁An a",
+      "▁L ok",
+      "▁Lo k",
+      "▁B T",
+      "▁ BT",
+      "Bo oks",
+      "Book s",
+      "B ooks",
+      "▁u nsure",
+      "▁uns ure",
+      "▁comp ares",
+      "▁compar es",
+      "▁compare s",
+      "▁Jeff erson",
+      "受 众",
+      "▁L aser",
+      "▁La ser",
+      "▁Las er",
+      "▁exec utable",
+      "▁execut able",
+      "▁ executable",
+      "▁B orn",
+      "▁Bo rn",
+      "▁Bor n",
+      "v r",
+      "在全 球",
+      "在 全球",
+      "ur bs",
+      "urb s",
+      "Call ed",
+      "Cal led",
+      "C alled",
+      "机 箱",
+      "过 往",
+      "gom ery",
+      "隔 壁",
+      "▁normal ize",
+      "▁ normalize",
+      "nL ook",
+      "nLo ok",
+      "n Look",
+      "▁orche str",
+      "▁orch estr",
+      "同意 终结本次执行程序",
+      "得 解",
+      "▁an atomy",
+      "成熟 的",
+      "成 熟的",
+      "ic ky",
+      "ick y",
+      "▁sp ouse",
+      "▁spo use",
+      "▁Sh ipping",
+      "▁Ship ping",
+      "▁Shi pping",
+      "▁vol atile",
+      "▁Birth day",
+      "周 一",
+      "才 华",
+      "水 位",
+      "PC I",
+      "P CI",
+      "▁ 房间",
+      "▁de pletion",
+      "▁dep letion",
+      "ou is",
+      "o uis",
+      "ell ows",
+      "ellow s",
+      "ello ws",
+      "▁t ouched",
+      "▁touch ed",
+      "▁tou ched",
+      "收 款",
+      "教 职工",
+      "bi es",
+      "bie s",
+      "b ies",
+      "cre dit",
+      "cr edit",
+      "cred it",
+      "c redit",
+      "又 会",
+      "挣 扎",
+      "敢 于",
+      "G s",
+      "co rd",
+      "cor d",
+      "c ord",
+      "et tes",
+      "ette s",
+      "ett es",
+      "e ttes",
+      "ua tor",
+      "u ator",
+      "平 民",
+      "▁S ustain",
+      "▁Sus tain",
+      "▁biom ass",
+      "▁bio mass",
+      "取 名",
+      "可以 是",
+      "▁sent enced",
+      "▁sentence d",
+      "的 法律",
+      "ra per",
+      "rap er",
+      "rape r",
+      "r aper",
+      "阴 影",
+      "第七 十八",
+      "Data set",
+      "Dat aset",
+      "比 为",
+      "\\\\ *",
+      "日被 刑事拘留",
+      "te chnology",
+      "techn ology",
+      "我 所",
+      "可用 于",
+      "可 用于",
+      "是 国内",
+      "▁rail road",
+      "▁G OP",
+      "▁GO P",
+      "EX PECT",
+      "EXP ECT",
+      "User Id",
+      "▁dens ities",
+      "格 尔",
+      "nat ural",
+      "n atural",
+      "Con stant",
+      "Const ant",
+      "▁th reading",
+      "▁thread ing",
+      "彩 礼",
+      "▁T EST",
+      "▁TE ST",
+      "▁ TEST",
+      "跳 舞",
+      "tp l",
+      "t pl",
+      "中国 经济",
+      "▁al loc",
+      "▁all oc",
+      "▁ alloc",
+      "▁Sal mon",
+      "流 感",
+      "ross over",
+      "rosso ver",
+      "▁c fg",
+      "▁cf g",
+      "▁ cfg",
+      "▁c ation",
+      "▁ca tion",
+      "▁cat ion",
+      "▁ cation",
+      "View ById",
+      "д и",
+      "faster xml",
+      "▁N as",
+      "▁Na s",
+      "的 白",
+      "奖 学金",
+      "管理 中心",
+      "▁comm emor",
+      "nUn incorporated",
+      "▁Guard ian",
+      "▁user Id",
+      "▁ userId",
+      "殖 民",
+      "▁C AP",
+      "▁CA P",
+      "▁ CAP",
+      "▁R eader",
+      "▁Re ader",
+      "▁Read er",
+      "▁ Reader",
+      "▁re actor",
+      "▁react or",
+      "自 古",
+      "▁Car eer",
+      "▁Care er",
+      "▁an atomical",
+      "▁anatomic al",
+      "国 道",
+      "而且 还",
+      "公诉机关指控 的",
+      "dir s",
+      "di rs",
+      "d irs",
+      "二 代",
+      "▁s tap",
+      "▁st ap",
+      "▁At hens",
+      "▁Ath ens",
+      "comfort able",
+      "nO ct",
+      "n Oct",
+      "ve rb",
+      "ver b",
+      "v erb",
+      "抗 菌",
+      "asp berry",
+      "清 朝",
+      "nT est",
+      "nTe st",
+      "n Test",
+      "▁h arness",
+      "▁har ness",
+      "偏 差",
+      "印 发",
+      "并 为",
+      "▁found ers",
+      "▁fo unders",
+      "▁founder s",
+      "▁understand s",
+      "▁d eline",
+      "▁de line",
+      "▁del ine",
+      "g overnment",
+      "▁N y",
+      "▁general ization",
+      "海 域",
+      "IG H",
+      "I GH",
+      "▁Cl uster",
+      "▁ Cluster",
+      "▁H our",
+      "▁Ho ur",
+      "▁Hou r",
+      "▁ Hour",
+      "第一百零 三",
+      "▁re sear",
+      "▁res ear",
+      "Mult iple",
+      "Multi ple",
+      "两 家",
+      "▁G am",
+      "▁Ga m",
+      "man ia",
+      "ma nia",
+      "mani a",
+      "m ania",
+      "准予 强制执行",
+      "采 样",
+      "session s",
+      "sess ions",
+      "s essions",
+      "▁Nurs ing",
+      "▁Nur sing",
+      "ni d",
+      "n id",
+      "oy a",
+      "o ya",
+      "▁m as",
+      "▁ma s",
+      "▁ mas",
+      "的 了",
+      "逃 避",
+      "的主 题",
+      "的 主题",
+      "的 症状",
+      "nS ur",
+      "n Sur",
+      "▁O ften",
+      "▁Of ten",
+      "▁dr unk",
+      "▁expl osion",
+      "▁explos ion",
+      "光 线",
+      "th reat",
+      "thr eat",
+      "▁seiz ures",
+      "▁seizure s",
+      "▁dep recated",
+      "▁deprec ated",
+      "▁ deprecated",
+      "▁re build",
+      "▁reb uild",
+      "一 副",
+      "限 期",
+      "▁b os",
+      "▁bo s",
+      "▁ bos",
+      "cer pt",
+      "▁Def ence",
+      "二〇一五年 十二月",
+      "胆 固醇",
+      "er ie",
+      "eri e",
+      "e rie",
+      "'' '",
+      "' ''",
+      "nB en",
+      "nBe n",
+      "n Ben",
+      "▁Paul o",
+      "▁Pa ulo",
+      "和 工作",
+      "▁collabor ate",
+      "人 之",
+      "要 使",
+      "np ro",
+      "npr o",
+      "n pro",
+      "暴 雨",
+      "▁r ope",
+      "▁ro pe",
+      "离婚 纠纷一案中",
+      "离婚纠纷一案 中",
+      "Sol ution",
+      "S olution",
+      "代 书",
+      "蒸 发",
+      "▁L is",
+      "▁Li s",
+      "户籍 所在地",
+      "ct uary",
+      "▁Br other",
+      "▁Bro ther",
+      "He r",
+      "H er",
+      "gan g",
+      "ga ng",
+      "g ang",
+      "iph one",
+      "i phone",
+      "ro gate",
+      "rog ate",
+      "▁fib rin",
+      "▁fibr in",
+      "世界 各地",
+      "不同 类型的",
+      "不 和",
+      "估 算",
+      "时时 彩",
+      "▁B yte",
+      "▁By te",
+      "▁ Byte",
+      "▁un used",
+      "▁unus ed",
+      "▁ unused",
+      ": <",
+      "冠 状",
+      "联系 方式",
+      "▁H illary",
+      "▁Hill ary",
+      "可以 把",
+      "影响 的",
+      "▁capt uring",
+      "▁dismiss ed",
+      "这 事",
+      "限 定",
+      "ig gs",
+      "igg s",
+      "▁pit uitary",
+      "下 辖",
+      "生产 经营",
+      "▁pharm acy",
+      "▁pharmac y",
+      "▁success ive",
+      "▁succ essive",
+      "抱 歉",
+      "电 线",
+      "麒 麟",
+      "批准 后",
+      "De ep",
+      "D eep",
+      "▁L ac",
+      "▁La c",
+      "▁ex tern",
+      "▁ext ern",
+      "▁ extern",
+      "▁mod ifying",
+      "▁modify ing",
+      "ı n",
+      "尺 度",
+      "督 查",
+      "自愿 认罪",
+      "捐 款",
+      "合 唱",
+      "▁ 型号",
+      "体现 在",
+      "体 现在",
+      "urs ors",
+      "ursor s",
+      "被 盗",
+      "日 依法",
+      "▁C ru",
+      "▁Cr u",
+      "一点 的",
+      "一 点的",
+      "fr am",
+      "fra m",
+      "f ram",
+      "ox ic",
+      "中国科学院 大学",
+      "anal ytics",
+      "analytic s",
+      "▁Develop er",
+      "▁ Developer",
+      "▁H V",
+      "▁ HV",
+      "▁Tes la",
+      "曲线 的",
+      "曲 线的",
+      "▁mon ster",
+      "菏 泽",
+      "▁aff iliates",
+      "▁affil iates",
+      "▁affiliate s",
+      "华 盛",
+      "任何 人",
+      "任 何人",
+      "特别 的",
+      "特 别的",
+      "▁in nings",
+      "▁inn ings",
+      "▁inning s",
+      "故 宫",
+      "▁s sh",
+      "▁ss h",
+      "▁ ssh",
+      "role um",
+      "万 的",
+      "获 利",
+      "衰 竭",
+      "LE T",
+      "L ET",
+      "▁S ak",
+      "▁Sa k",
+      "part ner",
+      "nDef unct",
+      "就 得",
+      "华 北",
+      "▁St atic",
+      "▁Stat ic",
+      "▁Sta tic",
+      "▁ Static",
+      "id as",
+      "ida s",
+      "i das",
+      "▁inaug ural",
+      "市 新",
+      "犹 如",
+      "▁s py",
+      "▁sp y",
+      "▁ spy",
+      "▁sen iors",
+      "▁senior s",
+      "派 遣",
+      "MO DEL",
+      "MODE L",
+      "MOD EL",
+      "▁w ast",
+      "▁was t",
+      "▁wa st",
+      "▁Ne ural",
+      "学 子",
+      "都会 有",
+      "都 会有",
+      "经 人介绍",
+      "▁ 青",
+      "该项 目",
+      "该 项目",
+      "ma cro",
+      "mac ro",
+      "产 的",
+      "nI d",
+      "n Id",
+      "amb oo",
+      "ambo o",
+      "hand ed",
+      "han ded",
+      "▁r ises",
+      "▁rise s",
+      "▁ris es",
+      "▁ri ses",
+      "▁se psis",
+      "▁sep sis",
+      "▁re alm",
+      "▁real m",
+      "▁lip ids",
+      "▁lipid s",
+      "粉 末",
+      "▁DO I",
+      "br ities",
+      "brit ies",
+      "▁spokes person",
+      "村 里",
+      "le sh",
+      "les h",
+      "l esh",
+      "▁mount ing",
+      "▁mo unting",
+      "面 子",
+      "npm js",
+      "sm ith",
+      "s mith",
+      "率 的",
+      "理论 和",
+      "中 和",
+      "就 医",
+      "▁Municip al",
+      "高 新",
+      "▁F it",
+      "▁Fi t",
+      "▁ Fit",
+      "fab ric",
+      "f abric",
+      "▁As set",
+      "▁Ass et",
+      "▁ Asset",
+      "博 弈",
+      "▁mod ulus",
+      "很 可能",
+      "的 职业",
+      "钢 结构",
+      "nD iv",
+      "nDi v",
+      "n Div",
+      "▁per imeter",
+      "▁peri meter",
+      "开 采",
+      "▁s per",
+      "▁sp er",
+      "▁spe r",
+      "堪 称",
+      "yz e",
+      "y ze",
+      "与 你",
+      "前 途",
+      "数 目",
+      "角 是",
+      "it as",
+      "ita s",
+      "i tas",
+      "▁fl ush",
+      "▁flu sh",
+      "▁ flush",
+      "应 为",
+      "行政 执法",
+      "▁cha os",
+      "▁Pro grams",
+      "▁Pr ograms",
+      "▁Program s",
+      "前列 腺",
+      "支付 令",
+      "ot ides",
+      "otide s",
+      "otid es",
+      "oti des",
+      "cc c",
+      "c cc",
+      "in j",
+      "i nj",
+      "nL ocal",
+      "nLo cal",
+      "n Local",
+      "▁P seud",
+      "▁recombin ation",
+      "▁M ong",
+      "▁Mon g",
+      "▁Mo ng",
+      "ered Reader",
+      "特 意",
+      "太多 的",
+      "太 多的",
+      "▁head ache",
+      "诉至 本院",
+      "食品 有限公司",
+      "▁mer ger",
+      "▁merge r",
+      "▁merg er",
+      "出 品",
+      "应 聘",
+      "时 有",
+      "▁V ideos",
+      "▁Video s",
+      "▁ap prove",
+      "▁app rove",
+      "▁appro ve",
+      "▁star red",
+      "名 额",
+      "如实供述自己的 罪行",
+      "底 下",
+      "▁f lap",
+      "▁fl ap",
+      "▁ 晋江文学城",
+      "植 株",
+      "os phere",
+      "osp here",
+      "osph ere",
+      "o sphere",
+      "vel ocity",
+      "的 神",
+      "é l",
+      "觉 醒",
+      "▁D ebug",
+      "▁De bug",
+      "▁Deb ug",
+      "▁ Debug",
+      "ā n",
+      "交通肇事 罪",
+      "▁al erts",
+      "▁alert s",
+      "▁py game",
+      "▁ pygame",
+      "nDem ographics",
+      "产业 化",
+      "uls ive",
+      "法定代表人 李",
+      "徐州 市",
+      "徐 州市",
+      "的对 象",
+      "的 对象",
+      "az or",
+      "azo r",
+      "▁v et",
+      "▁ve t",
+      "▁ vet",
+      "H uman",
+      "ens ure",
+      "e nsure",
+      "\\\\ \\\\\\",
+      "\\\\\\\\ \\",
+      "\\\\\\ \\\\",
+      "\\ \\\\\\\\",
+      "▁sn ake",
+      "▁Mot ors",
+      "▁Motor s",
+      "▁Moto rs",
+      "▁con greg",
+      "▁congr eg",
+      "▁cong reg",
+      "▁已 知",
+      "▁ 已知",
+      "he ads",
+      "head s",
+      "hea ds",
+      "磷 酸",
+      "▁rain fall",
+      "的 自",
+      "ée s",
+      "é es",
+      "的 文章",
+      "▁le pt",
+      "▁lo ader",
+      "▁load er",
+      "▁ loader",
+      "▁atheros clerosis",
+      "整 洁",
+      "il io",
+      "ili o",
+      "i lio",
+      "▁b oil",
+      "▁bo il",
+      "▁Latin o",
+      "▁Lat ino",
+      "us hi",
+      "ush i",
+      "u shi",
+      "in ski",
+      "ins ki",
+      "insk i",
+      "和 水",
+      "杀 死",
+      "▁Out door",
+      "真 理",
+      "色 素",
+      "停车 位",
+      "停 车位",
+      "▁en force",
+      "N L",
+      "as ta",
+      "ast a",
+      "a sta",
+      "ne ll",
+      "nel l",
+      "n ell",
+      "data s",
+      "da tas",
+      "dat as",
+      "d atas",
+      "交 点",
+      "BI T",
+      "B IT",
+      "▁Wi Fi",
+      "▁ WiFi",
+      "▁fab rics",
+      "▁fabric s",
+      "▁Phil lips",
+      "▁Phill ips",
+      "▁Phillip s",
+      "哪 儿",
+      "驳回 起诉",
+      "▁c arbox",
+      "▁car box",
+      "▁carb ox",
+      "▁bl essed",
+      "▁bless ed",
+      "▁fil med",
+      "▁film ed",
+      "好 用",
+      "就是 这样",
+      "就 是这样",
+      "▁Col ors",
+      "▁Color s",
+      "▁ Colors",
+      "nd efined",
+      "ndef ined",
+      "ndefine d",
+      "n defined",
+      "ib a",
+      "i ba",
+      "▁p ins",
+      "▁pi ns",
+      "▁pin s",
+      "▁ pins",
+      "▁renew ed",
+      "月 光",
+      "你 需要",
+      "马 路",
+      "他人 的",
+      "他 人的",
+      "共青 团",
+      "的 角色",
+      "共同 举办",
+      "▁Par ameter",
+      "▁Para meter",
+      "▁Param eter",
+      "▁ Parameter",
+      "▁F ro",
+      "▁Fr o",
+      "re load",
+      "rel oad",
+      "一审 被告",
+      "裁定 适用于",
+      "label ed",
+      "se l",
+      "s el",
+      "年 月",
+      "起 义",
+      "mod ern",
+      "mode rn",
+      "的 个数",
+      "▁sc ala",
+      "▁scal a",
+      "▁ scala",
+      "的重要 组成部分",
+      "Product s",
+      "Produ cts",
+      "an ni",
+      "ann i",
+      "h appy",
+      "▁l asted",
+      "▁last ed",
+      "▁la sted",
+      "▁las ted",
+      "▁ind ications",
+      "▁indic ations",
+      "▁indication s",
+      "▁access ory",
+      "▁debut ed",
+      "▁deb uted",
+      "▁observ ational",
+      "▁observation al",
+      ": '",
+      "议 员",
+      "▁M ik",
+      "▁Mi k",
+      "在 向",
+      "的 内",
+      "▁B utter",
+      "▁But ter",
+      "会 儿",
+      "ict ions",
+      "iction s",
+      "Ho use",
+      "H ouse",
+      "在 水",
+      "性 疾病",
+      "Lower Case",
+      "近 年",
+      "▁cr ashes",
+      "▁crash es",
+      "▁appar atus",
+      "但 如果",
+      "官方 网站",
+      "狭 窄",
+      "的 成功",
+      "a ção",
+      "G U",
+      "Sam e",
+      "Sa me",
+      "S ame",
+      "简 要",
+      "be e",
+      "b ee",
+      "的 语言",
+      "ig rate",
+      "igr ate",
+      "▁ 彭",
+      "有力 的",
+      "有 力的",
+      "的一 首",
+      "的 一首",
+      "Valid ate",
+      "▁hem oglobin",
+      "IF F",
+      "I FF",
+      "D J",
+      "河 县",
+      "M vc",
+      "ros ophila",
+      "oc ortic",
+      "▁Week ly",
+      "▁lymph ocyte",
+      "ut t",
+      "u tt",
+      "ac ier",
+      "aci er",
+      "流 转",
+      "龙 岩",
+      "任何 一个",
+      "▁p asta",
+      "▁past a",
+      "▁pa sta",
+      "▁pas ta",
+      "▁com orb",
+      "▁como rb",
+      "▁j umped",
+      "▁jump ed",
+      "任 教",
+      "行政 机关",
+      "▁p ound",
+      "▁po und",
+      "▁pou nd",
+      "▁ pound",
+      "会 产生",
+      "新中国 的",
+      "新 中国的",
+      "ter one",
+      "te rone",
+      "▁Cy ber",
+      "未作 答辩",
+      "coll apse",
+      "的 限制",
+      "le igh",
+      "lei gh",
+      "▁administr ators",
+      "▁administrator s",
+      "ou bt",
+      "nH ead",
+      "nHe ad",
+      "n Head",
+      "▁con strained",
+      "▁cons trained",
+      "▁constr ained",
+      "k afka",
+      "fin ish",
+      "ER C",
+      "E RC",
+      "ge v",
+      "g ev",
+      "销售 额",
+      "st own",
+      "sto wn",
+      "s town",
+      "缓刑考验 期限",
+      "am etric",
+      "amet ric",
+      "a metric",
+      "▁me dial",
+      "▁med ial",
+      "▁media l",
+      "▁medi al",
+      "冲 动",
+      "孵 化",
+      "杀 菌",
+      "ro e",
+      "r oe",
+      "高 尔夫",
+      "其实 就是",
+      "▁L atest",
+      "▁La test",
+      "▁Lat est",
+      "▁Late st",
+      "▁ Latest",
+      "▁in stitute",
+      "▁instit ute",
+      "▁ 董",
+      "\"] ;\\",
+      "\" ];\\",
+      "▁h alt",
+      "▁ha lt",
+      "▁hal t",
+      "▁ 形态特征",
+      "▁an tip",
+      "▁ant ip",
+      "▁anti p",
+      "▁antagon ists",
+      "▁antagonist s",
+      "er ness",
+      "ern ess",
+      "▁cyt oplasmic",
+      "▁cytoplasm ic",
+      "mo v",
+      "m ov",
+      "起 伏",
+      "▁R af",
+      "▁Ra f",
+      "co res",
+      "core s",
+      "cor es",
+      "c ores",
+      ") %",
+      "▁dis claimer",
+      "CA L",
+      "C AL",
+      "▁s clerosis",
+      "▁scler osis",
+      "re b",
+      "r eb",
+      "ei ght",
+      "e ight",
+      "mong odb",
+      "mongo db",
+      "亲 人",
+      "▁f ridge",
+      "▁fr idge",
+      "▁cr ushing",
+      "▁crush ing",
+      "科学 与",
+      "▁ur ged",
+      "▁urg ed",
+      "▁urge d",
+      "▁Phil osoph",
+      "▁anticip ate",
+      "nI ns",
+      "nIn s",
+      "n Ins",
+      "▁j est",
+      "▁je st",
+      "▁ jest",
+      "duct ory",
+      "ductor y",
+      "裁定适用于 下列范围",
+      "▁glut amate",
+      "第三 次",
+      "第 三次",
+      "Print f",
+      "▁p oses",
+      "▁pos es",
+      "▁po ses",
+      "▁pose s",
+      "▁ poses",
+      "▁per mutation",
+      "▁permut ation",
+      "▁perm utation",
+      "g m",
+      "华 侨",
+      "▁d iving",
+      "▁di ving",
+      "▁div ing",
+      "封 面",
+      "的 想法",
+      "▁broad band",
+      "▁ 内",
+      "oth ers",
+      "other s",
+      "othe rs",
+      "o thers",
+      "▁sent iment",
+      "▁& \\\\",
+      "▁&\\ \\",
+      "▁ &\\\\",
+      "请仔细 阅读",
+      "请 仔细阅读",
+      "▁m asks",
+      "▁mask s",
+      "▁mas ks",
+      "发行 的",
+      "发 行的",
+      "pa pers",
+      "paper s",
+      "p apers",
+      "▁d ialysis",
+      "▁dial ysis",
+      "▁exceed ed",
+      "▁Form ation",
+      "▁Format ion",
+      "迎来 了",
+      "迎 来了",
+      "▁S ara",
+      "▁Sar a",
+      "▁Sa ra",
+      "al gebra",
+      "alg ebra",
+      "▁neuro trans",
+      "▁neur otrans",
+      "▁neurot rans",
+      "cent ered",
+      "center ed",
+      "▁hierarch ical",
+      "建 有",
+      "泉 州",
+      "Fil l",
+      "Fi ll",
+      "F ill",
+      "▁matt ress",
+      "▁out reach",
+      "女性 的",
+      "女 性的",
+      "不好 意思",
+      "▁fib re",
+      "▁fibr e",
+      "▁fi bre",
+      "Text View",
+      "▁depos ited",
+      "▁deposit ed",
+      "Cam p",
+      "Ca mp",
+      "C amp",
+      "▁transcript s",
+      "▁tran scripts",
+      "光 辉",
+      "oint ers",
+      "oin ters",
+      "oi nters",
+      "ointer s",
+      "o inters",
+      "rot ation",
+      "r otation",
+      "Al ign",
+      "Ali gn",
+      "网 路",
+      "一共 有",
+      "一 共有",
+      "de leted",
+      "delete d",
+      "del eted",
+      "delet ed",
+      "▁c leans",
+      "▁cle ans",
+      "▁clean s",
+      "▁orig inated",
+      "▁origin ated",
+      "▁originate d",
+      "▁z h",
+      "▁ zh",
+      "PH ONE",
+      "od en",
+      "ode n",
+      "o den",
+      "的一 年",
+      "的 一年",
+      "▁hung ry",
+      "et ti",
+      "ett i",
+      "e tti",
+      "▁S AP",
+      "▁SA P",
+      "▁ SAP",
+      "技术 创新",
+      "Col umns",
+      "Column s",
+      "Colum ns",
+      "盯 着",
+      "辩 解",
+      "yl an",
+      "yla n",
+      "y lan",
+      "▁De pending",
+      "▁Dep ending",
+      "▁Depend ing",
+      "A Z",
+      "Co ver",
+      "Cov er",
+      "C over",
+      "▁strugg led",
+      "▁struggle d",
+      "丰 田",
+      "可 乐",
+      "的 处分",
+      "▁f ade",
+      "▁fa de",
+      "▁ fade",
+      "▁tri angular",
+      "▁triang ular",
+      "\"\" \\",
+      "\" \"\\",
+      "▁top ical",
+      "▁topic al",
+      "▁B uch",
+      "▁Bu ch",
+      "▁Buc h",
+      "ar ctic",
+      "arc tic",
+      "arct ic",
+      "本案 诉讼费用",
+      "本案诉讼费 用",
+      "▁en closed",
+      "▁prof iling",
+      "▁profil ing",
+      "▁S ans",
+      "▁San s",
+      "▁Sa ns",
+      "▁list eners",
+      "▁listen ers",
+      "▁listener s",
+      "▁sympt omatic",
+      "▁symptom atic",
+      "De r",
+      "D er",
+      "▁R d",
+      "目的 地",
+      "目 的地",
+      "▁T han",
+      "▁Th an",
+      "▁ Than",
+      "> •",
+      "名 牌",
+      "塑 胶",
+      "为我 们",
+      "为 我们",
+      "▁A Z",
+      "▁ AZ",
+      "大概 是",
+      "▁arch itects",
+      "▁architect s",
+      "白 了",
+      "的 实际",
+      "▁N ovel",
+      "▁No vel",
+      "▁Nov el",
+      "al though",
+      "alth ough",
+      "▁s ailing",
+      "▁sa iling",
+      "▁sail ing",
+      "▁nom inations",
+      "▁nomin ations",
+      "▁nomination s",
+      "▁o z",
+      "▁ oz",
+      "同 理",
+      "报 导",
+      "集中 在",
+      "共同 财产",
+      "社会 保险",
+      "nD raw",
+      "nDr aw",
+      "n Draw",
+      "具有 良好的",
+      "具 有良好的",
+      "▁inter leukin",
+      "营业 执照",
+      "▁ad op",
+      "Mat cher",
+      "Match er",
+      "M atcher",
+      "▁privile ge",
+      "nBl ack",
+      "n Black",
+      "im ports",
+      "import s",
+      "imp orts",
+      "▁re minded",
+      "▁remind ed",
+      "▁remin ded",
+      "油 画",
+      "第一 天",
+      "第 一天",
+      "▁par alle",
+      "▁paral le",
+      "▁ paralle",
+      "扣 留",
+      "▁l ol",
+      "▁lo l",
+      "▁ lol",
+      "▁Gi ants",
+      "▁Giant s",
+      "▁Gian ts",
+      "▁cardi omy",
+      "▁cardio my",
+      "▁in active",
+      "▁CON NECTION",
+      "jw t",
+      "j wt",
+      "▁ 性质",
+      "▁Ser ge",
+      "▁Serg e",
+      "t W",
+      "狐 狸",
+      "薄 弱",
+      "▁ 历史",
+      "收 回",
+      "时 节",
+      "▁W ells",
+      "▁Well s",
+      "▁Wel ls",
+      "层 层",
+      "确实 是",
+      "Ca st",
+      "Cas t",
+      "C ast",
+      "▁N ine",
+      "▁Ni ne",
+      "▁Nin e",
+      "▁double d",
+      "▁dou bled",
+      "▁doub led",
+      "工程 学院",
+      "▁taste s",
+      "▁tast es",
+      "lt imate",
+      "▁pul ses",
+      "▁pulse s",
+      "▁puls es",
+      "的 黑",
+      "GI S",
+      "G IS",
+      "Not es",
+      "No tes",
+      "Note s",
+      "N otes",
+      "ug oslav",
+      "▁multip lying",
+      "▁multiply ing",
+      "观 音",
+      "Ang le",
+      "phosph ate",
+      "刷 卡",
+      "/ ~",
+      "工 伤",
+      "在 国际",
+      "▁sh rub",
+      "▁shr ub",
+      "fe eding",
+      "feed ing",
+      "fee ding",
+      "▁ م",
+      "相 乘",
+      "od os",
+      "odo s",
+      "o dos",
+      "▁e co",
+      "▁ec o",
+      "▁ eco",
+      "ha gen",
+      "h agen",
+      "▁m ongoose",
+      "▁mongo ose",
+      "▁mong oose",
+      "▁ mongoose",
+      "▁guitar ist",
+      "▁inc ubation",
+      "▁incub ation",
+      "经历 过",
+      "还 请",
+      "un ders",
+      "und ers",
+      "under s",
+      "unde rs",
+      "▁Ang ular",
+      "体系 的",
+      "体 系的",
+      "▁s olver",
+      "▁sol ver",
+      "▁solve r",
+      "▁solv er",
+      "▁ solver",
+      "P b",
+      "S ensor",
+      "▁c ables",
+      "▁ca bles",
+      "▁cable s",
+      "▁cab les",
+      "▁t ribut",
+      "▁tri but",
+      "▁trib ut",
+      "马 鞍",
+      "Op s",
+      "O ps",
+      "▁ten ure",
+      "历 代",
+      "▁ 体重",
+      "ip ation",
+      "ipa tion",
+      "▁sw elling",
+      "] `",
+      "问 问",
+      "住所地 安徽省",
+      "的 红",
+      "▁loyal ty",
+      "▁我 们",
+      "▁ 我们",
+      "▁L OG",
+      "▁LO G",
+      "▁ LOG",
+      "ree ze",
+      "r eeze",
+      "▁S ongs",
+      "▁Son gs",
+      "▁Song s",
+      "ac l",
+      "a cl",
+      "▁gl ue",
+      "▁glu e",
+      "Mem bers",
+      "Member s",
+      "M embers",
+      "并 经",
+      "要 说",
+      "red dit",
+      "▁bal con",
+      "Com ments",
+      "Comment s",
+      "Comm ents",
+      "Se nt",
+      "Sen t",
+      "S ent",
+      "St rings",
+      "String s",
+      "Str ings",
+      "Pipe line",
+      "P ipeline",
+      "胰 岛",
+      "▁S HA",
+      "▁SH A",
+      "▁ SHA",
+      "结果 显示",
+      "只 想",
+      "nu p",
+      "n up",
+      "年 全国",
+      "判断 即可",
+      "我 将",
+      "NO S",
+      "N OS",
+      "▁rev enues",
+      "▁revenue s",
+      "▁reven ues",
+      "地 形",
+      "作为 一种",
+      "的那 个",
+      "的 那个",
+      "人均 纯收入",
+      "MPL ATE",
+      "堆 积",
+      "▁brand ing",
+      "▁interpol ation",
+      "st em",
+      "ste m",
+      "s tem",
+      "▁ а",
+      "查 控",
+      "▁A CC",
+      "▁AC C",
+      "▁ ACC",
+      "有所 不同",
+      "投 案",
+      "内容 为",
+      "女 方",
+      "先 求出",
+      "当然 是",
+      "朋友 的",
+      "C MAKE",
+      "br ains",
+      "bra ins",
+      "brain s",
+      "刻 意",
+      "又 在",
+      "孔 子",
+      "尸 体",
+      "明 日",
+      "ir teen",
+      "irt een",
+      "EA R",
+      "E AR",
+      "dr o",
+      "d ro",
+      "随 访",
+      "▁s ilk",
+      "▁sil k",
+      "治疗 后",
+      "有 大",
+      "z t",
+      "▁ σ",
+      "▁Ste phan",
+      "▁Step han",
+      "nEx pand",
+      "n Expand",
+      "Mo vie",
+      "M ovie",
+      "St ock",
+      "符合 减刑条件",
+      "符合减刑 条件",
+      "▁definit ive",
+      "▁defin itive",
+      "田 园",
+      "na ny",
+      "nan y",
+      "n any",
+      "▁en riched",
+      "▁enrich ed",
+      "▁m anga",
+      "▁man ga",
+      "▁mang a",
+      "▁ 黑",
+      "Au d",
+      "A ud",
+      "▁Phot os",
+      "▁Photo s",
+      "废 水",
+      "▁o cular",
+      "▁oc ular",
+      "▁ ocular",
+      "的事 项",
+      "的 事项",
+      "ar lier",
+      "arl ier",
+      "rot ate",
+      "▁Br exit",
+      "▁Bre xit",
+      "▁Th row",
+      "▁Thr ow",
+      "▁ Throw",
+      "▁conn ector",
+      "▁connect or",
+      "▁ connector",
+      "ch or",
+      "cho r",
+      "c hor",
+      "on na",
+      "onn a",
+      "Sh ift",
+      "S hift",
+      "nH ope",
+      "nHop e",
+      "▁web pack",
+      "▁ webpack",
+      "▁histor ically",
+      "▁historical ly",
+      "▁historic ally",
+      "游戏 大小",
+      "保 罗",
+      "市 的",
+      "本 期",
+      "气 味",
+      "sl ider",
+      "slide r",
+      "▁c amps",
+      "▁camp s",
+      "▁cam ps",
+      "min utes",
+      "minute s",
+      "▁sup plementation",
+      "▁supplement ation",
+      "入 职",
+      "}} </",
+      "} }</",
+      "▁M ine",
+      "▁Min e",
+      "▁Mi ne",
+      "▁R ac",
+      "▁Ra c",
+      "Ch rist",
+      "Chris t",
+      "▁st ained",
+      "▁stain ed",
+      "▁Disc ussion",
+      "脱 落",
+      "▁G MT",
+      "▁GM T",
+      "sc heme",
+      "sch eme",
+      "sche me",
+      "热 线",
+      "还是 会",
+      "还 是会",
+      "▁S car",
+      "▁Sc ar",
+      "的 他",
+      "fi b",
+      "f ib",
+      "ling s",
+      "lin gs",
+      "l ings",
+      "prom pt",
+      "val ued",
+      "value d",
+      "预交 案件受理费",
+      "减 法",
+      "nS am",
+      "n Sam",
+      "▁Tree Node",
+      "▁ TreeNode",
+      "Sub scription",
+      "局 势",
+      "的 积",
+      "or escence",
+      "ores cence",
+      "借贷 关系",
+      "无 一",
+      "om item",
+      "omi tem",
+      "▁\" ~",
+      "am ation",
+      "ama tion",
+      "G W",
+      "ç a",
+      "中 信",
+      "兴 起",
+      "▁B ee",
+      "▁Be e",
+      "Se nder",
+      "Send er",
+      "Sen der",
+      "S ender",
+      "▁exc av",
+      "不 分",
+      "网 的",
+      "无所 谓",
+      "无 所谓",
+      "▁b ikes",
+      "▁bi kes",
+      "▁bike s",
+      "▁bik es",
+      "杂 交",
+      "▁Gu est",
+      ",, ,,",
+      ",,, ,",
+      ", ,,,",
+      "ç ões",
+      "am burg",
+      "amb urg",
+      "第四百八 十七条",
+      "▁he ights",
+      "▁height s",
+      "被 打",
+      "up loads",
+      "upload s",
+      "弘 历",
+      "int ent",
+      "▁ex empl",
+      "▁proof s",
+      "cons ider",
+      "cr umb",
+      "crum b",
+      "▁Al ban",
+      "▁Alb an",
+      "▁ar omatic",
+      "▁arom atic",
+      "▁aroma tic",
+      "买 入",
+      "▁su bt",
+      "▁sub t",
+      "▁off spring",
+      "小小 的",
+      "小 小的",
+      "nL ong",
+      "nLo ng",
+      "n Long",
+      "▁ 纲",
+      "它 在",
+      "nn o",
+      "n no",
+      "▁是 否",
+      "▁ 是否",
+      "__ ()\\",
+      "__( )\\",
+      "ym phony",
+      "ymph ony",
+      "怪 兽",
+      "优质 的",
+      "优 质的",
+      "So ft",
+      "S oft",
+      "▁intent ions",
+      "▁intention s",
+      "墙 面",
+      "案 经过",
+      "▁Le af",
+      "▁ Leaf",
+      "身体 的",
+      "身 体的",
+      "pr iv",
+      "pri v",
+      "p riv",
+      "▁S election",
+      "▁Se lection",
+      "▁Select ion",
+      "▁Sel ection",
+      "▁ Selection",
+      "联 通",
+      "的一 面",
+      "的 一面",
+      "▁bu lb",
+      "▁bul b",
+      "st ead",
+      "ste ad",
+      "▁mon key",
+      "伴 随",
+      "引 流",
+      "装 的",
+      "ot ional",
+      "otion al",
+      "oti onal",
+      "omitem pty",
+      "▁ex clusion",
+      "▁exclus ion",
+      "▁qu ad",
+      "▁ quad",
+      "▁Flor ence",
+      "▁Flo rence",
+      "▁Flore nce",
+      "罪犯 张",
+      "at ex",
+      "ate x",
+      "a tex",
+      "歌 唱",
+      "▁M MP",
+      "▁MM P",
+      "▁ MMP",
+      "的 脸",
+      "▁a i",
+      "▁ ai",
+      "er gy",
+      "erg y",
+      "re shape",
+      "resh ape",
+      "nR un",
+      "n Run",
+      "▁l od",
+      "▁lo d",
+      "▁ lod",
+      "nS earch",
+      "nSe arch",
+      "n Search",
+      "▁so oner",
+      "▁soon er",
+      "▁soo ner",
+      "▁gar bage",
+      "互 补",
+      "ell ij",
+      "elli j",
+      "ns ource",
+      "n source",
+      "▁sh elves",
+      "▁shel ves",
+      "▁c ann",
+      "▁can n",
+      "▁ca nn",
+      "身份证 号码",
+      "身份证号 码",
+      "▁Th irty",
+      "申请执行人 李",
+      "审 议",
+      "红 楼",
+      "° \\",
+      "负 有",
+      "nF ROM",
+      "n FROM",
+      "▁Te aching",
+      "▁Tea ching",
+      "}, {",
+      "} ,{",
+      "座谈 会",
+      "bin s",
+      "bi ns",
+      "b ins",
+      "▁LO VE",
+      "▁所属 专辑",
+      "ic z",
+      "i cz",
+      "▁B D",
+      "▁ BD",
+      "▁lith ium",
+      "▁play ground",
+      "过 户",
+      "fl ies",
+      "f lies",
+      "▁T ile",
+      "▁Ti le",
+      "▁Til e",
+      "▁ Tile",
+      "Al gorithm",
+      "饺 子",
+      "▁d ich",
+      "▁di ch",
+      "▁dic h",
+      "我 把",
+      "了一 条",
+      "了 一条",
+      "▁h emis",
+      "▁he mis",
+      "▁hem is",
+      ": [",
+      "最 美",
+      "了一 系列",
+      "了 一系列",
+      "std lib",
+      "la p",
+      "l ap",
+      "莫 斯科",
+      "▁Z en",
+      "▁Ze n",
+      "▁View s",
+      "▁Vi ews",
+      "▁Vie ws",
+      "▁ Views",
+      "our cing",
+      "应 届",
+      "▁retail er",
+      "兼 顾",
+      "▁} $",
+      "▁ }$",
+      "▁P apers",
+      "▁Pa pers",
+      "▁Paper s",
+      "▁Pap ers",
+      "nS ometimes",
+      "nSome times",
+      "▁strugg les",
+      "▁struggle s",
+      "乘 方",
+      "▁Ex ternal",
+      "▁Ext ernal",
+      "▁ External",
+      "▁mon itors",
+      "▁monitor s",
+      "▁oblig ation",
+      "▁obl igation",
+      "▁日 本",
+      "▁ 日本",
+      "很多 人都",
+      "很多人 都",
+      "is mic",
+      "ism ic",
+      "ploy ed",
+      "▁o ptic",
+      "▁op tic",
+      "▁opt ic",
+      "▁ optic",
+      "nP roduction",
+      "nPro duction",
+      "nProduct ion",
+      "n Production",
+      "▁ill ustration",
+      "▁illustr ation",
+      "▁illust ration",
+      "Git hub",
+      "G ithub",
+      "ial ysis",
+      "该 校",
+      "Me asure",
+      "▁perc ussion",
+      "解 毒",
+      "▁equ ip",
+      "nGe orge",
+      "n George",
+      "几 千",
+      "索 要",
+      "一案 中",
+      "理事 会",
+      "的 功效",
+      "尚 欠",
+      "蓝 天",
+      "ing le",
+      "二〇一五年 十一月",
+      "CA T",
+      "C AT",
+      "mo b",
+      "m ob",
+      "▁I v",
+      "专 职",
+      "▁ret rieved",
+      "▁retriev ed",
+      "▁retrieve d",
+      "Http Servlet",
+      "个 股",
+      "ze s",
+      "z es",
+      "平方 分米",
+      "▁joint ly",
+      "三 四",
+      "是 家",
+      "Pe t",
+      "P et",
+      "大连 市",
+      "▁S ons",
+      "▁So ns",
+      "▁Son s",
+      "▁T all",
+      "▁Ta ll",
+      "▁Tal l",
+      "▁oper ative",
+      "▁ operative",
+      "▁Cou nsel",
+      "冰 雪",
+      "nB usiness",
+      "n Business",
+      "▁sub types",
+      "▁subt ypes",
+      "▁subtype s",
+      "此 款",
+      "MM ARY",
+      "ra ils",
+      "rail s",
+      "r ails",
+      "og an",
+      "oga n",
+      "o gan",
+      "Route s",
+      "Ro utes",
+      "R outes",
+      "Comp iler",
+      "Compile r",
+      "▁adren al",
+      "Do m",
+      "D om",
+      "▁electro des",
+      "▁electrode s",
+      "▁electr odes",
+      "硬 币",
+      "nd on",
+      "ndo n",
+      "n don",
+      "▁a zure",
+      "▁az ure",
+      "▁ azure",
+      "be am",
+      "▁ambit ious",
+      "en son",
+      "ens on",
+      "e nson",
+      "具体应用 法律",
+      "el lington",
+      "ell ington",
+      "elling ton",
+      "ser ialize",
+      "serial ize",
+      "s erialize",
+      "▁spons ors",
+      "▁sponsor s",
+      "wed ge",
+      "w edge",
+      "▁En cycl",
+      "▁Enc ycl",
+      "向 左",
+      "▁n inth",
+      "▁nin th",
+      "▁ni nth",
+      "▁Lith uan",
+      "用 过",
+      "站 长",
+      "是一 件",
+      "是 一件",
+      "ya rn",
+      "yar n",
+      "y arn",
+      "gi ene",
+      "gie ne",
+      "nR equ",
+      "nRe qu",
+      "n Requ",
+      "▁term inals",
+      "▁terminal s",
+      "▁compl imentary",
+      "▁compliment ary",
+      "▁K un",
+      "▁Ku n",
+      "生活 中的",
+      "生活中 的",
+      "▁f ighter",
+      "▁fight er",
+      "▁con serv",
+      "▁cons erv",
+      "▁prec eding",
+      "▁preced ing",
+      "▁ 余",
+      "▁G ang",
+      "▁Ga ng",
+      "▁Gan g",
+      "▁v ide",
+      "▁vi de",
+      "▁vid e",
+      "▁manif olds",
+      "▁manifold s",
+      "}\\\\ }",
+      "} \\\\}",
+      "▁M Hz",
+      "▁MH z",
+      "▁ MHz",
+      "ua ting",
+      "u ating",
+      "往 下",
+      "都 区",
+      "▁x xx",
+      "▁xx x",
+      "▁ xxx",
+      "▁Har bor",
+      "▁leak age",
+      "▁arch ives",
+      "▁archive s",
+      "▁ archives",
+      "▁ 邓",
+      "一 类",
+      "比 我",
+      "ab l",
+      "a bl",
+      "是我 的",
+      "是 我的",
+      "▁synth ase",
+      "▁associ ates",
+      "▁associate s",
+      "WE B",
+      "W EB",
+      "民政 局",
+      "▁L it",
+      "▁Li t",
+      "免疫 力",
+      "▁L ayer",
+      "▁La yer",
+      "▁Lay er",
+      "▁ Layer",
+      "▁l amin",
+      "▁la min",
+      "▁lam in",
+      "要求 与被告离婚",
+      "urrenc es",
+      "urrence s",
+      "▁select ivity",
+      "▁( %",
+      "▁ (%",
+      "▁M AN",
+      "▁MA N",
+      "▁ MAN",
+      "nS elect",
+      "nSe lect",
+      "n Select",
+      "诚 实",
+      "d ummy",
+      "▁房型 介绍",
+      "委托代理人 李",
+      "唐 山",
+      "文 科",
+      "▁A round",
+      "▁Ar ound",
+      "▁st enosis",
+      "▁merchand ise",
+      "奇 葩",
+      "上海 浦东",
+      "▁ad orable",
+      "Tur n",
+      "T urn",
+      "co rn",
+      "cor n",
+      "c orn",
+      "工业 大学",
+      "ome town",
+      "omet own",
+      "\\' >\\",
+      "\\'> \\",
+      "\\ '>\\",
+      "▁r ape",
+      "▁ra pe",
+      "▁rap e",
+      "▁ rape",
+      "▁But ler",
+      "光 源",
+      "克 尔",
+      "▁Q U",
+      "▁ QU",
+      "am ba",
+      "amb a",
+      "osa ic",
+      "▁v ars",
+      "▁var s",
+      "▁va rs",
+      "▁ vars",
+      "S printf",
+      "的范围内 处分自己的",
+      "nCol lect",
+      "nColl ect",
+      "n Collect",
+      "单位 和",
+      "单 位和",
+      "唯 有",
+      "$ /",
+      "还 挺",
+      "Wr ap",
+      "W rap",
+      "▁de ce",
+      "▁dec e",
+      "▁fif teen",
+      "▁foot print",
+      "▁foo tprint",
+      "无 声",
+      "unc ture",
+      "unct ure",
+      "▁sl opes",
+      "▁slope s",
+      "ic ates",
+      "ica tes",
+      "icate s",
+      "▁catch ing",
+      "▁cat ching",
+      "K G",
+      "ryst all",
+      "rystal l",
+      "rys tall",
+      "MESS AGE",
+      "▁Res pon",
+      "▁ Respon",
+      "▁f ictional",
+      "▁fiction al",
+      "追 逐",
+      "风格 的",
+      "风 格的",
+      "os por",
+      "osp or",
+      "nEx per",
+      "n Exper",
+      "至 上",
+      "fa n",
+      "f an",
+      "gr av",
+      "gra v",
+      "g rav",
+      "请你 根据",
+      "mi le",
+      "mil e",
+      "m ile",
+      "刑事 责任",
+      "余 弦",
+      "▁pe aked",
+      "▁peak ed",
+      "re ported",
+      "rep orted",
+      "report ed",
+      "在 自己",
+      "▁C ord",
+      "▁Co rd",
+      "▁Cor d",
+      "▁b uzz",
+      "▁bu zz",
+      "▁C atalog",
+      "▁Catal og",
+      "▁ Catalog",
+      "oe lectric",
+      "oelect ric",
+      "o electric",
+      "▁sl ave",
+      "▁ slave",
+      "▁stat ue",
+      "▁pal ette",
+      "▁pa lette",
+      "▁pale tte",
+      "▁ palette",
+      "▁mark down",
+      "▁ markdown",
+      "▁manip ulate",
+      "hs pace",
+      "h space",
+      "▁rem inds",
+      "▁remind s",
+      "▁remin ds",
+      "叙 述",
+      "al ways",
+      "scrib ed",
+      "scri bed",
+      "scribe d",
+      "s cribed",
+      "▁dissemin ation",
+      "到 一个",
+      "met al",
+      "me tal",
+      "meta l",
+      "m etal",
+      "vs code",
+      "v scode",
+      "dis cord",
+      "disc ord",
+      "▁ 이",
+      "Do t",
+      "D ot",
+      "的 生命",
+      "of lu",
+      "o flu",
+      "▁d rone",
+      "▁dr one",
+      "▁dro ne",
+      "第十 六",
+      "第 十六",
+      "▁class ifier",
+      "▁Exper iments",
+      "▁Experiment s",
+      "碎 片",
+      "A H",
+      "▁s unny",
+      "▁sun ny",
+      "▁in vert",
+      "▁inv ert",
+      "勤 奋",
+      "水 费",
+      "▁exc retion",
+      "▁process ors",
+      "▁processor s",
+      "▁ processors",
+      "na d",
+      "n ad",
+      "支 气管",
+      "▁h ike",
+      "▁hi ke",
+      "热 搜",
+      "▁success or",
+      "什么 都",
+      "ble r",
+      "bl er",
+      "b ler",
+      "▁perform ers",
+      "▁perfor mers",
+      "▁performer s",
+      "st rand",
+      "str and",
+      "stra nd",
+      "▁m olar",
+      "▁mo lar",
+      "▁mol ar",
+      "str ftime",
+      "抬 头",
+      "警 示",
+      "▁L ucas",
+      "▁Luc as",
+      "▁Lu cas",
+      "▁ment oring",
+      "▁mentor ing",
+      "L B",
+      "常 住",
+      "▁K ill",
+      "▁Ki ll",
+      "▁Kil l",
+      "在 刑罚",
+      "的心 情",
+      "的 心情",
+      "▁an tic",
+      "▁ant ic",
+      "▁anti c",
+      "▁ antic",
+      "d g",
+      "▁sh ake",
+      "▁sha ke",
+      "▁ shake",
+      "iling ual",
+      "▁hospital ity",
+      "▁hospit ality",
+      "▁ 肖",
+      "国 人",
+      "主人 公",
+      "都 配有",
+      "正式 批准",
+      "Reg istration",
+      "以 太",
+      "有 好",
+      "锦 标",
+      "胜利 的",
+      "胜 利的",
+      "∞ ）",
+      "清 算",
+      "nT our",
+      "nTo ur",
+      "n Tour",
+      "▁Nich olas",
+      "▁const itutional",
+      "▁constitution al",
+      "军 的",
+      "TR A",
+      "T RA",
+      "监 视",
+      "选择 一个",
+      "▁f eat",
+      "▁fe at",
+      "▁ feat",
+      "中心 的",
+      "中 心的",
+      "ufeff namespace",
+      "oe ing",
+      "▁p acks",
+      "▁pack s",
+      "▁pa cks",
+      "▁pac ks",
+      "inter pret",
+      "使 人",
+      "贴 近",
+      "犯罪 的",
+      "▁DE AL",
+      "sc atter",
+      "▁sur geons",
+      "▁surge ons",
+      "▁surgeon s",
+      "▁aug mented",
+      "▁augment ed",
+      "专业 化",
+      "▁R achel",
+      "▁Ra chel",
+      "▁Rac hel",
+      "▁Ro of",
+      "い る",
+      "▁cap illary",
+      "副 市长",
+      "▁L ub",
+      "▁Lu b",
+      "Calc ul",
+      "C alcul",
+      "履行 生效法律文书",
+      "▁Re solution",
+      "▁Res olution",
+      "▁ Resolution",
+      "▁parliament ary",
+      "何 必",
+      "比较 大",
+      "比 较大",
+      "▁Hop kins",
+      "无正当理由拒不到庭 的",
+      "ve re",
+      "ver e",
+      "v ere",
+      "▁Con test",
+      "▁Cont est",
+      "▁Co ntest",
+      "▁group ed",
+      "▁Init iative",
+      "入 户",
+      "木 质",
+      "风 采",
+      "第二 条",
+      "第 二条",
+      "技术 有限公司",
+      "▁Prov ide",
+      "求解 即可",
+      "▁ret ina",
+      "▁retin a",
+      "▁wh olesale",
+      "▁whole sale",
+      "▁wholes ale",
+      "Enum erable",
+      "▁v m",
+      "▁ vm",
+      "去 医院",
+      "附近 的",
+      "it ely",
+      "ite ly",
+      "nset up",
+      "n setup",
+      "▁P ATH",
+      "▁PA TH",
+      "▁PAT H",
+      "▁ PATH",
+      "▁sc ary",
+      "▁scar y",
+      "还有 就是",
+      "▁F ill",
+      "▁Fil l",
+      "▁Fi ll",
+      "▁ Fill",
+      "td efer",
+      "tdef er",
+      "t defer",
+      "▁grand father",
+      "抑郁 症",
+      "▁for ty",
+      "▁fort y",
+      "之日起 五日内",
+      "NE R",
+      "N ER",
+      "nO b",
+      "n Ob",
+      "▁mod alities",
+      "▁modal ities",
+      "色 谱",
+      "管 的",
+      "重 症",
+      "ic ers",
+      "ice rs",
+      "icer s",
+      "i cers",
+      "汇 聚",
+      "(\\' %",
+      "( \\'%",
+      "▁conf ined",
+      "▁con tested",
+      "▁cont ested",
+      "▁contest ed",
+      "▁estim ator",
+      "但被执行人 至今未履行",
+      "不 相",
+      "疑 似",
+      "wall et",
+      "wal let",
+      "abil istic",
+      "▁read ings",
+      "▁reading s",
+      "▁enzym atic",
+      "▁visual ize",
+      "IC A",
+      "I CA",
+      "▁T imer",
+      "▁Time r",
+      "▁Tim er",
+      "▁Ti mer",
+      "▁ Timer",
+      "▁hass le",
+      "nSh ort",
+      "n Short",
+      "transfer ase",
+      "还 算",
+      "▁def icient",
+      "▁ deficient",
+      "白 银",
+      "银行 账户",
+      "▁D istance",
+      "▁Dist ance",
+      "▁Di stance",
+      "▁ Distance",
+      "ch et",
+      "che t",
+      "c het",
+      "tic ket",
+      "tick et",
+      "t icket",
+      "▁Rail road",
+      "悠 久",
+      "月 息",
+      "自然 数",
+      "▁P lain",
+      "▁Pl ain",
+      "▁ Plain",
+      "▁K erala",
+      "▁Ker ala",
+      "word press",
+      "▁facilit ates",
+      "▁facilitate s",
+      "Pe n",
+      "P en",
+      "物业 公司",
+      "ic aid",
+      "ica id",
+      "っ た",
+      "换 成",
+      "▁Fl ask",
+      "▁disc rep",
+      "▁discre p",
+      "▁no where",
+      "▁now here",
+      "▁thr illed",
+      "▁thrill ed",
+      "翡 翠",
+      "静 静",
+      "一个 整数",
+      "▁D anny",
+      "▁Dan ny",
+      "的 本",
+      "衣 物",
+      "打印 机",
+      "▁i od",
+      "▁io d",
+      "▁ iod",
+      "传 送",
+      "贝壳 找房",
+      "▁NOT ICE",
+      "ä t",
+      "st ed",
+      "ste d",
+      "s ted",
+      "▁英 文",
+      "▁ 英文",
+      "▁cap tures",
+      "▁capt ures",
+      "▁capture s",
+      "关于 多学科",
+      "idel ity",
+      "关于多学科 知识的选择题",
+      "伯 特",
+      "▁tax p",
+      "▁ta xp",
+      "out heastern",
+      "outh eastern",
+      "outheast ern",
+      "li fy",
+      "lif y",
+      "l ify",
+      "▁P arts",
+      "▁Par ts",
+      "▁Part s",
+      "▁ Parts",
+      "流 体",
+      "您 需要",
+      "等 特点",
+      "▁j am",
+      "▁ja m",
+      "▁ jam",
+      "▁battle s",
+      "▁batt les",
+      "广 电",
+      "不合 格",
+      "不 合格",
+      "学生 在",
+      "ab lo",
+      "abl o",
+      "a blo",
+      "nA lex",
+      "nAl ex",
+      "n Alex",
+      "▁Har vey",
+      "Mod ified",
+      "▁vag inal",
+      "判处有期徒刑 一年",
+      "最高人民法院关于 执行",
+      "感 叹",
+      "行驶 至",
+      "▁d ispens",
+      "▁disp ens",
+      "▁quarter back",
+      "人们 对",
+      "▁M ental",
+      "▁Men tal",
+      "▁Ment al",
+      "当我 们",
+      "当 我们",
+      "▁b ats",
+      "▁bat s",
+      "▁ba ts",
+      "▁St ories",
+      "▁S hot",
+      "▁Sh ot",
+      "▁ Shot",
+      "█ █",
+      "cin e",
+      "ci ne",
+      "c ine",
+      "▁餐馆 简介",
+      "▁Trans action",
+      "▁ Transaction",
+      "▁illustr ates",
+      "▁illust rates",
+      "▁illustrate s",
+      "Foo ter",
+      "F ooter",
+      "nHon ours",
+      "会 是",
+      "▁D ot",
+      "▁Do t",
+      "▁ Dot",
+      "▁S ed",
+      "▁Se d",
+      "▁S pa",
+      "▁Sp a",
+      "▁G ABA",
+      "▁GA BA",
+      "▁ 卢",
+      "nA ir",
+      "n Air",
+      "ake spe",
+      "akes pe",
+      "▁ 再",
+      "古 今",
+      "折 磨",
+      "ex us",
+      "否 决",
+      "de al",
+      "dea l",
+      "d eal",
+      "nM al",
+      "nMa l",
+      "n Mal",
+      "解决 问题的",
+      "解决问题 的",
+      "穿 过",
+      "关系 式",
+      "可 作为",
+      "▁Ar ctic",
+      "▁Arc tic",
+      "和 的",
+      "编 著",
+      "延 期",
+      "▁d ire",
+      "▁di re",
+      "▁dir e",
+      "▁hist opath",
+      "eth ylene",
+      "ethyl ene",
+      "▁Develop ers",
+      "▁Developer s",
+      "{ $",
+      "nS em",
+      "nSe m",
+      "n Sem",
+      "▁h ay",
+      "▁ha y",
+      "▁ hay",
+      "np y",
+      "n py",
+      "▁L ed",
+      "▁Le d",
+      "▁ Led",
+      "以下 这道",
+      "▁prison ers",
+      "▁prisoner s",
+      "下面 这道选择题",
+      "▁ 或",
+      "bel ongs",
+      "▁De scribe",
+      "▁Des cribe",
+      "▁ Describe",
+      "预 告",
+      "部 部长",
+      "健康 发展",
+      "由 上海市",
+      "▁ 秦",
+      "保 暖",
+      "战 国",
+      "En ergy",
+      "E nergy",
+      "aa S",
+      "dev ices",
+      "device s",
+      "▁cor pus",
+      "▁ corpus",
+      "▁ab lation",
+      "▁stress ful",
+      "▁renov ation",
+      "\\\\ _{",
+      "\\\\_ {",
+      "▁her bs",
+      "▁herb s",
+      "▁met eor",
+      "▁Cal gary",
+      "滚 动",
+      "Dest roy",
+      "俗 称",
+      "三十 一日",
+      "三 十一日",
+      "▁Des ktop",
+      "▁Desk top",
+      "▁ Desktop",
+      "▁cor neal",
+      "▁corn eal",
+      "案件 的",
+      "案 件的",
+      "▁disturb ances",
+      "▁disturbance s",
+      "▁ è",
+      "Pa c",
+      "P ac",
+      "ung le",
+      "▁e val",
+      "▁ev al",
+      "▁ eval",
+      "▁C anyon",
+      "▁Can yon",
+      "▁cl icked",
+      "▁click ed",
+      "▁ clicked",
+      "eo n",
+      "e on",
+      "▁l int",
+      "▁li nt",
+      "▁lin t",
+      "▁ lint",
+      "▁neg lect",
+      "村民 小组",
+      "▁coord inator",
+      "▁R angers",
+      "▁Range rs",
+      "▁Ran gers",
+      "▁D ad",
+      "▁Da d",
+      "▁king dom",
+      "丙 烯",
+      "选 题",
+      "mo logy",
+      "mol ogy",
+      "m ology",
+      "沥 青",
+      "肾 脏",
+      "ou ted",
+      "out ed",
+      "oute d",
+      "o uted",
+      "路由 器",
+      "▁Interesting ly",
+      "▁Interest ingly",
+      "综合运用 多",
+      "Sur face",
+      "S urface",
+      "的 公",
+      "并从 给出的选项中",
+      "▁Program me",
+      "下 半",
+      "年 参加",
+      "▁p ier",
+      "▁pi er",
+      "▁pie r",
+      "▁premier ed",
+      "▁premiere d",
+      "}` );\\",
+      "} `);\\",
+      "▁ 网络",
+      "共同 建设",
+      "ct ype",
+      "c type",
+      "被执行人 有",
+      "管 委会",
+      "re new",
+      "ren ew",
+      "rene w",
+      "▁f ond",
+      "▁fo nd",
+      "Asp NetCore",
+      "yr o",
+      "y ro",
+      "(` ${",
+      "( `${",
+      "▁w ounded",
+      "▁wound ed",
+      "单 个",
+      "du plicate",
+      "d uplicate",
+      "清 爽",
+      "▁一 个",
+      "▁ 一个",
+      "ear th",
+      "e arth",
+      "au rus",
+      "aur us",
+      "a urus",
+      "熟悉 的",
+      "▁landscape s",
+      "▁landsc apes",
+      "nu rl",
+      "nur l",
+      "n url",
+      "附带 民事",
+      "ang ered",
+      "ange red",
+      "anger ed",
+      "VI S",
+      "V IS",
+      "是 需要",
+      "NO W",
+      "N OW",
+      "请你 仔细",
+      "ges terone",
+      "oc ene",
+      "t throw",
+      "N ational",
+      "良 心",
+      "自动 撤回上诉处理",
+      "专 著",
+      "ol is",
+      "oli s",
+      "o lis",
+      "** **\\",
+      "**** \\",
+      "*** *\\",
+      "* ***\\",
+      "▁内容 介绍",
+      "▁ 内容介绍",
+      "低 价",
+      "毒 素",
+      "▁URL s",
+      "▁s era",
+      "▁se ra",
+      "▁ser a",
+      "▁house d",
+      "▁hous ed",
+      "▁ho used",
+      "▁l ining",
+      "▁li ning",
+      "▁lin ing",
+      "▁ lining",
+      "北 美",
+      "行政 复议",
+      "▁Al ger",
+      "Per cent",
+      "产 妇",
+      "nIS BN",
+      "n ISBN",
+      "用 科学记数法",
+      "th ro",
+      "thr o",
+      "▁Pop ular",
+      "传 真",
+      "▁m emo",
+      "▁me mo",
+      "▁mem o",
+      "▁ memo",
+      "▁Pl ans",
+      "▁Plan s",
+      "酒店的 房型",
+      "man uel",
+      "ophys ical",
+      "o physical",
+      "陈某 某",
+      "陈 某某",
+      "了 就",
+      "将 近",
+      "▁ 表示",
+      "为一 体",
+      "为 一体",
+      "为您 提供",
+      "变化 的",
+      "变 化的",
+      "征 集",
+      "▁B anks",
+      "▁Bank s",
+      "▁Ban ks",
+      "樱 花",
+      "▁B og",
+      "▁Bo g",
+      "差异有 统计学意义",
+      "▁respons iveness",
+      "▁responsive ness",
+      "lic h",
+      "li ch",
+      "l ich",
+      "▁Type Error",
+      "▁ TypeError",
+      "不 禁",
+      "ou cher",
+      "ouch er",
+      "▁G D",
+      "▁ GD",
+      "▁w ages",
+      "▁wa ges",
+      "▁wage s",
+      "▁wag es",
+      "多 位",
+      "完全 平方",
+      "▁par al",
+      "▁para l",
+      "▁pa ral",
+      "制品 有限公司",
+      "▁Dop pler",
+      "▁' \\\\",
+      "▁'\\ \\",
+      "▁ '\\\\",
+      "▁c ass",
+      "▁ca ss",
+      "▁cas s",
+      "▁ cass",
+      "▁to uring",
+      "▁tour ing",
+      "▁tou ring",
+      "▁wrest ling",
+      "▁Hamilton ian",
+      "新 中国",
+      "获得 的",
+      "获 得的",
+      "▁cur r",
+      "▁cu rr",
+      "▁ curr",
+      "▁Croat ia",
+      "▁threat ening",
+      "▁threaten ing",
+      "▁ threatening",
+      "nSt ill",
+      "▁sacr ific",
+      "Wal l",
+      "W all",
+      "▁N it",
+      "▁Ni t",
+      "多 大",
+      "LC J",
+      "yl um",
+      "文化 传播",
+      "▁K arn",
+      "▁Kar n",
+      "▁Ka rn",
+      "▁b arn",
+      "▁bar n",
+      "▁ba rn",
+      "nH ouses",
+      "nHouse s",
+      "木 耳",
+      "▁ax ial",
+      "▁axi al",
+      "▁ axial",
+      "原 名",
+      "CL C",
+      "C LC",
+      "el ia",
+      "eli a",
+      "e lia",
+      "▁Vol unte",
+      "千 金",
+      "敏 捷",
+      "订 阅",
+      "ar u",
+      "a ru",
+      "今后 的",
+      "今 后的",
+      "终结 本案",
+      "bi shop",
+      "bis hop",
+      "bish op",
+      "b ishop",
+      "▁st eal",
+      "▁ste al",
+      "颈 椎",
+      "In clude",
+      "没 钱",
+      "▁f ost",
+      "▁fo st",
+      "▁ax ios",
+      "▁axi os",
+      "▁ axios",
+      "借 口",
+      "um ab",
+      "uma b",
+      "杀 了",
+      "▁K as",
+      "▁Ka s",
+      "逾期 利息",
+      "几个 选项",
+      "▁a erial",
+      "▁aer ial",
+      "▁pick le",
+      "▁pic kle",
+      "▁ pickle",
+      "Me n",
+      "M en",
+      "ph inx",
+      "phi nx",
+      "▁me tro",
+      "▁met ro",
+      "▁su perf",
+      "▁super f",
+      "{ →",
+      "п о",
+      "应 按",
+      ">) \\",
+      "> )\\",
+      "SI G",
+      "S IG",
+      "县 支行",
+      "因式 分解",
+      "▁C ells",
+      "▁Cell s",
+      "▁Cel ls",
+      "▁ Cells",
+      "▁has attr",
+      "▁G DP",
+      "▁GD P",
+      "▁ GDP",
+      "▁ab ort",
+      "▁ abort",
+      "波 兰",
+      "un gal",
+      "ung al",
+      "▁Em ily",
+      "镜 子",
+      "的 传统",
+      "的心 态",
+      "的 心态",
+      "ji ang",
+      "jian g",
+      "jia ng",
+      "j iang",
+      "▁ep och",
+      "▁ epoch",
+      "▁tact ics",
+      "nm e",
+      "n me",
+      "找 工作",
+      "Tr a",
+      "T ra",
+      "nB ase",
+      "nBas e",
+      "n Base",
+      "把 手",
+      "▁D ialog",
+      "▁Dial og",
+      "▁ Dialog",
+      "▁be ings",
+      "▁being s",
+      "签 下",
+      "色 泽",
+      "▁R EF",
+      "▁RE F",
+      "▁ REF",
+      "建设 项目",
+      "登记 信息",
+      "er ton",
+      "ert on",
+      "erto n",
+      "年经 教育部",
+      "▁ph rases",
+      "▁phrase s",
+      "▁de pressive",
+      "▁dep ressive",
+      "泉 水",
+      "离开 了",
+      "离 开了",
+      "br ed",
+      "bre d",
+      "b red",
+      "▁M oz",
+      "▁Mo z",
+      "▁sub space",
+      "▁subs pace",
+      "commun ications",
+      "communication s",
+      "▁author ization",
+      "▁ authorization",
+      "分钟 后",
+      "ID TH",
+      "▁c n",
+      "▁ cn",
+      "第十 一",
+      "第 十一",
+      "▁S ugar",
+      "▁Su gar",
+      "▁witness ed",
+      "FR ING",
+      "snap shot",
+      "s napshot",
+      "车 道",
+      "oth ic",
+      "额外 的",
+      "额 外的",
+      "In ternet",
+      "Inter net",
+      "im an",
+      "ima n",
+      "i man",
+      "road cast",
+      "右 手",
+      "▁c oping",
+      "▁co ping",
+      "▁cop ing",
+      "▁E ducational",
+      "▁Education al",
+      "▁Educ ational",
+      "同 心",
+      "Sl ot",
+      "S lot",
+      "REMO VED",
+      "▁Part ner",
+      "材料 有限公司",
+      "▁pra ised",
+      "▁praise d",
+      "因 被告",
+      "▁h ug",
+      "▁hu g",
+      "醉酒 驾驶机动车",
+      "置 业",
+      "ner gic",
+      "nerg ic",
+      "n ergic",
+      "▁transl ations",
+      "▁translation s",
+      "▁ translations",
+      "届 中国",
+      "nS tand",
+      "nSt and",
+      "nSta nd",
+      "n Stand",
+      "▁reward ing",
+      "▁retro spect",
+      "蓬 勃",
+      "新 农村",
+      "VID IA",
+      "cu be",
+      "cub e",
+      "c ube",
+      "▁sacr ifice",
+      "▁sacrific e",
+      "ern ate",
+      "中 奖",
+      "电 竞",
+      "PA CK",
+      "P ACK",
+      "端 正",
+      "的商 品",
+      "的 商品",
+      "▁k icked",
+      "▁kick ed",
+      "中 市",
+      "串 行",
+      "az er",
+      "aze r",
+      "a zer",
+      "▁ex cel",
+      "▁exc el",
+      "▁ excel",
+      "elastic search",
+      "了 点",
+      "不 建议",
+      "De velop",
+      "茅 台",
+      "]) ;",
+      "] );",
+      "Name d",
+      "Na med",
+      "N amed",
+      "▁craft s",
+      "▁cra fts",
+      "反 之",
+      "prec ision",
+      "p recision",
+      "▁work around",
+      "▁product ions",
+      "▁production s",
+      "前 辈",
+      "Fac e",
+      "Fa ce",
+      "F ace",
+      "Ph il",
+      "Phi l",
+      "▁home made",
+      "▁hom emade",
+      "▁medicine s",
+      "▁medic ines",
+      "Pa ne",
+      "Pan e",
+      "P ane",
+      "ur as",
+      "ura s",
+      "u ras",
+      "可 通过",
+      "群 岛",
+      "▁S old",
+      "▁So ld",
+      "▁Sol d",
+      "博 会",
+      "▁consult ants",
+      "▁consultant s",
+      "南昌 市",
+      "这个 城市",
+      "血 脂",
+      "▁v ault",
+      "▁va ult",
+      "▁ vault",
+      "系 的",
+      "Met ric",
+      "M etric",
+      "湾 区",
+      "time zone",
+      "在使用 现有的床铺",
+      "无 语",
+      "此 处",
+      "ID s",
+      "I Ds",
+      "▁房间 数量",
+      "▁S uit",
+      "▁Su it",
+      "no bject",
+      "nob ject",
+      "n object",
+      "Ma ss",
+      "Mas s",
+      "M ass",
+      "Ph ot",
+      "P hot",
+      "calc ul",
+      "c alcul",
+      "fl ower",
+      "flow er",
+      "flo wer",
+      "f lower",
+      "▁all iance",
+      "中国 大陆",
+      "mat ter",
+      "m atter",
+      "Fe atures",
+      "Feature s",
+      "▁a symmetric",
+      "▁as ymmetric",
+      "▁asym metric",
+      "▁asymmet ric",
+      "字 段",
+      "em ployee",
+      "▁tert iary",
+      "ag em",
+      "age m",
+      "a gem",
+      "▁ 雷",
+      "给 您",
+      "▁conv ince",
+      "▁convinc e",
+      "TR UE",
+      "▁Hind u",
+      "▁muc osa",
+      "蚌 埠",
+      "▁第 四十",
+      "▁ 第四十",
+      "▁S tein",
+      "▁St ein",
+      "▁Ste in",
+      "ñ o",
+      "▁ 超",
+      "▁l s",
+      "▁ ls",
+      "ua ng",
+      "uan g",
+      "u ang",
+      "man aged",
+      "manage d",
+      "▁Ar nold",
+      "能 被",
+      "该 院",
+      "▁e lic",
+      "▁el ic",
+      "▁ elic",
+      "▁v egan",
+      "▁ve gan",
+      "▁inter mitt",
+      "ote chnology",
+      "o technology",
+      "DA O",
+      "D AO",
+      "de ad",
+      "dea d",
+      "d ead",
+      "ro phys",
+      "roph ys",
+      "rophy s",
+      "r ophys",
+      "▁g p",
+      "▁ gp",
+      "有 帮助",
+      "al la",
+      "all a",
+      "a lla",
+      "ip ur",
+      "i pur",
+      "寻 常",
+      "就 此",
+      "愉 悦",
+      "ic ester",
+      "ice ster",
+      "ices ter",
+      "i cester",
+      "▁Ep isode",
+      "Data Frame",
+      "对角 线",
+      "给他 们",
+      "给 他们",
+      "▁d ice",
+      "▁di ce",
+      "▁dic e",
+      "ant ine",
+      "anti ne",
+      "ph rine",
+      "▁W inds",
+      "▁Win ds",
+      "▁Wind s",
+      "friend s",
+      "f riends",
+      "▁use fulness",
+      "▁useful ness",
+      "糯 米",
+      "间 隙",
+      "~\\ \\",
+      "~ \\\\",
+      "申请执行人 张",
+      "▁ τ",
+      "▁ 创建",
+      "▁s po",
+      "▁sp o",
+      "▁Qt Gui",
+      "▁ QtGui",
+      "卵 巢",
+      "重 重",
+      "▁D y",
+      "▁ Dy",
+      "双方 的",
+      "双 方的",
+      "▁Sh ore",
+      "第十 章",
+      "第 十章",
+      "▁sc ream",
+      "发 回",
+      "情况 的",
+      "程序 合法",
+      "▁P ars",
+      "▁Par s",
+      "▁Pa rs",
+      "▁ Pars",
+      "▁br ains",
+      "▁brain s",
+      "▁bra ins",
+      "▁ brains",
+      "▁Hop efully",
+      "▁Hope fully",
+      "re r",
+      "r er",
+      "▁j unk",
+      "▁ju nk",
+      "▁jun k",
+      "▁M atter",
+      "▁Mat ter",
+      "▁Matt er",
+      "▁where by",
+      "到底 是",
+      "查询 被执行人",
+      "nPer haps",
+      "习 性",
+      "渔 业",
+      "▁s ep",
+      "▁se p",
+      "▁ sep",
+      "▁ass isting",
+      "▁assist ing",
+      "剧 场",
+      "wi d",
+      "w id",
+      "▁F ALSE",
+      "▁ FALSE",
+      "ens itivity",
+      "ensit ivity",
+      "({ \\'",
+      "({\\ '",
+      "( {\\'",
+      "▁B iden",
+      "▁Bi den",
+      "▁Bid en",
+      "FRING EMENT",
+      "W ar",
+      "并 支付",
+      "ut ton",
+      "utt on",
+      "类 网络小说",
+      "▁pel vic",
+      "▁C atherine",
+      "政 权",
+      "▁文 件",
+      "▁ 文件",
+      "AS SERT",
+      "ASS ERT",
+      "nCom put",
+      "nComp ut",
+      "n Comput",
+      "深刻 的",
+      "的 平台",
+      "pa in",
+      "p ain",
+      "▁re plicate",
+      "▁replic ate",
+      "▁replica te",
+      "从选项中 选出你认为正确的答案",
+      "从选项中选出 你认为正确的答案",
+      "▁v om",
+      "▁vo m",
+      "不同意 离婚",
+      "合理 的解释",
+      "合理的 解释",
+      "照 射",
+      "Al so",
+      "故 本",
+      "tt es",
+      "tte s",
+      "t tes",
+      "▁m int",
+      "▁min t",
+      "▁mi nt",
+      "▁journal ism",
+      "修 行",
+      "千 古",
+      "需要 对",
+      "需 要对",
+      "as say",
+      "ass ay",
+      "assa y",
+      "▁c ouch",
+      "▁co uch",
+      "▁cou ch",
+      "CO RE",
+      "COR E",
+      "C ORE",
+      "▁Init ialize",
+      "▁Initial ize",
+      "▁ Initialize",
+      "宁 静",
+      "工业 园",
+      "质量 的",
+      "质 量的",
+      "▁low ering",
+      "▁lower ing",
+      "板 材",
+      "病 症",
+      "贵 阳",
+      "▁g room",
+      "▁gr oom",
+      "▁gro om",
+      "▁Chap el",
+      "▁Cha pel",
+      "exec ution",
+      "明 了",
+      "财务 管理",
+      "▁C ongo",
+      "▁Con go",
+      "▁Co ngo",
+      "▁Cong o",
+      "def icient",
+      "▁G ym",
+      "▁Gy m",
+      "现 行",
+      "▁n oble",
+      "▁no ble",
+      "▁nob le",
+      "▁li cence",
+      "▁lic ence",
+      "▁emphas ize",
+      "万 年",
+      "▁D ual",
+      "▁Du al",
+      "op ausal",
+      "opa usal",
+      "▁exhib itions",
+      "▁exhibit ions",
+      "▁exhibition s",
+      "总 共",
+      "回答 是",
+      "▁touch ing",
+      "▁tou ching",
+      "▁ 袁",
+      "tw idth",
+      "t width",
+      "ash ington",
+      "ashing ton",
+      "该 方法",
+      "▁AUTH ORS",
+      "功 耗",
+      "烟 草",
+      "gp u",
+      "g pu",
+      "五个 月",
+      "五 个月",
+      "邮政储蓄 银行股份有限公司",
+      "加 固",
+      "▁F emale",
+      "▁Fe male",
+      "▁Fem ale",
+      "stere otype",
+      "▁energ etic",
+      "▁energet ic",
+      "换 个",
+      "有效 率",
+      "有 效率",
+      "这才 是",
+      "这 才是",
+      "uss ions",
+      "ussion s",
+      "li ga",
+      "lig a",
+      "l iga",
+      "▁ab elian",
+      "end region",
+      "第一 名",
+      "第 一名",
+      "级别 的",
+      "级 别的",
+      "run ch",
+      "ru nch",
+      "r unch",
+      "州市 人民法院",
+      "州 市人民法院",
+      "物业服务 合同纠纷一案",
+      "二 战",
+      "却 不",
+      "被 誉为",
+      "Wal k",
+      "W alk",
+      "nw as",
+      "n was",
+      "▁M old",
+      "▁Mo ld",
+      "▁Mol d",
+      "▁dispar ities",
+      "夏 日",
+      "扫 码",
+      "▁Mer cedes",
+      "▁Merc edes",
+      "也 算",
+      "福 田",
+      "un ted",
+      "unt ed",
+      "unte d",
+      "▁tom atoes",
+      "▁tomat oes",
+      "▁tomato es",
+      "▁f ame",
+      "▁fam e",
+      "▁fa me",
+      "在 做",
+      "▁R anch",
+      "▁Ra nch",
+      "▁Ran ch",
+      "喜 悦",
+      "▁下 列",
+      "▁ 下列",
+      "发现 自己",
+      "hi ft",
+      "h ift",
+      "▁up s",
+      "▁u ps",
+      "▁ ups",
+      "第一百零 七条",
+      "▁E cuador",
+      "E U",
+      "nt e",
+      "n te",
+      "咨询 有限公司",
+      "准备 好",
+      "全 场",
+      "nd a",
+      "n da",
+      "并 进行",
+      "no pen",
+      "nop en",
+      "n open",
+      "DO WN",
+      "we ets",
+      "weet s",
+      "MOD ULE",
+      "oid es",
+      "oi des",
+      "o ides",
+      "▁sub urb",
+      "▁acc identally",
+      "▁accident ally",
+      "▁accidental ly",
+      "ic io",
+      "ici o",
+      "L G",
+      "发表 论文",
+      "梦 中",
+      "al ore",
+      "alo re",
+      "芒 果",
+      "leg iate",
+      "▁D rosophila",
+      "▁detect able",
+      "▁se lections",
+      "▁select ions",
+      "▁selection s",
+      "▁sel ections",
+      "居住 地",
+      "nCh rist",
+      "nChris t",
+      "n Christ",
+      "城 的",
+      "米 兰",
+      "▁p ag",
+      "▁pa g",
+      "▁ pag",
+      "tt ext",
+      "tte xt",
+      "t text",
+      "我 一",
+      "nN umber",
+      "n Number",
+      "▁theore ms",
+      "▁theorem s",
+      "▁theor ems",
+      "re nergic",
+      "ren ergic",
+      "_ %",
+      "mp h",
+      "m ph",
+      "▁xml ns",
+      "▁ xmlns",
+      "在 内",
+      "在 区间",
+      "扬 声器",
+      "▁y early",
+      "▁year ly",
+      "▁ye arly",
+      "▁cross es",
+      "株 洲",
+      "深 夜",
+      "▁re use",
+      "▁mention s",
+      "▁ment ions",
+      "忍 受",
+      "电影 作品",
+      "▁tight ly",
+      "▁tr acked",
+      "▁track ed",
+      "跨 国",
+      "结构 住房",
+      "▁Rh ode",
+      "▁Rho de",
+      "有多种 选择",
+      "gl ass",
+      "glas s",
+      "g lass",
+      "▁b utt",
+      "▁but t",
+      "▁bu tt",
+      "nM ount",
+      "n Mount",
+      "▁sett lements",
+      "▁settlement s",
+      "▁settle ments",
+      "d raft",
+      "uk ary",
+      "uka ry",
+      "DA Y",
+      "D AY",
+      "▁i c",
+      "▁ ic",
+      "an ton",
+      "ant on",
+      "anto n",
+      "a nton",
+      "ol v",
+      "o lv",
+      "的重 点",
+      "的 重点",
+      "处 方",
+      "长 途",
+      "nN eed",
+      "nNe ed",
+      "n Need",
+      "▁Reyn olds",
+      "▁Ch ef",
+      "▁Che f",
+      "万 千",
+      "un o",
+      "u no",
+      "Au gust",
+      "Aug ust",
+      "A ugust",
+      "▁ ’",
+      "▁us eless",
+      "▁use less",
+      "徘 徊",
+      "的 颜色",
+      "Un less",
+      "七 十",
+      "▁ap ples",
+      "▁app les",
+      "▁appl es",
+      "▁apple s",
+      "第一百四十 四条",
+      "第一百四十四 条",
+      "▁home page",
+      "▁hom epage",
+      "▁ homepage",
+      "能 为",
+      "ns ic",
+      "▁Fil ip",
+      "▁Fi lip",
+      "▁aure us",
+      "▁aur eus",
+      "▁pur ification",
+      "▁circum ference",
+      "具有 一定的",
+      "具有一定 的",
+      "具 有一定的",
+      "▁M ig",
+      "▁Mi g",
+      "▁m emor",
+      "▁me mor",
+      "▁mem or",
+      "▁memo r",
+      "▁Through out",
+      "缺 血",
+      "▁diagn ose",
+      "▁pass words",
+      "▁password s",
+      "▁con sequently",
+      "▁consequ ently",
+      "▁cl earing",
+      "▁cle aring",
+      "▁clear ing",
+      "工作 效率",
+      "▁In clude",
+      "▁ Include",
+      "▁cr icketer",
+      "▁cricket er",
+      "化 肥",
+      "许多 人",
+      "许 多人",
+      "Jul y",
+      "J uly",
+      "pop ulation",
+      "莫名 其",
+      "▁trans c",
+      "▁tra nsc",
+      "▁tran sc",
+      "万 余",
+      "就是 个",
+      "就 是个",
+      "法 执字第",
+      "▁beet le",
+      "▁bee tle",
+      "▁spec ifying",
+      "▁specify ing",
+      "▁T hom",
+      "▁Th om",
+      "List en",
+      "Li sten",
+      "ho ng",
+      "hon g",
+      "h ong",
+      "▁Sh ared",
+      "▁Share d",
+      "▁Sha red",
+      "▁Shar ed",
+      "▁ Shared",
+      "Dash board",
+      "D ashboard",
+      "▁should ers",
+      "▁shoulder s",
+      "▁Indust ries",
+      "结 点",
+      "和解 题",
+      "和 解题",
+      "医 护",
+      "ops ies",
+      "nPro fessional",
+      "之 星",
+      "店 的",
+      "推荐 过",
+      "la bs",
+      "lab s",
+      "l abs",
+      "nf ile",
+      "nfi le",
+      "nfil e",
+      "n file",
+      "Se gment",
+      "Seg ment",
+      "还有 一些",
+      "还 有一些",
+      "Re ply",
+      "Rep ly",
+      "跑 了",
+      "轴 上",
+      "服务 于",
+      "▁hy giene",
+      "总 投资",
+      "犯 抢劫罪",
+      "▁side w",
+      "▁sid ew",
+      "omon as",
+      "omo nas",
+      "▁C oron",
+      "▁Co ron",
+      "▁Cor on",
+      "be aring",
+      "bear ing",
+      "b earing",
+      "时 段",
+      "省 内",
+      "▁gl ory",
+      "▁glo ry",
+      "▁Com pare",
+      "▁Comp are",
+      "▁Compar e",
+      "▁ Compare",
+      "巨 头",
+      "换 了",
+      "ze g",
+      "z eg",
+      "关于 适用",
+      "▁k inda",
+      "▁kind a",
+      "▁kin da",
+      "▁ki nda",
+      "▁Palest inian",
+      "陆 军",
+      "▁m iddleware",
+      "▁middle ware",
+      "▁ middleware",
+      "留 意",
+      "▁travel ers",
+      "Z W",
+      "ce mia",
+      "c emia",
+      "▁electroph oresis",
+      "\" `",
+      "▁Aw esome",
+      "▁ Awesome",
+      "▁pe oples",
+      "▁people s",
+      "▁Se par",
+      "▁Sep ar",
+      "▁ Separ",
+      "长 远",
+      "在不 同",
+      "在 不同",
+      "的一 半",
+      "的 一半",
+      "▁Sab ha",
+      "世 上",
+      "▁定 义",
+      "▁ 定义",
+      "医生 的",
+      "医 生的",
+      "▁ter med",
+      "▁term ed",
+      "▁exp ired",
+      "▁expire d",
+      "望 着",
+      "本 法",
+      "nDist ribution",
+      "n Distribution",
+      "随 即",
+      "▁pol ls",
+      "▁poll s",
+      "▁rot ational",
+      "▁rotation al",
+      "A U",
+      "掌 控",
+      "维持 原判",
+      "ke t",
+      "k et",
+      "gr and",
+      "gra nd",
+      "gran d",
+      "g rand",
+      "而成 的",
+      "而 成的",
+      "▁re creation",
+      "▁rec reation",
+      "▁G host",
+      "▁Gh ost",
+      "Pr eferences",
+      "Pre ferences",
+      "Preference s",
+      "P references",
+      "七个 月",
+      "七 个月",
+      "催化 剂",
+      "ack ed",
+      "ff ield",
+      "ffi eld",
+      "f field",
+      "H V",
+      "it at",
+      "ita t",
+      "i tat",
+      "po ke",
+      "p oke",
+      "AT URE",
+      "良 性",
+      "重 心",
+      "▁Fil ms",
+      "▁Film s",
+      "错 的",
+      "▁flo oding",
+      "▁flood ing",
+      "▁resil ience",
+      "▁T ennis",
+      "▁Ten nis",
+      "▁sym met",
+      "▁autom obile",
+      "▁auto mobile",
+      "歌 舞",
+      "Circ le",
+      "C ircle",
+      "▁remote ly",
+      "▁remot ely",
+      "▁dist ortion",
+      "▁distort ion",
+      "刹 车",
+      "火 的",
+      "▁h et",
+      "▁he t",
+      "▁ het",
+      "ro mes",
+      "rom es",
+      "rome s",
+      "r omes",
+      "背 面",
+      "产业 的",
+      "产 业的",
+      "财政 局",
+      "es ta",
+      "est a",
+      "e sta",
+      "og rap",
+      "ogr ap",
+      "营 运",
+      "要 不",
+      "▁T urb",
+      "▁Tur b",
+      "▁Tu rb",
+      "▁Co vid",
+      "▁Cov id",
+      "GP T",
+      "G PT",
+      "ur an",
+      "ura n",
+      "u ran",
+      "阿里 巴巴",
+      "da ily",
+      "d aily",
+      "t D",
+      "顿 时",
+      "提升 了",
+      "顶 部",
+      "al ent",
+      "ale nt",
+      "号民事 裁定",
+      "细 分",
+      "W ater",
+      "的通 项公式",
+      "on acci",
+      "ona cci",
+      "▁p osed",
+      "▁pos ed",
+      "▁po sed",
+      "▁pose d",
+      "▁ posed",
+      "▁tr igon",
+      "▁tri gon",
+      "▁trig on",
+      "tt ar",
+      "t tar",
+      "focus ed",
+      "f ocused",
+      "akespe are",
+      "▁Bulg aria",
+      "aw i",
+      "a wi",
+      "nm ore",
+      "n more",
+      "仍 是",
+      "指 责",
+      "以来 的",
+      "以 来的",
+      "▁cock tail",
+      "nom bre",
+      "匿 名",
+      "全 日制",
+      "下 水",
+      "▁Su dan",
+      "▁Sud an",
+      "▁l uc",
+      "▁lu c",
+      "▁ luc",
+      "pen ding",
+      "pend ing",
+      "p ending",
+      "Args Constructor",
+      "下 图",
+      "up grade",
+      "川 县",
+      "fr ican",
+      "frica n",
+      "▁hyper bolic",
+      "唐山 市",
+      "唐 山市",
+      "ia ne",
+      "ian e",
+      "i ane",
+      "▁H elen",
+      "▁He len",
+      "▁Hel en",
+      "▁Create s",
+      "▁Creat es",
+      "▁Cre ates",
+      "nD ocument",
+      "n Document",
+      "▁出 场",
+      "▁ 出场",
+      "bo rg",
+      "bor g",
+      "b org",
+      "▁a us",
+      "▁au s",
+      "▁ aus",
+      "▁Gr ande",
+      "▁Grand e",
+      "▁Gra nde",
+      "▁Gran de",
+      "▁b other",
+      "▁both er",
+      "▁bo ther",
+      "▁bot her",
+      "▁capac itor",
+      "A w",
+      "xl s",
+      "x ls",
+      "我们 也",
+      "▁C hang",
+      "▁Ch ang",
+      "▁Cha ng",
+      "▁Chan g",
+      "▁br ake",
+      "▁bra ke",
+      "el lation",
+      "ell ation",
+      "ella tion",
+      "▁Altern atively",
+      "▁Alternative ly",
+      "来 回",
+      "dr y",
+      "d ry",
+      "▁T hr",
+      "▁Th r",
+      "las ses",
+      "lass es",
+      "▁les bian",
+      "▁sur prises",
+      "▁surpr ises",
+      "▁surprise s",
+      "全 校",
+      "对 她",
+      "hi m",
+      "h im",
+      "▁th roat",
+      "▁thro at",
+      "▁travel led",
+      "▁trav elled",
+      "we ak",
+      "B order",
+      "▁dynam ical",
+      "▁dynamic al",
+      "伴 有",
+      "名 誉",
+      "等多 个",
+      "等 多个",
+      "▁eff ector",
+      "▁effect or",
+      "推 销",
+      "wan g",
+      "wa ng",
+      "w ang",
+      "▁ 蒋",
+      "活 的",
+      "xr ef",
+      "x ref",
+      "▁st ellar",
+      "▁ stellar",
+      "我 这",
+      "肯定 会",
+      "▁R osen",
+      "▁Ro sen",
+      "▁Rose n",
+      "▁Ros en",
+      "nab out",
+      "n about",
+      "▁Mc Donald",
+      "▁McD onald",
+      "t P",
+      "▁ „",
+      "▁G ir",
+      "▁Gi r",
+      "▁oblig ations",
+      "▁obligation s",
+      "明 智",
+      "aa aa",
+      "aaa a",
+      "a aaa",
+      "▁D eb",
+      "▁De b",
+      "▁ Deb",
+      "暂无 财产可供执行",
+      "暂 无财产可供执行",
+      "投 保",
+      "追 加",
+      "记者 从",
+      "ocom plete",
+      "o complete",
+      "农 家",
+      "出 示",
+      "知 的",
+      "保证 了",
+      "没有 了",
+      "没 有了",
+      "▁B ash",
+      "▁Bas h",
+      "▁Ba sh",
+      "▁p ing",
+      "▁pi ng",
+      "▁pin g",
+      "▁ ping",
+      "E r",
+      "楼 层",
+      "风 湿",
+      "没 问题",
+      "酒店的房型 有多种选择",
+      "仙 人",
+      "光 纤",
+      "▁fl ame",
+      "▁inf initely",
+      "▁infinite ly",
+      "▁infinit ely",
+      "▁tweet ed",
+      "▁twe eted",
+      "卡 尔",
+      "摄影 师",
+      "▁S aw",
+      "▁Sa w",
+      "▁s por",
+      "▁sp or",
+      "▁spo r",
+      "技 艺",
+      "计 时",
+      "时间 长",
+      "的数 字",
+      "的 数字",
+      "▁B ench",
+      "▁Be nch",
+      "▁Ben ch",
+      "▁ Bench",
+      "他 和",
+      "我 心",
+      "茫 茫",
+      "▁政 治",
+      "▁ 政治",
+      "Re place",
+      "▁out look",
+      "IN FRINGEMENT",
+      "亚 太",
+      "礼 貌",
+      "▁L ak",
+      "▁La k",
+      "▁c iting",
+      "▁cit ing",
+      "▁ci ting",
+      "▁broad casting",
+      "▁broadcast ing",
+      "大 声",
+      "角 为",
+      "ic hes",
+      "ich es",
+      "iche s",
+      "i ches",
+      "高 频",
+      "安全 隐患",
+      "raw ler",
+      "rawl er",
+      "▁Val ues",
+      "▁Value s",
+      "▁ Values",
+      "▁mis under",
+      "t L",
+      "干 的",
+      "nIn stead",
+      "nInst ead",
+      "加 剧",
+      "▁ 打开",
+      "事 迹",
+      "建 平",
+      "t console",
+      "拼 命",
+      "ai du",
+      "aid u",
+      "▁T rend",
+      "▁Tr end",
+      "▁Tre nd",
+      "▁embry o",
+      "▁embr yo",
+      "独 具",
+      "随 身",
+      "▁P CI",
+      "▁PC I",
+      "▁ PCI",
+      "▁H tml",
+      "▁ Html",
+      "仙 侠",
+      "▁gal axies",
+      "▁hyd raulic",
+      "以 非法",
+      "▁b reach",
+      "▁br each",
+      "▁bre ach",
+      "alu able",
+      "▁M oving",
+      "▁Mo ving",
+      "▁M yst",
+      "▁My st",
+      "▁Mys t",
+      "▁midfield ers",
+      "▁midfielder s",
+      "▁wood s",
+      "▁wo ods",
+      "ther efore",
+      "there fore",
+      "土 家族",
+      "▁re forms",
+      "▁ref orms",
+      "▁reform s",
+      "▁Em ployee",
+      "▁Employ ee",
+      "▁ Employee",
+      "h c",
+      "无 缝",
+      "游 览",
+      "盐 水",
+      "mo z",
+      "m oz",
+      "的 决定",
+      "' ^",
+      "引起 了",
+      "引 起了",
+      "▁Excell ence",
+      "sub scription",
+      "搬 家",
+      "nE lect",
+      "nEl ect",
+      "n Elect",
+      "大 盘",
+      "放弃 了",
+      "▁haz ards",
+      "▁hazard s",
+      "ma re",
+      "mar e",
+      "m are",
+      "▁G MAT",
+      "▁GM AT",
+      "▁mir rors",
+      "▁mirror s",
+      "▁free zing",
+      ":: {",
+      ": :{",
+      "▁dis course",
+      "▁disc ourse",
+      "▁discour se",
+      "ad don",
+      "add on",
+      "▁s ellers",
+      "▁sell ers",
+      "▁seller s",
+      "▁sel lers",
+      "▁graph ene",
+      "▁grap hene",
+      "▁enthus iasm",
+      "▁enthusi asm",
+      "и с",
+      "关 卡",
+      "是 国家",
+      "为什么 不",
+      "▁sun light",
+      "奥 林",
+      "还 真",
+      "Id x",
+      "I dx",
+      "▁k a",
+      "▁ ka",
+      "的心 理",
+      "的 心理",
+      "▁re conc",
+      "▁rec onc",
+      "▁recon c",
+      "fa v",
+      "f av",
+      "的 音乐",
+      "▁W ing",
+      "▁Win g",
+      "▁Wi ng",
+      "▁ Wing",
+      "st a",
+      "s ta",
+      "Gi ven",
+      "Give n",
+      "G iven",
+      "第二百五十 七",
+      "▁rout inely",
+      "▁routine ly",
+      "Ch rome",
+      "问 道",
+      "A im",
+      "• \\",
+      "较 长",
+      "▁ski ing",
+      "生活 方式",
+      "nE val",
+      "nEv al",
+      "n Eval",
+      "end ants",
+      "endant s",
+      "先 将",
+      "ad ers",
+      "ade rs",
+      "ader s",
+      "a ders",
+      "▁a per",
+      "▁ap er",
+      "▁ aper",
+      "有 色",
+      "▁h aul",
+      "▁ha ul",
+      "▁ haul",
+      "▁ach ieves",
+      "▁achieve s",
+      "怀 念",
+      "阿 根廷",
+      "▁advoc ates",
+      "▁advocate s",
+      "▁med iating",
+      "▁media ting",
+      "▁medi ating",
+      "▁mix tures",
+      "▁mixture s",
+      "一 站",
+      "\\' %",
+      "为 目标",
+      "洛阳 市",
+      "洛 阳市",
+      "▁V ehicle",
+      "▁ Vehicle",
+      "取得 的",
+      "取 得的",
+      "的 网络",
+      "电子 邮件",
+      "▁[ ^",
+      "▁ [^",
+      "说的 是",
+      "说 的是",
+      "grow th",
+      "g rowth",
+      "com press",
+      "comp ress",
+      "▁Guide lines",
+      "▁Guid elines",
+      "弹 簧",
+      "题 型",
+      "▁macroph age",
+      "E q",
+      "企业 文化",
+      "TE GER",
+      "▁Le ave",
+      "▁ Leave",
+      "人 情",
+      "你 做出",
+      "ore tical",
+      "oretic al",
+      "▁mit igate",
+      "▁ 启动",
+      "rel ationship",
+      "relation ship",
+      "relations hip",
+      "▁Oak land",
+      "▁publish ers",
+      "▁publisher s",
+      "并 购",
+      "uff y",
+      "uf fy",
+      "▁The rm",
+      "▁Th erm",
+      "▁Ther m",
+      "▁ref use",
+      "program ming",
+      "▁hemat opoietic",
+      "见 解",
+      "车 位",
+      "IC T",
+      "I CT",
+      "▁n ost",
+      "▁no st",
+      "▁nos t",
+      "▁ nost",
+      "▁Ch arg",
+      "▁Char g",
+      "▁Cha rg",
+      "前 方",
+      "▁caps ule",
+      "▁cyto chrome",
+      "特 有的",
+      "运动 的",
+      "运 动的",
+      "▁prote ase",
+      "▁ 克",
+      "Ch oice",
+      "Cho ice",
+      "ur ry",
+      "▁B uffer",
+      "▁Buff er",
+      "▁ Buffer",
+      "ax ios",
+      "配 资",
+      "nS ort",
+      "nSo rt",
+      "n Sort",
+      "▁land mark",
+      "▁semin ars",
+      "▁seminar s",
+      "▁der ivation",
+      "▁deriv ation",
+      "ot er",
+      "ote r",
+      "o ter",
+      "予以 准许",
+      "Sup press",
+      "火 星",
+      "出具 了",
+      "添加 剂",
+      "的一 条",
+      "的 一条",
+      "J o",
+      "情 境",
+      ")/ /",
+      ") //",
+      "T s",
+      "的一 家",
+      "的 一家",
+      "Log ic",
+      "Lo gic",
+      "▁sanction s",
+      "▁voc abulary",
+      "作 词",
+      "螺 丝",
+      "陷 阱",
+      "▁Un ique",
+      "▁Uni que",
+      "▁ Unique",
+      "RO S",
+      "R OS",
+      "nD i",
+      "n Di",
+      "▁C red",
+      "▁Cr ed",
+      "▁Cre d",
+      "▁ Cred",
+      "▁H ook",
+      "▁Ho ok",
+      "▁ Hook",
+      "午 餐",
+      "如 上",
+      "愈 合",
+      "▁mat urity",
+      "nW ow",
+      "▁sk ull",
+      "▁sup ermark",
+      "▁super mark",
+      "三 观",
+      "请 从",
+      "有 兴趣",
+      "胡椒 粉",
+      "Word s",
+      "W ords",
+      "pw d",
+      "p wd",
+      "▁m ph",
+      "▁mp h",
+      "▁ mph",
+      "总 支",
+      "▁t au",
+      "▁ta u",
+      "▁ tau",
+      "▁cl icks",
+      "▁click s",
+      "▁cli cks",
+      "man if",
+      "ma nif",
+      "mani f",
+      "vers ation",
+      "涂 抹",
+      "邢 台",
+      "▁K ick",
+      "▁Ki ck",
+      "▁cr ush",
+      "▁cru sh",
+      "▁gu essing",
+      "▁guess ing",
+      "▁S UB",
+      "▁SU B",
+      "▁ SUB",
+      "▁S yl",
+      "▁Sy l",
+      "符合 法定",
+      "nS ection",
+      "nSe ction",
+      "nSec tion",
+      "n Section",
+      "▁fl avour",
+      "▁flav our",
+      "▁break through",
+      "▁cig arette",
+      "宣 讲",
+      "▁N ak",
+      "▁Na k",
+      "ed ish",
+      "edi sh",
+      "um ann",
+      "uman n",
+      "uma nn",
+      "u mann",
+      "亲 情",
+      "唱 片",
+      "▁A part",
+      "▁Ap art",
+      "▁aff air",
+      "▁af fair",
+      "却 依然",
+      "pt ides",
+      "ptide s",
+      "▁le thal",
+      "▁let hal",
+      "▁Contin ue",
+      "▁ Continue",
+      "花 序",
+      "即使 是",
+      "k v",
+      "就不 是",
+      "就 不是",
+      "阶段 的",
+      "阶 段的",
+      "▁b od",
+      "▁bo d",
+      "Event Args",
+      "▁Argent ine",
+      "高 山",
+      "▁sh rink",
+      "▁shr ink",
+      "昆明 市",
+      "nM ain",
+      "nMa in",
+      "n Main",
+      "▁T ampa",
+      "▁Tam pa",
+      "▁para site",
+      "▁paras ite",
+      "Bu y",
+      "B uy",
+      "DE P",
+      "D EP",
+      "ip ients",
+      "ipient s",
+      "▁hot test",
+      "▁ho ttest",
+      "▁stud ios",
+      "▁studio s",
+      "Java Script",
+      "有趣 的",
+      "ib us",
+      "ibu s",
+      "i bus",
+      "▁class ify",
+      "sw ick",
+      "s wick",
+      "▁b if",
+      "▁bi f",
+      "▁Pun jab",
+      "▁thromb osis",
+      "▁thrombo sis",
+      "反 击",
+      "有 何",
+      "发言 人",
+      "拿 起",
+      "ht ra",
+      "h tra",
+      "▁ass urance",
+      "{→ }{",
+      "富 士",
+      "紧 凑",
+      "TN F",
+      "T NF",
+      "一道 题",
+      "一 道题",
+      "的解 集",
+      "▁al together",
+      "▁alt ogether",
+      "散 发",
+      "无 忧",
+      "▁am ple",
+      "▁ampl e",
+      "▁amp le",
+      "▁ ample",
+      "可 使",
+      "行政 行为",
+      "max imum",
+      "牵 引",
+      "at ti",
+      "att i",
+      "a tti",
+      "分 许",
+      "Ps i",
+      "P si",
+      "▁L H",
+      "▁ LH",
+      "nD el",
+      "nDe l",
+      "n Del",
+      "出具 欠条",
+      "▁f led",
+      "▁fl ed",
+      "▁fle d",
+      "▁Object ive",
+      "▁Ob jective",
+      "得 的",
+      "不 舒服",
+      "▁K oh",
+      "▁Ko h",
+      "▁gr ains",
+      "▁grain s",
+      "▁gra ins",
+      "▁con verts",
+      "▁conver ts",
+      "▁convert s",
+      "▁conv erts",
+      "▁sn acks",
+      "▁snack s",
+      "▁Mil waukee",
+      "[ /",
+      "口 气",
+      "▁com pose",
+      "▁comp ose",
+      "▁compos e",
+      "▁ compose",
+      "▁m uscular",
+      "▁mus cular",
+      "▁musc ular",
+      "跑 到",
+      "lu c",
+      "l uc",
+      "全 集",
+      "名 片",
+      "有 必要",
+      "du plex",
+      "dup lex",
+      "d uplex",
+      "now led",
+      "Gate way",
+      "G ateway",
+      "中国人民 财产保险股份有限公司",
+      "好 转",
+      "*} \\",
+      "* }\\",
+      "公告 费",
+      "nj ust",
+      "n just",
+      "边 境",
+      "equ ation",
+      "eq uation",
+      "%\\ \\",
+      "% \\\\",
+      "par a",
+      "pa ra",
+      "p ara",
+      "人大 代表",
+      "经营 管理",
+      "not ice",
+      "▁co oler",
+      "▁cool er",
+      "山 药",
+      "手续 费",
+      "▁compet ent",
+      "▁compete nt",
+      "さ れ",
+      "隶 属",
+      "签订 了",
+      "ot opy",
+      "oto py",
+      "▁princ ip",
+      "▁mult imedia",
+      "▁multi media",
+      "▁multim edia",
+      "IN A",
+      "I NA",
+      "才能 够",
+      "才 能够",
+      "▁w ors",
+      "▁wor s",
+      "▁wo rs",
+      "pl ayers",
+      "play ers",
+      "player s",
+      "p layers",
+      "▁Phot ography",
+      "▁发 展",
+      "▁ 发展",
+      "▁| |\\",
+      "▁|| \\",
+      "FI LES",
+      "FILE S",
+      "FIL ES",
+      "▁L eeds",
+      "▁Le eds",
+      "▁Lee ds",
+      "▁} _{",
+      "▁ }_{",
+      "因为 他们",
+      "因为他 们",
+      "因 为他们",
+      "np art",
+      "n part",
+      "叙 事",
+      "nE p",
+      "n Ep",
+      "▁Z n",
+      "▁ Zn",
+      "又一 次",
+      "又 一次",
+      "nGo ogle",
+      "n Google",
+      "▁U ber",
+      "▁pro mo",
+      "▁pr omo",
+      "▁prom o",
+      "▁overl apping",
+      "▁overlap ping",
+      "▁ 军",
+      "商 丘",
+      "▁记 员",
+      "T w",
+      "尊 敬",
+      "就是 为了",
+      "就 是为了",
+      "Un ited",
+      "Unit ed",
+      "▁hydro ph",
+      "▁hyd roph",
+      "的 男人",
+      "▁G eo",
+      "▁Ge o",
+      "▁ Geo",
+      "▁p ause",
+      "▁pa use",
+      "▁ pause",
+      "▁phys i",
+      "▁phy si",
+      "ting ham",
+      "t ingham",
+      "▁Jer emy",
+      "畸 形",
+      "锦标 赛",
+      "▁C U",
+      "▁ CU",
+      "通知 后",
+      "▁air ports",
+      "▁airport s",
+      "▁con ting",
+      "▁cont ing",
+      "▁contin g",
+      "完 后",
+      "强制 法",
+      "iv ate",
+      "iva te",
+      "▁L odge",
+      "if ferences",
+      "iffer ences",
+      "ifference s",
+      "▁ 副",
+      "反 抗",
+      "二十 四",
+      "二 十四",
+      "▁Qu arter",
+      "▁Quart er",
+      "情 商",
+      "GU I",
+      "G UI",
+      "▁proport ions",
+      "▁proportion s",
+      "∞ ,",
+      "ur ling",
+      "url ing",
+      "press ure",
+      "▁par ametric",
+      "▁paramet ric",
+      "▁para metric",
+      "▁param etric",
+      "▁ parametric",
+      "战 队",
+      "缓 缓",
+      "▁a mph",
+      "▁am ph",
+      "▁amp h",
+      "int ellij",
+      "到 时候",
+      "▁environment ally",
+      "▁environmental ly",
+      "整 机",
+      "梦 见",
+      "▁) {\\",
+      "▁ ){\\",
+      "▁appl iance",
+      "却 在",
+      "用 作",
+      "Ha r",
+      "H ar",
+      "nT ext",
+      "nTe xt",
+      "n Text",
+      "Label s",
+      "Lab els",
+      "▁a irlines",
+      "▁air lines",
+      "▁airline s",
+      "os i",
+      "o si",
+      "pe nt",
+      "pen t",
+      "p ent",
+      "▁L iu",
+      "▁Li u",
+      "ear range",
+      "▁anal ogue",
+      "▁analog ue",
+      "西 门",
+      "▁s or",
+      "▁so r",
+      "那个 时候",
+      "一 头",
+      "ou g",
+      "o ug",
+      "ch oline",
+      "cho line",
+      "chol ine",
+      "有 种",
+      "▁x y",
+      "▁ xy",
+      "▁T yler",
+      "▁Ty ler",
+      "▁elect rophys",
+      "▁electro phys",
+      "▁electroph ys",
+      "▁electr ophys",
+      "▁ 电",
+      "表达 式",
+      "都 已经",
+      "▁P ont",
+      "▁Po nt",
+      "▁Pon t",
+      "▁d ign",
+      "▁di gn",
+      "▁dig n",
+      "CAS CADE",
+      "▁Again st",
+      "债权 人",
+      "Sp e",
+      "S pe",
+      "or te",
+      "ort e",
+      "ic one",
+      "icon e",
+      "ico ne",
+      "i cone",
+      "par s",
+      "pa rs",
+      "p ars",
+      "源 头",
+      "▁ac credited",
+      "▁accred ited",
+      "▁accr edited",
+      "征 服",
+      "抗 生素",
+      "▁E mma",
+      "▁Em ma",
+      "捕 捉",
+      "致 命",
+      "'\\ \\",
+      "' \\\\",
+      "商 学院",
+      "净 利润",
+      "到 手",
+      "与 应用",
+      "▁B ever",
+      "▁Be ver",
+      "抛物 线的",
+      "抛物线 的",
+      "da ting",
+      "dat ing",
+      "d ating",
+      "▁echo card",
+      "▁ech ocard",
+      "灭 火",
+      "▁prompt ly",
+      "Eval uation",
+      "ac i",
+      "a ci",
+      "▁con cludes",
+      "▁conclude s",
+      "I j",
+      "兴 业",
+      "▁L DL",
+      "▁LD L",
+      "▁ LDL",
+      "▁p orch",
+      "▁por ch",
+      "▁gly col",
+      "▁glyc ol",
+      "挣 钱",
+      "▁p ean",
+      "▁pe an",
+      "▁sl ug",
+      "▁ slug",
+      "mi ral",
+      "mir al",
+      "m iral",
+      "闪 光",
+      "▁H C",
+      "▁ HC",
+      "▁tr acker",
+      "▁track er",
+      "▁ tracker",
+      "▁L ines",
+      "▁Lin es",
+      "▁Line s",
+      "▁Li nes",
+      "▁ Lines",
+      "▁interfer on",
+      "▁interf eron",
+      "想要 的",
+      "▁I NF",
+      "▁IN F",
+      "▁ INF",
+      "▁t ib",
+      "▁ti b",
+      "出 院",
+      "▁line age",
+      "nI nf",
+      "nIn f",
+      "n Inf",
+      "▁d eer",
+      "▁de er",
+      "入 驻",
+      "▁C ream",
+      "▁Cre am",
+      "ograph ies",
+      "Trans lation",
+      "Transl ation",
+      "迄 今",
+      "um ed",
+      "ume d",
+      "u med",
+      ")) /((",
+      "))/ ((",
+      "))/( (",
+      ") )/((",
+      "笔 画",
+      "▁S pl",
+      "▁Sp l",
+      "▁ Spl",
+      "▁Cl osing",
+      "▁Clo sing",
+      "▁sea mless",
+      "▁seam less",
+      "▁M n",
+      "▁ Mn",
+      "其前 身为",
+      "▁Re dis",
+      "▁Red is",
+      "▁ Redis",
+      "▁adv ancing",
+      "▁Rel ationship",
+      "▁Relations hip",
+      "▁ Relationship",
+      "无 名",
+      "op ia",
+      "文 的",
+      "▁P roc",
+      "▁Pro c",
+      "▁Pr oc",
+      "▁ Proc",
+      "▁off shore",
+      "左 手",
+      "纸 张",
+      "私 信",
+      "违 背",
+      "ver sal",
+      "vers al",
+      "下 雨",
+      "▁Wh atever",
+      "▁What ever",
+      "诱 发",
+      "增 函数",
+      "乡 市",
+      "打 架",
+      "的 系统",
+      "考 上",
+      "屏 蔽",
+      "期 权",
+      "并按 照",
+      "并 按照",
+      "二〇一七年 十二月",
+      "二〇一七年十二 月",
+      "二〇一七年十 二月",
+      "ic als",
+      "ical s",
+      "ica ls",
+      "伴 侣",
+      "列 举",
+      "育 人",
+      "辩 论",
+      "新中国的 诞生",
+      "fil ling",
+      "fill ing",
+      "f illing",
+      "▁un comfortable",
+      "三 峡",
+      "跳 跃",
+      "▁I B",
+      "▁ IB",
+      "last ing",
+      "las ting",
+      "l asting",
+      "▁Jess ica",
+      "▁des ires",
+      "▁desire s",
+      "nS chedule",
+      "nSc hedule",
+      "n Schedule",
+      "▁C d",
+      "▁ Cd",
+      "幸运 的是",
+      "stan bul",
+      "▁predict s",
+      "▁comprom ised",
+      "▁compromise d",
+      "一 间",
+      "ge on",
+      "geo n",
+      "g eon",
+      "in sky",
+      "ins ky",
+      "insk y",
+      "▁ 进球",
+      "ul ose",
+      "ulo se",
+      "ulos e",
+      "u lose",
+      "好看 的",
+      "好 看的",
+      "▁P DE",
+      "▁PD E",
+      "may be",
+      "▁bal ancing",
+      "并 向",
+      "SE C",
+      "S EC",
+      "服务 等",
+      "min ster",
+      "mins ter",
+      "m inster",
+      "万 亩",
+      "路 边",
+      "Pix el",
+      "P ixel",
+      "正确答案 和",
+      "交通 大学",
+      "▁c him",
+      "▁ch im",
+      "▁chi m",
+      "La unch",
+      "L aunch",
+      "in vest",
+      "inv est",
+      "▁spec s",
+      "▁spe cs",
+      "▁ specs",
+      "▁tr aged",
+      "▁tra ged",
+      "操 纵",
+      "来 做",
+      "As sets",
+      "Ass ets",
+      "Asset s",
+      "Load ed",
+      "Lo aded",
+      "▁cor rosion",
+      "城乡 建设",
+      "执行 长",
+      "执 行长",
+      "即可 求解",
+      "nth anks",
+      "nthan ks",
+      "n thanks",
+      "▁cl ause",
+      "▁short age",
+      "▁M W",
+      "▁ MW",
+      "住 浙江省",
+      "▁unve iled",
+      "li ft",
+      "lif t",
+      "l ift",
+      "oc yst",
+      "o cyst",
+      "cri tical",
+      "c ritical",
+      "▁ves icles",
+      "▁vesicle s",
+      "叠 加",
+      "复 苏",
+      "准确 的",
+      "投资 人",
+      "金 牛",
+      "AA C",
+      "A AC",
+      "先生 的",
+      "先 生的",
+      "集 装箱",
+      "mid t",
+      "mi dt",
+      "Sp rite",
+      "S prite",
+      "▁Em peror",
+      "▁hemorrh age",
+      "zi g",
+      "z ig",
+      "慢慢 的",
+      "▁F ILE",
+      "▁FI LE",
+      "▁ FILE",
+      "铁 道",
+      "应予 支持",
+      "ch in",
+      "chi n",
+      "c hin",
+      "积极 参与",
+      "工业 出版社",
+      "▁attenu ated",
+      "▁atten uated",
+      "nFilm ography",
+      "新能源 汽车",
+      "roc ket",
+      "rock et",
+      "r ocket",
+      "▁strength ening",
+      "▁strengthen ing",
+      "▁Term inal",
+      "▁ Terminal",
+      "恰 当",
+      "选 育",
+      "ber y",
+      "be ry",
+      "b ery",
+      "gu ided",
+      "guide d",
+      "gui ded",
+      "guid ed",
+      "缓 冲",
+      "一 双",
+      "eg g",
+      "e gg",
+      "nPh ys",
+      "n Phys",
+      "co ords",
+      "coord s",
+      "▁C hester",
+      "▁Ch ester",
+      "▁Che ster",
+      "▁Chest er",
+      "▁Ches ter",
+      "亮 的",
+      "们 在",
+      "第三 项",
+      "第 三项",
+      "合作 关系",
+      "▁Aud ience",
+      "▁Audi ence",
+      "骨 骼",
+      "▁\" )",
+      "▁ \")",
+      "▁plan ets",
+      "▁plane ts",
+      "▁planet s",
+      "Fi t",
+      "F it",
+      "完全 可以",
+      "if rame",
+      "i frame",
+      "制 订",
+      "糟 糕",
+      "斯 的",
+      "樱 桃",
+      "▁conver gent",
+      "▁converge nt",
+      "分 层",
+      "余 年",
+      "▁K re",
+      "▁Kr e",
+      "ar ius",
+      "ari us",
+      "al tern",
+      "alt ern",
+      "alter n",
+      "投 注",
+      "音 效",
+      "▁M t",
+      "卫生 院",
+      "to LowerCase",
+      "食 欲",
+      "▁Al most",
+      "sa ss",
+      "s ass",
+      "▁I RS",
+      "▁IR S",
+      "▁ IRS",
+      "▁B ound",
+      "▁Bo und",
+      "▁Bou nd",
+      "▁ Bound",
+      "process ed",
+      "proc essed",
+      "DN S",
+      "D NS",
+      "Har d",
+      "Ha rd",
+      "H ard",
+      "ro cal",
+      "roc al",
+      "r ocal",
+      "起 重",
+      "che wan",
+      "La TeX",
+      "Op tim",
+      "Opt im",
+      "▁it al",
+      "▁i tal",
+      "▁ ital",
+      "▁re vel",
+      "▁reve l",
+      "▁rev el",
+      "ath ione",
+      "athi one",
+      "不 便",
+      "也 让",
+      "ek yll",
+      "cap acity",
+      "▁Engine ers",
+      "▁Engineer s",
+      "bo nd",
+      "bon d",
+      "b ond",
+      "▁R eed",
+      "▁Re ed",
+      "View Holder",
+      "和解题 的思路",
+      "spir acy",
+      "▁cere bell",
+      "印 章",
+      "发电 机",
+      "发 电机",
+      "我只 是",
+      "我 只是",
+      "}\\\\ ,",
+      "} \\\\,",
+      "琐事 发生",
+      "重要 的作用",
+      "重要的 作用",
+      "你 了",
+      "打 出",
+      "经验 的",
+      "维 亚",
+      "tr on",
+      "tro n",
+      "t ron",
+      "▁d ar",
+      "▁da r",
+      "▁ dar",
+      "▁f ut",
+      "▁fu t",
+      "水 肿",
+      "Sc i",
+      "S ci",
+      "Le ss",
+      "Les s",
+      "L ess",
+      "in tro",
+      "int ro",
+      "intr o",
+      "第一百零 八条",
+      "较 强",
+      "幸福 的",
+      "▁per cept",
+      "V ue",
+      "Fat al",
+      "Fa tal",
+      "F atal",
+      "最 爱",
+      "积极 的",
+      "title s",
+      "tit les",
+      "女人 的",
+      "女 人的",
+      "▁E EG",
+      "▁EE G",
+      "告知 申请执行人",
+      "全 天",
+      "▁co variance",
+      "▁covari ance",
+      "困 惑",
+      "aj i",
+      "a ji",
+      "身份 号码",
+      "act ly",
+      "д е",
+      "权 纠纷一案",
+      "本次执行程序 终结",
+      "伴随着 新中国的诞生",
+      "▁G az",
+      "▁Ga z",
+      "请给出 你的",
+      "▁ve loc",
+      "▁vel oc",
+      "必 不可",
+      "更 名为",
+      "By Name",
+      "go lang",
+      "g olang",
+      "▁magn esium",
+      "▁juris diction",
+      "▁jurisdict ion",
+      "检察 机关",
+      "▁inhib iting",
+      "▁inhibit ing",
+      "▁P interest",
+      "▁cont acting",
+      "▁contact ing",
+      "K L",
+      "j j",
+      "fl ight",
+      "f light",
+      "▁phot ographers",
+      "▁photograph ers",
+      "▁photographer s",
+      "导致 了",
+      "ac eous",
+      "ace ous",
+      "第一百五十四条第一款第 十一",
+      "第一百五十四条第一款 第十一",
+      "化 县",
+      "搬 运",
+      "美术 馆",
+      "某 与被告",
+      "▁Let ters",
+      "▁Letter s",
+      "▁magn ificent",
+      "充 值",
+      "奇 函数",
+      "▁Hug hes",
+      "▁Hugh es",
+      "▁Br ussels",
+      "该 款",
+      "av in",
+      "avi n",
+      "a vin",
+      "▁un cle",
+      "▁unc le",
+      "nA L",
+      "n AL",
+      "不 充分",
+      ">\" ,\"",
+      ">\", \"",
+      "> \",\"",
+      "▁l acks",
+      "▁la cks",
+      "▁lack s",
+      "▁lac ks",
+      "定 律",
+      "格 外",
+      "的 费用",
+      "呼 和浩特",
+      "▁Y ale",
+      "▁Ya le",
+      "步 入",
+      "信息 查询",
+      "te enth",
+      "tee nth",
+      "teen th",
+      "t eenth",
+      "▁No table",
+      "▁Not able",
+      "▁in ability",
+      "再次 提出",
+      "▁requ esting",
+      "▁request ing",
+      "▁m oll",
+      "▁mo ll",
+      "▁mol l",
+      "耐 用",
+      "▁je ans",
+      "反 驳",
+      "IF T",
+      "I FT",
+      "id em",
+      "ide m",
+      "i dem",
+      "联 保",
+      "▁Comp ute",
+      "▁Comput e",
+      "▁ Compute",
+      "▁ 静",
+      "▁m um",
+      "▁mu m",
+      "▁ 服务",
+      "▁P ete",
+      "▁Pet e",
+      "▁Pe te",
+      "▁met ropolitan",
+      "IT ION",
+      "▁T iger",
+      "▁Ti ger",
+      "▁Tig er",
+      "target s",
+      "tar gets",
+      "▁S event",
+      "▁Se vent",
+      "▁Seven t",
+      "▁Sev ent",
+      "主 打",
+      "BI N",
+      "B IN",
+      "▁function ally",
+      "▁functional ly",
+      "fr a",
+      "f ra",
+      "获 荣誉",
+      "▁D ict",
+      "▁Di ct",
+      "▁ Dict",
+      "▁S acr",
+      "▁Sa cr",
+      "▁Sac r",
+      "▁S oci",
+      "▁So ci",
+      "▁Soc i",
+      "▁Pos itive",
+      "▁ Positive",
+      "工 贸",
+      "▁] ,",
+      "▁ ],",
+      "完成 的",
+      "完 成的",
+      "和 使用",
+      "oc he",
+      "och e",
+      "o che",
+      "▁c AMP",
+      "十二 条之规定",
+      "十二条 之规定",
+      "▁P rec",
+      "▁Pr ec",
+      "▁Pre c",
+      "▁penal ties",
+      "经营 活动",
+      "▁t enth",
+      "▁te nth",
+      "▁ten th",
+      "▁tent h",
+      "▁f ungal",
+      "▁fun gal",
+      "▁fung al",
+      "▁under gone",
+      "▁undergo ne",
+      "Ma g",
+      "M ag",
+      "▁ 获取",
+      "奠定 了",
+      "奠 定了",
+      "▁he par",
+      "肝 炎",
+      "zh i",
+      "z hi",
+      "▁acid ic",
+      "il iar",
+      "ili ar",
+      "ilia r",
+      "Compar ison",
+      "inherit doc",
+      "交 际",
+      "▁is omorphism",
+      "▁K om",
+      "▁Ko m",
+      "▁Se bast",
+      "来 电",
+      "▁S hir",
+      "▁Sh ir",
+      "▁Shi r",
+      "▁micro M",
+      "于本判决生效后 十日内",
+      "v n",
+      "万 达",
+      "胸 部",
+      "pub lisher",
+      "publish er",
+      "'] ['",
+      "'][ '",
+      "' ]['",
+      "▁t ilt",
+      "▁til t",
+      "▁ti lt",
+      "View s",
+      "Vi ews",
+      "▁P ine",
+      "▁Pi ne",
+      "▁Pin e",
+      "te nant",
+      "ten ant",
+      "t enant",
+      "▁B ring",
+      "▁Br ing",
+      "▁Bri ng",
+      "▁s olic",
+      "▁so lic",
+      "▁sol ic",
+      "▁Four th",
+      "▁Public ations",
+      "▁Publication s",
+      "▁G and",
+      "▁Ga nd",
+      "▁Gan d",
+      "▁recycl ed",
+      "上 天",
+      "美 白",
+      "▁设施 与服务",
+      "开 本",
+      "▁h il",
+      "▁hi l",
+      "▁fem oral",
+      "多 层",
+      "不明 确",
+      "不 明确",
+      "▁exc use",
+      "com poser",
+      "comp oser",
+      "compose r",
+      "Function s",
+      "▁pol ygon",
+      "▁poly gon",
+      "▁polyg on",
+      "▁Azerbai jan",
+      "平台 上",
+      "平 台上",
+      "二〇一六年 十一月",
+      "二〇一六年十一 月",
+      "二〇一六年十 一月",
+      "和解题 过程",
+      "和 解题过程",
+      "原 数",
+      "实 证",
+      "board ing",
+      "bo arding",
+      "主要 从事",
+      "comp act",
+      "▁I ng",
+      "▁In g",
+      "▁ Ing",
+      "▁S ão",
+      "▁dis established",
+      "▁disestablish ed",
+      "自 卑",
+      "让 您",
+      "不 死",
+      ":: $",
+      ": :$",
+      "教育 厅",
+      "▁T ap",
+      "▁Ta p",
+      "▁ Tap",
+      "▁Fa ith",
+      "担保 有限公司",
+      "▁law makers",
+      "买 个",
+      "以 至于",
+      "Mb ps",
+      "M bps",
+      "岁 时",
+      "燃 油",
+      "pi e",
+      "p ie",
+      "▁M ull",
+      "▁Mu ll",
+      "▁Mul l",
+      "▁n oon",
+      "▁no on",
+      "▁ noon",
+      "▁stabil ization",
+      "▁stab ilization",
+      "▁L amb",
+      "▁La mb",
+      "▁Lam b",
+      "美 联",
+      "ns p",
+      "n sp",
+      "AB CD",
+      "ABC D",
+      "▁decor ating",
+      "nA sk",
+      "nAs k",
+      "n Ask",
+      "ver ting",
+      "vert ing",
+      "▁Princ eton",
+      "▁Prince ton",
+      "▁Pal mer",
+      "▁Palm er",
+      "你做出 选择的依据",
+      "▁Mark down",
+      "▁ Markdown",
+      "▁Stock holm",
+      "你做出选择的依据 和分析",
+      "稳 步",
+      "两 被告",
+      "Vis itor",
+      "Visit or",
+      "C ategories",
+      "N M",
+      "严 禁",
+      "gc d",
+      "g cd",
+      "L ENGTH",
+      "无 尽",
+      "▁C X",
+      "▁ CX",
+      "此外 请给出",
+      "▁n icht",
+      "▁nic ht",
+      "▁ni cht",
+      "▁St uart",
+      "▁survey ed",
+      "很 长",
+      "肺 癌",
+      "详 见",
+      "一对 一",
+      "▁iter ative",
+      "▁neut r",
+      "风 的",
+      "▁\" :",
+      "▁ \":",
+      "▁sp ike",
+      "▁sol ub",
+      "广东省 广州市",
+      "第一百一十九 条",
+      "第一百一 十九条",
+      "nDisc ography",
+      "帐 户",
+      "天 文",
+      "Pol y",
+      "Po ly",
+      "P oly",
+      "每个 客房",
+      "▁ERR OR",
+      "▁ ERROR",
+      "▁G overn",
+      "▁Go vern",
+      "▁Gover n",
+      "▁Gov ern",
+      "宏 伟",
+      "▁dimin ished",
+      "Q A",
+      "单 车",
+      "AI DS",
+      "A IDS",
+      "per ed",
+      "pe red",
+      "p ered",
+      "尔多 斯",
+      "tr acks",
+      "tra cks",
+      "track s",
+      "貌 似",
+      "an chor",
+      "anc hor",
+      "anch or",
+      "(\" ,",
+      "( \",",
+      "就不 要",
+      "就 不要",
+      "no p",
+      "n op",
+      "ib ling",
+      "ibli ng",
+      "ibl ing",
+      "i bling",
+      "▁ind ucing",
+      "▁indu cing",
+      "加 州",
+      "农民 收入",
+      "for est",
+      "fore st",
+      "fo rest",
+      "yl abel",
+      "yla bel",
+      "y label",
+      "▁VAL UES",
+      "% \"",
+      "▁ 武",
+      "(' #",
+      "gr pc",
+      "g rpc",
+      "过程 的",
+      "申请 再审",
+      "中国科学院 成立",
+      "ford shire",
+      "av ailability",
+      "▁C SF",
+      "▁CS F",
+      "▁ CSF",
+      "并给出 相应的",
+      "封 装",
+      "我 从",
+      "顺 德",
+      "▁bal lot",
+      "▁ball ot",
+      "▁E g",
+      "et heus",
+      "eth eus",
+      "ethe us",
+      "▁dam aging",
+      "▁qu arters",
+      "▁quarter s",
+      "▁quart ers",
+      "▁ quarters",
+      "浴 缸",
+      "陈 列",
+      "AC S",
+      "A CS",
+      "超过 了",
+      "超 过了",
+      "▁draft ed",
+      "▁print able",
+      "▁pri ntable",
+      "优 异",
+      "转 会",
+      "▁fac ilitating",
+      "▁facilit ating",
+      "毒 性",
+      "说不 定",
+      "说 不定",
+      "▁st ent",
+      "▁ste nt",
+      "前 两",
+      "四 级",
+      "nm od",
+      "n mod",
+      "四 方",
+      "沉 浸",
+      "nS E",
+      "n SE",
+      "漂亮 的",
+      "漂 亮的",
+      "▁neph ro",
+      "组 装",
+      "▁play offs",
+      "▁playoff s",
+      "N V",
+      "便 携",
+      "滑 雪",
+      "给 的",
+      "te c",
+      "t ec",
+      "▁Trad itional",
+      "lock ed",
+      "我 希望",
+      "身边 的",
+      "身 边的",
+      "W B",
+      "发现 的",
+      "每个客房 都配有",
+      "com pleted",
+      "comp leted",
+      "compl eted",
+      "complete d",
+      "as tery",
+      "ast ery",
+      "aster y",
+      "aste ry",
+      "▁graph ical",
+      "▁graphic al",
+      "瀑 布",
+      "是 自己",
+      "▁con duction",
+      "▁conduct ion",
+      "▁cond uction",
+      "锐 角",
+      "open hagen",
+      "▁Admin istrative",
+      "▁Administr ative",
+      "并 用",
+      "▁v a",
+      "▁ va",
+      "nd is",
+      "ndi s",
+      "n dis",
+      "其他 可供执行财产",
+      "违 纪",
+      "ha o",
+      "h ao",
+      "Da rk",
+      "Dar k",
+      "D ark",
+      "资本 主义",
+      "va c",
+      "v ac",
+      "ri ves",
+      "riv es",
+      "rive s",
+      "r ives",
+      "色 调",
+      "▁even ly",
+      "这篇 日志",
+      "{ `",
+      "海 尔",
+      "高 雄",
+      "长方 体",
+      "长方 体的",
+      "长方体 的",
+      "Col ors",
+      "Color s",
+      "▁green house",
+      "AR C",
+      "A RC",
+      "▁syst olic",
+      "冷 漠",
+      "数的 和",
+      "数 的和",
+      "▁E volution",
+      "▁Ev olution",
+      "▁private ly",
+      "▁priv ately",
+      "▁question naires",
+      "▁questionnaire s",
+      "清 晨",
+      "在 日本",
+      "件 的",
+      "由 来",
+      "Iss ue",
+      "▁( /",
+      "▁ (/",
+      "可以 向本院申请复议一次",
+      "▁terr itories",
+      "▁territ ories",
+      "AM D",
+      "A MD",
+      "华 中",
+      "算 了",
+      "鱼 类",
+      "▁D oug",
+      "▁Do ug",
+      "▁Dou g",
+      "患 病",
+      "\"] ,\"",
+      "\"], \"",
+      "\" ],\"",
+      "Be havior",
+      "交 汇",
+      "只 见",
+      "▁S ang",
+      "▁San g",
+      "▁Sa ng",
+      "▁ju venile",
+      "全 自动",
+      "主要 作品",
+      "▁benef ici",
+      "TH ON",
+      "▁Y oga",
+      "▁Yo ga",
+      "Sc hedule",
+      "S chedule",
+      "▁Am azing",
+      "水 上",
+      "▁W ag",
+      "▁Wa g",
+      "▁there after",
+      "▁sh aping",
+      "▁sha ping",
+      "oglob ulin",
+      "放 假",
+      "现 货",
+      "畅 通",
+      "比 喻",
+      "相 传",
+      "() ))",
+      "()) )",
+      "( )))",
+      "▁re mission",
+      "▁rem ission",
+      "面临 的",
+      "ps on",
+      "p son",
+      "供 热",
+      "制 裁",
+      "Pre vious",
+      "vers ible",
+      "手 感",
+      "▁s tro",
+      "▁st ro",
+      "▁str o",
+      "▁B undle",
+      "▁Bund le",
+      "▁ Bundle",
+      "▁P ossible",
+      "▁Poss ible",
+      "我 公司",
+      "oplas m",
+      "opl asm",
+      "▁H indi",
+      "▁Hi ndi",
+      "▁Hind i",
+      "社会 保障",
+      "人民 出版社",
+      "▁le agues",
+      "▁league s",
+      "up us",
+      "u pus",
+      "经济 社会发展",
+      "经济社会 发展",
+      "▁Ins ert",
+      "▁ Insert",
+      "万元 的",
+      "万 元的",
+      "ec on",
+      "eco n",
+      "e con",
+      "li te",
+      "lit e",
+      "l ite",
+      "▁be ads",
+      "▁bead s",
+      "▁bo tan",
+      "▁bot an",
+      "▁m ines",
+      "▁min es",
+      "▁mine s",
+      "▁mi nes",
+      "g allery",
+      "▁decor ations",
+      "▁decoration s",
+      "| $",
+      "国 立",
+      "美 景",
+      "营 收",
+      "吸引 力",
+      "吸 引力",
+      "is miss",
+      "ism iss",
+      "▁Re uters",
+      "▁ Reuters",
+      "Ho l",
+      "H ol",
+      "[] )",
+      "[ ])",
+      "tim ately",
+      "t imately",
+      "▁collabor ations",
+      "▁collaboration s",
+      "由上海市 人民政府",
+      "ancel lor",
+      "ancell or",
+      "西 游",
+      "他们还 推荐了",
+      "这篇日志 的人",
+      "推荐过 这篇日志的人",
+      "宿 迁",
+      "▁V PN",
+      "▁VP N",
+      "▁ VPN",
+      "素质 教育",
+      "ar ming",
+      "arm ing",
+      "▁conf irms",
+      "▁confirm s",
+      "鄂 尔多斯",
+      "▁pol yp",
+      "▁poly p",
+      "▁Pro ceedings",
+      "▁sand wich",
+      "▁sandw ich",
+      "原 生",
+      "气 势",
+      "高 消费",
+      "▁T itan",
+      "▁Ti tan",
+      "▁Tit an",
+      "▁Leban on",
+      "岁以下的儿童 在使用现有的床铺",
+      "▁ 있",
+      "教 你",
+      "比 值",
+      "▁expend iture",
+      "格 林",
+      "生 姜",
+      "▁输入 样例",
+      "▁ 输入样例",
+      "如 意",
+      "at ta",
+      "att a",
+      "LEV EL",
+      "ophil ic",
+      "▁F oster",
+      "▁Fo ster",
+      "Em ployee",
+      "脖 子",
+      "追 索",
+      "▁n ond",
+      "▁no nd",
+      "▁non d",
+      "att achment",
+      "attach ment",
+      "第二百六十二条第二款 之规定",
+      "自 豪",
+      "nThe ir",
+      "▁CLA IM",
+      "▁D ynam",
+      "▁Dy nam",
+      "▁refriger ator",
+      "家 伙",
+      "▁g lu",
+      "▁gl u",
+      "▁Commun ist",
+      "话 语",
+      "监狱 提出",
+      "sa les",
+      "sal es",
+      "sale s",
+      "s ales",
+      "依法 向被执行人",
+      "g d",
+      "× \\",
+      "寓 意",
+      "▁g ymn",
+      "▁gym n",
+      "▁gy mn",
+      "湘 潭",
+      "gu y",
+      "g uy",
+      "第六 十四",
+      "模型 的",
+      "模 型的",
+      "st re",
+      "str e",
+      "s tre",
+      "▁Ad vert",
+      "▁Adv ert",
+      "▁Mont gomery",
+      "▁T anz",
+      "▁Ta nz",
+      "▁Tan z",
+      "▁den oted",
+      "▁denote d",
+      "gr a",
+      "g ra",
+      "ac ry",
+      "acr y",
+      "och rom",
+      "o chrom",
+      "是 根据",
+      "}} }{",
+      "}}} {",
+      "} }}{",
+      "▁M ead",
+      "▁Me ad",
+      "um bled",
+      "umb led",
+      "umble d",
+      "第一 位",
+      "第 一位",
+      "du cers",
+      "duc ers",
+      "ducer s",
+      "年经教育部 正式批准",
+      "您 好",
+      "村 的",
+      "申 诉",
+      "▁t odd",
+      "▁to dd",
+      "▁拉丁 学名",
+      "一般 在",
+      "▁dis comfort",
+      "粘 贴",
+      "ch ars",
+      "char s",
+      "cha rs",
+      "▁e ditions",
+      "▁ed itions",
+      "▁edit ions",
+      "▁edition s",
+      "名 学生",
+      "▁che apest",
+      "▁cheap est",
+      "华 丽",
+      "喝 水",
+      "认定 的",
+      "认 定的",
+      "nJ ul",
+      "n Jul",
+      "▁text ures",
+      "▁texture s",
+      "▁tex tures",
+      "▁dend ritic",
+      "合作 的",
+      "合 作的",
+      "ad al",
+      "ada l",
+      "a dal",
+      "MET HO",
+      "M ETHO",
+      "▁uter ine",
+      "▁end oscopic",
+      "▁in activation",
+      "对 立",
+      "▁W S",
+      "▁ WS",
+      "re cv",
+      "rec v",
+      "▁s ang",
+      "▁sa ng",
+      "▁san g",
+      "oc ratic",
+      "ocr atic",
+      "Auto Field",
+      "祝 贺",
+      "疗 程",
+      "od ia",
+      "odi a",
+      "▁D aw",
+      "▁Da w",
+      "Requ ire",
+      "Re quire",
+      "ang ement",
+      "ange ment",
+      "#### ########",
+      "######## ####",
+      "###### ######",
+      "剧 烈",
+      "分布 式",
+      "▁X XX",
+      "▁XX X",
+      "▁ XXX",
+      "▁gl uten",
+      "▁glut en",
+      "▁glu ten",
+      "▁different iable",
+      "周 五",
+      "日 凌晨",
+      "}) \\\\)",
+      "})\\\\ )",
+      "} )\\\\)",
+      "主办 的",
+      "主 办的",
+      "▁Re spons",
+      "▁Res pons",
+      "▁Respon s",
+      "▁horm onal",
+      "▁key boards",
+      "▁keyboard s",
+      "完 工",
+      "组 和",
+      "un can",
+      "unc an",
+      "u ncan",
+      "▁Numer ical",
+      "▁cut aneous",
+      "▁ cutaneous",
+      "因为 它",
+      "▁wealth y",
+      "aw ks",
+      "awk s",
+      "an tes",
+      "ant es",
+      "ante s",
+      "▁输出 样例",
+      "▁ 输出样例",
+      "weight ed",
+      "▁mid night",
+      "▁ 等",
+      "第十 二",
+      "第 十二",
+      "Dep th",
+      "▁un incorporated",
+      "▁terror ism",
+      "o S",
+      "金 沙",
+      "di p",
+      "d ip",
+      "国内 的",
+      "国 内的",
+      "▁背景 设定",
+      "rad iol",
+      "radio l",
+      "▁counter part",
+      "▁psych osocial",
+      "▁psycho social",
+      "加入 了",
+      "加 入了",
+      "▁G or",
+      "▁Go r",
+      "瓶 颈",
+      "贵阳 市",
+      "贵 阳市",
+      "al ey",
+      "ale y",
+      "a ley",
+      "▁B ras",
+      "▁Br as",
+      "▁Bra s",
+      "▁her oes",
+      "▁hero es",
+      "▁constit utes",
+      "▁constitu tes",
+      "▁constitute s",
+      "BO OL",
+      "Li ter",
+      "Lite r",
+      "L iter",
+      "▁quant ification",
+      "合 力",
+      "更 强",
+      "正 版",
+      "AG ES",
+      "AGE S",
+      "义务 教育",
+      "▁sc aled",
+      "▁scale d",
+      "▁scal ed",
+      "▁ scaled",
+      "▁wond ered",
+      "▁wonder ed",
+      "▁cyt otoxicity",
+      "▁cytotox icity",
+      "▁cyto toxicity",
+      "▁cytotoxic ity",
+      "年 任",
+      "居 委会",
+      "▁Med icaid",
+      "▁Medic aid",
+      "改 成",
+      "^^ ^^",
+      "被执行人 未按",
+      "为 他",
+      "ent ric",
+      "▁A cid",
+      "▁Ac id",
+      "▁e ster",
+      "▁est er",
+      "▁es ter",
+      "▁este r",
+      "▁ ester",
+      "▁mi mic",
+      "▁mim ic",
+      "pr us",
+      "p rus",
+      "▁dem ocratic",
+      "▁democr atic",
+      "的一 致",
+      "的 一致",
+      "ko k",
+      "k ok",
+      "▁a te",
+      "▁at e",
+      "▁ ate",
+      "给出 解题",
+      "rid ges",
+      "ridge s",
+      "▁Scot ia",
+      "▁Nav igation",
+      "▁ Navigation",
+      "意 向",
+      "一般 都是",
+      "一般都 是",
+      "pr essive",
+      "press ive",
+      "p ressive",
+      "人力资源 和社会保障",
+      "▁con ve",
+      "▁conv e",
+      "▁Tour ism",
+      "Request Mapping",
+      "▁plan ting",
+      "▁plant ing",
+      "的 因素",
+      "湖北省 武汉市",
+      "▁sc andal",
+      "▁scan dal",
+      "潮 湿",
+      "me g",
+      "m eg",
+      "曹 操",
+      "Work s",
+      "о б",
+      "▁d ella",
+      "▁de lla",
+      "▁del la",
+      "▁dell a",
+      "sa le",
+      "sal e",
+      "s ale",
+      "五 星",
+      "av ar",
+      "ava r",
+      "a var",
+      "归还 原告",
+      "▁\" \")\\",
+      "▁\"\" )\\",
+      "▁\"\") \\",
+      "▁ \"\")\\",
+      "油 漆",
+      "医 务",
+      "搜 集",
+      "时 装",
+      "nEx amples",
+      "nExample s",
+      "n Examples",
+      "洗 手",
+      ">: \\\\(",
+      "> :\\\\(",
+      "aff eine",
+      "affe ine",
+      "与中国科学院 共同举办",
+      "落 户",
+      "做 不到",
+      "de mand",
+      "dem and",
+      "nst art",
+      "n start",
+      "桂 花",
+      "K ernel",
+      "甲 方",
+      "▁P b",
+      "▁ Pb",
+      "员工 的",
+      "]) :\\",
+      "]): \\",
+      "] ):\\",
+      "这样 才能",
+      "sc heduler",
+      "schedule r",
+      "小 子",
+      "自 制",
+      "nL og",
+      "nLo g",
+      "n Log",
+      "云中 书城",
+      "Alter Field",
+      "几 位",
+      "电 厂",
+      "一起 去",
+      "Ke ep",
+      "K eep",
+      "▁ex ogenous",
+      "数据 分析",
+      "▁c ough",
+      "▁co ugh",
+      "▁cou gh",
+      "us cular",
+      "usc ular",
+      "▁minim izing",
+      "▁Ind ependence",
+      "est ock",
+      "esto ck",
+      "e stock",
+      "▁wh isk",
+      "▁whis k",
+      "back ends",
+      "backend s",
+      "来 得",
+      "Cl aim",
+      "ring s",
+      "rin gs",
+      "r ings",
+      "▁G ent",
+      "▁Ge nt",
+      "▁Gen t",
+      "▁S wan",
+      "▁Sw an",
+      "▁Re ady",
+      "▁Read y",
+      "▁ Ready",
+      "▁im proper",
+      "▁impro per",
+      "▁s elenium",
+      "▁ selenium",
+      "第二百 三",
+      "ning s",
+      "nin gs",
+      "n ings",
+      "轻 伤",
+      "To r",
+      "T or",
+      "i Pad",
+      "for get",
+      "forge t",
+      "f orget",
+      "▁l adder",
+      "▁s ticks",
+      "▁st icks",
+      "▁stick s",
+      "十 位",
+      "迭 代",
+      "▁t ones",
+      "▁to nes",
+      "▁ton es",
+      "▁tone s",
+      "▁b logging",
+      "▁blog ging",
+      "四舍 五入",
+      "化 石",
+      "泰 山",
+      "质量 和",
+      "质 量和",
+      "Mus t",
+      "Mu st",
+      "M ust",
+      "stit ial",
+      "均 未",
+      "Ru b",
+      "R ub",
+      "比特 币",
+      "▁D ow",
+      "▁Do w",
+      "pass ed",
+      "pas sed",
+      "▁Bud get",
+      "▁con comit",
+      "es se",
+      "ess e",
+      "▁tr ump",
+      "社会 责任",
+      "先 算",
+      "递 归",
+      "中含 有",
+      "中 含有",
+      "中 兴",
+      "▁D ress",
+      "▁Dr ess",
+      "▁medi ators",
+      "▁mediator s",
+      "在 她",
+      "众 生",
+      "cy te",
+      "cyt e",
+      "c yte",
+      "进行了 调查",
+      "▁inn ocent",
+      "▁innoc ent",
+      "制 动",
+      "旅 程",
+      "Fl ex",
+      "F lex",
+      "oc cup",
+      "occ up",
+      "nSe ason",
+      "n Season",
+      "▁emb ark",
+      "▁adhes ive",
+      "▁supply ing",
+      "▁suppl ying",
+      "眼 球",
+      "安全 感",
+      "▁com plain",
+      "▁comp lain",
+      "▁compl ain",
+      "以 期",
+      "每个 人都",
+      "每个人 都",
+      "男 方",
+      "污水 处理",
+      "污 水处理",
+      "要求 离婚",
+      "看 不",
+      "现在 是",
+      "预订 取消",
+      "▁over expression",
+      "▁h obby",
+      "▁hob by",
+      "▁sw ift",
+      "▁ swift",
+      "▁In stant",
+      "▁Inst ant",
+      "▁ Instant",
+      "预付 政策",
+      ":\\ \\\\\\",
+      ":\\\\ \\\\",
+      ": \\\\\\\\",
+      "nIm age",
+      "n Image",
+      "▁m oles",
+      "▁mole s",
+      "▁mo les",
+      "▁mol es",
+      "的 可",
+      "ac ey",
+      "ace y",
+      "▁ к",
+      "▁ref ere",
+      "▁refer e",
+      "▁M agnetic",
+      "▁Mag netic",
+      "▁Magn etic",
+      "▁Magnet ic",
+      "届 时",
+      "OD Y",
+      "O DY",
+      "gr ass",
+      "gra ss",
+      "ind ust",
+      "im plicit",
+      "impl icit",
+      "▁spec ifies",
+      "▁ γ",
+      "开 朗",
+      "这是 我",
+      "这 是我",
+      "Assert ions",
+      "西 洋",
+      "Integr ation",
+      "▁adequ ately",
+      "▁adequate ly",
+      "个 大",
+      "税 率",
+      "in ston",
+      "ins ton",
+      "inst on",
+      "▁interfer e",
+      "▁interf ere",
+      "▁下 载",
+      "▁ 下载",
+      "lo dash",
+      "lod ash",
+      "另外 一个",
+      "易 懂",
+      "▁Eval uate",
+      "▁ Evaluate",
+      "▁dehydro genase",
+      "开 盘",
+      "请 参阅",
+      "al ia",
+      "ali a",
+      "a lia",
+      "▁In ventory",
+      "▁ Inventory",
+      "▁diss ertation",
+      "三 门",
+      "两 面",
+      "严 厉",
+      "全球 化",
+      "▁B ach",
+      "▁Ba ch",
+      "▁Bac h",
+      "yc in",
+      "y cin",
+      "Qu ick",
+      "笔记本 电脑",
+      "Ex pect",
+      "Exp ect",
+      "▁H aven",
+      "▁Ha ven",
+      "▁Have n",
+      "▁Hav en",
+      "▁illness es",
+      "▁asympt omatic",
+      "头 条",
+      "看 成",
+      "or de",
+      "ord e",
+      "东 县",
+      "▁ide ally",
+      "▁ideal ly",
+      "他 不",
+      "▁exc eeds",
+      "▁exceed s",
+      "延 安",
+      "In ner",
+      "re ports",
+      "rep orts",
+      "report s",
+      "项目 建设",
+      "▁U nits",
+      "▁Un its",
+      "▁Unit s",
+      "▁Uni ts",
+      "▁ Units",
+      "read line",
+      "▁prom otions",
+      "▁promot ions",
+      "▁promotion s",
+      "以后 的",
+      "以 后的",
+      "nR et",
+      "nRe t",
+      "n Ret",
+      "石家庄 市",
+      "▁show c",
+      "▁poly mers",
+      "▁polymer s",
+      "▁polym ers",
+      "听 过",
+      "▁i e",
+      "▁ ie",
+      "▁mamm ary",
+      "z d",
+      "室 的",
+      "捞 出",
+      "▁revolution ary",
+      "各种 各样的",
+      "AR G",
+      "A RG",
+      "\\' },\\",
+      "\\'} ,\\",
+      "\\ '},\\",
+      "photo s",
+      "phot os",
+      "LI BR",
+      "LIB R",
+      "om bie",
+      "omb ie",
+      "Tr acker",
+      "Track er",
+      "Dispatch er",
+      "} ）",
+      ";< /",
+      "; </",
+      "▁e vid",
+      "▁ev id",
+      "▁H app",
+      "▁Ha pp",
+      "▁an emia",
+      "同 胞",
+      "就 应该",
+      "nCh ild",
+      "n Child",
+      "▁firm ly",
+      "未履行 生效法律文书确定的义务",
+      "未 履行生效法律文书确定的义务",
+      "ua wei",
+      "▁length y",
+      "▁leng thy",
+      "ri ot",
+      "rio t",
+      "r iot",
+      "ord ering",
+      "order ing",
+      "orde ring",
+      "卡 通",
+      "只是 一",
+      "只 是一",
+      "竞争 对手",
+      "▁s ocks",
+      "▁so cks",
+      "▁soc ks",
+      "▁sock s",
+      "候选 人",
+      "这是 因为",
+      "这 是因为",
+      "▁游戏 大小",
+      "▁ 游戏大小",
+      "Be haviour",
+      "▁e mit",
+      "▁em it",
+      "▁ emit",
+      "儿童和 加床",
+      "镇 长",
+      "tr ip",
+      "tri p",
+      "t rip",
+      "酒店 位于",
+      "▁D ocs",
+      "▁Do cs",
+      "▁Doc s",
+      "▁ Docs",
+      "不 稳定",
+      "三包 服务",
+      "▁inv iting",
+      "▁R M",
+      "▁ RM",
+      "▁alter ation",
+      "▁alt eration",
+      "天 后",
+      "▁g um",
+      "▁gu m",
+      "点的 坐标",
+      "点 的坐标",
+      "▁d ash",
+      "▁da sh",
+      "▁das h",
+      "▁ dash",
+      "dig its",
+      "digit s",
+      "Vari ant",
+      "Var iant",
+      "有 毒",
+      "sa lt",
+      "sal t",
+      "s alt",
+      "▁ES PN",
+      "▁ESP N",
+      "Al t",
+      "A lt",
+      "却 没有",
+      "uc hi",
+      "uch i",
+      "u chi",
+      "il ion",
+      "ili on",
+      "ilio n",
+      "i lion",
+      "▁life s",
+      "▁lif es",
+      "▁pol ys",
+      "▁poly s",
+      "▁po lys",
+      "部门 的",
+      "部 门的",
+      "nC ode",
+      "nCo de",
+      "n Code",
+      "fa cing",
+      "fac ing",
+      "f acing",
+      "▁athlet ics",
+      "▁athletic s",
+      "▁vis cosity",
+      "nF amily",
+      "n Family",
+      "连 通",
+      "▁ex ported",
+      "▁exp orted",
+      "▁export ed",
+      "B ucket",
+      "榆 林",
+      "▁In ject",
+      "▁Inj ect",
+      "▁ Inject",
+      "运 河",
+      "的热 情",
+      "的 热情",
+      "\\\\ \">",
+      "\\\\\" >",
+      "ec al",
+      "eca l",
+      "e cal",
+      "g f",
+      "山 河",
+      "电 解",
+      "▁Y e",
+      "的 安装",
+      "类似 的",
+      "类 似的",
+      "La ng",
+      "L ang",
+      "VELO PM",
+      "▁phen otypic",
+      "管 教",
+      "▁s ie",
+      "▁si e",
+      "▁ 范",
+      "▁l av",
+      "▁la v",
+      "▁ lav",
+      "▁B ears",
+      "▁Be ars",
+      "▁Bear s",
+      "▁Bea rs",
+      "提起 行政诉讼",
+      "Oct ober",
+      "▁p x",
+      "▁ px",
+      "污染 物",
+      "▁s lim",
+      "▁sl im",
+      "▁histor ians",
+      "▁historian s",
+      "小 事",
+      "ru g",
+      "r ug",
+      "▁T D",
+      "▁ TD",
+      "el in",
+      "eli n",
+      "e lin",
+      "▁P ir",
+      "▁Pi r",
+      "▁H ein",
+      "▁He in",
+      "比较 高",
+      "比 较高",
+      "端 点",
+      "正方 体",
+      "被 申请执行人",
+      "求 值",
+      "阶 层",
+      "ie nte",
+      "ient e",
+      "ien te",
+      "i ente",
+      "on ne",
+      "onn e",
+      "变 态",
+      "帝 王",
+      "真 菌",
+      "ft p",
+      "f tp",
+      "ns td",
+      "nst d",
+      "n std",
+      "▁F irm",
+      "▁Fi rm",
+      "OUT PUT",
+      "▁l itter",
+      "▁lit ter",
+      "▁pass port",
+      "▁pas sport",
+      "▁ passport",
+      "北京 时间",
+      "final ly",
+      "fin ally",
+      "f inally",
+      "水 产",
+      "原 裁定",
+      "ic idal",
+      "ici dal",
+      "▁gu ards",
+      "▁guard s",
+      "lo o",
+      "l oo",
+      "in ventory",
+      "▁Monitor ing",
+      "预 备",
+      "▁R ams",
+      "▁Ram s",
+      "▁Ra ms",
+      "以下 有期徒刑",
+      "▁Territ ory",
+      "▁Terr itory",
+      "年 了",
+      "aque s",
+      "aq ues",
+      "aqu es",
+      "a ques",
+      "条 形",
+      "为 广大",
+      "表面 的",
+      "表 面的",
+      "造 就",
+      "▁e rupt",
+      "▁er upt",
+      "无 处",
+      "洁 净",
+      "te t",
+      "t et",
+      "▁be side",
+      "▁bes ide",
+      "▁Ap proved",
+      "▁Appro ved",
+      "R B",
+      "告知 书",
+      "▁Wy oming",
+      "str action",
+      "stract ion",
+      "stra ction",
+      "s traction",
+      "地 利",
+      "证 号",
+      "map sto",
+      "maps to",
+      "rand int",
+      "头 疼",
+      "CT V",
+      "C TV",
+      "Instance Of",
+      "守 法",
+      "寄 生",
+      "nM et",
+      "nMe t",
+      "n Met",
+      "计分 考核",
+      "▁pick up",
+      "不 一",
+      "出 轨",
+      "▁in voice",
+      "▁inv oice",
+      "▁ invoice",
+      "nB ur",
+      "n Bur",
+      "▁pre re",
+      "▁pr ere",
+      "anc ellation",
+      "ancel lation",
+      "ancell ation",
+      "上 手",
+      "▁C and",
+      "▁Can d",
+      "▁Ca nd",
+      "▁copy ing",
+      "▁cop ying",
+      "▁Av iation",
+      "▁Press ure",
+      "▁c her",
+      "▁ch er",
+      "▁che r",
+      "▁ cher",
+      "为 民",
+      "pa c",
+      "p ac",
+      "▁e u",
+      "▁ eu",
+      "开始 执行",
+      "▁C ork",
+      "▁Co rk",
+      "▁Cor k",
+      "tr acker",
+      "track er",
+      "▁irre ducible",
+      "效力 于",
+      "▁R ib",
+      "▁Ri b",
+      "value Of",
+      "data sets",
+      "dat asets",
+      "dataset s",
+      "datas ets",
+      "▁irr igation",
+      "背 叛",
+      "da s",
+      "d as",
+      "▁E B",
+      "▁ EB",
+      "旗下 的",
+      "旗 下的",
+      "降 临",
+      "该公司 员工",
+      "该 公司员工",
+      "上 半",
+      "的女 儿",
+      "的 女儿",
+      "好 玩",
+      "退 款",
+      "泉州 市",
+      "泉 州市",
+      "进入 到",
+      "布 尔",
+      "▁P ul",
+      "▁Pu l",
+      "▁B ryan",
+      "▁Br yan",
+      "▁Bry an",
+      "ort heastern",
+      "orth eastern",
+      "ortheast ern",
+      "he v",
+      "h ev",
+      "他们 是",
+      "▁ 功能",
+      "单调 递减",
+      "▁S hen",
+      "▁Sh en",
+      "▁She n",
+      "更重要 的是",
+      "更 重要的是",
+      "▁L ocated",
+      "▁Loc ated",
+      "▁im prison",
+      "万 家",
+      "▁st rap",
+      "▁str ap",
+      "▁ strap",
+      "又 一",
+      "发 烧",
+      "然后 把",
+      "专业 技术",
+      "景 色",
+      "课 时",
+      "▁con trace",
+      "▁cont race",
+      "▁contr ace",
+      "▁ab olished",
+      "三 项",
+      "丰 收",
+      "富 汗",
+      "ia bly",
+      "iab ly",
+      "i ably",
+      "AR CHAR",
+      "ARCH AR",
+      "\" [",
+      "翅 膀",
+      "pkg s",
+      "pk gs",
+      "中华人民共和国行政 强制法",
+      "背 光",
+      "上海 科技大学",
+      "▁Ad venture",
+      "▁Advent ure",
+      "共 鸣",
+      "▁Pl aza",
+      "▁Am ateur",
+      "▁prob able",
+      "严格 的",
+      "严 格的",
+      "li on",
+      "lio n",
+      "l ion",
+      "▁Orth odox",
+      "▁Measure ment",
+      "▁ Measurement",
+      "▁celebr ities",
+      "▁cele brities",
+      "▁Ass ume",
+      "▁stret ching",
+      "▁stretch ing",
+      "▁p ackets",
+      "▁pack ets",
+      "▁packet s",
+      "Bi n",
+      "B in",
+      "▁At tempt",
+      "▁Att empt",
+      "▁ Attempt",
+      "▁Not ification",
+      "▁ Notification",
+      "尽 力",
+      "着 他",
+      "▁L oop",
+      "▁Lo op",
+      "▁ Loop",
+      "▁Over view",
+      "▁ Overview",
+      "▁commun icating",
+      "▁communic ating",
+      "▁C av",
+      "▁Ca v",
+      "▁pol ype",
+      "▁poly pe",
+      "▁polyp e",
+      "os o",
+      "o so",
+      "新 西兰",
+      "re comm",
+      "rec omm",
+      "Bal ance",
+      "▁ 以",
+      "彼 得",
+      "大多数 人",
+      "▁C ritical",
+      "▁Crit ical",
+      "复 古",
+      "征 求",
+      "了下 来",
+      "了 下来",
+      "需要 在",
+      "需 要在",
+      "▁H od",
+      "▁Ho d",
+      "▁conc ise",
+      "等 财产",
+      "▁t ray",
+      "▁tr ay",
+      "▁tra y",
+      "With out",
+      "sup erv",
+      "super v",
+      "▁cul min",
+      "か ら",
+      "统计 局",
+      "an ca",
+      "anc a",
+      "美术 学院",
+      "▁inv oke",
+      "▁ invoke",
+      "▁dr illing",
+      "▁drill ing",
+      "就算 是",
+      "就 算是",
+      "PA RT",
+      "PAR T",
+      "P ART",
+      "副 秘书长",
+      "装 有",
+      "▁b aked",
+      "▁bake d",
+      "▁bak ed",
+      "▁l aying",
+      "▁la ying",
+      "▁lay ing",
+      "▁ter race",
+      "▁terr ace",
+      "▁recomm ends",
+      "▁recommend s",
+      "ans ible",
+      "ansi ble",
+      "▁roof ing",
+      "也 曾",
+      "恒 大",
+      "这首 歌",
+      "这 首歌",
+      "▁S ox",
+      "▁So x",
+      "ph rase",
+      "▁ar thro",
+      "▁arth ro",
+      "▁sp leen",
+      "▁spl een",
+      "▁acet ate",
+      "▁adv ancement",
+      "▁advance ment",
+      "Le arning",
+      "Learn ing",
+      "L earning",
+      "▁vit amins",
+      "▁vitamin s",
+      "▁therap ist",
+      "哪 家",
+      "商 量",
+      "ar ettes",
+      "are ttes",
+      "aret tes",
+      "arette s",
+      "cept ors",
+      "ceptor s",
+      "H a",
+      "红 枣",
+      "author ization",
+      "▁P OS",
+      "▁PO S",
+      "▁ POS",
+      "▁nec kl",
+      "▁neck l",
+      "▁eigen value",
+      "南 开",
+      "出色 的",
+      "出 色的",
+      "▁in def",
+      "▁i ndef",
+      "▁ind ef",
+      "则 有",
+      "nA D",
+      "n AD",
+      "▁m isc",
+      "▁mis c",
+      "▁mi sc",
+      "▁ misc",
+      "▁Z imbabwe",
+      "▁block ade",
+      "励 志",
+      "核 准",
+      "in voice",
+      "inv oice",
+      "▁plate lets",
+      "▁platelet s",
+      "杨某 某",
+      "杨 某某",
+      "ance ment",
+      "anc ement",
+      "▁ 陆",
+      "导 出",
+      "饰 品",
+      "▁Pr ague",
+      "▁Pra gue",
+      "▁immun odef",
+      "▁immuno def",
+      "▁di vergence",
+      "▁diver gence",
+      "ø r",
+      "车 子",
+      "RO W",
+      "R OW",
+      "▁V C",
+      "▁ VC",
+      "一次 方程",
+      "三角 函数",
+      "▁M issing",
+      "▁Miss ing",
+      "▁Mis sing",
+      "▁ Missing",
+      "▁epidem iology",
+      "毛 巾",
+      "▁lip oprotein",
+      "▁Design s",
+      "▁Oper ator",
+      "▁Op erator",
+      "▁Opera tor",
+      "▁ Operator",
+      "▁tre asure",
+      "▁treas ure",
+      "ograph ically",
+      "ographic ally",
+      "ographical ly",
+      "编 曲",
+      "ME S",
+      "M ES",
+      "兼 任",
+      "aw t",
+      "a wt",
+      "申请 书",
+      "进行 比较",
+      "▁A mer",
+      "▁Am er",
+      "并不 会",
+      "并 不会",
+      "▁B omb",
+      "▁Bo mb",
+      "▁Bom b",
+      "▁cr isp",
+      "▁cris p",
+      "阳市 中级人民法院",
+      "阳 市中级人民法院",
+      "早 点",
+      "\"} }",
+      "\" }}",
+      "▁a nnot",
+      "▁an not",
+      "▁ann ot",
+      "▁ annot",
+      "▁cl oses",
+      "▁clos es",
+      "▁close s",
+      "fa x",
+      "f ax",
+      "▁incons istent",
+      "履行 能力",
+      "第五 十三条",
+      "第五十 三条",
+      "ast olic",
+      "能 耗",
+      "年龄 的",
+      "了 呢",
+      "宽 松",
+      "di c",
+      "d ic",
+      "sol ve",
+      "s olve",
+      "▁discover ing",
+      "山 上",
+      "鼓 楼",
+      "det ect",
+      "二〇一六年 十二月",
+      "二〇一六年十二 月",
+      "二〇一六年十 二月",
+      "在那 里",
+      "在 那里",
+      "Sh ip",
+      "S hip",
+      "te lif",
+      "tel if",
+      "t elif",
+      "运 会",
+      "ne e",
+      "n ee",
+      "▁G PL",
+      "▁GP L",
+      "▁ GPL",
+      "Tw itter",
+      "T witter",
+      "E W",
+      "首 位",
+      "基本 信息",
+      "F oo",
+      "▁n urt",
+      "▁nur t",
+      "▁nu rt",
+      "▁ur ge",
+      "▁urg e",
+      "▁wor ries",
+      "快 餐",
+      "谈 到",
+      "Log o",
+      "Lo go",
+      "L ogo",
+      "is san",
+      "iss an",
+      "issa n",
+      "▁robust ness",
+      "im us",
+      "i mus",
+      "▁t ribe",
+      "▁tr ibe",
+      "▁tri be",
+      "▁trib e",
+      "▁nomin ee",
+      "▁nomine e",
+      "属于 自己的",
+      "▁l ending",
+      "▁len ding",
+      "▁lend ing",
+      "nO per",
+      "n Oper",
+      "符合法律规定 的",
+      "▁List Node",
+      "▁ ListNode",
+      "转 入",
+      "ud i",
+      "u di",
+      "IN PUT",
+      "▁R uth",
+      "▁Ru th",
+      "▁Rut h",
+      "海 滩",
+      "pm od",
+      "p mod",
+      "把自己 的",
+      "把 自己的",
+      "中 东",
+      "PL ICATION",
+      "必 将",
+      "站 的",
+      "il lar",
+      "ill ar",
+      "illa r",
+      "▁Graph ics",
+      "▁Graphic s",
+      "▁ Graphics",
+      "置 于",
+      "议 会",
+      "▁Jen kins",
+      "plic ative",
+      "▁ster oids",
+      "▁steroid s",
+      "TO R",
+      "T OR",
+      "ja y",
+      "j ay",
+      "绝 句",
+      "中国科学院 研究生院",
+      "环 比",
+      "▁Jul ie",
+      "▁Ju lie",
+      "▁ch ord",
+      "▁cho rd",
+      "▁chor d",
+      "▁phosph atase",
+      "▁phosphat ase",
+      "▁G mina",
+      "For ward",
+      "凭 什么",
+      "EN TER",
+      "ENT ER",
+      "tf unction",
+      "t function",
+      "RE AM",
+      "al ion",
+      "ali on",
+      "a lion",
+      "ot ent",
+      "ote nt",
+      "oten t",
+      "他 自己",
+      "ox in",
+      "o xin",
+      "▁r aid",
+      "▁ra id",
+      "归 属",
+      "钱 包",
+      "as sis",
+      "ass is",
+      "表 白",
+      "代表 作",
+      "om ycin",
+      "omy cin",
+      "diction ary",
+      "d ictionary",
+      "本题 的关键",
+      "▁b ride",
+      "▁br ide",
+      "▁brid e",
+      "▁g lands",
+      "▁gl ands",
+      "▁gland s",
+      "最高人民法院关于 审理",
+      "我 发现",
+      "▁Product ions",
+      "▁Production s",
+      "x z",
+      "▁O nInit",
+      "▁On Init",
+      "▁ OnInit",
+      "▁re solutions",
+      "▁res olutions",
+      "▁resolution s",
+      "再 现",
+      "▁你 的",
+      "▁ 你的",
+      "▁ve ins",
+      "▁vein s",
+      "▁kin ases",
+      "▁kinase s",
+      "▁h ooks",
+      "▁hook s",
+      "▁ho oks",
+      "▁ hooks",
+      "▁al lies",
+      "▁all ies",
+      "金 银",
+      "生产 力",
+      "▁r ug",
+      "▁ru g",
+      "▁ rug",
+      "恰 恰",
+      "有 能力",
+      "▁W inn",
+      "▁Win n",
+      "▁Wi nn",
+      "▁G ly",
+      "▁Gl y",
+      "▁ Gly",
+      "see ing",
+      ")_ {",
+      ") _{",
+      "sh i",
+      "s hi",
+      "id in",
+      "idi n",
+      "i din",
+      "▁b ending",
+      "▁ben ding",
+      "▁bend ing",
+      "▁al veolar",
+      "进 士",
+      "}) ^",
+      "} )^",
+      "en ry",
+      "▁Ad just",
+      "jo ur",
+      "j our",
+      "▁bright ness",
+      "▁question ed",
+      "▁quest ioned",
+      "| >",
+      "标志 着",
+      "\":\" \",\"",
+      "▁ret ire",
+      "显 现",
+      "semb ling",
+      "sem bling",
+      "sembl ing",
+      "▁_ \\",
+      "▁ _\\",
+      "使 之",
+      "烟 花",
+      "了一 份",
+      "了 一份",
+      "胰岛 素",
+      "▁Jeff rey",
+      "▁is omorphic",
+      "▁ 七",
+      "徐 某",
+      "红 尘",
+      "绽 放",
+      "▁p ant",
+      "▁pan t",
+      "▁pa nt",
+      "▁rad icals",
+      "▁radical s",
+      "▁South west",
+      "Serial ization",
+      "这 本",
+      "V el",
+      "▁cal ibr",
+      "pro blems",
+      "problem s",
+      "第三 个",
+      "第 三个",
+      "▁L eo",
+      "▁Le o",
+      "精 通",
+      "ov ic",
+      "ovi c",
+      "o vic",
+      "▁D ipl",
+      "▁Di pl",
+      "▁Dip l",
+      "ACC ESS",
+      "A CCESS",
+      "▁deal ers",
+      "▁dealer s",
+      "哺 乳",
+      "阿 富汗",
+      "min or",
+      "mi nor",
+      "▁int estine",
+      "▁form ations",
+      "▁format ions",
+      "▁formation s",
+      "▁ formations",
+      "技 法",
+      "白 发",
+      "是怎样 的",
+      "是 怎样的",
+      "App ly",
+      "Ap ply",
+      "us able",
+      "usa ble",
+      "学 长",
+      "— \\",
+      "红 柿",
+      "的可 能",
+      "的 可能",
+      "运动 队",
+      "to ut",
+      "t out",
+      "▁T win",
+      "▁Tw in",
+      "▁C ertain",
+      "▁Cert ain",
+      "▁Cer tain",
+      "MI C",
+      "M IC",
+      "Bi ll",
+      "B ill",
+      "▁n ap",
+      "▁na p",
+      "▁ nap",
+      "▁l asts",
+      "▁last s",
+      "▁la sts",
+      "▁las ts",
+      "▁web page",
+      "▁partner ed",
+      "克 隆",
+      "拌 匀",
+      "油 腻",
+      "GP U",
+      "G PU",
+      "nlog ger",
+      "n logger",
+      "创新 创业",
+      "ight house",
+      "igh thouse",
+      "ighth ouse",
+      "哭 了",
+      "▁l ub",
+      "▁lu b",
+      "▁y ii",
+      "▁ yii",
+      "抓获 经过",
+      "ca use",
+      "c ause",
+      "▁light ly",
+      "仅 为",
+      "▁per for",
+      "▁perf or",
+      "▁c ess",
+      "▁ce ss",
+      "▁ cess",
+      "▁C anon",
+      "▁Can on",
+      "▁Ca non",
+      "▁end ors",
+      "▁ endors",
+      "▁Invest ig",
+      "管理 者",
+      "经营 的",
+      "两 份",
+      "父母 的",
+      "▁end ocrine",
+      "▁ endocrine",
+      "年 夜",
+      "搬 迁",
+      "F un",
+      "▁Con tr",
+      "▁Cont r",
+      "▁Co ntr",
+      "国 学",
+      "▁n ails",
+      "▁na ils",
+      "▁nail s",
+      "▁initial ization",
+      "$ :",
+      "GN U",
+      "G NU",
+      "nPro perty",
+      "n Property",
+      "▁DEAL INGS",
+      "尾 巴",
+      "宣告 缓刑",
+      "▁hor iz",
+      "▁ho riz",
+      "问 答",
+      "Sch ool",
+      "S chool",
+      "Sing leton",
+      "Single ton",
+      "不 佳",
+      "壁 纸",
+      "外 来",
+      "acet yl",
+      "C urrency",
+      "及 照片",
+      "▁w it",
+      "▁wi t",
+      "▁ wit",
+      "ope nsource",
+      "open source",
+      "opens ource",
+      "应 注意",
+      "Al ias",
+      "Ali as",
+      "▁但 是",
+      "▁ 但是",
+      "▁Man it",
+      "▁Ma nit",
+      "▁th rive",
+      "▁thr ive",
+      "▁pathophys iology",
+      "星 辰",
+      "触 犯",
+      "华盛 顿",
+      "VELOPM ENT",
+      "合并 同类项",
+      "nin put",
+      "n input",
+      "油 烟",
+      "不 限",
+      "微 型",
+      "量 大",
+      "在 面对",
+      "▁推荐 菜",
+      "wa t",
+      "w at",
+      "短 视频",
+      "董事 会",
+      "Hy per",
+      "Hyp er",
+      "并 提出",
+      "av ia",
+      "avi a",
+      "a via",
+      "▁o un",
+      "▁ou n",
+      "▁ oun",
+      "▁e ukary",
+      "▁agon ists",
+      "▁agonist s",
+      "▁stand ings",
+      "▁standing s",
+      "殴 打",
+      "流动 性",
+      "民事权利 和",
+      "▁Com pl",
+      "▁Comp l",
+      "焦 作",
+      "老师 们",
+      "射 手",
+      "即可 解答",
+      "班 长",
+      "科 教",
+      "酸 奶",
+      "▁mel ting",
+      "▁melt ing",
+      "▁Roman ian",
+      "▁Rom anian",
+      "▁Romania n",
+      "ell ites",
+      "ellite s",
+      "elli tes",
+      "▁gu iding",
+      "▁guid ing",
+      "▁gui ding",
+      "维 尔",
+      "悬 挂",
+      "sim ilar",
+      "s imilar",
+      "▁traject ories",
+      "▁unsuccess ful",
+      "桌 子",
+      "访 谈",
+      "因 原告",
+      "开始 的",
+      "身份 信息",
+      "▁st airs",
+      "▁stair s",
+      "▁ stairs",
+      "在这 一",
+      "在 这一",
+      "Re pos",
+      "Rep os",
+      "Repo s",
+      "▁compl iant",
+      "▁imper ative",
+      "▁impe rative",
+      "IN SERT",
+      "INS ERT",
+      "位 移",
+      "嘴 巴",
+      "于 此",
+      "矿 物",
+      "太原 市",
+      "无 视",
+      "男人 的",
+      "男 人的",
+      "▁n at",
+      "▁na t",
+      "▁ nat",
+      "技术 咨询",
+      "四 海",
+      "家 务",
+      "报 销",
+      "▁W ide",
+      "▁Wi de",
+      "▁Wid e",
+      "▁Sh opping",
+      "▁Shop ping",
+      ")] )\\",
+      ") ])\\",
+      "aras htra",
+      "nS O",
+      "n SO",
+      "干 涉",
+      "▁scr ut",
+      "可以 考虑",
+      "后 记",
+      "魔 术",
+      "av ian",
+      "avi an",
+      "avia n",
+      "a vian",
+      "中 海",
+      "黄 昏",
+      "如果 想",
+      "▁酒店 星级",
+      "▁am end",
+      "▁amen d",
+      "List Item",
+      "做到 了",
+      "做 到了",
+      "的好 处",
+      "的 好处",
+      "er ent",
+      "ere nt",
+      "eren t",
+      "e rent",
+      "t R",
+      "用 车",
+      "re fn",
+      "ref n",
+      "运算 法则",
+      "运算法 则",
+      "Du ring",
+      "D uring",
+      "ot tle",
+      "ott le",
+      "o ttle",
+      "颠 覆",
+      "ch arts",
+      "char ts",
+      "chart s",
+      "▁Con cert",
+      "▁Conc ert",
+      "▁Concer t",
+      "编 织",
+      "数量 的",
+      "数 量的",
+      "To Props",
+      "机 能",
+      "这么 说",
+      "等 一系列",
+      "动 能",
+      "责令 其",
+      "不得 超过",
+      "▁Bu ndes",
+      "▁Bund es",
+      "▁Bun des",
+      "ja c",
+      "j ac",
+      "ku b",
+      "k ub",
+      "此题 主要",
+      "▁seroton in",
+      "▁п ро",
+      "▁t rio",
+      "▁tr io",
+      "▁tri o",
+      "▁Peters burg",
+      "的 程度",
+      "▁F em",
+      "▁Fe m",
+      "▁nic kel",
+      "▁nick el",
+      "ap proved",
+      "appro ved",
+      "中 南",
+      "▁R ails",
+      "▁Ra ils",
+      "▁Rail s",
+      "re marks",
+      "rem arks",
+      "remark s",
+      "▁std err",
+      "▁ stderr",
+      "游 玩",
+      "over view",
+      "nh e",
+      "n he",
+      "▁sh orts",
+      "▁short s",
+      "产 出",
+      "af a",
+      "a fa",
+      "nl ike",
+      "n like",
+      "▁h urd",
+      "▁hur d",
+      "▁hu rd",
+      "▁ 丹",
+      "擅 自",
+      "▁bi polar",
+      "▁bip olar",
+      "回 落",
+      "报 复",
+      "(\\' <",
+      "( \\'<",
+      "常务 委员会",
+      "▁ob serving",
+      "▁observ ing",
+      "▁obs erving",
+      "▁Dist ributed",
+      "r w",
+      "▁属 性",
+      "▁ 属性",
+      "的一 本",
+      "的 一本",
+      "▁Me chanical",
+      "▁Mechan ical",
+      "ie nto",
+      "ient o",
+      "ien to",
+      "i ento",
+      "Arch ive",
+      "Arc hive",
+      "▁s omatic",
+      "▁so matic",
+      "▁som atic",
+      "ub license",
+      "ublic ense",
+      "▁con formation",
+      "▁conform ation",
+      "▁ 崔",
+      "ro pic",
+      "rop ic",
+      "r opic",
+      "add itional",
+      "OR DER",
+      "ORD ER",
+      "▁p redis",
+      "▁pre dis",
+      "▁pred is",
+      "支 配",
+      "FI ELD",
+      "▁m ais",
+      "▁ma is",
+      "▁ne urom",
+      "▁neuro m",
+      "▁neur om",
+      "▁Pot ential",
+      "▁p orous",
+      "▁po rous",
+      "▁por ous",
+      "气 血",
+      "fe e",
+      "f ee",
+      "/\\ '\\",
+      "/\\' \\",
+      "/ \\'\\",
+      "▁C anal",
+      "▁Can al",
+      "▁Ca nal",
+      "nCom pet",
+      "nComp et",
+      "▁up ward",
+      "▁Brun swick",
+      "机动车交通事故责任 纠纷一案",
+      "普 陀",
+      "▁sh irts",
+      "▁shirt s",
+      "▁ shirts",
+      "反 转",
+      "省 钱",
+      "}} _{",
+      "} }_{",
+      "威 海",
+      "▁f lee",
+      "▁fl ee",
+      "▁fle e",
+      "▁lif ted",
+      "▁lift ed",
+      "▁low ered",
+      "▁lower ed",
+      "part ition",
+      "萎 缩",
+      "要 多",
+      "托 管",
+      "py pi",
+      "pyp i",
+      "告诉 记者",
+      "其前身为 中国科学院研究生院",
+      "毛 重",
+      "Ev en",
+      "E ven",
+      "书面 申请",
+      "▁v entric",
+      "▁vent ric",
+      "▁susp icious",
+      "S y",
+      "的 请求",
+      "▁sc affold",
+      "Kin g",
+      "Ki ng",
+      "K ing",
+      "Menu Item",
+      "▁Divis ión",
+      "▁sec urities",
+      "▁cylind rical",
+      "]; //",
+      "] ;//",
+      "▁It ems",
+      "▁Item s",
+      "▁ Items",
+      "该 书",
+      "SC s",
+      "S Cs",
+      "▁e en",
+      "▁ een",
+      "饮 酒",
+      "▁quant ified",
+      "▁Inter active",
+      "从 头",
+      "团 长",
+      "ul ae",
+      "ula e",
+      "▁b ee",
+      "▁be e",
+      "▁ bee",
+      "In g",
+      "I ng",
+      "▁M ED",
+      "▁ME D",
+      "▁ MED",
+      "▁b ored",
+      "▁bo red",
+      "▁bore d",
+      "▁bor ed",
+      "十二 条第一款",
+      "十二条 第一款",
+      "▁b illing",
+      "▁bill ing",
+      "▁bil ling",
+      "▁U T",
+      "▁ UT",
+      "不足 以",
+      "不 足以",
+      "se ven",
+      "sev en",
+      "s even",
+      "▁St ir",
+      "se o",
+      "s eo",
+      "▁pr ayers",
+      "▁pra yers",
+      "▁prayer s",
+      "▁pray ers",
+      "数 百",
+      "ow l",
+      "o wl",
+      "ast ore",
+      "a store",
+      "洪 水",
+      "百 花",
+      "▁rob otic",
+      "▁robot ic",
+      "Y I",
+      "侦 探",
+      "▁bl ah",
+      "▁liter al",
+      "▁lit eral",
+      "▁ literal",
+      "两 只",
+      "▁D b",
+      "▁ Db",
+      "ph is",
+      "phi s",
+      "p his",
+      "ain es",
+      "ai nes",
+      "aine s",
+      "a ines",
+      "三 亚",
+      "▁` \\",
+      "▁ `\\",
+      "▁chrom osomes",
+      "▁chromos omes",
+      "▁chromosome s",
+      "人 次",
+      "助 攻",
+      "▁P om",
+      "▁Po m",
+      "游 击",
+      "▁pro jective",
+      "▁project ive",
+      "▁proj ective",
+      "▁To pics",
+      "▁Top ics",
+      "▁Topic s",
+      "住所地 上海市",
+      "从 容",
+      "od or",
+      "odo r",
+      "o dor",
+      "▁pos ters",
+      "▁post ers",
+      "▁po sters",
+      "▁poster s",
+      ")) $",
+      ") )$",
+      "F our",
+      "十 堰",
+      "生 抽",
+      "在家 里",
+      "在 家里",
+      "af ood",
+      "a food",
+      "▁t its",
+      "▁tit s",
+      "▁ti ts",
+      "▁T ickets",
+      "▁Ticket s",
+      "▁O A",
+      "▁ OA",
+      "nS ch",
+      "nSc h",
+      "n Sch",
+      "▁index ed",
+      "既 往",
+      "流 入",
+      "学生 会",
+      "▁anomal ies",
+      "▁inter im",
+      "面临 着",
+      "▁Col oring",
+      "▁Color ing",
+      "▁we aken",
+      "▁weak en",
+      "受 人",
+      "最 喜欢",
+      "▁flow ering",
+      "▁flower ing",
+      "▁constitu encies",
+      "stim ulated",
+      "ha ul",
+      "h aul",
+      "以非法 占有",
+      "▁Pak istani",
+      "▁Pakistan i",
+      "▁Tem perature",
+      "▁ Temperature",
+      "▁facilit ated",
+      "▁facilitate d",
+      "▁E uclidean",
+      "▁Euclid ean",
+      "帮我 做",
+      "帮 我做",
+      "▁S ul",
+      "▁Su l",
+      "▁p oles",
+      "▁pol es",
+      "▁po les",
+      "▁pole s",
+      "▁pat ron",
+      "▁pa tron",
+      "第二百四十 四条",
+      "▁R GB",
+      "▁ RGB",
+      "▁habit ats",
+      "▁habitat s",
+      "无 烟",
+      "co b",
+      "c ob",
+      "学 到",
+      "▁中文 学名",
+      "▁Pl asma",
+      "▁Plas ma",
+      "制 剂",
+      "日本 人",
+      "日 本人",
+      "is tro",
+      "ist ro",
+      "istr o",
+      "ier archy",
+      "ierarch y",
+      "常 熟",
+      "多样 性",
+      "ad in",
+      "adi n",
+      "a din",
+      "▁S ensor",
+      "▁Sens or",
+      "▁ Sensor",
+      "▁micro grams",
+      "拉 伸",
+      "灰 尘",
+      "实用 性",
+      "相互 作用",
+      "▁s ublicense",
+      "▁sub license",
+      "是 把",
+      "▁Ven ice",
+      "▁Ve nice",
+      "▁Inter view",
+      "so f",
+      "s of",
+      "高 性能",
+      "nb ody",
+      "n body",
+      "▁good ness",
+      "医院 的",
+      "医 院的",
+      "VER Y",
+      "VE RY",
+      "V ERY",
+      "▁spons orship",
+      "▁sponsor ship",
+      "▁sponsors hip",
+      "▁O z",
+      "erc ises",
+      "ercise s",
+      "▁NON INFRINGEMENT",
+      "▁L ynn",
+      "▁Ly nn",
+      "▁Lyn n",
+      "▁Art icles",
+      "▁Article s",
+      "▁ Articles",
+      "▁enc ounters",
+      "▁encounter s",
+      "急 诊",
+      "鸿 蒙",
+      "ug a",
+      "u ga",
+      "▁C rew",
+      "▁Cr ew",
+      "▁Cre w",
+      "Observ able",
+      "河 道",
+      "代表 的",
+      "代 表的",
+      "vir us",
+      "vi rus",
+      "v irus",
+      "▁Ass ert",
+      "▁ Assert",
+      "吃 到",
+      "yd ia",
+      "T ITLE",
+      "▁attach ments",
+      "▁attachment s",
+      "▁ attachments",
+      "修 饰",
+      "抗 战",
+      "ka t",
+      "k at",
+      "at hi",
+      "ath i",
+      "a thi",
+      "蔓 延",
+      "黎 明",
+      "Ch unk",
+      "▁GO OD",
+      "冻结 被执行人",
+      "▁k otlin",
+      "▁ kotlin",
+      "进 的",
+      "▁Aut hentication",
+      "▁Authent ication",
+      "▁ Authentication",
+      "▁ 花",
+      "▁A id",
+      "杂 质",
+      "在本 案",
+      "在 本案",
+      "▁app rent",
+      "▁appre nt",
+      "检 疫",
+      "非常 高兴",
+      "非常高 兴",
+      "): \\\\(",
+      ") :\\\\(",
+      "▁aff ine",
+      "▁af fine",
+      "nR ichard",
+      "n Richard",
+      "重 审",
+      "]: \\\\(",
+      "] :\\\\(",
+      "nv ector",
+      "n vector",
+      "▁Integr ated",
+      "▁sc ared",
+      "▁scar ed",
+      "▁scare d",
+      "▁inv ented",
+      "▁invent ed",
+      "▁trans mitter",
+      "▁transmit ter",
+      "▁C G",
+      "▁ CG",
+      "sol id",
+      "s olid",
+      "可以 自收到",
+      "报告财产 令",
+      "或 有",
+      "管 家",
+      "nP rior",
+      "nPr ior",
+      "▁Muh ammad",
+      "▁file Name",
+      "▁ fileName",
+      "find er",
+      "fin der",
+      "fi nder",
+      "f inder",
+      "说 服",
+      "▁ON LY",
+      "▁ ONLY",
+      "▁java script",
+      "▁jav ascript",
+      "▁ javascript",
+      "▁carbohyd rate",
+      "宝 鸡",
+      "条 约",
+      "▁reg imens",
+      "▁regime ns",
+      "▁regimen s",
+      "▁be verage",
+      "▁ 香",
+      "▁ch ond",
+      "▁cho nd",
+      "tool tip",
+      "▁Lie utenant",
+      "丝 绸",
+      "主 干",
+      "会 所",
+      "其 所",
+      "提供 在线",
+      "nSpe aking",
+      "▁uncertain ties",
+      "决 字",
+      "影 院",
+      "um bs",
+      "umb s",
+      "St orm",
+      "▁C hev",
+      "▁Ch ev",
+      "▁Che v",
+      "定义 域",
+      "由题意 得",
+      "On Click",
+      "O nClick",
+      "▁ide als",
+      "▁idea ls",
+      "▁ideal s",
+      "▁qu icker",
+      "▁quick er",
+      "ER O",
+      "E RO",
+      "控制 的",
+      "控 制的",
+      "ch ard",
+      "char d",
+      "cha rd",
+      "c hard",
+      "er sen",
+      "ers en",
+      "mon go",
+      "mo ngo",
+      "mong o",
+      "m ongo",
+      "▁rational e",
+      "▁ration ale",
+      "后 又",
+      "PE S",
+      "P ES",
+      "Em itter",
+      "E mitter",
+      "▁生 长",
+      "▁ 生长",
+      "]] (/",
+      "] ](/",
+      "en to",
+      "ent o",
+      "e nto",
+      "we ed",
+      "w eed",
+      "名 叫",
+      "奥 迪",
+      "昌 县",
+      "内角 和",
+      "内 角和",
+      "▁angi ogenesis",
+      "二 分",
+      "she ets",
+      "sheet s",
+      "s heets",
+      "▁inter ven",
+      "le vard",
+      "lev ard",
+      "了 你",
+      "nR ef",
+      "nRe f",
+      "n Ref",
+      "▁B ert",
+      "▁Be rt",
+      "▁Ber t",
+      "▁crow ds",
+      "▁crowd s",
+      "领导 下",
+      "▁ana est",
+      "▁ve ntil",
+      "▁vent il",
+      "▁ven til",
+      "做 为",
+      "以非法占有 为目的",
+      "nComp anies",
+      "▁ 姚",
+      "裤 子",
+      "nM ass",
+      "nMa ss",
+      "n Mass",
+      "ns ome",
+      "nso me",
+      "n some",
+      "▁s plic",
+      "▁sp lic",
+      "▁spl ic",
+      "こ と",
+      "说 一下",
+      "平方 差",
+      "平 方差",
+      "等 功能",
+      "代理 检察员",
+      "▁b ou",
+      "▁bo u",
+      "行业 协会",
+      "ph alt",
+      "pha lt",
+      "phal t",
+      "▁R alph",
+      "col lections",
+      "collection s",
+      "coll ections",
+      "collect ions",
+      "系统 和",
+      "多 方",
+      "弱 势",
+      "▁这 个",
+      "▁ 这个",
+      "▁A sc",
+      "▁As c",
+      "▁ Asc",
+      "EV ENT",
+      "E VENT",
+      "▁incent ive",
+      "▁in version",
+      "▁i nversion",
+      "▁inv ersion",
+      "日 因",
+      "魔 龙",
+      "b w",
+      "课 外",
+      "yn a",
+      "y na",
+      "▁[^ ](/",
+      "nTh omas",
+      "n Thomas",
+      "RE C",
+      "R EC",
+      "br ight",
+      "b right",
+      "▁disag ree",
+      "> ﹣",
+      "▁ 现在",
+      "促进 了",
+      "促 进了",
+      "面积 公式",
+      "ne ver",
+      "n ever",
+      "▁gen otypes",
+      "▁genotype s",
+      "参 谋",
+      "头 皮",
+      "$( \"",
+      "$ (\"",
+      "ograph ed",
+      "ograp hed",
+      "▁over weight",
+      "▁overw eight",
+      "Bit map",
+      "▁all eles",
+      "▁alle les",
+      "▁allele s",
+      "分解 因式",
+      "▁Elect ronics",
+      "▁Electron ics",
+      "▁Electronic s",
+      "▁d ps",
+      "▁dp s",
+      "民营 企业",
+      "nP lay",
+      "nPl ay",
+      "n Play",
+      "▁W iki",
+      "▁Wi ki",
+      "▁Wik i",
+      "▁ Wiki",
+      "▁C ards",
+      "▁Car ds",
+      "▁Card s",
+      "▁ Cards",
+      "▁over lay",
+      "▁overl ay",
+      "▁ overlay",
+      "▁act ivating",
+      "▁activ ating",
+      "▁ad missions",
+      "▁admission s",
+      "▁adm issions",
+      "十 个",
+      "页 数",
+      "▁f ax",
+      "▁fa x",
+      "▁ fax",
+      "在刑罚 执行期间",
+      "相 声",
+      "nD O",
+      "n DO",
+      "婴 幼儿",
+      "ic ia",
+      "ici a",
+      "i cia",
+      "ag land",
+      "▁time zone",
+      "▁ timezone",
+      "▁Re tro",
+      "▁Ret ro",
+      "绑 架",
+      "和 企业",
+      "是 美国",
+      "EN CES",
+      "ENCE S",
+      "▁Subject s",
+      "| ^",
+      "临 近",
+      "癫痫 病",
+      "ant ib",
+      "anti b",
+      "文 体",
+      "▁r ider",
+      "▁ri der",
+      "▁ride r",
+      "▁rid er",
+      "很 明显",
+      "nt ers",
+      "nter s",
+      "nte rs",
+      "n ters",
+      "▁S uff",
+      "▁Su ff",
+      "▁out lines",
+      "▁outline s",
+      "实 效",
+      "au f",
+      "a uf",
+      "▁B rent",
+      "▁Br ent",
+      "▁Bre nt",
+      "▁Bren t",
+      "强 力",
+      "邪 恶",
+      "▁y r",
+      "▁ yr",
+      "▁d st",
+      "▁ds t",
+      "▁ dst",
+      "T uple",
+      "us tin",
+      "ust in",
+      "▁scal able",
+      "▁scala ble",
+      "▁Play Station",
+      "▁m ins",
+      "▁min s",
+      "▁mi ns",
+      "▁ mins",
+      "解题 的关键",
+      "复 试",
+      "就 可",
+      "赤峰 市",
+      "孙 子",
+      "ij a",
+      "i ja",
+      "▁Organ ic",
+      "▁Org anic",
+      "▁method ological",
+      "nb ecause",
+      "nbe cause",
+      "n because",
+      "nS elected",
+      "nSe lected",
+      "nSelect ed",
+      "n Selected",
+      "▁out dated",
+      "ad b",
+      "a db",
+      "急 需",
+      "▁gr ill",
+      "▁gri ll",
+      "酌情 从轻处罚",
+      "▁an onym",
+      "迷 失",
+      "nJ un",
+      "n Jun",
+      "未成年 人",
+      "未 成年人",
+      "偏 偏",
+      "的 非",
+      "总结 了",
+      ")< /",
+      ") </",
+      "▁E lite",
+      "▁El ite",
+      "▁Eli te",
+      "( |",
+      "▁E ST",
+      "▁ES T",
+      "▁ EST",
+      "▁f utures",
+      "▁future s",
+      "▁fu tures",
+      "▁fut ures",
+      "山 地",
+      "go al",
+      "oi t",
+      "o it",
+      "年轻 的",
+      "的 含量",
+      "葡萄 牙",
+      "期 末",
+      "Z ip",
+      "De cor",
+      "Dec or",
+      "Ba s",
+      "B as",
+      "nS k",
+      "n Sk",
+      "▁gl oves",
+      "▁glo ves",
+      "▁theore tically",
+      "▁theoretical ly",
+      "▁theor etically",
+      "家庭 的",
+      "家 庭的",
+      "语 气",
+      "类 小说",
+      "民事权利和 诉讼权利",
+      "传 来",
+      "nD E",
+      "n DE",
+      "]] )\\",
+      "]]) \\",
+      "] ])\\",
+      "pr ep",
+      "pre p",
+      "p rep",
+      "nJ ack",
+      "n Jack",
+      "yp ical",
+      "▁C olin",
+      "▁Col in",
+      "▁Co lin",
+      "▁fr ontal",
+      "▁front al",
+      "▁ frontal",
+      "normal ize",
+      "悟 空",
+      "▁P ortal",
+      "▁Port al",
+      "▁Por tal",
+      "▁ Portal",
+      "▁att acking",
+      "▁attack ing",
+      "▁irr ational",
+      ":< /",
+      ": </",
+      "▁K yle",
+      "▁Ky le",
+      "PU B",
+      "P UB",
+      "大家 在",
+      "▁econ omies",
+      "▁econom ies",
+      "早 晚",
+      "▁gly ph",
+      "▁ glyph",
+      "小 结",
+      "所 未",
+      "nf n",
+      "n fn",
+      "at urated",
+      "atur ated",
+      "atu rated",
+      "atura ted",
+      "加 西亚",
+      "融 券",
+      "pr ef",
+      "pre f",
+      "p ref",
+      "单项 式",
+      "▁awk ward",
+      "低 位",
+      "ER P",
+      "E RP",
+      "第一 批",
+      "第 一批",
+      "re ads",
+      "read s",
+      "rea ds",
+      "▁d amp",
+      "▁dam p",
+      "▁da mp",
+      "▁w atches",
+      "▁watch es",
+      "▁wat ches",
+      "三 款",
+      "ad h",
+      "a dh",
+      "工艺 品",
+      "() ),",
+      "()) ,",
+      "( )),",
+      "子 弹",
+      "宝 贵",
+      "st rom",
+      "str om",
+      "工 学",
+      "恭 喜",
+      "м и",
+      "阵 地",
+      "ut c",
+      "u tc",
+      "小时 内",
+      "U r",
+      "第十 四",
+      "第 十四",
+      "AS C",
+      "A SC",
+      "ib o",
+      "i bo",
+      "▁term inated",
+      "▁terminate d",
+      "pro g",
+      "pr og",
+      "p rog",
+      "tr acking",
+      "track ing",
+      "▁screen shots",
+      "▁screens hots",
+      "▁screenshot s",
+      "增 高",
+      "我 用",
+      "杜 绝",
+      "西 风",
+      "来看 看",
+      "来 看看",
+      "as so",
+      "ass o",
+      "▁t id",
+      "▁ti d",
+      "▁ tid",
+      "大 兴",
+      "mu tex",
+      "mut ex",
+      "▁c ollar",
+      "▁col lar",
+      "▁coll ar",
+      "▁pair ing",
+      "▁pa iring",
+      "网络 的",
+      "▁S ymphony",
+      "▁Sym phony",
+      "不 定",
+      "请 在",
+      "金 色",
+      "▁f og",
+      "▁fo g",
+      "及其 委托代理人",
+      "▁hyper tensive",
+      "▁hypert ensive",
+      "已 是",
+      "而 被",
+      "▁s arc",
+      "▁sa rc",
+      "▁sar c",
+      "▁psych iat",
+      "高 潮",
+      "E ast",
+      "科学 研究",
+      "科 学研究",
+      "▁de ny",
+      "▁den y",
+      "▁Der by",
+      "ycler View",
+      "▁Amb assador",
+      "bl r",
+      "b lr",
+      "▁c hi",
+      "▁ch i",
+      "▁ chi",
+      "先进 集体",
+      "医疗 机构",
+      "nStud ents",
+      "n Students",
+      "prehens ive",
+      "相 爱",
+      "的人 都",
+      "的 人都",
+      "▁hospital ization",
+      "日 用",
+      "棋 牌",
+      "▁M ü",
+      "un less",
+      "▁pione er",
+      "rp m",
+      "r pm",
+      "pi es",
+      "pie s",
+      "p ies",
+      "也 越来越",
+      "Z Z",
+      "副总 裁",
+      "副 总裁",
+      "▁ar gc",
+      "▁arg c",
+      "▁ argc",
+      "科技 有限公司于",
+      "科技有限公司 于",
+      "Conn ector",
+      "Connect or",
+      "（ ）",
+      "ed i",
+      "e di",
+      "▁p olished",
+      "▁pol ished",
+      "▁po lished",
+      "▁polish ed",
+      "▁in visible",
+      "▁inv isible",
+      "Map s",
+      "Ma ps",
+      "M aps",
+      "ph on",
+      "p hon",
+      "nCh arts",
+      "nChar ts",
+      "nChart s",
+      "▁we aker",
+      "▁weak er",
+      "▁aden osine",
+      "再 由",
+      "糖 果",
+      "威 尼斯",
+      "▁Sc heme",
+      "▁Sch eme",
+      "▁Sche me",
+      "▁ Scheme",
+      "nd ist",
+      "ndis t",
+      "ndi st",
+      "n dist",
+      "适用法律 正确",
+      "cre ation",
+      "creat ion",
+      "c reation",
+      "威 廉",
+      "▁a k",
+      "▁ ak",
+      "网络 上",
+      "▁s es",
+      "▁se s",
+      "▁ ses",
+      "▁ 因为",
+      "▁fort une",
+      "t j",
+      "UL D",
+      "U LD",
+      "二十 年",
+      "二 十年",
+      "解决 办法",
+      "az ione",
+      "azi one",
+      "▁P rodu",
+      "▁Pro du",
+      "▁ Produ",
+      "▁com prise",
+      "▁comp rise",
+      "▁compr ise",
+      "ni ne",
+      "nin e",
+      "n ine",
+      "part icle",
+      "p article",
+      "时 钟",
+      "▁t oe",
+      "▁to e",
+      "▁p encil",
+      "打 败",
+      "裂 缝",
+      "量刑 建议",
+      "▁y acht",
+      "▁ya cht",
+      "餐 桌",
+      "ac ia",
+      "aci a",
+      "a cia",
+      "at ra",
+      "a tra",
+      "有关 单位",
+      "tb order",
+      "t border",
+      "元 年",
+      "nA ut",
+      "n Aut",
+      "ul ner",
+      "us ually",
+      "▁prov inces",
+      "▁province s",
+      "▁thr iller",
+      "▁thrill er",
+      "▁biom arker",
+      "▁bio marker",
+      "▁biomark er",
+      "Re q",
+      "R eq",
+      "▁redund ant",
+      "▁term inology",
+      "锅 内",
+      "▁N SW",
+      "▁NS W",
+      "微 观",
+      "登 上",
+      "LO B",
+      "L OB",
+      "提交 了",
+      "▁M is",
+      "▁Mi s",
+      "▁B obby",
+      "▁Bob by",
+      "▁intrig uing",
+      "千 秋",
+      "赌 场",
+      "空间 的",
+      "空 间的",
+      "ix es",
+      "▁b ush",
+      "▁bu sh",
+      "▁bus h",
+      "▁ 巴",
+      "LE X",
+      "L EX",
+      "起到 了",
+      "起 到了",
+      "query set",
+      "quer yset",
+      "▁mot ility",
+      "▁mo tility",
+      "遍 布",
+      "Ho p",
+      "H op",
+      "▁instrument ation",
+      "拖 拉",
+      "Bu f",
+      "B uf",
+      "试 剂",
+      "用户 名",
+      "用 户名",
+      "▁p id",
+      "▁pi d",
+      "▁ pid",
+      "复杂 选择题",
+      "商 户",
+      "Widget s",
+      "▁miss ile",
+      "本院依法组成合议庭 对本案",
+      "新 村",
+      "青 铜",
+      "发生 率",
+      "▁academ ics",
+      "▁academic s",
+      "说 完",
+      "ew s",
+      "e ws",
+      "原有 的",
+      "原 有的",
+      "or ta",
+      "ort a",
+      "oint s",
+      "oin ts",
+      "o ints",
+      "也不 错",
+      "也 不错",
+      "nw ould",
+      "n would",
+      "type def",
+      "typed ef",
+      "ul arity",
+      "ular ity",
+      "nP layers",
+      "nPl ayers",
+      "nPlay ers",
+      "n Players",
+      "表 述",
+      "Ge o",
+      "G eo",
+      "的 土地",
+      "丰 县",
+      "果 断",
+      "那 我",
+      "可得 答案",
+      "▁Joh ns",
+      "▁John s",
+      "一 世",
+      "▁c osine",
+      "▁cos ine",
+      "▁under take",
+      "▁undert ake",
+      "▁epigen etic",
+      "c g",
+      "水 资源",
+      "▁a ids",
+      "▁aid s",
+      "▁ai ds",
+      "为客户 提供",
+      "为 客户提供",
+      "▁c ached",
+      "▁ca ched",
+      "▁cache d",
+      "▁ cached",
+      "出生 地",
+      "发改 委",
+      "nT itle",
+      "n Title",
+      "DO C",
+      "D OC",
+      "▁H unting",
+      "▁Hun ting",
+      "▁Hunt ing",
+      "Ne g",
+      "N eg",
+      "og e",
+      "o ge",
+      "领导 的",
+      "ab al",
+      "aba l",
+      "a bal",
+      "选项中挑 选出",
+      "▁ λ",
+      "之 王",
+      "后 经",
+      "▁p ir",
+      "▁pi r",
+      "▁ pir",
+      "▁con fer",
+      "▁conf er",
+      "▁Ray mond",
+      "▁D H",
+      "▁ DH",
+      "tprint f",
+      "t printf",
+      "▁u d",
+      "▁ ud",
+      "椭圆 形",
+      "椭 圆形",
+      "▁tat too",
+      "▁re plied",
+      "▁s ensation",
+      "▁sens ation",
+      "▁recogn izes",
+      "▁recognize s",
+      "IM G",
+      "I MG",
+      "喜爱 的",
+      "喜 爱的",
+      "公民 身份号码",
+      "▁mar athon",
+      ": ,",
+      "v p",
+      "一 笔",
+      "▁ 编辑",
+      "Jan uary",
+      "▁cat ering",
+      "▁cater ing",
+      "▁cate ring",
+      "Status Code",
+      "▁s ap",
+      "▁sa p",
+      "▁ sap",
+      "▁factor ization",
+      "全 会",
+      "板 上",
+      "▁real ise",
+      "团 的",
+      "找 我",
+      ")\\\\ )\\",
+      ")\\\\) \\",
+      ") \\\\)\\",
+      "▁Ke ys",
+      "▁Key s",
+      "▁ Keys",
+      "enc oder",
+      "encode r",
+      "vis ibility",
+      "▁Pub lisher",
+      "▁Publish er",
+      "▁ Publisher",
+      "▁Col lections",
+      "▁Coll ections",
+      "▁Collection s",
+      "▁Collect ions",
+      "▁ Collections",
+      "吉 利",
+      "性 状",
+      "▁ph ag",
+      "▁fr agr",
+      "▁frag r",
+      "▁fra gr",
+      "▁superv ised",
+      "▁ supervised",
+      "肢 体",
+      "nP eter",
+      "nPe ter",
+      "nPet er",
+      "n Peter",
+      "▁financial ly",
+      "▁financ ially",
+      "衬 衫",
+      "▁S ugg",
+      "▁Su gg",
+      "▁ Sugg",
+      "▁B asin",
+      "▁Bas in",
+      "▁Ba sin",
+      "▁f lora",
+      "▁fl ora",
+      "▁flo ra",
+      "▁flor a",
+      "道 题",
+      "分钟 左右",
+      "tab ular",
+      "▁R av",
+      "▁Ra v",
+      "▁L ights",
+      "▁Light s",
+      "玛 丽",
+      "▁F X",
+      "▁ FX",
+      "▁ir res",
+      "▁irre s",
+      "▁irr es",
+      "Se ptember",
+      "特 长",
+      "坚持 以",
+      "mor t",
+      "mo rt",
+      "m ort",
+      "Aut om",
+      "Auto m",
+      "Au tom",
+      "▁Jam ie",
+      "nB razil",
+      "▁su pers",
+      "▁sup ers",
+      "▁super s",
+      "通 车",
+      "成年 人",
+      "TA BLE",
+      "T ABLE",
+      "绘 本",
+      "落 败",
+      "大 城市",
+      "的 文件",
+      "治疗 方法",
+      "re ferences",
+      "reference s",
+      "refer ences",
+      "r eferences",
+      "万 多",
+      "做 起",
+      "尼 尔",
+      "be havior",
+      "▁flat ten",
+      "▁ flatten",
+      "▁Tem plates",
+      "▁Template s",
+      "▁ Templates",
+      "▁Calcul ator",
+      "▁Calc ulator",
+      "▁ Calculator",
+      "▁W ellington",
+      "▁Well ington",
+      "▁Wel lington",
+      "▁altern ating",
+      "的比 赛",
+      "的 比赛",
+      "并 处",
+      "拿 来",
+      "的 就",
+      "▁re min",
+      "▁rem in",
+      "学士 学位",
+      "Bre ak",
+      "B reak",
+      "▁sk irt",
+      "▁ski rt",
+      "fa b",
+      "f ab",
+      "▁Di ocese",
+      "# >",
+      "se ys",
+      "sey s",
+      "发 力",
+      "vol umes",
+      "volume s",
+      "▁Sh adow",
+      "▁Sha dow",
+      "▁ Shadow",
+      "co tt",
+      "cot t",
+      "c ott",
+      "orph ism",
+      "▁test imony",
+      "▁testimon y",
+      "固 化",
+      "▁C lo",
+      "▁Cl o",
+      "本院 经审理",
+      "▁Ch ase",
+      "▁Cha se",
+      "Add Field",
+      "民 用",
+      "Ar c",
+      "A rc",
+      "it os",
+      "ito s",
+      "i tos",
+      "V B",
+      "膜 炎",
+      "的 国际",
+      "▁rib bon",
+      "filter ed",
+      "▁Aut omatic",
+      "▁Auto matic",
+      "▁Autom atic",
+      "▁val uation",
+      "▁valu ation",
+      "新一 轮",
+      "新 一轮",
+      "注册 资本",
+      "▁r apper",
+      "▁rap per",
+      "▁ rapper",
+      "▁b iography",
+      "▁bi ography",
+      "一 贯",
+      "同 情",
+      "评 委",
+      "过 关",
+      "创业 者",
+      "创 业者",
+      "并 结合",
+      "有意 义",
+      "有 意义",
+      "▁j umps",
+      "▁jump s",
+      "在法律规定 的范围内处分自己的",
+      "▁J ake",
+      "▁Ja ke",
+      "▁Jak e",
+      "▁hand lers",
+      "▁handle rs",
+      "▁handler s",
+      "▁ handlers",
+      "仙 女",
+      "各 省",
+      "救 灾",
+      "本 事",
+      "▁r outines",
+      "▁rout ines",
+      "▁routine s",
+      "脑 海",
+      "现 阶段",
+      "Cl ip",
+      "Cli p",
+      "C lip",
+      "▁Leg isl",
+      "二 楼",
+      "ic ulty",
+      "icult y",
+      "icul ty",
+      "▁Dur ham",
+      "环 绕",
+      "ov sk",
+      "my answer",
+      "达成 和解协议",
+      "▁P review",
+      "▁Pre view",
+      "▁ Preview",
+      "▁Saskat chewan",
+      "等 各种",
+      "重庆市 渝",
+      "受 案",
+      "而 我",
+      "草 案",
+      "▁D P",
+      "▁ DP",
+      "/\" ,\\",
+      "/\", \\",
+      "/ \",\\",
+      "as ured",
+      "asure d",
+      "asu red",
+      "▁Con duct",
+      "▁Cond uct",
+      "▁now adays",
+      "background Color",
+      "内 衣",
+      "ux e",
+      "u xe",
+      "银川 市",
+      "银 川市",
+      "up al",
+      "upa l",
+      "u pal",
+      "▁A ri",
+      "▁Ar i",
+      "▁Est ab",
+      "▁Es tab",
+      "cr edited",
+      "cred ited",
+      "credit ed",
+      "复 活",
+      "sl ash",
+      "s lash",
+      "▁Sup erv",
+      "▁Super v",
+      "▁helic opter",
+      "IC O",
+      "I CO",
+      "nP ass",
+      "n Pass",
+      "第三 十二条",
+      "▁s b",
+      "▁ sb",
+      "什么 样",
+      "▁Em pty",
+      "▁Emp ty",
+      "▁ Empty",
+      "▁ ü",
+      "景 象",
+      "水 和",
+      "装 扮",
+      "解 锁",
+      "▁T T",
+      "▁ TT",
+      "▁qu er",
+      "▁q uer",
+      "▁que r",
+      "▁ quer",
+      "上诉于 浙江省",
+      "低 头",
+      "的 面",
+      "畜牧 业",
+      "▁L il",
+      "▁Li l",
+      "▁p ore",
+      "▁po re",
+      "▁por e",
+      "b rowse",
+      "blank Input",
+      "人 说",
+      "解 压",
+      "no b",
+      "n ob",
+      "第一百 九",
+      "▁admin ister",
+      "驾 车",
+      "结构 和",
+      "▁dark ness",
+      "ar ker",
+      "ark er",
+      "▁ 场上位置",
+      "ENT IAL",
+      "▁m ening",
+      "▁me ning",
+      "▁men ing",
+      "in formatics",
+      "inform atics",
+      "template Url",
+      "▁A CE",
+      "▁AC E",
+      "▁ ACE",
+      "▁W er",
+      "▁We r",
+      "VI CES",
+      "VICE S",
+      "V ICES",
+      "\\\\ }\\\\)",
+      "\\\\} \\\\)",
+      "\\\\}\\\\ )",
+      "op tera",
+      "opt era",
+      "opter a",
+      "▁Mar cus",
+      "▁Marc us",
+      "举 个",
+      "贫 血",
+      "▁' ,",
+      "▁ ',",
+      "会 自动",
+      "工作 和",
+      "并 提供",
+      "发 声",
+      "▁Op portun",
+      "\" |",
+      "L U",
+      "发 还",
+      "sin x",
+      "si nx",
+      "s inx",
+      "系列 产品",
+      "▁m mol",
+      "▁mm ol",
+      "nAny one",
+      "▁ î",
+      "罚 字",
+      "葡萄 糖",
+      "▁T ip",
+      "▁Ti p",
+      "▁ Tip",
+      "down loads",
+      "download s",
+      "▁Design er",
+      "CH O",
+      "C HO",
+      "LA S",
+      "L AS",
+      "adjust ed",
+      "adj usted",
+      "天 线",
+      "的 投资",
+      "Pi ck",
+      "Pic k",
+      "P ick",
+      "▁W rest",
+      "▁Wr est",
+      "▁DI RECT",
+      "▁ DIRECT",
+      "▁play off",
+      "处处 长",
+      "处 处长",
+      "tp adding",
+      "t padding",
+      "本 刊",
+      "鸡 汤",
+      "nC c",
+      "IN CLUDING",
+      "源 码",
+      "目前 在",
+      "ver te",
+      "vert e",
+      "ca tion",
+      "cat ion",
+      "c ation",
+      "▁pat rol",
+      "片 区",
+      "人民 网",
+      "完全平方 公式",
+      "▁He aven",
+      "▁epidem iological",
+      "▁epidemiologic al",
+      "▁div ides",
+      "▁divid es",
+      "▁divide s",
+      "▁harass ment",
+      "▁real ization",
+      "▁s ag",
+      "▁sa g",
+      "▁Rev ival",
+      "十 里",
+      "查 阅",
+      "本科 学",
+      "本 科学",
+      "Cl ock",
+      "C lock",
+      "▁Cont ents",
+      "▁Content s",
+      "▁ Contents",
+      "▁tr action",
+      "▁tra ction",
+      "▁tract ion",
+      "▁ traction",
+      "公 摊",
+      "卡 拉",
+      "着 眼",
+      "t weet",
+      "mb alib",
+      "▁Check list",
+      "油 气",
+      "命名 为",
+      "命 名为",
+      "吸 血",
+      "视 线",
+      "▁& #",
+      "▁ &#",
+      "▁t uition",
+      "▁tu ition",
+      "\\' }",
+      "\\ '}",
+      "nSpec ies",
+      "巧 妙",
+      "租 车",
+      "Bt n",
+      "B tn",
+      "▁M G",
+      "▁ MG",
+      "那样 的",
+      "那 样的",
+      "wi ld",
+      "w ild",
+      "pro ducer",
+      "produ cer",
+      "prod ucer",
+      "解 剖",
+      "那 年",
+      "jo n",
+      "j on",
+      "月中 旬",
+      "▁合作 人物",
+      "稳 健",
+      "at em",
+      "ate m",
+      "a tem",
+      "Dec oder",
+      "Decode r",
+      "ur names",
+      "urn ames",
+      "urname s",
+      "▁B lake",
+      "▁Bl ake",
+      "第一 中级人民法院",
+      "带 的",
+      "tl s",
+      "t ls",
+      "更多 的是",
+      "更多的 是",
+      "▁c ite",
+      "▁cit e",
+      "▁ci te",
+      "▁ cite",
+      "nAthlet es",
+      "▁Sign ificant",
+      "▁ch ic",
+      "▁chi c",
+      "▁rel uct",
+      "财 保",
+      "总 决赛",
+      "拿出 来",
+      "拿 出来",
+      "教师 的",
+      "教 师的",
+      "就是 因为",
+      "就 是因为",
+      "▁Bed room",
+      "▁inev itable",
+      "▁rein forcement",
+      "▁reinforce ment",
+      "后 在",
+      "满 族",
+      "享有 的",
+      "享 有的",
+      "▁H ass",
+      "▁Ha ss",
+      "▁Has s",
+      "进行了 分析",
+      "▁E ther",
+      "▁Eth er",
+      "▁Et her",
+      "▁R apid",
+      "▁Ra pid",
+      "▁Rap id",
+      "▁saf egu",
+      "▁safe gu",
+      "国 资",
+      "▁H M",
+      "▁ HM",
+      "中西 医",
+      "中 西医",
+      "蒙古 族",
+      "▁b ooth",
+      "▁bo oth",
+      "▁boot h",
+      "馆 藏",
+      "血液 中",
+      "血 液中",
+      "nSpec ial",
+      "n Special",
+      "▁dess ert",
+      "nRec ently",
+      "nRecent ly",
+      "可 口",
+      "著 有",
+      "▁z b",
+      "▁ zb",
+      "A W",
+      "nM ale",
+      "nMal e",
+      "nMa le",
+      "▁gr asp",
+      "▁gra sp",
+      "不 支持",
+      "__ ':\\",
+      "▁t andem",
+      "▁tan dem",
+      "▁conform ational",
+      "▁conformation al",
+      "估 值",
+      "挤 压",
+      "流 的",
+      "以为 是",
+      "Add ed",
+      "Ad ded",
+      "▁upgrad ing",
+      "妥 善",
+      "RE ST",
+      "RES T",
+      "R EST",
+      "tool bar",
+      "天 花",
+      "自 称",
+      "—— “",
+      "An im",
+      "A nim",
+      "Inter faces",
+      "Interface s",
+      "品 类",
+      "首先 是",
+      "首 先是",
+      "▁Par ser",
+      "▁Parse r",
+      "▁Pars er",
+      "▁ Parser",
+      "▁he rein",
+      "▁her ein",
+      "▁here in",
+      "房 源",
+      "BE R",
+      "B ER",
+      "▁viol in",
+      "▁vi olin",
+      "▁ac ademy",
+      "▁academ y",
+      "填 词",
+      "麻 辣",
+      "推荐 的",
+      "同时 也是",
+      "同时也 是",
+      "▁cere bro",
+      "get Item",
+      "dep artment",
+      "▁defe ating",
+      "▁defeat ing",
+      "滤 波",
+      "龙 泉",
+      "了 这个",
+      "To do",
+      "T odo",
+      "▁trans former",
+      "▁transform er",
+      "▁ transformer",
+      "重 合",
+      "anti ago",
+      "▁pack aged",
+      "▁package d",
+      "IS C",
+      "I SC",
+      "▁设 置",
+      "▁ 设置",
+      "nEnter ing",
+      "nEnt ering",
+      "▁f p",
+      "▁ fp",
+      "') :\\",
+      "' ):\\",
+      "Ph ase",
+      "品 尝",
+      "承 接",
+      "em o",
+      "e mo",
+      "▁cre ature",
+      "▁creat ure",
+      "将 对",
+      "开 阔",
+      "or f",
+      "o rf",
+      "ат ь",
+      "а ть",
+      "▁S pin",
+      "▁Sp in",
+      "▁ Spin",
+      "▁o wing",
+      "▁ow ing",
+      "▁ owing",
+      "▁ten ants",
+      "▁tenant s",
+      "▁elig ibility",
+      "bi gg",
+      "big g",
+      "b igg",
+      "}\\\\ }\\\\)",
+      "}\\\\} \\\\)",
+      "} \\\\}\\\\)",
+      "▁civil ian",
+      "▁C openhagen",
+      "▁frustr ating",
+      "嘴 里",
+      "脾 胃",
+      "结合 起来",
+      "▁rub y",
+      "▁ru by",
+      "▁ ruby",
+      "Mark et",
+      "Mar ket",
+      "dom ains",
+      "domain s",
+      "mem brane",
+      "▁Im plement",
+      "▁Imp lement",
+      "物 管",
+      "// #",
+      "/ /#",
+      "rt ype",
+      "r type",
+      "▁Te achers",
+      "▁Teacher s",
+      "▁ver tically",
+      "▁vert ically",
+      "▁vertical ly",
+      "逐 年",
+      "链 表",
+      "nd ouble",
+      "ndo uble",
+      "n double",
+      "▁diagn oses",
+      "▁diagnose s",
+      "又不 是",
+      "又 不是",
+      "ag ra",
+      "agr a",
+      "a gra",
+      "▁F ol",
+      "▁Fo l",
+      "an mar",
+      "▁bo iler",
+      "▁boil er",
+      "ass istant",
+      "is is",
+      "isi s",
+      "i sis",
+      "▁c urt",
+      "▁cur t",
+      "▁cu rt",
+      "en erated",
+      "ener ated",
+      "ene rated",
+      "enerate d",
+      "牛 皮",
+      "小姐 姐",
+      "小 姐姐",
+      "银行 的",
+      "银 行的",
+      "▁l bs",
+      "▁lb s",
+      "制作 人",
+      "DU CT",
+      "D UCT",
+      "▁p ads",
+      "▁pa ds",
+      "▁pad s",
+      "▁r anch",
+      "▁ra nch",
+      "▁ran ch",
+      "▁in accur",
+      "会 造成",
+      "health y",
+      "了 她",
+      "涨 价",
+      "不 代表",
+      "▁d orm",
+      "▁do rm",
+      "▁st itch",
+      "眼 科",
+      "满意 的",
+      "满 意的",
+      "() ]\\",
+      "()] \\",
+      "( )]\\",
+      "▁s ab",
+      "▁sa b",
+      "绝 大部分",
+      "会 更",
+      "的 特征",
+      ")) **",
+      "))* *",
+      ") )**",
+      "▁\\ '$",
+      "▁\\' $",
+      "▁ \\'$",
+      "▁R ent",
+      "▁Re nt",
+      "▁Ren t",
+      "▁red ox",
+      "Script s",
+      "▁en crypt",
+      "▁enc rypt",
+      "▁ encrypt",
+      "短 板",
+      "af ety",
+      "afe ty",
+      "Bound s",
+      "B ounds",
+      "▁belong ed",
+      "每 件",
+      "▁R I",
+      "▁ RI",
+      "提起 的",
+      "提 起的",
+      "的一 大",
+      "的 一大",
+      "▁Mc L",
+      "有 下列情形",
+      "▁read able",
+      "▁ readable",
+      "极 度",
+      "li es",
+      "lie s",
+      "l ies",
+      "总部 位于",
+      "▁h ull",
+      "▁hu ll",
+      "▁War riors",
+      "▁Warrior s",
+      "▁U ltimate",
+      "速度 快",
+      "ber ger",
+      "berg er",
+      "▁Sp encer",
+      "▁pun ishment",
+      "地方 政府",
+      "S G",
+      "志 刚",
+      "梦 里",
+      "当年 的",
+      "当 年的",
+      "▁F ailed",
+      "▁Fa iled",
+      "▁Fail ed",
+      "▁ Failed",
+      "▁cycl ists",
+      "▁cyclist s",
+      "B a",
+      "申请 费",
+      "TA IN",
+      "T AIN",
+      "ig ion",
+      "igi on",
+      "ret ch",
+      "r etch",
+      "X L",
+      "** ,",
+      "* *,",
+      "提出 异议",
+      "pag ination",
+      "p agination",
+      "& \\\\",
+      "养老 金",
+      "▁C omics",
+      "▁Com ics",
+      "▁Comic s",
+      "白色 的",
+      "白 色的",
+      "in in",
+      "ini n",
+      "i nin",
+      "视 图",
+      "首 要",
+      "nR ob",
+      "n Rob",
+      "ar ange",
+      "ara nge",
+      "aran ge",
+      "a range",
+      "▁be aten",
+      "▁beat en",
+      "▁economic al",
+      "▁ec onomical",
+      "▁econom ical",
+      "▁design ation",
+      "改 装",
+      "无 毛",
+      "楼 下",
+      "绿 地",
+      "SE M",
+      "S EM",
+      "的 办法",
+      "▁Fac ility",
+      "▁Go ing",
+      "▁Jama ica",
+      "产 于",
+      "黄 花",
+      "Ma d",
+      "M ad",
+      "ri na",
+      "rin a",
+      "r ina",
+      "▁parent ing",
+      "▁paren ting",
+      "并 能",
+      "法官 助理",
+      "特别 是在",
+      "特别是 在",
+      "il lon",
+      "ill on",
+      "illo n",
+      "▁sup pressor",
+      "▁suppress or",
+      "在 小",
+      "河 口",
+      "Par s",
+      "Pa rs",
+      "P ars",
+      "▁re lying",
+      "▁rel ying",
+      "▁rely ing",
+      "误 区",
+      "Br e",
+      "B re",
+      "损 耗",
+      "课 后",
+      "吸 入",
+      "规 避",
+      "Ap ache",
+      "商 的",
+      "很 低",
+      "电 极",
+      "AP H",
+      "A PH",
+      "所以 我们",
+      "所以我 们",
+      "▁E F",
+      "▁ EF",
+      "产业 园",
+      "cur ve",
+      "请仔细 研究",
+      "▁correct ions",
+      "▁correction s",
+      "▁constitu ents",
+      "▁constituent s",
+      "这 只",
+      "▁t ended",
+      "▁ten ded",
+      "▁tend ed",
+      "楼 上",
+      "'] )",
+      "' ])",
+      "启 蒙",
+      "收 据",
+      "uate s",
+      "ua tes",
+      "u ates",
+      "▁Hy brid",
+      "国民 经济",
+      "▁Cor ner",
+      "▁Corn er",
+      "▁Mi rror",
+      "▁Mir ror",
+      "▁ Mirror",
+      "▁re fine",
+      "▁ref ine",
+      "▁refin e",
+      "物业 费",
+      "昨 晚",
+      "ma gic",
+      "mag ic",
+      "m agic",
+      "▁on ions",
+      "▁onion s",
+      "嫌 弃",
+      "▁n n",
+      "▁ nn",
+      "本文 从",
+      "今年 以来",
+      "今 年以来",
+      "▁Th ought",
+      "▁Though t",
+      "el sen",
+      "else n",
+      "els en",
+      "e lsen",
+      "vol ution",
+      "v olution",
+      "st ant",
+      "stan t",
+      "sta nt",
+      "▁critic ized",
+      "国 旗",
+      "▁Tr iple",
+      "▁Tri ple",
+      "▁Trip le",
+      "未发现 被执行人",
+      "未 发现被执行人",
+      "▁un fair",
+      "▁unf air",
+      "夺 冠",
+      "认真 的",
+      "认 真的",
+      "T U",
+      "和 不",
+      "榜 样",
+      "ag ul",
+      "agu l",
+      "重要 意义",
+      "hydro x",
+      "hyd rox",
+      "▁h ometown",
+      "▁home town",
+      "经典 的",
+      "巡 逻",
+      "朝 着",
+      "▁social ly",
+      "▁soc ially",
+      "▁soci ally",
+      "▁s aturated",
+      "▁sat urated",
+      "次 会议",
+      "▁Sk ills",
+      "▁Skill s",
+      "执行案件 立案",
+      "▁C annot",
+      "▁Can not",
+      "▁Ca nnot",
+      "▁Cann ot",
+      "▁ Cannot",
+      "▁verte br",
+      "▁ vertebr",
+      ")\\\\ ;",
+      ") \\\\;",
+      "▁N okia",
+      "▁Sh akespeare",
+      "专 卖",
+      "八 卦",
+      "OP EN",
+      "Add itional",
+      "▁hippocamp al",
+      "▁P retty",
+      "▁Pre tty",
+      "▁Pret ty",
+      "第二 中级人民法院",
+      "▁re paired",
+      "▁rep aired",
+      "▁repair ed",
+      "▁malign ancy",
+      "变 了",
+      "死 后",
+      ")} $\\",
+      ")}$ \\",
+      ") }$\\",
+      "拒不 履行",
+      "拒 不履行",
+      "▁G PIO",
+      "▁GP IO",
+      "▁ GPIO",
+      "▁Art ists",
+      "▁Artist s",
+      "▁hol istic",
+      "▁D in",
+      "▁Di n",
+      "▁NS String",
+      "▁ NSString",
+      "▁architect ures",
+      "▁architecture s",
+      "也 挺",
+      "应 变",
+      "黏 膜",
+      "ry l",
+      "r yl",
+      "压 制",
+      "大 面积",
+      "脑 血管",
+      "Us ed",
+      "Use d",
+      "U sed",
+      "bur se",
+      "b urse",
+      "▁gr aded",
+      "▁grad ed",
+      "▁grade d",
+      "▁gra ded",
+      "▁ graded",
+      "▁Psych ology",
+      "拓 宽",
+      "it r",
+      "i tr",
+      "yo n",
+      "y on",
+      "▁F ork",
+      "▁For k",
+      "▁Fo rk",
+      "▁thank ful",
+      "偏 向",
+      "ra ls",
+      "ral s",
+      "r als",
+      "▁B ros",
+      "▁Br os",
+      "▁Bro s",
+      "▁Nor folk",
+      "▁infil tr",
+      "▁head lines",
+      "▁headline s",
+      "▁ 学校",
+      "ell ery",
+      "eller y",
+      "elle ry",
+      "nAny way",
+      "▁Paralym pics",
+      "高 分",
+      "js p",
+      "j sp",
+      "SER VICE",
+      "▁Pot ter",
+      "生 机",
+      "▁ 详细",
+      "▁v oy",
+      "▁vo y",
+      "▁r ook",
+      "▁ro ok",
+      "▁rein forced",
+      "▁reinforce d",
+      "相 见",
+      "划 算",
+      "火 花",
+      "绍兴 市",
+      "om eric",
+      "ome ric",
+      "omer ic",
+      "o meric",
+      "▁s keleton",
+      "▁ske leton",
+      "赞 同",
+      "St ub",
+      "S tub",
+      "be nz",
+      "ben z",
+      "b enz",
+      "▁g uit",
+      "▁gu it",
+      "▁gui t",
+      "▁S urre",
+      "▁Sur re",
+      "▁recip rocal",
+      "▁reciproc al",
+      "▁g h",
+      "▁ gh",
+      "nb uild",
+      "n build",
+      "ob server",
+      "obs erver",
+      "observ er",
+      "if ty",
+      "ift y",
+      "i fty",
+      "un used",
+      "▁under water",
+      "▁underw ater",
+      "无 职业",
+      "fa ir",
+      "f air",
+      "严格 按照",
+      "▁adult hood",
+      "▁adul thood",
+      ", ’",
+      "回 了",
+      "轮 廓",
+      "Wi Fi",
+      "li ved",
+      "live d",
+      "l ived",
+      "video s",
+      "v ideos",
+      "▁purs ued",
+      "▁pursue d",
+      "▁alk aline",
+      "▁alkal ine",
+      "▁convenient ly",
+      "▁contrad iction",
+      "▁contradict ion",
+      "不 由",
+      "施 肥",
+      "吃 过",
+      "氯 化",
+      "an ie",
+      "ani e",
+      "a nie",
+      "▁Brand on",
+      "▁Bra ndon",
+      "有很大 的",
+      "有 很大的",
+      "▁acqu iring",
+      "北 市",
+      "迫 切",
+      "▁e du",
+      "▁ed u",
+      "▁ edu",
+      "▁Te ams",
+      "▁Team s",
+      "▁Tea ms",
+      "▁em bol",
+      "▁emb ol",
+      "▁Or igin",
+      "▁Ori gin",
+      "▁ Origin",
+      "淋 漓",
+      "如果 我们",
+      "如果我 们",
+      "min ded",
+      "mind ed",
+      "可以 使",
+      "适配 器",
+      "▁game play",
+      "▁ext inction",
+      "▁extinct ion",
+      "国 会",
+      "▁A DD",
+      "▁AD D",
+      "▁ ADD",
+      "▁he ater",
+      "▁heat er",
+      "en ia",
+      "eni a",
+      "e nia",
+      "ibli cal",
+      "ibl ical",
+      "▁mod ulo",
+      "▁im pulse",
+      "▁imp ulse",
+      "nf e",
+      "n fe",
+      "工程 的",
+      "▁Dis cord",
+      "▁Disc ord",
+      "传染 病",
+      "内 分泌",
+      "广泛 应用于",
+      "广泛应用 于",
+      "▁T opic",
+      "▁To pic",
+      "▁Top ic",
+      "▁ Topic",
+      "分 娩",
+      "nc l",
+      "n cl",
+      "nC RE",
+      "n CRE",
+      "▁m arry",
+      "▁mar ry",
+      "ope napi",
+      "open api",
+      "ne apolis",
+      "fun ded",
+      "fund ed",
+      "git commit",
+      "▁ ра",
+      "▁adj uvant",
+      "▁trans location",
+      "▁transl ocation",
+      "Fr iday",
+      "▁c akes",
+      "▁ca kes",
+      "▁cake s",
+      "▁ cakes",
+      "▁h elm",
+      "▁he lm",
+      "▁hel m",
+      "▁ helm",
+      "▁sq lite",
+      "▁sql ite",
+      "▁ sqlite",
+      "▁squ amous",
+      "ang led",
+      "angle d",
+      "▁Vlad imir",
+      "▁qual ifications",
+      "▁qualification s",
+      "量 和",
+      "ol an",
+      "ola n",
+      "o lan",
+      "▁Ap ps",
+      "▁App s",
+      "▁ Apps",
+      "南 非",
+      "▁ 浙江",
+      "入 党",
+      "听 见",
+      "等 信息",
+      "羽毛 球",
+      "war f",
+      "wa rf",
+      "w arf",
+      "Pro cessing",
+      "Process ing",
+      "Proc essing",
+      "台 词",
+      "不 低于",
+      "对我 们",
+      "对 我们",
+      "▁D ent",
+      "▁De nt",
+      "▁Den t",
+      "Fil ters",
+      "Filter s",
+      "监 区",
+      "土地 承包",
+      "城县 人民法院",
+      "城 县人民法院",
+      "#### ##\\",
+      "### ###\\",
+      "##### #\\",
+      "###### \\",
+      "▁Tra ffic",
+      "▁chrom osomal",
+      "▁chromos omal",
+      "▁R ush",
+      "▁Ru sh",
+      "▁Rus h",
+      "▁M illion",
+      "▁Mil lion",
+      "▁Mill ion",
+      "▁pre natal",
+      "杨 柳",
+      "▁ac claimed",
+      "高 一",
+      "从 一个",
+      "理论 上",
+      "的水 平",
+      "的 水平",
+      "un ing",
+      "uni ng",
+      "u ning",
+      "▁pre clinical",
+      "▁rent als",
+      "▁rental s",
+      "▁teen ager",
+      "▁teenage r",
+      "translation type",
+      "隐 形",
+      "▁} }/",
+      "▁}} /",
+      "▁R ica",
+      "▁Ric a",
+      "▁Ri ca",
+      "篇 章",
+      "除 去",
+      "▁ 玄幻",
+      "zh en",
+      "zhe n",
+      "z hen",
+      "▁k om",
+      "▁ko m",
+      "▁ kom",
+      "ycl ic",
+      "yc lic",
+      "▁fl ank",
+      "({ \"",
+      "( {\"",
+      "没 关系",
+      "da sh",
+      "das h",
+      "d ash",
+      "▁A ub",
+      "▁Au b",
+      "▁Manufact uring",
+      "参 会",
+      "nA ND",
+      "n AND",
+      "opt imal",
+      "optim al",
+      "▁fr eight",
+      "▁fre ight",
+      "▁optim izing",
+      "nT im",
+      "n Tim",
+      "locfile hash",
+      "open locfilehash",
+      "source gitcommit",
+      "▁Tun is",
+      "▁Tu nis",
+      "▁Gab riel",
+      "▁Gabri el",
+      "▁Des ert",
+      "▁Main tenance",
+      "出 差",
+      "有不 少",
+      "有 不少",
+      "▁di lig",
+      "▁dil ig",
+      "▁B rowse",
+      "▁Brow se",
+      "▁ Browse",
+      "▁immun ore",
+      "▁immuno re",
+      "Da m",
+      "D am",
+      "lic kr",
+      "lick r",
+      "▁Ba iley",
+      "展 出",
+      "bu k",
+      "b uk",
+      "▁P ump",
+      "▁Pu mp",
+      "▁b ang",
+      "▁ban g",
+      "▁ba ng",
+      "▁ bang",
+      "▁back pack",
+      "光 的",
+      "Hi t",
+      "H it",
+      "双 曲线",
+      "日报 道",
+      "日 报道",
+      "本 公司",
+      "▁c hor",
+      "▁ch or",
+      "▁cho r",
+      "▁ chor",
+      "pre tty",
+      "pret ty",
+      "p retty",
+      "▁kn ees",
+      "▁kne es",
+      "▁knee s",
+      "▁ve rsa",
+      "▁ver sa",
+      "▁vers a",
+      "nE arlier",
+      "TRIBUT ORS",
+      "nW eek",
+      "nWe ek",
+      "n Week",
+      "▁ang iography",
+      "▁angi ography",
+      "▁\" >",
+      "▁ \">",
+      "有 自己的",
+      "法律 依据",
+      "资格 考试",
+      "软件 开发",
+      "▁S heet",
+      "▁She et",
+      "▁ Sheet",
+      "▁fun gi",
+      "▁fung i",
+      "ver ages",
+      "verage s",
+      "▁det ectors",
+      "▁detect ors",
+      "▁detector s",
+      "欢 喜",
+      "潇 湘",
+      "he e",
+      "h ee",
+      "驾驶 的",
+      "自然科学 基金",
+      "谈 谈",
+      "非 凡",
+      "▁T LR",
+      "▁TL R",
+      "▁K iss",
+      "▁Ki ss",
+      "▁Pe arl",
+      "▁Pear l",
+      "ns tring",
+      "nst ring",
+      "nstr ing",
+      "n string",
+      "▁un safe",
+      "▁uns afe",
+      "▁ unsafe",
+      "cob acter",
+      "c obacter",
+      "▁Philipp ine",
+      "▁Philip pine",
+      "▁grand mother",
+      "疲 惫",
+      "这里 是",
+      "ig roup",
+      "igr oup",
+      "i group",
+      "施 行",
+      "越 是",
+      "不断 提升",
+      "▁ce il",
+      "▁ ceil",
+      "ot ives",
+      "otive s",
+      "oti ves",
+      "p q",
+      "引 人",
+      "ST ER",
+      "STE R",
+      "S TER",
+      "ta ils",
+      "tail s",
+      "t ails",
+      "ul iar",
+      "uli ar",
+      "ulia r",
+      "▁b itter",
+      "▁bit ter",
+      "▁Synd rome",
+      "统 战",
+      "▁S UV",
+      "▁SU V",
+      "▁ SUV",
+      "nt rans",
+      "ntr ans",
+      "n trans",
+      "▁Tom my",
+      "on uclear",
+      "onucle ar",
+      "下 行",
+      "多 月",
+      "收 纳",
+      "木 兰",
+      "]} \\",
+      "] }\\",
+      "▁H REF",
+      "▁HR EF",
+      "Look ing",
+      "Lo oking",
+      "L ooking",
+      "▁like wise",
+      "▁lik ewise",
+      "▁ 米",
+      "口 径",
+      "接 头",
+      "ze d",
+      "z ed",
+      "更 何况",
+      "▁E xc",
+      "▁Ex c",
+      "ic orn",
+      "ico rn",
+      "i corn",
+      "▁A void",
+      "▁Av oid",
+      "▁De von",
+      "▁Dev on",
+      "▁migr ants",
+      "▁migrant s",
+      "Dep artment",
+      "之日起五日内 向本院申请复议一次",
+      "听 着",
+      "奔 跑",
+      "崇 拜",
+      "RP C",
+      "R PC",
+      "▁M V",
+      "▁ MV",
+      "allenge s",
+      "alleng es",
+      "▁WAR NING",
+      "▁ WARNING",
+      "水 源",
+      "透 气",
+      "▁P am",
+      "▁Pa m",
+      "id os",
+      "ido s",
+      "i dos",
+      "▁he els",
+      "▁heel s",
+      "▁ heels",
+      "▁plas mid",
+      "▁plasm id",
+      "diff erent",
+      "d ifferent",
+      "综 述",
+      "形式 为",
+      "▁ident ifiers",
+      "▁identifier s",
+      "动 了",
+      "油 的",
+      "sp r",
+      "s pr",
+      "ca ss",
+      "cas s",
+      "c ass",
+      "na mon",
+      "nam on",
+      "n amon",
+      "au coma",
+      "auc oma",
+      "WAR NING",
+      "eval uate",
+      "evalu ate",
+      "ie g",
+      "i eg",
+      "▁V oc",
+      "▁Vo c",
+      "ip pers",
+      "ipp ers",
+      "ipper s",
+      "恐 龙",
+      "▁seiz ure",
+      "▁butter fly",
+      "▁cont inent",
+      "▁contin ent",
+      "Cert ificate",
+      "上 报",
+      "机 车",
+      "就 觉得",
+      "ne ls",
+      "nel s",
+      "n els",
+      "▁Sub mit",
+      "▁ Submit",
+      "▁diagn ostics",
+      "▁diagnostic s",
+      "▁ 贾",
+      "克 的",
+      "秦 皇",
+      "▁L T",
+      "▁ LT",
+      "生产 和",
+      "国际 贸易",
+      "你 这",
+      "/( (",
+      "/ ((",
+      "留下 了",
+      "留 下了",
+      "市场 需求",
+      "group Id",
+      "平 淡",
+      "up les",
+      "uple s",
+      "u ples",
+      "▁b ins",
+      "▁bi ns",
+      "▁bin s",
+      "▁ bins",
+      "▁c ath",
+      "▁ca th",
+      "▁cat h",
+      "▁sc rub",
+      "▁scr ub",
+      "单纯 的",
+      "band s",
+      "ban ds",
+      "b ands",
+      "▁M aced",
+      "▁Mac ed",
+      "▁Ma ced",
+      "改 正",
+      "植 入",
+      "▁demand ed",
+      "▁W end",
+      "▁We nd",
+      "▁Wen d",
+      "Util ity",
+      "U tility",
+      "▁inter change",
+      "Com b",
+      "Co mb",
+      "C omb",
+      "ON LINE",
+      "al amic",
+      "ala mic",
+      "alam ic",
+      "s elenium",
+      "▁hor rible",
+      "▁horr ible",
+      "好 听",
+      "经 纬",
+      "冲 刺",
+      "发达 国家",
+      "▁ch unks",
+      "▁chunk s",
+      "▁ chunks",
+      "res trial",
+      "rest rial",
+      "▁esc aped",
+      "▁escape d",
+      "▁escap ed",
+      "T alk",
+      "pt ime",
+      "p time",
+      "nW ater",
+      "n Water",
+      "▁T icket",
+      "▁ Ticket",
+      "▁D OM",
+      "▁DO M",
+      "▁ DOM",
+      "说 自己",
+      "nOk ay",
+      "台 式",
+      "建 构",
+      "▁内 容",
+      "▁ 内容",
+      "▁C ars",
+      "▁Car s",
+      "▁Ca rs",
+      "▁Man uel",
+      "▁Pre viously",
+      "▁Previous ly",
+      "尿 酸",
+      "LI N",
+      "L IN",
+      "▁c li",
+      "▁cl i",
+      "▁ cli",
+      "▁ang iotensin",
+      "血 样",
+      "at os",
+      "ato s",
+      "a tos",
+      "▁Ser bian",
+      "▁Serbia n",
+      "航 母",
+      "pe m",
+      "p em",
+      "确定 了",
+      "确 定了",
+      "▁N azi",
+      "▁Na zi",
+      "▁Naz i",
+      "erg arten",
+      "▁trans porter",
+      "▁transport er",
+      "细 心",
+      "Re ceived",
+      "Receive d",
+      "▁pres ently",
+      "▁present ly",
+      "▁amb assador",
+      "再 加",
+      "▁mac OS",
+      "▁f asting",
+      "▁fast ing",
+      "协 定",
+      "嵌入 式",
+      "▁a board",
+      "▁ab oard",
+      "Create Model",
+      "▁ 德",
+      "斯 塔",
+      "突出 的",
+      "突 出的",
+      "Can not",
+      "Ca nnot",
+      "C annot",
+      "▁ab brev",
+      "re leased",
+      "release d",
+      "▁C orrect",
+      "▁Cor rect",
+      "▁ Correct",
+      "▁Fail ure",
+      "▁ Failure",
+      "▁carcin omas",
+      "▁carcinoma s",
+      "▁w t",
+      "▁ wt",
+      "im its",
+      "imit s",
+      "imi ts",
+      "i mits",
+      "指 甲",
+      "MS G",
+      "M SG",
+      "快速 的",
+      "外出 务工",
+      "▁v archar",
+      "▁var char",
+      "函数的 单调性",
+      "▁Dec ision",
+      "▁citizens hip",
+      "▁citizen ship",
+      "IF Y",
+      "pi x",
+      "p ix",
+      "▁D D",
+      "▁ DD",
+      "整体 的",
+      "整 体的",
+      "要 从",
+      "nE ND",
+      "n END",
+      "pl astic",
+      "▁ward robe",
+      "过来 的",
+      "过 来的",
+      "sh all",
+      "sha ll",
+      "shal l",
+      "s hall",
+      "col our",
+      "申请执行人 可以",
+      "申请执行 人可以",
+      "申请执行人可 以",
+      "inst ances",
+      "instance s",
+      "Q S",
+      "之 情",
+      "商 是",
+      "将 被",
+      "▁C od",
+      "▁Co d",
+      "▁ Cod",
+      "管理 制度",
+      "▁congr at",
+      "▁cong rat",
+      "▁pre school",
+      "▁pres chool",
+      "gr ey",
+      "gre y",
+      "g rey",
+      "▁Re tail",
+      "▁Ret ail",
+      "达 人",
+      "监 委",
+      "▁l b",
+      "▁ lb",
+      "en na",
+      "enn a",
+      "▁flag ship",
+      "▁flags hip",
+      "▁c ategorized",
+      "▁categor ized",
+      "公 认",
+      "▁finger print",
+      "二 的",
+      "馒 头",
+      "or ange",
+      "ora nge",
+      "oran ge",
+      "o range",
+      "不等式 组",
+      "▁C ort",
+      "▁Co rt",
+      "▁Cor t",
+      "▁w and",
+      "▁wa nd",
+      "form erly",
+      "former ly",
+      "str icted",
+      "strict ed",
+      "nScott ish",
+      ")) ]\\",
+      ") )]\\",
+      "▁Z oo",
+      "▁Re id",
+      "抓 紧",
+      "su gg",
+      "s ugg",
+      "▁Cap itol",
+      "尊 严",
+      "滋 润",
+      "mi s",
+      "m is",
+      "之 以",
+      "反 比例",
+      "不断 发展",
+      "\\' $",
+      "\\ '$",
+      "▁commit tees",
+      "▁committee s",
+      "无正当理由 未到庭参加诉讼",
+      "▁perturb ation",
+      "往 前",
+      "▁我 的",
+      "▁ 我的",
+      "创新 的",
+      "创 新的",
+      "lif ting",
+      "lift ing",
+      "l ifting",
+      "A O",
+      "倡 议",
+      "av ir",
+      "avi r",
+      "a vir",
+      "▁tail or",
+      "part icip",
+      "▁mil estone",
+      "▁miles tone",
+      "▁mile stone",
+      "术 中",
+      "▁d uck",
+      "▁du ck",
+      "par ation",
+      "pa ration",
+      "para tion",
+      "▁hippocamp us",
+      "▁practition er",
+      "id ia",
+      "idi a",
+      "▁F UN",
+      "▁ FUN",
+      "insp ired",
+      "thand off",
+      "深 情",
+      "特 质",
+      "路 与",
+      "▁sle eve",
+      "集 的",
+      "}% \\",
+      "} %\\",
+      "id ian",
+      "idi an",
+      "idia n",
+      "▁fin ances",
+      "▁finance s",
+      "▁financ es",
+      "▁frustr ated",
+      "风 貌",
+      "▁confirm ing",
+      "包 围",
+      "建筑 材料",
+      "ro cyte",
+      "roc yte",
+      "r ocyte",
+      "▁Object s",
+      "▁ Objects",
+      "是 企业",
+      "ap ur",
+      "a pur",
+      "▁alumin ium",
+      "▁ret aining",
+      "▁retain ing",
+      "▁ 北",
+      "▁O C",
+      "▁ OC",
+      "是不 会",
+      "是 不会",
+      "Br idge",
+      "B ridge",
+      "▁d ownt",
+      "▁down t",
+      "▁dow nt",
+      "▁susp ects",
+      "▁suspect s",
+      "▁infil tration",
+      "▁infiltr ation",
+      "▁@ @",
+      "▁ @@",
+      "las thandoff",
+      "能 使",
+      "采访 时",
+      "建筑 工程有限公司",
+      "建筑工程 有限公司",
+      "与 否",
+      "右 脚",
+      "英文 名",
+      "英 文名",
+      "TR AN",
+      "TRA N",
+      "T RAN",
+      "to ire",
+      "ur ate",
+      "ura te",
+      "u rate",
+      "ect ing",
+      "ec ting",
+      "ectin g",
+      "nUp date",
+      "n Update",
+      "nPro duct",
+      "n Product",
+      "赡 养",
+      "▁w his",
+      "▁wh is",
+      "可能 有",
+      "可 能有",
+      "bal anced",
+      "balance d",
+      "▁res iding",
+      "▁resid ing",
+      "第二百六 十四条",
+      "▁pre term",
+      "▁pr eterm",
+      "▁pret erm",
+      "浓 缩",
+      "国际 机场",
+      "gev ity",
+      "sh aring",
+      "sha ring",
+      "▁indirect ly",
+      "世 家",
+      "红色 的",
+      "红 色的",
+      "Br anch",
+      "▁B il",
+      "▁Bi l",
+      "▁Ed die",
+      "▁Edd ie",
+      "▁spir its",
+      "▁spirit s",
+      "▁jud icial",
+      "有线 电视",
+      "is ers",
+      "ise rs",
+      "iser s",
+      "content locale",
+      "拍 的",
+      "ab ove",
+      "abo ve",
+      "中 性",
+      "ka l",
+      "k al",
+      "Look up",
+      "▁th ats",
+      "▁that s",
+      "▁re name",
+      "▁r ename",
+      "▁ rename",
+      "▁st icky",
+      "▁stick y",
+      "ect l",
+      "ec tl",
+      "e ctl",
+      "▁G uitar",
+      "▁Gu itar",
+      "▁D owntown",
+      "▁Down town",
+      "疑 难",
+      "省 略",
+      "▁differ ed",
+      "▁diff ered",
+      "▁E PA",
+      "▁EP A",
+      "的最小 值",
+      "的 最小值",
+      "被执行人 的",
+      "re name",
+      "ren ame",
+      "rena me",
+      "r ename",
+      "▁me tre",
+      "▁met re",
+      "▁S ocket",
+      "▁Soc ket",
+      "▁ Socket",
+      "▁warm th",
+      "De cember",
+      "Dec ember",
+      "课 本",
+      "PV C",
+      "P VC",
+      "美国 人",
+      "美 国人",
+      "ho rn",
+      "hor n",
+      "h orn",
+      "从小 到大",
+      "at aka",
+      "ata ka",
+      "▁Consider ing",
+      "穿 上",
+      "自己 去",
+      "余 名",
+      "大 局",
+      "cap ture",
+      "capt ure",
+      "Com merce",
+      "增 生",
+      "▁政治 面貌",
+      "RE SULT",
+      "RES ULT",
+      "导 游",
+      "CL E",
+      "C LE",
+      "▁w ars",
+      "▁war s",
+      "▁wa rs",
+      "▁Le icester",
+      "爱 人",
+      "or etic",
+      "ore tic",
+      "▁Br ooks",
+      "▁Bro oks",
+      "▁Brook s",
+      "▁gen res",
+      "▁ge nres",
+      "▁genre s",
+      "▁Sil icon",
+      "Ta x",
+      "T ax",
+      "交通 警察",
+      "▁qu iz",
+      "▁qui z",
+      "▁ quiz",
+      "交纳 诉讼费用",
+      "tel ling",
+      "t elling",
+      "▁Jul ian",
+      "▁Julia n",
+      "▁sulf ur",
+      "▁sul fur",
+      "案件受理费 减半收取",
+      "诸 如",
+      "生产 企业",
+      "爱 尔",
+      "珊 瑚",
+      "ur dy",
+      "urd y",
+      "ic hel",
+      "ich el",
+      "iche l",
+      "i chel",
+      "▁个人 简介",
+      "▁semicon ductor",
+      "yl ated",
+      "ylate d",
+      "yla ted",
+      "y lated",
+      "▁mal ware",
+      "▁integr ates",
+      "▁integrate s",
+      "wo n",
+      "w on",
+      "▁h ey",
+      "▁he y",
+      "▁ hey",
+      "▁b acks",
+      "▁back s",
+      "▁ba cks",
+      "▁bac ks",
+      "▁ backs",
+      "▁grad ients",
+      "▁gradient s",
+      "得 名",
+      "▁Al ert",
+      "▁Ale rt",
+      "▁ Alert",
+      "▁t ornado",
+      "▁torn ado",
+      "▁lat itude",
+      "▁ latitude",
+      "大 臣",
+      "就 用",
+      "秋 冬",
+      "fr ared",
+      "fra red",
+      "nc ategories",
+      "n categories",
+      "je r",
+      "j er",
+      "▁Ok ay",
+      "nL ater",
+      "nLa ter",
+      "▁col league",
+      "▁colle ague",
+      "茄 子",
+      "追 偿",
+      "期间 内",
+      "for cing",
+      "▁deal ership",
+      "▁dealer ship",
+      "▁dealers hip",
+      "张 力",
+      "二〇一七年 八月",
+      "二〇一七年八 月",
+      "▁sub type",
+      "▁subt ype",
+      "停 留",
+      "▁Y a",
+      "es ar",
+      "esa r",
+      "住房 公积金",
+      "Fi nder",
+      "Find er",
+      "Fin der",
+      "F inder",
+      "从 重",
+      "切 口",
+      "突 变",
+      "so lete",
+      "sole te",
+      "sol ete",
+      "jet brains",
+      "▁even tual",
+      "▁event ual",
+      "闻 名",
+      "开 玩笑",
+      "ov ić",
+      "ovi ć",
+      "ad ors",
+      "ado rs",
+      "ador s",
+      "▁H ood",
+      "▁Ho od",
+      "▁b loom",
+      "▁bl oom",
+      "▁blo om",
+      "▁spin ning",
+      "ena issance",
+      "e naissance",
+      "▁Con ditions",
+      "▁Cond itions",
+      "▁Condition s",
+      "PI D",
+      "P ID",
+      "▁sp rings",
+      "▁spring s",
+      "▁spr ings",
+      "职业技术 学院",
+      "pat ients",
+      "patient s",
+      "m igration",
+      "峰 会",
+      "bi l",
+      "b il",
+      "节 假日",
+      "▁c emetery",
+      "和 尚",
+      "he it",
+      "从业 人员",
+      "nC reated",
+      "nCreate d",
+      "n Created",
+      "▁hydroly sis",
+      "无 理",
+      "PI I",
+      "P II",
+      "游泳 池",
+      "游 泳池",
+      "▁B ib",
+      "▁Bi b",
+      "▁L ay",
+      "▁La y",
+      "▁a le",
+      "▁al e",
+      "▁ ale",
+      "Inv oke",
+      "get All",
+      "▁Le sson",
+      "▁Less on",
+      "▁Les son",
+      "为 真",
+      "▁He ath",
+      "▁Heat h",
+      "prevent Default",
+      "▁s z",
+      "▁ sz",
+      "队伍 建设",
+      "▁Tr uth",
+      "▁ Truth",
+      "▁nick name",
+      "▁ nickname",
+      "▁context ual",
+      "в и",
+      "民 航",
+      "远 方",
+      "精神 的",
+      "精 神的",
+      "▁pert aining",
+      "ll er",
+      "l ler",
+      "Reg ular",
+      "pur e",
+      "pu re",
+      "p ure",
+      "工程 项目",
+      "▁an gel",
+      "▁ang el",
+      "▁ angel",
+      "▁Integr al",
+      "债务 人",
+      "以下 内容",
+      "第一百零 一条",
+      "▁Linked In",
+      "全 长",
+      "▁b rom",
+      "▁br om",
+      "▁bro m",
+      "▁Any way",
+      "▁bi opsies",
+      "营养 价值",
+      "is odes",
+      "iso des",
+      "isode s",
+      "▁spec ialize",
+      "▁special ize",
+      "if a",
+      "i fa",
+      "Br and",
+      "B rand",
+      "履行 法律文书确定的义务",
+      "要求 其",
+      "name se",
+      "nam ese",
+      "names e",
+      "Jo urnal",
+      "J ournal",
+      "get User",
+      "▁Man aging",
+      "▁Manag ing",
+      "lo e",
+      "l oe",
+      "用户 提供",
+      "sp rite",
+      "spr ite",
+      "s prite",
+      "两 项",
+      "猴 子",
+      "ul la",
+      "ull a",
+      "u lla",
+      "▁p rey",
+      "▁pre y",
+      "▁pr ey",
+      "▁overlook ed",
+      "▁rep etitive",
+      "▁complet eness",
+      "▁complete ness",
+      "W W",
+      "日 讯",
+      "用 餐",
+      ")` \\",
+      ") `\\",
+      "▁s hy",
+      "▁sh y",
+      "▁guitar s",
+      "▁guit ars",
+      "IB LE",
+      "I BLE",
+      "就 必须",
+      "Reg ex",
+      "▁disc our",
+      "落 下",
+      "在我 的",
+      "在 我的",
+      "科技 成果",
+      "参考 书",
+      "▁R ip",
+      "▁Ri p",
+      "美 化",
+      "se at",
+      "sea t",
+      "s eat",
+      "▁C RC",
+      "▁CR C",
+      "▁ CRC",
+      "▁R ut",
+      "▁Ru t",
+      "悔罪 表现",
+      "im eric",
+      "ime ric",
+      "imer ic",
+      "i meric",
+      "nest js",
+      "er tools",
+      "ert ools",
+      "erto ols",
+      "▁H av",
+      "▁Ha v",
+      "Le ague",
+      "▁Pear son",
+      "▁Encycl opedia",
+      "越 野",
+      "外交 部",
+      "产品 类型",
+      "检验 报告",
+      "▁Ga rage",
+      "▁Gar age",
+      "▁reg istr",
+      "▁retrospective ly",
+      "▁retrospect ively",
+      "减刑 一年",
+      "over lay",
+      "text rm",
+      "tex trm",
+      "▁shock ed",
+      "都 比",
+      "ion i",
+      "io ni",
+      "i oni",
+      "的 评价",
+      "▁A nth",
+      "▁An th",
+      "▁Ant h",
+      "立即 开始执行",
+      "▁st ance",
+      "▁ stance",
+      "▁w ashed",
+      "▁was hed",
+      "▁wash ed",
+      "▁P ere",
+      "▁Per e",
+      "▁Pe re",
+      "(\\' ,",
+      "(\\ ',",
+      "( \\',",
+      "非常 重要",
+      "▁ant ique",
+      "▁anti que",
+      "▁antiqu e",
+      "in escence",
+      "ines cence",
+      "▁pract iced",
+      "▁practice d",
+      "er yl",
+      "ery l",
+      "e ryl",
+      "Al ignment",
+      "Align ment",
+      "▁res idency",
+      "▁resid ency",
+      "▁reside ncy",
+      "EN AB",
+      "E NAB",
+      "iv ier",
+      "ivi er",
+      "▁Ban ach",
+      "▁st reak",
+      "▁stre ak",
+      "▁She ffield",
+      "带 走",
+      "电 工",
+      "由被告 负担",
+      "Trans ition",
+      "法的 意义",
+      "法 的意义",
+      "le tes",
+      "let es",
+      "lete s",
+      "▁W ake",
+      "▁Wa ke",
+      "▁person alities",
+      "▁personal ities",
+      "正方 体的",
+      "正方体 的",
+      "▁re im",
+      "束 缚",
+      "for k",
+      "fo rk",
+      "f ork",
+      "ra nce",
+      "ran ce",
+      "r ance",
+      "▁co zy",
+      "License d",
+      "L icensed",
+      "nG allery",
+      "n Gallery",
+      "未 婚",
+      "nb r",
+      "n br",
+      "人 死亡",
+      "保护 区",
+      "erg us",
+      "me ans",
+      "mean s",
+      "▁wed dings",
+      "▁wedding s",
+      "▁ Á",
+      "▁ 适用",
+      "ni nline",
+      "nin line",
+      "n inline",
+      "▁direct ing",
+      "▁dir ecting",
+      "和大 家",
+      "和 大家",
+      "帅 哥",
+      "前几 天",
+      "前 几天",
+      "▁y og",
+      "▁yo g",
+      "sw ing",
+      "s wing",
+      "▁typ o",
+      "▁ty po",
+      "价值 人民币",
+      "资产 管理",
+      "▁R ein",
+      "▁Re in",
+      "▁en tre",
+      "▁ent re",
+      "▁entr e",
+      "ar ag",
+      "ara g",
+      "a rag",
+      "mon ton",
+      "mo nton",
+      "mont on",
+      "▁oste opor",
+      "▁osteo por",
+      "{ [",
+      "因 果",
+      "▁hospital ized",
+      "整个 人",
+      "整 个人",
+      "as an",
+      "asa n",
+      "a san",
+      "▁h ollow",
+      "▁hol low",
+      "工程 款",
+      "人均 价格",
+      "▁p ave",
+      "▁pa ve",
+      "▁pav e",
+      "▁Men tor",
+      "▁Ment or",
+      "▁C ONT",
+      "▁CO NT",
+      "▁CON T",
+      "▁ CONT",
+      "▁Se rum",
+      "▁Ser um",
+      "▁And rea",
+      "▁Andre a",
+      "▁Andr ea",
+      "争 论",
+      "庞大 的",
+      "庞 大的",
+      "▁preced ed",
+      "▁vic inity",
+      "▁S EM",
+      "▁SE M",
+      "▁ SEM",
+      "▁p unk",
+      "▁pun k",
+      "▁pu nk",
+      "▁ punk",
+      "em aker",
+      "ema ker",
+      "e maker",
+      "▁Gil bert",
+      "德 华",
+      "的 开发",
+      "[] []",
+      "[ ][]",
+      "To ast",
+      "▁c ate",
+      "▁ca te",
+      "▁cat e",
+      "ot oxin",
+      "oto xin",
+      "otox in",
+      "▁an notations",
+      "▁annotation s",
+      "▁annot ations",
+      "▁ annotations",
+      "Y L",
+      "▁sp ider",
+      "一 幅",
+      "悬 浮",
+      "千万 别",
+      "媒体 报道",
+      "▁la unches",
+      "▁launch es",
+      "▁form ulations",
+      "▁formulation s",
+      "杠 杆",
+      "纺织 品",
+      "nG ame",
+      "n Game",
+      "▁Bar onet",
+      "▁Baron et",
+      "▁ 行政区",
+      "nTax onomy",
+      "In ternational",
+      "Inter national",
+      "▁W I",
+      "▁ WI",
+      "av ier",
+      "avi er",
+      "初 次",
+      "酵 母",
+      "都不 知道",
+      "都 不知道",
+      "▁a dap",
+      "▁ad ap",
+      "▁re boot",
+      "▁reb oot",
+      "电 容",
+      "载 荷",
+      "▁Greg ory",
+      "谋 划",
+      "和国 际",
+      "和 国际",
+      "Sl ice",
+      "▁analog ous",
+      "▁contamin ated",
+      "均 在",
+      "▁hand made",
+      "天 真",
+      "▁b ore",
+      "▁bo re",
+      "▁bor e",
+      "ol utely",
+      "olute ly",
+      "▁Key board",
+      "▁ Keyboard",
+      "树 林",
+      "▁pilot s",
+      "▁pil ots",
+      "第一百三十三条之一 第一款",
+      "Im m",
+      "I mm",
+      "▁all ergy",
+      "▁allerg y",
+      "▁colon ies",
+      "实 名",
+      "农业 大学",
+      "药物 治疗",
+      "uc ing",
+      "u cing",
+      "F ollow",
+      "折 腾",
+      "Ni l",
+      "N il",
+      "▁k ay",
+      "▁ka y",
+      "nh elp",
+      "nhe lp",
+      "n help",
+      "SD L",
+      "S DL",
+      "▁SP DX",
+      "▁SPD X",
+      "▁H oney",
+      "▁Hon ey",
+      "▁Ho ney",
+      "▁Fant asy",
+      "工商 银行股份有限公司",
+      "工商银行 股份有限公司",
+      "re presentation",
+      "rep resentation",
+      "represent ation",
+      "an del",
+      "and el",
+      "ande l",
+      "a ndel",
+      "▁r ivals",
+      "▁ri vals",
+      "▁riv als",
+      "▁rival s",
+      "商 人",
+      "就 跟",
+      "▁答 案",
+      "▁ 答案",
+      "▁Pe dro",
+      "▁Ped ro",
+      "▁mal icious",
+      "▁adopt ing",
+      "▁adop ting",
+      "v l",
+      "而 生",
+      "▁cess ation",
+      "三 轮",
+      "主 页",
+      "第二百 二十",
+      "▁ed ema",
+      "▁S IM",
+      "▁SI M",
+      "▁ SIM",
+      "▁C han",
+      "▁Ch an",
+      "▁Cha n",
+      "▁ Chan",
+      "bi ased",
+      "bia sed",
+      "bias ed",
+      "nReturn s",
+      "n Returns",
+      "AS F",
+      "A SF",
+      "J ames",
+      "▁compet itor",
+      "▁competit or",
+      "▁discover ies",
+      "▁植物 界",
+      "need ed",
+      "nee ded",
+      "▁Cl one",
+      "▁Clo ne",
+      "▁ Clone",
+      "同 日",
+      "某某 于",
+      "某 某于",
+      "P riority",
+      "性 感",
+      "人体 的",
+      "人 体的",
+      "成长 的",
+      "成 长的",
+      "▁B lo",
+      "▁Bl o",
+      "Ent ries",
+      "ogen etic",
+      "ogene tic",
+      "oge netic",
+      "尘 埃",
+      "▁I nj",
+      "▁In j",
+      "▁y ang",
+      "▁ya ng",
+      "▁ yang",
+      "omy ces",
+      "list ing",
+      "lis ting",
+      "l isting",
+      "O U",
+      "举 动",
+      "con g",
+      "co ng",
+      "c ong",
+      "▁In ner",
+      "▁Inn er",
+      "▁ Inner",
+      "▁Sc ene",
+      "▁ Scene",
+      "mo tor",
+      "mot or",
+      "m otor",
+      "pur pose",
+      "p urpose",
+      "大 肠",
+      "的 分",
+      "sp ons",
+      "s pons",
+      "▁learn t",
+      "▁lear nt",
+      "▁e rosion",
+      "▁er osion",
+      "▁Har per",
+      "▁enthusi asts",
+      "▁enthusiast s",
+      "一 楼",
+      "总 有",
+      "ро в",
+      "р ов",
+      "在 进行",
+      "ir ie",
+      "iri e",
+      "i rie",
+      "▁cir rh",
+      "▁obst acle",
+      "▁aux iliary",
+      "▁Altern ative",
+      "▁Alter native",
+      "膀 胱",
+      "路 过",
+      "办理 减刑",
+      "▁mod ulated",
+      "▁modulate d",
+      "泄 露",
+      "▁at ten",
+      "▁att en",
+      "答 复",
+      "▁Te V",
+      "科学发展 观",
+      "▁aver aging",
+      ")$ $,",
+      ")$$ ,",
+      ") $$,",
+      "▁s ci",
+      "▁sc i",
+      "▁ sci",
+      "nE DIT",
+      "n EDIT",
+      "▁sa cred",
+      "▁sacr ed",
+      "▁sac red",
+      "ra ce",
+      "rac e",
+      "r ace",
+      "▁H yd",
+      "▁Hy d",
+      "▁s ensible",
+      "▁sens ible",
+      "oe lect",
+      "o elect",
+      "▁VE GF",
+      "▁ VEGF",
+      "▁ch ees",
+      "▁che es",
+      "▁J ump",
+      "▁Ju mp",
+      "移动 互联网",
+      "▁W ords",
+      "▁Word s",
+      "▁Wor ds",
+      "▁ Words",
+      "▁c oded",
+      "▁co ded",
+      "▁code d",
+      "▁cod ed",
+      "▁ coded",
+      "▁the med",
+      "▁th emed",
+      "▁them ed",
+      "▁theme d",
+      "▁ themed",
+      "▁weak ly",
+      "▁tr acing",
+      "▁tra cing",
+      "dr ug",
+      "d rug",
+      "▁B end",
+      "▁Be nd",
+      "▁Ben d",
+      "Number s",
+      "Num bers",
+      "N umbers",
+      "▁quarter ly",
+      "▁quart erly",
+      "▁prec ursors",
+      "▁precursor s",
+      "再 看",
+      "我 为",
+      "▁B ald",
+      "▁Bal d",
+      "▁Ba ld",
+      "sign ed",
+      "sig ned",
+      "s igned",
+      "▁In cludes",
+      "▁Include s",
+      "football er",
+      "外 套",
+      "度 数",
+      "廉 洁",
+      "box ed",
+      "as mine",
+      "asm ine",
+      "nBe ing",
+      "n Being",
+      "▁sh ells",
+      "▁shell s",
+      "▁shel ls",
+      "class ifiers",
+      "▁c ot",
+      "▁co t",
+      "▁ cot",
+      "▁dig estion",
+      "▁digest ion",
+      "▁rheumat oid",
+      "ut z",
+      "u tz",
+      "sl ack",
+      "▁Am endment",
+      "▁kilomet res",
+      "▁mental ly",
+      "▁ment ally",
+      "▁stand alone",
+      "▁ standalone",
+      "价 为",
+      "贴 吧",
+      "逆 转",
+      "▁C hip",
+      "▁Ch ip",
+      "▁Chi p",
+      "▁ Chip",
+      "▁dam n",
+      "▁da mn",
+      "Com pare",
+      "Comp are",
+      "Compar e",
+      "干 货",
+      "的值 是",
+      "Pat ch",
+      "P atch",
+      "▁a mel",
+      "▁am el",
+      "▁ amel",
+      "▁Tr uck",
+      "他 又",
+      "针对 性的",
+      "针对性 的",
+      "▁custom s",
+      "▁cust oms",
+      "▁K or",
+      "▁Ko r",
+      "▁le aks",
+      "▁leak s",
+      "提供在线 菜单",
+      "▁ 杰",
+      "还不 是",
+      "还 不是",
+      "师范 学院",
+      "▁cost umes",
+      "▁costume s",
+      "地 段",
+      "适合 自己的",
+      "▁Ret rieved",
+      "▁g estation",
+      "▁ge station",
+      "▁gest ation",
+      "满 分",
+      "▁Indones ian",
+      "▁Indonesia n",
+      "涂 层",
+      "as ide",
+      "asi de",
+      "a side",
+      "▁pen etr",
+      "▁penet r",
+      "▁hard wood",
+      "干 脆",
+      "中国 建设",
+      "▁step ping",
+      "▁ste pping",
+      "▁immun ological",
+      "▁immuno logical",
+      "▁immunologic al",
+      "也 得",
+      "发展 趋势",
+      "尚 有",
+      "榜 单",
+      "怎么 看",
+      "Http Client",
+      "大 蒜",
+      "视 听",
+      "▁ 穿越",
+      "List Node",
+      "▁in ception",
+      "▁ 梅",
+      "长度 为",
+      "长 度为",
+      "▁F leet",
+      "▁Fle et",
+      "▁e aten",
+      "▁eat en",
+      "双 床",
+      "▁Tri angle",
+      "▁ther apeutics",
+      "▁therapeutic s",
+      "是 基于",
+      "▁J ur",
+      "▁Ju r",
+      "▁Bul let",
+      "▁Bull et",
+      "我 才",
+      "▁R R",
+      "▁ RR",
+      "▁d ynasty",
+      "▁Fin dings",
+      "▁Find ings",
+      "▁Finding s",
+      "▁Mod eling",
+      "▁Model ing",
+      "▁Mode ling",
+      "▁Pan ama",
+      "▁Pa nama",
+      "星 空",
+      "要 以",
+      "拥有 的",
+      "拥 有的",
+      "阴 性",
+      "▁s olves",
+      "▁sol ves",
+      "▁solve s",
+      "▁solv es",
+      "卵 形",
+      "冠 心病",
+      "ar mac",
+      "arm ac",
+      "arma c",
+      "村 村民委员会",
+      "▁c yan",
+      "▁cy an",
+      "▁enrich ment",
+      "山 大",
+      "胶 原",
+      "▁特 点",
+      "▁ 特点",
+      "ar xiv",
+      "Build ing",
+      "非常 有",
+      "非 常有",
+      "是这样 的",
+      "是 这样的",
+      "进行 调查",
+      "▁te tra",
+      "▁tet ra",
+      "▁tetr a",
+      "平 整",
+      "▁内 置",
+      "▁ 内置",
+      "屋 顶",
+      "nr es",
+      "nre s",
+      "n res",
+      "▁H ip",
+      "▁Hi p",
+      "一审 法院",
+      "ow ing",
+      "o wing",
+      "▁sp ans",
+      "▁span s",
+      "▁spa ns",
+      "display Name",
+      "和 信息化",
+      "▁第 一",
+      "▁ 第一",
+      "哪 一个",
+      "▁teen age",
+      "Vari ables",
+      "Variable s",
+      "▁Ant arctic",
+      "▁t ed",
+      "▁te d",
+      "▁ ted",
+      "本案按 撤诉处理",
+      "▁sp ermat",
+      "▁sperm at",
+      "▁sper mat",
+      "出 货",
+      "明 清",
+      "以外 的",
+      "以 外的",
+      "▁r im",
+      "▁ri m",
+      "▁ rim",
+      "铜 陵",
+      "el len",
+      "ell en",
+      "elle n",
+      "▁sw immers",
+      "▁swim mers",
+      "▁swimmer s",
+      "▁ 飞",
+      "在 与",
+      "▁ 网站",
+      "新 材料",
+      "▁t ors",
+      "▁to rs",
+      "▁tor s",
+      "CH ANGE",
+      "▁d arker",
+      "▁dark er",
+      "▁dar ker",
+      "nG overnment",
+      "▁Contin uous",
+      "累 了",
+      "um ption",
+      "ump tion",
+      "压 抑",
+      "尚 书",
+      "新 世纪",
+      "um ph",
+      "ump h",
+      "u mph",
+      "▁o nCreate",
+      "▁on Create",
+      "▁ onCreate",
+      "雨 水",
+      "▁Comb ined",
+      "▁Combin ed",
+      "▁Combine d",
+      "一 并",
+      "的 运动",
+      "▁b end",
+      "▁be nd",
+      "▁ben d",
+      "▁int uition",
+      "去 皮",
+      "原告 要求被告",
+      "asset id",
+      "asse tid",
+      "▁label led",
+      "▁lab elled",
+      "▁ labelled",
+      "再 一次",
+      "▁Gr ass",
+      "▁Gra ss",
+      "逐 一",
+      "的 形成",
+      "olec ules",
+      "olecule s",
+      "再次提出 执行申请",
+      "iv ot",
+      "ivo t",
+      "▁S ue",
+      "▁Su e",
+      "oc hem",
+      "och em",
+      "oche m",
+      "o chem",
+      "K K",
+      "刘 海",
+      "会 很",
+      "一个 大",
+      "一 个大",
+      "还 记得",
+      "▁O re",
+      "▁Or e",
+      "ul tan",
+      "ult an",
+      "优惠 券",
+      "nT EST",
+      "n TEST",
+      "▁fa una",
+      "re start",
+      "rest art",
+      "Inter ceptor",
+      "一 所",
+      "吃 苦",
+      "呵 护",
+      "起 始",
+      "ri mp",
+      "rim p",
+      "r imp",
+      "nEn joy",
+      "▁Bar ack",
+      "▁Ba rack",
+      "▁所属 运动队",
+      "Per missions",
+      "Permission s",
+      "Perm issions",
+      "▁rid iculous",
+      "亚 军",
+      "▁um bre",
+      "▁umb re",
+      "报 请",
+      "▁M BA",
+      "▁MB A",
+      "▁ MBA",
+      "▁\\\\ |",
+      "▁ \\\\|",
+      "你会 发现",
+      "你 会发现",
+      "放 进",
+      "') ;",
+      "' );",
+      "▁Ben nett",
+      "诉前 财产保全",
+      "▁ox idase",
+      "▁oxid ase",
+      "▁Trans lation",
+      "▁Transl ation",
+      "▁ Translation",
+      "小孩 子",
+      "小 孩子",
+      "的 思考",
+      "▁T aken",
+      "▁Take n",
+      "▁Ta ken",
+      "▁Tak en",
+      "nb efore",
+      "nbe fore",
+      "n before",
+      ":\\ ',",
+      ":\\' ,",
+      ": \\',",
+      "RE QUI",
+      "▁Di abetes",
+      "安 康",
+      "▁) ;",
+      "▁ );",
+      "到 这里",
+      "az ar",
+      "aza r",
+      "▁t amb",
+      "▁ta mb",
+      "▁tam b",
+      "▁Es sex",
+      "▁Ess ex",
+      "城 中",
+      "处 女",
+      "chn ical",
+      "chnic al",
+      "coh olic",
+      "cohol ic",
+      "放 过",
+      "▁harm ony",
+      "▁harmon y",
+      "▁k an",
+      "▁ka n",
+      "▁ kan",
+      "▁C reating",
+      "▁Creat ing",
+      "▁Cre ating",
+      "▁ Creating",
+      "▁emphas ized",
+      "▁emphasize d",
+      "▁ 毛",
+      "to ps",
+      "top s",
+      "t ops",
+      "▁\" \"\"",
+      "▁\"\" \"",
+      "▁ \"\"\"",
+      "小型 轿车",
+      "▁c ombo",
+      "▁com bo",
+      "▁comb o",
+      "歌 声",
+      "的手 机",
+      "的 手机",
+      "om ore",
+      "omo re",
+      "o more",
+      "进一步 加强",
+      "ident ified",
+      "▁H CC",
+      "▁HC C",
+      "▁W AY",
+      "▁WA Y",
+      "▁ WAY",
+      "▁dis closed",
+      "▁disclose d",
+      "一 刻",
+      "稀 释",
+      "好 朋友",
+      "nt on",
+      "nto n",
+      "n ton",
+      "▁c uff",
+      "▁cu ff",
+      "▁vir ulence",
+      "的 第",
+      "酸 性",
+      "aj e",
+      "a je",
+      "只剩 下",
+      "只 剩下",
+      "相关 性",
+      "泄 漏",
+      "IV ES",
+      "IVE S",
+      "缅 甸",
+      "da p",
+      "d ap",
+      "▁ 这",
+      "▁de duct",
+      "▁ded uct",
+      "▁fol ded",
+      "▁fold ed",
+      "导 读",
+      "汽车 销售",
+      "▁g lam",
+      "▁gl am",
+      "ol ition",
+      "olit ion",
+      "oli tion",
+      "▁Con fed",
+      "▁Conf ed",
+      "Re lated",
+      "Rel ated",
+      "也 只是",
+      "ap est",
+      "ape st",
+      "apes t",
+      "▁D ame",
+      "▁Dam e",
+      "▁Da me",
+      "的 办学",
+      "西 红柿",
+      "ra pe",
+      "rap e",
+      "r ape",
+      "体制 改革",
+      ")\\ \\\\\\",
+      ")\\\\ \\\\",
+      ") \\\\\\\\",
+      "一 如",
+      "▁R N",
+      "▁ RN",
+      "梗 死",
+      "没 能",
+      "nAl tern",
+      "n Altern",
+      "nS ingles",
+      "nSing les",
+      "nSingle s",
+      "会 给",
+      "▁ Š",
+      "手 持",
+      "▁R i",
+      "一步 步",
+      "一 步步",
+      "▁h oly",
+      "▁hol y",
+      "▁ho ly",
+      "z k",
+      "就 更",
+      "减刑 后",
+      "应有 的",
+      "应 有的",
+      "Man ag",
+      "Ma nag",
+      "M anag",
+      "un ordered",
+      "append Child",
+      "发生 变化",
+      "▁threshold s",
+      "说 你",
+      "▁Sus sex",
+      "印 尼",
+      "矿 产",
+      "▁fire f",
+      "▁fi ref",
+      "▁fir ef",
+      "valid ators",
+      "validator s",
+      "▁G ates",
+      "▁Ga tes",
+      "▁Gate s",
+      "▁Gat es",
+      "▁E nsure",
+      "▁Ens ure",
+      "老 爷",
+      "展现 了",
+      "平方差 公式",
+      "▁teen agers",
+      "▁teenager s",
+      "▁teenage rs",
+      "涌 现",
+      "Mi r",
+      "M ir",
+      "pl ants",
+      "plant s",
+      "plan ts",
+      "jen kins",
+      "▁Re ason",
+      "▁ Reason",
+      "▁Un known",
+      "▁ Unknown",
+      "▁check list",
+      "可以 看",
+      "▁G RE",
+      "▁GR E",
+      "▁ GRE",
+      "▁mist aken",
+      "▁mistake n",
+      "▁cur iosity",
+      "al ex",
+      "ale x",
+      "a lex",
+      "教学 质量",
+      "▁th oracic",
+      "饼 干",
+      "▁dis cord",
+      "▁disc ord",
+      "▁ discord",
+      "带 宽",
+      "摇 滚",
+      "花 期",
+      "配 偶",
+      "可以 的",
+      "▁coh omology",
+      "ends with",
+      "W ay",
+      "ozyg ous",
+      "Url s",
+      "Ur ls",
+      "▁f res",
+      "▁fr es",
+      "▁fre s",
+      "▁Clark e",
+      "▁Clar ke",
+      "▁Crim inal",
+      "三日 内",
+      "三 日内",
+      "ov ol",
+      "ovo l",
+      "o vol",
+      "号 行政处罚决定",
+      "▁G ust",
+      "▁Gu st",
+      "▁his tone",
+      "▁hist one",
+      "▁hi stone",
+      "class List",
+      "▁count able",
+      "▁cou ntable",
+      "▁capac ities",
+      "经验 和",
+      "证明 了",
+      "证 明了",
+      "▁A RM",
+      "▁AR M",
+      "▁ ARM",
+      "▁mis leading",
+      "九 州",
+      "元 二次方程",
+      "▁G uang",
+      "▁Gu ang",
+      "做 完",
+      "爆 料",
+      "be ar",
+      "b ear",
+      "pa use",
+      "p ause",
+      "▁we bin",
+      "▁web in",
+      "▁L ibraries",
+      "▁e cc",
+      "▁ec c",
+      "▁ ecc",
+      "最喜欢 的",
+      "最 喜欢的",
+      "oubt edly",
+      "二〇一七年 十一月",
+      "二〇一七年十一 月",
+      "二〇一七年十 一月",
+      "名 录",
+      "▁Kn ights",
+      "▁Knight s",
+      "▁perfect ion",
+      "▁perf ection",
+      "散 步",
+      "举办 了",
+      "代理 权限",
+      "▁Ash ley",
+      "▁get Name",
+      "▁ge tName",
+      "▁ getName",
+      "▁tr ailing",
+      "▁tra iling",
+      "▁trail ing",
+      "外 伤",
+      "ec o",
+      "e co",
+      "▁K end",
+      "▁Ke nd",
+      "▁Ken d",
+      "▁Ch uck",
+      "▁fuel s",
+      "▁fu els",
+      "▁e Bay",
+      "▁super visor",
+      "▁superv isor",
+      "nB as",
+      "n Bas",
+      "▁e ine",
+      "▁ein e",
+      "▁ eine",
+      "给我 的",
+      "给 我的",
+      "▁EL ISA",
+      "▁ ELISA",
+      "ym metric",
+      "自 私",
+      "紫外 线",
+      "▁T ips",
+      "▁Ti ps",
+      "▁Tip s",
+      "武 昌",
+      "▁ 案",
+      "容积 率",
+      "ic ides",
+      "ici des",
+      "icide s",
+      "之 称",
+      "▁b s",
+      "▁ bs",
+      "▁b ark",
+      "▁bar k",
+      "▁ba rk",
+      "▁cr ashed",
+      "▁crash ed",
+      "普 洱",
+      "药 材",
+      "淮安 市",
+      "淮 安市",
+      "Day s",
+      "Da ys",
+      "D ays",
+      "▁anti viral",
+      "▁antiv iral",
+      "▁Cy prus",
+      "▁f ighters",
+      "▁fight ers",
+      "▁fighter s",
+      "▁iso forms",
+      "▁isoform s",
+      "St ory",
+      "愿 景",
+      "月 饼",
+      "▁a o",
+      "▁ ao",
+      "已 缴纳",
+      "▁n eo",
+      "▁ne o",
+      "▁ neo",
+      "List s",
+      "Li sts",
+      "L ists",
+      "ew idth",
+      "e width",
+      "皮 革",
+      "on er",
+      "one r",
+      "o ner",
+      "nS uper",
+      "nSup er",
+      "n Super",
+      "▁Ca uchy",
+      "▁gl ance",
+      "▁dev iations",
+      "▁deviation s",
+      "▁N D",
+      "▁ ND",
+      "od erm",
+      "ode rm",
+      "oder m",
+      "▁Kash mir",
+      "▁sec uring",
+      "▁inher itance",
+      "▁inherit ance",
+      "绝对 是",
+      "公诉机关 指控被告人",
+      "公诉机关指控 被告人",
+      "▁M ovies",
+      "▁Movie s",
+      "低 压",
+      "ne ar",
+      "nea r",
+      "n ear",
+      "依法 适用",
+      "▁adip ose",
+      "预 定",
+      "▁Polit ics",
+      "nOver all",
+      "▁In stance",
+      "▁Inst ance",
+      "▁ Instance",
+      "▁Prob ably",
+      "▁direct ive",
+      "▁dir ective",
+      "▁ directive",
+      "我 之前",
+      "is ma",
+      "ism a",
+      "▁comb ustion",
+      "判 别",
+      "增 殖",
+      "May be",
+      "▁Red dit",
+      "▁ev oked",
+      "▁c rossover",
+      "▁cross over",
+      "每 组",
+      "沙 滩",
+      "nM ax",
+      "nMa x",
+      "n Max",
+      "nM el",
+      "nMe l",
+      "n Mel",
+      "他 认为",
+      "▁f oil",
+      "▁fo il",
+      "▁foi l",
+      "▁p rod",
+      "▁pro d",
+      "▁pr od",
+      "▁ prod",
+      "▁es ophageal",
+      "▁esophag eal",
+      "▁ esophageal",
+      "黄 油",
+      "我 记得",
+      "▁ 译者",
+      "▁Requ irements",
+      "山 林",
+      "镇 江",
+      "air es",
+      "ai res",
+      "aire s",
+      "a ires",
+      "巨 人",
+      "而 定",
+      "▁b p",
+      "▁ bp",
+      "我 坚信",
+      "▁U ttar",
+      "▁Ut tar",
+      "传 导",
+      "原 著",
+      "直线 与",
+      "▁T EXT",
+      "▁TE XT",
+      "▁ TEXT",
+      "▁Re place",
+      "▁ Replace",
+      "▁inv ites",
+      "▁invite s",
+      "企 图",
+      "只 在",
+      "▁H ours",
+      "▁Ho urs",
+      "▁Hour s",
+      "▁Hou rs",
+      "▁ Hours",
+      "上海浦东 发展",
+      "▁Not tingham",
+      "直 达",
+      "乳腺 癌",
+      "也是 一种",
+      "也是一 种",
+      "也 是一种",
+      "病 原",
+      "▁k not",
+      "▁kn ot",
+      "▁check point",
+      "▁ checkpoint",
+      "龙 门",
+      "▁R EG",
+      "▁RE G",
+      "▁ REG",
+      "完全 没有",
+      "▁am endment",
+      "▁amend ment",
+      "▁c ong",
+      "▁con g",
+      "▁co ng",
+      "▁ cong",
+      "▁l oses",
+      "▁lo ses",
+      "▁lose s",
+      "▁los es",
+      "原 神",
+      "OT A",
+      "O TA",
+      "▁R B",
+      "▁ RB",
+      "if inals",
+      "ifi nals",
+      "i finals",
+      "▁An imation",
+      "▁ Animation",
+      "× (",
+      "前 缀",
+      "▁PR OC",
+      "▁PRO C",
+      "▁ PROC",
+      "▁V O",
+      "▁ VO",
+      "▁S ac",
+      "▁Sa c",
+      "Un able",
+      "U nable",
+      "sm ooth",
+      "源 市",
+      "窃 取",
+      "车 祸",
+      "IC U",
+      "I CU",
+      "▁attenu ation",
+      "▁atten uation",
+      "oo s",
+      "o os",
+      "▁H b",
+      "▁ Hb",
+      "每年 的",
+      "每 年的",
+      "环境 变量",
+      "irect ional",
+      "irection al",
+      "▁disc retion",
+      "▁discre tion",
+      "▁discret ion",
+      "▁ С",
+      "安 德",
+      "date r",
+      "da ter",
+      "dat er",
+      "d ater",
+      "; <",
+      "▁ 사",
+      "光 芒",
+      "▁ver se",
+      "▁vers e",
+      "▁ verse",
+      "▁Inf antry",
+      "▁Vari ables",
+      "▁Variable s",
+      "▁ Variables",
+      "手 套",
+      "挫 折",
+      "通 电",
+      "Go d",
+      "G od",
+      "▁O g",
+      "▁K ra",
+      "▁Kr a",
+      "osc ope",
+      "o scope",
+      "催 讨",
+      "转型 升级",
+      "▁op posing",
+      "H M",
+      "▁dep ths",
+      "▁depth s",
+      "Jo h",
+      "J oh",
+      "ts x",
+      "t sx",
+      "▁tw ins",
+      "▁twin s",
+      "▁firm ware",
+      "阻 塞",
+      "不予 执行",
+      "▁int ram",
+      "▁intr am",
+      "▁intra m",
+      "▁r ou",
+      "▁ro u",
+      "▁ rou",
+      "▁W ool",
+      "▁Wo ol",
+      "Ro ck",
+      "R ock",
+      "值得 一提",
+      "▁instruct ors",
+      "▁instructor s",
+      "王 爷",
+      "nF l",
+      "n Fl",
+      "▁l emma",
+      "▁le mma",
+      "分 校",
+      "开 办",
+      "测 绘",
+      "的 单",
+      "he lf",
+      "hel f",
+      "h elf",
+      "Pa per",
+      "P aper",
+      "▁C had",
+      "▁Ch ad",
+      "▁Cha d",
+      "▁N est",
+      "▁Ne st",
+      "nA nn",
+      "nAn n",
+      "n Ann",
+      "U buntu",
+      "prom ise",
+      "p romise",
+      "ko m",
+      "k om",
+      "al an",
+      "ala n",
+      "a lan",
+      "icular ly",
+      "icul arly",
+      "▁de af",
+      "▁glut athione",
+      "开 出",
+      "▁C ay",
+      "▁Ca y",
+      "▁Joh annes",
+      "▁Johann es",
+      "金融 借款合同纠纷一案中",
+      "金融借款合同纠纷一案 中",
+      "▁a min",
+      "▁am in",
+      "▁ amin",
+      "▁Har dy",
+      "▁Hard y",
+      "▁cl ues",
+      "▁clue s",
+      "▁sculpt ures",
+      "▁sculpture s",
+      "积 的",
+      "是 这样",
+      "yl inder",
+      "▁fil ming",
+      "▁film ing",
+      "▁F abric",
+      "▁Fab ric",
+      "▁Sec ure",
+      "▁ Secure",
+      "户 数",
+      "萨 克",
+      "ft s",
+      "f ts",
+      "print ed",
+      "终结本次执行程序 后",
+      "▁contin uum",
+      "▁continu um",
+      "遇到 的",
+      "遇 到的",
+      "陌生 人",
+      "▁} $\\",
+      "▁}$ \\",
+      "▁ }$\\",
+      "▁j Query",
+      "▁ jQuery",
+      "▁lum bar",
+      "ec c",
+      "e cc",
+      "cal ls",
+      "call s",
+      "c alls",
+      "外国 人",
+      "外 国人",
+      "▁B le",
+      "▁Bl e",
+      "爸爸 妈妈",
+      "▁ 她",
+      "追 溯",
+      "大力 发展",
+      "xl abel",
+      "x label",
+      "Nov ember",
+      "N ovember",
+      ")) **(",
+      "))* *(",
+      "))** (",
+      ") )**(",
+      "▁' ';\\",
+      "▁'' ;\\",
+      "▁as sass",
+      "▁ass ass",
+      "▁G ateway",
+      "▁Gate way",
+      "▁Gat eway",
+      "▁ Gateway",
+      "▁ 技术",
+      "▁mon ot",
+      "▁mo not",
+      "▁mono t",
+      "▁Star ted",
+      "▁Start ed",
+      "▁ Started",
+      "▁cre ators",
+      "▁creat ors",
+      "▁creator s",
+      "▁commut ative",
+      "孙 某",
+      "无 形",
+      "开启 了",
+      "cr op",
+      "cro p",
+      "c rop",
+      "不仅 可以",
+      "基准 利率",
+      "gu ild",
+      "gui ld",
+      "g uild",
+      "比 重",
+      "CUR RENT",
+      "bound ed",
+      "b ounded",
+      "▁Gener ally",
+      "▁General ly",
+      "▁del iber",
+      "▁delib er",
+      "▁inc umbent",
+      "天 内",
+      "([ [",
+      "( [[",
+      "和 文化",
+      "请求 依法",
+      "ag ged",
+      "agg ed",
+      "的 抗",
+      "不合 理",
+      "不 合理",
+      "▁A RT",
+      "▁AR T",
+      "▁ ART",
+      "▁fr ustration",
+      "▁frustr ation",
+      "▁sh eer",
+      "▁she er",
+      "▁fun ctor",
+      "▁func tor",
+      "▁funct or",
+      "试 着",
+      "▁o lf",
+      "▁ol f",
+      "▁ olf",
+      "奥林 匹",
+      "的事 业",
+      "的 事业",
+      "移动 电话",
+      "▁Max well",
+      "ion es",
+      "io nes",
+      "ione s",
+      "i ones",
+      "▁S age",
+      "▁Sa ge",
+      "▁Sag e",
+      "at is",
+      "ati s",
+      "a tis",
+      "nInst all",
+      "nIns tall",
+      "n Install",
+      "管辖 权的",
+      "管辖权 的",
+      "▁wr aps",
+      "▁wrap s",
+      "▁secret ed",
+      "▁fellow ship",
+      "▁fellows hip",
+      "▁ 奥",
+      "弹 出",
+      "开幕 式",
+      "是一 项",
+      "是 一项",
+      "dec lar",
+      "decl ar",
+      "▁in die",
+      "▁ind ie",
+      "▁ �",
+      "规 程",
+      "▁co hes",
+      "▁coh es",
+      "▁per former",
+      "▁perform er",
+      "▁perfor mer",
+      "走 廊",
+      "▁pi gment",
+      "▁pig ment",
+      "tt ype",
+      "tty pe",
+      "t type",
+      "▁apolog ize",
+      "津 贴",
+      "代表 着",
+      "et ype",
+      "ety pe",
+      "e type",
+      "假释案件 具体应用法律",
+      "assert False",
+      "~~~~~~~~ ~~~~~~~~",
+      "一 书",
+      "都比 较",
+      "都 比较",
+      "ce il",
+      "et ed",
+      "ete d",
+      "e ted",
+      "th ed",
+      "the d",
+      "t hed",
+      "▁P ilot",
+      "▁Pi lot",
+      "▁Pil ot",
+      "▁E nable",
+      "▁En able",
+      "▁ Enable",
+      "事 了",
+      "av iolet",
+      "▁Cin ema",
+      "成 份",
+      "▁R D",
+      "▁ RD",
+      "pr ec",
+      "pre c",
+      "p rec",
+      "▁happ ily",
+      "▁l enders",
+      "▁len ders",
+      "▁lender s",
+      "▁lend ers",
+      "平安 银行股份有限公司",
+      "晓 东",
+      "路 灯",
+      "购买 的",
+      "购 买的",
+      "nC or",
+      "nCo r",
+      "n Cor",
+      "Table s",
+      "Tab les",
+      "Ta bles",
+      "T ables",
+      "ograph ers",
+      "ographer s",
+      "▁cort isol",
+      "工业 区",
+      "▁S ounds",
+      "▁Sound s",
+      "▁us able",
+      "▁ usable",
+      "收益 率",
+      "现实 中",
+      "▁con ventions",
+      "▁convention s",
+      "gu est",
+      "gue st",
+      "▁Leon ard",
+      "详 解",
+      "▁sur fact",
+      "▁surf act",
+      "ri le",
+      "ril e",
+      "r ile",
+      "nC ome",
+      "nCom e",
+      "nCo me",
+      "n Come",
+      "五 花",
+      "MB A",
+      "M BA",
+      "保全 费",
+      "兰州 市",
+      "兰 州市",
+      "请 登录",
+      "▁\\' \\',",
+      "▁\\'\\' ,",
+      "▁\\'\\ ',",
+      "▁ \\'\\',",
+      "▁ad her",
+      "国家 标准",
+      "哮 喘",
+      "CN N",
+      "C NN",
+      "▁剧 情",
+      "▁ 剧情",
+      "▁ind ul",
+      "▁indu l",
+      "cl osing",
+      "el eration",
+      "ele ration",
+      "eler ation",
+      "陵 区",
+      "然后 就",
+      "然 后就",
+      "认定 事实清楚",
+      "认定事实 清楚",
+      "▁as semble",
+      "▁assembl e",
+      "▁desper ate",
+      "▁grand children",
+      "地 毯",
+      "▁S IG",
+      "▁SI G",
+      "▁ SIG",
+      "▁c av",
+      "▁ca v",
+      "found ed",
+      "f ounded",
+      "F U",
+      "h z",
+      "Pi pe",
+      "P ipe",
+      "am bia",
+      "amb ia",
+      "▁pl atinum",
+      "人 设",
+      "pr em",
+      "pre m",
+      "p rem",
+      "▁H ir",
+      "▁Hi r",
+      "变 种",
+      "换 货",
+      "Foo d",
+      "F ood",
+      "sk in",
+      "ski n",
+      "s kin",
+      "▁a ka",
+      "▁ak a",
+      "▁ aka",
+      "住 广东省",
+      "oc yan",
+      "一 二",
+      "致 敬",
+      "▁建 筑",
+      "▁ 建筑",
+      "和 支持",
+      "IM ARY",
+      "能 有",
+      "gc c",
+      "g cc",
+      "总 工会",
+      "▁N ike",
+      "▁Ni ke",
+      "▁Nik e",
+      "▁M ixed",
+      "▁Mix ed",
+      "▁Ident ification",
+      "▁ Identification",
+      "nCon tin",
+      "nCont in",
+      "n Contin",
+      "conn ections",
+      "connect ions",
+      "connection s",
+      "源 自",
+      "up e",
+      "u pe",
+      "了解 了",
+      "re ceived",
+      "receive d",
+      "Sup ported",
+      "Support ed",
+      "▁it erator",
+      "▁iter ator",
+      "▁ iterator",
+      "J un",
+      "两位 数",
+      "两 位数",
+      "▁re dd",
+      "▁r edd",
+      "▁red d",
+      "pr actice",
+      "pract ice",
+      "能 把",
+      "IS TS",
+      "IST S",
+      "I STS",
+      "▁c ens",
+      "▁ce ns",
+      "mo zilla",
+      "moz illa",
+      "m ozilla",
+      "想 做",
+      "铺 设",
+      "间 距",
+      "Da n",
+      "D an",
+      "只有 在",
+      "▁言情 小说",
+      "▁ 言情小说",
+      "条第一款 的规定",
+      "▁re viewer",
+      "▁review er",
+      "不 看",
+      "▁D art",
+      "▁Dar t",
+      "▁Da rt",
+      "▁t asting",
+      "▁tast ing",
+      "ar um",
+      "aru m",
+      "a rum",
+      "主任 医师",
+      "▁b iod",
+      "▁bi od",
+      "▁bio d",
+      "▁py lint",
+      "▁pyl int",
+      "▁account ability",
+      "澎 湃",
+      "钢 丝",
+      "▁c asp",
+      "▁ca sp",
+      "▁cas p",
+      "并按对方当事人 的人数",
+      "孢 子",
+      "局 的",
+      "本判决 生效后",
+      "本 判决生效后",
+      "火 腿",
+      "见 识",
+      "▁T P",
+      "▁ TP",
+      "▁s tab",
+      "▁st ab",
+      "属于 基础题",
+      "char set",
+      "chars et",
+      "长 征",
+      "▁N U",
+      "▁ NU",
+      "ha rt",
+      "har t",
+      "h art",
+      "▁Bi om",
+      "▁Bio m",
+      "ad emic",
+      "ade mic",
+      "adem ic",
+      "▁H amburg",
+      "▁Ham burg",
+      "▁art ifact",
+      "▁ artifact",
+      "▁hard core",
+      "官 兵",
+      "戒 指",
+      "以及 对",
+      "以 及对",
+      "nC onn",
+      "nCon n",
+      "nCo nn",
+      "n Conn",
+      "▁na ive",
+      "并无 不当",
+      "re try",
+      "ret ry",
+      "r etry",
+      "背 部",
+      "交通 便利",
+      "借款 期限",
+      "容留 他人",
+      "▁rec urring",
+      "精 湛",
+      "维 奇",
+      "▁m ont",
+      "▁mon t",
+      "▁mo nt",
+      "▁ mont",
+      "▁ 钟",
+      "抵 制",
+      "zz le",
+      "▁De tailed",
+      "▁Det ailed",
+      "▁Detail ed",
+      "深 厚",
+      "▁AD HD",
+      "▁P air",
+      "▁Pa ir",
+      "▁ Pair",
+      "olec ule",
+      "▁ex empt",
+      "砂 糖",
+      "没 那么",
+      "tf rac",
+      "t frac",
+      "▁over looking",
+      "▁overl ooking",
+      "▁overlook ing",
+      "万 像素",
+      "ex cel",
+      "exc el",
+      "ater ials",
+      "aterial s",
+      "▁S SH",
+      "▁SS H",
+      "▁ SSH",
+      "() ){\\",
+      "()) {\\",
+      "( )){\\",
+      "▁ex pr",
+      "▁exp r",
+      "▁ expr",
+      "▁S trip",
+      "▁St rip",
+      "▁Str ip",
+      "▁Stri p",
+      "▁ Strip",
+      "Dig ital",
+      "D igital",
+      "▁country side",
+      "NO NE",
+      "NON E",
+      "N ONE",
+      "▁Wh ats",
+      "▁What s",
+      "LD L",
+      "L DL",
+      "精彩 的",
+      "行政 村",
+      "中 日",
+      "ci a",
+      "c ia",
+      "▁L iv",
+      "▁Li v",
+      "▁sh ooter",
+      "▁shoot er",
+      "▁inter rupt",
+      "右 下",
+      "AM E",
+      "A ME",
+      "看不 见",
+      "看 不见",
+      "ye rs",
+      "yer s",
+      "y ers",
+      "▁Rob ot",
+      "▁Ro bot",
+      "▁ Robot",
+      "▁recover ing",
+      "▁ 把",
+      "你是 否",
+      "你 是否",
+      "tt i",
+      "t ti",
+      "nChar les",
+      "▁conv inc",
+      "nObject ive",
+      "nOb jective",
+      "▁ 关于",
+      "不成 立",
+      "不 成立",
+      "的网络 小说",
+      "的 网络小说",
+      "nout put",
+      "n output",
+      "▁tr ache",
+      "▁tra che",
+      "成 正",
+      "▁F est",
+      "▁Fe st",
+      "Check ed",
+      "▁arr hythm",
+      "将 该",
+      "▁F oo",
+      "▁Fo o",
+      "▁ Foo",
+      "ught y",
+      "ugh ty",
+      "本院 依职权",
+      "am ming",
+      "amm ing",
+      "金 龙",
+      "集 聚",
+      "wy n",
+      "w yn",
+      "▁}} \">\\",
+      "▁}}\" >\\",
+      "ut ility",
+      "util ity",
+      "u tility",
+      "▁cyt ometry",
+      "▁cyto metry",
+      "蛋 黄",
+      "最好 是",
+      "来 实现",
+      "▁l ime",
+      "▁li me",
+      "▁lim e",
+      "▁ lime",
+      "er nels",
+      "ern els",
+      "ernel s",
+      "op enia",
+      "ope nia",
+      "open ia",
+      "▁incre mental",
+      "▁increment al",
+      "ö n",
+      "爱 我",
+      "▁L emma",
+      "▁Le mma",
+      "▁Lem ma",
+      "▁tox ins",
+      "▁toxin s",
+      "办 案",
+      "jd bc",
+      "j dbc",
+      "er many",
+      "erm any",
+      "erman y",
+      "▁local host",
+      "▁ localhost",
+      "外 星",
+      "nH ot",
+      "n Hot",
+      "▁diss ociation",
+      "HO T",
+      "H OT",
+      "▁incon ven",
+      "{{ '",
+      "{ {'",
+      "寝 室",
+      "收 敛",
+      "艺术 的",
+      "before Each",
+      "▁Hend erson",
+      "▁share holders",
+      "▁shareholder s",
+      "▁socio economic",
+      "▁elect rochemical",
+      "▁electro chemical",
+      "▁electr ochemical",
+      "▁electroc hemical",
+      "用 自己的",
+      "重 度",
+      "▁电 源",
+      "▁ 电源",
+      "▁me rit",
+      "▁mer it",
+      "k B",
+      "反 诉",
+      "节 的",
+      "香 油",
+      "To ur",
+      "T our",
+      "所以 我很",
+      "所以我 很",
+      "▁reconstruct ed",
+      "香 水",
+      "方法 是",
+      "UI Kit",
+      "▁C MOS",
+      "▁CM OS",
+      "el ected",
+      "elect ed",
+      "e lected",
+      "▁ δ",
+      "对象 的",
+      "▁l ent",
+      "▁le nt",
+      "▁len t",
+      "▁moderate ly",
+      "▁moder ately",
+      "三 层",
+      "腰 椎",
+      "▁G S",
+      "▁ GS",
+      "医疗 保险",
+      "研究 表明",
+      "▁neg lected",
+      "▁neglect ed",
+      "革 新",
+      "中国特色 社会主义",
+      "并且 在",
+      "并 且在",
+      "Com pleted",
+      "Comp leted",
+      "Complete d",
+      "▁De pression",
+      "▁Dep ression",
+      "知 晓",
+      "外国 语",
+      "ti on",
+      "t ion",
+      "ys ql",
+      "y sql",
+      "证据 证明",
+      "▁never theless",
+      "ur nal",
+      "urn al",
+      "▁Hen ri",
+      "汇 集",
+      "苏 宁",
+      "麻 将",
+      "/** /*",
+      "nr un",
+      "n run",
+      "研究 人员",
+      "▁book mark",
+      "▁initial ized",
+      "▁initialize d",
+      "▁ initialized",
+      "G Y",
+      "LE FT",
+      "L EFT",
+      "▁dipl omatic",
+      "▁diplom atic",
+      "▁diploma tic",
+      "▁diplomat ic",
+      "干 了",
+      "潜 水",
+      "li s",
+      "l is",
+      "tr ade",
+      "tra de",
+      "trad e",
+      "▁c aut",
+      "▁ca ut",
+      "node js",
+      "▁ref urb",
+      "的 压力",
+      "R atio",
+      "fl are",
+      "f lare",
+      "工程 技术",
+      "职业 技能",
+      "▁小说 进度",
+      "▁ad here",
+      "▁adher e",
+      "▁Her bert",
+      "v w",
+      "晚 报",
+      "{} )",
+      "{ })",
+      "ub ation",
+      "uba tion",
+      "第二百四十 三条",
+      "▁malign ancies",
+      "手 臂",
+      "\"] )",
+      "\" ])",
+      "▁l iner",
+      "▁li ner",
+      "▁line r",
+      "▁lin er",
+      "▁ liner",
+      "OPT IONS",
+      "矿物 质",
+      "矿 物质",
+      "用 脚",
+      "(\" {",
+      "( \"{",
+      "sg i",
+      "s gi",
+      "ne very",
+      "never y",
+      "n every",
+      "▁t owers",
+      "▁to wers",
+      "▁tow ers",
+      "▁tower s",
+      "In ventory",
+      "耽 误",
+      "快乐 的",
+      "▁T um",
+      "▁Tu m",
+      "▁W R",
+      "▁ WR",
+      "市场 监管",
+      "nCh ris",
+      "n Chris",
+      "▁Bas ically",
+      "▁Basic ally",
+      "强 奸",
+      "抽 烟",
+      "山县 人民法院",
+      "山 县人民法院",
+      "▁zb MATH",
+      "uc a",
+      "u ca",
+      "▁其 五",
+      "er tz",
+      "ert z",
+      "▁Moroc co",
+      "PP T",
+      "P PT",
+      "届 人民代表大会",
+      "硅 胶",
+      "▁ 面积",
+      "▁c itation",
+      "▁cit ation",
+      "▁ citation",
+      "IT HOUT",
+      "ITH OUT",
+      "▁differ ing",
+      "▁diff ering",
+      "ah r",
+      "a hr",
+      "▁M HC",
+      "▁MH C",
+      "▁C able",
+      "▁Ca ble",
+      "▁Cab le",
+      "▁Wall ace",
+      "▁clear er",
+      "▁pro gesterone",
+      "职员 表",
+      "▁J in",
+      "▁Ji n",
+      "en sit",
+      "ens it",
+      "为 证",
+      "格 里",
+      "▁c age",
+      "▁ca ge",
+      "locfile id",
+      "▁Reg arding",
+      "一 举",
+      ")} }",
+      ") }}",
+      "gr an",
+      "gra n",
+      "g ran",
+      "镇 人民政府",
+      "▁A reas",
+      "▁Ar eas",
+      "▁Are as",
+      "▁Area s",
+      "群 里",
+      "肩 膀",
+      "申请执行人 亦",
+      "好 在",
+      "配 电",
+      "∴ \\\\",
+      "▁U UID",
+      "▁ UUID",
+      "▁bo xing",
+      "▁box ing",
+      "全 员",
+      "问 他",
+      "FI L",
+      "F IL",
+      "▁e osin",
+      "▁piv otal",
+      "▁pivot al",
+      "▁Conc rete",
+      "作 案",
+      "▁w eed",
+      "▁we ed",
+      "▁ weed",
+      "县人民检察院 以",
+      "县 人民检察院以",
+      "att ering",
+      "atter ing",
+      "结案 若干问题的意见",
+      "X C",
+      "▁hist ories",
+      "▁histor ies",
+      "▁hi stories",
+      "上 世纪",
+      "当地 的",
+      "当 地的",
+      "曾经 的",
+      "▁D ell",
+      "▁De ll",
+      "▁Del l",
+      "sc hemas",
+      "schema s",
+      "sch emas",
+      "sche mas",
+      "清 凉",
+      "金 水",
+      "台 区",
+      "▁f lew",
+      "▁fl ew",
+      "▁fle w",
+      "▁n y",
+      "▁ ny",
+      "等 症状",
+      "讨论 了",
+      "ts witch",
+      "t switch",
+      "▁Ch icken",
+      "▁Chic ken",
+      "▁H ait",
+      "▁Ha it",
+      "▁Ge netic",
+      "▁Gen etic",
+      "▁Gene tic",
+      "▁orche stra",
+      "▁orchestr a",
+      "ID D",
+      "I DD",
+      "说明 了",
+      "说 明了",
+      "▁Nut rition",
+      "更新 的",
+      "更 新的",
+      "▁主要 作品",
+      "▁ 主要作品",
+      "n assert",
+      "设 想",
+      "ae a",
+      "a ea",
+      "iy a",
+      "i ya",
+      "分析 师",
+      "全国 联保",
+      "现在 已经",
+      "▁餐馆 类型",
+      "▁fa ults",
+      "▁fault s",
+      "} <",
+      "带 你",
+      "胃 肠",
+      "农民 人均纯收入",
+      "▁Agricult ural",
+      "▁Agr icultural",
+      "矿 石",
+      "不正 确",
+      "不 正确",
+      "水平 和",
+      "水 平和",
+      "ys es",
+      "yse s",
+      "y ses",
+      "▁N athan",
+      "▁Na than",
+      "▁Nat han",
+      "▁Nath an",
+      "▁hazard ous",
+      "回 执",
+      "有 小",
+      "首 饰",
+      "预交 上诉案件受理费",
+      "▁с т",
+      "▁ ст",
+      "Ti ck",
+      "T ick",
+      "nE NV",
+      "n ENV",
+      "nN ever",
+      "nNe ver",
+      "n Never",
+      "re store",
+      "rest ore",
+      "▁Lou ise",
+      "▁Louis e",
+      "▁poll ut",
+      "ieve ment",
+      "iev ement",
+      "活 血",
+      "起 着",
+      "es ch",
+      "esc h",
+      "e sch",
+      "np lot",
+      "npl ot",
+      "n plot",
+      "▁ill eg",
+      "▁il leg",
+      "地 在",
+      "发展 规划",
+      "at ype",
+      "aty pe",
+      "a type",
+      "▁graft s",
+      "▁gra fts",
+      "Type d",
+      "Typ ed",
+      "Ty ped",
+      "T yped",
+      "▁Ur ugu",
+      "▁euro s",
+      "▁eu ros",
+      "▁F actors",
+      "▁Factor s",
+      "▁Fa ctors",
+      "▁Fact ors",
+      "共同 努力",
+      "ri ans",
+      "ria ns",
+      "rian s",
+      "r ians",
+      "yl ate",
+      "yla te",
+      "y late",
+      "▁K rish",
+      "▁Kr ish",
+      "▁Kris h",
+      "▁com posites",
+      "▁compos ites",
+      "▁composite s",
+      "与 我",
+      "增强 了",
+      "无 机",
+      "致 辞",
+      "AS P",
+      "A SP",
+      "二 进制",
+      "/\\ ',",
+      "/\\' ,",
+      "/ \\',",
+      "丰富 多彩",
+      "▁coh orts",
+      "▁cohort s",
+      "doc uments",
+      "document s",
+      "▁celebr ations",
+      "▁celebration s",
+      "▁re agents",
+      "▁reagent s",
+      "恶 化",
+      "还 行",
+      "aa a",
+      "a aa",
+      "增长 率",
+      "fe st",
+      "f est",
+      "ri dor",
+      "rid or",
+      "r idor",
+      "▁bl onde",
+      "▁blo nde",
+      "社 科",
+      "留下 的",
+      "留 下的",
+      "▁Out look",
+      "巴 士",
+      "教育 和",
+      "View er",
+      "nRe cent",
+      "nRec ent",
+      "n Recent",
+      "▁faith ful",
+      "▁summar ize",
+      "V o",
+      "▁ 수",
+      "不 完",
+      "勾 选",
+      "▁ 人均价格",
+      "nCh inese",
+      "n Chinese",
+      "帷 幕",
+      "nM art",
+      "nMar t",
+      "nMa rt",
+      "n Mart",
+      "ro ots",
+      "root s",
+      "organ isms",
+      "伪 造",
+      "向 阳",
+      "ee per",
+      "eep er",
+      "e eper",
+      "▁Y ugoslav",
+      "▁head line",
+      "妇 联",
+      "Role s",
+      "Ro les",
+      "R oles",
+      "nN ice",
+      "att ack",
+      "atta ck",
+      "file Path",
+      "▁Treas ury",
+      "▁real ised",
+      "▁realise d",
+      "▁Ro chester",
+      "▁Roc hester",
+      "▁det ailing",
+      "▁detail ing",
+      "▁ 老",
+      "又 被",
+      "字 的",
+      "自 考",
+      "▁multiple x",
+      "▁multi plex",
+      "▁multip lex",
+      "▁ 汪",
+      "一 说",
+      "议 题",
+      "青 云",
+      "Bo b",
+      "B ob",
+      "GM P",
+      "G MP",
+      "法定代表人 刘",
+      "▁c anine",
+      "▁can ine",
+      "▁ca nine",
+      "▁implant ed",
+      "▁impl anted",
+      "le ston",
+      "les ton",
+      "lest on",
+      "l eston",
+      "新 娘",
+      "▁ §",
+      "半径 为",
+      "建立 起",
+      "▁pol ish",
+      "▁po lish",
+      "Conn ected",
+      "Connect ed",
+      "党 风",
+      "反 向",
+      "ok emon",
+      "oke mon",
+      "▁Gen eva",
+      "▁Gene va",
+      "▁re called",
+      "▁rec alled",
+      "▁recall ed",
+      "少年 儿童",
+      "▁R ating",
+      "▁Ra ting",
+      "▁Rat ing",
+      "▁ Rating",
+      "冰 糖",
+      "/_ _",
+      "/ __",
+      "Ba g",
+      "B ag",
+      "审定 委员会",
+      "▁traged y",
+      "]$ $",
+      "] $$",
+      "但是 他",
+      "但 是他",
+      ")] ,",
+      ") ],",
+      "▁At las",
+      "▁Atl as",
+      "▁Ti bet",
+      "▁Tib et",
+      "落 到",
+      "雇 佣",
+      "▁N V",
+      "▁ NV",
+      "▁b ees",
+      "▁be es",
+      "▁bee s",
+      "专业 委员会",
+      "am iliar",
+      "amil iar",
+      "永 不",
+      "▁Wars aw",
+      "n metadata",
+      "▁rel iably",
+      "▁WH AT",
+      "pat terns",
+      "pattern s",
+      "p atterns",
+      "位 是",
+      "式 中",
+      "立即 发生法律效力",
+      "立 即发生法律效力",
+      "language s",
+      "l anguages",
+      "Trans former",
+      "Transform er",
+      "▁[ #",
+      "▁ [#",
+      "▁analy zes",
+      "▁analyz es",
+      "▁analyze s",
+      "▁mis match",
+      "▁mism atch",
+      "▁Sh ape",
+      "▁Sha pe",
+      "▁ Shape",
+      "▁space craft",
+      "▁spac ecraft",
+      "英 格兰",
+      "原告 主张",
+      "▁j ets",
+      "▁je ts",
+      "▁jet s",
+      "▁Pl astic",
+      "▁Plas tic",
+      "▁panc reas",
+      "乔 治",
+      "峡 谷",
+      "纸 质",
+      "▁aw ful",
+      "交 代",
+      "…… ”",
+      "▁ 楼盘",
+      "th ag",
+      "tha g",
+      "bol ds",
+      "bold s",
+      "b olds",
+      "nL ooks",
+      "nLook s",
+      "nLo oks",
+      "▁en cephal",
+      "▁ encephal",
+      "为 首",
+      "乐 团",
+      "遇 上",
+      "不能 成立",
+      "or man",
+      "orm an",
+      "nEval uate",
+      "n Evaluate",
+      "黄 山",
+      "是一 样的",
+      "是 一样的",
+      "nInt roduction",
+      "n Introduction",
+      "▁p are",
+      "▁par e",
+      "▁pa re",
+      "▁ pare",
+      "混 沌",
+      "lic an",
+      "li can",
+      "l ican",
+      "Wal let",
+      "Wall et",
+      "▁\\' \\',\\",
+      "▁\\'\\' ,\\",
+      "▁\\'\\ ',\\",
+      "▁\\'\\', \\",
+      "atern ion",
+      "▁CON TRIBUTORS",
+      "Code s",
+      "Co des",
+      "Cod es",
+      "C odes",
+      "▁And rews",
+      "▁Andrew s",
+      "▁Andre ws",
+      "▁Andr ews",
+      "▁Brad ley",
+      "▁repe ating",
+      "▁repeat ing",
+      "三 明",
+      "nT ur",
+      "n Tur",
+      "yn ec",
+      "y nec",
+      "▁unique ness",
+      "▁uniqu eness",
+      "劳动 关系",
+      "交纳 案件受理费",
+      "▁Exec ute",
+      "▁ Execute",
+      "▁renov ated",
+      "光 照",
+      "购 置",
+      "nP rim",
+      "nPr im",
+      "n Prim",
+      "▁H ide",
+      "▁Hi de",
+      "▁ Hide",
+      "▁special izing",
+      "▁aden ocarcinoma",
+      "im ento",
+      "ime nto",
+      "iment o",
+      "imen to",
+      "对 付",
+      "洛 克",
+      "▁ 史",
+      "互 通",
+      "探 险",
+      "比利 时",
+      "to ur",
+      "t our",
+      "▁k ills",
+      "▁kil ls",
+      "▁kill s",
+      "review ed",
+      "学 姐",
+      "▁B eer",
+      "▁Be er",
+      "▁Bee r",
+      "珠海 市",
+      "珠 海市",
+      "▁C ho",
+      "▁Ch o",
+      "▁ Cho",
+      "▁B one",
+      "▁Bo ne",
+      "▁Bon e",
+      "cri bing",
+      "crib ing",
+      "▁Mar athon",
+      "旺 盛",
+      "讽 刺",
+      "In vest",
+      "Inv est",
+      "▁In door",
+      "▁Ind oor",
+      "▁ П",
+      "潇 洒",
+      "▁[ …",
+      "▁C um",
+      "▁Cu m",
+      "an son",
+      "ans on",
+      "a nson",
+      "▁cl oning",
+      "▁m otions",
+      "▁mot ions",
+      "▁motion s",
+      "▁pres enter",
+      "▁present er",
+      "▁Partners hip",
+      "▁Partner ship",
+      "▁Indian apolis",
+      "是一 些",
+      "是 一些",
+      "nC am",
+      "n Cam",
+      "▁par ab",
+      "▁para b",
+      "▁pa rab",
+      "Ch ecker",
+      "Check er",
+      "不 肯",
+      "其 后",
+      "微量 元素",
+      "▁l obe",
+      "▁lo be",
+      "▁lob e",
+      "▁m aths",
+      "▁mat hs",
+      "▁math s",
+      "▁ma ths",
+      "构 筑",
+      "▁cand le",
+      "▁可 以",
+      "▁ 可以",
+      "nsup ported",
+      "n supported",
+      "点 是",
+      "还 想",
+      "▁c hol",
+      "▁ch ol",
+      "▁cho l",
+      "▁ chol",
+      "▁remark ably",
+      "目前 已",
+      "虽然 是",
+      "▁[ ])\\",
+      "▁[] )\\",
+      "▁[]) \\",
+      "▁ [])\\",
+      "▁o dor",
+      "▁od or",
+      "▁ odor",
+      ",$ $",
+      ", $$",
+      "Be r",
+      "B er",
+      "▁ex ponents",
+      "▁exponent s",
+      "手机 的",
+      "手 机的",
+      "▁U FC",
+      "▁b oom",
+      "▁bo om",
+      "end foreach",
+      "不 全",
+      "▁free style",
+      "▁fre estyle",
+      "大 发",
+      "桥 区",
+      "发生 后",
+      "▁T RA",
+      "▁TR A",
+      "▁ TRA",
+      "▁G ross",
+      "▁Gr oss",
+      "▁Gro ss",
+      "▁e last",
+      "▁el ast",
+      "一 门",
+      "大 叔",
+      "环境 和",
+      "介绍 一下",
+      "对其 进行",
+      "char At",
+      "打 游戏",
+      "房屋 租赁",
+      "搜 狐",
+      "的 证言",
+      "Key board",
+      "减 排",
+      "较 差",
+      "每天 都",
+      "or is",
+      "ori s",
+      "o ris",
+      "计算 能力",
+      "ms gs",
+      "msg s",
+      "▁C ock",
+      "▁Co ck",
+      "▁Coc k",
+      "不等式 的解",
+      "不等式的 解",
+      "▁s ourced",
+      "▁source d",
+      "▁sour ced",
+      "而 去",
+      "命题 的",
+      "学会 了",
+      "▁re ef",
+      "xf f",
+      "x ff",
+      "▁tum ours",
+      "▁tumour s",
+      "mb H",
+      "te k",
+      "t ek",
+      "图形 的",
+      "图 形的",
+      "▁col oured",
+      "▁colour ed",
+      "好 几",
+      "是一 所",
+      "是 一所",
+      "▁ac rylic",
+      "物业服务 合同纠纷一案中",
+      "物业服务合同纠纷一案 中",
+      "重 力",
+      "Pe ter",
+      "Pet er",
+      "P eter",
+      "▁tow el",
+      "▁mart ial",
+      "将 为",
+      "oy l",
+      "o yl",
+      "等 离子",
+      "▁S ett",
+      "▁Se tt",
+      "▁Set t",
+      "▁S andy",
+      "▁San dy",
+      "▁Sand y",
+      "强 国",
+      "海 燕",
+      "颈 部",
+      ")\\ '",
+      ") \\'",
+      "ac qu",
+      "di n",
+      "d in",
+      "nP urpose",
+      "▁Par allel",
+      "▁ Parallel",
+      "共 用",
+      "一种 是",
+      "不用 担心",
+      "▁c ares",
+      "▁car es",
+      "▁care s",
+      "▁ca res",
+      "▁B elt",
+      "▁Be lt",
+      "▁Bel t",
+      "▁overs ight",
+      "头 晕",
+      "清 香",
+      "纵 坐标",
+      "风景 区",
+      "风 景区",
+      "nS tat",
+      "nSt at",
+      "nSta t",
+      "n Stat",
+      "cult ural",
+      "c ultural",
+      "派 的",
+      "nint er",
+      "ni nter",
+      "nin ter",
+      "n inter",
+      "▁phot ons",
+      "▁photo ns",
+      "▁photon s",
+      "工商行政管理局 登记成立",
+      "Token s",
+      "Tok ens",
+      "Per formance",
+      "以 致",
+      "必须 是",
+      "im edia",
+      "i media",
+      "▁h atch",
+      "▁hat ch",
+      "▁St ress",
+      "▁Str ess",
+      "▁Stre ss",
+      "丝 毫",
+      "两 张",
+      "▁P orter",
+      "▁Port er",
+      "▁Por ter",
+      "师 大",
+      "ci ble",
+      "c ible",
+      "▁me nus",
+      "▁men us",
+      "▁menu s",
+      "▁differ entially",
+      "▁different ially",
+      "▁differential ly",
+      "提 及",
+      "▁成 语",
+      "▁ 成语",
+      "线 城市",
+      "▁Jo el",
+      "▁Joe l",
+      "▁I Enumerable",
+      "▁ IEnumerable",
+      "dev Dependencies",
+      "入 境",
+      "山 庄",
+      "▁Mar co",
+      "▁Marc o",
+      "oth ermal",
+      "other mal",
+      "otherm al",
+      "o thermal",
+      "号 执行",
+      "▁HE AD",
+      "▁ HEAD",
+      "Dan iel",
+      "D aniel",
+      "▁Manit oba",
+      "▁S ullivan",
+      "会 把",
+      "理 科",
+      "心理 咨询",
+      "法定代表人 陈",
+      "▁success es",
+      "▁succ esses",
+      "▁ 登场",
+      "项目 中",
+      "ull ah",
+      "ulla h",
+      "房 贷",
+      "虽 说",
+      "会 觉得",
+      "▁y outube",
+      "▁you tube",
+      "▁ youtube",
+      "NO N",
+      "N ON",
+      "▁Sim pson",
+      "▁normal ization",
+      "加 分",
+      "因 家庭",
+      "方式 的",
+      "方 式的",
+      "an tha",
+      "ant ha",
+      "anth a",
+      "nR eal",
+      "nRe al",
+      "n Real",
+      "▁H off",
+      "▁Ho ff",
+      "▁Ab bey",
+      "▁Abb ey",
+      "▁d olor",
+      "▁do lor",
+      "▁dol or",
+      "▁h edge",
+      "▁hed ge",
+      "Contract s",
+      "oc ellular",
+      "o cellular",
+      "▁c ouncill",
+      "▁counc ill",
+      "▁council l",
+      "nS ir",
+      "▁C AT",
+      "▁CA T",
+      "▁ CAT",
+      "▁famil ial",
+      "bold symbol",
+      "bolds ymbol",
+      "它 可以",
+      "▁北京 市",
+      "▁ 北京市",
+      "而来 的",
+      "而 来的",
+      "con straints",
+      "constraint s",
+      "▁Fellow ship",
+      "技术 与",
+      "炸 弹",
+      "不得 不说",
+      "不得不 说",
+      "中国 银行",
+      "com mercial",
+      "▁inter acts",
+      "▁interact s",
+      "一 到",
+      "马 云",
+      "ac id",
+      "aci d",
+      "a cid",
+      "晋 级",
+      "▁life long",
+      "▁lif elong",
+      "▁be verages",
+      "▁beverage s",
+      "安装 的",
+      "安 装的",
+      "am ines",
+      "amin es",
+      "amine s",
+      "ami nes",
+      "▁jud ged",
+      "▁judge d",
+      "▁judg ed",
+      "中 了",
+      "▁W hole",
+      "▁Wh ole",
+      "▁Who le",
+      "▁veloc ities",
+      "Boolean Field",
+      "吸 食",
+      "多 重",
+      "靠 着",
+      "这几 年",
+      "这 几年",
+      "▁N atal",
+      "▁Na tal",
+      "▁Nat al",
+      "▁M é",
+      "el ize",
+      "eli ze",
+      "总 建筑面积",
+      "nM ethod",
+      "n Method",
+      "▁alter ing",
+      "▁alt ering",
+      "▁un available",
+      "甜 甜",
+      "▁s yll",
+      "▁sy ll",
+      "党支部 书记",
+      "ophag y",
+      "▁W arning",
+      "▁War ning",
+      "▁ Warning",
+      "▁spread sheet",
+      "▁spreads heet",
+      "▁collect ively",
+      "▁collective ly",
+      "m w",
+      "很 是",
+      "nP an",
+      "n Pan",
+      "▁M ang",
+      "▁Man g",
+      "▁Ma ng",
+      "spec ified",
+      "na g",
+      "n ag",
+      "▁M ight",
+      "▁Mi ght",
+      "▁Mig ht",
+      "▁ ∼",
+      "尔 特",
+      "更 具",
+      "Fig ure",
+      "F igure",
+      "晚 餐",
+      "骗 子",
+      "中国 古代",
+      "aff le",
+      "st uff",
+      "▁balcon y",
+      "标 配",
+      "▁ 免费",
+      "泡 泡",
+      "看 他",
+      "中关村 在线",
+      "Sc heme",
+      "Sch eme",
+      "妥 协",
+      "转 到",
+      "未 约定",
+      "▁Sup erior",
+      "▁Super ior",
+      "▁Redist ributions",
+      "加 水",
+      "懒 得",
+      "比 你",
+      "ch ef",
+      "che f",
+      "▁sp aw",
+      "▁spa w",
+      "幼 稚",
+      "Set s",
+      "Se ts",
+      "S ets",
+      "▁s cent",
+      "▁sc ent",
+      "▁scen t",
+      "▁ins ured",
+      "海 绵",
+      "hr a",
+      "h ra",
+      "旗 人民法院",
+      "▁so vere",
+      "T a",
+      "人 有",
+      "贴 心",
+      "履行 了",
+      "履 行了",
+      "不是 一个",
+      "不 是一个",
+      "得 益",
+      "自 媒体",
+      "us an",
+      "usa n",
+      "u san",
+      "nM ath",
+      "nMat h",
+      "nMa th",
+      "n Math",
+      "▁B attery",
+      "▁Bat tery",
+      "▁Batt ery",
+      "Fa t",
+      "F at",
+      "▁P aint",
+      "▁Pa int",
+      "▁Pain t",
+      "▁ Paint",
+      "▁n ginx",
+      "▁ng inx",
+      "▁ nginx",
+      "▁bre athe",
+      "▁breath e",
+      "▁breat he",
+      "鱼 的",
+      "oa k",
+      "o ak",
+      "来到 了一",
+      "来到了 一",
+      "以上事实 有",
+      "ience d",
+      "ien ced",
+      "i enced",
+      "log ged",
+      "安 防",
+      "提 案",
+      "充分 的",
+      "充 分的",
+      "win gs",
+      "wing s",
+      "w ings",
+      "amp agne",
+      "tdef ault",
+      "t default",
+      "reg istered",
+      "register ed",
+      "存 活",
+      "艾 滋",
+      "万元 及利息",
+      "万 元及利息",
+      "▁Ac tual",
+      "▁Act ual",
+      "▁ Actual",
+      "▁cas inos",
+      "▁casino s",
+      "可以自收到 裁定书",
+      "▁witness es",
+      "栖 息",
+      "for mer",
+      "form er",
+      "Creat ion",
+      "Cre ation",
+      "C reation",
+      "▁ex posing",
+      "▁exp osing",
+      "▁Bulg arian",
+      "▁Bulgaria n",
+      "病 虫",
+      "uc umber",
+      "▁Ex pect",
+      "▁Exp ect",
+      "▁ Expect",
+      "z A",
+      "Se par",
+      "Sep ar",
+      "r g",
+      "再 遭",
+      "梯 形",
+      "长 辈",
+      "▁s ich",
+      "▁si ch",
+      "▁sul ph",
+      "劳动 人事",
+      "浦东 新区",
+      "▁d ancers",
+      "▁dance rs",
+      "▁dan cers",
+      "▁dancer s",
+      "▁wr apping",
+      "▁wrap ping",
+      "▁dis section",
+      "▁diss ection",
+      "▁dissect ion",
+      "世 代",
+      "增 收",
+      "红 利",
+      "Sm ith",
+      "S mith",
+      "刑 满",
+      "▁s ore",
+      "▁so re",
+      "▁sor e",
+      "y k",
+      "统一 的",
+      "统 一的",
+      "▁sil ica",
+      "▁E E",
+      "▁ EE",
+      "四条 的规定",
+      "四 条的规定",
+      "onom ical",
+      "onomic al",
+      "▁F requency",
+      "▁ Frequency",
+      "视 网",
+      "ma le",
+      "mal e",
+      "m ale",
+      "轨道 交通",
+      "▁catal ogue",
+      "▁catalog ue",
+      "汇 款",
+      "▁M LS",
+      "▁ML S",
+      "Ch apter",
+      "▁E instein",
+      "{ |",
+      "写 法",
+      "困难 的",
+      "困 难的",
+      "▁prem ise",
+      "▁I stanbul",
+      "▁immun ization",
+      "出 炉",
+      "每 当",
+      "这个 是",
+      "▁Kaz akh",
+      "▁th rust",
+      "▁thr ust",
+      "▁thru st",
+      "灯 具",
+      "加 之",
+      "平 滑",
+      "血 浆",
+      "▁cr ore",
+      "▁cro re",
+      "魔 兽",
+      "工作 日",
+      "ov sky",
+      "ovsk y",
+      "适 时",
+      "Qu ality",
+      "Qual ity",
+      "Q uality",
+      "▁Boot strap",
+      "▁ Bootstrap",
+      "Or g",
+      "O rg",
+      "有多 大",
+      "有 多大",
+      "▁le ap",
+      "q p",
+      "几年 来",
+      "几 年来",
+      "▁\\' *",
+      "▁m asters",
+      "▁master s",
+      "▁ma sters",
+      "▁mast ers",
+      "▁mas ters",
+      "▁logarith m",
+      "开 支",
+      "▁C raw",
+      "▁Cr aw",
+      "▁Cra w",
+      "ide press",
+      "↓ ↓",
+      "▁Ch erry",
+      "▁Cher ry",
+      "如果 要",
+      "护肤 品",
+      "管理 法",
+      "▁loc king",
+      "▁lock ing",
+      "▁ locking",
+      "掀 起",
+      "IE EE",
+      "I EEE",
+      "▁cyst ic",
+      "应 准予",
+      "▁call er",
+      "▁ca ller",
+      "▁cal ler",
+      "电 量",
+      "▁g inger",
+      "▁ging er",
+      "▁gin ger",
+      "▁cere mon",
+      "▁cer emon",
+      "▁negot iate",
+      "▁program med",
+      "▁programme d",
+      "源 县",
+      "准备 工作",
+      "尤其 是在",
+      "尤其是 在",
+      "sa ving",
+      "sav ing",
+      "s aving",
+      "▁St ats",
+      "▁Stat s",
+      "▁Sta ts",
+      "▁ Stats",
+      "▁superf icial",
+      "不利 于",
+      "不 利于",
+      "▁micro g",
+      "▁mic rog",
+      "所 用",
+      "▁p hi",
+      "▁ph i",
+      "▁ phi",
+      "▁W alt",
+      "▁Wal t",
+      "▁Wa lt",
+      "们 都",
+      "大 爷",
+      "犹 太",
+      "cy cles",
+      "cycle s",
+      "cycl es",
+      "c ycles",
+      "rient ation",
+      "▁class rooms",
+      "▁classroom s",
+      "▁d well",
+      "▁dw ell",
+      "Fa r",
+      "F ar",
+      "对 这个",
+      "red ients",
+      "redient s",
+      "▁c ognition",
+      "▁cogn ition",
+      "的 身份",
+      "am en",
+      "ame n",
+      "a men",
+      "主 教",
+      "之 余",
+      "总的 来说",
+      "经济 管理",
+      "\\\\ {\\\\",
+      "\\\\{ \\\\",
+      "ay ette",
+      "aye tte",
+      "fix tures",
+      "fixture s",
+      "at ha",
+      "ath a",
+      "a tha",
+      "u w",
+      "]$ \\",
+      "] $\\",
+      "ns im",
+      "n sim",
+      "▁l ute",
+      "▁lu te",
+      "▁port raits",
+      "▁por traits",
+      "▁portrait s",
+      "commun ication",
+      ",\\\\ ,",
+      ", \\\\,",
+      "nl ong",
+      "n long",
+      "油 菜",
+      "au de",
+      "aud e",
+      "a ude",
+      "y ahoo",
+      "二〇一七年 九月",
+      "二〇一七年九 月",
+      "▁Ed monton",
+      "▁Arg uments",
+      "▁Argument s",
+      "▁ Arguments",
+      "不 绝",
+      "SR C",
+      "S RC",
+      "国家和 地区",
+      "res sing",
+      "ress ing",
+      "r essing",
+      "▁re built",
+      "▁reb uilt",
+      "▁f initely",
+      "▁fin itely",
+      "▁finite ly",
+      "▁locom otives",
+      "▁locomotive s",
+      "切 入",
+      "那 位",
+      "▁D L",
+      "▁ DL",
+      "▁ 角色",
+      "是否 存在",
+      "tt est",
+      "ttes t",
+      "tte st",
+      "t test",
+      "le ader",
+      "lead er",
+      "lea der",
+      "l eader",
+      "incre ment",
+      "▁dis close",
+      "▁disc lose",
+      "哥 伦",
+      "秒 钟",
+      "微 商",
+      "打 着",
+      "sn ap",
+      "s nap",
+      "▁H ugh",
+      "▁Hu gh",
+      "▁Hug h",
+      "▁m usc",
+      "▁mus c",
+      "▁mu sc",
+      "▁sh iny",
+      "▁W aters",
+      "▁Water s",
+      "▁Wat ers",
+      "▁Wa ters",
+      "threat ening",
+      "▁willing ness",
+      "fu rt",
+      "fur t",
+      "f urt",
+      "Mon go",
+      "Mo ngo",
+      "M ongo",
+      "qu ito",
+      "quit o",
+      "▁Con structor",
+      "▁Construct or",
+      "▁ Constructor",
+      "▁character izing",
+      "uc he",
+      "uch e",
+      "u che",
+      "婚姻 关系",
+      "▁hom ologous",
+      "▁homolog ous",
+      "草 地",
+      "对 企业",
+      "横 坐标",
+      "▁pot ency",
+      "▁de central",
+      "▁dec entral",
+      "▁decent ral",
+      "▁research ing",
+      "▁resear ching",
+      "▁A br",
+      "▁Ab r",
+      "▁U TF",
+      "▁UT F",
+      "▁ UTF",
+      "▁U EFA",
+      "▁At omic",
+      "▁Atom ic",
+      "▁re side",
+      "▁res ide",
+      "▁resid e",
+      "▁back drop",
+      "▁ 云",
+      "▁re un",
+      "▁acc us",
+      "▁ac cus",
+      "▁in box",
+      "▁concomit ant",
+      "▁intermitt ent",
+      "▁if f",
+      "▁i ff",
+      "▁ iff",
+      "ra ises",
+      "raise s",
+      "rais es",
+      "▁E A",
+      "▁ EA",
+      "ench ymal",
+      "▁Al cohol",
+      "～ ～",
+      "叙 利亚",
+      "▁comp iling",
+      "往 后",
+      "ri cia",
+      "ric ia",
+      "r icia",
+      "不受申请执行 时效",
+      "nAdd itional",
+      "n Additional",
+      "local Storage",
+      "遗 忘",
+      "et z",
+      "e tz",
+      "▁山 东",
+      "▁ 山东",
+      "▁MAT LAB",
+      "▁lapar oscopic",
+      "审 慎",
+      "uth ors",
+      "uthor s",
+      "tex cept",
+      "t except",
+      "▁att ent",
+      "▁atten t",
+      "下 达",
+      "四 处",
+      "op p",
+      "o pp",
+      "受 不了",
+      "Fr ank",
+      "F rank",
+      "▁T ube",
+      "▁Tu be",
+      "▁Tub e",
+      "▁ Tube",
+      "yn aptic",
+      "yna ptic",
+      "▁Po well",
+      "▁Pow ell",
+      "font Size",
+      "▁applic ability",
+      "A u",
+      "盘 中",
+      "结 实",
+      "▁A BO",
+      "▁AB O",
+      "▁J unction",
+      "▁Jun ction",
+      "人 社",
+      "催 收",
+      "隐 瞒",
+      "直 升",
+      "▁f ried",
+      "▁fr ied",
+      "▁fri ed",
+      "▁ fried",
+      "是 想",
+      "齐 齐",
+      "UN K",
+      "U NK",
+      "ol ation",
+      "ola tion",
+      "o lation",
+      "▁Art ificial",
+      "▁Special ist",
+      "▁disp utes",
+      "▁disput es",
+      "▁dispute s",
+      "按 中国人民银行同期",
+      "口 水",
+      "打 个",
+      "大幅 度",
+      "大 幅度",
+      "▁mm Hg",
+      "▁bul bs",
+      "▁bulb s",
+      "upy text",
+      "图 纸",
+      "大 自然",
+      "未 答辩",
+      "▁val or",
+      "▁va lor",
+      "面 条",
+      "hu a",
+      "h ua",
+      "市场监督管理局 登记成立",
+      "▁cr acks",
+      "▁crack s",
+      "▁cra cks",
+      "回 购",
+      "城 南",
+      "Co s",
+      "C os",
+      "▁fort unate",
+      "干 活",
+      "fl o",
+      "f lo",
+      "的 个人",
+      "int endent",
+      "可 为",
+      "比 上",
+      "▁rest oring",
+      "▁restor ing",
+      "拇 指",
+      "东 亚",
+      "梅 西",
+      "窗 帘",
+      "DD R",
+      "D DR",
+      "▁ 选择",
+      "马拉 松",
+      "▁l ender",
+      "▁le nder",
+      "▁len der",
+      "▁lend er",
+      "▁explo ited",
+      "▁exploit ed",
+      "▁Ab raham",
+      "师 父",
+      "日 军",
+      "▁» \\",
+      "事实 与",
+      "降水 量",
+      "降 水量",
+      "▁L um",
+      "▁Lu m",
+      "am ycin",
+      "amy cin",
+      "▁N ames",
+      "▁Name s",
+      "▁Na mes",
+      "▁Nam es",
+      "▁ Names",
+      "Stat istics",
+      "▁install er",
+      "▁ installer",
+      "▁ э",
+      "强 者",
+      "▁z e",
+      "▁ ze",
+      "音乐 学院",
+      "yard s",
+      "yar ds",
+      "y ards",
+      "▁Som erset",
+      "亦 有",
+      "依法 予以",
+      "▁apopt otic",
+      "▁dis charged",
+      "▁discharge d",
+      "▁discharg ed",
+      "▁ ب",
+      "知道 的",
+      "知 道的",
+      "at su",
+      "ats u",
+      "常住 人口",
+      "▁ag rees",
+      "▁agre es",
+      "▁agree s",
+      "▁agr ees",
+      "▁M X",
+      "▁ MX",
+      "几十 年",
+      "几 十年",
+      "ick ness",
+      "发 文",
+      "奇 的",
+      "蓬 莱",
+      "下一 代",
+      "下 一代",
+      "作 他用",
+      "所 说",
+      "百度 地图",
+      "这 意味着",
+      "at terns",
+      "atter ns",
+      "附 相关法律条文",
+      "▁e psilon",
+      "▁eps ilon",
+      "▁ epsilon",
+      "▁Class ical",
+      "▁Classic al",
+      "再 将",
+      "和 应用",
+      "已 全部",
+      "计划 的",
+      "跟我 说",
+      "跟 我说",
+      "本院 立案",
+      "▁\\' \\';\\",
+      "▁\\'\\' ;\\",
+      "▁\\'\\ ';\\",
+      "▁fresh ly",
+      "之 意",
+      "安 定",
+      "谁 知",
+      "▁pill ow",
+      "▁pil low",
+      "k W",
+      "б о",
+      "ig ated",
+      "igate d",
+      "iga ted",
+      "▁m obil",
+      "▁mo bil",
+      "▁mob il",
+      "▁Method ist",
+      "J D",
+      "人 选",
+      "眼 部",
+      "雄 厚",
+      "le ys",
+      "ley s",
+      "▁R us",
+      "▁Ru s",
+      "存在 于",
+      "存 在于",
+      "正 能量",
+      "ar iat",
+      "ari at",
+      "aria t",
+      "建设 集团有限公司",
+      "▁emigr ants",
+      "nH int",
+      "nHi nt",
+      "n Hint",
+      "成 一",
+      "襄 阳",
+      "▁h yster",
+      "▁hy ster",
+      "▁s pheres",
+      "▁spher es",
+      "▁sphere s",
+      "▁sph eres",
+      "amazon aws",
+      "等 量",
+      "领 土",
+      "做 生意",
+      "历史 文化",
+      "end ering",
+      "ender ing",
+      "ende ring",
+      "▁Spec ies",
+      "Q C",
+      "伊斯 兰",
+      "▁O bs",
+      "▁Ob s",
+      "▁ Obs",
+      "Com plex",
+      "Comp lex",
+      "▁un ited",
+      "▁unit ed",
+      "▁uni ted",
+      "▁g rounded",
+      "▁gr ounded",
+      "▁ground ed",
+      "小说 阅读网",
+      "▁pitch er",
+      "▁pit cher",
+      "▁progen itor",
+      "▁rif le",
+      "▁help ers",
+      "▁hel pers",
+      "▁helper s",
+      "▁ helpers",
+      "▁well being",
+      "奥 会",
+      "DE L",
+      "D EL",
+      "nN A",
+      "n NA",
+      "▁Bel arus",
+      "b h",
+      "▁LI KE",
+      "▁M ate",
+      "▁Mat e",
+      "▁Ma te",
+      "▁ Mate",
+      "▁recogn ise",
+      "四 肢",
+      "湿 润",
+      "▁M ET",
+      "▁ME T",
+      "▁ MET",
+      "▁T ub",
+      "▁Tu b",
+      "▁bud gets",
+      "▁budget s",
+      "▁Bou levard",
+      "▁ 勇",
+      "vi ol",
+      "v iol",
+      "枪 支",
+      "注意 我",
+      "函数 的性质",
+      "函数的 性质",
+      "ho oting",
+      "hoot ing",
+      "h ooting",
+      "▁phil anth",
+      "供 销",
+      "身 的",
+      "ys e",
+      "y se",
+      "纳税 人",
+      "芯片 组",
+      "▁d ile",
+      "▁di le",
+      "▁dil e",
+      "▁re load",
+      "▁rel oad",
+      "▁ reload",
+      "▁poss esses",
+      "▁possess es",
+      "▁trans fusion",
+      "▁transf usion",
+      "金 币",
+      "第一 部",
+      "第 一部",
+      "in strument",
+      "骨 髓",
+      "回到 了",
+      "回 到了",
+      "▁C otton",
+      "▁Cot ton",
+      "▁F itness",
+      "▁Fit ness",
+      "▁sub stituted",
+      "▁substit uted",
+      "▁substitute d",
+      "降 解",
+      "Se n",
+      "S en",
+      "▁Im agine",
+      "▁des criptor",
+      "▁descript or",
+      "▁ descriptor",
+      "固 然",
+      "▁ 系统",
+      "生成 的",
+      "生 成的",
+      "▁D ifference",
+      "正 经",
+      "▁T eh",
+      "▁Te h",
+      "Mark er",
+      "Mar ker",
+      "M arker",
+      "pro fessional",
+      "食 谱",
+      "农业 生产",
+      "挽 回",
+      "▁El lis",
+      "▁Ell is",
+      "▁Elli s",
+      "▁no isy",
+      "终结 本次执行",
+      "前 置",
+      "受 体",
+      "玉 林",
+      "排 版",
+      "热 力",
+      "续行 查封",
+      "不错 的选择",
+      "不错的 选择",
+      "x p",
+      "分 之",
+      "后 代",
+      "航空 公司",
+      "瞄 准",
+      "选择 上述",
+      "党 工委",
+      "的一个 重要",
+      "▁man eu",
+      "SS H",
+      "S SH",
+      "也 表示",
+      "会 发生",
+      "nB ig",
+      "nBi g",
+      "n Big",
+      "it ian",
+      "iti an",
+      "▁S hift",
+      "▁Sh ift",
+      "▁Shi ft",
+      "▁ Shift",
+      "▁Invest igation",
+      "▁Investig ation",
+      "▁i le",
+      "▁il e",
+      "▁ ile",
+      "证明 被告",
+      "Re ason",
+      "nRe cord",
+      "nRec ord",
+      "n Record",
+      "刘 备",
+      "代表 性",
+      "▁circ adian",
+      "▁Mod ified",
+      "▁ Modified",
+      "▁risk y",
+      "▁ris ky",
+      "▁ri sky",
+      "▁v ascul",
+      "▁fore casts",
+      "▁forecast s",
+      "学 前",
+      "订 购",
+      "不要 太",
+      "什么 呢",
+      "hm ad",
+      "hma d",
+      "h mad",
+      "is to",
+      "ist o",
+      "i sto",
+      "not her",
+      "no ther",
+      "n other",
+      "▁Par ents",
+      "▁Parent s",
+      "▁oscill ations",
+      "▁oscillation s",
+      "坐 堂",
+      "武 林",
+      "户 居住",
+      "找 一个",
+      "▁blank et",
+      "▁condition ed",
+      "▁ conditioned",
+      "商 圈",
+      "驾驶 人",
+      "▁B IG",
+      "▁BI G",
+      "最大 限度",
+      "pack ed",
+      "p acked",
+      "绿 茶",
+      "▁F B",
+      "▁ FB",
+      "PI PE",
+      "P IPE",
+      "ar at",
+      "ara t",
+      "a rat",
+      "ans wers",
+      "answer s",
+      "▁un covered",
+      "▁unc overed",
+      "▁uncover ed",
+      "出 走",
+      "▁T ank",
+      "▁Ta nk",
+      "▁Tan k",
+      "tikz picture",
+      "▁recogn izing",
+      "他人 财物",
+      "▁Arch itect",
+      "私 吞",
+      "et les",
+      "Quant ity",
+      "▁Salv ador",
+      "▁activ ator",
+      "▁P olar",
+      "▁Pol ar",
+      "▁Po lar",
+      "▁Pro ps",
+      "▁Pr ops",
+      "▁Prop s",
+      "▁ Props",
+      "per fect",
+      "perf ect",
+      "▁ 分",
+      "国 库",
+      "画 画",
+      "过 点",
+      "特点 是",
+      "特 点是",
+      "我们 应该",
+      "bound s",
+      "b ounds",
+      "cont acts",
+      "contact s",
+      "▁Ex amination",
+      "▁Exam ination",
+      "江 阴",
+      "ap id",
+      "api d",
+      "a pid",
+      "购物 中心",
+      "▁predecess or",
+      "红 豆",
+      "Y ES",
+      "置 换",
+      "\"> [",
+      "\" >[",
+      "th ouse",
+      "t house",
+      "▁z oo",
+      "▁ zoo",
+      "ny ear",
+      "n year",
+      "OB JECT",
+      "O BJECT",
+      "▁ampl ifier",
+      "▁prote sters",
+      "▁prot esters",
+      "▁protest ers",
+      "公布 了",
+      "▁Ang els",
+      "▁Angel s",
+      "▁dr agon",
+      "▁drag on",
+      "▁ dragon",
+      "治疗 效果",
+      "▁discount ed",
+      "F G",
+      "海 口",
+      "▁s ham",
+      "▁sh am",
+      "▁sha m",
+      "借款 到期后",
+      "conn ector",
+      "connect or",
+      "救 治",
+      "作用 的",
+      "作 用的",
+      "▁A CL",
+      "▁AC L",
+      "▁result ant",
+      ">( )\\",
+      ">() \\",
+      "> ()\\",
+      "一个 不断",
+      "del ivery",
+      "▁branch ing",
+      "iv el",
+      "ive l",
+      "i vel",
+      "▁Se oul",
+      "ud os",
+      "udo s",
+      "u dos",
+      "git lab",
+      "qu arter",
+      "这种 方法",
+      "▁He brew",
+      "温 水",
+      "版 次",
+      "豆 浆",
+      "论述 了",
+      "论 述了",
+      "nG reen",
+      "nGr een",
+      "n Green",
+      "▁Ar senal",
+      "▁Un icode",
+      "▁Uni code",
+      "▁ Unicode",
+      "▁myel oid",
+      "á l",
+      "但其 实",
+      "但 其实",
+      "▁pe dag",
+      "▁ped ag",
+      "▁fr amed",
+      "▁frame d",
+      "▁fra med",
+      "CA M",
+      "C AM",
+      "第十 一章",
+      "第十一 章",
+      "第 十一章",
+      "▁Lo an",
+      "真 爱",
+      "op oly",
+      "opol y",
+      "opo ly",
+      "o poly",
+      "争 执",
+      "掉 的",
+      "玩家 们",
+      "▁C afe",
+      "▁Ca fe",
+      "▁Caf e",
+      "▁Int ent",
+      "▁ Intent",
+      "Ref lection",
+      "qr t",
+      "q rt",
+      "着 一个",
+      "▁W are",
+      "▁War e",
+      "▁Wa re",
+      "▁ Ware",
+      "▁sp ice",
+      "铃 声",
+      "le ave",
+      "lea ve",
+      "不 计",
+      "}} )",
+      "} })",
+      "im etry",
+      "ime try",
+      "imet ry",
+      "i metry",
+      "不 远",
+      "群众 的",
+      "个人 觉得",
+      "▁p ivot",
+      "▁piv ot",
+      "▁ab norm",
+      "纪念 馆",
+      "里面 有",
+      "nO ption",
+      "n Option",
+      "▁Com posite",
+      "▁ Composite",
+      "aj o",
+      "a jo",
+      "组织 和",
+      "Be ar",
+      "B ear",
+      "▁G lad",
+      "▁Gl ad",
+      "re solver",
+      "res olver",
+      "resolve r",
+      "fr uit",
+      "f ruit",
+      "▁j upytext",
+      "▁discontin ued",
+      "▁period ically",
+      "▁periodic ally",
+      "控制 在",
+      "已经 开始",
+      "x b",
+      "自 贡",
+      "误 会",
+      "go d",
+      "g od",
+      "P ull",
+      "ro ne",
+      "ron e",
+      "r one",
+      "▁A ires",
+      "▁Air es",
+      "▁minim ally",
+      "▁minimal ly",
+      "▁toler ated",
+      "▁tolerate d",
+      "你可 能",
+      "你 可能",
+      "同时 还",
+      "▁For ty",
+      "▁Fort y",
+      "▁S LE",
+      "▁SL E",
+      "▁s olv",
+      "▁so lv",
+      "▁sol v",
+      "粉 尘",
+      "▁p ts",
+      "▁pt s",
+      "▁ pts",
+      "Di m",
+      "D im",
+      "废弃 了",
+      "ac ycl",
+      "acy cl",
+      "a cycl",
+      "▁man date",
+      "▁mand ate",
+      "▁ma ndate",
+      "科 尔",
+      "▁Philosoph y",
+      "五 千",
+      "出 水",
+      "nA h",
+      "n Ah",
+      "So up",
+      "S oup",
+      "arch ives",
+      "archive s",
+      "进 食",
+      "适量 的",
+      "适 量的",
+      "ab ic",
+      "abi c",
+      "a bic",
+      "im aging",
+      "ima ging",
+      "imag ing",
+      "gi a",
+      "g ia",
+      "nI T",
+      "n IT",
+      "w elcome",
+      "▁tur key",
+      "▁con spiracy",
+      "▁reper toire",
+      "▁quant itatively",
+      "▁quantitative ly",
+      "点 和",
+      "vi v",
+      "v iv",
+      "Al though",
+      "天 子",
+      "培养 学生",
+      "▁nom inal",
+      "▁nomin al",
+      "求 助",
+      "▁$ \\",
+      "▁ $\\",
+      "▁） \\",
+      "▁ ）\\",
+      "nf un",
+      "n fun",
+      "一审 判决",
+      "▁di pole",
+      "▁dip ole",
+      "者 在",
+      "▁D ed",
+      "▁De d",
+      "▁s ock",
+      "▁so ck",
+      "▁soc k",
+      "▁ sock",
+      "▁E mbed",
+      "▁Em bed",
+      "▁Emb ed",
+      "▁ Embed",
+      "▁tr illion",
+      "镇 政府",
+      "▁Min neapolis",
+      "学 问",
+      "伊 拉克",
+      "反映 给",
+      "和 质量",
+      "这样 的人",
+      "这样的 人",
+      "▁A udi",
+      "▁Aud i",
+      "▁Au di",
+      "▁B ent",
+      "▁Be nt",
+      "▁Ben t",
+      "弱 的",
+      "至 尊",
+      "▁gu inea",
+      "▁gui nea",
+      "▁BB Q",
+      "第二百零 七条",
+      "▁H eight",
+      "▁He ight",
+      "▁ Height",
+      "HEAD ER",
+      "挪 作他用",
+      "▁fact ories",
+      "▁factor ies",
+      "▁Rec ognition",
+      "▁Recogn ition",
+      "咽 喉",
+      "肥 料",
+      "去 括号",
+      "非 线性",
+      "im ize",
+      "imi ze",
+      "is chen",
+      "isc hen",
+      "isch en",
+      "ische n",
+      "叫 我",
+      "棱 长",
+      "我们 对",
+      "Hi nt",
+      "H int",
+      "一审 原告",
+      "decor ators",
+      "decorator s",
+      "▁其 它",
+      "▁ 其它",
+      "tr ight",
+      "tri ght",
+      "trig ht",
+      "t right",
+      "▁s enses",
+      "▁sens es",
+      "▁sense s",
+      "▁sen ses",
+      "co ordinate",
+      "coord inate",
+      "▁epid ermal",
+      "问 候",
+      "▁K O",
+      "▁ KO",
+      "ut ta",
+      "utt a",
+      "▁appet ite",
+      "▁ful filled",
+      "▁fulfill ed",
+      "▁fulfil led",
+      "回 避",
+      "到 案后",
+      "▁ 布",
+      "分 歧",
+      "火 力",
+      "点 燃",
+      "风 口",
+      "▁water proof",
+      "▁pro position",
+      "▁propos ition",
+      "▁prop osition",
+      "UC T",
+      "U CT",
+      "的最大 值",
+      "的 最大值",
+      "nD ear",
+      "nDe ar",
+      "n Dear",
+      "▁and rogen",
+      "not ifications",
+      "notification s",
+      "下 调",
+      "不 经",
+      "定 为",
+      "cr um",
+      "c rum",
+      "Pos ts",
+      "Post s",
+      "Po sts",
+      "}) ,",
+      "} ),",
+      "oy le",
+      "oyl e",
+      "o yle",
+      "▁W OR",
+      "▁WO R",
+      "▁ WOR",
+      "与 众",
+      "机 票",
+      "考 场",
+      "热水 器",
+      "目标 是",
+      "as ar",
+      "asa r",
+      "To Many",
+      "enti eth",
+      "▁c ipher",
+      "▁ci pher",
+      "▁ cipher",
+      "▁f ooter",
+      "▁foot er",
+      "▁fo oter",
+      "▁foo ter",
+      "▁ footer",
+      "▁period ontal",
+      "▁periodont al",
+      "Q P",
+      "九 章",
+      "睡 着",
+      "CM D",
+      "C MD",
+      "贸 市场",
+      "calc ulate",
+      "calcul ate",
+      "c alculate",
+      "针 织",
+      "ym pt",
+      "从重 处罚",
+      "leet code",
+      "四 面",
+      "大 片",
+      "起 草",
+      "怎么 会",
+      "▁re try",
+      "▁r etry",
+      "▁ret ry",
+      "▁ retry",
+      "▁t read",
+      "▁tr ead",
+      "▁tre ad",
+      "▁fresh man",
+      "▁Ob viously",
+      "▁un ilateral",
+      "W L",
+      "低 下",
+      "的 要",
+      "延长 赛",
+      "误 解",
+      "agr ang",
+      "agra ng",
+      "▁evalu ates",
+      "▁evaluate s",
+      "▁eval uates",
+      "▁dis establish",
+      "新 高",
+      "se v",
+      "s ev",
+      "▁H ang",
+      "▁Ha ng",
+      "▁Han g",
+      "▁su ites",
+      "▁suit es",
+      "▁suite s",
+      "政协 委员",
+      "本院 受理",
+      "▁b ishops",
+      "▁bi shops",
+      "▁bishop s",
+      "▁digest ive",
+      "苗 族",
+      "Pr im",
+      "P rim",
+      "co al",
+      "ic os",
+      "ico s",
+      "i cos",
+      "l ibraries",
+      "奥 特",
+      "ak k",
+      "a kk",
+      "于一体 的",
+      "于 一体的",
+      "以及 其他",
+      "以 及其他",
+      "▁l upus",
+      "▁lu pus",
+      "▁par ade",
+      "▁para de",
+      "▁parad e",
+      "该 卡",
+      "GE N",
+      "G EN",
+      "住址 同上",
+      "GRO UP",
+      "is ive",
+      "isi ve",
+      "▁Wor cester",
+      "读 物",
+      "▁Is le",
+      "庭 院",
+      "ne ur",
+      "n eur",
+      "追踪 报道",
+      "▁G ear",
+      "▁Ge ar",
+      "德 尔",
+      "田 间",
+      "网 吧",
+      "Ke l",
+      "K el",
+      "很不 错",
+      "很 不错",
+      "ar tan",
+      "art an",
+      "arta n",
+      "ighbor s",
+      "ighb ors",
+      "nCRE ATE",
+      "n CREATE",
+      "▁W IN",
+      "▁WI N",
+      "▁ WIN",
+      "Cent ral",
+      "C entral",
+      "▁Christian ity",
+      "搞 定",
+      "在他 的",
+      "在 他的",
+      "CK ET",
+      "es ses",
+      "ess es",
+      "esse s",
+      "二〇一六年 三月",
+      "海 区",
+      "▁r idge",
+      "▁rid ge",
+      "▁ ridge",
+      "exper iment",
+      "nU S",
+      "n US",
+      "▁F S",
+      "▁ FS",
+      "▁x i",
+      "▁ xi",
+      "▁In come",
+      "▁Inc ome",
+      "▁Reg ulation",
+      "▁human itarian",
+      "筷 子",
+      "认定 为",
+      "认 定为",
+      "an ck",
+      "anc k",
+      "比较 方便",
+      "国 强",
+      "星 光",
+      "极 坐标",
+      "▁Off er",
+      "▁Of fer",
+      "▁ Offer",
+      "认定事实 如下",
+      "讨论 版",
+      "vis ited",
+      "visit ed",
+      "▁Broad casting",
+      "测 验",
+      "驻 马",
+      "ien ne",
+      "i enne",
+      "▁ip sum",
+      "▁ips um",
+      "▁privile ged",
+      "▁privilege d",
+      "▁enthusi astic",
+      "▁enthusiast ic",
+      "▁reserv ations",
+      "▁reservation s",
+      "th el",
+      "the l",
+      "t hel",
+      "有限公司 负担",
+      "Screen shot",
+      "Screens hot",
+      "核 对",
+      "请 勿",
+      "以下 是",
+      "莆田 市",
+      "这 方面",
+      "nU SE",
+      "nUS E",
+      "n USE",
+      "nv al",
+      "n val",
+      "mar shal",
+      "一 两",
+      "▁F orms",
+      "▁For ms",
+      "▁Form s",
+      "▁ Forms",
+      "▁hom otopy",
+      "代表人 的人数提出副本",
+      "yu n",
+      "y un",
+      "安阳 市",
+      "安 阳市",
+      "ter a",
+      "te ra",
+      "t era",
+      "▁brand ed",
+      "▁ branded",
+      "Pres ident",
+      "廊 坊",
+      "▁k Da",
+      "▁t ee",
+      "▁te e",
+      "▁ tee",
+      "在 高",
+      "enn ium",
+      "▁el der",
+      "▁eld er",
+      "Serial ized",
+      "Serialize d",
+      "之 初",
+      "号 征收社会抚养费",
+      "▁educ ator",
+      "伤 亡",
+      "支 柱",
+      "est ones",
+      "estone s",
+      "esto nes",
+      "eston es",
+      "e stones",
+      "▁under taking",
+      "▁undert aking",
+      "▁pre treatment",
+      "▁pret reatment",
+      "野 外",
+      "IS P",
+      "I SP",
+      "nM ac",
+      "nMa c",
+      "n Mac",
+      "▁aqu atic",
+      "邮 票",
+      "全 网",
+      "历 年",
+      "战 的",
+      "烘 干",
+      "医学 会",
+      "医 学会",
+      "受害 者",
+      "▁ 营业时间",
+      "扭 曲",
+      "找 回",
+      "潜 能",
+      "▁l ace",
+      "▁la ce",
+      "▁lac e",
+      "▁人口 数量",
+      "▁ment ors",
+      "▁mentor s",
+      "▁Tanz ania",
+      "▁mut ually",
+      "▁mutual ly",
+      "多样 化",
+      "vo ice",
+      "v oice",
+      "▁C lif",
+      "▁Cl if",
+      "▁br ave",
+      "▁bra ve",
+      "偿还 借款本金",
+      "偿还借款 本金",
+      "▁bi ased",
+      "▁bias ed",
+      "▁ biased",
+      "▁parad ox",
+      "Im plementation",
+      "▁merc ury",
+      "ighbor hood",
+      "▁Comput ational",
+      "吃 吃",
+      "炒 作",
+      "西 侧",
+      "▁b ent",
+      "▁be nt",
+      "▁ben t",
+      "▁o val",
+      "▁ov al",
+      "▁ oval",
+      "店 里",
+      "绿 豆",
+      "IC H",
+      "I CH",
+      "也 开始",
+      "作品 的",
+      "作 品的",
+      "pop up",
+      "▁F act",
+      "▁Fac t",
+      "▁Fa ct",
+      "▁ Fact",
+      "▁Hol mes",
+      "▁Draw ing",
+      "▁ Drawing",
+      "▁Test Case",
+      "▁ TestCase",
+      "▁immigr ant",
+      "变 价",
+      "▁[ <",
+      "▁ [<",
+      "着名 的",
+      "着 名的",
+      "ln g",
+      "l ng",
+      "工商 户",
+      "工 商户",
+      "世界上 最",
+      "申请 仲裁",
+      "exper imental",
+      "experiment al",
+      "the mes",
+      "theme s",
+      "them es",
+      "t hemes",
+      "ch ief",
+      "chi ef",
+      "co urses",
+      "course s",
+      "c ourses",
+      "▁deterior ation",
+      "学生 们",
+      "co ck",
+      "coc k",
+      "c ock",
+      "nPro of",
+      "n Proof",
+      "▁E clipse",
+      "plant ation",
+      "▁In frastructure",
+      "▁ Infrastructure",
+      "卡 车",
+      "向 南",
+      "tb l",
+      "t bl",
+      "下面 是",
+      "▁T rek",
+      "▁Tr ek",
+      "▁Tre k",
+      "▁s word",
+      "▁sw ord",
+      "▁ret ard",
+      "▁ster eo",
+      "▁stere o",
+      "▁resign ation",
+      "其 一",
+      "有 不",
+      "/\" >",
+      "/ \">",
+      "方程 的解",
+      "方程的 解",
+      "get ting",
+      "g etting",
+      "精 益",
+      "▁C i",
+      "▁ Ci",
+      "▁M K",
+      "▁ MK",
+      "▁` @",
+      "履行 上述",
+      "▁c ada",
+      "▁ca da",
+      "▁cad a",
+      "Warning s",
+      "War nings",
+      "W arnings",
+      "包 子",
+      "小 伙",
+      "▁\" \")",
+      "▁\"\" )",
+      "科学 出版社",
+      "▁ 牛",
+      "这 片",
+      "▁f ights",
+      "▁fight s",
+      "▁pec uliar",
+      "▁under est",
+      "放 电",
+      "an j",
+      "a nj",
+      "入 围",
+      "放 开",
+      "万人 次",
+      "万 人次",
+      "▁synd romes",
+      "▁syndrome s",
+      "饮 品",
+      "▁H DL",
+      "▁HD L",
+      "▁ HDL",
+      "▁comput ations",
+      "▁computation s",
+      "秦皇 岛",
+      "而且 在",
+      "而 且在",
+      "ED IT",
+      "E DIT",
+      "]) {\\",
+      "] ){\\",
+      "ad ia",
+      "adi a",
+      "避 开",
+      "的 成本",
+      "还是 挺",
+      "er ville",
+      "erv ille",
+      "▁proud ly",
+      "▁Prov incial",
+      "替 补",
+      "▁ 最后",
+      "副 县长",
+      "▁B oss",
+      "▁Bo ss",
+      "▁Bos s",
+      "▁ Boss",
+      "▁ 神",
+      "▁non zero",
+      "和计划生育 委员会",
+      "up uncture",
+      "▁incorrect ly",
+      "辆 车",
+      "因为 是",
+      "▁G mbH",
+      "nE uropean",
+      "n European",
+      "▁ign oring",
+      "▁ignor ing",
+      "下 手",
+      "uck ed",
+      "USER NAME",
+      "е с",
+      "ta da",
+      "t ada",
+      "ti ny",
+      "tin y",
+      "t iny",
+      "▁N om",
+      "▁No m",
+      "▁ Nom",
+      "还有 什么",
+      "还 有什么",
+      "om eters",
+      "ome ters",
+      "omet ers",
+      "ometer s",
+      "元 为",
+      "西 祠",
+      "角 膜",
+      "ho n",
+      "h on",
+      "em ory",
+      "emor y",
+      "emo ry",
+      "二〇一七年 五月",
+      "年 会",
+      "纠 缠",
+      "呼吸 道",
+      "Mon ey",
+      "Mo ney",
+      "M oney",
+      "▁tub ular",
+      "▁co herence",
+      "的 政治",
+      "▁M aths",
+      "▁Mat hs",
+      "▁Math s",
+      "▁Ma ths",
+      "▁D ining",
+      "▁Di ning",
+      "▁Din ing",
+      "\\\\ \\\\\\\\\\\\",
+      "\\\\\\\\ \\\\\\\\",
+      "\\\\\\ \\\\\\\\\\",
+      "\\\\\\\\\\ \\\\\\",
+      "\\\\\\\\\\\\ \\\\",
+      "▁nurs ery",
+      "▁nur sery",
+      "▁nurse ry",
+      "▁allev iate",
+      "一 女",
+      "OR G",
+      "O RG",
+      "gs on",
+      "g son",
+      "情况 进行",
+      "▁sp ill",
+      "▁program mer",
+      "▁programme r",
+      "———————— ————————",
+      "中 之",
+      "抗 议",
+      "an di",
+      "and i",
+      "a ndi",
+      "camp aign",
+      "l j",
+      "磁 盘",
+      "▁in hal",
+      "▁inh al",
+      "▁hepar in",
+      "后 人",
+      "巴 拉",
+      "帮助 下",
+      "Av atar",
+      "▁her pes",
+      "▁S AT",
+      "▁SA T",
+      "▁ SAT",
+      "短 时间内",
+      "脚 下",
+      "O cc",
+      "双方 在",
+      "业主 收取",
+      "▁l iable",
+      "▁li able",
+      "▁lia ble",
+      "▁ liable",
+      "▁exped ition",
+      "侮 辱",
+      "MA IL",
+      "▁c pu",
+      "▁cp u",
+      "▁ cpu",
+      "▁l ou",
+      "▁lo u",
+      "非常高兴 能",
+      "▁S yntax",
+      "▁Syn tax",
+      "▁Synt ax",
+      "▁ Syntax",
+      "▁lon gevity",
+      "▁ Ö",
+      "事 的",
+      "和 人",
+      "盐城 市",
+      "盐 城市",
+      "nh as",
+      "n has",
+      "nP ower",
+      "nPo wer",
+      "n Power",
+      "▁Q atar",
+      "▁light ning",
+      "Af f",
+      "A ff",
+      "▁Anc ient",
+      "▁Chen nai",
+      "▁Vit amin",
+      "▁work load",
+      "FF ER",
+      "号案件 的执行",
+      "up dates",
+      "update s",
+      "▁Health y",
+      "ER ENCES",
+      "ERENCE S",
+      "▁proc urement",
+      "但 你",
+      "鉴 赏",
+      "两部 分",
+      "两 部分",
+      "▁numer ically",
+      "▁numerical ly",
+      "▁numeric ally",
+      "▁spec ulation",
+      "刻 画",
+      "陆 游",
+      "仅仅 是",
+      "op encv",
+      "open cv",
+      "▁For mer",
+      "▁Form er",
+      "▁Min imum",
+      "▁ Minimum",
+      "▁post gres",
+      "▁ postgres",
+      "向被执行人 发出执行通知书",
+      "实 木",
+      "食 管",
+      "▁s unset",
+      "▁su nset",
+      "▁sun set",
+      "▁文 学",
+      "▁ 文学",
+      "有时 间",
+      "有 时间",
+      "▁B av",
+      "▁Ba v",
+      "en burg",
+      "▁P atch",
+      "▁Pat ch",
+      "▁ Patch",
+      "dec imal",
+      "pat ibility",
+      "耐 药",
+      "Display Name",
+      "C y",
+      "各 主要",
+      "if ug",
+      "nk ind",
+      "n kind",
+      "▁P rote",
+      "▁Pro te",
+      "▁Pr ote",
+      "▁Prot e",
+      "represent ed",
+      "黑 的",
+      "门户 网站",
+      "▁n aval",
+      "▁nav al",
+      "▁na val",
+      "第十三 条第二款",
+      "第十三条 第二款",
+      "不断 累",
+      "▁Ap ollo",
+      "▁A SF",
+      "▁AS F",
+      "▁ ASF",
+      "▁turb ine",
+      "▁techn ician",
+      "住 建",
+      "招 待",
+      "退 赔",
+      "st ones",
+      "ston es",
+      "stone s",
+      "sto nes",
+      "▁C rist",
+      "▁Cr ist",
+      "▁K ubernetes",
+      "ol ith",
+      "olit h",
+      "oli th",
+      "un ts",
+      "unt s",
+      "ig hed",
+      "igh ed",
+      "ac iones",
+      "acion es",
+      "aci ones",
+      "acio nes",
+      "▁exceed ing",
+      "上 车",
+      "我的 经纪人",
+      "nTr ump",
+      "n Trump",
+      "▁M iles",
+      "▁Mil es",
+      "▁Mi les",
+      "▁Mile s",
+      "config s",
+      "conf igs",
+      "▁lingu istic",
+      "US ED",
+      "USE D",
+      "U SED",
+      "▁her bal",
+      "▁herb al",
+      "▁Ott oman",
+      "▁Otto man",
+      "produ cing",
+      "prod ucing",
+      "p roducing",
+      "隔 音",
+      "IM O",
+      "I MO",
+      "日 美股",
+      "pa irs",
+      "pair s",
+      "p airs",
+      "▁R aid",
+      "▁Ra id",
+      "我非常 感激",
+      "▁f erry",
+      "▁fer ry",
+      "▁ferr y",
+      "▁rupt ure",
+      "工业 化",
+      "应用 的",
+      "应 用的",
+      "chem ical",
+      "c hemical",
+      "棵 树",
+      "而 上",
+      "ll d",
+      "l ld",
+      "与 本案",
+      "ad ren",
+      "adr en",
+      "▁P erc",
+      "▁Per c",
+      "▁Pe rc",
+      "▁im pose",
+      "▁imp ose",
+      "define Property",
+      "对我 国",
+      "对 我国",
+      "man t",
+      "ma nt",
+      "m ant",
+      "这么 多年",
+      "这么多 年",
+      "as per",
+      "asp er",
+      "▁B elf",
+      "▁Be lf",
+      "▁Bel f",
+      "itc hens",
+      "itch ens",
+      "itchen s",
+      "▁M oment",
+      "▁Mo ment",
+      "▁Mom ent",
+      "▁C rit",
+      "▁Cr it",
+      "▁S ites",
+      "▁Site s",
+      "▁Si tes",
+      "▁Sit es",
+      "▁f rank",
+      "▁fr ank",
+      "▁fra nk",
+      "和 知识",
+      "状态 下",
+      "心 底",
+      "▁A E",
+      "▁ AE",
+      "违反 了",
+      "我们将 予以",
+      "▁L ynch",
+      "▁Ly nch",
+      "▁Lyn ch",
+      "T N",
+      "▁Jo an",
+      "▁av ail",
+      "▁Init ially",
+      "▁Initial ly",
+      "hi nt",
+      "hin t",
+      "h int",
+      "你 家",
+      "更 要",
+      "的 犯罪",
+      "Pr ep",
+      "Pre p",
+      "P rep",
+      "▁PT SD",
+      "二〇一七年 三月",
+      "二〇一七年三 月",
+      "nFootball ers",
+      "一 向",
+      "妨 碍",
+      "Cal c",
+      "Ca lc",
+      "C alc",
+      "▁cr uel",
+      "▁cru el",
+      "止 痛",
+      "的 案件",
+      "qu ez",
+      "que z",
+      "list dir",
+      "dim ension",
+      "d imension",
+      "调 制",
+      "igue z",
+      "igu ez",
+      "nRe lease",
+      "n Release",
+      "引发 的",
+      "引 发的",
+      "赢得 胜利的",
+      "▁Econ omy",
+      "▁Ec onomy",
+      "威 力",
+      "枝 花",
+      "是由 于",
+      "是 由于",
+      "的 控制",
+      "多边形 的",
+      "多 边形的",
+      "▁loc alities",
+      "▁local ities",
+      "▁neuro psych",
+      "浓 厚",
+      "读 了",
+      "研究 室",
+      "这部 分",
+      "这 部分",
+      "mo tion",
+      "mot ion",
+      "m otion",
+      "▁My anmar",
+      "▁respon ds",
+      "▁respond s",
+      "▁resp onds",
+      "PO R",
+      "P OR",
+      "防火 墙",
+      "Inf later",
+      "▁ 广州",
+      "得更 全面",
+      "Blue print",
+      "Oper ations",
+      "Operation s",
+      "酰 胺",
+      "Ba nd",
+      "Ban d",
+      "B and",
+      "人民 调解协议",
+      "人民调解 协议",
+      "▁Coal ition",
+      "▁waste water",
+      "▁wast ewater",
+      "责任 心",
+      "重组 案",
+      "The ta",
+      "Th eta",
+      "▁St efan",
+      "▁Ste fan",
+      "▁in flux",
+      "▁inf lux",
+      "▁influ x",
+      "▁infl ux",
+      "▁ret iring",
+      "▁Strateg ic",
+      "产 科",
+      "偶 函数",
+      "就 直接",
+      "缺席 审理",
+      "演唱 的歌曲",
+      "信 道",
+      "向 西",
+      "▁F ew",
+      "▁Fe w",
+      "明确的 被告",
+      "Lo ndon",
+      "L ondon",
+      "全 都",
+      "锅 里",
+      "专家 坐堂",
+      "南京 还有",
+      "an tage",
+      "ant age",
+      "anta ge",
+      "▁compens ate",
+      "▁P U",
+      "▁ PU",
+      "nB uy",
+      "n Buy",
+      "▁L abs",
+      "▁La bs",
+      "▁Lab s",
+      "所有的 位于",
+      "▁c ubes",
+      "▁cub es",
+      "▁cube s",
+      "▁cu bes",
+      "广东省 深圳市",
+      "prov ided",
+      "provide d",
+      "} '",
+      "先 说",
+      "吻 合",
+      "怀特 说",
+      "ip eg",
+      "ipe g",
+      "i peg",
+      "支出 色的",
+      "支 出色的",
+      "▁Sh arp",
+      "▁Sha rp",
+      "▁Shar p",
+      "▁ Sharp",
+      "nIm port",
+      "n Import",
+      "会 签下",
+      "nn ow",
+      "nno w",
+      "n now",
+      "▁B rief",
+      "▁Br ief",
+      "▁Bri ef",
+      "▁Hur ricane",
+      "五 行",
+      "嘲 讽",
+      "农业 农村",
+      "▁congress ional",
+      "图 表",
+      "巴 基",
+      "生产 劳动",
+      "十三 条规定",
+      "十三条 规定",
+      "▁mid st",
+      "▁mi dst",
+      "[ #",
+      "进 军",
+      "了一 批",
+      "了 一批",
+      "你 以为",
+      "Sp ot",
+      "S pot",
+      "yp ress",
+      "y press",
+      "号 行政",
+      "公摊 水费",
+      "延长赛 落败",
+      "证监会 否决",
+      "反映给 相关部门",
+      "把 她",
+      "白 玉",
+      "加深 天马",
+      "融券 筹码",
+      "实录 加西亚",
+      "重组案 再遭",
+      "▁R atio",
+      "▁Rat io",
+      "▁ Ratio",
+      "不断累 加深天马",
+      "日美股 专家坐堂",
+      "实录加西亚 延长赛落败",
+      "融券筹码 不断累加深天马",
+      "融券筹码不断累加深天马 重组案再遭",
+      "羊 毛",
+      "学术 论文",
+      "求 和",
+      "ln x",
+      "l nx",
+      "▁英 语",
+      "▁ 英语",
+      "in dy",
+      "ind y",
+      "与被告 王",
+      "pl atin",
+      "plat in",
+      "rs s",
+      "r ss",
+      "▁N Z",
+      "▁ NZ",
+      "不了 解",
+      "不 了解",
+      "▁D od",
+      "▁Do d",
+      "▁C lin",
+      "▁Cl in",
+      "▁D ale",
+      "▁Da le",
+      "▁Dal e",
+      "西祠 讨论版",
+      "支出色的 球队",
+      "▁Br east",
+      "▁Bre ast",
+      "我的经纪人 告诉我",
+      "万 能",
+      "推 介",
+      "特此 通知",
+      "签约 火箭",
+      "▁f ier",
+      "▁fi er",
+      "将自己 磨练",
+      "有几 支球队",
+      "bench mark",
+      "▁Cor rection",
+      "▁Correct ion",
+      "第一百五十四条 裁定适用于下列范围",
+      "il k",
+      "晚饭 团",
+      "月 获得",
+      "会签下 我",
+      "却被 物管",
+      "请求 判令",
+      "你的遭遇 吧",
+      "却依然 在向",
+      "扬子 晚饭团",
+      "明明 废弃了",
+      "一直在 关注我",
+      "却被物管 私吞",
+      "并将 你的遭遇",
+      "我坚信 有球队",
+      "我知道 有球队",
+      "所以我很 刻苦",
+      "而且 这个城市",
+      "怀特 非常高兴能",
+      "我现在 来到了一",
+      "说出 你的遭遇吧",
+      "赢得胜利的 球队",
+      "业主收取 公摊水费",
+      "我非常感激 这一切",
+      "租金 却被物管私吞",
+      "请登录 西祠讨论版",
+      "又有 多少小区的水景",
+      "将自己磨练 得更全面",
+      "我们将予以 追踪报道",
+      "我知道有球队 注意我",
+      "而且这个城市 也很好",
+      "我坚信有球队 会签下我",
+      "一个不断 赢得胜利的球队",
+      "南京还有 多少小区的水景",
+      "怀特非常高兴能 签约火箭",
+      "有几支球队 一直在关注我",
+      "却依然在向 业主收取公摊水费",
+      "又有多少小区的水景 挪作他用",
+      "并将你的遭遇 反映给相关部门",
+      "我现在来到了一 支出色的球队",
+      "南京还有多少小区的水景 明明废弃了",
+      "t G",
+      "会 谈",
+      "血 流",
+      "▁ 最大",
+      "▁T hing",
+      "▁Th ing",
+      "▁Thin g",
+      "▁mac ros",
+      "▁macro s",
+      "▁ macros",
+      "▁c ottage",
+      "▁cot tage",
+      "▁Buddh ist",
+      "▁Budd hist",
+      "▁Ch ocolate",
+      "b z",
+      "九 十",
+      "的 粉丝",
+      "nE qu",
+      "n Equ",
+      "▁b elly",
+      "▁bel ly",
+      "▁bell y",
+      "慎 重",
+      "氧 气",
+      "而 过",
+      "▁F ra",
+      "▁Fr a",
+      "ess ive",
+      "▁Rock y",
+      "▁Roc ky",
+      "cent ric",
+      "c entric",
+      "穿 刺",
+      "Li f",
+      "L if",
+      "sent ence",
+      "房 的",
+      "er mine",
+      "erm ine",
+      "nR ugby",
+      "▁Some one",
+      "▁hydroph obic",
+      "收 割",
+      "艰 辛",
+      "nJ im",
+      "n Jim",
+      "▁登场 作品",
+      "▁sc rews",
+      "▁scr ews",
+      "▁screw s",
+      "关节 炎",
+      "▁res istor",
+      "▁resist or",
+      "S a",
+      "收 盘",
+      "做出 的",
+      "做 出的",
+      "pi eces",
+      "piece s",
+      "pie ces",
+      "▁Res cue",
+      "▁s el",
+      "▁se l",
+      "▁ sel",
+      "▁P ag",
+      "▁Pa g",
+      "▁ Pag",
+      "C atalog",
+      "▁anal ogs",
+      "▁analog s",
+      "▁ana logs",
+      "是 人",
+      "不明 白",
+      "不 明白",
+      "也是 很",
+      "也 是很",
+      "激烈 的",
+      "激 烈的",
+      "搅拌 均匀",
+      "▁tool bar",
+      "▁ toolbar",
+      "▁Mah arashtra",
+      "No r",
+      "N or",
+      "ir ical",
+      "iri cal",
+      "i rical",
+      "im plementation",
+      "imple mentation",
+      "即可 求得",
+      "▁S ync",
+      "▁Sy nc",
+      "▁Syn c",
+      "▁ Sync",
+      "向 我",
+      "年 少",
+      "雪 花",
+      "ce ipt",
+      "к у",
+      "▁E cho",
+      "▁Ec ho",
+      "▁T YPE",
+      "▁ TYPE",
+      "▁s oda",
+      "▁so da",
+      "▁advert isement",
+      "▁advertis ement",
+      "▁advertise ment",
+      "pl ural",
+      "▁review ers",
+      "▁reviewer s",
+      "马 某",
+      "er es",
+      "ere s",
+      "e res",
+      "▁R oh",
+      "▁Ro h",
+      "▁pe el",
+      "▁inv aluable",
+      "▁pregn ancies",
+      "▁C ities",
+      "▁Cit ies",
+      "▁Ci ties",
+      "共 振",
+      "芜湖 市",
+      "设计 了",
+      "住 山东省",
+      "ball s",
+      "bal ls",
+      "b alls",
+      "nB ook",
+      "nBo ok",
+      "n Book",
+      "lan tic",
+      "l antic",
+      "▁valid ator",
+      "▁ validator",
+      "▁conv iction",
+      "T K",
+      "断 了",
+      "济 宁",
+      "联 手",
+      "请 用",
+      "IN ESS",
+      "INE SS",
+      "I NESS",
+      "nHome work",
+      "nHom ework",
+      "▁de leting",
+      "▁del eting",
+      "▁delet ing",
+      "分布 在",
+      "Spec ific",
+      "S pecific",
+      "▁Sym posium",
+      "if ax",
+      "ifa x",
+      "i fax",
+      "ind ividual",
+      "▁goal keeper",
+      "▁Legisl ature",
+      "整 顿",
+      "各自 的",
+      "iv ary",
+      "iva ry",
+      "ivar y",
+      "der ive",
+      "de rive",
+      "▁Abd ul",
+      "夫妻感情 破裂",
+      "原 型",
+      "CM S",
+      "C MS",
+      "▁S hin",
+      "▁Sh in",
+      "▁Shi n",
+      "▁kind ly",
+      "▁ion ization",
+      "该 局",
+      "▁dr ummer",
+      "▁drum mer",
+      "运 势",
+      "各 单位",
+      "▁l anes",
+      "▁la nes",
+      "▁lane s",
+      "▁lan es",
+      "搞 好",
+      "构 件",
+      "re views",
+      "review s",
+      "▁Pre pare",
+      "▁Prep are",
+      "▁ped estrian",
+      "▁pedest rian",
+      "祝 愿",
+      "过 期",
+      "ch ai",
+      "cha i",
+      "c hai",
+      "st ry",
+      "str y",
+      "s try",
+      "▁s ights",
+      "▁sight s",
+      "▁separ ating",
+      "率 达",
+      "父亲 的",
+      "市 值",
+      "更 适合",
+      "电动 机",
+      "电 动机",
+      "神秘 的",
+      "\\\\ ;\\\\",
+      "\\\\; \\\\",
+      "/ '",
+      "明 知",
+      "意识 形态",
+      "科技 股份有限公司",
+      "V GA",
+      "责任 的",
+      ">( \\",
+      "> (\\",
+      "专业 课",
+      "Access or",
+      "血 症",
+      "ie c",
+      "i ec",
+      "~~ ~",
+      "~ ~~",
+      "脂肪 酸",
+      "▁Na du",
+      "▁Nad u",
+      "ich let",
+      "▁determin istic",
+      "AN Y",
+      "A NY",
+      "ul ance",
+      "ula nce",
+      "▁Ch ess",
+      "▁Che ss",
+      "▁Ches s",
+      "▁ad mits",
+      "▁admit s",
+      "▁adm its",
+      "▁rac ism",
+      "等证据 予以证实",
+      "子 孙",
+      "▁f d",
+      "▁ fd",
+      "生前 为",
+      "▁re bounds",
+      "▁reb ounds",
+      "▁rebound s",
+      "▁post natal",
+      "▁P t",
+      "▁ Pt",
+      "心脏 病",
+      "tN ame",
+      "t Name",
+      "▁p eek",
+      "▁pe ek",
+      "▁ peek",
+      "nAdd itionally",
+      "nAdditional ly",
+      "执行 至",
+      "tr ad",
+      "tra d",
+      "t rad",
+      "一个 新的",
+      "学习 教育",
+      "▁sh owers",
+      "▁show ers",
+      "▁shower s",
+      "lim itations",
+      "limit ations",
+      "深 远",
+      "▁— —",
+      "▁ ——",
+      "▁n ont",
+      "▁no nt",
+      "▁non t",
+      "▁Pay Pal",
+      "▁p umping",
+      "▁pump ing",
+      "▁commod ity",
+      "▁repos itories",
+      "▁ repositories",
+      "▁transfer ring",
+      "▁transf erring",
+      "IN DEX",
+      "IND EX",
+      "nS ystem",
+      "n System",
+      "▁N ortheast",
+      "▁North east",
+      "略 有",
+      "创造 性",
+      "半个 月",
+      "半 个月",
+      "▁H err",
+      "▁He rr",
+      "▁Her r",
+      "▁diarr hea",
+      "和他 的",
+      "和 他的",
+      "▁re he",
+      "DATA BASE",
+      "D ATABASE",
+      "聘 请",
+      "J en",
+      "不知 不觉",
+      "▁f rog",
+      "▁fr og",
+      "▁fro g",
+      "▁ 中共党员",
+      "▁lim bs",
+      "▁limb s",
+      "主 线",
+      "nK now",
+      "nKn ow",
+      "n Know",
+      "制成 的",
+      "制 成的",
+      "Ver ify",
+      "rais al",
+      "日 开始",
+      "营养 成分",
+      "/ ,",
+      "今 生",
+      "是 怎样",
+      "▁Fr anç",
+      "▁weakness es",
+      "关系 到",
+      "ns age",
+      "▁build ers",
+      "▁builder s",
+      "▁ builders",
+      "减 弱",
+      "Ver s",
+      "V ers",
+      "con e",
+      "co ne",
+      "c one",
+      "e uclid",
+      "nS ocial",
+      "n Social",
+      "nPar ameters",
+      "n Parameters",
+      "场 馆",
+      "Th ursday",
+      "J ul",
+      "人 意",
+      "设 在",
+      "be ck",
+      "bec k",
+      "b eck",
+      "多 点",
+      "nF in",
+      "n Fin",
+      "▁O ral",
+      "▁Or al",
+      "▁List en",
+      "▁Li sten",
+      "▁Lis ten",
+      "▁ Listen",
+      "对被执行人 的财产",
+      "对 被执行人的财产",
+      "▁metabol ite",
+      "少 于",
+      "提到 的",
+      "提 到的",
+      "符合 下列",
+      "▁war mer",
+      "▁warm er",
+      "▁combin atorial",
+      "合 影",
+      "中止 或者",
+      "▁e lem",
+      "▁el em",
+      "▁ele m",
+      "▁ elem",
+      "▁consolid ation",
+      "▁C ec",
+      "▁Ce c",
+      "▁V II",
+      "▁VI I",
+      "er obic",
+      "ero bic",
+      "▁ex cluding",
+      "乘 车",
+      "这 三",
+      "▁a bd",
+      "▁ab d",
+      "什么 东西",
+      "net works",
+      "network s",
+      "▁inc lined",
+      "▁incl ined",
+      "▁coll apsed",
+      "▁collapse d",
+      "phys ics",
+      "▁ 尺寸",
+      "as type",
+      "ast ype",
+      "asty pe",
+      "re ceive",
+      "▁begin ner",
+      "▁beg inner",
+      "维持 原裁定",
+      "tt le",
+      "t tle",
+      "yn th",
+      "ynt h",
+      "y nth",
+      "▁A sync",
+      "▁As ync",
+      "▁ Async",
+      "direct ories",
+      "director ies",
+      "不 符",
+      "宽 容",
+      "ar ah",
+      "ara h",
+      "a rah",
+      "un ity",
+      "unit y",
+      "uni ty",
+      "an ions",
+      "ani ons",
+      "anion s",
+      "敬 老",
+      "压缩 机",
+      "nc heck",
+      "nch eck",
+      "nche ck",
+      "n check",
+      "▁行政区 类别",
+      "os exual",
+      "osex ual",
+      "o sexual",
+      "stud ents",
+      "student s",
+      "▁arg uing",
+      "▁argu ing",
+      "建 的",
+      "ak an",
+      "aka n",
+      "a kan",
+      "▁mod ality",
+      "▁modal ity",
+      "▁char itable",
+      "纵 向",
+      "整数 的",
+      "整 数的",
+      "治疗 的",
+      "br own",
+      "bro wn",
+      "b rown",
+      "▁c ouns",
+      "▁co uns",
+      "▁cou ns",
+      "▁prob abilistic",
+      "就 越",
+      "▁M U",
+      "▁ MU",
+      "因为 我们",
+      "因为我 们",
+      "因 为我们",
+      "▁T race",
+      "▁Tr ace",
+      "▁Tra ce",
+      "▁ Trace",
+      "▁sl eek",
+      "▁sle ek",
+      "▁car ved",
+      "▁refer endum",
+      "应 以",
+      "网 球",
+      "道 教",
+      "骚 扰",
+      "ME R",
+      "M ER",
+      "nDe ath",
+      "n Death",
+      "illi ams",
+      "illiam s",
+      "no l",
+      "n ol",
+      "▁R ED",
+      "▁RE D",
+      "▁ RED",
+      "▁g eo",
+      "▁ge o",
+      "▁ geo",
+      "▁aggreg ates",
+      "▁aggregate s",
+      "峰 值",
+      "▁V iv",
+      "▁Vi v",
+      "▁hel met",
+      "▁helm et",
+      "▁r esides",
+      "▁res ides",
+      "▁resid es",
+      "▁reside s",
+      "四边形 的",
+      "四 边形的",
+      "▁b ake",
+      "▁ba ke",
+      "▁bak e",
+      "▁gest ational",
+      "▁gestation al",
+      "花 瓣",
+      "评 级",
+      "SC I",
+      "S CI",
+      "▁G I",
+      "▁ GI",
+      "本质 上",
+      "罪犯 刘",
+      "St ates",
+      "State s",
+      "Stat es",
+      "O X",
+      "▁notice able",
+      "关键 字",
+      "ot ti",
+      "ott i",
+      "o tti",
+      "▁sc enery",
+      "▁scen ery",
+      "▁scene ry",
+      "con verter",
+      "convert er",
+      "▁separ ator",
+      "▁ separator",
+      "山 下",
+      "占用 的",
+      "占 用的",
+      "▁S ql",
+      "▁ Sql",
+      "二〇一七年 六月",
+      "二〇一七年六 月",
+      "▁Re verse",
+      "▁Rever se",
+      "▁ Reverse",
+      "▁transl ates",
+      "▁translate s",
+      "赞 助",
+      "系数 的",
+      "系 数的",
+      "▁est á",
+      "提交 的证据",
+      "提交的 证据",
+      "Cap ture",
+      "Capt ure",
+      "古老 的",
+      "古 老的",
+      "很好 地",
+      "an ey",
+      "ane y",
+      "a ney",
+      "stat istics",
+      "▁B od",
+      "▁Bo d",
+      "不只 是",
+      "不 只是",
+      "H N",
+      "gr owing",
+      "gro wing",
+      "grow ing",
+      "g rowing",
+      "▁murder ed",
+      "▁Louis ville",
+      "▁a fore",
+      "▁af ore",
+      "os us",
+      "o sus",
+      "▁V ern",
+      "▁Ver n",
+      "▁Ve rn",
+      "▁游戏 介绍",
+      "De code",
+      "Dec ode",
+      "D ecode",
+      "▁com ma",
+      "▁comm a",
+      "▁co mma",
+      "▁L S",
+      "▁ LS",
+      "▁F ul",
+      "▁Fu l",
+      "云 龙",
+      "风 波",
+      "教学 的",
+      "教 学的",
+      "但是 如果",
+      "▁avoid ance",
+      "▁ 松",
+      "沿 革",
+      "per f",
+      "pe rf",
+      "▁o vert",
+      "▁over t",
+      "▁ov ert",
+      "▁val ves",
+      "▁valve s",
+      "mong oose",
+      "mongo ose",
+      "m ongoose",
+      "ynt hetic",
+      "ynth etic",
+      "y nthetic",
+      "▁Squad ron",
+      "status Code",
+      "做 工",
+      "招 牌",
+      "▁D uncan",
+      "▁Dun can",
+      "▁Du ncan",
+      "erman ent",
+      "相 近",
+      "▁H L",
+      "▁ HL",
+      "ob ia",
+      "obi a",
+      "o bia",
+      "▁phot oc",
+      "▁photo c",
+      "曲 折",
+      "电 能",
+      "进 出",
+      "\"< <",
+      "\" <<",
+      "▁[ {",
+      "▁ [{",
+      "ment e",
+      "me nte",
+      "men te",
+      "m ente",
+      "是什么 意思",
+      "▁refr actory",
+      "写 着",
+      "模拟 器",
+      "sh irts",
+      "shirt s",
+      "回 车",
+      "民 宿",
+      "el ic",
+      "eli c",
+      "e lic",
+      "▁b ibli",
+      "▁bib li",
+      "▁L adies",
+      "▁La dies",
+      "▁Lad ies",
+      "▁P urchase",
+      "▁coll ateral",
+      "ou cester",
+      "ic ultural",
+      "icult ural",
+      "i cultural",
+      "i J",
+      "出 路",
+      "▁fem inine",
+      "▁femin ine",
+      "▁Consult ing",
+      "琉 璃",
+      "▁K lein",
+      "▁Kle in",
+      "▁Kl ein",
+      "unction al",
+      "unct ional",
+      "自 强",
+      "芹 菜",
+      "Pro b",
+      "Pr ob",
+      "P rob",
+      "▁a ls",
+      "▁al s",
+      "▁ als",
+      "nP rot",
+      "nPro t",
+      "nPr ot",
+      "n Prot",
+      "G erman",
+      "▁Com ponents",
+      "▁Component s",
+      "▁ Components",
+      "中 式",
+      "喷 雾",
+      "黄 石",
+      "ap o",
+      "a po",
+      "的问题 和",
+      "的 问题和",
+      "cou ld",
+      "c ould",
+      "ov ina",
+      "ovi na",
+      "玩 耍",
+      "▁p b",
+      "▁ pb",
+      "un al",
+      "una l",
+      "u nal",
+      "▁Gar cia",
+      "▁Garc ia",
+      "▁Le ather",
+      "▁re versal",
+      "▁revers al",
+      "▁rever sal",
+      "▁explor atory",
+      "▁O ber",
+      "▁Ob er",
+      "元 整",
+      "去 过",
+      "月 上",
+      "▁in herit",
+      "▁inher it",
+      "▁ inherit",
+      "争 霸",
+      "ol la",
+      "oll a",
+      "o lla",
+      "▁R ew",
+      "▁Re w",
+      "▁ Rew",
+      "▁Ethiop ia",
+      "评 测",
+      "大学 学历",
+      "数额 较大",
+      "▁st ove",
+      "▁sto ve",
+      "ath yroid",
+      "athy roid",
+      "▁render er",
+      "▁ renderer",
+      "▁( “",
+      "不论 是",
+      "▁b acon",
+      "▁ba con",
+      "▁bac on",
+      "公 因式",
+      "File Sync",
+      "▁a erobic",
+      "▁aer obic",
+      "▁kid neys",
+      "▁kidney s",
+      "离 散",
+      "▁R UN",
+      "▁ RUN",
+      "ME M",
+      "M EM",
+      "▁c f",
+      "▁ cf",
+      "文化 传媒",
+      "nF rank",
+      "nFr ank",
+      "n Frank",
+      "▁Re venue",
+      "▁Rev enue",
+      "▁dece ased",
+      "tR E",
+      "t RE",
+      "▁br och",
+      "▁bro ch",
+      "▁nav bar",
+      "▁ navbar",
+      "▁Os lo",
+      "▁commit ments",
+      "▁commitment s",
+      "淡 水",
+      "工作 时间",
+      "工作时 间",
+      "▁HV AC",
+      "▁Roman ized",
+      "▁synchron ization",
+      "相 较",
+      "c ategor",
+      "▁pic nic",
+      "exec utor",
+      "nDef inition",
+      "n Definition",
+      "▁deliver ies",
+      "一 子",
+      "第二 款",
+      "企 事业单位",
+      "建设 有限公司",
+      "▁D ashboard",
+      "▁Dash board",
+      "▁ Dashboard",
+      "▁exception ally",
+      "▁exceptional ly",
+      "$$ (",
+      "$ $(",
+      "ot ta",
+      "ott a",
+      "▁endorse d",
+      "▁endors ed",
+      "男 士",
+      "送 货",
+      "对人 体",
+      "对 人体",
+      "美联 储",
+      "▁bo iling",
+      "▁boil ing",
+      "w m",
+      "可 将",
+      "uk u",
+      "u ku",
+      "▁ar rows",
+      "▁arr ows",
+      "▁arrow s",
+      "▁weigh ing",
+      "▁Res istance",
+      "食 盐",
+      "接受 了",
+      "im ited",
+      "imit ed",
+      "imi ted",
+      "▁For ex",
+      "▁Fore x",
+      "▁Fo rex",
+      "An other",
+      "A nother",
+      "▁pyl ori",
+      "▁host name",
+      "▁ hostname",
+      "东 山",
+      "发 售",
+      "魔 鬼",
+      "▁B M",
+      "▁ BM",
+      "ar te",
+      "art e",
+      "带 入",
+      "V K",
+      "特 产",
+      "用 在",
+      "然后 将",
+      "然 后将",
+      "al li",
+      "all i",
+      "a lli",
+      "称 赞",
+      "▁P ray",
+      "▁Pr ay",
+      "▁Pra y",
+      "▁T hu",
+      "▁Th u",
+      "教育 培训",
+      "▁dis sect",
+      "▁diss ect",
+      "员 的",
+      "鲜 艳",
+      "那个 人",
+      "那 个人",
+      "公 用",
+      "lf loor",
+      "l floor",
+      "tc lass",
+      "t class",
+      "Rec ords",
+      "Record s",
+      "▁up side",
+      "▁ups ide",
+      "▁eng lish",
+      "▁ english",
+      "亩 产",
+      "ni ce",
+      "nic e",
+      "n ice",
+      "x d",
+      "谁 的",
+      "发展 到",
+      "发病 率",
+      "可行 的",
+      "可 行的",
+      "武 功",
+      "n install",
+      "余 款",
+      "都 想",
+      "keep ers",
+      "keeper s",
+      "kee pers",
+      "▁Mig uel",
+      "法定 代理人",
+      "▁qu ilt",
+      "▁q uilt",
+      "▁qui lt",
+      "▁quil t",
+      "}\" ,\\",
+      "}\", \\",
+      "} \",\\",
+      "成立 以来",
+      "LO CAL",
+      "LOC AL",
+      "▁Is aac",
+      "▁Isa ac",
+      "▁pull s",
+      "▁pul ls",
+      "▁just ified",
+      "DE VELOPMENT",
+      "se tt",
+      "set t",
+      "s ett",
+      "nu s",
+      "n us",
+      "it ee",
+      "ite e",
+      "i tee",
+      "▁F as",
+      "▁Fa s",
+      "Par ty",
+      "Part y",
+      "P arty",
+      "▁l amps",
+      "▁lam ps",
+      "▁lamp s",
+      "▁Hot els",
+      "▁Hotel s",
+      "八 章",
+      "私 立",
+      "co sx",
+      "cos x",
+      "c osx",
+      "name of",
+      "▁ext inct",
+      ">） <",
+      "> ）<",
+      "厘米 的",
+      "厘 米的",
+      "有限 的",
+      "▁M ood",
+      "▁Mo od",
+      "▁w ells",
+      "▁well s",
+      "▁wel ls",
+      "ass adors",
+      "assador s",
+      "istor ical",
+      "istoric al",
+      "isto rical",
+      "裙 子",
+      "▁Q R",
+      "▁ QR",
+      "▁b ounce",
+      "▁bou nce",
+      "▁ir relevant",
+      "▁irre levant",
+      "▁( ~",
+      "Be nch",
+      "Ben ch",
+      "B ench",
+      "▁P rices",
+      "▁Pr ices",
+      "▁Price s",
+      "▁Pri ces",
+      "▁Improve ment",
+      "首 家",
+      "我爱 你",
+      "我 爱你",
+      "▁o mitted",
+      "▁om itted",
+      "▁omit ted",
+      "▁S eed",
+      "▁Se ed",
+      "▁See d",
+      "▁ Seed",
+      "西 山",
+      ">\\ '\\",
+      ">\\' \\",
+      "> \\'\\",
+      "── ──",
+      "申请 复议",
+      "这是 一种",
+      "这是一 种",
+      "这 是一种",
+      "pro ving",
+      "pr oving",
+      "prov ing",
+      "▁Admin istrator",
+      "▁Administr ator",
+      "供 货",
+      "闵 行",
+      "毛 主席",
+      "的 坐标",
+      "▁输出 格式",
+      "▁ 输出格式",
+      "三条 之规定",
+      "三 条之规定",
+      "fore ign",
+      "▁F iction",
+      "▁Fi ction",
+      "执 勤",
+      "洽 谈",
+      "鸡 精",
+      "do ng",
+      "don g",
+      "d ong",
+      "实践 活动",
+      "▁e rad",
+      "▁er ad",
+      "▁era d",
+      "power s",
+      "po wers",
+      "pow ers",
+      "p owers",
+      "ter ror",
+      "te rror",
+      "terr or",
+      "t error",
+      "start ing",
+      "star ting",
+      "▁sh uttle",
+      "▁shut tle",
+      "Some thing",
+      "S omething",
+      "▁down ward",
+      "▁neglig ible",
+      "▁permut ations",
+      "▁permutation s",
+      "惩 处",
+      "}$ )",
+      "} $)",
+      "▁p ens",
+      "▁pe ns",
+      "▁pen s",
+      "物流 有限公司",
+      "▁La place",
+      "▁Lapl ace",
+      "▁D imension",
+      "▁Dim ension",
+      "▁ Dimension",
+      "▁unfort unate",
+      "干 旱",
+      "prot obuf",
+      "proto buf",
+      "W IDTH",
+      "nCom mon",
+      "n Common",
+      "▁Bat man",
+      "▁br okers",
+      "▁bro kers",
+      "▁broke rs",
+      "▁broker s",
+      "征 字",
+      "\\\\ \\'",
+      "\\\\\\ '",
+      "\\ \\\\'",
+      "▁deleg ates",
+      "▁delegate s",
+      "▁rev isions",
+      "▁revision s",
+      "▁revis ions",
+      "传 授",
+      "坦 白",
+      "eg an",
+      "ega n",
+      "e gan",
+      "St eps",
+      "Step s",
+      "Ste ps",
+      "▁发行 日期",
+      "▁gluc ocortic",
+      "以 防",
+      "Fr e",
+      "F re",
+      "▁D ob",
+      "▁Do b",
+      "▁P RE",
+      "▁PR E",
+      "▁ PRE",
+      "▁Pro xy",
+      "▁Pr oxy",
+      "▁ Proxy",
+      "▁lub ric",
+      "右 上",
+      "穿 搭",
+      "本案 移送",
+      "▁输入 格式",
+      "▁ 输入格式",
+      "▁debut s",
+      "▁deb uts",
+      "▁Car roll",
+      "▁Carr oll",
+      "▁P urpose",
+      "▁Pur pose",
+      "▁activ ates",
+      "▁activate s",
+      "▁sem antics",
+      "▁semantic s",
+      "nT w",
+      "n Tw",
+      "如果 一个",
+      "成为 中国",
+      "成 为中国",
+      "▁the or",
+      "▁th eor",
+      "品 位",
+      "存 续",
+      "柔 毛",
+      "肉 质",
+      "这部 剧",
+      "▁W ear",
+      "▁We ar",
+      "▁past or",
+      "▁pa stor",
+      "▁pas tor",
+      "▁Bal ance",
+      "▁ Balance",
+      "▁exhib iting",
+      "▁exhibit ing",
+      "al sy",
+      "als y",
+      "▁crow ded",
+      "▁crowd ed",
+      "▁morph ine",
+      "▁medic inal",
+      "忽 悠",
+      "消 极",
+      "高 大",
+      "人才 的",
+      "▁Account ing",
+      "核 查",
+      "点 半",
+      "民事 执行中",
+      "nF igure",
+      "n Figure",
+      "▁group ing",
+      "▁cong estion",
+      "▁congest ion",
+      "▁ 动物",
+      "oc in",
+      "oci n",
+      "o cin",
+      "▁R enaissance",
+      "▁Re naissance",
+      "Pa n",
+      "P an",
+      "文明 的",
+      "文 明的",
+      "nM iss",
+      "nMi ss",
+      "n Miss",
+      "▁re tic",
+      "▁r etic",
+      "▁ret ic",
+      "en crypt",
+      "enc rypt",
+      "▁de prec",
+      "▁dep rec",
+      "城 关",
+      "bl ade",
+      "bla de",
+      "党 和",
+      "二氧化 碳",
+      "大学 英语",
+      "资本 市场",
+      "Pol ygon",
+      "Poly gon",
+      "et eenth",
+      "ete enth",
+      "e teenth",
+      "日向本院提出 申请",
+      "赶 到",
+      "节点 的",
+      "节 点的",
+      "▁U AE",
+      "▁mess y",
+      "▁mes sy",
+      "厨 师",
+      "同 仁",
+      "学院 的",
+      "学 院的",
+      "工商 局",
+      "软 硬件",
+      "安全 管理",
+      "▁immun oglobulin",
+      "单 品",
+      "ar de",
+      "ard e",
+      "▁he el",
+      "▁ heel",
+      "nAs ian",
+      "▁tribut ary",
+      "▁B US",
+      "▁BU S",
+      "nT ags",
+      "n Tags",
+      "▁d umb",
+      "▁du mb",
+      "自 诉",
+      "精神 和",
+      "个体 工商户",
+      "▁hour ly",
+      "Util ities",
+      "看 我",
+      "Co p",
+      "C op",
+      "就算 了",
+      "就 算了",
+      "金华 市",
+      "sk ill",
+      "ski ll",
+      "s kill",
+      "Pr essed",
+      "Press ed",
+      "Pres sed",
+      "P ressed",
+      "▁layer ed",
+      "▁lay ered",
+      "▁accum ulate",
+      "的 超",
+      "纠 错",
+      "nA nal",
+      "nAn al",
+      "n Anal",
+      "▁f lesh",
+      "▁fl esh",
+      "▁fle sh",
+      "▁Spe ech",
+      "主 料",
+      "ef e",
+      "e fe",
+      "h g",
+      "民 字第",
+      "▁deb ates",
+      "▁debate s",
+      "▁Insp ector",
+      "打 磨",
+      "元 减半收取",
+      "▁Mon ster",
+      "▁Mons ter",
+      "四 人",
+      "闹 钟",
+      "过滤 器",
+      "写 在",
+      "所有 人都",
+      "所有人 都",
+      "ITH UB",
+      "▁d ick",
+      "▁di ck",
+      "▁dic k",
+      "Fe bruary",
+      "▁enjoy ment",
+      "▁explos ive",
+      "碱 性",
+      "▁L iquid",
+      "▁en amel",
+      "▁E lements",
+      "▁El ements",
+      "▁Element s",
+      "▁Ele ments",
+      "▁ Elements",
+      "编 排",
+      "ER SION",
+      "ERS ION",
+      "▁L ecture",
+      "▁Lect ure",
+      "临 界",
+      "流 氓",
+      "可视 化",
+      "的 感情",
+      "▁Cop per",
+      "精 装",
+      "ly s",
+      "l ys",
+      "}/ {",
+      "} /{",
+      "Ho ur",
+      "H our",
+      "核酸 检测",
+      "▁D rew",
+      "▁Dr ew",
+      "nVis it",
+      "n Visit",
+      "值得一提 的是",
+      "后 被",
+      "日 趋",
+      "▁te ndon",
+      "▁ten don",
+      "▁tend on",
+      "Pres enter",
+      "Present er",
+      "An notations",
+      "Annotation s",
+      "大 腿",
+      "荔 枝",
+      "土地 上",
+      "土 地上",
+      "的 习惯",
+      "▁dep icted",
+      "▁depict ed",
+      "▁live stock",
+      "▁liv estock",
+      "▁Sacr amento",
+      "★ ★",
+      "平均 数",
+      "▁Bul ld",
+      "▁Bu lld",
+      "▁Bull d",
+      "▁Gu jar",
+      "inf ection",
+      "infect ion",
+      "合伙 人",
+      "▁micro tub",
+      "气 球",
+      "缓 期",
+      "▁X en",
+      "Dist ribution",
+      "义 词",
+      "字 号",
+      "拿 了",
+      "一切 都",
+      "大家 好",
+      "住 湖南省",
+      "▁E DIT",
+      "▁ED IT",
+      "▁ EDIT",
+      "Not hing",
+      "No thing",
+      "N othing",
+      "端 午",
+      "IL S",
+      "I LS",
+      "in ho",
+      "inh o",
+      "bl ogs",
+      "blog s",
+      "blo gs",
+      "b logs",
+      "ycl es",
+      "ycle s",
+      "yc les",
+      "y cles",
+      "O x",
+      "赶 快",
+      "文明 建设",
+      "it ches",
+      "itc hes",
+      "itch es",
+      "le vels",
+      "lev els",
+      "level s",
+      "im etric",
+      "imet ric",
+      "i metric",
+      "典 雅",
+      "采取 了",
+      "ep isode",
+      "▁prot agon",
+      "古 老",
+      "石 墨",
+      "使用 者",
+      "▁a po",
+      "▁ap o",
+      "▁ apo",
+      "▁B erry",
+      "▁Ber ry",
+      "▁ Berry",
+      "flat ten",
+      "▁list ened",
+      "▁listen ed",
+      "▁secure ly",
+      "理 学",
+      "离 合",
+      "紊 乱",
+      "顺 应",
+      "他 表示",
+      "可以 做",
+      "可 直接",
+      ")\" )\\",
+      ") \")\\",
+      "ed ic",
+      "edi c",
+      "e dic",
+      "▁M ON",
+      "▁MO N",
+      "▁ MON",
+      "▁v oter",
+      "▁vote r",
+      "▁vot er",
+      "▁vo ter",
+      "▁Import ant",
+      "▁re solving",
+      "▁res olving",
+      "兴 隆",
+      "Me an",
+      "M ean",
+      "▁A ST",
+      "▁AS T",
+      "▁ AST",
+      "F ALSE",
+      "布 斯",
+      "膝 盖",
+      "自 营",
+      "▁P izza",
+      "▁synchron ized",
+      "▁ 法",
+      "接 纳",
+      "▁} ;",
+      "▁ };",
+      "Be en",
+      "B een",
+      "▁r hs",
+      "▁rh s",
+      "共产 党员",
+      "共产党 员",
+      "develop ers",
+      "developer s",
+      "石 灰",
+      "以上 就是",
+      "步 步",
+      "这 才",
+      "tf inal",
+      "t final",
+      "第一百五十四条第一款第十一 项",
+      "底 蕴",
+      "林 果",
+      "▁West minster",
+      "埔 寨",
+      "nP hot",
+      "nPh ot",
+      "n Phot",
+      "▁M esh",
+      "▁Me sh",
+      "▁Mes h",
+      "▁ Mesh",
+      "▁Re set",
+      "▁Res et",
+      "▁ Reset",
+      "▁Ind igenous",
+      "江 北",
+      "知名 的",
+      "知 名的",
+      "▁S SD",
+      "▁SS D",
+      "cle aned",
+      "clean ed",
+      "字 节",
+      "美 德",
+      "W ave",
+      "sub section",
+      "一 言",
+      "敬 请",
+      "扣押 其",
+      "Hi de",
+      "H ide",
+      "il og",
+      "ilo g",
+      "i log",
+      "LE MENT",
+      "L EMENT",
+      "id able",
+      "ida ble",
+      "▁c ivic",
+      "▁ci vic",
+      "ococc al",
+      "ococ cal",
+      "制定 了",
+      "制 定了",
+      "短期 内",
+      "短 期内",
+      "▁Mu eller",
+      "楼 市",
+      "被 骗",
+      "on o",
+      "o no",
+      "la ct",
+      "l act",
+      "▁Y emen",
+      "▁Ye men",
+      "段 的",
+      "vi z",
+      "v iz",
+      "bol a",
+      "bo la",
+      "b ola",
+      "▁re places",
+      "▁replac es",
+      "▁replace s",
+      "Rob ert",
+      "Ro bert",
+      "R obert",
+      "▁Qu est",
+      "▁Que st",
+      "▁ Quest",
+      "▁Ca ucas",
+      "Start ing",
+      "Star ting",
+      "▁end urance",
+      "海 事",
+      "mod ify",
+      "▁se wing",
+      "▁sew ing",
+      "м о",
+      "突 击",
+      "罕 见",
+      "宝宝 的",
+      "本院 依照",
+      "硬盘 容量",
+      "▁Gu arant",
+      "▁billion s",
+      "▁bill ions",
+      "洗 浴",
+      "疏 松",
+      "政策 和",
+      "如果 你是",
+      "如果你 是",
+      "强制 措施",
+      "本 月",
+      "花 样",
+      "公元 前",
+      "nE st",
+      "n Est",
+      "坐 着",
+      "这个 词",
+      "▁连载 网站",
+      "▁Cl aire",
+      "▁Clair e",
+      "▁rem edy",
+      "▁remed y",
+      "▁Kenn eth",
+      "▁plot ting",
+      "▁re produced",
+      "▁rep roduced",
+      "▁reprodu ced",
+      "▁reproduce d",
+      "▁reproduc ed",
+      "解 说",
+      "▁( );\\",
+      "▁() ;\\",
+      "▁ ();\\",
+      "歷 史",
+      "白 头",
+      "防 汛",
+      "的人 来说",
+      "直 辖",
+      "▁M odi",
+      "▁Mod i",
+      "▁Mo di",
+      "Comp uter",
+      "Comput er",
+      "Compute r",
+      "▁ter minus",
+      "▁term inus",
+      "▁Con figure",
+      "▁Config ure",
+      "▁Conf igure",
+      "▁ Configure",
+      "▁ 根据",
+      "nP RO",
+      "n PRO",
+      "form ats",
+      "format s",
+      "R h",
+      "的 供述",
+      "▁advertis ements",
+      "▁advertisement s",
+      "▁advertise ments",
+      "▁m r",
+      "▁ mr",
+      "及 应用",
+      "nQ ue",
+      "nQu e",
+      "n Que",
+      "start up",
+      "PN G",
+      "P NG",
+      "dp i",
+      "d pi",
+      "nL and",
+      "nLa nd",
+      "n Land",
+      "▁P eer",
+      "▁Pe er",
+      "▁ Peer",
+      "set Timeout",
+      "王 的",
+      "▁p pm",
+      "▁pp m",
+      "▁ ppm",
+      "Oper ating",
+      "药 用",
+      "在 未来",
+      "▁frag ile",
+      "区 人民政府",
+      "or tical",
+      "ort ical",
+      "ortic al",
+      "é g",
+      "思想 的",
+      "思 想的",
+      "▁dis joint",
+      "两 端",
+      "▁I CT",
+      "▁IC T",
+      "▁ ICT",
+      "纪检 监察",
+      "▁e mitted",
+      "▁em itted",
+      "▁emit ted",
+      "umber land",
+      "影响 力的",
+      "影响力 的",
+      "nD utch",
+      "▁van ity",
+      "▁back bone",
+      "▁Typ ically",
+      "▁summ ation",
+      "能 看到",
+      "▁m appings",
+      "▁mapping s",
+      "▁ref lective",
+      "▁reflect ive",
+      "▁ 到",
+      "财产 刑",
+      "▁D ani",
+      "▁Dan i",
+      "▁Da ni",
+      "▁c och",
+      "▁co ch",
+      "▁coc h",
+      "合 击",
+      "cp y",
+      "c py",
+      "第二 种",
+      "▁h ats",
+      "▁ha ts",
+      "▁hat s",
+      "prob ably",
+      "碳 酸",
+      "保修 卡",
+      "ew ish",
+      "▁song writer",
+      "▁ songwriter",
+      "邂 逅",
+      "身份 证号",
+      "身份证 号",
+      "ask ets",
+      "asket s",
+      "te acher",
+      "▁mem oir",
+      "▁memo ir",
+      "▁writ ings",
+      "▁writing s",
+      "▁CH AR",
+      "▁ CHAR",
+      "cl osures",
+      "closure s",
+      "K M",
+      "▁A rr",
+      "▁Ar r",
+      "▁ Arr",
+      "▁r het",
+      "▁rh et",
+      "▁rhe t",
+      "▁ 艾",
+      "体 型",
+      "萃 取",
+      "骨 质",
+      "▁Dr ama",
+      "▁pan or",
+      "▁pa nor",
+      "▁repe ats",
+      "▁repeat s",
+      "▁secret ory",
+      "▁ant idepress",
+      "细 化",
+      "追 缴",
+      "▁N W",
+      "▁ NW",
+      "▁ 地点",
+      "par tum",
+      "part um",
+      "▁M eyer",
+      "▁Me yer",
+      "▁Mey er",
+      "但 对",
+      "收 支",
+      "nt z",
+      "n tz",
+      "解题 思路",
+      "想 必",
+      "BO T",
+      "B OT",
+      "}) )\\",
+      "})) \\",
+      "} ))\\",
+      "ci s",
+      "c is",
+      "Ga te",
+      "G ate",
+      "op hen",
+      "oph en",
+      "o phen",
+      "▁H ern",
+      "▁He rn",
+      "▁Her n",
+      "▁Ow ner",
+      "▁Own er",
+      "▁ Owner",
+      "▁al coholic",
+      "▁alcohol ic",
+      "▁ol factory",
+      "▁olf actory",
+      "▁orig inating",
+      "▁origin ating",
+      "* {",
+      "参数 的",
+      "参 数的",
+      "输入 法",
+      "ay ed",
+      "aye d",
+      "▁P hen",
+      "▁Ph en",
+      "父 子",
+      "Fun d",
+      "F und",
+      "信息 咨询",
+      "或 直接向",
+      "Ind icator",
+      "▁def ended",
+      "▁defend ed",
+      "▁elim inates",
+      "▁eliminate s",
+      "ind le",
+      "te ams",
+      "team s",
+      "ain less",
+      "▁imag ined",
+      "▁imagine d",
+      "▁play list",
+      "▁ playlist",
+      "▁char ities",
+      "▁clar ification",
+      "的 房屋",
+      "▁C oc",
+      "▁Co c",
+      "▁l oos",
+      "▁lo os",
+      "min ton",
+      "mi nton",
+      "m inton",
+      "▁gl aucoma",
+      "保 洁",
+      "al lic",
+      "all ic",
+      "alli c",
+      "▁Richard son",
+      "▁Richards on",
+      "嫉 妒",
+      "前 所未",
+      "收到 了",
+      "收 到了",
+      "日因 涉嫌",
+      "日 因涉嫌",
+      "▁ch ase",
+      "▁cha se",
+      "Th rough",
+      "▁cre amy",
+      "▁cream y",
+      "气 流",
+      "▁Iraq i",
+      "▁v illa",
+      "▁vill a",
+      "▁vi lla",
+      "利 害",
+      "▁gl imp",
+      "▁gli mp",
+      "劳 人",
+      "和 县",
+      "意义 上",
+      "新 民",
+      "难 的",
+      "st mt",
+      "stm t",
+      "倾 听",
+      "家 装",
+      "撞 击",
+      "重 叠",
+      "顺 畅",
+      "if fe",
+      "iff e",
+      "Sl ider",
+      "Slide r",
+      "▁co ached",
+      "▁coach ed",
+      "杀 害",
+      "▁analog ues",
+      "▁analogue s",
+      "亲 身",
+      "随 手",
+      "本次 活动",
+      "pl iers",
+      "plier s",
+      "p liers",
+      "▁l ively",
+      "▁li vely",
+      "▁live ly",
+      "▁liv ely",
+      "▁knock ed",
+      "▁di amonds",
+      "▁diamond s",
+      "▁dia monds",
+      "奴 隶",
+      "▁K aw",
+      "▁Ka w",
+      "▁F ergus",
+      "▁Ch ronic",
+      "▁Chron ic",
+      "$$ )",
+      "$ $)",
+      "▁/ \\",
+      "▁ /\\",
+      "为 导向",
+      "▁T ong",
+      "▁To ng",
+      "▁Ton g",
+      "nJ ournal",
+      "nJo urnal",
+      "n Journal",
+      "screen shot",
+      "screens hot",
+      "▁neutroph ils",
+      "▁neutrophil s",
+      "少 先",
+      "而 有",
+      "的新 型",
+      "的 新型",
+      "红楼 梦",
+      "舒适 的",
+      "en zie",
+      "enz ie",
+      "▁Act ions",
+      "▁Action s",
+      "▁ Actions",
+      "▁A SP",
+      "▁AS P",
+      "▁ ASP",
+      "咨询 医生",
+      "合同 的",
+      "柬 埔寨",
+      "▁b orough",
+      "▁bo rough",
+      "▁bor ough",
+      "▁ borough",
+      "看看 吧",
+      "ns pec",
+      "nsp ec",
+      "n spec",
+      "低 级",
+      "米 尔",
+      "京 剧",
+      "护 栏",
+      "AT IVE",
+      "and ez",
+      "ande z",
+      "▁c afe",
+      "▁ca fe",
+      "▁caf e",
+      "▁Am anda",
+      "▁glimp se",
+      "向 其",
+      "故 依照",
+      "▁O dd",
+      "▁Od d",
+      "▁[ {\\",
+      "▁[{ \\",
+      "不予 支持",
+      "bl ood",
+      "blo od",
+      "b lood",
+      "▁ch oir",
+      "▁cho ir",
+      "▁Bu enos",
+      "▁d ancer",
+      "▁dance r",
+      "▁dan cer",
+      "掌 声",
+      "概 论",
+      "防 盗",
+      "在 没有",
+      "}/ ${",
+      "} /${",
+      "二次根 式",
+      "Rot ation",
+      "R otation",
+      "▁mess enger",
+      "is te",
+      "ist e",
+      "i ste",
+      "▁s poon",
+      "▁sp oon",
+      "▁spo on",
+      "▁w reck",
+      "▁wr eck",
+      "▁wre ck",
+      "▁appe als",
+      "▁appeal s",
+      "▁begin ners",
+      "▁beginner s",
+      "的 努力",
+      "Sk ip",
+      "S kip",
+      "申请执行人 刘",
+      "▁immun o",
+      "▁imm uno",
+      "▁c affeine",
+      "▁chicken s",
+      "▁chick ens",
+      "▁stim ulates",
+      "▁stimul ates",
+      "▁stimulate s",
+      "溶 性",
+      "▁w ip",
+      "▁wi p",
+      "F riend",
+      "▁de pot",
+      "▁dep ot",
+      "▁d rawer",
+      "▁draw er",
+      "Acc ording",
+      "界 限",
+      "nO K",
+      "n OK",
+      "Squ are",
+      "S quare",
+      "enh anced",
+      "意 为",
+      "预 热",
+      "带领 下",
+      "日期 间",
+      "日 期间",
+      "round ed",
+      "r ounded",
+      "read lines",
+      "readline s",
+      "山 顶",
+      "T ur",
+      "▁ 音乐",
+      "每一 天",
+      "每 一天",
+      "▁Est onia",
+      "▁Eston ia",
+      "力 气",
+      "不 具有",
+      "Bl ob",
+      "B lob",
+      "▁m ould",
+      "南 充",
+      "坚 果",
+      "▁E k",
+      "多少 人",
+      "颁发 的",
+      "颁 发的",
+      "re member",
+      "rem ember",
+      "reme mber",
+      "▁predict able",
+      "火 热",
+      "的 工资",
+      "▁m ascul",
+      "▁masc ul",
+      "first Name",
+      "在我 看来",
+      "op ent",
+      "ope nt",
+      "open t",
+      "o pent",
+      "co ffee",
+      "c offee",
+      "▁c itations",
+      "▁cit ations",
+      "▁citation s",
+      "北 极",
+      "洛 杉",
+      "都 得",
+      "▁d d",
+      "▁ dd",
+      "但 还是",
+      "告诉 我们",
+      "告诉我 们",
+      "▁text ile",
+      "▁tex tile",
+      "岛 屿",
+      "痘 痘",
+      "▁L av",
+      "▁La v",
+      "▁S ig",
+      "▁Si g",
+      "▁ Sig",
+      "▁z ebra",
+      "▁ze bra",
+      "▁H ospit",
+      "▁sp ices",
+      "▁spice s",
+      "▁obstruct ive",
+      "不受申请执行时效 期间的限制",
+      "得 太",
+      "高 管",
+      "▁f rost",
+      "▁fr ost",
+      "▁fro st",
+      "▁F ortunately",
+      "▁Fort unately",
+      "状 元",
+      "不平 衡",
+      "不 平衡",
+      "绿化 率",
+      "▁Ph armac",
+      "▁Pharm ac",
+      "rib utable",
+      "ribut able",
+      "红 花",
+      "or en",
+      "ore n",
+      "o ren",
+      "▁ins ult",
+      "▁part itions",
+      "▁partition s",
+      "it zer",
+      "itz er",
+      "▁W ORK",
+      "▁WOR K",
+      "▁ WORK",
+      "ut i",
+      "u ti",
+      "▁Occ up",
+      "▁Oc cup",
+      "cover ed",
+      "cov ered",
+      "c overed",
+      "尼 克",
+      "山 脉",
+      "应 是",
+      "清 淡",
+      "▁N ET",
+      "▁NE T",
+      "▁ NET",
+      "nL ine",
+      "nLin e",
+      "n Line",
+      "▁has ht",
+      "▁hash t",
+      "design ated",
+      "▁dis asters",
+      "▁disaster s",
+      "了我 的",
+      "了 我的",
+      "▁gen ital",
+      "▁B JP",
+      "▁V il",
+      "▁Vi l",
+      "As sign",
+      "Ass ign",
+      "ir y",
+      "i ry",
+      "停留 在",
+      "停 留在",
+      "(' \\\\",
+      "( '\\\\",
+      "▁optim istic",
+      "我不 想",
+      "我 不想",
+      "fl ip",
+      "f lip",
+      "▁M orr",
+      "▁Mor r",
+      "▁Mo rr",
+      "▁p end",
+      "▁pe nd",
+      "▁pen d",
+      "▁ pend",
+      "nad min",
+      "n admin",
+      "▁U tility",
+      "▁Ut ility",
+      "▁Util ity",
+      "▁ Utility",
+      "声 誉",
+      "车 内",
+      "▁ 存储",
+      "▁d ull",
+      "▁du ll",
+      "▁f ing",
+      "▁fin g",
+      "▁fi ng",
+      "_ )",
+      "屏 障",
+      "往 上",
+      "高 产",
+      "二〇一六年 八月",
+      "私 营",
+      "IN AL",
+      "INA L",
+      "I NAL",
+      "▁S ki",
+      "▁Sk i",
+      "▁C lasses",
+      "▁Class es",
+      "▁ Classes",
+      "▁pol ling",
+      "▁poll ing",
+      "▁Disc rete",
+      "▁cancel ed",
+      "▁neutroph il",
+      "▁neutr ophil",
+      "# :",
+      "J R",
+      "▁ 圣",
+      "出版 时间",
+      "▁p olo",
+      "▁pol o",
+      "▁po lo",
+      "中国人民 解放军",
+      "国有 企业",
+      "▁Pers ian",
+      "▁wel ding",
+      "▁weld ing",
+      "谁 是",
+      "▁* ,",
+      "▁ *,",
+      "中的 一",
+      "中 的一",
+      "▁e yel",
+      "▁ey el",
+      "▁eye l",
+      "oul try",
+      "▁Bar nes",
+      "▁Barn es",
+      "▁bl ended",
+      "▁blend ed",
+      "orient ation",
+      "o rientation",
+      "调 剂",
+      "鼓 舞",
+      "精确 到",
+      "▁Americ as",
+      "▁America s",
+      "▁aper ture",
+      "▁ ε",
+      "房 东",
+      "让 它",
+      "信息 服务",
+      "▁ubiquit ous",
+      "和 建议",
+      "故本 案",
+      "故 本案",
+      "高效 的",
+      "意义 上的",
+      "意义上 的",
+      "成 形",
+      "PA T",
+      "P AT",
+      "因此 在",
+      "da emon",
+      "战 绩",
+      "▁G ender",
+      "▁Ge nder",
+      "▁Gen der",
+      "▁ Gender",
+      "▁Sloven ia",
+      "原 地",
+      "创新 型",
+      "创 新型",
+      "由 已知",
+      "hb ar",
+      "h bar",
+      "rf loor",
+      "r floor",
+      "▁Mem phis",
+      "▁att ained",
+      "▁attain ed",
+      "微波 炉",
+      ">' ,\\",
+      ">', \\",
+      "> ',\\",
+      "es ty",
+      "est y",
+      "e sty",
+      "▁R he",
+      "▁Rh e",
+      "▁A way",
+      "▁Aw ay",
+      "节 课",
+      "马 里",
+      "和 完善",
+      "nPart icip",
+      "n Particip",
+      "▁ob servers",
+      "▁observ ers",
+      "▁observe rs",
+      "▁observer s",
+      "ar b",
+      "a rb",
+      "你 现在",
+      "注射 液",
+      "▁N eo",
+      "▁Ne o",
+      "clean up",
+      "▁Can vas",
+      "▁ Canvas",
+      "▁Ex hibition",
+      "▁Exhib ition",
+      "▁hypert rophy",
+      "求 法",
+      "▁C ycle",
+      "▁Cy cle",
+      "▁Cycl e",
+      "▁ Cycle",
+      "▁op ted",
+      "▁opt ed",
+      "属于 中档题",
+      "▁mar sh",
+      "▁mars h",
+      "▁fore casting",
+      "▁forecast ing",
+      "主义 者",
+      "的 整体",
+      "nM ary",
+      "nMar y",
+      "nMa ry",
+      "n Mary",
+      "▁hard est",
+      "▁har dest",
+      "▁ecosystem s",
+      "▁eco systems",
+      "年 终",
+      "ra x",
+      "r ax",
+      "命令 行",
+      "▁D awn",
+      "▁Da wn",
+      "▁Daw n",
+      "\":[ ],\"",
+      "▁R outes",
+      "▁Ro utes",
+      "▁Route s",
+      "▁Rou tes",
+      "▁ Routes",
+      "浑 身",
+      "▁S au",
+      "▁Sa u",
+      "▁S we",
+      "▁Sw e",
+      "野生 动物",
+      "ce ans",
+      "cean s",
+      "cea ns",
+      "en sed",
+      "ens ed",
+      "ense d",
+      "划拨 被执行人",
+      "▁comm ence",
+      "差 分",
+      "▁о б",
+      "▁ об",
+      "不忘 初心",
+      "Man ifest",
+      "棕 色",
+      "双人 间",
+      "双 人间",
+      "CA SE",
+      "CAS E",
+      "C ASE",
+      "▁C ul",
+      "▁Cu l",
+      "▁v ow",
+      "▁vo w",
+      "▁c atar",
+      "▁ca tar",
+      "▁cat ar",
+      "str ained",
+      "stra ined",
+      "strain ed",
+      "s trained",
+      "repre neur",
+      "reprene ur",
+      "▁type name",
+      "▁typ ename",
+      "▁ typename",
+      "▁dis placed",
+      "agrang ian",
+      "nOr iginal",
+      "n Original",
+      "金 的",
+      "元素 的",
+      "元 素的",
+      "公安 部",
+      "包含 一个",
+      "▁El len",
+      "▁Ell en",
+      "▁gar ner",
+      "有效 期",
+      "▁c arn",
+      "▁car n",
+      "▁ca rn",
+      "▁aut oc",
+      "▁auto c",
+      "▁au toc",
+      "▁pr imer",
+      "▁prim er",
+      "▁prime r",
+      "▁pri mer",
+      "K g",
+      "几 句",
+      "▁Key Error",
+      "变 现",
+      "庞 大",
+      "给 被告",
+      "▁be ers",
+      "▁beer s",
+      "▁bee rs",
+      "▁de tox",
+      "▁det ox",
+      "▁String Builder",
+      "▁ StringBuilder",
+      "▁N SA",
+      "▁NS A",
+      "Mo tor",
+      "M otor",
+      "embed ded",
+      "pro portion",
+      "prop ortion",
+      "间 断",
+      "作出 裁定",
+      "本身 就是",
+      "血液 循环",
+      "exec utable",
+      "领先 的",
+      "ap ro",
+      "a pro",
+      "Ma gic",
+      "Mag ic",
+      "M agic",
+      "▁scan f",
+      "▁ scanf",
+      "▁lock smith",
+      "▁locks mith",
+      "▁register ing",
+      "j d",
+      "上 部",
+      "真 情",
+      "Sl eep",
+      "S leep",
+      "▁mult is",
+      "▁multi s",
+      "▁mul tis",
+      "▁cho pped",
+      "▁chop ped",
+      "号 称",
+      "il is",
+      "ili s",
+      "i lis",
+      "明显 高于",
+      "Liter al",
+      "Lite ral",
+      "arch itecture",
+      "▁ 下",
+      "不一 致",
+      "不 一致",
+      "ve ns",
+      "ven s",
+      "v ens",
+      "北 斗",
+      "婷 婷",
+      "con c",
+      "co nc",
+      "c onc",
+      "ax ial",
+      "ex ter",
+      "ext er",
+      "▁a udi",
+      "▁aud i",
+      "▁au di",
+      "▁ audi",
+      "▁Sh ield",
+      "▁Shi eld",
+      "用 书",
+      "ge ries",
+      "ger ies",
+      "g eries",
+      "▁High land",
+      "▁ В",
+      "交 集",
+      "Fi g",
+      "F ig",
+      "tu r",
+      "t ur",
+      "ross e",
+      "ros se",
+      "r osse",
+      "▁P ixel",
+      "▁Pix el",
+      "▁ Pixel",
+      "iropr actic",
+      "▁reprodu cible",
+      "▁repro ducible",
+      "▁reproduc ible",
+      "年 末",
+      "提 要",
+      "都是 有",
+      "都 是有",
+      "进一步 提高",
+      "▁long time",
+      "嘲 笑",
+      "纸 上",
+      "椭圆 的",
+      "椭 圆的",
+      "等 行业",
+      "Sh adow",
+      "Sha dow",
+      "▁collabor ated",
+      "▁collaborate d",
+      "la ne",
+      "lan e",
+      "l ane",
+      "ol one",
+      "olo ne",
+      "ri ages",
+      "ria ges",
+      "riage s",
+      "▁b logger",
+      "▁blog ger",
+      "uff iciency",
+      "或 缺",
+      "沿 线",
+      "需 在",
+      "PI N",
+      "P IN",
+      "▁C s",
+      "▁ Cs",
+      "流 产",
+      "一些 人",
+      "倾 城",
+      "▁O nc",
+      "▁On c",
+      "停车 服务",
+      "停 车服务",
+      "nAr gent",
+      "▁Mark ets",
+      "▁Market s",
+      "ra v",
+      "r av",
+      "抗 氧化",
+      "(\" \");\\",
+      "(\"\" );\\",
+      "▁hung er",
+      "▁hun ger",
+      "▁pos ture",
+      "▁post ure",
+      "▁P ipeline",
+      "▁Pipe line",
+      "▁Pip eline",
+      "▁ Pipeline",
+      "出 锅",
+      "友 人",
+      "▁S antiago",
+      "▁wheel chair",
+      "oy e",
+      "o ye",
+      "上 发表",
+      "不是 说",
+      "须 经",
+      "ys tone",
+      "yst one",
+      "y stone",
+      "▁comfort ably",
+      "典 范",
+      "晓 明",
+      "渊 博",
+      "Wi re",
+      "W ire",
+      "nTe am",
+      "n Team",
+      "▁ign or",
+      "▁ig nor",
+      "Pos itive",
+      "▁term inate",
+      "ve ls",
+      "vel s",
+      "v els",
+      "比较 好的",
+      "比较好 的",
+      "比 较好的",
+      "中西医 结合",
+      "▁Cl aim",
+      "▁ Claim",
+      "▁legisl ature",
+      "修 补",
+      "须 知",
+      "▁Bo at",
+      "▁Gl enn",
+      "▁Glen n",
+      "▁Gle nn",
+      "▁g rief",
+      "▁gr ief",
+      "▁gri ef",
+      "▁In strument",
+      "▁ Instrument",
+      "] ^",
+      "磁 场",
+      "结果 是",
+      "ct omy",
+      "▁n ause",
+      "▁na use",
+      "▁sur render",
+      "▁surre nder",
+      "▁delight ful",
+      "狠 狠",
+      "讲 的",
+      "▁S eq",
+      "▁Se q",
+      "▁ Seq",
+      "nDe aths",
+      "nDeath s",
+      "▁Corn wall",
+      "▁bl own",
+      "▁blow n",
+      "▁blo wn",
+      "con straint",
+      "上 个",
+      "天 数",
+      "率 领",
+      "▁dec lining",
+      "▁decl ining",
+      "IS BN",
+      "nStart ing",
+      "n Starting",
+      "等 价",
+      "▁t utor",
+      "▁tu tor",
+      "▁tut or",
+      "▁lig ament",
+      "也 无",
+      "flu or",
+      "▁Struct ural",
+      "现实 的",
+      "现 实的",
+      "▁经营 状态",
+      "pro files",
+      "profile s",
+      "prof iles",
+      "东 侧",
+      "帅 气",
+      "想起 来",
+      "想 起来",
+      "▁C UR",
+      "▁CU R",
+      "▁ CUR",
+      "京 都",
+      "mm m",
+      "m mm",
+      "抗 击",
+      "▁f ucked",
+      "▁fuck ed",
+      "nBe etles",
+      "符合法定 减刑条件",
+      "宽 敞",
+      "per l",
+      "pe rl",
+      "hor se",
+      "h orse",
+      "Data Type",
+      "▁ 戴",
+      "来 访",
+      "等 候",
+      "铅 笔",
+      "我 以为",
+      "▁Josh ua",
+      "▁Jos hua",
+      "▁M apping",
+      "▁Map ping",
+      "▁Ma pping",
+      "▁ Mapping",
+      "交 友",
+      "▁四 川",
+      "▁ 四川",
+      "▁P KC",
+      "▁PK C",
+      "▁C isco",
+      "/ `",
+      "开 局",
+      "血 栓",
+      "bi z",
+      "b iz",
+      "中的 一个",
+      "中的一 个",
+      "中 的一个",
+      "▁sp ends",
+      "▁spend s",
+      "postgres ql",
+      "凡 是",
+      "tu m",
+      "t um",
+      "▁R P",
+      "▁ RP",
+      "re ed",
+      "ree d",
+      "r eed",
+      "▁Les lie",
+      "▁wait ed",
+      "▁wa ited",
+      "▁Edit orial",
+      "▁Editor ial",
+      "温 室",
+      "了一 点",
+      "了 一点",
+      "科 科长",
+      "墙 壁",
+      "时 应",
+      "就在 于",
+      "就 在于",
+      "nL ive",
+      "n Live",
+      "▁D esk",
+      "▁De sk",
+      "▁Des k",
+      "bet ter",
+      "b etter",
+      "▁F usion",
+      "上 人",
+      "▁r c",
+      "▁ rc",
+      "▁Bo ost",
+      "▁ Boost",
+      "ge ttext",
+      "get text",
+      "▁pr incess",
+      "▁princ ess",
+      "▁prince ss",
+      "▁ 斯",
+      "卫 东",
+      "的 她",
+      "▁F ake",
+      "▁Fa ke",
+      "▁ Fake",
+      "▁Em irates",
+      "的车 辆",
+      "的 车辆",
+      "建筑 设计",
+      "nQ ual",
+      "nQu al",
+      "n Qual",
+      "填 报",
+      "尚 可",
+      "度 高",
+      "日被 告",
+      "日 被告",
+      "本裁定书送达后 立即执行",
+      "▁comprehens ion",
+      "北 上",
+      "的重 量",
+      "的 重量",
+      "转换 为",
+      "工程 设计",
+      "▁Sh ower",
+      "▁Show er",
+      "▁Hard ware",
+      "▁tit anium",
+      "ä n",
+      "▁\" #{",
+      "▁\"# {",
+      "不能 提供",
+      "▁P ara",
+      "▁Par a",
+      "▁Pa ra",
+      "Con figure",
+      "Config ure",
+      "Conf igure",
+      "▁attract ing",
+      "▁attr acting",
+      "▁per itoneal",
+      "围 棋",
+      "推 测",
+      "▁E y",
+      "col ored",
+      "color ed",
+      "m olecular",
+      "为国 家",
+      "为 国家",
+      "ce ral",
+      "cer al",
+      "c eral",
+      "▁ar gu",
+      "▁arg u",
+      "相关 法律规定",
+      "▁ro oted",
+      "▁root ed",
+      "高 水平",
+      "▁Sem inar",
+      "▁Semi nar",
+      "▁coat ings",
+      "▁coating s",
+      "快 要",
+      "杰 克",
+      "金 奖",
+      "ij n",
+      "计算 方法",
+      "izar re",
+      "红 酒",
+      "霸 王",
+      "quis ites",
+      "quisite s",
+      "位 居",
+      "le z",
+      "l ez",
+      "ia ni",
+      "ian i",
+      "i ani",
+      "▁CO PD",
+      "LOG GER",
+      "Dest ination",
+      "▁ann ouncing",
+      "▁announc ing",
+      "▁ 异",
+      "披 针",
+      "ik k",
+      "i kk",
+      "}` \\",
+      "} `\\",
+      "▁[ :",
+      "▁ [:",
+      "an an",
+      "ana n",
+      "a nan",
+      "▁re pay",
+      "▁rep ay",
+      "▁t ensions",
+      "▁tens ions",
+      "▁tension s",
+      "擦 拭",
+      "an tis",
+      "ant is",
+      "anti s",
+      "nf ind",
+      "nfi nd",
+      "n find",
+      "山 人",
+      "巴 克",
+      "项 的",
+      "Un ion",
+      "▁k icks",
+      "▁kick s",
+      "▁ki cks",
+      "时 会",
+      "根 源",
+      "生 生",
+      "ad vert",
+      "adv ert",
+      "lin king",
+      "link ing",
+      "l inking",
+      "▁rel ieve",
+      "▁re cession",
+      "▁rec ession",
+      "M H",
+      "末 端",
+      "必要 性",
+      "第一 种",
+      "第 一种",
+      "in ki",
+      "ink i",
+      "▁hom olog",
+      "存 的",
+      "et ches",
+      "etch es",
+      "etc hes",
+      "find One",
+      "nCh oose",
+      "n Choose",
+      "▁dial ect",
+      "▁dia lect",
+      "▁o phthal",
+      "▁Bang alore",
+      "格 拉",
+      "男孩 子",
+      "男 孩子",
+      "ir is",
+      "iri s",
+      "i ris",
+      "▁Food s",
+      "▁Foo ds",
+      "▁Fo ods",
+      "评 判",
+      "Tr ip",
+      "Tri p",
+      "T rip",
+      "▁C ake",
+      "▁Ca ke",
+      "▁G erald",
+      "▁Ger ald",
+      "lp atterns",
+      "l patterns",
+      "▁over load",
+      "▁overl oad",
+      "螺 纹",
+      "科研 成果",
+      "nRec ipients",
+      "得 意",
+      "村 内",
+      "特 区",
+      "成功 率",
+      "成 功率",
+      "▁p ulp",
+      "▁pul p",
+      "▁pu lp",
+      "Re ceiver",
+      "Receive r",
+      "▁p ointers",
+      "▁point ers",
+      "▁po inters",
+      "▁pointer s",
+      "往 返",
+      "网 盘",
+      "ies el",
+      "ie sel",
+      "ki nson",
+      "kin son",
+      "kins on",
+      "k inson",
+      "▁Cub an",
+      "▁Cu ban",
+      "▁Cuba n",
+      "号刑事 附带民事",
+      "▁mon ocytes",
+      "▁monocyte s",
+      "按 下",
+      "ah i",
+      "a hi",
+      "oor s",
+      "oo rs",
+      "o ors",
+      "食品 药品",
+      "▁g ing",
+      "▁gi ng",
+      "▁gin g",
+      "▁ ging",
+      "vol tage",
+      "台 中",
+      "了一 套",
+      "了 一套",
+      "um bia",
+      "umb ia",
+      "生 了",
+      "▁G ren",
+      "▁Gr en",
+      "▁Gre n",
+      "▁数据 范围",
+      "▁ 数据范围",
+      "▁pitch ed",
+      "▁pit ched",
+      "▁thought ful",
+      "主 频",
+      "朝 廷",
+      "▁M ist",
+      "▁Mi st",
+      "▁Mis t",
+      "Cl ause",
+      "局 域",
+      "Kn ow",
+      "K now",
+      "和我 们",
+      "和 我们",
+      "可以 申请",
+      "qual ified",
+      "就是 你",
+      "就 是你",
+      "ier re",
+      "海 棠",
+      "菠 萝",
+      "研究 对象",
+      "ch urch",
+      "▁col span",
+      "▁cols pan",
+      "▁ colspan",
+      "▁Karn ataka",
+      "[ _",
+      "E asy",
+      "as tom",
+      "ast om",
+      "plete ly",
+      "plet ely",
+      "prising ly",
+      "画 质",
+      "软 骨",
+      "我一 个",
+      "我 一个",
+      "▁F lying",
+      "▁Fl ying",
+      "▁Fly ing",
+      "▁com merce",
+      "▁ commerce",
+      "有 感",
+      "沙 拉",
+      "几分 钟",
+      "几 分钟",
+      "▁N issan",
+      "▁inter stitial",
+      "▁nan ot",
+      "▁na not",
+      "▁nano t",
+      "▁und oubtedly",
+      "上 千",
+      "偿 付",
+      "四 是",
+      "回 升",
+      "通 关",
+      "无 明显",
+      "▁hand ic",
+      "▁han dic",
+      "K Y",
+      "tr ait",
+      "tra it",
+      "t rait",
+      "prop Types",
+      "inter active",
+      "▁disappoint ing",
+      "的 性能",
+      "▁sp lash",
+      "▁spl ash",
+      "▁ splash",
+      "▁comorb id",
+      "▁spot light",
+      "tL ist",
+      "t List",
+      "中国 邮政储蓄银行股份有限公司",
+      "店 面",
+      "曾 用",
+      "ra ts",
+      "rat s",
+      "r ats",
+      "总 和",
+      "过 硬",
+      "(( (",
+      "( ((",
+      "mA h",
+      "m Ah",
+      "▁X ia",
+      "▁Xi a",
+      "▁D ong",
+      "▁Do ng",
+      "▁Don g",
+      "财产 的规定",
+      "财产的 规定",
+      "▁W riter",
+      "▁Writ er",
+      "▁Write r",
+      "▁Wr iter",
+      "▁ Writer",
+      "hedul ing",
+      "▁Bloom berg",
+      "扩 建",
+      "西 县",
+      "同 底数",
+      "▁Lux em",
+      "nMus ical",
+      "nMusic al",
+      "w c",
+      "风 俗",
+      "怎么 能",
+      "t continue",
+      "写 入",
+      "数 千",
+      "闪 烁",
+      "时间 是",
+      "洛杉 矶",
+      "er eg",
+      "ere g",
+      "e reg",
+      "col um",
+      "c olum",
+      "ce phal",
+      "cep hal",
+      "全 能",
+      "旗 帜",
+      "<> (",
+      "< >(",
+      "在同 一",
+      "在 同一",
+      "▁afore mentioned",
+      "喇 叭",
+      "录 用",
+      "惊 人",
+      "鉴定 意见",
+      "▁parent heses",
+      "王 世",
+      "支付 货款",
+      "nD ay",
+      "n Day",
+      "上诉 人的",
+      "上诉人 的",
+      "住所地 北京市",
+      "▁end for",
+      "in ternet",
+      "inter net",
+      "intern et",
+      "排 气",
+      "毛 孔",
+      "▁n b",
+      "▁ nb",
+      "▁с о",
+      "▁f lick",
+      "▁fl ick",
+      "初 恋",
+      "王者 荣耀",
+      "or ate",
+      "ora te",
+      "o rate",
+      "our met",
+      "势 头",
+      "cell ent",
+      "强制 性",
+      "▁I Ds",
+      "▁ID s",
+      "▁ IDs",
+      "▁out per",
+      "Int roduction",
+      "▁b b",
+      "▁ bb",
+      "在 里面",
+      "我不 会",
+      "我 不会",
+      "旅游 业",
+      "tox ic",
+      "t oxic",
+      "▁cortic oster",
+      "交 替",
+      "南 阳",
+      "点 缀",
+      "op o",
+      "o po",
+      "sl ow",
+      "s low",
+      "▁M uk",
+      "▁Mu k",
+      "Rob ot",
+      "Ro bot",
+      "ra mer",
+      "ram er",
+      "rame r",
+      "r amer",
+      "Mill is",
+      "Mil lis",
+      "蕴 含",
+      "大家 都知道",
+      "大家都 知道",
+      "▁cap ita",
+      "石 膏",
+      "nJ eff",
+      "n Jeff",
+      "▁K urt",
+      "▁Ku rt",
+      "▁Kur t",
+      "平 装",
+      "田 径",
+      "喜欢 你",
+      "▁susp icion",
+      "J A",
+      "不再 是",
+      "ol oad",
+      "olo ad",
+      "o load",
+      "▁C lock",
+      "▁Cl ock",
+      "▁Clo ck",
+      "▁ Clock",
+      "▁un des",
+      "▁u ndes",
+      "▁und es",
+      "▁br akes",
+      "▁bra kes",
+      "▁brake s",
+      "排 在",
+      "降 水",
+      "▁f ox",
+      "▁fo x",
+      "▁ fox",
+      "万 个",
+      "物 证",
+      "▁La uren",
+      "▁Laure n",
+      "▁Laur en",
+      "▁Me chanics",
+      "▁Mechan ics",
+      "N g",
+      "文 集",
+      "Bal l",
+      "Ba ll",
+      "B all",
+      "认真 学习",
+      "best os",
+      "bes tos",
+      "▁D ating",
+      "▁Dat ing",
+      "▁Da ting",
+      "arth ritis",
+      "脸 色",
+      "ndef ine",
+      "nde fine",
+      "n define",
+      "本 草",
+      "hu i",
+      "h ui",
+      "▁已 注销",
+      "AD MIN",
+      "▁Po etry",
+      "LC D",
+      "L CD",
+      "OV A",
+      "O VA",
+      "充电 器",
+      "充 电器",
+      "▁lot tery",
+      "丹 麦",
+      "五 官",
+      "利用 率",
+      "业内 人士",
+      "▁Put in",
+      "▁Pu tin",
+      "▁Arm strong",
+      "宝 藏",
+      "意识 的",
+      "Template s",
+      "Tem plates",
+      "▁veter inary",
+      "▁veterin ary",
+      "▁Observ atory",
+      "▁P Y",
+      "▁ PY",
+      "掌握 了",
+      "▁** [",
+      "产业 结构",
+      "▁am azed",
+      "▁st icking",
+      "▁stick ing",
+      "d ifference",
+      "· \\",
+      "为 己",
+      "医 用",
+      "ys on",
+      "y son",
+      "列出 方程",
+      "ok ine",
+      "okin e",
+      "oki ne",
+      "化 管理",
+      "▁ch alk",
+      "▁ chalk",
+      "▁en codes",
+      "▁enc odes",
+      "▁encode s",
+      "▁un predict",
+      "ST A",
+      "S TA",
+      "最 适合",
+      "it als",
+      "ital s",
+      "ita ls",
+      "cookie s",
+      "cook ies",
+      "文 具",
+      "执 民字第",
+      "ex act",
+      "or acle",
+      "ora cle",
+      "orac le",
+      "货 架",
+      "▁ap ical",
+      "▁api cal",
+      "▁re located",
+      "▁rel ocated",
+      "AB S",
+      "A BS",
+      "罪犯 改造",
+      "nu cle",
+      "n ucle",
+      "had oop",
+      "h adoop",
+      "理论 的",
+      "en ne",
+      "enn e",
+      "这部 电影",
+      "nA ccess",
+      "nAcc ess",
+      "nAc cess",
+      "n Access",
+      "▁un affected",
+      "单 片",
+      "但是 你",
+      "但 是你",
+      "▁T et",
+      "▁Te t",
+      "个人 信息",
+      "▁st aged",
+      "▁stage d",
+      "▁radio active",
+      "缘 分",
+      "nL aw",
+      "nLa w",
+      "n Law",
+      "市场 竞争",
+      "▁reg isters",
+      "▁register s",
+      "▁sat ellites",
+      "▁satellite s",
+      "í s",
+      "wa v",
+      "w av",
+      "se cts",
+      "sec ts",
+      "sect s",
+      "s ects",
+      "EX PORT",
+      "EXP ORT",
+      "▁on click",
+      "▁onc lick",
+      "外 婆",
+      "和 自己",
+      "▁I dea",
+      "▁Id ea",
+      "▁Ide a",
+      "igm oid",
+      "▁di aling",
+      "▁dial ing",
+      "▁dia ling",
+      "▁emotional ly",
+      "▁emotion ally",
+      "▁spont aneously",
+      "▁spontaneous ly",
+      "采 摘",
+      "进行 判断",
+      "▁公司 简介",
+      "▁trig lycer",
+      "▁Profess ion",
+      "▁Prof ession",
+      "但 她",
+      "光 大",
+      "厂 的",
+      "fess or",
+      "ot ions",
+      "otion s",
+      "oti ons",
+      "▁att acker",
+      "▁attack er",
+      "Ra ises",
+      "ch allenge",
+      "拨 打",
+      "输 了",
+      "这 首",
+      "想 办法",
+      "▁l abs",
+      "▁la bs",
+      "▁lab s",
+      "▁ labs",
+      "ap tops",
+      "apt ops",
+      "aptop s",
+      "in side",
+      "ins ide",
+      "▁c hill",
+      "▁ch ill",
+      "▁chi ll",
+      "RE N",
+      "R EN",
+      "年 以上",
+      "INST ALL",
+      "out ines",
+      "outine s",
+      "第七 十二条第一款",
+      "第七十二条 第一款",
+      "第一百四十五条 宣判前",
+      "▁ 后",
+      "es p",
+      "e sp",
+      "▁F erm",
+      "▁Fe rm",
+      "▁Fer m",
+      "动 词",
+      "新 款",
+      "月 球",
+      "cha ft",
+      "▁Sur rey",
+      "▁Surre y",
+      "▁Ad visor",
+      "▁Adv isor",
+      "▁Advis or",
+      "▁ Advisor",
+      "▁Re gression",
+      "▁Reg ression",
+      "▁sign alling",
+      "▁signal ling",
+      "刊 登",
+      "哔 哩",
+      "▁L HC",
+      "▁LH C",
+      "取消 预订",
+      "sand box",
+      "s andbox",
+      "保 姆",
+      "拼 接",
+      "玩 了",
+      "聆 听",
+      "▁dr yer",
+      "▁dry er",
+      "想 像",
+      "og on",
+      "ogo n",
+      "o gon",
+      "▁read me",
+      "▁ readme",
+      "▁ 洪",
+      "全 额",
+      "写字 楼",
+      "▁{ })",
+      "▁{} )",
+      "▁ {})",
+      "项目 管理",
+      "▁f reed",
+      "▁fr eed",
+      "▁free d",
+      "▁fre ed",
+      "▁Eth ics",
+      "太 湖",
+      "有 据",
+      "可以 参考",
+      "bec ca",
+      "b ecca",
+      "▁Em ployment",
+      "▁Employ ment",
+      "▁Sil va",
+      "Draw ing",
+      "▁a sleep",
+      "▁as leep",
+      "▁Fl owers",
+      "▁Flow ers",
+      "▁Flo wers",
+      "▁Flower s",
+      "▁aband on",
+      "▁hil arious",
+      "F K",
+      "广 义",
+      "的 的",
+      "要 加强",
+      "配置 文件",
+      "▁day time",
+      "▁Act ivities",
+      "▁Activ ities",
+      "游 乐",
+      "▁Ext ended",
+      "▁ Extended",
+      "nPublic ations",
+      "nPublication s",
+      "宜 兴",
+      "登 山",
+      "粤 语",
+      "ha z",
+      "h az",
+      "H idden",
+      "▁p arity",
+      "▁par ity",
+      "▁pari ty",
+      "▁pro active",
+      "nU K",
+      "n UK",
+      "和 研究",
+      "▁Leg acy",
+      "Sc heduler",
+      "Schedule r",
+      "喂 养",
+      "护 照",
+      "邮 电",
+      "基本 都是",
+      "▁exc ision",
+      "▁G ö",
+      "ac os",
+      "aco s",
+      "a cos",
+      "上诉 理由",
+      "撤诉 条件",
+      "▁ac ne",
+      "Ex amples",
+      "Example s",
+      "春 晚",
+      "予以 查封",
+      "Sci Net",
+      "早 年",
+      "设备 和",
+      "▁P Rs",
+      "▁PR s",
+      "ff ffff",
+      "fff fff",
+      "ffff ff",
+      "▁t ales",
+      "▁tal es",
+      "▁ta les",
+      "▁tale s",
+      "抉 择",
+      "招 募",
+      "肉 类",
+      "▁R onald",
+      "▁Ron ald",
+      "▁scholar ships",
+      "▁scholars hips",
+      "▁scholarship s",
+      "ML E",
+      "M LE",
+      "本院将 依法",
+      "判处有期徒刑 三年",
+      "不 惜",
+      "开 后",
+      "▁R id",
+      "▁Ri d",
+      "用户 体验",
+      "▁bur ns",
+      "▁burn s",
+      "▁di ode",
+      "Work flow",
+      "through put",
+      "并 由",
+      "房 租",
+      "螃 蟹",
+      "({ \\\\",
+      "({\\ \\",
+      "( {\\\\",
+      "de ck",
+      "dec k",
+      "d eck",
+      "▁g ri",
+      "▁gr i",
+      "ph erd",
+      "pher d",
+      "▁Ge off",
+      "▁Geo ff",
+      "▁s urnames",
+      "▁sur names",
+      "▁surname s",
+      "▁Out standing",
+      "▁bi ologically",
+      "▁biological ly",
+      "▁biologic ally",
+      "皮 质",
+      "预 后",
+      "城市 建设",
+      "金融 服务",
+      "▁re de",
+      "▁r ede",
+      "▁red e",
+      "ACT ION",
+      "A CTION",
+      "▁F lower",
+      "▁Fl ower",
+      "▁Flow er",
+      "▁Flo wer",
+      "▁sw allow",
+      "▁Ar med",
+      "▁Arm ed",
+      "▁Hand ler",
+      "▁Handle r",
+      "▁ Handler",
+      "▁exp ands",
+      "▁expand s",
+      "▁dom inance",
+      "m A",
+      "一 瓶",
+      "还 剩",
+      "▁H H",
+      "▁ HH",
+      "▁l amb",
+      "▁la mb",
+      "▁lam b",
+      "▁rel ied",
+      "▁ 盐",
+      "或 多",
+      "盛 宴",
+      "和 预先",
+      "▁v ivid",
+      "▁vi vid",
+      "▁Event ually",
+      "苯 丙",
+      "is tar",
+      "ist ar",
+      "ista r",
+      "i star",
+      "的 名",
+      "▁in h",
+      "▁i nh",
+      "▁ inh",
+      "▁O wen",
+      "▁Ow en",
+      "▁\\' \\')\\",
+      "▁\\'\\' )\\",
+      "▁\\'\\ ')\\",
+      "▁\\'\\') \\",
+      "迹 象",
+      "TI NG",
+      "T ING",
+      "▁prospective ly",
+      "▁prospect ively",
+      "四 次",
+      "J et",
+      "下来 了",
+      "下 来了",
+      "ar ance",
+      "ara nce",
+      "aran ce",
+      "a rance",
+      "ca sh",
+      "cas h",
+      "c ash",
+      "客户 提供",
+      "范围 内的",
+      "范围内 的",
+      "ro uting",
+      "rou ting",
+      "r outing",
+      "▁tumor igen",
+      "球 菌",
+      "ma k",
+      "m ak",
+      "知道 自己",
+      "组织 开展",
+      "▁str omal",
+      "▁strom al",
+      "▁stro mal",
+      "刀 具",
+      "▁适宜 环境",
+      "一 圈",
+      "▁/ \\\\",
+      "▁/\\ \\",
+      "▁ /\\\\",
+      "▁anomal y",
+      "▁micro phone",
+      "▁cult ivation",
+      "▁cultiv ation",
+      "买 车",
+      "▁L ots",
+      "▁Lo ts",
+      "▁Lot s",
+      "graph s",
+      "▁imm ense",
+      "▁n x",
+      "▁ nx",
+      "ad oc",
+      "ado c",
+      "a doc",
+      "arct ica",
+      "arctic a",
+      "con scious",
+      "gi s",
+      "g is",
+      "an din",
+      "and in",
+      "andi n",
+      "▁THE Y",
+      "▁TH EY",
+      "金 黄",
+      "IP O",
+      "I PO",
+      "J ax",
+      "实际 行动",
+      "联系 电话",
+      "▁se ized",
+      "▁seiz ed",
+      "▁play wright",
+      "HO W",
+      "H OW",
+      "如果 能",
+      "明确 了",
+      "的时候 就",
+      "nC ase",
+      "n Case",
+      "to Have",
+      "▁B ened",
+      "▁Be ned",
+      "▁Ben ed",
+      "▁o mega",
+      "▁om ega",
+      "▁ omega",
+      "▁endot helium",
+      "▁Sub scription",
+      "▁ Subscription",
+      "印 花",
+      "▁P eg",
+      "▁Pe g",
+      "▁stre pt",
+      "▁w ishing",
+      "▁wish ing",
+      "▁wis hing",
+      "不 打",
+      "本科 学历",
+      "本科学 历",
+      "tr ust",
+      "t rust",
+      "cess ing",
+      "ces sing",
+      "c essing",
+      "不同类型的 客房",
+      "▁at ypical",
+      "▁long itude",
+      "▁ longitude",
+      "▁path ologic",
+      "nm sg",
+      "nms g",
+      "n msg",
+      "um ped",
+      "ump ed",
+      "re ating",
+      "reat ing",
+      "rea ting",
+      "nSw edish",
+      "▁Card inals",
+      "▁Cardinal s",
+      "▁Cardi nals",
+      "但 又",
+      "向 来",
+      "刑事 案件",
+      "Ch ris",
+      "▁re plica",
+      "▁replic a",
+      "显微 镜",
+      "巴基 斯坦",
+      "song writer",
+      "机 和",
+      "nT O",
+      "n TO",
+      "医疗 费",
+      "▁ 卫生间",
+      "▁Brig ade",
+      "A z",
+      "巡 查",
+      "def initions",
+      "definition s",
+      "亲 近",
+      "非 物质",
+      "历 时",
+      "限 量",
+      "Mi x",
+      "M ix",
+      "Qu al",
+      "Q ual",
+      "的一 系列",
+      "的 一系列",
+      "▁puzzle s",
+      "▁puzz les",
+      "▁L ighting",
+      "▁Light ing",
+      "omorph isms",
+      "omorphism s",
+      "EM P",
+      "E MP",
+      "当事人 陈述",
+      "SE ARCH",
+      "▁t high",
+      "▁th igh",
+      "▁thi gh",
+      "余 姚",
+      "遗 迹",
+      "其 主要",
+      "▁O EM",
+      "an imate",
+      "ani mate",
+      "anim ate",
+      "cap ital",
+      "▁cul inary",
+      "▁pred icate",
+      "▁ predicate",
+      "可 信",
+      "IV ATE",
+      "ov anni",
+      "ovan ni",
+      "到 我",
+      "直 肠",
+      "是不 能",
+      "是 不能",
+      "▁A BS",
+      "▁AB S",
+      "▁ ABS",
+      "▁rel ational",
+      "▁relation al",
+      "将 有",
+      "细 则",
+      "长 生",
+      "▁p eg",
+      "▁pe g",
+      "▁ peg",
+      "ri tes",
+      "rit es",
+      "rite s",
+      "r ites",
+      "▁Ar duino",
+      "▁c urated",
+      "▁cur ated",
+      "▁cu rated",
+      "▁g esture",
+      "▁gest ure",
+      "mb ic",
+      "m bic",
+      "▁\" /\"",
+      "▁\"/ \"",
+      "▁ \"/\"",
+      "▁T ir",
+      "▁Ti r",
+      "▁Ill ustr",
+      "forget table",
+      "图 画",
+      "No v",
+      "N ov",
+      "在 工作",
+      "Adv anced",
+      "otox icity",
+      "otoxic ity",
+      "o toxicity",
+      "▁Custom ers",
+      "▁Customer s",
+      "▁ 山",
+      "向上 的",
+      "向 上的",
+      "▁Qu adr",
+      "▁Quad r",
+      "▁anx ious",
+      "李 白",
+      "为一 个",
+      "为 一个",
+      "▁a ws",
+      "▁aw s",
+      "▁ aws",
+      "ug gage",
+      "ugg age",
+      "内 饰",
+      "无 私",
+      "确 切",
+      "nh ere",
+      "nhe re",
+      "n here",
+      "uc ceed",
+      "ucc eed",
+      "▁pat ents",
+      "▁patent s",
+      "▁re sembles",
+      "▁resemble s",
+      "周长 是",
+      "周 长是",
+      "nM ike",
+      "nMi ke",
+      "n Mike",
+      "▁Er nest",
+      "▁Spot ify",
+      "▁lect urer",
+      "▁lecture r",
+      "I Q",
+      "中 队",
+      "双 语",
+      "有 你",
+      "▁A ce",
+      "▁Ac e",
+      "当地 时间",
+      "▁grat itude",
+      "植 树",
+      "Ma h",
+      "M ah",
+      "ri ad",
+      "ria d",
+      "r iad",
+      "付费 政策",
+      "▁ant ico",
+      "▁anti co",
+      "▁antic o",
+      "add Class",
+      "▁aneurys m",
+      "▁交通 信息",
+      "劳动人事 争议",
+      "▁osteopor osis",
+      "失 调",
+      "随 地",
+      "an ing",
+      "ani ng",
+      "a ning",
+      "▁New port",
+      "▁prost agland",
+      "七 个",
+      "同样 是",
+      "So ng",
+      "Son g",
+      "S ong",
+      "刑 一",
+      "穴 位",
+      "形势 下",
+      "附带 不同的",
+      "▁De signed",
+      "▁Des igned",
+      "▁Design ed",
+      "附带不同的 取消预订",
+      "不同类型的客房 附带不同的取消预订",
+      "假 命题",
+      "的不 是",
+      "的 不是",
+      "劣 势",
+      "悬 疑",
+      "桥 镇",
+      "▁de hyd",
+      "▁mult in",
+      "▁multi n",
+      "▁mul tin",
+      "和预先 付费政策",
+      "便 可",
+      "农 牧",
+      "对 任意",
+      "▁bl ades",
+      "▁blade s",
+      "▁dec orate",
+      "▁decor ate",
+      "口 岸",
+      "yr us",
+      "y rus",
+      "止 血",
+      "福 特",
+      "sh an",
+      "sha n",
+      "s han",
+      "ik u",
+      "i ku",
+      "力 资源",
+      "or tal",
+      "ort al",
+      "orta l",
+      "▁所 获荣誉",
+      "nBrazil ian",
+      "三 代",
+      "睡 前",
+      "AM I",
+      "A MI",
+      "`: \\",
+      "` :\\",
+      "视为 放弃",
+      "▁rac ist",
+      "▁adv iser",
+      "▁advis er",
+      "▁advise r",
+      "nArt icles",
+      "nArticle s",
+      "n Articles",
+      "为 宜",
+      "平 平",
+      "此 文",
+      "▁X I",
+      "nJ oh",
+      "nJo h",
+      "n Joh",
+      "de ath",
+      "dea th",
+      "信用卡 纠纷一案",
+      "造 纸",
+      "▁H els",
+      "▁He ls",
+      "▁Hel s",
+      "wit hin",
+      "with in",
+      "wi thin",
+      "可 求",
+      "过 错",
+      "▁C OR",
+      "▁CO R",
+      "▁ COR",
+      "▁contin ental",
+      "▁continent al",
+      "岁 了",
+      "等 职",
+      "▁A U",
+      "▁ AU",
+      "▁江 苏",
+      "▁ 江苏",
+      "公共 服务",
+      "nG ive",
+      "n Give",
+      "Null Or",
+      "相 伴",
+      "TE RN",
+      "TER N",
+      "T ERN",
+      "以上 资料",
+      "▁me nstr",
+      "▁men str",
+      "▁mens tr",
+      "j c",
+      "ie le",
+      "iel e",
+      "i ele",
+      "▁str iker",
+      "▁stri ker",
+      "▁strike r",
+      "▁ec centric",
+      "▁ecc entric",
+      "打 好",
+      "众多 的",
+      "众 多的",
+      "有所 帮助",
+      "▁m kdir",
+      "▁mk dir",
+      "▁ mkdir",
+      "j m",
+      "军 团",
+      "▁d ol",
+      "▁do l",
+      "建筑 装饰",
+      "ol vers",
+      "olve rs",
+      "olver s",
+      "olv ers",
+      "yt ical",
+      "ytic al",
+      "y tical",
+      "▁Com edy",
+      "▁Come dy",
+      "▁B I",
+      "▁ BI",
+      "▁d ic",
+      "▁di c",
+      "▁ dic",
+      "损害 赔偿",
+      "款 人民币",
+      "▁bon uses",
+      "▁bonus es",
+      "▁assert ion",
+      "轻 度",
+      "辩 证",
+      "▁I Q",
+      "▁ IQ",
+      "▁account able",
+      "pro x",
+      "pr ox",
+      "p rox",
+      "国 度",
+      "▁方 法",
+      "▁ 方法",
+      "的 保护",
+      "周围 景观",
+      "▁a met",
+      "▁am et",
+      "▁ amet",
+      "Man age",
+      "Manag e",
+      "▁cl amp",
+      "▁T utorial",
+      "set Attribute",
+      "而 行",
+      "赢 了",
+      "罪犯 陈",
+      "▁E CG",
+      "▁EC G",
+      "深 层",
+      "清 市",
+      "演 化",
+      "BO X",
+      "B OX",
+      "陌生 的",
+      "陌 生的",
+      "农业 人口",
+      "reg ions",
+      "region s",
+      "脱 水",
+      "腌 制",
+      "nB ooks",
+      "nBook s",
+      "nBo oks",
+      "n Books",
+      "▁moist ur",
+      "办公 楼",
+      "nd ev",
+      "nde v",
+      "n dev",
+      "pe ated",
+      "peat ed",
+      "成功 了",
+      "▁o le",
+      "▁ol e",
+      "▁ ole",
+      "▁s pp",
+      "▁sp p",
+      "▁ph yt",
+      "▁phy t",
+      "▁all ograft",
+      "> *",
+      "日 夜",
+      "获 胜",
+      "ra vel",
+      "rav el",
+      "r avel",
+      "▁\\ ')\\",
+      "▁\\' )\\",
+      "▁\\') \\",
+      "▁ \\')\\",
+      "▁c rap",
+      "▁cr ap",
+      "▁cra p",
+      "二〇一六年 五月",
+      "庄 园",
+      "暧 昧",
+      "am on",
+      "amo n",
+      "a mon",
+      "en in",
+      "eni n",
+      "e nin",
+      "ár io",
+      "á rio",
+      "▁dis like",
+      "▁compet it",
+      "上 岸",
+      "form atted",
+      "format ted",
+      "＊ ＊",
+      "一站 式",
+      "基督 教",
+      "扬州 市",
+      "扬 州市",
+      "qu arters",
+      "quarter s",
+      "▁para sites",
+      "▁paras ites",
+      "▁parasite s",
+      "▁ 强",
+      "配 上",
+      "的 战略",
+      "os yl",
+      "td ouble",
+      "t double",
+      "▁cl utter",
+      "下 巴",
+      "儒 家",
+      "抗 辩",
+      "方 块",
+      "临床 上",
+      "临 床上",
+      "AT ES",
+      "ATE S",
+      "A TES",
+      "math sf",
+      "re ceiver",
+      "receive r",
+      "▁s ynchronous",
+      "▁synchron ous",
+      "▁e sp",
+      "▁es p",
+      "▁ esp",
+      "游戏 加载",
+      "pace d",
+      "pa ced",
+      "pac ed",
+      "p aced",
+      "▁st amps",
+      "▁stamp s",
+      "▁observ able",
+      "听 听",
+      "纵横 中文网",
+      "met ics",
+      "metic s",
+      "m etics",
+      "▁J esse",
+      "▁Jes se",
+      "▁Jess e",
+      "▁L ogan",
+      "▁Log an",
+      "▁Lo gan",
+      "▁stead ily",
+      "各 组",
+      "几个 人",
+      "几 个人",
+      "▁I nto",
+      "▁In to",
+      "▁Int o",
+      "▁ Into",
+      "isc opal",
+      "isco pal",
+      "▁sub class",
+      "▁phosph orus",
+      "果 汁",
+      "枣 庄",
+      "限 额",
+      "马 桶",
+      "此题 考查了",
+      "此题考查 了",
+      "de legate",
+      "▁sle eves",
+      "▁sleeve s",
+      "加 了",
+      "的 动作",
+      "nV er",
+      "n Ver",
+      "改造 期间",
+      "ware s",
+      "war es",
+      "wa res",
+      "w ares",
+      "▁tr ainers",
+      "▁train ers",
+      "▁trainer s",
+      "▁hepat ocytes",
+      "草 坪",
+      "com pet",
+      "comp et",
+      "▁Gib son",
+      "是 单",
+      "极 值",
+      "在 今年",
+      "▁S ob",
+      "▁So b",
+      "br ush",
+      "al ignment",
+      "align ment",
+      "▁Armen ian",
+      "▁Armenia n",
+      "送达之日起 发生法律效力",
+      "供 暖",
+      "变 色",
+      "古 镇",
+      "yy yy",
+      "▁en dings",
+      "▁end ings",
+      "▁ending s",
+      "▁laugh ing",
+      "觉得 很",
+      "觉 得很",
+      "▁real izing",
+      "▁privile ges",
+      "▁privilege s",
+      "▁ 拉",
+      "住房 和",
+      "He ro",
+      "Her o",
+      "H ero",
+      "al amus",
+      "ala mus",
+      "alam us",
+      "▁he avier",
+      "▁heav ier",
+      "nAct ually",
+      "特 大",
+      "qu eries",
+      "que ries",
+      "quer ies",
+      "▁bubble s",
+      "▁bub bles",
+      "respon ses",
+      "response s",
+      "respons es",
+      "▁Domin ican",
+      "ph en",
+      "p hen",
+      "耕地 面积",
+      "耕 地面积",
+      "nuse r",
+      "nu ser",
+      "nus er",
+      "n user",
+      "was her",
+      "wash er",
+      "▁Har old",
+      "图 为",
+      "得 住",
+      "蔷 薇",
+      "}} ,\\",
+      "}}, \\",
+      "} },\\",
+      "但 它",
+      "什么 叫",
+      "接受 的",
+      "▁C OM",
+      "▁CO M",
+      "▁ COM",
+      "▁S ib",
+      "▁Si b",
+      "▁Z ip",
+      "▁ Zip",
+      "返还 借款",
+      "▁St rat",
+      "▁Str at",
+      "▁Stra t",
+      "▁s pins",
+      "▁sp ins",
+      "▁spin s",
+      "ounc ing",
+      "oun cing",
+      "▁tr agic",
+      "▁tra gic",
+      "▁illum ination",
+      "读 卡",
+      "vt k",
+      "v tk",
+      "聚合 物",
+      "聚 合物",
+      "Bl ocks",
+      "Block s",
+      "B locks",
+      "▁r ails",
+      "▁ra ils",
+      "▁rail s",
+      "▁ rails",
+      "珠 江",
+      "▁y e",
+      "▁ ye",
+      "▁cirrh osis",
+      "友 们",
+      "有 以下",
+      "ar rays",
+      "arr ays",
+      "array s",
+      "arra ys",
+      "取 向",
+      "流 星",
+      "nd b",
+      "n db",
+      "▁over ly",
+      "▁overl y",
+      "▁ov erly",
+      "text less",
+      "▁sup porter",
+      "▁support er",
+      "NO TE",
+      "NOT E",
+      "N OTE",
+      "按 年利率",
+      "Ind ia",
+      "▁I deal",
+      "▁Id eal",
+      "▁Ide al",
+      "▁Idea l",
+      "▁fil ament",
+      "▁di astolic",
+      "倾向 于",
+      "分数 线",
+      "nW eb",
+      "nWe b",
+      "n Web",
+      "期 盼",
+      "母 婴",
+      "自 习",
+      "角 和",
+      "▁cha pel",
+      "▁chap el",
+      "▁trunc ated",
+      "} %",
+      "ht aking",
+      "h taking",
+      "二〇一五年 六月",
+      "▁Cit izens",
+      "▁Citizen s",
+      "å r",
+      "元 钱",
+      "投 身",
+      "起 飞",
+      "打造 的",
+      "十一 个月",
+      "十 一个月",
+      "▁n asty",
+      "▁na sty",
+      "▁nas ty",
+      "rec ipes",
+      "recip es",
+      "recipe s",
+      "▁py ramid",
+      "▁pyram id",
+      "ag in",
+      "agi n",
+      "a gin",
+      "or el",
+      "ore l",
+      "o rel",
+      "乌鲁木齐 市",
+      "▁Inter mediate",
+      "四 位",
+      "批 次",
+      "证据 材料",
+      "分析 法",
+      "圆柱 的",
+      "id is",
+      "idi s",
+      "i dis",
+      "共 和",
+      "IE num",
+      "I Enum",
+      "text greater",
+      "J C",
+      "路 易",
+      "No m",
+      "N om",
+      "▁H ex",
+      "▁He x",
+      "▁ Hex",
+      "We ather",
+      "▁qu artz",
+      "▁quart z",
+      "▁front end",
+      "▁ frontend",
+      "▁sen ators",
+      "▁senator s",
+      "缺 口",
+      "列 方程",
+      "是一 场",
+      "是 一场",
+      "ch ester",
+      "ches ter",
+      "che ster",
+      "c hester",
+      "▁ker atin",
+      "▁kerat in",
+      "co ins",
+      "coin s",
+      "hu rst",
+      "h urst",
+      "的最佳 选择",
+      "考 量",
+      "酒后 驾驶",
+      "▁M int",
+      "▁Min t",
+      "▁Mi nt",
+      "前 任",
+      "综合 素质",
+      "B EGIN",
+      "Ag greg",
+      "Agg reg",
+      "▁sc enic",
+      "▁scen ic",
+      "▁craft sm",
+      "▁crafts m",
+      "真 假",
+      "的一 位",
+      "的 一位",
+      "奥林匹 克",
+      "▁p uls",
+      "▁pul s",
+      "▁pu ls",
+      "ar ying",
+      "ary ing",
+      "a rying",
+      "▁furn ish",
+      "▁depriv ation",
+      "售 出",
+      "hed ron",
+      "龙 岗",
+      "免费的 停车服务",
+      "accept able",
+      "▁in patient",
+      "回 路",
+      "ON Y",
+      "O NY",
+      "Wi l",
+      "W il",
+      "In sp",
+      "Ins p",
+      "I nsp",
+      "artifact Id",
+      "▁therap ists",
+      "▁therapist s",
+      "▁intrac ranial",
+      "▁C DC",
+      "▁CD C",
+      "▁ CDC",
+      "▁O SI",
+      "▁OS I",
+      "执行 和解协议",
+      "▁tack les",
+      "▁tackle s",
+      "▁co median",
+      "▁comed ian",
+      "▁drop down",
+      "▁ dropdown",
+      "▁freel ance",
+      "▁Al gorithms",
+      "▁Algorithm s",
+      "▁dis solution",
+      "▁diss olution",
+      "▁dr astically",
+      "▁C W",
+      "▁ CW",
+      "▁f lor",
+      "▁fl or",
+      "▁flo r",
+      "委员会 主任",
+      "ce lain",
+      "cel ain",
+      "▁ 孟",
+      "征 信",
+      "就能 够",
+      "就 能够",
+      "Bit s",
+      "Bi ts",
+      "B its",
+      "▁Bang kok",
+      "Det ection",
+      "▁… \\",
+      "▁ …\\",
+      "▁第 五十",
+      "▁ 第五十",
+      "▁Py Qt",
+      "▁b lob",
+      "▁bl ob",
+      "▁blo b",
+      "▁ blob",
+      "▁jer sey",
+      "▁at rophy",
+      "A ustralia",
+      "▁end angered",
+      "定 额",
+      "内部 的",
+      "内 部的",
+      "▁sa nit",
+      "▁san it",
+      "Americ a",
+      "nUn less",
+      "n Unless",
+      "▁Leg ion",
+      "▁inert ia",
+      "▁in vention",
+      "▁inv ention",
+      "▁invent ion",
+      "tp ar",
+      "t par",
+      "▁t xt",
+      "▁tx t",
+      "▁ txt",
+      "▁amel ior",
+      "à n",
+      "电影 节",
+      "澄 清",
+      "萨 尔",
+      "独 一无",
+      "key board",
+      "▁parse Int",
+      "▁ parseInt",
+      "不 息",
+      "菱 形",
+      "▁通 过",
+      "▁ 通过",
+      "▁R SA",
+      "▁RS A",
+      "▁Mal ta",
+      "▁Malt a",
+      "▁Sc roll",
+      "▁Scr oll",
+      "▁ Scroll",
+      "Typ ography",
+      "hem atically",
+      "hemat ically",
+      "hema tically",
+      "hematical ly",
+      "hematic ally",
+      "的 去",
+      "高 出",
+      "▁r ic",
+      "▁ri c",
+      "▁ ric",
+      "▁archae ological",
+      "收 治",
+      "而对 于",
+      "而 对于",
+      "AT IC",
+      "ili near",
+      "iline ar",
+      "ilin ear",
+      "i linear",
+      "▁Elect ro",
+      "ativ istic",
+      "不 平",
+      "un gs",
+      "ung s",
+      "▁E xit",
+      "▁Ex it",
+      "▁ Exit",
+      "▁with stand",
+      "í t",
+      "石 榴",
+      "▁c oral",
+      "▁co ral",
+      "▁cor al",
+      "▁j okes",
+      "▁jo kes",
+      "▁joke s",
+      "灌 注",
+      "os ta",
+      "ost a",
+      "o sta",
+      "相关 负责人",
+      "▁f ines",
+      "▁fin es",
+      "▁fine s",
+      "▁fi nes",
+      "身份证 复印件",
+      "▁B inding",
+      "▁Bin ding",
+      "▁Bind ing",
+      "▁ Binding",
+      "好好 的",
+      "好 好的",
+      "▁time less",
+      "▁tim eless",
+      "▁crim inals",
+      "▁criminal s",
+      "上 层",
+      "变 质",
+      "▁ 浴室",
+      "按 规定",
+      "Co in",
+      "C oin",
+      "\\' )}}",
+      "\\') }}",
+      "tensor flow",
+      "前 台",
+      "金额 为",
+      "金 额为",
+      "op ez",
+      "ope z",
+      "o pez",
+      "nCol lege",
+      "同 济",
+      "沙 特",
+      "▁J i",
+      "wi fe",
+      "w ife",
+      "▁K och",
+      "▁Ko ch",
+      "▁encompass es",
+      "惠 民",
+      "黑 人",
+      "ob o",
+      "o bo",
+      "同时 在",
+      "是否 能",
+      "▁D ogs",
+      "▁Do gs",
+      "▁Dog s",
+      "嘿 嘿",
+      "意识 和",
+      "is su",
+      "iss u",
+      "▁Cl osed",
+      "▁Close d",
+      "▁Clo sed",
+      "▁ Closed",
+      "了一 句",
+      "了 一句",
+      "调研 员",
+      "摇 头",
+      "ok i",
+      "o ki",
+      "▁t hi",
+      "▁th i",
+      "▁ thi",
+      "an ian",
+      "ani an",
+      "ania n",
+      "▁A tom",
+      "▁At om",
+      "▁ Atom",
+      "ten ham",
+      "mat plotlib",
+      "上 百",
+      "唐 朝",
+      "社会 学",
+      "al in",
+      "ali n",
+      "a lin",
+      "长期 以来",
+      "Ar row",
+      "Arr ow",
+      "有 管辖权的",
+      "des criptor",
+      "思 议",
+      "的 模式",
+      "Log s",
+      "Lo gs",
+      "L ogs",
+      "Byte Array",
+      "少 爷",
+      "真的 太",
+      "这种 人",
+      "驻马 店",
+      "▁se ptic",
+      "▁sept ic",
+      "▁sep tic",
+      "震 动",
+      "首 相",
+      "即便 是",
+      "即 便是",
+      "▁T roy",
+      "▁Tr oy",
+      "▁Tro y",
+      "Pro ducer",
+      "Prod ucer",
+      "Produ cer",
+      "▁str etched",
+      "▁stret ched",
+      "▁stretch ed",
+      "▁App arently",
+      "▁ag gression",
+      "▁aggress ion",
+      "插 座",
+      "沸 腾",
+      "装 潢",
+      "但我 们",
+      "但 我们",
+      "▁Ex cellent",
+      "▁Excell ent",
+      "有 三",
+      "穷 尽",
+      "op ping",
+      "opp ing",
+      "o pping",
+      "大 成",
+      "委 书记",
+      "vert ices",
+      "像 我",
+      "海 涛",
+      "ut ilities",
+      "util ities",
+      "为 正",
+      "卫 浴",
+      "的 再审申请",
+      "der abad",
+      "bb ox",
+      "b box",
+      "od on",
+      "odo n",
+      "o don",
+      "Re ceive",
+      "合 规",
+      "op tic",
+      "opt ic",
+      "o ptic",
+      "▁bo arding",
+      "▁board ing",
+      "▁ boarding",
+      "于 法",
+      "好 久",
+      "ес т",
+      "е ст",
+      "▁L ap",
+      "▁La p",
+      "▁reduct ase",
+      "nPolit icians",
+      "天 鹅",
+      "斯 顿",
+      "超声 波",
+      "▁M ile",
+      "▁Mil e",
+      "▁Mi le",
+      "ath lon",
+      "▁S ector",
+      "▁Se ctor",
+      "▁Sec tor",
+      "Dis abled",
+      "Disable d",
+      "▁elic ited",
+      "▁elicit ed",
+      "aphyl ococcus",
+      "与 时",
+      "沐 浴",
+      "▁ref inement",
+      "▁refin ement",
+      "▁refine ment",
+      "▁non etheless",
+      "▁none theless",
+      "六 大",
+      "指 点",
+      "car ousel",
+      "借 钱",
+      "截 面",
+      "尚欠 原告",
+      "尚 欠原告",
+      "▁E GFR",
+      "▁EGF R",
+      "▁EG FR",
+      "ol ding",
+      "old ing",
+      "▁reg imes",
+      "▁regime s",
+      "▁us ability",
+      "真的 有",
+      "真 的有",
+      "▁n ovo",
+      "▁no vo",
+      "▁nov o",
+      "▁mot ifs",
+      "▁motif s",
+      "▁Mus ical",
+      "▁Music al",
+      "不 改",
+      "恋 人",
+      "ould er",
+      "oul der",
+      "▁economic ally",
+      "▁econom ically",
+      "▁economical ly",
+      "脸 部",
+      "的 趋势",
+      "▁C rypto",
+      "▁Crypt o",
+      "▁Cry pto",
+      "▁ Crypto",
+      "沧 海",
+      "白 雪",
+      "制定 的",
+      "制 定的",
+      "▁C hin",
+      "▁Ch in",
+      "▁Chi n",
+      "List View",
+      "phys ical",
+      "再 算",
+      "异 味",
+      "水 性",
+      "Sa t",
+      "S at",
+      "▁T rou",
+      "▁Tr ou",
+      "▁Tro u",
+      "▁ag gress",
+      "▁in verted",
+      "▁invert ed",
+      "佩 服",
+      "而 知",
+      "指挥 部",
+      "▁R ig",
+      "▁Ri g",
+      "直接 利用",
+      "nAv ailable",
+      "n Available",
+      "精 髓",
+      "草 木",
+      "三位 数",
+      "三 位数",
+      "紧 接着",
+      "▁liqu ids",
+      "▁liquid s",
+      "Spec ification",
+      "Specific ation",
+      "想 了",
+      "请 注意",
+      "达 不到",
+      "Fa ll",
+      "F all",
+      "Pl ugins",
+      "Plugin s",
+      "有 异议",
+      "比亚 迪",
+      "▁H EL",
+      "▁HE L",
+      "Rel ative",
+      "es Module",
+      "produ ced",
+      "prod uced",
+      "p roduced",
+      "主要 由",
+      "ar ga",
+      "arg a",
+      "<> \\",
+      "< >\\",
+      "aw ays",
+      "awa ys",
+      "away s",
+      "a ways",
+      "nP rint",
+      "nPr int",
+      "n Print",
+      "………… …………",
+      "但 从",
+      "tg l",
+      "t gl",
+      "file r",
+      "fil er",
+      "fi ler",
+      "f iler",
+      "op rop",
+      "opro p",
+      "o prop",
+      "▁phylogen etic",
+      "理 化",
+      "▁Rec ruit",
+      "老 头",
+      "脱 颖",
+      "CP P",
+      "C PP",
+      "ca ke",
+      "c ake",
+      "的方程 为",
+      "的 方程为",
+      "ce mic",
+      "c emic",
+      "▁get Id",
+      "▁ getId",
+      "Bl ank",
+      "▁s her",
+      "▁sh er",
+      "▁she r",
+      "an other",
+      "ano ther",
+      "a nother",
+      "▁mind set",
+      "▁minds et",
+      "p v",
+      "Version UID",
+      "▁polar ized",
+      "华 山",
+      "坚 固",
+      "多 篇",
+      "富 豪",
+      "花 香",
+      "▁rod ent",
+      "▁rode nt",
+      "▁trigon ometric",
+      "P Q",
+      "w u",
+      "с и",
+      "fa de",
+      "f ade",
+      "二〇一七年 七月",
+      "二〇一七年七 月",
+      "当 他",
+      "折 射",
+      "ve rn",
+      "ver n",
+      "v ern",
+      "所 需要的",
+      "mi rror",
+      "mir ror",
+      "pl icate",
+      "plic ate",
+      "plica te",
+      "▁Merc ury",
+      "▁trans gender",
+      "八 大",
+      "时 而",
+      "man ship",
+      "mans hip",
+      "m anship",
+      "▁M onica",
+      "▁Mon ica",
+      "▁Mo nica",
+      "▁inv oked",
+      "▁invoke d",
+      "冷 的",
+      "昆 山",
+      "RE M",
+      "R EM",
+      "▁G el",
+      "▁Ge l",
+      "Be cause",
+      "B ecause",
+      "Ro uting",
+      "R outing",
+      "op lasty",
+      "oplas ty",
+      "opl asty",
+      "oplast y",
+      "o plasty",
+      "▁Trans it",
+      "[ ^",
+      "写 明",
+      "双方 约定",
+      "▁C ult",
+      "▁Cu lt",
+      "▁Cul t",
+      "▁url s",
+      "▁ur ls",
+      "▁ urls",
+      "rel ations",
+      "relation s",
+      "▁shop pers",
+      "▁sub marine",
+      "做 一些",
+      "▁B es",
+      "▁Be s",
+      "▁mult itude",
+      "原 先",
+      "器 和",
+      "张 国",
+      "曙 光",
+      "ec d",
+      "e cd",
+      "ени я",
+      "▁B ean",
+      "▁Be an",
+      "▁Bea n",
+      "▁ Bean",
+      "▁An imals",
+      "▁Animal s",
+      "等腰 三角形",
+      "▁short cut",
+      "▁Reg ardless",
+      "药 剂",
+      "▁D j",
+      "▁S essions",
+      "▁Session s",
+      "▁m orphism",
+      "▁morph ism",
+      "▁ 章",
+      "声 称",
+      "逃 脱",
+      "tic ks",
+      "ti cks",
+      "tick s",
+      "t icks",
+      "▁s alon",
+      "▁sa lon",
+      "▁sal on",
+      "▁al pine",
+      "▁ alpine",
+      "▁Robert son",
+      "▁Roberts on",
+      "ol lar",
+      "oll ar",
+      "olla r",
+      "腹 痛",
+      "飞 速",
+      "AD A",
+      "A DA",
+      "ens ors",
+      "ensor s",
+      "▁r enders",
+      "▁render s",
+      "▁rend ers",
+      "▁ann ouncements",
+      "▁announce ments",
+      "▁announcement s",
+      "▁announc ements",
+      "Re n",
+      "R en",
+      "事件 的",
+      "事 件的",
+      ")* *(",
+      ")** (",
+      ") **(",
+      "书面 答辩",
+      "▁p laus",
+      "▁pl aus",
+      "spe aking",
+      "定 居",
+      "开 着",
+      "杰 伦",
+      "必要 时",
+      "sign up",
+      "sig nup",
+      "▁aff irm",
+      "▁af firm",
+      "▁C ognitive",
+      "▁forth coming",
+      "▁phys iologic",
+      "▁physi ologic",
+      "An s",
+      "A ns",
+      "也会 有",
+      "也 会有",
+      "新华 网",
+      "▁A TM",
+      "▁AT M",
+      "▁ ATM",
+      "没有 约定",
+      "▁h ect",
+      "▁he ct",
+      "▁doctor al",
+      "▁ doctoral",
+      "▁dem olished",
+      "▁demo lished",
+      "不 离",
+      "直 立",
+      "名词 审定委员会",
+      "记 作",
+      "党 总支",
+      "让 更多",
+      "nIn vest",
+      "n Invest",
+      "m V",
+      "埃 尔",
+      "}\\\\) （",
+      "同 伴",
+      "露 天",
+      "多少 个",
+      "▁st urdy",
+      "什么 事",
+      "的 工具",
+      "▁L on",
+      "▁Lo n",
+      "生活 质量",
+      "▁p lag",
+      "▁pl ag",
+      "pe ptide",
+      "acc uracy",
+      "accur acy",
+      "禁 用",
+      "\"} )\\",
+      "\" })\\",
+      "▁jew ellery",
+      "▁neuro path",
+      "▁neur opath",
+      "审 阅",
+      "甲 乙",
+      "羽 绒",
+      "▁D T",
+      "▁ DT",
+      "nH appy",
+      "n Happy",
+      "▁M arin",
+      "▁Mar in",
+      "▁Ma rin",
+      "▁Mari n",
+      "萨 斯",
+      "pg f",
+      "p gf",
+      "gly ph",
+      "▁h umble",
+      "▁hum ble",
+      "Dec laration",
+      "}` ;\\",
+      "} `;\\",
+      "▁divid end",
+      "▁divide nd",
+      "人 造",
+      "巧 合",
+      "]) );\\",
+      "])) ;\\",
+      "] ));\\",
+      "▁P ret",
+      "▁Pr et",
+      "▁Pre t",
+      "▁职业 生涯",
+      "▁ 职业生涯",
+      "og enes",
+      "ogen es",
+      "ogene s",
+      "oge nes",
+      "o genes",
+      "adv anced",
+      "▁intern ship",
+      "▁intro ductory",
+      "▁Stephan ie",
+      "▁garden ing",
+      "▁gard ening",
+      "开 店",
+      "is Valid",
+      "▁ge ography",
+      "▁rein force",
+      "星 际",
+      "见 于",
+      "系列 的",
+      "▁ful filling",
+      "▁fulfill ing",
+      "▁fulfil ling",
+      "q r",
+      "城 管",
+      "▁mix er",
+      "▁cur tain",
+      "▁curt ain",
+      "▁op acity",
+      "▁ opacity",
+      "je e",
+      "j ee",
+      "▁C ry",
+      "▁Cr y",
+      "直接 影响",
+      "▁spectro scopic",
+      "龙 的",
+      "Ad visor",
+      "Adv isor",
+      "▁bes poke",
+      "▁grad ual",
+      "▁gradu al",
+      "回 调",
+      "大 三",
+      "春 日",
+      "nM ag",
+      "nMa g",
+      "n Mag",
+      "变得 更加",
+      "变得更 加",
+      "en cers",
+      "ence rs",
+      "enc ers",
+      "encer s",
+      "党 内",
+      "达 尔",
+      "你 喜欢",
+      "▁G reens",
+      "▁Green s",
+      "▁Gree ns",
+      "▁Gre ens",
+      "▁hack ers",
+      "▁hacker s",
+      "▁According ly",
+      "张 艺",
+      "hist oric",
+      "h istoric",
+      "▁v entral",
+      "▁vent ral",
+      "凋 亡",
+      "\"] [",
+      "\" ][",
+      "也都 是",
+      "也 都是",
+      "艺术 品",
+      "▁Mc D",
+      "可以 随时",
+      "ro let",
+      "rol et",
+      "role t",
+      "Pub lisher",
+      "Publish er",
+      "上 台",
+      "思 明",
+      "▁G U",
+      "▁ GU",
+      "la us",
+      "l aus",
+      "▁M UST",
+      "▁MU ST",
+      "限制 高消费",
+      "▁Ang ela",
+      "▁Angel a",
+      "天 台",
+      "干 警",
+      "朱 某",
+      "泌 尿",
+      "闪 电",
+      "▁Sl ack",
+      "▁pump kin",
+      "nDist rict",
+      "nDi strict",
+      "打 扫",
+      "浦 区",
+      "Bu g",
+      "B ug",
+      "np s",
+      "n ps",
+      "▁t d",
+      "▁ td",
+      "}) $$",
+      "})$ $",
+      "} )$$",
+      "A verage",
+      "▁co arse",
+      "nA gain",
+      "nAg ain",
+      "▁Wal mart",
+      "勘 探",
+      "环 氧",
+      "的 吧",
+      ">\" ;\\",
+      "> \";\\",
+      "▁sy rup",
+      "夫妻感情 确已破裂",
+      "▁mean ings",
+      "▁meaning s",
+      "▁philosoph ical",
+      "永 康",
+      "gi o",
+      "g io",
+      "说 白了",
+      "de vel",
+      "dev el",
+      "社会抚养费 征收",
+      "就是 把",
+      "就 是把",
+      "的 法定",
+      "tif f",
+      "ti ff",
+      "t iff",
+      "▁c urrencies",
+      "▁curr encies",
+      "老 了",
+      "谱 曲",
+      "sa p",
+      "s ap",
+      "农业 部",
+      "电影 院",
+      "电 影院",
+      "等 材料",
+      "▁g ait",
+      "▁ga it",
+      "▁Break fast",
+      "▁landsc aping",
+      "交 所",
+      "看 重",
+      "拖拉 机",
+      "柳州 市",
+      "柳 州市",
+      "▁Jour ney",
+      "▁inter preter",
+      "▁interpret er",
+      "量 身",
+      "▁注 意",
+      "▁ 注意",
+      "▁P ly",
+      "▁Pl y",
+      "目前 为止",
+      "▁a cre",
+      "▁ac re",
+      "▁ acre",
+      "▁ 参考资料",
+      "▁qu ark",
+      "▁ quark",
+      "▁happ ier",
+      "救 护",
+      "习惯 了",
+      "nD en",
+      "nDe n",
+      "n Den",
+      "▁Pro gressive",
+      "▁Progress ive",
+      "给付 义务",
+      "▁L ect",
+      "▁Le ct",
+      "▁opt ics",
+      "▁optic s",
+      "▁fest ive",
+      "▁aut onomy",
+      "▁fore most",
+      "▁G ospel",
+      "▁Gos pel",
+      "▁t rophy",
+      "▁tro phy",
+      "▁troph y",
+      "▁trop hy",
+      "▁harvest ed",
+      "▁ 美",
+      "与 他",
+      "已 达",
+      "ah ren",
+      "ahr en",
+      "▁bread th",
+      "O V",
+      "Mu s",
+      "M us",
+      "保持 着",
+      "ep id",
+      "e pid",
+      "▁chem ically",
+      "▁chemical ly",
+      "头 顶",
+      "快 三",
+      "县 国土资源局",
+      "▁Mod ules",
+      "▁Module s",
+      "▁ Modules",
+      "可以 为",
+      "可 以为",
+      "▁c rews",
+      "▁cre ws",
+      "▁cr ews",
+      "▁crew s",
+      "奇 特",
+      "请 选择",
+      "Dr ag",
+      "D rag",
+      "uc es",
+      "uce s",
+      "u ces",
+      "nowled g",
+      "th eight",
+      "the ight",
+      "t height",
+      "▁curt ains",
+      "▁curtain s",
+      "工 件",
+      "Be d",
+      "B ed",
+      "▁b c",
+      "▁ bc",
+      "统计 图",
+      "nL in",
+      "n Lin",
+      "▁d angers",
+      "▁danger s",
+      "▁dan gers",
+      "湖 人",
+      "在 学习",
+      "▁sand w",
+      "▁san dw",
+      "助 推",
+      "苗 木",
+      "以 提高",
+      "满足 了",
+      "Li ttle",
+      "L ittle",
+      "▁tem ples",
+      "▁temp les",
+      "▁temple s",
+      "后 会",
+      "▁E co",
+      "▁Ec o",
+      "__ ))\\",
+      "__) )\\",
+      "METHO D",
+      "M ETHOD",
+      "CC C",
+      "C CC",
+      "从而 使",
+      "现代 化的",
+      "现代化 的",
+      "▁Where as",
+      "飞 船",
+      "▁E lli",
+      "▁El li",
+      "▁Ell i",
+      "▁W alking",
+      "▁Wal king",
+      "▁Walk ing",
+      "十几 年",
+      "十 几年",
+      "女主 角",
+      "女 主角",
+      "me al",
+      "m eal",
+      "文物 保护",
+      "▁demonstr ations",
+      "▁demonstration s",
+      "▁ ф",
+      "契 合",
+      "▁P OL",
+      "▁PO L",
+      "▁歌曲 语言",
+      "▁C oming",
+      "▁Com ing",
+      "▁Co ming",
+      "▁Dav ies",
+      "▁fore nsic",
+      "现 今",
+      "脚 踏",
+      "与 方法",
+      "无 统计学意义",
+      "▁Sp ider",
+      "▁ Spider",
+      "het amine",
+      "▁Re ality",
+      "▁Real ity",
+      "▁isot ope",
+      "罚金 限",
+      "Fa ke",
+      "F ake",
+      "▁l ibr",
+      "▁li br",
+      "▁lib r",
+      "SP I",
+      "S PI",
+      "分子 量",
+      "nEnd emic",
+      "▁poison ing",
+      "▁organ izers",
+      "▁organize rs",
+      "▁organizer s",
+      "主 的",
+      "地 为",
+      "极 高",
+      "▁ban ana",
+      "不 复",
+      "纳 斯",
+      "▁pal ace",
+      "▁se afood",
+      "▁sea food",
+      "纹 理",
+      "并 同意",
+      "我看 到",
+      "我 看到",
+      "▁I CU",
+      "▁IC U",
+      "▁ ICU",
+      "et ting",
+      "ett ing",
+      "etti ng",
+      "▁Sign ature",
+      "▁ Signature",
+      "祖 先",
+      "钢 笔",
+      "Mer ge",
+      "▁w asted",
+      "▁was ted",
+      "▁waste d",
+      "▁wa sted",
+      "▁wast ed",
+      "▁de coding",
+      "▁dec oding",
+      "三 边",
+      "宫 廷",
+      "况 且",
+      "▁ 普通",
+      "积累 了",
+      "积 累了",
+      "ex tern",
+      "ext ern",
+      "exter n",
+      "nInter esting",
+      "的 呢",
+      "▁A W",
+      "▁ AW",
+      "▁A ck",
+      "▁Ac k",
+      "▁ Ack",
+      "▁F uel",
+      "▁Fu el",
+      "▁v egg",
+      "▁ve gg",
+      "veh icle",
+      "v ehicle",
+      "▁sk ating",
+      "▁mutagen esis",
+      "再次 申请",
+      "▁d atas",
+      "▁data s",
+      "▁dat as",
+      "▁da tas",
+      "▁ datas",
+      "install er",
+      "上 路",
+      "le ague",
+      "lea gue",
+      "▁lact ate",
+      "▁percent ages",
+      "▁percentage s",
+      "适用 的",
+      "适 用的",
+      "团 员",
+      "既 能",
+      "全村 有",
+      "▁neckl ace",
+      "▁spl icing",
+      "▁splic ing",
+      "下面 就",
+      "大 有",
+      "一生 的",
+      "一 生的",
+      "nS port",
+      "nSp ort",
+      "n Sport",
+      "na fter",
+      "n after",
+      "▁H erald",
+      "▁Her ald",
+      "で す",
+      "et i",
+      "e ti",
+      "ve e",
+      "v ee",
+      "券 商",
+      "的 量",
+      "香 菜",
+      "Rec ipe",
+      "▁V ince",
+      "▁Vin ce",
+      "▁Vi nce",
+      "▁Vinc e",
+      "民 进",
+      "ST E",
+      "S TE",
+      "nt ext",
+      "nte xt",
+      "n text",
+      "▁D eck",
+      "▁De ck",
+      "▁Dec k",
+      "Make r",
+      "Ma ker",
+      "M aker",
+      "▁w asting",
+      "▁was ting",
+      "▁wast ing",
+      "交 谈",
+      "家 家",
+      "用的 是",
+      "用 的是",
+      "ogen icity",
+      "ogenic ity",
+      "响 起",
+      "text it",
+      "tex tit",
+      "被执行人 应当",
+      "▁sen ator",
+      "▁im balance",
+      "v v",
+      "会议 上",
+      "诸葛 亮",
+      "nS al",
+      "n Sal",
+      "实施 方案",
+      "Ra ting",
+      "R ating",
+      "harm onic",
+      "▁mod ulating",
+      "分 院",
+      "(\" :",
+      "( \":",
+      "容留他人 吸毒",
+      "Int ended",
+      "▁fresh water",
+      "▁ 正",
+      "打 通",
+      "诊 所",
+      "更 喜欢",
+      "▁contract ed",
+      "▁contr acted",
+      "死 者",
+      "犯 非法",
+      "▁se ismic",
+      "▁Phot oshop",
+      "▁Photo shop",
+      "▁Photos hop",
+      "▁F c",
+      "在 整个",
+      "ur acy",
+      "ura cy",
+      "专业 合作社",
+      "▁Bright on",
+      "▁Brig hton",
+      "周 刊",
+      "序 号",
+      "这种 方式",
+      "▁Z oom",
+      "▁Zoo m",
+      "▁ Zoom",
+      "▁pos tal",
+      "▁post al",
+      "▁ 雪",
+      "不完 全",
+      "不 完全",
+      "他们 会",
+      "▁R EL",
+      "▁RE L",
+      "▁ REL",
+      "▁T ah",
+      "▁Ta h",
+      "第七 十九",
+      "ver gence",
+      "俊 杰",
+      "我 跟",
+      "种 的",
+      "纯 净",
+      "降 至",
+      "col span",
+      "cols pan",
+      "▁gen omes",
+      "▁genome s",
+      "▁c ounters",
+      "▁count ers",
+      "▁cou nters",
+      "▁counter s",
+      "轰 炸",
+      "ay e",
+      "a ye",
+      "Dis k",
+      "Di sk",
+      "D isk",
+      "he aded",
+      "head ed",
+      "hea ded",
+      "Equal To",
+      "nB etween",
+      "n Between",
+      "sh utdown",
+      "白 日",
+      "也是 非常",
+      "也 是非常",
+      "高等 院校",
+      "▁th ru",
+      "▁thr u",
+      "适 度",
+      "GR AM",
+      "G RAM",
+      "▁portray ed",
+      "书 名",
+      "排 斥",
+      "相 距",
+      "神 州",
+      "▁X L",
+      "▁ XL",
+      "第三 届",
+      "▁Qu ote",
+      "▁ Quote",
+      "▁wor rying",
+      "▁worry ing",
+      "献 血",
+      "下 部",
+      "庭 的",
+      "贫 穷",
+      "es i",
+      "e si",
+      "▁R FC",
+      "▁RF C",
+      "il let",
+      "ill et",
+      "ille t",
+      "和 非",
+      "） ,",
+      "不就 是",
+      "不 就是",
+      "据 介绍",
+      "La test",
+      "Lat est",
+      "L atest",
+      "engu in",
+      "▁e ther",
+      "▁et her",
+      "▁eth er",
+      "▁ ether",
+      "▁oscill ator",
+      "C X",
+      "上 饶",
+      "许 昌",
+      "amp ire",
+      "委托代理人 刘",
+      "▁Re vision",
+      "▁Rev ision",
+      "▁ Revision",
+      "时 被",
+      "if s",
+      "i fs",
+      "▁M IC",
+      "▁MI C",
+      "▁ MIC",
+      "or thy",
+      "ort hy",
+      "orth y",
+      "▁Re becca",
+      "▁Reb ecca",
+      "▁behavi ours",
+      "▁behaviour s",
+      "方向 的",
+      "好 不容易",
+      "H appy",
+      "nE con",
+      "n Econ",
+      "▁Eff ective",
+      "▁Effect ive",
+      "▁synthes ize",
+      "▁ synthesize",
+      "舍 不得",
+      "▁B atch",
+      "▁Bat ch",
+      "▁ Batch",
+      "get Time",
+      "re spect",
+      "res pect",
+      "resp ect",
+      "位上 的",
+      "位 上的",
+      "}{ {{",
+      "}{{ {",
+      "} {{{",
+      "▁dis cs",
+      "▁disc s",
+      "▁sup plemented",
+      "▁supplement ed",
+      "的 组织",
+      "▁a vec",
+      "▁av ec",
+      "▁Ass uming",
+      "热 心",
+      "通 路",
+      "物理 学",
+      "物 理学",
+      "DIR S",
+      "DI RS",
+      "D IRS",
+      "Def ined",
+      "Define d",
+      "D efined",
+      "▁neighbour ing",
+      "▁ 冰",
+      "比 是",
+      "有理 数的",
+      "有理数 的",
+      "▁P ros",
+      "▁Pro s",
+      "▁Pr os",
+      "▁med iator",
+      "▁media tor",
+      "▁medi ator",
+      "▁Wrest ling",
+      "▁ref lections",
+      "▁reflect ions",
+      "▁reflection s",
+      "打 野",
+      "火 影",
+      "主要 为",
+      "在 回答",
+      "我的 心",
+      "我 的心",
+      "nN one",
+      "nNo ne",
+      "nNon e",
+      "n None",
+      "▁glyc oprotein",
+      "十五 年",
+      "十 五年",
+      "回 馈",
+      "手中 的",
+      "手 中的",
+      "的 制作",
+      "Mon t",
+      "Mo nt",
+      "M ont",
+      "▁F eder",
+      "▁Fe der",
+      "▁Fed er",
+      "▁learn s",
+      "▁lear ns",
+      "▁cig arettes",
+      "▁cigarette s",
+      "心 力",
+      "ra ck",
+      "rac k",
+      "r ack",
+      "re ns",
+      "ren s",
+      "r ens",
+      "▁ 有限公司",
+      "Res pon",
+      "▁pr inters",
+      "▁print ers",
+      "▁pri nters",
+      "▁printer s",
+      "▁profess ors",
+      "▁professor s",
+      "哈 佛",
+      "指 望",
+      "Collect or",
+      "Coll ector",
+      "ri ka",
+      "rik a",
+      "r ika",
+      "ter bury",
+      "平 顶",
+      "得 不",
+      "PE C",
+      "P EC",
+      "Th an",
+      "T han",
+      "uc king",
+      "uck ing",
+      "▁rescue d",
+      "▁resc ued",
+      "切 断",
+      "▁a bc",
+      "▁ab c",
+      "▁ abc",
+      "en ues",
+      "enu es",
+      "enue s",
+      "▁R ide",
+      "▁Ri de",
+      "▁Rid e",
+      "▁Att ack",
+      "▁ Attack",
+      "波 特",
+      "▁f ör",
+      "co ded",
+      "code d",
+      "cod ed",
+      "c oded",
+      "▁Luc y",
+      "▁Lu cy",
+      "或 财产线索",
+      "orph ic",
+      "▁Mar ina",
+      "▁Ma rina",
+      "▁Mari na",
+      "▁Marin a",
+      "▁cho oses",
+      "▁choose s",
+      "▁in quiries",
+      "房 权证",
+      "贫困 户",
+      "nS il",
+      "n Sil",
+      "征 程",
+      "技术 在",
+      "cr ow",
+      "cro w",
+      "c row",
+      "ford able",
+      "▁terr ific",
+      "桃 园",
+      "▁ex on",
+      "▁sk ate",
+      "nO nline",
+      "nOn line",
+      "n Online",
+      "zeg ovina",
+      "▁pay roll",
+      "▁anim ations",
+      "▁animation s",
+      "年 到",
+      "▁( $\\\\",
+      "▁($ \\\\",
+      "Fr ench",
+      "Fre nch",
+      "F rench",
+      "CS V",
+      "C SV",
+      "SE QU",
+      "▁med iates",
+      "▁media tes",
+      "▁medi ates",
+      "▁mediate s",
+      "追 寻",
+      "nW E",
+      "n WE",
+      "▁E yes",
+      "▁Eye s",
+      "▁Ey es",
+      "日被 取保候审",
+      "eng lish",
+      "穿 梭",
+      "OP LE",
+      "nD aniel",
+      "nDan iel",
+      "n Daniel",
+      "▁shoot s",
+      "▁Fern ando",
+      "м а",
+      "军 训",
+      "ij ay",
+      "ija y",
+      "i jay",
+      "sl ides",
+      "slide s",
+      "CA A",
+      "C AA",
+      "▁S AR",
+      "▁SA R",
+      "▁ SAR",
+      "科研 项目",
+      "▁s cam",
+      "▁sc am",
+      "nR oyal",
+      "▁re const",
+      "▁recon st",
+      "▁re semble",
+      "▁asc ertain",
+      "▁c rystalline",
+      "▁crystal line",
+      "▁crystall ine",
+      "▁sub cutaneous",
+      "文 人",
+      "ary ng",
+      "是一 次",
+      "是 一次",
+      "na nt",
+      "nan t",
+      "n ant",
+      "文化 建设",
+      "nF ore",
+      "nFor e",
+      "n Fore",
+      "▁T rin",
+      "▁Tr in",
+      "▁Tri n",
+      "▁explo itation",
+      "▁exploit ation",
+      "吸 取",
+      "Bu d",
+      "B ud",
+      "天文 学",
+      "天 文学",
+      "暴力 枚举",
+      "ma ble",
+      "m able",
+      "另 案",
+      "部 级",
+      "全面 发展",
+      "▁Har bour",
+      "▁vulner abilities",
+      "完 备",
+      "回来 的",
+      "回 来的",
+      "真实 性",
+      "▁i Pod",
+      "▁r otor",
+      "▁ro tor",
+      "▁rot or",
+      "▁t icks",
+      "▁tick s",
+      "▁ti cks",
+      "▁ ticks",
+      "犯 交通肇事罪",
+      "▁ec ology",
+      "▁eco logy",
+      "百 里",
+      "而 这",
+      "AT S",
+      "A TS",
+      "关于 人民法院",
+      "▁en vision",
+      "▁env ision",
+      "朴 素",
+      "an de",
+      "and e",
+      "a nde",
+      "▁L iz",
+      "▁Li z",
+      "ou thern",
+      "outh ern",
+      "Mo r",
+      "M or",
+      "展览 会",
+      "长期 的",
+      "长 期的",
+      "Pac ific",
+      "▁con dom",
+      "▁cond om",
+      "▁condo m",
+      "▁int ends",
+      "▁intend s",
+      "名 列",
+      "vi vo",
+      "viv o",
+      "v ivo",
+      "▁umbre lla",
+      "尽 早",
+      "期满 后",
+      "解除 保全",
+      "'] ]],\\",
+      "做 大",
+      "批 复",
+      "算 术",
+      "▁Y ES",
+      "▁ YES",
+      "▁M oss",
+      "▁Mo ss",
+      "▁Mos s",
+      "桌 上",
+      "开心 的",
+      "开 心的",
+      "集团 的",
+      "集 团的",
+      "刑满 释放",
+      "投资 基金",
+      "生产 基地",
+      "He art",
+      "mix ed",
+      "m ixed",
+      "▁Fr aser",
+      "▁Fra ser",
+      "▁central ized",
+      "OnClick Listener",
+      "费 的",
+      "oz o",
+      "o zo",
+      "▁m asc",
+      "▁ma sc",
+      "▁mas c",
+      "nT able",
+      "n Table",
+      "▁br aces",
+      "▁bra ces",
+      "▁brace s",
+      "汇 编",
+      "read Line",
+      "▁eleg ance",
+      "搜 寻",
+      "生 孩子",
+      "中 对",
+      "尽 情",
+      "nI nc",
+      "nIn c",
+      "n Inc",
+      "▁dub bed",
+      "▁mathematic ian",
+      "结构 性",
+      "dr ag",
+      "d rag",
+      "▁sun shine",
+      "口 区",
+      "草 药",
+      "en ary",
+      "ena ry",
+      "▁rust ic",
+      "Creat ing",
+      "Cre ating",
+      "C reating",
+      "▁Winds or",
+      "b k",
+      "志 明",
+      "秉 持",
+      "教育 工作",
+      "ed iatric",
+      "中 使用",
+      "机构 和",
+      "at hy",
+      "ath y",
+      "a thy",
+      "ma int",
+      "main t",
+      "m aint",
+      "▁can opy",
+      "▁onc ology",
+      "松 弛",
+      "淮 南",
+      "It s",
+      "I ts",
+      "精神 病",
+      "▁制作 人",
+      "▁ 制作人",
+      "But tons",
+      "Button s",
+      "好 后",
+      "新 春",
+      "So l",
+      "S ol",
+      "投资 的",
+      "▁IS SN",
+      "第一 行包含",
+      "第一行 包含",
+      "rodu cing",
+      "rod ucing",
+      "外 围",
+      "抽 奖",
+      "ru ce",
+      "r uce",
+      "▁L ic",
+      "▁Li c",
+      "▁c osm",
+      "▁co sm",
+      "▁cos m",
+      "▁M akes",
+      "▁Make s",
+      "▁Ma kes",
+      "▁Mak es",
+      "Work ing",
+      "▁gen ius",
+      "▁rain bow",
+      "花 朵",
+      "▁Ar row",
+      "▁Arr ow",
+      "▁ Arrow",
+      "反应 的",
+      "反 应的",
+      "果 园",
+      "中新 网",
+      "社会 各界",
+      "▁st ains",
+      "▁stain s",
+      "he els",
+      "heel s",
+      "hee ls",
+      "id ual",
+      "▁c ols",
+      "▁col s",
+      "▁co ls",
+      "▁ cols",
+      "▁Ad ults",
+      "▁Adult s",
+      "▁c rying",
+      "▁cr ying",
+      "▁cry ing",
+      "▁Success ful",
+      "柔 性",
+      "毗 邻",
+      "最大 化",
+      "AAAA AAAA",
+      "▁ex quisite",
+      "伸 缩",
+      "迟 迟",
+      "郊 区",
+      "ai den",
+      "aid en",
+      "a iden",
+      "▁M ell",
+      "▁Me ll",
+      "▁Mel l",
+      "▁sh adows",
+      "▁shadow s",
+      "厂 长",
+      "阻 挡",
+      "▁r ushed",
+      "▁rush ed",
+      "▁worth while",
+      "的 材料",
+      "\\\\( {{",
+      "ar chy",
+      "arch y",
+      "arc hy",
+      "▁Ex po",
+      "▁Exp o",
+      "▁casp ase",
+      "Q T",
+      "… (",
+      "投 产",
+      "落 日",
+      "DE V",
+      "D EV",
+      "中 出现",
+      "并 取得",
+      "es are",
+      "esa re",
+      "esar e",
+      "▁C heap",
+      "▁Che ap",
+      "▁di vine",
+      "▁div ine",
+      "▁convinc ing",
+      "彝 族",
+      "水 土",
+      "领 军",
+      "▁K ab",
+      "▁Ka b",
+      "▁ Kab",
+      "信息 进行",
+      "认真 分析",
+      "▁Re serv",
+      "▁Res erv",
+      "▁G ur",
+      "▁Gu r",
+      "▁Cor nell",
+      "▁Corn ell",
+      "ени е",
+      "由 申请人",
+      "▁ch iral",
+      "▁chi ral",
+      "▁fore front",
+      "▁summar ized",
+      "▁summarize d",
+      "▁ 涛",
+      "哥 们",
+      "Me l",
+      "M el",
+      "il da",
+      "ild a",
+      "▁C arp",
+      "▁Car p",
+      "▁Ca rp",
+      "▁yield ing",
+      "▁Sec urities",
+      "r n",
+      "年 薪",
+      "领导 者",
+      "xi ng",
+      "xin g",
+      "x ing",
+      "▁Ster ling",
+      "s x",
+      "极 品",
+      "▁P ID",
+      "▁PI D",
+      "▁ PID",
+      "at ars",
+      "ata rs",
+      "atar s",
+      "Not ify",
+      "si zing",
+      "s izing",
+      "find All",
+      "use Effect",
+      "盆 地",
+      "ma d",
+      "m ad",
+      "▁ 接口",
+      "▁un anim",
+      "▁una nim",
+      "▁nit rate",
+      "▁st acked",
+      "▁stack ed",
+      "▁scholar ly",
+      "▁Sub mission",
+      "▁inter rupted",
+      "▁interrupt ed",
+      "始 于",
+      "影响 因素",
+      "pr inc",
+      "pri nc",
+      "▁U SER",
+      "▁US ER",
+      "▁USE R",
+      "▁ USER",
+      "Am azon",
+      "▁not ch",
+      "科学记数 法的",
+      "科学记数法 的",
+      "▁O ptical",
+      "▁Opt ical",
+      "▁Op tical",
+      "▁bout ique",
+      "店 内",
+      "集 资",
+      "ens ation",
+      "nE ducational",
+      "nEducation al",
+      "两 类",
+      "被 查封",
+      "nT urn",
+      "nTur n",
+      "n Turn",
+      "om nia",
+      "▁P ride",
+      "▁Pr ide",
+      "▁Pri de",
+      "▁Pro test",
+      "▁Prot est",
+      "▁Prote st",
+      "印 证",
+      "了 出来",
+      "安县 人民法院",
+      "安 县人民法院",
+      "▁str anger",
+      "▁strange r",
+      "▁strang er",
+      "趋 于",
+      "区域 内",
+      "润滑 油",
+      "un ix",
+      "uni x",
+      "u nix",
+      "来 往",
+      "挑战 性的",
+      "提供 一个",
+      "or ers",
+      "ore rs",
+      "orer s",
+      "Not Exist",
+      "区人民检察院 指控",
+      "▁question ing",
+      "奢 华",
+      "小 板",
+      "西 医",
+      "in stead",
+      "inst ead",
+      "▁index ing",
+      "▁ 加",
+      "万 辆",
+      "In M",
+      "I nM",
+      "ne o",
+      "n eo",
+      "Ch inese",
+      "val ence",
+      "form ations",
+      "formation s",
+      "format ions",
+      "▁trigger ing",
+      "__ )",
+      "_ _)",
+      "ap k",
+      "a pk",
+      "eg al",
+      "ega l",
+      "e gal",
+      "TY PES",
+      "TYPE S",
+      "▁Mal ay",
+      "▁Ma lay",
+      "▁g otta",
+      "▁got ta",
+      "运输 有限公司",
+      "▁F inite",
+      "▁Fin ite",
+      "Fin ished",
+      "Finish ed",
+      "▁import ing",
+      "▁ab straction",
+      "▁abs traction",
+      "▁abstract ion",
+      "应 试",
+      "的 山",
+      "博士 后",
+      "就会 有",
+      "就 会有",
+      "的事 物",
+      "的 事物",
+      "▁Gl obe",
+      "▁Glo be",
+      "▁Glob e",
+      "▁SE LECT",
+      "▁ SELECT",
+      "业 态",
+      "为 整数",
+      "市场 化",
+      "能有 效",
+      "能 有效",
+      "就是 这个",
+      "就 是这个",
+      "▁dom inate",
+      "▁Test ament",
+      "▁Cert ification",
+      "商 机",
+      "工 信",
+      "▁be lie",
+      "▁bel ie",
+      "整 车",
+      "晚 年",
+      "道 中",
+      "日 正式",
+      "与被告 张",
+      "政府 采购",
+      "酒店提供 免费",
+      "ba tis",
+      "bat is",
+      "b atis",
+      "separ ator",
+      "class ification",
+      "人 参",
+      "迪 士",
+      "只能 是",
+      "no ur",
+      "n our",
+      "▁bl ends",
+      "▁blend s",
+      "确认 人民调解协议",
+      "o L",
+      "保修 政策",
+      "▁sl ider",
+      "▁slide r",
+      "▁ slider",
+      "▁Up dates",
+      "▁Update s",
+      "▁ Updates",
+      "ap at",
+      "apa t",
+      "a pat",
+      "出席 会议",
+      "指派 代理检察员",
+      "按 自动撤回上诉处理",
+      "鸳 鸯",
+      "SN A",
+      "S NA",
+      "▁W r",
+      "▁ Wr",
+      "▁B ASE",
+      "▁BAS E",
+      "▁BA SE",
+      "▁ BASE",
+      "▁P oor",
+      "▁Po or",
+      "▁R econ",
+      "▁Re con",
+      "▁Rec on",
+      "▁Public ation",
+      "▁h one",
+      "▁hon e",
+      "▁ho ne",
+      "▁dig ging",
+      "▁ 伊",
+      "挂 在",
+      "▁U rl",
+      "▁Ur l",
+      "▁ Url",
+      "▁won ders",
+      "▁wond ers",
+      "▁wonder s",
+      "壮 族",
+      "大 庆",
+      "的 学校",
+      "▁H ull",
+      "▁Hu ll",
+      "Pr incipal",
+      "▁Viet namese",
+      "▁Vietnam ese",
+      "希 尔",
+      "▁V III",
+      "▁VI II",
+      "▁VII I",
+      "航 线",
+      "证明 材料",
+      "in ker",
+      "ink er",
+      "根据题意 得",
+      "nChild ren",
+      "n Children",
+      "▁junction s",
+      "▁Salmon ella",
+      "浓 浓",
+      "高 尚",
+      "▁P ig",
+      "▁Pi g",
+      "week ly",
+      "每 分钟",
+      "运行 的",
+      "运 行的",
+      "▁f auc",
+      "▁fa uc",
+      "▁Gl oucester",
+      "▁Sub sequently",
+      "立 功",
+      "怎么 了",
+      "▁s tip",
+      "▁st ip",
+      "丢 了",
+      "分 量",
+      "古 籍",
+      "换 届",
+      "的 回答",
+      "组 委会",
+      "PR OC",
+      "PRO C",
+      "nOver view",
+      "n Overview",
+      "_ \"",
+      "ql i",
+      "q li",
+      "你 应该",
+      "创建 于",
+      "创 建于",
+      "tl ocal",
+      "t local",
+      "▁am put",
+      "▁amp ut",
+      "other apeutic",
+      "字 幕",
+      "篮 板",
+      "ho lst",
+      "hol st",
+      "假 装",
+      "开 除",
+      "掩 盖",
+      "FR A",
+      "F RA",
+      "世界 大战",
+      "co ated",
+      "the ory",
+      "Hor izontal",
+      "▁text books",
+      "▁textbook s",
+      "二十 一",
+      "二 十一",
+      "ac ha",
+      "ach a",
+      "a cha",
+      "菠 菜",
+      "说 要",
+      "小时 的",
+      "小 时的",
+      "wo men",
+      "w omen",
+      "▁to ast",
+      "▁ toast",
+      "全 景",
+      "地 基",
+      "的老 师",
+      "的 老师",
+      "ib an",
+      "iba n",
+      "i ban",
+      "▁M PI",
+      "▁MP I",
+      "▁ MPI",
+      "▁anti tumor",
+      "▁antit umor",
+      "▁guid eline",
+      "▁guide line",
+      "▁co agulation",
+      ", [",
+      "K F",
+      "一直 到",
+      "一 直到",
+      "▁compl ained",
+      "▁complain ed",
+      "中 标",
+      "增大 而",
+      "总 觉得",
+      "淡淡 的",
+      "真的 好",
+      "真 的好",
+      "二审 上诉人",
+      "▁as sort",
+      "▁ass ort",
+      "烦 躁",
+      "Pro c",
+      "Pr oc",
+      "P roc",
+      "平方 千米",
+      "fr ontal",
+      "front al",
+      "公 有",
+      "细 细",
+      "莱 坞",
+      "正弦 定理",
+      "认罪 态度",
+      "graph ic",
+      "▁pean ut",
+      "▁cred ibility",
+      "预 言",
+      "▁S G",
+      "▁ SG",
+      "发挥 了",
+      "▁p es",
+      "▁pe s",
+      "▁ pes",
+      "us uario",
+      "二〇一六年 十一",
+      "二〇一六年十 一",
+      "nm anager",
+      "n manager",
+      "▁Spect rum",
+      "吃 什么",
+      "烟台 市",
+      "▁D il",
+      "▁Di l",
+      "▁He ather",
+      "▁Heat her",
+      "▁Heath er",
+      "填 补",
+      "ki ck",
+      "k ick",
+      "▁Coord inator",
+      "夺 得",
+      "德 镇",
+      "永 安",
+      "su s",
+      "s us",
+      "干 细胞",
+      "管理 与",
+      "▁{ })\\",
+      "▁{} )\\",
+      "▁{}) \\",
+      "▁ {})\\",
+      "as hire",
+      "ash ire",
+      "ashi re",
+      "a shire",
+      "al sa",
+      "als a",
+      "rx js",
+      "mu r",
+      "m ur",
+      "▁D EL",
+      "▁DE L",
+      "▁ DEL",
+      "▁R oth",
+      "▁Ro th",
+      "▁Rot h",
+      "▁sil icone",
+      "▁silicon e",
+      "▁minor ities",
+      "一般 为",
+      "▁al beit",
+      "Extract or",
+      "Extra ctor",
+      "冰 冰",
+      "港 区",
+      "基础 的",
+      "卫生 健康",
+      "as car",
+      "asc ar",
+      "a scar",
+      "▁G rab",
+      "▁Gr ab",
+      "▁Gra b",
+      "▁conc aten",
+      "▁elev ator",
+      "▁no except",
+      "枯 燥",
+      "ra el",
+      "r ael",
+      "几乎 没有",
+      "注销 信息",
+      "齐齐 哈尔",
+      "▁Trans l",
+      "▁Tra nsl",
+      "▁ Transl",
+      "▁in effective",
+      "上 调",
+      "不可 或缺",
+      "() ):\\",
+      "()) :\\",
+      "( )):\\",
+      "▁c lan",
+      "▁cl an",
+      ":: ~",
+      "人均 耕地",
+      "▁Aut hent",
+      "▁Auth ent",
+      "▁ Authent",
+      "Th reshold",
+      "▁aff ection",
+      "▁affect ion",
+      "念 头",
+      "瘦 肉",
+      "ba b",
+      "b ab",
+      "wa gen",
+      "w agen",
+      "▁Rol ling",
+      "▁Roll ing",
+      "安 利",
+      "抗 拒",
+      "J im",
+      "有理 数",
+      "克 里",
+      "页 无",
+      "在一 次",
+      "在 一次",
+      "Cl one",
+      "▁f ills",
+      "▁fil ls",
+      "▁fill s",
+      "▁ fills",
+      "西 装",
+      "TE M",
+      "T EM",
+      "将 面临",
+      "nW ORK",
+      "n WORK",
+      "▁an ne",
+      "▁ann e",
+      "▁ anne",
+      "▁ar Xiv",
+      "▁open ly",
+      "▁squ ash",
+      "第五百一十九 条",
+      "R X",
+      "t F",
+      "带 到",
+      "渐 进",
+      "空 的",
+      "第七 十三",
+      "▁go nad",
+      "▁gon ad",
+      "ch ecker",
+      "check er",
+      "▁rem inis",
+      "▁remin is",
+      "煤 气",
+      "实力 的",
+      "实 力的",
+      "▁p am",
+      "▁pa m",
+      "▁SPE CIAL",
+      "▁S elected",
+      "▁Se lected",
+      "▁Select ed",
+      "▁Sel ected",
+      "▁ Selected",
+      "最高人民法院关于 办理减刑",
+      "▁S ev",
+      "▁Se v",
+      "nTh ink",
+      "n Think",
+      "词条 注销信息",
+      "ut ation",
+      "uta tion",
+      "M n",
+      "都 觉得",
+      "工程 专业",
+      "els ius",
+      "t H",
+      "▁ 首",
+      "双 击",
+      "▁A ES",
+      "▁AE S",
+      "▁ AES",
+      "▁C AL",
+      "▁CA L",
+      "▁ CAL",
+      "▁L aur",
+      "▁La ur",
+      "▁n our",
+      "▁no ur",
+      "▁ nour",
+      "tr avel",
+      "tra vel",
+      "t ravel",
+      "教 务",
+      "也没 什么",
+      "也 没什么",
+      "▁le uc",
+      "应 及时",
+      "的 房子",
+      "▁ab sc",
+      "▁abs c",
+      "▁bl essing",
+      "▁bless ing",
+      "▁life span",
+      "▁lifes pan",
+      "x m",
+      "SD N",
+      "S DN",
+      "邯郸 市",
+      "▁H AS",
+      "▁HA S",
+      "考 前",
+      "▁D AY",
+      "▁DA Y",
+      "▁ DAY",
+      "这件事 情",
+      "这件 事情",
+      "这 件事情",
+      "OA uth",
+      "O Auth",
+      "▁ar sen",
+      "craft ed",
+      "▁program mers",
+      "▁programme rs",
+      "▁programmer s",
+      "每 种",
+      "管理 学",
+      "管 理学",
+      "不可 避免",
+      "▁rum ors",
+      "凝 胶",
+      "ar na",
+      "arn a",
+      "被执行人 名下",
+      "▁G aming",
+      "▁Ga ming",
+      "▁Gam ing",
+      "▁vine gar",
+      "nPres ident",
+      "n President",
+      "▁hist ogram",
+      "两个 数",
+      "两 个数",
+      "你就 是",
+      "你 就是",
+      "刑期 执行至",
+      "▁exp ire",
+      "▁Ch arter",
+      "▁Char ter",
+      "▁Chart er",
+      "▁hope ful",
+      "▁hop eful",
+      "八 个",
+      "的 合理",
+      "的 形象",
+      "▁Ig E",
+      "nar ray",
+      "n array",
+      "▁Ver ify",
+      "▁ Verify",
+      "▁al ters",
+      "▁alter s",
+      "▁alt ers",
+      "方 正",
+      "点 为",
+      "PC A",
+      "P CA",
+      "应该 如何",
+      "应 该如何",
+      "▁prom oters",
+      "▁promot ers",
+      "▁promote rs",
+      "▁promoter s",
+      "▁promo ters",
+      "看了 一",
+      "看 了一",
+      "▁B elle",
+      "▁Bel le",
+      "▁Bell e",
+      "▁post pon",
+      "▁evidence d",
+      "▁evid enced",
+      "v u",
+      "ar ton",
+      "art on",
+      "▁Man ila",
+      "▁creat inine",
+      "就 拿",
+      "在 世界",
+      "▁obsc ure",
+      "D V",
+      "一 曲",
+      "且 不",
+      "▁RNA s",
+      "▁RN As",
+      "▁ RNAs",
+      "mark ed",
+      "进 货",
+      "的 未来",
+      "数据 中心",
+      "病 房",
+      "轻 薄",
+      "ID I",
+      "I DI",
+      "Su ch",
+      "S uch",
+      "nl ibrary",
+      "n library",
+      "▁t etr",
+      "▁te tr",
+      "▁tet r",
+      "▁Nic ole",
+      "▁Ni cole",
+      "▁Nicol e",
+      "▁n otions",
+      "▁not ions",
+      "▁notion s",
+      "一 朝",
+      "基 建",
+      "好 感",
+      "玩家 的",
+      "玩 家的",
+      "gr own",
+      "gro wn",
+      "grow n",
+      "g rown",
+      "▁emphas izes",
+      "▁emphasize s",
+      "大 体",
+      "\"/ \"",
+      "\" /\"",
+      "/\" ,",
+      "/ \",",
+      "▁fem inist",
+      "▁femin ist",
+      "▁C yt",
+      "▁Cy t",
+      "公共 交通",
+      "Fr ance",
+      "F rance",
+      "▁Croat ian",
+      "▁Croatia n",
+      "▁encour agement",
+      "▁encourage ment",
+      "展 厅",
+      "装 机",
+      "林业 局",
+      "() );",
+      "()) ;",
+      "( ));",
+      "党建 工作",
+      "持 股",
+      "提 炼",
+      "是 世界",
+      "大部分 人",
+      "Gu ide",
+      "Gui de",
+      "Guid e",
+      "▁O Auth",
+      "▁OA uth",
+      "▁ OAuth",
+      "▁Sol omon",
+      "▁Solo mon",
+      "▁enc oder",
+      "▁encode r",
+      "▁ encoder",
+      "乖 乖",
+      "标 杆",
+      "留 守",
+      "预 案",
+      "ta dd",
+      "t add",
+      "Tr ait",
+      "Tra it",
+      "T rait",
+      "▁U nix",
+      "▁Un ix",
+      "▁Uni x",
+      "▁ Unix",
+      "▁H idden",
+      "▁ Hidden",
+      "▁Slovak ia",
+      "自 适应",
+      "将自己 的",
+      "将 自己的",
+      "▁产品 类型",
+      "▁ 产品类型",
+      "▁Sur ve",
+      "▁Surv e",
+      "▁T ables",
+      "▁Table s",
+      "▁Ta bles",
+      "▁Tab les",
+      "▁ Tables",
+      "▁genu inely",
+      "▁genuine ly",
+      "▁super conduct",
+      "一 见",
+      "圣 母",
+      "不少 人",
+      "如今 的",
+      "▁\" ,\"",
+      "▁\", \"",
+      "▁ \",\"",
+      "t rivial",
+      "▁mean while",
+      "二手 车",
+      "I iw",
+      "▁\" ,\\",
+      "▁\", \\",
+      "▁ \",\\",
+      "党风 廉政",
+      "▁King ston",
+      "▁Kings ton",
+      "▁seam lessly",
+      "▁seamless ly",
+      "用 工",
+      "迪士 尼",
+      "▁sh red",
+      "▁shr ed",
+      "linked in",
+      "以上资料 仅供参考",
+      "申请人 于",
+      "百度 小说",
+      "▁公司 性质",
+      "▁sn ail",
+      "▁per cutaneous",
+      "光 绪",
+      "能 用",
+      "那一 刻",
+      "那 一刻",
+      "▁l asers",
+      "▁laser s",
+      "▁las ers",
+      "金 陵",
+      "to ne",
+      "ton e",
+      "t one",
+      "具体情况 还请",
+      "▁d irectional",
+      "▁direct ional",
+      "▁direction al",
+      "化 了",
+      "复旦 大学",
+      "WIN DO",
+      "Start ed",
+      "Star ted",
+      "儿 科",
+      "班 车",
+      "ra r",
+      "r ar",
+      "元 左右",
+      "▁NA TO",
+      "▁NAT O",
+      "ú n",
+      "没 用",
+      "中国 国际",
+      "Game s",
+      "Ga mes",
+      "G ames",
+      "Pic ture",
+      "P icture",
+      "nPro ject",
+      "n Project",
+      "具体情况还请 咨询医生",
+      "▁ 薛",
+      "反 感",
+      "向 北",
+      "ca tal",
+      "cat al",
+      "c atal",
+      "ex ican",
+      "▁M arks",
+      "▁Mar ks",
+      "▁Mark s",
+      "▁V intage",
+      "▁Vin tage",
+      "▁am plified",
+      "▁ampl ified",
+      "▁paper work",
+      "第十 二章",
+      "第十二 章",
+      "第 十二章",
+      "▁or ch",
+      "▁ orch",
+      "nE mail",
+      "nEm ail",
+      "n Email",
+      "▁el bow",
+      "贷款 基准利率",
+      "爆 破",
+      "红 薯",
+      "do w",
+      "d ow",
+      "变速 箱",
+      "time d",
+      "tim ed",
+      "ti med",
+      "▁R osa",
+      "▁Ro sa",
+      "▁Ros a",
+      "conf idence",
+      "汉 堡",
+      "oh o",
+      "o ho",
+      "也是 有",
+      "也 是有",
+      "▁e rect",
+      "▁er ect",
+      "Source s",
+      "S ources",
+      "▁W idget",
+      "▁Wid get",
+      "▁ Widget",
+      "▁refr active",
+      "台 州",
+      "回 味",
+      "▁P ipe",
+      "▁Pi pe",
+      "▁Pip e",
+      "▁ Pipe",
+      "▁f ellows",
+      "▁fellow s",
+      "▁fell ows",
+      "Organ ization",
+      "▁U m",
+      "▁im perfect",
+      "▁imper fect",
+      "▁success ion",
+      "▁suc cession",
+      "▁succ ession",
+      "▁F o",
+      "南通 市",
+      "▁Se ems",
+      "▁See ms",
+      "▁dis ks",
+      "▁disk s",
+      "cult ure",
+      "c ulture",
+      "▁glomer ular",
+      "R J",
+      "讲 授",
+      "经常 会",
+      "ink les",
+      "inkle s",
+      "筛 查",
+      "谈 论",
+      "▁m ás",
+      "▁K ang",
+      "▁Ka ng",
+      "▁Kan g",
+      "▁n ore",
+      "▁no re",
+      "▁nor e",
+      "Over lay",
+      "▁all ocate",
+      "▁alloc ate",
+      "▁ allocate",
+      "access ible",
+      "Z n",
+      "票 价",
+      "ug h",
+      "u gh",
+      "Pro to",
+      "Pr oto",
+      "Prot o",
+      "▁ 汤",
+      "不 光",
+      "▁t ram",
+      "▁tr am",
+      "▁tra m",
+      "fr ag",
+      "fra g",
+      "f rag",
+      "▁Pol ynomial",
+      "▁pest icides",
+      "▁pestic ides",
+      "的风 格",
+      "的 风格",
+      "▁IP v",
+      "▁ IPv",
+      "▁L aws",
+      "▁La ws",
+      "▁Law s",
+      "▁li bert",
+      "▁lib ert",
+      "▁cre ations",
+      "▁creat ions",
+      "▁creation s",
+      "参与 者",
+      "对 学生",
+      "▁B ike",
+      "▁Bi ke",
+      "▁S ally",
+      "▁Sal ly",
+      "▁Dar win",
+      "▁R BI",
+      "▁RB I",
+      "ost ruct",
+      "o struct",
+      "▁Ang lican",
+      "文 库",
+      "离 去",
+      "的 良好",
+      "在自己 的",
+      "在 自己的",
+      "每一个 人",
+      "每一 个人",
+      "每 一个人",
+      "al most",
+      "alm ost",
+      "▁Ac ute",
+      "▁Per form",
+      "滋 事",
+      "et ur",
+      "e tur",
+      "▁A AA",
+      "▁AA A",
+      "▁ AAA",
+      "景点 的距离",
+      "景 点的距离",
+      "▁main land",
+      "三 家",
+      "新 房",
+      "离 谱",
+      "较 轻",
+      "对我 来说",
+      "对 我来说",
+      "build s",
+      "▁Me rit",
+      "▁Mer it",
+      "circ uit",
+      "tlog ger",
+      "t logger",
+      "vet ica",
+      "号行政处罚决定 书",
+      "号 行政处罚决定书",
+      "就 别",
+      "本 赛季",
+      "赵某 某",
+      "赵 某某",
+      "ve al",
+      "v eal",
+      "as one",
+      "ason e",
+      "▁Month ly",
+      "▁Parad ise",
+      "▁aff ective",
+      "▁affect ive",
+      "R x",
+      "复 读",
+      "评 比",
+      "▁Z ag",
+      "il ogy",
+      "ilo gy",
+      "ilog y",
+      "i logy",
+      "景点的距离 如下",
+      "安 宁",
+      "data b",
+      "da tab",
+      "dat ab",
+      "▁DE BUG",
+      "▁ DEBUG",
+      "拥有 移动电话",
+      "▁Alb any",
+      "▁Alban y",
+      "▁S olving",
+      "▁Sol ving",
+      "▁a verages",
+      "▁average s",
+      "▁aver ages",
+      "▁国 际",
+      "▁ 国际",
+      "多样 的",
+      "多 样的",
+      "▁Work er",
+      "▁Wor ker",
+      "▁ Worker",
+      "▁professional ism",
+      "K V",
+      "op ies",
+      "o pies",
+      "ot hesis",
+      "oth esis",
+      "othe sis",
+      "othes is",
+      "▁st ickers",
+      "▁stick ers",
+      "▁sticker s",
+      "草 本",
+      "链 条",
+      "St yled",
+      "Sty led",
+      "Style d",
+      "▁s andy",
+      "▁sand y",
+      "▁san dy",
+      "▁sp icy",
+      "▁psych o",
+      "▁psy cho",
+      "present ation",
+      "p resentation",
+      "之 谜",
+      "▁G W",
+      "▁ GW",
+      "中央 电视台",
+      "▁F ifty",
+      "▁Fif ty",
+      "▁Fi fty",
+      "撤回起诉 处理",
+      "▁Bos nia",
+      "war nings",
+      "warning s",
+      "warn ings",
+      "w arnings",
+      "▁lymph oid",
+      "▁Ann iversary",
+      "▁inter sections",
+      "▁intersection s",
+      "▁intersect ions",
+      "等 设施",
+      "商业 模式",
+      "▁dist ributor",
+      "▁distribut or",
+      "Par k",
+      "Pa rk",
+      "P ark",
+      "▁N acional",
+      "故本 院",
+      "故 本院",
+      "的比 较",
+      "的 比较",
+      "▁D um",
+      "▁Du m",
+      "▁was her",
+      "▁wash er",
+      "▁ washer",
+      "▁cha otic",
+      "▁not ices",
+      "▁notice s",
+      "红 星",
+      "个数 是",
+      "个 数是",
+      "▁H aus",
+      "▁Ha us",
+      "▁Ar rays",
+      "▁Array s",
+      "▁Arr ays",
+      "▁ Arrays",
+      "匈 牙",
+      "示范 区",
+      "▁C edar",
+      "▁Ce dar",
+      "on ge",
+      "ong e",
+      "o nge",
+      "▁it erate",
+      "▁iter ate",
+      "▁ iterate",
+      "▁fragment ation",
+      "▁frag mentation",
+      "办 的",
+      "大 棚",
+      "看 病",
+      "by ter",
+      "byte r",
+      "▁D iss",
+      "▁Dis s",
+      "▁Di ss",
+      "▁a orta",
+      "▁Sh annon",
+      "▁Shan non",
+      "▁up wards",
+      "▁upward s",
+      "▁ 付",
+      "知 己",
+      "ev o",
+      "e vo",
+      "ak ov",
+      "ako v",
+      "a kov",
+      "LIBR ARY",
+      "▁Prov idence",
+      "▁Provide nce",
+      "升 本",
+      "块 的",
+      "扭 转",
+      "爱情 的",
+      "爱 情的",
+      "nT own",
+      "nTo wn",
+      "▁H ugo",
+      "▁Hu go",
+      "▁Hug o",
+      "▁er ected",
+      "▁erect ed",
+      "▁the aters",
+      "▁theater s",
+      "▁theat ers",
+      "转 折",
+      "和 经济",
+      "LA ST",
+      "LAS T",
+      "L AST",
+      "ic ans",
+      "ica ns",
+      "ican s",
+      "im show",
+      "ims how",
+      "is file",
+      "isf ile",
+      "▁Any thing",
+      "不 放",
+      "树 叶",
+      "浪 潮",
+      "配 色",
+      "的 免费",
+      "这是 个",
+      "这 是个",
+      "ex ual",
+      "▁P f",
+      "▁S ic",
+      "▁Si c",
+      "这是 什么",
+      "这 是什么",
+      "▁U CLA",
+      "▁UC LA",
+      "dis miss",
+      "d ismiss",
+      "▁pian ist",
+      "▁transl ator",
+      "E I",
+      "▁S AS",
+      "▁SA S",
+      "和谐 社会",
+      "号民事 裁定书",
+      "号民事裁定 书",
+      "▁c runch",
+      "▁cr unch",
+      "▁cru nch",
+      "▁ crunch",
+      "▁destr uctive",
+      "分 段",
+      "来 袭",
+      "燕 子",
+      "qi ng",
+      "q ing",
+      "Diagn ostics",
+      "分 销",
+      "正常 使用",
+      "Cr edit",
+      "Cred it",
+      "Cre dit",
+      "C redit",
+      "fin ger",
+      "f inger",
+      "▁progress ively",
+      "▁progressive ly",
+      "是 她",
+      "车 管",
+      "防 爆",
+      "() }",
+      "( )}",
+      "的 版本",
+      "IF IC",
+      "IFI C",
+      "mg mt",
+      "▁where in",
+      "▁whe rein",
+      "declar ations",
+      "Br ien",
+      "B rien",
+      "▁ex piration",
+      "▁exp iration",
+      "严 密",
+      "各 行",
+      "蜜 蜂",
+      "ob y",
+      "o by",
+      "▁用 料",
+      "▁ 用料",
+      "▁G RO",
+      "▁GR O",
+      "▁ GRO",
+      "专业 领域",
+      "▁B har",
+      "▁Bh ar",
+      "你 所",
+      "后 发现",
+      "▁D im",
+      "▁Di m",
+      "▁ Dim",
+      "▁W ish",
+      "▁Wis h",
+      "▁Wi sh",
+      "▁su bs",
+      "▁sub s",
+      "if orms",
+      "iform s",
+      "i forms",
+      "▁ins isted",
+      "▁insist ed",
+      "▁cyt oplasm",
+      "▁def endant",
+      "▁defend ant",
+      "桃 源",
+      "欠 据",
+      "CO D",
+      "C OD",
+      "另一 半",
+      "另 一半",
+      "ce rol",
+      "cer ol",
+      "▁Jac ques",
+      "▁str ipped",
+      "▁stri pped",
+      "▁strip ped",
+      "适用 缓刑",
+      "ge nre",
+      "gen re",
+      "▁P are",
+      "▁Par e",
+      "▁Pa re",
+      "▁cor rupt",
+      "▁fibr illation",
+      "nU l",
+      "n Ul",
+      "申请执行人 陈",
+      "▁form ulate",
+      "▁formula te",
+      "▁head aches",
+      "▁headache s",
+      "▁% \\",
+      "▁ %\\",
+      "▁W on",
+      "▁Wo n",
+      "va ult",
+      "v ault",
+      "▁cut off",
+      "商务 部",
+      "nE xt",
+      "nEx t",
+      "n Ext",
+      "▁G an",
+      "▁Ga n",
+      "od ied",
+      "odi ed",
+      "o died",
+      "▁Ident ifier",
+      "▁ Identifier",
+      "▁carcin ogenesis",
+      "▁Nor dic",
+      "▁Nord ic",
+      "Eval uate",
+      "▁over time",
+      "▁overt ime",
+      "acchar ides",
+      "accharide s",
+      "全 社会",
+      "ro ts",
+      "rot s",
+      "r ots",
+      "的是 什么",
+      "的 是什么",
+      "由 被执行人",
+      "▁sub div",
+      "下 设",
+      "海 滨",
+      "▁Ch ern",
+      "▁Che rn",
+      "▁Cher n",
+      "nA frican",
+      "族自治 县人民法院",
+      "族自治县 人民法院",
+      "一定 能",
+      "双 十一",
+      "的 调查",
+      "ba ng",
+      "ban g",
+      "b ang",
+      "▁I sh",
+      "▁Is h",
+      "▁Day ton",
+      "▁Morris on",
+      "▁Morr ison",
+      "位 列",
+      "厅 长",
+      "打 得",
+      "sp ort",
+      "s port",
+      "▁B uzz",
+      "▁Bu zz",
+      "▁M ilk",
+      "▁Mil k",
+      "分 枝",
+      "恋 情",
+      "整 整",
+      "▁ 制片",
+      "国土 资",
+      "ti ght",
+      "t ight",
+      "▁N eck",
+      "▁Ne ck",
+      "▁Nec k",
+      "微信 小程序",
+      "表 的",
+      "在这 些",
+      "在 这些",
+      "▁A MS",
+      "▁AM S",
+      "▁ AMS",
+      "nc ategory",
+      "n category",
+      "▁pres erves",
+      "▁preserve s",
+      "污 泥",
+      "减 函数",
+      "披针 形",
+      "▁E PS",
+      "▁EP S",
+      "▁ EPS",
+      "re ceptor",
+      "▁phot ore",
+      "▁photo re",
+      "▁Car oline",
+      "▁Carol ine",
+      "硫 化",
+      "ed en",
+      "ede n",
+      "e den",
+      "个人 认为",
+      "in atal",
+      "ina tal",
+      "i natal",
+      "▁pe dest",
+      "▁ped est",
+      "温 岭",
+      "环 路",
+      "vo tes",
+      "vote s",
+      "v otes",
+      "▁sl owing",
+      "▁slow ing",
+      "傍 晚",
+      "大 风",
+      "常务 理事",
+      "▁refuge e",
+      "移 项",
+      "▁B G",
+      "▁ BG",
+      "黑色 的",
+      "黑 色的",
+      "fl ux",
+      "flu x",
+      "f lux",
+      "▁rep etition",
+      "十 首",
+      "总 计",
+      "▁H SV",
+      "▁HS V",
+      "nav igate",
+      "▁cush ion",
+      "预 处理",
+      ">\\ ',",
+      ">\\' ,",
+      "> \\',",
+      "ou ns",
+      "oun s",
+      "o uns",
+      "▁R abbit",
+      "▁Rab bit",
+      "▁ Rabbit",
+      "Mod ifier",
+      "▁cr ushed",
+      "▁crush ed",
+      "维 斯",
+      "而且 是",
+      "Me ss",
+      "M ess",
+      "▁an ion",
+      "▁ anion",
+      "报 到",
+      "col m",
+      "co lm",
+      "ed es",
+      "ede s",
+      "e des",
+      "ik an",
+      "ika n",
+      "i kan",
+      "提出 了一种",
+      "提出了 一种",
+      "it ably",
+      "ita bly",
+      "orb ent",
+      "▁Al lan",
+      "▁All an",
+      "▁im part",
+      "▁imp art",
+      "▁grab bed",
+      "▁fer mentation",
+      "提 琴",
+      "雷 锋",
+      "CO R",
+      "C OR",
+      "▁f right",
+      "▁fr ight",
+      "▁fri ght",
+      "挺 好",
+      "农户 数",
+      "农 户数",
+      "▁SV G",
+      "▁ SVG",
+      "cr umbs",
+      "crumb s",
+      "crum bs",
+      "▁sal aries",
+      "▁de duction",
+      "▁ded uction",
+      "▁deduct ion",
+      "▁construct ive",
+      "▁constr uctive",
+      "C d",
+      "时 分",
+      "ho ot",
+      "h oot",
+      "ial is",
+      "ia lis",
+      "i alis",
+      "▁u rea",
+      "▁ur ea",
+      "▁ure a",
+      "写 下",
+      "读 作",
+      "转 运",
+      "民主 党",
+      "NO WN",
+      "NOW N",
+      "oqu e",
+      "o que",
+      "arn ish",
+      "积极参加 政治",
+      "▁V isa",
+      "▁Vis a",
+      "▁Vi sa",
+      "nReg ular",
+      "n Regular",
+      "不 带",
+      "潜 艇",
+      "将 上述",
+      "道路 为",
+      "nV ir",
+      "n Vir",
+      "▁J ets",
+      "▁Je ts",
+      "▁Jet s",
+      "▁Elli ott",
+      "▁Trans mission",
+      "乔 木",
+      "现 身",
+      "侵犯 了",
+      "侵 犯了",
+      "he tical",
+      "het ical",
+      "hetic al",
+      "▁as phalt",
+      "甘 草",
+      "▁w s",
+      "▁ ws",
+      "的大 力",
+      "的 大力",
+      "nm ax",
+      "n max",
+      "被害 人的",
+      "被害人 的",
+      "nEx pl",
+      "n Expl",
+      "Qt Core",
+      "▁t unes",
+      "▁tun es",
+      "▁tune s",
+      "▁tu nes",
+      "▁Expl ore",
+      "▁authent icated",
+      "▁authentic ated",
+      "▁authenticate d",
+      "▁ authenticated",
+      "UM N",
+      "U MN",
+      "do u",
+      "d ou",
+      "克里 斯",
+      "克 里斯",
+      "旗舰 店",
+      "line ss",
+      "li ness",
+      "lin ess",
+      "lines s",
+      "l iness",
+      "▁pup py",
+      "QU AL",
+      "Q UAL",
+      "▁B uc",
+      "▁Bu c",
+      "can cer",
+      "c ancer",
+      "Default s",
+      "mut ation",
+      "m utation",
+      "留 给",
+      "诺 贝尔",
+      "in qu",
+      "um bers",
+      "umb ers",
+      "umber s",
+      "▁Lux ury",
+      "二〇一七年 十月",
+      "二〇一七年十 月",
+      "▁Car diff",
+      "▁Card iff",
+      "▁Cardi ff",
+      "▁Re creation",
+      "▁Rec reation",
+      "▁F i",
+      "▁ Fi",
+      "则 实数",
+      "▁hair st",
+      "▁ha irst",
+      "▁hairs t",
+      "▁releg ated",
+      "在 实际",
+      "ba gs",
+      "bag s",
+      "b ags",
+      "ov el",
+      "ove l",
+      "o vel",
+      "产品 重量",
+      "▁sur geries",
+      "▁surg eries",
+      "▁surge ries",
+      "е к",
+      "不 忍",
+      "幕 后",
+      "打 下",
+      "▁l end",
+      "▁le nd",
+      "▁len d",
+      "▁C odes",
+      "▁Co des",
+      "▁Code s",
+      "▁Cod es",
+      "▁ Codes",
+      "Re lation",
+      "Rel ation",
+      "踏 上",
+      "不要 再",
+      "较多 的",
+      "较 多的",
+      "nd el",
+      "nde l",
+      "n del",
+      "▁Brad ford",
+      "▁short ened",
+      "根本 没有",
+      "美国 总统",
+      "行政 拘留",
+      "▁impair ments",
+      "▁impairment s",
+      "Gi rl",
+      "G irl",
+      "▁T rav",
+      "▁Tr av",
+      "▁Tra v",
+      "▁w ont",
+      "▁won t",
+      "▁wo nt",
+      "管理 委员会",
+      "太 小",
+      "约 有",
+      "不少 于",
+      "不 少于",
+      "就 成了",
+      "这几 天",
+      "这 几天",
+      "越来 越大",
+      "越来越 大",
+      "品 德",
+      "拦 截",
+      "的 经历",
+      "ST OR",
+      "S TOR",
+      "He ading",
+      "Head ing",
+      "陈 代谢",
+      "Pr ime",
+      "Prim e",
+      "ns how",
+      "nsh ow",
+      "n show",
+      "家协会 会员",
+      "乙 方",
+      "挪 威",
+      "香 的",
+      "我这 个",
+      "我 这个",
+      "效果 好",
+      "特 异性",
+      "看 电影",
+      "▁tax a",
+      "▁ta xa",
+      "▁C ategories",
+      "▁ Categories",
+      "IB M",
+      "I BM",
+      "US H",
+      "U SH",
+      "wr ong",
+      "w rong",
+      "▁mac rom",
+      "▁macro m",
+      "残 忍",
+      "英 烈",
+      "Mi ss",
+      "M iss",
+      "▁| \\\\",
+      "▁|\\ \\",
+      "▁ |\\\\",
+      "▁or bits",
+      "▁orbit s",
+      "▁Nic olas",
+      "▁Nicol as",
+      "▁asym metry",
+      "▁asymmet ry",
+      "▁pro posing",
+      "▁propos ing",
+      "▁prop osing",
+      "d L",
+      "世 俗",
+      "侧 重",
+      "认识 的",
+      "ry an",
+      "r yan",
+      "所选 定的",
+      "你 我",
+      "计 画",
+      "▁о т",
+      "▁Ga uss",
+      "Min imum",
+      "▁r ookie",
+      "▁rook ie",
+      "▁When ever",
+      "w l",
+      "胚 胎",
+      "ki e",
+      "k ie",
+      "nu r",
+      "n ur",
+      "腹腔 镜",
+      "▁h rs",
+      "▁hr s",
+      "前 段时间",
+      "武 侯",
+      "nS l",
+      "n Sl",
+      "先 根据",
+      "App s",
+      "Ap ps",
+      "▁c ep",
+      "▁ce p",
+      "▁ cep",
+      "gu ides",
+      "guide s",
+      "gui des",
+      "guid es",
+      "▁ceiling s",
+      "▁ceil ings",
+      "▁even ings",
+      "▁evening s",
+      "w Y",
+      "南 宋",
+      "神 圣",
+      "跌 幅",
+      "躲 避",
+      "该村 到",
+      "页无 正文",
+      "nH and",
+      "nHa nd",
+      "n Hand",
+      "Are Equal",
+      "▁rad ians",
+      "▁radi ans",
+      "found ation",
+      "责任 感",
+      "朋友 聚餐",
+      "成 名",
+      "老 爸",
+      "事业 的",
+      "事 业的",
+      "ia de",
+      "iad e",
+      "i ade",
+      "ific ance",
+      "▁spatial ly",
+      "▁spat ially",
+      "厌 恶",
+      "cy st",
+      "c yst",
+      "ec hes",
+      "ech es",
+      "e ches",
+      "都有 自己的",
+      "都 有自己的",
+      "延 吉",
+      "鼠标 点击",
+      "▁d eng",
+      "▁de ng",
+      "▁den g",
+      "nA long",
+      "nAl ong",
+      "sur name",
+      "s urname",
+      "▁S print",
+      "▁Sp rint",
+      "▁Spr int",
+      "▁g astrop",
+      "▁gastr op",
+      "▁gastro p",
+      "▁gast rop",
+      "▁ 被",
+      "私 有",
+      "阳市 人民法院",
+      "阳 市人民法院",
+      "▁ac quis",
+      "▁acqu is",
+      "F requency",
+      "送 上",
+      "最后 一次",
+      "最后一 次",
+      "金融 危机",
+      "or nings",
+      "orn ings",
+      "▁Mar tha",
+      "▁Mart ha",
+      "▁sol ids",
+      "▁solid s",
+      "水 管",
+      "沧 桑",
+      "装 卸",
+      "辨 别",
+      "ab h",
+      "a bh",
+      "分割 线",
+      "ion a",
+      "io na",
+      "i ona",
+      "▁O le",
+      "▁Ol e",
+      "▁re pr",
+      "▁rep r",
+      "▁ repr",
+      "▁v ague",
+      "▁vag ue",
+      "▁va gue",
+      "▁ans ible",
+      "▁ ansible",
+      "发 芽",
+      "地 层",
+      "比 的",
+      "ot oxic",
+      "otox ic",
+      "o toxic",
+      "our ced",
+      "ource d",
+      "乌 拉",
+      "机 油",
+      "nc ia",
+      "n cia",
+      "▁G TP",
+      "▁GT P",
+      "tf ile",
+      "t file",
+      "▁Y outube",
+      "▁You tube",
+      "心 目",
+      "▁工作 分工",
+      "ph eres",
+      "phere s",
+      "pher es",
+      "▁d eter",
+      "▁de ter",
+      "▁det er",
+      "ns ystem",
+      "nsys tem",
+      "n system",
+      "万 亿",
+      "参与 了",
+      "洗 干净",
+      "▁A us",
+      "▁Au s",
+      "tc out",
+      "t cout",
+      "▁Valid ate",
+      "▁ Validate",
+      "城 里",
+      "ver ters",
+      "vert ers",
+      "verter s",
+      "verte rs",
+      "▁Out comes",
+      "▁Outcome s",
+      "个 赞",
+      "教 养",
+      "tr et",
+      "tre t",
+      "t ret",
+      "get Data",
+      "安 娜",
+      "法 语",
+      "研究 与",
+      "鲜明 的",
+      "鲜 明的",
+      "与被告 李",
+      "▁r ods",
+      "▁ro ds",
+      "▁rod s",
+      "eal ous",
+      "▁Rep roduce",
+      "▁s f",
+      "▁ sf",
+      "ol ey",
+      "ole y",
+      "o ley",
+      "无烟 客房",
+      "sp read",
+      "spr ead",
+      "s pread",
+      "▁O live",
+      "▁Ol ive",
+      "nVol ume",
+      "n Volume",
+      "▁sub urban",
+      "▁suburb an",
+      "▁rep utable",
+      "▁reput able",
+      "▁P ant",
+      "▁Pan t",
+      "▁Pa nt",
+      "in ative",
+      "i native",
+      "▁ 查看",
+      "不相 同",
+      "不 相同",
+      "▁O DE",
+      "▁OD E",
+      "如果 你的",
+      "如果你 的",
+      "pl ica",
+      "plic a",
+      "sur vey",
+      "▁sex uality",
+      "▁sexual ity",
+      "▁un forgettable",
+      "部 份",
+      "▁a a",
+      "▁ aa",
+      "▁k h",
+      "▁ kh",
+      "er mo",
+      "erm o",
+      "▁S paces",
+      "▁Sp aces",
+      "▁Space s",
+      "▁Spa ces",
+      "▁t oured",
+      "▁to ured",
+      "▁tour ed",
+      "▁tou red",
+      "b j",
+      "(\\\\ )",
+      "( \\\\)",
+      "▁) ,\\",
+      "▁), \\",
+      "▁ ),\\",
+      "创建 一个",
+      "S orry",
+      "▁Ang le",
+      "▁ Angle",
+      "摸 索",
+      "▁r uler",
+      "▁rule r",
+      "▁rul er",
+      "▁ru ler",
+      "不允许 携带宠物",
+      "出 错",
+      "最 简",
+      "芬 兰",
+      "ri zz",
+      "riz z",
+      "r izz",
+      "彻底 破裂",
+      "▁ 敏",
+      "切 尔",
+      "同 盟",
+      "th r",
+      "t hr",
+      "▁é t",
+      "▁ ét",
+      "读卡 器",
+      "▁m ens",
+      "▁me ns",
+      "▁men s",
+      "▁ mens",
+      "rab bit",
+      "r abbit",
+      "兴 建",
+      "周 六",
+      "an us",
+      "anu s",
+      "a nus",
+      "▁bi ome",
+      "▁biom e",
+      "▁bio me",
+      "通 往",
+      "卫生 部",
+      "▁Pr ison",
+      "▁Pri son",
+      "往 年",
+      "了一 张",
+      "了 一张",
+      "创新 能力",
+      "▁Fr agment",
+      "▁Fra gment",
+      "▁ Fragment",
+      "▁enlarg ed",
+      "nIn formation",
+      "n Information",
+      "▁Contin ental",
+      "▁conflic ting",
+      "▁conflict ing",
+      "双曲线 的",
+      "双 曲线的",
+      "年的 时间",
+      "年 的时间",
+      "Un its",
+      "Unit s",
+      "U nits",
+      "能够 认罪悔罪",
+      "石 英",
+      "▁M VP",
+      "▁MV P",
+      "元 或查封",
+      "▁C ann",
+      "▁Can n",
+      "▁Ca nn",
+      "Con struct",
+      "Co nstruct",
+      "Const ruct",
+      "Repos itories",
+      "个 个",
+      "植 被",
+      "Cu t",
+      "C ut",
+      "Pop up",
+      "ed ged",
+      "edge d",
+      "上诉于 山东省",
+      "建设 工程施工",
+      "建设工程 施工",
+      "▁out fits",
+      "▁outfit s",
+      "MS C",
+      "M SC",
+      "分析 与",
+      "已 完成",
+      "nE vents",
+      "nEven ts",
+      "nEvent s",
+      "nEv ents",
+      "n Events",
+      "ont ally",
+      "ontal ly",
+      "▁pal ate",
+      "▁pa late",
+      "▁tens ile",
+      "engine ering",
+      "nPop ulation",
+      "在 下",
+      "水 域",
+      "海 量",
+      "锦 绣",
+      "价格 的",
+      "价 格的",
+      "高度 的",
+      "高 度的",
+      "i ère",
+      "▁JO IN",
+      "互为 相反数",
+      "ou tfile",
+      "out file",
+      "▁ch rist",
+      "▁chr ist",
+      "晶 晶",
+      "▁其 六",
+      "后来 的",
+      "后 来的",
+      "▁B ottom",
+      "▁Bot tom",
+      "▁ Bottom",
+      "▁San tos",
+      "▁Sant os",
+      "个 子",
+      "nHe alth",
+      "n Health",
+      "▁extract ing",
+      "▁extr acting",
+      "▁extrac ting",
+      "果 树",
+      "构成 的",
+      "构 成的",
+      "▁P ie",
+      "▁Pi e",
+      "id ium",
+      "idi um",
+      "▁S tick",
+      "▁St ick",
+      "▁ru ins",
+      "▁ruin s",
+      "▁Stre pt",
+      "内 膜",
+      "本 章",
+      "▁` $",
+      "yn es",
+      "y nes",
+      "▁l an",
+      "▁la n",
+      "▁ lan",
+      "▁as tro",
+      "▁ast ro",
+      "▁astr o",
+      "▁ astro",
+      "张 文",
+      "\"> {",
+      "\" >{",
+      "is her",
+      "ish er",
+      "▁No tre",
+      "▁Not re",
+      ">{ {",
+      "> {{",
+      "▁t m",
+      "▁ tm",
+      "妈妈 的",
+      "nL ess",
+      "nLe ss",
+      "nLes s",
+      "n Less",
+      "▁fluor ide",
+      "▁con ceived",
+      "▁conce ived",
+      "▁Re placement",
+      "▁Replace ment",
+      "庆 典",
+      "开 荒",
+      "▁t k",
+      "▁ tk",
+      "▁C oul",
+      "▁Co ul",
+      "▁Cou l",
+      "admin istration",
+      ") ∵",
+      "学 了",
+      "风 电",
+      "▁up right",
+      "本裁定送达后 立即执行",
+      "女 装",
+      "师 兄",
+      "猎 人",
+      "la ke",
+      "l ake",
+      "▁J al",
+      "▁Ja l",
+      "琢 磨",
+      "表 皮",
+      "身 处",
+      "型 企业",
+      "带头 人",
+      "参加 会议",
+      "是什么 样的",
+      "是 什么样的",
+      "nW indows",
+      "n Windows",
+      "develop ed",
+      "devel oped",
+      "▁endomet rial",
+      "▁endometri al",
+      "▁ 钱",
+      "大 王",
+      "her oku",
+      "hero ku",
+      "▁红 袖添香网",
+      "▁ 红袖添香网",
+      "Res olved",
+      "Resolve d",
+      "▁Colon el",
+      "▁Ul timately",
+      "▁Ultimate ly",
+      "主 讲",
+      "年 由",
+      "Z oom",
+      "wan t",
+      "wa nt",
+      "w ant",
+      "▁A ds",
+      "▁Ad s",
+      "很大 程度上",
+      "Con nell",
+      "Conn ell",
+      "nNor wegian",
+      "_ ]",
+      "不 留",
+      "马 达",
+      "nN O",
+      "n NO",
+      "但对 于",
+      "但 对于",
+      "fe at",
+      "f eat",
+      "▁Bur ns",
+      "▁Burn s",
+      "▁Her zegovina",
+      "ue n",
+      "u en",
+      "▁X i",
+      "▁ Xi",
+      "和 学习",
+      "或者 代表人的人数提出副本",
+      "常态 化",
+      "采用 的",
+      "采 用的",
+      "▁Z ach",
+      "▁rece ivers",
+      "▁receive rs",
+      "▁receiver s",
+      "▁ 莫",
+      "出 不",
+      "山 川",
+      "称 呼",
+      "证人 证言",
+      "▁rec tal",
+      "▁rect al",
+      "▁adj unct",
+      "五 首",
+      "贯 通",
+      "报道 称",
+      "OD AY",
+      "O DAY",
+      "▁P AC",
+      "▁PA C",
+      "ode grad",
+      "▁Writ ers",
+      "▁Write rs",
+      "▁Writer s",
+      "▁exponential ly",
+      "▁exponent ially",
+      "▁immun otherapy",
+      "▁immuno therapy",
+      "宴 会",
+      "oi se",
+      "ois e",
+      "o ise",
+      "可以 作为",
+      "▁av atar",
+      "▁ avatar",
+      "▁harvest ing",
+      "网 贷",
+      "ji t",
+      "j it",
+      "电子 书",
+      "▁p om",
+      "▁po m",
+      "▁ pom",
+      "家 教",
+      "搞 得",
+      "爱 护",
+      "虐 待",
+      "▁N B",
+      "▁ NB",
+      "▁sal ts",
+      "▁salt s",
+      "▁stir ring",
+      "▁comp artments",
+      "▁compartment s",
+      "太 难",
+      "长 治",
+      "ak is",
+      "aki s",
+      "▁T CR",
+      "▁TC R",
+      "av ors",
+      "avor s",
+      "avo rs",
+      "仍 不",
+      "油 耗",
+      "重阳 节",
+      "懂 的",
+      "片 子",
+      "督 察",
+      "收费 是",
+      "政府 部门",
+      "Re uters",
+      "find all",
+      "集团 股份有限公司",
+      "▁in organic",
+      "▁struct urally",
+      "▁structural ly",
+      "番 禺",
+      "大家 一起",
+      "电动 汽车",
+      "on set",
+      "ons et",
+      "o nset",
+      "▁Trans ition",
+      "▁Transit ion",
+      "▁ Transition",
+      "东 盟",
+      "PC s",
+      "P Cs",
+      "关系 是",
+      "理性 的",
+      "理 性的",
+      "▁s log",
+      "▁sl og",
+      "LO N",
+      "L ON",
+      "at y",
+      "a ty",
+      "爱尔 兰",
+      "▁编辑 推荐",
+      "德 州",
+      "又一 个",
+      "又 一个",
+      "证据 的",
+      "▁A SD",
+      "▁AS D",
+      "synt hes",
+      "▁virt ue",
+      "▁antib acterial",
+      "nD ES",
+      "nDE S",
+      "n DES",
+      "▁Ap pend",
+      "▁App end",
+      "▁Appe nd",
+      "▁ Append",
+      "▁RE SULTS",
+      "▁Second ly",
+      ", %",
+      "二 项",
+      "日 出",
+      "牵 头",
+      "bl o",
+      "b lo",
+      "产品 在",
+      "体内 的",
+      "体 内的",
+      "可靠 的",
+      ")) ))\\",
+      "))) )\\",
+      ")))) \\",
+      ") )))\\",
+      "▁re stor",
+      "▁res tor",
+      "▁rest or",
+      "▁breath taking",
+      "▁breat htaking",
+      "两 句",
+      "之 事",
+      "他 能",
+      "秸 秆",
+      "▁a ld",
+      "▁al d",
+      "▁ ald",
+      "在卷 佐证",
+      "▁May o",
+      "▁Ma yo",
+      "▁ammon ia",
+      "ass ignment",
+      "assign ment",
+      "▁elastic ity",
+      "▁elast icity",
+      "Event Handler",
+      "名 胜",
+      "▁p urity",
+      "▁pur ity",
+      "▁re size",
+      "▁res ize",
+      "▁ resize",
+      "nCont ent",
+      "n Content",
+      "ingu ished",
+      "传 闻",
+      "re on",
+      "r eon",
+      "ron ym",
+      "r onym",
+      "ne sted",
+      "nes ted",
+      "nest ed",
+      "n ested",
+      "▁D ollar",
+      "▁Doll ar",
+      "▁Dol lar",
+      "▁ 丽",
+      "切 开",
+      "拖 延",
+      "多少 元",
+      "防腐 剂",
+      "▁v ibe",
+      "▁vi be",
+      "▁vib e",
+      "nSt eve",
+      "nSte ve",
+      "n Steve",
+      "也 对",
+      "至少 有",
+      "认定 书",
+      "ut ory",
+      "utor y",
+      "uto ry",
+      "nS ix",
+      "nf loat",
+      "n float",
+      "▁Mon roe",
+      "▁Sh arma",
+      "▁Shar ma",
+      "▁Java script",
+      "▁Jav ascript",
+      "球 形",
+      "直角坐标 系",
+      "br oken",
+      "bro ken",
+      "ver sed",
+      "vers ed",
+      "verse d",
+      "v ersed",
+      "▁st orms",
+      "▁storm s",
+      "▁season ed",
+      "▁seas oned",
+      "夜 里",
+      "马 力",
+      "oz a",
+      "o za",
+      "yu m",
+      "y um",
+      "一下 吧",
+      "▁b ure",
+      "▁bu re",
+      "▁bur e",
+      "Comp ile",
+      "▁tele scope",
+      "▁tel escope",
+      "▁telesc ope",
+      "冬 瓜",
+      "我 再",
+      "大学 毕业",
+      "▁R andy",
+      "▁Rand y",
+      "▁Ran dy",
+      "河 镇",
+      "然 的",
+      "▁micro array",
+      "托 福",
+      "nC H",
+      "n CH",
+      "▁ 填词",
+      "ec ko",
+      "eck o",
+      "ext ended",
+      "extend ed",
+      "▁Out side",
+      "▁accomplish ments",
+      "▁accomplishment s",
+      "tu int",
+      "t uint",
+      "▁idi opathic",
+      "长 效",
+      "▁ton nes",
+      "二〇一六年 九月",
+      "二〇一六年九 月",
+      "昌 平",
+      "有 啥",
+      "落 花",
+      "▁or ally",
+      "▁oral ly",
+      "本裁定 立即执行",
+      "▁con structions",
+      "▁construct ions",
+      "▁construction s",
+      "q c",
+      "心 事",
+      "然后 从",
+      "▁h av",
+      "▁ha v",
+      "▁ hav",
+      "输出 格式",
+      "▁\\' \\')",
+      "▁\\'\\' )",
+      "▁\\'\\ ')",
+      "教学 模式",
+      "▁P article",
+      "▁Part icle",
+      "▁ Particle",
+      "皱 纹",
+      "up o",
+      "u po",
+      "爱国 主义",
+      "或 不",
+      "▁run way",
+      "Cl inical",
+      "Service Provider",
+      "err no",
+      "act ice",
+      "actic e",
+      "▁De pot",
+      "▁Dep ot",
+      "▁wavelength s",
+      "▁interpret ations",
+      "▁interpretation s",
+      "便 民",
+      "奇 妙",
+      "改 制",
+      "患者 在",
+      "患 者在",
+      "▁le aked",
+      "▁leak ed",
+      "▁ancest ors",
+      "▁ancestor s",
+      "▁Serial izable",
+      "▁ Serializable",
+      "de en",
+      "d een",
+      "▁Order ed",
+      "▁Ord ered",
+      "▁ Ordered",
+      "R N",
+      "兵 器",
+      "eg o",
+      "e go",
+      "Icon s",
+      "I cons",
+      "ee nth",
+      "een th",
+      "e enth",
+      "tr unc",
+      "cert ificate",
+      "▁coord inating",
+      "胆 囊",
+      "可以 先",
+      "的 业务",
+      "计算 公式",
+      "▁n ome",
+      "▁no me",
+      "▁nom e",
+      "▁ nome",
+      "count y",
+      "cou nty",
+      "c ounty",
+      "▁S creens",
+      "▁Sc reens",
+      "▁Screen s",
+      "▁Scre ens",
+      "▁ Screens",
+      "渤 海",
+      "结 肠",
+      "对你 的",
+      "对 你的",
+      "海淀 区",
+      "▁brut al",
+      "▁bru tal",
+      "▁Top ology",
+      "气 泡",
+      "知 情",
+      "借款 人民币",
+      "借 款人民币",
+      "太 好",
+      "自 负",
+      "表现 的",
+      "体系 建设",
+      "R H",
+      "ul k",
+      "nCh ina",
+      "n China",
+      "nWork ing",
+      "n Working",
+      "▁en light",
+      "史 料",
+      "抗 诉",
+      "画 风",
+      "其次 是",
+      "ro gens",
+      "rogen s",
+      "rog ens",
+      "r ogens",
+      "Pen ding",
+      "P ending",
+      "▁hydro xy",
+      "▁hydrox y",
+      "▁ hydroxy",
+      "所 给",
+      "新 月",
+      "缤 纷",
+      "黑 客",
+      "第二 步",
+      "no op",
+      "n oop",
+      "▁mult il",
+      "▁multi l",
+      "▁mul til",
+      "▁lim estone",
+      "▁lime stone",
+      "▁coll ectors",
+      "▁collect ors",
+      "▁colle ctors",
+      "▁collector s",
+      "▁hem odynamic",
+      "▁hemod ynamic",
+      "▁ 谭",
+      "开 会",
+      ">( \"",
+      "> (\"",
+      "▁m c",
+      "▁ mc",
+      "大 辞典",
+      "es sing",
+      "ess ing",
+      "所选定的 客房",
+      "hold ing",
+      "hol ding",
+      "h olding",
+      "iat rics",
+      "iatric s",
+      "D Y",
+      "不可 思议",
+      "中国 社会",
+      "▁doubt s",
+      "▁doub ts",
+      "购 车",
+      "博士 生导师",
+      "nRad io",
+      "n Radio",
+      "nt able",
+      "n table",
+      "▁un expl",
+      "ri mental",
+      "rim ental",
+      "r imental",
+      "▁side bar",
+      "▁ sidebar",
+      "port folio",
+      "▁Malays ian",
+      "▁Malaysia n",
+      "耦 合",
+      "还 说",
+      "法律 允许",
+      "▁s tall",
+      "▁st all",
+      "nN atural",
+      "n Natural",
+      "▁ex pires",
+      "▁exp ires",
+      "▁expire s",
+      "▁ expires",
+      "ja m",
+      "j am",
+      "地 进行",
+      "教研 室",
+      "危害 程度",
+      "▁B udd",
+      "▁Bud d",
+      "▁Bu dd",
+      "▁re mix",
+      "▁rem ix",
+      "单 击",
+      "第六 节",
+      "网站 上",
+      "▁A er",
+      "▁viol ated",
+      "▁violate d",
+      "浮 云",
+      "▁F ischer",
+      "随 笔",
+      "▁C ir",
+      "▁Ci r",
+      "新闻 出版",
+      "ad apt",
+      "ada pt",
+      "▁S low",
+      "▁Sl ow",
+      "▁Slo w",
+      "▁ab used",
+      "▁abuse d",
+      "房 管",
+      "糊 涂",
+      "原审 判决",
+      "▁Char leston",
+      "▁Charles ton",
+      "▁Charl eston",
+      "▁fundamental ly",
+      "▁fundament ally",
+      "平 日",
+      "霸 道",
+      "有其 他",
+      "有 其他",
+      "sh aw",
+      "sha w",
+      "s haw",
+      "▁Sebast ian",
+      "▁overwhel med",
+      "▁polit ically",
+      "▁political ly",
+      "曾 因",
+      "淇 淋",
+      "较 快",
+      "ause s",
+      "au ses",
+      "aus es",
+      "a uses",
+      "pic kle",
+      "pick le",
+      "▁g ospel",
+      "▁Mid west",
+      "文 笔",
+      "od il",
+      "odi l",
+      "▁B ea",
+      "▁Be a",
+      "▁Dev ices",
+      "▁Device s",
+      "▁ Devices",
+      "在 社会",
+      "▁S yr",
+      "▁Sy r",
+      "▁b ib",
+      "▁bi b",
+      "▁ bib",
+      "二审 被上诉人",
+      "化工 有限公司",
+      "▁S amples",
+      "▁Sam ples",
+      "▁Sample s",
+      "▁ Samples",
+      "▁supposed ly",
+      "▁ و",
+      "▁ 👍",
+      "了 啊",
+      "孕 育",
+      "那 段",
+      "▁W P",
+      "▁ WP",
+      "都 无法",
+      "_ [",
+      "惠 州",
+      "▁S ap",
+      "▁Sa p",
+      "▁D evil",
+      "▁De vil",
+      "▁Dev il",
+      "COMM AND",
+      "二〇一六年 十二",
+      "二〇一六年十 二",
+      "Mi d",
+      "M id",
+      "好像 是",
+      "好 像是",
+      "服务 员",
+      "ur ai",
+      "ura i",
+      "农村 合作",
+      "sh ield",
+      "shi eld",
+      "▁g reens",
+      "▁green s",
+      "▁gre ens",
+      "招 收",
+      "▁I X",
+      "▁ IX",
+      "数据 范围",
+      "▁j upyter",
+      "▁ jupyter",
+      "▁super oxide",
+      "ü ck",
+      "▁文 章",
+      "▁ 文章",
+      "▁Frank furt",
+      "虚 构",
+      "▁人物 生平",
+      "▁S erie",
+      "▁Se rie",
+      "▁Ser ie",
+      "▁Con firm",
+      "▁Conf irm",
+      "▁ Confirm",
+      "▁intr icate",
+      "▁micro organisms",
+      "健 脾",
+      "惠 普",
+      "麻 痹",
+      "病人 的",
+      "病 人的",
+      "nS ET",
+      "nSE T",
+      "n SET",
+      "▁R OM",
+      "▁RO M",
+      "▁ ROM",
+      "B LOCK",
+      "▁furn ace",
+      "oper ators",
+      "operator s",
+      "▁Recomm end",
+      "▁H U",
+      "▁ HU",
+      "注意 的是",
+      "表现 在",
+      "表 现在",
+      "▁h ij",
+      "▁hi j",
+      "nd array",
+      "二〇一六年 六月",
+      "二〇一六年六 月",
+      "nCom plete",
+      "nComp lete",
+      "n Complete",
+      "▁Card inal",
+      "▁Cardi nal",
+      "▁sett ling",
+      "▁a h",
+      "▁ ah",
+      "世纪 的",
+      "功能 性",
+      "ut sche",
+      "uts che",
+      "utsch e",
+      "stack rel",
+      "▁half way",
+      "西 斯",
+      "▁I ntr",
+      "▁In tr",
+      "▁Int r",
+      "由于 被执行人",
+      "请求 判令被告",
+      "请求判令 被告",
+      "hard ware",
+      "▁c aching",
+      "▁ca ching",
+      "审 视",
+      "火 烧",
+      "}{ |",
+      "} {|",
+      "科学记数法的 表示",
+      "第一百七十 条第一款第",
+      "℃ \\",
+      "元 未",
+      "谁知 道",
+      "谁 知道",
+      "spect rum",
+      "▁com post",
+      "▁comp ost",
+      "▁compos t",
+      "▁celebr ates",
+      "▁celebrate s",
+      "▁lie utenant",
+      "nW H",
+      "n WH",
+      "▁b arr",
+      "▁bar r",
+      "▁ba rr",
+      "法律 的规定",
+      "法律的 规定",
+      "expand ed",
+      "一 碗",
+      "效 能",
+      "ja h",
+      "j ah",
+      "对他 们",
+      "对 他们",
+      "截至 目前",
+      "▁Go als",
+      "▁Goal s",
+      "▁Goa ls",
+      "▁mini ature",
+      "增 幅",
+      "族 的",
+      "皮 带",
+      "内容 和",
+      "攀 枝花",
+      "or get",
+      "org et",
+      "orge t",
+      "▁P iano",
+      "▁Pi ano",
+      "▁or phan",
+      "▁vac ant",
+      "▁va cant",
+      "▁mod erator",
+      "▁moder ator",
+      "出 任",
+      "投 递",
+      "胃 癌",
+      "重 伤",
+      "\"\" >\\",
+      "\"\"> \\",
+      "\" \">\\",
+      "为 空",
+      "与他 人",
+      "与 他人",
+      "理解 的",
+      "as sa",
+      "ass a",
+      "卫星 频道",
+      "文化 交流",
+      "pl ess",
+      "ple ss",
+      "ples s",
+      "p less",
+      "▁L yon",
+      "▁Ly on",
+      "J K",
+      "听 课",
+      "▁ 现代",
+      "余弦 定理",
+      "抗日 战争",
+      "▁p ills",
+      "▁pill s",
+      "▁pil ls",
+      "日 公开开庭审理了本案",
+      "三 份",
+      "目录 下",
+      "送达 地址",
+      "WR ITE",
+      "nD igital",
+      "n Digital",
+      "▁comm ute",
+      "▁commut e",
+      "上 图",
+      "之 大",
+      "总 队",
+      "ri tz",
+      "rit z",
+      "r itz",
+      "▁V ine",
+      "▁Vin e",
+      "▁Vi ne",
+      "▁D oors",
+      "▁Do ors",
+      "▁Door s",
+      "上 列",
+      "▁I U",
+      "▁ IU",
+      "st itu",
+      "stit u",
+      "建材 有限公司",
+      "ist ency",
+      "iste ncy",
+      "▁ 尹",
+      "学 好",
+      "重 来",
+      "Bu r",
+      "B ur",
+      "x FF",
+      "和 朋友",
+      "▁M eth",
+      "▁Me th",
+      "▁Met h",
+      "ed itary",
+      "edit ary",
+      "▁fib res",
+      "▁fibr es",
+      "▁fibre s",
+      "T c",
+      "京 津",
+      "累 累",
+      "▁E nv",
+      "▁En v",
+      "▁ Env",
+      "三个 方面",
+      "▁Go al",
+      "▁Goa l",
+      "▁ Goal",
+      "er text",
+      "ert ext",
+      "▁f used",
+      "▁fu sed",
+      "▁fuse d",
+      "▁fus ed",
+      "recogn ized",
+      "前 世",
+      "高 素质",
+      "tp ass",
+      "t pass",
+      "▁out rage",
+      "▁Pro ducer",
+      "▁Produ cer",
+      "▁ Producer",
+      "制 止",
+      "高 通",
+      "而言 之",
+      "而 言之",
+      "nper malink",
+      "n permalink",
+      "不 作",
+      "居 室",
+      "汉 服",
+      "▁M ats",
+      "▁Mat s",
+      "▁Ma ts",
+      "▁Ex plain",
+      "▁Expl ain",
+      "▁Exp lain",
+      "▁Pre dict",
+      "▁Pred ict",
+      "na bstract",
+      "nab stract",
+      "n abstract",
+      "连 环",
+      "医护 人员",
+      "▁D ana",
+      "▁Dan a",
+      "▁Da na",
+      "En code",
+      "Enc ode",
+      "E ncode",
+      "Pub lished",
+      "Publish ed",
+      "▁therm odynamic",
+      "▁ Å",
+      "摧 毁",
+      "收 音",
+      "盖 上",
+      "ad ores",
+      "ado res",
+      "ador es",
+      "第二百四十 二",
+      "▁file path",
+      "▁ filepath",
+      "▁imm ature",
+      "侵 蚀",
+      "专业 人士",
+      "▁wer den",
+      "sh ake",
+      "sha ke",
+      "▁G amma",
+      "▁Ga mma",
+      "▁Gam ma",
+      "▁ Gamma",
+      "▁for ex",
+      "▁fore x",
+      "▁fo rex",
+      "▁dec ays",
+      "▁decay s",
+      "▁ref res",
+      "▁refr es",
+      "▁Winn ipeg",
+      "立 志",
+      "宝贵 的",
+      "宝 贵的",
+      "由 上诉人",
+      "的主要 原因",
+      "▁c onda",
+      "▁con da",
+      "▁co nda",
+      "▁cond a",
+      "▁ conda",
+      "▁tr ibal",
+      "▁tri bal",
+      "▁trib al",
+      "小 巧",
+      "收 视",
+      "高 楼",
+      "▁tri bes",
+      "▁trib es",
+      "▁tribe s",
+      "▁argu ably",
+      "S dk",
+      "cy t",
+      "c yt",
+      "or no",
+      "orn o",
+      "▁D ro",
+      "▁Dr o",
+      "Code c",
+      "Co dec",
+      "Cod ec",
+      "C odec",
+      "二〇一七年 十一",
+      "二〇一七年十 一",
+      "▁Feed back",
+      "▁ Feedback",
+      "> ]",
+      "东 坡",
+      "向 社会",
+      "nn um",
+      "n num",
+      "▁an tif",
+      "▁ant if",
+      "▁anti f",
+      "▁cycl o",
+      "在服刑 期间的",
+      "在服刑期间 的",
+      "nS ample",
+      "nSam ple",
+      "n Sample",
+      "▁prosec utors",
+      "▁prosecutor s",
+      "一 辈",
+      "打 折",
+      "颜色 的",
+      "颜 色的",
+      "l iquid",
+      "您 所选定的客房",
+      "▁sal ivary",
+      "▁saliva ry",
+      "区 和",
+      "赛 道",
+      "mg r",
+      "m gr",
+      "▁H BO",
+      "▁HB O",
+      "▁intention ally",
+      "▁intentional ly",
+      "▁ 解",
+      "行 列",
+      "{{ \\",
+      "{ {\\",
+      "▁sk iers",
+      "▁ski ers",
+      "双 层",
+      "对 各",
+      "沉 迷",
+      "车 厢",
+      "国庆 节",
+      "教学 方法",
+      "▁B arb",
+      "▁Bar b",
+      "▁Ba rb",
+      "######## ######\\",
+      "############ ##\\",
+      "spec s",
+      "spe cs",
+      "▁progress ed",
+      "也 叫",
+      "何 以",
+      "护 航",
+      "立 项",
+      "▁k on",
+      "▁ko n",
+      "▁ kon",
+      "▁ 就",
+      "孤 单",
+      "CA N",
+      "C AN",
+      "也有 很多",
+      "也 有很多",
+      "All owed",
+      "Allow ed",
+      "▁an astom",
+      "复 查",
+      "小 人",
+      "报 案",
+      "▁k o",
+      "▁ ko",
+      "in is",
+      "ini s",
+      "i nis",
+      "nH as",
+      "nHa s",
+      "n Has",
+      "nN ov",
+      "nNo v",
+      "n Nov",
+      "之 所",
+      "唿 吁",
+      "赞 美",
+      "个体 户",
+      "ér ie",
+      "é rie",
+      "中心 主任",
+      "▁R ates",
+      "▁Ra tes",
+      "▁Rate s",
+      "▁Rat es",
+      "▁Sustain able",
+      "得 利",
+      "对 管辖权",
+      "nf ield",
+      "nfi eld",
+      "n field",
+      "生 子",
+      "建筑 安装",
+      "▁d osing",
+      "▁do sing",
+      "▁dos ing",
+      "▁Com prehensive",
+      "▁electro static",
+      "▁electr ostatic",
+      "坏 死",
+      "bu ck",
+      "b uck",
+      "am ics",
+      "ami cs",
+      "amic s",
+      "▁under lie",
+      "▁首 先",
+      "▁ 首先",
+      "参考 答案",
+      "cre st",
+      "cr est",
+      "c rest",
+      "▁S TEM",
+      "▁ST EM",
+      "▁STE M",
+      "▁ STEM",
+      "▁open ings",
+      "▁opening s",
+      "nTransport ation",
+      "定 性",
+      "为 实现",
+      "in ez",
+      "ine z",
+      "i nez",
+      "▁t aper",
+      "▁ta per",
+      "▁tap er",
+      "▁tape r",
+      "号民事判决书 的执行",
+      "摄像 机",
+      "▁sc anned",
+      "▁scan ned",
+      "▁civil ians",
+      "▁civilian s",
+      "敦 煌",
+      "Gl y",
+      "G ly",
+      "nd i",
+      "n di",
+      "国家 重点",
+      "享受 三包服务",
+      "罪犯 奖惩审批表",
+      "▁sub urbs",
+      "▁suburb s",
+      "cent ration",
+      "▁inter play",
+      "向 外",
+      "安 居",
+      "底 盘",
+      "极 了",
+      "是 完全",
+      "gr ep",
+      "gre p",
+      "g rep",
+      "symbol s",
+      "建设 工程有限公司",
+      "建设工程 有限公司",
+      "▁T L",
+      "▁ TL",
+      "审理 中",
+      "月下 旬",
+      "购买 了",
+      "购 买了",
+      "nK e",
+      "n Ke",
+      "于 一体",
+      "产品 和",
+      "的了 解",
+      "的 了解",
+      "pa st",
+      "pas t",
+      "p ast",
+      "▁G ST",
+      "▁GS T",
+      "国家 安全",
+      "am arin",
+      "ama rin",
+      "amar in",
+      "▁rec urs",
+      "▁deleg ation",
+      "款 的",
+      "f amilies",
+      "三 课",
+      "外 墙",
+      "正 值",
+      "章 程",
+      "BU T",
+      "B UT",
+      "▁n ig",
+      "▁ni g",
+      "il ers",
+      "ile rs",
+      "iler s",
+      "i lers",
+      "par en",
+      "pa ren",
+      "pare n",
+      "p aren",
+      "pre set",
+      "pres et",
+      "p reset",
+      "据此 解答即可",
+      "据此解答 即可",
+      "▁ce ased",
+      "▁cease d",
+      "▁cult urally",
+      "▁cultural ly",
+      "数 和",
+      "会 将",
+      "点 多",
+      "▁ 适量",
+      "▁K ot",
+      "▁Ko t",
+      "几个 方面",
+      "▁g ulp",
+      "▁gu lp",
+      "▁ gulp",
+      "▁squ eeze",
+      "▁squee ze",
+      "▁un aware",
+      "▁una ware",
+      "á r",
+      "nCon vert",
+      "n Convert",
+      "▁f athers",
+      "▁fa thers",
+      "▁father s",
+      "▁oscill ation",
+      "亦 可",
+      "池 塘",
+      "CL I",
+      "C LI",
+      "▁authent icate",
+      "▁authentic ate",
+      "▁ authenticate",
+      "六 合",
+      "称 作",
+      "▁ 命令",
+      "▁un pack",
+      "▁ unpack",
+      "进 阶",
+      "这对 于",
+      "这 对于",
+      "ti ps",
+      "tip s",
+      "t ips",
+      "▁{ $",
+      "▁ {$",
+      "▁P all",
+      "▁Pal l",
+      "▁Pa ll",
+      "▁M aker",
+      "▁Make r",
+      "▁Ma ker",
+      "▁Mak er",
+      "▁ Maker",
+      "减为 有期徒刑",
+      "非物质 文化遗产",
+      "▁ell ipse",
+      "CA LL",
+      "CAL L",
+      "C ALL",
+      "▁发行 时间",
+      "▁free zer",
+      "▁freeze r",
+      "▁ 侯",
+      "祖国 的",
+      "祖 国的",
+      "/{ {",
+      "/ {{",
+      "ic l",
+      "i cl",
+      ", “",
+      "构 思",
+      "禁 毒",
+      "GP L",
+      "G PL",
+      "神奇 的",
+      "神 奇的",
+      "规模 的",
+      "nPol ish",
+      "nPo lish",
+      "add resses",
+      "address es",
+      "addr esses",
+      "嘻 嘻",
+      "安装 在",
+      "▁c lash",
+      "▁cl ash",
+      "▁For bes",
+      "▁Hol dings",
+      "▁Hold ings",
+      "f k",
+      "▁ 带",
+      "布 朗",
+      "故 此",
+      "身 世",
+      "降 噪",
+      "也算 是",
+      "也 算是",
+      "区域 的",
+      "往往 是",
+      "的一 份",
+      "的 一份",
+      "__ ),",
+      "__) ,",
+      "t struct",
+      "▁m osaic",
+      "▁Fort une",
+      "好 运",
+      "RA M",
+      "R AM",
+      "▁B rom",
+      "▁Br om",
+      "▁Bro m",
+      "皇 宫",
+      "▁w orm",
+      "▁wor m",
+      "▁wo rm",
+      "▁ worm",
+      "▁Do ing",
+      "▁mot ivate",
+      "▁motiv ate",
+      "td b",
+      "t db",
+      "an za",
+      "anz a",
+      "nB ill",
+      "nBi ll",
+      "n Bill",
+      "sub title",
+      "▁ultr asonic",
+      "▁ultras onic",
+      "▁ultrason ic",
+      "死 去",
+      "证明 其",
+      "▁A post",
+      "▁Ap ost",
+      "fav icon",
+      "跟 进",
+      "召开 了",
+      "召 开了",
+      "▁c ondo",
+      "▁con do",
+      "▁co ndo",
+      "▁cond o",
+      "Commun ity",
+      "Comm unity",
+      "▁withdraw n",
+      "▁withd rawn",
+      "▁disadvant ages",
+      "▁disadvantage s",
+      "学 霸",
+      "推 迟",
+      "FT ER",
+      "F TER",
+      "工艺 美术",
+      "▁b reeze",
+      "▁br eeze",
+      "▁Sat ellite",
+      "高分 子",
+      "高 分子",
+      "ta ll",
+      "tal l",
+      "t all",
+      "生命 周期",
+      "资金 周转",
+      "St reet",
+      "▁Strateg ies",
+      "▁K ad",
+      "▁Ka d",
+      "ovol ta",
+      "▁Sc ala",
+      "▁mush rooms",
+      "▁mushroom s",
+      "国 安",
+      "距离 最近的",
+      "文化传播 有限公司",
+      "稀 土",
+      "不能 再",
+      "▁l ing",
+      "▁li ng",
+      "▁lin g",
+      "▁ ling",
+      "Part ial",
+      "▁Bud apest",
+      "积 水",
+      "等 优点",
+      "nS ep",
+      "nSe p",
+      "n Sep",
+      "▁c ogn",
+      "▁co gn",
+      "olph ins",
+      "olphin s",
+      "不 解",
+      "稳 固",
+      "还 被",
+      "会比 较",
+      "会 比较",
+      "作用 下",
+      "群体 的",
+      "群 体的",
+      "▁T yr",
+      "▁Ty r",
+      "▁k ar",
+      "▁ka r",
+      "▁ kar",
+      "▁ethnic ity",
+      "▁ethn icity",
+      "ant lr",
+      "▁team mates",
+      "▁teammate s",
+      "中 区",
+      "日 元",
+      "运 送",
+      "tr as",
+      "tra s",
+      "t ras",
+      "ér ic",
+      "é ric",
+      "▁S him",
+      "▁Sh im",
+      "▁Shi m",
+      "Cor rect",
+      "C orrect",
+      "▁Ident ify",
+      "公 差",
+      "寻 衅",
+      "为 原告",
+      "编号 为",
+      "编 号为",
+      "p ulse",
+      "▁pl ural",
+      "▁plur al",
+      "▁ plural",
+      "▁Christ ine",
+      "二手 房",
+      "优 缺点",
+      "依然 是",
+      "fu el",
+      "f uel",
+      "▁b ik",
+      "▁bi k",
+      "▁Ab ove",
+      "▁K atie",
+      "▁Kat ie",
+      "▁Ka tie",
+      "▁log os",
+      "▁lo gos",
+      "▁logo s",
+      "▁intermediate s",
+      "▁intermed iates",
+      "Sal es",
+      "Sa les",
+      "Sale s",
+      "S ales",
+      "▁D iana",
+      "▁Di ana",
+      "本裁定 立即开始执行",
+      "未 见",
+      "查 到",
+      "胰 腺",
+      "用 一个",
+      "▁aut ologous",
+      "▁K S",
+      "▁ KS",
+      "体育 馆",
+      "hl er",
+      "h ler",
+      "▁d iary",
+      "▁di ary",
+      "▁dia ry",
+      "it tings",
+      "itt ings",
+      "itting s",
+      "ord inate",
+      "题 文",
+      "ł aw",
+      "Min i",
+      "Mi ni",
+      "M ini",
+      "▁M ask",
+      "▁Ma sk",
+      "▁Mas k",
+      "▁ Mask",
+      "▁启动 方式",
+      "快 点",
+      "车 载",
+      "酷 睿",
+      "IF E",
+      "I FE",
+      "生活 费",
+      "** /\\",
+      "* */\\",
+      "oc cer",
+      "occ er",
+      "nf inal",
+      "nfi nal",
+      "n final",
+      "▁D erek",
+      "▁De rek",
+      "▁Der ek",
+      "扭 矩",
+      "行业 中",
+      "ma ch",
+      "mac h",
+      "m ach",
+      "nf ig",
+      "nfi g",
+      "n fig",
+      "你是 不是",
+      "你 是不是",
+      "这 批",
+      "▁i k",
+      "▁ ik",
+      "月 开始",
+      "▁prolifer ative",
+      "胡 同",
+      "($ {",
+      "( ${",
+      "这些 东西",
+      "之日起 十五日内",
+      "▁c atches",
+      "▁catch es",
+      "▁cat ches",
+      "w k",
+      "她 就",
+      "价格 为",
+      "ab us",
+      "a bus",
+      "ns erver",
+      "n server",
+      "nt arget",
+      "n target",
+      "Http Request",
+      "最美 的",
+      "最 美的",
+      "一直 没有",
+      "Par ts",
+      "Part s",
+      "P arts",
+      "之 年",
+      "án d",
+      "á nd",
+      "ra no",
+      "ran o",
+      "r ano",
+      "第六 十条",
+      "me aning",
+      "mean ing",
+      "凝 固",
+      "建 新",
+      "江 汉",
+      "尼 西亚",
+      "的 产生",
+      "重 金属",
+      "ac use",
+      "应准予 离婚",
+      "oli nergic",
+      "olin ergic",
+      "洗 碗",
+      "产品 有",
+      "日 消息",
+      ":: __",
+      "哥伦 比亚",
+      "清偿 责任",
+      "at hing",
+      "ath ing",
+      "athi ng",
+      "a thing",
+      "▁Dr ake",
+      "mo unted",
+      "mount ed",
+      "▁post doctoral",
+      "亲 手",
+      "无 双",
+      "bird s",
+      "bir ds",
+      "b irds",
+      "nH elp",
+      "nHe lp",
+      "nHel p",
+      "n Help",
+      "二〇一七年 九",
+      "W s",
+      "考 入",
+      "建议 你",
+      "Ro ll",
+      "R oll",
+      "▁f name",
+      "▁fn ame",
+      "▁ fname",
+      "▁br eeds",
+      "▁bre eds",
+      "▁breed s",
+      "氧化 物",
+      "nG en",
+      "nGe n",
+      "n Gen",
+      "消防 安全",
+      "▁temp file",
+      "短 缺",
+      "ス ト",
+      "等 我",
+      "香 烟",
+      "▁C VD",
+      "▁CV D",
+      "▁e rotic",
+      "▁er otic",
+      "nB esides",
+      "▁Bel fast",
+      "▁Belf ast",
+      "▁pl anner",
+      "▁plan ner",
+      "▁bench marks",
+      "▁benchmark s",
+      "▁ 凯",
+      "弥 漫",
+      "情 愿",
+      "拓 扑",
+      "纲 要",
+      "损害 国家",
+      "▁accommod ations",
+      "▁accommodation s",
+      "破 损",
+      "镀 锌",
+      "nN OTE",
+      "nNO TE",
+      "n NOTE",
+      "▁M usk",
+      "▁Mus k",
+      "▁Mu sk",
+      "有什么 区别",
+      "▁par cel",
+      "recip itation",
+      "椅 子",
+      "段 子",
+      "中文 版",
+      "只是 在",
+      "只 是在",
+      "set ter",
+      "sett er",
+      "s etter",
+      "nM unicip",
+      "身 子",
+      "如实供述 自己",
+      "nCom ment",
+      "n Comment",
+      "短 线",
+      "▁I b",
+      "amin ergic",
+      "ami nergic",
+      "▁per ceive",
+      "▁im plication",
+      "▁impl ication",
+      "▁dem ographics",
+      "▁demo graphics",
+      "▁demographic s",
+      "清 真",
+      "芳 草",
+      "pa s",
+      "p as",
+      "Fl at",
+      "F lat",
+      "▁s ir",
+      "▁si r",
+      "▁sh rimp",
+      "▁shr imp",
+      "区人民检察院 指派检察员",
+      "t N",
+      "治 病",
+      "Po t",
+      "P ot",
+      "全 过程",
+      "避免 了",
+      "▁j ur",
+      "▁ju r",
+      "EN TRY",
+      "ENT RY",
+      "▁M ant",
+      "▁Man t",
+      "▁Ma nt",
+      "Let ter",
+      "L etter",
+      "▁R IGHT",
+      "▁ RIGHT",
+      "rop olis",
+      "▁cin namon",
+      "▁P ediatric",
+      "▁Ped iatric",
+      "黄 豆",
+      "cu m",
+      "c um",
+      "宣判 后",
+      "输出 样例",
+      "▁L ank",
+      "▁La nk",
+      "▁Lan k",
+      "olk ata",
+      "so ever",
+      "nL ocated",
+      "red ient",
+      "exp ressing",
+      "express ing",
+      "expr essing",
+      "半 月",
+      "的本 质",
+      "的 本质",
+      "bul l",
+      "bu ll",
+      "b ull",
+      "▁re public",
+      "▁rep ublic",
+      "▁B ool",
+      "▁Bo ol",
+      "▁ Bool",
+      "▁L ives",
+      "▁Li ves",
+      "▁Live s",
+      "▁Liv es",
+      "二〇一四年 十二月",
+      "会 见",
+      "▁a bst",
+      "▁ab st",
+      "▁abs t",
+      "▁stap le",
+      "Read Only",
+      "奔 腾",
+      "清 澈",
+      "音 质",
+      "IV ER",
+      "IVE R",
+      "I VER",
+      "in en",
+      "ine n",
+      "i nen",
+      "▁E NV",
+      "▁EN V",
+      "▁ ENV",
+      "cons uming",
+      "consum ing",
+      "▁frag rance",
+      "▁fragr ance",
+      "了 下",
+      "BO DY",
+      "B ODY",
+      "▁g uild",
+      "▁gu ild",
+      "▁gui ld",
+      "▁ guild",
+      "杭州 市中级人民法院",
+      "杭州市 中级人民法院",
+      "杭 州市中级人民法院",
+      "社 群",
+      "淮南 市",
+      "淮 南市",
+      "▁V id",
+      "▁Vi d",
+      "▁fin ale",
+      "▁final e",
+      "▁decor ator",
+      "▁ decorator",
+      "▁capac itance",
+      "▁synerg istic",
+      "汤 匙",
+      "胡 某",
+      "▁ 专业",
+      "mo cha",
+      "m ocha",
+      "ob jective",
+      "object ive",
+      "obj ective",
+      "▁fact orial",
+      "▁factor ial",
+      "▁headqu artered",
+      "一件 事",
+      "一 件事",
+      "舞台 上",
+      "舞 台上",
+      "▁T ik",
+      "▁Ti k",
+      "nP age",
+      "n Page",
+      "nAp ple",
+      "nApp le",
+      "n Apple",
+      "▁s quir",
+      "▁squ ir",
+      "▁on wards",
+      "▁del imiter",
+      "▁Import antly",
+      "▁Important ly",
+      "Y X",
+      "人 品",
+      "德 拉",
+      "总 值",
+      "格 子",
+      "鸡 腿",
+      "으 로",
+      "发展 中",
+      "co urt",
+      "cou rt",
+      "c ourt",
+      "▁W ere",
+      "▁We re",
+      "▁Wer e",
+      "▁l ush",
+      "▁lu sh",
+      "Sc alar",
+      "▁cate chol",
+      "▁con ception",
+      "▁concept ion",
+      "▁conce ption",
+      "十 佳",
+      "忍 者",
+      "监 局",
+      "防 空",
+      "▁D N",
+      "▁ DN",
+      "ro tic",
+      "rot ic",
+      "r otic",
+      "DO MAIN",
+      "DOM AIN",
+      "可 控",
+      "fa ker",
+      "fake r",
+      "f aker",
+      "tr ail",
+      "tra il",
+      "t rail",
+      "Set ter",
+      "S etter",
+      "▁ST ART",
+      "▁STAR T",
+      "▁ START",
+      "▁cand les",
+      "▁candle s",
+      "▁as piration",
+      "▁aspir ation",
+      "▁asp iration",
+      "▁W T",
+      "▁ WT",
+      "完全 是",
+      "完 全是",
+      "th irds",
+      "third s",
+      "authent icated",
+      "authenticate d",
+      "▁ 非",
+      "格 斗",
+      "▁D ear",
+      "▁De ar",
+      "▁ Dear",
+      "Cl osed",
+      "Close d",
+      "igraph y",
+      "北 侧",
+      "▁j i",
+      "▁ ji",
+      "ht on",
+      "h ton",
+      "▁vir gin",
+      "下 厨房",
+      "▁occ upy",
+      "▁occup y",
+      "巴 尔",
+      "合格 的",
+      "合 格的",
+      "严重 影响",
+      "▁tr aps",
+      "▁tra ps",
+      "▁trap s",
+      "▁gr ading",
+      "▁grad ing",
+      "▁gra ding",
+      "▁enhance ments",
+      "▁enhancement s",
+      "新 版",
+      "昆 仑",
+      "健身 房",
+      "last Name",
+      "las tName",
+      "▁ultr aviolet",
+      "▁ 康",
+      "他 家",
+      "에 서",
+      "▁ 湖南",
+      "也应 该",
+      "也 应该",
+      "nEx press",
+      "n Express",
+      "各主要 景点的距离如下",
+      "先 端",
+      "喷 涂",
+      "平 和",
+      "日 渐",
+      "院 内",
+      "Hand lers",
+      "Handler s",
+      "Handle rs",
+      "ex plicit",
+      "expl icit",
+      "我有 一",
+      "我 有一",
+      "▁R SV",
+      "▁RS V",
+      "可以 减刑",
+      "开 端",
+      "便宜 的",
+      "Le arn",
+      "L earn",
+      "接 连",
+      "曼 联",
+      "猜 想",
+      "▁m k",
+      "▁ mk",
+      "▁( ((",
+      "▁(( (",
+      "▁ (((",
+      "▁上 海市",
+      "▁上海 市",
+      "▁ 上海市",
+      "▁C ave",
+      "▁Ca ve",
+      "▁Cav e",
+      "▁N amed",
+      "▁Name d",
+      "▁Na med",
+      "▁Nam ed",
+      "▁ Named",
+      "住所地 重庆市",
+      "dir ector",
+      "direct or",
+      "▁Ant arctica",
+      "▁Antarctic a",
+      "频 频",
+      "CX X",
+      "C XX",
+      "▁D w",
+      "第二 季",
+      "不同 的是",
+      "不同的 是",
+      "le vation",
+      "lev ation",
+      "▁ 约",
+      "吃 饱",
+      "大 笑",
+      "ke es",
+      "kee s",
+      "k ees",
+      "▁D uty",
+      "▁Du ty",
+      "in ement",
+      "ine ment",
+      "AP PLICATION",
+      "| :",
+      "▁в ы",
+      "▁H BV",
+      "▁HB V",
+      "▁oste o",
+      "▁ost eo",
+      "▁pe nis",
+      "▁pen is",
+      "Tool tip",
+      "▁ab rupt",
+      "Ver tical",
+      "Vert ical",
+      "▁congreg ation",
+      "▁cong regation",
+      "并 被",
+      "匈牙 利",
+      "英 特尔",
+      "}\\ ',",
+      "}\\' ,",
+      "} \\',",
+      "▁T LS",
+      "▁TL S",
+      "▁ TLS",
+      "▁a in",
+      "▁ai n",
+      "▁ ain",
+      "工程 机械",
+      "容留他人吸毒 罪",
+      "▁nucle ic",
+      "▁nuclei c",
+      "▁Patri ots",
+      "repos itories",
+      "苏 轼",
+      "ro x",
+      "r ox",
+      "不 怎么",
+      "没有 异议",
+      "没 有异议",
+      "输入 样例",
+      "被执行人 的财产",
+      "被执行人的 财产",
+      "nUn ivers",
+      "▁rot ated",
+      "▁rotate d",
+      "organ ized",
+      "亚 科",
+      "引 资",
+      "星 的",
+      "谎 言",
+      "并 根据",
+      "显卡 芯片",
+      "这段 时间",
+      "这 段时间",
+      "▁t ide",
+      "▁ti de",
+      "▁tid e",
+      "▁pol yg",
+      "▁poly g",
+      "AB STRACT",
+      "ABS TRACT",
+      "▁ 顾",
+      "也 比",
+      "底 座",
+      "还 需",
+      "rs a",
+      "r sa",
+      "HT TPS",
+      "HTTP S",
+      "▁Se xy",
+      "▁Sex y",
+      "nf irst",
+      "nfi rst",
+      "n first",
+      "▁sax ophone",
+      "▁calcul ates",
+      "▁calc ulates",
+      "▁calculate s",
+      "np ip",
+      "n pip",
+      "▁Mel issa",
+      ":] )\\",
+      ": ])\\",
+      "am bo",
+      "amb o",
+      "zi er",
+      "zie r",
+      "z ier",
+      "第六 十七",
+      "▁f use",
+      "▁fu se",
+      "▁fus e",
+      "the docs",
+      "thed ocs",
+      "二〇一五年 七月",
+      "▁fig uring",
+      "之 城",
+      "书 桌",
+      "大 到",
+      "署 名",
+      "进行 治疗",
+      "▁W ong",
+      "▁Won g",
+      "▁Wo ng",
+      "tex pect",
+      "t expect",
+      "▁en velop",
+      "▁up holst",
+      "▁Social ist",
+      "▁Soci alist",
+      "▁phot ographic",
+      "▁photo graphic",
+      "▁photograph ic",
+      "位 上",
+      "墙 体",
+      "红 梅",
+      "▁F K",
+      "▁ FK",
+      "min s",
+      "mi ns",
+      "m ins",
+      "ou til",
+      "out il",
+      "o util",
+      "nU nlike",
+      "nUn like",
+      "▁Mar ion",
+      "▁Mario n",
+      "▁Mari on",
+      ";; \\",
+      "; ;\\",
+      "Job s",
+      "Jo bs",
+      "J obs",
+      "▁V ia",
+      "▁Vi a",
+      "优秀 教师",
+      "▁un biased",
+      "素 描",
+      "遵 纪",
+      "el m",
+      "e lm",
+      "可能 在",
+      "可 能在",
+      "马某 某",
+      "马 某某",
+      "裁判 文书",
+      "pit ch",
+      "p itch",
+      "▁h ose",
+      "▁ho se",
+      "ov ered",
+      "ove red",
+      "over ed",
+      "▁s hark",
+      "▁sh ark",
+      "▁sha rk",
+      "▁shar k",
+      "▁java fx",
+      "▁Ev ening",
+      "▁Even ing",
+      "▁Eve ning",
+      "▁amb iguous",
+      "▁ambig uous",
+      "德 斯",
+      "CR P",
+      "C RP",
+      "▁的 值",
+      "▁ 的值",
+      "还 存在",
+      "始终 坚持",
+      "en ums",
+      "enu ms",
+      "enum s",
+      "e nums",
+      "▁Railway s",
+      "▁Rail ways",
+      "▁Luxem bourg",
+      "花 花",
+      "监督 检查",
+      "不知道 怎么",
+      "fe male",
+      "fem ale",
+      "f emale",
+      "pt iles",
+      "p tiles",
+      "act oring",
+      "actor ing",
+      "▁ 但",
+      "少 吃",
+      "王 小",
+      "的 在",
+      "容易 被",
+      "▁b ak",
+      "▁ba k",
+      "▁ bak",
+      "▁dis rupted",
+      "▁disrupt ed",
+      "呼 吁",
+      "通 辽",
+      "青 蛙",
+      "大理 石",
+      "据 统计",
+      "▁Ge V",
+      "▁P on",
+      "▁Po n",
+      "▁al ly",
+      "▁all y",
+      "▁ ally",
+      "▁soph omore",
+      "喜欢 吃",
+      "ep it",
+      "e pit",
+      "ri que",
+      "r ique",
+      "▁M ond",
+      "▁Mon d",
+      "▁Mo nd",
+      "nM odel",
+      "nMod el",
+      "n Model",
+      "í c",
+      "TR Y",
+      "T RY",
+      "▁ 联赛",
+      "在本 院",
+      "在 本院",
+      "配套 设施",
+      "了 大",
+      "人 来",
+      "的 金",
+      "SI M",
+      "S IM",
+      "中国 家",
+      "中 国家",
+      "La w",
+      "L aw",
+      "加油 站",
+      "ju mp",
+      "j ump",
+      "global s",
+      "glob als",
+      "拼音 是",
+      "标的 为",
+      "st ill",
+      "▁Al umin",
+      "▁B ru",
+      "▁Br u",
+      "nm essage",
+      "n message",
+      "符合法律规定的 撤诉条件",
+      "ste am",
+      "s team",
+      "增 减",
+      "结 节",
+      "风 力",
+      "ve z",
+      "v ez",
+      "line r",
+      "li ner",
+      "lin er",
+      "l iner",
+      "▁suff ers",
+      "▁suffer s",
+      "有 可供执行的财产",
+      "M u",
+      "无 辜",
+      "爱 豆",
+      "▁Can ad",
+      "▁Ca nad",
+      "▁vi gil",
+      "▁vig il",
+      "▁dipl om",
+      "▁det rimental",
+      "豪 门",
+      "本人 的",
+      "本 人的",
+      "▁S hi",
+      "▁Sh i",
+      "at tempt",
+      "att empt",
+      "投 机",
+      "雷 斯",
+      "的 观点",
+      "▁w ow",
+      "▁wo w",
+      "▁ wow",
+      "顺 着",
+      "ff t",
+      "f ft",
+      "▁ 运行",
+      "潍坊 市",
+      "帝国 主义",
+      "▁S kip",
+      "▁Sk ip",
+      "▁Ski p",
+      "▁ Skip",
+      "▁abd omen",
+      "水 池",
+      "流 出",
+      "▁IN FO",
+      "▁INF O",
+      "▁ INFO",
+      "pl anes",
+      "plan es",
+      "plane s",
+      "ri cing",
+      "ric ing",
+      "r icing",
+      "os opher",
+      "osoph er",
+      "oso pher",
+      "▁b ipart",
+      "▁bi part",
+      "▁bip art",
+      "填 空",
+      "近 来",
+      "de cl",
+      "dec l",
+      "d ecl",
+      "de ns",
+      "den s",
+      "d ens",
+      "▁\" <<",
+      "▁\"< <",
+      "▁ \"<<",
+      "▁O rt",
+      "▁Or t",
+      "▁in tox",
+      "▁int ox",
+      "▁into x",
+      "▁Rh odes",
+      "▁Rhode s",
+      "▁Rho des",
+      "▁res orts",
+      "▁resort s",
+      "▁migr ated",
+      "▁migrate d",
+      "遏 制",
+      "▁E pic",
+      "▁Ep ic",
+      "▁ru in",
+      "▁pup il",
+      "▁ret ains",
+      "▁retain s",
+      "▁D ifferences",
+      "▁Difference s",
+      "奇 艺",
+      "荷 花",
+      "ib i",
+      "i bi",
+      "He ap",
+      "▁un m",
+      "▁u nm",
+      "开展 经营活动",
+      "普 京",
+      "深 渊",
+      "自 理",
+      "]} ,",
+      "] },",
+      "两人 的",
+      "两 人的",
+      "Pe er",
+      "P eer",
+      "▁reson ant",
+      ") &",
+      "绝 不",
+      "谣 言",
+      "($ \"",
+      "( $\"",
+      "▁ 常",
+      "▁s v",
+      "▁ sv",
+      "({ },",
+      "( {},",
+      "ji an",
+      "jia n",
+      "j ian",
+      "To One",
+      "▁Cel tic",
+      "相应 价值的财产",
+      "▁mod ifier",
+      "▁ 辅料",
+      "养殖 业",
+      "eg ro",
+      "egr o",
+      "e gro",
+      "nF re",
+      "nFr e",
+      "n Fre",
+      "▁C IA",
+      "▁CI A",
+      "▁de bit",
+      "▁deb it",
+      "re porter",
+      "rep orter",
+      "report er",
+      "▁gas oline",
+      "sc i",
+      "s ci",
+      "sk a",
+      "s ka",
+      "▁G F",
+      "▁ GF",
+      "▁ 牺牲",
+      "幸福 感",
+      "教学 楼",
+      "看出 来",
+      "看 出来",
+      "国 债",
+      "的 头",
+      "管理 处",
+      "▁I on",
+      "▁Io n",
+      "▁H annah",
+      "▁Han nah",
+      "▁Hann ah",
+      "Vis ibility",
+      "r z",
+      "▁n M",
+      "▁ nM",
+      "向量 的",
+      "向 量的",
+      "▁S OL",
+      "▁SO L",
+      "▁ SOL",
+      "▁turb ulent",
+      "图 解",
+      "重要 性",
+      "大家 可以",
+      "▁c rest",
+      "▁cre st",
+      "▁cr est",
+      "▁ crest",
+      "▁B oeing",
+      "▁ag grav",
+      "▁p addle",
+      "▁padd le",
+      "▁V ARCHAR",
+      "▁comp utes",
+      "▁comput es",
+      "▁compute s",
+      "▁neighb ours",
+      "▁neighbour s",
+      "单 曲",
+      "J eff",
+      "ov ern",
+      "ove rn",
+      "over n",
+      "o vern",
+      "志愿服务 队",
+      "▁nause a",
+      "风 气",
+      "sub plot",
+      "▁Batt alion",
+      "in p",
+      "i np",
+      "▁s ph",
+      "▁sp h",
+      "Qu ote",
+      "▁n ood",
+      "▁no od",
+      "▁as sim",
+      "▁ass im",
+      "▁d imer",
+      "▁di mer",
+      "▁dim er",
+      "▁Th ermal",
+      "▁Ther mal",
+      "▁Therm al",
+      "SO L",
+      "S OL",
+      "gr af",
+      "gra f",
+      "g raf",
+      "国 华",
+      "抗 震",
+      "自己 也",
+      "Check s",
+      "Che cks",
+      "▁str okes",
+      "▁stroke s",
+      "▁stro kes",
+      "的 刑罚",
+      "都不 敢",
+      "都 不敢",
+      "ME TA",
+      "MET A",
+      "M ETA",
+      "▁B AL",
+      "▁BA L",
+      "▁P ione",
+      "▁Pi one",
+      "▁t roph",
+      "▁tr oph",
+      "▁tro ph",
+      "▁trop h",
+      "mock ito",
+      "▁get All",
+      "▁ getAll",
+      "微 创",
+      "无 为",
+      "情况 说明",
+      "nd one",
+      "ndo ne",
+      "ndon e",
+      "n done",
+      "ow itz",
+      "o witz",
+      "▁First ly",
+      "海 波",
+      "ub in",
+      "ubi n",
+      "u bin",
+      "▁o sm",
+      "▁os m",
+      "▁ osm",
+      "▁bl owing",
+      "▁blow ing",
+      "▁blo wing",
+      "▁R ational",
+      "▁Rat ional",
+      "▁Ratio nal",
+      "▁grand son",
+      "▁hard ness",
+      "condition al",
+      "cond itional",
+      "▁hemis phere",
+      "途 经",
+      "▁N N",
+      "▁ NN",
+      "▁t omb",
+      "▁to mb",
+      "▁tom b",
+      "党的 十九大",
+      "执行机关 提供的",
+      "是人 类",
+      "是 人类",
+      "办公室 副主任",
+      "▁manif estation",
+      "▁manifest ation",
+      "无 心",
+      "▁t b",
+      "▁ tb",
+      "nl ist",
+      "n list",
+      "▁L amp",
+      "▁La mp",
+      "▁Lam p",
+      "▁n arc",
+      "▁na rc",
+      "H s",
+      "因为 这个",
+      "▁St af",
+      "▁Sta f",
+      "Decor ator",
+      "▁Kazakh stan",
+      "查 证",
+      "▁h ob",
+      "▁ho b",
+      "PER TY",
+      "▁L ogo",
+      "▁Log o",
+      "▁Lo go",
+      "▁ Logo",
+      "it ures",
+      "iture s",
+      "itu res",
+      "i tures",
+      "▁m RNAs",
+      "▁mRNA s",
+      "照 样",
+      "私 募",
+      "产 阶级",
+      "这一 天",
+      "这 一天",
+      "专业 学习",
+      "完成 任务",
+      "West ern",
+      "W estern",
+      "dr ivers",
+      "driver s",
+      "driv ers",
+      "drive rs",
+      "d rivers",
+      "▁ant ire",
+      "▁anti re",
+      "▁SU MMARY",
+      "第一百四十五 条的规定",
+      "第一百四十五条 的规定",
+      "▁perturb ations",
+      "▁perturbation s",
+      "▁Writ ten",
+      "▁Wr itten",
+      "Com e",
+      "Co me",
+      "C ome",
+      "两个 整数",
+      "▁Ex ped",
+      "▁Exp ed",
+      "join ing",
+      "jo ining",
+      "经 庭审",
+      "▁cons ortium",
+      "▁Scholar ship",
+      "▁St aphylococcus",
+      "墙 上",
+      "▁ 版本",
+      "▁Per cent",
+      "▁Perc ent",
+      "▁ Percent",
+      "▁serv icing",
+      "▁neuro science",
+      "河 西",
+      "社 长",
+      "▁H I",
+      "▁ HI",
+      "选择 出",
+      "as aki",
+      "asa ki",
+      "▁anten nas",
+      "▁antenna s",
+      "J B",
+      "奠 定",
+      "打 动",
+      "法 医",
+      "th ought",
+      "though t",
+      "▁el dest",
+      "▁eld est",
+      "高级人民法院 于",
+      "名 气",
+      "英 镑",
+      "▁P SA",
+      "▁PS A",
+      "▁b amboo",
+      "nUp dated",
+      "nUpdate d",
+      "n Updated",
+      "人 称",
+      "水 泵",
+      "al ks",
+      "alk s",
+      "ta king",
+      "t aking",
+      "▁D odge",
+      "▁Dod ge",
+      "▁R efer",
+      "▁Re fer",
+      "▁Ref er",
+      "新冠肺炎 疫情",
+      "▁S urgical",
+      "▁disturb ing",
+      "之 夜",
+      "老 龄",
+      "淋巴 结",
+      "aw an",
+      "awa n",
+      "a wan",
+      "▁B om",
+      "▁Bo m",
+      "▁Ham mer",
+      "▁Hamm er",
+      "▁Pal estine",
+      "▁Palest ine",
+      "物 权",
+      "设置 了",
+      "\"# \">",
+      "\"#\" >",
+      "]{ \\\\",
+      "] {\\\\",
+      "▁game rs",
+      "▁gam ers",
+      "▁ga mers",
+      "▁seg regation",
+      "独有 的",
+      "独 有的",
+      "的 资金",
+      "重大 的",
+      "重 大的",
+      "国家 图书馆",
+      "ga ther",
+      "g ather",
+      "有下列情形 之一的",
+      "▁autom ate",
+      "▁auto mate",
+      "nInd ividual",
+      "无 缘",
+      "Lu c",
+      "L uc",
+      "学习 了",
+      "Al tern",
+      "Alter n",
+      "Alt ern",
+      "▁Pres idential",
+      "▁President ial",
+      "▁Mono Behaviour",
+      "才 行",
+      "nJ o",
+      "n Jo",
+      "满满 的",
+      "的 性格",
+      "▁文学 体裁",
+      "ess ential",
+      "专 心",
+      "▁c asc",
+      "▁ca sc",
+      "▁cas c",
+      "Tick et",
+      "T icket",
+      "▁S ultan",
+      "▁Sul tan",
+      "v x",
+      "两 周",
+      "改 性",
+      "▁ 周围景观",
+      "nSE LECT",
+      "n SELECT",
+      "打 字",
+      "鱼 肉",
+      "的 身体",
+      "可以 得到",
+      "▁Cl ara",
+      "▁Clar a",
+      "ategor ized",
+      "宜 春",
+      "淮 安",
+      "依法 可",
+      "故本 选项错误",
+      "▁terror ists",
+      "▁terrorist s",
+      "▁biod iversity",
+      "超 高",
+      "首 付",
+      "膝 关节",
+      "m RNA",
+      "gr ant",
+      "gra nt",
+      "gran t",
+      "g rant",
+      "▁n avy",
+      "▁nav y",
+      "▁na vy",
+      "UT ION",
+      "▁Her oes",
+      "▁Hero es",
+      "▁con versions",
+      "▁convers ions",
+      "▁conversion s",
+      "玲 珑",
+      "最 主要",
+      "▁Or ders",
+      "▁Order s",
+      "▁Ord ers",
+      "▁ Orders",
+      "▁bed ding",
+      "调 和",
+      "ba um",
+      "▁P owers",
+      "▁Power s",
+      "▁Po wers",
+      "▁Pow ers",
+      "双 色",
+      "不仅 能",
+      "Ni ck",
+      "N ick",
+      "农业 银行",
+      "ass ic",
+      "▁C atch",
+      "▁Cat ch",
+      "本院 予以采纳",
+      "本院予以 采纳",
+      "Tool bar",
+      "公开 发",
+      "公 开发",
+      "卖 的",
+      "固 态",
+      "▁ 没有",
+      "第一 届",
+      "nW ar",
+      "n War",
+      "▁t rab",
+      "▁tr ab",
+      "▁tra b",
+      "ar coma",
+      "arc oma",
+      "En h",
+      "E nh",
+      "最 开始",
+      "▁H olly",
+      "▁Hol ly",
+      "▁Holl y",
+      "ake dirs",
+      "aked irs",
+      "会 同",
+      "地 主",
+      "大 便",
+      "恩 施",
+      "车 展",
+      "国资 委",
+      "AR GS",
+      "ARG S",
+      "▁aw ake",
+      "▁awa ke",
+      "blog spot",
+      "blogs pot",
+      "江 上",
+      "特 约",
+      "鲜 美",
+      "因为 她",
+      "因 为她",
+      "$ ;",
+      "月 薪",
+      "To o",
+      "T oo",
+      "本来 就是",
+      "本来就 是",
+      "Component Model",
+      "莫名其 妙",
+      "le ast",
+      "lea st",
+      "l east",
+      "是怎么 回事",
+      "是 怎么回事",
+      "▁High ly",
+      "▁sk ipped",
+      "▁skip ped",
+      "▁ski pped",
+      "▁Member ship",
+      "▁Members hip",
+      "体 能",
+      "题 解",
+      "所以 要",
+      "se lective",
+      "select ive",
+      "sel ective",
+      "▁th irteen",
+      "イ ン",
+      "招 呼",
+      "el n",
+      "e ln",
+      "子 序列",
+      "中的 作用",
+      "中 的作用",
+      "生长 发育",
+      "th ttp",
+      "t http",
+      "nNot ice",
+      "n Notice",
+      "▁profit ability",
+      "先 是",
+      "基 亚",
+      "总 分",
+      "浮 动",
+      "爵 士",
+      "▁F uj",
+      "▁Fu j",
+      "▁O VER",
+      "▁ OVER",
+      "向本院 申请再审",
+      "向本院申请 再审",
+      "持 卡",
+      "科 级",
+      "但 没有",
+      "or ously",
+      "orous ly",
+      "分 隔",
+      "去 买",
+      "片 刻",
+      "家 企业",
+      "的 参数",
+      "ra nt",
+      "ran t",
+      "r ant",
+      "公告 送达",
+      "▁in doors",
+      "▁ind oors",
+      "▁indoor s",
+      "Sp l",
+      "S pl",
+      "你就 会",
+      "你 就会",
+      "健康 成长",
+      "的工作 人员",
+      "的 工作人员",
+      "pect ral",
+      "▁blog gers",
+      "▁blogger s",
+      "▁behaviour al",
+      "▁Measure ments",
+      "▁Measurement s",
+      "说 得",
+      "量 表",
+      "▁' %",
+      "的自 然",
+      "的 自然",
+      "还是 一",
+      "还 是一",
+      "▁t rek",
+      "▁tr ek",
+      "▁tre k",
+      "▁分布 范围",
+      "回顾 性分析",
+      "初 衷",
+      "孤 儿",
+      "葱 花",
+      "▁In tro",
+      "▁Int ro",
+      "▁Intr o",
+      "▁ch ars",
+      "▁char s",
+      "▁cha rs",
+      "▁ chars",
+      "IB ILITY",
+      "二 中",
+      "透 视",
+      "▁F D",
+      "▁ FD",
+      "忘记 了",
+      "▁I CC",
+      "▁IC C",
+      "▁ ICC",
+      "二〇一五年 八月",
+      "则 为",
+      "▁法 国",
+      "▁ 法国",
+      "▁L enn",
+      "▁Le nn",
+      "▁Len n",
+      "nWith out",
+      "n Without",
+      "▁w inding",
+      "▁win ding",
+      "▁wind ing",
+      "▁comp anions",
+      "▁companion s",
+      "▁P resentation",
+      "▁Present ation",
+      "域 为",
+      "领 悟",
+      "Case s",
+      "Ca ses",
+      "Cas es",
+      "C ases",
+      "sc alar",
+      "scal ar",
+      "scala r",
+      "▁gift ed",
+      "▁assembl ies",
+      "h k",
+      "到 家",
+      "北 宋",
+      "▁h p",
+      "▁ hp",
+      "▁so bre",
+      "▁Re duction",
+      "▁Red uction",
+      "火 了",
+      "寻衅 滋事",
+      "▁f airy",
+      "▁fa iry",
+      "▁fair y",
+      "ish able",
+      "isha ble",
+      "▁st arch",
+      "▁star ch",
+      "▁ut most",
+      "▁voc alist",
+      "▁vocal ist",
+      "天 长",
+      "是 老",
+      "ug o",
+      "u go",
+      "▁p tr",
+      "▁pt r",
+      "▁ ptr",
+      "▁sp reads",
+      "▁spread s",
+      "▁s outheastern",
+      "▁south eastern",
+      "▁southeast ern",
+      "专 人",
+      "药 师",
+      "青 青",
+      "ye t",
+      "y et",
+      "ad ay",
+      "ada y",
+      "a day",
+      "▁Sk ype",
+      "▁Sky pe",
+      "取决于 被执行人",
+      "国家 自然科学基金",
+      "▁Lithuan ia",
+      "温暖 的",
+      "方面 的知识",
+      "方面的 知识",
+      "Mar tin",
+      "Mart in",
+      "▁acquisition s",
+      "▁acquis itions",
+      "ч и",
+      "并 行",
+      "煮 熟",
+      "▁K w",
+      "只能 在",
+      "只 能在",
+      "x html",
+      "or sche",
+      "ors che",
+      "▁k icking",
+      "▁kick ing",
+      "▁Fergus on",
+      "▁research ed",
+      "▁resear ched",
+      "▁mathematic ians",
+      "▁mathematician s",
+      "抛 光",
+      "▁U ri",
+      "▁Ur i",
+      "▁ Uri",
+      "竞争 力的",
+      "竞争力 的",
+      "▁Ch arge",
+      "▁Char ge",
+      "▁Charg e",
+      "▁Pur ple",
+      "▁aug mentation",
+      "▁augment ation",
+      "苯丙 胺",
+      "▁W et",
+      "▁We t",
+      "R V",
+      "五 四",
+      "着 自己",
+      "CK ER",
+      "▁S emi",
+      "▁Se mi",
+      "▁Sem i",
+      "▁t oes",
+      "▁to es",
+      "▁toe s",
+      "ali ases",
+      "alias es",
+      "alia ses",
+      "▁syn onym",
+      "乱 世",
+      "RA W",
+      "R AW",
+      "▁C RM",
+      "▁CR M",
+      "▁ CRM",
+      "▁sub family",
+      "▁Jackson ville",
+      "三 元",
+      "承 揽",
+      "还 包括",
+      "Mi ke",
+      "M ike",
+      "违约 责任",
+      "tu ser",
+      "t user",
+      "Un icode",
+      "Over flow",
+      "丰 满",
+      "在 第",
+      "▁r ode",
+      "▁ro de",
+      "▁rod e",
+      "leg acy",
+      "▁g ravel",
+      "▁gr avel",
+      "▁grave l",
+      "▁gra vel",
+      "▁grav el",
+      "Package s",
+      "Pack ages",
+      "年 中",
+      "窗 外",
+      "到 案经过",
+      "▁Cur tis",
+      "▁Curt is",
+      "▁Pan thers",
+      "▁Const ants",
+      "▁Constant s",
+      "▁ Constants",
+      "▁altern atively",
+      "▁alternative ly",
+      "扫 地",
+      "准确 性",
+      "▁cos mic",
+      "▁cosm ic",
+      "▁f ading",
+      "▁fa ding",
+      "t display",
+      "▁es ophag",
+      "▁layout s",
+      "▁lay outs",
+      "▁ layouts",
+      "▁coll ects",
+      "▁collect s",
+      "▁colle cts",
+      "不 掉",
+      "科 的",
+      "Si n",
+      "S in",
+      "fi f",
+      "f if",
+      "债权 债务",
+      "▁PR OF",
+      "▁PRO F",
+      "▁Fun eral",
+      "▁vom iting",
+      "外 在",
+      "nO ld",
+      "n Old",
+      "np ost",
+      "n post",
+      "en zyme",
+      "enz yme",
+      "▁own ing",
+      "▁ow ning",
+      "主 权",
+      "传 入",
+      "轻 重",
+      "▁规格 参数",
+      "c andidate",
+      "▁adv isors",
+      "▁advis ors",
+      "▁advisor s",
+      "狂 欢",
+      "ed d",
+      "e dd",
+      "▁sl ack",
+      "▁ slack",
+      "会 说",
+      "供 需",
+      "吉 尔",
+      "期 为",
+      "构成 了",
+      "构 成了",
+      "ma gn",
+      "mag n",
+      "m agn",
+      "om eth",
+      "ome th",
+      "omet h",
+      "o meth",
+      "▁Bl air",
+      "Sub scriber",
+      "Subscribe r",
+      "▁ ω",
+      "国外 的",
+      "国 外的",
+      "fl ake",
+      "f lake",
+      "serial izers",
+      "serializer s",
+      "serialize rs",
+      "哈 利",
+      "白 马",
+      "}) )",
+      "} ))",
+      "▁ 配置",
+      "专科 学",
+      "专 科学",
+      "类 游戏",
+      "通过 网络",
+      "不 听",
+      "水 文",
+      "副 处长",
+      "网站 建设",
+      "▁T ail",
+      "▁Ta il",
+      "▁Tai l",
+      "被执行人 杨",
+      "nE vent",
+      "nEven t",
+      "nEv ent",
+      "n Event",
+      "变 身",
+      "高 空",
+      "仔细 地",
+      "告诉 他",
+      "概率 为",
+      "概 率为",
+      "爱的 人",
+      "爱 的人",
+      "由原告 承担",
+      "▁Ca sey",
+      "▁Cas ey",
+      "▁Case y",
+      "二〇一七年 四月",
+      "心 跳",
+      "这 名",
+      "TL S",
+      "T LS",
+      "in sk",
+      "ins k",
+      "▁da emon",
+      "▁ daemon",
+      "服务 商",
+      "▁D ew",
+      "▁De w",
+      "其行为已构成 危险驾驶罪",
+      "No rm",
+      "Nor m",
+      "N orm",
+      "▁* )\\",
+      "▁*) \\",
+      "▁ *)\\",
+      "bai du",
+      "b aidu",
+      "▁en umer",
+      "▁enum er",
+      "ex pressed",
+      "exp ressed",
+      "express ed",
+      "expr essed",
+      "pr imes",
+      "pri mes",
+      "prime s",
+      "prim es",
+      "pro per",
+      "pr oper",
+      "prop er",
+      "▁hom omorphism",
+      "街 上",
+      "在外 面",
+      "在 外面",
+      "▁P CA",
+      "▁PC A",
+      "▁ PCA",
+      "pl ets",
+      "ple ts",
+      "plet s",
+      "p lets",
+      "size s",
+      "si zes",
+      "s izes",
+      "▁R unner",
+      "▁Run ner",
+      "▁ Runner",
+      "▁Teh ran",
+      "ang ements",
+      "ange ments",
+      "angement s",
+      "ent ricular",
+      "entric ular",
+      "私 下",
+      "▁H PLC",
+      "▁HP LC",
+      "▁ HPLC",
+      "▁indic ative",
+      "收录 于",
+      "考试 的",
+      "as in",
+      "asi n",
+      "a sin",
+      "因本案 于",
+      "复合 材料",
+      "▁lept in",
+      "▁Hy derabad",
+      "名 著",
+      "辛 勤",
+      "ov i",
+      "o vi",
+      "▁N GO",
+      "▁NG O",
+      "修 真",
+      "St amp",
+      "la ravel",
+      "lar avel",
+      "▁meth ane",
+      "nWh ite",
+      "n White",
+      "▁S igma",
+      "▁Sig ma",
+      "▁ Sigma",
+      "▁Incre ased",
+      "▁Increase d",
+      "▁Appro ximately",
+      "@ \\",
+      "气 压",
+      "藏 在",
+      "专门 的",
+      "专 门的",
+      "▁S EE",
+      "▁SE E",
+      "经济 林果",
+      "上 前",
+      "升 温",
+      "变 焦",
+      "均 已",
+      "um ming",
+      "umm ing",
+      "▁f etus",
+      "▁fet us",
+      "▁hum id",
+      "▁hu mid",
+      "▁lock down",
+      "▁B iological",
+      "▁Bi ological",
+      "▁Bio logical",
+      "▁ground water",
+      "▁intent ional",
+      "▁intention al",
+      "元 和",
+      "对 人",
+      "课 文",
+      "Fl y",
+      "F ly",
+      "ro mb",
+      "rom b",
+      "r omb",
+      "感受 到了",
+      "感受到 了",
+      "感 受到了",
+      "▁incub ated",
+      "加 点",
+      "\"} ]",
+      "\" }]",
+      "▁e V",
+      "这么 大",
+      "▁Re new",
+      "▁Ren ew",
+      "ac ement",
+      "ace ment",
+      "bi ology",
+      "bio logy",
+      "b iology",
+      "▁b ibliography",
+      "▁bibli ography",
+      "▁film maker",
+      "▁filmm aker",
+      "莱 斯",
+      "司法 局",
+      "一个 数的",
+      "一个数 的",
+      "北京 赛车",
+      "质量 管理",
+      "副主任 医师",
+      "副 主任医师",
+      "▁h alls",
+      "▁hall s",
+      "▁hal ls",
+      "Comp ute",
+      "Comput e",
+      "res cent",
+      "r escent",
+      "File System",
+      "▁allerg ies",
+      "▁Incre asing",
+      "下 肢",
+      "但 凡",
+      "后 将",
+      "烘 焙",
+      "Su gg",
+      "S ugg",
+      "年 上半年",
+      "sequ ences",
+      "sequence s",
+      "过 头",
+      "bi as",
+      "bia s",
+      "b ias",
+      "在案 佐证",
+      "法律 援助",
+      "▁D uck",
+      "▁Du ck",
+      "▁Duc k",
+      "ac lass",
+      "acl ass",
+      "a class",
+      "▁Ken ny",
+      "▁Kenn y",
+      "项 链",
+      "▁{ }\\'",
+      "▁{}\\ '",
+      "▁{} \\'",
+      "▁ {}\\'",
+      "债权 的实现",
+      "\\' )),\\",
+      "\\') ),\\",
+      "\\')) ,\\",
+      "\\ ')),\\",
+      "养 猪",
+      "嘉 定",
+      "欧 冠",
+      "fr o",
+      "f ro",
+      "▁к а",
+      "▁ ка",
+      "com pressed",
+      "comp ressed",
+      "compress ed",
+      "▁ 号",
+      "他 妈",
+      "器 具",
+      "▁O D",
+      "▁ OD",
+      "动 不动",
+      "海 内外",
+      "第十 三条",
+      "第十三 条",
+      "第 十三条",
+      "BU FFER",
+      "▁Ad rian",
+      "▁Adri an",
+      "▁story telling",
+      "▁instruction al",
+      "▁instruct ional",
+      "偏 好",
+      "波 澜",
+      "炒 股",
+      ")} {(",
+      ")}{ (",
+      ") }{(",
+      "El se",
+      "▁wid gets",
+      "▁widget s",
+      "▁ widgets",
+      "▁myth ology",
+      "意 境",
+      "no h",
+      "n oh",
+      "ne eds",
+      "need s",
+      "nee ds",
+      "n eeds",
+      "▁Dise ases",
+      "▁Disease s",
+      "▁Rodr iguez",
+      "▁trans membrane",
+      "▁ г",
+      "—— —",
+      "— ——",
+      "一个 问题",
+      "▁hook ed",
+      "▁ho oked",
+      "▁Pod cast",
+      "▁manuscript s",
+      "十 余",
+      "她 也",
+      "阶 梯",
+      "/\\\\ /",
+      "/ \\\\/",
+      "▁D ON",
+      "▁DO N",
+      "▁Mar x",
+      "▁Ma rx",
+      "▁im mers",
+      "▁imm ers",
+      "▁ immers",
+      "▁reprodu c",
+      "▁repro duc",
+      "▁R aspberry",
+      "表 态",
+      "可 不是",
+      "\\' \"\\",
+      "\\'\" \\",
+      "ht ub",
+      "h tub",
+      "def er",
+      "de fer",
+      "d efer",
+      "▁N oise",
+      "▁No ise",
+      "W V",
+      "盘 点",
+      "]\" ,",
+      "] \",",
+      "zo n",
+      "z on",
+      "还不 够",
+      "还 不够",
+      "nR est",
+      "nRe st",
+      "nRes t",
+      "n Rest",
+      "▁EV EN",
+      "▁dign ity",
+      "▁Like wise",
+      "▁Lik ewise",
+      "▁high ways",
+      "▁highway s",
+      "当 晚",
+      "这 话",
+      "间 盘",
+      "ap per",
+      "app er",
+      "appe r",
+      "▁Br ady",
+      "▁Brad y",
+      "▁Bra dy",
+      "所 欠",
+      "RO L",
+      "R OL",
+      "中国 农业银行",
+      "中国农业 银行",
+      "▁mark up",
+      "▁ markup",
+      "Ex ceptions",
+      "Exception s",
+      "苦 难",
+      "wa ld",
+      "wal d",
+      "w ald",
+      "es tim",
+      "est im",
+      "e stim",
+      "加 拉",
+      "后 就",
+      "辨 识",
+      "遥 远",
+      "第九 十七条",
+      "ex posed",
+      "exp osed",
+      "par alle",
+      "pret tier",
+      "▁ ö",
+      "元 旦",
+      "较 短",
+      "CR C",
+      "C RC",
+      "rd f",
+      "r df",
+      "normal ized",
+      "normalize d",
+      "K a",
+      "异 能",
+      "面 具",
+      "的 元素",
+      "艺术 学院",
+      "UST OM",
+      "m igrate",
+      "▁un install",
+      "▁u ninstall",
+      "▁Arch bishop",
+      "▁Spring field",
+      "赞 赏",
+      "异议 人",
+      "的 广告",
+      "▁bal let",
+      "▁ball et",
+      "▁Eth ereum",
+      "file system",
+      "files ystem",
+      "▁negot iation",
+      "sp i",
+      "s pi",
+      "高中 生",
+      "高 中生",
+      "▁met am",
+      "▁meta m",
+      "▁Graph QL",
+      "▁mill iseconds",
+      "Cs s",
+      "C ss",
+      "脱颖 而出",
+      "av ings",
+      "aving s",
+      "avin gs",
+      "B etween",
+      "死 于",
+      "脱 发",
+      "在 接受",
+      "在 银行",
+      "▁ap ex",
+      "F N",
+      "您 选择",
+      "▁# ##\\",
+      "▁## #\\",
+      "▁### \\",
+      "▁ ###\\",
+      "as than",
+      "ast han",
+      "▁b loss",
+      "▁bl oss",
+      "▁blo ss",
+      "▁pat ri",
+      "▁pa tri",
+      "▁rod ents",
+      "▁rodent s",
+      "生 猪",
+      "▁V AT",
+      "▁VA T",
+      "ju ven",
+      "▁F ight",
+      "▁Fig ht",
+      "▁Fi ght",
+      "▁irre versible",
+      "▁z a",
+      "▁ za",
+      "动物 门",
+      "nL ead",
+      "nLe ad",
+      "n Lead",
+      "▁aut ophagy",
+      "七 星",
+      "体会 到",
+      "▁e lit",
+      "▁el it",
+      "会 比",
+      "溶 于",
+      "不了 的",
+      "不 了的",
+      "mi me",
+      "m ime",
+      "nS yn",
+      "n Syn",
+      "做的 事情",
+      "做的事 情",
+      "做 的事情",
+      "pat ched",
+      "patch ed",
+      "p atched",
+      "▁EX ISTS",
+      "▁cult ivated",
+      "▁cultiv ated",
+      "牢 牢",
+      "种 质",
+      "C tx",
+      "无线 电",
+      "艾滋 病",
+      "ad am",
+      "ada m",
+      "a dam",
+      "ee ds",
+      "eed s",
+      "e eds",
+      "▁t hy",
+      "▁th y",
+      "▁ thy",
+      "▁rock y",
+      "▁vict ories",
+      "巴 斯",
+      "油 价",
+      "ni p",
+      "n ip",
+      "求 实数",
+      "纤维 素",
+      "自动 驾驶",
+      "diff usion",
+      "像 个",
+      "IN C",
+      "I NC",
+      "`` \\",
+      "` `\\",
+      "▁W C",
+      "▁ WC",
+      "营业 部",
+      "in ky",
+      "ink y",
+      "质量 问题",
+      "monitor ing",
+      "nD ifferent",
+      "nDiff erent",
+      "n Different",
+      "▁re perfusion",
+      "▁reper fusion",
+      "▁ 特",
+      "不仅 要",
+      "的 函数",
+      "的 道路",
+      "▁bu rial",
+      "▁bur ial",
+      "打 死",
+      "漫 长",
+      "▁W WE",
+      "▁WW E",
+      "详细 介绍",
+      "▁S cout",
+      "▁Sc out",
+      "▁six teen",
+      "大 厂",
+      "拉 着",
+      "立 新",
+      "老 是",
+      "我们 一起",
+      "提出 减刑建议书",
+      "set Value",
+      "点 数",
+      "谴 责",
+      "韧 性",
+      "家长 们",
+      "▁F ear",
+      "▁Fe ar",
+      "Fin ish",
+      "movie s",
+      "mov ies",
+      "m ovies",
+      "▁a utor",
+      "▁aut or",
+      "▁auto r",
+      "▁au tor",
+      "trans parent",
+      "一 字",
+      "nU ser",
+      "nUse r",
+      "nUs er",
+      "n User",
+      "ycl ing",
+      "yc ling",
+      "可 笑",
+      "调 侃",
+      "万元 整",
+      "万 元整",
+      "亚 热带",
+      "二次根 式的",
+      "二次根式 的",
+      "ic iary",
+      "ici ary",
+      "icia ry",
+      "▁it ertools",
+      "▁iter tools",
+      "州 的",
+      "湖 泊",
+      "▁E q",
+      "▁ Eq",
+      "元 器件",
+      "▁M ales",
+      "▁Mal es",
+      "▁Ma les",
+      "▁Male s",
+      "宁 愿",
+      "读 题",
+      "▁i z",
+      "▁ iz",
+      "一般 人",
+      "洗衣 间",
+      "es so",
+      "ess o",
+      "mp eg",
+      "m peg",
+      "▁su prem",
+      "▁sup rem",
+      ") '",
+      "购 进",
+      "企业 发展",
+      "查控 系统",
+      "get Text",
+      "▁ad duct",
+      "▁add uct",
+      "连 连",
+      "都 为",
+      "nB i",
+      "n Bi",
+      "三十 年",
+      "三 十年",
+      ")\\\\ ,",
+      ") \\\\,",
+      "▁Sch midt",
+      "il las",
+      "ill as",
+      "illa s",
+      "▁C omic",
+      "▁Com ic",
+      "▁Co mic",
+      "Gener ation",
+      "Gen eration",
+      "Gene ration",
+      "G eneration",
+      "▁colon ization",
+      "入 场",
+      "向 法庭",
+      "nE ff",
+      "n Eff",
+      "纪委 书记",
+      "纪 委书记",
+      "ress or",
+      "J M",
+      "an dr",
+      "and r",
+      "▁G os",
+      "▁Go s",
+      "▁G ut",
+      "▁Gu t",
+      "政府 信息公开",
+      "In stant",
+      "Inst ant",
+      "▁Lanc aster",
+      "▁Lan caster",
+      "卡 的",
+      "开户 行",
+      "结果 的",
+      "ig ible",
+      "igi ble",
+      "▁F IRST",
+      "▁Vol ks",
+      "uit able",
+      "ui table",
+      "u itable",
+      "地 将",
+      "灾 区",
+      "▁注 解",
+      "nJ on",
+      "nJo n",
+      "n Jon",
+      "神经 网络",
+      "prov ide",
+      "▁re flux",
+      "▁ref lux",
+      "中 枢",
+      "保险 箱",
+      "几年 的",
+      "几 年的",
+      "▁Gram my",
+      "▁inter personal",
+      "无 事",
+      "▁Last ly",
+      "▁b izarre",
+      "▁Dis count",
+      "▁Disc ount",
+      "▁aster oid",
+      "塑 性",
+      "▁S olo",
+      "▁So lo",
+      "▁Sol o",
+      "ind say",
+      "inds ay",
+      "os lovak",
+      "med iately",
+      "mediate ly",
+      "▁merch ants",
+      "▁merchant s",
+      "get Attribute",
+      "孕 期",
+      "钻 研",
+      "劳人 仲",
+      "d aughter",
+      "▁dent istry",
+      "▁dentist ry",
+      "De ad",
+      "D ead",
+      "▁Y ii",
+      "▁Yi i",
+      "▁as ylum",
+      "▁r ugged",
+      "▁rug ged",
+      "nBase ball",
+      "▁prophyl axis",
+      "▁antioxid ants",
+      "▁antioxidant s",
+      "摄 氏",
+      "sr v",
+      "s rv",
+      "us b",
+      "u sb",
+      "nP al",
+      "n Pal",
+      "▁G ul",
+      "▁Gu l",
+      "分析 方法",
+      "如何 选择",
+      "ew are",
+      "e ware",
+      "▁P ale",
+      "▁Pal e",
+      "▁Pa le",
+      "▁g rac",
+      "▁gr ac",
+      "▁gra c",
+      "▁inf used",
+      "地 表",
+      "先后 在",
+      "先 后在",
+      "和 美国",
+      "▁re r",
+      "▁r er",
+      "▁ rer",
+      "因与 被申请人",
+      "因 与被申请人",
+      "▁a waiting",
+      "▁await ing",
+      "▁awa iting",
+      "▁sub traction",
+      "▁subtract ion",
+      "在地 上",
+      "在 地上",
+      "语言 的",
+      "▁Filip ino",
+      "▁det ached",
+      "▁advert ised",
+      "▁advertis ed",
+      "▁advertise d",
+      ", …",
+      "其 对",
+      "给 别人",
+      "in ia",
+      "ini a",
+      "i nia",
+      "▁B rick",
+      "▁Br ick",
+      "▁Bri ck",
+      "aur ants",
+      "aurant s",
+      "交叉 口",
+      "nM arch",
+      "nMar ch",
+      "n March",
+      "▁Rem oval",
+      "他 要",
+      "环 卫",
+      "采 暖",
+      "FL AG",
+      "du ty",
+      "d uty",
+      "▁$$ |",
+      "据 新华社",
+      "ch emy",
+      "che my",
+      "chem y",
+      "c hemy",
+      "tv alue",
+      "t value",
+      "br anches",
+      "branch es",
+      "▁l aptops",
+      "▁lap tops",
+      "▁laptop s",
+      ") ”",
+      "以 罪犯",
+      "相对 来说",
+      "struct ures",
+      "structure s",
+      "每 隔",
+      "穿 的",
+      "▁k b",
+      "▁ kb",
+      "内存 类型",
+      "土地 管理法",
+      "▁H ands",
+      "▁Hand s",
+      "▁Han ds",
+      "▁sc int",
+      "▁sci nt",
+      "▁un itary",
+      "▁unit ary",
+      "dis covery",
+      "disc overy",
+      "农 林",
+      "宣 言",
+      "人民 卫生",
+      "▁pro tons",
+      "▁prot ons",
+      "▁proton s",
+      "▁proto ns",
+      "▁hand ball",
+      "▁pat ented",
+      "▁patent ed",
+      "▁Lanc ashire",
+      "庄 镇",
+      "行 了",
+      "tu rl",
+      "tur l",
+      "t url",
+      "罪犯 奖励审批表",
+      "▁reim burse",
+      "根 的",
+      "ek t",
+      "e kt",
+      "▁O T",
+      "▁ OT",
+      "选择 性",
+      "研究 发现",
+      "▁R ocket",
+      "▁Rock et",
+      "▁Roc ket",
+      "坐落 在",
+      "坐 落在",
+      "这次 的",
+      "这 次的",
+      "ar ine",
+      "ari ne",
+      "arin e",
+      "a rine",
+      "游戏大小 为",
+      "▁out going",
+      "武 帝",
+      "co x",
+      "c ox",
+      "▁W arm",
+      "▁War m",
+      "▁Wa rm",
+      "▁H iggs",
+      "▁K erry",
+      "▁Ker ry",
+      "▁Kerr y",
+      "不值 得",
+      "不 值得",
+      "周围 的",
+      "▁L af",
+      "▁La f",
+      "iv atives",
+      "ivative s",
+      "▁Ch anging",
+      "▁Chang ing",
+      "▁Chan ging",
+      "广泛 应用",
+      "ol ics",
+      "olic s",
+      "oli cs",
+      "连载 的小说",
+      "连载的 小说",
+      "▁Ch arl",
+      "▁Char l",
+      "▁Cha rl",
+      "rodu ced",
+      "rod uced",
+      "roduce d",
+      "int ensity",
+      "软件 的",
+      "软 件的",
+      "▁M orph",
+      "▁Mor ph",
+      "▁W ow",
+      "▁Wo w",
+      "▁IS IS",
+      "▁inst inct",
+      "▁assort ment",
+      "oflu orescence",
+      "写 真",
+      "nS mall",
+      "n Small",
+      "▁F ixes",
+      "▁Fix es",
+      "on Create",
+      "o nCreate",
+      "奋 进",
+      "巴 萨",
+      "bn b",
+      "b nb",
+      "▁b ob",
+      "▁bo b",
+      "▁ bob",
+      "E t",
+      "受 贿",
+      "和 睦",
+      "肺 部",
+      "▁D ock",
+      "▁Do ck",
+      "▁Doc k",
+      "▁weigh s",
+      "nF unction",
+      "n Function",
+      "无 知",
+      "▁a y",
+      "▁ ay",
+      "br oker",
+      "bro ker",
+      "osp ice",
+      "▁rhythm s",
+      "▁rhyth ms",
+      "▁Organ isation",
+      "▁W nt",
+      "▁a ce",
+      "▁ac e",
+      "▁ ace",
+      "▁Kn ox",
+      "▁bul lying",
+      "▁bull ying",
+      "trans actions",
+      "transaction s",
+      "Co u",
+      "C ou",
+      "nWith in",
+      "play list",
+      "▁cycl ist",
+      "▁trade marks",
+      "▁tradem arks",
+      "▁trademark s",
+      "开 来",
+      "▁wave let",
+      "х о",
+      "洋 洋",
+      "Fl u",
+      "F lu",
+      "生效 的",
+      "还 提供",
+      "▁ 赛季",
+      "比赛 的",
+      "比 赛的",
+      "可以 分为",
+      "混合 运算",
+      "▁diver gent",
+      "归 去",
+      "日 光",
+      "▁ 口味",
+      "▁电 话",
+      "▁ 电话",
+      "in se",
+      "ins e",
+      "i nse",
+      "环境 污染",
+      "至关 重要",
+      "▁Gand hi",
+      "▁Ether net",
+      "{ <",
+      "再 多",
+      "MM P",
+      "M MP",
+      "教育 学院",
+      "教育学 院",
+      "▁my osin",
+      "▁str etches",
+      "▁stret ches",
+      "▁stretch es",
+      "天 道",
+      "夯 实",
+      "次 日",
+      "画 像",
+      "▁Z ur",
+      "为主题 的",
+      "cl air",
+      "委员会 副主任",
+      "中国 工商银行股份有限公司",
+      "在 某",
+      "用 具",
+      "只要 是",
+      "只 要是",
+      "视网 膜",
+      "act ually",
+      "actual ly",
+      "▁Re direct",
+      "▁Red irect",
+      "▁ Redirect",
+      "子 树",
+      "学 说",
+      "菜 肴",
+      "观 的",
+      "这 群",
+      "毕竟 是",
+      "har dt",
+      "hard t",
+      "is tem",
+      "ist em",
+      "iste m",
+      "i stem",
+      "由被告 住所地",
+      "G i",
+      "余 杭",
+      "AR E",
+      "A RE",
+      "▁Y un",
+      "▁Yu n",
+      "外观 设计",
+      "生活 习惯",
+      "明智 的选择",
+      "甲基 苯丙胺",
+      "▁th under",
+      "▁Ess ential",
+      "▁comp letes",
+      "▁complet es",
+      "▁complete s",
+      "适 中",
+      "一千 元",
+      "一 千元",
+      "rupt ing",
+      "rup ting",
+      "研发 的",
+      "研 发的",
+      "要 及时",
+      "其他 需要",
+      "▁re vert",
+      "▁reve rt",
+      "▁rev ert",
+      "▁rever t",
+      "Det ector",
+      "▁pri ests",
+      "▁priest s",
+      "活 塞",
+      "aj s",
+      "a js",
+      "▁I G",
+      "▁ IG",
+      "▁德 国",
+      "▁ 德国",
+      "▁re ck",
+      "▁r eck",
+      "▁rec k",
+      "▁ reck",
+      "REG IST",
+      "▁Cons ortium",
+      "▁approxim ations",
+      "▁approximation s",
+      "▁ 名",
+      "成 因",
+      "▁H K",
+      "▁ HK",
+      "立方 分米",
+      "il ian",
+      "ili an",
+      "ilia n",
+      "til es",
+      "ti les",
+      "tile s",
+      "t iles",
+      "▁fung us",
+      "▁mult iples",
+      "▁multiple s",
+      "▁multi ples",
+      "▁multip les",
+      "村 落",
+      "▁anti cancer",
+      "▁antic ancer",
+      "倒 霉",
+      "看 清",
+      "陵 县",
+      "▁b og",
+      "▁bo g",
+      "CA CHE",
+      "CAC HE",
+      "C ACHE",
+      "tc har",
+      "t char",
+      "▁im perial",
+      "▁imp erial",
+      "▁imper ial",
+      "▁impe rial",
+      "都 可",
+      "AD C",
+      "A DC",
+      "同学 的",
+      "同 学的",
+      "CE PT",
+      "▁S PI",
+      "▁SP I",
+      "▁ SPI",
+      "te sted",
+      "test ed",
+      "tes ted",
+      "t ested",
+      "▁Bo nus",
+      "▁Bon us",
+      "om erase",
+      "ome rase",
+      "omer ase",
+      "劳动人事争议 仲裁委员会",
+      "Authent icated",
+      "募 集",
+      "大 军",
+      "yl p",
+      "y lp",
+      "予以 采纳",
+      "▁p enn",
+      "▁pe nn",
+      "▁pen n",
+      "▁bomb s",
+      "▁bom bs",
+      "miss ible",
+      "界 定",
+      "BM I",
+      "B MI",
+      "切换 到",
+      "方程 的",
+      "简直 是",
+      "▁Bry ant",
+      "▁Bryan t",
+      "Image View",
+      "▁super market",
+      "▁supermark et",
+      "年 和",
+      "高 地",
+      "这样 可以",
+      "酒店 交通",
+      "▁S par",
+      "▁Sp ar",
+      "▁Spa r",
+      "▁Mon ten",
+      "▁Mont en",
+      "▁Monte n",
+      "▁ 那",
+      "学 分",
+      "构 图",
+      "祭 祀",
+      "是 全国",
+      "十三 条的规定",
+      "十三条 的规定",
+      "▁her editary",
+      "以 各主要景点的距离如下",
+      "踊 跃",
+      "▁动物 界",
+      "▁Cru ise",
+      "▁compl aining",
+      "▁complain ing",
+      "楼 房",
+      "上 大学",
+      "景 德镇",
+      "更 重要",
+      "▁bond ed",
+      "▁bon ded",
+      "▁land marks",
+      "▁landmark s",
+      "▁arbitr arily",
+      "▁breast feeding",
+      "能 效",
+      "el og",
+      "elo g",
+      "e log",
+      "▁Cl early",
+      "▁Cle arly",
+      "▁Clear ly",
+      "▁L agrangian",
+      "应 邀",
+      "新鲜 的",
+      "问题 进行分析",
+      "read able",
+      "▁Arm enia",
+      "▁Armen ia",
+      "▁Con verter",
+      "▁Convert er",
+      "▁ Converter",
+      "P t",
+      "魔 王",
+      "▁Sof ia",
+      "▁app arel",
+      "▁appar el",
+      "玉 环",
+      "理 赔",
+      "芝 士",
+      "血 小板",
+      "▁G iant",
+      "▁Gi ant",
+      "▁Gian t",
+      "▁it iner",
+      "Part ition",
+      "碳 水",
+      "首 尔",
+      "Var s",
+      "V ars",
+      "as ser",
+      "ass er",
+      "asse r",
+      "▁S che",
+      "▁Sc he",
+      "▁Sch e",
+      "▁T ier",
+      "▁Ti er",
+      "▁R ally",
+      "笼 罩",
+      "车 库",
+      "高级 工程师",
+      "搭 档",
+      "少年 的",
+      "少 年的",
+      "la sh",
+      "las h",
+      "l ash",
+      "▁G IS",
+      "▁GI S",
+      "▁ GIS",
+      "系统 地",
+      "▁Bur ton",
+      "南 侧",
+      "操作 的",
+      "操 作的",
+      "应予 终结",
+      "nH uman",
+      "n Human",
+      "ugin osa",
+      "▁bal ances",
+      "▁balance s",
+      "▁pall iative",
+      "sm s",
+      "s ms",
+      "▁B j",
+      "▁ 位置",
+      "情况 和",
+      "根本 上",
+      "ENG INE",
+      "▁Eug ene",
+      "研 磨",
+      "▁ 丛书",
+      "类 专业",
+      "in verse",
+      "▁respon ders",
+      "▁respond ers",
+      "出 游",
+      "海 边",
+      "酿 酒",
+      "里 克",
+      "▁W O",
+      "▁ WO",
+      "比较 有",
+      "的 疗效",
+      "oth erm",
+      "other m",
+      "othe rm",
+      "▁in sign",
+      "▁i nsign",
+      "▁ins ign",
+      "人 也",
+      "便 会",
+      "每 行",
+      "诸 暨",
+      "小 店",
+      "依法 应予准许",
+      "依法应予 准许",
+      "▁card io",
+      "▁cardi o",
+      "▁quiet ly",
+      "▁eukary otic",
+      "▁mention ing",
+      "路 店",
+      "等 方法",
+      "es que",
+      "▁we ighed",
+      "▁weigh ed",
+      "▁mit igation",
+      "▁collabor ating",
+      "中 铁",
+      "足 迹",
+      "▁h u",
+      "▁ hu",
+      "▁д о",
+      "▁ до",
+      "为 学生",
+      "南北 朝",
+      "▁S py",
+      "▁Sp y",
+      "▁ Spy",
+      "▁u mb",
+      "▁um b",
+      "▁ umb",
+      "都有 一个",
+      "都 有一个",
+      "back ed",
+      "b acked",
+      "▁K indle",
+      "▁Kind le",
+      "水 深",
+      "个月 的",
+      "个 月的",
+      "给予 了",
+      "详细 信息",
+      "▁Sch war",
+      "▁Schw ar",
+      "▁l uggage",
+      "Dim ension",
+      "D imension",
+      "▁as ynchronous",
+      "几 家",
+      "联 谊",
+      "韶 关",
+      "实际 的",
+      "触摸 屏",
+      "提供 参考",
+      "▁L und",
+      "▁Lu nd",
+      "律 诗",
+      "目标 的",
+      "目 标的",
+      "}^{ {",
+      "}^ {{",
+      "▁S can",
+      "▁Sc an",
+      "▁ Scan",
+      "纷纷 表示",
+      "th anks",
+      "than ks",
+      "▁Aur ora",
+      "con ference",
+      "▁inspiration al",
+      "▁inspir ational",
+      "入 院",
+      "▁C AM",
+      "▁CA M",
+      "▁ CAM",
+      "nUs age",
+      "n Usage",
+      "Wh ether",
+      "W hether",
+      "▁Re peat",
+      "▁Rep eat",
+      "▁re design",
+      "▁rede sign",
+      "▁Quant itative",
+      "多 半",
+      "qu il",
+      "re li",
+      "rel i",
+      "r eli",
+      "一个月 内",
+      "家用 电器",
+      "▁serv ant",
+      "Content Type",
+      "出 卖",
+      "孤 立",
+      "edu cation",
+      "educ ation",
+      "e ducation",
+      "两 会",
+      "灌 木",
+      "▁P our",
+      "▁Po ur",
+      "▁dec oder",
+      "▁decode r",
+      "▁ decoder",
+      "▁ex agger",
+      "▁mult iplier",
+      "▁multi plier",
+      "▁multip lier",
+      "代 号",
+      "铃 薯",
+      "办 主任",
+      "直播 间",
+      "▁п ри",
+      "Sc ience",
+      "Sci ence",
+      "▁Mil ton",
+      "▁Pet erson",
+      "▁Peter son",
+      "▁Peters on",
+      "▁sub scriber",
+      "▁subscrib er",
+      "▁subscribe r",
+      "守 门",
+      "柔 和",
+      "补 水",
+      "超 大",
+      "阐 释",
+      "有 很大",
+      "ca ps",
+      "cap s",
+      "c aps",
+      "al king",
+      "alk ing",
+      "cer ely",
+      "仇 恨",
+      "复 位",
+      "fp s",
+      "f ps",
+      "减轻 处罚",
+      "减 轻处罚",
+      "be low",
+      "bel ow",
+      "G ITHUB",
+      "投资 管理有限公司",
+      "投资管理 有限公司",
+      "就不 用",
+      "就 不用",
+      "▁E RA",
+      "▁ER A",
+      "▁ ERA",
+      "F a",
+      "岭 南",
+      "排 球",
+      "放 学",
+      "▁G lob",
+      "▁Gl ob",
+      "▁Glo b",
+      "▁An nie",
+      "▁Ann ie",
+      "▁m ills",
+      "▁mill s",
+      "▁mil ls",
+      "ic illin",
+      "▁Div ide",
+      "▁Divid e",
+      "▁ch ambers",
+      "▁chamber s",
+      "罚 息",
+      "▁V T",
+      "▁ VT",
+      "和 治疗",
+      "▁w ished",
+      "▁wish ed",
+      "▁wis hed",
+      "enc rypted",
+      "encrypt ed",
+      "▁Inf inity",
+      "云 端",
+      "它 能",
+      "ne z",
+      "n ez",
+      "消化 道",
+      "ta ng",
+      "tan g",
+      "t ang",
+      "ex ico",
+      "▁C alled",
+      "▁Cal led",
+      "▁Call ed",
+      "▁ Called",
+      "二〇一八年 五月",
+      "▁Dir ichlet",
+      "姓 氏",
+      "皮 的",
+      "园林 绿化",
+      "in ges",
+      "ing es",
+      "inge s",
+      "▁play back",
+      "▁Inter action",
+      "▁ Interaction",
+      "\\\\ '",
+      "\\ \\'",
+      "▁% %",
+      "▁ %%",
+      "机械 化",
+      "解决 的",
+      "取 暖",
+      "后 备",
+      "本 体",
+      "局域 网",
+      "eb ug",
+      "e bug",
+      "▁W rong",
+      "▁Wr ong",
+      "▁ Wrong",
+      "▁Bre aking",
+      "▁Break ing",
+      "▁pres idency",
+      "Y A",
+      "怀 抱",
+      "本 能",
+      "当然 了",
+      "的人 类",
+      "的 人类",
+      "▁se cular",
+      "▁sec ular",
+      "▁ter restrial",
+      "▁interpret ing",
+      "佣 金",
+      "佳 人",
+      "法 案",
+      "立方 厘米",
+      "▁turb ulence",
+      "交 界",
+      "资金 的",
+      "资 金的",
+      "▁H AL",
+      "▁HA L",
+      "▁ HAL",
+      "▁B ing",
+      "▁Bi ng",
+      "▁Bin g",
+      "▁spokes woman",
+      "巴 马",
+      "市 东",
+      "an yl",
+      "any l",
+      "ov ir",
+      "ovi r",
+      "o vir",
+      "文化 产业",
+      "▁Jan et",
+      "▁Jane t",
+      "▁Ja net",
+      "▁B anking",
+      "▁Bank ing",
+      "▁Ban king",
+      "Pred icate",
+      "不 需",
+      "出 借",
+      "后 端",
+      "知乎 上",
+      "di al",
+      "d ial",
+      "根据 自己的",
+      "un safe",
+      "uns afe",
+      "▁No bel",
+      "▁Nob el",
+      "▁te lev",
+      "▁tele v",
+      "▁tel ev",
+      "nut rition",
+      "大 唐",
+      "▁p ores",
+      "▁po res",
+      "▁por es",
+      "▁pore s",
+      "在服刑 改造期间",
+      "▁hist ologic",
+      "▁The oretical",
+      "后 方",
+      "情 趣",
+      "血 的",
+      "讯 息",
+      "▁c ared",
+      "▁car ed",
+      "▁care d",
+      "▁ca red",
+      "▁U L",
+      "▁ UL",
+      "过多 的",
+      "过 多的",
+      "TING S",
+      "T INGS",
+      "IN VALID",
+      "▁Com fort",
+      "▁Con versely",
+      "▁unpredict able",
+      "皇 上",
+      "OT S",
+      "O TS",
+      "▁第 六",
+      "▁ 第六",
+      "信息 安全",
+      "▁b olt",
+      "▁bo lt",
+      "▁bol t",
+      "▁ bolt",
+      "▁g raz",
+      "▁gr az",
+      "▁gra z",
+      "▁Old er",
+      "▁Ol der",
+      "上 缴",
+      "如 其",
+      "支持 下",
+      "▁t rop",
+      "▁tr op",
+      "▁tro p",
+      "it su",
+      "its u",
+      "▁R ag",
+      "▁Ra g",
+      "▁sub scriptions",
+      "▁subscription s",
+      "▁subs criptions",
+      "▁ subscriptions",
+      "某某 诉被告",
+      "nH y",
+      "n Hy",
+      "nn ode",
+      "nno de",
+      "n node",
+      "▁clin ician",
+      "▁clinic ian",
+      "▁sc rolling",
+      "▁scroll ing",
+      "▁def iciencies",
+      "固 件",
+      "ek a",
+      "e ka",
+      "在 过去",
+      "问题 了",
+      "ST IT",
+      "ol um",
+      "np eople",
+      "n people",
+      "▁Wag ner",
+      "▁f rench",
+      "▁fr ench",
+      "▁fre nch",
+      "pi ration",
+      "pir ation",
+      "p iration",
+      "▁Fried rich",
+      "▁comm enced",
+      "▁commence d",
+      "远 县",
+      "阵 营",
+      "TE X",
+      "T EX",
+      "领导 和",
+      "▁j wt",
+      "▁ jwt",
+      "刑 终字第",
+      "司法 解释",
+      "▁n ortheastern",
+      "▁north eastern",
+      "▁northeast ern",
+      "江 东",
+      "而 易",
+      "银 杏",
+      "▁E van",
+      "▁Ev an",
+      "▁Eva n",
+      "▁Rom ney",
+      "▁F amilies",
+      "气 动",
+      "飞 的",
+      "▁p ued",
+      "▁pu ed",
+      "▁arch ived",
+      "▁archive d",
+      "单 人",
+      "空 心",
+      "nr m",
+      "n rm",
+      "发表 了",
+      "▁L ah",
+      "▁La h",
+      "R ussian",
+      "TE MPLATE",
+      "▁th riving",
+      "▁thr iving",
+      "兑 现",
+      "废 话",
+      "网易 云",
+      "en vs",
+      "env s",
+      "for mal",
+      "form al",
+      "f ormal",
+      "ne utral",
+      "neut ral",
+      "▁ch assis",
+      "就 大",
+      "垃圾 分类",
+      "ur ses",
+      "urs es",
+      "urse s",
+      "gre SQL",
+      "▁theat rical",
+      "改 动",
+      "花 落",
+      "o qu",
+      "▁C DP",
+      "▁CD P",
+      "ave ry",
+      "av ery",
+      "aver y",
+      "a very",
+      "▁er yt",
+      "▁ery t",
+      "▁ex tras",
+      "▁ext ras",
+      "▁extra s",
+      "▁extr as",
+      "▁ extras",
+      "▁p endant",
+      "▁pend ant",
+      "▁para bola",
+      "▁parab ola",
+      "▁neuro pathy",
+      "▁neur opathy",
+      "▁neuropath y",
+      "与 研究",
+      "em os",
+      "emo s",
+      "e mos",
+      "本案 申请执行人",
+      "跑 道",
+      "FO UND",
+      "▁k nit",
+      "▁kn it",
+      "▁A ware",
+      "▁Aw are",
+      "▁ Aware",
+      "将 从",
+      "推 向",
+      "省 长",
+      "自 如",
+      "\\' ,$",
+      "\\', $",
+      "\\ ',$",
+      "up ta",
+      "upt a",
+      "bur st",
+      "bu rst",
+      "b urst",
+      "▁L ambda",
+      "▁Lamb da",
+      "▁ Lambda",
+      "今 朝",
+      "EC D",
+      "E CD",
+      "惯 用脚",
+      "</ >\\",
+      "< />\\",
+      "高职 院校",
+      "Re cent",
+      "Rec ent",
+      "▁Bl ind",
+      "短 路",
+      "' >\\",
+      "[] >",
+      "[ ]>",
+      "管理 协会",
+      "▁h orr",
+      "▁hor r",
+      "▁ho rr",
+      "▁col itis",
+      "▁coli tis",
+      "享受 到",
+      "享 受到",
+      "总 人口",
+      "法律 责任",
+      "De ath",
+      "est ly",
+      "二〇一五年 九月",
+      "▁in versely",
+      "▁inverse ly",
+      "地 亚",
+      "外 接",
+      "安装 包",
+      "有的 是",
+      "有 的是",
+      "JA VA",
+      "J AVA",
+      "ny es",
+      "n yes",
+      "rgb a",
+      "rg ba",
+      "r gba",
+      "et ers",
+      "eter s",
+      "ete rs",
+      "e ters",
+      "▁P ing",
+      "▁Pi ng",
+      "▁Pin g",
+      "▁ Ping",
+      "nTh ird",
+      "n Third",
+      "nc reate",
+      "n create",
+      "other ap",
+      "othe rap",
+      "功 课",
+      "岁 以上",
+      "来 不及",
+      "inter op",
+      "▁solub ility",
+      "部 和",
+      "ES CO",
+      "所 提出的",
+      "的时间 内",
+      "的 时间内",
+      "▁tr aced",
+      "▁tra ced",
+      "▁trace d",
+      "▁proced ural",
+      "红 灯",
+      "本报 讯",
+      "输入 格式",
+      "事实与 理由",
+      "危险 化学品",
+      "Co ok",
+      "C ook",
+      "比 率",
+      "AP S",
+      "A PS",
+      "二 千元",
+      "▁P LC",
+      "▁PL C",
+      "▁ PLC",
+      "▁S DL",
+      "▁SD L",
+      "▁ SDL",
+      "un pack",
+      "▁p ollen",
+      "▁pol len",
+      "▁poll en",
+      "▁pollut ants",
+      "Tr im",
+      "Tri m",
+      "T rim",
+      "ding er",
+      "din ger",
+      "d inger",
+      "▁categor ical",
+      "nC ross",
+      "nCr oss",
+      "n Cross",
+      "▁D ylan",
+      "▁Dy lan",
+      "▁bre asts",
+      "▁breast s",
+      "▁sc heduler",
+      "▁schedul er",
+      "▁schedule r",
+      "▁ scheduler",
+      "乏 力",
+      "密 钥",
+      "还有 人",
+      "还 有人",
+      "hash Code",
+      "▁irre spective",
+      "▁irres pective",
+      "军 官",
+      "专业 从事",
+      "er ground",
+      "erg round",
+      "▁ele phant",
+      "respon ding",
+      "respond ing",
+      "ap plications",
+      "application s",
+      "监 听",
+      "▁ 广西",
+      "▁M TV",
+      "▁MT V",
+      "一点 都不",
+      "行业 发展",
+      "▁Cam ero",
+      "的人 口",
+      "的 人口",
+      "▁E NG",
+      "▁EN G",
+      "▁ ENG",
+      "▁J aw",
+      "▁Ja w",
+      "这个 东西",
+      "源 性",
+      "预 览",
+      "As t",
+      "A st",
+      "文 史",
+      "往 期",
+      "我 已",
+      "Em p",
+      "E mp",
+      "死亡 率",
+      "的 动力",
+      "jo hn",
+      "joh n",
+      "j ohn",
+      "▁I nk",
+      "▁In k",
+      "▁M ega",
+      "▁Me ga",
+      "▁Meg a",
+      "ia ries",
+      "iar ies",
+      "i aries",
+      "▁el icit",
+      "▁elic it",
+      "▁author ed",
+      "▁auth ored",
+      "▁mon uments",
+      "▁monument s",
+      "▁Ab original",
+      "诚 意",
+      "▁r os",
+      "▁ro s",
+      "▁ ros",
+      "怎么 可能",
+      "神经 系统",
+      "▁go at",
+      "umbn ails",
+      "umbnail s",
+      "▁bar rels",
+      "▁barrel s",
+      "▁barr els",
+      "迷 宫",
+      "da g",
+      "d ag",
+      "水 处理",
+      "▁install ment",
+      "ry n",
+      "r yn",
+      "▁S ach",
+      "▁Sa ch",
+      "▁Sac h",
+      "补 正",
+      "▁L M",
+      "▁ LM",
+      "一小 时",
+      "一 小时",
+      "在他 们",
+      "在 他们",
+      "▁* ***",
+      "▁** **",
+      "▁*** *",
+      "▁ ****",
+      "▁pharmacokin etic",
+      "也 称",
+      "子 叶",
+      "ia d",
+      "i ad",
+      "▁l ign",
+      "▁li gn",
+      "▁lig n",
+      "prov ince",
+      "▁b iliary",
+      "▁bil iary",
+      "修 仙",
+      "四 百",
+      "定 于",
+      "徒 弟",
+      "Mat e",
+      "Ma te",
+      "M ate",
+      "合同 签订",
+      "科学技术 出版社",
+      "优美 的",
+      "优 美的",
+      "机器 翻译",
+      "op rot",
+      "opro t",
+      "o prot",
+      "▁r ushing",
+      "▁rush ing",
+      "▁rem ovable",
+      "▁remov able",
+      "▁prosec ution",
+      "中 被",
+      "李 文",
+      "盛 大",
+      "▁ 意大利",
+      "故 诉至法院",
+      "▁V ander",
+      "▁Van der",
+      "▁Va nder",
+      "▁Vand er",
+      "Europe an",
+      "E uropean",
+      "▁ultr ason",
+      "▁ultra son",
+      "▁ultras on",
+      "伸 出",
+      "年 是",
+      "这 天",
+      "▁p k",
+      "▁ pk",
+      "最 合适的",
+      "但 未",
+      "声 道",
+      "并 获得",
+      "的 儿子",
+      "Con g",
+      "Co ng",
+      "C ong",
+      "保修 信息",
+      "▁class ics",
+      "▁classic s",
+      "取决于被执行人 是否有",
+      "▁confront ed",
+      "春 光",
+      "\\\\ (|",
+      "\\\\( |",
+      "▁Na Cl",
+      "▁ne oplastic",
+      "▁neoplas tic",
+      "▁neo plastic",
+      "大 妈",
+      "母 乳",
+      "肾 功能",
+      "▁what soever",
+      "单 机",
+      "{} \"",
+      "{ }\"",
+      "思想 和",
+      "nC zech",
+      "▁All ah",
+      "世 事",
+      "的最 后",
+      "的 最后",
+      "之 名",
+      "电 图",
+      "经 由",
+      "冰 淇淋",
+      "辖区 内",
+      "辖 区内",
+      "Ui nt",
+      "U int",
+      "nd an",
+      "nda n",
+      "n dan",
+      "ou nty",
+      "ount y",
+      "oun ty",
+      "ING TON",
+      "辛 辣",
+      "高 了",
+      "▁ 谱曲",
+      "阿里 云",
+      "IT EM",
+      "ITE M",
+      "I TEM",
+      "就会 出现",
+      "就 会出现",
+      "▁Report ing",
+      "自 尊",
+      "▁L ov",
+      "▁Lo v",
+      "依据 问题",
+      "▁S ources",
+      "▁Source s",
+      "▁ Sources",
+      "▁det ention",
+      "I U",
+      "洋 溢",
+      "爱 吃",
+      "驱 逐",
+      "公司 和",
+      "品牌 策划",
+      "申请 行政复议",
+      "▁micro n",
+      "▁mic ron",
+      "▁mi cron",
+      "▁sh ader",
+      "▁sha der",
+      "▁shade r",
+      "▁ shader",
+      "郭 某",
+      "Bo n",
+      "B on",
+      "I vy",
+      "ga m",
+      "g am",
+      "▁O UR",
+      "不会 出现",
+      "不 会出现",
+      "遵纪 守法",
+      "nEn ron",
+      "n Enron",
+      "▁Ant ib",
+      "▁Anti b",
+      "st reams",
+      "stream s",
+      "stre ams",
+      ") ×",
+      "挂 钩",
+      "此 案",
+      "要 将",
+      "有 这么",
+      "干部 职工",
+      "▁C ove",
+      "▁Co ve",
+      "▁Cov e",
+      "oxy gen",
+      "o xygen",
+      "cl icked",
+      "click ed",
+      "op ening",
+      "ope ning",
+      "open ing",
+      "▁Jane iro",
+      "病 菌",
+      "不在 乎",
+      "不 在乎",
+      "为 原料",
+      "Se qu",
+      "Seq u",
+      "S equ",
+      "▁I ND",
+      "▁IN D",
+      "▁ IND",
+      "▁refer rals",
+      "▁referral s",
+      "一 词",
+      "螺 栓",
+      "就像 是",
+      "就 像是",
+      "▁L az",
+      "▁La z",
+      "ra dor",
+      "rad or",
+      "r ador",
+      "}\" );\\",
+      "} \");\\",
+      "▁C ERN",
+      "▁CE RN",
+      "▁P ill",
+      "▁Pi ll",
+      "▁Pil l",
+      "▁bear ings",
+      "▁bearing s",
+      "课 件",
+      "▁F res",
+      "▁Fr es",
+      "▁Fre s",
+      "▁Re ply",
+      "▁Rep ly",
+      "▁ Reply",
+      "▁Bron ze",
+      "拆 卸",
+      "未 央",
+      "运 维",
+      "VA R",
+      "V AR",
+      "部 主任",
+      "gi st",
+      "gis t",
+      "g ist",
+      "ns ize",
+      "n size",
+      "Provider s",
+      "Prov iders",
+      "陪 你",
+      "体现 出",
+      "pl en",
+      "ple n",
+      "p len",
+      "▁` \\\\",
+      "▁`\\ \\",
+      "非常 重要的",
+      "非常重要 的",
+      "日提出 减刑建议",
+      "answer ed",
+      "函 授",
+      "囊 肿",
+      "导 管",
+      "巡 察",
+      "火 炬",
+      "Bi o",
+      "B io",
+      "的 合作",
+      "▁P AT",
+      "▁PA T",
+      "▁ PAT",
+      "血 量",
+      "在 香港",
+      "nDe velop",
+      "n Develop",
+      "▁Cow boys",
+      "▁Tim othy",
+      "▁air plane",
+      "▁Question naire",
+      "不 理",
+      "何 人",
+      "家 村",
+      "球 衣",
+      "奥 地利",
+      "or iasis",
+      "ori asis",
+      "oria sis",
+      "▁Sh aron",
+      "▁Sha ron",
+      "▁Shar on",
+      "▁Griff in",
+      "汗 水",
+      "骁 龙",
+      "▁c ue",
+      "▁cu e",
+      "▁ cue",
+      "De g",
+      "D eg",
+      "gu st",
+      "g ust",
+      "▁s lash",
+      "▁sl ash",
+      "▁ slash",
+      "▁Bulld ogs",
+      "ad renergic",
+      "adren ergic",
+      "会 场",
+      "很 棒",
+      "IF N",
+      "I FN",
+      "安庆 市",
+      "安 庆市",
+      "Mi me",
+      "M ime",
+      "Pod s",
+      "Po ds",
+      "P ods",
+      "代表 性的",
+      "代表性 的",
+      "nS ure",
+      "nSur e",
+      "▁Advent ures",
+      "▁Adventure s",
+      "H o",
+      "上 衣",
+      "口 红",
+      "线 和",
+      "支持 和",
+      "而这 些",
+      "而 这些",
+      "ag ascar",
+      "aga scar",
+      "▁Per l",
+      "▁Pe rl",
+      "▁dis proportion",
+      "烟 火",
+      "J ac",
+      "共同 体",
+      "▁p ots",
+      "▁pot s",
+      "▁po ts",
+      "公诉机关指控的 罪名成立",
+      "建立 在",
+      "联系 人",
+      "ri ven",
+      "riv en",
+      "rive n",
+      "r iven",
+      "H Z",
+      "et tle",
+      "ett le",
+      "e ttle",
+      "伪 装",
+      "肝 癌",
+      "重点 是",
+      "重 点是",
+      "ak in",
+      "aki n",
+      "a kin",
+      "原审 裁定",
+      "▁B ren",
+      "▁Br en",
+      "▁Bre n",
+      "▁sp awn",
+      "▁spa wn",
+      "▁spaw n",
+      "▁ spawn",
+      "▁spark ed",
+      "▁kind ness",
+      "一 想",
+      "舍 得",
+      "邻 近",
+      "给你 们",
+      "给 你们",
+      "▁I sn",
+      "▁Is n",
+      "▁Rel ative",
+      "▁ Relative",
+      "长 假",
+      "▁D U",
+      "▁ DU",
+      "的高 级",
+      "的 高级",
+      "蚌埠 市",
+      "an on",
+      "ano n",
+      "a non",
+      "▁g ra",
+      "▁gr a",
+      "▁ gra",
+      "占 了",
+      "可 比",
+      "逛 街",
+      "高 贵",
+      "nR O",
+      "n RO",
+      "动物 园",
+      "当选 为",
+      "试题 分析",
+      "越来 越好",
+      "越来越 好",
+      "▁Br uno",
+      "▁Brun o",
+      "▁Bru no",
+      "▁m apper",
+      "▁map per",
+      "▁ mapper",
+      "▁rot ations",
+      "▁rotation s",
+      "六 级",
+      "高 点",
+      "▁cl erk",
+      "▁cle rk",
+      "▁cler k",
+      "市人民法院 于",
+      "▁sp ells",
+      "▁spell s",
+      "bit ration",
+      "bi tration",
+      "bitr ation",
+      "群 的",
+      "预 示",
+      "太极 拳",
+      "HD MI",
+      "罪犯 考核",
+      "▁Hil ton",
+      "▁acceler ating",
+      "W O",
+      "做 梦",
+      "很 正常",
+      "ar ty",
+      "art y",
+      "nL ib",
+      "n Lib",
+      "▁sad ly",
+      "▁cont empl",
+      "▁Chev rolet",
+      "轴 的",
+      "分别 于",
+      "设置 为",
+      "TO OL",
+      "▁J iang",
+      "▁Ji ang",
+      "▁d read",
+      "▁dr ead",
+      "创业 板",
+      "▁tem po",
+      "▁temp o",
+      "就 该",
+      "这 笔",
+      "bar d",
+      "ba rd",
+      "b ard",
+      "on ate",
+      "ona te",
+      "ne ider",
+      "Num eric",
+      "Numer ic",
+      "Nu meric",
+      "仍 旧",
+      "在 场",
+      "▁s d",
+      "▁ sd",
+      "举个 例子",
+      "住 江苏省",
+      "Re ver",
+      "Rev er",
+      "R ever",
+      "▁Data set",
+      "▁Dat aset",
+      "▁ Dataset",
+      "lif ecycle",
+      "life cycle",
+      "▁shock ing",
+      "▁bel ieving",
+      "▁belie ving",
+      "校 正",
+      "箱 子",
+      "通 透",
+      "D jango",
+      "ei ther",
+      "e ither",
+      "ا ن",
+      "眼 界",
+      "碾 压",
+      "把 一个",
+      "▁fe min",
+      "▁fem in",
+      "▁gro ove",
+      "mat ching",
+      "match ing",
+      "cl amation",
+      "Com patible",
+      "Compat ible",
+      "何 种",
+      "提 议",
+      "\\' _",
+      "\\ '_",
+      "() }\\",
+      "()} \\",
+      "( )}\\",
+      "nd iv",
+      "ndi v",
+      "n div",
+      "C e",
+      "不会 再",
+      "会 引起",
+      "▁G aza",
+      "▁Ga za",
+      "▁Gaz a",
+      "对 申请执行人",
+      "育 儿",
+      "健康 教育",
+      "在 天",
+      "驻 地",
+      "高 某",
+      "za n",
+      "z an",
+      "因为 有",
+      "nA dv",
+      "nAd v",
+      "n Adv",
+      "▁mel ted",
+      "▁melt ed",
+      "曾 获",
+      "ab o",
+      "a bo",
+      "▁E du",
+      "▁Ed u",
+      "中央 空调",
+      "▁T ill",
+      "▁Ti ll",
+      "▁Til l",
+      "ok ines",
+      "okin es",
+      "okine s",
+      "oki nes",
+      "▁ax ons",
+      "▁axon s",
+      "fit ting",
+      "f itting",
+      "▁Hand book",
+      "▁dis connect",
+      "▁ disconnect",
+      "Document ation",
+      "上 为",
+      "包 包",
+      "无 不",
+      "竞 选",
+      "▁& \\",
+      "与 企业",
+      "ik it",
+      "iki t",
+      "i kit",
+      "re ward",
+      "rew ard",
+      "r eward",
+      "▁bud dy",
+      "▁map le",
+      "▁ma ple",
+      "转 弯",
+      "直升 机",
+      "政协 副主席",
+      "ord ion",
+      "Tem perature",
+      "I o",
+      "北 约",
+      "杀 伤",
+      "正 宗",
+      "nJ ud",
+      "Ro man",
+      "Rom an",
+      "R oman",
+      "und ai",
+      "unda i",
+      "▁l ion",
+      "▁li on",
+      "▁ lion",
+      "▁Man or",
+      "▁Ma nor",
+      "▁trace back",
+      "▁hyper plasia",
+      "▁Im plications",
+      "要 对",
+      "钝 角",
+      "第六 十四条",
+      "第六十四 条",
+      "SY STEM",
+      "SYS TEM",
+      "ex istent",
+      "exist ent",
+      "Ass ociation",
+      "▁Beautiful Soup",
+      "我 问",
+      "曲 目",
+      "织 物",
+      "▁B oh",
+      "▁Bo h",
+      ">\\ ',\\",
+      ">\\' ,\\",
+      ">\\', \\",
+      "> \\',\\",
+      "▁scal p",
+      "声 乐",
+      "绳 子",
+      "曾用 名",
+      "th ews",
+      "the ws",
+      "thew s",
+      "t hews",
+      "ulos keletal",
+      "耶 稣",
+      "不会 被",
+      "不 会被",
+      "执行 异议",
+      "rs trip",
+      "rst rip",
+      "r strip",
+      "▁C aval",
+      "▁Ca val",
+      "▁Cav al",
+      "▁ar mor",
+      "▁arm or",
+      "▁kn ots",
+      "▁knot s",
+      "▁stuff ed",
+      "出 境",
+      "as n",
+      "a sn",
+      "Wi ki",
+      "W iki",
+      "▁di aph",
+      "▁dia ph",
+      "▁e lective",
+      "▁elect ive",
+      "▁el ective",
+      "▁sh ipment",
+      "▁ship ment",
+      "Value Error",
+      "▁collabor ators",
+      "免 交",
+      "圈 的",
+      "▁E gg",
+      "▁Eg g",
+      "ri bly",
+      "rib ly",
+      "r ibly",
+      "准许 上诉人",
+      "byte rian",
+      "byter ian",
+      "他们 对",
+      "边长 为",
+      "边 长为",
+      "ut en",
+      "ute n",
+      "u ten",
+      "▁w ag",
+      "▁wa g",
+      "第七十九条 和",
+      "hou ette",
+      "▁Sm ooth",
+      "▁tr ader",
+      "▁trad er",
+      "▁tra der",
+      "▁trade r",
+      "▁state wide",
+      "中 段",
+      "打 到",
+      "财经 大学",
+      "igh bour",
+      "ighb our",
+      "th ermal",
+      "ther mal",
+      "nReg ards",
+      "▁Gujar at",
+      "od ds",
+      "odd s",
+      "▁r oses",
+      "▁ro ses",
+      "▁rose s",
+      "▁ros es",
+      "▁respect s",
+      "▁resp ects",
+      "Ti O",
+      "T iO",
+      "▁B eg",
+      "▁Be g",
+      "CM ake",
+      "C Make",
+      "oss ible",
+      "rec ursive",
+      "医 治",
+      "触 及",
+      "th or",
+      "t hor",
+      "SE CON",
+      "SEC ON",
+      "▁pl eas",
+      "▁ple as",
+      "▁plea s",
+      "▁bow ling",
+      "▁bowl ing",
+      "▁cock tails",
+      "▁cocktail s",
+      "▁Ach ievement",
+      "▁commission er",
+      "发现被执行人有 财产可供执行",
+      "Di e",
+      "D ie",
+      "等 活动",
+      "中国 企业",
+      "税务 总局",
+      "▁sc av",
+      "▁A hmad",
+      "▁Ah mad",
+      "▁fibr oblast",
+      "▁fibro blast",
+      "世界 里",
+      "ut an",
+      "uta n",
+      "u tan",
+      "▁M odal",
+      "▁Mod al",
+      "▁Mo dal",
+      "▁ Modal",
+      "二〇一六年 十月",
+      "二〇一六年十 月",
+      "find ViewById",
+      "不 自",
+      "在 被",
+      "红 十字",
+      "at ham",
+      "ath am",
+      "atha m",
+      "iaz ep",
+      "▁l one",
+      "▁lo ne",
+      "▁lon e",
+      "Res olve",
+      "Re solve",
+      "▁Phil lip",
+      "▁Phill ip",
+      "▁h acking",
+      "▁hack ing",
+      "▁electroly te",
+      "ancellation Token",
+      "一 扫",
+      "纯 属",
+      "ne gie",
+      "neg ie",
+      "▁Non linear",
+      "▁s mo",
+      "▁sm o",
+      "raw ling",
+      "rawl ing",
+      "in cluded",
+      "include d",
+      "▁entrepreneur ial",
+      "大 都",
+      "越 小",
+      "ato nin",
+      "aton in",
+      "ist ine",
+      "习近平 总书记",
+      "Get ting",
+      "G etting",
+      "é c",
+      "ó d",
+      "EO F",
+      "E OF",
+      "Le x",
+      "L ex",
+      "Cu be",
+      "C ube",
+      "ws gi",
+      "w sgi",
+      "▁R up",
+      "▁Ru p",
+      "向本院 提供被执行人",
+      "向本院提供 被执行人",
+      "▁Techn iques",
+      "▁Technique s",
+      "梦 境",
+      "考 评",
+      "牛仔 裤",
+      "▁R ough",
+      "▁Ro ugh",
+      "▁Rou gh",
+      "▁surf ing",
+      "▁tw entieth",
+      "同 名",
+      "▁M F",
+      "▁ MF",
+      "▁scrut iny",
+      "乌 兰",
+      "人 多",
+      "宾 客",
+      "弊 端",
+      "us ty",
+      "ust y",
+      "u sty",
+      "第一 部分",
+      "第一部 分",
+      "第 一部分",
+      "nMount ains",
+      "▁view point",
+      "李 小",
+      "皇 马",
+      "当前 的",
+      "当 前的",
+      "nP ay",
+      "n Pay",
+      "Cy cle",
+      "C ycle",
+      "▁h airy",
+      "▁hair y",
+      "▁ha iry",
+      "oblast s",
+      "obl asts",
+      "o blasts",
+      "▁har bor",
+      "▁harb or",
+      "e j",
+      "嘉 奖",
+      "如 期",
+      "蝙 蝠",
+      "▁ 俱乐部",
+      "所 产生的",
+      "dag ger",
+      "d agger",
+      "▁Can terbury",
+      "_ *",
+      "过 量",
+      "▁s late",
+      "▁sl ate",
+      "二〇一七年 六",
+      "本 领",
+      "ip v",
+      "i pv",
+      "条件 和",
+      "所 带来的",
+      "▁p eric",
+      "▁per ic",
+      "▁pe ric",
+      "▁peri c",
+      "▁Med ian",
+      "▁Media n",
+      "Im porter",
+      "Import er",
+      "Imp orter",
+      "ndef ault",
+      "nde fault",
+      "n default",
+      "men opausal",
+      "偏 见",
+      "赋 值",
+      "nG l",
+      "n Gl",
+      "审核 表",
+      "▁P ST",
+      "▁PS T",
+      "知识 渊博",
+      "▁E igen",
+      "TIME OUT",
+      "▁pul sed",
+      "▁pulse d",
+      "▁puls ed",
+      "nB asketball",
+      "▁acc idental",
+      "▁accident al",
+      "▁different iating",
+      "DU CTION",
+      "DUCT ION",
+      "经原告多次 催要",
+      "H U",
+      "乙 肝",
+      "分别 在",
+      "▁a wa",
+      "▁aw a",
+      "▁ awa",
+      "Th ink",
+      "Un til",
+      "U ntil",
+      "nI tem",
+      "nIt em",
+      "n Item",
+      "而 有所不同",
+      "附带 民事诉讼",
+      "附带民事 诉讼",
+      "tem poral",
+      "temp oral",
+      "t emporal",
+      "Inter action",
+      "▁ 风",
+      "车 轮",
+      "▁都市 言情",
+      "pr esso",
+      "press o",
+      "pres so",
+      "判决生效后 十日内",
+      "介 意",
+      "提供 的证据",
+      "提供的 证据",
+      "第一百四十 四条的规定",
+      "第一百四十四 条的规定",
+      "第一百四十四条 的规定",
+      "▁Cont ributors",
+      "一 枝",
+      "较小 的",
+      "较 小的",
+      "op py",
+      "opp y",
+      "▁G ian",
+      "▁Gi an",
+      "▁te amed",
+      "▁team ed",
+      "▁tea med",
+      "▁four teen",
+      "与 国际",
+      "\": [\\",
+      "\":[ \\",
+      "债权 依法",
+      "在 人民法院",
+      "批准 的项目",
+      "具有 挑战性的",
+      "该犯 在服刑期间",
+      "read thedocs",
+      "副 总",
+      "lish er",
+      "lis her",
+      "l isher",
+      "▁W ings",
+      "▁Win gs",
+      "▁Wing s",
+      "▁w ears",
+      "▁we ars",
+      "▁wear s",
+      "el iness",
+      "eline ss",
+      "elines s",
+      "eli ness",
+      "elin ess",
+      "e liness",
+      "▁tor rent",
+      "压 根",
+      "床 房",
+      "的 技能",
+      "英雄 联盟",
+      "▁per oxide",
+      "▁perox ide",
+      "▁ 蓝",
+      "花 纹",
+      "$) ,",
+      "$ ),",
+      "原来 是",
+      "很快 就",
+      "▁S CI",
+      "▁SC I",
+      "▁ SCI",
+      "ar ger",
+      "arg er",
+      "arge r",
+      "nH eat",
+      "nHe at",
+      "符合下列 条件",
+      "mat cher",
+      "match er",
+      "m atcher",
+      "▁F inals",
+      "▁Fin als",
+      "▁Final s",
+      "▁Fi nals",
+      "▁Luther an",
+      "get Property",
+      "下 场",
+      "宁 区",
+      "Ca s",
+      "C as",
+      "也许 是",
+      "的 种类",
+      "nY et",
+      "lic he",
+      "li che",
+      "lich e",
+      "l iche",
+      "▁m alt",
+      "▁mal t",
+      "▁ma lt",
+      "具备 执行条件",
+      "make dirs",
+      "m akedirs",
+      "▁Pr eston",
+      "▁Pres ton",
+      "▁Pre ston",
+      "▁Prest on",
+      "▁Sw imming",
+      "▁Ut ilities",
+      "▁Util ities",
+      "▁ Utilities",
+      "▁Active Record",
+      "▁hepat ocellular",
+      "军 民",
+      "押 金",
+      "cre ative",
+      "creat ive",
+      "▁Rain bow",
+      "tg z",
+      "t gz",
+      "惠州 市",
+      "惠 州市",
+      "io la",
+      "iol a",
+      "i ola",
+      "请 告诉我",
+      "骑 行",
+      "ae s",
+      "a es",
+      "也 希望",
+      "非常 适合",
+      "但 债权的实现",
+      "min imal",
+      "mini mal",
+      "▁Exec ution",
+      "▁ Execution",
+      "▁aer uginosa",
+      "周 日",
+      "投 降",
+      "盐 酸",
+      "支行 的",
+      "支 行的",
+      "的一 句话",
+      "的 一句话",
+      "ab bit",
+      "abb it",
+      "理论 与实践",
+      "理论与 实践",
+      "确定 的内容",
+      "确定的 内容",
+      "▁F etch",
+      "▁ Fetch",
+      "▁micro gl",
+      "▁microg l",
+      "▁test imon",
+      "穿 透",
+      "▁T G",
+      "▁ TG",
+      "nB ro",
+      "nBr o",
+      "n Bro",
+      "Sp ark",
+      "S park",
+      "Th omas",
+      "▁Tas mania",
+      "▁aberr ant",
+      "Pi c",
+      "P ic",
+      "▁ 化学",
+      "规章 制度",
+      "re cent",
+      "rec ent",
+      "▁attract s",
+      "▁attr acts",
+      "▁multi plicative",
+      "勘 察",
+      "FA I",
+      "F AI",
+      "三年 级",
+      "三 年级",
+      "借款 利息",
+      "▁my riad",
+      "cle rotic",
+      "cler otic",
+      "K I",
+      "pl ayed",
+      "play ed",
+      "平行 四边形的",
+      "平行四边形 的",
+      "ous ands",
+      "ousand s",
+      "▁S ister",
+      "▁Si ster",
+      "▁empir e",
+      "▁emp ire",
+      "▁tow els",
+      "▁towel s",
+      "AP E",
+      "A PE",
+      "ye l",
+      "y el",
+      "连云 港",
+      "▁c urb",
+      "▁cur b",
+      "▁cu rb",
+      ",\\\\ ,\\\\",
+      ",\\\\, \\\\",
+      ", \\\\,\\\\",
+      "nthe ir",
+      "n their",
+      "▁pr ism",
+      "▁pri sm",
+      "angan ese",
+      "▁aeros ol",
+      "▁Ply mouth",
+      "▁Invest igator",
+      "▁Investig ator",
+      "的 理想",
+      "▁t ant",
+      "▁ta nt",
+      "▁tan t",
+      "▁Gr ammar",
+      "▁Gram mar",
+      "▁adoles cence",
+      "只 剩",
+      "望 远",
+      "波 罗",
+      "对 身体",
+      "▁t achy",
+      "▁ta chy",
+      "▁neon ates",
+      "▁surfact ant",
+      "每 晚",
+      "in z",
+      "i nz",
+      "▁c ó",
+      "▁S ME",
+      "▁SM E",
+      "㈰㈰ ㈰㈰",
+      "怎么 回事",
+      "▁A mber",
+      "▁Am ber",
+      "▁Amb er",
+      "▁al gae",
+      "▁alg ae",
+      "▁ver dict",
+      "内 侧",
+      "奇 偶",
+      "收 条",
+      "灯 泡",
+      "ER A",
+      "E RA",
+      "局限 性",
+      "男主 角",
+      "男 主角",
+      "错过 了",
+      "错 过了",
+      "UE ST",
+      "UES T",
+      "U EST",
+      "ie ri",
+      "ier i",
+      "i eri",
+      "Mo tion",
+      "M otion",
+      "▁infer tility",
+      "K W",
+      "np l",
+      "n pl",
+      "也 只有",
+      "南 极",
+      "颇 有",
+      "太大 的",
+      "太 大的",
+      "ich ever",
+      "iche ver",
+      "requ ent",
+      "req uent",
+      "▁leuk ocyte",
+      "的女 性",
+      "的 女性",
+      "这样 的话",
+      "这样的 话",
+      "ou lli",
+      "oul li",
+      "金 石",
+      "R ich",
+      "▁P CB",
+      "▁PC B",
+      "▁ PCB",
+      "▁图书 信息",
+      "oca ust",
+      "▁pert inent",
+      "万 户",
+      "暑 期",
+      "nM or",
+      "n Mor",
+      "br ids",
+      "brid s",
+      "▁sc rat",
+      "▁scr at",
+      "nCon cept",
+      "n Concept",
+      "人 行",
+      "从 我",
+      "大 雨",
+      "▁B MP",
+      "▁BM P",
+      "▁ BMP",
+      "▁Ang lo",
+      "mu table",
+      "mut able",
+      "m utable",
+      "▁judgment s",
+      "▁judg ments",
+      "丛 林",
+      "瑕 疵",
+      "襄阳 市",
+      "襄 阳市",
+      "so up",
+      "s oup",
+      "▁P oker",
+      "▁Po ker",
+      "▁Pok er",
+      "有 明确的被告",
+      "▁f encing",
+      "▁fen cing",
+      "▁init iating",
+      "▁initi ating",
+      "▁In struction",
+      "▁ Instruction",
+      "上 行",
+      "有 其",
+      "▁b t",
+      "▁ bt",
+      "ca nt",
+      "can t",
+      "c ant",
+      "住 江西省",
+      "▁B oris",
+      "▁Bo ris",
+      "▁Bor is",
+      "com mits",
+      "comm its",
+      "commit s",
+      "▁str ipes",
+      "▁stri pes",
+      "▁strip es",
+      "▁stripe s",
+      "子 弟",
+      "揭 露",
+      "珍贵 的",
+      "珍 贵的",
+      "▁K nown",
+      "▁Kn own",
+      "▁Know n",
+      "▁ Known",
+      "percent age",
+      "r k",
+      "代 会",
+      "\"# \"",
+      "能力 与",
+      "WH ERE",
+      "W HERE",
+      "▁P iece",
+      "▁Pie ce",
+      "▁ Piece",
+      "▁Week end",
+      "mess aging",
+      "platform s",
+      "plat forms",
+      "▁med iation",
+      "▁media tion",
+      "▁medi ation",
+      "反 派",
+      "ecom ment",
+      "e comment",
+      "▁aut onomic",
+      "▁he uristic",
+      "▁heur istic",
+      "▁ 得",
+      "之 乡",
+      "奉 贤",
+      "振 华",
+      "提 现",
+      "橱 柜",
+      "▁Fr anz",
+      "▁Fra nz",
+      "▁Suff olk",
+      "▁syn apses",
+      "▁synapse s",
+      "乐 于",
+      "公 分",
+      "含量 的",
+      "含 量的",
+      "完毕 后",
+      "旅游 局",
+      "病虫 害",
+      "spe ech",
+      "延 缓",
+      "货 源",
+      "▁O B",
+      "▁ OB",
+      "个 国家",
+      "特别授权 代理",
+      "nArgent ine",
+      "▁manip ulated",
+      "▁manipulate d",
+      "拉 萨",
+      "的正 确",
+      "的 正确",
+      "▁F ruit",
+      "▁Fr uit",
+      "bel lion",
+      "bell ion",
+      "▁Tr acking",
+      "▁Track ing",
+      "▁ Tracking",
+      "▁Cart esian",
+      "兽 医",
+      "飞 翔",
+      "） ”",
+      "准备 了",
+      "就 比较",
+      "方向 键",
+      "▁a gar",
+      "▁ag ar",
+      "▁ agar",
+      "▁tr yp",
+      "▁try p",
+      "▁gr ind",
+      "▁gri nd",
+      "两 千",
+      "无 故",
+      "作家 协会",
+      "作 家协会",
+      "民事 行为",
+      "连云 港市",
+      "连云港 市",
+      "▁pers isted",
+      "▁persist ed",
+      "天 王",
+      "转 眼",
+      "▁* _",
+      "nS ave",
+      "n Save",
+      "▁Moh ammed",
+      "Com pletion",
+      "Comp letion",
+      "PS C",
+      "P SC",
+      "war t",
+      "wa rt",
+      "w art",
+      "▁se ated",
+      "▁sea ted",
+      "▁seat ed",
+      "lin space",
+      "▁L ANGUAGE",
+      "体 操",
+      "在 读",
+      "待 机",
+      "攻 关",
+      "样 板",
+      "一流 的",
+      "一 流的",
+      "就会 被",
+      "就 会被",
+      "饮用 水",
+      "饮 用水",
+      "▁c off",
+      "▁co ff",
+      "具有 重要的",
+      "技术 转让",
+      "▁Bur ke",
+      "▁Burk e",
+      "▁Mar cel",
+      "▁Marc el",
+      "对 所",
+      "掩 饰",
+      "缺 损",
+      "hd l",
+      "h dl",
+      "放射 性",
+      "有 任何",
+      "At om",
+      "A tom",
+      "行政 法规",
+      "▁B rett",
+      "▁Br ett",
+      "▁Bre tt",
+      "▁Bret t",
+      "▁st acks",
+      "▁stack s",
+      "▁am ended",
+      "▁amen ded",
+      "▁amend ed",
+      "无疑 是",
+      "长大 了",
+      "长 大了",
+      "AT EG",
+      "ATE G",
+      "re ls",
+      "rel s",
+      "r els",
+      "▁A DC",
+      "▁AD C",
+      "▁ ADC",
+      "的 绝对值",
+      "▁deb ts",
+      "▁debt s",
+      "使 者",
+      "年 产",
+      "us sed",
+      "uss ed",
+      "▁v ous",
+      "▁vo us",
+      "em atics",
+      "emat ics",
+      "ematic s",
+      "共 创",
+      "河 东",
+      "ан и",
+      "а ни",
+      "▁mark eted",
+      "▁market ed",
+      "王 一",
+      "精 盐",
+      "紧 迫",
+      "是一 条",
+      "是 一条",
+      "▁mit ral",
+      "; /",
+      "抗 原",
+      "逃 跑",
+      "和 相关",
+      "客房 时",
+      "清晰 的",
+      "个 村民小组",
+      "小数点 移动",
+      "▁cl iff",
+      "▁cli ff",
+      "网 课",
+      "▁原 创",
+      "▁ 原创",
+      "不明 显",
+      "不 明显",
+      "该 问题",
+      "销售 的",
+      "▁N urse",
+      "▁Nurs e",
+      "▁Nur se",
+      "nInt egr",
+      "n Integr",
+      "ts tring",
+      "t string",
+      "师 的",
+      "▁B ast",
+      "▁Bas t",
+      "▁Ba st",
+      "iz ards",
+      "izard s",
+      "izar ds",
+      "▁sc oop",
+      "▁av oids",
+      "▁avoid s",
+      "Class ifier",
+      "升 学",
+      "no c",
+      "n oc",
+      "▁T OP",
+      "▁TO P",
+      "▁ TOP",
+      "非法 占用的",
+      "elect ro",
+      "同 性",
+      "萧 萧",
+      "角 相等",
+      "AT ING",
+      "A TING",
+      "哭 泣",
+      "对 症",
+      "霸 气",
+      "自由 的",
+      "nM ont",
+      "nMon t",
+      "n Mont",
+      "end ars",
+      "endar s",
+      "enda rs",
+      "▁Ass istance",
+      "牛 排",
+      "▁A J",
+      "▁ AJ",
+      "万亿 元",
+      "万 亿元",
+      "容易 出现",
+      "ty ard",
+      "t yard",
+      "▁S OME",
+      "▁SO ME",
+      "▁D inner",
+      "▁Din ner",
+      "▁pop ulate",
+      "▁popul ate",
+      "▁deploy ing",
+      "▁ К",
+      "马 丁",
+      "就好 像",
+      "就 好像",
+      "深厚 的",
+      "深 厚的",
+      "▁p ou",
+      "▁po u",
+      "▁Jew el",
+      "▁split s",
+      "▁spl its",
+      "也 想",
+      "装 甲",
+      "IR A",
+      "I RA",
+      "今天 是",
+      "wan a",
+      "wa na",
+      "w ana",
+      "▁N il",
+      "▁Ni l",
+      "▁ Nil",
+      "tu mor",
+      "tum or",
+      "t umor",
+      "ut sch",
+      "uts ch",
+      "▁B less",
+      "▁Bl ess",
+      "▁Ble ss",
+      "▁S pons",
+      "▁Sp ons",
+      "▁invert ible",
+      "Suppress Warnings",
+      "保健 品",
+      "▁t riv",
+      "▁tr iv",
+      "▁tri v",
+      "▁c olum",
+      "▁col um",
+      "▁ colum",
+      "▁Kim ber",
+      "▁Ki mber",
+      "全 称",
+      "分 针",
+      "胳 膊",
+      "自 驾",
+      "透 析",
+      "AD DR",
+      "ADD R",
+      "A DDR",
+      "交通 管理",
+      "开奖 结果",
+      "nS uch",
+      "n Such",
+      "ro ads",
+      "road s",
+      "酒店交通 比较方便",
+      "Res olution",
+      "Re solution",
+      "ET S",
+      "E TS",
+      "▁D F",
+      "▁ DF",
+      "保定 市",
+      "▁G CC",
+      "▁GC C",
+      "学生 学习",
+      "ab ort",
+      "abor t",
+      "abo rt",
+      "orn ado",
+      "orna do",
+      "升 华",
+      "Gl u",
+      "G lu",
+      "同时 向",
+      "Hi st",
+      "His t",
+      "H ist",
+      "± \\",
+      "坎 坷",
+      "车 队",
+      "BR E",
+      "B RE",
+      "投资 或",
+      "▁g ems",
+      "▁ge ms",
+      "▁gem s",
+      "▁ gems",
+      "酒店 为您提供",
+      "ugg ling",
+      "协助 执行通知书",
+      "协助执行 通知书",
+      "整 式",
+      "客房 条款",
+      "前所未 有的",
+      "Trans l",
+      "Tra nsl",
+      "选择上述 客房时",
+      "▁sweep ing",
+      "Part icip",
+      "傻 子",
+      "ha y",
+      "h ay",
+      "pd b",
+      "p db",
+      "的 陈述",
+      "pi ns",
+      "pin s",
+      "p ins",
+      "igg ins",
+      "set attr",
+      "削 弱",
+      "川 普",
+      "DI Y",
+      "的 温度",
+      "▁Open ing",
+      "▁Op ening",
+      "▁rehe ars",
+      "hy a",
+      "h ya",
+      "准许 或者",
+      "▁Qu inn",
+      "▁neph ew",
+      "▁nep hew",
+      "▁catal ysts",
+      "▁catalyst s",
+      "▁cooper ate",
+      "▁logarith mic",
+      "▁logarithm ic",
+      "的条件 下",
+      "的 条件下",
+      "▁\\' /\\'",
+      "▁\\'/ \\'",
+      "▁web inar",
+      "▁webin ar",
+      "阅 览",
+      "▁g uilt",
+      "▁gu ilt",
+      "▁gui lt",
+      "请参阅 客房条款",
+      "nCh ristian",
+      "nChrist ian",
+      "n Christian",
+      "▁public ity",
+      "找 了",
+      "滥 用",
+      "我也 不知道",
+      "我也不 知道",
+      "我 也不知道",
+      "提起的 诉讼",
+      "你 来",
+      "佳 节",
+      "灯 火",
+      "老 妈",
+      "的 特殊",
+      "这么 做",
+      "马 铃薯",
+      "▁P hp",
+      "▁Ph p",
+      "▁ Php",
+      "两种 情况",
+      "质保 时间",
+      "an king",
+      "ank ing",
+      "arc tan",
+      "arct an",
+      "上诉于 河南省",
+      "元 宵",
+      "唐 诗",
+      "尽 头",
+      "承 德",
+      "▁r t",
+      "▁ rt",
+      "好用 的",
+      "好 用的",
+      "For t",
+      "F ort",
+      "保证 责任",
+      "着自己 的",
+      "着 自己的",
+      "st ash",
+      "sta sh",
+      "▁Hor izon",
+      "作 息",
+      "过 剩",
+      "om ew",
+      "ome w",
+      "▁r isen",
+      "▁rise n",
+      "▁ris en",
+      "▁ri sen",
+      "参 保",
+      "奶 酪",
+      "cess ion",
+      "c ession",
+      "Mark down",
+      "▁H B",
+      "▁ HB",
+      "为我 国",
+      "为 我国",
+      "的 实力",
+      "设备 及",
+      "BO SS",
+      "B OSS",
+      "土地 使用权",
+      "In puts",
+      "Input s",
+      "Log ged",
+      "▁Griff ith",
+      "附 有",
+      "旁边 的",
+      "旁 边的",
+      "节目 的",
+      "节 目的",
+      "毫无 疑问",
+      "货币 政策",
+      "nc ase",
+      "n case",
+      "▁S pread",
+      "▁Sp read",
+      "▁Spr ead",
+      "水 箱",
+      "ea s",
+      "e as",
+      "▁F CC",
+      "▁FC C",
+      "查询 回执",
+      "▁L ucky",
+      "▁Luc ky",
+      "▁Luck y",
+      "▁un bel",
+      "▁ph antom",
+      "▁ phantom",
+      "▁ear rings",
+      "免 去",
+      "广 元",
+      "水 温",
+      "珍 藏",
+      "战略 合作",
+      "无论 如何",
+      "表扬 奖励",
+      "av oid",
+      "avo id",
+      "a void",
+      "▁lon ely",
+      "▁lone ly",
+      "▁Tex ture",
+      "▁Text ure",
+      "▁ Texture",
+      "有管辖权的 人民法院",
+      "生态 系统",
+      "▁insp ections",
+      "▁inspection s",
+      "▁inspect ions",
+      "qu a",
+      "q ua",
+      "ab er",
+      "abe r",
+      "a ber",
+      "Qu est",
+      "Que st",
+      "▁fac ets",
+      "▁face ts",
+      "▁facet s",
+      "Re vision",
+      "Rev ision",
+      "▁elucid ated",
+      "▁elucidate d",
+      "酒店为您提供 免费的停车服务",
+      "虚 弱",
+      "非 负",
+      "仍然 是",
+      "市场 中",
+      "Cur ve",
+      "al ive",
+      "ali ve",
+      "a live",
+      "▁小说 阅读网",
+      "▁ 小说阅读网",
+      "的 逆",
+      "象 棋",
+      "趁 着",
+      "▁Luck ily",
+      "▁det ective",
+      "▁detect ive",
+      "♀ ♀",
+      "万 名",
+      "既 要",
+      "分析 这些",
+      "▁L ips",
+      "▁Li ps",
+      "▁Lip s",
+      "Se eder",
+      "See der",
+      "Seed er",
+      "os cale",
+      "osc ale",
+      "o scale",
+      "nC ritical",
+      "无 果",
+      "让你 的",
+      "让 你的",
+      "▁S cr",
+      "▁Sc r",
+      "▁L ists",
+      "▁List s",
+      "▁Li sts",
+      "▁Lis ts",
+      "▁ Lists",
+      "ns cript",
+      "nsc ript",
+      "n script",
+      "▁ch orus",
+      "▁cho rus",
+      "▁chor us",
+      "cit ation",
+      "c itation",
+      "nd ocument",
+      "n document",
+      "唯 独",
+      "小 三",
+      "WA Y",
+      "W AY",
+      "fi rm",
+      "f irm",
+      "▁pol arity",
+      "▁polar ity",
+      "▁att ributable",
+      "三 部",
+      "农 行",
+      "中 科院",
+      "有人 在",
+      "有 人在",
+      "CD MA",
+      "▁L EG",
+      "▁LE G",
+      "▁g rit",
+      "▁gr it",
+      "▁gri t",
+      "▁Urugu ay",
+      "▁pict ured",
+      "▁picture d",
+      "蒙 特",
+      "▁Craw ford",
+      "▁vers atility",
+      "▁versa tility",
+      "换 热",
+      "璀 璨",
+      "融 化",
+      "较 量",
+      "附 着",
+      "buk kit",
+      "nc olor",
+      "ncol or",
+      "nco lor",
+      "n color",
+      "fa ctors",
+      "factor s",
+      "fact ors",
+      "f actors",
+      "开 场",
+      "航 运",
+      "nu id",
+      "n uid",
+      "就是 这么",
+      "▁w oven",
+      "▁wo ven",
+      "▁Georg ian",
+      "▁Georgia n",
+      "充 要",
+      "抽 屉",
+      "▁l g",
+      "▁ lg",
+      "听说 过",
+      "听 说过",
+      "ha na",
+      "han a",
+      "h ana",
+      "ony ms",
+      "onym s",
+      "▁M err",
+      "▁Me rr",
+      "▁Mer r",
+      "T Z",
+      "走 来",
+      "CH E",
+      "C HE",
+      "这种 事情",
+      "这种事 情",
+      "ipher text",
+      "iph ertext",
+      "企 查",
+      "超 标",
+      "查 笔录",
+      "当庭 陈述",
+      "▁K ann",
+      "▁Ka nn",
+      "▁Kan n",
+      "ust ering",
+      "uster ing",
+      "▁favor ed",
+      "▁fav ored",
+      "▁after math",
+      "肿 胀",
+      "追 随",
+      "镇 痛",
+      "时 之前",
+      "生产 建设",
+      "科技 公司",
+      "▁Exper ts",
+      "▁Exp erts",
+      "▁Expert s",
+      "▁honor ary",
+      "麻 木",
+      "极高 的",
+      "极 高的",
+      "电子 有限公司",
+      "琵 琶",
+      "青 龙",
+      "发出 的",
+      "发 出的",
+      "is mo",
+      "ism o",
+      "▁S ew",
+      "▁Se w",
+      "▁g ods",
+      "▁go ds",
+      "▁god s",
+      "如下图 所示",
+      "▁re ver",
+      "▁r ever",
+      "▁reve r",
+      "▁rev er",
+      "inos aur",
+      "Rect angle",
+      "M U",
+      "t A",
+      "л у",
+      "美 团",
+      "SA T",
+      "S AT",
+      "理解 为",
+      "▁W inner",
+      "▁Win ner",
+      "▁Winn er",
+      "▁Cont ains",
+      "▁ Contains",
+      "tt f",
+      "t tf",
+      "ys c",
+      "y sc",
+      "▁ 广东",
+      "服务 和",
+      "ou sand",
+      "ous and",
+      "▁s tool",
+      "▁st ool",
+      "▁sto ol",
+      "ere bral",
+      "User Name",
+      "▁olig onucle",
+      "上 也",
+      "滤 镜",
+      "▁bow ls",
+      "▁bowl s",
+      "vent ure",
+      "ven ture",
+      "小 块",
+      "昼 夜",
+      "并将 其",
+      "并 将其",
+      "的 顺序",
+      "▁C BC",
+      "▁CB C",
+      "▁T uc",
+      "▁Tu c",
+      "规划 设计",
+      "▁H ann",
+      "▁Ha nn",
+      "▁Han n",
+      "Or ders",
+      "Order s",
+      "Ord ers",
+      "nC ities",
+      "▁Ex cept",
+      "▁mut ated",
+      "▁BUS INESS",
+      "▁mon sters",
+      "▁monster s",
+      "▁conduct ance",
+      "▁south western",
+      "▁southwest ern",
+      "云 霄",
+      "▁r ugs",
+      "▁ru gs",
+      "▁rug s",
+      "pa lette",
+      "pal ette",
+      "▁plate au",
+      "▁Neg ative",
+      "▁ Negative",
+      "确认人民调解协议 效力的",
+      "▁ κ",
+      "涨 停",
+      "协助 执行",
+      "并提出 了",
+      "并 提出了",
+      "nA ugust",
+      "nAug ust",
+      "n August",
+      "一 朵",
+      "点 了",
+      "po ons",
+      "poon s",
+      "p oons",
+      "ü n",
+      "天 河",
+      "本 国",
+      "▁C her",
+      "▁Ch er",
+      "▁Che r",
+      "Act ual",
+      "Ac tual",
+      "▁bin aries",
+      "一年 级",
+      "一 年级",
+      "方位 的",
+      "方 位的",
+      "第四 届",
+      "工作 报告",
+      "▁B irds",
+      "▁Bird s",
+      "▁Bir ds",
+      "N atural",
+      "/ [",
+      "时 髦",
+      "真 伪",
+      "ST REAM",
+      "▁code cs",
+      "▁codec s",
+      "▁det erg",
+      "▁deter g",
+      "小 花",
+      "民事 判决书",
+      "▁pre tend",
+      "▁pret end",
+      "oblast oma",
+      "o blastoma",
+      "▁nine teenth",
+      "▁nin eteenth",
+      "住 的",
+      "aw ai",
+      "awa i",
+      "▁Mat lab",
+      "▁transf ected",
+      "t q",
+      "少 林",
+      "烧 热",
+      "支持 的",
+      "去年 同期",
+      "get Elements",
+      "getElement s",
+      "是 与",
+      "该 项",
+      "支付 了",
+      "政府 和",
+      "组 组长",
+      "绵阳 市",
+      "绵 阳市",
+      "od ule",
+      "o dule",
+      "ood le",
+      "▁L ung",
+      "▁Lu ng",
+      "Im mutable",
+      "Imm utable",
+      "bread crumb",
+      "缝 隙",
+      "中国 传统",
+      "▁gl itter",
+      "大 明",
+      "og h",
+      "o gh",
+      "进行 计算",
+      "所 欲",
+      "处 三年",
+      "▁w ipe",
+      "▁wip e",
+      "▁wi pe",
+      "▁explo iting",
+      "▁exploit ing",
+      "▁instruct ed",
+      "以 南",
+      "作 弊",
+      "司 职",
+      "角 逐",
+      "人性 化",
+      "▁M LB",
+      "▁ML B",
+      ";\\\\ ;\\\\",
+      "; \\\\;\\\\",
+      "Max imum",
+      "requ ently",
+      "requent ly",
+      "就 和",
+      "点 在",
+      "nC entral",
+      "n Central",
+      "▁Camb odia",
+      "▁SER VICES",
+      "▁soci etal",
+      "扁 平",
+      "有 事",
+      "观 影",
+      "在一 些",
+      "在 一些",
+      "cs dn",
+      "nI NS",
+      "nIN S",
+      "n INS",
+      "▁E RP",
+      "▁ER P",
+      "▁ ERP",
+      "应当 承担",
+      "一定 要注意",
+      "一定要 注意",
+      "▁Re model",
+      "▁Rem odel",
+      "不 值",
+      "之 分",
+      "听 证",
+      "辜 负",
+      "Jo n",
+      "J on",
+      "是 属于",
+      "此次 活动",
+      "ol ate",
+      "ola te",
+      "o late",
+      "▁W aste",
+      "▁Was te",
+      "▁Wa ste",
+      "▁he mod",
+      "▁hem od",
+      "nCh ange",
+      "n Change",
+      "不起 来",
+      "不 起来",
+      "▁R ET",
+      "▁RE T",
+      "▁ RET",
+      "招商 引资",
+      "ist ors",
+      "istor s",
+      "isto rs",
+      "▁emerg encies",
+      "抽 查",
+      "没有 一个",
+      "没 有一个",
+      "▁S chr",
+      "▁Sc hr",
+      "▁Sch r",
+      "▁R overs",
+      "▁Ro vers",
+      "▁Rover s",
+      "亲 友",
+      "保 利",
+      "联 立",
+      "no se",
+      "nos e",
+      "n ose",
+      "▁Ig A",
+      "市场 份额",
+      "ra eli",
+      "rael i",
+      "其他需要 裁定",
+      "▁Hawai ian",
+      "▁Hawaii an",
+      "em m",
+      "e mm",
+      "撤回 执行申请",
+      "系 该",
+      "^{ *",
+      "新 时期",
+      "Ber ry",
+      "B erry",
+      "▁qu int",
+      "▁q uint",
+      "▁qui nt",
+      "▁quin t",
+      "nT aking",
+      "▁estab lishes",
+      "▁establish es",
+      "include graphics",
+      "防 线",
+      "成功 后",
+      "有 那么",
+      "nP rob",
+      "nPro b",
+      "nPr ob",
+      "n Prob",
+      "等腰 三角形的",
+      "等腰三角形 的",
+      "C riteria",
+      "▁Act ress",
+      "▁sol itary",
+      "中 也",
+      "松 原",
+      "nM s",
+      "n Ms",
+      "ig ar",
+      "iga r",
+      "i gar",
+      "除以 进率",
+      "▁E stim",
+      "▁Est im",
+      "▁Es tim",
+      "▁ Estim",
+      "▁t ails",
+      "▁tail s",
+      "▁ta ils",
+      "▁ tails",
+      "▁up loading",
+      "▁upload ing",
+      "乡 县",
+      "几 张",
+      "推 崇",
+      "油 炸",
+      "Ta p",
+      "T ap",
+      "▁H ib",
+      "▁Hi b",
+      "▁P VC",
+      "▁PV C",
+      "▁ PVC",
+      "▁h an",
+      "▁ha n",
+      "▁ han",
+      "▁H iro",
+      "▁Hi ro",
+      "▁Hir o",
+      "之间 的距离",
+      "之间的 距离",
+      "▁Rap ids",
+      "▁Rapid s",
+      "住房和 城乡建设",
+      "nArt icle",
+      "n Article",
+      "Boot strap",
+      "public ation",
+      "封 锁",
+      "比对 照",
+      "比 对照",
+      "nD an",
+      "n Dan",
+      "恶性 肿瘤",
+      "born s",
+      "bor ns",
+      "▁no qa",
+      "wrap ped",
+      "wr apped",
+      "▁custom ization",
+      "杏 仁",
+      "nS ong",
+      "nSo ng",
+      "n Song",
+      "▁Tr acy",
+      "▁Tra cy",
+      "▁se rine",
+      "▁ser ine",
+      "▁Free man",
+      "▁Fre eman",
+      "tra umatic",
+      "determ ined",
+      "▁plac ental",
+      "▁placenta l",
+      "泪 水",
+      "▁F IX",
+      "▁FI X",
+      "▁ FIX",
+      "▁U ns",
+      "▁Un s",
+      "▁ Uns",
+      "▁i ch",
+      "▁ic h",
+      "▁ ich",
+      "▁L omb",
+      "▁Lo mb",
+      "ns elect",
+      "nsel ect",
+      "nse lect",
+      "n select",
+      "▁Comp act",
+      "▁ Compact",
+      "▁ende avor",
+      "▁Throw able",
+      "▁ Throwable",
+      "暂 行",
+      "眼 看",
+      "骨 科",
+      "rn g",
+      "r ng",
+      "买 受人",
+      "化学 物质",
+      "fa ult",
+      "f ault",
+      "▁Ch rys",
+      "ent ropy",
+      "金 秋",
+      "上一 层",
+      "上 一层",
+      "nJ oe",
+      "nJo e",
+      "n Joe",
+      "默 契",
+      "家庭 成员",
+      "pir atory",
+      "▁reve nge",
+      "▁rev enge",
+      "▁reven ge",
+      "▁ventric le",
+      "好 坏",
+      "梧 桐",
+      "ud s",
+      "u ds",
+      "▁W o",
+      "弹 幕",
+      "ne ts",
+      "net s",
+      "n ets",
+      "管理 模式",
+      "▁C ases",
+      "▁Ca ses",
+      "▁Cas es",
+      "▁Case s",
+      "▁ Cases",
+      "ax el",
+      "▁h arb",
+      "▁har b",
+      "▁ha rb",
+      "get env",
+      "▁co ats",
+      "▁coat s",
+      "▁f laws",
+      "▁fl aws",
+      "▁flaw s",
+      "▁divor ced",
+      "▁divorce d",
+      "中 位数",
+      "中 发现",
+      "web socket",
+      "webs ocket",
+      "▁driv eway",
+      "▁drive way",
+      "▁time delta",
+      "▁timed elta",
+      "v k",
+      "继 而",
+      "▁I z",
+      "女生 的",
+      "女 生的",
+      "▁N eeds",
+      "▁Ne eds",
+      "▁Need s",
+      "orpor ate",
+      "orp orate",
+      "▁det ained",
+      "虾 仁",
+      "fa q",
+      "f aq",
+      "▁F en",
+      "▁Fe n",
+      "且 在",
+      "治 国",
+      "薄 荷",
+      "cr on",
+      "cro n",
+      "c ron",
+      "▁[ {\"",
+      "▁[{ \"",
+      "▁ [{\"",
+      "▁p ym",
+      "▁py m",
+      "ar ris",
+      "arr is",
+      "▁Car son",
+      "▁Cars on",
+      "▁percept ual",
+      "▁h apl",
+      "▁ha pl",
+      "▁hap l",
+      "▁veget arian",
+      "一 轮",
+      "要 好",
+      "in de",
+      "ind e",
+      "i nde",
+      "▁La gos",
+      "▁Lag os",
+      "▁p henyl",
+      "▁phen yl",
+      "▁ phenyl",
+      "M ERCHANTABILITY",
+      "▁y o",
+      "▁ yo",
+      "等 进行",
+      "nd ec",
+      "nde c",
+      "n dec",
+      "ta pe",
+      "tap e",
+      "t ape",
+      "▁sub division",
+      "▁subdiv ision",
+      "串 联",
+      "崇 尚",
+      "观 摩",
+      "的情 绪",
+      "的 情绪",
+      "nR oad",
+      "n Road",
+      "an ched",
+      "anc hed",
+      "anch ed",
+      "anche d",
+      "▁de graded",
+      "▁degrad ed",
+      "▁serial izers",
+      "▁serialize rs",
+      "▁serializer s",
+      "▁ serializers",
+      "校 验",
+      "累 犯",
+      "教育 出版社",
+      "Car thy",
+      "Cart hy",
+      "To List",
+      "▁Ep idem",
+      "▁Light ning",
+      "滋 养",
+      "镶 嵌",
+      "Mi l",
+      "M il",
+      "低 成本",
+      "要 坚持",
+      "▁ 社会",
+      "力量 的",
+      "力 量的",
+      "公共 卫生",
+      "gov uk",
+      "▁Mid lands",
+      "▁deliber ately",
+      "▁deliberate ly",
+      "脊 柱",
+      "首 创",
+      "该 系统",
+      "▁F ans",
+      "▁Fa ns",
+      "▁Fan s",
+      "低 保",
+      "先 予",
+      "▁i g",
+      "▁ ig",
+      "到 直线",
+      "直辖 市",
+      "nT er",
+      "nTe r",
+      "n Ter",
+      "bu cks",
+      "buck s",
+      "b ucks",
+      "ch dir",
+      "▁F olk",
+      "▁Fol k",
+      "▁id le",
+      "▁ idle",
+      "▁l inen",
+      "▁li nen",
+      "▁line n",
+      "▁lin en",
+      "▁Ber nie",
+      "▁Bern ie",
+      "▁M ozilla",
+      "▁Mo zilla",
+      "▁Moz illa",
+      "▁dev ised",
+      "▁re charge",
+      "卖 了",
+      "拉 动",
+      "接 线",
+      "效率 高",
+      "效 率高",
+      "Tr ump",
+      "▁kid n",
+      "▁ki dn",
+      "nPh oto",
+      "nPhot o",
+      "n Photo",
+      "▁t odos",
+      "▁to dos",
+      "▁todo s",
+      "光 华",
+      "陆 地",
+      "ba c",
+      "b ac",
+      "▁E vil",
+      "▁Ev il",
+      "р ы",
+      "道路 交通安全",
+      "▁boost ing",
+      "▁fall back",
+      "▁Illegal Argument",
+      "伸 手",
+      "初 学者",
+      "▁Inj ury",
+      "> @",
+      "屠 杀",
+      "流 泪",
+      "划分 为",
+      "划 分为",
+      "可 随时",
+      "的 科学",
+      "nK im",
+      "n Kim",
+      "不 必要的",
+      "对外 开放",
+      "Pad ding",
+      "P adding",
+      "▁P riority",
+      "▁Prior ity",
+      "▁ Priority",
+      "Cred ential",
+      "利 好",
+      "后 天",
+      "了他 的",
+      "了 他的",
+      "ri os",
+      "rio s",
+      "r ios",
+      "▁C af",
+      "▁Ca f",
+      "af fin",
+      "aff in",
+      "ogene ic",
+      "本次执行程序 中",
+      "▁intra operative",
+      "\" ^",
+      "光 彩",
+      "安 东",
+      "漫 步",
+      "di ms",
+      "dim s",
+      "d ims",
+      "▁p inch",
+      "▁pi nch",
+      "▁pin ch",
+      "arr ison",
+      "arris on",
+      "▁aut ore",
+      "▁auto re",
+      "▁autor e",
+      "▁Ver izon",
+      "一 刀",
+      "与 王",
+      "史 诗",
+      "使用 时",
+      "多年 前",
+      "多 年前",
+      "pro f",
+      "pr of",
+      "但是 对于",
+      "▁Em ploy",
+      "▁Emp loy",
+      "▁all oys",
+      "▁alloy s",
+      "▁Game Object",
+      "▁ GameObject",
+      "挥 发",
+      "舍 友",
+      "起 身",
+      "学 研究",
+      "▁M add",
+      "▁Mad d",
+      "▁Ma dd",
+      "th ings",
+      "thing s",
+      "thin gs",
+      "▁san ctuary",
+      "村 人",
+      "li o",
+      "l io",
+      "▁P ole",
+      "▁Pol e",
+      "▁Po le",
+      "SE rror",
+      "S Error",
+      "中 天",
+      "不太 好",
+      "不 太好",
+      "请 不要",
+      "▁L uk",
+      "▁Lu k",
+      "py torch",
+      "▁Pat ricia",
+      "▁Patri cia",
+      "德 的",
+      "甘 油",
+      "bl y",
+      "b ly",
+      "据 报道",
+      "issue comment",
+      "issu ecomment",
+      "侵 入",
+      "地 里",
+      "▁U R",
+      "▁ UR",
+      "▁这 里",
+      "▁ 这里",
+      "Le ad",
+      "L ead",
+      "▁山东 省",
+      "▁ 山东省",
+      "计数 单位",
+      "ı r",
+      "加 持",
+      "店 家",
+      "茂 名",
+      "▁n ip",
+      "▁ni p",
+      "▁ nip",
+      "tr end",
+      "tre nd",
+      "t rend",
+      "▁ 包装清单",
+      "委托代理人 陈",
+      "▁ing estion",
+      "高 高",
+      "ST D",
+      "S TD",
+      "经营 项目",
+      "amp oo",
+      "oph ilia",
+      "ophil ia",
+      "▁Laure nt",
+      "▁Lauren t",
+      "▁Laur ent",
+      "▁Surv ival",
+      "▁b iologic",
+      "▁bi ologic",
+      "▁bio logic",
+      "社 员",
+      "召开 的",
+      "召 开的",
+      "▁f err",
+      "▁fe rr",
+      "▁fer r",
+      "▁f unct",
+      "▁fun ct",
+      "▁func t",
+      "atem ala",
+      "Work space",
+      "Works pace",
+      "Serialized Name",
+      "绘 图",
+      "充足 的",
+      "各行 各",
+      "效果 图",
+      "th us",
+      "t hus",
+      "ur en",
+      "ure n",
+      "u ren",
+      "▁lumin osity",
+      "花 鸟",
+      "衰 退",
+      "学 专业",
+      "ur us",
+      "uru s",
+      "u rus",
+      "▁C hest",
+      "▁Ch est",
+      "▁Che st",
+      "▁Ches t",
+      "▁nost alg",
+      "unc iation",
+      "Rel ationship",
+      "Relation ship",
+      "单 单",
+      "复 工",
+      "结果 为",
+      "▁Wal sh",
+      "▁Mass age",
+      "▁welcome s",
+      "▁wel comes",
+      "▁请 你",
+      "▁ 请你",
+      "等 原因",
+      "转移 到",
+      "Tr ain",
+      "Tra in",
+      "T rain",
+      "in alg",
+      "inal g",
+      "ina lg",
+      "解决 的事项",
+      "解决的 事项",
+      "at ured",
+      "ature d",
+      "atur ed",
+      "atu red",
+      "▁de lib",
+      "▁del ib",
+      "▁bacter ium",
+      "▁South ampton",
+      "代 的",
+      "剑 桥",
+      "大 二",
+      "▁n h",
+      "▁ nh",
+      "ad ir",
+      "adi r",
+      "a dir",
+      "as pect",
+      "asp ect",
+      "a spect",
+      "ion ine",
+      "io nine",
+      "ioni ne",
+      "▁organ izer",
+      "▁organize r",
+      "克 制",
+      "无 证",
+      "胖 子",
+      "集 结",
+      "一家 人",
+      "一 家人",
+      "都 应该",
+      "年被 评为",
+      "年 被评为",
+      "▁C hop",
+      "▁Ch op",
+      "▁Cho p",
+      "odont ic",
+      "odon tic",
+      "▁negot iating",
+      "AD S",
+      "A DS",
+      "但他 们",
+      "但 他们",
+      "str uments",
+      "strument s",
+      "▁ 点",
+      "也 别",
+      "战 线",
+      "李 明",
+      "肛 门",
+      "鹦 鹉",
+      "▁( :",
+      "▁ (:",
+      "sol ver",
+      "solve r",
+      "s olver",
+      "' }",
+      "▁ 重",
+      "R ing",
+      "▁pip elines",
+      "▁pipeline s",
+      "▁pipe lines",
+      "暗 恋",
+      "ac cum",
+      "acc um",
+      "▁stabil ized",
+      "▁stabilize d",
+      "跟 她",
+      "▁E b",
+      "但 现在",
+      "证件 号码",
+      "罚金 人民币",
+      "▁Lib ert",
+      "▁Li bert",
+      "▁p ledge",
+      "▁pl edge",
+      "get Current",
+      "切 削",
+      "吃 喝",
+      "眼 底",
+      "你这 个",
+      "你 这个",
+      "ot en",
+      "ote n",
+      "o ten",
+      "▁n ach",
+      "▁na ch",
+      "▁a ston",
+      "▁as ton",
+      "▁ast on",
+      "▁ aston",
+      "▁swe pt",
+      "▁p oultry",
+      "▁Off icers",
+      "▁Office rs",
+      "▁Officer s",
+      "病 害",
+      "ci r",
+      "c ir",
+      "pol e",
+      "po le",
+      "p ole",
+      "▁taxp ayers",
+      "▁taxpayer s",
+      "不 曾",
+      "睫 毛",
+      "语 义",
+      "nR H",
+      "n RH",
+      "慢慢 地",
+      "爱 奇艺",
+      "中国 平安",
+      "er nal",
+      "ern al",
+      "ac tical",
+      "act ical",
+      "actic al",
+      "ο υ",
+      "用 什么",
+      "nS un",
+      "n Sun",
+      "et able",
+      "eta ble",
+      "e table",
+      "▁in ning",
+      "▁inn ing",
+      "▁p uppet",
+      "▁pup pet",
+      "▁ puppet",
+      "▁depict s",
+      "* >",
+      "▁Z h",
+      "看了 看",
+      "起诉 的",
+      "su lf",
+      "s ulf",
+      "▁novel ists",
+      "▁novelist s",
+      "U U",
+      "民 三",
+      "粉 色",
+      "PUB LIC",
+      "▁eff iciencies",
+      "娘 家",
+      "科 协",
+      "}) }{",
+      "} )}{",
+      "▁Ad miral",
+      "▁Ra iders",
+      "▁Raid ers",
+      "\" *",
+      "早 早",
+      "死 神",
+      "句话 说",
+      "句 话说",
+      "打开 了",
+      "打 开了",
+      "▁J upiter",
+      "Att achment",
+      "Attach ment",
+      "▁strat ified",
+      "中 立",
+      "省 会",
+      "矢 量",
+      "├ ──",
+      "▁ 排名",
+      "只要 有",
+      "只 要有",
+      "ri cks",
+      "ric ks",
+      "rick s",
+      "r icks",
+      "▁C ron",
+      "▁Cr on",
+      "▁Cro n",
+      "▁d ude",
+      "▁du de",
+      "▁Ha yes",
+      "▁Hay es",
+      "pop ular",
+      "▁ob solete",
+      "一 味",
+      "八 角",
+      "二〇一五年 十月",
+      "平安银行股份有限公司 上海",
+      "的 时刻",
+      "▁Z IP",
+      "违法 犯罪",
+      "no ver",
+      "nov er",
+      "n over",
+      "ol len",
+      "oll en",
+      "▁J ung",
+      "▁Jun g",
+      "▁Ju ng",
+      "nsh ould",
+      "n should",
+      "▁ins ane",
+      "▁disc ret",
+      "▁discre t",
+      "予以 减刑",
+      "ke ras",
+      "ker as",
+      "k eras",
+      "▁ST UD",
+      "▁six ty",
+      "▁gl orious",
+      "th emed",
+      "the med",
+      "theme d",
+      "them ed",
+      "▁F rost",
+      "▁Fr ost",
+      "▁Fro st",
+      "nF actor",
+      "nFac tor",
+      "n Factor",
+      "▁partner ing",
+      "长 是",
+      "pe z",
+      "p ez",
+      "▁R w",
+      "▁ 职称",
+      "Phys ics",
+      "▁mon astery",
+      "繁 琐",
+      "执行 措施",
+      "金属 材料",
+      "or bit",
+      "orb it",
+      "▁S isters",
+      "▁Si sters",
+      "▁Sister s",
+      "代 为",
+      "向 后",
+      "尼 奥",
+      "想 看",
+      "▁re sh",
+      "▁r esh",
+      "▁res h",
+      "▁ resh",
+      "iv ating",
+      "iva ting",
+      "▁Pat rol",
+      "▁iss uing",
+      "前 科",
+      "PR S",
+      "P RS",
+      "▁R if",
+      "▁Ri f",
+      "▁R iv",
+      "▁Ri v",
+      "▁m ug",
+      "▁mu g",
+      "函数 的图象",
+      "函数的 图象",
+      "存 入",
+      "请 假",
+      "雨 中",
+      "范围 的",
+      "▁ex h",
+      "nU pon",
+      "nUp on",
+      "遵守监规 纪律",
+      "K n",
+      "继承 人",
+      "▁c athe",
+      "▁ca the",
+      "▁cat he",
+      "▁cath e",
+      "nSup port",
+      "n Support",
+      "ind icator",
+      "揭 开",
+      "贵 人",
+      "ct s",
+      "c ts",
+      "ar ya",
+      "ary a",
+      "▁D und",
+      "▁Dun d",
+      "▁Du nd",
+      "▁b atches",
+      "▁batch es",
+      "▁bat ches",
+      "▁out flow",
+      "▁mos quito",
+      "▁mosquit o",
+      "▁cis platin",
+      "万 象",
+      "重 工",
+      "义乌 市",
+      "的 秘密",
+      "▁g rape",
+      "▁gr ape",
+      "▁grap e",
+      "▁gra pe",
+      "nLe ague",
+      "n League",
+      "Vel ocity",
+      "nWORK DIR",
+      "▁camp uses",
+      "▁campus es",
+      "IEnum erable",
+      "I Enumerable",
+      "和他 们",
+      "和 他们",
+      "的 父亲",
+      "皮肤 病",
+      "资源 和",
+      "邓 小平",
+      "▁D EF",
+      "▁DE F",
+      "▁ DEF",
+      "lect ive",
+      "l ective",
+      "▁ 宁",
+      "▁h h",
+      "▁ hh",
+      "极大 地",
+      "极 大地",
+      "por a",
+      "po ra",
+      "p ora",
+      "▁M ud",
+      "▁Mu d",
+      "Le ave",
+      "▁aest hetics",
+      "▁aesthetic s",
+      "▁endorse ment",
+      "▁endors ement",
+      "米 粉",
+      "创新 性",
+      "须经 批准的项目",
+      "安 保",
+      "EL S",
+      "E LS",
+      "ba m",
+      "b am",
+      "你没 有",
+      "你 没有",
+      "遇到 过",
+      "os est",
+      "ose st",
+      "oses t",
+      "▁h eme",
+      "▁he me",
+      "▁hem e",
+      "▁ heme",
+      "▁m uff",
+      "▁mu ff",
+      "se ctor",
+      "sec tor",
+      "sect or",
+      "s ector",
+      "Loc ator",
+      "▁ac uity",
+      "▁b ounding",
+      "▁bo unding",
+      "▁bound ing",
+      "cont ributors",
+      "contrib utors",
+      "contributor s",
+      "监狱 认为",
+      "nA pril",
+      "nAp ril",
+      "nApr il",
+      "n April",
+      "▁cost ing",
+      "▁cos ting",
+      "该 车",
+      "vi k",
+      "v ik",
+      "对 社会",
+      "科研 工作",
+      "tr ial",
+      "tri al",
+      "t rial",
+      "nC olor",
+      "nCol or",
+      "nCo lor",
+      "n Color",
+      "loc ated",
+      "l ocated",
+      "▁t urtle",
+      "▁tur tle",
+      "▁binding s",
+      "▁bin dings",
+      "▁bind ings",
+      "▁ bindings",
+      "con version",
+      "co nversion",
+      "conv ersion",
+      "▁ex changed",
+      "▁exchange d",
+      "橘 子",
+      "lo l",
+      "l ol",
+      "日 受理",
+      "▁_ (\"",
+      "▁ _(\"",
+      "▁Gu atemala",
+      "之 光",
+      "at oms",
+      "ato ms",
+      "atom s",
+      "同底数 幂的",
+      "▁C HECK",
+      "▁ CHECK",
+      "▁Q uart",
+      "▁Qu art",
+      "▁E arlier",
+      "▁Ear lier",
+      "▁Earl ier",
+      "▁el apsed",
+      "▁ elapsed",
+      "打 赢",
+      "骨 架",
+      "打破 了",
+      "打 破了",
+      "Card s",
+      "Car ds",
+      "C ards",
+      "千 瓦",
+      "药 店",
+      "阳 春",
+      "\\' {",
+      "Bo th",
+      "Bot h",
+      "B oth",
+      "▁R ao",
+      "▁Ra o",
+      "▁S ARS",
+      "▁SA RS",
+      "▁SAR S",
+      "▁as bestos",
+      "微信 号",
+      "微 信号",
+      "此题 考查",
+      "▁H ond",
+      "▁Hon d",
+      "▁Ho nd",
+      "▁mes enchymal",
+      "技术 推广",
+      "ce t",
+      "c et",
+      "把 这些",
+      "根据 不同",
+      "east ern",
+      "eas tern",
+      "e astern",
+      "▁l iking",
+      "▁li king",
+      "▁lik ing",
+      "nWrit ers",
+      "nWrite rs",
+      "iter items",
+      "▁mar itime",
+      "com pletion",
+      "comp letion",
+      "▁character ised",
+      "在 国家",
+      "ém on",
+      "é mon",
+      "我们 必须",
+      "pro be",
+      "pr obe",
+      "prob e",
+      "p robe",
+      "▁Met eor",
+      "到 场",
+      "藏 书",
+      "也 只能",
+      "bl ic",
+      "b lic",
+      "▁J UST",
+      "▁b aths",
+      "▁bath s",
+      "▁bat hs",
+      "▁ba ths",
+      "poly mer",
+      "p olymer",
+      "▁ste aling",
+      "▁steal ing",
+      "bo s",
+      "b os",
+      "不满 意",
+      "不 满意",
+      "eo us",
+      "e ous",
+      "ri nk",
+      "rin k",
+      "r ink",
+      "▁Mc N",
+      "▁g cd",
+      "▁ gcd",
+      "负责 同志",
+      "oc ado",
+      "oca do",
+      "二〇一七年 十二",
+      "二〇一七年十 二",
+      "▁sharp ly",
+      "▁shar ply",
+      "敏 锐",
+      "▁\" $(",
+      "▁\"$ (",
+      "明确 表示",
+      "是一家 集",
+      "OL DER",
+      "OLD ER",
+      "AT FORM",
+      "ep rint",
+      "e print",
+      "to Array",
+      "▁contract ing",
+      "▁contr acting",
+      "龙 华",
+      "]\" \\",
+      "] \"\\",
+      "▁b ac",
+      "▁ba c",
+      "▁ bac",
+      "nPl aying",
+      "nPlay ing",
+      "▁Win ners",
+      "▁Winn ers",
+      "▁Winner s",
+      "▁poly peptide",
+      "▁polype ptide",
+      "in di",
+      "ind i",
+      "i ndi",
+      "on ance",
+      "ona nce",
+      "Feed back",
+      "▁o ocytes",
+      "▁est radiol",
+      "经 商",
+      "Li te",
+      "L ite",
+      "UI nt",
+      "U Int",
+      "▁H OME",
+      "▁HO ME",
+      "▁ HOME",
+      "工 匠",
+      "从 这个",
+      "周 杰伦",
+      "ox el",
+      "▁h erm",
+      "▁he rm",
+      "▁her m",
+      "H W",
+      "kb d",
+      "k bd",
+      "▁W id",
+      "▁Wi d",
+      "不同 程度的",
+      "不同程度 的",
+      "▁Algebra ic",
+      "将 要",
+      "直 销",
+      "▁p erm",
+      "▁per m",
+      "▁pe rm",
+      "▁ perm",
+      "▁C heng",
+      "▁Ch eng",
+      "▁Che ng",
+      "▁Chen g",
+      "▁e ffic",
+      "▁eff ic",
+      "▁Ver tical",
+      "▁Vert ical",
+      "▁ Vertical",
+      "想 念",
+      "ki d",
+      "k id",
+      "百分 比",
+      "▁c mp",
+      "▁cm p",
+      "▁ cmp",
+      "感觉 自己",
+      "fol ios",
+      "folio s",
+      "杜 鹃",
+      "me et",
+      "▁Em my",
+      "▁S erve",
+      "▁Ser ve",
+      "▁Serv e",
+      "▁ Serve",
+      "Key word",
+      "▁ul cers",
+      "▁ulcer s",
+      "▁Maur ice",
+      "▁conn ective",
+      "▁connect ive",
+      "最初 的",
+      "还会 有",
+      "还 会有",
+      "em is",
+      "emi s",
+      "e mis",
+      "确诊 病例",
+      "nDe tails",
+      "nDet ails",
+      "n Details",
+      "▁dis connected",
+      "▁disconnect ed",
+      "专 为",
+      "冷 暖",
+      "地 去",
+      "通讯 员",
+      "}} }}",
+      "}}} }",
+      "} }}}",
+      "GR APH",
+      "en den",
+      "end en",
+      "ende n",
+      "▁H askell",
+      "á t",
+      "▁A ges",
+      "▁Ag es",
+      "▁Age s",
+      "nc ount",
+      "nco unt",
+      "n count",
+      "▁Car pet",
+      "▁Carp et",
+      "▁fire wall",
+      "contin uous",
+      "百 强",
+      "翻 身",
+      "喜欢 的人",
+      "喜欢的 人",
+      "▁H uff",
+      "▁Hu ff",
+      "▁生长 环境",
+      "▁sequ enced",
+      "▁sequence d",
+      "EX P",
+      "E XP",
+      "第一 家",
+      "第 一家",
+      "▁St yl",
+      "ed ited",
+      "edit ed",
+      "edi ted",
+      "▁in jective",
+      "▁inj ective",
+      "▁inject ive",
+      "▁ 概述",
+      "小 女孩",
+      "自我 的",
+      "自 我的",
+      "EM PL",
+      "EMP L",
+      "E MPL",
+      "▁z ur",
+      "▁zu r",
+      "AMP LE",
+      "A MPLE",
+      "plic ant",
+      "plica nt",
+      "▁Me asures",
+      "▁Measure s",
+      "▁Alexand ria",
+      "▁ Ł",
+      "他就 是",
+      "他 就是",
+      "上诉 请求",
+      "▁con tests",
+      "▁cont ests",
+      "▁contest s",
+      "整 套",
+      "ve lt",
+      "vel t",
+      "v elt",
+      "九 江",
+      "我 好",
+      "经 络",
+      "与时 俱",
+      "营业 额",
+      "透明 的",
+      "透 明的",
+      "un ov",
+      "uno v",
+      "u nov",
+      "▁[… ]\\",
+      "st ress",
+      "str ess",
+      "stre ss",
+      "▁learn er",
+      "▁lear ner",
+      "▁adapt ing",
+      "▁adap ting",
+      "▁dw elling",
+      "▁dwell ing",
+      "▁liqu idity",
+      "▁liquid ity",
+      "▁reminis cent",
+      "罗 拉",
+      "酸 钠",
+      "可谓 是",
+      "t ogether",
+      "▁bi ochemistry",
+      "多 糖",
+      "Tr uth",
+      "▁D ial",
+      "▁Di al",
+      "▁ Dial",
+      "np r",
+      "n pr",
+      "的 品质",
+      "ri be",
+      "rib e",
+      "r ibe",
+      "RE LEASE",
+      "pre ferred",
+      "prefer red",
+      "^ \\",
+      "yi i",
+      "y ii",
+      "有三 个",
+      "有 三个",
+      "的 含义",
+      "▁} $$",
+      "▁}$ $",
+      "▁ }$$",
+      "Vis it",
+      "Vi sit",
+      "▁O tto",
+      "▁Ott o",
+      "▁Ot to",
+      "▁comm itting",
+      "▁commit ting",
+      "对 数",
+      "秘 诀",
+      "移 除",
+      "随 心",
+      "变量 的",
+      "变 量的",
+      "▁P ip",
+      "▁Pi p",
+      "▁C ham",
+      "▁Ch am",
+      "▁Cha m",
+      "op ened",
+      "ope ned",
+      "open ed",
+      "▁ste ak",
+      "▁tr acer",
+      "▁tra cer",
+      "▁trace r",
+      "at ization",
+      "▁Mort gage",
+      "指数 幂",
+      "这些 年",
+      "值得 关注",
+      "▁d amping",
+      "▁dam ping",
+      "▁damp ing",
+      "▁Sub scribe",
+      "▁ Subscribe",
+      "▁ ∈",
+      "平行 的",
+      "平 行的",
+      "▁does nt",
+      "▁doesn t",
+      "准许或者 不准许撤诉",
+      "通 货",
+      "au dit",
+      "aud it",
+      "audi t",
+      "er mis",
+      "erm is",
+      "qu iet",
+      "本 田",
+      "Se x",
+      "S ex",
+      "音乐 会",
+      "问题 如下",
+      "ac ute",
+      "▁che fs",
+      "▁chef s",
+      "▁gr apes",
+      "▁grap es",
+      "▁gra pes",
+      "▁grape s",
+      "optim izer",
+      "optimize r",
+      "▁crypt ography",
+      "▁information al",
+      "▁inform ational",
+      "不 孕",
+      "处 级",
+      "富 强",
+      "抵 御",
+      "第二百零 五条",
+      "▁St ainless",
+      "银 耳",
+      "CR M",
+      "C RM",
+      "不 规则",
+      "染色 体",
+      "等 服务",
+      "第二 部分",
+      "▁l eng",
+      "▁le ng",
+      "▁len g",
+      "▁ü ber",
+      "▁po orer",
+      "▁poor er",
+      "Person al",
+      "Pers onal",
+      "P ersonal",
+      "▁induct ed",
+      "打 扰",
+      "SO D",
+      "S OD",
+      "ic her",
+      "ich er",
+      "iche r",
+      "i cher",
+      "▁fe ather",
+      "▁feat her",
+      "▁ro asted",
+      "▁roast ed",
+      "外 加",
+      "E uro",
+      "square d",
+      "squ ared",
+      "▁Arch ae",
+      "▁tors ion",
+      "IR S",
+      "I RS",
+      "▁м о",
+      "▁ мо",
+      "cl usters",
+      "clus ters",
+      "cluster s",
+      "▁mark eters",
+      "▁market ers",
+      "自 拍",
+      "正常 人",
+      "理解 和",
+      "相比 较",
+      "相 比较",
+      "▁F unc",
+      "▁Fun c",
+      "▁Fu nc",
+      "▁ Func",
+      "▁fund amentals",
+      "▁fundamental s",
+      "▁fundament als",
+      "缓 交",
+      "S vg",
+      "第十 八",
+      "第 十八",
+      "课程 的",
+      "an sen",
+      "ans en",
+      "tt ime",
+      "tti me",
+      "t time",
+      "en cycl",
+      "enc ycl",
+      "ency cl",
+      "ig uous",
+      "igu ous",
+      "▁R aven",
+      "▁Ra ven",
+      "▁Rav en",
+      "▁cit rus",
+      "呜 呜",
+      "le w",
+      "l ew",
+      "事物 的",
+      "事 物的",
+      "好吃 的",
+      "好 吃的",
+      "百分 百",
+      "这 辈子",
+      "▁B is",
+      "▁Bi s",
+      "对该 领域的",
+      "▁P irates",
+      "▁Pi rates",
+      "▁Pir ates",
+      "▁prep ares",
+      "▁prepar es",
+      "▁prepare s",
+      "▁pet roleum",
+      "国 情",
+      "彼 岸",
+      "我 真",
+      "este em",
+      "十二 五",
+      "▁dou bling",
+      "▁doub ling",
+      "均 需",
+      "关联 性",
+      "志愿 军",
+      "执行 款",
+      "面上 的",
+      "面 上的",
+      "▁J ill",
+      "▁Ji ll",
+      "末 日",
+      "十二 年",
+      "十 二年",
+      "sn ow",
+      "s now",
+      "▁v or",
+      "▁vo r",
+      "▁ vor",
+      "Abs olute",
+      "orge town",
+      "orget own",
+      "▁Hels inki",
+      "▁Autom ation",
+      "机 房",
+      "▁c hin",
+      "▁ch in",
+      "▁chi n",
+      "▁ chin",
+      "loc als",
+      "local s",
+      "▁wid ow",
+      "▁wi dow",
+      "▁med all",
+      "▁medal l",
+      "tre quire",
+      "t require",
+      "▁k itchens",
+      "▁kitchen s",
+      "从 左",
+      "也 一样",
+      "涵盖 了",
+      "通过 分析",
+      "Al bum",
+      "full Name",
+      "▁Benef its",
+      "tright arrow",
+      "t rightarrow",
+      "R W",
+      "缘 故",
+      "离家 出走",
+      "In struction",
+      "分 流",
+      "迷 惑",
+      "不 至于",
+      "出了 一",
+      "出 了一",
+      "拿到 了",
+      "拿 到了",
+      "), \\\\",
+      "),\\ \\",
+      ") ,\\\\",
+      "▁c oup",
+      "▁co up",
+      "▁cou p",
+      "▁Ag greg",
+      "▁ Aggreg",
+      "▁hash lib",
+      "▁expl orer",
+      "▁explore r",
+      "▁explo rer",
+      "▁explor er",
+      "▁ explorer",
+      "▁Individual s",
+      "千 载",
+      "汇 票",
+      "de sk",
+      "des k",
+      "d esk",
+      "管理 学院",
+      "管理学 院",
+      "▁in fl",
+      "▁i nfl",
+      "▁inf l",
+      "▁s lab",
+      "▁sl ab",
+      "leg ant",
+      "sp inal",
+      "spin al",
+      "spi nal",
+      "▁M iche",
+      "▁Mich e",
+      "▁Mi che",
+      "▁Mic he",
+      "▁scar ce",
+      "▁scarc e",
+      "▁trend y",
+      "我 县",
+      "CR R",
+      "C RR",
+      "活性 炭",
+      "的根 本",
+      "的 根本",
+      "不断 完善",
+      "Tr ade",
+      "Tra de",
+      "▁C ube",
+      "▁Cub e",
+      "▁Cu be",
+      "▁ Cube",
+      "▁J oined",
+      "▁Jo ined",
+      "▁Join ed",
+      "冻结 财产的规定",
+      "▁anal ogy",
+      "▁analog y",
+      "▁ana logy",
+      "det ection",
+      "detect ion",
+      "▁off enders",
+      "н у",
+      "降 到",
+      "也未 能",
+      "也 未能",
+      "月份 的",
+      "▁wh istle",
+      "▁whis tle",
+      "▁hist ology",
+      "▁resil ient",
+      "▁irrad iated",
+      "也 太",
+      "il ight",
+      "ili ght",
+      "i light",
+      "尖 锐",
+      "抗 疫",
+      "在上 诉",
+      "在 上诉",
+      "精美 的",
+      "精 美的",
+      "▁S NP",
+      "▁SN P",
+      "▁S UP",
+      "▁SU P",
+      "▁ SUP",
+      "▁o we",
+      "▁ow e",
+      "▁ owe",
+      "相关 规定",
+      "▁Pa so",
+      "▁Pas o",
+      "▁R ising",
+      "▁Ri sing",
+      "▁Ris ing",
+      "依法由 代理审判员",
+      "▁destroy ing",
+      "X A",
+      "万 科",
+      "公司 与",
+      "聊城 市",
+      "聊 城市",
+      "▁c left",
+      "▁cl eft",
+      "▁cle ft",
+      "in jection",
+      "inject ion",
+      "inj ection",
+      "io graphic",
+      "i ographic",
+      "▁exhaust ed",
+      "其他需要裁定 解决的事项",
+      "案件适用法律 若干问题的规定",
+      "悬 崖",
+      "自己 喜欢的",
+      "自己喜欢 的",
+      "▁Pl ants",
+      "▁Plan ts",
+      "▁Plant s",
+      "▁age ing",
+      "▁pap ill",
+      "ns ervices",
+      "n services",
+      "剂 的",
+      "重 磅",
+      "老师 和",
+      "▁A LS",
+      "▁AL S",
+      "▁ ALS",
+      "▁s to",
+      "▁st o",
+      "▁ sto",
+      "▁s vg",
+      "▁sv g",
+      "▁ svg",
+      "一条 直线",
+      "可以 实现",
+      "▁b rig",
+      "▁br ig",
+      "▁ke rat",
+      "▁ker at",
+      "事 发",
+      "年 均",
+      "点 上",
+      "电影 的",
+      "ud er",
+      "ude r",
+      "u der",
+      "Or acle",
+      "nG rand",
+      "nGr and",
+      "n Grand",
+      "选择 正确答案",
+      "▁exp resses",
+      "▁express es",
+      "▁expr esses",
+      "▁comment ing",
+      "▁constit uent",
+      "▁constitu ent",
+      "翻 转",
+      "▁南 京",
+      "▁ 南京",
+      "请 大家",
+      "这 都是",
+      "Access Token",
+      "nW ashington",
+      "▁discrim inate",
+      "® ,",
+      "▁Ac qu",
+      "▁Ad ds",
+      "▁Add s",
+      "▁de rives",
+      "▁der ives",
+      "▁deriv es",
+      "▁derive s",
+      "int roduction",
+      "intro duction",
+      "Request Handler",
+      "J J",
+      "借 用",
+      "势 必",
+      "um bo",
+      "umb o",
+      "是最 好的",
+      "是 最好的",
+      "De leted",
+      "Delete d",
+      "Del eted",
+      "timestamp s",
+      "为 她",
+      "住 户",
+      "尿 道",
+      "摩 尔",
+      "第三 条",
+      "第 三条",
+      "罪行 的",
+      "罪 行的",
+      "签字 确认",
+      "nL oad",
+      "nLo ad",
+      "n Load",
+      "▁根据 您的",
+      "▁ 根据您的",
+      "▁horizontal ly",
+      "▁horiz ontally",
+      "V L",
+      "且 仅",
+      "RE L",
+      "R EL",
+      "所 依据",
+      "▁F ur",
+      "▁Fu r",
+      "▁T iny",
+      "▁Ti ny",
+      "▁Tin y",
+      "pher ical",
+      "▁Un iform",
+      "▁Uni form",
+      "▁ Uniform",
+      "▁computer ized",
+      "怀 着",
+      "教科 书",
+      "▁G othic",
+      "▁plaus ible",
+      "大 树",
+      "在 收到",
+      "▁Chief s",
+      "▁Mon ter",
+      "▁Mont er",
+      "▁Mo nter",
+      "▁Monte r",
+      "▁Le aders",
+      "▁Lead ers",
+      "▁Leader s",
+      "▁str angers",
+      "▁strange rs",
+      "▁stranger s",
+      "▁strang ers",
+      "适用简易程序 公开开庭进行了审理",
+      "春 雨",
+      "成本 低",
+      "某的 证言",
+      "某 的证言",
+      "▁V ale",
+      "▁Val e",
+      "▁Va le",
+      "▁bar bec",
+      "在道路上 醉酒驾驶机动车",
+      "▁discrep ancy",
+      "传 销",
+      "翩 翩",
+      "聘 用",
+      "通常 是",
+      "tr aff",
+      "tra ff",
+      "▁E rik",
+      "▁Er ik",
+      "▁b ets",
+      "▁be ts",
+      "▁bet s",
+      "单 体",
+      "es k",
+      "e sk",
+      "▁重 庆",
+      "▁ 重庆",
+      "长 三角",
+      "▁A FC",
+      "▁AF C",
+      "▁P ert",
+      "▁Per t",
+      "▁Pe rt",
+      "▁M yers",
+      "▁My ers",
+      "▁Ku wait",
+      "▁Emb assy",
+      "新 旧",
+      "触 控",
+      "▁K Y",
+      "▁ KY",
+      "Gr eg",
+      "G reg",
+      "am oto",
+      "amo to",
+      "向本院 提供",
+      "▁con sect",
+      "▁cons ect",
+      "▁consec t",
+      "▁Diagn ostic",
+      "▁unf amiliar",
+      "gt k",
+      "g tk",
+      "Mut able",
+      "Mu table",
+      "M utable",
+      "▁C andid",
+      "▁Can did",
+      "▁Cand id",
+      "▁mar ital",
+      "▁re acted",
+      "▁react ed",
+      "Ka b",
+      "K ab",
+      "an sk",
+      "ans k",
+      "mo ving",
+      "mov ing",
+      "m oving",
+      "王 晓",
+      "母亲 的",
+      "表达 了",
+      "MAR K",
+      "M ARK",
+      "被 他",
+      "\"{ %",
+      "\" {%",
+      "与 直线",
+      "it in",
+      "iti n",
+      "i tin",
+      "▁maneu ver",
+      "▁ â",
+      "校 内",
+      "芳 香",
+      "故事 的",
+      "故 事的",
+      "is tas",
+      "ist as",
+      "ista s",
+      "ri val",
+      "riv al",
+      "r ival",
+      "ug uese",
+      "ugu ese",
+      "▁post partum",
+      "外 侧",
+      "理 所",
+      "cx x",
+      "c xx",
+      "漫长 的",
+      "漫 长的",
+      "ans om",
+      "▁Ag ents",
+      "▁Agent s",
+      "▁cr acking",
+      "▁crack ing",
+      "▁metaph or",
+      "浙江 大学",
+      "ot ropy",
+      "otrop y",
+      "otr opy",
+      "可 可",
+      "岳 阳",
+      "益 气",
+      "媒体 的",
+      "媒 体的",
+      "fa rm",
+      "far m",
+      "f arm",
+      "▁A gn",
+      "▁Ag n",
+      "▁L ily",
+      "▁Li ly",
+      "▁Lil y",
+      "parse rs",
+      "parser s",
+      "pars ers",
+      "河 水",
+      "▁D ee",
+      "▁De e",
+      "▁We bb",
+      "▁Web b",
+      "nS aint",
+      "nU ntil",
+      "nUn til",
+      "n Until",
+      "▁give away",
+      "再也 不",
+      "再 也不",
+      "ST AMP",
+      "STA MP",
+      "▁Green e",
+      "▁Gree ne",
+      "▁Gre ene",
+      "勋 章",
+      "代表 了",
+      "ik ed",
+      "ike d",
+      "▁s inc",
+      "▁sin c",
+      "▁si nc",
+      "Mock ito",
+      "riter ion",
+      "▁flav ours",
+      "▁flavour s",
+      "分 红",
+      "长 了",
+      "隔 热",
+      "opol ys",
+      "opoly s",
+      "opo lys",
+      "▁V irus",
+      "▁Vir us",
+      "▁Vi rus",
+      "block ing",
+      "b locking",
+      "由题意 可得",
+      "▁J D",
+      "▁ JD",
+      "AT IONAL",
+      "ATION AL",
+      "▁Gal ois",
+      "▁Di agram",
+      "一 出",
+      "上 古",
+      "上 都",
+      "之 妻",
+      "农 商",
+      "▁S uk",
+      "▁Su k",
+      "▁O mega",
+      "▁Om ega",
+      "▁ Omega",
+      "▁gal lon",
+      "▁gall on",
+      "对该领域的 了解",
+      "随 风",
+      "齐 鲁",
+      "有限 元",
+      "八条 之规定",
+      "八 条之规定",
+      "请仔细 阅读选项",
+      "请仔细阅读 选项",
+      "sc enario",
+      "您 对该领域的了解",
+      "▁dec lines",
+      "▁decl ines",
+      "▁decline s",
+      "这是一道 关于多学科知识的选择题",
+      "中 去",
+      "居 多",
+      "牌 照",
+      "x or",
+      "C rypt",
+      "B rowse",
+      "▁N orton",
+      "▁Nor ton",
+      "传 出",
+      "依 题意",
+      "oy er",
+      "oye r",
+      "o yer",
+      "vin e",
+      "vi ne",
+      "v ine",
+      "▁A FP",
+      "▁AF P",
+      "一如 既往",
+      "审慎 考虑",
+      "▁C airo",
+      "▁Ca iro",
+      "▁C ivic",
+      "▁Ci vic",
+      "In jection",
+      "Inject ion",
+      "停 在",
+      "现 存",
+      "▁用 法",
+      "▁ 用法",
+      "因 生活",
+      "nL es",
+      "nLe s",
+      "n Les",
+      "po on",
+      "p oon",
+      "▁B ax",
+      "▁Ba x",
+      "ut tle",
+      "utt le",
+      "u ttle",
+      "证据确实 充分",
+      "▁over write",
+      "▁overw rite",
+      "Ap plications",
+      "Application s",
+      "▁hypoth alamic",
+      "失 常",
+      "美 方",
+      "致 的",
+      "Au g",
+      "A ug",
+      "▁/ /\\",
+      "▁// \\",
+      "▁ //\\",
+      "nMod ern",
+      "▁For ums",
+      "▁Forum s",
+      "▁Fo rums",
+      "二〇一八年 八月",
+      "bound ary",
+      "p urchase",
+      "自 贸",
+      "le a",
+      "l ea",
+      "为 宗旨",
+      "▁c alf",
+      "▁ca lf",
+      "▁cal f",
+      "▁Rich ards",
+      "▁Richard s",
+      "/ …",
+      "Sp y",
+      "S py",
+      "ul ed",
+      "ule d",
+      "u led",
+      "都有 哪些",
+      "都 有哪些",
+      "是一道 选择题",
+      "是 一道选择题",
+      "▁anal yzer",
+      "▁analy zer",
+      "▁analyz er",
+      "▁analyze r",
+      "▁ analyzer",
+      "▁proceed ed",
+      "▁du plication",
+      "▁duplic ation",
+      "名 声",
+      "咫 尺",
+      "霹 雳",
+      "AL K",
+      "IT A",
+      "I TA",
+      "▁B rock",
+      "▁Br ock",
+      "▁Bro ck",
+      "Spring Boot",
+      "月 明",
+      "交汇 处",
+      "出席 了",
+      "▁e book",
+      "▁eb ook",
+      "▁select s",
+      "▁sel ects",
+      "白 石",
+      "▁k eras",
+      "▁ke ras",
+      "▁ker as",
+      "▁ keras",
+      "办理 结婚登记",
+      "act ivities",
+      "activ ities",
+      "▁ 燕",
+      "向 着",
+      "秀 丽",
+      "质 检",
+      "实 实在",
+      "ge ons",
+      "geo ns",
+      "geon s",
+      "▁pe dal",
+      "▁ped al",
+      "created At",
+      "▁Fe aturing",
+      "责 编",
+      "高 二",
+      "ho g",
+      "h og",
+      "微博 上",
+      "nres ult",
+      "n result",
+      "▁Bol ivia",
+      "▁Volunte er",
+      "▁sub scribed",
+      "▁subscrib ed",
+      "▁subscribe d",
+      "▁subs cribed",
+      "| <",
+      "▁ 🙂",
+      "服 药",
+      "ib m",
+      "i bm",
+      "м ен",
+      "听 起来",
+      "规划 和",
+      "▁C HO",
+      "▁CH O",
+      "▁ CHO",
+      "▁\\ ')",
+      "▁\\' )",
+      "▁ \\')",
+      "iss ued",
+      "issue d",
+      "issu ed",
+      "is ement",
+      "ise ment",
+      "▁H ansen",
+      "▁Han sen",
+      "▁Hans en",
+      "▁pat rons",
+      "▁patron s",
+      "王 文",
+      "肤 色",
+      "驾 驭",
+      "$ ^{",
+      "买了 一",
+      "买 了一",
+      "二十 五",
+      "二 十五",
+      "会 发现",
+      "or bed",
+      "orb ed",
+      "▁w ird",
+      "▁wi rd",
+      "▁pep pers",
+      "▁pepper s",
+      "▁number ing",
+      "你 得",
+      "进 驻",
+      "ES P",
+      "E SP",
+      "nE ast",
+      "n East",
+      "host ed",
+      "ho sted",
+      "pro ton",
+      "pr oton",
+      "prot on",
+      "proto n",
+      "▁S ympt",
+      "▁Sym pt",
+      "▁Sup plement",
+      "察 觉",
+      "接 力",
+      "日 均",
+      "▁A PC",
+      "▁AP C",
+      "▁D SL",
+      "▁DS L",
+      "大学 生的",
+      "大学生 的",
+      "大 学生的",
+      "请你仔细 读题",
+      "从选项中 选择出",
+      "▁sail ors",
+      "如上 是一道选择题",
+      "ivari able",
+      "ivar iable",
+      "i variable",
+      "le f",
+      "l ef",
+      "▁` [",
+      "并 确定",
+      "的路 程",
+      "的 路程",
+      "可以 达到",
+      "可 以达到",
+      "判处有期徒刑 六个月",
+      "MP I",
+      "M PI",
+      "由 本院",
+      "SI GN",
+      "SIG N",
+      "S IGN",
+      "贯彻 执行",
+      "▁Pl atinum",
+      "▁disp ersed",
+      "▁dispers ed",
+      "▁insp ector",
+      "▁inspect or",
+      "▁transplant ed",
+      "通 告",
+      "BA R",
+      "B AR",
+      "读者 的",
+      "读 者的",
+      "at ore",
+      "ator e",
+      "ato re",
+      "▁L atex",
+      "▁La tex",
+      "▁Lat ex",
+      "▁Late x",
+      "▁gl ial",
+      "▁gli al",
+      "▁m ongo",
+      "▁mon go",
+      "▁mo ngo",
+      "▁mong o",
+      "▁ mongo",
+      "具有 重要意义",
+      "为 社会",
+      "是指 在",
+      "▁O st",
+      "▁Os t",
+      "▁T ue",
+      "▁Tu e",
+      "▁M arl",
+      "▁Mar l",
+      "▁Ma rl",
+      "is Array",
+      "uc lease",
+      "ucle ase",
+      "▁H uawei",
+      "▁night mare",
+      "壁 画",
+      "桃 李",
+      "航 海",
+      "和 孩子",
+      "中心 小学",
+      "tl ist",
+      "t list",
+      "▁THE N",
+      "▁TH EN",
+      "▁pop up",
+      "▁ popup",
+      "▁tor ture",
+      "▁tort ure",
+      "part icles",
+      "particle s",
+      "p articles",
+      "▁fil enames",
+      "▁file names",
+      "▁filename s",
+      "▁waters hed",
+      "B W",
+      "免 除",
+      "嫁 给",
+      "窒 息",
+      "血 脉",
+      "re lu",
+      "rel u",
+      "▁T odo",
+      "▁To do",
+      "▁ Todo",
+      "▁w edge",
+      "▁wed ge",
+      "▁ wedge",
+      "内容 如下所示",
+      "民事执行中 查封",
+      "Read Line",
+      "▁brace let",
+      "ke a",
+      "k ea",
+      "▁H ew",
+      "▁He w",
+      "nw ork",
+      "n work",
+      "▁c ones",
+      "▁con es",
+      "▁co nes",
+      "▁cone s",
+      "▁st agger",
+      "▁North western",
+      "▁Northwest ern",
+      "他 有",
+      "日 到",
+      "▁G ig",
+      "▁Gi g",
+      "▁E den",
+      "▁Ed en",
+      "Is NullOr",
+      "Part icle",
+      "P article",
+      "H b",
+      "庄 严",
+      "政 党",
+      "青 梅",
+      "EU R",
+      "E UR",
+      "外面 的",
+      "外 面的",
+      "▁l ifts",
+      "▁li fts",
+      "▁lif ts",
+      "▁lift s",
+      "▁ana erobic",
+      "La zy",
+      "L azy",
+      "▁д ля",
+      "▁d warf",
+      "▁dw arf",
+      "let cher",
+      "ex ercise",
+      "▁come back",
+      "必 有",
+      "等 同",
+      "醒 了",
+      "中途 退",
+      "的女 子",
+      "的 女子",
+      "▁Equ ity",
+      "▁Eq uity",
+      "ress ions",
+      "ression s",
+      "r essions",
+      "▁Im proved",
+      "▁Improve d",
+      "▁vis ceral",
+      "一 连",
+      "免 税",
+      "冲 锋",
+      "妨 害",
+      "涡 轮",
+      "DA T",
+      "D AT",
+      "^* $",
+      "^ *$",
+      "想象 力",
+      "ac ian",
+      "aci an",
+      "acia n",
+      "▁Ad aptive",
+      "▁Ada ptive",
+      "▁Adapt ive",
+      "▁Pain ting",
+      "▁Paint ing",
+      "▁Pharm aceutical",
+      "永 生",
+      "二次 元",
+      "带 孩子",
+      "pl ug",
+      "▁N in",
+      "▁Ni n",
+      "tr uth",
+      "▁k ernels",
+      "▁kernel s",
+      "▁ker nels",
+      "he on",
+      "h eon",
+      "ens ory",
+      "ensor y",
+      "▁R iley",
+      "▁Ri ley",
+      "▁d rones",
+      "▁dr ones",
+      "▁dro nes",
+      "▁drone s",
+      "本次执行程序 应予终结",
+      "▁pancreat itis",
+      "eb a",
+      "e ba",
+      "yn ch",
+      "ync h",
+      "y nch",
+      "进行 研究",
+      "▁w itch",
+      "▁wit ch",
+      "▁ witch",
+      "Pl ayers",
+      "Player s",
+      "Play ers",
+      "P layers",
+      "bad ges",
+      "badge s",
+      "▁Cal if",
+      "▁Ca lif",
+      "▁condition er",
+      "就 让",
+      "身 旁",
+      "MC s",
+      "M Cs",
+      "俗 话说",
+      "me mo",
+      "mem o",
+      "m emo",
+      "nP rom",
+      "nPro m",
+      "nPr om",
+      "n Prom",
+      "经济 发展的",
+      "经济发展 的",
+      "▁cl utch",
+      "▁on board",
+      "或 在",
+      "看 向",
+      "ra j",
+      "r aj",
+      "好 莱坞",
+      "▁P II",
+      "▁PI I",
+      "▁ PII",
+      "nw ell",
+      "nwe ll",
+      "n well",
+      "tro viral",
+      "▁H ampton",
+      "▁Ham pton",
+      "▁Hamp ton",
+      "▁spec ials",
+      "▁special s",
+      "圣 经",
+      "太 后",
+      "幼 虫",
+      "数据 线",
+      "显示 出",
+      "\\\\ }$$",
+      "\\\\} $$",
+      "\\\\}$ $",
+      "fr eeze",
+      "free ze",
+      "f reeze",
+      "▁oxid ized",
+      "瘙 痒",
+      "追 回",
+      "飞行 员",
+      "▁} }{",
+      "▁}} {",
+      "▁ }}{",
+      "由 审判员",
+      "li ers",
+      "lier s",
+      "lie rs",
+      "l iers",
+      "族自治州 中级人民法院",
+      "哈 萨克",
+      "空气 中",
+      "▁web cam",
+      "▁dipl oma",
+      "▁diplom a",
+      "▁ 试",
+      "俄 国",
+      "的 组合",
+      "主要 内容",
+      "问卷 调查",
+      "pl asty",
+      "p lasty",
+      "▁Mer ge",
+      "▁Merg e",
+      "▁ Merge",
+      "enn ials",
+      "ennial s",
+      "▁det ects",
+      "▁detect s",
+      "劳务 合同纠纷一案",
+      "人 脸",
+      "so on",
+      "s oon",
+      "it ating",
+      "ita ting",
+      "itat ing",
+      "礼 拜",
+      "Mu r",
+      "M ur",
+      "就 够了",
+      "▁pr ince",
+      "▁princ e",
+      "▁pri nce",
+      "▁crit ique",
+      "监 制",
+      "做好 了",
+      "做 好了",
+      "▁ext rav",
+      "▁extra v",
+      "▁extr av",
+      "▁Re porter",
+      "▁Rep orter",
+      "▁Report er",
+      "▁ Reporter",
+      "▁contract ile",
+      "▁unexpected ly",
+      "鼻 炎",
+      "我们 有",
+      "▁d ei",
+      "▁de i",
+      "sc anner",
+      "scan ner",
+      "▁abs urd",
+      "▁ins erts",
+      "▁insert s",
+      "鸟 类",
+      "bul k",
+      "b ulk",
+      "▁S lot",
+      "▁Sl ot",
+      "▁Slo t",
+      "▁ Slot",
+      "▁h eur",
+      "▁he ur",
+      "可供 您选择",
+      "ou ched",
+      "ouch ed",
+      "sec rets",
+      "secret s",
+      "Pr eference",
+      "Pre ference",
+      "P reference",
+      "Compat Activity",
+      "红 茶",
+      "青 椒",
+      "▁{ %",
+      "▁ {%",
+      "中国 画",
+      "中 国画",
+      "是 这个",
+      "的 优点",
+      "st ay",
+      "sta y",
+      "荆 州",
+      "nU N",
+      "n UN",
+      "桂林 市",
+      "▁C ly",
+      "▁Cl y",
+      "▁F unk",
+      "▁Fun k",
+      "▁Fu nk",
+      "▁rad ios",
+      "▁radio s",
+      "▁radi os",
+      "▁tr actor",
+      "▁tra ctor",
+      "▁tract or",
+      "starts With",
+      "▁for bidden",
+      "香 料",
+      "医疗 卫生",
+      "然后 告诉我",
+      "请告诉我 您的选择",
+      "conf irmed",
+      "confirm ed",
+      "日向本院提出 财产保全",
+      "▁narr atives",
+      "▁narrative s",
+      "请仔细阅读 下面这道选择题",
+      "请 仔细阅读下面这道选择题",
+      "文 旅",
+      "中 全会",
+      "的 课程",
+      "某某 负担",
+      "某 某负担",
+      "▁Sal em",
+      "▁Sa lem",
+      "▁Sale m",
+      "▁st igma",
+      "▁phosphory lated",
+      "瓷 器",
+      "长 子",
+      "Sk y",
+      "S ky",
+      "高 精度",
+      "▁a ur",
+      "▁au r",
+      "▁ aur",
+      "▁l ord",
+      "▁lo rd",
+      "▁ lord",
+      "De ploy",
+      "Dep loy",
+      "card ia",
+      "古 装",
+      "备 战",
+      "武 警",
+      "金 字",
+      "▁号 码",
+      "▁ 号码",
+      "去年 的",
+      "去 年的",
+      "▁Tw elve",
+      "▁s prite",
+      "▁sp rite",
+      "▁spr ite",
+      "▁ sprite",
+      "中华人民共和国 土地管理法",
+      "追 捧",
+      "速度 和",
+      "速 度和",
+      "▁J ag",
+      "▁Ja g",
+      "Back up",
+      "▁Be tty",
+      "▁Bet ty",
+      "知识渊博 的人",
+      "▁Bal let",
+      "▁Ball et",
+      "等 诸多",
+      "组织 中",
+      "il in",
+      "ili n",
+      "i lin",
+      "更多 信息",
+      "注册 资金",
+      "▁he ir",
+      "▁p erv",
+      "▁per v",
+      "▁pe rv",
+      "▁pay day",
+      "园 艺",
+      "iz i",
+      "i zi",
+      "多万 元",
+      "多 万元",
+      "添加 了",
+      "添 加了",
+      "您认为 的",
+      "▁ 占地面积",
+      "人民卫生 出版社",
+      "▁anticip ation",
+      "▁antic ipation",
+      "N X",
+      "读 写",
+      "EP S",
+      "E PS",
+      "应 依法",
+      "Lo ss",
+      "L oss",
+      "继续 进行",
+      "AC TER",
+      "ACT ER",
+      "▁pl ur",
+      "▁L opez",
+      "▁Lo pez",
+      "re levant",
+      "▁hon esty",
+      "▁honest y",
+      "▁hone sty",
+      "▁individual ized",
+      "白 领",
+      "nK en",
+      "nKe n",
+      "n Ken",
+      "请用 您的",
+      "ar box",
+      "arb ox",
+      "▁M ens",
+      "▁Me ns",
+      "▁Men s",
+      "▁E qual",
+      "▁Equ al",
+      "▁Eq ual",
+      "▁ Equal",
+      "▁ 维",
+      "Ne ver",
+      "N ever",
+      "请您务必 注意",
+      "在回答 以下这道",
+      "问题和选项 的内容",
+      "▁work sheet",
+      "▁works heet",
+      "从选项中 选定正确答案",
+      "具有挑战性的 选择题时",
+      "▁ast rocytes",
+      "▁astr ocytes",
+      "请您务必注意 问题和选项的内容",
+      "CO PY",
+      "C OPY",
+      "pon ge",
+      "po nge",
+      "p onge",
+      "Sec ure",
+      "▁o zone",
+      "▁oz one",
+      "▁L ounge",
+      "▁Lou nge",
+      "▁Lo unge",
+      "%%%% %%%%",
+      "▁David son",
+      "▁in efficient",
+      "交 织",
+      "相比 于",
+      "醉酒 后",
+      "醉 酒后",
+      "▁B ali",
+      "▁Bal i",
+      "▁Ba li",
+      "bra him",
+      "工贸 有限公司",
+      "▁De bian",
+      "▁Deb ian",
+      "▁bat ting",
+      "▁batt ing",
+      "审慎考虑 您的选择",
+      "并告诉我 您认为的",
+      "请用您的 专业知识",
+      "nEvery one",
+      "▁work outs",
+      "▁workout s",
+      "▁summar izes",
+      "▁summarize s",
+      "以下备选答案 可供您选择",
+      "请用您的专业知识 回答以下问题",
+      "南阳 市",
+      "南 阳市",
+      "以下 命令",
+      "▁l aughter",
+      "▁laugh ter",
+      "▁imp ressions",
+      "▁impression s",
+      "▁impress ions",
+      "约 占",
+      "理由 正当",
+      "dor ff",
+      "dorf f",
+      "gu ess",
+      "gue ss",
+      "▁a vid",
+      "▁av id",
+      "▁ avid",
+      "▁Wes ley",
+      "current Time",
+      "混 淆",
+      "迷 人",
+      "▁X Y",
+      "▁ XY",
+      "开放 式",
+      "指出 了",
+      "指 出了",
+      "ox id",
+      "li kes",
+      "like s",
+      "lik es",
+      "l ikes",
+      "▁C hal",
+      "▁Ch al",
+      "▁Cha l",
+      "▁V ERY",
+      "▁VE RY",
+      "▁ VERY",
+      "▁f erm",
+      "▁fe rm",
+      "▁fer m",
+      "nAd min",
+      "n Admin",
+      "▁s arcoma",
+      "▁sar coma",
+      "▁sarc oma",
+      "▁advert ise",
+      "▁advertis e",
+      "长 短",
+      "领 略",
+      "灵敏 度",
+      "▁S eth",
+      "▁Se th",
+      "▁Set h",
+      "open ssl",
+      "opens sl",
+      "XXXX XXXX",
+      "ell aneous",
+      "季 后",
+      "解 散",
+      "▁L t",
+      "的一 段",
+      "的 一段",
+      "起诉 必须",
+      "请选择 你认为",
+      "found land",
+      "д у",
+      "不 归",
+      "此 地",
+      "专 升本",
+      "如果 你想",
+      "如果你 想",
+      "请选择你认为 合适的答案",
+      "人 权",
+      "丫 头",
+      "丽 莎",
+      "营 利",
+      "转 子",
+      "都 喜欢",
+      "森林 公园",
+      "▁J agu",
+      "▁Ja gu",
+      "▁Jag u",
+      "p uppet",
+      "| [",
+      "LE N",
+      "L EN",
+      "fl uent",
+      "flu ent",
+      "f luent",
+      "▁pione ering",
+      "▁pioneer ing",
+      "心 率",
+      "OT H",
+      "O TH",
+      "▁{ /",
+      "▁ {/",
+      "吸血 鬼",
+      "请 审查",
+      "近 距离",
+      "严格 执行",
+      "▁Thom son",
+      "astruct ural",
+      "a structural",
+      "▁file system",
+      "▁files ystem",
+      "▁ filesystem",
+      "▁cent imeters",
+      "准 时",
+      "坏 人",
+      "移 交",
+      "…… [",
+      "fin e",
+      "fi ne",
+      "f ine",
+      "▁h oc",
+      "▁ho c",
+      "▁ hoc",
+      "ne num",
+      "nen um",
+      "n enum",
+      "urt les",
+      "urtle s",
+      "实际应用 领域",
+      "▁r ented",
+      "▁rent ed",
+      "他 俩",
+      "录 入",
+      "面 目",
+      "co rs",
+      "cor s",
+      "c ors",
+      "▁B yz",
+      "▁By z",
+      "▁b lu",
+      "▁bl u",
+      "发展 方向",
+      "人民检察院 以",
+      "▁d oping",
+      "▁do ping",
+      "▁dop ing",
+      "▁pet rol",
+      "最高人民法院关于 执行案件立案",
+      "发 给",
+      "失 控",
+      "次 于",
+      "起 初",
+      "▁ 开发商",
+      "第十 三章",
+      "第十三 章",
+      "第 十三章",
+      "选择 是什么",
+      "▁F loyd",
+      "▁Flo yd",
+      "get Type",
+      "▁cr acked",
+      "▁crack ed",
+      "▁contr asting",
+      "▁contrast ing",
+      "参 见",
+      "博物 院",
+      "在 生活中",
+      "▁Joy ce",
+      "▁str aps",
+      "▁strap s",
+      "▁disadvant age",
+      "刷 牙",
+      "在 同",
+      "提取 物",
+      "nI ce",
+      "随着 我国",
+      "▁t idal",
+      "▁ti dal",
+      "▁tid al",
+      "Drop down",
+      "Ext ended",
+      "▁obs essed",
+      "▁Attribute Error",
+      "随 处",
+      "非常 好的",
+      "非常好 的",
+      "▁stabil ize",
+      "凌 云",
+      "环 形",
+      "竹 林",
+      "乐 意",
+      "Ow n",
+      "O wn",
+      "Draw er",
+      "D rawer",
+      "st ance",
+      "stan ce",
+      "sta nce",
+      "{{' }}(",
+      "▁ch eek",
+      "▁che ek",
+      "cont ained",
+      "▁sandwich es",
+      "▁sandw iches",
+      "他 将",
+      "口 中",
+      "引 言",
+      "水 务",
+      "MP a",
+      "M Pa",
+      "乘 进率",
+      "▁R ise",
+      "▁Ri se",
+      "▁Ris e",
+      "itor ium",
+      "ito rium",
+      "请审查 以下问题",
+      "▁el usive",
+      "▁ens ured",
+      "▁ensure d",
+      "▁St ability",
+      "作为一名 知识渊博的人",
+      "Ow nProperty",
+      "Own Property",
+      "南 岸",
+      "垫 付",
+      "烟 雾",
+      "智能 家居",
+      "an dan",
+      "and an",
+      "anda n",
+      "a ndan",
+      "ef fer",
+      "eff er",
+      "▁h ips",
+      "▁hip s",
+      "▁hi ps",
+      "▁ hips",
+      "On Init",
+      "O nInit",
+      "工业 和信息化",
+      "▁am azon",
+      "▁ amazon",
+      "人 流",
+      "eb o",
+      "e bo",
+      "▁A da",
+      "▁Ad a",
+      "▁S ham",
+      "▁Sh am",
+      "▁Sha m",
+      "Par allel",
+      "▁m ornings",
+      "▁mor nings",
+      "▁morning s",
+      "恐 慌",
+      "nA g",
+      "n Ag",
+      "▁c t",
+      "▁ ct",
+      "有一 条",
+      "有 一条",
+      "ce ls",
+      "cel s",
+      "c els",
+      "只是 为了",
+      "只 是为了",
+      "▁Al leg",
+      "▁All eg",
+      "▁Sc andin",
+      "▁Scan din",
+      "前 瞻",
+      "房 地",
+      "雨 后",
+      "tt o",
+      "t to",
+      "nH ol",
+      "n Hol",
+      "得到 结果",
+      "选项 涉及到",
+      "▁rad iographic",
+      "▁radio graphic",
+      "▁radi ographic",
+      "推 翻",
+      "人口 与",
+      "和 个人",
+      "ep pe",
+      "与 对照组",
+      "▁read iness",
+      "延 边",
+      "疤 痕",
+      "以下 的问题和",
+      "以下的 问题和",
+      "审阅 这道选择题",
+      "ter raform",
+      "以下的问题和 选项涉及到",
+      "并从给出的选项中 选定正确答案",
+      "交 响",
+      "▁p u",
+      "▁ pu",
+      "的最 高",
+      "的 最高",
+      "roll up",
+      "▁Fl uid",
+      "▁Flu id",
+      "math scr",
+      "▁S inger",
+      "▁Sing er",
+      "▁Sin ger",
+      "' _",
+      "▁ 廖",
+      "日 为",
+      "瓦 斯",
+      "跑 去",
+      "迷 信",
+      "特别 好",
+      "▁acc ents",
+      "▁ac cents",
+      "▁accent s",
+      "▁int imid",
+      "▁intim id",
+      "共 分",
+      "有 空",
+      "诅 咒",
+      "你不 能",
+      "你 不能",
+      "不良 影响",
+      "St rict",
+      "Str ict",
+      "▁centr ifug",
+      "▁digital ly",
+      "▁digit ally",
+      "征 地",
+      "飞 扬",
+      "在有 限",
+      "在 有限",
+      "▁G ap",
+      "▁Ga p",
+      "希望 通过",
+      "▁nom bre",
+      "▁ nombre",
+      "▁coinc ide",
+      "塔 尔",
+      "事业 发展",
+      "kr ainian",
+      "▁non invasive",
+      "切 的",
+      "d ream",
+      "▁head phones",
+      "▁re nin",
+      "▁r enin",
+      "Game Object",
+      "▁advert isers",
+      "▁advertis ers",
+      "▁advertise rs",
+      "国有 资产",
+      "▁ev angel",
+      "▁Mad agascar",
+      "▁authent icity",
+      "▁authentic ity",
+      "库 里",
+      "角 质",
+      "饭 后",
+      "at u",
+      "a tu",
+      "这 玩意",
+      "▁Ar ms",
+      "▁Arm s",
+      "▁S han",
+      "▁Sh an",
+      "▁Sha n",
+      "▁neuro logic",
+      "▁neur ologic",
+      "Error Message",
+      "So n",
+      "S on",
+      "吃 起来",
+      "植物 门",
+      "的 现状",
+      "sm tp",
+      "积极 开展",
+      "ass oci",
+      "asso ci",
+      "ment al",
+      "men tal",
+      "m ental",
+      "▁alk yl",
+      "▁b rides",
+      "▁br ides",
+      "▁brid es",
+      "▁bride s",
+      "▁sur plus",
+      "排 污",
+      "松 下",
+      "隐 蔽",
+      "li ps",
+      "lip s",
+      "l ips",
+      "▁just ification",
+      "出 众",
+      "液 化",
+      "nG O",
+      "n GO",
+      "圆的 半径",
+      "学生 的学习",
+      "学生的 学习",
+      "▁an nex",
+      "▁ann ex",
+      "▁anne x",
+      "▁ge ared",
+      "▁gear ed",
+      "z s",
+      "▁ 邹",
+      "几 分",
+      "鸡 翅",
+      "CI S",
+      "C IS",
+      "Un s",
+      "U ns",
+      "tg o",
+      "t go",
+      "问题 上",
+      "▁M ys",
+      "▁My s",
+      "推广 应用",
+      "oss ip",
+      "▁E duc",
+      "▁Ed uc",
+      "▁Edu c",
+      "▁Joh ann",
+      "it eration",
+      "ite ration",
+      "iter ation",
+      "w f",
+      "中 层",
+      "打 压",
+      "大大 的",
+      "大 大的",
+      "en ko",
+      "as InstanceOf",
+      "▁ ∠",
+      "tm ax",
+      "t max",
+      "▁M am",
+      "▁Ma m",
+      "▁CO DE",
+      "▁ CODE",
+      "▁t weak",
+      "▁twe ak",
+      "Sample s",
+      "Sam ples",
+      "S amples",
+      "in spect",
+      "ins pect",
+      "insp ect",
+      "▁Br owns",
+      "▁Brown s",
+      "▁Brow ns",
+      "▁fac ade",
+      "▁per ennial",
+      "▁un conscious",
+      "▁uncon scious",
+      "张 伟",
+      "民 意",
+      "它 本身",
+      "{} \\'",
+      "{}\\ '",
+      "{ }\\'",
+      "▁第 一章",
+      "▁第一 章",
+      "▁ 第一章",
+      ")^ {\\\\",
+      ")^{ \\\\",
+      ") ^{\\\\",
+      "stad t",
+      "sta dt",
+      "▁Recomm ended",
+      "▁Recommend ed",
+      "履 职",
+      "拓 片",
+      "活 在",
+      "S qu",
+      "Pro d",
+      "Pr od",
+      "P rod",
+      "▁A val",
+      "▁Av al",
+      "你 都",
+      "BL E",
+      "B LE",
+      "等式 的",
+      "等 式的",
+      "an gs",
+      "ang s",
+      "en et",
+      "ene t",
+      "e net",
+      "▁a vg",
+      "▁av g",
+      "▁ avg",
+      "印度 尼西亚",
+      "▁sim plex",
+      "▁simple x",
+      "▁simpl ex",
+      "调 配",
+      "最好 不要",
+      "▁M etric",
+      "▁Met ric",
+      "▁ Metric",
+      "▁Na ples",
+      "▁Nap les",
+      "▁disturb ed",
+      "▁touchdown s",
+      "远 处",
+      "雨 花",
+      "HD L",
+      "H DL",
+      "Ki n",
+      "K in",
+      "平均 每天",
+      "平均每 天",
+      "built in",
+      "▁reward ed",
+      "$: \\",
+      "$ :\\",
+      "的 记忆",
+      "的 诊断",
+      "▁L ey",
+      "▁Le y",
+      "▁Jen ny",
+      "▁Jenn y",
+      "其中 农业人口",
+      "▁un even",
+      "▁une ven",
+      "▁Pr event",
+      "▁Pre vent",
+      "▁nons ense",
+      "▁partition ing",
+      "▁Trans formation",
+      "▁Transform ation",
+      "不 买",
+      "留 住",
+      "▁CO NF",
+      "▁CON F",
+      "▁ CONF",
+      "▁e sta",
+      "▁est a",
+      "▁es ta",
+      "▁ esta",
+      "▁m ods",
+      "▁mod s",
+      "▁mo ds",
+      "▁ mods",
+      "▁C ardi",
+      "▁Car di",
+      "▁Card i",
+      "▁proceed ing",
+      "▁disappe arance",
+      "▁disappear ance",
+      "嘴 角",
+      "王 建",
+      "迎 合",
+      "so v",
+      "s ov",
+      "可以 进行",
+      "▁l ista",
+      "▁li sta",
+      "▁list a",
+      "▁ lista",
+      "海 贼",
+      "lb l",
+      "l bl",
+      "▁A rd",
+      "▁Ar d",
+      "城市 管理",
+      "是 错误的",
+      "越来 越高",
+      "越来越 高",
+      "且 申请执行人",
+      "▁or acle",
+      "▁ oracle",
+      "▁Act ivation",
+      "▁Activ ation",
+      "▁ Activation",
+      "下 山",
+      "人 居",
+      "穷 人",
+      "Ge orge",
+      "▁N oble",
+      "▁No ble",
+      "▁Nob le",
+      "▁H ollow",
+      "▁Hol low",
+      "▁Holl ow",
+      "▁retic ulum",
+      "白 斑",
+      "▁S ag",
+      "▁Sa g",
+      "他 把",
+      "啦 啦",
+      "木 马",
+      "案 款",
+      "试 管",
+      "驾 照",
+      "转换 成",
+      "转 换成",
+      "es us",
+      "e sus",
+      "struct uring",
+      "入 坑",
+      "演 义",
+      "的 类型",
+      "根据 原告",
+      "研究 报告",
+      "as uring",
+      "asu ring",
+      "Not Empty",
+      "▁migr aine",
+      "依法 须经批准的项目",
+      "| }",
+      "创造 出",
+      "病 患者",
+      "nc he",
+      "nch e",
+      "n che",
+      "▁E CM",
+      "▁EC M",
+      "ur ous",
+      "uro us",
+      "u rous",
+      "nu nder",
+      "nun der",
+      "n under",
+      "计算 器",
+      "中国 文化",
+      "通知 如下",
+      "fi ber",
+      "fib er",
+      "f iber",
+      "在 本次执行程序中",
+      "▁Ac counts",
+      "▁Account s",
+      "▁ Accounts",
+      "▁gang lion",
+      "▁Inter state",
+      "两 款",
+      "这 组",
+      "▁ 代表",
+      "依赖 于",
+      "▁Col our",
+      "▁Start up",
+      "▁ Startup",
+      "生效之日起 十日内",
+      "▁Chrom ium",
+      "inter action",
+      "围 观",
+      "得 罪",
+      "om od",
+      "omo d",
+      "o mod",
+      "▁H ers",
+      "▁He rs",
+      "▁Her s",
+      "if olia",
+      "dim ensions",
+      "dimension s",
+      "d imensions",
+      "sp awn",
+      "不 甘",
+      "汉 书",
+      "语 境",
+      "In B",
+      "I nB",
+      "了一 年",
+      "了 一年",
+      "小数 的",
+      "小 数的",
+      "▁i ris",
+      "▁ir is",
+      "▁ iris",
+      "▁游戏 平台",
+      "司法鉴定 中心",
+      "un credited",
+      "东 湖",
+      "玄 武",
+      "这一 步",
+      "这 一步",
+      "投资或 控股",
+      "申请 执行的",
+      "申请执行 的",
+      "ncon fig",
+      "nco nfig",
+      "n config",
+      "▁borrow ed",
+      "▁trust worthy",
+      "精 油",
+      "▁Gl acier",
+      "▁first ly",
+      "回 转",
+      "稿 件",
+      "切除 术",
+      "ar yl",
+      "ary l",
+      "a ryl",
+      "▁b uds",
+      "▁bu ds",
+      "▁bud s",
+      "您 的最佳选择",
+      "送达了 执行通知书",
+      "▁inter iors",
+      "▁interior s",
+      "他 没有",
+      "又 包含了",
+      "nal so",
+      "nals o",
+      "n also",
+      "▁s ucks",
+      "▁su cks",
+      "▁suc ks",
+      "▁suck s",
+      "自然人 投资或控股",
+      "工 况",
+      "nL ow",
+      "nLo w",
+      "n Low",
+      "cl ang",
+      "c lang",
+      "▁p ops",
+      "▁pop s",
+      "▁po ps",
+      "▁b ricks",
+      "▁br icks",
+      "▁brick s",
+      "▁dil ution",
+      "V i",
+      "с ь",
+      "一 章",
+      "粮 油",
+      "限 时",
+      "全 明星",
+      "成功 地",
+      "社会 组织",
+      "Sk ill",
+      "S kill",
+      "lox acin",
+      "▁inev itably",
+      "▁serial ized",
+      "▁serialize d",
+      "▁ serialized",
+      "小 微",
+      "水电 站",
+      "水 电站",
+      "之间 存在",
+      "▁L ands",
+      "▁Land s",
+      "▁Lan ds",
+      "▁Ro oms",
+      "▁Room s",
+      "▁T uple",
+      "▁Tu ple",
+      "▁ Tuple",
+      "▁Jac obs",
+      "▁Jacob s",
+      "▁gall ons",
+      "▁gallon s",
+      "▁warrant ed",
+      "话 剧",
+      "看不 懂",
+      "看 不懂",
+      "nC ity",
+      "n City",
+      "科学 研究院",
+      "科学研究 院",
+      "▁Gib bs",
+      "know ledge",
+      "低 端",
+      "▁L R",
+      "▁ LR",
+      "江门 市",
+      "江 门市",
+      "一次 函数",
+      "▁R oma",
+      "▁Ro ma",
+      "▁Rom a",
+      "▁qu ir",
+      "▁qui r",
+      "▁ quir",
+      "▁liqu or",
+      "eth asone",
+      "ver ified",
+      "▁gad gets",
+      "▁transl ating",
+      "戴 着",
+      "细 微",
+      "no id",
+      "n oid",
+      "▁C OX",
+      "▁CO X",
+      "既 包含了",
+      "▁L ing",
+      "▁Lin g",
+      "▁Li ng",
+      "▁ed its",
+      "▁edit s",
+      "par ametric",
+      "param etric",
+      "para metric",
+      "paramet ric",
+      "失 落",
+      "点 头",
+      "准确 地",
+      "要 根据",
+      "发布 会上",
+      "发布会 上",
+      "anch ez",
+      "anche z",
+      "▁Py thag",
+      "▁ev apor",
+      "▁evap or",
+      "▁Col eman",
+      "▁Cole man",
+      "▁pos itivity",
+      "▁posit ivity",
+      "隐 隐",
+      "再 进行",
+      "政策 根据",
+      "ex tras",
+      "ext ras",
+      "extra s",
+      "extr as",
+      "▁re plicated",
+      "▁replic ated",
+      "▁replicate d",
+      "▁replica ted",
+      "不 还",
+      "保 质",
+      "辨 证",
+      "▁ 调料",
+      "ia st",
+      "ias t",
+      "i ast",
+      "▁L ua",
+      "▁Lu a",
+      "行政 管理",
+      "请仔细 浏览",
+      "并给出 正确答案",
+      "古 巴",
+      "缠 绕",
+      "位数 是",
+      "位 数是",
+      "县委 常委",
+      "县 委常委",
+      "在面对 以下",
+      "请您 认真分析",
+      "▁Dec imal",
+      "▁ Decimal",
+      "并告诉我 您的选择",
+      "在面对以下 选择题时",
+      "▁econom ist",
+      "请仔细浏览 以下备选答案",
+      "请您认真分析 并给出正确答案",
+      "假 冒",
+      "全 班",
+      "来 院",
+      "金黄 色",
+      "金 黄色",
+      "主要 种植",
+      "住 上海市",
+      "fl ask",
+      "Rich ard",
+      "R ichard",
+      "▁Bever ly",
+      "install ation",
+      "Not ifications",
+      "Notification s",
+      "此 生",
+      "翼 翼",
+      "Co V",
+      "成交 量",
+      "▁S ta",
+      "▁St a",
+      "如何 使用",
+      "▁DO WN",
+      "▁ DOWN",
+      "▁T ested",
+      "▁Te sted",
+      "▁Test ed",
+      "▁Tes ted",
+      "Draw able",
+      "ovolta ic",
+      "渣 男",
+      "英 勇",
+      "物质 的",
+      "物 质的",
+      "Aw are",
+      "A ware",
+      "▁d uplex",
+      "▁du plex",
+      "▁ duplex",
+      "▁fire arms",
+      "▁firearm s",
+      "▁Land scape",
+      "▁disappoint ment",
+      "他 却",
+      "汲 取",
+      "bo b",
+      "b ob",
+      "▁N ec",
+      "▁Ne c",
+      "后 可",
+      "奢侈 品",
+      "▁e ps",
+      "▁ep s",
+      "▁ eps",
+      "不断 创新",
+      "▁Bl ank",
+      "▁ Blank",
+      "▁sm arter",
+      "▁smart er",
+      "劳务 费",
+      "▁B org",
+      "▁Bo rg",
+      "▁Bor g",
+      "▁Un iv",
+      "▁Uni v",
+      "per ing",
+      "pe ring",
+      "p ering",
+      "① ②",
+      "区 位",
+      "纯 洁",
+      "诗 集",
+      "AC A",
+      "A CA",
+      "wo rm",
+      "wor m",
+      "w orm",
+      "▁P Cs",
+      "▁PC s",
+      "▁ PCs",
+      "可以 裁定",
+      "▁c ysts",
+      "▁cy sts",
+      "▁cyst s",
+      "▁sp oil",
+      "▁spo il",
+      "点 亮",
+      "保持 了",
+      "Ij oi",
+      "ad an",
+      "ada n",
+      "a dan",
+      "ur ger",
+      "urg er",
+      "▁M are",
+      "▁Mar e",
+      "▁Ma re",
+      "法 民初字第",
+      "current ly",
+      "curr ently",
+      "▁custom izable",
+      "▁A UTO",
+      "▁AU TO",
+      "▁ AUTO",
+      "▁p lat",
+      "▁pl at",
+      "▁ plat",
+      "▁E VERY",
+      "▁EV ERY",
+      "房地产开发 有限责任公司",
+      "但债权的实现 取决于被执行人是否有",
+      "▁' _",
+      "▁ '_",
+      "就是 我",
+      "就 是我",
+      "并 表示",
+      "在校 学生",
+      "ot ron",
+      "otr on",
+      "o tron",
+      "ale igh",
+      "a leigh",
+      "▁con serve",
+      "▁cons erve",
+      "▁conserv e",
+      "▁over haul",
+      "parallel ed",
+      "paralle led",
+      "最 先",
+      "%) ;",
+      "% );",
+      "nF eb",
+      "nFe b",
+      "n Feb",
+      "使用 的是",
+      "使用的 是",
+      "使 用的是",
+      "▁m ega",
+      "▁me ga",
+      "▁meg a",
+      "▁ mega",
+      "nAnd rew",
+      "n Andrew",
+      "▁D reams",
+      "▁Dream s",
+      "あ る",
+      "盛 会",
+      "nK n",
+      "n Kn",
+      "不 方便",
+      "ob ox",
+      "obo x",
+      "o box",
+      "▁play ful",
+      "pre diction",
+      "pred iction",
+      "predict ion",
+      "▁duplic ates",
+      "▁duplicate s",
+      "之 声",
+      "京 城",
+      "插 图",
+      "Ac k",
+      "A ck",
+      "眼中 的",
+      "眼 中的",
+      "nA ust",
+      "nC amp",
+      "nCam p",
+      "n Camp",
+      "St able",
+      "S table",
+      "pro pag",
+      "prop ag",
+      "▁sh uffle",
+      "▁ shuffle",
+      "▁sm iling",
+      "▁res piration",
+      "▁respir ation",
+      "▁resp iration",
+      "▁un subscribe",
+      "上 进",
+      "很有 可能",
+      "很 有可能",
+      "我是 一个",
+      "我 是一个",
+      "isp ens",
+      "▁ful fil",
+      "号 牌",
+      "诗 文",
+      "_{ (",
+      "_ {(",
+      "▁L ig",
+      "▁Li g",
+      "企业 形象",
+      "执行 程序",
+      "▁s ued",
+      "▁su ed",
+      "▁sue d",
+      "生产建设 兵团",
+      "nG etting",
+      "nGet ting",
+      "n Getting",
+      "▁Hal ifax",
+      "中 进行",
+      "民事 责任",
+      "▁a uch",
+      "▁au ch",
+      "▁De mand",
+      "▁Dem and",
+      "▁Hash Set",
+      "▁ HashSet",
+      "包 厢",
+      "方式 和",
+      "看不 起",
+      "看 不起",
+      "▁Ne ut",
+      "▁bl ur",
+      "▁blu r",
+      "▁K ernel",
+      "▁Ker nel",
+      "▁ Kernel",
+      "▁ 要",
+      "手 足",
+      "跳 过",
+      "进 场",
+      "wh it",
+      "w hit",
+      "in tel",
+      "int el",
+      "著作权 集体",
+      "▁Ferr ari",
+      "▁there of",
+      "▁per inatal",
+      "▁peri natal",
+      "五 六",
+      "大 象",
+      "选 址",
+      "▁o ps",
+      "▁op s",
+      "▁ ops",
+      "nS imple",
+      "nSim ple",
+      "nSimpl e",
+      "n Simple",
+      "▁Min ecraft",
+      "▁Mine craft",
+      "< >",
+      "族 群",
+      "ym n",
+      "y mn",
+      "为 重点",
+      "动画 片",
+      "安装 了",
+      "现代 农业",
+      "▁P itch",
+      "▁Pit ch",
+      "▁pil gr",
+      "▁Ed mund",
+      "▁compre hend",
+      "可 由",
+      "小 哥",
+      "废 气",
+      "想 吃",
+      "扶 养",
+      "▁K um",
+      "▁Ku m",
+      "draw able",
+      "can onical",
+      "▁inc ision",
+      "科学记数法的表示 形式为",
+      "与 李",
+      "在 各",
+      "担 负",
+      "为主 料的",
+      "上面 这个问题",
+      "财 运",
+      "Jo e",
+      "J oe",
+      "g kin",
+      "若干 选项",
+      "主要 负责人",
+      "的其他 财产",
+      "▁polymer ization",
+      "宋 朝",
+      "巡 回",
+      "捆 绑",
+      "邮 局",
+      "os m",
+      "o sm",
+      "坚定 的",
+      "坚 定的",
+      "▁L AN",
+      "▁LA N",
+      "▁ LAN",
+      "开户 银行",
+      "职业 道德",
+      "he sus",
+      "hes us",
+      "h esus",
+      "pt ical",
+      "ptic al",
+      "p tical",
+      "sh ader",
+      "sha der",
+      "ter ase",
+      "te rase",
+      "tera se",
+      "t erase",
+      "▁A eros",
+      "▁Aer os",
+      "att ices",
+      "attice s",
+      "atti ces",
+      "ki p",
+      "k ip",
+      "一直 都是",
+      "然后 点击",
+      "计算机 网络",
+      "这是一道 需",
+      "分析这些 选项",
+      "综合运用多 方面",
+      "▁Par king",
+      "▁Park ing",
+      "▁pl otted",
+      "▁plot ted",
+      "▁br ighter",
+      "▁bright er",
+      "▁dipl omat",
+      "▁diplom at",
+      "▁diploma t",
+      "请仔细 分析这些选项",
+      "然后告诉我 您的最佳选择",
+      "这是一道需 综合运用多方面",
+      "th i",
+      "t hi",
+      "保持 在",
+      "SA TA",
+      "SAT A",
+      "S ATA",
+      "Da ily",
+      "D aily",
+      "内容 判断出",
+      "包含一个 问题和",
+      "内容判断出 您认为",
+      "以下内容 包含一个问题和",
+      "▁Pseud omonas",
+      "依据问题 内容判断出您认为",
+      "以下内容包含一个问题和 若干选项",
+      "前 夕",
+      "感 性",
+      "这 儿",
+      "El l",
+      "E ll",
+      "法律 的",
+      "魔法 师",
+      "魔 法师",
+      "or ax",
+      "ora x",
+      "o rax",
+      "▁A FL",
+      "▁AF L",
+      "▁A gu",
+      "▁Ag u",
+      "▁V oy",
+      "▁Vo y",
+      "fil ls",
+      "fill s",
+      "f ills",
+      "▁M aint",
+      "▁Main t",
+      "▁Ma int",
+      "▁Mai nt",
+      "short cuts",
+      "▁W ik",
+      "▁Wi k",
+      "UM ENT",
+      "U MENT",
+      "in itis",
+      "init is",
+      "ini tis",
+      "le ctic",
+      "lect ic",
+      "lec tic",
+      "▁Alex a",
+      "▁Ale xa",
+      "▁em pathy",
+      "▁emp athy",
+      "▁Est onian",
+      "▁Eston ian",
+      "▁Estonia n",
+      ", {",
+      "▁A ML",
+      "▁AM L",
+      "▁ AML",
+      "周边 配套",
+      "▁D ash",
+      "▁Da sh",
+      "▁Das h",
+      "▁ Dash",
+      "turn ed",
+      "tur ned",
+      "▁f b",
+      "▁ fb",
+      "erial izer",
+      "erialize r",
+      "▁re create",
+      "▁rec reate",
+      "毛 细",
+      "短 片",
+      "cs c",
+      "c sc",
+      "▁H utch",
+      "▁Chap man",
+      "▁None theless",
+      "▁Non etheless",
+      "县 的",
+      "拳 头",
+      "JD K",
+      "J DK",
+      "交流 与",
+      "这个 地图",
+      "▁H elm",
+      "▁He lm",
+      "▁Hel m",
+      "sp here",
+      "s phere",
+      "▁pres umed",
+      "▁presum ed",
+      "依 依",
+      "沙 龙",
+      "营 商",
+      "仍不 能",
+      "仍 不能",
+      "所用 的",
+      "所 用的",
+      "TE MP",
+      "TEM P",
+      "T EMP",
+      "▁AB OUT",
+      "▁ABO UT",
+      "ut ively",
+      "utive ly",
+      "uti vely",
+      "▁Bol lywood",
+      "▁Sett lement",
+      "▁document ing",
+      "▁fashion able",
+      "二 层",
+      "疏 散",
+      "的 还是",
+      "nA mb",
+      "nAm b",
+      "n Amb",
+      "oblast ic",
+      "obl astic",
+      "▁vacc inated",
+      "▁vaccin ated",
+      "烤 肉",
+      "热 衷",
+      "为他 们",
+      "为 他们",
+      "审判 庭",
+      "与被 上诉人",
+      "与 被上诉人",
+      "access or",
+      "▁Stand ing",
+      "▁Stan ding",
+      "▁compass ionate",
+      "▁compassion ate",
+      "机动车交通事故责任 纠纷一案中",
+      "机动车交通事故责任纠纷一案 中",
+      "无 言",
+      "猕 猴",
+      "着 作",
+      "得出 答案",
+      "龙头 企业",
+      "is dir",
+      "on ium",
+      "oni um",
+      "给定 的问题",
+      "给定的 问题",
+      "▁sup erc",
+      "▁super c",
+      "▁non atomic",
+      "▁ nonatomic",
+      "毛 衣",
+      "的 快",
+      "皮 下",
+      "IL Y",
+      "I LY",
+      "▁Qu ad",
+      "▁C ookie",
+      "▁Cook ie",
+      "▁ Cookie",
+      "▁climb ed",
+      "▁clim bed",
+      "▁cli mbed",
+      "领 会",
+      "tc tx",
+      "t ctx",
+      "nj obs",
+      "n jobs",
+      "eh icles",
+      "ehicle s",
+      "lar gest",
+      "large st",
+      "larg est",
+      "▁p aints",
+      "▁pain ts",
+      "▁paint s",
+      "▁pa ints",
+      "| _",
+      "超 额",
+      "在 国外",
+      "给定 问题",
+      "▁H undred",
+      "▁json ify",
+      "▁Ap artment",
+      "▁Apart ment",
+      "D y",
+      "东 市",
+      "孔 雀",
+      "无 毒",
+      "Pa g",
+      "P ag",
+      "ot to",
+      "ott o",
+      "o tto",
+      "这个 行业",
+      "sequ ent",
+      "seq uent",
+      "L y",
+      "嘴 唇",
+      "GR E",
+      "G RE",
+      "付出 的",
+      "付 出的",
+      "▁P lat",
+      "▁Pl at",
+      "现在 我会给你",
+      "依据 给定的问题",
+      "Api Model",
+      "仔细阅读 上面这个问题",
+      "现在我会给你 几个选项",
+      "请你根据 问题进行分析",
+      "从选项中选出 合适的答案",
+      "揭 晓",
+      "新 书",
+      "比 方",
+      "▁C omm",
+      "▁Com m",
+      "▁Co mm",
+      "▁ Comm",
+      "上面 这段话",
+      "▁the ology",
+      "Collect ors",
+      "Coll ectors",
+      "Collector s",
+      "nSimilar ly",
+      "attach ments",
+      "attachment s",
+      "▁cred ential",
+      "▁ φ",
+      "双 面",
+      "eu s",
+      "e us",
+      "ass uming",
+      "att erson",
+      "atter son",
+      "'} \\",
+      "' }\\",
+      "em ap",
+      "ema p",
+      "e map",
+      "mon ds",
+      "mond s",
+      "m onds",
+      "▁o bey",
+      "▁ob ey",
+      "DE VICE",
+      "DEV ICE",
+      "Ir eland",
+      "I reland",
+      "▁Fin ish",
+      "▁ Finish",
+      "▁PE OPLE",
+      "▁sal iva",
+      "▁Dragon s",
+      "▁Drag ons",
+      "已发生法律效力 的",
+      "▁ins ufficiency",
+      "与 传统",
+      "▁catar act",
+      "▁Dist inguished",
+      "N b",
+      "▁ 星",
+      "深 海",
+      "使用 方法",
+      "Version s",
+      "Vers ions",
+      "▁imm obil",
+      "依法适用 普通程序",
+      "操 场",
+      "有 声",
+      "一分 钟",
+      "一 分钟",
+      "至少 要",
+      "是否 可以",
+      "in ium",
+      "ini um",
+      "▁p leasing",
+      "▁ple asing",
+      "▁pleas ing",
+      "▁plea sing",
+      "▁assist ants",
+      "▁assistant s",
+      "二 三",
+      "奋 力",
+      "j Query",
+      "▁p ests",
+      "▁pe sts",
+      "▁pest s",
+      "▁pes ts",
+      "\\\\\": \\\\\"",
+      "▁depict ing",
+      "▁ 来",
+      "近 些",
+      "逃 生",
+      "fe ng",
+      "fen g",
+      "f eng",
+      "▁c aud",
+      "▁ca ud",
+      "opath ology",
+      "▁Rec ording",
+      "▁Record ing",
+      "▁Ge ographic",
+      "▁Geo graphic",
+      "▁interfer ing",
+      "▁interf ering",
+      "刻 录",
+      "ka s",
+      "k as",
+      "nA pr",
+      "nAp r",
+      "n Apr",
+      "▁D isk",
+      "▁Dis k",
+      "▁Di sk",
+      "▁ Disk",
+      "ic ators",
+      "icator s",
+      "superv ised",
+      "▁River side",
+      "▁Rivers ide",
+      "建 工",
+      "显存 容量",
+      "rem oved",
+      "remove d",
+      "▁Eff icient",
+      "波 长",
+      "科 创",
+      "红 颜",
+      "贮 藏",
+      "M ajor",
+      "▁m aze",
+      "▁ma ze",
+      "▁prosec utor",
+      "[] {",
+      "[ ]{",
+      "▁m oi",
+      "▁mo i",
+      "期限届满 前",
+      "nView s",
+      "n Views",
+      "esar ean",
+      "esare an",
+      "▁De ploy",
+      "▁Dep loy",
+      "▁ Deploy",
+      "poss ibly",
+      "亦 是",
+      "低 迷",
+      "写 道",
+      "加 尔",
+      "筹 划",
+      "▁ 创办",
+      "al ist",
+      "ali st",
+      "alis t",
+      "a list",
+      "▁D onna",
+      "▁Don na",
+      "hand led",
+      "handle d",
+      "▁sm oked",
+      "▁smoke d",
+      "▁gran ules",
+      "▁granul es",
+      "▁integr in",
+      "oo oo",
+      "Re try",
+      "Ret ry",
+      "R etry",
+      "▁N ath",
+      "▁Na th",
+      "▁Nat h",
+      "申请 执行费",
+      "申请执行 费",
+      "出具借条 一份",
+      "▁rev ival",
+      "▁t c",
+      "▁ tc",
+      "ngo ing",
+      "n going",
+      "▁Res idential",
+      "小 憩",
+      "都 用",
+      "第一 季",
+      "ix ir",
+      "Sc ott",
+      "S cott",
+      "▁trans itional",
+      "▁transition al",
+      "▁transit ional",
+      "九 天",
+      "欺 诈",
+      "账 单",
+      "这样 就",
+      "ro at",
+      "olph in",
+      "web app",
+      "▁Complex ity",
+      "▁inher ently",
+      "▁inherent ly",
+      "▁carbohyd rates",
+      "▁carbohydrate s",
+      "k V",
+      "校 外",
+      "牙 膏",
+      "配 角",
+      "决策 和",
+      "ca rs",
+      "car s",
+      "c ars",
+      "从而 达到",
+      "▁new com",
+      "▁general ize",
+      "二 胎",
+      "对 战",
+      "数的 定义",
+      "数 的定义",
+      "社区 矫正",
+      "*\\\\ *\\\\",
+      "res ume",
+      "sh util",
+      "shu til",
+      "▁HT TPS",
+      "▁HTTP S",
+      "▁ HTTPS",
+      "▁bl ending",
+      "▁blend ing",
+      "we d",
+      "w ed",
+      "以 达到",
+      "Sl ug",
+      "▁ac upuncture",
+      "生产 能力",
+      "▁V all",
+      "▁Val l",
+      "▁Va ll",
+      "▁Cal vin",
+      "▁hon ours",
+      "▁honour s",
+      "▁refer ee",
+      "▁refere e",
+      "▁t actical",
+      "▁tact ical",
+      "丽 江",
+      "变得 更",
+      "变 得更",
+      "在 现场",
+      "老 干部",
+      "Rub y",
+      "Ru by",
+      "新冠 病毒",
+      "car go",
+      "c argo",
+      "▁Gi ovanni",
+      "▁in compatible",
+      "up p",
+      "u pp",
+      "加 起来",
+      "时间 段",
+      "高级 的",
+      "高 级的",
+      "uc ci",
+      "ucc i",
+      "u cci",
+      "▁R ex",
+      "▁Re x",
+      "▁Y uk",
+      "▁Yu k",
+      "执行 到位",
+      "▁qu as",
+      "\\' ]))\\",
+      "\\'] ))\\",
+      "\\']) )\\",
+      "▁extra pol",
+      "▁extr apol",
+      "eval uation",
+      "evalu ation",
+      "太 守",
+      "慷 慨",
+      "▁R eb",
+      "▁Re b",
+      "▁ Reb",
+      "or set",
+      "ors et",
+      "orse t",
+      "▁P ond",
+      "▁Po nd",
+      "▁Pon d",
+      "▁Tre aty",
+      "▁Treat y",
+      "▁p ushes",
+      "▁push es",
+      "mun ition",
+      "▁Fib onacci",
+      "新 风",
+      "航空 航天",
+      "▁p lex",
+      "▁pl ex",
+      "▁ple x",
+      "▁ plex",
+      "吃 得",
+      "成 套",
+      "泥 土",
+      "号 房屋",
+      "an no",
+      "ann o",
+      "a nno",
+      "im ab",
+      "ima b",
+      "有的 时候",
+      "有 的时候",
+      "正是 因为",
+      "正 是因为",
+      "nV ari",
+      "n Vari",
+      "▁M ing",
+      "▁Min g",
+      "▁Mi ng",
+      "orph ous",
+      "oxy genase",
+      "oxygen ase",
+      "▁sol vents",
+      "▁solvent s",
+      "▁solv ents",
+      "▁Autom otive",
+      "总 能",
+      "cu s",
+      "c us",
+      "Le af",
+      "▁E GF",
+      "▁EG F",
+      "▁C trl",
+      "▁ Ctrl",
+      "▁Fr anco",
+      "▁Franc o",
+      "▁Fra nco",
+      "Phys ical",
+      "▁s ickness",
+      "▁sick ness",
+      "▁volcan ic",
+      "▁hyd ration",
+      "▁zebra fish",
+      "减 压",
+      "橡 皮",
+      "认 清",
+      "贴 合",
+      "ux t",
+      "u xt",
+      "但是 这",
+      "▁] ];",
+      "什么 都不",
+      "什么都 不",
+      "民终 字第",
+      "民 终字第",
+      "刑 二",
+      "售 卖",
+      "洗 脑",
+      "问 了",
+      "tt y",
+      "t ty",
+      "有效 性",
+      "▁E UR",
+      "▁EU R",
+      "▁ EUR",
+      "▁D ensity",
+      "▁aspir in",
+      "斑 点",
+      "也能 够",
+      "也 能够",
+      "奥斯 卡",
+      "▁b abe",
+      "▁bab e",
+      "▁ba be",
+      "Imp act",
+      "block quote",
+      "(\" @",
+      "( \"@",
+      "zo o",
+      "z oo",
+      "观众 的",
+      "▁职务 任免",
+      "受案 登记表",
+      "在一定 程度上",
+      "在 一定程度上",
+      "▁Bar rett",
+      "▁Barr ett",
+      "▁por cine",
+      "▁clar ified",
+      "洞 察",
+      "肉 眼",
+      "闭 幕",
+      "哪怕 是",
+      "▁sy nov",
+      "▁syn ov",
+      "请你 根据题意",
+      "请你根据 题意",
+      "st ories",
+      "sto ries",
+      "stor ies",
+      "▁giant s",
+      "▁gi ants",
+      "▁Ext reme",
+      "▁Extr eme",
+      "▁imm ersion",
+      "▁immers ion",
+      "皆 是",
+      "▁T in",
+      "▁Ti n",
+      "▁bl amed",
+      "▁blame d",
+      "▁C reation",
+      "▁Creat ion",
+      "▁Cre ation",
+      "▁ Creation",
+      "▁in ferred",
+      "▁infer red",
+      "nPublic ation",
+      "层面 的",
+      "层 面的",
+      "ge ns",
+      "gen s",
+      "g ens",
+      "分类 讨论",
+      "nd oes",
+      "ndo es",
+      "n does",
+      "t Assert",
+      "▁K olkata",
+      "nF acebook",
+      "n Facebook",
+      "周 三",
+      "贪 污",
+      "▁T I",
+      "▁ TI",
+      "bi ke",
+      "b ike",
+      "▁war fare",
+      "不 休",
+      "为 大",
+      "奖 章",
+      "小 球",
+      "fr m",
+      "f rm",
+      "层次 的",
+      "层 次的",
+      "消失 了",
+      "设置 的",
+      "▁f oi",
+      "▁fo i",
+      "▁T ours",
+      "▁To urs",
+      "▁Tour s",
+      "▁Tou rs",
+      "又包含了 选项",
+      "既包含了 问题",
+      "Us uario",
+      "上面这段话 既包含了问题",
+      "请你根据题意 选出正确的答案",
+      "书 本",
+      "喝 茶",
+      "患 上",
+      "枕 头",
+      "IZ E",
+      "I ZE",
+      "的 危险",
+      "▁\\' [",
+      "nt het",
+      "nthe t",
+      "nth et",
+      "▁Tr ials",
+      "▁Tri als",
+      "▁Trial s",
+      "▁bure auc",
+      "▁Input Stream",
+      "▁ InputStream",
+      "▁phot ographed",
+      "▁photograph ed",
+      "е й",
+      "▁ 紫",
+      "西 市",
+      "贯 注",
+      "▁A H",
+      "▁ AH",
+      "中文 名",
+      "中 文名",
+      "▁E rd",
+      "▁Er d",
+      "gluc ose",
+      "此题主要 考查了",
+      "▁hand some",
+      "▁hands ome",
+      "▁consult ations",
+      "▁consultation s",
+      "▁instant aneous",
+      "倍数 的",
+      "倍 数的",
+      "临床 资料",
+      "职业 学院",
+      "ail le",
+      "a ille",
+      "▁P itt",
+      "▁Pi tt",
+      "▁Pit t",
+      "▁sl avery",
+      "▁slave ry",
+      "李 世",
+      "漫 漫",
+      "ey ond",
+      "e yond",
+      "▁c hel",
+      "▁ch el",
+      "▁che l",
+      "▁ chel",
+      "格 斯",
+      "分子 的",
+      "分 子的",
+      "zb ek",
+      "Dir ective",
+      "Direct ive",
+      "▁grand parents",
+      "nF ire",
+      "n Fire",
+      "Ch oose",
+      "Cho ose",
+      "ham mer",
+      "▁e clipse",
+      "▁ecl ipse",
+      "▁ eclipse",
+      "some times",
+      "s ometimes",
+      "未发现 可供执行的财产",
+      "▁Univers ities",
+      "不 长",
+      "你说 的",
+      "你 说的",
+      "办理 了",
+      "在各 种",
+      "在 各种",
+      "没 见过",
+      "统计 分析",
+      "App ro",
+      "Ap pro",
+      "od end",
+      "ode nd",
+      "oden d",
+      "红 袖添香网",
+      "▁doctor ate",
+      "▁ampl itudes",
+      "▁amplit udes",
+      "▁amplitude s",
+      "上 司",
+      "的 差",
+      "表 征",
+      "CM V",
+      "C MV",
+      "功能 和",
+      "nc om",
+      "nco m",
+      "n com",
+      "▁S ys",
+      "▁Sy s",
+      "▁ Sys",
+      "▁a pk",
+      "▁ap k",
+      "▁ apk",
+      "▁H ess",
+      "▁He ss",
+      "li mate",
+      "lim ate",
+      "l imate",
+      "闪 耀",
+      "ur f",
+      "u rf",
+      "一口 气",
+      "一 口气",
+      "行业 内",
+      "行 业内",
+      "th in",
+      "thi n",
+      "t hin",
+      "▁T ul",
+      "▁Tu l",
+      "全方位 的",
+      "全 方位的",
+      "▁sl aves",
+      "▁slave s",
+      "▁ple aded",
+      "▁plea ded",
+      "▁che ating",
+      "▁cheat ing",
+      "Ass ignment",
+      "Assign ment",
+      "重 返",
+      "▁中国 大陆",
+      "▁ 中国大陆",
+      "▁emb assy",
+      "▁upper case",
+      "之 父",
+      "倒 闭",
+      "寺 庙",
+      "放 的",
+      "手机 上",
+      "手 机上",
+      "▁/ ></",
+      "▁/> </",
+      "▁/>< /",
+      "▁B old",
+      "▁Bo ld",
+      "▁Bol d",
+      "▁ Bold",
+      "系 夫妻关系",
+      "▁av enues",
+      "▁out ward",
+      "▁s ibling",
+      "▁si bling",
+      "▁novel ist",
+      "▁nov elist",
+      "▁par abolic",
+      "▁para bolic",
+      "▁parab olic",
+      ": )",
+      "一 脚",
+      "何 如",
+      "去 向",
+      "▁ 河南",
+      "阜 阳市",
+      "st ores",
+      "store s",
+      "sto res",
+      "stor es",
+      "多 是",
+      "漏 水",
+      "药 业",
+      "dy n",
+      "d yn",
+      "在 公司",
+      "▁T rees",
+      "▁Tr ees",
+      "▁Tre es",
+      "▁Tree s",
+      "小额贷款 有限公司",
+      "▁hist amine",
+      "黄 帝",
+      "ev t",
+      "e vt",
+      "ig ms",
+      "igm s",
+      "符合 减刑",
+      "▁F lip",
+      "▁Fl ip",
+      "▁Canadian s",
+      "▁Canad ians",
+      "体 长",
+      "_( \"",
+      "_ (\"",
+      "经过 了",
+      "经 过了",
+      "▁To night",
+      "▁Ton ight",
+      "▁consect etur",
+      "专 线",
+      "天 龙",
+      "空 腹",
+      "靠谱 的",
+      "▁S AM",
+      "▁SA M",
+      "pos ites",
+      "po sites",
+      "posite s",
+      "▁re bound",
+      "▁reb ound",
+      "z l",
+      "助 学",
+      "实 操",
+      "导 电",
+      "\"] :\\",
+      "\" ]:\\",
+      "nT or",
+      "nTo r",
+      "n Tor",
+      "的最小 值为",
+      "的最小值 为",
+      "▁Bened ict",
+      "▁colour ful",
+      "大 词典",
+      "▁tr out",
+      "▁tro ut",
+      "▁trou t",
+      "nSc ient",
+      "▁P okemon",
+      "▁Pok emon",
+      "coll ector",
+      "collect or",
+      "▁Behavior al",
+      "▁inv itations",
+      "▁invitation s",
+      "丸 子",
+      "兼 具",
+      "如 是",
+      "时 就",
+      "请 教",
+      "np er",
+      "n per",
+      "▁B id",
+      "▁Bi d",
+      "▁acc r",
+      "▁ac cr",
+      "ip lier",
+      "ipl ier",
+      "i plier",
+      "▁Found er",
+      "▁Fo under",
+      "▁Ex posure",
+      "▁Exp osure",
+      "▁dead lines",
+      "▁deadline s",
+      "▁judge ment",
+      "▁judg ement",
+      "▁experi mentation",
+      "▁experiment ation",
+      "以 求",
+      "有 心",
+      "JP G",
+      "J PG",
+      "Ob s",
+      "O bs",
+      "办 实事",
+      "▁log out",
+      "▁logo ut",
+      "▁ logout",
+      "▁showc asing",
+      "关 机",
+      "带 队",
+      "nc ode",
+      "nco de",
+      "n code",
+      "▁sm iles",
+      "▁smile s",
+      "set Level",
+      "公 章",
+      "周 四",
+      "胡 子",
+      "}$ {",
+      "} ${",
+      "ni st",
+      "nis t",
+      "n ist",
+      "单位 出具",
+      "没有 办法",
+      "犯罪 情节",
+      "▁M ick",
+      "▁Mi ck",
+      "▁Mic k",
+      "oph ore",
+      "sc opes",
+      "scope s",
+      "▁f acet",
+      "▁fac et",
+      "▁face t",
+      "▁fa cet",
+      "▁Br oken",
+      "▁Bro ken",
+      "内 脏",
+      "格 力",
+      "站 起来",
+      "▁g cc",
+      "▁ gcc",
+      "ar cer",
+      "arc er",
+      "Match ers",
+      "Matcher s",
+      "舒 缓",
+      "Hour s",
+      "Ho urs",
+      "H ours",
+      "ben ef",
+      "▁f lock",
+      "▁fl ock",
+      "▁flo ck",
+      "作出 明智的选择",
+      "less ness",
+      "belongs To",
+      "▁decentral ized",
+      "▁fert ilization",
+      "▁fertil ization",
+      "哈 希",
+      "民 申",
+      "文化 旅游",
+      "▁No ah",
+      "人力资源 管理",
+      "▁inf amous",
+      "▁qu otation",
+      "▁quot ation",
+      "▁quota tion",
+      "▁Collabor ation",
+      "再 也",
+      "喝 了",
+      "座 的",
+      "胜 负",
+      "诸 侯",
+      "▁L IB",
+      "▁LI B",
+      "▁ LIB",
+      "ew ood",
+      "e wood",
+      "ur eka",
+      "ure ka",
+      "▁com missions",
+      "▁comm issions",
+      "▁commission s",
+      "Be ta",
+      "B eta",
+      "}` ,\\",
+      "}`, \\",
+      "} `,\\",
+      "▁b unk",
+      "▁bu nk",
+      "▁{ }\",",
+      "▁{} \",",
+      "▁{}\" ,",
+      "高新技术 企业",
+      "path name",
+      "▁C ourses",
+      "▁Co urses",
+      "▁Cour ses",
+      "▁Course s",
+      "此 项",
+      "王 妃",
+      "ast ric",
+      "astr ic",
+      "中医药学 名词审定委员会",
+      "也 更",
+      "听 的",
+      "条 纹",
+      "餐 具",
+      "▁ST ATE",
+      "▁STAT E",
+      "▁ STATE",
+      "▁He arts",
+      "▁Heart s",
+      "▁Hear ts",
+      "▁Abs olute",
+      "▁ Absolute",
+      "著作权集体 管理协会",
+      "凶 手",
+      "天 平",
+      "计 较",
+      "▁и з",
+      "▁Dis patch",
+      "▁ Dispatch",
+      "▁Th inking",
+      "▁Think ing",
+      "▁Thin king",
+      "▁understand able",
+      "▁D G",
+      "▁ DG",
+      "▁k Hz",
+      "▁ kHz",
+      "re cht",
+      "rec ht",
+      "▁f ats",
+      "▁fa ts",
+      "▁fat s",
+      "ri ates",
+      "ria tes",
+      "riate s",
+      "r iates",
+      "人力资源和社会保障 局",
+      "海 盗",
+      "▁东 方",
+      "▁ 东方",
+      "深入 开展",
+      "al coh",
+      "alc oh",
+      "收费是 每人",
+      "ab bage",
+      "abb age",
+      "▁t rape",
+      "▁tr ape",
+      "▁tra pe",
+      "▁trap e",
+      "att ered",
+      "atter ed",
+      "▁And reas",
+      "▁Andre as",
+      "▁Andrea s",
+      "▁Andr eas",
+      "其他 财产可供执行",
+      "hib ernate",
+      "ounc ement",
+      "ounce ment",
+      "▁mon onuclear",
+      "求 是",
+      "而 对",
+      "ar med",
+      "arm ed",
+      "nF ound",
+      "n Found",
+      "请您 根据您的",
+      "verte br",
+      "nl anguage",
+      "n language",
+      "▁ 淋浴",
+      "可以 刷卡",
+      "判断 和知识",
+      "浏览 备选答案",
+      "▁CON FIG",
+      "▁CONF IG",
+      "▁ CONFIG",
+      "▁Bald win",
+      "▁pub lishes",
+      "▁publish es",
+      "▁sur rogate",
+      "判断和知识 回答以下问题",
+      "State ToProps",
+      "authent icate",
+      "读 本",
+      "▁St yles",
+      "▁Style s",
+      "▁Styl es",
+      "▁ Styles",
+      "▁res umed",
+      "▁resume d",
+      "workflow s",
+      "▁myocard ium",
+      "抗 癌",
+      "热 处理",
+      "ac ea",
+      "ace a",
+      "a cea",
+      "转为 普通程序",
+      "▁D iversity",
+      "▁Spec ification",
+      "▁Specific ation",
+      "▁ Specification",
+      "子 集",
+      "开 元",
+      "梯 度",
+      "竞 价",
+      "AD I",
+      "A DI",
+      "大 教堂",
+      "资格 证",
+      "▁cyst eine",
+      "prot ection",
+      "prote ction",
+      "各 家",
+      "宣 称",
+      "深 思",
+      "赛 场",
+      "毕业 的",
+      "毕 业的",
+      "Tr aining",
+      "Tra ining",
+      "Train ing",
+      "格 的",
+      "R pc",
+      "rel ax",
+      "年平均 气温",
+      "经济 的发展",
+      "经济的 发展",
+      "Hash Code",
+      "ing ement",
+      "inge ment",
+      "▁certain ty",
+      "怀 里",
+      "摆 在",
+      "裂 纹",
+      "pr efer",
+      "pre fer",
+      "pref er",
+      "p refer",
+      "▁sk ies",
+      "▁ski es",
+      "▁Le vels",
+      "▁Level s",
+      "▁Lev els",
+      "提出 的撤诉申请",
+      "提出的 撤诉申请",
+      "▁C alling",
+      "▁Cal ling",
+      "▁Call ing",
+      "▁pl aques",
+      "▁plaque s",
+      "▁Met adata",
+      "▁Meta data",
+      "▁ Metadata",
+      "痊 愈",
+      "风险 管理",
+      "er ated",
+      "era ted",
+      "erate d",
+      "e rated",
+      "▁nod ules",
+      "▁M essages",
+      "▁Message s",
+      "▁Mess ages",
+      "▁ Messages",
+      "Date Format",
+      "文 联",
+      "潜 伏",
+      "选 了",
+      "Lo rem",
+      "L orem",
+      "em oji",
+      "emo ji",
+      "le ter",
+      "let er",
+      "lete r",
+      "l eter",
+      "my batis",
+      "ost atin",
+      "ostat in",
+      "osta tin",
+      "▁occ urrences",
+      "▁occurrence s",
+      "dynamic s",
+      "dyn amics",
+      "dynam ics",
+      "d ynamics",
+      "A k",
+      "拍 了",
+      "ig t",
+      "i gt",
+      "▁r x",
+      "▁ rx",
+      "共同 的",
+      "▁S omal",
+      "▁So mal",
+      "▁Som al",
+      "▁We ber",
+      "▁Web er",
+      "▁adm ire",
+      "▁admir e",
+      "▁astr onaut",
+      "▁remember ing",
+      "恒 温",
+      "qu ist",
+      "quis t",
+      "▁c ans",
+      "▁can s",
+      "▁ca ns",
+      "▁s age",
+      "▁sa ge",
+      "▁sag e",
+      "遥 感",
+      "em i",
+      "e mi",
+      "FA CE",
+      "F ACE",
+      "am mers",
+      "amm ers",
+      "▁rough ness",
+      "▁re placements",
+      "▁replac ements",
+      "▁replacement s",
+      "▁replace ments",
+      "藏 着",
+      "▁W ei",
+      "▁We i",
+      "▁H oll",
+      "▁Hol l",
+      "▁Ho ll",
+      "Can ada",
+      "op ause",
+      "opa use",
+      "o pause",
+      "▁Alger ia",
+      "音像 著作权集体管理协会",
+      "的 电",
+      "自 发",
+      "言 之",
+      "通 胀",
+      "DO T",
+      "D OT",
+      "nF ort",
+      "nFor t",
+      "n Fort",
+      "▁al kal",
+      "▁alk al",
+      "▁neutr ino",
+      "▁lever aging",
+      "他 想",
+      "叹 息",
+      "并 就",
+      "救 济",
+      "药 房",
+      ")( \"",
+      ") (\"",
+      "创建 的",
+      "创 建的",
+      "▁H ole",
+      "▁Hol e",
+      "▁Ho le",
+      "▁v oor",
+      "▁vo or",
+      "ont ology",
+      "onto logy",
+      "▁B ET",
+      "▁BE T",
+      "el man",
+      "elm an",
+      "▁wh olly",
+      "▁C urrency",
+      "▁ Currency",
+      "Calcul ator",
+      "Calc ulator",
+      "又 将",
+      "越 秀",
+      "软 化",
+      "名下 位于",
+      "基础设施 建设",
+      "add Widget",
+      "Migration s",
+      "M igrations",
+      "成 龙",
+      "水 体",
+      "王 家",
+      "正 处于",
+      "▁co ils",
+      "▁coil s",
+      "▁a while",
+      "▁staff ing",
+      "▁legisl ators",
+      "S r",
+      "只 看",
+      "常 有",
+      "棋 子",
+      "站 立",
+      "聊 聊",
+      "作品 中",
+      "条形 码",
+      "那么 在",
+      "深入 推进",
+      "▁o mit",
+      "▁om it",
+      "▁is let",
+      "nEvery thing",
+      "n Everything",
+      "亲 和",
+      "近 百",
+      "音乐 的",
+      "▁gen omics",
+      "▁genomic s",
+      "▁î n",
+      "曾 多次",
+      "▁J ong",
+      "▁Jo ng",
+      "▁Jon g",
+      "▁\" \");\\",
+      "▁\"\" );\\",
+      "▁\"\") ;\\",
+      "▁un int",
+      "▁u nint",
+      "▁uni nt",
+      "▁c aries",
+      "▁car ies",
+      "▁ca ries",
+      "▁nucle otides",
+      "▁nucleotide s",
+      "三 农",
+      "nB ob",
+      "nBo b",
+      "n Bob",
+      "nJ une",
+      "nJun e",
+      "n June",
+      "▁And ré",
+      "▁Andr é",
+      "▁g reeting",
+      "▁gre eting",
+      "▁greet ing",
+      "惆 怅",
+      "惩 戒",
+      "文 凭",
+      "皇 冠",
+      "济宁 市",
+      "济 宁市",
+      "▁d are",
+      "▁da re",
+      "▁dar e",
+      "▁d yes",
+      "▁dy es",
+      "▁dye s",
+      "▁n ets",
+      "▁ne ts",
+      "▁net s",
+      "▁ nets",
+      "▁sm ells",
+      "▁smell s",
+      "▁str ipe",
+      "▁stri pe",
+      "▁strip e",
+      "▁ stripe",
+      "出 击",
+      "微 分",
+      "高中 的",
+      "高 中的",
+      "ip el",
+      "ipe l",
+      "i pel",
+      "▁S lim",
+      "▁Sl im",
+      "na ming",
+      "nam ing",
+      "n aming",
+      "▁Lore ntz",
+      "▁Loren tz",
+      "▁N G",
+      "▁ NG",
+      "▁P s",
+      "▁ Ps",
+      "不 包括",
+      "公里 的",
+      "公 里的",
+      "刷 题",
+      "拥 挤",
+      "木 瓜",
+      "** （",
+      "工作 时",
+      "nC le",
+      "nCl e",
+      "as List",
+      "▁Lib ya",
+      "Sup plier",
+      "▁re count",
+      "▁rec ount",
+      "▁correspond ent",
+      "两 车",
+      "出 让",
+      "答 疑",
+      "装 载",
+      "豌 豆",
+      "遗 失",
+      "WO R",
+      "W OR",
+      "人数 的",
+      "人 数的",
+      "re mark",
+      "rem ark",
+      "▁her oin",
+      "▁hero in",
+      "▁e ternal",
+      "▁et ernal",
+      "▁workflow s",
+      "▁ workflows",
+      "Does NotExist",
+      "▁volunte ering",
+      "▁volunteer ing",
+      "▁ 先",
+      "记 账",
+      "贴 士",
+      "的 几个",
+      "ot in",
+      "oti n",
+      "o tin",
+      "资格 证书",
+      "资格证 书",
+      "ro red",
+      "ror ed",
+      "r ored",
+      "▁L ars",
+      "▁La rs",
+      "▁Lar s",
+      "▁a top",
+      "▁at op",
+      "Sp here",
+      "S phere",
+      "ull ary",
+      "ulla ry",
+      "二〇一六年 六",
+      "K D",
+      "全 神",
+      "天 府",
+      "造 影",
+      "(\" \",",
+      "(\"\" ,",
+      "( \"\",",
+      "▁M IL",
+      "▁MI L",
+      "刑事 判决书",
+      "▁S hack",
+      "▁Sh ack",
+      "▁Sha ck",
+      "CONT ENT",
+      "nStat istics",
+      "n Statistics",
+      "冷 凝",
+      "出 汗",
+      "教 的",
+      "UN IT",
+      "▁T aj",
+      "▁Ta j",
+      "那么 这个",
+      "dec lare",
+      "declar e",
+      "decl are",
+      "not hing",
+      "no thing",
+      "n othing",
+      "Web Socket",
+      "王 牌",
+      "编 者",
+      "平均 每",
+      "依据 是什么",
+      "imp act",
+      "▁< %",
+      "▁ <%",
+      "▁P GA",
+      "▁PG A",
+      "▁p ods",
+      "▁po ds",
+      "▁pod s",
+      "通过本院 或直接向",
+      "▁sl aughter",
+      "▁prot otypes",
+      "▁prototype s",
+      "▁proto types",
+      "她 和",
+      "悉 尼",
+      "清 扫",
+      "私 家",
+      "远 古",
+      "hi d",
+      "h id",
+      "on io",
+      "oni o",
+      "o nio",
+      "违法 所得",
+      "am ble",
+      "amb le",
+      "ー ル",
+      "化 身",
+      "岛 上",
+      "▁ 言情",
+      "这 东西",
+      "环境 卫生",
+      "公诉机关 认为",
+      "▁Ch anged",
+      "▁Change d",
+      "▁Chang ed",
+      "▁Chan ged",
+      "▁ Changed",
+      "▁Coast al",
+      "▁Machine s",
+      "▁Mach ines",
+      "dep recated",
+      "nM icrosoft",
+      "nMicro soft",
+      "n Microsoft",
+      "有一 位",
+      "有 一位",
+      "▁T BI",
+      "▁TB I",
+      "未经 法庭",
+      "二〇一六年 四月",
+      "二〇一六年四 月",
+      "▁jud ging",
+      "▁judg ing",
+      "▁carbon ate",
+      "▁carb onate",
+      "和 控制",
+      "Pro du",
+      "Prod u",
+      "P rodu",
+      "▁T uring",
+      "▁Tur ing",
+      "▁Tu ring",
+      "▁ т",
+      "的手 段",
+      "的 手段",
+      "▁p rag",
+      "▁pr ag",
+      "▁pra g",
+      "▁un le",
+      "nAc cept",
+      "n Accept",
+      "▁inter rog",
+      "AU TO",
+      "A UTO",
+      "nP rove",
+      "nPro ve",
+      "nPr ove",
+      "nProv e",
+      "▁a rose",
+      "▁ar ose",
+      "依据 已经发生法律效力的",
+      "互 换",
+      "野 心",
+      "▁Not ch",
+      "▁f east",
+      "▁fe ast",
+      "▁feas t",
+      "▁re play",
+      "▁rep lay",
+      "▁Im migration",
+      "▁Imm igration",
+      "交 接",
+      "必 修",
+      "直 言",
+      "看 懂",
+      "操作 简单",
+      "nC oll",
+      "nCol l",
+      "nCo ll",
+      "n Coll",
+      "▁G ael",
+      "▁Ga el",
+      "▁Qu iz",
+      "▁Kn ock",
+      "▁epit opes",
+      "▁epitope s",
+      "庆 幸",
+      "日 历",
+      "稳 产",
+      "送 餐",
+      "ze k",
+      "z ek",
+      "的 收入",
+      "实习 律师",
+      "\\' })\\",
+      "\\'} )\\",
+      "▁* */\\",
+      "▁** /\\",
+      "▁ **/\\",
+      "▁Pow der",
+      "plus plus",
+      "用科学记数法 表示为",
+      "▁cat hedral",
+      "▁Runtime Error",
+      "则 在",
+      "朗 诵",
+      "Do g",
+      "D og",
+      "命运 的",
+      "拼 多多",
+      "nw ww",
+      "n www",
+      "ri tt",
+      "rit t",
+      "r itt",
+      "▁f us",
+      "▁fu s",
+      "市委 常委",
+      "市 委常委",
+      "▁c ron",
+      "▁cr on",
+      "▁cro n",
+      "▁ cron",
+      "▁d ell",
+      "▁de ll",
+      "▁del l",
+      "rystal line",
+      "rystall ine",
+      "▁cart ridge",
+      "▁adapt ations",
+      "▁adaptation s",
+      "编 纂",
+      "赶 上",
+      "QS tring",
+      "Q String",
+      "本院 审理过程中",
+      "本院审理 过程中",
+      "▁cerebell ar",
+      "▁leuk ocytes",
+      "▁leukocyte s",
+      "完 好",
+      "新 老",
+      "诺 基亚",
+      "ar bon",
+      "arb on",
+      "ri ved",
+      "riv ed",
+      "rive d",
+      "r ived",
+      "为终审 判决",
+      "▁Euro pa",
+      "▁Eur opa",
+      "妇 幼",
+      "消 炎",
+      "等 边",
+      "ul g",
+      "u lg",
+      "自然 灾害",
+      "pan das",
+      "pand as",
+      "▁c rawl",
+      "▁cra wl",
+      "▁Lat via",
+      "▁st arters",
+      "▁start ers",
+      "▁star ters",
+      "▁starter s",
+      "多 达",
+      "旅 途",
+      "BA D",
+      "B AD",
+      "ll vm",
+      "FAIL ED",
+      "FAI LED",
+      "Index ed",
+      "▁maxim izing",
+      "▁multi plicity",
+      "客 家",
+      "神 木",
+      "选 的",
+      "郑 某",
+      "nP e",
+      "n Pe",
+      "▁W ii",
+      "▁Wi i",
+      "oot ers",
+      "oo ters",
+      "ooter s",
+      "o oters",
+      "▁bi ases",
+      "▁bias es",
+      "▁fr aming",
+      "▁fra ming",
+      "卡 斯",
+      "日 程",
+      "是 好",
+      "读 完",
+      "nC ap",
+      "n Cap",
+      "上市 时间",
+      "nT owns",
+      "nTown s",
+      "roll ment",
+      "▁particip ates",
+      "▁participate s",
+      "警 务",
+      "遗 漏",
+      "小时 后",
+      "最早 的",
+      "等量 关系",
+      "nm kdir",
+      "n mkdir",
+      "Term inal",
+      "istr ator",
+      "给你 的",
+      "给 你的",
+      "as tings",
+      "ast ings",
+      "asting s",
+      "▁s immer",
+      "▁sim mer",
+      "▁treat y",
+      "▁tre aty",
+      "time line",
+      "tim eline",
+      "我 被",
+      "藏 族",
+      "远程 仓库",
+      "he avy",
+      "hea vy",
+      "li pid",
+      "lip id",
+      "▁Max im",
+      "▁Ma xim",
+      "▁re cap",
+      "▁rec ap",
+      "▁default dict",
+      "详 尽",
+      "跨 界",
+      "雪 山",
+      "op edic",
+      "ope dic",
+      "oped ic",
+      "▁C hall",
+      "▁Ch all",
+      "▁Cha ll",
+      "▁Chal l",
+      "nJan uary",
+      "n January",
+      "▁Creat or",
+      "▁Cre ator",
+      "▁ Creator",
+      "以 待",
+      "养 分",
+      "ve nv",
+      "ven v",
+      "v env",
+      "▁R im",
+      "▁Ri m",
+      "▁M arian",
+      "▁Mar ian",
+      "▁Ma rian",
+      "▁Maria n",
+      "▁Mari an",
+      "▁bott len",
+      "▁bottle n",
+      "号民事调解书 已经发生法律效力",
+      "ó s",
+      "游 离",
+      "ri ke",
+      "rik e",
+      "r ike",
+      "ad emy",
+      "ade my",
+      "adem y",
+      "▁DE LETE",
+      "▁ DELETE",
+      "▁inaccur ate",
+      "正 品",
+      "ny t",
+      "n yt",
+      "会 提供",
+      "安全 和",
+      "加快 推进",
+      "进行 调解",
+      "▁A SAP",
+      "▁AS AP",
+      "▁eleg ans",
+      "▁wear able",
+      "债权依法 受法律保护",
+      "号民事调解书 的本次执行程序",
+      "▁ 乔",
+      "乡村 人口",
+      "Qt Gui",
+      "▁Ch ic",
+      "▁Chi c",
+      "add ons",
+      "addon s",
+      "▁Al ter",
+      "▁Alt er",
+      "▁ Alter",
+      "▁sp rink",
+      "▁spr ink",
+      "▁card board",
+      "▁ident ifiable",
+      "▁ 邵",
+      "的 行业",
+      "教学 工作",
+      "▁NE WS",
+      "▁NEW S",
+      "▁m ating",
+      "▁mat ing",
+      "▁ma ting",
+      "assert Raises",
+      "CH A",
+      "C HA",
+      "PL A",
+      "P LA",
+      "ok s",
+      "o ks",
+      "技术 的发展",
+      "技术的 发展",
+      "▁S ense",
+      "▁Se nse",
+      "▁Sen se",
+      "▁Sens e",
+      "▁ Sense",
+      "Del ivery",
+      "▁toler ate",
+      "Object Name",
+      "▁dys plasia",
+      "以 保证",
+      "认证 的",
+      "充分 肯定",
+      "特色 产业",
+      "\\' \\',",
+      "\\'\\ ',",
+      "\\'\\' ,",
+      "io ids",
+      "ioid s",
+      "i oids",
+      "▁so uls",
+      "▁sou ls",
+      "▁soul s",
+      "cont inence",
+      "contin ence",
+      "光 阴",
+      "接触 到",
+      "管理 器",
+      "nm ay",
+      "n may",
+      "▁FP GA",
+      "▁ FPGA",
+      "▁l unar",
+      "▁lu nar",
+      "是在 法律允许",
+      "▁Chile an",
+      "▁Chi lean",
+      "sim ulation",
+      "以 北",
+      "底 的",
+      "涉 外",
+      "高 雅",
+      "HP V",
+      "H PV",
+      "GE NER",
+      "GEN ER",
+      "G ENER",
+      "▁sus pend",
+      "▁susp end",
+      "re stricted",
+      "▁l u",
+      "▁ lu",
+      "大家 对",
+      "nT rad",
+      "nTr ad",
+      "▁A viv",
+      "▁Av iv",
+      "▁ach ievable",
+      "er b",
+      "e rb",
+      "的 特性",
+      "的 经典",
+      "es tro",
+      "est ro",
+      "estr o",
+      "▁a roma",
+      "▁ar oma",
+      "▁arom a",
+      "▁aff iliation",
+      "▁affil iation",
+      "开 具",
+      "赔 付",
+      "阵 阵",
+      "▁o rn",
+      "▁or n",
+      "▁ orn",
+      "被执行人 无",
+      "▁Ven us",
+      "▁Ve nus",
+      "▁l odge",
+      "▁lod ge",
+      "现 申请执行人",
+      "▁Check s",
+      "▁Che cks",
+      "▁ Checks",
+      "▁repair ing",
+      "城 东",
+      "城镇 化",
+      "是无 法",
+      "是 无法",
+      "高 强度",
+      "数据 处理",
+      "B ritish",
+      "um etric",
+      "u metric",
+      "dispatch er",
+      "disp atcher",
+      "▁ 欧",
+      "说 一句",
+      "▁P IN",
+      "▁PI N",
+      "▁ PIN",
+      "▁Pharm acy",
+      "▁Pharmac y",
+      "中 含",
+      "繁 忙",
+      "有 特殊",
+      "耐 腐蚀",
+      "本院 查明",
+      "ot ation",
+      "ota tion",
+      "▁sub way",
+      "▁B iography",
+      "▁Bi ography",
+      "春 夏",
+      "体育 场",
+      "▁b apt",
+      "▁ba pt",
+      "ch ains",
+      "chain s",
+      "cha ins",
+      "chai ns",
+      "▁Jack ie",
+      "▁Jac kie",
+      "二 本",
+      "树 种",
+      "止 损",
+      "因数 的",
+      "因 数的",
+      "近似 数",
+      "希望 能够",
+      "希望能 够",
+      "考试 成绩",
+      "Car bon",
+      "C arbon",
+      "med ical",
+      "m edical",
+      "▁F ormal",
+      "▁For mal",
+      "▁Form al",
+      "▁Tr igon",
+      "▁Tri gon",
+      "两 根",
+      "小 华",
+      "铁 皮",
+      "WI TH",
+      "W ITH",
+      "ra il",
+      "r ail",
+      "品牌 形象",
+      "▁im pe",
+      "▁imp e",
+      "许可 中途退",
+      "land er",
+      "la nder",
+      "lan der",
+      "l ander",
+      "▁Br ass",
+      "▁Bra ss",
+      "▁Bras s",
+      "il iated",
+      "ili ated",
+      "ilia ted",
+      "iliate d",
+      "nSchool s",
+      "nSch ools",
+      "▁Colon ial",
+      "▁explo its",
+      "▁exploit s",
+      "H Q",
+      "▁A j",
+      "▁安 徽",
+      "▁ 安徽",
+      "社区 居民",
+      "em ann",
+      "ema nn",
+      "eman n",
+      "e mann",
+      "▁empir ically",
+      "▁empirical ly",
+      "O l",
+      "何 为",
+      "剪 切",
+      "安 稳",
+      "RN G",
+      "R NG",
+      "▁O c",
+      "属 植物",
+      "高 标准",
+      "可以 上诉",
+      "▁rad iol",
+      "▁radio l",
+      "▁radi ol",
+      "▁ radiol",
+      "Inter rupt",
+      "inst agram",
+      "▁rob otics",
+      "▁robot ics",
+      "▁robotic s",
+      "▁separ able",
+      "x u",
+      "绝 世",
+      "出现 问题",
+      "▁H ire",
+      "▁Hi re",
+      "▁Hir e",
+      "▁c rab",
+      "▁cr ab",
+      "▁cra b",
+      "如 发现被执行人",
+      "▁s andbox",
+      "▁sand box",
+      "▁ sandbox",
+      "不 时",
+      "宫 殿",
+      "给 药",
+      "这 与",
+      "一直 未",
+      "宜昌 市",
+      "走出 去",
+      "走 出去",
+      "la ws",
+      "law s",
+      "l aws",
+      "Gr and",
+      "G rand",
+      "▁congr uent",
+      "▁congru ent",
+      "z c",
+      "nB o",
+      "n Bo",
+      "abol ism",
+      "同 龄",
+      "炫 耀",
+      "播放 器",
+      "base Url",
+      "Number Of",
+      "▁Prime ra",
+      "▁Prim era",
+      "序 幕",
+      "ns l",
+      "n sl",
+      "▁P riv",
+      "▁Pr iv",
+      "▁Pri v",
+      "一定 程度上",
+      "一定程度 上",
+      "um ping",
+      "ump ing",
+      "县人民法院 作出的",
+      "▁meth anol",
+      "廉 价",
+      "此 书",
+      "ni mg",
+      "nim g",
+      "n img",
+      "进行 计算即可",
+      "进行计算 即可",
+      "它 会",
+      "室 温",
+      "杯 子",
+      "会展 中心",
+      "▁Mar itime",
+      "▁Mari time",
+      "▁news letters",
+      "▁newsletter s",
+      "依法 应",
+      "U uid",
+      "▁ted ious",
+      "▁ 六",
+      "三 方",
+      "借 到",
+      "国 画",
+      "小 狗",
+      "机 等",
+      "白 宫",
+      "预 留",
+      "ur m",
+      "u rm",
+      "人物 的",
+      "人 物的",
+      "结婚 了",
+      "设计 中",
+      "公共 场所",
+      "▁P OSS",
+      "▁PO SS",
+      "▁POS S",
+      "▁Ut ils",
+      "▁Util s",
+      "▁ Utils",
+      "▁ra iny",
+      "▁rain y",
+      "Con cept",
+      "▁c apped",
+      "▁cap ped",
+      "▁ca pped",
+      "▁por celain",
+      "Io T",
+      "I oT",
+      "九龙 坡",
+      "编辑 器",
+      "CL AI",
+      "CLA I",
+      "详细 内容",
+      "▁w izard",
+      "▁ wizard",
+      "sub mission",
+      "▁Advert ising",
+      "个 案",
+      "放 手",
+      "▁re location",
+      "▁rel ocation",
+      "性 情",
+      "金 星",
+      "治疗 前",
+      "np ath",
+      "n path",
+      "▁ з",
+      "五 彩",
+      "抬 起",
+      "腰 部",
+      "Le e",
+      "L ee",
+      "]] ;\\",
+      "] ];\\",
+      "id an",
+      "ida n",
+      "i dan",
+      "车牌 号为",
+      "PY THON",
+      "To Array",
+      "▁P ocket",
+      "▁wave form",
+      "类 比",
+      "繁 多",
+      "不能 用",
+      "不 能用",
+      "'} ,\\",
+      "' },\\",
+      "il ee",
+      "ile e",
+      "i lee",
+      "一步 一步",
+      "▁ext rac",
+      "▁extra c",
+      "▁extr ac",
+      "▁craft ing",
+      "▁phosphat idyl",
+      "▁ 干",
+      "编写 的",
+      "编 写的",
+      "ic eps",
+      "ice ps",
+      "big cup",
+      "nS hips",
+      "nSh ips",
+      "▁desper ately",
+      "▁desperate ly",
+      "倍 的",
+      "太 低",
+      "说 她",
+      "▁p lea",
+      "▁pl ea",
+      "▁ple a",
+      "▁T ales",
+      "▁Ta les",
+      "▁Tal es",
+      "▁Tale s",
+      "▁broad casts",
+      "▁broadcast s",
+      "> ×",
+      "王 府",
+      "双床 间",
+      "来 解决",
+      "要做 好",
+      "要 做好",
+      "nF ac",
+      "n Fac",
+      "▁P TH",
+      "▁PT H",
+      "▁s paced",
+      "▁sp aced",
+      "▁space d",
+      "▁spac ed",
+      "▁spa ced",
+      "▁th rone",
+      "▁thr one",
+      "▁thro ne",
+      "▁John ston",
+      "▁Johns ton",
+      "▁F OX",
+      "▁FO X",
+      "没有 那么",
+      "没 有那么",
+      "研究 领域",
+      "▁tip o",
+      "▁ti po",
+      "▁ tipo",
+      "ital ic",
+      "ita lic",
+      "ex pires",
+      "exp ires",
+      "▁yog urt",
+      "Sub scribe",
+      "▁do ctrine",
+      "▁ М",
+      "为 先",
+      "地 把",
+      "MI S",
+      "M IS",
+      "服务 质量",
+      "综合 考虑",
+      "pr ior",
+      "pri or",
+      "p rior",
+      "▁M oses",
+      "▁Mo ses",
+      "▁Mos es",
+      "▁c locks",
+      "▁cl ocks",
+      "▁clock s",
+      "▁s lated",
+      "▁sl ated",
+      "▁slate d",
+      "Prim itive",
+      "▁foot wear",
+      "▁mult ipro",
+      "▁multi pro",
+      "▁multip ro",
+      "▁mascul ine",
+      "布 莱",
+      "闲 置",
+      "消费 品",
+      "等 形式",
+      "az ol",
+      "azo l",
+      "▁P NG",
+      "▁PN G",
+      "▁ PNG",
+      "▁B read",
+      "▁Br ead",
+      "▁Bre ad",
+      "经人介绍 相识",
+      "违 章",
+      "逆 天",
+      "▁ 教师",
+      "对 这些",
+      "汕头 市",
+      "汕 头市",
+      "ank ton",
+      "▁sh util",
+      "▁shut il",
+      "▁ shutil",
+      "▁disc ern",
+      "▁Concept s",
+      "利 斯",
+      "即 成",
+      "奇 异",
+      "所 见",
+      "置 身",
+      "▁m oms",
+      "▁mom s",
+      "▁mo ms",
+      "nI rish",
+      "om ized",
+      "omi zed",
+      "▁ch arger",
+      "▁char ger",
+      "▁charg er",
+      "▁charge r",
+      "prob ability",
+      "▁distingu ishing",
+      "▁distinguish ing",
+      "有 待",
+      "ji e",
+      "j ie",
+      "参观 了",
+      "情人 节",
+      "▁G SM",
+      "▁GS M",
+      "▁ GSM",
+      "▁New man",
+      "▁Appe als",
+      "▁Appeal s",
+      "某某 到庭参加诉讼",
+      "▁ord ained",
+      "▁Arch itects",
+      "▁Architect s",
+      "▁Struct ures",
+      "▁Structure s",
+      "UG H",
+      "U GH",
+      "Op acity",
+      "▁Soph ie",
+      "▁park ed",
+      "▁Up grade",
+      "▁ Upgrade",
+      "▁Port folio",
+      "▁insert ing",
+      "攀 升",
+      "iO iJ",
+      "疫情防控 工作",
+      "Cl icked",
+      "Click ed",
+      "本判决 为终审判决",
+      "De ployment",
+      "Deploy ment",
+      "内 向",
+      "进 门",
+      "GA N",
+      "G AN",
+      "Ph p",
+      "P hp",
+      "传说 中的",
+      "▁w ifi",
+      "▁wi fi",
+      "▁ wifi",
+      "in arily",
+      "inar ily",
+      "▁Esc ort",
+      "▁Christ ie",
+      "▁Chris tie",
+      "_) \\",
+      "_ )\\",
+      "能 达到",
+      "▁D up",
+      "▁Du p",
+      "as ome",
+      "a some",
+      "▁b urger",
+      "▁bur ger",
+      "▁burg er",
+      "INST ANCE",
+      "厉害 的",
+      "nW ood",
+      "n Wood",
+      "▁C razy",
+      "▁Cr azy",
+      "▁Cra zy",
+      "▁Com mod",
+      "▁Comm od",
+      "commit tee",
+      "▁Con version",
+      "▁Co nversion",
+      "▁Conv ersion",
+      "▁ Conversion",
+      "ч а",
+      "华 尔",
+      "真 好",
+      "饥 饿",
+      "VE GF",
+      "ib ia",
+      "ibi a",
+      "i bia",
+      "▁cor ridor",
+      "Inject able",
+      "▁un paralleled",
+      "中 卫",
+      "譬 如",
+      "目的 观察",
+      "▁K err",
+      "▁Ke rr",
+      "▁Ker r",
+      "▁Ph osph",
+      "▁expect ancy",
+      "对 焦",
+      "有 各",
+      "甘 蔗",
+      "从来 不",
+      "自己 做",
+      "nK ing",
+      "n King",
+      "po und",
+      "p ound",
+      "ri ding",
+      "rid ing",
+      "r iding",
+      "▁expl ode",
+      "▁explo de",
+      "U F",
+      "凝 结",
+      "力 强",
+      "KT V",
+      "K TV",
+      "想象 的",
+      "没有 再",
+      "al ez",
+      "ale z",
+      "a lez",
+      "You ng",
+      "做题 的思路",
+      "▁Ar lington",
+      "吸 尘",
+      "ng r",
+      "n gr",
+      "反 过来",
+      "感觉 很",
+      "ère s",
+      "è res",
+      "▁de cks",
+      "▁dec ks",
+      "▁deck s",
+      "▁de vote",
+      "▁dev ote",
+      "▁Custom s",
+      "沙 坪",
+      "mi xin",
+      "mix in",
+      "m ixin",
+      "▁w oke",
+      "▁wo ke",
+      "分析和 依据",
+      "▁Ha iti",
+      "▁Hait i",
+      "通 气",
+      "EN U",
+      "E NU",
+      "▁h d",
+      "▁ hd",
+      "主干 道",
+      "他们 都",
+      "他 们都",
+      "关心 的",
+      "关 心的",
+      "创作 者",
+      "创 作者",
+      "nD er",
+      "nDe r",
+      "n Der",
+      "质量 控制",
+      "lo ver",
+      "lov er",
+      "love r",
+      "l over",
+      "▁res usc",
+      "并给出你的 解题",
+      "an aconda",
+      "ana conda",
+      "▁Land ing",
+      "▁Lan ding",
+      "从选项中选择出 合适的答案",
+      "并给出你的解题 分析和依据",
+      "拖 鞋",
+      "超 人",
+      ",, ,",
+      ", ,,",
+      "st or",
+      "sto r",
+      "s tor",
+      "到底 是什么",
+      "到底是 什么",
+      "▁na ughty",
+      "红 球",
+      "街 区",
+      "ho e",
+      "h oe",
+      "st y",
+      "s ty",
+      "优异 的",
+      "优 异的",
+      "宅 基地",
+      "▁fl own",
+      "▁flo wn",
+      "▁flow n",
+      "▁t reas",
+      "▁tr eas",
+      "▁tre as",
+      "s urgical",
+      "帮 他",
+      "战 机",
+      "无 偿",
+      "求 精",
+      "着 她",
+      "▁Con cer",
+      "▁Conc er",
+      "▁Fal con",
+      "▁show cases",
+      "▁showcase s",
+      "▁showc ases",
+      "以 赴",
+      "Min d",
+      "Mi nd",
+      "M ind",
+      "Bu ilt",
+      "B uilt",
+      "并做出 解释",
+      "▁sne ak",
+      "极 少",
+      "LP L",
+      "L PL",
+      "性能 和",
+      "nD ou",
+      "nDo u",
+      "缓刑 一年",
+      "▁F ors",
+      "▁For s",
+      "▁Fo rs",
+      "▁re hab",
+      "book ing",
+      "bo oking",
+      "b ooking",
+      "选出正确的答案 并做出解释",
+      "告 白",
+      "重 构",
+      "ID A",
+      "I DA",
+      "会 选择",
+      "的 直线",
+      "▁U SS",
+      "▁US S",
+      "数的 绝对值",
+      "数 的绝对值",
+      "ATP ase",
+      "上诉于 江苏省",
+      "▁cred ible",
+      "▁ credible",
+      "华 语",
+      "沉 重",
+      "ch y",
+      "c hy",
+      "马鞍 山市",
+      "pl ings",
+      "pling s",
+      "p lings",
+      "Tr acking",
+      "Track ing",
+      "▁Pract ices",
+      "▁Practice s",
+      "py l",
+      "p yl",
+      "of ib",
+      "o fib",
+      "Tr avel",
+      "Tra vel",
+      "T ravel",
+      "▁Be hind",
+      "位 和",
+      "议 论",
+      "on et",
+      "one t",
+      "o net",
+      "nF ish",
+      "申请执行人 亦未",
+      "申请执行人亦 未",
+      "Get Mapping",
+      "h q",
+      "▁ 韦",
+      "久 久",
+      "你 和",
+      "全 等",
+      "勤 劳",
+      "怪 的",
+      "点 心",
+      "才 开始",
+      "现场 照片",
+      "▁Di rac",
+      "▁Dir ac",
+      "▁b abel",
+      "▁bab el",
+      "▁ba bel",
+      "▁babe l",
+      "▁ babel",
+      "▁W izard",
+      "▁ Wizard",
+      "ku be",
+      "kub e",
+      "k ube",
+      "▁M OS",
+      "▁MO S",
+      "▁ MOS",
+      "融资 租赁",
+      "通俗 易懂",
+      "au ght",
+      "augh t",
+      "aug ht",
+      "a ught",
+      "Or Default",
+      "▁Raj asthan",
+      "▁propag anda",
+      "万 字",
+      "普 惠",
+      "这 在",
+      "自诉 人",
+      "▁pl eth",
+      "▁ple th",
+      "坚 硬",
+      "家 政",
+      "缺 氧",
+      "先天 性",
+      "否 命题",
+      "Form ula",
+      "▁rec ursion",
+      "▁recurs ion",
+      "▁territ orial",
+      "▁Revolution ary",
+      "动 人",
+      "悲 哀",
+      "武 士",
+      "状 的",
+      "要 先",
+      "为己 任",
+      "有 网友",
+      "CL AS",
+      "CLA S",
+      "C LAS",
+      "ph ony",
+      "phon y",
+      "▁Al ps",
+      "▁Alp s",
+      "Mar shal",
+      "▁Lanka n",
+      "▁Lan kan",
+      "▁Lank an",
+      "ung arian",
+      "正确答案和 做题的思路",
+      "不 加",
+      "寄 托",
+      "▁武 汉",
+      "▁ 武汉",
+      "理论 研究",
+      "gr aded",
+      "grad ed",
+      "grade d",
+      "gra ded",
+      "▁World wide",
+      "年获 得",
+      "年 获得",
+      "▁h ap",
+      "▁ha p",
+      "co eff",
+      "LO AT",
+      "也有 一些",
+      "也 有一些",
+      "▁C ran",
+      "▁Cr an",
+      "▁Cra n",
+      "▁RE AL",
+      "间 谍",
+      "▁' ',",
+      "▁'' ,",
+      "▁ '',",
+      "▁ma ize",
+      "申请 撤回上诉",
+      "申请撤回 上诉",
+      "f ashion",
+      "rec ipient",
+      "recip ient",
+      "▁abnormal ity",
+      "▁abnorm ality",
+      "W r",
+      "Z A",
+      "大会 上",
+      "大 会上",
+      "面对 面",
+      "面 对面",
+      "▁S KF",
+      "▁SK F",
+      "ber ra",
+      "er men",
+      "erm en",
+      "二〇一五年 三月",
+      "▁sing leton",
+      "▁single ton",
+      "▁ singleton",
+      "▁short ening",
+      "何 事",
+      "听 众",
+      "建 明",
+      "倍数 是",
+      "倍 数是",
+      "性 问题",
+      "此时 的",
+      "此 时的",
+      "uc us",
+      "u cus",
+      "也 请给出你的",
+      "▁ch asing",
+      "▁cha sing",
+      "解题过程 和分析思路",
+      "也请给出你的 解题过程和分析思路",
+      "fin ance",
+      "f inance",
+      "nL ittle",
+      "n Little",
+      "▁Princ iples",
+      "▁Principle s",
+      "▁contrad ict",
+      "▁ 即",
+      "个 好",
+      "病 学",
+      "为 契机",
+      "为了 避免",
+      "▁R are",
+      "▁Ra re",
+      "▁s addle",
+      "▁ber ries",
+      "▁ berries",
+      "▁em ailed",
+      "▁email ed",
+      "Logger Factory",
+      "本 是",
+      "建立 的",
+      "Op ts",
+      "Opt s",
+      "O pts",
+      "▁Y AML",
+      "L H",
+      "动 静",
+      "学生 对",
+      "的地 位",
+      "的 地位",
+      "▁T ac",
+      "▁Ta c",
+      "dec oder",
+      "decode r",
+      "▁Mongo DB",
+      "ML Element",
+      "MLE lement",
+      "getElements By",
+      "忙 着",
+      "自 此",
+      "▁OR DER",
+      "▁ ORDER",
+      "条第一款 规定",
+      "未经法庭 许可中途退",
+      "▁poly ester",
+      "一 侧",
+      "建 房",
+      "浓度 为",
+      "浓 度为",
+      "▁C oh",
+      "▁Co h",
+      "ly mph",
+      "lymp h",
+      "l ymph",
+      "HE IGHT",
+      "▁diss ip",
+      "serial ization",
+      "泰 安",
+      "实验 组",
+      "你也 可以",
+      "你 也可以",
+      "get Context",
+      "▁Em ployees",
+      "▁Employee s",
+      "▁Employ ees",
+      "捷 克",
+      "▁外 观",
+      "▁ 外观",
+      "收到 的",
+      "收 到的",
+      "第四 款",
+      "中国 当代",
+      "数据 传输",
+      "▁Q uint",
+      "▁Qu int",
+      "▁sp indle",
+      "▁Onc ology",
+      "▁da unting",
+      "请告诉我您的选择 和解题的思路",
+      "文 华",
+      "更 深",
+      "友好 的",
+      "友 好的",
+      ":\\\\ /\\\\/",
+      "▁fier ce",
+      "end ocrine",
+      "▁Whats App",
+      "▁ 温",
+      "他 与",
+      "/' ,",
+      "/ ',",
+      "SV G",
+      "详情 请",
+      "at ro",
+      "a tro",
+      "仪器 仪表",
+      "▁t urf",
+      "▁tur f",
+      "▁tu rf",
+      "▁f olds",
+      "▁fol ds",
+      "▁fold s",
+      "▁fa ulty",
+      "▁fault y",
+      "▁need les",
+      "▁needle s",
+      "▁special ised",
+      "丽 丽",
+      "斜 阳",
+      "昏 迷",
+      "了一 段",
+      "了 一段",
+      "永远 是",
+      "▁B un",
+      "▁Bu n",
+      "▁W ins",
+      "▁Win s",
+      "▁Wi ns",
+      "att rib",
+      "attr ib",
+      "roid ery",
+      "▁ 斌",
+      "一 等",
+      "中国 市场",
+      "▁Bur k",
+      "▁Bu rk",
+      "▁T ort",
+      "▁To rt",
+      "▁Tor t",
+      "必不可 少的",
+      "t client",
+      "佛 罗",
+      "心 想",
+      "感 官",
+      "甜 美",
+      "鄙 视",
+      "하 는",
+      "nw ho",
+      "n who",
+      "Dis able",
+      "▁Pre diction",
+      "▁Pred iction",
+      "▁Predict ion",
+      "UN C",
+      "U NC",
+      "ga e",
+      "g ae",
+      "▁R ings",
+      "▁Ring s",
+      "▁Aber deen",
+      "上 皮",
+      "关 门",
+      "签 收",
+      "需要 进行",
+      "▁L IVE",
+      "▁LI VE",
+      "ast rophic",
+      "op ulmonary",
+      "玉 兰",
+      "石 子",
+      "算 出",
+      "原则 上",
+      "▁mar vel",
+      "▁Lam bert",
+      "▁Lamb ert",
+      "▁rent ing",
+      "▁s ounded",
+      "▁sound ed",
+      "▁quadr ant",
+      "▁quad rant",
+      "▁fertil izer",
+      "城 建",
+      "很 差",
+      "科 夫",
+      "孙 悟空",
+      "▁R um",
+      "▁Ru m",
+      "主要作品 有",
+      "▁ex its",
+      "▁exit s",
+      "st ations",
+      "stat ions",
+      "station s",
+      "▁para graphs",
+      "▁paragraph s",
+      "▁un acceptable",
+      "并做出 分析",
+      "re ctomy",
+      "rect omy",
+      "r ectomy",
+      "▁Sk etch",
+      "▁amb ition",
+      "▁ambit ion",
+      "▁acetyl choline",
+      "初 三",
+      "快 手",
+      "mi um",
+      "m ium",
+      "具有 较强的",
+      "▁Pac kers",
+      "▁Pack ers",
+      "▁life cycle",
+      "▁lif ecycle",
+      "▁ lifecycle",
+      "选择正确答案 并做出分析",
+      "en ol",
+      "eno l",
+      "e nol",
+      "身边 的人",
+      "身边的 人",
+      "▁m our",
+      "▁mo ur",
+      "▁Pl ains",
+      "▁Plain s",
+      "▁host ile",
+      "▁clean ers",
+      "▁cleaner s",
+      "二 甲",
+      "工 农",
+      "握 手",
+      "特 邀",
+      "ni ts",
+      "nit s",
+      "n its",
+      "▁G lor",
+      "▁Gl or",
+      "▁Glo r",
+      "tp anic",
+      "t panic",
+      "▁beh aves",
+      "▁behav es",
+      "▁behave s",
+      "▁amb ulatory",
+      "第二百六十二条第二款 和",
+      "水 下",
+      ":$ {",
+      ": ${",
+      "F ox",
+      "zi e",
+      "z ie",
+      "迈 克尔",
+      "▁oste oc",
+      "▁osteo c",
+      "▁restrict ive",
+      "▁Trans actions",
+      "▁Transaction s",
+      "他 被",
+      "并给出 依据",
+      "▁con sp",
+      "▁co nsp",
+      "▁cons p",
+      "月 子",
+      "yn b",
+      "y nb",
+      "服务 费",
+      "服刑 期间",
+      "em erg",
+      "eme rg",
+      "nw indow",
+      "nwin dow",
+      "n window",
+      "▁remed ies",
+      "▁disc overs",
+      "▁discover s",
+      "并给出依据 和解题过程",
+      "军 工",
+      "慕 容",
+      "格 雷",
+      "优化 的",
+      "优 化的",
+      "$\\ ',",
+      "$ \\',",
+      "▁E at",
+      "at ched",
+      "atch ed",
+      "▁bomb ing",
+      "▁bom bing",
+      "▁strength ened",
+      "▁strengthen ed",
+      "▁U X",
+      "▁ UX",
+      "▁h airs",
+      "▁hair s",
+      "▁ha irs",
+      "▁reg ularity",
+      "▁regular ity",
+      "▁regul arity",
+      "h Y",
+      "登 高",
+      "站 着",
+      "说出 来",
+      "说 出来",
+      "build ers",
+      "builder s",
+      "▁menstr ual",
+      "▁dis posable",
+      "高 龄",
+      "加工 厂",
+      "加 工厂",
+      "ha ss",
+      "has s",
+      "h ass",
+      "he ws",
+      "hew s",
+      "h ews",
+      "给人 一种",
+      "▁Mar qu",
+      "▁In verse",
+      "▁ 제",
+      "不 移",
+      "昔 日",
+      "赃 物",
+      "我们 家",
+      "的水 果",
+      "的 水果",
+      "全神 贯注",
+      "技术 研究",
+      "ha ven",
+      "hav en",
+      "have n",
+      "h aven",
+      "▁P iet",
+      "▁Pi et",
+      "▁Pie t",
+      "▁T alent",
+      "▁Tal ent",
+      "▁Tale nt",
+      "▁Mon ument",
+      "▁f idelity",
+      "▁fire works",
+      "晓 得",
+      "表示 出",
+      "▁Is abel",
+      "▁Isa bel",
+      "▁Ver tex",
+      "▁Vert ex",
+      "▁ Vertex",
+      "nA uthors",
+      "nAuthor s",
+      "▁prosper ity",
+      "▁reproduc ibility",
+      "史 记",
+      "AR P",
+      "A RP",
+      "▁R L",
+      "▁ RL",
+      "时不 时",
+      "时 不时",
+      "Wi ld",
+      "Wil d",
+      "W ild",
+      "▁Pa stor",
+      "▁Pas tor",
+      "▁Past or",
+      "▁Val encia",
+      "▁Vale ncia",
+      "后 世",
+      "管理 层",
+      "▁I CE",
+      "▁IC E",
+      "▁ ICE",
+      "al ese",
+      "ale se",
+      "ales e",
+      "▁Ber ks",
+      "▁Pap ua",
+      "nS ingle",
+      "nSing le",
+      "n Single",
+      "▁iso form",
+      "此外请给出 你做出选择的依据和分析",
+      "一 堂",
+      "为 零",
+      "受 力",
+      "手 脚",
+      "地下 水",
+      "地 下水",
+      "▁A de",
+      "▁Ad e",
+      "▁L TE",
+      "▁LT E",
+      "▁ LTE",
+      "ch ang",
+      "chan g",
+      "cha ng",
+      "c hang",
+      "it emap",
+      "ite map",
+      "item ap",
+      "单位出具 的财产",
+      "武 夷",
+      "素 有",
+      "听取 了",
+      "ur eth",
+      "ure th",
+      "▁S ail",
+      "▁Sa il",
+      "▁p aren",
+      "▁par en",
+      "▁pa ren",
+      "▁pare n",
+      "▁ paren",
+      "synthes ize",
+      "▁Reg istered",
+      "▁Register ed",
+      "▁dist ributing",
+      "▁distribut ing",
+      "よ う",
+      "瘦 身",
+      "经 开",
+      "bud get",
+      "c sharp",
+      "oc ative",
+      "生物 科技有限公司",
+      "▁amend ments",
+      "▁amendment s",
+      "▁hypot hetical",
+      "相 望",
+      "pi g",
+      "p ig",
+      "提供 商",
+      "ym es",
+      "yme s",
+      "y mes",
+      "▁联赛 等级",
+      "▁sher iff",
+      "prom etheus",
+      "▁consult ancy",
+      "不 强",
+      "加 成",
+      "ok k",
+      "o kk",
+      "上诉 称",
+      "上 诉称",
+      "惊人 的",
+      "惊 人的",
+      "ba tim",
+      "bat im",
+      "ew orthy",
+      "e worthy",
+      "▁todd ler",
+      "耗 费",
+      "自 知",
+      "十一 条的规定",
+      "十一条 的规定",
+      "▁sh ocks",
+      "▁shock s",
+      "▁exert ed",
+      "▁dish washer",
+      "死 人",
+      "高 价",
+      "▁w k",
+      "▁ wk",
+      "nM er",
+      "nMe r",
+      "n Mer",
+      "▁E NA",
+      "▁EN A",
+      "nD epend",
+      "nDe pend",
+      "nDep end",
+      "n Depend",
+      "ot ically",
+      "otic ally",
+      "o tically",
+      "对 孩子",
+      "然后 我",
+      "▁w re",
+      "▁wr e",
+      "▁Com pet",
+      "▁Comp et",
+      "▁Yan kees",
+      "▁Re strict",
+      "▁Rest rict",
+      "▁ Restrict",
+      "▁ed itable",
+      "▁edit able",
+      "▁ editable",
+      "熟 知",
+      "瓶 子",
+      "动物 的",
+      "动 物的",
+      "的人 工",
+      "的 人工",
+      "社会 化",
+      "gr ow",
+      "gro w",
+      "g row",
+      "▁L al",
+      "▁La l",
+      "小伙伴 们",
+      "▁M aya",
+      "▁May a",
+      "▁Ma ya",
+      "nPl aces",
+      "nPlace s",
+      "▁m ansion",
+      "▁mans ion",
+      "拜 访",
+      "▁n ylon",
+      "▁ny lon",
+      "▁Estab lish",
+      "全 心",
+      "资本 家",
+      "▁gastro pod",
+      "▁gastrop od",
+      "▁instit utes",
+      "▁institute s",
+      "▁manip ulating",
+      "倒 了",
+      "急 剧",
+      "无 脑",
+      "里 亚",
+      "浓厚 的",
+      "浓 厚的",
+      "nMus icians",
+      "nMusic ians",
+      "草 堂",
+      "nP en",
+      "nPe n",
+      "n Pen",
+      "▁f als",
+      "▁fa ls",
+      "▁fal s",
+      "代表作品 有",
+      "▁Rob erto",
+      "▁Robert o",
+      "▁Rober to",
+      "▁new borns",
+      "▁newborn s",
+      "▁o nChange",
+      "▁on Change",
+      "▁ onChange",
+      "跳 出",
+      "ct ime",
+      "c time",
+      "▁C ats",
+      "▁Ca ts",
+      "▁Cat s",
+      "c ognitive",
+      "▁pend ulum",
+      "▁occup ancy",
+      "上 架",
+      "买 单",
+      "sv n",
+      "s vn",
+      "但不 能",
+      "但 不能",
+      "季后 赛",
+      "无法 送达",
+      "▁Ed gar",
+      "▁w ired",
+      "▁wire d",
+      "▁wi red",
+      "▁ wired",
+      "▁re tract",
+      "▁ide ology",
+      "▁eigen vectors",
+      "胶 带",
+      "解 密",
+      "项 为",
+      "▁ 球队",
+      "无 任何",
+      "聪明 的",
+      "聪 明的",
+      "▁ov ary",
+      "▁ovar y",
+      "▁F ighting",
+      "▁Fight ing",
+      "修 剪",
+      "小 鱼",
+      "尾 部",
+      "撒 上",
+      "渐 近",
+      "阻 燃",
+      "建设 中",
+      "管理 所",
+      "管 理所",
+      "ac io",
+      "aci o",
+      "Service Impl",
+      "▁REF ERENCES",
+      "吊 顶",
+      "CA C",
+      "C AC",
+      "▁t eas",
+      "▁te as",
+      "▁tea s",
+      "▁ap nea",
+      "▁mer its",
+      "▁merit s",
+      "Anal yzer",
+      "▁Cap acity",
+      "▁ Capacity",
+      "▁Maced onia",
+      "石 门",
+      "▁ 更新",
+      "单片 机",
+      "▁g rav",
+      "▁gr av",
+      "▁gra v",
+      "▁ grav",
+      "▁Ev angel",
+      "▁Evan gel",
+      "万 古",
+      "外 商",
+      "早 起",
+      "ON S",
+      "O NS",
+      "的人 文",
+      "的 人文",
+      "电池 类型",
+      "cm ath",
+      "c math",
+      "▁Ob server",
+      "▁Observ er",
+      "▁Obs erver",
+      "▁ Observer",
+      "▁hur ricane",
+      "▁integr able",
+      "▁multi player",
+      "▁multip layer",
+      "nL o",
+      "n Lo",
+      "▁青 春",
+      "▁ 青春",
+      "▁A cross",
+      "▁Ac ross",
+      "▁Aut umn",
+      "Get Value",
+      "▁specific s",
+      "林 区",
+      "st m",
+      "s tm",
+      "本 研究",
+      "表面 上",
+      "表 面上",
+      "也可能 是",
+      "也 可能是",
+      "▁K NOW",
+      "▁Cr isis",
+      "▁st akes",
+      "▁stake s",
+      "增 量",
+      "工 期",
+      "nt y",
+      "n ty",
+      "}> {",
+      "} >{",
+      "圣诞 节",
+      "的 特色",
+      "▁W ade",
+      "▁Wa de",
+      "解题 过程中的",
+      "解题过程 中的",
+      "▁follic ular",
+      "台 县",
+      "复 方",
+      "异 物",
+      "避 孕",
+      "人 可以",
+      "泰 州市",
+      "▁RO I",
+      "▁R ho",
+      "▁Rh o",
+      "京东 图书",
+      "▁inc urred",
+      "循 序",
+      "收 货",
+      "笑 脸",
+      "申请 撤回",
+      "in ence",
+      "ine nce",
+      "inen ce",
+      "▁con currently",
+      "▁concurrent ly",
+      "无 愧",
+      "脊 椎",
+      "跳 转",
+      "▁Y uan",
+      "▁Yu an",
+      "Rest Controller",
+      "N Z",
+      "▁ ре",
+      "恢复 正常",
+      "日向 原告",
+      "日 向原告",
+      "nIN SERT",
+      "nINS ERT",
+      "n INSERT",
+      "▁H erman",
+      "▁Her man",
+      "▁Herm an",
+      "Account s",
+      "Ac counts",
+      "▁Bron cos",
+      "▁Port able",
+      "▁Por table",
+      "五 代",
+      "是 没",
+      "点 火",
+      "鱿 鱼",
+      "ng x",
+      "base s",
+      "ba ses",
+      "bas es",
+      "b ases",
+      "nL ight",
+      "n Light",
+      "▁arc ade",
+      "▁outbreak s",
+      "大使 馆",
+      "FO NT",
+      "F ONT",
+      "▁M itt",
+      "▁Mi tt",
+      "▁Mit t",
+      "Ban ner",
+      "B anner",
+      "布 袋",
+      "发展 成为",
+      "▁W inston",
+      "▁Win ston",
+      "▁Wins ton",
+      "▁Ch ancellor",
+      "▁ge ological",
+      "▁geo logical",
+      "▁sub cellular",
+      "议 院",
+      "▁R H",
+      "▁ RH",
+      "的 各项",
+      "AP PS",
+      "APP S",
+      "Con v",
+      "Co nv",
+      "▁C opa",
+      "▁Co pa",
+      "▁Cop a",
+      "▁en chant",
+      "▁salv age",
+      "▁New foundland",
+      "偏 离",
+      "树 枝",
+      "El em",
+      "Ele m",
+      "E lem",
+      "me ns",
+      "men s",
+      "m ens",
+      "▁P GE",
+      "▁PG E",
+      "电子 信息",
+      "▁W ife",
+      "▁Wi fe",
+      "cat enin",
+      "▁ill usion",
+      "月 第",
+      "美 金",
+      "赃 款",
+      ">] (",
+      "> ](",
+      "▁T reg",
+      "▁Tr eg",
+      "▁Tre g",
+      "ple asant",
+      "▁pod casts",
+      "▁podcast s",
+      "解题过程中的 依据是什么",
+      "▁desc endants",
+      "性 及",
+      "特 地",
+      "UP S",
+      "U PS",
+      "是否 会",
+      "os se",
+      "oss e",
+      "ot opic",
+      "oto pic",
+      "o topic",
+      "▁H imal",
+      "▁Hi mal",
+      "▁Him al",
+      "▁Ap artments",
+      "▁Apart ments",
+      "▁Apartment s",
+      "ff i",
+      "f fi",
+      "og i",
+      "o gi",
+      "可 获得",
+      "的最 好",
+      "的 最好",
+      "了解 更多",
+      "Ab ility",
+      "▁Far mer",
+      "▁Farm er",
+      "▁Gard ner",
+      "德 惠",
+      "▁ 系列",
+      "十一 五",
+      "和 生产",
+      "罪犯 杨",
+      "▁G row",
+      "▁Gr ow",
+      "▁Gro w",
+      "▁$ ${\\\\",
+      "▁$$ {\\\\",
+      "▁$${ \\\\",
+      "stit uted",
+      "stitute d",
+      "stitu ted",
+      "separ ated",
+      "跟 上",
+      "骗 取",
+      "▁其 七",
+      "事业 部",
+      "win ner",
+      "w inner",
+      "▁Per forming",
+      "▁Perform ing",
+      "Pl ane",
+      "Plan e",
+      "P lane",
+      "▁N ano",
+      "▁Na no",
+      "▁Nan o",
+      "和分析 的过程",
+      "chrom ium",
+      "▁Trib une",
+      "▁permit ting",
+      "▁perm itting",
+      "特 许",
+      "讲 课",
+      "▁T ian",
+      "▁Ti an",
+      "▁D iane",
+      "▁Di ane",
+      "▁Re lay",
+      "▁Rel ay",
+      "nJ oseph",
+      "不 耐",
+      "尿 素",
+      "为 单位",
+      "表情 包",
+      "ab is",
+      "abi s",
+      "a bis",
+      "▁Gu ests",
+      "▁Guest s",
+      "▁tri umph",
+      "▁ 熊",
+      "女 排",
+      "龙 潭",
+      "住 天津市",
+      "▁F ully",
+      "▁Full y",
+      "▁Ful ly",
+      "Aw esome",
+      "▁Ad dition",
+      "▁Add ition",
+      "奖 品",
+      "U tf",
+      "的知识 和",
+      "的 知识和",
+      "部 副部长",
+      "▁K ard",
+      "▁Kar d",
+      "▁Ka rd",
+      "Att ach",
+      "▁Leg ends",
+      "▁Legend s",
+      "▁tang ible",
+      "▁comput ationally",
+      "▁computational ly",
+      "▁computation ally",
+      "书 包",
+      "芝 加",
+      "团队 的",
+      "团 队的",
+      "展现 出",
+      "doc k",
+      "do ck",
+      "d ock",
+      "▁hom ogen",
+      "互 利",
+      "床 垫",
+      "更 低",
+      "砂 浆",
+      "考 的",
+      "舰 队",
+      "不是 我",
+      "不 是我",
+      "深深 的",
+      "深 深的",
+      "he us",
+      "h eus",
+      "se ll",
+      "sel l",
+      "s ell",
+      "sp atial",
+      "▁cool ed",
+      "▁re juven",
+      "▁s erializer",
+      "▁serial izer",
+      "▁serialize r",
+      "▁ serializer",
+      "ッ ク",
+      "插 画",
+      "江 水",
+      "nF ive",
+      "n Five",
+      "▁ASC II",
+      "El astic",
+      "nV ersion",
+      "n Version",
+      "不 育",
+      "病 程",
+      "研究生 学历",
+      "prot ective",
+      "▁prof iciency",
+      "▁ 가",
+      "年 头",
+      "是 新",
+      "波 浪",
+      "ra k",
+      "r ak",
+      "是 两个",
+      "行政 区域",
+      "行政区 域",
+      "as poon",
+      "asp oon",
+      "▁F iber",
+      "▁Fib er",
+      "▁Fi ber",
+      "en stein",
+      "ens tein",
+      "▁De tail",
+      "▁Det ail",
+      "▁ Detail",
+      "▁econom ists",
+      "▁economist s",
+      "▁antire troviral",
+      "已 成",
+      "闭 合",
+      "▁V ote",
+      "▁Vo te",
+      "▁ Vote",
+      "▁us ize",
+      "▁u size",
+      "▁trouble d",
+      "▁trou bled",
+      "改 判",
+      "春 色",
+      "智 库",
+      "歉 意",
+      "保 字第",
+      "▁bu ys",
+      "▁buy s",
+      "Or d",
+      "O rd",
+      "开庭 传票",
+      "ai ser",
+      "ais er",
+      "aise r",
+      "a iser",
+      "▁Tr igger",
+      "▁ Trigger",
+      "▁de crypt",
+      "▁dec rypt",
+      "▁wild erness",
+      "K R",
+      "▁ Р",
+      "全 线",
+      "已 然",
+      "月 被",
+      "清 远",
+      "▁p aved",
+      "▁pa ved",
+      "▁pave d",
+      "▁pav ed",
+      "ot rophic",
+      "▁get User",
+      "▁ getUser",
+      "▁Count ies",
+      "▁smooth ing",
+      "▁smo othing",
+      "并告诉我您的选择 和分析的过程",
+      "纽 带",
+      "Ar m",
+      "A rm",
+      "中医 院",
+      "中 医院",
+      "MP LE",
+      "MPL E",
+      "全力 以赴",
+      "牢记 使命",
+      "w izard",
+      "▁Con cord",
+      "▁Conc ord",
+      "▁Me aning",
+      "▁Mean ing",
+      "▁air flow",
+      "▁cr aving",
+      "▁cra ving",
+      "▁flo oded",
+      "▁flood ed",
+      "▁book ings",
+      "▁booking s",
+      "▁careg iver",
+      "光 度",
+      "染 料",
+      "树 上",
+      "AS A",
+      "A SA",
+      "ta s",
+      "t as",
+      "和解 为由",
+      "ot ech",
+      "ote ch",
+      "o tech",
+      "▁ch ina",
+      "▁chi na",
+      "▁chin a",
+      "aff inity",
+      "affin ity",
+      "▁ceremon ies",
+      "▁ 左",
+      "眉 毛",
+      "北京 的",
+      "就有 了",
+      "就 有了",
+      "最快 的",
+      "最 快的",
+      "要 怎么",
+      "被执行人 黄",
+      "▁Bas il",
+      "▁Ba sil",
+      "▁P orsche",
+      "▁Por sche",
+      "▁Ex ercises",
+      "▁Exercise s",
+      "▁travel lers",
+      "▁trav ellers",
+      "其 为",
+      "▁开 始",
+      "▁ 开始",
+      "的 看法",
+      "并给出 合理的解释",
+      "▁string ent",
+      "▁motiv ational",
+      "▁motivation al",
+      "▁ 邱",
+      "▁E h",
+      "▁主 体",
+      "▁ 主体",
+      "这里 有",
+      "这 里有",
+      "赔偿 被害人",
+      "gen esis",
+      "gene sis",
+      "genes is",
+      "▁T ropical",
+      "业 者",
+      "对 前款",
+      "得益 于",
+      "看 电视",
+      "em ma",
+      "emm a",
+      "e mma",
+      "no nce",
+      "non ce",
+      "n once",
+      "▁B ooth",
+      "▁Bo oth",
+      "▁Boot h",
+      "▁Mal colm",
+      "并给出你的 解题过程",
+      "并给出你的解题 过程",
+      "▁coinc idence",
+      "▁coincide nce",
+      "秋 日",
+      "we re",
+      "wer e",
+      "w ere",
+      "▁j elly",
+      "▁U PDATE",
+      "▁UP DATE",
+      "▁ UPDATE",
+      "Cap acity",
+      "Ret ention",
+      "▁land lord",
+      "In strument",
+      "W G",
+      "节 水",
+      "已 交纳",
+      "谅解 书",
+      "EX EC",
+      "ve ster",
+      "ves ter",
+      "vest er",
+      "v ester",
+      "▁He rein",
+      "▁Here in",
+      "▁Her ein",
+      "▁spor adic",
+      "惯 性",
+      "打 入",
+      "信息 网",
+      "▁Commun ities",
+      "Pr ef",
+      "Pre f",
+      "P ref",
+      "可以 提高",
+      "可 以提高",
+      "▁op pose",
+      "▁sed iments",
+      "▁sediment s",
+      "刺 客",
+      "沉 降",
+      "▁ 综合",
+      "一 季度",
+      "▁v ib",
+      "▁vi b",
+      "中国 政府",
+      "但是 我们",
+      "但是我 们",
+      "但 是我们",
+      "nH ouse",
+      "n House",
+      "▁D riving",
+      "▁Dr iving",
+      "▁chlor ine",
+      "▁in ducible",
+      "▁indu cible",
+      "▁Ent repreneur",
+      "型 和",
+      "总 站",
+      "洞 庭",
+      "矿 区",
+      "造 林",
+      "iv ered",
+      "ive red",
+      "iver ed",
+      "▁A ging",
+      "▁Ag ing",
+      "im ester",
+      "ime ster",
+      "imes ter",
+      "次 要",
+      "茶 匙",
+      "另一 方",
+      "另 一方",
+      "投入 到",
+      "持卡 人",
+      "芝加 哥",
+      "]\\\\ )",
+      "] \\\\)",
+      "Un defined",
+      "U ndefined",
+      "▁sil encing",
+      "独 有",
+      "处理 好",
+      "老人 的",
+      "老 人的",
+      "▁arr ests",
+      "▁arrest s",
+      "▁bronch ial",
+      "▁prefer ably",
+      "然后告诉我您的最佳选择 和分析思路",
+      "ld ap",
+      "l dap",
+      "▁r all",
+      "▁ra ll",
+      "empt yset",
+      "empty set",
+      "Primary Key",
+      "▁Wh olesale",
+      "▁Whole sale",
+      "▁scal ability",
+      "▁ 包",
+      "赛 区",
+      "冻结 的",
+      "ir ds",
+      "ird s",
+      "▁iod ine",
+      "亿 万",
+      "成 败",
+      "*( (",
+      "* ((",
+      "▁广州 市",
+      "▁ 广州市",
+      "end ra",
+      "e ndra",
+      "▁Bo yd",
+      "▁Boy d",
+      "▁E scape",
+      "▁Esc ape",
+      "▁ed ible",
+      "▁cytos olic",
+      "白 血",
+      "都 属于",
+      "li sta",
+      "list a",
+      "lis ta",
+      "l ista",
+      "phen yl",
+      "p henyl",
+      "▁gran ular",
+      "▁granul ar",
+      "签 到",
+      "脱 硫",
+      "问 责",
+      "优良 的",
+      "对他 的",
+      "对 他的",
+      "in tern",
+      "int ern",
+      "inter n",
+      "pur ple",
+      "▁N exus",
+      "▁C astro",
+      "▁Cast ro",
+      "▁Cas tro",
+      "▁N early",
+      "▁Ne arly",
+      "▁Near ly",
+      "▁W elfare",
+      "▁Wel fare",
+      "nBas ically",
+      "▁imp acting",
+      "▁impact ing",
+      "▁n c",
+      "▁ nc",
+      "十分 钟",
+      "十 分钟",
+      "有 需要",
+      "▁int rav",
+      "▁intr av",
+      "▁intra v",
+      "诉 刑诉",
+      "二〇一七年 三",
+      "es cence",
+      "esc ence",
+      "▁Screen ing",
+      "▁Scre ening",
+      "林 场",
+      "比 拟",
+      "航 行",
+      "路 桥",
+      "AE S",
+      "A ES",
+      "努力 的",
+      "努 力的",
+      "这种 事",
+      "目前 正在",
+      "▁H uang",
+      "▁Hu ang",
+      "nS tring",
+      "nSt ring",
+      "nStr ing",
+      "n String",
+      "m z",
+      "大 卫",
+      "忠 实",
+      "颜 料",
+      "▁C MV",
+      "▁CM V",
+      "▁ CMV",
+      "▁Ex tr",
+      "▁Ext r",
+      "▁Q ing",
+      "trans lated",
+      "translate d",
+      "transl ated",
+      "Get Component",
+      "元 气",
+      "荷 载",
+      "▁F iji",
+      "▁Fi ji",
+      "▁We ak",
+      "ab ella",
+      "abel la",
+      "abe lla",
+      "il ical",
+      "ili cal",
+      "▁Re leased",
+      "▁Release d",
+      "effective ness",
+      "effect iveness",
+      "▁trans porters",
+      "▁transport ers",
+      "▁transporter s",
+      "k d",
+      "▁ 则",
+      "▼ \\",
+      "双 核",
+      "山 坡",
+      "洗 脸",
+      "读 懂",
+      "增长 的",
+      "增 长的",
+      "建筑 师",
+      "Pl ain",
+      "P lain",
+      "▁In dies",
+      "▁Ind ies",
+      "▁Ver non",
+      "▁Vern on",
+      "未经法庭许可中途退 庭的",
+      "转变 为",
+      "转 变为",
+      "▁A kt",
+      "▁Ak t",
+      "▁主要 参数",
+      "ex istence",
+      "exist ence",
+      "pres ident",
+      "SEQU ENTIAL",
+      "▁Ep iscopal",
+      "冷 酷",
+      "华 润",
+      "hh h",
+      "h hh",
+      "以 确保",
+      "和 安全",
+      "更 方便",
+      "的主 人",
+      "的 主人",
+      "▁se wer",
+      "▁sew er",
+      "▁Abb ott",
+      "▁P acket",
+      "▁Pac ket",
+      "▁Pack et",
+      "▁ Packet",
+      "▁bl ender",
+      "▁ble nder",
+      "▁blend er",
+      "本案 在审理过程中",
+      "曾 被",
+      "]< <",
+      "] <<",
+      "}` )\\",
+      "} `)\\",
+      "中国 移动",
+      "但是 现在",
+      "iv able",
+      "iva ble",
+      "▁Sat urn",
+      "▁Sa turn",
+      "regn ancy",
+      "rem aining",
+      "▁form ulae",
+      "▁formula e",
+      "为 凭",
+      "拷 贝",
+      "甚 么",
+      "Ga s",
+      "G as",
+      "时 免费",
+      "im eo",
+      "ime o",
+      "un dy",
+      "und y",
+      "▁s eper",
+      "▁se per",
+      "▁sep er",
+      "差异 无统计学意义",
+      "▁Tom orrow",
+      "亲 朋",
+      "到 此",
+      "定 金",
+      "不等 于",
+      "不 等于",
+      "睡 不着",
+      "密切 相关",
+      "es try",
+      "est ry",
+      "estr y",
+      "e stry",
+      "▁M ast",
+      "▁Ma st",
+      "▁Mas t",
+      "▁B ritt",
+      "▁Br itt",
+      "▁Brit t",
+      "▁Bri tt",
+      "▁st eer",
+      "▁ste er",
+      "▁phil osopher",
+      "▁philosoph er",
+      "@ {",
+      "凹 陷",
+      "▁B een",
+      "▁Be en",
+      "▁Bee n",
+      "▁ Been",
+      "▁T ODAY",
+      "▁TO DAY",
+      "▁key note",
+      "▁val ence",
+      "▁ valence",
+      "在 西",
+      "鹏 飞",
+      "▁H TC",
+      "▁HT C",
+      "制度 改革",
+      "▁ri bs",
+      "▁rib s",
+      "pec ific",
+      "tout put",
+      "t output",
+      "▁Wal let",
+      "▁Wall et",
+      "▁ Wallet",
+      "nWin ners",
+      "icul ously",
+      "iculous ly",
+      "起诉必须 符合下列条件",
+      "▁substit utions",
+      "▁substitution s",
+      "T y",
+      "并 茂",
+      "解 码",
+      "ct r",
+      "c tr",
+      "的 这个",
+      "ma xn",
+      "max n",
+      "▁F SH",
+      "▁FS H",
+      "▁副 标题",
+      "可能 导致",
+      "核心 价值观",
+      "▁Less ons",
+      "▁Lesson s",
+      "▁un happy",
+      "follow ing",
+      "f ollowing",
+      "其行为已构成 盗窃罪",
+      "▁stat utory",
+      "凄 凉",
+      "方 形",
+      "权 属",
+      "离 别",
+      "EE K",
+      "▁ 杭州",
+      "▁比 赛",
+      "▁ 比赛",
+      "综合 治理",
+      "设计 方案",
+      "ORM AL",
+      "Par cel",
+      "▁h olog",
+      "▁hol og",
+      "▁ho log",
+      "▁anti genic",
+      "▁antigen ic",
+      "花 草",
+      "还 很",
+      "ce p",
+      "c ep",
+      "▁} :",
+      "▁ }:",
+      "ge ms",
+      "gem s",
+      "g ems",
+      "一点 也不",
+      "分别 求出",
+      "你 选择的依据",
+      "aff olds",
+      "affold s",
+      "▁Pres byterian",
+      "对 口",
+      "山 楂",
+      "饭 菜",
+      "nM a",
+      "n Ma",
+      "▁r v",
+      "▁ rv",
+      "▁M ey",
+      "▁Me y",
+      "儿童 文学",
+      "共产 主义",
+      "第二 十五",
+      "第 二十五",
+      "交通警察 大队",
+      "仔细地 阅读选项",
+      "▁Str ength",
+      "▁Stre ngth",
+      "并给出 你选择的依据",
+      "▁ind ispens",
+      "请您务必 仔细地阅读选项",
+      "并给出你选择的依据 和分析思路",
+      "命 的",
+      "肯定 的",
+      "肯 定的",
+      "节目 中",
+      "要不 然",
+      "要 不然",
+      "▁p on",
+      "▁po n",
+      "▁ pon",
+      "Grad ient",
+      "八 字",
+      "小 幅",
+      "ne ck",
+      "nec k",
+      "n eck",
+      "rupt s",
+      "ru pts",
+      "rup ts",
+      "▁st din",
+      "▁std in",
+      "▁ stdin",
+      "▁task ed",
+      "scrib ers",
+      "scri bers",
+      "scribe rs",
+      "scriber s",
+      "▁Sp atial",
+      "▁prop ensity",
+      "▁apprent ices",
+      "作 法",
+      "(\" _",
+      "( \"_",
+      "充要 条件",
+      "冠军 联赛",
+      "st amp",
+      "sta mp",
+      "▁dist ributors",
+      "▁distribut ors",
+      "▁distributor s",
+      "▁ 血",
+      "al m",
+      "a lm",
+      "▁异 界",
+      "▁ 异界",
+      "没有 被",
+      "NN NN",
+      "▁spark ling",
+      "师 德",
+      "湖 州",
+      "自动 履行",
+      "▁br oth",
+      "▁bro th",
+      "▁sp elled",
+      "▁spell ed",
+      "内 径",
+      "和 李",
+      "宇 航",
+      "ст и",
+      "с ти",
+      "经济 学家",
+      "经济学 家",
+      "hy brid",
+      "ne rror",
+      "ner ror",
+      "n error",
+      "▁b ites",
+      "▁bi tes",
+      "▁bit es",
+      "▁bite s",
+      "sign als",
+      "signal s",
+      "sig nals",
+      "一 听",
+      "抗 美",
+      "推 拿",
+      "抵抗 力",
+      "▁M AG",
+      "▁MA G",
+      "▁ MAG",
+      "公司 员工",
+      "uv ian",
+      "u vian",
+      "▁In cluded",
+      "▁Include d",
+      "▁business people",
+      "一 角",
+      "封 印",
+      "os x",
+      "o sx",
+      "人们 在",
+      "人 们在",
+      "科技 的",
+      "等人 的",
+      "等 人的",
+      "故 对原告",
+      "▁W rap",
+      "▁Wr ap",
+      "▁ Wrap",
+      "进一步 提升",
+      "之 风",
+      "终 生",
+      "者 为",
+      "背 影",
+      "AN A",
+      "A NA",
+      "Crypt o",
+      "C rypto",
+      "▁reluct ant",
+      "IT T",
+      "I TT",
+      "gl m",
+      "g lm",
+      "预交 的",
+      "Vo id",
+      "V oid",
+      "place ments",
+      "plac ements",
+      "placement s",
+      "任 性",
+      "台 的",
+      "职能 部门",
+      "▁g aze",
+      "▁ga ze",
+      "▁gaz e",
+      "解决 的问题",
+      "解决的 问题",
+      "De term",
+      "Det erm",
+      "D eterm",
+      "lor ida",
+      "▁l acked",
+      "▁lack ed",
+      "▁sculpt or",
+      "▁furnish ings",
+      "ki l",
+      "k il",
+      "▁y y",
+      "▁ yy",
+      "在 什么",
+      "能量 的",
+      "能 量的",
+      "镇江 市",
+      "镇 江市",
+      "ab it",
+      "abi t",
+      "a bit",
+      "▁WH EN",
+      "▁parad ise",
+      "▁trans istor",
+      "▁wh itespace",
+      "▁white space",
+      "▁whites pace",
+      "海 参",
+      "着 实",
+      "额 为",
+      "高 明",
+      "▁C how",
+      "▁Ch ow",
+      "▁Cho w",
+      "▁Keep ing",
+      "地 貌",
+      "巨 星",
+      "雄 性",
+      "▁ 状态",
+      "出发 点",
+      "线上 的",
+      "线 上的",
+      "▁$ ${",
+      "▁$$ {",
+      "诗词 正文",
+      "▁B ACK",
+      "▁BA CK",
+      "▁ BACK",
+      "▁Str aight",
+      "▁Stra ight",
+      "pine phrine",
+      "▁Pr actical",
+      "▁Pract ical",
+      "long rightarrow",
+      "A J",
+      "亚 纲",
+      "小 品",
+      "曲 面",
+      "贵 宾",
+      "蓄 电池",
+      "SC HE",
+      "S CHE",
+      "TE RS",
+      "TER S",
+      "T ERS",
+      "otrop in",
+      "▁dent ists",
+      "▁dentist s",
+      "▁neoplas ms",
+      "七 夕",
+      "使 我",
+      "重 现",
+      "ar f",
+      "a rf",
+      "玩 手机",
+      "非常 大",
+      "nw in",
+      "n win",
+      "▁F ry",
+      "▁Fr y",
+      "Dam age",
+      "user Name",
+      "▁bio film",
+      "俱 全",
+      "则 会",
+      "温 差",
+      "耗 材",
+      "身 穿",
+      "阶段 性",
+      "▁S ind",
+      "▁Si nd",
+      "▁Sin d",
+      "环境保护 局",
+      "▁Co operative",
+      "▁Cooper ative",
+      "上 岗",
+      "友 善",
+      "▁S tern",
+      "▁St ern",
+      "▁Ste rn",
+      "▁Ster n",
+      "▁mos que",
+      "▁sample r",
+      "▁sam pler",
+      "author ize",
+      "set Content",
+      "▁lower case",
+      "▁before hand",
+      "get ter",
+      "g etter",
+      "nS ound",
+      "nSo und",
+      "n Sound",
+      "浙江省 杭州市",
+      "Upper Case",
+      "▁pneumonia e",
+      "▁pictures que",
+      "Re b",
+      "R eb",
+      "可 实现",
+      "文件 中",
+      "qu ark",
+      "qua rk",
+      "in variant",
+      "▁Aware ness",
+      "▁protagon ist",
+      "平面 内",
+      "▁p ys",
+      "▁py s",
+      "由原告 抚养",
+      "需要 注意的是",
+      "需要注意 的是",
+      "▁conf ocal",
+      "▁pass ages",
+      "▁passage s",
+      "专 员",
+      "也 成为",
+      "言 律诗",
+      "或者 未经法庭许可中途退庭的",
+      "出 栏",
+      "礼 包",
+      "位置 的",
+      "F old",
+      "▁St range",
+      "▁Str ange",
+      "▁Stra nge",
+      "▁De ployment",
+      "▁Deploy ment",
+      "▁ Deployment",
+      "▁government al",
+      "▁govern mental",
+      "可 予",
+      "已 将",
+      "老 区",
+      "被 我",
+      "J av",
+      "▁K ö",
+      ">\\ ';\\",
+      ">\\' ;\\",
+      "> \\';\\",
+      "第七 十二条",
+      "越来越多 的人",
+      "越来越多的 人",
+      "Pat ient",
+      "▁M agnet",
+      "▁Mag net",
+      "▁Magn et",
+      "▁tra ilers",
+      "▁trail ers",
+      "▁trailer s",
+      "▁exhaust ive",
+      "余 家",
+      "双 赢",
+      "太 仓",
+      "补 肾",
+      "过 着",
+      "in gen",
+      "ing en",
+      "inge n",
+      "re ements",
+      "ree ments",
+      "reement s",
+      "想 来",
+      "hu ang",
+      "hua ng",
+      "h uang",
+      "▁E arn",
+      "▁Ear n",
+      "ar ching",
+      "arch ing",
+      "arc hing",
+      "▁T error",
+      "▁Te rror",
+      "▁Ter ror",
+      "▁Terr or",
+      "bin dings",
+      "bind ings",
+      "binding s",
+      "▁Object ives",
+      "▁Objective s",
+      "想 买",
+      "方法 将",
+      "中小 学生",
+      "中小学 生",
+      "中 小学生",
+      "Al loc",
+      "All oc",
+      "ag net",
+      "agn et",
+      "agne t",
+      "iff any",
+      "▁P ablo",
+      "▁Pa blo",
+      "新疆 生产建设兵团",
+      "blue print",
+      "▁An onymous",
+      "▁un healthy",
+      "功 名",
+      "勾 股",
+      "的 发",
+      "是 日本",
+      "比较 适合",
+      "▁ur eth",
+      "▁ure th",
+      "▁ ureth",
+      "▁brew ery",
+      "安 检",
+      "肩 负",
+      "轻 型",
+      "截止 到",
+      "In to",
+      "Int o",
+      "I nto",
+      "▁hemod ialysis",
+      "用 料",
+      "组 分",
+      "家人 的",
+      "家 人的",
+      "让 别人",
+      "▁S ard",
+      "▁Sar d",
+      "▁Sa rd",
+      "履行债务 的能力",
+      "▁Bed ford",
+      "▁col onic",
+      "▁colon ic",
+      "▁rel iance",
+      "Function al",
+      "F unctional",
+      "Null Exception",
+      "es m",
+      "e sm",
+      "ia ge",
+      "i age",
+      "▁K ub",
+      "▁Ku b",
+      "▁R BC",
+      "▁RB C",
+      "青年 教师",
+      "le name",
+      "len ame",
+      "l ename",
+      "▁Os aka",
+      "▁educ ating",
+      "▁ 乌",
+      "在 建",
+      "心 头",
+      "IT ER",
+      "ITE R",
+      "I TER",
+      "▁s unk",
+      "▁su nk",
+      "▁sun k",
+      "▁J orge",
+      "▁st ride",
+      "▁str ide",
+      "▁stri de",
+      "忧 伤",
+      "提 拔",
+      "▁f f",
+      "▁ ff",
+      "垃圾 桶",
+      "不等式 的解集",
+      "不等式的解 集",
+      "▁cr ater",
+      "▁crate r",
+      "▁cra ter",
+      "▁Inter pret",
+      "接 送",
+      "的 情",
+      "中央 党校",
+      "但 实际上",
+      "▁Veg et",
+      "▁Ve get",
+      ", )",
+      "地 道",
+      "影 音",
+      "▁E ns",
+      "▁En s",
+      "UR CES",
+      "URCE S",
+      "co bra",
+      "cob ra",
+      "ie der",
+      "ied er",
+      "i eder",
+      "▁t idy",
+      "▁ti dy",
+      "▁tid y",
+      "alc ulus",
+      "alcul us",
+      "▁undergo es",
+      "滞 后",
+      "▁本 名",
+      "VE NT",
+      "V ENT",
+      "hi ts",
+      "hit s",
+      "h its",
+      "▁L OT",
+      "▁LO T",
+      "▁D eer",
+      "▁De er",
+      "▁Dee r",
+      "的免费 政策",
+      "lam ide",
+      "l amide",
+      "追索 劳动报酬",
+      "▁And hra",
+      "▁cardiomy opathy",
+      "痉 挛",
+      "ou ve",
+      "一项 住宿",
+      "▁W ays",
+      "▁Way s",
+      "▁Wa ys",
+      "for ces",
+      "force s",
+      "qu otes",
+      "quot es",
+      "quote s",
+      "quo tes",
+      "▁P omer",
+      "▁Po mer",
+      "▁Pom er",
+      "▁Rec ipe",
+      "▁ Recipe",
+      "你 最",
+      "品 格",
+      "对 新",
+      "世界 观",
+      "的 医疗",
+      "的 成长",
+      "的高 速",
+      "的 高速",
+      "▁For ever",
+      "▁Fore ver",
+      "▁pr imate",
+      "▁prim ate",
+      "▁pri mate",
+      "▁Ch ambers",
+      "▁Chamber s",
+      "▁Cham bers",
+      "▁reg iment",
+      "▁regime nt",
+      "▁regimen t",
+      "简 明",
+      "远 了",
+      "LO Y",
+      "▁S AY",
+      "▁SA Y",
+      "重点 项目",
+      "▁Pier ce",
+      "inter pol",
+      "▁f ountain",
+      "▁expat riates",
+      "▁expatriate s",
+      "书 房",
+      "卖 给",
+      "吃 亏",
+      "无 解",
+      "Co v",
+      "C ov",
+      "ya m",
+      "y am",
+      "木 地板",
+      "本案 系",
+      "正文 语种",
+      "▁bi fur",
+      "▁bif ur",
+      "▁sh ining",
+      "▁fair ness",
+      "一项住宿 的免费政策",
+      "▁section al",
+      "▁ sectional",
+      "享有的 债权依法受法律保护",
+      "三 类",
+      "二 字",
+      "工 科",
+      "牙 刷",
+      "]$ ,",
+      "] $,",
+      "▁ER K",
+      "▁j ug",
+      "▁ju g",
+      "▁ jug",
+      "▁o mn",
+      "▁om n",
+      "提供 被告",
+      "终结执行 后",
+      "ins ulin",
+      "fe atured",
+      "feature d",
+      "feat ured",
+      "卧 龙",
+      "双 边",
+      "菌 株",
+      "飞 跃",
+      "的 文字",
+      "sin k",
+      "si nk",
+      "s ink",
+      "em por",
+      "emp or",
+      "▁Be au",
+      "▁Bea u",
+      "▁S tim",
+      "▁St im",
+      "sh apes",
+      "shape s",
+      "sha pes",
+      "▁hurd les",
+      "▁Support ing",
+      "▁cart ridges",
+      "▁cartridge s",
+      "单位出具的财产 查询回执",
+      "拟 合",
+      "V an",
+      "不停 地",
+      "都能 够",
+      "都 能够",
+      "\": \"/",
+      "\":\" /",
+      "▁第 二章",
+      "▁第二 章",
+      "▁ 第二章",
+      "提请 减刑",
+      "▁T rap",
+      "▁Tr ap",
+      "▁Tra p",
+      "ur etic",
+      "ure tic",
+      "aut ical",
+      "au tical",
+      "▁in mates",
+      "▁juris dict",
+      "并给出相应的 解题过程",
+      "▁cardinal ity",
+      "▁commercial s",
+      "咖 喱",
+      "日 日",
+      "赵 丽",
+      "ig u",
+      "i gu",
+      "五千 元",
+      "五 千元",
+      "知名 品牌",
+      "ch nic",
+      "chn ic",
+      "nH enry",
+      "▁question able",
+      "侦 察",
+      "▁F TP",
+      "▁FT P",
+      "▁ FTP",
+      "▁p yt",
+      "▁py t",
+      "▁L OSS",
+      "▁LO SS",
+      "s phinx",
+      "▁p urse",
+      "▁pur se",
+      "▁purs e",
+      "art ifacts",
+      "artifact s",
+      "▁expans ive",
+      "N u",
+      "RA N",
+      "R AN",
+      "▁sugar s",
+      "▁sug ars",
+      "▁Adv ance",
+      "▁Mat thews",
+      "▁Matt hews",
+      "▁Matthew s",
+      "▁empower ment",
+      ">` \\",
+      "> `\\",
+      "结果 单",
+      "请求 撤销",
+      "▁SU CH",
+      "▁Dem onstr",
+      "▁Demo nstr",
+      "▁re usable",
+      "▁simpl ifies",
+      "恒 星",
+      "Ge ne",
+      "Gen e",
+      "G ene",
+      "ba by",
+      "bab y",
+      "b aby",
+      "▁A nk",
+      "▁An k",
+      "▁Ig M",
+      "▁B EST",
+      "▁BE ST",
+      "FAIL URE",
+      "被执行人应当 履行义务",
+      "▁Est imation",
+      "▁Estim ation",
+      "第二百五十七 条第一款第",
+      "后 因",
+      "志 勇",
+      "或 将",
+      "未 曾",
+      "进 村",
+      "造 假",
+      "▁A mp",
+      "▁Am p",
+      "▁neuro t",
+      "▁neur ot",
+      "▁ur ging",
+      "▁urg ing",
+      "comp uted",
+      "comput ed",
+      "compute d",
+      "▁end uring",
+      "▁entire ty",
+      "▁Ge ometric",
+      "▁Geo metric",
+      "at ie",
+      "ati e",
+      "a tie",
+      "▁S HE",
+      "▁SH E",
+      "一线 城市",
+      "一 线城市",
+      "▁Ax is",
+      "▁ Axis",
+      "▁T umor",
+      "▁Tu mor",
+      "▁Tum or",
+      "ncl ient",
+      "n client",
+      "▁Man age",
+      "▁Manag e",
+      "▁ Manage",
+      "▁extrem ity",
+      "▁associ ative",
+      "▁ 好",
+      "上 加",
+      "在 后",
+      "▁G IF",
+      "▁GI F",
+      "管理 体制",
+      "这个 数是",
+      "这个数 是",
+      "这 个数是",
+      "nInd ex",
+      "n Index",
+      "▁Middle s",
+      "▁Midd les",
+      "▁separ ates",
+      "▁separate s",
+      "▁ 施",
+      "演 习",
+      "青 菜",
+      "nl l",
+      "n ll",
+      "▁S erg",
+      "▁Se rg",
+      "▁Ser g",
+      "▁牺牲 情况",
+      "▁ps oriasis",
+      "String Builder",
+      "填 上",
+      "准备 的",
+      "爱上 了",
+      "爱 上了",
+      "St eve",
+      "Ste ve",
+      "om egal",
+      "ome gal",
+      "omega l",
+      "▁hyp oxic",
+      "▁hypox ic",
+      "▁Professional s",
+      "▁Profession als",
+      "朦 胧",
+      "画 出",
+      "订 立",
+      "早期 的",
+      "早 期的",
+      "▁I nfl",
+      "▁In fl",
+      "▁Inf l",
+      "▁r RNA",
+      "▁bo son",
+      "▁bos on",
+      "本院依法 受理",
+      "X M",
+      "f q",
+      "い た",
+      "飘 飘",
+      "儿子 的",
+      "儿 子的",
+      "▁U DP",
+      "市场 上的",
+      "市场上 的",
+      "Tr ust",
+      "T rust",
+      "▁An dr",
+      "▁And r",
+      "▁occup ies",
+      "nFurther more",
+      "▁confidential ity",
+      "国 军",
+      "在 手",
+      "篇 幅",
+      "苦 恼",
+      "▁` \"",
+      "im ple",
+      "impl e",
+      "imp le",
+      "▁HD MI",
+      "▁ HDMI",
+      "先进 工作者",
+      "▁Gr ill",
+      "▁ag ile",
+      "服饰 有限公司",
+      "▁NAS CAR",
+      "▁Sund ays",
+      "▁Sunday s",
+      "▁Sun days",
+      "又 叫",
+      "平 湖",
+      "有 道",
+      "计算 的",
+      "大大 提高",
+      "VER TI",
+      "▁o der",
+      "▁od er",
+      "▁ oder",
+      "sm outh",
+      "s mouth",
+      "▁Error s",
+      "▁Er rors",
+      "▁Err ors",
+      "▁ Errors",
+      "▁async io",
+      "下 发",
+      "乳 胶",
+      "可 调",
+      "放 映",
+      "立 式",
+      "的 房产",
+      "的 看着",
+      "Har m",
+      "Ha rm",
+      "H arm",
+      "▁L ad",
+      "▁La d",
+      "▁Ch ance",
+      "▁Cha nce",
+      "▁Chan ce",
+      "▁cont ral",
+      "▁contr al",
+      "▁Mart inez",
+      "▁Martin ez",
+      "Al e",
+      "A le",
+      "分式 的",
+      "分 式的",
+      "建设 银行",
+      "紧紧 围绕",
+      "PRO TO",
+      "▁K afka",
+      "▁ Kafka",
+      "▁geometric al",
+      "▁geomet rical",
+      "ш и",
+      "由 其",
+      "▁sp ikes",
+      "▁spike s",
+      "月 山",
+      "秒 杀",
+      "tt e",
+      "t te",
+      "完 之后",
+      "er en",
+      "ere n",
+      "e ren",
+      "▁B uk",
+      "▁Bu k",
+      "经营 理念",
+      "▁Moh ammad",
+      "▁elong ation",
+      "不 入",
+      "谁 说",
+      "已经 很",
+      "OL ED",
+      "O LED",
+      "iction aries",
+      "华 硕",
+      "矿 井",
+      "网 店",
+      "鲤 鱼",
+      "▁m v",
+      "▁ mv",
+      "co rp",
+      "cor p",
+      "c orp",
+      "故意 杀人",
+      "程序 设计",
+      "自我 介绍",
+      "▁flo ats",
+      "▁float s",
+      "▁un authorized",
+      "八 百",
+      "张 玉",
+      "意 料",
+      "nD ev",
+      "nDe v",
+      "n Dev",
+      "vir t",
+      "vi rt",
+      "v irt",
+      "相关 信息",
+      "▁laws uits",
+      "▁lawsuit s",
+      "▁ 博",
+      "容 忍",
+      "就 将",
+      "满 天",
+      "该 剧",
+      "黑 夜",
+      "SU S",
+      "S US",
+      "nf l",
+      "n fl",
+      "▁т о",
+      "▁ то",
+      "可能 需要",
+      "家庭 聚会",
+      "\\\\_ \\\\_",
+      "▁pop ped",
+      "▁po pped",
+      "nPr imary",
+      "nPrim ary",
+      "n Primary",
+      "▁graph ql",
+      "▁ graphql",
+      "▁viol ate",
+      "▁vi olate",
+      "▁serv ants",
+      "▁servant s",
+      "▁Dem ocracy",
+      "▁Democr acy",
+      "▁dis abling",
+      "▁vertebr ate",
+      "▁accred itation",
+      "▁ 艳",
+      "你 把",
+      "实 景",
+      "]] )",
+      "] ])",
+      "▁平 面",
+      "▁ 平面",
+      "Sh ot",
+      "S hot",
+      "ob le",
+      "obl e",
+      "o ble",
+      "at ase",
+      "ata se",
+      "atas e",
+      "man ufact",
+      "▁Under ground",
+      "▁Und erground",
+      "裂 片",
+      "Tab s",
+      "Ta bs",
+      "T abs",
+      "id ades",
+      "ida des",
+      "idad es",
+      "idade s",
+      "缩 写",
+      "▁P BS",
+      "▁PB S",
+      "集成 电路",
+      "IG ENCE",
+      "▁j ealous",
+      "▁Mc Carthy",
+      "k j",
+      "▁s sl",
+      "▁ss l",
+      "▁ ssl",
+      "ugg le",
+      "▁Y ard",
+      "▁Ya rd",
+      "床 头",
+      "晚 了",
+      "某 人",
+      "选 秀",
+      "城市 化",
+      "守门 员",
+      "过渡 期",
+      "go ods",
+      "good s",
+      "▁NE ED",
+      "Develop er",
+      "志 军",
+      "Fe b",
+      "F eb",
+      "描述 和",
+      "最佳 的",
+      "▁W en",
+      "▁We n",
+      "fa sta",
+      "fast a",
+      "fas ta",
+      "f asta",
+      "▁pool ed",
+      "th inking",
+      "think ing",
+      "thin king",
+      "▁hy brids",
+      "▁hybrid s",
+      "团 伙",
+      "手 下",
+      "杂 物",
+      "爱 与",
+      "无限 的",
+      "是 利用",
+      "分享 一下",
+      "olog ne",
+      "▁st ole",
+      "▁sto le",
+      "order By",
+      "▁sh aded",
+      "▁sha ded",
+      "▁shade d",
+      "▁sl owed",
+      "▁slow ed",
+      "▁amb ulance",
+      "他 从",
+      "实施 的",
+      "已 超过",
+      "所学 校",
+      "所 学校",
+      "竞争 优势",
+      "str ipe",
+      "strip e",
+      "stri pe",
+      "▁Sh ows",
+      "▁Show s",
+      "▁p oured",
+      "▁po ured",
+      "▁pour ed",
+      "▁pou red",
+      "▁Pra irie",
+      "▁go ats",
+      "▁goat s",
+      "▁novel ty",
+      "▁spectro meter",
+      "▁spectrom eter",
+      "▁ ا",
+      "日 式",
+      "甲 级",
+      "阿 克",
+      "的大 量",
+      "的 大量",
+      "ch unks",
+      "chunk s",
+      "nOr der",
+      "n Order",
+      "给出解题 步骤",
+      "压 实",
+      "年 鉴",
+      "你要 是",
+      "你 要是",
+      "地面 积",
+      "地 面积",
+      ";; ;;",
+      "▁K ane",
+      "▁Ka ne",
+      "▁Kan e",
+      "led ger",
+      "ledge r",
+      "▁lib erty",
+      "▁libert y",
+      "▁Yugoslav ia",
+      "未发现被执行人 有可供执行的财产",
+      "未发现被执行人有 可供执行的财产",
+      "未 发现被执行人有可供执行的财产",
+      "↓ \\",
+      "▁ 陶",
+      "节 气",
+      "质 押",
+      "仅 次于",
+      "Bear er",
+      "▁ 九",
+      "勃 勃",
+      "开 裂",
+      "交换 机",
+      "▁A SS",
+      "▁AS S",
+      "▁ ASS",
+      "▁R ack",
+      "▁Ra ck",
+      "▁Rac k",
+      "▁perox idase",
+      "效果 的",
+      "ab ine",
+      "abin e",
+      "abi ne",
+      "nb egin",
+      "nbe gin",
+      "n begin",
+      "nF ellows",
+      "▁leg ends",
+      "▁legend s",
+      "合 数",
+      "村 位于",
+      "重要 讲话",
+      "Dev ices",
+      "Device s",
+      "▁de ferred",
+      "▁defer red",
+      "▁withd rew",
+      "适应 性",
+      ":\" )\\",
+      ": \")\\",
+      "nTe ams",
+      "nTeam s",
+      "▁B east",
+      "▁Be ast",
+      "▁Bea st",
+      "ole rant",
+      "oler ant",
+      "pl ified",
+      "剪 刀",
+      "寿 司",
+      "且 原告",
+      "也 因此",
+      "分别 占",
+      "Si nk",
+      "Sin k",
+      "S ink",
+      "ec ed",
+      "e ced",
+      "▁C Ds",
+      "▁CD s",
+      "怎么说 呢",
+      "AR RAY",
+      "da vid",
+      "d avid",
+      "ject ed",
+      "j ected",
+      "get attr",
+      "▁sw immer",
+      "▁swim mer",
+      "Un ix",
+      "U nix",
+      "We bsite",
+      "Web site",
+      "▁r ation",
+      "▁rat ion",
+      "▁ra tion",
+      "▁ratio n",
+      "▁ ration",
+      "pre serve",
+      "pres erve",
+      "▁home owner",
+      "是 出于",
+      "▁ 吹风机",
+      "▁Does n",
+      "Data Set",
+      "us ement",
+      "use ment",
+      "交 于",
+      "茉 莉",
+      "nSt atus",
+      "nStat us",
+      "n Status",
+      "▁premium s",
+      "政 委",
+      "比 对",
+      "hi r",
+      "h ir",
+      "▁ 世界",
+      "离婚 后",
+      "离 婚后",
+      "jo rd",
+      "j ord",
+      "NAME SPACE",
+      "▁Dis order",
+      "▁illegal ly",
+      "▁illeg ally",
+      "Y C",
+      "风 尘",
+      "SU RE",
+      "S URE",
+      "et us",
+      "严重 后果",
+      "st alk",
+      "s talk",
+      "you ng",
+      "yo ung",
+      "▁Fel ix",
+      "▁down side",
+      "▁downs ide",
+      "▁graft ing",
+      "▁viol ating",
+      "没 过",
+      "▁V L",
+      "▁ VL",
+      "中 输入",
+      "就是 说",
+      "是一 座",
+      "是 一座",
+      "能做 到",
+      "能 做到",
+      "黑色 素",
+      "黑 色素",
+      "▁j og",
+      "▁jo g",
+      "投资 管理",
+      "▁har bour",
+      "▁harb our",
+      "▁prolif ic",
+      "▁prol ific",
+      "保 税",
+      "支付 的",
+      "Ma de",
+      "Mad e",
+      "M ade",
+      "nr oot",
+      "n root",
+      "vo cab",
+      "voc ab",
+      "▁ 柳",
+      "受 过",
+      "小 腿",
+      "成 才",
+      "手 写",
+      "攻 读",
+      "野 蛮",
+      "是 全球",
+      "第五 届",
+      "▁P IL",
+      "▁PI L",
+      "优惠 政策",
+      "调解 无效",
+      "▁P une",
+      "▁Pu ne",
+      "▁Pun e",
+      "▁t art",
+      "▁ta rt",
+      "▁tar t",
+      "nn umber",
+      "nnum ber",
+      "n number",
+      "▁Bal tic",
+      "▁Balt ic",
+      "一 模",
+      "只 用",
+      "战 神",
+      "马 蹄",
+      "的话 语",
+      "的 话语",
+      "cu lt",
+      "c ult",
+      "do rf",
+      "dor f",
+      "d orf",
+      "nc omp",
+      "ncom p",
+      "nco mp",
+      "n comp",
+      "▁ax on",
+      "▁ axon",
+      "▁} }>\\",
+      "▁}} >\\",
+      "二〇一八年 四月",
+      "▁qu arant",
+      "▁Car negie",
+      "产 区",
+      "山 谷",
+      "只 知道",
+      "nW as",
+      "▁p ari",
+      "▁par i",
+      "▁pa ri",
+      "▁t ally",
+      "▁tal ly",
+      "▁tall y",
+      "▁uter us",
+      "dis count",
+      "disc ount",
+      "sol utions",
+      "solution s",
+      "s olutions",
+      "▁car riage",
+      "▁command ed",
+      "▁test icular",
+      "做 强",
+      "出 新",
+      "同 桌",
+      "张 小",
+      "▁ 标准",
+      "如果 我",
+      "蓝色 的",
+      "蓝 色的",
+      "得出 结论",
+      "nFr iday",
+      "n Friday",
+      "y b",
+      "协 力",
+      "叛 逆",
+      "把 控",
+      "贮 存",
+      "马 的",
+      "驾 校",
+      "垂直 于",
+      "W ant",
+      "▁b ead",
+      "▁be ad",
+      "▁T issue",
+      "有限责任 公司于",
+      "有限责任公司 于",
+      "▁miss iles",
+      "▁missile s",
+      "大 白",
+      "捍 卫",
+      "教育 学",
+      "亿元 人民币",
+      "亿 元人民币",
+      "局 党组成员",
+      "fac ial",
+      "f acial",
+      "▁ant iqu",
+      "▁anti qu",
+      "> ;",
+      "低 音",
+      "求 证",
+      "ri to",
+      "rit o",
+      "r ito",
+      "缓期 二年",
+      "nR ural",
+      "▁em bro",
+      "▁emb ro",
+      "▁embr o",
+      "▁Off set",
+      "▁ Offset",
+      "▁Camero on",
+      "中 度",
+      "或 少",
+      "考 了",
+      "() ]",
+      "( )]",
+      "ra tic",
+      "rat ic",
+      "r atic",
+      "▁Sh aring",
+      "▁Sha ring",
+      "▁Shar ing",
+      "▁Power Shell",
+      "▁ 정",
+      "多年 生",
+      "多 年生",
+      "nf il",
+      "nfi l",
+      "n fil",
+      "附属 医院",
+      "▁d ams",
+      "▁dam s",
+      "▁da ms",
+      "at osis",
+      "ato sis",
+      "atos is",
+      "▁Base l",
+      "▁Bas el",
+      "▁Ba sel",
+      "▁ax onal",
+      "▁axon al",
+      "▁hyp ogly",
+      "▁ind uctive",
+      "▁induct ive",
+      "▁xen ograft",
+      "▁re structuring",
+      "事 事",
+      "兰 特",
+      "脉 络",
+      "re k",
+      "r ek",
+      "为 保证",
+      "泸州 市",
+      "泸 州市",
+      "这 四个",
+      "re in",
+      "rei n",
+      "r ein",
+      "内容 丰富",
+      "法 刑初字第",
+      "▁den ial",
+      "▁suit ability",
+      "上班 族",
+      "张家 界",
+      "拥有 了",
+      "拥 有了",
+      "rel lo",
+      "rell o",
+      "r ello",
+      "▁LED s",
+      "▁LE Ds",
+      "[ <",
+      "分配 律",
+      "生产 工艺",
+      "▁B ark",
+      "▁Bar k",
+      "▁Ba rk",
+      "时间 复杂度",
+      "▁Do ctrine",
+      "▁Sun shine",
+      "▁Rest oration",
+      "▁In structions",
+      "▁Instruction s",
+      "归 于",
+      "毛 发",
+      "WH O",
+      "W HO",
+      "的 季节",
+      "请 写出",
+      "人际 关系",
+      "▁Aud it",
+      "▁Au dit",
+      "▁Audi t",
+      "▁ Audit",
+      "▁gr ids",
+      "▁grid s",
+      "▁gri ds",
+      "▁sh out",
+      "▁Ne umann",
+      "itution al",
+      "▁assign ing",
+      "cobacter ium",
+      "倘 若",
+      "天 门",
+      "太 少",
+      "悄 然",
+      "海 峰",
+      "nc p",
+      "n cp",
+      "ua nt",
+      "uan t",
+      "u ant",
+      "yl ic",
+      "y lic",
+      "or gia",
+      "org ia",
+      "V ehicle",
+      "ier ship",
+      "iers hip",
+      "i ership",
+      "▁check box",
+      "▁ checkbox",
+      "▁sk etches",
+      "▁sketch es",
+      "▁school ing",
+      "测 算",
+      "相 册",
+      "记 为",
+      "\"\" >",
+      "\" \">",
+      "▁ 如何",
+      "在 台湾",
+      "ie ra",
+      "ier a",
+      "i era",
+      "两边 同时",
+      "本院 审查",
+      "de bian",
+      "deb ian",
+      "ow ards",
+      "o wards",
+      "农村经济 总收入",
+      "decor ator",
+      "▁S ynthesis",
+      "▁Synt hesis",
+      "▁exerc ising",
+      "▁c x",
+      "▁ cx",
+      "展开 了",
+      "展 开了",
+      "(\\' _",
+      "(\\ '_",
+      "( \\'_",
+      "提供 不出",
+      "英语 学习",
+      "▁c ellar",
+      "▁cell ar",
+      "▁cel lar",
+      "了 么",
+      "性质 的",
+      "性 质的",
+      "▁W an",
+      "▁Wa n",
+      "▁an ts",
+      "▁ant s",
+      "▁ ants",
+      "经常 居住地",
+      "上 万",
+      "年 期",
+      "每 条",
+      "▁Y o",
+      "ec ker",
+      "eck er",
+      "user Data",
+      "实 测",
+      "译 文",
+      ")\\ ',",
+      ")\\' ,",
+      ") \\',",
+      "▁[ \\\\",
+      "▁[\\ \\",
+      "▁ [\\\\",
+      "▁r pm",
+      "▁ rpm",
+      "plan et",
+      "plane t",
+      "to List",
+      "end ance",
+      "enda nce",
+      "二〇一八年 三月",
+      "▁ Č",
+      "元 宝",
+      "婚生 子",
+      "婚 生子",
+      "眼前 的",
+      "眼 前的",
+      "▁C CR",
+      "▁CC R",
+      "▁Aut onomous",
+      "▁work sheets",
+      "▁works heets",
+      "▁worksheet s",
+      "英 超",
+      "主办 方",
+      "gi ve",
+      "g ive",
+      "▁c ured",
+      "▁cur ed",
+      "▁cure d",
+      "▁cu red",
+      "夫妻 共同财产",
+      "nN othing",
+      "nNo thing",
+      "nNot hing",
+      "n Nothing",
+      "▁Http Client",
+      "▁ HttpClient",
+      "ocom patibility",
+      "o compatibility",
+      "N W",
+      "▁ 初",
+      "边 际",
+      "橄榄 油",
+      "▁B erm",
+      "▁Be rm",
+      "▁Ber m",
+      "lin alg",
+      "l inalg",
+      "▁fl ats",
+      "▁flat s",
+      "▁myel oma",
+      "▁ch olinergic",
+      "停 电",
+      "纯 正",
+      "针 刺",
+      "▁P aram",
+      "▁Par am",
+      "▁Pa ram",
+      "▁Para m",
+      "▁ Param",
+      "▁Pa scal",
+      "▁Pas cal",
+      "▁in variants",
+      "▁invari ants",
+      "▁invariant s",
+      "理 发",
+      "腹 黑",
+      "颇 为",
+      "任务 的",
+      "▁Hy undai",
+      "tassert Equals",
+      "t assertEquals",
+      "反 响",
+      "扇 形",
+      "汽车 的",
+      "汽 车的",
+      "In voice",
+      "Inv oice",
+      "debug ger",
+      "可选 择",
+      "可 选择",
+      "Ro ad",
+      "R oad",
+      "我们 知道",
+      "执行 法院",
+      "lor ds",
+      "lord s",
+      "l ords",
+      "培养 学生的",
+      "培养学生 的",
+      "指定 期限内",
+      "▁rhet oric",
+      "引 力",
+      "白 金",
+      "黄 冈",
+      "it he",
+      "ith e",
+      "i the",
+      "与时俱 进",
+      "▁st icker",
+      "▁stick er",
+      "} ÷",
+      "出 头",
+      "苦 瓜",
+      "边 上",
+      "年代 初",
+      "采用 的是",
+      "采用的 是",
+      "采 用的是",
+      "▁qu in",
+      "▁qui n",
+      "▁ quin",
+      "▁huge ly",
+      "▁hug ely",
+      "滋 补",
+      "相 撞",
+      "NL P",
+      "N LP",
+      "▁ 空调",
+      "发出 了",
+      "发 出了",
+      "我 怎么",
+      "nA ug",
+      "n Aug",
+      "▁T iO",
+      "▁Ti O",
+      "▁ TiO",
+      "专家 学者",
+      "emor rh",
+      "osa urs",
+      "▁use Effect",
+      "▁ useEffect",
+      "想 在",
+      "▁g y",
+      "▁ gy",
+      "最 容易",
+      "货物 运输",
+      "相等的 实数根",
+      "▁Pep per",
+      "▁th icker",
+      "▁thick er",
+      "pro vision",
+      "prov ision",
+      "▁Begin ning",
+      "申请执行人 向本院申请执行",
+      "不 小",
+      "视 察",
+      "第三百 零",
+      "▁Jud y",
+      "▁Ju dy",
+      "▁mel ody",
+      "im mers",
+      "immer s",
+      "imm ers",
+      "▁Publish ers",
+      "▁Publisher s",
+      "拳 击",
+      "管 线",
+      "赞 誉",
+      "▁ 城市",
+      "便利 店",
+      "▁poss essed",
+      "▁possess ed",
+      "和 无",
+      "密 的",
+      "ec an",
+      "eca n",
+      "e can",
+      "▁un do",
+      "▁u ndo",
+      "▁und o",
+      "▁ undo",
+      "▁V ERSION",
+      "▁ VERSION",
+      "▁good bye",
+      "援 朝",
+      "送 去",
+      "为 提高",
+      "▁T ou",
+      "▁To u",
+      "利害 关系",
+      "总有 效率",
+      "总 有效率",
+      "is cus",
+      "isc us",
+      "▁Myst ery",
+      "▁Mys tery",
+      "教 委",
+      "衣 橱",
+      "赞 扬",
+      "Com bo",
+      "Comb o",
+      "C ombo",
+      "▁Om aha",
+      "mult ipart",
+      "multi part",
+      "供 求",
+      "前 身",
+      "均 无",
+      "有 明显",
+      "比较 少",
+      "比 较少",
+      "验证 码",
+      "nu ts",
+      "nut s",
+      "n uts",
+      "tu ce",
+      "t uce",
+      "▁d ug",
+      "▁du g",
+      "▁CH ANGE",
+      "▁ CHANGE",
+      "▁aw aken",
+      "▁awake n",
+      "▁awa ken",
+      "▁s iding",
+      "▁si ding",
+      "▁sid ing",
+      "ur rection",
+      "urre ction",
+      "▁sub species",
+      "吃 药",
+      "征 税",
+      "最 受",
+      "▁Ad apt",
+      "▁Ada pt",
+      "民事行为 能力",
+      "▁Ale mbic",
+      "▁IN DIRECT",
+      "同时 对",
+      "但是 由于",
+      "但 是由于",
+      "nId ent",
+      "n Ident",
+      "▁k ings",
+      "▁kin gs",
+      "▁king s",
+      "▁in verter",
+      "▁invert er",
+      "▁Hunt ington",
+      "▁Hunting ton",
+      "* \"",
+      "岁 以下",
+      "试验 区",
+      "▁R PG",
+      "▁RP G",
+      "nS ervice",
+      "n Service",
+      "Def initions",
+      "Definition s",
+      "她 会",
+      "流 速",
+      "Te l",
+      "T el",
+      "or an",
+      "ora n",
+      "o ran",
+      "he mer",
+      "hem er",
+      "heme r",
+      "人口与 计划生育",
+      "▁mos quit",
+      "$ \"",
+      "之 歌",
+      "写 得",
+      "此 举",
+      "衡 水",
+      "进 水",
+      "全 覆盖",
+      "而 导致",
+      "um ers",
+      "ume rs",
+      "umer s",
+      "u mers",
+      "南县 人民法院",
+      "南 县人民法院",
+      "ryst als",
+      "rystal s",
+      "人 头",
+      "旧 的",
+      "破 了",
+      "中国 国家",
+      "为自己 的",
+      "为 自己的",
+      "Page r",
+      "Pa ger",
+      "Pag er",
+      "P ager",
+      "er din",
+      "erd in",
+      "priv acy",
+      "p rivacy",
+      "▁simpl ifying",
+      "▁simplify ing",
+      "景 气",
+      "甲 醇",
+      "EE P",
+      "E EP",
+      "湖州 市",
+      "湖 州市",
+      "生产 厂家",
+      "dr agon",
+      "drag on",
+      "▁m ater",
+      "▁mat er",
+      "▁ma ter",
+      "▁mate r",
+      "▁wh oles",
+      "▁who les",
+      "▁whole s",
+      "% ^",
+      "预 估",
+      "是否 是",
+      "wo lf",
+      "w olf",
+      "高级 教师",
+      "▁nan oc",
+      "▁na noc",
+      "▁nano c",
+      "▁over see",
+      "▁overs ee",
+      "▁overse e",
+      "大 展",
+      "an kar",
+      "ank ar",
+      "anka r",
+      "reg exp",
+      "regex p",
+      "▁AT LAS",
+      "nQu estions",
+      "nQuestion s",
+      "▁app raisal",
+      "▁kind ergarten",
+      "探 寻",
+      "极 易",
+      "格 栅",
+      "特 么",
+      "诗 句",
+      "BO OK",
+      "▁box er",
+      "nRe move",
+      "nRem ove",
+      "n Remove",
+      "▁bundle d",
+      "▁bund led",
+      "滚 滚",
+      "趋 向",
+      "MC A",
+      "M CA",
+      "vi an",
+      "via n",
+      "v ian",
+      "▁L ima",
+      "▁Lim a",
+      "▁Li ma",
+      "▁contest ants",
+      "卫 计",
+      "台 南",
+      "就 从",
+      "浸 润",
+      "腿 部",
+      "的 母亲",
+      "▁ч то",
+      "即可 开始游戏",
+      "▁bl inds",
+      "▁blind s",
+      "云 山",
+      "协 和",
+      "的 圆",
+      "vo n",
+      "v on",
+      "的商 业",
+      "的 商业",
+      "车 服务",
+      "非常 高",
+      "▁R PM",
+      "▁RP M",
+      "▁b ids",
+      "▁bi ds",
+      "▁bid s",
+      "il iate",
+      "ili ate",
+      "ilia te",
+      "▁b east",
+      "▁be ast",
+      "xim ation",
+      "x imation",
+      "nH istoric",
+      "Buff eredReader",
+      "之 久",
+      "一次 次",
+      "却 发现",
+      "商丘 市",
+      "▁H ed",
+      "▁He d",
+      "犯 诈骗罪",
+      "E arth",
+      "to ast",
+      "ud ing",
+      "udi ng",
+      "u ding",
+      "Dis covery",
+      "Disc overy",
+      "▁anne aling",
+      "何 某",
+      "先 前",
+      "▁< >",
+      "▁ <>",
+      "nG od",
+      "nGo d",
+      "n God",
+      "sa nd",
+      "san d",
+      "s and",
+      "tp ath",
+      "t path",
+      "▁ev ap",
+      "An chor",
+      "就 餐",
+      "雷 霆",
+      "*) \\",
+      "* )\\",
+      "ot i",
+      "o ti",
+      "为 指导",
+      "在 企业",
+      "证明 原",
+      "sl ots",
+      "slot s",
+      "nP rivate",
+      "nPr ivate",
+      "n Private",
+      "第七十八条 之规定",
+      "第七十八 条之规定",
+      "▁Prov ides",
+      "▁Provide s",
+      "▁Block chain",
+      "▁percent ile",
+      "和 为",
+      "灵 芝",
+      "魅 族",
+      "也有 人",
+      "也 有人",
+      "转化 成",
+      "转 化成",
+      "nDet ermine",
+      "v f",
+      "春 华",
+      "罐 头",
+      "在 现代",
+      "▁s dk",
+      "▁sd k",
+      "▁ sdk",
+      "▁v id",
+      "▁vi d",
+      "▁ vid",
+      "▁b iking",
+      "▁bi king",
+      "▁bik ing",
+      "等 号",
+      "闪 闪",
+      "SP A",
+      "S PA",
+      "的原 理",
+      "的 原理",
+      "过 一个",
+      "教学 改革",
+      "dis on",
+      "di son",
+      "d ison",
+      "▁M its",
+      "▁Mi ts",
+      "▁Mit s",
+      "▁act u",
+      "▁ac tu",
+      "并支付 利息",
+      "并 支付利息",
+      "轴对称 图形",
+      "cor pus",
+      "corp us",
+      "▁hydro chlor",
+      "重 温",
+      "本文 以",
+      "不同 程度",
+      "im uth",
+      "▁Raf ael",
+      "▁He ating",
+      "▁Heat ing",
+      "▁orbit als",
+      "▁orbital s",
+      "stand alone",
+      "▁intens ities",
+      "育 种",
+      "该 产品",
+      "那么 大",
+      "IN ET",
+      "INE T",
+      "I NET",
+      "ps in",
+      "psi n",
+      "p sin",
+      "Pa int",
+      "P aint",
+      "▁T weet",
+      "▁v oucher",
+      "YL E",
+      "Y LE",
+      "ex as",
+      "实实在 在",
+      "▁St uff",
+      "nInd eed",
+      "▁Tim ber",
+      "▁Ti mber",
+      "▁on site",
+      "▁Print ing",
+      "主 意",
+      "你 别",
+      "▁ 需要",
+      "工作 量",
+      "Cl ub",
+      "上述 证据",
+      "▁H awks",
+      "▁Haw ks",
+      "▁Mac ro",
+      "▁Ma cro",
+      "▁ Macro",
+      "▁Graph ic",
+      "▁eight een",
+      "申请执行人可 向本院申请恢复执行",
+      "当 了",
+      "不 重要",
+      "推动 了",
+      "推 动了",
+      "的情 感",
+      "的 情感",
+      "达成 的",
+      "达 成的",
+      "ab an",
+      "aba n",
+      "a ban",
+      "UR ITY",
+      "URI TY",
+      "▁Lo ans",
+      "▁Loan s",
+      "gr avity",
+      "grav ity",
+      "▁im print",
+      "▁imp rint",
+      "▁P recision",
+      "▁Prec ision",
+      "▁bi odegrad",
+      "▁Depend ency",
+      "▁ Dependency",
+      "天 安",
+      "应 收",
+      "Te n",
+      "T en",
+      "经 原告",
+      "ir en",
+      "ire n",
+      "i ren",
+      "临床 表现",
+      "Pop en",
+      "Po pen",
+      "P open",
+      "nP ain",
+      "▁c heat",
+      "▁ch eat",
+      "▁che at",
+      "aut ions",
+      "a utions",
+      "▁sl udge",
+      "▁must ard",
+      "▁Ori ental",
+      "▁elect rom",
+      "▁electro m",
+      "▁electr om",
+      "IP P",
+      "I PP",
+      "新 陈代谢",
+      "方法 采用",
+      "we eks",
+      "week s",
+      "w eeks",
+      "▁O man",
+      "▁Om an",
+      "▁dr ains",
+      "▁drain s",
+      "pro jection",
+      "project ion",
+      "proj ection",
+      "一 旁",
+      "所 含",
+      "生 源",
+      "退 化",
+      "不 接受",
+      "手机 号",
+      "的可 以",
+      "的 可以",
+      "ho ff",
+      "h off",
+      "▁ce ase",
+      "▁box ers",
+      "▁boxer s",
+      "▁b idding",
+      "▁bid ding",
+      "pot ential",
+      "减 退",
+      "靠 自己",
+      "Win d",
+      "Wi nd",
+      "W ind",
+      "al us",
+      "alu s",
+      "▁演员 表",
+      "TH ING",
+      "ét ico",
+      "▁R ear",
+      "▁Re ar",
+      "▁dir name",
+      "▁ dirname",
+      "Anal ytics",
+      "▁broad caster",
+      "▁broadcast er",
+      "灭 亡",
+      "让 其",
+      "yl a",
+      "y la",
+      "还有 个",
+      "还 有个",
+      "per g",
+      "pe rg",
+      "p erg",
+      "▁A lf",
+      "▁Al f",
+      "ite space",
+      "ites pace",
+      "二〇一四年 十一月",
+      "ym a",
+      "y ma",
+      "▁b isc",
+      "▁bi sc",
+      "▁bis c",
+      "▁t int",
+      "▁ti nt",
+      "▁tin t",
+      "▁ tint",
+      "▁S exual",
+      "▁Sex ual",
+      "iv y",
+      "i vy",
+      "多 领域",
+      "▁D ip",
+      "▁Di p",
+      "or ian",
+      "ori an",
+      "oria n",
+      "o rian",
+      "▁B alk",
+      "▁Bal k",
+      "▁s ind",
+      "▁sin d",
+      "▁si nd",
+      "Com bin",
+      "Comb in",
+      "ST ATIC",
+      "STAT IC",
+      "▁Pl aying",
+      "▁Play ing",
+      "▁Mar riage",
+      "▁imp lying",
+      "▁impl ying",
+      "▁imply ing",
+      "▁scream ing",
+      "Pers istence",
+      "保 费",
+      "升 值",
+      "晴 天",
+      "集 镇",
+      "与 环境",
+      "发明 专利",
+      "年轻 人的",
+      "年轻人 的",
+      "ou los",
+      "oul os",
+      "o ulos",
+      "▁re ps",
+      "▁r eps",
+      "▁rep s",
+      "xi n",
+      "x in",
+      "不 考虑",
+      "并处 没收",
+      "技术 研发",
+      "短篇 小说",
+      "短 篇小说",
+      "Pos ted",
+      "Post ed",
+      "Po sted",
+      "est ation",
+      "esta tion",
+      "e station",
+      "▁pl astics",
+      "▁plastic s",
+      "▁fil aments",
+      "▁filament s",
+      "▁postpon ed",
+      "▁sung lasses",
+      "学 堂",
+      "em u",
+      "e mu",
+      "的 画面",
+      "▁d it",
+      "▁di t",
+      "▁ dit",
+      "▁w olf",
+      "▁wo lf",
+      "▁ wolf",
+      "▁V ijay",
+      "▁Vi jay",
+      "江 河",
+      "虚 无",
+      "▁R ey",
+      "▁Re y",
+      "小心 翼翼",
+      "有 停车位",
+      "ba sis",
+      "bas is",
+      "b asis",
+      "Med ium",
+      "PACK AGE",
+      "增 效",
+      "魔 幻",
+      "使 学生",
+      "ak ar",
+      "aka r",
+      "a kar",
+      "nEx ec",
+      "n Exec",
+      "sh uffle",
+      "▁l inger",
+      "▁lin ger",
+      "▁ling er",
+      "▁ linger",
+      "案件受理费 人民币",
+      "DIRECT ORY",
+      "▁an atomic",
+      "在 京",
+      "该 笔",
+      "▁\" &",
+      "AN TS",
+      "ANT S",
+      "▁f ry",
+      "▁fr y",
+      "集体 经济",
+      "al pine",
+      "nH ouston",
+      "▁far ther",
+      "▁priorit ize",
+      "▁re pression",
+      "▁rep ression",
+      "▁repr ession",
+      "▁vibr ational",
+      "▁vibration al",
+      "▁eryth rocytes",
+      "▁erythrocyte s",
+      "致 癌",
+      "行 之",
+      "出不 穷",
+      "更加 的",
+      "死亡 的",
+      "or eal",
+      "ore al",
+      "o real",
+      "依法 可以从轻处罚",
+      "▁auto antib",
+      "红 线",
+      "jp a",
+      "j pa",
+      "dis s",
+      "di ss",
+      "d iss",
+      "ur ic",
+      "uri c",
+      "u ric",
+      "▁h ue",
+      "▁hu e",
+      "将 本案移送",
+      "data bases",
+      "dat abases",
+      "database s",
+      "datab ases",
+      "▁Leban ese",
+      "▁vac ancies",
+      "中国 队",
+      "新型 的",
+      "新 型的",
+      "yr in",
+      "y rin",
+      "基础 教育",
+      "▁mile age",
+      "▁String IO",
+      "Ident ification",
+      "发 觉",
+      "客 栈",
+      "票 务",
+      "▁G le",
+      "▁Gl e",
+      "▁L er",
+      "▁Le r",
+      "中国 电信",
+      "差不多 的",
+      "差不 多的",
+      "换 句话说",
+      "▁h eck",
+      "▁he ck",
+      "▁ heck",
+      "STR UCT",
+      "Pro duction",
+      "Product ion",
+      "Prod uction",
+      "Produ ction",
+      "P roduction",
+      "外 径",
+      "DF S",
+      "D FS",
+      "制造 的",
+      ")) ))",
+      "))) )",
+      ") )))",
+      "▁V ictory",
+      "▁Victor y",
+      "H G",
+      "偏 低",
+      "扩 充",
+      "派 对",
+      "go s",
+      "g os",
+      "▁N I",
+      "▁ NI",
+      "法律 顾问",
+      "随便 吃吃",
+      "二〇一六年 九",
+      "纪念 碑",
+      "if en",
+      "ife n",
+      "i fen",
+      "欢迎 大家",
+      "ator ics",
+      "ato rics",
+      "▁bar gain",
+      "▁barg ain",
+      "▁m acular",
+      "▁mac ular",
+      "▁ma cular",
+      "cor relation",
+      "▁over coming",
+      "exam ethasone",
+      "大 山",
+      "组 团",
+      "和 专业",
+      "年 北京",
+      "Ver y",
+      "V ery",
+      "▁M MA",
+      "▁MM A",
+      "Part ai",
+      "Inter op",
+      "▁Pro per",
+      "▁Pr oper",
+      "▁Prop er",
+      "▁Kn owing",
+      "▁Know ing",
+      "凡 事",
+      "并 可",
+      "身 亡",
+      "cc i",
+      "c ci",
+      "会让 你",
+      "会 让你",
+      "项目 部",
+      "Bo ost",
+      "EF ORE",
+      "▁H ank",
+      "▁Ha nk",
+      "▁Han k",
+      "中 选择一个",
+      "un ding",
+      "und ing",
+      "undi ng",
+      "op ically",
+      "opic ally",
+      "而 异",
+      "钻 井",
+      "你 已经",
+      "添加 到",
+      "▁a lot",
+      "▁al ot",
+      "▁pre jud",
+      "ist rative",
+      "istr ative",
+      "则 该",
+      "胜 任",
+      "永远 不会",
+      "▁f oci",
+      "▁foc i",
+      "▁fo ci",
+      "▁ 云中书城",
+      "▁人物 简介",
+      "▁Le one",
+      "▁Leon e",
+      "▁Leo ne",
+      "中国人民 大学",
+      "ed itable",
+      "edit able",
+      "edi table",
+      "▁Opt imal",
+      "▁Optim al",
+      "▁dil emma",
+      "▁dile mma",
+      "电子 科技有限公司",
+      "contract s",
+      "contr acts",
+      "▁Rad iation",
+      "执行情况 告知申请执行人",
+      "S u",
+      "姜 片",
+      "雷 电",
+      "\"\" ,",
+      "\" \",",
+      "▁ 修改",
+      "Sl ide",
+      "S lide",
+      "ir con",
+      "irc on",
+      "sp arse",
+      "s parse",
+      "▁fel ony",
+      "盗 版",
+      "到 哪里",
+      "我 曾经",
+      "农村 居民",
+      "ip zig",
+      "▁L oren",
+      "▁Lo ren",
+      "▁Lore n",
+      "▁Lor en",
+      "▁Gen esis",
+      "▁Gene sis",
+      "▁Ass ignment",
+      "▁Assign ment",
+      "▁ Assignment",
+      "三 期",
+      "代 办",
+      "新 乡",
+      "江 津",
+      "除 湿",
+      "和 创新",
+      "新建 的",
+      "新 建的",
+      "的好 坏",
+      "的 好坏",
+      "De ar",
+      "D ear",
+      "▁H MS",
+      "▁HM S",
+      "▁T AG",
+      "▁TA G",
+      "▁ TAG",
+      "▁ 处理器",
+      "ulf ide",
+      "▁Chron icle",
+      "▁Chronic le",
+      "保 值",
+      "得 过",
+      "误 导",
+      "n ée",
+      "预期 的",
+      "预 期的",
+      "经济 贸易",
+      "om ies",
+      "omi es",
+      "▁Al to",
+      "▁Alt o",
+      "▁P las",
+      "▁Pl as",
+      "Check box",
+      "▁ch ecker",
+      "▁check er",
+      "▁ checker",
+      "▁soy bean",
+      "▁aut osomal",
+      "伺 服",
+      "你 没",
+      "服 役",
+      "lo st",
+      "los t",
+      "l ost",
+      "不确定 性",
+      "检 诉刑诉",
+      "身心 健康",
+      "▁Cur ve",
+      "▁ Curve",
+      "▁Dev ils",
+      "▁Devil s",
+      "二〇一六年 七月",
+      "JSON Object",
+      "▁mat hematically",
+      "▁mathematical ly",
+      "▁mathematic ally",
+      "后 半",
+      "年 版",
+      "to ml",
+      "tom l",
+      "▁K od",
+      "▁Ko d",
+      "▁V on",
+      "▁Vo n",
+      "采取保全措施 后",
+      "▁cyl inders",
+      "▁cylinder s",
+      "▁cylind ers",
+      "▁th umbnail",
+      "▁ thumbnail",
+      "▁experiment ing",
+      "烟 雨",
+      "粉 的",
+      "镜 片",
+      "回家 的",
+      "回 家的",
+      "▁M ai",
+      "▁Ma i",
+      "▁s care",
+      "▁sc are",
+      "▁scar e",
+      "fill able",
+      "▁academ ia",
+      "本 校",
+      ")) }\\",
+      ") )}\\",
+      "ba th",
+      "bat h",
+      "b ath",
+      "▁M SN",
+      "▁MS N",
+      "▁S DS",
+      "▁SD S",
+      "nF ood",
+      "n Food",
+      "op han",
+      "oph an",
+      "o phan",
+      "▁T hin",
+      "▁Th in",
+      "▁W ick",
+      "▁Wi ck",
+      "住所地 福建省",
+      "▁cycl ase",
+      "制 服",
+      "是 原",
+      "汉 化",
+      "精 子",
+      "▁大 小",
+      "▁ 大小",
+      "np re",
+      "npr e",
+      "n pre",
+      "▁P ORT",
+      "▁PO RT",
+      "▁ PORT",
+      "人 去",
+      "但 因",
+      "号之 二",
+      "号 之二",
+      "我来 说",
+      "我 来说",
+      "un ner",
+      "unn er",
+      "▁Par ade",
+      "▁Parad e",
+      "▁Para de",
+      "▁Wall paper",
+      "▁contamin ants",
+      "乳 酸",
+      "王 俊",
+      "dd d",
+      "d dd",
+      "▁% )",
+      "▁ %)",
+      "到 医院",
+      "▁E va",
+      "▁Ev a",
+      "下列 说法",
+      "年 降水量",
+      "co dep",
+      "code p",
+      "cod ep",
+      "ip ynb",
+      "ipy nb",
+      "▁expend itures",
+      "▁expenditure s",
+      "▁ 沙",
+      "对 阵",
+      "汤 汁",
+      "up a",
+      "u pa",
+      "AT TR",
+      "ATT R",
+      "分式 方程",
+      "▁pe ac",
+      "▁商品 毛重",
+      "进行了 深入",
+      "Follow ing",
+      "F ollowing",
+      "▁ 次",
+      "交 办",
+      "Ch o",
+      "C ho",
+      "SA R",
+      "S AR",
+      "ies is",
+      "ie sis",
+      "i esis",
+      "▁sur pass",
+      "中 线",
+      "先 看",
+      "和 弦",
+      "而 无",
+      "公司 于",
+      "es to",
+      "est o",
+      "e sto",
+      "Mar gin",
+      "M argin",
+      "pat ches",
+      "patch es",
+      "p atches",
+      "sp inner",
+      "spin ner",
+      "▁f al",
+      "▁fa l",
+      "▁C lip",
+      "▁Cl ip",
+      "▁ Clip",
+      "▁emer ges",
+      "▁emerg es",
+      "▁emerge s",
+      "点 儿",
+      "▁对 于",
+      "▁ 对于",
+      "省 部级",
+      "nP ak",
+      "这个 游戏",
+      "cou pled",
+      "▁tar iffs",
+      "dis connect",
+      "▁Dis orders",
+      "▁Disorder s",
+      "▁prot ections",
+      "▁protect ions",
+      "▁protection s",
+      "他 了",
+      "日 落",
+      "活 化",
+      "保险 人",
+      "公里 处",
+      "各 领域",
+      "位置 关系",
+      "和 相应的",
+      "ne uro",
+      "neur o",
+      "▁N VIDIA",
+      "▁fist ula",
+      "▁decl aring",
+      "▁declar ing",
+      "分 別",
+      "获得 者",
+      "有效 数字",
+      "自己 喜欢",
+      "F OLDER",
+      "G allery",
+      "Re ading",
+      "Read ing",
+      "▁S andra",
+      "▁Sand ra",
+      "▁Sa ndra",
+      "> |",
+      "休 假",
+      "文 昌",
+      "波 斯",
+      "▁L U",
+      "▁ LU",
+      "周某 某",
+      "周 某某",
+      "在 生产",
+      "方向 盘",
+      "}\\\\ ;",
+      "} \\\\;",
+      "产品 设计",
+      "寻衅滋事 罪",
+      "eff icacy",
+      "改 建",
+      "渊 源",
+      "▁L anguages",
+      "▁Language s",
+      "当 归",
+      "放 宽",
+      "溪 镇",
+      ":' ,",
+      ": ',",
+      "创造 力",
+      "达成 了",
+      "达 成了",
+      "de struct",
+      "dest ruct",
+      "▁cell ulose",
+      "appro ximately",
+      "approx imately",
+      "是 集",
+      "焦作 市",
+      "予以 证明",
+      "公 报",
+      "升 起",
+      "几乎 是",
+      "co us",
+      "cou s",
+      "c ous",
+      "wa iting",
+      "wait ing",
+      "▁qu arry",
+      "sh ipping",
+      "ship ping",
+      "shi pping",
+      "▁L indsay",
+      "▁Lind say",
+      "▁annot ated",
+      "加 息",
+      "宣 扬",
+      "语 的",
+      "浓度 的",
+      "浓 度的",
+      ")( \\\\",
+      ") (\\\\",
+      "em br",
+      "emb r",
+      "sum m",
+      "su mm",
+      "s umm",
+      "get Key",
+      "▁TH REE",
+      "安 得",
+      "ha w",
+      "h aw",
+      "物业 类别",
+      "▁h erd",
+      "▁he rd",
+      "▁her d",
+      "Separ ator",
+      "nE lections",
+      "nEl ections",
+      "nElect ions",
+      "nElection s",
+      "▁inhal ation",
+      "二 轮",
+      "寒 假",
+      "打 法",
+      "断 面",
+      "▁F W",
+      "▁ FW",
+      "is in",
+      "isi n",
+      "i sin",
+      "nM uch",
+      "▁Un cle",
+      "▁Unc le",
+      "▁pair wise",
+      "▁D imensions",
+      "▁Dimension s",
+      "▁Dim ensions",
+      "▁correct ness",
+      "方 舟",
+      "来 个",
+      "j vm",
+      "从严 治",
+      "▁X ml",
+      "▁ Xml",
+      "自由 贸易",
+      "nP rem",
+      "nPre m",
+      "nPr em",
+      "n Prem",
+      "▁d ome",
+      "▁do me",
+      "▁dom e",
+      "▁b ucks",
+      "▁bu cks",
+      "▁buck s",
+      "▁ bucks",
+      "▁e jection",
+      "▁pre scriptions",
+      "▁pres criptions",
+      "▁prescription s",
+      "▁ 夜",
+      "充 斥",
+      "同 位",
+      "巴 彦",
+      "幸 好",
+      "我 与",
+      "胞 胎",
+      "也 已经",
+      "▁coc oa",
+      "第二百四十 条",
+      "IN TEGER",
+      "Ch ristian",
+      "Christ ian",
+      "▁gather ings",
+      "▁gathering s",
+      "▁config urable",
+      "体 贴",
+      "吞 噬",
+      "意外 的",
+      "意 外的",
+      "▁C KD",
+      "▁CK D",
+      "in form",
+      "inf orm",
+      "info rm",
+      "▁ord inal",
+      "入 睡",
+      "中国 人的",
+      "中国人 的",
+      "可能 会有",
+      "可能会 有",
+      "▁dis reg",
+      "▁favour ites",
+      "▁favourite s",
+      "推 断",
+      "秋 水",
+      "aa n",
+      "a an",
+      "这只 是",
+      "这 只是",
+      "▁Incre ase",
+      "发生法律效力 之日起",
+      "▁discrim inant",
+      "一 式",
+      "和 多",
+      "教育 活动",
+      "被告 辩称",
+      "达成 协议",
+      "ne tty",
+      "net ty",
+      "nett y",
+      "▁商品 产地",
+      "曾 是",
+      "黄 土",
+      "ст а",
+      "с та",
+      "▁W B",
+      "▁ WB",
+      "又 因为",
+      "家常 菜",
+      "IA LS",
+      "IAL S",
+      "I ALS",
+      "产品 尺寸",
+      "zi lla",
+      "z illa",
+      "kin ase",
+      "▁se als",
+      "▁sea ls",
+      "▁seal s",
+      "▁ref using",
+      "太 白",
+      "很 像",
+      "CO ME",
+      "COM E",
+      "C OME",
+      "{} )\\",
+      "{}) \\",
+      "{ })\\",
+      "▁contract ions",
+      "▁contr actions",
+      "▁contraction s",
+      "▁osteo arthritis",
+      "别 克",
+      "战 友",
+      "抚 顺",
+      "背 诵",
+      "在 经济",
+      "ob ac",
+      "oba c",
+      "o bac",
+      "od ers",
+      "ode rs",
+      "oder s",
+      "o ders",
+      "Vis ion",
+      "V ision",
+      "bus ter",
+      "bu ster",
+      "b uster",
+      "▁blood y",
+      "▁blo ody",
+      "▁pay able",
+      "▁hol omorphic",
+      "很 想",
+      "言 行",
+      "nH op",
+      "n Hop",
+      "▁B ic",
+      "▁Bi c",
+      "数据 进行",
+      "之 神",
+      "公 会",
+      "利 物",
+      "危险 的",
+      "W ould",
+      "nL ink",
+      "nLin k",
+      "n Link",
+      "▁ro ast",
+      "rect angle",
+      "▁tack ling",
+      "马 尾",
+      "的 研发",
+      "nM ap",
+      "nMa p",
+      "n Map",
+      "og er",
+      "oge r",
+      "o ger",
+      "征求 意见",
+      "ETHO D",
+      "ETH OD",
+      "ef ile",
+      "e file",
+      "▁s inks",
+      "▁sin ks",
+      "▁sink s",
+      "▁bet ray",
+      "report ing",
+      "▁reg enerative",
+      "架空 历史",
+      "▁l oft",
+      "▁lo ft",
+      "查封 被申请人",
+      "fol ders",
+      "fold ers",
+      "folder s",
+      "▁char coal",
+      "▁prim ates",
+      "▁pri mates",
+      "▁primate s",
+      "▁deline ate",
+      "丞 相",
+      "五 种",
+      "开 销",
+      "杂志 社",
+      "矿 泉水",
+      "▁con ced",
+      "▁conc ed",
+      "▁conce d",
+      "end points",
+      "endpoint s",
+      "▁Dis ability",
+      "▁north western",
+      "▁northwest ern",
+      "传 记",
+      "DI V",
+      "D IV",
+      "in ster",
+      "ins ter",
+      "inst er",
+      "join ed",
+      "jo ined",
+      "j oined",
+      "le ting",
+      "let ing",
+      "l eting",
+      "▁sem inal",
+      "▁semi nal",
+      "▁semin al",
+      "nN ovember",
+      "nNov ember",
+      "n November",
+      "IA N",
+      "I AN",
+      "▁B cl",
+      "二审 法院",
+      "社交 媒体",
+      "Port al",
+      "P ortal",
+      "rid den",
+      "r idden",
+      "▁post operatively",
+      "▁postoperative ly",
+      "一 发",
+      "入 库",
+      "家 有",
+      "你 必须",
+      "教学 班",
+      "for th",
+      "fort h",
+      "f orth",
+      "▁R over",
+      "▁Ro ver",
+      "▁c anon",
+      "▁can on",
+      "▁ca non",
+      "▁aut opsy",
+      "▁auto psy",
+      "功 底",
+      "歌 剧",
+      "就 一直",
+      "new line",
+      "▁St rike",
+      "▁Str ike",
+      "▁Stri ke",
+      "▁for matter",
+      "▁form atter",
+      "▁format ter",
+      "▁ formatter",
+      "▁implicit ly",
+      "▁Research ers",
+      "═ ═",
+      "热 议",
+      "不到 位",
+      "不 到位",
+      "▁g az",
+      "▁ga z",
+      "项目 经理",
+      "nJ uly",
+      "nJul y",
+      "n July",
+      "X B",
+      "决 战",
+      "扁 舟",
+      "Ha m",
+      "H am",
+      "P ur",
+      "▁D EC",
+      "▁DE C",
+      "此 页无正文",
+      "土木 结构住房",
+      "▁Unc ategorized",
+      "取 胜",
+      "阻 抗",
+      "高 专",
+      "iv i",
+      "i vi",
+      "征 决字",
+      "起源 于",
+      "起 源于",
+      "WE VER",
+      "Run With",
+      "Ru nWith",
+      "▁who ever",
+      "有 为",
+      "罪 恶",
+      "耐 火",
+      "的 约定",
+      "劳务 派遣",
+      "nH ard",
+      "nHar d",
+      "nHa rd",
+      "n Hard",
+      "▁w ards",
+      "▁war ds",
+      "▁ward s",
+      "▁ wards",
+      "▁Bol ton",
+      "不 许",
+      "过 长",
+      ")} ,",
+      ") },",
+      "▁R ican",
+      "▁Ric an",
+      "▁Rica n",
+      "▁Ri can",
+      "▁axi om",
+      "▁Bull etin",
+      "▁Bullet in",
+      "▁air borne",
+      "感 人",
+      "我 叫",
+      "设 法",
+      "麦 当",
+      "vi p",
+      "v ip",
+      "▁k i",
+      "▁ ki",
+      "建筑 的",
+      "▁r ud",
+      "▁ru d",
+      "mt ime",
+      "m time",
+      "▁M oor",
+      "▁Mo or",
+      "▁qu il",
+      "▁qui l",
+      "▁ quil",
+      "ur ator",
+      "ura tor",
+      "▁y ummy",
+      "▁yum my",
+      "nprint f",
+      "n printf",
+      "▁caut ious",
+      "乳 房",
+      "劳 工",
+      "率 是",
+      "or h",
+      "o rh",
+      "▁т а",
+      "▁ та",
+      "了一 大",
+      "了 一大",
+      "人与 人",
+      "人 与人",
+      "验证 了",
+      "建设 工作",
+      "user id",
+      "use rid",
+      "▁bel ts",
+      "▁belt s",
+      "判处 无期徒刑",
+      "▁Det ective",
+      "绿 灯",
+      "it u",
+      "i tu",
+      "▁s r",
+      "▁ sr",
+      "▁T il",
+      "▁Ti l",
+      "Jo nes",
+      "Jon es",
+      "J ones",
+      "ph oton",
+      "photo n",
+      "phot on",
+      "char ged",
+      "charge d",
+      "charg ed",
+      "▁Al umni",
+      "▁Liter ary",
+      "▁Hol ocaust",
+      "第二百五十三 条的规定",
+      "曾 于",
+      "难 怪",
+      "我们 来",
+      "营业 收入",
+      "营销 策划",
+      "de nse",
+      "den se",
+      "dens e",
+      "d ense",
+      "th ese",
+      "the se",
+      "▁d awn",
+      "▁da wn",
+      "犯盗窃罪 被",
+      "▁ 比",
+      "推 导",
+      "at ura",
+      "atur a",
+      "atu ra",
+      "▁St ores",
+      "▁Store s",
+      "▁fire arm",
+      "▁Elastic search",
+      "三 楼",
+      "内 阁",
+      "ao l",
+      "a ol",
+      "冷 空气",
+      "nm in",
+      "n min",
+      "好 自己的",
+      "▁li tre",
+      "▁lit re",
+      "▁mid point",
+      "克 罗",
+      "官 僚",
+      "了大 量",
+      "了 大量",
+      "}} {{\\\\",
+      "}}{ {\\\\",
+      "}}{{ \\\\",
+      "Con version",
+      "Co nversion",
+      "Conv ersion",
+      "▁neurom uscular",
+      "▁ 成",
+      "口 镇",
+      "]) :",
+      "] ):",
+      "十三 年",
+      "十 三年",
+      "投入 使用",
+      "▁H its",
+      "▁Hi ts",
+      "▁Hit s",
+      "▁p odium",
+      "▁pod ium",
+      "▁tumorigen esis",
+      "三 者",
+      "临 汾",
+      "墓 志",
+      "洗 发",
+      "爱 民",
+      "脊 髓",
+      "▁j ungle",
+      "▁Vik ings",
+      "sub scriptions",
+      "subscription s",
+      "н я",
+      "他 为",
+      "甜 品",
+      "空气 质量",
+      "vid ia",
+      "v idia",
+      "▁f aux",
+      "▁fa ux",
+      "▁parad igms",
+      "▁paradigm s",
+      "该犯 在服刑期间确有悔改表现",
+      "该犯在服刑期间 确有悔改表现",
+      "妖 怪",
+      "绵 阳",
+      "正规 的",
+      "DE SC",
+      "DES C",
+      "水利 水电",
+      "研究 机构",
+      "部分 地区",
+      "▁ass ayed",
+      "▁assay ed",
+      "▁qual ifier",
+      "中 说",
+      "流 向",
+      "祖 宗",
+      "通 的",
+      "和 一些",
+      "LOW ED",
+      "SE MENT",
+      "SEM ENT",
+      "S EMENT",
+      "in file",
+      "inf ile",
+      "i nfile",
+      "▁buff et",
+      "▁ge odes",
+      "▁geo des",
+      "▁v ortex",
+      "▁vor tex",
+      "▁un usually",
+      "▁unus ually",
+      "▁unusual ly",
+      "利 民",
+      "念 念",
+      "美 感",
+      "前缀 和",
+      "ik en",
+      "ike n",
+      "i ken",
+      "▁Ber gen",
+      "▁Berg en",
+      "▁ref uses",
+      "▁refuse s",
+      "▁call able",
+      "▁di ameters",
+      "▁diameter s",
+      "例 行",
+      "含 金",
+      "应 激",
+      "粪 便",
+      "静 止",
+      "层 出不穷",
+      "铁路 运输",
+      "LI MIT",
+      "L IMIT",
+      "sof Ivy",
+      "▁Ch arity",
+      "▁Char ity",
+      "▁Sup ported",
+      "▁Support ed",
+      "▁ Supported",
+      "▁par amount",
+      "▁para mount",
+      "▁param ount",
+      "赣州 市",
+      "赣 州市",
+      "im id",
+      "imi d",
+      "i mid",
+      "pr ay",
+      "p ray",
+      "▁E ra",
+      "▁Er a",
+      "▁p add",
+      "▁pa dd",
+      "▁pad d",
+      "▁主要 原料",
+      "▁Suz uki",
+      "▁hum our",
+      "Db Context",
+      "▁line back",
+      "▁st umbled",
+      "▁ambig uity",
+      "▁Volks wagen",
+      "F x",
+      "将 与",
+      "qt y",
+      "q ty",
+      "在其 中",
+      "在 其中",
+      "某 负担",
+      "的 营养",
+      "这两 年",
+      "这 两年",
+      "er ta",
+      "ert a",
+      "Gr ade",
+      "Grad e",
+      "nTH IS",
+      "yp ass",
+      "y pass",
+      "▁gli oma",
+      "▁h unter",
+      "▁hun ter",
+      "▁hunt er",
+      "▁hu nter",
+      "酿 造",
+      "阳 明",
+      "▁C TL",
+      "▁CT L",
+      "整个 过程",
+      "不准许撤诉 的",
+      "nh eader",
+      "nhe ader",
+      "n header",
+      "▁E uclid",
+      "▁dro plets",
+      "▁drop lets",
+      "憧 憬",
+      "演 的",
+      "工作 了",
+      "工 作了",
+      "方 方面",
+      "次日 起",
+      "次 日起",
+      "全面 提升",
+      "▁制片 地区",
+      "get cwd",
+      "如果 把",
+      "想到 的",
+      "想 到的",
+      "手术 后",
+      "手 术后",
+      "半个 小时",
+      "半 个小时",
+      "ode nal",
+      "oden al",
+      "▁gang lia",
+      "▁condem ned",
+      "▁external ly",
+      "▁extern ally",
+      "灭 绝",
+      "铁 丝",
+      "专卖 店",
+      "代理 商",
+      "的中 心",
+      "的 中心",
+      "语言 学",
+      "Be nz",
+      "Ben z",
+      "B enz",
+      "解析式 为",
+      "nM ATER",
+      "▁rich ness",
+      "▁valid ators",
+      "▁validator s",
+      "▁ validators",
+      "一 览",
+      "bb c",
+      "b bc",
+      "ji a",
+      "j ia",
+      "我就 不",
+      "我 就不",
+      "来 表示",
+      "第二 审",
+      "第 二审",
+      "al on",
+      "alo n",
+      "a lon",
+      "培训 学校",
+      "▁g riev",
+      "▁gr iev",
+      "▁gri ev",
+      "nSen ior",
+      "▁Lis bon",
+      "▁flo ods",
+      "▁flood s",
+      "▁m A",
+      "▁ mA",
+      "是什么 呢",
+      "是 什么呢",
+      "被 认为是",
+      "EL ISA",
+      "fr ica",
+      "ch anger",
+      "chan ger",
+      "change r",
+      "chang er",
+      "▁Far mers",
+      "▁Farm ers",
+      "▁Farmer s",
+      "就读 于",
+      "▁F IN",
+      "▁FI N",
+      "▁s ont",
+      "▁so nt",
+      "▁son t",
+      "url open",
+      "▁tool kit",
+      "▁ toolkit",
+      "te chnical",
+      "techn ical",
+      "o C",
+      "伙 同",
+      "王 玉",
+      "PR I",
+      "P RI",
+      "è me",
+      "▁H Q",
+      "▁ HQ",
+      "▁W itch",
+      "▁Wit ch",
+      "▁Al bums",
+      "▁Alb ums",
+      "▁Album s",
+      "▁retard ation",
+      "御 史",
+      "热 潮",
+      "时 需要",
+      "War e",
+      "W are",
+      "ac yl",
+      "acy l",
+      "et in",
+      "eti n",
+      "e tin",
+      "值得 注意的是",
+      "广东省 佛山市",
+      "乘 除",
+      "余 个",
+      "分 包",
+      "就 职",
+      ":\" \\",
+      ": \"\\",
+      "世界 各国",
+      "公司 总经理",
+      "已 减半收取",
+      "▁Control s",
+      "▁Contr ols",
+      "▁ Controls",
+      "▁prison er",
+      "▁constit utive",
+      "nCong ratulations",
+      "▁ 贺",
+      "佳 作",
+      "TI M",
+      "T IM",
+      "▁R AF",
+      "▁RA F",
+      "▁v ue",
+      "▁ vue",
+      "尤为 重要",
+      "fr ied",
+      "f ried",
+      "nDiv ide",
+      "▁In form",
+      "▁Inf orm",
+      "▁Info rm",
+      "▁re agent",
+      "切 记",
+      "轻 盈",
+      "法 测定",
+      "par d",
+      "pa rd",
+      "p ard",
+      "魔龙 令牌",
+      "▁B low",
+      "▁Bl ow",
+      "▁Blo w",
+      "▁m ixes",
+      "▁mix es",
+      "▁run off",
+      "▁st rives",
+      "▁str ives",
+      "▁stri ves",
+      "▁strive s",
+      "▁mod ulates",
+      "▁modulate s",
+      "▁rad iological",
+      "▁radio logical",
+      "▁radi ological",
+      "▁radiol ogical",
+      "极 管",
+      "相 符",
+      "UC N",
+      "U CN",
+      "的有 关",
+      "的 有关",
+      "nS oviet",
+      "▁Tre vor",
+      "▁Min isters",
+      "▁Minister s",
+      "▁Mini sters",
+      "▁aggreg ated",
+      "▁aggregate d",
+      "▁negot iated",
+      "▁negotiate d",
+      "多 情",
+      "而 今",
+      "表 决",
+      "DT D",
+      "D TD",
+      "HA N",
+      "H AN",
+      "一本 书",
+      "一 本书",
+      "在 诉讼",
+      "专科 学校",
+      "专科学 校",
+      "是一名 演员",
+      "uls ory",
+      "▁Al pine",
+      "▁Alp ine",
+      "▁Web ster",
+      "▁vac ancy",
+      "IP v",
+      "很 适合",
+      "冠状 动脉",
+      "ac ons",
+      "acon s",
+      "aco ns",
+      "a cons",
+      "▁O mar",
+      "▁Om ar",
+      "\\\\ \\\\\\\\",
+      "\\\\\\\\ \\\\",
+      "\\\\\\ \\\\\\",
+      "\\\\\\\\\\ \\",
+      "\\ \\\\\\\\\\",
+      "▁s ponge",
+      "▁sp onge",
+      "▁spo nge",
+      "▁Syr acuse",
+      "坏 事",
+      "杏 花",
+      "村 干部",
+      "第三 边",
+      "第 三边",
+      "▁c hr",
+      "▁ch r",
+      "▁ chr",
+      "爷爷 奶奶",
+      "相关 证据",
+      "▁m ong",
+      "▁mon g",
+      "▁mo ng",
+      "▁ mong",
+      "公 办",
+      "切 勿",
+      "Go t",
+      "G ot",
+      "ib u",
+      "i bu",
+      "▁K C",
+      "▁ KC",
+      "▁e h",
+      "▁ eh",
+      "工商 联",
+      "由此 可见",
+      "in ged",
+      "ing ed",
+      "inge d",
+      "st ret",
+      "str et",
+      "stre t",
+      "s tret",
+      "糖尿 病患者",
+      "糖尿病 患者",
+      "npro cess",
+      "n process",
+      "▁substit uting",
+      "漂 流",
+      "ar gc",
+      "arg c",
+      "no va",
+      "nov a",
+      "n ova",
+      "om bo",
+      "omb o",
+      "▁Ro land",
+      "▁Rol and",
+      "▁adv ising",
+      "▁advis ing",
+      "中国 银行股份有限公司",
+      "中国银行 股份有限公司",
+      "双 双",
+      "小 猫",
+      "拉 黑",
+      "福 德",
+      "挖掘 机",
+      "非常 多",
+      "ar aj",
+      "ara j",
+      "a raj",
+      "{\\\\ \"",
+      "{ \\\\\"",
+      "▁k ub",
+      "▁ kub",
+      "▁Sh ane",
+      "▁Sha ne",
+      "▁Shan e",
+      "小 林",
+      "我 以前",
+      "ap sing",
+      "aps ing",
+      "ret rieve",
+      "▁sh aking",
+      "▁sha king",
+      "▁throm bin",
+      "▁thromb in",
+      "▁Compar ative",
+      "K i",
+      "也 行",
+      "青 天",
+      "TE S",
+      "T ES",
+      "可以 有",
+      "持有 的",
+      "持 有的",
+      "nR en",
+      "nRe n",
+      "n Ren",
+      "Nav igator",
+      "▁pitch ing",
+      "▁pit ching",
+      "▁analges ia",
+      "抢 购",
+      "而 下",
+      "本 项目",
+      "系列 活动",
+      "Vo ice",
+      "V oice",
+      "▁C laus",
+      "▁Cl aus",
+      "二〇一六年 十",
+      "ic ating",
+      "ica ting",
+      "▁ar ranging",
+      "▁arr anging",
+      "▁arrang ing",
+      "S Z",
+      "状态 的",
+      "Mar y",
+      "Ma ry",
+      "M ary",
+      "ic ki",
+      "ick i",
+      "第十 四章",
+      "第十四 章",
+      "第 十四章",
+      "mor tem",
+      "mort em",
+      "▁hor ny",
+      "▁horn y",
+      "▁diaph rag",
+      "▁dis traction",
+      "▁di straction",
+      "▁distr action",
+      "州市中级人民法院 提出上诉",
+      "州 市中级人民法院提出上诉",
+      "焚 烧",
+      "▁V es",
+      "▁Ve s",
+      "下面 是一个",
+      "下面是 一个",
+      "index es",
+      "Inst ances",
+      "Instance s",
+      "▁Inter val",
+      "▁ Interval",
+      "▁microbi ota",
+      "az o",
+      "a zo",
+      "销售 等",
+      "▁Ro ots",
+      "▁Root s",
+      "▁am alg",
+      "▁a rche",
+      "▁ar che",
+      "▁arch e",
+      "▁arc he",
+      "▁w ives",
+      "▁wi ves",
+      "▁ wives",
+      "ны й",
+      "结果 与",
+      "裁决 书",
+      "fa tal",
+      "fat al",
+      "f atal",
+      "▁se aling",
+      "▁sea ling",
+      "▁seal ing",
+      "de scribed",
+      "des cribed",
+      "describe d",
+      "▁embrace d",
+      "▁embr aced",
+      "▁circuit ry",
+      "▁panor amic",
+      "▁Eng agement",
+      "两 三",
+      "你不 是",
+      "你 不是",
+      "▁M AR",
+      "▁MA R",
+      "▁ MAR",
+      "vd ots",
+      "v dots",
+      "▁draw ers",
+      "▁drawer s",
+      "▁toile ts",
+      "▁toilet s",
+      "二 号",
+      "对 决",
+      "铸 铁",
+      "][ :",
+      "] [:",
+      "sv c",
+      "s vc",
+      "区间 为",
+      "的一 样",
+      "的 一样",
+      "的数 是",
+      "的 数是",
+      "污染 防治",
+      "简直 就是",
+      "do ing",
+      "doi ng",
+      "▁h ype",
+      "▁hy pe",
+      "▁hyp e",
+      "▁congr u",
+      "▁cong ru",
+      "▁tra pez",
+      "▁trap ez",
+      "▁trape z",
+      "prepare d",
+      "prep ared",
+      "本案 诉讼费由被告承担",
+      "本案诉讼费 由被告承担",
+      "中 风",
+      "拟 定",
+      "捕 获",
+      "是 会",
+      "皮 书",
+      "福 音",
+      "荣 华",
+      "EA A",
+      "E AA",
+      "UN G",
+      "U NG",
+      "▁d m",
+      "▁ dm",
+      "加大 对",
+      "▁D ul",
+      "▁Du l",
+      "PA RAM",
+      "PAR AM",
+      "▁func tools",
+      "▁funct ools",
+      "拟 任",
+      "是 种",
+      "gr en",
+      "gre n",
+      "g ren",
+      "▁A IM",
+      "▁AI M",
+      "ml ess",
+      "m less",
+      "住所地 江苏省",
+      "▁h acker",
+      "▁hack er",
+      "Over view",
+      "▁for give",
+      "▁forg ive",
+      "▁veter in",
+      "▁reservoir s",
+      "▁civil ization",
+      "专 访",
+      "养 的",
+      "泊 尔",
+      "圆柱 形",
+      "随时 随地",
+      "厦门市 思明",
+      "nIn cre",
+      "nInc re",
+      "n Incre",
+      "上诉于 上海市",
+      "▁brief ing",
+      "▁div isors",
+      "▁divisor s",
+      "▁dis position",
+      "▁disp osition",
+      "と い",
+      "创 设",
+      "特 拉",
+      "泰安 市",
+      "泰 安市",
+      "▁V augh",
+      "▁Va ugh",
+      "▁threat en",
+      "▁sim ulating",
+      "一 汽",
+      "基 石",
+      "大 牌",
+      "食 疗",
+      "ous se",
+      "▁m ans",
+      "▁man s",
+      "▁ma ns",
+      "▁ mans",
+      "et ched",
+      "etch ed",
+      "etc hed",
+      "▁原 料",
+      "▁ 原料",
+      "皮肤 的",
+      "▁e go",
+      "▁eg o",
+      "▁ ego",
+      "行政 审批",
+      "▁G ain",
+      "▁Ga in",
+      "iz umab",
+      "▁py ram",
+      "▁pyr am",
+      "sam pling",
+      "▁sn ippets",
+      "▁snippet s",
+      "hom ogeneous",
+      "乳 化",
+      "保 研",
+      "减 灾",
+      "执 教",
+      "站 位",
+      "这 支",
+      "NA S",
+      "N AS",
+      "▁ST AR",
+      "▁ STAR",
+      "▁K athy",
+      "▁Kat hy",
+      "▁Ka thy",
+      "▁Kath y",
+      "二〇一四年 九月",
+      "▁Krish na",
+      "▁musc uloskeletal",
+      "抽 出",
+      "时 可",
+      "和 时间",
+      "nas ium",
+      "ge ometric",
+      "geo metric",
+      "geom etric",
+      "并 入",
+      "永 嘉",
+      "会议 由",
+      "长大 的",
+      "长 大的",
+      "▁B ri",
+      "▁Br i",
+      "市场 价格",
+      "▁Min eral",
+      "▁Mine ral",
+      "▁prom pts",
+      "▁prompt s",
+      "▁tr apping",
+      "▁tra pping",
+      "▁trap ping",
+      "∈ （",
+      "十堰 市",
+      "红 细胞",
+      "bi sh",
+      "bis h",
+      "b ish",
+      "pl acing",
+      "plac ing",
+      "GL IGENCE",
+      "▁Pok émon",
+      "▁plac enta",
+      "上 火",
+      "超 强",
+      "bl att",
+      "bla tt",
+      "gr ave",
+      "grav e",
+      "gra ve",
+      "np ass",
+      "n pass",
+      "▁B ind",
+      "▁Bi nd",
+      "▁Bin d",
+      "▁ Bind",
+      "劳动争议 纠纷一案",
+      "▁day light",
+      "▁announce s",
+      "▁announc es",
+      "▁wave guide",
+      "▁ ş",
+      "cd f",
+      "c df",
+      "▁N L",
+      "▁ NL",
+      "别 忘了",
+      "这个 小",
+      "这 个小",
+      "▁C omes",
+      "▁Com es",
+      "▁Co mes",
+      "▁Come s",
+      "▁B anner",
+      "▁Ban ner",
+      "▁ Banner",
+      "▁CON SEQUENTIAL",
+      "ö m",
+      "欧 式",
+      "▁Q A",
+      "▁ QA",
+      "工资 款",
+      "需 付费",
+      "nW il",
+      "n Wil",
+      "▁S atisf",
+      "▁Sat isf",
+      "▁out right",
+      "▁ou tright",
+      "こ の",
+      "不 正",
+      "克 林",
+      "老 天",
+      "输 液",
+      "Co A",
+      "占有 率",
+      "第一 行",
+      "第 一行",
+      "nV ue",
+      "n Vue",
+      "ap opt",
+      "apo pt",
+      "怎样的 体验",
+      "管辖权 异议",
+      "op ters",
+      "opt ers",
+      "opter s",
+      "expl orer",
+      "Request Body",
+      "排 泄",
+      "机 上",
+      "极 小",
+      "软 体",
+      "包括 了",
+      "等 产品",
+      "与 被害人",
+      "感情 基础",
+      "▁sw agger",
+      "▁ swagger",
+      "▁exec utor",
+      "▁execut or",
+      "▁ executor",
+      "千 人",
+      "换 乘",
+      "\"/ /",
+      "\" //",
+      "下辖 村",
+      "是用 来",
+      "是 用来",
+      "Mar c",
+      "Ma rc",
+      "M arc",
+      "or ce",
+      "形成 一个",
+      "形 成一个",
+      "Log out",
+      "Logo ut",
+      "河南省 郑州市",
+      "▁che mot",
+      "▁chem ot",
+      "▁chemo t",
+      "roll able",
+      "open stack",
+      "▁paras itic",
+      "张 晓",
+      "杜 甫",
+      "在 选择",
+      "ub icin",
+      "ubi cin",
+      "▁Re mix",
+      "▁Rem ix",
+      "▁cor ps",
+      "ens ible",
+      "tra ffic",
+      "traff ic",
+      "▁C lement",
+      "▁Cl ement",
+      "▁Cle ment",
+      "▁hect ares",
+      "ー ド",
+      "东 汉",
+      "坚 韧",
+      "第一项 至",
+      "mor ph",
+      "m orph",
+      "om otor",
+      "omo tor",
+      "o motor",
+      "▁R oose",
+      "▁Ro ose",
+      "tutorial s",
+      "Cal ifornia",
+      "祖 父",
+      "誓 言",
+      "风 筝",
+      "️ \\",
+      "ip ei",
+      "ipe i",
+      "苏州市 吴",
+      "w ould",
+      "被执行人 吴",
+      "li ttle",
+      "lit tle",
+      "l ittle",
+      "py game",
+      "▁prost hesis",
+      "▁investig ative",
+      "三 章",
+      "龙 山",
+      "可以 对",
+      "计算 在",
+      "▁g ol",
+      "▁go l",
+      "j q",
+      "▁ 段",
+      "影 人",
+      "ts c",
+      "t sc",
+      "也 存在",
+      "但不 是",
+      "但 不是",
+      "▁K ai",
+      "▁Ka i",
+      "怎样 才能",
+      "英语 教学",
+      "执行机关 认为",
+      "被执行人 暂无",
+      "pl itude",
+      "plit ude",
+      "exper ience",
+      "各 样",
+      "思 索",
+      "ic um",
+      "i cum",
+      "broad cast",
+      "b roadcast",
+      "▁pers ists",
+      "▁persist s",
+      "▁amb itions",
+      "▁ambit ions",
+      "▁ambition s",
+      "平 坦",
+      "来 宾",
+      "▁g ui",
+      "▁gu i",
+      "▁ gui",
+      "好好 学习",
+      "tag ged",
+      "t agged",
+      "▁Wend y",
+      "▁Wen dy",
+      "Non null",
+      "STIT UTE",
+      "▁ver ses",
+      "▁vers es",
+      "▁verse s",
+      "▁ verses",
+      "▁sing ularity",
+      "▁singular ity",
+      "l w",
+      "与 原",
+      "处 境",
+      "▁ är",
+      "敏感 性",
+      "敏 感性",
+      "完全 破裂",
+      "▁O ste",
+      "▁Os te",
+      "▁Ost e",
+      "▁re gain",
+      "▁reg ain",
+      "▁ubiquit in",
+      "正 月",
+      "▁w p",
+      "▁ wp",
+      "▁P ix",
+      "▁Pi x",
+      "▁ Pix",
+      "▁p si",
+      "▁ps i",
+      "▁ psi",
+      "县委 书记",
+      "县 委书记",
+      "▁Six ty",
+      "▁T ucker",
+      "▁Tuc ker",
+      "z f",
+      "▁ 建",
+      "▁ 闫",
+      "田 县",
+      "花 岗",
+      "装 入",
+      "对本 院",
+      "对 本院",
+      "的小 孩",
+      "的 小孩",
+      "Ni ght",
+      "N ight",
+      "▁c amel",
+      "▁ca mel",
+      "▁came l",
+      "▁cam el",
+      "▁ camel",
+      "ran ging",
+      "r anging",
+      "▁cr ashing",
+      "▁crash ing",
+      "▁des erved",
+      "▁deserve d",
+      "▁Insp ection",
+      "一 滴",
+      "六年 级",
+      "六 年级",
+      "man s",
+      "ma ns",
+      "m ans",
+      "▁G ad",
+      "▁Ga d",
+      "气候 变化",
+      "▁C andy",
+      "▁Can dy",
+      "▁Cand y",
+      "▁S PECT",
+      "▁SP ECT",
+      "▁SPE CT",
+      "cre ating",
+      "creat ing",
+      "c reating",
+      "▁buff ers",
+      "▁buffer s",
+      "▁buf fers",
+      "游 行",
+      "El e",
+      "E le",
+      "好 消息",
+      "汽车 站",
+      "汽 车站",
+      "▁H elic",
+      "▁He lic",
+      "▁Hel ic",
+      "▁Law yers",
+      "FORM ATION",
+      "FORMAT ION",
+      "ar ra",
+      "arr a",
+      "ta pp",
+      "tap p",
+      "t app",
+      "▁P PAR",
+      "▁PP AR",
+      "▁bl unt",
+      "▁blu nt",
+      "▁bur nt",
+      "▁burn t",
+      "utter ing",
+      "utt ering",
+      "宴 请",
+      "sb t",
+      "s bt",
+      "十一 年",
+      "十 一年",
+      "和 要求",
+      "适合 于",
+      "都 具有",
+      "Z ONE",
+      "oh an",
+      "oha n",
+      "o han",
+      "▁ph age",
+      "▁phag e",
+      "▁Willi e",
+      "▁Will ie",
+      "▁Wil lie",
+      "▁en anti",
+      "initial ized",
+      "initialize d",
+      "前 卫",
+      "无 色",
+      "Ba n",
+      "B an",
+      "崭 新的",
+      "胶原 蛋白",
+      "▁r age",
+      "▁ra ge",
+      "▁rag e",
+      "▁ rage",
+      "▁in noc",
+      "▁inn oc",
+      "LOC ATION",
+      "▁P atterns",
+      "▁Pat terns",
+      "▁Pattern s",
+      "▁ Patterns",
+      "▁crafts manship",
+      "▁craftsm anship",
+      "异 步",
+      "我是 个",
+      "我 是个",
+      "是 关于",
+      "进程 中",
+      "ON LY",
+      "R isk",
+      "fa ng",
+      "fan g",
+      "f ang",
+      "do ped",
+      "d oped",
+      "ub ert",
+      "ube rt",
+      "uber t",
+      "u bert",
+      "het ized",
+      "nMar tin",
+      "nMart in",
+      "n Martin",
+      "瓦 尔",
+      "福 彩",
+      "PC B",
+      "P CB",
+      "也是 在",
+      "也 是在",
+      "右下 角",
+      "这一 年",
+      "这 一年",
+      "og el",
+      "oge l",
+      "o gel",
+      "TAIN ER",
+      "le tion",
+      "let ion",
+      "▁Bul ls",
+      "▁Bull s",
+      "~ (",
+      "田 野",
+      "程 式",
+      "自然 而",
+      "自 然而",
+      "纪委 监委",
+      "pa ired",
+      "pair ed",
+      "p aired",
+      "▁eth yl",
+      "▁ ethyl",
+      "▁excit atory",
+      "赏 析",
+      "奥特 曼",
+      "__ (*",
+      "__( *",
+      "▁O WN",
+      "理论 知识",
+      "au nch",
+      "a unch",
+      "ie ren",
+      "ier en",
+      "iere n",
+      "i eren",
+      "▁R eds",
+      "▁Re ds",
+      "▁Red s",
+      "▁p osing",
+      "▁pos ing",
+      "▁po sing",
+      "▁ posing",
+      "之 于",
+      "抚 育",
+      "有关 系",
+      "有 关系",
+      "让自己 的",
+      "让 自己的",
+      "▁N arr",
+      "▁Na rr",
+      "▁Nar r",
+      "▁qu ota",
+      "▁quot a",
+      "▁quo ta",
+      "FUN CTION",
+      "▁Re action",
+      "▁React ion",
+      "▁multi center",
+      "▁multic enter",
+      "唯 物",
+      "村 级",
+      "▁ 函数",
+      "物体 的",
+      "物 体的",
+      "呼和浩特 市",
+      "ind rome",
+      "remove Class",
+      "创 刊",
+      "差 价",
+      "手 绘",
+      "nH E",
+      "n HE",
+      "ow o",
+      "o wo",
+      "之后 再",
+      "之 后再",
+      "el dom",
+      "eld om",
+      "▁am yg",
+      "▁amy g",
+      "vis iae",
+      "▁sh unt",
+      "依法 提起诉讼",
+      "nV ictor",
+      "n Victor",
+      "▁vegg ies",
+      "文化传媒 有限公司",
+      "出 名",
+      "处 以",
+      "家 住",
+      "异 侠",
+      "▁E lm",
+      "▁El m",
+      "ca kes",
+      "cake s",
+      "c akes",
+      "set Item",
+      "▁suc rose",
+      "▁favour able",
+      "克 莱",
+      "挂 靠",
+      "可以 被",
+      "西游 记",
+      "▁N em",
+      "▁Ne m",
+      "损伤 程度",
+      "▁Be am",
+      "▁Bea m",
+      "▁C ater",
+      "▁Ca ter",
+      "▁Cat er",
+      "▁Exped ition",
+      "▁expansion s",
+      "▁expans ions",
+      "▁ 吉",
+      "按 期",
+      "自 救",
+      "du p",
+      "d up",
+      "都 一样",
+      "▁U PS",
+      "▁UP S",
+      "▁ UPS",
+      "▁g ir",
+      "▁gi r",
+      "三大 阶段",
+      "第三项 裁定",
+      "▁Tor res",
+      "uit ively",
+      "uitive ly",
+      "u itively",
+      "▁kotlin x",
+      "mark eting",
+      "market ing",
+      "▁team mate",
+      "trad itional",
+      "永 定",
+      "盗 走",
+      "赋 能",
+      "de a",
+      "d ea",
+      "▁天 津",
+      "▁ 天津",
+      "海口 市",
+      "海 口市",
+      "nV an",
+      "n Van",
+      "ns tr",
+      "nst r",
+      "n str",
+      "该局 局长",
+      "该 局局长",
+      "▁g els",
+      "▁ge ls",
+      "▁gel s",
+      "par ing",
+      "pa ring",
+      "p aring",
+      "▁doc king",
+      "▁dock ing",
+      "In frastructure",
+      "▁Parliament ary",
+      "忧 郁",
+      "肾 病",
+      "这 将",
+      "连 结",
+      "▁ 显示",
+      "绿色 的",
+      "绿 色的",
+      "▁Zur ich",
+      "▁f iller",
+      "▁fil ler",
+      "▁fill er",
+      "▁fi ller",
+      "▁prev ailing",
+      "▁ 兰",
+      "幕 墙",
+      "老 挝",
+      "平板 电脑",
+      "▁W idth",
+      "▁Wid th",
+      "▁ Width",
+      "▁Bel grade",
+      "▁mel atonin",
+      "au i",
+      "a ui",
+      "不 公平",
+      "人员 在",
+      "卫生 所",
+      "iv os",
+      "ivo s",
+      "ur in",
+      "uri n",
+      "u rin",
+      "X gate",
+      "ip ient",
+      "▁D ixon",
+      "▁s tint",
+      "▁st int",
+      "▁Wh ilst",
+      "segment s",
+      "seg ments",
+      "▁batt ling",
+      "▁anest hetic",
+      "ç o",
+      "▁ 光",
+      "到 头",
+      "太 行",
+      "相 切",
+      "nA S",
+      "n AS",
+      "vey or",
+      "民间借贷 纠纷",
+      "Man aged",
+      "Manag ed",
+      "Manage d",
+      "▁Czech oslovak",
+      "财 税",
+      "长 发",
+      "▁f aint",
+      "▁fa int",
+      "▁ 答",
+      "ッ ト",
+      "丽 娜",
+      "名 城",
+      "换 装",
+      "最 值",
+      "烈 的",
+      "等 为",
+      "足 额",
+      "AT M",
+      "A TM",
+      "at as",
+      "ata s",
+      "a tas",
+      "ti po",
+      "tip o",
+      "▁Ar bor",
+      "IZ ATION",
+      "cast ers",
+      "ca sters",
+      "cas ters",
+      "caster s",
+      "c asters",
+      "▁car amel",
+      "Exper imental",
+      "蓝 光",
+      "入住 时",
+      "有用 的",
+      "有 用的",
+      "因为 没有",
+      "差不多 了",
+      "差不 多了",
+      "de cre",
+      "dec re",
+      "how to",
+      "▁S ell",
+      "▁Se ll",
+      "▁Sel l",
+      "ose cond",
+      "o second",
+      "▁l ocker",
+      "▁loc ker",
+      "▁lock er",
+      "科技 发展有限公司",
+      "玉 石",
+      "▁ 深圳",
+      "去看 看",
+      "去 看看",
+      "网友 们",
+      "网 友们",
+      "骨干 教师",
+      "▁F old",
+      "▁Fol d",
+      "▁Fo ld",
+      "▁ Fold",
+      "▁s eas",
+      "▁se as",
+      "▁sea s",
+      "▁F lood",
+      "▁Fl ood",
+      "▁Flo od",
+      "▁fl ames",
+      "▁flame s",
+      "X D",
+      "▁ 磊",
+      "因 病",
+      "艳 丽",
+      "na k",
+      "n ak",
+      "中 没有",
+      "引发 了",
+      "引 发了",
+      "滁 州市",
+      "De cl",
+      "Dec l",
+      "D ecl",
+      "im ap",
+      "ima p",
+      "i map",
+      "ast atin",
+      "asta tin",
+      "▁F ighter",
+      "▁Fight er",
+      "Fi ve",
+      "F ive",
+      "▁j an",
+      "▁ja n",
+      "▁ jan",
+      "ic ola",
+      "ico la",
+      "icol a",
+      "▁Whit ney",
+      "准 入",
+      "贬 值",
+      "的 客户",
+      "en as",
+      "ena s",
+      "e nas",
+      "wo rt",
+      "wor t",
+      "w ort",
+      "各个 方面",
+      "▁St roke",
+      "▁Str oke",
+      "nAltern atively",
+      "查 验",
+      "每 秒",
+      "气 场",
+      "疏 通",
+      "秋 色",
+      "失败 的",
+      "RO LL",
+      "ROL L",
+      "ed ay",
+      "eda y",
+      "e day",
+      "还会 提供",
+      "还 会提供",
+      "▁Ch ak",
+      "▁Cha k",
+      "▁l umen",
+      "▁lum en",
+      "▁lu men",
+      "in stant",
+      "inst ant",
+      "anal yzer",
+      "base line",
+      "bas eline",
+      "struct ural",
+      "等 单位",
+      "▁D ys",
+      "▁Dy s",
+      "公交 线路",
+      "限制 消费",
+      "Est im",
+      "Es tim",
+      "E stim",
+      "▁J eep",
+      "▁Je ep",
+      "el late",
+      "ell ate",
+      "ella te",
+      "▁rad ii",
+      "▁radi i",
+      "▁New ark",
+      "▁in continence",
+      "▁ İ",
+      "心 意",
+      "手 艺",
+      "Ke n",
+      "K en",
+      "就 只能",
+      "to os",
+      "too s",
+      "t oos",
+      "▁R az",
+      "▁Ra z",
+      "正确 判断",
+      "法律 后果",
+      "aut oload",
+      "auto load",
+      "combin ed",
+      "comb ined",
+      "combine d",
+      "▁don ating",
+      "Model Admin",
+      "▁geomet ries",
+      "▁deg enerative",
+      "农 夫",
+      "博 彩",
+      "故 居",
+      "本 的",
+      "al om",
+      "alo m",
+      "var n",
+      "va rn",
+      "v arn",
+      "▁b fs",
+      "▁ bfs",
+      "hen ko",
+      "h enko",
+      "at cher",
+      "atch er",
+      "▁Com bat",
+      "▁Comb at",
+      "▁voy age",
+      "亚 历",
+      "动 感",
+      "北 欧",
+      "销售 量",
+      "Bo ld",
+      "B old",
+      "pi nk",
+      "pin k",
+      "p ink",
+      "▁A CM",
+      "▁AC M",
+      "▁coch lear",
+      "nBe autiful",
+      "n Beautiful",
+      "▁Rel igious",
+      "户 名",
+      "祝 你",
+      "id f",
+      "i df",
+      "的 魅力",
+      "▁A MA",
+      "▁AM A",
+      "认定 被告人",
+      "Tag Name",
+      "▁Ref uge",
+      "▁Diff usion",
+      "▁enthus iast",
+      "▁enthusi ast",
+      "性 命",
+      "希望 你",
+      "都 必须",
+      "子叶 植物",
+      "▁L ens",
+      "▁Le ns",
+      "▁Len s",
+      "权纠纷一案 中",
+      "权 纠纷一案中",
+      "▁Em manuel",
+      "▁counc ils",
+      "▁council s",
+      "▁Bench mark",
+      "▁ Benchmark",
+      "▁empower ing",
+      "司 长",
+      "选 修",
+      "bi lt",
+      "bil t",
+      "b ilt",
+      "▁in i",
+      "▁i ni",
+      "▁ ini",
+      "▁r ash",
+      "▁ra sh",
+      "▁Th ames",
+      "Public Key",
+      "▁compens ated",
+      "▁compensate d",
+      "▁indispens able",
+      "各 县",
+      "补 办",
+      "反映 出",
+      "▁B arton",
+      "▁Bar ton",
+      "▁Bart on",
+      "▁Dam age",
+      "▁ Damage",
+      "▁broch ure",
+      "▁ 半",
+      "与 张",
+      "人 寿",
+      "本 组",
+      "相 色谱",
+      "al lo",
+      "all o",
+      "nF oot",
+      "▁L azy",
+      "▁La zy",
+      "▁Laz y",
+      "▁ Lazy",
+      "omat ous",
+      "▁str ands",
+      "▁strand s",
+      "▁A ircraft",
+      "▁Air craft",
+      "O rientation",
+      "▁electron ically",
+      "▁electronic ally",
+      "我 做",
+      "Ma s",
+      "M as",
+      "▁ 房价",
+      "ual a",
+      "ua la",
+      "u ala",
+      "S uffix",
+      "nr ight",
+      "n right",
+      "▁Cl aud",
+      "第一项至 第三项裁定",
+      "▁ 千",
+      "小 鸟",
+      "铁 矿",
+      "在 任何",
+      "▁C ot",
+      "▁Co t",
+      "▁E as",
+      "em poral",
+      "emp oral",
+      "empor al",
+      "▁com plemented",
+      "▁complement ed",
+      "了 然",
+      "发 泄",
+      "物 件",
+      "bf s",
+      "b fs",
+      "st k",
+      "s tk",
+      "vi ll",
+      "vil l",
+      "v ill",
+      "▁U ni",
+      "▁Un i",
+      "也是 如此",
+      "▁un to",
+      "▁u nto",
+      "▁unt o",
+      "▁p alsy",
+      "▁pal sy",
+      "▁tal lest",
+      "▁tall est",
+      "▁Rel igion",
+      "感 触",
+      "采 光",
+      "黄 芪",
+      "未知 数",
+      "求 函数",
+      "▁h urry",
+      "▁hur ry",
+      "▁tun nels",
+      "▁tunnel s",
+      "丽 萍",
+      "的 设备",
+      "gin ia",
+      "gi nia",
+      "g inia",
+      "▁b askets",
+      "▁basket s",
+      "▁master ing",
+      "▁mast ering",
+      "▁nav igating",
+      "自 幼",
+      "宿迁 市",
+      ")} $$",
+      ")}$ $",
+      ") }$$",
+      "重点 工作",
+      "er ring",
+      "err ing",
+      "es tring",
+      "est ring",
+      "estr ing",
+      "e string",
+      "▁ins ist",
+      "▁Un ified",
+      "▁En forcement",
+      "▁complex ities",
+      "安 吉",
+      "找 准",
+      "前 公示",
+      "实在 太",
+      "第一百 二十",
+      "▁en thal",
+      "▁ent hal",
+      "nP URPOSE",
+      "▁Comp iler",
+      "▁ Compiler",
+      "▁disc arded",
+      "▁discard ed",
+      "▁anis otropy",
+      "可 疑",
+      "的 优秀",
+      "he id",
+      "LOB AL",
+      "▁L emon",
+      "▁Le mon",
+      "▁Lem on",
+      "Sim ilar",
+      "S imilar",
+      "syn chron",
+      "s ynchron",
+      "▁c ranial",
+      "▁cran ial",
+      "▁de polar",
+      "▁dep olar",
+      "▁loose ly",
+      "▁loos ely",
+      "▁b iblical",
+      "▁bibli cal",
+      "▁minim ized",
+      "▁minimize d",
+      "▁symmetric al",
+      "▁symmet rical",
+      "Sw ift",
+      "▁a unt",
+      "▁au nt",
+      "M iddle",
+      "▁Sh ock",
+      "二〇一四年 七月",
+      "▁Reg ulatory",
+      "粘 膜",
+      "跨 度",
+      "pr incipal",
+      "princ ipal",
+      "一 文",
+      "pt o",
+      "p to",
+      "enc ils",
+      "encil s",
+      "ile vel",
+      "i level",
+      "condition ing",
+      "Re w",
+      "R ew",
+      "最强 的",
+      "最 强的",
+      "遥控 器",
+      "it en",
+      "ite n",
+      "i ten",
+      "任免 信息",
+      "ur face",
+      "urf ace",
+      "▁To ast",
+      "▁ Toast",
+      "步 兵",
+      "等 值",
+      "这 让",
+      "▁D ud",
+      "▁Du d",
+      "▁S CC",
+      "▁SC C",
+      "住 吉林省",
+      "nE asy",
+      "n Easy",
+      "向本院 起诉",
+      "▁d owns",
+      "▁down s",
+      "▁dow ns",
+      "▁k appa",
+      "▁ kappa",
+      "▁Pol bot",
+      "▁w ording",
+      "▁word ing",
+      "▁wor ding",
+      "▁Christ ina",
+      "▁Scient ist",
+      "▁Af fordable",
+      "▁accus ations",
+      "▁personal ised",
+      "坏 的",
+      "Sa y",
+      "S ay",
+      "所以 他",
+      "请 分析",
+      "an si",
+      "ans i",
+      "市委 书记",
+      "市 委书记",
+      "zhi hu",
+      "▁v ene",
+      "▁ve ne",
+      "▁ven e",
+      "cons ult",
+      "SET TINGS",
+      "opt imize",
+      "optim ize",
+      "▁trump et",
+      "▁spe eches",
+      "▁speech es",
+      "Serialize Field",
+      "▁prefer entially",
+      "独 生",
+      "在其 他",
+      "在 其他",
+      "at he",
+      "ath e",
+      "a the",
+      "经 相关部门",
+      "nPl ants",
+      "▁Par agu",
+      "▁Para gu",
+      "▁Tr ends",
+      "▁Trend s",
+      "▁Mess enger",
+      "▁sensit ization",
+      "榆 树",
+      "追 星",
+      "将 持续",
+      "试用 期",
+      "配备 了",
+      "▁C LE",
+      "▁CL E",
+      "▁ CLE",
+      "com ings",
+      "coming s",
+      "▁rob bery",
+      "▁inc arcer",
+      "nS omething",
+      "nSome thing",
+      "n Something",
+      "VERTI SEMENT",
+      "喷 射",
+      "山 羊",
+      "BR O",
+      "B RO",
+      "Et h",
+      "E th",
+      "XY Z",
+      "da d",
+      "d ad",
+      "无数 的",
+      "无 数的",
+      "▁m yc",
+      "▁my c",
+      "我们 已经",
+      "▁s oak",
+      "▁so ak",
+      "▁V inyl",
+      "▁Vin yl",
+      "▁Fire base",
+      "▁Colomb ian",
+      "▁Colombia n",
+      "▁anomal ous",
+      "声 卡",
+      "正 直",
+      "了一 起",
+      "了 一起",
+      "▁Al eks",
+      "▁Ale ks",
+      "▁PROF ITS",
+      "▁gr illed",
+      "▁grill ed",
+      "印 记",
+      "夫 子",
+      "沥 干",
+      "莱 芜",
+      "▁G V",
+      "▁ GV",
+      "at ican",
+      "atic an",
+      "ati can",
+      "Un iform",
+      "game Object",
+      "▁pre defined",
+      "▁pred efined",
+      "夺 取",
+      "小 众",
+      "果 蔬",
+      "▁用 量",
+      "▁ 用量",
+      "▁T WO",
+      "▁TW O",
+      "你知道 吗",
+      "an imal",
+      "ani mal",
+      "anim al",
+      "文物保护 单位",
+      "▁co work",
+      "▁cow ork",
+      "▁Ple asant",
+      "▁fore ground",
+      "▁bio informatics",
+      "滨 江",
+      "能 手",
+      "id l",
+      "i dl",
+      "和 精神",
+      "的 措施",
+      "Ad am",
+      "A dam",
+      "Gi ve",
+      "G ive",
+      "▁A NA",
+      "▁AN A",
+      "▁ ANA",
+      "基本 情况",
+      "ap iro",
+      "api ro",
+      "€ ™",
+      "三 五",
+      "中 超",
+      "别 离",
+      "核 定",
+      "孙某 某",
+      "孙 某某",
+      "工作 站",
+      "的 资源",
+      "▁D ag",
+      "▁Da g",
+      "本科 毕业",
+      "▁thr ill",
+      "积极参加 生产劳动",
+      "Com posite",
+      "上 新",
+      "杨 幂",
+      "流 派",
+      "突 起",
+      "nA w",
+      "n Aw",
+      "▁P up",
+      "▁Pu p",
+      "节能 减排",
+      "enron Xgate",
+      "▁bio availability",
+      "穿 戴",
+      "这道 题目",
+      "这道题 目",
+      "▁swe ater",
+      "▁sweat er",
+      "▁Speed way",
+      "▁jour neys",
+      "▁journey s",
+      "Mapping URL",
+      "▁transf ection",
+      "nB er",
+      "nBe r",
+      "n Ber",
+      "▁T emp",
+      "▁Te mp",
+      "▁Tem p",
+      "▁ Temp",
+      "em ails",
+      "email s",
+      "ema ils",
+      "e mails",
+      "web hook",
+      "▁url parse",
+      "▁addict ive",
+      "和 外",
+      "WN r",
+      "与 美国",
+      "罗马 尼亚",
+      "▁L ite",
+      "▁Li te",
+      "▁Lit e",
+      "▁ Lite",
+      "nCom mand",
+      "n Command",
+      "omet rics",
+      "ometric s",
+      "o metrics",
+      "nMult iply",
+      "▁pharmac ologic",
+      "而 未",
+      "莲 子",
+      "从来 没",
+      "利物 浦",
+      "nSt ay",
+      "nSta y",
+      "▁_ (\\'",
+      "▁ _(\\'",
+      "▁Jak arta",
+      "condition ed",
+      "▁ess entials",
+      "▁essential s",
+      "再 者",
+      "托 盘",
+      "民进 党",
+      "op ress",
+      "opre ss",
+      "o press",
+      "\"_ _",
+      "\" __",
+      "直接 把",
+      "创新 发展",
+      "进行 审查",
+      "▁In fect",
+      "▁Inf ect",
+      "nick name",
+      "▁irr itation",
+      "▁irrit ation",
+      "▁pres cribing",
+      "对前款 第一项至第三项裁定",
+      "▁firef ighters",
+      "后 宫",
+      "谁 来",
+      "发生 过",
+      "是 基础题",
+      "▁de riving",
+      "▁der iving",
+      "▁deriv ing",
+      "▁Vol leyball",
+      "▁Congress ional",
+      "周 二",
+      "芬 芳",
+      "长 兴",
+      "风 向",
+      "SY S",
+      "S YS",
+      "可以 不",
+      "的 建筑",
+      "le in",
+      "lei n",
+      "l ein",
+      "协商 解决",
+      "解 不等式",
+      "国家 税务总局",
+      "▁min ers",
+      "▁mine rs",
+      "▁mi ners",
+      "▁Alb ania",
+      "▁Alban ia",
+      "▁inter connected",
+      "а й",
+      "息 息",
+      "挤 出",
+      "仙侠 异侠",
+      "▁En semble",
+      "受 偿",
+      "▁в о",
+      "▁ во",
+      "▁发 布",
+      "▁ 发布",
+      "和 各种",
+      "的 氛围",
+      "社会 发展",
+      "At tempt",
+      "Att empt",
+      "c riteria",
+      "▁en closure",
+      "此 种",
+      "的 年龄",
+      "nQu ick",
+      "n Quick",
+      "▁THE ORY",
+      "▁UN ESCO",
+      "char acters",
+      "character s",
+      "▁delet ions",
+      "▁deletion s",
+      "source forge",
+      "▁advantage ous",
+      "无 忌",
+      "oh a",
+      "o ha",
+      "嘉兴 市",
+      "依法 应予",
+      "依法应 予",
+      "nM icro",
+      "nMi cro",
+      "n Micro",
+      "申请执行人 申请",
+      "▁aneurys ms",
+      "▁aneurysm s",
+      "▁verte bral",
+      "▁vertebr al",
+      "减 免",
+      "左 上",
+      "烟 气",
+      "rf c",
+      "r fc",
+      "的 日常",
+      "US IC",
+      "al ty",
+      "alt y",
+      "决定 如下",
+      "nt han",
+      "nth an",
+      "n than",
+      "▁Ka plan",
+      "▁Kap lan",
+      "▁par ench",
+      "▁pa rench",
+      "▁pare nch",
+      "▁paren ch",
+      "▁te aspoon",
+      "▁teas poon",
+      "▁ment orship",
+      "▁mentor ship",
+      "▁mentors hip",
+      "Z Y",
+      "你 又",
+      "削 减",
+      "奋 战",
+      "每 亩",
+      "理 会",
+      "wa p",
+      "w ap",
+      "他 已经",
+      "Ch ip",
+      "C hip",
+      "空气 净化",
+      "\\\\ }\\\\",
+      "\\\\} \\\\",
+      "现实 生活中",
+      "nYou ng",
+      "n Young",
+      "▁S table",
+      "▁St able",
+      "▁Sta ble",
+      "▁ Stable",
+      "▁ch icks",
+      "▁chick s",
+      "▁chi cks",
+      "▁chic ks",
+      "系其 真实意思表示",
+      "nRe public",
+      "nRep ublic",
+      "▁sk ipping",
+      "▁skip ping",
+      "▁ski pping",
+      "▁cour tyard",
+      "▁court yard",
+      "▁supplement ary",
+      "内 含",
+      ",) \\",
+      ", )\\",
+      "▁{ |",
+      "▁ {|",
+      "ce le",
+      "cel e",
+      "c ele",
+      "可以 获得",
+      "外形 尺寸",
+      "k p",
+      "多 地",
+      "手 腕",
+      "报 废",
+      "张艺 兴",
+      "这一 个",
+      "这 一个",
+      "食用 油",
+      "交易 中心",
+      "un set",
+      "uns et",
+      "u nset",
+      "▁玄幻 小说",
+      "信息 技术有限公司",
+      "信息技术 有限公司",
+      "sing leton",
+      "single ton",
+      "▁Ex isting",
+      "▁analges ic",
+      "▁ 辉",
+      "叶 柄",
+      "源 泉",
+      "整数 部分",
+      "▁c reek",
+      "▁cre ek",
+      "▁cr eek",
+      "▁ded uce",
+      "▁mir acle",
+      "▁smooth er",
+      "▁smo other",
+      "薄 的",
+      "与 现代",
+      "▁V alle",
+      "▁Val le",
+      "▁Vall e",
+      "▁axi oms",
+      "▁axiom s",
+      "▁major s",
+      "▁maj ors",
+      "Meta Data",
+      "tp osition",
+      "t position",
+      "▁compet encies",
+      "是 经",
+      "笑 声",
+      "红 了",
+      "W ood",
+      "ad eon",
+      "ade on",
+      "dr iving",
+      "driv ing",
+      "d riving",
+      "▁en forced",
+      "▁enforce d",
+      "▁fundra iser",
+      "松 动",
+      "病 灶",
+      "元 人民币",
+      "减刑 幅度",
+      "教育 资源",
+      "sc rap",
+      "scr ap",
+      "▁hel ix",
+      "▁G ather",
+      "▁Ga ther",
+      "▁Gat her",
+      "C raft",
+      "▁S NPs",
+      "▁SN Ps",
+      "▁SNP s",
+      "Res ize",
+      "Re size",
+      "▁d evil",
+      "▁de vil",
+      "▁dev il",
+      "▁Sen ators",
+      "▁Senator s",
+      "% （",
+      "示 威",
+      "▁ 经济",
+      "机场 等",
+      "ro ve",
+      "rov e",
+      "r ove",
+      "出租车 约",
+      "是不 可能",
+      "是 不可能",
+      "▁cor ro",
+      "并按照 对方当事人",
+      "High light",
+      "▁North ampton",
+      "▁benefici aries",
+      "滚 筒",
+      "慈溪 市",
+      "慈 溪市",
+      "▁R ental",
+      "▁Ren tal",
+      "▁Rent al",
+      "二〇一八年 一月",
+      "东 升",
+      "ee e",
+      "e ee",
+      "▁用 途",
+      "▁ 用途",
+      "本来 就",
+      "wo ff",
+      "wof f",
+      "w off",
+      "▁N MDA",
+      "▁NM DA",
+      "va tion",
+      "v ation",
+      "▁Rect angle",
+      "▁ Rectangle",
+      "▁allow ance",
+      "上 场",
+      "厚 重",
+      "的美 国",
+      "的 美国",
+      "Ba by",
+      "B aby",
+      "nI ss",
+      "nIs s",
+      "n Iss",
+      "in hib",
+      "inh ib",
+      "▁t ense",
+      "▁te nse",
+      "▁ten se",
+      "▁tens e",
+      "▁ 结",
+      "这 台",
+      "不合 适",
+      "不 合适",
+      "买 东西",
+      "出现 过",
+      "和 别人",
+      "的反 应",
+      "的 反应",
+      "重新 启动",
+      "▁Ro uge",
+      "▁Rou ge",
+      "anc elled",
+      "ancel led",
+      "ancell ed",
+      "▁Institute s",
+      "▁Instit utes",
+      "好 歹",
+      "增加 到",
+      "战斗 机",
+      "情侣 约会",
+      "▁Love ly",
+      "▁Lo vely",
+      "▁Lov ely",
+      "npro ject",
+      "n project",
+      "▁K atherine",
+      "臭 氧",
+      "的分 类",
+      "的 分类",
+      "▁mi RNA",
+      "▁miR NA",
+      "▁s pars",
+      "▁sp ars",
+      "▁spa rs",
+      "▁F ooter",
+      "▁Foot er",
+      "▁Foo ter",
+      "▁Fo oter",
+      "▁ Footer",
+      "▁random ised",
+      "堡 垒",
+      "无 常",
+      "是 天",
+      "短 篇",
+      "阳 泉",
+      "探测 器",
+      "深度 学习",
+      "▁T ate",
+      "▁Ta te",
+      "▁Tat e",
+      "BUT TON",
+      "Th rows",
+      "Throw s",
+      "▁st roll",
+      "▁str oll",
+      "▁stro ll",
+      "动 荡",
+      "宁 海",
+      "无 害",
+      "im ar",
+      "ima r",
+      "i mar",
+      "ne nv",
+      "nen v",
+      "n env",
+      "▁F uk",
+      "▁Fu k",
+      "与被告 刘",
+      "var ez",
+      "va rez",
+      "vare z",
+      "oly mer",
+      "ounc ill",
+      "ouncil l",
+      "不 求",
+      "后 排",
+      "MO S",
+      "M OS",
+      "▁I sa",
+      "▁Is a",
+      "ph ire",
+      "phi re",
+      "p hire",
+      "可能 是因为",
+      "可能是 因为",
+      "ther apy",
+      "▁migr ant",
+      "▁mi grant",
+      "▁progress es",
+      "杨 紫",
+      "病 历",
+      "落 幕",
+      "你 真的",
+      "催告 书",
+      "新品 种",
+      "新 品种",
+      "▁Trin idad",
+      "data source",
+      "datas ource",
+      "▁contral ateral",
+      "▁ ₹",
+      "两 层",
+      "卖 场",
+      "▁S NA",
+      "▁SN A",
+      "▁ SNA",
+      "gl oss",
+      "g loss",
+      "▁z eta",
+      "▁ze ta",
+      "▁ zeta",
+      "号执行 案件",
+      "号 执行案件",
+      "▁p print",
+      "▁pp rint",
+      "▁DIS CLAI",
+      "C W",
+      "借 此",
+      "成 活",
+      "ha re",
+      "har e",
+      "h are",
+      ")\\ ')\\",
+      ")\\' )\\",
+      ") \\')\\",
+      "bo wer",
+      "bow er",
+      "b ower",
+      "um ina",
+      "umin a",
+      "umi na",
+      "u mina",
+      "显着性 差异",
+      "给付 之日止",
+      "▁PRO JECT",
+      "▁ PROJECT",
+      "▁minim ization",
+      "不 须",
+      "居 士",
+      "浇 水",
+      "音 色",
+      "GI T",
+      "G IT",
+      "和 维护",
+      "▁t cp",
+      "▁tc p",
+      "▁ tcp",
+      "实践 经验",
+      "▁C asa",
+      "▁Ca sa",
+      "▁Cas a",
+      "▁小说 类型",
+      "▁up regulation",
+      "下 沉",
+      "烧 伤",
+      "▁f ir",
+      "▁fi r",
+      "从 根本上",
+      "tool kit",
+      "▁tra verse",
+      "▁ ½",
+      "urd ish",
+      "▁N icol",
+      "▁Nic ol",
+      "▁Ni col",
+      "▁a ided",
+      "▁aid ed",
+      "▁ai ded",
+      "▁com fy",
+      "▁sn ails",
+      "▁snail s",
+      "▁distr acted",
+      "编 委",
+      "SM S",
+      "S MS",
+      "可以 向",
+      "编辑 部",
+      "sb in",
+      "s bin",
+      "xl sx",
+      "xls x",
+      "▁Cor relation",
+      "但 有",
+      "末 世",
+      "汕 头",
+      "磁 性",
+      "袜 子",
+      "nL OG",
+      "n LOG",
+      "ced ures",
+      "cedure s",
+      "▁eff lux",
+      "▁s ounding",
+      "▁so unding",
+      "▁sound ing",
+      "▁accel erator",
+      "▁acceler ator",
+      "▁micro vascular",
+      "的 眼睛",
+      "大学 教授",
+      "▁tele com",
+      "▁tel ecom",
+      "▁aspir ations",
+      "▁aspiration s",
+      "▁ 乐",
+      "附 上",
+      "停 下来",
+      "▁s oo",
+      "▁so o",
+      "抗美 援朝",
+      "▁ad orn",
+      "nAss ume",
+      "第六十七条 第一款",
+      "第六十七 条第一款",
+      "▁Insp ired",
+      "▁vibr ations",
+      "▁vibration s",
+      "受 邀",
+      "多 变",
+      "逆 袭",
+      "好几 个",
+      "好 几个",
+      "sk ins",
+      "ski ns",
+      "skin s",
+      "s kins",
+      "▁Au burn",
+      "▁Aub urn",
+      "滤 芯",
+      "研 修",
+      "雇 主",
+      "ce a",
+      "c ea",
+      "双方 已",
+      "予以 冻结",
+      "然后 利用",
+      "ah ead",
+      "a head",
+      "ne lif",
+      "nel if",
+      "n elif",
+      "二〇一五年 四月",
+      "▁Min utes",
+      "▁ Minutes",
+      "最高人民法院关于审理 民间借贷",
+      "о р",
+      "一 试",
+      "燕 窝",
+      "跳 槽",
+      "La r",
+      "L ar",
+      "ja s",
+      "j as",
+      "立足 于",
+      "YW dl",
+      "▁f oul",
+      "▁fo ul",
+      "nAp ply",
+      "nApp ly",
+      "n Apply",
+      "在这种 情况下",
+      "在 这种情况下",
+      "▁ham let",
+      "▁not orious",
+      "体 征",
+      "的女 孩子",
+      "的女孩 子",
+      "的 女孩子",
+      "uit ies",
+      "ui ties",
+      "u ities",
+      "▁nit ro",
+      "▁ni tro",
+      "▁mechan ic",
+      "基 质",
+      "领 事",
+      "寄生 虫",
+      "培训 中心",
+      "工作 原理",
+      "承担 责任",
+      "▁g odd",
+      "▁go dd",
+      "▁god d",
+      "▁inf ancy",
+      "ASH INGTON",
+      "▁Elect oral",
+      "卫 国",
+      "在 南",
+      "并 把",
+      "相 约",
+      "IC C",
+      "I CC",
+      "不停 的",
+      "二年 级",
+      "二 年级",
+      "uc hs",
+      "uch s",
+      "债权 文书",
+      "学术 交流",
+      "D ummy",
+      "E arly",
+      "上 单调递增",
+      "▁p rose",
+      "▁pro se",
+      "▁pr ose",
+      "▁pros e",
+      "nC ategory",
+      "n Category",
+      "es ophageal",
+      "两 级",
+      "最 难",
+      "副 产品",
+      "的那 些",
+      "的 那些",
+      "）× （",
+      "） ×（",
+      "▁S eat",
+      "▁Se at",
+      "▁Sea t",
+      "▁wh ale",
+      "利 浦",
+      "we ep",
+      "w eep",
+      "re action",
+      "react ion",
+      "rea ction",
+      "▁ins pected",
+      "▁insp ected",
+      "▁inspect ed",
+      "▁am munition",
+      "安 抚",
+      "惬 意",
+      "cu b",
+      "c ub",
+      "▁B F",
+      "▁ BF",
+      "灵活 性",
+      "灵 活性",
+      "du it",
+      "d uit",
+      "实验 小学",
+      "虽然 没有",
+      "LOG IN",
+      "▁S oil",
+      "▁So il",
+      "▁l akh",
+      "▁la kh",
+      "lic ted",
+      "lict ed",
+      "l icted",
+      "▁g mail",
+      "▁ gmail",
+      "nNo vels",
+      "nNov els",
+      "Sim ulation",
+      "\") （",
+      "是 关键",
+      "pres ence",
+      "á g",
+      "永 乐",
+      "给 它",
+      "辱 骂",
+      "沙坪 坝",
+      "▁O bt",
+      "▁Ob t",
+      "▁Cl aude",
+      "▁Claud e",
+      "▁g ourmet",
+      "▁re union",
+      "▁reun ion",
+      "▁strateg ically",
+      "▁strategic ally",
+      "七八 糟",
+      "十三 五",
+      "十 三五",
+      "在 路上",
+      "▁b ei",
+      "▁be i",
+      "跨境 电商",
+      "Ne ill",
+      "▁m ang",
+      "▁man g",
+      "▁ma ng",
+      "▁m otto",
+      "▁mot to",
+      "▁mo tto",
+      "▁inconven ience",
+      "不 言",
+      "不 走",
+      "吓 得",
+      "非 遗",
+      "qt t",
+      "q tt",
+      "ip ay",
+      "ipa y",
+      "i pay",
+      "▁pres idents",
+      "▁president s",
+      "封 信",
+      "江 岸",
+      "降 落",
+      "百科 全书",
+      "nD iff",
+      "nDi ff",
+      "n Diff",
+      "剧 组",
+      "战 后",
+      "送 往",
+      "问 世",
+      "鲜 血",
+      "因此 而",
+      "责任 制",
+      "ánd ez",
+      "cri ptive",
+      "cript ive",
+      "▁Dipl oma",
+      "愚 蠢",
+      "滋 生",
+      "找到 一个",
+      "找 到一个",
+      "VI SED",
+      "VIS ED",
+      "wi ves",
+      "w ives",
+      "▁C aps",
+      "▁Ca ps",
+      "▁Cap s",
+      "nC hart",
+      "nCh art",
+      "nChar t",
+      "n Chart",
+      "二〇一七年 一月",
+      "家 境",
+      "降 压",
+      "做 单位",
+      "▁S kill",
+      "▁Sk ill",
+      "▁Ski ll",
+      "▁ Skill",
+      "▁A mount",
+      "▁Am ount",
+      "▁ Amount",
+      "▁prob ation",
+      "▁top ography",
+      "烧 开",
+      "超 时",
+      "威海 市",
+      "威 海市",
+      "ú mer",
+      "家庭 暴力",
+      "▁Veter inary",
+      "▁ 而",
+      "你 可",
+      "牙 龈",
+      "笑 笑",
+      "都 好",
+      "在 规定",
+      "方便 的",
+      "▁R is",
+      "▁Ri s",
+      "▁Mc Kin",
+      "▁McK in",
+      "▁high s",
+      "C ulture",
+      "▁cov alent",
+      "邑 县",
+      "顺 风",
+      "Pi x",
+      "P ix",
+      "是你 的",
+      "是 你的",
+      "▁L ob",
+      "▁Lo b",
+      "tl ine",
+      "t line",
+      "▁OP EN",
+      "▁ OPEN",
+      "nF iles",
+      "nFil es",
+      "nFile s",
+      "n Files",
+      "save fig",
+      "▁Math Jax",
+      "▁ammon ium",
+      "▁port folios",
+      "▁portfolio s",
+      "下 垂",
+      "地方 的",
+      "地 方的",
+      "奇怪 的",
+      "奇 怪的",
+      "di es",
+      "die s",
+      "d ies",
+      "▁I vy",
+      "▁Iv y",
+      "▁ Ivy",
+      "▁Y ak",
+      "▁Ya k",
+      "▁l ia",
+      "▁li a",
+      "▁ lia",
+      "存储 容量",
+      "▁p ane",
+      "▁pan e",
+      "▁pa ne",
+      "▁As sign",
+      "▁Ass ign",
+      "▁ Assign",
+      "▁cere visiae",
+      "线 圈",
+      "'] [",
+      "' ][",
+      "▁T ay",
+      "▁Ta y",
+      "具有 一定",
+      "具 有一定",
+      "分支 机构",
+      "本文 主要",
+      "data t",
+      "da tat",
+      "dat at",
+      "en tin",
+      "ent in",
+      "enti n",
+      "▁pred ators",
+      "▁prostagland in",
+      "现代 人",
+      "▁P EG",
+      "▁PE G",
+      "▁ PEG",
+      "m olecule",
+      "▁pl aster",
+      "▁plas ter",
+      "▁so othing",
+      "▁soo thing",
+      "▁conn ectors",
+      "▁connect ors",
+      "▁connector s",
+      "不受申请执行 期间的限制",
+      "设立 了",
+      "bol t",
+      "bo lt",
+      "b olt",
+      "▁g lm",
+      "▁gl m",
+      "▁ glm",
+      "▁f MRI",
+      "▁dis gu",
+      "nP ython",
+      "nPy thon",
+      "n Python",
+      "▁re make",
+      "▁rem ake",
+      "痴 呆",
+      "骗 局",
+      "所以 就",
+      "我们 现在",
+      "▁剧情 简介",
+      "▁mis use",
+      "▁con stituted",
+      "▁constit uted",
+      "▁constitu ted",
+      "▁constitute d",
+      "向 你",
+      "旺 季",
+      "对我 的",
+      "对 我的",
+      "精致 的",
+      "精 致的",
+      "▁E BV",
+      "▁EB V",
+      "销售 收入",
+      "AL LED",
+      "ALL ED",
+      "op hon",
+      "oph on",
+      "o phon",
+      "chan ics",
+      "Contin ue",
+      "▁central ly",
+      "▁centr ally",
+      "String Utils",
+      "在 全",
+      "她 还",
+      "远 东",
+      "下午 茶",
+      "两个 方面",
+      "Not ice",
+      "tra its",
+      "trait s",
+      "t raits",
+      "▁ro oft",
+      "▁roof t",
+      "▁shel ters",
+      "▁shelter s",
+      "▁consolid ated",
+      "X S",
+      "耳 边",
+      "该 片",
+      "se ss",
+      "ses s",
+      "s ess",
+      "▁Mar shal",
+      "▁Marsh al",
+      "▁Mars hal",
+      "▁ Marshal",
+      "▁c ations",
+      "▁cat ions",
+      "▁cation s",
+      "▁sign age",
+      "长 老",
+      "收音 机",
+      "甚至 连",
+      "▁C SR",
+      "▁CS R",
+      "▁I ps",
+      "om bies",
+      "omb ies",
+      "ombie s",
+      "▁El der",
+      "▁Eld er",
+      "eff iciency",
+      "挑 衅",
+      "牵 手",
+      "nE r",
+      "n Er",
+      "在 执行",
+      "直接 在",
+      "ric anes",
+      "ricane s",
+      "买 来",
+      "改 名",
+      "薪 水",
+      "CP A",
+      "C PA",
+      "ON D",
+      "O ND",
+      "也是 个",
+      "也 是个",
+      "帮助 你",
+      "近几年 来",
+      "近 几年来",
+      "▁B hat",
+      "▁Bh at",
+      "▁G odd",
+      "▁Go dd",
+      "▁God d",
+      "▁Sa cred",
+      "▁Sacr ed",
+      "▁Sac red",
+      "家 和",
+      "▁k r",
+      "▁ kr",
+      "你想 要",
+      "你 想要",
+      "nV al",
+      "n Val",
+      "▁第 三章",
+      "▁第三 章",
+      "▁ 第三章",
+      "教学 内容",
+      "ig non",
+      "ign on",
+      "被子 植物门",
+      "▁S alon",
+      "▁Sal on",
+      "▁Sa lon",
+      "Re ferences",
+      "Reference s",
+      "R eferences",
+      "东南 部",
+      "东 南部",
+      "巡视 员",
+      "很好 看",
+      "很 好看",
+      "go og",
+      "正常 运行",
+      "▁Can berra",
+      "▁retro fit",
+      "▁up stairs",
+      "▁Monten egro",
+      "gi d",
+      "g id",
+      "结果 和",
+      "▁创办 时间",
+      "ph ants",
+      "phant s",
+      "phan ts",
+      "▁voltage s",
+      "▁volt ages",
+      "▁Consult ant",
+      "存 货",
+      "心 仪",
+      "消 肿",
+      "▁F Y",
+      "下一 次",
+      "下 一次",
+      "\"\" )\\",
+      "\" \")\\",
+      "re ws",
+      "rew s",
+      "r ews",
+      "行业 标准",
+      "nG reek",
+      "nGr eek",
+      "▁F uzzy",
+      "本院依法 送达",
+      "湖南省 长沙市",
+      "▁nomin ees",
+      "▁nomine es",
+      "▁nominee s",
+      "▁physic ist",
+      "▁mob ilization",
+      "▁mobil ization",
+      "姨 妈",
+      "女 演员",
+      "不会 自动",
+      "不 会自动",
+      "系统 集成",
+      "买卖 合同纠纷",
+      "买卖合同 纠纷",
+      "priv ile",
+      "▁Con way",
+      "▁might y",
+      "nFe bruary",
+      "n February",
+      "交 错",
+      "壁 垒",
+      "既 不",
+      "端 子",
+      "活 下去",
+      "是我 们的",
+      "是我们 的",
+      "是 我们的",
+      "党史 学习教育",
+      "▁Recruit ment",
+      "▁Buff eredReader",
+      "▁ BufferedReader",
+      "丽 华",
+      "凯 旋",
+      "绿 水",
+      "▁sl iced",
+      "▁slic ed",
+      "▁slice d",
+      "第一百一十九 条第",
+      "第一百一十九条 第",
+      "▁vent ures",
+      "▁ven tures",
+      "▁venture s",
+      "h f",
+      "召 集",
+      "推 移",
+      "校 企",
+      "甄 嬛",
+      "轮 椅",
+      "工信 部",
+      "的 状况",
+      "ml in",
+      "m lin",
+      "lip sum",
+      "lips um",
+      "null ptr",
+      "▁DA MAGE",
+      "▁in dist",
+      "▁i ndist",
+      "▁ind ist",
+      "菩 提",
+      "ze w",
+      "z ew",
+      "▁K em",
+      "▁Ke m",
+      ")} $$\\",
+      ")}$ $\\",
+      ")}$$ \\",
+      ") }$$\\",
+      "mo ves",
+      "move s",
+      "mov es",
+      "m oves",
+      "kappa B",
+      "CON NECT",
+      "aggreg ate",
+      "▁house holder",
+      "▁household er",
+      "▁n ud",
+      "▁nu d",
+      "xml ns",
+      "仲裁 裁决书",
+      "仲裁裁决 书",
+      "nOct ober",
+      "n October",
+      "▁caps ules",
+      "▁capsule s",
+      "n protected",
+      "抓 手",
+      "王 永",
+      "缝 合",
+      "题 为",
+      "ox ia",
+      "o xia",
+      "▁M use",
+      "▁Mus e",
+      "▁Mu se",
+      "▁C rack",
+      "▁Cr ack",
+      "▁Cra ck",
+      "▁master piece",
+      "如 梦",
+      "还 应",
+      "ib al",
+      "iba l",
+      "i bal",
+      "公共 安全",
+      "PL ACE",
+      "PLA CE",
+      "act us",
+      "period ic",
+      "nStand ard",
+      "n Standard",
+      "▁Inf inite",
+      "enter prise",
+      "▁Prote stant",
+      "▁Protest ant",
+      "小 雨",
+      "工 夫",
+      "燕 麦",
+      "绒 毛",
+      "sa f",
+      "s af",
+      "价值 和",
+      "转 基因",
+      "sp am",
+      "▁d econ",
+      "▁de con",
+      "▁dec on",
+      "▁Min imal",
+      "▁Mini mal",
+      "▁Event Args",
+      "▁ EventArgs",
+      "▁ex pressive",
+      "▁exp ressive",
+      "▁express ive",
+      "▁expr essive",
+      "▁post menopausal",
+      "带 了",
+      "演 戏",
+      "Sal e",
+      "Sa le",
+      "S ale",
+      "▁Pl anck",
+      "▁Plan ck",
+      "Japan ese",
+      "J apanese",
+      "▁dis band",
+      "▁wood land",
+      "compl iance",
+      "▁Reserv oir",
+      "洗 礼",
+      "EO S",
+      "E OS",
+      "中文 系",
+      "规模 化",
+      "nint o",
+      "ni nto",
+      "nin to",
+      "n into",
+      "▁f req",
+      "▁fr eq",
+      "▁fre q",
+      "▁ freq",
+      "▁Ver de",
+      "▁Hit ler",
+      "▁Street s",
+      "▁Stre ets",
+      "▁visual s",
+      "pop ulations",
+      "population s",
+      "default Value",
+      "中国建设 银行股份有限公司",
+      "会 用",
+      "未 还",
+      "问 你",
+      "ee le",
+      "e ele",
+      "ge ar",
+      "g ear",
+      "▁M iy",
+      "▁Mi y",
+      "ac eted",
+      "ace ted",
+      "acet ed",
+      "ol ateral",
+      "olate ral",
+      "▁D ynasty",
+      "Throw able",
+      "放 了",
+      "更 需要",
+      "nd ra",
+      "tr ig",
+      "tri g",
+      "t rig",
+      "都是 一个",
+      "都是一 个",
+      "都 是一个",
+      "K afka",
+      "▁counsel ling",
+      "▁couns elling",
+      "伟 业",
+      "居 易",
+      "ch el",
+      "che l",
+      "c hel",
+      "证据 不足",
+      "nAd vert",
+      "nAdv ert",
+      "▁J ensen",
+      "▁Jen sen",
+      "▁calibr ated",
+      "曲 靖",
+      "近 乎",
+      "▁en listed",
+      "▁NE GLIGENCE",
+      "小 刚",
+      "小 号",
+      "满 脸",
+      "自 立",
+      "职业 学校",
+      "银行 贷款",
+      "▁B url",
+      "▁Bur l",
+      "▁Bu rl",
+      "sex ual",
+      "s exual",
+      "▁Act iv",
+      "二〇一四年 六月",
+      "comp iled",
+      "compile d",
+      "▁op ioids",
+      "▁opioid s",
+      "▁Document s",
+      "▁Doc uments",
+      "▁ Documents",
+      "白 细胞",
+      "负 事故",
+      "过了 一",
+      "过 了一",
+      "▁Tel ugu",
+      "▁f luent",
+      "▁fl uent",
+      "▁flu ent",
+      "▁ fluent",
+      "▁mush room",
+      "▁pr istine",
+      "▁Diagn osis",
+      "PO P",
+      "P OP",
+      "以及 在",
+      "di ag",
+      "疫情 期间",
+      "cc oli",
+      "▁App eal",
+      "▁Appe al",
+      "Pag ination",
+      "P agination",
+      "还会提供 一项住宿的免费政策",
+      "▁Re habilitation",
+      "执 保",
+      "母 校",
+      "TG F",
+      "T GF",
+      "um l",
+      "u ml",
+      "充分 不",
+      "现 如今",
+      "的事 儿",
+      "的 事儿",
+      "pe on",
+      "p eon",
+      "pr it",
+      "pri t",
+      "p rit",
+      "un def",
+      "und ef",
+      "unde f",
+      "u ndef",
+      "▁Red uce",
+      "▁M anning",
+      "▁Man ning",
+      "▁Mann ing",
+      "nMunicip alities",
+      "中 大",
+      "畜 禽",
+      "蜗 牛",
+      "题 干",
+      "发现 问题",
+      "▁Sh oes",
+      "▁Co verage",
+      "▁Cover age",
+      "▁Cove rage",
+      "▁ Coverage",
+      "K h",
+      "专 场",
+      "徒 步",
+      "我 先",
+      "cycl ic",
+      "c yclic",
+      "nSim ply",
+      "nSimpl y",
+      "▁cont ours",
+      "▁contour s",
+      "B t",
+      "仓 位",
+      "超 长",
+      "作者 的",
+      "作 者的",
+      "几年 前",
+      "几 年前",
+      "因数 是",
+      "因 数是",
+      "等 主编",
+      "顺利 的",
+      "顺 利的",
+      "▁T ox",
+      "▁To x",
+      "功 德",
+      "增 设",
+      "头 盔",
+      "地球 上",
+      "地铁 站",
+      "▁P sy",
+      "▁Ps y",
+      "严格 落实",
+      "存续 期间",
+      "金融 市场",
+      "▁T rying",
+      "▁Tr ying",
+      "▁Try ing",
+      "抛 开",
+      "VO L",
+      "V OL",
+      "信息 和",
+      "凝聚 力",
+      "pon t",
+      "po nt",
+      "p ont",
+      "洗浴 用品",
+      "▁Port o",
+      "▁Por to",
+      "▁ly ric",
+      "▁sc affolds",
+      "▁scaffold s",
+      "县人民检察院 指派检察员",
+      "( #",
+      "微 风",
+      "这 下",
+      "建筑 业",
+      "要不 是",
+      "要 不是",
+      "▁I TS",
+      "▁IT S",
+      "▁ ITS",
+      "司法 救助",
+      "更多 的人",
+      "更多的 人",
+      "▁文章 目录",
+      "▁cycl in",
+      "▁borrow ing",
+      "▁bor rowing",
+      "I r",
+      "双 子",
+      "一大 堆",
+      "妇 产科",
+      "玻璃 钢",
+      "EN SION",
+      "ENS ION",
+      "Un lock",
+      "▁d emos",
+      "▁de mos",
+      "▁dem os",
+      "▁demo s",
+      "期限 不得超过",
+      "press or",
+      "presso r",
+      "p ressor",
+      "道 上",
+      "п ис",
+      "Go al",
+      "Pol l",
+      "Po ll",
+      "P oll",
+      "un di",
+      "und i",
+      "u ndi",
+      "并提供 了",
+      "并 提供了",
+      "▁Col ony",
+      "▁Colon y",
+      "虽 有",
+      "▁n r",
+      "▁ nr",
+      "付出 了",
+      "付 出了",
+      "dis p",
+      "di sp",
+      "d isp",
+      "so lo",
+      "sol o",
+      "s olo",
+      "ion ate",
+      "iona te",
+      "i onate",
+      "it erate",
+      "ite rate",
+      "iter ate",
+      "ser ving",
+      "serv ing",
+      "s erving",
+      "▁Che ese",
+      "▁oun ces",
+      "▁ounce s",
+      "▁diss ipation",
+      "▁dissip ation",
+      "拥 堵",
+      "成本 的",
+      "成 本的",
+      "时间 里",
+      "}/ >\\",
+      "} />\\",
+      "特别 喜欢",
+      "ocr acy",
+      "▁We iss",
+      "▁Wei ss",
+      "Expl orer",
+      "几 度",
+      "重 置",
+      "ov á",
+      "uz i",
+      "u zi",
+      "vi l",
+      "v il",
+      "▁B TC",
+      "▁BT C",
+      "▁g ynec",
+      "▁gy nec",
+      "▁La ravel",
+      "▁Lar avel",
+      "▁chronic ally",
+      "▁chron ically",
+      "无 损",
+      "沼 气",
+      "调 入",
+      "__ ,\\",
+      "__, \\",
+      "个人 感觉",
+      "ag ging",
+      "agg ing",
+      "ys tick",
+      "yst ick",
+      "y stick",
+      "ro physics",
+      "rophys ics",
+      "▁Eff iciency",
+      "及其 它",
+      "及 其它",
+      "才会 有",
+      "才 会有",
+      "来说 是",
+      "来 说是",
+      "第二 批",
+      "nH el",
+      "nHe l",
+      "n Hel",
+      "▁I rr",
+      "▁Ir r",
+      "▁F alk",
+      "▁Fal k",
+      "ec olor",
+      "eco lor",
+      "e color",
+      "se ller",
+      "sel ler",
+      "sell er",
+      "s eller",
+      "▁Ni elsen",
+      "▁dist ort",
+      "▁dop aminergic",
+      "成 大",
+      "水 壶",
+      "干净 的",
+      "风险 的",
+      "的是 一个",
+      "的 是一个",
+      "MENT S",
+      "M ENTS",
+      "nend if",
+      "n endif",
+      "欧洲 冠军联赛",
+      "Up grade",
+      "▁per sec",
+      "▁pers ec",
+      "▁comprom ising",
+      "B H",
+      "反 面",
+      "正 向",
+      "缩 减",
+      "▁d z",
+      "▁ dz",
+      "意 想不到",
+      "al ach",
+      "ala ch",
+      "▁B are",
+      "▁Bar e",
+      "▁Ba re",
+      "▁sex es",
+      "▁Rel ief",
+      "▁turb ines",
+      "▁turbine s",
+      "较好 地",
+      "br oad",
+      "bro ad",
+      "b road",
+      "Def ine",
+      "De fine",
+      "▁comm unal",
+      "▁commun al",
+      "▁advance ments",
+      "▁advancement s",
+      "出 一",
+      "庄 子",
+      "新 安",
+      "脱 口",
+      "乡 政府",
+      "id le",
+      "idl e",
+      "nm ap",
+      "n map",
+      "知识 分子",
+      "饮食 习惯",
+      "▁S eal",
+      "▁Se al",
+      "▁Sea l",
+      "nA ward",
+      "nAw ard",
+      "▁f inely",
+      "▁fin ely",
+      "▁fine ly",
+      "▁onto logy",
+      "▁ont ology",
+      "▁ ontology",
+      "▁dermat itis",
+      "F H",
+      "ä ä",
+      "德 语",
+      "ü ss",
+      "能 满足",
+      "可能 就是",
+      "dy nam",
+      "dyn am",
+      "d ynam",
+      "▁D anger",
+      "▁Dan ger",
+      "▁He aders",
+      "▁Head ers",
+      "▁Header s",
+      "▁ Headers",
+      "▁She pherd",
+      "▁cer amics",
+      "▁ceramic s",
+      "▁mis conduct",
+      "▁miscon duct",
+      "之 门",
+      "估 价",
+      "所 占",
+      "深 造",
+      "稀 缺",
+      "购 销",
+      "辗 转",
+      "OM E",
+      "O ME",
+      "nR ock",
+      "n Rock",
+      "Att ack",
+      "nRes ult",
+      "n Result",
+      "▁sa usage",
+      "满 怀",
+      "qu iz",
+      "▁E MP",
+      "▁EM P",
+      "▁ EMP",
+      "Comp act",
+      "存 量",
+      "鲜 肉",
+      "▁m V",
+      "▁ mV",
+      "▁I BD",
+      "▁IB D",
+      "▁K ras",
+      "▁Kr as",
+      "▁Kra s",
+      "▁T rim",
+      "▁Tr im",
+      "▁Tri m",
+      "▁ Trim",
+      "▁ce phal",
+      "▁cep hal",
+      "▁ cephal",
+      "▁Al ready",
+      "▁Che shire",
+      "▁Ches hire",
+      "▁Life time",
+      "▁Lif etime",
+      "宜 居",
+      "那 份",
+      "最 准确",
+      "追究 其",
+      "·· ··",
+      "el ier",
+      "eli er",
+      "e lier",
+      "pass wd",
+      "并 列",
+      "是 该",
+      "锦 赛",
+      "在 设计",
+      "em as",
+      "ema s",
+      "e mas",
+      "迅速 发展",
+      "▁S aid",
+      "▁Sa id",
+      "▁All ied",
+      "▁W olver",
+      "▁Wol ver",
+      "▁short cuts",
+      "▁shortcut s",
+      "▁ shortcuts",
+      "▁Confed erate",
+      "字 样",
+      "民 辖",
+      "给 力",
+      "阴 凉",
+      "本文 对",
+      "▁'@ /",
+      "__ ___",
+      "____ _",
+      "___ __",
+      "_ ____",
+      "▁G mail",
+      "▁amp hib",
+      "▁amph ib",
+      "今 夜",
+      "鲜 活",
+      "▁F eld",
+      "▁Fe ld",
+      "▁Fel d",
+      "▁front ier",
+      "▁advis able",
+      "东 昌",
+      "停 产",
+      "红 米",
+      "造 谣",
+      "） ;",
+      "体系 和",
+      "最 关键",
+      "tb ody",
+      "t body",
+      "提供 了一个",
+      "提供了 一个",
+      "具备执行条件 时",
+      "▁de coded",
+      "▁dec oded",
+      "▁decode d",
+      "内 障",
+      "峨 眉",
+      "sh u",
+      "s hu",
+      ")* *\\",
+      ")** \\",
+      ") **\\",
+      "nE ver",
+      "nEv er",
+      "nc ould",
+      "n could",
+      "t ensive",
+      "▁prefer able",
+      "侵 袭",
+      "图 上",
+      "型 材",
+      "大 头",
+      "形 体",
+      "葱 姜",
+      "血 腥",
+      "UA L",
+      "U AL",
+      "▁H OW",
+      "▁HO W",
+      "▁ HOW",
+      "QU ERY",
+      "QUE RY",
+      "▁s pear",
+      "▁sp ear",
+      "▁spe ar",
+      "style Urls",
+      "号刑事附带民事 判决",
+      "▁diagn osing",
+      "大 招",
+      "小 黑",
+      "摄 取",
+      "纸 箱",
+      "非 得",
+      "$, \\",
+      "$ ,\\",
+      "▁P OP",
+      "▁PO P",
+      "▁ POP",
+      "▁S TD",
+      "▁ST D",
+      "▁ STD",
+      "▁Cl ients",
+      "▁Client s",
+      "▁ Clients",
+      "▁sl ipped",
+      "▁slip ped",
+      "▁elev ations",
+      "▁elevation s",
+      "监 事",
+      "酸 甜",
+      "支持 其",
+      "}) $\\",
+      "})$ \\",
+      "} )$\\",
+      "自首 情节",
+      "cos mos",
+      "▁p ouch",
+      "▁po uch",
+      "▁pou ch",
+      "▁Re agan",
+      "▁hyper gly",
+      "▁s ourcing",
+      "▁sour cing",
+      "伯 格",
+      "侵 占",
+      "根 治",
+      "遗 弃",
+      "▁P N",
+      "▁ PN",
+      "pro n",
+      "pr on",
+      "p ron",
+      "in fer",
+      "inf er",
+      "年龄 较大的",
+      "日 刑满释放",
+      "▁height ened",
+      "▁res idences",
+      "▁resid ences",
+      "▁residence s",
+      "反 省",
+      "液 的",
+      "物 语",
+      "', $",
+      "' ,$",
+      "▁ 办学",
+      "▁ 距离",
+      "审理 期间",
+      "积极 探索",
+      "▁Sh ak",
+      "▁Sha k",
+      "▁V inc",
+      "▁Vin c",
+      "▁Vi nc",
+      "的法定 条件",
+      "os ystem",
+      "o system",
+      "find ById",
+      "▁Enh anced",
+      "协助执行 单位出具的财产查询回执",
+      "变 大",
+      "多 一点",
+      "▁L ys",
+      "▁Ly s",
+      "▁H IGH",
+      "▁HI GH",
+      "▁a val",
+      "▁av al",
+      "▁ aval",
+      "时不 需付费",
+      "pixel s",
+      "pix els",
+      "p ixels",
+      "▁De nis",
+      "▁Den is",
+      "城市 人民法院",
+      "城 市人民法院",
+      "str ength",
+      "stre ngth",
+      "▁conc ave",
+      "▁un pleasant",
+      "▁supplement al",
+      "姥 姥",
+      "苍 穹",
+      "创业 的",
+      "创 业的",
+      "开封 市",
+      "点 个赞",
+      "狐 月山",
+      "nar gs",
+      "n args",
+      "Mon day",
+      "M onday",
+      "nG raph",
+      "nGr aph",
+      "n Graph",
+      "约 是",
+      "罗 汉",
+      "而 至",
+      "订 货",
+      "nB ox",
+      "nBo x",
+      "n Box",
+      "wit z",
+      "wi tz",
+      "w itz",
+      "或者 约定",
+      "ov able",
+      "ova ble",
+      "▁mer cy",
+      "▁merc y",
+      "▁man tle",
+      "▁mant le",
+      "▁text ured",
+      "▁texture d",
+      "几 人",
+      "地 势",
+      "智 利",
+      "穿 衣",
+      "愉快 的",
+      "愉 快的",
+      "的 表情",
+      "ee ze",
+      "make at",
+      "mak eat",
+      "id ental",
+      "ident al",
+      "iden tal",
+      "▁Cou pon",
+      "▁arsen ic",
+      "▁C riteria",
+      "▁Crit eria",
+      "▁ Criteria",
+      "▁int raper",
+      "▁intr aper",
+      "▁intra per",
+      "该 文",
+      "▁ 颜色",
+      "he ed",
+      "hee d",
+      "h eed",
+      "ook e",
+      "oo ke",
+      "o oke",
+      "家庭 教育",
+      "▁Le vi",
+      "▁Lev i",
+      "▁Ky oto",
+      "▁St okes",
+      "作为 被执行人的",
+      "▁dess erts",
+      "▁dessert s",
+      "▁renov ations",
+      "▁renovation s",
+      "大 意",
+      "昆山 市",
+      "昆 山市",
+      "骨质 疏松",
+      "oc ice",
+      "oci ce",
+      "▁qu ench",
+      "▁que nch",
+      "▁com posing",
+      "▁comp osing",
+      "▁compos ing",
+      "五 次",
+      "音 量",
+      "也不 想",
+      "也 不想",
+      "au ff",
+      "auf f",
+      "a uff",
+      "fr esh",
+      "fre sh",
+      "f resh",
+      "▁MA PK",
+      "▁MAP K",
+      "▁c ereal",
+      "▁cere al",
+      "▁cer eal",
+      "▁ce real",
+      "掉 落",
+      "深入 浅",
+      "受欢迎 的",
+      "ot helial",
+      "▁peri operative",
+      "G y",
+      "不 只",
+      "回 想",
+      "的 联系",
+      "一段 文字",
+      "总体 规划",
+      "▁L act",
+      "▁La ct",
+      "▁Lac t",
+      "▁Scient ists",
+      "▁Scientist s",
+      "预 装",
+      "最小 公",
+      "的 关注",
+      "问 一下",
+      "mq tt",
+      "m qtt",
+      "▁g ust",
+      "▁gu st",
+      "▁ gust",
+      "住 黑龙江省",
+      "▁Mon key",
+      "▁Z ambia",
+      "▁Zam bia",
+      "▁pr aying",
+      "▁pra ying",
+      "▁pray ing",
+      "nCh urches",
+      "▁Pro motion",
+      "▁Prom otion",
+      "一 再",
+      "学 区",
+      "的 商",
+      "▁C PR",
+      "▁CP R",
+      "▁A uction",
+      "▁Au ction",
+      "▁Wil helm",
+      "▁d ilation",
+      "▁di lation",
+      "▁dil ation",
+      "是连载于 晋江文学城",
+      "▁conjug ated",
+      "▁conjugate d",
+      "▁manif ested",
+      "▁manifest ed",
+      "▁interview ing",
+      "剥 削",
+      "县 域",
+      "见 习",
+      "▁E G",
+      "▁ EG",
+      "原 发性",
+      "▁ad mir",
+      "▁adm ir",
+      "blast oma",
+      "▁Har vest",
+      "▁conduct s",
+      "时 隔",
+      "Ka r",
+      "K ar",
+      "生动 的",
+      "生 动的",
+      "的 危害",
+      "的大 部分",
+      "的 大部分",
+      "▁Roose velt",
+      "光 环",
+      "多 岁",
+      "特 工",
+      "Ki m",
+      "K im",
+      "作品 集",
+      "▁H IS",
+      "▁HI S",
+      "满足 条件",
+      "nL ists",
+      "nList s",
+      "n Lists",
+      "med icine",
+      "▁Tr acker",
+      "▁Track er",
+      "▁ Tracker",
+      "▁mag nets",
+      "▁magn ets",
+      "▁magnet s",
+      "▁add itives",
+      "▁additive s",
+      "▁General ized",
+      "▁ А",
+      "可 与",
+      "解 脱",
+      "瑞 安市",
+      "英烈 网",
+      "▁A DP",
+      "▁AD P",
+      "chr ane",
+      "▁A ston",
+      "▁As ton",
+      "▁Ast on",
+      "▁M itch",
+      "▁Mit ch",
+      "▁St ark",
+      "▁Star k",
+      "▁Sta rk",
+      "经过 财产调查",
+      "est imate",
+      "estim ate",
+      "▁dis card",
+      "▁disc ard",
+      "X i",
+      "功 用",
+      "圣 人",
+      "FD A",
+      "F DA",
+      "和 学生",
+      "ec ca",
+      "ecc a",
+      "给定 一个",
+      "▁b ait",
+      "▁ba it",
+      "房地产 市场",
+      "re ally",
+      "real ly",
+      "nd epend",
+      "nde pend",
+      "n depend",
+      "▁L imits",
+      "▁Lim its",
+      "▁Li mits",
+      "▁Limit s",
+      "▁br illi",
+      "▁Arab ian",
+      "▁Arabia n",
+      "▁file Path",
+      "▁ filePath",
+      "z x",
+      "▁M H",
+      "▁ MH",
+      "▁来 源",
+      "▁ 来源",
+      ">\" >\\",
+      "> \">\\",
+      "ol ev",
+      "ole v",
+      "o lev",
+      "pe ek",
+      "p eek",
+      "▁s op",
+      "▁so p",
+      "▁Cand ida",
+      "▁Candid a",
+      "▁Pro cedure",
+      "▁Ge orgetown",
+      "▁George town",
+      "▁redund ancy",
+      "地 坪",
+      "外 貌",
+      "和 提高",
+      "这个 名字",
+      "国家 统计局",
+      "er mann",
+      "erm ann",
+      "erman n",
+      "▁bre wing",
+      "▁brew ing",
+      "nBuild ing",
+      "n Building",
+      "▁Norm ally",
+      "▁Normal ly",
+      "催 款",
+      "绝 佳",
+      "采 收",
+      "很 开心",
+      "es ton",
+      "est on",
+      "esto n",
+      "e ston",
+      "▁L unch",
+      "▁Lu nch",
+      "as signed",
+      "ass igned",
+      "assign ed",
+      "Co ordinate",
+      "Coord inate",
+      "所 学",
+      "更 像",
+      "碳水 化合物",
+      "WITH OUT",
+      "W ITHOUT",
+      "▁No body",
+      "▁Nob ody",
+      "▁asym met",
+      "で き",
+      "守 望",
+      "就 近",
+      "GT M",
+      "G TM",
+      "等多 项",
+      "等 多项",
+      "公司 拥有",
+      "可能 出现",
+      "▁cl osures",
+      "▁clos ures",
+      "▁closure s",
+      "▁ closures",
+      "台 币",
+      "大 涨",
+      "年 金",
+      "戒 毒",
+      "是一 门",
+      "是 一门",
+      "第一 期",
+      "第 一期",
+      "ch alk",
+      "nf ull",
+      "n full",
+      "sever ity",
+      "▁Tunis ia",
+      "` <",
+      "▁ 词",
+      "害 羞",
+      "院 的",
+      "ло в",
+      "л ов",
+      "udd le",
+      "▁E lim",
+      "▁El im",
+      "▁Eli m",
+      "▁s cler",
+      "▁sc ler",
+      "Co verage",
+      "Cover age",
+      "▁j ackets",
+      "▁jack ets",
+      "▁jacket s",
+      "▁Man agers",
+      "▁Manager s",
+      "▁Manag ers",
+      "▁Manage rs",
+      "撤 离",
+      "杂 草",
+      "也不 敢",
+      "也 不敢",
+      "营商 环境",
+      "the ir",
+      "▁vac u",
+      "▁va cu",
+      "采取保全措施后 三十日内",
+      "▁partic ulate",
+      "标 价",
+      "范 例",
+      "霉 菌",
+      "▁P é",
+      "优秀 奖",
+      "要 考虑",
+      "nS ay",
+      "n Say",
+      "国土 面积",
+      "▁Pen ny",
+      "▁Penn y",
+      "rom ycin",
+      "r omycin",
+      "▁p acing",
+      "▁pa cing",
+      "▁pac ing",
+      "▁w ikipedia",
+      "▁ wikipedia",
+      "天 成",
+      "纷 争",
+      "_) ;\\",
+      "_ );\\",
+      "劳动 模范",
+      "圆的 周长",
+      "十分 重要的",
+      "d ollar",
+      "▁B ella",
+      "▁Be lla",
+      "▁Bel la",
+      "▁Bell a",
+      "el apsed",
+      "▁super hero",
+      "救 命",
+      "无 须",
+      "荷 叶",
+      "天花 板",
+      "▁re visit",
+      "▁revis it",
+      "▁up regulated",
+      "扎 根",
+      "能 做",
+      "be g",
+      "b eg",
+      "为 确保",
+      "的 生长",
+      "开发 利用",
+      "nChe ers",
+      "ple ctic",
+      "p lectic",
+      "▁note books",
+      "▁noteb ooks",
+      "▁notebook s",
+      "刚 性",
+      "▁重 量",
+      "▁ 重量",
+      "Pi ng",
+      "Pin g",
+      "P ing",
+      "师资 力量",
+      "▁n icer",
+      "▁nice r",
+      "▁nic er",
+      "▁ni cer",
+      "▁B oulder",
+      "▁mon arch",
+      "trans pose",
+      "聚 氨",
+      "的 少年",
+      "es on",
+      "e son",
+      "▁U CI",
+      "▁UC I",
+      "企业 提供",
+      "各行各 业",
+      "▁w orms",
+      "▁wor ms",
+      "▁worm s",
+      "▁Air ways",
+      "l ı",
+      "干 啥",
+      "or io",
+      "ori o",
+      "o rio",
+      "nWeek ly",
+      "▁\\' \\'\\'\\",
+      "▁\\'\\'\\ '\\",
+      "▁\\'\\' \\'\\",
+      "▁ \\'\\'\\'\\",
+      "▁text ual",
+      "▁tex tual",
+      "▁follic les",
+      "▁follicle s",
+      "▁prere quisite",
+      "交 货",
+      "木 头",
+      "祈 祷",
+      "请 看",
+      "酱 香",
+      "▁R é",
+      "不 那么",
+      "出去 玩",
+      "刺激 性",
+      "当 一个",
+      "▁A th",
+      "▁At h",
+      "▁H IF",
+      "▁HI F",
+      "侵权 行为",
+      "▁Response s",
+      "▁Respon ses",
+      "▁Respons es",
+      "▁ ⟨",
+      "药 学",
+      "gr ab",
+      "gra b",
+      "g rab",
+      "临床 症状",
+      "浪费 时间",
+      "精益 求精",
+      "▁circ a",
+      "▁cir ca",
+      "▁HO WEVER",
+      "sc heduled",
+      "schedule d",
+      "▁F letcher",
+      "▁ 民",
+      "八 一",
+      "虹 桥",
+      "遵义 市",
+      "新开 传奇",
+      "▁roof s",
+      "▁cond ensation",
+      "▁condens ation",
+      "da la",
+      "dal a",
+      "d ala",
+      "的人 力资源",
+      "的 人力资源",
+      "▁M egan",
+      "▁Me gan",
+      "▁Meg an",
+      "▁Mega n",
+      "小 猪",
+      "毕 节",
+      "的 实施",
+      "▁H ou",
+      "▁Ho u",
+      "ci pher",
+      "c ipher",
+      "off line",
+      "不 问",
+      "射 频",
+      "打 球",
+      "跟 鞋",
+      "CD C",
+      "C DC",
+      "re ts",
+      "ret s",
+      "r ets",
+      "um py",
+      "ump y",
+      "决定 执行",
+      "第二 十九",
+      "nBl ue",
+      "n Blue",
+      "Al gebra",
+      "▁hel ium",
+      "▁ helium",
+      "▁car rots",
+      "▁Enter prises",
+      "▁Enterprise s",
+      "▁vari ational",
+      "▁variation al",
+      "▁成 都",
+      "▁ 成都",
+      "会 因为",
+      "在本 次",
+      "在 本次",
+      "▁BR CA",
+      "tfor each",
+      "t foreach",
+      "call backs",
+      "callback s",
+      "▁am orphous",
+      "是在法律允许 的范围内",
+      "中 是",
+      "地 审查",
+      "日用 品",
+      "日 用品",
+      "课堂 上",
+      "ogen y",
+      "oge ny",
+      "▁d eps",
+      "▁de ps",
+      "▁dep s",
+      "▁ deps",
+      "▁Franç ois",
+      "▁re opened",
+      "▁reopen ed",
+      "对 边",
+      "新 星",
+      "求 生",
+      "额 头",
+      "家 公司",
+      "欧洲 三大",
+      "年 参演电影",
+      "\\\\\", \\\\\"",
+      "▁bul lets",
+      "▁bull ets",
+      "▁bullet s",
+      "听 得",
+      "颠 倒",
+      "Pe d",
+      "P ed",
+      "与 实践",
+      "则 需要",
+      "心灵 的",
+      "Be ing",
+      "▁te nor",
+      "▁ten or",
+      "nR unning",
+      "nRun ning",
+      "n Running",
+      "▁manual s",
+      "Prop Types",
+      "▁voc ational",
+      "一 族",
+      "企 鹅",
+      "资源 共享",
+      "▁pen ny",
+      "▁penn y",
+      "▁Tele graph",
+      "▁all ogeneic",
+      "▁prost hetic",
+      "▁Opportun ity",
+      "一般 都",
+      "交警 大队",
+      "▁Qt Core",
+      "▁ QtCore",
+      "▁Tit ans",
+      "▁Titan s",
+      "send Message",
+      "▁commod ities",
+      "充 沛",
+      "丝绸 之路",
+      "二〇一七年 八",
+      "终结本案 的本次执行程序",
+      "▁cat astrophic",
+      "▁atheros clerotic",
+      "沿 途",
+      "版 纸",
+      "电 费",
+      "他 指出",
+      "Er ic",
+      "E ric",
+      "▁A CS",
+      "▁AC S",
+      "▁ ACS",
+      "▁O CT",
+      "▁OC T",
+      "适用 范围",
+      "oc ode",
+      "oco de",
+      "o code",
+      "ynth ia",
+      "▁Mur der",
+      "▁immun ologic",
+      "▁immuno logic",
+      "▁Condition ing",
+      "桂 园",
+      "盐 城",
+      "简 算",
+      "▁L é",
+      "he tti",
+      "het ti",
+      "h etti",
+      "en ches",
+      "enc hes",
+      "ench es",
+      "▁F erry",
+      "▁Fer ry",
+      "▁Ferr y",
+      "开 个",
+      "结 识",
+      "绞 痛",
+      "ny a",
+      "n ya",
+      "▁B K",
+      "▁ BK",
+      "都有 着",
+      "都 有着",
+      "ha bi",
+      "hab i",
+      "h abi",
+      "还款 期限",
+      "Art ist",
+      "▁mi RNAs",
+      "▁miRNA s",
+      "vari ants",
+      "variant s",
+      "律师事务所 实习律师",
+      "▁Un limited",
+      "确有悔改表现 的事实清楚",
+      "能 给",
+      "要 让",
+      "▁D SM",
+      "▁DS M",
+      "在一起 的",
+      "在一 起的",
+      "▁软件 名称",
+      "▁p itches",
+      "▁pitch es",
+      "▁pit ches",
+      "▁immers ive",
+      "rupted Exception",
+      "云 飞",
+      "大 宗",
+      "太 差",
+      "狠 抓",
+      "鸽 子",
+      "ca b",
+      "c ab",
+      "引导 学生",
+      "忘记 密码",
+      "▁g own",
+      "▁go wn",
+      "ming ton",
+      "m ington",
+      "▁p antry",
+      "▁pan try",
+      "▁pa ntry",
+      "▁pant ry",
+      "承担连带 清偿责任",
+      "▁recip roc",
+      "cloud flare",
+      "宝 典",
+      "[] )\\",
+      "[]) \\",
+      "[ ])\\",
+      "nN ews",
+      "nNew s",
+      "nNe ws",
+      "n News",
+      "▁Mc Cl",
+      "▁McC l",
+      "对该犯 减刑",
+      "的经营 理念",
+      "的 经营理念",
+      "▁ur eter",
+      "▁ure ter",
+      "nPro gram",
+      "nPr ogram",
+      "n Program",
+      "▁rev ital",
+      "MD A",
+      "M DA",
+      "cast s",
+      "ca sts",
+      "cas ts",
+      "c asts",
+      "Euro pe",
+      "E urope",
+      "▁h omic",
+      "▁hom ic",
+      "▁ho mic",
+      "anal ytic",
+      "▁Sh erman",
+      "▁Sher man",
+      "▁p opping",
+      "▁pop ping",
+      "▁po pping",
+      "▁Philipp e",
+      "▁Philip pe",
+      "同 月",
+      "石 狮",
+      "防 卫",
+      "的 宗旨",
+      "▁c ops",
+      "▁co ps",
+      "▁cop s",
+      "▁g zip",
+      "est ead",
+      "este ad",
+      "e stead",
+      "▁W ander",
+      "▁Wa nder",
+      "▁Wan der",
+      "▁Dom estic",
+      "出 动",
+      "呼 气",
+      "巴 勒",
+      "幽 灵",
+      "思 义",
+      "钱 了",
+      "mo on",
+      "m oon",
+      "en ity",
+      "eni ty",
+      "ip ine",
+      "i pine",
+      "▁Y aml",
+      "▁Yam l",
+      "▁Ya ml",
+      "ー ト",
+      "导 热",
+      "昨 夜",
+      "是 中",
+      "了一 位",
+      "了 一位",
+      "武汉 大学",
+      "ome ncl",
+      "omen cl",
+      "ife rous",
+      "ifer ous",
+      "▁bad ges",
+      "▁badge s",
+      "▁ badges",
+      "▁ple ural",
+      "▁sem igroup",
+      "▁semi group",
+      "与 前",
+      "修 身",
+      "都 快",
+      "门 面",
+      "op re",
+      "o pre",
+      "od ore",
+      "odo re",
+      "odor e",
+      "wa ves",
+      "wave s",
+      "wav es",
+      "w aves",
+      "▁Mic key",
+      "▁Mick ey",
+      "▁troubles hooting",
+      "一 趟",
+      "连续 的",
+      "▁Rec ogn",
+      "▁Tal iban",
+      "长 和",
+      "） ∵",
+      "▁w o",
+      "▁ wo",
+      "记忆 力",
+      "IFI ED",
+      "ag ara",
+      "aga ra",
+      "agar a",
+      "gh ost",
+      "g host",
+      "te xit",
+      "tex it",
+      "t exit",
+      "▁Quant ity",
+      "▁ Quantity",
+      "下 限",
+      "黄 岛",
+      "ер е",
+      "е ре",
+      "▁爱 情",
+      "▁ 爱情",
+      "绝对 的",
+      "绝 对的",
+      "突出 贡献",
+      "▁al ph",
+      "▁ alph",
+      "▁Ch lor",
+      "{ ,",
+      "了 其",
+      "%) \\",
+      "% )\\",
+      "联系 的",
+      "联 系的",
+      "ir an",
+      "ira n",
+      "i ran",
+      "可以 做到",
+      "可以做 到",
+      "og raf",
+      "ogr af",
+      "o graf",
+      "▁m ason",
+      "▁ma son",
+      "▁mas on",
+      "C i",
+      "三 江",
+      "常 温",
+      "弱 点",
+      "水 墨",
+      "而 为",
+      "与被告 陈",
+      "▁d ass",
+      "▁da ss",
+      "▁das s",
+      "▁d umps",
+      "▁dump s",
+      "▁ dumps",
+      "cred ible",
+      "▁see kers",
+      "▁seek ers",
+      "▁Process es",
+      "▁Proc esses",
+      "放 缓",
+      "LAB EL",
+      "ric ula",
+      "▁flux es",
+      ": *",
+      "介 于",
+      "每 到",
+      "表 层",
+      "▁第 七",
+      "▁ 第七",
+      "不仅 有",
+      "不 仅有",
+      "不好 看",
+      "不 好看",
+      "双方 均",
+      "正式 版",
+      "be lt",
+      "bel t",
+      "b elt",
+      "▁l aps",
+      "▁la ps",
+      "▁lap s",
+      "un geon",
+      "ung eon",
+      "unge on",
+      "(\\' /\\',",
+      "(\\'/ \\',",
+      "Mat ches",
+      "Match es",
+      "M atches",
+      "Start up",
+      "▁Dor othy",
+      "▁ign ition",
+      "л ю",
+      "未 有",
+      "顽 强",
+      "od us",
+      "▁M ESS",
+      "▁ME SS",
+      "▁ MESS",
+      "▁T akes",
+      "▁Take s",
+      "▁Ta kes",
+      "▁Tak es",
+      "申请 撤回对被告",
+      "申请撤回 对被告",
+      "令 我",
+      "省 心",
+      "▁O PT",
+      "▁OP T",
+      "▁ OPT",
+      "在执行 中",
+      "在 执行中",
+      "长期 从事",
+      "▁n mol",
+      "▁nm ol",
+      "▁s ess",
+      "▁se ss",
+      "▁ses s",
+      "▁ sess",
+      "▁ar ach",
+      "▁wh ats",
+      "▁what s",
+      "arr ants",
+      "arrant s",
+      "▁Att ention",
+      "[ ,",
+      "AD O",
+      "A DO",
+      "模式 下",
+      "cc cc",
+      "ccc c",
+      "c ccc",
+      "可以 有效",
+      "可以有 效",
+      "经相关部门 批准后",
+      "下 乡",
+      "佳 能",
+      "lo y",
+      "l oy",
+      "分子 式",
+      "台湾 的",
+      "对于 一个",
+      "nFe el",
+      "iss ors",
+      "▁Gr aves",
+      "▁Gra ves",
+      "▁St akes",
+      "▁Sta kes",
+      "▁li tres",
+      "▁lit res",
+      "▁litre s",
+      "▁contin gent",
+      "▁conting ent",
+      "▁micro structure",
+      "堕 落",
+      "▁ 湖北",
+      "开发 了",
+      "开 发了",
+      "日期 为",
+      "日 期为",
+      "▁F ee",
+      "▁Fe e",
+      "▁ Fee",
+      "乱 七八糟",
+      "有异议 的",
+      "▁IN ST",
+      "▁ INST",
+      "▁pl asm",
+      "▁plas m",
+      "▁orig inate",
+      "▁origin ate",
+      "▁dissemin ated",
+      "丁 香",
+      "之 争",
+      "▁o st",
+      "▁os t",
+      "▁ ost",
+      "以上 学历",
+      "文明 单位",
+      "pat rick",
+      "▁Bav aria",
+      "▁Co chrane",
+      "▁Oper ators",
+      "▁Operator s",
+      "▁Collabor ative",
+      "我 司",
+      "真 挚",
+      "mn t",
+      "m nt",
+      "即可 得解",
+      "环境 影响",
+      "D ream",
+      "▁英文 参考",
+      "▁pre frontal",
+      "一 新",
+      "谁 家",
+      "收费 标准",
+      "Con nor",
+      "Conn or",
+      "▁As sets",
+      "▁Ass ets",
+      "▁Asset s",
+      "▁ Assets",
+      "▁STR ICT",
+      "nSc ience",
+      "n Science",
+      "▁wr inkles",
+      "▁ 易",
+      "入 味",
+      "后 有",
+      "试 探",
+      "赢 家",
+      "那么 的",
+      "深入 了解",
+      "网络 游戏",
+      "该行 行长",
+      "Re verse",
+      "Rever se",
+      "▁pre fers",
+      "▁prefer s",
+      "▁ste rile",
+      "▁ster ile",
+      "▁vill ain",
+      "▁villa in",
+      "▁benef ited",
+      "▁benefit ed",
+      "▁forward ed",
+      "列 宁",
+      "纬 度",
+      "希望 对",
+      "时间 短",
+      "Mat t",
+      "Ma tt",
+      "M att",
+      "col e",
+      "co le",
+      "c ole",
+      "▁H ipp",
+      "▁Hi pp",
+      "▁Hip p",
+      "安市 人民法院",
+      "安 市人民法院",
+      "it ively",
+      "itive ly",
+      "iti vely",
+      "▁G iving",
+      "▁Gi ving",
+      "nDisc over",
+      "nDis cover",
+      "napi Version",
+      "▁illustr ator",
+      "x n",
+      "南 湖",
+      "▁用 户",
+      "▁ 用户",
+      "情况 及",
+      "ate urs",
+      "ateur s",
+      "▁c make",
+      "▁cm ake",
+      "▁ cmake",
+      "ath olic",
+      "▁Na pole",
+      "▁Nap ole",
+      "▁U nable",
+      "▁Un able",
+      "▁ Unable",
+      "▁PR IMARY",
+      "translation al",
+      "transl ational",
+      "个 点",
+      "优 酷",
+      "则 可",
+      "拼 图",
+      "玩 过",
+      "白 沙",
+      "pp t",
+      "p pt",
+      "去哪 里",
+      "去 哪里",
+      "法学 院",
+      "法 学院",
+      "经过 点",
+      "经 过点",
+      "(/ \\\\",
+      "( /\\\\",
+      "V ault",
+      "Sum mer",
+      "S ummer",
+      "spe aker",
+      "在服刑期间 的表现",
+      "在服刑期间的 表现",
+      "侍 郎",
+      "每 股",
+      "走 私",
+      "}` ,",
+      "} `,",
+      "是 正确的",
+      "▁voc ê",
+      "ling er",
+      "lin ger",
+      "l inger",
+      "nDe cember",
+      "nDec ember",
+      "n December",
+      "▁com pressor",
+      "▁comp ressor",
+      "▁compress or",
+      "д о",
+      "▁ 字",
+      "即 事",
+      "受 限",
+      "巨 额",
+      "得 失",
+      "蚊 子",
+      "bs d",
+      "b sd",
+      "oc ide",
+      "oci de",
+      "▁E lev",
+      "▁El ev",
+      "▁Ele v",
+      "▁Bi har",
+      "▁em oji",
+      "▁ emoji",
+      "com patibility",
+      "compat ibility",
+      "观 望",
+      "pu nk",
+      "p unk",
+      "有一 部分",
+      "有 一部分",
+      "▁Y acht",
+      "▁Ya cht",
+      "ce iving",
+      "▁Moh amed",
+      "▁suggest ive",
+      "供给 侧",
+      "受害 人",
+      "bar t",
+      "ba rt",
+      "b art",
+      "st yl",
+      "sty l",
+      "不是 什么",
+      "不 是什么",
+      "pag inate",
+      "▁aud ition",
+      "▁au dition",
+      "▁audit ion",
+      "▁audi tion",
+      "▁prol actin",
+      "▁ 葛",
+      "去 何",
+      "用 语",
+      "来得 及",
+      "的 衣服",
+      "pon d",
+      "po nd",
+      "p ond",
+      "目前 已经",
+      "目前已 经",
+      "▁Great est",
+      "▁Gre atest",
+      "▁med ically",
+      "▁medical ly",
+      "▁medic ally",
+      "变 性",
+      "无 能",
+      "笔 误",
+      "党组 副书记",
+      "▁Spec tro",
+      "▁Spect ro",
+      "▁pl anners",
+      "▁plan ners",
+      "▁planner s",
+      "▁t olerant",
+      "▁toler ant",
+      "不 败",
+      "多 用",
+      "有 情",
+      "要 走",
+      "▁w i",
+      "▁ wi",
+      "大多 数的",
+      "大多数 的",
+      "成为 一名",
+      "▁P LAY",
+      "▁PL AY",
+      "▁ PLAY",
+      "▁Well ness",
+      "▁Newton soft",
+      "▁ 会",
+      "地 对",
+      "说的 话",
+      "说 的话",
+      "ID EO",
+      "IDE O",
+      "都是 为了",
+      "都 是为了",
+      "▁L ocks",
+      "▁Lo cks",
+      "▁Lock s",
+      "▁Loc ks",
+      "PROC ESS",
+      "enum erate",
+      "Inv ocation",
+      "в е",
+      "开 播",
+      "认 购",
+      "的不 断",
+      "的 不断",
+      "被 授予",
+      "com be",
+      "comb e",
+      "nt hem",
+      "nthe m",
+      "nth em",
+      "n them",
+      "▁Pol yte",
+      "▁Poly te",
+      "▁width s",
+      "▁wid ths",
+      "▁gar ment",
+      "▁jurisdiction s",
+      "▁jurisdict ions",
+      "卡 牌",
+      "东营 市",
+      "▁g tk",
+      "▁gt k",
+      "▁ gtk",
+      "儿童 或者",
+      "▁s ailed",
+      "▁sa iled",
+      "▁sail ed",
+      "▁Bac illus",
+      "▁surg ically",
+      "▁surgical ly",
+      "酒店 还会提供一项住宿的免费政策",
+      "人 脉",
+      "无 用",
+      "玩 乐",
+      "重 回",
+      "▁| >",
+      "▁ |>",
+      "bo ts",
+      "bot s",
+      "b ots",
+      "li bc",
+      "lib c",
+      "▁A ster",
+      "▁As ter",
+      "▁Ast er",
+      "▁tele sc",
+      "▁tel esc",
+      "▁Circ ular",
+      "▁Cir cular",
+      "▁sett lers",
+      "▁settle rs",
+      "J ean",
+      "▁U AV",
+      "尽量 不要",
+      "ug hed",
+      "ugh ed",
+      "评审 委员会",
+      "FF FFFF",
+      "FFFF FF",
+      "in ians",
+      "ini ans",
+      "inian s",
+      "inia ns",
+      "▁R ocks",
+      "▁Rock s",
+      "▁Ro cks",
+      "▁Roc ks",
+      "或者 申请仲裁",
+      "Pr inter",
+      "Print er",
+      "ran king",
+      "rank ing",
+      "r anking",
+      "▁Dec laration",
+      "▁ Declaration",
+      "哎 呀",
+      "售 楼",
+      "案 情",
+      "步 枪",
+      "阳 镇",
+      "的 场景",
+      "ge nes",
+      "gen es",
+      "gene s",
+      "g enes",
+      "mi ght",
+      "m ight",
+      "进一步 完善",
+      "ost omy",
+      "▁r ounding",
+      "▁ro unding",
+      "▁round ing",
+      "nC itations",
+      "Con straints",
+      "Constraint s",
+      "学 风",
+      "树 下",
+      "pp m",
+      "p pm",
+      "发挥 着",
+      "系统 性",
+      "每 平方米",
+      "病毒 感染",
+      "nh tml",
+      "n html",
+      "sp ots",
+      "spot s",
+      "▁Bl vd",
+      "▁dis pose",
+      "▁disp ose",
+      "dir ective",
+      "direct ive",
+      "▁Work flow",
+      "▁ Workflow",
+      "▁rect angles",
+      "▁rectangle s",
+      "边 框",
+      "连 线",
+      "长 篇",
+      "摄氏 度",
+      "的 次数",
+      "“ \\\\(",
+      "率 分别为",
+      "含有 丰富的",
+      "nA part",
+      "nAp art",
+      "▁CL ASS",
+      "▁CLA SS",
+      "▁ CLASS",
+      "▁stair case",
+      "で は",
+      "一 动",
+      "开 方",
+      "银 色",
+      "而是 在",
+      "而 是在",
+      "hi ve",
+      "h ive",
+      "宣传 活动",
+      "工商 管理",
+      "裁定 驳回",
+      "Te chn",
+      "Tech n",
+      "▁LGBT Q",
+      "▁c aves",
+      "▁ca ves",
+      "▁cave s",
+      "▁cav es",
+      "▁IO Error",
+      "▁Water loo",
+      "R u",
+      "不 愧",
+      "卒 中",
+      "增 压",
+      "菜 鸟",
+      "Da r",
+      "D ar",
+      "义务 的",
+      "es ser",
+      "ess er",
+      "esse r",
+      "ur ved",
+      "▁gu essed",
+      "▁guess ed",
+      "▁ 简",
+      "之 恋",
+      "瘫 痪",
+      "E ye",
+      "与 客户",
+      "我 开始",
+      "▁/ ><",
+      "▁/> <",
+      "fa cts",
+      "fac ts",
+      "fact s",
+      "f acts",
+      "▁ca lor",
+      "▁cal or",
+      "German y",
+      "G ermany",
+      "▁estim ators",
+      "▁estimator s",
+      "万 方",
+      "SC A",
+      "S CA",
+      "监护 人",
+      "Ca sh",
+      "Cas h",
+      "C ash",
+      "OV ER",
+      "O VER",
+      "▁英 格兰",
+      "▁ 英格兰",
+      "ar eth",
+      "are th",
+      "aret h",
+      "▁im prov",
+      "▁impro v",
+      "▁imp rov",
+      "▁final ists",
+      "▁finalist s",
+      "▁upholst ery",
+      "刘 邦",
+      "nS U",
+      "n SU",
+      "▁或 者",
+      "▁ 或者",
+      "率达 到",
+      "率 达到",
+      "nA DD",
+      "nAD D",
+      "n ADD",
+      "▁plas min",
+      "▁plasm in",
+      "▁pro bing",
+      "▁prob ing",
+      "▁SEC URITY",
+      "▁f ittings",
+      "▁fit tings",
+      "▁fitting s",
+      "▁pen insula",
+      "A f",
+      "不 就",
+      "诡 异",
+      "高中 文化",
+      "高 中文化",
+      "cl ave",
+      "ng roup",
+      "ngr oup",
+      "n group",
+      "▁re bel",
+      "▁reb el",
+      "▁path ologies",
+      "对被执行人的财产 进行了调查",
+      "密 闭",
+      "敬 畏",
+      "红 糖",
+      "成立 后",
+      "本 决定",
+      "B irth",
+      "t config",
+      "ad aptive",
+      "ada ptive",
+      "adapt ive",
+      "▁du plicated",
+      "▁duplic ated",
+      "▁duplicate d",
+      "▁ 烤",
+      "山 海",
+      "王 志",
+      "造 船",
+      "ны х",
+      "上进 行",
+      "上 进行",
+      "ve nth",
+      "vent h",
+      "ven th",
+      "v enth",
+      "电源 适配器",
+      "ern ary",
+      "▁Schw artz",
+      "▁Schwar tz",
+      "▁imm ersed",
+      "▁immers ed",
+      "▁organ ise",
+      "第二百五十七条 第六",
+      "第二百五十七条第 六",
+      "亚 种",
+      "过 节",
+      "阻 滞",
+      "开展 的",
+      ")\\ '\\",
+      ")\\' \\",
+      ") \\'\\",
+      "▁] ;\\",
+      "▁]; \\",
+      "▁ ];\\",
+      "nS ide",
+      "n Side",
+      "▁P anc",
+      "▁Pan c",
+      "▁Pa nc",
+      "▁s lick",
+      "▁sl ick",
+      "▁slic k",
+      "ron ectin",
+      "▁Star bucks",
+      "▁ ℃",
+      "Si O",
+      "S iO",
+      "iz h",
+      "i zh",
+      "丰富 了",
+      "出版 物",
+      "ov is",
+      "ovi s",
+      "o vis",
+      "▁L TD",
+      "▁LT D",
+      "情况 属实",
+      "执行 一案",
+      "中医药 大学",
+      "乙醇 含量为",
+      "▁le asing",
+      "▁ leasing",
+      "气 和",
+      "黄 埔",
+      "牛皮 癣",
+      "ur llib",
+      "url lib",
+      "what wg",
+      "▁e iner",
+      "▁ein er",
+      "▁eine r",
+      "小说网 的一部",
+      "▁reb els",
+      "▁rebel s",
+      "List eners",
+      "Listener s",
+      "Listen ers",
+      "号 之",
+      "â te",
+      "▁ 返回",
+      "认识 了",
+      "nL ee",
+      "nLe e",
+      "n Lee",
+      "▁R MS",
+      "▁RM S",
+      "有 不同的",
+      "nJ ean",
+      "n Jean",
+      "web ui",
+      "让我们 一起",
+      "让 我们一起",
+      "oc using",
+      "ocus ing",
+      "par amet",
+      "param et",
+      "para met",
+      "nPolit ics",
+      "▁Col lective",
+      "▁Coll ective",
+      "▁Collect ive",
+      "▁wonder fully",
+      "▁wonderful ly",
+      "▁ 向",
+      "君 主",
+      "如 水",
+      "▁ 暖气",
+      "指导 和",
+      "▁s ue",
+      "▁su e",
+      "in fect",
+      "inf ect",
+      "▁Tur bo",
+      "▁Turb o",
+      "Trans late",
+      "Transl ate",
+      "▁Call back",
+      "▁ Callback",
+      "▁Front ier",
+      "▁arch ival",
+      "& :",
+      "电 镀",
+      "\"> &",
+      "\" >&",
+      "签署 了",
+      "rum s",
+      "ru ms",
+      "r ums",
+      "美国 政府",
+      "em ing",
+      "emi ng",
+      "e ming",
+      "ipher y",
+      "iph ery",
+      "▁G ENER",
+      "▁GE NER",
+      "▁ GENER",
+      "▁op aque",
+      "▁Ide ally",
+      "▁Ideal ly",
+      "▁soft ball",
+      "g w",
+      "想 把",
+      "胜 地",
+      "在你 的",
+      "在 你的",
+      "宜宾 市",
+      "▁T HERE",
+      "▁THE RE",
+      "▁TH ERE",
+      "第一百零 三条",
+      "第一百零三 条",
+      "bit coin",
+      "b itcoin",
+      "▁F older",
+      "▁Fol der",
+      "▁Fold er",
+      "▁ Folder",
+      "▁hash Code",
+      "▁ hashCode",
+      "但 仍",
+      "刘 晓",
+      "大 洋",
+      "大 类",
+      "水 浒",
+      "点 滴",
+      "id or",
+      "ido r",
+      "i dor",
+      "▁s pur",
+      "▁sp ur",
+      "年 出版的图书",
+      "可以在 判决书送达之日起十五日内",
+      "ow e",
+      "o we",
+      "收录 了",
+      "的 焦点",
+      "HP LC",
+      "H PLC",
+      "历史 悠久",
+      "学习 能力",
+      "现在 很多",
+      "现 在很多",
+      "ca mel",
+      "cam el",
+      "c amel",
+      "▁P ASS",
+      "▁PA SS",
+      "▁ PASS",
+      "An imal",
+      "Anim al",
+      "如 未按本判决",
+      "▁barbec ue",
+      "▁conv ection",
+      "▁conve ction",
+      "按 月",
+      "游 人",
+      "跳 绳",
+      "gi c",
+      "g ic",
+      "re i",
+      "r ei",
+      "要求 和",
+      "要 求和",
+      "▁L eh",
+      "▁Le h",
+      "▁M eh",
+      "▁Me h",
+      "nC omb",
+      "nCom b",
+      "nCo mb",
+      "n Comb",
+      "▁G one",
+      "▁Go ne",
+      "▁Gon e",
+      "▁sc orer",
+      "▁score r",
+      "nIn sects",
+      "nIns ects",
+      "▁pl edged",
+      "▁pledge d",
+      "▁In cluding",
+      "小 波",
+      "层 级",
+      "抢 夺",
+      "漳 州",
+      "在 当地",
+      "说到 底",
+      "说 到底",
+      "主体 责任",
+      "深入 研究",
+      "思想政治 教育",
+      "▁ur gency",
+      "▁urg ency",
+      "▁urge ncy",
+      "县人民检察院 指控",
+      "年龄较大的 儿童或者",
+      "▁abstract s",
+      "fe m",
+      "f em",
+      "双眼 皮",
+      "Sp in",
+      "S pin",
+      "▁E SR",
+      "▁ES R",
+      "ION AL",
+      "IO NAL",
+      "▁c log",
+      "▁cl og",
+      "ic kers",
+      "ick ers",
+      "icker s",
+      "na tion",
+      "nat ion",
+      "n ation",
+      "▁w ickets",
+      "▁William son",
+      "▁Williams on",
+      "流 年",
+      "陪 我",
+      "魔 力",
+      "in ders",
+      "ind ers",
+      "inder s",
+      "inde rs",
+      "la table",
+      "lat able",
+      "l atable",
+      "nThrough out",
+      "报 仇",
+      "yn n",
+      "y nn",
+      "不能 说",
+      "其实 我",
+      "担保 法",
+      "ist ema",
+      "iste ma",
+      "istem a",
+      "wide hat",
+      "▁Pav ilion",
+      "▁Mult ivariate",
+      "无 异",
+      "注 塑",
+      "背 负",
+      "но в",
+      "н ов",
+      "副 区长",
+      "▁HE LP",
+      "▁HEL P",
+      "nC orrect",
+      "nCor rect",
+      "n Correct",
+      "直 觉",
+      "稀 有",
+      "解 开",
+      "as u",
+      "a su",
+      "sh it",
+      "shi t",
+      "s hit",
+      "▁G oa",
+      "▁Go a",
+      "▁d al",
+      "▁da l",
+      "▁ dal",
+      "小朋友 们",
+      "小 朋友们",
+      "我们 的生活",
+      "我们的 生活",
+      "▁Shir ley",
+      "count ries",
+      "▁aspir ing",
+      "▁asp iring",
+      "▁aspirin g",
+      "▁ 기",
+      "按 揭",
+      "骆 驼",
+      "十八 大",
+      "十 八大",
+      "的 唯一",
+      "离 退休",
+      "能 求出",
+      "tc md",
+      "t cmd",
+      "▁Ter esa",
+      "▁di vert",
+      "▁div ert",
+      "▁diver t",
+      "▁dive rt",
+      "▁w ander",
+      "▁wa nder",
+      "▁wand er",
+      "record ed",
+      "▁utter ly",
+      "▁poss essing",
+      "▁possess ing",
+      "施 加",
+      "篆 刻",
+      "解 法",
+      "分数 的",
+      "分 数的",
+      "MA NA",
+      "MAN A",
+      "M ANA",
+      "l é",
+      "土 豪",
+      "看 点",
+      "都 对",
+      "魏 晋",
+      "Ap r",
+      "A pr",
+      "▁特 性",
+      "▁ 特性",
+      "ns ee",
+      "nse e",
+      "n see",
+      "Fatal f",
+      "Fat alf",
+      "pl anned",
+      "plan ned",
+      "yroid ism",
+      "nCon struction",
+      "▁rel ativistic",
+      "前 十",
+      "求 婚",
+      "深 耕",
+      "cr t",
+      "c rt",
+      "来 找出",
+      "的 网站",
+      "ig or",
+      "igo r",
+      "i gor",
+      "细致 分析",
+      "spons ored",
+      "▁dec isive",
+      "▁pro static",
+      "▁pr ostatic",
+      "▁prost atic",
+      "{ /",
+      "天 际",
+      "就 当",
+      "片 长",
+      "完全 不",
+      "主要 还是",
+      "bed room",
+      "▁e states",
+      "▁est ates",
+      "▁estate s",
+      "▁esta tes",
+      "▁mark ings",
+      "▁marking s",
+      "▁con verters",
+      "▁conver ters",
+      "▁convert ers",
+      "▁converter s",
+      "nDe velopment",
+      "nDevelop ment",
+      "n Development",
+      "▁share holder",
+      "MMMMMMMM MMMMMMMM",
+      "с е",
+      "山 寨",
+      "狙 击",
+      "解 出",
+      "赛 后",
+      "ON SE",
+      "ONS E",
+      "▁T us",
+      "▁Tu s",
+      "▁出版 日期",
+      "ill ian",
+      "illi an",
+      "nAl ways",
+      "▁Tun nel",
+      "▁di agon",
+      "▁dia gon",
+      "▁phospholip id",
+      "一 的",
+      "北 仑",
+      "柯 南",
+      "第三 步",
+      "车辆 的",
+      "另行 支付",
+      "潇湘 书院",
+      "▁G ore",
+      "▁Go re",
+      "▁Gor e",
+      "ta iled",
+      "tail ed",
+      "t ailed",
+      "▁s emic",
+      "▁se mic",
+      "▁sem ic",
+      "▁semi c",
+      "用 房",
+      "开放 的",
+      "开 放的",
+      "Be rn",
+      "Ber n",
+      "B ern",
+      "FA CT",
+      "F ACT",
+      "两个 人的",
+      "两个人 的",
+      "两 个人的",
+      "ma ils",
+      "mail s",
+      "m ails",
+      "nA rea",
+      "nAr ea",
+      "nAre a",
+      "n Area",
+      "▁dist orted",
+      "▁distort ed",
+      "天 山",
+      "漫 游",
+      "牧 场",
+      "\\\\ @",
+      "这里 面",
+      "这 里面",
+      "不仅 如此",
+      "赔偿 责任",
+      "▁B ret",
+      "▁Br et",
+      "▁Bre t",
+      "欧洲三大 杯",
+      "▁Lud wig",
+      "▁Mob ility",
+      "选项中挑选出 正确答案",
+      "confirm ation",
+      "生效法律文书 所确定的义务",
+      "Validation Error",
+      "盛 开",
+      "za p",
+      "z ap",
+      "英文 名称",
+      "英文名 称",
+      "▁B ard",
+      "▁Bar d",
+      "▁Ba rd",
+      "▁j ars",
+      "▁ja rs",
+      "▁jar s",
+      "ich igan",
+      "ichi gan",
+      "▁elect r",
+      "▁ele ctr",
+      "归 根",
+      "育 苗",
+      "▁N OR",
+      "▁NO R",
+      "回答 一个",
+      "▁t rypt",
+      "▁try pt",
+      "▁tryp t",
+      "Hash Set",
+      "ê n",
+      "乱 的",
+      "作 答",
+      "城 墙",
+      "藏 品",
+      "Le an",
+      "L ean",
+      "▁N GC",
+      "▁NG C",
+      "Sol id",
+      "S olid",
+      "al arm",
+      "ala rm",
+      "alar m",
+      "pro fits",
+      "profit s",
+      "prof its",
+      "vis itor",
+      "visit or",
+      "▁o ceans",
+      "▁ocean s",
+      "var iance",
+      "vari ance",
+      "Diff erent",
+      "D ifferent",
+      "▁im posing",
+      "▁imp osing",
+      "x r",
+      "一 战",
+      "大 跌",
+      "ab r",
+      "a br",
+      "工作 岗位",
+      "比较 容易",
+      "custom ers",
+      "customer s",
+      "一 目",
+      "变 小",
+      "撕 裂",
+      "淡 化",
+      "粘 度",
+      "用手 机",
+      "用 手机",
+      "MED IA",
+      "nF ield",
+      "n Field",
+      "▁cloud y",
+      "sat ellite",
+      "▁NS Object",
+      "▁Reg ulations",
+      "▁Regulation s",
+      "▁earthqu akes",
+      "▁earthquake s",
+      "内 裤",
+      "大 衣",
+      "实习 生",
+      "▁O rg",
+      "▁Or g",
+      "▁ Org",
+      "▁qu o",
+      "▁q uo",
+      "▁ quo",
+      "第六十四 条之规定",
+      "第六十四条 之规定",
+      "▁Port smouth",
+      "▁neglig ence",
+      "兴 旺",
+      "静 音",
+      "都不 一样",
+      "都 不一样",
+      "ST AND",
+      "STA ND",
+      "▁typ h",
+      "▁ty ph",
+      "Will iam",
+      "W illiam",
+      "▁polic ym",
+      "▁policy m",
+      "▁outper forms",
+      "▁refer encing",
+      "中 将",
+      "nP o",
+      "n Po",
+      "我只 能",
+      "我 只能",
+      "清楚 了",
+      "▁Machine ry",
+      "▁Mach inery",
+      "▁forward ing",
+      "▁P lane",
+      "▁Pl ane",
+      "▁Plan e",
+      "▁ Plane",
+      "▁w aking",
+      "▁wa king",
+      "最高 人民检察院",
+      "如 皋",
+      "宣传 片",
+      "要求 您",
+      "宏观 经济",
+      "▁W inc",
+      "▁Win c",
+      "▁Wi nc",
+      "al anine",
+      "ala nine",
+      "alan ine",
+      "interop Require",
+      "普 法",
+      "眩 晕",
+      "成员 国",
+      "Pr em",
+      "Pre m",
+      "P rem",
+      "WA RD",
+      "WAR D",
+      "W ARD",
+      "ny der",
+      "▁O aks",
+      "▁Oak s",
+      "合法的 借贷关系",
+      "▁Re ceived",
+      "▁ Received",
+      "▁ 别",
+      "▁M oto",
+      "▁Mot o",
+      "▁Mo to",
+      "▁ell ipt",
+      "▁Gold man",
+      "▁Riemann ian",
+      "B u",
+      "不 舍",
+      "系 原告",
+      "▁J EE",
+      "可以 看看",
+      "可以看 看",
+      "la sty",
+      "last y",
+      "las ty",
+      "l asty",
+      "▁a nthem",
+      "▁an them",
+      "▁ant hem",
+      "▁anth em",
+      "il itating",
+      "ilit ating",
+      "▁Byz antine",
+      "▁thr illing",
+      "▁thrill ing",
+      "▁fulfill ment",
+      "学 府",
+      "理 应",
+      "翻 新",
+      "Es s",
+      "E ss",
+      "` ${",
+      "▁b ru",
+      "▁br u",
+      "▁g in",
+      "▁gi n",
+      "▁ gin",
+      "sh ock",
+      "assert Is",
+      "▁Germ ans",
+      "▁German s",
+      "▁Ger mans",
+      "▁ancest or",
+      "▁over sized",
+      "▁overs ized",
+      "▁Lib eration",
+      "Sp ort",
+      "S port",
+      "id ered",
+      "ide red",
+      "ider ed",
+      "▁Hart ford",
+      "▁deliber ate",
+      "▁delib erate",
+      "▁retin opathy",
+      "▁pharmac ology",
+      "IDD LE",
+      "at ches",
+      "atch es",
+      "Re porter",
+      "Rep orter",
+      "Report er",
+      "▁gr azing",
+      "▁graz ing",
+      "▁gra zing",
+      "▁cav ities",
+      "▁b ilingual",
+      "塞 罗",
+      "极 佳",
+      "屏幕 比例",
+      "Qu een",
+      "Que en",
+      "▁fr inge",
+      "▁fri nge",
+      "Art icles",
+      "Article s",
+      "▁C elsius",
+      "可 有",
+      "城 北",
+      "的 现实",
+      "IT CH",
+      "ix OS",
+      "avor ites",
+      "avorite s",
+      "avo rites",
+      "▁carpet s",
+      "▁carp ets",
+      "nAss uming",
+      "外 媒",
+      "林 某",
+      "饭 圈",
+      "）÷ （",
+      "） ÷（",
+      "总体 来说",
+      "um bling",
+      "umb ling",
+      "▁rel ativity",
+      "中药 材",
+      "中 药材",
+      "底面 积",
+      "底 面积",
+      "判处 死刑",
+      "pos ter",
+      "post er",
+      "po ster",
+      "p oster",
+      "▁hur ts",
+      "▁hurt s",
+      "▁deep est",
+      "作为被执行人的 公民",
+      "士 顿",
+      "走 去",
+      "AA F",
+      "A AF",
+      "J ar",
+      "go to",
+      "got o",
+      "g oto",
+      "▁Z ion",
+      "▁pet als",
+      "交 房",
+      "动 摇",
+      "同 为",
+      "造 福",
+      "po m",
+      "p om",
+      "us o",
+      "u so",
+      "相较 于",
+      "部位 的",
+      "部 位的",
+      "nT ri",
+      "nTr i",
+      "n Tri",
+      "▁P DT",
+      "▁PD T",
+      "服务 机构",
+      "ta uto",
+      "tau to",
+      "t auto",
+      "▁c trl",
+      "▁ct rl",
+      "▁ ctrl",
+      "▁since re",
+      "▁sinc ere",
+      "ch angelog",
+      "change log",
+      "chang elog",
+      "nCal ifornia",
+      "n California",
+      "▁Power Point",
+      "▁evac uation",
+      "in structions",
+      "instruction s",
+      "伤 残",
+      "蟑 螂",
+      "Le s",
+      "L es",
+      "▁O M",
+      "▁ OM",
+      "的 增大而",
+      "▁p eas",
+      "▁pe as",
+      "nRe ally",
+      "nReal ly",
+      "nDe fault",
+      "nDef ault",
+      "n Default",
+      "Q M",
+      "帐 篷",
+      "西 汉",
+      "sc m",
+      "s cm",
+      "了我 们",
+      "了 我们",
+      "儿童 的",
+      "多 边形",
+      "宿 州市",
+      "思考 题",
+      "ies ta",
+      "ie sta",
+      "iest a",
+      "i esta",
+      "▁C one",
+      "▁Con e",
+      "▁Co ne",
+      "▁further more",
+      "▁un identified",
+      "异 世",
+      "弗 兰",
+      "袋 子",
+      "躺 着",
+      "af i",
+      "a fi",
+      "且仅 当",
+      "某一 个",
+      "某 一个",
+      "的 普通",
+      "▁br idal",
+      "▁brid al",
+      "▁contr asts",
+      "▁contrast s",
+      "▁ 玉",
+      "一 班",
+      "发 帖",
+      "湿 热",
+      "Z bl",
+      "sa v",
+      "s av",
+      "▁S part",
+      "▁Sp art",
+      "▁Spa rt",
+      "▁Spar t",
+      "▁t ester",
+      "▁te ster",
+      "▁test er",
+      "Mut ation",
+      "M utation",
+      "▁Carl isle",
+      "▁serv iced",
+      "▁service d",
+      "▁surviv or",
+      "▁coast line",
+      "▁Phys icians",
+      "▁Response Entity",
+      "J U",
+      "他 所",
+      "白 衣",
+      "nM il",
+      "nMi l",
+      "n Mil",
+      "ul sa",
+      "uls a",
+      "▁Con v",
+      "▁Co nv",
+      "▁ Conv",
+      "▁M ATH",
+      "▁MA TH",
+      "▁MAT H",
+      "▁ MATH",
+      "Up dates",
+      "Update s",
+      "▁p umped",
+      "▁pump ed",
+      "▁s eldom",
+      "▁sel dom",
+      "▁Col leges",
+      "▁College s",
+      "▁aeros pace",
+      "▁Author ization",
+      "▁ Authorization",
+      "中国平安 财产保险股份有限公司",
+      "▁ 生",
+      "舌 头",
+      "ho c",
+      "h oc",
+      "▁L B",
+      "▁ LB",
+      "▁ 查询",
+      "og ie",
+      "ogi e",
+      "o gie",
+      "▁S iem",
+      "▁Si em",
+      "▁Sie m",
+      "number Of",
+      "ns omething",
+      "nsome thing",
+      "n something",
+      "om position",
+      "omp osition",
+      "nAD VERTISEMENT",
+      "自由 基",
+      "那么 就",
+      "dam age",
+      "▁C urry",
+      "▁Cur ry",
+      "nS pring",
+      "nSp ring",
+      "n Spring",
+      "▁Dynam o",
+      "关 羽",
+      "启 迪",
+      "李 玉",
+      "还 给",
+      "三门 峡",
+      "应该 在",
+      "淮 北市",
+      ":: ::",
+      "pro te",
+      "pr ote",
+      "prot e",
+      "p rote",
+      "ts ize",
+      "t size",
+      "CLUD ED",
+      "CLUDE D",
+      "▁T rent",
+      "▁Tr ent",
+      "▁Tre nt",
+      "▁sc arc",
+      "▁scar c",
+      "▁en acted",
+      "ochemical ly",
+      "ochem ically",
+      "▁com plements",
+      "▁comp lements",
+      "▁compl ements",
+      "▁complement s",
+      "讨 好",
+      "参加 过",
+      "开发 和",
+      "P iece",
+      "Fr ames",
+      "Frame s",
+      "解 本题的关键",
+      "act ively",
+      "active ly",
+      "activ ely",
+      "ree tings",
+      "reet ings",
+      "reeting s",
+      "▁caus ative",
+      "▁un official",
+      "几 款",
+      "远 近",
+      "▁X u",
+      "发展 与",
+      "可有 效",
+      "可 有效",
+      "正 因为",
+      "聚 乙烯",
+      "里程 碑",
+      "需要 您",
+      "▁w ie",
+      "▁wi e",
+      "还有 一种",
+      "还 有一种",
+      "▁n oun",
+      "▁no un",
+      "中华 英烈网",
+      "ca nnot",
+      "can not",
+      "c annot",
+      "▁b link",
+      "▁bl ink",
+      "cl assic",
+      "class ic",
+      "sol ving",
+      "s olving",
+      "▁stream line",
+      "▁rearr angement",
+      "受 访",
+      "挽 救",
+      "被 困",
+      "送 的",
+      "▁d l",
+      "▁ dl",
+      "五花 肉",
+      "房产 证",
+      "▁F FT",
+      "▁FF T",
+      "服务 体系",
+      "nI ran",
+      "▁Le ading",
+      "▁Lead ing",
+      "▁mission ary",
+      "▁ 峰",
+      "▁ 龚",
+      "安 庆",
+      "那 只",
+      "主演 的",
+      "主 演的",
+      "以 满足",
+      "附加 费用",
+      "and ex",
+      "ande x",
+      "▁INC IDENT",
+      "▁teach ings",
+      "▁teaching s",
+      "多 说",
+      "酸 碱",
+      ": &#",
+      "▁Bay es",
+      "▁Ba yes",
+      "▁P ulse",
+      "▁Pul se",
+      "二〇一六年 一",
+      "▁Ab ility",
+      "▁ Ability",
+      "▁compar atively",
+      "▁comparative ly",
+      "本裁定送达后 立即发生法律效力",
+      "中 获得",
+      "冬 奥会",
+      "▁b or",
+      "▁bo r",
+      "▁ bor",
+      "很重要 的",
+      "很 重要的",
+      "▁tr iang",
+      "▁tri ang",
+      "▁Log istic",
+      "凭 着",
+      "献 给",
+      "盖 章",
+      "an ed",
+      "ane d",
+      "a ned",
+      "▁N PC",
+      "▁NP C",
+      "▁ NPC",
+      "突发 事件",
+      "find ing",
+      "fin ding",
+      "f inding",
+      "Hand ling",
+      "▁Brown ian",
+      "专 区",
+      "换 取",
+      "格 调",
+      "讲 堂",
+      "会 成为",
+      "银行 业",
+      "银 行业",
+      "ha rk",
+      "har k",
+      "h ark",
+      "je to",
+      "jet o",
+      "j eto",
+      "nP et",
+      "nPe t",
+      "n Pet",
+      "most ly",
+      "▁sym posium",
+      "召 回",
+      "学 和",
+      "ET F",
+      "E TF",
+      "nH a",
+      "n Ha",
+      "也不 好",
+      "也 不好",
+      "执行 官",
+      "民族 的",
+      "民 族的",
+      "es tag",
+      "est ag",
+      "esta g",
+      "▁ded uced",
+      "▁deduce d",
+      "▁un iforms",
+      "▁uniform s",
+      "▁uni forms",
+      "▁inhab ited",
+      "▁inhabit ed",
+      "▁migr ating",
+      "提 早",
+      "AR A",
+      "A RA",
+      "NP s",
+      "N Ps",
+      "▁H W",
+      "▁ HW",
+      "▁ 教育",
+      "小 姑娘",
+      "的心 灵",
+      "的 心灵",
+      "nR ay",
+      "n Ray",
+      "▁Q CD",
+      "四边形 是",
+      "期限为 一年",
+      "Sp orts",
+      "Sport s",
+      "S ports",
+      "sc enes",
+      "scene s",
+      "▁Cap ture",
+      "▁Capt ure",
+      "▁ Capture",
+      "▁ster ling",
+      "▁Alex andra",
+      "▁Alexand ra",
+      "▁Alexa ndra",
+      "▁lymph atic",
+      "活 得",
+      "▁u v",
+      "▁ uv",
+      "多么 的",
+      "nr eg",
+      "nre g",
+      "n reg",
+      "▁re sts",
+      "▁r ests",
+      "▁res ts",
+      "▁rest s",
+      "▁Z immer",
+      "根 基",
+      "波 形",
+      "IG F",
+      "I GF",
+      "▁k J",
+      "自己 对",
+      "var ying",
+      "va rying",
+      "v arying",
+      "▁liv elihood",
+      "▁arrhythm ias",
+      "▁trans cribed",
+      "▁tran scribed",
+      "▁Logger Factory",
+      "▁ LoggerFactory",
+      "指 着",
+      "相 减",
+      "迟 到",
+      "{% \\",
+      "{ %\\",
+      "ên cia",
+      "ê ncia",
+      "▁b run",
+      "▁br un",
+      "▁bru n",
+      "的一 元二次方程",
+      "的 一元二次方程",
+      "Pat terns",
+      "Pattern s",
+      "P atterns",
+      "▁Bundes liga",
+      "▁respect ful",
+      "▁sur jective",
+      "▁anis otropic",
+      "也 应",
+      "盛 行",
+      "功能 障碍",
+      "▁M esa",
+      "▁Me sa",
+      "▁Mes a",
+      "两 江",
+      "打 过",
+      "时 速",
+      "聊 城",
+      "读 研",
+      "的 斜率",
+      "Wh it",
+      "W hit",
+      "▁rel ieved",
+      "▁relieve d",
+      "▁n l",
+      "▁ nl",
+      "天 左右",
+      "大概 就是",
+      "▁M ice",
+      "▁Mi ce",
+      "▁Mic e",
+      "基层 党组织",
+      "▁mon os",
+      "▁mo nos",
+      "▁mono s",
+      "av orable",
+      "avor able",
+      "大 巴",
+      "拜 登",
+      "新 课",
+      "自 已",
+      "连 忙",
+      "金 凤",
+      "又 开始",
+      "履行 的",
+      "履 行的",
+      "猕猴 桃",
+      "为了 提高",
+      "ir cle",
+      "irc le",
+      "▁房价 信息",
+      "二〇一七年 十",
+      "unt arily",
+      "▁d uality",
+      "▁du ality",
+      "▁dual ity",
+      "▁mur ders",
+      "▁murder s",
+      "▁Emer ging",
+      "▁polym orphic",
+      "▁polymorph ic",
+      "因 地",
+      "多 做",
+      "na h",
+      "n ah",
+      "自然 是",
+      "起诉 状",
+      "(\\' [",
+      "▁e cl",
+      "▁ec l",
+      "▁ ecl",
+      "栽培 技术",
+      "▁p uff",
+      "▁pu ff",
+      "本院已 依法",
+      "▁m ural",
+      "▁mur al",
+      "▁mu ral",
+      "▁nov ice",
+      "IF ICATION",
+      "IFIC ATION",
+      "▁cond ensed",
+      "▁condens ed",
+      "相 依",
+      "这 门",
+      "公司 以",
+      "麦当 劳",
+      "▁K ob",
+      "▁Ko b",
+      "文化 程度",
+      "amps ia",
+      "▁ 心",
+      "午 后",
+      "迈 进",
+      "au k",
+      "a uk",
+      "cd c",
+      "c dc",
+      "战争 的",
+      "▁De als",
+      "▁Deal s",
+      "▁Es say",
+      "▁Ess ay",
+      "的 委托代理人",
+      "▁McC ain",
+      "▁X avier",
+      "▁gre ase",
+      "tres ponse",
+      "t response",
+      "▁effic acious",
+      "▁Depend encies",
+      "▁ Dependencies",
+      "特 级",
+      "遍 及",
+      "k Hz",
+      "上来 说",
+      "上 来说",
+      "提供 给",
+      "提 供给",
+      "es an",
+      "esa n",
+      "e san",
+      "先予 执行",
+      "ap ons",
+      "apon s",
+      "apo ns",
+      "a pons",
+      "nT ell",
+      "nTe ll",
+      "▁Jo ey",
+      "▁Joe y",
+      "tox icity",
+      "toxic ity",
+      "▁Clay ton",
+      "▁en quiry",
+      "诉讼费 由被告承担",
+      "▁optim ism",
+      "check point",
+      "▁water fall",
+      "把 关",
+      "ad c",
+      "a dc",
+      "▁\\\\ &",
+      "学习 成绩",
+      "▁Br oker",
+      "▁Bro ker",
+      "▁ Broker",
+      "▁p iston",
+      "▁pi ston",
+      "▁st amped",
+      "▁stamp ed",
+      "▁l igation",
+      "▁lig ation",
+      "▁Ch allenges",
+      "▁Challenge s",
+      "▁insight ful",
+      "痛 点",
+      "控制 和",
+      "植物 油",
+      "kl ass",
+      "k lass",
+      "ng OnInit",
+      "▁There sa",
+      "▁Ther esa",
+      "▁cr owned",
+      "▁crow ned",
+      "▁crown ed",
+      "▁fol iage",
+      "▁tr ending",
+      "▁trend ing",
+      "▁cluster ed",
+      "▁clust ered",
+      "▁m aternity",
+      "▁norm ative",
+      "▁summ aries",
+      "▁summar ies",
+      "▁Develop ing",
+      "团 圆",
+      "均 以",
+      "法 令",
+      "交易 的",
+      "大约 是",
+      "大 约是",
+      "ut tg",
+      "utt g",
+      "pa ble",
+      "p able",
+      "▁Go ods",
+      "▁Good s",
+      "认罪态度 较好",
+      "ject ory",
+      "▁An ders",
+      "▁And ers",
+      "抢 险",
+      "觉得 他",
+      "▁D LL",
+      "▁DL L",
+      "▁ DLL",
+      "监控 系统",
+      "发展和 改革",
+      "Br oker",
+      "Bro ker",
+      "▁k eeper",
+      "▁ke eper",
+      "▁keep er",
+      "▁ keeper",
+      "高新技术产业 开发区",
+      "▁anth ology",
+      "▁con formal",
+      "▁conf ormal",
+      "▁conform al",
+      "▁thromb ocyt",
+      "▁thrombo cyt",
+      "小 车",
+      "海 湾",
+      "鳞 片",
+      "▁X V",
+      "世纪 初",
+      "培养 基",
+      "宝鸡 市",
+      "耐 高温",
+      "\\' ];",
+      "\\'] ;",
+      "ji ra",
+      "j ira",
+      "▁a we",
+      "▁aw e",
+      "nC apt",
+      "nCap t",
+      "n Capt",
+      "pl ist",
+      "p list",
+      "▁D rum",
+      "▁Dr um",
+      "CR EMENT",
+      "CRE MENT",
+      "▁C anton",
+      "▁Can ton",
+      "▁Ca nton",
+      "▁Cant on",
+      "▁l umber",
+      "▁lum ber",
+      "▁lu mber",
+      "astrop ods",
+      "▁disp arity",
+      "▁dispar ity",
+      "j w",
+      "但 要",
+      "更 长",
+      "SU M",
+      "S UM",
+      "co c",
+      "c oc",
+      "做出 最佳",
+      "服从 管教",
+      "rem ely",
+      "reme ly",
+      "▁Open GL",
+      "▁p istol",
+      "op erated",
+      "ope rated",
+      "oper ated",
+      "▁IN TEGER",
+      "▁ INTEGER",
+      "▁ec lectic",
+      "▁short comings",
+      "多 发",
+      "Com m",
+      "Co mm",
+      "C omm",
+      "od ian",
+      "odi an",
+      "odia n",
+      "ma pped",
+      "map ped",
+      "m apped",
+      "appro val",
+      "erdin and",
+      "ン ト",
+      "可以 买",
+      "注册 表",
+      "CO VID",
+      "web driver",
+      "▁solid arity",
+      "い て",
+      "发 财",
+      "报 关",
+      "星 系",
+      "的 区域",
+      "医务 人员",
+      "滨海 新区",
+      "▁re acts",
+      "▁react s",
+      "▁sh ores",
+      "▁shore s",
+      "▁Re plies",
+      "▁Rep lies",
+      "▁show room",
+      "▁t hematic",
+      "▁the matic",
+      "▁th ematic",
+      "▁them atic",
+      "▁P atterson",
+      "也 纳",
+      "抗 压",
+      "红 木",
+      "du b",
+      "d ub",
+      "▁] </",
+      "▁sequ entially",
+      "▁sequential ly",
+      "惯 例",
+      "的 二",
+      "飞 来",
+      "▁V B",
+      "▁ VB",
+      "并 要求",
+      "ga rd",
+      "gar d",
+      "g ard",
+      "os et",
+      "ose t",
+      "o set",
+      "pl at",
+      "p lat",
+      "然后 就是",
+      "然后就 是",
+      "he rin",
+      "her in",
+      "ot ers",
+      "ote rs",
+      "oter s",
+      "o ters",
+      "执行 标的为",
+      "执行标的 为",
+      "▁Do ctors",
+      "▁Doctor s",
+      "cont ributor",
+      "contrib utor",
+      "承 租",
+      "IR T",
+      "I RT",
+      "从业 者",
+      "从 业者",
+      "周长 为",
+      "周 长为",
+      "我们 所",
+      "由于 其",
+      "ms dn",
+      "为了 保证",
+      "ANN EL",
+      "nC ert",
+      "n Cert",
+      "▁M olly",
+      "▁Mol ly",
+      "story book",
+      "ophosph ate",
+      "o phosphate",
+      "° ,",
+      "惊 叹",
+      "mu d",
+      "m ud",
+      "▁n oct",
+      "▁no ct",
+      "FIL TER",
+      "▁rel ent",
+      "未发现 被执行人有",
+      "未发现被执行人 有",
+      "未 发现被执行人有",
+      "pred icate",
+      "▁Up loaded",
+      "▁Upload ed",
+      "ا ر",
+      "城 西",
+      "礼 盒",
+      "费 等",
+      "be come",
+      "bec ome",
+      "peon ato",
+      "▁Valid ator",
+      "▁ Validator",
+      "乡 土",
+      "法 条",
+      "\\' #",
+      "le c",
+      "l ec",
+      "也 随之",
+      "而 现在",
+      "▁M ash",
+      "▁Ma sh",
+      "▁Mas h",
+      "▁ 保修信息",
+      "res erved",
+      "▁F unding",
+      "▁Fund ing",
+      "▁Fun ding",
+      "Bench mark",
+      "▁Inter vention",
+      "失 衡",
+      "学 派",
+      "病 死",
+      "借款 后",
+      "株洲 市",
+      "▁C ors",
+      "▁Co rs",
+      "▁Cor s",
+      "Te nant",
+      "Ten ant",
+      "T enant",
+      "▁Comb ining",
+      "▁Combin ing",
+      "▁g g",
+      "▁ gg",
+      "▁R HS",
+      "▁RH S",
+      "ist rar",
+      "istr ar",
+      "▁Bl ade",
+      "▁p ains",
+      "▁pain s",
+      "▁pa ins",
+      "deg rees",
+      "degree s",
+      "st aging",
+      "sta ging",
+      "iment ary",
+      "▁gre eted",
+      "▁greet ed",
+      "▁shed ding",
+      "m h",
+      "党 务",
+      "月 刊",
+      "梨 花",
+      "tk ey",
+      "t key",
+      "▁v ap",
+      "▁va p",
+      "现场勘 验",
+      "管理 条例",
+      "Al ive",
+      "Ali ve",
+      "A live",
+      "izz ard",
+      "izza rd",
+      "二〇一八年 六月",
+      "atal ytic",
+      "▁t apping",
+      "▁ta pping",
+      "▁tap ping",
+      "▁de pleted",
+      "▁dep leted",
+      "▁glyc ogen",
+      "▁ 由",
+      "张 云",
+      "谁 都",
+      "如果 他",
+      "on ica",
+      "onic a",
+      "oni ca",
+      "o nica",
+      "▁D unn",
+      "▁Dun n",
+      "▁Du nn",
+      "▁ 被子植物门",
+      "nTour ist",
+      "pass port",
+      "pas sport",
+      "▁down time",
+      "▁downt ime",
+      "▁dow ntime",
+      "之 势",
+      "盘 锦",
+      "记 入",
+      "ACT IVE",
+      "▁Br ands",
+      "▁Brand s",
+      "▁mask ed",
+      "▁re consider",
+      "想 让",
+      "日 和",
+      "液 中",
+      "ST OP",
+      "S TOP",
+      "wY WNr",
+      "▁K amp",
+      "▁Kam p",
+      "▁Ka mp",
+      "▁K ara",
+      "▁Kar a",
+      "▁Ka ra",
+      "nGr ade",
+      "n Grade",
+      "▁O xygen",
+      "▁R aleigh",
+      "▁Ra leigh",
+      "▁pill ows",
+      "▁pillow s",
+      "▁sy nthet",
+      "▁synt het",
+      "▁synth et",
+      "禁 区",
+      "金 庸",
+      "ET H",
+      "E TH",
+      "an ders",
+      "and ers",
+      "ander s",
+      "ande rs",
+      "epoch s",
+      "ou rage",
+      "our age",
+      "▁Sc ope",
+      "▁ Scope",
+      "▁can oe",
+      "oc onstr",
+      "ocon str",
+      "oco nstr",
+      "▁C andidate",
+      "▁Candid ate",
+      "▁Candida te",
+      "to UpperCase",
+      "作 的",
+      "') [",
+      "' )[",
+      "EF T",
+      "E FT",
+      "RI A",
+      "R IA",
+      "pi s",
+      "p is",
+      "zs h",
+      "z sh",
+      "提高 到",
+      "ip ad",
+      "ipa d",
+      "i pad",
+      "▁Sem inary",
+      "▁Seminar y",
+      "▁capital ize",
+      "▁disappe ars",
+      "▁disappear s",
+      "同 城",
+      "率 低",
+      "wo w",
+      "w ow",
+      "的不 良",
+      "的 不良",
+      "sp lash",
+      "spl ash",
+      "▁Paul a",
+      "▁Pa ula",
+      "opolys accharide",
+      "▁ ú",
+      "为 国",
+      "楼 的",
+      "▁I o",
+      "▁ Io",
+      "以内 的",
+      "以 内的",
+      "缴纳 罚金",
+      "ev ille",
+      "evil le",
+      "e ville",
+      "▁em bell",
+      "▁emb ell",
+      "▁ging ival",
+      "乌 龟",
+      "安 顺",
+      "胃 口",
+      "▁' *",
+      "▁cal ves",
+      "tT uesday",
+      "▁inf init",
+      "戴 上",
+      "手 掌",
+      "▁电 视",
+      "▁ 电视",
+      "来 判断",
+      "ng ulp",
+      "n gulp",
+      "▁ital ic",
+      "▁ italic",
+      "target ed",
+      "▁night ly",
+      "咀 嚼",
+      "溢 出",
+      "if ter",
+      "ift er",
+      "i fter",
+      "▁Ad ri",
+      "▁M amm",
+      "▁Ma mm",
+      "▁Mam m",
+      "▁S cha",
+      "▁Sc ha",
+      "▁Sch a",
+      "我们 都知道",
+      "我们都 知道",
+      "sc aled",
+      "scale d",
+      "scal ed",
+      "▁d otted",
+      "▁dot ted",
+      "accept ed",
+      "▁dis mant",
+      "几 万",
+      "风 尚",
+      "as ci",
+      "asc i",
+      "a sci",
+      "▁S urf",
+      "▁Sur f",
+      "▁Su rf",
+      "IL ABLE",
+      "nc lean",
+      "ncl ean",
+      "n clean",
+      "解除 财产保全",
+      "▁mit otic",
+      "ounc ements",
+      "ounce ments",
+      "ouncement s",
+      "▁in scribed",
+      "▁ins cribed",
+      "▁declar ations",
+      "▁declaration s",
+      "▁ declarations",
+      "为 新",
+      "精 良",
+      "我们 从",
+      "是他 们",
+      "是 他们",
+      "▁I CD",
+      "▁IC D",
+      "业务 范围",
+      "ga ges",
+      "gage s",
+      "g ages",
+      "TH READ",
+      "▁Spe aking",
+      "▁aut omorphism",
+      "▁autom orphism",
+      "▁accompl ishment",
+      "▁accomplish ment",
+      "▁ Н",
+      "因 人",
+      "张 学",
+      "中 找到",
+      "研究 工作",
+      "通货 膨胀",
+      "▁r ipe",
+      "▁ri pe",
+      "▁rip e",
+      "▁Ad apter",
+      "▁Adapt er",
+      "▁ Adapter",
+      "▁pos itives",
+      "▁positive s",
+      "▁posit ives",
+      "▁re alities",
+      "▁real ities",
+      "▁orig inates",
+      "▁origin ates",
+      "▁originate s",
+      "▁ 鹏",
+      "保 单",
+      "盘 的",
+      "石 器",
+      "▁其 八",
+      "▁第 四章",
+      "▁ 第四章",
+      "营养 丰富",
+      "ff iti",
+      "ffi ti",
+      "▁P ACK",
+      "▁PA CK",
+      "▁PAC K",
+      "▁ PACK",
+      "▁f amed",
+      "▁fam ed",
+      "▁fa med",
+      "▁fame d",
+      "▁f lare",
+      "▁fl are",
+      "▁ flare",
+      "ig ators",
+      "igator s",
+      "sn ippet",
+      "▁h acked",
+      "▁hack ed",
+      "达成 执行和解协议",
+      "############ ###\\",
+      "# ##############\\",
+      "全 套",
+      "南 站",
+      "红 烧",
+      "配 以",
+      "ны е",
+      "Da sh",
+      "D ash",
+      "▁ra ft",
+      "▁ raft",
+      "此题 的关键",
+      "▁bas il",
+      "▁ba sil",
+      "▁S utton",
+      "▁local ize",
+      "李 晓",
+      "▁S UN",
+      "▁SU N",
+      "机电 设备",
+      "wYWNr YWdl",
+      "▁clean sing",
+      "▁cleans ing",
+      "▁prec autions",
+      "▁special ization",
+      "戴 尔",
+      "NP C",
+      "N PC",
+      "nc o",
+      "n co",
+      "但是 却",
+      "的 图片",
+      "有 良好的",
+      "这种 现象",
+      "Th ird",
+      "doc type",
+      "do ctype",
+      "留 着",
+      "穿 了",
+      "道 人",
+      "疯狂 的",
+      "▁b ere",
+      "▁be re",
+      "▁ber e",
+      "法人 或者其他",
+      "▁ex erts",
+      "▁exert s",
+      "▁Cit izen",
+      "▁alias es",
+      "▁ali ases",
+      "▁ aliases",
+      "▁book let",
+      "▁pl ugged",
+      "▁plug ged",
+      "请从 给出的选项中",
+      "nFe atures",
+      "n Features",
+      "宁夏回族自治区 银川市",
+      "选出 你认为正确的答案",
+      "▁glucocortic oid",
+      "救 赎",
+      "电 荷",
+      "之 目的",
+      "Hol d",
+      "Ho ld",
+      "H old",
+      "nG ot",
+      "nGo t",
+      "n Got",
+      "▁S le",
+      "▁Sl e",
+      "涉及 广泛",
+      "▁L och",
+      "▁Lo ch",
+      "▁Loc h",
+      "▁la ughed",
+      "▁laugh ed",
+      "á z",
+      "军 医",
+      "送 礼",
+      "EN E",
+      "E NE",
+      "▁N EC",
+      "▁NE C",
+      "人民 调解",
+      "IENT S",
+      "I ENTS",
+      "fire s",
+      "fi res",
+      "f ires",
+      "Solve r",
+      "Sol ver",
+      "S olver",
+      "Wait ing",
+      "▁bod ily",
+      "▁gre edy",
+      "Main Window",
+      "▁cos metics",
+      "▁cosmetic s",
+      "▁cosm etics",
+      "ж и",
+      "即 有",
+      "家 门",
+      "泸 州",
+      "（ ﹣",
+      "肝 硬化",
+      "pe ar",
+      "p ear",
+      "nMethod s",
+      "n Methods",
+      "▁Re presentation",
+      "▁Rep resentation",
+      "▁Represent ation",
+      "优 选",
+      "心 律",
+      "部 委",
+      "包 头市",
+      "ba ss",
+      "bas s",
+      "b ass",
+      "▁H ag",
+      "▁Ha g",
+      "临床 疗效",
+      "休 克",
+      "张 扬",
+      "高 涨",
+      "▁ 机械",
+      "名下 有",
+      "名 下有",
+      "无 固定",
+      "ro ma",
+      "rom a",
+      "r oma",
+      "一个 挑战",
+      "View Set",
+      "▁Ad oles",
+      "▁thym us",
+      "▁thy mus",
+      "▁gest ures",
+      "▁gesture s",
+      "▁mon ocyte",
+      "▁mono cyte",
+      "▁ 设",
+      "汉 中",
+      "爆 出",
+      "返 利",
+      "'' ,",
+      "' ',",
+      "和 教育",
+      "CS DN",
+      "C SDN",
+      "kl ore",
+      "▁C ory",
+      "▁Co ry",
+      "▁Cor y",
+      "▁S unny",
+      "▁Sun ny",
+      "▁Pro fit",
+      "▁Prof it",
+      "▁con cord",
+      "▁conc ord",
+      "resource manager",
+      "众 数",
+      "依 约",
+      "标 示",
+      "海 龙",
+      "莉 莉",
+      "被 抓",
+      "Me et",
+      "er ase",
+      "era se",
+      "eras e",
+      "e rase",
+      "▁m oss",
+      "▁mo ss",
+      "▁mos s",
+      "▁分布 区域",
+      "▁po ised",
+      "oc ations",
+      "ocation s",
+      "nPat ients",
+      "D u",
+      "噩 梦",
+      "学 年",
+      "心 愿",
+      "打 完",
+      "拥 护",
+      "ia b",
+      "i ab",
+      "正 相关",
+      "ue vo",
+      "u evo",
+      "ch itz",
+      "chi tz",
+      "质量 管理体系",
+      "质量管理 体系",
+      "▁depend able",
+      "隔 断",
+      "RS S",
+      "R SS",
+      "一万 元",
+      "一 万元",
+      "年龄 段",
+      "构建 了",
+      "ai se",
+      "ais e",
+      "a ise",
+      "企业 经营",
+      "有色 金属",
+      "额外 费用",
+      "▁M erg",
+      "▁Me rg",
+      "▁Mer g",
+      "▁Er nst",
+      "loc king",
+      "lock ing",
+      "nE lection",
+      "nEl ection",
+      "nElect ion",
+      "兴 安",
+      "无 从",
+      "柳 州",
+      "最 起码",
+      "or as",
+      "ora s",
+      "o ras",
+      "▁H ear",
+      "▁He ar",
+      "▁cs rf",
+      "▁ csrf",
+      "nG lobal",
+      "n Global",
+      "▁un real",
+      "▁unre al",
+      "nap plication",
+      "n application",
+      "▁manip ulations",
+      "▁manipulation s",
+      "Q B",
+      "▁K E",
+      "▁ KE",
+      "▁分 钟",
+      "▁ 分钟",
+      "▁T ek",
+      "▁Te k",
+      "活动 现场",
+      "▁pi ez",
+      "▁pie z",
+      "一道 复杂的",
+      "▁did nt",
+      "▁didn t",
+      "在有限 时间内",
+      "▁end ure",
+      "nIm agine",
+      "▁Comb ine",
+      "▁Combin e",
+      ", _",
+      "与 大",
+      "及 对",
+      "咸 阳",
+      "太 久",
+      "Le d",
+      "L ed",
+      "▁Cub s",
+      "▁Cu bs",
+      "nM onday",
+      "nMon day",
+      "n Monday",
+      "▁bio active",
+      "▁nanop article",
+      "▁nano particle",
+      "新 编",
+      "菜 的",
+      "跑 路",
+      "集 市",
+      "uk es",
+      "uke s",
+      "u kes",
+      "我将 给您",
+      "还有 其他",
+      "还 有其他",
+      "▁ch emo",
+      "▁che mo",
+      "▁chem o",
+      "▁Daw son",
+      "▁pot tery",
+      "▁dr agging",
+      "▁drag ging",
+      "应 交",
+      "网 购",
+      "主体 的",
+      "主 体的",
+      "之外 的",
+      "之 外的",
+      "TA CT",
+      "T ACT",
+      "cr unch",
+      "c runch",
+      "请从 所提供的",
+      "选项中 做出最佳",
+      "一道复杂的 选择题",
+      "方面的知识 来找出",
+      "要求您 综合运用多",
+      "▁Sport ing",
+      "我将给您 一道复杂的选择题",
+      "方面的知识来找出 正确的答案",
+      "请从所提供的 选项中做出最佳",
+      "花 粉",
+      "衣 食",
+      "CM C",
+      "C MC",
+      "▁* (",
+      "▁ *(",
+      "▁U i",
+      "▁ Ui",
+      "▁R PC",
+      "▁RP C",
+      "▁ RPC",
+      "机器 学习",
+      "深入 学习",
+      "▁C hes",
+      "▁Ch es",
+      "▁Che s",
+      "▁L INE",
+      "▁LI NE",
+      "▁LIN E",
+      "▁ LINE",
+      "▁e ats",
+      "▁eat s",
+      "▁pro tr",
+      "▁pr otr",
+      "▁prot r",
+      "▁le aking",
+      "▁leak ing",
+      "▁mand ibular",
+      "▁v iz",
+      "▁vi z",
+      "▁ viz",
+      "▁y um",
+      "▁ yum",
+      "▁H int",
+      "▁Hi nt",
+      "▁ Hint",
+      "is ational",
+      "isation al",
+      "start Time",
+      "▁endomet ri",
+      "▁const ellation",
+      "▁ 鲁",
+      "ip a",
+      "i pa",
+      "ow ler",
+      "owl er",
+      "Inf luence",
+      "▁Ven tures",
+      "▁Vent ures",
+      "▁umb ilical",
+      "▁young sters",
+      "○ ○",
+      "初 审",
+      "即 墨",
+      "果 肉",
+      "不下 去",
+      "不 下去",
+      "不小 的",
+      "不 小的",
+      "清楚 的",
+      "os arcoma",
+      "▁belong ings",
+      "▁belonging s",
+      "▁default Value",
+      "▁ defaultValue",
+      "▁hypoth alamus",
+      "卵 管",
+      "浮 现",
+      "Sy n",
+      "S yn",
+      "▁D rivers",
+      "▁Dr ivers",
+      "▁Drive rs",
+      "▁Driver s",
+      "本院经审理 认定事实如下",
+      "第十 七",
+      "第 十七",
+      "ci ón",
+      "c ión",
+      "▁El even",
+      "▁Ele ven",
+      "▁Elev en",
+      "▁Gen tle",
+      "▁Gent le",
+      "▁past ry",
+      "▁pa stry",
+      "▁pas try",
+      "▁hash tag",
+      "▁hasht ag",
+      "▁inst alls",
+      "▁install s",
+      "▁analy sing",
+      "▁analys ing",
+      "徐 汇",
+      "得 天",
+      "疙 瘩",
+      "IS M",
+      "I SM",
+      "登记 在",
+      "nS ty",
+      "nSt y",
+      "n Sty",
+      "Aud it",
+      "Au dit",
+      "罗 某",
+      "▁L F",
+      "▁ LF",
+      "[] ,\\",
+      "[], \\",
+      "[ ],\\",
+      "in tr",
+      "int r",
+      "i ntr",
+      "nD ie",
+      "nDi e",
+      "n Die",
+      "绝对 不会",
+      "RE SET",
+      "RES ET",
+      "▁Bro nx",
+      "▁Bron x",
+      "▁fe eder",
+      "▁feed er",
+      "▁fee der",
+      "喜欢 他",
+      "多大 的",
+      "多 大的",
+      "下面 这道题",
+      "car bons",
+      "carbon s",
+      "▁Out come",
+      "▁Type Script",
+      "午 夜",
+      "外 币",
+      "大 帝",
+      "我 俩",
+      "轮 流",
+      "ni h",
+      "n ih",
+      "忽略 了",
+      "bb ie",
+      "b bie",
+      "降低 成本",
+      "降 低成本",
+      "▁Sch ro",
+      "▁Schr o",
+      "▁Bah rain",
+      "stream ing",
+      "stre aming",
+      "▁wrest ler",
+      "▁Vol tage",
+      "▁lam inate",
+      "▁lamin ate",
+      "▁enum eration",
+      "▁enumer ation",
+      "冶 炼",
+      "湿 疹",
+      "就是 不",
+      "就 是不",
+      "技术 及",
+      "的 运用",
+      "Ra in",
+      "R ain",
+      "请你 帮我",
+      "▁Fund s",
+      "▁Fun ds",
+      "▁Mon aco",
+      "▁k night",
+      "▁kn ight",
+      "▁V atican",
+      "买 过",
+      "加 权",
+      "紧 邻",
+      "利益 的",
+      "名 同学",
+      "实体 店",
+      "菏泽 市",
+      "▁K ok",
+      "▁Ko k",
+      "▁t ak",
+      "▁ta k",
+      "ob iles",
+      "obile s",
+      "obi les",
+      "obil es",
+      "太 和",
+      "撰 稿",
+      "是 宋代",
+      "庭审 笔录",
+      "Cell s",
+      "C ells",
+      "dar win",
+      "▁res olves",
+      "▁resolve s",
+      "▁present ers",
+      "▁presenter s",
+      "▁ 费",
+      "人 所",
+      "凝 血",
+      "加 装",
+      "培 根",
+      "海 运",
+      "特 勒",
+      "▁A UC",
+      "▁AU C",
+      "▁l il",
+      "▁li l",
+      "综艺 节目",
+      "nM att",
+      "nMat t",
+      "nMa tt",
+      "n Matt",
+      "▁GOOD S",
+      "▁in cap",
+      "▁inc ap",
+      "▁t apes",
+      "▁ta pes",
+      "▁tap es",
+      "▁tape s",
+      "▁in ward",
+      "▁Sim mons",
+      "▁nood les",
+      "有 名",
+      "盛 典",
+      "美 发",
+      "原告 人",
+      "和解 析",
+      "和 解析",
+      "王世 贞",
+      "考察 了",
+      "[[ \\'",
+      "[ [\\'",
+      "▁A IR",
+      "▁AI R",
+      "平均 亩产",
+      "案例 分析",
+      "nD est",
+      "nDes t",
+      "nDe st",
+      "n Dest",
+      "▁作品 体裁",
+      "r outine",
+      "▁Th reat",
+      "▁Thr eat",
+      "▁air ways",
+      "▁airway s",
+      "▁anaest hesia",
+      "▁epidem iologic",
+      "万 岁",
+      "听 觉",
+      "讲 了",
+      "顺 义",
+      "Sh a",
+      "S ha",
+      "▁$$ [",
+      "▁R ak",
+      "▁Ra k",
+      "▁m ah",
+      "▁ma h",
+      "选项 正确",
+      "nAtt ached",
+      "▁garner ed",
+      "上 任",
+      "多 处",
+      "伦 贝尔",
+      "en go",
+      "eng o",
+      "e ngo",
+      "min ent",
+      "mine nt",
+      "m inent",
+      "也 给",
+      "听 话",
+      "现 年",
+      "用 处",
+      "礼 服",
+      "跟 风",
+      "的 开始",
+      "▁V or",
+      "▁Vo r",
+      "▁M orm",
+      "▁Mor m",
+      "▁Mo rm",
+      "▁ac yl",
+      "▁ acyl",
+      "▁In flu",
+      "▁Inf lu",
+      "▁Infl u",
+      "查封 被执行人",
+      "▁hun ters",
+      "▁hunt ers",
+      "▁hu nters",
+      "▁hunter s",
+      "▁Flo oring",
+      "▁Floor ing",
+      "▁pave ment",
+      "▁pav ement",
+      "▁tear Down",
+      "丧 尸",
+      "名 思义",
+      "小学 校",
+      "小 学校",
+      "ocard ial",
+      "ocar dial",
+      "▁sad ness",
+      "仅 限",
+      "拜 托",
+      "木 斯",
+      "的人 格",
+      "的 人格",
+      "nI nsp",
+      "nIn sp",
+      "nIns p",
+      "n Insp",
+      "▁B ots",
+      "▁Bo ts",
+      "▁Bot s",
+      "▁D rain",
+      "▁Dr ain",
+      "▁De utsche",
+      "▁chat ting",
+      "Ass essment",
+      "▁re building",
+      "▁rebuild ing",
+      "古 迹",
+      "泳 池",
+      "香 甜",
+      "ba h",
+      "b ah",
+      "老 太太",
+      "}} }$\\",
+      "}}} $\\",
+      "} }}$\\",
+      "子叶植物 纲",
+      "▁Cam den",
+      "olit ical",
+      "oli tical",
+      "▁C ellular",
+      "▁Cell ular",
+      "Act ivation",
+      "制 图",
+      "ug gy",
+      "ugg y",
+      "机场等 接",
+      "提取 公因式",
+      "Sun day",
+      "S unday",
+      "▁mod uli",
+      "▁su icidal",
+      "acchar omyces",
+      "▁aggressive ly",
+      "▁aggress ively",
+      "▁psych ologist",
+      "宝 玉",
+      "度 为",
+      "硝 酸",
+      "的 创新",
+      "▁G lu",
+      "▁Gl u",
+      "▁ Glu",
+      "进行 检查",
+      "fe eds",
+      "feed s",
+      "fee ds",
+      "f eeds",
+      "▁br ushes",
+      "▁brush es",
+      "ass essment",
+      "▁relig ions",
+      "▁religion s",
+      "▁incre ments",
+      "▁increment s",
+      "▁pa ediatric",
+      "▁warrant ies",
+      "浮 生",
+      "连 同",
+      "不得 已",
+      "往往 会",
+      "PL US",
+      "世界 第一",
+      "为 代表的",
+      "▁T RAN",
+      "▁TR AN",
+      "▁TRA N",
+      "▁ TRAN",
+      "▁f ecal",
+      "▁fe cal",
+      "并给出 正确的答案",
+      "▁advent urous",
+      "不 振",
+      "成就 感",
+      "网络 通信",
+      "at ant",
+      "ata nt",
+      "atan t",
+      "le don",
+      "led on",
+      "▁m ats",
+      "▁mat s",
+      "▁ma ts",
+      "ol fram",
+      "olf ram",
+      "▁bo obs",
+      "请分析 下面这道题",
+      "▁sed ation",
+      "文 创",
+      "ya r",
+      "y ar",
+      "大队 长",
+      "大 队长",
+      "he nol",
+      "hen ol",
+      "h enol",
+      "nd own",
+      "ndo wn",
+      "n down",
+      "▁V est",
+      "▁Ve st",
+      "▁Ves t",
+      "▁ey eb",
+      "▁eye b",
+      "一道 选择题",
+      "▁par ishes",
+      "▁parish es",
+      "▁absor bing",
+      "▁absorb ing",
+      "分 队",
+      "流 逝",
+      "穿 插",
+      "行 销",
+      "赛 的",
+      "Bo ss",
+      "B oss",
+      "▁C ros",
+      "▁Cr os",
+      "▁Cro s",
+      "▁t oug",
+      "▁to ug",
+      "▁tou g",
+      "Re comm",
+      "Rec omm",
+      "▁de cou",
+      "▁dec ou",
+      "▁Car men",
+      "▁Carm en",
+      "c ylinder",
+      "▁bucket s",
+      "▁buck ets",
+      "▁recursive ly",
+      "▁recurs ively",
+      "停 机",
+      "而 得名",
+      "还不 能",
+      "还 不能",
+      "所用 到的",
+      "所 用到的",
+      "(\" \")\\",
+      "(\"\" )\\",
+      "( \"\")\\",
+      "ise ase",
+      "▁L ymph",
+      "▁Ly mph",
+      "▁Nik ol",
+      "▁Beat les",
+      "▁Fal cons",
+      "▁Falcon s",
+      "▁for each",
+      "▁fore ach",
+      "▁fo reach",
+      "▁ foreach",
+      "▁in forming",
+      "▁inform ing",
+      "▁tread mill",
+      "▁mil estones",
+      "▁mile stones",
+      "▁milestone s",
+      "实 干",
+      "萨 拉",
+      "被 除",
+      "AV I",
+      "A VI",
+      "及 技术",
+      "手机 版",
+      "nP DF",
+      "n PDF",
+      "效力 球队",
+      "的影响 力",
+      "的 影响力",
+      "ch osen",
+      "cho sen",
+      "sp aper",
+      "s paper",
+      "▁D ivid",
+      "▁Di vid",
+      "▁Div id",
+      "▁Nor wich",
+      "▁Bound ary",
+      "▁vest ibular",
+      "优 劣",
+      "目 睹",
+      "相 聚",
+      "ke h",
+      "k eh",
+      "▁d v",
+      "▁ dv",
+      "土地 利用",
+      "聊天 记录",
+      "pa int",
+      "pain t",
+      "p aint",
+      "▁Le igh",
+      "id opsis",
+      "ido psis",
+      "▁real izes",
+      "▁realize s",
+      "栗 子",
+      "次 年",
+      "玩 玩",
+      "花 果",
+      "长 白",
+      "阅 历",
+      "首 款",
+      "请 立即",
+      "队员 们",
+      "UG IN",
+      "国际 标准",
+      "en did",
+      "end id",
+      "igg le",
+      "代表 国家队",
+      "ius eppe",
+      "▁fract al",
+      "公布 失信被执行人名单",
+      "管 网",
+      "多 利亚",
+      "政治 局",
+      "方式 为",
+      "等 技术",
+      "▁h en",
+      "▁he n",
+      "▁ hen",
+      "scr atch",
+      "可 酌情从轻处罚",
+      "▁ins omnia",
+      "▁nore pinephrine",
+      "之 气",
+      "楼 宇",
+      "ba k",
+      "b ak",
+      "▁ 河北",
+      "行驶 证",
+      "▁G ott",
+      "▁Go tt",
+      "▁Got t",
+      "未到庭 应诉",
+      "est yles",
+      "estyle s",
+      "esty les",
+      "e styles",
+      "id ences",
+      "idence s",
+      "iden ces",
+      "一 斤",
+      "冰 冷",
+      "吕 布",
+      "尿 管",
+      "教 书",
+      "纳 德",
+      "Ro m",
+      "R om",
+      "的 眼神",
+      "MS Cs",
+      "MSC s",
+      "M SCs",
+      "生产 总值",
+      "Br oad",
+      "Bro ad",
+      "B road",
+      "▁disrupt ive",
+      "全 职",
+      "受 灾",
+      "基 底",
+      "建 伟",
+      "水 解",
+      "st s",
+      "s ts",
+      "瞩 目的",
+      "过敏 性",
+      "▁N eon",
+      "▁Ne on",
+      "▁Neo n",
+      "方可 开展经营活动",
+      "剧 团",
+      "置 疑",
+      "对该 案",
+      "对 该案",
+      "一个 小时",
+      "一个小 时",
+      "一 个小时",
+      "aut er",
+      "au ter",
+      "a uter",
+      "ot hes",
+      "oth es",
+      "othe s",
+      "▁L ans",
+      "▁La ns",
+      "▁Lan s",
+      "▁V ish",
+      "▁Vis h",
+      "▁Vi sh",
+      "▁N agar",
+      "▁Na gar",
+      "▁Nag ar",
+      "RU PTION",
+      "欠 缺",
+      "▁' \\",
+      "▁ '\\",
+      "东北 部",
+      "东 北部",
+      "▁S onic",
+      "▁So nic",
+      "▁Son ic",
+      "▁U zbek",
+      "ain ment",
+      "ai nment",
+      "打 成",
+      "脑 残",
+      "自 成",
+      "▁主 料",
+      "▁ 主料",
+      "发表 的",
+      "发 表的",
+      "交易 平台",
+      "Of fer",
+      "Off er",
+      "▁A side",
+      "▁As ide",
+      "▁Mus cle",
+      "▁ham ster",
+      "▁week day",
+      "吸 水",
+      "海 的",
+      "益 生",
+      "顽 固",
+      "鼻 腔",
+      "As c",
+      "A sc",
+      "在上 面",
+      "在 上面",
+      "公益 事业",
+      "\\' \\'\\",
+      "\\'\\ '\\",
+      "\\'\\' \\",
+      "it tal",
+      "itt al",
+      "▁st aring",
+      "▁star ing",
+      "▁ ت",
+      "▁N d",
+      "nc at",
+      "n cat",
+      "nQ uant",
+      "nQu ant",
+      "n Quant",
+      "缓期二年 执行",
+      "▁Cook ies",
+      "▁Cookie s",
+      "▁Ins ights",
+      "▁wrest lers",
+      "▁wrestler s",
+      "nExper ience",
+      "不 比",
+      "写 上",
+      "是否 为",
+      "住 河北省",
+      "进行 处理",
+      "▁s way",
+      "▁sw ay",
+      "经本院 依法",
+      "经 本院依法",
+      "▁Ab use",
+      "▁Abu se",
+      "item ize",
+      "银 联",
+      "OR A",
+      "O RA",
+      "▁ 视频",
+      "某某 诉称",
+      "比较 大的",
+      "比较大 的",
+      "比 较大的",
+      "监狱 表扬",
+      "▁N ile",
+      "▁Ni le",
+      "▁Nil e",
+      "bes gue",
+      "Current ly",
+      "▁competitive ness",
+      "▁competit iveness",
+      "夜 市",
+      "批 表",
+      "时 事",
+      "<\\ '",
+      "< \\'",
+      "na no",
+      "nan o",
+      "n ano",
+      "un ky",
+      "unk y",
+      "▁g ov",
+      "▁go v",
+      "▁ gov",
+      "in ches",
+      "inc hes",
+      "inch es",
+      "▁t agging",
+      "▁tag ging",
+      "▁temp ted",
+      "▁tempt ed",
+      "▁outs ourcing",
+      "▁My cobacterium",
+      "DR S",
+      "D RS",
+      "▁H SP",
+      "▁HS P",
+      "共同 债务",
+      "鼠标 操作",
+      "sk ills",
+      "skill s",
+      "▁hydrox yl",
+      "▁hydroxy l",
+      "二 章",
+      "来 之",
+      "这 人",
+      ">' ,",
+      "> ',",
+      "nn p",
+      "n np",
+      "(\\' :",
+      "(\\ ':",
+      "( \\':",
+      "▁M ae",
+      "▁Ma e",
+      "元为 基数",
+      "▁Caf é",
+      "对应的 选项",
+      "▁m illing",
+      "▁mill ing",
+      "▁mil ling",
+      "▁univers ally",
+      "▁universal ly",
+      "▁neutral izing",
+      "兴 国",
+      "再 有",
+      "定 罪",
+      "nE v",
+      "n Ev",
+      "第八 条",
+      "第 八条",
+      "▁C CD",
+      "▁CC D",
+      "claim s",
+      "▁g rating",
+      "▁gr ating",
+      "▁grat ing",
+      "▁gra ting",
+      "▁syn apse",
+      "▁hand held",
+      "▁poly ethylene",
+      "偏 高",
+      "转 行",
+      "PH A",
+      "P HA",
+      "OT AL",
+      "OTA L",
+      "▁De pt",
+      "▁Dep t",
+      "自然 保护区",
+      "重庆 时时彩",
+      "▁ch ords",
+      "▁chord s",
+      "▁chor ds",
+      "▁Cath olics",
+      "▁Catholic s",
+      "之 手",
+      "美 式",
+      "▁C Y",
+      "▁ CY",
+      "六 安市",
+      "方式 是",
+      "羽绒 服",
+      "▁L ean",
+      "▁Le an",
+      "▁ Lean",
+      "可以 向人民法院",
+      "可以向 人民法院",
+      "▁sem ifinals",
+      "▁semi finals",
+      "▁cert ifications",
+      "▁certification s",
+      "二 个",
+      "MO VE",
+      "印象 深刻",
+      "▁con qu",
+      "▁sup reme",
+      "▁suprem e",
+      "▁blind ness",
+      "之 首",
+      "正 气",
+      "迪 亚",
+      "ko t",
+      "k ot",
+      "判别 式",
+      "大小 的",
+      "大 小的",
+      "Se ason",
+      "▁Be aver",
+      "▁Bea ver",
+      "▁fe ared",
+      "▁fear ed",
+      "片 中",
+      "觉 悟",
+      "顺 手",
+      ")) ^",
+      ") )^",
+      "▁[ %",
+      "一起 吃",
+      "当天 的",
+      "当 天的",
+      "▁S TO",
+      "▁ST O",
+      "ic hen",
+      "ich en",
+      "iche n",
+      "i chen",
+      "▁s cars",
+      "▁sc ars",
+      "▁scar s",
+      "combin e",
+      "comb ine",
+      "▁As sess",
+      "▁Ass ess",
+      "▁Qu otes",
+      "▁Quote s",
+      "▁line ages",
+      "▁lineage s",
+      "▁data frame",
+      "▁rib osomal",
+      "上 过",
+      "嘉 善",
+      "暴 躁",
+      "警 报",
+      "的天 空",
+      "的 天空",
+      "的路 上",
+      "的 路上",
+      "▁d ow",
+      "▁do w",
+      "▁ dow",
+      "暂 不具备",
+      "nA ction",
+      "nAct ion",
+      "nAc tion",
+      "n Action",
+      "▁m otives",
+      "▁mot ives",
+      "▁motiv es",
+      "▁new line",
+      "▁ newline",
+      "▁strand ed",
+      "▁ stranded",
+      "▁mod ifiers",
+      "▁modifier s",
+      "Des erialize",
+      "De serialize",
+      "▁prom inence",
+      "▁promin ence",
+      "▁conf inement",
+      "三 世",
+      "下 周",
+      "害 虫",
+      "建 民",
+      "放 着",
+      "很高 兴",
+      "很 高兴",
+      "金字 塔",
+      "Know n",
+      "Kn own",
+      "K nown",
+      "▁T oll",
+      "▁To ll",
+      "▁Tol l",
+      "Vi ctor",
+      "V ictor",
+      "作出 正确判断",
+      "▁sh ooters",
+      "▁shoot ers",
+      "▁shooter s",
+      "▁Princ iple",
+      "▁Ext ensions",
+      "▁Extension s",
+      "▁ Extensions",
+      "洛 夫",
+      "与 文化",
+      "总价 中",
+      "pa ul",
+      "p aul",
+      "中心 对称",
+      "负面 影响",
+      "刑罚 执行机关",
+      "▁sn akes",
+      "▁snake s",
+      "Type Name",
+      "没有约定 或者约定",
+      "▁Gen etics",
+      "▁Genetic s",
+      "▁Green land",
+      "▁Network ing",
+      "▁Net working",
+      "▁un controlled",
+      "张 志",
+      "雾 霾",
+      "第六 届",
+      "HO LD",
+      "H OLD",
+      "▁F GF",
+      "所 确定的",
+      "▁bu oy",
+      "▁ju icy",
+      "▁manufact ures",
+      "▁manufacture s",
+      "H J",
+      "▁ 利",
+      "玩 转",
+      "轮 候",
+      "顶 着",
+      "FT P",
+      "F TP",
+      "▁M p",
+      "在 网络",
+      "望远 镜",
+      "不是 因为",
+      "不 是因为",
+      "▁F ounded",
+      "▁Found ed",
+      "▁Techn ique",
+      "东 城",
+      "充 当",
+      "带 回",
+      "德 基",
+      "捷 径",
+      "省 的",
+      "的 身影",
+      "聚氨 酯",
+      "让 用户",
+      "you t",
+      "yo ut",
+      "y out",
+      "▁D FT",
+      "▁DF T",
+      "您 将面临",
+      "文学 作品",
+      "检验 检疫",
+      "淋浴 设施",
+      "en sing",
+      "ens ing",
+      "ensin g",
+      "B K",
+      "三 生",
+      "不 乏",
+      "尽 责",
+      "大多 是",
+      "大 多是",
+      "an ye",
+      "any e",
+      "nB re",
+      "nBr e",
+      "n Bre",
+      "▁I OS",
+      "▁IO S",
+      "▁ IOS",
+      "详细 评估",
+      "两次 以上的",
+      "es cent",
+      "esc ent",
+      "in care",
+      "inc are",
+      "nS eeds",
+      "nSee ds",
+      "nSe eds",
+      "▁un beat",
+      "▁arg inine",
+      "沦 为",
+      "骑 兵",
+      "在 英国",
+      "葫芦 岛",
+      "am us",
+      "a mus",
+      "▁Z ar",
+      "▁n th",
+      "▁ nth",
+      "ig nty",
+      "ign ty",
+      "▁合作 关系",
+      "▁ 合作关系",
+      "▁mail box",
+      "▁Sup plier",
+      "▁ Supplier",
+      "epit helial",
+      "哈 密",
+      "憔 悴",
+      "熟 的",
+      "而 他",
+      "到 自己",
+      "未 提出",
+      "▁C ra",
+      "▁Cr a",
+      "不断 扩大",
+      "给你 一段",
+      "最佳 的答案",
+      "最佳的 答案",
+      "▁s ucking",
+      "▁suc king",
+      "▁suck ing",
+      "type param",
+      "▁stress ors",
+      "直接 用",
+      "br on",
+      "bro n",
+      "b ron",
+      "回答 一道",
+      "大家 都是",
+      "大家都 是",
+      "这个 事情",
+      "这方面 的",
+      "这 方面的",
+      "J apan",
+      "Term s",
+      "Ter ms",
+      "▁b red",
+      "▁br ed",
+      "▁bre d",
+      "▁ bred",
+      "▁C ancel",
+      "▁Can cel",
+      "▁ Cancel",
+      "平面 直角坐标系",
+      "机场等接 车服务",
+      "▁Daniel s",
+      "▁Dani els",
+      "c plusplus",
+      "▁depart ed",
+      "写 过",
+      "西 欧",
+      "尽 全力",
+      "▁S ey",
+      "▁Se y",
+      "这个 过程",
+      "Br own",
+      "Bro wn",
+      "B rown",
+      "nc all",
+      "n call",
+      "▁m age",
+      "▁mag e",
+      "▁ma ge",
+      "电动 自行车",
+      "▁ex poses",
+      "▁exp oses",
+      "▁expose s",
+      "▁Requ ests",
+      "▁Request s",
+      "▁ Requests",
+      "▁Tele gram",
+      "将 它",
+      "数 中",
+      "海 带",
+      "羟 基",
+      "陵 园",
+      "ev il",
+      "e vil",
+      "▁D UI",
+      "▁DU I",
+      "元的 事实",
+      "元 的事实",
+      "▁ge om",
+      "▁geo m",
+      "▁ geom",
+      "▁pro j",
+      "▁pr oj",
+      "▁ proj",
+      "需在 入住时",
+      "计算在 总价中",
+      "App Data",
+      "▁fauc et",
+      "附加费用 不会自动",
+      "需在入住时 另行支付",
+      "set ObjectName",
+      "附加费用不会自动 计算在总价中",
+      "▁ 郝",
+      "不 识",
+      "张 开",
+      "五年 级",
+      "五 年级",
+      "Ha lf",
+      "H alf",
+      "▁c ra",
+      "▁cr a",
+      "▁Re ich",
+      "▁t iger",
+      "▁ti ger",
+      "scal ajs",
+      "scala js",
+      "▁sk illet",
+      "▁skill et",
+      "▁High lands",
+      "▁Highland s",
+      "Rec yclerView",
+      "物业管理有限公司 撤回起诉",
+      "班 组",
+      "▁e b",
+      "▁ eb",
+      "三千 元",
+      "三 千元",
+      "这条 路",
+      "直线 距离",
+      "▁CU DA",
+      "RET URN",
+      "reg nant",
+      "regn ant",
+      "▁gen eal",
+      "▁gene al",
+      "▁ge neal",
+      "▁mast urb",
+      "nS aturday",
+      "n Saturday",
+      "re duction",
+      "red uction",
+      "redu ction",
+      "▁ 阳",
+      "内 镜",
+      "断 电",
+      "而 你",
+      "肉 牛",
+      "风 月",
+      "华尔 街",
+      "▁M Pa",
+      "▁MP a",
+      "▁ MPa",
+      "如果 真的",
+      "主要 是因为",
+      "主要是 因为",
+      "然后从 提供的",
+      "请您 细致分析",
+      "oint ment",
+      "▁AD VISED",
+      "您将面临 一个挑战",
+      "回答一道 复杂选择题",
+      "▁phot ovoltaic",
+      "在有限时间内 回答一道复杂选择题",
+      "人 会",
+      "夜 深",
+      "张 大",
+      "某某 在",
+      "某 某在",
+      "au re",
+      "aur e",
+      "a ure",
+      "act eria",
+      "acter ia",
+      "该 国",
+      "赛 中",
+      "敌人 的",
+      "敌 人的",
+      "发展 前景",
+      "歌曲 歌词",
+      "找出 您认为",
+      "请您 全神贯注",
+      "地审查 以下这道",
+      "Stud ents",
+      "Student s",
+      "▁bl ocker",
+      "▁block er",
+      "找出您认为 最佳的答案",
+      "涉及广泛 知识的选择题",
+      "▁correct ing",
+      "▁re directed",
+      "▁red irected",
+      "▁redirect ed",
+      "请您全神贯注 地审查以下这道",
+      "▁ 微",
+      "宝 塔",
+      "▁; )",
+      "专门 为",
+      "在 深圳",
+      "必须 有",
+      "是 现代",
+      "cr ate",
+      "c rate",
+      "nS end",
+      "nSe nd",
+      "nSen d",
+      "n Send",
+      "▁F lint",
+      "▁Fl int",
+      "▁St Object",
+      "公 尺",
+      "废 旧",
+      "活 了",
+      "% }\\",
+      "ea t",
+      "e at",
+      "上升 到",
+      "立方 根",
+      "请 思考",
+      "▁G at",
+      "▁Ga t",
+      "an ford",
+      "uous ly",
+      "u ously",
+      "▁n emat",
+      "▁ne mat",
+      "at ation",
+      "ata tion",
+      "童 装",
+      "面 面",
+      "▁B W",
+      "▁ BW",
+      "▁F ault",
+      "▁Fa ult",
+      "▁b ells",
+      "▁bel ls",
+      "▁bell s",
+      "▁sh ampoo",
+      "▁invent or",
+      "▁export ing",
+      "CC A",
+      "C CA",
+      "GF P",
+      "G FP",
+      "▁D X",
+      "▁ DX",
+      "可能 与",
+      "可 能与",
+      "el es",
+      "ele s",
+      "e les",
+      "hb ox",
+      "h box",
+      "▁K ah",
+      "▁Ka h",
+      "▁G nRH",
+      "不动产 登记",
+      "公司 董事长",
+      "证券 交易所",
+      "ger ald",
+      "g erald",
+      "▁conf igs",
+      "▁config s",
+      "▁ configs",
+      "人大常委会 副主任",
+      "巴 掌",
+      "货 品",
+      "也 并不",
+      "恨 不得",
+      "所以 你",
+      "▁I MO",
+      "▁IM O",
+      "▁ IMO",
+      "以下 有关",
+      "数学 模型",
+      "ec ode",
+      "eco de",
+      "e code",
+      "▁Hy de",
+      "▁Hyd e",
+      "▁T ale",
+      "▁Ta le",
+      "▁Tal e",
+      "▁dom est",
+      "▁dome st",
+      "ament als",
+      "amental s",
+      "▁Intellig ent",
+      "▁screen writer",
+      "之 差",
+      "照 亮",
+      "迫 使",
+      "bm p",
+      "b mp",
+      "上诉 费",
+      "流量 的",
+      "流 量的",
+      "\\\\ )”",
+      "\\\\) ”",
+      "ig ne",
+      "ign e",
+      "不可 替代",
+      "年 下半年",
+      "批发 市场",
+      "▁A jax",
+      "▁Aj ax",
+      "▁S ilk",
+      "▁Sil k",
+      "▁mis con",
+      "▁misc on",
+      "power shell",
+      "powers hell",
+      "LE SS",
+      "LES S",
+      "L ESS",
+      "ne ut",
+      "▁Ex hib",
+      "▁unem ployed",
+      "储 能",
+      "铭 记",
+      "快捷 键",
+      "简单 来说",
+      "这个 角色",
+      "▁Ge tty",
+      "▁Get ty",
+      "▁he ats",
+      "▁heat s",
+      "Can adian",
+      "▁S anchez",
+      "帮我做 一道选择题",
+      "nMore over",
+      "▁att aching",
+      "▁attach ing",
+      "一 回",
+      "按 压",
+      "接 管",
+      "整 容",
+      "筹 资",
+      "白 内障",
+      "ha ra",
+      "har a",
+      "h ara",
+      "nE nt",
+      "nEn t",
+      "n Ent",
+      "▁M our",
+      "▁Mo ur",
+      "▁Y osh",
+      "▁Yo sh",
+      "▁regular ization",
+      "▁Y i",
+      "权利 的",
+      "权 利的",
+      "tw ig",
+      "t wig",
+      "▁s uture",
+      "▁su ture",
+      "▁sut ure",
+      "信息的 若干规定",
+      "给出 正确的选项",
+      "Ga ussian",
+      "atur days",
+      "aturday s",
+      "▁foss ils",
+      "▁fossil s",
+      "▁spin ach",
+      "▁film makers",
+      "▁filmm akers",
+      "▁filmmaker s",
+      "仔细阅读 下面这道选择题",
+      "▁Access ories",
+      "振 荡",
+      "▁d ns",
+      "▁ dns",
+      "进行了 研究",
+      "ich lor",
+      "i chlor",
+      "rac ial",
+      "r acial",
+      "是一部 连载于",
+      "ident ally",
+      "idental ly",
+      "▁elect roc",
+      "▁electro c",
+      "▁electr oc",
+      "▁soc iology",
+      "▁soci ology",
+      "▁socio logy",
+      "▁abbrev iated",
+      "照 相",
+      "了解 的",
+      "▁D uc",
+      "▁Du c",
+      "不当 得利",
+      "as ton",
+      "ast on",
+      "a ston",
+      "▁b ots",
+      "▁bo ts",
+      "▁bot s",
+      "▁ bots",
+      "▁p yro",
+      "▁py ro",
+      "▁pyr o",
+      "▁in let",
+      "▁i nlet",
+      "▁un det",
+      "▁und et",
+      "请您 仔细阅读",
+      "▁cele ry",
+      "▁cel ery",
+      "占 总",
+      "记 事",
+      "wo r",
+      "w or",
+      "呈 批表",
+      "YY YY",
+      "▁人物 名称",
+      "ag ency",
+      "age ncy",
+      "agen cy",
+      "a gency",
+      "▁Mac OS",
+      "第一百四十 三条",
+      "▁road map",
+      "▁entr ants",
+      "您 需在入住时另行支付",
+      "▁de generate",
+      "▁deg enerate",
+      "▁re sembling",
+      "天 一",
+      "托 运",
+      "登 临",
+      "红 叶",
+      "谁 会",
+      "vr e",
+      "v re",
+      "分享 了",
+      "nM ad",
+      "nMa d",
+      "n Mad",
+      "到了 一个",
+      "到 了一个",
+      "我有 一道题",
+      "我有一 道题",
+      "▁in land",
+      "▁Sch neider",
+      "两 委",
+      "侥 幸",
+      "想 你",
+      "旭 东",
+      "热 电",
+      "闭 环",
+      "顾 虑",
+      "SA P",
+      "S AP",
+      "虽然 在",
+      "面积 约",
+      "受到 影响",
+      "▁B ates",
+      "▁Ba tes",
+      "▁Bat es",
+      "十一 条之规定",
+      "十一条 之规定",
+      "▁H acker",
+      "▁Hack er",
+      "▁CON CLUS",
+      "▁rhythm ic",
+      "▁rhyth mic",
+      "▁Conn ected",
+      "▁Connect ed",
+      "▁ Connected",
+      "最 深",
+      "洞 口",
+      "粗 暴",
+      "衣 裙",
+      "十八 年",
+      "十 八年",
+      "我 老公",
+      "Se ver",
+      "S ever",
+      "is ans",
+      "isa ns",
+      "isan s",
+      "▁El se",
+      "▁ Else",
+      "▁E rin",
+      "▁Er in",
+      "cl iffe",
+      "int osh",
+      "into sh",
+      "▁nut rit",
+      "▁nutr it",
+      "n content",
+      "▁re solver",
+      "▁res olver",
+      "▁resolve r",
+      "▁ resolver",
+      "相 位",
+      "石 桥",
+      "▁书 评",
+      "尤其 在",
+      "尤 其在",
+      "月上 旬",
+      "Sp ider",
+      "rat ive",
+      "r ative",
+      "及 早",
+      "各 异",
+      "晶 莹",
+      "退 伍",
+      "EF F",
+      "E FF",
+      "发现 有",
+      "发 现有",
+      "详细 介绍了",
+      "详细介绍 了",
+      "nSt eps",
+      "nStep s",
+      "nSte ps",
+      "n Steps",
+      "▁H eter",
+      "▁He ter",
+      "▁N ucle",
+      "▁Nu cle",
+      "▁W olfram",
+      "▁Wolf ram",
+      "▁Wol fram",
+      "▁compet ency",
+      "▁compete ncy",
+      "▁Fund amental",
+      "扑 克",
+      "am mu",
+      "amm u",
+      "结构 设计",
+      "nl ine",
+      "n line",
+      "▁C rus",
+      "▁Cr us",
+      "▁Cru s",
+      "二〇一七年 七",
+      "各种 信息进行",
+      "REQUI RE",
+      "Re member",
+      "Rem ember",
+      "▁stat ues",
+      "▁statue s",
+      "三 联",
+      "大庆 市",
+      "大 庆市",
+      "检察 官",
+      "的 感受",
+      "OP ER",
+      "O PER",
+      "▁I st",
+      "▁Is t",
+      "相关 专业",
+      "多领域 知识",
+      "▁Bud dy",
+      "▁Budd y",
+      "▁H older",
+      "▁Hol der",
+      "▁Hold er",
+      "▁ Holder",
+      "inf inity",
+      "▁Bre wing",
+      "▁Brew ing",
+      "an u",
+      "a nu",
+      "的 软件",
+      "Win g",
+      "Wi ng",
+      "W ing",
+      "▁v tk",
+      "▁ vtk",
+      "强 有力的",
+      "最 常见的",
+      "W omen",
+      "▁商品 编号",
+      "并在 选项中",
+      "进行 解答即可",
+      "ex porter",
+      "export er",
+      "exp orter",
+      "▁Re fresh",
+      "▁Ref resh",
+      "▁ Refresh",
+      "Percent age",
+      "报 记者",
+      "un de",
+      "und e",
+      "u nde",
+      "▁A ux",
+      "▁Au x",
+      "▁pitch ers",
+      "▁pitcher s",
+      "▁element al",
+      "▁ele mental",
+      "▁elem ental",
+      "▁m anganese",
+      "▁Compl iance",
+      "请你帮我 选出正确的选项",
+      "re newcommand",
+      "renew command",
+      "去 哪",
+      "泛 滥",
+      "(\" \"",
+      "( \"\"",
+      "J WT",
+      "但 随着",
+      "这种 东西",
+      "ng ood",
+      "ngo od",
+      "n good",
+      "综合运用 各种",
+      "Layout Inflater",
+      "不 失",
+      "入 伍",
+      "户 通",
+      "源 源",
+      "补 习",
+      ")) (",
+      ") )(",
+      "可怕 的",
+      "ha ir",
+      "hai r",
+      "h air",
+      "ls en",
+      "l sen",
+      "Lo uis",
+      "L ouis",
+      "Tor ch",
+      "T orch",
+      "dot env",
+      "mark up",
+      "最准确 的答案",
+      "▁sit com",
+      "二〇一五年 二月",
+      "▁Met rics",
+      "▁Metric s",
+      "▁ Metrics",
+      "per fusion",
+      "perf usion",
+      "一 米",
+      "兰 花",
+      "新 诗",
+      "普 拉",
+      "溯 源",
+      "虚 伪",
+      "▁` _",
+      "天然 的",
+      "天 然的",
+      "数据 和",
+      "子宫 内膜",
+      "Tra its",
+      "Trait s",
+      "T raits",
+      "▁M ature",
+      "▁Mat ure",
+      "▁Ma ture",
+      "▁d ermal",
+      "▁der mal",
+      "▁Haus dorff",
+      "▁call backs",
+      "▁callback s",
+      "▁ callbacks",
+      "剥 离",
+      "君 王",
+      "土 路",
+      "排 毒",
+      "复制 到",
+      "AN CH",
+      "FP GA",
+      "下面 是一道",
+      "下面是 一道",
+      "nAdd ed",
+      "nAd ded",
+      "n Added",
+      "▁c urry",
+      "▁cur ry",
+      "▁curr y",
+      "heart ed",
+      "ost atic",
+      "ostat ic",
+      "osta tic",
+      "o static",
+      "re eting",
+      "ree ting",
+      "reet ing",
+      "并在选项中 确定",
+      "涉及 多领域知识",
+      "Be autiful",
+      "▁pleth ora",
+      "▁st acking",
+      "▁stack ing",
+      "▁P ermanent",
+      "▁attack ers",
+      "▁attacker s",
+      "▁nanot ubes",
+      "下面是一道 复杂选择题",
+      "请仔细阅读 问题和选项",
+      "translation s",
+      "transl ations",
+      "▁Organ izations",
+      "▁Organization s",
+      "L u",
+      "水 印",
+      "法 宝",
+      "独 创",
+      "ot r",
+      "o tr",
+      "体育 运动",
+      "nl ast",
+      "n last",
+      "美术 出版社",
+      "ap ollo",
+      "apol lo",
+      "贷款利率 计算",
+      "▁sub merged",
+      "▁Buck ingham",
+      "劳 累",
+      "善 意",
+      "实 处",
+      "时 下",
+      "▁ 添加",
+      "li de",
+      "l ide",
+      "▁RF ID",
+      "wh atever",
+      "what ever",
+      "nL icensed",
+      "n Licensed",
+      "▁plant ation",
+      "▁ plantation",
+      "▁Austral ians",
+      "▁Australia ns",
+      "▁Australian s",
+      "▁ 齐",
+      "南 沙",
+      "多 于",
+      "纹 身",
+      "gi l",
+      "g il",
+      "出入 境",
+      "出 入境",
+      "▁G in",
+      "▁Gi n",
+      "OT HER",
+      "OTH ER",
+      "O THER",
+      "▁s erm",
+      "▁se rm",
+      "▁ser m",
+      "提出 撤诉申请",
+      "最合适的 选项",
+      "▁my elin",
+      "▁myel in",
+      "这道题目 是一道",
+      "信息 科技有限公司",
+      "Coll ision",
+      "▁App lying",
+      "▁Appl ying",
+      "▁Apply ing",
+      "▁un bounded",
+      "▁neuro imaging",
+      "VP N",
+      "V PN",
+      "在 广州",
+      "▁R ats",
+      "▁Ra ts",
+      "▁Rat s",
+      "▁fun nel",
+      "▁no ises",
+      "▁noise s",
+      "al gebras",
+      "algebra s",
+      "▁milit ants",
+      "第六十七条第三款 之规定",
+      "▁explan atory",
+      "主 食",
+      "滚 子",
+      "箭 头",
+      "逼 近",
+      "IN O",
+      "I NO",
+      "▁S OD",
+      "▁SO D",
+      "▁ SOD",
+      "广播 电台",
+      "首先 根据",
+      "首 先根据",
+      "▁f unk",
+      "▁fun k",
+      "▁fu nk",
+      "然后 会提供",
+      "▁s aint",
+      "▁sa int",
+      "并确定 您认为",
+      "接下来 这道题目是一道",
+      "▁Hyp othesis",
+      "请您仔细阅读 问题和选项",
+      "▁Per spective",
+      "▁Pers pective",
+      "并确定您认为 最准确的答案",
+      "买 方",
+      "棒 的",
+      "每 夜",
+      "肯 干",
+      "nC D",
+      "n CD",
+      "▁o ss",
+      "▁os s",
+      "▁ oss",
+      "建立 健全",
+      "液晶 电视",
+      "膳食 纤维",
+      "日 交付执行",
+      "▁Mar ía",
+      "第二百零 八条",
+      "给你一段 描述",
+      "请你根据 这段话",
+      "▁fl ipped",
+      "▁flip ped",
+      "然后会提供 几个选项",
+      "▁Log istics",
+      "▁Logistic s",
+      "从选项中选出 最合适的选项",
+      "执行通知 履行法律文书确定的义务",
+      "á c",
+      "三 相",
+      "国 营",
+      "烧 结",
+      "腾 飞",
+      "国家 对",
+      "特性 的",
+      "特 性的",
+      "JP EG",
+      "J PEG",
+      "▁t ad",
+      "▁ta d",
+      "Ser ve",
+      "Serv e",
+      "S erve",
+      "▁un inter",
+      "▁u ninter",
+      "▁unint er",
+      "▁uni nter",
+      "▁Friend ly",
+      "▁Count ries",
+      "迅 猛",
+      "▁ 删除",
+      "项和 为",
+      "项 和为",
+      "日起至 实际",
+      "▁gar ments",
+      "▁garment s",
+      "▁C andidates",
+      "▁Candid ates",
+      "▁Candida tes",
+      "▁Candidate s",
+      "和 王",
+      "安 阳",
+      "药 理",
+      "Am b",
+      "A mb",
+      "了一 家",
+      "了 一家",
+      "ee ks",
+      "eek s",
+      "e eks",
+      "的朋友 们",
+      "的 朋友们",
+      "仔细 的思考",
+      "二〇一五年 五月",
+      "t K",
+      "题 库",
+      "的新 闻",
+      "的 新闻",
+      "], [\"",
+      "],[ \"",
+      "▁P aw",
+      "▁Pa w",
+      "相比 之下",
+      "▁nan ow",
+      "▁na now",
+      "▁nano w",
+      "▁ins ider",
+      "▁inside r",
+      "ahren heit",
+      "nTh ursday",
+      "n Thursday",
+      "Every thing",
+      "参与 到",
+      "Di al",
+      "D ial",
+      "▁E TH",
+      "▁ET H",
+      "▁ ETH",
+      "nb ase",
+      "n base",
+      "请在 给定的",
+      "接下来 您需要",
+      "该问题 需要对",
+      "回答一个 选择题",
+      "继续执行 的条件",
+      "各种信息进行 详细评估",
+      "▁stream lined",
+      "▁streamline d",
+      "接下来您需要 回答一个选择题",
+      "请在给定的 选项中挑选出正确答案",
+      "该问题需要对 各种信息进行详细评估",
+      "改 版",
+      "硕 果",
+      "体育 局",
+      "有 原告",
+      "步行 街",
+      "视为 其",
+      "视 为其",
+      "大 户",
+      "的 规律",
+      "▁li ters",
+      "▁liter s",
+      "▁lit ers",
+      "▁under grad",
+      "trans former",
+      "transform er",
+      "▁friend ships",
+      "▁friends hips",
+      "▁friendship s",
+      "恩 怨",
+      "常见 病",
+      "nd at",
+      "nda t",
+      "n dat",
+      "▁D ish",
+      "▁Dis h",
+      "▁Di sh",
+      "▁F avorite",
+      "▁cerebro spinal",
+      "平 价",
+      "犯 了",
+      "笑 语",
+      "萧 条",
+      "重 塑",
+      "的高 端",
+      "的 高端",
+      "▁s ua",
+      "▁su a",
+      "}} }\\\\",
+      "}}} \\\\",
+      "} }}\\\\",
+      "▁a vian",
+      "▁av ian",
+      "▁ avian",
+      "▁e sters",
+      "▁est ers",
+      "▁es ters",
+      "▁este rs",
+      "▁ester s",
+      "▁ esters",
+      "▁INCIDENT AL",
+      "房 内",
+      "汉 代",
+      "▁i os",
+      "▁io s",
+      "▁ ios",
+      "在线 汉语",
+      "og rad",
+      "ogr ad",
+      "o grad",
+      "ip otent",
+      "▁sew age",
+      "▁C ycling",
+      "▁Cycl ing",
+      "American s",
+      "Americ ans",
+      "America ns",
+      "▁knock down",
+      "皇 子",
+      "阴 暗",
+      "雅 典",
+      "{} /",
+      "{ }/",
+      "Mut ex",
+      "Mu tex",
+      "▁M ali",
+      "▁Mal i",
+      "▁Ma li",
+      "▁Th ur",
+      "▁Thu r",
+      "▁C ause",
+      "▁Ca use",
+      "▁p aran",
+      "▁par an",
+      "▁para n",
+      "▁pa ran",
+      "两次以上的 影人",
+      "预测 问题的答案",
+      "Re strict",
+      "Rest rict",
+      "res istance",
+      "ostruct ures",
+      "o structures",
+      "▁imprison ed",
+      "▁ 图",
+      "展 区",
+      "越 发",
+      "经营 部",
+      "平方 米的",
+      "平方米 的",
+      "资产 阶级",
+      "资 产阶级",
+      "▁R ash",
+      "▁Ra sh",
+      "▁Ras h",
+      "并从 提供的",
+      "▁Sax ony",
+      "一段文字 描述和",
+      "请思考 以下有关",
+      "▁ap ology",
+      "▁apolog y",
+      "▁apo logy",
+      "下面是一个 选择题",
+      "▁under pin",
+      "仔细 预测问题的答案",
+      "备选答案 中选择一个",
+      "专业领域 知识的选择题",
+      "▁un ivariate",
+      "我会给你 一段文字描述和",
+      "▁mind fulness",
+      "备选答案中选择一个 正确的选项",
+      "我会给你一段文字描述和 对应的选项",
+      "请你 从选项中选出你认为正确的答案",
+      "万 分",
+      "乔 丹",
+      "住所 地在",
+      "住所地 在",
+      "额外的 床",
+      "麻烦 帮我做",
+      "con vex",
+      "conv ex",
+      "▁tw ists",
+      "▁twist s",
+      "麻烦帮我做 一道题",
+      "nin stance",
+      "n instance",
+      "▁wing span",
+      "▁wings pan",
+      "se tdefault",
+      "set default",
+      "从选项中 选出正确的选项",
+      "从选项中选出 正确的选项",
+      "▁imprison ment",
+      "储 藏",
+      "卑 微",
+      "护 卫",
+      "AB B",
+      "A BB",
+      "LT E",
+      "L TE",
+      "我不 能",
+      "我 不能",
+      "设立 的",
+      "man i",
+      "ma ni",
+      "m ani",
+      "工作 温度",
+      "▁K ell",
+      "▁Ke ll",
+      "▁Kel l",
+      "nSe ptember",
+      "n September",
+      "会 社",
+      "分 发",
+      "前 线",
+      "包 的",
+      "即 在",
+      "打 仗",
+      "▁g t",
+      "▁ gt",
+      "的解集 为",
+      "这个 地方",
+      "ick son",
+      "icks on",
+      "label led",
+      "lab elled",
+      "г и",
+      "▁ Ç",
+      "宽 的",
+      "手 柄",
+      "难 民",
+      "年代 的",
+      "年 代的",
+      "▁个 整数",
+      "张 额外的床",
+      "▁sw arm",
+      "委托代理人 杨",
+      "▁re sect",
+      "▁res ect",
+      "launch er",
+      "▁straw berry",
+      "也 从",
+      "产 学",
+      "我 却",
+      "玉 龙",
+      "ño s",
+      "ñ os",
+      "ot us",
+      "一个 好的",
+      "英雄 合击",
+      "▁go og",
+      "▁ goog",
+      "▁} ),\\",
+      "▁}) ,\\",
+      "▁ }),\\",
+      "申请再审 称",
+      "▁more over",
+      "说 来",
+      "iz u",
+      "i zu",
+      "jo h",
+      "j oh",
+      "承德 市",
+      "承 德市",
+      "on ga",
+      "ong a",
+      "▁[ ])",
+      "▁[] )",
+      "▁ [])",
+      "返还 原告",
+      "顾 名思义",
+      "▁K ris",
+      "▁Kr is",
+      "▁Cr ohn",
+      "▁Cro hn",
+      "▁Graph s",
+      "▁prior i",
+      "▁pri ori",
+      "▁Mod ular",
+      "▁un locked",
+      "▁unlock ed",
+      "▁合作 两次以上的影人",
+      "▁Character istics",
+      "▁ 하",
+      "也 知道",
+      "有效 果",
+      "有 效果",
+      "网游 之",
+      "▁m ars",
+      "▁mar s",
+      "▁ma rs",
+      "ikh ail",
+      "st ellar",
+      "▁S unset",
+      "▁Sun set",
+      "▁Su nset",
+      "▁fet uses",
+      "▁fetus es",
+      "▁locom otive",
+      "▁vol umetric",
+      "▁eryth rocyte",
+      "▁Manufact urers",
+      "nP y",
+      "n Py",
+      "输 卵管",
+      "ch arm",
+      "char m",
+      "cha rm",
+      "c harm",
+      "en ers",
+      "ener s",
+      "ene rs",
+      "e ners",
+      "▁ju ices",
+      "▁juice s",
+      "▁s olute",
+      "▁sol ute",
+      "Media Type",
+      "七日内 预交案件受理费",
+      "备 忘",
+      "多 米",
+      "科 比",
+      "([ ^",
+      "( [^",
+      "三轮 车",
+      "依次 为",
+      "加床 或",
+      "系数 为",
+      "系 数为",
+      "▁N CI",
+      "▁NC I",
+      "正在 进行",
+      "正 在进行",
+      "第十五 章",
+      "第 十五章",
+      "\\' )]\\",
+      "\\') ]\\",
+      "rel ude",
+      "relu de",
+      "▁St arter",
+      "▁Star ter",
+      "▁Start er",
+      "▁\\\\ \\\\\\\\\\\\",
+      "▁\\\\\\\\ \\\\\\\\",
+      "▁\\\\\\ \\\\\\\\\\",
+      "▁\\\\\\\\\\ \\\\\\",
+      "▁ \\\\\\\\\\\\\\\\",
+      "▁antidepress ant",
+      "上 官",
+      "人 道",
+      "特 权",
+      "玉 树",
+      "网 讯",
+      "赤 壁",
+      "Du e",
+      "D ue",
+      "多余 的",
+      "青春 期",
+      "IO US",
+      "一定 程度",
+      "新农村 建设",
+      "▁M aven",
+      "▁Ma ven",
+      "申请执行人 同意",
+      "▁blot ting",
+      "iter ations",
+      "iteration s",
+      "Http Response",
+      "▁transport ing",
+      "▁ 魔",
+      "与 被",
+      "冲 压",
+      "国 标",
+      "技 师",
+      "村 村",
+      "颇 具",
+      "Sl f",
+      "S lf",
+      "以及 几个",
+      "机械 工业",
+      "运用 扎实",
+      "▁H olt",
+      "▁Hol t",
+      "▁Ho lt",
+      "Ar thur",
+      "▁c utter",
+      "▁cut ter",
+      "▁wh ichever",
+      "▁which ever",
+      "延 时",
+      "男 篮",
+      "缓 和",
+      "az zo",
+      "azz o",
+      "当 且仅当",
+      "ig ins",
+      "igin s",
+      "igi ns",
+      "▁C orb",
+      "▁Co rb",
+      "▁Cor b",
+      "履行 生效法律文书所确定的义务",
+      "履行生效法律文书 所确定的义务",
+      "宛 如",
+      "月 末",
+      "背 着",
+      "DL L",
+      "D LL",
+      "人心 的",
+      "人 心的",
+      "人性 的",
+      "人 性的",
+      "▁N aked",
+      "▁Nak ed",
+      "▁test is",
+      "▁Gaz ette",
+      "▁model led",
+      "▁mod elled",
+      "全 军",
+      "动 向",
+      "改 写成",
+      "程序 的",
+      "RE SP",
+      "RES P",
+      "R ESP",
+      "▁f en",
+      "▁fe n",
+      "▁ fen",
+      "全体 员工",
+      "至少 需要",
+      "ge nome",
+      "gen ome",
+      "▁off ence",
+      "L j",
+      "庆 区",
+      "温 带",
+      "睡 了",
+      "铁 岭",
+      "根据 其",
+      "鉴定 书",
+      "▁K oz",
+      "▁Ko z",
+      "的真实 性",
+      "的 真实性",
+      "PS HOT",
+      "ui ser",
+      "uis er",
+      "u iser",
+      "▁T ina",
+      "▁Ti na",
+      "▁Tin a",
+      "▁se ab",
+      "▁sea b",
+      "DO CKER",
+      "▁L ever",
+      "▁Le ver",
+      "▁Lev er",
+      "所提出的 任何",
+      "As String",
+      "sym metric",
+      "s ymmetric",
+      "nM ur",
+      "n Mur",
+      "上诉 案件",
+      "均需 获得",
+      "并从 所给",
+      "▁W heat",
+      "▁Wh eat",
+      "这道题 需要您",
+      "ic ially",
+      "ici ally",
+      "icial ly",
+      "接下来的 选择题",
+      "FILE NAME",
+      "nDis play",
+      "n Display",
+      "以便 作出正确判断",
+      "并从所给 选项中挑",
+      "运用扎实 的知识和",
+      "▁Pack ages",
+      "▁Package s",
+      "▁ Packages",
+      "Pro jection",
+      "Project ion",
+      "请仔细研究 接下来的选择题",
+      "运用扎实的知识和 仔细的思考",
+      "并从所给选项中挑 选出正确的答案",
+      "久 之",
+      "容 貌",
+      "nL os",
+      "nLo s",
+      "▁E is",
+      "tM onday",
+      "t Monday",
+      "▁glyc oproteins",
+      "▁glycoprotein s",
+      "报 社",
+      "禁 烟",
+      "电脑 上",
+      "政法 大学",
+      "最 简单的",
+      "Join t",
+      "Jo int",
+      "J oint",
+      "乘法 分配律",
+      "书 香",
+      "冬 梅",
+      "锯 齿",
+      "fu r",
+      "f ur",
+      "▁$ <",
+      "我想 说",
+      "我 想说",
+      "起重 机",
+      "ni ca",
+      "nic a",
+      "n ica",
+      "▁P ras",
+      "▁Pr as",
+      "▁Pra s",
+      "酒店的 确认",
+      "ap unov",
+      "mac ros",
+      "macro s",
+      "加床或 婴儿床",
+      "的要求 均需获得",
+      "▁Base ment",
+      "▁Bas ement",
+      "▁both ered",
+      "▁bother ed",
+      "▁mach ining",
+      "\\\\)\\\\( {\\\\,\\\\",
+      "所提出的任何 加床或婴儿床",
+      "丽 娟",
+      "使 君",
+      "朝 天",
+      "co w",
+      "c ow",
+      "vi e",
+      "v ie",
+      "做的 事",
+      "做 的事",
+      "如下 文",
+      "） **",
+      "Sn ap",
+      "S nap",
+      "这种 问题",
+      "具有 较高的",
+      "spl ice",
+      "▁Di vine",
+      "▁Div ine",
+      "▁cr ises",
+      "▁cris es",
+      "人 中",
+      "<% @",
+      "R abbit",
+      "▁f avors",
+      "▁favor s",
+      "▁fav ors",
+      "▁str ang",
+      "pro grams",
+      "pr ograms",
+      "program s",
+      "▁Ge Force",
+      "▁Ph antom",
+      "otrans fer",
+      "o transfer",
+      "▁intra ocular",
+      "不 爽",
+      "两 部",
+      "屋 子",
+      "绪 论",
+      "遍 地",
+      "西 城区",
+      "可以 用来",
+      "可以用 来",
+      "是世界 上",
+      "是 世界上",
+      "▁US DA",
+      "▁USD A",
+      "月获得 表扬",
+      "ga lact",
+      "gal act",
+      "的要求均需获得 酒店的确认",
+      "欣 慰",
+      "辛 亥",
+      "可以 多",
+      "而 努力",
+      "Vo te",
+      "V ote",
+      "物品 清单",
+      "系统 工程",
+      "▁sal ads",
+      "▁salad s",
+      "集团 有限责任公司",
+      "View Group",
+      "Entity Framework",
+      "内 皮",
+      "牛 津",
+      "遮 挡",
+      "mp i",
+      "m pi",
+      "在 很多",
+      "我们 去",
+      "活力 的",
+      "活 力的",
+      "科尔 沁",
+      "\\\\ ([",
+      "\\\\( [",
+      "sl ope",
+      "▁all ied",
+      "▁Pione er",
+      "▁re calls",
+      "▁rec alls",
+      "▁recall s",
+      "eterm inate",
+      "圣 地",
+      "巢 湖",
+      "慈 溪",
+      "细 雨",
+      "迈 向",
+      "齐 聚",
+      "▁H aj",
+      "▁Ha j",
+      "互联网 金融",
+      "▁Rel ax",
+      "▁Bel ieve",
+      "▁agree ing",
+      "美 满",
+      "赣 州",
+      "走 着",
+      "▁V acc",
+      "▁Va cc",
+      "▁Vac c",
+      "▁Jul ius",
+      "▁G rowing",
+      "▁Gr owing",
+      "▁Gro wing",
+      "▁Grow ing",
+      "▁I brahim",
+      "▁cl auses",
+      "▁clause s",
+      "▁Fac ilities",
+      "宽 广",
+      "od r",
+      "o dr",
+      "▁K ul",
+      "▁Ku l",
+      "nr ead",
+      "nre ad",
+      "n read",
+      "▁软件 大小",
+      "Mi rror",
+      "Mir ror",
+      "▁m ates",
+      "▁mat es",
+      "▁ma tes",
+      "▁mate s",
+      "▁ mates",
+      "▁c types",
+      "▁ct ypes",
+      "博 大",
+      "开 挖",
+      "AR N",
+      "A RN",
+      "ía s",
+      "í as",
+      "▁k l",
+      "▁ kl",
+      "▁사 용",
+      "年前 的",
+      "年 前的",
+      "连载 的一部",
+      "连载的 一部",
+      "Prom pt",
+      "ol ogie",
+      "olog ie",
+      "olo gie",
+      "▁sc ram",
+      "▁scr am",
+      "Sc enario",
+      "▁en tails",
+      "▁ent ails",
+      "▁prophyl actic",
+      "敲 门",
+      "正 交",
+      "药 水",
+      "集 训",
+      "但因 为",
+      "但 因为",
+      "房屋 的",
+      "▁Me V",
+      "PAT CH",
+      "P ATCH",
+      "fil ms",
+      "film s",
+      "鄂尔多斯 市",
+      "Ill egal",
+      "Il legal",
+      "PRO PERTY",
+      "strand ed",
+      "ieve ments",
+      "iev ements",
+      "ievement s",
+      "serial ized",
+      "serialize d",
+      "PM I",
+      "P MI",
+      "拒绝 了",
+      "脑子 里",
+      "脑 子里",
+      "LO GO",
+      "LOG O",
+      "▁h ors",
+      "▁hor s",
+      "▁ho rs",
+      "▁eryth emat",
+      "▁eryt hemat",
+      "▁plan etary",
+      "▁planet ary",
+      "会 话",
+      "寿 光",
+      "知 网",
+      "不开 心",
+      "不 开心",
+      "战略 性",
+      "▁[ **",
+      "认为 自己",
+      "认 为自己",
+      "▁J PEG",
+      "▁ JPEG",
+      "▁ap pre",
+      "▁app re",
+      "▁d emon",
+      "▁de mon",
+      "▁dem on",
+      "▁demo n",
+      "▁t uples",
+      "▁tuple s",
+      "▁tu ples",
+      "▁conduct ors",
+      "▁conductor s",
+      "无 端",
+      "欣 喜",
+      "的美 好",
+      "的 美好",
+      "连载 的",
+      "▁War rior",
+      "▁dist utils",
+      "▁mal formations",
+      "奠 基",
+      "蠕 动",
+      "光驱 类型",
+      "Br ain",
+      "B rain",
+      "rank ed",
+      "ole cyst",
+      "olec yst",
+      "▁mor tar",
+      "▁mort ar",
+      "▁F utures",
+      "▁Future s",
+      "▁Fu tures",
+      "▁over arching",
+      "喜 庆",
+      "Cl i",
+      "C li",
+      "今天 就",
+      "an os",
+      "ano s",
+      "a nos",
+      "ts ys",
+      "t sys",
+      "迄今 为止",
+      "退役 军人",
+      "▁V and",
+      "▁Van d",
+      "▁Va nd",
+      "me eting",
+      "meet ing",
+      "▁ru ined",
+      "▁ruin ed",
+      "▁def enses",
+      "▁defense s",
+      "▁transc end",
+      "in struction",
+      "两 头",
+      "几 倍",
+      "勒 索",
+      "秦 淮",
+      "表 单",
+      "し ます",
+      "▁t ul",
+      "▁tu l",
+      "Render ing",
+      "R endering",
+      "房屋 买卖合同纠纷一案",
+      "身 在",
+      "人家 的",
+      "人 家的",
+      "▁p ng",
+      "▁ png",
+      "dir ty",
+      "d irty",
+      "代理权限 为",
+      "W izard",
+      "ent eric",
+      "enter ic",
+      "ente ric",
+      "▁M orton",
+      "▁Mor ton",
+      "▁Mort on",
+      "三 连",
+      "推 特",
+      "研 判",
+      "站 台",
+      "蚝 油",
+      "Co d",
+      "C od",
+      "FI X",
+      "F IX",
+      "ds l",
+      "d sl",
+      "▁电 子",
+      "▁ 电子",
+      "长时间 的",
+      "长 时间的",
+      "▁V era",
+      "▁Ver a",
+      "▁Ve ra",
+      "发展 中国家",
+      "发展中 国家",
+      "▁o auth",
+      "▁ oauth",
+      "▁St ones",
+      "▁Stone s",
+      "Ch annels",
+      "Channel s",
+      "Chan nels",
+      "▁em peror",
+      "乳 头",
+      "之后 就",
+      "之 后就",
+      "我心 里",
+      "我 心里",
+      "DE ST",
+      "DES T",
+      "D EST",
+      "Mon o",
+      "Mo no",
+      "M ono",
+      "▁( []",
+      "▁([ ]",
+      "▁ ([]",
+      "▁r av",
+      "▁ra v",
+      "▁ rav",
+      "将 进一步",
+      "▁d atos",
+      "▁dat os",
+      "▁da tos",
+      "▁po uring",
+      "▁pour ing",
+      "▁pou ring",
+      "amp hetamine",
+      "]* *",
+      "] **",
+      "▁A cad",
+      "▁Ac ad",
+      "▁b lew",
+      "▁bl ew",
+      "▁ble w",
+      "对其 从轻处罚",
+      "miss ive",
+      "▁Back up",
+      "▁ Backup",
+      "▁n esting",
+      "▁nest ing",
+      "嗓 子",
+      "某 于",
+      "SC R",
+      "S CR",
+      "▁l p",
+      "▁ lp",
+      "她们 的",
+      "她 们的",
+      "软 组织",
+      "LO NG",
+      "LON G",
+      "L ONG",
+      "理论 基础",
+      "Sp anish",
+      "Span ish",
+      "rest aurant",
+      "▁tre asures",
+      "▁treasure s",
+      "▁treas ures",
+      "▁mosquito es",
+      "▁mosquit oes",
+      "▁not eworthy",
+      "▁note worthy",
+      "▁ 严",
+      "八 九",
+      "毅 然",
+      "钟 情",
+      "': '",
+      "' :'",
+      "也很 难",
+      "也 很难",
+      "爱 企查",
+      "As ia",
+      "▁A Ch",
+      "▁AC h",
+      "▁N ina",
+      "▁Ni na",
+      "▁Nin a",
+      "▁o wed",
+      "▁ow ed",
+      "▁owe d",
+      "▁ owed",
+      "Ge nder",
+      "Gen der",
+      "G ender",
+      "Ne sted",
+      "N ested",
+      "nL ouis",
+      "nLo uis",
+      "n Louis",
+      "▁Bo tan",
+      "▁Bot an",
+      "▁R each",
+      "▁Re ach",
+      "▁Le ipzig",
+      "▁Vel ocity",
+      "▁ Velocity",
+      "▁ ②",
+      "方 剂",
+      "不相 信",
+      "不 相信",
+      "商业 街",
+      "地下 室",
+      "资料 来源",
+      "▁W eld",
+      "▁We ld",
+      "▁Wel d",
+      "由题意 可知",
+      "UR EMENT",
+      "URE MENT",
+      "quant um",
+      "in frared",
+      "短 发",
+      "耐 热",
+      "bi r",
+      "b ir",
+      "▁M ö",
+      "▁试 读",
+      "的单 位",
+      "的 单位",
+      "休闲 小憩",
+      "▁M eat",
+      "▁Me at",
+      "中 刑执字第",
+      "▁N inja",
+      "▁Nin ja",
+      "note book",
+      "▁prompt ing",
+      "▁in sensitive",
+      "▁ins ensitive",
+      "从 上",
+      "法 检测",
+      "特点 和",
+      "特 点和",
+      "的人 是",
+      "的 人是",
+      "看到 这个",
+      "这种 情况下",
+      "这种情况 下",
+      "▁Viol ence",
+      "▁tax ation",
+      "▁taxa tion",
+      "南 宫",
+      "娄 底",
+      "比 尔",
+      "灵 异",
+      "里 奥",
+      "执行 逮捕",
+      "ess ie",
+      "occup ied",
+      "▁Not Found",
+      "▁ NotFound",
+      "▁Numer ous",
+      "▁ground breaking",
+      "平 庸",
+      "林 的",
+      "江 海",
+      "点 开",
+      "球 星",
+      "RI C",
+      "R IC",
+      "be h",
+      "b eh",
+      "▁V ie",
+      "▁Vi e",
+      "行政 部门",
+      "▁M alt",
+      "▁Mal t",
+      "▁Ma lt",
+      "nAsk ed",
+      "nDep artment",
+      "n Department",
+      "▁prote olytic",
+      "▁approxim ated",
+      "▁approx imated",
+      "▁approximate d",
+      "更 佳",
+      "西 宁",
+      "aq u",
+      "a qu",
+      "▁C z",
+      "外星 人",
+      "工作 组",
+      "时间 去",
+      "但是 没有",
+      "但 是没有",
+      "在不同 的",
+      "在 不同的",
+      "显着 高于",
+      "服务 器的",
+      "服务器 的",
+      "这组 数据",
+      "这 组数据",
+      "tun signed",
+      "t unsigned",
+      "▁inter oper",
+      "t E",
+      "中 生",
+      "我 更",
+      "普 罗",
+      "涂 装",
+      "喜欢 上",
+      "aut y",
+      "au ty",
+      "a uty",
+      "mer e",
+      "me re",
+      "m ere",
+      "nD ave",
+      "nE ric",
+      "nEr ic",
+      "n Eric",
+      "▁th irst",
+      "▁thi rst",
+      "di methyl",
+      "dim ethyl",
+      "又 好",
+      "吃 掉",
+      "过 失",
+      "▁C p",
+      "Br on",
+      "Bro n",
+      "B ron",
+      "▁X in",
+      "▁Xi n",
+      "▁K ant",
+      "▁Ka nt",
+      "▁Kan t",
+      "如下文 所示",
+      "▁Anth rop",
+      "▁spect ators",
+      "乡 长",
+      "滑 坡",
+      "苍 茫",
+      "▁i ç",
+      "塑料 袋",
+      "定义 在",
+      "管理 等",
+      "nN et",
+      "nNe t",
+      "n Net",
+      "tr ees",
+      "tree s",
+      "tre es",
+      "t rees",
+      "▁caf é",
+      "▁V ault",
+      "▁Va ult",
+      "▁ Vault",
+      "▁r ains",
+      "▁ra ins",
+      "▁rain s",
+      "▁obl ique",
+      "▁Http Servlet",
+      "▁ HttpServlet",
+      "不 以",
+      "之 都",
+      "当下 的",
+      "当 下的",
+      "新 模式",
+      "循序 渐进",
+      "ne mail",
+      "nem ail",
+      "n email",
+      "Make file",
+      "check sum",
+      "checks um",
+      "ere quisites",
+      "▁ 健",
+      "下 颌",
+      "书 生",
+      "分析 仪",
+      "在 欧洲",
+      "委 党校",
+      "LIC K",
+      "LI CK",
+      "L ICK",
+      "▁I PO",
+      "▁IP O",
+      "▁ IPO",
+      "▁Z am",
+      "冻结 被告",
+      "积极 作用",
+      "nB rian",
+      "nBr ian",
+      "▁Ch oir",
+      "▁Cho ir",
+      "▁s ided",
+      "▁side d",
+      "▁si ded",
+      "▁sid ed",
+      "▁ sided",
+      "Object Id",
+      "Token izer",
+      "左 键",
+      "贪 婪",
+      "逆 向",
+      "▁ 医院",
+      "不住 了",
+      "不 住了",
+      "乐 山市",
+      "打 不过",
+      "▁V od",
+      "▁Vo d",
+      "▁p av",
+      "▁pa v",
+      "以为 自己",
+      "以 为自己",
+      "分析 步骤",
+      "等 农作物",
+      "▁s lit",
+      "▁sl it",
+      "做一个 选择题",
+      "pr inter",
+      "print er",
+      "pri nter",
+      "▁ 剑",
+      "上 当",
+      "伴 奏",
+      "抓 取",
+      "十九 年",
+      "十 九年",
+      "bys hev",
+      "ing les",
+      "ingle s",
+      "▁graph ite",
+      "▁clip board",
+      "▁emphas izing",
+      "≤ \\",
+      "岳 麓",
+      "木 工",
+      "CB A",
+      "C BA",
+      "收入 的",
+      "邢台 市",
+      "▁L or",
+      "▁Lo r",
+      "是单 选题",
+      "▁D ust",
+      "▁Du st",
+      "EMPL ARY",
+      "▁fe athers",
+      "▁feather s",
+      "à o",
+      "寄 存",
+      "布 料",
+      "悠 然",
+      "我 行",
+      "河 边",
+      "达 拉",
+      "不对 称",
+      "不 对称",
+      "体积 是",
+      "▁a gr",
+      "▁ag r",
+      "▁ agr",
+      "▁ID EA",
+      "▁IDE A",
+      "nL iter",
+      "n Liter",
+      "▁g ears",
+      "▁ge ars",
+      "▁gear s",
+      "▁H ouses",
+      "▁House s",
+      "▁Ho uses",
+      "▁Hou ses",
+      "▁trim med",
+      "▁stand point",
+      "揭 牌",
+      "fe lt",
+      "f elt",
+      "身体 素质",
+      "▁w hip",
+      "▁wh ip",
+      "ver ses",
+      "vers es",
+      "verse s",
+      "▁sh utter",
+      "▁shut ter",
+      "in tensive",
+      "int ensive",
+      "valid ated",
+      "validate d",
+      "▁w arrants",
+      "▁warrant s",
+      "▁def endants",
+      "▁defend ants",
+      "▁defendant s",
+      "为 谁",
+      "划 线",
+      "妈 咪",
+      "足 协",
+      "题 名",
+      "le i",
+      "l ei",
+      "sin A",
+      "si nA",
+      "▁$ {{\\\\",
+      "▁${ {\\\\",
+      "▁${{ \\\\",
+      "悔改表现 突出",
+      "pat ible",
+      "▁bass ist",
+      "▁ å",
+      "网易云 音乐",
+      "▁Vol vo",
+      "nM obile",
+      "n Mobile",
+      "▁免费 洗浴用品",
+      "ellig ence",
+      "elli gence",
+      "何 在",
+      "毅 力",
+      "畏 惧",
+      "总 承包",
+      "用 除法",
+      "的 局面",
+      "def s",
+      "de fs",
+      "ic om",
+      "ico m",
+      "i com",
+      "nO UT",
+      "n OUT",
+      "no st",
+      "nos t",
+      "n ost",
+      "nthe se",
+      "nth ese",
+      "n these",
+      "right s",
+      "r ights",
+      "▁Cl erk",
+      "▁Cle rk",
+      "nEx ercise",
+      "nSyn opsis",
+      "▁option ally",
+      "▁optional ly",
+      "▁sil houette",
+      "▁recomm ending",
+      "▁recommend ing",
+      "不 回",
+      "心 血",
+      "所 著",
+      "是 哪",
+      "神经 元",
+      "Ch ief",
+      "cent os",
+      "这道题 是单选题",
+      "请你 做一个选择题",
+      "请你 选出正确的选项",
+      "▁ship ments",
+      "▁shipment s",
+      "问题和选项 如下文所示",
+      "▁Ar ithmetic",
+      "悲 惨",
+      "找 寻",
+      "素 的",
+      "虚 空",
+      "pe p",
+      "p ep",
+      "ta nd",
+      "tan d",
+      "t and",
+      "uc kle",
+      "uck le",
+      "Des pite",
+      "all ocate",
+      "alloc ate",
+      "▁optim ally",
+      "▁optimal ly",
+      "▁ver ifying",
+      "▁verify ing",
+      "史 学",
+      "竹 马",
+      "件事 情",
+      "件 事情",
+      "的热 爱",
+      "的 热爱",
+      "\\' \",",
+      "\\'\" ,",
+      "or na",
+      "orn a",
+      "全面 推进",
+      "nW ait",
+      "n Wait",
+      "im ilar",
+      "imi lar",
+      "nM useum",
+      "▁disc iples",
+      "▁discipl es",
+      "▁dem olition",
+      "江 门",
+      "调 速",
+      "▁p its",
+      "▁pi ts",
+      "▁pit s",
+      "ex posure",
+      "exp osure",
+      "▁reimburse ment",
+      "天 蝎",
+      "扫 黑",
+      "遴 选",
+      "▁r ut",
+      "▁ru t",
+      "ns ign",
+      "n sign",
+      "这是一个 涉及",
+      "▁cass ette",
+      "了 很久",
+      "将 达到",
+      "ST AR",
+      "STA R",
+      "ac om",
+      "aco m",
+      "a com",
+      "建立 一个",
+      "艺术 设计",
+      "▁ves icle",
+      "▁POSS IBILITY",
+      "川 市",
+      "nA bs",
+      "nAb s",
+      "n Abs",
+      "就是 我们",
+      "就是我 们",
+      "就 是我们",
+      "▁cal endars",
+      "▁calendar s",
+      "▁butter flies",
+      "伤 寒",
+      "特 价",
+      "眉 头",
+      "骑 车",
+      "Es c",
+      "E sc",
+      "我都 会",
+      "我 都会",
+      "la in",
+      "l ain",
+      "分析 并从",
+      "yz ing",
+      "y zing",
+      "▁L aud",
+      "▁La ud",
+      "▁fant as",
+      "▁fan tas",
+      "user Info",
+      "▁loc ating",
+      "▁descript ors",
+      "▁descriptor s",
+      "▁progen itors",
+      "▁progenitor s",
+      "▁R X",
+      "▁ RX",
+      "赡养 费",
+      "Se nse",
+      "Sen se",
+      "S ense",
+      "故意杀人 罪",
+      "▁kn ives",
+      "bi ological",
+      "bio logical",
+      "b iological",
+      "▁Berks hire",
+      "▁conjug ates",
+      "▁conjugate s",
+      "▁incl ination",
+      "▁sub projects",
+      "▁subproject s",
+      "▁ 春",
+      "扰 乱",
+      "睁 开",
+      "虫 子",
+      "BM P",
+      "B MP",
+      "MA R",
+      "M AR",
+      "人员 和",
+      "文学 奖",
+      "管理 水平",
+      "as ers",
+      "ase rs",
+      "aser s",
+      "ron ics",
+      "ronic s",
+      "▁C Make",
+      "▁CM ake",
+      "▁ CMake",
+      "▁st ark",
+      "▁star k",
+      "▁Cor pus",
+      "▁Corp us",
+      "▁S odium",
+      "四 章",
+      "泰 兴",
+      "纤 细",
+      ":] ,",
+      ": ],",
+      "年 上海",
+      "示 意图",
+      "等 其他",
+      "结构 与",
+      "信息的 选择题",
+      "给定的 选项中",
+      "▁Tuc son",
+      "您需要 分析并从",
+      "▁Q String",
+      "▁ QString",
+      "请仔细研究 以下问题",
+      "综合运用各种 信息的选择题",
+      "您需要分析并从 给定的选项中",
+      "▁' $",
+      "▁ '$",
+      "与我 们",
+      "与 我们",
+      "UB LE",
+      "U BLE",
+      "▁b ye",
+      "▁by e",
+      "▁ bye",
+      "候选 答案",
+      "候 选答案",
+      "基本 参数",
+      "说法 错误",
+      "进一步 的",
+      "▁K hal",
+      "▁Kh al",
+      "▁Gab ri",
+      "▁toss ed",
+      "▁con gest",
+      "▁cong est",
+      "pr eferences",
+      "pre ferences",
+      "prefer ences",
+      "p references",
+      "▁cyber security",
+      "鸭 子",
+      "▁[ /",
+      "▁ [/",
+      "Pro of",
+      "nM aking",
+      "nMa king",
+      "n Making",
+      "set Data",
+      "▁gener osity",
+      "▁integr ative",
+      "▁H Y",
+      "▁ HY",
+      "压力 大",
+      "▁B am",
+      "▁Ba m",
+      "伟大 复兴",
+      "感兴趣 的",
+      "抽样 调查",
+      "▁bake ry",
+      "▁bak ery",
+      "nAl gebra",
+      "n Algebra",
+      "Type Error",
+      "▁Web Socket",
+      "▁ WebSocket",
+      "▁embr acing",
+      "物业管理 有限责任公司",
+      "▁counter example",
+      "苏 丹",
+      "Si g",
+      "S ig",
+      "[{ \"",
+      "[ {\"",
+      "▁名 称",
+      "▁ 名称",
+      "▁E MS",
+      "▁EM S",
+      "交通 工具",
+      "目前 没有",
+      "if era",
+      "ife ra",
+      "ifer a",
+      "of ten",
+      "oft en",
+      "o ften",
+      "下面 我会给你",
+      "的撤诉申请 符合法律规定",
+      "▁perox idation",
+      "匠 心",
+      "防 潮",
+      "GS M",
+      "G SM",
+      "只是 个",
+      "只 是个",
+      "的 方案",
+      "ur se",
+      "urs e",
+      "中山 大学",
+      "wh ole",
+      "who le",
+      "w hole",
+      "请你仔细 分析",
+      "ir ected",
+      "irect ed",
+      "▁Iceland ic",
+      "微 小",
+      "筹 集",
+      "▁ 最新",
+      "如果 想要",
+      "如果想 要",
+      "▁golf ers",
+      "▁gol fers",
+      "以及几个 候选答案",
+      "▁over dose",
+      "下面我会给你 一道题",
+      "然后 选出正确的选项",
+      "▁counsel or",
+      "▁mar riages",
+      "▁marriage s",
+      "打 发",
+      "贪 心",
+      "输 血",
+      "黑 板",
+      "Hy p",
+      "H yp",
+      "bl a",
+      "b la",
+      "rn e",
+      "r ne",
+      "可以 给",
+      "衢 州市",
+      "过来 了",
+      "过 来了",
+      "重生 之",
+      "St rip",
+      "Str ip",
+      "S trip",
+      "ul ton",
+      "ult on",
+      "ys sey",
+      "▁M ons",
+      "▁Mon s",
+      "▁Mo ns",
+      "▁m anic",
+      "▁man ic",
+      "▁ma nic",
+      "uttg art",
+      "▁comorbid ities",
+      "j r",
+      "农 资",
+      "断 层",
+      "▁U M",
+      "▁ UM",
+      "到 什么",
+      "目标 和",
+      "▁S UR",
+      "▁SU R",
+      "城市 规划",
+      "nM ade",
+      "nMa de",
+      "nMad e",
+      "n Made",
+      "▁Ram sey",
+      "▁Rams ey",
+      "▁you ths",
+      "▁youth s",
+      "▁valid ating",
+      "对 本",
+      "放 出",
+      "缔 造",
+      "mv c",
+      "m vc",
+      "的地 步",
+      "行为 规范",
+      "银行 同期",
+      "▁d ean",
+      "▁de an",
+      "▁pe dig",
+      "▁ped ig",
+      "▁ali ens",
+      "▁alien s",
+      "▁d ances",
+      "▁dance s",
+      "▁dan ces",
+      "信息查询 结果单",
+      "Neg ative",
+      "nun iform",
+      "n uniform",
+      "unning ham",
+      "unn ingham",
+      "▁Ware house",
+      "▁period ont",
+      "▁peri odont",
+      "▁ultrason ography",
+      "又 说",
+      "▁g ib",
+      "▁gi b",
+      "ub ern",
+      "ube rn",
+      "uber n",
+      "▁c ler",
+      "▁cl er",
+      "▁cle r",
+      "▁ cler",
+      "# ,",
+      "七 天",
+      "声 中",
+      "大 名",
+      "悲 观",
+      "与 世界",
+      "准 考证",
+      "gp io",
+      "▁W it",
+      "▁Wi t",
+      "th ick",
+      "thi ck",
+      "申请执行人 杨",
+      "违反 法律规定",
+      "real path",
+      "▁royal ty",
+      "▁kin ematic",
+      "栖 霞",
+      "疏 导",
+      "▁c z",
+      "▁ cz",
+      "基本 功",
+      "(\" {}",
+      "(\"{ }",
+      "▁L una",
+      "▁Lu na",
+      "▁thr ombo",
+      "▁throm bo",
+      "▁thromb o",
+      "text width",
+      "tex twidth",
+      "▁d isciplinary",
+      "▁disc iplinary",
+      "▁discipl inary",
+      "▁ disciplinary",
+      "仙 子",
+      "舅 舅",
+      "颤 抖",
+      "IC P",
+      "I CP",
+      "两 方面",
+      "还能 够",
+      "还 能够",
+      "▁w get",
+      "▁ch ili",
+      "▁chi li",
+      "oplas ma",
+      "opl asma",
+      "oplasm a",
+      "▁Se gment",
+      "▁Seg ment",
+      "▁ Segment",
+      "▁fin alist",
+      "▁final ist",
+      "九 九",
+      "固 有",
+      "李 子",
+      "症 的",
+      "变更 为",
+      "变 更为",
+      "赔偿 金",
+      "E con",
+      "▁L OL",
+      "▁LO L",
+      "▁U SC",
+      "▁US C",
+      "高职 高专",
+      "▁S lide",
+      "▁Sl ide",
+      "▁ Slide",
+      "▁t orus",
+      "▁to rus",
+      "▁tor us",
+      "▁Har ley",
+      "▁Willi s",
+      "▁Will is",
+      "▁Wil lis",
+      "ac oustic",
+      "▁emb roidery",
+      "北 辰",
+      "各 区",
+      "妆 容",
+      "截 然",
+      "$_ {",
+      "$ _{",
+      "委 常委",
+      "审查 核实",
+      "▁Cl iff",
+      "▁Clif f",
+      "扣划 被执行人",
+      "▁Buddh a",
+      "▁Budd ha",
+      "▁is lets",
+      "▁islet s",
+      "Enum erator",
+      "▁Record ings",
+      "▁Recording s",
+      "墓 地",
+      "官 司",
+      "霍 尔",
+      "▁E z",
+      "这两 天",
+      "这 两天",
+      "mu ted",
+      "mut ed",
+      "m uted",
+      "en dale",
+      "end ale",
+      "enda le",
+      "▁Re alty",
+      "▁Real ty",
+      "▁L isting",
+      "▁List ing",
+      "▁Lis ting",
+      "▁Man aged",
+      "▁Manag ed",
+      "▁Manage d",
+      "▁ Managed",
+      "不 讲",
+      "光 亮",
+      "大 观",
+      "日 产",
+      "杀 虫",
+      "Cr e",
+      "C re",
+      "equ e",
+      "eq ue",
+      "e que",
+      "um ba",
+      "umb a",
+      "▁un i",
+      "▁u ni",
+      "▁ uni",
+      "一定 要有",
+      "一定要 有",
+      "自由 恋爱",
+      "▁tr ilogy",
+      "▁tri logy",
+      "即 得",
+      "可 观",
+      "大 黄",
+      "轻 便",
+      "迎 春",
+      "ud p",
+      "u dp",
+      "▁ 第五章",
+      "从而 导致",
+      "从 而导致",
+      "▁s talk",
+      "▁st alk",
+      "▁ stalk",
+      "nc urrent",
+      "n current",
+      "▁All ison",
+      "▁dep ended",
+      "▁depend ed",
+      "▁catal ysis",
+      "▁under line",
+      "▁ underline",
+      "伊 始",
+      "和 张",
+      "白 白",
+      "CS I",
+      "C SI",
+      "HA L",
+      "H AL",
+      "也 同样",
+      "▁P UT",
+      "▁PU T",
+      "▁ PUT",
+      "▁d ia",
+      "▁di a",
+      "nK evin",
+      "nKe vin",
+      "▁Nic kel",
+      "▁Nick el",
+      "Docker file",
+      "▁visual ized",
+      "▁visualize d",
+      "l x",
+      "刀 片",
+      "青 花",
+      "调味 料",
+      "mod s",
+      "mo ds",
+      "m ods",
+      "为真 命题",
+      "质量 监督",
+      "▁si RNA",
+      "法定代表人 杨",
+      "▁war rior",
+      "水 区",
+      "肉 体",
+      "了 不起",
+      "永远 的",
+      "永 远的",
+      "Ch an",
+      "C han",
+      "nI ll",
+      "n Ill",
+      "优质 服务",
+      "▁W istar",
+      "▁Wis tar",
+      "▁Wi star",
+      "▁inv oices",
+      "▁invoice s",
+      "▁inter ruption",
+      "▁interrupt ion",
+      "略 微",
+      "租 用",
+      "än d",
+      "ä nd",
+      "▁B N",
+      "▁ BN",
+      "in tl",
+      "int l",
+      "▁P AN",
+      "▁PA N",
+      "两组 患者",
+      "工业 企业",
+      "最大限度 地",
+      "Cl ients",
+      "Client s",
+      "Cli ents",
+      "▁sum mers",
+      "▁summer s",
+      "▁summ ers",
+      "▁a merican",
+      "▁foster ing",
+      "▁fost ering",
+      "nG astropods",
+      "▁councill ors",
+      "养 育",
+      "差 额",
+      "心 痛",
+      "的热 点",
+      "的 热点",
+      "临床 应用",
+      "比较 简单",
+      "规划 教材",
+      "Save d",
+      "Sa ved",
+      "S aved",
+      "wire d",
+      "wi red",
+      "w ired",
+      "▁Cr ash",
+      "▁Cra sh",
+      "enn ifer",
+      "▁min ors",
+      "▁minor s",
+      "▁pill ars",
+      "▁Child hood",
+      "一 架",
+      "加 斯",
+      "待 定",
+      "意 气",
+      "中国 梦",
+      "热烈 的",
+      "热 烈的",
+      "▁c han",
+      "▁ch an",
+      "▁cha n",
+      "▁ chan",
+      "ud ding",
+      "udd ing",
+      "▁C oral",
+      "▁Co ral",
+      "▁Cor al",
+      "brand ed",
+      "v matrix",
+      "▁tax onomy",
+      "Measure ment",
+      "▁trans plants",
+      "▁transplant s",
+      "小 便",
+      "不过 在",
+      "为人 民",
+      "为 人民",
+      "▁F rem",
+      "▁Fr em",
+      "▁Fre m",
+      "es ture",
+      "est ure",
+      "▁is cha",
+      "Min utes",
+      "other wise",
+      "▁Le besgue",
+      "Australia n",
+      "A ustralian",
+      "oph aryngeal",
+      "trans formed",
+      "transform ed",
+      "▁ 商",
+      "制 宜",
+      "备 课",
+      "nM i",
+      "n Mi",
+      "教学 法",
+      "涉及 的",
+      "ar ie",
+      "ari e",
+      "a rie",
+      "水平的 提高",
+      "符合 条件的",
+      "▁cob alt",
+      "▁Ham mond",
+      "▁Hamm ond",
+      "▁Power ed",
+      "▁Pow ered",
+      "▁let tuce",
+      "▁plas mon",
+      "▁plasm on",
+      "▁safe guard",
+      "▁safegu ard",
+      "í k",
+      "八 首",
+      "ax on",
+      "ha el",
+      "h ael",
+      "\"] );\\",
+      "\"]) ;\\",
+      "\" ]);\\",
+      "▁g ast",
+      "▁gas t",
+      "▁ga st",
+      "olog na",
+      "▁D ancing",
+      "▁Dan cing",
+      "▁t opping",
+      "▁to pping",
+      "▁top ping",
+      "astructure s",
+      "astruct ures",
+      "a structures",
+      "亚 目",
+      "会 和",
+      "抗 病",
+      "缅 怀",
+      "英 才",
+      ">} \\",
+      "> }\\",
+      "单 职业",
+      "喜欢 看",
+      "数字 是",
+      "▁N eb",
+      "▁Ne b",
+      "Sp ect",
+      "Spec t",
+      "Spe ct",
+      "S pect",
+      "ur sed",
+      "urs ed",
+      "urse d",
+      "BIN ARY",
+      "cent ering",
+      "center ing",
+      "Linked List",
+      "墨 水",
+      "抒 情",
+      "省 道",
+      "股 指",
+      "Sa h",
+      "S ah",
+      "太大 了",
+      "太 大了",
+      "可以 利用",
+      "法律 关系",
+      "▁de nim",
+      "▁den im",
+      "▁ou nce",
+      "▁oun ce",
+      "▁ ounce",
+      "▁Vis itors",
+      "▁Visit ors",
+      "▁inoc ulated",
+      "▁macro scopic",
+      "吃 不",
+      "昂 贵",
+      "蒸 馏",
+      "NT s",
+      "N Ts",
+      "收入 来源",
+      "▁K ens",
+      "▁Ke ns",
+      "▁Ken s",
+      "作品 展",
+      "▁par athyroid",
+      "壮 观",
+      "散 户",
+      "沧 州",
+      "痛 风",
+      "书法 家",
+      "剩余 的",
+      "文化 教育",
+      "st rain",
+      "str ain",
+      "stra in",
+      "s train",
+      "ENAB LED",
+      "nTur kish",
+      "▁ball oons",
+      "▁balloon s",
+      "▁taxp ayer",
+      "遵守 法律法规及监规",
+      "▁dilig ence",
+      "归 案",
+      "朝 代",
+      "赫 尔",
+      "韵 味",
+      "竞争 中",
+      "▁E ur",
+      "Test er",
+      "Te ster",
+      "T ester",
+      "nP ress",
+      "nPre ss",
+      "nPr ess",
+      "nPres s",
+      "n Press",
+      "▁sk aters",
+      "▁skate rs",
+      "▁Com mands",
+      "▁Command s",
+      "▁Comm ands",
+      "▁ Commands",
+      "▁prof icient",
+      "文 革",
+      "耐 久",
+      "耗 时",
+      "阻 断",
+      "雄 蕊",
+      "碧 桂园",
+      ">\" )\\",
+      "> \")\\",
+      "▁L IN",
+      "▁LI N",
+      "▁ LIN",
+      "住 湖北省",
+      "看到 一个",
+      "看 到一个",
+      "▁N ass",
+      "▁Na ss",
+      "▁Nas s",
+      "ann ual",
+      "工作 领导小组",
+      "ilib ili",
+      "▁get away",
+      "▁electro card",
+      "▁electr ocard",
+      "▁electroc ard",
+      "警 官",
+      "可以 和",
+      "light er",
+      "l ighter",
+      "▁Bo xing",
+      "▁Box ing",
+      "▁r hesus",
+      "▁rh esus",
+      "▁rhe sus",
+      "▁Log ging",
+      "▁ Logging",
+      "声 声",
+      "ST S",
+      "S TS",
+      "原 标题",
+      "应 支付",
+      "▁H ak",
+      "▁Ha k",
+      "sl ave",
+      "共计 人民币",
+      "是不 一样的",
+      "是 不一样的",
+      "▁Card iac",
+      "▁Cardi ac",
+      "▁check sum",
+      "▁checks um",
+      "▁ checksum",
+      "▁chrom ium",
+      "▁ chromium",
+      "经相关部门批准后 方可开展经营活动",
+      "中 点",
+      "乙 酸",
+      "台 账",
+      "在 有",
+      "bs t",
+      "b st",
+      "▁т е",
+      "▁ те",
+      "No me",
+      "Nom e",
+      "N ome",
+      "▁E TF",
+      "▁ET F",
+      "▁ ETF",
+      "欢迎 所有",
+      "环境 优美",
+      "稳产 农田",
+      "并给出 解析",
+      "▁Tool kit",
+      "▁Sup erman",
+      "▁Super man",
+      "▁onc ogenic",
+      "▁re written",
+      "▁trans istors",
+      "▁transistor s",
+      "* $",
+      "M b",
+      "▁ 车",
+      "上 榜",
+      "武 陵",
+      "连 城",
+      "ary nx",
+      "▁F uck",
+      "▁Fu ck",
+      "▁P ied",
+      "▁Pi ed",
+      "▁Pie d",
+      "▁BU ILD",
+      "▁ BUILD",
+      "se tName",
+      "set Name",
+      "▁d umped",
+      "▁dump ed",
+      "▁gl owing",
+      "▁glo wing",
+      "▁glow ing",
+      "▁hum oral",
+      "▁humor al",
+      "toHave Been",
+      "▁com pressive",
+      "▁comp ressive",
+      "▁compr essive",
+      "▁compress ive",
+      "ä l",
+      "下 有",
+      "切 碎",
+      "钱 财",
+      "雪 梅",
+      "风 起",
+      "生活 和",
+      "_( \\'",
+      "_ (\\'",
+      "▁M VC",
+      "▁MV C",
+      "Mar ie",
+      "Ma rie",
+      "M arie",
+      "ER ENCE",
+      "▁Bo hem",
+      "▁Boh em",
+      "▁CA USED",
+      "▁epit ope",
+      "▁go odies",
+      "▁good ies",
+      "▁follic le",
+      "equ ivalent",
+      "equiv alent",
+      "▁C omposition",
+      "▁Com position",
+      "▁Comp osition",
+      "选出你认为正确的答案 并给出解析",
+      "S X",
+      "乐 曲",
+      "Fe e",
+      "F ee",
+      "是 负数",
+      "顺德 区",
+      "iv ar",
+      "iva r",
+      "i var",
+      "其他 国家",
+      "er eal",
+      "ere al",
+      "e real",
+      "engu ins",
+      "enguin s",
+      "ra ising",
+      "rais ing",
+      "▁course work",
+      "宝 妈",
+      "应 从",
+      "时 限",
+      "统 统",
+      "黄 酮",
+      "西北 部",
+      "西 北部",
+      "bm od",
+      "b mod",
+      "nB al",
+      "n Bal",
+      "▁A gr",
+      "▁Ag r",
+      "Mac ro",
+      "Ma cro",
+      "ag her",
+      "agh er",
+      "有限公司 与被告",
+      "▁special ties",
+      "又 无",
+      "朱 元",
+      "能 与",
+      "锦 江",
+      "的 计划",
+      "个人 观点",
+      "文明 城市",
+      "Order ed",
+      "Ord ered",
+      "▁Alber to",
+      "▁Alb erto",
+      "▁Albert o",
+      "▁Default s",
+      "▁ Defaults",
+      "▁piece wise",
+      "拨 款",
+      "版 块",
+      "只是 想",
+      "只 是想",
+      "pin ion",
+      "▁hal ves",
+      "犯 容留他人吸毒罪",
+      "▁bi ophysical",
+      "▁bio physical",
+      "后 才",
+      "国 泰",
+      "则 这个",
+      "OP PO",
+      "▁s ut",
+      "▁su t",
+      "谈话 笔录",
+      "▁r aster",
+      "▁ra ster",
+      "▁Gr avity",
+      "▁ex iting",
+      "▁exit ing",
+      "县 市",
+      "可 想",
+      "日 将",
+      "某 在",
+      "肥 皂",
+      "蜂 窝",
+      ")} (",
+      ") }(",
+      "社会 上",
+      "社 会上",
+      "财产 的",
+      "财 产的",
+      "es tr",
+      "est r",
+      "e str",
+      "▁E Ps",
+      "▁EP s",
+      "写出 你的",
+      "▁注意 事项",
+      "▁ 注意事项",
+      "Rot ate",
+      "opa edic",
+      "▁ch angelog",
+      "▁chang elog",
+      "▁change log",
+      "▁ changelog",
+      "▁pos itional",
+      "▁position al",
+      "▁posit ional",
+      "▁initial State",
+      "拖 动",
+      "hu n",
+      "h un",
+      "▁S v",
+      "}\" ;\\",
+      "} \";\\",
+      "应该 怎么",
+      "应 该怎么",
+      "nST ATIC",
+      "n STATIC",
+      "c w",
+      "某 天",
+      "及时 的",
+      "及 时的",
+      "都 十分",
+      "▁g an",
+      "▁ga n",
+      "▁ gan",
+      "二〇一六年 四",
+      "im iento",
+      "imi ento",
+      "▁rot ary",
+      "S aturday",
+      "aut omatic",
+      "auto matic",
+      "autom atic",
+      "挂 着",
+      "耕 耘",
+      "锻 造",
+      "ub i",
+      "u bi",
+      "з ов",
+      "▁ 纠错",
+      "的 培养",
+      "RO LE",
+      "ROL E",
+      "▁P LEASE",
+      "充分不 必要条件",
+      "▁In jection",
+      "▁Inject ion",
+      "▁Inj ection",
+      "▁ Injection",
+      "自 闭",
+      "诗 意",
+      "差异 化",
+      "系 主任",
+      "图文 并茂",
+      "▁Rub ber",
+      "▁Mich ele",
+      "▁Michel e",
+      "▁Miche le",
+      "▁den ying",
+      "▁deny ing",
+      "na issance",
+      "号征收社会抚养费 决定书",
+      "▁multi national",
+      "▁multin ational",
+      "▁function alities",
+      "▁functional ities",
+      "痛 经",
+      "yc h",
+      "y ch",
+      "自贡 市",
+      "▁r ho",
+      "▁rh o",
+      "▁ rho",
+      "no vel",
+      "nov el",
+      "n ovel",
+      "Ma king",
+      "M aking",
+      "li able",
+      "lia ble",
+      "l iable",
+      "▁inhal ed",
+      "▁inh aled",
+      "ne ighbors",
+      "▁Wild erness",
+      "▁suppress es",
+      "▁ 同",
+      "剖 宫",
+      "为 公司",
+      "待 今后",
+      "第四 次",
+      "第 四次",
+      "le on",
+      "l eon",
+      "▁C NC",
+      "▁CN C",
+      "ble rs",
+      "bl ers",
+      "bler s",
+      "b lers",
+      "▁h aha",
+      "▁ha ha",
+      "▁作品 原文",
+      "▁pol yt",
+      "▁poly t",
+      "安徽省 合肥市",
+      "▁bo oster",
+      "▁boost er",
+      "▁fascin ated",
+      "▁fasc inated",
+      "▁ Ω",
+      "思 绪",
+      "桥 头",
+      "芭 蕾",
+      "三个 人",
+      "三 个人",
+      "被 发现",
+      "▁P es",
+      "▁Pe s",
+      "▁s aga",
+      "▁sa ga",
+      "▁sag a",
+      "or ylation",
+      "ory lation",
+      "▁allerg en",
+      "▁re distribution",
+      "▁redist ribution",
+      "湛 江",
+      "补 给",
+      "▁r ag",
+      "▁ra g",
+      "▁ rag",
+      "lang s",
+      "lan gs",
+      "l angs",
+      "And rew",
+      "▁Cl air",
+      "▁ign ores",
+      "▁ignore s",
+      "▁ignor es",
+      "▁re factor",
+      "▁ref actor",
+      "▁ad versely",
+      "▁adverse ly",
+      "▁advers ely",
+      "▁end oscopy",
+      "▁ 集",
+      "跳 动",
+      "鹤 壁",
+      "你 分析",
+      "▁orient ations",
+      "▁orientation s",
+      "▁quant ization",
+      "有点 像",
+      "贵 金属",
+      "습 니다",
+      "▁C RP",
+      "▁CR P",
+      "▁ CRP",
+      "▁J ah",
+      "▁Ja h",
+      "▁A min",
+      "▁Am in",
+      "主张 的事实",
+      "ouch ers",
+      "oucher s",
+      "▁Harm ony",
+      "▁spe eding",
+      "▁speed ing",
+      "▁after ward",
+      "卫生 和计划生育委员会",
+      "是连载于起点中文网 的一部",
+      "第一百五十四条第一款第 五",
+      "第一百五十四条第一款 第五",
+      "前 几",
+      "生 怕",
+      "都 与",
+      "休闲 娱乐",
+      "Hy dro",
+      "H ydro",
+      "Wr ong",
+      "W rong",
+      "第八 十四条",
+      "▁pop ul",
+      "▁V ehicles",
+      "▁Vehicle s",
+      "▁ О",
+      "吞 吐",
+      "在 点"
+    ]
+  }
+}
\ No newline at end of file
diff --git a/tokenizer_config.json b/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1f46debaef91a9fa2ddbd4646878f27693b9404d
--- /dev/null
+++ b/tokenizer_config.json
@@ -0,0 +1,52 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{'<用户>' + message['content'].strip() + '<AI>'}}{% else %}{{message['content'].strip()}}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/trainer_state.json b/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..dd34f426b17ea20ee61da403df29f80a04105b67
--- /dev/null
+++ b/trainer_state.json
@@ -0,0 +1,318169 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 14.0,
+  "eval_steps": 500,
+  "global_step": 136346,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0003082297338949964,
+      "grad_norm": 0.5791122913360596,
+      "learning_rate": 1.2325390304026295e-05,
+      "loss": 11.5149,
+      "step": 3
+    },
+    {
+      "epoch": 0.0006164594677899928,
+      "grad_norm": 1.7733268737792969,
+      "learning_rate": 2.465078060805259e-05,
+      "loss": 11.504,
+      "step": 6
+    },
+    {
+      "epoch": 0.0009246892016849893,
+      "grad_norm": 3.047468423843384,
+      "learning_rate": 3.697617091207888e-05,
+      "loss": 11.4391,
+      "step": 9
+    },
+    {
+      "epoch": 0.0012329189355799856,
+      "grad_norm": 3.346104621887207,
+      "learning_rate": 4.930156121610518e-05,
+      "loss": 11.2635,
+      "step": 12
+    },
+    {
+      "epoch": 0.001541148669474982,
+      "grad_norm": 3.2847304344177246,
+      "learning_rate": 6.162695152013147e-05,
+      "loss": 10.9864,
+      "step": 15
+    },
+    {
+      "epoch": 0.0018493784033699785,
+      "grad_norm": 3.2979729175567627,
+      "learning_rate": 7.395234182415776e-05,
+      "loss": 10.6487,
+      "step": 18
+    },
+    {
+      "epoch": 0.0021576081372649747,
+      "grad_norm": 3.445396661758423,
+      "learning_rate": 8.627773212818406e-05,
+      "loss": 10.2667,
+      "step": 21
+    },
+    {
+      "epoch": 0.0024658378711599712,
+      "grad_norm": 3.620011568069458,
+      "learning_rate": 9.860312243221036e-05,
+      "loss": 9.8679,
+      "step": 24
+    },
+    {
+      "epoch": 0.002774067605054968,
+      "grad_norm": 3.66719913482666,
+      "learning_rate": 0.00011092851273623665,
+      "loss": 9.4532,
+      "step": 27
+    },
+    {
+      "epoch": 0.003082297338949964,
+      "grad_norm": 3.709792375564575,
+      "learning_rate": 0.00012325390304026294,
+      "loss": 9.0393,
+      "step": 30
+    },
+    {
+      "epoch": 0.0033905270728449605,
+      "grad_norm": 3.5757813453674316,
+      "learning_rate": 0.00013557929334428925,
+      "loss": 8.6449,
+      "step": 33
+    },
+    {
+      "epoch": 0.003698756806739957,
+      "grad_norm": 3.200732707977295,
+      "learning_rate": 0.00014790468364831553,
+      "loss": 8.2932,
+      "step": 36
+    },
+    {
+      "epoch": 0.004006986540634953,
+      "grad_norm": 2.7463502883911133,
+      "learning_rate": 0.0001602300739523418,
+      "loss": 8.0151,
+      "step": 39
+    },
+    {
+      "epoch": 0.004315216274529949,
+      "grad_norm": 1.961776852607727,
+      "learning_rate": 0.00017255546425636812,
+      "loss": 7.7988,
+      "step": 42
+    },
+    {
+      "epoch": 0.004623446008424946,
+      "grad_norm": 1.5326848030090332,
+      "learning_rate": 0.0001848808545603944,
+      "loss": 7.6688,
+      "step": 45
+    },
+    {
+      "epoch": 0.0049316757423199424,
+      "grad_norm": 2.266148090362549,
+      "learning_rate": 0.00019720624486442071,
+      "loss": 7.6276,
+      "step": 48
+    },
+    {
+      "epoch": 0.005239905476214939,
+      "grad_norm": 2.563206672668457,
+      "learning_rate": 0.000209531635168447,
+      "loss": 7.5954,
+      "step": 51
+    },
+    {
+      "epoch": 0.005548135210109936,
+      "grad_norm": 1.9326294660568237,
+      "learning_rate": 0.0002218570254724733,
+      "loss": 7.5136,
+      "step": 54
+    },
+    {
+      "epoch": 0.005856364944004932,
+      "grad_norm": 1.0131096839904785,
+      "learning_rate": 0.0002341824157764996,
+      "loss": 7.4167,
+      "step": 57
+    },
+    {
+      "epoch": 0.006164594677899928,
+      "grad_norm": 1.1923354864120483,
+      "learning_rate": 0.00024650780608052587,
+      "loss": 7.3273,
+      "step": 60
+    },
+    {
+      "epoch": 0.006472824411794925,
+      "grad_norm": 1.2690105438232422,
+      "learning_rate": 0.00025883319638455215,
+      "loss": 7.2615,
+      "step": 63
+    },
+    {
+      "epoch": 0.006781054145689921,
+      "grad_norm": 0.8375588059425354,
+      "learning_rate": 0.0002711585866885785,
+      "loss": 7.2088,
+      "step": 66
+    },
+    {
+      "epoch": 0.007089283879584917,
+      "grad_norm": 0.719011127948761,
+      "learning_rate": 0.0002834839769926048,
+      "loss": 7.1474,
+      "step": 69
+    },
+    {
+      "epoch": 0.007397513613479914,
+      "grad_norm": 0.7095780372619629,
+      "learning_rate": 0.00029580936729663106,
+      "loss": 7.0923,
+      "step": 72
+    },
+    {
+      "epoch": 0.00770574334737491,
+      "grad_norm": 0.5094121098518372,
+      "learning_rate": 0.00030813475760065734,
+      "loss": 7.0406,
+      "step": 75
+    },
+    {
+      "epoch": 0.008013973081269906,
+      "grad_norm": 0.4085525572299957,
+      "learning_rate": 0.0003204601479046836,
+      "loss": 6.9586,
+      "step": 78
+    },
+    {
+      "epoch": 0.008322202815164903,
+      "grad_norm": 0.4122146666049957,
+      "learning_rate": 0.0003327855382087099,
+      "loss": 6.9295,
+      "step": 81
+    },
+    {
+      "epoch": 0.008630432549059899,
+      "grad_norm": 0.41669413447380066,
+      "learning_rate": 0.00034511092851273624,
+      "loss": 6.8629,
+      "step": 84
+    },
+    {
+      "epoch": 0.008938662282954896,
+      "grad_norm": 0.40379494428634644,
+      "learning_rate": 0.0003574363188167625,
+      "loss": 6.7987,
+      "step": 87
+    },
+    {
+      "epoch": 0.009246892016849893,
+      "grad_norm": 0.4202142059803009,
+      "learning_rate": 0.0003697617091207888,
+      "loss": 6.7779,
+      "step": 90
+    },
+    {
+      "epoch": 0.009555121750744888,
+      "grad_norm": 0.3880140781402588,
+      "learning_rate": 0.0003820870994248151,
+      "loss": 6.7204,
+      "step": 93
+    },
+    {
+      "epoch": 0.009863351484639885,
+      "grad_norm": 0.419883668422699,
+      "learning_rate": 0.00039441248972884143,
+      "loss": 6.6553,
+      "step": 96
+    },
+    {
+      "epoch": 0.010171581218534882,
+      "grad_norm": 0.4570387601852417,
+      "learning_rate": 0.0004067378800328677,
+      "loss": 6.5936,
+      "step": 99
+    },
+    {
+      "epoch": 0.010479810952429877,
+      "grad_norm": 0.5889524221420288,
+      "learning_rate": 0.000419063270336894,
+      "loss": 6.5271,
+      "step": 102
+    },
+    {
+      "epoch": 0.010788040686324874,
+      "grad_norm": 0.5578189492225647,
+      "learning_rate": 0.0004313886606409203,
+      "loss": 6.4701,
+      "step": 105
+    },
+    {
+      "epoch": 0.011096270420219871,
+      "grad_norm": 0.4710846245288849,
+      "learning_rate": 0.0004437140509449466,
+      "loss": 6.4254,
+      "step": 108
+    },
+    {
+      "epoch": 0.011404500154114866,
+      "grad_norm": 0.30558162927627563,
+      "learning_rate": 0.0004560394412489729,
+      "loss": 6.4217,
+      "step": 111
+    },
+    {
+      "epoch": 0.011712729888009863,
+      "grad_norm": 0.3634621500968933,
+      "learning_rate": 0.0004683648315529992,
+      "loss": 6.3422,
+      "step": 114
+    },
+    {
+      "epoch": 0.01202095962190486,
+      "grad_norm": 0.369437575340271,
+      "learning_rate": 0.00048069022185702546,
+      "loss": 6.303,
+      "step": 117
+    },
+    {
+      "epoch": 0.012329189355799856,
+      "grad_norm": 0.7418856024742126,
+      "learning_rate": 0.0004930156121610517,
+      "loss": 6.2535,
+      "step": 120
+    },
+    {
+      "epoch": 0.012637419089694853,
+      "grad_norm": 0.855678915977478,
+      "learning_rate": 0.000505341002465078,
+      "loss": 6.2037,
+      "step": 123
+    },
+    {
+      "epoch": 0.01294564882358985,
+      "grad_norm": 0.42414671182632446,
+      "learning_rate": 0.0005176663927691043,
+      "loss": 6.175,
+      "step": 126
+    },
+    {
+      "epoch": 0.013253878557484845,
+      "grad_norm": 0.44572392106056213,
+      "learning_rate": 0.0005299917830731307,
+      "loss": 6.1548,
+      "step": 129
+    },
+    {
+      "epoch": 0.013562108291379842,
+      "grad_norm": 0.24437515437602997,
+      "learning_rate": 0.000542317173377157,
+      "loss": 6.0967,
+      "step": 132
+    },
+    {
+      "epoch": 0.013870338025274839,
+      "grad_norm": 0.4750615954399109,
+      "learning_rate": 0.0005546425636811833,
+      "loss": 6.0776,
+      "step": 135
+    },
+    {
+      "epoch": 0.014178567759169834,
+      "grad_norm": 0.9504273533821106,
+      "learning_rate": 0.0005669679539852095,
+      "loss": 6.0471,
+      "step": 138
+    },
+    {
+      "epoch": 0.014486797493064831,
+      "grad_norm": 0.5452646613121033,
+      "learning_rate": 0.0005792933442892358,
+      "loss": 5.9987,
+      "step": 141
+    },
+    {
+      "epoch": 0.014795027226959828,
+      "grad_norm": 0.4314074218273163,
+      "learning_rate": 0.0005916187345932621,
+      "loss": 5.9874,
+      "step": 144
+    },
+    {
+      "epoch": 0.015103256960854823,
+      "grad_norm": 0.48243793845176697,
+      "learning_rate": 0.0006039441248972884,
+      "loss": 5.962,
+      "step": 147
+    },
+    {
+      "epoch": 0.01541148669474982,
+      "grad_norm": 0.7570855617523193,
+      "learning_rate": 0.0006162695152013147,
+      "loss": 5.921,
+      "step": 150
+    },
+    {
+      "epoch": 0.015719716428644816,
+      "grad_norm": 0.4731276333332062,
+      "learning_rate": 0.000628594905505341,
+      "loss": 5.9032,
+      "step": 153
+    },
+    {
+      "epoch": 0.016027946162539813,
+      "grad_norm": 0.3078465759754181,
+      "learning_rate": 0.0006409202958093672,
+      "loss": 5.8932,
+      "step": 156
+    },
+    {
+      "epoch": 0.01633617589643481,
+      "grad_norm": 0.3441495895385742,
+      "learning_rate": 0.0006532456861133935,
+      "loss": 5.8695,
+      "step": 159
+    },
+    {
+      "epoch": 0.016644405630329807,
+      "grad_norm": 0.43056854605674744,
+      "learning_rate": 0.0006655710764174198,
+      "loss": 5.8642,
+      "step": 162
+    },
+    {
+      "epoch": 0.016952635364224804,
+      "grad_norm": 1.3783445358276367,
+      "learning_rate": 0.0006778964667214461,
+      "loss": 5.845,
+      "step": 165
+    },
+    {
+      "epoch": 0.017260865098119797,
+      "grad_norm": 0.77984619140625,
+      "learning_rate": 0.0006902218570254725,
+      "loss": 5.8216,
+      "step": 168
+    },
+    {
+      "epoch": 0.017569094832014794,
+      "grad_norm": 0.2990098297595978,
+      "learning_rate": 0.0007025472473294988,
+      "loss": 5.7801,
+      "step": 171
+    },
+    {
+      "epoch": 0.01787732456590979,
+      "grad_norm": 0.29485219717025757,
+      "learning_rate": 0.000714872637633525,
+      "loss": 5.7757,
+      "step": 174
+    },
+    {
+      "epoch": 0.018185554299804788,
+      "grad_norm": 0.4363936483860016,
+      "learning_rate": 0.0007271980279375513,
+      "loss": 5.7598,
+      "step": 177
+    },
+    {
+      "epoch": 0.018493784033699785,
+      "grad_norm": 0.8902605175971985,
+      "learning_rate": 0.0007395234182415776,
+      "loss": 5.7719,
+      "step": 180
+    },
+    {
+      "epoch": 0.018802013767594782,
+      "grad_norm": 0.7133249044418335,
+      "learning_rate": 0.0007518488085456039,
+      "loss": 5.7568,
+      "step": 183
+    },
+    {
+      "epoch": 0.019110243501489776,
+      "grad_norm": 0.9105846881866455,
+      "learning_rate": 0.0007641741988496302,
+      "loss": 5.7191,
+      "step": 186
+    },
+    {
+      "epoch": 0.019418473235384773,
+      "grad_norm": 0.9680726528167725,
+      "learning_rate": 0.0007764995891536565,
+      "loss": 5.7077,
+      "step": 189
+    },
+    {
+      "epoch": 0.01972670296927977,
+      "grad_norm": 0.535446047782898,
+      "learning_rate": 0.0007888249794576829,
+      "loss": 5.6982,
+      "step": 192
+    },
+    {
+      "epoch": 0.020034932703174767,
+      "grad_norm": 0.7894541621208191,
+      "learning_rate": 0.0008011503697617091,
+      "loss": 5.668,
+      "step": 195
+    },
+    {
+      "epoch": 0.020343162437069764,
+      "grad_norm": 0.6975138187408447,
+      "learning_rate": 0.0008134757600657354,
+      "loss": 5.6432,
+      "step": 198
+    },
+    {
+      "epoch": 0.02065139217096476,
+      "grad_norm": 0.6306262016296387,
+      "learning_rate": 0.0008258011503697617,
+      "loss": 5.6647,
+      "step": 201
+    },
+    {
+      "epoch": 0.020959621904859754,
+      "grad_norm": 0.5615081787109375,
+      "learning_rate": 0.000838126540673788,
+      "loss": 5.6208,
+      "step": 204
+    },
+    {
+      "epoch": 0.02126785163875475,
+      "grad_norm": 0.6468993425369263,
+      "learning_rate": 0.0008504519309778143,
+      "loss": 5.6272,
+      "step": 207
+    },
+    {
+      "epoch": 0.02157608137264975,
+      "grad_norm": 0.8359414339065552,
+      "learning_rate": 0.0008627773212818406,
+      "loss": 5.6114,
+      "step": 210
+    },
+    {
+      "epoch": 0.021884311106544745,
+      "grad_norm": 0.8909689784049988,
+      "learning_rate": 0.0008751027115858668,
+      "loss": 5.5957,
+      "step": 213
+    },
+    {
+      "epoch": 0.022192540840439742,
+      "grad_norm": 0.39673465490341187,
+      "learning_rate": 0.0008874281018898932,
+      "loss": 5.568,
+      "step": 216
+    },
+    {
+      "epoch": 0.022500770574334736,
+      "grad_norm": 0.9037743806838989,
+      "learning_rate": 0.0008997534921939195,
+      "loss": 5.5746,
+      "step": 219
+    },
+    {
+      "epoch": 0.022809000308229733,
+      "grad_norm": 0.6929497122764587,
+      "learning_rate": 0.0009120788824979458,
+      "loss": 5.5482,
+      "step": 222
+    },
+    {
+      "epoch": 0.02311723004212473,
+      "grad_norm": 0.5773665308952332,
+      "learning_rate": 0.0009244042728019721,
+      "loss": 5.5389,
+      "step": 225
+    },
+    {
+      "epoch": 0.023425459776019727,
+      "grad_norm": 0.9532020092010498,
+      "learning_rate": 0.0009367296631059984,
+      "loss": 5.5369,
+      "step": 228
+    },
+    {
+      "epoch": 0.023733689509914724,
+      "grad_norm": 1.2347012758255005,
+      "learning_rate": 0.0009490550534100246,
+      "loss": 5.5251,
+      "step": 231
+    },
+    {
+      "epoch": 0.02404191924380972,
+      "grad_norm": 1.2062091827392578,
+      "learning_rate": 0.0009613804437140509,
+      "loss": 5.5042,
+      "step": 234
+    },
+    {
+      "epoch": 0.024350148977704714,
+      "grad_norm": 1.4920969009399414,
+      "learning_rate": 0.0009737058340180772,
+      "loss": 5.4851,
+      "step": 237
+    },
+    {
+      "epoch": 0.02465837871159971,
+      "grad_norm": 0.5619600415229797,
+      "learning_rate": 0.0009860312243221035,
+      "loss": 5.4889,
+      "step": 240
+    },
+    {
+      "epoch": 0.02496660844549471,
+      "grad_norm": 0.8607615828514099,
+      "learning_rate": 0.0009983566146261299,
+      "loss": 5.4747,
+      "step": 243
+    },
+    {
+      "epoch": 0.025274838179389705,
+      "grad_norm": 0.6228588223457336,
+      "learning_rate": 0.001010682004930156,
+      "loss": 5.4502,
+      "step": 246
+    },
+    {
+      "epoch": 0.025583067913284702,
+      "grad_norm": 1.1925005912780762,
+      "learning_rate": 0.0010230073952341824,
+      "loss": 5.4449,
+      "step": 249
+    },
+    {
+      "epoch": 0.0258912976471797,
+      "grad_norm": 0.7956414818763733,
+      "learning_rate": 0.0010353327855382086,
+      "loss": 5.4623,
+      "step": 252
+    },
+    {
+      "epoch": 0.026199527381074693,
+      "grad_norm": 0.654242992401123,
+      "learning_rate": 0.001047658175842235,
+      "loss": 5.4287,
+      "step": 255
+    },
+    {
+      "epoch": 0.02650775711496969,
+      "grad_norm": 0.592880368232727,
+      "learning_rate": 0.0010599835661462614,
+      "loss": 5.3891,
+      "step": 258
+    },
+    {
+      "epoch": 0.026815986848864687,
+      "grad_norm": 0.9015865921974182,
+      "learning_rate": 0.0010723089564502876,
+      "loss": 5.4127,
+      "step": 261
+    },
+    {
+      "epoch": 0.027124216582759684,
+      "grad_norm": 0.593488335609436,
+      "learning_rate": 0.001084634346754314,
+      "loss": 5.3887,
+      "step": 264
+    },
+    {
+      "epoch": 0.02743244631665468,
+      "grad_norm": 0.7008156180381775,
+      "learning_rate": 0.0010969597370583401,
+      "loss": 5.386,
+      "step": 267
+    },
+    {
+      "epoch": 0.027740676050549678,
+      "grad_norm": 0.32653194665908813,
+      "learning_rate": 0.0011092851273623665,
+      "loss": 5.3479,
+      "step": 270
+    },
+    {
+      "epoch": 0.02804890578444467,
+      "grad_norm": 0.551142692565918,
+      "learning_rate": 0.0011216105176663927,
+      "loss": 5.3613,
+      "step": 273
+    },
+    {
+      "epoch": 0.02835713551833967,
+      "grad_norm": 2.3521084785461426,
+      "learning_rate": 0.001133935907970419,
+      "loss": 5.3653,
+      "step": 276
+    },
+    {
+      "epoch": 0.028665365252234665,
+      "grad_norm": 1.3452407121658325,
+      "learning_rate": 0.0011462612982744455,
+      "loss": 5.3559,
+      "step": 279
+    },
+    {
+      "epoch": 0.028973594986129662,
+      "grad_norm": 1.0670260190963745,
+      "learning_rate": 0.0011585866885784717,
+      "loss": 5.3299,
+      "step": 282
+    },
+    {
+      "epoch": 0.02928182472002466,
+      "grad_norm": 0.7768902778625488,
+      "learning_rate": 0.001170912078882498,
+      "loss": 5.3346,
+      "step": 285
+    },
+    {
+      "epoch": 0.029590054453919656,
+      "grad_norm": 0.48641496896743774,
+      "learning_rate": 0.0011832374691865242,
+      "loss": 5.3178,
+      "step": 288
+    },
+    {
+      "epoch": 0.02989828418781465,
+      "grad_norm": 0.5284126400947571,
+      "learning_rate": 0.0011955628594905506,
+      "loss": 5.3061,
+      "step": 291
+    },
+    {
+      "epoch": 0.030206513921709647,
+      "grad_norm": 0.9099608659744263,
+      "learning_rate": 0.0012078882497945768,
+      "loss": 5.2764,
+      "step": 294
+    },
+    {
+      "epoch": 0.030514743655604644,
+      "grad_norm": 0.7352691888809204,
+      "learning_rate": 0.0012202136400986032,
+      "loss": 5.2853,
+      "step": 297
+    },
+    {
+      "epoch": 0.03082297338949964,
+      "grad_norm": 0.8361043334007263,
+      "learning_rate": 0.0012325390304026294,
+      "loss": 5.2838,
+      "step": 300
+    },
+    {
+      "epoch": 0.031131203123394638,
+      "grad_norm": 1.525067925453186,
+      "learning_rate": 0.0012448644207066558,
+      "loss": 5.2612,
+      "step": 303
+    },
+    {
+      "epoch": 0.03143943285728963,
+      "grad_norm": 0.6117688417434692,
+      "learning_rate": 0.001257189811010682,
+      "loss": 5.2488,
+      "step": 306
+    },
+    {
+      "epoch": 0.03174766259118463,
+      "grad_norm": 0.9976358413696289,
+      "learning_rate": 0.0012695152013147083,
+      "loss": 5.2327,
+      "step": 309
+    },
+    {
+      "epoch": 0.032055892325079625,
+      "grad_norm": 0.8152816891670227,
+      "learning_rate": 0.0012818405916187345,
+      "loss": 5.2095,
+      "step": 312
+    },
+    {
+      "epoch": 0.03236412205897462,
+      "grad_norm": 0.8640046715736389,
+      "learning_rate": 0.0012941659819227609,
+      "loss": 5.1932,
+      "step": 315
+    },
+    {
+      "epoch": 0.03267235179286962,
+      "grad_norm": 0.9461572170257568,
+      "learning_rate": 0.001306491372226787,
+      "loss": 5.1822,
+      "step": 318
+    },
+    {
+      "epoch": 0.03298058152676461,
+      "grad_norm": 0.7717807292938232,
+      "learning_rate": 0.0013188167625308134,
+      "loss": 5.183,
+      "step": 321
+    },
+    {
+      "epoch": 0.03328881126065961,
+      "grad_norm": 0.9057526588439941,
+      "learning_rate": 0.0013311421528348396,
+      "loss": 5.1686,
+      "step": 324
+    },
+    {
+      "epoch": 0.03359704099455461,
+      "grad_norm": 0.5352618098258972,
+      "learning_rate": 0.001343467543138866,
+      "loss": 5.1378,
+      "step": 327
+    },
+    {
+      "epoch": 0.03390527072844961,
+      "grad_norm": 1.2399810552597046,
+      "learning_rate": 0.0013557929334428922,
+      "loss": 5.1436,
+      "step": 330
+    },
+    {
+      "epoch": 0.0342135004623446,
+      "grad_norm": 0.6678963303565979,
+      "learning_rate": 0.0013681183237469186,
+      "loss": 5.1488,
+      "step": 333
+    },
+    {
+      "epoch": 0.034521730196239594,
+      "grad_norm": 0.6166791915893555,
+      "learning_rate": 0.001380443714050945,
+      "loss": 5.1239,
+      "step": 336
+    },
+    {
+      "epoch": 0.034829959930134595,
+      "grad_norm": 1.1305850744247437,
+      "learning_rate": 0.0013927691043549711,
+      "loss": 5.1145,
+      "step": 339
+    },
+    {
+      "epoch": 0.03513818966402959,
+      "grad_norm": 0.46510085463523865,
+      "learning_rate": 0.0014050944946589975,
+      "loss": 5.1041,
+      "step": 342
+    },
+    {
+      "epoch": 0.03544641939792459,
+      "grad_norm": 0.4835362136363983,
+      "learning_rate": 0.0014174198849630237,
+      "loss": 5.0699,
+      "step": 345
+    },
+    {
+      "epoch": 0.03575464913181958,
+      "grad_norm": 0.6595330238342285,
+      "learning_rate": 0.00142974527526705,
+      "loss": 5.0744,
+      "step": 348
+    },
+    {
+      "epoch": 0.036062878865714576,
+      "grad_norm": 0.7306437492370605,
+      "learning_rate": 0.0014420706655710763,
+      "loss": 5.0703,
+      "step": 351
+    },
+    {
+      "epoch": 0.036371108599609576,
+      "grad_norm": 0.5263068079948425,
+      "learning_rate": 0.0014543960558751027,
+      "loss": 5.0666,
+      "step": 354
+    },
+    {
+      "epoch": 0.03667933833350457,
+      "grad_norm": 0.5896726250648499,
+      "learning_rate": 0.001466721446179129,
+      "loss": 5.0326,
+      "step": 357
+    },
+    {
+      "epoch": 0.03698756806739957,
+      "grad_norm": 0.9357500672340393,
+      "learning_rate": 0.0014790468364831552,
+      "loss": 5.0363,
+      "step": 360
+    },
+    {
+      "epoch": 0.037295797801294564,
+      "grad_norm": 0.7629897594451904,
+      "learning_rate": 0.0014913722267871816,
+      "loss": 5.0304,
+      "step": 363
+    },
+    {
+      "epoch": 0.037604027535189564,
+      "grad_norm": 0.6347280144691467,
+      "learning_rate": 0.0015036976170912078,
+      "loss": 4.9962,
+      "step": 366
+    },
+    {
+      "epoch": 0.03791225726908456,
+      "grad_norm": 0.4810947775840759,
+      "learning_rate": 0.0015160230073952342,
+      "loss": 4.9856,
+      "step": 369
+    },
+    {
+      "epoch": 0.03822048700297955,
+      "grad_norm": 0.5907162427902222,
+      "learning_rate": 0.0015283483976992604,
+      "loss": 4.9712,
+      "step": 372
+    },
+    {
+      "epoch": 0.03852871673687455,
+      "grad_norm": 0.5781192183494568,
+      "learning_rate": 0.0015406737880032868,
+      "loss": 5.0078,
+      "step": 375
+    },
+    {
+      "epoch": 0.038836946470769546,
+      "grad_norm": 0.6017566323280334,
+      "learning_rate": 0.001552999178307313,
+      "loss": 4.9563,
+      "step": 378
+    },
+    {
+      "epoch": 0.039145176204664546,
+      "grad_norm": 1.208348035812378,
+      "learning_rate": 0.0015653245686113393,
+      "loss": 4.9696,
+      "step": 381
+    },
+    {
+      "epoch": 0.03945340593855954,
+      "grad_norm": 0.6113926768302917,
+      "learning_rate": 0.0015776499589153657,
+      "loss": 4.9461,
+      "step": 384
+    },
+    {
+      "epoch": 0.03976163567245453,
+      "grad_norm": 0.6794010996818542,
+      "learning_rate": 0.0015899753492193919,
+      "loss": 4.9668,
+      "step": 387
+    },
+    {
+      "epoch": 0.040069865406349534,
+      "grad_norm": 0.4383271038532257,
+      "learning_rate": 0.0016023007395234183,
+      "loss": 4.9283,
+      "step": 390
+    },
+    {
+      "epoch": 0.04037809514024453,
+      "grad_norm": 0.9564613699913025,
+      "learning_rate": 0.0016146261298274444,
+      "loss": 4.8814,
+      "step": 393
+    },
+    {
+      "epoch": 0.04068632487413953,
+      "grad_norm": 0.6730177402496338,
+      "learning_rate": 0.0016269515201314708,
+      "loss": 4.9158,
+      "step": 396
+    },
+    {
+      "epoch": 0.04099455460803452,
+      "grad_norm": 0.5306158661842346,
+      "learning_rate": 0.001639276910435497,
+      "loss": 4.904,
+      "step": 399
+    },
+    {
+      "epoch": 0.04130278434192952,
+      "grad_norm": 0.48708540201187134,
+      "learning_rate": 0.0016516023007395234,
+      "loss": 4.9002,
+      "step": 402
+    },
+    {
+      "epoch": 0.041611014075824515,
+      "grad_norm": 0.4917944371700287,
+      "learning_rate": 0.0016639276910435496,
+      "loss": 4.8913,
+      "step": 405
+    },
+    {
+      "epoch": 0.04191924380971951,
+      "grad_norm": 1.0929678678512573,
+      "learning_rate": 0.001676253081347576,
+      "loss": 4.8986,
+      "step": 408
+    },
+    {
+      "epoch": 0.04222747354361451,
+      "grad_norm": 0.5417898297309875,
+      "learning_rate": 0.0016885784716516024,
+      "loss": 4.8702,
+      "step": 411
+    },
+    {
+      "epoch": 0.0425357032775095,
+      "grad_norm": 1.1427472829818726,
+      "learning_rate": 0.0017009038619556285,
+      "loss": 4.8396,
+      "step": 414
+    },
+    {
+      "epoch": 0.0428439330114045,
+      "grad_norm": 0.8225170969963074,
+      "learning_rate": 0.001713229252259655,
+      "loss": 4.8439,
+      "step": 417
+    },
+    {
+      "epoch": 0.0431521627452995,
+      "grad_norm": 0.5638198256492615,
+      "learning_rate": 0.001725554642563681,
+      "loss": 4.8271,
+      "step": 420
+    },
+    {
+      "epoch": 0.04346039247919449,
+      "grad_norm": 0.3389821946620941,
+      "learning_rate": 0.0017378800328677075,
+      "loss": 4.8207,
+      "step": 423
+    },
+    {
+      "epoch": 0.04376862221308949,
+      "grad_norm": 0.38620057702064514,
+      "learning_rate": 0.0017502054231717337,
+      "loss": 4.8082,
+      "step": 426
+    },
+    {
+      "epoch": 0.044076851946984484,
+      "grad_norm": 1.1568442583084106,
+      "learning_rate": 0.00176253081347576,
+      "loss": 4.7946,
+      "step": 429
+    },
+    {
+      "epoch": 0.044385081680879485,
+      "grad_norm": 0.650175154209137,
+      "learning_rate": 0.0017748562037797865,
+      "loss": 4.7798,
+      "step": 432
+    },
+    {
+      "epoch": 0.04469331141477448,
+      "grad_norm": 0.5364396572113037,
+      "learning_rate": 0.0017871815940838126,
+      "loss": 4.7732,
+      "step": 435
+    },
+    {
+      "epoch": 0.04500154114866947,
+      "grad_norm": 0.7013806700706482,
+      "learning_rate": 0.001799506984387839,
+      "loss": 4.7733,
+      "step": 438
+    },
+    {
+      "epoch": 0.04530977088256447,
+      "grad_norm": 0.4559784233570099,
+      "learning_rate": 0.0018118323746918652,
+      "loss": 4.7789,
+      "step": 441
+    },
+    {
+      "epoch": 0.045618000616459466,
+      "grad_norm": 0.3456243872642517,
+      "learning_rate": 0.0018241577649958916,
+      "loss": 4.7456,
+      "step": 444
+    },
+    {
+      "epoch": 0.045926230350354466,
+      "grad_norm": 0.6245532631874084,
+      "learning_rate": 0.0018364831552999178,
+      "loss": 4.7408,
+      "step": 447
+    },
+    {
+      "epoch": 0.04623446008424946,
+      "grad_norm": 1.1933598518371582,
+      "learning_rate": 0.0018488085456039441,
+      "loss": 4.7728,
+      "step": 450
+    },
+    {
+      "epoch": 0.04654268981814446,
+      "grad_norm": 0.8743248581886292,
+      "learning_rate": 0.0018611339359079703,
+      "loss": 4.7595,
+      "step": 453
+    },
+    {
+      "epoch": 0.046850919552039454,
+      "grad_norm": 0.4980567693710327,
+      "learning_rate": 0.0018734593262119967,
+      "loss": 4.7222,
+      "step": 456
+    },
+    {
+      "epoch": 0.04715914928593445,
+      "grad_norm": 0.6380690932273865,
+      "learning_rate": 0.001885784716516023,
+      "loss": 4.7175,
+      "step": 459
+    },
+    {
+      "epoch": 0.04746737901982945,
+      "grad_norm": 0.3606894612312317,
+      "learning_rate": 0.0018981101068200493,
+      "loss": 4.7075,
+      "step": 462
+    },
+    {
+      "epoch": 0.04777560875372444,
+      "grad_norm": 0.5618919730186462,
+      "learning_rate": 0.0019104354971240757,
+      "loss": 4.6939,
+      "step": 465
+    },
+    {
+      "epoch": 0.04808383848761944,
+      "grad_norm": 0.639410138130188,
+      "learning_rate": 0.0019227608874281018,
+      "loss": 4.6748,
+      "step": 468
+    },
+    {
+      "epoch": 0.048392068221514435,
+      "grad_norm": 0.7849680185317993,
+      "learning_rate": 0.0019350862777321282,
+      "loss": 4.6895,
+      "step": 471
+    },
+    {
+      "epoch": 0.04870029795540943,
+      "grad_norm": 0.5419800877571106,
+      "learning_rate": 0.0019474116680361544,
+      "loss": 4.64,
+      "step": 474
+    },
+    {
+      "epoch": 0.04900852768930443,
+      "grad_norm": 0.40359726548194885,
+      "learning_rate": 0.001959737058340181,
+      "loss": 4.6564,
+      "step": 477
+    },
+    {
+      "epoch": 0.04931675742319942,
+      "grad_norm": 0.742076575756073,
+      "learning_rate": 0.001972062448644207,
+      "loss": 4.6434,
+      "step": 480
+    },
+    {
+      "epoch": 0.04962498715709442,
+      "grad_norm": 0.620801568031311,
+      "learning_rate": 0.0019843878389482336,
+      "loss": 4.6509,
+      "step": 483
+    },
+    {
+      "epoch": 0.04993321689098942,
+      "grad_norm": 0.5293563008308411,
+      "learning_rate": 0.0019967132292522598,
+      "loss": 4.6459,
+      "step": 486
+    },
+    {
+      "epoch": 0.05024144662488442,
+      "grad_norm": 0.7527710795402527,
+      "learning_rate": 0.002009038619556286,
+      "loss": 4.6557,
+      "step": 489
+    },
+    {
+      "epoch": 0.05054967635877941,
+      "grad_norm": 0.47365424036979675,
+      "learning_rate": 0.002021364009860312,
+      "loss": 4.6223,
+      "step": 492
+    },
+    {
+      "epoch": 0.050857906092674404,
+      "grad_norm": 0.5232967734336853,
+      "learning_rate": 0.0020336894001643387,
+      "loss": 4.6186,
+      "step": 495
+    },
+    {
+      "epoch": 0.051166135826569405,
+      "grad_norm": 0.40717506408691406,
+      "learning_rate": 0.002046014790468365,
+      "loss": 4.6125,
+      "step": 498
+    },
+    {
+      "epoch": 0.0514743655604644,
+      "grad_norm": 0.5403701066970825,
+      "learning_rate": 0.002058340180772391,
+      "loss": 4.6143,
+      "step": 501
+    },
+    {
+      "epoch": 0.0517825952943594,
+      "grad_norm": 0.7209203839302063,
+      "learning_rate": 0.0020706655710764172,
+      "loss": 4.5713,
+      "step": 504
+    },
+    {
+      "epoch": 0.05209082502825439,
+      "grad_norm": 0.6991008520126343,
+      "learning_rate": 0.002082990961380444,
+      "loss": 4.6044,
+      "step": 507
+    },
+    {
+      "epoch": 0.052399054762149386,
+      "grad_norm": 0.7478086352348328,
+      "learning_rate": 0.00209531635168447,
+      "loss": 4.5685,
+      "step": 510
+    },
+    {
+      "epoch": 0.052707284496044386,
+      "grad_norm": 0.5864932537078857,
+      "learning_rate": 0.002107641741988496,
+      "loss": 4.588,
+      "step": 513
+    },
+    {
+      "epoch": 0.05301551422993938,
+      "grad_norm": 0.44748950004577637,
+      "learning_rate": 0.002119967132292523,
+      "loss": 4.5823,
+      "step": 516
+    },
+    {
+      "epoch": 0.05332374396383438,
+      "grad_norm": 0.32787564396858215,
+      "learning_rate": 0.002132292522596549,
+      "loss": 4.5522,
+      "step": 519
+    },
+    {
+      "epoch": 0.053631973697729374,
+      "grad_norm": 0.30747687816619873,
+      "learning_rate": 0.002144617912900575,
+      "loss": 4.5429,
+      "step": 522
+    },
+    {
+      "epoch": 0.05394020343162437,
+      "grad_norm": 0.3548784554004669,
+      "learning_rate": 0.0021569433032046013,
+      "loss": 4.5207,
+      "step": 525
+    },
+    {
+      "epoch": 0.05424843316551937,
+      "grad_norm": 0.6617491841316223,
+      "learning_rate": 0.002169268693508628,
+      "loss": 4.5373,
+      "step": 528
+    },
+    {
+      "epoch": 0.05455666289941436,
+      "grad_norm": 0.9917429089546204,
+      "learning_rate": 0.002181594083812654,
+      "loss": 4.5504,
+      "step": 531
+    },
+    {
+      "epoch": 0.05486489263330936,
+      "grad_norm": 0.6506537795066833,
+      "learning_rate": 0.0021939194741166803,
+      "loss": 4.5385,
+      "step": 534
+    },
+    {
+      "epoch": 0.055173122367204355,
+      "grad_norm": 0.3738003075122833,
+      "learning_rate": 0.002206244864420707,
+      "loss": 4.5169,
+      "step": 537
+    },
+    {
+      "epoch": 0.055481352101099356,
+      "grad_norm": 0.3488200008869171,
+      "learning_rate": 0.002218570254724733,
+      "loss": 4.5119,
+      "step": 540
+    },
+    {
+      "epoch": 0.05578958183499435,
+      "grad_norm": 0.31217944622039795,
+      "learning_rate": 0.0022308956450287592,
+      "loss": 4.4796,
+      "step": 543
+    },
+    {
+      "epoch": 0.05609781156888934,
+      "grad_norm": 0.26770153641700745,
+      "learning_rate": 0.0022432210353327854,
+      "loss": 4.4699,
+      "step": 546
+    },
+    {
+      "epoch": 0.05640604130278434,
+      "grad_norm": 0.3656662702560425,
+      "learning_rate": 0.002255546425636812,
+      "loss": 4.4817,
+      "step": 549
+    },
+    {
+      "epoch": 0.05671427103667934,
+      "grad_norm": 0.5845988392829895,
+      "learning_rate": 0.002267871815940838,
+      "loss": 4.4525,
+      "step": 552
+    },
+    {
+      "epoch": 0.05702250077057434,
+      "grad_norm": 0.41006627678871155,
+      "learning_rate": 0.0022801972062448644,
+      "loss": 4.4649,
+      "step": 555
+    },
+    {
+      "epoch": 0.05733073050446933,
+      "grad_norm": 1.2013694047927856,
+      "learning_rate": 0.002292522596548891,
+      "loss": 4.4666,
+      "step": 558
+    },
+    {
+      "epoch": 0.057638960238364324,
+      "grad_norm": 0.6116489171981812,
+      "learning_rate": 0.002304847986852917,
+      "loss": 4.4853,
+      "step": 561
+    },
+    {
+      "epoch": 0.057947189972259325,
+      "grad_norm": 0.30115845799446106,
+      "learning_rate": 0.0023171733771569433,
+      "loss": 4.4409,
+      "step": 564
+    },
+    {
+      "epoch": 0.05825541970615432,
+      "grad_norm": 0.2863396108150482,
+      "learning_rate": 0.0023294987674609695,
+      "loss": 4.4358,
+      "step": 567
+    },
+    {
+      "epoch": 0.05856364944004932,
+      "grad_norm": 0.3191300928592682,
+      "learning_rate": 0.002341824157764996,
+      "loss": 4.454,
+      "step": 570
+    },
+    {
+      "epoch": 0.05887187917394431,
+      "grad_norm": 0.4280944764614105,
+      "learning_rate": 0.0023541495480690223,
+      "loss": 4.3943,
+      "step": 573
+    },
+    {
+      "epoch": 0.05918010890783931,
+      "grad_norm": 0.49310484528541565,
+      "learning_rate": 0.0023664749383730485,
+      "loss": 4.4097,
+      "step": 576
+    },
+    {
+      "epoch": 0.059488338641734306,
+      "grad_norm": 0.4923991858959198,
+      "learning_rate": 0.002378800328677075,
+      "loss": 4.4454,
+      "step": 579
+    },
+    {
+      "epoch": 0.0597965683756293,
+      "grad_norm": 0.5043625235557556,
+      "learning_rate": 0.0023911257189811012,
+      "loss": 4.3975,
+      "step": 582
+    },
+    {
+      "epoch": 0.0601047981095243,
+      "grad_norm": 0.5404270887374878,
+      "learning_rate": 0.0024034511092851274,
+      "loss": 4.3957,
+      "step": 585
+    },
+    {
+      "epoch": 0.060413027843419294,
+      "grad_norm": 0.9954332709312439,
+      "learning_rate": 0.0024157764995891536,
+      "loss": 4.3864,
+      "step": 588
+    },
+    {
+      "epoch": 0.060721257577314294,
+      "grad_norm": 0.3632584512233734,
+      "learning_rate": 0.00242810188989318,
+      "loss": 4.38,
+      "step": 591
+    },
+    {
+      "epoch": 0.06102948731120929,
+      "grad_norm": 0.2620343267917633,
+      "learning_rate": 0.0024404272801972064,
+      "loss": 4.3538,
+      "step": 594
+    },
+    {
+      "epoch": 0.06133771704510428,
+      "grad_norm": 0.25050923228263855,
+      "learning_rate": 0.0024527526705012325,
+      "loss": 4.351,
+      "step": 597
+    },
+    {
+      "epoch": 0.06164594677899928,
+      "grad_norm": 0.27279627323150635,
+      "learning_rate": 0.0024650780608052587,
+      "loss": 4.3335,
+      "step": 600
+    },
+    {
+      "epoch": 0.061954176512894275,
+      "grad_norm": 0.6038771271705627,
+      "learning_rate": 0.0024774034511092853,
+      "loss": 4.3409,
+      "step": 603
+    },
+    {
+      "epoch": 0.062262406246789276,
+      "grad_norm": 0.6948337554931641,
+      "learning_rate": 0.0024897288414133115,
+      "loss": 4.3555,
+      "step": 606
+    },
+    {
+      "epoch": 0.06257063598068427,
+      "grad_norm": 0.5086238980293274,
+      "learning_rate": 0.0025020542317173377,
+      "loss": 4.3491,
+      "step": 609
+    },
+    {
+      "epoch": 0.06287886571457926,
+      "grad_norm": 0.475999116897583,
+      "learning_rate": 0.002514379622021364,
+      "loss": 4.3412,
+      "step": 612
+    },
+    {
+      "epoch": 0.06318709544847426,
+      "grad_norm": 0.3968357741832733,
+      "learning_rate": 0.0025267050123253905,
+      "loss": 4.3139,
+      "step": 615
+    },
+    {
+      "epoch": 0.06349532518236926,
+      "grad_norm": 0.6681760549545288,
+      "learning_rate": 0.0025390304026294166,
+      "loss": 4.2999,
+      "step": 618
+    },
+    {
+      "epoch": 0.06380355491626426,
+      "grad_norm": 0.3453294634819031,
+      "learning_rate": 0.002551355792933443,
+      "loss": 4.2873,
+      "step": 621
+    },
+    {
+      "epoch": 0.06411178465015925,
+      "grad_norm": 0.3346744775772095,
+      "learning_rate": 0.002563681183237469,
+      "loss": 4.2868,
+      "step": 624
+    },
+    {
+      "epoch": 0.06442001438405424,
+      "grad_norm": 0.39689645171165466,
+      "learning_rate": 0.0025760065735414956,
+      "loss": 4.2846,
+      "step": 627
+    },
+    {
+      "epoch": 0.06472824411794924,
+      "grad_norm": 0.4017212688922882,
+      "learning_rate": 0.0025883319638455218,
+      "loss": 4.2625,
+      "step": 630
+    },
+    {
+      "epoch": 0.06503647385184425,
+      "grad_norm": 0.3414025902748108,
+      "learning_rate": 0.0026006573541495484,
+      "loss": 4.2657,
+      "step": 633
+    },
+    {
+      "epoch": 0.06534470358573924,
+      "grad_norm": 0.4091610312461853,
+      "learning_rate": 0.002612982744453574,
+      "loss": 4.2414,
+      "step": 636
+    },
+    {
+      "epoch": 0.06565293331963423,
+      "grad_norm": 0.3916926085948944,
+      "learning_rate": 0.0026253081347576007,
+      "loss": 4.1801,
+      "step": 639
+    },
+    {
+      "epoch": 0.06596116305352923,
+      "grad_norm": 1.0324465036392212,
+      "learning_rate": 0.002637633525061627,
+      "loss": 4.2162,
+      "step": 642
+    },
+    {
+      "epoch": 0.06626939278742423,
+      "grad_norm": 0.4595172107219696,
+      "learning_rate": 0.0026499589153656535,
+      "loss": 4.2352,
+      "step": 645
+    },
+    {
+      "epoch": 0.06657762252131923,
+      "grad_norm": 0.3215947151184082,
+      "learning_rate": 0.0026622843056696792,
+      "loss": 4.1865,
+      "step": 648
+    },
+    {
+      "epoch": 0.06688585225521422,
+      "grad_norm": 0.2739149034023285,
+      "learning_rate": 0.002674609695973706,
+      "loss": 4.1644,
+      "step": 651
+    },
+    {
+      "epoch": 0.06719408198910921,
+      "grad_norm": 0.250794917345047,
+      "learning_rate": 0.002686935086277732,
+      "loss": 4.1164,
+      "step": 654
+    },
+    {
+      "epoch": 0.06750231172300421,
+      "grad_norm": 0.38465654850006104,
+      "learning_rate": 0.0026992604765817586,
+      "loss": 4.0844,
+      "step": 657
+    },
+    {
+      "epoch": 0.06781054145689921,
+      "grad_norm": 0.5341691970825195,
+      "learning_rate": 0.0027115858668857844,
+      "loss": 4.1149,
+      "step": 660
+    },
+    {
+      "epoch": 0.06811877119079421,
+      "grad_norm": 0.3479110896587372,
+      "learning_rate": 0.002723911257189811,
+      "loss": 4.1186,
+      "step": 663
+    },
+    {
+      "epoch": 0.0684270009246892,
+      "grad_norm": 1.026038646697998,
+      "learning_rate": 0.002736236647493837,
+      "loss": 4.1293,
+      "step": 666
+    },
+    {
+      "epoch": 0.0687352306585842,
+      "grad_norm": 0.445689857006073,
+      "learning_rate": 0.0027485620377978638,
+      "loss": 4.1319,
+      "step": 669
+    },
+    {
+      "epoch": 0.06904346039247919,
+      "grad_norm": 0.3061058819293976,
+      "learning_rate": 0.00276088742810189,
+      "loss": 4.0338,
+      "step": 672
+    },
+    {
+      "epoch": 0.0693516901263742,
+      "grad_norm": 0.26792746782302856,
+      "learning_rate": 0.002773212818405916,
+      "loss": 4.0154,
+      "step": 675
+    },
+    {
+      "epoch": 0.06965991986026919,
+      "grad_norm": 0.2843894064426422,
+      "learning_rate": 0.0027855382087099423,
+      "loss": 4.0364,
+      "step": 678
+    },
+    {
+      "epoch": 0.06996814959416418,
+      "grad_norm": 0.3073459565639496,
+      "learning_rate": 0.002797863599013969,
+      "loss": 3.9836,
+      "step": 681
+    },
+    {
+      "epoch": 0.07027637932805918,
+      "grad_norm": 0.5893545746803284,
+      "learning_rate": 0.002810188989317995,
+      "loss": 4.0062,
+      "step": 684
+    },
+    {
+      "epoch": 0.07058460906195417,
+      "grad_norm": 0.5386547446250916,
+      "learning_rate": 0.0028225143796220217,
+      "loss": 4.0066,
+      "step": 687
+    },
+    {
+      "epoch": 0.07089283879584918,
+      "grad_norm": 0.7944250106811523,
+      "learning_rate": 0.0028348397699260474,
+      "loss": 3.9816,
+      "step": 690
+    },
+    {
+      "epoch": 0.07120106852974417,
+      "grad_norm": 0.32200196385383606,
+      "learning_rate": 0.002847165160230074,
+      "loss": 3.9551,
+      "step": 693
+    },
+    {
+      "epoch": 0.07150929826363916,
+      "grad_norm": 0.28814995288848877,
+      "learning_rate": 0.0028594905505341,
+      "loss": 3.912,
+      "step": 696
+    },
+    {
+      "epoch": 0.07181752799753416,
+      "grad_norm": 0.2727998197078705,
+      "learning_rate": 0.002871815940838127,
+      "loss": 3.9203,
+      "step": 699
+    },
+    {
+      "epoch": 0.07212575773142915,
+      "grad_norm": 0.2785607576370239,
+      "learning_rate": 0.0028841413311421525,
+      "loss": 3.8865,
+      "step": 702
+    },
+    {
+      "epoch": 0.07243398746532416,
+      "grad_norm": 0.4318368136882782,
+      "learning_rate": 0.002896466721446179,
+      "loss": 3.8364,
+      "step": 705
+    },
+    {
+      "epoch": 0.07274221719921915,
+      "grad_norm": 0.5888954997062683,
+      "learning_rate": 0.0029087921117502053,
+      "loss": 3.8772,
+      "step": 708
+    },
+    {
+      "epoch": 0.07305044693311415,
+      "grad_norm": 0.5866847634315491,
+      "learning_rate": 0.002921117502054232,
+      "loss": 3.9039,
+      "step": 711
+    },
+    {
+      "epoch": 0.07335867666700914,
+      "grad_norm": 0.40300968289375305,
+      "learning_rate": 0.002933442892358258,
+      "loss": 3.8332,
+      "step": 714
+    },
+    {
+      "epoch": 0.07366690640090415,
+      "grad_norm": 0.2894107401371002,
+      "learning_rate": 0.0029457682826622843,
+      "loss": 3.8533,
+      "step": 717
+    },
+    {
+      "epoch": 0.07397513613479914,
+      "grad_norm": 0.2637479901313782,
+      "learning_rate": 0.0029580936729663105,
+      "loss": 3.7962,
+      "step": 720
+    },
+    {
+      "epoch": 0.07428336586869413,
+      "grad_norm": 0.5004228353500366,
+      "learning_rate": 0.002970419063270337,
+      "loss": 3.7759,
+      "step": 723
+    },
+    {
+      "epoch": 0.07459159560258913,
+      "grad_norm": 0.30835986137390137,
+      "learning_rate": 0.0029827444535743632,
+      "loss": 3.7819,
+      "step": 726
+    },
+    {
+      "epoch": 0.07489982533648412,
+      "grad_norm": 0.5601911544799805,
+      "learning_rate": 0.00299506984387839,
+      "loss": 3.7716,
+      "step": 729
+    },
+    {
+      "epoch": 0.07520805507037913,
+      "grad_norm": 0.48242396116256714,
+      "learning_rate": 0.0030073952341824156,
+      "loss": 3.803,
+      "step": 732
+    },
+    {
+      "epoch": 0.07551628480427412,
+      "grad_norm": 0.355916827917099,
+      "learning_rate": 0.003019720624486442,
+      "loss": 3.7532,
+      "step": 735
+    },
+    {
+      "epoch": 0.07582451453816912,
+      "grad_norm": 0.4205069839954376,
+      "learning_rate": 0.0030320460147904684,
+      "loss": 3.7657,
+      "step": 738
+    },
+    {
+      "epoch": 0.07613274427206411,
+      "grad_norm": 0.35680562257766724,
+      "learning_rate": 0.003044371405094495,
+      "loss": 3.7348,
+      "step": 741
+    },
+    {
+      "epoch": 0.0764409740059591,
+      "grad_norm": 0.36372673511505127,
+      "learning_rate": 0.0030566967953985207,
+      "loss": 3.7569,
+      "step": 744
+    },
+    {
+      "epoch": 0.07674920373985411,
+      "grad_norm": 0.2887914776802063,
+      "learning_rate": 0.0030690221857025473,
+      "loss": 3.7704,
+      "step": 747
+    },
+    {
+      "epoch": 0.0770574334737491,
+      "grad_norm": 0.255290687084198,
+      "learning_rate": 0.0030813475760065735,
+      "loss": 3.7054,
+      "step": 750
+    },
+    {
+      "epoch": 0.0773656632076441,
+      "grad_norm": 0.2969897389411926,
+      "learning_rate": 0.0030936729663106,
+      "loss": 3.708,
+      "step": 753
+    },
+    {
+      "epoch": 0.07767389294153909,
+      "grad_norm": 0.491763710975647,
+      "learning_rate": 0.003105998356614626,
+      "loss": 3.7279,
+      "step": 756
+    },
+    {
+      "epoch": 0.07798212267543408,
+      "grad_norm": 0.6437285542488098,
+      "learning_rate": 0.0031183237469186525,
+      "loss": 3.7129,
+      "step": 759
+    },
+    {
+      "epoch": 0.07829035240932909,
+      "grad_norm": 0.3605806827545166,
+      "learning_rate": 0.0031306491372226786,
+      "loss": 3.6754,
+      "step": 762
+    },
+    {
+      "epoch": 0.07859858214322409,
+      "grad_norm": 0.26162126660346985,
+      "learning_rate": 0.0031429745275267052,
+      "loss": 3.6869,
+      "step": 765
+    },
+    {
+      "epoch": 0.07890681187711908,
+      "grad_norm": 0.3107220530509949,
+      "learning_rate": 0.0031552999178307314,
+      "loss": 3.6278,
+      "step": 768
+    },
+    {
+      "epoch": 0.07921504161101407,
+      "grad_norm": 0.30417200922966003,
+      "learning_rate": 0.0031676253081347576,
+      "loss": 3.6046,
+      "step": 771
+    },
+    {
+      "epoch": 0.07952327134490907,
+      "grad_norm": 0.5612326860427856,
+      "learning_rate": 0.0031799506984387838,
+      "loss": 3.6578,
+      "step": 774
+    },
+    {
+      "epoch": 0.07983150107880407,
+      "grad_norm": 0.6136355996131897,
+      "learning_rate": 0.0031922760887428104,
+      "loss": 3.6897,
+      "step": 777
+    },
+    {
+      "epoch": 0.08013973081269907,
+      "grad_norm": 0.4560060501098633,
+      "learning_rate": 0.0032046014790468366,
+      "loss": 3.6976,
+      "step": 780
+    },
+    {
+      "epoch": 0.08044796054659406,
+      "grad_norm": 0.23871034383773804,
+      "learning_rate": 0.003216926869350863,
+      "loss": 3.657,
+      "step": 783
+    },
+    {
+      "epoch": 0.08075619028048905,
+      "grad_norm": 0.17063000798225403,
+      "learning_rate": 0.003229252259654889,
+      "loss": 3.5905,
+      "step": 786
+    },
+    {
+      "epoch": 0.08106442001438405,
+      "grad_norm": 0.35351842641830444,
+      "learning_rate": 0.0032415776499589155,
+      "loss": 3.603,
+      "step": 789
+    },
+    {
+      "epoch": 0.08137264974827906,
+      "grad_norm": 0.340762197971344,
+      "learning_rate": 0.0032539030402629417,
+      "loss": 3.5978,
+      "step": 792
+    },
+    {
+      "epoch": 0.08168087948217405,
+      "grad_norm": 0.22542034089565277,
+      "learning_rate": 0.0032662284305669683,
+      "loss": 3.5821,
+      "step": 795
+    },
+    {
+      "epoch": 0.08198910921606904,
+      "grad_norm": 0.25130555033683777,
+      "learning_rate": 0.003278553820870994,
+      "loss": 3.5491,
+      "step": 798
+    },
+    {
+      "epoch": 0.08229733894996404,
+      "grad_norm": 0.5155714750289917,
+      "learning_rate": 0.0032908792111750206,
+      "loss": 3.5605,
+      "step": 801
+    },
+    {
+      "epoch": 0.08260556868385904,
+      "grad_norm": 0.3964254856109619,
+      "learning_rate": 0.003303204601479047,
+      "loss": 3.581,
+      "step": 804
+    },
+    {
+      "epoch": 0.08291379841775404,
+      "grad_norm": 0.27110666036605835,
+      "learning_rate": 0.0033155299917830734,
+      "loss": 3.5995,
+      "step": 807
+    },
+    {
+      "epoch": 0.08322202815164903,
+      "grad_norm": 0.38535767793655396,
+      "learning_rate": 0.003327855382087099,
+      "loss": 3.6029,
+      "step": 810
+    },
+    {
+      "epoch": 0.08353025788554402,
+      "grad_norm": 0.6176712512969971,
+      "learning_rate": 0.0033401807723911258,
+      "loss": 3.5694,
+      "step": 813
+    },
+    {
+      "epoch": 0.08383848761943902,
+      "grad_norm": 0.33828550577163696,
+      "learning_rate": 0.003352506162695152,
+      "loss": 3.5507,
+      "step": 816
+    },
+    {
+      "epoch": 0.08414671735333402,
+      "grad_norm": 0.2286808043718338,
+      "learning_rate": 0.0033648315529991786,
+      "loss": 3.5345,
+      "step": 819
+    },
+    {
+      "epoch": 0.08445494708722902,
+      "grad_norm": 0.30232542753219604,
+      "learning_rate": 0.0033771569433032047,
+      "loss": 3.5154,
+      "step": 822
+    },
+    {
+      "epoch": 0.08476317682112401,
+      "grad_norm": 0.31767842173576355,
+      "learning_rate": 0.0033894823336072313,
+      "loss": 3.5442,
+      "step": 825
+    },
+    {
+      "epoch": 0.085071406555019,
+      "grad_norm": 0.4275444746017456,
+      "learning_rate": 0.003401807723911257,
+      "loss": 3.5715,
+      "step": 828
+    },
+    {
+      "epoch": 0.085379636288914,
+      "grad_norm": 0.3426364064216614,
+      "learning_rate": 0.0034141331142152837,
+      "loss": 3.5224,
+      "step": 831
+    },
+    {
+      "epoch": 0.085687866022809,
+      "grad_norm": 0.33871403336524963,
+      "learning_rate": 0.00342645850451931,
+      "loss": 3.5119,
+      "step": 834
+    },
+    {
+      "epoch": 0.085996095756704,
+      "grad_norm": 0.2641143202781677,
+      "learning_rate": 0.0034387838948233365,
+      "loss": 3.5179,
+      "step": 837
+    },
+    {
+      "epoch": 0.086304325490599,
+      "grad_norm": 0.22955679893493652,
+      "learning_rate": 0.003451109285127362,
+      "loss": 3.4807,
+      "step": 840
+    },
+    {
+      "epoch": 0.08661255522449399,
+      "grad_norm": 0.3795819878578186,
+      "learning_rate": 0.003463434675431389,
+      "loss": 3.4916,
+      "step": 843
+    },
+    {
+      "epoch": 0.08692078495838898,
+      "grad_norm": 0.2942325174808502,
+      "learning_rate": 0.003475760065735415,
+      "loss": 3.4601,
+      "step": 846
+    },
+    {
+      "epoch": 0.08722901469228399,
+      "grad_norm": 0.49732574820518494,
+      "learning_rate": 0.0034880854560394416,
+      "loss": 3.5021,
+      "step": 849
+    },
+    {
+      "epoch": 0.08753724442617898,
+      "grad_norm": 0.4395911991596222,
+      "learning_rate": 0.0035004108463434673,
+      "loss": 3.4976,
+      "step": 852
+    },
+    {
+      "epoch": 0.08784547416007397,
+      "grad_norm": 0.24201816320419312,
+      "learning_rate": 0.003512736236647494,
+      "loss": 3.4648,
+      "step": 855
+    },
+    {
+      "epoch": 0.08815370389396897,
+      "grad_norm": 0.32818078994750977,
+      "learning_rate": 0.00352506162695152,
+      "loss": 3.4826,
+      "step": 858
+    },
+    {
+      "epoch": 0.08846193362786396,
+      "grad_norm": 0.4433400630950928,
+      "learning_rate": 0.0035373870172555467,
+      "loss": 3.4763,
+      "step": 861
+    },
+    {
+      "epoch": 0.08877016336175897,
+      "grad_norm": 0.2911035716533661,
+      "learning_rate": 0.003549712407559573,
+      "loss": 3.5024,
+      "step": 864
+    },
+    {
+      "epoch": 0.08907839309565396,
+      "grad_norm": 0.27419009804725647,
+      "learning_rate": 0.003562037797863599,
+      "loss": 3.4382,
+      "step": 867
+    },
+    {
+      "epoch": 0.08938662282954896,
+      "grad_norm": 0.2970244586467743,
+      "learning_rate": 0.0035743631881676253,
+      "loss": 3.4362,
+      "step": 870
+    },
+    {
+      "epoch": 0.08969485256344395,
+      "grad_norm": 0.34221401810646057,
+      "learning_rate": 0.003586688578471652,
+      "loss": 3.4469,
+      "step": 873
+    },
+    {
+      "epoch": 0.09000308229733894,
+      "grad_norm": 0.31807199120521545,
+      "learning_rate": 0.003599013968775678,
+      "loss": 3.3974,
+      "step": 876
+    },
+    {
+      "epoch": 0.09031131203123395,
+      "grad_norm": 0.31519362330436707,
+      "learning_rate": 0.0036113393590797046,
+      "loss": 3.4275,
+      "step": 879
+    },
+    {
+      "epoch": 0.09061954176512894,
+      "grad_norm": 0.5152423977851868,
+      "learning_rate": 0.0036236647493837304,
+      "loss": 3.4468,
+      "step": 882
+    },
+    {
+      "epoch": 0.09092777149902394,
+      "grad_norm": 0.32447418570518494,
+      "learning_rate": 0.003635990139687757,
+      "loss": 3.4505,
+      "step": 885
+    },
+    {
+      "epoch": 0.09123600123291893,
+      "grad_norm": 0.19884614646434784,
+      "learning_rate": 0.003648315529991783,
+      "loss": 3.4228,
+      "step": 888
+    },
+    {
+      "epoch": 0.09154423096681394,
+      "grad_norm": 0.2726935148239136,
+      "learning_rate": 0.0036606409202958098,
+      "loss": 3.3957,
+      "step": 891
+    },
+    {
+      "epoch": 0.09185246070070893,
+      "grad_norm": 0.29470425844192505,
+      "learning_rate": 0.0036729663105998355,
+      "loss": 3.3813,
+      "step": 894
+    },
+    {
+      "epoch": 0.09216069043460393,
+      "grad_norm": 0.27806392312049866,
+      "learning_rate": 0.003685291700903862,
+      "loss": 3.3871,
+      "step": 897
+    },
+    {
+      "epoch": 0.09246892016849892,
+      "grad_norm": 0.23773950338363647,
+      "learning_rate": 0.0036976170912078883,
+      "loss": 3.3941,
+      "step": 900
+    },
+    {
+      "epoch": 0.09277714990239391,
+      "grad_norm": 0.45804303884506226,
+      "learning_rate": 0.003709942481511915,
+      "loss": 3.3752,
+      "step": 903
+    },
+    {
+      "epoch": 0.09308537963628892,
+      "grad_norm": 0.45320865511894226,
+      "learning_rate": 0.0037222678718159406,
+      "loss": 3.4068,
+      "step": 906
+    },
+    {
+      "epoch": 0.09339360937018391,
+      "grad_norm": 0.277089387178421,
+      "learning_rate": 0.0037345932621199673,
+      "loss": 3.4052,
+      "step": 909
+    },
+    {
+      "epoch": 0.09370183910407891,
+      "grad_norm": 0.26548513770103455,
+      "learning_rate": 0.0037469186524239934,
+      "loss": 3.3753,
+      "step": 912
+    },
+    {
+      "epoch": 0.0940100688379739,
+      "grad_norm": 0.24219335615634918,
+      "learning_rate": 0.00375924404272802,
+      "loss": 3.3913,
+      "step": 915
+    },
+    {
+      "epoch": 0.0943182985718689,
+      "grad_norm": 0.2855617105960846,
+      "learning_rate": 0.003771569433032046,
+      "loss": 3.3636,
+      "step": 918
+    },
+    {
+      "epoch": 0.0946265283057639,
+      "grad_norm": 0.35244864225387573,
+      "learning_rate": 0.003783894823336073,
+      "loss": 3.3603,
+      "step": 921
+    },
+    {
+      "epoch": 0.0949347580396589,
+      "grad_norm": 0.3226896822452545,
+      "learning_rate": 0.0037962202136400986,
+      "loss": 3.3267,
+      "step": 924
+    },
+    {
+      "epoch": 0.09524298777355389,
+      "grad_norm": 0.279863178730011,
+      "learning_rate": 0.003808545603944125,
+      "loss": 3.3192,
+      "step": 927
+    },
+    {
+      "epoch": 0.09555121750744888,
+      "grad_norm": 0.35309404134750366,
+      "learning_rate": 0.0038208709942481513,
+      "loss": 3.2978,
+      "step": 930
+    },
+    {
+      "epoch": 0.09585944724134388,
+      "grad_norm": 0.2359645515680313,
+      "learning_rate": 0.003833196384552178,
+      "loss": 3.3627,
+      "step": 933
+    },
+    {
+      "epoch": 0.09616767697523888,
+      "grad_norm": 0.22583429515361786,
+      "learning_rate": 0.0038455217748562037,
+      "loss": 3.2669,
+      "step": 936
+    },
+    {
+      "epoch": 0.09647590670913388,
+      "grad_norm": 0.2914174199104309,
+      "learning_rate": 0.0038578471651602303,
+      "loss": 3.3238,
+      "step": 939
+    },
+    {
+      "epoch": 0.09678413644302887,
+      "grad_norm": 0.37748411297798157,
+      "learning_rate": 0.0038701725554642565,
+      "loss": 3.3232,
+      "step": 942
+    },
+    {
+      "epoch": 0.09709236617692386,
+      "grad_norm": 0.28686878085136414,
+      "learning_rate": 0.003882497945768283,
+      "loss": 3.3143,
+      "step": 945
+    },
+    {
+      "epoch": 0.09740059591081886,
+      "grad_norm": 0.22591544687747955,
+      "learning_rate": 0.003894823336072309,
+      "loss": 3.3285,
+      "step": 948
+    },
+    {
+      "epoch": 0.09770882564471386,
+      "grad_norm": 0.24365665018558502,
+      "learning_rate": 0.003907148726376336,
+      "loss": 3.2799,
+      "step": 951
+    },
+    {
+      "epoch": 0.09801705537860886,
+      "grad_norm": 0.3929263651371002,
+      "learning_rate": 0.003919474116680362,
+      "loss": 3.29,
+      "step": 954
+    },
+    {
+      "epoch": 0.09832528511250385,
+      "grad_norm": 0.20268237590789795,
+      "learning_rate": 0.003931799506984388,
+      "loss": 3.2412,
+      "step": 957
+    },
+    {
+      "epoch": 0.09863351484639885,
+      "grad_norm": 0.3333010673522949,
+      "learning_rate": 0.003944124897288414,
+      "loss": 3.2523,
+      "step": 960
+    },
+    {
+      "epoch": 0.09894174458029384,
+      "grad_norm": 0.32760193943977356,
+      "learning_rate": 0.0039564502875924406,
+      "loss": 3.2958,
+      "step": 963
+    },
+    {
+      "epoch": 0.09924997431418885,
+      "grad_norm": 0.27670565247535706,
+      "learning_rate": 0.003968775677896467,
+      "loss": 3.2683,
+      "step": 966
+    },
+    {
+      "epoch": 0.09955820404808384,
+      "grad_norm": 0.32110410928726196,
+      "learning_rate": 0.003981101068200493,
+      "loss": 3.2576,
+      "step": 969
+    },
+    {
+      "epoch": 0.09986643378197883,
+      "grad_norm": 0.43541696667671204,
+      "learning_rate": 0.0039934264585045195,
+      "loss": 3.2924,
+      "step": 972
+    },
+    {
+      "epoch": 0.10017466351587383,
+      "grad_norm": 0.3483084738254547,
+      "learning_rate": 0.004005751848808546,
+      "loss": 3.2936,
+      "step": 975
+    },
+    {
+      "epoch": 0.10048289324976883,
+      "grad_norm": 0.29586124420166016,
+      "learning_rate": 0.004018077239112572,
+      "loss": 3.2511,
+      "step": 978
+    },
+    {
+      "epoch": 0.10079112298366383,
+      "grad_norm": 0.21434040367603302,
+      "learning_rate": 0.0040304026294165985,
+      "loss": 3.242,
+      "step": 981
+    },
+    {
+      "epoch": 0.10109935271755882,
+      "grad_norm": 0.35204213857650757,
+      "learning_rate": 0.004042728019720624,
+      "loss": 3.2156,
+      "step": 984
+    },
+    {
+      "epoch": 0.10140758245145381,
+      "grad_norm": 0.25223758816719055,
+      "learning_rate": 0.004055053410024651,
+      "loss": 3.257,
+      "step": 987
+    },
+    {
+      "epoch": 0.10171581218534881,
+      "grad_norm": 0.2969653010368347,
+      "learning_rate": 0.004067378800328677,
+      "loss": 3.2576,
+      "step": 990
+    },
+    {
+      "epoch": 0.10202404191924382,
+      "grad_norm": 0.26683250069618225,
+      "learning_rate": 0.004079704190632704,
+      "loss": 3.1998,
+      "step": 993
+    },
+    {
+      "epoch": 0.10233227165313881,
+      "grad_norm": 0.26404044032096863,
+      "learning_rate": 0.00409202958093673,
+      "loss": 3.2303,
+      "step": 996
+    },
+    {
+      "epoch": 0.1026405013870338,
+      "grad_norm": 0.2442736029624939,
+      "learning_rate": 0.004104354971240756,
+      "loss": 3.2428,
+      "step": 999
+    },
+    {
+      "epoch": 0.1029487311209288,
+      "grad_norm": 0.2192964255809784,
+      "learning_rate": 0.004116680361544782,
+      "loss": 3.2661,
+      "step": 1002
+    },
+    {
+      "epoch": 0.10325696085482379,
+      "grad_norm": 0.21057608723640442,
+      "learning_rate": 0.004129005751848809,
+      "loss": 3.1995,
+      "step": 1005
+    },
+    {
+      "epoch": 0.1035651905887188,
+      "grad_norm": 0.3122745454311371,
+      "learning_rate": 0.0041413311421528345,
+      "loss": 3.2104,
+      "step": 1008
+    },
+    {
+      "epoch": 0.10387342032261379,
+      "grad_norm": 0.643337607383728,
+      "learning_rate": 0.004153656532456861,
+      "loss": 3.2196,
+      "step": 1011
+    },
+    {
+      "epoch": 0.10418165005650878,
+      "grad_norm": 0.265302449464798,
+      "learning_rate": 0.004165981922760888,
+      "loss": 3.2163,
+      "step": 1014
+    },
+    {
+      "epoch": 0.10448987979040378,
+      "grad_norm": 0.27250421047210693,
+      "learning_rate": 0.004178307313064914,
+      "loss": 3.1781,
+      "step": 1017
+    },
+    {
+      "epoch": 0.10479810952429877,
+      "grad_norm": 0.3951704800128937,
+      "learning_rate": 0.00419063270336894,
+      "loss": 3.2405,
+      "step": 1020
+    },
+    {
+      "epoch": 0.10510633925819378,
+      "grad_norm": 0.20837850868701935,
+      "learning_rate": 0.004202958093672967,
+      "loss": 3.2269,
+      "step": 1023
+    },
+    {
+      "epoch": 0.10541456899208877,
+      "grad_norm": 0.3887670338153839,
+      "learning_rate": 0.004215283483976992,
+      "loss": 3.219,
+      "step": 1026
+    },
+    {
+      "epoch": 0.10572279872598377,
+      "grad_norm": 0.18901754915714264,
+      "learning_rate": 0.004227608874281019,
+      "loss": 3.1759,
+      "step": 1029
+    },
+    {
+      "epoch": 0.10603102845987876,
+      "grad_norm": 0.3570176362991333,
+      "learning_rate": 0.004239934264585046,
+      "loss": 3.1544,
+      "step": 1032
+    },
+    {
+      "epoch": 0.10633925819377375,
+      "grad_norm": 0.2346538007259369,
+      "learning_rate": 0.004252259654889072,
+      "loss": 3.1834,
+      "step": 1035
+    },
+    {
+      "epoch": 0.10664748792766876,
+      "grad_norm": 0.1956055760383606,
+      "learning_rate": 0.004264585045193098,
+      "loss": 3.1597,
+      "step": 1038
+    },
+    {
+      "epoch": 0.10695571766156375,
+      "grad_norm": 0.19475719332695007,
+      "learning_rate": 0.0042769104354971246,
+      "loss": 3.1818,
+      "step": 1041
+    },
+    {
+      "epoch": 0.10726394739545875,
+      "grad_norm": 0.20991206169128418,
+      "learning_rate": 0.00428923582580115,
+      "loss": 3.148,
+      "step": 1044
+    },
+    {
+      "epoch": 0.10757217712935374,
+      "grad_norm": 0.45754027366638184,
+      "learning_rate": 0.004301561216105177,
+      "loss": 3.1838,
+      "step": 1047
+    },
+    {
+      "epoch": 0.10788040686324873,
+      "grad_norm": 0.2500004470348358,
+      "learning_rate": 0.004313886606409203,
+      "loss": 3.158,
+      "step": 1050
+    },
+    {
+      "epoch": 0.10818863659714374,
+      "grad_norm": 0.29174116253852844,
+      "learning_rate": 0.004326211996713229,
+      "loss": 3.1619,
+      "step": 1053
+    },
+    {
+      "epoch": 0.10849686633103874,
+      "grad_norm": 0.1642913520336151,
+      "learning_rate": 0.004338537387017256,
+      "loss": 3.1313,
+      "step": 1056
+    },
+    {
+      "epoch": 0.10880509606493373,
+      "grad_norm": 0.20638629794120789,
+      "learning_rate": 0.004350862777321282,
+      "loss": 3.1553,
+      "step": 1059
+    },
+    {
+      "epoch": 0.10911332579882872,
+      "grad_norm": 0.2534577548503876,
+      "learning_rate": 0.004363188167625308,
+      "loss": 3.146,
+      "step": 1062
+    },
+    {
+      "epoch": 0.10942155553272373,
+      "grad_norm": 0.3894107937812805,
+      "learning_rate": 0.004375513557929334,
+      "loss": 3.1702,
+      "step": 1065
+    },
+    {
+      "epoch": 0.10972978526661872,
+      "grad_norm": 0.18316411972045898,
+      "learning_rate": 0.0043878389482333606,
+      "loss": 3.1306,
+      "step": 1068
+    },
+    {
+      "epoch": 0.11003801500051372,
+      "grad_norm": 0.22901946306228638,
+      "learning_rate": 0.004400164338537387,
+      "loss": 3.1012,
+      "step": 1071
+    },
+    {
+      "epoch": 0.11034624473440871,
+      "grad_norm": 0.3013692796230316,
+      "learning_rate": 0.004412489728841414,
+      "loss": 3.1266,
+      "step": 1074
+    },
+    {
+      "epoch": 0.1106544744683037,
+      "grad_norm": 0.26568275690078735,
+      "learning_rate": 0.0044248151191454395,
+      "loss": 3.1161,
+      "step": 1077
+    },
+    {
+      "epoch": 0.11096270420219871,
+      "grad_norm": 0.23559318482875824,
+      "learning_rate": 0.004437140509449466,
+      "loss": 3.125,
+      "step": 1080
+    },
+    {
+      "epoch": 0.1112709339360937,
+      "grad_norm": 0.29804936051368713,
+      "learning_rate": 0.004449465899753492,
+      "loss": 3.1212,
+      "step": 1083
+    },
+    {
+      "epoch": 0.1115791636699887,
+      "grad_norm": 0.2965604066848755,
+      "learning_rate": 0.0044617912900575185,
+      "loss": 3.1435,
+      "step": 1086
+    },
+    {
+      "epoch": 0.11188739340388369,
+      "grad_norm": 0.22977206110954285,
+      "learning_rate": 0.004474116680361544,
+      "loss": 3.1355,
+      "step": 1089
+    },
+    {
+      "epoch": 0.11219562313777869,
+      "grad_norm": 0.2511363923549652,
+      "learning_rate": 0.004486442070665571,
+      "loss": 3.1041,
+      "step": 1092
+    },
+    {
+      "epoch": 0.11250385287167369,
+      "grad_norm": 0.13533104956150055,
+      "learning_rate": 0.004498767460969597,
+      "loss": 3.1006,
+      "step": 1095
+    },
+    {
+      "epoch": 0.11281208260556869,
+      "grad_norm": 0.1323193609714508,
+      "learning_rate": 0.004511092851273624,
+      "loss": 3.0623,
+      "step": 1098
+    },
+    {
+      "epoch": 0.11312031233946368,
+      "grad_norm": 0.24355067312717438,
+      "learning_rate": 0.00452341824157765,
+      "loss": 3.109,
+      "step": 1101
+    },
+    {
+      "epoch": 0.11342854207335867,
+      "grad_norm": 0.45989617705345154,
+      "learning_rate": 0.004535743631881676,
+      "loss": 3.1102,
+      "step": 1104
+    },
+    {
+      "epoch": 0.11373677180725367,
+      "grad_norm": 0.27389761805534363,
+      "learning_rate": 0.004548069022185702,
+      "loss": 3.1058,
+      "step": 1107
+    },
+    {
+      "epoch": 0.11404500154114867,
+      "grad_norm": 0.3120715320110321,
+      "learning_rate": 0.004560394412489729,
+      "loss": 3.0936,
+      "step": 1110
+    },
+    {
+      "epoch": 0.11435323127504367,
+      "grad_norm": 0.3641244173049927,
+      "learning_rate": 0.004572719802793755,
+      "loss": 3.0895,
+      "step": 1113
+    },
+    {
+      "epoch": 0.11466146100893866,
+      "grad_norm": 0.16439078748226166,
+      "learning_rate": 0.004585045193097782,
+      "loss": 3.0697,
+      "step": 1116
+    },
+    {
+      "epoch": 0.11496969074283366,
+      "grad_norm": 0.21766935288906097,
+      "learning_rate": 0.004597370583401808,
+      "loss": 3.0952,
+      "step": 1119
+    },
+    {
+      "epoch": 0.11527792047672865,
+      "grad_norm": 0.1682632714509964,
+      "learning_rate": 0.004609695973705834,
+      "loss": 3.0644,
+      "step": 1122
+    },
+    {
+      "epoch": 0.11558615021062366,
+      "grad_norm": 0.18391060829162598,
+      "learning_rate": 0.00462202136400986,
+      "loss": 3.0565,
+      "step": 1125
+    },
+    {
+      "epoch": 0.11589437994451865,
+      "grad_norm": 0.2503467798233032,
+      "learning_rate": 0.004634346754313887,
+      "loss": 3.0798,
+      "step": 1128
+    },
+    {
+      "epoch": 0.11620260967841364,
+      "grad_norm": 0.3139159083366394,
+      "learning_rate": 0.004646672144617912,
+      "loss": 3.0784,
+      "step": 1131
+    },
+    {
+      "epoch": 0.11651083941230864,
+      "grad_norm": 0.2205217182636261,
+      "learning_rate": 0.004658997534921939,
+      "loss": 3.0696,
+      "step": 1134
+    },
+    {
+      "epoch": 0.11681906914620364,
+      "grad_norm": 0.322355180978775,
+      "learning_rate": 0.004671322925225966,
+      "loss": 3.0811,
+      "step": 1137
+    },
+    {
+      "epoch": 0.11712729888009864,
+      "grad_norm": 0.27023863792419434,
+      "learning_rate": 0.004683648315529992,
+      "loss": 3.0955,
+      "step": 1140
+    },
+    {
+      "epoch": 0.11743552861399363,
+      "grad_norm": 0.2672137916088104,
+      "learning_rate": 0.004695973705834018,
+      "loss": 3.0584,
+      "step": 1143
+    },
+    {
+      "epoch": 0.11774375834788862,
+      "grad_norm": 0.271323561668396,
+      "learning_rate": 0.0047082990961380446,
+      "loss": 3.0483,
+      "step": 1146
+    },
+    {
+      "epoch": 0.11805198808178362,
+      "grad_norm": 0.1428508758544922,
+      "learning_rate": 0.00472062448644207,
+      "loss": 3.0661,
+      "step": 1149
+    },
+    {
+      "epoch": 0.11836021781567863,
+      "grad_norm": 0.29395970702171326,
+      "learning_rate": 0.004732949876746097,
+      "loss": 3.0391,
+      "step": 1152
+    },
+    {
+      "epoch": 0.11866844754957362,
+      "grad_norm": 0.22083403170108795,
+      "learning_rate": 0.0047452752670501235,
+      "loss": 3.0579,
+      "step": 1155
+    },
+    {
+      "epoch": 0.11897667728346861,
+      "grad_norm": 0.2015424370765686,
+      "learning_rate": 0.00475760065735415,
+      "loss": 3.0356,
+      "step": 1158
+    },
+    {
+      "epoch": 0.1192849070173636,
+      "grad_norm": 0.21997034549713135,
+      "learning_rate": 0.004769926047658176,
+      "loss": 3.0301,
+      "step": 1161
+    },
+    {
+      "epoch": 0.1195931367512586,
+      "grad_norm": 0.16206422448158264,
+      "learning_rate": 0.0047822514379622025,
+      "loss": 3.0407,
+      "step": 1164
+    },
+    {
+      "epoch": 0.11990136648515361,
+      "grad_norm": 0.22591377794742584,
+      "learning_rate": 0.004794576828266228,
+      "loss": 3.0414,
+      "step": 1167
+    },
+    {
+      "epoch": 0.1202095962190486,
+      "grad_norm": 0.2582632601261139,
+      "learning_rate": 0.004806902218570255,
+      "loss": 3.0148,
+      "step": 1170
+    },
+    {
+      "epoch": 0.1205178259529436,
+      "grad_norm": 0.273416131734848,
+      "learning_rate": 0.004819227608874281,
+      "loss": 3.0023,
+      "step": 1173
+    },
+    {
+      "epoch": 0.12082605568683859,
+      "grad_norm": 0.16373753547668457,
+      "learning_rate": 0.004831552999178307,
+      "loss": 3.0127,
+      "step": 1176
+    },
+    {
+      "epoch": 0.12113428542073358,
+      "grad_norm": 0.2623594105243683,
+      "learning_rate": 0.004843878389482334,
+      "loss": 3.0635,
+      "step": 1179
+    },
+    {
+      "epoch": 0.12144251515462859,
+      "grad_norm": 0.34809616208076477,
+      "learning_rate": 0.00485620377978636,
+      "loss": 3.0222,
+      "step": 1182
+    },
+    {
+      "epoch": 0.12175074488852358,
+      "grad_norm": 0.23841938376426697,
+      "learning_rate": 0.004868529170090386,
+      "loss": 3.019,
+      "step": 1185
+    },
+    {
+      "epoch": 0.12205897462241858,
+      "grad_norm": 0.2161986231803894,
+      "learning_rate": 0.004880854560394413,
+      "loss": 2.9934,
+      "step": 1188
+    },
+    {
+      "epoch": 0.12236720435631357,
+      "grad_norm": 0.2870507836341858,
+      "learning_rate": 0.0048931799506984385,
+      "loss": 3.0438,
+      "step": 1191
+    },
+    {
+      "epoch": 0.12267543409020856,
+      "grad_norm": 0.20796675980091095,
+      "learning_rate": 0.004905505341002465,
+      "loss": 2.9947,
+      "step": 1194
+    },
+    {
+      "epoch": 0.12298366382410357,
+      "grad_norm": 0.1762983798980713,
+      "learning_rate": 0.004917830731306492,
+      "loss": 2.9729,
+      "step": 1197
+    },
+    {
+      "epoch": 0.12329189355799856,
+      "grad_norm": 0.1240881159901619,
+      "learning_rate": 0.0049301561216105174,
+      "loss": 3.0149,
+      "step": 1200
+    },
+    {
+      "epoch": 0.12360012329189356,
+      "grad_norm": 0.16968263685703278,
+      "learning_rate": 0.004942481511914544,
+      "loss": 2.9944,
+      "step": 1203
+    },
+    {
+      "epoch": 0.12390835302578855,
+      "grad_norm": 0.1743592470884323,
+      "learning_rate": 0.004954806902218571,
+      "loss": 2.9947,
+      "step": 1206
+    },
+    {
+      "epoch": 0.12421658275968354,
+      "grad_norm": 0.29677319526672363,
+      "learning_rate": 0.004967132292522596,
+      "loss": 2.9922,
+      "step": 1209
+    },
+    {
+      "epoch": 0.12452481249357855,
+      "grad_norm": 0.273882657289505,
+      "learning_rate": 0.004979457682826623,
+      "loss": 2.9698,
+      "step": 1212
+    },
+    {
+      "epoch": 0.12483304222747355,
+      "grad_norm": 0.3060019910335541,
+      "learning_rate": 0.004991783073130649,
+      "loss": 2.9925,
+      "step": 1215
+    },
+    {
+      "epoch": 0.12514127196136854,
+      "grad_norm": 0.13856515288352966,
+      "learning_rate": 0.005004108463434675,
+      "loss": 3.0212,
+      "step": 1218
+    },
+    {
+      "epoch": 0.12544950169526353,
+      "grad_norm": 0.12940354645252228,
+      "learning_rate": 0.005016433853738702,
+      "loss": 2.9472,
+      "step": 1221
+    },
+    {
+      "epoch": 0.12575773142915853,
+      "grad_norm": 0.15493866801261902,
+      "learning_rate": 0.005028759244042728,
+      "loss": 2.9859,
+      "step": 1224
+    },
+    {
+      "epoch": 0.12606596116305352,
+      "grad_norm": 0.4994816184043884,
+      "learning_rate": 0.005041084634346754,
+      "loss": 2.949,
+      "step": 1227
+    },
+    {
+      "epoch": 0.1263741908969485,
+      "grad_norm": 0.37235137820243835,
+      "learning_rate": 0.005053410024650781,
+      "loss": 3.006,
+      "step": 1230
+    },
+    {
+      "epoch": 0.12668242063084353,
+      "grad_norm": 0.24599948525428772,
+      "learning_rate": 0.0050657354149548075,
+      "loss": 2.9954,
+      "step": 1233
+    },
+    {
+      "epoch": 0.12699065036473853,
+      "grad_norm": 0.1838703751564026,
+      "learning_rate": 0.005078060805258833,
+      "loss": 2.9886,
+      "step": 1236
+    },
+    {
+      "epoch": 0.12729888009863352,
+      "grad_norm": 0.19366377592086792,
+      "learning_rate": 0.005090386195562859,
+      "loss": 2.9715,
+      "step": 1239
+    },
+    {
+      "epoch": 0.12760710983252851,
+      "grad_norm": 0.11911759525537491,
+      "learning_rate": 0.005102711585866886,
+      "loss": 2.965,
+      "step": 1242
+    },
+    {
+      "epoch": 0.1279153395664235,
+      "grad_norm": 0.12456653267145157,
+      "learning_rate": 0.005115036976170912,
+      "loss": 2.9343,
+      "step": 1245
+    },
+    {
+      "epoch": 0.1282235693003185,
+      "grad_norm": 0.322380393743515,
+      "learning_rate": 0.005127362366474938,
+      "loss": 2.9604,
+      "step": 1248
+    },
+    {
+      "epoch": 0.1285317990342135,
+      "grad_norm": 0.40975773334503174,
+      "learning_rate": 0.005139687756778965,
+      "loss": 2.9386,
+      "step": 1251
+    },
+    {
+      "epoch": 0.1288400287681085,
+      "grad_norm": 0.2045045793056488,
+      "learning_rate": 0.005152013147082991,
+      "loss": 2.9459,
+      "step": 1254
+    },
+    {
+      "epoch": 0.12914825850200348,
+      "grad_norm": 0.20005717873573303,
+      "learning_rate": 0.005164338537387018,
+      "loss": 2.9631,
+      "step": 1257
+    },
+    {
+      "epoch": 0.12945648823589848,
+      "grad_norm": 0.18930204212665558,
+      "learning_rate": 0.0051766639276910435,
+      "loss": 2.9014,
+      "step": 1260
+    },
+    {
+      "epoch": 0.1297647179697935,
+      "grad_norm": 0.3180810213088989,
+      "learning_rate": 0.00518898931799507,
+      "loss": 2.9242,
+      "step": 1263
+    },
+    {
+      "epoch": 0.1300729477036885,
+      "grad_norm": 0.17843572795391083,
+      "learning_rate": 0.005201314708299097,
+      "loss": 2.9063,
+      "step": 1266
+    },
+    {
+      "epoch": 0.13038117743758348,
+      "grad_norm": 0.12591248750686646,
+      "learning_rate": 0.005213640098603123,
+      "loss": 2.9095,
+      "step": 1269
+    },
+    {
+      "epoch": 0.13068940717147848,
+      "grad_norm": 0.17976878583431244,
+      "learning_rate": 0.005225965488907148,
+      "loss": 2.928,
+      "step": 1272
+    },
+    {
+      "epoch": 0.13099763690537347,
+      "grad_norm": 0.16759532690048218,
+      "learning_rate": 0.005238290879211175,
+      "loss": 2.9202,
+      "step": 1275
+    },
+    {
+      "epoch": 0.13130586663926846,
+      "grad_norm": 0.27441859245300293,
+      "learning_rate": 0.0052506162695152014,
+      "loss": 2.9242,
+      "step": 1278
+    },
+    {
+      "epoch": 0.13161409637316346,
+      "grad_norm": 0.23654502630233765,
+      "learning_rate": 0.005262941659819228,
+      "loss": 2.9175,
+      "step": 1281
+    },
+    {
+      "epoch": 0.13192232610705845,
+      "grad_norm": 0.3399145007133484,
+      "learning_rate": 0.005275267050123254,
+      "loss": 2.9277,
+      "step": 1284
+    },
+    {
+      "epoch": 0.13223055584095345,
+      "grad_norm": 0.199320450425148,
+      "learning_rate": 0.00528759244042728,
+      "loss": 2.9184,
+      "step": 1287
+    },
+    {
+      "epoch": 0.13253878557484847,
+      "grad_norm": 0.16563403606414795,
+      "learning_rate": 0.005299917830731307,
+      "loss": 2.9166,
+      "step": 1290
+    },
+    {
+      "epoch": 0.13284701530874346,
+      "grad_norm": 0.18119758367538452,
+      "learning_rate": 0.005312243221035334,
+      "loss": 2.9239,
+      "step": 1293
+    },
+    {
+      "epoch": 0.13315524504263845,
+      "grad_norm": 0.1558375358581543,
+      "learning_rate": 0.0053245686113393585,
+      "loss": 2.9028,
+      "step": 1296
+    },
+    {
+      "epoch": 0.13346347477653345,
+      "grad_norm": 0.36665746569633484,
+      "learning_rate": 0.005336894001643385,
+      "loss": 2.9081,
+      "step": 1299
+    },
+    {
+      "epoch": 0.13377170451042844,
+      "grad_norm": 0.186012864112854,
+      "learning_rate": 0.005349219391947412,
+      "loss": 2.8836,
+      "step": 1302
+    },
+    {
+      "epoch": 0.13407993424432343,
+      "grad_norm": 0.14102259278297424,
+      "learning_rate": 0.005361544782251438,
+      "loss": 2.8906,
+      "step": 1305
+    },
+    {
+      "epoch": 0.13438816397821843,
+      "grad_norm": 0.12519022822380066,
+      "learning_rate": 0.005373870172555464,
+      "loss": 2.9148,
+      "step": 1308
+    },
+    {
+      "epoch": 0.13469639371211342,
+      "grad_norm": 0.14027029275894165,
+      "learning_rate": 0.005386195562859491,
+      "loss": 2.9108,
+      "step": 1311
+    },
+    {
+      "epoch": 0.13500462344600841,
+      "grad_norm": 0.2553085684776306,
+      "learning_rate": 0.005398520953163517,
+      "loss": 2.8837,
+      "step": 1314
+    },
+    {
+      "epoch": 0.1353128531799034,
+      "grad_norm": 0.2809675335884094,
+      "learning_rate": 0.005410846343467544,
+      "loss": 2.8795,
+      "step": 1317
+    },
+    {
+      "epoch": 0.13562108291379843,
+      "grad_norm": 0.19451378285884857,
+      "learning_rate": 0.005423171733771569,
+      "loss": 2.8648,
+      "step": 1320
+    },
+    {
+      "epoch": 0.13592931264769342,
+      "grad_norm": 0.22285006940364838,
+      "learning_rate": 0.005435497124075595,
+      "loss": 2.8994,
+      "step": 1323
+    },
+    {
+      "epoch": 0.13623754238158842,
+      "grad_norm": 0.14703693985939026,
+      "learning_rate": 0.005447822514379622,
+      "loss": 2.8984,
+      "step": 1326
+    },
+    {
+      "epoch": 0.1365457721154834,
+      "grad_norm": 0.23260341584682465,
+      "learning_rate": 0.005460147904683649,
+      "loss": 2.863,
+      "step": 1329
+    },
+    {
+      "epoch": 0.1368540018493784,
+      "grad_norm": 0.16448146104812622,
+      "learning_rate": 0.005472473294987674,
+      "loss": 2.8895,
+      "step": 1332
+    },
+    {
+      "epoch": 0.1371622315832734,
+      "grad_norm": 0.1994483470916748,
+      "learning_rate": 0.005484798685291701,
+      "loss": 2.9012,
+      "step": 1335
+    },
+    {
+      "epoch": 0.1374704613171684,
+      "grad_norm": 0.2786753177642822,
+      "learning_rate": 0.0054971240755957275,
+      "loss": 2.8753,
+      "step": 1338
+    },
+    {
+      "epoch": 0.13777869105106338,
+      "grad_norm": 0.13169367611408234,
+      "learning_rate": 0.005509449465899754,
+      "loss": 2.8567,
+      "step": 1341
+    },
+    {
+      "epoch": 0.13808692078495838,
+      "grad_norm": 0.21205192804336548,
+      "learning_rate": 0.00552177485620378,
+      "loss": 2.8523,
+      "step": 1344
+    },
+    {
+      "epoch": 0.1383951505188534,
+      "grad_norm": 0.3462331295013428,
+      "learning_rate": 0.0055341002465078065,
+      "loss": 2.881,
+      "step": 1347
+    },
+    {
+      "epoch": 0.1387033802527484,
+      "grad_norm": 0.26768332719802856,
+      "learning_rate": 0.005546425636811832,
+      "loss": 2.8803,
+      "step": 1350
+    },
+    {
+      "epoch": 0.1390116099866434,
+      "grad_norm": 0.22518084943294525,
+      "learning_rate": 0.005558751027115859,
+      "loss": 2.874,
+      "step": 1353
+    },
+    {
+      "epoch": 0.13931983972053838,
+      "grad_norm": 0.1767919361591339,
+      "learning_rate": 0.005571076417419885,
+      "loss": 2.8593,
+      "step": 1356
+    },
+    {
+      "epoch": 0.13962806945443337,
+      "grad_norm": 0.14405187964439392,
+      "learning_rate": 0.005583401807723911,
+      "loss": 2.8576,
+      "step": 1359
+    },
+    {
+      "epoch": 0.13993629918832837,
+      "grad_norm": 0.15364724397659302,
+      "learning_rate": 0.005595727198027938,
+      "loss": 2.856,
+      "step": 1362
+    },
+    {
+      "epoch": 0.14024452892222336,
+      "grad_norm": 0.26737314462661743,
+      "learning_rate": 0.005608052588331964,
+      "loss": 2.8225,
+      "step": 1365
+    },
+    {
+      "epoch": 0.14055275865611835,
+      "grad_norm": 0.14594382047653198,
+      "learning_rate": 0.00562037797863599,
+      "loss": 2.8397,
+      "step": 1368
+    },
+    {
+      "epoch": 0.14086098839001335,
+      "grad_norm": 0.1974790245294571,
+      "learning_rate": 0.005632703368940017,
+      "loss": 2.8294,
+      "step": 1371
+    },
+    {
+      "epoch": 0.14116921812390834,
+      "grad_norm": 0.12267682701349258,
+      "learning_rate": 0.005645028759244043,
+      "loss": 2.8543,
+      "step": 1374
+    },
+    {
+      "epoch": 0.14147744785780336,
+      "grad_norm": 0.14111129939556122,
+      "learning_rate": 0.00565735414954807,
+      "loss": 2.8181,
+      "step": 1377
+    },
+    {
+      "epoch": 0.14178567759169836,
+      "grad_norm": 0.1846015751361847,
+      "learning_rate": 0.005669679539852095,
+      "loss": 2.8272,
+      "step": 1380
+    },
+    {
+      "epoch": 0.14209390732559335,
+      "grad_norm": 0.26931676268577576,
+      "learning_rate": 0.0056820049301561214,
+      "loss": 2.8286,
+      "step": 1383
+    },
+    {
+      "epoch": 0.14240213705948834,
+      "grad_norm": 0.17969557642936707,
+      "learning_rate": 0.005694330320460148,
+      "loss": 2.8315,
+      "step": 1386
+    },
+    {
+      "epoch": 0.14271036679338334,
+      "grad_norm": 0.2056432068347931,
+      "learning_rate": 0.005706655710764175,
+      "loss": 2.835,
+      "step": 1389
+    },
+    {
+      "epoch": 0.14301859652727833,
+      "grad_norm": 0.29306477308273315,
+      "learning_rate": 0.0057189811010682,
+      "loss": 2.8294,
+      "step": 1392
+    },
+    {
+      "epoch": 0.14332682626117332,
+      "grad_norm": 0.1792561262845993,
+      "learning_rate": 0.005731306491372227,
+      "loss": 2.8321,
+      "step": 1395
+    },
+    {
+      "epoch": 0.14363505599506832,
+      "grad_norm": 0.11323501914739609,
+      "learning_rate": 0.005743631881676254,
+      "loss": 2.83,
+      "step": 1398
+    },
+    {
+      "epoch": 0.1439432857289633,
+      "grad_norm": 0.2804841101169586,
+      "learning_rate": 0.00575595727198028,
+      "loss": 2.8271,
+      "step": 1401
+    },
+    {
+      "epoch": 0.1442515154628583,
+      "grad_norm": 0.33056163787841797,
+      "learning_rate": 0.005768282662284305,
+      "loss": 2.7976,
+      "step": 1404
+    },
+    {
+      "epoch": 0.14455974519675333,
+      "grad_norm": 0.12834665179252625,
+      "learning_rate": 0.005780608052588332,
+      "loss": 2.8169,
+      "step": 1407
+    },
+    {
+      "epoch": 0.14486797493064832,
+      "grad_norm": 0.15917035937309265,
+      "learning_rate": 0.005792933442892358,
+      "loss": 2.8124,
+      "step": 1410
+    },
+    {
+      "epoch": 0.1451762046645433,
+      "grad_norm": 0.28015008568763733,
+      "learning_rate": 0.005805258833196385,
+      "loss": 2.8019,
+      "step": 1413
+    },
+    {
+      "epoch": 0.1454844343984383,
+      "grad_norm": 0.16829009354114532,
+      "learning_rate": 0.005817584223500411,
+      "loss": 2.8357,
+      "step": 1416
+    },
+    {
+      "epoch": 0.1457926641323333,
+      "grad_norm": 0.14804339408874512,
+      "learning_rate": 0.005829909613804437,
+      "loss": 2.8102,
+      "step": 1419
+    },
+    {
+      "epoch": 0.1461008938662283,
+      "grad_norm": 0.20360830426216125,
+      "learning_rate": 0.005842235004108464,
+      "loss": 2.8211,
+      "step": 1422
+    },
+    {
+      "epoch": 0.1464091236001233,
+      "grad_norm": 0.22152036428451538,
+      "learning_rate": 0.0058545603944124905,
+      "loss": 2.8103,
+      "step": 1425
+    },
+    {
+      "epoch": 0.14671735333401828,
+      "grad_norm": 0.20746375620365143,
+      "learning_rate": 0.005866885784716516,
+      "loss": 2.7994,
+      "step": 1428
+    },
+    {
+      "epoch": 0.14702558306791327,
+      "grad_norm": 0.16845661401748657,
+      "learning_rate": 0.005879211175020542,
+      "loss": 2.8286,
+      "step": 1431
+    },
+    {
+      "epoch": 0.1473338128018083,
+      "grad_norm": 0.1094370111823082,
+      "learning_rate": 0.005891536565324569,
+      "loss": 2.7888,
+      "step": 1434
+    },
+    {
+      "epoch": 0.1476420425357033,
+      "grad_norm": 0.14844520390033722,
+      "learning_rate": 0.005903861955628595,
+      "loss": 2.8035,
+      "step": 1437
+    },
+    {
+      "epoch": 0.14795027226959828,
+      "grad_norm": 0.12289691716432571,
+      "learning_rate": 0.005916187345932621,
+      "loss": 2.7852,
+      "step": 1440
+    },
+    {
+      "epoch": 0.14825850200349328,
+      "grad_norm": 0.1203322485089302,
+      "learning_rate": 0.0059285127362366475,
+      "loss": 2.8101,
+      "step": 1443
+    },
+    {
+      "epoch": 0.14856673173738827,
+      "grad_norm": 0.1871965080499649,
+      "learning_rate": 0.005940838126540674,
+      "loss": 2.7485,
+      "step": 1446
+    },
+    {
+      "epoch": 0.14887496147128326,
+      "grad_norm": 0.1567300707101822,
+      "learning_rate": 0.005953163516844701,
+      "loss": 2.8097,
+      "step": 1449
+    },
+    {
+      "epoch": 0.14918319120517826,
+      "grad_norm": 0.18046674132347107,
+      "learning_rate": 0.0059654889071487265,
+      "loss": 2.8118,
+      "step": 1452
+    },
+    {
+      "epoch": 0.14949142093907325,
+      "grad_norm": 0.23180244863033295,
+      "learning_rate": 0.005977814297452753,
+      "loss": 2.7836,
+      "step": 1455
+    },
+    {
+      "epoch": 0.14979965067296824,
+      "grad_norm": 0.2300175577402115,
+      "learning_rate": 0.00599013968775678,
+      "loss": 2.7675,
+      "step": 1458
+    },
+    {
+      "epoch": 0.15010788040686324,
+      "grad_norm": 0.11340396106243134,
+      "learning_rate": 0.006002465078060806,
+      "loss": 2.8012,
+      "step": 1461
+    },
+    {
+      "epoch": 0.15041611014075826,
+      "grad_norm": 0.10667074471712112,
+      "learning_rate": 0.006014790468364831,
+      "loss": 2.8154,
+      "step": 1464
+    },
+    {
+      "epoch": 0.15072433987465325,
+      "grad_norm": 0.10800652205944061,
+      "learning_rate": 0.006027115858668858,
+      "loss": 2.7646,
+      "step": 1467
+    },
+    {
+      "epoch": 0.15103256960854824,
+      "grad_norm": 0.2588643431663513,
+      "learning_rate": 0.006039441248972884,
+      "loss": 2.7912,
+      "step": 1470
+    },
+    {
+      "epoch": 0.15134079934244324,
+      "grad_norm": 0.32462435960769653,
+      "learning_rate": 0.006051766639276911,
+      "loss": 2.7666,
+      "step": 1473
+    },
+    {
+      "epoch": 0.15164902907633823,
+      "grad_norm": 0.23754975199699402,
+      "learning_rate": 0.006064092029580937,
+      "loss": 2.7694,
+      "step": 1476
+    },
+    {
+      "epoch": 0.15195725881023323,
+      "grad_norm": 0.14895015954971313,
+      "learning_rate": 0.006076417419884963,
+      "loss": 2.7678,
+      "step": 1479
+    },
+    {
+      "epoch": 0.15226548854412822,
+      "grad_norm": 0.3228299021720886,
+      "learning_rate": 0.00608874281018899,
+      "loss": 2.7786,
+      "step": 1482
+    },
+    {
+      "epoch": 0.1525737182780232,
+      "grad_norm": 0.15597562491893768,
+      "learning_rate": 0.006101068200493017,
+      "loss": 2.7967,
+      "step": 1485
+    },
+    {
+      "epoch": 0.1528819480119182,
+      "grad_norm": 0.09748488664627075,
+      "learning_rate": 0.0061133935907970414,
+      "loss": 2.7673,
+      "step": 1488
+    },
+    {
+      "epoch": 0.1531901777458132,
+      "grad_norm": 0.12523339688777924,
+      "learning_rate": 0.006125718981101068,
+      "loss": 2.7391,
+      "step": 1491
+    },
+    {
+      "epoch": 0.15349840747970822,
+      "grad_norm": 0.16529253125190735,
+      "learning_rate": 0.006138044371405095,
+      "loss": 2.7642,
+      "step": 1494
+    },
+    {
+      "epoch": 0.15380663721360321,
+      "grad_norm": 0.2083311527967453,
+      "learning_rate": 0.006150369761709121,
+      "loss": 2.764,
+      "step": 1497
+    },
+    {
+      "epoch": 0.1541148669474982,
+      "grad_norm": 0.13263079524040222,
+      "learning_rate": 0.006162695152013147,
+      "loss": 2.7828,
+      "step": 1500
+    },
+    {
+      "epoch": 0.1544230966813932,
+      "grad_norm": 0.1473417580127716,
+      "learning_rate": 0.006175020542317174,
+      "loss": 2.7574,
+      "step": 1503
+    },
+    {
+      "epoch": 0.1547313264152882,
+      "grad_norm": 0.22629734873771667,
+      "learning_rate": 0.0061873459326212,
+      "loss": 2.7792,
+      "step": 1506
+    },
+    {
+      "epoch": 0.1550395561491832,
+      "grad_norm": 0.21652548015117645,
+      "learning_rate": 0.006199671322925227,
+      "loss": 2.7785,
+      "step": 1509
+    },
+    {
+      "epoch": 0.15534778588307818,
+      "grad_norm": 0.1948641836643219,
+      "learning_rate": 0.006211996713229252,
+      "loss": 2.7969,
+      "step": 1512
+    },
+    {
+      "epoch": 0.15565601561697318,
+      "grad_norm": 0.13890105485916138,
+      "learning_rate": 0.006224322103533278,
+      "loss": 2.7856,
+      "step": 1515
+    },
+    {
+      "epoch": 0.15596424535086817,
+      "grad_norm": 0.09859870374202728,
+      "learning_rate": 0.006236647493837305,
+      "loss": 2.7523,
+      "step": 1518
+    },
+    {
+      "epoch": 0.1562724750847632,
+      "grad_norm": 0.10258977860212326,
+      "learning_rate": 0.0062489728841413315,
+      "loss": 2.7466,
+      "step": 1521
+    },
+    {
+      "epoch": 0.15658070481865818,
+      "grad_norm": 0.11476584523916245,
+      "learning_rate": 0.006261298274445357,
+      "loss": 2.7314,
+      "step": 1524
+    },
+    {
+      "epoch": 0.15688893455255318,
+      "grad_norm": 0.1920320987701416,
+      "learning_rate": 0.006273623664749384,
+      "loss": 2.7647,
+      "step": 1527
+    },
+    {
+      "epoch": 0.15719716428644817,
+      "grad_norm": 0.18576020002365112,
+      "learning_rate": 0.0062859490550534105,
+      "loss": 2.7632,
+      "step": 1530
+    },
+    {
+      "epoch": 0.15750539402034316,
+      "grad_norm": 0.128046452999115,
+      "learning_rate": 0.006298274445357437,
+      "loss": 2.7237,
+      "step": 1533
+    },
+    {
+      "epoch": 0.15781362375423816,
+      "grad_norm": 0.30617430806159973,
+      "learning_rate": 0.006310599835661463,
+      "loss": 2.7907,
+      "step": 1536
+    },
+    {
+      "epoch": 0.15812185348813315,
+      "grad_norm": 0.140928253531456,
+      "learning_rate": 0.0063229252259654894,
+      "loss": 2.7879,
+      "step": 1539
+    },
+    {
+      "epoch": 0.15843008322202815,
+      "grad_norm": 0.2537645399570465,
+      "learning_rate": 0.006335250616269515,
+      "loss": 2.7513,
+      "step": 1542
+    },
+    {
+      "epoch": 0.15873831295592314,
+      "grad_norm": 0.40944191813468933,
+      "learning_rate": 0.006347576006573542,
+      "loss": 2.7418,
+      "step": 1545
+    },
+    {
+      "epoch": 0.15904654268981813,
+      "grad_norm": 0.1284068077802658,
+      "learning_rate": 0.0063599013968775675,
+      "loss": 2.7235,
+      "step": 1548
+    },
+    {
+      "epoch": 0.15935477242371315,
+      "grad_norm": 0.08984164893627167,
+      "learning_rate": 0.006372226787181594,
+      "loss": 2.7414,
+      "step": 1551
+    },
+    {
+      "epoch": 0.15966300215760815,
+      "grad_norm": 0.13366155326366425,
+      "learning_rate": 0.006384552177485621,
+      "loss": 2.7456,
+      "step": 1554
+    },
+    {
+      "epoch": 0.15997123189150314,
+      "grad_norm": 0.1179983913898468,
+      "learning_rate": 0.006396877567789647,
+      "loss": 2.7313,
+      "step": 1557
+    },
+    {
+      "epoch": 0.16027946162539813,
+      "grad_norm": 0.15718503296375275,
+      "learning_rate": 0.006409202958093673,
+      "loss": 2.7315,
+      "step": 1560
+    },
+    {
+      "epoch": 0.16058769135929313,
+      "grad_norm": 0.14405110478401184,
+      "learning_rate": 0.0064215283483977,
+      "loss": 2.7275,
+      "step": 1563
+    },
+    {
+      "epoch": 0.16089592109318812,
+      "grad_norm": 0.13050544261932373,
+      "learning_rate": 0.006433853738701726,
+      "loss": 2.6935,
+      "step": 1566
+    },
+    {
+      "epoch": 0.16120415082708311,
+      "grad_norm": 0.2343079298734665,
+      "learning_rate": 0.006446179129005751,
+      "loss": 2.6932,
+      "step": 1569
+    },
+    {
+      "epoch": 0.1615123805609781,
+      "grad_norm": 0.2493698000907898,
+      "learning_rate": 0.006458504519309778,
+      "loss": 2.7414,
+      "step": 1572
+    },
+    {
+      "epoch": 0.1618206102948731,
+      "grad_norm": 0.17371931672096252,
+      "learning_rate": 0.006470829909613804,
+      "loss": 2.7522,
+      "step": 1575
+    },
+    {
+      "epoch": 0.1621288400287681,
+      "grad_norm": 0.16282691061496735,
+      "learning_rate": 0.006483155299917831,
+      "loss": 2.7659,
+      "step": 1578
+    },
+    {
+      "epoch": 0.16243706976266312,
+      "grad_norm": 0.12791027128696442,
+      "learning_rate": 0.006495480690221857,
+      "loss": 2.7077,
+      "step": 1581
+    },
+    {
+      "epoch": 0.1627452994965581,
+      "grad_norm": 0.09789251536130905,
+      "learning_rate": 0.006507806080525883,
+      "loss": 2.7041,
+      "step": 1584
+    },
+    {
+      "epoch": 0.1630535292304531,
+      "grad_norm": 0.10156393051147461,
+      "learning_rate": 0.00652013147082991,
+      "loss": 2.685,
+      "step": 1587
+    },
+    {
+      "epoch": 0.1633617589643481,
+      "grad_norm": 0.1974211484193802,
+      "learning_rate": 0.006532456861133937,
+      "loss": 2.7183,
+      "step": 1590
+    },
+    {
+      "epoch": 0.1636699886982431,
+      "grad_norm": 0.1420728713274002,
+      "learning_rate": 0.0065447822514379615,
+      "loss": 2.7095,
+      "step": 1593
+    },
+    {
+      "epoch": 0.16397821843213808,
+      "grad_norm": 0.3637617528438568,
+      "learning_rate": 0.006557107641741988,
+      "loss": 2.7578,
+      "step": 1596
+    },
+    {
+      "epoch": 0.16428644816603308,
+      "grad_norm": 0.09830935299396515,
+      "learning_rate": 0.006569433032046015,
+      "loss": 2.6937,
+      "step": 1599
+    },
+    {
+      "epoch": 0.16459467789992807,
+      "grad_norm": 0.15821218490600586,
+      "learning_rate": 0.006581758422350041,
+      "loss": 2.7031,
+      "step": 1602
+    },
+    {
+      "epoch": 0.16490290763382306,
+      "grad_norm": 0.17226357758045197,
+      "learning_rate": 0.006594083812654067,
+      "loss": 2.6702,
+      "step": 1605
+    },
+    {
+      "epoch": 0.16521113736771809,
+      "grad_norm": 0.21252015233039856,
+      "learning_rate": 0.006606409202958094,
+      "loss": 2.6893,
+      "step": 1608
+    },
+    {
+      "epoch": 0.16551936710161308,
+      "grad_norm": 0.11433108150959015,
+      "learning_rate": 0.00661873459326212,
+      "loss": 2.6852,
+      "step": 1611
+    },
+    {
+      "epoch": 0.16582759683550807,
+      "grad_norm": 0.15884144604206085,
+      "learning_rate": 0.006631059983566147,
+      "loss": 2.7164,
+      "step": 1614
+    },
+    {
+      "epoch": 0.16613582656940307,
+      "grad_norm": 0.1429038643836975,
+      "learning_rate": 0.006643385373870173,
+      "loss": 2.6976,
+      "step": 1617
+    },
+    {
+      "epoch": 0.16644405630329806,
+      "grad_norm": 0.09187953919172287,
+      "learning_rate": 0.006655710764174198,
+      "loss": 2.7134,
+      "step": 1620
+    },
+    {
+      "epoch": 0.16675228603719305,
+      "grad_norm": 0.13670755922794342,
+      "learning_rate": 0.006668036154478225,
+      "loss": 2.6951,
+      "step": 1623
+    },
+    {
+      "epoch": 0.16706051577108805,
+      "grad_norm": 0.17965632677078247,
+      "learning_rate": 0.0066803615447822515,
+      "loss": 2.6911,
+      "step": 1626
+    },
+    {
+      "epoch": 0.16736874550498304,
+      "grad_norm": 0.21141032874584198,
+      "learning_rate": 0.006692686935086277,
+      "loss": 2.67,
+      "step": 1629
+    },
+    {
+      "epoch": 0.16767697523887803,
+      "grad_norm": 0.30064719915390015,
+      "learning_rate": 0.006705012325390304,
+      "loss": 2.6837,
+      "step": 1632
+    },
+    {
+      "epoch": 0.16798520497277303,
+      "grad_norm": 0.11874115467071533,
+      "learning_rate": 0.0067173377156943305,
+      "loss": 2.6968,
+      "step": 1635
+    },
+    {
+      "epoch": 0.16829343470666805,
+      "grad_norm": 0.10265806317329407,
+      "learning_rate": 0.006729663105998357,
+      "loss": 2.6632,
+      "step": 1638
+    },
+    {
+      "epoch": 0.16860166444056304,
+      "grad_norm": 0.10916320979595184,
+      "learning_rate": 0.006741988496302383,
+      "loss": 2.6749,
+      "step": 1641
+    },
+    {
+      "epoch": 0.16890989417445804,
+      "grad_norm": 0.2549231946468353,
+      "learning_rate": 0.0067543138866064095,
+      "loss": 2.636,
+      "step": 1644
+    },
+    {
+      "epoch": 0.16921812390835303,
+      "grad_norm": 0.15071339905261993,
+      "learning_rate": 0.006766639276910436,
+      "loss": 2.6933,
+      "step": 1647
+    },
+    {
+      "epoch": 0.16952635364224802,
+      "grad_norm": 0.1088666021823883,
+      "learning_rate": 0.006778964667214463,
+      "loss": 2.6477,
+      "step": 1650
+    },
+    {
+      "epoch": 0.16983458337614302,
+      "grad_norm": 0.0984036773443222,
+      "learning_rate": 0.0067912900575184875,
+      "loss": 2.6801,
+      "step": 1653
+    },
+    {
+      "epoch": 0.170142813110038,
+      "grad_norm": 0.15402089059352875,
+      "learning_rate": 0.006803615447822514,
+      "loss": 2.6877,
+      "step": 1656
+    },
+    {
+      "epoch": 0.170451042843933,
+      "grad_norm": 0.1299775093793869,
+      "learning_rate": 0.006815940838126541,
+      "loss": 2.6717,
+      "step": 1659
+    },
+    {
+      "epoch": 0.170759272577828,
+      "grad_norm": 0.15615323185920715,
+      "learning_rate": 0.006828266228430567,
+      "loss": 2.6578,
+      "step": 1662
+    },
+    {
+      "epoch": 0.171067502311723,
+      "grad_norm": 0.122567318379879,
+      "learning_rate": 0.006840591618734593,
+      "loss": 2.6959,
+      "step": 1665
+    },
+    {
+      "epoch": 0.171375732045618,
+      "grad_norm": 0.1386043280363083,
+      "learning_rate": 0.00685291700903862,
+      "loss": 2.6491,
+      "step": 1668
+    },
+    {
+      "epoch": 0.171683961779513,
+      "grad_norm": 0.1900375783443451,
+      "learning_rate": 0.006865242399342646,
+      "loss": 2.6643,
+      "step": 1671
+    },
+    {
+      "epoch": 0.171992191513408,
+      "grad_norm": 0.1118064671754837,
+      "learning_rate": 0.006877567789646673,
+      "loss": 2.6496,
+      "step": 1674
+    },
+    {
+      "epoch": 0.172300421247303,
+      "grad_norm": 0.1593448519706726,
+      "learning_rate": 0.006889893179950698,
+      "loss": 2.6833,
+      "step": 1677
+    },
+    {
+      "epoch": 0.172608650981198,
+      "grad_norm": 0.17275281250476837,
+      "learning_rate": 0.006902218570254724,
+      "loss": 2.6909,
+      "step": 1680
+    },
+    {
+      "epoch": 0.17291688071509298,
+      "grad_norm": 0.13396479189395905,
+      "learning_rate": 0.006914543960558751,
+      "loss": 2.692,
+      "step": 1683
+    },
+    {
+      "epoch": 0.17322511044898797,
+      "grad_norm": 0.09812068939208984,
+      "learning_rate": 0.006926869350862778,
+      "loss": 2.6939,
+      "step": 1686
+    },
+    {
+      "epoch": 0.17353334018288297,
+      "grad_norm": 0.08181022852659225,
+      "learning_rate": 0.006939194741166803,
+      "loss": 2.6408,
+      "step": 1689
+    },
+    {
+      "epoch": 0.17384156991677796,
+      "grad_norm": 0.15573051571846008,
+      "learning_rate": 0.00695152013147083,
+      "loss": 2.6647,
+      "step": 1692
+    },
+    {
+      "epoch": 0.17414979965067298,
+      "grad_norm": 0.2834240198135376,
+      "learning_rate": 0.006963845521774857,
+      "loss": 2.6585,
+      "step": 1695
+    },
+    {
+      "epoch": 0.17445802938456798,
+      "grad_norm": 0.23794801533222198,
+      "learning_rate": 0.006976170912078883,
+      "loss": 2.6559,
+      "step": 1698
+    },
+    {
+      "epoch": 0.17476625911846297,
+      "grad_norm": 0.1332167536020279,
+      "learning_rate": 0.006988496302382908,
+      "loss": 2.6695,
+      "step": 1701
+    },
+    {
+      "epoch": 0.17507448885235796,
+      "grad_norm": 0.09555593878030777,
+      "learning_rate": 0.007000821692686935,
+      "loss": 2.6811,
+      "step": 1704
+    },
+    {
+      "epoch": 0.17538271858625296,
+      "grad_norm": 0.10987939685583115,
+      "learning_rate": 0.007013147082990961,
+      "loss": 2.6524,
+      "step": 1707
+    },
+    {
+      "epoch": 0.17569094832014795,
+      "grad_norm": 0.11458218097686768,
+      "learning_rate": 0.007025472473294988,
+      "loss": 2.6085,
+      "step": 1710
+    },
+    {
+      "epoch": 0.17599917805404294,
+      "grad_norm": 0.12646709382534027,
+      "learning_rate": 0.007037797863599014,
+      "loss": 2.6561,
+      "step": 1713
+    },
+    {
+      "epoch": 0.17630740778793794,
+      "grad_norm": 0.15338967740535736,
+      "learning_rate": 0.00705012325390304,
+      "loss": 2.6471,
+      "step": 1716
+    },
+    {
+      "epoch": 0.17661563752183293,
+      "grad_norm": 0.14660318195819855,
+      "learning_rate": 0.007062448644207067,
+      "loss": 2.6532,
+      "step": 1719
+    },
+    {
+      "epoch": 0.17692386725572792,
+      "grad_norm": 0.2730877995491028,
+      "learning_rate": 0.0070747740345110935,
+      "loss": 2.6565,
+      "step": 1722
+    },
+    {
+      "epoch": 0.17723209698962294,
+      "grad_norm": 0.26743727922439575,
+      "learning_rate": 0.007087099424815119,
+      "loss": 2.6707,
+      "step": 1725
+    },
+    {
+      "epoch": 0.17754032672351794,
+      "grad_norm": 0.13842618465423584,
+      "learning_rate": 0.007099424815119146,
+      "loss": 2.6652,
+      "step": 1728
+    },
+    {
+      "epoch": 0.17784855645741293,
+      "grad_norm": 0.15871621668338776,
+      "learning_rate": 0.0071117502054231715,
+      "loss": 2.6464,
+      "step": 1731
+    },
+    {
+      "epoch": 0.17815678619130793,
+      "grad_norm": 0.11526347696781158,
+      "learning_rate": 0.007124075595727198,
+      "loss": 2.662,
+      "step": 1734
+    },
+    {
+      "epoch": 0.17846501592520292,
+      "grad_norm": 0.21620534360408783,
+      "learning_rate": 0.007136400986031224,
+      "loss": 2.6603,
+      "step": 1737
+    },
+    {
+      "epoch": 0.1787732456590979,
+      "grad_norm": 0.0905444398522377,
+      "learning_rate": 0.0071487263763352505,
+      "loss": 2.6523,
+      "step": 1740
+    },
+    {
+      "epoch": 0.1790814753929929,
+      "grad_norm": 0.28233054280281067,
+      "learning_rate": 0.007161051766639277,
+      "loss": 2.6597,
+      "step": 1743
+    },
+    {
+      "epoch": 0.1793897051268879,
+      "grad_norm": 0.2363336831331253,
+      "learning_rate": 0.007173377156943304,
+      "loss": 2.6483,
+      "step": 1746
+    },
+    {
+      "epoch": 0.1796979348607829,
+      "grad_norm": 0.11012139916419983,
+      "learning_rate": 0.0071857025472473295,
+      "loss": 2.6513,
+      "step": 1749
+    },
+    {
+      "epoch": 0.1800061645946779,
+      "grad_norm": 0.09720948338508606,
+      "learning_rate": 0.007198027937551356,
+      "loss": 2.6511,
+      "step": 1752
+    },
+    {
+      "epoch": 0.1803143943285729,
+      "grad_norm": 0.13130852580070496,
+      "learning_rate": 0.007210353327855383,
+      "loss": 2.6509,
+      "step": 1755
+    },
+    {
+      "epoch": 0.1806226240624679,
+      "grad_norm": 0.14865098893642426,
+      "learning_rate": 0.007222678718159409,
+      "loss": 2.6253,
+      "step": 1758
+    },
+    {
+      "epoch": 0.1809308537963629,
+      "grad_norm": 0.20482710003852844,
+      "learning_rate": 0.007235004108463434,
+      "loss": 2.6312,
+      "step": 1761
+    },
+    {
+      "epoch": 0.1812390835302579,
+      "grad_norm": 0.12063097953796387,
+      "learning_rate": 0.007247329498767461,
+      "loss": 2.6007,
+      "step": 1764
+    },
+    {
+      "epoch": 0.18154731326415288,
+      "grad_norm": 0.23084934055805206,
+      "learning_rate": 0.007259654889071487,
+      "loss": 2.6129,
+      "step": 1767
+    },
+    {
+      "epoch": 0.18185554299804788,
+      "grad_norm": 0.10387217253446579,
+      "learning_rate": 0.007271980279375514,
+      "loss": 2.6309,
+      "step": 1770
+    },
+    {
+      "epoch": 0.18216377273194287,
+      "grad_norm": 0.14229682087898254,
+      "learning_rate": 0.00728430566967954,
+      "loss": 2.6074,
+      "step": 1773
+    },
+    {
+      "epoch": 0.18247200246583786,
+      "grad_norm": 0.12009115517139435,
+      "learning_rate": 0.007296631059983566,
+      "loss": 2.6407,
+      "step": 1776
+    },
+    {
+      "epoch": 0.18278023219973286,
+      "grad_norm": 0.15677185356616974,
+      "learning_rate": 0.007308956450287593,
+      "loss": 2.6268,
+      "step": 1779
+    },
+    {
+      "epoch": 0.18308846193362788,
+      "grad_norm": 0.13304303586483002,
+      "learning_rate": 0.0073212818405916195,
+      "loss": 2.6463,
+      "step": 1782
+    },
+    {
+      "epoch": 0.18339669166752287,
+      "grad_norm": 0.15444768965244293,
+      "learning_rate": 0.007333607230895644,
+      "loss": 2.6218,
+      "step": 1785
+    },
+    {
+      "epoch": 0.18370492140141786,
+      "grad_norm": 0.1738140732049942,
+      "learning_rate": 0.007345932621199671,
+      "loss": 2.6525,
+      "step": 1788
+    },
+    {
+      "epoch": 0.18401315113531286,
+      "grad_norm": 0.13087227940559387,
+      "learning_rate": 0.007358258011503698,
+      "loss": 2.6266,
+      "step": 1791
+    },
+    {
+      "epoch": 0.18432138086920785,
+      "grad_norm": 0.1026511862874031,
+      "learning_rate": 0.007370583401807724,
+      "loss": 2.6017,
+      "step": 1794
+    },
+    {
+      "epoch": 0.18462961060310285,
+      "grad_norm": 0.11183813214302063,
+      "learning_rate": 0.00738290879211175,
+      "loss": 2.5966,
+      "step": 1797
+    },
+    {
+      "epoch": 0.18493784033699784,
+      "grad_norm": 0.12239934504032135,
+      "learning_rate": 0.007395234182415777,
+      "loss": 2.6205,
+      "step": 1800
+    },
+    {
+      "epoch": 0.18524607007089283,
+      "grad_norm": 0.2630854845046997,
+      "learning_rate": 0.007407559572719803,
+      "loss": 2.609,
+      "step": 1803
+    },
+    {
+      "epoch": 0.18555429980478783,
+      "grad_norm": 0.24282613396644592,
+      "learning_rate": 0.00741988496302383,
+      "loss": 2.6405,
+      "step": 1806
+    },
+    {
+      "epoch": 0.18586252953868282,
+      "grad_norm": 0.2825084328651428,
+      "learning_rate": 0.007432210353327855,
+      "loss": 2.5933,
+      "step": 1809
+    },
+    {
+      "epoch": 0.18617075927257784,
+      "grad_norm": 0.26462721824645996,
+      "learning_rate": 0.007444535743631881,
+      "loss": 2.6021,
+      "step": 1812
+    },
+    {
+      "epoch": 0.18647898900647283,
+      "grad_norm": 0.11797992140054703,
+      "learning_rate": 0.007456861133935908,
+      "loss": 2.6246,
+      "step": 1815
+    },
+    {
+      "epoch": 0.18678721874036783,
+      "grad_norm": 0.14044708013534546,
+      "learning_rate": 0.0074691865242399345,
+      "loss": 2.6028,
+      "step": 1818
+    },
+    {
+      "epoch": 0.18709544847426282,
+      "grad_norm": 0.1374548226594925,
+      "learning_rate": 0.00748151191454396,
+      "loss": 2.6092,
+      "step": 1821
+    },
+    {
+      "epoch": 0.18740367820815781,
+      "grad_norm": 0.10084279626607895,
+      "learning_rate": 0.007493837304847987,
+      "loss": 2.6162,
+      "step": 1824
+    },
+    {
+      "epoch": 0.1877119079420528,
+      "grad_norm": 0.1052001565694809,
+      "learning_rate": 0.0075061626951520135,
+      "loss": 2.5742,
+      "step": 1827
+    },
+    {
+      "epoch": 0.1880201376759478,
+      "grad_norm": 0.11738535761833191,
+      "learning_rate": 0.00751848808545604,
+      "loss": 2.5715,
+      "step": 1830
+    },
+    {
+      "epoch": 0.1883283674098428,
+      "grad_norm": 0.10453224182128906,
+      "learning_rate": 0.007530813475760066,
+      "loss": 2.5896,
+      "step": 1833
+    },
+    {
+      "epoch": 0.1886365971437378,
+      "grad_norm": 0.10509374737739563,
+      "learning_rate": 0.007543138866064092,
+      "loss": 2.6047,
+      "step": 1836
+    },
+    {
+      "epoch": 0.18894482687763278,
+      "grad_norm": 0.11291799694299698,
+      "learning_rate": 0.007555464256368119,
+      "loss": 2.6062,
+      "step": 1839
+    },
+    {
+      "epoch": 0.1892530566115278,
+      "grad_norm": 0.11998583376407623,
+      "learning_rate": 0.007567789646672146,
+      "loss": 2.629,
+      "step": 1842
+    },
+    {
+      "epoch": 0.1895612863454228,
+      "grad_norm": 0.21776226162910461,
+      "learning_rate": 0.0075801150369761705,
+      "loss": 2.5847,
+      "step": 1845
+    },
+    {
+      "epoch": 0.1898695160793178,
+      "grad_norm": 0.210985466837883,
+      "learning_rate": 0.007592440427280197,
+      "loss": 2.5901,
+      "step": 1848
+    },
+    {
+      "epoch": 0.19017774581321278,
+      "grad_norm": 0.11799308657646179,
+      "learning_rate": 0.007604765817584224,
+      "loss": 2.5893,
+      "step": 1851
+    },
+    {
+      "epoch": 0.19048597554710778,
+      "grad_norm": 0.10019934922456741,
+      "learning_rate": 0.00761709120788825,
+      "loss": 2.6327,
+      "step": 1854
+    },
+    {
+      "epoch": 0.19079420528100277,
+      "grad_norm": 0.07964596897363663,
+      "learning_rate": 0.007629416598192276,
+      "loss": 2.5921,
+      "step": 1857
+    },
+    {
+      "epoch": 0.19110243501489776,
+      "grad_norm": 0.16393065452575684,
+      "learning_rate": 0.007641741988496303,
+      "loss": 2.5912,
+      "step": 1860
+    },
+    {
+      "epoch": 0.19141066474879276,
+      "grad_norm": 0.324639230966568,
+      "learning_rate": 0.007654067378800329,
+      "loss": 2.5998,
+      "step": 1863
+    },
+    {
+      "epoch": 0.19171889448268775,
+      "grad_norm": 0.14071421325206757,
+      "learning_rate": 0.007666392769104356,
+      "loss": 2.5803,
+      "step": 1866
+    },
+    {
+      "epoch": 0.19202712421658277,
+      "grad_norm": 0.20063026249408722,
+      "learning_rate": 0.007678718159408381,
+      "loss": 2.6019,
+      "step": 1869
+    },
+    {
+      "epoch": 0.19233535395047777,
+      "grad_norm": 0.11311519891023636,
+      "learning_rate": 0.007691043549712407,
+      "loss": 2.5645,
+      "step": 1872
+    },
+    {
+      "epoch": 0.19264358368437276,
+      "grad_norm": 0.08542342483997345,
+      "learning_rate": 0.007703368940016434,
+      "loss": 2.6122,
+      "step": 1875
+    },
+    {
+      "epoch": 0.19295181341826775,
+      "grad_norm": 0.08306868374347687,
+      "learning_rate": 0.007715694330320461,
+      "loss": 2.5859,
+      "step": 1878
+    },
+    {
+      "epoch": 0.19326004315216275,
+      "grad_norm": 0.11635984480381012,
+      "learning_rate": 0.007728019720624486,
+      "loss": 2.5855,
+      "step": 1881
+    },
+    {
+      "epoch": 0.19356827288605774,
+      "grad_norm": 0.08945252746343613,
+      "learning_rate": 0.007740345110928513,
+      "loss": 2.5509,
+      "step": 1884
+    },
+    {
+      "epoch": 0.19387650261995273,
+      "grad_norm": 0.19044962525367737,
+      "learning_rate": 0.0077526705012325395,
+      "loss": 2.559,
+      "step": 1887
+    },
+    {
+      "epoch": 0.19418473235384773,
+      "grad_norm": 0.1462780088186264,
+      "learning_rate": 0.007764995891536566,
+      "loss": 2.5749,
+      "step": 1890
+    },
+    {
+      "epoch": 0.19449296208774272,
+      "grad_norm": 0.15944691002368927,
+      "learning_rate": 0.007777321281840591,
+      "loss": 2.5801,
+      "step": 1893
+    },
+    {
+      "epoch": 0.19480119182163771,
+      "grad_norm": 0.10125305503606796,
+      "learning_rate": 0.007789646672144618,
+      "loss": 2.5821,
+      "step": 1896
+    },
+    {
+      "epoch": 0.19510942155553274,
+      "grad_norm": 0.17344938218593597,
+      "learning_rate": 0.007801972062448644,
+      "loss": 2.5905,
+      "step": 1899
+    },
+    {
+      "epoch": 0.19541765128942773,
+      "grad_norm": 0.16651591658592224,
+      "learning_rate": 0.007814297452752672,
+      "loss": 2.5668,
+      "step": 1902
+    },
+    {
+      "epoch": 0.19572588102332272,
+      "grad_norm": 0.17417702078819275,
+      "learning_rate": 0.007826622843056696,
+      "loss": 2.568,
+      "step": 1905
+    },
+    {
+      "epoch": 0.19603411075721772,
+      "grad_norm": 0.11182334274053574,
+      "learning_rate": 0.007838948233360723,
+      "loss": 2.5547,
+      "step": 1908
+    },
+    {
+      "epoch": 0.1963423404911127,
+      "grad_norm": 0.23256631195545197,
+      "learning_rate": 0.007851273623664749,
+      "loss": 2.5722,
+      "step": 1911
+    },
+    {
+      "epoch": 0.1966505702250077,
+      "grad_norm": 0.18180392682552338,
+      "learning_rate": 0.007863599013968776,
+      "loss": 2.558,
+      "step": 1914
+    },
+    {
+      "epoch": 0.1969587999589027,
+      "grad_norm": 0.12168890237808228,
+      "learning_rate": 0.007875924404272802,
+      "loss": 2.5977,
+      "step": 1917
+    },
+    {
+      "epoch": 0.1972670296927977,
+      "grad_norm": 0.11032187938690186,
+      "learning_rate": 0.007888249794576828,
+      "loss": 2.5846,
+      "step": 1920
+    },
+    {
+      "epoch": 0.19757525942669268,
+      "grad_norm": 0.0740116760134697,
+      "learning_rate": 0.007900575184880855,
+      "loss": 2.5824,
+      "step": 1923
+    },
+    {
+      "epoch": 0.19788348916058768,
+      "grad_norm": 0.05902474746108055,
+      "learning_rate": 0.007912900575184881,
+      "loss": 2.5497,
+      "step": 1926
+    },
+    {
+      "epoch": 0.1981917188944827,
+      "grad_norm": 0.09003309905529022,
+      "learning_rate": 0.007925225965488907,
+      "loss": 2.5523,
+      "step": 1929
+    },
+    {
+      "epoch": 0.1984999486283777,
+      "grad_norm": 0.4191035330295563,
+      "learning_rate": 0.007937551355792934,
+      "loss": 2.6223,
+      "step": 1932
+    },
+    {
+      "epoch": 0.1988081783622727,
+      "grad_norm": 0.17093214392662048,
+      "learning_rate": 0.00794987674609696,
+      "loss": 2.5647,
+      "step": 1935
+    },
+    {
+      "epoch": 0.19911640809616768,
+      "grad_norm": 0.0921127051115036,
+      "learning_rate": 0.007962202136400986,
+      "loss": 2.564,
+      "step": 1938
+    },
+    {
+      "epoch": 0.19942463783006267,
+      "grad_norm": 0.14204134047031403,
+      "learning_rate": 0.007974527526705012,
+      "loss": 2.5972,
+      "step": 1941
+    },
+    {
+      "epoch": 0.19973286756395767,
+      "grad_norm": 0.07556895911693573,
+      "learning_rate": 0.007986852917009039,
+      "loss": 2.5796,
+      "step": 1944
+    },
+    {
+      "epoch": 0.20004109729785266,
+      "grad_norm": 0.07290320843458176,
+      "learning_rate": 0.007999178307313065,
+      "loss": 2.5564,
+      "step": 1947
+    },
+    {
+      "epoch": 0.20034932703174765,
+      "grad_norm": 0.1624913364648819,
+      "learning_rate": 0.008011503697617092,
+      "loss": 2.5849,
+      "step": 1950
+    },
+    {
+      "epoch": 0.20065755676564265,
+      "grad_norm": 0.11839967221021652,
+      "learning_rate": 0.008023829087921118,
+      "loss": 2.5611,
+      "step": 1953
+    },
+    {
+      "epoch": 0.20096578649953767,
+      "grad_norm": 0.14280788600444794,
+      "learning_rate": 0.008036154478225144,
+      "loss": 2.5289,
+      "step": 1956
+    },
+    {
+      "epoch": 0.20127401623343266,
+      "grad_norm": 0.11515247821807861,
+      "learning_rate": 0.008048479868529171,
+      "loss": 2.5678,
+      "step": 1959
+    },
+    {
+      "epoch": 0.20158224596732766,
+      "grad_norm": 0.1147715225815773,
+      "learning_rate": 0.008060805258833197,
+      "loss": 2.5452,
+      "step": 1962
+    },
+    {
+      "epoch": 0.20189047570122265,
+      "grad_norm": 0.09767001122236252,
+      "learning_rate": 0.008073130649137223,
+      "loss": 2.6023,
+      "step": 1965
+    },
+    {
+      "epoch": 0.20219870543511764,
+      "grad_norm": 0.0866391509771347,
+      "learning_rate": 0.008085456039441248,
+      "loss": 2.5518,
+      "step": 1968
+    },
+    {
+      "epoch": 0.20250693516901264,
+      "grad_norm": 0.1610632985830307,
+      "learning_rate": 0.008097781429745276,
+      "loss": 2.5271,
+      "step": 1971
+    },
+    {
+      "epoch": 0.20281516490290763,
+      "grad_norm": 0.20238341391086578,
+      "learning_rate": 0.008110106820049302,
+      "loss": 2.5597,
+      "step": 1974
+    },
+    {
+      "epoch": 0.20312339463680262,
+      "grad_norm": 0.11807162314653397,
+      "learning_rate": 0.008122432210353327,
+      "loss": 2.5663,
+      "step": 1977
+    },
+    {
+      "epoch": 0.20343162437069762,
+      "grad_norm": 0.14654900133609772,
+      "learning_rate": 0.008134757600657355,
+      "loss": 2.5729,
+      "step": 1980
+    },
+    {
+      "epoch": 0.2037398541045926,
+      "grad_norm": 0.17804567515850067,
+      "learning_rate": 0.00814708299096138,
+      "loss": 2.5658,
+      "step": 1983
+    },
+    {
+      "epoch": 0.20404808383848763,
+      "grad_norm": 0.12376303225755692,
+      "learning_rate": 0.008159408381265408,
+      "loss": 2.5703,
+      "step": 1986
+    },
+    {
+      "epoch": 0.20435631357238263,
+      "grad_norm": 0.1248418316245079,
+      "learning_rate": 0.008171733771569432,
+      "loss": 2.5328,
+      "step": 1989
+    },
+    {
+      "epoch": 0.20466454330627762,
+      "grad_norm": 0.08159278333187103,
+      "learning_rate": 0.00818405916187346,
+      "loss": 2.5349,
+      "step": 1992
+    },
+    {
+      "epoch": 0.2049727730401726,
+      "grad_norm": 0.11184779554605484,
+      "learning_rate": 0.008196384552177485,
+      "loss": 2.5557,
+      "step": 1995
+    },
+    {
+      "epoch": 0.2052810027740676,
+      "grad_norm": 0.09568610787391663,
+      "learning_rate": 0.008208709942481513,
+      "loss": 2.5415,
+      "step": 1998
+    },
+    {
+      "epoch": 0.2055892325079626,
+      "grad_norm": 0.08708583563566208,
+      "learning_rate": 0.008221035332785539,
+      "loss": 2.5369,
+      "step": 2001
+    },
+    {
+      "epoch": 0.2058974622418576,
+      "grad_norm": 0.11849135160446167,
+      "learning_rate": 0.008233360723089564,
+      "loss": 2.5617,
+      "step": 2004
+    },
+    {
+      "epoch": 0.2062056919757526,
+      "grad_norm": 0.1407340168952942,
+      "learning_rate": 0.008245686113393592,
+      "loss": 2.5374,
+      "step": 2007
+    },
+    {
+      "epoch": 0.20651392170964758,
+      "grad_norm": 0.13198955357074738,
+      "learning_rate": 0.008258011503697617,
+      "loss": 2.57,
+      "step": 2010
+    },
+    {
+      "epoch": 0.20682215144354257,
+      "grad_norm": 0.12408044934272766,
+      "learning_rate": 0.008270336894001643,
+      "loss": 2.5344,
+      "step": 2013
+    },
+    {
+      "epoch": 0.2071303811774376,
+      "grad_norm": 0.149169921875,
+      "learning_rate": 0.008282662284305669,
+      "loss": 2.5357,
+      "step": 2016
+    },
+    {
+      "epoch": 0.2074386109113326,
+      "grad_norm": 0.10010293871164322,
+      "learning_rate": 0.008294987674609696,
+      "loss": 2.5166,
+      "step": 2019
+    },
+    {
+      "epoch": 0.20774684064522758,
+      "grad_norm": 0.17650344967842102,
+      "learning_rate": 0.008307313064913722,
+      "loss": 2.5664,
+      "step": 2022
+    },
+    {
+      "epoch": 0.20805507037912258,
+      "grad_norm": 0.09946206212043762,
+      "learning_rate": 0.008319638455217748,
+      "loss": 2.5378,
+      "step": 2025
+    },
+    {
+      "epoch": 0.20836330011301757,
+      "grad_norm": 0.07705225795507431,
+      "learning_rate": 0.008331963845521775,
+      "loss": 2.5088,
+      "step": 2028
+    },
+    {
+      "epoch": 0.20867152984691256,
+      "grad_norm": 0.18174925446510315,
+      "learning_rate": 0.008344289235825801,
+      "loss": 2.5264,
+      "step": 2031
+    },
+    {
+      "epoch": 0.20897975958080756,
+      "grad_norm": 0.14415894448757172,
+      "learning_rate": 0.008356614626129829,
+      "loss": 2.5549,
+      "step": 2034
+    },
+    {
+      "epoch": 0.20928798931470255,
+      "grad_norm": 0.17721933126449585,
+      "learning_rate": 0.008368940016433854,
+      "loss": 2.5476,
+      "step": 2037
+    },
+    {
+      "epoch": 0.20959621904859754,
+      "grad_norm": 0.1727544367313385,
+      "learning_rate": 0.00838126540673788,
+      "loss": 2.5809,
+      "step": 2040
+    },
+    {
+      "epoch": 0.20990444878249256,
+      "grad_norm": 0.20624054968357086,
+      "learning_rate": 0.008393590797041908,
+      "loss": 2.5256,
+      "step": 2043
+    },
+    {
+      "epoch": 0.21021267851638756,
+      "grad_norm": 0.08070924133062363,
+      "learning_rate": 0.008405916187345933,
+      "loss": 2.5537,
+      "step": 2046
+    },
+    {
+      "epoch": 0.21052090825028255,
+      "grad_norm": 0.07868220657110214,
+      "learning_rate": 0.008418241577649959,
+      "loss": 2.5266,
+      "step": 2049
+    },
+    {
+      "epoch": 0.21082913798417754,
+      "grad_norm": 0.19941876828670502,
+      "learning_rate": 0.008430566967953985,
+      "loss": 2.5344,
+      "step": 2052
+    },
+    {
+      "epoch": 0.21113736771807254,
+      "grad_norm": 0.08758697658777237,
+      "learning_rate": 0.008442892358258012,
+      "loss": 2.5409,
+      "step": 2055
+    },
+    {
+      "epoch": 0.21144559745196753,
+      "grad_norm": 0.11635969579219818,
+      "learning_rate": 0.008455217748562038,
+      "loss": 2.5497,
+      "step": 2058
+    },
+    {
+      "epoch": 0.21175382718586253,
+      "grad_norm": 0.16910326480865479,
+      "learning_rate": 0.008467543138866064,
+      "loss": 2.5509,
+      "step": 2061
+    },
+    {
+      "epoch": 0.21206205691975752,
+      "grad_norm": 0.14605827629566193,
+      "learning_rate": 0.008479868529170091,
+      "loss": 2.5589,
+      "step": 2064
+    },
+    {
+      "epoch": 0.2123702866536525,
+      "grad_norm": 0.18890123069286346,
+      "learning_rate": 0.008492193919474117,
+      "loss": 2.5454,
+      "step": 2067
+    },
+    {
+      "epoch": 0.2126785163875475,
+      "grad_norm": 0.09277717024087906,
+      "learning_rate": 0.008504519309778144,
+      "loss": 2.4984,
+      "step": 2070
+    },
+    {
+      "epoch": 0.21298674612144253,
+      "grad_norm": 0.07268327474594116,
+      "learning_rate": 0.008516844700082168,
+      "loss": 2.5323,
+      "step": 2073
+    },
+    {
+      "epoch": 0.21329497585533752,
+      "grad_norm": 0.0807403028011322,
+      "learning_rate": 0.008529170090386196,
+      "loss": 2.5083,
+      "step": 2076
+    },
+    {
+      "epoch": 0.21360320558923251,
+      "grad_norm": 0.12681947648525238,
+      "learning_rate": 0.008541495480690222,
+      "loss": 2.5386,
+      "step": 2079
+    },
+    {
+      "epoch": 0.2139114353231275,
+      "grad_norm": 0.25378334522247314,
+      "learning_rate": 0.008553820870994249,
+      "loss": 2.5188,
+      "step": 2082
+    },
+    {
+      "epoch": 0.2142196650570225,
+      "grad_norm": 0.15101733803749084,
+      "learning_rate": 0.008566146261298275,
+      "loss": 2.5457,
+      "step": 2085
+    },
+    {
+      "epoch": 0.2145278947909175,
+      "grad_norm": 0.17336703836917877,
+      "learning_rate": 0.0085784716516023,
+      "loss": 2.5206,
+      "step": 2088
+    },
+    {
+      "epoch": 0.2148361245248125,
+      "grad_norm": 0.07735245674848557,
+      "learning_rate": 0.008590797041906328,
+      "loss": 2.5297,
+      "step": 2091
+    },
+    {
+      "epoch": 0.21514435425870748,
+      "grad_norm": 0.15841136872768402,
+      "learning_rate": 0.008603122432210354,
+      "loss": 2.5086,
+      "step": 2094
+    },
+    {
+      "epoch": 0.21545258399260248,
+      "grad_norm": 0.15941859781742096,
+      "learning_rate": 0.00861544782251438,
+      "loss": 2.5316,
+      "step": 2097
+    },
+    {
+      "epoch": 0.21576081372649747,
+      "grad_norm": 0.13837756216526031,
+      "learning_rate": 0.008627773212818405,
+      "loss": 2.4818,
+      "step": 2100
+    },
+    {
+      "epoch": 0.2160690434603925,
+      "grad_norm": 0.14743675291538239,
+      "learning_rate": 0.008640098603122433,
+      "loss": 2.5351,
+      "step": 2103
+    },
+    {
+      "epoch": 0.21637727319428748,
+      "grad_norm": 0.15961112082004547,
+      "learning_rate": 0.008652423993426459,
+      "loss": 2.4916,
+      "step": 2106
+    },
+    {
+      "epoch": 0.21668550292818248,
+      "grad_norm": 0.16091223061084747,
+      "learning_rate": 0.008664749383730484,
+      "loss": 2.5026,
+      "step": 2109
+    },
+    {
+      "epoch": 0.21699373266207747,
+      "grad_norm": 0.1695915311574936,
+      "learning_rate": 0.008677074774034512,
+      "loss": 2.4994,
+      "step": 2112
+    },
+    {
+      "epoch": 0.21730196239597246,
+      "grad_norm": 0.1457175761461258,
+      "learning_rate": 0.008689400164338537,
+      "loss": 2.5225,
+      "step": 2115
+    },
+    {
+      "epoch": 0.21761019212986746,
+      "grad_norm": 0.0995342880487442,
+      "learning_rate": 0.008701725554642563,
+      "loss": 2.5373,
+      "step": 2118
+    },
+    {
+      "epoch": 0.21791842186376245,
+      "grad_norm": 0.11527393013238907,
+      "learning_rate": 0.00871405094494659,
+      "loss": 2.5207,
+      "step": 2121
+    },
+    {
+      "epoch": 0.21822665159765745,
+      "grad_norm": 0.07951527088880539,
+      "learning_rate": 0.008726376335250616,
+      "loss": 2.4868,
+      "step": 2124
+    },
+    {
+      "epoch": 0.21853488133155244,
+      "grad_norm": 0.11319970339536667,
+      "learning_rate": 0.008738701725554644,
+      "loss": 2.4965,
+      "step": 2127
+    },
+    {
+      "epoch": 0.21884311106544746,
+      "grad_norm": 0.14932893216609955,
+      "learning_rate": 0.008751027115858668,
+      "loss": 2.5164,
+      "step": 2130
+    },
+    {
+      "epoch": 0.21915134079934245,
+      "grad_norm": 0.1703396886587143,
+      "learning_rate": 0.008763352506162695,
+      "loss": 2.5175,
+      "step": 2133
+    },
+    {
+      "epoch": 0.21945957053323745,
+      "grad_norm": 0.2208787351846695,
+      "learning_rate": 0.008775677896466721,
+      "loss": 2.521,
+      "step": 2136
+    },
+    {
+      "epoch": 0.21976780026713244,
+      "grad_norm": 0.0884699895977974,
+      "learning_rate": 0.008788003286770749,
+      "loss": 2.5356,
+      "step": 2139
+    },
+    {
+      "epoch": 0.22007603000102743,
+      "grad_norm": 0.06739311665296555,
+      "learning_rate": 0.008800328677074774,
+      "loss": 2.5102,
+      "step": 2142
+    },
+    {
+      "epoch": 0.22038425973492243,
+      "grad_norm": 0.09653139859437943,
+      "learning_rate": 0.0088126540673788,
+      "loss": 2.5047,
+      "step": 2145
+    },
+    {
+      "epoch": 0.22069248946881742,
+      "grad_norm": 0.11972832679748535,
+      "learning_rate": 0.008824979457682828,
+      "loss": 2.5086,
+      "step": 2148
+    },
+    {
+      "epoch": 0.22100071920271241,
+      "grad_norm": 0.13725396990776062,
+      "learning_rate": 0.008837304847986853,
+      "loss": 2.5034,
+      "step": 2151
+    },
+    {
+      "epoch": 0.2213089489366074,
+      "grad_norm": 0.09293966740369797,
+      "learning_rate": 0.008849630238290879,
+      "loss": 2.5004,
+      "step": 2154
+    },
+    {
+      "epoch": 0.2216171786705024,
+      "grad_norm": 0.07625159621238708,
+      "learning_rate": 0.008861955628594905,
+      "loss": 2.508,
+      "step": 2157
+    },
+    {
+      "epoch": 0.22192540840439742,
+      "grad_norm": 0.08581928163766861,
+      "learning_rate": 0.008874281018898932,
+      "loss": 2.4973,
+      "step": 2160
+    },
+    {
+      "epoch": 0.22223363813829242,
+      "grad_norm": 0.12700457870960236,
+      "learning_rate": 0.008886606409202958,
+      "loss": 2.5174,
+      "step": 2163
+    },
+    {
+      "epoch": 0.2225418678721874,
+      "grad_norm": 0.17155064642429352,
+      "learning_rate": 0.008898931799506984,
+      "loss": 2.4969,
+      "step": 2166
+    },
+    {
+      "epoch": 0.2228500976060824,
+      "grad_norm": 0.13356278836727142,
+      "learning_rate": 0.008911257189811011,
+      "loss": 2.4876,
+      "step": 2169
+    },
+    {
+      "epoch": 0.2231583273399774,
+      "grad_norm": 0.07805536687374115,
+      "learning_rate": 0.008923582580115037,
+      "loss": 2.5151,
+      "step": 2172
+    },
+    {
+      "epoch": 0.2234665570738724,
+      "grad_norm": 0.10661714524030685,
+      "learning_rate": 0.008935907970419064,
+      "loss": 2.4607,
+      "step": 2175
+    },
+    {
+      "epoch": 0.22377478680776738,
+      "grad_norm": 0.15095242857933044,
+      "learning_rate": 0.008948233360723088,
+      "loss": 2.5358,
+      "step": 2178
+    },
+    {
+      "epoch": 0.22408301654166238,
+      "grad_norm": 0.11287077516317368,
+      "learning_rate": 0.008960558751027116,
+      "loss": 2.5289,
+      "step": 2181
+    },
+    {
+      "epoch": 0.22439124627555737,
+      "grad_norm": 0.16408318281173706,
+      "learning_rate": 0.008972884141331142,
+      "loss": 2.5256,
+      "step": 2184
+    },
+    {
+      "epoch": 0.22469947600945236,
+      "grad_norm": 0.1227622851729393,
+      "learning_rate": 0.008985209531635169,
+      "loss": 2.5091,
+      "step": 2187
+    },
+    {
+      "epoch": 0.22500770574334739,
+      "grad_norm": 0.06549924612045288,
+      "learning_rate": 0.008997534921939195,
+      "loss": 2.4908,
+      "step": 2190
+    },
+    {
+      "epoch": 0.22531593547724238,
+      "grad_norm": 0.09310626983642578,
+      "learning_rate": 0.00900986031224322,
+      "loss": 2.4903,
+      "step": 2193
+    },
+    {
+      "epoch": 0.22562416521113737,
+      "grad_norm": 0.12637357413768768,
+      "learning_rate": 0.009022185702547248,
+      "loss": 2.5089,
+      "step": 2196
+    },
+    {
+      "epoch": 0.22593239494503237,
+      "grad_norm": 0.1691301167011261,
+      "learning_rate": 0.009034511092851274,
+      "loss": 2.4984,
+      "step": 2199
+    },
+    {
+      "epoch": 0.22624062467892736,
+      "grad_norm": 0.18173068761825562,
+      "learning_rate": 0.0090468364831553,
+      "loss": 2.4552,
+      "step": 2202
+    },
+    {
+      "epoch": 0.22654885441282235,
+      "grad_norm": 0.19549600780010223,
+      "learning_rate": 0.009059161873459327,
+      "loss": 2.4642,
+      "step": 2205
+    },
+    {
+      "epoch": 0.22685708414671735,
+      "grad_norm": 0.09038446098566055,
+      "learning_rate": 0.009071487263763353,
+      "loss": 2.5017,
+      "step": 2208
+    },
+    {
+      "epoch": 0.22716531388061234,
+      "grad_norm": 0.07959726452827454,
+      "learning_rate": 0.009083812654067379,
+      "loss": 2.4934,
+      "step": 2211
+    },
+    {
+      "epoch": 0.22747354361450733,
+      "grad_norm": 0.07991699874401093,
+      "learning_rate": 0.009096138044371404,
+      "loss": 2.498,
+      "step": 2214
+    },
+    {
+      "epoch": 0.22778177334840236,
+      "grad_norm": 0.09022307395935059,
+      "learning_rate": 0.009108463434675432,
+      "loss": 2.4832,
+      "step": 2217
+    },
+    {
+      "epoch": 0.22809000308229735,
+      "grad_norm": 0.11399543285369873,
+      "learning_rate": 0.009120788824979457,
+      "loss": 2.4929,
+      "step": 2220
+    },
+    {
+      "epoch": 0.22839823281619234,
+      "grad_norm": 0.10349836200475693,
+      "learning_rate": 0.009133114215283485,
+      "loss": 2.4622,
+      "step": 2223
+    },
+    {
+      "epoch": 0.22870646255008734,
+      "grad_norm": 0.17096632719039917,
+      "learning_rate": 0.00914543960558751,
+      "loss": 2.5103,
+      "step": 2226
+    },
+    {
+      "epoch": 0.22901469228398233,
+      "grad_norm": 0.13803228735923767,
+      "learning_rate": 0.009157764995891536,
+      "loss": 2.5034,
+      "step": 2229
+    },
+    {
+      "epoch": 0.22932292201787732,
+      "grad_norm": 0.16332487761974335,
+      "learning_rate": 0.009170090386195564,
+      "loss": 2.5051,
+      "step": 2232
+    },
+    {
+      "epoch": 0.22963115175177232,
+      "grad_norm": 0.12147244811058044,
+      "learning_rate": 0.00918241577649959,
+      "loss": 2.4794,
+      "step": 2235
+    },
+    {
+      "epoch": 0.2299393814856673,
+      "grad_norm": 0.08943907916545868,
+      "learning_rate": 0.009194741166803615,
+      "loss": 2.5331,
+      "step": 2238
+    },
+    {
+      "epoch": 0.2302476112195623,
+      "grad_norm": 0.08069117367267609,
+      "learning_rate": 0.009207066557107641,
+      "loss": 2.4807,
+      "step": 2241
+    },
+    {
+      "epoch": 0.2305558409534573,
+      "grad_norm": 0.11125557869672775,
+      "learning_rate": 0.009219391947411669,
+      "loss": 2.4567,
+      "step": 2244
+    },
+    {
+      "epoch": 0.23086407068735232,
+      "grad_norm": 0.2825096547603607,
+      "learning_rate": 0.009231717337715694,
+      "loss": 2.5101,
+      "step": 2247
+    },
+    {
+      "epoch": 0.2311723004212473,
+      "grad_norm": 0.10534384101629257,
+      "learning_rate": 0.00924404272801972,
+      "loss": 2.5272,
+      "step": 2250
+    },
+    {
+      "epoch": 0.2314805301551423,
+      "grad_norm": 0.07159514725208282,
+      "learning_rate": 0.009256368118323748,
+      "loss": 2.4707,
+      "step": 2253
+    },
+    {
+      "epoch": 0.2317887598890373,
+      "grad_norm": 0.06435802578926086,
+      "learning_rate": 0.009268693508627773,
+      "loss": 2.4788,
+      "step": 2256
+    },
+    {
+      "epoch": 0.2320969896229323,
+      "grad_norm": 0.09402693063020706,
+      "learning_rate": 0.0092810188989318,
+      "loss": 2.4639,
+      "step": 2259
+    },
+    {
+      "epoch": 0.2324052193568273,
+      "grad_norm": 0.18836408853530884,
+      "learning_rate": 0.009293344289235825,
+      "loss": 2.4747,
+      "step": 2262
+    },
+    {
+      "epoch": 0.23271344909072228,
+      "grad_norm": 0.09705471992492676,
+      "learning_rate": 0.009305669679539852,
+      "loss": 2.5041,
+      "step": 2265
+    },
+    {
+      "epoch": 0.23302167882461727,
+      "grad_norm": 0.09185091406106949,
+      "learning_rate": 0.009317995069843878,
+      "loss": 2.4625,
+      "step": 2268
+    },
+    {
+      "epoch": 0.23332990855851227,
+      "grad_norm": 0.0848812386393547,
+      "learning_rate": 0.009330320460147905,
+      "loss": 2.4876,
+      "step": 2271
+    },
+    {
+      "epoch": 0.2336381382924073,
+      "grad_norm": 0.07989475131034851,
+      "learning_rate": 0.009342645850451931,
+      "loss": 2.4697,
+      "step": 2274
+    },
+    {
+      "epoch": 0.23394636802630228,
+      "grad_norm": 0.09660454094409943,
+      "learning_rate": 0.009354971240755957,
+      "loss": 2.4917,
+      "step": 2277
+    },
+    {
+      "epoch": 0.23425459776019728,
+      "grad_norm": 0.09550273418426514,
+      "learning_rate": 0.009367296631059984,
+      "loss": 2.4806,
+      "step": 2280
+    },
+    {
+      "epoch": 0.23456282749409227,
+      "grad_norm": 0.16650651395320892,
+      "learning_rate": 0.00937962202136401,
+      "loss": 2.4424,
+      "step": 2283
+    },
+    {
+      "epoch": 0.23487105722798726,
+      "grad_norm": 0.1455817073583603,
+      "learning_rate": 0.009391947411668036,
+      "loss": 2.4907,
+      "step": 2286
+    },
+    {
+      "epoch": 0.23517928696188226,
+      "grad_norm": 0.075865738093853,
+      "learning_rate": 0.009404272801972062,
+      "loss": 2.5004,
+      "step": 2289
+    },
+    {
+      "epoch": 0.23548751669577725,
+      "grad_norm": 0.188491553068161,
+      "learning_rate": 0.009416598192276089,
+      "loss": 2.5111,
+      "step": 2292
+    },
+    {
+      "epoch": 0.23579574642967224,
+      "grad_norm": 0.07567702233791351,
+      "learning_rate": 0.009428923582580115,
+      "loss": 2.4966,
+      "step": 2295
+    },
+    {
+      "epoch": 0.23610397616356724,
+      "grad_norm": 0.0682358667254448,
+      "learning_rate": 0.00944124897288414,
+      "loss": 2.4781,
+      "step": 2298
+    },
+    {
+      "epoch": 0.23641220589746223,
+      "grad_norm": 0.173895925283432,
+      "learning_rate": 0.009453574363188168,
+      "loss": 2.4471,
+      "step": 2301
+    },
+    {
+      "epoch": 0.23672043563135725,
+      "grad_norm": 0.15088587999343872,
+      "learning_rate": 0.009465899753492194,
+      "loss": 2.4783,
+      "step": 2304
+    },
+    {
+      "epoch": 0.23702866536525224,
+      "grad_norm": 0.09947361797094345,
+      "learning_rate": 0.009478225143796221,
+      "loss": 2.4757,
+      "step": 2307
+    },
+    {
+      "epoch": 0.23733689509914724,
+      "grad_norm": 0.0709480568766594,
+      "learning_rate": 0.009490550534100247,
+      "loss": 2.4617,
+      "step": 2310
+    },
+    {
+      "epoch": 0.23764512483304223,
+      "grad_norm": 0.11335324496030807,
+      "learning_rate": 0.009502875924404273,
+      "loss": 2.4506,
+      "step": 2313
+    },
+    {
+      "epoch": 0.23795335456693723,
+      "grad_norm": 0.10329569876194,
+      "learning_rate": 0.0095152013147083,
+      "loss": 2.4444,
+      "step": 2316
+    },
+    {
+      "epoch": 0.23826158430083222,
+      "grad_norm": 0.18935157358646393,
+      "learning_rate": 0.009527526705012326,
+      "loss": 2.4739,
+      "step": 2319
+    },
+    {
+      "epoch": 0.2385698140347272,
+      "grad_norm": 0.10977230221033096,
+      "learning_rate": 0.009539852095316352,
+      "loss": 2.4849,
+      "step": 2322
+    },
+    {
+      "epoch": 0.2388780437686222,
+      "grad_norm": 0.1623351126909256,
+      "learning_rate": 0.009552177485620377,
+      "loss": 2.4856,
+      "step": 2325
+    },
+    {
+      "epoch": 0.2391862735025172,
+      "grad_norm": 0.12067209929227829,
+      "learning_rate": 0.009564502875924405,
+      "loss": 2.427,
+      "step": 2328
+    },
+    {
+      "epoch": 0.2394945032364122,
+      "grad_norm": 0.12578649818897247,
+      "learning_rate": 0.00957682826622843,
+      "loss": 2.4719,
+      "step": 2331
+    },
+    {
+      "epoch": 0.23980273297030721,
+      "grad_norm": 0.09442924708127975,
+      "learning_rate": 0.009589153656532456,
+      "loss": 2.475,
+      "step": 2334
+    },
+    {
+      "epoch": 0.2401109627042022,
+      "grad_norm": 0.06693053990602493,
+      "learning_rate": 0.009601479046836484,
+      "loss": 2.4949,
+      "step": 2337
+    },
+    {
+      "epoch": 0.2404191924380972,
+      "grad_norm": 0.09371168911457062,
+      "learning_rate": 0.00961380443714051,
+      "loss": 2.4611,
+      "step": 2340
+    },
+    {
+      "epoch": 0.2407274221719922,
+      "grad_norm": 0.11009377986192703,
+      "learning_rate": 0.009626129827444537,
+      "loss": 2.4998,
+      "step": 2343
+    },
+    {
+      "epoch": 0.2410356519058872,
+      "grad_norm": 0.08789053559303284,
+      "learning_rate": 0.009638455217748561,
+      "loss": 2.4891,
+      "step": 2346
+    },
+    {
+      "epoch": 0.24134388163978218,
+      "grad_norm": 0.2513992488384247,
+      "learning_rate": 0.009650780608052589,
+      "loss": 2.4613,
+      "step": 2349
+    },
+    {
+      "epoch": 0.24165211137367718,
+      "grad_norm": 0.09223336726427078,
+      "learning_rate": 0.009663105998356614,
+      "loss": 2.4874,
+      "step": 2352
+    },
+    {
+      "epoch": 0.24196034110757217,
+      "grad_norm": 0.08941586315631866,
+      "learning_rate": 0.009675431388660642,
+      "loss": 2.4777,
+      "step": 2355
+    },
+    {
+      "epoch": 0.24226857084146716,
+      "grad_norm": 0.09664765000343323,
+      "learning_rate": 0.009687756778964668,
+      "loss": 2.4728,
+      "step": 2358
+    },
+    {
+      "epoch": 0.24257680057536218,
+      "grad_norm": 0.08079587668180466,
+      "learning_rate": 0.009700082169268693,
+      "loss": 2.4621,
+      "step": 2361
+    },
+    {
+      "epoch": 0.24288503030925718,
+      "grad_norm": 0.07663597911596298,
+      "learning_rate": 0.00971240755957272,
+      "loss": 2.487,
+      "step": 2364
+    },
+    {
+      "epoch": 0.24319326004315217,
+      "grad_norm": 0.07564109563827515,
+      "learning_rate": 0.009724732949876747,
+      "loss": 2.4123,
+      "step": 2367
+    },
+    {
+      "epoch": 0.24350148977704716,
+      "grad_norm": 0.1025756299495697,
+      "learning_rate": 0.009737058340180772,
+      "loss": 2.4669,
+      "step": 2370
+    },
+    {
+      "epoch": 0.24380971951094216,
+      "grad_norm": 0.1370251476764679,
+      "learning_rate": 0.009749383730484798,
+      "loss": 2.4664,
+      "step": 2373
+    },
+    {
+      "epoch": 0.24411794924483715,
+      "grad_norm": 0.11926325410604477,
+      "learning_rate": 0.009761709120788825,
+      "loss": 2.4483,
+      "step": 2376
+    },
+    {
+      "epoch": 0.24442617897873214,
+      "grad_norm": 0.16847510635852814,
+      "learning_rate": 0.009774034511092851,
+      "loss": 2.4421,
+      "step": 2379
+    },
+    {
+      "epoch": 0.24473440871262714,
+      "grad_norm": 0.14343461394309998,
+      "learning_rate": 0.009786359901396877,
+      "loss": 2.452,
+      "step": 2382
+    },
+    {
+      "epoch": 0.24504263844652213,
+      "grad_norm": 0.0658588707447052,
+      "learning_rate": 0.009798685291700904,
+      "loss": 2.4717,
+      "step": 2385
+    },
+    {
+      "epoch": 0.24535086818041713,
+      "grad_norm": 0.09394209086894989,
+      "learning_rate": 0.00981101068200493,
+      "loss": 2.4467,
+      "step": 2388
+    },
+    {
+      "epoch": 0.24565909791431215,
+      "grad_norm": 0.0717134177684784,
+      "learning_rate": 0.009823336072308958,
+      "loss": 2.4505,
+      "step": 2391
+    },
+    {
+      "epoch": 0.24596732764820714,
+      "grad_norm": 0.07518400996923447,
+      "learning_rate": 0.009835661462612983,
+      "loss": 2.431,
+      "step": 2394
+    },
+    {
+      "epoch": 0.24627555738210213,
+      "grad_norm": 0.10242413729429245,
+      "learning_rate": 0.00984798685291701,
+      "loss": 2.451,
+      "step": 2397
+    },
+    {
+      "epoch": 0.24658378711599713,
+      "grad_norm": 0.11668457090854645,
+      "learning_rate": 0.009860312243221035,
+      "loss": 2.4574,
+      "step": 2400
+    },
+    {
+      "epoch": 0.24689201684989212,
+      "grad_norm": 0.1074887290596962,
+      "learning_rate": 0.009872637633525062,
+      "loss": 2.4688,
+      "step": 2403
+    },
+    {
+      "epoch": 0.24720024658378711,
+      "grad_norm": 0.143118217587471,
+      "learning_rate": 0.009884963023829088,
+      "loss": 2.4614,
+      "step": 2406
+    },
+    {
+      "epoch": 0.2475084763176821,
+      "grad_norm": 0.08865509182214737,
+      "learning_rate": 0.009897288414133114,
+      "loss": 2.4768,
+      "step": 2409
+    },
+    {
+      "epoch": 0.2478167060515771,
+      "grad_norm": 0.10735021531581879,
+      "learning_rate": 0.009909613804437141,
+      "loss": 2.457,
+      "step": 2412
+    },
+    {
+      "epoch": 0.2481249357854721,
+      "grad_norm": 0.11766096949577332,
+      "learning_rate": 0.009921939194741167,
+      "loss": 2.4661,
+      "step": 2415
+    },
+    {
+      "epoch": 0.2484331655193671,
+      "grad_norm": 0.11476657539606094,
+      "learning_rate": 0.009934264585045193,
+      "loss": 2.4488,
+      "step": 2418
+    },
+    {
+      "epoch": 0.2487413952532621,
+      "grad_norm": 0.05828983336687088,
+      "learning_rate": 0.00994658997534922,
+      "loss": 2.4167,
+      "step": 2421
+    },
+    {
+      "epoch": 0.2490496249871571,
+      "grad_norm": 0.05311143398284912,
+      "learning_rate": 0.009958915365653246,
+      "loss": 2.451,
+      "step": 2424
+    },
+    {
+      "epoch": 0.2493578547210521,
+      "grad_norm": 0.14447921514511108,
+      "learning_rate": 0.009971240755957273,
+      "loss": 2.4448,
+      "step": 2427
+    },
+    {
+      "epoch": 0.2496660844549471,
+      "grad_norm": 0.178679421544075,
+      "learning_rate": 0.009983566146261297,
+      "loss": 2.4577,
+      "step": 2430
+    },
+    {
+      "epoch": 0.24997431418884208,
+      "grad_norm": 0.18707922101020813,
+      "learning_rate": 0.009995891536565325,
+      "loss": 2.4544,
+      "step": 2433
+    },
+    {
+      "epoch": 0.2502825439227371,
+      "grad_norm": 0.11012792587280273,
+      "learning_rate": 0.01,
+      "loss": 2.4636,
+      "step": 2436
+    },
+    {
+      "epoch": 0.25059077365663207,
+      "grad_norm": 0.1133418157696724,
+      "learning_rate": 0.01,
+      "loss": 2.4694,
+      "step": 2439
+    },
+    {
+      "epoch": 0.25089900339052706,
+      "grad_norm": 0.09263787418603897,
+      "learning_rate": 0.01,
+      "loss": 2.4174,
+      "step": 2442
+    },
+    {
+      "epoch": 0.25120723312442206,
+      "grad_norm": 0.07637537270784378,
+      "learning_rate": 0.01,
+      "loss": 2.4546,
+      "step": 2445
+    },
+    {
+      "epoch": 0.25151546285831705,
+      "grad_norm": 0.05083318054676056,
+      "learning_rate": 0.01,
+      "loss": 2.4517,
+      "step": 2448
+    },
+    {
+      "epoch": 0.25182369259221205,
+      "grad_norm": 0.11429949849843979,
+      "learning_rate": 0.01,
+      "loss": 2.3998,
+      "step": 2451
+    },
+    {
+      "epoch": 0.25213192232610704,
+      "grad_norm": 0.0740060955286026,
+      "learning_rate": 0.01,
+      "loss": 2.4572,
+      "step": 2454
+    },
+    {
+      "epoch": 0.25244015206000203,
+      "grad_norm": 0.23151956498622894,
+      "learning_rate": 0.01,
+      "loss": 2.4507,
+      "step": 2457
+    },
+    {
+      "epoch": 0.252748381793897,
+      "grad_norm": 0.09557089954614639,
+      "learning_rate": 0.01,
+      "loss": 2.438,
+      "step": 2460
+    },
+    {
+      "epoch": 0.2530566115277921,
+      "grad_norm": 0.06453042477369308,
+      "learning_rate": 0.01,
+      "loss": 2.4444,
+      "step": 2463
+    },
+    {
+      "epoch": 0.25336484126168707,
+      "grad_norm": 0.06805883347988129,
+      "learning_rate": 0.01,
+      "loss": 2.4333,
+      "step": 2466
+    },
+    {
+      "epoch": 0.25367307099558206,
+      "grad_norm": 0.12063002586364746,
+      "learning_rate": 0.01,
+      "loss": 2.4349,
+      "step": 2469
+    },
+    {
+      "epoch": 0.25398130072947706,
+      "grad_norm": 0.12683679163455963,
+      "learning_rate": 0.01,
+      "loss": 2.4615,
+      "step": 2472
+    },
+    {
+      "epoch": 0.25428953046337205,
+      "grad_norm": 0.19388514757156372,
+      "learning_rate": 0.01,
+      "loss": 2.4251,
+      "step": 2475
+    },
+    {
+      "epoch": 0.25459776019726704,
+      "grad_norm": 0.15118692815303802,
+      "learning_rate": 0.01,
+      "loss": 2.4493,
+      "step": 2478
+    },
+    {
+      "epoch": 0.25490598993116204,
+      "grad_norm": 0.0716528594493866,
+      "learning_rate": 0.01,
+      "loss": 2.4177,
+      "step": 2481
+    },
+    {
+      "epoch": 0.25521421966505703,
+      "grad_norm": 0.06410454958677292,
+      "learning_rate": 0.01,
+      "loss": 2.4472,
+      "step": 2484
+    },
+    {
+      "epoch": 0.255522449398952,
+      "grad_norm": 0.0613977424800396,
+      "learning_rate": 0.01,
+      "loss": 2.4374,
+      "step": 2487
+    },
+    {
+      "epoch": 0.255830679132847,
+      "grad_norm": 0.18522503972053528,
+      "learning_rate": 0.01,
+      "loss": 2.4237,
+      "step": 2490
+    },
+    {
+      "epoch": 0.256138908866742,
+      "grad_norm": 0.10789433866739273,
+      "learning_rate": 0.01,
+      "loss": 2.4035,
+      "step": 2493
+    },
+    {
+      "epoch": 0.256447138600637,
+      "grad_norm": 0.17734338343143463,
+      "learning_rate": 0.01,
+      "loss": 2.4809,
+      "step": 2496
+    },
+    {
+      "epoch": 0.256755368334532,
+      "grad_norm": 0.09952409565448761,
+      "learning_rate": 0.01,
+      "loss": 2.4586,
+      "step": 2499
+    },
+    {
+      "epoch": 0.257063598068427,
+      "grad_norm": 0.15578734874725342,
+      "learning_rate": 0.01,
+      "loss": 2.4451,
+      "step": 2502
+    },
+    {
+      "epoch": 0.257371827802322,
+      "grad_norm": 0.11684698611497879,
+      "learning_rate": 0.01,
+      "loss": 2.4422,
+      "step": 2505
+    },
+    {
+      "epoch": 0.257680057536217,
+      "grad_norm": 0.06539366394281387,
+      "learning_rate": 0.01,
+      "loss": 2.4369,
+      "step": 2508
+    },
+    {
+      "epoch": 0.25798828727011197,
+      "grad_norm": 0.15363268554210663,
+      "learning_rate": 0.01,
+      "loss": 2.4307,
+      "step": 2511
+    },
+    {
+      "epoch": 0.25829651700400696,
+      "grad_norm": 0.07657501846551895,
+      "learning_rate": 0.01,
+      "loss": 2.4287,
+      "step": 2514
+    },
+    {
+      "epoch": 0.25860474673790196,
+      "grad_norm": 0.11238528788089752,
+      "learning_rate": 0.01,
+      "loss": 2.415,
+      "step": 2517
+    },
+    {
+      "epoch": 0.25891297647179695,
+      "grad_norm": 0.08362044394016266,
+      "learning_rate": 0.01,
+      "loss": 2.4603,
+      "step": 2520
+    },
+    {
+      "epoch": 0.259221206205692,
+      "grad_norm": 0.07373514771461487,
+      "learning_rate": 0.01,
+      "loss": 2.3937,
+      "step": 2523
+    },
+    {
+      "epoch": 0.259529435939587,
+      "grad_norm": 0.062842458486557,
+      "learning_rate": 0.01,
+      "loss": 2.4096,
+      "step": 2526
+    },
+    {
+      "epoch": 0.259837665673482,
+      "grad_norm": 0.12551096081733704,
+      "learning_rate": 0.01,
+      "loss": 2.4379,
+      "step": 2529
+    },
+    {
+      "epoch": 0.260145895407377,
+      "grad_norm": 0.06409156322479248,
+      "learning_rate": 0.01,
+      "loss": 2.4212,
+      "step": 2532
+    },
+    {
+      "epoch": 0.260454125141272,
+      "grad_norm": 0.10057753324508667,
+      "learning_rate": 0.01,
+      "loss": 2.4349,
+      "step": 2535
+    },
+    {
+      "epoch": 0.26076235487516697,
+      "grad_norm": 0.1575561910867691,
+      "learning_rate": 0.01,
+      "loss": 2.44,
+      "step": 2538
+    },
+    {
+      "epoch": 0.26107058460906196,
+      "grad_norm": 0.25684165954589844,
+      "learning_rate": 0.01,
+      "loss": 2.4308,
+      "step": 2541
+    },
+    {
+      "epoch": 0.26137881434295696,
+      "grad_norm": 0.07472192496061325,
+      "learning_rate": 0.01,
+      "loss": 2.4065,
+      "step": 2544
+    },
+    {
+      "epoch": 0.26168704407685195,
+      "grad_norm": 0.060896482318639755,
+      "learning_rate": 0.01,
+      "loss": 2.4347,
+      "step": 2547
+    },
+    {
+      "epoch": 0.26199527381074694,
+      "grad_norm": 0.12883131206035614,
+      "learning_rate": 0.01,
+      "loss": 2.42,
+      "step": 2550
+    },
+    {
+      "epoch": 0.26230350354464194,
+      "grad_norm": 0.10772990435361862,
+      "learning_rate": 0.01,
+      "loss": 2.3982,
+      "step": 2553
+    },
+    {
+      "epoch": 0.26261173327853693,
+      "grad_norm": 0.20955395698547363,
+      "learning_rate": 0.01,
+      "loss": 2.4204,
+      "step": 2556
+    },
+    {
+      "epoch": 0.2629199630124319,
+      "grad_norm": 0.08120223879814148,
+      "learning_rate": 0.01,
+      "loss": 2.4192,
+      "step": 2559
+    },
+    {
+      "epoch": 0.2632281927463269,
+      "grad_norm": 0.059099119156599045,
+      "learning_rate": 0.01,
+      "loss": 2.4252,
+      "step": 2562
+    },
+    {
+      "epoch": 0.2635364224802219,
+      "grad_norm": 0.08729352802038193,
+      "learning_rate": 0.01,
+      "loss": 2.4227,
+      "step": 2565
+    },
+    {
+      "epoch": 0.2638446522141169,
+      "grad_norm": 0.1920178234577179,
+      "learning_rate": 0.01,
+      "loss": 2.4017,
+      "step": 2568
+    },
+    {
+      "epoch": 0.2641528819480119,
+      "grad_norm": 0.15997105836868286,
+      "learning_rate": 0.01,
+      "loss": 2.444,
+      "step": 2571
+    },
+    {
+      "epoch": 0.2644611116819069,
+      "grad_norm": 0.12249890714883804,
+      "learning_rate": 0.01,
+      "loss": 2.3957,
+      "step": 2574
+    },
+    {
+      "epoch": 0.2647693414158019,
+      "grad_norm": 0.05974414199590683,
+      "learning_rate": 0.01,
+      "loss": 2.421,
+      "step": 2577
+    },
+    {
+      "epoch": 0.26507757114969693,
+      "grad_norm": 0.13711535930633545,
+      "learning_rate": 0.01,
+      "loss": 2.4234,
+      "step": 2580
+    },
+    {
+      "epoch": 0.2653858008835919,
+      "grad_norm": 0.15437988936901093,
+      "learning_rate": 0.01,
+      "loss": 2.4216,
+      "step": 2583
+    },
+    {
+      "epoch": 0.2656940306174869,
+      "grad_norm": 0.10766157507896423,
+      "learning_rate": 0.01,
+      "loss": 2.4086,
+      "step": 2586
+    },
+    {
+      "epoch": 0.2660022603513819,
+      "grad_norm": 0.0736764669418335,
+      "learning_rate": 0.01,
+      "loss": 2.4227,
+      "step": 2589
+    },
+    {
+      "epoch": 0.2663104900852769,
+      "grad_norm": 0.06279190629720688,
+      "learning_rate": 0.01,
+      "loss": 2.4371,
+      "step": 2592
+    },
+    {
+      "epoch": 0.2666187198191719,
+      "grad_norm": 0.11150863766670227,
+      "learning_rate": 0.01,
+      "loss": 2.3941,
+      "step": 2595
+    },
+    {
+      "epoch": 0.2669269495530669,
+      "grad_norm": 0.1527506411075592,
+      "learning_rate": 0.01,
+      "loss": 2.4287,
+      "step": 2598
+    },
+    {
+      "epoch": 0.2672351792869619,
+      "grad_norm": 0.13321219384670258,
+      "learning_rate": 0.01,
+      "loss": 2.3995,
+      "step": 2601
+    },
+    {
+      "epoch": 0.2675434090208569,
+      "grad_norm": 0.1157502606511116,
+      "learning_rate": 0.01,
+      "loss": 2.4284,
+      "step": 2604
+    },
+    {
+      "epoch": 0.2678516387547519,
+      "grad_norm": 0.10027257353067398,
+      "learning_rate": 0.01,
+      "loss": 2.3877,
+      "step": 2607
+    },
+    {
+      "epoch": 0.26815986848864687,
+      "grad_norm": 0.10909545421600342,
+      "learning_rate": 0.01,
+      "loss": 2.4134,
+      "step": 2610
+    },
+    {
+      "epoch": 0.26846809822254186,
+      "grad_norm": 0.09810952842235565,
+      "learning_rate": 0.01,
+      "loss": 2.4231,
+      "step": 2613
+    },
+    {
+      "epoch": 0.26877632795643686,
+      "grad_norm": 0.06906435638666153,
+      "learning_rate": 0.01,
+      "loss": 2.3989,
+      "step": 2616
+    },
+    {
+      "epoch": 0.26908455769033185,
+      "grad_norm": 0.10627961158752441,
+      "learning_rate": 0.01,
+      "loss": 2.4333,
+      "step": 2619
+    },
+    {
+      "epoch": 0.26939278742422684,
+      "grad_norm": 0.10462147742509842,
+      "learning_rate": 0.01,
+      "loss": 2.394,
+      "step": 2622
+    },
+    {
+      "epoch": 0.26970101715812184,
+      "grad_norm": 0.10885953158140182,
+      "learning_rate": 0.01,
+      "loss": 2.4172,
+      "step": 2625
+    },
+    {
+      "epoch": 0.27000924689201683,
+      "grad_norm": 0.0981958881020546,
+      "learning_rate": 0.01,
+      "loss": 2.4112,
+      "step": 2628
+    },
+    {
+      "epoch": 0.2703174766259118,
+      "grad_norm": 0.14177650213241577,
+      "learning_rate": 0.01,
+      "loss": 2.3748,
+      "step": 2631
+    },
+    {
+      "epoch": 0.2706257063598068,
+      "grad_norm": 0.06374615430831909,
+      "learning_rate": 0.01,
+      "loss": 2.4054,
+      "step": 2634
+    },
+    {
+      "epoch": 0.27093393609370187,
+      "grad_norm": 0.23363849520683289,
+      "learning_rate": 0.01,
+      "loss": 2.4194,
+      "step": 2637
+    },
+    {
+      "epoch": 0.27124216582759686,
+      "grad_norm": 0.10294153541326523,
+      "learning_rate": 0.01,
+      "loss": 2.384,
+      "step": 2640
+    },
+    {
+      "epoch": 0.27155039556149185,
+      "grad_norm": 0.17200984060764313,
+      "learning_rate": 0.01,
+      "loss": 2.4122,
+      "step": 2643
+    },
+    {
+      "epoch": 0.27185862529538685,
+      "grad_norm": 0.06513970345258713,
+      "learning_rate": 0.01,
+      "loss": 2.4152,
+      "step": 2646
+    },
+    {
+      "epoch": 0.27216685502928184,
+      "grad_norm": 0.08533628284931183,
+      "learning_rate": 0.01,
+      "loss": 2.4508,
+      "step": 2649
+    },
+    {
+      "epoch": 0.27247508476317683,
+      "grad_norm": 0.07299966365098953,
+      "learning_rate": 0.01,
+      "loss": 2.4091,
+      "step": 2652
+    },
+    {
+      "epoch": 0.2727833144970718,
+      "grad_norm": 0.06617329269647598,
+      "learning_rate": 0.01,
+      "loss": 2.4,
+      "step": 2655
+    },
+    {
+      "epoch": 0.2730915442309668,
+      "grad_norm": 0.07062381505966187,
+      "learning_rate": 0.01,
+      "loss": 2.4002,
+      "step": 2658
+    },
+    {
+      "epoch": 0.2733997739648618,
+      "grad_norm": 0.11162712424993515,
+      "learning_rate": 0.01,
+      "loss": 2.386,
+      "step": 2661
+    },
+    {
+      "epoch": 0.2737080036987568,
+      "grad_norm": 0.07827174663543701,
+      "learning_rate": 0.01,
+      "loss": 2.4111,
+      "step": 2664
+    },
+    {
+      "epoch": 0.2740162334326518,
+      "grad_norm": 0.07248109579086304,
+      "learning_rate": 0.01,
+      "loss": 2.3968,
+      "step": 2667
+    },
+    {
+      "epoch": 0.2743244631665468,
+      "grad_norm": 0.1251075118780136,
+      "learning_rate": 0.01,
+      "loss": 2.4216,
+      "step": 2670
+    },
+    {
+      "epoch": 0.2746326929004418,
+      "grad_norm": 0.1280512660741806,
+      "learning_rate": 0.01,
+      "loss": 2.4233,
+      "step": 2673
+    },
+    {
+      "epoch": 0.2749409226343368,
+      "grad_norm": 0.06290891766548157,
+      "learning_rate": 0.01,
+      "loss": 2.412,
+      "step": 2676
+    },
+    {
+      "epoch": 0.2752491523682318,
+      "grad_norm": 0.09324091672897339,
+      "learning_rate": 0.01,
+      "loss": 2.4025,
+      "step": 2679
+    },
+    {
+      "epoch": 0.27555738210212677,
+      "grad_norm": 0.06253890693187714,
+      "learning_rate": 0.01,
+      "loss": 2.4197,
+      "step": 2682
+    },
+    {
+      "epoch": 0.27586561183602176,
+      "grad_norm": 0.10279545187950134,
+      "learning_rate": 0.01,
+      "loss": 2.4099,
+      "step": 2685
+    },
+    {
+      "epoch": 0.27617384156991676,
+      "grad_norm": 0.07942310720682144,
+      "learning_rate": 0.01,
+      "loss": 2.4052,
+      "step": 2688
+    },
+    {
+      "epoch": 0.27648207130381175,
+      "grad_norm": 0.10373161733150482,
+      "learning_rate": 0.01,
+      "loss": 2.3899,
+      "step": 2691
+    },
+    {
+      "epoch": 0.2767903010377068,
+      "grad_norm": 0.312575101852417,
+      "learning_rate": 0.01,
+      "loss": 2.4243,
+      "step": 2694
+    },
+    {
+      "epoch": 0.2770985307716018,
+      "grad_norm": 0.07417728751897812,
+      "learning_rate": 0.01,
+      "loss": 2.3604,
+      "step": 2697
+    },
+    {
+      "epoch": 0.2774067605054968,
+      "grad_norm": 0.09007294476032257,
+      "learning_rate": 0.01,
+      "loss": 2.3863,
+      "step": 2700
+    },
+    {
+      "epoch": 0.2777149902393918,
+      "grad_norm": 0.10452757775783539,
+      "learning_rate": 0.01,
+      "loss": 2.41,
+      "step": 2703
+    },
+    {
+      "epoch": 0.2780232199732868,
+      "grad_norm": 0.09276364743709564,
+      "learning_rate": 0.01,
+      "loss": 2.3878,
+      "step": 2706
+    },
+    {
+      "epoch": 0.27833144970718177,
+      "grad_norm": 0.08949960023164749,
+      "learning_rate": 0.01,
+      "loss": 2.3823,
+      "step": 2709
+    },
+    {
+      "epoch": 0.27863967944107676,
+      "grad_norm": 0.0589129813015461,
+      "learning_rate": 0.01,
+      "loss": 2.401,
+      "step": 2712
+    },
+    {
+      "epoch": 0.27894790917497175,
+      "grad_norm": 0.08298425376415253,
+      "learning_rate": 0.01,
+      "loss": 2.4,
+      "step": 2715
+    },
+    {
+      "epoch": 0.27925613890886675,
+      "grad_norm": 0.07719019800424576,
+      "learning_rate": 0.01,
+      "loss": 2.3726,
+      "step": 2718
+    },
+    {
+      "epoch": 0.27956436864276174,
+      "grad_norm": 0.09369128197431564,
+      "learning_rate": 0.01,
+      "loss": 2.3893,
+      "step": 2721
+    },
+    {
+      "epoch": 0.27987259837665673,
+      "grad_norm": 0.11461931467056274,
+      "learning_rate": 0.01,
+      "loss": 2.4017,
+      "step": 2724
+    },
+    {
+      "epoch": 0.2801808281105517,
+      "grad_norm": 0.050078991800546646,
+      "learning_rate": 0.01,
+      "loss": 2.3852,
+      "step": 2727
+    },
+    {
+      "epoch": 0.2804890578444467,
+      "grad_norm": 0.08188966661691666,
+      "learning_rate": 0.01,
+      "loss": 2.3469,
+      "step": 2730
+    },
+    {
+      "epoch": 0.2807972875783417,
+      "grad_norm": 0.0805756077170372,
+      "learning_rate": 0.01,
+      "loss": 2.3632,
+      "step": 2733
+    },
+    {
+      "epoch": 0.2811055173122367,
+      "grad_norm": 0.07377249747514725,
+      "learning_rate": 0.01,
+      "loss": 2.3852,
+      "step": 2736
+    },
+    {
+      "epoch": 0.2814137470461317,
+      "grad_norm": 0.17040085792541504,
+      "learning_rate": 0.01,
+      "loss": 2.3904,
+      "step": 2739
+    },
+    {
+      "epoch": 0.2817219767800267,
+      "grad_norm": 0.1419583261013031,
+      "learning_rate": 0.01,
+      "loss": 2.3735,
+      "step": 2742
+    },
+    {
+      "epoch": 0.2820302065139217,
+      "grad_norm": 0.13182134926319122,
+      "learning_rate": 0.01,
+      "loss": 2.3904,
+      "step": 2745
+    },
+    {
+      "epoch": 0.2823384362478167,
+      "grad_norm": 0.1058223620057106,
+      "learning_rate": 0.01,
+      "loss": 2.3922,
+      "step": 2748
+    },
+    {
+      "epoch": 0.2826466659817117,
+      "grad_norm": 0.08037062734365463,
+      "learning_rate": 0.01,
+      "loss": 2.3692,
+      "step": 2751
+    },
+    {
+      "epoch": 0.2829548957156067,
+      "grad_norm": 0.10247037559747696,
+      "learning_rate": 0.01,
+      "loss": 2.3712,
+      "step": 2754
+    },
+    {
+      "epoch": 0.2832631254495017,
+      "grad_norm": 0.09925279021263123,
+      "learning_rate": 0.01,
+      "loss": 2.3632,
+      "step": 2757
+    },
+    {
+      "epoch": 0.2835713551833967,
+      "grad_norm": 0.05111562833189964,
+      "learning_rate": 0.01,
+      "loss": 2.3622,
+      "step": 2760
+    },
+    {
+      "epoch": 0.2838795849172917,
+      "grad_norm": 0.060480840504169464,
+      "learning_rate": 0.01,
+      "loss": 2.3592,
+      "step": 2763
+    },
+    {
+      "epoch": 0.2841878146511867,
+      "grad_norm": 0.13488496840000153,
+      "learning_rate": 0.01,
+      "loss": 2.3822,
+      "step": 2766
+    },
+    {
+      "epoch": 0.2844960443850817,
+      "grad_norm": 0.08369171619415283,
+      "learning_rate": 0.01,
+      "loss": 2.3922,
+      "step": 2769
+    },
+    {
+      "epoch": 0.2848042741189767,
+      "grad_norm": 0.19474861025810242,
+      "learning_rate": 0.01,
+      "loss": 2.387,
+      "step": 2772
+    },
+    {
+      "epoch": 0.2851125038528717,
+      "grad_norm": 0.17801512777805328,
+      "learning_rate": 0.01,
+      "loss": 2.3745,
+      "step": 2775
+    },
+    {
+      "epoch": 0.2854207335867667,
+      "grad_norm": 0.0658038854598999,
+      "learning_rate": 0.01,
+      "loss": 2.3857,
+      "step": 2778
+    },
+    {
+      "epoch": 0.28572896332066167,
+      "grad_norm": 0.0510118305683136,
+      "learning_rate": 0.01,
+      "loss": 2.3735,
+      "step": 2781
+    },
+    {
+      "epoch": 0.28603719305455666,
+      "grad_norm": 0.0649714320898056,
+      "learning_rate": 0.01,
+      "loss": 2.4002,
+      "step": 2784
+    },
+    {
+      "epoch": 0.28634542278845165,
+      "grad_norm": 0.11462211608886719,
+      "learning_rate": 0.01,
+      "loss": 2.3642,
+      "step": 2787
+    },
+    {
+      "epoch": 0.28665365252234665,
+      "grad_norm": 0.0745900496840477,
+      "learning_rate": 0.01,
+      "loss": 2.4058,
+      "step": 2790
+    },
+    {
+      "epoch": 0.28696188225624164,
+      "grad_norm": 0.2475040704011917,
+      "learning_rate": 0.01,
+      "loss": 2.3778,
+      "step": 2793
+    },
+    {
+      "epoch": 0.28727011199013663,
+      "grad_norm": 0.08792129158973694,
+      "learning_rate": 0.01,
+      "loss": 2.3932,
+      "step": 2796
+    },
+    {
+      "epoch": 0.2875783417240316,
+      "grad_norm": 0.04952983185648918,
+      "learning_rate": 0.01,
+      "loss": 2.3631,
+      "step": 2799
+    },
+    {
+      "epoch": 0.2878865714579266,
+      "grad_norm": 0.053665559738874435,
+      "learning_rate": 0.01,
+      "loss": 2.3673,
+      "step": 2802
+    },
+    {
+      "epoch": 0.2881948011918216,
+      "grad_norm": 0.0579262301325798,
+      "learning_rate": 0.01,
+      "loss": 2.3234,
+      "step": 2805
+    },
+    {
+      "epoch": 0.2885030309257166,
+      "grad_norm": 0.13837358355522156,
+      "learning_rate": 0.01,
+      "loss": 2.3854,
+      "step": 2808
+    },
+    {
+      "epoch": 0.28881126065961166,
+      "grad_norm": 0.09924750030040741,
+      "learning_rate": 0.01,
+      "loss": 2.3819,
+      "step": 2811
+    },
+    {
+      "epoch": 0.28911949039350665,
+      "grad_norm": 0.14742402732372284,
+      "learning_rate": 0.01,
+      "loss": 2.3853,
+      "step": 2814
+    },
+    {
+      "epoch": 0.28942772012740164,
+      "grad_norm": 0.11731177568435669,
+      "learning_rate": 0.01,
+      "loss": 2.4082,
+      "step": 2817
+    },
+    {
+      "epoch": 0.28973594986129664,
+      "grad_norm": 0.16812686622142792,
+      "learning_rate": 0.01,
+      "loss": 2.3855,
+      "step": 2820
+    },
+    {
+      "epoch": 0.29004417959519163,
+      "grad_norm": 0.06864415854215622,
+      "learning_rate": 0.01,
+      "loss": 2.3911,
+      "step": 2823
+    },
+    {
+      "epoch": 0.2903524093290866,
+      "grad_norm": 0.050597239285707474,
+      "learning_rate": 0.01,
+      "loss": 2.3627,
+      "step": 2826
+    },
+    {
+      "epoch": 0.2906606390629816,
+      "grad_norm": 0.06927742809057236,
+      "learning_rate": 0.01,
+      "loss": 2.3653,
+      "step": 2829
+    },
+    {
+      "epoch": 0.2909688687968766,
+      "grad_norm": 0.049216922372579575,
+      "learning_rate": 0.01,
+      "loss": 2.3527,
+      "step": 2832
+    },
+    {
+      "epoch": 0.2912770985307716,
+      "grad_norm": 0.06790090352296829,
+      "learning_rate": 0.01,
+      "loss": 2.4087,
+      "step": 2835
+    },
+    {
+      "epoch": 0.2915853282646666,
+      "grad_norm": 0.14112398028373718,
+      "learning_rate": 0.01,
+      "loss": 2.3777,
+      "step": 2838
+    },
+    {
+      "epoch": 0.2918935579985616,
+      "grad_norm": 0.07459170371294022,
+      "learning_rate": 0.01,
+      "loss": 2.3776,
+      "step": 2841
+    },
+    {
+      "epoch": 0.2922017877324566,
+      "grad_norm": 0.05480146035552025,
+      "learning_rate": 0.01,
+      "loss": 2.3831,
+      "step": 2844
+    },
+    {
+      "epoch": 0.2925100174663516,
+      "grad_norm": 0.11372058093547821,
+      "learning_rate": 0.01,
+      "loss": 2.3667,
+      "step": 2847
+    },
+    {
+      "epoch": 0.2928182472002466,
+      "grad_norm": 0.05589181184768677,
+      "learning_rate": 0.01,
+      "loss": 2.3744,
+      "step": 2850
+    },
+    {
+      "epoch": 0.29312647693414157,
+      "grad_norm": 0.10505107790231705,
+      "learning_rate": 0.01,
+      "loss": 2.3461,
+      "step": 2853
+    },
+    {
+      "epoch": 0.29343470666803656,
+      "grad_norm": 0.06241190806031227,
+      "learning_rate": 0.01,
+      "loss": 2.3616,
+      "step": 2856
+    },
+    {
+      "epoch": 0.29374293640193155,
+      "grad_norm": 0.2687353193759918,
+      "learning_rate": 0.01,
+      "loss": 2.3728,
+      "step": 2859
+    },
+    {
+      "epoch": 0.29405116613582655,
+      "grad_norm": 0.13569511473178864,
+      "learning_rate": 0.01,
+      "loss": 2.3758,
+      "step": 2862
+    },
+    {
+      "epoch": 0.29435939586972154,
+      "grad_norm": 0.08852502703666687,
+      "learning_rate": 0.01,
+      "loss": 2.3805,
+      "step": 2865
+    },
+    {
+      "epoch": 0.2946676256036166,
+      "grad_norm": 0.0690246969461441,
+      "learning_rate": 0.01,
+      "loss": 2.3694,
+      "step": 2868
+    },
+    {
+      "epoch": 0.2949758553375116,
+      "grad_norm": 0.13508114218711853,
+      "learning_rate": 0.01,
+      "loss": 2.3626,
+      "step": 2871
+    },
+    {
+      "epoch": 0.2952840850714066,
+      "grad_norm": 0.06574945896863937,
+      "learning_rate": 0.01,
+      "loss": 2.3661,
+      "step": 2874
+    },
+    {
+      "epoch": 0.29559231480530157,
+      "grad_norm": 0.08492054790258408,
+      "learning_rate": 0.01,
+      "loss": 2.3737,
+      "step": 2877
+    },
+    {
+      "epoch": 0.29590054453919656,
+      "grad_norm": 0.11930177360773087,
+      "learning_rate": 0.01,
+      "loss": 2.3684,
+      "step": 2880
+    },
+    {
+      "epoch": 0.29620877427309156,
+      "grad_norm": 0.06913982331752777,
+      "learning_rate": 0.01,
+      "loss": 2.3597,
+      "step": 2883
+    },
+    {
+      "epoch": 0.29651700400698655,
+      "grad_norm": 0.1508978009223938,
+      "learning_rate": 0.01,
+      "loss": 2.3807,
+      "step": 2886
+    },
+    {
+      "epoch": 0.29682523374088154,
+      "grad_norm": 0.059416841715574265,
+      "learning_rate": 0.01,
+      "loss": 2.3672,
+      "step": 2889
+    },
+    {
+      "epoch": 0.29713346347477654,
+      "grad_norm": 0.07864934206008911,
+      "learning_rate": 0.01,
+      "loss": 2.3542,
+      "step": 2892
+    },
+    {
+      "epoch": 0.29744169320867153,
+      "grad_norm": 0.15172207355499268,
+      "learning_rate": 0.01,
+      "loss": 2.3816,
+      "step": 2895
+    },
+    {
+      "epoch": 0.2977499229425665,
+      "grad_norm": 0.08946362882852554,
+      "learning_rate": 0.01,
+      "loss": 2.3854,
+      "step": 2898
+    },
+    {
+      "epoch": 0.2980581526764615,
+      "grad_norm": 0.06231836602091789,
+      "learning_rate": 0.01,
+      "loss": 2.3803,
+      "step": 2901
+    },
+    {
+      "epoch": 0.2983663824103565,
+      "grad_norm": 0.06673764437437057,
+      "learning_rate": 0.01,
+      "loss": 2.3506,
+      "step": 2904
+    },
+    {
+      "epoch": 0.2986746121442515,
+      "grad_norm": 0.11514609307050705,
+      "learning_rate": 0.01,
+      "loss": 2.3345,
+      "step": 2907
+    },
+    {
+      "epoch": 0.2989828418781465,
+      "grad_norm": 0.05702753737568855,
+      "learning_rate": 0.01,
+      "loss": 2.353,
+      "step": 2910
+    },
+    {
+      "epoch": 0.2992910716120415,
+      "grad_norm": 0.09202984720468521,
+      "learning_rate": 0.01,
+      "loss": 2.3978,
+      "step": 2913
+    },
+    {
+      "epoch": 0.2995993013459365,
+      "grad_norm": 0.09088042378425598,
+      "learning_rate": 0.01,
+      "loss": 2.3508,
+      "step": 2916
+    },
+    {
+      "epoch": 0.2999075310798315,
+      "grad_norm": 0.09106214344501495,
+      "learning_rate": 0.01,
+      "loss": 2.3695,
+      "step": 2919
+    },
+    {
+      "epoch": 0.3002157608137265,
+      "grad_norm": 0.12793834507465363,
+      "learning_rate": 0.01,
+      "loss": 2.3585,
+      "step": 2922
+    },
+    {
+      "epoch": 0.30052399054762147,
+      "grad_norm": 0.16437458992004395,
+      "learning_rate": 0.01,
+      "loss": 2.3708,
+      "step": 2925
+    },
+    {
+      "epoch": 0.3008322202815165,
+      "grad_norm": 0.10168170928955078,
+      "learning_rate": 0.01,
+      "loss": 2.3839,
+      "step": 2928
+    },
+    {
+      "epoch": 0.3011404500154115,
+      "grad_norm": 0.12716282904148102,
+      "learning_rate": 0.01,
+      "loss": 2.3653,
+      "step": 2931
+    },
+    {
+      "epoch": 0.3014486797493065,
+      "grad_norm": 0.05094976723194122,
+      "learning_rate": 0.01,
+      "loss": 2.3315,
+      "step": 2934
+    },
+    {
+      "epoch": 0.3017569094832015,
+      "grad_norm": 0.11750750988721848,
+      "learning_rate": 0.01,
+      "loss": 2.3544,
+      "step": 2937
+    },
+    {
+      "epoch": 0.3020651392170965,
+      "grad_norm": 0.0688977912068367,
+      "learning_rate": 0.01,
+      "loss": 2.3485,
+      "step": 2940
+    },
+    {
+      "epoch": 0.3023733689509915,
+      "grad_norm": 0.09537909924983978,
+      "learning_rate": 0.01,
+      "loss": 2.3357,
+      "step": 2943
+    },
+    {
+      "epoch": 0.3026815986848865,
+      "grad_norm": 0.15028056502342224,
+      "learning_rate": 0.01,
+      "loss": 2.3029,
+      "step": 2946
+    },
+    {
+      "epoch": 0.30298982841878147,
+      "grad_norm": 0.2069140523672104,
+      "learning_rate": 0.01,
+      "loss": 2.3658,
+      "step": 2949
+    },
+    {
+      "epoch": 0.30329805815267646,
+      "grad_norm": 0.04774792492389679,
+      "learning_rate": 0.01,
+      "loss": 2.3488,
+      "step": 2952
+    },
+    {
+      "epoch": 0.30360628788657146,
+      "grad_norm": 0.04033259302377701,
+      "learning_rate": 0.01,
+      "loss": 2.3536,
+      "step": 2955
+    },
+    {
+      "epoch": 0.30391451762046645,
+      "grad_norm": 0.04587483033537865,
+      "learning_rate": 0.01,
+      "loss": 2.3377,
+      "step": 2958
+    },
+    {
+      "epoch": 0.30422274735436144,
+      "grad_norm": 0.08392881602048874,
+      "learning_rate": 0.01,
+      "loss": 2.3323,
+      "step": 2961
+    },
+    {
+      "epoch": 0.30453097708825644,
+      "grad_norm": 0.16665025055408478,
+      "learning_rate": 0.01,
+      "loss": 2.3763,
+      "step": 2964
+    },
+    {
+      "epoch": 0.30483920682215143,
+      "grad_norm": 0.19268077611923218,
+      "learning_rate": 0.01,
+      "loss": 2.3545,
+      "step": 2967
+    },
+    {
+      "epoch": 0.3051474365560464,
+      "grad_norm": 0.14428319036960602,
+      "learning_rate": 0.01,
+      "loss": 2.3481,
+      "step": 2970
+    },
+    {
+      "epoch": 0.3054556662899414,
+      "grad_norm": 0.08958342671394348,
+      "learning_rate": 0.01,
+      "loss": 2.3704,
+      "step": 2973
+    },
+    {
+      "epoch": 0.3057638960238364,
+      "grad_norm": 0.06964152306318283,
+      "learning_rate": 0.01,
+      "loss": 2.3649,
+      "step": 2976
+    },
+    {
+      "epoch": 0.3060721257577314,
+      "grad_norm": 0.1336866170167923,
+      "learning_rate": 0.01,
+      "loss": 2.3426,
+      "step": 2979
+    },
+    {
+      "epoch": 0.3063803554916264,
+      "grad_norm": 0.06913724541664124,
+      "learning_rate": 0.01,
+      "loss": 2.363,
+      "step": 2982
+    },
+    {
+      "epoch": 0.30668858522552145,
+      "grad_norm": 0.0705854743719101,
+      "learning_rate": 0.01,
+      "loss": 2.327,
+      "step": 2985
+    },
+    {
+      "epoch": 0.30699681495941644,
+      "grad_norm": 0.06596222519874573,
+      "learning_rate": 0.01,
+      "loss": 2.3669,
+      "step": 2988
+    },
+    {
+      "epoch": 0.30730504469331144,
+      "grad_norm": 0.12716993689537048,
+      "learning_rate": 0.01,
+      "loss": 2.3395,
+      "step": 2991
+    },
+    {
+      "epoch": 0.30761327442720643,
+      "grad_norm": 0.09933049976825714,
+      "learning_rate": 0.01,
+      "loss": 2.3532,
+      "step": 2994
+    },
+    {
+      "epoch": 0.3079215041611014,
+      "grad_norm": 0.19280697405338287,
+      "learning_rate": 0.01,
+      "loss": 2.3513,
+      "step": 2997
+    },
+    {
+      "epoch": 0.3082297338949964,
+      "grad_norm": 0.08448618650436401,
+      "learning_rate": 0.01,
+      "loss": 2.3828,
+      "step": 3000
+    },
+    {
+      "epoch": 0.3085379636288914,
+      "grad_norm": 0.14882349967956543,
+      "learning_rate": 0.01,
+      "loss": 2.3398,
+      "step": 3003
+    },
+    {
+      "epoch": 0.3088461933627864,
+      "grad_norm": 0.08360068500041962,
+      "learning_rate": 0.01,
+      "loss": 2.3414,
+      "step": 3006
+    },
+    {
+      "epoch": 0.3091544230966814,
+      "grad_norm": 0.1378074288368225,
+      "learning_rate": 0.01,
+      "loss": 2.3461,
+      "step": 3009
+    },
+    {
+      "epoch": 0.3094626528305764,
+      "grad_norm": 0.13160692155361176,
+      "learning_rate": 0.01,
+      "loss": 2.3517,
+      "step": 3012
+    },
+    {
+      "epoch": 0.3097708825644714,
+      "grad_norm": 0.0702040046453476,
+      "learning_rate": 0.01,
+      "loss": 2.3524,
+      "step": 3015
+    },
+    {
+      "epoch": 0.3100791122983664,
+      "grad_norm": 0.06959223002195358,
+      "learning_rate": 0.01,
+      "loss": 2.3398,
+      "step": 3018
+    },
+    {
+      "epoch": 0.31038734203226137,
+      "grad_norm": 0.10830830782651901,
+      "learning_rate": 0.01,
+      "loss": 2.3437,
+      "step": 3021
+    },
+    {
+      "epoch": 0.31069557176615636,
+      "grad_norm": 0.09298605471849442,
+      "learning_rate": 0.01,
+      "loss": 2.3473,
+      "step": 3024
+    },
+    {
+      "epoch": 0.31100380150005136,
+      "grad_norm": 0.06620427966117859,
+      "learning_rate": 0.01,
+      "loss": 2.3341,
+      "step": 3027
+    },
+    {
+      "epoch": 0.31131203123394635,
+      "grad_norm": 0.21722812950611115,
+      "learning_rate": 0.01,
+      "loss": 2.3281,
+      "step": 3030
+    },
+    {
+      "epoch": 0.31162026096784134,
+      "grad_norm": 0.1475544422864914,
+      "learning_rate": 0.01,
+      "loss": 2.3383,
+      "step": 3033
+    },
+    {
+      "epoch": 0.31192849070173634,
+      "grad_norm": 0.13449987769126892,
+      "learning_rate": 0.01,
+      "loss": 2.314,
+      "step": 3036
+    },
+    {
+      "epoch": 0.31223672043563133,
+      "grad_norm": 0.06219559907913208,
+      "learning_rate": 0.01,
+      "loss": 2.3526,
+      "step": 3039
+    },
+    {
+      "epoch": 0.3125449501695264,
+      "grad_norm": 0.05337538942694664,
+      "learning_rate": 0.01,
+      "loss": 2.3386,
+      "step": 3042
+    },
+    {
+      "epoch": 0.3128531799034214,
+      "grad_norm": 0.11457488685846329,
+      "learning_rate": 0.01,
+      "loss": 2.3261,
+      "step": 3045
+    },
+    {
+      "epoch": 0.31316140963731637,
+      "grad_norm": 0.23809069395065308,
+      "learning_rate": 0.01,
+      "loss": 2.3411,
+      "step": 3048
+    },
+    {
+      "epoch": 0.31346963937121136,
+      "grad_norm": 0.11100046336650848,
+      "learning_rate": 0.01,
+      "loss": 2.3269,
+      "step": 3051
+    },
+    {
+      "epoch": 0.31377786910510636,
+      "grad_norm": 0.05229029804468155,
+      "learning_rate": 0.01,
+      "loss": 2.3339,
+      "step": 3054
+    },
+    {
+      "epoch": 0.31408609883900135,
+      "grad_norm": 0.05956039950251579,
+      "learning_rate": 0.01,
+      "loss": 2.3566,
+      "step": 3057
+    },
+    {
+      "epoch": 0.31439432857289634,
+      "grad_norm": 0.13084881007671356,
+      "learning_rate": 0.01,
+      "loss": 2.332,
+      "step": 3060
+    },
+    {
+      "epoch": 0.31470255830679134,
+      "grad_norm": 0.13889305293560028,
+      "learning_rate": 0.01,
+      "loss": 2.3118,
+      "step": 3063
+    },
+    {
+      "epoch": 0.31501078804068633,
+      "grad_norm": 0.10443049669265747,
+      "learning_rate": 0.01,
+      "loss": 2.3246,
+      "step": 3066
+    },
+    {
+      "epoch": 0.3153190177745813,
+      "grad_norm": 0.04321267828345299,
+      "learning_rate": 0.01,
+      "loss": 2.3353,
+      "step": 3069
+    },
+    {
+      "epoch": 0.3156272475084763,
+      "grad_norm": 0.046873319894075394,
+      "learning_rate": 0.01,
+      "loss": 2.3144,
+      "step": 3072
+    },
+    {
+      "epoch": 0.3159354772423713,
+      "grad_norm": 0.06548158824443817,
+      "learning_rate": 0.01,
+      "loss": 2.3285,
+      "step": 3075
+    },
+    {
+      "epoch": 0.3162437069762663,
+      "grad_norm": 0.19105824828147888,
+      "learning_rate": 0.01,
+      "loss": 2.349,
+      "step": 3078
+    },
+    {
+      "epoch": 0.3165519367101613,
+      "grad_norm": 0.14477142691612244,
+      "learning_rate": 0.01,
+      "loss": 2.3074,
+      "step": 3081
+    },
+    {
+      "epoch": 0.3168601664440563,
+      "grad_norm": 0.08536936342716217,
+      "learning_rate": 0.01,
+      "loss": 2.3462,
+      "step": 3084
+    },
+    {
+      "epoch": 0.3171683961779513,
+      "grad_norm": 0.0595535933971405,
+      "learning_rate": 0.01,
+      "loss": 2.3522,
+      "step": 3087
+    },
+    {
+      "epoch": 0.3174766259118463,
+      "grad_norm": 0.058548733592033386,
+      "learning_rate": 0.01,
+      "loss": 2.3314,
+      "step": 3090
+    },
+    {
+      "epoch": 0.31778485564574127,
+      "grad_norm": 0.04651311784982681,
+      "learning_rate": 0.01,
+      "loss": 2.3058,
+      "step": 3093
+    },
+    {
+      "epoch": 0.31809308537963626,
+      "grad_norm": 0.0516805462539196,
+      "learning_rate": 0.01,
+      "loss": 2.3243,
+      "step": 3096
+    },
+    {
+      "epoch": 0.31840131511353126,
+      "grad_norm": 0.16851970553398132,
+      "learning_rate": 0.01,
+      "loss": 2.3315,
+      "step": 3099
+    },
+    {
+      "epoch": 0.3187095448474263,
+      "grad_norm": 0.08350600302219391,
+      "learning_rate": 0.01,
+      "loss": 2.315,
+      "step": 3102
+    },
+    {
+      "epoch": 0.3190177745813213,
+      "grad_norm": 0.08899964392185211,
+      "learning_rate": 0.01,
+      "loss": 2.3218,
+      "step": 3105
+    },
+    {
+      "epoch": 0.3193260043152163,
+      "grad_norm": 0.2424800843000412,
+      "learning_rate": 0.01,
+      "loss": 2.3207,
+      "step": 3108
+    },
+    {
+      "epoch": 0.3196342340491113,
+      "grad_norm": 0.213782399892807,
+      "learning_rate": 0.01,
+      "loss": 2.3728,
+      "step": 3111
+    },
+    {
+      "epoch": 0.3199424637830063,
+      "grad_norm": 0.15629780292510986,
+      "learning_rate": 0.01,
+      "loss": 2.3453,
+      "step": 3114
+    },
+    {
+      "epoch": 0.3202506935169013,
+      "grad_norm": 0.06920924782752991,
+      "learning_rate": 0.01,
+      "loss": 2.3111,
+      "step": 3117
+    },
+    {
+      "epoch": 0.32055892325079627,
+      "grad_norm": 0.04514181613922119,
+      "learning_rate": 0.01,
+      "loss": 2.33,
+      "step": 3120
+    },
+    {
+      "epoch": 0.32086715298469126,
+      "grad_norm": 0.05500979721546173,
+      "learning_rate": 0.01,
+      "loss": 2.3078,
+      "step": 3123
+    },
+    {
+      "epoch": 0.32117538271858626,
+      "grad_norm": 0.09148071706295013,
+      "learning_rate": 0.01,
+      "loss": 2.3457,
+      "step": 3126
+    },
+    {
+      "epoch": 0.32148361245248125,
+      "grad_norm": 0.10582035779953003,
+      "learning_rate": 0.01,
+      "loss": 2.3114,
+      "step": 3129
+    },
+    {
+      "epoch": 0.32179184218637624,
+      "grad_norm": 0.1557345986366272,
+      "learning_rate": 0.01,
+      "loss": 2.3334,
+      "step": 3132
+    },
+    {
+      "epoch": 0.32210007192027124,
+      "grad_norm": 0.11304829269647598,
+      "learning_rate": 0.01,
+      "loss": 2.2797,
+      "step": 3135
+    },
+    {
+      "epoch": 0.32240830165416623,
+      "grad_norm": 0.08236223459243774,
+      "learning_rate": 0.01,
+      "loss": 2.3357,
+      "step": 3138
+    },
+    {
+      "epoch": 0.3227165313880612,
+      "grad_norm": 0.09718946367502213,
+      "learning_rate": 0.01,
+      "loss": 2.3096,
+      "step": 3141
+    },
+    {
+      "epoch": 0.3230247611219562,
+      "grad_norm": 0.07455772161483765,
+      "learning_rate": 0.01,
+      "loss": 2.3127,
+      "step": 3144
+    },
+    {
+      "epoch": 0.3233329908558512,
+      "grad_norm": 0.0556890033185482,
+      "learning_rate": 0.01,
+      "loss": 2.3088,
+      "step": 3147
+    },
+    {
+      "epoch": 0.3236412205897462,
+      "grad_norm": 0.07595494389533997,
+      "learning_rate": 0.01,
+      "loss": 2.3159,
+      "step": 3150
+    },
+    {
+      "epoch": 0.3239494503236412,
+      "grad_norm": 0.07064896821975708,
+      "learning_rate": 0.01,
+      "loss": 2.3336,
+      "step": 3153
+    },
+    {
+      "epoch": 0.3242576800575362,
+      "grad_norm": 0.06646276265382767,
+      "learning_rate": 0.01,
+      "loss": 2.324,
+      "step": 3156
+    },
+    {
+      "epoch": 0.32456590979143124,
+      "grad_norm": 0.08837945014238358,
+      "learning_rate": 0.01,
+      "loss": 2.3191,
+      "step": 3159
+    },
+    {
+      "epoch": 0.32487413952532623,
+      "grad_norm": 0.13228796422481537,
+      "learning_rate": 0.01,
+      "loss": 2.3231,
+      "step": 3162
+    },
+    {
+      "epoch": 0.3251823692592212,
+      "grad_norm": 0.1080455407500267,
+      "learning_rate": 0.01,
+      "loss": 2.3341,
+      "step": 3165
+    },
+    {
+      "epoch": 0.3254905989931162,
+      "grad_norm": 0.1073957234621048,
+      "learning_rate": 0.01,
+      "loss": 2.3237,
+      "step": 3168
+    },
+    {
+      "epoch": 0.3257988287270112,
+      "grad_norm": 0.12472347915172577,
+      "learning_rate": 0.01,
+      "loss": 2.3315,
+      "step": 3171
+    },
+    {
+      "epoch": 0.3261070584609062,
+      "grad_norm": 0.09123571217060089,
+      "learning_rate": 0.01,
+      "loss": 2.3588,
+      "step": 3174
+    },
+    {
+      "epoch": 0.3264152881948012,
+      "grad_norm": 0.07830306142568588,
+      "learning_rate": 0.01,
+      "loss": 2.3273,
+      "step": 3177
+    },
+    {
+      "epoch": 0.3267235179286962,
+      "grad_norm": 0.11552650481462479,
+      "learning_rate": 0.01,
+      "loss": 2.3407,
+      "step": 3180
+    },
+    {
+      "epoch": 0.3270317476625912,
+      "grad_norm": 0.13251489400863647,
+      "learning_rate": 0.01,
+      "loss": 2.3241,
+      "step": 3183
+    },
+    {
+      "epoch": 0.3273399773964862,
+      "grad_norm": 0.12775808572769165,
+      "learning_rate": 0.01,
+      "loss": 2.331,
+      "step": 3186
+    },
+    {
+      "epoch": 0.3276482071303812,
+      "grad_norm": 0.12069859355688095,
+      "learning_rate": 0.01,
+      "loss": 2.3486,
+      "step": 3189
+    },
+    {
+      "epoch": 0.32795643686427617,
+      "grad_norm": 0.059109434485435486,
+      "learning_rate": 0.01,
+      "loss": 2.2969,
+      "step": 3192
+    },
+    {
+      "epoch": 0.32826466659817116,
+      "grad_norm": 0.12731850147247314,
+      "learning_rate": 0.01,
+      "loss": 2.3269,
+      "step": 3195
+    },
+    {
+      "epoch": 0.32857289633206616,
+      "grad_norm": 0.15247757732868195,
+      "learning_rate": 0.01,
+      "loss": 2.3312,
+      "step": 3198
+    },
+    {
+      "epoch": 0.32888112606596115,
+      "grad_norm": 0.128463476896286,
+      "learning_rate": 0.01,
+      "loss": 2.3275,
+      "step": 3201
+    },
+    {
+      "epoch": 0.32918935579985614,
+      "grad_norm": 0.09406638145446777,
+      "learning_rate": 0.01,
+      "loss": 2.3205,
+      "step": 3204
+    },
+    {
+      "epoch": 0.32949758553375114,
+      "grad_norm": 0.10524141043424606,
+      "learning_rate": 0.01,
+      "loss": 2.3423,
+      "step": 3207
+    },
+    {
+      "epoch": 0.32980581526764613,
+      "grad_norm": 0.11357913911342621,
+      "learning_rate": 0.01,
+      "loss": 2.3071,
+      "step": 3210
+    },
+    {
+      "epoch": 0.3301140450015411,
+      "grad_norm": 0.06979521363973618,
+      "learning_rate": 0.01,
+      "loss": 2.3319,
+      "step": 3213
+    },
+    {
+      "epoch": 0.33042227473543617,
+      "grad_norm": 0.07000034302473068,
+      "learning_rate": 0.01,
+      "loss": 2.3523,
+      "step": 3216
+    },
+    {
+      "epoch": 0.33073050446933117,
+      "grad_norm": 0.07495003193616867,
+      "learning_rate": 0.01,
+      "loss": 2.305,
+      "step": 3219
+    },
+    {
+      "epoch": 0.33103873420322616,
+      "grad_norm": 0.07131810486316681,
+      "learning_rate": 0.01,
+      "loss": 2.2896,
+      "step": 3222
+    },
+    {
+      "epoch": 0.33134696393712115,
+      "grad_norm": 0.051389019936323166,
+      "learning_rate": 0.01,
+      "loss": 2.2974,
+      "step": 3225
+    },
+    {
+      "epoch": 0.33165519367101615,
+      "grad_norm": 0.05159701779484749,
+      "learning_rate": 0.01,
+      "loss": 2.3344,
+      "step": 3228
+    },
+    {
+      "epoch": 0.33196342340491114,
+      "grad_norm": 0.07632975280284882,
+      "learning_rate": 0.01,
+      "loss": 2.3091,
+      "step": 3231
+    },
+    {
+      "epoch": 0.33227165313880613,
+      "grad_norm": 0.08053800463676453,
+      "learning_rate": 0.01,
+      "loss": 2.298,
+      "step": 3234
+    },
+    {
+      "epoch": 0.3325798828727011,
+      "grad_norm": 0.1371622234582901,
+      "learning_rate": 0.01,
+      "loss": 2.3095,
+      "step": 3237
+    },
+    {
+      "epoch": 0.3328881126065961,
+      "grad_norm": 0.11367069184780121,
+      "learning_rate": 0.01,
+      "loss": 2.3212,
+      "step": 3240
+    },
+    {
+      "epoch": 0.3331963423404911,
+      "grad_norm": 0.13252900540828705,
+      "learning_rate": 0.01,
+      "loss": 2.3238,
+      "step": 3243
+    },
+    {
+      "epoch": 0.3335045720743861,
+      "grad_norm": 0.15517258644104004,
+      "learning_rate": 0.01,
+      "loss": 2.3263,
+      "step": 3246
+    },
+    {
+      "epoch": 0.3338128018082811,
+      "grad_norm": 0.14029370248317719,
+      "learning_rate": 0.01,
+      "loss": 2.3457,
+      "step": 3249
+    },
+    {
+      "epoch": 0.3341210315421761,
+      "grad_norm": 0.105759397149086,
+      "learning_rate": 0.01,
+      "loss": 2.3008,
+      "step": 3252
+    },
+    {
+      "epoch": 0.3344292612760711,
+      "grad_norm": 0.04762979596853256,
+      "learning_rate": 0.01,
+      "loss": 2.3306,
+      "step": 3255
+    },
+    {
+      "epoch": 0.3347374910099661,
+      "grad_norm": 0.12065446376800537,
+      "learning_rate": 0.01,
+      "loss": 2.2904,
+      "step": 3258
+    },
+    {
+      "epoch": 0.3350457207438611,
+      "grad_norm": 0.08886688947677612,
+      "learning_rate": 0.01,
+      "loss": 2.3243,
+      "step": 3261
+    },
+    {
+      "epoch": 0.33535395047775607,
+      "grad_norm": 0.08021339774131775,
+      "learning_rate": 0.01,
+      "loss": 2.3313,
+      "step": 3264
+    },
+    {
+      "epoch": 0.33566218021165106,
+      "grad_norm": 0.04490290582180023,
+      "learning_rate": 0.01,
+      "loss": 2.2888,
+      "step": 3267
+    },
+    {
+      "epoch": 0.33597040994554606,
+      "grad_norm": 0.061480812728405,
+      "learning_rate": 0.01,
+      "loss": 2.2898,
+      "step": 3270
+    },
+    {
+      "epoch": 0.33627863967944105,
+      "grad_norm": 0.04230419546365738,
+      "learning_rate": 0.01,
+      "loss": 2.3062,
+      "step": 3273
+    },
+    {
+      "epoch": 0.3365868694133361,
+      "grad_norm": 0.12344948202371597,
+      "learning_rate": 0.01,
+      "loss": 2.3105,
+      "step": 3276
+    },
+    {
+      "epoch": 0.3368950991472311,
+      "grad_norm": 0.13087160885334015,
+      "learning_rate": 0.01,
+      "loss": 2.3388,
+      "step": 3279
+    },
+    {
+      "epoch": 0.3372033288811261,
+      "grad_norm": 0.06671308726072311,
+      "learning_rate": 0.01,
+      "loss": 2.3062,
+      "step": 3282
+    },
+    {
+      "epoch": 0.3375115586150211,
+      "grad_norm": 0.055828843265771866,
+      "learning_rate": 0.01,
+      "loss": 2.3227,
+      "step": 3285
+    },
+    {
+      "epoch": 0.3378197883489161,
+      "grad_norm": 0.07760481536388397,
+      "learning_rate": 0.01,
+      "loss": 2.307,
+      "step": 3288
+    },
+    {
+      "epoch": 0.33812801808281107,
+      "grad_norm": 0.08074722439050674,
+      "learning_rate": 0.01,
+      "loss": 2.3363,
+      "step": 3291
+    },
+    {
+      "epoch": 0.33843624781670606,
+      "grad_norm": 0.046514566987752914,
+      "learning_rate": 0.01,
+      "loss": 2.3152,
+      "step": 3294
+    },
+    {
+      "epoch": 0.33874447755060105,
+      "grad_norm": 0.15358585119247437,
+      "learning_rate": 0.01,
+      "loss": 2.3114,
+      "step": 3297
+    },
+    {
+      "epoch": 0.33905270728449605,
+      "grad_norm": 0.09048300981521606,
+      "learning_rate": 0.01,
+      "loss": 2.3218,
+      "step": 3300
+    },
+    {
+      "epoch": 0.33936093701839104,
+      "grad_norm": 0.08199465274810791,
+      "learning_rate": 0.01,
+      "loss": 2.3133,
+      "step": 3303
+    },
+    {
+      "epoch": 0.33966916675228603,
+      "grad_norm": 0.13738159835338593,
+      "learning_rate": 0.01,
+      "loss": 2.3108,
+      "step": 3306
+    },
+    {
+      "epoch": 0.339977396486181,
+      "grad_norm": 0.11493804305791855,
+      "learning_rate": 0.01,
+      "loss": 2.2996,
+      "step": 3309
+    },
+    {
+      "epoch": 0.340285626220076,
+      "grad_norm": 0.06872740387916565,
+      "learning_rate": 0.01,
+      "loss": 2.306,
+      "step": 3312
+    },
+    {
+      "epoch": 0.340593855953971,
+      "grad_norm": 0.055139992386102676,
+      "learning_rate": 0.01,
+      "loss": 2.3129,
+      "step": 3315
+    },
+    {
+      "epoch": 0.340902085687866,
+      "grad_norm": 0.16477546095848083,
+      "learning_rate": 0.01,
+      "loss": 2.3138,
+      "step": 3318
+    },
+    {
+      "epoch": 0.341210315421761,
+      "grad_norm": 0.06387230008840561,
+      "learning_rate": 0.01,
+      "loss": 2.3025,
+      "step": 3321
+    },
+    {
+      "epoch": 0.341518545155656,
+      "grad_norm": 0.1657593995332718,
+      "learning_rate": 0.01,
+      "loss": 2.3255,
+      "step": 3324
+    },
+    {
+      "epoch": 0.341826774889551,
+      "grad_norm": 0.08980764448642731,
+      "learning_rate": 0.01,
+      "loss": 2.3024,
+      "step": 3327
+    },
+    {
+      "epoch": 0.342135004623446,
+      "grad_norm": 0.05479981005191803,
+      "learning_rate": 0.01,
+      "loss": 2.2955,
+      "step": 3330
+    },
+    {
+      "epoch": 0.34244323435734103,
+      "grad_norm": 0.05986113101243973,
+      "learning_rate": 0.01,
+      "loss": 2.3078,
+      "step": 3333
+    },
+    {
+      "epoch": 0.342751464091236,
+      "grad_norm": 0.1339874267578125,
+      "learning_rate": 0.01,
+      "loss": 2.2974,
+      "step": 3336
+    },
+    {
+      "epoch": 0.343059693825131,
+      "grad_norm": 0.11250229179859161,
+      "learning_rate": 0.01,
+      "loss": 2.3162,
+      "step": 3339
+    },
+    {
+      "epoch": 0.343367923559026,
+      "grad_norm": 0.12179972976446152,
+      "learning_rate": 0.01,
+      "loss": 2.2746,
+      "step": 3342
+    },
+    {
+      "epoch": 0.343676153292921,
+      "grad_norm": 0.10306143760681152,
+      "learning_rate": 0.01,
+      "loss": 2.3008,
+      "step": 3345
+    },
+    {
+      "epoch": 0.343984383026816,
+      "grad_norm": 0.08372616767883301,
+      "learning_rate": 0.01,
+      "loss": 2.2962,
+      "step": 3348
+    },
+    {
+      "epoch": 0.344292612760711,
+      "grad_norm": 0.05286876857280731,
+      "learning_rate": 0.01,
+      "loss": 2.3067,
+      "step": 3351
+    },
+    {
+      "epoch": 0.344600842494606,
+      "grad_norm": 0.06248036026954651,
+      "learning_rate": 0.01,
+      "loss": 2.3432,
+      "step": 3354
+    },
+    {
+      "epoch": 0.344909072228501,
+      "grad_norm": 0.1287723332643509,
+      "learning_rate": 0.01,
+      "loss": 2.3064,
+      "step": 3357
+    },
+    {
+      "epoch": 0.345217301962396,
+      "grad_norm": 0.08843682706356049,
+      "learning_rate": 0.01,
+      "loss": 2.3059,
+      "step": 3360
+    },
+    {
+      "epoch": 0.34552553169629097,
+      "grad_norm": 0.07060680538415909,
+      "learning_rate": 0.01,
+      "loss": 2.2627,
+      "step": 3363
+    },
+    {
+      "epoch": 0.34583376143018596,
+      "grad_norm": 0.10443838685750961,
+      "learning_rate": 0.01,
+      "loss": 2.2876,
+      "step": 3366
+    },
+    {
+      "epoch": 0.34614199116408095,
+      "grad_norm": 0.06748315691947937,
+      "learning_rate": 0.01,
+      "loss": 2.3182,
+      "step": 3369
+    },
+    {
+      "epoch": 0.34645022089797595,
+      "grad_norm": 0.06599223613739014,
+      "learning_rate": 0.01,
+      "loss": 2.2997,
+      "step": 3372
+    },
+    {
+      "epoch": 0.34675845063187094,
+      "grad_norm": 0.08530016988515854,
+      "learning_rate": 0.01,
+      "loss": 2.2959,
+      "step": 3375
+    },
+    {
+      "epoch": 0.34706668036576593,
+      "grad_norm": 0.10694181174039841,
+      "learning_rate": 0.01,
+      "loss": 2.3248,
+      "step": 3378
+    },
+    {
+      "epoch": 0.3473749100996609,
+      "grad_norm": 0.06598237156867981,
+      "learning_rate": 0.01,
+      "loss": 2.2837,
+      "step": 3381
+    },
+    {
+      "epoch": 0.3476831398335559,
+      "grad_norm": 0.0782204419374466,
+      "learning_rate": 0.01,
+      "loss": 2.2926,
+      "step": 3384
+    },
+    {
+      "epoch": 0.3479913695674509,
+      "grad_norm": 0.09585436433553696,
+      "learning_rate": 0.01,
+      "loss": 2.2984,
+      "step": 3387
+    },
+    {
+      "epoch": 0.34829959930134596,
+      "grad_norm": 0.061477720737457275,
+      "learning_rate": 0.01,
+      "loss": 2.2693,
+      "step": 3390
+    },
+    {
+      "epoch": 0.34860782903524096,
+      "grad_norm": 0.104725681245327,
+      "learning_rate": 0.01,
+      "loss": 2.2887,
+      "step": 3393
+    },
+    {
+      "epoch": 0.34891605876913595,
+      "grad_norm": 0.12205322831869125,
+      "learning_rate": 0.01,
+      "loss": 2.3052,
+      "step": 3396
+    },
+    {
+      "epoch": 0.34922428850303094,
+      "grad_norm": 0.16279913485050201,
+      "learning_rate": 0.01,
+      "loss": 2.2771,
+      "step": 3399
+    },
+    {
+      "epoch": 0.34953251823692594,
+      "grad_norm": 0.059565551578998566,
+      "learning_rate": 0.01,
+      "loss": 2.3027,
+      "step": 3402
+    },
+    {
+      "epoch": 0.34984074797082093,
+      "grad_norm": 0.06318376958370209,
+      "learning_rate": 0.01,
+      "loss": 2.3131,
+      "step": 3405
+    },
+    {
+      "epoch": 0.3501489777047159,
+      "grad_norm": 0.05476443096995354,
+      "learning_rate": 0.01,
+      "loss": 2.2953,
+      "step": 3408
+    },
+    {
+      "epoch": 0.3504572074386109,
+      "grad_norm": 0.07989142090082169,
+      "learning_rate": 0.01,
+      "loss": 2.31,
+      "step": 3411
+    },
+    {
+      "epoch": 0.3507654371725059,
+      "grad_norm": 0.15566086769104004,
+      "learning_rate": 0.01,
+      "loss": 2.2839,
+      "step": 3414
+    },
+    {
+      "epoch": 0.3510736669064009,
+      "grad_norm": 0.060441337525844574,
+      "learning_rate": 0.01,
+      "loss": 2.2952,
+      "step": 3417
+    },
+    {
+      "epoch": 0.3513818966402959,
+      "grad_norm": 0.06277213245630264,
+      "learning_rate": 0.01,
+      "loss": 2.32,
+      "step": 3420
+    },
+    {
+      "epoch": 0.3516901263741909,
+      "grad_norm": 0.04959907755255699,
+      "learning_rate": 0.01,
+      "loss": 2.3116,
+      "step": 3423
+    },
+    {
+      "epoch": 0.3519983561080859,
+      "grad_norm": 0.06766139715909958,
+      "learning_rate": 0.01,
+      "loss": 2.3201,
+      "step": 3426
+    },
+    {
+      "epoch": 0.3523065858419809,
+      "grad_norm": 0.053323931992053986,
+      "learning_rate": 0.01,
+      "loss": 2.3271,
+      "step": 3429
+    },
+    {
+      "epoch": 0.3526148155758759,
+      "grad_norm": 0.06396596878767014,
+      "learning_rate": 0.01,
+      "loss": 2.2929,
+      "step": 3432
+    },
+    {
+      "epoch": 0.35292304530977087,
+      "grad_norm": 0.07360636442899704,
+      "learning_rate": 0.01,
+      "loss": 2.2918,
+      "step": 3435
+    },
+    {
+      "epoch": 0.35323127504366586,
+      "grad_norm": 0.10262563079595566,
+      "learning_rate": 0.01,
+      "loss": 2.2871,
+      "step": 3438
+    },
+    {
+      "epoch": 0.35353950477756085,
+      "grad_norm": 0.09783780574798584,
+      "learning_rate": 0.01,
+      "loss": 2.3229,
+      "step": 3441
+    },
+    {
+      "epoch": 0.35384773451145585,
+      "grad_norm": 0.08542583137750626,
+      "learning_rate": 0.01,
+      "loss": 2.2887,
+      "step": 3444
+    },
+    {
+      "epoch": 0.35415596424535084,
+      "grad_norm": 0.11864805966615677,
+      "learning_rate": 0.01,
+      "loss": 2.2848,
+      "step": 3447
+    },
+    {
+      "epoch": 0.3544641939792459,
+      "grad_norm": 0.10997387021780014,
+      "learning_rate": 0.01,
+      "loss": 2.2897,
+      "step": 3450
+    },
+    {
+      "epoch": 0.3547724237131409,
+      "grad_norm": 0.10915081202983856,
+      "learning_rate": 0.01,
+      "loss": 2.3114,
+      "step": 3453
+    },
+    {
+      "epoch": 0.3550806534470359,
+      "grad_norm": 0.15109725296497345,
+      "learning_rate": 0.01,
+      "loss": 2.2933,
+      "step": 3456
+    },
+    {
+      "epoch": 0.35538888318093087,
+      "grad_norm": 0.04911811649799347,
+      "learning_rate": 0.01,
+      "loss": 2.3035,
+      "step": 3459
+    },
+    {
+      "epoch": 0.35569711291482586,
+      "grad_norm": 0.12352598458528519,
+      "learning_rate": 0.01,
+      "loss": 2.2897,
+      "step": 3462
+    },
+    {
+      "epoch": 0.35600534264872086,
+      "grad_norm": 0.10834213346242905,
+      "learning_rate": 0.01,
+      "loss": 2.2879,
+      "step": 3465
+    },
+    {
+      "epoch": 0.35631357238261585,
+      "grad_norm": 0.10665787756443024,
+      "learning_rate": 0.01,
+      "loss": 2.2614,
+      "step": 3468
+    },
+    {
+      "epoch": 0.35662180211651084,
+      "grad_norm": 0.0898185670375824,
+      "learning_rate": 0.01,
+      "loss": 2.2943,
+      "step": 3471
+    },
+    {
+      "epoch": 0.35693003185040584,
+      "grad_norm": 0.07015782594680786,
+      "learning_rate": 0.01,
+      "loss": 2.298,
+      "step": 3474
+    },
+    {
+      "epoch": 0.35723826158430083,
+      "grad_norm": 0.1292288452386856,
+      "learning_rate": 0.01,
+      "loss": 2.3122,
+      "step": 3477
+    },
+    {
+      "epoch": 0.3575464913181958,
+      "grad_norm": 0.09300121665000916,
+      "learning_rate": 0.01,
+      "loss": 2.2769,
+      "step": 3480
+    },
+    {
+      "epoch": 0.3578547210520908,
+      "grad_norm": 0.0449809767305851,
+      "learning_rate": 0.01,
+      "loss": 2.2564,
+      "step": 3483
+    },
+    {
+      "epoch": 0.3581629507859858,
+      "grad_norm": 0.051362160593271255,
+      "learning_rate": 0.01,
+      "loss": 2.2739,
+      "step": 3486
+    },
+    {
+      "epoch": 0.3584711805198808,
+      "grad_norm": 0.12473469972610474,
+      "learning_rate": 0.01,
+      "loss": 2.2844,
+      "step": 3489
+    },
+    {
+      "epoch": 0.3587794102537758,
+      "grad_norm": 0.0925057902932167,
+      "learning_rate": 0.01,
+      "loss": 2.2618,
+      "step": 3492
+    },
+    {
+      "epoch": 0.3590876399876708,
+      "grad_norm": 0.1026608943939209,
+      "learning_rate": 0.01,
+      "loss": 2.2814,
+      "step": 3495
+    },
+    {
+      "epoch": 0.3593958697215658,
+      "grad_norm": 0.0995681881904602,
+      "learning_rate": 0.01,
+      "loss": 2.2861,
+      "step": 3498
+    },
+    {
+      "epoch": 0.3597040994554608,
+      "grad_norm": 0.06513385474681854,
+      "learning_rate": 0.01,
+      "loss": 2.2827,
+      "step": 3501
+    },
+    {
+      "epoch": 0.3600123291893558,
+      "grad_norm": 0.06724824756383896,
+      "learning_rate": 0.01,
+      "loss": 2.2799,
+      "step": 3504
+    },
+    {
+      "epoch": 0.3603205589232508,
+      "grad_norm": 0.06367610394954681,
+      "learning_rate": 0.01,
+      "loss": 2.2846,
+      "step": 3507
+    },
+    {
+      "epoch": 0.3606287886571458,
+      "grad_norm": 0.07489916682243347,
+      "learning_rate": 0.01,
+      "loss": 2.2816,
+      "step": 3510
+    },
+    {
+      "epoch": 0.3609370183910408,
+      "grad_norm": 0.11221667379140854,
+      "learning_rate": 0.01,
+      "loss": 2.2869,
+      "step": 3513
+    },
+    {
+      "epoch": 0.3612452481249358,
+      "grad_norm": 0.09854032099246979,
+      "learning_rate": 0.01,
+      "loss": 2.2646,
+      "step": 3516
+    },
+    {
+      "epoch": 0.3615534778588308,
+      "grad_norm": 0.09218656271696091,
+      "learning_rate": 0.01,
+      "loss": 2.2844,
+      "step": 3519
+    },
+    {
+      "epoch": 0.3618617075927258,
+      "grad_norm": 0.1531379073858261,
+      "learning_rate": 0.01,
+      "loss": 2.279,
+      "step": 3522
+    },
+    {
+      "epoch": 0.3621699373266208,
+      "grad_norm": 0.07070820778608322,
+      "learning_rate": 0.01,
+      "loss": 2.2747,
+      "step": 3525
+    },
+    {
+      "epoch": 0.3624781670605158,
+      "grad_norm": 0.1057102233171463,
+      "learning_rate": 0.01,
+      "loss": 2.275,
+      "step": 3528
+    },
+    {
+      "epoch": 0.36278639679441077,
+      "grad_norm": 0.049471016973257065,
+      "learning_rate": 0.01,
+      "loss": 2.3013,
+      "step": 3531
+    },
+    {
+      "epoch": 0.36309462652830576,
+      "grad_norm": 0.08196526020765305,
+      "learning_rate": 0.01,
+      "loss": 2.2571,
+      "step": 3534
+    },
+    {
+      "epoch": 0.36340285626220076,
+      "grad_norm": 0.09507983922958374,
+      "learning_rate": 0.01,
+      "loss": 2.3196,
+      "step": 3537
+    },
+    {
+      "epoch": 0.36371108599609575,
+      "grad_norm": 0.089228555560112,
+      "learning_rate": 0.01,
+      "loss": 2.2539,
+      "step": 3540
+    },
+    {
+      "epoch": 0.36401931572999074,
+      "grad_norm": 0.0866270586848259,
+      "learning_rate": 0.01,
+      "loss": 2.284,
+      "step": 3543
+    },
+    {
+      "epoch": 0.36432754546388574,
+      "grad_norm": 0.13805072009563446,
+      "learning_rate": 0.01,
+      "loss": 2.2723,
+      "step": 3546
+    },
+    {
+      "epoch": 0.36463577519778073,
+      "grad_norm": 0.09308724105358124,
+      "learning_rate": 0.01,
+      "loss": 2.2969,
+      "step": 3549
+    },
+    {
+      "epoch": 0.3649440049316757,
+      "grad_norm": 0.07004178315401077,
+      "learning_rate": 0.01,
+      "loss": 2.2959,
+      "step": 3552
+    },
+    {
+      "epoch": 0.3652522346655707,
+      "grad_norm": 0.09345975518226624,
+      "learning_rate": 0.01,
+      "loss": 2.2656,
+      "step": 3555
+    },
+    {
+      "epoch": 0.3655604643994657,
+      "grad_norm": 0.07694482058286667,
+      "learning_rate": 0.01,
+      "loss": 2.2921,
+      "step": 3558
+    },
+    {
+      "epoch": 0.3658686941333607,
+      "grad_norm": 0.05591150000691414,
+      "learning_rate": 0.01,
+      "loss": 2.2869,
+      "step": 3561
+    },
+    {
+      "epoch": 0.36617692386725575,
+      "grad_norm": 0.06863993406295776,
+      "learning_rate": 0.01,
+      "loss": 2.2897,
+      "step": 3564
+    },
+    {
+      "epoch": 0.36648515360115075,
+      "grad_norm": 0.06258527934551239,
+      "learning_rate": 0.01,
+      "loss": 2.2994,
+      "step": 3567
+    },
+    {
+      "epoch": 0.36679338333504574,
+      "grad_norm": 0.1049329936504364,
+      "learning_rate": 0.01,
+      "loss": 2.2794,
+      "step": 3570
+    },
+    {
+      "epoch": 0.36710161306894074,
+      "grad_norm": 0.1229025200009346,
+      "learning_rate": 0.01,
+      "loss": 2.2949,
+      "step": 3573
+    },
+    {
+      "epoch": 0.36740984280283573,
+      "grad_norm": 0.13274389505386353,
+      "learning_rate": 0.01,
+      "loss": 2.2791,
+      "step": 3576
+    },
+    {
+      "epoch": 0.3677180725367307,
+      "grad_norm": 0.09388844668865204,
+      "learning_rate": 0.01,
+      "loss": 2.3067,
+      "step": 3579
+    },
+    {
+      "epoch": 0.3680263022706257,
+      "grad_norm": 0.05375714227557182,
+      "learning_rate": 0.01,
+      "loss": 2.2946,
+      "step": 3582
+    },
+    {
+      "epoch": 0.3683345320045207,
+      "grad_norm": 0.059105634689331055,
+      "learning_rate": 0.01,
+      "loss": 2.2821,
+      "step": 3585
+    },
+    {
+      "epoch": 0.3686427617384157,
+      "grad_norm": 0.055578552186489105,
+      "learning_rate": 0.01,
+      "loss": 2.2694,
+      "step": 3588
+    },
+    {
+      "epoch": 0.3689509914723107,
+      "grad_norm": 0.08778764307498932,
+      "learning_rate": 0.01,
+      "loss": 2.2712,
+      "step": 3591
+    },
+    {
+      "epoch": 0.3692592212062057,
+      "grad_norm": 0.1044803187251091,
+      "learning_rate": 0.01,
+      "loss": 2.2797,
+      "step": 3594
+    },
+    {
+      "epoch": 0.3695674509401007,
+      "grad_norm": 0.15398399531841278,
+      "learning_rate": 0.01,
+      "loss": 2.3042,
+      "step": 3597
+    },
+    {
+      "epoch": 0.3698756806739957,
+      "grad_norm": 0.11562564969062805,
+      "learning_rate": 0.01,
+      "loss": 2.2609,
+      "step": 3600
+    },
+    {
+      "epoch": 0.37018391040789067,
+      "grad_norm": 0.060630831867456436,
+      "learning_rate": 0.01,
+      "loss": 2.2663,
+      "step": 3603
+    },
+    {
+      "epoch": 0.37049214014178566,
+      "grad_norm": 0.0576477013528347,
+      "learning_rate": 0.01,
+      "loss": 2.2974,
+      "step": 3606
+    },
+    {
+      "epoch": 0.37080036987568066,
+      "grad_norm": 0.059915438294410706,
+      "learning_rate": 0.01,
+      "loss": 2.3031,
+      "step": 3609
+    },
+    {
+      "epoch": 0.37110859960957565,
+      "grad_norm": 0.10807155817747116,
+      "learning_rate": 0.01,
+      "loss": 2.2739,
+      "step": 3612
+    },
+    {
+      "epoch": 0.37141682934347064,
+      "grad_norm": 0.09196165949106216,
+      "learning_rate": 0.01,
+      "loss": 2.3054,
+      "step": 3615
+    },
+    {
+      "epoch": 0.37172505907736564,
+      "grad_norm": 0.07379795610904694,
+      "learning_rate": 0.01,
+      "loss": 2.2805,
+      "step": 3618
+    },
+    {
+      "epoch": 0.3720332888112607,
+      "grad_norm": 0.06034912168979645,
+      "learning_rate": 0.01,
+      "loss": 2.2549,
+      "step": 3621
+    },
+    {
+      "epoch": 0.3723415185451557,
+      "grad_norm": 0.13983361423015594,
+      "learning_rate": 0.01,
+      "loss": 2.269,
+      "step": 3624
+    },
+    {
+      "epoch": 0.3726497482790507,
+      "grad_norm": 0.11592069268226624,
+      "learning_rate": 0.01,
+      "loss": 2.2903,
+      "step": 3627
+    },
+    {
+      "epoch": 0.37295797801294567,
+      "grad_norm": 0.15428505837917328,
+      "learning_rate": 0.01,
+      "loss": 2.2918,
+      "step": 3630
+    },
+    {
+      "epoch": 0.37326620774684066,
+      "grad_norm": 0.19936774671077728,
+      "learning_rate": 0.01,
+      "loss": 2.2782,
+      "step": 3633
+    },
+    {
+      "epoch": 0.37357443748073565,
+      "grad_norm": 0.15364627540111542,
+      "learning_rate": 0.01,
+      "loss": 2.2736,
+      "step": 3636
+    },
+    {
+      "epoch": 0.37388266721463065,
+      "grad_norm": 0.047554273158311844,
+      "learning_rate": 0.01,
+      "loss": 2.3172,
+      "step": 3639
+    },
+    {
+      "epoch": 0.37419089694852564,
+      "grad_norm": 0.0555570051074028,
+      "learning_rate": 0.01,
+      "loss": 2.2731,
+      "step": 3642
+    },
+    {
+      "epoch": 0.37449912668242064,
+      "grad_norm": 0.052204012870788574,
+      "learning_rate": 0.01,
+      "loss": 2.281,
+      "step": 3645
+    },
+    {
+      "epoch": 0.37480735641631563,
+      "grad_norm": 0.09206510335206985,
+      "learning_rate": 0.01,
+      "loss": 2.2639,
+      "step": 3648
+    },
+    {
+      "epoch": 0.3751155861502106,
+      "grad_norm": 0.1199311912059784,
+      "learning_rate": 0.01,
+      "loss": 2.2873,
+      "step": 3651
+    },
+    {
+      "epoch": 0.3754238158841056,
+      "grad_norm": 0.08949270099401474,
+      "learning_rate": 0.01,
+      "loss": 2.2668,
+      "step": 3654
+    },
+    {
+      "epoch": 0.3757320456180006,
+      "grad_norm": 0.08521883934736252,
+      "learning_rate": 0.01,
+      "loss": 2.247,
+      "step": 3657
+    },
+    {
+      "epoch": 0.3760402753518956,
+      "grad_norm": 0.07689694315195084,
+      "learning_rate": 0.01,
+      "loss": 2.2904,
+      "step": 3660
+    },
+    {
+      "epoch": 0.3763485050857906,
+      "grad_norm": 0.08761987835168839,
+      "learning_rate": 0.01,
+      "loss": 2.2761,
+      "step": 3663
+    },
+    {
+      "epoch": 0.3766567348196856,
+      "grad_norm": 0.056420013308525085,
+      "learning_rate": 0.01,
+      "loss": 2.259,
+      "step": 3666
+    },
+    {
+      "epoch": 0.3769649645535806,
+      "grad_norm": 0.06192856654524803,
+      "learning_rate": 0.01,
+      "loss": 2.2294,
+      "step": 3669
+    },
+    {
+      "epoch": 0.3772731942874756,
+      "grad_norm": 0.1021333709359169,
+      "learning_rate": 0.01,
+      "loss": 2.2649,
+      "step": 3672
+    },
+    {
+      "epoch": 0.37758142402137057,
+      "grad_norm": 0.10071670264005661,
+      "learning_rate": 0.01,
+      "loss": 2.2584,
+      "step": 3675
+    },
+    {
+      "epoch": 0.37788965375526556,
+      "grad_norm": 0.05968625843524933,
+      "learning_rate": 0.01,
+      "loss": 2.2699,
+      "step": 3678
+    },
+    {
+      "epoch": 0.3781978834891606,
+      "grad_norm": 0.07489661872386932,
+      "learning_rate": 0.01,
+      "loss": 2.2663,
+      "step": 3681
+    },
+    {
+      "epoch": 0.3785061132230556,
+      "grad_norm": 0.07880943268537521,
+      "learning_rate": 0.01,
+      "loss": 2.2709,
+      "step": 3684
+    },
+    {
+      "epoch": 0.3788143429569506,
+      "grad_norm": 0.055632054805755615,
+      "learning_rate": 0.01,
+      "loss": 2.272,
+      "step": 3687
+    },
+    {
+      "epoch": 0.3791225726908456,
+      "grad_norm": 0.05365302786231041,
+      "learning_rate": 0.01,
+      "loss": 2.2268,
+      "step": 3690
+    },
+    {
+      "epoch": 0.3794308024247406,
+      "grad_norm": 0.0802481397986412,
+      "learning_rate": 0.01,
+      "loss": 2.2631,
+      "step": 3693
+    },
+    {
+      "epoch": 0.3797390321586356,
+      "grad_norm": 0.1312764585018158,
+      "learning_rate": 0.01,
+      "loss": 2.2985,
+      "step": 3696
+    },
+    {
+      "epoch": 0.3800472618925306,
+      "grad_norm": 0.14543971419334412,
+      "learning_rate": 0.01,
+      "loss": 2.25,
+      "step": 3699
+    },
+    {
+      "epoch": 0.38035549162642557,
+      "grad_norm": 0.05727002024650574,
+      "learning_rate": 0.01,
+      "loss": 2.2556,
+      "step": 3702
+    },
+    {
+      "epoch": 0.38066372136032056,
+      "grad_norm": 0.07309607416391373,
+      "learning_rate": 0.01,
+      "loss": 2.2574,
+      "step": 3705
+    },
+    {
+      "epoch": 0.38097195109421556,
+      "grad_norm": 0.03849095106124878,
+      "learning_rate": 0.01,
+      "loss": 2.2501,
+      "step": 3708
+    },
+    {
+      "epoch": 0.38128018082811055,
+      "grad_norm": 0.0623021237552166,
+      "learning_rate": 0.01,
+      "loss": 2.2672,
+      "step": 3711
+    },
+    {
+      "epoch": 0.38158841056200554,
+      "grad_norm": 0.08916610479354858,
+      "learning_rate": 0.01,
+      "loss": 2.2683,
+      "step": 3714
+    },
+    {
+      "epoch": 0.38189664029590054,
+      "grad_norm": 0.08126388490200043,
+      "learning_rate": 0.01,
+      "loss": 2.2574,
+      "step": 3717
+    },
+    {
+      "epoch": 0.38220487002979553,
+      "grad_norm": 0.07121114432811737,
+      "learning_rate": 0.01,
+      "loss": 2.2358,
+      "step": 3720
+    },
+    {
+      "epoch": 0.3825130997636905,
+      "grad_norm": 0.07406505942344666,
+      "learning_rate": 0.01,
+      "loss": 2.2736,
+      "step": 3723
+    },
+    {
+      "epoch": 0.3828213294975855,
+      "grad_norm": 0.13355331122875214,
+      "learning_rate": 0.01,
+      "loss": 2.2685,
+      "step": 3726
+    },
+    {
+      "epoch": 0.3831295592314805,
+      "grad_norm": 0.05672430619597435,
+      "learning_rate": 0.01,
+      "loss": 2.2913,
+      "step": 3729
+    },
+    {
+      "epoch": 0.3834377889653755,
+      "grad_norm": 0.047647468745708466,
+      "learning_rate": 0.01,
+      "loss": 2.2533,
+      "step": 3732
+    },
+    {
+      "epoch": 0.3837460186992705,
+      "grad_norm": 0.059008341282606125,
+      "learning_rate": 0.01,
+      "loss": 2.2867,
+      "step": 3735
+    },
+    {
+      "epoch": 0.38405424843316555,
+      "grad_norm": 0.06551840156316757,
+      "learning_rate": 0.01,
+      "loss": 2.2742,
+      "step": 3738
+    },
+    {
+      "epoch": 0.38436247816706054,
+      "grad_norm": 0.08781883865594864,
+      "learning_rate": 0.01,
+      "loss": 2.2427,
+      "step": 3741
+    },
+    {
+      "epoch": 0.38467070790095553,
+      "grad_norm": 0.06808102875947952,
+      "learning_rate": 0.01,
+      "loss": 2.2493,
+      "step": 3744
+    },
+    {
+      "epoch": 0.3849789376348505,
+      "grad_norm": 0.06570697575807571,
+      "learning_rate": 0.01,
+      "loss": 2.2445,
+      "step": 3747
+    },
+    {
+      "epoch": 0.3852871673687455,
+      "grad_norm": 0.08742080628871918,
+      "learning_rate": 0.01,
+      "loss": 2.2576,
+      "step": 3750
+    },
+    {
+      "epoch": 0.3855953971026405,
+      "grad_norm": 0.1518019735813141,
+      "learning_rate": 0.01,
+      "loss": 2.2819,
+      "step": 3753
+    },
+    {
+      "epoch": 0.3859036268365355,
+      "grad_norm": 0.10349754244089127,
+      "learning_rate": 0.01,
+      "loss": 2.2465,
+      "step": 3756
+    },
+    {
+      "epoch": 0.3862118565704305,
+      "grad_norm": 0.06008581072092056,
+      "learning_rate": 0.01,
+      "loss": 2.2817,
+      "step": 3759
+    },
+    {
+      "epoch": 0.3865200863043255,
+      "grad_norm": 0.0450257770717144,
+      "learning_rate": 0.01,
+      "loss": 2.2585,
+      "step": 3762
+    },
+    {
+      "epoch": 0.3868283160382205,
+      "grad_norm": 0.04145176708698273,
+      "learning_rate": 0.01,
+      "loss": 2.2634,
+      "step": 3765
+    },
+    {
+      "epoch": 0.3871365457721155,
+      "grad_norm": 0.17084141075611115,
+      "learning_rate": 0.01,
+      "loss": 2.2355,
+      "step": 3768
+    },
+    {
+      "epoch": 0.3874447755060105,
+      "grad_norm": 0.06679602712392807,
+      "learning_rate": 0.01,
+      "loss": 2.2737,
+      "step": 3771
+    },
+    {
+      "epoch": 0.38775300523990547,
+      "grad_norm": 0.05363382026553154,
+      "learning_rate": 0.01,
+      "loss": 2.244,
+      "step": 3774
+    },
+    {
+      "epoch": 0.38806123497380046,
+      "grad_norm": 0.05722133815288544,
+      "learning_rate": 0.01,
+      "loss": 2.2515,
+      "step": 3777
+    },
+    {
+      "epoch": 0.38836946470769546,
+      "grad_norm": 0.06288215517997742,
+      "learning_rate": 0.01,
+      "loss": 2.2625,
+      "step": 3780
+    },
+    {
+      "epoch": 0.38867769444159045,
+      "grad_norm": 0.05087801814079285,
+      "learning_rate": 0.01,
+      "loss": 2.2883,
+      "step": 3783
+    },
+    {
+      "epoch": 0.38898592417548544,
+      "grad_norm": 0.08160998672246933,
+      "learning_rate": 0.01,
+      "loss": 2.2462,
+      "step": 3786
+    },
+    {
+      "epoch": 0.38929415390938044,
+      "grad_norm": 0.22291240096092224,
+      "learning_rate": 0.01,
+      "loss": 2.2613,
+      "step": 3789
+    },
+    {
+      "epoch": 0.38960238364327543,
+      "grad_norm": 0.11482773721218109,
+      "learning_rate": 0.01,
+      "loss": 2.2633,
+      "step": 3792
+    },
+    {
+      "epoch": 0.3899106133771705,
+      "grad_norm": 0.056299589574337006,
+      "learning_rate": 0.01,
+      "loss": 2.2896,
+      "step": 3795
+    },
+    {
+      "epoch": 0.39021884311106547,
+      "grad_norm": 0.04524017125368118,
+      "learning_rate": 0.01,
+      "loss": 2.2543,
+      "step": 3798
+    },
+    {
+      "epoch": 0.39052707284496047,
+      "grad_norm": 0.0903107225894928,
+      "learning_rate": 0.01,
+      "loss": 2.2801,
+      "step": 3801
+    },
+    {
+      "epoch": 0.39083530257885546,
+      "grad_norm": 0.0645504966378212,
+      "learning_rate": 0.01,
+      "loss": 2.2628,
+      "step": 3804
+    },
+    {
+      "epoch": 0.39114353231275045,
+      "grad_norm": 0.06752094626426697,
+      "learning_rate": 0.01,
+      "loss": 2.2732,
+      "step": 3807
+    },
+    {
+      "epoch": 0.39145176204664545,
+      "grad_norm": 0.04459339380264282,
+      "learning_rate": 0.01,
+      "loss": 2.2601,
+      "step": 3810
+    },
+    {
+      "epoch": 0.39175999178054044,
+      "grad_norm": 0.07300913333892822,
+      "learning_rate": 0.01,
+      "loss": 2.2437,
+      "step": 3813
+    },
+    {
+      "epoch": 0.39206822151443543,
+      "grad_norm": 0.16804097592830658,
+      "learning_rate": 0.01,
+      "loss": 2.26,
+      "step": 3816
+    },
+    {
+      "epoch": 0.3923764512483304,
+      "grad_norm": 0.10682248324155807,
+      "learning_rate": 0.01,
+      "loss": 2.2764,
+      "step": 3819
+    },
+    {
+      "epoch": 0.3926846809822254,
+      "grad_norm": 0.046895258128643036,
+      "learning_rate": 0.01,
+      "loss": 2.2654,
+      "step": 3822
+    },
+    {
+      "epoch": 0.3929929107161204,
+      "grad_norm": 0.05799179524183273,
+      "learning_rate": 0.01,
+      "loss": 2.2254,
+      "step": 3825
+    },
+    {
+      "epoch": 0.3933011404500154,
+      "grad_norm": 0.0474528968334198,
+      "learning_rate": 0.01,
+      "loss": 2.2604,
+      "step": 3828
+    },
+    {
+      "epoch": 0.3936093701839104,
+      "grad_norm": 0.1437537968158722,
+      "learning_rate": 0.01,
+      "loss": 2.2532,
+      "step": 3831
+    },
+    {
+      "epoch": 0.3939175999178054,
+      "grad_norm": 0.06202014535665512,
+      "learning_rate": 0.01,
+      "loss": 2.2486,
+      "step": 3834
+    },
+    {
+      "epoch": 0.3942258296517004,
+      "grad_norm": 0.09379147738218307,
+      "learning_rate": 0.01,
+      "loss": 2.2602,
+      "step": 3837
+    },
+    {
+      "epoch": 0.3945340593855954,
+      "grad_norm": 0.07898830622434616,
+      "learning_rate": 0.01,
+      "loss": 2.2605,
+      "step": 3840
+    },
+    {
+      "epoch": 0.3948422891194904,
+      "grad_norm": 0.10186600685119629,
+      "learning_rate": 0.01,
+      "loss": 2.2807,
+      "step": 3843
+    },
+    {
+      "epoch": 0.39515051885338537,
+      "grad_norm": 0.08611535280942917,
+      "learning_rate": 0.01,
+      "loss": 2.2571,
+      "step": 3846
+    },
+    {
+      "epoch": 0.39545874858728036,
+      "grad_norm": 0.10435480624437332,
+      "learning_rate": 0.01,
+      "loss": 2.2721,
+      "step": 3849
+    },
+    {
+      "epoch": 0.39576697832117536,
+      "grad_norm": 0.11543019860982895,
+      "learning_rate": 0.01,
+      "loss": 2.2598,
+      "step": 3852
+    },
+    {
+      "epoch": 0.3960752080550704,
+      "grad_norm": 0.11996404081583023,
+      "learning_rate": 0.01,
+      "loss": 2.2536,
+      "step": 3855
+    },
+    {
+      "epoch": 0.3963834377889654,
+      "grad_norm": 0.05615765228867531,
+      "learning_rate": 0.01,
+      "loss": 2.2637,
+      "step": 3858
+    },
+    {
+      "epoch": 0.3966916675228604,
+      "grad_norm": 0.06568838655948639,
+      "learning_rate": 0.01,
+      "loss": 2.2756,
+      "step": 3861
+    },
+    {
+      "epoch": 0.3969998972567554,
+      "grad_norm": 0.07747132331132889,
+      "learning_rate": 0.01,
+      "loss": 2.2816,
+      "step": 3864
+    },
+    {
+      "epoch": 0.3973081269906504,
+      "grad_norm": 0.057373497635126114,
+      "learning_rate": 0.01,
+      "loss": 2.255,
+      "step": 3867
+    },
+    {
+      "epoch": 0.3976163567245454,
+      "grad_norm": 0.11501277983188629,
+      "learning_rate": 0.01,
+      "loss": 2.2494,
+      "step": 3870
+    },
+    {
+      "epoch": 0.39792458645844037,
+      "grad_norm": 0.07761958241462708,
+      "learning_rate": 0.01,
+      "loss": 2.2459,
+      "step": 3873
+    },
+    {
+      "epoch": 0.39823281619233536,
+      "grad_norm": 0.06263428926467896,
+      "learning_rate": 0.01,
+      "loss": 2.2649,
+      "step": 3876
+    },
+    {
+      "epoch": 0.39854104592623035,
+      "grad_norm": 0.04552373290061951,
+      "learning_rate": 0.01,
+      "loss": 2.2578,
+      "step": 3879
+    },
+    {
+      "epoch": 0.39884927566012535,
+      "grad_norm": 0.0631655901670456,
+      "learning_rate": 0.01,
+      "loss": 2.2648,
+      "step": 3882
+    },
+    {
+      "epoch": 0.39915750539402034,
+      "grad_norm": 0.06519417464733124,
+      "learning_rate": 0.01,
+      "loss": 2.2438,
+      "step": 3885
+    },
+    {
+      "epoch": 0.39946573512791533,
+      "grad_norm": 0.10446424037218094,
+      "learning_rate": 0.01,
+      "loss": 2.2815,
+      "step": 3888
+    },
+    {
+      "epoch": 0.3997739648618103,
+      "grad_norm": 0.07533372938632965,
+      "learning_rate": 0.01,
+      "loss": 2.272,
+      "step": 3891
+    },
+    {
+      "epoch": 0.4000821945957053,
+      "grad_norm": 0.05748215690255165,
+      "learning_rate": 0.01,
+      "loss": 2.2971,
+      "step": 3894
+    },
+    {
+      "epoch": 0.4003904243296003,
+      "grad_norm": 0.051343973726034164,
+      "learning_rate": 0.01,
+      "loss": 2.2316,
+      "step": 3897
+    },
+    {
+      "epoch": 0.4006986540634953,
+      "grad_norm": 0.04799075797200203,
+      "learning_rate": 0.01,
+      "loss": 2.2333,
+      "step": 3900
+    },
+    {
+      "epoch": 0.4010068837973903,
+      "grad_norm": 0.12885436415672302,
+      "learning_rate": 0.01,
+      "loss": 2.247,
+      "step": 3903
+    },
+    {
+      "epoch": 0.4013151135312853,
+      "grad_norm": 0.07175249606370926,
+      "learning_rate": 0.01,
+      "loss": 2.2407,
+      "step": 3906
+    },
+    {
+      "epoch": 0.4016233432651803,
+      "grad_norm": 0.10784266144037247,
+      "learning_rate": 0.01,
+      "loss": 2.2458,
+      "step": 3909
+    },
+    {
+      "epoch": 0.40193157299907534,
+      "grad_norm": 0.08646712452173233,
+      "learning_rate": 0.01,
+      "loss": 2.2571,
+      "step": 3912
+    },
+    {
+      "epoch": 0.40223980273297033,
+      "grad_norm": 0.05365338176488876,
+      "learning_rate": 0.01,
+      "loss": 2.2585,
+      "step": 3915
+    },
+    {
+      "epoch": 0.4025480324668653,
+      "grad_norm": 0.07037780433893204,
+      "learning_rate": 0.01,
+      "loss": 2.2277,
+      "step": 3918
+    },
+    {
+      "epoch": 0.4028562622007603,
+      "grad_norm": 0.040290024131536484,
+      "learning_rate": 0.01,
+      "loss": 2.2508,
+      "step": 3921
+    },
+    {
+      "epoch": 0.4031644919346553,
+      "grad_norm": 0.050338853150606155,
+      "learning_rate": 0.01,
+      "loss": 2.2356,
+      "step": 3924
+    },
+    {
+      "epoch": 0.4034727216685503,
+      "grad_norm": 0.1420246660709381,
+      "learning_rate": 0.01,
+      "loss": 2.2531,
+      "step": 3927
+    },
+    {
+      "epoch": 0.4037809514024453,
+      "grad_norm": 0.07432923465967178,
+      "learning_rate": 0.01,
+      "loss": 2.2766,
+      "step": 3930
+    },
+    {
+      "epoch": 0.4040891811363403,
+      "grad_norm": 0.04954257979989052,
+      "learning_rate": 0.01,
+      "loss": 2.2825,
+      "step": 3933
+    },
+    {
+      "epoch": 0.4043974108702353,
+      "grad_norm": 0.05988876149058342,
+      "learning_rate": 0.01,
+      "loss": 2.2342,
+      "step": 3936
+    },
+    {
+      "epoch": 0.4047056406041303,
+      "grad_norm": 0.09800540655851364,
+      "learning_rate": 0.01,
+      "loss": 2.2268,
+      "step": 3939
+    },
+    {
+      "epoch": 0.4050138703380253,
+      "grad_norm": 0.09171874821186066,
+      "learning_rate": 0.01,
+      "loss": 2.2648,
+      "step": 3942
+    },
+    {
+      "epoch": 0.40532210007192027,
+      "grad_norm": 0.07430606335401535,
+      "learning_rate": 0.01,
+      "loss": 2.2523,
+      "step": 3945
+    },
+    {
+      "epoch": 0.40563032980581526,
+      "grad_norm": 0.043649185448884964,
+      "learning_rate": 0.01,
+      "loss": 2.2303,
+      "step": 3948
+    },
+    {
+      "epoch": 0.40593855953971025,
+      "grad_norm": 0.04120480641722679,
+      "learning_rate": 0.01,
+      "loss": 2.2299,
+      "step": 3951
+    },
+    {
+      "epoch": 0.40624678927360525,
+      "grad_norm": 0.0692945346236229,
+      "learning_rate": 0.01,
+      "loss": 2.2466,
+      "step": 3954
+    },
+    {
+      "epoch": 0.40655501900750024,
+      "grad_norm": 0.08884318917989731,
+      "learning_rate": 0.01,
+      "loss": 2.2802,
+      "step": 3957
+    },
+    {
+      "epoch": 0.40686324874139523,
+      "grad_norm": 0.05542384088039398,
+      "learning_rate": 0.01,
+      "loss": 2.2303,
+      "step": 3960
+    },
+    {
+      "epoch": 0.4071714784752902,
+      "grad_norm": 0.08013599365949631,
+      "learning_rate": 0.01,
+      "loss": 2.2361,
+      "step": 3963
+    },
+    {
+      "epoch": 0.4074797082091852,
+      "grad_norm": 0.15963242948055267,
+      "learning_rate": 0.01,
+      "loss": 2.2608,
+      "step": 3966
+    },
+    {
+      "epoch": 0.40778793794308027,
+      "grad_norm": 0.05428241938352585,
+      "learning_rate": 0.01,
+      "loss": 2.2415,
+      "step": 3969
+    },
+    {
+      "epoch": 0.40809616767697526,
+      "grad_norm": 0.09297880530357361,
+      "learning_rate": 0.01,
+      "loss": 2.2804,
+      "step": 3972
+    },
+    {
+      "epoch": 0.40840439741087026,
+      "grad_norm": 0.11259882897138596,
+      "learning_rate": 0.01,
+      "loss": 2.2562,
+      "step": 3975
+    },
+    {
+      "epoch": 0.40871262714476525,
+      "grad_norm": 0.0546397790312767,
+      "learning_rate": 0.01,
+      "loss": 2.2423,
+      "step": 3978
+    },
+    {
+      "epoch": 0.40902085687866024,
+      "grad_norm": 0.13870957493782043,
+      "learning_rate": 0.01,
+      "loss": 2.2431,
+      "step": 3981
+    },
+    {
+      "epoch": 0.40932908661255524,
+      "grad_norm": 0.05527504161000252,
+      "learning_rate": 0.01,
+      "loss": 2.2649,
+      "step": 3984
+    },
+    {
+      "epoch": 0.40963731634645023,
+      "grad_norm": 0.08060980588197708,
+      "learning_rate": 0.01,
+      "loss": 2.2708,
+      "step": 3987
+    },
+    {
+      "epoch": 0.4099455460803452,
+      "grad_norm": 0.05611690506339073,
+      "learning_rate": 0.01,
+      "loss": 2.2683,
+      "step": 3990
+    },
+    {
+      "epoch": 0.4102537758142402,
+      "grad_norm": 0.08760816603899002,
+      "learning_rate": 0.01,
+      "loss": 2.2392,
+      "step": 3993
+    },
+    {
+      "epoch": 0.4105620055481352,
+      "grad_norm": 0.07327746599912643,
+      "learning_rate": 0.01,
+      "loss": 2.2587,
+      "step": 3996
+    },
+    {
+      "epoch": 0.4108702352820302,
+      "grad_norm": 0.05924748629331589,
+      "learning_rate": 0.01,
+      "loss": 2.2435,
+      "step": 3999
+    },
+    {
+      "epoch": 0.4111784650159252,
+      "grad_norm": 0.08269370347261429,
+      "learning_rate": 0.01,
+      "loss": 2.2365,
+      "step": 4002
+    },
+    {
+      "epoch": 0.4114866947498202,
+      "grad_norm": 0.06834371387958527,
+      "learning_rate": 0.01,
+      "loss": 2.2579,
+      "step": 4005
+    },
+    {
+      "epoch": 0.4117949244837152,
+      "grad_norm": 0.06737885624170303,
+      "learning_rate": 0.01,
+      "loss": 2.2585,
+      "step": 4008
+    },
+    {
+      "epoch": 0.4121031542176102,
+      "grad_norm": 0.0919148176908493,
+      "learning_rate": 0.01,
+      "loss": 2.2524,
+      "step": 4011
+    },
+    {
+      "epoch": 0.4124113839515052,
+      "grad_norm": 0.0744348093867302,
+      "learning_rate": 0.01,
+      "loss": 2.2328,
+      "step": 4014
+    },
+    {
+      "epoch": 0.41271961368540017,
+      "grad_norm": 0.08952994644641876,
+      "learning_rate": 0.01,
+      "loss": 2.2556,
+      "step": 4017
+    },
+    {
+      "epoch": 0.41302784341929516,
+      "grad_norm": 0.054230738431215286,
+      "learning_rate": 0.01,
+      "loss": 2.2559,
+      "step": 4020
+    },
+    {
+      "epoch": 0.41333607315319015,
+      "grad_norm": 0.11185753345489502,
+      "learning_rate": 0.01,
+      "loss": 2.2599,
+      "step": 4023
+    },
+    {
+      "epoch": 0.41364430288708515,
+      "grad_norm": 0.11211541295051575,
+      "learning_rate": 0.01,
+      "loss": 2.2456,
+      "step": 4026
+    },
+    {
+      "epoch": 0.4139525326209802,
+      "grad_norm": 0.08211257308721542,
+      "learning_rate": 0.01,
+      "loss": 2.2636,
+      "step": 4029
+    },
+    {
+      "epoch": 0.4142607623548752,
+      "grad_norm": 0.07233046740293503,
+      "learning_rate": 0.01,
+      "loss": 2.2148,
+      "step": 4032
+    },
+    {
+      "epoch": 0.4145689920887702,
+      "grad_norm": 0.1062379851937294,
+      "learning_rate": 0.01,
+      "loss": 2.2382,
+      "step": 4035
+    },
+    {
+      "epoch": 0.4148772218226652,
+      "grad_norm": 0.07079877704381943,
+      "learning_rate": 0.01,
+      "loss": 2.2462,
+      "step": 4038
+    },
+    {
+      "epoch": 0.41518545155656017,
+      "grad_norm": 0.04237307608127594,
+      "learning_rate": 0.01,
+      "loss": 2.2523,
+      "step": 4041
+    },
+    {
+      "epoch": 0.41549368129045516,
+      "grad_norm": 0.12513239681720734,
+      "learning_rate": 0.01,
+      "loss": 2.2614,
+      "step": 4044
+    },
+    {
+      "epoch": 0.41580191102435016,
+      "grad_norm": 0.07134360820055008,
+      "learning_rate": 0.01,
+      "loss": 2.2533,
+      "step": 4047
+    },
+    {
+      "epoch": 0.41611014075824515,
+      "grad_norm": 0.07371515780687332,
+      "learning_rate": 0.01,
+      "loss": 2.2333,
+      "step": 4050
+    },
+    {
+      "epoch": 0.41641837049214014,
+      "grad_norm": 0.05744464695453644,
+      "learning_rate": 0.01,
+      "loss": 2.2292,
+      "step": 4053
+    },
+    {
+      "epoch": 0.41672660022603514,
+      "grad_norm": 0.0790088102221489,
+      "learning_rate": 0.01,
+      "loss": 2.2217,
+      "step": 4056
+    },
+    {
+      "epoch": 0.41703482995993013,
+      "grad_norm": 0.12540112435817719,
+      "learning_rate": 0.01,
+      "loss": 2.2367,
+      "step": 4059
+    },
+    {
+      "epoch": 0.4173430596938251,
+      "grad_norm": 0.06895852833986282,
+      "learning_rate": 0.01,
+      "loss": 2.2354,
+      "step": 4062
+    },
+    {
+      "epoch": 0.4176512894277201,
+      "grad_norm": 0.09068478643894196,
+      "learning_rate": 0.01,
+      "loss": 2.2605,
+      "step": 4065
+    },
+    {
+      "epoch": 0.4179595191616151,
+      "grad_norm": 0.051881443709135056,
+      "learning_rate": 0.01,
+      "loss": 2.2501,
+      "step": 4068
+    },
+    {
+      "epoch": 0.4182677488955101,
+      "grad_norm": 0.20433951914310455,
+      "learning_rate": 0.01,
+      "loss": 2.2582,
+      "step": 4071
+    },
+    {
+      "epoch": 0.4185759786294051,
+      "grad_norm": 0.08301309496164322,
+      "learning_rate": 0.01,
+      "loss": 2.2424,
+      "step": 4074
+    },
+    {
+      "epoch": 0.4188842083633001,
+      "grad_norm": 0.07062964886426926,
+      "learning_rate": 0.01,
+      "loss": 2.2345,
+      "step": 4077
+    },
+    {
+      "epoch": 0.4191924380971951,
+      "grad_norm": 0.09770773351192474,
+      "learning_rate": 0.01,
+      "loss": 2.264,
+      "step": 4080
+    },
+    {
+      "epoch": 0.4195006678310901,
+      "grad_norm": 0.0847458690404892,
+      "learning_rate": 0.01,
+      "loss": 2.2329,
+      "step": 4083
+    },
+    {
+      "epoch": 0.41980889756498513,
+      "grad_norm": 0.06491915881633759,
+      "learning_rate": 0.01,
+      "loss": 2.2174,
+      "step": 4086
+    },
+    {
+      "epoch": 0.4201171272988801,
+      "grad_norm": 0.11355047672986984,
+      "learning_rate": 0.01,
+      "loss": 2.2653,
+      "step": 4089
+    },
+    {
+      "epoch": 0.4204253570327751,
+      "grad_norm": 0.10509520024061203,
+      "learning_rate": 0.01,
+      "loss": 2.2435,
+      "step": 4092
+    },
+    {
+      "epoch": 0.4207335867666701,
+      "grad_norm": 0.07456620037555695,
+      "learning_rate": 0.01,
+      "loss": 2.2348,
+      "step": 4095
+    },
+    {
+      "epoch": 0.4210418165005651,
+      "grad_norm": 0.07531027495861053,
+      "learning_rate": 0.01,
+      "loss": 2.2524,
+      "step": 4098
+    },
+    {
+      "epoch": 0.4213500462344601,
+      "grad_norm": 0.06129564717411995,
+      "learning_rate": 0.01,
+      "loss": 2.2577,
+      "step": 4101
+    },
+    {
+      "epoch": 0.4216582759683551,
+      "grad_norm": 0.03984616696834564,
+      "learning_rate": 0.01,
+      "loss": 2.2354,
+      "step": 4104
+    },
+    {
+      "epoch": 0.4219665057022501,
+      "grad_norm": 0.1273418813943863,
+      "learning_rate": 0.01,
+      "loss": 2.2478,
+      "step": 4107
+    },
+    {
+      "epoch": 0.4222747354361451,
+      "grad_norm": 0.08859774470329285,
+      "learning_rate": 0.01,
+      "loss": 2.2504,
+      "step": 4110
+    },
+    {
+      "epoch": 0.42258296517004007,
+      "grad_norm": 0.10512147098779678,
+      "learning_rate": 0.01,
+      "loss": 2.2435,
+      "step": 4113
+    },
+    {
+      "epoch": 0.42289119490393506,
+      "grad_norm": 0.11181578040122986,
+      "learning_rate": 0.01,
+      "loss": 2.2396,
+      "step": 4116
+    },
+    {
+      "epoch": 0.42319942463783006,
+      "grad_norm": 0.07474307715892792,
+      "learning_rate": 0.01,
+      "loss": 2.2518,
+      "step": 4119
+    },
+    {
+      "epoch": 0.42350765437172505,
+      "grad_norm": 0.07233690470457077,
+      "learning_rate": 0.01,
+      "loss": 2.2283,
+      "step": 4122
+    },
+    {
+      "epoch": 0.42381588410562004,
+      "grad_norm": 0.06051602587103844,
+      "learning_rate": 0.01,
+      "loss": 2.2429,
+      "step": 4125
+    },
+    {
+      "epoch": 0.42412411383951504,
+      "grad_norm": 0.0492120198905468,
+      "learning_rate": 0.01,
+      "loss": 2.2312,
+      "step": 4128
+    },
+    {
+      "epoch": 0.42443234357341003,
+      "grad_norm": 0.07249493151903152,
+      "learning_rate": 0.01,
+      "loss": 2.244,
+      "step": 4131
+    },
+    {
+      "epoch": 0.424740573307305,
+      "grad_norm": 0.0993468165397644,
+      "learning_rate": 0.01,
+      "loss": 2.2441,
+      "step": 4134
+    },
+    {
+      "epoch": 0.4250488030412,
+      "grad_norm": 0.07051920145750046,
+      "learning_rate": 0.01,
+      "loss": 2.2188,
+      "step": 4137
+    },
+    {
+      "epoch": 0.425357032775095,
+      "grad_norm": 0.08267249912023544,
+      "learning_rate": 0.01,
+      "loss": 2.2472,
+      "step": 4140
+    },
+    {
+      "epoch": 0.42566526250899006,
+      "grad_norm": 0.1307336390018463,
+      "learning_rate": 0.01,
+      "loss": 2.2359,
+      "step": 4143
+    },
+    {
+      "epoch": 0.42597349224288505,
+      "grad_norm": 0.09383214265108109,
+      "learning_rate": 0.01,
+      "loss": 2.2519,
+      "step": 4146
+    },
+    {
+      "epoch": 0.42628172197678005,
+      "grad_norm": 0.08928582817316055,
+      "learning_rate": 0.01,
+      "loss": 2.2322,
+      "step": 4149
+    },
+    {
+      "epoch": 0.42658995171067504,
+      "grad_norm": 0.10554556548595428,
+      "learning_rate": 0.01,
+      "loss": 2.2219,
+      "step": 4152
+    },
+    {
+      "epoch": 0.42689818144457004,
+      "grad_norm": 0.06501816213130951,
+      "learning_rate": 0.01,
+      "loss": 2.2351,
+      "step": 4155
+    },
+    {
+      "epoch": 0.42720641117846503,
+      "grad_norm": 0.10736589878797531,
+      "learning_rate": 0.01,
+      "loss": 2.2327,
+      "step": 4158
+    },
+    {
+      "epoch": 0.42751464091236,
+      "grad_norm": 0.11834681034088135,
+      "learning_rate": 0.01,
+      "loss": 2.2617,
+      "step": 4161
+    },
+    {
+      "epoch": 0.427822870646255,
+      "grad_norm": 0.07011161744594574,
+      "learning_rate": 0.01,
+      "loss": 2.2218,
+      "step": 4164
+    },
+    {
+      "epoch": 0.42813110038015,
+      "grad_norm": 0.0653071179986,
+      "learning_rate": 0.01,
+      "loss": 2.2115,
+      "step": 4167
+    },
+    {
+      "epoch": 0.428439330114045,
+      "grad_norm": 0.057517893612384796,
+      "learning_rate": 0.01,
+      "loss": 2.2533,
+      "step": 4170
+    },
+    {
+      "epoch": 0.42874755984794,
+      "grad_norm": 0.060261376202106476,
+      "learning_rate": 0.01,
+      "loss": 2.2199,
+      "step": 4173
+    },
+    {
+      "epoch": 0.429055789581835,
+      "grad_norm": 0.12384762614965439,
+      "learning_rate": 0.01,
+      "loss": 2.2124,
+      "step": 4176
+    },
+    {
+      "epoch": 0.42936401931573,
+      "grad_norm": 0.06436473876237869,
+      "learning_rate": 0.01,
+      "loss": 2.2558,
+      "step": 4179
+    },
+    {
+      "epoch": 0.429672249049625,
+      "grad_norm": 0.049704987555742264,
+      "learning_rate": 0.01,
+      "loss": 2.2434,
+      "step": 4182
+    },
+    {
+      "epoch": 0.42998047878351997,
+      "grad_norm": 0.0809103325009346,
+      "learning_rate": 0.01,
+      "loss": 2.2461,
+      "step": 4185
+    },
+    {
+      "epoch": 0.43028870851741496,
+      "grad_norm": 0.04888701066374779,
+      "learning_rate": 0.01,
+      "loss": 2.2342,
+      "step": 4188
+    },
+    {
+      "epoch": 0.43059693825130996,
+      "grad_norm": 0.04951067641377449,
+      "learning_rate": 0.01,
+      "loss": 2.2292,
+      "step": 4191
+    },
+    {
+      "epoch": 0.43090516798520495,
+      "grad_norm": 0.13740333914756775,
+      "learning_rate": 0.01,
+      "loss": 2.2243,
+      "step": 4194
+    },
+    {
+      "epoch": 0.43121339771909994,
+      "grad_norm": 0.09912848472595215,
+      "learning_rate": 0.01,
+      "loss": 2.2065,
+      "step": 4197
+    },
+    {
+      "epoch": 0.43152162745299494,
+      "grad_norm": 0.1031954362988472,
+      "learning_rate": 0.01,
+      "loss": 2.2247,
+      "step": 4200
+    },
+    {
+      "epoch": 0.43182985718689,
+      "grad_norm": 0.04378229379653931,
+      "learning_rate": 0.01,
+      "loss": 2.2485,
+      "step": 4203
+    },
+    {
+      "epoch": 0.432138086920785,
+      "grad_norm": 0.05430865287780762,
+      "learning_rate": 0.01,
+      "loss": 2.2178,
+      "step": 4206
+    },
+    {
+      "epoch": 0.43244631665468,
+      "grad_norm": 0.05675321817398071,
+      "learning_rate": 0.01,
+      "loss": 2.2568,
+      "step": 4209
+    },
+    {
+      "epoch": 0.43275454638857497,
+      "grad_norm": 0.07637004554271698,
+      "learning_rate": 0.01,
+      "loss": 2.2567,
+      "step": 4212
+    },
+    {
+      "epoch": 0.43306277612246996,
+      "grad_norm": 0.06263475120067596,
+      "learning_rate": 0.01,
+      "loss": 2.2597,
+      "step": 4215
+    },
+    {
+      "epoch": 0.43337100585636495,
+      "grad_norm": 0.09689760208129883,
+      "learning_rate": 0.01,
+      "loss": 2.2376,
+      "step": 4218
+    },
+    {
+      "epoch": 0.43367923559025995,
+      "grad_norm": 0.13923399150371552,
+      "learning_rate": 0.01,
+      "loss": 2.2394,
+      "step": 4221
+    },
+    {
+      "epoch": 0.43398746532415494,
+      "grad_norm": 0.0607299767434597,
+      "learning_rate": 0.01,
+      "loss": 2.2366,
+      "step": 4224
+    },
+    {
+      "epoch": 0.43429569505804994,
+      "grad_norm": 0.05221550166606903,
+      "learning_rate": 0.01,
+      "loss": 2.2587,
+      "step": 4227
+    },
+    {
+      "epoch": 0.43460392479194493,
+      "grad_norm": 0.05556831881403923,
+      "learning_rate": 0.01,
+      "loss": 2.2422,
+      "step": 4230
+    },
+    {
+      "epoch": 0.4349121545258399,
+      "grad_norm": 0.0843261182308197,
+      "learning_rate": 0.01,
+      "loss": 2.2399,
+      "step": 4233
+    },
+    {
+      "epoch": 0.4352203842597349,
+      "grad_norm": 0.08864692598581314,
+      "learning_rate": 0.01,
+      "loss": 2.2155,
+      "step": 4236
+    },
+    {
+      "epoch": 0.4355286139936299,
+      "grad_norm": 0.11530198156833649,
+      "learning_rate": 0.01,
+      "loss": 2.2612,
+      "step": 4239
+    },
+    {
+      "epoch": 0.4358368437275249,
+      "grad_norm": 0.11549337208271027,
+      "learning_rate": 0.01,
+      "loss": 2.2233,
+      "step": 4242
+    },
+    {
+      "epoch": 0.4361450734614199,
+      "grad_norm": 0.11105350404977798,
+      "learning_rate": 0.01,
+      "loss": 2.2426,
+      "step": 4245
+    },
+    {
+      "epoch": 0.4364533031953149,
+      "grad_norm": 0.1190980076789856,
+      "learning_rate": 0.01,
+      "loss": 2.2353,
+      "step": 4248
+    },
+    {
+      "epoch": 0.4367615329292099,
+      "grad_norm": 0.08560021221637726,
+      "learning_rate": 0.01,
+      "loss": 2.2542,
+      "step": 4251
+    },
+    {
+      "epoch": 0.4370697626631049,
+      "grad_norm": 0.05514337494969368,
+      "learning_rate": 0.01,
+      "loss": 2.2171,
+      "step": 4254
+    },
+    {
+      "epoch": 0.43737799239699987,
+      "grad_norm": 0.06764981150627136,
+      "learning_rate": 0.01,
+      "loss": 2.2363,
+      "step": 4257
+    },
+    {
+      "epoch": 0.4376862221308949,
+      "grad_norm": 0.04801105335354805,
+      "learning_rate": 0.01,
+      "loss": 2.2352,
+      "step": 4260
+    },
+    {
+      "epoch": 0.4379944518647899,
+      "grad_norm": 0.04782482981681824,
+      "learning_rate": 0.01,
+      "loss": 2.2458,
+      "step": 4263
+    },
+    {
+      "epoch": 0.4383026815986849,
+      "grad_norm": 0.12880820035934448,
+      "learning_rate": 0.01,
+      "loss": 2.2384,
+      "step": 4266
+    },
+    {
+      "epoch": 0.4386109113325799,
+      "grad_norm": 0.06714754551649094,
+      "learning_rate": 0.01,
+      "loss": 2.2214,
+      "step": 4269
+    },
+    {
+      "epoch": 0.4389191410664749,
+      "grad_norm": 0.08878037333488464,
+      "learning_rate": 0.01,
+      "loss": 2.2597,
+      "step": 4272
+    },
+    {
+      "epoch": 0.4392273708003699,
+      "grad_norm": 0.051335882395505905,
+      "learning_rate": 0.01,
+      "loss": 2.2065,
+      "step": 4275
+    },
+    {
+      "epoch": 0.4395356005342649,
+      "grad_norm": 0.058174654841423035,
+      "learning_rate": 0.01,
+      "loss": 2.246,
+      "step": 4278
+    },
+    {
+      "epoch": 0.4398438302681599,
+      "grad_norm": 0.053695593029260635,
+      "learning_rate": 0.01,
+      "loss": 2.2406,
+      "step": 4281
+    },
+    {
+      "epoch": 0.44015206000205487,
+      "grad_norm": 0.07685926556587219,
+      "learning_rate": 0.01,
+      "loss": 2.2212,
+      "step": 4284
+    },
+    {
+      "epoch": 0.44046028973594986,
+      "grad_norm": 0.13495223224163055,
+      "learning_rate": 0.01,
+      "loss": 2.2486,
+      "step": 4287
+    },
+    {
+      "epoch": 0.44076851946984485,
+      "grad_norm": 0.0707453116774559,
+      "learning_rate": 0.01,
+      "loss": 2.247,
+      "step": 4290
+    },
+    {
+      "epoch": 0.44107674920373985,
+      "grad_norm": 0.04909240081906319,
+      "learning_rate": 0.01,
+      "loss": 2.2528,
+      "step": 4293
+    },
+    {
+      "epoch": 0.44138497893763484,
+      "grad_norm": 0.06148238107562065,
+      "learning_rate": 0.01,
+      "loss": 2.2462,
+      "step": 4296
+    },
+    {
+      "epoch": 0.44169320867152984,
+      "grad_norm": 0.07306285202503204,
+      "learning_rate": 0.01,
+      "loss": 2.199,
+      "step": 4299
+    },
+    {
+      "epoch": 0.44200143840542483,
+      "grad_norm": 0.12965865433216095,
+      "learning_rate": 0.01,
+      "loss": 2.2156,
+      "step": 4302
+    },
+    {
+      "epoch": 0.4423096681393198,
+      "grad_norm": 0.059606775641441345,
+      "learning_rate": 0.01,
+      "loss": 2.2209,
+      "step": 4305
+    },
+    {
+      "epoch": 0.4426178978732148,
+      "grad_norm": 0.06866457313299179,
+      "learning_rate": 0.01,
+      "loss": 2.2508,
+      "step": 4308
+    },
+    {
+      "epoch": 0.4429261276071098,
+      "grad_norm": 0.08940677344799042,
+      "learning_rate": 0.01,
+      "loss": 2.2244,
+      "step": 4311
+    },
+    {
+      "epoch": 0.4432343573410048,
+      "grad_norm": 0.10428988933563232,
+      "learning_rate": 0.01,
+      "loss": 2.2106,
+      "step": 4314
+    },
+    {
+      "epoch": 0.44354258707489985,
+      "grad_norm": 0.1565064787864685,
+      "learning_rate": 0.01,
+      "loss": 2.2745,
+      "step": 4317
+    },
+    {
+      "epoch": 0.44385081680879485,
+      "grad_norm": 0.11433500796556473,
+      "learning_rate": 0.01,
+      "loss": 2.2655,
+      "step": 4320
+    },
+    {
+      "epoch": 0.44415904654268984,
+      "grad_norm": 0.07315809279680252,
+      "learning_rate": 0.01,
+      "loss": 2.2523,
+      "step": 4323
+    },
+    {
+      "epoch": 0.44446727627658483,
+      "grad_norm": 0.048583708703517914,
+      "learning_rate": 0.01,
+      "loss": 2.2345,
+      "step": 4326
+    },
+    {
+      "epoch": 0.4447755060104798,
+      "grad_norm": 0.03422848507761955,
+      "learning_rate": 0.01,
+      "loss": 2.1883,
+      "step": 4329
+    },
+    {
+      "epoch": 0.4450837357443748,
+      "grad_norm": 0.05057518929243088,
+      "learning_rate": 0.01,
+      "loss": 2.2288,
+      "step": 4332
+    },
+    {
+      "epoch": 0.4453919654782698,
+      "grad_norm": 0.10407044738531113,
+      "learning_rate": 0.01,
+      "loss": 2.1974,
+      "step": 4335
+    },
+    {
+      "epoch": 0.4457001952121648,
+      "grad_norm": 0.06545260548591614,
+      "learning_rate": 0.01,
+      "loss": 2.2121,
+      "step": 4338
+    },
+    {
+      "epoch": 0.4460084249460598,
+      "grad_norm": 0.09442485123872757,
+      "learning_rate": 0.01,
+      "loss": 2.2145,
+      "step": 4341
+    },
+    {
+      "epoch": 0.4463166546799548,
+      "grad_norm": 0.11353209614753723,
+      "learning_rate": 0.01,
+      "loss": 2.227,
+      "step": 4344
+    },
+    {
+      "epoch": 0.4466248844138498,
+      "grad_norm": 0.11243279278278351,
+      "learning_rate": 0.01,
+      "loss": 2.242,
+      "step": 4347
+    },
+    {
+      "epoch": 0.4469331141477448,
+      "grad_norm": 0.14264856278896332,
+      "learning_rate": 0.01,
+      "loss": 2.2405,
+      "step": 4350
+    },
+    {
+      "epoch": 0.4472413438816398,
+      "grad_norm": 0.048186566680669785,
+      "learning_rate": 0.01,
+      "loss": 2.1921,
+      "step": 4353
+    },
+    {
+      "epoch": 0.44754957361553477,
+      "grad_norm": 0.0693448930978775,
+      "learning_rate": 0.01,
+      "loss": 2.2404,
+      "step": 4356
+    },
+    {
+      "epoch": 0.44785780334942976,
+      "grad_norm": 0.04426461458206177,
+      "learning_rate": 0.01,
+      "loss": 2.2114,
+      "step": 4359
+    },
+    {
+      "epoch": 0.44816603308332476,
+      "grad_norm": 0.06392990797758102,
+      "learning_rate": 0.01,
+      "loss": 2.224,
+      "step": 4362
+    },
+    {
+      "epoch": 0.44847426281721975,
+      "grad_norm": 0.16224262118339539,
+      "learning_rate": 0.01,
+      "loss": 2.261,
+      "step": 4365
+    },
+    {
+      "epoch": 0.44878249255111474,
+      "grad_norm": 0.06382444500923157,
+      "learning_rate": 0.01,
+      "loss": 2.2067,
+      "step": 4368
+    },
+    {
+      "epoch": 0.44909072228500974,
+      "grad_norm": 0.09267281740903854,
+      "learning_rate": 0.01,
+      "loss": 2.2403,
+      "step": 4371
+    },
+    {
+      "epoch": 0.44939895201890473,
+      "grad_norm": 0.09785914421081543,
+      "learning_rate": 0.01,
+      "loss": 2.2276,
+      "step": 4374
+    },
+    {
+      "epoch": 0.4497071817527998,
+      "grad_norm": 0.06673259288072586,
+      "learning_rate": 0.01,
+      "loss": 2.214,
+      "step": 4377
+    },
+    {
+      "epoch": 0.45001541148669477,
+      "grad_norm": 0.05463524907827377,
+      "learning_rate": 0.01,
+      "loss": 2.2048,
+      "step": 4380
+    },
+    {
+      "epoch": 0.45032364122058977,
+      "grad_norm": 0.05466567724943161,
+      "learning_rate": 0.01,
+      "loss": 2.2062,
+      "step": 4383
+    },
+    {
+      "epoch": 0.45063187095448476,
+      "grad_norm": 0.07413290441036224,
+      "learning_rate": 0.01,
+      "loss": 2.2178,
+      "step": 4386
+    },
+    {
+      "epoch": 0.45094010068837975,
+      "grad_norm": 0.06564678996801376,
+      "learning_rate": 0.01,
+      "loss": 2.2304,
+      "step": 4389
+    },
+    {
+      "epoch": 0.45124833042227475,
+      "grad_norm": 0.12468644231557846,
+      "learning_rate": 0.01,
+      "loss": 2.2301,
+      "step": 4392
+    },
+    {
+      "epoch": 0.45155656015616974,
+      "grad_norm": 0.06898069381713867,
+      "learning_rate": 0.01,
+      "loss": 2.2255,
+      "step": 4395
+    },
+    {
+      "epoch": 0.45186478989006473,
+      "grad_norm": 0.13579058647155762,
+      "learning_rate": 0.01,
+      "loss": 2.2021,
+      "step": 4398
+    },
+    {
+      "epoch": 0.4521730196239597,
+      "grad_norm": 0.07980421930551529,
+      "learning_rate": 0.01,
+      "loss": 2.2598,
+      "step": 4401
+    },
+    {
+      "epoch": 0.4524812493578547,
+      "grad_norm": 0.07771994173526764,
+      "learning_rate": 0.01,
+      "loss": 2.2166,
+      "step": 4404
+    },
+    {
+      "epoch": 0.4527894790917497,
+      "grad_norm": 0.08967602998018265,
+      "learning_rate": 0.01,
+      "loss": 2.2095,
+      "step": 4407
+    },
+    {
+      "epoch": 0.4530977088256447,
+      "grad_norm": 0.10909977555274963,
+      "learning_rate": 0.01,
+      "loss": 2.2064,
+      "step": 4410
+    },
+    {
+      "epoch": 0.4534059385595397,
+      "grad_norm": 0.11167363077402115,
+      "learning_rate": 0.01,
+      "loss": 2.2021,
+      "step": 4413
+    },
+    {
+      "epoch": 0.4537141682934347,
+      "grad_norm": 0.10310694575309753,
+      "learning_rate": 0.01,
+      "loss": 2.2582,
+      "step": 4416
+    },
+    {
+      "epoch": 0.4540223980273297,
+      "grad_norm": 0.06411474943161011,
+      "learning_rate": 0.01,
+      "loss": 2.2203,
+      "step": 4419
+    },
+    {
+      "epoch": 0.4543306277612247,
+      "grad_norm": 0.11141805350780487,
+      "learning_rate": 0.01,
+      "loss": 2.2163,
+      "step": 4422
+    },
+    {
+      "epoch": 0.4546388574951197,
+      "grad_norm": 0.09054200351238251,
+      "learning_rate": 0.01,
+      "loss": 2.2054,
+      "step": 4425
+    },
+    {
+      "epoch": 0.45494708722901467,
+      "grad_norm": 0.06952405720949173,
+      "learning_rate": 0.01,
+      "loss": 2.2488,
+      "step": 4428
+    },
+    {
+      "epoch": 0.45525531696290966,
+      "grad_norm": 0.08597440272569656,
+      "learning_rate": 0.01,
+      "loss": 2.2044,
+      "step": 4431
+    },
+    {
+      "epoch": 0.4555635466968047,
+      "grad_norm": 0.06718187034130096,
+      "learning_rate": 0.01,
+      "loss": 2.2419,
+      "step": 4434
+    },
+    {
+      "epoch": 0.4558717764306997,
+      "grad_norm": 0.0558515265583992,
+      "learning_rate": 0.01,
+      "loss": 2.2102,
+      "step": 4437
+    },
+    {
+      "epoch": 0.4561800061645947,
+      "grad_norm": 0.0560682937502861,
+      "learning_rate": 0.01,
+      "loss": 2.2324,
+      "step": 4440
+    },
+    {
+      "epoch": 0.4564882358984897,
+      "grad_norm": 0.058881547302007675,
+      "learning_rate": 0.01,
+      "loss": 2.1966,
+      "step": 4443
+    },
+    {
+      "epoch": 0.4567964656323847,
+      "grad_norm": 0.07034582644701004,
+      "learning_rate": 0.01,
+      "loss": 2.2021,
+      "step": 4446
+    },
+    {
+      "epoch": 0.4571046953662797,
+      "grad_norm": 0.09703799337148666,
+      "learning_rate": 0.01,
+      "loss": 2.21,
+      "step": 4449
+    },
+    {
+      "epoch": 0.45741292510017467,
+      "grad_norm": 0.06268820911645889,
+      "learning_rate": 0.01,
+      "loss": 2.2237,
+      "step": 4452
+    },
+    {
+      "epoch": 0.45772115483406967,
+      "grad_norm": 0.123359814286232,
+      "learning_rate": 0.01,
+      "loss": 2.2063,
+      "step": 4455
+    },
+    {
+      "epoch": 0.45802938456796466,
+      "grad_norm": 0.0536644384264946,
+      "learning_rate": 0.01,
+      "loss": 2.2002,
+      "step": 4458
+    },
+    {
+      "epoch": 0.45833761430185965,
+      "grad_norm": 0.0957527682185173,
+      "learning_rate": 0.01,
+      "loss": 2.2484,
+      "step": 4461
+    },
+    {
+      "epoch": 0.45864584403575465,
+      "grad_norm": 0.12607458233833313,
+      "learning_rate": 0.01,
+      "loss": 2.2241,
+      "step": 4464
+    },
+    {
+      "epoch": 0.45895407376964964,
+      "grad_norm": 0.07415255159139633,
+      "learning_rate": 0.01,
+      "loss": 2.2083,
+      "step": 4467
+    },
+    {
+      "epoch": 0.45926230350354463,
+      "grad_norm": 0.10248073190450668,
+      "learning_rate": 0.01,
+      "loss": 2.2253,
+      "step": 4470
+    },
+    {
+      "epoch": 0.4595705332374396,
+      "grad_norm": 0.05264243111014366,
+      "learning_rate": 0.01,
+      "loss": 2.2166,
+      "step": 4473
+    },
+    {
+      "epoch": 0.4598787629713346,
+      "grad_norm": 0.0557783767580986,
+      "learning_rate": 0.01,
+      "loss": 2.2213,
+      "step": 4476
+    },
+    {
+      "epoch": 0.4601869927052296,
+      "grad_norm": 0.06835830211639404,
+      "learning_rate": 0.01,
+      "loss": 2.2255,
+      "step": 4479
+    },
+    {
+      "epoch": 0.4604952224391246,
+      "grad_norm": 0.12045460939407349,
+      "learning_rate": 0.01,
+      "loss": 2.2331,
+      "step": 4482
+    },
+    {
+      "epoch": 0.4608034521730196,
+      "grad_norm": 0.11495090276002884,
+      "learning_rate": 0.01,
+      "loss": 2.2191,
+      "step": 4485
+    },
+    {
+      "epoch": 0.4611116819069146,
+      "grad_norm": 0.07859046757221222,
+      "learning_rate": 0.01,
+      "loss": 2.2282,
+      "step": 4488
+    },
+    {
+      "epoch": 0.46141991164080964,
+      "grad_norm": 0.03789819777011871,
+      "learning_rate": 0.01,
+      "loss": 2.2188,
+      "step": 4491
+    },
+    {
+      "epoch": 0.46172814137470464,
+      "grad_norm": 0.03617655113339424,
+      "learning_rate": 0.01,
+      "loss": 2.2496,
+      "step": 4494
+    },
+    {
+      "epoch": 0.46203637110859963,
+      "grad_norm": 0.06894705444574356,
+      "learning_rate": 0.01,
+      "loss": 2.2007,
+      "step": 4497
+    },
+    {
+      "epoch": 0.4623446008424946,
+      "grad_norm": 0.1143706887960434,
+      "learning_rate": 0.01,
+      "loss": 2.2247,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4626528305763896,
+      "grad_norm": 0.10069230943918228,
+      "learning_rate": 0.01,
+      "loss": 2.2114,
+      "step": 4503
+    },
+    {
+      "epoch": 0.4629610603102846,
+      "grad_norm": 0.10068007558584213,
+      "learning_rate": 0.01,
+      "loss": 2.2438,
+      "step": 4506
+    },
+    {
+      "epoch": 0.4632692900441796,
+      "grad_norm": 0.05319290608167648,
+      "learning_rate": 0.01,
+      "loss": 2.2422,
+      "step": 4509
+    },
+    {
+      "epoch": 0.4635775197780746,
+      "grad_norm": 0.06933122128248215,
+      "learning_rate": 0.01,
+      "loss": 2.2059,
+      "step": 4512
+    },
+    {
+      "epoch": 0.4638857495119696,
+      "grad_norm": 0.11921056360006332,
+      "learning_rate": 0.01,
+      "loss": 2.2137,
+      "step": 4515
+    },
+    {
+      "epoch": 0.4641939792458646,
+      "grad_norm": 0.06092121824622154,
+      "learning_rate": 0.01,
+      "loss": 2.1941,
+      "step": 4518
+    },
+    {
+      "epoch": 0.4645022089797596,
+      "grad_norm": 0.06017937511205673,
+      "learning_rate": 0.01,
+      "loss": 2.2539,
+      "step": 4521
+    },
+    {
+      "epoch": 0.4648104387136546,
+      "grad_norm": 0.05721915140748024,
+      "learning_rate": 0.01,
+      "loss": 2.2348,
+      "step": 4524
+    },
+    {
+      "epoch": 0.46511866844754957,
+      "grad_norm": 0.07706714421510696,
+      "learning_rate": 0.01,
+      "loss": 2.2169,
+      "step": 4527
+    },
+    {
+      "epoch": 0.46542689818144456,
+      "grad_norm": 0.07279779762029648,
+      "learning_rate": 0.01,
+      "loss": 2.2163,
+      "step": 4530
+    },
+    {
+      "epoch": 0.46573512791533955,
+      "grad_norm": 0.06781268864870071,
+      "learning_rate": 0.01,
+      "loss": 2.1682,
+      "step": 4533
+    },
+    {
+      "epoch": 0.46604335764923455,
+      "grad_norm": 0.0807657316327095,
+      "learning_rate": 0.01,
+      "loss": 2.2123,
+      "step": 4536
+    },
+    {
+      "epoch": 0.46635158738312954,
+      "grad_norm": 0.06467099487781525,
+      "learning_rate": 0.01,
+      "loss": 2.2152,
+      "step": 4539
+    },
+    {
+      "epoch": 0.46665981711702453,
+      "grad_norm": 0.10680168867111206,
+      "learning_rate": 0.01,
+      "loss": 2.2062,
+      "step": 4542
+    },
+    {
+      "epoch": 0.4669680468509195,
+      "grad_norm": 0.11668167263269424,
+      "learning_rate": 0.01,
+      "loss": 2.206,
+      "step": 4545
+    },
+    {
+      "epoch": 0.4672762765848146,
+      "grad_norm": 0.06468226760625839,
+      "learning_rate": 0.01,
+      "loss": 2.2011,
+      "step": 4548
+    },
+    {
+      "epoch": 0.46758450631870957,
+      "grad_norm": 0.07668601721525192,
+      "learning_rate": 0.01,
+      "loss": 2.2128,
+      "step": 4551
+    },
+    {
+      "epoch": 0.46789273605260456,
+      "grad_norm": 0.05631673336029053,
+      "learning_rate": 0.01,
+      "loss": 2.1812,
+      "step": 4554
+    },
+    {
+      "epoch": 0.46820096578649956,
+      "grad_norm": 0.12898530066013336,
+      "learning_rate": 0.01,
+      "loss": 2.2312,
+      "step": 4557
+    },
+    {
+      "epoch": 0.46850919552039455,
+      "grad_norm": 0.07105603069067001,
+      "learning_rate": 0.01,
+      "loss": 2.1949,
+      "step": 4560
+    },
+    {
+      "epoch": 0.46881742525428954,
+      "grad_norm": 0.07172367721796036,
+      "learning_rate": 0.01,
+      "loss": 2.2509,
+      "step": 4563
+    },
+    {
+      "epoch": 0.46912565498818454,
+      "grad_norm": 0.1219574511051178,
+      "learning_rate": 0.01,
+      "loss": 2.2147,
+      "step": 4566
+    },
+    {
+      "epoch": 0.46943388472207953,
+      "grad_norm": 0.05777307227253914,
+      "learning_rate": 0.01,
+      "loss": 2.2071,
+      "step": 4569
+    },
+    {
+      "epoch": 0.4697421144559745,
+      "grad_norm": 0.12805253267288208,
+      "learning_rate": 0.01,
+      "loss": 2.2166,
+      "step": 4572
+    },
+    {
+      "epoch": 0.4700503441898695,
+      "grad_norm": 0.11360877752304077,
+      "learning_rate": 0.01,
+      "loss": 2.1827,
+      "step": 4575
+    },
+    {
+      "epoch": 0.4703585739237645,
+      "grad_norm": 0.07203348726034164,
+      "learning_rate": 0.01,
+      "loss": 2.2378,
+      "step": 4578
+    },
+    {
+      "epoch": 0.4706668036576595,
+      "grad_norm": 0.05645303055644035,
+      "learning_rate": 0.01,
+      "loss": 2.2044,
+      "step": 4581
+    },
+    {
+      "epoch": 0.4709750333915545,
+      "grad_norm": 0.06103040650486946,
+      "learning_rate": 0.01,
+      "loss": 2.2302,
+      "step": 4584
+    },
+    {
+      "epoch": 0.4712832631254495,
+      "grad_norm": 0.0621771402657032,
+      "learning_rate": 0.01,
+      "loss": 2.2147,
+      "step": 4587
+    },
+    {
+      "epoch": 0.4715914928593445,
+      "grad_norm": 0.08458666503429413,
+      "learning_rate": 0.01,
+      "loss": 2.1781,
+      "step": 4590
+    },
+    {
+      "epoch": 0.4718997225932395,
+      "grad_norm": 0.092729851603508,
+      "learning_rate": 0.01,
+      "loss": 2.2326,
+      "step": 4593
+    },
+    {
+      "epoch": 0.4722079523271345,
+      "grad_norm": 0.09255766123533249,
+      "learning_rate": 0.01,
+      "loss": 2.2082,
+      "step": 4596
+    },
+    {
+      "epoch": 0.47251618206102947,
+      "grad_norm": 0.11929985135793686,
+      "learning_rate": 0.01,
+      "loss": 2.2064,
+      "step": 4599
+    },
+    {
+      "epoch": 0.47282441179492446,
+      "grad_norm": 0.12234004586935043,
+      "learning_rate": 0.01,
+      "loss": 2.1513,
+      "step": 4602
+    },
+    {
+      "epoch": 0.47313264152881945,
+      "grad_norm": 0.07648742944002151,
+      "learning_rate": 0.01,
+      "loss": 2.2376,
+      "step": 4605
+    },
+    {
+      "epoch": 0.4734408712627145,
+      "grad_norm": 0.05717691034078598,
+      "learning_rate": 0.01,
+      "loss": 2.231,
+      "step": 4608
+    },
+    {
+      "epoch": 0.4737491009966095,
+      "grad_norm": 0.048224568367004395,
+      "learning_rate": 0.01,
+      "loss": 2.2126,
+      "step": 4611
+    },
+    {
+      "epoch": 0.4740573307305045,
+      "grad_norm": 0.07530826330184937,
+      "learning_rate": 0.01,
+      "loss": 2.2155,
+      "step": 4614
+    },
+    {
+      "epoch": 0.4743655604643995,
+      "grad_norm": 0.08617862313985825,
+      "learning_rate": 0.01,
+      "loss": 2.2286,
+      "step": 4617
+    },
+    {
+      "epoch": 0.4746737901982945,
+      "grad_norm": 0.10041820257902145,
+      "learning_rate": 0.01,
+      "loss": 2.1917,
+      "step": 4620
+    },
+    {
+      "epoch": 0.47498201993218947,
+      "grad_norm": 0.04470205307006836,
+      "learning_rate": 0.01,
+      "loss": 2.2188,
+      "step": 4623
+    },
+    {
+      "epoch": 0.47529024966608446,
+      "grad_norm": 0.060269374400377274,
+      "learning_rate": 0.01,
+      "loss": 2.2267,
+      "step": 4626
+    },
+    {
+      "epoch": 0.47559847939997946,
+      "grad_norm": 0.06320520490407944,
+      "learning_rate": 0.01,
+      "loss": 2.2054,
+      "step": 4629
+    },
+    {
+      "epoch": 0.47590670913387445,
+      "grad_norm": 0.05642838776111603,
+      "learning_rate": 0.01,
+      "loss": 2.2062,
+      "step": 4632
+    },
+    {
+      "epoch": 0.47621493886776944,
+      "grad_norm": 0.064301997423172,
+      "learning_rate": 0.01,
+      "loss": 2.2296,
+      "step": 4635
+    },
+    {
+      "epoch": 0.47652316860166444,
+      "grad_norm": 0.07448214292526245,
+      "learning_rate": 0.01,
+      "loss": 2.197,
+      "step": 4638
+    },
+    {
+      "epoch": 0.47683139833555943,
+      "grad_norm": 0.08586326986551285,
+      "learning_rate": 0.01,
+      "loss": 2.1743,
+      "step": 4641
+    },
+    {
+      "epoch": 0.4771396280694544,
+      "grad_norm": 0.13179326057434082,
+      "learning_rate": 0.01,
+      "loss": 2.2299,
+      "step": 4644
+    },
+    {
+      "epoch": 0.4774478578033494,
+      "grad_norm": 0.1163720041513443,
+      "learning_rate": 0.01,
+      "loss": 2.2089,
+      "step": 4647
+    },
+    {
+      "epoch": 0.4777560875372444,
+      "grad_norm": 0.04846031963825226,
+      "learning_rate": 0.01,
+      "loss": 2.1564,
+      "step": 4650
+    },
+    {
+      "epoch": 0.4780643172711394,
+      "grad_norm": 0.13724131882190704,
+      "learning_rate": 0.01,
+      "loss": 2.2078,
+      "step": 4653
+    },
+    {
+      "epoch": 0.4783725470050344,
+      "grad_norm": 0.062840536236763,
+      "learning_rate": 0.01,
+      "loss": 2.2252,
+      "step": 4656
+    },
+    {
+      "epoch": 0.4786807767389294,
+      "grad_norm": 0.06721820682287216,
+      "learning_rate": 0.01,
+      "loss": 2.1781,
+      "step": 4659
+    },
+    {
+      "epoch": 0.4789890064728244,
+      "grad_norm": 0.09086044877767563,
+      "learning_rate": 0.01,
+      "loss": 2.2179,
+      "step": 4662
+    },
+    {
+      "epoch": 0.47929723620671943,
+      "grad_norm": 0.07732655107975006,
+      "learning_rate": 0.01,
+      "loss": 2.2334,
+      "step": 4665
+    },
+    {
+      "epoch": 0.47960546594061443,
+      "grad_norm": 0.04763714596629143,
+      "learning_rate": 0.01,
+      "loss": 2.2262,
+      "step": 4668
+    },
+    {
+      "epoch": 0.4799136956745094,
+      "grad_norm": 0.09649144858121872,
+      "learning_rate": 0.01,
+      "loss": 2.2141,
+      "step": 4671
+    },
+    {
+      "epoch": 0.4802219254084044,
+      "grad_norm": 0.05458167567849159,
+      "learning_rate": 0.01,
+      "loss": 2.1967,
+      "step": 4674
+    },
+    {
+      "epoch": 0.4805301551422994,
+      "grad_norm": 0.08577650040388107,
+      "learning_rate": 0.01,
+      "loss": 2.2183,
+      "step": 4677
+    },
+    {
+      "epoch": 0.4808383848761944,
+      "grad_norm": 0.0733698159456253,
+      "learning_rate": 0.01,
+      "loss": 2.2185,
+      "step": 4680
+    },
+    {
+      "epoch": 0.4811466146100894,
+      "grad_norm": 0.06648692488670349,
+      "learning_rate": 0.01,
+      "loss": 2.1904,
+      "step": 4683
+    },
+    {
+      "epoch": 0.4814548443439844,
+      "grad_norm": 0.08376996219158173,
+      "learning_rate": 0.01,
+      "loss": 2.2097,
+      "step": 4686
+    },
+    {
+      "epoch": 0.4817630740778794,
+      "grad_norm": 0.05270134285092354,
+      "learning_rate": 0.01,
+      "loss": 2.2304,
+      "step": 4689
+    },
+    {
+      "epoch": 0.4820713038117744,
+      "grad_norm": 0.05531509965658188,
+      "learning_rate": 0.01,
+      "loss": 2.2039,
+      "step": 4692
+    },
+    {
+      "epoch": 0.48237953354566937,
+      "grad_norm": 0.05848492309451103,
+      "learning_rate": 0.01,
+      "loss": 2.2113,
+      "step": 4695
+    },
+    {
+      "epoch": 0.48268776327956436,
+      "grad_norm": 0.06692120432853699,
+      "learning_rate": 0.01,
+      "loss": 2.1972,
+      "step": 4698
+    },
+    {
+      "epoch": 0.48299599301345936,
+      "grad_norm": 0.07243851572275162,
+      "learning_rate": 0.01,
+      "loss": 2.223,
+      "step": 4701
+    },
+    {
+      "epoch": 0.48330422274735435,
+      "grad_norm": 0.06565523892641068,
+      "learning_rate": 0.01,
+      "loss": 2.1913,
+      "step": 4704
+    },
+    {
+      "epoch": 0.48361245248124934,
+      "grad_norm": 0.04595122113823891,
+      "learning_rate": 0.01,
+      "loss": 2.1782,
+      "step": 4707
+    },
+    {
+      "epoch": 0.48392068221514434,
+      "grad_norm": 0.06658844649791718,
+      "learning_rate": 0.01,
+      "loss": 2.224,
+      "step": 4710
+    },
+    {
+      "epoch": 0.48422891194903933,
+      "grad_norm": 0.0807071253657341,
+      "learning_rate": 0.01,
+      "loss": 2.217,
+      "step": 4713
+    },
+    {
+      "epoch": 0.4845371416829343,
+      "grad_norm": 0.0562782846391201,
+      "learning_rate": 0.01,
+      "loss": 2.2033,
+      "step": 4716
+    },
+    {
+      "epoch": 0.4848453714168293,
+      "grad_norm": 0.07851718366146088,
+      "learning_rate": 0.01,
+      "loss": 2.1847,
+      "step": 4719
+    },
+    {
+      "epoch": 0.48515360115072437,
+      "grad_norm": 0.07649900764226913,
+      "learning_rate": 0.01,
+      "loss": 2.2222,
+      "step": 4722
+    },
+    {
+      "epoch": 0.48546183088461936,
+      "grad_norm": 0.07279150187969208,
+      "learning_rate": 0.01,
+      "loss": 2.1951,
+      "step": 4725
+    },
+    {
+      "epoch": 0.48577006061851435,
+      "grad_norm": 0.053628645837306976,
+      "learning_rate": 0.01,
+      "loss": 2.1681,
+      "step": 4728
+    },
+    {
+      "epoch": 0.48607829035240935,
+      "grad_norm": 0.09401357173919678,
+      "learning_rate": 0.01,
+      "loss": 2.1943,
+      "step": 4731
+    },
+    {
+      "epoch": 0.48638652008630434,
+      "grad_norm": 0.1156088337302208,
+      "learning_rate": 0.01,
+      "loss": 2.2317,
+      "step": 4734
+    },
+    {
+      "epoch": 0.48669474982019933,
+      "grad_norm": 0.12672138214111328,
+      "learning_rate": 0.01,
+      "loss": 2.2085,
+      "step": 4737
+    },
+    {
+      "epoch": 0.48700297955409433,
+      "grad_norm": 0.06799574196338654,
+      "learning_rate": 0.01,
+      "loss": 2.2161,
+      "step": 4740
+    },
+    {
+      "epoch": 0.4873112092879893,
+      "grad_norm": 0.06479325145483017,
+      "learning_rate": 0.01,
+      "loss": 2.1663,
+      "step": 4743
+    },
+    {
+      "epoch": 0.4876194390218843,
+      "grad_norm": 0.09143824130296707,
+      "learning_rate": 0.01,
+      "loss": 2.2193,
+      "step": 4746
+    },
+    {
+      "epoch": 0.4879276687557793,
+      "grad_norm": 0.09262688457965851,
+      "learning_rate": 0.01,
+      "loss": 2.218,
+      "step": 4749
+    },
+    {
+      "epoch": 0.4882358984896743,
+      "grad_norm": 0.11519678682088852,
+      "learning_rate": 0.01,
+      "loss": 2.1937,
+      "step": 4752
+    },
+    {
+      "epoch": 0.4885441282235693,
+      "grad_norm": 0.07646415382623672,
+      "learning_rate": 0.01,
+      "loss": 2.2133,
+      "step": 4755
+    },
+    {
+      "epoch": 0.4888523579574643,
+      "grad_norm": 0.08090809732675552,
+      "learning_rate": 0.01,
+      "loss": 2.193,
+      "step": 4758
+    },
+    {
+      "epoch": 0.4891605876913593,
+      "grad_norm": 0.08812209218740463,
+      "learning_rate": 0.01,
+      "loss": 2.2215,
+      "step": 4761
+    },
+    {
+      "epoch": 0.4894688174252543,
+      "grad_norm": 0.14427846670150757,
+      "learning_rate": 0.01,
+      "loss": 2.2115,
+      "step": 4764
+    },
+    {
+      "epoch": 0.48977704715914927,
+      "grad_norm": 0.08065719902515411,
+      "learning_rate": 0.01,
+      "loss": 2.1861,
+      "step": 4767
+    },
+    {
+      "epoch": 0.49008527689304426,
+      "grad_norm": 0.04888691008090973,
+      "learning_rate": 0.01,
+      "loss": 2.1911,
+      "step": 4770
+    },
+    {
+      "epoch": 0.49039350662693926,
+      "grad_norm": 0.04742259159684181,
+      "learning_rate": 0.01,
+      "loss": 2.2152,
+      "step": 4773
+    },
+    {
+      "epoch": 0.49070173636083425,
+      "grad_norm": 0.061714138835668564,
+      "learning_rate": 0.01,
+      "loss": 2.2009,
+      "step": 4776
+    },
+    {
+      "epoch": 0.49100996609472924,
+      "grad_norm": 0.07582443952560425,
+      "learning_rate": 0.01,
+      "loss": 2.2189,
+      "step": 4779
+    },
+    {
+      "epoch": 0.4913181958286243,
+      "grad_norm": 0.1390780359506607,
+      "learning_rate": 0.01,
+      "loss": 2.211,
+      "step": 4782
+    },
+    {
+      "epoch": 0.4916264255625193,
+      "grad_norm": 0.03784565255045891,
+      "learning_rate": 0.01,
+      "loss": 2.2011,
+      "step": 4785
+    },
+    {
+      "epoch": 0.4919346552964143,
+      "grad_norm": 0.07413594424724579,
+      "learning_rate": 0.01,
+      "loss": 2.2103,
+      "step": 4788
+    },
+    {
+      "epoch": 0.4922428850303093,
+      "grad_norm": 0.09402404725551605,
+      "learning_rate": 0.01,
+      "loss": 2.1912,
+      "step": 4791
+    },
+    {
+      "epoch": 0.49255111476420427,
+      "grad_norm": 0.0717400312423706,
+      "learning_rate": 0.01,
+      "loss": 2.1868,
+      "step": 4794
+    },
+    {
+      "epoch": 0.49285934449809926,
+      "grad_norm": 0.05179424583911896,
+      "learning_rate": 0.01,
+      "loss": 2.2298,
+      "step": 4797
+    },
+    {
+      "epoch": 0.49316757423199425,
+      "grad_norm": 0.12123433500528336,
+      "learning_rate": 0.01,
+      "loss": 2.2005,
+      "step": 4800
+    },
+    {
+      "epoch": 0.49347580396588925,
+      "grad_norm": 0.04941033944487572,
+      "learning_rate": 0.01,
+      "loss": 2.2113,
+      "step": 4803
+    },
+    {
+      "epoch": 0.49378403369978424,
+      "grad_norm": 0.10987304151058197,
+      "learning_rate": 0.01,
+      "loss": 2.209,
+      "step": 4806
+    },
+    {
+      "epoch": 0.49409226343367924,
+      "grad_norm": 0.09235193580389023,
+      "learning_rate": 0.01,
+      "loss": 2.1967,
+      "step": 4809
+    },
+    {
+      "epoch": 0.49440049316757423,
+      "grad_norm": 0.057354703545570374,
+      "learning_rate": 0.01,
+      "loss": 2.219,
+      "step": 4812
+    },
+    {
+      "epoch": 0.4947087229014692,
+      "grad_norm": 0.04692654311656952,
+      "learning_rate": 0.01,
+      "loss": 2.173,
+      "step": 4815
+    },
+    {
+      "epoch": 0.4950169526353642,
+      "grad_norm": 0.09447453171014786,
+      "learning_rate": 0.01,
+      "loss": 2.1806,
+      "step": 4818
+    },
+    {
+      "epoch": 0.4953251823692592,
+      "grad_norm": 0.09967079013586044,
+      "learning_rate": 0.01,
+      "loss": 2.1809,
+      "step": 4821
+    },
+    {
+      "epoch": 0.4956334121031542,
+      "grad_norm": 0.06462189555168152,
+      "learning_rate": 0.01,
+      "loss": 2.1922,
+      "step": 4824
+    },
+    {
+      "epoch": 0.4959416418370492,
+      "grad_norm": 0.038030870258808136,
+      "learning_rate": 0.01,
+      "loss": 2.2239,
+      "step": 4827
+    },
+    {
+      "epoch": 0.4962498715709442,
+      "grad_norm": 0.06828872114419937,
+      "learning_rate": 0.01,
+      "loss": 2.1881,
+      "step": 4830
+    },
+    {
+      "epoch": 0.4965581013048392,
+      "grad_norm": 0.10087070614099503,
+      "learning_rate": 0.01,
+      "loss": 2.22,
+      "step": 4833
+    },
+    {
+      "epoch": 0.4968663310387342,
+      "grad_norm": 0.07630455493927002,
+      "learning_rate": 0.01,
+      "loss": 2.188,
+      "step": 4836
+    },
+    {
+      "epoch": 0.4971745607726292,
+      "grad_norm": 0.05040668696165085,
+      "learning_rate": 0.01,
+      "loss": 2.2012,
+      "step": 4839
+    },
+    {
+      "epoch": 0.4974827905065242,
+      "grad_norm": 0.05160282924771309,
+      "learning_rate": 0.01,
+      "loss": 2.2119,
+      "step": 4842
+    },
+    {
+      "epoch": 0.4977910202404192,
+      "grad_norm": 0.04949258640408516,
+      "learning_rate": 0.01,
+      "loss": 2.1959,
+      "step": 4845
+    },
+    {
+      "epoch": 0.4980992499743142,
+      "grad_norm": 0.07766029983758926,
+      "learning_rate": 0.01,
+      "loss": 2.1896,
+      "step": 4848
+    },
+    {
+      "epoch": 0.4984074797082092,
+      "grad_norm": 0.06274580955505371,
+      "learning_rate": 0.01,
+      "loss": 2.2014,
+      "step": 4851
+    },
+    {
+      "epoch": 0.4987157094421042,
+      "grad_norm": 0.1071280762553215,
+      "learning_rate": 0.01,
+      "loss": 2.2045,
+      "step": 4854
+    },
+    {
+      "epoch": 0.4990239391759992,
+      "grad_norm": 0.10645020008087158,
+      "learning_rate": 0.01,
+      "loss": 2.1895,
+      "step": 4857
+    },
+    {
+      "epoch": 0.4993321689098942,
+      "grad_norm": 0.1151091679930687,
+      "learning_rate": 0.01,
+      "loss": 2.1954,
+      "step": 4860
+    },
+    {
+      "epoch": 0.4996403986437892,
+      "grad_norm": 0.09699530899524689,
+      "learning_rate": 0.01,
+      "loss": 2.1833,
+      "step": 4863
+    },
+    {
+      "epoch": 0.49994862837768417,
+      "grad_norm": 0.06568959355354309,
+      "learning_rate": 0.01,
+      "loss": 2.1862,
+      "step": 4866
+    },
+    {
+      "epoch": 0.5002568581115792,
+      "grad_norm": 0.0421447716653347,
+      "learning_rate": 0.01,
+      "loss": 2.1819,
+      "step": 4869
+    },
+    {
+      "epoch": 0.5005650878454742,
+      "grad_norm": 0.04529868811368942,
+      "learning_rate": 0.01,
+      "loss": 2.1852,
+      "step": 4872
+    },
+    {
+      "epoch": 0.5008733175793691,
+      "grad_norm": 0.059541650116443634,
+      "learning_rate": 0.01,
+      "loss": 2.1955,
+      "step": 4875
+    },
+    {
+      "epoch": 0.5011815473132641,
+      "grad_norm": 0.061823770403862,
+      "learning_rate": 0.01,
+      "loss": 2.2039,
+      "step": 4878
+    },
+    {
+      "epoch": 0.5014897770471591,
+      "grad_norm": 0.05892050638794899,
+      "learning_rate": 0.01,
+      "loss": 2.178,
+      "step": 4881
+    },
+    {
+      "epoch": 0.5017980067810541,
+      "grad_norm": 0.04842402786016464,
+      "learning_rate": 0.01,
+      "loss": 2.1948,
+      "step": 4884
+    },
+    {
+      "epoch": 0.5021062365149491,
+      "grad_norm": 0.05962050333619118,
+      "learning_rate": 0.01,
+      "loss": 2.1932,
+      "step": 4887
+    },
+    {
+      "epoch": 0.5024144662488441,
+      "grad_norm": 0.056295089423656464,
+      "learning_rate": 0.01,
+      "loss": 2.1757,
+      "step": 4890
+    },
+    {
+      "epoch": 0.5027226959827391,
+      "grad_norm": 0.07448049634695053,
+      "learning_rate": 0.01,
+      "loss": 2.2181,
+      "step": 4893
+    },
+    {
+      "epoch": 0.5030309257166341,
+      "grad_norm": 0.07998815923929214,
+      "learning_rate": 0.01,
+      "loss": 2.1714,
+      "step": 4896
+    },
+    {
+      "epoch": 0.5033391554505291,
+      "grad_norm": 0.08058517426252365,
+      "learning_rate": 0.01,
+      "loss": 2.2131,
+      "step": 4899
+    },
+    {
+      "epoch": 0.5036473851844241,
+      "grad_norm": 0.07899410277605057,
+      "learning_rate": 0.01,
+      "loss": 2.1727,
+      "step": 4902
+    },
+    {
+      "epoch": 0.5039556149183191,
+      "grad_norm": 0.05830831080675125,
+      "learning_rate": 0.01,
+      "loss": 2.176,
+      "step": 4905
+    },
+    {
+      "epoch": 0.5042638446522141,
+      "grad_norm": 0.05831579118967056,
+      "learning_rate": 0.01,
+      "loss": 2.1754,
+      "step": 4908
+    },
+    {
+      "epoch": 0.5045720743861091,
+      "grad_norm": 0.052614904940128326,
+      "learning_rate": 0.01,
+      "loss": 2.1935,
+      "step": 4911
+    },
+    {
+      "epoch": 0.5048803041200041,
+      "grad_norm": 0.0830332413315773,
+      "learning_rate": 0.01,
+      "loss": 2.2274,
+      "step": 4914
+    },
+    {
+      "epoch": 0.5051885338538991,
+      "grad_norm": 0.1138230562210083,
+      "learning_rate": 0.01,
+      "loss": 2.1976,
+      "step": 4917
+    },
+    {
+      "epoch": 0.505496763587794,
+      "grad_norm": 0.07024016976356506,
+      "learning_rate": 0.01,
+      "loss": 2.1969,
+      "step": 4920
+    },
+    {
+      "epoch": 0.505804993321689,
+      "grad_norm": 0.07235170155763626,
+      "learning_rate": 0.01,
+      "loss": 2.2163,
+      "step": 4923
+    },
+    {
+      "epoch": 0.5061132230555841,
+      "grad_norm": 0.06894835084676743,
+      "learning_rate": 0.01,
+      "loss": 2.2232,
+      "step": 4926
+    },
+    {
+      "epoch": 0.5064214527894791,
+      "grad_norm": 0.0825890600681305,
+      "learning_rate": 0.01,
+      "loss": 2.1896,
+      "step": 4929
+    },
+    {
+      "epoch": 0.5067296825233741,
+      "grad_norm": 0.05901159718632698,
+      "learning_rate": 0.01,
+      "loss": 2.1988,
+      "step": 4932
+    },
+    {
+      "epoch": 0.5070379122572691,
+      "grad_norm": 0.048157334327697754,
+      "learning_rate": 0.01,
+      "loss": 2.1904,
+      "step": 4935
+    },
+    {
+      "epoch": 0.5073461419911641,
+      "grad_norm": 0.10036749392747879,
+      "learning_rate": 0.01,
+      "loss": 2.1977,
+      "step": 4938
+    },
+    {
+      "epoch": 0.5076543717250591,
+      "grad_norm": 0.10984963923692703,
+      "learning_rate": 0.01,
+      "loss": 2.1957,
+      "step": 4941
+    },
+    {
+      "epoch": 0.5079626014589541,
+      "grad_norm": 0.09587367624044418,
+      "learning_rate": 0.01,
+      "loss": 2.2,
+      "step": 4944
+    },
+    {
+      "epoch": 0.5082708311928491,
+      "grad_norm": 0.06347552686929703,
+      "learning_rate": 0.01,
+      "loss": 2.1918,
+      "step": 4947
+    },
+    {
+      "epoch": 0.5085790609267441,
+      "grad_norm": 0.0658629834651947,
+      "learning_rate": 0.01,
+      "loss": 2.1945,
+      "step": 4950
+    },
+    {
+      "epoch": 0.5088872906606391,
+      "grad_norm": 0.045971643179655075,
+      "learning_rate": 0.01,
+      "loss": 2.2114,
+      "step": 4953
+    },
+    {
+      "epoch": 0.5091955203945341,
+      "grad_norm": 0.04058291018009186,
+      "learning_rate": 0.01,
+      "loss": 2.2066,
+      "step": 4956
+    },
+    {
+      "epoch": 0.5095037501284291,
+      "grad_norm": 0.052851296961307526,
+      "learning_rate": 0.01,
+      "loss": 2.1884,
+      "step": 4959
+    },
+    {
+      "epoch": 0.5098119798623241,
+      "grad_norm": 0.033158350735902786,
+      "learning_rate": 0.01,
+      "loss": 2.2078,
+      "step": 4962
+    },
+    {
+      "epoch": 0.5101202095962191,
+      "grad_norm": 0.05409036949276924,
+      "learning_rate": 0.01,
+      "loss": 2.181,
+      "step": 4965
+    },
+    {
+      "epoch": 0.5104284393301141,
+      "grad_norm": 0.0731736570596695,
+      "learning_rate": 0.01,
+      "loss": 2.1825,
+      "step": 4968
+    },
+    {
+      "epoch": 0.510736669064009,
+      "grad_norm": 0.05854470282793045,
+      "learning_rate": 0.01,
+      "loss": 2.2119,
+      "step": 4971
+    },
+    {
+      "epoch": 0.511044898797904,
+      "grad_norm": 0.05071520060300827,
+      "learning_rate": 0.01,
+      "loss": 2.1886,
+      "step": 4974
+    },
+    {
+      "epoch": 0.511353128531799,
+      "grad_norm": 0.060792725533246994,
+      "learning_rate": 0.01,
+      "loss": 2.2066,
+      "step": 4977
+    },
+    {
+      "epoch": 0.511661358265694,
+      "grad_norm": 0.0910191684961319,
+      "learning_rate": 0.01,
+      "loss": 2.209,
+      "step": 4980
+    },
+    {
+      "epoch": 0.511969587999589,
+      "grad_norm": 0.12366749346256256,
+      "learning_rate": 0.01,
+      "loss": 2.2263,
+      "step": 4983
+    },
+    {
+      "epoch": 0.512277817733484,
+      "grad_norm": 0.11254429817199707,
+      "learning_rate": 0.01,
+      "loss": 2.175,
+      "step": 4986
+    },
+    {
+      "epoch": 0.512586047467379,
+      "grad_norm": 0.11091643571853638,
+      "learning_rate": 0.01,
+      "loss": 2.2046,
+      "step": 4989
+    },
+    {
+      "epoch": 0.512894277201274,
+      "grad_norm": 0.045611754059791565,
+      "learning_rate": 0.01,
+      "loss": 2.1861,
+      "step": 4992
+    },
+    {
+      "epoch": 0.513202506935169,
+      "grad_norm": 0.09836157411336899,
+      "learning_rate": 0.01,
+      "loss": 2.1931,
+      "step": 4995
+    },
+    {
+      "epoch": 0.513510736669064,
+      "grad_norm": 0.11932815611362457,
+      "learning_rate": 0.01,
+      "loss": 2.2087,
+      "step": 4998
+    },
+    {
+      "epoch": 0.513818966402959,
+      "grad_norm": 0.10955359041690826,
+      "learning_rate": 0.01,
+      "loss": 2.19,
+      "step": 5001
+    },
+    {
+      "epoch": 0.514127196136854,
+      "grad_norm": 0.09804633259773254,
+      "learning_rate": 0.01,
+      "loss": 2.1572,
+      "step": 5004
+    },
+    {
+      "epoch": 0.514435425870749,
+      "grad_norm": 0.04835839942097664,
+      "learning_rate": 0.01,
+      "loss": 2.2115,
+      "step": 5007
+    },
+    {
+      "epoch": 0.514743655604644,
+      "grad_norm": 0.04645110294222832,
+      "learning_rate": 0.01,
+      "loss": 2.2012,
+      "step": 5010
+    },
+    {
+      "epoch": 0.515051885338539,
+      "grad_norm": 0.05947386845946312,
+      "learning_rate": 0.01,
+      "loss": 2.2039,
+      "step": 5013
+    },
+    {
+      "epoch": 0.515360115072434,
+      "grad_norm": 0.05693971738219261,
+      "learning_rate": 0.01,
+      "loss": 2.1733,
+      "step": 5016
+    },
+    {
+      "epoch": 0.515668344806329,
+      "grad_norm": 0.0724320039153099,
+      "learning_rate": 0.01,
+      "loss": 2.1944,
+      "step": 5019
+    },
+    {
+      "epoch": 0.5159765745402239,
+      "grad_norm": 0.06627337634563446,
+      "learning_rate": 0.01,
+      "loss": 2.1932,
+      "step": 5022
+    },
+    {
+      "epoch": 0.5162848042741189,
+      "grad_norm": 0.10879958420991898,
+      "learning_rate": 0.01,
+      "loss": 2.2024,
+      "step": 5025
+    },
+    {
+      "epoch": 0.5165930340080139,
+      "grad_norm": 0.12266898900270462,
+      "learning_rate": 0.01,
+      "loss": 2.1938,
+      "step": 5028
+    },
+    {
+      "epoch": 0.5169012637419089,
+      "grad_norm": 0.06240540370345116,
+      "learning_rate": 0.01,
+      "loss": 2.18,
+      "step": 5031
+    },
+    {
+      "epoch": 0.5172094934758039,
+      "grad_norm": 0.05043266713619232,
+      "learning_rate": 0.01,
+      "loss": 2.1936,
+      "step": 5034
+    },
+    {
+      "epoch": 0.5175177232096989,
+      "grad_norm": 0.052652738988399506,
+      "learning_rate": 0.01,
+      "loss": 2.1631,
+      "step": 5037
+    },
+    {
+      "epoch": 0.5178259529435939,
+      "grad_norm": 0.04598904401063919,
+      "learning_rate": 0.01,
+      "loss": 2.2067,
+      "step": 5040
+    },
+    {
+      "epoch": 0.518134182677489,
+      "grad_norm": 0.07040087133646011,
+      "learning_rate": 0.01,
+      "loss": 2.1737,
+      "step": 5043
+    },
+    {
+      "epoch": 0.518442412411384,
+      "grad_norm": 0.04827702417969704,
+      "learning_rate": 0.01,
+      "loss": 2.2128,
+      "step": 5046
+    },
+    {
+      "epoch": 0.518750642145279,
+      "grad_norm": 0.09803622215986252,
+      "learning_rate": 0.01,
+      "loss": 2.1851,
+      "step": 5049
+    },
+    {
+      "epoch": 0.519058871879174,
+      "grad_norm": 0.1019926443696022,
+      "learning_rate": 0.01,
+      "loss": 2.181,
+      "step": 5052
+    },
+    {
+      "epoch": 0.519367101613069,
+      "grad_norm": 0.08847504109144211,
+      "learning_rate": 0.01,
+      "loss": 2.1874,
+      "step": 5055
+    },
+    {
+      "epoch": 0.519675331346964,
+      "grad_norm": 0.06151921674609184,
+      "learning_rate": 0.01,
+      "loss": 2.1785,
+      "step": 5058
+    },
+    {
+      "epoch": 0.519983561080859,
+      "grad_norm": 0.04823022335767746,
+      "learning_rate": 0.01,
+      "loss": 2.1662,
+      "step": 5061
+    },
+    {
+      "epoch": 0.520291790814754,
+      "grad_norm": 0.12454935908317566,
+      "learning_rate": 0.01,
+      "loss": 2.1864,
+      "step": 5064
+    },
+    {
+      "epoch": 0.520600020548649,
+      "grad_norm": 0.0716002956032753,
+      "learning_rate": 0.01,
+      "loss": 2.1866,
+      "step": 5067
+    },
+    {
+      "epoch": 0.520908250282544,
+      "grad_norm": 0.055079616606235504,
+      "learning_rate": 0.01,
+      "loss": 2.2137,
+      "step": 5070
+    },
+    {
+      "epoch": 0.5212164800164389,
+      "grad_norm": 0.05969909206032753,
+      "learning_rate": 0.01,
+      "loss": 2.1972,
+      "step": 5073
+    },
+    {
+      "epoch": 0.5215247097503339,
+      "grad_norm": 0.07373122125864029,
+      "learning_rate": 0.01,
+      "loss": 2.2312,
+      "step": 5076
+    },
+    {
+      "epoch": 0.5218329394842289,
+      "grad_norm": 0.1899929642677307,
+      "learning_rate": 0.01,
+      "loss": 2.2141,
+      "step": 5079
+    },
+    {
+      "epoch": 0.5221411692181239,
+      "grad_norm": 0.05221979692578316,
+      "learning_rate": 0.01,
+      "loss": 2.1899,
+      "step": 5082
+    },
+    {
+      "epoch": 0.5224493989520189,
+      "grad_norm": 0.04537337273359299,
+      "learning_rate": 0.01,
+      "loss": 2.1571,
+      "step": 5085
+    },
+    {
+      "epoch": 0.5227576286859139,
+      "grad_norm": 0.05490431934595108,
+      "learning_rate": 0.01,
+      "loss": 2.1705,
+      "step": 5088
+    },
+    {
+      "epoch": 0.5230658584198089,
+      "grad_norm": 0.03813198208808899,
+      "learning_rate": 0.01,
+      "loss": 2.1773,
+      "step": 5091
+    },
+    {
+      "epoch": 0.5233740881537039,
+      "grad_norm": 0.045411352068185806,
+      "learning_rate": 0.01,
+      "loss": 2.2065,
+      "step": 5094
+    },
+    {
+      "epoch": 0.5236823178875989,
+      "grad_norm": 0.05433456227183342,
+      "learning_rate": 0.01,
+      "loss": 2.1901,
+      "step": 5097
+    },
+    {
+      "epoch": 0.5239905476214939,
+      "grad_norm": 0.10771681368350983,
+      "learning_rate": 0.01,
+      "loss": 2.171,
+      "step": 5100
+    },
+    {
+      "epoch": 0.5242987773553889,
+      "grad_norm": 0.06446761637926102,
+      "learning_rate": 0.01,
+      "loss": 2.2033,
+      "step": 5103
+    },
+    {
+      "epoch": 0.5246070070892839,
+      "grad_norm": 0.06428392231464386,
+      "learning_rate": 0.01,
+      "loss": 2.1877,
+      "step": 5106
+    },
+    {
+      "epoch": 0.5249152368231789,
+      "grad_norm": 0.0525304451584816,
+      "learning_rate": 0.01,
+      "loss": 2.206,
+      "step": 5109
+    },
+    {
+      "epoch": 0.5252234665570739,
+      "grad_norm": 0.07332491129636765,
+      "learning_rate": 0.01,
+      "loss": 2.1992,
+      "step": 5112
+    },
+    {
+      "epoch": 0.5255316962909689,
+      "grad_norm": 0.1738174557685852,
+      "learning_rate": 0.01,
+      "loss": 2.176,
+      "step": 5115
+    },
+    {
+      "epoch": 0.5258399260248638,
+      "grad_norm": 0.08102334290742874,
+      "learning_rate": 0.01,
+      "loss": 2.2067,
+      "step": 5118
+    },
+    {
+      "epoch": 0.5261481557587588,
+      "grad_norm": 0.06945500522851944,
+      "learning_rate": 0.01,
+      "loss": 2.206,
+      "step": 5121
+    },
+    {
+      "epoch": 0.5264563854926538,
+      "grad_norm": 0.07017000019550323,
+      "learning_rate": 0.01,
+      "loss": 2.2002,
+      "step": 5124
+    },
+    {
+      "epoch": 0.5267646152265488,
+      "grad_norm": 0.03883346915245056,
+      "learning_rate": 0.01,
+      "loss": 2.1608,
+      "step": 5127
+    },
+    {
+      "epoch": 0.5270728449604438,
+      "grad_norm": 0.050974566489458084,
+      "learning_rate": 0.01,
+      "loss": 2.1973,
+      "step": 5130
+    },
+    {
+      "epoch": 0.5273810746943388,
+      "grad_norm": 0.0665312334895134,
+      "learning_rate": 0.01,
+      "loss": 2.2064,
+      "step": 5133
+    },
+    {
+      "epoch": 0.5276893044282338,
+      "grad_norm": 0.03946761414408684,
+      "learning_rate": 0.01,
+      "loss": 2.1794,
+      "step": 5136
+    },
+    {
+      "epoch": 0.5279975341621288,
+      "grad_norm": 0.046160902827978134,
+      "learning_rate": 0.01,
+      "loss": 2.1919,
+      "step": 5139
+    },
+    {
+      "epoch": 0.5283057638960238,
+      "grad_norm": 0.046186063438653946,
+      "learning_rate": 0.01,
+      "loss": 2.1687,
+      "step": 5142
+    },
+    {
+      "epoch": 0.5286139936299188,
+      "grad_norm": 0.07073774188756943,
+      "learning_rate": 0.01,
+      "loss": 2.2154,
+      "step": 5145
+    },
+    {
+      "epoch": 0.5289222233638138,
+      "grad_norm": 0.08319617807865143,
+      "learning_rate": 0.01,
+      "loss": 2.2104,
+      "step": 5148
+    },
+    {
+      "epoch": 0.5292304530977088,
+      "grad_norm": 0.08551453799009323,
+      "learning_rate": 0.01,
+      "loss": 2.1743,
+      "step": 5151
+    },
+    {
+      "epoch": 0.5295386828316038,
+      "grad_norm": 0.06613323837518692,
+      "learning_rate": 0.01,
+      "loss": 2.1774,
+      "step": 5154
+    },
+    {
+      "epoch": 0.5298469125654989,
+      "grad_norm": 0.0601351298391819,
+      "learning_rate": 0.01,
+      "loss": 2.1831,
+      "step": 5157
+    },
+    {
+      "epoch": 0.5301551422993939,
+      "grad_norm": 0.04963411018252373,
+      "learning_rate": 0.01,
+      "loss": 2.1915,
+      "step": 5160
+    },
+    {
+      "epoch": 0.5304633720332889,
+      "grad_norm": 0.06755329668521881,
+      "learning_rate": 0.01,
+      "loss": 2.163,
+      "step": 5163
+    },
+    {
+      "epoch": 0.5307716017671839,
+      "grad_norm": 0.04755258187651634,
+      "learning_rate": 0.01,
+      "loss": 2.1518,
+      "step": 5166
+    },
+    {
+      "epoch": 0.5310798315010788,
+      "grad_norm": 0.15916316211223602,
+      "learning_rate": 0.01,
+      "loss": 2.1835,
+      "step": 5169
+    },
+    {
+      "epoch": 0.5313880612349738,
+      "grad_norm": 0.0807122215628624,
+      "learning_rate": 0.01,
+      "loss": 2.2193,
+      "step": 5172
+    },
+    {
+      "epoch": 0.5316962909688688,
+      "grad_norm": 0.05207689106464386,
+      "learning_rate": 0.01,
+      "loss": 2.1754,
+      "step": 5175
+    },
+    {
+      "epoch": 0.5320045207027638,
+      "grad_norm": 0.045082803815603256,
+      "learning_rate": 0.01,
+      "loss": 2.1525,
+      "step": 5178
+    },
+    {
+      "epoch": 0.5323127504366588,
+      "grad_norm": 0.07747700810432434,
+      "learning_rate": 0.01,
+      "loss": 2.1438,
+      "step": 5181
+    },
+    {
+      "epoch": 0.5326209801705538,
+      "grad_norm": 0.13366450369358063,
+      "learning_rate": 0.01,
+      "loss": 2.1904,
+      "step": 5184
+    },
+    {
+      "epoch": 0.5329292099044488,
+      "grad_norm": 0.06902889162302017,
+      "learning_rate": 0.01,
+      "loss": 2.1786,
+      "step": 5187
+    },
+    {
+      "epoch": 0.5332374396383438,
+      "grad_norm": 0.04604712501168251,
+      "learning_rate": 0.01,
+      "loss": 2.1848,
+      "step": 5190
+    },
+    {
+      "epoch": 0.5335456693722388,
+      "grad_norm": 0.08803047984838486,
+      "learning_rate": 0.01,
+      "loss": 2.1798,
+      "step": 5193
+    },
+    {
+      "epoch": 0.5338538991061338,
+      "grad_norm": 0.08366485685110092,
+      "learning_rate": 0.01,
+      "loss": 2.2008,
+      "step": 5196
+    },
+    {
+      "epoch": 0.5341621288400288,
+      "grad_norm": 0.06176333501935005,
+      "learning_rate": 0.01,
+      "loss": 2.1722,
+      "step": 5199
+    },
+    {
+      "epoch": 0.5344703585739238,
+      "grad_norm": 0.0837249681353569,
+      "learning_rate": 0.01,
+      "loss": 2.1783,
+      "step": 5202
+    },
+    {
+      "epoch": 0.5347785883078188,
+      "grad_norm": 0.1322035938501358,
+      "learning_rate": 0.01,
+      "loss": 2.1948,
+      "step": 5205
+    },
+    {
+      "epoch": 0.5350868180417138,
+      "grad_norm": 0.11094444990158081,
+      "learning_rate": 0.01,
+      "loss": 2.1937,
+      "step": 5208
+    },
+    {
+      "epoch": 0.5353950477756088,
+      "grad_norm": 0.05182232707738876,
+      "learning_rate": 0.01,
+      "loss": 2.1945,
+      "step": 5211
+    },
+    {
+      "epoch": 0.5357032775095038,
+      "grad_norm": 0.08261944353580475,
+      "learning_rate": 0.01,
+      "loss": 2.1855,
+      "step": 5214
+    },
+    {
+      "epoch": 0.5360115072433987,
+      "grad_norm": 0.11097295582294464,
+      "learning_rate": 0.01,
+      "loss": 2.1902,
+      "step": 5217
+    },
+    {
+      "epoch": 0.5363197369772937,
+      "grad_norm": 0.05825675278902054,
+      "learning_rate": 0.01,
+      "loss": 2.1984,
+      "step": 5220
+    },
+    {
+      "epoch": 0.5366279667111887,
+      "grad_norm": 0.11612821370363235,
+      "learning_rate": 0.01,
+      "loss": 2.1679,
+      "step": 5223
+    },
+    {
+      "epoch": 0.5369361964450837,
+      "grad_norm": 0.09120064228773117,
+      "learning_rate": 0.01,
+      "loss": 2.1586,
+      "step": 5226
+    },
+    {
+      "epoch": 0.5372444261789787,
+      "grad_norm": 0.055181995034217834,
+      "learning_rate": 0.01,
+      "loss": 2.2169,
+      "step": 5229
+    },
+    {
+      "epoch": 0.5375526559128737,
+      "grad_norm": 0.055780068039894104,
+      "learning_rate": 0.01,
+      "loss": 2.1748,
+      "step": 5232
+    },
+    {
+      "epoch": 0.5378608856467687,
+      "grad_norm": 0.06303024291992188,
+      "learning_rate": 0.01,
+      "loss": 2.1782,
+      "step": 5235
+    },
+    {
+      "epoch": 0.5381691153806637,
+      "grad_norm": 0.10456321388483047,
+      "learning_rate": 0.01,
+      "loss": 2.1975,
+      "step": 5238
+    },
+    {
+      "epoch": 0.5384773451145587,
+      "grad_norm": 0.054177962243556976,
+      "learning_rate": 0.01,
+      "loss": 2.1641,
+      "step": 5241
+    },
+    {
+      "epoch": 0.5387855748484537,
+      "grad_norm": 0.06265738606452942,
+      "learning_rate": 0.01,
+      "loss": 2.183,
+      "step": 5244
+    },
+    {
+      "epoch": 0.5390938045823487,
+      "grad_norm": 0.13720418512821198,
+      "learning_rate": 0.01,
+      "loss": 2.1698,
+      "step": 5247
+    },
+    {
+      "epoch": 0.5394020343162437,
+      "grad_norm": 0.04917861148715019,
+      "learning_rate": 0.01,
+      "loss": 2.1692,
+      "step": 5250
+    },
+    {
+      "epoch": 0.5397102640501387,
+      "grad_norm": 0.04919945448637009,
+      "learning_rate": 0.01,
+      "loss": 2.1652,
+      "step": 5253
+    },
+    {
+      "epoch": 0.5400184937840337,
+      "grad_norm": 0.06462734192609787,
+      "learning_rate": 0.01,
+      "loss": 2.1987,
+      "step": 5256
+    },
+    {
+      "epoch": 0.5403267235179287,
+      "grad_norm": 0.05275480076670647,
+      "learning_rate": 0.01,
+      "loss": 2.1955,
+      "step": 5259
+    },
+    {
+      "epoch": 0.5406349532518236,
+      "grad_norm": 0.12235717475414276,
+      "learning_rate": 0.01,
+      "loss": 2.1937,
+      "step": 5262
+    },
+    {
+      "epoch": 0.5409431829857186,
+      "grad_norm": 0.05300014466047287,
+      "learning_rate": 0.01,
+      "loss": 2.1589,
+      "step": 5265
+    },
+    {
+      "epoch": 0.5412514127196136,
+      "grad_norm": 0.0429493710398674,
+      "learning_rate": 0.01,
+      "loss": 2.1618,
+      "step": 5268
+    },
+    {
+      "epoch": 0.5415596424535086,
+      "grad_norm": 0.07041259855031967,
+      "learning_rate": 0.01,
+      "loss": 2.1661,
+      "step": 5271
+    },
+    {
+      "epoch": 0.5418678721874037,
+      "grad_norm": 0.05304478853940964,
+      "learning_rate": 0.01,
+      "loss": 2.183,
+      "step": 5274
+    },
+    {
+      "epoch": 0.5421761019212987,
+      "grad_norm": 0.12009457498788834,
+      "learning_rate": 0.01,
+      "loss": 2.1862,
+      "step": 5277
+    },
+    {
+      "epoch": 0.5424843316551937,
+      "grad_norm": 0.11629784107208252,
+      "learning_rate": 0.01,
+      "loss": 2.1897,
+      "step": 5280
+    },
+    {
+      "epoch": 0.5427925613890887,
+      "grad_norm": 0.07305426150560379,
+      "learning_rate": 0.01,
+      "loss": 2.1832,
+      "step": 5283
+    },
+    {
+      "epoch": 0.5431007911229837,
+      "grad_norm": 0.0855623185634613,
+      "learning_rate": 0.01,
+      "loss": 2.1884,
+      "step": 5286
+    },
+    {
+      "epoch": 0.5434090208568787,
+      "grad_norm": 0.04178578779101372,
+      "learning_rate": 0.01,
+      "loss": 2.1872,
+      "step": 5289
+    },
+    {
+      "epoch": 0.5437172505907737,
+      "grad_norm": 0.05382310971617699,
+      "learning_rate": 0.01,
+      "loss": 2.1901,
+      "step": 5292
+    },
+    {
+      "epoch": 0.5440254803246687,
+      "grad_norm": 0.10682760924100876,
+      "learning_rate": 0.01,
+      "loss": 2.1957,
+      "step": 5295
+    },
+    {
+      "epoch": 0.5443337100585637,
+      "grad_norm": 0.15037471055984497,
+      "learning_rate": 0.01,
+      "loss": 2.2085,
+      "step": 5298
+    },
+    {
+      "epoch": 0.5446419397924587,
+      "grad_norm": 0.08333491533994675,
+      "learning_rate": 0.01,
+      "loss": 2.1964,
+      "step": 5301
+    },
+    {
+      "epoch": 0.5449501695263537,
+      "grad_norm": 0.08964785933494568,
+      "learning_rate": 0.01,
+      "loss": 2.1613,
+      "step": 5304
+    },
+    {
+      "epoch": 0.5452583992602487,
+      "grad_norm": 0.06194687634706497,
+      "learning_rate": 0.01,
+      "loss": 2.1711,
+      "step": 5307
+    },
+    {
+      "epoch": 0.5455666289941437,
+      "grad_norm": 0.047254305332899094,
+      "learning_rate": 0.01,
+      "loss": 2.1956,
+      "step": 5310
+    },
+    {
+      "epoch": 0.5458748587280386,
+      "grad_norm": 0.052926719188690186,
+      "learning_rate": 0.01,
+      "loss": 2.1767,
+      "step": 5313
+    },
+    {
+      "epoch": 0.5461830884619336,
+      "grad_norm": 0.08765383809804916,
+      "learning_rate": 0.01,
+      "loss": 2.1782,
+      "step": 5316
+    },
+    {
+      "epoch": 0.5464913181958286,
+      "grad_norm": 0.0749160572886467,
+      "learning_rate": 0.01,
+      "loss": 2.1875,
+      "step": 5319
+    },
+    {
+      "epoch": 0.5467995479297236,
+      "grad_norm": 0.09781020879745483,
+      "learning_rate": 0.01,
+      "loss": 2.1748,
+      "step": 5322
+    },
+    {
+      "epoch": 0.5471077776636186,
+      "grad_norm": 0.04605260491371155,
+      "learning_rate": 0.01,
+      "loss": 2.145,
+      "step": 5325
+    },
+    {
+      "epoch": 0.5474160073975136,
+      "grad_norm": 0.13507331907749176,
+      "learning_rate": 0.01,
+      "loss": 2.1769,
+      "step": 5328
+    },
+    {
+      "epoch": 0.5477242371314086,
+      "grad_norm": 0.05028709024190903,
+      "learning_rate": 0.01,
+      "loss": 2.1925,
+      "step": 5331
+    },
+    {
+      "epoch": 0.5480324668653036,
+      "grad_norm": 0.08754327893257141,
+      "learning_rate": 0.01,
+      "loss": 2.159,
+      "step": 5334
+    },
+    {
+      "epoch": 0.5483406965991986,
+      "grad_norm": 0.10449190437793732,
+      "learning_rate": 0.01,
+      "loss": 2.1898,
+      "step": 5337
+    },
+    {
+      "epoch": 0.5486489263330936,
+      "grad_norm": 0.10263057053089142,
+      "learning_rate": 0.01,
+      "loss": 2.1776,
+      "step": 5340
+    },
+    {
+      "epoch": 0.5489571560669886,
+      "grad_norm": 0.0547097772359848,
+      "learning_rate": 0.01,
+      "loss": 2.1941,
+      "step": 5343
+    },
+    {
+      "epoch": 0.5492653858008836,
+      "grad_norm": 0.06682941317558289,
+      "learning_rate": 0.01,
+      "loss": 2.1901,
+      "step": 5346
+    },
+    {
+      "epoch": 0.5495736155347786,
+      "grad_norm": 0.06421027332544327,
+      "learning_rate": 0.01,
+      "loss": 2.1981,
+      "step": 5349
+    },
+    {
+      "epoch": 0.5498818452686736,
+      "grad_norm": 0.041218411177396774,
+      "learning_rate": 0.01,
+      "loss": 2.1844,
+      "step": 5352
+    },
+    {
+      "epoch": 0.5501900750025686,
+      "grad_norm": 0.042902372777462006,
+      "learning_rate": 0.01,
+      "loss": 2.1981,
+      "step": 5355
+    },
+    {
+      "epoch": 0.5504983047364636,
+      "grad_norm": 0.05338321253657341,
+      "learning_rate": 0.01,
+      "loss": 2.168,
+      "step": 5358
+    },
+    {
+      "epoch": 0.5508065344703585,
+      "grad_norm": 0.06692427396774292,
+      "learning_rate": 0.01,
+      "loss": 2.1891,
+      "step": 5361
+    },
+    {
+      "epoch": 0.5511147642042535,
+      "grad_norm": 0.07927200943231583,
+      "learning_rate": 0.01,
+      "loss": 2.1853,
+      "step": 5364
+    },
+    {
+      "epoch": 0.5514229939381485,
+      "grad_norm": 0.05655739828944206,
+      "learning_rate": 0.01,
+      "loss": 2.1838,
+      "step": 5367
+    },
+    {
+      "epoch": 0.5517312236720435,
+      "grad_norm": 0.04488144442439079,
+      "learning_rate": 0.01,
+      "loss": 2.1754,
+      "step": 5370
+    },
+    {
+      "epoch": 0.5520394534059385,
+      "grad_norm": 0.09253795444965363,
+      "learning_rate": 0.01,
+      "loss": 2.1742,
+      "step": 5373
+    },
+    {
+      "epoch": 0.5523476831398335,
+      "grad_norm": 0.07396019250154495,
+      "learning_rate": 0.01,
+      "loss": 2.1582,
+      "step": 5376
+    },
+    {
+      "epoch": 0.5526559128737285,
+      "grad_norm": 0.053663600236177444,
+      "learning_rate": 0.01,
+      "loss": 2.1508,
+      "step": 5379
+    },
+    {
+      "epoch": 0.5529641426076235,
+      "grad_norm": 0.062076181173324585,
+      "learning_rate": 0.01,
+      "loss": 2.1772,
+      "step": 5382
+    },
+    {
+      "epoch": 0.5532723723415185,
+      "grad_norm": 0.08481581509113312,
+      "learning_rate": 0.01,
+      "loss": 2.1836,
+      "step": 5385
+    },
+    {
+      "epoch": 0.5535806020754136,
+      "grad_norm": 0.08981155604124069,
+      "learning_rate": 0.01,
+      "loss": 2.204,
+      "step": 5388
+    },
+    {
+      "epoch": 0.5538888318093086,
+      "grad_norm": 0.10067261755466461,
+      "learning_rate": 0.01,
+      "loss": 2.1527,
+      "step": 5391
+    },
+    {
+      "epoch": 0.5541970615432036,
+      "grad_norm": 0.06231047958135605,
+      "learning_rate": 0.01,
+      "loss": 2.194,
+      "step": 5394
+    },
+    {
+      "epoch": 0.5545052912770986,
+      "grad_norm": 0.049111973494291306,
+      "learning_rate": 0.01,
+      "loss": 2.1889,
+      "step": 5397
+    },
+    {
+      "epoch": 0.5548135210109936,
+      "grad_norm": 0.06446948647499084,
+      "learning_rate": 0.01,
+      "loss": 2.2103,
+      "step": 5400
+    },
+    {
+      "epoch": 0.5551217507448886,
+      "grad_norm": 0.051946625113487244,
+      "learning_rate": 0.01,
+      "loss": 2.1977,
+      "step": 5403
+    },
+    {
+      "epoch": 0.5554299804787836,
+      "grad_norm": 0.1369265466928482,
+      "learning_rate": 0.01,
+      "loss": 2.1771,
+      "step": 5406
+    },
+    {
+      "epoch": 0.5557382102126786,
+      "grad_norm": 0.08489779382944107,
+      "learning_rate": 0.01,
+      "loss": 2.1782,
+      "step": 5409
+    },
+    {
+      "epoch": 0.5560464399465735,
+      "grad_norm": 0.10673670470714569,
+      "learning_rate": 0.01,
+      "loss": 2.173,
+      "step": 5412
+    },
+    {
+      "epoch": 0.5563546696804685,
+      "grad_norm": 0.055250637233257294,
+      "learning_rate": 0.01,
+      "loss": 2.1539,
+      "step": 5415
+    },
+    {
+      "epoch": 0.5566628994143635,
+      "grad_norm": 0.05136672407388687,
+      "learning_rate": 0.01,
+      "loss": 2.2035,
+      "step": 5418
+    },
+    {
+      "epoch": 0.5569711291482585,
+      "grad_norm": 0.040590591728687286,
+      "learning_rate": 0.01,
+      "loss": 2.1778,
+      "step": 5421
+    },
+    {
+      "epoch": 0.5572793588821535,
+      "grad_norm": 0.048333633691072464,
+      "learning_rate": 0.01,
+      "loss": 2.191,
+      "step": 5424
+    },
+    {
+      "epoch": 0.5575875886160485,
+      "grad_norm": 0.0582728311419487,
+      "learning_rate": 0.01,
+      "loss": 2.1734,
+      "step": 5427
+    },
+    {
+      "epoch": 0.5578958183499435,
+      "grad_norm": 0.05272262915968895,
+      "learning_rate": 0.01,
+      "loss": 2.1714,
+      "step": 5430
+    },
+    {
+      "epoch": 0.5582040480838385,
+      "grad_norm": 0.08472342789173126,
+      "learning_rate": 0.01,
+      "loss": 2.1624,
+      "step": 5433
+    },
+    {
+      "epoch": 0.5585122778177335,
+      "grad_norm": 0.10869960486888885,
+      "learning_rate": 0.01,
+      "loss": 2.164,
+      "step": 5436
+    },
+    {
+      "epoch": 0.5588205075516285,
+      "grad_norm": 0.0569114163517952,
+      "learning_rate": 0.01,
+      "loss": 2.1933,
+      "step": 5439
+    },
+    {
+      "epoch": 0.5591287372855235,
+      "grad_norm": 0.14485467970371246,
+      "learning_rate": 0.01,
+      "loss": 2.1779,
+      "step": 5442
+    },
+    {
+      "epoch": 0.5594369670194185,
+      "grad_norm": 0.08184878528118134,
+      "learning_rate": 0.01,
+      "loss": 2.1779,
+      "step": 5445
+    },
+    {
+      "epoch": 0.5597451967533135,
+      "grad_norm": 0.06575775146484375,
+      "learning_rate": 0.01,
+      "loss": 2.136,
+      "step": 5448
+    },
+    {
+      "epoch": 0.5600534264872085,
+      "grad_norm": 0.08628299832344055,
+      "learning_rate": 0.01,
+      "loss": 2.1696,
+      "step": 5451
+    },
+    {
+      "epoch": 0.5603616562211035,
+      "grad_norm": 0.06078352406620979,
+      "learning_rate": 0.01,
+      "loss": 2.1865,
+      "step": 5454
+    },
+    {
+      "epoch": 0.5606698859549984,
+      "grad_norm": 0.05207353085279465,
+      "learning_rate": 0.01,
+      "loss": 2.167,
+      "step": 5457
+    },
+    {
+      "epoch": 0.5609781156888934,
+      "grad_norm": 0.059535857290029526,
+      "learning_rate": 0.01,
+      "loss": 2.1977,
+      "step": 5460
+    },
+    {
+      "epoch": 0.5612863454227884,
+      "grad_norm": 0.05342729389667511,
+      "learning_rate": 0.01,
+      "loss": 2.1823,
+      "step": 5463
+    },
+    {
+      "epoch": 0.5615945751566834,
+      "grad_norm": 0.04207632318139076,
+      "learning_rate": 0.01,
+      "loss": 2.1849,
+      "step": 5466
+    },
+    {
+      "epoch": 0.5619028048905784,
+      "grad_norm": 0.1334255486726761,
+      "learning_rate": 0.01,
+      "loss": 2.1886,
+      "step": 5469
+    },
+    {
+      "epoch": 0.5622110346244734,
+      "grad_norm": 0.06532323360443115,
+      "learning_rate": 0.01,
+      "loss": 2.1979,
+      "step": 5472
+    },
+    {
+      "epoch": 0.5625192643583684,
+      "grad_norm": 0.0793483555316925,
+      "learning_rate": 0.01,
+      "loss": 2.188,
+      "step": 5475
+    },
+    {
+      "epoch": 0.5628274940922634,
+      "grad_norm": 0.04637480154633522,
+      "learning_rate": 0.01,
+      "loss": 2.1562,
+      "step": 5478
+    },
+    {
+      "epoch": 0.5631357238261584,
+      "grad_norm": 0.0482000894844532,
+      "learning_rate": 0.01,
+      "loss": 2.1587,
+      "step": 5481
+    },
+    {
+      "epoch": 0.5634439535600534,
+      "grad_norm": 0.06253401190042496,
+      "learning_rate": 0.01,
+      "loss": 2.1978,
+      "step": 5484
+    },
+    {
+      "epoch": 0.5637521832939484,
+      "grad_norm": 0.15622715651988983,
+      "learning_rate": 0.01,
+      "loss": 2.171,
+      "step": 5487
+    },
+    {
+      "epoch": 0.5640604130278434,
+      "grad_norm": 0.10941077768802643,
+      "learning_rate": 0.01,
+      "loss": 2.1952,
+      "step": 5490
+    },
+    {
+      "epoch": 0.5643686427617384,
+      "grad_norm": 0.08030713349580765,
+      "learning_rate": 0.01,
+      "loss": 2.1948,
+      "step": 5493
+    },
+    {
+      "epoch": 0.5646768724956334,
+      "grad_norm": 0.13679014146327972,
+      "learning_rate": 0.01,
+      "loss": 2.1767,
+      "step": 5496
+    },
+    {
+      "epoch": 0.5649851022295284,
+      "grad_norm": 0.04662426933646202,
+      "learning_rate": 0.01,
+      "loss": 2.1926,
+      "step": 5499
+    },
+    {
+      "epoch": 0.5652933319634234,
+      "grad_norm": 0.05347858741879463,
+      "learning_rate": 0.01,
+      "loss": 2.1825,
+      "step": 5502
+    },
+    {
+      "epoch": 0.5656015616973185,
+      "grad_norm": 0.06205238029360771,
+      "learning_rate": 0.01,
+      "loss": 2.1537,
+      "step": 5505
+    },
+    {
+      "epoch": 0.5659097914312134,
+      "grad_norm": 0.05525955557823181,
+      "learning_rate": 0.01,
+      "loss": 2.1802,
+      "step": 5508
+    },
+    {
+      "epoch": 0.5662180211651084,
+      "grad_norm": 0.055693045258522034,
+      "learning_rate": 0.01,
+      "loss": 2.1727,
+      "step": 5511
+    },
+    {
+      "epoch": 0.5665262508990034,
+      "grad_norm": 0.051134396344423294,
+      "learning_rate": 0.01,
+      "loss": 2.1695,
+      "step": 5514
+    },
+    {
+      "epoch": 0.5668344806328984,
+      "grad_norm": 0.05469521880149841,
+      "learning_rate": 0.01,
+      "loss": 2.1664,
+      "step": 5517
+    },
+    {
+      "epoch": 0.5671427103667934,
+      "grad_norm": 0.039416272193193436,
+      "learning_rate": 0.01,
+      "loss": 2.1708,
+      "step": 5520
+    },
+    {
+      "epoch": 0.5674509401006884,
+      "grad_norm": 0.10661659389734268,
+      "learning_rate": 0.01,
+      "loss": 2.1753,
+      "step": 5523
+    },
+    {
+      "epoch": 0.5677591698345834,
+      "grad_norm": 0.07567829638719559,
+      "learning_rate": 0.01,
+      "loss": 2.1733,
+      "step": 5526
+    },
+    {
+      "epoch": 0.5680673995684784,
+      "grad_norm": 0.06030309572815895,
+      "learning_rate": 0.01,
+      "loss": 2.1795,
+      "step": 5529
+    },
+    {
+      "epoch": 0.5683756293023734,
+      "grad_norm": 0.07429811358451843,
+      "learning_rate": 0.01,
+      "loss": 2.1936,
+      "step": 5532
+    },
+    {
+      "epoch": 0.5686838590362684,
+      "grad_norm": 0.08618849515914917,
+      "learning_rate": 0.01,
+      "loss": 2.2009,
+      "step": 5535
+    },
+    {
+      "epoch": 0.5689920887701634,
+      "grad_norm": 0.04969833791255951,
+      "learning_rate": 0.01,
+      "loss": 2.1711,
+      "step": 5538
+    },
+    {
+      "epoch": 0.5693003185040584,
+      "grad_norm": 0.11154712736606598,
+      "learning_rate": 0.01,
+      "loss": 2.1802,
+      "step": 5541
+    },
+    {
+      "epoch": 0.5696085482379534,
+      "grad_norm": 0.07754155993461609,
+      "learning_rate": 0.01,
+      "loss": 2.164,
+      "step": 5544
+    },
+    {
+      "epoch": 0.5699167779718484,
+      "grad_norm": 0.04600273445248604,
+      "learning_rate": 0.01,
+      "loss": 2.1918,
+      "step": 5547
+    },
+    {
+      "epoch": 0.5702250077057434,
+      "grad_norm": 0.06788063049316406,
+      "learning_rate": 0.01,
+      "loss": 2.1477,
+      "step": 5550
+    },
+    {
+      "epoch": 0.5705332374396384,
+      "grad_norm": 0.11349419504404068,
+      "learning_rate": 0.01,
+      "loss": 2.1603,
+      "step": 5553
+    },
+    {
+      "epoch": 0.5708414671735333,
+      "grad_norm": 0.11178430914878845,
+      "learning_rate": 0.01,
+      "loss": 2.1439,
+      "step": 5556
+    },
+    {
+      "epoch": 0.5711496969074283,
+      "grad_norm": 0.050257500261068344,
+      "learning_rate": 0.01,
+      "loss": 2.1851,
+      "step": 5559
+    },
+    {
+      "epoch": 0.5714579266413233,
+      "grad_norm": 0.08327756822109222,
+      "learning_rate": 0.01,
+      "loss": 2.158,
+      "step": 5562
+    },
+    {
+      "epoch": 0.5717661563752183,
+      "grad_norm": 0.06866388767957687,
+      "learning_rate": 0.01,
+      "loss": 2.1584,
+      "step": 5565
+    },
+    {
+      "epoch": 0.5720743861091133,
+      "grad_norm": 0.1139674037694931,
+      "learning_rate": 0.01,
+      "loss": 2.1897,
+      "step": 5568
+    },
+    {
+      "epoch": 0.5723826158430083,
+      "grad_norm": 0.07029612362384796,
+      "learning_rate": 0.01,
+      "loss": 2.1723,
+      "step": 5571
+    },
+    {
+      "epoch": 0.5726908455769033,
+      "grad_norm": 0.10171212255954742,
+      "learning_rate": 0.01,
+      "loss": 2.1788,
+      "step": 5574
+    },
+    {
+      "epoch": 0.5729990753107983,
+      "grad_norm": 0.11364202946424484,
+      "learning_rate": 0.01,
+      "loss": 2.1659,
+      "step": 5577
+    },
+    {
+      "epoch": 0.5733073050446933,
+      "grad_norm": 0.08066857606172562,
+      "learning_rate": 0.01,
+      "loss": 2.1902,
+      "step": 5580
+    },
+    {
+      "epoch": 0.5736155347785883,
+      "grad_norm": 0.09207342565059662,
+      "learning_rate": 0.01,
+      "loss": 2.1519,
+      "step": 5583
+    },
+    {
+      "epoch": 0.5739237645124833,
+      "grad_norm": 0.06945987790822983,
+      "learning_rate": 0.01,
+      "loss": 2.1677,
+      "step": 5586
+    },
+    {
+      "epoch": 0.5742319942463783,
+      "grad_norm": 0.05137445777654648,
+      "learning_rate": 0.01,
+      "loss": 2.1686,
+      "step": 5589
+    },
+    {
+      "epoch": 0.5745402239802733,
+      "grad_norm": 0.10192268341779709,
+      "learning_rate": 0.01,
+      "loss": 2.1758,
+      "step": 5592
+    },
+    {
+      "epoch": 0.5748484537141683,
+      "grad_norm": 0.056787896901369095,
+      "learning_rate": 0.01,
+      "loss": 2.1642,
+      "step": 5595
+    },
+    {
+      "epoch": 0.5751566834480633,
+      "grad_norm": 0.07727455347776413,
+      "learning_rate": 0.01,
+      "loss": 2.1662,
+      "step": 5598
+    },
+    {
+      "epoch": 0.5754649131819582,
+      "grad_norm": 0.1311456710100174,
+      "learning_rate": 0.01,
+      "loss": 2.1713,
+      "step": 5601
+    },
+    {
+      "epoch": 0.5757731429158532,
+      "grad_norm": 0.1014258936047554,
+      "learning_rate": 0.01,
+      "loss": 2.1751,
+      "step": 5604
+    },
+    {
+      "epoch": 0.5760813726497482,
+      "grad_norm": 0.06325560063123703,
+      "learning_rate": 0.01,
+      "loss": 2.1757,
+      "step": 5607
+    },
+    {
+      "epoch": 0.5763896023836432,
+      "grad_norm": 0.07262448221445084,
+      "learning_rate": 0.01,
+      "loss": 2.1575,
+      "step": 5610
+    },
+    {
+      "epoch": 0.5766978321175382,
+      "grad_norm": 0.07160039991140366,
+      "learning_rate": 0.01,
+      "loss": 2.1706,
+      "step": 5613
+    },
+    {
+      "epoch": 0.5770060618514332,
+      "grad_norm": 0.050024017691612244,
+      "learning_rate": 0.01,
+      "loss": 2.1811,
+      "step": 5616
+    },
+    {
+      "epoch": 0.5773142915853282,
+      "grad_norm": 0.09685138612985611,
+      "learning_rate": 0.01,
+      "loss": 2.1549,
+      "step": 5619
+    },
+    {
+      "epoch": 0.5776225213192233,
+      "grad_norm": 0.058329988270998,
+      "learning_rate": 0.01,
+      "loss": 2.1813,
+      "step": 5622
+    },
+    {
+      "epoch": 0.5779307510531183,
+      "grad_norm": 0.06637705117464066,
+      "learning_rate": 0.01,
+      "loss": 2.1717,
+      "step": 5625
+    },
+    {
+      "epoch": 0.5782389807870133,
+      "grad_norm": 0.0906175896525383,
+      "learning_rate": 0.01,
+      "loss": 2.1677,
+      "step": 5628
+    },
+    {
+      "epoch": 0.5785472105209083,
+      "grad_norm": 0.06751519441604614,
+      "learning_rate": 0.01,
+      "loss": 2.1584,
+      "step": 5631
+    },
+    {
+      "epoch": 0.5788554402548033,
+      "grad_norm": 0.04437318444252014,
+      "learning_rate": 0.01,
+      "loss": 2.2013,
+      "step": 5634
+    },
+    {
+      "epoch": 0.5791636699886983,
+      "grad_norm": 0.04365368187427521,
+      "learning_rate": 0.01,
+      "loss": 2.1746,
+      "step": 5637
+    },
+    {
+      "epoch": 0.5794718997225933,
+      "grad_norm": 0.04844829812645912,
+      "learning_rate": 0.01,
+      "loss": 2.1818,
+      "step": 5640
+    },
+    {
+      "epoch": 0.5797801294564883,
+      "grad_norm": 0.04154437035322189,
+      "learning_rate": 0.01,
+      "loss": 2.1536,
+      "step": 5643
+    },
+    {
+      "epoch": 0.5800883591903833,
+      "grad_norm": 0.07691098004579544,
+      "learning_rate": 0.01,
+      "loss": 2.1883,
+      "step": 5646
+    },
+    {
+      "epoch": 0.5803965889242783,
+      "grad_norm": 0.07065980136394501,
+      "learning_rate": 0.01,
+      "loss": 2.154,
+      "step": 5649
+    },
+    {
+      "epoch": 0.5807048186581732,
+      "grad_norm": 0.1051129475235939,
+      "learning_rate": 0.01,
+      "loss": 2.1447,
+      "step": 5652
+    },
+    {
+      "epoch": 0.5810130483920682,
+      "grad_norm": 0.10921964794397354,
+      "learning_rate": 0.01,
+      "loss": 2.1777,
+      "step": 5655
+    },
+    {
+      "epoch": 0.5813212781259632,
+      "grad_norm": 0.1120898649096489,
+      "learning_rate": 0.01,
+      "loss": 2.1617,
+      "step": 5658
+    },
+    {
+      "epoch": 0.5816295078598582,
+      "grad_norm": 0.09593590348958969,
+      "learning_rate": 0.01,
+      "loss": 2.1457,
+      "step": 5661
+    },
+    {
+      "epoch": 0.5819377375937532,
+      "grad_norm": 0.054108936339616776,
+      "learning_rate": 0.01,
+      "loss": 2.1793,
+      "step": 5664
+    },
+    {
+      "epoch": 0.5822459673276482,
+      "grad_norm": 0.07890141755342484,
+      "learning_rate": 0.01,
+      "loss": 2.1676,
+      "step": 5667
+    },
+    {
+      "epoch": 0.5825541970615432,
+      "grad_norm": 0.07864063233137131,
+      "learning_rate": 0.01,
+      "loss": 2.1766,
+      "step": 5670
+    },
+    {
+      "epoch": 0.5828624267954382,
+      "grad_norm": 0.08160068094730377,
+      "learning_rate": 0.01,
+      "loss": 2.166,
+      "step": 5673
+    },
+    {
+      "epoch": 0.5831706565293332,
+      "grad_norm": 0.08126121759414673,
+      "learning_rate": 0.01,
+      "loss": 2.1691,
+      "step": 5676
+    },
+    {
+      "epoch": 0.5834788862632282,
+      "grad_norm": 0.05922897160053253,
+      "learning_rate": 0.01,
+      "loss": 2.1854,
+      "step": 5679
+    },
+    {
+      "epoch": 0.5837871159971232,
+      "grad_norm": 0.1024496778845787,
+      "learning_rate": 0.01,
+      "loss": 2.1818,
+      "step": 5682
+    },
+    {
+      "epoch": 0.5840953457310182,
+      "grad_norm": 0.08880037069320679,
+      "learning_rate": 0.01,
+      "loss": 2.2054,
+      "step": 5685
+    },
+    {
+      "epoch": 0.5844035754649132,
+      "grad_norm": 0.04404953494668007,
+      "learning_rate": 0.01,
+      "loss": 2.1524,
+      "step": 5688
+    },
+    {
+      "epoch": 0.5847118051988082,
+      "grad_norm": 0.05817687511444092,
+      "learning_rate": 0.01,
+      "loss": 2.1813,
+      "step": 5691
+    },
+    {
+      "epoch": 0.5850200349327032,
+      "grad_norm": 0.047581762075424194,
+      "learning_rate": 0.01,
+      "loss": 2.1545,
+      "step": 5694
+    },
+    {
+      "epoch": 0.5853282646665982,
+      "grad_norm": 0.11034911125898361,
+      "learning_rate": 0.01,
+      "loss": 2.1803,
+      "step": 5697
+    },
+    {
+      "epoch": 0.5856364944004931,
+      "grad_norm": 0.05118945613503456,
+      "learning_rate": 0.01,
+      "loss": 2.1314,
+      "step": 5700
+    },
+    {
+      "epoch": 0.5859447241343881,
+      "grad_norm": 0.048316795378923416,
+      "learning_rate": 0.01,
+      "loss": 2.1711,
+      "step": 5703
+    },
+    {
+      "epoch": 0.5862529538682831,
+      "grad_norm": 0.12578584253787994,
+      "learning_rate": 0.01,
+      "loss": 2.1636,
+      "step": 5706
+    },
+    {
+      "epoch": 0.5865611836021781,
+      "grad_norm": 0.06594375520944595,
+      "learning_rate": 0.01,
+      "loss": 2.1977,
+      "step": 5709
+    },
+    {
+      "epoch": 0.5868694133360731,
+      "grad_norm": 0.060622964054346085,
+      "learning_rate": 0.01,
+      "loss": 2.1408,
+      "step": 5712
+    },
+    {
+      "epoch": 0.5871776430699681,
+      "grad_norm": 0.10055366903543472,
+      "learning_rate": 0.01,
+      "loss": 2.1999,
+      "step": 5715
+    },
+    {
+      "epoch": 0.5874858728038631,
+      "grad_norm": 0.10235504060983658,
+      "learning_rate": 0.01,
+      "loss": 2.1337,
+      "step": 5718
+    },
+    {
+      "epoch": 0.5877941025377581,
+      "grad_norm": 0.07707791030406952,
+      "learning_rate": 0.01,
+      "loss": 2.1387,
+      "step": 5721
+    },
+    {
+      "epoch": 0.5881023322716531,
+      "grad_norm": 0.05508594587445259,
+      "learning_rate": 0.01,
+      "loss": 2.1494,
+      "step": 5724
+    },
+    {
+      "epoch": 0.5884105620055481,
+      "grad_norm": 0.06580860912799835,
+      "learning_rate": 0.01,
+      "loss": 2.1598,
+      "step": 5727
+    },
+    {
+      "epoch": 0.5887187917394431,
+      "grad_norm": 0.07102775573730469,
+      "learning_rate": 0.01,
+      "loss": 2.1618,
+      "step": 5730
+    },
+    {
+      "epoch": 0.5890270214733381,
+      "grad_norm": 0.06750554591417313,
+      "learning_rate": 0.01,
+      "loss": 2.1782,
+      "step": 5733
+    },
+    {
+      "epoch": 0.5893352512072332,
+      "grad_norm": 0.07100195437669754,
+      "learning_rate": 0.01,
+      "loss": 2.1456,
+      "step": 5736
+    },
+    {
+      "epoch": 0.5896434809411282,
+      "grad_norm": 0.10585575550794601,
+      "learning_rate": 0.01,
+      "loss": 2.1751,
+      "step": 5739
+    },
+    {
+      "epoch": 0.5899517106750232,
+      "grad_norm": 0.055082373321056366,
+      "learning_rate": 0.01,
+      "loss": 2.1808,
+      "step": 5742
+    },
+    {
+      "epoch": 0.5902599404089182,
+      "grad_norm": 0.06285014003515244,
+      "learning_rate": 0.01,
+      "loss": 2.1588,
+      "step": 5745
+    },
+    {
+      "epoch": 0.5905681701428132,
+      "grad_norm": 0.13328112661838531,
+      "learning_rate": 0.01,
+      "loss": 2.177,
+      "step": 5748
+    },
+    {
+      "epoch": 0.5908763998767081,
+      "grad_norm": 0.08568006008863449,
+      "learning_rate": 0.01,
+      "loss": 2.1559,
+      "step": 5751
+    },
+    {
+      "epoch": 0.5911846296106031,
+      "grad_norm": 0.07850711792707443,
+      "learning_rate": 0.01,
+      "loss": 2.2047,
+      "step": 5754
+    },
+    {
+      "epoch": 0.5914928593444981,
+      "grad_norm": 0.07706760615110397,
+      "learning_rate": 0.01,
+      "loss": 2.1602,
+      "step": 5757
+    },
+    {
+      "epoch": 0.5918010890783931,
+      "grad_norm": 0.07512292265892029,
+      "learning_rate": 0.01,
+      "loss": 2.1871,
+      "step": 5760
+    },
+    {
+      "epoch": 0.5921093188122881,
+      "grad_norm": 0.059620197862386703,
+      "learning_rate": 0.01,
+      "loss": 2.1484,
+      "step": 5763
+    },
+    {
+      "epoch": 0.5924175485461831,
+      "grad_norm": 0.04021789878606796,
+      "learning_rate": 0.01,
+      "loss": 2.1651,
+      "step": 5766
+    },
+    {
+      "epoch": 0.5927257782800781,
+      "grad_norm": 0.050683967769145966,
+      "learning_rate": 0.01,
+      "loss": 2.1693,
+      "step": 5769
+    },
+    {
+      "epoch": 0.5930340080139731,
+      "grad_norm": 0.07091210782527924,
+      "learning_rate": 0.01,
+      "loss": 2.1851,
+      "step": 5772
+    },
+    {
+      "epoch": 0.5933422377478681,
+      "grad_norm": 0.09877889603376389,
+      "learning_rate": 0.01,
+      "loss": 2.1642,
+      "step": 5775
+    },
+    {
+      "epoch": 0.5936504674817631,
+      "grad_norm": 0.08692251145839691,
+      "learning_rate": 0.01,
+      "loss": 2.1755,
+      "step": 5778
+    },
+    {
+      "epoch": 0.5939586972156581,
+      "grad_norm": 0.06255677342414856,
+      "learning_rate": 0.01,
+      "loss": 2.1634,
+      "step": 5781
+    },
+    {
+      "epoch": 0.5942669269495531,
+      "grad_norm": 0.05615478754043579,
+      "learning_rate": 0.01,
+      "loss": 2.1909,
+      "step": 5784
+    },
+    {
+      "epoch": 0.5945751566834481,
+      "grad_norm": 0.04576956480741501,
+      "learning_rate": 0.01,
+      "loss": 2.1519,
+      "step": 5787
+    },
+    {
+      "epoch": 0.5948833864173431,
+      "grad_norm": 0.044911667704582214,
+      "learning_rate": 0.01,
+      "loss": 2.1697,
+      "step": 5790
+    },
+    {
+      "epoch": 0.5951916161512381,
+      "grad_norm": 0.07787128537893295,
+      "learning_rate": 0.01,
+      "loss": 2.1611,
+      "step": 5793
+    },
+    {
+      "epoch": 0.595499845885133,
+      "grad_norm": 0.06199866533279419,
+      "learning_rate": 0.01,
+      "loss": 2.1576,
+      "step": 5796
+    },
+    {
+      "epoch": 0.595808075619028,
+      "grad_norm": 0.07048948854207993,
+      "learning_rate": 0.01,
+      "loss": 2.1721,
+      "step": 5799
+    },
+    {
+      "epoch": 0.596116305352923,
+      "grad_norm": 0.1173306256532669,
+      "learning_rate": 0.01,
+      "loss": 2.1573,
+      "step": 5802
+    },
+    {
+      "epoch": 0.596424535086818,
+      "grad_norm": 0.06866045296192169,
+      "learning_rate": 0.01,
+      "loss": 2.1606,
+      "step": 5805
+    },
+    {
+      "epoch": 0.596732764820713,
+      "grad_norm": 0.06821485608816147,
+      "learning_rate": 0.01,
+      "loss": 2.1842,
+      "step": 5808
+    },
+    {
+      "epoch": 0.597040994554608,
+      "grad_norm": 0.09566816687583923,
+      "learning_rate": 0.01,
+      "loss": 2.1569,
+      "step": 5811
+    },
+    {
+      "epoch": 0.597349224288503,
+      "grad_norm": 0.1130233108997345,
+      "learning_rate": 0.01,
+      "loss": 2.1649,
+      "step": 5814
+    },
+    {
+      "epoch": 0.597657454022398,
+      "grad_norm": 0.07310149073600769,
+      "learning_rate": 0.01,
+      "loss": 2.1798,
+      "step": 5817
+    },
+    {
+      "epoch": 0.597965683756293,
+      "grad_norm": 0.04523763060569763,
+      "learning_rate": 0.01,
+      "loss": 2.1515,
+      "step": 5820
+    },
+    {
+      "epoch": 0.598273913490188,
+      "grad_norm": 0.05843660235404968,
+      "learning_rate": 0.01,
+      "loss": 2.1403,
+      "step": 5823
+    },
+    {
+      "epoch": 0.598582143224083,
+      "grad_norm": 0.03981595113873482,
+      "learning_rate": 0.01,
+      "loss": 2.1598,
+      "step": 5826
+    },
+    {
+      "epoch": 0.598890372957978,
+      "grad_norm": 0.057108644396066666,
+      "learning_rate": 0.01,
+      "loss": 2.1619,
+      "step": 5829
+    },
+    {
+      "epoch": 0.599198602691873,
+      "grad_norm": 0.12298591434955597,
+      "learning_rate": 0.01,
+      "loss": 2.125,
+      "step": 5832
+    },
+    {
+      "epoch": 0.599506832425768,
+      "grad_norm": 0.06120186299085617,
+      "learning_rate": 0.01,
+      "loss": 2.181,
+      "step": 5835
+    },
+    {
+      "epoch": 0.599815062159663,
+      "grad_norm": 0.05780164897441864,
+      "learning_rate": 0.01,
+      "loss": 2.1555,
+      "step": 5838
+    },
+    {
+      "epoch": 0.600123291893558,
+      "grad_norm": 0.0962534248828888,
+      "learning_rate": 0.01,
+      "loss": 2.1626,
+      "step": 5841
+    },
+    {
+      "epoch": 0.600431521627453,
+      "grad_norm": 0.07417720556259155,
+      "learning_rate": 0.01,
+      "loss": 2.1996,
+      "step": 5844
+    },
+    {
+      "epoch": 0.6007397513613479,
+      "grad_norm": 0.08221522718667984,
+      "learning_rate": 0.01,
+      "loss": 2.1562,
+      "step": 5847
+    },
+    {
+      "epoch": 0.6010479810952429,
+      "grad_norm": 0.11511900275945663,
+      "learning_rate": 0.01,
+      "loss": 2.196,
+      "step": 5850
+    },
+    {
+      "epoch": 0.601356210829138,
+      "grad_norm": 0.1331305354833603,
+      "learning_rate": 0.01,
+      "loss": 2.1649,
+      "step": 5853
+    },
+    {
+      "epoch": 0.601664440563033,
+      "grad_norm": 0.07239941507577896,
+      "learning_rate": 0.01,
+      "loss": 2.1695,
+      "step": 5856
+    },
+    {
+      "epoch": 0.601972670296928,
+      "grad_norm": 0.05865192040801048,
+      "learning_rate": 0.01,
+      "loss": 2.1808,
+      "step": 5859
+    },
+    {
+      "epoch": 0.602280900030823,
+      "grad_norm": 0.047268107533454895,
+      "learning_rate": 0.01,
+      "loss": 2.176,
+      "step": 5862
+    },
+    {
+      "epoch": 0.602589129764718,
+      "grad_norm": 0.046770863234996796,
+      "learning_rate": 0.01,
+      "loss": 2.1689,
+      "step": 5865
+    },
+    {
+      "epoch": 0.602897359498613,
+      "grad_norm": 0.04817832633852959,
+      "learning_rate": 0.01,
+      "loss": 2.1566,
+      "step": 5868
+    },
+    {
+      "epoch": 0.603205589232508,
+      "grad_norm": 0.05692889541387558,
+      "learning_rate": 0.01,
+      "loss": 2.1564,
+      "step": 5871
+    },
+    {
+      "epoch": 0.603513818966403,
+      "grad_norm": 0.056694116443395615,
+      "learning_rate": 0.01,
+      "loss": 2.1591,
+      "step": 5874
+    },
+    {
+      "epoch": 0.603822048700298,
+      "grad_norm": 0.08296339213848114,
+      "learning_rate": 0.01,
+      "loss": 2.1695,
+      "step": 5877
+    },
+    {
+      "epoch": 0.604130278434193,
+      "grad_norm": 0.0934629738330841,
+      "learning_rate": 0.01,
+      "loss": 2.1472,
+      "step": 5880
+    },
+    {
+      "epoch": 0.604438508168088,
+      "grad_norm": 0.10192359238862991,
+      "learning_rate": 0.01,
+      "loss": 2.1441,
+      "step": 5883
+    },
+    {
+      "epoch": 0.604746737901983,
+      "grad_norm": 0.04818946123123169,
+      "learning_rate": 0.01,
+      "loss": 2.1747,
+      "step": 5886
+    },
+    {
+      "epoch": 0.605054967635878,
+      "grad_norm": 0.10131523758172989,
+      "learning_rate": 0.01,
+      "loss": 2.1546,
+      "step": 5889
+    },
+    {
+      "epoch": 0.605363197369773,
+      "grad_norm": 0.07115977257490158,
+      "learning_rate": 0.01,
+      "loss": 2.1597,
+      "step": 5892
+    },
+    {
+      "epoch": 0.605671427103668,
+      "grad_norm": 0.03929082304239273,
+      "learning_rate": 0.01,
+      "loss": 2.171,
+      "step": 5895
+    },
+    {
+      "epoch": 0.6059796568375629,
+      "grad_norm": 0.04109720513224602,
+      "learning_rate": 0.01,
+      "loss": 2.134,
+      "step": 5898
+    },
+    {
+      "epoch": 0.6062878865714579,
+      "grad_norm": 0.05026080831885338,
+      "learning_rate": 0.01,
+      "loss": 2.1491,
+      "step": 5901
+    },
+    {
+      "epoch": 0.6065961163053529,
+      "grad_norm": 0.08281126618385315,
+      "learning_rate": 0.01,
+      "loss": 2.1732,
+      "step": 5904
+    },
+    {
+      "epoch": 0.6069043460392479,
+      "grad_norm": 0.04994012042880058,
+      "learning_rate": 0.01,
+      "loss": 2.1664,
+      "step": 5907
+    },
+    {
+      "epoch": 0.6072125757731429,
+      "grad_norm": 0.06299131363630295,
+      "learning_rate": 0.01,
+      "loss": 2.1669,
+      "step": 5910
+    },
+    {
+      "epoch": 0.6075208055070379,
+      "grad_norm": 0.059428080916404724,
+      "learning_rate": 0.01,
+      "loss": 2.1731,
+      "step": 5913
+    },
+    {
+      "epoch": 0.6078290352409329,
+      "grad_norm": 0.07036252319812775,
+      "learning_rate": 0.01,
+      "loss": 2.1787,
+      "step": 5916
+    },
+    {
+      "epoch": 0.6081372649748279,
+      "grad_norm": 0.04721888527274132,
+      "learning_rate": 0.01,
+      "loss": 2.1531,
+      "step": 5919
+    },
+    {
+      "epoch": 0.6084454947087229,
+      "grad_norm": 0.06953759491443634,
+      "learning_rate": 0.01,
+      "loss": 2.1573,
+      "step": 5922
+    },
+    {
+      "epoch": 0.6087537244426179,
+      "grad_norm": 0.11679168790578842,
+      "learning_rate": 0.01,
+      "loss": 2.155,
+      "step": 5925
+    },
+    {
+      "epoch": 0.6090619541765129,
+      "grad_norm": 0.09196575731039047,
+      "learning_rate": 0.01,
+      "loss": 2.1574,
+      "step": 5928
+    },
+    {
+      "epoch": 0.6093701839104079,
+      "grad_norm": 0.05219469591975212,
+      "learning_rate": 0.01,
+      "loss": 2.1605,
+      "step": 5931
+    },
+    {
+      "epoch": 0.6096784136443029,
+      "grad_norm": 0.09352759271860123,
+      "learning_rate": 0.01,
+      "loss": 2.1456,
+      "step": 5934
+    },
+    {
+      "epoch": 0.6099866433781979,
+      "grad_norm": 0.07393237948417664,
+      "learning_rate": 0.01,
+      "loss": 2.1611,
+      "step": 5937
+    },
+    {
+      "epoch": 0.6102948731120929,
+      "grad_norm": 0.06727741658687592,
+      "learning_rate": 0.01,
+      "loss": 2.1599,
+      "step": 5940
+    },
+    {
+      "epoch": 0.6106031028459878,
+      "grad_norm": 0.09024669975042343,
+      "learning_rate": 0.01,
+      "loss": 2.1621,
+      "step": 5943
+    },
+    {
+      "epoch": 0.6109113325798828,
+      "grad_norm": 0.04514656960964203,
+      "learning_rate": 0.01,
+      "loss": 2.1809,
+      "step": 5946
+    },
+    {
+      "epoch": 0.6112195623137778,
+      "grad_norm": 0.04011565446853638,
+      "learning_rate": 0.01,
+      "loss": 2.1715,
+      "step": 5949
+    },
+    {
+      "epoch": 0.6115277920476728,
+      "grad_norm": 0.04640655592083931,
+      "learning_rate": 0.01,
+      "loss": 2.15,
+      "step": 5952
+    },
+    {
+      "epoch": 0.6118360217815678,
+      "grad_norm": 0.0471080057322979,
+      "learning_rate": 0.01,
+      "loss": 2.1805,
+      "step": 5955
+    },
+    {
+      "epoch": 0.6121442515154628,
+      "grad_norm": 0.17398513853549957,
+      "learning_rate": 0.01,
+      "loss": 2.1497,
+      "step": 5958
+    },
+    {
+      "epoch": 0.6124524812493578,
+      "grad_norm": 0.06299551576375961,
+      "learning_rate": 0.01,
+      "loss": 2.1387,
+      "step": 5961
+    },
+    {
+      "epoch": 0.6127607109832528,
+      "grad_norm": 0.07517322897911072,
+      "learning_rate": 0.01,
+      "loss": 2.1348,
+      "step": 5964
+    },
+    {
+      "epoch": 0.6130689407171478,
+      "grad_norm": 0.050515878945589066,
+      "learning_rate": 0.01,
+      "loss": 2.1725,
+      "step": 5967
+    },
+    {
+      "epoch": 0.6133771704510429,
+      "grad_norm": 0.04682675376534462,
+      "learning_rate": 0.01,
+      "loss": 2.1759,
+      "step": 5970
+    },
+    {
+      "epoch": 0.6136854001849379,
+      "grad_norm": 0.05297816917300224,
+      "learning_rate": 0.01,
+      "loss": 2.1491,
+      "step": 5973
+    },
+    {
+      "epoch": 0.6139936299188329,
+      "grad_norm": 0.07467235624790192,
+      "learning_rate": 0.01,
+      "loss": 2.1556,
+      "step": 5976
+    },
+    {
+      "epoch": 0.6143018596527279,
+      "grad_norm": 0.06621374934911728,
+      "learning_rate": 0.01,
+      "loss": 2.1498,
+      "step": 5979
+    },
+    {
+      "epoch": 0.6146100893866229,
+      "grad_norm": 0.0538405105471611,
+      "learning_rate": 0.01,
+      "loss": 2.1694,
+      "step": 5982
+    },
+    {
+      "epoch": 0.6149183191205179,
+      "grad_norm": 0.09891212731599808,
+      "learning_rate": 0.01,
+      "loss": 2.1598,
+      "step": 5985
+    },
+    {
+      "epoch": 0.6152265488544129,
+      "grad_norm": 0.042064208537340164,
+      "learning_rate": 0.01,
+      "loss": 2.1375,
+      "step": 5988
+    },
+    {
+      "epoch": 0.6155347785883079,
+      "grad_norm": 0.06750064343214035,
+      "learning_rate": 0.01,
+      "loss": 2.1371,
+      "step": 5991
+    },
+    {
+      "epoch": 0.6158430083222028,
+      "grad_norm": 0.0626809298992157,
+      "learning_rate": 0.01,
+      "loss": 2.1455,
+      "step": 5994
+    },
+    {
+      "epoch": 0.6161512380560978,
+      "grad_norm": 0.04291335120797157,
+      "learning_rate": 0.01,
+      "loss": 2.1397,
+      "step": 5997
+    },
+    {
+      "epoch": 0.6164594677899928,
+      "grad_norm": 0.05945251137018204,
+      "learning_rate": 0.01,
+      "loss": 2.1346,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6167676975238878,
+      "grad_norm": 0.15699933469295502,
+      "learning_rate": 0.01,
+      "loss": 2.1384,
+      "step": 6003
+    },
+    {
+      "epoch": 0.6170759272577828,
+      "grad_norm": 0.06863987445831299,
+      "learning_rate": 0.01,
+      "loss": 2.1401,
+      "step": 6006
+    },
+    {
+      "epoch": 0.6173841569916778,
+      "grad_norm": 0.04850529506802559,
+      "learning_rate": 0.01,
+      "loss": 2.1637,
+      "step": 6009
+    },
+    {
+      "epoch": 0.6176923867255728,
+      "grad_norm": 0.05660491809248924,
+      "learning_rate": 0.01,
+      "loss": 2.1721,
+      "step": 6012
+    },
+    {
+      "epoch": 0.6180006164594678,
+      "grad_norm": 0.050568364560604095,
+      "learning_rate": 0.01,
+      "loss": 2.1676,
+      "step": 6015
+    },
+    {
+      "epoch": 0.6183088461933628,
+      "grad_norm": 0.060765717178583145,
+      "learning_rate": 0.01,
+      "loss": 2.127,
+      "step": 6018
+    },
+    {
+      "epoch": 0.6186170759272578,
+      "grad_norm": 0.0731448233127594,
+      "learning_rate": 0.01,
+      "loss": 2.1531,
+      "step": 6021
+    },
+    {
+      "epoch": 0.6189253056611528,
+      "grad_norm": 0.055431608110666275,
+      "learning_rate": 0.01,
+      "loss": 2.1662,
+      "step": 6024
+    },
+    {
+      "epoch": 0.6192335353950478,
+      "grad_norm": 0.05376220867037773,
+      "learning_rate": 0.01,
+      "loss": 2.1465,
+      "step": 6027
+    },
+    {
+      "epoch": 0.6195417651289428,
+      "grad_norm": 0.09729186445474625,
+      "learning_rate": 0.01,
+      "loss": 2.161,
+      "step": 6030
+    },
+    {
+      "epoch": 0.6198499948628378,
+      "grad_norm": 0.08046093583106995,
+      "learning_rate": 0.01,
+      "loss": 2.1435,
+      "step": 6033
+    },
+    {
+      "epoch": 0.6201582245967328,
+      "grad_norm": 0.09514495730400085,
+      "learning_rate": 0.01,
+      "loss": 2.1511,
+      "step": 6036
+    },
+    {
+      "epoch": 0.6204664543306277,
+      "grad_norm": 0.056993287056684494,
+      "learning_rate": 0.01,
+      "loss": 2.1439,
+      "step": 6039
+    },
+    {
+      "epoch": 0.6207746840645227,
+      "grad_norm": 0.06429582834243774,
+      "learning_rate": 0.01,
+      "loss": 2.1393,
+      "step": 6042
+    },
+    {
+      "epoch": 0.6210829137984177,
+      "grad_norm": 0.1299380660057068,
+      "learning_rate": 0.01,
+      "loss": 2.1831,
+      "step": 6045
+    },
+    {
+      "epoch": 0.6213911435323127,
+      "grad_norm": 0.13815906643867493,
+      "learning_rate": 0.01,
+      "loss": 2.1645,
+      "step": 6048
+    },
+    {
+      "epoch": 0.6216993732662077,
+      "grad_norm": 0.056314874440431595,
+      "learning_rate": 0.01,
+      "loss": 2.1417,
+      "step": 6051
+    },
+    {
+      "epoch": 0.6220076030001027,
+      "grad_norm": 0.06146218627691269,
+      "learning_rate": 0.01,
+      "loss": 2.1418,
+      "step": 6054
+    },
+    {
+      "epoch": 0.6223158327339977,
+      "grad_norm": 0.062167149037122726,
+      "learning_rate": 0.01,
+      "loss": 2.1778,
+      "step": 6057
+    },
+    {
+      "epoch": 0.6226240624678927,
+      "grad_norm": 0.059581879526376724,
+      "learning_rate": 0.01,
+      "loss": 2.1725,
+      "step": 6060
+    },
+    {
+      "epoch": 0.6229322922017877,
+      "grad_norm": 0.044389910995960236,
+      "learning_rate": 0.01,
+      "loss": 2.1553,
+      "step": 6063
+    },
+    {
+      "epoch": 0.6232405219356827,
+      "grad_norm": 0.036525238305330276,
+      "learning_rate": 0.01,
+      "loss": 2.1545,
+      "step": 6066
+    },
+    {
+      "epoch": 0.6235487516695777,
+      "grad_norm": 0.0995573177933693,
+      "learning_rate": 0.01,
+      "loss": 2.1566,
+      "step": 6069
+    },
+    {
+      "epoch": 0.6238569814034727,
+      "grad_norm": 0.10412520170211792,
+      "learning_rate": 0.01,
+      "loss": 2.1525,
+      "step": 6072
+    },
+    {
+      "epoch": 0.6241652111373677,
+      "grad_norm": 0.10417335480451584,
+      "learning_rate": 0.01,
+      "loss": 2.1535,
+      "step": 6075
+    },
+    {
+      "epoch": 0.6244734408712627,
+      "grad_norm": 0.09024351090192795,
+      "learning_rate": 0.01,
+      "loss": 2.1551,
+      "step": 6078
+    },
+    {
+      "epoch": 0.6247816706051577,
+      "grad_norm": 0.04889573156833649,
+      "learning_rate": 0.01,
+      "loss": 2.1549,
+      "step": 6081
+    },
+    {
+      "epoch": 0.6250899003390528,
+      "grad_norm": 0.05154373124241829,
+      "learning_rate": 0.01,
+      "loss": 2.1461,
+      "step": 6084
+    },
+    {
+      "epoch": 0.6253981300729478,
+      "grad_norm": 0.04337237402796745,
+      "learning_rate": 0.01,
+      "loss": 2.1733,
+      "step": 6087
+    },
+    {
+      "epoch": 0.6257063598068427,
+      "grad_norm": 0.06173473224043846,
+      "learning_rate": 0.01,
+      "loss": 2.1657,
+      "step": 6090
+    },
+    {
+      "epoch": 0.6260145895407377,
+      "grad_norm": 0.06174352392554283,
+      "learning_rate": 0.01,
+      "loss": 2.1528,
+      "step": 6093
+    },
+    {
+      "epoch": 0.6263228192746327,
+      "grad_norm": 0.07301110029220581,
+      "learning_rate": 0.01,
+      "loss": 2.1489,
+      "step": 6096
+    },
+    {
+      "epoch": 0.6266310490085277,
+      "grad_norm": 0.04265190288424492,
+      "learning_rate": 0.01,
+      "loss": 2.1624,
+      "step": 6099
+    },
+    {
+      "epoch": 0.6269392787424227,
+      "grad_norm": 0.056723251938819885,
+      "learning_rate": 0.01,
+      "loss": 2.1624,
+      "step": 6102
+    },
+    {
+      "epoch": 0.6272475084763177,
+      "grad_norm": 0.06809309124946594,
+      "learning_rate": 0.01,
+      "loss": 2.1525,
+      "step": 6105
+    },
+    {
+      "epoch": 0.6275557382102127,
+      "grad_norm": 0.06820474565029144,
+      "learning_rate": 0.01,
+      "loss": 2.1472,
+      "step": 6108
+    },
+    {
+      "epoch": 0.6278639679441077,
+      "grad_norm": 0.05961904302239418,
+      "learning_rate": 0.01,
+      "loss": 2.1561,
+      "step": 6111
+    },
+    {
+      "epoch": 0.6281721976780027,
+      "grad_norm": 0.04617665335536003,
+      "learning_rate": 0.01,
+      "loss": 2.1475,
+      "step": 6114
+    },
+    {
+      "epoch": 0.6284804274118977,
+      "grad_norm": 0.040670618414878845,
+      "learning_rate": 0.01,
+      "loss": 2.153,
+      "step": 6117
+    },
+    {
+      "epoch": 0.6287886571457927,
+      "grad_norm": 0.09909021109342575,
+      "learning_rate": 0.01,
+      "loss": 2.141,
+      "step": 6120
+    },
+    {
+      "epoch": 0.6290968868796877,
+      "grad_norm": 0.04966261237859726,
+      "learning_rate": 0.01,
+      "loss": 2.1264,
+      "step": 6123
+    },
+    {
+      "epoch": 0.6294051166135827,
+      "grad_norm": 0.0570046491920948,
+      "learning_rate": 0.01,
+      "loss": 2.1572,
+      "step": 6126
+    },
+    {
+      "epoch": 0.6297133463474777,
+      "grad_norm": 0.10374405980110168,
+      "learning_rate": 0.01,
+      "loss": 2.149,
+      "step": 6129
+    },
+    {
+      "epoch": 0.6300215760813727,
+      "grad_norm": 0.061325542628765106,
+      "learning_rate": 0.01,
+      "loss": 2.1521,
+      "step": 6132
+    },
+    {
+      "epoch": 0.6303298058152677,
+      "grad_norm": 0.16151310503482819,
+      "learning_rate": 0.01,
+      "loss": 2.1825,
+      "step": 6135
+    },
+    {
+      "epoch": 0.6306380355491626,
+      "grad_norm": 0.0921199768781662,
+      "learning_rate": 0.01,
+      "loss": 2.1773,
+      "step": 6138
+    },
+    {
+      "epoch": 0.6309462652830576,
+      "grad_norm": 0.05603238567709923,
+      "learning_rate": 0.01,
+      "loss": 2.1452,
+      "step": 6141
+    },
+    {
+      "epoch": 0.6312544950169526,
+      "grad_norm": 0.12173126637935638,
+      "learning_rate": 0.01,
+      "loss": 2.1713,
+      "step": 6144
+    },
+    {
+      "epoch": 0.6315627247508476,
+      "grad_norm": 0.04609265923500061,
+      "learning_rate": 0.01,
+      "loss": 2.1518,
+      "step": 6147
+    },
+    {
+      "epoch": 0.6318709544847426,
+      "grad_norm": 0.06445127725601196,
+      "learning_rate": 0.01,
+      "loss": 2.162,
+      "step": 6150
+    },
+    {
+      "epoch": 0.6321791842186376,
+      "grad_norm": 0.05396106466650963,
+      "learning_rate": 0.01,
+      "loss": 2.1248,
+      "step": 6153
+    },
+    {
+      "epoch": 0.6324874139525326,
+      "grad_norm": 0.06955734640359879,
+      "learning_rate": 0.01,
+      "loss": 2.1497,
+      "step": 6156
+    },
+    {
+      "epoch": 0.6327956436864276,
+      "grad_norm": 0.04371445253491402,
+      "learning_rate": 0.01,
+      "loss": 2.1167,
+      "step": 6159
+    },
+    {
+      "epoch": 0.6331038734203226,
+      "grad_norm": 0.07146921008825302,
+      "learning_rate": 0.01,
+      "loss": 2.1633,
+      "step": 6162
+    },
+    {
+      "epoch": 0.6334121031542176,
+      "grad_norm": 0.08056561648845673,
+      "learning_rate": 0.01,
+      "loss": 2.1506,
+      "step": 6165
+    },
+    {
+      "epoch": 0.6337203328881126,
+      "grad_norm": 0.08875605463981628,
+      "learning_rate": 0.01,
+      "loss": 2.1834,
+      "step": 6168
+    },
+    {
+      "epoch": 0.6340285626220076,
+      "grad_norm": 0.05090434104204178,
+      "learning_rate": 0.01,
+      "loss": 2.1514,
+      "step": 6171
+    },
+    {
+      "epoch": 0.6343367923559026,
+      "grad_norm": 0.11710961163043976,
+      "learning_rate": 0.01,
+      "loss": 2.1589,
+      "step": 6174
+    },
+    {
+      "epoch": 0.6346450220897976,
+      "grad_norm": 0.04704523831605911,
+      "learning_rate": 0.01,
+      "loss": 2.1469,
+      "step": 6177
+    },
+    {
+      "epoch": 0.6349532518236926,
+      "grad_norm": 0.045143596827983856,
+      "learning_rate": 0.01,
+      "loss": 2.1311,
+      "step": 6180
+    },
+    {
+      "epoch": 0.6352614815575875,
+      "grad_norm": 0.04246919974684715,
+      "learning_rate": 0.01,
+      "loss": 2.1481,
+      "step": 6183
+    },
+    {
+      "epoch": 0.6355697112914825,
+      "grad_norm": 0.04303867742419243,
+      "learning_rate": 0.01,
+      "loss": 2.1557,
+      "step": 6186
+    },
+    {
+      "epoch": 0.6358779410253775,
+      "grad_norm": 0.17376503348350525,
+      "learning_rate": 0.01,
+      "loss": 2.1616,
+      "step": 6189
+    },
+    {
+      "epoch": 0.6361861707592725,
+      "grad_norm": 0.11983154714107513,
+      "learning_rate": 0.01,
+      "loss": 2.1569,
+      "step": 6192
+    },
+    {
+      "epoch": 0.6364944004931675,
+      "grad_norm": 0.0443497858941555,
+      "learning_rate": 0.01,
+      "loss": 2.1454,
+      "step": 6195
+    },
+    {
+      "epoch": 0.6368026302270625,
+      "grad_norm": 0.04241250827908516,
+      "learning_rate": 0.01,
+      "loss": 2.1409,
+      "step": 6198
+    },
+    {
+      "epoch": 0.6371108599609576,
+      "grad_norm": 0.07058902829885483,
+      "learning_rate": 0.01,
+      "loss": 2.1246,
+      "step": 6201
+    },
+    {
+      "epoch": 0.6374190896948526,
+      "grad_norm": 0.060852985829114914,
+      "learning_rate": 0.01,
+      "loss": 2.1512,
+      "step": 6204
+    },
+    {
+      "epoch": 0.6377273194287476,
+      "grad_norm": 0.058703117072582245,
+      "learning_rate": 0.01,
+      "loss": 2.1114,
+      "step": 6207
+    },
+    {
+      "epoch": 0.6380355491626426,
+      "grad_norm": 0.08501632511615753,
+      "learning_rate": 0.01,
+      "loss": 2.1818,
+      "step": 6210
+    },
+    {
+      "epoch": 0.6383437788965376,
+      "grad_norm": 0.07715412974357605,
+      "learning_rate": 0.01,
+      "loss": 2.1661,
+      "step": 6213
+    },
+    {
+      "epoch": 0.6386520086304326,
+      "grad_norm": 0.06822165101766586,
+      "learning_rate": 0.01,
+      "loss": 2.1652,
+      "step": 6216
+    },
+    {
+      "epoch": 0.6389602383643276,
+      "grad_norm": 0.048459213227033615,
+      "learning_rate": 0.01,
+      "loss": 2.1311,
+      "step": 6219
+    },
+    {
+      "epoch": 0.6392684680982226,
+      "grad_norm": 0.08208850026130676,
+      "learning_rate": 0.01,
+      "loss": 2.1316,
+      "step": 6222
+    },
+    {
+      "epoch": 0.6395766978321176,
+      "grad_norm": 0.06399821490049362,
+      "learning_rate": 0.01,
+      "loss": 2.1354,
+      "step": 6225
+    },
+    {
+      "epoch": 0.6398849275660126,
+      "grad_norm": 0.12036826461553574,
+      "learning_rate": 0.01,
+      "loss": 2.1509,
+      "step": 6228
+    },
+    {
+      "epoch": 0.6401931572999076,
+      "grad_norm": 0.08180755376815796,
+      "learning_rate": 0.01,
+      "loss": 2.1571,
+      "step": 6231
+    },
+    {
+      "epoch": 0.6405013870338025,
+      "grad_norm": 0.053771521896123886,
+      "learning_rate": 0.01,
+      "loss": 2.1485,
+      "step": 6234
+    },
+    {
+      "epoch": 0.6408096167676975,
+      "grad_norm": 0.042291607707738876,
+      "learning_rate": 0.01,
+      "loss": 2.1606,
+      "step": 6237
+    },
+    {
+      "epoch": 0.6411178465015925,
+      "grad_norm": 0.044655315577983856,
+      "learning_rate": 0.01,
+      "loss": 2.1592,
+      "step": 6240
+    },
+    {
+      "epoch": 0.6414260762354875,
+      "grad_norm": 0.07763859629631042,
+      "learning_rate": 0.01,
+      "loss": 2.1543,
+      "step": 6243
+    },
+    {
+      "epoch": 0.6417343059693825,
+      "grad_norm": 0.055368274450302124,
+      "learning_rate": 0.01,
+      "loss": 2.1643,
+      "step": 6246
+    },
+    {
+      "epoch": 0.6420425357032775,
+      "grad_norm": 0.047774944454431534,
+      "learning_rate": 0.01,
+      "loss": 2.1542,
+      "step": 6249
+    },
+    {
+      "epoch": 0.6423507654371725,
+      "grad_norm": 0.06478223204612732,
+      "learning_rate": 0.01,
+      "loss": 2.1501,
+      "step": 6252
+    },
+    {
+      "epoch": 0.6426589951710675,
+      "grad_norm": 0.03782160207629204,
+      "learning_rate": 0.01,
+      "loss": 2.1455,
+      "step": 6255
+    },
+    {
+      "epoch": 0.6429672249049625,
+      "grad_norm": 0.11297930777072906,
+      "learning_rate": 0.01,
+      "loss": 2.1667,
+      "step": 6258
+    },
+    {
+      "epoch": 0.6432754546388575,
+      "grad_norm": 0.09408997744321823,
+      "learning_rate": 0.01,
+      "loss": 2.146,
+      "step": 6261
+    },
+    {
+      "epoch": 0.6435836843727525,
+      "grad_norm": 0.06677352637052536,
+      "learning_rate": 0.01,
+      "loss": 2.1723,
+      "step": 6264
+    },
+    {
+      "epoch": 0.6438919141066475,
+      "grad_norm": 0.08687873184680939,
+      "learning_rate": 0.01,
+      "loss": 2.1517,
+      "step": 6267
+    },
+    {
+      "epoch": 0.6442001438405425,
+      "grad_norm": 0.06850516051054001,
+      "learning_rate": 0.01,
+      "loss": 2.148,
+      "step": 6270
+    },
+    {
+      "epoch": 0.6445083735744375,
+      "grad_norm": 0.07705084979534149,
+      "learning_rate": 0.01,
+      "loss": 2.1567,
+      "step": 6273
+    },
+    {
+      "epoch": 0.6448166033083325,
+      "grad_norm": 0.1622423529624939,
+      "learning_rate": 0.01,
+      "loss": 2.1676,
+      "step": 6276
+    },
+    {
+      "epoch": 0.6451248330422275,
+      "grad_norm": 0.11197759211063385,
+      "learning_rate": 0.01,
+      "loss": 2.1376,
+      "step": 6279
+    },
+    {
+      "epoch": 0.6454330627761224,
+      "grad_norm": 0.06562814861536026,
+      "learning_rate": 0.01,
+      "loss": 2.1652,
+      "step": 6282
+    },
+    {
+      "epoch": 0.6457412925100174,
+      "grad_norm": 0.0867902860045433,
+      "learning_rate": 0.01,
+      "loss": 2.1733,
+      "step": 6285
+    },
+    {
+      "epoch": 0.6460495222439124,
+      "grad_norm": 0.08153738081455231,
+      "learning_rate": 0.01,
+      "loss": 2.1442,
+      "step": 6288
+    },
+    {
+      "epoch": 0.6463577519778074,
+      "grad_norm": 0.09800709784030914,
+      "learning_rate": 0.01,
+      "loss": 2.1262,
+      "step": 6291
+    },
+    {
+      "epoch": 0.6466659817117024,
+      "grad_norm": 0.07728230953216553,
+      "learning_rate": 0.01,
+      "loss": 2.139,
+      "step": 6294
+    },
+    {
+      "epoch": 0.6469742114455974,
+      "grad_norm": 0.09658671170473099,
+      "learning_rate": 0.01,
+      "loss": 2.1421,
+      "step": 6297
+    },
+    {
+      "epoch": 0.6472824411794924,
+      "grad_norm": 0.0448787659406662,
+      "learning_rate": 0.01,
+      "loss": 2.1415,
+      "step": 6300
+    },
+    {
+      "epoch": 0.6475906709133874,
+      "grad_norm": 0.03848707675933838,
+      "learning_rate": 0.01,
+      "loss": 2.1209,
+      "step": 6303
+    },
+    {
+      "epoch": 0.6478989006472824,
+      "grad_norm": 0.07465004920959473,
+      "learning_rate": 0.01,
+      "loss": 2.1395,
+      "step": 6306
+    },
+    {
+      "epoch": 0.6482071303811774,
+      "grad_norm": 0.060424912720918655,
+      "learning_rate": 0.01,
+      "loss": 2.1806,
+      "step": 6309
+    },
+    {
+      "epoch": 0.6485153601150724,
+      "grad_norm": 0.05204974114894867,
+      "learning_rate": 0.01,
+      "loss": 2.1287,
+      "step": 6312
+    },
+    {
+      "epoch": 0.6488235898489675,
+      "grad_norm": 0.06045055389404297,
+      "learning_rate": 0.01,
+      "loss": 2.1727,
+      "step": 6315
+    },
+    {
+      "epoch": 0.6491318195828625,
+      "grad_norm": 0.04978582262992859,
+      "learning_rate": 0.01,
+      "loss": 2.1264,
+      "step": 6318
+    },
+    {
+      "epoch": 0.6494400493167575,
+      "grad_norm": 0.08131048828363419,
+      "learning_rate": 0.01,
+      "loss": 2.137,
+      "step": 6321
+    },
+    {
+      "epoch": 0.6497482790506525,
+      "grad_norm": 0.09749994426965714,
+      "learning_rate": 0.01,
+      "loss": 2.1557,
+      "step": 6324
+    },
+    {
+      "epoch": 0.6500565087845475,
+      "grad_norm": 0.06079535186290741,
+      "learning_rate": 0.01,
+      "loss": 2.1432,
+      "step": 6327
+    },
+    {
+      "epoch": 0.6503647385184425,
+      "grad_norm": 0.08241060376167297,
+      "learning_rate": 0.01,
+      "loss": 2.1551,
+      "step": 6330
+    },
+    {
+      "epoch": 0.6506729682523374,
+      "grad_norm": 0.12339378148317337,
+      "learning_rate": 0.01,
+      "loss": 2.1216,
+      "step": 6333
+    },
+    {
+      "epoch": 0.6509811979862324,
+      "grad_norm": 0.0660511702299118,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 6336
+    },
+    {
+      "epoch": 0.6512894277201274,
+      "grad_norm": 0.06279938668012619,
+      "learning_rate": 0.01,
+      "loss": 2.1778,
+      "step": 6339
+    },
+    {
+      "epoch": 0.6515976574540224,
+      "grad_norm": 0.068712018430233,
+      "learning_rate": 0.01,
+      "loss": 2.1348,
+      "step": 6342
+    },
+    {
+      "epoch": 0.6519058871879174,
+      "grad_norm": 0.05808734893798828,
+      "learning_rate": 0.01,
+      "loss": 2.135,
+      "step": 6345
+    },
+    {
+      "epoch": 0.6522141169218124,
+      "grad_norm": 0.044942643493413925,
+      "learning_rate": 0.01,
+      "loss": 2.1613,
+      "step": 6348
+    },
+    {
+      "epoch": 0.6525223466557074,
+      "grad_norm": 0.11666214466094971,
+      "learning_rate": 0.01,
+      "loss": 2.1399,
+      "step": 6351
+    },
+    {
+      "epoch": 0.6528305763896024,
+      "grad_norm": 0.06776747852563858,
+      "learning_rate": 0.01,
+      "loss": 2.1369,
+      "step": 6354
+    },
+    {
+      "epoch": 0.6531388061234974,
+      "grad_norm": 0.10171874612569809,
+      "learning_rate": 0.01,
+      "loss": 2.1273,
+      "step": 6357
+    },
+    {
+      "epoch": 0.6534470358573924,
+      "grad_norm": 0.04611232131719589,
+      "learning_rate": 0.01,
+      "loss": 2.1482,
+      "step": 6360
+    },
+    {
+      "epoch": 0.6537552655912874,
+      "grad_norm": 0.042139992117881775,
+      "learning_rate": 0.01,
+      "loss": 2.1235,
+      "step": 6363
+    },
+    {
+      "epoch": 0.6540634953251824,
+      "grad_norm": 0.057816632091999054,
+      "learning_rate": 0.01,
+      "loss": 2.1449,
+      "step": 6366
+    },
+    {
+      "epoch": 0.6543717250590774,
+      "grad_norm": 0.11400949209928513,
+      "learning_rate": 0.01,
+      "loss": 2.1566,
+      "step": 6369
+    },
+    {
+      "epoch": 0.6546799547929724,
+      "grad_norm": 0.07320736348628998,
+      "learning_rate": 0.01,
+      "loss": 2.1682,
+      "step": 6372
+    },
+    {
+      "epoch": 0.6549881845268674,
+      "grad_norm": 0.07262291014194489,
+      "learning_rate": 0.01,
+      "loss": 2.1514,
+      "step": 6375
+    },
+    {
+      "epoch": 0.6552964142607623,
+      "grad_norm": 0.05559679865837097,
+      "learning_rate": 0.01,
+      "loss": 2.1347,
+      "step": 6378
+    },
+    {
+      "epoch": 0.6556046439946573,
+      "grad_norm": 0.049424149096012115,
+      "learning_rate": 0.01,
+      "loss": 2.1423,
+      "step": 6381
+    },
+    {
+      "epoch": 0.6559128737285523,
+      "grad_norm": 0.05457301065325737,
+      "learning_rate": 0.01,
+      "loss": 2.1425,
+      "step": 6384
+    },
+    {
+      "epoch": 0.6562211034624473,
+      "grad_norm": 0.058564141392707825,
+      "learning_rate": 0.01,
+      "loss": 2.1158,
+      "step": 6387
+    },
+    {
+      "epoch": 0.6565293331963423,
+      "grad_norm": 0.10944786667823792,
+      "learning_rate": 0.01,
+      "loss": 2.146,
+      "step": 6390
+    },
+    {
+      "epoch": 0.6568375629302373,
+      "grad_norm": 0.07760695368051529,
+      "learning_rate": 0.01,
+      "loss": 2.176,
+      "step": 6393
+    },
+    {
+      "epoch": 0.6571457926641323,
+      "grad_norm": 0.07621042430400848,
+      "learning_rate": 0.01,
+      "loss": 2.1779,
+      "step": 6396
+    },
+    {
+      "epoch": 0.6574540223980273,
+      "grad_norm": 0.09723789244890213,
+      "learning_rate": 0.01,
+      "loss": 2.1455,
+      "step": 6399
+    },
+    {
+      "epoch": 0.6577622521319223,
+      "grad_norm": 0.05648832768201828,
+      "learning_rate": 0.01,
+      "loss": 2.154,
+      "step": 6402
+    },
+    {
+      "epoch": 0.6580704818658173,
+      "grad_norm": 0.04370080679655075,
+      "learning_rate": 0.01,
+      "loss": 2.1374,
+      "step": 6405
+    },
+    {
+      "epoch": 0.6583787115997123,
+      "grad_norm": 0.03729141131043434,
+      "learning_rate": 0.01,
+      "loss": 2.1275,
+      "step": 6408
+    },
+    {
+      "epoch": 0.6586869413336073,
+      "grad_norm": 0.055584125220775604,
+      "learning_rate": 0.01,
+      "loss": 2.1442,
+      "step": 6411
+    },
+    {
+      "epoch": 0.6589951710675023,
+      "grad_norm": 0.07981918007135391,
+      "learning_rate": 0.01,
+      "loss": 2.1618,
+      "step": 6414
+    },
+    {
+      "epoch": 0.6593034008013973,
+      "grad_norm": 0.09241674095392227,
+      "learning_rate": 0.01,
+      "loss": 2.1519,
+      "step": 6417
+    },
+    {
+      "epoch": 0.6596116305352923,
+      "grad_norm": 0.10454630106687546,
+      "learning_rate": 0.01,
+      "loss": 2.1309,
+      "step": 6420
+    },
+    {
+      "epoch": 0.6599198602691873,
+      "grad_norm": 0.08674053847789764,
+      "learning_rate": 0.01,
+      "loss": 2.1617,
+      "step": 6423
+    },
+    {
+      "epoch": 0.6602280900030822,
+      "grad_norm": 0.06003529578447342,
+      "learning_rate": 0.01,
+      "loss": 2.1475,
+      "step": 6426
+    },
+    {
+      "epoch": 0.6605363197369772,
+      "grad_norm": 0.07370956987142563,
+      "learning_rate": 0.01,
+      "loss": 2.1466,
+      "step": 6429
+    },
+    {
+      "epoch": 0.6608445494708723,
+      "grad_norm": 0.05090004578232765,
+      "learning_rate": 0.01,
+      "loss": 2.1506,
+      "step": 6432
+    },
+    {
+      "epoch": 0.6611527792047673,
+      "grad_norm": 0.06062362715601921,
+      "learning_rate": 0.01,
+      "loss": 2.1601,
+      "step": 6435
+    },
+    {
+      "epoch": 0.6614610089386623,
+      "grad_norm": 0.05484107881784439,
+      "learning_rate": 0.01,
+      "loss": 2.1452,
+      "step": 6438
+    },
+    {
+      "epoch": 0.6617692386725573,
+      "grad_norm": 0.1367156058549881,
+      "learning_rate": 0.01,
+      "loss": 2.1586,
+      "step": 6441
+    },
+    {
+      "epoch": 0.6620774684064523,
+      "grad_norm": 0.05140338093042374,
+      "learning_rate": 0.01,
+      "loss": 2.1463,
+      "step": 6444
+    },
+    {
+      "epoch": 0.6623856981403473,
+      "grad_norm": 0.09168683737516403,
+      "learning_rate": 0.01,
+      "loss": 2.1467,
+      "step": 6447
+    },
+    {
+      "epoch": 0.6626939278742423,
+      "grad_norm": 0.04098822921514511,
+      "learning_rate": 0.01,
+      "loss": 2.1648,
+      "step": 6450
+    },
+    {
+      "epoch": 0.6630021576081373,
+      "grad_norm": 0.049763478338718414,
+      "learning_rate": 0.01,
+      "loss": 2.1289,
+      "step": 6453
+    },
+    {
+      "epoch": 0.6633103873420323,
+      "grad_norm": 0.060069404542446136,
+      "learning_rate": 0.01,
+      "loss": 2.1467,
+      "step": 6456
+    },
+    {
+      "epoch": 0.6636186170759273,
+      "grad_norm": 0.06611450761556625,
+      "learning_rate": 0.01,
+      "loss": 2.1599,
+      "step": 6459
+    },
+    {
+      "epoch": 0.6639268468098223,
+      "grad_norm": 0.04955270141363144,
+      "learning_rate": 0.01,
+      "loss": 2.136,
+      "step": 6462
+    },
+    {
+      "epoch": 0.6642350765437173,
+      "grad_norm": 0.04004522040486336,
+      "learning_rate": 0.01,
+      "loss": 2.1457,
+      "step": 6465
+    },
+    {
+      "epoch": 0.6645433062776123,
+      "grad_norm": 0.06539756804704666,
+      "learning_rate": 0.01,
+      "loss": 2.1458,
+      "step": 6468
+    },
+    {
+      "epoch": 0.6648515360115073,
+      "grad_norm": 0.10684728622436523,
+      "learning_rate": 0.01,
+      "loss": 2.1279,
+      "step": 6471
+    },
+    {
+      "epoch": 0.6651597657454023,
+      "grad_norm": 0.09936464577913284,
+      "learning_rate": 0.01,
+      "loss": 2.1767,
+      "step": 6474
+    },
+    {
+      "epoch": 0.6654679954792972,
+      "grad_norm": 0.04908827692270279,
+      "learning_rate": 0.01,
+      "loss": 2.1259,
+      "step": 6477
+    },
+    {
+      "epoch": 0.6657762252131922,
+      "grad_norm": 0.048053622245788574,
+      "learning_rate": 0.01,
+      "loss": 2.1718,
+      "step": 6480
+    },
+    {
+      "epoch": 0.6660844549470872,
+      "grad_norm": 0.05524458363652229,
+      "learning_rate": 0.01,
+      "loss": 2.1673,
+      "step": 6483
+    },
+    {
+      "epoch": 0.6663926846809822,
+      "grad_norm": 0.05107030272483826,
+      "learning_rate": 0.01,
+      "loss": 2.13,
+      "step": 6486
+    },
+    {
+      "epoch": 0.6667009144148772,
+      "grad_norm": 0.12472579628229141,
+      "learning_rate": 0.01,
+      "loss": 2.149,
+      "step": 6489
+    },
+    {
+      "epoch": 0.6670091441487722,
+      "grad_norm": 0.05257454514503479,
+      "learning_rate": 0.01,
+      "loss": 2.1343,
+      "step": 6492
+    },
+    {
+      "epoch": 0.6673173738826672,
+      "grad_norm": 0.05986837297677994,
+      "learning_rate": 0.01,
+      "loss": 2.1265,
+      "step": 6495
+    },
+    {
+      "epoch": 0.6676256036165622,
+      "grad_norm": 0.08322940021753311,
+      "learning_rate": 0.01,
+      "loss": 2.1317,
+      "step": 6498
+    },
+    {
+      "epoch": 0.6679338333504572,
+      "grad_norm": 0.0466473363339901,
+      "learning_rate": 0.01,
+      "loss": 2.1235,
+      "step": 6501
+    },
+    {
+      "epoch": 0.6682420630843522,
+      "grad_norm": 0.05092160776257515,
+      "learning_rate": 0.01,
+      "loss": 2.1672,
+      "step": 6504
+    },
+    {
+      "epoch": 0.6685502928182472,
+      "grad_norm": 0.08392294496297836,
+      "learning_rate": 0.01,
+      "loss": 2.1473,
+      "step": 6507
+    },
+    {
+      "epoch": 0.6688585225521422,
+      "grad_norm": 0.042165517807006836,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 6510
+    },
+    {
+      "epoch": 0.6691667522860372,
+      "grad_norm": 0.06214481219649315,
+      "learning_rate": 0.01,
+      "loss": 2.138,
+      "step": 6513
+    },
+    {
+      "epoch": 0.6694749820199322,
+      "grad_norm": 0.06087846681475639,
+      "learning_rate": 0.01,
+      "loss": 2.15,
+      "step": 6516
+    },
+    {
+      "epoch": 0.6697832117538272,
+      "grad_norm": 0.047256652265787125,
+      "learning_rate": 0.01,
+      "loss": 2.1433,
+      "step": 6519
+    },
+    {
+      "epoch": 0.6700914414877222,
+      "grad_norm": 0.10626421123743057,
+      "learning_rate": 0.01,
+      "loss": 2.156,
+      "step": 6522
+    },
+    {
+      "epoch": 0.6703996712216171,
+      "grad_norm": 0.09426552802324295,
+      "learning_rate": 0.01,
+      "loss": 2.1472,
+      "step": 6525
+    },
+    {
+      "epoch": 0.6707079009555121,
+      "grad_norm": 0.0632442831993103,
+      "learning_rate": 0.01,
+      "loss": 2.1536,
+      "step": 6528
+    },
+    {
+      "epoch": 0.6710161306894071,
+      "grad_norm": 0.07149971276521683,
+      "learning_rate": 0.01,
+      "loss": 2.1694,
+      "step": 6531
+    },
+    {
+      "epoch": 0.6713243604233021,
+      "grad_norm": 0.04060966521501541,
+      "learning_rate": 0.01,
+      "loss": 2.164,
+      "step": 6534
+    },
+    {
+      "epoch": 0.6716325901571971,
+      "grad_norm": 0.20043891668319702,
+      "learning_rate": 0.01,
+      "loss": 2.125,
+      "step": 6537
+    },
+    {
+      "epoch": 0.6719408198910921,
+      "grad_norm": 0.06755783408880234,
+      "learning_rate": 0.01,
+      "loss": 2.15,
+      "step": 6540
+    },
+    {
+      "epoch": 0.6722490496249871,
+      "grad_norm": 0.0509268082678318,
+      "learning_rate": 0.01,
+      "loss": 2.1405,
+      "step": 6543
+    },
+    {
+      "epoch": 0.6725572793588821,
+      "grad_norm": 0.04033916816115379,
+      "learning_rate": 0.01,
+      "loss": 2.136,
+      "step": 6546
+    },
+    {
+      "epoch": 0.6728655090927772,
+      "grad_norm": 0.04707946255803108,
+      "learning_rate": 0.01,
+      "loss": 2.1514,
+      "step": 6549
+    },
+    {
+      "epoch": 0.6731737388266722,
+      "grad_norm": 0.04360898956656456,
+      "learning_rate": 0.01,
+      "loss": 2.1518,
+      "step": 6552
+    },
+    {
+      "epoch": 0.6734819685605672,
+      "grad_norm": 0.11959343403577805,
+      "learning_rate": 0.01,
+      "loss": 2.1377,
+      "step": 6555
+    },
+    {
+      "epoch": 0.6737901982944622,
+      "grad_norm": 0.06620760262012482,
+      "learning_rate": 0.01,
+      "loss": 2.1419,
+      "step": 6558
+    },
+    {
+      "epoch": 0.6740984280283572,
+      "grad_norm": 0.056747015565633774,
+      "learning_rate": 0.01,
+      "loss": 2.138,
+      "step": 6561
+    },
+    {
+      "epoch": 0.6744066577622522,
+      "grad_norm": 0.05230560526251793,
+      "learning_rate": 0.01,
+      "loss": 2.1335,
+      "step": 6564
+    },
+    {
+      "epoch": 0.6747148874961472,
+      "grad_norm": 0.0526299811899662,
+      "learning_rate": 0.01,
+      "loss": 2.131,
+      "step": 6567
+    },
+    {
+      "epoch": 0.6750231172300422,
+      "grad_norm": 0.15683774650096893,
+      "learning_rate": 0.01,
+      "loss": 2.1167,
+      "step": 6570
+    },
+    {
+      "epoch": 0.6753313469639372,
+      "grad_norm": 0.10133557766675949,
+      "learning_rate": 0.01,
+      "loss": 2.1219,
+      "step": 6573
+    },
+    {
+      "epoch": 0.6756395766978321,
+      "grad_norm": 0.06826774775981903,
+      "learning_rate": 0.01,
+      "loss": 2.1416,
+      "step": 6576
+    },
+    {
+      "epoch": 0.6759478064317271,
+      "grad_norm": 0.046236682683229446,
+      "learning_rate": 0.01,
+      "loss": 2.1704,
+      "step": 6579
+    },
+    {
+      "epoch": 0.6762560361656221,
+      "grad_norm": 0.07654762268066406,
+      "learning_rate": 0.01,
+      "loss": 2.1411,
+      "step": 6582
+    },
+    {
+      "epoch": 0.6765642658995171,
+      "grad_norm": 0.07760706543922424,
+      "learning_rate": 0.01,
+      "loss": 2.168,
+      "step": 6585
+    },
+    {
+      "epoch": 0.6768724956334121,
+      "grad_norm": 0.04213540256023407,
+      "learning_rate": 0.01,
+      "loss": 2.1899,
+      "step": 6588
+    },
+    {
+      "epoch": 0.6771807253673071,
+      "grad_norm": 0.0517420619726181,
+      "learning_rate": 0.01,
+      "loss": 2.1561,
+      "step": 6591
+    },
+    {
+      "epoch": 0.6774889551012021,
+      "grad_norm": 0.04073292762041092,
+      "learning_rate": 0.01,
+      "loss": 2.1475,
+      "step": 6594
+    },
+    {
+      "epoch": 0.6777971848350971,
+      "grad_norm": 0.11223835498094559,
+      "learning_rate": 0.01,
+      "loss": 2.1102,
+      "step": 6597
+    },
+    {
+      "epoch": 0.6781054145689921,
+      "grad_norm": 0.08094224333763123,
+      "learning_rate": 0.01,
+      "loss": 2.1537,
+      "step": 6600
+    },
+    {
+      "epoch": 0.6784136443028871,
+      "grad_norm": 0.036313675343990326,
+      "learning_rate": 0.01,
+      "loss": 2.1471,
+      "step": 6603
+    },
+    {
+      "epoch": 0.6787218740367821,
+      "grad_norm": 0.09553749114274979,
+      "learning_rate": 0.01,
+      "loss": 2.1445,
+      "step": 6606
+    },
+    {
+      "epoch": 0.6790301037706771,
+      "grad_norm": 0.07334265112876892,
+      "learning_rate": 0.01,
+      "loss": 2.1594,
+      "step": 6609
+    },
+    {
+      "epoch": 0.6793383335045721,
+      "grad_norm": 0.12031051516532898,
+      "learning_rate": 0.01,
+      "loss": 2.1321,
+      "step": 6612
+    },
+    {
+      "epoch": 0.6796465632384671,
+      "grad_norm": 0.08834968507289886,
+      "learning_rate": 0.01,
+      "loss": 2.1474,
+      "step": 6615
+    },
+    {
+      "epoch": 0.679954792972362,
+      "grad_norm": 0.05016850307583809,
+      "learning_rate": 0.01,
+      "loss": 2.1582,
+      "step": 6618
+    },
+    {
+      "epoch": 0.680263022706257,
+      "grad_norm": 0.039213377982378006,
+      "learning_rate": 0.01,
+      "loss": 2.1461,
+      "step": 6621
+    },
+    {
+      "epoch": 0.680571252440152,
+      "grad_norm": 0.035611145198345184,
+      "learning_rate": 0.01,
+      "loss": 2.137,
+      "step": 6624
+    },
+    {
+      "epoch": 0.680879482174047,
+      "grad_norm": 0.09345167875289917,
+      "learning_rate": 0.01,
+      "loss": 2.1357,
+      "step": 6627
+    },
+    {
+      "epoch": 0.681187711907942,
+      "grad_norm": 0.04311450198292732,
+      "learning_rate": 0.01,
+      "loss": 2.1413,
+      "step": 6630
+    },
+    {
+      "epoch": 0.681495941641837,
+      "grad_norm": 0.040315765887498856,
+      "learning_rate": 0.01,
+      "loss": 2.1091,
+      "step": 6633
+    },
+    {
+      "epoch": 0.681804171375732,
+      "grad_norm": 0.11044291406869888,
+      "learning_rate": 0.01,
+      "loss": 2.1392,
+      "step": 6636
+    },
+    {
+      "epoch": 0.682112401109627,
+      "grad_norm": 0.1288553774356842,
+      "learning_rate": 0.01,
+      "loss": 2.1129,
+      "step": 6639
+    },
+    {
+      "epoch": 0.682420630843522,
+      "grad_norm": 0.0698169469833374,
+      "learning_rate": 0.01,
+      "loss": 2.137,
+      "step": 6642
+    },
+    {
+      "epoch": 0.682728860577417,
+      "grad_norm": 0.037890784442424774,
+      "learning_rate": 0.01,
+      "loss": 2.1195,
+      "step": 6645
+    },
+    {
+      "epoch": 0.683037090311312,
+      "grad_norm": 0.07425201684236526,
+      "learning_rate": 0.01,
+      "loss": 2.1194,
+      "step": 6648
+    },
+    {
+      "epoch": 0.683345320045207,
+      "grad_norm": 0.058168716728687286,
+      "learning_rate": 0.01,
+      "loss": 2.1371,
+      "step": 6651
+    },
+    {
+      "epoch": 0.683653549779102,
+      "grad_norm": 0.05515358969569206,
+      "learning_rate": 0.01,
+      "loss": 2.137,
+      "step": 6654
+    },
+    {
+      "epoch": 0.683961779512997,
+      "grad_norm": 0.0501445047557354,
+      "learning_rate": 0.01,
+      "loss": 2.1539,
+      "step": 6657
+    },
+    {
+      "epoch": 0.684270009246892,
+      "grad_norm": 0.06167145445942879,
+      "learning_rate": 0.01,
+      "loss": 2.1413,
+      "step": 6660
+    },
+    {
+      "epoch": 0.6845782389807871,
+      "grad_norm": 0.0841723158955574,
+      "learning_rate": 0.01,
+      "loss": 2.1194,
+      "step": 6663
+    },
+    {
+      "epoch": 0.6848864687146821,
+      "grad_norm": 0.06027607619762421,
+      "learning_rate": 0.01,
+      "loss": 2.158,
+      "step": 6666
+    },
+    {
+      "epoch": 0.685194698448577,
+      "grad_norm": 0.1187741607427597,
+      "learning_rate": 0.01,
+      "loss": 2.1651,
+      "step": 6669
+    },
+    {
+      "epoch": 0.685502928182472,
+      "grad_norm": 0.10789939761161804,
+      "learning_rate": 0.01,
+      "loss": 2.1465,
+      "step": 6672
+    },
+    {
+      "epoch": 0.685811157916367,
+      "grad_norm": 0.06254967302083969,
+      "learning_rate": 0.01,
+      "loss": 2.1639,
+      "step": 6675
+    },
+    {
+      "epoch": 0.686119387650262,
+      "grad_norm": 0.04242802783846855,
+      "learning_rate": 0.01,
+      "loss": 2.1563,
+      "step": 6678
+    },
+    {
+      "epoch": 0.686427617384157,
+      "grad_norm": 0.03538980334997177,
+      "learning_rate": 0.01,
+      "loss": 2.1373,
+      "step": 6681
+    },
+    {
+      "epoch": 0.686735847118052,
+      "grad_norm": 0.04609490931034088,
+      "learning_rate": 0.01,
+      "loss": 2.1345,
+      "step": 6684
+    },
+    {
+      "epoch": 0.687044076851947,
+      "grad_norm": 0.1298975795507431,
+      "learning_rate": 0.01,
+      "loss": 2.1446,
+      "step": 6687
+    },
+    {
+      "epoch": 0.687352306585842,
+      "grad_norm": 0.10049281269311905,
+      "learning_rate": 0.01,
+      "loss": 2.1432,
+      "step": 6690
+    },
+    {
+      "epoch": 0.687660536319737,
+      "grad_norm": 0.05908266827464104,
+      "learning_rate": 0.01,
+      "loss": 2.1288,
+      "step": 6693
+    },
+    {
+      "epoch": 0.687968766053632,
+      "grad_norm": 0.0546141043305397,
+      "learning_rate": 0.01,
+      "loss": 2.1086,
+      "step": 6696
+    },
+    {
+      "epoch": 0.688276995787527,
+      "grad_norm": 0.04135862737894058,
+      "learning_rate": 0.01,
+      "loss": 2.1187,
+      "step": 6699
+    },
+    {
+      "epoch": 0.688585225521422,
+      "grad_norm": 0.03824761137366295,
+      "learning_rate": 0.01,
+      "loss": 2.1162,
+      "step": 6702
+    },
+    {
+      "epoch": 0.688893455255317,
+      "grad_norm": 0.041454900056123734,
+      "learning_rate": 0.01,
+      "loss": 2.1304,
+      "step": 6705
+    },
+    {
+      "epoch": 0.689201684989212,
+      "grad_norm": 0.08948934823274612,
+      "learning_rate": 0.01,
+      "loss": 2.1538,
+      "step": 6708
+    },
+    {
+      "epoch": 0.689509914723107,
+      "grad_norm": 0.07379783689975739,
+      "learning_rate": 0.01,
+      "loss": 2.145,
+      "step": 6711
+    },
+    {
+      "epoch": 0.689818144457002,
+      "grad_norm": 0.0833912044763565,
+      "learning_rate": 0.01,
+      "loss": 2.1218,
+      "step": 6714
+    },
+    {
+      "epoch": 0.690126374190897,
+      "grad_norm": 0.05899098515510559,
+      "learning_rate": 0.01,
+      "loss": 2.1516,
+      "step": 6717
+    },
+    {
+      "epoch": 0.690434603924792,
+      "grad_norm": 0.06462058424949646,
+      "learning_rate": 0.01,
+      "loss": 2.1496,
+      "step": 6720
+    },
+    {
+      "epoch": 0.6907428336586869,
+      "grad_norm": 0.04040443152189255,
+      "learning_rate": 0.01,
+      "loss": 2.1311,
+      "step": 6723
+    },
+    {
+      "epoch": 0.6910510633925819,
+      "grad_norm": 0.05336814373731613,
+      "learning_rate": 0.01,
+      "loss": 2.1227,
+      "step": 6726
+    },
+    {
+      "epoch": 0.6913592931264769,
+      "grad_norm": 0.05057406798005104,
+      "learning_rate": 0.01,
+      "loss": 2.1281,
+      "step": 6729
+    },
+    {
+      "epoch": 0.6916675228603719,
+      "grad_norm": 0.08063513040542603,
+      "learning_rate": 0.01,
+      "loss": 2.1318,
+      "step": 6732
+    },
+    {
+      "epoch": 0.6919757525942669,
+      "grad_norm": 0.08304840326309204,
+      "learning_rate": 0.01,
+      "loss": 2.1179,
+      "step": 6735
+    },
+    {
+      "epoch": 0.6922839823281619,
+      "grad_norm": 0.04266434162855148,
+      "learning_rate": 0.01,
+      "loss": 2.1447,
+      "step": 6738
+    },
+    {
+      "epoch": 0.6925922120620569,
+      "grad_norm": 0.07502007484436035,
+      "learning_rate": 0.01,
+      "loss": 2.1173,
+      "step": 6741
+    },
+    {
+      "epoch": 0.6929004417959519,
+      "grad_norm": 0.10870220512151718,
+      "learning_rate": 0.01,
+      "loss": 2.1555,
+      "step": 6744
+    },
+    {
+      "epoch": 0.6932086715298469,
+      "grad_norm": 0.15824924409389496,
+      "learning_rate": 0.01,
+      "loss": 2.1668,
+      "step": 6747
+    },
+    {
+      "epoch": 0.6935169012637419,
+      "grad_norm": 0.06319935619831085,
+      "learning_rate": 0.01,
+      "loss": 2.1788,
+      "step": 6750
+    },
+    {
+      "epoch": 0.6938251309976369,
+      "grad_norm": 0.06392507255077362,
+      "learning_rate": 0.01,
+      "loss": 2.1398,
+      "step": 6753
+    },
+    {
+      "epoch": 0.6941333607315319,
+      "grad_norm": 0.044481996446847916,
+      "learning_rate": 0.01,
+      "loss": 2.147,
+      "step": 6756
+    },
+    {
+      "epoch": 0.6944415904654269,
+      "grad_norm": 0.09093592315912247,
+      "learning_rate": 0.01,
+      "loss": 2.1399,
+      "step": 6759
+    },
+    {
+      "epoch": 0.6947498201993219,
+      "grad_norm": 0.09249415248632431,
+      "learning_rate": 0.01,
+      "loss": 2.1274,
+      "step": 6762
+    },
+    {
+      "epoch": 0.6950580499332168,
+      "grad_norm": 0.06134162098169327,
+      "learning_rate": 0.01,
+      "loss": 2.142,
+      "step": 6765
+    },
+    {
+      "epoch": 0.6953662796671118,
+      "grad_norm": 0.048883359879255295,
+      "learning_rate": 0.01,
+      "loss": 2.1357,
+      "step": 6768
+    },
+    {
+      "epoch": 0.6956745094010068,
+      "grad_norm": 0.04553356394171715,
+      "learning_rate": 0.01,
+      "loss": 2.132,
+      "step": 6771
+    },
+    {
+      "epoch": 0.6959827391349018,
+      "grad_norm": 0.10365505516529083,
+      "learning_rate": 0.01,
+      "loss": 2.1568,
+      "step": 6774
+    },
+    {
+      "epoch": 0.6962909688687968,
+      "grad_norm": 0.07474958896636963,
+      "learning_rate": 0.01,
+      "loss": 2.1209,
+      "step": 6777
+    },
+    {
+      "epoch": 0.6965991986026919,
+      "grad_norm": 0.11140461266040802,
+      "learning_rate": 0.01,
+      "loss": 2.1585,
+      "step": 6780
+    },
+    {
+      "epoch": 0.6969074283365869,
+      "grad_norm": 0.0529690645635128,
+      "learning_rate": 0.01,
+      "loss": 2.1228,
+      "step": 6783
+    },
+    {
+      "epoch": 0.6972156580704819,
+      "grad_norm": 0.06484264135360718,
+      "learning_rate": 0.01,
+      "loss": 2.117,
+      "step": 6786
+    },
+    {
+      "epoch": 0.6975238878043769,
+      "grad_norm": 0.0467400886118412,
+      "learning_rate": 0.01,
+      "loss": 2.1367,
+      "step": 6789
+    },
+    {
+      "epoch": 0.6978321175382719,
+      "grad_norm": 0.09690822660923004,
+      "learning_rate": 0.01,
+      "loss": 2.1275,
+      "step": 6792
+    },
+    {
+      "epoch": 0.6981403472721669,
+      "grad_norm": 0.053299982100725174,
+      "learning_rate": 0.01,
+      "loss": 2.1557,
+      "step": 6795
+    },
+    {
+      "epoch": 0.6984485770060619,
+      "grad_norm": 0.08451724797487259,
+      "learning_rate": 0.01,
+      "loss": 2.1235,
+      "step": 6798
+    },
+    {
+      "epoch": 0.6987568067399569,
+      "grad_norm": 0.11180119216442108,
+      "learning_rate": 0.01,
+      "loss": 2.1389,
+      "step": 6801
+    },
+    {
+      "epoch": 0.6990650364738519,
+      "grad_norm": 0.04366112872958183,
+      "learning_rate": 0.01,
+      "loss": 2.1345,
+      "step": 6804
+    },
+    {
+      "epoch": 0.6993732662077469,
+      "grad_norm": 0.057021014392375946,
+      "learning_rate": 0.01,
+      "loss": 2.145,
+      "step": 6807
+    },
+    {
+      "epoch": 0.6996814959416419,
+      "grad_norm": 0.050035975873470306,
+      "learning_rate": 0.01,
+      "loss": 2.1245,
+      "step": 6810
+    },
+    {
+      "epoch": 0.6999897256755369,
+      "grad_norm": 0.16434957087039948,
+      "learning_rate": 0.01,
+      "loss": 2.099,
+      "step": 6813
+    },
+    {
+      "epoch": 0.7002979554094318,
+      "grad_norm": 0.0473979152739048,
+      "learning_rate": 0.01,
+      "loss": 2.124,
+      "step": 6816
+    },
+    {
+      "epoch": 0.7006061851433268,
+      "grad_norm": 0.06207640469074249,
+      "learning_rate": 0.01,
+      "loss": 2.1528,
+      "step": 6819
+    },
+    {
+      "epoch": 0.7009144148772218,
+      "grad_norm": 0.09829109162092209,
+      "learning_rate": 0.01,
+      "loss": 2.1359,
+      "step": 6822
+    },
+    {
+      "epoch": 0.7012226446111168,
+      "grad_norm": 0.0563257597386837,
+      "learning_rate": 0.01,
+      "loss": 2.1639,
+      "step": 6825
+    },
+    {
+      "epoch": 0.7015308743450118,
+      "grad_norm": 0.12371699512004852,
+      "learning_rate": 0.01,
+      "loss": 2.1479,
+      "step": 6828
+    },
+    {
+      "epoch": 0.7018391040789068,
+      "grad_norm": 0.07342347502708435,
+      "learning_rate": 0.01,
+      "loss": 2.1786,
+      "step": 6831
+    },
+    {
+      "epoch": 0.7021473338128018,
+      "grad_norm": 0.05420146882534027,
+      "learning_rate": 0.01,
+      "loss": 2.1261,
+      "step": 6834
+    },
+    {
+      "epoch": 0.7024555635466968,
+      "grad_norm": 0.04500873014330864,
+      "learning_rate": 0.01,
+      "loss": 2.1356,
+      "step": 6837
+    },
+    {
+      "epoch": 0.7027637932805918,
+      "grad_norm": 0.10648415237665176,
+      "learning_rate": 0.01,
+      "loss": 2.1205,
+      "step": 6840
+    },
+    {
+      "epoch": 0.7030720230144868,
+      "grad_norm": 0.05089351162314415,
+      "learning_rate": 0.01,
+      "loss": 2.1403,
+      "step": 6843
+    },
+    {
+      "epoch": 0.7033802527483818,
+      "grad_norm": 0.10011807084083557,
+      "learning_rate": 0.01,
+      "loss": 2.1508,
+      "step": 6846
+    },
+    {
+      "epoch": 0.7036884824822768,
+      "grad_norm": 0.06787194311618805,
+      "learning_rate": 0.01,
+      "loss": 2.1391,
+      "step": 6849
+    },
+    {
+      "epoch": 0.7039967122161718,
+      "grad_norm": 0.08248817175626755,
+      "learning_rate": 0.01,
+      "loss": 2.1782,
+      "step": 6852
+    },
+    {
+      "epoch": 0.7043049419500668,
+      "grad_norm": 0.04949905723333359,
+      "learning_rate": 0.01,
+      "loss": 2.1401,
+      "step": 6855
+    },
+    {
+      "epoch": 0.7046131716839618,
+      "grad_norm": 0.043910931795835495,
+      "learning_rate": 0.01,
+      "loss": 2.108,
+      "step": 6858
+    },
+    {
+      "epoch": 0.7049214014178568,
+      "grad_norm": 0.05133078247308731,
+      "learning_rate": 0.01,
+      "loss": 2.1088,
+      "step": 6861
+    },
+    {
+      "epoch": 0.7052296311517517,
+      "grad_norm": 0.11582443863153458,
+      "learning_rate": 0.01,
+      "loss": 2.1301,
+      "step": 6864
+    },
+    {
+      "epoch": 0.7055378608856467,
+      "grad_norm": 0.04287354275584221,
+      "learning_rate": 0.01,
+      "loss": 2.124,
+      "step": 6867
+    },
+    {
+      "epoch": 0.7058460906195417,
+      "grad_norm": 0.09393726289272308,
+      "learning_rate": 0.01,
+      "loss": 2.1326,
+      "step": 6870
+    },
+    {
+      "epoch": 0.7061543203534367,
+      "grad_norm": 0.1286250203847885,
+      "learning_rate": 0.01,
+      "loss": 2.1292,
+      "step": 6873
+    },
+    {
+      "epoch": 0.7064625500873317,
+      "grad_norm": 0.14816388487815857,
+      "learning_rate": 0.01,
+      "loss": 2.1439,
+      "step": 6876
+    },
+    {
+      "epoch": 0.7067707798212267,
+      "grad_norm": 0.062444012612104416,
+      "learning_rate": 0.01,
+      "loss": 2.1421,
+      "step": 6879
+    },
+    {
+      "epoch": 0.7070790095551217,
+      "grad_norm": 0.053750455379486084,
+      "learning_rate": 0.01,
+      "loss": 2.1185,
+      "step": 6882
+    },
+    {
+      "epoch": 0.7073872392890167,
+      "grad_norm": 0.051356710493564606,
+      "learning_rate": 0.01,
+      "loss": 2.1298,
+      "step": 6885
+    },
+    {
+      "epoch": 0.7076954690229117,
+      "grad_norm": 0.061504025012254715,
+      "learning_rate": 0.01,
+      "loss": 2.1132,
+      "step": 6888
+    },
+    {
+      "epoch": 0.7080036987568067,
+      "grad_norm": 0.056496761739254,
+      "learning_rate": 0.01,
+      "loss": 2.1019,
+      "step": 6891
+    },
+    {
+      "epoch": 0.7083119284907017,
+      "grad_norm": 0.048710647970438004,
+      "learning_rate": 0.01,
+      "loss": 2.126,
+      "step": 6894
+    },
+    {
+      "epoch": 0.7086201582245968,
+      "grad_norm": 0.06260757148265839,
+      "learning_rate": 0.01,
+      "loss": 2.1534,
+      "step": 6897
+    },
+    {
+      "epoch": 0.7089283879584918,
+      "grad_norm": 0.06622278690338135,
+      "learning_rate": 0.01,
+      "loss": 2.1247,
+      "step": 6900
+    },
+    {
+      "epoch": 0.7092366176923868,
+      "grad_norm": 0.0810452550649643,
+      "learning_rate": 0.01,
+      "loss": 2.1336,
+      "step": 6903
+    },
+    {
+      "epoch": 0.7095448474262818,
+      "grad_norm": 0.04692875221371651,
+      "learning_rate": 0.01,
+      "loss": 2.1096,
+      "step": 6906
+    },
+    {
+      "epoch": 0.7098530771601768,
+      "grad_norm": 0.04757360368967056,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 6909
+    },
+    {
+      "epoch": 0.7101613068940718,
+      "grad_norm": 0.05597659945487976,
+      "learning_rate": 0.01,
+      "loss": 2.1425,
+      "step": 6912
+    },
+    {
+      "epoch": 0.7104695366279667,
+      "grad_norm": 0.051605843007564545,
+      "learning_rate": 0.01,
+      "loss": 2.1118,
+      "step": 6915
+    },
+    {
+      "epoch": 0.7107777663618617,
+      "grad_norm": 0.06179991737008095,
+      "learning_rate": 0.01,
+      "loss": 2.1362,
+      "step": 6918
+    },
+    {
+      "epoch": 0.7110859960957567,
+      "grad_norm": 0.05455191805958748,
+      "learning_rate": 0.01,
+      "loss": 2.1279,
+      "step": 6921
+    },
+    {
+      "epoch": 0.7113942258296517,
+      "grad_norm": 0.11560655385255814,
+      "learning_rate": 0.01,
+      "loss": 2.1316,
+      "step": 6924
+    },
+    {
+      "epoch": 0.7117024555635467,
+      "grad_norm": 0.12203246355056763,
+      "learning_rate": 0.01,
+      "loss": 2.1173,
+      "step": 6927
+    },
+    {
+      "epoch": 0.7120106852974417,
+      "grad_norm": 0.07024069130420685,
+      "learning_rate": 0.01,
+      "loss": 2.1395,
+      "step": 6930
+    },
+    {
+      "epoch": 0.7123189150313367,
+      "grad_norm": 0.04773107171058655,
+      "learning_rate": 0.01,
+      "loss": 2.1455,
+      "step": 6933
+    },
+    {
+      "epoch": 0.7126271447652317,
+      "grad_norm": 0.06106821820139885,
+      "learning_rate": 0.01,
+      "loss": 2.1352,
+      "step": 6936
+    },
+    {
+      "epoch": 0.7129353744991267,
+      "grad_norm": 0.11438222974538803,
+      "learning_rate": 0.01,
+      "loss": 2.149,
+      "step": 6939
+    },
+    {
+      "epoch": 0.7132436042330217,
+      "grad_norm": 0.07224932312965393,
+      "learning_rate": 0.01,
+      "loss": 2.1234,
+      "step": 6942
+    },
+    {
+      "epoch": 0.7135518339669167,
+      "grad_norm": 0.06790932267904282,
+      "learning_rate": 0.01,
+      "loss": 2.1222,
+      "step": 6945
+    },
+    {
+      "epoch": 0.7138600637008117,
+      "grad_norm": 0.12322958558797836,
+      "learning_rate": 0.01,
+      "loss": 2.106,
+      "step": 6948
+    },
+    {
+      "epoch": 0.7141682934347067,
+      "grad_norm": 0.07186157256364822,
+      "learning_rate": 0.01,
+      "loss": 2.1365,
+      "step": 6951
+    },
+    {
+      "epoch": 0.7144765231686017,
+      "grad_norm": 0.05366130173206329,
+      "learning_rate": 0.01,
+      "loss": 2.1264,
+      "step": 6954
+    },
+    {
+      "epoch": 0.7147847529024967,
+      "grad_norm": 0.06682512164115906,
+      "learning_rate": 0.01,
+      "loss": 2.1163,
+      "step": 6957
+    },
+    {
+      "epoch": 0.7150929826363916,
+      "grad_norm": 0.04629479721188545,
+      "learning_rate": 0.01,
+      "loss": 2.126,
+      "step": 6960
+    },
+    {
+      "epoch": 0.7154012123702866,
+      "grad_norm": 0.053164754062891006,
+      "learning_rate": 0.01,
+      "loss": 2.1262,
+      "step": 6963
+    },
+    {
+      "epoch": 0.7157094421041816,
+      "grad_norm": 0.08918699622154236,
+      "learning_rate": 0.01,
+      "loss": 2.157,
+      "step": 6966
+    },
+    {
+      "epoch": 0.7160176718380766,
+      "grad_norm": 0.06226164847612381,
+      "learning_rate": 0.01,
+      "loss": 2.1391,
+      "step": 6969
+    },
+    {
+      "epoch": 0.7163259015719716,
+      "grad_norm": 0.08120178431272507,
+      "learning_rate": 0.01,
+      "loss": 2.1118,
+      "step": 6972
+    },
+    {
+      "epoch": 0.7166341313058666,
+      "grad_norm": 0.06390135735273361,
+      "learning_rate": 0.01,
+      "loss": 2.1302,
+      "step": 6975
+    },
+    {
+      "epoch": 0.7169423610397616,
+      "grad_norm": 0.039068643003702164,
+      "learning_rate": 0.01,
+      "loss": 2.1304,
+      "step": 6978
+    },
+    {
+      "epoch": 0.7172505907736566,
+      "grad_norm": 0.05006824806332588,
+      "learning_rate": 0.01,
+      "loss": 2.1352,
+      "step": 6981
+    },
+    {
+      "epoch": 0.7175588205075516,
+      "grad_norm": 0.03946538642048836,
+      "learning_rate": 0.01,
+      "loss": 2.1513,
+      "step": 6984
+    },
+    {
+      "epoch": 0.7178670502414466,
+      "grad_norm": 0.05072702839970589,
+      "learning_rate": 0.01,
+      "loss": 2.1298,
+      "step": 6987
+    },
+    {
+      "epoch": 0.7181752799753416,
+      "grad_norm": 0.06457548588514328,
+      "learning_rate": 0.01,
+      "loss": 2.1276,
+      "step": 6990
+    },
+    {
+      "epoch": 0.7184835097092366,
+      "grad_norm": 0.05759236589074135,
+      "learning_rate": 0.01,
+      "loss": 2.1198,
+      "step": 6993
+    },
+    {
+      "epoch": 0.7187917394431316,
+      "grad_norm": 0.1151571124792099,
+      "learning_rate": 0.01,
+      "loss": 2.1217,
+      "step": 6996
+    },
+    {
+      "epoch": 0.7190999691770266,
+      "grad_norm": 0.04867241531610489,
+      "learning_rate": 0.01,
+      "loss": 2.1343,
+      "step": 6999
+    },
+    {
+      "epoch": 0.7194081989109216,
+      "grad_norm": 0.074817955493927,
+      "learning_rate": 0.01,
+      "loss": 2.1474,
+      "step": 7002
+    },
+    {
+      "epoch": 0.7197164286448166,
+      "grad_norm": 0.04749060794711113,
+      "learning_rate": 0.01,
+      "loss": 2.1403,
+      "step": 7005
+    },
+    {
+      "epoch": 0.7200246583787115,
+      "grad_norm": 0.04965493455529213,
+      "learning_rate": 0.01,
+      "loss": 2.142,
+      "step": 7008
+    },
+    {
+      "epoch": 0.7203328881126067,
+      "grad_norm": 0.044914234429597855,
+      "learning_rate": 0.01,
+      "loss": 2.1397,
+      "step": 7011
+    },
+    {
+      "epoch": 0.7206411178465016,
+      "grad_norm": 0.06727777421474457,
+      "learning_rate": 0.01,
+      "loss": 2.1443,
+      "step": 7014
+    },
+    {
+      "epoch": 0.7209493475803966,
+      "grad_norm": 0.10670837014913559,
+      "learning_rate": 0.01,
+      "loss": 2.1316,
+      "step": 7017
+    },
+    {
+      "epoch": 0.7212575773142916,
+      "grad_norm": 0.05047740787267685,
+      "learning_rate": 0.01,
+      "loss": 2.1268,
+      "step": 7020
+    },
+    {
+      "epoch": 0.7215658070481866,
+      "grad_norm": 0.055116791278123856,
+      "learning_rate": 0.01,
+      "loss": 2.1194,
+      "step": 7023
+    },
+    {
+      "epoch": 0.7218740367820816,
+      "grad_norm": 0.04873311519622803,
+      "learning_rate": 0.01,
+      "loss": 2.1122,
+      "step": 7026
+    },
+    {
+      "epoch": 0.7221822665159766,
+      "grad_norm": 0.0893159881234169,
+      "learning_rate": 0.01,
+      "loss": 2.1413,
+      "step": 7029
+    },
+    {
+      "epoch": 0.7224904962498716,
+      "grad_norm": 0.07278893142938614,
+      "learning_rate": 0.01,
+      "loss": 2.1394,
+      "step": 7032
+    },
+    {
+      "epoch": 0.7227987259837666,
+      "grad_norm": 0.09431196749210358,
+      "learning_rate": 0.01,
+      "loss": 2.1489,
+      "step": 7035
+    },
+    {
+      "epoch": 0.7231069557176616,
+      "grad_norm": 0.03588537499308586,
+      "learning_rate": 0.01,
+      "loss": 2.1585,
+      "step": 7038
+    },
+    {
+      "epoch": 0.7234151854515566,
+      "grad_norm": 0.044003136456012726,
+      "learning_rate": 0.01,
+      "loss": 2.1442,
+      "step": 7041
+    },
+    {
+      "epoch": 0.7237234151854516,
+      "grad_norm": 0.10805044323205948,
+      "learning_rate": 0.01,
+      "loss": 2.127,
+      "step": 7044
+    },
+    {
+      "epoch": 0.7240316449193466,
+      "grad_norm": 0.06328746676445007,
+      "learning_rate": 0.01,
+      "loss": 2.1166,
+      "step": 7047
+    },
+    {
+      "epoch": 0.7243398746532416,
+      "grad_norm": 0.08782347291707993,
+      "learning_rate": 0.01,
+      "loss": 2.1474,
+      "step": 7050
+    },
+    {
+      "epoch": 0.7246481043871366,
+      "grad_norm": 0.06585227698087692,
+      "learning_rate": 0.01,
+      "loss": 2.1228,
+      "step": 7053
+    },
+    {
+      "epoch": 0.7249563341210316,
+      "grad_norm": 0.06324558705091476,
+      "learning_rate": 0.01,
+      "loss": 2.1313,
+      "step": 7056
+    },
+    {
+      "epoch": 0.7252645638549265,
+      "grad_norm": 0.057287219911813736,
+      "learning_rate": 0.01,
+      "loss": 2.1241,
+      "step": 7059
+    },
+    {
+      "epoch": 0.7255727935888215,
+      "grad_norm": 0.07684747129678726,
+      "learning_rate": 0.01,
+      "loss": 2.1299,
+      "step": 7062
+    },
+    {
+      "epoch": 0.7258810233227165,
+      "grad_norm": 0.10347555577754974,
+      "learning_rate": 0.01,
+      "loss": 2.12,
+      "step": 7065
+    },
+    {
+      "epoch": 0.7261892530566115,
+      "grad_norm": 0.06019530072808266,
+      "learning_rate": 0.01,
+      "loss": 2.138,
+      "step": 7068
+    },
+    {
+      "epoch": 0.7264974827905065,
+      "grad_norm": 0.04816723242402077,
+      "learning_rate": 0.01,
+      "loss": 2.1161,
+      "step": 7071
+    },
+    {
+      "epoch": 0.7268057125244015,
+      "grad_norm": 0.05839864909648895,
+      "learning_rate": 0.01,
+      "loss": 2.136,
+      "step": 7074
+    },
+    {
+      "epoch": 0.7271139422582965,
+      "grad_norm": 0.061795271933078766,
+      "learning_rate": 0.01,
+      "loss": 2.1315,
+      "step": 7077
+    },
+    {
+      "epoch": 0.7274221719921915,
+      "grad_norm": 0.05736471712589264,
+      "learning_rate": 0.01,
+      "loss": 2.1403,
+      "step": 7080
+    },
+    {
+      "epoch": 0.7277304017260865,
+      "grad_norm": 0.059238459914922714,
+      "learning_rate": 0.01,
+      "loss": 2.1101,
+      "step": 7083
+    },
+    {
+      "epoch": 0.7280386314599815,
+      "grad_norm": 0.10844148695468903,
+      "learning_rate": 0.01,
+      "loss": 2.1454,
+      "step": 7086
+    },
+    {
+      "epoch": 0.7283468611938765,
+      "grad_norm": 0.047568898648023605,
+      "learning_rate": 0.01,
+      "loss": 2.1183,
+      "step": 7089
+    },
+    {
+      "epoch": 0.7286550909277715,
+      "grad_norm": 0.05178900063037872,
+      "learning_rate": 0.01,
+      "loss": 2.1346,
+      "step": 7092
+    },
+    {
+      "epoch": 0.7289633206616665,
+      "grad_norm": 0.04113532230257988,
+      "learning_rate": 0.01,
+      "loss": 2.0915,
+      "step": 7095
+    },
+    {
+      "epoch": 0.7292715503955615,
+      "grad_norm": 0.10488615930080414,
+      "learning_rate": 0.01,
+      "loss": 2.1239,
+      "step": 7098
+    },
+    {
+      "epoch": 0.7295797801294565,
+      "grad_norm": 0.13013161718845367,
+      "learning_rate": 0.01,
+      "loss": 2.1251,
+      "step": 7101
+    },
+    {
+      "epoch": 0.7298880098633515,
+      "grad_norm": 0.10956915467977524,
+      "learning_rate": 0.01,
+      "loss": 2.1113,
+      "step": 7104
+    },
+    {
+      "epoch": 0.7301962395972464,
+      "grad_norm": 0.06996689736843109,
+      "learning_rate": 0.01,
+      "loss": 2.118,
+      "step": 7107
+    },
+    {
+      "epoch": 0.7305044693311414,
+      "grad_norm": 0.07773365080356598,
+      "learning_rate": 0.01,
+      "loss": 2.1144,
+      "step": 7110
+    },
+    {
+      "epoch": 0.7308126990650364,
+      "grad_norm": 0.06922838091850281,
+      "learning_rate": 0.01,
+      "loss": 2.1148,
+      "step": 7113
+    },
+    {
+      "epoch": 0.7311209287989314,
+      "grad_norm": 0.08941454440355301,
+      "learning_rate": 0.01,
+      "loss": 2.1493,
+      "step": 7116
+    },
+    {
+      "epoch": 0.7314291585328264,
+      "grad_norm": 0.04264171048998833,
+      "learning_rate": 0.01,
+      "loss": 2.136,
+      "step": 7119
+    },
+    {
+      "epoch": 0.7317373882667214,
+      "grad_norm": 0.04473461955785751,
+      "learning_rate": 0.01,
+      "loss": 2.1294,
+      "step": 7122
+    },
+    {
+      "epoch": 0.7320456180006164,
+      "grad_norm": 0.0396125465631485,
+      "learning_rate": 0.01,
+      "loss": 2.1439,
+      "step": 7125
+    },
+    {
+      "epoch": 0.7323538477345115,
+      "grad_norm": 0.04613679647445679,
+      "learning_rate": 0.01,
+      "loss": 2.1503,
+      "step": 7128
+    },
+    {
+      "epoch": 0.7326620774684065,
+      "grad_norm": 0.04897918924689293,
+      "learning_rate": 0.01,
+      "loss": 2.1214,
+      "step": 7131
+    },
+    {
+      "epoch": 0.7329703072023015,
+      "grad_norm": 0.05057375133037567,
+      "learning_rate": 0.01,
+      "loss": 2.1112,
+      "step": 7134
+    },
+    {
+      "epoch": 0.7332785369361965,
+      "grad_norm": 0.05711055174469948,
+      "learning_rate": 0.01,
+      "loss": 2.102,
+      "step": 7137
+    },
+    {
+      "epoch": 0.7335867666700915,
+      "grad_norm": 0.08658434450626373,
+      "learning_rate": 0.01,
+      "loss": 2.1574,
+      "step": 7140
+    },
+    {
+      "epoch": 0.7338949964039865,
+      "grad_norm": 0.07044188678264618,
+      "learning_rate": 0.01,
+      "loss": 2.1037,
+      "step": 7143
+    },
+    {
+      "epoch": 0.7342032261378815,
+      "grad_norm": 0.03941315785050392,
+      "learning_rate": 0.01,
+      "loss": 2.1369,
+      "step": 7146
+    },
+    {
+      "epoch": 0.7345114558717765,
+      "grad_norm": 0.04527783393859863,
+      "learning_rate": 0.01,
+      "loss": 2.1212,
+      "step": 7149
+    },
+    {
+      "epoch": 0.7348196856056715,
+      "grad_norm": 0.07909847050905228,
+      "learning_rate": 0.01,
+      "loss": 2.1316,
+      "step": 7152
+    },
+    {
+      "epoch": 0.7351279153395665,
+      "grad_norm": 0.12793006002902985,
+      "learning_rate": 0.01,
+      "loss": 2.1254,
+      "step": 7155
+    },
+    {
+      "epoch": 0.7354361450734614,
+      "grad_norm": 0.0639350563287735,
+      "learning_rate": 0.01,
+      "loss": 2.1319,
+      "step": 7158
+    },
+    {
+      "epoch": 0.7357443748073564,
+      "grad_norm": 0.0342305451631546,
+      "learning_rate": 0.01,
+      "loss": 2.1386,
+      "step": 7161
+    },
+    {
+      "epoch": 0.7360526045412514,
+      "grad_norm": 0.049001939594745636,
+      "learning_rate": 0.01,
+      "loss": 2.1485,
+      "step": 7164
+    },
+    {
+      "epoch": 0.7363608342751464,
+      "grad_norm": 0.047717638313770294,
+      "learning_rate": 0.01,
+      "loss": 2.1368,
+      "step": 7167
+    },
+    {
+      "epoch": 0.7366690640090414,
+      "grad_norm": 0.04402822256088257,
+      "learning_rate": 0.01,
+      "loss": 2.1162,
+      "step": 7170
+    },
+    {
+      "epoch": 0.7369772937429364,
+      "grad_norm": 0.06922505795955658,
+      "learning_rate": 0.01,
+      "loss": 2.1279,
+      "step": 7173
+    },
+    {
+      "epoch": 0.7372855234768314,
+      "grad_norm": 0.06231709569692612,
+      "learning_rate": 0.01,
+      "loss": 2.0946,
+      "step": 7176
+    },
+    {
+      "epoch": 0.7375937532107264,
+      "grad_norm": 0.11480400711297989,
+      "learning_rate": 0.01,
+      "loss": 2.146,
+      "step": 7179
+    },
+    {
+      "epoch": 0.7379019829446214,
+      "grad_norm": 0.05144179239869118,
+      "learning_rate": 0.01,
+      "loss": 2.1128,
+      "step": 7182
+    },
+    {
+      "epoch": 0.7382102126785164,
+      "grad_norm": 0.05130591616034508,
+      "learning_rate": 0.01,
+      "loss": 2.0964,
+      "step": 7185
+    },
+    {
+      "epoch": 0.7385184424124114,
+      "grad_norm": 0.0549122579395771,
+      "learning_rate": 0.01,
+      "loss": 2.165,
+      "step": 7188
+    },
+    {
+      "epoch": 0.7388266721463064,
+      "grad_norm": 0.1378844678401947,
+      "learning_rate": 0.01,
+      "loss": 2.1367,
+      "step": 7191
+    },
+    {
+      "epoch": 0.7391349018802014,
+      "grad_norm": 0.06231486052274704,
+      "learning_rate": 0.01,
+      "loss": 2.1341,
+      "step": 7194
+    },
+    {
+      "epoch": 0.7394431316140964,
+      "grad_norm": 0.10189559310674667,
+      "learning_rate": 0.01,
+      "loss": 2.1161,
+      "step": 7197
+    },
+    {
+      "epoch": 0.7397513613479914,
+      "grad_norm": 0.053364284336566925,
+      "learning_rate": 0.01,
+      "loss": 2.1043,
+      "step": 7200
+    },
+    {
+      "epoch": 0.7400595910818863,
+      "grad_norm": 0.046057943254709244,
+      "learning_rate": 0.01,
+      "loss": 2.1011,
+      "step": 7203
+    },
+    {
+      "epoch": 0.7403678208157813,
+      "grad_norm": 0.04084615036845207,
+      "learning_rate": 0.01,
+      "loss": 2.1253,
+      "step": 7206
+    },
+    {
+      "epoch": 0.7406760505496763,
+      "grad_norm": 0.04594961181282997,
+      "learning_rate": 0.01,
+      "loss": 2.1228,
+      "step": 7209
+    },
+    {
+      "epoch": 0.7409842802835713,
+      "grad_norm": 0.06608622521162033,
+      "learning_rate": 0.01,
+      "loss": 2.1188,
+      "step": 7212
+    },
+    {
+      "epoch": 0.7412925100174663,
+      "grad_norm": 0.125398188829422,
+      "learning_rate": 0.01,
+      "loss": 2.1057,
+      "step": 7215
+    },
+    {
+      "epoch": 0.7416007397513613,
+      "grad_norm": 0.08068963885307312,
+      "learning_rate": 0.01,
+      "loss": 2.0947,
+      "step": 7218
+    },
+    {
+      "epoch": 0.7419089694852563,
+      "grad_norm": 0.07993921637535095,
+      "learning_rate": 0.01,
+      "loss": 2.1214,
+      "step": 7221
+    },
+    {
+      "epoch": 0.7422171992191513,
+      "grad_norm": 0.04969675466418266,
+      "learning_rate": 0.01,
+      "loss": 2.1099,
+      "step": 7224
+    },
+    {
+      "epoch": 0.7425254289530463,
+      "grad_norm": 0.054677605628967285,
+      "learning_rate": 0.01,
+      "loss": 2.1229,
+      "step": 7227
+    },
+    {
+      "epoch": 0.7428336586869413,
+      "grad_norm": 0.04562999680638313,
+      "learning_rate": 0.01,
+      "loss": 2.1409,
+      "step": 7230
+    },
+    {
+      "epoch": 0.7431418884208363,
+      "grad_norm": 0.07618910074234009,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 7233
+    },
+    {
+      "epoch": 0.7434501181547313,
+      "grad_norm": 0.14368098974227905,
+      "learning_rate": 0.01,
+      "loss": 2.1348,
+      "step": 7236
+    },
+    {
+      "epoch": 0.7437583478886263,
+      "grad_norm": 0.05517590045928955,
+      "learning_rate": 0.01,
+      "loss": 2.116,
+      "step": 7239
+    },
+    {
+      "epoch": 0.7440665776225214,
+      "grad_norm": 0.17316390573978424,
+      "learning_rate": 0.01,
+      "loss": 2.1363,
+      "step": 7242
+    },
+    {
+      "epoch": 0.7443748073564164,
+      "grad_norm": 0.15268415212631226,
+      "learning_rate": 0.01,
+      "loss": 2.1033,
+      "step": 7245
+    },
+    {
+      "epoch": 0.7446830370903114,
+      "grad_norm": 0.06212317943572998,
+      "learning_rate": 0.01,
+      "loss": 2.0971,
+      "step": 7248
+    },
+    {
+      "epoch": 0.7449912668242064,
+      "grad_norm": 0.04282272607088089,
+      "learning_rate": 0.01,
+      "loss": 2.1434,
+      "step": 7251
+    },
+    {
+      "epoch": 0.7452994965581013,
+      "grad_norm": 0.04305952787399292,
+      "learning_rate": 0.01,
+      "loss": 2.1406,
+      "step": 7254
+    },
+    {
+      "epoch": 0.7456077262919963,
+      "grad_norm": 0.048668697476387024,
+      "learning_rate": 0.01,
+      "loss": 2.1303,
+      "step": 7257
+    },
+    {
+      "epoch": 0.7459159560258913,
+      "grad_norm": 0.05524542182683945,
+      "learning_rate": 0.01,
+      "loss": 2.1331,
+      "step": 7260
+    },
+    {
+      "epoch": 0.7462241857597863,
+      "grad_norm": 0.0438026525080204,
+      "learning_rate": 0.01,
+      "loss": 2.109,
+      "step": 7263
+    },
+    {
+      "epoch": 0.7465324154936813,
+      "grad_norm": 0.08154566586017609,
+      "learning_rate": 0.01,
+      "loss": 2.1053,
+      "step": 7266
+    },
+    {
+      "epoch": 0.7468406452275763,
+      "grad_norm": 0.11754357814788818,
+      "learning_rate": 0.01,
+      "loss": 2.1298,
+      "step": 7269
+    },
+    {
+      "epoch": 0.7471488749614713,
+      "grad_norm": 0.06593465805053711,
+      "learning_rate": 0.01,
+      "loss": 2.1323,
+      "step": 7272
+    },
+    {
+      "epoch": 0.7474571046953663,
+      "grad_norm": 0.08065393567085266,
+      "learning_rate": 0.01,
+      "loss": 2.1297,
+      "step": 7275
+    },
+    {
+      "epoch": 0.7477653344292613,
+      "grad_norm": 0.10624121129512787,
+      "learning_rate": 0.01,
+      "loss": 2.1175,
+      "step": 7278
+    },
+    {
+      "epoch": 0.7480735641631563,
+      "grad_norm": 0.06357972323894501,
+      "learning_rate": 0.01,
+      "loss": 2.164,
+      "step": 7281
+    },
+    {
+      "epoch": 0.7483817938970513,
+      "grad_norm": 0.03753754869103432,
+      "learning_rate": 0.01,
+      "loss": 2.1519,
+      "step": 7284
+    },
+    {
+      "epoch": 0.7486900236309463,
+      "grad_norm": 0.04756931588053703,
+      "learning_rate": 0.01,
+      "loss": 2.0901,
+      "step": 7287
+    },
+    {
+      "epoch": 0.7489982533648413,
+      "grad_norm": 0.0494108609855175,
+      "learning_rate": 0.01,
+      "loss": 2.1474,
+      "step": 7290
+    },
+    {
+      "epoch": 0.7493064830987363,
+      "grad_norm": 0.063727006316185,
+      "learning_rate": 0.01,
+      "loss": 2.1425,
+      "step": 7293
+    },
+    {
+      "epoch": 0.7496147128326313,
+      "grad_norm": 0.06327082961797714,
+      "learning_rate": 0.01,
+      "loss": 2.1346,
+      "step": 7296
+    },
+    {
+      "epoch": 0.7499229425665263,
+      "grad_norm": 0.10383486747741699,
+      "learning_rate": 0.01,
+      "loss": 2.1245,
+      "step": 7299
+    },
+    {
+      "epoch": 0.7502311723004212,
+      "grad_norm": 0.10473886877298355,
+      "learning_rate": 0.01,
+      "loss": 2.1302,
+      "step": 7302
+    },
+    {
+      "epoch": 0.7505394020343162,
+      "grad_norm": 0.04905236139893532,
+      "learning_rate": 0.01,
+      "loss": 2.119,
+      "step": 7305
+    },
+    {
+      "epoch": 0.7508476317682112,
+      "grad_norm": 0.04571664705872536,
+      "learning_rate": 0.01,
+      "loss": 2.1505,
+      "step": 7308
+    },
+    {
+      "epoch": 0.7511558615021062,
+      "grad_norm": 0.06305412203073502,
+      "learning_rate": 0.01,
+      "loss": 2.1389,
+      "step": 7311
+    },
+    {
+      "epoch": 0.7514640912360012,
+      "grad_norm": 0.05825283005833626,
+      "learning_rate": 0.01,
+      "loss": 2.1361,
+      "step": 7314
+    },
+    {
+      "epoch": 0.7517723209698962,
+      "grad_norm": 0.059476301074028015,
+      "learning_rate": 0.01,
+      "loss": 2.1191,
+      "step": 7317
+    },
+    {
+      "epoch": 0.7520805507037912,
+      "grad_norm": 0.042396873235702515,
+      "learning_rate": 0.01,
+      "loss": 2.1309,
+      "step": 7320
+    },
+    {
+      "epoch": 0.7523887804376862,
+      "grad_norm": 0.04611228406429291,
+      "learning_rate": 0.01,
+      "loss": 2.1438,
+      "step": 7323
+    },
+    {
+      "epoch": 0.7526970101715812,
+      "grad_norm": 0.09147686511278152,
+      "learning_rate": 0.01,
+      "loss": 2.1288,
+      "step": 7326
+    },
+    {
+      "epoch": 0.7530052399054762,
+      "grad_norm": 0.08085332810878754,
+      "learning_rate": 0.01,
+      "loss": 2.1427,
+      "step": 7329
+    },
+    {
+      "epoch": 0.7533134696393712,
+      "grad_norm": 0.03873496130108833,
+      "learning_rate": 0.01,
+      "loss": 2.1257,
+      "step": 7332
+    },
+    {
+      "epoch": 0.7536216993732662,
+      "grad_norm": 0.05457824096083641,
+      "learning_rate": 0.01,
+      "loss": 2.1373,
+      "step": 7335
+    },
+    {
+      "epoch": 0.7539299291071612,
+      "grad_norm": 0.049249522387981415,
+      "learning_rate": 0.01,
+      "loss": 2.1185,
+      "step": 7338
+    },
+    {
+      "epoch": 0.7542381588410562,
+      "grad_norm": 0.07082841545343399,
+      "learning_rate": 0.01,
+      "loss": 2.1157,
+      "step": 7341
+    },
+    {
+      "epoch": 0.7545463885749512,
+      "grad_norm": 0.046108178794384,
+      "learning_rate": 0.01,
+      "loss": 2.1238,
+      "step": 7344
+    },
+    {
+      "epoch": 0.7548546183088461,
+      "grad_norm": 0.05572620406746864,
+      "learning_rate": 0.01,
+      "loss": 2.1445,
+      "step": 7347
+    },
+    {
+      "epoch": 0.7551628480427411,
+      "grad_norm": 0.1091703474521637,
+      "learning_rate": 0.01,
+      "loss": 2.1281,
+      "step": 7350
+    },
+    {
+      "epoch": 0.7554710777766361,
+      "grad_norm": 0.09372757375240326,
+      "learning_rate": 0.01,
+      "loss": 2.1231,
+      "step": 7353
+    },
+    {
+      "epoch": 0.7557793075105311,
+      "grad_norm": 0.0482059009373188,
+      "learning_rate": 0.01,
+      "loss": 2.1003,
+      "step": 7356
+    },
+    {
+      "epoch": 0.7560875372444262,
+      "grad_norm": 0.041941821575164795,
+      "learning_rate": 0.01,
+      "loss": 2.1382,
+      "step": 7359
+    },
+    {
+      "epoch": 0.7563957669783212,
+      "grad_norm": 0.07122782617807388,
+      "learning_rate": 0.01,
+      "loss": 2.1419,
+      "step": 7362
+    },
+    {
+      "epoch": 0.7567039967122162,
+      "grad_norm": 0.06854265183210373,
+      "learning_rate": 0.01,
+      "loss": 2.1328,
+      "step": 7365
+    },
+    {
+      "epoch": 0.7570122264461112,
+      "grad_norm": 0.10073423385620117,
+      "learning_rate": 0.01,
+      "loss": 2.1322,
+      "step": 7368
+    },
+    {
+      "epoch": 0.7573204561800062,
+      "grad_norm": 0.038869407027959824,
+      "learning_rate": 0.01,
+      "loss": 2.1273,
+      "step": 7371
+    },
+    {
+      "epoch": 0.7576286859139012,
+      "grad_norm": 0.09483812749385834,
+      "learning_rate": 0.01,
+      "loss": 2.1465,
+      "step": 7374
+    },
+    {
+      "epoch": 0.7579369156477962,
+      "grad_norm": 0.07226487994194031,
+      "learning_rate": 0.01,
+      "loss": 2.1386,
+      "step": 7377
+    },
+    {
+      "epoch": 0.7582451453816912,
+      "grad_norm": 0.05041668191552162,
+      "learning_rate": 0.01,
+      "loss": 2.1249,
+      "step": 7380
+    },
+    {
+      "epoch": 0.7585533751155862,
+      "grad_norm": 0.03839525580406189,
+      "learning_rate": 0.01,
+      "loss": 2.1125,
+      "step": 7383
+    },
+    {
+      "epoch": 0.7588616048494812,
+      "grad_norm": 0.047746479511260986,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 7386
+    },
+    {
+      "epoch": 0.7591698345833762,
+      "grad_norm": 0.05524810031056404,
+      "learning_rate": 0.01,
+      "loss": 2.166,
+      "step": 7389
+    },
+    {
+      "epoch": 0.7594780643172712,
+      "grad_norm": 0.050045181065797806,
+      "learning_rate": 0.01,
+      "loss": 2.1411,
+      "step": 7392
+    },
+    {
+      "epoch": 0.7597862940511662,
+      "grad_norm": 0.09187906980514526,
+      "learning_rate": 0.01,
+      "loss": 2.13,
+      "step": 7395
+    },
+    {
+      "epoch": 0.7600945237850611,
+      "grad_norm": 0.15085643529891968,
+      "learning_rate": 0.01,
+      "loss": 2.1198,
+      "step": 7398
+    },
+    {
+      "epoch": 0.7604027535189561,
+      "grad_norm": 0.05295104160904884,
+      "learning_rate": 0.01,
+      "loss": 2.1067,
+      "step": 7401
+    },
+    {
+      "epoch": 0.7607109832528511,
+      "grad_norm": 0.03696104511618614,
+      "learning_rate": 0.01,
+      "loss": 2.1159,
+      "step": 7404
+    },
+    {
+      "epoch": 0.7610192129867461,
+      "grad_norm": 0.04209265485405922,
+      "learning_rate": 0.01,
+      "loss": 2.1243,
+      "step": 7407
+    },
+    {
+      "epoch": 0.7613274427206411,
+      "grad_norm": 0.056943077594041824,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 7410
+    },
+    {
+      "epoch": 0.7616356724545361,
+      "grad_norm": 0.12749402225017548,
+      "learning_rate": 0.01,
+      "loss": 2.1087,
+      "step": 7413
+    },
+    {
+      "epoch": 0.7619439021884311,
+      "grad_norm": 0.09119253605604172,
+      "learning_rate": 0.01,
+      "loss": 2.1252,
+      "step": 7416
+    },
+    {
+      "epoch": 0.7622521319223261,
+      "grad_norm": 0.04251190647482872,
+      "learning_rate": 0.01,
+      "loss": 2.1384,
+      "step": 7419
+    },
+    {
+      "epoch": 0.7625603616562211,
+      "grad_norm": 0.04010685533285141,
+      "learning_rate": 0.01,
+      "loss": 2.1449,
+      "step": 7422
+    },
+    {
+      "epoch": 0.7628685913901161,
+      "grad_norm": 0.05524475499987602,
+      "learning_rate": 0.01,
+      "loss": 2.0841,
+      "step": 7425
+    },
+    {
+      "epoch": 0.7631768211240111,
+      "grad_norm": 0.10250036418437958,
+      "learning_rate": 0.01,
+      "loss": 2.0827,
+      "step": 7428
+    },
+    {
+      "epoch": 0.7634850508579061,
+      "grad_norm": 0.0748668685555458,
+      "learning_rate": 0.01,
+      "loss": 2.128,
+      "step": 7431
+    },
+    {
+      "epoch": 0.7637932805918011,
+      "grad_norm": 0.08616036176681519,
+      "learning_rate": 0.01,
+      "loss": 2.1087,
+      "step": 7434
+    },
+    {
+      "epoch": 0.7641015103256961,
+      "grad_norm": 0.09491308033466339,
+      "learning_rate": 0.01,
+      "loss": 2.1247,
+      "step": 7437
+    },
+    {
+      "epoch": 0.7644097400595911,
+      "grad_norm": 0.08575759083032608,
+      "learning_rate": 0.01,
+      "loss": 2.1419,
+      "step": 7440
+    },
+    {
+      "epoch": 0.764717969793486,
+      "grad_norm": 0.04314613714814186,
+      "learning_rate": 0.01,
+      "loss": 2.1462,
+      "step": 7443
+    },
+    {
+      "epoch": 0.765026199527381,
+      "grad_norm": 0.035719119012355804,
+      "learning_rate": 0.01,
+      "loss": 2.1337,
+      "step": 7446
+    },
+    {
+      "epoch": 0.765334429261276,
+      "grad_norm": 0.04597650095820427,
+      "learning_rate": 0.01,
+      "loss": 2.1258,
+      "step": 7449
+    },
+    {
+      "epoch": 0.765642658995171,
+      "grad_norm": 0.10039210319519043,
+      "learning_rate": 0.01,
+      "loss": 2.1238,
+      "step": 7452
+    },
+    {
+      "epoch": 0.765950888729066,
+      "grad_norm": 0.07157409191131592,
+      "learning_rate": 0.01,
+      "loss": 2.1349,
+      "step": 7455
+    },
+    {
+      "epoch": 0.766259118462961,
+      "grad_norm": 0.09058292210102081,
+      "learning_rate": 0.01,
+      "loss": 2.1232,
+      "step": 7458
+    },
+    {
+      "epoch": 0.766567348196856,
+      "grad_norm": 0.06009940057992935,
+      "learning_rate": 0.01,
+      "loss": 2.1386,
+      "step": 7461
+    },
+    {
+      "epoch": 0.766875577930751,
+      "grad_norm": 0.1165439561009407,
+      "learning_rate": 0.01,
+      "loss": 2.1293,
+      "step": 7464
+    },
+    {
+      "epoch": 0.767183807664646,
+      "grad_norm": 0.06138407811522484,
+      "learning_rate": 0.01,
+      "loss": 2.1066,
+      "step": 7467
+    },
+    {
+      "epoch": 0.767492037398541,
+      "grad_norm": 0.06058945506811142,
+      "learning_rate": 0.01,
+      "loss": 2.1044,
+      "step": 7470
+    },
+    {
+      "epoch": 0.767800267132436,
+      "grad_norm": 0.06741827726364136,
+      "learning_rate": 0.01,
+      "loss": 2.1191,
+      "step": 7473
+    },
+    {
+      "epoch": 0.7681084968663311,
+      "grad_norm": 0.047926925122737885,
+      "learning_rate": 0.01,
+      "loss": 2.1333,
+      "step": 7476
+    },
+    {
+      "epoch": 0.7684167266002261,
+      "grad_norm": 0.06450969725847244,
+      "learning_rate": 0.01,
+      "loss": 2.1061,
+      "step": 7479
+    },
+    {
+      "epoch": 0.7687249563341211,
+      "grad_norm": 0.11133641749620438,
+      "learning_rate": 0.01,
+      "loss": 2.1129,
+      "step": 7482
+    },
+    {
+      "epoch": 0.7690331860680161,
+      "grad_norm": 0.049795158207416534,
+      "learning_rate": 0.01,
+      "loss": 2.136,
+      "step": 7485
+    },
+    {
+      "epoch": 0.7693414158019111,
+      "grad_norm": 0.06083859130740166,
+      "learning_rate": 0.01,
+      "loss": 2.1459,
+      "step": 7488
+    },
+    {
+      "epoch": 0.7696496455358061,
+      "grad_norm": 0.04686833918094635,
+      "learning_rate": 0.01,
+      "loss": 2.1073,
+      "step": 7491
+    },
+    {
+      "epoch": 0.769957875269701,
+      "grad_norm": 0.05475611612200737,
+      "learning_rate": 0.01,
+      "loss": 2.1167,
+      "step": 7494
+    },
+    {
+      "epoch": 0.770266105003596,
+      "grad_norm": 0.04683786630630493,
+      "learning_rate": 0.01,
+      "loss": 2.1491,
+      "step": 7497
+    },
+    {
+      "epoch": 0.770574334737491,
+      "grad_norm": 0.10841275751590729,
+      "learning_rate": 0.01,
+      "loss": 2.0967,
+      "step": 7500
+    },
+    {
+      "epoch": 0.770882564471386,
+      "grad_norm": 0.09716581553220749,
+      "learning_rate": 0.01,
+      "loss": 2.1133,
+      "step": 7503
+    },
+    {
+      "epoch": 0.771190794205281,
+      "grad_norm": 0.04913085699081421,
+      "learning_rate": 0.01,
+      "loss": 2.1401,
+      "step": 7506
+    },
+    {
+      "epoch": 0.771499023939176,
+      "grad_norm": 0.04710682854056358,
+      "learning_rate": 0.01,
+      "loss": 2.109,
+      "step": 7509
+    },
+    {
+      "epoch": 0.771807253673071,
+      "grad_norm": 0.054945673793554306,
+      "learning_rate": 0.01,
+      "loss": 2.1169,
+      "step": 7512
+    },
+    {
+      "epoch": 0.772115483406966,
+      "grad_norm": 0.04265155643224716,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 7515
+    },
+    {
+      "epoch": 0.772423713140861,
+      "grad_norm": 0.03544042259454727,
+      "learning_rate": 0.01,
+      "loss": 2.1172,
+      "step": 7518
+    },
+    {
+      "epoch": 0.772731942874756,
+      "grad_norm": 0.05048484355211258,
+      "learning_rate": 0.01,
+      "loss": 2.1015,
+      "step": 7521
+    },
+    {
+      "epoch": 0.773040172608651,
+      "grad_norm": 0.14160272479057312,
+      "learning_rate": 0.01,
+      "loss": 2.1475,
+      "step": 7524
+    },
+    {
+      "epoch": 0.773348402342546,
+      "grad_norm": 0.08693049848079681,
+      "learning_rate": 0.01,
+      "loss": 2.1266,
+      "step": 7527
+    },
+    {
+      "epoch": 0.773656632076441,
+      "grad_norm": 0.06437800824642181,
+      "learning_rate": 0.01,
+      "loss": 2.1273,
+      "step": 7530
+    },
+    {
+      "epoch": 0.773964861810336,
+      "grad_norm": 0.04450656846165657,
+      "learning_rate": 0.01,
+      "loss": 2.1192,
+      "step": 7533
+    },
+    {
+      "epoch": 0.774273091544231,
+      "grad_norm": 0.05369933694601059,
+      "learning_rate": 0.01,
+      "loss": 2.1264,
+      "step": 7536
+    },
+    {
+      "epoch": 0.774581321278126,
+      "grad_norm": 0.04080953076481819,
+      "learning_rate": 0.01,
+      "loss": 2.1319,
+      "step": 7539
+    },
+    {
+      "epoch": 0.774889551012021,
+      "grad_norm": 0.03433745354413986,
+      "learning_rate": 0.01,
+      "loss": 2.1024,
+      "step": 7542
+    },
+    {
+      "epoch": 0.7751977807459159,
+      "grad_norm": 0.1574896275997162,
+      "learning_rate": 0.01,
+      "loss": 2.1182,
+      "step": 7545
+    },
+    {
+      "epoch": 0.7755060104798109,
+      "grad_norm": 0.1207810789346695,
+      "learning_rate": 0.01,
+      "loss": 2.1052,
+      "step": 7548
+    },
+    {
+      "epoch": 0.7758142402137059,
+      "grad_norm": 0.07270894944667816,
+      "learning_rate": 0.01,
+      "loss": 2.1331,
+      "step": 7551
+    },
+    {
+      "epoch": 0.7761224699476009,
+      "grad_norm": 0.07062831521034241,
+      "learning_rate": 0.01,
+      "loss": 2.099,
+      "step": 7554
+    },
+    {
+      "epoch": 0.7764306996814959,
+      "grad_norm": 0.04142964631319046,
+      "learning_rate": 0.01,
+      "loss": 2.1192,
+      "step": 7557
+    },
+    {
+      "epoch": 0.7767389294153909,
+      "grad_norm": 0.04645151272416115,
+      "learning_rate": 0.01,
+      "loss": 2.1117,
+      "step": 7560
+    },
+    {
+      "epoch": 0.7770471591492859,
+      "grad_norm": 0.046251073479652405,
+      "learning_rate": 0.01,
+      "loss": 2.1399,
+      "step": 7563
+    },
+    {
+      "epoch": 0.7773553888831809,
+      "grad_norm": 0.07185769826173782,
+      "learning_rate": 0.01,
+      "loss": 2.1261,
+      "step": 7566
+    },
+    {
+      "epoch": 0.7776636186170759,
+      "grad_norm": 0.045216575264930725,
+      "learning_rate": 0.01,
+      "loss": 2.1302,
+      "step": 7569
+    },
+    {
+      "epoch": 0.7779718483509709,
+      "grad_norm": 0.04923580586910248,
+      "learning_rate": 0.01,
+      "loss": 2.1482,
+      "step": 7572
+    },
+    {
+      "epoch": 0.7782800780848659,
+      "grad_norm": 0.06434139609336853,
+      "learning_rate": 0.01,
+      "loss": 2.1325,
+      "step": 7575
+    },
+    {
+      "epoch": 0.7785883078187609,
+      "grad_norm": 0.11186740547418594,
+      "learning_rate": 0.01,
+      "loss": 2.1168,
+      "step": 7578
+    },
+    {
+      "epoch": 0.7788965375526559,
+      "grad_norm": 0.06694278120994568,
+      "learning_rate": 0.01,
+      "loss": 2.15,
+      "step": 7581
+    },
+    {
+      "epoch": 0.7792047672865509,
+      "grad_norm": 0.05431769788265228,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 7584
+    },
+    {
+      "epoch": 0.7795129970204459,
+      "grad_norm": 0.05853963643312454,
+      "learning_rate": 0.01,
+      "loss": 2.1145,
+      "step": 7587
+    },
+    {
+      "epoch": 0.779821226754341,
+      "grad_norm": 0.04059399664402008,
+      "learning_rate": 0.01,
+      "loss": 2.1373,
+      "step": 7590
+    },
+    {
+      "epoch": 0.780129456488236,
+      "grad_norm": 0.06444236636161804,
+      "learning_rate": 0.01,
+      "loss": 2.1167,
+      "step": 7593
+    },
+    {
+      "epoch": 0.7804376862221309,
+      "grad_norm": 0.09885245561599731,
+      "learning_rate": 0.01,
+      "loss": 2.1146,
+      "step": 7596
+    },
+    {
+      "epoch": 0.7807459159560259,
+      "grad_norm": 0.08536794036626816,
+      "learning_rate": 0.01,
+      "loss": 2.1282,
+      "step": 7599
+    },
+    {
+      "epoch": 0.7810541456899209,
+      "grad_norm": 0.04299011081457138,
+      "learning_rate": 0.01,
+      "loss": 2.103,
+      "step": 7602
+    },
+    {
+      "epoch": 0.7813623754238159,
+      "grad_norm": 0.060757700353860855,
+      "learning_rate": 0.01,
+      "loss": 2.0923,
+      "step": 7605
+    },
+    {
+      "epoch": 0.7816706051577109,
+      "grad_norm": 0.037401244044303894,
+      "learning_rate": 0.01,
+      "loss": 2.1343,
+      "step": 7608
+    },
+    {
+      "epoch": 0.7819788348916059,
+      "grad_norm": 0.12264932692050934,
+      "learning_rate": 0.01,
+      "loss": 2.1193,
+      "step": 7611
+    },
+    {
+      "epoch": 0.7822870646255009,
+      "grad_norm": 0.052691470831632614,
+      "learning_rate": 0.01,
+      "loss": 2.1097,
+      "step": 7614
+    },
+    {
+      "epoch": 0.7825952943593959,
+      "grad_norm": 0.05509025603532791,
+      "learning_rate": 0.01,
+      "loss": 2.1208,
+      "step": 7617
+    },
+    {
+      "epoch": 0.7829035240932909,
+      "grad_norm": 0.10352631658315659,
+      "learning_rate": 0.01,
+      "loss": 2.1277,
+      "step": 7620
+    },
+    {
+      "epoch": 0.7832117538271859,
+      "grad_norm": 0.05865751951932907,
+      "learning_rate": 0.01,
+      "loss": 2.138,
+      "step": 7623
+    },
+    {
+      "epoch": 0.7835199835610809,
+      "grad_norm": 0.09445837885141373,
+      "learning_rate": 0.01,
+      "loss": 2.1131,
+      "step": 7626
+    },
+    {
+      "epoch": 0.7838282132949759,
+      "grad_norm": 0.11066542565822601,
+      "learning_rate": 0.01,
+      "loss": 2.1315,
+      "step": 7629
+    },
+    {
+      "epoch": 0.7841364430288709,
+      "grad_norm": 0.05489170923829079,
+      "learning_rate": 0.01,
+      "loss": 2.1264,
+      "step": 7632
+    },
+    {
+      "epoch": 0.7844446727627659,
+      "grad_norm": 0.06804061681032181,
+      "learning_rate": 0.01,
+      "loss": 2.1491,
+      "step": 7635
+    },
+    {
+      "epoch": 0.7847529024966609,
+      "grad_norm": 0.07411237061023712,
+      "learning_rate": 0.01,
+      "loss": 2.126,
+      "step": 7638
+    },
+    {
+      "epoch": 0.7850611322305558,
+      "grad_norm": 0.050356972962617874,
+      "learning_rate": 0.01,
+      "loss": 2.1237,
+      "step": 7641
+    },
+    {
+      "epoch": 0.7853693619644508,
+      "grad_norm": 0.06125912442803383,
+      "learning_rate": 0.01,
+      "loss": 2.1328,
+      "step": 7644
+    },
+    {
+      "epoch": 0.7856775916983458,
+      "grad_norm": 0.05983618274331093,
+      "learning_rate": 0.01,
+      "loss": 2.1152,
+      "step": 7647
+    },
+    {
+      "epoch": 0.7859858214322408,
+      "grad_norm": 0.04065684601664543,
+      "learning_rate": 0.01,
+      "loss": 2.1213,
+      "step": 7650
+    },
+    {
+      "epoch": 0.7862940511661358,
+      "grad_norm": 0.05535745993256569,
+      "learning_rate": 0.01,
+      "loss": 2.1106,
+      "step": 7653
+    },
+    {
+      "epoch": 0.7866022809000308,
+      "grad_norm": 0.09727519005537033,
+      "learning_rate": 0.01,
+      "loss": 2.1202,
+      "step": 7656
+    },
+    {
+      "epoch": 0.7869105106339258,
+      "grad_norm": 0.07764584571123123,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 7659
+    },
+    {
+      "epoch": 0.7872187403678208,
+      "grad_norm": 0.04933121055364609,
+      "learning_rate": 0.01,
+      "loss": 2.1217,
+      "step": 7662
+    },
+    {
+      "epoch": 0.7875269701017158,
+      "grad_norm": 0.12199501693248749,
+      "learning_rate": 0.01,
+      "loss": 2.1412,
+      "step": 7665
+    },
+    {
+      "epoch": 0.7878351998356108,
+      "grad_norm": 0.14431309700012207,
+      "learning_rate": 0.01,
+      "loss": 2.1249,
+      "step": 7668
+    },
+    {
+      "epoch": 0.7881434295695058,
+      "grad_norm": 0.07583998888731003,
+      "learning_rate": 0.01,
+      "loss": 2.139,
+      "step": 7671
+    },
+    {
+      "epoch": 0.7884516593034008,
+      "grad_norm": 0.10426465421915054,
+      "learning_rate": 0.01,
+      "loss": 2.1053,
+      "step": 7674
+    },
+    {
+      "epoch": 0.7887598890372958,
+      "grad_norm": 0.06411170959472656,
+      "learning_rate": 0.01,
+      "loss": 2.1105,
+      "step": 7677
+    },
+    {
+      "epoch": 0.7890681187711908,
+      "grad_norm": 0.07436025142669678,
+      "learning_rate": 0.01,
+      "loss": 2.1301,
+      "step": 7680
+    },
+    {
+      "epoch": 0.7893763485050858,
+      "grad_norm": 0.10409426689147949,
+      "learning_rate": 0.01,
+      "loss": 2.1319,
+      "step": 7683
+    },
+    {
+      "epoch": 0.7896845782389807,
+      "grad_norm": 0.05232664570212364,
+      "learning_rate": 0.01,
+      "loss": 2.1458,
+      "step": 7686
+    },
+    {
+      "epoch": 0.7899928079728757,
+      "grad_norm": 0.06705309450626373,
+      "learning_rate": 0.01,
+      "loss": 2.1231,
+      "step": 7689
+    },
+    {
+      "epoch": 0.7903010377067707,
+      "grad_norm": 0.04422546550631523,
+      "learning_rate": 0.01,
+      "loss": 2.0836,
+      "step": 7692
+    },
+    {
+      "epoch": 0.7906092674406657,
+      "grad_norm": 0.04316714033484459,
+      "learning_rate": 0.01,
+      "loss": 2.1117,
+      "step": 7695
+    },
+    {
+      "epoch": 0.7909174971745607,
+      "grad_norm": 0.058282140642404556,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 7698
+    },
+    {
+      "epoch": 0.7912257269084557,
+      "grad_norm": 0.07676571607589722,
+      "learning_rate": 0.01,
+      "loss": 2.1402,
+      "step": 7701
+    },
+    {
+      "epoch": 0.7915339566423507,
+      "grad_norm": 0.07258665561676025,
+      "learning_rate": 0.01,
+      "loss": 2.1458,
+      "step": 7704
+    },
+    {
+      "epoch": 0.7918421863762458,
+      "grad_norm": 0.04850257560610771,
+      "learning_rate": 0.01,
+      "loss": 2.0886,
+      "step": 7707
+    },
+    {
+      "epoch": 0.7921504161101408,
+      "grad_norm": 0.05658482015132904,
+      "learning_rate": 0.01,
+      "loss": 2.1174,
+      "step": 7710
+    },
+    {
+      "epoch": 0.7924586458440358,
+      "grad_norm": 0.06475166231393814,
+      "learning_rate": 0.01,
+      "loss": 2.0995,
+      "step": 7713
+    },
+    {
+      "epoch": 0.7927668755779308,
+      "grad_norm": 0.10428962856531143,
+      "learning_rate": 0.01,
+      "loss": 2.109,
+      "step": 7716
+    },
+    {
+      "epoch": 0.7930751053118258,
+      "grad_norm": 0.04227283224463463,
+      "learning_rate": 0.01,
+      "loss": 2.1124,
+      "step": 7719
+    },
+    {
+      "epoch": 0.7933833350457208,
+      "grad_norm": 0.0594823881983757,
+      "learning_rate": 0.01,
+      "loss": 2.0944,
+      "step": 7722
+    },
+    {
+      "epoch": 0.7936915647796158,
+      "grad_norm": 0.08695527911186218,
+      "learning_rate": 0.01,
+      "loss": 2.1077,
+      "step": 7725
+    },
+    {
+      "epoch": 0.7939997945135108,
+      "grad_norm": 0.06003952398896217,
+      "learning_rate": 0.01,
+      "loss": 2.088,
+      "step": 7728
+    },
+    {
+      "epoch": 0.7943080242474058,
+      "grad_norm": 0.058509476482868195,
+      "learning_rate": 0.01,
+      "loss": 2.1471,
+      "step": 7731
+    },
+    {
+      "epoch": 0.7946162539813008,
+      "grad_norm": 0.048057131469249725,
+      "learning_rate": 0.01,
+      "loss": 2.1252,
+      "step": 7734
+    },
+    {
+      "epoch": 0.7949244837151958,
+      "grad_norm": 0.11144626140594482,
+      "learning_rate": 0.01,
+      "loss": 2.1209,
+      "step": 7737
+    },
+    {
+      "epoch": 0.7952327134490907,
+      "grad_norm": 0.041008081287145615,
+      "learning_rate": 0.01,
+      "loss": 2.1139,
+      "step": 7740
+    },
+    {
+      "epoch": 0.7955409431829857,
+      "grad_norm": 0.04088988155126572,
+      "learning_rate": 0.01,
+      "loss": 2.0927,
+      "step": 7743
+    },
+    {
+      "epoch": 0.7958491729168807,
+      "grad_norm": 0.1495555192232132,
+      "learning_rate": 0.01,
+      "loss": 2.0977,
+      "step": 7746
+    },
+    {
+      "epoch": 0.7961574026507757,
+      "grad_norm": 0.042645204812288284,
+      "learning_rate": 0.01,
+      "loss": 2.1021,
+      "step": 7749
+    },
+    {
+      "epoch": 0.7964656323846707,
+      "grad_norm": 0.04671596363186836,
+      "learning_rate": 0.01,
+      "loss": 2.1015,
+      "step": 7752
+    },
+    {
+      "epoch": 0.7967738621185657,
+      "grad_norm": 0.07249152660369873,
+      "learning_rate": 0.01,
+      "loss": 2.1278,
+      "step": 7755
+    },
+    {
+      "epoch": 0.7970820918524607,
+      "grad_norm": 0.05848756060004234,
+      "learning_rate": 0.01,
+      "loss": 2.1168,
+      "step": 7758
+    },
+    {
+      "epoch": 0.7973903215863557,
+      "grad_norm": 0.05428781732916832,
+      "learning_rate": 0.01,
+      "loss": 2.1228,
+      "step": 7761
+    },
+    {
+      "epoch": 0.7976985513202507,
+      "grad_norm": 0.04751111939549446,
+      "learning_rate": 0.01,
+      "loss": 2.1178,
+      "step": 7764
+    },
+    {
+      "epoch": 0.7980067810541457,
+      "grad_norm": 0.08653240650892258,
+      "learning_rate": 0.01,
+      "loss": 2.1081,
+      "step": 7767
+    },
+    {
+      "epoch": 0.7983150107880407,
+      "grad_norm": 0.04038892313838005,
+      "learning_rate": 0.01,
+      "loss": 2.1028,
+      "step": 7770
+    },
+    {
+      "epoch": 0.7986232405219357,
+      "grad_norm": 0.05703849345445633,
+      "learning_rate": 0.01,
+      "loss": 2.1249,
+      "step": 7773
+    },
+    {
+      "epoch": 0.7989314702558307,
+      "grad_norm": 0.06425055861473083,
+      "learning_rate": 0.01,
+      "loss": 2.1291,
+      "step": 7776
+    },
+    {
+      "epoch": 0.7992396999897257,
+      "grad_norm": 0.05537475273013115,
+      "learning_rate": 0.01,
+      "loss": 2.1122,
+      "step": 7779
+    },
+    {
+      "epoch": 0.7995479297236207,
+      "grad_norm": 0.05172963812947273,
+      "learning_rate": 0.01,
+      "loss": 2.1218,
+      "step": 7782
+    },
+    {
+      "epoch": 0.7998561594575156,
+      "grad_norm": 0.05907023698091507,
+      "learning_rate": 0.01,
+      "loss": 2.1041,
+      "step": 7785
+    },
+    {
+      "epoch": 0.8001643891914106,
+      "grad_norm": 0.10618621110916138,
+      "learning_rate": 0.01,
+      "loss": 2.1266,
+      "step": 7788
+    },
+    {
+      "epoch": 0.8004726189253056,
+      "grad_norm": 0.06189849600195885,
+      "learning_rate": 0.01,
+      "loss": 2.1327,
+      "step": 7791
+    },
+    {
+      "epoch": 0.8007808486592006,
+      "grad_norm": 0.10624901950359344,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 7794
+    },
+    {
+      "epoch": 0.8010890783930956,
+      "grad_norm": 0.04061825945973396,
+      "learning_rate": 0.01,
+      "loss": 2.0859,
+      "step": 7797
+    },
+    {
+      "epoch": 0.8013973081269906,
+      "grad_norm": 0.04402461647987366,
+      "learning_rate": 0.01,
+      "loss": 2.1303,
+      "step": 7800
+    },
+    {
+      "epoch": 0.8017055378608856,
+      "grad_norm": 0.05029004439711571,
+      "learning_rate": 0.01,
+      "loss": 2.1224,
+      "step": 7803
+    },
+    {
+      "epoch": 0.8020137675947806,
+      "grad_norm": 0.055786702781915665,
+      "learning_rate": 0.01,
+      "loss": 2.1296,
+      "step": 7806
+    },
+    {
+      "epoch": 0.8023219973286756,
+      "grad_norm": 0.11740477383136749,
+      "learning_rate": 0.01,
+      "loss": 2.1222,
+      "step": 7809
+    },
+    {
+      "epoch": 0.8026302270625706,
+      "grad_norm": 0.10261218994855881,
+      "learning_rate": 0.01,
+      "loss": 2.1405,
+      "step": 7812
+    },
+    {
+      "epoch": 0.8029384567964656,
+      "grad_norm": 0.05233708769083023,
+      "learning_rate": 0.01,
+      "loss": 2.1118,
+      "step": 7815
+    },
+    {
+      "epoch": 0.8032466865303606,
+      "grad_norm": 0.04390858858823776,
+      "learning_rate": 0.01,
+      "loss": 2.1299,
+      "step": 7818
+    },
+    {
+      "epoch": 0.8035549162642556,
+      "grad_norm": 0.05893026292324066,
+      "learning_rate": 0.01,
+      "loss": 2.1184,
+      "step": 7821
+    },
+    {
+      "epoch": 0.8038631459981507,
+      "grad_norm": 0.06398338079452515,
+      "learning_rate": 0.01,
+      "loss": 2.1057,
+      "step": 7824
+    },
+    {
+      "epoch": 0.8041713757320457,
+      "grad_norm": 0.07129772752523422,
+      "learning_rate": 0.01,
+      "loss": 2.1056,
+      "step": 7827
+    },
+    {
+      "epoch": 0.8044796054659407,
+      "grad_norm": 0.07481534779071808,
+      "learning_rate": 0.01,
+      "loss": 2.1272,
+      "step": 7830
+    },
+    {
+      "epoch": 0.8047878351998357,
+      "grad_norm": 0.049200594425201416,
+      "learning_rate": 0.01,
+      "loss": 2.0942,
+      "step": 7833
+    },
+    {
+      "epoch": 0.8050960649337306,
+      "grad_norm": 0.05124384164810181,
+      "learning_rate": 0.01,
+      "loss": 2.0859,
+      "step": 7836
+    },
+    {
+      "epoch": 0.8054042946676256,
+      "grad_norm": 0.07997792959213257,
+      "learning_rate": 0.01,
+      "loss": 2.1412,
+      "step": 7839
+    },
+    {
+      "epoch": 0.8057125244015206,
+      "grad_norm": 0.12280064076185226,
+      "learning_rate": 0.01,
+      "loss": 2.0826,
+      "step": 7842
+    },
+    {
+      "epoch": 0.8060207541354156,
+      "grad_norm": 0.05292202904820442,
+      "learning_rate": 0.01,
+      "loss": 2.0965,
+      "step": 7845
+    },
+    {
+      "epoch": 0.8063289838693106,
+      "grad_norm": 0.04903187230229378,
+      "learning_rate": 0.01,
+      "loss": 2.0911,
+      "step": 7848
+    },
+    {
+      "epoch": 0.8066372136032056,
+      "grad_norm": 0.06882268935441971,
+      "learning_rate": 0.01,
+      "loss": 2.108,
+      "step": 7851
+    },
+    {
+      "epoch": 0.8069454433371006,
+      "grad_norm": 0.06937083601951599,
+      "learning_rate": 0.01,
+      "loss": 2.1234,
+      "step": 7854
+    },
+    {
+      "epoch": 0.8072536730709956,
+      "grad_norm": 0.10075647383928299,
+      "learning_rate": 0.01,
+      "loss": 2.0983,
+      "step": 7857
+    },
+    {
+      "epoch": 0.8075619028048906,
+      "grad_norm": 0.07185733318328857,
+      "learning_rate": 0.01,
+      "loss": 2.0998,
+      "step": 7860
+    },
+    {
+      "epoch": 0.8078701325387856,
+      "grad_norm": 0.07266184687614441,
+      "learning_rate": 0.01,
+      "loss": 2.1056,
+      "step": 7863
+    },
+    {
+      "epoch": 0.8081783622726806,
+      "grad_norm": 0.05049808695912361,
+      "learning_rate": 0.01,
+      "loss": 2.1126,
+      "step": 7866
+    },
+    {
+      "epoch": 0.8084865920065756,
+      "grad_norm": 0.07260838896036148,
+      "learning_rate": 0.01,
+      "loss": 2.1311,
+      "step": 7869
+    },
+    {
+      "epoch": 0.8087948217404706,
+      "grad_norm": 0.0659325122833252,
+      "learning_rate": 0.01,
+      "loss": 2.1317,
+      "step": 7872
+    },
+    {
+      "epoch": 0.8091030514743656,
+      "grad_norm": 0.056960709393024445,
+      "learning_rate": 0.01,
+      "loss": 2.0988,
+      "step": 7875
+    },
+    {
+      "epoch": 0.8094112812082606,
+      "grad_norm": 0.1266620010137558,
+      "learning_rate": 0.01,
+      "loss": 2.1274,
+      "step": 7878
+    },
+    {
+      "epoch": 0.8097195109421556,
+      "grad_norm": 0.05951874330639839,
+      "learning_rate": 0.01,
+      "loss": 2.1342,
+      "step": 7881
+    },
+    {
+      "epoch": 0.8100277406760505,
+      "grad_norm": 0.06081915274262428,
+      "learning_rate": 0.01,
+      "loss": 2.1036,
+      "step": 7884
+    },
+    {
+      "epoch": 0.8103359704099455,
+      "grad_norm": 0.07136547565460205,
+      "learning_rate": 0.01,
+      "loss": 2.1067,
+      "step": 7887
+    },
+    {
+      "epoch": 0.8106442001438405,
+      "grad_norm": 0.08835722506046295,
+      "learning_rate": 0.01,
+      "loss": 2.1123,
+      "step": 7890
+    },
+    {
+      "epoch": 0.8109524298777355,
+      "grad_norm": 0.04469553008675575,
+      "learning_rate": 0.01,
+      "loss": 2.1117,
+      "step": 7893
+    },
+    {
+      "epoch": 0.8112606596116305,
+      "grad_norm": 0.042171087116003036,
+      "learning_rate": 0.01,
+      "loss": 2.0875,
+      "step": 7896
+    },
+    {
+      "epoch": 0.8115688893455255,
+      "grad_norm": 0.0847015529870987,
+      "learning_rate": 0.01,
+      "loss": 2.0998,
+      "step": 7899
+    },
+    {
+      "epoch": 0.8118771190794205,
+      "grad_norm": 0.09157509356737137,
+      "learning_rate": 0.01,
+      "loss": 2.121,
+      "step": 7902
+    },
+    {
+      "epoch": 0.8121853488133155,
+      "grad_norm": 0.06001126766204834,
+      "learning_rate": 0.01,
+      "loss": 2.1441,
+      "step": 7905
+    },
+    {
+      "epoch": 0.8124935785472105,
+      "grad_norm": 0.03552449122071266,
+      "learning_rate": 0.01,
+      "loss": 2.1371,
+      "step": 7908
+    },
+    {
+      "epoch": 0.8128018082811055,
+      "grad_norm": 0.034304428845644,
+      "learning_rate": 0.01,
+      "loss": 2.1066,
+      "step": 7911
+    },
+    {
+      "epoch": 0.8131100380150005,
+      "grad_norm": 0.04897907376289368,
+      "learning_rate": 0.01,
+      "loss": 2.1054,
+      "step": 7914
+    },
+    {
+      "epoch": 0.8134182677488955,
+      "grad_norm": 0.06674344837665558,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 7917
+    },
+    {
+      "epoch": 0.8137264974827905,
+      "grad_norm": 0.06437379866838455,
+      "learning_rate": 0.01,
+      "loss": 2.1189,
+      "step": 7920
+    },
+    {
+      "epoch": 0.8140347272166855,
+      "grad_norm": 0.06402087956666946,
+      "learning_rate": 0.01,
+      "loss": 2.1111,
+      "step": 7923
+    },
+    {
+      "epoch": 0.8143429569505805,
+      "grad_norm": 0.11063557863235474,
+      "learning_rate": 0.01,
+      "loss": 2.1131,
+      "step": 7926
+    },
+    {
+      "epoch": 0.8146511866844754,
+      "grad_norm": 0.10625256597995758,
+      "learning_rate": 0.01,
+      "loss": 2.113,
+      "step": 7929
+    },
+    {
+      "epoch": 0.8149594164183704,
+      "grad_norm": 0.0682268813252449,
+      "learning_rate": 0.01,
+      "loss": 2.0929,
+      "step": 7932
+    },
+    {
+      "epoch": 0.8152676461522654,
+      "grad_norm": 0.08721883594989777,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 7935
+    },
+    {
+      "epoch": 0.8155758758861605,
+      "grad_norm": 0.07372716814279556,
+      "learning_rate": 0.01,
+      "loss": 2.1173,
+      "step": 7938
+    },
+    {
+      "epoch": 0.8158841056200555,
+      "grad_norm": 0.049299102276563644,
+      "learning_rate": 0.01,
+      "loss": 2.1172,
+      "step": 7941
+    },
+    {
+      "epoch": 0.8161923353539505,
+      "grad_norm": 0.06552339345216751,
+      "learning_rate": 0.01,
+      "loss": 2.1035,
+      "step": 7944
+    },
+    {
+      "epoch": 0.8165005650878455,
+      "grad_norm": 0.08362871408462524,
+      "learning_rate": 0.01,
+      "loss": 2.0942,
+      "step": 7947
+    },
+    {
+      "epoch": 0.8168087948217405,
+      "grad_norm": 0.07610680162906647,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 7950
+    },
+    {
+      "epoch": 0.8171170245556355,
+      "grad_norm": 0.058830149471759796,
+      "learning_rate": 0.01,
+      "loss": 2.121,
+      "step": 7953
+    },
+    {
+      "epoch": 0.8174252542895305,
+      "grad_norm": 0.10281010717153549,
+      "learning_rate": 0.01,
+      "loss": 2.1084,
+      "step": 7956
+    },
+    {
+      "epoch": 0.8177334840234255,
+      "grad_norm": 0.04509102553129196,
+      "learning_rate": 0.01,
+      "loss": 2.0917,
+      "step": 7959
+    },
+    {
+      "epoch": 0.8180417137573205,
+      "grad_norm": 0.034059979021549225,
+      "learning_rate": 0.01,
+      "loss": 2.1286,
+      "step": 7962
+    },
+    {
+      "epoch": 0.8183499434912155,
+      "grad_norm": 0.09370562434196472,
+      "learning_rate": 0.01,
+      "loss": 2.1298,
+      "step": 7965
+    },
+    {
+      "epoch": 0.8186581732251105,
+      "grad_norm": 0.09386254847049713,
+      "learning_rate": 0.01,
+      "loss": 2.1018,
+      "step": 7968
+    },
+    {
+      "epoch": 0.8189664029590055,
+      "grad_norm": 0.1801362931728363,
+      "learning_rate": 0.01,
+      "loss": 2.1125,
+      "step": 7971
+    },
+    {
+      "epoch": 0.8192746326929005,
+      "grad_norm": 0.12590090930461884,
+      "learning_rate": 0.01,
+      "loss": 2.1145,
+      "step": 7974
+    },
+    {
+      "epoch": 0.8195828624267955,
+      "grad_norm": 0.09913074970245361,
+      "learning_rate": 0.01,
+      "loss": 2.1049,
+      "step": 7977
+    },
+    {
+      "epoch": 0.8198910921606904,
+      "grad_norm": 0.05249069631099701,
+      "learning_rate": 0.01,
+      "loss": 2.1348,
+      "step": 7980
+    },
+    {
+      "epoch": 0.8201993218945854,
+      "grad_norm": 0.05334639549255371,
+      "learning_rate": 0.01,
+      "loss": 2.0952,
+      "step": 7983
+    },
+    {
+      "epoch": 0.8205075516284804,
+      "grad_norm": 0.03963373601436615,
+      "learning_rate": 0.01,
+      "loss": 2.1133,
+      "step": 7986
+    },
+    {
+      "epoch": 0.8208157813623754,
+      "grad_norm": 0.03334924206137657,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 7989
+    },
+    {
+      "epoch": 0.8211240110962704,
+      "grad_norm": 0.0628419816493988,
+      "learning_rate": 0.01,
+      "loss": 2.1298,
+      "step": 7992
+    },
+    {
+      "epoch": 0.8214322408301654,
+      "grad_norm": 0.07143758237361908,
+      "learning_rate": 0.01,
+      "loss": 2.0891,
+      "step": 7995
+    },
+    {
+      "epoch": 0.8217404705640604,
+      "grad_norm": 0.06662650406360626,
+      "learning_rate": 0.01,
+      "loss": 2.0976,
+      "step": 7998
+    },
+    {
+      "epoch": 0.8220487002979554,
+      "grad_norm": 0.10575726628303528,
+      "learning_rate": 0.01,
+      "loss": 2.0946,
+      "step": 8001
+    },
+    {
+      "epoch": 0.8223569300318504,
+      "grad_norm": 0.056455157697200775,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 8004
+    },
+    {
+      "epoch": 0.8226651597657454,
+      "grad_norm": 0.10326797515153885,
+      "learning_rate": 0.01,
+      "loss": 2.0823,
+      "step": 8007
+    },
+    {
+      "epoch": 0.8229733894996404,
+      "grad_norm": 0.08464314043521881,
+      "learning_rate": 0.01,
+      "loss": 2.1274,
+      "step": 8010
+    },
+    {
+      "epoch": 0.8232816192335354,
+      "grad_norm": 0.052144117653369904,
+      "learning_rate": 0.01,
+      "loss": 2.0952,
+      "step": 8013
+    },
+    {
+      "epoch": 0.8235898489674304,
+      "grad_norm": 0.05464213341474533,
+      "learning_rate": 0.01,
+      "loss": 2.1117,
+      "step": 8016
+    },
+    {
+      "epoch": 0.8238980787013254,
+      "grad_norm": 0.06700276583433151,
+      "learning_rate": 0.01,
+      "loss": 2.1289,
+      "step": 8019
+    },
+    {
+      "epoch": 0.8242063084352204,
+      "grad_norm": 0.05322539806365967,
+      "learning_rate": 0.01,
+      "loss": 2.1346,
+      "step": 8022
+    },
+    {
+      "epoch": 0.8245145381691154,
+      "grad_norm": 0.040953267365694046,
+      "learning_rate": 0.01,
+      "loss": 2.1035,
+      "step": 8025
+    },
+    {
+      "epoch": 0.8248227679030103,
+      "grad_norm": 0.043644580990076065,
+      "learning_rate": 0.01,
+      "loss": 2.1238,
+      "step": 8028
+    },
+    {
+      "epoch": 0.8251309976369053,
+      "grad_norm": 0.060951683670282364,
+      "learning_rate": 0.01,
+      "loss": 2.1169,
+      "step": 8031
+    },
+    {
+      "epoch": 0.8254392273708003,
+      "grad_norm": 0.11269400268793106,
+      "learning_rate": 0.01,
+      "loss": 2.1184,
+      "step": 8034
+    },
+    {
+      "epoch": 0.8257474571046953,
+      "grad_norm": 0.05554080754518509,
+      "learning_rate": 0.01,
+      "loss": 2.112,
+      "step": 8037
+    },
+    {
+      "epoch": 0.8260556868385903,
+      "grad_norm": 0.08155755698680878,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 8040
+    },
+    {
+      "epoch": 0.8263639165724853,
+      "grad_norm": 0.0661015510559082,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 8043
+    },
+    {
+      "epoch": 0.8266721463063803,
+      "grad_norm": 0.06186169385910034,
+      "learning_rate": 0.01,
+      "loss": 2.1225,
+      "step": 8046
+    },
+    {
+      "epoch": 0.8269803760402753,
+      "grad_norm": 0.06658541411161423,
+      "learning_rate": 0.01,
+      "loss": 2.1189,
+      "step": 8049
+    },
+    {
+      "epoch": 0.8272886057741703,
+      "grad_norm": 0.09120085090398788,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 8052
+    },
+    {
+      "epoch": 0.8275968355080654,
+      "grad_norm": 0.059662993997335434,
+      "learning_rate": 0.01,
+      "loss": 2.1257,
+      "step": 8055
+    },
+    {
+      "epoch": 0.8279050652419604,
+      "grad_norm": 0.08305416256189346,
+      "learning_rate": 0.01,
+      "loss": 2.1108,
+      "step": 8058
+    },
+    {
+      "epoch": 0.8282132949758554,
+      "grad_norm": 0.047110967338085175,
+      "learning_rate": 0.01,
+      "loss": 2.0786,
+      "step": 8061
+    },
+    {
+      "epoch": 0.8285215247097504,
+      "grad_norm": 0.042120445519685745,
+      "learning_rate": 0.01,
+      "loss": 2.1081,
+      "step": 8064
+    },
+    {
+      "epoch": 0.8288297544436454,
+      "grad_norm": 0.04596862941980362,
+      "learning_rate": 0.01,
+      "loss": 2.1025,
+      "step": 8067
+    },
+    {
+      "epoch": 0.8291379841775404,
+      "grad_norm": 0.055216096341609955,
+      "learning_rate": 0.01,
+      "loss": 2.1416,
+      "step": 8070
+    },
+    {
+      "epoch": 0.8294462139114354,
+      "grad_norm": 0.05959683656692505,
+      "learning_rate": 0.01,
+      "loss": 2.0832,
+      "step": 8073
+    },
+    {
+      "epoch": 0.8297544436453304,
+      "grad_norm": 0.045481909066438675,
+      "learning_rate": 0.01,
+      "loss": 2.1391,
+      "step": 8076
+    },
+    {
+      "epoch": 0.8300626733792253,
+      "grad_norm": 0.049282100051641464,
+      "learning_rate": 0.01,
+      "loss": 2.1227,
+      "step": 8079
+    },
+    {
+      "epoch": 0.8303709031131203,
+      "grad_norm": 0.058084890246391296,
+      "learning_rate": 0.01,
+      "loss": 2.1211,
+      "step": 8082
+    },
+    {
+      "epoch": 0.8306791328470153,
+      "grad_norm": 0.11113768070936203,
+      "learning_rate": 0.01,
+      "loss": 2.132,
+      "step": 8085
+    },
+    {
+      "epoch": 0.8309873625809103,
+      "grad_norm": 0.07015852630138397,
+      "learning_rate": 0.01,
+      "loss": 2.0988,
+      "step": 8088
+    },
+    {
+      "epoch": 0.8312955923148053,
+      "grad_norm": 0.09509722143411636,
+      "learning_rate": 0.01,
+      "loss": 2.1064,
+      "step": 8091
+    },
+    {
+      "epoch": 0.8316038220487003,
+      "grad_norm": 0.03616593778133392,
+      "learning_rate": 0.01,
+      "loss": 2.1078,
+      "step": 8094
+    },
+    {
+      "epoch": 0.8319120517825953,
+      "grad_norm": 0.0486396960914135,
+      "learning_rate": 0.01,
+      "loss": 2.1279,
+      "step": 8097
+    },
+    {
+      "epoch": 0.8322202815164903,
+      "grad_norm": 0.050647489726543427,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 8100
+    },
+    {
+      "epoch": 0.8325285112503853,
+      "grad_norm": 0.08125802874565125,
+      "learning_rate": 0.01,
+      "loss": 2.1099,
+      "step": 8103
+    },
+    {
+      "epoch": 0.8328367409842803,
+      "grad_norm": 0.08078313618898392,
+      "learning_rate": 0.01,
+      "loss": 2.1162,
+      "step": 8106
+    },
+    {
+      "epoch": 0.8331449707181753,
+      "grad_norm": 0.06157573312520981,
+      "learning_rate": 0.01,
+      "loss": 2.1288,
+      "step": 8109
+    },
+    {
+      "epoch": 0.8334532004520703,
+      "grad_norm": 0.057771824300289154,
+      "learning_rate": 0.01,
+      "loss": 2.1136,
+      "step": 8112
+    },
+    {
+      "epoch": 0.8337614301859653,
+      "grad_norm": 0.06634260714054108,
+      "learning_rate": 0.01,
+      "loss": 2.1029,
+      "step": 8115
+    },
+    {
+      "epoch": 0.8340696599198603,
+      "grad_norm": 0.12186034023761749,
+      "learning_rate": 0.01,
+      "loss": 2.1236,
+      "step": 8118
+    },
+    {
+      "epoch": 0.8343778896537553,
+      "grad_norm": 0.03940106928348541,
+      "learning_rate": 0.01,
+      "loss": 2.1169,
+      "step": 8121
+    },
+    {
+      "epoch": 0.8346861193876502,
+      "grad_norm": 0.06003478914499283,
+      "learning_rate": 0.01,
+      "loss": 2.1108,
+      "step": 8124
+    },
+    {
+      "epoch": 0.8349943491215452,
+      "grad_norm": 0.04963524639606476,
+      "learning_rate": 0.01,
+      "loss": 2.0893,
+      "step": 8127
+    },
+    {
+      "epoch": 0.8353025788554402,
+      "grad_norm": 0.04543556645512581,
+      "learning_rate": 0.01,
+      "loss": 2.1062,
+      "step": 8130
+    },
+    {
+      "epoch": 0.8356108085893352,
+      "grad_norm": 0.05210501328110695,
+      "learning_rate": 0.01,
+      "loss": 2.1065,
+      "step": 8133
+    },
+    {
+      "epoch": 0.8359190383232302,
+      "grad_norm": 0.10866094380617142,
+      "learning_rate": 0.01,
+      "loss": 2.1154,
+      "step": 8136
+    },
+    {
+      "epoch": 0.8362272680571252,
+      "grad_norm": 0.07595928758382797,
+      "learning_rate": 0.01,
+      "loss": 2.1079,
+      "step": 8139
+    },
+    {
+      "epoch": 0.8365354977910202,
+      "grad_norm": 0.04948664829134941,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 8142
+    },
+    {
+      "epoch": 0.8368437275249152,
+      "grad_norm": 0.11719872057437897,
+      "learning_rate": 0.01,
+      "loss": 2.1106,
+      "step": 8145
+    },
+    {
+      "epoch": 0.8371519572588102,
+      "grad_norm": 0.04469067603349686,
+      "learning_rate": 0.01,
+      "loss": 2.1145,
+      "step": 8148
+    },
+    {
+      "epoch": 0.8374601869927052,
+      "grad_norm": 0.038385387510061264,
+      "learning_rate": 0.01,
+      "loss": 2.0933,
+      "step": 8151
+    },
+    {
+      "epoch": 0.8377684167266002,
+      "grad_norm": 0.04040665924549103,
+      "learning_rate": 0.01,
+      "loss": 2.1119,
+      "step": 8154
+    },
+    {
+      "epoch": 0.8380766464604952,
+      "grad_norm": 0.042900413274765015,
+      "learning_rate": 0.01,
+      "loss": 2.1053,
+      "step": 8157
+    },
+    {
+      "epoch": 0.8383848761943902,
+      "grad_norm": 0.06709393113851547,
+      "learning_rate": 0.01,
+      "loss": 2.0858,
+      "step": 8160
+    },
+    {
+      "epoch": 0.8386931059282852,
+      "grad_norm": 0.08388926833868027,
+      "learning_rate": 0.01,
+      "loss": 2.1287,
+      "step": 8163
+    },
+    {
+      "epoch": 0.8390013356621802,
+      "grad_norm": 0.0701015368103981,
+      "learning_rate": 0.01,
+      "loss": 2.0965,
+      "step": 8166
+    },
+    {
+      "epoch": 0.8393095653960753,
+      "grad_norm": 0.0841115415096283,
+      "learning_rate": 0.01,
+      "loss": 2.1136,
+      "step": 8169
+    },
+    {
+      "epoch": 0.8396177951299703,
+      "grad_norm": 0.08542285114526749,
+      "learning_rate": 0.01,
+      "loss": 2.1166,
+      "step": 8172
+    },
+    {
+      "epoch": 0.8399260248638653,
+      "grad_norm": 0.046626705676317215,
+      "learning_rate": 0.01,
+      "loss": 2.1318,
+      "step": 8175
+    },
+    {
+      "epoch": 0.8402342545977602,
+      "grad_norm": 0.08752947300672531,
+      "learning_rate": 0.01,
+      "loss": 2.0873,
+      "step": 8178
+    },
+    {
+      "epoch": 0.8405424843316552,
+      "grad_norm": 0.04728331416845322,
+      "learning_rate": 0.01,
+      "loss": 2.0951,
+      "step": 8181
+    },
+    {
+      "epoch": 0.8408507140655502,
+      "grad_norm": 0.04881293699145317,
+      "learning_rate": 0.01,
+      "loss": 2.1011,
+      "step": 8184
+    },
+    {
+      "epoch": 0.8411589437994452,
+      "grad_norm": 0.049758728593587875,
+      "learning_rate": 0.01,
+      "loss": 2.1206,
+      "step": 8187
+    },
+    {
+      "epoch": 0.8414671735333402,
+      "grad_norm": 0.037589117884635925,
+      "learning_rate": 0.01,
+      "loss": 2.085,
+      "step": 8190
+    },
+    {
+      "epoch": 0.8417754032672352,
+      "grad_norm": 0.11785265803337097,
+      "learning_rate": 0.01,
+      "loss": 2.1195,
+      "step": 8193
+    },
+    {
+      "epoch": 0.8420836330011302,
+      "grad_norm": 0.09815037995576859,
+      "learning_rate": 0.01,
+      "loss": 2.065,
+      "step": 8196
+    },
+    {
+      "epoch": 0.8423918627350252,
+      "grad_norm": 0.07950727641582489,
+      "learning_rate": 0.01,
+      "loss": 2.1081,
+      "step": 8199
+    },
+    {
+      "epoch": 0.8427000924689202,
+      "grad_norm": 0.04057254642248154,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 8202
+    },
+    {
+      "epoch": 0.8430083222028152,
+      "grad_norm": 0.07260222733020782,
+      "learning_rate": 0.01,
+      "loss": 2.1018,
+      "step": 8205
+    },
+    {
+      "epoch": 0.8433165519367102,
+      "grad_norm": 0.057693734765052795,
+      "learning_rate": 0.01,
+      "loss": 2.112,
+      "step": 8208
+    },
+    {
+      "epoch": 0.8436247816706052,
+      "grad_norm": 0.039680637419223785,
+      "learning_rate": 0.01,
+      "loss": 2.0974,
+      "step": 8211
+    },
+    {
+      "epoch": 0.8439330114045002,
+      "grad_norm": 0.07584577798843384,
+      "learning_rate": 0.01,
+      "loss": 2.129,
+      "step": 8214
+    },
+    {
+      "epoch": 0.8442412411383952,
+      "grad_norm": 0.044016819447278976,
+      "learning_rate": 0.01,
+      "loss": 2.1139,
+      "step": 8217
+    },
+    {
+      "epoch": 0.8445494708722902,
+      "grad_norm": 0.04449582099914551,
+      "learning_rate": 0.01,
+      "loss": 2.1085,
+      "step": 8220
+    },
+    {
+      "epoch": 0.8448577006061851,
+      "grad_norm": 0.044676005840301514,
+      "learning_rate": 0.01,
+      "loss": 2.1081,
+      "step": 8223
+    },
+    {
+      "epoch": 0.8451659303400801,
+      "grad_norm": 0.04926025867462158,
+      "learning_rate": 0.01,
+      "loss": 2.1187,
+      "step": 8226
+    },
+    {
+      "epoch": 0.8454741600739751,
+      "grad_norm": 0.10964366793632507,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 8229
+    },
+    {
+      "epoch": 0.8457823898078701,
+      "grad_norm": 0.09405852109193802,
+      "learning_rate": 0.01,
+      "loss": 2.1117,
+      "step": 8232
+    },
+    {
+      "epoch": 0.8460906195417651,
+      "grad_norm": 0.09241268038749695,
+      "learning_rate": 0.01,
+      "loss": 2.1225,
+      "step": 8235
+    },
+    {
+      "epoch": 0.8463988492756601,
+      "grad_norm": 0.05887102335691452,
+      "learning_rate": 0.01,
+      "loss": 2.0944,
+      "step": 8238
+    },
+    {
+      "epoch": 0.8467070790095551,
+      "grad_norm": 0.13513131439685822,
+      "learning_rate": 0.01,
+      "loss": 2.1217,
+      "step": 8241
+    },
+    {
+      "epoch": 0.8470153087434501,
+      "grad_norm": 0.06370443850755692,
+      "learning_rate": 0.01,
+      "loss": 2.121,
+      "step": 8244
+    },
+    {
+      "epoch": 0.8473235384773451,
+      "grad_norm": 0.0426030196249485,
+      "learning_rate": 0.01,
+      "loss": 2.0937,
+      "step": 8247
+    },
+    {
+      "epoch": 0.8476317682112401,
+      "grad_norm": 0.049412764608860016,
+      "learning_rate": 0.01,
+      "loss": 2.108,
+      "step": 8250
+    },
+    {
+      "epoch": 0.8479399979451351,
+      "grad_norm": 0.061407607048749924,
+      "learning_rate": 0.01,
+      "loss": 2.1009,
+      "step": 8253
+    },
+    {
+      "epoch": 0.8482482276790301,
+      "grad_norm": 0.12416908144950867,
+      "learning_rate": 0.01,
+      "loss": 2.0819,
+      "step": 8256
+    },
+    {
+      "epoch": 0.8485564574129251,
+      "grad_norm": 0.05728744715452194,
+      "learning_rate": 0.01,
+      "loss": 2.1132,
+      "step": 8259
+    },
+    {
+      "epoch": 0.8488646871468201,
+      "grad_norm": 0.06724981963634491,
+      "learning_rate": 0.01,
+      "loss": 2.0932,
+      "step": 8262
+    },
+    {
+      "epoch": 0.8491729168807151,
+      "grad_norm": 0.055260930210351944,
+      "learning_rate": 0.01,
+      "loss": 2.0933,
+      "step": 8265
+    },
+    {
+      "epoch": 0.84948114661461,
+      "grad_norm": 0.04230106249451637,
+      "learning_rate": 0.01,
+      "loss": 2.1135,
+      "step": 8268
+    },
+    {
+      "epoch": 0.849789376348505,
+      "grad_norm": 0.04593104496598244,
+      "learning_rate": 0.01,
+      "loss": 2.1091,
+      "step": 8271
+    },
+    {
+      "epoch": 0.8500976060824,
+      "grad_norm": 0.09625285863876343,
+      "learning_rate": 0.01,
+      "loss": 2.0946,
+      "step": 8274
+    },
+    {
+      "epoch": 0.850405835816295,
+      "grad_norm": 0.04556501284241676,
+      "learning_rate": 0.01,
+      "loss": 2.0975,
+      "step": 8277
+    },
+    {
+      "epoch": 0.85071406555019,
+      "grad_norm": 0.09413543343544006,
+      "learning_rate": 0.01,
+      "loss": 2.0863,
+      "step": 8280
+    },
+    {
+      "epoch": 0.851022295284085,
+      "grad_norm": 0.08400101214647293,
+      "learning_rate": 0.01,
+      "loss": 2.1045,
+      "step": 8283
+    },
+    {
+      "epoch": 0.8513305250179801,
+      "grad_norm": 0.06278138607740402,
+      "learning_rate": 0.01,
+      "loss": 2.116,
+      "step": 8286
+    },
+    {
+      "epoch": 0.8516387547518751,
+      "grad_norm": 0.04442959651350975,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 8289
+    },
+    {
+      "epoch": 0.8519469844857701,
+      "grad_norm": 0.045807912945747375,
+      "learning_rate": 0.01,
+      "loss": 2.0823,
+      "step": 8292
+    },
+    {
+      "epoch": 0.8522552142196651,
+      "grad_norm": 0.0426551029086113,
+      "learning_rate": 0.01,
+      "loss": 2.1109,
+      "step": 8295
+    },
+    {
+      "epoch": 0.8525634439535601,
+      "grad_norm": 0.12200357019901276,
+      "learning_rate": 0.01,
+      "loss": 2.1146,
+      "step": 8298
+    },
+    {
+      "epoch": 0.8528716736874551,
+      "grad_norm": 0.04152747616171837,
+      "learning_rate": 0.01,
+      "loss": 2.1204,
+      "step": 8301
+    },
+    {
+      "epoch": 0.8531799034213501,
+      "grad_norm": 0.08464021235704422,
+      "learning_rate": 0.01,
+      "loss": 2.085,
+      "step": 8304
+    },
+    {
+      "epoch": 0.8534881331552451,
+      "grad_norm": 0.050391390919685364,
+      "learning_rate": 0.01,
+      "loss": 2.0957,
+      "step": 8307
+    },
+    {
+      "epoch": 0.8537963628891401,
+      "grad_norm": 0.08581732958555222,
+      "learning_rate": 0.01,
+      "loss": 2.1015,
+      "step": 8310
+    },
+    {
+      "epoch": 0.8541045926230351,
+      "grad_norm": 0.10992308706045151,
+      "learning_rate": 0.01,
+      "loss": 2.0939,
+      "step": 8313
+    },
+    {
+      "epoch": 0.8544128223569301,
+      "grad_norm": 0.053225912153720856,
+      "learning_rate": 0.01,
+      "loss": 2.1106,
+      "step": 8316
+    },
+    {
+      "epoch": 0.854721052090825,
+      "grad_norm": 0.06759096682071686,
+      "learning_rate": 0.01,
+      "loss": 2.1035,
+      "step": 8319
+    },
+    {
+      "epoch": 0.85502928182472,
+      "grad_norm": 0.058069922029972076,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 8322
+    },
+    {
+      "epoch": 0.855337511558615,
+      "grad_norm": 0.0657680481672287,
+      "learning_rate": 0.01,
+      "loss": 2.0918,
+      "step": 8325
+    },
+    {
+      "epoch": 0.85564574129251,
+      "grad_norm": 0.1428556591272354,
+      "learning_rate": 0.01,
+      "loss": 2.0895,
+      "step": 8328
+    },
+    {
+      "epoch": 0.855953971026405,
+      "grad_norm": 0.10311869531869888,
+      "learning_rate": 0.01,
+      "loss": 2.1041,
+      "step": 8331
+    },
+    {
+      "epoch": 0.8562622007603,
+      "grad_norm": 0.12024179100990295,
+      "learning_rate": 0.01,
+      "loss": 2.1179,
+      "step": 8334
+    },
+    {
+      "epoch": 0.856570430494195,
+      "grad_norm": 0.08294446766376495,
+      "learning_rate": 0.01,
+      "loss": 2.1042,
+      "step": 8337
+    },
+    {
+      "epoch": 0.85687866022809,
+      "grad_norm": 0.05203935503959656,
+      "learning_rate": 0.01,
+      "loss": 2.1275,
+      "step": 8340
+    },
+    {
+      "epoch": 0.857186889961985,
+      "grad_norm": 0.061564356088638306,
+      "learning_rate": 0.01,
+      "loss": 2.0951,
+      "step": 8343
+    },
+    {
+      "epoch": 0.85749511969588,
+      "grad_norm": 0.043616339564323425,
+      "learning_rate": 0.01,
+      "loss": 2.1082,
+      "step": 8346
+    },
+    {
+      "epoch": 0.857803349429775,
+      "grad_norm": 0.0885004997253418,
+      "learning_rate": 0.01,
+      "loss": 2.1071,
+      "step": 8349
+    },
+    {
+      "epoch": 0.85811157916367,
+      "grad_norm": 0.06275481730699539,
+      "learning_rate": 0.01,
+      "loss": 2.137,
+      "step": 8352
+    },
+    {
+      "epoch": 0.858419808897565,
+      "grad_norm": 0.054776523262262344,
+      "learning_rate": 0.01,
+      "loss": 2.1117,
+      "step": 8355
+    },
+    {
+      "epoch": 0.85872803863146,
+      "grad_norm": 0.07782801240682602,
+      "learning_rate": 0.01,
+      "loss": 2.0822,
+      "step": 8358
+    },
+    {
+      "epoch": 0.859036268365355,
+      "grad_norm": 0.12301263958215714,
+      "learning_rate": 0.01,
+      "loss": 2.1126,
+      "step": 8361
+    },
+    {
+      "epoch": 0.85934449809925,
+      "grad_norm": 0.07181745767593384,
+      "learning_rate": 0.01,
+      "loss": 2.1359,
+      "step": 8364
+    },
+    {
+      "epoch": 0.859652727833145,
+      "grad_norm": 0.07232604175806046,
+      "learning_rate": 0.01,
+      "loss": 2.0849,
+      "step": 8367
+    },
+    {
+      "epoch": 0.8599609575670399,
+      "grad_norm": 0.06810937821865082,
+      "learning_rate": 0.01,
+      "loss": 2.1088,
+      "step": 8370
+    },
+    {
+      "epoch": 0.8602691873009349,
+      "grad_norm": 0.048163384199142456,
+      "learning_rate": 0.01,
+      "loss": 2.1173,
+      "step": 8373
+    },
+    {
+      "epoch": 0.8605774170348299,
+      "grad_norm": 0.05688156560063362,
+      "learning_rate": 0.01,
+      "loss": 2.114,
+      "step": 8376
+    },
+    {
+      "epoch": 0.8608856467687249,
+      "grad_norm": 0.065540611743927,
+      "learning_rate": 0.01,
+      "loss": 2.0989,
+      "step": 8379
+    },
+    {
+      "epoch": 0.8611938765026199,
+      "grad_norm": 0.09561596065759659,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 8382
+    },
+    {
+      "epoch": 0.8615021062365149,
+      "grad_norm": 0.06719313561916351,
+      "learning_rate": 0.01,
+      "loss": 2.0928,
+      "step": 8385
+    },
+    {
+      "epoch": 0.8618103359704099,
+      "grad_norm": 0.05895761027932167,
+      "learning_rate": 0.01,
+      "loss": 2.1037,
+      "step": 8388
+    },
+    {
+      "epoch": 0.8621185657043049,
+      "grad_norm": 0.09232669323682785,
+      "learning_rate": 0.01,
+      "loss": 2.1272,
+      "step": 8391
+    },
+    {
+      "epoch": 0.8624267954381999,
+      "grad_norm": 0.06715840846300125,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 8394
+    },
+    {
+      "epoch": 0.8627350251720949,
+      "grad_norm": 0.04794420674443245,
+      "learning_rate": 0.01,
+      "loss": 2.1087,
+      "step": 8397
+    },
+    {
+      "epoch": 0.8630432549059899,
+      "grad_norm": 0.037383124232292175,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 8400
+    },
+    {
+      "epoch": 0.863351484639885,
+      "grad_norm": 0.05601905286312103,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 8403
+    },
+    {
+      "epoch": 0.86365971437378,
+      "grad_norm": 0.0839313194155693,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 8406
+    },
+    {
+      "epoch": 0.863967944107675,
+      "grad_norm": 0.07600929588079453,
+      "learning_rate": 0.01,
+      "loss": 2.1143,
+      "step": 8409
+    },
+    {
+      "epoch": 0.86427617384157,
+      "grad_norm": 0.06851659715175629,
+      "learning_rate": 0.01,
+      "loss": 2.0921,
+      "step": 8412
+    },
+    {
+      "epoch": 0.864584403575465,
+      "grad_norm": 0.05021858587861061,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 8415
+    },
+    {
+      "epoch": 0.86489263330936,
+      "grad_norm": 0.04881426692008972,
+      "learning_rate": 0.01,
+      "loss": 2.1047,
+      "step": 8418
+    },
+    {
+      "epoch": 0.8652008630432549,
+      "grad_norm": 0.04262546822428703,
+      "learning_rate": 0.01,
+      "loss": 2.0852,
+      "step": 8421
+    },
+    {
+      "epoch": 0.8655090927771499,
+      "grad_norm": 0.050467535853385925,
+      "learning_rate": 0.01,
+      "loss": 2.1045,
+      "step": 8424
+    },
+    {
+      "epoch": 0.8658173225110449,
+      "grad_norm": 0.0725008100271225,
+      "learning_rate": 0.01,
+      "loss": 2.077,
+      "step": 8427
+    },
+    {
+      "epoch": 0.8661255522449399,
+      "grad_norm": 0.07234456390142441,
+      "learning_rate": 0.01,
+      "loss": 2.13,
+      "step": 8430
+    },
+    {
+      "epoch": 0.8664337819788349,
+      "grad_norm": 0.060751501470804214,
+      "learning_rate": 0.01,
+      "loss": 2.0948,
+      "step": 8433
+    },
+    {
+      "epoch": 0.8667420117127299,
+      "grad_norm": 0.058911584317684174,
+      "learning_rate": 0.01,
+      "loss": 2.0908,
+      "step": 8436
+    },
+    {
+      "epoch": 0.8670502414466249,
+      "grad_norm": 0.08380532264709473,
+      "learning_rate": 0.01,
+      "loss": 2.1438,
+      "step": 8439
+    },
+    {
+      "epoch": 0.8673584711805199,
+      "grad_norm": 0.058240536600351334,
+      "learning_rate": 0.01,
+      "loss": 2.1384,
+      "step": 8442
+    },
+    {
+      "epoch": 0.8676667009144149,
+      "grad_norm": 0.0422792062163353,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 8445
+    },
+    {
+      "epoch": 0.8679749306483099,
+      "grad_norm": 0.07096652686595917,
+      "learning_rate": 0.01,
+      "loss": 2.1195,
+      "step": 8448
+    },
+    {
+      "epoch": 0.8682831603822049,
+      "grad_norm": 0.13370642066001892,
+      "learning_rate": 0.01,
+      "loss": 2.1367,
+      "step": 8451
+    },
+    {
+      "epoch": 0.8685913901160999,
+      "grad_norm": 0.0597628615796566,
+      "learning_rate": 0.01,
+      "loss": 2.087,
+      "step": 8454
+    },
+    {
+      "epoch": 0.8688996198499949,
+      "grad_norm": 0.039561979472637177,
+      "learning_rate": 0.01,
+      "loss": 2.112,
+      "step": 8457
+    },
+    {
+      "epoch": 0.8692078495838899,
+      "grad_norm": 0.04080485925078392,
+      "learning_rate": 0.01,
+      "loss": 2.1024,
+      "step": 8460
+    },
+    {
+      "epoch": 0.8695160793177849,
+      "grad_norm": 0.05293022468686104,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 8463
+    },
+    {
+      "epoch": 0.8698243090516798,
+      "grad_norm": 0.06960830092430115,
+      "learning_rate": 0.01,
+      "loss": 2.1255,
+      "step": 8466
+    },
+    {
+      "epoch": 0.8701325387855748,
+      "grad_norm": 0.09768849611282349,
+      "learning_rate": 0.01,
+      "loss": 2.1217,
+      "step": 8469
+    },
+    {
+      "epoch": 0.8704407685194698,
+      "grad_norm": 0.11970885097980499,
+      "learning_rate": 0.01,
+      "loss": 2.0932,
+      "step": 8472
+    },
+    {
+      "epoch": 0.8707489982533648,
+      "grad_norm": 0.12014521658420563,
+      "learning_rate": 0.01,
+      "loss": 2.1009,
+      "step": 8475
+    },
+    {
+      "epoch": 0.8710572279872598,
+      "grad_norm": 0.04288540408015251,
+      "learning_rate": 0.01,
+      "loss": 2.1111,
+      "step": 8478
+    },
+    {
+      "epoch": 0.8713654577211548,
+      "grad_norm": 0.033004507422447205,
+      "learning_rate": 0.01,
+      "loss": 2.1029,
+      "step": 8481
+    },
+    {
+      "epoch": 0.8716736874550498,
+      "grad_norm": 0.03685779869556427,
+      "learning_rate": 0.01,
+      "loss": 2.1077,
+      "step": 8484
+    },
+    {
+      "epoch": 0.8719819171889448,
+      "grad_norm": 0.06450948119163513,
+      "learning_rate": 0.01,
+      "loss": 2.102,
+      "step": 8487
+    },
+    {
+      "epoch": 0.8722901469228398,
+      "grad_norm": 0.04806706681847572,
+      "learning_rate": 0.01,
+      "loss": 2.1056,
+      "step": 8490
+    },
+    {
+      "epoch": 0.8725983766567348,
+      "grad_norm": 0.05847964435815811,
+      "learning_rate": 0.01,
+      "loss": 2.095,
+      "step": 8493
+    },
+    {
+      "epoch": 0.8729066063906298,
+      "grad_norm": 0.11569567024707794,
+      "learning_rate": 0.01,
+      "loss": 2.1058,
+      "step": 8496
+    },
+    {
+      "epoch": 0.8732148361245248,
+      "grad_norm": 0.04440119490027428,
+      "learning_rate": 0.01,
+      "loss": 2.1127,
+      "step": 8499
+    },
+    {
+      "epoch": 0.8735230658584198,
+      "grad_norm": 0.13856938481330872,
+      "learning_rate": 0.01,
+      "loss": 2.1072,
+      "step": 8502
+    },
+    {
+      "epoch": 0.8738312955923148,
+      "grad_norm": 0.06448937207460403,
+      "learning_rate": 0.01,
+      "loss": 2.0813,
+      "step": 8505
+    },
+    {
+      "epoch": 0.8741395253262098,
+      "grad_norm": 0.05872811749577522,
+      "learning_rate": 0.01,
+      "loss": 2.1227,
+      "step": 8508
+    },
+    {
+      "epoch": 0.8744477550601047,
+      "grad_norm": 0.06387540698051453,
+      "learning_rate": 0.01,
+      "loss": 2.099,
+      "step": 8511
+    },
+    {
+      "epoch": 0.8747559847939997,
+      "grad_norm": 0.044399481266736984,
+      "learning_rate": 0.01,
+      "loss": 2.0989,
+      "step": 8514
+    },
+    {
+      "epoch": 0.8750642145278948,
+      "grad_norm": 0.118850938975811,
+      "learning_rate": 0.01,
+      "loss": 2.1261,
+      "step": 8517
+    },
+    {
+      "epoch": 0.8753724442617898,
+      "grad_norm": 0.05479248985648155,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 8520
+    },
+    {
+      "epoch": 0.8756806739956848,
+      "grad_norm": 0.06442543119192123,
+      "learning_rate": 0.01,
+      "loss": 2.0844,
+      "step": 8523
+    },
+    {
+      "epoch": 0.8759889037295798,
+      "grad_norm": 0.054294027388095856,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 8526
+    },
+    {
+      "epoch": 0.8762971334634748,
+      "grad_norm": 0.04776893928647041,
+      "learning_rate": 0.01,
+      "loss": 2.1056,
+      "step": 8529
+    },
+    {
+      "epoch": 0.8766053631973698,
+      "grad_norm": 0.06740310043096542,
+      "learning_rate": 0.01,
+      "loss": 2.0956,
+      "step": 8532
+    },
+    {
+      "epoch": 0.8769135929312648,
+      "grad_norm": 0.048034511506557465,
+      "learning_rate": 0.01,
+      "loss": 2.1223,
+      "step": 8535
+    },
+    {
+      "epoch": 0.8772218226651598,
+      "grad_norm": 0.05819391459226608,
+      "learning_rate": 0.01,
+      "loss": 2.1133,
+      "step": 8538
+    },
+    {
+      "epoch": 0.8775300523990548,
+      "grad_norm": 0.06093437224626541,
+      "learning_rate": 0.01,
+      "loss": 2.0889,
+      "step": 8541
+    },
+    {
+      "epoch": 0.8778382821329498,
+      "grad_norm": 0.04628787562251091,
+      "learning_rate": 0.01,
+      "loss": 2.1202,
+      "step": 8544
+    },
+    {
+      "epoch": 0.8781465118668448,
+      "grad_norm": 0.0903085321187973,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 8547
+    },
+    {
+      "epoch": 0.8784547416007398,
+      "grad_norm": 0.06924945116043091,
+      "learning_rate": 0.01,
+      "loss": 2.1004,
+      "step": 8550
+    },
+    {
+      "epoch": 0.8787629713346348,
+      "grad_norm": 0.04104374721646309,
+      "learning_rate": 0.01,
+      "loss": 2.0954,
+      "step": 8553
+    },
+    {
+      "epoch": 0.8790712010685298,
+      "grad_norm": 0.11671441793441772,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 8556
+    },
+    {
+      "epoch": 0.8793794308024248,
+      "grad_norm": 0.10247964411973953,
+      "learning_rate": 0.01,
+      "loss": 2.0861,
+      "step": 8559
+    },
+    {
+      "epoch": 0.8796876605363197,
+      "grad_norm": 0.03979288041591644,
+      "learning_rate": 0.01,
+      "loss": 2.1307,
+      "step": 8562
+    },
+    {
+      "epoch": 0.8799958902702147,
+      "grad_norm": 0.0406351312994957,
+      "learning_rate": 0.01,
+      "loss": 2.0868,
+      "step": 8565
+    },
+    {
+      "epoch": 0.8803041200041097,
+      "grad_norm": 0.04127006232738495,
+      "learning_rate": 0.01,
+      "loss": 2.0899,
+      "step": 8568
+    },
+    {
+      "epoch": 0.8806123497380047,
+      "grad_norm": 0.04559047520160675,
+      "learning_rate": 0.01,
+      "loss": 2.1071,
+      "step": 8571
+    },
+    {
+      "epoch": 0.8809205794718997,
+      "grad_norm": 0.12507610023021698,
+      "learning_rate": 0.01,
+      "loss": 2.0944,
+      "step": 8574
+    },
+    {
+      "epoch": 0.8812288092057947,
+      "grad_norm": 0.042683400213718414,
+      "learning_rate": 0.01,
+      "loss": 2.078,
+      "step": 8577
+    },
+    {
+      "epoch": 0.8815370389396897,
+      "grad_norm": 0.04022818058729172,
+      "learning_rate": 0.01,
+      "loss": 2.0797,
+      "step": 8580
+    },
+    {
+      "epoch": 0.8818452686735847,
+      "grad_norm": 0.0382862351834774,
+      "learning_rate": 0.01,
+      "loss": 2.0859,
+      "step": 8583
+    },
+    {
+      "epoch": 0.8821534984074797,
+      "grad_norm": 0.05260771885514259,
+      "learning_rate": 0.01,
+      "loss": 2.0832,
+      "step": 8586
+    },
+    {
+      "epoch": 0.8824617281413747,
+      "grad_norm": 0.05381648615002632,
+      "learning_rate": 0.01,
+      "loss": 2.1211,
+      "step": 8589
+    },
+    {
+      "epoch": 0.8827699578752697,
+      "grad_norm": 0.055818814784288406,
+      "learning_rate": 0.01,
+      "loss": 2.1108,
+      "step": 8592
+    },
+    {
+      "epoch": 0.8830781876091647,
+      "grad_norm": 0.16680215299129486,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 8595
+    },
+    {
+      "epoch": 0.8833864173430597,
+      "grad_norm": 0.10034742951393127,
+      "learning_rate": 0.01,
+      "loss": 2.1187,
+      "step": 8598
+    },
+    {
+      "epoch": 0.8836946470769547,
+      "grad_norm": 0.0827341303229332,
+      "learning_rate": 0.01,
+      "loss": 2.1112,
+      "step": 8601
+    },
+    {
+      "epoch": 0.8840028768108497,
+      "grad_norm": 0.07657956331968307,
+      "learning_rate": 0.01,
+      "loss": 2.0711,
+      "step": 8604
+    },
+    {
+      "epoch": 0.8843111065447447,
+      "grad_norm": 0.036220960319042206,
+      "learning_rate": 0.01,
+      "loss": 2.1097,
+      "step": 8607
+    },
+    {
+      "epoch": 0.8846193362786396,
+      "grad_norm": 0.04672658443450928,
+      "learning_rate": 0.01,
+      "loss": 2.1099,
+      "step": 8610
+    },
+    {
+      "epoch": 0.8849275660125346,
+      "grad_norm": 0.04827800393104553,
+      "learning_rate": 0.01,
+      "loss": 2.1081,
+      "step": 8613
+    },
+    {
+      "epoch": 0.8852357957464296,
+      "grad_norm": 0.04962724447250366,
+      "learning_rate": 0.01,
+      "loss": 2.0895,
+      "step": 8616
+    },
+    {
+      "epoch": 0.8855440254803246,
+      "grad_norm": 0.03474809601902962,
+      "learning_rate": 0.01,
+      "loss": 2.0942,
+      "step": 8619
+    },
+    {
+      "epoch": 0.8858522552142196,
+      "grad_norm": 0.07395246624946594,
+      "learning_rate": 0.01,
+      "loss": 2.1145,
+      "step": 8622
+    },
+    {
+      "epoch": 0.8861604849481146,
+      "grad_norm": 0.09853484481573105,
+      "learning_rate": 0.01,
+      "loss": 2.0991,
+      "step": 8625
+    },
+    {
+      "epoch": 0.8864687146820096,
+      "grad_norm": 0.11892013251781464,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 8628
+    },
+    {
+      "epoch": 0.8867769444159046,
+      "grad_norm": 0.12780621647834778,
+      "learning_rate": 0.01,
+      "loss": 2.1154,
+      "step": 8631
+    },
+    {
+      "epoch": 0.8870851741497997,
+      "grad_norm": 0.04470033943653107,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 8634
+    },
+    {
+      "epoch": 0.8873934038836947,
+      "grad_norm": 0.054323747754096985,
+      "learning_rate": 0.01,
+      "loss": 2.0952,
+      "step": 8637
+    },
+    {
+      "epoch": 0.8877016336175897,
+      "grad_norm": 0.08175788819789886,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 8640
+    },
+    {
+      "epoch": 0.8880098633514847,
+      "grad_norm": 0.07456079125404358,
+      "learning_rate": 0.01,
+      "loss": 2.141,
+      "step": 8643
+    },
+    {
+      "epoch": 0.8883180930853797,
+      "grad_norm": 0.055910736322402954,
+      "learning_rate": 0.01,
+      "loss": 2.1102,
+      "step": 8646
+    },
+    {
+      "epoch": 0.8886263228192747,
+      "grad_norm": 0.05231192335486412,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 8649
+    },
+    {
+      "epoch": 0.8889345525531697,
+      "grad_norm": 0.05306578800082207,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 8652
+    },
+    {
+      "epoch": 0.8892427822870647,
+      "grad_norm": 0.05569072067737579,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 8655
+    },
+    {
+      "epoch": 0.8895510120209597,
+      "grad_norm": 0.050971515476703644,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 8658
+    },
+    {
+      "epoch": 0.8898592417548546,
+      "grad_norm": 0.061436936259269714,
+      "learning_rate": 0.01,
+      "loss": 2.1167,
+      "step": 8661
+    },
+    {
+      "epoch": 0.8901674714887496,
+      "grad_norm": 0.04307536780834198,
+      "learning_rate": 0.01,
+      "loss": 2.0972,
+      "step": 8664
+    },
+    {
+      "epoch": 0.8904757012226446,
+      "grad_norm": 0.1459832638502121,
+      "learning_rate": 0.01,
+      "loss": 2.1306,
+      "step": 8667
+    },
+    {
+      "epoch": 0.8907839309565396,
+      "grad_norm": 0.05527958646416664,
+      "learning_rate": 0.01,
+      "loss": 2.0974,
+      "step": 8670
+    },
+    {
+      "epoch": 0.8910921606904346,
+      "grad_norm": 0.1319393813610077,
+      "learning_rate": 0.01,
+      "loss": 2.1259,
+      "step": 8673
+    },
+    {
+      "epoch": 0.8914003904243296,
+      "grad_norm": 0.06124665215611458,
+      "learning_rate": 0.01,
+      "loss": 2.0997,
+      "step": 8676
+    },
+    {
+      "epoch": 0.8917086201582246,
+      "grad_norm": 0.08667455613613129,
+      "learning_rate": 0.01,
+      "loss": 2.0941,
+      "step": 8679
+    },
+    {
+      "epoch": 0.8920168498921196,
+      "grad_norm": 0.06631213426589966,
+      "learning_rate": 0.01,
+      "loss": 2.1196,
+      "step": 8682
+    },
+    {
+      "epoch": 0.8923250796260146,
+      "grad_norm": 0.060188647359609604,
+      "learning_rate": 0.01,
+      "loss": 2.0971,
+      "step": 8685
+    },
+    {
+      "epoch": 0.8926333093599096,
+      "grad_norm": 0.039312943816185,
+      "learning_rate": 0.01,
+      "loss": 2.1119,
+      "step": 8688
+    },
+    {
+      "epoch": 0.8929415390938046,
+      "grad_norm": 0.03959662839770317,
+      "learning_rate": 0.01,
+      "loss": 2.0897,
+      "step": 8691
+    },
+    {
+      "epoch": 0.8932497688276996,
+      "grad_norm": 0.09711046516895294,
+      "learning_rate": 0.01,
+      "loss": 2.1133,
+      "step": 8694
+    },
+    {
+      "epoch": 0.8935579985615946,
+      "grad_norm": 0.07965920865535736,
+      "learning_rate": 0.01,
+      "loss": 2.0635,
+      "step": 8697
+    },
+    {
+      "epoch": 0.8938662282954896,
+      "grad_norm": 0.08770687133073807,
+      "learning_rate": 0.01,
+      "loss": 2.0885,
+      "step": 8700
+    },
+    {
+      "epoch": 0.8941744580293846,
+      "grad_norm": 0.04591045528650284,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 8703
+    },
+    {
+      "epoch": 0.8944826877632795,
+      "grad_norm": 0.09602218866348267,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 8706
+    },
+    {
+      "epoch": 0.8947909174971745,
+      "grad_norm": 0.09482742100954056,
+      "learning_rate": 0.01,
+      "loss": 2.0966,
+      "step": 8709
+    },
+    {
+      "epoch": 0.8950991472310695,
+      "grad_norm": 0.03937089815735817,
+      "learning_rate": 0.01,
+      "loss": 2.1043,
+      "step": 8712
+    },
+    {
+      "epoch": 0.8954073769649645,
+      "grad_norm": 0.056832704693078995,
+      "learning_rate": 0.01,
+      "loss": 2.1165,
+      "step": 8715
+    },
+    {
+      "epoch": 0.8957156066988595,
+      "grad_norm": 0.06370353698730469,
+      "learning_rate": 0.01,
+      "loss": 2.1144,
+      "step": 8718
+    },
+    {
+      "epoch": 0.8960238364327545,
+      "grad_norm": 0.06752549856901169,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 8721
+    },
+    {
+      "epoch": 0.8963320661666495,
+      "grad_norm": 0.13301892578601837,
+      "learning_rate": 0.01,
+      "loss": 2.11,
+      "step": 8724
+    },
+    {
+      "epoch": 0.8966402959005445,
+      "grad_norm": 0.05210836976766586,
+      "learning_rate": 0.01,
+      "loss": 2.0925,
+      "step": 8727
+    },
+    {
+      "epoch": 0.8969485256344395,
+      "grad_norm": 0.03570270165801048,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 8730
+    },
+    {
+      "epoch": 0.8972567553683345,
+      "grad_norm": 0.05898820236325264,
+      "learning_rate": 0.01,
+      "loss": 2.0786,
+      "step": 8733
+    },
+    {
+      "epoch": 0.8975649851022295,
+      "grad_norm": 0.05087563395500183,
+      "learning_rate": 0.01,
+      "loss": 2.1071,
+      "step": 8736
+    },
+    {
+      "epoch": 0.8978732148361245,
+      "grad_norm": 0.09473355114459991,
+      "learning_rate": 0.01,
+      "loss": 2.103,
+      "step": 8739
+    },
+    {
+      "epoch": 0.8981814445700195,
+      "grad_norm": 0.09793075919151306,
+      "learning_rate": 0.01,
+      "loss": 2.0972,
+      "step": 8742
+    },
+    {
+      "epoch": 0.8984896743039145,
+      "grad_norm": 0.05115204304456711,
+      "learning_rate": 0.01,
+      "loss": 2.0979,
+      "step": 8745
+    },
+    {
+      "epoch": 0.8987979040378095,
+      "grad_norm": 0.057413987815380096,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 8748
+    },
+    {
+      "epoch": 0.8991061337717046,
+      "grad_norm": 0.04136224836111069,
+      "learning_rate": 0.01,
+      "loss": 2.1269,
+      "step": 8751
+    },
+    {
+      "epoch": 0.8994143635055996,
+      "grad_norm": 0.06866753846406937,
+      "learning_rate": 0.01,
+      "loss": 2.1092,
+      "step": 8754
+    },
+    {
+      "epoch": 0.8997225932394946,
+      "grad_norm": 0.0757627934217453,
+      "learning_rate": 0.01,
+      "loss": 2.0933,
+      "step": 8757
+    },
+    {
+      "epoch": 0.9000308229733895,
+      "grad_norm": 0.08082983642816544,
+      "learning_rate": 0.01,
+      "loss": 2.1124,
+      "step": 8760
+    },
+    {
+      "epoch": 0.9003390527072845,
+      "grad_norm": 0.046828944236040115,
+      "learning_rate": 0.01,
+      "loss": 2.0978,
+      "step": 8763
+    },
+    {
+      "epoch": 0.9006472824411795,
+      "grad_norm": 0.11039458215236664,
+      "learning_rate": 0.01,
+      "loss": 2.0989,
+      "step": 8766
+    },
+    {
+      "epoch": 0.9009555121750745,
+      "grad_norm": 0.048537638038396835,
+      "learning_rate": 0.01,
+      "loss": 2.0946,
+      "step": 8769
+    },
+    {
+      "epoch": 0.9012637419089695,
+      "grad_norm": 0.06700310111045837,
+      "learning_rate": 0.01,
+      "loss": 2.1184,
+      "step": 8772
+    },
+    {
+      "epoch": 0.9015719716428645,
+      "grad_norm": 0.044369909912347794,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 8775
+    },
+    {
+      "epoch": 0.9018802013767595,
+      "grad_norm": 0.041071876883506775,
+      "learning_rate": 0.01,
+      "loss": 2.0774,
+      "step": 8778
+    },
+    {
+      "epoch": 0.9021884311106545,
+      "grad_norm": 0.04735315591096878,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 8781
+    },
+    {
+      "epoch": 0.9024966608445495,
+      "grad_norm": 0.11621284484863281,
+      "learning_rate": 0.01,
+      "loss": 2.0766,
+      "step": 8784
+    },
+    {
+      "epoch": 0.9028048905784445,
+      "grad_norm": 0.11453153938055038,
+      "learning_rate": 0.01,
+      "loss": 2.0866,
+      "step": 8787
+    },
+    {
+      "epoch": 0.9031131203123395,
+      "grad_norm": 0.057418763637542725,
+      "learning_rate": 0.01,
+      "loss": 2.081,
+      "step": 8790
+    },
+    {
+      "epoch": 0.9034213500462345,
+      "grad_norm": 0.041579000651836395,
+      "learning_rate": 0.01,
+      "loss": 2.1154,
+      "step": 8793
+    },
+    {
+      "epoch": 0.9037295797801295,
+      "grad_norm": 0.045673951506614685,
+      "learning_rate": 0.01,
+      "loss": 2.1241,
+      "step": 8796
+    },
+    {
+      "epoch": 0.9040378095140245,
+      "grad_norm": 0.05963718518614769,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 8799
+    },
+    {
+      "epoch": 0.9043460392479195,
+      "grad_norm": 0.04776541888713837,
+      "learning_rate": 0.01,
+      "loss": 2.1138,
+      "step": 8802
+    },
+    {
+      "epoch": 0.9046542689818144,
+      "grad_norm": 0.09103482216596603,
+      "learning_rate": 0.01,
+      "loss": 2.1192,
+      "step": 8805
+    },
+    {
+      "epoch": 0.9049624987157094,
+      "grad_norm": 0.09218809008598328,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 8808
+    },
+    {
+      "epoch": 0.9052707284496044,
+      "grad_norm": 0.10253725945949554,
+      "learning_rate": 0.01,
+      "loss": 2.1189,
+      "step": 8811
+    },
+    {
+      "epoch": 0.9055789581834994,
+      "grad_norm": 0.09638465940952301,
+      "learning_rate": 0.01,
+      "loss": 2.1008,
+      "step": 8814
+    },
+    {
+      "epoch": 0.9058871879173944,
+      "grad_norm": 0.0947449579834938,
+      "learning_rate": 0.01,
+      "loss": 2.1222,
+      "step": 8817
+    },
+    {
+      "epoch": 0.9061954176512894,
+      "grad_norm": 0.04588090255856514,
+      "learning_rate": 0.01,
+      "loss": 2.1198,
+      "step": 8820
+    },
+    {
+      "epoch": 0.9065036473851844,
+      "grad_norm": 0.05041109770536423,
+      "learning_rate": 0.01,
+      "loss": 2.0843,
+      "step": 8823
+    },
+    {
+      "epoch": 0.9068118771190794,
+      "grad_norm": 0.038898076862096786,
+      "learning_rate": 0.01,
+      "loss": 2.125,
+      "step": 8826
+    },
+    {
+      "epoch": 0.9071201068529744,
+      "grad_norm": 0.03356321156024933,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 8829
+    },
+    {
+      "epoch": 0.9074283365868694,
+      "grad_norm": 0.04668448120355606,
+      "learning_rate": 0.01,
+      "loss": 2.1071,
+      "step": 8832
+    },
+    {
+      "epoch": 0.9077365663207644,
+      "grad_norm": 0.051277391612529755,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 8835
+    },
+    {
+      "epoch": 0.9080447960546594,
+      "grad_norm": 0.049883171916007996,
+      "learning_rate": 0.01,
+      "loss": 2.1111,
+      "step": 8838
+    },
+    {
+      "epoch": 0.9083530257885544,
+      "grad_norm": 0.04149313643574715,
+      "learning_rate": 0.01,
+      "loss": 2.0991,
+      "step": 8841
+    },
+    {
+      "epoch": 0.9086612555224494,
+      "grad_norm": 0.09206261485815048,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 8844
+    },
+    {
+      "epoch": 0.9089694852563444,
+      "grad_norm": 0.1830751895904541,
+      "learning_rate": 0.01,
+      "loss": 2.1093,
+      "step": 8847
+    },
+    {
+      "epoch": 0.9092777149902393,
+      "grad_norm": 0.0757865458726883,
+      "learning_rate": 0.01,
+      "loss": 2.115,
+      "step": 8850
+    },
+    {
+      "epoch": 0.9095859447241343,
+      "grad_norm": 0.06030673533678055,
+      "learning_rate": 0.01,
+      "loss": 2.0874,
+      "step": 8853
+    },
+    {
+      "epoch": 0.9098941744580293,
+      "grad_norm": 0.03440079465508461,
+      "learning_rate": 0.01,
+      "loss": 2.0997,
+      "step": 8856
+    },
+    {
+      "epoch": 0.9102024041919243,
+      "grad_norm": 0.040004558861255646,
+      "learning_rate": 0.01,
+      "loss": 2.0767,
+      "step": 8859
+    },
+    {
+      "epoch": 0.9105106339258193,
+      "grad_norm": 0.033261023461818695,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 8862
+    },
+    {
+      "epoch": 0.9108188636597144,
+      "grad_norm": 0.04814066365361214,
+      "learning_rate": 0.01,
+      "loss": 2.0868,
+      "step": 8865
+    },
+    {
+      "epoch": 0.9111270933936094,
+      "grad_norm": 0.04939806088805199,
+      "learning_rate": 0.01,
+      "loss": 2.0944,
+      "step": 8868
+    },
+    {
+      "epoch": 0.9114353231275044,
+      "grad_norm": 0.05242007225751877,
+      "learning_rate": 0.01,
+      "loss": 2.1035,
+      "step": 8871
+    },
+    {
+      "epoch": 0.9117435528613994,
+      "grad_norm": 0.04576495289802551,
+      "learning_rate": 0.01,
+      "loss": 2.0881,
+      "step": 8874
+    },
+    {
+      "epoch": 0.9120517825952944,
+      "grad_norm": 0.0369776152074337,
+      "learning_rate": 0.01,
+      "loss": 2.1017,
+      "step": 8877
+    },
+    {
+      "epoch": 0.9123600123291894,
+      "grad_norm": 0.08296829462051392,
+      "learning_rate": 0.01,
+      "loss": 2.1199,
+      "step": 8880
+    },
+    {
+      "epoch": 0.9126682420630844,
+      "grad_norm": 0.07186676561832428,
+      "learning_rate": 0.01,
+      "loss": 2.0906,
+      "step": 8883
+    },
+    {
+      "epoch": 0.9129764717969794,
+      "grad_norm": 0.06849399209022522,
+      "learning_rate": 0.01,
+      "loss": 2.0944,
+      "step": 8886
+    },
+    {
+      "epoch": 0.9132847015308744,
+      "grad_norm": 0.1285102367401123,
+      "learning_rate": 0.01,
+      "loss": 2.0959,
+      "step": 8889
+    },
+    {
+      "epoch": 0.9135929312647694,
+      "grad_norm": 0.045700203627347946,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 8892
+    },
+    {
+      "epoch": 0.9139011609986644,
+      "grad_norm": 0.04561945050954819,
+      "learning_rate": 0.01,
+      "loss": 2.1126,
+      "step": 8895
+    },
+    {
+      "epoch": 0.9142093907325594,
+      "grad_norm": 0.0417817123234272,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 8898
+    },
+    {
+      "epoch": 0.9145176204664544,
+      "grad_norm": 0.07923369109630585,
+      "learning_rate": 0.01,
+      "loss": 2.1059,
+      "step": 8901
+    },
+    {
+      "epoch": 0.9148258502003493,
+      "grad_norm": 0.052836060523986816,
+      "learning_rate": 0.01,
+      "loss": 2.1089,
+      "step": 8904
+    },
+    {
+      "epoch": 0.9151340799342443,
+      "grad_norm": 0.04591790586709976,
+      "learning_rate": 0.01,
+      "loss": 2.1007,
+      "step": 8907
+    },
+    {
+      "epoch": 0.9154423096681393,
+      "grad_norm": 0.09871240705251694,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 8910
+    },
+    {
+      "epoch": 0.9157505394020343,
+      "grad_norm": 0.044554613530635834,
+      "learning_rate": 0.01,
+      "loss": 2.0956,
+      "step": 8913
+    },
+    {
+      "epoch": 0.9160587691359293,
+      "grad_norm": 0.10009585320949554,
+      "learning_rate": 0.01,
+      "loss": 2.0838,
+      "step": 8916
+    },
+    {
+      "epoch": 0.9163669988698243,
+      "grad_norm": 0.07252159714698792,
+      "learning_rate": 0.01,
+      "loss": 2.0973,
+      "step": 8919
+    },
+    {
+      "epoch": 0.9166752286037193,
+      "grad_norm": 0.09162852168083191,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 8922
+    },
+    {
+      "epoch": 0.9169834583376143,
+      "grad_norm": 0.06149733439087868,
+      "learning_rate": 0.01,
+      "loss": 2.1377,
+      "step": 8925
+    },
+    {
+      "epoch": 0.9172916880715093,
+      "grad_norm": 0.09315814077854156,
+      "learning_rate": 0.01,
+      "loss": 2.0901,
+      "step": 8928
+    },
+    {
+      "epoch": 0.9175999178054043,
+      "grad_norm": 0.056877728551626205,
+      "learning_rate": 0.01,
+      "loss": 2.0934,
+      "step": 8931
+    },
+    {
+      "epoch": 0.9179081475392993,
+      "grad_norm": 0.0976705476641655,
+      "learning_rate": 0.01,
+      "loss": 2.0791,
+      "step": 8934
+    },
+    {
+      "epoch": 0.9182163772731943,
+      "grad_norm": 0.0493176206946373,
+      "learning_rate": 0.01,
+      "loss": 2.0937,
+      "step": 8937
+    },
+    {
+      "epoch": 0.9185246070070893,
+      "grad_norm": 0.06268187612295151,
+      "learning_rate": 0.01,
+      "loss": 2.1053,
+      "step": 8940
+    },
+    {
+      "epoch": 0.9188328367409843,
+      "grad_norm": 0.049251820892095566,
+      "learning_rate": 0.01,
+      "loss": 2.1104,
+      "step": 8943
+    },
+    {
+      "epoch": 0.9191410664748793,
+      "grad_norm": 0.05342431366443634,
+      "learning_rate": 0.01,
+      "loss": 2.1005,
+      "step": 8946
+    },
+    {
+      "epoch": 0.9194492962087742,
+      "grad_norm": 0.036090634763240814,
+      "learning_rate": 0.01,
+      "loss": 2.0815,
+      "step": 8949
+    },
+    {
+      "epoch": 0.9197575259426692,
+      "grad_norm": 0.0320359505712986,
+      "learning_rate": 0.01,
+      "loss": 2.0704,
+      "step": 8952
+    },
+    {
+      "epoch": 0.9200657556765642,
+      "grad_norm": 0.03514352813363075,
+      "learning_rate": 0.01,
+      "loss": 2.1046,
+      "step": 8955
+    },
+    {
+      "epoch": 0.9203739854104592,
+      "grad_norm": 0.06132291629910469,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 8958
+    },
+    {
+      "epoch": 0.9206822151443542,
+      "grad_norm": 0.07312822341918945,
+      "learning_rate": 0.01,
+      "loss": 2.1079,
+      "step": 8961
+    },
+    {
+      "epoch": 0.9209904448782492,
+      "grad_norm": 0.09670150279998779,
+      "learning_rate": 0.01,
+      "loss": 2.1195,
+      "step": 8964
+    },
+    {
+      "epoch": 0.9212986746121442,
+      "grad_norm": 0.1106385663151741,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 8967
+    },
+    {
+      "epoch": 0.9216069043460392,
+      "grad_norm": 0.05964332073926926,
+      "learning_rate": 0.01,
+      "loss": 2.1108,
+      "step": 8970
+    },
+    {
+      "epoch": 0.9219151340799342,
+      "grad_norm": 0.05584556236863136,
+      "learning_rate": 0.01,
+      "loss": 2.1274,
+      "step": 8973
+    },
+    {
+      "epoch": 0.9222233638138292,
+      "grad_norm": 0.04485652595758438,
+      "learning_rate": 0.01,
+      "loss": 2.0627,
+      "step": 8976
+    },
+    {
+      "epoch": 0.9225315935477242,
+      "grad_norm": 0.07286686450242996,
+      "learning_rate": 0.01,
+      "loss": 2.1087,
+      "step": 8979
+    },
+    {
+      "epoch": 0.9228398232816193,
+      "grad_norm": 0.10815869271755219,
+      "learning_rate": 0.01,
+      "loss": 2.1057,
+      "step": 8982
+    },
+    {
+      "epoch": 0.9231480530155143,
+      "grad_norm": 0.1037832722067833,
+      "learning_rate": 0.01,
+      "loss": 2.0836,
+      "step": 8985
+    },
+    {
+      "epoch": 0.9234562827494093,
+      "grad_norm": 0.08297618478536606,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 8988
+    },
+    {
+      "epoch": 0.9237645124833043,
+      "grad_norm": 0.04203306511044502,
+      "learning_rate": 0.01,
+      "loss": 2.1112,
+      "step": 8991
+    },
+    {
+      "epoch": 0.9240727422171993,
+      "grad_norm": 0.06641580909490585,
+      "learning_rate": 0.01,
+      "loss": 2.1004,
+      "step": 8994
+    },
+    {
+      "epoch": 0.9243809719510943,
+      "grad_norm": 0.04921744763851166,
+      "learning_rate": 0.01,
+      "loss": 2.1116,
+      "step": 8997
+    },
+    {
+      "epoch": 0.9246892016849892,
+      "grad_norm": 0.03472235053777695,
+      "learning_rate": 0.01,
+      "loss": 2.0777,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9249974314188842,
+      "grad_norm": 0.03650922700762749,
+      "learning_rate": 0.01,
+      "loss": 2.0802,
+      "step": 9003
+    },
+    {
+      "epoch": 0.9253056611527792,
+      "grad_norm": 0.04657342657446861,
+      "learning_rate": 0.01,
+      "loss": 2.0773,
+      "step": 9006
+    },
+    {
+      "epoch": 0.9256138908866742,
+      "grad_norm": 0.05943501368165016,
+      "learning_rate": 0.01,
+      "loss": 2.0753,
+      "step": 9009
+    },
+    {
+      "epoch": 0.9259221206205692,
+      "grad_norm": 0.04763554409146309,
+      "learning_rate": 0.01,
+      "loss": 2.0959,
+      "step": 9012
+    },
+    {
+      "epoch": 0.9262303503544642,
+      "grad_norm": 0.1267511248588562,
+      "learning_rate": 0.01,
+      "loss": 2.0971,
+      "step": 9015
+    },
+    {
+      "epoch": 0.9265385800883592,
+      "grad_norm": 0.055529460310935974,
+      "learning_rate": 0.01,
+      "loss": 2.1327,
+      "step": 9018
+    },
+    {
+      "epoch": 0.9268468098222542,
+      "grad_norm": 0.15508927404880524,
+      "learning_rate": 0.01,
+      "loss": 2.0947,
+      "step": 9021
+    },
+    {
+      "epoch": 0.9271550395561492,
+      "grad_norm": 0.0593777671456337,
+      "learning_rate": 0.01,
+      "loss": 2.1171,
+      "step": 9024
+    },
+    {
+      "epoch": 0.9274632692900442,
+      "grad_norm": 0.08907107263803482,
+      "learning_rate": 0.01,
+      "loss": 2.093,
+      "step": 9027
+    },
+    {
+      "epoch": 0.9277714990239392,
+      "grad_norm": 0.07041808217763901,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 9030
+    },
+    {
+      "epoch": 0.9280797287578342,
+      "grad_norm": 0.03434208780527115,
+      "learning_rate": 0.01,
+      "loss": 2.0928,
+      "step": 9033
+    },
+    {
+      "epoch": 0.9283879584917292,
+      "grad_norm": 0.07591548562049866,
+      "learning_rate": 0.01,
+      "loss": 2.0857,
+      "step": 9036
+    },
+    {
+      "epoch": 0.9286961882256242,
+      "grad_norm": 0.08999443799257278,
+      "learning_rate": 0.01,
+      "loss": 2.0984,
+      "step": 9039
+    },
+    {
+      "epoch": 0.9290044179595192,
+      "grad_norm": 0.11046464741230011,
+      "learning_rate": 0.01,
+      "loss": 2.1009,
+      "step": 9042
+    },
+    {
+      "epoch": 0.9293126476934142,
+      "grad_norm": 0.08271370083093643,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 9045
+    },
+    {
+      "epoch": 0.9296208774273091,
+      "grad_norm": 0.046337299048900604,
+      "learning_rate": 0.01,
+      "loss": 2.0826,
+      "step": 9048
+    },
+    {
+      "epoch": 0.9299291071612041,
+      "grad_norm": 0.037284769117832184,
+      "learning_rate": 0.01,
+      "loss": 2.1015,
+      "step": 9051
+    },
+    {
+      "epoch": 0.9302373368950991,
+      "grad_norm": 0.04956496134400368,
+      "learning_rate": 0.01,
+      "loss": 2.1036,
+      "step": 9054
+    },
+    {
+      "epoch": 0.9305455666289941,
+      "grad_norm": 0.12329571694135666,
+      "learning_rate": 0.01,
+      "loss": 2.0917,
+      "step": 9057
+    },
+    {
+      "epoch": 0.9308537963628891,
+      "grad_norm": 0.06971380859613419,
+      "learning_rate": 0.01,
+      "loss": 2.114,
+      "step": 9060
+    },
+    {
+      "epoch": 0.9311620260967841,
+      "grad_norm": 0.06084508076310158,
+      "learning_rate": 0.01,
+      "loss": 2.1122,
+      "step": 9063
+    },
+    {
+      "epoch": 0.9314702558306791,
+      "grad_norm": 0.049602411687374115,
+      "learning_rate": 0.01,
+      "loss": 2.1268,
+      "step": 9066
+    },
+    {
+      "epoch": 0.9317784855645741,
+      "grad_norm": 0.05200349539518356,
+      "learning_rate": 0.01,
+      "loss": 2.0979,
+      "step": 9069
+    },
+    {
+      "epoch": 0.9320867152984691,
+      "grad_norm": 0.05793909728527069,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 9072
+    },
+    {
+      "epoch": 0.9323949450323641,
+      "grad_norm": 0.10819883644580841,
+      "learning_rate": 0.01,
+      "loss": 2.1096,
+      "step": 9075
+    },
+    {
+      "epoch": 0.9327031747662591,
+      "grad_norm": 0.07809442281723022,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 9078
+    },
+    {
+      "epoch": 0.9330114045001541,
+      "grad_norm": 0.09595733880996704,
+      "learning_rate": 0.01,
+      "loss": 2.0769,
+      "step": 9081
+    },
+    {
+      "epoch": 0.9333196342340491,
+      "grad_norm": 0.11658616364002228,
+      "learning_rate": 0.01,
+      "loss": 2.0945,
+      "step": 9084
+    },
+    {
+      "epoch": 0.9336278639679441,
+      "grad_norm": 0.07642678171396255,
+      "learning_rate": 0.01,
+      "loss": 2.0811,
+      "step": 9087
+    },
+    {
+      "epoch": 0.933936093701839,
+      "grad_norm": 0.03174865245819092,
+      "learning_rate": 0.01,
+      "loss": 2.1017,
+      "step": 9090
+    },
+    {
+      "epoch": 0.934244323435734,
+      "grad_norm": 0.05137626454234123,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 9093
+    },
+    {
+      "epoch": 0.9345525531696292,
+      "grad_norm": 0.05306951329112053,
+      "learning_rate": 0.01,
+      "loss": 2.1163,
+      "step": 9096
+    },
+    {
+      "epoch": 0.9348607829035241,
+      "grad_norm": 0.0716642439365387,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 9099
+    },
+    {
+      "epoch": 0.9351690126374191,
+      "grad_norm": 0.10328514873981476,
+      "learning_rate": 0.01,
+      "loss": 2.0789,
+      "step": 9102
+    },
+    {
+      "epoch": 0.9354772423713141,
+      "grad_norm": 0.04914560168981552,
+      "learning_rate": 0.01,
+      "loss": 2.0963,
+      "step": 9105
+    },
+    {
+      "epoch": 0.9357854721052091,
+      "grad_norm": 0.04810576140880585,
+      "learning_rate": 0.01,
+      "loss": 2.1119,
+      "step": 9108
+    },
+    {
+      "epoch": 0.9360937018391041,
+      "grad_norm": 0.05689787119626999,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 9111
+    },
+    {
+      "epoch": 0.9364019315729991,
+      "grad_norm": 0.06455382704734802,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 9114
+    },
+    {
+      "epoch": 0.9367101613068941,
+      "grad_norm": 0.044911760836839676,
+      "learning_rate": 0.01,
+      "loss": 2.0967,
+      "step": 9117
+    },
+    {
+      "epoch": 0.9370183910407891,
+      "grad_norm": 0.06244887784123421,
+      "learning_rate": 0.01,
+      "loss": 2.0921,
+      "step": 9120
+    },
+    {
+      "epoch": 0.9373266207746841,
+      "grad_norm": 0.052621614187955856,
+      "learning_rate": 0.01,
+      "loss": 2.1296,
+      "step": 9123
+    },
+    {
+      "epoch": 0.9376348505085791,
+      "grad_norm": 0.05098232626914978,
+      "learning_rate": 0.01,
+      "loss": 2.0807,
+      "step": 9126
+    },
+    {
+      "epoch": 0.9379430802424741,
+      "grad_norm": 0.058582011610269547,
+      "learning_rate": 0.01,
+      "loss": 2.0973,
+      "step": 9129
+    },
+    {
+      "epoch": 0.9382513099763691,
+      "grad_norm": 0.10984500497579575,
+      "learning_rate": 0.01,
+      "loss": 2.0789,
+      "step": 9132
+    },
+    {
+      "epoch": 0.9385595397102641,
+      "grad_norm": 0.045173123478889465,
+      "learning_rate": 0.01,
+      "loss": 2.0937,
+      "step": 9135
+    },
+    {
+      "epoch": 0.9388677694441591,
+      "grad_norm": 0.06749478727579117,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 9138
+    },
+    {
+      "epoch": 0.939175999178054,
+      "grad_norm": 0.06236808001995087,
+      "learning_rate": 0.01,
+      "loss": 2.1099,
+      "step": 9141
+    },
+    {
+      "epoch": 0.939484228911949,
+      "grad_norm": 0.06205837428569794,
+      "learning_rate": 0.01,
+      "loss": 2.0893,
+      "step": 9144
+    },
+    {
+      "epoch": 0.939792458645844,
+      "grad_norm": 0.0742972195148468,
+      "learning_rate": 0.01,
+      "loss": 2.1034,
+      "step": 9147
+    },
+    {
+      "epoch": 0.940100688379739,
+      "grad_norm": 0.06998419016599655,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 9150
+    },
+    {
+      "epoch": 0.940408918113634,
+      "grad_norm": 0.04214362055063248,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 9153
+    },
+    {
+      "epoch": 0.940717147847529,
+      "grad_norm": 0.055913276970386505,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 9156
+    },
+    {
+      "epoch": 0.941025377581424,
+      "grad_norm": 0.0941486805677414,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 9159
+    },
+    {
+      "epoch": 0.941333607315319,
+      "grad_norm": 0.05609782040119171,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 9162
+    },
+    {
+      "epoch": 0.941641837049214,
+      "grad_norm": 0.05714662745594978,
+      "learning_rate": 0.01,
+      "loss": 2.0939,
+      "step": 9165
+    },
+    {
+      "epoch": 0.941950066783109,
+      "grad_norm": 0.05364496633410454,
+      "learning_rate": 0.01,
+      "loss": 2.0838,
+      "step": 9168
+    },
+    {
+      "epoch": 0.942258296517004,
+      "grad_norm": 0.050090350210666656,
+      "learning_rate": 0.01,
+      "loss": 2.087,
+      "step": 9171
+    },
+    {
+      "epoch": 0.942566526250899,
+      "grad_norm": 0.07287559658288956,
+      "learning_rate": 0.01,
+      "loss": 2.098,
+      "step": 9174
+    },
+    {
+      "epoch": 0.942874755984794,
+      "grad_norm": 0.04061901941895485,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 9177
+    },
+    {
+      "epoch": 0.943182985718689,
+      "grad_norm": 0.10750306397676468,
+      "learning_rate": 0.01,
+      "loss": 2.1105,
+      "step": 9180
+    },
+    {
+      "epoch": 0.943491215452584,
+      "grad_norm": 0.10353365540504456,
+      "learning_rate": 0.01,
+      "loss": 2.0712,
+      "step": 9183
+    },
+    {
+      "epoch": 0.943799445186479,
+      "grad_norm": 0.07502592355012894,
+      "learning_rate": 0.01,
+      "loss": 2.1115,
+      "step": 9186
+    },
+    {
+      "epoch": 0.944107674920374,
+      "grad_norm": 0.046962104737758636,
+      "learning_rate": 0.01,
+      "loss": 2.0937,
+      "step": 9189
+    },
+    {
+      "epoch": 0.944415904654269,
+      "grad_norm": 0.05084332078695297,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 9192
+    },
+    {
+      "epoch": 0.9447241343881639,
+      "grad_norm": 0.0458371527493,
+      "learning_rate": 0.01,
+      "loss": 2.0967,
+      "step": 9195
+    },
+    {
+      "epoch": 0.9450323641220589,
+      "grad_norm": 0.040458545088768005,
+      "learning_rate": 0.01,
+      "loss": 2.0949,
+      "step": 9198
+    },
+    {
+      "epoch": 0.9453405938559539,
+      "grad_norm": 0.046158358454704285,
+      "learning_rate": 0.01,
+      "loss": 2.0912,
+      "step": 9201
+    },
+    {
+      "epoch": 0.9456488235898489,
+      "grad_norm": 0.10080043226480484,
+      "learning_rate": 0.01,
+      "loss": 2.1,
+      "step": 9204
+    },
+    {
+      "epoch": 0.9459570533237439,
+      "grad_norm": 0.07679333537817001,
+      "learning_rate": 0.01,
+      "loss": 2.1013,
+      "step": 9207
+    },
+    {
+      "epoch": 0.9462652830576389,
+      "grad_norm": 0.07189175486564636,
+      "learning_rate": 0.01,
+      "loss": 2.1008,
+      "step": 9210
+    },
+    {
+      "epoch": 0.946573512791534,
+      "grad_norm": 0.07828579097986221,
+      "learning_rate": 0.01,
+      "loss": 2.1063,
+      "step": 9213
+    },
+    {
+      "epoch": 0.946881742525429,
+      "grad_norm": 0.07649674266576767,
+      "learning_rate": 0.01,
+      "loss": 2.1146,
+      "step": 9216
+    },
+    {
+      "epoch": 0.947189972259324,
+      "grad_norm": 0.06558651477098465,
+      "learning_rate": 0.01,
+      "loss": 2.0705,
+      "step": 9219
+    },
+    {
+      "epoch": 0.947498201993219,
+      "grad_norm": 0.03276702016592026,
+      "learning_rate": 0.01,
+      "loss": 2.1065,
+      "step": 9222
+    },
+    {
+      "epoch": 0.947806431727114,
+      "grad_norm": 0.03779645636677742,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 9225
+    },
+    {
+      "epoch": 0.948114661461009,
+      "grad_norm": 0.048466913402080536,
+      "learning_rate": 0.01,
+      "loss": 2.1037,
+      "step": 9228
+    },
+    {
+      "epoch": 0.948422891194904,
+      "grad_norm": 0.04391203075647354,
+      "learning_rate": 0.01,
+      "loss": 2.0722,
+      "step": 9231
+    },
+    {
+      "epoch": 0.948731120928799,
+      "grad_norm": 0.11353743076324463,
+      "learning_rate": 0.01,
+      "loss": 2.113,
+      "step": 9234
+    },
+    {
+      "epoch": 0.949039350662694,
+      "grad_norm": 0.045930709689855576,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 9237
+    },
+    {
+      "epoch": 0.949347580396589,
+      "grad_norm": 0.06440164893865585,
+      "learning_rate": 0.01,
+      "loss": 2.0786,
+      "step": 9240
+    },
+    {
+      "epoch": 0.949655810130484,
+      "grad_norm": 0.08666238933801651,
+      "learning_rate": 0.01,
+      "loss": 2.1049,
+      "step": 9243
+    },
+    {
+      "epoch": 0.9499640398643789,
+      "grad_norm": 0.11012524366378784,
+      "learning_rate": 0.01,
+      "loss": 2.1018,
+      "step": 9246
+    },
+    {
+      "epoch": 0.9502722695982739,
+      "grad_norm": 0.047307875007390976,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 9249
+    },
+    {
+      "epoch": 0.9505804993321689,
+      "grad_norm": 0.04565277695655823,
+      "learning_rate": 0.01,
+      "loss": 2.1174,
+      "step": 9252
+    },
+    {
+      "epoch": 0.9508887290660639,
+      "grad_norm": 0.03389623388648033,
+      "learning_rate": 0.01,
+      "loss": 2.0896,
+      "step": 9255
+    },
+    {
+      "epoch": 0.9511969587999589,
+      "grad_norm": 0.04582008346915245,
+      "learning_rate": 0.01,
+      "loss": 2.0888,
+      "step": 9258
+    },
+    {
+      "epoch": 0.9515051885338539,
+      "grad_norm": 0.07722247391939163,
+      "learning_rate": 0.01,
+      "loss": 2.0843,
+      "step": 9261
+    },
+    {
+      "epoch": 0.9518134182677489,
+      "grad_norm": 0.03505149856209755,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 9264
+    },
+    {
+      "epoch": 0.9521216480016439,
+      "grad_norm": 0.08010539412498474,
+      "learning_rate": 0.01,
+      "loss": 2.1249,
+      "step": 9267
+    },
+    {
+      "epoch": 0.9524298777355389,
+      "grad_norm": 0.0723007321357727,
+      "learning_rate": 0.01,
+      "loss": 2.0951,
+      "step": 9270
+    },
+    {
+      "epoch": 0.9527381074694339,
+      "grad_norm": 0.05629736930131912,
+      "learning_rate": 0.01,
+      "loss": 2.0948,
+      "step": 9273
+    },
+    {
+      "epoch": 0.9530463372033289,
+      "grad_norm": 0.05514506623148918,
+      "learning_rate": 0.01,
+      "loss": 2.1214,
+      "step": 9276
+    },
+    {
+      "epoch": 0.9533545669372239,
+      "grad_norm": 0.1107834130525589,
+      "learning_rate": 0.01,
+      "loss": 2.0876,
+      "step": 9279
+    },
+    {
+      "epoch": 0.9536627966711189,
+      "grad_norm": 0.046309590339660645,
+      "learning_rate": 0.01,
+      "loss": 2.0669,
+      "step": 9282
+    },
+    {
+      "epoch": 0.9539710264050139,
+      "grad_norm": 0.06956466287374496,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 9285
+    },
+    {
+      "epoch": 0.9542792561389088,
+      "grad_norm": 0.086011603474617,
+      "learning_rate": 0.01,
+      "loss": 2.0896,
+      "step": 9288
+    },
+    {
+      "epoch": 0.9545874858728038,
+      "grad_norm": 0.04768074303865433,
+      "learning_rate": 0.01,
+      "loss": 2.0923,
+      "step": 9291
+    },
+    {
+      "epoch": 0.9548957156066988,
+      "grad_norm": 0.0958017110824585,
+      "learning_rate": 0.01,
+      "loss": 2.1134,
+      "step": 9294
+    },
+    {
+      "epoch": 0.9552039453405938,
+      "grad_norm": 0.06098558008670807,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 9297
+    },
+    {
+      "epoch": 0.9555121750744888,
+      "grad_norm": 0.05258086323738098,
+      "learning_rate": 0.01,
+      "loss": 2.0998,
+      "step": 9300
+    },
+    {
+      "epoch": 0.9558204048083838,
+      "grad_norm": 0.06664231419563293,
+      "learning_rate": 0.01,
+      "loss": 2.1215,
+      "step": 9303
+    },
+    {
+      "epoch": 0.9561286345422788,
+      "grad_norm": 0.05491488054394722,
+      "learning_rate": 0.01,
+      "loss": 2.0837,
+      "step": 9306
+    },
+    {
+      "epoch": 0.9564368642761738,
+      "grad_norm": 0.0436725877225399,
+      "learning_rate": 0.01,
+      "loss": 2.1268,
+      "step": 9309
+    },
+    {
+      "epoch": 0.9567450940100688,
+      "grad_norm": 0.08737560361623764,
+      "learning_rate": 0.01,
+      "loss": 2.0901,
+      "step": 9312
+    },
+    {
+      "epoch": 0.9570533237439638,
+      "grad_norm": 0.08130110800266266,
+      "learning_rate": 0.01,
+      "loss": 2.0766,
+      "step": 9315
+    },
+    {
+      "epoch": 0.9573615534778588,
+      "grad_norm": 0.07826768606901169,
+      "learning_rate": 0.01,
+      "loss": 2.0836,
+      "step": 9318
+    },
+    {
+      "epoch": 0.9576697832117538,
+      "grad_norm": 0.09330857545137405,
+      "learning_rate": 0.01,
+      "loss": 2.0794,
+      "step": 9321
+    },
+    {
+      "epoch": 0.9579780129456488,
+      "grad_norm": 0.03914652019739151,
+      "learning_rate": 0.01,
+      "loss": 2.102,
+      "step": 9324
+    },
+    {
+      "epoch": 0.9582862426795438,
+      "grad_norm": 0.03853154182434082,
+      "learning_rate": 0.01,
+      "loss": 2.0915,
+      "step": 9327
+    },
+    {
+      "epoch": 0.9585944724134389,
+      "grad_norm": 0.07349935919046402,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 9330
+    },
+    {
+      "epoch": 0.9589027021473339,
+      "grad_norm": 0.1473885178565979,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 9333
+    },
+    {
+      "epoch": 0.9592109318812289,
+      "grad_norm": 0.11091527342796326,
+      "learning_rate": 0.01,
+      "loss": 2.0934,
+      "step": 9336
+    },
+    {
+      "epoch": 0.9595191616151239,
+      "grad_norm": 0.0400085523724556,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 9339
+    },
+    {
+      "epoch": 0.9598273913490188,
+      "grad_norm": 0.05025499314069748,
+      "learning_rate": 0.01,
+      "loss": 2.0845,
+      "step": 9342
+    },
+    {
+      "epoch": 0.9601356210829138,
+      "grad_norm": 0.03745681792497635,
+      "learning_rate": 0.01,
+      "loss": 2.1006,
+      "step": 9345
+    },
+    {
+      "epoch": 0.9604438508168088,
+      "grad_norm": 0.05147318169474602,
+      "learning_rate": 0.01,
+      "loss": 2.0912,
+      "step": 9348
+    },
+    {
+      "epoch": 0.9607520805507038,
+      "grad_norm": 0.06338364630937576,
+      "learning_rate": 0.01,
+      "loss": 2.1169,
+      "step": 9351
+    },
+    {
+      "epoch": 0.9610603102845988,
+      "grad_norm": 0.09458258748054504,
+      "learning_rate": 0.01,
+      "loss": 2.1005,
+      "step": 9354
+    },
+    {
+      "epoch": 0.9613685400184938,
+      "grad_norm": 0.09883291274309158,
+      "learning_rate": 0.01,
+      "loss": 2.0934,
+      "step": 9357
+    },
+    {
+      "epoch": 0.9616767697523888,
+      "grad_norm": 0.048908524215221405,
+      "learning_rate": 0.01,
+      "loss": 2.0863,
+      "step": 9360
+    },
+    {
+      "epoch": 0.9619849994862838,
+      "grad_norm": 0.11762084811925888,
+      "learning_rate": 0.01,
+      "loss": 2.1182,
+      "step": 9363
+    },
+    {
+      "epoch": 0.9622932292201788,
+      "grad_norm": 0.0835133045911789,
+      "learning_rate": 0.01,
+      "loss": 2.0728,
+      "step": 9366
+    },
+    {
+      "epoch": 0.9626014589540738,
+      "grad_norm": 0.0580466203391552,
+      "learning_rate": 0.01,
+      "loss": 2.0756,
+      "step": 9369
+    },
+    {
+      "epoch": 0.9629096886879688,
+      "grad_norm": 0.051043394953012466,
+      "learning_rate": 0.01,
+      "loss": 2.0936,
+      "step": 9372
+    },
+    {
+      "epoch": 0.9632179184218638,
+      "grad_norm": 0.1081843450665474,
+      "learning_rate": 0.01,
+      "loss": 2.107,
+      "step": 9375
+    },
+    {
+      "epoch": 0.9635261481557588,
+      "grad_norm": 0.04656577482819557,
+      "learning_rate": 0.01,
+      "loss": 2.1084,
+      "step": 9378
+    },
+    {
+      "epoch": 0.9638343778896538,
+      "grad_norm": 0.03988798335194588,
+      "learning_rate": 0.01,
+      "loss": 2.1015,
+      "step": 9381
+    },
+    {
+      "epoch": 0.9641426076235488,
+      "grad_norm": 0.07686126232147217,
+      "learning_rate": 0.01,
+      "loss": 2.1417,
+      "step": 9384
+    },
+    {
+      "epoch": 0.9644508373574437,
+      "grad_norm": 0.057407401502132416,
+      "learning_rate": 0.01,
+      "loss": 2.1191,
+      "step": 9387
+    },
+    {
+      "epoch": 0.9647590670913387,
+      "grad_norm": 0.0947386845946312,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 9390
+    },
+    {
+      "epoch": 0.9650672968252337,
+      "grad_norm": 0.05064699798822403,
+      "learning_rate": 0.01,
+      "loss": 2.1001,
+      "step": 9393
+    },
+    {
+      "epoch": 0.9653755265591287,
+      "grad_norm": 0.04948986694216728,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 9396
+    },
+    {
+      "epoch": 0.9656837562930237,
+      "grad_norm": 0.10736438632011414,
+      "learning_rate": 0.01,
+      "loss": 2.0939,
+      "step": 9399
+    },
+    {
+      "epoch": 0.9659919860269187,
+      "grad_norm": 0.039317477494478226,
+      "learning_rate": 0.01,
+      "loss": 2.1077,
+      "step": 9402
+    },
+    {
+      "epoch": 0.9663002157608137,
+      "grad_norm": 0.06933067739009857,
+      "learning_rate": 0.01,
+      "loss": 2.1056,
+      "step": 9405
+    },
+    {
+      "epoch": 0.9666084454947087,
+      "grad_norm": 0.03649623692035675,
+      "learning_rate": 0.01,
+      "loss": 2.0838,
+      "step": 9408
+    },
+    {
+      "epoch": 0.9669166752286037,
+      "grad_norm": 0.09309684485197067,
+      "learning_rate": 0.01,
+      "loss": 2.0913,
+      "step": 9411
+    },
+    {
+      "epoch": 0.9672249049624987,
+      "grad_norm": 0.11532922834157944,
+      "learning_rate": 0.01,
+      "loss": 2.1127,
+      "step": 9414
+    },
+    {
+      "epoch": 0.9675331346963937,
+      "grad_norm": 0.053582970052957535,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 9417
+    },
+    {
+      "epoch": 0.9678413644302887,
+      "grad_norm": 0.07581201195716858,
+      "learning_rate": 0.01,
+      "loss": 2.1148,
+      "step": 9420
+    },
+    {
+      "epoch": 0.9681495941641837,
+      "grad_norm": 0.051002178341150284,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 9423
+    },
+    {
+      "epoch": 0.9684578238980787,
+      "grad_norm": 0.06385383754968643,
+      "learning_rate": 0.01,
+      "loss": 2.0826,
+      "step": 9426
+    },
+    {
+      "epoch": 0.9687660536319737,
+      "grad_norm": 0.10576994717121124,
+      "learning_rate": 0.01,
+      "loss": 2.0768,
+      "step": 9429
+    },
+    {
+      "epoch": 0.9690742833658686,
+      "grad_norm": 0.054983410984277725,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 9432
+    },
+    {
+      "epoch": 0.9693825130997636,
+      "grad_norm": 0.09159716218709946,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 9435
+    },
+    {
+      "epoch": 0.9696907428336586,
+      "grad_norm": 0.07718406617641449,
+      "learning_rate": 0.01,
+      "loss": 2.1132,
+      "step": 9438
+    },
+    {
+      "epoch": 0.9699989725675536,
+      "grad_norm": 0.0788009986281395,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 9441
+    },
+    {
+      "epoch": 0.9703072023014487,
+      "grad_norm": 0.040717653930187225,
+      "learning_rate": 0.01,
+      "loss": 2.084,
+      "step": 9444
+    },
+    {
+      "epoch": 0.9706154320353437,
+      "grad_norm": 0.09677381813526154,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 9447
+    },
+    {
+      "epoch": 0.9709236617692387,
+      "grad_norm": 0.0706525593996048,
+      "learning_rate": 0.01,
+      "loss": 2.0647,
+      "step": 9450
+    },
+    {
+      "epoch": 0.9712318915031337,
+      "grad_norm": 0.04624510183930397,
+      "learning_rate": 0.01,
+      "loss": 2.0818,
+      "step": 9453
+    },
+    {
+      "epoch": 0.9715401212370287,
+      "grad_norm": 0.04585500434041023,
+      "learning_rate": 0.01,
+      "loss": 2.0927,
+      "step": 9456
+    },
+    {
+      "epoch": 0.9718483509709237,
+      "grad_norm": 0.03468145430088043,
+      "learning_rate": 0.01,
+      "loss": 2.0759,
+      "step": 9459
+    },
+    {
+      "epoch": 0.9721565807048187,
+      "grad_norm": 0.06956649571657181,
+      "learning_rate": 0.01,
+      "loss": 2.092,
+      "step": 9462
+    },
+    {
+      "epoch": 0.9724648104387137,
+      "grad_norm": 0.04509080946445465,
+      "learning_rate": 0.01,
+      "loss": 2.1095,
+      "step": 9465
+    },
+    {
+      "epoch": 0.9727730401726087,
+      "grad_norm": 0.09959586709737778,
+      "learning_rate": 0.01,
+      "loss": 2.0921,
+      "step": 9468
+    },
+    {
+      "epoch": 0.9730812699065037,
+      "grad_norm": 0.08427727967500687,
+      "learning_rate": 0.01,
+      "loss": 2.1031,
+      "step": 9471
+    },
+    {
+      "epoch": 0.9733894996403987,
+      "grad_norm": 0.14798741042613983,
+      "learning_rate": 0.01,
+      "loss": 2.091,
+      "step": 9474
+    },
+    {
+      "epoch": 0.9736977293742937,
+      "grad_norm": 0.057735662907361984,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 9477
+    },
+    {
+      "epoch": 0.9740059591081887,
+      "grad_norm": 0.04484837129712105,
+      "learning_rate": 0.01,
+      "loss": 2.1015,
+      "step": 9480
+    },
+    {
+      "epoch": 0.9743141888420837,
+      "grad_norm": 0.04166285693645477,
+      "learning_rate": 0.01,
+      "loss": 2.1021,
+      "step": 9483
+    },
+    {
+      "epoch": 0.9746224185759786,
+      "grad_norm": 0.05640358105301857,
+      "learning_rate": 0.01,
+      "loss": 2.0925,
+      "step": 9486
+    },
+    {
+      "epoch": 0.9749306483098736,
+      "grad_norm": 0.040314216166734695,
+      "learning_rate": 0.01,
+      "loss": 2.0797,
+      "step": 9489
+    },
+    {
+      "epoch": 0.9752388780437686,
+      "grad_norm": 0.04522860422730446,
+      "learning_rate": 0.01,
+      "loss": 2.0935,
+      "step": 9492
+    },
+    {
+      "epoch": 0.9755471077776636,
+      "grad_norm": 0.03492886200547218,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 9495
+    },
+    {
+      "epoch": 0.9758553375115586,
+      "grad_norm": 0.03252703323960304,
+      "learning_rate": 0.01,
+      "loss": 2.125,
+      "step": 9498
+    },
+    {
+      "epoch": 0.9761635672454536,
+      "grad_norm": 0.04002056270837784,
+      "learning_rate": 0.01,
+      "loss": 2.0651,
+      "step": 9501
+    },
+    {
+      "epoch": 0.9764717969793486,
+      "grad_norm": 0.07364718616008759,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 9504
+    },
+    {
+      "epoch": 0.9767800267132436,
+      "grad_norm": 0.05577448755502701,
+      "learning_rate": 0.01,
+      "loss": 2.0726,
+      "step": 9507
+    },
+    {
+      "epoch": 0.9770882564471386,
+      "grad_norm": 0.13259132206439972,
+      "learning_rate": 0.01,
+      "loss": 2.1075,
+      "step": 9510
+    },
+    {
+      "epoch": 0.9773964861810336,
+      "grad_norm": 0.06911557912826538,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 9513
+    },
+    {
+      "epoch": 0.9777047159149286,
+      "grad_norm": 0.10592345148324966,
+      "learning_rate": 0.01,
+      "loss": 2.0982,
+      "step": 9516
+    },
+    {
+      "epoch": 0.9780129456488236,
+      "grad_norm": 0.05682144686579704,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 9519
+    },
+    {
+      "epoch": 0.9783211753827186,
+      "grad_norm": 0.07456633448600769,
+      "learning_rate": 0.01,
+      "loss": 2.0983,
+      "step": 9522
+    },
+    {
+      "epoch": 0.9786294051166136,
+      "grad_norm": 0.062031425535678864,
+      "learning_rate": 0.01,
+      "loss": 2.1163,
+      "step": 9525
+    },
+    {
+      "epoch": 0.9789376348505086,
+      "grad_norm": 0.0570233091711998,
+      "learning_rate": 0.01,
+      "loss": 2.1046,
+      "step": 9528
+    },
+    {
+      "epoch": 0.9792458645844035,
+      "grad_norm": 0.04668619483709335,
+      "learning_rate": 0.01,
+      "loss": 2.081,
+      "step": 9531
+    },
+    {
+      "epoch": 0.9795540943182985,
+      "grad_norm": 0.04718153178691864,
+      "learning_rate": 0.01,
+      "loss": 2.0678,
+      "step": 9534
+    },
+    {
+      "epoch": 0.9798623240521935,
+      "grad_norm": 0.054066251963377,
+      "learning_rate": 0.01,
+      "loss": 2.0911,
+      "step": 9537
+    },
+    {
+      "epoch": 0.9801705537860885,
+      "grad_norm": 0.1274210512638092,
+      "learning_rate": 0.01,
+      "loss": 2.097,
+      "step": 9540
+    },
+    {
+      "epoch": 0.9804787835199835,
+      "grad_norm": 0.07543773949146271,
+      "learning_rate": 0.01,
+      "loss": 2.0824,
+      "step": 9543
+    },
+    {
+      "epoch": 0.9807870132538785,
+      "grad_norm": 0.07845018804073334,
+      "learning_rate": 0.01,
+      "loss": 2.0749,
+      "step": 9546
+    },
+    {
+      "epoch": 0.9810952429877735,
+      "grad_norm": 0.08444254100322723,
+      "learning_rate": 0.01,
+      "loss": 2.1019,
+      "step": 9549
+    },
+    {
+      "epoch": 0.9814034727216685,
+      "grad_norm": 0.07719142735004425,
+      "learning_rate": 0.01,
+      "loss": 2.0811,
+      "step": 9552
+    },
+    {
+      "epoch": 0.9817117024555635,
+      "grad_norm": 0.05624673515558243,
+      "learning_rate": 0.01,
+      "loss": 2.0752,
+      "step": 9555
+    },
+    {
+      "epoch": 0.9820199321894585,
+      "grad_norm": 0.0419309176504612,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 9558
+    },
+    {
+      "epoch": 0.9823281619233536,
+      "grad_norm": 0.0343257375061512,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 9561
+    },
+    {
+      "epoch": 0.9826363916572486,
+      "grad_norm": 0.059452395886182785,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 9564
+    },
+    {
+      "epoch": 0.9829446213911436,
+      "grad_norm": 0.09073518216609955,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 9567
+    },
+    {
+      "epoch": 0.9832528511250386,
+      "grad_norm": 0.10660509765148163,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 9570
+    },
+    {
+      "epoch": 0.9835610808589336,
+      "grad_norm": 0.04380667209625244,
+      "learning_rate": 0.01,
+      "loss": 2.0992,
+      "step": 9573
+    },
+    {
+      "epoch": 0.9838693105928286,
+      "grad_norm": 0.06383811682462692,
+      "learning_rate": 0.01,
+      "loss": 2.0722,
+      "step": 9576
+    },
+    {
+      "epoch": 0.9841775403267236,
+      "grad_norm": 0.07926032692193985,
+      "learning_rate": 0.01,
+      "loss": 2.0571,
+      "step": 9579
+    },
+    {
+      "epoch": 0.9844857700606185,
+      "grad_norm": 0.05310386046767235,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 9582
+    },
+    {
+      "epoch": 0.9847939997945135,
+      "grad_norm": 0.03591843321919441,
+      "learning_rate": 0.01,
+      "loss": 2.0757,
+      "step": 9585
+    },
+    {
+      "epoch": 0.9851022295284085,
+      "grad_norm": 0.04773431271314621,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 9588
+    },
+    {
+      "epoch": 0.9854104592623035,
+      "grad_norm": 0.04679710045456886,
+      "learning_rate": 0.01,
+      "loss": 2.0771,
+      "step": 9591
+    },
+    {
+      "epoch": 0.9857186889961985,
+      "grad_norm": 0.05671774223446846,
+      "learning_rate": 0.01,
+      "loss": 2.1106,
+      "step": 9594
+    },
+    {
+      "epoch": 0.9860269187300935,
+      "grad_norm": 0.049488577991724014,
+      "learning_rate": 0.01,
+      "loss": 2.0695,
+      "step": 9597
+    },
+    {
+      "epoch": 0.9863351484639885,
+      "grad_norm": 0.04207129031419754,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 9600
+    },
+    {
+      "epoch": 0.9866433781978835,
+      "grad_norm": 0.10019747167825699,
+      "learning_rate": 0.01,
+      "loss": 2.073,
+      "step": 9603
+    },
+    {
+      "epoch": 0.9869516079317785,
+      "grad_norm": 0.051381729543209076,
+      "learning_rate": 0.01,
+      "loss": 2.0626,
+      "step": 9606
+    },
+    {
+      "epoch": 0.9872598376656735,
+      "grad_norm": 0.13477744162082672,
+      "learning_rate": 0.01,
+      "loss": 2.1098,
+      "step": 9609
+    },
+    {
+      "epoch": 0.9875680673995685,
+      "grad_norm": 0.09002148360013962,
+      "learning_rate": 0.01,
+      "loss": 2.0927,
+      "step": 9612
+    },
+    {
+      "epoch": 0.9878762971334635,
+      "grad_norm": 0.05230112001299858,
+      "learning_rate": 0.01,
+      "loss": 2.0902,
+      "step": 9615
+    },
+    {
+      "epoch": 0.9881845268673585,
+      "grad_norm": 0.0639885738492012,
+      "learning_rate": 0.01,
+      "loss": 2.1179,
+      "step": 9618
+    },
+    {
+      "epoch": 0.9884927566012535,
+      "grad_norm": 0.0553070530295372,
+      "learning_rate": 0.01,
+      "loss": 2.0923,
+      "step": 9621
+    },
+    {
+      "epoch": 0.9888009863351485,
+      "grad_norm": 0.04541468620300293,
+      "learning_rate": 0.01,
+      "loss": 2.0965,
+      "step": 9624
+    },
+    {
+      "epoch": 0.9891092160690435,
+      "grad_norm": 0.08656930178403854,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 9627
+    },
+    {
+      "epoch": 0.9894174458029384,
+      "grad_norm": 0.04954921826720238,
+      "learning_rate": 0.01,
+      "loss": 2.0759,
+      "step": 9630
+    },
+    {
+      "epoch": 0.9897256755368334,
+      "grad_norm": 0.07971720397472382,
+      "learning_rate": 0.01,
+      "loss": 2.0837,
+      "step": 9633
+    },
+    {
+      "epoch": 0.9900339052707284,
+      "grad_norm": 0.12388944625854492,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 9636
+    },
+    {
+      "epoch": 0.9903421350046234,
+      "grad_norm": 0.040693242102861404,
+      "learning_rate": 0.01,
+      "loss": 2.0806,
+      "step": 9639
+    },
+    {
+      "epoch": 0.9906503647385184,
+      "grad_norm": 0.032711997628211975,
+      "learning_rate": 0.01,
+      "loss": 2.0925,
+      "step": 9642
+    },
+    {
+      "epoch": 0.9909585944724134,
+      "grad_norm": 0.04089382663369179,
+      "learning_rate": 0.01,
+      "loss": 2.0841,
+      "step": 9645
+    },
+    {
+      "epoch": 0.9912668242063084,
+      "grad_norm": 0.05480481684207916,
+      "learning_rate": 0.01,
+      "loss": 2.0769,
+      "step": 9648
+    },
+    {
+      "epoch": 0.9915750539402034,
+      "grad_norm": 0.04627472907304764,
+      "learning_rate": 0.01,
+      "loss": 2.094,
+      "step": 9651
+    },
+    {
+      "epoch": 0.9918832836740984,
+      "grad_norm": 0.0517272874712944,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 9654
+    },
+    {
+      "epoch": 0.9921915134079934,
+      "grad_norm": 0.051012761890888214,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 9657
+    },
+    {
+      "epoch": 0.9924997431418884,
+      "grad_norm": 0.08666348457336426,
+      "learning_rate": 0.01,
+      "loss": 2.0875,
+      "step": 9660
+    },
+    {
+      "epoch": 0.9928079728757834,
+      "grad_norm": 0.0972173810005188,
+      "learning_rate": 0.01,
+      "loss": 2.0995,
+      "step": 9663
+    },
+    {
+      "epoch": 0.9931162026096784,
+      "grad_norm": 0.0765865370631218,
+      "learning_rate": 0.01,
+      "loss": 2.0729,
+      "step": 9666
+    },
+    {
+      "epoch": 0.9934244323435734,
+      "grad_norm": 0.04532674700021744,
+      "learning_rate": 0.01,
+      "loss": 2.0656,
+      "step": 9669
+    },
+    {
+      "epoch": 0.9937326620774684,
+      "grad_norm": 0.08642619848251343,
+      "learning_rate": 0.01,
+      "loss": 2.1036,
+      "step": 9672
+    },
+    {
+      "epoch": 0.9940408918113633,
+      "grad_norm": 0.04758689925074577,
+      "learning_rate": 0.01,
+      "loss": 2.0683,
+      "step": 9675
+    },
+    {
+      "epoch": 0.9943491215452585,
+      "grad_norm": 0.07701463252305984,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 9678
+    },
+    {
+      "epoch": 0.9946573512791534,
+      "grad_norm": 0.05999990925192833,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 9681
+    },
+    {
+      "epoch": 0.9949655810130484,
+      "grad_norm": 0.08793257176876068,
+      "learning_rate": 0.01,
+      "loss": 2.0689,
+      "step": 9684
+    },
+    {
+      "epoch": 0.9952738107469434,
+      "grad_norm": 0.06139199063181877,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 9687
+    },
+    {
+      "epoch": 0.9955820404808384,
+      "grad_norm": 0.09202239662408829,
+      "learning_rate": 0.01,
+      "loss": 2.0837,
+      "step": 9690
+    },
+    {
+      "epoch": 0.9958902702147334,
+      "grad_norm": 0.09284163266420364,
+      "learning_rate": 0.01,
+      "loss": 2.107,
+      "step": 9693
+    },
+    {
+      "epoch": 0.9961984999486284,
+      "grad_norm": 0.08113729953765869,
+      "learning_rate": 0.01,
+      "loss": 2.076,
+      "step": 9696
+    },
+    {
+      "epoch": 0.9965067296825234,
+      "grad_norm": 0.10663104802370071,
+      "learning_rate": 0.01,
+      "loss": 2.0973,
+      "step": 9699
+    },
+    {
+      "epoch": 0.9968149594164184,
+      "grad_norm": 0.11791951954364777,
+      "learning_rate": 0.01,
+      "loss": 2.0885,
+      "step": 9702
+    },
+    {
+      "epoch": 0.9971231891503134,
+      "grad_norm": 0.09039194136857986,
+      "learning_rate": 0.01,
+      "loss": 2.0957,
+      "step": 9705
+    },
+    {
+      "epoch": 0.9974314188842084,
+      "grad_norm": 0.08142858743667603,
+      "learning_rate": 0.01,
+      "loss": 2.0721,
+      "step": 9708
+    },
+    {
+      "epoch": 0.9977396486181034,
+      "grad_norm": 0.07347192615270615,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 9711
+    },
+    {
+      "epoch": 0.9980478783519984,
+      "grad_norm": 0.04449746012687683,
+      "learning_rate": 0.01,
+      "loss": 2.0728,
+      "step": 9714
+    },
+    {
+      "epoch": 0.9983561080858934,
+      "grad_norm": 0.040178634226322174,
+      "learning_rate": 0.01,
+      "loss": 2.0773,
+      "step": 9717
+    },
+    {
+      "epoch": 0.9986643378197884,
+      "grad_norm": 0.0577414333820343,
+      "learning_rate": 0.01,
+      "loss": 2.0854,
+      "step": 9720
+    },
+    {
+      "epoch": 0.9989725675536834,
+      "grad_norm": 0.07444582879543304,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 9723
+    },
+    {
+      "epoch": 0.9992807972875783,
+      "grad_norm": 0.10387948155403137,
+      "learning_rate": 0.01,
+      "loss": 2.0698,
+      "step": 9726
+    },
+    {
+      "epoch": 0.9995890270214733,
+      "grad_norm": 0.11066528409719467,
+      "learning_rate": 0.01,
+      "loss": 2.1035,
+      "step": 9729
+    },
+    {
+      "epoch": 0.9998972567553683,
+      "grad_norm": 0.06454616039991379,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 9732
+    },
+    {
+      "epoch": 0.9990763546798029,
+      "grad_norm": 0.048325520008802414,
+      "learning_rate": 0.01,
+      "loss": 2.1225,
+      "step": 9735
+    },
+    {
+      "epoch": 0.999384236453202,
+      "grad_norm": 0.03542228788137436,
+      "learning_rate": 0.01,
+      "loss": 2.1024,
+      "step": 9738
+    },
+    {
+      "epoch": 0.999692118226601,
+      "grad_norm": 0.042020559310913086,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 9741
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.04916913062334061,
+      "learning_rate": 0.01,
+      "loss": 2.1244,
+      "step": 9744
+    },
+    {
+      "epoch": 1.000307881773399,
+      "grad_norm": 0.08905553072690964,
+      "learning_rate": 0.01,
+      "loss": 2.0867,
+      "step": 9747
+    },
+    {
+      "epoch": 1.000615763546798,
+      "grad_norm": 0.07140953093767166,
+      "learning_rate": 0.01,
+      "loss": 2.0863,
+      "step": 9750
+    },
+    {
+      "epoch": 1.000923645320197,
+      "grad_norm": 0.05284767597913742,
+      "learning_rate": 0.01,
+      "loss": 2.1131,
+      "step": 9753
+    },
+    {
+      "epoch": 1.001231527093596,
+      "grad_norm": 0.1293289214372635,
+      "learning_rate": 0.01,
+      "loss": 2.1036,
+      "step": 9756
+    },
+    {
+      "epoch": 1.001539408866995,
+      "grad_norm": 0.06052086502313614,
+      "learning_rate": 0.01,
+      "loss": 2.1189,
+      "step": 9759
+    },
+    {
+      "epoch": 1.0018472906403941,
+      "grad_norm": 0.07361391931772232,
+      "learning_rate": 0.01,
+      "loss": 2.0962,
+      "step": 9762
+    },
+    {
+      "epoch": 1.0021551724137931,
+      "grad_norm": 0.06513562798500061,
+      "learning_rate": 0.01,
+      "loss": 2.129,
+      "step": 9765
+    },
+    {
+      "epoch": 1.0024630541871922,
+      "grad_norm": 0.036649156361818314,
+      "learning_rate": 0.01,
+      "loss": 2.0964,
+      "step": 9768
+    },
+    {
+      "epoch": 1.0027709359605912,
+      "grad_norm": 0.05371764674782753,
+      "learning_rate": 0.01,
+      "loss": 2.0976,
+      "step": 9771
+    },
+    {
+      "epoch": 1.0030788177339902,
+      "grad_norm": 0.06316730380058289,
+      "learning_rate": 0.01,
+      "loss": 2.097,
+      "step": 9774
+    },
+    {
+      "epoch": 1.0033866995073892,
+      "grad_norm": 0.03097986802458763,
+      "learning_rate": 0.01,
+      "loss": 2.1128,
+      "step": 9777
+    },
+    {
+      "epoch": 1.0036945812807883,
+      "grad_norm": 0.046021945774555206,
+      "learning_rate": 0.01,
+      "loss": 2.1296,
+      "step": 9780
+    },
+    {
+      "epoch": 1.0040024630541873,
+      "grad_norm": 0.06580191850662231,
+      "learning_rate": 0.01,
+      "loss": 2.1106,
+      "step": 9783
+    },
+    {
+      "epoch": 1.0043103448275863,
+      "grad_norm": 0.054073531180620193,
+      "learning_rate": 0.01,
+      "loss": 2.0986,
+      "step": 9786
+    },
+    {
+      "epoch": 1.0046182266009853,
+      "grad_norm": 0.10088641196489334,
+      "learning_rate": 0.01,
+      "loss": 2.1301,
+      "step": 9789
+    },
+    {
+      "epoch": 1.0049261083743843,
+      "grad_norm": 0.03944807127118111,
+      "learning_rate": 0.01,
+      "loss": 2.1337,
+      "step": 9792
+    },
+    {
+      "epoch": 1.0052339901477831,
+      "grad_norm": 0.07183028757572174,
+      "learning_rate": 0.01,
+      "loss": 2.1272,
+      "step": 9795
+    },
+    {
+      "epoch": 1.0055418719211822,
+      "grad_norm": 0.13821956515312195,
+      "learning_rate": 0.01,
+      "loss": 2.1016,
+      "step": 9798
+    },
+    {
+      "epoch": 1.0058497536945812,
+      "grad_norm": 0.14031893014907837,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 9801
+    },
+    {
+      "epoch": 1.0061576354679802,
+      "grad_norm": 0.06494525820016861,
+      "learning_rate": 0.01,
+      "loss": 2.128,
+      "step": 9804
+    },
+    {
+      "epoch": 1.0064655172413792,
+      "grad_norm": 0.05946667864918709,
+      "learning_rate": 0.01,
+      "loss": 2.1335,
+      "step": 9807
+    },
+    {
+      "epoch": 1.0067733990147782,
+      "grad_norm": 0.05583272874355316,
+      "learning_rate": 0.01,
+      "loss": 2.1186,
+      "step": 9810
+    },
+    {
+      "epoch": 1.0070812807881773,
+      "grad_norm": 0.06858284026384354,
+      "learning_rate": 0.01,
+      "loss": 2.1207,
+      "step": 9813
+    },
+    {
+      "epoch": 1.0073891625615763,
+      "grad_norm": 0.05864641070365906,
+      "learning_rate": 0.01,
+      "loss": 2.0869,
+      "step": 9816
+    },
+    {
+      "epoch": 1.0076970443349753,
+      "grad_norm": 0.043661102652549744,
+      "learning_rate": 0.01,
+      "loss": 2.1067,
+      "step": 9819
+    },
+    {
+      "epoch": 1.0080049261083743,
+      "grad_norm": 0.07878375053405762,
+      "learning_rate": 0.01,
+      "loss": 2.1149,
+      "step": 9822
+    },
+    {
+      "epoch": 1.0083128078817734,
+      "grad_norm": 0.04246210679411888,
+      "learning_rate": 0.01,
+      "loss": 2.1241,
+      "step": 9825
+    },
+    {
+      "epoch": 1.0086206896551724,
+      "grad_norm": 0.06508597731590271,
+      "learning_rate": 0.01,
+      "loss": 2.1232,
+      "step": 9828
+    },
+    {
+      "epoch": 1.0089285714285714,
+      "grad_norm": 0.07472758740186691,
+      "learning_rate": 0.01,
+      "loss": 2.0893,
+      "step": 9831
+    },
+    {
+      "epoch": 1.0092364532019704,
+      "grad_norm": 0.13144147396087646,
+      "learning_rate": 0.01,
+      "loss": 2.1194,
+      "step": 9834
+    },
+    {
+      "epoch": 1.0095443349753694,
+      "grad_norm": 0.08961367607116699,
+      "learning_rate": 0.01,
+      "loss": 2.1215,
+      "step": 9837
+    },
+    {
+      "epoch": 1.0098522167487685,
+      "grad_norm": 0.053439076989889145,
+      "learning_rate": 0.01,
+      "loss": 2.1173,
+      "step": 9840
+    },
+    {
+      "epoch": 1.0101600985221675,
+      "grad_norm": 0.03234443441033363,
+      "learning_rate": 0.01,
+      "loss": 2.104,
+      "step": 9843
+    },
+    {
+      "epoch": 1.0104679802955665,
+      "grad_norm": 0.07516933977603912,
+      "learning_rate": 0.01,
+      "loss": 2.1186,
+      "step": 9846
+    },
+    {
+      "epoch": 1.0107758620689655,
+      "grad_norm": 0.12221794575452805,
+      "learning_rate": 0.01,
+      "loss": 2.0934,
+      "step": 9849
+    },
+    {
+      "epoch": 1.0110837438423645,
+      "grad_norm": 0.08198120445013046,
+      "learning_rate": 0.01,
+      "loss": 2.1495,
+      "step": 9852
+    },
+    {
+      "epoch": 1.0113916256157636,
+      "grad_norm": 0.058380696922540665,
+      "learning_rate": 0.01,
+      "loss": 2.1234,
+      "step": 9855
+    },
+    {
+      "epoch": 1.0116995073891626,
+      "grad_norm": 0.04831172525882721,
+      "learning_rate": 0.01,
+      "loss": 2.0977,
+      "step": 9858
+    },
+    {
+      "epoch": 1.0120073891625616,
+      "grad_norm": 0.045920804142951965,
+      "learning_rate": 0.01,
+      "loss": 2.0842,
+      "step": 9861
+    },
+    {
+      "epoch": 1.0123152709359606,
+      "grad_norm": 0.12969541549682617,
+      "learning_rate": 0.01,
+      "loss": 2.1005,
+      "step": 9864
+    },
+    {
+      "epoch": 1.0126231527093597,
+      "grad_norm": 0.09659627079963684,
+      "learning_rate": 0.01,
+      "loss": 2.1126,
+      "step": 9867
+    },
+    {
+      "epoch": 1.0129310344827587,
+      "grad_norm": 0.033160608261823654,
+      "learning_rate": 0.01,
+      "loss": 2.1244,
+      "step": 9870
+    },
+    {
+      "epoch": 1.0132389162561577,
+      "grad_norm": 0.03523699939250946,
+      "learning_rate": 0.01,
+      "loss": 2.1009,
+      "step": 9873
+    },
+    {
+      "epoch": 1.0135467980295567,
+      "grad_norm": 0.04670235142111778,
+      "learning_rate": 0.01,
+      "loss": 2.1107,
+      "step": 9876
+    },
+    {
+      "epoch": 1.0138546798029557,
+      "grad_norm": 0.05278048664331436,
+      "learning_rate": 0.01,
+      "loss": 2.133,
+      "step": 9879
+    },
+    {
+      "epoch": 1.0141625615763548,
+      "grad_norm": 0.1409105509519577,
+      "learning_rate": 0.01,
+      "loss": 2.105,
+      "step": 9882
+    },
+    {
+      "epoch": 1.0144704433497538,
+      "grad_norm": 0.08208174258470535,
+      "learning_rate": 0.01,
+      "loss": 2.1202,
+      "step": 9885
+    },
+    {
+      "epoch": 1.0147783251231528,
+      "grad_norm": 0.052980221807956696,
+      "learning_rate": 0.01,
+      "loss": 2.1108,
+      "step": 9888
+    },
+    {
+      "epoch": 1.0150862068965518,
+      "grad_norm": 0.03402642160654068,
+      "learning_rate": 0.01,
+      "loss": 2.1058,
+      "step": 9891
+    },
+    {
+      "epoch": 1.0153940886699508,
+      "grad_norm": 0.05165582895278931,
+      "learning_rate": 0.01,
+      "loss": 2.0962,
+      "step": 9894
+    },
+    {
+      "epoch": 1.0157019704433496,
+      "grad_norm": 0.0488906130194664,
+      "learning_rate": 0.01,
+      "loss": 2.1157,
+      "step": 9897
+    },
+    {
+      "epoch": 1.0160098522167487,
+      "grad_norm": 0.06578544527292252,
+      "learning_rate": 0.01,
+      "loss": 2.0783,
+      "step": 9900
+    },
+    {
+      "epoch": 1.0163177339901477,
+      "grad_norm": 0.05930023267865181,
+      "learning_rate": 0.01,
+      "loss": 2.115,
+      "step": 9903
+    },
+    {
+      "epoch": 1.0166256157635467,
+      "grad_norm": 0.07461842894554138,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 9906
+    },
+    {
+      "epoch": 1.0169334975369457,
+      "grad_norm": 0.04523751139640808,
+      "learning_rate": 0.01,
+      "loss": 2.1062,
+      "step": 9909
+    },
+    {
+      "epoch": 1.0172413793103448,
+      "grad_norm": 0.05342249572277069,
+      "learning_rate": 0.01,
+      "loss": 2.1127,
+      "step": 9912
+    },
+    {
+      "epoch": 1.0175492610837438,
+      "grad_norm": 0.040748368948698044,
+      "learning_rate": 0.01,
+      "loss": 2.0893,
+      "step": 9915
+    },
+    {
+      "epoch": 1.0178571428571428,
+      "grad_norm": 0.03435824438929558,
+      "learning_rate": 0.01,
+      "loss": 2.1151,
+      "step": 9918
+    },
+    {
+      "epoch": 1.0181650246305418,
+      "grad_norm": 0.04769265651702881,
+      "learning_rate": 0.01,
+      "loss": 2.0984,
+      "step": 9921
+    },
+    {
+      "epoch": 1.0184729064039408,
+      "grad_norm": 0.07814217358827591,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 9924
+    },
+    {
+      "epoch": 1.0187807881773399,
+      "grad_norm": 0.12953363358974457,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 9927
+    },
+    {
+      "epoch": 1.0190886699507389,
+      "grad_norm": 0.11376773566007614,
+      "learning_rate": 0.01,
+      "loss": 2.0989,
+      "step": 9930
+    },
+    {
+      "epoch": 1.019396551724138,
+      "grad_norm": 0.05323106423020363,
+      "learning_rate": 0.01,
+      "loss": 2.1135,
+      "step": 9933
+    },
+    {
+      "epoch": 1.019704433497537,
+      "grad_norm": 0.07705114781856537,
+      "learning_rate": 0.01,
+      "loss": 2.1046,
+      "step": 9936
+    },
+    {
+      "epoch": 1.020012315270936,
+      "grad_norm": 0.05934451147913933,
+      "learning_rate": 0.01,
+      "loss": 2.1207,
+      "step": 9939
+    },
+    {
+      "epoch": 1.020320197044335,
+      "grad_norm": 0.10474961996078491,
+      "learning_rate": 0.01,
+      "loss": 2.1134,
+      "step": 9942
+    },
+    {
+      "epoch": 1.020628078817734,
+      "grad_norm": 0.05283385515213013,
+      "learning_rate": 0.01,
+      "loss": 2.1085,
+      "step": 9945
+    },
+    {
+      "epoch": 1.020935960591133,
+      "grad_norm": 0.043369196355342865,
+      "learning_rate": 0.01,
+      "loss": 2.1265,
+      "step": 9948
+    },
+    {
+      "epoch": 1.021243842364532,
+      "grad_norm": 0.0366055853664875,
+      "learning_rate": 0.01,
+      "loss": 2.1214,
+      "step": 9951
+    },
+    {
+      "epoch": 1.021551724137931,
+      "grad_norm": 0.06200672313570976,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 9954
+    },
+    {
+      "epoch": 1.02185960591133,
+      "grad_norm": 0.06652572005987167,
+      "learning_rate": 0.01,
+      "loss": 2.1139,
+      "step": 9957
+    },
+    {
+      "epoch": 1.022167487684729,
+      "grad_norm": 0.04040740057826042,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 9960
+    },
+    {
+      "epoch": 1.0224753694581281,
+      "grad_norm": 0.049162358045578,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 9963
+    },
+    {
+      "epoch": 1.0227832512315271,
+      "grad_norm": 0.05465700104832649,
+      "learning_rate": 0.01,
+      "loss": 2.1109,
+      "step": 9966
+    },
+    {
+      "epoch": 1.0230911330049262,
+      "grad_norm": 0.0575067512691021,
+      "learning_rate": 0.01,
+      "loss": 2.0956,
+      "step": 9969
+    },
+    {
+      "epoch": 1.0233990147783252,
+      "grad_norm": 0.14622198045253754,
+      "learning_rate": 0.01,
+      "loss": 2.1031,
+      "step": 9972
+    },
+    {
+      "epoch": 1.0237068965517242,
+      "grad_norm": 0.04765618219971657,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 9975
+    },
+    {
+      "epoch": 1.0240147783251232,
+      "grad_norm": 0.04039911553263664,
+      "learning_rate": 0.01,
+      "loss": 2.0933,
+      "step": 9978
+    },
+    {
+      "epoch": 1.0243226600985222,
+      "grad_norm": 0.06009029969573021,
+      "learning_rate": 0.01,
+      "loss": 2.115,
+      "step": 9981
+    },
+    {
+      "epoch": 1.0246305418719213,
+      "grad_norm": 0.06187298893928528,
+      "learning_rate": 0.01,
+      "loss": 2.079,
+      "step": 9984
+    },
+    {
+      "epoch": 1.0249384236453203,
+      "grad_norm": 0.05368026718497276,
+      "learning_rate": 0.01,
+      "loss": 2.0875,
+      "step": 9987
+    },
+    {
+      "epoch": 1.0252463054187193,
+      "grad_norm": 0.051921263337135315,
+      "learning_rate": 0.01,
+      "loss": 2.1243,
+      "step": 9990
+    },
+    {
+      "epoch": 1.0255541871921183,
+      "grad_norm": 0.09820009768009186,
+      "learning_rate": 0.01,
+      "loss": 2.0983,
+      "step": 9993
+    },
+    {
+      "epoch": 1.0258620689655173,
+      "grad_norm": 0.10601375997066498,
+      "learning_rate": 0.01,
+      "loss": 2.1288,
+      "step": 9996
+    },
+    {
+      "epoch": 1.0261699507389161,
+      "grad_norm": 0.05488260090351105,
+      "learning_rate": 0.01,
+      "loss": 2.1033,
+      "step": 9999
+    },
+    {
+      "epoch": 1.0264778325123152,
+      "grad_norm": 0.07482553273439407,
+      "learning_rate": 0.01,
+      "loss": 2.1181,
+      "step": 10002
+    },
+    {
+      "epoch": 1.0267857142857142,
+      "grad_norm": 0.044733475893735886,
+      "learning_rate": 0.01,
+      "loss": 2.1237,
+      "step": 10005
+    },
+    {
+      "epoch": 1.0270935960591132,
+      "grad_norm": 0.04775967076420784,
+      "learning_rate": 0.01,
+      "loss": 2.1288,
+      "step": 10008
+    },
+    {
+      "epoch": 1.0274014778325122,
+      "grad_norm": 0.05972621962428093,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 10011
+    },
+    {
+      "epoch": 1.0277093596059113,
+      "grad_norm": 0.12219330668449402,
+      "learning_rate": 0.01,
+      "loss": 2.1034,
+      "step": 10014
+    },
+    {
+      "epoch": 1.0280172413793103,
+      "grad_norm": 0.05171920731663704,
+      "learning_rate": 0.01,
+      "loss": 2.0925,
+      "step": 10017
+    },
+    {
+      "epoch": 1.0283251231527093,
+      "grad_norm": 0.04166760668158531,
+      "learning_rate": 0.01,
+      "loss": 2.0928,
+      "step": 10020
+    },
+    {
+      "epoch": 1.0286330049261083,
+      "grad_norm": 0.05231022089719772,
+      "learning_rate": 0.01,
+      "loss": 2.0945,
+      "step": 10023
+    },
+    {
+      "epoch": 1.0289408866995073,
+      "grad_norm": 0.1091604232788086,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 10026
+    },
+    {
+      "epoch": 1.0292487684729064,
+      "grad_norm": 0.07104350626468658,
+      "learning_rate": 0.01,
+      "loss": 2.1125,
+      "step": 10029
+    },
+    {
+      "epoch": 1.0295566502463054,
+      "grad_norm": 0.0466371588408947,
+      "learning_rate": 0.01,
+      "loss": 2.0973,
+      "step": 10032
+    },
+    {
+      "epoch": 1.0298645320197044,
+      "grad_norm": 0.05548730120062828,
+      "learning_rate": 0.01,
+      "loss": 2.0846,
+      "step": 10035
+    },
+    {
+      "epoch": 1.0301724137931034,
+      "grad_norm": 0.06483764201402664,
+      "learning_rate": 0.01,
+      "loss": 2.105,
+      "step": 10038
+    },
+    {
+      "epoch": 1.0304802955665024,
+      "grad_norm": 0.05243910476565361,
+      "learning_rate": 0.01,
+      "loss": 2.1011,
+      "step": 10041
+    },
+    {
+      "epoch": 1.0307881773399015,
+      "grad_norm": 0.09996815025806427,
+      "learning_rate": 0.01,
+      "loss": 2.1389,
+      "step": 10044
+    },
+    {
+      "epoch": 1.0310960591133005,
+      "grad_norm": 0.04864559695124626,
+      "learning_rate": 0.01,
+      "loss": 2.1069,
+      "step": 10047
+    },
+    {
+      "epoch": 1.0314039408866995,
+      "grad_norm": 0.14447607100009918,
+      "learning_rate": 0.01,
+      "loss": 2.1075,
+      "step": 10050
+    },
+    {
+      "epoch": 1.0317118226600985,
+      "grad_norm": 0.050261352211236954,
+      "learning_rate": 0.01,
+      "loss": 2.1147,
+      "step": 10053
+    },
+    {
+      "epoch": 1.0320197044334976,
+      "grad_norm": 0.07719244807958603,
+      "learning_rate": 0.01,
+      "loss": 2.1061,
+      "step": 10056
+    },
+    {
+      "epoch": 1.0323275862068966,
+      "grad_norm": 0.10620381683111191,
+      "learning_rate": 0.01,
+      "loss": 2.1129,
+      "step": 10059
+    },
+    {
+      "epoch": 1.0326354679802956,
+      "grad_norm": 0.05358508229255676,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 10062
+    },
+    {
+      "epoch": 1.0329433497536946,
+      "grad_norm": 0.04341145232319832,
+      "learning_rate": 0.01,
+      "loss": 2.1046,
+      "step": 10065
+    },
+    {
+      "epoch": 1.0332512315270936,
+      "grad_norm": 0.04785105213522911,
+      "learning_rate": 0.01,
+      "loss": 2.0804,
+      "step": 10068
+    },
+    {
+      "epoch": 1.0335591133004927,
+      "grad_norm": 0.04886849224567413,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 10071
+    },
+    {
+      "epoch": 1.0338669950738917,
+      "grad_norm": 0.03917735815048218,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 10074
+    },
+    {
+      "epoch": 1.0341748768472907,
+      "grad_norm": 0.10696244239807129,
+      "learning_rate": 0.01,
+      "loss": 2.085,
+      "step": 10077
+    },
+    {
+      "epoch": 1.0344827586206897,
+      "grad_norm": 0.14525163173675537,
+      "learning_rate": 0.01,
+      "loss": 2.1246,
+      "step": 10080
+    },
+    {
+      "epoch": 1.0347906403940887,
+      "grad_norm": 0.06464140862226486,
+      "learning_rate": 0.01,
+      "loss": 2.1088,
+      "step": 10083
+    },
+    {
+      "epoch": 1.0350985221674878,
+      "grad_norm": 0.055628299713134766,
+      "learning_rate": 0.01,
+      "loss": 2.1013,
+      "step": 10086
+    },
+    {
+      "epoch": 1.0354064039408868,
+      "grad_norm": 0.0457589291036129,
+      "learning_rate": 0.01,
+      "loss": 2.118,
+      "step": 10089
+    },
+    {
+      "epoch": 1.0357142857142858,
+      "grad_norm": 0.07108809798955917,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 10092
+    },
+    {
+      "epoch": 1.0360221674876848,
+      "grad_norm": 0.07304032146930695,
+      "learning_rate": 0.01,
+      "loss": 2.1632,
+      "step": 10095
+    },
+    {
+      "epoch": 1.0363300492610836,
+      "grad_norm": 0.04778844490647316,
+      "learning_rate": 0.01,
+      "loss": 2.1076,
+      "step": 10098
+    },
+    {
+      "epoch": 1.0366379310344827,
+      "grad_norm": 0.0444946251809597,
+      "learning_rate": 0.01,
+      "loss": 2.1092,
+      "step": 10101
+    },
+    {
+      "epoch": 1.0369458128078817,
+      "grad_norm": 0.03863450884819031,
+      "learning_rate": 0.01,
+      "loss": 2.0973,
+      "step": 10104
+    },
+    {
+      "epoch": 1.0372536945812807,
+      "grad_norm": 0.11049003899097443,
+      "learning_rate": 0.01,
+      "loss": 2.1069,
+      "step": 10107
+    },
+    {
+      "epoch": 1.0375615763546797,
+      "grad_norm": 0.055413637310266495,
+      "learning_rate": 0.01,
+      "loss": 2.0935,
+      "step": 10110
+    },
+    {
+      "epoch": 1.0378694581280787,
+      "grad_norm": 0.1212301105260849,
+      "learning_rate": 0.01,
+      "loss": 2.1033,
+      "step": 10113
+    },
+    {
+      "epoch": 1.0381773399014778,
+      "grad_norm": 0.06444283574819565,
+      "learning_rate": 0.01,
+      "loss": 2.0821,
+      "step": 10116
+    },
+    {
+      "epoch": 1.0384852216748768,
+      "grad_norm": 0.048522353172302246,
+      "learning_rate": 0.01,
+      "loss": 2.1129,
+      "step": 10119
+    },
+    {
+      "epoch": 1.0387931034482758,
+      "grad_norm": 0.03755674138665199,
+      "learning_rate": 0.01,
+      "loss": 2.0773,
+      "step": 10122
+    },
+    {
+      "epoch": 1.0391009852216748,
+      "grad_norm": 0.03873259574174881,
+      "learning_rate": 0.01,
+      "loss": 2.0877,
+      "step": 10125
+    },
+    {
+      "epoch": 1.0394088669950738,
+      "grad_norm": 0.062387898564338684,
+      "learning_rate": 0.01,
+      "loss": 2.1119,
+      "step": 10128
+    },
+    {
+      "epoch": 1.0397167487684729,
+      "grad_norm": 0.037559203803539276,
+      "learning_rate": 0.01,
+      "loss": 2.1165,
+      "step": 10131
+    },
+    {
+      "epoch": 1.0400246305418719,
+      "grad_norm": 0.0703917145729065,
+      "learning_rate": 0.01,
+      "loss": 2.0877,
+      "step": 10134
+    },
+    {
+      "epoch": 1.040332512315271,
+      "grad_norm": 0.05063795670866966,
+      "learning_rate": 0.01,
+      "loss": 2.1282,
+      "step": 10137
+    },
+    {
+      "epoch": 1.04064039408867,
+      "grad_norm": 0.08476493507623672,
+      "learning_rate": 0.01,
+      "loss": 2.1217,
+      "step": 10140
+    },
+    {
+      "epoch": 1.040948275862069,
+      "grad_norm": 0.09482383728027344,
+      "learning_rate": 0.01,
+      "loss": 2.1002,
+      "step": 10143
+    },
+    {
+      "epoch": 1.041256157635468,
+      "grad_norm": 0.1094396710395813,
+      "learning_rate": 0.01,
+      "loss": 2.1138,
+      "step": 10146
+    },
+    {
+      "epoch": 1.041564039408867,
+      "grad_norm": 0.17252720892429352,
+      "learning_rate": 0.01,
+      "loss": 2.1079,
+      "step": 10149
+    },
+    {
+      "epoch": 1.041871921182266,
+      "grad_norm": 0.11076754331588745,
+      "learning_rate": 0.01,
+      "loss": 2.1198,
+      "step": 10152
+    },
+    {
+      "epoch": 1.042179802955665,
+      "grad_norm": 0.06879215687513351,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 10155
+    },
+    {
+      "epoch": 1.042487684729064,
+      "grad_norm": 0.07402212172746658,
+      "learning_rate": 0.01,
+      "loss": 2.0869,
+      "step": 10158
+    },
+    {
+      "epoch": 1.042795566502463,
+      "grad_norm": 0.04562051594257355,
+      "learning_rate": 0.01,
+      "loss": 2.1139,
+      "step": 10161
+    },
+    {
+      "epoch": 1.043103448275862,
+      "grad_norm": 0.04578396677970886,
+      "learning_rate": 0.01,
+      "loss": 2.0974,
+      "step": 10164
+    },
+    {
+      "epoch": 1.0434113300492611,
+      "grad_norm": 0.051678020507097244,
+      "learning_rate": 0.01,
+      "loss": 2.0995,
+      "step": 10167
+    },
+    {
+      "epoch": 1.0437192118226601,
+      "grad_norm": 0.03445015102624893,
+      "learning_rate": 0.01,
+      "loss": 2.106,
+      "step": 10170
+    },
+    {
+      "epoch": 1.0440270935960592,
+      "grad_norm": 0.03868851810693741,
+      "learning_rate": 0.01,
+      "loss": 2.0732,
+      "step": 10173
+    },
+    {
+      "epoch": 1.0443349753694582,
+      "grad_norm": 0.058904558420181274,
+      "learning_rate": 0.01,
+      "loss": 2.085,
+      "step": 10176
+    },
+    {
+      "epoch": 1.0446428571428572,
+      "grad_norm": 0.10729484260082245,
+      "learning_rate": 0.01,
+      "loss": 2.0909,
+      "step": 10179
+    },
+    {
+      "epoch": 1.0449507389162562,
+      "grad_norm": 0.10037554055452347,
+      "learning_rate": 0.01,
+      "loss": 2.0945,
+      "step": 10182
+    },
+    {
+      "epoch": 1.0452586206896552,
+      "grad_norm": 0.07336730509996414,
+      "learning_rate": 0.01,
+      "loss": 2.0885,
+      "step": 10185
+    },
+    {
+      "epoch": 1.0455665024630543,
+      "grad_norm": 0.11717227101325989,
+      "learning_rate": 0.01,
+      "loss": 2.1019,
+      "step": 10188
+    },
+    {
+      "epoch": 1.0458743842364533,
+      "grad_norm": 0.06263696402311325,
+      "learning_rate": 0.01,
+      "loss": 2.1113,
+      "step": 10191
+    },
+    {
+      "epoch": 1.0461822660098523,
+      "grad_norm": 0.07939436286687851,
+      "learning_rate": 0.01,
+      "loss": 2.0803,
+      "step": 10194
+    },
+    {
+      "epoch": 1.0464901477832513,
+      "grad_norm": 0.05761004984378815,
+      "learning_rate": 0.01,
+      "loss": 2.1155,
+      "step": 10197
+    },
+    {
+      "epoch": 1.0467980295566504,
+      "grad_norm": 0.04293765127658844,
+      "learning_rate": 0.01,
+      "loss": 2.0944,
+      "step": 10200
+    },
+    {
+      "epoch": 1.0471059113300492,
+      "grad_norm": 0.04638001322746277,
+      "learning_rate": 0.01,
+      "loss": 2.113,
+      "step": 10203
+    },
+    {
+      "epoch": 1.0474137931034482,
+      "grad_norm": 0.047882046550512314,
+      "learning_rate": 0.01,
+      "loss": 2.0733,
+      "step": 10206
+    },
+    {
+      "epoch": 1.0477216748768472,
+      "grad_norm": 0.07461071759462357,
+      "learning_rate": 0.01,
+      "loss": 2.107,
+      "step": 10209
+    },
+    {
+      "epoch": 1.0480295566502462,
+      "grad_norm": 0.10987289249897003,
+      "learning_rate": 0.01,
+      "loss": 2.105,
+      "step": 10212
+    },
+    {
+      "epoch": 1.0483374384236452,
+      "grad_norm": 0.04183235019445419,
+      "learning_rate": 0.01,
+      "loss": 2.0953,
+      "step": 10215
+    },
+    {
+      "epoch": 1.0486453201970443,
+      "grad_norm": 0.049700990319252014,
+      "learning_rate": 0.01,
+      "loss": 2.1067,
+      "step": 10218
+    },
+    {
+      "epoch": 1.0489532019704433,
+      "grad_norm": 0.08448828011751175,
+      "learning_rate": 0.01,
+      "loss": 2.1113,
+      "step": 10221
+    },
+    {
+      "epoch": 1.0492610837438423,
+      "grad_norm": 0.05486508831381798,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 10224
+    },
+    {
+      "epoch": 1.0495689655172413,
+      "grad_norm": 0.057925377041101456,
+      "learning_rate": 0.01,
+      "loss": 2.127,
+      "step": 10227
+    },
+    {
+      "epoch": 1.0498768472906403,
+      "grad_norm": 0.05322302505373955,
+      "learning_rate": 0.01,
+      "loss": 2.0861,
+      "step": 10230
+    },
+    {
+      "epoch": 1.0501847290640394,
+      "grad_norm": 0.046823181211948395,
+      "learning_rate": 0.01,
+      "loss": 2.089,
+      "step": 10233
+    },
+    {
+      "epoch": 1.0504926108374384,
+      "grad_norm": 0.05037027224898338,
+      "learning_rate": 0.01,
+      "loss": 2.0841,
+      "step": 10236
+    },
+    {
+      "epoch": 1.0508004926108374,
+      "grad_norm": 0.05172303318977356,
+      "learning_rate": 0.01,
+      "loss": 2.105,
+      "step": 10239
+    },
+    {
+      "epoch": 1.0511083743842364,
+      "grad_norm": 0.07993052154779434,
+      "learning_rate": 0.01,
+      "loss": 2.1097,
+      "step": 10242
+    },
+    {
+      "epoch": 1.0514162561576355,
+      "grad_norm": 0.039322953671216965,
+      "learning_rate": 0.01,
+      "loss": 2.0951,
+      "step": 10245
+    },
+    {
+      "epoch": 1.0517241379310345,
+      "grad_norm": 0.05829343572258949,
+      "learning_rate": 0.01,
+      "loss": 2.1257,
+      "step": 10248
+    },
+    {
+      "epoch": 1.0520320197044335,
+      "grad_norm": 0.12303601950407028,
+      "learning_rate": 0.01,
+      "loss": 2.1143,
+      "step": 10251
+    },
+    {
+      "epoch": 1.0523399014778325,
+      "grad_norm": 0.07176418602466583,
+      "learning_rate": 0.01,
+      "loss": 2.1297,
+      "step": 10254
+    },
+    {
+      "epoch": 1.0526477832512315,
+      "grad_norm": 0.05229344964027405,
+      "learning_rate": 0.01,
+      "loss": 2.0934,
+      "step": 10257
+    },
+    {
+      "epoch": 1.0529556650246306,
+      "grad_norm": 0.041665658354759216,
+      "learning_rate": 0.01,
+      "loss": 2.116,
+      "step": 10260
+    },
+    {
+      "epoch": 1.0532635467980296,
+      "grad_norm": 0.04542261362075806,
+      "learning_rate": 0.01,
+      "loss": 2.1277,
+      "step": 10263
+    },
+    {
+      "epoch": 1.0535714285714286,
+      "grad_norm": 0.0501495897769928,
+      "learning_rate": 0.01,
+      "loss": 2.0911,
+      "step": 10266
+    },
+    {
+      "epoch": 1.0538793103448276,
+      "grad_norm": 0.06474924832582474,
+      "learning_rate": 0.01,
+      "loss": 2.1254,
+      "step": 10269
+    },
+    {
+      "epoch": 1.0541871921182266,
+      "grad_norm": 0.0736108273267746,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 10272
+    },
+    {
+      "epoch": 1.0544950738916257,
+      "grad_norm": 0.07487022131681442,
+      "learning_rate": 0.01,
+      "loss": 2.113,
+      "step": 10275
+    },
+    {
+      "epoch": 1.0548029556650247,
+      "grad_norm": 0.04876410961151123,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 10278
+    },
+    {
+      "epoch": 1.0551108374384237,
+      "grad_norm": 0.056595779955387115,
+      "learning_rate": 0.01,
+      "loss": 2.0864,
+      "step": 10281
+    },
+    {
+      "epoch": 1.0554187192118227,
+      "grad_norm": 0.06958241015672684,
+      "learning_rate": 0.01,
+      "loss": 2.1,
+      "step": 10284
+    },
+    {
+      "epoch": 1.0557266009852218,
+      "grad_norm": 0.08811846375465393,
+      "learning_rate": 0.01,
+      "loss": 2.1021,
+      "step": 10287
+    },
+    {
+      "epoch": 1.0560344827586208,
+      "grad_norm": 0.061557747423648834,
+      "learning_rate": 0.01,
+      "loss": 2.1063,
+      "step": 10290
+    },
+    {
+      "epoch": 1.0563423645320198,
+      "grad_norm": 0.07043389976024628,
+      "learning_rate": 0.01,
+      "loss": 2.106,
+      "step": 10293
+    },
+    {
+      "epoch": 1.0566502463054188,
+      "grad_norm": 0.0916379988193512,
+      "learning_rate": 0.01,
+      "loss": 2.0851,
+      "step": 10296
+    },
+    {
+      "epoch": 1.0569581280788178,
+      "grad_norm": 0.050577979534864426,
+      "learning_rate": 0.01,
+      "loss": 2.0966,
+      "step": 10299
+    },
+    {
+      "epoch": 1.0572660098522166,
+      "grad_norm": 0.06576110422611237,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 10302
+    },
+    {
+      "epoch": 1.0575738916256157,
+      "grad_norm": 0.09315023571252823,
+      "learning_rate": 0.01,
+      "loss": 2.1341,
+      "step": 10305
+    },
+    {
+      "epoch": 1.0578817733990147,
+      "grad_norm": 0.0649820864200592,
+      "learning_rate": 0.01,
+      "loss": 2.1064,
+      "step": 10308
+    },
+    {
+      "epoch": 1.0581896551724137,
+      "grad_norm": 0.07930494844913483,
+      "learning_rate": 0.01,
+      "loss": 2.107,
+      "step": 10311
+    },
+    {
+      "epoch": 1.0584975369458127,
+      "grad_norm": 0.09142257273197174,
+      "learning_rate": 0.01,
+      "loss": 2.1162,
+      "step": 10314
+    },
+    {
+      "epoch": 1.0588054187192117,
+      "grad_norm": 0.05011974647641182,
+      "learning_rate": 0.01,
+      "loss": 2.0686,
+      "step": 10317
+    },
+    {
+      "epoch": 1.0591133004926108,
+      "grad_norm": 0.1002635508775711,
+      "learning_rate": 0.01,
+      "loss": 2.138,
+      "step": 10320
+    },
+    {
+      "epoch": 1.0594211822660098,
+      "grad_norm": 0.07570278644561768,
+      "learning_rate": 0.01,
+      "loss": 2.0693,
+      "step": 10323
+    },
+    {
+      "epoch": 1.0597290640394088,
+      "grad_norm": 0.05086719989776611,
+      "learning_rate": 0.01,
+      "loss": 2.0991,
+      "step": 10326
+    },
+    {
+      "epoch": 1.0600369458128078,
+      "grad_norm": 0.03596855327486992,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 10329
+    },
+    {
+      "epoch": 1.0603448275862069,
+      "grad_norm": 0.05059434473514557,
+      "learning_rate": 0.01,
+      "loss": 2.1,
+      "step": 10332
+    },
+    {
+      "epoch": 1.0606527093596059,
+      "grad_norm": 0.058818116784095764,
+      "learning_rate": 0.01,
+      "loss": 2.0855,
+      "step": 10335
+    },
+    {
+      "epoch": 1.060960591133005,
+      "grad_norm": 0.14139403402805328,
+      "learning_rate": 0.01,
+      "loss": 2.0755,
+      "step": 10338
+    },
+    {
+      "epoch": 1.061268472906404,
+      "grad_norm": 0.12123113870620728,
+      "learning_rate": 0.01,
+      "loss": 2.0896,
+      "step": 10341
+    },
+    {
+      "epoch": 1.061576354679803,
+      "grad_norm": 0.04767270013689995,
+      "learning_rate": 0.01,
+      "loss": 2.11,
+      "step": 10344
+    },
+    {
+      "epoch": 1.061884236453202,
+      "grad_norm": 0.03506815433502197,
+      "learning_rate": 0.01,
+      "loss": 2.0953,
+      "step": 10347
+    },
+    {
+      "epoch": 1.062192118226601,
+      "grad_norm": 0.08807789534330368,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 10350
+    },
+    {
+      "epoch": 1.0625,
+      "grad_norm": 0.1130862608551979,
+      "learning_rate": 0.01,
+      "loss": 2.0888,
+      "step": 10353
+    },
+    {
+      "epoch": 1.062807881773399,
+      "grad_norm": 0.05720696598291397,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 10356
+    },
+    {
+      "epoch": 1.063115763546798,
+      "grad_norm": 0.057933416217565536,
+      "learning_rate": 0.01,
+      "loss": 2.1138,
+      "step": 10359
+    },
+    {
+      "epoch": 1.063423645320197,
+      "grad_norm": 0.056713253259658813,
+      "learning_rate": 0.01,
+      "loss": 2.0965,
+      "step": 10362
+    },
+    {
+      "epoch": 1.063731527093596,
+      "grad_norm": 0.05062280595302582,
+      "learning_rate": 0.01,
+      "loss": 2.1058,
+      "step": 10365
+    },
+    {
+      "epoch": 1.064039408866995,
+      "grad_norm": 0.03439073637127876,
+      "learning_rate": 0.01,
+      "loss": 2.0945,
+      "step": 10368
+    },
+    {
+      "epoch": 1.0643472906403941,
+      "grad_norm": 0.10244173556566238,
+      "learning_rate": 0.01,
+      "loss": 2.0916,
+      "step": 10371
+    },
+    {
+      "epoch": 1.0646551724137931,
+      "grad_norm": 0.04706069454550743,
+      "learning_rate": 0.01,
+      "loss": 2.103,
+      "step": 10374
+    },
+    {
+      "epoch": 1.0649630541871922,
+      "grad_norm": 0.11580058932304382,
+      "learning_rate": 0.01,
+      "loss": 2.0995,
+      "step": 10377
+    },
+    {
+      "epoch": 1.0652709359605912,
+      "grad_norm": 0.044736508280038834,
+      "learning_rate": 0.01,
+      "loss": 2.0906,
+      "step": 10380
+    },
+    {
+      "epoch": 1.0655788177339902,
+      "grad_norm": 0.08990567922592163,
+      "learning_rate": 0.01,
+      "loss": 2.1197,
+      "step": 10383
+    },
+    {
+      "epoch": 1.0658866995073892,
+      "grad_norm": 0.06923419237136841,
+      "learning_rate": 0.01,
+      "loss": 2.0997,
+      "step": 10386
+    },
+    {
+      "epoch": 1.0661945812807883,
+      "grad_norm": 0.059495240449905396,
+      "learning_rate": 0.01,
+      "loss": 2.1106,
+      "step": 10389
+    },
+    {
+      "epoch": 1.0665024630541873,
+      "grad_norm": 0.07906550914049149,
+      "learning_rate": 0.01,
+      "loss": 2.1196,
+      "step": 10392
+    },
+    {
+      "epoch": 1.0668103448275863,
+      "grad_norm": 0.08792297542095184,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 10395
+    },
+    {
+      "epoch": 1.0671182266009853,
+      "grad_norm": 0.06077072396874428,
+      "learning_rate": 0.01,
+      "loss": 2.088,
+      "step": 10398
+    },
+    {
+      "epoch": 1.0674261083743843,
+      "grad_norm": 0.03865751996636391,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 10401
+    },
+    {
+      "epoch": 1.0677339901477834,
+      "grad_norm": 0.03158612549304962,
+      "learning_rate": 0.01,
+      "loss": 2.0861,
+      "step": 10404
+    },
+    {
+      "epoch": 1.0680418719211822,
+      "grad_norm": 0.03455328568816185,
+      "learning_rate": 0.01,
+      "loss": 2.0819,
+      "step": 10407
+    },
+    {
+      "epoch": 1.0683497536945812,
+      "grad_norm": 0.062100328505039215,
+      "learning_rate": 0.01,
+      "loss": 2.0967,
+      "step": 10410
+    },
+    {
+      "epoch": 1.0686576354679802,
+      "grad_norm": 0.10934283584356308,
+      "learning_rate": 0.01,
+      "loss": 2.1135,
+      "step": 10413
+    },
+    {
+      "epoch": 1.0689655172413792,
+      "grad_norm": 0.07184179127216339,
+      "learning_rate": 0.01,
+      "loss": 2.0949,
+      "step": 10416
+    },
+    {
+      "epoch": 1.0692733990147782,
+      "grad_norm": 0.06610151380300522,
+      "learning_rate": 0.01,
+      "loss": 2.1001,
+      "step": 10419
+    },
+    {
+      "epoch": 1.0695812807881773,
+      "grad_norm": 0.06064629554748535,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 10422
+    },
+    {
+      "epoch": 1.0698891625615763,
+      "grad_norm": 0.0531432181596756,
+      "learning_rate": 0.01,
+      "loss": 2.1105,
+      "step": 10425
+    },
+    {
+      "epoch": 1.0701970443349753,
+      "grad_norm": 0.056448470801115036,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 10428
+    },
+    {
+      "epoch": 1.0705049261083743,
+      "grad_norm": 0.03736816346645355,
+      "learning_rate": 0.01,
+      "loss": 2.1243,
+      "step": 10431
+    },
+    {
+      "epoch": 1.0708128078817734,
+      "grad_norm": 0.12693117558956146,
+      "learning_rate": 0.01,
+      "loss": 2.1156,
+      "step": 10434
+    },
+    {
+      "epoch": 1.0711206896551724,
+      "grad_norm": 0.0428193174302578,
+      "learning_rate": 0.01,
+      "loss": 2.1025,
+      "step": 10437
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 0.0464596189558506,
+      "learning_rate": 0.01,
+      "loss": 2.1067,
+      "step": 10440
+    },
+    {
+      "epoch": 1.0717364532019704,
+      "grad_norm": 0.07535267621278763,
+      "learning_rate": 0.01,
+      "loss": 2.0785,
+      "step": 10443
+    },
+    {
+      "epoch": 1.0720443349753694,
+      "grad_norm": 0.0537327378988266,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 10446
+    },
+    {
+      "epoch": 1.0723522167487685,
+      "grad_norm": 0.03783145919442177,
+      "learning_rate": 0.01,
+      "loss": 2.0921,
+      "step": 10449
+    },
+    {
+      "epoch": 1.0726600985221675,
+      "grad_norm": 0.052689142525196075,
+      "learning_rate": 0.01,
+      "loss": 2.116,
+      "step": 10452
+    },
+    {
+      "epoch": 1.0729679802955665,
+      "grad_norm": 0.1437288373708725,
+      "learning_rate": 0.01,
+      "loss": 2.071,
+      "step": 10455
+    },
+    {
+      "epoch": 1.0732758620689655,
+      "grad_norm": 0.07633062452077866,
+      "learning_rate": 0.01,
+      "loss": 2.088,
+      "step": 10458
+    },
+    {
+      "epoch": 1.0735837438423645,
+      "grad_norm": 0.061189718544483185,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 10461
+    },
+    {
+      "epoch": 1.0738916256157636,
+      "grad_norm": 0.06256800144910812,
+      "learning_rate": 0.01,
+      "loss": 2.1056,
+      "step": 10464
+    },
+    {
+      "epoch": 1.0741995073891626,
+      "grad_norm": 0.0745188519358635,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 10467
+    },
+    {
+      "epoch": 1.0745073891625616,
+      "grad_norm": 0.0663486197590828,
+      "learning_rate": 0.01,
+      "loss": 2.0704,
+      "step": 10470
+    },
+    {
+      "epoch": 1.0748152709359606,
+      "grad_norm": 0.05472427234053612,
+      "learning_rate": 0.01,
+      "loss": 2.082,
+      "step": 10473
+    },
+    {
+      "epoch": 1.0751231527093597,
+      "grad_norm": 0.10171230137348175,
+      "learning_rate": 0.01,
+      "loss": 2.1135,
+      "step": 10476
+    },
+    {
+      "epoch": 1.0754310344827587,
+      "grad_norm": 0.05689026787877083,
+      "learning_rate": 0.01,
+      "loss": 2.0748,
+      "step": 10479
+    },
+    {
+      "epoch": 1.0757389162561577,
+      "grad_norm": 0.0593440905213356,
+      "learning_rate": 0.01,
+      "loss": 2.0922,
+      "step": 10482
+    },
+    {
+      "epoch": 1.0760467980295567,
+      "grad_norm": 0.07408995181322098,
+      "learning_rate": 0.01,
+      "loss": 2.0781,
+      "step": 10485
+    },
+    {
+      "epoch": 1.0763546798029557,
+      "grad_norm": 0.05688070133328438,
+      "learning_rate": 0.01,
+      "loss": 2.1085,
+      "step": 10488
+    },
+    {
+      "epoch": 1.0766625615763548,
+      "grad_norm": 0.05378828942775726,
+      "learning_rate": 0.01,
+      "loss": 2.1084,
+      "step": 10491
+    },
+    {
+      "epoch": 1.0769704433497538,
+      "grad_norm": 0.057735592126846313,
+      "learning_rate": 0.01,
+      "loss": 2.1023,
+      "step": 10494
+    },
+    {
+      "epoch": 1.0772783251231528,
+      "grad_norm": 0.0586666576564312,
+      "learning_rate": 0.01,
+      "loss": 2.1003,
+      "step": 10497
+    },
+    {
+      "epoch": 1.0775862068965518,
+      "grad_norm": 0.12087473273277283,
+      "learning_rate": 0.01,
+      "loss": 2.0974,
+      "step": 10500
+    },
+    {
+      "epoch": 1.0778940886699506,
+      "grad_norm": 0.07307861000299454,
+      "learning_rate": 0.01,
+      "loss": 2.0913,
+      "step": 10503
+    },
+    {
+      "epoch": 1.0782019704433496,
+      "grad_norm": 0.06621012091636658,
+      "learning_rate": 0.01,
+      "loss": 2.1173,
+      "step": 10506
+    },
+    {
+      "epoch": 1.0785098522167487,
+      "grad_norm": 0.0647876039147377,
+      "learning_rate": 0.01,
+      "loss": 2.1006,
+      "step": 10509
+    },
+    {
+      "epoch": 1.0788177339901477,
+      "grad_norm": 0.06163914501667023,
+      "learning_rate": 0.01,
+      "loss": 2.0892,
+      "step": 10512
+    },
+    {
+      "epoch": 1.0791256157635467,
+      "grad_norm": 0.04312353581190109,
+      "learning_rate": 0.01,
+      "loss": 2.0901,
+      "step": 10515
+    },
+    {
+      "epoch": 1.0794334975369457,
+      "grad_norm": 0.0760812908411026,
+      "learning_rate": 0.01,
+      "loss": 2.0995,
+      "step": 10518
+    },
+    {
+      "epoch": 1.0797413793103448,
+      "grad_norm": 0.0802140161395073,
+      "learning_rate": 0.01,
+      "loss": 2.0905,
+      "step": 10521
+    },
+    {
+      "epoch": 1.0800492610837438,
+      "grad_norm": 0.09008529782295227,
+      "learning_rate": 0.01,
+      "loss": 2.08,
+      "step": 10524
+    },
+    {
+      "epoch": 1.0803571428571428,
+      "grad_norm": 0.07469696551561356,
+      "learning_rate": 0.01,
+      "loss": 2.0725,
+      "step": 10527
+    },
+    {
+      "epoch": 1.0806650246305418,
+      "grad_norm": 0.08821582794189453,
+      "learning_rate": 0.01,
+      "loss": 2.1086,
+      "step": 10530
+    },
+    {
+      "epoch": 1.0809729064039408,
+      "grad_norm": 0.04690997302532196,
+      "learning_rate": 0.01,
+      "loss": 2.1095,
+      "step": 10533
+    },
+    {
+      "epoch": 1.0812807881773399,
+      "grad_norm": 0.04316158965229988,
+      "learning_rate": 0.01,
+      "loss": 2.0818,
+      "step": 10536
+    },
+    {
+      "epoch": 1.0815886699507389,
+      "grad_norm": 0.06996279209852219,
+      "learning_rate": 0.01,
+      "loss": 2.0993,
+      "step": 10539
+    },
+    {
+      "epoch": 1.081896551724138,
+      "grad_norm": 0.10073279589414597,
+      "learning_rate": 0.01,
+      "loss": 2.112,
+      "step": 10542
+    },
+    {
+      "epoch": 1.082204433497537,
+      "grad_norm": 0.0448322668671608,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 10545
+    },
+    {
+      "epoch": 1.082512315270936,
+      "grad_norm": 0.11411638557910919,
+      "learning_rate": 0.01,
+      "loss": 2.1082,
+      "step": 10548
+    },
+    {
+      "epoch": 1.082820197044335,
+      "grad_norm": 0.10779088735580444,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 10551
+    },
+    {
+      "epoch": 1.083128078817734,
+      "grad_norm": 0.041448626667261124,
+      "learning_rate": 0.01,
+      "loss": 2.1041,
+      "step": 10554
+    },
+    {
+      "epoch": 1.083435960591133,
+      "grad_norm": 0.07522560656070709,
+      "learning_rate": 0.01,
+      "loss": 2.0794,
+      "step": 10557
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.048221901059150696,
+      "learning_rate": 0.01,
+      "loss": 2.0936,
+      "step": 10560
+    },
+    {
+      "epoch": 1.084051724137931,
+      "grad_norm": 0.05512038618326187,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 10563
+    },
+    {
+      "epoch": 1.08435960591133,
+      "grad_norm": 0.07599300891160965,
+      "learning_rate": 0.01,
+      "loss": 2.1246,
+      "step": 10566
+    },
+    {
+      "epoch": 1.084667487684729,
+      "grad_norm": 0.06631644070148468,
+      "learning_rate": 0.01,
+      "loss": 2.0861,
+      "step": 10569
+    },
+    {
+      "epoch": 1.0849753694581281,
+      "grad_norm": 0.04972488060593605,
+      "learning_rate": 0.01,
+      "loss": 2.11,
+      "step": 10572
+    },
+    {
+      "epoch": 1.0852832512315271,
+      "grad_norm": 0.08250217139720917,
+      "learning_rate": 0.01,
+      "loss": 2.1142,
+      "step": 10575
+    },
+    {
+      "epoch": 1.0855911330049262,
+      "grad_norm": 0.09104974567890167,
+      "learning_rate": 0.01,
+      "loss": 2.0822,
+      "step": 10578
+    },
+    {
+      "epoch": 1.0858990147783252,
+      "grad_norm": 0.057310063391923904,
+      "learning_rate": 0.01,
+      "loss": 2.0819,
+      "step": 10581
+    },
+    {
+      "epoch": 1.0862068965517242,
+      "grad_norm": 0.08102291077375412,
+      "learning_rate": 0.01,
+      "loss": 2.0931,
+      "step": 10584
+    },
+    {
+      "epoch": 1.0865147783251232,
+      "grad_norm": 0.045641325414180756,
+      "learning_rate": 0.01,
+      "loss": 2.1096,
+      "step": 10587
+    },
+    {
+      "epoch": 1.0868226600985222,
+      "grad_norm": 0.05350523442029953,
+      "learning_rate": 0.01,
+      "loss": 2.1151,
+      "step": 10590
+    },
+    {
+      "epoch": 1.0871305418719213,
+      "grad_norm": 0.045734379440546036,
+      "learning_rate": 0.01,
+      "loss": 2.1043,
+      "step": 10593
+    },
+    {
+      "epoch": 1.0874384236453203,
+      "grad_norm": 0.044645924121141434,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 10596
+    },
+    {
+      "epoch": 1.0877463054187193,
+      "grad_norm": 0.046704743057489395,
+      "learning_rate": 0.01,
+      "loss": 2.0823,
+      "step": 10599
+    },
+    {
+      "epoch": 1.0880541871921183,
+      "grad_norm": 0.09600807726383209,
+      "learning_rate": 0.01,
+      "loss": 2.09,
+      "step": 10602
+    },
+    {
+      "epoch": 1.0883620689655173,
+      "grad_norm": 0.062323443591594696,
+      "learning_rate": 0.01,
+      "loss": 2.091,
+      "step": 10605
+    },
+    {
+      "epoch": 1.0886699507389164,
+      "grad_norm": 0.08459887653589249,
+      "learning_rate": 0.01,
+      "loss": 2.094,
+      "step": 10608
+    },
+    {
+      "epoch": 1.0889778325123152,
+      "grad_norm": 0.0621943362057209,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 10611
+    },
+    {
+      "epoch": 1.0892857142857142,
+      "grad_norm": 0.10963741689920425,
+      "learning_rate": 0.01,
+      "loss": 2.0769,
+      "step": 10614
+    },
+    {
+      "epoch": 1.0895935960591132,
+      "grad_norm": 0.07325689494609833,
+      "learning_rate": 0.01,
+      "loss": 2.0905,
+      "step": 10617
+    },
+    {
+      "epoch": 1.0899014778325122,
+      "grad_norm": 0.08307964354753494,
+      "learning_rate": 0.01,
+      "loss": 2.0977,
+      "step": 10620
+    },
+    {
+      "epoch": 1.0902093596059113,
+      "grad_norm": 0.18072094023227692,
+      "learning_rate": 0.01,
+      "loss": 2.1096,
+      "step": 10623
+    },
+    {
+      "epoch": 1.0905172413793103,
+      "grad_norm": 0.10427471250295639,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 10626
+    },
+    {
+      "epoch": 1.0908251231527093,
+      "grad_norm": 0.0732191875576973,
+      "learning_rate": 0.01,
+      "loss": 2.1031,
+      "step": 10629
+    },
+    {
+      "epoch": 1.0911330049261083,
+      "grad_norm": 0.03703717514872551,
+      "learning_rate": 0.01,
+      "loss": 2.0734,
+      "step": 10632
+    },
+    {
+      "epoch": 1.0914408866995073,
+      "grad_norm": 0.04907006770372391,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 10635
+    },
+    {
+      "epoch": 1.0917487684729064,
+      "grad_norm": 0.04126304015517235,
+      "learning_rate": 0.01,
+      "loss": 2.0824,
+      "step": 10638
+    },
+    {
+      "epoch": 1.0920566502463054,
+      "grad_norm": 0.04017401486635208,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 10641
+    },
+    {
+      "epoch": 1.0923645320197044,
+      "grad_norm": 0.036132264882326126,
+      "learning_rate": 0.01,
+      "loss": 2.0792,
+      "step": 10644
+    },
+    {
+      "epoch": 1.0926724137931034,
+      "grad_norm": 0.06275150179862976,
+      "learning_rate": 0.01,
+      "loss": 2.1172,
+      "step": 10647
+    },
+    {
+      "epoch": 1.0929802955665024,
+      "grad_norm": 0.08319203555583954,
+      "learning_rate": 0.01,
+      "loss": 2.0868,
+      "step": 10650
+    },
+    {
+      "epoch": 1.0932881773399015,
+      "grad_norm": 0.08663000166416168,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 10653
+    },
+    {
+      "epoch": 1.0935960591133005,
+      "grad_norm": 0.10765951871871948,
+      "learning_rate": 0.01,
+      "loss": 2.0891,
+      "step": 10656
+    },
+    {
+      "epoch": 1.0939039408866995,
+      "grad_norm": 0.035412587225437164,
+      "learning_rate": 0.01,
+      "loss": 2.0912,
+      "step": 10659
+    },
+    {
+      "epoch": 1.0942118226600985,
+      "grad_norm": 0.051735054701566696,
+      "learning_rate": 0.01,
+      "loss": 2.0986,
+      "step": 10662
+    },
+    {
+      "epoch": 1.0945197044334976,
+      "grad_norm": 0.04320614039897919,
+      "learning_rate": 0.01,
+      "loss": 2.0912,
+      "step": 10665
+    },
+    {
+      "epoch": 1.0948275862068966,
+      "grad_norm": 0.03285462409257889,
+      "learning_rate": 0.01,
+      "loss": 2.0957,
+      "step": 10668
+    },
+    {
+      "epoch": 1.0951354679802956,
+      "grad_norm": 0.05172726511955261,
+      "learning_rate": 0.01,
+      "loss": 2.0706,
+      "step": 10671
+    },
+    {
+      "epoch": 1.0954433497536946,
+      "grad_norm": 0.04941645637154579,
+      "learning_rate": 0.01,
+      "loss": 2.1018,
+      "step": 10674
+    },
+    {
+      "epoch": 1.0957512315270936,
+      "grad_norm": 0.04746576398611069,
+      "learning_rate": 0.01,
+      "loss": 2.1002,
+      "step": 10677
+    },
+    {
+      "epoch": 1.0960591133004927,
+      "grad_norm": 0.10900839418172836,
+      "learning_rate": 0.01,
+      "loss": 2.1188,
+      "step": 10680
+    },
+    {
+      "epoch": 1.0963669950738917,
+      "grad_norm": 0.06924229860305786,
+      "learning_rate": 0.01,
+      "loss": 2.097,
+      "step": 10683
+    },
+    {
+      "epoch": 1.0966748768472907,
+      "grad_norm": 0.11047599464654922,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 10686
+    },
+    {
+      "epoch": 1.0969827586206897,
+      "grad_norm": 0.10662158578634262,
+      "learning_rate": 0.01,
+      "loss": 2.078,
+      "step": 10689
+    },
+    {
+      "epoch": 1.0972906403940887,
+      "grad_norm": 0.07408568263053894,
+      "learning_rate": 0.01,
+      "loss": 2.0918,
+      "step": 10692
+    },
+    {
+      "epoch": 1.0975985221674878,
+      "grad_norm": 0.0471009686589241,
+      "learning_rate": 0.01,
+      "loss": 2.1248,
+      "step": 10695
+    },
+    {
+      "epoch": 1.0979064039408868,
+      "grad_norm": 0.049591194838285446,
+      "learning_rate": 0.01,
+      "loss": 2.082,
+      "step": 10698
+    },
+    {
+      "epoch": 1.0982142857142858,
+      "grad_norm": 0.0919683426618576,
+      "learning_rate": 0.01,
+      "loss": 2.1229,
+      "step": 10701
+    },
+    {
+      "epoch": 1.0985221674876848,
+      "grad_norm": 0.05292963236570358,
+      "learning_rate": 0.01,
+      "loss": 2.1097,
+      "step": 10704
+    },
+    {
+      "epoch": 1.0988300492610836,
+      "grad_norm": 0.053880974650382996,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 10707
+    },
+    {
+      "epoch": 1.0991379310344827,
+      "grad_norm": 0.05608196556568146,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 10710
+    },
+    {
+      "epoch": 1.0994458128078817,
+      "grad_norm": 0.06456641852855682,
+      "learning_rate": 0.01,
+      "loss": 2.1148,
+      "step": 10713
+    },
+    {
+      "epoch": 1.0997536945812807,
+      "grad_norm": 0.08165917545557022,
+      "learning_rate": 0.01,
+      "loss": 2.1199,
+      "step": 10716
+    },
+    {
+      "epoch": 1.1000615763546797,
+      "grad_norm": 0.0773044228553772,
+      "learning_rate": 0.01,
+      "loss": 2.0972,
+      "step": 10719
+    },
+    {
+      "epoch": 1.1003694581280787,
+      "grad_norm": 0.07669848203659058,
+      "learning_rate": 0.01,
+      "loss": 2.101,
+      "step": 10722
+    },
+    {
+      "epoch": 1.1006773399014778,
+      "grad_norm": 0.0773942843079567,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 10725
+    },
+    {
+      "epoch": 1.1009852216748768,
+      "grad_norm": 0.06698640435934067,
+      "learning_rate": 0.01,
+      "loss": 2.1189,
+      "step": 10728
+    },
+    {
+      "epoch": 1.1012931034482758,
+      "grad_norm": 0.098200224339962,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 10731
+    },
+    {
+      "epoch": 1.1016009852216748,
+      "grad_norm": 0.06676481664180756,
+      "learning_rate": 0.01,
+      "loss": 2.097,
+      "step": 10734
+    },
+    {
+      "epoch": 1.1019088669950738,
+      "grad_norm": 0.03925321251153946,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 10737
+    },
+    {
+      "epoch": 1.1022167487684729,
+      "grad_norm": 0.08387935161590576,
+      "learning_rate": 0.01,
+      "loss": 2.1069,
+      "step": 10740
+    },
+    {
+      "epoch": 1.1025246305418719,
+      "grad_norm": 0.06382130831480026,
+      "learning_rate": 0.01,
+      "loss": 2.091,
+      "step": 10743
+    },
+    {
+      "epoch": 1.102832512315271,
+      "grad_norm": 0.04457903653383255,
+      "learning_rate": 0.01,
+      "loss": 2.074,
+      "step": 10746
+    },
+    {
+      "epoch": 1.10314039408867,
+      "grad_norm": 0.057858239859342575,
+      "learning_rate": 0.01,
+      "loss": 2.1021,
+      "step": 10749
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.055992983281612396,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 10752
+    },
+    {
+      "epoch": 1.103756157635468,
+      "grad_norm": 0.10200835764408112,
+      "learning_rate": 0.01,
+      "loss": 2.0948,
+      "step": 10755
+    },
+    {
+      "epoch": 1.104064039408867,
+      "grad_norm": 0.11163626611232758,
+      "learning_rate": 0.01,
+      "loss": 2.0963,
+      "step": 10758
+    },
+    {
+      "epoch": 1.104371921182266,
+      "grad_norm": 0.11462046951055527,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 10761
+    },
+    {
+      "epoch": 1.104679802955665,
+      "grad_norm": 0.08823121339082718,
+      "learning_rate": 0.01,
+      "loss": 2.1136,
+      "step": 10764
+    },
+    {
+      "epoch": 1.104987684729064,
+      "grad_norm": 0.08843538910150528,
+      "learning_rate": 0.01,
+      "loss": 2.0767,
+      "step": 10767
+    },
+    {
+      "epoch": 1.105295566502463,
+      "grad_norm": 0.05961614102125168,
+      "learning_rate": 0.01,
+      "loss": 2.1067,
+      "step": 10770
+    },
+    {
+      "epoch": 1.105603448275862,
+      "grad_norm": 0.08095360547304153,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 10773
+    },
+    {
+      "epoch": 1.1059113300492611,
+      "grad_norm": 0.08094312995672226,
+      "learning_rate": 0.01,
+      "loss": 2.0849,
+      "step": 10776
+    },
+    {
+      "epoch": 1.1062192118226601,
+      "grad_norm": 0.05718453973531723,
+      "learning_rate": 0.01,
+      "loss": 2.1097,
+      "step": 10779
+    },
+    {
+      "epoch": 1.1065270935960592,
+      "grad_norm": 0.0537499338388443,
+      "learning_rate": 0.01,
+      "loss": 2.082,
+      "step": 10782
+    },
+    {
+      "epoch": 1.1068349753694582,
+      "grad_norm": 0.06437748670578003,
+      "learning_rate": 0.01,
+      "loss": 2.0982,
+      "step": 10785
+    },
+    {
+      "epoch": 1.1071428571428572,
+      "grad_norm": 0.03420199081301689,
+      "learning_rate": 0.01,
+      "loss": 2.0919,
+      "step": 10788
+    },
+    {
+      "epoch": 1.1074507389162562,
+      "grad_norm": 0.049510665237903595,
+      "learning_rate": 0.01,
+      "loss": 2.0777,
+      "step": 10791
+    },
+    {
+      "epoch": 1.1077586206896552,
+      "grad_norm": 0.044145356863737106,
+      "learning_rate": 0.01,
+      "loss": 2.0994,
+      "step": 10794
+    },
+    {
+      "epoch": 1.1080665024630543,
+      "grad_norm": 0.0494622103869915,
+      "learning_rate": 0.01,
+      "loss": 2.103,
+      "step": 10797
+    },
+    {
+      "epoch": 1.1083743842364533,
+      "grad_norm": 0.039029188454151154,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 10800
+    },
+    {
+      "epoch": 1.1086822660098523,
+      "grad_norm": 0.05786842480301857,
+      "learning_rate": 0.01,
+      "loss": 2.0767,
+      "step": 10803
+    },
+    {
+      "epoch": 1.1089901477832513,
+      "grad_norm": 0.07576561719179153,
+      "learning_rate": 0.01,
+      "loss": 2.1078,
+      "step": 10806
+    },
+    {
+      "epoch": 1.1092980295566504,
+      "grad_norm": 0.084762342274189,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 10809
+    },
+    {
+      "epoch": 1.1096059113300494,
+      "grad_norm": 0.05042179673910141,
+      "learning_rate": 0.01,
+      "loss": 2.0742,
+      "step": 10812
+    },
+    {
+      "epoch": 1.1099137931034482,
+      "grad_norm": 0.07194402068853378,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 10815
+    },
+    {
+      "epoch": 1.1102216748768472,
+      "grad_norm": 0.13966146111488342,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 10818
+    },
+    {
+      "epoch": 1.1105295566502462,
+      "grad_norm": 0.060582250356674194,
+      "learning_rate": 0.01,
+      "loss": 2.1039,
+      "step": 10821
+    },
+    {
+      "epoch": 1.1108374384236452,
+      "grad_norm": 0.03663609176874161,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 10824
+    },
+    {
+      "epoch": 1.1111453201970443,
+      "grad_norm": 0.09468091279268265,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 10827
+    },
+    {
+      "epoch": 1.1114532019704433,
+      "grad_norm": 0.07199615240097046,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 10830
+    },
+    {
+      "epoch": 1.1117610837438423,
+      "grad_norm": 0.06624965369701385,
+      "learning_rate": 0.01,
+      "loss": 2.1286,
+      "step": 10833
+    },
+    {
+      "epoch": 1.1120689655172413,
+      "grad_norm": 0.0414128340780735,
+      "learning_rate": 0.01,
+      "loss": 2.0922,
+      "step": 10836
+    },
+    {
+      "epoch": 1.1123768472906403,
+      "grad_norm": 0.06416642665863037,
+      "learning_rate": 0.01,
+      "loss": 2.0908,
+      "step": 10839
+    },
+    {
+      "epoch": 1.1126847290640394,
+      "grad_norm": 0.05309692397713661,
+      "learning_rate": 0.01,
+      "loss": 2.117,
+      "step": 10842
+    },
+    {
+      "epoch": 1.1129926108374384,
+      "grad_norm": 0.04576392099261284,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 10845
+    },
+    {
+      "epoch": 1.1133004926108374,
+      "grad_norm": 0.0887250304222107,
+      "learning_rate": 0.01,
+      "loss": 2.0815,
+      "step": 10848
+    },
+    {
+      "epoch": 1.1136083743842364,
+      "grad_norm": 0.061223480850458145,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 10851
+    },
+    {
+      "epoch": 1.1139162561576355,
+      "grad_norm": 0.12983545660972595,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 10854
+    },
+    {
+      "epoch": 1.1142241379310345,
+      "grad_norm": 0.09382637590169907,
+      "learning_rate": 0.01,
+      "loss": 2.0838,
+      "step": 10857
+    },
+    {
+      "epoch": 1.1145320197044335,
+      "grad_norm": 0.04275491461157799,
+      "learning_rate": 0.01,
+      "loss": 2.0905,
+      "step": 10860
+    },
+    {
+      "epoch": 1.1148399014778325,
+      "grad_norm": 0.044315680861473083,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 10863
+    },
+    {
+      "epoch": 1.1151477832512315,
+      "grad_norm": 0.05177663639187813,
+      "learning_rate": 0.01,
+      "loss": 2.0985,
+      "step": 10866
+    },
+    {
+      "epoch": 1.1154556650246306,
+      "grad_norm": 0.08161107450723648,
+      "learning_rate": 0.01,
+      "loss": 2.105,
+      "step": 10869
+    },
+    {
+      "epoch": 1.1157635467980296,
+      "grad_norm": 0.08273576200008392,
+      "learning_rate": 0.01,
+      "loss": 2.0991,
+      "step": 10872
+    },
+    {
+      "epoch": 1.1160714285714286,
+      "grad_norm": 0.04973771795630455,
+      "learning_rate": 0.01,
+      "loss": 2.0849,
+      "step": 10875
+    },
+    {
+      "epoch": 1.1163793103448276,
+      "grad_norm": 0.036696773022413254,
+      "learning_rate": 0.01,
+      "loss": 2.0651,
+      "step": 10878
+    },
+    {
+      "epoch": 1.1166871921182266,
+      "grad_norm": 0.03647401183843613,
+      "learning_rate": 0.01,
+      "loss": 2.0772,
+      "step": 10881
+    },
+    {
+      "epoch": 1.1169950738916257,
+      "grad_norm": 0.03360895812511444,
+      "learning_rate": 0.01,
+      "loss": 2.0952,
+      "step": 10884
+    },
+    {
+      "epoch": 1.1173029556650247,
+      "grad_norm": 0.037918057292699814,
+      "learning_rate": 0.01,
+      "loss": 2.0776,
+      "step": 10887
+    },
+    {
+      "epoch": 1.1176108374384237,
+      "grad_norm": 0.10544890910387039,
+      "learning_rate": 0.01,
+      "loss": 2.1079,
+      "step": 10890
+    },
+    {
+      "epoch": 1.1179187192118227,
+      "grad_norm": 0.15091745555400848,
+      "learning_rate": 0.01,
+      "loss": 2.1231,
+      "step": 10893
+    },
+    {
+      "epoch": 1.1182266009852218,
+      "grad_norm": 0.07386527210474014,
+      "learning_rate": 0.01,
+      "loss": 2.0922,
+      "step": 10896
+    },
+    {
+      "epoch": 1.1185344827586208,
+      "grad_norm": 0.04889804869890213,
+      "learning_rate": 0.01,
+      "loss": 2.1016,
+      "step": 10899
+    },
+    {
+      "epoch": 1.1188423645320198,
+      "grad_norm": 0.04805940017104149,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 10902
+    },
+    {
+      "epoch": 1.1191502463054188,
+      "grad_norm": 0.040073320269584656,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 10905
+    },
+    {
+      "epoch": 1.1194581280788178,
+      "grad_norm": 0.046124961227178574,
+      "learning_rate": 0.01,
+      "loss": 2.0891,
+      "step": 10908
+    },
+    {
+      "epoch": 1.1197660098522166,
+      "grad_norm": 0.04982076957821846,
+      "learning_rate": 0.01,
+      "loss": 2.0595,
+      "step": 10911
+    },
+    {
+      "epoch": 1.1200738916256157,
+      "grad_norm": 0.036569107323884964,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 10914
+    },
+    {
+      "epoch": 1.1203817733990147,
+      "grad_norm": 0.033519893884658813,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 10917
+    },
+    {
+      "epoch": 1.1206896551724137,
+      "grad_norm": 0.0513744130730629,
+      "learning_rate": 0.01,
+      "loss": 2.1119,
+      "step": 10920
+    },
+    {
+      "epoch": 1.1209975369458127,
+      "grad_norm": 0.08677095174789429,
+      "learning_rate": 0.01,
+      "loss": 2.0791,
+      "step": 10923
+    },
+    {
+      "epoch": 1.1213054187192117,
+      "grad_norm": 0.1263512223958969,
+      "learning_rate": 0.01,
+      "loss": 2.0912,
+      "step": 10926
+    },
+    {
+      "epoch": 1.1216133004926108,
+      "grad_norm": 0.0737731009721756,
+      "learning_rate": 0.01,
+      "loss": 2.1193,
+      "step": 10929
+    },
+    {
+      "epoch": 1.1219211822660098,
+      "grad_norm": 0.045122213661670685,
+      "learning_rate": 0.01,
+      "loss": 2.1029,
+      "step": 10932
+    },
+    {
+      "epoch": 1.1222290640394088,
+      "grad_norm": 0.04616571217775345,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 10935
+    },
+    {
+      "epoch": 1.1225369458128078,
+      "grad_norm": 0.03985420614480972,
+      "learning_rate": 0.01,
+      "loss": 2.0868,
+      "step": 10938
+    },
+    {
+      "epoch": 1.1228448275862069,
+      "grad_norm": 0.11042526364326477,
+      "learning_rate": 0.01,
+      "loss": 2.1057,
+      "step": 10941
+    },
+    {
+      "epoch": 1.1231527093596059,
+      "grad_norm": 0.08071359992027283,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 10944
+    },
+    {
+      "epoch": 1.123460591133005,
+      "grad_norm": 0.049534909427165985,
+      "learning_rate": 0.01,
+      "loss": 2.1055,
+      "step": 10947
+    },
+    {
+      "epoch": 1.123768472906404,
+      "grad_norm": 0.08341135829687119,
+      "learning_rate": 0.01,
+      "loss": 2.0948,
+      "step": 10950
+    },
+    {
+      "epoch": 1.124076354679803,
+      "grad_norm": 0.03842156007885933,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 10953
+    },
+    {
+      "epoch": 1.124384236453202,
+      "grad_norm": 0.04978267103433609,
+      "learning_rate": 0.01,
+      "loss": 2.0927,
+      "step": 10956
+    },
+    {
+      "epoch": 1.124692118226601,
+      "grad_norm": 0.04545191302895546,
+      "learning_rate": 0.01,
+      "loss": 2.0847,
+      "step": 10959
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 0.10103368014097214,
+      "learning_rate": 0.01,
+      "loss": 2.105,
+      "step": 10962
+    },
+    {
+      "epoch": 1.125307881773399,
+      "grad_norm": 0.05956938862800598,
+      "learning_rate": 0.01,
+      "loss": 2.1043,
+      "step": 10965
+    },
+    {
+      "epoch": 1.125615763546798,
+      "grad_norm": 0.048797741532325745,
+      "learning_rate": 0.01,
+      "loss": 2.1044,
+      "step": 10968
+    },
+    {
+      "epoch": 1.125923645320197,
+      "grad_norm": 0.041901495307683945,
+      "learning_rate": 0.01,
+      "loss": 2.0847,
+      "step": 10971
+    },
+    {
+      "epoch": 1.126231527093596,
+      "grad_norm": 0.14950989186763763,
+      "learning_rate": 0.01,
+      "loss": 2.0919,
+      "step": 10974
+    },
+    {
+      "epoch": 1.126539408866995,
+      "grad_norm": 0.049760669469833374,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 10977
+    },
+    {
+      "epoch": 1.1268472906403941,
+      "grad_norm": 0.07016187906265259,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 10980
+    },
+    {
+      "epoch": 1.1271551724137931,
+      "grad_norm": 0.057528458535671234,
+      "learning_rate": 0.01,
+      "loss": 2.0759,
+      "step": 10983
+    },
+    {
+      "epoch": 1.1274630541871922,
+      "grad_norm": 0.06690733879804611,
+      "learning_rate": 0.01,
+      "loss": 2.102,
+      "step": 10986
+    },
+    {
+      "epoch": 1.1277709359605912,
+      "grad_norm": 0.05225450173020363,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 10989
+    },
+    {
+      "epoch": 1.1280788177339902,
+      "grad_norm": 0.048363544046878815,
+      "learning_rate": 0.01,
+      "loss": 2.0634,
+      "step": 10992
+    },
+    {
+      "epoch": 1.1283866995073892,
+      "grad_norm": 0.05356382206082344,
+      "learning_rate": 0.01,
+      "loss": 2.1003,
+      "step": 10995
+    },
+    {
+      "epoch": 1.1286945812807883,
+      "grad_norm": 0.06921149045228958,
+      "learning_rate": 0.01,
+      "loss": 2.0934,
+      "step": 10998
+    },
+    {
+      "epoch": 1.1290024630541873,
+      "grad_norm": 0.04210525006055832,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 11001
+    },
+    {
+      "epoch": 1.1293103448275863,
+      "grad_norm": 0.11790584027767181,
+      "learning_rate": 0.01,
+      "loss": 2.0746,
+      "step": 11004
+    },
+    {
+      "epoch": 1.1296182266009853,
+      "grad_norm": 0.08045307546854019,
+      "learning_rate": 0.01,
+      "loss": 2.0929,
+      "step": 11007
+    },
+    {
+      "epoch": 1.1299261083743843,
+      "grad_norm": 0.10474243015050888,
+      "learning_rate": 0.01,
+      "loss": 2.1418,
+      "step": 11010
+    },
+    {
+      "epoch": 1.1302339901477834,
+      "grad_norm": 0.06073759123682976,
+      "learning_rate": 0.01,
+      "loss": 2.1089,
+      "step": 11013
+    },
+    {
+      "epoch": 1.1305418719211824,
+      "grad_norm": 0.057685475796461105,
+      "learning_rate": 0.01,
+      "loss": 2.0959,
+      "step": 11016
+    },
+    {
+      "epoch": 1.1308497536945814,
+      "grad_norm": 0.04218476638197899,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 11019
+    },
+    {
+      "epoch": 1.1311576354679802,
+      "grad_norm": 0.04814853519201279,
+      "learning_rate": 0.01,
+      "loss": 2.1134,
+      "step": 11022
+    },
+    {
+      "epoch": 1.1314655172413792,
+      "grad_norm": 0.1344536989927292,
+      "learning_rate": 0.01,
+      "loss": 2.1121,
+      "step": 11025
+    },
+    {
+      "epoch": 1.1317733990147782,
+      "grad_norm": 0.057088855654001236,
+      "learning_rate": 0.01,
+      "loss": 2.0978,
+      "step": 11028
+    },
+    {
+      "epoch": 1.1320812807881773,
+      "grad_norm": 0.04567364603281021,
+      "learning_rate": 0.01,
+      "loss": 2.0837,
+      "step": 11031
+    },
+    {
+      "epoch": 1.1323891625615763,
+      "grad_norm": 0.07506916671991348,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 11034
+    },
+    {
+      "epoch": 1.1326970443349753,
+      "grad_norm": 0.05837171897292137,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 11037
+    },
+    {
+      "epoch": 1.1330049261083743,
+      "grad_norm": 0.0457015223801136,
+      "learning_rate": 0.01,
+      "loss": 2.101,
+      "step": 11040
+    },
+    {
+      "epoch": 1.1333128078817734,
+      "grad_norm": 0.061310991644859314,
+      "learning_rate": 0.01,
+      "loss": 2.1128,
+      "step": 11043
+    },
+    {
+      "epoch": 1.1336206896551724,
+      "grad_norm": 0.05517786741256714,
+      "learning_rate": 0.01,
+      "loss": 2.0844,
+      "step": 11046
+    },
+    {
+      "epoch": 1.1339285714285714,
+      "grad_norm": 0.07835637778043747,
+      "learning_rate": 0.01,
+      "loss": 2.0996,
+      "step": 11049
+    },
+    {
+      "epoch": 1.1342364532019704,
+      "grad_norm": 0.05821641907095909,
+      "learning_rate": 0.01,
+      "loss": 2.1074,
+      "step": 11052
+    },
+    {
+      "epoch": 1.1345443349753694,
+      "grad_norm": 0.04394884407520294,
+      "learning_rate": 0.01,
+      "loss": 2.0799,
+      "step": 11055
+    },
+    {
+      "epoch": 1.1348522167487685,
+      "grad_norm": 0.05148720741271973,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 11058
+    },
+    {
+      "epoch": 1.1351600985221675,
+      "grad_norm": 0.05766841769218445,
+      "learning_rate": 0.01,
+      "loss": 2.0973,
+      "step": 11061
+    },
+    {
+      "epoch": 1.1354679802955665,
+      "grad_norm": 0.09894710779190063,
+      "learning_rate": 0.01,
+      "loss": 2.0831,
+      "step": 11064
+    },
+    {
+      "epoch": 1.1357758620689655,
+      "grad_norm": 0.11916875094175339,
+      "learning_rate": 0.01,
+      "loss": 2.1044,
+      "step": 11067
+    },
+    {
+      "epoch": 1.1360837438423645,
+      "grad_norm": 0.03926829248666763,
+      "learning_rate": 0.01,
+      "loss": 2.0866,
+      "step": 11070
+    },
+    {
+      "epoch": 1.1363916256157636,
+      "grad_norm": 0.05105220153927803,
+      "learning_rate": 0.01,
+      "loss": 2.0911,
+      "step": 11073
+    },
+    {
+      "epoch": 1.1366995073891626,
+      "grad_norm": 0.04516123607754707,
+      "learning_rate": 0.01,
+      "loss": 2.0693,
+      "step": 11076
+    },
+    {
+      "epoch": 1.1370073891625616,
+      "grad_norm": 0.046173594892024994,
+      "learning_rate": 0.01,
+      "loss": 2.092,
+      "step": 11079
+    },
+    {
+      "epoch": 1.1373152709359606,
+      "grad_norm": 0.05173357576131821,
+      "learning_rate": 0.01,
+      "loss": 2.1007,
+      "step": 11082
+    },
+    {
+      "epoch": 1.1376231527093597,
+      "grad_norm": 0.06486919522285461,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 11085
+    },
+    {
+      "epoch": 1.1379310344827587,
+      "grad_norm": 0.09763675928115845,
+      "learning_rate": 0.01,
+      "loss": 2.0942,
+      "step": 11088
+    },
+    {
+      "epoch": 1.1382389162561577,
+      "grad_norm": 0.1281820833683014,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 11091
+    },
+    {
+      "epoch": 1.1385467980295567,
+      "grad_norm": 0.05734977498650551,
+      "learning_rate": 0.01,
+      "loss": 2.0891,
+      "step": 11094
+    },
+    {
+      "epoch": 1.1388546798029557,
+      "grad_norm": 0.06809762120246887,
+      "learning_rate": 0.01,
+      "loss": 2.0855,
+      "step": 11097
+    },
+    {
+      "epoch": 1.1391625615763548,
+      "grad_norm": 0.05105281621217728,
+      "learning_rate": 0.01,
+      "loss": 2.0975,
+      "step": 11100
+    },
+    {
+      "epoch": 1.1394704433497538,
+      "grad_norm": 0.07381090521812439,
+      "learning_rate": 0.01,
+      "loss": 2.0762,
+      "step": 11103
+    },
+    {
+      "epoch": 1.1397783251231528,
+      "grad_norm": 0.050722070038318634,
+      "learning_rate": 0.01,
+      "loss": 2.0778,
+      "step": 11106
+    },
+    {
+      "epoch": 1.1400862068965516,
+      "grad_norm": 0.03850618377327919,
+      "learning_rate": 0.01,
+      "loss": 2.0626,
+      "step": 11109
+    },
+    {
+      "epoch": 1.1403940886699506,
+      "grad_norm": 0.08264841884374619,
+      "learning_rate": 0.01,
+      "loss": 2.1258,
+      "step": 11112
+    },
+    {
+      "epoch": 1.1407019704433496,
+      "grad_norm": 0.06493505835533142,
+      "learning_rate": 0.01,
+      "loss": 2.0961,
+      "step": 11115
+    },
+    {
+      "epoch": 1.1410098522167487,
+      "grad_norm": 0.06895186007022858,
+      "learning_rate": 0.01,
+      "loss": 2.1207,
+      "step": 11118
+    },
+    {
+      "epoch": 1.1413177339901477,
+      "grad_norm": 0.042232003062963486,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 11121
+    },
+    {
+      "epoch": 1.1416256157635467,
+      "grad_norm": 0.10296539217233658,
+      "learning_rate": 0.01,
+      "loss": 2.1126,
+      "step": 11124
+    },
+    {
+      "epoch": 1.1419334975369457,
+      "grad_norm": 0.043095991015434265,
+      "learning_rate": 0.01,
+      "loss": 2.0928,
+      "step": 11127
+    },
+    {
+      "epoch": 1.1422413793103448,
+      "grad_norm": 0.046020470559597015,
+      "learning_rate": 0.01,
+      "loss": 2.092,
+      "step": 11130
+    },
+    {
+      "epoch": 1.1425492610837438,
+      "grad_norm": 0.04754204675555229,
+      "learning_rate": 0.01,
+      "loss": 2.126,
+      "step": 11133
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.03732014447450638,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 11136
+    },
+    {
+      "epoch": 1.1431650246305418,
+      "grad_norm": 0.039080169051885605,
+      "learning_rate": 0.01,
+      "loss": 2.0877,
+      "step": 11139
+    },
+    {
+      "epoch": 1.1434729064039408,
+      "grad_norm": 0.04575611278414726,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 11142
+    },
+    {
+      "epoch": 1.1437807881773399,
+      "grad_norm": 0.11764515191316605,
+      "learning_rate": 0.01,
+      "loss": 2.0964,
+      "step": 11145
+    },
+    {
+      "epoch": 1.1440886699507389,
+      "grad_norm": 0.10087098181247711,
+      "learning_rate": 0.01,
+      "loss": 2.069,
+      "step": 11148
+    },
+    {
+      "epoch": 1.144396551724138,
+      "grad_norm": 0.05462269112467766,
+      "learning_rate": 0.01,
+      "loss": 2.0902,
+      "step": 11151
+    },
+    {
+      "epoch": 1.144704433497537,
+      "grad_norm": 0.06296168267726898,
+      "learning_rate": 0.01,
+      "loss": 2.1257,
+      "step": 11154
+    },
+    {
+      "epoch": 1.145012315270936,
+      "grad_norm": 0.041026949882507324,
+      "learning_rate": 0.01,
+      "loss": 2.0672,
+      "step": 11157
+    },
+    {
+      "epoch": 1.145320197044335,
+      "grad_norm": 0.05761269852519035,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 11160
+    },
+    {
+      "epoch": 1.145628078817734,
+      "grad_norm": 0.12491103261709213,
+      "learning_rate": 0.01,
+      "loss": 2.0681,
+      "step": 11163
+    },
+    {
+      "epoch": 1.145935960591133,
+      "grad_norm": 0.09269531071186066,
+      "learning_rate": 0.01,
+      "loss": 2.0618,
+      "step": 11166
+    },
+    {
+      "epoch": 1.146243842364532,
+      "grad_norm": 0.05623659864068031,
+      "learning_rate": 0.01,
+      "loss": 2.0861,
+      "step": 11169
+    },
+    {
+      "epoch": 1.146551724137931,
+      "grad_norm": 0.04075248911976814,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 11172
+    },
+    {
+      "epoch": 1.14685960591133,
+      "grad_norm": 0.04557061940431595,
+      "learning_rate": 0.01,
+      "loss": 2.0876,
+      "step": 11175
+    },
+    {
+      "epoch": 1.147167487684729,
+      "grad_norm": 0.05686535686254501,
+      "learning_rate": 0.01,
+      "loss": 2.0746,
+      "step": 11178
+    },
+    {
+      "epoch": 1.1474753694581281,
+      "grad_norm": 0.04164785146713257,
+      "learning_rate": 0.01,
+      "loss": 2.0684,
+      "step": 11181
+    },
+    {
+      "epoch": 1.1477832512315271,
+      "grad_norm": 0.05453825742006302,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 11184
+    },
+    {
+      "epoch": 1.1480911330049262,
+      "grad_norm": 0.15215769410133362,
+      "learning_rate": 0.01,
+      "loss": 2.0806,
+      "step": 11187
+    },
+    {
+      "epoch": 1.1483990147783252,
+      "grad_norm": 0.14634385704994202,
+      "learning_rate": 0.01,
+      "loss": 2.11,
+      "step": 11190
+    },
+    {
+      "epoch": 1.1487068965517242,
+      "grad_norm": 0.06655893474817276,
+      "learning_rate": 0.01,
+      "loss": 2.0953,
+      "step": 11193
+    },
+    {
+      "epoch": 1.1490147783251232,
+      "grad_norm": 0.07074826955795288,
+      "learning_rate": 0.01,
+      "loss": 2.0984,
+      "step": 11196
+    },
+    {
+      "epoch": 1.1493226600985222,
+      "grad_norm": 0.044581200927495956,
+      "learning_rate": 0.01,
+      "loss": 2.0816,
+      "step": 11199
+    },
+    {
+      "epoch": 1.1496305418719213,
+      "grad_norm": 0.06769565492868423,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 11202
+    },
+    {
+      "epoch": 1.1499384236453203,
+      "grad_norm": 0.11437363177537918,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 11205
+    },
+    {
+      "epoch": 1.1502463054187193,
+      "grad_norm": 0.040699586272239685,
+      "learning_rate": 0.01,
+      "loss": 2.1028,
+      "step": 11208
+    },
+    {
+      "epoch": 1.1505541871921183,
+      "grad_norm": 0.06996385753154755,
+      "learning_rate": 0.01,
+      "loss": 2.0755,
+      "step": 11211
+    },
+    {
+      "epoch": 1.1508620689655173,
+      "grad_norm": 0.04621044918894768,
+      "learning_rate": 0.01,
+      "loss": 2.11,
+      "step": 11214
+    },
+    {
+      "epoch": 1.1511699507389164,
+      "grad_norm": 0.07714984565973282,
+      "learning_rate": 0.01,
+      "loss": 2.1008,
+      "step": 11217
+    },
+    {
+      "epoch": 1.1514778325123154,
+      "grad_norm": 0.11194620281457901,
+      "learning_rate": 0.01,
+      "loss": 2.093,
+      "step": 11220
+    },
+    {
+      "epoch": 1.1517857142857142,
+      "grad_norm": 0.08710391074419022,
+      "learning_rate": 0.01,
+      "loss": 2.1017,
+      "step": 11223
+    },
+    {
+      "epoch": 1.1520935960591132,
+      "grad_norm": 0.06653989106416702,
+      "learning_rate": 0.01,
+      "loss": 2.1084,
+      "step": 11226
+    },
+    {
+      "epoch": 1.1524014778325122,
+      "grad_norm": 0.056591540575027466,
+      "learning_rate": 0.01,
+      "loss": 2.107,
+      "step": 11229
+    },
+    {
+      "epoch": 1.1527093596059113,
+      "grad_norm": 0.056475669145584106,
+      "learning_rate": 0.01,
+      "loss": 2.0774,
+      "step": 11232
+    },
+    {
+      "epoch": 1.1530172413793103,
+      "grad_norm": 0.08408259600400925,
+      "learning_rate": 0.01,
+      "loss": 2.1193,
+      "step": 11235
+    },
+    {
+      "epoch": 1.1533251231527093,
+      "grad_norm": 0.06853178143501282,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 11238
+    },
+    {
+      "epoch": 1.1536330049261083,
+      "grad_norm": 0.11699818074703217,
+      "learning_rate": 0.01,
+      "loss": 2.0932,
+      "step": 11241
+    },
+    {
+      "epoch": 1.1539408866995073,
+      "grad_norm": 0.06365542113780975,
+      "learning_rate": 0.01,
+      "loss": 2.079,
+      "step": 11244
+    },
+    {
+      "epoch": 1.1542487684729064,
+      "grad_norm": 0.040505316108465195,
+      "learning_rate": 0.01,
+      "loss": 2.0688,
+      "step": 11247
+    },
+    {
+      "epoch": 1.1545566502463054,
+      "grad_norm": 0.06958888471126556,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 11250
+    },
+    {
+      "epoch": 1.1548645320197044,
+      "grad_norm": 0.08968232572078705,
+      "learning_rate": 0.01,
+      "loss": 2.064,
+      "step": 11253
+    },
+    {
+      "epoch": 1.1551724137931034,
+      "grad_norm": 0.05143412947654724,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 11256
+    },
+    {
+      "epoch": 1.1554802955665024,
+      "grad_norm": 0.1219927966594696,
+      "learning_rate": 0.01,
+      "loss": 2.1038,
+      "step": 11259
+    },
+    {
+      "epoch": 1.1557881773399015,
+      "grad_norm": 0.11974184960126877,
+      "learning_rate": 0.01,
+      "loss": 2.1099,
+      "step": 11262
+    },
+    {
+      "epoch": 1.1560960591133005,
+      "grad_norm": 0.09557066112756729,
+      "learning_rate": 0.01,
+      "loss": 2.0976,
+      "step": 11265
+    },
+    {
+      "epoch": 1.1564039408866995,
+      "grad_norm": 0.10169565677642822,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 11268
+    },
+    {
+      "epoch": 1.1567118226600985,
+      "grad_norm": 0.0586596317589283,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 11271
+    },
+    {
+      "epoch": 1.1570197044334976,
+      "grad_norm": 0.0423048697412014,
+      "learning_rate": 0.01,
+      "loss": 2.1084,
+      "step": 11274
+    },
+    {
+      "epoch": 1.1573275862068966,
+      "grad_norm": 0.07394707947969437,
+      "learning_rate": 0.01,
+      "loss": 2.0802,
+      "step": 11277
+    },
+    {
+      "epoch": 1.1576354679802956,
+      "grad_norm": 0.05507487431168556,
+      "learning_rate": 0.01,
+      "loss": 2.0777,
+      "step": 11280
+    },
+    {
+      "epoch": 1.1579433497536946,
+      "grad_norm": 0.03251083940267563,
+      "learning_rate": 0.01,
+      "loss": 2.1128,
+      "step": 11283
+    },
+    {
+      "epoch": 1.1582512315270936,
+      "grad_norm": 0.04480468109250069,
+      "learning_rate": 0.01,
+      "loss": 2.1162,
+      "step": 11286
+    },
+    {
+      "epoch": 1.1585591133004927,
+      "grad_norm": 0.08275749534368515,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 11289
+    },
+    {
+      "epoch": 1.1588669950738917,
+      "grad_norm": 0.08199379593133926,
+      "learning_rate": 0.01,
+      "loss": 2.1006,
+      "step": 11292
+    },
+    {
+      "epoch": 1.1591748768472907,
+      "grad_norm": 0.1015235111117363,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 11295
+    },
+    {
+      "epoch": 1.1594827586206897,
+      "grad_norm": 0.08872919529676437,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 11298
+    },
+    {
+      "epoch": 1.1597906403940887,
+      "grad_norm": 0.04640579596161842,
+      "learning_rate": 0.01,
+      "loss": 2.09,
+      "step": 11301
+    },
+    {
+      "epoch": 1.1600985221674878,
+      "grad_norm": 0.044142261147499084,
+      "learning_rate": 0.01,
+      "loss": 2.115,
+      "step": 11304
+    },
+    {
+      "epoch": 1.1604064039408868,
+      "grad_norm": 0.1030762568116188,
+      "learning_rate": 0.01,
+      "loss": 2.1358,
+      "step": 11307
+    },
+    {
+      "epoch": 1.1607142857142858,
+      "grad_norm": 0.06712359189987183,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 11310
+    },
+    {
+      "epoch": 1.1610221674876846,
+      "grad_norm": 0.05579240992665291,
+      "learning_rate": 0.01,
+      "loss": 2.0915,
+      "step": 11313
+    },
+    {
+      "epoch": 1.1613300492610836,
+      "grad_norm": 0.04237228259444237,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 11316
+    },
+    {
+      "epoch": 1.1616379310344827,
+      "grad_norm": 0.08990642428398132,
+      "learning_rate": 0.01,
+      "loss": 2.0753,
+      "step": 11319
+    },
+    {
+      "epoch": 1.1619458128078817,
+      "grad_norm": 0.09788185358047485,
+      "learning_rate": 0.01,
+      "loss": 2.0907,
+      "step": 11322
+    },
+    {
+      "epoch": 1.1622536945812807,
+      "grad_norm": 0.07207074761390686,
+      "learning_rate": 0.01,
+      "loss": 2.0858,
+      "step": 11325
+    },
+    {
+      "epoch": 1.1625615763546797,
+      "grad_norm": 0.07704904675483704,
+      "learning_rate": 0.01,
+      "loss": 2.0938,
+      "step": 11328
+    },
+    {
+      "epoch": 1.1628694581280787,
+      "grad_norm": 0.07003269344568253,
+      "learning_rate": 0.01,
+      "loss": 2.1031,
+      "step": 11331
+    },
+    {
+      "epoch": 1.1631773399014778,
+      "grad_norm": 0.05584646388888359,
+      "learning_rate": 0.01,
+      "loss": 2.0852,
+      "step": 11334
+    },
+    {
+      "epoch": 1.1634852216748768,
+      "grad_norm": 0.10223853588104248,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 11337
+    },
+    {
+      "epoch": 1.1637931034482758,
+      "grad_norm": 0.13336369395256042,
+      "learning_rate": 0.01,
+      "loss": 2.0902,
+      "step": 11340
+    },
+    {
+      "epoch": 1.1641009852216748,
+      "grad_norm": 0.04471458122134209,
+      "learning_rate": 0.01,
+      "loss": 2.0884,
+      "step": 11343
+    },
+    {
+      "epoch": 1.1644088669950738,
+      "grad_norm": 0.04342315346002579,
+      "learning_rate": 0.01,
+      "loss": 2.0885,
+      "step": 11346
+    },
+    {
+      "epoch": 1.1647167487684729,
+      "grad_norm": 0.05172869563102722,
+      "learning_rate": 0.01,
+      "loss": 2.069,
+      "step": 11349
+    },
+    {
+      "epoch": 1.1650246305418719,
+      "grad_norm": 0.04304314777255058,
+      "learning_rate": 0.01,
+      "loss": 2.0915,
+      "step": 11352
+    },
+    {
+      "epoch": 1.165332512315271,
+      "grad_norm": 0.14592792093753815,
+      "learning_rate": 0.01,
+      "loss": 2.0841,
+      "step": 11355
+    },
+    {
+      "epoch": 1.16564039408867,
+      "grad_norm": 0.05238402634859085,
+      "learning_rate": 0.01,
+      "loss": 2.0772,
+      "step": 11358
+    },
+    {
+      "epoch": 1.165948275862069,
+      "grad_norm": 0.04800669103860855,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 11361
+    },
+    {
+      "epoch": 1.166256157635468,
+      "grad_norm": 0.06648313999176025,
+      "learning_rate": 0.01,
+      "loss": 2.0956,
+      "step": 11364
+    },
+    {
+      "epoch": 1.166564039408867,
+      "grad_norm": 0.03402326628565788,
+      "learning_rate": 0.01,
+      "loss": 2.0847,
+      "step": 11367
+    },
+    {
+      "epoch": 1.166871921182266,
+      "grad_norm": 0.05076766759157181,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 11370
+    },
+    {
+      "epoch": 1.167179802955665,
+      "grad_norm": 0.07221470773220062,
+      "learning_rate": 0.01,
+      "loss": 2.0778,
+      "step": 11373
+    },
+    {
+      "epoch": 1.167487684729064,
+      "grad_norm": 0.04556736722588539,
+      "learning_rate": 0.01,
+      "loss": 2.0957,
+      "step": 11376
+    },
+    {
+      "epoch": 1.167795566502463,
+      "grad_norm": 0.03702834993600845,
+      "learning_rate": 0.01,
+      "loss": 2.0843,
+      "step": 11379
+    },
+    {
+      "epoch": 1.168103448275862,
+      "grad_norm": 0.046527571976184845,
+      "learning_rate": 0.01,
+      "loss": 2.084,
+      "step": 11382
+    },
+    {
+      "epoch": 1.1684113300492611,
+      "grad_norm": 0.09520363062620163,
+      "learning_rate": 0.01,
+      "loss": 2.0924,
+      "step": 11385
+    },
+    {
+      "epoch": 1.1687192118226601,
+      "grad_norm": 0.12759263813495636,
+      "learning_rate": 0.01,
+      "loss": 2.1265,
+      "step": 11388
+    },
+    {
+      "epoch": 1.1690270935960592,
+      "grad_norm": 0.03981192037463188,
+      "learning_rate": 0.01,
+      "loss": 2.076,
+      "step": 11391
+    },
+    {
+      "epoch": 1.1693349753694582,
+      "grad_norm": 0.04739897698163986,
+      "learning_rate": 0.01,
+      "loss": 2.1013,
+      "step": 11394
+    },
+    {
+      "epoch": 1.1696428571428572,
+      "grad_norm": 0.04937390610575676,
+      "learning_rate": 0.01,
+      "loss": 2.0832,
+      "step": 11397
+    },
+    {
+      "epoch": 1.1699507389162562,
+      "grad_norm": 0.07097122073173523,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 11400
+    },
+    {
+      "epoch": 1.1702586206896552,
+      "grad_norm": 0.07773783802986145,
+      "learning_rate": 0.01,
+      "loss": 2.1215,
+      "step": 11403
+    },
+    {
+      "epoch": 1.1705665024630543,
+      "grad_norm": 0.04591994732618332,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 11406
+    },
+    {
+      "epoch": 1.1708743842364533,
+      "grad_norm": 0.08724237233400345,
+      "learning_rate": 0.01,
+      "loss": 2.0817,
+      "step": 11409
+    },
+    {
+      "epoch": 1.1711822660098523,
+      "grad_norm": 0.06528041511774063,
+      "learning_rate": 0.01,
+      "loss": 2.0962,
+      "step": 11412
+    },
+    {
+      "epoch": 1.1714901477832513,
+      "grad_norm": 0.0660424679517746,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 11415
+    },
+    {
+      "epoch": 1.1717980295566504,
+      "grad_norm": 0.08304266631603241,
+      "learning_rate": 0.01,
+      "loss": 2.0912,
+      "step": 11418
+    },
+    {
+      "epoch": 1.1721059113300494,
+      "grad_norm": 0.05073266103863716,
+      "learning_rate": 0.01,
+      "loss": 2.0747,
+      "step": 11421
+    },
+    {
+      "epoch": 1.1724137931034484,
+      "grad_norm": 0.07305736094713211,
+      "learning_rate": 0.01,
+      "loss": 2.0852,
+      "step": 11424
+    },
+    {
+      "epoch": 1.1727216748768472,
+      "grad_norm": 0.09365279227495193,
+      "learning_rate": 0.01,
+      "loss": 2.0883,
+      "step": 11427
+    },
+    {
+      "epoch": 1.1730295566502462,
+      "grad_norm": 0.12279067188501358,
+      "learning_rate": 0.01,
+      "loss": 2.0999,
+      "step": 11430
+    },
+    {
+      "epoch": 1.1733374384236452,
+      "grad_norm": 0.0589769072830677,
+      "learning_rate": 0.01,
+      "loss": 2.0948,
+      "step": 11433
+    },
+    {
+      "epoch": 1.1736453201970443,
+      "grad_norm": 0.06621567159891129,
+      "learning_rate": 0.01,
+      "loss": 2.0823,
+      "step": 11436
+    },
+    {
+      "epoch": 1.1739532019704433,
+      "grad_norm": 0.051341816782951355,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 11439
+    },
+    {
+      "epoch": 1.1742610837438423,
+      "grad_norm": 0.06027314066886902,
+      "learning_rate": 0.01,
+      "loss": 2.0798,
+      "step": 11442
+    },
+    {
+      "epoch": 1.1745689655172413,
+      "grad_norm": 0.10131573677062988,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 11445
+    },
+    {
+      "epoch": 1.1748768472906403,
+      "grad_norm": 0.08082377910614014,
+      "learning_rate": 0.01,
+      "loss": 2.0949,
+      "step": 11448
+    },
+    {
+      "epoch": 1.1751847290640394,
+      "grad_norm": 0.07095243781805038,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 11451
+    },
+    {
+      "epoch": 1.1754926108374384,
+      "grad_norm": 0.07132910192012787,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 11454
+    },
+    {
+      "epoch": 1.1758004926108374,
+      "grad_norm": 0.10488838702440262,
+      "learning_rate": 0.01,
+      "loss": 2.0639,
+      "step": 11457
+    },
+    {
+      "epoch": 1.1761083743842364,
+      "grad_norm": 0.12755680084228516,
+      "learning_rate": 0.01,
+      "loss": 2.0989,
+      "step": 11460
+    },
+    {
+      "epoch": 1.1764162561576355,
+      "grad_norm": 0.12174911797046661,
+      "learning_rate": 0.01,
+      "loss": 2.1026,
+      "step": 11463
+    },
+    {
+      "epoch": 1.1767241379310345,
+      "grad_norm": 0.07873964309692383,
+      "learning_rate": 0.01,
+      "loss": 2.0908,
+      "step": 11466
+    },
+    {
+      "epoch": 1.1770320197044335,
+      "grad_norm": 0.04275409132242203,
+      "learning_rate": 0.01,
+      "loss": 2.0899,
+      "step": 11469
+    },
+    {
+      "epoch": 1.1773399014778325,
+      "grad_norm": 0.046134103089571,
+      "learning_rate": 0.01,
+      "loss": 2.1064,
+      "step": 11472
+    },
+    {
+      "epoch": 1.1776477832512315,
+      "grad_norm": 0.07631804049015045,
+      "learning_rate": 0.01,
+      "loss": 2.0811,
+      "step": 11475
+    },
+    {
+      "epoch": 1.1779556650246306,
+      "grad_norm": 0.04843062162399292,
+      "learning_rate": 0.01,
+      "loss": 2.1078,
+      "step": 11478
+    },
+    {
+      "epoch": 1.1782635467980296,
+      "grad_norm": 0.04664747416973114,
+      "learning_rate": 0.01,
+      "loss": 2.0807,
+      "step": 11481
+    },
+    {
+      "epoch": 1.1785714285714286,
+      "grad_norm": 0.042328983545303345,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 11484
+    },
+    {
+      "epoch": 1.1788793103448276,
+      "grad_norm": 0.04443054646253586,
+      "learning_rate": 0.01,
+      "loss": 2.092,
+      "step": 11487
+    },
+    {
+      "epoch": 1.1791871921182266,
+      "grad_norm": 0.03439139202237129,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 11490
+    },
+    {
+      "epoch": 1.1794950738916257,
+      "grad_norm": 0.1651001274585724,
+      "learning_rate": 0.01,
+      "loss": 2.1068,
+      "step": 11493
+    },
+    {
+      "epoch": 1.1798029556650247,
+      "grad_norm": 0.04535198211669922,
+      "learning_rate": 0.01,
+      "loss": 2.0951,
+      "step": 11496
+    },
+    {
+      "epoch": 1.1801108374384237,
+      "grad_norm": 0.04346736520528793,
+      "learning_rate": 0.01,
+      "loss": 2.1032,
+      "step": 11499
+    },
+    {
+      "epoch": 1.1804187192118227,
+      "grad_norm": 0.08131600171327591,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 11502
+    },
+    {
+      "epoch": 1.1807266009852218,
+      "grad_norm": 0.0638989582657814,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 11505
+    },
+    {
+      "epoch": 1.1810344827586208,
+      "grad_norm": 0.06305757910013199,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 11508
+    },
+    {
+      "epoch": 1.1813423645320198,
+      "grad_norm": 0.048068735748529434,
+      "learning_rate": 0.01,
+      "loss": 2.0803,
+      "step": 11511
+    },
+    {
+      "epoch": 1.1816502463054186,
+      "grad_norm": 0.1859566867351532,
+      "learning_rate": 0.01,
+      "loss": 2.1206,
+      "step": 11514
+    },
+    {
+      "epoch": 1.1819581280788176,
+      "grad_norm": 0.13449640572071075,
+      "learning_rate": 0.01,
+      "loss": 2.1047,
+      "step": 11517
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.09624927490949631,
+      "learning_rate": 0.01,
+      "loss": 2.1338,
+      "step": 11520
+    },
+    {
+      "epoch": 1.1825738916256157,
+      "grad_norm": 0.04823729023337364,
+      "learning_rate": 0.01,
+      "loss": 2.1485,
+      "step": 11523
+    },
+    {
+      "epoch": 1.1828817733990147,
+      "grad_norm": 0.04005538672208786,
+      "learning_rate": 0.01,
+      "loss": 2.1062,
+      "step": 11526
+    },
+    {
+      "epoch": 1.1831896551724137,
+      "grad_norm": 0.035647980868816376,
+      "learning_rate": 0.01,
+      "loss": 2.1002,
+      "step": 11529
+    },
+    {
+      "epoch": 1.1834975369458127,
+      "grad_norm": 0.03485687077045441,
+      "learning_rate": 0.01,
+      "loss": 2.085,
+      "step": 11532
+    },
+    {
+      "epoch": 1.1838054187192117,
+      "grad_norm": 0.03271855041384697,
+      "learning_rate": 0.01,
+      "loss": 2.1083,
+      "step": 11535
+    },
+    {
+      "epoch": 1.1841133004926108,
+      "grad_norm": 0.14434824883937836,
+      "learning_rate": 0.01,
+      "loss": 2.0917,
+      "step": 11538
+    },
+    {
+      "epoch": 1.1844211822660098,
+      "grad_norm": 0.16373054683208466,
+      "learning_rate": 0.01,
+      "loss": 2.0889,
+      "step": 11541
+    },
+    {
+      "epoch": 1.1847290640394088,
+      "grad_norm": 0.1128426045179367,
+      "learning_rate": 0.01,
+      "loss": 2.0605,
+      "step": 11544
+    },
+    {
+      "epoch": 1.1850369458128078,
+      "grad_norm": 0.03807492554187775,
+      "learning_rate": 0.01,
+      "loss": 2.0792,
+      "step": 11547
+    },
+    {
+      "epoch": 1.1853448275862069,
+      "grad_norm": 0.0678490698337555,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 11550
+    },
+    {
+      "epoch": 1.1856527093596059,
+      "grad_norm": 0.04688851907849312,
+      "learning_rate": 0.01,
+      "loss": 2.0648,
+      "step": 11553
+    },
+    {
+      "epoch": 1.185960591133005,
+      "grad_norm": 0.03682132810354233,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 11556
+    },
+    {
+      "epoch": 1.186268472906404,
+      "grad_norm": 0.0797944962978363,
+      "learning_rate": 0.01,
+      "loss": 2.0632,
+      "step": 11559
+    },
+    {
+      "epoch": 1.186576354679803,
+      "grad_norm": 0.09593506157398224,
+      "learning_rate": 0.01,
+      "loss": 2.0923,
+      "step": 11562
+    },
+    {
+      "epoch": 1.186884236453202,
+      "grad_norm": 0.10455387830734253,
+      "learning_rate": 0.01,
+      "loss": 2.0927,
+      "step": 11565
+    },
+    {
+      "epoch": 1.187192118226601,
+      "grad_norm": 0.05642642080783844,
+      "learning_rate": 0.01,
+      "loss": 2.0688,
+      "step": 11568
+    },
+    {
+      "epoch": 1.1875,
+      "grad_norm": 0.09467128664255142,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 11571
+    },
+    {
+      "epoch": 1.187807881773399,
+      "grad_norm": 0.061598166823387146,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 11574
+    },
+    {
+      "epoch": 1.188115763546798,
+      "grad_norm": 0.0875246673822403,
+      "learning_rate": 0.01,
+      "loss": 2.0753,
+      "step": 11577
+    },
+    {
+      "epoch": 1.188423645320197,
+      "grad_norm": 0.05889583006501198,
+      "learning_rate": 0.01,
+      "loss": 2.0705,
+      "step": 11580
+    },
+    {
+      "epoch": 1.188731527093596,
+      "grad_norm": 0.0796559602022171,
+      "learning_rate": 0.01,
+      "loss": 2.0939,
+      "step": 11583
+    },
+    {
+      "epoch": 1.189039408866995,
+      "grad_norm": 0.04127117991447449,
+      "learning_rate": 0.01,
+      "loss": 2.0905,
+      "step": 11586
+    },
+    {
+      "epoch": 1.1893472906403941,
+      "grad_norm": 0.06161842495203018,
+      "learning_rate": 0.01,
+      "loss": 2.0954,
+      "step": 11589
+    },
+    {
+      "epoch": 1.1896551724137931,
+      "grad_norm": 0.05344879627227783,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 11592
+    },
+    {
+      "epoch": 1.1899630541871922,
+      "grad_norm": 0.03660701587796211,
+      "learning_rate": 0.01,
+      "loss": 2.0748,
+      "step": 11595
+    },
+    {
+      "epoch": 1.1902709359605912,
+      "grad_norm": 0.04792351648211479,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 11598
+    },
+    {
+      "epoch": 1.1905788177339902,
+      "grad_norm": 0.04336618259549141,
+      "learning_rate": 0.01,
+      "loss": 2.1017,
+      "step": 11601
+    },
+    {
+      "epoch": 1.1908866995073892,
+      "grad_norm": 0.0654226765036583,
+      "learning_rate": 0.01,
+      "loss": 2.0922,
+      "step": 11604
+    },
+    {
+      "epoch": 1.1911945812807883,
+      "grad_norm": 0.08879897743463516,
+      "learning_rate": 0.01,
+      "loss": 2.0842,
+      "step": 11607
+    },
+    {
+      "epoch": 1.1915024630541873,
+      "grad_norm": 0.15568268299102783,
+      "learning_rate": 0.01,
+      "loss": 2.0947,
+      "step": 11610
+    },
+    {
+      "epoch": 1.1918103448275863,
+      "grad_norm": 0.11712448298931122,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 11613
+    },
+    {
+      "epoch": 1.1921182266009853,
+      "grad_norm": 0.04966702312231064,
+      "learning_rate": 0.01,
+      "loss": 2.1093,
+      "step": 11616
+    },
+    {
+      "epoch": 1.1924261083743843,
+      "grad_norm": 0.04304838925600052,
+      "learning_rate": 0.01,
+      "loss": 2.0828,
+      "step": 11619
+    },
+    {
+      "epoch": 1.1927339901477834,
+      "grad_norm": 0.04981999844312668,
+      "learning_rate": 0.01,
+      "loss": 2.0964,
+      "step": 11622
+    },
+    {
+      "epoch": 1.1930418719211824,
+      "grad_norm": 0.045383159071207047,
+      "learning_rate": 0.01,
+      "loss": 2.0967,
+      "step": 11625
+    },
+    {
+      "epoch": 1.1933497536945814,
+      "grad_norm": 0.0348484069108963,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 11628
+    },
+    {
+      "epoch": 1.1936576354679802,
+      "grad_norm": 0.04802081733942032,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 11631
+    },
+    {
+      "epoch": 1.1939655172413792,
+      "grad_norm": 0.0615711510181427,
+      "learning_rate": 0.01,
+      "loss": 2.1153,
+      "step": 11634
+    },
+    {
+      "epoch": 1.1942733990147782,
+      "grad_norm": 0.15608100593090057,
+      "learning_rate": 0.01,
+      "loss": 2.0753,
+      "step": 11637
+    },
+    {
+      "epoch": 1.1945812807881773,
+      "grad_norm": 0.10449741780757904,
+      "learning_rate": 0.01,
+      "loss": 2.1083,
+      "step": 11640
+    },
+    {
+      "epoch": 1.1948891625615763,
+      "grad_norm": 0.062145963311195374,
+      "learning_rate": 0.01,
+      "loss": 2.1104,
+      "step": 11643
+    },
+    {
+      "epoch": 1.1951970443349753,
+      "grad_norm": 0.04744469001889229,
+      "learning_rate": 0.01,
+      "loss": 2.0742,
+      "step": 11646
+    },
+    {
+      "epoch": 1.1955049261083743,
+      "grad_norm": 0.036814477294683456,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 11649
+    },
+    {
+      "epoch": 1.1958128078817734,
+      "grad_norm": 0.037870246917009354,
+      "learning_rate": 0.01,
+      "loss": 2.0906,
+      "step": 11652
+    },
+    {
+      "epoch": 1.1961206896551724,
+      "grad_norm": 0.17372412979602814,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 11655
+    },
+    {
+      "epoch": 1.1964285714285714,
+      "grad_norm": 0.04681265726685524,
+      "learning_rate": 0.01,
+      "loss": 2.1029,
+      "step": 11658
+    },
+    {
+      "epoch": 1.1967364532019704,
+      "grad_norm": 0.058284103870391846,
+      "learning_rate": 0.01,
+      "loss": 2.0829,
+      "step": 11661
+    },
+    {
+      "epoch": 1.1970443349753694,
+      "grad_norm": 0.07531574368476868,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 11664
+    },
+    {
+      "epoch": 1.1973522167487685,
+      "grad_norm": 0.053437430411577225,
+      "learning_rate": 0.01,
+      "loss": 2.0726,
+      "step": 11667
+    },
+    {
+      "epoch": 1.1976600985221675,
+      "grad_norm": 0.047438427805900574,
+      "learning_rate": 0.01,
+      "loss": 2.107,
+      "step": 11670
+    },
+    {
+      "epoch": 1.1979679802955665,
+      "grad_norm": 0.04474404826760292,
+      "learning_rate": 0.01,
+      "loss": 2.08,
+      "step": 11673
+    },
+    {
+      "epoch": 1.1982758620689655,
+      "grad_norm": 0.15452256798744202,
+      "learning_rate": 0.01,
+      "loss": 2.0788,
+      "step": 11676
+    },
+    {
+      "epoch": 1.1985837438423645,
+      "grad_norm": 0.05446213111281395,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 11679
+    },
+    {
+      "epoch": 1.1988916256157636,
+      "grad_norm": 0.059178926050662994,
+      "learning_rate": 0.01,
+      "loss": 2.0867,
+      "step": 11682
+    },
+    {
+      "epoch": 1.1991995073891626,
+      "grad_norm": 0.05807918682694435,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 11685
+    },
+    {
+      "epoch": 1.1995073891625616,
+      "grad_norm": 0.046843890100717545,
+      "learning_rate": 0.01,
+      "loss": 2.0729,
+      "step": 11688
+    },
+    {
+      "epoch": 1.1998152709359606,
+      "grad_norm": 0.042494870722293854,
+      "learning_rate": 0.01,
+      "loss": 2.079,
+      "step": 11691
+    },
+    {
+      "epoch": 1.2001231527093597,
+      "grad_norm": 0.04506772756576538,
+      "learning_rate": 0.01,
+      "loss": 2.0862,
+      "step": 11694
+    },
+    {
+      "epoch": 1.2004310344827587,
+      "grad_norm": 0.04237942770123482,
+      "learning_rate": 0.01,
+      "loss": 2.0712,
+      "step": 11697
+    },
+    {
+      "epoch": 1.2007389162561577,
+      "grad_norm": 0.03488307446241379,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 11700
+    },
+    {
+      "epoch": 1.2010467980295567,
+      "grad_norm": 0.03693482652306557,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 11703
+    },
+    {
+      "epoch": 1.2013546798029557,
+      "grad_norm": 0.10410935431718826,
+      "learning_rate": 0.01,
+      "loss": 2.0884,
+      "step": 11706
+    },
+    {
+      "epoch": 1.2016625615763548,
+      "grad_norm": 0.11816459894180298,
+      "learning_rate": 0.01,
+      "loss": 2.0687,
+      "step": 11709
+    },
+    {
+      "epoch": 1.2019704433497538,
+      "grad_norm": 0.06567800790071487,
+      "learning_rate": 0.01,
+      "loss": 2.063,
+      "step": 11712
+    },
+    {
+      "epoch": 1.2022783251231528,
+      "grad_norm": 0.06639432907104492,
+      "learning_rate": 0.01,
+      "loss": 2.078,
+      "step": 11715
+    },
+    {
+      "epoch": 1.2025862068965516,
+      "grad_norm": 0.05059380456805229,
+      "learning_rate": 0.01,
+      "loss": 2.1163,
+      "step": 11718
+    },
+    {
+      "epoch": 1.2028940886699506,
+      "grad_norm": 0.04076917842030525,
+      "learning_rate": 0.01,
+      "loss": 2.0784,
+      "step": 11721
+    },
+    {
+      "epoch": 1.2032019704433496,
+      "grad_norm": 0.05994633212685585,
+      "learning_rate": 0.01,
+      "loss": 2.0819,
+      "step": 11724
+    },
+    {
+      "epoch": 1.2035098522167487,
+      "grad_norm": 0.05682201310992241,
+      "learning_rate": 0.01,
+      "loss": 2.0625,
+      "step": 11727
+    },
+    {
+      "epoch": 1.2038177339901477,
+      "grad_norm": 0.05393010750412941,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 11730
+    },
+    {
+      "epoch": 1.2041256157635467,
+      "grad_norm": 0.04697128012776375,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 11733
+    },
+    {
+      "epoch": 1.2044334975369457,
+      "grad_norm": 0.04945002868771553,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 11736
+    },
+    {
+      "epoch": 1.2047413793103448,
+      "grad_norm": 0.06519649177789688,
+      "learning_rate": 0.01,
+      "loss": 2.0873,
+      "step": 11739
+    },
+    {
+      "epoch": 1.2050492610837438,
+      "grad_norm": 0.1188720241189003,
+      "learning_rate": 0.01,
+      "loss": 2.0967,
+      "step": 11742
+    },
+    {
+      "epoch": 1.2053571428571428,
+      "grad_norm": 0.1045864149928093,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 11745
+    },
+    {
+      "epoch": 1.2056650246305418,
+      "grad_norm": 0.04561993479728699,
+      "learning_rate": 0.01,
+      "loss": 2.0872,
+      "step": 11748
+    },
+    {
+      "epoch": 1.2059729064039408,
+      "grad_norm": 0.04972228407859802,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 11751
+    },
+    {
+      "epoch": 1.2062807881773399,
+      "grad_norm": 0.05342618376016617,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 11754
+    },
+    {
+      "epoch": 1.2065886699507389,
+      "grad_norm": 0.05637587606906891,
+      "learning_rate": 0.01,
+      "loss": 2.0836,
+      "step": 11757
+    },
+    {
+      "epoch": 1.206896551724138,
+      "grad_norm": 0.11595457047224045,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 11760
+    },
+    {
+      "epoch": 1.207204433497537,
+      "grad_norm": 0.11803465336561203,
+      "learning_rate": 0.01,
+      "loss": 2.0741,
+      "step": 11763
+    },
+    {
+      "epoch": 1.207512315270936,
+      "grad_norm": 0.045427508652210236,
+      "learning_rate": 0.01,
+      "loss": 2.0721,
+      "step": 11766
+    },
+    {
+      "epoch": 1.207820197044335,
+      "grad_norm": 0.0365883894264698,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 11769
+    },
+    {
+      "epoch": 1.208128078817734,
+      "grad_norm": 0.03729262575507164,
+      "learning_rate": 0.01,
+      "loss": 2.087,
+      "step": 11772
+    },
+    {
+      "epoch": 1.208435960591133,
+      "grad_norm": 0.05842882767319679,
+      "learning_rate": 0.01,
+      "loss": 2.0762,
+      "step": 11775
+    },
+    {
+      "epoch": 1.208743842364532,
+      "grad_norm": 0.07687997072935104,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 11778
+    },
+    {
+      "epoch": 1.209051724137931,
+      "grad_norm": 0.06832735985517502,
+      "learning_rate": 0.01,
+      "loss": 2.0974,
+      "step": 11781
+    },
+    {
+      "epoch": 1.20935960591133,
+      "grad_norm": 0.10200455039739609,
+      "learning_rate": 0.01,
+      "loss": 2.0866,
+      "step": 11784
+    },
+    {
+      "epoch": 1.209667487684729,
+      "grad_norm": 0.10769661515951157,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 11787
+    },
+    {
+      "epoch": 1.2099753694581281,
+      "grad_norm": 0.12233126908540726,
+      "learning_rate": 0.01,
+      "loss": 2.0911,
+      "step": 11790
+    },
+    {
+      "epoch": 1.2102832512315271,
+      "grad_norm": 0.046646762639284134,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 11793
+    },
+    {
+      "epoch": 1.2105911330049262,
+      "grad_norm": 0.030627859756350517,
+      "learning_rate": 0.01,
+      "loss": 2.0805,
+      "step": 11796
+    },
+    {
+      "epoch": 1.2108990147783252,
+      "grad_norm": 0.03977693244814873,
+      "learning_rate": 0.01,
+      "loss": 2.0465,
+      "step": 11799
+    },
+    {
+      "epoch": 1.2112068965517242,
+      "grad_norm": 0.06213162839412689,
+      "learning_rate": 0.01,
+      "loss": 2.1123,
+      "step": 11802
+    },
+    {
+      "epoch": 1.2115147783251232,
+      "grad_norm": 0.04708317294716835,
+      "learning_rate": 0.01,
+      "loss": 2.1315,
+      "step": 11805
+    },
+    {
+      "epoch": 1.2118226600985222,
+      "grad_norm": 0.10807790607213974,
+      "learning_rate": 0.01,
+      "loss": 2.0825,
+      "step": 11808
+    },
+    {
+      "epoch": 1.2121305418719213,
+      "grad_norm": 0.08579280972480774,
+      "learning_rate": 0.01,
+      "loss": 2.1366,
+      "step": 11811
+    },
+    {
+      "epoch": 1.2124384236453203,
+      "grad_norm": 0.05783751606941223,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 11814
+    },
+    {
+      "epoch": 1.2127463054187193,
+      "grad_norm": 0.04836808145046234,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 11817
+    },
+    {
+      "epoch": 1.2130541871921183,
+      "grad_norm": 0.04026191681623459,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 11820
+    },
+    {
+      "epoch": 1.2133620689655173,
+      "grad_norm": 0.03309273347258568,
+      "learning_rate": 0.01,
+      "loss": 2.08,
+      "step": 11823
+    },
+    {
+      "epoch": 1.2136699507389164,
+      "grad_norm": 0.10611164569854736,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 11826
+    },
+    {
+      "epoch": 1.2139778325123154,
+      "grad_norm": 0.0817422866821289,
+      "learning_rate": 0.01,
+      "loss": 2.069,
+      "step": 11829
+    },
+    {
+      "epoch": 1.2142857142857142,
+      "grad_norm": 0.05528125911951065,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 11832
+    },
+    {
+      "epoch": 1.2145935960591132,
+      "grad_norm": 0.05016999691724777,
+      "learning_rate": 0.01,
+      "loss": 2.0794,
+      "step": 11835
+    },
+    {
+      "epoch": 1.2149014778325122,
+      "grad_norm": 0.06376414746046066,
+      "learning_rate": 0.01,
+      "loss": 2.1097,
+      "step": 11838
+    },
+    {
+      "epoch": 1.2152093596059113,
+      "grad_norm": 0.06668904423713684,
+      "learning_rate": 0.01,
+      "loss": 2.0745,
+      "step": 11841
+    },
+    {
+      "epoch": 1.2155172413793103,
+      "grad_norm": 0.046260587871074677,
+      "learning_rate": 0.01,
+      "loss": 2.0711,
+      "step": 11844
+    },
+    {
+      "epoch": 1.2158251231527093,
+      "grad_norm": 0.039374321699142456,
+      "learning_rate": 0.01,
+      "loss": 2.087,
+      "step": 11847
+    },
+    {
+      "epoch": 1.2161330049261083,
+      "grad_norm": 0.10993562638759613,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 11850
+    },
+    {
+      "epoch": 1.2164408866995073,
+      "grad_norm": 0.03676668182015419,
+      "learning_rate": 0.01,
+      "loss": 2.0717,
+      "step": 11853
+    },
+    {
+      "epoch": 1.2167487684729064,
+      "grad_norm": 0.10777715593576431,
+      "learning_rate": 0.01,
+      "loss": 2.1016,
+      "step": 11856
+    },
+    {
+      "epoch": 1.2170566502463054,
+      "grad_norm": 0.07948705554008484,
+      "learning_rate": 0.01,
+      "loss": 2.083,
+      "step": 11859
+    },
+    {
+      "epoch": 1.2173645320197044,
+      "grad_norm": 0.11646637320518494,
+      "learning_rate": 0.01,
+      "loss": 2.0552,
+      "step": 11862
+    },
+    {
+      "epoch": 1.2176724137931034,
+      "grad_norm": 0.07525186985731125,
+      "learning_rate": 0.01,
+      "loss": 2.0877,
+      "step": 11865
+    },
+    {
+      "epoch": 1.2179802955665024,
+      "grad_norm": 0.048124101012945175,
+      "learning_rate": 0.01,
+      "loss": 2.0652,
+      "step": 11868
+    },
+    {
+      "epoch": 1.2182881773399015,
+      "grad_norm": 0.04603361710906029,
+      "learning_rate": 0.01,
+      "loss": 2.0922,
+      "step": 11871
+    },
+    {
+      "epoch": 1.2185960591133005,
+      "grad_norm": 0.07067687064409256,
+      "learning_rate": 0.01,
+      "loss": 2.0946,
+      "step": 11874
+    },
+    {
+      "epoch": 1.2189039408866995,
+      "grad_norm": 0.0959327220916748,
+      "learning_rate": 0.01,
+      "loss": 2.096,
+      "step": 11877
+    },
+    {
+      "epoch": 1.2192118226600985,
+      "grad_norm": 0.08565320819616318,
+      "learning_rate": 0.01,
+      "loss": 2.09,
+      "step": 11880
+    },
+    {
+      "epoch": 1.2195197044334976,
+      "grad_norm": 0.06728377193212509,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 11883
+    },
+    {
+      "epoch": 1.2198275862068966,
+      "grad_norm": 0.03809618949890137,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 11886
+    },
+    {
+      "epoch": 1.2201354679802956,
+      "grad_norm": 0.049925826489925385,
+      "learning_rate": 0.01,
+      "loss": 2.0625,
+      "step": 11889
+    },
+    {
+      "epoch": 1.2204433497536946,
+      "grad_norm": 0.05949478596448898,
+      "learning_rate": 0.01,
+      "loss": 2.0687,
+      "step": 11892
+    },
+    {
+      "epoch": 1.2207512315270936,
+      "grad_norm": 0.08161807060241699,
+      "learning_rate": 0.01,
+      "loss": 2.0789,
+      "step": 11895
+    },
+    {
+      "epoch": 1.2210591133004927,
+      "grad_norm": 0.05829952284693718,
+      "learning_rate": 0.01,
+      "loss": 2.0846,
+      "step": 11898
+    },
+    {
+      "epoch": 1.2213669950738917,
+      "grad_norm": 0.05801619216799736,
+      "learning_rate": 0.01,
+      "loss": 2.1014,
+      "step": 11901
+    },
+    {
+      "epoch": 1.2216748768472907,
+      "grad_norm": 0.04123099148273468,
+      "learning_rate": 0.01,
+      "loss": 2.0867,
+      "step": 11904
+    },
+    {
+      "epoch": 1.2219827586206897,
+      "grad_norm": 0.05088057741522789,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 11907
+    },
+    {
+      "epoch": 1.2222906403940887,
+      "grad_norm": 0.07357197254896164,
+      "learning_rate": 0.01,
+      "loss": 2.1023,
+      "step": 11910
+    },
+    {
+      "epoch": 1.2225985221674878,
+      "grad_norm": 0.060078103095293045,
+      "learning_rate": 0.01,
+      "loss": 2.0784,
+      "step": 11913
+    },
+    {
+      "epoch": 1.2229064039408868,
+      "grad_norm": 0.12629617750644684,
+      "learning_rate": 0.01,
+      "loss": 2.0767,
+      "step": 11916
+    },
+    {
+      "epoch": 1.2232142857142858,
+      "grad_norm": 0.07202067971229553,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 11919
+    },
+    {
+      "epoch": 1.2235221674876846,
+      "grad_norm": 0.06407934427261353,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 11922
+    },
+    {
+      "epoch": 1.2238300492610836,
+      "grad_norm": 0.053789231926202774,
+      "learning_rate": 0.01,
+      "loss": 2.092,
+      "step": 11925
+    },
+    {
+      "epoch": 1.2241379310344827,
+      "grad_norm": 0.04130502790212631,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 11928
+    },
+    {
+      "epoch": 1.2244458128078817,
+      "grad_norm": 0.05235166475176811,
+      "learning_rate": 0.01,
+      "loss": 2.0829,
+      "step": 11931
+    },
+    {
+      "epoch": 1.2247536945812807,
+      "grad_norm": 0.04508119449019432,
+      "learning_rate": 0.01,
+      "loss": 2.0857,
+      "step": 11934
+    },
+    {
+      "epoch": 1.2250615763546797,
+      "grad_norm": 0.03570512309670448,
+      "learning_rate": 0.01,
+      "loss": 2.0734,
+      "step": 11937
+    },
+    {
+      "epoch": 1.2253694581280787,
+      "grad_norm": 0.04690218344330788,
+      "learning_rate": 0.01,
+      "loss": 2.0988,
+      "step": 11940
+    },
+    {
+      "epoch": 1.2256773399014778,
+      "grad_norm": 0.10231764614582062,
+      "learning_rate": 0.01,
+      "loss": 2.0716,
+      "step": 11943
+    },
+    {
+      "epoch": 1.2259852216748768,
+      "grad_norm": 0.05221893638372421,
+      "learning_rate": 0.01,
+      "loss": 2.0693,
+      "step": 11946
+    },
+    {
+      "epoch": 1.2262931034482758,
+      "grad_norm": 0.0647406056523323,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 11949
+    },
+    {
+      "epoch": 1.2266009852216748,
+      "grad_norm": 0.06388009339570999,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 11952
+    },
+    {
+      "epoch": 1.2269088669950738,
+      "grad_norm": 0.06904192268848419,
+      "learning_rate": 0.01,
+      "loss": 2.0917,
+      "step": 11955
+    },
+    {
+      "epoch": 1.2272167487684729,
+      "grad_norm": 0.08780385553836823,
+      "learning_rate": 0.01,
+      "loss": 2.0881,
+      "step": 11958
+    },
+    {
+      "epoch": 1.2275246305418719,
+      "grad_norm": 0.037958092987537384,
+      "learning_rate": 0.01,
+      "loss": 2.0813,
+      "step": 11961
+    },
+    {
+      "epoch": 1.227832512315271,
+      "grad_norm": 0.04035305231809616,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 11964
+    },
+    {
+      "epoch": 1.22814039408867,
+      "grad_norm": 0.056451354175806046,
+      "learning_rate": 0.01,
+      "loss": 2.0653,
+      "step": 11967
+    },
+    {
+      "epoch": 1.228448275862069,
+      "grad_norm": 0.06248374283313751,
+      "learning_rate": 0.01,
+      "loss": 2.0794,
+      "step": 11970
+    },
+    {
+      "epoch": 1.228756157635468,
+      "grad_norm": 0.05662978067994118,
+      "learning_rate": 0.01,
+      "loss": 2.0873,
+      "step": 11973
+    },
+    {
+      "epoch": 1.229064039408867,
+      "grad_norm": 0.06416438519954681,
+      "learning_rate": 0.01,
+      "loss": 2.0933,
+      "step": 11976
+    },
+    {
+      "epoch": 1.229371921182266,
+      "grad_norm": 0.04529969021677971,
+      "learning_rate": 0.01,
+      "loss": 2.0892,
+      "step": 11979
+    },
+    {
+      "epoch": 1.229679802955665,
+      "grad_norm": 0.03636370226740837,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 11982
+    },
+    {
+      "epoch": 1.229987684729064,
+      "grad_norm": 0.03992651402950287,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 11985
+    },
+    {
+      "epoch": 1.230295566502463,
+      "grad_norm": 0.19436125457286835,
+      "learning_rate": 0.01,
+      "loss": 2.0779,
+      "step": 11988
+    },
+    {
+      "epoch": 1.230603448275862,
+      "grad_norm": 0.15459048748016357,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 11991
+    },
+    {
+      "epoch": 1.2309113300492611,
+      "grad_norm": 0.11131371557712555,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 11994
+    },
+    {
+      "epoch": 1.2312192118226601,
+      "grad_norm": 0.06876586377620697,
+      "learning_rate": 0.01,
+      "loss": 2.1081,
+      "step": 11997
+    },
+    {
+      "epoch": 1.2315270935960592,
+      "grad_norm": 0.03379599004983902,
+      "learning_rate": 0.01,
+      "loss": 2.0977,
+      "step": 12000
+    },
+    {
+      "epoch": 1.2318349753694582,
+      "grad_norm": 0.06905510276556015,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 12003
+    },
+    {
+      "epoch": 1.2321428571428572,
+      "grad_norm": 0.05859539657831192,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 12006
+    },
+    {
+      "epoch": 1.2324507389162562,
+      "grad_norm": 0.1200842559337616,
+      "learning_rate": 0.01,
+      "loss": 2.1045,
+      "step": 12009
+    },
+    {
+      "epoch": 1.2327586206896552,
+      "grad_norm": 0.09969060868024826,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 12012
+    },
+    {
+      "epoch": 1.2330665024630543,
+      "grad_norm": 0.08915867656469345,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 12015
+    },
+    {
+      "epoch": 1.2333743842364533,
+      "grad_norm": 0.10951671004295349,
+      "learning_rate": 0.01,
+      "loss": 2.1007,
+      "step": 12018
+    },
+    {
+      "epoch": 1.2336822660098523,
+      "grad_norm": 0.15262556076049805,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 12021
+    },
+    {
+      "epoch": 1.2339901477832513,
+      "grad_norm": 0.05622226372361183,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 12024
+    },
+    {
+      "epoch": 1.2342980295566504,
+      "grad_norm": 0.05918841436505318,
+      "learning_rate": 0.01,
+      "loss": 2.0911,
+      "step": 12027
+    },
+    {
+      "epoch": 1.2346059113300494,
+      "grad_norm": 0.04867622256278992,
+      "learning_rate": 0.01,
+      "loss": 2.0872,
+      "step": 12030
+    },
+    {
+      "epoch": 1.2349137931034484,
+      "grad_norm": 0.04389597848057747,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 12033
+    },
+    {
+      "epoch": 1.2352216748768472,
+      "grad_norm": 0.1209108904004097,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 12036
+    },
+    {
+      "epoch": 1.2355295566502462,
+      "grad_norm": 0.08446931838989258,
+      "learning_rate": 0.01,
+      "loss": 2.0744,
+      "step": 12039
+    },
+    {
+      "epoch": 1.2358374384236452,
+      "grad_norm": 0.07141686230897903,
+      "learning_rate": 0.01,
+      "loss": 2.1099,
+      "step": 12042
+    },
+    {
+      "epoch": 1.2361453201970443,
+      "grad_norm": 0.1216801181435585,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 12045
+    },
+    {
+      "epoch": 1.2364532019704433,
+      "grad_norm": 0.10539086163043976,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 12048
+    },
+    {
+      "epoch": 1.2367610837438423,
+      "grad_norm": 0.09336747229099274,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 12051
+    },
+    {
+      "epoch": 1.2370689655172413,
+      "grad_norm": 0.091059610247612,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 12054
+    },
+    {
+      "epoch": 1.2373768472906403,
+      "grad_norm": 0.08291159570217133,
+      "learning_rate": 0.01,
+      "loss": 2.1039,
+      "step": 12057
+    },
+    {
+      "epoch": 1.2376847290640394,
+      "grad_norm": 0.07626821845769882,
+      "learning_rate": 0.01,
+      "loss": 2.0784,
+      "step": 12060
+    },
+    {
+      "epoch": 1.2379926108374384,
+      "grad_norm": 0.05197496339678764,
+      "learning_rate": 0.01,
+      "loss": 2.0968,
+      "step": 12063
+    },
+    {
+      "epoch": 1.2383004926108374,
+      "grad_norm": 0.061275266110897064,
+      "learning_rate": 0.01,
+      "loss": 2.095,
+      "step": 12066
+    },
+    {
+      "epoch": 1.2386083743842364,
+      "grad_norm": 0.04282483085989952,
+      "learning_rate": 0.01,
+      "loss": 2.083,
+      "step": 12069
+    },
+    {
+      "epoch": 1.2389162561576355,
+      "grad_norm": 0.037066422402858734,
+      "learning_rate": 0.01,
+      "loss": 2.0653,
+      "step": 12072
+    },
+    {
+      "epoch": 1.2392241379310345,
+      "grad_norm": 0.0467105396091938,
+      "learning_rate": 0.01,
+      "loss": 2.0913,
+      "step": 12075
+    },
+    {
+      "epoch": 1.2395320197044335,
+      "grad_norm": 0.053995974361896515,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 12078
+    },
+    {
+      "epoch": 1.2398399014778325,
+      "grad_norm": 0.08583737164735794,
+      "learning_rate": 0.01,
+      "loss": 2.089,
+      "step": 12081
+    },
+    {
+      "epoch": 1.2401477832512315,
+      "grad_norm": 0.07264076173305511,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 12084
+    },
+    {
+      "epoch": 1.2404556650246306,
+      "grad_norm": 0.062001802027225494,
+      "learning_rate": 0.01,
+      "loss": 2.0969,
+      "step": 12087
+    },
+    {
+      "epoch": 1.2407635467980296,
+      "grad_norm": 0.05311381444334984,
+      "learning_rate": 0.01,
+      "loss": 2.0833,
+      "step": 12090
+    },
+    {
+      "epoch": 1.2410714285714286,
+      "grad_norm": 0.04272656887769699,
+      "learning_rate": 0.01,
+      "loss": 2.0883,
+      "step": 12093
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.10696172714233398,
+      "learning_rate": 0.01,
+      "loss": 2.0925,
+      "step": 12096
+    },
+    {
+      "epoch": 1.2416871921182266,
+      "grad_norm": 0.08625493943691254,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 12099
+    },
+    {
+      "epoch": 1.2419950738916257,
+      "grad_norm": 0.06818173080682755,
+      "learning_rate": 0.01,
+      "loss": 2.0943,
+      "step": 12102
+    },
+    {
+      "epoch": 1.2423029556650247,
+      "grad_norm": 0.050731681287288666,
+      "learning_rate": 0.01,
+      "loss": 2.0802,
+      "step": 12105
+    },
+    {
+      "epoch": 1.2426108374384237,
+      "grad_norm": 0.08426883816719055,
+      "learning_rate": 0.01,
+      "loss": 2.0825,
+      "step": 12108
+    },
+    {
+      "epoch": 1.2429187192118227,
+      "grad_norm": 0.09432832151651382,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 12111
+    },
+    {
+      "epoch": 1.2432266009852218,
+      "grad_norm": 0.06951441615819931,
+      "learning_rate": 0.01,
+      "loss": 2.0951,
+      "step": 12114
+    },
+    {
+      "epoch": 1.2435344827586208,
+      "grad_norm": 0.06427393108606339,
+      "learning_rate": 0.01,
+      "loss": 2.0706,
+      "step": 12117
+    },
+    {
+      "epoch": 1.2438423645320198,
+      "grad_norm": 0.03967609629034996,
+      "learning_rate": 0.01,
+      "loss": 2.0897,
+      "step": 12120
+    },
+    {
+      "epoch": 1.2441502463054186,
+      "grad_norm": 0.036665160208940506,
+      "learning_rate": 0.01,
+      "loss": 2.0791,
+      "step": 12123
+    },
+    {
+      "epoch": 1.2444581280788176,
+      "grad_norm": 0.072290800511837,
+      "learning_rate": 0.01,
+      "loss": 2.0889,
+      "step": 12126
+    },
+    {
+      "epoch": 1.2447660098522166,
+      "grad_norm": 0.07136868685483932,
+      "learning_rate": 0.01,
+      "loss": 2.0688,
+      "step": 12129
+    },
+    {
+      "epoch": 1.2450738916256157,
+      "grad_norm": 0.15400493144989014,
+      "learning_rate": 0.01,
+      "loss": 2.0821,
+      "step": 12132
+    },
+    {
+      "epoch": 1.2453817733990147,
+      "grad_norm": 0.07114578038454056,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 12135
+    },
+    {
+      "epoch": 1.2456896551724137,
+      "grad_norm": 0.043961767107248306,
+      "learning_rate": 0.01,
+      "loss": 2.0877,
+      "step": 12138
+    },
+    {
+      "epoch": 1.2459975369458127,
+      "grad_norm": 0.056267060339450836,
+      "learning_rate": 0.01,
+      "loss": 2.0688,
+      "step": 12141
+    },
+    {
+      "epoch": 1.2463054187192117,
+      "grad_norm": 0.035889722406864166,
+      "learning_rate": 0.01,
+      "loss": 2.0783,
+      "step": 12144
+    },
+    {
+      "epoch": 1.2466133004926108,
+      "grad_norm": 0.1781640499830246,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 12147
+    },
+    {
+      "epoch": 1.2469211822660098,
+      "grad_norm": 0.0891503393650055,
+      "learning_rate": 0.01,
+      "loss": 2.0957,
+      "step": 12150
+    },
+    {
+      "epoch": 1.2472290640394088,
+      "grad_norm": 0.047431472688913345,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 12153
+    },
+    {
+      "epoch": 1.2475369458128078,
+      "grad_norm": 0.04693286865949631,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 12156
+    },
+    {
+      "epoch": 1.2478448275862069,
+      "grad_norm": 0.0382777564227581,
+      "learning_rate": 0.01,
+      "loss": 2.0851,
+      "step": 12159
+    },
+    {
+      "epoch": 1.2481527093596059,
+      "grad_norm": 0.04085429012775421,
+      "learning_rate": 0.01,
+      "loss": 2.0846,
+      "step": 12162
+    },
+    {
+      "epoch": 1.248460591133005,
+      "grad_norm": 0.05329781025648117,
+      "learning_rate": 0.01,
+      "loss": 2.0732,
+      "step": 12165
+    },
+    {
+      "epoch": 1.248768472906404,
+      "grad_norm": 0.06961992383003235,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 12168
+    },
+    {
+      "epoch": 1.249076354679803,
+      "grad_norm": 0.05290938913822174,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 12171
+    },
+    {
+      "epoch": 1.249384236453202,
+      "grad_norm": 0.1247674971818924,
+      "learning_rate": 0.01,
+      "loss": 2.0679,
+      "step": 12174
+    },
+    {
+      "epoch": 1.249692118226601,
+      "grad_norm": 0.04983863607048988,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 12177
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.08552074432373047,
+      "learning_rate": 0.01,
+      "loss": 2.1084,
+      "step": 12180
+    },
+    {
+      "epoch": 1.250307881773399,
+      "grad_norm": 0.1376069337129593,
+      "learning_rate": 0.01,
+      "loss": 2.0997,
+      "step": 12183
+    },
+    {
+      "epoch": 1.250615763546798,
+      "grad_norm": 0.07097752392292023,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 12186
+    },
+    {
+      "epoch": 1.250923645320197,
+      "grad_norm": 0.03953644260764122,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 12189
+    },
+    {
+      "epoch": 1.251231527093596,
+      "grad_norm": 0.04611526057124138,
+      "learning_rate": 0.01,
+      "loss": 2.0717,
+      "step": 12192
+    },
+    {
+      "epoch": 1.251539408866995,
+      "grad_norm": 0.07573895156383514,
+      "learning_rate": 0.01,
+      "loss": 2.0785,
+      "step": 12195
+    },
+    {
+      "epoch": 1.2518472906403941,
+      "grad_norm": 0.07144660502672195,
+      "learning_rate": 0.01,
+      "loss": 2.0815,
+      "step": 12198
+    },
+    {
+      "epoch": 1.2521551724137931,
+      "grad_norm": 0.05297645181417465,
+      "learning_rate": 0.01,
+      "loss": 2.093,
+      "step": 12201
+    },
+    {
+      "epoch": 1.2524630541871922,
+      "grad_norm": 0.044887710362672806,
+      "learning_rate": 0.01,
+      "loss": 2.0752,
+      "step": 12204
+    },
+    {
+      "epoch": 1.2527709359605912,
+      "grad_norm": 0.04305564984679222,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 12207
+    },
+    {
+      "epoch": 1.2530788177339902,
+      "grad_norm": 0.057785287499427795,
+      "learning_rate": 0.01,
+      "loss": 2.094,
+      "step": 12210
+    },
+    {
+      "epoch": 1.2533866995073892,
+      "grad_norm": 0.04404570534825325,
+      "learning_rate": 0.01,
+      "loss": 2.1037,
+      "step": 12213
+    },
+    {
+      "epoch": 1.2536945812807883,
+      "grad_norm": 0.055468104779720306,
+      "learning_rate": 0.01,
+      "loss": 2.0851,
+      "step": 12216
+    },
+    {
+      "epoch": 1.2540024630541873,
+      "grad_norm": 0.17121906578540802,
+      "learning_rate": 0.01,
+      "loss": 2.081,
+      "step": 12219
+    },
+    {
+      "epoch": 1.2543103448275863,
+      "grad_norm": 0.09411416202783585,
+      "learning_rate": 0.01,
+      "loss": 2.0974,
+      "step": 12222
+    },
+    {
+      "epoch": 1.2546182266009853,
+      "grad_norm": 0.07855021953582764,
+      "learning_rate": 0.01,
+      "loss": 2.0733,
+      "step": 12225
+    },
+    {
+      "epoch": 1.2549261083743843,
+      "grad_norm": 0.052616432309150696,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 12228
+    },
+    {
+      "epoch": 1.2552339901477834,
+      "grad_norm": 0.047992121428251266,
+      "learning_rate": 0.01,
+      "loss": 2.0918,
+      "step": 12231
+    },
+    {
+      "epoch": 1.2555418719211824,
+      "grad_norm": 0.04336715489625931,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 12234
+    },
+    {
+      "epoch": 1.2558497536945814,
+      "grad_norm": 0.03128316253423691,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 12237
+    },
+    {
+      "epoch": 1.2561576354679804,
+      "grad_norm": 0.06315557658672333,
+      "learning_rate": 0.01,
+      "loss": 2.0918,
+      "step": 12240
+    },
+    {
+      "epoch": 1.2564655172413794,
+      "grad_norm": 0.0528687946498394,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 12243
+    },
+    {
+      "epoch": 1.2567733990147782,
+      "grad_norm": 0.17166706919670105,
+      "learning_rate": 0.01,
+      "loss": 2.068,
+      "step": 12246
+    },
+    {
+      "epoch": 1.2570812807881773,
+      "grad_norm": 0.11394128203392029,
+      "learning_rate": 0.01,
+      "loss": 2.1179,
+      "step": 12249
+    },
+    {
+      "epoch": 1.2573891625615763,
+      "grad_norm": 0.08554805815219879,
+      "learning_rate": 0.01,
+      "loss": 2.0857,
+      "step": 12252
+    },
+    {
+      "epoch": 1.2576970443349753,
+      "grad_norm": 0.05203767865896225,
+      "learning_rate": 0.01,
+      "loss": 2.0975,
+      "step": 12255
+    },
+    {
+      "epoch": 1.2580049261083743,
+      "grad_norm": 0.06072428077459335,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 12258
+    },
+    {
+      "epoch": 1.2583128078817734,
+      "grad_norm": 0.044136617332696915,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 12261
+    },
+    {
+      "epoch": 1.2586206896551724,
+      "grad_norm": 0.038774993270635605,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 12264
+    },
+    {
+      "epoch": 1.2589285714285714,
+      "grad_norm": 0.03669529780745506,
+      "learning_rate": 0.01,
+      "loss": 2.0949,
+      "step": 12267
+    },
+    {
+      "epoch": 1.2592364532019704,
+      "grad_norm": 0.050722066313028336,
+      "learning_rate": 0.01,
+      "loss": 2.0772,
+      "step": 12270
+    },
+    {
+      "epoch": 1.2595443349753694,
+      "grad_norm": 0.12684300541877747,
+      "learning_rate": 0.01,
+      "loss": 2.0814,
+      "step": 12273
+    },
+    {
+      "epoch": 1.2598522167487685,
+      "grad_norm": 0.07431039214134216,
+      "learning_rate": 0.01,
+      "loss": 2.1001,
+      "step": 12276
+    },
+    {
+      "epoch": 1.2601600985221675,
+      "grad_norm": 0.07050034403800964,
+      "learning_rate": 0.01,
+      "loss": 2.0507,
+      "step": 12279
+    },
+    {
+      "epoch": 1.2604679802955665,
+      "grad_norm": 0.0553620308637619,
+      "learning_rate": 0.01,
+      "loss": 2.0757,
+      "step": 12282
+    },
+    {
+      "epoch": 1.2607758620689655,
+      "grad_norm": 0.10946903377771378,
+      "learning_rate": 0.01,
+      "loss": 2.0659,
+      "step": 12285
+    },
+    {
+      "epoch": 1.2610837438423645,
+      "grad_norm": 0.06346802413463593,
+      "learning_rate": 0.01,
+      "loss": 2.0871,
+      "step": 12288
+    },
+    {
+      "epoch": 1.2613916256157636,
+      "grad_norm": 0.09386668354272842,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 12291
+    },
+    {
+      "epoch": 1.2616995073891626,
+      "grad_norm": 0.05672946944832802,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 12294
+    },
+    {
+      "epoch": 1.2620073891625616,
+      "grad_norm": 0.1079539805650711,
+      "learning_rate": 0.01,
+      "loss": 2.0851,
+      "step": 12297
+    },
+    {
+      "epoch": 1.2623152709359606,
+      "grad_norm": 0.043078985065221786,
+      "learning_rate": 0.01,
+      "loss": 2.0846,
+      "step": 12300
+    },
+    {
+      "epoch": 1.2626231527093597,
+      "grad_norm": 0.043098706752061844,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 12303
+    },
+    {
+      "epoch": 1.2629310344827587,
+      "grad_norm": 0.04908977448940277,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 12306
+    },
+    {
+      "epoch": 1.2632389162561577,
+      "grad_norm": 0.09897246211767197,
+      "learning_rate": 0.01,
+      "loss": 2.0899,
+      "step": 12309
+    },
+    {
+      "epoch": 1.2635467980295567,
+      "grad_norm": 0.0861278846859932,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 12312
+    },
+    {
+      "epoch": 1.2638546798029557,
+      "grad_norm": 0.06973280757665634,
+      "learning_rate": 0.01,
+      "loss": 2.0598,
+      "step": 12315
+    },
+    {
+      "epoch": 1.2641625615763548,
+      "grad_norm": 0.048658501356840134,
+      "learning_rate": 0.01,
+      "loss": 2.0838,
+      "step": 12318
+    },
+    {
+      "epoch": 1.2644704433497536,
+      "grad_norm": 0.05053295940160751,
+      "learning_rate": 0.01,
+      "loss": 2.0665,
+      "step": 12321
+    },
+    {
+      "epoch": 1.2647783251231526,
+      "grad_norm": 0.0536380410194397,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 12324
+    },
+    {
+      "epoch": 1.2650862068965516,
+      "grad_norm": 0.13452892005443573,
+      "learning_rate": 0.01,
+      "loss": 2.0928,
+      "step": 12327
+    },
+    {
+      "epoch": 1.2653940886699506,
+      "grad_norm": 0.056635159999132156,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 12330
+    },
+    {
+      "epoch": 1.2657019704433496,
+      "grad_norm": 0.09460306912660599,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 12333
+    },
+    {
+      "epoch": 1.2660098522167487,
+      "grad_norm": 0.09019794315099716,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 12336
+    },
+    {
+      "epoch": 1.2663177339901477,
+      "grad_norm": 0.04020017758011818,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 12339
+    },
+    {
+      "epoch": 1.2666256157635467,
+      "grad_norm": 0.05063892900943756,
+      "learning_rate": 0.01,
+      "loss": 2.0846,
+      "step": 12342
+    },
+    {
+      "epoch": 1.2669334975369457,
+      "grad_norm": 0.06472054123878479,
+      "learning_rate": 0.01,
+      "loss": 2.0884,
+      "step": 12345
+    },
+    {
+      "epoch": 1.2672413793103448,
+      "grad_norm": 0.0523315854370594,
+      "learning_rate": 0.01,
+      "loss": 2.0756,
+      "step": 12348
+    },
+    {
+      "epoch": 1.2675492610837438,
+      "grad_norm": 0.040240950882434845,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 12351
+    },
+    {
+      "epoch": 1.2678571428571428,
+      "grad_norm": 0.061988551169633865,
+      "learning_rate": 0.01,
+      "loss": 2.0899,
+      "step": 12354
+    },
+    {
+      "epoch": 1.2681650246305418,
+      "grad_norm": 0.03831657022237778,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 12357
+    },
+    {
+      "epoch": 1.2684729064039408,
+      "grad_norm": 0.105617955327034,
+      "learning_rate": 0.01,
+      "loss": 2.0553,
+      "step": 12360
+    },
+    {
+      "epoch": 1.2687807881773399,
+      "grad_norm": 0.09372366219758987,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 12363
+    },
+    {
+      "epoch": 1.2690886699507389,
+      "grad_norm": 0.10305638611316681,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 12366
+    },
+    {
+      "epoch": 1.269396551724138,
+      "grad_norm": 0.07187418639659882,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 12369
+    },
+    {
+      "epoch": 1.269704433497537,
+      "grad_norm": 0.03744306415319443,
+      "learning_rate": 0.01,
+      "loss": 2.0771,
+      "step": 12372
+    },
+    {
+      "epoch": 1.270012315270936,
+      "grad_norm": 0.03059488907456398,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 12375
+    },
+    {
+      "epoch": 1.270320197044335,
+      "grad_norm": 0.09767211973667145,
+      "learning_rate": 0.01,
+      "loss": 2.0889,
+      "step": 12378
+    },
+    {
+      "epoch": 1.270628078817734,
+      "grad_norm": 0.05093003436923027,
+      "learning_rate": 0.01,
+      "loss": 2.0852,
+      "step": 12381
+    },
+    {
+      "epoch": 1.270935960591133,
+      "grad_norm": 0.04648155719041824,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 12384
+    },
+    {
+      "epoch": 1.271243842364532,
+      "grad_norm": 0.04745417460799217,
+      "learning_rate": 0.01,
+      "loss": 2.0888,
+      "step": 12387
+    },
+    {
+      "epoch": 1.271551724137931,
+      "grad_norm": 0.05171092227101326,
+      "learning_rate": 0.01,
+      "loss": 2.075,
+      "step": 12390
+    },
+    {
+      "epoch": 1.27185960591133,
+      "grad_norm": 0.05051850154995918,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 12393
+    },
+    {
+      "epoch": 1.272167487684729,
+      "grad_norm": 0.06048591807484627,
+      "learning_rate": 0.01,
+      "loss": 2.1025,
+      "step": 12396
+    },
+    {
+      "epoch": 1.2724753694581281,
+      "grad_norm": 0.09673628211021423,
+      "learning_rate": 0.01,
+      "loss": 2.0866,
+      "step": 12399
+    },
+    {
+      "epoch": 1.2727832512315271,
+      "grad_norm": 0.11826295405626297,
+      "learning_rate": 0.01,
+      "loss": 2.0757,
+      "step": 12402
+    },
+    {
+      "epoch": 1.2730911330049262,
+      "grad_norm": 0.06271769106388092,
+      "learning_rate": 0.01,
+      "loss": 2.0797,
+      "step": 12405
+    },
+    {
+      "epoch": 1.2733990147783252,
+      "grad_norm": 0.0463738851249218,
+      "learning_rate": 0.01,
+      "loss": 2.054,
+      "step": 12408
+    },
+    {
+      "epoch": 1.2737068965517242,
+      "grad_norm": 0.035649579018354416,
+      "learning_rate": 0.01,
+      "loss": 2.0747,
+      "step": 12411
+    },
+    {
+      "epoch": 1.2740147783251232,
+      "grad_norm": 0.08586122840642929,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 12414
+    },
+    {
+      "epoch": 1.2743226600985222,
+      "grad_norm": 0.08538860082626343,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 12417
+    },
+    {
+      "epoch": 1.2746305418719213,
+      "grad_norm": 0.05974709242582321,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 12420
+    },
+    {
+      "epoch": 1.2749384236453203,
+      "grad_norm": 0.057023100554943085,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 12423
+    },
+    {
+      "epoch": 1.2752463054187193,
+      "grad_norm": 0.062105972319841385,
+      "learning_rate": 0.01,
+      "loss": 2.0946,
+      "step": 12426
+    },
+    {
+      "epoch": 1.2755541871921183,
+      "grad_norm": 0.048221688717603683,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 12429
+    },
+    {
+      "epoch": 1.2758620689655173,
+      "grad_norm": 0.04440610110759735,
+      "learning_rate": 0.01,
+      "loss": 2.0852,
+      "step": 12432
+    },
+    {
+      "epoch": 1.2761699507389164,
+      "grad_norm": 0.06074054539203644,
+      "learning_rate": 0.01,
+      "loss": 2.0667,
+      "step": 12435
+    },
+    {
+      "epoch": 1.2764778325123154,
+      "grad_norm": 0.09477720409631729,
+      "learning_rate": 0.01,
+      "loss": 2.0815,
+      "step": 12438
+    },
+    {
+      "epoch": 1.2767857142857144,
+      "grad_norm": 0.08470610529184341,
+      "learning_rate": 0.01,
+      "loss": 2.1023,
+      "step": 12441
+    },
+    {
+      "epoch": 1.2770935960591134,
+      "grad_norm": 0.09079881012439728,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 12444
+    },
+    {
+      "epoch": 1.2774014778325122,
+      "grad_norm": 0.06132930517196655,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 12447
+    },
+    {
+      "epoch": 1.2777093596059113,
+      "grad_norm": 0.04702606052160263,
+      "learning_rate": 0.01,
+      "loss": 2.1051,
+      "step": 12450
+    },
+    {
+      "epoch": 1.2780172413793103,
+      "grad_norm": 0.06621818989515305,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 12453
+    },
+    {
+      "epoch": 1.2783251231527093,
+      "grad_norm": 0.047375794500112534,
+      "learning_rate": 0.01,
+      "loss": 2.0982,
+      "step": 12456
+    },
+    {
+      "epoch": 1.2786330049261083,
+      "grad_norm": 0.03766035661101341,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 12459
+    },
+    {
+      "epoch": 1.2789408866995073,
+      "grad_norm": 0.10769324004650116,
+      "learning_rate": 0.01,
+      "loss": 2.0848,
+      "step": 12462
+    },
+    {
+      "epoch": 1.2792487684729064,
+      "grad_norm": 0.06848851591348648,
+      "learning_rate": 0.01,
+      "loss": 2.0664,
+      "step": 12465
+    },
+    {
+      "epoch": 1.2795566502463054,
+      "grad_norm": 0.09864836931228638,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 12468
+    },
+    {
+      "epoch": 1.2798645320197044,
+      "grad_norm": 0.04042387008666992,
+      "learning_rate": 0.01,
+      "loss": 2.0672,
+      "step": 12471
+    },
+    {
+      "epoch": 1.2801724137931034,
+      "grad_norm": 0.09212526679039001,
+      "learning_rate": 0.01,
+      "loss": 2.0789,
+      "step": 12474
+    },
+    {
+      "epoch": 1.2804802955665024,
+      "grad_norm": 0.08713985234498978,
+      "learning_rate": 0.01,
+      "loss": 2.0765,
+      "step": 12477
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.04133505001664162,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 12480
+    },
+    {
+      "epoch": 1.2810960591133005,
+      "grad_norm": 0.07466418296098709,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 12483
+    },
+    {
+      "epoch": 1.2814039408866995,
+      "grad_norm": 0.08685484528541565,
+      "learning_rate": 0.01,
+      "loss": 2.0689,
+      "step": 12486
+    },
+    {
+      "epoch": 1.2817118226600985,
+      "grad_norm": 0.13300663232803345,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 12489
+    },
+    {
+      "epoch": 1.2820197044334976,
+      "grad_norm": 0.1024186760187149,
+      "learning_rate": 0.01,
+      "loss": 2.0763,
+      "step": 12492
+    },
+    {
+      "epoch": 1.2823275862068966,
+      "grad_norm": 0.05908042937517166,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 12495
+    },
+    {
+      "epoch": 1.2826354679802956,
+      "grad_norm": 0.03715427219867706,
+      "learning_rate": 0.01,
+      "loss": 2.0925,
+      "step": 12498
+    },
+    {
+      "epoch": 1.2829433497536946,
+      "grad_norm": 0.05033004283905029,
+      "learning_rate": 0.01,
+      "loss": 2.0875,
+      "step": 12501
+    },
+    {
+      "epoch": 1.2832512315270936,
+      "grad_norm": 0.04851456359028816,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 12504
+    },
+    {
+      "epoch": 1.2835591133004927,
+      "grad_norm": 0.08153282105922699,
+      "learning_rate": 0.01,
+      "loss": 2.0828,
+      "step": 12507
+    },
+    {
+      "epoch": 1.2838669950738917,
+      "grad_norm": 0.07549238950014114,
+      "learning_rate": 0.01,
+      "loss": 2.0681,
+      "step": 12510
+    },
+    {
+      "epoch": 1.2841748768472907,
+      "grad_norm": 0.08571973443031311,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 12513
+    },
+    {
+      "epoch": 1.2844827586206897,
+      "grad_norm": 0.1036754697561264,
+      "learning_rate": 0.01,
+      "loss": 2.0712,
+      "step": 12516
+    },
+    {
+      "epoch": 1.2847906403940887,
+      "grad_norm": 0.04329349100589752,
+      "learning_rate": 0.01,
+      "loss": 2.0711,
+      "step": 12519
+    },
+    {
+      "epoch": 1.2850985221674878,
+      "grad_norm": 0.03718428686261177,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 12522
+    },
+    {
+      "epoch": 1.2854064039408866,
+      "grad_norm": 0.07977878302335739,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 12525
+    },
+    {
+      "epoch": 1.2857142857142856,
+      "grad_norm": 0.11540202796459198,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 12528
+    },
+    {
+      "epoch": 1.2860221674876846,
+      "grad_norm": 0.07631656527519226,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 12531
+    },
+    {
+      "epoch": 1.2863300492610836,
+      "grad_norm": 0.051738426089286804,
+      "learning_rate": 0.01,
+      "loss": 2.0862,
+      "step": 12534
+    },
+    {
+      "epoch": 1.2866379310344827,
+      "grad_norm": 0.05070396512746811,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 12537
+    },
+    {
+      "epoch": 1.2869458128078817,
+      "grad_norm": 0.039027947932481766,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 12540
+    },
+    {
+      "epoch": 1.2872536945812807,
+      "grad_norm": 0.052646003663539886,
+      "learning_rate": 0.01,
+      "loss": 2.0745,
+      "step": 12543
+    },
+    {
+      "epoch": 1.2875615763546797,
+      "grad_norm": 0.06035429984331131,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 12546
+    },
+    {
+      "epoch": 1.2878694581280787,
+      "grad_norm": 0.0818098783493042,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 12549
+    },
+    {
+      "epoch": 1.2881773399014778,
+      "grad_norm": 0.030897030606865883,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 12552
+    },
+    {
+      "epoch": 1.2884852216748768,
+      "grad_norm": 0.042795561254024506,
+      "learning_rate": 0.01,
+      "loss": 2.0673,
+      "step": 12555
+    },
+    {
+      "epoch": 1.2887931034482758,
+      "grad_norm": 0.11560031026601791,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 12558
+    },
+    {
+      "epoch": 1.2891009852216748,
+      "grad_norm": 0.03919963166117668,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 12561
+    },
+    {
+      "epoch": 1.2894088669950738,
+      "grad_norm": 0.03816407918930054,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 12564
+    },
+    {
+      "epoch": 1.2897167487684729,
+      "grad_norm": 0.04641805216670036,
+      "learning_rate": 0.01,
+      "loss": 2.0779,
+      "step": 12567
+    },
+    {
+      "epoch": 1.2900246305418719,
+      "grad_norm": 0.16832102835178375,
+      "learning_rate": 0.01,
+      "loss": 2.0658,
+      "step": 12570
+    },
+    {
+      "epoch": 1.290332512315271,
+      "grad_norm": 0.06651032716035843,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 12573
+    },
+    {
+      "epoch": 1.29064039408867,
+      "grad_norm": 0.05333925411105156,
+      "learning_rate": 0.01,
+      "loss": 2.0842,
+      "step": 12576
+    },
+    {
+      "epoch": 1.290948275862069,
+      "grad_norm": 0.03173685073852539,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 12579
+    },
+    {
+      "epoch": 1.291256157635468,
+      "grad_norm": 0.0836583599448204,
+      "learning_rate": 0.01,
+      "loss": 2.0605,
+      "step": 12582
+    },
+    {
+      "epoch": 1.291564039408867,
+      "grad_norm": 0.09913724660873413,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 12585
+    },
+    {
+      "epoch": 1.291871921182266,
+      "grad_norm": 0.04268624261021614,
+      "learning_rate": 0.01,
+      "loss": 2.1027,
+      "step": 12588
+    },
+    {
+      "epoch": 1.292179802955665,
+      "grad_norm": 0.03744608163833618,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 12591
+    },
+    {
+      "epoch": 1.292487684729064,
+      "grad_norm": 0.11833969503641129,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 12594
+    },
+    {
+      "epoch": 1.292795566502463,
+      "grad_norm": 0.11814229190349579,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 12597
+    },
+    {
+      "epoch": 1.293103448275862,
+      "grad_norm": 0.04583375155925751,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 12600
+    },
+    {
+      "epoch": 1.2934113300492611,
+      "grad_norm": 0.11559072881937027,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 12603
+    },
+    {
+      "epoch": 1.2937192118226601,
+      "grad_norm": 0.050608474761247635,
+      "learning_rate": 0.01,
+      "loss": 2.0879,
+      "step": 12606
+    },
+    {
+      "epoch": 1.2940270935960592,
+      "grad_norm": 0.04284593090415001,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 12609
+    },
+    {
+      "epoch": 1.2943349753694582,
+      "grad_norm": 0.042168114334344864,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 12612
+    },
+    {
+      "epoch": 1.2946428571428572,
+      "grad_norm": 0.06251826882362366,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 12615
+    },
+    {
+      "epoch": 1.2949507389162562,
+      "grad_norm": 0.11159554868936539,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 12618
+    },
+    {
+      "epoch": 1.2952586206896552,
+      "grad_norm": 0.0715017095208168,
+      "learning_rate": 0.01,
+      "loss": 2.0844,
+      "step": 12621
+    },
+    {
+      "epoch": 1.2955665024630543,
+      "grad_norm": 0.12078917771577835,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 12624
+    },
+    {
+      "epoch": 1.2958743842364533,
+      "grad_norm": 0.11613093316555023,
+      "learning_rate": 0.01,
+      "loss": 2.0753,
+      "step": 12627
+    },
+    {
+      "epoch": 1.2961822660098523,
+      "grad_norm": 0.053543299436569214,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 12630
+    },
+    {
+      "epoch": 1.2964901477832513,
+      "grad_norm": 0.05464399605989456,
+      "learning_rate": 0.01,
+      "loss": 2.0845,
+      "step": 12633
+    },
+    {
+      "epoch": 1.2967980295566504,
+      "grad_norm": 0.0487944521009922,
+      "learning_rate": 0.01,
+      "loss": 2.0859,
+      "step": 12636
+    },
+    {
+      "epoch": 1.2971059113300494,
+      "grad_norm": 0.05266605690121651,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 12639
+    },
+    {
+      "epoch": 1.2974137931034484,
+      "grad_norm": 0.07863074541091919,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 12642
+    },
+    {
+      "epoch": 1.2977216748768474,
+      "grad_norm": 0.05371072143316269,
+      "learning_rate": 0.01,
+      "loss": 2.1016,
+      "step": 12645
+    },
+    {
+      "epoch": 1.2980295566502464,
+      "grad_norm": 0.05592924728989601,
+      "learning_rate": 0.01,
+      "loss": 2.0832,
+      "step": 12648
+    },
+    {
+      "epoch": 1.2983374384236452,
+      "grad_norm": 0.0675397738814354,
+      "learning_rate": 0.01,
+      "loss": 2.0784,
+      "step": 12651
+    },
+    {
+      "epoch": 1.2986453201970443,
+      "grad_norm": 0.04399113729596138,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 12654
+    },
+    {
+      "epoch": 1.2989532019704433,
+      "grad_norm": 0.04609301686286926,
+      "learning_rate": 0.01,
+      "loss": 2.0683,
+      "step": 12657
+    },
+    {
+      "epoch": 1.2992610837438423,
+      "grad_norm": 0.05637912079691887,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 12660
+    },
+    {
+      "epoch": 1.2995689655172413,
+      "grad_norm": 0.1115126758813858,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 12663
+    },
+    {
+      "epoch": 1.2998768472906403,
+      "grad_norm": 0.07762522250413895,
+      "learning_rate": 0.01,
+      "loss": 2.0853,
+      "step": 12666
+    },
+    {
+      "epoch": 1.3001847290640394,
+      "grad_norm": 0.06717406958341599,
+      "learning_rate": 0.01,
+      "loss": 2.0881,
+      "step": 12669
+    },
+    {
+      "epoch": 1.3004926108374384,
+      "grad_norm": 0.0727803036570549,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 12672
+    },
+    {
+      "epoch": 1.3008004926108374,
+      "grad_norm": 0.06588024646043777,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 12675
+    },
+    {
+      "epoch": 1.3011083743842364,
+      "grad_norm": 0.09718842804431915,
+      "learning_rate": 0.01,
+      "loss": 2.0793,
+      "step": 12678
+    },
+    {
+      "epoch": 1.3014162561576355,
+      "grad_norm": 0.05804411321878433,
+      "learning_rate": 0.01,
+      "loss": 2.0825,
+      "step": 12681
+    },
+    {
+      "epoch": 1.3017241379310345,
+      "grad_norm": 0.07549803704023361,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 12684
+    },
+    {
+      "epoch": 1.3020320197044335,
+      "grad_norm": 0.04496621713042259,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 12687
+    },
+    {
+      "epoch": 1.3023399014778325,
+      "grad_norm": 0.1283668428659439,
+      "learning_rate": 0.01,
+      "loss": 2.0639,
+      "step": 12690
+    },
+    {
+      "epoch": 1.3026477832512315,
+      "grad_norm": 0.1276516169309616,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 12693
+    },
+    {
+      "epoch": 1.3029556650246306,
+      "grad_norm": 0.12865981459617615,
+      "learning_rate": 0.01,
+      "loss": 2.0697,
+      "step": 12696
+    },
+    {
+      "epoch": 1.3032635467980296,
+      "grad_norm": 0.05869213864207268,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 12699
+    },
+    {
+      "epoch": 1.3035714285714286,
+      "grad_norm": 0.042082637548446655,
+      "learning_rate": 0.01,
+      "loss": 2.0847,
+      "step": 12702
+    },
+    {
+      "epoch": 1.3038793103448276,
+      "grad_norm": 0.11474558711051941,
+      "learning_rate": 0.01,
+      "loss": 2.0767,
+      "step": 12705
+    },
+    {
+      "epoch": 1.3041871921182266,
+      "grad_norm": 0.12426330894231796,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 12708
+    },
+    {
+      "epoch": 1.3044950738916257,
+      "grad_norm": 0.06731969118118286,
+      "learning_rate": 0.01,
+      "loss": 2.064,
+      "step": 12711
+    },
+    {
+      "epoch": 1.3048029556650247,
+      "grad_norm": 0.049505047500133514,
+      "learning_rate": 0.01,
+      "loss": 2.0848,
+      "step": 12714
+    },
+    {
+      "epoch": 1.3051108374384237,
+      "grad_norm": 0.04460617154836655,
+      "learning_rate": 0.01,
+      "loss": 2.0958,
+      "step": 12717
+    },
+    {
+      "epoch": 1.3054187192118227,
+      "grad_norm": 0.05461740121245384,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 12720
+    },
+    {
+      "epoch": 1.3057266009852218,
+      "grad_norm": 0.04074997082352638,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 12723
+    },
+    {
+      "epoch": 1.3060344827586206,
+      "grad_norm": 0.04925557225942612,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 12726
+    },
+    {
+      "epoch": 1.3063423645320196,
+      "grad_norm": 0.06570678949356079,
+      "learning_rate": 0.01,
+      "loss": 2.0703,
+      "step": 12729
+    },
+    {
+      "epoch": 1.3066502463054186,
+      "grad_norm": 0.076143778860569,
+      "learning_rate": 0.01,
+      "loss": 2.0639,
+      "step": 12732
+    },
+    {
+      "epoch": 1.3069581280788176,
+      "grad_norm": 0.06691130995750427,
+      "learning_rate": 0.01,
+      "loss": 2.0858,
+      "step": 12735
+    },
+    {
+      "epoch": 1.3072660098522166,
+      "grad_norm": 0.06733332574367523,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 12738
+    },
+    {
+      "epoch": 1.3075738916256157,
+      "grad_norm": 0.08008868247270584,
+      "learning_rate": 0.01,
+      "loss": 2.085,
+      "step": 12741
+    },
+    {
+      "epoch": 1.3078817733990147,
+      "grad_norm": 0.10713468492031097,
+      "learning_rate": 0.01,
+      "loss": 2.0642,
+      "step": 12744
+    },
+    {
+      "epoch": 1.3081896551724137,
+      "grad_norm": 0.101436547935009,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 12747
+    },
+    {
+      "epoch": 1.3084975369458127,
+      "grad_norm": 0.0552450455725193,
+      "learning_rate": 0.01,
+      "loss": 2.0553,
+      "step": 12750
+    },
+    {
+      "epoch": 1.3088054187192117,
+      "grad_norm": 0.08755996823310852,
+      "learning_rate": 0.01,
+      "loss": 2.0564,
+      "step": 12753
+    },
+    {
+      "epoch": 1.3091133004926108,
+      "grad_norm": 0.03980748727917671,
+      "learning_rate": 0.01,
+      "loss": 2.084,
+      "step": 12756
+    },
+    {
+      "epoch": 1.3094211822660098,
+      "grad_norm": 0.0971774309873581,
+      "learning_rate": 0.01,
+      "loss": 2.0858,
+      "step": 12759
+    },
+    {
+      "epoch": 1.3097290640394088,
+      "grad_norm": 0.05845404043793678,
+      "learning_rate": 0.01,
+      "loss": 2.0792,
+      "step": 12762
+    },
+    {
+      "epoch": 1.3100369458128078,
+      "grad_norm": 0.08498022705316544,
+      "learning_rate": 0.01,
+      "loss": 2.0932,
+      "step": 12765
+    },
+    {
+      "epoch": 1.3103448275862069,
+      "grad_norm": 0.05135398730635643,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 12768
+    },
+    {
+      "epoch": 1.3106527093596059,
+      "grad_norm": 0.048872511833906174,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 12771
+    },
+    {
+      "epoch": 1.310960591133005,
+      "grad_norm": 0.0529983788728714,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 12774
+    },
+    {
+      "epoch": 1.311268472906404,
+      "grad_norm": 0.03580658137798309,
+      "learning_rate": 0.01,
+      "loss": 2.0741,
+      "step": 12777
+    },
+    {
+      "epoch": 1.311576354679803,
+      "grad_norm": 0.03904234617948532,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 12780
+    },
+    {
+      "epoch": 1.311884236453202,
+      "grad_norm": 0.04073551669716835,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 12783
+    },
+    {
+      "epoch": 1.312192118226601,
+      "grad_norm": 0.058973487466573715,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 12786
+    },
+    {
+      "epoch": 1.3125,
+      "grad_norm": 0.05673586577177048,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 12789
+    },
+    {
+      "epoch": 1.312807881773399,
+      "grad_norm": 0.13680770993232727,
+      "learning_rate": 0.01,
+      "loss": 2.0673,
+      "step": 12792
+    },
+    {
+      "epoch": 1.313115763546798,
+      "grad_norm": 0.0454241968691349,
+      "learning_rate": 0.01,
+      "loss": 2.0766,
+      "step": 12795
+    },
+    {
+      "epoch": 1.313423645320197,
+      "grad_norm": 0.04074293375015259,
+      "learning_rate": 0.01,
+      "loss": 2.0593,
+      "step": 12798
+    },
+    {
+      "epoch": 1.313731527093596,
+      "grad_norm": 0.04174893721938133,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 12801
+    },
+    {
+      "epoch": 1.314039408866995,
+      "grad_norm": 0.06716062128543854,
+      "learning_rate": 0.01,
+      "loss": 2.0642,
+      "step": 12804
+    },
+    {
+      "epoch": 1.3143472906403941,
+      "grad_norm": 0.09866861253976822,
+      "learning_rate": 0.01,
+      "loss": 2.0889,
+      "step": 12807
+    },
+    {
+      "epoch": 1.3146551724137931,
+      "grad_norm": 0.13097235560417175,
+      "learning_rate": 0.01,
+      "loss": 2.0744,
+      "step": 12810
+    },
+    {
+      "epoch": 1.3149630541871922,
+      "grad_norm": 0.07859724014997482,
+      "learning_rate": 0.01,
+      "loss": 2.0899,
+      "step": 12813
+    },
+    {
+      "epoch": 1.3152709359605912,
+      "grad_norm": 0.06141912192106247,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 12816
+    },
+    {
+      "epoch": 1.3155788177339902,
+      "grad_norm": 0.05112985149025917,
+      "learning_rate": 0.01,
+      "loss": 2.0658,
+      "step": 12819
+    },
+    {
+      "epoch": 1.3158866995073892,
+      "grad_norm": 0.03482404351234436,
+      "learning_rate": 0.01,
+      "loss": 2.0802,
+      "step": 12822
+    },
+    {
+      "epoch": 1.3161945812807883,
+      "grad_norm": 0.05562854930758476,
+      "learning_rate": 0.01,
+      "loss": 2.0659,
+      "step": 12825
+    },
+    {
+      "epoch": 1.3165024630541873,
+      "grad_norm": 0.04841645434498787,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 12828
+    },
+    {
+      "epoch": 1.3168103448275863,
+      "grad_norm": 0.0756571963429451,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 12831
+    },
+    {
+      "epoch": 1.3171182266009853,
+      "grad_norm": 0.09575197845697403,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 12834
+    },
+    {
+      "epoch": 1.3174261083743843,
+      "grad_norm": 0.07003197073936462,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 12837
+    },
+    {
+      "epoch": 1.3177339901477834,
+      "grad_norm": 0.12592460215091705,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 12840
+    },
+    {
+      "epoch": 1.3180418719211824,
+      "grad_norm": 0.07086621969938278,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 12843
+    },
+    {
+      "epoch": 1.3183497536945814,
+      "grad_norm": 0.07367062568664551,
+      "learning_rate": 0.01,
+      "loss": 2.0759,
+      "step": 12846
+    },
+    {
+      "epoch": 1.3186576354679804,
+      "grad_norm": 0.06731852889060974,
+      "learning_rate": 0.01,
+      "loss": 2.077,
+      "step": 12849
+    },
+    {
+      "epoch": 1.3189655172413794,
+      "grad_norm": 0.07499510049819946,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 12852
+    },
+    {
+      "epoch": 1.3192733990147782,
+      "grad_norm": 0.07604499906301498,
+      "learning_rate": 0.01,
+      "loss": 2.0717,
+      "step": 12855
+    },
+    {
+      "epoch": 1.3195812807881773,
+      "grad_norm": 0.0770401731133461,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 12858
+    },
+    {
+      "epoch": 1.3198891625615763,
+      "grad_norm": 0.06753168255090714,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 12861
+    },
+    {
+      "epoch": 1.3201970443349753,
+      "grad_norm": 0.14175836741924286,
+      "learning_rate": 0.01,
+      "loss": 2.0811,
+      "step": 12864
+    },
+    {
+      "epoch": 1.3205049261083743,
+      "grad_norm": 0.04258207604289055,
+      "learning_rate": 0.01,
+      "loss": 2.0831,
+      "step": 12867
+    },
+    {
+      "epoch": 1.3208128078817734,
+      "grad_norm": 0.03372815623879433,
+      "learning_rate": 0.01,
+      "loss": 2.0827,
+      "step": 12870
+    },
+    {
+      "epoch": 1.3211206896551724,
+      "grad_norm": 0.040649689733982086,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 12873
+    },
+    {
+      "epoch": 1.3214285714285714,
+      "grad_norm": 0.060684412717819214,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 12876
+    },
+    {
+      "epoch": 1.3217364532019704,
+      "grad_norm": 0.036452196538448334,
+      "learning_rate": 0.01,
+      "loss": 2.0726,
+      "step": 12879
+    },
+    {
+      "epoch": 1.3220443349753694,
+      "grad_norm": 0.09457682818174362,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 12882
+    },
+    {
+      "epoch": 1.3223522167487685,
+      "grad_norm": 0.049776870757341385,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 12885
+    },
+    {
+      "epoch": 1.3226600985221675,
+      "grad_norm": 0.05794042348861694,
+      "learning_rate": 0.01,
+      "loss": 2.0781,
+      "step": 12888
+    },
+    {
+      "epoch": 1.3229679802955665,
+      "grad_norm": 0.052100833505392075,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 12891
+    },
+    {
+      "epoch": 1.3232758620689655,
+      "grad_norm": 0.06558441370725632,
+      "learning_rate": 0.01,
+      "loss": 2.076,
+      "step": 12894
+    },
+    {
+      "epoch": 1.3235837438423645,
+      "grad_norm": 0.1107778549194336,
+      "learning_rate": 0.01,
+      "loss": 2.0859,
+      "step": 12897
+    },
+    {
+      "epoch": 1.3238916256157636,
+      "grad_norm": 0.052982959896326065,
+      "learning_rate": 0.01,
+      "loss": 2.0749,
+      "step": 12900
+    },
+    {
+      "epoch": 1.3241995073891626,
+      "grad_norm": 0.0976911410689354,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 12903
+    },
+    {
+      "epoch": 1.3245073891625616,
+      "grad_norm": 0.05620484799146652,
+      "learning_rate": 0.01,
+      "loss": 2.0892,
+      "step": 12906
+    },
+    {
+      "epoch": 1.3248152709359606,
+      "grad_norm": 0.12753431499004364,
+      "learning_rate": 0.01,
+      "loss": 2.073,
+      "step": 12909
+    },
+    {
+      "epoch": 1.3251231527093597,
+      "grad_norm": 0.14311249554157257,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 12912
+    },
+    {
+      "epoch": 1.3254310344827587,
+      "grad_norm": 0.049242276698350906,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 12915
+    },
+    {
+      "epoch": 1.3257389162561577,
+      "grad_norm": 0.07875422388315201,
+      "learning_rate": 0.01,
+      "loss": 2.0889,
+      "step": 12918
+    },
+    {
+      "epoch": 1.3260467980295567,
+      "grad_norm": 0.036451369524002075,
+      "learning_rate": 0.01,
+      "loss": 2.0634,
+      "step": 12921
+    },
+    {
+      "epoch": 1.3263546798029557,
+      "grad_norm": 0.03659353405237198,
+      "learning_rate": 0.01,
+      "loss": 2.0725,
+      "step": 12924
+    },
+    {
+      "epoch": 1.3266625615763548,
+      "grad_norm": 0.049371130764484406,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 12927
+    },
+    {
+      "epoch": 1.3269704433497536,
+      "grad_norm": 0.05998126044869423,
+      "learning_rate": 0.01,
+      "loss": 2.0894,
+      "step": 12930
+    },
+    {
+      "epoch": 1.3272783251231526,
+      "grad_norm": 0.08878383040428162,
+      "learning_rate": 0.01,
+      "loss": 2.0627,
+      "step": 12933
+    },
+    {
+      "epoch": 1.3275862068965516,
+      "grad_norm": 0.059183619916439056,
+      "learning_rate": 0.01,
+      "loss": 2.0829,
+      "step": 12936
+    },
+    {
+      "epoch": 1.3278940886699506,
+      "grad_norm": 0.05783310905098915,
+      "learning_rate": 0.01,
+      "loss": 2.0972,
+      "step": 12939
+    },
+    {
+      "epoch": 1.3282019704433496,
+      "grad_norm": 0.07206647843122482,
+      "learning_rate": 0.01,
+      "loss": 2.0667,
+      "step": 12942
+    },
+    {
+      "epoch": 1.3285098522167487,
+      "grad_norm": 0.07303550839424133,
+      "learning_rate": 0.01,
+      "loss": 2.0745,
+      "step": 12945
+    },
+    {
+      "epoch": 1.3288177339901477,
+      "grad_norm": 0.04319525510072708,
+      "learning_rate": 0.01,
+      "loss": 2.0673,
+      "step": 12948
+    },
+    {
+      "epoch": 1.3291256157635467,
+      "grad_norm": 0.044222913682460785,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 12951
+    },
+    {
+      "epoch": 1.3294334975369457,
+      "grad_norm": 0.08791283518075943,
+      "learning_rate": 0.01,
+      "loss": 2.0841,
+      "step": 12954
+    },
+    {
+      "epoch": 1.3297413793103448,
+      "grad_norm": 0.05172525718808174,
+      "learning_rate": 0.01,
+      "loss": 2.0828,
+      "step": 12957
+    },
+    {
+      "epoch": 1.3300492610837438,
+      "grad_norm": 0.053524646908044815,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 12960
+    },
+    {
+      "epoch": 1.3303571428571428,
+      "grad_norm": 0.03858666867017746,
+      "learning_rate": 0.01,
+      "loss": 2.1045,
+      "step": 12963
+    },
+    {
+      "epoch": 1.3306650246305418,
+      "grad_norm": 0.037037089467048645,
+      "learning_rate": 0.01,
+      "loss": 2.0742,
+      "step": 12966
+    },
+    {
+      "epoch": 1.3309729064039408,
+      "grad_norm": 0.04609520733356476,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 12969
+    },
+    {
+      "epoch": 1.3312807881773399,
+      "grad_norm": 0.03773853927850723,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 12972
+    },
+    {
+      "epoch": 1.3315886699507389,
+      "grad_norm": 0.040222879499197006,
+      "learning_rate": 0.01,
+      "loss": 2.073,
+      "step": 12975
+    },
+    {
+      "epoch": 1.331896551724138,
+      "grad_norm": 0.11522398144006729,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 12978
+    },
+    {
+      "epoch": 1.332204433497537,
+      "grad_norm": 0.0496886670589447,
+      "learning_rate": 0.01,
+      "loss": 2.0873,
+      "step": 12981
+    },
+    {
+      "epoch": 1.332512315270936,
+      "grad_norm": 0.0955866202712059,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 12984
+    },
+    {
+      "epoch": 1.332820197044335,
+      "grad_norm": 0.06260306388139725,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 12987
+    },
+    {
+      "epoch": 1.333128078817734,
+      "grad_norm": 0.049784719944000244,
+      "learning_rate": 0.01,
+      "loss": 2.0807,
+      "step": 12990
+    },
+    {
+      "epoch": 1.333435960591133,
+      "grad_norm": 0.04571852833032608,
+      "learning_rate": 0.01,
+      "loss": 2.0648,
+      "step": 12993
+    },
+    {
+      "epoch": 1.333743842364532,
+      "grad_norm": 0.11032246053218842,
+      "learning_rate": 0.01,
+      "loss": 2.0777,
+      "step": 12996
+    },
+    {
+      "epoch": 1.334051724137931,
+      "grad_norm": 0.04030182585120201,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 12999
+    },
+    {
+      "epoch": 1.33435960591133,
+      "grad_norm": 0.06681946665048599,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 13002
+    },
+    {
+      "epoch": 1.334667487684729,
+      "grad_norm": 0.04532696306705475,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 13005
+    },
+    {
+      "epoch": 1.3349753694581281,
+      "grad_norm": 0.03890594094991684,
+      "learning_rate": 0.01,
+      "loss": 2.0792,
+      "step": 13008
+    },
+    {
+      "epoch": 1.3352832512315271,
+      "grad_norm": 0.08290864527225494,
+      "learning_rate": 0.01,
+      "loss": 2.0903,
+      "step": 13011
+    },
+    {
+      "epoch": 1.3355911330049262,
+      "grad_norm": 0.09402919560670853,
+      "learning_rate": 0.01,
+      "loss": 2.0678,
+      "step": 13014
+    },
+    {
+      "epoch": 1.3358990147783252,
+      "grad_norm": 0.06945643573999405,
+      "learning_rate": 0.01,
+      "loss": 2.0823,
+      "step": 13017
+    },
+    {
+      "epoch": 1.3362068965517242,
+      "grad_norm": 0.0540471225976944,
+      "learning_rate": 0.01,
+      "loss": 2.0705,
+      "step": 13020
+    },
+    {
+      "epoch": 1.3365147783251232,
+      "grad_norm": 0.04104168713092804,
+      "learning_rate": 0.01,
+      "loss": 2.0808,
+      "step": 13023
+    },
+    {
+      "epoch": 1.3368226600985222,
+      "grad_norm": 0.04578167945146561,
+      "learning_rate": 0.01,
+      "loss": 2.0821,
+      "step": 13026
+    },
+    {
+      "epoch": 1.3371305418719213,
+      "grad_norm": 0.07289981096982956,
+      "learning_rate": 0.01,
+      "loss": 2.0686,
+      "step": 13029
+    },
+    {
+      "epoch": 1.3374384236453203,
+      "grad_norm": 0.09114310890436172,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 13032
+    },
+    {
+      "epoch": 1.3377463054187193,
+      "grad_norm": 0.06305088102817535,
+      "learning_rate": 0.01,
+      "loss": 2.0593,
+      "step": 13035
+    },
+    {
+      "epoch": 1.3380541871921183,
+      "grad_norm": 0.06495746225118637,
+      "learning_rate": 0.01,
+      "loss": 2.0728,
+      "step": 13038
+    },
+    {
+      "epoch": 1.3383620689655173,
+      "grad_norm": 0.05586539953947067,
+      "learning_rate": 0.01,
+      "loss": 2.0747,
+      "step": 13041
+    },
+    {
+      "epoch": 1.3386699507389164,
+      "grad_norm": 0.04566524177789688,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 13044
+    },
+    {
+      "epoch": 1.3389778325123154,
+      "grad_norm": 0.060839373618364334,
+      "learning_rate": 0.01,
+      "loss": 2.0717,
+      "step": 13047
+    },
+    {
+      "epoch": 1.3392857142857144,
+      "grad_norm": 0.11224903166294098,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 13050
+    },
+    {
+      "epoch": 1.3395935960591134,
+      "grad_norm": 0.09223728626966476,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 13053
+    },
+    {
+      "epoch": 1.3399014778325122,
+      "grad_norm": 0.0413731187582016,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 13056
+    },
+    {
+      "epoch": 1.3402093596059113,
+      "grad_norm": 0.050937358289957047,
+      "learning_rate": 0.01,
+      "loss": 2.0774,
+      "step": 13059
+    },
+    {
+      "epoch": 1.3405172413793103,
+      "grad_norm": 0.0407971516251564,
+      "learning_rate": 0.01,
+      "loss": 2.0766,
+      "step": 13062
+    },
+    {
+      "epoch": 1.3408251231527093,
+      "grad_norm": 0.0623883455991745,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 13065
+    },
+    {
+      "epoch": 1.3411330049261083,
+      "grad_norm": 0.09325427561998367,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 13068
+    },
+    {
+      "epoch": 1.3414408866995073,
+      "grad_norm": 0.06965765357017517,
+      "learning_rate": 0.01,
+      "loss": 2.0653,
+      "step": 13071
+    },
+    {
+      "epoch": 1.3417487684729064,
+      "grad_norm": 0.12671297788619995,
+      "learning_rate": 0.01,
+      "loss": 2.1028,
+      "step": 13074
+    },
+    {
+      "epoch": 1.3420566502463054,
+      "grad_norm": 0.04154878482222557,
+      "learning_rate": 0.01,
+      "loss": 2.0783,
+      "step": 13077
+    },
+    {
+      "epoch": 1.3423645320197044,
+      "grad_norm": 0.04698561131954193,
+      "learning_rate": 0.01,
+      "loss": 2.0799,
+      "step": 13080
+    },
+    {
+      "epoch": 1.3426724137931034,
+      "grad_norm": 0.031127501279115677,
+      "learning_rate": 0.01,
+      "loss": 2.0756,
+      "step": 13083
+    },
+    {
+      "epoch": 1.3429802955665024,
+      "grad_norm": 0.05258537083864212,
+      "learning_rate": 0.01,
+      "loss": 2.0821,
+      "step": 13086
+    },
+    {
+      "epoch": 1.3432881773399015,
+      "grad_norm": 0.06848637759685516,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 13089
+    },
+    {
+      "epoch": 1.3435960591133005,
+      "grad_norm": 0.07738485932350159,
+      "learning_rate": 0.01,
+      "loss": 2.0674,
+      "step": 13092
+    },
+    {
+      "epoch": 1.3439039408866995,
+      "grad_norm": 0.09635680168867111,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 13095
+    },
+    {
+      "epoch": 1.3442118226600985,
+      "grad_norm": 0.04388611391186714,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 13098
+    },
+    {
+      "epoch": 1.3445197044334976,
+      "grad_norm": 0.0776490792632103,
+      "learning_rate": 0.01,
+      "loss": 2.0753,
+      "step": 13101
+    },
+    {
+      "epoch": 1.3448275862068966,
+      "grad_norm": 0.11331035196781158,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 13104
+    },
+    {
+      "epoch": 1.3451354679802956,
+      "grad_norm": 0.04267279431223869,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 13107
+    },
+    {
+      "epoch": 1.3454433497536946,
+      "grad_norm": 0.05454112961888313,
+      "learning_rate": 0.01,
+      "loss": 2.0711,
+      "step": 13110
+    },
+    {
+      "epoch": 1.3457512315270936,
+      "grad_norm": 0.07470305263996124,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 13113
+    },
+    {
+      "epoch": 1.3460591133004927,
+      "grad_norm": 0.057337477803230286,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 13116
+    },
+    {
+      "epoch": 1.3463669950738917,
+      "grad_norm": 0.09155120700597763,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 13119
+    },
+    {
+      "epoch": 1.3466748768472907,
+      "grad_norm": 0.09644894301891327,
+      "learning_rate": 0.01,
+      "loss": 2.0905,
+      "step": 13122
+    },
+    {
+      "epoch": 1.3469827586206897,
+      "grad_norm": 0.0579628124833107,
+      "learning_rate": 0.01,
+      "loss": 2.0834,
+      "step": 13125
+    },
+    {
+      "epoch": 1.3472906403940887,
+      "grad_norm": 0.09968624264001846,
+      "learning_rate": 0.01,
+      "loss": 2.0797,
+      "step": 13128
+    },
+    {
+      "epoch": 1.3475985221674878,
+      "grad_norm": 0.04834052175283432,
+      "learning_rate": 0.01,
+      "loss": 2.0552,
+      "step": 13131
+    },
+    {
+      "epoch": 1.3479064039408866,
+      "grad_norm": 0.05561887100338936,
+      "learning_rate": 0.01,
+      "loss": 2.0564,
+      "step": 13134
+    },
+    {
+      "epoch": 1.3482142857142856,
+      "grad_norm": 0.14990638196468353,
+      "learning_rate": 0.01,
+      "loss": 2.0738,
+      "step": 13137
+    },
+    {
+      "epoch": 1.3485221674876846,
+      "grad_norm": 0.07530777156352997,
+      "learning_rate": 0.01,
+      "loss": 2.0765,
+      "step": 13140
+    },
+    {
+      "epoch": 1.3488300492610836,
+      "grad_norm": 0.09080106765031815,
+      "learning_rate": 0.01,
+      "loss": 2.083,
+      "step": 13143
+    },
+    {
+      "epoch": 1.3491379310344827,
+      "grad_norm": 0.042014699429273605,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 13146
+    },
+    {
+      "epoch": 1.3494458128078817,
+      "grad_norm": 0.08905219286680222,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 13149
+    },
+    {
+      "epoch": 1.3497536945812807,
+      "grad_norm": 0.1093059629201889,
+      "learning_rate": 0.01,
+      "loss": 2.102,
+      "step": 13152
+    },
+    {
+      "epoch": 1.3500615763546797,
+      "grad_norm": 0.09834617376327515,
+      "learning_rate": 0.01,
+      "loss": 2.0601,
+      "step": 13155
+    },
+    {
+      "epoch": 1.3503694581280787,
+      "grad_norm": 0.0754542201757431,
+      "learning_rate": 0.01,
+      "loss": 2.103,
+      "step": 13158
+    },
+    {
+      "epoch": 1.3506773399014778,
+      "grad_norm": 0.09639342129230499,
+      "learning_rate": 0.01,
+      "loss": 2.0743,
+      "step": 13161
+    },
+    {
+      "epoch": 1.3509852216748768,
+      "grad_norm": 0.05084405094385147,
+      "learning_rate": 0.01,
+      "loss": 2.0898,
+      "step": 13164
+    },
+    {
+      "epoch": 1.3512931034482758,
+      "grad_norm": 0.04796381667256355,
+      "learning_rate": 0.01,
+      "loss": 2.0788,
+      "step": 13167
+    },
+    {
+      "epoch": 1.3516009852216748,
+      "grad_norm": 0.05373486131429672,
+      "learning_rate": 0.01,
+      "loss": 2.0632,
+      "step": 13170
+    },
+    {
+      "epoch": 1.3519088669950738,
+      "grad_norm": 0.05145580321550369,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 13173
+    },
+    {
+      "epoch": 1.3522167487684729,
+      "grad_norm": 0.03861214593052864,
+      "learning_rate": 0.01,
+      "loss": 2.0678,
+      "step": 13176
+    },
+    {
+      "epoch": 1.3525246305418719,
+      "grad_norm": 0.04394346475601196,
+      "learning_rate": 0.01,
+      "loss": 2.0601,
+      "step": 13179
+    },
+    {
+      "epoch": 1.352832512315271,
+      "grad_norm": 0.08851804584264755,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 13182
+    },
+    {
+      "epoch": 1.35314039408867,
+      "grad_norm": 0.059799451380968094,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 13185
+    },
+    {
+      "epoch": 1.353448275862069,
+      "grad_norm": 0.13764169812202454,
+      "learning_rate": 0.01,
+      "loss": 2.1001,
+      "step": 13188
+    },
+    {
+      "epoch": 1.353756157635468,
+      "grad_norm": 0.05652278661727905,
+      "learning_rate": 0.01,
+      "loss": 2.0771,
+      "step": 13191
+    },
+    {
+      "epoch": 1.354064039408867,
+      "grad_norm": 0.04755775257945061,
+      "learning_rate": 0.01,
+      "loss": 2.0467,
+      "step": 13194
+    },
+    {
+      "epoch": 1.354371921182266,
+      "grad_norm": 0.058131635189056396,
+      "learning_rate": 0.01,
+      "loss": 2.0884,
+      "step": 13197
+    },
+    {
+      "epoch": 1.354679802955665,
+      "grad_norm": 0.041266053915023804,
+      "learning_rate": 0.01,
+      "loss": 2.0684,
+      "step": 13200
+    },
+    {
+      "epoch": 1.354987684729064,
+      "grad_norm": 0.034990034997463226,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 13203
+    },
+    {
+      "epoch": 1.355295566502463,
+      "grad_norm": 0.13107064366340637,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 13206
+    },
+    {
+      "epoch": 1.355603448275862,
+      "grad_norm": 0.05397200584411621,
+      "learning_rate": 0.01,
+      "loss": 2.0953,
+      "step": 13209
+    },
+    {
+      "epoch": 1.3559113300492611,
+      "grad_norm": 0.04137737303972244,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 13212
+    },
+    {
+      "epoch": 1.3562192118226601,
+      "grad_norm": 0.05001407861709595,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 13215
+    },
+    {
+      "epoch": 1.3565270935960592,
+      "grad_norm": 0.10387953370809555,
+      "learning_rate": 0.01,
+      "loss": 2.0818,
+      "step": 13218
+    },
+    {
+      "epoch": 1.3568349753694582,
+      "grad_norm": 0.052998363971710205,
+      "learning_rate": 0.01,
+      "loss": 2.0711,
+      "step": 13221
+    },
+    {
+      "epoch": 1.3571428571428572,
+      "grad_norm": 0.06805765628814697,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 13224
+    },
+    {
+      "epoch": 1.3574507389162562,
+      "grad_norm": 0.06597940623760223,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 13227
+    },
+    {
+      "epoch": 1.3577586206896552,
+      "grad_norm": 0.10083628445863724,
+      "learning_rate": 0.01,
+      "loss": 2.086,
+      "step": 13230
+    },
+    {
+      "epoch": 1.3580665024630543,
+      "grad_norm": 0.05467986315488815,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 13233
+    },
+    {
+      "epoch": 1.3583743842364533,
+      "grad_norm": 0.08951261639595032,
+      "learning_rate": 0.01,
+      "loss": 2.1256,
+      "step": 13236
+    },
+    {
+      "epoch": 1.3586822660098523,
+      "grad_norm": 0.052532244473695755,
+      "learning_rate": 0.01,
+      "loss": 2.0555,
+      "step": 13239
+    },
+    {
+      "epoch": 1.3589901477832513,
+      "grad_norm": 0.038159146904945374,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 13242
+    },
+    {
+      "epoch": 1.3592980295566504,
+      "grad_norm": 0.09895820915699005,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 13245
+    },
+    {
+      "epoch": 1.3596059113300494,
+      "grad_norm": 0.07522387057542801,
+      "learning_rate": 0.01,
+      "loss": 2.0634,
+      "step": 13248
+    },
+    {
+      "epoch": 1.3599137931034484,
+      "grad_norm": 0.04762687534093857,
+      "learning_rate": 0.01,
+      "loss": 2.0848,
+      "step": 13251
+    },
+    {
+      "epoch": 1.3602216748768474,
+      "grad_norm": 0.0716032013297081,
+      "learning_rate": 0.01,
+      "loss": 2.0783,
+      "step": 13254
+    },
+    {
+      "epoch": 1.3605295566502464,
+      "grad_norm": 0.08518968522548676,
+      "learning_rate": 0.01,
+      "loss": 2.0818,
+      "step": 13257
+    },
+    {
+      "epoch": 1.3608374384236452,
+      "grad_norm": 0.06927520781755447,
+      "learning_rate": 0.01,
+      "loss": 2.0728,
+      "step": 13260
+    },
+    {
+      "epoch": 1.3611453201970443,
+      "grad_norm": 0.10368376970291138,
+      "learning_rate": 0.01,
+      "loss": 2.0419,
+      "step": 13263
+    },
+    {
+      "epoch": 1.3614532019704433,
+      "grad_norm": 0.04249117895960808,
+      "learning_rate": 0.01,
+      "loss": 2.0695,
+      "step": 13266
+    },
+    {
+      "epoch": 1.3617610837438423,
+      "grad_norm": 0.06504488736391068,
+      "learning_rate": 0.01,
+      "loss": 2.0658,
+      "step": 13269
+    },
+    {
+      "epoch": 1.3620689655172413,
+      "grad_norm": 0.03990466147661209,
+      "learning_rate": 0.01,
+      "loss": 2.0796,
+      "step": 13272
+    },
+    {
+      "epoch": 1.3623768472906403,
+      "grad_norm": 0.042559072375297546,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 13275
+    },
+    {
+      "epoch": 1.3626847290640394,
+      "grad_norm": 0.046650230884552,
+      "learning_rate": 0.01,
+      "loss": 2.0756,
+      "step": 13278
+    },
+    {
+      "epoch": 1.3629926108374384,
+      "grad_norm": 0.08641167730093002,
+      "learning_rate": 0.01,
+      "loss": 2.0638,
+      "step": 13281
+    },
+    {
+      "epoch": 1.3633004926108374,
+      "grad_norm": 0.11438708007335663,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 13284
+    },
+    {
+      "epoch": 1.3636083743842364,
+      "grad_norm": 0.05360870808362961,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 13287
+    },
+    {
+      "epoch": 1.3639162561576355,
+      "grad_norm": 0.07226021587848663,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 13290
+    },
+    {
+      "epoch": 1.3642241379310345,
+      "grad_norm": 0.04532739892601967,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 13293
+    },
+    {
+      "epoch": 1.3645320197044335,
+      "grad_norm": 0.06119906157255173,
+      "learning_rate": 0.01,
+      "loss": 2.0507,
+      "step": 13296
+    },
+    {
+      "epoch": 1.3648399014778325,
+      "grad_norm": 0.05576052516698837,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 13299
+    },
+    {
+      "epoch": 1.3651477832512315,
+      "grad_norm": 0.038748834282159805,
+      "learning_rate": 0.01,
+      "loss": 2.0632,
+      "step": 13302
+    },
+    {
+      "epoch": 1.3654556650246306,
+      "grad_norm": 0.09733711183071136,
+      "learning_rate": 0.01,
+      "loss": 2.0733,
+      "step": 13305
+    },
+    {
+      "epoch": 1.3657635467980296,
+      "grad_norm": 0.043375931680202484,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 13308
+    },
+    {
+      "epoch": 1.3660714285714286,
+      "grad_norm": 0.11930018663406372,
+      "learning_rate": 0.01,
+      "loss": 2.0931,
+      "step": 13311
+    },
+    {
+      "epoch": 1.3663793103448276,
+      "grad_norm": 0.06754540652036667,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 13314
+    },
+    {
+      "epoch": 1.3666871921182266,
+      "grad_norm": 0.07226148992776871,
+      "learning_rate": 0.01,
+      "loss": 2.0786,
+      "step": 13317
+    },
+    {
+      "epoch": 1.3669950738916257,
+      "grad_norm": 0.08159705251455307,
+      "learning_rate": 0.01,
+      "loss": 2.0594,
+      "step": 13320
+    },
+    {
+      "epoch": 1.3673029556650247,
+      "grad_norm": 0.044994477182626724,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 13323
+    },
+    {
+      "epoch": 1.3676108374384237,
+      "grad_norm": 0.05308050662279129,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 13326
+    },
+    {
+      "epoch": 1.3679187192118227,
+      "grad_norm": 0.09141236543655396,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 13329
+    },
+    {
+      "epoch": 1.3682266009852218,
+      "grad_norm": 0.040702883154153824,
+      "learning_rate": 0.01,
+      "loss": 2.0497,
+      "step": 13332
+    },
+    {
+      "epoch": 1.3685344827586206,
+      "grad_norm": 0.06111524999141693,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 13335
+    },
+    {
+      "epoch": 1.3688423645320196,
+      "grad_norm": 0.10802275687456131,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 13338
+    },
+    {
+      "epoch": 1.3691502463054186,
+      "grad_norm": 0.09049344807863235,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 13341
+    },
+    {
+      "epoch": 1.3694581280788176,
+      "grad_norm": 0.055894456803798676,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 13344
+    },
+    {
+      "epoch": 1.3697660098522166,
+      "grad_norm": 0.054729919880628586,
+      "learning_rate": 0.01,
+      "loss": 2.0674,
+      "step": 13347
+    },
+    {
+      "epoch": 1.3700738916256157,
+      "grad_norm": 0.05745011568069458,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 13350
+    },
+    {
+      "epoch": 1.3703817733990147,
+      "grad_norm": 0.06573651731014252,
+      "learning_rate": 0.01,
+      "loss": 2.0493,
+      "step": 13353
+    },
+    {
+      "epoch": 1.3706896551724137,
+      "grad_norm": 0.20495210587978363,
+      "learning_rate": 0.01,
+      "loss": 2.0798,
+      "step": 13356
+    },
+    {
+      "epoch": 1.3709975369458127,
+      "grad_norm": 0.10678639262914658,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 13359
+    },
+    {
+      "epoch": 1.3713054187192117,
+      "grad_norm": 0.10948281735181808,
+      "learning_rate": 0.01,
+      "loss": 2.087,
+      "step": 13362
+    },
+    {
+      "epoch": 1.3716133004926108,
+      "grad_norm": 0.07788719981908798,
+      "learning_rate": 0.01,
+      "loss": 2.0745,
+      "step": 13365
+    },
+    {
+      "epoch": 1.3719211822660098,
+      "grad_norm": 0.04947768524289131,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 13368
+    },
+    {
+      "epoch": 1.3722290640394088,
+      "grad_norm": 0.04789843037724495,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 13371
+    },
+    {
+      "epoch": 1.3725369458128078,
+      "grad_norm": 0.05217898637056351,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 13374
+    },
+    {
+      "epoch": 1.3728448275862069,
+      "grad_norm": 0.04018987715244293,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 13377
+    },
+    {
+      "epoch": 1.3731527093596059,
+      "grad_norm": 0.035446904599666595,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 13380
+    },
+    {
+      "epoch": 1.373460591133005,
+      "grad_norm": 0.038600854575634,
+      "learning_rate": 0.01,
+      "loss": 2.0865,
+      "step": 13383
+    },
+    {
+      "epoch": 1.373768472906404,
+      "grad_norm": 0.055341143161058426,
+      "learning_rate": 0.01,
+      "loss": 2.0591,
+      "step": 13386
+    },
+    {
+      "epoch": 1.374076354679803,
+      "grad_norm": 0.11673317849636078,
+      "learning_rate": 0.01,
+      "loss": 2.0897,
+      "step": 13389
+    },
+    {
+      "epoch": 1.374384236453202,
+      "grad_norm": 0.06797752529382706,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 13392
+    },
+    {
+      "epoch": 1.374692118226601,
+      "grad_norm": 0.041885413229465485,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 13395
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 0.07391481846570969,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 13398
+    },
+    {
+      "epoch": 1.375307881773399,
+      "grad_norm": 0.07959283888339996,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 13401
+    },
+    {
+      "epoch": 1.375615763546798,
+      "grad_norm": 0.09504754096269608,
+      "learning_rate": 0.01,
+      "loss": 2.0841,
+      "step": 13404
+    },
+    {
+      "epoch": 1.375923645320197,
+      "grad_norm": 0.08874071389436722,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 13407
+    },
+    {
+      "epoch": 1.376231527093596,
+      "grad_norm": 0.05350710079073906,
+      "learning_rate": 0.01,
+      "loss": 2.0763,
+      "step": 13410
+    },
+    {
+      "epoch": 1.376539408866995,
+      "grad_norm": 0.06738609820604324,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 13413
+    },
+    {
+      "epoch": 1.3768472906403941,
+      "grad_norm": 0.04335073009133339,
+      "learning_rate": 0.01,
+      "loss": 2.0601,
+      "step": 13416
+    },
+    {
+      "epoch": 1.3771551724137931,
+      "grad_norm": 0.056045398116111755,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 13419
+    },
+    {
+      "epoch": 1.3774630541871922,
+      "grad_norm": 0.06593155860900879,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 13422
+    },
+    {
+      "epoch": 1.3777709359605912,
+      "grad_norm": 0.08942624181509018,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 13425
+    },
+    {
+      "epoch": 1.3780788177339902,
+      "grad_norm": 0.10098058730363846,
+      "learning_rate": 0.01,
+      "loss": 2.064,
+      "step": 13428
+    },
+    {
+      "epoch": 1.3783866995073892,
+      "grad_norm": 0.04804680123925209,
+      "learning_rate": 0.01,
+      "loss": 2.0684,
+      "step": 13431
+    },
+    {
+      "epoch": 1.3786945812807883,
+      "grad_norm": 0.08567364513874054,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 13434
+    },
+    {
+      "epoch": 1.3790024630541873,
+      "grad_norm": 0.06027091667056084,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 13437
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.08809462189674377,
+      "learning_rate": 0.01,
+      "loss": 2.0862,
+      "step": 13440
+    },
+    {
+      "epoch": 1.3796182266009853,
+      "grad_norm": 0.053466469049453735,
+      "learning_rate": 0.01,
+      "loss": 2.0497,
+      "step": 13443
+    },
+    {
+      "epoch": 1.3799261083743843,
+      "grad_norm": 0.033619511872529984,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 13446
+    },
+    {
+      "epoch": 1.3802339901477834,
+      "grad_norm": 0.04768878221511841,
+      "learning_rate": 0.01,
+      "loss": 2.054,
+      "step": 13449
+    },
+    {
+      "epoch": 1.3805418719211824,
+      "grad_norm": 0.07854757457971573,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 13452
+    },
+    {
+      "epoch": 1.3808497536945814,
+      "grad_norm": 0.05409559607505798,
+      "learning_rate": 0.01,
+      "loss": 2.082,
+      "step": 13455
+    },
+    {
+      "epoch": 1.3811576354679804,
+      "grad_norm": 0.057855118066072464,
+      "learning_rate": 0.01,
+      "loss": 2.0814,
+      "step": 13458
+    },
+    {
+      "epoch": 1.3814655172413794,
+      "grad_norm": 0.047502126544713974,
+      "learning_rate": 0.01,
+      "loss": 2.0955,
+      "step": 13461
+    },
+    {
+      "epoch": 1.3817733990147782,
+      "grad_norm": 0.040939487516880035,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 13464
+    },
+    {
+      "epoch": 1.3820812807881773,
+      "grad_norm": 0.1307850480079651,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 13467
+    },
+    {
+      "epoch": 1.3823891625615763,
+      "grad_norm": 0.04386845603585243,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 13470
+    },
+    {
+      "epoch": 1.3826970443349753,
+      "grad_norm": 0.08174968510866165,
+      "learning_rate": 0.01,
+      "loss": 2.0544,
+      "step": 13473
+    },
+    {
+      "epoch": 1.3830049261083743,
+      "grad_norm": 0.1113237589597702,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 13476
+    },
+    {
+      "epoch": 1.3833128078817734,
+      "grad_norm": 0.06756308674812317,
+      "learning_rate": 0.01,
+      "loss": 2.0813,
+      "step": 13479
+    },
+    {
+      "epoch": 1.3836206896551724,
+      "grad_norm": 0.05931835621595383,
+      "learning_rate": 0.01,
+      "loss": 2.0544,
+      "step": 13482
+    },
+    {
+      "epoch": 1.3839285714285714,
+      "grad_norm": 0.043539129197597504,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 13485
+    },
+    {
+      "epoch": 1.3842364532019704,
+      "grad_norm": 0.0510721355676651,
+      "learning_rate": 0.01,
+      "loss": 2.075,
+      "step": 13488
+    },
+    {
+      "epoch": 1.3845443349753694,
+      "grad_norm": 0.10811501741409302,
+      "learning_rate": 0.01,
+      "loss": 2.0626,
+      "step": 13491
+    },
+    {
+      "epoch": 1.3848522167487685,
+      "grad_norm": 0.08322811871767044,
+      "learning_rate": 0.01,
+      "loss": 2.0882,
+      "step": 13494
+    },
+    {
+      "epoch": 1.3851600985221675,
+      "grad_norm": 0.05101979896426201,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 13497
+    },
+    {
+      "epoch": 1.3854679802955665,
+      "grad_norm": 0.036535851657390594,
+      "learning_rate": 0.01,
+      "loss": 2.0791,
+      "step": 13500
+    },
+    {
+      "epoch": 1.3857758620689655,
+      "grad_norm": 0.05161239951848984,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 13503
+    },
+    {
+      "epoch": 1.3860837438423645,
+      "grad_norm": 0.06677153706550598,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 13506
+    },
+    {
+      "epoch": 1.3863916256157636,
+      "grad_norm": 0.04239841178059578,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 13509
+    },
+    {
+      "epoch": 1.3866995073891626,
+      "grad_norm": 0.04252205416560173,
+      "learning_rate": 0.01,
+      "loss": 2.0875,
+      "step": 13512
+    },
+    {
+      "epoch": 1.3870073891625616,
+      "grad_norm": 0.10147176682949066,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 13515
+    },
+    {
+      "epoch": 1.3873152709359606,
+      "grad_norm": 0.046371154487133026,
+      "learning_rate": 0.01,
+      "loss": 2.0784,
+      "step": 13518
+    },
+    {
+      "epoch": 1.3876231527093597,
+      "grad_norm": 0.0997064933180809,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 13521
+    },
+    {
+      "epoch": 1.3879310344827587,
+      "grad_norm": 0.05733582749962807,
+      "learning_rate": 0.01,
+      "loss": 2.0742,
+      "step": 13524
+    },
+    {
+      "epoch": 1.3882389162561577,
+      "grad_norm": 0.05061260983347893,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 13527
+    },
+    {
+      "epoch": 1.3885467980295567,
+      "grad_norm": 0.04336051642894745,
+      "learning_rate": 0.01,
+      "loss": 2.0846,
+      "step": 13530
+    },
+    {
+      "epoch": 1.3888546798029557,
+      "grad_norm": 0.04115337133407593,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 13533
+    },
+    {
+      "epoch": 1.3891625615763548,
+      "grad_norm": 0.04914069175720215,
+      "learning_rate": 0.01,
+      "loss": 2.0815,
+      "step": 13536
+    },
+    {
+      "epoch": 1.3894704433497536,
+      "grad_norm": 0.0677042305469513,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 13539
+    },
+    {
+      "epoch": 1.3897783251231526,
+      "grad_norm": 0.04985208064317703,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 13542
+    },
+    {
+      "epoch": 1.3900862068965516,
+      "grad_norm": 0.08042199164628983,
+      "learning_rate": 0.01,
+      "loss": 2.0828,
+      "step": 13545
+    },
+    {
+      "epoch": 1.3903940886699506,
+      "grad_norm": 0.03648814186453819,
+      "learning_rate": 0.01,
+      "loss": 2.0669,
+      "step": 13548
+    },
+    {
+      "epoch": 1.3907019704433496,
+      "grad_norm": 0.03645399957895279,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 13551
+    },
+    {
+      "epoch": 1.3910098522167487,
+      "grad_norm": 0.04866683483123779,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 13554
+    },
+    {
+      "epoch": 1.3913177339901477,
+      "grad_norm": 0.07728299498558044,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 13557
+    },
+    {
+      "epoch": 1.3916256157635467,
+      "grad_norm": 0.12097810208797455,
+      "learning_rate": 0.01,
+      "loss": 2.0935,
+      "step": 13560
+    },
+    {
+      "epoch": 1.3919334975369457,
+      "grad_norm": 0.12485776096582413,
+      "learning_rate": 0.01,
+      "loss": 2.0712,
+      "step": 13563
+    },
+    {
+      "epoch": 1.3922413793103448,
+      "grad_norm": 0.053524475544691086,
+      "learning_rate": 0.01,
+      "loss": 2.0595,
+      "step": 13566
+    },
+    {
+      "epoch": 1.3925492610837438,
+      "grad_norm": 0.04277713969349861,
+      "learning_rate": 0.01,
+      "loss": 2.0791,
+      "step": 13569
+    },
+    {
+      "epoch": 1.3928571428571428,
+      "grad_norm": 0.09847384691238403,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 13572
+    },
+    {
+      "epoch": 1.3931650246305418,
+      "grad_norm": 0.03410463035106659,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 13575
+    },
+    {
+      "epoch": 1.3934729064039408,
+      "grad_norm": 0.10606678575277328,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 13578
+    },
+    {
+      "epoch": 1.3937807881773399,
+      "grad_norm": 0.06554549187421799,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 13581
+    },
+    {
+      "epoch": 1.3940886699507389,
+      "grad_norm": 0.07487329095602036,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 13584
+    },
+    {
+      "epoch": 1.394396551724138,
+      "grad_norm": 0.07526996731758118,
+      "learning_rate": 0.01,
+      "loss": 2.0733,
+      "step": 13587
+    },
+    {
+      "epoch": 1.394704433497537,
+      "grad_norm": 0.0581665076315403,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 13590
+    },
+    {
+      "epoch": 1.395012315270936,
+      "grad_norm": 0.057513732463121414,
+      "learning_rate": 0.01,
+      "loss": 2.087,
+      "step": 13593
+    },
+    {
+      "epoch": 1.395320197044335,
+      "grad_norm": 0.037192508578300476,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 13596
+    },
+    {
+      "epoch": 1.395628078817734,
+      "grad_norm": 0.0965125560760498,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 13599
+    },
+    {
+      "epoch": 1.395935960591133,
+      "grad_norm": 0.04594407603144646,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 13602
+    },
+    {
+      "epoch": 1.396243842364532,
+      "grad_norm": 0.08442319929599762,
+      "learning_rate": 0.01,
+      "loss": 2.0627,
+      "step": 13605
+    },
+    {
+      "epoch": 1.396551724137931,
+      "grad_norm": 0.08673713356256485,
+      "learning_rate": 0.01,
+      "loss": 2.054,
+      "step": 13608
+    },
+    {
+      "epoch": 1.39685960591133,
+      "grad_norm": 0.07299968600273132,
+      "learning_rate": 0.01,
+      "loss": 2.0672,
+      "step": 13611
+    },
+    {
+      "epoch": 1.397167487684729,
+      "grad_norm": 0.052630744874477386,
+      "learning_rate": 0.01,
+      "loss": 2.079,
+      "step": 13614
+    },
+    {
+      "epoch": 1.3974753694581281,
+      "grad_norm": 0.0626215934753418,
+      "learning_rate": 0.01,
+      "loss": 2.0732,
+      "step": 13617
+    },
+    {
+      "epoch": 1.3977832512315271,
+      "grad_norm": 0.0866907387971878,
+      "learning_rate": 0.01,
+      "loss": 2.0743,
+      "step": 13620
+    },
+    {
+      "epoch": 1.3980911330049262,
+      "grad_norm": 0.05650071054697037,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 13623
+    },
+    {
+      "epoch": 1.3983990147783252,
+      "grad_norm": 0.07526635378599167,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 13626
+    },
+    {
+      "epoch": 1.3987068965517242,
+      "grad_norm": 0.07472112774848938,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 13629
+    },
+    {
+      "epoch": 1.3990147783251232,
+      "grad_norm": 0.07251216471195221,
+      "learning_rate": 0.01,
+      "loss": 2.086,
+      "step": 13632
+    },
+    {
+      "epoch": 1.3993226600985222,
+      "grad_norm": 0.08701921999454498,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 13635
+    },
+    {
+      "epoch": 1.3996305418719213,
+      "grad_norm": 0.033323436975479126,
+      "learning_rate": 0.01,
+      "loss": 2.0716,
+      "step": 13638
+    },
+    {
+      "epoch": 1.3999384236453203,
+      "grad_norm": 0.04960713908076286,
+      "learning_rate": 0.01,
+      "loss": 2.0557,
+      "step": 13641
+    },
+    {
+      "epoch": 1.4002463054187193,
+      "grad_norm": 0.1418198049068451,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 13644
+    },
+    {
+      "epoch": 1.4005541871921183,
+      "grad_norm": 0.06056910380721092,
+      "learning_rate": 0.01,
+      "loss": 2.0825,
+      "step": 13647
+    },
+    {
+      "epoch": 1.4008620689655173,
+      "grad_norm": 0.062474992126226425,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 13650
+    },
+    {
+      "epoch": 1.4011699507389164,
+      "grad_norm": 0.05380658805370331,
+      "learning_rate": 0.01,
+      "loss": 2.0644,
+      "step": 13653
+    },
+    {
+      "epoch": 1.4014778325123154,
+      "grad_norm": 0.046230513602495193,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 13656
+    },
+    {
+      "epoch": 1.4017857142857144,
+      "grad_norm": 0.05238807573914528,
+      "learning_rate": 0.01,
+      "loss": 2.0976,
+      "step": 13659
+    },
+    {
+      "epoch": 1.4020935960591134,
+      "grad_norm": 0.045423392206430435,
+      "learning_rate": 0.01,
+      "loss": 2.0729,
+      "step": 13662
+    },
+    {
+      "epoch": 1.4024014778325122,
+      "grad_norm": 0.1077577993273735,
+      "learning_rate": 0.01,
+      "loss": 2.0481,
+      "step": 13665
+    },
+    {
+      "epoch": 1.4027093596059113,
+      "grad_norm": 0.04370421916246414,
+      "learning_rate": 0.01,
+      "loss": 2.0634,
+      "step": 13668
+    },
+    {
+      "epoch": 1.4030172413793103,
+      "grad_norm": 0.061496302485466,
+      "learning_rate": 0.01,
+      "loss": 2.0449,
+      "step": 13671
+    },
+    {
+      "epoch": 1.4033251231527093,
+      "grad_norm": 0.048742834478616714,
+      "learning_rate": 0.01,
+      "loss": 2.0683,
+      "step": 13674
+    },
+    {
+      "epoch": 1.4036330049261083,
+      "grad_norm": 0.14942513406276703,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 13677
+    },
+    {
+      "epoch": 1.4039408866995073,
+      "grad_norm": 0.04235846549272537,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 13680
+    },
+    {
+      "epoch": 1.4042487684729064,
+      "grad_norm": 0.05509978160262108,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 13683
+    },
+    {
+      "epoch": 1.4045566502463054,
+      "grad_norm": 0.09692233055830002,
+      "learning_rate": 0.01,
+      "loss": 2.0703,
+      "step": 13686
+    },
+    {
+      "epoch": 1.4048645320197044,
+      "grad_norm": 0.11141908913850784,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 13689
+    },
+    {
+      "epoch": 1.4051724137931034,
+      "grad_norm": 0.06601562350988388,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 13692
+    },
+    {
+      "epoch": 1.4054802955665024,
+      "grad_norm": 0.04997260868549347,
+      "learning_rate": 0.01,
+      "loss": 2.056,
+      "step": 13695
+    },
+    {
+      "epoch": 1.4057881773399015,
+      "grad_norm": 0.07198163866996765,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 13698
+    },
+    {
+      "epoch": 1.4060960591133005,
+      "grad_norm": 0.03802650794386864,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 13701
+    },
+    {
+      "epoch": 1.4064039408866995,
+      "grad_norm": 0.06259030848741531,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 13704
+    },
+    {
+      "epoch": 1.4067118226600985,
+      "grad_norm": 0.09554235637187958,
+      "learning_rate": 0.01,
+      "loss": 2.0476,
+      "step": 13707
+    },
+    {
+      "epoch": 1.4070197044334976,
+      "grad_norm": 0.056935038417577744,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 13710
+    },
+    {
+      "epoch": 1.4073275862068966,
+      "grad_norm": 0.11038411408662796,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 13713
+    },
+    {
+      "epoch": 1.4076354679802956,
+      "grad_norm": 0.05257488042116165,
+      "learning_rate": 0.01,
+      "loss": 2.09,
+      "step": 13716
+    },
+    {
+      "epoch": 1.4079433497536946,
+      "grad_norm": 0.0573866032063961,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 13719
+    },
+    {
+      "epoch": 1.4082512315270936,
+      "grad_norm": 0.04933631047606468,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 13722
+    },
+    {
+      "epoch": 1.4085591133004927,
+      "grad_norm": 0.05909980088472366,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 13725
+    },
+    {
+      "epoch": 1.4088669950738917,
+      "grad_norm": 0.09598751366138458,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 13728
+    },
+    {
+      "epoch": 1.4091748768472907,
+      "grad_norm": 0.05608231574296951,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 13731
+    },
+    {
+      "epoch": 1.4094827586206897,
+      "grad_norm": 0.08262834697961807,
+      "learning_rate": 0.01,
+      "loss": 2.0661,
+      "step": 13734
+    },
+    {
+      "epoch": 1.4097906403940887,
+      "grad_norm": 0.041144959628582,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 13737
+    },
+    {
+      "epoch": 1.4100985221674878,
+      "grad_norm": 0.03748650476336479,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 13740
+    },
+    {
+      "epoch": 1.4104064039408866,
+      "grad_norm": 0.04054822400212288,
+      "learning_rate": 0.01,
+      "loss": 2.0564,
+      "step": 13743
+    },
+    {
+      "epoch": 1.4107142857142856,
+      "grad_norm": 0.07961263507604599,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 13746
+    },
+    {
+      "epoch": 1.4110221674876846,
+      "grad_norm": 0.049971841275691986,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 13749
+    },
+    {
+      "epoch": 1.4113300492610836,
+      "grad_norm": 0.040059734135866165,
+      "learning_rate": 0.01,
+      "loss": 2.0791,
+      "step": 13752
+    },
+    {
+      "epoch": 1.4116379310344827,
+      "grad_norm": 0.0400179885327816,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 13755
+    },
+    {
+      "epoch": 1.4119458128078817,
+      "grad_norm": 0.04587862268090248,
+      "learning_rate": 0.01,
+      "loss": 2.0572,
+      "step": 13758
+    },
+    {
+      "epoch": 1.4122536945812807,
+      "grad_norm": 0.08982817828655243,
+      "learning_rate": 0.01,
+      "loss": 2.0864,
+      "step": 13761
+    },
+    {
+      "epoch": 1.4125615763546797,
+      "grad_norm": 0.05488836392760277,
+      "learning_rate": 0.01,
+      "loss": 2.0529,
+      "step": 13764
+    },
+    {
+      "epoch": 1.4128694581280787,
+      "grad_norm": 0.06559593975543976,
+      "learning_rate": 0.01,
+      "loss": 2.0564,
+      "step": 13767
+    },
+    {
+      "epoch": 1.4131773399014778,
+      "grad_norm": 0.10647718608379364,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 13770
+    },
+    {
+      "epoch": 1.4134852216748768,
+      "grad_norm": 0.05944173410534859,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 13773
+    },
+    {
+      "epoch": 1.4137931034482758,
+      "grad_norm": 0.05548718199133873,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 13776
+    },
+    {
+      "epoch": 1.4141009852216748,
+      "grad_norm": 0.0694265142083168,
+      "learning_rate": 0.01,
+      "loss": 2.0647,
+      "step": 13779
+    },
+    {
+      "epoch": 1.4144088669950738,
+      "grad_norm": 0.10526683181524277,
+      "learning_rate": 0.01,
+      "loss": 2.0768,
+      "step": 13782
+    },
+    {
+      "epoch": 1.4147167487684729,
+      "grad_norm": 0.08820123970508575,
+      "learning_rate": 0.01,
+      "loss": 2.0693,
+      "step": 13785
+    },
+    {
+      "epoch": 1.4150246305418719,
+      "grad_norm": 0.04513731971383095,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 13788
+    },
+    {
+      "epoch": 1.415332512315271,
+      "grad_norm": 0.05737076327204704,
+      "learning_rate": 0.01,
+      "loss": 2.0698,
+      "step": 13791
+    },
+    {
+      "epoch": 1.41564039408867,
+      "grad_norm": 0.0431799478828907,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 13794
+    },
+    {
+      "epoch": 1.415948275862069,
+      "grad_norm": 0.09012471139431,
+      "learning_rate": 0.01,
+      "loss": 2.0634,
+      "step": 13797
+    },
+    {
+      "epoch": 1.416256157635468,
+      "grad_norm": 0.05895904824137688,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 13800
+    },
+    {
+      "epoch": 1.416564039408867,
+      "grad_norm": 0.1610986888408661,
+      "learning_rate": 0.01,
+      "loss": 2.0743,
+      "step": 13803
+    },
+    {
+      "epoch": 1.416871921182266,
+      "grad_norm": 0.07852904498577118,
+      "learning_rate": 0.01,
+      "loss": 2.064,
+      "step": 13806
+    },
+    {
+      "epoch": 1.417179802955665,
+      "grad_norm": 0.06620481610298157,
+      "learning_rate": 0.01,
+      "loss": 2.0688,
+      "step": 13809
+    },
+    {
+      "epoch": 1.417487684729064,
+      "grad_norm": 0.033222537487745285,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 13812
+    },
+    {
+      "epoch": 1.417795566502463,
+      "grad_norm": 0.02942623570561409,
+      "learning_rate": 0.01,
+      "loss": 2.0654,
+      "step": 13815
+    },
+    {
+      "epoch": 1.418103448275862,
+      "grad_norm": 0.03543059900403023,
+      "learning_rate": 0.01,
+      "loss": 2.0776,
+      "step": 13818
+    },
+    {
+      "epoch": 1.4184113300492611,
+      "grad_norm": 0.13414567708969116,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 13821
+    },
+    {
+      "epoch": 1.4187192118226601,
+      "grad_norm": 0.06474481523036957,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 13824
+    },
+    {
+      "epoch": 1.4190270935960592,
+      "grad_norm": 0.11285994201898575,
+      "learning_rate": 0.01,
+      "loss": 2.0533,
+      "step": 13827
+    },
+    {
+      "epoch": 1.4193349753694582,
+      "grad_norm": 0.05104577913880348,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 13830
+    },
+    {
+      "epoch": 1.4196428571428572,
+      "grad_norm": 0.05463656783103943,
+      "learning_rate": 0.01,
+      "loss": 2.0788,
+      "step": 13833
+    },
+    {
+      "epoch": 1.4199507389162562,
+      "grad_norm": 0.07886187732219696,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 13836
+    },
+    {
+      "epoch": 1.4202586206896552,
+      "grad_norm": 0.06960279494524002,
+      "learning_rate": 0.01,
+      "loss": 2.0927,
+      "step": 13839
+    },
+    {
+      "epoch": 1.4205665024630543,
+      "grad_norm": 0.07481426745653152,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 13842
+    },
+    {
+      "epoch": 1.4208743842364533,
+      "grad_norm": 0.04317006468772888,
+      "learning_rate": 0.01,
+      "loss": 2.0665,
+      "step": 13845
+    },
+    {
+      "epoch": 1.4211822660098523,
+      "grad_norm": 0.10644064098596573,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 13848
+    },
+    {
+      "epoch": 1.4214901477832513,
+      "grad_norm": 0.09246213734149933,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 13851
+    },
+    {
+      "epoch": 1.4217980295566504,
+      "grad_norm": 0.05824518948793411,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 13854
+    },
+    {
+      "epoch": 1.4221059113300494,
+      "grad_norm": 0.06316854059696198,
+      "learning_rate": 0.01,
+      "loss": 2.076,
+      "step": 13857
+    },
+    {
+      "epoch": 1.4224137931034484,
+      "grad_norm": 0.058339640498161316,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 13860
+    },
+    {
+      "epoch": 1.4227216748768474,
+      "grad_norm": 0.05458427220582962,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 13863
+    },
+    {
+      "epoch": 1.4230295566502464,
+      "grad_norm": 0.041047900915145874,
+      "learning_rate": 0.01,
+      "loss": 2.0793,
+      "step": 13866
+    },
+    {
+      "epoch": 1.4233374384236452,
+      "grad_norm": 0.04485390707850456,
+      "learning_rate": 0.01,
+      "loss": 2.0723,
+      "step": 13869
+    },
+    {
+      "epoch": 1.4236453201970443,
+      "grad_norm": 0.13340137898921967,
+      "learning_rate": 0.01,
+      "loss": 2.0555,
+      "step": 13872
+    },
+    {
+      "epoch": 1.4239532019704433,
+      "grad_norm": 0.05519254505634308,
+      "learning_rate": 0.01,
+      "loss": 2.0931,
+      "step": 13875
+    },
+    {
+      "epoch": 1.4242610837438423,
+      "grad_norm": 0.07989728450775146,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 13878
+    },
+    {
+      "epoch": 1.4245689655172413,
+      "grad_norm": 0.09447802603244781,
+      "learning_rate": 0.01,
+      "loss": 2.061,
+      "step": 13881
+    },
+    {
+      "epoch": 1.4248768472906403,
+      "grad_norm": 0.05240226909518242,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 13884
+    },
+    {
+      "epoch": 1.4251847290640394,
+      "grad_norm": 0.07171013206243515,
+      "learning_rate": 0.01,
+      "loss": 2.0829,
+      "step": 13887
+    },
+    {
+      "epoch": 1.4254926108374384,
+      "grad_norm": 0.05098895728588104,
+      "learning_rate": 0.01,
+      "loss": 2.0716,
+      "step": 13890
+    },
+    {
+      "epoch": 1.4258004926108374,
+      "grad_norm": 0.08569507300853729,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 13893
+    },
+    {
+      "epoch": 1.4261083743842364,
+      "grad_norm": 0.09055166691541672,
+      "learning_rate": 0.01,
+      "loss": 2.0843,
+      "step": 13896
+    },
+    {
+      "epoch": 1.4264162561576355,
+      "grad_norm": 0.03242780640721321,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 13899
+    },
+    {
+      "epoch": 1.4267241379310345,
+      "grad_norm": 0.04612202197313309,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 13902
+    },
+    {
+      "epoch": 1.4270320197044335,
+      "grad_norm": 0.05800663307309151,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 13905
+    },
+    {
+      "epoch": 1.4273399014778325,
+      "grad_norm": 0.04150572046637535,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 13908
+    },
+    {
+      "epoch": 1.4276477832512315,
+      "grad_norm": 0.08542584627866745,
+      "learning_rate": 0.01,
+      "loss": 2.0406,
+      "step": 13911
+    },
+    {
+      "epoch": 1.4279556650246306,
+      "grad_norm": 0.11966803669929504,
+      "learning_rate": 0.01,
+      "loss": 2.0829,
+      "step": 13914
+    },
+    {
+      "epoch": 1.4282635467980296,
+      "grad_norm": 0.12066449970006943,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 13917
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.04103751480579376,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 13920
+    },
+    {
+      "epoch": 1.4288793103448276,
+      "grad_norm": 0.05432034656405449,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 13923
+    },
+    {
+      "epoch": 1.4291871921182266,
+      "grad_norm": 0.03935731574892998,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 13926
+    },
+    {
+      "epoch": 1.4294950738916257,
+      "grad_norm": 0.1475706547498703,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 13929
+    },
+    {
+      "epoch": 1.4298029556650247,
+      "grad_norm": 0.06562622636556625,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 13932
+    },
+    {
+      "epoch": 1.4301108374384237,
+      "grad_norm": 0.051726315170526505,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 13935
+    },
+    {
+      "epoch": 1.4304187192118227,
+      "grad_norm": 0.0998329371213913,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 13938
+    },
+    {
+      "epoch": 1.4307266009852218,
+      "grad_norm": 0.04965333640575409,
+      "learning_rate": 0.01,
+      "loss": 2.0568,
+      "step": 13941
+    },
+    {
+      "epoch": 1.4310344827586206,
+      "grad_norm": 0.04430006071925163,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 13944
+    },
+    {
+      "epoch": 1.4313423645320196,
+      "grad_norm": 0.05260150134563446,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 13947
+    },
+    {
+      "epoch": 1.4316502463054186,
+      "grad_norm": 0.04135138541460037,
+      "learning_rate": 0.01,
+      "loss": 2.055,
+      "step": 13950
+    },
+    {
+      "epoch": 1.4319581280788176,
+      "grad_norm": 0.08347123116254807,
+      "learning_rate": 0.01,
+      "loss": 2.0826,
+      "step": 13953
+    },
+    {
+      "epoch": 1.4322660098522166,
+      "grad_norm": 0.12328385561704636,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 13956
+    },
+    {
+      "epoch": 1.4325738916256157,
+      "grad_norm": 0.05809056758880615,
+      "learning_rate": 0.01,
+      "loss": 2.0798,
+      "step": 13959
+    },
+    {
+      "epoch": 1.4328817733990147,
+      "grad_norm": 0.038590408861637115,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 13962
+    },
+    {
+      "epoch": 1.4331896551724137,
+      "grad_norm": 0.11158851534128189,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 13965
+    },
+    {
+      "epoch": 1.4334975369458127,
+      "grad_norm": 0.0880589634180069,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 13968
+    },
+    {
+      "epoch": 1.4338054187192117,
+      "grad_norm": 0.059966232627630234,
+      "learning_rate": 0.01,
+      "loss": 2.0659,
+      "step": 13971
+    },
+    {
+      "epoch": 1.4341133004926108,
+      "grad_norm": 0.04661833122372627,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 13974
+    },
+    {
+      "epoch": 1.4344211822660098,
+      "grad_norm": 0.04508896544575691,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 13977
+    },
+    {
+      "epoch": 1.4347290640394088,
+      "grad_norm": 0.051987554877996445,
+      "learning_rate": 0.01,
+      "loss": 2.0449,
+      "step": 13980
+    },
+    {
+      "epoch": 1.4350369458128078,
+      "grad_norm": 0.04814029112458229,
+      "learning_rate": 0.01,
+      "loss": 2.0587,
+      "step": 13983
+    },
+    {
+      "epoch": 1.4353448275862069,
+      "grad_norm": 0.09631717205047607,
+      "learning_rate": 0.01,
+      "loss": 2.0747,
+      "step": 13986
+    },
+    {
+      "epoch": 1.4356527093596059,
+      "grad_norm": 0.06581971794366837,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 13989
+    },
+    {
+      "epoch": 1.435960591133005,
+      "grad_norm": 0.06483247131109238,
+      "learning_rate": 0.01,
+      "loss": 2.0627,
+      "step": 13992
+    },
+    {
+      "epoch": 1.436268472906404,
+      "grad_norm": 0.1000155657529831,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 13995
+    },
+    {
+      "epoch": 1.436576354679803,
+      "grad_norm": 0.07297492027282715,
+      "learning_rate": 0.01,
+      "loss": 2.0686,
+      "step": 13998
+    },
+    {
+      "epoch": 1.436884236453202,
+      "grad_norm": 0.054907578974962234,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 14001
+    },
+    {
+      "epoch": 1.437192118226601,
+      "grad_norm": 0.051127828657627106,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 14004
+    },
+    {
+      "epoch": 1.4375,
+      "grad_norm": 0.04157300665974617,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 14007
+    },
+    {
+      "epoch": 1.437807881773399,
+      "grad_norm": 0.07996746897697449,
+      "learning_rate": 0.01,
+      "loss": 2.0625,
+      "step": 14010
+    },
+    {
+      "epoch": 1.438115763546798,
+      "grad_norm": 0.0764036774635315,
+      "learning_rate": 0.01,
+      "loss": 2.076,
+      "step": 14013
+    },
+    {
+      "epoch": 1.438423645320197,
+      "grad_norm": 0.10736891627311707,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 14016
+    },
+    {
+      "epoch": 1.438731527093596,
+      "grad_norm": 0.0598980113863945,
+      "learning_rate": 0.01,
+      "loss": 2.0774,
+      "step": 14019
+    },
+    {
+      "epoch": 1.439039408866995,
+      "grad_norm": 0.10858605802059174,
+      "learning_rate": 0.01,
+      "loss": 2.0708,
+      "step": 14022
+    },
+    {
+      "epoch": 1.4393472906403941,
+      "grad_norm": 0.03999786823987961,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 14025
+    },
+    {
+      "epoch": 1.4396551724137931,
+      "grad_norm": 0.053138673305511475,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 14028
+    },
+    {
+      "epoch": 1.4399630541871922,
+      "grad_norm": 0.06477091461420059,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 14031
+    },
+    {
+      "epoch": 1.4402709359605912,
+      "grad_norm": 0.035987384617328644,
+      "learning_rate": 0.01,
+      "loss": 2.0897,
+      "step": 14034
+    },
+    {
+      "epoch": 1.4405788177339902,
+      "grad_norm": 0.06938667595386505,
+      "learning_rate": 0.01,
+      "loss": 2.0895,
+      "step": 14037
+    },
+    {
+      "epoch": 1.4408866995073892,
+      "grad_norm": 0.041746124625205994,
+      "learning_rate": 0.01,
+      "loss": 2.0705,
+      "step": 14040
+    },
+    {
+      "epoch": 1.4411945812807883,
+      "grad_norm": 0.04503123462200165,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 14043
+    },
+    {
+      "epoch": 1.4415024630541873,
+      "grad_norm": 0.04153100401163101,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 14046
+    },
+    {
+      "epoch": 1.4418103448275863,
+      "grad_norm": 0.04992615804076195,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 14049
+    },
+    {
+      "epoch": 1.4421182266009853,
+      "grad_norm": 0.0718725174665451,
+      "learning_rate": 0.01,
+      "loss": 2.049,
+      "step": 14052
+    },
+    {
+      "epoch": 1.4424261083743843,
+      "grad_norm": 0.13080431520938873,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 14055
+    },
+    {
+      "epoch": 1.4427339901477834,
+      "grad_norm": 0.04203762486577034,
+      "learning_rate": 0.01,
+      "loss": 2.0713,
+      "step": 14058
+    },
+    {
+      "epoch": 1.4430418719211824,
+      "grad_norm": 0.04111120104789734,
+      "learning_rate": 0.01,
+      "loss": 2.0845,
+      "step": 14061
+    },
+    {
+      "epoch": 1.4433497536945814,
+      "grad_norm": 0.044398125261068344,
+      "learning_rate": 0.01,
+      "loss": 2.0679,
+      "step": 14064
+    },
+    {
+      "epoch": 1.4436576354679804,
+      "grad_norm": 0.031682152301073074,
+      "learning_rate": 0.01,
+      "loss": 2.0855,
+      "step": 14067
+    },
+    {
+      "epoch": 1.4439655172413794,
+      "grad_norm": 0.07848865538835526,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 14070
+    },
+    {
+      "epoch": 1.4442733990147782,
+      "grad_norm": 0.08814079314470291,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 14073
+    },
+    {
+      "epoch": 1.4445812807881773,
+      "grad_norm": 0.05344429612159729,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 14076
+    },
+    {
+      "epoch": 1.4448891625615763,
+      "grad_norm": 0.05509471520781517,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 14079
+    },
+    {
+      "epoch": 1.4451970443349753,
+      "grad_norm": 0.08177798241376877,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 14082
+    },
+    {
+      "epoch": 1.4455049261083743,
+      "grad_norm": 0.07953787595033646,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 14085
+    },
+    {
+      "epoch": 1.4458128078817734,
+      "grad_norm": 0.06984551250934601,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 14088
+    },
+    {
+      "epoch": 1.4461206896551724,
+      "grad_norm": 0.07923319190740585,
+      "learning_rate": 0.01,
+      "loss": 2.06,
+      "step": 14091
+    },
+    {
+      "epoch": 1.4464285714285714,
+      "grad_norm": 0.04370349645614624,
+      "learning_rate": 0.01,
+      "loss": 2.0839,
+      "step": 14094
+    },
+    {
+      "epoch": 1.4467364532019704,
+      "grad_norm": 0.045787643641233444,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 14097
+    },
+    {
+      "epoch": 1.4470443349753694,
+      "grad_norm": 0.04126288741827011,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 14100
+    },
+    {
+      "epoch": 1.4473522167487685,
+      "grad_norm": 0.039805784821510315,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 14103
+    },
+    {
+      "epoch": 1.4476600985221675,
+      "grad_norm": 0.038430992513895035,
+      "learning_rate": 0.01,
+      "loss": 2.0697,
+      "step": 14106
+    },
+    {
+      "epoch": 1.4479679802955665,
+      "grad_norm": 0.07664498686790466,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 14109
+    },
+    {
+      "epoch": 1.4482758620689655,
+      "grad_norm": 0.04592788219451904,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 14112
+    },
+    {
+      "epoch": 1.4485837438423645,
+      "grad_norm": 0.06161922961473465,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 14115
+    },
+    {
+      "epoch": 1.4488916256157636,
+      "grad_norm": 0.07906373590230942,
+      "learning_rate": 0.01,
+      "loss": 2.0758,
+      "step": 14118
+    },
+    {
+      "epoch": 1.4491995073891626,
+      "grad_norm": 0.09529503434896469,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 14121
+    },
+    {
+      "epoch": 1.4495073891625616,
+      "grad_norm": 0.05416659638285637,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 14124
+    },
+    {
+      "epoch": 1.4498152709359606,
+      "grad_norm": 0.07085006684064865,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 14127
+    },
+    {
+      "epoch": 1.4501231527093597,
+      "grad_norm": 0.07722880691289902,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 14130
+    },
+    {
+      "epoch": 1.4504310344827587,
+      "grad_norm": 0.06599342823028564,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 14133
+    },
+    {
+      "epoch": 1.4507389162561577,
+      "grad_norm": 0.11217498779296875,
+      "learning_rate": 0.01,
+      "loss": 2.068,
+      "step": 14136
+    },
+    {
+      "epoch": 1.4510467980295567,
+      "grad_norm": 0.13082845509052277,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 14139
+    },
+    {
+      "epoch": 1.4513546798029557,
+      "grad_norm": 0.06812801957130432,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 14142
+    },
+    {
+      "epoch": 1.4516625615763548,
+      "grad_norm": 0.043554674834012985,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 14145
+    },
+    {
+      "epoch": 1.4519704433497536,
+      "grad_norm": 0.04550860822200775,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 14148
+    },
+    {
+      "epoch": 1.4522783251231526,
+      "grad_norm": 0.04838492348790169,
+      "learning_rate": 0.01,
+      "loss": 2.0642,
+      "step": 14151
+    },
+    {
+      "epoch": 1.4525862068965516,
+      "grad_norm": 0.08433537930250168,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 14154
+    },
+    {
+      "epoch": 1.4528940886699506,
+      "grad_norm": 0.06986009329557419,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 14157
+    },
+    {
+      "epoch": 1.4532019704433496,
+      "grad_norm": 0.06860263645648956,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 14160
+    },
+    {
+      "epoch": 1.4535098522167487,
+      "grad_norm": 0.06218327581882477,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 14163
+    },
+    {
+      "epoch": 1.4538177339901477,
+      "grad_norm": 0.10177832096815109,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 14166
+    },
+    {
+      "epoch": 1.4541256157635467,
+      "grad_norm": 0.047695957124233246,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 14169
+    },
+    {
+      "epoch": 1.4544334975369457,
+      "grad_norm": 0.09761510044336319,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 14172
+    },
+    {
+      "epoch": 1.4547413793103448,
+      "grad_norm": 0.050296783447265625,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 14175
+    },
+    {
+      "epoch": 1.4550492610837438,
+      "grad_norm": 0.13070064783096313,
+      "learning_rate": 0.01,
+      "loss": 2.0579,
+      "step": 14178
+    },
+    {
+      "epoch": 1.4553571428571428,
+      "grad_norm": 0.1080620214343071,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 14181
+    },
+    {
+      "epoch": 1.4556650246305418,
+      "grad_norm": 0.06132792308926582,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 14184
+    },
+    {
+      "epoch": 1.4559729064039408,
+      "grad_norm": 0.07258635014295578,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 14187
+    },
+    {
+      "epoch": 1.4562807881773399,
+      "grad_norm": 0.05423443764448166,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 14190
+    },
+    {
+      "epoch": 1.4565886699507389,
+      "grad_norm": 0.06038088724017143,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 14193
+    },
+    {
+      "epoch": 1.456896551724138,
+      "grad_norm": 0.043958742171525955,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 14196
+    },
+    {
+      "epoch": 1.457204433497537,
+      "grad_norm": 0.03985238075256348,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 14199
+    },
+    {
+      "epoch": 1.457512315270936,
+      "grad_norm": 0.05626663193106651,
+      "learning_rate": 0.01,
+      "loss": 2.0518,
+      "step": 14202
+    },
+    {
+      "epoch": 1.457820197044335,
+      "grad_norm": 0.06365952640771866,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 14205
+    },
+    {
+      "epoch": 1.458128078817734,
+      "grad_norm": 0.0698857232928276,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 14208
+    },
+    {
+      "epoch": 1.458435960591133,
+      "grad_norm": 0.08149702101945877,
+      "learning_rate": 0.01,
+      "loss": 2.0659,
+      "step": 14211
+    },
+    {
+      "epoch": 1.458743842364532,
+      "grad_norm": 0.055818330496549606,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 14214
+    },
+    {
+      "epoch": 1.459051724137931,
+      "grad_norm": 0.046251330524683,
+      "learning_rate": 0.01,
+      "loss": 2.0598,
+      "step": 14217
+    },
+    {
+      "epoch": 1.45935960591133,
+      "grad_norm": 0.10986322909593582,
+      "learning_rate": 0.01,
+      "loss": 2.0768,
+      "step": 14220
+    },
+    {
+      "epoch": 1.459667487684729,
+      "grad_norm": 0.06735626608133316,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 14223
+    },
+    {
+      "epoch": 1.4599753694581281,
+      "grad_norm": 0.1051633432507515,
+      "learning_rate": 0.01,
+      "loss": 2.0807,
+      "step": 14226
+    },
+    {
+      "epoch": 1.4602832512315271,
+      "grad_norm": 0.0956743136048317,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 14229
+    },
+    {
+      "epoch": 1.4605911330049262,
+      "grad_norm": 0.04349840059876442,
+      "learning_rate": 0.01,
+      "loss": 2.0723,
+      "step": 14232
+    },
+    {
+      "epoch": 1.4608990147783252,
+      "grad_norm": 0.0388668030500412,
+      "learning_rate": 0.01,
+      "loss": 2.0612,
+      "step": 14235
+    },
+    {
+      "epoch": 1.4612068965517242,
+      "grad_norm": 0.04311763867735863,
+      "learning_rate": 0.01,
+      "loss": 2.0553,
+      "step": 14238
+    },
+    {
+      "epoch": 1.4615147783251232,
+      "grad_norm": 0.09116464853286743,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 14241
+    },
+    {
+      "epoch": 1.4618226600985222,
+      "grad_norm": 0.08582088351249695,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 14244
+    },
+    {
+      "epoch": 1.4621305418719213,
+      "grad_norm": 0.0435602031648159,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 14247
+    },
+    {
+      "epoch": 1.4624384236453203,
+      "grad_norm": 0.067762091755867,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 14250
+    },
+    {
+      "epoch": 1.4627463054187193,
+      "grad_norm": 0.03980677202343941,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 14253
+    },
+    {
+      "epoch": 1.4630541871921183,
+      "grad_norm": 0.08521614968776703,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 14256
+    },
+    {
+      "epoch": 1.4633620689655173,
+      "grad_norm": 0.06770948320627213,
+      "learning_rate": 0.01,
+      "loss": 2.0535,
+      "step": 14259
+    },
+    {
+      "epoch": 1.4636699507389164,
+      "grad_norm": 0.053458839654922485,
+      "learning_rate": 0.01,
+      "loss": 2.0594,
+      "step": 14262
+    },
+    {
+      "epoch": 1.4639778325123154,
+      "grad_norm": 0.06733859330415726,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 14265
+    },
+    {
+      "epoch": 1.4642857142857144,
+      "grad_norm": 0.08033892512321472,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 14268
+    },
+    {
+      "epoch": 1.4645935960591134,
+      "grad_norm": 0.0832366794347763,
+      "learning_rate": 0.01,
+      "loss": 2.0867,
+      "step": 14271
+    },
+    {
+      "epoch": 1.4649014778325122,
+      "grad_norm": 0.055291153490543365,
+      "learning_rate": 0.01,
+      "loss": 2.0618,
+      "step": 14274
+    },
+    {
+      "epoch": 1.4652093596059113,
+      "grad_norm": 0.07180801033973694,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 14277
+    },
+    {
+      "epoch": 1.4655172413793103,
+      "grad_norm": 0.048950713127851486,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 14280
+    },
+    {
+      "epoch": 1.4658251231527093,
+      "grad_norm": 0.05428025498986244,
+      "learning_rate": 0.01,
+      "loss": 2.052,
+      "step": 14283
+    },
+    {
+      "epoch": 1.4661330049261083,
+      "grad_norm": 0.06631309539079666,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 14286
+    },
+    {
+      "epoch": 1.4664408866995073,
+      "grad_norm": 0.06743253022432327,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 14289
+    },
+    {
+      "epoch": 1.4667487684729064,
+      "grad_norm": 0.10901882499456406,
+      "learning_rate": 0.01,
+      "loss": 2.076,
+      "step": 14292
+    },
+    {
+      "epoch": 1.4670566502463054,
+      "grad_norm": 0.08234187960624695,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 14295
+    },
+    {
+      "epoch": 1.4673645320197044,
+      "grad_norm": 0.07249965518712997,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 14298
+    },
+    {
+      "epoch": 1.4676724137931034,
+      "grad_norm": 0.0705137550830841,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 14301
+    },
+    {
+      "epoch": 1.4679802955665024,
+      "grad_norm": 0.10730472952127457,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 14304
+    },
+    {
+      "epoch": 1.4682881773399015,
+      "grad_norm": 0.048364557325839996,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 14307
+    },
+    {
+      "epoch": 1.4685960591133005,
+      "grad_norm": 0.03604978322982788,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 14310
+    },
+    {
+      "epoch": 1.4689039408866995,
+      "grad_norm": 0.09732489287853241,
+      "learning_rate": 0.01,
+      "loss": 2.0627,
+      "step": 14313
+    },
+    {
+      "epoch": 1.4692118226600985,
+      "grad_norm": 0.06590714305639267,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 14316
+    },
+    {
+      "epoch": 1.4695197044334976,
+      "grad_norm": 0.075086310505867,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 14319
+    },
+    {
+      "epoch": 1.4698275862068966,
+      "grad_norm": 0.10288450121879578,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 14322
+    },
+    {
+      "epoch": 1.4701354679802956,
+      "grad_norm": 0.0535271093249321,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 14325
+    },
+    {
+      "epoch": 1.4704433497536946,
+      "grad_norm": 0.04609391465783119,
+      "learning_rate": 0.01,
+      "loss": 2.049,
+      "step": 14328
+    },
+    {
+      "epoch": 1.4707512315270936,
+      "grad_norm": 0.044252909719944,
+      "learning_rate": 0.01,
+      "loss": 2.0586,
+      "step": 14331
+    },
+    {
+      "epoch": 1.4710591133004927,
+      "grad_norm": 0.07837995141744614,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 14334
+    },
+    {
+      "epoch": 1.4713669950738917,
+      "grad_norm": 0.06548511236906052,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 14337
+    },
+    {
+      "epoch": 1.4716748768472907,
+      "grad_norm": 0.09237763285636902,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 14340
+    },
+    {
+      "epoch": 1.4719827586206897,
+      "grad_norm": 0.04163452237844467,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 14343
+    },
+    {
+      "epoch": 1.4722906403940887,
+      "grad_norm": 0.13814504444599152,
+      "learning_rate": 0.01,
+      "loss": 2.0667,
+      "step": 14346
+    },
+    {
+      "epoch": 1.4725985221674878,
+      "grad_norm": 0.054490260779857635,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 14349
+    },
+    {
+      "epoch": 1.4729064039408866,
+      "grad_norm": 0.07470995932817459,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 14352
+    },
+    {
+      "epoch": 1.4732142857142856,
+      "grad_norm": 0.12089511752128601,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 14355
+    },
+    {
+      "epoch": 1.4735221674876846,
+      "grad_norm": 0.05507595092058182,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 14358
+    },
+    {
+      "epoch": 1.4738300492610836,
+      "grad_norm": 0.05130109563469887,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 14361
+    },
+    {
+      "epoch": 1.4741379310344827,
+      "grad_norm": 0.04399503022432327,
+      "learning_rate": 0.01,
+      "loss": 2.0669,
+      "step": 14364
+    },
+    {
+      "epoch": 1.4744458128078817,
+      "grad_norm": 0.06161755695939064,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 14367
+    },
+    {
+      "epoch": 1.4747536945812807,
+      "grad_norm": 0.045603156089782715,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 14370
+    },
+    {
+      "epoch": 1.4750615763546797,
+      "grad_norm": 0.06444186717271805,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 14373
+    },
+    {
+      "epoch": 1.4753694581280787,
+      "grad_norm": 0.07450753450393677,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 14376
+    },
+    {
+      "epoch": 1.4756773399014778,
+      "grad_norm": 0.10367168486118317,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 14379
+    },
+    {
+      "epoch": 1.4759852216748768,
+      "grad_norm": 0.08999089151620865,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 14382
+    },
+    {
+      "epoch": 1.4762931034482758,
+      "grad_norm": 0.09803617745637894,
+      "learning_rate": 0.01,
+      "loss": 2.0972,
+      "step": 14385
+    },
+    {
+      "epoch": 1.4766009852216748,
+      "grad_norm": 0.042447153478860855,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 14388
+    },
+    {
+      "epoch": 1.4769088669950738,
+      "grad_norm": 0.04479740187525749,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 14391
+    },
+    {
+      "epoch": 1.4772167487684729,
+      "grad_norm": 0.04113270714879036,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 14394
+    },
+    {
+      "epoch": 1.4775246305418719,
+      "grad_norm": 0.1087644025683403,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 14397
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.05737099424004555,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 14400
+    },
+    {
+      "epoch": 1.47814039408867,
+      "grad_norm": 0.11025606095790863,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 14403
+    },
+    {
+      "epoch": 1.478448275862069,
+      "grad_norm": 0.06662195175886154,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 14406
+    },
+    {
+      "epoch": 1.478756157635468,
+      "grad_norm": 0.05261904001235962,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 14409
+    },
+    {
+      "epoch": 1.479064039408867,
+      "grad_norm": 0.048272691667079926,
+      "learning_rate": 0.01,
+      "loss": 2.0858,
+      "step": 14412
+    },
+    {
+      "epoch": 1.479371921182266,
+      "grad_norm": 0.048300545662641525,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 14415
+    },
+    {
+      "epoch": 1.479679802955665,
+      "grad_norm": 0.08697368204593658,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 14418
+    },
+    {
+      "epoch": 1.479987684729064,
+      "grad_norm": 0.07993713021278381,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 14421
+    },
+    {
+      "epoch": 1.480295566502463,
+      "grad_norm": 0.10037390887737274,
+      "learning_rate": 0.01,
+      "loss": 2.0584,
+      "step": 14424
+    },
+    {
+      "epoch": 1.480603448275862,
+      "grad_norm": 0.06048484891653061,
+      "learning_rate": 0.01,
+      "loss": 2.0589,
+      "step": 14427
+    },
+    {
+      "epoch": 1.4809113300492611,
+      "grad_norm": 0.08982612937688828,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 14430
+    },
+    {
+      "epoch": 1.4812192118226601,
+      "grad_norm": 0.06678975373506546,
+      "learning_rate": 0.01,
+      "loss": 2.0544,
+      "step": 14433
+    },
+    {
+      "epoch": 1.4815270935960592,
+      "grad_norm": 0.07890944927930832,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 14436
+    },
+    {
+      "epoch": 1.4818349753694582,
+      "grad_norm": 0.05838685482740402,
+      "learning_rate": 0.01,
+      "loss": 2.0481,
+      "step": 14439
+    },
+    {
+      "epoch": 1.4821428571428572,
+      "grad_norm": 0.06483394652605057,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 14442
+    },
+    {
+      "epoch": 1.4824507389162562,
+      "grad_norm": 0.07320713996887207,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 14445
+    },
+    {
+      "epoch": 1.4827586206896552,
+      "grad_norm": 0.07484092563390732,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 14448
+    },
+    {
+      "epoch": 1.4830665024630543,
+      "grad_norm": 0.07702804356813431,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 14451
+    },
+    {
+      "epoch": 1.4833743842364533,
+      "grad_norm": 0.05470692366361618,
+      "learning_rate": 0.01,
+      "loss": 2.0767,
+      "step": 14454
+    },
+    {
+      "epoch": 1.4836822660098523,
+      "grad_norm": 0.055773910135030746,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 14457
+    },
+    {
+      "epoch": 1.4839901477832513,
+      "grad_norm": 0.03712743893265724,
+      "learning_rate": 0.01,
+      "loss": 2.0723,
+      "step": 14460
+    },
+    {
+      "epoch": 1.4842980295566504,
+      "grad_norm": 0.035963475704193115,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 14463
+    },
+    {
+      "epoch": 1.4846059113300494,
+      "grad_norm": 0.08578921854496002,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 14466
+    },
+    {
+      "epoch": 1.4849137931034484,
+      "grad_norm": 0.08239159733057022,
+      "learning_rate": 0.01,
+      "loss": 2.0654,
+      "step": 14469
+    },
+    {
+      "epoch": 1.4852216748768474,
+      "grad_norm": 0.05097891017794609,
+      "learning_rate": 0.01,
+      "loss": 2.0837,
+      "step": 14472
+    },
+    {
+      "epoch": 1.4855295566502464,
+      "grad_norm": 0.051847904920578,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 14475
+    },
+    {
+      "epoch": 1.4858374384236452,
+      "grad_norm": 0.04754810780286789,
+      "learning_rate": 0.01,
+      "loss": 2.0658,
+      "step": 14478
+    },
+    {
+      "epoch": 1.4861453201970443,
+      "grad_norm": 0.046647075563669205,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 14481
+    },
+    {
+      "epoch": 1.4864532019704433,
+      "grad_norm": 0.06013277545571327,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 14484
+    },
+    {
+      "epoch": 1.4867610837438423,
+      "grad_norm": 0.13433513045310974,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 14487
+    },
+    {
+      "epoch": 1.4870689655172413,
+      "grad_norm": 0.046518564224243164,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 14490
+    },
+    {
+      "epoch": 1.4873768472906403,
+      "grad_norm": 0.09483514726161957,
+      "learning_rate": 0.01,
+      "loss": 2.0839,
+      "step": 14493
+    },
+    {
+      "epoch": 1.4876847290640394,
+      "grad_norm": 0.07147302478551865,
+      "learning_rate": 0.01,
+      "loss": 2.0741,
+      "step": 14496
+    },
+    {
+      "epoch": 1.4879926108374384,
+      "grad_norm": 0.12423846870660782,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 14499
+    },
+    {
+      "epoch": 1.4883004926108374,
+      "grad_norm": 0.07726770639419556,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 14502
+    },
+    {
+      "epoch": 1.4886083743842364,
+      "grad_norm": 0.059802982956171036,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 14505
+    },
+    {
+      "epoch": 1.4889162561576355,
+      "grad_norm": 0.050745993852615356,
+      "learning_rate": 0.01,
+      "loss": 2.049,
+      "step": 14508
+    },
+    {
+      "epoch": 1.4892241379310345,
+      "grad_norm": 0.052064161747694016,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 14511
+    },
+    {
+      "epoch": 1.4895320197044335,
+      "grad_norm": 0.06646674871444702,
+      "learning_rate": 0.01,
+      "loss": 2.0689,
+      "step": 14514
+    },
+    {
+      "epoch": 1.4898399014778325,
+      "grad_norm": 0.043484605848789215,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 14517
+    },
+    {
+      "epoch": 1.4901477832512315,
+      "grad_norm": 0.14452145993709564,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 14520
+    },
+    {
+      "epoch": 1.4904556650246306,
+      "grad_norm": 0.08289093524217606,
+      "learning_rate": 0.01,
+      "loss": 2.0654,
+      "step": 14523
+    },
+    {
+      "epoch": 1.4907635467980296,
+      "grad_norm": 0.05047908052802086,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 14526
+    },
+    {
+      "epoch": 1.4910714285714286,
+      "grad_norm": 0.04830252006649971,
+      "learning_rate": 0.01,
+      "loss": 2.0529,
+      "step": 14529
+    },
+    {
+      "epoch": 1.4913793103448276,
+      "grad_norm": 0.0430610254406929,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 14532
+    },
+    {
+      "epoch": 1.4916871921182266,
+      "grad_norm": 0.04282008111476898,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 14535
+    },
+    {
+      "epoch": 1.4919950738916257,
+      "grad_norm": 0.037373676896095276,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 14538
+    },
+    {
+      "epoch": 1.4923029556650247,
+      "grad_norm": 0.04186755418777466,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 14541
+    },
+    {
+      "epoch": 1.4926108374384237,
+      "grad_norm": 0.05514196678996086,
+      "learning_rate": 0.01,
+      "loss": 2.0647,
+      "step": 14544
+    },
+    {
+      "epoch": 1.4929187192118227,
+      "grad_norm": 0.07391703873872757,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 14547
+    },
+    {
+      "epoch": 1.4932266009852218,
+      "grad_norm": 0.1295444518327713,
+      "learning_rate": 0.01,
+      "loss": 2.0571,
+      "step": 14550
+    },
+    {
+      "epoch": 1.4935344827586206,
+      "grad_norm": 0.06389490514993668,
+      "learning_rate": 0.01,
+      "loss": 2.0756,
+      "step": 14553
+    },
+    {
+      "epoch": 1.4938423645320196,
+      "grad_norm": 0.09335155785083771,
+      "learning_rate": 0.01,
+      "loss": 2.0904,
+      "step": 14556
+    },
+    {
+      "epoch": 1.4941502463054186,
+      "grad_norm": 0.059700366109609604,
+      "learning_rate": 0.01,
+      "loss": 2.0598,
+      "step": 14559
+    },
+    {
+      "epoch": 1.4944581280788176,
+      "grad_norm": 0.07785683870315552,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 14562
+    },
+    {
+      "epoch": 1.4947660098522166,
+      "grad_norm": 0.11935362964868546,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 14565
+    },
+    {
+      "epoch": 1.4950738916256157,
+      "grad_norm": 0.06188122183084488,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 14568
+    },
+    {
+      "epoch": 1.4953817733990147,
+      "grad_norm": 0.05302932485938072,
+      "learning_rate": 0.01,
+      "loss": 2.0704,
+      "step": 14571
+    },
+    {
+      "epoch": 1.4956896551724137,
+      "grad_norm": 0.03871694207191467,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 14574
+    },
+    {
+      "epoch": 1.4959975369458127,
+      "grad_norm": 0.03942064568400383,
+      "learning_rate": 0.01,
+      "loss": 2.0725,
+      "step": 14577
+    },
+    {
+      "epoch": 1.4963054187192117,
+      "grad_norm": 0.05354088917374611,
+      "learning_rate": 0.01,
+      "loss": 2.054,
+      "step": 14580
+    },
+    {
+      "epoch": 1.4966133004926108,
+      "grad_norm": 0.07863521575927734,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 14583
+    },
+    {
+      "epoch": 1.4969211822660098,
+      "grad_norm": 0.0440685860812664,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 14586
+    },
+    {
+      "epoch": 1.4972290640394088,
+      "grad_norm": 0.0724552571773529,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 14589
+    },
+    {
+      "epoch": 1.4975369458128078,
+      "grad_norm": 0.06099352613091469,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 14592
+    },
+    {
+      "epoch": 1.4978448275862069,
+      "grad_norm": 0.05534674599766731,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 14595
+    },
+    {
+      "epoch": 1.4981527093596059,
+      "grad_norm": 0.07876823097467422,
+      "learning_rate": 0.01,
+      "loss": 2.0686,
+      "step": 14598
+    },
+    {
+      "epoch": 1.498460591133005,
+      "grad_norm": 0.07860377430915833,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 14601
+    },
+    {
+      "epoch": 1.498768472906404,
+      "grad_norm": 0.054005399346351624,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 14604
+    },
+    {
+      "epoch": 1.499076354679803,
+      "grad_norm": 0.10550951957702637,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 14607
+    },
+    {
+      "epoch": 1.499384236453202,
+      "grad_norm": 0.056426841765642166,
+      "learning_rate": 0.01,
+      "loss": 2.0589,
+      "step": 14610
+    },
+    {
+      "epoch": 1.499692118226601,
+      "grad_norm": 0.09640904515981674,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 14613
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.0822538211941719,
+      "learning_rate": 0.01,
+      "loss": 2.0684,
+      "step": 14616
+    },
+    {
+      "epoch": 1.500307881773399,
+      "grad_norm": 0.05105495825409889,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 14619
+    },
+    {
+      "epoch": 1.500615763546798,
+      "grad_norm": 0.07851336896419525,
+      "learning_rate": 0.01,
+      "loss": 2.0635,
+      "step": 14622
+    },
+    {
+      "epoch": 1.500923645320197,
+      "grad_norm": 0.051046207547187805,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 14625
+    },
+    {
+      "epoch": 1.501231527093596,
+      "grad_norm": 0.12335740774869919,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 14628
+    },
+    {
+      "epoch": 1.501539408866995,
+      "grad_norm": 0.04044636711478233,
+      "learning_rate": 0.01,
+      "loss": 2.0635,
+      "step": 14631
+    },
+    {
+      "epoch": 1.5018472906403941,
+      "grad_norm": 0.0532059408724308,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 14634
+    },
+    {
+      "epoch": 1.5021551724137931,
+      "grad_norm": 0.0446847639977932,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 14637
+    },
+    {
+      "epoch": 1.5024630541871922,
+      "grad_norm": 0.05464153364300728,
+      "learning_rate": 0.01,
+      "loss": 2.0464,
+      "step": 14640
+    },
+    {
+      "epoch": 1.5027709359605912,
+      "grad_norm": 0.08923088759183884,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 14643
+    },
+    {
+      "epoch": 1.5030788177339902,
+      "grad_norm": 0.06256496161222458,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 14646
+    },
+    {
+      "epoch": 1.5033866995073892,
+      "grad_norm": 0.05338229984045029,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 14649
+    },
+    {
+      "epoch": 1.5036945812807883,
+      "grad_norm": 0.04416535049676895,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 14652
+    },
+    {
+      "epoch": 1.5040024630541873,
+      "grad_norm": 0.07076221704483032,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 14655
+    },
+    {
+      "epoch": 1.5043103448275863,
+      "grad_norm": 0.08566464483737946,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 14658
+    },
+    {
+      "epoch": 1.5046182266009853,
+      "grad_norm": 0.049552109092473984,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 14661
+    },
+    {
+      "epoch": 1.5049261083743843,
+      "grad_norm": 0.11802852898836136,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 14664
+    },
+    {
+      "epoch": 1.5052339901477834,
+      "grad_norm": 0.05280107632279396,
+      "learning_rate": 0.01,
+      "loss": 2.0722,
+      "step": 14667
+    },
+    {
+      "epoch": 1.5055418719211824,
+      "grad_norm": 0.036458853632211685,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 14670
+    },
+    {
+      "epoch": 1.5058497536945814,
+      "grad_norm": 0.0465536043047905,
+      "learning_rate": 0.01,
+      "loss": 2.0765,
+      "step": 14673
+    },
+    {
+      "epoch": 1.5061576354679804,
+      "grad_norm": 0.09052444994449615,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 14676
+    },
+    {
+      "epoch": 1.5064655172413794,
+      "grad_norm": 0.08750707656145096,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 14679
+    },
+    {
+      "epoch": 1.5067733990147785,
+      "grad_norm": 0.07876972109079361,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 14682
+    },
+    {
+      "epoch": 1.5070812807881775,
+      "grad_norm": 0.045561011880636215,
+      "learning_rate": 0.01,
+      "loss": 2.0656,
+      "step": 14685
+    },
+    {
+      "epoch": 1.5073891625615765,
+      "grad_norm": 0.04548237472772598,
+      "learning_rate": 0.01,
+      "loss": 2.0811,
+      "step": 14688
+    },
+    {
+      "epoch": 1.5076970443349755,
+      "grad_norm": 0.04897540062665939,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 14691
+    },
+    {
+      "epoch": 1.5080049261083743,
+      "grad_norm": 0.08820399641990662,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 14694
+    },
+    {
+      "epoch": 1.5083128078817734,
+      "grad_norm": 0.0701432004570961,
+      "learning_rate": 0.01,
+      "loss": 2.0644,
+      "step": 14697
+    },
+    {
+      "epoch": 1.5086206896551724,
+      "grad_norm": 0.10921904444694519,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 14700
+    },
+    {
+      "epoch": 1.5089285714285714,
+      "grad_norm": 0.08308566361665726,
+      "learning_rate": 0.01,
+      "loss": 2.101,
+      "step": 14703
+    },
+    {
+      "epoch": 1.5092364532019704,
+      "grad_norm": 0.12545743584632874,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 14706
+    },
+    {
+      "epoch": 1.5095443349753694,
+      "grad_norm": 0.11245466768741608,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 14709
+    },
+    {
+      "epoch": 1.5098522167487685,
+      "grad_norm": 0.10128718614578247,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 14712
+    },
+    {
+      "epoch": 1.5101600985221675,
+      "grad_norm": 0.07226911187171936,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 14715
+    },
+    {
+      "epoch": 1.5104679802955665,
+      "grad_norm": 0.056605782359838486,
+      "learning_rate": 0.01,
+      "loss": 2.0686,
+      "step": 14718
+    },
+    {
+      "epoch": 1.5107758620689655,
+      "grad_norm": 0.08795683085918427,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 14721
+    },
+    {
+      "epoch": 1.5110837438423645,
+      "grad_norm": 0.07311341166496277,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 14724
+    },
+    {
+      "epoch": 1.5113916256157636,
+      "grad_norm": 0.07164688408374786,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 14727
+    },
+    {
+      "epoch": 1.5116995073891626,
+      "grad_norm": 0.04577312618494034,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 14730
+    },
+    {
+      "epoch": 1.5120073891625616,
+      "grad_norm": 0.043186552822589874,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 14733
+    },
+    {
+      "epoch": 1.5123152709359606,
+      "grad_norm": 0.06101042777299881,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 14736
+    },
+    {
+      "epoch": 1.5126231527093597,
+      "grad_norm": 0.12651608884334564,
+      "learning_rate": 0.01,
+      "loss": 2.0914,
+      "step": 14739
+    },
+    {
+      "epoch": 1.5129310344827587,
+      "grad_norm": 0.039495594799518585,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 14742
+    },
+    {
+      "epoch": 1.5132389162561575,
+      "grad_norm": 0.10854054987430573,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 14745
+    },
+    {
+      "epoch": 1.5135467980295565,
+      "grad_norm": 0.142778679728508,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 14748
+    },
+    {
+      "epoch": 1.5138546798029555,
+      "grad_norm": 0.06473375856876373,
+      "learning_rate": 0.01,
+      "loss": 2.075,
+      "step": 14751
+    },
+    {
+      "epoch": 1.5141625615763545,
+      "grad_norm": 0.050436701625585556,
+      "learning_rate": 0.01,
+      "loss": 2.0823,
+      "step": 14754
+    },
+    {
+      "epoch": 1.5144704433497536,
+      "grad_norm": 0.057088159024715424,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 14757
+    },
+    {
+      "epoch": 1.5147783251231526,
+      "grad_norm": 0.051623161882162094,
+      "learning_rate": 0.01,
+      "loss": 2.0816,
+      "step": 14760
+    },
+    {
+      "epoch": 1.5150862068965516,
+      "grad_norm": 0.0770149901509285,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 14763
+    },
+    {
+      "epoch": 1.5153940886699506,
+      "grad_norm": 0.06827536970376968,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 14766
+    },
+    {
+      "epoch": 1.5157019704433496,
+      "grad_norm": 0.06987358629703522,
+      "learning_rate": 0.01,
+      "loss": 2.0618,
+      "step": 14769
+    },
+    {
+      "epoch": 1.5160098522167487,
+      "grad_norm": 0.05388219282031059,
+      "learning_rate": 0.01,
+      "loss": 2.084,
+      "step": 14772
+    },
+    {
+      "epoch": 1.5163177339901477,
+      "grad_norm": 0.11866139620542526,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 14775
+    },
+    {
+      "epoch": 1.5166256157635467,
+      "grad_norm": 0.12754911184310913,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 14778
+    },
+    {
+      "epoch": 1.5169334975369457,
+      "grad_norm": 0.03591502830386162,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 14781
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.09142038226127625,
+      "learning_rate": 0.01,
+      "loss": 2.05,
+      "step": 14784
+    },
+    {
+      "epoch": 1.5175492610837438,
+      "grad_norm": 0.13140954077243805,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 14787
+    },
+    {
+      "epoch": 1.5178571428571428,
+      "grad_norm": 0.08330459147691727,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 14790
+    },
+    {
+      "epoch": 1.5181650246305418,
+      "grad_norm": 0.0779498815536499,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 14793
+    },
+    {
+      "epoch": 1.5184729064039408,
+      "grad_norm": 0.05396762117743492,
+      "learning_rate": 0.01,
+      "loss": 2.0586,
+      "step": 14796
+    },
+    {
+      "epoch": 1.5187807881773399,
+      "grad_norm": 0.06744614988565445,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 14799
+    },
+    {
+      "epoch": 1.5190886699507389,
+      "grad_norm": 0.04777420684695244,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 14802
+    },
+    {
+      "epoch": 1.519396551724138,
+      "grad_norm": 0.044643301516771317,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 14805
+    },
+    {
+      "epoch": 1.519704433497537,
+      "grad_norm": 0.05263877660036087,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 14808
+    },
+    {
+      "epoch": 1.520012315270936,
+      "grad_norm": 0.07794903963804245,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 14811
+    },
+    {
+      "epoch": 1.520320197044335,
+      "grad_norm": 0.03846001625061035,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 14814
+    },
+    {
+      "epoch": 1.520628078817734,
+      "grad_norm": 0.03806301951408386,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 14817
+    },
+    {
+      "epoch": 1.520935960591133,
+      "grad_norm": 0.08289408683776855,
+      "learning_rate": 0.01,
+      "loss": 2.0712,
+      "step": 14820
+    },
+    {
+      "epoch": 1.521243842364532,
+      "grad_norm": 0.04307285323739052,
+      "learning_rate": 0.01,
+      "loss": 2.0926,
+      "step": 14823
+    },
+    {
+      "epoch": 1.521551724137931,
+      "grad_norm": 0.04523704573512077,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 14826
+    },
+    {
+      "epoch": 1.52185960591133,
+      "grad_norm": 0.0813162624835968,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 14829
+    },
+    {
+      "epoch": 1.522167487684729,
+      "grad_norm": 0.08958449214696884,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 14832
+    },
+    {
+      "epoch": 1.5224753694581281,
+      "grad_norm": 0.1036042720079422,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 14835
+    },
+    {
+      "epoch": 1.5227832512315271,
+      "grad_norm": 0.06528764218091965,
+      "learning_rate": 0.01,
+      "loss": 2.0389,
+      "step": 14838
+    },
+    {
+      "epoch": 1.5230911330049262,
+      "grad_norm": 0.04857415333390236,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 14841
+    },
+    {
+      "epoch": 1.5233990147783252,
+      "grad_norm": 0.11137302964925766,
+      "learning_rate": 0.01,
+      "loss": 2.0706,
+      "step": 14844
+    },
+    {
+      "epoch": 1.5237068965517242,
+      "grad_norm": 0.05258537456393242,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 14847
+    },
+    {
+      "epoch": 1.5240147783251232,
+      "grad_norm": 0.05203690007328987,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 14850
+    },
+    {
+      "epoch": 1.5243226600985222,
+      "grad_norm": 0.036557264626026154,
+      "learning_rate": 0.01,
+      "loss": 2.0827,
+      "step": 14853
+    },
+    {
+      "epoch": 1.5246305418719213,
+      "grad_norm": 0.05553048849105835,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 14856
+    },
+    {
+      "epoch": 1.5249384236453203,
+      "grad_norm": 0.07551626116037369,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 14859
+    },
+    {
+      "epoch": 1.5252463054187193,
+      "grad_norm": 0.09335839748382568,
+      "learning_rate": 0.01,
+      "loss": 2.069,
+      "step": 14862
+    },
+    {
+      "epoch": 1.5255541871921183,
+      "grad_norm": 0.07123745232820511,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 14865
+    },
+    {
+      "epoch": 1.5258620689655173,
+      "grad_norm": 0.06792188435792923,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 14868
+    },
+    {
+      "epoch": 1.5261699507389164,
+      "grad_norm": 0.035666827112436295,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 14871
+    },
+    {
+      "epoch": 1.5264778325123154,
+      "grad_norm": 0.039600104093551636,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 14874
+    },
+    {
+      "epoch": 1.5267857142857144,
+      "grad_norm": 0.03983796760439873,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 14877
+    },
+    {
+      "epoch": 1.5270935960591134,
+      "grad_norm": 0.07013492286205292,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 14880
+    },
+    {
+      "epoch": 1.5274014778325125,
+      "grad_norm": 0.07064792513847351,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 14883
+    },
+    {
+      "epoch": 1.5277093596059115,
+      "grad_norm": 0.10502810031175613,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 14886
+    },
+    {
+      "epoch": 1.5280172413793105,
+      "grad_norm": 0.050288375467061996,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 14889
+    },
+    {
+      "epoch": 1.5283251231527095,
+      "grad_norm": 0.07382049411535263,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 14892
+    },
+    {
+      "epoch": 1.5286330049261085,
+      "grad_norm": 0.08693026751279831,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 14895
+    },
+    {
+      "epoch": 1.5289408866995073,
+      "grad_norm": 0.04283773526549339,
+      "learning_rate": 0.01,
+      "loss": 2.068,
+      "step": 14898
+    },
+    {
+      "epoch": 1.5292487684729064,
+      "grad_norm": 0.1044667437672615,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 14901
+    },
+    {
+      "epoch": 1.5295566502463054,
+      "grad_norm": 0.06316410005092621,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 14904
+    },
+    {
+      "epoch": 1.5298645320197044,
+      "grad_norm": 0.04687780514359474,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 14907
+    },
+    {
+      "epoch": 1.5301724137931034,
+      "grad_norm": 0.04785927012562752,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 14910
+    },
+    {
+      "epoch": 1.5304802955665024,
+      "grad_norm": 0.03788747265934944,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 14913
+    },
+    {
+      "epoch": 1.5307881773399015,
+      "grad_norm": 0.05618858337402344,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 14916
+    },
+    {
+      "epoch": 1.5310960591133005,
+      "grad_norm": 0.10059016942977905,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 14919
+    },
+    {
+      "epoch": 1.5314039408866995,
+      "grad_norm": 0.06718064099550247,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 14922
+    },
+    {
+      "epoch": 1.5317118226600985,
+      "grad_norm": 0.09006398916244507,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 14925
+    },
+    {
+      "epoch": 1.5320197044334976,
+      "grad_norm": 0.036577485501766205,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 14928
+    },
+    {
+      "epoch": 1.5323275862068966,
+      "grad_norm": 0.07979925721883774,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 14931
+    },
+    {
+      "epoch": 1.5326354679802956,
+      "grad_norm": 0.08286473900079727,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 14934
+    },
+    {
+      "epoch": 1.5329433497536946,
+      "grad_norm": 0.042596206068992615,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 14937
+    },
+    {
+      "epoch": 1.5332512315270936,
+      "grad_norm": 0.043695200234651566,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 14940
+    },
+    {
+      "epoch": 1.5335591133004927,
+      "grad_norm": 0.13041776418685913,
+      "learning_rate": 0.01,
+      "loss": 2.0966,
+      "step": 14943
+    },
+    {
+      "epoch": 1.5338669950738915,
+      "grad_norm": 0.06076871603727341,
+      "learning_rate": 0.01,
+      "loss": 2.081,
+      "step": 14946
+    },
+    {
+      "epoch": 1.5341748768472905,
+      "grad_norm": 0.08744147419929504,
+      "learning_rate": 0.01,
+      "loss": 2.0698,
+      "step": 14949
+    },
+    {
+      "epoch": 1.5344827586206895,
+      "grad_norm": 0.041992440819740295,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 14952
+    },
+    {
+      "epoch": 1.5347906403940885,
+      "grad_norm": 0.0773782953619957,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 14955
+    },
+    {
+      "epoch": 1.5350985221674875,
+      "grad_norm": 0.04371657967567444,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 14958
+    },
+    {
+      "epoch": 1.5354064039408866,
+      "grad_norm": 0.04753347858786583,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 14961
+    },
+    {
+      "epoch": 1.5357142857142856,
+      "grad_norm": 0.10644037276506424,
+      "learning_rate": 0.01,
+      "loss": 2.0838,
+      "step": 14964
+    },
+    {
+      "epoch": 1.5360221674876846,
+      "grad_norm": 0.037067610770463943,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 14967
+    },
+    {
+      "epoch": 1.5363300492610836,
+      "grad_norm": 0.06745267659425735,
+      "learning_rate": 0.01,
+      "loss": 2.0464,
+      "step": 14970
+    },
+    {
+      "epoch": 1.5366379310344827,
+      "grad_norm": 0.10053039342164993,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 14973
+    },
+    {
+      "epoch": 1.5369458128078817,
+      "grad_norm": 0.08785562962293625,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 14976
+    },
+    {
+      "epoch": 1.5372536945812807,
+      "grad_norm": 0.12240536510944366,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 14979
+    },
+    {
+      "epoch": 1.5375615763546797,
+      "grad_norm": 0.08541588485240936,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 14982
+    },
+    {
+      "epoch": 1.5378694581280787,
+      "grad_norm": 0.0469081737101078,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 14985
+    },
+    {
+      "epoch": 1.5381773399014778,
+      "grad_norm": 0.04396476596593857,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 14988
+    },
+    {
+      "epoch": 1.5384852216748768,
+      "grad_norm": 0.033920448273420334,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 14991
+    },
+    {
+      "epoch": 1.5387931034482758,
+      "grad_norm": 0.035798102617263794,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 14994
+    },
+    {
+      "epoch": 1.5391009852216748,
+      "grad_norm": 0.030788132920861244,
+      "learning_rate": 0.01,
+      "loss": 2.0856,
+      "step": 14997
+    },
+    {
+      "epoch": 1.5394088669950738,
+      "grad_norm": 0.06127138063311577,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 15000
+    },
+    {
+      "epoch": 1.5397167487684729,
+      "grad_norm": 0.07792042940855026,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 15003
+    },
+    {
+      "epoch": 1.5400246305418719,
+      "grad_norm": 0.1915716826915741,
+      "learning_rate": 0.01,
+      "loss": 2.06,
+      "step": 15006
+    },
+    {
+      "epoch": 1.540332512315271,
+      "grad_norm": 0.13130734860897064,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 15009
+    },
+    {
+      "epoch": 1.54064039408867,
+      "grad_norm": 0.08750183880329132,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 15012
+    },
+    {
+      "epoch": 1.540948275862069,
+      "grad_norm": 0.0678631141781807,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 15015
+    },
+    {
+      "epoch": 1.541256157635468,
+      "grad_norm": 0.04352593049407005,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 15018
+    },
+    {
+      "epoch": 1.541564039408867,
+      "grad_norm": 0.059398628771305084,
+      "learning_rate": 0.01,
+      "loss": 2.0542,
+      "step": 15021
+    },
+    {
+      "epoch": 1.541871921182266,
+      "grad_norm": 0.09179355949163437,
+      "learning_rate": 0.01,
+      "loss": 2.0623,
+      "step": 15024
+    },
+    {
+      "epoch": 1.542179802955665,
+      "grad_norm": 0.08243024349212646,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 15027
+    },
+    {
+      "epoch": 1.542487684729064,
+      "grad_norm": 0.05103360861539841,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 15030
+    },
+    {
+      "epoch": 1.542795566502463,
+      "grad_norm": 0.04206395894289017,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 15033
+    },
+    {
+      "epoch": 1.543103448275862,
+      "grad_norm": 0.03659799322485924,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 15036
+    },
+    {
+      "epoch": 1.5434113300492611,
+      "grad_norm": 0.1279965043067932,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 15039
+    },
+    {
+      "epoch": 1.5437192118226601,
+      "grad_norm": 0.09509512782096863,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 15042
+    },
+    {
+      "epoch": 1.5440270935960592,
+      "grad_norm": 0.07963217794895172,
+      "learning_rate": 0.01,
+      "loss": 2.0632,
+      "step": 15045
+    },
+    {
+      "epoch": 1.5443349753694582,
+      "grad_norm": 0.06425557285547256,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 15048
+    },
+    {
+      "epoch": 1.5446428571428572,
+      "grad_norm": 0.1166144609451294,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 15051
+    },
+    {
+      "epoch": 1.5449507389162562,
+      "grad_norm": 0.0558270663022995,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 15054
+    },
+    {
+      "epoch": 1.5452586206896552,
+      "grad_norm": 0.05666494369506836,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 15057
+    },
+    {
+      "epoch": 1.5455665024630543,
+      "grad_norm": 0.048931702971458435,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 15060
+    },
+    {
+      "epoch": 1.5458743842364533,
+      "grad_norm": 0.10072410106658936,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 15063
+    },
+    {
+      "epoch": 1.5461822660098523,
+      "grad_norm": 0.06339754164218903,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 15066
+    },
+    {
+      "epoch": 1.5464901477832513,
+      "grad_norm": 0.04913650080561638,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 15069
+    },
+    {
+      "epoch": 1.5467980295566504,
+      "grad_norm": 0.1012924313545227,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 15072
+    },
+    {
+      "epoch": 1.5471059113300494,
+      "grad_norm": 0.048015668988227844,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 15075
+    },
+    {
+      "epoch": 1.5474137931034484,
+      "grad_norm": 0.09666754305362701,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 15078
+    },
+    {
+      "epoch": 1.5477216748768474,
+      "grad_norm": 0.07722094655036926,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 15081
+    },
+    {
+      "epoch": 1.5480295566502464,
+      "grad_norm": 0.06525082141160965,
+      "learning_rate": 0.01,
+      "loss": 2.056,
+      "step": 15084
+    },
+    {
+      "epoch": 1.5483374384236455,
+      "grad_norm": 0.04979628697037697,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 15087
+    },
+    {
+      "epoch": 1.5486453201970445,
+      "grad_norm": 0.05903888866305351,
+      "learning_rate": 0.01,
+      "loss": 2.0551,
+      "step": 15090
+    },
+    {
+      "epoch": 1.5489532019704435,
+      "grad_norm": 0.09098793566226959,
+      "learning_rate": 0.01,
+      "loss": 2.0758,
+      "step": 15093
+    },
+    {
+      "epoch": 1.5492610837438425,
+      "grad_norm": 0.08262350410223007,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 15096
+    },
+    {
+      "epoch": 1.5495689655172413,
+      "grad_norm": 0.057414278388023376,
+      "learning_rate": 0.01,
+      "loss": 2.0887,
+      "step": 15099
+    },
+    {
+      "epoch": 1.5498768472906403,
+      "grad_norm": 0.06718642264604568,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 15102
+    },
+    {
+      "epoch": 1.5501847290640394,
+      "grad_norm": 0.07351098954677582,
+      "learning_rate": 0.01,
+      "loss": 2.0589,
+      "step": 15105
+    },
+    {
+      "epoch": 1.5504926108374384,
+      "grad_norm": 0.03318174555897713,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 15108
+    },
+    {
+      "epoch": 1.5508004926108374,
+      "grad_norm": 0.11198091506958008,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 15111
+    },
+    {
+      "epoch": 1.5511083743842364,
+      "grad_norm": 0.056512147188186646,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 15114
+    },
+    {
+      "epoch": 1.5514162561576355,
+      "grad_norm": 0.08460327982902527,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 15117
+    },
+    {
+      "epoch": 1.5517241379310345,
+      "grad_norm": 0.08536583930253983,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 15120
+    },
+    {
+      "epoch": 1.5520320197044335,
+      "grad_norm": 0.10857357084751129,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 15123
+    },
+    {
+      "epoch": 1.5523399014778325,
+      "grad_norm": 0.04923904314637184,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 15126
+    },
+    {
+      "epoch": 1.5526477832512315,
+      "grad_norm": 0.05313669145107269,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 15129
+    },
+    {
+      "epoch": 1.5529556650246306,
+      "grad_norm": 0.058348409831523895,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 15132
+    },
+    {
+      "epoch": 1.5532635467980296,
+      "grad_norm": 0.04621830955147743,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 15135
+    },
+    {
+      "epoch": 1.5535714285714286,
+      "grad_norm": 0.079473577439785,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 15138
+    },
+    {
+      "epoch": 1.5538793103448276,
+      "grad_norm": 0.10176654160022736,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 15141
+    },
+    {
+      "epoch": 1.5541871921182266,
+      "grad_norm": 0.048997897654771805,
+      "learning_rate": 0.01,
+      "loss": 2.0832,
+      "step": 15144
+    },
+    {
+      "epoch": 1.5544950738916257,
+      "grad_norm": 0.06281887739896774,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 15147
+    },
+    {
+      "epoch": 1.5548029556650245,
+      "grad_norm": 0.07801464200019836,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 15150
+    },
+    {
+      "epoch": 1.5551108374384235,
+      "grad_norm": 0.08535271883010864,
+      "learning_rate": 0.01,
+      "loss": 2.0597,
+      "step": 15153
+    },
+    {
+      "epoch": 1.5554187192118225,
+      "grad_norm": 0.06536438316106796,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 15156
+    },
+    {
+      "epoch": 1.5557266009852215,
+      "grad_norm": 0.045388113707304,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 15159
+    },
+    {
+      "epoch": 1.5560344827586206,
+      "grad_norm": 0.041893068701028824,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 15162
+    },
+    {
+      "epoch": 1.5563423645320196,
+      "grad_norm": 0.04261694848537445,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 15165
+    },
+    {
+      "epoch": 1.5566502463054186,
+      "grad_norm": 0.03123985230922699,
+      "learning_rate": 0.01,
+      "loss": 2.0653,
+      "step": 15168
+    },
+    {
+      "epoch": 1.5569581280788176,
+      "grad_norm": 0.048562515527009964,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 15171
+    },
+    {
+      "epoch": 1.5572660098522166,
+      "grad_norm": 0.1343316286802292,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 15174
+    },
+    {
+      "epoch": 1.5575738916256157,
+      "grad_norm": 0.10992839932441711,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 15177
+    },
+    {
+      "epoch": 1.5578817733990147,
+      "grad_norm": 0.09098651260137558,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 15180
+    },
+    {
+      "epoch": 1.5581896551724137,
+      "grad_norm": 0.05405926704406738,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 15183
+    },
+    {
+      "epoch": 1.5584975369458127,
+      "grad_norm": 0.04776093736290932,
+      "learning_rate": 0.01,
+      "loss": 2.0775,
+      "step": 15186
+    },
+    {
+      "epoch": 1.5588054187192117,
+      "grad_norm": 0.04614724963903427,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 15189
+    },
+    {
+      "epoch": 1.5591133004926108,
+      "grad_norm": 0.05032865330576897,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 15192
+    },
+    {
+      "epoch": 1.5594211822660098,
+      "grad_norm": 0.051392171531915665,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 15195
+    },
+    {
+      "epoch": 1.5597290640394088,
+      "grad_norm": 0.10255944728851318,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 15198
+    },
+    {
+      "epoch": 1.5600369458128078,
+      "grad_norm": 0.04222560301423073,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 15201
+    },
+    {
+      "epoch": 1.5603448275862069,
+      "grad_norm": 0.045385826379060745,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 15204
+    },
+    {
+      "epoch": 1.5606527093596059,
+      "grad_norm": 0.04241577908396721,
+      "learning_rate": 0.01,
+      "loss": 2.0727,
+      "step": 15207
+    },
+    {
+      "epoch": 1.560960591133005,
+      "grad_norm": 0.12387125194072723,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 15210
+    },
+    {
+      "epoch": 1.561268472906404,
+      "grad_norm": 0.14704599976539612,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 15213
+    },
+    {
+      "epoch": 1.561576354679803,
+      "grad_norm": 0.049915559589862823,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 15216
+    },
+    {
+      "epoch": 1.561884236453202,
+      "grad_norm": 0.057630617171525955,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 15219
+    },
+    {
+      "epoch": 1.562192118226601,
+      "grad_norm": 0.04765351861715317,
+      "learning_rate": 0.01,
+      "loss": 2.0763,
+      "step": 15222
+    },
+    {
+      "epoch": 1.5625,
+      "grad_norm": 0.06213679164648056,
+      "learning_rate": 0.01,
+      "loss": 2.0831,
+      "step": 15225
+    },
+    {
+      "epoch": 1.562807881773399,
+      "grad_norm": 0.07990710437297821,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 15228
+    },
+    {
+      "epoch": 1.563115763546798,
+      "grad_norm": 0.0683673620223999,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 15231
+    },
+    {
+      "epoch": 1.563423645320197,
+      "grad_norm": 0.12503905594348907,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 15234
+    },
+    {
+      "epoch": 1.563731527093596,
+      "grad_norm": 0.03973531350493431,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 15237
+    },
+    {
+      "epoch": 1.564039408866995,
+      "grad_norm": 0.07055282592773438,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 15240
+    },
+    {
+      "epoch": 1.5643472906403941,
+      "grad_norm": 0.06088467687368393,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 15243
+    },
+    {
+      "epoch": 1.5646551724137931,
+      "grad_norm": 0.06393450498580933,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 15246
+    },
+    {
+      "epoch": 1.5649630541871922,
+      "grad_norm": 0.08600255101919174,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 15249
+    },
+    {
+      "epoch": 1.5652709359605912,
+      "grad_norm": 0.07075429707765579,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 15252
+    },
+    {
+      "epoch": 1.5655788177339902,
+      "grad_norm": 0.058057527989149094,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 15255
+    },
+    {
+      "epoch": 1.5658866995073892,
+      "grad_norm": 0.04670482128858566,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 15258
+    },
+    {
+      "epoch": 1.5661945812807883,
+      "grad_norm": 0.08971681445837021,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 15261
+    },
+    {
+      "epoch": 1.5665024630541873,
+      "grad_norm": 0.12580984830856323,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 15264
+    },
+    {
+      "epoch": 1.5668103448275863,
+      "grad_norm": 0.05133863538503647,
+      "learning_rate": 0.01,
+      "loss": 2.056,
+      "step": 15267
+    },
+    {
+      "epoch": 1.5671182266009853,
+      "grad_norm": 0.07821512222290039,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 15270
+    },
+    {
+      "epoch": 1.5674261083743843,
+      "grad_norm": 0.07024712860584259,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 15273
+    },
+    {
+      "epoch": 1.5677339901477834,
+      "grad_norm": 0.09332927316427231,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 15276
+    },
+    {
+      "epoch": 1.5680418719211824,
+      "grad_norm": 0.06875135749578476,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 15279
+    },
+    {
+      "epoch": 1.5683497536945814,
+      "grad_norm": 0.08868546783924103,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 15282
+    },
+    {
+      "epoch": 1.5686576354679804,
+      "grad_norm": 0.07729046791791916,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 15285
+    },
+    {
+      "epoch": 1.5689655172413794,
+      "grad_norm": 0.07686775177717209,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 15288
+    },
+    {
+      "epoch": 1.5692733990147785,
+      "grad_norm": 0.0839667096734047,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 15291
+    },
+    {
+      "epoch": 1.5695812807881775,
+      "grad_norm": 0.0704370066523552,
+      "learning_rate": 0.01,
+      "loss": 2.0638,
+      "step": 15294
+    },
+    {
+      "epoch": 1.5698891625615765,
+      "grad_norm": 0.05312497168779373,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 15297
+    },
+    {
+      "epoch": 1.5701970443349755,
+      "grad_norm": 0.049166906625032425,
+      "learning_rate": 0.01,
+      "loss": 2.0544,
+      "step": 15300
+    },
+    {
+      "epoch": 1.5705049261083743,
+      "grad_norm": 0.041398897767066956,
+      "learning_rate": 0.01,
+      "loss": 2.0652,
+      "step": 15303
+    },
+    {
+      "epoch": 1.5708128078817734,
+      "grad_norm": 0.08617027848958969,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 15306
+    },
+    {
+      "epoch": 1.5711206896551724,
+      "grad_norm": 0.0348927266895771,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 15309
+    },
+    {
+      "epoch": 1.5714285714285714,
+      "grad_norm": 0.060787077993154526,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 15312
+    },
+    {
+      "epoch": 1.5717364532019704,
+      "grad_norm": 0.050898227840662,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 15315
+    },
+    {
+      "epoch": 1.5720443349753694,
+      "grad_norm": 0.04594309255480766,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 15318
+    },
+    {
+      "epoch": 1.5723522167487685,
+      "grad_norm": 0.1161418929696083,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 15321
+    },
+    {
+      "epoch": 1.5726600985221675,
+      "grad_norm": 0.05419136583805084,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 15324
+    },
+    {
+      "epoch": 1.5729679802955665,
+      "grad_norm": 0.07661257684230804,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 15327
+    },
+    {
+      "epoch": 1.5732758620689655,
+      "grad_norm": 0.09760436415672302,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 15330
+    },
+    {
+      "epoch": 1.5735837438423645,
+      "grad_norm": 0.07211121916770935,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 15333
+    },
+    {
+      "epoch": 1.5738916256157636,
+      "grad_norm": 0.08360971510410309,
+      "learning_rate": 0.01,
+      "loss": 2.0465,
+      "step": 15336
+    },
+    {
+      "epoch": 1.5741995073891626,
+      "grad_norm": 0.05901337414979935,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 15339
+    },
+    {
+      "epoch": 1.5745073891625616,
+      "grad_norm": 0.07543767988681793,
+      "learning_rate": 0.01,
+      "loss": 2.0437,
+      "step": 15342
+    },
+    {
+      "epoch": 1.5748152709359606,
+      "grad_norm": 0.04725690186023712,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 15345
+    },
+    {
+      "epoch": 1.5751231527093597,
+      "grad_norm": 0.051067035645246506,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 15348
+    },
+    {
+      "epoch": 1.5754310344827587,
+      "grad_norm": 0.03145357966423035,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 15351
+    },
+    {
+      "epoch": 1.5757389162561575,
+      "grad_norm": 0.09291981905698776,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 15354
+    },
+    {
+      "epoch": 1.5760467980295565,
+      "grad_norm": 0.03409574180841446,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 15357
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.045993223786354065,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 15360
+    },
+    {
+      "epoch": 1.5766625615763545,
+      "grad_norm": 0.03174202889204025,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 15363
+    },
+    {
+      "epoch": 1.5769704433497536,
+      "grad_norm": 0.057013627141714096,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 15366
+    },
+    {
+      "epoch": 1.5772783251231526,
+      "grad_norm": 0.05778640881180763,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 15369
+    },
+    {
+      "epoch": 1.5775862068965516,
+      "grad_norm": 0.11259103566408157,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 15372
+    },
+    {
+      "epoch": 1.5778940886699506,
+      "grad_norm": 0.05728684365749359,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 15375
+    },
+    {
+      "epoch": 1.5782019704433496,
+      "grad_norm": 0.07037964463233948,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 15378
+    },
+    {
+      "epoch": 1.5785098522167487,
+      "grad_norm": 0.05147834122180939,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 15381
+    },
+    {
+      "epoch": 1.5788177339901477,
+      "grad_norm": 0.0663742870092392,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 15384
+    },
+    {
+      "epoch": 1.5791256157635467,
+      "grad_norm": 0.04665178433060646,
+      "learning_rate": 0.01,
+      "loss": 2.0698,
+      "step": 15387
+    },
+    {
+      "epoch": 1.5794334975369457,
+      "grad_norm": 0.037410903722047806,
+      "learning_rate": 0.01,
+      "loss": 2.0529,
+      "step": 15390
+    },
+    {
+      "epoch": 1.5797413793103448,
+      "grad_norm": 0.03849703446030617,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 15393
+    },
+    {
+      "epoch": 1.5800492610837438,
+      "grad_norm": 0.10119396448135376,
+      "learning_rate": 0.01,
+      "loss": 2.0752,
+      "step": 15396
+    },
+    {
+      "epoch": 1.5803571428571428,
+      "grad_norm": 0.03294401615858078,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 15399
+    },
+    {
+      "epoch": 1.5806650246305418,
+      "grad_norm": 0.06310781091451645,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 15402
+    },
+    {
+      "epoch": 1.5809729064039408,
+      "grad_norm": 0.050191253423690796,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 15405
+    },
+    {
+      "epoch": 1.5812807881773399,
+      "grad_norm": 0.09952792525291443,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 15408
+    },
+    {
+      "epoch": 1.5815886699507389,
+      "grad_norm": 0.12618017196655273,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 15411
+    },
+    {
+      "epoch": 1.581896551724138,
+      "grad_norm": 0.19296571612358093,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 15414
+    },
+    {
+      "epoch": 1.582204433497537,
+      "grad_norm": 0.1124732494354248,
+      "learning_rate": 0.01,
+      "loss": 2.0451,
+      "step": 15417
+    },
+    {
+      "epoch": 1.582512315270936,
+      "grad_norm": 0.06556060910224915,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 15420
+    },
+    {
+      "epoch": 1.582820197044335,
+      "grad_norm": 0.05607735365629196,
+      "learning_rate": 0.01,
+      "loss": 2.0663,
+      "step": 15423
+    },
+    {
+      "epoch": 1.583128078817734,
+      "grad_norm": 0.04731295630335808,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 15426
+    },
+    {
+      "epoch": 1.583435960591133,
+      "grad_norm": 0.060452669858932495,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 15429
+    },
+    {
+      "epoch": 1.583743842364532,
+      "grad_norm": 0.056996360421180725,
+      "learning_rate": 0.01,
+      "loss": 2.0752,
+      "step": 15432
+    },
+    {
+      "epoch": 1.584051724137931,
+      "grad_norm": 0.11382313817739487,
+      "learning_rate": 0.01,
+      "loss": 2.0723,
+      "step": 15435
+    },
+    {
+      "epoch": 1.58435960591133,
+      "grad_norm": 0.05176008865237236,
+      "learning_rate": 0.01,
+      "loss": 2.0706,
+      "step": 15438
+    },
+    {
+      "epoch": 1.584667487684729,
+      "grad_norm": 0.05329972878098488,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 15441
+    },
+    {
+      "epoch": 1.5849753694581281,
+      "grad_norm": 0.05416284501552582,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 15444
+    },
+    {
+      "epoch": 1.5852832512315271,
+      "grad_norm": 0.052267350256443024,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 15447
+    },
+    {
+      "epoch": 1.5855911330049262,
+      "grad_norm": 0.1025579646229744,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 15450
+    },
+    {
+      "epoch": 1.5858990147783252,
+      "grad_norm": 0.06416940689086914,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 15453
+    },
+    {
+      "epoch": 1.5862068965517242,
+      "grad_norm": 0.05699596926569939,
+      "learning_rate": 0.01,
+      "loss": 2.0678,
+      "step": 15456
+    },
+    {
+      "epoch": 1.5865147783251232,
+      "grad_norm": 0.036711812019348145,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 15459
+    },
+    {
+      "epoch": 1.5868226600985222,
+      "grad_norm": 0.1025582030415535,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 15462
+    },
+    {
+      "epoch": 1.5871305418719213,
+      "grad_norm": 0.03923096880316734,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 15465
+    },
+    {
+      "epoch": 1.5874384236453203,
+      "grad_norm": 0.08267144113779068,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 15468
+    },
+    {
+      "epoch": 1.5877463054187193,
+      "grad_norm": 0.15374930202960968,
+      "learning_rate": 0.01,
+      "loss": 2.0384,
+      "step": 15471
+    },
+    {
+      "epoch": 1.5880541871921183,
+      "grad_norm": 0.10246127098798752,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 15474
+    },
+    {
+      "epoch": 1.5883620689655173,
+      "grad_norm": 0.09784136712551117,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 15477
+    },
+    {
+      "epoch": 1.5886699507389164,
+      "grad_norm": 0.08747786283493042,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 15480
+    },
+    {
+      "epoch": 1.5889778325123154,
+      "grad_norm": 0.0755406990647316,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 15483
+    },
+    {
+      "epoch": 1.5892857142857144,
+      "grad_norm": 0.05593521520495415,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 15486
+    },
+    {
+      "epoch": 1.5895935960591134,
+      "grad_norm": 0.04462866857647896,
+      "learning_rate": 0.01,
+      "loss": 2.056,
+      "step": 15489
+    },
+    {
+      "epoch": 1.5899014778325125,
+      "grad_norm": 0.040571678429841995,
+      "learning_rate": 0.01,
+      "loss": 2.0906,
+      "step": 15492
+    },
+    {
+      "epoch": 1.5902093596059115,
+      "grad_norm": 0.038100458681583405,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 15495
+    },
+    {
+      "epoch": 1.5905172413793105,
+      "grad_norm": 0.04336906597018242,
+      "learning_rate": 0.01,
+      "loss": 2.0716,
+      "step": 15498
+    },
+    {
+      "epoch": 1.5908251231527095,
+      "grad_norm": 0.1424836665391922,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 15501
+    },
+    {
+      "epoch": 1.5911330049261085,
+      "grad_norm": 0.09235331416130066,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 15504
+    },
+    {
+      "epoch": 1.5914408866995073,
+      "grad_norm": 0.07932816445827484,
+      "learning_rate": 0.01,
+      "loss": 2.0595,
+      "step": 15507
+    },
+    {
+      "epoch": 1.5917487684729064,
+      "grad_norm": 0.057297565042972565,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 15510
+    },
+    {
+      "epoch": 1.5920566502463054,
+      "grad_norm": 0.05323542281985283,
+      "learning_rate": 0.01,
+      "loss": 2.0568,
+      "step": 15513
+    },
+    {
+      "epoch": 1.5923645320197044,
+      "grad_norm": 0.039465416222810745,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 15516
+    },
+    {
+      "epoch": 1.5926724137931034,
+      "grad_norm": 0.044614970684051514,
+      "learning_rate": 0.01,
+      "loss": 2.0784,
+      "step": 15519
+    },
+    {
+      "epoch": 1.5929802955665024,
+      "grad_norm": 0.044074248522520065,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 15522
+    },
+    {
+      "epoch": 1.5932881773399015,
+      "grad_norm": 0.04098647087812424,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 15525
+    },
+    {
+      "epoch": 1.5935960591133005,
+      "grad_norm": 0.18400810658931732,
+      "learning_rate": 0.01,
+      "loss": 2.0623,
+      "step": 15528
+    },
+    {
+      "epoch": 1.5939039408866995,
+      "grad_norm": 0.10264160484075546,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 15531
+    },
+    {
+      "epoch": 1.5942118226600985,
+      "grad_norm": 0.10086511820554733,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 15534
+    },
+    {
+      "epoch": 1.5945197044334976,
+      "grad_norm": 0.03823179379105568,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 15537
+    },
+    {
+      "epoch": 1.5948275862068966,
+      "grad_norm": 0.0449577271938324,
+      "learning_rate": 0.01,
+      "loss": 2.0635,
+      "step": 15540
+    },
+    {
+      "epoch": 1.5951354679802956,
+      "grad_norm": 0.04791559278964996,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 15543
+    },
+    {
+      "epoch": 1.5954433497536946,
+      "grad_norm": 0.04523475095629692,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 15546
+    },
+    {
+      "epoch": 1.5957512315270936,
+      "grad_norm": 0.10654012113809586,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 15549
+    },
+    {
+      "epoch": 1.5960591133004927,
+      "grad_norm": 0.06602972745895386,
+      "learning_rate": 0.01,
+      "loss": 2.0565,
+      "step": 15552
+    },
+    {
+      "epoch": 1.5963669950738915,
+      "grad_norm": 0.10605626553297043,
+      "learning_rate": 0.01,
+      "loss": 2.0792,
+      "step": 15555
+    },
+    {
+      "epoch": 1.5966748768472905,
+      "grad_norm": 0.05995124578475952,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 15558
+    },
+    {
+      "epoch": 1.5969827586206895,
+      "grad_norm": 0.05426995828747749,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 15561
+    },
+    {
+      "epoch": 1.5972906403940885,
+      "grad_norm": 0.08749561756849289,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 15564
+    },
+    {
+      "epoch": 1.5975985221674875,
+      "grad_norm": 0.0735105574131012,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 15567
+    },
+    {
+      "epoch": 1.5979064039408866,
+      "grad_norm": 0.05417585372924805,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 15570
+    },
+    {
+      "epoch": 1.5982142857142856,
+      "grad_norm": 0.04170646145939827,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 15573
+    },
+    {
+      "epoch": 1.5985221674876846,
+      "grad_norm": 0.05886775627732277,
+      "learning_rate": 0.01,
+      "loss": 2.0419,
+      "step": 15576
+    },
+    {
+      "epoch": 1.5988300492610836,
+      "grad_norm": 0.04231201857328415,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 15579
+    },
+    {
+      "epoch": 1.5991379310344827,
+      "grad_norm": 0.06620585918426514,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 15582
+    },
+    {
+      "epoch": 1.5994458128078817,
+      "grad_norm": 0.10536913573741913,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 15585
+    },
+    {
+      "epoch": 1.5997536945812807,
+      "grad_norm": 0.14467884600162506,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 15588
+    },
+    {
+      "epoch": 1.6000615763546797,
+      "grad_norm": 0.11715273559093475,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 15591
+    },
+    {
+      "epoch": 1.6003694581280787,
+      "grad_norm": 0.04978121817111969,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 15594
+    },
+    {
+      "epoch": 1.6006773399014778,
+      "grad_norm": 0.06248803436756134,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 15597
+    },
+    {
+      "epoch": 1.6009852216748768,
+      "grad_norm": 0.05408048257231712,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 15600
+    },
+    {
+      "epoch": 1.6012931034482758,
+      "grad_norm": 0.05805948004126549,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 15603
+    },
+    {
+      "epoch": 1.6016009852216748,
+      "grad_norm": 0.03809194639325142,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 15606
+    },
+    {
+      "epoch": 1.6019088669950738,
+      "grad_norm": 0.07981141656637192,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 15609
+    },
+    {
+      "epoch": 1.6022167487684729,
+      "grad_norm": 0.04769575223326683,
+      "learning_rate": 0.01,
+      "loss": 2.0703,
+      "step": 15612
+    },
+    {
+      "epoch": 1.6025246305418719,
+      "grad_norm": 0.09913644194602966,
+      "learning_rate": 0.01,
+      "loss": 2.074,
+      "step": 15615
+    },
+    {
+      "epoch": 1.602832512315271,
+      "grad_norm": 0.12298569083213806,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 15618
+    },
+    {
+      "epoch": 1.60314039408867,
+      "grad_norm": 0.0525309219956398,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 15621
+    },
+    {
+      "epoch": 1.603448275862069,
+      "grad_norm": 0.07430320978164673,
+      "learning_rate": 0.01,
+      "loss": 2.0465,
+      "step": 15624
+    },
+    {
+      "epoch": 1.603756157635468,
+      "grad_norm": 0.036753058433532715,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 15627
+    },
+    {
+      "epoch": 1.604064039408867,
+      "grad_norm": 0.04560523107647896,
+      "learning_rate": 0.01,
+      "loss": 2.074,
+      "step": 15630
+    },
+    {
+      "epoch": 1.604371921182266,
+      "grad_norm": 0.07089810073375702,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 15633
+    },
+    {
+      "epoch": 1.604679802955665,
+      "grad_norm": 0.10833004862070084,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 15636
+    },
+    {
+      "epoch": 1.604987684729064,
+      "grad_norm": 0.06033416837453842,
+      "learning_rate": 0.01,
+      "loss": 2.0893,
+      "step": 15639
+    },
+    {
+      "epoch": 1.605295566502463,
+      "grad_norm": 0.06819162517786026,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 15642
+    },
+    {
+      "epoch": 1.605603448275862,
+      "grad_norm": 0.08949002623558044,
+      "learning_rate": 0.01,
+      "loss": 2.0891,
+      "step": 15645
+    },
+    {
+      "epoch": 1.6059113300492611,
+      "grad_norm": 0.04749004542827606,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 15648
+    },
+    {
+      "epoch": 1.6062192118226601,
+      "grad_norm": 0.06903103739023209,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 15651
+    },
+    {
+      "epoch": 1.6065270935960592,
+      "grad_norm": 0.10074819624423981,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 15654
+    },
+    {
+      "epoch": 1.6068349753694582,
+      "grad_norm": 0.0390753298997879,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 15657
+    },
+    {
+      "epoch": 1.6071428571428572,
+      "grad_norm": 0.04776669666171074,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 15660
+    },
+    {
+      "epoch": 1.6074507389162562,
+      "grad_norm": 0.1191340908408165,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 15663
+    },
+    {
+      "epoch": 1.6077586206896552,
+      "grad_norm": 0.08326657861471176,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 15666
+    },
+    {
+      "epoch": 1.6080665024630543,
+      "grad_norm": 0.044734589755535126,
+      "learning_rate": 0.01,
+      "loss": 2.0626,
+      "step": 15669
+    },
+    {
+      "epoch": 1.6083743842364533,
+      "grad_norm": 0.047262392938137054,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 15672
+    },
+    {
+      "epoch": 1.6086822660098523,
+      "grad_norm": 0.0908563882112503,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 15675
+    },
+    {
+      "epoch": 1.6089901477832513,
+      "grad_norm": 0.06681264191865921,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 15678
+    },
+    {
+      "epoch": 1.6092980295566504,
+      "grad_norm": 0.09569018334150314,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 15681
+    },
+    {
+      "epoch": 1.6096059113300494,
+      "grad_norm": 0.04303963482379913,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 15684
+    },
+    {
+      "epoch": 1.6099137931034484,
+      "grad_norm": 0.09924867749214172,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 15687
+    },
+    {
+      "epoch": 1.6102216748768474,
+      "grad_norm": 0.041328392922878265,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 15690
+    },
+    {
+      "epoch": 1.6105295566502464,
+      "grad_norm": 0.056367840617895126,
+      "learning_rate": 0.01,
+      "loss": 2.0746,
+      "step": 15693
+    },
+    {
+      "epoch": 1.6108374384236455,
+      "grad_norm": 0.06074264645576477,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 15696
+    },
+    {
+      "epoch": 1.6111453201970445,
+      "grad_norm": 0.06541740894317627,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 15699
+    },
+    {
+      "epoch": 1.6114532019704435,
+      "grad_norm": 0.06279835850000381,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 15702
+    },
+    {
+      "epoch": 1.6117610837438425,
+      "grad_norm": 0.03825109452009201,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 15705
+    },
+    {
+      "epoch": 1.6120689655172413,
+      "grad_norm": 0.03792817145586014,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 15708
+    },
+    {
+      "epoch": 1.6123768472906403,
+      "grad_norm": 0.05229473114013672,
+      "learning_rate": 0.01,
+      "loss": 2.0765,
+      "step": 15711
+    },
+    {
+      "epoch": 1.6126847290640394,
+      "grad_norm": 0.11285384744405746,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 15714
+    },
+    {
+      "epoch": 1.6129926108374384,
+      "grad_norm": 0.07333546876907349,
+      "learning_rate": 0.01,
+      "loss": 2.05,
+      "step": 15717
+    },
+    {
+      "epoch": 1.6133004926108374,
+      "grad_norm": 0.07698936760425568,
+      "learning_rate": 0.01,
+      "loss": 2.0757,
+      "step": 15720
+    },
+    {
+      "epoch": 1.6136083743842364,
+      "grad_norm": 0.06517963856458664,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 15723
+    },
+    {
+      "epoch": 1.6139162561576355,
+      "grad_norm": 0.07242800295352936,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 15726
+    },
+    {
+      "epoch": 1.6142241379310345,
+      "grad_norm": 0.03956649452447891,
+      "learning_rate": 0.01,
+      "loss": 2.0437,
+      "step": 15729
+    },
+    {
+      "epoch": 1.6145320197044335,
+      "grad_norm": 0.10249898582696915,
+      "learning_rate": 0.01,
+      "loss": 2.0716,
+      "step": 15732
+    },
+    {
+      "epoch": 1.6148399014778325,
+      "grad_norm": 0.09716839343309402,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 15735
+    },
+    {
+      "epoch": 1.6151477832512315,
+      "grad_norm": 0.0809134840965271,
+      "learning_rate": 0.01,
+      "loss": 2.0553,
+      "step": 15738
+    },
+    {
+      "epoch": 1.6154556650246306,
+      "grad_norm": 0.07891330122947693,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 15741
+    },
+    {
+      "epoch": 1.6157635467980296,
+      "grad_norm": 0.06289231032133102,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 15744
+    },
+    {
+      "epoch": 1.6160714285714286,
+      "grad_norm": 0.032251689583063126,
+      "learning_rate": 0.01,
+      "loss": 2.049,
+      "step": 15747
+    },
+    {
+      "epoch": 1.6163793103448276,
+      "grad_norm": 0.032203931361436844,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 15750
+    },
+    {
+      "epoch": 1.6166871921182266,
+      "grad_norm": 0.042572617530822754,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 15753
+    },
+    {
+      "epoch": 1.6169950738916257,
+      "grad_norm": 0.06869769096374512,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 15756
+    },
+    {
+      "epoch": 1.6173029556650245,
+      "grad_norm": 0.09649953991174698,
+      "learning_rate": 0.01,
+      "loss": 2.0811,
+      "step": 15759
+    },
+    {
+      "epoch": 1.6176108374384235,
+      "grad_norm": 0.060255225747823715,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 15762
+    },
+    {
+      "epoch": 1.6179187192118225,
+      "grad_norm": 0.0548517182469368,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 15765
+    },
+    {
+      "epoch": 1.6182266009852215,
+      "grad_norm": 0.09392546862363815,
+      "learning_rate": 0.01,
+      "loss": 2.0722,
+      "step": 15768
+    },
+    {
+      "epoch": 1.6185344827586206,
+      "grad_norm": 0.052100926637649536,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 15771
+    },
+    {
+      "epoch": 1.6188423645320196,
+      "grad_norm": 0.05099212005734444,
+      "learning_rate": 0.01,
+      "loss": 2.0589,
+      "step": 15774
+    },
+    {
+      "epoch": 1.6191502463054186,
+      "grad_norm": 0.0486266165971756,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 15777
+    },
+    {
+      "epoch": 1.6194581280788176,
+      "grad_norm": 0.044072605669498444,
+      "learning_rate": 0.01,
+      "loss": 2.0877,
+      "step": 15780
+    },
+    {
+      "epoch": 1.6197660098522166,
+      "grad_norm": 0.09196856617927551,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 15783
+    },
+    {
+      "epoch": 1.6200738916256157,
+      "grad_norm": 0.05948984995484352,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 15786
+    },
+    {
+      "epoch": 1.6203817733990147,
+      "grad_norm": 0.043075162917375565,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 15789
+    },
+    {
+      "epoch": 1.6206896551724137,
+      "grad_norm": 0.06739038228988647,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 15792
+    },
+    {
+      "epoch": 1.6209975369458127,
+      "grad_norm": 0.05961238220334053,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 15795
+    },
+    {
+      "epoch": 1.6213054187192117,
+      "grad_norm": 0.06527238339185715,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 15798
+    },
+    {
+      "epoch": 1.6216133004926108,
+      "grad_norm": 0.09234929084777832,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 15801
+    },
+    {
+      "epoch": 1.6219211822660098,
+      "grad_norm": 0.08050446212291718,
+      "learning_rate": 0.01,
+      "loss": 2.0769,
+      "step": 15804
+    },
+    {
+      "epoch": 1.6222290640394088,
+      "grad_norm": 0.06419754028320312,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 15807
+    },
+    {
+      "epoch": 1.6225369458128078,
+      "grad_norm": 0.06302323937416077,
+      "learning_rate": 0.01,
+      "loss": 2.0535,
+      "step": 15810
+    },
+    {
+      "epoch": 1.6228448275862069,
+      "grad_norm": 0.051602717489004135,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 15813
+    },
+    {
+      "epoch": 1.6231527093596059,
+      "grad_norm": 0.12424405664205551,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 15816
+    },
+    {
+      "epoch": 1.623460591133005,
+      "grad_norm": 0.10444232821464539,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 15819
+    },
+    {
+      "epoch": 1.623768472906404,
+      "grad_norm": 0.06170908361673355,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 15822
+    },
+    {
+      "epoch": 1.624076354679803,
+      "grad_norm": 0.05145244672894478,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 15825
+    },
+    {
+      "epoch": 1.624384236453202,
+      "grad_norm": 0.0459282286465168,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 15828
+    },
+    {
+      "epoch": 1.624692118226601,
+      "grad_norm": 0.05250949412584305,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 15831
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 0.03222022205591202,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 15834
+    },
+    {
+      "epoch": 1.625307881773399,
+      "grad_norm": 0.05849120765924454,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 15837
+    },
+    {
+      "epoch": 1.625615763546798,
+      "grad_norm": 0.04638088122010231,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 15840
+    },
+    {
+      "epoch": 1.625923645320197,
+      "grad_norm": 0.046597037464380264,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 15843
+    },
+    {
+      "epoch": 1.626231527093596,
+      "grad_norm": 0.10477445274591446,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 15846
+    },
+    {
+      "epoch": 1.626539408866995,
+      "grad_norm": 0.03439783677458763,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 15849
+    },
+    {
+      "epoch": 1.6268472906403941,
+      "grad_norm": 0.05810544639825821,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 15852
+    },
+    {
+      "epoch": 1.6271551724137931,
+      "grad_norm": 0.04557522386312485,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 15855
+    },
+    {
+      "epoch": 1.6274630541871922,
+      "grad_norm": 0.04236530885100365,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 15858
+    },
+    {
+      "epoch": 1.6277709359605912,
+      "grad_norm": 0.04338948428630829,
+      "learning_rate": 0.01,
+      "loss": 2.0584,
+      "step": 15861
+    },
+    {
+      "epoch": 1.6280788177339902,
+      "grad_norm": 0.039782583713531494,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 15864
+    },
+    {
+      "epoch": 1.6283866995073892,
+      "grad_norm": 0.06858891248703003,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 15867
+    },
+    {
+      "epoch": 1.6286945812807883,
+      "grad_norm": 0.0510399155318737,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 15870
+    },
+    {
+      "epoch": 1.6290024630541873,
+      "grad_norm": 0.12568604946136475,
+      "learning_rate": 0.01,
+      "loss": 2.0597,
+      "step": 15873
+    },
+    {
+      "epoch": 1.6293103448275863,
+      "grad_norm": 0.09245727956295013,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 15876
+    },
+    {
+      "epoch": 1.6296182266009853,
+      "grad_norm": 0.05763734132051468,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 15879
+    },
+    {
+      "epoch": 1.6299261083743843,
+      "grad_norm": 0.06099852919578552,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 15882
+    },
+    {
+      "epoch": 1.6302339901477834,
+      "grad_norm": 0.05738021805882454,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 15885
+    },
+    {
+      "epoch": 1.6305418719211824,
+      "grad_norm": 0.04953853040933609,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 15888
+    },
+    {
+      "epoch": 1.6308497536945814,
+      "grad_norm": 0.08572196215391159,
+      "learning_rate": 0.01,
+      "loss": 2.0618,
+      "step": 15891
+    },
+    {
+      "epoch": 1.6311576354679804,
+      "grad_norm": 0.09245479106903076,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 15894
+    },
+    {
+      "epoch": 1.6314655172413794,
+      "grad_norm": 0.057964712381362915,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 15897
+    },
+    {
+      "epoch": 1.6317733990147785,
+      "grad_norm": 0.05189305916428566,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 15900
+    },
+    {
+      "epoch": 1.6320812807881775,
+      "grad_norm": 0.07327884435653687,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 15903
+    },
+    {
+      "epoch": 1.6323891625615765,
+      "grad_norm": 0.07089177519083023,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 15906
+    },
+    {
+      "epoch": 1.6326970443349755,
+      "grad_norm": 0.09783073514699936,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 15909
+    },
+    {
+      "epoch": 1.6330049261083743,
+      "grad_norm": 0.06617991626262665,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 15912
+    },
+    {
+      "epoch": 1.6333128078817734,
+      "grad_norm": 0.10033921152353287,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 15915
+    },
+    {
+      "epoch": 1.6336206896551724,
+      "grad_norm": 0.054432835429906845,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 15918
+    },
+    {
+      "epoch": 1.6339285714285714,
+      "grad_norm": 0.056940387934446335,
+      "learning_rate": 0.01,
+      "loss": 2.0566,
+      "step": 15921
+    },
+    {
+      "epoch": 1.6342364532019704,
+      "grad_norm": 0.12047278136014938,
+      "learning_rate": 0.01,
+      "loss": 2.0464,
+      "step": 15924
+    },
+    {
+      "epoch": 1.6345443349753694,
+      "grad_norm": 0.04637087881565094,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 15927
+    },
+    {
+      "epoch": 1.6348522167487685,
+      "grad_norm": 0.03925006836652756,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 15930
+    },
+    {
+      "epoch": 1.6351600985221675,
+      "grad_norm": 0.04180562496185303,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 15933
+    },
+    {
+      "epoch": 1.6354679802955665,
+      "grad_norm": 0.08031155914068222,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 15936
+    },
+    {
+      "epoch": 1.6357758620689655,
+      "grad_norm": 0.0812869518995285,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 15939
+    },
+    {
+      "epoch": 1.6360837438423645,
+      "grad_norm": 0.07887094467878342,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 15942
+    },
+    {
+      "epoch": 1.6363916256157636,
+      "grad_norm": 0.12604457139968872,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 15945
+    },
+    {
+      "epoch": 1.6366995073891626,
+      "grad_norm": 0.1262006163597107,
+      "learning_rate": 0.01,
+      "loss": 2.0806,
+      "step": 15948
+    },
+    {
+      "epoch": 1.6370073891625616,
+      "grad_norm": 0.07335629314184189,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 15951
+    },
+    {
+      "epoch": 1.6373152709359606,
+      "grad_norm": 0.043172985315322876,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 15954
+    },
+    {
+      "epoch": 1.6376231527093597,
+      "grad_norm": 0.07475942373275757,
+      "learning_rate": 0.01,
+      "loss": 2.0842,
+      "step": 15957
+    },
+    {
+      "epoch": 1.6379310344827587,
+      "grad_norm": 0.06113087013363838,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 15960
+    },
+    {
+      "epoch": 1.6382389162561575,
+      "grad_norm": 0.08709672093391418,
+      "learning_rate": 0.01,
+      "loss": 2.0886,
+      "step": 15963
+    },
+    {
+      "epoch": 1.6385467980295565,
+      "grad_norm": 0.05810529738664627,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 15966
+    },
+    {
+      "epoch": 1.6388546798029555,
+      "grad_norm": 0.0831620916724205,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 15969
+    },
+    {
+      "epoch": 1.6391625615763545,
+      "grad_norm": 0.040577951818704605,
+      "learning_rate": 0.01,
+      "loss": 2.0676,
+      "step": 15972
+    },
+    {
+      "epoch": 1.6394704433497536,
+      "grad_norm": 0.03428385406732559,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 15975
+    },
+    {
+      "epoch": 1.6397783251231526,
+      "grad_norm": 0.04800771176815033,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 15978
+    },
+    {
+      "epoch": 1.6400862068965516,
+      "grad_norm": 0.05769934877753258,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 15981
+    },
+    {
+      "epoch": 1.6403940886699506,
+      "grad_norm": 0.08842117339372635,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 15984
+    },
+    {
+      "epoch": 1.6407019704433496,
+      "grad_norm": 0.09740414470434189,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 15987
+    },
+    {
+      "epoch": 1.6410098522167487,
+      "grad_norm": 0.11128890514373779,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 15990
+    },
+    {
+      "epoch": 1.6413177339901477,
+      "grad_norm": 0.03690354898571968,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 15993
+    },
+    {
+      "epoch": 1.6416256157635467,
+      "grad_norm": 0.07311075925827026,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 15996
+    },
+    {
+      "epoch": 1.6419334975369457,
+      "grad_norm": 0.045825451612472534,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 15999
+    },
+    {
+      "epoch": 1.6422413793103448,
+      "grad_norm": 0.09123300760984421,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 16002
+    },
+    {
+      "epoch": 1.6425492610837438,
+      "grad_norm": 0.0702185183763504,
+      "learning_rate": 0.01,
+      "loss": 2.0715,
+      "step": 16005
+    },
+    {
+      "epoch": 1.6428571428571428,
+      "grad_norm": 0.0355604812502861,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 16008
+    },
+    {
+      "epoch": 1.6431650246305418,
+      "grad_norm": 0.03151632100343704,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 16011
+    },
+    {
+      "epoch": 1.6434729064039408,
+      "grad_norm": 0.04302441328763962,
+      "learning_rate": 0.01,
+      "loss": 2.0824,
+      "step": 16014
+    },
+    {
+      "epoch": 1.6437807881773399,
+      "grad_norm": 0.06012306734919548,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 16017
+    },
+    {
+      "epoch": 1.6440886699507389,
+      "grad_norm": 0.04698712378740311,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 16020
+    },
+    {
+      "epoch": 1.644396551724138,
+      "grad_norm": 0.03930363059043884,
+      "learning_rate": 0.01,
+      "loss": 2.0612,
+      "step": 16023
+    },
+    {
+      "epoch": 1.644704433497537,
+      "grad_norm": 0.0881473496556282,
+      "learning_rate": 0.01,
+      "loss": 2.0724,
+      "step": 16026
+    },
+    {
+      "epoch": 1.645012315270936,
+      "grad_norm": 0.04207085818052292,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 16029
+    },
+    {
+      "epoch": 1.645320197044335,
+      "grad_norm": 0.04729215428233147,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 16032
+    },
+    {
+      "epoch": 1.645628078817734,
+      "grad_norm": 0.050990305840969086,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 16035
+    },
+    {
+      "epoch": 1.645935960591133,
+      "grad_norm": 0.05049813538789749,
+      "learning_rate": 0.01,
+      "loss": 2.0609,
+      "step": 16038
+    },
+    {
+      "epoch": 1.646243842364532,
+      "grad_norm": 0.07787630707025528,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 16041
+    },
+    {
+      "epoch": 1.646551724137931,
+      "grad_norm": 0.06656081229448318,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 16044
+    },
+    {
+      "epoch": 1.64685960591133,
+      "grad_norm": 0.08293109387159348,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 16047
+    },
+    {
+      "epoch": 1.647167487684729,
+      "grad_norm": 0.0775810182094574,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 16050
+    },
+    {
+      "epoch": 1.6474753694581281,
+      "grad_norm": 0.07917825132608414,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 16053
+    },
+    {
+      "epoch": 1.6477832512315271,
+      "grad_norm": 0.07658717036247253,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 16056
+    },
+    {
+      "epoch": 1.6480911330049262,
+      "grad_norm": 0.07735300809144974,
+      "learning_rate": 0.01,
+      "loss": 2.0481,
+      "step": 16059
+    },
+    {
+      "epoch": 1.6483990147783252,
+      "grad_norm": 0.07964644581079483,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 16062
+    },
+    {
+      "epoch": 1.6487068965517242,
+      "grad_norm": 0.0601799339056015,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 16065
+    },
+    {
+      "epoch": 1.6490147783251232,
+      "grad_norm": 0.1039920225739479,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 16068
+    },
+    {
+      "epoch": 1.6493226600985222,
+      "grad_norm": 0.055755455046892166,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 16071
+    },
+    {
+      "epoch": 1.6496305418719213,
+      "grad_norm": 0.0998646542429924,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 16074
+    },
+    {
+      "epoch": 1.6499384236453203,
+      "grad_norm": 0.04582648724317551,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 16077
+    },
+    {
+      "epoch": 1.6502463054187193,
+      "grad_norm": 0.08638078719377518,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 16080
+    },
+    {
+      "epoch": 1.6505541871921183,
+      "grad_norm": 0.053813617676496506,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 16083
+    },
+    {
+      "epoch": 1.6508620689655173,
+      "grad_norm": 0.08186789602041245,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 16086
+    },
+    {
+      "epoch": 1.6511699507389164,
+      "grad_norm": 0.037794895470142365,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 16089
+    },
+    {
+      "epoch": 1.6514778325123154,
+      "grad_norm": 0.1052238717675209,
+      "learning_rate": 0.01,
+      "loss": 2.0614,
+      "step": 16092
+    },
+    {
+      "epoch": 1.6517857142857144,
+      "grad_norm": 0.07596205919981003,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 16095
+    },
+    {
+      "epoch": 1.6520935960591134,
+      "grad_norm": 0.047295670956373215,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 16098
+    },
+    {
+      "epoch": 1.6524014778325125,
+      "grad_norm": 0.05572659894824028,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 16101
+    },
+    {
+      "epoch": 1.6527093596059115,
+      "grad_norm": 0.0429069958627224,
+      "learning_rate": 0.01,
+      "loss": 2.0681,
+      "step": 16104
+    },
+    {
+      "epoch": 1.6530172413793105,
+      "grad_norm": 0.055060967803001404,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 16107
+    },
+    {
+      "epoch": 1.6533251231527095,
+      "grad_norm": 0.05243745073676109,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 16110
+    },
+    {
+      "epoch": 1.6536330049261085,
+      "grad_norm": 0.052228983491659164,
+      "learning_rate": 0.01,
+      "loss": 2.06,
+      "step": 16113
+    },
+    {
+      "epoch": 1.6539408866995073,
+      "grad_norm": 0.065925233066082,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 16116
+    },
+    {
+      "epoch": 1.6542487684729064,
+      "grad_norm": 0.05819106101989746,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 16119
+    },
+    {
+      "epoch": 1.6545566502463054,
+      "grad_norm": 0.04320794716477394,
+      "learning_rate": 0.01,
+      "loss": 2.0691,
+      "step": 16122
+    },
+    {
+      "epoch": 1.6548645320197044,
+      "grad_norm": 0.04202846437692642,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 16125
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.12747296690940857,
+      "learning_rate": 0.01,
+      "loss": 2.0419,
+      "step": 16128
+    },
+    {
+      "epoch": 1.6554802955665024,
+      "grad_norm": 0.07199030369520187,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 16131
+    },
+    {
+      "epoch": 1.6557881773399015,
+      "grad_norm": 0.085335373878479,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 16134
+    },
+    {
+      "epoch": 1.6560960591133005,
+      "grad_norm": 0.061818841844797134,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 16137
+    },
+    {
+      "epoch": 1.6564039408866995,
+      "grad_norm": 0.06255804747343063,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 16140
+    },
+    {
+      "epoch": 1.6567118226600985,
+      "grad_norm": 0.08308485150337219,
+      "learning_rate": 0.01,
+      "loss": 2.0814,
+      "step": 16143
+    },
+    {
+      "epoch": 1.6570197044334976,
+      "grad_norm": 0.06358073651790619,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 16146
+    },
+    {
+      "epoch": 1.6573275862068966,
+      "grad_norm": 0.085427425801754,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 16149
+    },
+    {
+      "epoch": 1.6576354679802956,
+      "grad_norm": 0.043243568390607834,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 16152
+    },
+    {
+      "epoch": 1.6579433497536946,
+      "grad_norm": 0.06593325734138489,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 16155
+    },
+    {
+      "epoch": 1.6582512315270936,
+      "grad_norm": 0.14644569158554077,
+      "learning_rate": 0.01,
+      "loss": 2.0689,
+      "step": 16158
+    },
+    {
+      "epoch": 1.6585591133004927,
+      "grad_norm": 0.1211152896285057,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 16161
+    },
+    {
+      "epoch": 1.6588669950738915,
+      "grad_norm": 0.11020830273628235,
+      "learning_rate": 0.01,
+      "loss": 2.0572,
+      "step": 16164
+    },
+    {
+      "epoch": 1.6591748768472905,
+      "grad_norm": 0.08850467950105667,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 16167
+    },
+    {
+      "epoch": 1.6594827586206895,
+      "grad_norm": 0.050562698394060135,
+      "learning_rate": 0.01,
+      "loss": 2.0542,
+      "step": 16170
+    },
+    {
+      "epoch": 1.6597906403940885,
+      "grad_norm": 0.048076871782541275,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 16173
+    },
+    {
+      "epoch": 1.6600985221674875,
+      "grad_norm": 0.03727034851908684,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 16176
+    },
+    {
+      "epoch": 1.6604064039408866,
+      "grad_norm": 0.048614371567964554,
+      "learning_rate": 0.01,
+      "loss": 2.0552,
+      "step": 16179
+    },
+    {
+      "epoch": 1.6607142857142856,
+      "grad_norm": 0.05649641901254654,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 16182
+    },
+    {
+      "epoch": 1.6610221674876846,
+      "grad_norm": 0.05329003930091858,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 16185
+    },
+    {
+      "epoch": 1.6613300492610836,
+      "grad_norm": 0.06444583833217621,
+      "learning_rate": 0.01,
+      "loss": 2.055,
+      "step": 16188
+    },
+    {
+      "epoch": 1.6616379310344827,
+      "grad_norm": 0.045777902007102966,
+      "learning_rate": 0.01,
+      "loss": 2.0476,
+      "step": 16191
+    },
+    {
+      "epoch": 1.6619458128078817,
+      "grad_norm": 0.04831868037581444,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 16194
+    },
+    {
+      "epoch": 1.6622536945812807,
+      "grad_norm": 0.10648196935653687,
+      "learning_rate": 0.01,
+      "loss": 2.0579,
+      "step": 16197
+    },
+    {
+      "epoch": 1.6625615763546797,
+      "grad_norm": 0.08369257301092148,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 16200
+    },
+    {
+      "epoch": 1.6628694581280787,
+      "grad_norm": 0.13716475665569305,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 16203
+    },
+    {
+      "epoch": 1.6631773399014778,
+      "grad_norm": 0.05025027319788933,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 16206
+    },
+    {
+      "epoch": 1.6634852216748768,
+      "grad_norm": 0.03850054368376732,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 16209
+    },
+    {
+      "epoch": 1.6637931034482758,
+      "grad_norm": 0.046656832098960876,
+      "learning_rate": 0.01,
+      "loss": 2.0595,
+      "step": 16212
+    },
+    {
+      "epoch": 1.6641009852216748,
+      "grad_norm": 0.03826647624373436,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 16215
+    },
+    {
+      "epoch": 1.6644088669950738,
+      "grad_norm": 0.061087023466825485,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 16218
+    },
+    {
+      "epoch": 1.6647167487684729,
+      "grad_norm": 0.03787006065249443,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 16221
+    },
+    {
+      "epoch": 1.6650246305418719,
+      "grad_norm": 0.09619265049695969,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 16224
+    },
+    {
+      "epoch": 1.665332512315271,
+      "grad_norm": 0.04012330621480942,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 16227
+    },
+    {
+      "epoch": 1.66564039408867,
+      "grad_norm": 0.062126293778419495,
+      "learning_rate": 0.01,
+      "loss": 2.0726,
+      "step": 16230
+    },
+    {
+      "epoch": 1.665948275862069,
+      "grad_norm": 0.050277624279260635,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 16233
+    },
+    {
+      "epoch": 1.666256157635468,
+      "grad_norm": 0.03983129933476448,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 16236
+    },
+    {
+      "epoch": 1.666564039408867,
+      "grad_norm": 0.13119915127754211,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 16239
+    },
+    {
+      "epoch": 1.666871921182266,
+      "grad_norm": 0.0525536946952343,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 16242
+    },
+    {
+      "epoch": 1.667179802955665,
+      "grad_norm": 0.056762780994176865,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 16245
+    },
+    {
+      "epoch": 1.667487684729064,
+      "grad_norm": 0.08652041852474213,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 16248
+    },
+    {
+      "epoch": 1.667795566502463,
+      "grad_norm": 0.14455944299697876,
+      "learning_rate": 0.01,
+      "loss": 2.0406,
+      "step": 16251
+    },
+    {
+      "epoch": 1.668103448275862,
+      "grad_norm": 0.03951118513941765,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 16254
+    },
+    {
+      "epoch": 1.6684113300492611,
+      "grad_norm": 0.040585123002529144,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 16257
+    },
+    {
+      "epoch": 1.6687192118226601,
+      "grad_norm": 0.05393810570240021,
+      "learning_rate": 0.01,
+      "loss": 2.0679,
+      "step": 16260
+    },
+    {
+      "epoch": 1.6690270935960592,
+      "grad_norm": 0.050093088299036026,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 16263
+    },
+    {
+      "epoch": 1.6693349753694582,
+      "grad_norm": 0.04196159914135933,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 16266
+    },
+    {
+      "epoch": 1.6696428571428572,
+      "grad_norm": 0.03978092968463898,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 16269
+    },
+    {
+      "epoch": 1.6699507389162562,
+      "grad_norm": 0.05054232105612755,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 16272
+    },
+    {
+      "epoch": 1.6702586206896552,
+      "grad_norm": 0.0746975764632225,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 16275
+    },
+    {
+      "epoch": 1.6705665024630543,
+      "grad_norm": 0.05685516446828842,
+      "learning_rate": 0.01,
+      "loss": 2.0591,
+      "step": 16278
+    },
+    {
+      "epoch": 1.6708743842364533,
+      "grad_norm": 0.031971871852874756,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 16281
+    },
+    {
+      "epoch": 1.6711822660098523,
+      "grad_norm": 0.03947863727807999,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 16284
+    },
+    {
+      "epoch": 1.6714901477832513,
+      "grad_norm": 0.11271070688962936,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 16287
+    },
+    {
+      "epoch": 1.6717980295566504,
+      "grad_norm": 0.05308755114674568,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 16290
+    },
+    {
+      "epoch": 1.6721059113300494,
+      "grad_norm": 0.042826078832149506,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 16293
+    },
+    {
+      "epoch": 1.6724137931034484,
+      "grad_norm": 0.0458630695939064,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 16296
+    },
+    {
+      "epoch": 1.6727216748768474,
+      "grad_norm": 0.05401900038123131,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 16299
+    },
+    {
+      "epoch": 1.6730295566502464,
+      "grad_norm": 0.05380195751786232,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 16302
+    },
+    {
+      "epoch": 1.6733374384236455,
+      "grad_norm": 0.038716450333595276,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 16305
+    },
+    {
+      "epoch": 1.6736453201970445,
+      "grad_norm": 0.04034694656729698,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 16308
+    },
+    {
+      "epoch": 1.6739532019704435,
+      "grad_norm": 0.06753403693437576,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 16311
+    },
+    {
+      "epoch": 1.6742610837438425,
+      "grad_norm": 0.10001173615455627,
+      "learning_rate": 0.01,
+      "loss": 2.0467,
+      "step": 16314
+    },
+    {
+      "epoch": 1.6745689655172413,
+      "grad_norm": 0.04366351664066315,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 16317
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.07137630879878998,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 16320
+    },
+    {
+      "epoch": 1.6751847290640394,
+      "grad_norm": 0.049938492476940155,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 16323
+    },
+    {
+      "epoch": 1.6754926108374384,
+      "grad_norm": 0.03337172046303749,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 16326
+    },
+    {
+      "epoch": 1.6758004926108374,
+      "grad_norm": 0.07407473772764206,
+      "learning_rate": 0.01,
+      "loss": 2.0705,
+      "step": 16329
+    },
+    {
+      "epoch": 1.6761083743842364,
+      "grad_norm": 0.07006946206092834,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 16332
+    },
+    {
+      "epoch": 1.6764162561576355,
+      "grad_norm": 0.05342825874686241,
+      "learning_rate": 0.01,
+      "loss": 2.0481,
+      "step": 16335
+    },
+    {
+      "epoch": 1.6767241379310345,
+      "grad_norm": 0.052405234426259995,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 16338
+    },
+    {
+      "epoch": 1.6770320197044335,
+      "grad_norm": 0.20231324434280396,
+      "learning_rate": 0.01,
+      "loss": 2.0472,
+      "step": 16341
+    },
+    {
+      "epoch": 1.6773399014778325,
+      "grad_norm": 0.07893595844507217,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 16344
+    },
+    {
+      "epoch": 1.6776477832512315,
+      "grad_norm": 0.06872416287660599,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 16347
+    },
+    {
+      "epoch": 1.6779556650246306,
+      "grad_norm": 0.041687123477458954,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 16350
+    },
+    {
+      "epoch": 1.6782635467980296,
+      "grad_norm": 0.04184873029589653,
+      "learning_rate": 0.01,
+      "loss": 2.0769,
+      "step": 16353
+    },
+    {
+      "epoch": 1.6785714285714286,
+      "grad_norm": 0.036598458886146545,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 16356
+    },
+    {
+      "epoch": 1.6788793103448276,
+      "grad_norm": 0.062203384935855865,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 16359
+    },
+    {
+      "epoch": 1.6791871921182266,
+      "grad_norm": 0.04513971135020256,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 16362
+    },
+    {
+      "epoch": 1.6794950738916257,
+      "grad_norm": 0.043875399976968765,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 16365
+    },
+    {
+      "epoch": 1.6798029556650245,
+      "grad_norm": 0.030207300558686256,
+      "learning_rate": 0.01,
+      "loss": 2.0733,
+      "step": 16368
+    },
+    {
+      "epoch": 1.6801108374384235,
+      "grad_norm": 0.07749854028224945,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 16371
+    },
+    {
+      "epoch": 1.6804187192118225,
+      "grad_norm": 0.10269973427057266,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 16374
+    },
+    {
+      "epoch": 1.6807266009852215,
+      "grad_norm": 0.043558500707149506,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 16377
+    },
+    {
+      "epoch": 1.6810344827586206,
+      "grad_norm": 0.0490686409175396,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 16380
+    },
+    {
+      "epoch": 1.6813423645320196,
+      "grad_norm": 0.062107689678668976,
+      "learning_rate": 0.01,
+      "loss": 2.0592,
+      "step": 16383
+    },
+    {
+      "epoch": 1.6816502463054186,
+      "grad_norm": 0.0856776013970375,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 16386
+    },
+    {
+      "epoch": 1.6819581280788176,
+      "grad_norm": 0.11694356054067612,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 16389
+    },
+    {
+      "epoch": 1.6822660098522166,
+      "grad_norm": 0.07279752194881439,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 16392
+    },
+    {
+      "epoch": 1.6825738916256157,
+      "grad_norm": 0.06813056766986847,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 16395
+    },
+    {
+      "epoch": 1.6828817733990147,
+      "grad_norm": 0.045916907489299774,
+      "learning_rate": 0.01,
+      "loss": 2.0714,
+      "step": 16398
+    },
+    {
+      "epoch": 1.6831896551724137,
+      "grad_norm": 0.04464447498321533,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 16401
+    },
+    {
+      "epoch": 1.6834975369458127,
+      "grad_norm": 0.04815223440527916,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 16404
+    },
+    {
+      "epoch": 1.6838054187192117,
+      "grad_norm": 0.06025001034140587,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 16407
+    },
+    {
+      "epoch": 1.6841133004926108,
+      "grad_norm": 0.05691540613770485,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 16410
+    },
+    {
+      "epoch": 1.6844211822660098,
+      "grad_norm": 0.04643694683909416,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 16413
+    },
+    {
+      "epoch": 1.6847290640394088,
+      "grad_norm": 0.03540325164794922,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 16416
+    },
+    {
+      "epoch": 1.6850369458128078,
+      "grad_norm": 0.034472569823265076,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 16419
+    },
+    {
+      "epoch": 1.6853448275862069,
+      "grad_norm": 0.04316902533173561,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 16422
+    },
+    {
+      "epoch": 1.6856527093596059,
+      "grad_norm": 0.04943558946251869,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 16425
+    },
+    {
+      "epoch": 1.685960591133005,
+      "grad_norm": 0.11482315510511398,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 16428
+    },
+    {
+      "epoch": 1.686268472906404,
+      "grad_norm": 0.10594377666711807,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 16431
+    },
+    {
+      "epoch": 1.686576354679803,
+      "grad_norm": 0.09860610961914062,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 16434
+    },
+    {
+      "epoch": 1.686884236453202,
+      "grad_norm": 0.06849053502082825,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 16437
+    },
+    {
+      "epoch": 1.687192118226601,
+      "grad_norm": 0.05089464411139488,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 16440
+    },
+    {
+      "epoch": 1.6875,
+      "grad_norm": 0.04762034863233566,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 16443
+    },
+    {
+      "epoch": 1.687807881773399,
+      "grad_norm": 0.09014497697353363,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 16446
+    },
+    {
+      "epoch": 1.688115763546798,
+      "grad_norm": 0.06832917779684067,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 16449
+    },
+    {
+      "epoch": 1.688423645320197,
+      "grad_norm": 0.0529920794069767,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 16452
+    },
+    {
+      "epoch": 1.688731527093596,
+      "grad_norm": 0.03208652511239052,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 16455
+    },
+    {
+      "epoch": 1.689039408866995,
+      "grad_norm": 0.13702784478664398,
+      "learning_rate": 0.01,
+      "loss": 2.0393,
+      "step": 16458
+    },
+    {
+      "epoch": 1.6893472906403941,
+      "grad_norm": 0.05972970649600029,
+      "learning_rate": 0.01,
+      "loss": 2.0795,
+      "step": 16461
+    },
+    {
+      "epoch": 1.6896551724137931,
+      "grad_norm": 0.043536797165870667,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 16464
+    },
+    {
+      "epoch": 1.6899630541871922,
+      "grad_norm": 0.0556536540389061,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 16467
+    },
+    {
+      "epoch": 1.6902709359605912,
+      "grad_norm": 0.06583042442798615,
+      "learning_rate": 0.01,
+      "loss": 2.0568,
+      "step": 16470
+    },
+    {
+      "epoch": 1.6905788177339902,
+      "grad_norm": 0.0535028837621212,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 16473
+    },
+    {
+      "epoch": 1.6908866995073892,
+      "grad_norm": 0.09974632412195206,
+      "learning_rate": 0.01,
+      "loss": 2.0589,
+      "step": 16476
+    },
+    {
+      "epoch": 1.6911945812807883,
+      "grad_norm": 0.058350350707769394,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 16479
+    },
+    {
+      "epoch": 1.6915024630541873,
+      "grad_norm": 0.10049036890268326,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 16482
+    },
+    {
+      "epoch": 1.6918103448275863,
+      "grad_norm": 0.061119675636291504,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 16485
+    },
+    {
+      "epoch": 1.6921182266009853,
+      "grad_norm": 0.07189033925533295,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 16488
+    },
+    {
+      "epoch": 1.6924261083743843,
+      "grad_norm": 0.08962611109018326,
+      "learning_rate": 0.01,
+      "loss": 2.0586,
+      "step": 16491
+    },
+    {
+      "epoch": 1.6927339901477834,
+      "grad_norm": 0.05600450560450554,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 16494
+    },
+    {
+      "epoch": 1.6930418719211824,
+      "grad_norm": 0.1281098574399948,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 16497
+    },
+    {
+      "epoch": 1.6933497536945814,
+      "grad_norm": 0.036696117371320724,
+      "learning_rate": 0.01,
+      "loss": 2.065,
+      "step": 16500
+    },
+    {
+      "epoch": 1.6936576354679804,
+      "grad_norm": 0.12428770959377289,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 16503
+    },
+    {
+      "epoch": 1.6939655172413794,
+      "grad_norm": 0.07593953609466553,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 16506
+    },
+    {
+      "epoch": 1.6942733990147785,
+      "grad_norm": 0.0686376765370369,
+      "learning_rate": 0.01,
+      "loss": 2.05,
+      "step": 16509
+    },
+    {
+      "epoch": 1.6945812807881775,
+      "grad_norm": 0.044805269688367844,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 16512
+    },
+    {
+      "epoch": 1.6948891625615765,
+      "grad_norm": 0.04698259010910988,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 16515
+    },
+    {
+      "epoch": 1.6951970443349755,
+      "grad_norm": 0.04546966403722763,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 16518
+    },
+    {
+      "epoch": 1.6955049261083743,
+      "grad_norm": 0.07239431142807007,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 16521
+    },
+    {
+      "epoch": 1.6958128078817734,
+      "grad_norm": 0.08790195733308792,
+      "learning_rate": 0.01,
+      "loss": 2.0721,
+      "step": 16524
+    },
+    {
+      "epoch": 1.6961206896551724,
+      "grad_norm": 0.05445432290434837,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 16527
+    },
+    {
+      "epoch": 1.6964285714285714,
+      "grad_norm": 0.048141270875930786,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 16530
+    },
+    {
+      "epoch": 1.6967364532019704,
+      "grad_norm": 0.05230564624071121,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 16533
+    },
+    {
+      "epoch": 1.6970443349753694,
+      "grad_norm": 0.1007009968161583,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 16536
+    },
+    {
+      "epoch": 1.6973522167487685,
+      "grad_norm": 0.03878286853432655,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 16539
+    },
+    {
+      "epoch": 1.6976600985221675,
+      "grad_norm": 0.08503543585538864,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 16542
+    },
+    {
+      "epoch": 1.6979679802955665,
+      "grad_norm": 0.06239473819732666,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 16545
+    },
+    {
+      "epoch": 1.6982758620689655,
+      "grad_norm": 0.06893055140972137,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 16548
+    },
+    {
+      "epoch": 1.6985837438423645,
+      "grad_norm": 0.08434829860925674,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 16551
+    },
+    {
+      "epoch": 1.6988916256157636,
+      "grad_norm": 0.031773362308740616,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 16554
+    },
+    {
+      "epoch": 1.6991995073891626,
+      "grad_norm": 0.11598584800958633,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 16557
+    },
+    {
+      "epoch": 1.6995073891625616,
+      "grad_norm": 0.07008111476898193,
+      "learning_rate": 0.01,
+      "loss": 2.0787,
+      "step": 16560
+    },
+    {
+      "epoch": 1.6998152709359606,
+      "grad_norm": 0.03940622881054878,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 16563
+    },
+    {
+      "epoch": 1.7001231527093597,
+      "grad_norm": 0.05206933617591858,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 16566
+    },
+    {
+      "epoch": 1.7004310344827587,
+      "grad_norm": 0.04568307474255562,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 16569
+    },
+    {
+      "epoch": 1.7007389162561575,
+      "grad_norm": 0.031628433614969254,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 16572
+    },
+    {
+      "epoch": 1.7010467980295565,
+      "grad_norm": 0.05636722221970558,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 16575
+    },
+    {
+      "epoch": 1.7013546798029555,
+      "grad_norm": 0.11134552955627441,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 16578
+    },
+    {
+      "epoch": 1.7016625615763545,
+      "grad_norm": 0.06964823603630066,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 16581
+    },
+    {
+      "epoch": 1.7019704433497536,
+      "grad_norm": 0.041148003190755844,
+      "learning_rate": 0.01,
+      "loss": 2.0693,
+      "step": 16584
+    },
+    {
+      "epoch": 1.7022783251231526,
+      "grad_norm": 0.03673578426241875,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 16587
+    },
+    {
+      "epoch": 1.7025862068965516,
+      "grad_norm": 0.03659043833613396,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 16590
+    },
+    {
+      "epoch": 1.7028940886699506,
+      "grad_norm": 0.03824566677212715,
+      "learning_rate": 0.01,
+      "loss": 2.0587,
+      "step": 16593
+    },
+    {
+      "epoch": 1.7032019704433496,
+      "grad_norm": 0.07334180176258087,
+      "learning_rate": 0.01,
+      "loss": 2.0626,
+      "step": 16596
+    },
+    {
+      "epoch": 1.7035098522167487,
+      "grad_norm": 0.055927857756614685,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 16599
+    },
+    {
+      "epoch": 1.7038177339901477,
+      "grad_norm": 0.07610691338777542,
+      "learning_rate": 0.01,
+      "loss": 2.0419,
+      "step": 16602
+    },
+    {
+      "epoch": 1.7041256157635467,
+      "grad_norm": 0.06405298411846161,
+      "learning_rate": 0.01,
+      "loss": 2.0693,
+      "step": 16605
+    },
+    {
+      "epoch": 1.7044334975369457,
+      "grad_norm": 0.06193486601114273,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 16608
+    },
+    {
+      "epoch": 1.7047413793103448,
+      "grad_norm": 0.12181366235017776,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 16611
+    },
+    {
+      "epoch": 1.7050492610837438,
+      "grad_norm": 0.049060508608818054,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 16614
+    },
+    {
+      "epoch": 1.7053571428571428,
+      "grad_norm": 0.05021090805530548,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 16617
+    },
+    {
+      "epoch": 1.7056650246305418,
+      "grad_norm": 0.045171257108449936,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 16620
+    },
+    {
+      "epoch": 1.7059729064039408,
+      "grad_norm": 0.04944808408617973,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 16623
+    },
+    {
+      "epoch": 1.7062807881773399,
+      "grad_norm": 0.03556932508945465,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 16626
+    },
+    {
+      "epoch": 1.7065886699507389,
+      "grad_norm": 0.10005172342061996,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 16629
+    },
+    {
+      "epoch": 1.706896551724138,
+      "grad_norm": 0.04088572412729263,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 16632
+    },
+    {
+      "epoch": 1.707204433497537,
+      "grad_norm": 0.04937949404120445,
+      "learning_rate": 0.01,
+      "loss": 2.0656,
+      "step": 16635
+    },
+    {
+      "epoch": 1.707512315270936,
+      "grad_norm": 0.07822302728891373,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 16638
+    },
+    {
+      "epoch": 1.707820197044335,
+      "grad_norm": 0.05767158418893814,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 16641
+    },
+    {
+      "epoch": 1.708128078817734,
+      "grad_norm": 0.08512212336063385,
+      "learning_rate": 0.01,
+      "loss": 2.0579,
+      "step": 16644
+    },
+    {
+      "epoch": 1.708435960591133,
+      "grad_norm": 0.06758993119001389,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 16647
+    },
+    {
+      "epoch": 1.708743842364532,
+      "grad_norm": 0.08142005652189255,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 16650
+    },
+    {
+      "epoch": 1.709051724137931,
+      "grad_norm": 0.06357218325138092,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 16653
+    },
+    {
+      "epoch": 1.70935960591133,
+      "grad_norm": 0.10591546446084976,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 16656
+    },
+    {
+      "epoch": 1.709667487684729,
+      "grad_norm": 0.06571496278047562,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 16659
+    },
+    {
+      "epoch": 1.7099753694581281,
+      "grad_norm": 0.049450814723968506,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 16662
+    },
+    {
+      "epoch": 1.7102832512315271,
+      "grad_norm": 0.11850273609161377,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 16665
+    },
+    {
+      "epoch": 1.7105911330049262,
+      "grad_norm": 0.0952281728386879,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 16668
+    },
+    {
+      "epoch": 1.7108990147783252,
+      "grad_norm": 0.09431217610836029,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 16671
+    },
+    {
+      "epoch": 1.7112068965517242,
+      "grad_norm": 0.07080823183059692,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 16674
+    },
+    {
+      "epoch": 1.7115147783251232,
+      "grad_norm": 0.049033813178539276,
+      "learning_rate": 0.01,
+      "loss": 2.0489,
+      "step": 16677
+    },
+    {
+      "epoch": 1.7118226600985222,
+      "grad_norm": 0.04356718435883522,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 16680
+    },
+    {
+      "epoch": 1.7121305418719213,
+      "grad_norm": 0.03276592493057251,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 16683
+    },
+    {
+      "epoch": 1.7124384236453203,
+      "grad_norm": 0.04438839852809906,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 16686
+    },
+    {
+      "epoch": 1.7127463054187193,
+      "grad_norm": 0.07276454567909241,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 16689
+    },
+    {
+      "epoch": 1.7130541871921183,
+      "grad_norm": 0.11324001848697662,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 16692
+    },
+    {
+      "epoch": 1.7133620689655173,
+      "grad_norm": 0.14715081453323364,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 16695
+    },
+    {
+      "epoch": 1.7136699507389164,
+      "grad_norm": 0.07661852240562439,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 16698
+    },
+    {
+      "epoch": 1.7139778325123154,
+      "grad_norm": 0.05308947339653969,
+      "learning_rate": 0.01,
+      "loss": 2.0601,
+      "step": 16701
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.06816977262496948,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 16704
+    },
+    {
+      "epoch": 1.7145935960591134,
+      "grad_norm": 0.05123249441385269,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 16707
+    },
+    {
+      "epoch": 1.7149014778325125,
+      "grad_norm": 0.05118009075522423,
+      "learning_rate": 0.01,
+      "loss": 2.0659,
+      "step": 16710
+    },
+    {
+      "epoch": 1.7152093596059115,
+      "grad_norm": 0.03276235982775688,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 16713
+    },
+    {
+      "epoch": 1.7155172413793105,
+      "grad_norm": 0.049824655055999756,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 16716
+    },
+    {
+      "epoch": 1.7158251231527095,
+      "grad_norm": 0.1416471302509308,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 16719
+    },
+    {
+      "epoch": 1.7161330049261085,
+      "grad_norm": 0.04109251871705055,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 16722
+    },
+    {
+      "epoch": 1.7164408866995073,
+      "grad_norm": 0.08853971213102341,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 16725
+    },
+    {
+      "epoch": 1.7167487684729064,
+      "grad_norm": 0.05180136114358902,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 16728
+    },
+    {
+      "epoch": 1.7170566502463054,
+      "grad_norm": 0.0667758584022522,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 16731
+    },
+    {
+      "epoch": 1.7173645320197044,
+      "grad_norm": 0.039203155785799026,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 16734
+    },
+    {
+      "epoch": 1.7176724137931034,
+      "grad_norm": 0.05210564285516739,
+      "learning_rate": 0.01,
+      "loss": 2.0666,
+      "step": 16737
+    },
+    {
+      "epoch": 1.7179802955665024,
+      "grad_norm": 0.0668390691280365,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 16740
+    },
+    {
+      "epoch": 1.7182881773399015,
+      "grad_norm": 0.05041831359267235,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 16743
+    },
+    {
+      "epoch": 1.7185960591133005,
+      "grad_norm": 0.04496284946799278,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 16746
+    },
+    {
+      "epoch": 1.7189039408866995,
+      "grad_norm": 0.08660906553268433,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 16749
+    },
+    {
+      "epoch": 1.7192118226600985,
+      "grad_norm": 0.054843079298734665,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 16752
+    },
+    {
+      "epoch": 1.7195197044334976,
+      "grad_norm": 0.05377354100346565,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 16755
+    },
+    {
+      "epoch": 1.7198275862068966,
+      "grad_norm": 0.060547634959220886,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 16758
+    },
+    {
+      "epoch": 1.7201354679802956,
+      "grad_norm": 0.06253104656934738,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 16761
+    },
+    {
+      "epoch": 1.7204433497536946,
+      "grad_norm": 0.052225805819034576,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 16764
+    },
+    {
+      "epoch": 1.7207512315270936,
+      "grad_norm": 0.04988449066877365,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 16767
+    },
+    {
+      "epoch": 1.7210591133004927,
+      "grad_norm": 0.049726665019989014,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 16770
+    },
+    {
+      "epoch": 1.7213669950738915,
+      "grad_norm": 0.05053321272134781,
+      "learning_rate": 0.01,
+      "loss": 2.0518,
+      "step": 16773
+    },
+    {
+      "epoch": 1.7216748768472905,
+      "grad_norm": 0.060702208429574966,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 16776
+    },
+    {
+      "epoch": 1.7219827586206895,
+      "grad_norm": 0.06695824861526489,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 16779
+    },
+    {
+      "epoch": 1.7222906403940885,
+      "grad_norm": 0.07498425990343094,
+      "learning_rate": 0.01,
+      "loss": 2.0652,
+      "step": 16782
+    },
+    {
+      "epoch": 1.7225985221674875,
+      "grad_norm": 0.04544251784682274,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 16785
+    },
+    {
+      "epoch": 1.7229064039408866,
+      "grad_norm": 0.09211461246013641,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 16788
+    },
+    {
+      "epoch": 1.7232142857142856,
+      "grad_norm": 0.19020068645477295,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 16791
+    },
+    {
+      "epoch": 1.7235221674876846,
+      "grad_norm": 0.08979224413633347,
+      "learning_rate": 0.01,
+      "loss": 2.052,
+      "step": 16794
+    },
+    {
+      "epoch": 1.7238300492610836,
+      "grad_norm": 0.06724744290113449,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 16797
+    },
+    {
+      "epoch": 1.7241379310344827,
+      "grad_norm": 0.05009309947490692,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 16800
+    },
+    {
+      "epoch": 1.7244458128078817,
+      "grad_norm": 0.04953375831246376,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 16803
+    },
+    {
+      "epoch": 1.7247536945812807,
+      "grad_norm": 0.12439639121294022,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 16806
+    },
+    {
+      "epoch": 1.7250615763546797,
+      "grad_norm": 0.126046821475029,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 16809
+    },
+    {
+      "epoch": 1.7253694581280787,
+      "grad_norm": 0.06693071871995926,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 16812
+    },
+    {
+      "epoch": 1.7256773399014778,
+      "grad_norm": 0.04883793368935585,
+      "learning_rate": 0.01,
+      "loss": 2.0768,
+      "step": 16815
+    },
+    {
+      "epoch": 1.7259852216748768,
+      "grad_norm": 0.06895219534635544,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 16818
+    },
+    {
+      "epoch": 1.7262931034482758,
+      "grad_norm": 0.06979874521493912,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 16821
+    },
+    {
+      "epoch": 1.7266009852216748,
+      "grad_norm": 0.06453370302915573,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 16824
+    },
+    {
+      "epoch": 1.7269088669950738,
+      "grad_norm": 0.11405997723340988,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 16827
+    },
+    {
+      "epoch": 1.7272167487684729,
+      "grad_norm": 0.10612863302230835,
+      "learning_rate": 0.01,
+      "loss": 2.0825,
+      "step": 16830
+    },
+    {
+      "epoch": 1.7275246305418719,
+      "grad_norm": 0.09805281460285187,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 16833
+    },
+    {
+      "epoch": 1.727832512315271,
+      "grad_norm": 0.08369138836860657,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 16836
+    },
+    {
+      "epoch": 1.72814039408867,
+      "grad_norm": 0.08893299847841263,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 16839
+    },
+    {
+      "epoch": 1.728448275862069,
+      "grad_norm": 0.06902442872524261,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 16842
+    },
+    {
+      "epoch": 1.728756157635468,
+      "grad_norm": 0.058703918009996414,
+      "learning_rate": 0.01,
+      "loss": 2.0406,
+      "step": 16845
+    },
+    {
+      "epoch": 1.729064039408867,
+      "grad_norm": 0.03979343920946121,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 16848
+    },
+    {
+      "epoch": 1.729371921182266,
+      "grad_norm": 0.08049800246953964,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 16851
+    },
+    {
+      "epoch": 1.729679802955665,
+      "grad_norm": 0.07980278134346008,
+      "learning_rate": 0.01,
+      "loss": 2.0471,
+      "step": 16854
+    },
+    {
+      "epoch": 1.729987684729064,
+      "grad_norm": 0.051650844514369965,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 16857
+    },
+    {
+      "epoch": 1.730295566502463,
+      "grad_norm": 0.06828897446393967,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 16860
+    },
+    {
+      "epoch": 1.730603448275862,
+      "grad_norm": 0.10987185686826706,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 16863
+    },
+    {
+      "epoch": 1.7309113300492611,
+      "grad_norm": 0.055677276104688644,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 16866
+    },
+    {
+      "epoch": 1.7312192118226601,
+      "grad_norm": 0.056478437036275864,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 16869
+    },
+    {
+      "epoch": 1.7315270935960592,
+      "grad_norm": 0.1247091144323349,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 16872
+    },
+    {
+      "epoch": 1.7318349753694582,
+      "grad_norm": 0.06603918969631195,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 16875
+    },
+    {
+      "epoch": 1.7321428571428572,
+      "grad_norm": 0.04174170270562172,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 16878
+    },
+    {
+      "epoch": 1.7324507389162562,
+      "grad_norm": 0.03841250762343407,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 16881
+    },
+    {
+      "epoch": 1.7327586206896552,
+      "grad_norm": 0.03429241105914116,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 16884
+    },
+    {
+      "epoch": 1.7330665024630543,
+      "grad_norm": 0.05175672471523285,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 16887
+    },
+    {
+      "epoch": 1.7333743842364533,
+      "grad_norm": 0.06318958103656769,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 16890
+    },
+    {
+      "epoch": 1.7336822660098523,
+      "grad_norm": 0.08888188004493713,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 16893
+    },
+    {
+      "epoch": 1.7339901477832513,
+      "grad_norm": 0.04479747265577316,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 16896
+    },
+    {
+      "epoch": 1.7342980295566504,
+      "grad_norm": 0.05286455899477005,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 16899
+    },
+    {
+      "epoch": 1.7346059113300494,
+      "grad_norm": 0.04719952121376991,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 16902
+    },
+    {
+      "epoch": 1.7349137931034484,
+      "grad_norm": 0.06204066798090935,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 16905
+    },
+    {
+      "epoch": 1.7352216748768474,
+      "grad_norm": 0.1099078357219696,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 16908
+    },
+    {
+      "epoch": 1.7355295566502464,
+      "grad_norm": 0.08311746269464493,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 16911
+    },
+    {
+      "epoch": 1.7358374384236455,
+      "grad_norm": 0.09649046510457993,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 16914
+    },
+    {
+      "epoch": 1.7361453201970445,
+      "grad_norm": 0.06716254353523254,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 16917
+    },
+    {
+      "epoch": 1.7364532019704435,
+      "grad_norm": 0.05400918424129486,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 16920
+    },
+    {
+      "epoch": 1.7367610837438425,
+      "grad_norm": 0.06857965141534805,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 16923
+    },
+    {
+      "epoch": 1.7370689655172413,
+      "grad_norm": 0.05842095986008644,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 16926
+    },
+    {
+      "epoch": 1.7373768472906403,
+      "grad_norm": 0.06279005855321884,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 16929
+    },
+    {
+      "epoch": 1.7376847290640394,
+      "grad_norm": 0.04209805652499199,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 16932
+    },
+    {
+      "epoch": 1.7379926108374384,
+      "grad_norm": 0.07557601481676102,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 16935
+    },
+    {
+      "epoch": 1.7383004926108374,
+      "grad_norm": 0.05998634174466133,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 16938
+    },
+    {
+      "epoch": 1.7386083743842364,
+      "grad_norm": 0.05477887764573097,
+      "learning_rate": 0.01,
+      "loss": 2.0659,
+      "step": 16941
+    },
+    {
+      "epoch": 1.7389162561576355,
+      "grad_norm": 0.09489591419696808,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 16944
+    },
+    {
+      "epoch": 1.7392241379310345,
+      "grad_norm": 0.13861320912837982,
+      "learning_rate": 0.01,
+      "loss": 2.0778,
+      "step": 16947
+    },
+    {
+      "epoch": 1.7395320197044335,
+      "grad_norm": 0.1302078664302826,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 16950
+    },
+    {
+      "epoch": 1.7398399014778325,
+      "grad_norm": 0.08335380256175995,
+      "learning_rate": 0.01,
+      "loss": 2.0612,
+      "step": 16953
+    },
+    {
+      "epoch": 1.7401477832512315,
+      "grad_norm": 0.06901963800191879,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 16956
+    },
+    {
+      "epoch": 1.7404556650246306,
+      "grad_norm": 0.06619597226381302,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 16959
+    },
+    {
+      "epoch": 1.7407635467980296,
+      "grad_norm": 0.045459166169166565,
+      "learning_rate": 0.01,
+      "loss": 2.071,
+      "step": 16962
+    },
+    {
+      "epoch": 1.7410714285714286,
+      "grad_norm": 0.06302090734243393,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 16965
+    },
+    {
+      "epoch": 1.7413793103448276,
+      "grad_norm": 0.07184530049562454,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 16968
+    },
+    {
+      "epoch": 1.7416871921182266,
+      "grad_norm": 0.07319154590368271,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 16971
+    },
+    {
+      "epoch": 1.7419950738916257,
+      "grad_norm": 0.0667872503399849,
+      "learning_rate": 0.01,
+      "loss": 2.0406,
+      "step": 16974
+    },
+    {
+      "epoch": 1.7423029556650245,
+      "grad_norm": 0.0988057404756546,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 16977
+    },
+    {
+      "epoch": 1.7426108374384235,
+      "grad_norm": 0.09142420440912247,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 16980
+    },
+    {
+      "epoch": 1.7429187192118225,
+      "grad_norm": 0.07581423968076706,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 16983
+    },
+    {
+      "epoch": 1.7432266009852215,
+      "grad_norm": 0.05793863534927368,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 16986
+    },
+    {
+      "epoch": 1.7435344827586206,
+      "grad_norm": 0.0717446580529213,
+      "learning_rate": 0.01,
+      "loss": 2.061,
+      "step": 16989
+    },
+    {
+      "epoch": 1.7438423645320196,
+      "grad_norm": 0.03572774678468704,
+      "learning_rate": 0.01,
+      "loss": 2.0663,
+      "step": 16992
+    },
+    {
+      "epoch": 1.7441502463054186,
+      "grad_norm": 0.09349701553583145,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 16995
+    },
+    {
+      "epoch": 1.7444581280788176,
+      "grad_norm": 0.08374779671430588,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 16998
+    },
+    {
+      "epoch": 1.7447660098522166,
+      "grad_norm": 0.045029062777757645,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 17001
+    },
+    {
+      "epoch": 1.7450738916256157,
+      "grad_norm": 0.06359710544347763,
+      "learning_rate": 0.01,
+      "loss": 2.073,
+      "step": 17004
+    },
+    {
+      "epoch": 1.7453817733990147,
+      "grad_norm": 0.0948299989104271,
+      "learning_rate": 0.01,
+      "loss": 2.0326,
+      "step": 17007
+    },
+    {
+      "epoch": 1.7456896551724137,
+      "grad_norm": 0.05561600998044014,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 17010
+    },
+    {
+      "epoch": 1.7459975369458127,
+      "grad_norm": 0.05672406032681465,
+      "learning_rate": 0.01,
+      "loss": 2.0591,
+      "step": 17013
+    },
+    {
+      "epoch": 1.7463054187192117,
+      "grad_norm": 0.047100841999053955,
+      "learning_rate": 0.01,
+      "loss": 2.0597,
+      "step": 17016
+    },
+    {
+      "epoch": 1.7466133004926108,
+      "grad_norm": 0.09477028995752335,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 17019
+    },
+    {
+      "epoch": 1.7469211822660098,
+      "grad_norm": 0.07235399633646011,
+      "learning_rate": 0.01,
+      "loss": 2.0778,
+      "step": 17022
+    },
+    {
+      "epoch": 1.7472290640394088,
+      "grad_norm": 0.08015649020671844,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 17025
+    },
+    {
+      "epoch": 1.7475369458128078,
+      "grad_norm": 0.07459922134876251,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 17028
+    },
+    {
+      "epoch": 1.7478448275862069,
+      "grad_norm": 0.10745642334222794,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 17031
+    },
+    {
+      "epoch": 1.7481527093596059,
+      "grad_norm": 0.049479395151138306,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 17034
+    },
+    {
+      "epoch": 1.748460591133005,
+      "grad_norm": 0.03935602307319641,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 17037
+    },
+    {
+      "epoch": 1.748768472906404,
+      "grad_norm": 0.05755804106593132,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 17040
+    },
+    {
+      "epoch": 1.749076354679803,
+      "grad_norm": 0.07461614906787872,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 17043
+    },
+    {
+      "epoch": 1.749384236453202,
+      "grad_norm": 0.07078621536493301,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 17046
+    },
+    {
+      "epoch": 1.749692118226601,
+      "grad_norm": 0.12035417556762695,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 17049
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 0.054343827068805695,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 17052
+    },
+    {
+      "epoch": 1.750307881773399,
+      "grad_norm": 0.056529540568590164,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 17055
+    },
+    {
+      "epoch": 1.750615763546798,
+      "grad_norm": 0.09392616152763367,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 17058
+    },
+    {
+      "epoch": 1.750923645320197,
+      "grad_norm": 0.0874391570687294,
+      "learning_rate": 0.01,
+      "loss": 2.0733,
+      "step": 17061
+    },
+    {
+      "epoch": 1.751231527093596,
+      "grad_norm": 0.03889552876353264,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 17064
+    },
+    {
+      "epoch": 1.751539408866995,
+      "grad_norm": 0.06299902498722076,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 17067
+    },
+    {
+      "epoch": 1.7518472906403941,
+      "grad_norm": 0.05655315890908241,
+      "learning_rate": 0.01,
+      "loss": 2.0571,
+      "step": 17070
+    },
+    {
+      "epoch": 1.7521551724137931,
+      "grad_norm": 0.04646646976470947,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 17073
+    },
+    {
+      "epoch": 1.7524630541871922,
+      "grad_norm": 0.04910219460725784,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 17076
+    },
+    {
+      "epoch": 1.7527709359605912,
+      "grad_norm": 0.03616022691130638,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 17079
+    },
+    {
+      "epoch": 1.7530788177339902,
+      "grad_norm": 0.05539899319410324,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 17082
+    },
+    {
+      "epoch": 1.7533866995073892,
+      "grad_norm": 0.09409206360578537,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 17085
+    },
+    {
+      "epoch": 1.7536945812807883,
+      "grad_norm": 0.048882581293582916,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 17088
+    },
+    {
+      "epoch": 1.7540024630541873,
+      "grad_norm": 0.09687826037406921,
+      "learning_rate": 0.01,
+      "loss": 2.0472,
+      "step": 17091
+    },
+    {
+      "epoch": 1.7543103448275863,
+      "grad_norm": 0.06646592915058136,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 17094
+    },
+    {
+      "epoch": 1.7546182266009853,
+      "grad_norm": 0.07316572964191437,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 17097
+    },
+    {
+      "epoch": 1.7549261083743843,
+      "grad_norm": 0.058037593960762024,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 17100
+    },
+    {
+      "epoch": 1.7552339901477834,
+      "grad_norm": 0.05486461892724037,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 17103
+    },
+    {
+      "epoch": 1.7555418719211824,
+      "grad_norm": 0.0397610180079937,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 17106
+    },
+    {
+      "epoch": 1.7558497536945814,
+      "grad_norm": 0.11639061570167542,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 17109
+    },
+    {
+      "epoch": 1.7561576354679804,
+      "grad_norm": 0.04207361862063408,
+      "learning_rate": 0.01,
+      "loss": 2.064,
+      "step": 17112
+    },
+    {
+      "epoch": 1.7564655172413794,
+      "grad_norm": 0.09336013346910477,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 17115
+    },
+    {
+      "epoch": 1.7567733990147785,
+      "grad_norm": 0.059693820774555206,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 17118
+    },
+    {
+      "epoch": 1.7570812807881775,
+      "grad_norm": 0.05269778147339821,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 17121
+    },
+    {
+      "epoch": 1.7573891625615765,
+      "grad_norm": 0.0548628568649292,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 17124
+    },
+    {
+      "epoch": 1.7576970443349755,
+      "grad_norm": 0.049783483147621155,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 17127
+    },
+    {
+      "epoch": 1.7580049261083743,
+      "grad_norm": 0.11240525543689728,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 17130
+    },
+    {
+      "epoch": 1.7583128078817734,
+      "grad_norm": 0.04133368283510208,
+      "learning_rate": 0.01,
+      "loss": 2.064,
+      "step": 17133
+    },
+    {
+      "epoch": 1.7586206896551724,
+      "grad_norm": 0.042926132678985596,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 17136
+    },
+    {
+      "epoch": 1.7589285714285714,
+      "grad_norm": 0.053613241761922836,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 17139
+    },
+    {
+      "epoch": 1.7592364532019704,
+      "grad_norm": 0.03950737044215202,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 17142
+    },
+    {
+      "epoch": 1.7595443349753694,
+      "grad_norm": 0.045378413051366806,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 17145
+    },
+    {
+      "epoch": 1.7598522167487685,
+      "grad_norm": 0.036304816603660583,
+      "learning_rate": 0.01,
+      "loss": 2.0789,
+      "step": 17148
+    },
+    {
+      "epoch": 1.7601600985221675,
+      "grad_norm": 0.03886290267109871,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 17151
+    },
+    {
+      "epoch": 1.7604679802955665,
+      "grad_norm": 0.0885484591126442,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 17154
+    },
+    {
+      "epoch": 1.7607758620689655,
+      "grad_norm": 0.06733599305152893,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 17157
+    },
+    {
+      "epoch": 1.7610837438423645,
+      "grad_norm": 0.10319662094116211,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 17160
+    },
+    {
+      "epoch": 1.7613916256157636,
+      "grad_norm": 0.047492869198322296,
+      "learning_rate": 0.01,
+      "loss": 2.0726,
+      "step": 17163
+    },
+    {
+      "epoch": 1.7616995073891626,
+      "grad_norm": 0.04345547780394554,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 17166
+    },
+    {
+      "epoch": 1.7620073891625616,
+      "grad_norm": 0.0452197827398777,
+      "learning_rate": 0.01,
+      "loss": 2.0393,
+      "step": 17169
+    },
+    {
+      "epoch": 1.7623152709359606,
+      "grad_norm": 0.0703844428062439,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 17172
+    },
+    {
+      "epoch": 1.7626231527093597,
+      "grad_norm": 0.07864879071712494,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 17175
+    },
+    {
+      "epoch": 1.7629310344827587,
+      "grad_norm": 0.1282995641231537,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 17178
+    },
+    {
+      "epoch": 1.7632389162561575,
+      "grad_norm": 0.0837298184633255,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 17181
+    },
+    {
+      "epoch": 1.7635467980295565,
+      "grad_norm": 0.05081562697887421,
+      "learning_rate": 0.01,
+      "loss": 2.0656,
+      "step": 17184
+    },
+    {
+      "epoch": 1.7638546798029555,
+      "grad_norm": 0.07952243834733963,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 17187
+    },
+    {
+      "epoch": 1.7641625615763545,
+      "grad_norm": 0.06592147797346115,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 17190
+    },
+    {
+      "epoch": 1.7644704433497536,
+      "grad_norm": 0.04341195523738861,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 17193
+    },
+    {
+      "epoch": 1.7647783251231526,
+      "grad_norm": 0.04649266228079796,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 17196
+    },
+    {
+      "epoch": 1.7650862068965516,
+      "grad_norm": 0.04569242149591446,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 17199
+    },
+    {
+      "epoch": 1.7653940886699506,
+      "grad_norm": 0.040291350334882736,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 17202
+    },
+    {
+      "epoch": 1.7657019704433496,
+      "grad_norm": 0.05328141525387764,
+      "learning_rate": 0.01,
+      "loss": 2.0384,
+      "step": 17205
+    },
+    {
+      "epoch": 1.7660098522167487,
+      "grad_norm": 0.04405885562300682,
+      "learning_rate": 0.01,
+      "loss": 2.0566,
+      "step": 17208
+    },
+    {
+      "epoch": 1.7663177339901477,
+      "grad_norm": 0.06635614484548569,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 17211
+    },
+    {
+      "epoch": 1.7666256157635467,
+      "grad_norm": 0.09231774508953094,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 17214
+    },
+    {
+      "epoch": 1.7669334975369457,
+      "grad_norm": 0.056320998817682266,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 17217
+    },
+    {
+      "epoch": 1.7672413793103448,
+      "grad_norm": 0.049784105271101,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 17220
+    },
+    {
+      "epoch": 1.7675492610837438,
+      "grad_norm": 0.03728071227669716,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 17223
+    },
+    {
+      "epoch": 1.7678571428571428,
+      "grad_norm": 0.06607525050640106,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 17226
+    },
+    {
+      "epoch": 1.7681650246305418,
+      "grad_norm": 0.07367686927318573,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 17229
+    },
+    {
+      "epoch": 1.7684729064039408,
+      "grad_norm": 0.039499782025814056,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 17232
+    },
+    {
+      "epoch": 1.7687807881773399,
+      "grad_norm": 0.04863186180591583,
+      "learning_rate": 0.01,
+      "loss": 2.0595,
+      "step": 17235
+    },
+    {
+      "epoch": 1.7690886699507389,
+      "grad_norm": 0.03877348452806473,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 17238
+    },
+    {
+      "epoch": 1.769396551724138,
+      "grad_norm": 0.049965135753154755,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 17241
+    },
+    {
+      "epoch": 1.769704433497537,
+      "grad_norm": 0.0697547048330307,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 17244
+    },
+    {
+      "epoch": 1.770012315270936,
+      "grad_norm": 0.0562531016767025,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 17247
+    },
+    {
+      "epoch": 1.770320197044335,
+      "grad_norm": 0.12931805849075317,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 17250
+    },
+    {
+      "epoch": 1.770628078817734,
+      "grad_norm": 0.06590058654546738,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 17253
+    },
+    {
+      "epoch": 1.770935960591133,
+      "grad_norm": 0.045246463268995285,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 17256
+    },
+    {
+      "epoch": 1.771243842364532,
+      "grad_norm": 0.03972258046269417,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 17259
+    },
+    {
+      "epoch": 1.771551724137931,
+      "grad_norm": 0.030682874843478203,
+      "learning_rate": 0.01,
+      "loss": 2.0665,
+      "step": 17262
+    },
+    {
+      "epoch": 1.77185960591133,
+      "grad_norm": 0.08989464491605759,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 17265
+    },
+    {
+      "epoch": 1.772167487684729,
+      "grad_norm": 0.05595966801047325,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 17268
+    },
+    {
+      "epoch": 1.7724753694581281,
+      "grad_norm": 0.16923703253269196,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 17271
+    },
+    {
+      "epoch": 1.7727832512315271,
+      "grad_norm": 0.08722022920846939,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 17274
+    },
+    {
+      "epoch": 1.7730911330049262,
+      "grad_norm": 0.0741046667098999,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 17277
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.06061973422765732,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 17280
+    },
+    {
+      "epoch": 1.7737068965517242,
+      "grad_norm": 0.036843594163656235,
+      "learning_rate": 0.01,
+      "loss": 2.056,
+      "step": 17283
+    },
+    {
+      "epoch": 1.7740147783251232,
+      "grad_norm": 0.03937767818570137,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 17286
+    },
+    {
+      "epoch": 1.7743226600985222,
+      "grad_norm": 0.03801162540912628,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 17289
+    },
+    {
+      "epoch": 1.7746305418719213,
+      "grad_norm": 0.045572392642498016,
+      "learning_rate": 0.01,
+      "loss": 2.0665,
+      "step": 17292
+    },
+    {
+      "epoch": 1.7749384236453203,
+      "grad_norm": 0.06430240720510483,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 17295
+    },
+    {
+      "epoch": 1.7752463054187193,
+      "grad_norm": 0.09266401827335358,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 17298
+    },
+    {
+      "epoch": 1.7755541871921183,
+      "grad_norm": 0.09686179459095001,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 17301
+    },
+    {
+      "epoch": 1.7758620689655173,
+      "grad_norm": 0.04640132188796997,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 17304
+    },
+    {
+      "epoch": 1.7761699507389164,
+      "grad_norm": 0.03891894221305847,
+      "learning_rate": 0.01,
+      "loss": 2.0529,
+      "step": 17307
+    },
+    {
+      "epoch": 1.7764778325123154,
+      "grad_norm": 0.06023077294230461,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 17310
+    },
+    {
+      "epoch": 1.7767857142857144,
+      "grad_norm": 0.12215135246515274,
+      "learning_rate": 0.01,
+      "loss": 2.0472,
+      "step": 17313
+    },
+    {
+      "epoch": 1.7770935960591134,
+      "grad_norm": 0.04197768121957779,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 17316
+    },
+    {
+      "epoch": 1.7774014778325125,
+      "grad_norm": 0.0429445244371891,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 17319
+    },
+    {
+      "epoch": 1.7777093596059115,
+      "grad_norm": 0.04674970358610153,
+      "learning_rate": 0.01,
+      "loss": 2.0493,
+      "step": 17322
+    },
+    {
+      "epoch": 1.7780172413793105,
+      "grad_norm": 0.11712675541639328,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 17325
+    },
+    {
+      "epoch": 1.7783251231527095,
+      "grad_norm": 0.04812907800078392,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 17328
+    },
+    {
+      "epoch": 1.7786330049261085,
+      "grad_norm": 0.04147825017571449,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 17331
+    },
+    {
+      "epoch": 1.7789408866995073,
+      "grad_norm": 0.07262876629829407,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 17334
+    },
+    {
+      "epoch": 1.7792487684729064,
+      "grad_norm": 0.08528011292219162,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 17337
+    },
+    {
+      "epoch": 1.7795566502463054,
+      "grad_norm": 0.046615902334451675,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 17340
+    },
+    {
+      "epoch": 1.7798645320197044,
+      "grad_norm": 0.06018273904919624,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 17343
+    },
+    {
+      "epoch": 1.7801724137931034,
+      "grad_norm": 0.07272887974977493,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 17346
+    },
+    {
+      "epoch": 1.7804802955665024,
+      "grad_norm": 0.07152794301509857,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 17349
+    },
+    {
+      "epoch": 1.7807881773399015,
+      "grad_norm": 0.07950329035520554,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 17352
+    },
+    {
+      "epoch": 1.7810960591133005,
+      "grad_norm": 0.040778059512376785,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 17355
+    },
+    {
+      "epoch": 1.7814039408866995,
+      "grad_norm": 0.06180460751056671,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 17358
+    },
+    {
+      "epoch": 1.7817118226600985,
+      "grad_norm": 0.06950334459543228,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 17361
+    },
+    {
+      "epoch": 1.7820197044334976,
+      "grad_norm": 0.037724483758211136,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 17364
+    },
+    {
+      "epoch": 1.7823275862068966,
+      "grad_norm": 0.05991238355636597,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 17367
+    },
+    {
+      "epoch": 1.7826354679802956,
+      "grad_norm": 0.047278665006160736,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 17370
+    },
+    {
+      "epoch": 1.7829433497536946,
+      "grad_norm": 0.05376293137669563,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 17373
+    },
+    {
+      "epoch": 1.7832512315270936,
+      "grad_norm": 0.04049403965473175,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 17376
+    },
+    {
+      "epoch": 1.7835591133004927,
+      "grad_norm": 0.04954640567302704,
+      "learning_rate": 0.01,
+      "loss": 2.0393,
+      "step": 17379
+    },
+    {
+      "epoch": 1.7838669950738915,
+      "grad_norm": 0.049089133739471436,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 17382
+    },
+    {
+      "epoch": 1.7841748768472905,
+      "grad_norm": 0.0531185045838356,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 17385
+    },
+    {
+      "epoch": 1.7844827586206895,
+      "grad_norm": 0.060973040759563446,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 17388
+    },
+    {
+      "epoch": 1.7847906403940885,
+      "grad_norm": 0.044274650514125824,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 17391
+    },
+    {
+      "epoch": 1.7850985221674875,
+      "grad_norm": 0.08154580742120743,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 17394
+    },
+    {
+      "epoch": 1.7854064039408866,
+      "grad_norm": 0.05253531411290169,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 17397
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 0.056620582938194275,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 17400
+    },
+    {
+      "epoch": 1.7860221674876846,
+      "grad_norm": 0.069371297955513,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 17403
+    },
+    {
+      "epoch": 1.7863300492610836,
+      "grad_norm": 0.04726189747452736,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 17406
+    },
+    {
+      "epoch": 1.7866379310344827,
+      "grad_norm": 0.11150949448347092,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 17409
+    },
+    {
+      "epoch": 1.7869458128078817,
+      "grad_norm": 0.07482532411813736,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 17412
+    },
+    {
+      "epoch": 1.7872536945812807,
+      "grad_norm": 0.03803645819425583,
+      "learning_rate": 0.01,
+      "loss": 2.0555,
+      "step": 17415
+    },
+    {
+      "epoch": 1.7875615763546797,
+      "grad_norm": 0.08635829389095306,
+      "learning_rate": 0.01,
+      "loss": 2.0551,
+      "step": 17418
+    },
+    {
+      "epoch": 1.7878694581280787,
+      "grad_norm": 0.08558929711580276,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 17421
+    },
+    {
+      "epoch": 1.7881773399014778,
+      "grad_norm": 0.051051054149866104,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 17424
+    },
+    {
+      "epoch": 1.7884852216748768,
+      "grad_norm": 0.0584864616394043,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 17427
+    },
+    {
+      "epoch": 1.7887931034482758,
+      "grad_norm": 0.04015490040183067,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 17430
+    },
+    {
+      "epoch": 1.7891009852216748,
+      "grad_norm": 0.0499749630689621,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 17433
+    },
+    {
+      "epoch": 1.7894088669950738,
+      "grad_norm": 0.08796360343694687,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 17436
+    },
+    {
+      "epoch": 1.7897167487684729,
+      "grad_norm": 0.08200754970312119,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 17439
+    },
+    {
+      "epoch": 1.7900246305418719,
+      "grad_norm": 0.09300393611192703,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 17442
+    },
+    {
+      "epoch": 1.790332512315271,
+      "grad_norm": 0.08223576098680496,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 17445
+    },
+    {
+      "epoch": 1.79064039408867,
+      "grad_norm": 0.05235210806131363,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 17448
+    },
+    {
+      "epoch": 1.790948275862069,
+      "grad_norm": 0.047677502036094666,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 17451
+    },
+    {
+      "epoch": 1.791256157635468,
+      "grad_norm": 0.044341955333948135,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 17454
+    },
+    {
+      "epoch": 1.791564039408867,
+      "grad_norm": 0.09555595368146896,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 17457
+    },
+    {
+      "epoch": 1.791871921182266,
+      "grad_norm": 0.05652477219700813,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 17460
+    },
+    {
+      "epoch": 1.792179802955665,
+      "grad_norm": 0.0979117676615715,
+      "learning_rate": 0.01,
+      "loss": 2.0747,
+      "step": 17463
+    },
+    {
+      "epoch": 1.792487684729064,
+      "grad_norm": 0.0674947127699852,
+      "learning_rate": 0.01,
+      "loss": 2.0723,
+      "step": 17466
+    },
+    {
+      "epoch": 1.792795566502463,
+      "grad_norm": 0.05617907643318176,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 17469
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.10979234427213669,
+      "learning_rate": 0.01,
+      "loss": 2.0638,
+      "step": 17472
+    },
+    {
+      "epoch": 1.7934113300492611,
+      "grad_norm": 0.056006476283073425,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 17475
+    },
+    {
+      "epoch": 1.7937192118226601,
+      "grad_norm": 0.10030517727136612,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 17478
+    },
+    {
+      "epoch": 1.7940270935960592,
+      "grad_norm": 0.042350657284259796,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 17481
+    },
+    {
+      "epoch": 1.7943349753694582,
+      "grad_norm": 0.03725098446011543,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 17484
+    },
+    {
+      "epoch": 1.7946428571428572,
+      "grad_norm": 0.09215757250785828,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 17487
+    },
+    {
+      "epoch": 1.7949507389162562,
+      "grad_norm": 0.08012344688177109,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 17490
+    },
+    {
+      "epoch": 1.7952586206896552,
+      "grad_norm": 0.128404900431633,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 17493
+    },
+    {
+      "epoch": 1.7955665024630543,
+      "grad_norm": 0.08718766272068024,
+      "learning_rate": 0.01,
+      "loss": 2.0557,
+      "step": 17496
+    },
+    {
+      "epoch": 1.7958743842364533,
+      "grad_norm": 0.030426733195781708,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 17499
+    },
+    {
+      "epoch": 1.7961822660098523,
+      "grad_norm": 0.03950949385762215,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 17502
+    },
+    {
+      "epoch": 1.7964901477832513,
+      "grad_norm": 0.049466658383607864,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 17505
+    },
+    {
+      "epoch": 1.7967980295566504,
+      "grad_norm": 0.06188172101974487,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 17508
+    },
+    {
+      "epoch": 1.7971059113300494,
+      "grad_norm": 0.06420351564884186,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 17511
+    },
+    {
+      "epoch": 1.7974137931034484,
+      "grad_norm": 0.04329871013760567,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 17514
+    },
+    {
+      "epoch": 1.7977216748768474,
+      "grad_norm": 0.04420280084013939,
+      "learning_rate": 0.01,
+      "loss": 2.0481,
+      "step": 17517
+    },
+    {
+      "epoch": 1.7980295566502464,
+      "grad_norm": 0.04043954238295555,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 17520
+    },
+    {
+      "epoch": 1.7983374384236455,
+      "grad_norm": 0.049305226653814316,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 17523
+    },
+    {
+      "epoch": 1.7986453201970445,
+      "grad_norm": 0.1928088515996933,
+      "learning_rate": 0.01,
+      "loss": 2.0869,
+      "step": 17526
+    },
+    {
+      "epoch": 1.7989532019704435,
+      "grad_norm": 0.12283357232809067,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 17529
+    },
+    {
+      "epoch": 1.7992610837438425,
+      "grad_norm": 0.07897382229566574,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 17532
+    },
+    {
+      "epoch": 1.7995689655172413,
+      "grad_norm": 0.0749836266040802,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 17535
+    },
+    {
+      "epoch": 1.7998768472906403,
+      "grad_norm": 0.06578727811574936,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 17538
+    },
+    {
+      "epoch": 1.8001847290640394,
+      "grad_norm": 0.06609571725130081,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 17541
+    },
+    {
+      "epoch": 1.8004926108374384,
+      "grad_norm": 0.047696053981781006,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 17544
+    },
+    {
+      "epoch": 1.8008004926108374,
+      "grad_norm": 0.05110754072666168,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 17547
+    },
+    {
+      "epoch": 1.8011083743842364,
+      "grad_norm": 0.03783520683646202,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 17550
+    },
+    {
+      "epoch": 1.8014162561576355,
+      "grad_norm": 0.03145405650138855,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 17553
+    },
+    {
+      "epoch": 1.8017241379310345,
+      "grad_norm": 0.09492892026901245,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 17556
+    },
+    {
+      "epoch": 1.8020320197044335,
+      "grad_norm": 0.06920488178730011,
+      "learning_rate": 0.01,
+      "loss": 2.0809,
+      "step": 17559
+    },
+    {
+      "epoch": 1.8023399014778325,
+      "grad_norm": 0.0583655945956707,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 17562
+    },
+    {
+      "epoch": 1.8026477832512315,
+      "grad_norm": 0.08449242264032364,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 17565
+    },
+    {
+      "epoch": 1.8029556650246306,
+      "grad_norm": 0.12186135351657867,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 17568
+    },
+    {
+      "epoch": 1.8032635467980296,
+      "grad_norm": 0.09926268458366394,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 17571
+    },
+    {
+      "epoch": 1.8035714285714286,
+      "grad_norm": 0.06820474565029144,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 17574
+    },
+    {
+      "epoch": 1.8038793103448276,
+      "grad_norm": 0.050847604870796204,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 17577
+    },
+    {
+      "epoch": 1.8041871921182266,
+      "grad_norm": 0.053053803741931915,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 17580
+    },
+    {
+      "epoch": 1.8044950738916257,
+      "grad_norm": 0.047114890068769455,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 17583
+    },
+    {
+      "epoch": 1.8048029556650245,
+      "grad_norm": 0.05182573199272156,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 17586
+    },
+    {
+      "epoch": 1.8051108374384235,
+      "grad_norm": 0.12609605491161346,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 17589
+    },
+    {
+      "epoch": 1.8054187192118225,
+      "grad_norm": 0.0496569462120533,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 17592
+    },
+    {
+      "epoch": 1.8057266009852215,
+      "grad_norm": 0.0490572527050972,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 17595
+    },
+    {
+      "epoch": 1.8060344827586206,
+      "grad_norm": 0.038300756365060806,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 17598
+    },
+    {
+      "epoch": 1.8063423645320196,
+      "grad_norm": 0.03666609153151512,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 17601
+    },
+    {
+      "epoch": 1.8066502463054186,
+      "grad_norm": 0.036330632865428925,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 17604
+    },
+    {
+      "epoch": 1.8069581280788176,
+      "grad_norm": 0.0605342797935009,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 17607
+    },
+    {
+      "epoch": 1.8072660098522166,
+      "grad_norm": 0.04346880316734314,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 17610
+    },
+    {
+      "epoch": 1.8075738916256157,
+      "grad_norm": 0.06400660425424576,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 17613
+    },
+    {
+      "epoch": 1.8078817733990147,
+      "grad_norm": 0.0812198668718338,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 17616
+    },
+    {
+      "epoch": 1.8081896551724137,
+      "grad_norm": 0.06756972521543503,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 17619
+    },
+    {
+      "epoch": 1.8084975369458127,
+      "grad_norm": 0.05277147516608238,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 17622
+    },
+    {
+      "epoch": 1.8088054187192117,
+      "grad_norm": 0.07904385775327682,
+      "learning_rate": 0.01,
+      "loss": 2.0393,
+      "step": 17625
+    },
+    {
+      "epoch": 1.8091133004926108,
+      "grad_norm": 0.06955704092979431,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 17628
+    },
+    {
+      "epoch": 1.8094211822660098,
+      "grad_norm": 0.06605497002601624,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 17631
+    },
+    {
+      "epoch": 1.8097290640394088,
+      "grad_norm": 0.03861093521118164,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 17634
+    },
+    {
+      "epoch": 1.8100369458128078,
+      "grad_norm": 0.04323074221611023,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 17637
+    },
+    {
+      "epoch": 1.8103448275862069,
+      "grad_norm": 0.03443233296275139,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 17640
+    },
+    {
+      "epoch": 1.8106527093596059,
+      "grad_norm": 0.04190131649374962,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 17643
+    },
+    {
+      "epoch": 1.810960591133005,
+      "grad_norm": 0.09095717966556549,
+      "learning_rate": 0.01,
+      "loss": 2.0529,
+      "step": 17646
+    },
+    {
+      "epoch": 1.811268472906404,
+      "grad_norm": 0.05452005937695503,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 17649
+    },
+    {
+      "epoch": 1.811576354679803,
+      "grad_norm": 0.05032350867986679,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 17652
+    },
+    {
+      "epoch": 1.811884236453202,
+      "grad_norm": 0.05733015760779381,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 17655
+    },
+    {
+      "epoch": 1.812192118226601,
+      "grad_norm": 0.09373817592859268,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 17658
+    },
+    {
+      "epoch": 1.8125,
+      "grad_norm": 0.07385890185832977,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 17661
+    },
+    {
+      "epoch": 1.812807881773399,
+      "grad_norm": 0.08643963187932968,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 17664
+    },
+    {
+      "epoch": 1.813115763546798,
+      "grad_norm": 0.09909530729055405,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 17667
+    },
+    {
+      "epoch": 1.813423645320197,
+      "grad_norm": 0.04600978642702103,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 17670
+    },
+    {
+      "epoch": 1.813731527093596,
+      "grad_norm": 0.033060222864151,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 17673
+    },
+    {
+      "epoch": 1.814039408866995,
+      "grad_norm": 0.03789517655968666,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 17676
+    },
+    {
+      "epoch": 1.8143472906403941,
+      "grad_norm": 0.0502844899892807,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 17679
+    },
+    {
+      "epoch": 1.8146551724137931,
+      "grad_norm": 0.0627695843577385,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 17682
+    },
+    {
+      "epoch": 1.8149630541871922,
+      "grad_norm": 0.15737055242061615,
+      "learning_rate": 0.01,
+      "loss": 2.0572,
+      "step": 17685
+    },
+    {
+      "epoch": 1.8152709359605912,
+      "grad_norm": 0.09944868832826614,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 17688
+    },
+    {
+      "epoch": 1.8155788177339902,
+      "grad_norm": 0.12345952540636063,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 17691
+    },
+    {
+      "epoch": 1.8158866995073892,
+      "grad_norm": 0.06330909579992294,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 17694
+    },
+    {
+      "epoch": 1.8161945812807883,
+      "grad_norm": 0.0584748238325119,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 17697
+    },
+    {
+      "epoch": 1.8165024630541873,
+      "grad_norm": 0.07284627109766006,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 17700
+    },
+    {
+      "epoch": 1.8168103448275863,
+      "grad_norm": 0.07302995771169662,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 17703
+    },
+    {
+      "epoch": 1.8171182266009853,
+      "grad_norm": 0.06292667984962463,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 17706
+    },
+    {
+      "epoch": 1.8174261083743843,
+      "grad_norm": 0.04821958392858505,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 17709
+    },
+    {
+      "epoch": 1.8177339901477834,
+      "grad_norm": 0.03572079911828041,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 17712
+    },
+    {
+      "epoch": 1.8180418719211824,
+      "grad_norm": 0.12643416225910187,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 17715
+    },
+    {
+      "epoch": 1.8183497536945814,
+      "grad_norm": 0.08803770691156387,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 17718
+    },
+    {
+      "epoch": 1.8186576354679804,
+      "grad_norm": 0.061583418399095535,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 17721
+    },
+    {
+      "epoch": 1.8189655172413794,
+      "grad_norm": 0.04947415366768837,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 17724
+    },
+    {
+      "epoch": 1.8192733990147785,
+      "grad_norm": 0.06042906641960144,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 17727
+    },
+    {
+      "epoch": 1.8195812807881775,
+      "grad_norm": 0.03236406669020653,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 17730
+    },
+    {
+      "epoch": 1.8198891625615765,
+      "grad_norm": 0.05975859984755516,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 17733
+    },
+    {
+      "epoch": 1.8201970443349755,
+      "grad_norm": 0.11028258502483368,
+      "learning_rate": 0.01,
+      "loss": 2.0654,
+      "step": 17736
+    },
+    {
+      "epoch": 1.8205049261083743,
+      "grad_norm": 0.055842846632003784,
+      "learning_rate": 0.01,
+      "loss": 2.0589,
+      "step": 17739
+    },
+    {
+      "epoch": 1.8208128078817734,
+      "grad_norm": 0.09189102053642273,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 17742
+    },
+    {
+      "epoch": 1.8211206896551724,
+      "grad_norm": 0.07795927673578262,
+      "learning_rate": 0.01,
+      "loss": 2.0628,
+      "step": 17745
+    },
+    {
+      "epoch": 1.8214285714285714,
+      "grad_norm": 0.06452701985836029,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 17748
+    },
+    {
+      "epoch": 1.8217364532019704,
+      "grad_norm": 0.056360337883234024,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 17751
+    },
+    {
+      "epoch": 1.8220443349753694,
+      "grad_norm": 0.08861987292766571,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 17754
+    },
+    {
+      "epoch": 1.8223522167487685,
+      "grad_norm": 0.07276416569948196,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 17757
+    },
+    {
+      "epoch": 1.8226600985221675,
+      "grad_norm": 0.07501320540904999,
+      "learning_rate": 0.01,
+      "loss": 2.0592,
+      "step": 17760
+    },
+    {
+      "epoch": 1.8229679802955665,
+      "grad_norm": 0.08408310264348984,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 17763
+    },
+    {
+      "epoch": 1.8232758620689655,
+      "grad_norm": 0.039008378982543945,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 17766
+    },
+    {
+      "epoch": 1.8235837438423645,
+      "grad_norm": 0.05153367295861244,
+      "learning_rate": 0.01,
+      "loss": 2.0614,
+      "step": 17769
+    },
+    {
+      "epoch": 1.8238916256157636,
+      "grad_norm": 0.05685068294405937,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 17772
+    },
+    {
+      "epoch": 1.8241995073891626,
+      "grad_norm": 0.10836745798587799,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 17775
+    },
+    {
+      "epoch": 1.8245073891625616,
+      "grad_norm": 0.13855011761188507,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 17778
+    },
+    {
+      "epoch": 1.8248152709359606,
+      "grad_norm": 0.07912803441286087,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 17781
+    },
+    {
+      "epoch": 1.8251231527093597,
+      "grad_norm": 0.065729521214962,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 17784
+    },
+    {
+      "epoch": 1.8254310344827587,
+      "grad_norm": 0.04546307399868965,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 17787
+    },
+    {
+      "epoch": 1.8257389162561575,
+      "grad_norm": 0.03415641188621521,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 17790
+    },
+    {
+      "epoch": 1.8260467980295565,
+      "grad_norm": 0.038325123488903046,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 17793
+    },
+    {
+      "epoch": 1.8263546798029555,
+      "grad_norm": 0.057417213916778564,
+      "learning_rate": 0.01,
+      "loss": 2.0465,
+      "step": 17796
+    },
+    {
+      "epoch": 1.8266625615763545,
+      "grad_norm": 0.07312962412834167,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 17799
+    },
+    {
+      "epoch": 1.8269704433497536,
+      "grad_norm": 0.06465096771717072,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 17802
+    },
+    {
+      "epoch": 1.8272783251231526,
+      "grad_norm": 0.049065001308918,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 17805
+    },
+    {
+      "epoch": 1.8275862068965516,
+      "grad_norm": 0.05004505068063736,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 17808
+    },
+    {
+      "epoch": 1.8278940886699506,
+      "grad_norm": 0.12177273631095886,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 17811
+    },
+    {
+      "epoch": 1.8282019704433496,
+      "grad_norm": 0.09006219357252121,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 17814
+    },
+    {
+      "epoch": 1.8285098522167487,
+      "grad_norm": 0.07000398635864258,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 17817
+    },
+    {
+      "epoch": 1.8288177339901477,
+      "grad_norm": 0.03561507910490036,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 17820
+    },
+    {
+      "epoch": 1.8291256157635467,
+      "grad_norm": 0.050965216010808945,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 17823
+    },
+    {
+      "epoch": 1.8294334975369457,
+      "grad_norm": 0.04437123239040375,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 17826
+    },
+    {
+      "epoch": 1.8297413793103448,
+      "grad_norm": 0.046157170087099075,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 17829
+    },
+    {
+      "epoch": 1.8300492610837438,
+      "grad_norm": 0.0641985610127449,
+      "learning_rate": 0.01,
+      "loss": 2.0619,
+      "step": 17832
+    },
+    {
+      "epoch": 1.8303571428571428,
+      "grad_norm": 0.10295763611793518,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 17835
+    },
+    {
+      "epoch": 1.8306650246305418,
+      "grad_norm": 0.08395816385746002,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 17838
+    },
+    {
+      "epoch": 1.8309729064039408,
+      "grad_norm": 0.07087874412536621,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 17841
+    },
+    {
+      "epoch": 1.8312807881773399,
+      "grad_norm": 0.04754515364766121,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 17844
+    },
+    {
+      "epoch": 1.8315886699507389,
+      "grad_norm": 0.042998362332582474,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 17847
+    },
+    {
+      "epoch": 1.831896551724138,
+      "grad_norm": 0.044786881655454636,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 17850
+    },
+    {
+      "epoch": 1.832204433497537,
+      "grad_norm": 0.05035366117954254,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 17853
+    },
+    {
+      "epoch": 1.832512315270936,
+      "grad_norm": 0.08760454505681992,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 17856
+    },
+    {
+      "epoch": 1.832820197044335,
+      "grad_norm": 0.07182349264621735,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 17859
+    },
+    {
+      "epoch": 1.833128078817734,
+      "grad_norm": 0.0653420239686966,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 17862
+    },
+    {
+      "epoch": 1.833435960591133,
+      "grad_norm": 0.07664595544338226,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 17865
+    },
+    {
+      "epoch": 1.833743842364532,
+      "grad_norm": 0.052884750068187714,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 17868
+    },
+    {
+      "epoch": 1.834051724137931,
+      "grad_norm": 0.049432456493377686,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 17871
+    },
+    {
+      "epoch": 1.83435960591133,
+      "grad_norm": 0.10208621621131897,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 17874
+    },
+    {
+      "epoch": 1.834667487684729,
+      "grad_norm": 0.0663546770811081,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 17877
+    },
+    {
+      "epoch": 1.8349753694581281,
+      "grad_norm": 0.0952199399471283,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 17880
+    },
+    {
+      "epoch": 1.8352832512315271,
+      "grad_norm": 0.04969238117337227,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 17883
+    },
+    {
+      "epoch": 1.8355911330049262,
+      "grad_norm": 0.05101123824715614,
+      "learning_rate": 0.01,
+      "loss": 2.0642,
+      "step": 17886
+    },
+    {
+      "epoch": 1.8358990147783252,
+      "grad_norm": 0.1026005819439888,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 17889
+    },
+    {
+      "epoch": 1.8362068965517242,
+      "grad_norm": 0.06481184810400009,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 17892
+    },
+    {
+      "epoch": 1.8365147783251232,
+      "grad_norm": 0.0684402734041214,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 17895
+    },
+    {
+      "epoch": 1.8368226600985222,
+      "grad_norm": 0.1051085963845253,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 17898
+    },
+    {
+      "epoch": 1.8371305418719213,
+      "grad_norm": 0.06582857668399811,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 17901
+    },
+    {
+      "epoch": 1.8374384236453203,
+      "grad_norm": 0.05665391683578491,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 17904
+    },
+    {
+      "epoch": 1.8377463054187193,
+      "grad_norm": 0.06239892914891243,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 17907
+    },
+    {
+      "epoch": 1.8380541871921183,
+      "grad_norm": 0.08531507849693298,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 17910
+    },
+    {
+      "epoch": 1.8383620689655173,
+      "grad_norm": 0.07379250973463058,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 17913
+    },
+    {
+      "epoch": 1.8386699507389164,
+      "grad_norm": 0.052789974957704544,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 17916
+    },
+    {
+      "epoch": 1.8389778325123154,
+      "grad_norm": 0.09525316208600998,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 17919
+    },
+    {
+      "epoch": 1.8392857142857144,
+      "grad_norm": 0.05700648948550224,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 17922
+    },
+    {
+      "epoch": 1.8395935960591134,
+      "grad_norm": 0.061519671231508255,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 17925
+    },
+    {
+      "epoch": 1.8399014778325125,
+      "grad_norm": 0.05594256520271301,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 17928
+    },
+    {
+      "epoch": 1.8402093596059115,
+      "grad_norm": 0.06823567301034927,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 17931
+    },
+    {
+      "epoch": 1.8405172413793105,
+      "grad_norm": 0.061398666352033615,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 17934
+    },
+    {
+      "epoch": 1.8408251231527095,
+      "grad_norm": 0.10590513050556183,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 17937
+    },
+    {
+      "epoch": 1.8411330049261085,
+      "grad_norm": 0.0579022578895092,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 17940
+    },
+    {
+      "epoch": 1.8414408866995073,
+      "grad_norm": 0.07047640532255173,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 17943
+    },
+    {
+      "epoch": 1.8417487684729064,
+      "grad_norm": 0.07486578077077866,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 17946
+    },
+    {
+      "epoch": 1.8420566502463054,
+      "grad_norm": 0.057884715497493744,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 17949
+    },
+    {
+      "epoch": 1.8423645320197044,
+      "grad_norm": 0.10381656140089035,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 17952
+    },
+    {
+      "epoch": 1.8426724137931034,
+      "grad_norm": 0.041863467544317245,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 17955
+    },
+    {
+      "epoch": 1.8429802955665024,
+      "grad_norm": 0.10012530535459518,
+      "learning_rate": 0.01,
+      "loss": 2.0648,
+      "step": 17958
+    },
+    {
+      "epoch": 1.8432881773399015,
+      "grad_norm": 0.05597177520394325,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 17961
+    },
+    {
+      "epoch": 1.8435960591133005,
+      "grad_norm": 0.05338521674275398,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 17964
+    },
+    {
+      "epoch": 1.8439039408866995,
+      "grad_norm": 0.049141060560941696,
+      "learning_rate": 0.01,
+      "loss": 2.0486,
+      "step": 17967
+    },
+    {
+      "epoch": 1.8442118226600985,
+      "grad_norm": 0.0784049779176712,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 17970
+    },
+    {
+      "epoch": 1.8445197044334976,
+      "grad_norm": 0.038596317172050476,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 17973
+    },
+    {
+      "epoch": 1.8448275862068966,
+      "grad_norm": 0.08521022647619247,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 17976
+    },
+    {
+      "epoch": 1.8451354679802956,
+      "grad_norm": 0.05890432372689247,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 17979
+    },
+    {
+      "epoch": 1.8454433497536946,
+      "grad_norm": 0.09090931713581085,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 17982
+    },
+    {
+      "epoch": 1.8457512315270936,
+      "grad_norm": 0.06019595265388489,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 17985
+    },
+    {
+      "epoch": 1.8460591133004927,
+      "grad_norm": 0.07712443917989731,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 17988
+    },
+    {
+      "epoch": 1.8463669950738915,
+      "grad_norm": 0.06155428662896156,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 17991
+    },
+    {
+      "epoch": 1.8466748768472905,
+      "grad_norm": 0.07221681624650955,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 17994
+    },
+    {
+      "epoch": 1.8469827586206895,
+      "grad_norm": 0.056776583194732666,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 17997
+    },
+    {
+      "epoch": 1.8472906403940885,
+      "grad_norm": 0.12099254876375198,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 18000
+    },
+    {
+      "epoch": 1.8475985221674875,
+      "grad_norm": 0.060344647616147995,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 18003
+    },
+    {
+      "epoch": 1.8479064039408866,
+      "grad_norm": 0.042333200573921204,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 18006
+    },
+    {
+      "epoch": 1.8482142857142856,
+      "grad_norm": 0.046059176325798035,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 18009
+    },
+    {
+      "epoch": 1.8485221674876846,
+      "grad_norm": 0.03853166475892067,
+      "learning_rate": 0.01,
+      "loss": 2.0681,
+      "step": 18012
+    },
+    {
+      "epoch": 1.8488300492610836,
+      "grad_norm": 0.05197960138320923,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 18015
+    },
+    {
+      "epoch": 1.8491379310344827,
+      "grad_norm": 0.03316551446914673,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 18018
+    },
+    {
+      "epoch": 1.8494458128078817,
+      "grad_norm": 0.05977516993880272,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 18021
+    },
+    {
+      "epoch": 1.8497536945812807,
+      "grad_norm": 0.12331486493349075,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 18024
+    },
+    {
+      "epoch": 1.8500615763546797,
+      "grad_norm": 0.08065730333328247,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 18027
+    },
+    {
+      "epoch": 1.8503694581280787,
+      "grad_norm": 0.07649014890193939,
+      "learning_rate": 0.01,
+      "loss": 2.0629,
+      "step": 18030
+    },
+    {
+      "epoch": 1.8506773399014778,
+      "grad_norm": 0.08431357145309448,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 18033
+    },
+    {
+      "epoch": 1.8509852216748768,
+      "grad_norm": 0.041856877505779266,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 18036
+    },
+    {
+      "epoch": 1.8512931034482758,
+      "grad_norm": 0.03598650172352791,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 18039
+    },
+    {
+      "epoch": 1.8516009852216748,
+      "grad_norm": 0.0360511913895607,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 18042
+    },
+    {
+      "epoch": 1.8519088669950738,
+      "grad_norm": 0.0409335158765316,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 18045
+    },
+    {
+      "epoch": 1.8522167487684729,
+      "grad_norm": 0.04646136611700058,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 18048
+    },
+    {
+      "epoch": 1.8525246305418719,
+      "grad_norm": 0.08265028148889542,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 18051
+    },
+    {
+      "epoch": 1.852832512315271,
+      "grad_norm": 0.08118387311697006,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 18054
+    },
+    {
+      "epoch": 1.85314039408867,
+      "grad_norm": 0.05400428548455238,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 18057
+    },
+    {
+      "epoch": 1.853448275862069,
+      "grad_norm": 0.04605553671717644,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 18060
+    },
+    {
+      "epoch": 1.853756157635468,
+      "grad_norm": 0.06259449571371078,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 18063
+    },
+    {
+      "epoch": 1.854064039408867,
+      "grad_norm": 0.04901091381907463,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 18066
+    },
+    {
+      "epoch": 1.854371921182266,
+      "grad_norm": 0.13033097982406616,
+      "learning_rate": 0.01,
+      "loss": 2.0696,
+      "step": 18069
+    },
+    {
+      "epoch": 1.854679802955665,
+      "grad_norm": 0.04114639014005661,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 18072
+    },
+    {
+      "epoch": 1.854987684729064,
+      "grad_norm": 0.05269275978207588,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 18075
+    },
+    {
+      "epoch": 1.855295566502463,
+      "grad_norm": 0.052419982850551605,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 18078
+    },
+    {
+      "epoch": 1.855603448275862,
+      "grad_norm": 0.11109264940023422,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 18081
+    },
+    {
+      "epoch": 1.8559113300492611,
+      "grad_norm": 0.09544682502746582,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 18084
+    },
+    {
+      "epoch": 1.8562192118226601,
+      "grad_norm": 0.08282047510147095,
+      "learning_rate": 0.01,
+      "loss": 2.0601,
+      "step": 18087
+    },
+    {
+      "epoch": 1.8565270935960592,
+      "grad_norm": 0.04255926236510277,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 18090
+    },
+    {
+      "epoch": 1.8568349753694582,
+      "grad_norm": 0.04899003729224205,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 18093
+    },
+    {
+      "epoch": 1.8571428571428572,
+      "grad_norm": 0.05081721395254135,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 18096
+    },
+    {
+      "epoch": 1.8574507389162562,
+      "grad_norm": 0.06633096188306808,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 18099
+    },
+    {
+      "epoch": 1.8577586206896552,
+      "grad_norm": 0.06513489037752151,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 18102
+    },
+    {
+      "epoch": 1.8580665024630543,
+      "grad_norm": 0.09429512917995453,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 18105
+    },
+    {
+      "epoch": 1.8583743842364533,
+      "grad_norm": 0.06042760610580444,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 18108
+    },
+    {
+      "epoch": 1.8586822660098523,
+      "grad_norm": 0.04098260775208473,
+      "learning_rate": 0.01,
+      "loss": 2.0712,
+      "step": 18111
+    },
+    {
+      "epoch": 1.8589901477832513,
+      "grad_norm": 0.04882989823818207,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 18114
+    },
+    {
+      "epoch": 1.8592980295566504,
+      "grad_norm": 0.06196373701095581,
+      "learning_rate": 0.01,
+      "loss": 2.0535,
+      "step": 18117
+    },
+    {
+      "epoch": 1.8596059113300494,
+      "grad_norm": 0.10515942424535751,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 18120
+    },
+    {
+      "epoch": 1.8599137931034484,
+      "grad_norm": 0.08962828665971756,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 18123
+    },
+    {
+      "epoch": 1.8602216748768474,
+      "grad_norm": 0.07672600448131561,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 18126
+    },
+    {
+      "epoch": 1.8605295566502464,
+      "grad_norm": 0.04769902676343918,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 18129
+    },
+    {
+      "epoch": 1.8608374384236455,
+      "grad_norm": 0.06558270752429962,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 18132
+    },
+    {
+      "epoch": 1.8611453201970445,
+      "grad_norm": 0.06909210234880447,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 18135
+    },
+    {
+      "epoch": 1.8614532019704435,
+      "grad_norm": 0.07571686059236526,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 18138
+    },
+    {
+      "epoch": 1.8617610837438425,
+      "grad_norm": 0.11170367896556854,
+      "learning_rate": 0.01,
+      "loss": 2.05,
+      "step": 18141
+    },
+    {
+      "epoch": 1.8620689655172413,
+      "grad_norm": 0.12978370487689972,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 18144
+    },
+    {
+      "epoch": 1.8623768472906403,
+      "grad_norm": 0.056673552840948105,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 18147
+    },
+    {
+      "epoch": 1.8626847290640394,
+      "grad_norm": 0.04110763967037201,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 18150
+    },
+    {
+      "epoch": 1.8629926108374384,
+      "grad_norm": 0.06502550840377808,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 18153
+    },
+    {
+      "epoch": 1.8633004926108374,
+      "grad_norm": 0.059242501854896545,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 18156
+    },
+    {
+      "epoch": 1.8636083743842364,
+      "grad_norm": 0.05173099413514137,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 18159
+    },
+    {
+      "epoch": 1.8639162561576355,
+      "grad_norm": 0.0403546541929245,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 18162
+    },
+    {
+      "epoch": 1.8642241379310345,
+      "grad_norm": 0.10920348763465881,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 18165
+    },
+    {
+      "epoch": 1.8645320197044335,
+      "grad_norm": 0.05528813973069191,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 18168
+    },
+    {
+      "epoch": 1.8648399014778325,
+      "grad_norm": 0.04583175480365753,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 18171
+    },
+    {
+      "epoch": 1.8651477832512315,
+      "grad_norm": 0.04503155127167702,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 18174
+    },
+    {
+      "epoch": 1.8654556650246306,
+      "grad_norm": 0.038750261068344116,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 18177
+    },
+    {
+      "epoch": 1.8657635467980296,
+      "grad_norm": 0.0650535374879837,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 18180
+    },
+    {
+      "epoch": 1.8660714285714286,
+      "grad_norm": 0.08589319884777069,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 18183
+    },
+    {
+      "epoch": 1.8663793103448276,
+      "grad_norm": 0.06245085969567299,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 18186
+    },
+    {
+      "epoch": 1.8666871921182266,
+      "grad_norm": 0.07419238984584808,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 18189
+    },
+    {
+      "epoch": 1.8669950738916257,
+      "grad_norm": 0.06336040049791336,
+      "learning_rate": 0.01,
+      "loss": 2.0822,
+      "step": 18192
+    },
+    {
+      "epoch": 1.8673029556650245,
+      "grad_norm": 0.09494315087795258,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 18195
+    },
+    {
+      "epoch": 1.8676108374384235,
+      "grad_norm": 0.06543273478746414,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 18198
+    },
+    {
+      "epoch": 1.8679187192118225,
+      "grad_norm": 0.05602452531456947,
+      "learning_rate": 0.01,
+      "loss": 2.0626,
+      "step": 18201
+    },
+    {
+      "epoch": 1.8682266009852215,
+      "grad_norm": 0.041735779494047165,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 18204
+    },
+    {
+      "epoch": 1.8685344827586206,
+      "grad_norm": 0.03998032957315445,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 18207
+    },
+    {
+      "epoch": 1.8688423645320196,
+      "grad_norm": 0.07010776549577713,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 18210
+    },
+    {
+      "epoch": 1.8691502463054186,
+      "grad_norm": 0.08516181260347366,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 18213
+    },
+    {
+      "epoch": 1.8694581280788176,
+      "grad_norm": 0.08233955502510071,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 18216
+    },
+    {
+      "epoch": 1.8697660098522166,
+      "grad_norm": 0.06725854426622391,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 18219
+    },
+    {
+      "epoch": 1.8700738916256157,
+      "grad_norm": 0.03345496207475662,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 18222
+    },
+    {
+      "epoch": 1.8703817733990147,
+      "grad_norm": 0.09758662432432175,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 18225
+    },
+    {
+      "epoch": 1.8706896551724137,
+      "grad_norm": 0.05376002565026283,
+      "learning_rate": 0.01,
+      "loss": 2.0592,
+      "step": 18228
+    },
+    {
+      "epoch": 1.8709975369458127,
+      "grad_norm": 0.04620193690061569,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 18231
+    },
+    {
+      "epoch": 1.8713054187192117,
+      "grad_norm": 0.0853218212723732,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 18234
+    },
+    {
+      "epoch": 1.8716133004926108,
+      "grad_norm": 0.15689584612846375,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 18237
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.05037194490432739,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 18240
+    },
+    {
+      "epoch": 1.8722290640394088,
+      "grad_norm": 0.031591251492500305,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 18243
+    },
+    {
+      "epoch": 1.8725369458128078,
+      "grad_norm": 0.05832947790622711,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 18246
+    },
+    {
+      "epoch": 1.8728448275862069,
+      "grad_norm": 0.059619709849357605,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 18249
+    },
+    {
+      "epoch": 1.8731527093596059,
+      "grad_norm": 0.04232211783528328,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 18252
+    },
+    {
+      "epoch": 1.873460591133005,
+      "grad_norm": 0.03756287693977356,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 18255
+    },
+    {
+      "epoch": 1.873768472906404,
+      "grad_norm": 0.04177022725343704,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 18258
+    },
+    {
+      "epoch": 1.874076354679803,
+      "grad_norm": 0.054342493414878845,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 18261
+    },
+    {
+      "epoch": 1.874384236453202,
+      "grad_norm": 0.11872408539056778,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 18264
+    },
+    {
+      "epoch": 1.874692118226601,
+      "grad_norm": 0.05143645405769348,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 18267
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 0.06726546585559845,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 18270
+    },
+    {
+      "epoch": 1.875307881773399,
+      "grad_norm": 0.10031245648860931,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 18273
+    },
+    {
+      "epoch": 1.875615763546798,
+      "grad_norm": 0.09145006537437439,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 18276
+    },
+    {
+      "epoch": 1.875923645320197,
+      "grad_norm": 0.0797610953450203,
+      "learning_rate": 0.01,
+      "loss": 2.0507,
+      "step": 18279
+    },
+    {
+      "epoch": 1.876231527093596,
+      "grad_norm": 0.07170062512159348,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 18282
+    },
+    {
+      "epoch": 1.876539408866995,
+      "grad_norm": 0.043757934123277664,
+      "learning_rate": 0.01,
+      "loss": 2.0825,
+      "step": 18285
+    },
+    {
+      "epoch": 1.8768472906403941,
+      "grad_norm": 0.059610821306705475,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 18288
+    },
+    {
+      "epoch": 1.8771551724137931,
+      "grad_norm": 0.06014898791909218,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 18291
+    },
+    {
+      "epoch": 1.8774630541871922,
+      "grad_norm": 0.0823112353682518,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 18294
+    },
+    {
+      "epoch": 1.8777709359605912,
+      "grad_norm": 0.11342828720808029,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 18297
+    },
+    {
+      "epoch": 1.8780788177339902,
+      "grad_norm": 0.1533091962337494,
+      "learning_rate": 0.01,
+      "loss": 2.0593,
+      "step": 18300
+    },
+    {
+      "epoch": 1.8783866995073892,
+      "grad_norm": 0.09665971249341965,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 18303
+    },
+    {
+      "epoch": 1.8786945812807883,
+      "grad_norm": 0.043992988765239716,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 18306
+    },
+    {
+      "epoch": 1.8790024630541873,
+      "grad_norm": 0.03452041372656822,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 18309
+    },
+    {
+      "epoch": 1.8793103448275863,
+      "grad_norm": 0.05596618726849556,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 18312
+    },
+    {
+      "epoch": 1.8796182266009853,
+      "grad_norm": 0.0542016327381134,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 18315
+    },
+    {
+      "epoch": 1.8799261083743843,
+      "grad_norm": 0.049744654446840286,
+      "learning_rate": 0.01,
+      "loss": 2.0625,
+      "step": 18318
+    },
+    {
+      "epoch": 1.8802339901477834,
+      "grad_norm": 0.04219472035765648,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 18321
+    },
+    {
+      "epoch": 1.8805418719211824,
+      "grad_norm": 0.11828272044658661,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 18324
+    },
+    {
+      "epoch": 1.8808497536945814,
+      "grad_norm": 0.042180564254522324,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 18327
+    },
+    {
+      "epoch": 1.8811576354679804,
+      "grad_norm": 0.05486786365509033,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 18330
+    },
+    {
+      "epoch": 1.8814655172413794,
+      "grad_norm": 0.09456659108400345,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 18333
+    },
+    {
+      "epoch": 1.8817733990147785,
+      "grad_norm": 0.03962776064872742,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 18336
+    },
+    {
+      "epoch": 1.8820812807881775,
+      "grad_norm": 0.06588723510503769,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 18339
+    },
+    {
+      "epoch": 1.8823891625615765,
+      "grad_norm": 0.0490611270070076,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 18342
+    },
+    {
+      "epoch": 1.8826970443349755,
+      "grad_norm": 0.10546906292438507,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 18345
+    },
+    {
+      "epoch": 1.8830049261083743,
+      "grad_norm": 0.05751054733991623,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 18348
+    },
+    {
+      "epoch": 1.8833128078817734,
+      "grad_norm": 0.102676160633564,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 18351
+    },
+    {
+      "epoch": 1.8836206896551724,
+      "grad_norm": 0.0536825954914093,
+      "learning_rate": 0.01,
+      "loss": 2.0677,
+      "step": 18354
+    },
+    {
+      "epoch": 1.8839285714285714,
+      "grad_norm": 0.039357978850603104,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 18357
+    },
+    {
+      "epoch": 1.8842364532019704,
+      "grad_norm": 0.03515158221125603,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 18360
+    },
+    {
+      "epoch": 1.8845443349753694,
+      "grad_norm": 0.04234091565012932,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 18363
+    },
+    {
+      "epoch": 1.8848522167487685,
+      "grad_norm": 0.06893119215965271,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 18366
+    },
+    {
+      "epoch": 1.8851600985221675,
+      "grad_norm": 0.08353175222873688,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 18369
+    },
+    {
+      "epoch": 1.8854679802955665,
+      "grad_norm": 0.0567467026412487,
+      "learning_rate": 0.01,
+      "loss": 2.0672,
+      "step": 18372
+    },
+    {
+      "epoch": 1.8857758620689655,
+      "grad_norm": 0.05529101565480232,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 18375
+    },
+    {
+      "epoch": 1.8860837438423645,
+      "grad_norm": 0.09581262618303299,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 18378
+    },
+    {
+      "epoch": 1.8863916256157636,
+      "grad_norm": 0.052034202963113785,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 18381
+    },
+    {
+      "epoch": 1.8866995073891626,
+      "grad_norm": 0.10314558446407318,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 18384
+    },
+    {
+      "epoch": 1.8870073891625616,
+      "grad_norm": 0.04631441831588745,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 18387
+    },
+    {
+      "epoch": 1.8873152709359606,
+      "grad_norm": 0.08067111670970917,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 18390
+    },
+    {
+      "epoch": 1.8876231527093597,
+      "grad_norm": 0.07892802357673645,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 18393
+    },
+    {
+      "epoch": 1.8879310344827587,
+      "grad_norm": 0.06167163327336311,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 18396
+    },
+    {
+      "epoch": 1.8882389162561575,
+      "grad_norm": 0.06746269762516022,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 18399
+    },
+    {
+      "epoch": 1.8885467980295565,
+      "grad_norm": 0.06389199942350388,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 18402
+    },
+    {
+      "epoch": 1.8888546798029555,
+      "grad_norm": 0.05448369309306145,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 18405
+    },
+    {
+      "epoch": 1.8891625615763545,
+      "grad_norm": 0.05535599961876869,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 18408
+    },
+    {
+      "epoch": 1.8894704433497536,
+      "grad_norm": 0.0533415786921978,
+      "learning_rate": 0.01,
+      "loss": 2.0648,
+      "step": 18411
+    },
+    {
+      "epoch": 1.8897783251231526,
+      "grad_norm": 0.06423043459653854,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 18414
+    },
+    {
+      "epoch": 1.8900862068965516,
+      "grad_norm": 0.04682399705052376,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 18417
+    },
+    {
+      "epoch": 1.8903940886699506,
+      "grad_norm": 0.09395507723093033,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 18420
+    },
+    {
+      "epoch": 1.8907019704433496,
+      "grad_norm": 0.07777848839759827,
+      "learning_rate": 0.01,
+      "loss": 2.068,
+      "step": 18423
+    },
+    {
+      "epoch": 1.8910098522167487,
+      "grad_norm": 0.052006904035806656,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 18426
+    },
+    {
+      "epoch": 1.8913177339901477,
+      "grad_norm": 0.05325109511613846,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 18429
+    },
+    {
+      "epoch": 1.8916256157635467,
+      "grad_norm": 0.05496850982308388,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 18432
+    },
+    {
+      "epoch": 1.8919334975369457,
+      "grad_norm": 0.0395922027528286,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 18435
+    },
+    {
+      "epoch": 1.8922413793103448,
+      "grad_norm": 0.0664554312825203,
+      "learning_rate": 0.01,
+      "loss": 2.0555,
+      "step": 18438
+    },
+    {
+      "epoch": 1.8925492610837438,
+      "grad_norm": 0.06475098431110382,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 18441
+    },
+    {
+      "epoch": 1.8928571428571428,
+      "grad_norm": 0.06291298568248749,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 18444
+    },
+    {
+      "epoch": 1.8931650246305418,
+      "grad_norm": 0.07891640067100525,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 18447
+    },
+    {
+      "epoch": 1.8934729064039408,
+      "grad_norm": 0.10065143555402756,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 18450
+    },
+    {
+      "epoch": 1.8937807881773399,
+      "grad_norm": 0.0962565690279007,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 18453
+    },
+    {
+      "epoch": 1.8940886699507389,
+      "grad_norm": 0.06286763399839401,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 18456
+    },
+    {
+      "epoch": 1.894396551724138,
+      "grad_norm": 0.05067887529730797,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 18459
+    },
+    {
+      "epoch": 1.894704433497537,
+      "grad_norm": 0.03956342115998268,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 18462
+    },
+    {
+      "epoch": 1.895012315270936,
+      "grad_norm": 0.06670361012220383,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 18465
+    },
+    {
+      "epoch": 1.895320197044335,
+      "grad_norm": 0.07848145067691803,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 18468
+    },
+    {
+      "epoch": 1.895628078817734,
+      "grad_norm": 0.05415938422083855,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 18471
+    },
+    {
+      "epoch": 1.895935960591133,
+      "grad_norm": 0.07549092918634415,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 18474
+    },
+    {
+      "epoch": 1.896243842364532,
+      "grad_norm": 0.057871196419000626,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 18477
+    },
+    {
+      "epoch": 1.896551724137931,
+      "grad_norm": 0.059748612344264984,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 18480
+    },
+    {
+      "epoch": 1.89685960591133,
+      "grad_norm": 0.04501016065478325,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 18483
+    },
+    {
+      "epoch": 1.897167487684729,
+      "grad_norm": 0.06361118704080582,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 18486
+    },
+    {
+      "epoch": 1.8974753694581281,
+      "grad_norm": 0.09649393707513809,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 18489
+    },
+    {
+      "epoch": 1.8977832512315271,
+      "grad_norm": 0.08288730680942535,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 18492
+    },
+    {
+      "epoch": 1.8980911330049262,
+      "grad_norm": 0.037788692861795425,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 18495
+    },
+    {
+      "epoch": 1.8983990147783252,
+      "grad_norm": 0.05678097531199455,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 18498
+    },
+    {
+      "epoch": 1.8987068965517242,
+      "grad_norm": 0.05753886699676514,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 18501
+    },
+    {
+      "epoch": 1.8990147783251232,
+      "grad_norm": 0.0542941652238369,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 18504
+    },
+    {
+      "epoch": 1.8993226600985222,
+      "grad_norm": 0.06856728345155716,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 18507
+    },
+    {
+      "epoch": 1.8996305418719213,
+      "grad_norm": 0.09270088374614716,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 18510
+    },
+    {
+      "epoch": 1.8999384236453203,
+      "grad_norm": 0.04473109543323517,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 18513
+    },
+    {
+      "epoch": 1.9002463054187193,
+      "grad_norm": 0.06040007993578911,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 18516
+    },
+    {
+      "epoch": 1.9005541871921183,
+      "grad_norm": 0.032143257558345795,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 18519
+    },
+    {
+      "epoch": 1.9008620689655173,
+      "grad_norm": 0.04205821454524994,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 18522
+    },
+    {
+      "epoch": 1.9011699507389164,
+      "grad_norm": 0.03920583799481392,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 18525
+    },
+    {
+      "epoch": 1.9014778325123154,
+      "grad_norm": 0.17323125898838043,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 18528
+    },
+    {
+      "epoch": 1.9017857142857144,
+      "grad_norm": 0.04595707729458809,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 18531
+    },
+    {
+      "epoch": 1.9020935960591134,
+      "grad_norm": 0.03803316131234169,
+      "learning_rate": 0.01,
+      "loss": 2.0268,
+      "step": 18534
+    },
+    {
+      "epoch": 1.9024014778325125,
+      "grad_norm": 0.04623658210039139,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 18537
+    },
+    {
+      "epoch": 1.9027093596059115,
+      "grad_norm": 0.04388248175382614,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 18540
+    },
+    {
+      "epoch": 1.9030172413793105,
+      "grad_norm": 0.03582540154457092,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 18543
+    },
+    {
+      "epoch": 1.9033251231527095,
+      "grad_norm": 0.033453166484832764,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 18546
+    },
+    {
+      "epoch": 1.9036330049261085,
+      "grad_norm": 0.04929531365633011,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 18549
+    },
+    {
+      "epoch": 1.9039408866995073,
+      "grad_norm": 0.0575069934129715,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 18552
+    },
+    {
+      "epoch": 1.9042487684729064,
+      "grad_norm": 0.044883664697408676,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 18555
+    },
+    {
+      "epoch": 1.9045566502463054,
+      "grad_norm": 0.06335309147834778,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 18558
+    },
+    {
+      "epoch": 1.9048645320197044,
+      "grad_norm": 0.07315582036972046,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 18561
+    },
+    {
+      "epoch": 1.9051724137931034,
+      "grad_norm": 0.08378446102142334,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 18564
+    },
+    {
+      "epoch": 1.9054802955665024,
+      "grad_norm": 0.09492503106594086,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 18567
+    },
+    {
+      "epoch": 1.9057881773399015,
+      "grad_norm": 0.0497819185256958,
+      "learning_rate": 0.01,
+      "loss": 2.0551,
+      "step": 18570
+    },
+    {
+      "epoch": 1.9060960591133005,
+      "grad_norm": 0.06625241041183472,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 18573
+    },
+    {
+      "epoch": 1.9064039408866995,
+      "grad_norm": 0.13533645868301392,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 18576
+    },
+    {
+      "epoch": 1.9067118226600985,
+      "grad_norm": 0.129546657204628,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 18579
+    },
+    {
+      "epoch": 1.9070197044334976,
+      "grad_norm": 0.0862266942858696,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 18582
+    },
+    {
+      "epoch": 1.9073275862068966,
+      "grad_norm": 0.04262632504105568,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 18585
+    },
+    {
+      "epoch": 1.9076354679802956,
+      "grad_norm": 0.06536297500133514,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 18588
+    },
+    {
+      "epoch": 1.9079433497536946,
+      "grad_norm": 0.04408801719546318,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 18591
+    },
+    {
+      "epoch": 1.9082512315270936,
+      "grad_norm": 0.0382089763879776,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 18594
+    },
+    {
+      "epoch": 1.9085591133004927,
+      "grad_norm": 0.05695042014122009,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 18597
+    },
+    {
+      "epoch": 1.9088669950738915,
+      "grad_norm": 0.06890982389450073,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 18600
+    },
+    {
+      "epoch": 1.9091748768472905,
+      "grad_norm": 0.06517864018678665,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 18603
+    },
+    {
+      "epoch": 1.9094827586206895,
+      "grad_norm": 0.03709007799625397,
+      "learning_rate": 0.01,
+      "loss": 2.0579,
+      "step": 18606
+    },
+    {
+      "epoch": 1.9097906403940885,
+      "grad_norm": 0.040355831384658813,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 18609
+    },
+    {
+      "epoch": 1.9100985221674875,
+      "grad_norm": 0.08973202854394913,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 18612
+    },
+    {
+      "epoch": 1.9104064039408866,
+      "grad_norm": 0.08074682205915451,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 18615
+    },
+    {
+      "epoch": 1.9107142857142856,
+      "grad_norm": 0.07134959101676941,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 18618
+    },
+    {
+      "epoch": 1.9110221674876846,
+      "grad_norm": 0.10389960557222366,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 18621
+    },
+    {
+      "epoch": 1.9113300492610836,
+      "grad_norm": 0.052665699273347855,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 18624
+    },
+    {
+      "epoch": 1.9116379310344827,
+      "grad_norm": 0.06355523318052292,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 18627
+    },
+    {
+      "epoch": 1.9119458128078817,
+      "grad_norm": 0.06806465983390808,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 18630
+    },
+    {
+      "epoch": 1.9122536945812807,
+      "grad_norm": 0.08892465382814407,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 18633
+    },
+    {
+      "epoch": 1.9125615763546797,
+      "grad_norm": 0.09806855767965317,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 18636
+    },
+    {
+      "epoch": 1.9128694581280787,
+      "grad_norm": 0.036283593624830246,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 18639
+    },
+    {
+      "epoch": 1.9131773399014778,
+      "grad_norm": 0.06654248386621475,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 18642
+    },
+    {
+      "epoch": 1.9134852216748768,
+      "grad_norm": 0.07729227095842361,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 18645
+    },
+    {
+      "epoch": 1.9137931034482758,
+      "grad_norm": 0.05296695604920387,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 18648
+    },
+    {
+      "epoch": 1.9141009852216748,
+      "grad_norm": 0.05579183250665665,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 18651
+    },
+    {
+      "epoch": 1.9144088669950738,
+      "grad_norm": 0.04230615124106407,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 18654
+    },
+    {
+      "epoch": 1.9147167487684729,
+      "grad_norm": 0.0412709042429924,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 18657
+    },
+    {
+      "epoch": 1.9150246305418719,
+      "grad_norm": 0.056640543043613434,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 18660
+    },
+    {
+      "epoch": 1.915332512315271,
+      "grad_norm": 0.04353609308600426,
+      "learning_rate": 0.01,
+      "loss": 2.0703,
+      "step": 18663
+    },
+    {
+      "epoch": 1.91564039408867,
+      "grad_norm": 0.04212663322687149,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 18666
+    },
+    {
+      "epoch": 1.915948275862069,
+      "grad_norm": 0.07639022916555405,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 18669
+    },
+    {
+      "epoch": 1.916256157635468,
+      "grad_norm": 0.09405479580163956,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 18672
+    },
+    {
+      "epoch": 1.916564039408867,
+      "grad_norm": 0.07501058280467987,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 18675
+    },
+    {
+      "epoch": 1.916871921182266,
+      "grad_norm": 0.06343735009431839,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 18678
+    },
+    {
+      "epoch": 1.917179802955665,
+      "grad_norm": 0.07794613391160965,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 18681
+    },
+    {
+      "epoch": 1.917487684729064,
+      "grad_norm": 0.1063399538397789,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 18684
+    },
+    {
+      "epoch": 1.917795566502463,
+      "grad_norm": 0.06528618931770325,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 18687
+    },
+    {
+      "epoch": 1.918103448275862,
+      "grad_norm": 0.063084177672863,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 18690
+    },
+    {
+      "epoch": 1.9184113300492611,
+      "grad_norm": 0.06663991510868073,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 18693
+    },
+    {
+      "epoch": 1.9187192118226601,
+      "grad_norm": 0.09827464818954468,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 18696
+    },
+    {
+      "epoch": 1.9190270935960592,
+      "grad_norm": 0.052305273711681366,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 18699
+    },
+    {
+      "epoch": 1.9193349753694582,
+      "grad_norm": 0.07126889377832413,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 18702
+    },
+    {
+      "epoch": 1.9196428571428572,
+      "grad_norm": 0.06262009590864182,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 18705
+    },
+    {
+      "epoch": 1.9199507389162562,
+      "grad_norm": 0.056034356355667114,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 18708
+    },
+    {
+      "epoch": 1.9202586206896552,
+      "grad_norm": 0.07673577219247818,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 18711
+    },
+    {
+      "epoch": 1.9205665024630543,
+      "grad_norm": 0.06006854772567749,
+      "learning_rate": 0.01,
+      "loss": 2.0493,
+      "step": 18714
+    },
+    {
+      "epoch": 1.9208743842364533,
+      "grad_norm": 0.07149071991443634,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 18717
+    },
+    {
+      "epoch": 1.9211822660098523,
+      "grad_norm": 0.04935576766729355,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 18720
+    },
+    {
+      "epoch": 1.9214901477832513,
+      "grad_norm": 0.1052050068974495,
+      "learning_rate": 0.01,
+      "loss": 2.0384,
+      "step": 18723
+    },
+    {
+      "epoch": 1.9217980295566504,
+      "grad_norm": 0.07589028030633926,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 18726
+    },
+    {
+      "epoch": 1.9221059113300494,
+      "grad_norm": 0.0862005278468132,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 18729
+    },
+    {
+      "epoch": 1.9224137931034484,
+      "grad_norm": 0.07210662961006165,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 18732
+    },
+    {
+      "epoch": 1.9227216748768474,
+      "grad_norm": 0.0924825370311737,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 18735
+    },
+    {
+      "epoch": 1.9230295566502464,
+      "grad_norm": 0.05819706991314888,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 18738
+    },
+    {
+      "epoch": 1.9233374384236455,
+      "grad_norm": 0.04784362018108368,
+      "learning_rate": 0.01,
+      "loss": 2.0633,
+      "step": 18741
+    },
+    {
+      "epoch": 1.9236453201970445,
+      "grad_norm": 0.07216835021972656,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 18744
+    },
+    {
+      "epoch": 1.9239532019704435,
+      "grad_norm": 0.05539752170443535,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 18747
+    },
+    {
+      "epoch": 1.9242610837438425,
+      "grad_norm": 0.07037390768527985,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 18750
+    },
+    {
+      "epoch": 1.9245689655172413,
+      "grad_norm": 0.0941152572631836,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 18753
+    },
+    {
+      "epoch": 1.9248768472906403,
+      "grad_norm": 0.05315488949418068,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 18756
+    },
+    {
+      "epoch": 1.9251847290640394,
+      "grad_norm": 0.04986554756760597,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 18759
+    },
+    {
+      "epoch": 1.9254926108374384,
+      "grad_norm": 0.0750490128993988,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 18762
+    },
+    {
+      "epoch": 1.9258004926108374,
+      "grad_norm": 0.13903938233852386,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 18765
+    },
+    {
+      "epoch": 1.9261083743842364,
+      "grad_norm": 0.08733932673931122,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 18768
+    },
+    {
+      "epoch": 1.9264162561576355,
+      "grad_norm": 0.04527903348207474,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 18771
+    },
+    {
+      "epoch": 1.9267241379310345,
+      "grad_norm": 0.036207813769578934,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 18774
+    },
+    {
+      "epoch": 1.9270320197044335,
+      "grad_norm": 0.04572034999728203,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 18777
+    },
+    {
+      "epoch": 1.9273399014778325,
+      "grad_norm": 0.03662864491343498,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 18780
+    },
+    {
+      "epoch": 1.9276477832512315,
+      "grad_norm": 0.12049257755279541,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 18783
+    },
+    {
+      "epoch": 1.9279556650246306,
+      "grad_norm": 0.1005631759762764,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 18786
+    },
+    {
+      "epoch": 1.9282635467980296,
+      "grad_norm": 0.13943985104560852,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 18789
+    },
+    {
+      "epoch": 1.9285714285714286,
+      "grad_norm": 0.13312341272830963,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 18792
+    },
+    {
+      "epoch": 1.9288793103448276,
+      "grad_norm": 0.0517788864672184,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 18795
+    },
+    {
+      "epoch": 1.9291871921182266,
+      "grad_norm": 0.05525217577815056,
+      "learning_rate": 0.01,
+      "loss": 2.0707,
+      "step": 18798
+    },
+    {
+      "epoch": 1.9294950738916257,
+      "grad_norm": 0.04876135662198067,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 18801
+    },
+    {
+      "epoch": 1.9298029556650245,
+      "grad_norm": 0.04944787919521332,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 18804
+    },
+    {
+      "epoch": 1.9301108374384235,
+      "grad_norm": 0.06437812745571136,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 18807
+    },
+    {
+      "epoch": 1.9304187192118225,
+      "grad_norm": 0.04027709737420082,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 18810
+    },
+    {
+      "epoch": 1.9307266009852215,
+      "grad_norm": 0.05178974196314812,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 18813
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.05347009375691414,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 18816
+    },
+    {
+      "epoch": 1.9313423645320196,
+      "grad_norm": 0.03055960312485695,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 18819
+    },
+    {
+      "epoch": 1.9316502463054186,
+      "grad_norm": 0.10861945152282715,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 18822
+    },
+    {
+      "epoch": 1.9319581280788176,
+      "grad_norm": 0.05932777374982834,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 18825
+    },
+    {
+      "epoch": 1.9322660098522166,
+      "grad_norm": 0.046545740216970444,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 18828
+    },
+    {
+      "epoch": 1.9325738916256157,
+      "grad_norm": 0.07582221925258636,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 18831
+    },
+    {
+      "epoch": 1.9328817733990147,
+      "grad_norm": 0.07975540310144424,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 18834
+    },
+    {
+      "epoch": 1.9331896551724137,
+      "grad_norm": 0.07365059852600098,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 18837
+    },
+    {
+      "epoch": 1.9334975369458127,
+      "grad_norm": 0.09160298854112625,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 18840
+    },
+    {
+      "epoch": 1.9338054187192117,
+      "grad_norm": 0.07767198234796524,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 18843
+    },
+    {
+      "epoch": 1.9341133004926108,
+      "grad_norm": 0.07545919716358185,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 18846
+    },
+    {
+      "epoch": 1.9344211822660098,
+      "grad_norm": 0.06564575433731079,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 18849
+    },
+    {
+      "epoch": 1.9347290640394088,
+      "grad_norm": 0.04205799475312233,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 18852
+    },
+    {
+      "epoch": 1.9350369458128078,
+      "grad_norm": 0.07428024709224701,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 18855
+    },
+    {
+      "epoch": 1.9353448275862069,
+      "grad_norm": 0.08150817453861237,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 18858
+    },
+    {
+      "epoch": 1.9356527093596059,
+      "grad_norm": 0.07489453256130219,
+      "learning_rate": 0.01,
+      "loss": 2.0489,
+      "step": 18861
+    },
+    {
+      "epoch": 1.935960591133005,
+      "grad_norm": 0.09983116388320923,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 18864
+    },
+    {
+      "epoch": 1.936268472906404,
+      "grad_norm": 0.09074544906616211,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 18867
+    },
+    {
+      "epoch": 1.936576354679803,
+      "grad_norm": 0.056056223809719086,
+      "learning_rate": 0.01,
+      "loss": 2.0531,
+      "step": 18870
+    },
+    {
+      "epoch": 1.936884236453202,
+      "grad_norm": 0.04818575084209442,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 18873
+    },
+    {
+      "epoch": 1.937192118226601,
+      "grad_norm": 0.04811173304915428,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 18876
+    },
+    {
+      "epoch": 1.9375,
+      "grad_norm": 0.05799747258424759,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 18879
+    },
+    {
+      "epoch": 1.937807881773399,
+      "grad_norm": 0.07357611507177353,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 18882
+    },
+    {
+      "epoch": 1.938115763546798,
+      "grad_norm": 0.10275068879127502,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 18885
+    },
+    {
+      "epoch": 1.938423645320197,
+      "grad_norm": 0.03685866296291351,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 18888
+    },
+    {
+      "epoch": 1.938731527093596,
+      "grad_norm": 0.03603344038128853,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 18891
+    },
+    {
+      "epoch": 1.939039408866995,
+      "grad_norm": 0.06932532042264938,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 18894
+    },
+    {
+      "epoch": 1.9393472906403941,
+      "grad_norm": 0.05983889847993851,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 18897
+    },
+    {
+      "epoch": 1.9396551724137931,
+      "grad_norm": 0.1199260875582695,
+      "learning_rate": 0.01,
+      "loss": 2.0467,
+      "step": 18900
+    },
+    {
+      "epoch": 1.9399630541871922,
+      "grad_norm": 0.06222264841198921,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 18903
+    },
+    {
+      "epoch": 1.9402709359605912,
+      "grad_norm": 0.0764993354678154,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 18906
+    },
+    {
+      "epoch": 1.9405788177339902,
+      "grad_norm": 0.04790098965167999,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 18909
+    },
+    {
+      "epoch": 1.9408866995073892,
+      "grad_norm": 0.03822425380349159,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 18912
+    },
+    {
+      "epoch": 1.9411945812807883,
+      "grad_norm": 0.05336176976561546,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 18915
+    },
+    {
+      "epoch": 1.9415024630541873,
+      "grad_norm": 0.08732246607542038,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 18918
+    },
+    {
+      "epoch": 1.9418103448275863,
+      "grad_norm": 0.08886411786079407,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 18921
+    },
+    {
+      "epoch": 1.9421182266009853,
+      "grad_norm": 0.08069706708192825,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 18924
+    },
+    {
+      "epoch": 1.9424261083743843,
+      "grad_norm": 0.08464798331260681,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 18927
+    },
+    {
+      "epoch": 1.9427339901477834,
+      "grad_norm": 0.07051963359117508,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 18930
+    },
+    {
+      "epoch": 1.9430418719211824,
+      "grad_norm": 0.037250157445669174,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 18933
+    },
+    {
+      "epoch": 1.9433497536945814,
+      "grad_norm": 0.07512888312339783,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 18936
+    },
+    {
+      "epoch": 1.9436576354679804,
+      "grad_norm": 0.12079732865095139,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 18939
+    },
+    {
+      "epoch": 1.9439655172413794,
+      "grad_norm": 0.0511600561439991,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 18942
+    },
+    {
+      "epoch": 1.9442733990147785,
+      "grad_norm": 0.06442293524742126,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 18945
+    },
+    {
+      "epoch": 1.9445812807881775,
+      "grad_norm": 0.04928497597575188,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 18948
+    },
+    {
+      "epoch": 1.9448891625615765,
+      "grad_norm": 0.07882185280323029,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 18951
+    },
+    {
+      "epoch": 1.9451970443349755,
+      "grad_norm": 0.03649712726473808,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 18954
+    },
+    {
+      "epoch": 1.9455049261083743,
+      "grad_norm": 0.10654021054506302,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 18957
+    },
+    {
+      "epoch": 1.9458128078817734,
+      "grad_norm": 0.09655455499887466,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 18960
+    },
+    {
+      "epoch": 1.9461206896551724,
+      "grad_norm": 0.06114486977458,
+      "learning_rate": 0.01,
+      "loss": 2.0578,
+      "step": 18963
+    },
+    {
+      "epoch": 1.9464285714285714,
+      "grad_norm": 0.04167640954256058,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 18966
+    },
+    {
+      "epoch": 1.9467364532019704,
+      "grad_norm": 0.054138265550136566,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 18969
+    },
+    {
+      "epoch": 1.9470443349753694,
+      "grad_norm": 0.04728518798947334,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 18972
+    },
+    {
+      "epoch": 1.9473522167487685,
+      "grad_norm": 0.09992729872465134,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 18975
+    },
+    {
+      "epoch": 1.9476600985221675,
+      "grad_norm": 0.039344482123851776,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 18978
+    },
+    {
+      "epoch": 1.9479679802955665,
+      "grad_norm": 0.10921066254377365,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 18981
+    },
+    {
+      "epoch": 1.9482758620689655,
+      "grad_norm": 0.03921816125512123,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 18984
+    },
+    {
+      "epoch": 1.9485837438423645,
+      "grad_norm": 0.08293361961841583,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 18987
+    },
+    {
+      "epoch": 1.9488916256157636,
+      "grad_norm": 0.08310680091381073,
+      "learning_rate": 0.01,
+      "loss": 2.065,
+      "step": 18990
+    },
+    {
+      "epoch": 1.9491995073891626,
+      "grad_norm": 0.05105976015329361,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 18993
+    },
+    {
+      "epoch": 1.9495073891625616,
+      "grad_norm": 0.0387946255505085,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 18996
+    },
+    {
+      "epoch": 1.9498152709359606,
+      "grad_norm": 0.039592646062374115,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 18999
+    },
+    {
+      "epoch": 1.9501231527093597,
+      "grad_norm": 0.042499393224716187,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 19002
+    },
+    {
+      "epoch": 1.9504310344827587,
+      "grad_norm": 0.1202671155333519,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 19005
+    },
+    {
+      "epoch": 1.9507389162561575,
+      "grad_norm": 0.05047677457332611,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 19008
+    },
+    {
+      "epoch": 1.9510467980295565,
+      "grad_norm": 0.13210178911685944,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 19011
+    },
+    {
+      "epoch": 1.9513546798029555,
+      "grad_norm": 0.07093524187803268,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 19014
+    },
+    {
+      "epoch": 1.9516625615763545,
+      "grad_norm": 0.05528571456670761,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 19017
+    },
+    {
+      "epoch": 1.9519704433497536,
+      "grad_norm": 0.08988847583532333,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 19020
+    },
+    {
+      "epoch": 1.9522783251231526,
+      "grad_norm": 0.05767255648970604,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 19023
+    },
+    {
+      "epoch": 1.9525862068965516,
+      "grad_norm": 0.07641880214214325,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 19026
+    },
+    {
+      "epoch": 1.9528940886699506,
+      "grad_norm": 0.08751394599676132,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 19029
+    },
+    {
+      "epoch": 1.9532019704433496,
+      "grad_norm": 0.06144971400499344,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 19032
+    },
+    {
+      "epoch": 1.9535098522167487,
+      "grad_norm": 0.04502955824136734,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 19035
+    },
+    {
+      "epoch": 1.9538177339901477,
+      "grad_norm": 0.05031814053654671,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 19038
+    },
+    {
+      "epoch": 1.9541256157635467,
+      "grad_norm": 0.08885148912668228,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 19041
+    },
+    {
+      "epoch": 1.9544334975369457,
+      "grad_norm": 0.05610232427716255,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 19044
+    },
+    {
+      "epoch": 1.9547413793103448,
+      "grad_norm": 0.07169227302074432,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 19047
+    },
+    {
+      "epoch": 1.9550492610837438,
+      "grad_norm": 0.07136573642492294,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 19050
+    },
+    {
+      "epoch": 1.9553571428571428,
+      "grad_norm": 0.07761941850185394,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 19053
+    },
+    {
+      "epoch": 1.9556650246305418,
+      "grad_norm": 0.062269363552331924,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 19056
+    },
+    {
+      "epoch": 1.9559729064039408,
+      "grad_norm": 0.08797021210193634,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 19059
+    },
+    {
+      "epoch": 1.9562807881773399,
+      "grad_norm": 0.04401189833879471,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 19062
+    },
+    {
+      "epoch": 1.9565886699507389,
+      "grad_norm": 0.08460985869169235,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 19065
+    },
+    {
+      "epoch": 1.956896551724138,
+      "grad_norm": 0.07027498632669449,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 19068
+    },
+    {
+      "epoch": 1.957204433497537,
+      "grad_norm": 0.06747457385063171,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 19071
+    },
+    {
+      "epoch": 1.957512315270936,
+      "grad_norm": 0.10890615731477737,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 19074
+    },
+    {
+      "epoch": 1.957820197044335,
+      "grad_norm": 0.1049666553735733,
+      "learning_rate": 0.01,
+      "loss": 2.0683,
+      "step": 19077
+    },
+    {
+      "epoch": 1.958128078817734,
+      "grad_norm": 0.04320647194981575,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 19080
+    },
+    {
+      "epoch": 1.958435960591133,
+      "grad_norm": 0.06038364768028259,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 19083
+    },
+    {
+      "epoch": 1.958743842364532,
+      "grad_norm": 0.04486788064241409,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 19086
+    },
+    {
+      "epoch": 1.959051724137931,
+      "grad_norm": 0.045702993869781494,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 19089
+    },
+    {
+      "epoch": 1.95935960591133,
+      "grad_norm": 0.10784240067005157,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 19092
+    },
+    {
+      "epoch": 1.959667487684729,
+      "grad_norm": 0.07740773260593414,
+      "learning_rate": 0.01,
+      "loss": 2.079,
+      "step": 19095
+    },
+    {
+      "epoch": 1.9599753694581281,
+      "grad_norm": 0.0517832413315773,
+      "learning_rate": 0.01,
+      "loss": 2.061,
+      "step": 19098
+    },
+    {
+      "epoch": 1.9602832512315271,
+      "grad_norm": 0.04660172387957573,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 19101
+    },
+    {
+      "epoch": 1.9605911330049262,
+      "grad_norm": 0.084842748939991,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 19104
+    },
+    {
+      "epoch": 1.9608990147783252,
+      "grad_norm": 0.10866342484951019,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 19107
+    },
+    {
+      "epoch": 1.9612068965517242,
+      "grad_norm": 0.060094647109508514,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 19110
+    },
+    {
+      "epoch": 1.9615147783251232,
+      "grad_norm": 0.04082890599966049,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 19113
+    },
+    {
+      "epoch": 1.9618226600985222,
+      "grad_norm": 0.09193670004606247,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 19116
+    },
+    {
+      "epoch": 1.9621305418719213,
+      "grad_norm": 0.049036990851163864,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 19119
+    },
+    {
+      "epoch": 1.9624384236453203,
+      "grad_norm": 0.06658133864402771,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 19122
+    },
+    {
+      "epoch": 1.9627463054187193,
+      "grad_norm": 0.07290081679821014,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 19125
+    },
+    {
+      "epoch": 1.9630541871921183,
+      "grad_norm": 0.05635548382997513,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 19128
+    },
+    {
+      "epoch": 1.9633620689655173,
+      "grad_norm": 0.07143761217594147,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 19131
+    },
+    {
+      "epoch": 1.9636699507389164,
+      "grad_norm": 0.15296097099781036,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 19134
+    },
+    {
+      "epoch": 1.9639778325123154,
+      "grad_norm": 0.1795274019241333,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 19137
+    },
+    {
+      "epoch": 1.9642857142857144,
+      "grad_norm": 0.04691818729043007,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 19140
+    },
+    {
+      "epoch": 1.9645935960591134,
+      "grad_norm": 0.05018999055027962,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 19143
+    },
+    {
+      "epoch": 1.9649014778325125,
+      "grad_norm": 0.0349762961268425,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 19146
+    },
+    {
+      "epoch": 1.9652093596059115,
+      "grad_norm": 0.04055612534284592,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 19149
+    },
+    {
+      "epoch": 1.9655172413793105,
+      "grad_norm": 0.04818587005138397,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 19152
+    },
+    {
+      "epoch": 1.9658251231527095,
+      "grad_norm": 0.03579457104206085,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 19155
+    },
+    {
+      "epoch": 1.9661330049261085,
+      "grad_norm": 0.04382238909602165,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 19158
+    },
+    {
+      "epoch": 1.9664408866995073,
+      "grad_norm": 0.03784547746181488,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 19161
+    },
+    {
+      "epoch": 1.9667487684729064,
+      "grad_norm": 0.049413155764341354,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 19164
+    },
+    {
+      "epoch": 1.9670566502463054,
+      "grad_norm": 0.10560319572687149,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 19167
+    },
+    {
+      "epoch": 1.9673645320197044,
+      "grad_norm": 0.07912679761648178,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 19170
+    },
+    {
+      "epoch": 1.9676724137931034,
+      "grad_norm": 0.051868222653865814,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 19173
+    },
+    {
+      "epoch": 1.9679802955665024,
+      "grad_norm": 0.09925144910812378,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 19176
+    },
+    {
+      "epoch": 1.9682881773399015,
+      "grad_norm": 0.09824500232934952,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 19179
+    },
+    {
+      "epoch": 1.9685960591133005,
+      "grad_norm": 0.04710378497838974,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 19182
+    },
+    {
+      "epoch": 1.9689039408866995,
+      "grad_norm": 0.09339728951454163,
+      "learning_rate": 0.01,
+      "loss": 2.0623,
+      "step": 19185
+    },
+    {
+      "epoch": 1.9692118226600985,
+      "grad_norm": 0.04485667496919632,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 19188
+    },
+    {
+      "epoch": 1.9695197044334976,
+      "grad_norm": 0.06367155909538269,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 19191
+    },
+    {
+      "epoch": 1.9698275862068966,
+      "grad_norm": 0.06692302227020264,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 19194
+    },
+    {
+      "epoch": 1.9701354679802956,
+      "grad_norm": 0.06107610464096069,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 19197
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.06362861394882202,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 19200
+    },
+    {
+      "epoch": 1.9707512315270936,
+      "grad_norm": 0.07524324208498001,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 19203
+    },
+    {
+      "epoch": 1.9710591133004927,
+      "grad_norm": 0.09118182212114334,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 19206
+    },
+    {
+      "epoch": 1.9713669950738915,
+      "grad_norm": 0.0823112204670906,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 19209
+    },
+    {
+      "epoch": 1.9716748768472905,
+      "grad_norm": 0.061318982392549515,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 19212
+    },
+    {
+      "epoch": 1.9719827586206895,
+      "grad_norm": 0.09838750958442688,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 19215
+    },
+    {
+      "epoch": 1.9722906403940885,
+      "grad_norm": 0.061727046966552734,
+      "learning_rate": 0.01,
+      "loss": 2.0671,
+      "step": 19218
+    },
+    {
+      "epoch": 1.9725985221674875,
+      "grad_norm": 0.044177260249853134,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 19221
+    },
+    {
+      "epoch": 1.9729064039408866,
+      "grad_norm": 0.031012659892439842,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 19224
+    },
+    {
+      "epoch": 1.9732142857142856,
+      "grad_norm": 0.0593150295317173,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 19227
+    },
+    {
+      "epoch": 1.9735221674876846,
+      "grad_norm": 0.09283222258090973,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 19230
+    },
+    {
+      "epoch": 1.9738300492610836,
+      "grad_norm": 0.07416541129350662,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 19233
+    },
+    {
+      "epoch": 1.9741379310344827,
+      "grad_norm": 0.08513590693473816,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 19236
+    },
+    {
+      "epoch": 1.9744458128078817,
+      "grad_norm": 0.08401728421449661,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 19239
+    },
+    {
+      "epoch": 1.9747536945812807,
+      "grad_norm": 0.08488047868013382,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 19242
+    },
+    {
+      "epoch": 1.9750615763546797,
+      "grad_norm": 0.11438726633787155,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 19245
+    },
+    {
+      "epoch": 1.9753694581280787,
+      "grad_norm": 0.0416182205080986,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 19248
+    },
+    {
+      "epoch": 1.9756773399014778,
+      "grad_norm": 0.046806883066892624,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 19251
+    },
+    {
+      "epoch": 1.9759852216748768,
+      "grad_norm": 0.04319307208061218,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 19254
+    },
+    {
+      "epoch": 1.9762931034482758,
+      "grad_norm": 0.11832991987466812,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 19257
+    },
+    {
+      "epoch": 1.9766009852216748,
+      "grad_norm": 0.04716213047504425,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 19260
+    },
+    {
+      "epoch": 1.9769088669950738,
+      "grad_norm": 0.08626002073287964,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 19263
+    },
+    {
+      "epoch": 1.9772167487684729,
+      "grad_norm": 0.0981634259223938,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 19266
+    },
+    {
+      "epoch": 1.9775246305418719,
+      "grad_norm": 0.0657229796051979,
+      "learning_rate": 0.01,
+      "loss": 2.065,
+      "step": 19269
+    },
+    {
+      "epoch": 1.977832512315271,
+      "grad_norm": 0.0652332603931427,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 19272
+    },
+    {
+      "epoch": 1.97814039408867,
+      "grad_norm": 0.06810397654771805,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 19275
+    },
+    {
+      "epoch": 1.978448275862069,
+      "grad_norm": 0.04740637540817261,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 19278
+    },
+    {
+      "epoch": 1.978756157635468,
+      "grad_norm": 0.039233241230249405,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 19281
+    },
+    {
+      "epoch": 1.979064039408867,
+      "grad_norm": 0.07533819228410721,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 19284
+    },
+    {
+      "epoch": 1.979371921182266,
+      "grad_norm": 0.0820235162973404,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 19287
+    },
+    {
+      "epoch": 1.979679802955665,
+      "grad_norm": 0.057419124990701675,
+      "learning_rate": 0.01,
+      "loss": 2.0692,
+      "step": 19290
+    },
+    {
+      "epoch": 1.979987684729064,
+      "grad_norm": 0.10119790583848953,
+      "learning_rate": 0.01,
+      "loss": 2.0752,
+      "step": 19293
+    },
+    {
+      "epoch": 1.980295566502463,
+      "grad_norm": 0.116152822971344,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 19296
+    },
+    {
+      "epoch": 1.980603448275862,
+      "grad_norm": 0.05364501103758812,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 19299
+    },
+    {
+      "epoch": 1.9809113300492611,
+      "grad_norm": 0.09089913219213486,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 19302
+    },
+    {
+      "epoch": 1.9812192118226601,
+      "grad_norm": 0.06570890545845032,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 19305
+    },
+    {
+      "epoch": 1.9815270935960592,
+      "grad_norm": 0.10739763081073761,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 19308
+    },
+    {
+      "epoch": 1.9818349753694582,
+      "grad_norm": 0.0396854430437088,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 19311
+    },
+    {
+      "epoch": 1.9821428571428572,
+      "grad_norm": 0.11273244023323059,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 19314
+    },
+    {
+      "epoch": 1.9824507389162562,
+      "grad_norm": 0.10009465366601944,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 19317
+    },
+    {
+      "epoch": 1.9827586206896552,
+      "grad_norm": 0.053756825625896454,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 19320
+    },
+    {
+      "epoch": 1.9830665024630543,
+      "grad_norm": 0.06603456288576126,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 19323
+    },
+    {
+      "epoch": 1.9833743842364533,
+      "grad_norm": 0.038810715079307556,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 19326
+    },
+    {
+      "epoch": 1.9836822660098523,
+      "grad_norm": 0.04284658655524254,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 19329
+    },
+    {
+      "epoch": 1.9839901477832513,
+      "grad_norm": 0.04441271349787712,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 19332
+    },
+    {
+      "epoch": 1.9842980295566504,
+      "grad_norm": 0.04501213878393173,
+      "learning_rate": 0.01,
+      "loss": 2.0517,
+      "step": 19335
+    },
+    {
+      "epoch": 1.9846059113300494,
+      "grad_norm": 0.05109642818570137,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 19338
+    },
+    {
+      "epoch": 1.9849137931034484,
+      "grad_norm": 0.03543083369731903,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 19341
+    },
+    {
+      "epoch": 1.9852216748768474,
+      "grad_norm": 0.04665149003267288,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 19344
+    },
+    {
+      "epoch": 1.9855295566502464,
+      "grad_norm": 0.035318441689014435,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 19347
+    },
+    {
+      "epoch": 1.9858374384236455,
+      "grad_norm": 0.035862043499946594,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 19350
+    },
+    {
+      "epoch": 1.9861453201970445,
+      "grad_norm": 0.128739133477211,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 19353
+    },
+    {
+      "epoch": 1.9864532019704435,
+      "grad_norm": 0.08115250617265701,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 19356
+    },
+    {
+      "epoch": 1.9867610837438425,
+      "grad_norm": 0.04203096404671669,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 19359
+    },
+    {
+      "epoch": 1.9870689655172413,
+      "grad_norm": 0.03801970183849335,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 19362
+    },
+    {
+      "epoch": 1.9873768472906403,
+      "grad_norm": 0.05322232097387314,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 19365
+    },
+    {
+      "epoch": 1.9876847290640394,
+      "grad_norm": 0.037100568413734436,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 19368
+    },
+    {
+      "epoch": 1.9879926108374384,
+      "grad_norm": 0.03714398667216301,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 19371
+    },
+    {
+      "epoch": 1.9883004926108374,
+      "grad_norm": 0.050371263176202774,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 19374
+    },
+    {
+      "epoch": 1.9886083743842364,
+      "grad_norm": 0.03875119984149933,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 19377
+    },
+    {
+      "epoch": 1.9889162561576355,
+      "grad_norm": 0.06838756054639816,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 19380
+    },
+    {
+      "epoch": 1.9892241379310345,
+      "grad_norm": 0.04749476909637451,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 19383
+    },
+    {
+      "epoch": 1.9895320197044335,
+      "grad_norm": 0.041247084736824036,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 19386
+    },
+    {
+      "epoch": 1.9898399014778325,
+      "grad_norm": 0.08582460135221481,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 19389
+    },
+    {
+      "epoch": 1.9901477832512315,
+      "grad_norm": 0.042033273726701736,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 19392
+    },
+    {
+      "epoch": 1.9904556650246306,
+      "grad_norm": 0.08395756036043167,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 19395
+    },
+    {
+      "epoch": 1.9907635467980296,
+      "grad_norm": 0.07154903560876846,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 19398
+    },
+    {
+      "epoch": 1.9910714285714286,
+      "grad_norm": 0.06137581169605255,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 19401
+    },
+    {
+      "epoch": 1.9913793103448276,
+      "grad_norm": 0.1226835623383522,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 19404
+    },
+    {
+      "epoch": 1.9916871921182266,
+      "grad_norm": 0.06524399667978287,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 19407
+    },
+    {
+      "epoch": 1.9919950738916257,
+      "grad_norm": 0.060310300439596176,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 19410
+    },
+    {
+      "epoch": 1.9923029556650245,
+      "grad_norm": 0.10605314373970032,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 19413
+    },
+    {
+      "epoch": 1.9926108374384235,
+      "grad_norm": 0.07056690007448196,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 19416
+    },
+    {
+      "epoch": 1.9929187192118225,
+      "grad_norm": 0.04367789626121521,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 19419
+    },
+    {
+      "epoch": 1.9932266009852215,
+      "grad_norm": 0.07856806367635727,
+      "learning_rate": 0.01,
+      "loss": 2.061,
+      "step": 19422
+    },
+    {
+      "epoch": 1.9935344827586206,
+      "grad_norm": 0.07237541675567627,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 19425
+    },
+    {
+      "epoch": 1.9938423645320196,
+      "grad_norm": 0.04983443021774292,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 19428
+    },
+    {
+      "epoch": 1.9941502463054186,
+      "grad_norm": 0.049045633524656296,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 19431
+    },
+    {
+      "epoch": 1.9944581280788176,
+      "grad_norm": 0.07521536946296692,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 19434
+    },
+    {
+      "epoch": 1.9947660098522166,
+      "grad_norm": 0.03521602228283882,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 19437
+    },
+    {
+      "epoch": 1.9950738916256157,
+      "grad_norm": 0.06075441092252731,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 19440
+    },
+    {
+      "epoch": 1.9953817733990147,
+      "grad_norm": 0.08800282329320908,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 19443
+    },
+    {
+      "epoch": 1.9956896551724137,
+      "grad_norm": 0.04632639139890671,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 19446
+    },
+    {
+      "epoch": 1.9959975369458127,
+      "grad_norm": 0.05275778844952583,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 19449
+    },
+    {
+      "epoch": 1.9963054187192117,
+      "grad_norm": 0.04615132138133049,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 19452
+    },
+    {
+      "epoch": 1.9966133004926108,
+      "grad_norm": 0.12586715817451477,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 19455
+    },
+    {
+      "epoch": 1.9969211822660098,
+      "grad_norm": 0.09406362473964691,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 19458
+    },
+    {
+      "epoch": 1.9972290640394088,
+      "grad_norm": 0.032408129423856735,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 19461
+    },
+    {
+      "epoch": 1.9975369458128078,
+      "grad_norm": 0.08810164034366608,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 19464
+    },
+    {
+      "epoch": 1.9978448275862069,
+      "grad_norm": 0.11263968795537949,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 19467
+    },
+    {
+      "epoch": 1.9981527093596059,
+      "grad_norm": 0.06618282198905945,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 19470
+    },
+    {
+      "epoch": 1.998460591133005,
+      "grad_norm": 0.03649067133665085,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 19473
+    },
+    {
+      "epoch": 1.998768472906404,
+      "grad_norm": 0.03718538209795952,
+      "learning_rate": 0.01,
+      "loss": 2.0517,
+      "step": 19476
+    },
+    {
+      "epoch": 1.999076354679803,
+      "grad_norm": 0.047908563166856766,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 19479
+    },
+    {
+      "epoch": 1.999384236453202,
+      "grad_norm": 0.04926212877035141,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 19482
+    },
+    {
+      "epoch": 1.999692118226601,
+      "grad_norm": 0.08558470755815506,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 19485
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.07099032402038574,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 19488
+    },
+    {
+      "epoch": 2.003391921060746,
+      "grad_norm": 0.08354249596595764,
+      "learning_rate": 0.01,
+      "loss": 2.065,
+      "step": 19491
+    },
+    {
+      "epoch": 2.003700277520814,
+      "grad_norm": 0.09245558828115463,
+      "learning_rate": 0.01,
+      "loss": 2.0571,
+      "step": 19494
+    },
+    {
+      "epoch": 2.004008633980882,
+      "grad_norm": 0.09228463470935822,
+      "learning_rate": 0.01,
+      "loss": 2.0749,
+      "step": 19497
+    },
+    {
+      "epoch": 2.00431699044095,
+      "grad_norm": 0.05558445304632187,
+      "learning_rate": 0.01,
+      "loss": 2.0754,
+      "step": 19500
+    },
+    {
+      "epoch": 2.0046253469010176,
+      "grad_norm": 0.04827789589762688,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 19503
+    },
+    {
+      "epoch": 2.0049337033610852,
+      "grad_norm": 0.063465915620327,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 19506
+    },
+    {
+      "epoch": 2.0052420598211533,
+      "grad_norm": 0.047648850828409195,
+      "learning_rate": 0.01,
+      "loss": 2.0614,
+      "step": 19509
+    },
+    {
+      "epoch": 2.005550416281221,
+      "grad_norm": 0.038050852715969086,
+      "learning_rate": 0.01,
+      "loss": 2.0699,
+      "step": 19512
+    },
+    {
+      "epoch": 2.005858772741289,
+      "grad_norm": 0.036580201238393784,
+      "learning_rate": 0.01,
+      "loss": 2.0832,
+      "step": 19515
+    },
+    {
+      "epoch": 2.0061671292013568,
+      "grad_norm": 0.033919982612133026,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 19518
+    },
+    {
+      "epoch": 2.0064754856614244,
+      "grad_norm": 0.05007147789001465,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 19521
+    },
+    {
+      "epoch": 2.0067838421214925,
+      "grad_norm": 0.042562540620565414,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 19524
+    },
+    {
+      "epoch": 2.00709219858156,
+      "grad_norm": 0.0872044712305069,
+      "learning_rate": 0.01,
+      "loss": 2.0628,
+      "step": 19527
+    },
+    {
+      "epoch": 2.0074005550416283,
+      "grad_norm": 0.06331472843885422,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 19530
+    },
+    {
+      "epoch": 2.007708911501696,
+      "grad_norm": 0.08928905427455902,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 19533
+    },
+    {
+      "epoch": 2.0080172679617636,
+      "grad_norm": 0.0869508758187294,
+      "learning_rate": 0.01,
+      "loss": 2.0497,
+      "step": 19536
+    },
+    {
+      "epoch": 2.0083256244218317,
+      "grad_norm": 0.04267793521285057,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 19539
+    },
+    {
+      "epoch": 2.0086339808818994,
+      "grad_norm": 0.046922095119953156,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 19542
+    },
+    {
+      "epoch": 2.0089423373419675,
+      "grad_norm": 0.07374055683612823,
+      "learning_rate": 0.01,
+      "loss": 2.0571,
+      "step": 19545
+    },
+    {
+      "epoch": 2.009250693802035,
+      "grad_norm": 0.042078323662281036,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 19548
+    },
+    {
+      "epoch": 2.009559050262103,
+      "grad_norm": 0.052491504698991776,
+      "learning_rate": 0.01,
+      "loss": 2.0996,
+      "step": 19551
+    },
+    {
+      "epoch": 2.009867406722171,
+      "grad_norm": 0.04900294169783592,
+      "learning_rate": 0.01,
+      "loss": 2.0476,
+      "step": 19554
+    },
+    {
+      "epoch": 2.0101757631822386,
+      "grad_norm": 0.13067513704299927,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 19557
+    },
+    {
+      "epoch": 2.0104841196423067,
+      "grad_norm": 0.09229371696710587,
+      "learning_rate": 0.01,
+      "loss": 2.0708,
+      "step": 19560
+    },
+    {
+      "epoch": 2.0107924761023743,
+      "grad_norm": 0.05014317110180855,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 19563
+    },
+    {
+      "epoch": 2.011100832562442,
+      "grad_norm": 0.06385400146245956,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 19566
+    },
+    {
+      "epoch": 2.01140918902251,
+      "grad_norm": 0.04037034139037132,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 19569
+    },
+    {
+      "epoch": 2.0117175454825778,
+      "grad_norm": 0.02967817150056362,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 19572
+    },
+    {
+      "epoch": 2.012025901942646,
+      "grad_norm": 0.04519663751125336,
+      "learning_rate": 0.01,
+      "loss": 2.0497,
+      "step": 19575
+    },
+    {
+      "epoch": 2.0123342584027135,
+      "grad_norm": 0.07825223356485367,
+      "learning_rate": 0.01,
+      "loss": 2.06,
+      "step": 19578
+    },
+    {
+      "epoch": 2.012642614862781,
+      "grad_norm": 0.13088425993919373,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 19581
+    },
+    {
+      "epoch": 2.0129509713228493,
+      "grad_norm": 0.10113450884819031,
+      "learning_rate": 0.01,
+      "loss": 2.0628,
+      "step": 19584
+    },
+    {
+      "epoch": 2.013259327782917,
+      "grad_norm": 0.06662772595882416,
+      "learning_rate": 0.01,
+      "loss": 2.0655,
+      "step": 19587
+    },
+    {
+      "epoch": 2.013567684242985,
+      "grad_norm": 0.04824177175760269,
+      "learning_rate": 0.01,
+      "loss": 2.0853,
+      "step": 19590
+    },
+    {
+      "epoch": 2.0138760407030527,
+      "grad_norm": 0.07255363464355469,
+      "learning_rate": 0.01,
+      "loss": 2.0614,
+      "step": 19593
+    },
+    {
+      "epoch": 2.0141843971631204,
+      "grad_norm": 0.061763517558574677,
+      "learning_rate": 0.01,
+      "loss": 2.0761,
+      "step": 19596
+    },
+    {
+      "epoch": 2.0144927536231885,
+      "grad_norm": 0.058266837149858475,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 19599
+    },
+    {
+      "epoch": 2.014801110083256,
+      "grad_norm": 0.05644237995147705,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 19602
+    },
+    {
+      "epoch": 2.0151094665433242,
+      "grad_norm": 0.102548748254776,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 19605
+    },
+    {
+      "epoch": 2.015417823003392,
+      "grad_norm": 0.06959159672260284,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 19608
+    },
+    {
+      "epoch": 2.0157261794634596,
+      "grad_norm": 0.09066049009561539,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 19611
+    },
+    {
+      "epoch": 2.0160345359235277,
+      "grad_norm": 0.061073388904333115,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 19614
+    },
+    {
+      "epoch": 2.0163428923835953,
+      "grad_norm": 0.09000861644744873,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 19617
+    },
+    {
+      "epoch": 2.0166512488436634,
+      "grad_norm": 0.040078576654195786,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 19620
+    },
+    {
+      "epoch": 2.016959605303731,
+      "grad_norm": 0.045984748750925064,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 19623
+    },
+    {
+      "epoch": 2.0172679617637987,
+      "grad_norm": 0.045942965894937515,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 19626
+    },
+    {
+      "epoch": 2.017576318223867,
+      "grad_norm": 0.059133514761924744,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 19629
+    },
+    {
+      "epoch": 2.0178846746839345,
+      "grad_norm": 0.07386631518602371,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 19632
+    },
+    {
+      "epoch": 2.0181930311440026,
+      "grad_norm": 0.06906817853450775,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 19635
+    },
+    {
+      "epoch": 2.0185013876040703,
+      "grad_norm": 0.06259379535913467,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 19638
+    },
+    {
+      "epoch": 2.018809744064138,
+      "grad_norm": 0.07011371850967407,
+      "learning_rate": 0.01,
+      "loss": 2.0716,
+      "step": 19641
+    },
+    {
+      "epoch": 2.019118100524206,
+      "grad_norm": 0.05763932690024376,
+      "learning_rate": 0.01,
+      "loss": 2.0618,
+      "step": 19644
+    },
+    {
+      "epoch": 2.0194264569842737,
+      "grad_norm": 0.09810350090265274,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 19647
+    },
+    {
+      "epoch": 2.019734813444342,
+      "grad_norm": 0.11434987187385559,
+      "learning_rate": 0.01,
+      "loss": 2.057,
+      "step": 19650
+    },
+    {
+      "epoch": 2.0200431699044095,
+      "grad_norm": 0.10692505538463593,
+      "learning_rate": 0.01,
+      "loss": 2.0669,
+      "step": 19653
+    },
+    {
+      "epoch": 2.020351526364477,
+      "grad_norm": 0.06918302178382874,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 19656
+    },
+    {
+      "epoch": 2.0206598828245452,
+      "grad_norm": 0.06661045551300049,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 19659
+    },
+    {
+      "epoch": 2.020968239284613,
+      "grad_norm": 0.03996479883790016,
+      "learning_rate": 0.01,
+      "loss": 2.0552,
+      "step": 19662
+    },
+    {
+      "epoch": 2.021276595744681,
+      "grad_norm": 0.041359271854162216,
+      "learning_rate": 0.01,
+      "loss": 2.0542,
+      "step": 19665
+    },
+    {
+      "epoch": 2.0215849522047487,
+      "grad_norm": 0.046693217009305954,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 19668
+    },
+    {
+      "epoch": 2.0218933086648163,
+      "grad_norm": 0.031555816531181335,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 19671
+    },
+    {
+      "epoch": 2.0222016651248844,
+      "grad_norm": 0.036573588848114014,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 19674
+    },
+    {
+      "epoch": 2.022510021584952,
+      "grad_norm": 0.09064050763845444,
+      "learning_rate": 0.01,
+      "loss": 2.0609,
+      "step": 19677
+    },
+    {
+      "epoch": 2.02281837804502,
+      "grad_norm": 0.11865704506635666,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 19680
+    },
+    {
+      "epoch": 2.023126734505088,
+      "grad_norm": 0.08720502257347107,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 19683
+    },
+    {
+      "epoch": 2.0234350909651555,
+      "grad_norm": 0.06953457742929459,
+      "learning_rate": 0.01,
+      "loss": 2.0769,
+      "step": 19686
+    },
+    {
+      "epoch": 2.0237434474252236,
+      "grad_norm": 0.04386308416724205,
+      "learning_rate": 0.01,
+      "loss": 2.0595,
+      "step": 19689
+    },
+    {
+      "epoch": 2.0240518038852913,
+      "grad_norm": 0.047490183264017105,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 19692
+    },
+    {
+      "epoch": 2.0243601603453594,
+      "grad_norm": 0.061406608670949936,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 19695
+    },
+    {
+      "epoch": 2.024668516805427,
+      "grad_norm": 0.0626315325498581,
+      "learning_rate": 0.01,
+      "loss": 2.0908,
+      "step": 19698
+    },
+    {
+      "epoch": 2.0249768732654947,
+      "grad_norm": 0.048075366765260696,
+      "learning_rate": 0.01,
+      "loss": 2.0703,
+      "step": 19701
+    },
+    {
+      "epoch": 2.025285229725563,
+      "grad_norm": 0.06243044510483742,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 19704
+    },
+    {
+      "epoch": 2.0255935861856305,
+      "grad_norm": 0.06498084217309952,
+      "learning_rate": 0.01,
+      "loss": 2.0812,
+      "step": 19707
+    },
+    {
+      "epoch": 2.0259019426456986,
+      "grad_norm": 0.05091014504432678,
+      "learning_rate": 0.01,
+      "loss": 2.0451,
+      "step": 19710
+    },
+    {
+      "epoch": 2.026210299105766,
+      "grad_norm": 0.04733705893158913,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 19713
+    },
+    {
+      "epoch": 2.026518655565834,
+      "grad_norm": 0.10866537690162659,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 19716
+    },
+    {
+      "epoch": 2.026827012025902,
+      "grad_norm": 0.07504774630069733,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 19719
+    },
+    {
+      "epoch": 2.0271353684859696,
+      "grad_norm": 0.07938455790281296,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 19722
+    },
+    {
+      "epoch": 2.0274437249460378,
+      "grad_norm": 0.10090157389640808,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 19725
+    },
+    {
+      "epoch": 2.0277520814061054,
+      "grad_norm": 0.04133126139640808,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 19728
+    },
+    {
+      "epoch": 2.0280604378661735,
+      "grad_norm": 0.12022694200277328,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 19731
+    },
+    {
+      "epoch": 2.028368794326241,
+      "grad_norm": 0.05904841795563698,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 19734
+    },
+    {
+      "epoch": 2.028677150786309,
+      "grad_norm": 0.08344896882772446,
+      "learning_rate": 0.01,
+      "loss": 2.0721,
+      "step": 19737
+    },
+    {
+      "epoch": 2.028985507246377,
+      "grad_norm": 0.045264534652233124,
+      "learning_rate": 0.01,
+      "loss": 2.0597,
+      "step": 19740
+    },
+    {
+      "epoch": 2.0292938637064446,
+      "grad_norm": 0.05907116085290909,
+      "learning_rate": 0.01,
+      "loss": 2.0561,
+      "step": 19743
+    },
+    {
+      "epoch": 2.0296022201665127,
+      "grad_norm": 0.04975851625204086,
+      "learning_rate": 0.01,
+      "loss": 2.0642,
+      "step": 19746
+    },
+    {
+      "epoch": 2.0299105766265804,
+      "grad_norm": 0.08190937340259552,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 19749
+    },
+    {
+      "epoch": 2.030218933086648,
+      "grad_norm": 0.14594541490077972,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 19752
+    },
+    {
+      "epoch": 2.030527289546716,
+      "grad_norm": 0.11920581012964249,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 19755
+    },
+    {
+      "epoch": 2.030835646006784,
+      "grad_norm": 0.04334663227200508,
+      "learning_rate": 0.01,
+      "loss": 2.0835,
+      "step": 19758
+    },
+    {
+      "epoch": 2.031144002466852,
+      "grad_norm": 0.05323721095919609,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 19761
+    },
+    {
+      "epoch": 2.0314523589269196,
+      "grad_norm": 0.09565315395593643,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 19764
+    },
+    {
+      "epoch": 2.031760715386987,
+      "grad_norm": 0.14498768746852875,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 19767
+    },
+    {
+      "epoch": 2.0320690718470553,
+      "grad_norm": 0.16146855056285858,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 19770
+    },
+    {
+      "epoch": 2.032377428307123,
+      "grad_norm": 0.09023015946149826,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 19773
+    },
+    {
+      "epoch": 2.032685784767191,
+      "grad_norm": 0.05290327966213226,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 19776
+    },
+    {
+      "epoch": 2.0329941412272587,
+      "grad_norm": 0.06551158428192139,
+      "learning_rate": 0.01,
+      "loss": 2.062,
+      "step": 19779
+    },
+    {
+      "epoch": 2.0333024976873264,
+      "grad_norm": 0.09268030524253845,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 19782
+    },
+    {
+      "epoch": 2.0336108541473945,
+      "grad_norm": 0.05402594432234764,
+      "learning_rate": 0.01,
+      "loss": 2.0739,
+      "step": 19785
+    },
+    {
+      "epoch": 2.033919210607462,
+      "grad_norm": 0.052478570491075516,
+      "learning_rate": 0.01,
+      "loss": 2.0709,
+      "step": 19788
+    },
+    {
+      "epoch": 2.0342275670675303,
+      "grad_norm": 0.03243448957800865,
+      "learning_rate": 0.01,
+      "loss": 2.049,
+      "step": 19791
+    },
+    {
+      "epoch": 2.034535923527598,
+      "grad_norm": 0.08627558499574661,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 19794
+    },
+    {
+      "epoch": 2.0348442799876656,
+      "grad_norm": 0.04757314547896385,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 19797
+    },
+    {
+      "epoch": 2.0351526364477337,
+      "grad_norm": 0.11217369884252548,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 19800
+    },
+    {
+      "epoch": 2.0354609929078014,
+      "grad_norm": 0.07525690644979477,
+      "learning_rate": 0.01,
+      "loss": 2.0782,
+      "step": 19803
+    },
+    {
+      "epoch": 2.0357693493678695,
+      "grad_norm": 0.0945955365896225,
+      "learning_rate": 0.01,
+      "loss": 2.0594,
+      "step": 19806
+    },
+    {
+      "epoch": 2.036077705827937,
+      "grad_norm": 0.07789472490549088,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 19809
+    },
+    {
+      "epoch": 2.036386062288005,
+      "grad_norm": 0.06672658026218414,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 19812
+    },
+    {
+      "epoch": 2.036694418748073,
+      "grad_norm": 0.06361529976129532,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 19815
+    },
+    {
+      "epoch": 2.0370027752081405,
+      "grad_norm": 0.03530391305685043,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 19818
+    },
+    {
+      "epoch": 2.0373111316682087,
+      "grad_norm": 0.08201812207698822,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 19821
+    },
+    {
+      "epoch": 2.0376194881282763,
+      "grad_norm": 0.09198293834924698,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 19824
+    },
+    {
+      "epoch": 2.037927844588344,
+      "grad_norm": 0.058875374495983124,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 19827
+    },
+    {
+      "epoch": 2.038236201048412,
+      "grad_norm": 0.04453382268548012,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 19830
+    },
+    {
+      "epoch": 2.0385445575084797,
+      "grad_norm": 0.03713817149400711,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 19833
+    },
+    {
+      "epoch": 2.038852913968548,
+      "grad_norm": 0.056827936321496964,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 19836
+    },
+    {
+      "epoch": 2.0391612704286155,
+      "grad_norm": 0.08166830986738205,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 19839
+    },
+    {
+      "epoch": 2.039469626888683,
+      "grad_norm": 0.06837287545204163,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 19842
+    },
+    {
+      "epoch": 2.0397779833487513,
+      "grad_norm": 0.08867949992418289,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 19845
+    },
+    {
+      "epoch": 2.040086339808819,
+      "grad_norm": 0.07119370251893997,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 19848
+    },
+    {
+      "epoch": 2.040394696268887,
+      "grad_norm": 0.07701986283063889,
+      "learning_rate": 0.01,
+      "loss": 2.0695,
+      "step": 19851
+    },
+    {
+      "epoch": 2.0407030527289547,
+      "grad_norm": 0.04700729623436928,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 19854
+    },
+    {
+      "epoch": 2.0410114091890224,
+      "grad_norm": 0.05898338556289673,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 19857
+    },
+    {
+      "epoch": 2.0413197656490905,
+      "grad_norm": 0.11953815072774887,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 19860
+    },
+    {
+      "epoch": 2.041628122109158,
+      "grad_norm": 0.09704854339361191,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 19863
+    },
+    {
+      "epoch": 2.041936478569226,
+      "grad_norm": 0.1362537145614624,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 19866
+    },
+    {
+      "epoch": 2.042244835029294,
+      "grad_norm": 0.09366025030612946,
+      "learning_rate": 0.01,
+      "loss": 2.0628,
+      "step": 19869
+    },
+    {
+      "epoch": 2.0425531914893615,
+      "grad_norm": 0.05397522822022438,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 19872
+    },
+    {
+      "epoch": 2.0428615479494296,
+      "grad_norm": 0.07723390311002731,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 19875
+    },
+    {
+      "epoch": 2.0431699044094973,
+      "grad_norm": 0.08418615907430649,
+      "learning_rate": 0.01,
+      "loss": 2.0878,
+      "step": 19878
+    },
+    {
+      "epoch": 2.0434782608695654,
+      "grad_norm": 0.06149798631668091,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 19881
+    },
+    {
+      "epoch": 2.043786617329633,
+      "grad_norm": 0.0474097803235054,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 19884
+    },
+    {
+      "epoch": 2.0440949737897007,
+      "grad_norm": 0.04854200407862663,
+      "learning_rate": 0.01,
+      "loss": 2.0637,
+      "step": 19887
+    },
+    {
+      "epoch": 2.044403330249769,
+      "grad_norm": 0.04509511590003967,
+      "learning_rate": 0.01,
+      "loss": 2.0679,
+      "step": 19890
+    },
+    {
+      "epoch": 2.0447116867098365,
+      "grad_norm": 0.05422825738787651,
+      "learning_rate": 0.01,
+      "loss": 2.0449,
+      "step": 19893
+    },
+    {
+      "epoch": 2.0450200431699046,
+      "grad_norm": 0.06556607037782669,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 19896
+    },
+    {
+      "epoch": 2.0453283996299723,
+      "grad_norm": 0.03906751424074173,
+      "learning_rate": 0.01,
+      "loss": 2.0694,
+      "step": 19899
+    },
+    {
+      "epoch": 2.04563675609004,
+      "grad_norm": 0.05207069590687752,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 19902
+    },
+    {
+      "epoch": 2.045945112550108,
+      "grad_norm": 0.04187217727303505,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 19905
+    },
+    {
+      "epoch": 2.0462534690101757,
+      "grad_norm": 0.04163263365626335,
+      "learning_rate": 0.01,
+      "loss": 2.0535,
+      "step": 19908
+    },
+    {
+      "epoch": 2.046561825470244,
+      "grad_norm": 0.037544943392276764,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 19911
+    },
+    {
+      "epoch": 2.0468701819303114,
+      "grad_norm": 0.03623516857624054,
+      "learning_rate": 0.01,
+      "loss": 2.0758,
+      "step": 19914
+    },
+    {
+      "epoch": 2.047178538390379,
+      "grad_norm": 0.08026546239852905,
+      "learning_rate": 0.01,
+      "loss": 2.0654,
+      "step": 19917
+    },
+    {
+      "epoch": 2.047486894850447,
+      "grad_norm": 0.05316372588276863,
+      "learning_rate": 0.01,
+      "loss": 2.0751,
+      "step": 19920
+    },
+    {
+      "epoch": 2.047795251310515,
+      "grad_norm": 0.062127552926540375,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 19923
+    },
+    {
+      "epoch": 2.048103607770583,
+      "grad_norm": 0.049675267189741135,
+      "learning_rate": 0.01,
+      "loss": 2.0741,
+      "step": 19926
+    },
+    {
+      "epoch": 2.0484119642306506,
+      "grad_norm": 0.0425347164273262,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 19929
+    },
+    {
+      "epoch": 2.0487203206907183,
+      "grad_norm": 0.03532329574227333,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 19932
+    },
+    {
+      "epoch": 2.0490286771507864,
+      "grad_norm": 0.05779660493135452,
+      "learning_rate": 0.01,
+      "loss": 2.0801,
+      "step": 19935
+    },
+    {
+      "epoch": 2.049337033610854,
+      "grad_norm": 0.07841507345438004,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 19938
+    },
+    {
+      "epoch": 2.049645390070922,
+      "grad_norm": 0.0883709266781807,
+      "learning_rate": 0.01,
+      "loss": 2.0594,
+      "step": 19941
+    },
+    {
+      "epoch": 2.04995374653099,
+      "grad_norm": 0.09949532151222229,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 19944
+    },
+    {
+      "epoch": 2.0502621029910575,
+      "grad_norm": 0.04350358247756958,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 19947
+    },
+    {
+      "epoch": 2.0505704594511256,
+      "grad_norm": 0.042655814439058304,
+      "learning_rate": 0.01,
+      "loss": 2.0821,
+      "step": 19950
+    },
+    {
+      "epoch": 2.0508788159111933,
+      "grad_norm": 0.060070816427469254,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 19953
+    },
+    {
+      "epoch": 2.0511871723712614,
+      "grad_norm": 0.06479921191930771,
+      "learning_rate": 0.01,
+      "loss": 2.0783,
+      "step": 19956
+    },
+    {
+      "epoch": 2.051495528831329,
+      "grad_norm": 0.0982329398393631,
+      "learning_rate": 0.01,
+      "loss": 2.0756,
+      "step": 19959
+    },
+    {
+      "epoch": 2.0518038852913967,
+      "grad_norm": 0.10483184456825256,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 19962
+    },
+    {
+      "epoch": 2.052112241751465,
+      "grad_norm": 0.06383049488067627,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 19965
+    },
+    {
+      "epoch": 2.0524205982115324,
+      "grad_norm": 0.13797828555107117,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 19968
+    },
+    {
+      "epoch": 2.0527289546716005,
+      "grad_norm": 0.037840090692043304,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 19971
+    },
+    {
+      "epoch": 2.053037311131668,
+      "grad_norm": 0.043872520327568054,
+      "learning_rate": 0.01,
+      "loss": 2.0609,
+      "step": 19974
+    },
+    {
+      "epoch": 2.053345667591736,
+      "grad_norm": 0.03223152458667755,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 19977
+    },
+    {
+      "epoch": 2.053654024051804,
+      "grad_norm": 0.05935351178050041,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 19980
+    },
+    {
+      "epoch": 2.0539623805118716,
+      "grad_norm": 0.054079607129096985,
+      "learning_rate": 0.01,
+      "loss": 2.0665,
+      "step": 19983
+    },
+    {
+      "epoch": 2.0542707369719397,
+      "grad_norm": 0.04307890310883522,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 19986
+    },
+    {
+      "epoch": 2.0545790934320074,
+      "grad_norm": 0.06624720245599747,
+      "learning_rate": 0.01,
+      "loss": 2.052,
+      "step": 19989
+    },
+    {
+      "epoch": 2.054887449892075,
+      "grad_norm": 0.08096028864383698,
+      "learning_rate": 0.01,
+      "loss": 2.0947,
+      "step": 19992
+    },
+    {
+      "epoch": 2.055195806352143,
+      "grad_norm": 0.0872364342212677,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 19995
+    },
+    {
+      "epoch": 2.055504162812211,
+      "grad_norm": 0.04538879171013832,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 19998
+    },
+    {
+      "epoch": 2.055812519272279,
+      "grad_norm": 0.11873256415128708,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 20001
+    },
+    {
+      "epoch": 2.0561208757323466,
+      "grad_norm": 0.05929452180862427,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 20004
+    },
+    {
+      "epoch": 2.0564292321924142,
+      "grad_norm": 0.05131294205784798,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 20007
+    },
+    {
+      "epoch": 2.0567375886524824,
+      "grad_norm": 0.05690256133675575,
+      "learning_rate": 0.01,
+      "loss": 2.0779,
+      "step": 20010
+    },
+    {
+      "epoch": 2.05704594511255,
+      "grad_norm": 0.04414551705121994,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 20013
+    },
+    {
+      "epoch": 2.057354301572618,
+      "grad_norm": 0.04017036780714989,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 20016
+    },
+    {
+      "epoch": 2.0576626580326858,
+      "grad_norm": 0.06785457581281662,
+      "learning_rate": 0.01,
+      "loss": 2.058,
+      "step": 20019
+    },
+    {
+      "epoch": 2.0579710144927534,
+      "grad_norm": 0.06258828938007355,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 20022
+    },
+    {
+      "epoch": 2.0582793709528215,
+      "grad_norm": 0.11196446418762207,
+      "learning_rate": 0.01,
+      "loss": 2.0624,
+      "step": 20025
+    },
+    {
+      "epoch": 2.058587727412889,
+      "grad_norm": 0.08678428828716278,
+      "learning_rate": 0.01,
+      "loss": 2.0667,
+      "step": 20028
+    },
+    {
+      "epoch": 2.0588960838729573,
+      "grad_norm": 0.13598018884658813,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 20031
+    },
+    {
+      "epoch": 2.059204440333025,
+      "grad_norm": 0.06666143238544464,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 20034
+    },
+    {
+      "epoch": 2.0595127967930926,
+      "grad_norm": 0.05994727462530136,
+      "learning_rate": 0.01,
+      "loss": 2.0674,
+      "step": 20037
+    },
+    {
+      "epoch": 2.0598211532531607,
+      "grad_norm": 0.03867008537054062,
+      "learning_rate": 0.01,
+      "loss": 2.056,
+      "step": 20040
+    },
+    {
+      "epoch": 2.0601295097132284,
+      "grad_norm": 0.13077500462532043,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 20043
+    },
+    {
+      "epoch": 2.0604378661732965,
+      "grad_norm": 0.057164691388607025,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 20046
+    },
+    {
+      "epoch": 2.060746222633364,
+      "grad_norm": 0.07206998765468597,
+      "learning_rate": 0.01,
+      "loss": 2.0472,
+      "step": 20049
+    },
+    {
+      "epoch": 2.061054579093432,
+      "grad_norm": 0.08844766765832901,
+      "learning_rate": 0.01,
+      "loss": 2.0535,
+      "step": 20052
+    },
+    {
+      "epoch": 2.0613629355535,
+      "grad_norm": 0.07533573359251022,
+      "learning_rate": 0.01,
+      "loss": 2.0698,
+      "step": 20055
+    },
+    {
+      "epoch": 2.0616712920135676,
+      "grad_norm": 0.03260966017842293,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 20058
+    },
+    {
+      "epoch": 2.0619796484736357,
+      "grad_norm": 0.03491971641778946,
+      "learning_rate": 0.01,
+      "loss": 2.0612,
+      "step": 20061
+    },
+    {
+      "epoch": 2.0622880049337033,
+      "grad_norm": 0.04022398218512535,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 20064
+    },
+    {
+      "epoch": 2.062596361393771,
+      "grad_norm": 0.06325655430555344,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 20067
+    },
+    {
+      "epoch": 2.062904717853839,
+      "grad_norm": 0.06704667955636978,
+      "learning_rate": 0.01,
+      "loss": 2.0641,
+      "step": 20070
+    },
+    {
+      "epoch": 2.0632130743139068,
+      "grad_norm": 0.06883389502763748,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 20073
+    },
+    {
+      "epoch": 2.063521430773975,
+      "grad_norm": 0.05242495611310005,
+      "learning_rate": 0.01,
+      "loss": 2.0701,
+      "step": 20076
+    },
+    {
+      "epoch": 2.0638297872340425,
+      "grad_norm": 0.06587128341197968,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 20079
+    },
+    {
+      "epoch": 2.06413814369411,
+      "grad_norm": 0.03571178764104843,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 20082
+    },
+    {
+      "epoch": 2.0644465001541783,
+      "grad_norm": 0.061605412513017654,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 20085
+    },
+    {
+      "epoch": 2.064754856614246,
+      "grad_norm": 0.05552279204130173,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 20088
+    },
+    {
+      "epoch": 2.065063213074314,
+      "grad_norm": 0.047950152307748795,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 20091
+    },
+    {
+      "epoch": 2.0653715695343817,
+      "grad_norm": 0.09466604888439178,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 20094
+    },
+    {
+      "epoch": 2.0656799259944494,
+      "grad_norm": 0.04828859865665436,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 20097
+    },
+    {
+      "epoch": 2.0659882824545175,
+      "grad_norm": 0.03933820128440857,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 20100
+    },
+    {
+      "epoch": 2.066296638914585,
+      "grad_norm": 0.05044875666499138,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 20103
+    },
+    {
+      "epoch": 2.0666049953746533,
+      "grad_norm": 0.04152398556470871,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 20106
+    },
+    {
+      "epoch": 2.066913351834721,
+      "grad_norm": 0.10098916292190552,
+      "learning_rate": 0.01,
+      "loss": 2.0726,
+      "step": 20109
+    },
+    {
+      "epoch": 2.0672217082947886,
+      "grad_norm": 0.06381060183048248,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 20112
+    },
+    {
+      "epoch": 2.0675300647548567,
+      "grad_norm": 0.13991308212280273,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 20115
+    },
+    {
+      "epoch": 2.0678384212149243,
+      "grad_norm": 0.061171598732471466,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 20118
+    },
+    {
+      "epoch": 2.0681467776749924,
+      "grad_norm": 0.04276692867279053,
+      "learning_rate": 0.01,
+      "loss": 2.0667,
+      "step": 20121
+    },
+    {
+      "epoch": 2.06845513413506,
+      "grad_norm": 0.03582247719168663,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 20124
+    },
+    {
+      "epoch": 2.0687634905951278,
+      "grad_norm": 0.037077244371175766,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 20127
+    },
+    {
+      "epoch": 2.069071847055196,
+      "grad_norm": 0.11291185766458511,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 20130
+    },
+    {
+      "epoch": 2.0693802035152635,
+      "grad_norm": 0.06811921298503876,
+      "learning_rate": 0.01,
+      "loss": 2.0437,
+      "step": 20133
+    },
+    {
+      "epoch": 2.0696885599753316,
+      "grad_norm": 0.049292147159576416,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 20136
+    },
+    {
+      "epoch": 2.0699969164353993,
+      "grad_norm": 0.08937390893697739,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 20139
+    },
+    {
+      "epoch": 2.070305272895467,
+      "grad_norm": 0.04353107511997223,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 20142
+    },
+    {
+      "epoch": 2.070613629355535,
+      "grad_norm": 0.03737090900540352,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 20145
+    },
+    {
+      "epoch": 2.0709219858156027,
+      "grad_norm": 0.038217127323150635,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 20148
+    },
+    {
+      "epoch": 2.071230342275671,
+      "grad_norm": 0.07162989675998688,
+      "learning_rate": 0.01,
+      "loss": 2.0556,
+      "step": 20151
+    },
+    {
+      "epoch": 2.0715386987357385,
+      "grad_norm": 0.06507647782564163,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 20154
+    },
+    {
+      "epoch": 2.071847055195806,
+      "grad_norm": 0.07880635559558868,
+      "learning_rate": 0.01,
+      "loss": 2.0599,
+      "step": 20157
+    },
+    {
+      "epoch": 2.0721554116558742,
+      "grad_norm": 0.11247913539409637,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 20160
+    },
+    {
+      "epoch": 2.072463768115942,
+      "grad_norm": 0.04084709286689758,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 20163
+    },
+    {
+      "epoch": 2.07277212457601,
+      "grad_norm": 0.06349261850118637,
+      "learning_rate": 0.01,
+      "loss": 2.0656,
+      "step": 20166
+    },
+    {
+      "epoch": 2.0730804810360777,
+      "grad_norm": 0.03916813060641289,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 20169
+    },
+    {
+      "epoch": 2.0733888374961453,
+      "grad_norm": 0.041867464780807495,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 20172
+    },
+    {
+      "epoch": 2.0736971939562134,
+      "grad_norm": 0.09670063108205795,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 20175
+    },
+    {
+      "epoch": 2.074005550416281,
+      "grad_norm": 0.05259916931390762,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 20178
+    },
+    {
+      "epoch": 2.074313906876349,
+      "grad_norm": 0.0970730185508728,
+      "learning_rate": 0.01,
+      "loss": 2.0557,
+      "step": 20181
+    },
+    {
+      "epoch": 2.074622263336417,
+      "grad_norm": 0.1317344307899475,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 20184
+    },
+    {
+      "epoch": 2.0749306197964845,
+      "grad_norm": 0.0787033885717392,
+      "learning_rate": 0.01,
+      "loss": 2.072,
+      "step": 20187
+    },
+    {
+      "epoch": 2.0752389762565526,
+      "grad_norm": 0.04037567600607872,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 20190
+    },
+    {
+      "epoch": 2.0755473327166203,
+      "grad_norm": 0.03588324785232544,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 20193
+    },
+    {
+      "epoch": 2.0758556891766884,
+      "grad_norm": 0.05277855321764946,
+      "learning_rate": 0.01,
+      "loss": 2.0702,
+      "step": 20196
+    },
+    {
+      "epoch": 2.076164045636756,
+      "grad_norm": 0.050833381712436676,
+      "learning_rate": 0.01,
+      "loss": 2.0748,
+      "step": 20199
+    },
+    {
+      "epoch": 2.076472402096824,
+      "grad_norm": 0.089606374502182,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 20202
+    },
+    {
+      "epoch": 2.076780758556892,
+      "grad_norm": 0.05270789936184883,
+      "learning_rate": 0.01,
+      "loss": 2.0555,
+      "step": 20205
+    },
+    {
+      "epoch": 2.0770891150169595,
+      "grad_norm": 0.06895376741886139,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 20208
+    },
+    {
+      "epoch": 2.0773974714770276,
+      "grad_norm": 0.11967889964580536,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 20211
+    },
+    {
+      "epoch": 2.0777058279370952,
+      "grad_norm": 0.11328759789466858,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 20214
+    },
+    {
+      "epoch": 2.078014184397163,
+      "grad_norm": 0.045189183205366135,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 20217
+    },
+    {
+      "epoch": 2.078322540857231,
+      "grad_norm": 0.05716565251350403,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 20220
+    },
+    {
+      "epoch": 2.0786308973172987,
+      "grad_norm": 0.07340056449174881,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 20223
+    },
+    {
+      "epoch": 2.0789392537773668,
+      "grad_norm": 0.05395069718360901,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 20226
+    },
+    {
+      "epoch": 2.0792476102374344,
+      "grad_norm": 0.03723681718111038,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 20229
+    },
+    {
+      "epoch": 2.0795559666975025,
+      "grad_norm": 0.10626024752855301,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 20232
+    },
+    {
+      "epoch": 2.07986432315757,
+      "grad_norm": 0.09987606853246689,
+      "learning_rate": 0.01,
+      "loss": 2.067,
+      "step": 20235
+    },
+    {
+      "epoch": 2.080172679617638,
+      "grad_norm": 0.07282060384750366,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 20238
+    },
+    {
+      "epoch": 2.080481036077706,
+      "grad_norm": 0.04192940518260002,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 20241
+    },
+    {
+      "epoch": 2.0807893925377736,
+      "grad_norm": 0.06585846096277237,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 20244
+    },
+    {
+      "epoch": 2.0810977489978413,
+      "grad_norm": 0.04427814856171608,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 20247
+    },
+    {
+      "epoch": 2.0814061054579094,
+      "grad_norm": 0.05114896968007088,
+      "learning_rate": 0.01,
+      "loss": 2.059,
+      "step": 20250
+    },
+    {
+      "epoch": 2.081714461917977,
+      "grad_norm": 0.0445995107293129,
+      "learning_rate": 0.01,
+      "loss": 2.0544,
+      "step": 20253
+    },
+    {
+      "epoch": 2.082022818378045,
+      "grad_norm": 0.04904405400156975,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 20256
+    },
+    {
+      "epoch": 2.082331174838113,
+      "grad_norm": 0.03620357811450958,
+      "learning_rate": 0.01,
+      "loss": 2.0493,
+      "step": 20259
+    },
+    {
+      "epoch": 2.082639531298181,
+      "grad_norm": 0.10994633287191391,
+      "learning_rate": 0.01,
+      "loss": 2.0719,
+      "step": 20262
+    },
+    {
+      "epoch": 2.0829478877582486,
+      "grad_norm": 0.05244474112987518,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 20265
+    },
+    {
+      "epoch": 2.0832562442183162,
+      "grad_norm": 0.05937792733311653,
+      "learning_rate": 0.01,
+      "loss": 2.0593,
+      "step": 20268
+    },
+    {
+      "epoch": 2.0835646006783843,
+      "grad_norm": 0.08669353276491165,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 20271
+    },
+    {
+      "epoch": 2.083872957138452,
+      "grad_norm": 0.054145876318216324,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 20274
+    },
+    {
+      "epoch": 2.08418131359852,
+      "grad_norm": 0.040682870894670486,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 20277
+    },
+    {
+      "epoch": 2.0844896700585878,
+      "grad_norm": 0.04110307991504669,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 20280
+    },
+    {
+      "epoch": 2.0847980265186554,
+      "grad_norm": 0.13420680165290833,
+      "learning_rate": 0.01,
+      "loss": 2.0568,
+      "step": 20283
+    },
+    {
+      "epoch": 2.0851063829787235,
+      "grad_norm": 0.049191731959581375,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 20286
+    },
+    {
+      "epoch": 2.085414739438791,
+      "grad_norm": 0.04682133346796036,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 20289
+    },
+    {
+      "epoch": 2.0857230958988593,
+      "grad_norm": 0.043646588921546936,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 20292
+    },
+    {
+      "epoch": 2.086031452358927,
+      "grad_norm": 0.05107354745268822,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 20295
+    },
+    {
+      "epoch": 2.0863398088189946,
+      "grad_norm": 0.06274458020925522,
+      "learning_rate": 0.01,
+      "loss": 2.0621,
+      "step": 20298
+    },
+    {
+      "epoch": 2.0866481652790627,
+      "grad_norm": 0.11294244229793549,
+      "learning_rate": 0.01,
+      "loss": 2.0685,
+      "step": 20301
+    },
+    {
+      "epoch": 2.0869565217391304,
+      "grad_norm": 0.04948057234287262,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 20304
+    },
+    {
+      "epoch": 2.0872648781991985,
+      "grad_norm": 0.04451402649283409,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 20307
+    },
+    {
+      "epoch": 2.087573234659266,
+      "grad_norm": 0.04940638318657875,
+      "learning_rate": 0.01,
+      "loss": 2.0584,
+      "step": 20310
+    },
+    {
+      "epoch": 2.087881591119334,
+      "grad_norm": 0.06530692428350449,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 20313
+    },
+    {
+      "epoch": 2.088189947579402,
+      "grad_norm": 0.13395404815673828,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 20316
+    },
+    {
+      "epoch": 2.0884983040394696,
+      "grad_norm": 0.09222474694252014,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 20319
+    },
+    {
+      "epoch": 2.0888066604995377,
+      "grad_norm": 0.06510595977306366,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 20322
+    },
+    {
+      "epoch": 2.0891150169596053,
+      "grad_norm": 0.055552888661623,
+      "learning_rate": 0.01,
+      "loss": 2.0612,
+      "step": 20325
+    },
+    {
+      "epoch": 2.089423373419673,
+      "grad_norm": 0.04411375895142555,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 20328
+    },
+    {
+      "epoch": 2.089731729879741,
+      "grad_norm": 0.05151544511318207,
+      "learning_rate": 0.01,
+      "loss": 2.0539,
+      "step": 20331
+    },
+    {
+      "epoch": 2.0900400863398088,
+      "grad_norm": 0.09836700558662415,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 20334
+    },
+    {
+      "epoch": 2.090348442799877,
+      "grad_norm": 0.06430090218782425,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 20337
+    },
+    {
+      "epoch": 2.0906567992599445,
+      "grad_norm": 0.09683403372764587,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 20340
+    },
+    {
+      "epoch": 2.090965155720012,
+      "grad_norm": 0.08345566689968109,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 20343
+    },
+    {
+      "epoch": 2.0912735121800803,
+      "grad_norm": 0.045199088752269745,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 20346
+    },
+    {
+      "epoch": 2.091581868640148,
+      "grad_norm": 0.0399625338613987,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 20349
+    },
+    {
+      "epoch": 2.091890225100216,
+      "grad_norm": 0.03815968707203865,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 20352
+    },
+    {
+      "epoch": 2.0921985815602837,
+      "grad_norm": 0.054826896637678146,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 20355
+    },
+    {
+      "epoch": 2.0925069380203514,
+      "grad_norm": 0.05717878043651581,
+      "learning_rate": 0.01,
+      "loss": 2.0613,
+      "step": 20358
+    },
+    {
+      "epoch": 2.0928152944804195,
+      "grad_norm": 0.09222474694252014,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 20361
+    },
+    {
+      "epoch": 2.093123650940487,
+      "grad_norm": 0.0983637347817421,
+      "learning_rate": 0.01,
+      "loss": 2.0652,
+      "step": 20364
+    },
+    {
+      "epoch": 2.0934320074005552,
+      "grad_norm": 0.05338272079825401,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 20367
+    },
+    {
+      "epoch": 2.093740363860623,
+      "grad_norm": 0.047821927815675735,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 20370
+    },
+    {
+      "epoch": 2.0940487203206906,
+      "grad_norm": 0.10657365620136261,
+      "learning_rate": 0.01,
+      "loss": 2.0533,
+      "step": 20373
+    },
+    {
+      "epoch": 2.0943570767807587,
+      "grad_norm": 0.08650174736976624,
+      "learning_rate": 0.01,
+      "loss": 2.0486,
+      "step": 20376
+    },
+    {
+      "epoch": 2.0946654332408263,
+      "grad_norm": 0.06975332647562027,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 20379
+    },
+    {
+      "epoch": 2.0949737897008944,
+      "grad_norm": 0.08443387597799301,
+      "learning_rate": 0.01,
+      "loss": 2.0529,
+      "step": 20382
+    },
+    {
+      "epoch": 2.095282146160962,
+      "grad_norm": 0.04376668483018875,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 20385
+    },
+    {
+      "epoch": 2.0955905026210297,
+      "grad_norm": 0.1501801759004593,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 20388
+    },
+    {
+      "epoch": 2.095898859081098,
+      "grad_norm": 0.08488426357507706,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 20391
+    },
+    {
+      "epoch": 2.0962072155411655,
+      "grad_norm": 0.05569930747151375,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 20394
+    },
+    {
+      "epoch": 2.0965155720012336,
+      "grad_norm": 0.059826262295246124,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 20397
+    },
+    {
+      "epoch": 2.0968239284613013,
+      "grad_norm": 0.0911981388926506,
+      "learning_rate": 0.01,
+      "loss": 2.0568,
+      "step": 20400
+    },
+    {
+      "epoch": 2.097132284921369,
+      "grad_norm": 0.03921716660261154,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 20403
+    },
+    {
+      "epoch": 2.097440641381437,
+      "grad_norm": 0.07355164736509323,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 20406
+    },
+    {
+      "epoch": 2.0977489978415047,
+      "grad_norm": 0.031198322772979736,
+      "learning_rate": 0.01,
+      "loss": 2.0755,
+      "step": 20409
+    },
+    {
+      "epoch": 2.098057354301573,
+      "grad_norm": 0.07405471056699753,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 20412
+    },
+    {
+      "epoch": 2.0983657107616405,
+      "grad_norm": 0.08160628378391266,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 20415
+    },
+    {
+      "epoch": 2.098674067221708,
+      "grad_norm": 0.11125149577856064,
+      "learning_rate": 0.01,
+      "loss": 2.0612,
+      "step": 20418
+    },
+    {
+      "epoch": 2.0989824236817762,
+      "grad_norm": 0.04484894871711731,
+      "learning_rate": 0.01,
+      "loss": 2.061,
+      "step": 20421
+    },
+    {
+      "epoch": 2.099290780141844,
+      "grad_norm": 0.07138056308031082,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 20424
+    },
+    {
+      "epoch": 2.099599136601912,
+      "grad_norm": 0.0794389471411705,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 20427
+    },
+    {
+      "epoch": 2.0999074930619797,
+      "grad_norm": 0.05673963576555252,
+      "learning_rate": 0.01,
+      "loss": 2.0708,
+      "step": 20430
+    },
+    {
+      "epoch": 2.1002158495220473,
+      "grad_norm": 0.054527074098587036,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 20433
+    },
+    {
+      "epoch": 2.1005242059821154,
+      "grad_norm": 0.04964460805058479,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 20436
+    },
+    {
+      "epoch": 2.100832562442183,
+      "grad_norm": 0.07147829979658127,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 20439
+    },
+    {
+      "epoch": 2.101140918902251,
+      "grad_norm": 0.07474019378423691,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 20442
+    },
+    {
+      "epoch": 2.101449275362319,
+      "grad_norm": 0.06944689154624939,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 20445
+    },
+    {
+      "epoch": 2.1017576318223865,
+      "grad_norm": 0.07336383312940598,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 20448
+    },
+    {
+      "epoch": 2.1020659882824546,
+      "grad_norm": 0.08889491856098175,
+      "learning_rate": 0.01,
+      "loss": 2.0822,
+      "step": 20451
+    },
+    {
+      "epoch": 2.1023743447425223,
+      "grad_norm": 0.07154878228902817,
+      "learning_rate": 0.01,
+      "loss": 2.0672,
+      "step": 20454
+    },
+    {
+      "epoch": 2.1026827012025904,
+      "grad_norm": 0.0471792072057724,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 20457
+    },
+    {
+      "epoch": 2.102991057662658,
+      "grad_norm": 0.10836900025606155,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 20460
+    },
+    {
+      "epoch": 2.1032994141227257,
+      "grad_norm": 0.05292079225182533,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 20463
+    },
+    {
+      "epoch": 2.103607770582794,
+      "grad_norm": 0.13195812702178955,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 20466
+    },
+    {
+      "epoch": 2.1039161270428615,
+      "grad_norm": 0.1316298395395279,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 20469
+    },
+    {
+      "epoch": 2.1042244835029296,
+      "grad_norm": 0.04061713069677353,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 20472
+    },
+    {
+      "epoch": 2.1045328399629972,
+      "grad_norm": 0.09250857681035995,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 20475
+    },
+    {
+      "epoch": 2.104841196423065,
+      "grad_norm": 0.058364611119031906,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 20478
+    },
+    {
+      "epoch": 2.105149552883133,
+      "grad_norm": 0.046974651515483856,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 20481
+    },
+    {
+      "epoch": 2.1054579093432007,
+      "grad_norm": 0.04835136979818344,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 20484
+    },
+    {
+      "epoch": 2.1057662658032688,
+      "grad_norm": 0.04643654450774193,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 20487
+    },
+    {
+      "epoch": 2.1060746222633364,
+      "grad_norm": 0.10667752474546432,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 20490
+    },
+    {
+      "epoch": 2.106382978723404,
+      "grad_norm": 0.07521391659975052,
+      "learning_rate": 0.01,
+      "loss": 2.0552,
+      "step": 20493
+    },
+    {
+      "epoch": 2.106691335183472,
+      "grad_norm": 0.10269229114055634,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 20496
+    },
+    {
+      "epoch": 2.10699969164354,
+      "grad_norm": 0.040783487260341644,
+      "learning_rate": 0.01,
+      "loss": 2.0567,
+      "step": 20499
+    },
+    {
+      "epoch": 2.107308048103608,
+      "grad_norm": 0.05012373626232147,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 20502
+    },
+    {
+      "epoch": 2.1076164045636756,
+      "grad_norm": 0.042675044387578964,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 20505
+    },
+    {
+      "epoch": 2.1079247610237433,
+      "grad_norm": 0.06164225935935974,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 20508
+    },
+    {
+      "epoch": 2.1082331174838114,
+      "grad_norm": 0.06368019431829453,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 20511
+    },
+    {
+      "epoch": 2.108541473943879,
+      "grad_norm": 0.049319278448820114,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 20514
+    },
+    {
+      "epoch": 2.108849830403947,
+      "grad_norm": 0.06070362776517868,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 20517
+    },
+    {
+      "epoch": 2.109158186864015,
+      "grad_norm": 0.07560203224420547,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 20520
+    },
+    {
+      "epoch": 2.1094665433240825,
+      "grad_norm": 0.05563758686184883,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 20523
+    },
+    {
+      "epoch": 2.1097748997841506,
+      "grad_norm": 0.04349389672279358,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 20526
+    },
+    {
+      "epoch": 2.110083256244218,
+      "grad_norm": 0.05849798396229744,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 20529
+    },
+    {
+      "epoch": 2.1103916127042863,
+      "grad_norm": 0.06813669949769974,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 20532
+    },
+    {
+      "epoch": 2.110699969164354,
+      "grad_norm": 0.08951954543590546,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 20535
+    },
+    {
+      "epoch": 2.1110083256244216,
+      "grad_norm": 0.08673957735300064,
+      "learning_rate": 0.01,
+      "loss": 2.0602,
+      "step": 20538
+    },
+    {
+      "epoch": 2.1113166820844897,
+      "grad_norm": 0.06135937571525574,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 20541
+    },
+    {
+      "epoch": 2.1116250385445574,
+      "grad_norm": 0.08979734778404236,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 20544
+    },
+    {
+      "epoch": 2.1119333950046255,
+      "grad_norm": 0.07878229022026062,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 20547
+    },
+    {
+      "epoch": 2.112241751464693,
+      "grad_norm": 0.07120572775602341,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 20550
+    },
+    {
+      "epoch": 2.112550107924761,
+      "grad_norm": 0.11559943854808807,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 20553
+    },
+    {
+      "epoch": 2.112858464384829,
+      "grad_norm": 0.09005344659090042,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 20556
+    },
+    {
+      "epoch": 2.1131668208448966,
+      "grad_norm": 0.05143802613019943,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 20559
+    },
+    {
+      "epoch": 2.1134751773049647,
+      "grad_norm": 0.03870050981640816,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 20562
+    },
+    {
+      "epoch": 2.1137835337650324,
+      "grad_norm": 0.06471268832683563,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 20565
+    },
+    {
+      "epoch": 2.1140918902251,
+      "grad_norm": 0.07047493010759354,
+      "learning_rate": 0.01,
+      "loss": 2.052,
+      "step": 20568
+    },
+    {
+      "epoch": 2.114400246685168,
+      "grad_norm": 0.08189850300550461,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 20571
+    },
+    {
+      "epoch": 2.114708603145236,
+      "grad_norm": 0.04041110724210739,
+      "learning_rate": 0.01,
+      "loss": 2.0604,
+      "step": 20574
+    },
+    {
+      "epoch": 2.115016959605304,
+      "grad_norm": 0.06573761254549026,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 20577
+    },
+    {
+      "epoch": 2.1153253160653716,
+      "grad_norm": 0.06190131977200508,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 20580
+    },
+    {
+      "epoch": 2.115633672525439,
+      "grad_norm": 0.044485028833150864,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 20583
+    },
+    {
+      "epoch": 2.1159420289855073,
+      "grad_norm": 0.03833581507205963,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 20586
+    },
+    {
+      "epoch": 2.116250385445575,
+      "grad_norm": 0.07705090194940567,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 20589
+    },
+    {
+      "epoch": 2.116558741905643,
+      "grad_norm": 0.12635158002376556,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 20592
+    },
+    {
+      "epoch": 2.1168670983657107,
+      "grad_norm": 0.06076742708683014,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 20595
+    },
+    {
+      "epoch": 2.1171754548257784,
+      "grad_norm": 0.06634259968996048,
+      "learning_rate": 0.01,
+      "loss": 2.0603,
+      "step": 20598
+    },
+    {
+      "epoch": 2.1174838112858465,
+      "grad_norm": 0.06167810782790184,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 20601
+    },
+    {
+      "epoch": 2.117792167745914,
+      "grad_norm": 0.0474831759929657,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 20604
+    },
+    {
+      "epoch": 2.1181005242059823,
+      "grad_norm": 0.04799888655543327,
+      "learning_rate": 0.01,
+      "loss": 2.0628,
+      "step": 20607
+    },
+    {
+      "epoch": 2.11840888066605,
+      "grad_norm": 0.055144913494586945,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 20610
+    },
+    {
+      "epoch": 2.1187172371261176,
+      "grad_norm": 0.1116517037153244,
+      "learning_rate": 0.01,
+      "loss": 2.0467,
+      "step": 20613
+    },
+    {
+      "epoch": 2.1190255935861857,
+      "grad_norm": 0.06136411800980568,
+      "learning_rate": 0.01,
+      "loss": 2.0489,
+      "step": 20616
+    },
+    {
+      "epoch": 2.1193339500462534,
+      "grad_norm": 0.09005284309387207,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 20619
+    },
+    {
+      "epoch": 2.1196423065063215,
+      "grad_norm": 0.19345355033874512,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 20622
+    },
+    {
+      "epoch": 2.119950662966389,
+      "grad_norm": 0.14011520147323608,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 20625
+    },
+    {
+      "epoch": 2.120259019426457,
+      "grad_norm": 0.06036897376179695,
+      "learning_rate": 0.01,
+      "loss": 2.0566,
+      "step": 20628
+    },
+    {
+      "epoch": 2.120567375886525,
+      "grad_norm": 0.05589490756392479,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 20631
+    },
+    {
+      "epoch": 2.1208757323465925,
+      "grad_norm": 0.07571965456008911,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 20634
+    },
+    {
+      "epoch": 2.1211840888066607,
+      "grad_norm": 0.05558032542467117,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 20637
+    },
+    {
+      "epoch": 2.1214924452667283,
+      "grad_norm": 0.048844993114471436,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 20640
+    },
+    {
+      "epoch": 2.121800801726796,
+      "grad_norm": 0.04454483836889267,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 20643
+    },
+    {
+      "epoch": 2.122109158186864,
+      "grad_norm": 0.04565678536891937,
+      "learning_rate": 0.01,
+      "loss": 2.0498,
+      "step": 20646
+    },
+    {
+      "epoch": 2.1224175146469317,
+      "grad_norm": 0.04082358628511429,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 20649
+    },
+    {
+      "epoch": 2.122725871107,
+      "grad_norm": 0.03753536194562912,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 20652
+    },
+    {
+      "epoch": 2.1230342275670675,
+      "grad_norm": 0.04898180440068245,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 20655
+    },
+    {
+      "epoch": 2.123342584027135,
+      "grad_norm": 0.0748276561498642,
+      "learning_rate": 0.01,
+      "loss": 2.0632,
+      "step": 20658
+    },
+    {
+      "epoch": 2.1236509404872033,
+      "grad_norm": 0.19312037527561188,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 20661
+    },
+    {
+      "epoch": 2.123959296947271,
+      "grad_norm": 0.1444995105266571,
+      "learning_rate": 0.01,
+      "loss": 2.0713,
+      "step": 20664
+    },
+    {
+      "epoch": 2.124267653407339,
+      "grad_norm": 0.03868752345442772,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 20667
+    },
+    {
+      "epoch": 2.1245760098674067,
+      "grad_norm": 0.04851067438721657,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 20670
+    },
+    {
+      "epoch": 2.1248843663274743,
+      "grad_norm": 0.03709336742758751,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 20673
+    },
+    {
+      "epoch": 2.1251927227875425,
+      "grad_norm": 0.03498173505067825,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 20676
+    },
+    {
+      "epoch": 2.12550107924761,
+      "grad_norm": 0.04838285222649574,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 20679
+    },
+    {
+      "epoch": 2.125809435707678,
+      "grad_norm": 0.04562096297740936,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 20682
+    },
+    {
+      "epoch": 2.126117792167746,
+      "grad_norm": 0.06128044053912163,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 20685
+    },
+    {
+      "epoch": 2.1264261486278135,
+      "grad_norm": 0.04903801530599594,
+      "learning_rate": 0.01,
+      "loss": 2.049,
+      "step": 20688
+    },
+    {
+      "epoch": 2.1267345050878816,
+      "grad_norm": 0.038953814655542374,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 20691
+    },
+    {
+      "epoch": 2.1270428615479493,
+      "grad_norm": 0.06137779355049133,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 20694
+    },
+    {
+      "epoch": 2.1273512180080174,
+      "grad_norm": 0.11357403546571732,
+      "learning_rate": 0.01,
+      "loss": 2.0572,
+      "step": 20697
+    },
+    {
+      "epoch": 2.127659574468085,
+      "grad_norm": 0.16223950684070587,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 20700
+    },
+    {
+      "epoch": 2.127967930928153,
+      "grad_norm": 0.06391070038080215,
+      "learning_rate": 0.01,
+      "loss": 2.0682,
+      "step": 20703
+    },
+    {
+      "epoch": 2.128276287388221,
+      "grad_norm": 0.03951489180326462,
+      "learning_rate": 0.01,
+      "loss": 2.0645,
+      "step": 20706
+    },
+    {
+      "epoch": 2.1285846438482885,
+      "grad_norm": 0.03921591490507126,
+      "learning_rate": 0.01,
+      "loss": 2.0538,
+      "step": 20709
+    },
+    {
+      "epoch": 2.1288930003083566,
+      "grad_norm": 0.0479004830121994,
+      "learning_rate": 0.01,
+      "loss": 2.0587,
+      "step": 20712
+    },
+    {
+      "epoch": 2.1292013567684243,
+      "grad_norm": 0.04370121285319328,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 20715
+    },
+    {
+      "epoch": 2.129509713228492,
+      "grad_norm": 0.05750906467437744,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 20718
+    },
+    {
+      "epoch": 2.12981806968856,
+      "grad_norm": 0.06549614667892456,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 20721
+    },
+    {
+      "epoch": 2.1301264261486277,
+      "grad_norm": 0.05763638764619827,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 20724
+    },
+    {
+      "epoch": 2.130434782608696,
+      "grad_norm": 0.16448546946048737,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 20727
+    },
+    {
+      "epoch": 2.1307431390687634,
+      "grad_norm": 0.057152118533849716,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 20730
+    },
+    {
+      "epoch": 2.1310514955288316,
+      "grad_norm": 0.054196059703826904,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 20733
+    },
+    {
+      "epoch": 2.131359851988899,
+      "grad_norm": 0.03887069597840309,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 20736
+    },
+    {
+      "epoch": 2.131668208448967,
+      "grad_norm": 0.03529683127999306,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 20739
+    },
+    {
+      "epoch": 2.131976564909035,
+      "grad_norm": 0.053463347256183624,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 20742
+    },
+    {
+      "epoch": 2.1322849213691026,
+      "grad_norm": 0.03482777252793312,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 20745
+    },
+    {
+      "epoch": 2.1325932778291703,
+      "grad_norm": 0.056043028831481934,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 20748
+    },
+    {
+      "epoch": 2.1329016342892384,
+      "grad_norm": 0.04648544266819954,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 20751
+    },
+    {
+      "epoch": 2.133209990749306,
+      "grad_norm": 0.0878090187907219,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 20754
+    },
+    {
+      "epoch": 2.133518347209374,
+      "grad_norm": 0.059173766523599625,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 20757
+    },
+    {
+      "epoch": 2.133826703669442,
+      "grad_norm": 0.09119824320077896,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 20760
+    },
+    {
+      "epoch": 2.13413506012951,
+      "grad_norm": 0.04082552343606949,
+      "learning_rate": 0.01,
+      "loss": 2.0501,
+      "step": 20763
+    },
+    {
+      "epoch": 2.1344434165895776,
+      "grad_norm": 0.0831313356757164,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 20766
+    },
+    {
+      "epoch": 2.1347517730496453,
+      "grad_norm": 0.0635252296924591,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 20769
+    },
+    {
+      "epoch": 2.1350601295097134,
+      "grad_norm": 0.09516814351081848,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 20772
+    },
+    {
+      "epoch": 2.135368485969781,
+      "grad_norm": 0.05734236165881157,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 20775
+    },
+    {
+      "epoch": 2.1356768424298487,
+      "grad_norm": 0.07606332749128342,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 20778
+    },
+    {
+      "epoch": 2.135985198889917,
+      "grad_norm": 0.041968777775764465,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 20781
+    },
+    {
+      "epoch": 2.1362935553499844,
+      "grad_norm": 0.10549111664295197,
+      "learning_rate": 0.01,
+      "loss": 2.0695,
+      "step": 20784
+    },
+    {
+      "epoch": 2.1366019118100525,
+      "grad_norm": 0.0783652663230896,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 20787
+    },
+    {
+      "epoch": 2.13691026827012,
+      "grad_norm": 0.08309295773506165,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 20790
+    },
+    {
+      "epoch": 2.1372186247301883,
+      "grad_norm": 0.038408368825912476,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 20793
+    },
+    {
+      "epoch": 2.137526981190256,
+      "grad_norm": 0.1311808079481125,
+      "learning_rate": 0.01,
+      "loss": 2.0608,
+      "step": 20796
+    },
+    {
+      "epoch": 2.1378353376503236,
+      "grad_norm": 0.08983028680086136,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 20799
+    },
+    {
+      "epoch": 2.1381436941103917,
+      "grad_norm": 0.12422356754541397,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 20802
+    },
+    {
+      "epoch": 2.1384520505704594,
+      "grad_norm": 0.0687076672911644,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 20805
+    },
+    {
+      "epoch": 2.138760407030527,
+      "grad_norm": 0.04577179625630379,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 20808
+    },
+    {
+      "epoch": 2.139068763490595,
+      "grad_norm": 0.046294886618852615,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 20811
+    },
+    {
+      "epoch": 2.139377119950663,
+      "grad_norm": 0.048477739095687866,
+      "learning_rate": 0.01,
+      "loss": 2.055,
+      "step": 20814
+    },
+    {
+      "epoch": 2.139685476410731,
+      "grad_norm": 0.035206399857997894,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 20817
+    },
+    {
+      "epoch": 2.1399938328707986,
+      "grad_norm": 0.0451078936457634,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 20820
+    },
+    {
+      "epoch": 2.1403021893308667,
+      "grad_norm": 0.09469619393348694,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 20823
+    },
+    {
+      "epoch": 2.1406105457909343,
+      "grad_norm": 0.03966006636619568,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 20826
+    },
+    {
+      "epoch": 2.140918902251002,
+      "grad_norm": 0.0793539360165596,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 20829
+    },
+    {
+      "epoch": 2.14122725871107,
+      "grad_norm": 0.08547502756118774,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 20832
+    },
+    {
+      "epoch": 2.1415356151711378,
+      "grad_norm": 0.055916544049978256,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 20835
+    },
+    {
+      "epoch": 2.141843971631206,
+      "grad_norm": 0.0369451642036438,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 20838
+    },
+    {
+      "epoch": 2.1421523280912735,
+      "grad_norm": 0.05001531541347504,
+      "learning_rate": 0.01,
+      "loss": 2.06,
+      "step": 20841
+    },
+    {
+      "epoch": 2.142460684551341,
+      "grad_norm": 0.09491688758134842,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 20844
+    },
+    {
+      "epoch": 2.1427690410114093,
+      "grad_norm": 0.05615471303462982,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 20847
+    },
+    {
+      "epoch": 2.143077397471477,
+      "grad_norm": 0.08531193435192108,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 20850
+    },
+    {
+      "epoch": 2.143385753931545,
+      "grad_norm": 0.08138839155435562,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 20853
+    },
+    {
+      "epoch": 2.1436941103916127,
+      "grad_norm": 0.04965364560484886,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 20856
+    },
+    {
+      "epoch": 2.1440024668516804,
+      "grad_norm": 0.07006839662790298,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 20859
+    },
+    {
+      "epoch": 2.1443108233117485,
+      "grad_norm": 0.07477371394634247,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 20862
+    },
+    {
+      "epoch": 2.144619179771816,
+      "grad_norm": 0.07817060500383377,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 20865
+    },
+    {
+      "epoch": 2.1449275362318843,
+      "grad_norm": 0.04393570497632027,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 20868
+    },
+    {
+      "epoch": 2.145235892691952,
+      "grad_norm": 0.0777692198753357,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 20871
+    },
+    {
+      "epoch": 2.1455442491520196,
+      "grad_norm": 0.05267953500151634,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 20874
+    },
+    {
+      "epoch": 2.1458526056120877,
+      "grad_norm": 0.053839076310396194,
+      "learning_rate": 0.01,
+      "loss": 2.0593,
+      "step": 20877
+    },
+    {
+      "epoch": 2.1461609620721553,
+      "grad_norm": 0.05793678015470505,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 20880
+    },
+    {
+      "epoch": 2.1464693185322234,
+      "grad_norm": 0.045539624989032745,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 20883
+    },
+    {
+      "epoch": 2.146777674992291,
+      "grad_norm": 0.062209442257881165,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 20886
+    },
+    {
+      "epoch": 2.1470860314523588,
+      "grad_norm": 0.11053802073001862,
+      "learning_rate": 0.01,
+      "loss": 2.0764,
+      "step": 20889
+    },
+    {
+      "epoch": 2.147394387912427,
+      "grad_norm": 0.037597738206386566,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 20892
+    },
+    {
+      "epoch": 2.1477027443724945,
+      "grad_norm": 0.050160784274339676,
+      "learning_rate": 0.01,
+      "loss": 2.0797,
+      "step": 20895
+    },
+    {
+      "epoch": 2.1480111008325626,
+      "grad_norm": 0.053438689559698105,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 20898
+    },
+    {
+      "epoch": 2.1483194572926303,
+      "grad_norm": 0.06175898388028145,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 20901
+    },
+    {
+      "epoch": 2.148627813752698,
+      "grad_norm": 0.062430836260318756,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 20904
+    },
+    {
+      "epoch": 2.148936170212766,
+      "grad_norm": 0.09347888082265854,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 20907
+    },
+    {
+      "epoch": 2.1492445266728337,
+      "grad_norm": 0.13494189083576202,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 20910
+    },
+    {
+      "epoch": 2.149552883132902,
+      "grad_norm": 0.06735777854919434,
+      "learning_rate": 0.01,
+      "loss": 2.0799,
+      "step": 20913
+    },
+    {
+      "epoch": 2.1498612395929695,
+      "grad_norm": 0.08108525723218918,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 20916
+    },
+    {
+      "epoch": 2.150169596053037,
+      "grad_norm": 0.043221328407526016,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 20919
+    },
+    {
+      "epoch": 2.1504779525131053,
+      "grad_norm": 0.046568017452955246,
+      "learning_rate": 0.01,
+      "loss": 2.0393,
+      "step": 20922
+    },
+    {
+      "epoch": 2.150786308973173,
+      "grad_norm": 0.040786582976579666,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 20925
+    },
+    {
+      "epoch": 2.151094665433241,
+      "grad_norm": 0.043856412172317505,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 20928
+    },
+    {
+      "epoch": 2.1514030218933087,
+      "grad_norm": 0.03569160774350166,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 20931
+    },
+    {
+      "epoch": 2.1517113783533763,
+      "grad_norm": 0.059598151594400406,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 20934
+    },
+    {
+      "epoch": 2.1520197348134444,
+      "grad_norm": 0.12518006563186646,
+      "learning_rate": 0.01,
+      "loss": 2.0389,
+      "step": 20937
+    },
+    {
+      "epoch": 2.152328091273512,
+      "grad_norm": 0.04442603886127472,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 20940
+    },
+    {
+      "epoch": 2.15263644773358,
+      "grad_norm": 0.1175926923751831,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 20943
+    },
+    {
+      "epoch": 2.152944804193648,
+      "grad_norm": 0.09045865386724472,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 20946
+    },
+    {
+      "epoch": 2.1532531606537155,
+      "grad_norm": 0.07519534230232239,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 20949
+    },
+    {
+      "epoch": 2.1535615171137836,
+      "grad_norm": 0.03551176190376282,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 20952
+    },
+    {
+      "epoch": 2.1538698735738513,
+      "grad_norm": 0.11210478097200394,
+      "learning_rate": 0.01,
+      "loss": 2.0453,
+      "step": 20955
+    },
+    {
+      "epoch": 2.1541782300339194,
+      "grad_norm": 0.09921340644359589,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 20958
+    },
+    {
+      "epoch": 2.154486586493987,
+      "grad_norm": 0.04131443426012993,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 20961
+    },
+    {
+      "epoch": 2.1547949429540547,
+      "grad_norm": 0.07579990476369858,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 20964
+    },
+    {
+      "epoch": 2.155103299414123,
+      "grad_norm": 0.04955494403839111,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 20967
+    },
+    {
+      "epoch": 2.1554116558741905,
+      "grad_norm": 0.038999974727630615,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 20970
+    },
+    {
+      "epoch": 2.1557200123342586,
+      "grad_norm": 0.04656311497092247,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 20973
+    },
+    {
+      "epoch": 2.1560283687943262,
+      "grad_norm": 0.08853477984666824,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 20976
+    },
+    {
+      "epoch": 2.156336725254394,
+      "grad_norm": 0.07657559216022491,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 20979
+    },
+    {
+      "epoch": 2.156645081714462,
+      "grad_norm": 0.03593476489186287,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 20982
+    },
+    {
+      "epoch": 2.1569534381745297,
+      "grad_norm": 0.08656775951385498,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 20985
+    },
+    {
+      "epoch": 2.1572617946345978,
+      "grad_norm": 0.07760220021009445,
+      "learning_rate": 0.01,
+      "loss": 2.068,
+      "step": 20988
+    },
+    {
+      "epoch": 2.1575701510946654,
+      "grad_norm": 0.09167666733264923,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 20991
+    },
+    {
+      "epoch": 2.157878507554733,
+      "grad_norm": 0.052361227571964264,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 20994
+    },
+    {
+      "epoch": 2.158186864014801,
+      "grad_norm": 0.05431961268186569,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 20997
+    },
+    {
+      "epoch": 2.158495220474869,
+      "grad_norm": 0.03956061974167824,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 21000
+    },
+    {
+      "epoch": 2.158803576934937,
+      "grad_norm": 0.03906402364373207,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 21003
+    },
+    {
+      "epoch": 2.1591119333950046,
+      "grad_norm": 0.07876679301261902,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 21006
+    },
+    {
+      "epoch": 2.1594202898550723,
+      "grad_norm": 0.051768235862255096,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 21009
+    },
+    {
+      "epoch": 2.1597286463151404,
+      "grad_norm": 0.05242472141981125,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 21012
+    },
+    {
+      "epoch": 2.160037002775208,
+      "grad_norm": 0.07014864683151245,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 21015
+    },
+    {
+      "epoch": 2.160345359235276,
+      "grad_norm": 0.1024969145655632,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 21018
+    },
+    {
+      "epoch": 2.160653715695344,
+      "grad_norm": 0.059565525501966476,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 21021
+    },
+    {
+      "epoch": 2.1609620721554115,
+      "grad_norm": 0.04438649117946625,
+      "learning_rate": 0.01,
+      "loss": 2.0736,
+      "step": 21024
+    },
+    {
+      "epoch": 2.1612704286154796,
+      "grad_norm": 0.06552638858556747,
+      "learning_rate": 0.01,
+      "loss": 2.0449,
+      "step": 21027
+    },
+    {
+      "epoch": 2.1615787850755472,
+      "grad_norm": 0.044327035546302795,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 21030
+    },
+    {
+      "epoch": 2.1618871415356153,
+      "grad_norm": 0.04915094003081322,
+      "learning_rate": 0.01,
+      "loss": 2.0634,
+      "step": 21033
+    },
+    {
+      "epoch": 2.162195497995683,
+      "grad_norm": 0.052909743040800095,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 21036
+    },
+    {
+      "epoch": 2.1625038544557507,
+      "grad_norm": 0.09883973747491837,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 21039
+    },
+    {
+      "epoch": 2.1628122109158188,
+      "grad_norm": 0.07719819992780685,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 21042
+    },
+    {
+      "epoch": 2.1631205673758864,
+      "grad_norm": 0.09743590652942657,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 21045
+    },
+    {
+      "epoch": 2.1634289238359545,
+      "grad_norm": 0.06583153456449509,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 21048
+    },
+    {
+      "epoch": 2.163737280296022,
+      "grad_norm": 0.10464660078287125,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 21051
+    },
+    {
+      "epoch": 2.16404563675609,
+      "grad_norm": 0.057555560022592545,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 21054
+    },
+    {
+      "epoch": 2.164353993216158,
+      "grad_norm": 0.04753732308745384,
+      "learning_rate": 0.01,
+      "loss": 2.054,
+      "step": 21057
+    },
+    {
+      "epoch": 2.1646623496762256,
+      "grad_norm": 0.05236852541565895,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 21060
+    },
+    {
+      "epoch": 2.1649707061362937,
+      "grad_norm": 0.05231897532939911,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 21063
+    },
+    {
+      "epoch": 2.1652790625963614,
+      "grad_norm": 0.058431778103113174,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 21066
+    },
+    {
+      "epoch": 2.165587419056429,
+      "grad_norm": 0.04048459604382515,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 21069
+    },
+    {
+      "epoch": 2.165895775516497,
+      "grad_norm": 0.0941488966345787,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 21072
+    },
+    {
+      "epoch": 2.166204131976565,
+      "grad_norm": 0.04498032480478287,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 21075
+    },
+    {
+      "epoch": 2.166512488436633,
+      "grad_norm": 0.03668253496289253,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 21078
+    },
+    {
+      "epoch": 2.1668208448967006,
+      "grad_norm": 0.06661085039377213,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 21081
+    },
+    {
+      "epoch": 2.1671292013567682,
+      "grad_norm": 0.06425791233778,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 21084
+    },
+    {
+      "epoch": 2.1674375578168363,
+      "grad_norm": 0.046152301132678986,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 21087
+    },
+    {
+      "epoch": 2.167745914276904,
+      "grad_norm": 0.12584535777568817,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 21090
+    },
+    {
+      "epoch": 2.168054270736972,
+      "grad_norm": 0.04133312404155731,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 21093
+    },
+    {
+      "epoch": 2.1683626271970398,
+      "grad_norm": 0.04181768745183945,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 21096
+    },
+    {
+      "epoch": 2.1686709836571074,
+      "grad_norm": 0.04648204892873764,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 21099
+    },
+    {
+      "epoch": 2.1689793401171755,
+      "grad_norm": 0.044040024280548096,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 21102
+    },
+    {
+      "epoch": 2.169287696577243,
+      "grad_norm": 0.04033771902322769,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 21105
+    },
+    {
+      "epoch": 2.1695960530373113,
+      "grad_norm": 0.04693743214011192,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 21108
+    },
+    {
+      "epoch": 2.169904409497379,
+      "grad_norm": 0.07494813948869705,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 21111
+    },
+    {
+      "epoch": 2.1702127659574466,
+      "grad_norm": 0.09254445135593414,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 21114
+    },
+    {
+      "epoch": 2.1705211224175147,
+      "grad_norm": 0.13350296020507812,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 21117
+    },
+    {
+      "epoch": 2.1708294788775824,
+      "grad_norm": 0.051686085760593414,
+      "learning_rate": 0.01,
+      "loss": 2.0507,
+      "step": 21120
+    },
+    {
+      "epoch": 2.1711378353376505,
+      "grad_norm": 0.043330930173397064,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 21123
+    },
+    {
+      "epoch": 2.171446191797718,
+      "grad_norm": 0.05219477042555809,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 21126
+    },
+    {
+      "epoch": 2.171754548257786,
+      "grad_norm": 0.04195651784539223,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 21129
+    },
+    {
+      "epoch": 2.172062904717854,
+      "grad_norm": 0.15802520513534546,
+      "learning_rate": 0.01,
+      "loss": 2.0609,
+      "step": 21132
+    },
+    {
+      "epoch": 2.1723712611779216,
+      "grad_norm": 0.08619748800992966,
+      "learning_rate": 0.01,
+      "loss": 2.0542,
+      "step": 21135
+    },
+    {
+      "epoch": 2.1726796176379897,
+      "grad_norm": 0.08852280676364899,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 21138
+    },
+    {
+      "epoch": 2.1729879740980573,
+      "grad_norm": 0.040646422654390335,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 21141
+    },
+    {
+      "epoch": 2.173296330558125,
+      "grad_norm": 0.06913924962282181,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 21144
+    },
+    {
+      "epoch": 2.173604687018193,
+      "grad_norm": 0.05461576208472252,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 21147
+    },
+    {
+      "epoch": 2.1739130434782608,
+      "grad_norm": 0.08613748103380203,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 21150
+    },
+    {
+      "epoch": 2.174221399938329,
+      "grad_norm": 0.06459856033325195,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 21153
+    },
+    {
+      "epoch": 2.1745297563983965,
+      "grad_norm": 0.08450223505496979,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 21156
+    },
+    {
+      "epoch": 2.174838112858464,
+      "grad_norm": 0.10049585998058319,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 21159
+    },
+    {
+      "epoch": 2.1751464693185323,
+      "grad_norm": 0.06031005084514618,
+      "learning_rate": 0.01,
+      "loss": 2.0668,
+      "step": 21162
+    },
+    {
+      "epoch": 2.1754548257786,
+      "grad_norm": 0.06921012699604034,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 21165
+    },
+    {
+      "epoch": 2.175763182238668,
+      "grad_norm": 0.044479697942733765,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 21168
+    },
+    {
+      "epoch": 2.1760715386987357,
+      "grad_norm": 0.0857187807559967,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 21171
+    },
+    {
+      "epoch": 2.176379895158804,
+      "grad_norm": 0.057432882487773895,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 21174
+    },
+    {
+      "epoch": 2.1766882516188715,
+      "grad_norm": 0.0905427411198616,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 21177
+    },
+    {
+      "epoch": 2.176996608078939,
+      "grad_norm": 0.05289644002914429,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 21180
+    },
+    {
+      "epoch": 2.1773049645390072,
+      "grad_norm": 0.06351148337125778,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 21183
+    },
+    {
+      "epoch": 2.177613320999075,
+      "grad_norm": 0.1098824068903923,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 21186
+    },
+    {
+      "epoch": 2.1779216774591426,
+      "grad_norm": 0.03698734566569328,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 21189
+    },
+    {
+      "epoch": 2.1782300339192107,
+      "grad_norm": 0.09595025330781937,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 21192
+    },
+    {
+      "epoch": 2.1785383903792783,
+      "grad_norm": 0.05725647136569023,
+      "learning_rate": 0.01,
+      "loss": 2.0568,
+      "step": 21195
+    },
+    {
+      "epoch": 2.1788467468393464,
+      "grad_norm": 0.06952492892742157,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 21198
+    },
+    {
+      "epoch": 2.179155103299414,
+      "grad_norm": 0.06831461936235428,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 21201
+    },
+    {
+      "epoch": 2.179463459759482,
+      "grad_norm": 0.05125569924712181,
+      "learning_rate": 0.01,
+      "loss": 2.0593,
+      "step": 21204
+    },
+    {
+      "epoch": 2.17977181621955,
+      "grad_norm": 0.053290076553821564,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 21207
+    },
+    {
+      "epoch": 2.1800801726796175,
+      "grad_norm": 0.05718007683753967,
+      "learning_rate": 0.01,
+      "loss": 2.0472,
+      "step": 21210
+    },
+    {
+      "epoch": 2.1803885291396856,
+      "grad_norm": 0.0895228236913681,
+      "learning_rate": 0.01,
+      "loss": 2.0673,
+      "step": 21213
+    },
+    {
+      "epoch": 2.1806968855997533,
+      "grad_norm": 0.05205732583999634,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 21216
+    },
+    {
+      "epoch": 2.181005242059821,
+      "grad_norm": 0.05785641819238663,
+      "learning_rate": 0.01,
+      "loss": 2.0557,
+      "step": 21219
+    },
+    {
+      "epoch": 2.181313598519889,
+      "grad_norm": 0.0756557285785675,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 21222
+    },
+    {
+      "epoch": 2.1816219549799567,
+      "grad_norm": 0.07016823440790176,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 21225
+    },
+    {
+      "epoch": 2.181930311440025,
+      "grad_norm": 0.06859459728002548,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 21228
+    },
+    {
+      "epoch": 2.1822386679000925,
+      "grad_norm": 0.14140251278877258,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 21231
+    },
+    {
+      "epoch": 2.1825470243601606,
+      "grad_norm": 0.0544712096452713,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 21234
+    },
+    {
+      "epoch": 2.1828553808202282,
+      "grad_norm": 0.04346593841910362,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 21237
+    },
+    {
+      "epoch": 2.183163737280296,
+      "grad_norm": 0.04261700063943863,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 21240
+    },
+    {
+      "epoch": 2.183472093740364,
+      "grad_norm": 0.04763154685497284,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 21243
+    },
+    {
+      "epoch": 2.1837804502004317,
+      "grad_norm": 0.06876801699399948,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 21246
+    },
+    {
+      "epoch": 2.1840888066604993,
+      "grad_norm": 0.07819974422454834,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 21249
+    },
+    {
+      "epoch": 2.1843971631205674,
+      "grad_norm": 0.06239667907357216,
+      "learning_rate": 0.01,
+      "loss": 2.0268,
+      "step": 21252
+    },
+    {
+      "epoch": 2.184705519580635,
+      "grad_norm": 0.1095786988735199,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 21255
+    },
+    {
+      "epoch": 2.185013876040703,
+      "grad_norm": 0.055070340633392334,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 21258
+    },
+    {
+      "epoch": 2.185322232500771,
+      "grad_norm": 0.08038482069969177,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 21261
+    },
+    {
+      "epoch": 2.185630588960839,
+      "grad_norm": 0.06929390877485275,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 21264
+    },
+    {
+      "epoch": 2.1859389454209066,
+      "grad_norm": 0.054179031401872635,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 21267
+    },
+    {
+      "epoch": 2.1862473018809743,
+      "grad_norm": 0.06956303864717484,
+      "learning_rate": 0.01,
+      "loss": 2.0586,
+      "step": 21270
+    },
+    {
+      "epoch": 2.1865556583410424,
+      "grad_norm": 0.14279653131961823,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 21273
+    },
+    {
+      "epoch": 2.18686401480111,
+      "grad_norm": 0.04268357530236244,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 21276
+    },
+    {
+      "epoch": 2.1871723712611777,
+      "grad_norm": 0.05189354717731476,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 21279
+    },
+    {
+      "epoch": 2.187480727721246,
+      "grad_norm": 0.05047158896923065,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 21282
+    },
+    {
+      "epoch": 2.1877890841813135,
+      "grad_norm": 0.041077565401792526,
+      "learning_rate": 0.01,
+      "loss": 2.0558,
+      "step": 21285
+    },
+    {
+      "epoch": 2.1880974406413816,
+      "grad_norm": 0.03880000859498978,
+      "learning_rate": 0.01,
+      "loss": 2.0472,
+      "step": 21288
+    },
+    {
+      "epoch": 2.1884057971014492,
+      "grad_norm": 0.1096898689866066,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 21291
+    },
+    {
+      "epoch": 2.1887141535615173,
+      "grad_norm": 0.04502374678850174,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 21294
+    },
+    {
+      "epoch": 2.189022510021585,
+      "grad_norm": 0.037158042192459106,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 21297
+    },
+    {
+      "epoch": 2.1893308664816526,
+      "grad_norm": 0.03381425887346268,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 21300
+    },
+    {
+      "epoch": 2.1896392229417208,
+      "grad_norm": 0.05572035536170006,
+      "learning_rate": 0.01,
+      "loss": 2.0575,
+      "step": 21303
+    },
+    {
+      "epoch": 2.1899475794017884,
+      "grad_norm": 0.06287326663732529,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 21306
+    },
+    {
+      "epoch": 2.190255935861856,
+      "grad_norm": 0.07691732794046402,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 21309
+    },
+    {
+      "epoch": 2.190564292321924,
+      "grad_norm": 0.058651309460401535,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 21312
+    },
+    {
+      "epoch": 2.190872648781992,
+      "grad_norm": 0.033279962837696075,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 21315
+    },
+    {
+      "epoch": 2.19118100524206,
+      "grad_norm": 0.08742087334394455,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 21318
+    },
+    {
+      "epoch": 2.1914893617021276,
+      "grad_norm": 0.0864923968911171,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 21321
+    },
+    {
+      "epoch": 2.1917977181621957,
+      "grad_norm": 0.10759606957435608,
+      "learning_rate": 0.01,
+      "loss": 2.0582,
+      "step": 21324
+    },
+    {
+      "epoch": 2.1921060746222634,
+      "grad_norm": 0.058335281908512115,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 21327
+    },
+    {
+      "epoch": 2.192414431082331,
+      "grad_norm": 0.04506481811404228,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 21330
+    },
+    {
+      "epoch": 2.192722787542399,
+      "grad_norm": 0.0454195998609066,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 21333
+    },
+    {
+      "epoch": 2.193031144002467,
+      "grad_norm": 0.051547158509492874,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 21336
+    },
+    {
+      "epoch": 2.193339500462535,
+      "grad_norm": 0.09826447069644928,
+      "learning_rate": 0.01,
+      "loss": 2.0518,
+      "step": 21339
+    },
+    {
+      "epoch": 2.1936478569226026,
+      "grad_norm": 0.05799272283911705,
+      "learning_rate": 0.01,
+      "loss": 2.0649,
+      "step": 21342
+    },
+    {
+      "epoch": 2.19395621338267,
+      "grad_norm": 0.12493482232093811,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 21345
+    },
+    {
+      "epoch": 2.1942645698427383,
+      "grad_norm": 0.04930184409022331,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 21348
+    },
+    {
+      "epoch": 2.194572926302806,
+      "grad_norm": 0.04257272928953171,
+      "learning_rate": 0.01,
+      "loss": 2.0579,
+      "step": 21351
+    },
+    {
+      "epoch": 2.194881282762874,
+      "grad_norm": 0.04625258222222328,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 21354
+    },
+    {
+      "epoch": 2.1951896392229417,
+      "grad_norm": 0.04295830428600311,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 21357
+    },
+    {
+      "epoch": 2.1954979956830094,
+      "grad_norm": 0.21663565933704376,
+      "learning_rate": 0.01,
+      "loss": 2.0489,
+      "step": 21360
+    },
+    {
+      "epoch": 2.1958063521430775,
+      "grad_norm": 0.15513652563095093,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 21363
+    },
+    {
+      "epoch": 2.196114708603145,
+      "grad_norm": 0.08245841413736343,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 21366
+    },
+    {
+      "epoch": 2.1964230650632133,
+      "grad_norm": 0.03768035024404526,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 21369
+    },
+    {
+      "epoch": 2.196731421523281,
+      "grad_norm": 0.0586925707757473,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 21372
+    },
+    {
+      "epoch": 2.1970397779833486,
+      "grad_norm": 0.045760899782180786,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 21375
+    },
+    {
+      "epoch": 2.1973481344434167,
+      "grad_norm": 0.04357283189892769,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 21378
+    },
+    {
+      "epoch": 2.1976564909034844,
+      "grad_norm": 0.04477246478199959,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 21381
+    },
+    {
+      "epoch": 2.1979648473635525,
+      "grad_norm": 0.06785521656274796,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 21384
+    },
+    {
+      "epoch": 2.19827320382362,
+      "grad_norm": 0.04677508771419525,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 21387
+    },
+    {
+      "epoch": 2.198581560283688,
+      "grad_norm": 0.049355942755937576,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 21390
+    },
+    {
+      "epoch": 2.198889916743756,
+      "grad_norm": 0.11005277186632156,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 21393
+    },
+    {
+      "epoch": 2.1991982732038236,
+      "grad_norm": 0.10170245915651321,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 21396
+    },
+    {
+      "epoch": 2.1995066296638917,
+      "grad_norm": 0.045158207416534424,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 21399
+    },
+    {
+      "epoch": 2.1998149861239593,
+      "grad_norm": 0.0780436098575592,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 21402
+    },
+    {
+      "epoch": 2.200123342584027,
+      "grad_norm": 0.06062453240156174,
+      "learning_rate": 0.01,
+      "loss": 2.0437,
+      "step": 21405
+    },
+    {
+      "epoch": 2.200431699044095,
+      "grad_norm": 0.09208519756793976,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 21408
+    },
+    {
+      "epoch": 2.2007400555041627,
+      "grad_norm": 0.05279922112822533,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 21411
+    },
+    {
+      "epoch": 2.201048411964231,
+      "grad_norm": 0.0831415057182312,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 21414
+    },
+    {
+      "epoch": 2.2013567684242985,
+      "grad_norm": 0.06481669098138809,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 21417
+    },
+    {
+      "epoch": 2.201665124884366,
+      "grad_norm": 0.08297551423311234,
+      "learning_rate": 0.01,
+      "loss": 2.0741,
+      "step": 21420
+    },
+    {
+      "epoch": 2.2019734813444343,
+      "grad_norm": 0.06962350755929947,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 21423
+    },
+    {
+      "epoch": 2.202281837804502,
+      "grad_norm": 0.08055757731199265,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 21426
+    },
+    {
+      "epoch": 2.20259019426457,
+      "grad_norm": 0.09755895286798477,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 21429
+    },
+    {
+      "epoch": 2.2028985507246377,
+      "grad_norm": 0.04400679096579552,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 21432
+    },
+    {
+      "epoch": 2.2032069071847054,
+      "grad_norm": 0.04204344376921654,
+      "learning_rate": 0.01,
+      "loss": 2.063,
+      "step": 21435
+    },
+    {
+      "epoch": 2.2035152636447735,
+      "grad_norm": 0.029974184930324554,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 21438
+    },
+    {
+      "epoch": 2.203823620104841,
+      "grad_norm": 0.07232589274644852,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 21441
+    },
+    {
+      "epoch": 2.2041319765649092,
+      "grad_norm": 0.06404844671487808,
+      "learning_rate": 0.01,
+      "loss": 2.0596,
+      "step": 21444
+    },
+    {
+      "epoch": 2.204440333024977,
+      "grad_norm": 0.08751700818538666,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 21447
+    },
+    {
+      "epoch": 2.2047486894850445,
+      "grad_norm": 0.0371503084897995,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 21450
+    },
+    {
+      "epoch": 2.2050570459451126,
+      "grad_norm": 0.06034844368696213,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 21453
+    },
+    {
+      "epoch": 2.2053654024051803,
+      "grad_norm": 0.04261939972639084,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 21456
+    },
+    {
+      "epoch": 2.2056737588652484,
+      "grad_norm": 0.0612785667181015,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 21459
+    },
+    {
+      "epoch": 2.205982115325316,
+      "grad_norm": 0.05828654393553734,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 21462
+    },
+    {
+      "epoch": 2.2062904717853837,
+      "grad_norm": 0.06474754214286804,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 21465
+    },
+    {
+      "epoch": 2.206598828245452,
+      "grad_norm": 0.04646962508559227,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 21468
+    },
+    {
+      "epoch": 2.2069071847055195,
+      "grad_norm": 0.07051596790552139,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 21471
+    },
+    {
+      "epoch": 2.2072155411655876,
+      "grad_norm": 0.05658755078911781,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 21474
+    },
+    {
+      "epoch": 2.2075238976256553,
+      "grad_norm": 0.039348311722278595,
+      "learning_rate": 0.01,
+      "loss": 2.0587,
+      "step": 21477
+    },
+    {
+      "epoch": 2.207832254085723,
+      "grad_norm": 0.030550241470336914,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 21480
+    },
+    {
+      "epoch": 2.208140610545791,
+      "grad_norm": 0.11341346800327301,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 21483
+    },
+    {
+      "epoch": 2.2084489670058587,
+      "grad_norm": 0.07061111927032471,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 21486
+    },
+    {
+      "epoch": 2.208757323465927,
+      "grad_norm": 0.10256624966859818,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 21489
+    },
+    {
+      "epoch": 2.2090656799259945,
+      "grad_norm": 0.06658724695444107,
+      "learning_rate": 0.01,
+      "loss": 2.0544,
+      "step": 21492
+    },
+    {
+      "epoch": 2.209374036386062,
+      "grad_norm": 0.12220150977373123,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 21495
+    },
+    {
+      "epoch": 2.20968239284613,
+      "grad_norm": 0.05570116639137268,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 21498
+    },
+    {
+      "epoch": 2.209990749306198,
+      "grad_norm": 0.04273837059736252,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 21501
+    },
+    {
+      "epoch": 2.210299105766266,
+      "grad_norm": 0.04138748720288277,
+      "learning_rate": 0.01,
+      "loss": 2.0497,
+      "step": 21504
+    },
+    {
+      "epoch": 2.2106074622263336,
+      "grad_norm": 0.08082164078950882,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 21507
+    },
+    {
+      "epoch": 2.2109158186864013,
+      "grad_norm": 0.06663049012422562,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 21510
+    },
+    {
+      "epoch": 2.2112241751464694,
+      "grad_norm": 0.0444667711853981,
+      "learning_rate": 0.01,
+      "loss": 2.0469,
+      "step": 21513
+    },
+    {
+      "epoch": 2.211532531606537,
+      "grad_norm": 0.04407314583659172,
+      "learning_rate": 0.01,
+      "loss": 2.0667,
+      "step": 21516
+    },
+    {
+      "epoch": 2.211840888066605,
+      "grad_norm": 0.03877383843064308,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 21519
+    },
+    {
+      "epoch": 2.212149244526673,
+      "grad_norm": 0.059297189116477966,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 21522
+    },
+    {
+      "epoch": 2.2124576009867405,
+      "grad_norm": 0.06609878689050674,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 21525
+    },
+    {
+      "epoch": 2.2127659574468086,
+      "grad_norm": 0.06935823708772659,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 21528
+    },
+    {
+      "epoch": 2.2130743139068763,
+      "grad_norm": 0.07610715180635452,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 21531
+    },
+    {
+      "epoch": 2.2133826703669444,
+      "grad_norm": 0.10587569326162338,
+      "learning_rate": 0.01,
+      "loss": 2.0636,
+      "step": 21534
+    },
+    {
+      "epoch": 2.213691026827012,
+      "grad_norm": 0.05116620659828186,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 21537
+    },
+    {
+      "epoch": 2.2139993832870797,
+      "grad_norm": 0.03773776814341545,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 21540
+    },
+    {
+      "epoch": 2.214307739747148,
+      "grad_norm": 0.05412130430340767,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 21543
+    },
+    {
+      "epoch": 2.2146160962072154,
+      "grad_norm": 0.03664164990186691,
+      "learning_rate": 0.01,
+      "loss": 2.0535,
+      "step": 21546
+    },
+    {
+      "epoch": 2.2149244526672835,
+      "grad_norm": 0.04415920004248619,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 21549
+    },
+    {
+      "epoch": 2.215232809127351,
+      "grad_norm": 0.05737615004181862,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 21552
+    },
+    {
+      "epoch": 2.215541165587419,
+      "grad_norm": 0.032385457307100296,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 21555
+    },
+    {
+      "epoch": 2.215849522047487,
+      "grad_norm": 0.0982925221323967,
+      "learning_rate": 0.01,
+      "loss": 2.0653,
+      "step": 21558
+    },
+    {
+      "epoch": 2.2161578785075546,
+      "grad_norm": 0.03911735862493515,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 21561
+    },
+    {
+      "epoch": 2.2164662349676227,
+      "grad_norm": 0.07814744859933853,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 21564
+    },
+    {
+      "epoch": 2.2167745914276904,
+      "grad_norm": 0.05368256941437721,
+      "learning_rate": 0.01,
+      "loss": 2.0689,
+      "step": 21567
+    },
+    {
+      "epoch": 2.217082947887758,
+      "grad_norm": 0.046178530901670456,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 21570
+    },
+    {
+      "epoch": 2.217391304347826,
+      "grad_norm": 0.047109801322221756,
+      "learning_rate": 0.01,
+      "loss": 2.0721,
+      "step": 21573
+    },
+    {
+      "epoch": 2.217699660807894,
+      "grad_norm": 0.05443650484085083,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 21576
+    },
+    {
+      "epoch": 2.218008017267962,
+      "grad_norm": 0.13156400620937347,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 21579
+    },
+    {
+      "epoch": 2.2183163737280296,
+      "grad_norm": 0.04178638756275177,
+      "learning_rate": 0.01,
+      "loss": 2.0517,
+      "step": 21582
+    },
+    {
+      "epoch": 2.2186247301880972,
+      "grad_norm": 0.042627740651369095,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 21585
+    },
+    {
+      "epoch": 2.2189330866481654,
+      "grad_norm": 0.05318658426403999,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 21588
+    },
+    {
+      "epoch": 2.219241443108233,
+      "grad_norm": 0.061288055032491684,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 21591
+    },
+    {
+      "epoch": 2.219549799568301,
+      "grad_norm": 0.06663260608911514,
+      "learning_rate": 0.01,
+      "loss": 2.0549,
+      "step": 21594
+    },
+    {
+      "epoch": 2.219858156028369,
+      "grad_norm": 0.04567466303706169,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 21597
+    },
+    {
+      "epoch": 2.2201665124884364,
+      "grad_norm": 0.12566886842250824,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 21600
+    },
+    {
+      "epoch": 2.2204748689485045,
+      "grad_norm": 0.03933155536651611,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 21603
+    },
+    {
+      "epoch": 2.220783225408572,
+      "grad_norm": 0.04617391526699066,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 21606
+    },
+    {
+      "epoch": 2.2210915818686403,
+      "grad_norm": 0.05472411960363388,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 21609
+    },
+    {
+      "epoch": 2.221399938328708,
+      "grad_norm": 0.05556654930114746,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 21612
+    },
+    {
+      "epoch": 2.2217082947887756,
+      "grad_norm": 0.05096900090575218,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 21615
+    },
+    {
+      "epoch": 2.2220166512488437,
+      "grad_norm": 0.039425577968358994,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 21618
+    },
+    {
+      "epoch": 2.2223250077089114,
+      "grad_norm": 0.05080854892730713,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 21621
+    },
+    {
+      "epoch": 2.2226333641689795,
+      "grad_norm": 0.03824865445494652,
+      "learning_rate": 0.01,
+      "loss": 2.0497,
+      "step": 21624
+    },
+    {
+      "epoch": 2.222941720629047,
+      "grad_norm": 0.03814932331442833,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 21627
+    },
+    {
+      "epoch": 2.223250077089115,
+      "grad_norm": 0.057797808200120926,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 21630
+    },
+    {
+      "epoch": 2.223558433549183,
+      "grad_norm": 0.06596177071332932,
+      "learning_rate": 0.01,
+      "loss": 2.0572,
+      "step": 21633
+    },
+    {
+      "epoch": 2.2238667900092506,
+      "grad_norm": 0.13438032567501068,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 21636
+    },
+    {
+      "epoch": 2.2241751464693187,
+      "grad_norm": 0.06200256571173668,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 21639
+    },
+    {
+      "epoch": 2.2244835029293863,
+      "grad_norm": 0.06471807509660721,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 21642
+    },
+    {
+      "epoch": 2.224791859389454,
+      "grad_norm": 0.06439206004142761,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 21645
+    },
+    {
+      "epoch": 2.225100215849522,
+      "grad_norm": 0.0768360123038292,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 21648
+    },
+    {
+      "epoch": 2.2254085723095898,
+      "grad_norm": 0.10393831878900528,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 21651
+    },
+    {
+      "epoch": 2.225716928769658,
+      "grad_norm": 0.03999519720673561,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 21654
+    },
+    {
+      "epoch": 2.2260252852297255,
+      "grad_norm": 0.04620358720421791,
+      "learning_rate": 0.01,
+      "loss": 2.0576,
+      "step": 21657
+    },
+    {
+      "epoch": 2.226333641689793,
+      "grad_norm": 0.05364964157342911,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 21660
+    },
+    {
+      "epoch": 2.2266419981498613,
+      "grad_norm": 0.049792349338531494,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 21663
+    },
+    {
+      "epoch": 2.226950354609929,
+      "grad_norm": 0.0651509091258049,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 21666
+    },
+    {
+      "epoch": 2.227258711069997,
+      "grad_norm": 0.035688720643520355,
+      "learning_rate": 0.01,
+      "loss": 2.0509,
+      "step": 21669
+    },
+    {
+      "epoch": 2.2275670675300647,
+      "grad_norm": 0.06412792950868607,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 21672
+    },
+    {
+      "epoch": 2.2278754239901324,
+      "grad_norm": 0.08088821917772293,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 21675
+    },
+    {
+      "epoch": 2.2281837804502005,
+      "grad_norm": 0.060282256454229355,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 21678
+    },
+    {
+      "epoch": 2.228492136910268,
+      "grad_norm": 0.09816791117191315,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 21681
+    },
+    {
+      "epoch": 2.2288004933703363,
+      "grad_norm": 0.06672241538763046,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 21684
+    },
+    {
+      "epoch": 2.229108849830404,
+      "grad_norm": 0.06586040556430817,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 21687
+    },
+    {
+      "epoch": 2.2294172062904716,
+      "grad_norm": 0.0655137374997139,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 21690
+    },
+    {
+      "epoch": 2.2297255627505397,
+      "grad_norm": 0.08448750525712967,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 21693
+    },
+    {
+      "epoch": 2.2300339192106073,
+      "grad_norm": 0.05828822776675224,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 21696
+    },
+    {
+      "epoch": 2.2303422756706754,
+      "grad_norm": 0.0816628485918045,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 21699
+    },
+    {
+      "epoch": 2.230650632130743,
+      "grad_norm": 0.038368817418813705,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 21702
+    },
+    {
+      "epoch": 2.230958988590811,
+      "grad_norm": 0.08584286272525787,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 21705
+    },
+    {
+      "epoch": 2.231267345050879,
+      "grad_norm": 0.08528412878513336,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 21708
+    },
+    {
+      "epoch": 2.2315757015109465,
+      "grad_norm": 0.07158780843019485,
+      "learning_rate": 0.01,
+      "loss": 2.0437,
+      "step": 21711
+    },
+    {
+      "epoch": 2.2318840579710146,
+      "grad_norm": 0.07810889184474945,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 21714
+    },
+    {
+      "epoch": 2.2321924144310823,
+      "grad_norm": 0.07769618928432465,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 21717
+    },
+    {
+      "epoch": 2.23250077089115,
+      "grad_norm": 0.04515406861901283,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 21720
+    },
+    {
+      "epoch": 2.232809127351218,
+      "grad_norm": 0.11805865168571472,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 21723
+    },
+    {
+      "epoch": 2.2331174838112857,
+      "grad_norm": 0.11899860948324203,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 21726
+    },
+    {
+      "epoch": 2.233425840271354,
+      "grad_norm": 0.03309144452214241,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 21729
+    },
+    {
+      "epoch": 2.2337341967314215,
+      "grad_norm": 0.03389447182416916,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 21732
+    },
+    {
+      "epoch": 2.2340425531914896,
+      "grad_norm": 0.03472166880965233,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 21735
+    },
+    {
+      "epoch": 2.2343509096515572,
+      "grad_norm": 0.11815425753593445,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 21738
+    },
+    {
+      "epoch": 2.234659266111625,
+      "grad_norm": 0.08007802814245224,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 21741
+    },
+    {
+      "epoch": 2.234967622571693,
+      "grad_norm": 0.05053863301873207,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 21744
+    },
+    {
+      "epoch": 2.2352759790317607,
+      "grad_norm": 0.12116878479719162,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 21747
+    },
+    {
+      "epoch": 2.2355843354918283,
+      "grad_norm": 0.06733332574367523,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 21750
+    },
+    {
+      "epoch": 2.2358926919518964,
+      "grad_norm": 0.05141659080982208,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 21753
+    },
+    {
+      "epoch": 2.236201048411964,
+      "grad_norm": 0.03695325180888176,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 21756
+    },
+    {
+      "epoch": 2.236509404872032,
+      "grad_norm": 0.033388737589120865,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 21759
+    },
+    {
+      "epoch": 2.2368177613321,
+      "grad_norm": 0.03913639858365059,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 21762
+    },
+    {
+      "epoch": 2.237126117792168,
+      "grad_norm": 0.048513270914554596,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 21765
+    },
+    {
+      "epoch": 2.2374344742522356,
+      "grad_norm": 0.061340250074863434,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 21768
+    },
+    {
+      "epoch": 2.2377428307123033,
+      "grad_norm": 0.042784787714481354,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 21771
+    },
+    {
+      "epoch": 2.2380511871723714,
+      "grad_norm": 0.04124082252383232,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 21774
+    },
+    {
+      "epoch": 2.238359543632439,
+      "grad_norm": 0.03749304264783859,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 21777
+    },
+    {
+      "epoch": 2.2386679000925067,
+      "grad_norm": 0.12073452770709991,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 21780
+    },
+    {
+      "epoch": 2.238976256552575,
+      "grad_norm": 0.06613750755786896,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 21783
+    },
+    {
+      "epoch": 2.2392846130126425,
+      "grad_norm": 0.11562148481607437,
+      "learning_rate": 0.01,
+      "loss": 2.0542,
+      "step": 21786
+    },
+    {
+      "epoch": 2.2395929694727106,
+      "grad_norm": 0.06355132162570953,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 21789
+    },
+    {
+      "epoch": 2.2399013259327782,
+      "grad_norm": 0.09228016436100006,
+      "learning_rate": 0.01,
+      "loss": 2.066,
+      "step": 21792
+    },
+    {
+      "epoch": 2.2402096823928463,
+      "grad_norm": 0.07648111879825592,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 21795
+    },
+    {
+      "epoch": 2.240518038852914,
+      "grad_norm": 0.0650177150964737,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 21798
+    },
+    {
+      "epoch": 2.2408263953129817,
+      "grad_norm": 0.09675853699445724,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 21801
+    },
+    {
+      "epoch": 2.2411347517730498,
+      "grad_norm": 0.045449864119291306,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 21804
+    },
+    {
+      "epoch": 2.2414431082331174,
+      "grad_norm": 0.0982547327876091,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 21807
+    },
+    {
+      "epoch": 2.241751464693185,
+      "grad_norm": 0.060201246291399,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 21810
+    },
+    {
+      "epoch": 2.242059821153253,
+      "grad_norm": 0.04503628611564636,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 21813
+    },
+    {
+      "epoch": 2.242368177613321,
+      "grad_norm": 0.04854295030236244,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 21816
+    },
+    {
+      "epoch": 2.242676534073389,
+      "grad_norm": 0.12228553742170334,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 21819
+    },
+    {
+      "epoch": 2.2429848905334566,
+      "grad_norm": 0.10491406172513962,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 21822
+    },
+    {
+      "epoch": 2.2432932469935247,
+      "grad_norm": 0.1188177838921547,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 21825
+    },
+    {
+      "epoch": 2.2436016034535924,
+      "grad_norm": 0.10877541452646255,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 21828
+    },
+    {
+      "epoch": 2.24390995991366,
+      "grad_norm": 0.03375115245580673,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 21831
+    },
+    {
+      "epoch": 2.244218316373728,
+      "grad_norm": 0.07588639855384827,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 21834
+    },
+    {
+      "epoch": 2.244526672833796,
+      "grad_norm": 0.04841979220509529,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 21837
+    },
+    {
+      "epoch": 2.2448350292938635,
+      "grad_norm": 0.041367027908563614,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 21840
+    },
+    {
+      "epoch": 2.2451433857539316,
+      "grad_norm": 0.044548399746418,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 21843
+    },
+    {
+      "epoch": 2.2454517422139992,
+      "grad_norm": 0.049429114907979965,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 21846
+    },
+    {
+      "epoch": 2.2457600986740673,
+      "grad_norm": 0.07232322543859482,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 21849
+    },
+    {
+      "epoch": 2.246068455134135,
+      "grad_norm": 0.05866502597928047,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 21852
+    },
+    {
+      "epoch": 2.246376811594203,
+      "grad_norm": 0.08406642079353333,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 21855
+    },
+    {
+      "epoch": 2.2466851680542708,
+      "grad_norm": 0.08692225068807602,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 21858
+    },
+    {
+      "epoch": 2.2469935245143384,
+      "grad_norm": 0.09182562679052353,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 21861
+    },
+    {
+      "epoch": 2.2473018809744065,
+      "grad_norm": 0.07580506801605225,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 21864
+    },
+    {
+      "epoch": 2.247610237434474,
+      "grad_norm": 0.09604424238204956,
+      "learning_rate": 0.01,
+      "loss": 2.054,
+      "step": 21867
+    },
+    {
+      "epoch": 2.2479185938945423,
+      "grad_norm": 0.05383382737636566,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 21870
+    },
+    {
+      "epoch": 2.24822695035461,
+      "grad_norm": 0.035557687282562256,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 21873
+    },
+    {
+      "epoch": 2.2485353068146776,
+      "grad_norm": 0.0330585315823555,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 21876
+    },
+    {
+      "epoch": 2.2488436632747457,
+      "grad_norm": 0.07125352323055267,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 21879
+    },
+    {
+      "epoch": 2.2491520197348134,
+      "grad_norm": 0.07722420245409012,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 21882
+    },
+    {
+      "epoch": 2.2494603761948815,
+      "grad_norm": 0.05086112394928932,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 21885
+    },
+    {
+      "epoch": 2.249768732654949,
+      "grad_norm": 0.055818263441324234,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 21888
+    },
+    {
+      "epoch": 2.250077089115017,
+      "grad_norm": 0.06524482369422913,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 21891
+    },
+    {
+      "epoch": 2.250385445575085,
+      "grad_norm": 0.051445212215185165,
+      "learning_rate": 0.01,
+      "loss": 2.0594,
+      "step": 21894
+    },
+    {
+      "epoch": 2.2506938020351526,
+      "grad_norm": 0.057153645902872086,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 21897
+    },
+    {
+      "epoch": 2.2510021584952202,
+      "grad_norm": 0.08242445439100266,
+      "learning_rate": 0.01,
+      "loss": 2.0465,
+      "step": 21900
+    },
+    {
+      "epoch": 2.2513105149552883,
+      "grad_norm": 0.06253752112388611,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 21903
+    },
+    {
+      "epoch": 2.251618871415356,
+      "grad_norm": 0.05388238653540611,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 21906
+    },
+    {
+      "epoch": 2.251927227875424,
+      "grad_norm": 0.05570909380912781,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 21909
+    },
+    {
+      "epoch": 2.2522355843354918,
+      "grad_norm": 0.038132745772600174,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 21912
+    },
+    {
+      "epoch": 2.25254394079556,
+      "grad_norm": 0.07434429973363876,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 21915
+    },
+    {
+      "epoch": 2.2528522972556275,
+      "grad_norm": 0.05396199971437454,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 21918
+    },
+    {
+      "epoch": 2.253160653715695,
+      "grad_norm": 0.06541625410318375,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 21921
+    },
+    {
+      "epoch": 2.2534690101757633,
+      "grad_norm": 0.07078956067562103,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 21924
+    },
+    {
+      "epoch": 2.253777366635831,
+      "grad_norm": 0.06990516930818558,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 21927
+    },
+    {
+      "epoch": 2.254085723095899,
+      "grad_norm": 0.049635425209999084,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 21930
+    },
+    {
+      "epoch": 2.2543940795559667,
+      "grad_norm": 0.04051049426198006,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 21933
+    },
+    {
+      "epoch": 2.2547024360160344,
+      "grad_norm": 0.11082252860069275,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 21936
+    },
+    {
+      "epoch": 2.2550107924761025,
+      "grad_norm": 0.0768849179148674,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 21939
+    },
+    {
+      "epoch": 2.25531914893617,
+      "grad_norm": 0.0420587994158268,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 21942
+    },
+    {
+      "epoch": 2.2556275053962382,
+      "grad_norm": 0.03924650698900223,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 21945
+    },
+    {
+      "epoch": 2.255935861856306,
+      "grad_norm": 0.058580152690410614,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 21948
+    },
+    {
+      "epoch": 2.2562442183163736,
+      "grad_norm": 0.050332482904195786,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 21951
+    },
+    {
+      "epoch": 2.2565525747764417,
+      "grad_norm": 0.0472397580742836,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 21954
+    },
+    {
+      "epoch": 2.2568609312365093,
+      "grad_norm": 0.0500427670776844,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 21957
+    },
+    {
+      "epoch": 2.2571692876965774,
+      "grad_norm": 0.08899175375699997,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 21960
+    },
+    {
+      "epoch": 2.257477644156645,
+      "grad_norm": 0.06647983938455582,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 21963
+    },
+    {
+      "epoch": 2.2577860006167128,
+      "grad_norm": 0.12703396379947662,
+      "learning_rate": 0.01,
+      "loss": 2.0532,
+      "step": 21966
+    },
+    {
+      "epoch": 2.258094357076781,
+      "grad_norm": 0.06023648753762245,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 21969
+    },
+    {
+      "epoch": 2.2584027135368485,
+      "grad_norm": 0.05025608092546463,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 21972
+    },
+    {
+      "epoch": 2.2587110699969166,
+      "grad_norm": 0.043917398899793625,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 21975
+    },
+    {
+      "epoch": 2.2590194264569843,
+      "grad_norm": 0.03574421629309654,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 21978
+    },
+    {
+      "epoch": 2.259327782917052,
+      "grad_norm": 0.03546285256743431,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 21981
+    },
+    {
+      "epoch": 2.25963613937712,
+      "grad_norm": 0.039241861552000046,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 21984
+    },
+    {
+      "epoch": 2.2599444958371877,
+      "grad_norm": 0.09643664956092834,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 21987
+    },
+    {
+      "epoch": 2.260252852297256,
+      "grad_norm": 0.08918357640504837,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 21990
+    },
+    {
+      "epoch": 2.2605612087573235,
+      "grad_norm": 0.05018826946616173,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 21993
+    },
+    {
+      "epoch": 2.260869565217391,
+      "grad_norm": 0.05288619175553322,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 21996
+    },
+    {
+      "epoch": 2.2611779216774592,
+      "grad_norm": 0.053800489753484726,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 21999
+    },
+    {
+      "epoch": 2.261486278137527,
+      "grad_norm": 0.11081501096487045,
+      "learning_rate": 0.01,
+      "loss": 2.0517,
+      "step": 22002
+    },
+    {
+      "epoch": 2.261794634597595,
+      "grad_norm": 0.045440319925546646,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 22005
+    },
+    {
+      "epoch": 2.2621029910576627,
+      "grad_norm": 0.07007669657468796,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 22008
+    },
+    {
+      "epoch": 2.2624113475177303,
+      "grad_norm": 0.04412767291069031,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 22011
+    },
+    {
+      "epoch": 2.2627197039777984,
+      "grad_norm": 0.034647636115550995,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 22014
+    },
+    {
+      "epoch": 2.263028060437866,
+      "grad_norm": 0.04766898602247238,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 22017
+    },
+    {
+      "epoch": 2.263336416897934,
+      "grad_norm": 0.0641711950302124,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 22020
+    },
+    {
+      "epoch": 2.263644773358002,
+      "grad_norm": 0.05316673591732979,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 22023
+    },
+    {
+      "epoch": 2.2639531298180695,
+      "grad_norm": 0.03488560765981674,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 22026
+    },
+    {
+      "epoch": 2.2642614862781376,
+      "grad_norm": 0.05415144935250282,
+      "learning_rate": 0.01,
+      "loss": 2.0551,
+      "step": 22029
+    },
+    {
+      "epoch": 2.2645698427382053,
+      "grad_norm": 0.08091796189546585,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 22032
+    },
+    {
+      "epoch": 2.2648781991982734,
+      "grad_norm": 0.0850050300359726,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 22035
+    },
+    {
+      "epoch": 2.265186555658341,
+      "grad_norm": 0.13837046921253204,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 22038
+    },
+    {
+      "epoch": 2.2654949121184087,
+      "grad_norm": 0.05676966533064842,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 22041
+    },
+    {
+      "epoch": 2.265803268578477,
+      "grad_norm": 0.1171020120382309,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 22044
+    },
+    {
+      "epoch": 2.2661116250385445,
+      "grad_norm": 0.04077135771512985,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 22047
+    },
+    {
+      "epoch": 2.2664199814986126,
+      "grad_norm": 0.03570380434393883,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 22050
+    },
+    {
+      "epoch": 2.2667283379586802,
+      "grad_norm": 0.0496290847659111,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 22053
+    },
+    {
+      "epoch": 2.267036694418748,
+      "grad_norm": 0.06205829232931137,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 22056
+    },
+    {
+      "epoch": 2.267345050878816,
+      "grad_norm": 0.04645274206995964,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 22059
+    },
+    {
+      "epoch": 2.2676534073388837,
+      "grad_norm": 0.04779735952615738,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 22062
+    },
+    {
+      "epoch": 2.2679617637989518,
+      "grad_norm": 0.05166240781545639,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 22065
+    },
+    {
+      "epoch": 2.2682701202590194,
+      "grad_norm": 0.14004342257976532,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 22068
+    },
+    {
+      "epoch": 2.268578476719087,
+      "grad_norm": 0.06366194784641266,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 22071
+    },
+    {
+      "epoch": 2.268886833179155,
+      "grad_norm": 0.04914015159010887,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 22074
+    },
+    {
+      "epoch": 2.269195189639223,
+      "grad_norm": 0.07121274620294571,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 22077
+    },
+    {
+      "epoch": 2.269503546099291,
+      "grad_norm": 0.048453398048877716,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 22080
+    },
+    {
+      "epoch": 2.2698119025593586,
+      "grad_norm": 0.03777848929166794,
+      "learning_rate": 0.01,
+      "loss": 2.0643,
+      "step": 22083
+    },
+    {
+      "epoch": 2.2701202590194263,
+      "grad_norm": 0.04788368567824364,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 22086
+    },
+    {
+      "epoch": 2.2704286154794944,
+      "grad_norm": 0.036985646933317184,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 22089
+    },
+    {
+      "epoch": 2.270736971939562,
+      "grad_norm": 0.08654552698135376,
+      "learning_rate": 0.01,
+      "loss": 2.0268,
+      "step": 22092
+    },
+    {
+      "epoch": 2.27104532839963,
+      "grad_norm": 0.12740878760814667,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 22095
+    },
+    {
+      "epoch": 2.271353684859698,
+      "grad_norm": 0.058829743415117264,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 22098
+    },
+    {
+      "epoch": 2.2716620413197655,
+      "grad_norm": 0.04173153638839722,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 22101
+    },
+    {
+      "epoch": 2.2719703977798336,
+      "grad_norm": 0.04551135376095772,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 22104
+    },
+    {
+      "epoch": 2.272278754239901,
+      "grad_norm": 0.04404577612876892,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 22107
+    },
+    {
+      "epoch": 2.2725871106999693,
+      "grad_norm": 0.033306755125522614,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 22110
+    },
+    {
+      "epoch": 2.272895467160037,
+      "grad_norm": 0.03450062498450279,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 22113
+    },
+    {
+      "epoch": 2.273203823620105,
+      "grad_norm": 0.04665246978402138,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 22116
+    },
+    {
+      "epoch": 2.2735121800801728,
+      "grad_norm": 0.04256023094058037,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 22119
+    },
+    {
+      "epoch": 2.2738205365402404,
+      "grad_norm": 0.04408838599920273,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 22122
+    },
+    {
+      "epoch": 2.2741288930003085,
+      "grad_norm": 0.17131304740905762,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 22125
+    },
+    {
+      "epoch": 2.274437249460376,
+      "grad_norm": 0.12661625444889069,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 22128
+    },
+    {
+      "epoch": 2.274745605920444,
+      "grad_norm": 0.1100076287984848,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 22131
+    },
+    {
+      "epoch": 2.275053962380512,
+      "grad_norm": 0.06313984096050262,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 22134
+    },
+    {
+      "epoch": 2.2753623188405796,
+      "grad_norm": 0.04563833773136139,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 22137
+    },
+    {
+      "epoch": 2.2756706753006477,
+      "grad_norm": 0.034519702196121216,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 22140
+    },
+    {
+      "epoch": 2.2759790317607154,
+      "grad_norm": 0.033708199858665466,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 22143
+    },
+    {
+      "epoch": 2.2762873882207835,
+      "grad_norm": 0.036958202719688416,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 22146
+    },
+    {
+      "epoch": 2.276595744680851,
+      "grad_norm": 0.052069418132305145,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 22149
+    },
+    {
+      "epoch": 2.276904101140919,
+      "grad_norm": 0.03941415995359421,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 22152
+    },
+    {
+      "epoch": 2.277212457600987,
+      "grad_norm": 0.041270628571510315,
+      "learning_rate": 0.01,
+      "loss": 2.0574,
+      "step": 22155
+    },
+    {
+      "epoch": 2.2775208140610546,
+      "grad_norm": 0.05398832634091377,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 22158
+    },
+    {
+      "epoch": 2.277829170521122,
+      "grad_norm": 0.048906486481428146,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 22161
+    },
+    {
+      "epoch": 2.2781375269811903,
+      "grad_norm": 0.15073837339878082,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 22164
+    },
+    {
+      "epoch": 2.278445883441258,
+      "grad_norm": 0.06301066279411316,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 22167
+    },
+    {
+      "epoch": 2.278754239901326,
+      "grad_norm": 0.07525215297937393,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 22170
+    },
+    {
+      "epoch": 2.2790625963613937,
+      "grad_norm": 0.059356629848480225,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 22173
+    },
+    {
+      "epoch": 2.279370952821462,
+      "grad_norm": 0.04340675100684166,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 22176
+    },
+    {
+      "epoch": 2.2796793092815295,
+      "grad_norm": 0.052859678864479065,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 22179
+    },
+    {
+      "epoch": 2.279987665741597,
+      "grad_norm": 0.05917971953749657,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 22182
+    },
+    {
+      "epoch": 2.2802960222016653,
+      "grad_norm": 0.04111889749765396,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 22185
+    },
+    {
+      "epoch": 2.280604378661733,
+      "grad_norm": 0.06371071189641953,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 22188
+    },
+    {
+      "epoch": 2.2809127351218006,
+      "grad_norm": 0.039862968027591705,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 22191
+    },
+    {
+      "epoch": 2.2812210915818687,
+      "grad_norm": 0.0944151058793068,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 22194
+    },
+    {
+      "epoch": 2.2815294480419364,
+      "grad_norm": 0.041564084589481354,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 22197
+    },
+    {
+      "epoch": 2.2818378045020045,
+      "grad_norm": 0.1109161302447319,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 22200
+    },
+    {
+      "epoch": 2.282146160962072,
+      "grad_norm": 0.14463242888450623,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 22203
+    },
+    {
+      "epoch": 2.2824545174221402,
+      "grad_norm": 0.06254401057958603,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 22206
+    },
+    {
+      "epoch": 2.282762873882208,
+      "grad_norm": 0.03546387329697609,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 22209
+    },
+    {
+      "epoch": 2.2830712303422755,
+      "grad_norm": 0.041129205375909805,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 22212
+    },
+    {
+      "epoch": 2.2833795868023437,
+      "grad_norm": 0.048473335802555084,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 22215
+    },
+    {
+      "epoch": 2.2836879432624113,
+      "grad_norm": 0.0633682832121849,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 22218
+    },
+    {
+      "epoch": 2.283996299722479,
+      "grad_norm": 0.06431914120912552,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 22221
+    },
+    {
+      "epoch": 2.284304656182547,
+      "grad_norm": 0.09050017595291138,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 22224
+    },
+    {
+      "epoch": 2.2846130126426147,
+      "grad_norm": 0.07363586872816086,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 22227
+    },
+    {
+      "epoch": 2.284921369102683,
+      "grad_norm": 0.07984726130962372,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 22230
+    },
+    {
+      "epoch": 2.2852297255627505,
+      "grad_norm": 0.0845586284995079,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 22233
+    },
+    {
+      "epoch": 2.2855380820228186,
+      "grad_norm": 0.13392622768878937,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 22236
+    },
+    {
+      "epoch": 2.2858464384828863,
+      "grad_norm": 0.06791893392801285,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 22239
+    },
+    {
+      "epoch": 2.286154794942954,
+      "grad_norm": 0.044396497309207916,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 22242
+    },
+    {
+      "epoch": 2.286463151403022,
+      "grad_norm": 0.04964762553572655,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 22245
+    },
+    {
+      "epoch": 2.2867715078630897,
+      "grad_norm": 0.06517814844846725,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 22248
+    },
+    {
+      "epoch": 2.2870798643231574,
+      "grad_norm": 0.043722327798604965,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 22251
+    },
+    {
+      "epoch": 2.2873882207832255,
+      "grad_norm": 0.06852597743272781,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 22254
+    },
+    {
+      "epoch": 2.287696577243293,
+      "grad_norm": 0.13437853753566742,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 22257
+    },
+    {
+      "epoch": 2.288004933703361,
+      "grad_norm": 0.03924357146024704,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 22260
+    },
+    {
+      "epoch": 2.288313290163429,
+      "grad_norm": 0.039702124893665314,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 22263
+    },
+    {
+      "epoch": 2.288621646623497,
+      "grad_norm": 0.05187009647488594,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 22266
+    },
+    {
+      "epoch": 2.2889300030835646,
+      "grad_norm": 0.0682847648859024,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 22269
+    },
+    {
+      "epoch": 2.2892383595436323,
+      "grad_norm": 0.046652939170598984,
+      "learning_rate": 0.01,
+      "loss": 2.0481,
+      "step": 22272
+    },
+    {
+      "epoch": 2.2895467160037004,
+      "grad_norm": 0.04348958283662796,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 22275
+    },
+    {
+      "epoch": 2.289855072463768,
+      "grad_norm": 0.05141732096672058,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 22278
+    },
+    {
+      "epoch": 2.2901634289238357,
+      "grad_norm": 0.04899610951542854,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 22281
+    },
+    {
+      "epoch": 2.290471785383904,
+      "grad_norm": 0.05614173039793968,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 22284
+    },
+    {
+      "epoch": 2.2907801418439715,
+      "grad_norm": 0.08255594223737717,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 22287
+    },
+    {
+      "epoch": 2.2910884983040396,
+      "grad_norm": 0.12105574458837509,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 22290
+    },
+    {
+      "epoch": 2.2913968547641073,
+      "grad_norm": 0.07021155953407288,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 22293
+    },
+    {
+      "epoch": 2.2917052112241754,
+      "grad_norm": 0.08364082872867584,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 22296
+    },
+    {
+      "epoch": 2.292013567684243,
+      "grad_norm": 0.1123763918876648,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 22299
+    },
+    {
+      "epoch": 2.2923219241443107,
+      "grad_norm": 0.06539402157068253,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 22302
+    },
+    {
+      "epoch": 2.292630280604379,
+      "grad_norm": 0.08749070018529892,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 22305
+    },
+    {
+      "epoch": 2.2929386370644464,
+      "grad_norm": 0.05580052733421326,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 22308
+    },
+    {
+      "epoch": 2.293246993524514,
+      "grad_norm": 0.07866258919239044,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 22311
+    },
+    {
+      "epoch": 2.293555349984582,
+      "grad_norm": 0.06253038346767426,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 22314
+    },
+    {
+      "epoch": 2.29386370644465,
+      "grad_norm": 0.0412861630320549,
+      "learning_rate": 0.01,
+      "loss": 2.0504,
+      "step": 22317
+    },
+    {
+      "epoch": 2.294172062904718,
+      "grad_norm": 0.05568886548280716,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 22320
+    },
+    {
+      "epoch": 2.2944804193647856,
+      "grad_norm": 0.060274332761764526,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 22323
+    },
+    {
+      "epoch": 2.2947887758248537,
+      "grad_norm": 0.09785215556621552,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 22326
+    },
+    {
+      "epoch": 2.2950971322849214,
+      "grad_norm": 0.06171462684869766,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 22329
+    },
+    {
+      "epoch": 2.295405488744989,
+      "grad_norm": 0.1034122183918953,
+      "learning_rate": 0.01,
+      "loss": 2.0493,
+      "step": 22332
+    },
+    {
+      "epoch": 2.295713845205057,
+      "grad_norm": 0.04855991527438164,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 22335
+    },
+    {
+      "epoch": 2.296022201665125,
+      "grad_norm": 0.05000189319252968,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 22338
+    },
+    {
+      "epoch": 2.2963305581251925,
+      "grad_norm": 0.031126966699957848,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 22341
+    },
+    {
+      "epoch": 2.2966389145852606,
+      "grad_norm": 0.08688051998615265,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 22344
+    },
+    {
+      "epoch": 2.2969472710453283,
+      "grad_norm": 0.08918049931526184,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 22347
+    },
+    {
+      "epoch": 2.2972556275053964,
+      "grad_norm": 0.0592593289911747,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 22350
+    },
+    {
+      "epoch": 2.297563983965464,
+      "grad_norm": 0.11501101404428482,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 22353
+    },
+    {
+      "epoch": 2.297872340425532,
+      "grad_norm": 0.12966668605804443,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 22356
+    },
+    {
+      "epoch": 2.2981806968856,
+      "grad_norm": 0.08889560401439667,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 22359
+    },
+    {
+      "epoch": 2.2984890533456674,
+      "grad_norm": 0.07578025758266449,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 22362
+    },
+    {
+      "epoch": 2.2987974098057355,
+      "grad_norm": 0.04283175244927406,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 22365
+    },
+    {
+      "epoch": 2.299105766265803,
+      "grad_norm": 0.038169119507074356,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 22368
+    },
+    {
+      "epoch": 2.299414122725871,
+      "grad_norm": 0.07047520577907562,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 22371
+    },
+    {
+      "epoch": 2.299722479185939,
+      "grad_norm": 0.06367038190364838,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 22374
+    },
+    {
+      "epoch": 2.3000308356460066,
+      "grad_norm": 0.06601911783218384,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 22377
+    },
+    {
+      "epoch": 2.3003391921060747,
+      "grad_norm": 0.06333800405263901,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 22380
+    },
+    {
+      "epoch": 2.3006475485661424,
+      "grad_norm": 0.0635596215724945,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 22383
+    },
+    {
+      "epoch": 2.3009559050262105,
+      "grad_norm": 0.09258962422609329,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 22386
+    },
+    {
+      "epoch": 2.301264261486278,
+      "grad_norm": 0.04462733119726181,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 22389
+    },
+    {
+      "epoch": 2.301572617946346,
+      "grad_norm": 0.04802479222416878,
+      "learning_rate": 0.01,
+      "loss": 2.0557,
+      "step": 22392
+    },
+    {
+      "epoch": 2.301880974406414,
+      "grad_norm": 0.0396190769970417,
+      "learning_rate": 0.01,
+      "loss": 2.0543,
+      "step": 22395
+    },
+    {
+      "epoch": 2.3021893308664816,
+      "grad_norm": 0.05610959604382515,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 22398
+    },
+    {
+      "epoch": 2.3024976873265492,
+      "grad_norm": 0.03454664349555969,
+      "learning_rate": 0.01,
+      "loss": 2.0592,
+      "step": 22401
+    },
+    {
+      "epoch": 2.3028060437866174,
+      "grad_norm": 0.049920883029699326,
+      "learning_rate": 0.01,
+      "loss": 2.0533,
+      "step": 22404
+    },
+    {
+      "epoch": 2.303114400246685,
+      "grad_norm": 0.04548259451985359,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 22407
+    },
+    {
+      "epoch": 2.303422756706753,
+      "grad_norm": 0.07624956220388412,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 22410
+    },
+    {
+      "epoch": 2.3037311131668208,
+      "grad_norm": 0.1172046959400177,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 22413
+    },
+    {
+      "epoch": 2.304039469626889,
+      "grad_norm": 0.07652290165424347,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 22416
+    },
+    {
+      "epoch": 2.3043478260869565,
+      "grad_norm": 0.07458405196666718,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 22419
+    },
+    {
+      "epoch": 2.304656182547024,
+      "grad_norm": 0.08384053409099579,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 22422
+    },
+    {
+      "epoch": 2.3049645390070923,
+      "grad_norm": 0.0492565892636776,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 22425
+    },
+    {
+      "epoch": 2.30527289546716,
+      "grad_norm": 0.05125366151332855,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 22428
+    },
+    {
+      "epoch": 2.3055812519272276,
+      "grad_norm": 0.03920417279005051,
+      "learning_rate": 0.01,
+      "loss": 2.0471,
+      "step": 22431
+    },
+    {
+      "epoch": 2.3058896083872957,
+      "grad_norm": 0.07206401228904724,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 22434
+    },
+    {
+      "epoch": 2.3061979648473634,
+      "grad_norm": 0.08888402581214905,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 22437
+    },
+    {
+      "epoch": 2.3065063213074315,
+      "grad_norm": 0.1788371056318283,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 22440
+    },
+    {
+      "epoch": 2.306814677767499,
+      "grad_norm": 0.11030272394418716,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 22443
+    },
+    {
+      "epoch": 2.3071230342275673,
+      "grad_norm": 0.08602694422006607,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 22446
+    },
+    {
+      "epoch": 2.307431390687635,
+      "grad_norm": 0.07269123196601868,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 22449
+    },
+    {
+      "epoch": 2.3077397471477026,
+      "grad_norm": 0.043196290731430054,
+      "learning_rate": 0.01,
+      "loss": 2.0617,
+      "step": 22452
+    },
+    {
+      "epoch": 2.3080481036077707,
+      "grad_norm": 0.03932726010680199,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 22455
+    },
+    {
+      "epoch": 2.3083564600678383,
+      "grad_norm": 0.07751597464084625,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 22458
+    },
+    {
+      "epoch": 2.3086648165279064,
+      "grad_norm": 0.06549305468797684,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 22461
+    },
+    {
+      "epoch": 2.308973172987974,
+      "grad_norm": 0.05819348245859146,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 22464
+    },
+    {
+      "epoch": 2.3092815294480418,
+      "grad_norm": 0.04251622408628464,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 22467
+    },
+    {
+      "epoch": 2.30958988590811,
+      "grad_norm": 0.03997926414012909,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 22470
+    },
+    {
+      "epoch": 2.3098982423681775,
+      "grad_norm": 0.060847315937280655,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 22473
+    },
+    {
+      "epoch": 2.3102065988282456,
+      "grad_norm": 0.188557431101799,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 22476
+    },
+    {
+      "epoch": 2.3105149552883133,
+      "grad_norm": 0.050608254969120026,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 22479
+    },
+    {
+      "epoch": 2.310823311748381,
+      "grad_norm": 0.0930873304605484,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 22482
+    },
+    {
+      "epoch": 2.311131668208449,
+      "grad_norm": 0.05231544002890587,
+      "learning_rate": 0.01,
+      "loss": 2.0646,
+      "step": 22485
+    },
+    {
+      "epoch": 2.3114400246685167,
+      "grad_norm": 0.038996320217847824,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 22488
+    },
+    {
+      "epoch": 2.311748381128585,
+      "grad_norm": 0.04804287478327751,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 22491
+    },
+    {
+      "epoch": 2.3120567375886525,
+      "grad_norm": 0.05822224169969559,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 22494
+    },
+    {
+      "epoch": 2.31236509404872,
+      "grad_norm": 0.04677027836441994,
+      "learning_rate": 0.01,
+      "loss": 2.0552,
+      "step": 22497
+    },
+    {
+      "epoch": 2.3126734505087883,
+      "grad_norm": 0.05270523950457573,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 22500
+    },
+    {
+      "epoch": 2.312981806968856,
+      "grad_norm": 0.04068690538406372,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 22503
+    },
+    {
+      "epoch": 2.313290163428924,
+      "grad_norm": 0.03488868847489357,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 22506
+    },
+    {
+      "epoch": 2.3135985198889917,
+      "grad_norm": 0.1007746085524559,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 22509
+    },
+    {
+      "epoch": 2.3139068763490593,
+      "grad_norm": 0.04351113364100456,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 22512
+    },
+    {
+      "epoch": 2.3142152328091274,
+      "grad_norm": 0.042396917939186096,
+      "learning_rate": 0.01,
+      "loss": 2.0427,
+      "step": 22515
+    },
+    {
+      "epoch": 2.314523589269195,
+      "grad_norm": 0.05541342496871948,
+      "learning_rate": 0.01,
+      "loss": 2.0577,
+      "step": 22518
+    },
+    {
+      "epoch": 2.314831945729263,
+      "grad_norm": 0.05093314126133919,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 22521
+    },
+    {
+      "epoch": 2.315140302189331,
+      "grad_norm": 0.09227500855922699,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 22524
+    },
+    {
+      "epoch": 2.3154486586493985,
+      "grad_norm": 0.07242649048566818,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 22527
+    },
+    {
+      "epoch": 2.3157570151094666,
+      "grad_norm": 0.07468149811029434,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 22530
+    },
+    {
+      "epoch": 2.3160653715695343,
+      "grad_norm": 0.0861506536602974,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 22533
+    },
+    {
+      "epoch": 2.3163737280296024,
+      "grad_norm": 0.05024004355072975,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 22536
+    },
+    {
+      "epoch": 2.31668208448967,
+      "grad_norm": 0.07255478948354721,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 22539
+    },
+    {
+      "epoch": 2.3169904409497377,
+      "grad_norm": 0.08199785649776459,
+      "learning_rate": 0.01,
+      "loss": 2.0465,
+      "step": 22542
+    },
+    {
+      "epoch": 2.317298797409806,
+      "grad_norm": 0.05194063484668732,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 22545
+    },
+    {
+      "epoch": 2.3176071538698735,
+      "grad_norm": 0.05785224214196205,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 22548
+    },
+    {
+      "epoch": 2.3179155103299416,
+      "grad_norm": 0.0767458900809288,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 22551
+    },
+    {
+      "epoch": 2.3182238667900092,
+      "grad_norm": 0.04905517399311066,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 22554
+    },
+    {
+      "epoch": 2.318532223250077,
+      "grad_norm": 0.12444967031478882,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 22557
+    },
+    {
+      "epoch": 2.318840579710145,
+      "grad_norm": 0.07965657114982605,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 22560
+    },
+    {
+      "epoch": 2.3191489361702127,
+      "grad_norm": 0.12167848646640778,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 22563
+    },
+    {
+      "epoch": 2.3194572926302808,
+      "grad_norm": 0.05483706668019295,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 22566
+    },
+    {
+      "epoch": 2.3197656490903484,
+      "grad_norm": 0.043046776205301285,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 22569
+    },
+    {
+      "epoch": 2.320074005550416,
+      "grad_norm": 0.03646457940340042,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 22572
+    },
+    {
+      "epoch": 2.320382362010484,
+      "grad_norm": 0.05777902901172638,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 22575
+    },
+    {
+      "epoch": 2.320690718470552,
+      "grad_norm": 0.08646814525127411,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 22578
+    },
+    {
+      "epoch": 2.32099907493062,
+      "grad_norm": 0.10770904272794724,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 22581
+    },
+    {
+      "epoch": 2.3213074313906876,
+      "grad_norm": 0.11546061187982559,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 22584
+    },
+    {
+      "epoch": 2.3216157878507553,
+      "grad_norm": 0.08060076832771301,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 22587
+    },
+    {
+      "epoch": 2.3219241443108234,
+      "grad_norm": 0.05546190217137337,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 22590
+    },
+    {
+      "epoch": 2.322232500770891,
+      "grad_norm": 0.06310277432203293,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 22593
+    },
+    {
+      "epoch": 2.322540857230959,
+      "grad_norm": 0.037245072424411774,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 22596
+    },
+    {
+      "epoch": 2.322849213691027,
+      "grad_norm": 0.05082730948925018,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 22599
+    },
+    {
+      "epoch": 2.3231575701510945,
+      "grad_norm": 0.054748065769672394,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 22602
+    },
+    {
+      "epoch": 2.3234659266111626,
+      "grad_norm": 0.06093902140855789,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 22605
+    },
+    {
+      "epoch": 2.3237742830712302,
+      "grad_norm": 0.10238637775182724,
+      "learning_rate": 0.01,
+      "loss": 2.0559,
+      "step": 22608
+    },
+    {
+      "epoch": 2.3240826395312983,
+      "grad_norm": 0.0850609764456749,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 22611
+    },
+    {
+      "epoch": 2.324390995991366,
+      "grad_norm": 0.04108014330267906,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 22614
+    },
+    {
+      "epoch": 2.3246993524514337,
+      "grad_norm": 0.03669281303882599,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 22617
+    },
+    {
+      "epoch": 2.3250077089115018,
+      "grad_norm": 0.053524449467659,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 22620
+    },
+    {
+      "epoch": 2.3253160653715694,
+      "grad_norm": 0.03545799478888512,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 22623
+    },
+    {
+      "epoch": 2.3256244218316375,
+      "grad_norm": 0.0569877028465271,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 22626
+    },
+    {
+      "epoch": 2.325932778291705,
+      "grad_norm": 0.06943153589963913,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 22629
+    },
+    {
+      "epoch": 2.326241134751773,
+      "grad_norm": 0.07389956712722778,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 22632
+    },
+    {
+      "epoch": 2.326549491211841,
+      "grad_norm": 0.08397506922483444,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 22635
+    },
+    {
+      "epoch": 2.3268578476719086,
+      "grad_norm": 0.13598600029945374,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 22638
+    },
+    {
+      "epoch": 2.3271662041319767,
+      "grad_norm": 0.08953214436769485,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 22641
+    },
+    {
+      "epoch": 2.3274745605920444,
+      "grad_norm": 0.09812797605991364,
+      "learning_rate": 0.01,
+      "loss": 2.0631,
+      "step": 22644
+    },
+    {
+      "epoch": 2.3277829170521125,
+      "grad_norm": 0.06696080416440964,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 22647
+    },
+    {
+      "epoch": 2.32809127351218,
+      "grad_norm": 0.07981070876121521,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 22650
+    },
+    {
+      "epoch": 2.328399629972248,
+      "grad_norm": 0.06655466556549072,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 22653
+    },
+    {
+      "epoch": 2.328707986432316,
+      "grad_norm": 0.061447955667972565,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 22656
+    },
+    {
+      "epoch": 2.3290163428923836,
+      "grad_norm": 0.08812141418457031,
+      "learning_rate": 0.01,
+      "loss": 2.0471,
+      "step": 22659
+    },
+    {
+      "epoch": 2.3293246993524512,
+      "grad_norm": 0.04082084447145462,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 22662
+    },
+    {
+      "epoch": 2.3296330558125193,
+      "grad_norm": 0.04827709123492241,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 22665
+    },
+    {
+      "epoch": 2.329941412272587,
+      "grad_norm": 0.04531668871641159,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 22668
+    },
+    {
+      "epoch": 2.330249768732655,
+      "grad_norm": 0.06739286333322525,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 22671
+    },
+    {
+      "epoch": 2.3305581251927228,
+      "grad_norm": 0.07753113657236099,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 22674
+    },
+    {
+      "epoch": 2.330866481652791,
+      "grad_norm": 0.1572093814611435,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 22677
+    },
+    {
+      "epoch": 2.3311748381128585,
+      "grad_norm": 0.0898297056555748,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 22680
+    },
+    {
+      "epoch": 2.331483194572926,
+      "grad_norm": 0.0803251639008522,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 22683
+    },
+    {
+      "epoch": 2.3317915510329943,
+      "grad_norm": 0.06155823543667793,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 22686
+    },
+    {
+      "epoch": 2.332099907493062,
+      "grad_norm": 0.03681538626551628,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 22689
+    },
+    {
+      "epoch": 2.3324082639531296,
+      "grad_norm": 0.03595598414540291,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 22692
+    },
+    {
+      "epoch": 2.3327166204131977,
+      "grad_norm": 0.06950604170560837,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 22695
+    },
+    {
+      "epoch": 2.3330249768732654,
+      "grad_norm": 0.07386364042758942,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 22698
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.06711214780807495,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 22701
+    },
+    {
+      "epoch": 2.333641689793401,
+      "grad_norm": 0.07357903569936752,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 22704
+    },
+    {
+      "epoch": 2.3339500462534692,
+      "grad_norm": 0.0660550519824028,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 22707
+    },
+    {
+      "epoch": 2.334258402713537,
+      "grad_norm": 0.05066053569316864,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 22710
+    },
+    {
+      "epoch": 2.3345667591736046,
+      "grad_norm": 0.04992471635341644,
+      "learning_rate": 0.01,
+      "loss": 2.0764,
+      "step": 22713
+    },
+    {
+      "epoch": 2.3348751156336727,
+      "grad_norm": 0.03477726876735687,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 22716
+    },
+    {
+      "epoch": 2.3351834720937403,
+      "grad_norm": 0.05454208329319954,
+      "learning_rate": 0.01,
+      "loss": 2.0476,
+      "step": 22719
+    },
+    {
+      "epoch": 2.335491828553808,
+      "grad_norm": 0.09009736031293869,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 22722
+    },
+    {
+      "epoch": 2.335800185013876,
+      "grad_norm": 0.10271045565605164,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 22725
+    },
+    {
+      "epoch": 2.3361085414739438,
+      "grad_norm": 0.05939007177948952,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 22728
+    },
+    {
+      "epoch": 2.336416897934012,
+      "grad_norm": 0.06454264372587204,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 22731
+    },
+    {
+      "epoch": 2.3367252543940795,
+      "grad_norm": 0.04892132803797722,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 22734
+    },
+    {
+      "epoch": 2.3370336108541476,
+      "grad_norm": 0.060863394290208817,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 22737
+    },
+    {
+      "epoch": 2.3373419673142153,
+      "grad_norm": 0.06073556840419769,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 22740
+    },
+    {
+      "epoch": 2.337650323774283,
+      "grad_norm": 0.04620682820677757,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 22743
+    },
+    {
+      "epoch": 2.337958680234351,
+      "grad_norm": 0.04834865778684616,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 22746
+    },
+    {
+      "epoch": 2.3382670366944187,
+      "grad_norm": 0.08154502511024475,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 22749
+    },
+    {
+      "epoch": 2.3385753931544864,
+      "grad_norm": 0.062378283590078354,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 22752
+    },
+    {
+      "epoch": 2.3388837496145545,
+      "grad_norm": 0.11560855805873871,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 22755
+    },
+    {
+      "epoch": 2.339192106074622,
+      "grad_norm": 0.14084307849407196,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 22758
+    },
+    {
+      "epoch": 2.3395004625346902,
+      "grad_norm": 0.08354218304157257,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 22761
+    },
+    {
+      "epoch": 2.339808818994758,
+      "grad_norm": 0.04660021886229515,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 22764
+    },
+    {
+      "epoch": 2.340117175454826,
+      "grad_norm": 0.06319580972194672,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 22767
+    },
+    {
+      "epoch": 2.3404255319148937,
+      "grad_norm": 0.06721585988998413,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 22770
+    },
+    {
+      "epoch": 2.3407338883749613,
+      "grad_norm": 0.08593994379043579,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 22773
+    },
+    {
+      "epoch": 2.3410422448350294,
+      "grad_norm": 0.07594014704227448,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 22776
+    },
+    {
+      "epoch": 2.341350601295097,
+      "grad_norm": 0.05942634493112564,
+      "learning_rate": 0.01,
+      "loss": 2.0557,
+      "step": 22779
+    },
+    {
+      "epoch": 2.3416589577551647,
+      "grad_norm": 0.09516235440969467,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 22782
+    },
+    {
+      "epoch": 2.341967314215233,
+      "grad_norm": 0.11480774730443954,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 22785
+    },
+    {
+      "epoch": 2.3422756706753005,
+      "grad_norm": 0.078018918633461,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 22788
+    },
+    {
+      "epoch": 2.3425840271353686,
+      "grad_norm": 0.04268670827150345,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 22791
+    },
+    {
+      "epoch": 2.3428923835954363,
+      "grad_norm": 0.04221808537840843,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 22794
+    },
+    {
+      "epoch": 2.3432007400555044,
+      "grad_norm": 0.04617379978299141,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 22797
+    },
+    {
+      "epoch": 2.343509096515572,
+      "grad_norm": 0.05600395053625107,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 22800
+    },
+    {
+      "epoch": 2.3438174529756397,
+      "grad_norm": 0.04997321218252182,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 22803
+    },
+    {
+      "epoch": 2.344125809435708,
+      "grad_norm": 0.08891329169273376,
+      "learning_rate": 0.01,
+      "loss": 2.0464,
+      "step": 22806
+    },
+    {
+      "epoch": 2.3444341658957755,
+      "grad_norm": 0.06883488595485687,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 22809
+    },
+    {
+      "epoch": 2.344742522355843,
+      "grad_norm": 0.05532229691743851,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 22812
+    },
+    {
+      "epoch": 2.3450508788159112,
+      "grad_norm": 0.04480346292257309,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 22815
+    },
+    {
+      "epoch": 2.345359235275979,
+      "grad_norm": 0.06533756107091904,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 22818
+    },
+    {
+      "epoch": 2.345667591736047,
+      "grad_norm": 0.09851729869842529,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 22821
+    },
+    {
+      "epoch": 2.3459759481961147,
+      "grad_norm": 0.06026868894696236,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 22824
+    },
+    {
+      "epoch": 2.3462843046561828,
+      "grad_norm": 0.10084753483533859,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 22827
+    },
+    {
+      "epoch": 2.3465926611162504,
+      "grad_norm": 0.09302054345607758,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 22830
+    },
+    {
+      "epoch": 2.346901017576318,
+      "grad_norm": 0.06304951757192612,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 22833
+    },
+    {
+      "epoch": 2.347209374036386,
+      "grad_norm": 0.03625926375389099,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 22836
+    },
+    {
+      "epoch": 2.347517730496454,
+      "grad_norm": 0.0602838359773159,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 22839
+    },
+    {
+      "epoch": 2.3478260869565215,
+      "grad_norm": 0.08285371959209442,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 22842
+    },
+    {
+      "epoch": 2.3481344434165896,
+      "grad_norm": 0.05780784785747528,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 22845
+    },
+    {
+      "epoch": 2.3484427998766573,
+      "grad_norm": 0.07491873949766159,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 22848
+    },
+    {
+      "epoch": 2.3487511563367254,
+      "grad_norm": 0.04745258018374443,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 22851
+    },
+    {
+      "epoch": 2.349059512796793,
+      "grad_norm": 0.030006757006049156,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 22854
+    },
+    {
+      "epoch": 2.349367869256861,
+      "grad_norm": 0.035286273807287216,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 22857
+    },
+    {
+      "epoch": 2.349676225716929,
+      "grad_norm": 0.08970153331756592,
+      "learning_rate": 0.01,
+      "loss": 2.0473,
+      "step": 22860
+    },
+    {
+      "epoch": 2.3499845821769965,
+      "grad_norm": 0.06819456070661545,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 22863
+    },
+    {
+      "epoch": 2.3502929386370646,
+      "grad_norm": 0.07130144536495209,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 22866
+    },
+    {
+      "epoch": 2.3506012950971322,
+      "grad_norm": 0.046634428203105927,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 22869
+    },
+    {
+      "epoch": 2.3509096515572,
+      "grad_norm": 0.10333798825740814,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 22872
+    },
+    {
+      "epoch": 2.351218008017268,
+      "grad_norm": 0.06691887974739075,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 22875
+    },
+    {
+      "epoch": 2.3515263644773357,
+      "grad_norm": 0.06780122220516205,
+      "learning_rate": 0.01,
+      "loss": 2.052,
+      "step": 22878
+    },
+    {
+      "epoch": 2.3518347209374038,
+      "grad_norm": 0.04448529705405235,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 22881
+    },
+    {
+      "epoch": 2.3521430773974714,
+      "grad_norm": 0.09176638722419739,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 22884
+    },
+    {
+      "epoch": 2.3524514338575395,
+      "grad_norm": 0.05699750408530235,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 22887
+    },
+    {
+      "epoch": 2.352759790317607,
+      "grad_norm": 0.0556836873292923,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 22890
+    },
+    {
+      "epoch": 2.353068146777675,
+      "grad_norm": 0.05335136875510216,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 22893
+    },
+    {
+      "epoch": 2.353376503237743,
+      "grad_norm": 0.06900188326835632,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 22896
+    },
+    {
+      "epoch": 2.3536848596978106,
+      "grad_norm": 0.07300775498151779,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 22899
+    },
+    {
+      "epoch": 2.3539932161578783,
+      "grad_norm": 0.0912926122546196,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 22902
+    },
+    {
+      "epoch": 2.3543015726179464,
+      "grad_norm": 0.0803481861948967,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 22905
+    },
+    {
+      "epoch": 2.354609929078014,
+      "grad_norm": 0.05067334324121475,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 22908
+    },
+    {
+      "epoch": 2.354918285538082,
+      "grad_norm": 0.10282410681247711,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 22911
+    },
+    {
+      "epoch": 2.35522664199815,
+      "grad_norm": 0.04224551469087601,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 22914
+    },
+    {
+      "epoch": 2.355534998458218,
+      "grad_norm": 0.06768529862165451,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 22917
+    },
+    {
+      "epoch": 2.3558433549182856,
+      "grad_norm": 0.10372413694858551,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 22920
+    },
+    {
+      "epoch": 2.356151711378353,
+      "grad_norm": 0.05974859744310379,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 22923
+    },
+    {
+      "epoch": 2.3564600678384213,
+      "grad_norm": 0.06408294290304184,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 22926
+    },
+    {
+      "epoch": 2.356768424298489,
+      "grad_norm": 0.053037382662296295,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 22929
+    },
+    {
+      "epoch": 2.3570767807585566,
+      "grad_norm": 0.038871798664331436,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 22932
+    },
+    {
+      "epoch": 2.3573851372186247,
+      "grad_norm": 0.046300627291202545,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 22935
+    },
+    {
+      "epoch": 2.3576934936786924,
+      "grad_norm": 0.05344367399811745,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 22938
+    },
+    {
+      "epoch": 2.3580018501387605,
+      "grad_norm": 0.0878051146864891,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 22941
+    },
+    {
+      "epoch": 2.358310206598828,
+      "grad_norm": 0.09469857066869736,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 22944
+    },
+    {
+      "epoch": 2.3586185630588963,
+      "grad_norm": 0.04746483266353607,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 22947
+    },
+    {
+      "epoch": 2.358926919518964,
+      "grad_norm": 0.051294367760419846,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 22950
+    },
+    {
+      "epoch": 2.3592352759790316,
+      "grad_norm": 0.04378626495599747,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 22953
+    },
+    {
+      "epoch": 2.3595436324390997,
+      "grad_norm": 0.054627690464258194,
+      "learning_rate": 0.01,
+      "loss": 2.0592,
+      "step": 22956
+    },
+    {
+      "epoch": 2.3598519888991674,
+      "grad_norm": 0.04732866585254669,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 22959
+    },
+    {
+      "epoch": 2.3601603453592355,
+      "grad_norm": 0.11569249629974365,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 22962
+    },
+    {
+      "epoch": 2.360468701819303,
+      "grad_norm": 0.10606824606657028,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 22965
+    },
+    {
+      "epoch": 2.360777058279371,
+      "grad_norm": 0.05072540044784546,
+      "learning_rate": 0.01,
+      "loss": 2.0581,
+      "step": 22968
+    },
+    {
+      "epoch": 2.361085414739439,
+      "grad_norm": 0.0738227441906929,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 22971
+    },
+    {
+      "epoch": 2.3613937711995066,
+      "grad_norm": 0.06440632790327072,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 22974
+    },
+    {
+      "epoch": 2.3617021276595747,
+      "grad_norm": 0.05426304414868355,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 22977
+    },
+    {
+      "epoch": 2.3620104841196423,
+      "grad_norm": 0.06414210051298141,
+      "learning_rate": 0.01,
+      "loss": 2.0735,
+      "step": 22980
+    },
+    {
+      "epoch": 2.36231884057971,
+      "grad_norm": 0.034726858139038086,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 22983
+    },
+    {
+      "epoch": 2.362627197039778,
+      "grad_norm": 0.04326315596699715,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 22986
+    },
+    {
+      "epoch": 2.3629355534998457,
+      "grad_norm": 0.06470071524381638,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 22989
+    },
+    {
+      "epoch": 2.363243909959914,
+      "grad_norm": 0.09161586314439774,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 22992
+    },
+    {
+      "epoch": 2.3635522664199815,
+      "grad_norm": 0.035171300172805786,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 22995
+    },
+    {
+      "epoch": 2.363860622880049,
+      "grad_norm": 0.059095609933137894,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 22998
+    },
+    {
+      "epoch": 2.3641689793401173,
+      "grad_norm": 0.094328872859478,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 23001
+    },
+    {
+      "epoch": 2.364477335800185,
+      "grad_norm": 0.059930335730314255,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 23004
+    },
+    {
+      "epoch": 2.364785692260253,
+      "grad_norm": 0.08969483524560928,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 23007
+    },
+    {
+      "epoch": 2.3650940487203207,
+      "grad_norm": 0.054753273725509644,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 23010
+    },
+    {
+      "epoch": 2.3654024051803884,
+      "grad_norm": 0.09794525057077408,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 23013
+    },
+    {
+      "epoch": 2.3657107616404565,
+      "grad_norm": 0.07700082659721375,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 23016
+    },
+    {
+      "epoch": 2.366019118100524,
+      "grad_norm": 0.03871360793709755,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 23019
+    },
+    {
+      "epoch": 2.3663274745605922,
+      "grad_norm": 0.08822629600763321,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 23022
+    },
+    {
+      "epoch": 2.36663583102066,
+      "grad_norm": 0.08018068969249725,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 23025
+    },
+    {
+      "epoch": 2.3669441874807275,
+      "grad_norm": 0.046322405338287354,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 23028
+    },
+    {
+      "epoch": 2.3672525439407957,
+      "grad_norm": 0.08056683838367462,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 23031
+    },
+    {
+      "epoch": 2.3675609004008633,
+      "grad_norm": 0.059787072241306305,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 23034
+    },
+    {
+      "epoch": 2.3678692568609314,
+      "grad_norm": 0.07793419063091278,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 23037
+    },
+    {
+      "epoch": 2.368177613320999,
+      "grad_norm": 0.1337706595659256,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 23040
+    },
+    {
+      "epoch": 2.3684859697810667,
+      "grad_norm": 0.0608229860663414,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 23043
+    },
+    {
+      "epoch": 2.368794326241135,
+      "grad_norm": 0.07523812353610992,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 23046
+    },
+    {
+      "epoch": 2.3691026827012025,
+      "grad_norm": 0.05788956955075264,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 23049
+    },
+    {
+      "epoch": 2.3694110391612706,
+      "grad_norm": 0.05372630059719086,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 23052
+    },
+    {
+      "epoch": 2.3697193956213383,
+      "grad_norm": 0.0775051936507225,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 23055
+    },
+    {
+      "epoch": 2.370027752081406,
+      "grad_norm": 0.042055148631334305,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 23058
+    },
+    {
+      "epoch": 2.370336108541474,
+      "grad_norm": 0.03500501438975334,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 23061
+    },
+    {
+      "epoch": 2.3706444650015417,
+      "grad_norm": 0.05405488237738609,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 23064
+    },
+    {
+      "epoch": 2.37095282146161,
+      "grad_norm": 0.04199660196900368,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 23067
+    },
+    {
+      "epoch": 2.3712611779216775,
+      "grad_norm": 0.10155687481164932,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 23070
+    },
+    {
+      "epoch": 2.371569534381745,
+      "grad_norm": 0.04267631843686104,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 23073
+    },
+    {
+      "epoch": 2.371877890841813,
+      "grad_norm": 0.08961319178342819,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 23076
+    },
+    {
+      "epoch": 2.372186247301881,
+      "grad_norm": 0.06389858573675156,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 23079
+    },
+    {
+      "epoch": 2.372494603761949,
+      "grad_norm": 0.0733552873134613,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 23082
+    },
+    {
+      "epoch": 2.3728029602220166,
+      "grad_norm": 0.04577986150979996,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 23085
+    },
+    {
+      "epoch": 2.3731113166820843,
+      "grad_norm": 0.07929688692092896,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 23088
+    },
+    {
+      "epoch": 2.3734196731421524,
+      "grad_norm": 0.07446785271167755,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 23091
+    },
+    {
+      "epoch": 2.37372802960222,
+      "grad_norm": 0.08491658419370651,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 23094
+    },
+    {
+      "epoch": 2.374036386062288,
+      "grad_norm": 0.04434317350387573,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 23097
+    },
+    {
+      "epoch": 2.374344742522356,
+      "grad_norm": 0.031715601682662964,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 23100
+    },
+    {
+      "epoch": 2.3746530989824235,
+      "grad_norm": 0.05073494091629982,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 23103
+    },
+    {
+      "epoch": 2.3749614554424916,
+      "grad_norm": 0.0484505333006382,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 23106
+    },
+    {
+      "epoch": 2.3752698119025593,
+      "grad_norm": 0.036467526108026505,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 23109
+    },
+    {
+      "epoch": 2.3755781683626274,
+      "grad_norm": 0.048451874405145645,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 23112
+    },
+    {
+      "epoch": 2.375886524822695,
+      "grad_norm": 0.048714105039834976,
+      "learning_rate": 0.01,
+      "loss": 2.0546,
+      "step": 23115
+    },
+    {
+      "epoch": 2.3761948812827627,
+      "grad_norm": 0.07499600946903229,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 23118
+    },
+    {
+      "epoch": 2.376503237742831,
+      "grad_norm": 0.052757177501916885,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 23121
+    },
+    {
+      "epoch": 2.3768115942028984,
+      "grad_norm": 0.062441807240247726,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 23124
+    },
+    {
+      "epoch": 2.3771199506629666,
+      "grad_norm": 0.04984834045171738,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 23127
+    },
+    {
+      "epoch": 2.377428307123034,
+      "grad_norm": 0.12042737007141113,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 23130
+    },
+    {
+      "epoch": 2.377736663583102,
+      "grad_norm": 0.08114577829837799,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 23133
+    },
+    {
+      "epoch": 2.37804502004317,
+      "grad_norm": 0.044427551329135895,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 23136
+    },
+    {
+      "epoch": 2.3783533765032376,
+      "grad_norm": 0.04814080893993378,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 23139
+    },
+    {
+      "epoch": 2.3786617329633057,
+      "grad_norm": 0.03783731907606125,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 23142
+    },
+    {
+      "epoch": 2.3789700894233734,
+      "grad_norm": 0.03743833675980568,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 23145
+    },
+    {
+      "epoch": 2.3792784458834415,
+      "grad_norm": 0.05341466888785362,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 23148
+    },
+    {
+      "epoch": 2.379586802343509,
+      "grad_norm": 0.04965018108487129,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 23151
+    },
+    {
+      "epoch": 2.379895158803577,
+      "grad_norm": 0.06330037117004395,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 23154
+    },
+    {
+      "epoch": 2.380203515263645,
+      "grad_norm": 0.04018980264663696,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 23157
+    },
+    {
+      "epoch": 2.3805118717237126,
+      "grad_norm": 0.04469529166817665,
+      "learning_rate": 0.01,
+      "loss": 2.0437,
+      "step": 23160
+    },
+    {
+      "epoch": 2.3808202281837803,
+      "grad_norm": 0.07242292910814285,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 23163
+    },
+    {
+      "epoch": 2.3811285846438484,
+      "grad_norm": 0.05165792256593704,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 23166
+    },
+    {
+      "epoch": 2.381436941103916,
+      "grad_norm": 0.12728899717330933,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 23169
+    },
+    {
+      "epoch": 2.381745297563984,
+      "grad_norm": 0.07357775419950485,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 23172
+    },
+    {
+      "epoch": 2.382053654024052,
+      "grad_norm": 0.05125182494521141,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 23175
+    },
+    {
+      "epoch": 2.38236201048412,
+      "grad_norm": 0.061880383640527725,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 23178
+    },
+    {
+      "epoch": 2.3826703669441875,
+      "grad_norm": 0.03774267062544823,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 23181
+    },
+    {
+      "epoch": 2.382978723404255,
+      "grad_norm": 0.04750019684433937,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 23184
+    },
+    {
+      "epoch": 2.3832870798643233,
+      "grad_norm": 0.06727109849452972,
+      "learning_rate": 0.01,
+      "loss": 2.0545,
+      "step": 23187
+    },
+    {
+      "epoch": 2.383595436324391,
+      "grad_norm": 0.08848878741264343,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 23190
+    },
+    {
+      "epoch": 2.3839037927844586,
+      "grad_norm": 0.07577743381261826,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 23193
+    },
+    {
+      "epoch": 2.3842121492445267,
+      "grad_norm": 0.07908739149570465,
+      "learning_rate": 0.01,
+      "loss": 2.0516,
+      "step": 23196
+    },
+    {
+      "epoch": 2.3845205057045944,
+      "grad_norm": 0.08100943267345428,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 23199
+    },
+    {
+      "epoch": 2.3848288621646625,
+      "grad_norm": 0.12388890236616135,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 23202
+    },
+    {
+      "epoch": 2.38513721862473,
+      "grad_norm": 0.0538008026778698,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 23205
+    },
+    {
+      "epoch": 2.3854455750847983,
+      "grad_norm": 0.04448952525854111,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 23208
+    },
+    {
+      "epoch": 2.385753931544866,
+      "grad_norm": 0.038850218057632446,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 23211
+    },
+    {
+      "epoch": 2.3860622880049336,
+      "grad_norm": 0.10436736047267914,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 23214
+    },
+    {
+      "epoch": 2.3863706444650017,
+      "grad_norm": 0.03787916898727417,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 23217
+    },
+    {
+      "epoch": 2.3866790009250693,
+      "grad_norm": 0.0874573364853859,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 23220
+    },
+    {
+      "epoch": 2.386987357385137,
+      "grad_norm": 0.08322203904390335,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 23223
+    },
+    {
+      "epoch": 2.387295713845205,
+      "grad_norm": 0.034374579787254333,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 23226
+    },
+    {
+      "epoch": 2.3876040703052728,
+      "grad_norm": 0.03772469237446785,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 23229
+    },
+    {
+      "epoch": 2.387912426765341,
+      "grad_norm": 0.12290430814027786,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 23232
+    },
+    {
+      "epoch": 2.3882207832254085,
+      "grad_norm": 0.04306014999747276,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 23235
+    },
+    {
+      "epoch": 2.3885291396854766,
+      "grad_norm": 0.10940956324338913,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 23238
+    },
+    {
+      "epoch": 2.3888374961455443,
+      "grad_norm": 0.06950829923152924,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 23241
+    },
+    {
+      "epoch": 2.389145852605612,
+      "grad_norm": 0.11061616986989975,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 23244
+    },
+    {
+      "epoch": 2.38945420906568,
+      "grad_norm": 0.051813628524541855,
+      "learning_rate": 0.01,
+      "loss": 2.0657,
+      "step": 23247
+    },
+    {
+      "epoch": 2.3897625655257477,
+      "grad_norm": 0.10411377251148224,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 23250
+    },
+    {
+      "epoch": 2.3900709219858154,
+      "grad_norm": 0.03951719403266907,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 23253
+    },
+    {
+      "epoch": 2.3903792784458835,
+      "grad_norm": 0.08920100331306458,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 23256
+    },
+    {
+      "epoch": 2.390687634905951,
+      "grad_norm": 0.03843710198998451,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 23259
+    },
+    {
+      "epoch": 2.3909959913660193,
+      "grad_norm": 0.09093856811523438,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 23262
+    },
+    {
+      "epoch": 2.391304347826087,
+      "grad_norm": 0.06239760294556618,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 23265
+    },
+    {
+      "epoch": 2.391612704286155,
+      "grad_norm": 0.07422836124897003,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 23268
+    },
+    {
+      "epoch": 2.3919210607462227,
+      "grad_norm": 0.04418382793664932,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 23271
+    },
+    {
+      "epoch": 2.3922294172062903,
+      "grad_norm": 0.08631134033203125,
+      "learning_rate": 0.01,
+      "loss": 2.0675,
+      "step": 23274
+    },
+    {
+      "epoch": 2.3925377736663584,
+      "grad_norm": 0.05164894834160805,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 23277
+    },
+    {
+      "epoch": 2.392846130126426,
+      "grad_norm": 0.056628335267305374,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 23280
+    },
+    {
+      "epoch": 2.3931544865864938,
+      "grad_norm": 0.10568714141845703,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 23283
+    },
+    {
+      "epoch": 2.393462843046562,
+      "grad_norm": 0.1235639825463295,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 23286
+    },
+    {
+      "epoch": 2.3937711995066295,
+      "grad_norm": 0.04766182228922844,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 23289
+    },
+    {
+      "epoch": 2.3940795559666976,
+      "grad_norm": 0.040086500346660614,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 23292
+    },
+    {
+      "epoch": 2.3943879124267653,
+      "grad_norm": 0.04474460333585739,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 23295
+    },
+    {
+      "epoch": 2.3946962688868334,
+      "grad_norm": 0.03349655494093895,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 23298
+    },
+    {
+      "epoch": 2.395004625346901,
+      "grad_norm": 0.04361939802765846,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 23301
+    },
+    {
+      "epoch": 2.3953129818069687,
+      "grad_norm": 0.08834217488765717,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 23304
+    },
+    {
+      "epoch": 2.395621338267037,
+      "grad_norm": 0.09799984842538834,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 23307
+    },
+    {
+      "epoch": 2.3959296947271045,
+      "grad_norm": 0.06085311621427536,
+      "learning_rate": 0.01,
+      "loss": 2.0326,
+      "step": 23310
+    },
+    {
+      "epoch": 2.396238051187172,
+      "grad_norm": 0.04466985911130905,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 23313
+    },
+    {
+      "epoch": 2.3965464076472403,
+      "grad_norm": 0.05248266085982323,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 23316
+    },
+    {
+      "epoch": 2.396854764107308,
+      "grad_norm": 0.06024044379591942,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 23319
+    },
+    {
+      "epoch": 2.397163120567376,
+      "grad_norm": 0.05061257630586624,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 23322
+    },
+    {
+      "epoch": 2.3974714770274437,
+      "grad_norm": 0.04136064276099205,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 23325
+    },
+    {
+      "epoch": 2.397779833487512,
+      "grad_norm": 0.048327211290597916,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 23328
+    },
+    {
+      "epoch": 2.3980881899475794,
+      "grad_norm": 0.038865648210048676,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 23331
+    },
+    {
+      "epoch": 2.398396546407647,
+      "grad_norm": 0.04522191360592842,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 23334
+    },
+    {
+      "epoch": 2.398704902867715,
+      "grad_norm": 0.033488884568214417,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 23337
+    },
+    {
+      "epoch": 2.399013259327783,
+      "grad_norm": 0.05207303538918495,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 23340
+    },
+    {
+      "epoch": 2.3993216157878505,
+      "grad_norm": 0.09892462193965912,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 23343
+    },
+    {
+      "epoch": 2.3996299722479186,
+      "grad_norm": 0.07379446923732758,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 23346
+    },
+    {
+      "epoch": 2.3999383287079863,
+      "grad_norm": 0.07604874670505524,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 23349
+    },
+    {
+      "epoch": 2.4002466851680544,
+      "grad_norm": 0.05876392871141434,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 23352
+    },
+    {
+      "epoch": 2.400555041628122,
+      "grad_norm": 0.06497185677289963,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 23355
+    },
+    {
+      "epoch": 2.40086339808819,
+      "grad_norm": 0.059574466198682785,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 23358
+    },
+    {
+      "epoch": 2.401171754548258,
+      "grad_norm": 0.03427527844905853,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 23361
+    },
+    {
+      "epoch": 2.4014801110083255,
+      "grad_norm": 0.04248304292559624,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 23364
+    },
+    {
+      "epoch": 2.4017884674683936,
+      "grad_norm": 0.11010278761386871,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 23367
+    },
+    {
+      "epoch": 2.4020968239284612,
+      "grad_norm": 0.03992384672164917,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 23370
+    },
+    {
+      "epoch": 2.402405180388529,
+      "grad_norm": 0.0774572566151619,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 23373
+    },
+    {
+      "epoch": 2.402713536848597,
+      "grad_norm": 0.0753302052617073,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 23376
+    },
+    {
+      "epoch": 2.4030218933086647,
+      "grad_norm": 0.06872344017028809,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 23379
+    },
+    {
+      "epoch": 2.4033302497687328,
+      "grad_norm": 0.03949934244155884,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 23382
+    },
+    {
+      "epoch": 2.4036386062288004,
+      "grad_norm": 0.06230955943465233,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 23385
+    },
+    {
+      "epoch": 2.4039469626888685,
+      "grad_norm": 0.0821446031332016,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 23388
+    },
+    {
+      "epoch": 2.404255319148936,
+      "grad_norm": 0.0392254963517189,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 23391
+    },
+    {
+      "epoch": 2.404563675609004,
+      "grad_norm": 0.04208219051361084,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 23394
+    },
+    {
+      "epoch": 2.404872032069072,
+      "grad_norm": 0.053683795034885406,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 23397
+    },
+    {
+      "epoch": 2.4051803885291396,
+      "grad_norm": 0.0707748755812645,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 23400
+    },
+    {
+      "epoch": 2.4054887449892073,
+      "grad_norm": 0.04077059030532837,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 23403
+    },
+    {
+      "epoch": 2.4057971014492754,
+      "grad_norm": 0.041255395859479904,
+      "learning_rate": 0.01,
+      "loss": 2.0542,
+      "step": 23406
+    },
+    {
+      "epoch": 2.406105457909343,
+      "grad_norm": 0.04560267925262451,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 23409
+    },
+    {
+      "epoch": 2.406413814369411,
+      "grad_norm": 0.1152443066239357,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 23412
+    },
+    {
+      "epoch": 2.406722170829479,
+      "grad_norm": 0.06491648405790329,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 23415
+    },
+    {
+      "epoch": 2.407030527289547,
+      "grad_norm": 0.10551664233207703,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 23418
+    },
+    {
+      "epoch": 2.4073388837496146,
+      "grad_norm": 0.05565710365772247,
+      "learning_rate": 0.01,
+      "loss": 2.051,
+      "step": 23421
+    },
+    {
+      "epoch": 2.4076472402096822,
+      "grad_norm": 0.07831353694200516,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 23424
+    },
+    {
+      "epoch": 2.4079555966697503,
+      "grad_norm": 0.04822281748056412,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 23427
+    },
+    {
+      "epoch": 2.408263953129818,
+      "grad_norm": 0.10326611250638962,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 23430
+    },
+    {
+      "epoch": 2.4085723095898857,
+      "grad_norm": 0.03122270293533802,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 23433
+    },
+    {
+      "epoch": 2.4088806660499538,
+      "grad_norm": 0.04365687072277069,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 23436
+    },
+    {
+      "epoch": 2.4091890225100214,
+      "grad_norm": 0.055597368627786636,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 23439
+    },
+    {
+      "epoch": 2.4094973789700895,
+      "grad_norm": 0.05635388568043709,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 23442
+    },
+    {
+      "epoch": 2.409805735430157,
+      "grad_norm": 0.059105440974235535,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 23445
+    },
+    {
+      "epoch": 2.4101140918902253,
+      "grad_norm": 0.03262312710285187,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 23448
+    },
+    {
+      "epoch": 2.410422448350293,
+      "grad_norm": 0.12139608711004257,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 23451
+    },
+    {
+      "epoch": 2.4107308048103606,
+      "grad_norm": 0.0331861712038517,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 23454
+    },
+    {
+      "epoch": 2.4110391612704287,
+      "grad_norm": 0.14580830931663513,
+      "learning_rate": 0.01,
+      "loss": 2.0268,
+      "step": 23457
+    },
+    {
+      "epoch": 2.4113475177304964,
+      "grad_norm": 0.08361393213272095,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 23460
+    },
+    {
+      "epoch": 2.4116558741905645,
+      "grad_norm": 0.04604450613260269,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 23463
+    },
+    {
+      "epoch": 2.411964230650632,
+      "grad_norm": 0.0660676434636116,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 23466
+    },
+    {
+      "epoch": 2.4122725871107,
+      "grad_norm": 0.03829890862107277,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 23469
+    },
+    {
+      "epoch": 2.412580943570768,
+      "grad_norm": 0.03486289829015732,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 23472
+    },
+    {
+      "epoch": 2.4128893000308356,
+      "grad_norm": 0.043325070291757584,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 23475
+    },
+    {
+      "epoch": 2.4131976564909037,
+      "grad_norm": 0.062313832342624664,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 23478
+    },
+    {
+      "epoch": 2.4135060129509713,
+      "grad_norm": 0.07333476096391678,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 23481
+    },
+    {
+      "epoch": 2.413814369411039,
+      "grad_norm": 0.05855254456400871,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 23484
+    },
+    {
+      "epoch": 2.414122725871107,
+      "grad_norm": 0.07818280905485153,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 23487
+    },
+    {
+      "epoch": 2.4144310823311748,
+      "grad_norm": 0.17364929616451263,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 23490
+    },
+    {
+      "epoch": 2.414739438791243,
+      "grad_norm": 0.04748675599694252,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 23493
+    },
+    {
+      "epoch": 2.4150477952513105,
+      "grad_norm": 0.04472067952156067,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 23496
+    },
+    {
+      "epoch": 2.415356151711378,
+      "grad_norm": 0.04624779149889946,
+      "learning_rate": 0.01,
+      "loss": 2.0606,
+      "step": 23499
+    },
+    {
+      "epoch": 2.4156645081714463,
+      "grad_norm": 0.04642438516020775,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 23502
+    },
+    {
+      "epoch": 2.415972864631514,
+      "grad_norm": 0.03872397541999817,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 23505
+    },
+    {
+      "epoch": 2.416281221091582,
+      "grad_norm": 0.055659957230091095,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 23508
+    },
+    {
+      "epoch": 2.4165895775516497,
+      "grad_norm": 0.05187131464481354,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 23511
+    },
+    {
+      "epoch": 2.4168979340117174,
+      "grad_norm": 0.04747389629483223,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 23514
+    },
+    {
+      "epoch": 2.4172062904717855,
+      "grad_norm": 0.04594513028860092,
+      "learning_rate": 0.01,
+      "loss": 2.0489,
+      "step": 23517
+    },
+    {
+      "epoch": 2.417514646931853,
+      "grad_norm": 0.08976871520280838,
+      "learning_rate": 0.01,
+      "loss": 2.0728,
+      "step": 23520
+    },
+    {
+      "epoch": 2.4178230033919212,
+      "grad_norm": 0.09563881158828735,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 23523
+    },
+    {
+      "epoch": 2.418131359851989,
+      "grad_norm": 0.05488200485706329,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 23526
+    },
+    {
+      "epoch": 2.4184397163120566,
+      "grad_norm": 0.10557366907596588,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 23529
+    },
+    {
+      "epoch": 2.4187480727721247,
+      "grad_norm": 0.06399507075548172,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 23532
+    },
+    {
+      "epoch": 2.4190564292321923,
+      "grad_norm": 0.06832768023014069,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 23535
+    },
+    {
+      "epoch": 2.4193647856922604,
+      "grad_norm": 0.04164440929889679,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 23538
+    },
+    {
+      "epoch": 2.419673142152328,
+      "grad_norm": 0.03020538203418255,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 23541
+    },
+    {
+      "epoch": 2.4199814986123958,
+      "grad_norm": 0.05985027551651001,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 23544
+    },
+    {
+      "epoch": 2.420289855072464,
+      "grad_norm": 0.08126893639564514,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 23547
+    },
+    {
+      "epoch": 2.4205982115325315,
+      "grad_norm": 0.03897944465279579,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 23550
+    },
+    {
+      "epoch": 2.4209065679925996,
+      "grad_norm": 0.05052472651004791,
+      "learning_rate": 0.01,
+      "loss": 2.0536,
+      "step": 23553
+    },
+    {
+      "epoch": 2.4212149244526673,
+      "grad_norm": 0.03434956073760986,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 23556
+    },
+    {
+      "epoch": 2.421523280912735,
+      "grad_norm": 0.0911344438791275,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 23559
+    },
+    {
+      "epoch": 2.421831637372803,
+      "grad_norm": 0.11097840219736099,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 23562
+    },
+    {
+      "epoch": 2.4221399938328707,
+      "grad_norm": 0.09742028266191483,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 23565
+    },
+    {
+      "epoch": 2.422448350292939,
+      "grad_norm": 0.04184804484248161,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 23568
+    },
+    {
+      "epoch": 2.4227567067530065,
+      "grad_norm": 0.04188203811645508,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 23571
+    },
+    {
+      "epoch": 2.423065063213074,
+      "grad_norm": 0.05326924845576286,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 23574
+    },
+    {
+      "epoch": 2.4233734196731422,
+      "grad_norm": 0.04598800092935562,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 23577
+    },
+    {
+      "epoch": 2.42368177613321,
+      "grad_norm": 0.1110750213265419,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 23580
+    },
+    {
+      "epoch": 2.423990132593278,
+      "grad_norm": 0.0473935566842556,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 23583
+    },
+    {
+      "epoch": 2.4242984890533457,
+      "grad_norm": 0.08599594235420227,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 23586
+    },
+    {
+      "epoch": 2.4246068455134133,
+      "grad_norm": 0.03763740509748459,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 23589
+    },
+    {
+      "epoch": 2.4249152019734814,
+      "grad_norm": 0.056663963943719864,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 23592
+    },
+    {
+      "epoch": 2.425223558433549,
+      "grad_norm": 0.09664086252450943,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 23595
+    },
+    {
+      "epoch": 2.425531914893617,
+      "grad_norm": 0.047283969819545746,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 23598
+    },
+    {
+      "epoch": 2.425840271353685,
+      "grad_norm": 0.03726861625909805,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 23601
+    },
+    {
+      "epoch": 2.4261486278137525,
+      "grad_norm": 0.0514906644821167,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 23604
+    },
+    {
+      "epoch": 2.4264569842738206,
+      "grad_norm": 0.11972562968730927,
+      "learning_rate": 0.01,
+      "loss": 2.07,
+      "step": 23607
+    },
+    {
+      "epoch": 2.4267653407338883,
+      "grad_norm": 0.17023397982120514,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 23610
+    },
+    {
+      "epoch": 2.4270736971939564,
+      "grad_norm": 0.1355310082435608,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 23613
+    },
+    {
+      "epoch": 2.427382053654024,
+      "grad_norm": 0.039905257523059845,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 23616
+    },
+    {
+      "epoch": 2.4276904101140917,
+      "grad_norm": 0.0339365154504776,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 23619
+    },
+    {
+      "epoch": 2.42799876657416,
+      "grad_norm": 0.03365493193268776,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 23622
+    },
+    {
+      "epoch": 2.4283071230342275,
+      "grad_norm": 0.041596464812755585,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 23625
+    },
+    {
+      "epoch": 2.4286154794942956,
+      "grad_norm": 0.0537833645939827,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 23628
+    },
+    {
+      "epoch": 2.4289238359543632,
+      "grad_norm": 0.049658019095659256,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 23631
+    },
+    {
+      "epoch": 2.429232192414431,
+      "grad_norm": 0.03983505442738533,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 23634
+    },
+    {
+      "epoch": 2.429540548874499,
+      "grad_norm": 0.0780608057975769,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 23637
+    },
+    {
+      "epoch": 2.4298489053345667,
+      "grad_norm": 0.13898798823356628,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 23640
+    },
+    {
+      "epoch": 2.4301572617946348,
+      "grad_norm": 0.07781122624874115,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 23643
+    },
+    {
+      "epoch": 2.4304656182547024,
+      "grad_norm": 0.045428015291690826,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 23646
+    },
+    {
+      "epoch": 2.43077397471477,
+      "grad_norm": 0.03437395393848419,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 23649
+    },
+    {
+      "epoch": 2.431082331174838,
+      "grad_norm": 0.06030704453587532,
+      "learning_rate": 0.01,
+      "loss": 2.0563,
+      "step": 23652
+    },
+    {
+      "epoch": 2.431390687634906,
+      "grad_norm": 0.046194083988666534,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 23655
+    },
+    {
+      "epoch": 2.431699044094974,
+      "grad_norm": 0.10041481256484985,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 23658
+    },
+    {
+      "epoch": 2.4320074005550416,
+      "grad_norm": 0.07109228521585464,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 23661
+    },
+    {
+      "epoch": 2.4323157570151093,
+      "grad_norm": 0.03319769352674484,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 23664
+    },
+    {
+      "epoch": 2.4326241134751774,
+      "grad_norm": 0.047379735857248306,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 23667
+    },
+    {
+      "epoch": 2.432932469935245,
+      "grad_norm": 0.04836783930659294,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 23670
+    },
+    {
+      "epoch": 2.433240826395313,
+      "grad_norm": 0.07264747470617294,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 23673
+    },
+    {
+      "epoch": 2.433549182855381,
+      "grad_norm": 0.09331446141004562,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 23676
+    },
+    {
+      "epoch": 2.433857539315449,
+      "grad_norm": 0.07618245482444763,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 23679
+    },
+    {
+      "epoch": 2.4341658957755166,
+      "grad_norm": 0.13270626962184906,
+      "learning_rate": 0.01,
+      "loss": 2.05,
+      "step": 23682
+    },
+    {
+      "epoch": 2.4344742522355842,
+      "grad_norm": 0.06691613793373108,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 23685
+    },
+    {
+      "epoch": 2.4347826086956523,
+      "grad_norm": 0.04376102611422539,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 23688
+    },
+    {
+      "epoch": 2.43509096515572,
+      "grad_norm": 0.04829081892967224,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 23691
+    },
+    {
+      "epoch": 2.4353993216157876,
+      "grad_norm": 0.04604188725352287,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 23694
+    },
+    {
+      "epoch": 2.4357076780758558,
+      "grad_norm": 0.06296957284212112,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 23697
+    },
+    {
+      "epoch": 2.4360160345359234,
+      "grad_norm": 0.07184949517250061,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 23700
+    },
+    {
+      "epoch": 2.4363243909959915,
+      "grad_norm": 0.11544491350650787,
+      "learning_rate": 0.01,
+      "loss": 2.0512,
+      "step": 23703
+    },
+    {
+      "epoch": 2.436632747456059,
+      "grad_norm": 0.03796609491109848,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 23706
+    },
+    {
+      "epoch": 2.4369411039161273,
+      "grad_norm": 0.09068101644515991,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 23709
+    },
+    {
+      "epoch": 2.437249460376195,
+      "grad_norm": 0.06220867484807968,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 23712
+    },
+    {
+      "epoch": 2.4375578168362626,
+      "grad_norm": 0.040711212903261185,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 23715
+    },
+    {
+      "epoch": 2.4378661732963307,
+      "grad_norm": 0.03610675781965256,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 23718
+    },
+    {
+      "epoch": 2.4381745297563984,
+      "grad_norm": 0.06709878146648407,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 23721
+    },
+    {
+      "epoch": 2.438482886216466,
+      "grad_norm": 0.10784180462360382,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 23724
+    },
+    {
+      "epoch": 2.438791242676534,
+      "grad_norm": 0.0911094918847084,
+      "learning_rate": 0.01,
+      "loss": 2.0594,
+      "step": 23727
+    },
+    {
+      "epoch": 2.439099599136602,
+      "grad_norm": 0.059867773205041885,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 23730
+    },
+    {
+      "epoch": 2.43940795559667,
+      "grad_norm": 0.06214331462979317,
+      "learning_rate": 0.01,
+      "loss": 2.0698,
+      "step": 23733
+    },
+    {
+      "epoch": 2.4397163120567376,
+      "grad_norm": 0.058527860790491104,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 23736
+    },
+    {
+      "epoch": 2.4400246685168057,
+      "grad_norm": 0.09416967630386353,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 23739
+    },
+    {
+      "epoch": 2.4403330249768733,
+      "grad_norm": 0.11225190758705139,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 23742
+    },
+    {
+      "epoch": 2.440641381436941,
+      "grad_norm": 0.056495241820812225,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 23745
+    },
+    {
+      "epoch": 2.440949737897009,
+      "grad_norm": 0.05432302877306938,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 23748
+    },
+    {
+      "epoch": 2.4412580943570767,
+      "grad_norm": 0.04850140959024429,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 23751
+    },
+    {
+      "epoch": 2.4415664508171444,
+      "grad_norm": 0.05038965865969658,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 23754
+    },
+    {
+      "epoch": 2.4418748072772125,
+      "grad_norm": 0.050226423889398575,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 23757
+    },
+    {
+      "epoch": 2.44218316373728,
+      "grad_norm": 0.05979544296860695,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 23760
+    },
+    {
+      "epoch": 2.4424915201973483,
+      "grad_norm": 0.04254556819796562,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 23763
+    },
+    {
+      "epoch": 2.442799876657416,
+      "grad_norm": 0.1078273206949234,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 23766
+    },
+    {
+      "epoch": 2.443108233117484,
+      "grad_norm": 0.0591372586786747,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 23769
+    },
+    {
+      "epoch": 2.4434165895775517,
+      "grad_norm": 0.05430880934000015,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 23772
+    },
+    {
+      "epoch": 2.4437249460376194,
+      "grad_norm": 0.07246505469083786,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 23775
+    },
+    {
+      "epoch": 2.4440333024976875,
+      "grad_norm": 0.04457786679267883,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 23778
+    },
+    {
+      "epoch": 2.444341658957755,
+      "grad_norm": 0.03915979340672493,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 23781
+    },
+    {
+      "epoch": 2.444650015417823,
+      "grad_norm": 0.08284246176481247,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 23784
+    },
+    {
+      "epoch": 2.444958371877891,
+      "grad_norm": 0.08254294097423553,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 23787
+    },
+    {
+      "epoch": 2.4452667283379586,
+      "grad_norm": 0.12263736873865128,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 23790
+    },
+    {
+      "epoch": 2.4455750847980267,
+      "grad_norm": 0.08192913979291916,
+      "learning_rate": 0.01,
+      "loss": 2.047,
+      "step": 23793
+    },
+    {
+      "epoch": 2.4458834412580943,
+      "grad_norm": 0.09507100284099579,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 23796
+    },
+    {
+      "epoch": 2.4461917977181624,
+      "grad_norm": 0.10554829984903336,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 23799
+    },
+    {
+      "epoch": 2.44650015417823,
+      "grad_norm": 0.051542408764362335,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 23802
+    },
+    {
+      "epoch": 2.4468085106382977,
+      "grad_norm": 0.05467440187931061,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 23805
+    },
+    {
+      "epoch": 2.447116867098366,
+      "grad_norm": 0.033462487161159515,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 23808
+    },
+    {
+      "epoch": 2.4474252235584335,
+      "grad_norm": 0.04688438028097153,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 23811
+    },
+    {
+      "epoch": 2.447733580018501,
+      "grad_norm": 0.04699881002306938,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 23814
+    },
+    {
+      "epoch": 2.4480419364785693,
+      "grad_norm": 0.03932753950357437,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 23817
+    },
+    {
+      "epoch": 2.448350292938637,
+      "grad_norm": 0.11096165329217911,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 23820
+    },
+    {
+      "epoch": 2.448658649398705,
+      "grad_norm": 0.03774998337030411,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 23823
+    },
+    {
+      "epoch": 2.4489670058587727,
+      "grad_norm": 0.03317665681242943,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 23826
+    },
+    {
+      "epoch": 2.449275362318841,
+      "grad_norm": 0.06077956408262253,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 23829
+    },
+    {
+      "epoch": 2.4495837187789085,
+      "grad_norm": 0.08997214585542679,
+      "learning_rate": 0.01,
+      "loss": 2.0451,
+      "step": 23832
+    },
+    {
+      "epoch": 2.449892075238976,
+      "grad_norm": 0.12234194576740265,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 23835
+    },
+    {
+      "epoch": 2.4502004316990442,
+      "grad_norm": 0.13833922147750854,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 23838
+    },
+    {
+      "epoch": 2.450508788159112,
+      "grad_norm": 0.06707292795181274,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 23841
+    },
+    {
+      "epoch": 2.4508171446191795,
+      "grad_norm": 0.04679076373577118,
+      "learning_rate": 0.01,
+      "loss": 2.0624,
+      "step": 23844
+    },
+    {
+      "epoch": 2.4511255010792476,
+      "grad_norm": 0.08921289443969727,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 23847
+    },
+    {
+      "epoch": 2.4514338575393153,
+      "grad_norm": 0.056680746376514435,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 23850
+    },
+    {
+      "epoch": 2.4517422139993834,
+      "grad_norm": 0.07349438965320587,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 23853
+    },
+    {
+      "epoch": 2.452050570459451,
+      "grad_norm": 0.057414181530475616,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 23856
+    },
+    {
+      "epoch": 2.452358926919519,
+      "grad_norm": 0.09144090861082077,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 23859
+    },
+    {
+      "epoch": 2.452667283379587,
+      "grad_norm": 0.05499831214547157,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 23862
+    },
+    {
+      "epoch": 2.4529756398396545,
+      "grad_norm": 0.03803296014666557,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 23865
+    },
+    {
+      "epoch": 2.4532839962997226,
+      "grad_norm": 0.11475666612386703,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 23868
+    },
+    {
+      "epoch": 2.4535923527597903,
+      "grad_norm": 0.13063554465770721,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 23871
+    },
+    {
+      "epoch": 2.453900709219858,
+      "grad_norm": 0.04982873052358627,
+      "learning_rate": 0.01,
+      "loss": 2.0718,
+      "step": 23874
+    },
+    {
+      "epoch": 2.454209065679926,
+      "grad_norm": 0.04090685769915581,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 23877
+    },
+    {
+      "epoch": 2.4545174221399937,
+      "grad_norm": 0.044233810156583786,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 23880
+    },
+    {
+      "epoch": 2.454825778600062,
+      "grad_norm": 0.045451819896698,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 23883
+    },
+    {
+      "epoch": 2.4551341350601295,
+      "grad_norm": 0.04413032531738281,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 23886
+    },
+    {
+      "epoch": 2.4554424915201976,
+      "grad_norm": 0.07937499135732651,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 23889
+    },
+    {
+      "epoch": 2.455750847980265,
+      "grad_norm": 0.09389673918485641,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 23892
+    },
+    {
+      "epoch": 2.456059204440333,
+      "grad_norm": 0.061937954276800156,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 23895
+    },
+    {
+      "epoch": 2.456367560900401,
+      "grad_norm": 0.038002993911504745,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 23898
+    },
+    {
+      "epoch": 2.4566759173604686,
+      "grad_norm": 0.05679142847657204,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 23901
+    },
+    {
+      "epoch": 2.4569842738205363,
+      "grad_norm": 0.07343320548534393,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 23904
+    },
+    {
+      "epoch": 2.4572926302806044,
+      "grad_norm": 0.04992273077368736,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 23907
+    },
+    {
+      "epoch": 2.457600986740672,
+      "grad_norm": 0.07384973764419556,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 23910
+    },
+    {
+      "epoch": 2.45790934320074,
+      "grad_norm": 0.05728161707520485,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 23913
+    },
+    {
+      "epoch": 2.458217699660808,
+      "grad_norm": 0.06384929269552231,
+      "learning_rate": 0.01,
+      "loss": 2.0471,
+      "step": 23916
+    },
+    {
+      "epoch": 2.458526056120876,
+      "grad_norm": 0.07515307515859604,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 23919
+    },
+    {
+      "epoch": 2.4588344125809436,
+      "grad_norm": 0.09443585574626923,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 23922
+    },
+    {
+      "epoch": 2.4591427690410113,
+      "grad_norm": 0.059930965304374695,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 23925
+    },
+    {
+      "epoch": 2.4594511255010794,
+      "grad_norm": 0.09930091351270676,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 23928
+    },
+    {
+      "epoch": 2.459759481961147,
+      "grad_norm": 0.03487955033779144,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 23931
+    },
+    {
+      "epoch": 2.4600678384212147,
+      "grad_norm": 0.08312834799289703,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 23934
+    },
+    {
+      "epoch": 2.460376194881283,
+      "grad_norm": 0.06459874659776688,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 23937
+    },
+    {
+      "epoch": 2.4606845513413504,
+      "grad_norm": 0.10331536084413528,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 23940
+    },
+    {
+      "epoch": 2.4609929078014185,
+      "grad_norm": 0.06738618016242981,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 23943
+    },
+    {
+      "epoch": 2.461301264261486,
+      "grad_norm": 0.05211193859577179,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 23946
+    },
+    {
+      "epoch": 2.4616096207215543,
+      "grad_norm": 0.05060914158821106,
+      "learning_rate": 0.01,
+      "loss": 2.0518,
+      "step": 23949
+    },
+    {
+      "epoch": 2.461917977181622,
+      "grad_norm": 0.037570733577013016,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 23952
+    },
+    {
+      "epoch": 2.4622263336416896,
+      "grad_norm": 0.06046308949589729,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 23955
+    },
+    {
+      "epoch": 2.4625346901017577,
+      "grad_norm": 0.08765476942062378,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 23958
+    },
+    {
+      "epoch": 2.4628430465618254,
+      "grad_norm": 0.07397017627954483,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 23961
+    },
+    {
+      "epoch": 2.463151403021893,
+      "grad_norm": 0.10959914326667786,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 23964
+    },
+    {
+      "epoch": 2.463459759481961,
+      "grad_norm": 0.04373926669359207,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 23967
+    },
+    {
+      "epoch": 2.463768115942029,
+      "grad_norm": 0.049903422594070435,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 23970
+    },
+    {
+      "epoch": 2.464076472402097,
+      "grad_norm": 0.037193864583969116,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 23973
+    },
+    {
+      "epoch": 2.4643848288621646,
+      "grad_norm": 0.06762266159057617,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 23976
+    },
+    {
+      "epoch": 2.4646931853222327,
+      "grad_norm": 0.03918517008423805,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 23979
+    },
+    {
+      "epoch": 2.4650015417823004,
+      "grad_norm": 0.09819602221250534,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 23982
+    },
+    {
+      "epoch": 2.465309898242368,
+      "grad_norm": 0.12719838321208954,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 23985
+    },
+    {
+      "epoch": 2.465618254702436,
+      "grad_norm": 0.05890420451760292,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 23988
+    },
+    {
+      "epoch": 2.465926611162504,
+      "grad_norm": 0.05101997032761574,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 23991
+    },
+    {
+      "epoch": 2.466234967622572,
+      "grad_norm": 0.04597810283303261,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 23994
+    },
+    {
+      "epoch": 2.4665433240826395,
+      "grad_norm": 0.0504877045750618,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 23997
+    },
+    {
+      "epoch": 2.466851680542707,
+      "grad_norm": 0.07908850163221359,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 24000
+    },
+    {
+      "epoch": 2.4671600370027753,
+      "grad_norm": 0.07396575808525085,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 24003
+    },
+    {
+      "epoch": 2.467468393462843,
+      "grad_norm": 0.06695767492055893,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 24006
+    },
+    {
+      "epoch": 2.467776749922911,
+      "grad_norm": 0.0907633900642395,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 24009
+    },
+    {
+      "epoch": 2.4680851063829787,
+      "grad_norm": 0.07585210353136063,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 24012
+    },
+    {
+      "epoch": 2.4683934628430464,
+      "grad_norm": 0.049894414842128754,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 24015
+    },
+    {
+      "epoch": 2.4687018193031145,
+      "grad_norm": 0.038147564977407455,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 24018
+    },
+    {
+      "epoch": 2.469010175763182,
+      "grad_norm": 0.03651060909032822,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 24021
+    },
+    {
+      "epoch": 2.4693185322232503,
+      "grad_norm": 0.09396708011627197,
+      "learning_rate": 0.01,
+      "loss": 2.0419,
+      "step": 24024
+    },
+    {
+      "epoch": 2.469626888683318,
+      "grad_norm": 0.03474831208586693,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 24027
+    },
+    {
+      "epoch": 2.4699352451433856,
+      "grad_norm": 0.09970984607934952,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 24030
+    },
+    {
+      "epoch": 2.4702436016034537,
+      "grad_norm": 0.10633893311023712,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 24033
+    },
+    {
+      "epoch": 2.4705519580635213,
+      "grad_norm": 0.0787937119603157,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 24036
+    },
+    {
+      "epoch": 2.4708603145235895,
+      "grad_norm": 0.05214501917362213,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 24039
+    },
+    {
+      "epoch": 2.471168670983657,
+      "grad_norm": 0.02967134490609169,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 24042
+    },
+    {
+      "epoch": 2.4714770274437248,
+      "grad_norm": 0.03181852400302887,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 24045
+    },
+    {
+      "epoch": 2.471785383903793,
+      "grad_norm": 0.03905104100704193,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 24048
+    },
+    {
+      "epoch": 2.4720937403638605,
+      "grad_norm": 0.051532018929719925,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 24051
+    },
+    {
+      "epoch": 2.4724020968239286,
+      "grad_norm": 0.13418106734752655,
+      "learning_rate": 0.01,
+      "loss": 2.0486,
+      "step": 24054
+    },
+    {
+      "epoch": 2.4727104532839963,
+      "grad_norm": 0.0745147094130516,
+      "learning_rate": 0.01,
+      "loss": 2.0448,
+      "step": 24057
+    },
+    {
+      "epoch": 2.473018809744064,
+      "grad_norm": 0.06666639447212219,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 24060
+    },
+    {
+      "epoch": 2.473327166204132,
+      "grad_norm": 0.08356054127216339,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 24063
+    },
+    {
+      "epoch": 2.4736355226641997,
+      "grad_norm": 0.04733874648809433,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 24066
+    },
+    {
+      "epoch": 2.473943879124268,
+      "grad_norm": 0.07551899552345276,
+      "learning_rate": 0.01,
+      "loss": 2.0389,
+      "step": 24069
+    },
+    {
+      "epoch": 2.4742522355843355,
+      "grad_norm": 0.03114013373851776,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 24072
+    },
+    {
+      "epoch": 2.474560592044403,
+      "grad_norm": 0.062112826853990555,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 24075
+    },
+    {
+      "epoch": 2.4748689485044713,
+      "grad_norm": 0.0542120561003685,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 24078
+    },
+    {
+      "epoch": 2.475177304964539,
+      "grad_norm": 0.06153399124741554,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 24081
+    },
+    {
+      "epoch": 2.475485661424607,
+      "grad_norm": 0.06306985020637512,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 24084
+    },
+    {
+      "epoch": 2.4757940178846747,
+      "grad_norm": 0.03326687961816788,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 24087
+    },
+    {
+      "epoch": 2.4761023743447423,
+      "grad_norm": 0.10279777646064758,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 24090
+    },
+    {
+      "epoch": 2.4764107308048104,
+      "grad_norm": 0.045618560165166855,
+      "learning_rate": 0.01,
+      "loss": 2.0476,
+      "step": 24093
+    },
+    {
+      "epoch": 2.476719087264878,
+      "grad_norm": 0.05065792426466942,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 24096
+    },
+    {
+      "epoch": 2.477027443724946,
+      "grad_norm": 0.04114675521850586,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 24099
+    },
+    {
+      "epoch": 2.477335800185014,
+      "grad_norm": 0.11044265329837799,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 24102
+    },
+    {
+      "epoch": 2.4776441566450815,
+      "grad_norm": 0.053481362760066986,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 24105
+    },
+    {
+      "epoch": 2.4779525131051496,
+      "grad_norm": 0.14947016537189484,
+      "learning_rate": 0.01,
+      "loss": 2.0511,
+      "step": 24108
+    },
+    {
+      "epoch": 2.4782608695652173,
+      "grad_norm": 0.08009488135576248,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 24111
+    },
+    {
+      "epoch": 2.4785692260252854,
+      "grad_norm": 0.06397935748100281,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 24114
+    },
+    {
+      "epoch": 2.478877582485353,
+      "grad_norm": 0.0409528985619545,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 24117
+    },
+    {
+      "epoch": 2.4791859389454207,
+      "grad_norm": 0.11498477309942245,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 24120
+    },
+    {
+      "epoch": 2.479494295405489,
+      "grad_norm": 0.02937469258904457,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 24123
+    },
+    {
+      "epoch": 2.4798026518655565,
+      "grad_norm": 0.03659180551767349,
+      "learning_rate": 0.01,
+      "loss": 2.0638,
+      "step": 24126
+    },
+    {
+      "epoch": 2.4801110083256246,
+      "grad_norm": 0.10866084694862366,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 24129
+    },
+    {
+      "epoch": 2.4804193647856922,
+      "grad_norm": 0.07002250105142593,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 24132
+    },
+    {
+      "epoch": 2.48072772124576,
+      "grad_norm": 0.07129926234483719,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 24135
+    },
+    {
+      "epoch": 2.481036077705828,
+      "grad_norm": 0.061107151210308075,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 24138
+    },
+    {
+      "epoch": 2.4813444341658957,
+      "grad_norm": 0.04937992990016937,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 24141
+    },
+    {
+      "epoch": 2.481652790625964,
+      "grad_norm": 0.04651058465242386,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 24144
+    },
+    {
+      "epoch": 2.4819611470860314,
+      "grad_norm": 0.06658688187599182,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 24147
+    },
+    {
+      "epoch": 2.482269503546099,
+      "grad_norm": 0.05241367965936661,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 24150
+    },
+    {
+      "epoch": 2.482577860006167,
+      "grad_norm": 0.03564739227294922,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 24153
+    },
+    {
+      "epoch": 2.482886216466235,
+      "grad_norm": 0.03984629735350609,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 24156
+    },
+    {
+      "epoch": 2.483194572926303,
+      "grad_norm": 0.10245262831449509,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 24159
+    },
+    {
+      "epoch": 2.4835029293863706,
+      "grad_norm": 0.0727042481303215,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 24162
+    },
+    {
+      "epoch": 2.4838112858464383,
+      "grad_norm": 0.041190728545188904,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 24165
+    },
+    {
+      "epoch": 2.4841196423065064,
+      "grad_norm": 0.04986109584569931,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 24168
+    },
+    {
+      "epoch": 2.484427998766574,
+      "grad_norm": 0.0689210444688797,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 24171
+    },
+    {
+      "epoch": 2.484736355226642,
+      "grad_norm": 0.043997250497341156,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 24174
+    },
+    {
+      "epoch": 2.48504471168671,
+      "grad_norm": 0.06231982633471489,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 24177
+    },
+    {
+      "epoch": 2.485353068146778,
+      "grad_norm": 0.09958053380250931,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 24180
+    },
+    {
+      "epoch": 2.4856614246068456,
+      "grad_norm": 0.05297970771789551,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 24183
+    },
+    {
+      "epoch": 2.4859697810669132,
+      "grad_norm": 0.07420172542333603,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 24186
+    },
+    {
+      "epoch": 2.4862781375269813,
+      "grad_norm": 0.049017585813999176,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 24189
+    },
+    {
+      "epoch": 2.486586493987049,
+      "grad_norm": 0.05398377403616905,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 24192
+    },
+    {
+      "epoch": 2.4868948504471167,
+      "grad_norm": 0.03338189795613289,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 24195
+    },
+    {
+      "epoch": 2.4872032069071848,
+      "grad_norm": 0.1013825535774231,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 24198
+    },
+    {
+      "epoch": 2.4875115633672524,
+      "grad_norm": 0.06101495400071144,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 24201
+    },
+    {
+      "epoch": 2.4878199198273205,
+      "grad_norm": 0.06915189325809479,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 24204
+    },
+    {
+      "epoch": 2.488128276287388,
+      "grad_norm": 0.09961054474115372,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 24207
+    },
+    {
+      "epoch": 2.4884366327474563,
+      "grad_norm": 0.039923045784235,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 24210
+    },
+    {
+      "epoch": 2.488744989207524,
+      "grad_norm": 0.07982566952705383,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 24213
+    },
+    {
+      "epoch": 2.4890533456675916,
+      "grad_norm": 0.06360599398612976,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 24216
+    },
+    {
+      "epoch": 2.4893617021276597,
+      "grad_norm": 0.06489767879247665,
+      "learning_rate": 0.01,
+      "loss": 2.0615,
+      "step": 24219
+    },
+    {
+      "epoch": 2.4896700585877274,
+      "grad_norm": 0.08873300999403,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 24222
+    },
+    {
+      "epoch": 2.489978415047795,
+      "grad_norm": 0.04309386387467384,
+      "learning_rate": 0.01,
+      "loss": 2.0478,
+      "step": 24225
+    },
+    {
+      "epoch": 2.490286771507863,
+      "grad_norm": 0.042991675436496735,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 24228
+    },
+    {
+      "epoch": 2.490595127967931,
+      "grad_norm": 0.04507692903280258,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 24231
+    },
+    {
+      "epoch": 2.490903484427999,
+      "grad_norm": 0.09288784116506577,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 24234
+    },
+    {
+      "epoch": 2.4912118408880666,
+      "grad_norm": 0.040776100009679794,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 24237
+    },
+    {
+      "epoch": 2.4915201973481347,
+      "grad_norm": 0.09847230464220047,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 24240
+    },
+    {
+      "epoch": 2.4918285538082023,
+      "grad_norm": 0.05252716690301895,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 24243
+    },
+    {
+      "epoch": 2.49213691026827,
+      "grad_norm": 0.04028663411736488,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 24246
+    },
+    {
+      "epoch": 2.492445266728338,
+      "grad_norm": 0.04721330851316452,
+      "learning_rate": 0.01,
+      "loss": 2.045,
+      "step": 24249
+    },
+    {
+      "epoch": 2.4927536231884058,
+      "grad_norm": 0.0453517884016037,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 24252
+    },
+    {
+      "epoch": 2.4930619796484734,
+      "grad_norm": 0.09120344370603561,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 24255
+    },
+    {
+      "epoch": 2.4933703361085415,
+      "grad_norm": 0.1132507398724556,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 24258
+    },
+    {
+      "epoch": 2.493678692568609,
+      "grad_norm": 0.07698217034339905,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 24261
+    },
+    {
+      "epoch": 2.4939870490286773,
+      "grad_norm": 0.04104442894458771,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 24264
+    },
+    {
+      "epoch": 2.494295405488745,
+      "grad_norm": 0.04681272804737091,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 24267
+    },
+    {
+      "epoch": 2.494603761948813,
+      "grad_norm": 0.05137484520673752,
+      "learning_rate": 0.01,
+      "loss": 2.0616,
+      "step": 24270
+    },
+    {
+      "epoch": 2.4949121184088807,
+      "grad_norm": 0.05654723569750786,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 24273
+    },
+    {
+      "epoch": 2.4952204748689484,
+      "grad_norm": 0.052294518798589706,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 24276
+    },
+    {
+      "epoch": 2.4955288313290165,
+      "grad_norm": 0.06906304508447647,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 24279
+    },
+    {
+      "epoch": 2.495837187789084,
+      "grad_norm": 0.0664992555975914,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 24282
+    },
+    {
+      "epoch": 2.496145544249152,
+      "grad_norm": 0.06502712517976761,
+      "learning_rate": 0.01,
+      "loss": 2.0439,
+      "step": 24285
+    },
+    {
+      "epoch": 2.49645390070922,
+      "grad_norm": 0.047290291637182236,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 24288
+    },
+    {
+      "epoch": 2.4967622571692876,
+      "grad_norm": 0.07271420210599899,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 24291
+    },
+    {
+      "epoch": 2.4970706136293557,
+      "grad_norm": 0.042307768017053604,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 24294
+    },
+    {
+      "epoch": 2.4973789700894233,
+      "grad_norm": 0.04834264889359474,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 24297
+    },
+    {
+      "epoch": 2.4976873265494914,
+      "grad_norm": 0.045435305684804916,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 24300
+    },
+    {
+      "epoch": 2.497995683009559,
+      "grad_norm": 0.07511632144451141,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 24303
+    },
+    {
+      "epoch": 2.4983040394696268,
+      "grad_norm": 0.14621250331401825,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 24306
+    },
+    {
+      "epoch": 2.498612395929695,
+      "grad_norm": 0.05969877541065216,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 24309
+    },
+    {
+      "epoch": 2.4989207523897625,
+      "grad_norm": 0.040014345198869705,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 24312
+    },
+    {
+      "epoch": 2.49922910884983,
+      "grad_norm": 0.05185026675462723,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 24315
+    },
+    {
+      "epoch": 2.4995374653098983,
+      "grad_norm": 0.04459863528609276,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 24318
+    },
+    {
+      "epoch": 2.499845821769966,
+      "grad_norm": 0.06208227947354317,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 24321
+    },
+    {
+      "epoch": 2.500154178230034,
+      "grad_norm": 0.0788293108344078,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 24324
+    },
+    {
+      "epoch": 2.5004625346901017,
+      "grad_norm": 0.0636831745505333,
+      "learning_rate": 0.01,
+      "loss": 2.0522,
+      "step": 24327
+    },
+    {
+      "epoch": 2.50077089115017,
+      "grad_norm": 0.05828903615474701,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 24330
+    },
+    {
+      "epoch": 2.5010792476102375,
+      "grad_norm": 0.06897569447755814,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 24333
+    },
+    {
+      "epoch": 2.501387604070305,
+      "grad_norm": 0.03793172910809517,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 24336
+    },
+    {
+      "epoch": 2.5016959605303732,
+      "grad_norm": 0.05384537950158119,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 24339
+    },
+    {
+      "epoch": 2.502004316990441,
+      "grad_norm": 0.07979964464902878,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 24342
+    },
+    {
+      "epoch": 2.5023126734505086,
+      "grad_norm": 0.054392259567976,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 24345
+    },
+    {
+      "epoch": 2.5026210299105767,
+      "grad_norm": 0.06897418200969696,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 24348
+    },
+    {
+      "epoch": 2.5029293863706443,
+      "grad_norm": 0.0801873579621315,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 24351
+    },
+    {
+      "epoch": 2.5032377428307124,
+      "grad_norm": 0.03708551451563835,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 24354
+    },
+    {
+      "epoch": 2.50354609929078,
+      "grad_norm": 0.09656143933534622,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 24357
+    },
+    {
+      "epoch": 2.503854455750848,
+      "grad_norm": 0.0770551860332489,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 24360
+    },
+    {
+      "epoch": 2.504162812210916,
+      "grad_norm": 0.08103878796100616,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 24363
+    },
+    {
+      "epoch": 2.5044711686709835,
+      "grad_norm": 0.10928300768136978,
+      "learning_rate": 0.01,
+      "loss": 2.0389,
+      "step": 24366
+    },
+    {
+      "epoch": 2.5047795251310516,
+      "grad_norm": 0.07541976869106293,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 24369
+    },
+    {
+      "epoch": 2.5050878815911193,
+      "grad_norm": 0.05937611311674118,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 24372
+    },
+    {
+      "epoch": 2.505396238051187,
+      "grad_norm": 0.05396249517798424,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 24375
+    },
+    {
+      "epoch": 2.505704594511255,
+      "grad_norm": 0.04004419967532158,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 24378
+    },
+    {
+      "epoch": 2.5060129509713227,
+      "grad_norm": 0.05242707580327988,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 24381
+    },
+    {
+      "epoch": 2.506321307431391,
+      "grad_norm": 0.038752540946006775,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 24384
+    },
+    {
+      "epoch": 2.5066296638914585,
+      "grad_norm": 0.07296596467494965,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 24387
+    },
+    {
+      "epoch": 2.5069380203515266,
+      "grad_norm": 0.116209976375103,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 24390
+    },
+    {
+      "epoch": 2.5072463768115942,
+      "grad_norm": 0.09165041148662567,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 24393
+    },
+    {
+      "epoch": 2.507554733271662,
+      "grad_norm": 0.14264173805713654,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 24396
+    },
+    {
+      "epoch": 2.50786308973173,
+      "grad_norm": 0.045917656272649765,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 24399
+    },
+    {
+      "epoch": 2.5081714461917977,
+      "grad_norm": 0.057148344814777374,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 24402
+    },
+    {
+      "epoch": 2.5084798026518653,
+      "grad_norm": 0.03626836836338043,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 24405
+    },
+    {
+      "epoch": 2.5087881591119334,
+      "grad_norm": 0.03621996194124222,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 24408
+    },
+    {
+      "epoch": 2.509096515572001,
+      "grad_norm": 0.05835467949509621,
+      "learning_rate": 0.01,
+      "loss": 2.0506,
+      "step": 24411
+    },
+    {
+      "epoch": 2.509404872032069,
+      "grad_norm": 0.03973361849784851,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 24414
+    },
+    {
+      "epoch": 2.509713228492137,
+      "grad_norm": 0.09463023394346237,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 24417
+    },
+    {
+      "epoch": 2.510021584952205,
+      "grad_norm": 0.07552462071180344,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 24420
+    },
+    {
+      "epoch": 2.5103299414122726,
+      "grad_norm": 0.1288609802722931,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 24423
+    },
+    {
+      "epoch": 2.5106382978723403,
+      "grad_norm": 0.06943691521883011,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 24426
+    },
+    {
+      "epoch": 2.5109466543324084,
+      "grad_norm": 0.08581320196390152,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 24429
+    },
+    {
+      "epoch": 2.511255010792476,
+      "grad_norm": 0.07379914820194244,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 24432
+    },
+    {
+      "epoch": 2.5115633672525437,
+      "grad_norm": 0.09235703945159912,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 24435
+    },
+    {
+      "epoch": 2.511871723712612,
+      "grad_norm": 0.05038774758577347,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 24438
+    },
+    {
+      "epoch": 2.5121800801726795,
+      "grad_norm": 0.08711019903421402,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 24441
+    },
+    {
+      "epoch": 2.5124884366327476,
+      "grad_norm": 0.05432825908064842,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 24444
+    },
+    {
+      "epoch": 2.5127967930928152,
+      "grad_norm": 0.10115527361631393,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 24447
+    },
+    {
+      "epoch": 2.5131051495528833,
+      "grad_norm": 0.10387013852596283,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 24450
+    },
+    {
+      "epoch": 2.513413506012951,
+      "grad_norm": 0.0740542933344841,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 24453
+    },
+    {
+      "epoch": 2.5137218624730187,
+      "grad_norm": 0.07834311574697495,
+      "learning_rate": 0.01,
+      "loss": 2.0731,
+      "step": 24456
+    },
+    {
+      "epoch": 2.5140302189330868,
+      "grad_norm": 0.06743736565113068,
+      "learning_rate": 0.01,
+      "loss": 2.0588,
+      "step": 24459
+    },
+    {
+      "epoch": 2.5143385753931544,
+      "grad_norm": 0.051791246980428696,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 24462
+    },
+    {
+      "epoch": 2.514646931853222,
+      "grad_norm": 0.0731598362326622,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 24465
+    },
+    {
+      "epoch": 2.51495528831329,
+      "grad_norm": 0.04995987191796303,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 24468
+    },
+    {
+      "epoch": 2.515263644773358,
+      "grad_norm": 0.04953973367810249,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 24471
+    },
+    {
+      "epoch": 2.515572001233426,
+      "grad_norm": 0.05432116240262985,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 24474
+    },
+    {
+      "epoch": 2.5158803576934936,
+      "grad_norm": 0.048791225999593735,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 24477
+    },
+    {
+      "epoch": 2.5161887141535617,
+      "grad_norm": 0.11742904037237167,
+      "learning_rate": 0.01,
+      "loss": 2.04,
+      "step": 24480
+    },
+    {
+      "epoch": 2.5164970706136294,
+      "grad_norm": 0.12617075443267822,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 24483
+    },
+    {
+      "epoch": 2.516805427073697,
+      "grad_norm": 0.051573995500802994,
+      "learning_rate": 0.01,
+      "loss": 2.0662,
+      "step": 24486
+    },
+    {
+      "epoch": 2.517113783533765,
+      "grad_norm": 0.09131506085395813,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 24489
+    },
+    {
+      "epoch": 2.517422139993833,
+      "grad_norm": 0.06593006104230881,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 24492
+    },
+    {
+      "epoch": 2.5177304964539005,
+      "grad_norm": 0.038310687988996506,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 24495
+    },
+    {
+      "epoch": 2.5180388529139686,
+      "grad_norm": 0.05975675210356712,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 24498
+    },
+    {
+      "epoch": 2.518347209374036,
+      "grad_norm": 0.04541294649243355,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 24501
+    },
+    {
+      "epoch": 2.5186555658341043,
+      "grad_norm": 0.053723473101854324,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 24504
+    },
+    {
+      "epoch": 2.518963922294172,
+      "grad_norm": 0.030095964670181274,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 24507
+    },
+    {
+      "epoch": 2.51927227875424,
+      "grad_norm": 0.10244923830032349,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 24510
+    },
+    {
+      "epoch": 2.5195806352143078,
+      "grad_norm": 0.06249944120645523,
+      "learning_rate": 0.01,
+      "loss": 2.0449,
+      "step": 24513
+    },
+    {
+      "epoch": 2.5198889916743754,
+      "grad_norm": 0.08720767498016357,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 24516
+    },
+    {
+      "epoch": 2.5201973481344435,
+      "grad_norm": 0.07686194777488708,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 24519
+    },
+    {
+      "epoch": 2.520505704594511,
+      "grad_norm": 0.08837150782346725,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 24522
+    },
+    {
+      "epoch": 2.520814061054579,
+      "grad_norm": 0.0794796347618103,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 24525
+    },
+    {
+      "epoch": 2.521122417514647,
+      "grad_norm": 0.05655858293175697,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 24528
+    },
+    {
+      "epoch": 2.521430773974715,
+      "grad_norm": 0.08295401185750961,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 24531
+    },
+    {
+      "epoch": 2.5217391304347827,
+      "grad_norm": 0.03982521593570709,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 24534
+    },
+    {
+      "epoch": 2.5220474868948504,
+      "grad_norm": 0.04791923984885216,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 24537
+    },
+    {
+      "epoch": 2.5223558433549185,
+      "grad_norm": 0.11436691880226135,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 24540
+    },
+    {
+      "epoch": 2.522664199814986,
+      "grad_norm": 0.1064198762178421,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 24543
+    },
+    {
+      "epoch": 2.522972556275054,
+      "grad_norm": 0.08036024123430252,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 24546
+    },
+    {
+      "epoch": 2.523280912735122,
+      "grad_norm": 0.061799556016922,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 24549
+    },
+    {
+      "epoch": 2.5235892691951896,
+      "grad_norm": 0.04592469707131386,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 24552
+    },
+    {
+      "epoch": 2.523897625655257,
+      "grad_norm": 0.036766473203897476,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 24555
+    },
+    {
+      "epoch": 2.5242059821153253,
+      "grad_norm": 0.09619138389825821,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 24558
+    },
+    {
+      "epoch": 2.5245143385753934,
+      "grad_norm": 0.03915918245911598,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 24561
+    },
+    {
+      "epoch": 2.524822695035461,
+      "grad_norm": 0.04883084446191788,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 24564
+    },
+    {
+      "epoch": 2.5251310514955287,
+      "grad_norm": 0.05630512908101082,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 24567
+    },
+    {
+      "epoch": 2.525439407955597,
+      "grad_norm": 0.04064425081014633,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 24570
+    },
+    {
+      "epoch": 2.5257477644156645,
+      "grad_norm": 0.05302917957305908,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 24573
+    },
+    {
+      "epoch": 2.526056120875732,
+      "grad_norm": 0.07677201181650162,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 24576
+    },
+    {
+      "epoch": 2.5263644773358003,
+      "grad_norm": 0.07950242608785629,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 24579
+    },
+    {
+      "epoch": 2.526672833795868,
+      "grad_norm": 0.07068518549203873,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 24582
+    },
+    {
+      "epoch": 2.5269811902559356,
+      "grad_norm": 0.08623625338077545,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 24585
+    },
+    {
+      "epoch": 2.5272895467160037,
+      "grad_norm": 0.05549190193414688,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 24588
+    },
+    {
+      "epoch": 2.527597903176072,
+      "grad_norm": 0.05710297450423241,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 24591
+    },
+    {
+      "epoch": 2.5279062596361395,
+      "grad_norm": 0.05071646347641945,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 24594
+    },
+    {
+      "epoch": 2.528214616096207,
+      "grad_norm": 0.09765972197055817,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 24597
+    },
+    {
+      "epoch": 2.5285229725562752,
+      "grad_norm": 0.05874921754002571,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 24600
+    },
+    {
+      "epoch": 2.528831329016343,
+      "grad_norm": 0.10598991811275482,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 24603
+    },
+    {
+      "epoch": 2.5291396854764105,
+      "grad_norm": 0.07071609050035477,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 24606
+    },
+    {
+      "epoch": 2.5294480419364787,
+      "grad_norm": 0.07132923603057861,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 24609
+    },
+    {
+      "epoch": 2.5297563983965463,
+      "grad_norm": 0.06741276383399963,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 24612
+    },
+    {
+      "epoch": 2.530064754856614,
+      "grad_norm": 0.10399371385574341,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 24615
+    },
+    {
+      "epoch": 2.530373111316682,
+      "grad_norm": 0.054513610899448395,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 24618
+    },
+    {
+      "epoch": 2.53068146777675,
+      "grad_norm": 0.03990021347999573,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 24621
+    },
+    {
+      "epoch": 2.530989824236818,
+      "grad_norm": 0.0329439677298069,
+      "learning_rate": 0.01,
+      "loss": 2.0653,
+      "step": 24624
+    },
+    {
+      "epoch": 2.5312981806968855,
+      "grad_norm": 0.08065532892942429,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 24627
+    },
+    {
+      "epoch": 2.5316065371569536,
+      "grad_norm": 0.04455409198999405,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 24630
+    },
+    {
+      "epoch": 2.5319148936170213,
+      "grad_norm": 0.09395566582679749,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 24633
+    },
+    {
+      "epoch": 2.532223250077089,
+      "grad_norm": 0.04042106121778488,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 24636
+    },
+    {
+      "epoch": 2.532531606537157,
+      "grad_norm": 0.09208521991968155,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 24639
+    },
+    {
+      "epoch": 2.5328399629972247,
+      "grad_norm": 0.06603435426950455,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 24642
+    },
+    {
+      "epoch": 2.533148319457293,
+      "grad_norm": 0.039963483810424805,
+      "learning_rate": 0.01,
+      "loss": 2.0491,
+      "step": 24645
+    },
+    {
+      "epoch": 2.5334566759173605,
+      "grad_norm": 0.14821624755859375,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 24648
+    },
+    {
+      "epoch": 2.5337650323774286,
+      "grad_norm": 0.06644035130739212,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 24651
+    },
+    {
+      "epoch": 2.534073388837496,
+      "grad_norm": 0.0375928059220314,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 24654
+    },
+    {
+      "epoch": 2.534381745297564,
+      "grad_norm": 0.06041393801569939,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 24657
+    },
+    {
+      "epoch": 2.534690101757632,
+      "grad_norm": 0.06117352098226547,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 24660
+    },
+    {
+      "epoch": 2.5349984582176996,
+      "grad_norm": 0.05386986956000328,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 24663
+    },
+    {
+      "epoch": 2.5353068146777673,
+      "grad_norm": 0.03399750217795372,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 24666
+    },
+    {
+      "epoch": 2.5356151711378354,
+      "grad_norm": 0.06256785988807678,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 24669
+    },
+    {
+      "epoch": 2.535923527597903,
+      "grad_norm": 0.08575739711523056,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 24672
+    },
+    {
+      "epoch": 2.536231884057971,
+      "grad_norm": 0.0895959809422493,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 24675
+    },
+    {
+      "epoch": 2.536540240518039,
+      "grad_norm": 0.06579075753688812,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 24678
+    },
+    {
+      "epoch": 2.536848596978107,
+      "grad_norm": 0.04509506747126579,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 24681
+    },
+    {
+      "epoch": 2.5371569534381746,
+      "grad_norm": 0.03535350412130356,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 24684
+    },
+    {
+      "epoch": 2.5374653098982423,
+      "grad_norm": 0.03496406227350235,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 24687
+    },
+    {
+      "epoch": 2.5377736663583104,
+      "grad_norm": 0.04595872759819031,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 24690
+    },
+    {
+      "epoch": 2.538082022818378,
+      "grad_norm": 0.07009676098823547,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 24693
+    },
+    {
+      "epoch": 2.5383903792784457,
+      "grad_norm": 0.07328460365533829,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 24696
+    },
+    {
+      "epoch": 2.538698735738514,
+      "grad_norm": 0.09521552175283432,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 24699
+    },
+    {
+      "epoch": 2.5390070921985815,
+      "grad_norm": 0.09087500721216202,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 24702
+    },
+    {
+      "epoch": 2.5393154486586496,
+      "grad_norm": 0.05657880753278732,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 24705
+    },
+    {
+      "epoch": 2.539623805118717,
+      "grad_norm": 0.13524407148361206,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 24708
+    },
+    {
+      "epoch": 2.5399321615787853,
+      "grad_norm": 0.04498621076345444,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 24711
+    },
+    {
+      "epoch": 2.540240518038853,
+      "grad_norm": 0.04117140173912048,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 24714
+    },
+    {
+      "epoch": 2.5405488744989206,
+      "grad_norm": 0.03630746528506279,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 24717
+    },
+    {
+      "epoch": 2.5408572309589887,
+      "grad_norm": 0.03791969269514084,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 24720
+    },
+    {
+      "epoch": 2.5411655874190564,
+      "grad_norm": 0.045213595032691956,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 24723
+    },
+    {
+      "epoch": 2.541473943879124,
+      "grad_norm": 0.08232447504997253,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 24726
+    },
+    {
+      "epoch": 2.541782300339192,
+      "grad_norm": 0.0790674090385437,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 24729
+    },
+    {
+      "epoch": 2.54209065679926,
+      "grad_norm": 0.09643759578466415,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 24732
+    },
+    {
+      "epoch": 2.542399013259328,
+      "grad_norm": 0.09790430217981339,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 24735
+    },
+    {
+      "epoch": 2.5427073697193956,
+      "grad_norm": 0.04904096946120262,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 24738
+    },
+    {
+      "epoch": 2.5430157261794637,
+      "grad_norm": 0.042802706360816956,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 24741
+    },
+    {
+      "epoch": 2.5433240826395314,
+      "grad_norm": 0.04947663098573685,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 24744
+    },
+    {
+      "epoch": 2.543632439099599,
+      "grad_norm": 0.040841687470674515,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 24747
+    },
+    {
+      "epoch": 2.543940795559667,
+      "grad_norm": 0.051419809460639954,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 24750
+    },
+    {
+      "epoch": 2.544249152019735,
+      "grad_norm": 0.07173865288496017,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 24753
+    },
+    {
+      "epoch": 2.5445575084798024,
+      "grad_norm": 0.07664339989423752,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 24756
+    },
+    {
+      "epoch": 2.5448658649398705,
+      "grad_norm": 0.05180468037724495,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 24759
+    },
+    {
+      "epoch": 2.545174221399938,
+      "grad_norm": 0.03839515894651413,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 24762
+    },
+    {
+      "epoch": 2.5454825778600063,
+      "grad_norm": 0.08712394535541534,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 24765
+    },
+    {
+      "epoch": 2.545790934320074,
+      "grad_norm": 0.06906873732805252,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 24768
+    },
+    {
+      "epoch": 2.546099290780142,
+      "grad_norm": 0.04779994115233421,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 24771
+    },
+    {
+      "epoch": 2.5464076472402097,
+      "grad_norm": 0.03945513069629669,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 24774
+    },
+    {
+      "epoch": 2.5467160037002774,
+      "grad_norm": 0.04089882969856262,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 24777
+    },
+    {
+      "epoch": 2.5470243601603455,
+      "grad_norm": 0.04492718353867531,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 24780
+    },
+    {
+      "epoch": 2.547332716620413,
+      "grad_norm": 0.0761101022362709,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 24783
+    },
+    {
+      "epoch": 2.547641073080481,
+      "grad_norm": 0.09586388617753983,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 24786
+    },
+    {
+      "epoch": 2.547949429540549,
+      "grad_norm": 0.0410308912396431,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 24789
+    },
+    {
+      "epoch": 2.5482577860006166,
+      "grad_norm": 0.0583110935986042,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 24792
+    },
+    {
+      "epoch": 2.5485661424606847,
+      "grad_norm": 0.03310194984078407,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 24795
+    },
+    {
+      "epoch": 2.5488744989207524,
+      "grad_norm": 0.0849560797214508,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 24798
+    },
+    {
+      "epoch": 2.5491828553808205,
+      "grad_norm": 0.052898190915584564,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 24801
+    },
+    {
+      "epoch": 2.549491211840888,
+      "grad_norm": 0.09630381315946579,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 24804
+    },
+    {
+      "epoch": 2.5497995683009558,
+      "grad_norm": 0.04892333596944809,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 24807
+    },
+    {
+      "epoch": 2.550107924761024,
+      "grad_norm": 0.09465577453374863,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 24810
+    },
+    {
+      "epoch": 2.5504162812210915,
+      "grad_norm": 0.0832308977842331,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 24813
+    },
+    {
+      "epoch": 2.550724637681159,
+      "grad_norm": 0.11276236176490784,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 24816
+    },
+    {
+      "epoch": 2.5510329941412273,
+      "grad_norm": 0.08327414095401764,
+      "learning_rate": 0.01,
+      "loss": 2.0607,
+      "step": 24819
+    },
+    {
+      "epoch": 2.551341350601295,
+      "grad_norm": 0.05502014979720116,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 24822
+    },
+    {
+      "epoch": 2.551649707061363,
+      "grad_norm": 0.03681863471865654,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 24825
+    },
+    {
+      "epoch": 2.5519580635214307,
+      "grad_norm": 0.08096860349178314,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 24828
+    },
+    {
+      "epoch": 2.552266419981499,
+      "grad_norm": 0.05901675671339035,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 24831
+    },
+    {
+      "epoch": 2.5525747764415665,
+      "grad_norm": 0.09850065410137177,
+      "learning_rate": 0.01,
+      "loss": 2.0495,
+      "step": 24834
+    },
+    {
+      "epoch": 2.552883132901634,
+      "grad_norm": 0.08438712358474731,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 24837
+    },
+    {
+      "epoch": 2.5531914893617023,
+      "grad_norm": 0.04949135333299637,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 24840
+    },
+    {
+      "epoch": 2.55349984582177,
+      "grad_norm": 0.044099997729063034,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 24843
+    },
+    {
+      "epoch": 2.5538082022818376,
+      "grad_norm": 0.08626649528741837,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 24846
+    },
+    {
+      "epoch": 2.5541165587419057,
+      "grad_norm": 0.08545881509780884,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 24849
+    },
+    {
+      "epoch": 2.5544249152019733,
+      "grad_norm": 0.06181343272328377,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 24852
+    },
+    {
+      "epoch": 2.5547332716620414,
+      "grad_norm": 0.0839785784482956,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 24855
+    },
+    {
+      "epoch": 2.555041628122109,
+      "grad_norm": 0.055504992604255676,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 24858
+    },
+    {
+      "epoch": 2.555349984582177,
+      "grad_norm": 0.04236135631799698,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 24861
+    },
+    {
+      "epoch": 2.555658341042245,
+      "grad_norm": 0.035614918917417526,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 24864
+    },
+    {
+      "epoch": 2.5559666975023125,
+      "grad_norm": 0.04150492325425148,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 24867
+    },
+    {
+      "epoch": 2.5562750539623806,
+      "grad_norm": 0.07994359731674194,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 24870
+    },
+    {
+      "epoch": 2.5565834104224483,
+      "grad_norm": 0.08954035490751266,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 24873
+    },
+    {
+      "epoch": 2.556891766882516,
+      "grad_norm": 0.1362268626689911,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 24876
+    },
+    {
+      "epoch": 2.557200123342584,
+      "grad_norm": 0.11425944417715073,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 24879
+    },
+    {
+      "epoch": 2.5575084798026517,
+      "grad_norm": 0.07083035260438919,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 24882
+    },
+    {
+      "epoch": 2.55781683626272,
+      "grad_norm": 0.06250528246164322,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 24885
+    },
+    {
+      "epoch": 2.5581251927227875,
+      "grad_norm": 0.047506481409072876,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 24888
+    },
+    {
+      "epoch": 2.5584335491828556,
+      "grad_norm": 0.04237549751996994,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 24891
+    },
+    {
+      "epoch": 2.5587419056429233,
+      "grad_norm": 0.04128411412239075,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 24894
+    },
+    {
+      "epoch": 2.559050262102991,
+      "grad_norm": 0.03886473551392555,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 24897
+    },
+    {
+      "epoch": 2.559358618563059,
+      "grad_norm": 0.1163051575422287,
+      "learning_rate": 0.01,
+      "loss": 2.0456,
+      "step": 24900
+    },
+    {
+      "epoch": 2.5596669750231267,
+      "grad_norm": 0.04279797896742821,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 24903
+    },
+    {
+      "epoch": 2.5599753314831943,
+      "grad_norm": 0.08159471303224564,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 24906
+    },
+    {
+      "epoch": 2.5602836879432624,
+      "grad_norm": 0.06161525472998619,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 24909
+    },
+    {
+      "epoch": 2.56059204440333,
+      "grad_norm": 0.05011424049735069,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 24912
+    },
+    {
+      "epoch": 2.560900400863398,
+      "grad_norm": 0.05973159521818161,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 24915
+    },
+    {
+      "epoch": 2.561208757323466,
+      "grad_norm": 0.07461394369602203,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 24918
+    },
+    {
+      "epoch": 2.561517113783534,
+      "grad_norm": 0.09631699323654175,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 24921
+    },
+    {
+      "epoch": 2.5618254702436016,
+      "grad_norm": 0.05727219581604004,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 24924
+    },
+    {
+      "epoch": 2.5621338267036693,
+      "grad_norm": 0.08594338595867157,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 24927
+    },
+    {
+      "epoch": 2.5624421831637374,
+      "grad_norm": 0.1109083890914917,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 24930
+    },
+    {
+      "epoch": 2.562750539623805,
+      "grad_norm": 0.0593339204788208,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 24933
+    },
+    {
+      "epoch": 2.5630588960838727,
+      "grad_norm": 0.0381302647292614,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 24936
+    },
+    {
+      "epoch": 2.563367252543941,
+      "grad_norm": 0.056093595921993256,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 24939
+    },
+    {
+      "epoch": 2.5636756090040085,
+      "grad_norm": 0.11212731897830963,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 24942
+    },
+    {
+      "epoch": 2.5639839654640766,
+      "grad_norm": 0.07110024988651276,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 24945
+    },
+    {
+      "epoch": 2.5642923219241442,
+      "grad_norm": 0.05951390787959099,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 24948
+    },
+    {
+      "epoch": 2.5646006783842124,
+      "grad_norm": 0.07836031913757324,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 24951
+    },
+    {
+      "epoch": 2.56490903484428,
+      "grad_norm": 0.06882999837398529,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 24954
+    },
+    {
+      "epoch": 2.5652173913043477,
+      "grad_norm": 0.0868605375289917,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 24957
+    },
+    {
+      "epoch": 2.5655257477644158,
+      "grad_norm": 0.10812171548604965,
+      "learning_rate": 0.01,
+      "loss": 2.0462,
+      "step": 24960
+    },
+    {
+      "epoch": 2.5658341042244834,
+      "grad_norm": 0.07124783843755722,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 24963
+    },
+    {
+      "epoch": 2.566142460684551,
+      "grad_norm": 0.037611838430166245,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 24966
+    },
+    {
+      "epoch": 2.566450817144619,
+      "grad_norm": 0.03723758086562157,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 24969
+    },
+    {
+      "epoch": 2.566759173604687,
+      "grad_norm": 0.08805309981107712,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 24972
+    },
+    {
+      "epoch": 2.567067530064755,
+      "grad_norm": 0.10381683707237244,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 24975
+    },
+    {
+      "epoch": 2.5673758865248226,
+      "grad_norm": 0.11186369508504868,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 24978
+    },
+    {
+      "epoch": 2.5676842429848907,
+      "grad_norm": 0.07986850291490555,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 24981
+    },
+    {
+      "epoch": 2.5679925994449584,
+      "grad_norm": 0.04094192385673523,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 24984
+    },
+    {
+      "epoch": 2.568300955905026,
+      "grad_norm": 0.049611032009124756,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 24987
+    },
+    {
+      "epoch": 2.568609312365094,
+      "grad_norm": 0.05613689869642258,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 24990
+    },
+    {
+      "epoch": 2.568917668825162,
+      "grad_norm": 0.051894500851631165,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 24993
+    },
+    {
+      "epoch": 2.5692260252852295,
+      "grad_norm": 0.04092536121606827,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 24996
+    },
+    {
+      "epoch": 2.5695343817452976,
+      "grad_norm": 0.03474249318242073,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 24999
+    },
+    {
+      "epoch": 2.5698427382053652,
+      "grad_norm": 0.04513520747423172,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 25002
+    },
+    {
+      "epoch": 2.5701510946654333,
+      "grad_norm": 0.06130135431885719,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 25005
+    },
+    {
+      "epoch": 2.570459451125501,
+      "grad_norm": 0.07398026436567307,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 25008
+    },
+    {
+      "epoch": 2.570767807585569,
+      "grad_norm": 0.06060103699564934,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 25011
+    },
+    {
+      "epoch": 2.5710761640456368,
+      "grad_norm": 0.051868923008441925,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 25014
+    },
+    {
+      "epoch": 2.5713845205057044,
+      "grad_norm": 0.09465671330690384,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 25017
+    },
+    {
+      "epoch": 2.5716928769657725,
+      "grad_norm": 0.048888836055994034,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 25020
+    },
+    {
+      "epoch": 2.57200123342584,
+      "grad_norm": 0.04938677325844765,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 25023
+    },
+    {
+      "epoch": 2.572309589885908,
+      "grad_norm": 0.1066848635673523,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 25026
+    },
+    {
+      "epoch": 2.572617946345976,
+      "grad_norm": 0.044199470430612564,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 25029
+    },
+    {
+      "epoch": 2.572926302806044,
+      "grad_norm": 0.06313291937112808,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 25032
+    },
+    {
+      "epoch": 2.5732346592661117,
+      "grad_norm": 0.08843620121479034,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 25035
+    },
+    {
+      "epoch": 2.5735430157261794,
+      "grad_norm": 0.028659775853157043,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 25038
+    },
+    {
+      "epoch": 2.5738513721862475,
+      "grad_norm": 0.09034299850463867,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 25041
+    },
+    {
+      "epoch": 2.574159728646315,
+      "grad_norm": 0.08496701717376709,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 25044
+    },
+    {
+      "epoch": 2.574468085106383,
+      "grad_norm": 0.04633186012506485,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 25047
+    },
+    {
+      "epoch": 2.574776441566451,
+      "grad_norm": 0.05091328173875809,
+      "learning_rate": 0.01,
+      "loss": 2.0622,
+      "step": 25050
+    },
+    {
+      "epoch": 2.5750847980265186,
+      "grad_norm": 0.03941154107451439,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 25053
+    },
+    {
+      "epoch": 2.5753931544865862,
+      "grad_norm": 0.07574623823165894,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 25056
+    },
+    {
+      "epoch": 2.5757015109466543,
+      "grad_norm": 0.07106275856494904,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 25059
+    },
+    {
+      "epoch": 2.5760098674067224,
+      "grad_norm": 0.06767601519823074,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 25062
+    },
+    {
+      "epoch": 2.57631822386679,
+      "grad_norm": 0.05537039414048195,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 25065
+    },
+    {
+      "epoch": 2.5766265803268578,
+      "grad_norm": 0.06547438353300095,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 25068
+    },
+    {
+      "epoch": 2.576934936786926,
+      "grad_norm": 0.0862760990858078,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 25071
+    },
+    {
+      "epoch": 2.5772432932469935,
+      "grad_norm": 0.041683733463287354,
+      "learning_rate": 0.01,
+      "loss": 2.0484,
+      "step": 25074
+    },
+    {
+      "epoch": 2.577551649707061,
+      "grad_norm": 0.049321915954351425,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 25077
+    },
+    {
+      "epoch": 2.5778600061671293,
+      "grad_norm": 0.09261754900217056,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 25080
+    },
+    {
+      "epoch": 2.578168362627197,
+      "grad_norm": 0.07979609072208405,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 25083
+    },
+    {
+      "epoch": 2.5784767190872646,
+      "grad_norm": 0.06629879772663116,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 25086
+    },
+    {
+      "epoch": 2.5787850755473327,
+      "grad_norm": 0.07896976172924042,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 25089
+    },
+    {
+      "epoch": 2.579093432007401,
+      "grad_norm": 0.06102503091096878,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 25092
+    },
+    {
+      "epoch": 2.5794017884674685,
+      "grad_norm": 0.07823985069990158,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 25095
+    },
+    {
+      "epoch": 2.579710144927536,
+      "grad_norm": 0.08163253217935562,
+      "learning_rate": 0.01,
+      "loss": 2.0583,
+      "step": 25098
+    },
+    {
+      "epoch": 2.5800185013876042,
+      "grad_norm": 0.06111651286482811,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 25101
+    },
+    {
+      "epoch": 2.580326857847672,
+      "grad_norm": 0.03768099471926689,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 25104
+    },
+    {
+      "epoch": 2.5806352143077396,
+      "grad_norm": 0.0871853157877922,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 25107
+    },
+    {
+      "epoch": 2.5809435707678077,
+      "grad_norm": 0.05394020304083824,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 25110
+    },
+    {
+      "epoch": 2.5812519272278753,
+      "grad_norm": 0.07910983264446259,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 25113
+    },
+    {
+      "epoch": 2.581560283687943,
+      "grad_norm": 0.06922271102666855,
+      "learning_rate": 0.01,
+      "loss": 2.0493,
+      "step": 25116
+    },
+    {
+      "epoch": 2.581868640148011,
+      "grad_norm": 0.05517781525850296,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 25119
+    },
+    {
+      "epoch": 2.582176996608079,
+      "grad_norm": 0.05166595056653023,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 25122
+    },
+    {
+      "epoch": 2.582485353068147,
+      "grad_norm": 0.045153357088565826,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 25125
+    },
+    {
+      "epoch": 2.5827937095282145,
+      "grad_norm": 0.07232387363910675,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 25128
+    },
+    {
+      "epoch": 2.5831020659882826,
+      "grad_norm": 0.035037536174058914,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 25131
+    },
+    {
+      "epoch": 2.5834104224483503,
+      "grad_norm": 0.039313822984695435,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 25134
+    },
+    {
+      "epoch": 2.583718778908418,
+      "grad_norm": 0.0632469579577446,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 25137
+    },
+    {
+      "epoch": 2.584027135368486,
+      "grad_norm": 0.10993051528930664,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 25140
+    },
+    {
+      "epoch": 2.5843354918285537,
+      "grad_norm": 0.0852990597486496,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 25143
+    },
+    {
+      "epoch": 2.5846438482886214,
+      "grad_norm": 0.0442403107881546,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 25146
+    },
+    {
+      "epoch": 2.5849522047486895,
+      "grad_norm": 0.03534874692559242,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 25149
+    },
+    {
+      "epoch": 2.5852605612087576,
+      "grad_norm": 0.031708016991615295,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 25152
+    },
+    {
+      "epoch": 2.5855689176688252,
+      "grad_norm": 0.056695304811000824,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 25155
+    },
+    {
+      "epoch": 2.585877274128893,
+      "grad_norm": 0.12697716057300568,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 25158
+    },
+    {
+      "epoch": 2.586185630588961,
+      "grad_norm": 0.07686912268400192,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 25161
+    },
+    {
+      "epoch": 2.5864939870490287,
+      "grad_norm": 0.10015466809272766,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 25164
+    },
+    {
+      "epoch": 2.5868023435090963,
+      "grad_norm": 0.05786514654755592,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 25167
+    },
+    {
+      "epoch": 2.5871106999691644,
+      "grad_norm": 0.05359407886862755,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 25170
+    },
+    {
+      "epoch": 2.587419056429232,
+      "grad_norm": 0.10763208568096161,
+      "learning_rate": 0.01,
+      "loss": 2.0611,
+      "step": 25173
+    },
+    {
+      "epoch": 2.5877274128893,
+      "grad_norm": 0.06255360692739487,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 25176
+    },
+    {
+      "epoch": 2.588035769349368,
+      "grad_norm": 0.0519418902695179,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 25179
+    },
+    {
+      "epoch": 2.588344125809436,
+      "grad_norm": 0.09810636937618256,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 25182
+    },
+    {
+      "epoch": 2.5886524822695036,
+      "grad_norm": 0.05091201886534691,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 25185
+    },
+    {
+      "epoch": 2.5889608387295713,
+      "grad_norm": 0.046215660870075226,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 25188
+    },
+    {
+      "epoch": 2.5892691951896394,
+      "grad_norm": 0.06873856484889984,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 25191
+    },
+    {
+      "epoch": 2.589577551649707,
+      "grad_norm": 0.08075796812772751,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 25194
+    },
+    {
+      "epoch": 2.5898859081097747,
+      "grad_norm": 0.10317845642566681,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 25197
+    },
+    {
+      "epoch": 2.590194264569843,
+      "grad_norm": 0.07780349254608154,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 25200
+    },
+    {
+      "epoch": 2.5905026210299105,
+      "grad_norm": 0.0646161437034607,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 25203
+    },
+    {
+      "epoch": 2.5908109774899786,
+      "grad_norm": 0.10224328190088272,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 25206
+    },
+    {
+      "epoch": 2.5911193339500462,
+      "grad_norm": 0.0714394599199295,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 25209
+    },
+    {
+      "epoch": 2.5914276904101143,
+      "grad_norm": 0.06261731684207916,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 25212
+    },
+    {
+      "epoch": 2.591736046870182,
+      "grad_norm": 0.07271763682365417,
+      "learning_rate": 0.01,
+      "loss": 2.0489,
+      "step": 25215
+    },
+    {
+      "epoch": 2.5920444033302497,
+      "grad_norm": 0.07044383883476257,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 25218
+    },
+    {
+      "epoch": 2.5923527597903178,
+      "grad_norm": 0.10808674246072769,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 25221
+    },
+    {
+      "epoch": 2.5926611162503854,
+      "grad_norm": 0.049697790294885635,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 25224
+    },
+    {
+      "epoch": 2.592969472710453,
+      "grad_norm": 0.08951854705810547,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 25227
+    },
+    {
+      "epoch": 2.593277829170521,
+      "grad_norm": 0.06834417581558228,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 25230
+    },
+    {
+      "epoch": 2.593586185630589,
+      "grad_norm": 0.037199974060058594,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 25233
+    },
+    {
+      "epoch": 2.593894542090657,
+      "grad_norm": 0.056284189224243164,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 25236
+    },
+    {
+      "epoch": 2.5942028985507246,
+      "grad_norm": 0.07686278969049454,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 25239
+    },
+    {
+      "epoch": 2.5945112550107927,
+      "grad_norm": 0.043646443635225296,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 25242
+    },
+    {
+      "epoch": 2.5948196114708604,
+      "grad_norm": 0.0594371035695076,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 25245
+    },
+    {
+      "epoch": 2.595127967930928,
+      "grad_norm": 0.08617819100618362,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 25248
+    },
+    {
+      "epoch": 2.595436324390996,
+      "grad_norm": 0.13672196865081787,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 25251
+    },
+    {
+      "epoch": 2.595744680851064,
+      "grad_norm": 0.062405068427324295,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 25254
+    },
+    {
+      "epoch": 2.5960530373111315,
+      "grad_norm": 0.042263686656951904,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 25257
+    },
+    {
+      "epoch": 2.5963613937711996,
+      "grad_norm": 0.04821668192744255,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 25260
+    },
+    {
+      "epoch": 2.5966697502312672,
+      "grad_norm": 0.048817023634910583,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 25263
+    },
+    {
+      "epoch": 2.5969781066913353,
+      "grad_norm": 0.05794850364327431,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 25266
+    },
+    {
+      "epoch": 2.597286463151403,
+      "grad_norm": 0.05057196319103241,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 25269
+    },
+    {
+      "epoch": 2.597594819611471,
+      "grad_norm": 0.05443112552165985,
+      "learning_rate": 0.01,
+      "loss": 2.0381,
+      "step": 25272
+    },
+    {
+      "epoch": 2.5979031760715388,
+      "grad_norm": 0.0533830001950264,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 25275
+    },
+    {
+      "epoch": 2.5982115325316064,
+      "grad_norm": 0.040888138115406036,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 25278
+    },
+    {
+      "epoch": 2.5985198889916745,
+      "grad_norm": 0.07154986262321472,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 25281
+    },
+    {
+      "epoch": 2.598828245451742,
+      "grad_norm": 0.17314322292804718,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 25284
+    },
+    {
+      "epoch": 2.59913660191181,
+      "grad_norm": 0.051602523773908615,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 25287
+    },
+    {
+      "epoch": 2.599444958371878,
+      "grad_norm": 0.047731827944517136,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 25290
+    },
+    {
+      "epoch": 2.5997533148319456,
+      "grad_norm": 0.0334257036447525,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 25293
+    },
+    {
+      "epoch": 2.6000616712920137,
+      "grad_norm": 0.03708617389202118,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 25296
+    },
+    {
+      "epoch": 2.6003700277520814,
+      "grad_norm": 0.08540193736553192,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 25299
+    },
+    {
+      "epoch": 2.6006783842121495,
+      "grad_norm": 0.1036924496293068,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 25302
+    },
+    {
+      "epoch": 2.600986740672217,
+      "grad_norm": 0.056603506207466125,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 25305
+    },
+    {
+      "epoch": 2.601295097132285,
+      "grad_norm": 0.1030723974108696,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 25308
+    },
+    {
+      "epoch": 2.601603453592353,
+      "grad_norm": 0.060525115579366684,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 25311
+    },
+    {
+      "epoch": 2.6019118100524206,
+      "grad_norm": 0.061082128435373306,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 25314
+    },
+    {
+      "epoch": 2.602220166512488,
+      "grad_norm": 0.045477550476789474,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 25317
+    },
+    {
+      "epoch": 2.6025285229725563,
+      "grad_norm": 0.03306104615330696,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 25320
+    },
+    {
+      "epoch": 2.602836879432624,
+      "grad_norm": 0.052543554455041885,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 25323
+    },
+    {
+      "epoch": 2.603145235892692,
+      "grad_norm": 0.04408182203769684,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 25326
+    },
+    {
+      "epoch": 2.6034535923527597,
+      "grad_norm": 0.05216488614678383,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 25329
+    },
+    {
+      "epoch": 2.603761948812828,
+      "grad_norm": 0.12084914743900299,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 25332
+    },
+    {
+      "epoch": 2.6040703052728955,
+      "grad_norm": 0.09642963856458664,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 25335
+    },
+    {
+      "epoch": 2.604378661732963,
+      "grad_norm": 0.06409110128879547,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 25338
+    },
+    {
+      "epoch": 2.6046870181930313,
+      "grad_norm": 0.07277770340442657,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 25341
+    },
+    {
+      "epoch": 2.604995374653099,
+      "grad_norm": 0.049252595752477646,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 25344
+    },
+    {
+      "epoch": 2.6053037311131666,
+      "grad_norm": 0.0495469830930233,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 25347
+    },
+    {
+      "epoch": 2.6056120875732347,
+      "grad_norm": 0.06318475306034088,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 25350
+    },
+    {
+      "epoch": 2.6059204440333024,
+      "grad_norm": 0.07843147218227386,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 25353
+    },
+    {
+      "epoch": 2.6062288004933705,
+      "grad_norm": 0.055239688605070114,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 25356
+    },
+    {
+      "epoch": 2.606537156953438,
+      "grad_norm": 0.03876148536801338,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 25359
+    },
+    {
+      "epoch": 2.6068455134135062,
+      "grad_norm": 0.12309783697128296,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 25362
+    },
+    {
+      "epoch": 2.607153869873574,
+      "grad_norm": 0.09926038980484009,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 25365
+    },
+    {
+      "epoch": 2.6074622263336416,
+      "grad_norm": 0.07237336784601212,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 25368
+    },
+    {
+      "epoch": 2.6077705827937097,
+      "grad_norm": 0.09653117507696152,
+      "learning_rate": 0.01,
+      "loss": 2.0547,
+      "step": 25371
+    },
+    {
+      "epoch": 2.6080789392537773,
+      "grad_norm": 0.0454515665769577,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 25374
+    },
+    {
+      "epoch": 2.608387295713845,
+      "grad_norm": 0.04541772976517677,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 25377
+    },
+    {
+      "epoch": 2.608695652173913,
+      "grad_norm": 0.05721856653690338,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 25380
+    },
+    {
+      "epoch": 2.6090040086339807,
+      "grad_norm": 0.0526764839887619,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 25383
+    },
+    {
+      "epoch": 2.609312365094049,
+      "grad_norm": 0.08415975421667099,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 25386
+    },
+    {
+      "epoch": 2.6096207215541165,
+      "grad_norm": 0.07950541377067566,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 25389
+    },
+    {
+      "epoch": 2.6099290780141846,
+      "grad_norm": 0.07684530317783356,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 25392
+    },
+    {
+      "epoch": 2.6102374344742523,
+      "grad_norm": 0.0458965003490448,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 25395
+    },
+    {
+      "epoch": 2.61054579093432,
+      "grad_norm": 0.11776190251111984,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 25398
+    },
+    {
+      "epoch": 2.610854147394388,
+      "grad_norm": 0.03954809904098511,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 25401
+    },
+    {
+      "epoch": 2.6111625038544557,
+      "grad_norm": 0.08056820929050446,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 25404
+    },
+    {
+      "epoch": 2.6114708603145234,
+      "grad_norm": 0.03288201987743378,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 25407
+    },
+    {
+      "epoch": 2.6117792167745915,
+      "grad_norm": 0.06156465783715248,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 25410
+    },
+    {
+      "epoch": 2.612087573234659,
+      "grad_norm": 0.04141581431031227,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 25413
+    },
+    {
+      "epoch": 2.6123959296947272,
+      "grad_norm": 0.06731928139925003,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 25416
+    },
+    {
+      "epoch": 2.612704286154795,
+      "grad_norm": 0.07682816684246063,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 25419
+    },
+    {
+      "epoch": 2.613012642614863,
+      "grad_norm": 0.10766996443271637,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 25422
+    },
+    {
+      "epoch": 2.6133209990749307,
+      "grad_norm": 0.11409672349691391,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 25425
+    },
+    {
+      "epoch": 2.6136293555349983,
+      "grad_norm": 0.07693130522966385,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 25428
+    },
+    {
+      "epoch": 2.6139377119950664,
+      "grad_norm": 0.034606434404850006,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 25431
+    },
+    {
+      "epoch": 2.614246068455134,
+      "grad_norm": 0.0957694724202156,
+      "learning_rate": 0.01,
+      "loss": 2.0527,
+      "step": 25434
+    },
+    {
+      "epoch": 2.6145544249152017,
+      "grad_norm": 0.05739649757742882,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 25437
+    },
+    {
+      "epoch": 2.61486278137527,
+      "grad_norm": 0.05702357366681099,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 25440
+    },
+    {
+      "epoch": 2.6151711378353375,
+      "grad_norm": 0.10596863180398941,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 25443
+    },
+    {
+      "epoch": 2.6154794942954056,
+      "grad_norm": 0.07135487347841263,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 25446
+    },
+    {
+      "epoch": 2.6157878507554733,
+      "grad_norm": 0.04034152254462242,
+      "learning_rate": 0.01,
+      "loss": 2.034,
+      "step": 25449
+    },
+    {
+      "epoch": 2.6160962072155414,
+      "grad_norm": 0.05510259047150612,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 25452
+    },
+    {
+      "epoch": 2.616404563675609,
+      "grad_norm": 0.03920494019985199,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 25455
+    },
+    {
+      "epoch": 2.6167129201356767,
+      "grad_norm": 0.06935703754425049,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 25458
+    },
+    {
+      "epoch": 2.617021276595745,
+      "grad_norm": 0.04524112120270729,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 25461
+    },
+    {
+      "epoch": 2.6173296330558125,
+      "grad_norm": 0.03639009967446327,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 25464
+    },
+    {
+      "epoch": 2.61763798951588,
+      "grad_norm": 0.0551072359085083,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 25467
+    },
+    {
+      "epoch": 2.617946345975948,
+      "grad_norm": 0.03943365439772606,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 25470
+    },
+    {
+      "epoch": 2.618254702436016,
+      "grad_norm": 0.05363466218113899,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 25473
+    },
+    {
+      "epoch": 2.618563058896084,
+      "grad_norm": 0.10065023601055145,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 25476
+    },
+    {
+      "epoch": 2.6188714153561516,
+      "grad_norm": 0.06447052210569382,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 25479
+    },
+    {
+      "epoch": 2.6191797718162197,
+      "grad_norm": 0.03966406360268593,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 25482
+    },
+    {
+      "epoch": 2.6194881282762874,
+      "grad_norm": 0.05280005559325218,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 25485
+    },
+    {
+      "epoch": 2.619796484736355,
+      "grad_norm": 0.1111968457698822,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 25488
+    },
+    {
+      "epoch": 2.620104841196423,
+      "grad_norm": 0.11483361572027206,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 25491
+    },
+    {
+      "epoch": 2.620413197656491,
+      "grad_norm": 0.039019446820020676,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 25494
+    },
+    {
+      "epoch": 2.6207215541165585,
+      "grad_norm": 0.05683600530028343,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 25497
+    },
+    {
+      "epoch": 2.6210299105766266,
+      "grad_norm": 0.042798321694135666,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 25500
+    },
+    {
+      "epoch": 2.6213382670366943,
+      "grad_norm": 0.040838126093149185,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 25503
+    },
+    {
+      "epoch": 2.6216466234967624,
+      "grad_norm": 0.07249750196933746,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 25506
+    },
+    {
+      "epoch": 2.62195497995683,
+      "grad_norm": 0.08197704702615738,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 25509
+    },
+    {
+      "epoch": 2.622263336416898,
+      "grad_norm": 0.11885122954845428,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 25512
+    },
+    {
+      "epoch": 2.622571692876966,
+      "grad_norm": 0.07574759423732758,
+      "learning_rate": 0.01,
+      "loss": 2.0443,
+      "step": 25515
+    },
+    {
+      "epoch": 2.6228800493370334,
+      "grad_norm": 0.05106687173247337,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 25518
+    },
+    {
+      "epoch": 2.6231884057971016,
+      "grad_norm": 0.03381403908133507,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 25521
+    },
+    {
+      "epoch": 2.623496762257169,
+      "grad_norm": 0.04259166494011879,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 25524
+    },
+    {
+      "epoch": 2.623805118717237,
+      "grad_norm": 0.06276170909404755,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 25527
+    },
+    {
+      "epoch": 2.624113475177305,
+      "grad_norm": 0.05478539317846298,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 25530
+    },
+    {
+      "epoch": 2.6244218316373726,
+      "grad_norm": 0.05915123224258423,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 25533
+    },
+    {
+      "epoch": 2.6247301880974407,
+      "grad_norm": 0.09082819521427155,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 25536
+    },
+    {
+      "epoch": 2.6250385445575084,
+      "grad_norm": 0.04604744166135788,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 25539
+    },
+    {
+      "epoch": 2.6253469010175765,
+      "grad_norm": 0.04548676684498787,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 25542
+    },
+    {
+      "epoch": 2.625655257477644,
+      "grad_norm": 0.03629077225923538,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 25545
+    },
+    {
+      "epoch": 2.625963613937712,
+      "grad_norm": 0.060490988194942474,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 25548
+    },
+    {
+      "epoch": 2.62627197039778,
+      "grad_norm": 0.12501440942287445,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 25551
+    },
+    {
+      "epoch": 2.6265803268578476,
+      "grad_norm": 0.052221618592739105,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 25554
+    },
+    {
+      "epoch": 2.6268886833179153,
+      "grad_norm": 0.08127149194478989,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 25557
+    },
+    {
+      "epoch": 2.6271970397779834,
+      "grad_norm": 0.13186825811862946,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 25560
+    },
+    {
+      "epoch": 2.6275053962380515,
+      "grad_norm": 0.10297831892967224,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 25563
+    },
+    {
+      "epoch": 2.627813752698119,
+      "grad_norm": 0.04737088829278946,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 25566
+    },
+    {
+      "epoch": 2.628122109158187,
+      "grad_norm": 0.055748652666807175,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 25569
+    },
+    {
+      "epoch": 2.628430465618255,
+      "grad_norm": 0.08968318998813629,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 25572
+    },
+    {
+      "epoch": 2.6287388220783225,
+      "grad_norm": 0.04962621256709099,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 25575
+    },
+    {
+      "epoch": 2.62904717853839,
+      "grad_norm": 0.033645693212747574,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 25578
+    },
+    {
+      "epoch": 2.6293555349984583,
+      "grad_norm": 0.05864016339182854,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 25581
+    },
+    {
+      "epoch": 2.629663891458526,
+      "grad_norm": 0.12207403779029846,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 25584
+    },
+    {
+      "epoch": 2.6299722479185936,
+      "grad_norm": 0.049948643893003464,
+      "learning_rate": 0.01,
+      "loss": 2.0534,
+      "step": 25587
+    },
+    {
+      "epoch": 2.6302806043786617,
+      "grad_norm": 0.08774693310260773,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 25590
+    },
+    {
+      "epoch": 2.63058896083873,
+      "grad_norm": 0.053613871335983276,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 25593
+    },
+    {
+      "epoch": 2.6308973172987975,
+      "grad_norm": 0.06298764050006866,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 25596
+    },
+    {
+      "epoch": 2.631205673758865,
+      "grad_norm": 0.03511015325784683,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 25599
+    },
+    {
+      "epoch": 2.6315140302189333,
+      "grad_norm": 0.09345996379852295,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 25602
+    },
+    {
+      "epoch": 2.631822386679001,
+      "grad_norm": 0.06202877685427666,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 25605
+    },
+    {
+      "epoch": 2.6321307431390686,
+      "grad_norm": 0.10231085866689682,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 25608
+    },
+    {
+      "epoch": 2.6324390995991367,
+      "grad_norm": 0.12403954565525055,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 25611
+    },
+    {
+      "epoch": 2.6327474560592043,
+      "grad_norm": 0.059275902807712555,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 25614
+    },
+    {
+      "epoch": 2.633055812519272,
+      "grad_norm": 0.04833563044667244,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 25617
+    },
+    {
+      "epoch": 2.63336416897934,
+      "grad_norm": 0.04218841344118118,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 25620
+    },
+    {
+      "epoch": 2.633672525439408,
+      "grad_norm": 0.1189088523387909,
+      "learning_rate": 0.01,
+      "loss": 2.0407,
+      "step": 25623
+    },
+    {
+      "epoch": 2.633980881899476,
+      "grad_norm": 0.11460559070110321,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 25626
+    },
+    {
+      "epoch": 2.6342892383595435,
+      "grad_norm": 0.13970693945884705,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 25629
+    },
+    {
+      "epoch": 2.6345975948196116,
+      "grad_norm": 0.06081441789865494,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 25632
+    },
+    {
+      "epoch": 2.6349059512796793,
+      "grad_norm": 0.055472832173109055,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 25635
+    },
+    {
+      "epoch": 2.635214307739747,
+      "grad_norm": 0.06767648458480835,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 25638
+    },
+    {
+      "epoch": 2.635522664199815,
+      "grad_norm": 0.060980260372161865,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 25641
+    },
+    {
+      "epoch": 2.6358310206598827,
+      "grad_norm": 0.04839742183685303,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 25644
+    },
+    {
+      "epoch": 2.6361393771199504,
+      "grad_norm": 0.04725825786590576,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 25647
+    },
+    {
+      "epoch": 2.6364477335800185,
+      "grad_norm": 0.04045959562063217,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 25650
+    },
+    {
+      "epoch": 2.6367560900400866,
+      "grad_norm": 0.07219504565000534,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 25653
+    },
+    {
+      "epoch": 2.6370644465001543,
+      "grad_norm": 0.0438094437122345,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 25656
+    },
+    {
+      "epoch": 2.637372802960222,
+      "grad_norm": 0.06555571407079697,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 25659
+    },
+    {
+      "epoch": 2.63768115942029,
+      "grad_norm": 0.07731533795595169,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 25662
+    },
+    {
+      "epoch": 2.6379895158803577,
+      "grad_norm": 0.045945990830659866,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 25665
+    },
+    {
+      "epoch": 2.6382978723404253,
+      "grad_norm": 0.07175582647323608,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 25668
+    },
+    {
+      "epoch": 2.6386062288004934,
+      "grad_norm": 0.07860400527715683,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 25671
+    },
+    {
+      "epoch": 2.638914585260561,
+      "grad_norm": 0.05635571479797363,
+      "learning_rate": 0.01,
+      "loss": 2.0519,
+      "step": 25674
+    },
+    {
+      "epoch": 2.639222941720629,
+      "grad_norm": 0.07352261245250702,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 25677
+    },
+    {
+      "epoch": 2.639531298180697,
+      "grad_norm": 0.07502644509077072,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 25680
+    },
+    {
+      "epoch": 2.639839654640765,
+      "grad_norm": 0.05903060361742973,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 25683
+    },
+    {
+      "epoch": 2.6401480111008326,
+      "grad_norm": 0.07454460859298706,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 25686
+    },
+    {
+      "epoch": 2.6404563675609003,
+      "grad_norm": 0.06615950912237167,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 25689
+    },
+    {
+      "epoch": 2.6407647240209684,
+      "grad_norm": 0.07897205650806427,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 25692
+    },
+    {
+      "epoch": 2.641073080481036,
+      "grad_norm": 0.12898573279380798,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 25695
+    },
+    {
+      "epoch": 2.6413814369411037,
+      "grad_norm": 0.10731782764196396,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 25698
+    },
+    {
+      "epoch": 2.641689793401172,
+      "grad_norm": 0.07936359196901321,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 25701
+    },
+    {
+      "epoch": 2.6419981498612395,
+      "grad_norm": 0.07443369179964066,
+      "learning_rate": 0.01,
+      "loss": 2.0586,
+      "step": 25704
+    },
+    {
+      "epoch": 2.6423065063213076,
+      "grad_norm": 0.07027627527713776,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 25707
+    },
+    {
+      "epoch": 2.6426148627813753,
+      "grad_norm": 0.04909298196434975,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 25710
+    },
+    {
+      "epoch": 2.6429232192414434,
+      "grad_norm": 0.09595979005098343,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 25713
+    },
+    {
+      "epoch": 2.643231575701511,
+      "grad_norm": 0.07217621803283691,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 25716
+    },
+    {
+      "epoch": 2.6435399321615787,
+      "grad_norm": 0.10547403246164322,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 25719
+    },
+    {
+      "epoch": 2.643848288621647,
+      "grad_norm": 0.048883359879255295,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 25722
+    },
+    {
+      "epoch": 2.6441566450817144,
+      "grad_norm": 0.09399348497390747,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 25725
+    },
+    {
+      "epoch": 2.644465001541782,
+      "grad_norm": 0.05369775742292404,
+      "learning_rate": 0.01,
+      "loss": 2.0483,
+      "step": 25728
+    },
+    {
+      "epoch": 2.64477335800185,
+      "grad_norm": 0.08048120886087418,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 25731
+    },
+    {
+      "epoch": 2.645081714461918,
+      "grad_norm": 0.07802169024944305,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 25734
+    },
+    {
+      "epoch": 2.645390070921986,
+      "grad_norm": 0.06137097254395485,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 25737
+    },
+    {
+      "epoch": 2.6456984273820536,
+      "grad_norm": 0.05498100444674492,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 25740
+    },
+    {
+      "epoch": 2.6460067838421217,
+      "grad_norm": 0.057225704193115234,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 25743
+    },
+    {
+      "epoch": 2.6463151403021894,
+      "grad_norm": 0.06727226078510284,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 25746
+    },
+    {
+      "epoch": 2.646623496762257,
+      "grad_norm": 0.04740385711193085,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 25749
+    },
+    {
+      "epoch": 2.646931853222325,
+      "grad_norm": 0.04808951914310455,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 25752
+    },
+    {
+      "epoch": 2.647240209682393,
+      "grad_norm": 0.04689471423625946,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 25755
+    },
+    {
+      "epoch": 2.6475485661424605,
+      "grad_norm": 0.10928355902433395,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 25758
+    },
+    {
+      "epoch": 2.6478569226025286,
+      "grad_norm": 0.054571595042943954,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 25761
+    },
+    {
+      "epoch": 2.6481652790625962,
+      "grad_norm": 0.0809570774435997,
+      "learning_rate": 0.01,
+      "loss": 2.0451,
+      "step": 25764
+    },
+    {
+      "epoch": 2.6484736355226643,
+      "grad_norm": 0.10407903790473938,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 25767
+    },
+    {
+      "epoch": 2.648781991982732,
+      "grad_norm": 0.045180514454841614,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 25770
+    },
+    {
+      "epoch": 2.6490903484428,
+      "grad_norm": 0.07642797380685806,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 25773
+    },
+    {
+      "epoch": 2.6493987049028678,
+      "grad_norm": 0.06943316757678986,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 25776
+    },
+    {
+      "epoch": 2.6497070613629354,
+      "grad_norm": 0.06536037474870682,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 25779
+    },
+    {
+      "epoch": 2.6500154178230035,
+      "grad_norm": 0.04698020964860916,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 25782
+    },
+    {
+      "epoch": 2.650323774283071,
+      "grad_norm": 0.05523454770445824,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 25785
+    },
+    {
+      "epoch": 2.650632130743139,
+      "grad_norm": 0.043647054582834244,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 25788
+    },
+    {
+      "epoch": 2.650940487203207,
+      "grad_norm": 0.05682402104139328,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 25791
+    },
+    {
+      "epoch": 2.6512488436632746,
+      "grad_norm": 0.10919704288244247,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 25794
+    },
+    {
+      "epoch": 2.6515572001233427,
+      "grad_norm": 0.0895003080368042,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 25797
+    },
+    {
+      "epoch": 2.6518655565834104,
+      "grad_norm": 0.07750852406024933,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 25800
+    },
+    {
+      "epoch": 2.6521739130434785,
+      "grad_norm": 0.07465649396181107,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 25803
+    },
+    {
+      "epoch": 2.652482269503546,
+      "grad_norm": 0.042654186487197876,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 25806
+    },
+    {
+      "epoch": 2.652790625963614,
+      "grad_norm": 0.04910847172141075,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 25809
+    },
+    {
+      "epoch": 2.653098982423682,
+      "grad_norm": 0.04584549367427826,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 25812
+    },
+    {
+      "epoch": 2.6534073388837496,
+      "grad_norm": 0.03400958329439163,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 25815
+    },
+    {
+      "epoch": 2.6537156953438172,
+      "grad_norm": 0.11867182701826096,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 25818
+    },
+    {
+      "epoch": 2.6540240518038853,
+      "grad_norm": 0.09303310513496399,
+      "learning_rate": 0.01,
+      "loss": 2.0533,
+      "step": 25821
+    },
+    {
+      "epoch": 2.654332408263953,
+      "grad_norm": 0.04951944947242737,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 25824
+    },
+    {
+      "epoch": 2.654640764724021,
+      "grad_norm": 0.05893867462873459,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 25827
+    },
+    {
+      "epoch": 2.6549491211840888,
+      "grad_norm": 0.034138891845941544,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 25830
+    },
+    {
+      "epoch": 2.655257477644157,
+      "grad_norm": 0.039834585040807724,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 25833
+    },
+    {
+      "epoch": 2.6555658341042245,
+      "grad_norm": 0.10098033398389816,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 25836
+    },
+    {
+      "epoch": 2.655874190564292,
+      "grad_norm": 0.07205129414796829,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 25839
+    },
+    {
+      "epoch": 2.6561825470243603,
+      "grad_norm": 0.04370702803134918,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 25842
+    },
+    {
+      "epoch": 2.656490903484428,
+      "grad_norm": 0.10662799328565598,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 25845
+    },
+    {
+      "epoch": 2.6567992599444956,
+      "grad_norm": 0.1833692342042923,
+      "learning_rate": 0.01,
+      "loss": 2.0585,
+      "step": 25848
+    },
+    {
+      "epoch": 2.6571076164045637,
+      "grad_norm": 0.12229418754577637,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 25851
+    },
+    {
+      "epoch": 2.6574159728646314,
+      "grad_norm": 0.05533694103360176,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 25854
+    },
+    {
+      "epoch": 2.6577243293246995,
+      "grad_norm": 0.042578935623168945,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 25857
+    },
+    {
+      "epoch": 2.658032685784767,
+      "grad_norm": 0.034579649567604065,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 25860
+    },
+    {
+      "epoch": 2.6583410422448353,
+      "grad_norm": 0.052627481520175934,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 25863
+    },
+    {
+      "epoch": 2.658649398704903,
+      "grad_norm": 0.04877294600009918,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 25866
+    },
+    {
+      "epoch": 2.6589577551649706,
+      "grad_norm": 0.05192401632666588,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 25869
+    },
+    {
+      "epoch": 2.6592661116250387,
+      "grad_norm": 0.08188097178936005,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 25872
+    },
+    {
+      "epoch": 2.6595744680851063,
+      "grad_norm": 0.06215586140751839,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 25875
+    },
+    {
+      "epoch": 2.659882824545174,
+      "grad_norm": 0.10561207681894302,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 25878
+    },
+    {
+      "epoch": 2.660191181005242,
+      "grad_norm": 0.08467070758342743,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 25881
+    },
+    {
+      "epoch": 2.6604995374653098,
+      "grad_norm": 0.05138308182358742,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 25884
+    },
+    {
+      "epoch": 2.660807893925378,
+      "grad_norm": 0.07387588173151016,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 25887
+    },
+    {
+      "epoch": 2.6611162503854455,
+      "grad_norm": 0.06682645529508591,
+      "learning_rate": 0.01,
+      "loss": 2.0508,
+      "step": 25890
+    },
+    {
+      "epoch": 2.6614246068455136,
+      "grad_norm": 0.056953735649585724,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 25893
+    },
+    {
+      "epoch": 2.6617329633055813,
+      "grad_norm": 0.03709590435028076,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 25896
+    },
+    {
+      "epoch": 2.662041319765649,
+      "grad_norm": 0.05895649641752243,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 25899
+    },
+    {
+      "epoch": 2.662349676225717,
+      "grad_norm": 0.08740679174661636,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 25902
+    },
+    {
+      "epoch": 2.6626580326857847,
+      "grad_norm": 0.04745563119649887,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 25905
+    },
+    {
+      "epoch": 2.6629663891458524,
+      "grad_norm": 0.08748458325862885,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 25908
+    },
+    {
+      "epoch": 2.6632747456059205,
+      "grad_norm": 0.04833563417196274,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 25911
+    },
+    {
+      "epoch": 2.663583102065988,
+      "grad_norm": 0.03873222693800926,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 25914
+    },
+    {
+      "epoch": 2.6638914585260562,
+      "grad_norm": 0.07129956781864166,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 25917
+    },
+    {
+      "epoch": 2.664199814986124,
+      "grad_norm": 0.08984299004077911,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 25920
+    },
+    {
+      "epoch": 2.664508171446192,
+      "grad_norm": 0.03720667213201523,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 25923
+    },
+    {
+      "epoch": 2.6648165279062597,
+      "grad_norm": 0.09795575588941574,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 25926
+    },
+    {
+      "epoch": 2.6651248843663273,
+      "grad_norm": 0.1043957769870758,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 25929
+    },
+    {
+      "epoch": 2.6654332408263954,
+      "grad_norm": 0.04737646505236626,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 25932
+    },
+    {
+      "epoch": 2.665741597286463,
+      "grad_norm": 0.04864402487874031,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 25935
+    },
+    {
+      "epoch": 2.6660499537465308,
+      "grad_norm": 0.0386229082942009,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 25938
+    },
+    {
+      "epoch": 2.666358310206599,
+      "grad_norm": 0.04127798601984978,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 25941
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.07537275552749634,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 25944
+    },
+    {
+      "epoch": 2.6669750231267346,
+      "grad_norm": 0.041533395648002625,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 25947
+    },
+    {
+      "epoch": 2.6672833795868023,
+      "grad_norm": 0.045062318444252014,
+      "learning_rate": 0.01,
+      "loss": 2.0428,
+      "step": 25950
+    },
+    {
+      "epoch": 2.6675917360468704,
+      "grad_norm": 0.08328119665384293,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 25953
+    },
+    {
+      "epoch": 2.667900092506938,
+      "grad_norm": 0.08402698487043381,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 25956
+    },
+    {
+      "epoch": 2.6682084489670057,
+      "grad_norm": 0.10571663826704025,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 25959
+    },
+    {
+      "epoch": 2.668516805427074,
+      "grad_norm": 0.05094289034605026,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 25962
+    },
+    {
+      "epoch": 2.6688251618871415,
+      "grad_norm": 0.034051552414894104,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 25965
+    },
+    {
+      "epoch": 2.669133518347209,
+      "grad_norm": 0.05703623592853546,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 25968
+    },
+    {
+      "epoch": 2.6694418748072772,
+      "grad_norm": 0.10606861114501953,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 25971
+    },
+    {
+      "epoch": 2.669750231267345,
+      "grad_norm": 0.11793088167905807,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 25974
+    },
+    {
+      "epoch": 2.670058587727413,
+      "grad_norm": 0.09058693051338196,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 25977
+    },
+    {
+      "epoch": 2.6703669441874807,
+      "grad_norm": 0.0677378848195076,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 25980
+    },
+    {
+      "epoch": 2.6706753006475488,
+      "grad_norm": 0.05636313930153847,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 25983
+    },
+    {
+      "epoch": 2.6709836571076164,
+      "grad_norm": 0.04547611251473427,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 25986
+    },
+    {
+      "epoch": 2.671292013567684,
+      "grad_norm": 0.04852728173136711,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 25989
+    },
+    {
+      "epoch": 2.671600370027752,
+      "grad_norm": 0.04465510696172714,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 25992
+    },
+    {
+      "epoch": 2.67190872648782,
+      "grad_norm": 0.043789103627204895,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 25995
+    },
+    {
+      "epoch": 2.6722170829478875,
+      "grad_norm": 0.12845320999622345,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 25998
+    },
+    {
+      "epoch": 2.6725254394079556,
+      "grad_norm": 0.054881785064935684,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 26001
+    },
+    {
+      "epoch": 2.6728337958680233,
+      "grad_norm": 0.0845002606511116,
+      "learning_rate": 0.01,
+      "loss": 2.0475,
+      "step": 26004
+    },
+    {
+      "epoch": 2.6731421523280914,
+      "grad_norm": 0.04782318323850632,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 26007
+    },
+    {
+      "epoch": 2.673450508788159,
+      "grad_norm": 0.04288490489125252,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 26010
+    },
+    {
+      "epoch": 2.673758865248227,
+      "grad_norm": 0.10904238373041153,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 26013
+    },
+    {
+      "epoch": 2.674067221708295,
+      "grad_norm": 0.08276703208684921,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 26016
+    },
+    {
+      "epoch": 2.6743755781683625,
+      "grad_norm": 0.06076609715819359,
+      "learning_rate": 0.01,
+      "loss": 2.0507,
+      "step": 26019
+    },
+    {
+      "epoch": 2.6746839346284306,
+      "grad_norm": 0.043946780264377594,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 26022
+    },
+    {
+      "epoch": 2.6749922910884982,
+      "grad_norm": 0.03716238588094711,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 26025
+    },
+    {
+      "epoch": 2.675300647548566,
+      "grad_norm": 0.04982864856719971,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 26028
+    },
+    {
+      "epoch": 2.675609004008634,
+      "grad_norm": 0.05211315676569939,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 26031
+    },
+    {
+      "epoch": 2.6759173604687017,
+      "grad_norm": 0.10215940326452255,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 26034
+    },
+    {
+      "epoch": 2.6762257169287698,
+      "grad_norm": 0.07742898166179657,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 26037
+    },
+    {
+      "epoch": 2.6765340733888374,
+      "grad_norm": 0.047653328627347946,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 26040
+    },
+    {
+      "epoch": 2.6768424298489055,
+      "grad_norm": 0.07179111242294312,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 26043
+    },
+    {
+      "epoch": 2.677150786308973,
+      "grad_norm": 0.044947609305381775,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 26046
+    },
+    {
+      "epoch": 2.677459142769041,
+      "grad_norm": 0.03128395974636078,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 26049
+    },
+    {
+      "epoch": 2.677767499229109,
+      "grad_norm": 0.037758342921733856,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 26052
+    },
+    {
+      "epoch": 2.6780758556891766,
+      "grad_norm": 0.059724219143390656,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 26055
+    },
+    {
+      "epoch": 2.6783842121492443,
+      "grad_norm": 0.09603296220302582,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 26058
+    },
+    {
+      "epoch": 2.6786925686093124,
+      "grad_norm": 0.059488292783498764,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 26061
+    },
+    {
+      "epoch": 2.6790009250693805,
+      "grad_norm": 0.07343467324972153,
+      "learning_rate": 0.01,
+      "loss": 2.0494,
+      "step": 26064
+    },
+    {
+      "epoch": 2.679309281529448,
+      "grad_norm": 0.054782915860414505,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 26067
+    },
+    {
+      "epoch": 2.679617637989516,
+      "grad_norm": 0.042636722326278687,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 26070
+    },
+    {
+      "epoch": 2.679925994449584,
+      "grad_norm": 0.06803017109632492,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 26073
+    },
+    {
+      "epoch": 2.6802343509096516,
+      "grad_norm": 0.10990459471940994,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 26076
+    },
+    {
+      "epoch": 2.6805427073697192,
+      "grad_norm": 0.035727545619010925,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 26079
+    },
+    {
+      "epoch": 2.6808510638297873,
+      "grad_norm": 0.11241263896226883,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 26082
+    },
+    {
+      "epoch": 2.681159420289855,
+      "grad_norm": 0.06156554073095322,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 26085
+    },
+    {
+      "epoch": 2.6814677767499226,
+      "grad_norm": 0.0609101839363575,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 26088
+    },
+    {
+      "epoch": 2.6817761332099908,
+      "grad_norm": 0.046859260648489,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 26091
+    },
+    {
+      "epoch": 2.682084489670059,
+      "grad_norm": 0.05803006887435913,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 26094
+    },
+    {
+      "epoch": 2.6823928461301265,
+      "grad_norm": 0.09425931423902512,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 26097
+    },
+    {
+      "epoch": 2.682701202590194,
+      "grad_norm": 0.07893898338079453,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 26100
+    },
+    {
+      "epoch": 2.6830095590502623,
+      "grad_norm": 0.044163595885038376,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 26103
+    },
+    {
+      "epoch": 2.68331791551033,
+      "grad_norm": 0.061135292053222656,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 26106
+    },
+    {
+      "epoch": 2.6836262719703976,
+      "grad_norm": 0.03806147351861,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 26109
+    },
+    {
+      "epoch": 2.6839346284304657,
+      "grad_norm": 0.050713907927274704,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 26112
+    },
+    {
+      "epoch": 2.6842429848905334,
+      "grad_norm": 0.05512484535574913,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 26115
+    },
+    {
+      "epoch": 2.684551341350601,
+      "grad_norm": 0.048581019043922424,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 26118
+    },
+    {
+      "epoch": 2.684859697810669,
+      "grad_norm": 0.04121122509241104,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 26121
+    },
+    {
+      "epoch": 2.6851680542707372,
+      "grad_norm": 0.036977533251047134,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 26124
+    },
+    {
+      "epoch": 2.685476410730805,
+      "grad_norm": 0.10436423867940903,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 26127
+    },
+    {
+      "epoch": 2.6857847671908726,
+      "grad_norm": 0.05357207730412483,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 26130
+    },
+    {
+      "epoch": 2.6860931236509407,
+      "grad_norm": 0.05386963114142418,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 26133
+    },
+    {
+      "epoch": 2.6864014801110083,
+      "grad_norm": 0.03591128811240196,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 26136
+    },
+    {
+      "epoch": 2.686709836571076,
+      "grad_norm": 0.036740854382514954,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 26139
+    },
+    {
+      "epoch": 2.687018193031144,
+      "grad_norm": 0.05966171249747276,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 26142
+    },
+    {
+      "epoch": 2.6873265494912117,
+      "grad_norm": 0.10529398918151855,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 26145
+    },
+    {
+      "epoch": 2.6876349059512794,
+      "grad_norm": 0.07834989577531815,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 26148
+    },
+    {
+      "epoch": 2.6879432624113475,
+      "grad_norm": 0.10702015459537506,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 26151
+    },
+    {
+      "epoch": 2.6882516188714156,
+      "grad_norm": 0.04647925868630409,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 26154
+    },
+    {
+      "epoch": 2.6885599753314833,
+      "grad_norm": 0.04523247852921486,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 26157
+    },
+    {
+      "epoch": 2.688868331791551,
+      "grad_norm": 0.04293215274810791,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 26160
+    },
+    {
+      "epoch": 2.689176688251619,
+      "grad_norm": 0.06344863772392273,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 26163
+    },
+    {
+      "epoch": 2.6894850447116867,
+      "grad_norm": 0.047177109867334366,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 26166
+    },
+    {
+      "epoch": 2.6897934011717544,
+      "grad_norm": 0.09941892325878143,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 26169
+    },
+    {
+      "epoch": 2.6901017576318225,
+      "grad_norm": 0.04238753393292427,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 26172
+    },
+    {
+      "epoch": 2.69041011409189,
+      "grad_norm": 0.0550382174551487,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 26175
+    },
+    {
+      "epoch": 2.690718470551958,
+      "grad_norm": 0.09788551181554794,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 26178
+    },
+    {
+      "epoch": 2.691026827012026,
+      "grad_norm": 0.044239919632673264,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 26181
+    },
+    {
+      "epoch": 2.691335183472094,
+      "grad_norm": 0.10677097737789154,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 26184
+    },
+    {
+      "epoch": 2.6916435399321617,
+      "grad_norm": 0.04113283380866051,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 26187
+    },
+    {
+      "epoch": 2.6919518963922293,
+      "grad_norm": 0.05748388543725014,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 26190
+    },
+    {
+      "epoch": 2.6922602528522974,
+      "grad_norm": 0.04931147024035454,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 26193
+    },
+    {
+      "epoch": 2.692568609312365,
+      "grad_norm": 0.051582470536231995,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 26196
+    },
+    {
+      "epoch": 2.6928769657724327,
+      "grad_norm": 0.1054171696305275,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 26199
+    },
+    {
+      "epoch": 2.693185322232501,
+      "grad_norm": 0.09005532413721085,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 26202
+    },
+    {
+      "epoch": 2.6934936786925685,
+      "grad_norm": 0.08124802261590958,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 26205
+    },
+    {
+      "epoch": 2.6938020351526366,
+      "grad_norm": 0.06238120049238205,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 26208
+    },
+    {
+      "epoch": 2.6941103916127043,
+      "grad_norm": 0.0375673770904541,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 26211
+    },
+    {
+      "epoch": 2.6944187480727724,
+      "grad_norm": 0.04710889235138893,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 26214
+    },
+    {
+      "epoch": 2.69472710453284,
+      "grad_norm": 0.0509529784321785,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 26217
+    },
+    {
+      "epoch": 2.6950354609929077,
+      "grad_norm": 0.043452613055706024,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 26220
+    },
+    {
+      "epoch": 2.695343817452976,
+      "grad_norm": 0.07350075244903564,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 26223
+    },
+    {
+      "epoch": 2.6956521739130435,
+      "grad_norm": 0.07672197371721268,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 26226
+    },
+    {
+      "epoch": 2.695960530373111,
+      "grad_norm": 0.1255597323179245,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 26229
+    },
+    {
+      "epoch": 2.6962688868331792,
+      "grad_norm": 0.0636356994509697,
+      "learning_rate": 0.01,
+      "loss": 2.0438,
+      "step": 26232
+    },
+    {
+      "epoch": 2.696577243293247,
+      "grad_norm": 0.06334753334522247,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 26235
+    },
+    {
+      "epoch": 2.696885599753315,
+      "grad_norm": 0.0592232346534729,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 26238
+    },
+    {
+      "epoch": 2.6971939562133826,
+      "grad_norm": 0.06983037292957306,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 26241
+    },
+    {
+      "epoch": 2.6975023126734508,
+      "grad_norm": 0.042423997074365616,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 26244
+    },
+    {
+      "epoch": 2.6978106691335184,
+      "grad_norm": 0.12152963131666183,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 26247
+    },
+    {
+      "epoch": 2.698119025593586,
+      "grad_norm": 0.03922561556100845,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 26250
+    },
+    {
+      "epoch": 2.698427382053654,
+      "grad_norm": 0.04790705442428589,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 26253
+    },
+    {
+      "epoch": 2.698735738513722,
+      "grad_norm": 0.0596231147646904,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 26256
+    },
+    {
+      "epoch": 2.6990440949737895,
+      "grad_norm": 0.06809087842702866,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 26259
+    },
+    {
+      "epoch": 2.6993524514338576,
+      "grad_norm": 0.08235184103250504,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 26262
+    },
+    {
+      "epoch": 2.6996608078939253,
+      "grad_norm": 0.07293444871902466,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 26265
+    },
+    {
+      "epoch": 2.6999691643539934,
+      "grad_norm": 0.04277535527944565,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 26268
+    },
+    {
+      "epoch": 2.700277520814061,
+      "grad_norm": 0.10841301828622818,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 26271
+    },
+    {
+      "epoch": 2.700585877274129,
+      "grad_norm": 0.04575572907924652,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 26274
+    },
+    {
+      "epoch": 2.700894233734197,
+      "grad_norm": 0.0422062948346138,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 26277
+    },
+    {
+      "epoch": 2.7012025901942645,
+      "grad_norm": 0.034286245703697205,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 26280
+    },
+    {
+      "epoch": 2.7015109466543326,
+      "grad_norm": 0.040598925203084946,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 26283
+    },
+    {
+      "epoch": 2.7018193031144,
+      "grad_norm": 0.20375369489192963,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 26286
+    },
+    {
+      "epoch": 2.702127659574468,
+      "grad_norm": 0.1908276230096817,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 26289
+    },
+    {
+      "epoch": 2.702436016034536,
+      "grad_norm": 0.06826306879520416,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 26292
+    },
+    {
+      "epoch": 2.7027443724946036,
+      "grad_norm": 0.03917758911848068,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 26295
+    },
+    {
+      "epoch": 2.7030527289546717,
+      "grad_norm": 0.041581716388463974,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 26298
+    },
+    {
+      "epoch": 2.7033610854147394,
+      "grad_norm": 0.03759411349892616,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 26301
+    },
+    {
+      "epoch": 2.7036694418748075,
+      "grad_norm": 0.043274398893117905,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 26304
+    },
+    {
+      "epoch": 2.703977798334875,
+      "grad_norm": 0.03156547620892525,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 26307
+    },
+    {
+      "epoch": 2.704286154794943,
+      "grad_norm": 0.042134515941143036,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 26310
+    },
+    {
+      "epoch": 2.704594511255011,
+      "grad_norm": 0.15473207831382751,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 26313
+    },
+    {
+      "epoch": 2.7049028677150786,
+      "grad_norm": 0.16576655209064484,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 26316
+    },
+    {
+      "epoch": 2.7052112241751463,
+      "grad_norm": 0.2106962651014328,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 26319
+    },
+    {
+      "epoch": 2.7055195806352144,
+      "grad_norm": 0.06543057411909103,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 26322
+    },
+    {
+      "epoch": 2.705827937095282,
+      "grad_norm": 0.05747131630778313,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 26325
+    },
+    {
+      "epoch": 2.70613629355535,
+      "grad_norm": 0.03801654651761055,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 26328
+    },
+    {
+      "epoch": 2.706444650015418,
+      "grad_norm": 0.047034505754709244,
+      "learning_rate": 0.01,
+      "loss": 2.0524,
+      "step": 26331
+    },
+    {
+      "epoch": 2.706753006475486,
+      "grad_norm": 0.03980104252696037,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 26334
+    },
+    {
+      "epoch": 2.7070613629355536,
+      "grad_norm": 0.04470902681350708,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 26337
+    },
+    {
+      "epoch": 2.707369719395621,
+      "grad_norm": 0.03287111967802048,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 26340
+    },
+    {
+      "epoch": 2.7076780758556893,
+      "grad_norm": 0.04395715892314911,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 26343
+    },
+    {
+      "epoch": 2.707986432315757,
+      "grad_norm": 0.06358876079320908,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 26346
+    },
+    {
+      "epoch": 2.7082947887758246,
+      "grad_norm": 0.03925269469618797,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 26349
+    },
+    {
+      "epoch": 2.7086031452358927,
+      "grad_norm": 0.16810861229896545,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 26352
+    },
+    {
+      "epoch": 2.7089115016959604,
+      "grad_norm": 0.09612049907445908,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 26355
+    },
+    {
+      "epoch": 2.7092198581560285,
+      "grad_norm": 0.048840370029211044,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 26358
+    },
+    {
+      "epoch": 2.709528214616096,
+      "grad_norm": 0.03775126487016678,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 26361
+    },
+    {
+      "epoch": 2.7098365710761643,
+      "grad_norm": 0.04588973522186279,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 26364
+    },
+    {
+      "epoch": 2.710144927536232,
+      "grad_norm": 0.061829451471567154,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 26367
+    },
+    {
+      "epoch": 2.7104532839962996,
+      "grad_norm": 0.053572434931993484,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 26370
+    },
+    {
+      "epoch": 2.7107616404563677,
+      "grad_norm": 0.0460902638733387,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 26373
+    },
+    {
+      "epoch": 2.7110699969164354,
+      "grad_norm": 0.07622378319501877,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 26376
+    },
+    {
+      "epoch": 2.711378353376503,
+      "grad_norm": 0.04137422889471054,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 26379
+    },
+    {
+      "epoch": 2.711686709836571,
+      "grad_norm": 0.052519541233778,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 26382
+    },
+    {
+      "epoch": 2.711995066296639,
+      "grad_norm": 0.13548482954502106,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 26385
+    },
+    {
+      "epoch": 2.712303422756707,
+      "grad_norm": 0.05124654993414879,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 26388
+    },
+    {
+      "epoch": 2.7126117792167745,
+      "grad_norm": 0.04233216866850853,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 26391
+    },
+    {
+      "epoch": 2.7129201356768426,
+      "grad_norm": 0.04341414198279381,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 26394
+    },
+    {
+      "epoch": 2.7132284921369103,
+      "grad_norm": 0.078862264752388,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 26397
+    },
+    {
+      "epoch": 2.713536848596978,
+      "grad_norm": 0.08351139724254608,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 26400
+    },
+    {
+      "epoch": 2.713845205057046,
+      "grad_norm": 0.10899659991264343,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 26403
+    },
+    {
+      "epoch": 2.7141535615171137,
+      "grad_norm": 0.11437533795833588,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 26406
+    },
+    {
+      "epoch": 2.7144619179771814,
+      "grad_norm": 0.04197553172707558,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 26409
+    },
+    {
+      "epoch": 2.7147702744372495,
+      "grad_norm": 0.054637305438518524,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 26412
+    },
+    {
+      "epoch": 2.715078630897317,
+      "grad_norm": 0.058579228818416595,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 26415
+    },
+    {
+      "epoch": 2.7153869873573853,
+      "grad_norm": 0.03501029685139656,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 26418
+    },
+    {
+      "epoch": 2.715695343817453,
+      "grad_norm": 0.04865582287311554,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 26421
+    },
+    {
+      "epoch": 2.716003700277521,
+      "grad_norm": 0.08878735452890396,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 26424
+    },
+    {
+      "epoch": 2.7163120567375887,
+      "grad_norm": 0.06919421255588531,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 26427
+    },
+    {
+      "epoch": 2.7166204131976563,
+      "grad_norm": 0.05429752171039581,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 26430
+    },
+    {
+      "epoch": 2.7169287696577245,
+      "grad_norm": 0.10294333100318909,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 26433
+    },
+    {
+      "epoch": 2.717237126117792,
+      "grad_norm": 0.0625937432050705,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 26436
+    },
+    {
+      "epoch": 2.7175454825778598,
+      "grad_norm": 0.048780061304569244,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 26439
+    },
+    {
+      "epoch": 2.717853839037928,
+      "grad_norm": 0.03840133175253868,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 26442
+    },
+    {
+      "epoch": 2.7181621954979955,
+      "grad_norm": 0.04021105915307999,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 26445
+    },
+    {
+      "epoch": 2.7184705519580636,
+      "grad_norm": 0.04241623356938362,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 26448
+    },
+    {
+      "epoch": 2.7187789084181313,
+      "grad_norm": 0.04488721489906311,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 26451
+    },
+    {
+      "epoch": 2.7190872648781994,
+      "grad_norm": 0.09730737656354904,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 26454
+    },
+    {
+      "epoch": 2.719395621338267,
+      "grad_norm": 0.05743822455406189,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 26457
+    },
+    {
+      "epoch": 2.7197039777983347,
+      "grad_norm": 0.042568083852529526,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 26460
+    },
+    {
+      "epoch": 2.720012334258403,
+      "grad_norm": 0.050452932715415955,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 26463
+    },
+    {
+      "epoch": 2.7203206907184705,
+      "grad_norm": 0.04751294478774071,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 26466
+    },
+    {
+      "epoch": 2.720629047178538,
+      "grad_norm": 0.12711678445339203,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 26469
+    },
+    {
+      "epoch": 2.7209374036386063,
+      "grad_norm": 0.043656837195158005,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 26472
+    },
+    {
+      "epoch": 2.721245760098674,
+      "grad_norm": 0.061641763895750046,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 26475
+    },
+    {
+      "epoch": 2.721554116558742,
+      "grad_norm": 0.04843028262257576,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 26478
+    },
+    {
+      "epoch": 2.7218624730188097,
+      "grad_norm": 0.05009736493229866,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 26481
+    },
+    {
+      "epoch": 2.722170829478878,
+      "grad_norm": 0.0330355204641819,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 26484
+    },
+    {
+      "epoch": 2.7224791859389454,
+      "grad_norm": 0.05958884581923485,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 26487
+    },
+    {
+      "epoch": 2.722787542399013,
+      "grad_norm": 0.10207222402095795,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 26490
+    },
+    {
+      "epoch": 2.723095898859081,
+      "grad_norm": 0.06909897923469543,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 26493
+    },
+    {
+      "epoch": 2.723404255319149,
+      "grad_norm": 0.09485501050949097,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 26496
+    },
+    {
+      "epoch": 2.7237126117792165,
+      "grad_norm": 0.0501030869781971,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 26499
+    },
+    {
+      "epoch": 2.7240209682392846,
+      "grad_norm": 0.07009368389844894,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 26502
+    },
+    {
+      "epoch": 2.7243293246993523,
+      "grad_norm": 0.0766390711069107,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 26505
+    },
+    {
+      "epoch": 2.7246376811594204,
+      "grad_norm": 0.050900235772132874,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 26508
+    },
+    {
+      "epoch": 2.724946037619488,
+      "grad_norm": 0.043463658541440964,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 26511
+    },
+    {
+      "epoch": 2.725254394079556,
+      "grad_norm": 0.07331634312868118,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 26514
+    },
+    {
+      "epoch": 2.725562750539624,
+      "grad_norm": 0.06863999366760254,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 26517
+    },
+    {
+      "epoch": 2.7258711069996915,
+      "grad_norm": 0.12374615669250488,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 26520
+    },
+    {
+      "epoch": 2.7261794634597596,
+      "grad_norm": 0.05284014344215393,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 26523
+    },
+    {
+      "epoch": 2.7264878199198272,
+      "grad_norm": 0.07776612043380737,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 26526
+    },
+    {
+      "epoch": 2.726796176379895,
+      "grad_norm": 0.0325651615858078,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 26529
+    },
+    {
+      "epoch": 2.727104532839963,
+      "grad_norm": 0.03690047189593315,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 26532
+    },
+    {
+      "epoch": 2.7274128893000307,
+      "grad_norm": 0.12211088836193085,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 26535
+    },
+    {
+      "epoch": 2.727721245760099,
+      "grad_norm": 0.05561113730072975,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 26538
+    },
+    {
+      "epoch": 2.7280296022201664,
+      "grad_norm": 0.05541878193616867,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 26541
+    },
+    {
+      "epoch": 2.7283379586802345,
+      "grad_norm": 0.037987880408763885,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 26544
+    },
+    {
+      "epoch": 2.728646315140302,
+      "grad_norm": 0.0748986229300499,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 26547
+    },
+    {
+      "epoch": 2.72895467160037,
+      "grad_norm": 0.05377615988254547,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 26550
+    },
+    {
+      "epoch": 2.729263028060438,
+      "grad_norm": 0.11756724119186401,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 26553
+    },
+    {
+      "epoch": 2.7295713845205056,
+      "grad_norm": 0.046288661658763885,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 26556
+    },
+    {
+      "epoch": 2.7298797409805733,
+      "grad_norm": 0.041496362537145615,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 26559
+    },
+    {
+      "epoch": 2.7301880974406414,
+      "grad_norm": 0.036013491451740265,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 26562
+    },
+    {
+      "epoch": 2.7304964539007095,
+      "grad_norm": 0.044654857367277145,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 26565
+    },
+    {
+      "epoch": 2.730804810360777,
+      "grad_norm": 0.10827535390853882,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 26568
+    },
+    {
+      "epoch": 2.731113166820845,
+      "grad_norm": 0.053766943514347076,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 26571
+    },
+    {
+      "epoch": 2.731421523280913,
+      "grad_norm": 0.10854397714138031,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 26574
+    },
+    {
+      "epoch": 2.7317298797409806,
+      "grad_norm": 0.10521306842565536,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 26577
+    },
+    {
+      "epoch": 2.7320382362010482,
+      "grad_norm": 0.06532178819179535,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 26580
+    },
+    {
+      "epoch": 2.7323465926611163,
+      "grad_norm": 0.06266357749700546,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 26583
+    },
+    {
+      "epoch": 2.732654949121184,
+      "grad_norm": 0.04027451574802399,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 26586
+    },
+    {
+      "epoch": 2.7329633055812517,
+      "grad_norm": 0.06135137006640434,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 26589
+    },
+    {
+      "epoch": 2.7332716620413198,
+      "grad_norm": 0.08157611638307571,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 26592
+    },
+    {
+      "epoch": 2.733580018501388,
+      "grad_norm": 0.060507632791996,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 26595
+    },
+    {
+      "epoch": 2.7338883749614555,
+      "grad_norm": 0.07385814189910889,
+      "learning_rate": 0.01,
+      "loss": 2.0496,
+      "step": 26598
+    },
+    {
+      "epoch": 2.734196731421523,
+      "grad_norm": 0.12423606961965561,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 26601
+    },
+    {
+      "epoch": 2.7345050878815913,
+      "grad_norm": 0.09299054741859436,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 26604
+    },
+    {
+      "epoch": 2.734813444341659,
+      "grad_norm": 0.07102067768573761,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 26607
+    },
+    {
+      "epoch": 2.7351218008017266,
+      "grad_norm": 0.036449629813432693,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 26610
+    },
+    {
+      "epoch": 2.7354301572617947,
+      "grad_norm": 0.11397985368967056,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 26613
+    },
+    {
+      "epoch": 2.7357385137218624,
+      "grad_norm": 0.0781063437461853,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 26616
+    },
+    {
+      "epoch": 2.73604687018193,
+      "grad_norm": 0.05996118485927582,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 26619
+    },
+    {
+      "epoch": 2.736355226641998,
+      "grad_norm": 0.05300947278738022,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 26622
+    },
+    {
+      "epoch": 2.7366635831020663,
+      "grad_norm": 0.048238106071949005,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 26625
+    },
+    {
+      "epoch": 2.736971939562134,
+      "grad_norm": 0.09588511288166046,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 26628
+    },
+    {
+      "epoch": 2.7372802960222016,
+      "grad_norm": 0.04981528967618942,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 26631
+    },
+    {
+      "epoch": 2.7375886524822697,
+      "grad_norm": 0.06809774041175842,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 26634
+    },
+    {
+      "epoch": 2.7378970089423373,
+      "grad_norm": 0.08924854546785355,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 26637
+    },
+    {
+      "epoch": 2.738205365402405,
+      "grad_norm": 0.06717575341463089,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 26640
+    },
+    {
+      "epoch": 2.738513721862473,
+      "grad_norm": 0.08358625322580338,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 26643
+    },
+    {
+      "epoch": 2.7388220783225408,
+      "grad_norm": 0.14468573033809662,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 26646
+    },
+    {
+      "epoch": 2.7391304347826084,
+      "grad_norm": 0.0533704049885273,
+      "learning_rate": 0.01,
+      "loss": 2.0268,
+      "step": 26649
+    },
+    {
+      "epoch": 2.7394387912426765,
+      "grad_norm": 0.052842382341623306,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 26652
+    },
+    {
+      "epoch": 2.7397471477027446,
+      "grad_norm": 0.04230334609746933,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 26655
+    },
+    {
+      "epoch": 2.7400555041628123,
+      "grad_norm": 0.1004776731133461,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 26658
+    },
+    {
+      "epoch": 2.74036386062288,
+      "grad_norm": 0.03724560886621475,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 26661
+    },
+    {
+      "epoch": 2.740672217082948,
+      "grad_norm": 0.04997468739748001,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 26664
+    },
+    {
+      "epoch": 2.7409805735430157,
+      "grad_norm": 0.08138549327850342,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 26667
+    },
+    {
+      "epoch": 2.7412889300030834,
+      "grad_norm": 0.06195824220776558,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 26670
+    },
+    {
+      "epoch": 2.7415972864631515,
+      "grad_norm": 0.07340802997350693,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 26673
+    },
+    {
+      "epoch": 2.741905642923219,
+      "grad_norm": 0.09338308125734329,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 26676
+    },
+    {
+      "epoch": 2.742213999383287,
+      "grad_norm": 0.03742242231965065,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 26679
+    },
+    {
+      "epoch": 2.742522355843355,
+      "grad_norm": 0.04461616277694702,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 26682
+    },
+    {
+      "epoch": 2.742830712303423,
+      "grad_norm": 0.062195319682359695,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 26685
+    },
+    {
+      "epoch": 2.7431390687634907,
+      "grad_norm": 0.12009324878454208,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 26688
+    },
+    {
+      "epoch": 2.7434474252235583,
+      "grad_norm": 0.04727381840348244,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 26691
+    },
+    {
+      "epoch": 2.7437557816836264,
+      "grad_norm": 0.05080636590719223,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 26694
+    },
+    {
+      "epoch": 2.744064138143694,
+      "grad_norm": 0.11046820133924484,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 26697
+    },
+    {
+      "epoch": 2.7443724946037618,
+      "grad_norm": 0.06709396839141846,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 26700
+    },
+    {
+      "epoch": 2.74468085106383,
+      "grad_norm": 0.05088644102215767,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 26703
+    },
+    {
+      "epoch": 2.7449892075238975,
+      "grad_norm": 0.06446198374032974,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 26706
+    },
+    {
+      "epoch": 2.7452975639839656,
+      "grad_norm": 0.07209669798612595,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 26709
+    },
+    {
+      "epoch": 2.7456059204440333,
+      "grad_norm": 0.0958387553691864,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 26712
+    },
+    {
+      "epoch": 2.7459142769041014,
+      "grad_norm": 0.04666278511285782,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 26715
+    },
+    {
+      "epoch": 2.746222633364169,
+      "grad_norm": 0.0713481456041336,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 26718
+    },
+    {
+      "epoch": 2.7465309898242367,
+      "grad_norm": 0.060584791004657745,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 26721
+    },
+    {
+      "epoch": 2.746839346284305,
+      "grad_norm": 0.08711333572864532,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 26724
+    },
+    {
+      "epoch": 2.7471477027443725,
+      "grad_norm": 0.04297727718949318,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 26727
+    },
+    {
+      "epoch": 2.74745605920444,
+      "grad_norm": 0.06777355074882507,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 26730
+    },
+    {
+      "epoch": 2.7477644156645082,
+      "grad_norm": 0.05725346878170967,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 26733
+    },
+    {
+      "epoch": 2.748072772124576,
+      "grad_norm": 0.0601140521466732,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 26736
+    },
+    {
+      "epoch": 2.748381128584644,
+      "grad_norm": 0.04107888415455818,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 26739
+    },
+    {
+      "epoch": 2.7486894850447117,
+      "grad_norm": 0.03484128415584564,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 26742
+    },
+    {
+      "epoch": 2.7489978415047798,
+      "grad_norm": 0.05098375305533409,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 26745
+    },
+    {
+      "epoch": 2.7493061979648474,
+      "grad_norm": 0.1751684993505478,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 26748
+    },
+    {
+      "epoch": 2.749614554424915,
+      "grad_norm": 0.09155084937810898,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 26751
+    },
+    {
+      "epoch": 2.749922910884983,
+      "grad_norm": 0.05078737810254097,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 26754
+    },
+    {
+      "epoch": 2.750231267345051,
+      "grad_norm": 0.03406425565481186,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 26757
+    },
+    {
+      "epoch": 2.7505396238051185,
+      "grad_norm": 0.07126470655202866,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 26760
+    },
+    {
+      "epoch": 2.7508479802651866,
+      "grad_norm": 0.05165224149823189,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 26763
+    },
+    {
+      "epoch": 2.7511563367252543,
+      "grad_norm": 0.04024217650294304,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 26766
+    },
+    {
+      "epoch": 2.7514646931853224,
+      "grad_norm": 0.05387435853481293,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 26769
+    },
+    {
+      "epoch": 2.75177304964539,
+      "grad_norm": 0.11387746036052704,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 26772
+    },
+    {
+      "epoch": 2.752081406105458,
+      "grad_norm": 0.06504768133163452,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 26775
+    },
+    {
+      "epoch": 2.752389762565526,
+      "grad_norm": 0.07217580825090408,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 26778
+    },
+    {
+      "epoch": 2.7526981190255935,
+      "grad_norm": 0.07751470804214478,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 26781
+    },
+    {
+      "epoch": 2.7530064754856616,
+      "grad_norm": 0.06282947957515717,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 26784
+    },
+    {
+      "epoch": 2.7533148319457292,
+      "grad_norm": 0.04301406815648079,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 26787
+    },
+    {
+      "epoch": 2.753623188405797,
+      "grad_norm": 0.06154756247997284,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 26790
+    },
+    {
+      "epoch": 2.753931544865865,
+      "grad_norm": 0.04729504510760307,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 26793
+    },
+    {
+      "epoch": 2.7542399013259327,
+      "grad_norm": 0.06597118079662323,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 26796
+    },
+    {
+      "epoch": 2.7545482577860008,
+      "grad_norm": 0.050065554678440094,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 26799
+    },
+    {
+      "epoch": 2.7548566142460684,
+      "grad_norm": 0.048531875014305115,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 26802
+    },
+    {
+      "epoch": 2.7551649707061365,
+      "grad_norm": 0.04954374581575394,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 26805
+    },
+    {
+      "epoch": 2.755473327166204,
+      "grad_norm": 0.06998781114816666,
+      "learning_rate": 0.01,
+      "loss": 2.0537,
+      "step": 26808
+    },
+    {
+      "epoch": 2.755781683626272,
+      "grad_norm": 0.08933461457490921,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 26811
+    },
+    {
+      "epoch": 2.75609004008634,
+      "grad_norm": 0.06067082658410072,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 26814
+    },
+    {
+      "epoch": 2.7563983965464076,
+      "grad_norm": 0.07710520923137665,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 26817
+    },
+    {
+      "epoch": 2.7567067530064753,
+      "grad_norm": 0.05703970044851303,
+      "learning_rate": 0.01,
+      "loss": 2.0505,
+      "step": 26820
+    },
+    {
+      "epoch": 2.7570151094665434,
+      "grad_norm": 0.1251015067100525,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 26823
+    },
+    {
+      "epoch": 2.757323465926611,
+      "grad_norm": 0.05942771956324577,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 26826
+    },
+    {
+      "epoch": 2.757631822386679,
+      "grad_norm": 0.03195277228951454,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 26829
+    },
+    {
+      "epoch": 2.757940178846747,
+      "grad_norm": 0.03609882667660713,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 26832
+    },
+    {
+      "epoch": 2.758248535306815,
+      "grad_norm": 0.04004659876227379,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 26835
+    },
+    {
+      "epoch": 2.7585568917668826,
+      "grad_norm": 0.06475567817687988,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 26838
+    },
+    {
+      "epoch": 2.7588652482269502,
+      "grad_norm": 0.04894067347049713,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 26841
+    },
+    {
+      "epoch": 2.7591736046870183,
+      "grad_norm": 0.11116486042737961,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 26844
+    },
+    {
+      "epoch": 2.759481961147086,
+      "grad_norm": 0.0832393690943718,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 26847
+    },
+    {
+      "epoch": 2.7597903176071537,
+      "grad_norm": 0.06433276832103729,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 26850
+    },
+    {
+      "epoch": 2.7600986740672218,
+      "grad_norm": 0.06255584210157394,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 26853
+    },
+    {
+      "epoch": 2.7604070305272894,
+      "grad_norm": 0.06610293686389923,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 26856
+    },
+    {
+      "epoch": 2.7607153869873575,
+      "grad_norm": 0.048567045480012894,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 26859
+    },
+    {
+      "epoch": 2.761023743447425,
+      "grad_norm": 0.06945166736841202,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 26862
+    },
+    {
+      "epoch": 2.7613320999074933,
+      "grad_norm": 0.07567547261714935,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 26865
+    },
+    {
+      "epoch": 2.761640456367561,
+      "grad_norm": 0.09087162464857101,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 26868
+    },
+    {
+      "epoch": 2.7619488128276286,
+      "grad_norm": 0.07904476672410965,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 26871
+    },
+    {
+      "epoch": 2.7622571692876967,
+      "grad_norm": 0.08835271000862122,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 26874
+    },
+    {
+      "epoch": 2.7625655257477644,
+      "grad_norm": 0.042680736631155014,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 26877
+    },
+    {
+      "epoch": 2.762873882207832,
+      "grad_norm": 0.034312695264816284,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 26880
+    },
+    {
+      "epoch": 2.7631822386679,
+      "grad_norm": 0.04073645919561386,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 26883
+    },
+    {
+      "epoch": 2.763490595127968,
+      "grad_norm": 0.11242496222257614,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 26886
+    },
+    {
+      "epoch": 2.763798951588036,
+      "grad_norm": 0.048365022987127304,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 26889
+    },
+    {
+      "epoch": 2.7641073080481036,
+      "grad_norm": 0.06508596241474152,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 26892
+    },
+    {
+      "epoch": 2.7644156645081717,
+      "grad_norm": 0.08372494578361511,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 26895
+    },
+    {
+      "epoch": 2.7647240209682393,
+      "grad_norm": 0.08499015122652054,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 26898
+    },
+    {
+      "epoch": 2.765032377428307,
+      "grad_norm": 0.08214948326349258,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 26901
+    },
+    {
+      "epoch": 2.765340733888375,
+      "grad_norm": 0.0902111828327179,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 26904
+    },
+    {
+      "epoch": 2.7656490903484428,
+      "grad_norm": 0.0556274875998497,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 26907
+    },
+    {
+      "epoch": 2.7659574468085104,
+      "grad_norm": 0.06845957040786743,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 26910
+    },
+    {
+      "epoch": 2.7662658032685785,
+      "grad_norm": 0.051283106207847595,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 26913
+    },
+    {
+      "epoch": 2.766574159728646,
+      "grad_norm": 0.03301118686795235,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 26916
+    },
+    {
+      "epoch": 2.7668825161887143,
+      "grad_norm": 0.05253078415989876,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 26919
+    },
+    {
+      "epoch": 2.767190872648782,
+      "grad_norm": 0.1089513972401619,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 26922
+    },
+    {
+      "epoch": 2.76749922910885,
+      "grad_norm": 0.06042607128620148,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 26925
+    },
+    {
+      "epoch": 2.7678075855689177,
+      "grad_norm": 0.09722250699996948,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 26928
+    },
+    {
+      "epoch": 2.7681159420289854,
+      "grad_norm": 0.04542861133813858,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 26931
+    },
+    {
+      "epoch": 2.7684242984890535,
+      "grad_norm": 0.14705310761928558,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 26934
+    },
+    {
+      "epoch": 2.768732654949121,
+      "grad_norm": 0.03334478661417961,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 26937
+    },
+    {
+      "epoch": 2.769041011409189,
+      "grad_norm": 0.054583244025707245,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 26940
+    },
+    {
+      "epoch": 2.769349367869257,
+      "grad_norm": 0.03972140699625015,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 26943
+    },
+    {
+      "epoch": 2.7696577243293246,
+      "grad_norm": 0.04227004200220108,
+      "learning_rate": 0.01,
+      "loss": 2.052,
+      "step": 26946
+    },
+    {
+      "epoch": 2.7699660807893927,
+      "grad_norm": 0.035285089164972305,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 26949
+    },
+    {
+      "epoch": 2.7702744372494603,
+      "grad_norm": 0.08037975430488586,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 26952
+    },
+    {
+      "epoch": 2.7705827937095284,
+      "grad_norm": 0.06296487152576447,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 26955
+    },
+    {
+      "epoch": 2.770891150169596,
+      "grad_norm": 0.08724946528673172,
+      "learning_rate": 0.01,
+      "loss": 2.0515,
+      "step": 26958
+    },
+    {
+      "epoch": 2.7711995066296637,
+      "grad_norm": 0.06455433368682861,
+      "learning_rate": 0.01,
+      "loss": 2.0573,
+      "step": 26961
+    },
+    {
+      "epoch": 2.771507863089732,
+      "grad_norm": 0.07615622133016586,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 26964
+    },
+    {
+      "epoch": 2.7718162195497995,
+      "grad_norm": 0.06695935130119324,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 26967
+    },
+    {
+      "epoch": 2.772124576009867,
+      "grad_norm": 0.04257410392165184,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 26970
+    },
+    {
+      "epoch": 2.7724329324699353,
+      "grad_norm": 0.1329718828201294,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 26973
+    },
+    {
+      "epoch": 2.772741288930003,
+      "grad_norm": 0.06937066465616226,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 26976
+    },
+    {
+      "epoch": 2.773049645390071,
+      "grad_norm": 0.07017137110233307,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 26979
+    },
+    {
+      "epoch": 2.7733580018501387,
+      "grad_norm": 0.03589661046862602,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 26982
+    },
+    {
+      "epoch": 2.773666358310207,
+      "grad_norm": 0.04110831022262573,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 26985
+    },
+    {
+      "epoch": 2.7739747147702745,
+      "grad_norm": 0.06182010471820831,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 26988
+    },
+    {
+      "epoch": 2.774283071230342,
+      "grad_norm": 0.08161917328834534,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 26991
+    },
+    {
+      "epoch": 2.7745914276904102,
+      "grad_norm": 0.05832149460911751,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 26994
+    },
+    {
+      "epoch": 2.774899784150478,
+      "grad_norm": 0.08298707008361816,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 26997
+    },
+    {
+      "epoch": 2.7752081406105455,
+      "grad_norm": 0.04794152081012726,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 27000
+    },
+    {
+      "epoch": 2.7755164970706137,
+      "grad_norm": 0.04620853811502457,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 27003
+    },
+    {
+      "epoch": 2.7758248535306813,
+      "grad_norm": 0.045553676784038544,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 27006
+    },
+    {
+      "epoch": 2.7761332099907494,
+      "grad_norm": 0.058516908437013626,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 27009
+    },
+    {
+      "epoch": 2.776441566450817,
+      "grad_norm": 0.05661854147911072,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 27012
+    },
+    {
+      "epoch": 2.776749922910885,
+      "grad_norm": 0.05626550316810608,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 27015
+    },
+    {
+      "epoch": 2.777058279370953,
+      "grad_norm": 0.08505663275718689,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 27018
+    },
+    {
+      "epoch": 2.7773666358310205,
+      "grad_norm": 0.08449403196573257,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 27021
+    },
+    {
+      "epoch": 2.7776749922910886,
+      "grad_norm": 0.08090987801551819,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 27024
+    },
+    {
+      "epoch": 2.7779833487511563,
+      "grad_norm": 0.03180038183927536,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 27027
+    },
+    {
+      "epoch": 2.778291705211224,
+      "grad_norm": 0.03052234649658203,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 27030
+    },
+    {
+      "epoch": 2.778600061671292,
+      "grad_norm": 0.07030376046895981,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 27033
+    },
+    {
+      "epoch": 2.7789084181313597,
+      "grad_norm": 0.1097898781299591,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 27036
+    },
+    {
+      "epoch": 2.779216774591428,
+      "grad_norm": 0.10938852280378342,
+      "learning_rate": 0.01,
+      "loss": 2.0358,
+      "step": 27039
+    },
+    {
+      "epoch": 2.7795251310514955,
+      "grad_norm": 0.17036540806293488,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 27042
+    },
+    {
+      "epoch": 2.7798334875115636,
+      "grad_norm": 0.07512038946151733,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 27045
+    },
+    {
+      "epoch": 2.780141843971631,
+      "grad_norm": 0.03864987567067146,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 27048
+    },
+    {
+      "epoch": 2.780450200431699,
+      "grad_norm": 0.07508687674999237,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 27051
+    },
+    {
+      "epoch": 2.780758556891767,
+      "grad_norm": 0.03866630047559738,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 27054
+    },
+    {
+      "epoch": 2.7810669133518346,
+      "grad_norm": 0.044196490198373795,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 27057
+    },
+    {
+      "epoch": 2.7813752698119023,
+      "grad_norm": 0.03751998767256737,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 27060
+    },
+    {
+      "epoch": 2.7816836262719704,
+      "grad_norm": 0.04259081184864044,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 27063
+    },
+    {
+      "epoch": 2.781991982732038,
+      "grad_norm": 0.04910355806350708,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 27066
+    },
+    {
+      "epoch": 2.782300339192106,
+      "grad_norm": 0.07969753444194794,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 27069
+    },
+    {
+      "epoch": 2.782608695652174,
+      "grad_norm": 0.04511536657810211,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 27072
+    },
+    {
+      "epoch": 2.782917052112242,
+      "grad_norm": 0.03906785696744919,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 27075
+    },
+    {
+      "epoch": 2.7832254085723096,
+      "grad_norm": 0.045391857624053955,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 27078
+    },
+    {
+      "epoch": 2.7835337650323773,
+      "grad_norm": 0.04607332870364189,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 27081
+    },
+    {
+      "epoch": 2.7838421214924454,
+      "grad_norm": 0.10543697327375412,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 27084
+    },
+    {
+      "epoch": 2.784150477952513,
+      "grad_norm": 0.045777421444654465,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 27087
+    },
+    {
+      "epoch": 2.7844588344125807,
+      "grad_norm": 0.11416932195425034,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 27090
+    },
+    {
+      "epoch": 2.784767190872649,
+      "grad_norm": 0.04630710557103157,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 27093
+    },
+    {
+      "epoch": 2.785075547332717,
+      "grad_norm": 0.1050376445055008,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 27096
+    },
+    {
+      "epoch": 2.7853839037927846,
+      "grad_norm": 0.04592348635196686,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 27099
+    },
+    {
+      "epoch": 2.785692260252852,
+      "grad_norm": 0.058079514652490616,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 27102
+    },
+    {
+      "epoch": 2.7860006167129203,
+      "grad_norm": 0.12133990228176117,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 27105
+    },
+    {
+      "epoch": 2.786308973172988,
+      "grad_norm": 0.09523560851812363,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 27108
+    },
+    {
+      "epoch": 2.7866173296330556,
+      "grad_norm": 0.05585847795009613,
+      "learning_rate": 0.01,
+      "loss": 2.0458,
+      "step": 27111
+    },
+    {
+      "epoch": 2.7869256860931237,
+      "grad_norm": 0.03669681027531624,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 27114
+    },
+    {
+      "epoch": 2.7872340425531914,
+      "grad_norm": 0.05370241776108742,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 27117
+    },
+    {
+      "epoch": 2.787542399013259,
+      "grad_norm": 0.04395952448248863,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 27120
+    },
+    {
+      "epoch": 2.787850755473327,
+      "grad_norm": 0.04012378677725792,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 27123
+    },
+    {
+      "epoch": 2.7881591119333953,
+      "grad_norm": 0.08143822848796844,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 27126
+    },
+    {
+      "epoch": 2.788467468393463,
+      "grad_norm": 0.05250631272792816,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 27129
+    },
+    {
+      "epoch": 2.7887758248535306,
+      "grad_norm": 0.09945302456617355,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 27132
+    },
+    {
+      "epoch": 2.7890841813135987,
+      "grad_norm": 0.06001584604382515,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 27135
+    },
+    {
+      "epoch": 2.7893925377736664,
+      "grad_norm": 0.05244366079568863,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 27138
+    },
+    {
+      "epoch": 2.789700894233734,
+      "grad_norm": 0.06054399535059929,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 27141
+    },
+    {
+      "epoch": 2.790009250693802,
+      "grad_norm": 0.10409083217382431,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 27144
+    },
+    {
+      "epoch": 2.79031760715387,
+      "grad_norm": 0.06229739263653755,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 27147
+    },
+    {
+      "epoch": 2.7906259636139374,
+      "grad_norm": 0.06340862810611725,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 27150
+    },
+    {
+      "epoch": 2.7909343200740055,
+      "grad_norm": 0.035666175186634064,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 27153
+    },
+    {
+      "epoch": 2.7912426765340737,
+      "grad_norm": 0.043176453560590744,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 27156
+    },
+    {
+      "epoch": 2.7915510329941413,
+      "grad_norm": 0.042664237320423126,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 27159
+    },
+    {
+      "epoch": 2.791859389454209,
+      "grad_norm": 0.08263540267944336,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 27162
+    },
+    {
+      "epoch": 2.792167745914277,
+      "grad_norm": 0.05592063441872597,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 27165
+    },
+    {
+      "epoch": 2.7924761023743447,
+      "grad_norm": 0.0740666538476944,
+      "learning_rate": 0.01,
+      "loss": 2.0348,
+      "step": 27168
+    },
+    {
+      "epoch": 2.7927844588344124,
+      "grad_norm": 0.07478248327970505,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 27171
+    },
+    {
+      "epoch": 2.7930928152944805,
+      "grad_norm": 0.08470360189676285,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 27174
+    },
+    {
+      "epoch": 2.793401171754548,
+      "grad_norm": 0.04486660659313202,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 27177
+    },
+    {
+      "epoch": 2.793709528214616,
+      "grad_norm": 0.12479998171329498,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 27180
+    },
+    {
+      "epoch": 2.794017884674684,
+      "grad_norm": 0.05652941018342972,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 27183
+    },
+    {
+      "epoch": 2.794326241134752,
+      "grad_norm": 0.054577797651290894,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 27186
+    },
+    {
+      "epoch": 2.7946345975948197,
+      "grad_norm": 0.03588878735899925,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 27189
+    },
+    {
+      "epoch": 2.7949429540548874,
+      "grad_norm": 0.0890214741230011,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 27192
+    },
+    {
+      "epoch": 2.7952513105149555,
+      "grad_norm": 0.0901033952832222,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 27195
+    },
+    {
+      "epoch": 2.795559666975023,
+      "grad_norm": 0.09306314587593079,
+      "learning_rate": 0.01,
+      "loss": 2.0337,
+      "step": 27198
+    },
+    {
+      "epoch": 2.7958680234350908,
+      "grad_norm": 0.07897783815860748,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 27201
+    },
+    {
+      "epoch": 2.796176379895159,
+      "grad_norm": 0.08055876195430756,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 27204
+    },
+    {
+      "epoch": 2.7964847363552265,
+      "grad_norm": 0.058927252888679504,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 27207
+    },
+    {
+      "epoch": 2.7967930928152946,
+      "grad_norm": 0.1303315907716751,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 27210
+    },
+    {
+      "epoch": 2.7971014492753623,
+      "grad_norm": 0.08321187645196915,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 27213
+    },
+    {
+      "epoch": 2.7974098057354304,
+      "grad_norm": 0.0715765729546547,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 27216
+    },
+    {
+      "epoch": 2.797718162195498,
+      "grad_norm": 0.047501109540462494,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 27219
+    },
+    {
+      "epoch": 2.7980265186555657,
+      "grad_norm": 0.04297725111246109,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 27222
+    },
+    {
+      "epoch": 2.798334875115634,
+      "grad_norm": 0.08612347394227982,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 27225
+    },
+    {
+      "epoch": 2.7986432315757015,
+      "grad_norm": 0.06801822036504745,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 27228
+    },
+    {
+      "epoch": 2.798951588035769,
+      "grad_norm": 0.103379026055336,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 27231
+    },
+    {
+      "epoch": 2.7992599444958373,
+      "grad_norm": 0.040429919958114624,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 27234
+    },
+    {
+      "epoch": 2.799568300955905,
+      "grad_norm": 0.09817671030759811,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 27237
+    },
+    {
+      "epoch": 2.799876657415973,
+      "grad_norm": 0.06191599741578102,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 27240
+    },
+    {
+      "epoch": 2.8001850138760407,
+      "grad_norm": 0.07992551475763321,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 27243
+    },
+    {
+      "epoch": 2.800493370336109,
+      "grad_norm": 0.055606722831726074,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 27246
+    },
+    {
+      "epoch": 2.8008017267961764,
+      "grad_norm": 0.1322477012872696,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 27249
+    },
+    {
+      "epoch": 2.801110083256244,
+      "grad_norm": 0.05603817105293274,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 27252
+    },
+    {
+      "epoch": 2.801418439716312,
+      "grad_norm": 0.06935392320156097,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 27255
+    },
+    {
+      "epoch": 2.80172679617638,
+      "grad_norm": 0.04515109956264496,
+      "learning_rate": 0.01,
+      "loss": 2.053,
+      "step": 27258
+    },
+    {
+      "epoch": 2.8020351526364475,
+      "grad_norm": 0.050864629447460175,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 27261
+    },
+    {
+      "epoch": 2.8023435090965156,
+      "grad_norm": 0.038876548409461975,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 27264
+    },
+    {
+      "epoch": 2.8026518655565833,
+      "grad_norm": 0.03906438872218132,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 27267
+    },
+    {
+      "epoch": 2.8029602220166514,
+      "grad_norm": 0.08978519588708878,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 27270
+    },
+    {
+      "epoch": 2.803268578476719,
+      "grad_norm": 0.07353704422712326,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 27273
+    },
+    {
+      "epoch": 2.803576934936787,
+      "grad_norm": 0.0424012616276741,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 27276
+    },
+    {
+      "epoch": 2.803885291396855,
+      "grad_norm": 0.05939796194434166,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 27279
+    },
+    {
+      "epoch": 2.8041936478569225,
+      "grad_norm": 0.06859377771615982,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 27282
+    },
+    {
+      "epoch": 2.8045020043169906,
+      "grad_norm": 0.1016155332326889,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 27285
+    },
+    {
+      "epoch": 2.8048103607770583,
+      "grad_norm": 0.05608060583472252,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 27288
+    },
+    {
+      "epoch": 2.805118717237126,
+      "grad_norm": 0.043284934014081955,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 27291
+    },
+    {
+      "epoch": 2.805427073697194,
+      "grad_norm": 0.04967540502548218,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 27294
+    },
+    {
+      "epoch": 2.8057354301572617,
+      "grad_norm": 0.040739741176366806,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 27297
+    },
+    {
+      "epoch": 2.80604378661733,
+      "grad_norm": 0.039170749485492706,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 27300
+    },
+    {
+      "epoch": 2.8063521430773974,
+      "grad_norm": 0.040820952504873276,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 27303
+    },
+    {
+      "epoch": 2.8066604995374655,
+      "grad_norm": 0.07061021775007248,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 27306
+    },
+    {
+      "epoch": 2.806968855997533,
+      "grad_norm": 0.054698627442121506,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 27309
+    },
+    {
+      "epoch": 2.807277212457601,
+      "grad_norm": 0.04737408831715584,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 27312
+    },
+    {
+      "epoch": 2.807585568917669,
+      "grad_norm": 0.11695525050163269,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 27315
+    },
+    {
+      "epoch": 2.8078939253777366,
+      "grad_norm": 0.0921463891863823,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 27318
+    },
+    {
+      "epoch": 2.8082022818378043,
+      "grad_norm": 0.03874325007200241,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 27321
+    },
+    {
+      "epoch": 2.8085106382978724,
+      "grad_norm": 0.04040201008319855,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 27324
+    },
+    {
+      "epoch": 2.80881899475794,
+      "grad_norm": 0.035648833960294724,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 27327
+    },
+    {
+      "epoch": 2.809127351218008,
+      "grad_norm": 0.0893108919262886,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 27330
+    },
+    {
+      "epoch": 2.809435707678076,
+      "grad_norm": 0.1203017309308052,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 27333
+    },
+    {
+      "epoch": 2.809744064138144,
+      "grad_norm": 0.045630257576704025,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 27336
+    },
+    {
+      "epoch": 2.8100524205982116,
+      "grad_norm": 0.04097437858581543,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 27339
+    },
+    {
+      "epoch": 2.8103607770582792,
+      "grad_norm": 0.05476262420415878,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 27342
+    },
+    {
+      "epoch": 2.8106691335183474,
+      "grad_norm": 0.05577860027551651,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 27345
+    },
+    {
+      "epoch": 2.810977489978415,
+      "grad_norm": 0.04357774555683136,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 27348
+    },
+    {
+      "epoch": 2.8112858464384827,
+      "grad_norm": 0.04594416916370392,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 27351
+    },
+    {
+      "epoch": 2.8115942028985508,
+      "grad_norm": 0.041454486548900604,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 27354
+    },
+    {
+      "epoch": 2.8119025593586184,
+      "grad_norm": 0.05279424414038658,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 27357
+    },
+    {
+      "epoch": 2.8122109158186865,
+      "grad_norm": 0.10371188074350357,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 27360
+    },
+    {
+      "epoch": 2.812519272278754,
+      "grad_norm": 0.053941383957862854,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 27363
+    },
+    {
+      "epoch": 2.8128276287388223,
+      "grad_norm": 0.12082924693822861,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 27366
+    },
+    {
+      "epoch": 2.81313598519889,
+      "grad_norm": 0.11066311597824097,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 27369
+    },
+    {
+      "epoch": 2.8134443416589576,
+      "grad_norm": 0.07547413557767868,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 27372
+    },
+    {
+      "epoch": 2.8137526981190257,
+      "grad_norm": 0.06772000342607498,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 27375
+    },
+    {
+      "epoch": 2.8140610545790934,
+      "grad_norm": 0.03906597942113876,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 27378
+    },
+    {
+      "epoch": 2.814369411039161,
+      "grad_norm": 0.048235367983579636,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 27381
+    },
+    {
+      "epoch": 2.814677767499229,
+      "grad_norm": 0.08060193806886673,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 27384
+    },
+    {
+      "epoch": 2.814986123959297,
+      "grad_norm": 0.06527181714773178,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 27387
+    },
+    {
+      "epoch": 2.815294480419365,
+      "grad_norm": 0.04199523106217384,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 27390
+    },
+    {
+      "epoch": 2.8156028368794326,
+      "grad_norm": 0.07436412572860718,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 27393
+    },
+    {
+      "epoch": 2.8159111933395007,
+      "grad_norm": 0.05724874138832092,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 27396
+    },
+    {
+      "epoch": 2.8162195497995683,
+      "grad_norm": 0.03609538823366165,
+      "learning_rate": 0.01,
+      "loss": 2.0376,
+      "step": 27399
+    },
+    {
+      "epoch": 2.816527906259636,
+      "grad_norm": 0.03008181042969227,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 27402
+    },
+    {
+      "epoch": 2.816836262719704,
+      "grad_norm": 0.0635017454624176,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 27405
+    },
+    {
+      "epoch": 2.8171446191797718,
+      "grad_norm": 0.07800552248954773,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 27408
+    },
+    {
+      "epoch": 2.8174529756398394,
+      "grad_norm": 0.09289496392011642,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 27411
+    },
+    {
+      "epoch": 2.8177613320999075,
+      "grad_norm": 0.0791633352637291,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 27414
+    },
+    {
+      "epoch": 2.818069688559975,
+      "grad_norm": 0.06715335696935654,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 27417
+    },
+    {
+      "epoch": 2.8183780450200433,
+      "grad_norm": 0.0814514085650444,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 27420
+    },
+    {
+      "epoch": 2.818686401480111,
+      "grad_norm": 0.09124473482370377,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 27423
+    },
+    {
+      "epoch": 2.818994757940179,
+      "grad_norm": 0.03711434826254845,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 27426
+    },
+    {
+      "epoch": 2.8193031144002467,
+      "grad_norm": 0.05914726108312607,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 27429
+    },
+    {
+      "epoch": 2.8196114708603144,
+      "grad_norm": 0.04358556494116783,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 27432
+    },
+    {
+      "epoch": 2.8199198273203825,
+      "grad_norm": 0.1319088488817215,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 27435
+    },
+    {
+      "epoch": 2.82022818378045,
+      "grad_norm": 0.07464331388473511,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 27438
+    },
+    {
+      "epoch": 2.820536540240518,
+      "grad_norm": 0.08253163844347,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 27441
+    },
+    {
+      "epoch": 2.820844896700586,
+      "grad_norm": 0.042612046003341675,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 27444
+    },
+    {
+      "epoch": 2.8211532531606536,
+      "grad_norm": 0.04726095870137215,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 27447
+    },
+    {
+      "epoch": 2.8214616096207217,
+      "grad_norm": 0.06982813775539398,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 27450
+    },
+    {
+      "epoch": 2.8217699660807893,
+      "grad_norm": 0.051457736641168594,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 27453
+    },
+    {
+      "epoch": 2.8220783225408574,
+      "grad_norm": 0.07463426142930984,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 27456
+    },
+    {
+      "epoch": 2.822386679000925,
+      "grad_norm": 0.048826638609170914,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 27459
+    },
+    {
+      "epoch": 2.8226950354609928,
+      "grad_norm": 0.05443079397082329,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 27462
+    },
+    {
+      "epoch": 2.823003391921061,
+      "grad_norm": 0.10064379125833511,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 27465
+    },
+    {
+      "epoch": 2.8233117483811285,
+      "grad_norm": 0.0755578950047493,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 27468
+    },
+    {
+      "epoch": 2.823620104841196,
+      "grad_norm": 0.05240621045231819,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 27471
+    },
+    {
+      "epoch": 2.8239284613012643,
+      "grad_norm": 0.08781099319458008,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 27474
+    },
+    {
+      "epoch": 2.824236817761332,
+      "grad_norm": 0.04524267092347145,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 27477
+    },
+    {
+      "epoch": 2.8245451742214,
+      "grad_norm": 0.05119558051228523,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 27480
+    },
+    {
+      "epoch": 2.8248535306814677,
+      "grad_norm": 0.06969384104013443,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 27483
+    },
+    {
+      "epoch": 2.825161887141536,
+      "grad_norm": 0.09887497872114182,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 27486
+    },
+    {
+      "epoch": 2.8254702436016035,
+      "grad_norm": 0.04231324791908264,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 27489
+    },
+    {
+      "epoch": 2.825778600061671,
+      "grad_norm": 0.11022655665874481,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 27492
+    },
+    {
+      "epoch": 2.8260869565217392,
+      "grad_norm": 0.16821467876434326,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 27495
+    },
+    {
+      "epoch": 2.826395312981807,
+      "grad_norm": 0.13946877419948578,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 27498
+    },
+    {
+      "epoch": 2.8267036694418746,
+      "grad_norm": 0.07387804985046387,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 27501
+    },
+    {
+      "epoch": 2.8270120259019427,
+      "grad_norm": 0.06437870115041733,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 27504
+    },
+    {
+      "epoch": 2.8273203823620103,
+      "grad_norm": 0.04310622811317444,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 27507
+    },
+    {
+      "epoch": 2.8276287388220784,
+      "grad_norm": 0.04511455446481705,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 27510
+    },
+    {
+      "epoch": 2.827937095282146,
+      "grad_norm": 0.05232998728752136,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 27513
+    },
+    {
+      "epoch": 2.828245451742214,
+      "grad_norm": 0.03596142679452896,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 27516
+    },
+    {
+      "epoch": 2.828553808202282,
+      "grad_norm": 0.035050373524427414,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 27519
+    },
+    {
+      "epoch": 2.8288621646623495,
+      "grad_norm": 0.03525279834866524,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 27522
+    },
+    {
+      "epoch": 2.8291705211224176,
+      "grad_norm": 0.1629800945520401,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 27525
+    },
+    {
+      "epoch": 2.8294788775824853,
+      "grad_norm": 0.06687575578689575,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 27528
+    },
+    {
+      "epoch": 2.829787234042553,
+      "grad_norm": 0.05308271199464798,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 27531
+    },
+    {
+      "epoch": 2.830095590502621,
+      "grad_norm": 0.05350314825773239,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 27534
+    },
+    {
+      "epoch": 2.8304039469626887,
+      "grad_norm": 0.03929050639271736,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 27537
+    },
+    {
+      "epoch": 2.830712303422757,
+      "grad_norm": 0.037572652101516724,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 27540
+    },
+    {
+      "epoch": 2.8310206598828245,
+      "grad_norm": 0.05316625535488129,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 27543
+    },
+    {
+      "epoch": 2.8313290163428926,
+      "grad_norm": 0.04561341553926468,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 27546
+    },
+    {
+      "epoch": 2.8316373728029602,
+      "grad_norm": 0.055242884904146194,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 27549
+    },
+    {
+      "epoch": 2.831945729263028,
+      "grad_norm": 0.057326629757881165,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 27552
+    },
+    {
+      "epoch": 2.832254085723096,
+      "grad_norm": 0.050966937094926834,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 27555
+    },
+    {
+      "epoch": 2.8325624421831637,
+      "grad_norm": 0.07518629729747772,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 27558
+    },
+    {
+      "epoch": 2.8328707986432313,
+      "grad_norm": 0.10435313731431961,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 27561
+    },
+    {
+      "epoch": 2.8331791551032994,
+      "grad_norm": 0.07228770107030869,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 27564
+    },
+    {
+      "epoch": 2.833487511563367,
+      "grad_norm": 0.06778567284345627,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 27567
+    },
+    {
+      "epoch": 2.833795868023435,
+      "grad_norm": 0.03698448836803436,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 27570
+    },
+    {
+      "epoch": 2.834104224483503,
+      "grad_norm": 0.027820177376270294,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 27573
+    },
+    {
+      "epoch": 2.834412580943571,
+      "grad_norm": 0.07996013760566711,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 27576
+    },
+    {
+      "epoch": 2.8347209374036386,
+      "grad_norm": 0.04629092290997505,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 27579
+    },
+    {
+      "epoch": 2.8350292938637063,
+      "grad_norm": 0.04143141210079193,
+      "learning_rate": 0.01,
+      "loss": 2.039,
+      "step": 27582
+    },
+    {
+      "epoch": 2.8353376503237744,
+      "grad_norm": 0.12624233961105347,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 27585
+    },
+    {
+      "epoch": 2.835646006783842,
+      "grad_norm": 0.1815887838602066,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 27588
+    },
+    {
+      "epoch": 2.8359543632439097,
+      "grad_norm": 0.14360418915748596,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 27591
+    },
+    {
+      "epoch": 2.836262719703978,
+      "grad_norm": 0.10748742520809174,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 27594
+    },
+    {
+      "epoch": 2.836571076164046,
+      "grad_norm": 0.0488545261323452,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 27597
+    },
+    {
+      "epoch": 2.8368794326241136,
+      "grad_norm": 0.03699369728565216,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 27600
+    },
+    {
+      "epoch": 2.8371877890841812,
+      "grad_norm": 0.04813402146100998,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 27603
+    },
+    {
+      "epoch": 2.8374961455442493,
+      "grad_norm": 0.06493838876485825,
+      "learning_rate": 0.01,
+      "loss": 2.0502,
+      "step": 27606
+    },
+    {
+      "epoch": 2.837804502004317,
+      "grad_norm": 0.05451219528913498,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 27609
+    },
+    {
+      "epoch": 2.8381128584643847,
+      "grad_norm": 0.04877667874097824,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 27612
+    },
+    {
+      "epoch": 2.8384212149244528,
+      "grad_norm": 0.05827799066901207,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 27615
+    },
+    {
+      "epoch": 2.8387295713845204,
+      "grad_norm": 0.040546271950006485,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 27618
+    },
+    {
+      "epoch": 2.839037927844588,
+      "grad_norm": 0.03764180466532707,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 27621
+    },
+    {
+      "epoch": 2.839346284304656,
+      "grad_norm": 0.06349320709705353,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 27624
+    },
+    {
+      "epoch": 2.8396546407647243,
+      "grad_norm": 0.07079531252384186,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 27627
+    },
+    {
+      "epoch": 2.839962997224792,
+      "grad_norm": 0.05074724927544594,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 27630
+    },
+    {
+      "epoch": 2.8402713536848596,
+      "grad_norm": 0.10624159872531891,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 27633
+    },
+    {
+      "epoch": 2.8405797101449277,
+      "grad_norm": 0.08726594597101212,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 27636
+    },
+    {
+      "epoch": 2.8408880666049954,
+      "grad_norm": 0.04581126943230629,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 27639
+    },
+    {
+      "epoch": 2.841196423065063,
+      "grad_norm": 0.042876582592725754,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 27642
+    },
+    {
+      "epoch": 2.841504779525131,
+      "grad_norm": 0.03951287642121315,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 27645
+    },
+    {
+      "epoch": 2.841813135985199,
+      "grad_norm": 0.04439757391810417,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 27648
+    },
+    {
+      "epoch": 2.8421214924452665,
+      "grad_norm": 0.05910231173038483,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 27651
+    },
+    {
+      "epoch": 2.8424298489053346,
+      "grad_norm": 0.1063590720295906,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 27654
+    },
+    {
+      "epoch": 2.8427382053654027,
+      "grad_norm": 0.042655881494283676,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 27657
+    },
+    {
+      "epoch": 2.8430465618254703,
+      "grad_norm": 0.037402208894491196,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 27660
+    },
+    {
+      "epoch": 2.843354918285538,
+      "grad_norm": 0.03912290558218956,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 27663
+    },
+    {
+      "epoch": 2.843663274745606,
+      "grad_norm": 0.1316000372171402,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 27666
+    },
+    {
+      "epoch": 2.8439716312056738,
+      "grad_norm": 0.10243986546993256,
+      "learning_rate": 0.01,
+      "loss": 2.0419,
+      "step": 27669
+    },
+    {
+      "epoch": 2.8442799876657414,
+      "grad_norm": 0.09918151050806046,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 27672
+    },
+    {
+      "epoch": 2.8445883441258095,
+      "grad_norm": 0.0572928749024868,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 27675
+    },
+    {
+      "epoch": 2.844896700585877,
+      "grad_norm": 0.09781496226787567,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 27678
+    },
+    {
+      "epoch": 2.845205057045945,
+      "grad_norm": 0.04969809949398041,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 27681
+    },
+    {
+      "epoch": 2.845513413506013,
+      "grad_norm": 0.0894259363412857,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 27684
+    },
+    {
+      "epoch": 2.845821769966081,
+      "grad_norm": 0.05788953974843025,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 27687
+    },
+    {
+      "epoch": 2.8461301264261487,
+      "grad_norm": 0.07628165185451508,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 27690
+    },
+    {
+      "epoch": 2.8464384828862164,
+      "grad_norm": 0.07799988240003586,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 27693
+    },
+    {
+      "epoch": 2.8467468393462845,
+      "grad_norm": 0.08373581618070602,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 27696
+    },
+    {
+      "epoch": 2.847055195806352,
+      "grad_norm": 0.054138489067554474,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 27699
+    },
+    {
+      "epoch": 2.84736355226642,
+      "grad_norm": 0.03457311540842056,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 27702
+    },
+    {
+      "epoch": 2.847671908726488,
+      "grad_norm": 0.04888417571783066,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 27705
+    },
+    {
+      "epoch": 2.8479802651865556,
+      "grad_norm": 0.09924766421318054,
+      "learning_rate": 0.01,
+      "loss": 2.0514,
+      "step": 27708
+    },
+    {
+      "epoch": 2.848288621646623,
+      "grad_norm": 0.0531487911939621,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 27711
+    },
+    {
+      "epoch": 2.8485969781066913,
+      "grad_norm": 0.07373910397291183,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 27714
+    },
+    {
+      "epoch": 2.8489053345667594,
+      "grad_norm": 0.09154459089040756,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 27717
+    },
+    {
+      "epoch": 2.849213691026827,
+      "grad_norm": 0.04007700830698013,
+      "learning_rate": 0.01,
+      "loss": 2.0447,
+      "step": 27720
+    },
+    {
+      "epoch": 2.8495220474868947,
+      "grad_norm": 0.06406736373901367,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 27723
+    },
+    {
+      "epoch": 2.849830403946963,
+      "grad_norm": 0.06983067840337753,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 27726
+    },
+    {
+      "epoch": 2.8501387604070305,
+      "grad_norm": 0.04832616075873375,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 27729
+    },
+    {
+      "epoch": 2.850447116867098,
+      "grad_norm": 0.09404818713665009,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 27732
+    },
+    {
+      "epoch": 2.8507554733271663,
+      "grad_norm": 0.12212470918893814,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 27735
+    },
+    {
+      "epoch": 2.851063829787234,
+      "grad_norm": 0.08778437972068787,
+      "learning_rate": 0.01,
+      "loss": 2.0503,
+      "step": 27738
+    },
+    {
+      "epoch": 2.851372186247302,
+      "grad_norm": 0.08406031876802444,
+      "learning_rate": 0.01,
+      "loss": 2.0432,
+      "step": 27741
+    },
+    {
+      "epoch": 2.8516805427073697,
+      "grad_norm": 0.07672454416751862,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 27744
+    },
+    {
+      "epoch": 2.851988899167438,
+      "grad_norm": 0.12199164927005768,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 27747
+    },
+    {
+      "epoch": 2.8522972556275055,
+      "grad_norm": 0.05166463926434517,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 27750
+    },
+    {
+      "epoch": 2.852605612087573,
+      "grad_norm": 0.11618038266897202,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 27753
+    },
+    {
+      "epoch": 2.8529139685476412,
+      "grad_norm": 0.06255706399679184,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 27756
+    },
+    {
+      "epoch": 2.853222325007709,
+      "grad_norm": 0.04340292140841484,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 27759
+    },
+    {
+      "epoch": 2.8535306814677766,
+      "grad_norm": 0.06595482677221298,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 27762
+    },
+    {
+      "epoch": 2.8538390379278447,
+      "grad_norm": 0.03980451449751854,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 27765
+    },
+    {
+      "epoch": 2.8541473943879123,
+      "grad_norm": 0.035016849637031555,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 27768
+    },
+    {
+      "epoch": 2.8544557508479804,
+      "grad_norm": 0.07189803570508957,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 27771
+    },
+    {
+      "epoch": 2.854764107308048,
+      "grad_norm": 0.0759616568684578,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 27774
+    },
+    {
+      "epoch": 2.855072463768116,
+      "grad_norm": 0.05985911190509796,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 27777
+    },
+    {
+      "epoch": 2.855380820228184,
+      "grad_norm": 0.0529702752828598,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 27780
+    },
+    {
+      "epoch": 2.8556891766882515,
+      "grad_norm": 0.046540237963199615,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 27783
+    },
+    {
+      "epoch": 2.8559975331483196,
+      "grad_norm": 0.04505128040909767,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 27786
+    },
+    {
+      "epoch": 2.8563058896083873,
+      "grad_norm": 0.041494566947221756,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 27789
+    },
+    {
+      "epoch": 2.856614246068455,
+      "grad_norm": 0.038743190467357635,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 27792
+    },
+    {
+      "epoch": 2.856922602528523,
+      "grad_norm": 0.062206536531448364,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 27795
+    },
+    {
+      "epoch": 2.8572309589885907,
+      "grad_norm": 0.19090187549591064,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 27798
+    },
+    {
+      "epoch": 2.857539315448659,
+      "grad_norm": 0.06649639457464218,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 27801
+    },
+    {
+      "epoch": 2.8578476719087265,
+      "grad_norm": 0.07956647872924805,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 27804
+    },
+    {
+      "epoch": 2.8581560283687946,
+      "grad_norm": 0.050361573696136475,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 27807
+    },
+    {
+      "epoch": 2.8584643848288622,
+      "grad_norm": 0.036132823675870895,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 27810
+    },
+    {
+      "epoch": 2.85877274128893,
+      "grad_norm": 0.04948203265666962,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 27813
+    },
+    {
+      "epoch": 2.859081097748998,
+      "grad_norm": 0.07491700351238251,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 27816
+    },
+    {
+      "epoch": 2.8593894542090657,
+      "grad_norm": 0.03278898447751999,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 27819
+    },
+    {
+      "epoch": 2.8596978106691333,
+      "grad_norm": 0.04186544567346573,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 27822
+    },
+    {
+      "epoch": 2.8600061671292014,
+      "grad_norm": 0.07123599201440811,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 27825
+    },
+    {
+      "epoch": 2.860314523589269,
+      "grad_norm": 0.038315288722515106,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 27828
+    },
+    {
+      "epoch": 2.860622880049337,
+      "grad_norm": 0.06678714603185654,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 27831
+    },
+    {
+      "epoch": 2.860931236509405,
+      "grad_norm": 0.10680954158306122,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 27834
+    },
+    {
+      "epoch": 2.861239592969473,
+      "grad_norm": 0.13149844110012054,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 27837
+    },
+    {
+      "epoch": 2.8615479494295406,
+      "grad_norm": 0.04700513929128647,
+      "learning_rate": 0.01,
+      "loss": 2.0548,
+      "step": 27840
+    },
+    {
+      "epoch": 2.8618563058896083,
+      "grad_norm": 0.03596799075603485,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 27843
+    },
+    {
+      "epoch": 2.8621646623496764,
+      "grad_norm": 0.042376477271318436,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 27846
+    },
+    {
+      "epoch": 2.862473018809744,
+      "grad_norm": 0.054233819246292114,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 27849
+    },
+    {
+      "epoch": 2.8627813752698117,
+      "grad_norm": 0.04501786455512047,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 27852
+    },
+    {
+      "epoch": 2.86308973172988,
+      "grad_norm": 0.06443289667367935,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 27855
+    },
+    {
+      "epoch": 2.8633980881899475,
+      "grad_norm": 0.05772462114691734,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 27858
+    },
+    {
+      "epoch": 2.8637064446500156,
+      "grad_norm": 0.039729420095682144,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 27861
+    },
+    {
+      "epoch": 2.864014801110083,
+      "grad_norm": 0.1330452859401703,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 27864
+    },
+    {
+      "epoch": 2.8643231575701513,
+      "grad_norm": 0.03952759504318237,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 27867
+    },
+    {
+      "epoch": 2.864631514030219,
+      "grad_norm": 0.05294906720519066,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 27870
+    },
+    {
+      "epoch": 2.8649398704902866,
+      "grad_norm": 0.07945666462182999,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 27873
+    },
+    {
+      "epoch": 2.8652482269503547,
+      "grad_norm": 0.09529011696577072,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 27876
+    },
+    {
+      "epoch": 2.8655565834104224,
+      "grad_norm": 0.038869790732860565,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 27879
+    },
+    {
+      "epoch": 2.86586493987049,
+      "grad_norm": 0.0421532541513443,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 27882
+    },
+    {
+      "epoch": 2.866173296330558,
+      "grad_norm": 0.037143412977457047,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 27885
+    },
+    {
+      "epoch": 2.866481652790626,
+      "grad_norm": 0.0359092615544796,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 27888
+    },
+    {
+      "epoch": 2.866790009250694,
+      "grad_norm": 0.04631698876619339,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 27891
+    },
+    {
+      "epoch": 2.8670983657107616,
+      "grad_norm": 0.06015830859541893,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 27894
+    },
+    {
+      "epoch": 2.8674067221708297,
+      "grad_norm": 0.09565176069736481,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 27897
+    },
+    {
+      "epoch": 2.8677150786308974,
+      "grad_norm": 0.05290444567799568,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 27900
+    },
+    {
+      "epoch": 2.868023435090965,
+      "grad_norm": 0.03383943438529968,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 27903
+    },
+    {
+      "epoch": 2.868331791551033,
+      "grad_norm": 0.05390051752328873,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 27906
+    },
+    {
+      "epoch": 2.868640148011101,
+      "grad_norm": 0.08597470074892044,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 27909
+    },
+    {
+      "epoch": 2.8689485044711684,
+      "grad_norm": 0.059518035501241684,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 27912
+    },
+    {
+      "epoch": 2.8692568609312366,
+      "grad_norm": 0.04992047697305679,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 27915
+    },
+    {
+      "epoch": 2.869565217391304,
+      "grad_norm": 0.04728172719478607,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 27918
+    },
+    {
+      "epoch": 2.8698735738513723,
+      "grad_norm": 0.038830891251564026,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 27921
+    },
+    {
+      "epoch": 2.87018193031144,
+      "grad_norm": 0.039699580520391464,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 27924
+    },
+    {
+      "epoch": 2.870490286771508,
+      "grad_norm": 0.10109611600637436,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 27927
+    },
+    {
+      "epoch": 2.8707986432315757,
+      "grad_norm": 0.07930929958820343,
+      "learning_rate": 0.01,
+      "loss": 2.0463,
+      "step": 27930
+    },
+    {
+      "epoch": 2.8711069996916434,
+      "grad_norm": 0.05544453486800194,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 27933
+    },
+    {
+      "epoch": 2.8714153561517115,
+      "grad_norm": 0.11004924774169922,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 27936
+    },
+    {
+      "epoch": 2.871723712611779,
+      "grad_norm": 0.09423034638166428,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 27939
+    },
+    {
+      "epoch": 2.872032069071847,
+      "grad_norm": 0.05001804232597351,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 27942
+    },
+    {
+      "epoch": 2.872340425531915,
+      "grad_norm": 0.04487982019782066,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 27945
+    },
+    {
+      "epoch": 2.8726487819919826,
+      "grad_norm": 0.033674515783786774,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 27948
+    },
+    {
+      "epoch": 2.8729571384520507,
+      "grad_norm": 0.040895912796258926,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 27951
+    },
+    {
+      "epoch": 2.8732654949121184,
+      "grad_norm": 0.03655741363763809,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 27954
+    },
+    {
+      "epoch": 2.8735738513721865,
+      "grad_norm": 0.06042269244790077,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 27957
+    },
+    {
+      "epoch": 2.873882207832254,
+      "grad_norm": 0.0660943016409874,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 27960
+    },
+    {
+      "epoch": 2.874190564292322,
+      "grad_norm": 0.0681222453713417,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 27963
+    },
+    {
+      "epoch": 2.87449892075239,
+      "grad_norm": 0.05666188523173332,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 27966
+    },
+    {
+      "epoch": 2.8748072772124575,
+      "grad_norm": 0.09911254048347473,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 27969
+    },
+    {
+      "epoch": 2.875115633672525,
+      "grad_norm": 0.0730089545249939,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 27972
+    },
+    {
+      "epoch": 2.8754239901325933,
+      "grad_norm": 0.05434833839535713,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 27975
+    },
+    {
+      "epoch": 2.875732346592661,
+      "grad_norm": 0.07539419084787369,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 27978
+    },
+    {
+      "epoch": 2.876040703052729,
+      "grad_norm": 0.04300360381603241,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 27981
+    },
+    {
+      "epoch": 2.8763490595127967,
+      "grad_norm": 0.1077357679605484,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 27984
+    },
+    {
+      "epoch": 2.876657415972865,
+      "grad_norm": 0.08229992538690567,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 27987
+    },
+    {
+      "epoch": 2.8769657724329325,
+      "grad_norm": 0.12404177337884903,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 27990
+    },
+    {
+      "epoch": 2.877274128893,
+      "grad_norm": 0.06843412667512894,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 27993
+    },
+    {
+      "epoch": 2.8775824853530683,
+      "grad_norm": 0.09625189751386642,
+      "learning_rate": 0.01,
+      "loss": 2.0435,
+      "step": 27996
+    },
+    {
+      "epoch": 2.877890841813136,
+      "grad_norm": 0.04479416087269783,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 27999
+    },
+    {
+      "epoch": 2.8781991982732036,
+      "grad_norm": 0.04933414235711098,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 28002
+    },
+    {
+      "epoch": 2.8785075547332717,
+      "grad_norm": 0.05558469891548157,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 28005
+    },
+    {
+      "epoch": 2.8788159111933393,
+      "grad_norm": 0.04813915863633156,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 28008
+    },
+    {
+      "epoch": 2.8791242676534075,
+      "grad_norm": 0.06402178108692169,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 28011
+    },
+    {
+      "epoch": 2.879432624113475,
+      "grad_norm": 0.03481597453355789,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 28014
+    },
+    {
+      "epoch": 2.879740980573543,
+      "grad_norm": 0.03682602196931839,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 28017
+    },
+    {
+      "epoch": 2.880049337033611,
+      "grad_norm": 0.11671306192874908,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 28020
+    },
+    {
+      "epoch": 2.8803576934936785,
+      "grad_norm": 0.051900725811719894,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 28023
+    },
+    {
+      "epoch": 2.8806660499537466,
+      "grad_norm": 0.034372419118881226,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 28026
+    },
+    {
+      "epoch": 2.8809744064138143,
+      "grad_norm": 0.05165507644414902,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 28029
+    },
+    {
+      "epoch": 2.881282762873882,
+      "grad_norm": 0.046683549880981445,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 28032
+    },
+    {
+      "epoch": 2.88159111933395,
+      "grad_norm": 0.04153186455368996,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 28035
+    },
+    {
+      "epoch": 2.8818994757940177,
+      "grad_norm": 0.07401791960000992,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 28038
+    },
+    {
+      "epoch": 2.882207832254086,
+      "grad_norm": 0.08125065267086029,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 28041
+    },
+    {
+      "epoch": 2.8825161887141535,
+      "grad_norm": 0.076799176633358,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 28044
+    },
+    {
+      "epoch": 2.8828245451742216,
+      "grad_norm": 0.09814302623271942,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 28047
+    },
+    {
+      "epoch": 2.8831329016342893,
+      "grad_norm": 0.06998278200626373,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 28050
+    },
+    {
+      "epoch": 2.883441258094357,
+      "grad_norm": 0.057476770132780075,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 28053
+    },
+    {
+      "epoch": 2.883749614554425,
+      "grad_norm": 0.037239111959934235,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 28056
+    },
+    {
+      "epoch": 2.8840579710144927,
+      "grad_norm": 0.08055179566144943,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 28059
+    },
+    {
+      "epoch": 2.8843663274745603,
+      "grad_norm": 0.07050419598817825,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 28062
+    },
+    {
+      "epoch": 2.8846746839346284,
+      "grad_norm": 0.05187792330980301,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 28065
+    },
+    {
+      "epoch": 2.884983040394696,
+      "grad_norm": 0.03746993839740753,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 28068
+    },
+    {
+      "epoch": 2.885291396854764,
+      "grad_norm": 0.09767074137926102,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 28071
+    },
+    {
+      "epoch": 2.885599753314832,
+      "grad_norm": 0.08471439778804779,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 28074
+    },
+    {
+      "epoch": 2.8859081097749,
+      "grad_norm": 0.061311863362789154,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 28077
+    },
+    {
+      "epoch": 2.8862164662349676,
+      "grad_norm": 0.06402043253183365,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 28080
+    },
+    {
+      "epoch": 2.8865248226950353,
+      "grad_norm": 0.06756860762834549,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 28083
+    },
+    {
+      "epoch": 2.8868331791551034,
+      "grad_norm": 0.07851870357990265,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 28086
+    },
+    {
+      "epoch": 2.887141535615171,
+      "grad_norm": 0.08102571219205856,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 28089
+    },
+    {
+      "epoch": 2.8874498920752387,
+      "grad_norm": 0.04745829850435257,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 28092
+    },
+    {
+      "epoch": 2.887758248535307,
+      "grad_norm": 0.04496223106980324,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 28095
+    },
+    {
+      "epoch": 2.8880666049953745,
+      "grad_norm": 0.09044700860977173,
+      "learning_rate": 0.01,
+      "loss": 2.0354,
+      "step": 28098
+    },
+    {
+      "epoch": 2.8883749614554426,
+      "grad_norm": 0.05591853708028793,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 28101
+    },
+    {
+      "epoch": 2.8886833179155103,
+      "grad_norm": 0.07982437312602997,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 28104
+    },
+    {
+      "epoch": 2.8889916743755784,
+      "grad_norm": 0.039345480501651764,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 28107
+    },
+    {
+      "epoch": 2.889300030835646,
+      "grad_norm": 0.046677373349666595,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 28110
+    },
+    {
+      "epoch": 2.8896083872957137,
+      "grad_norm": 0.060302551835775375,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 28113
+    },
+    {
+      "epoch": 2.889916743755782,
+      "grad_norm": 0.09422065317630768,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 28116
+    },
+    {
+      "epoch": 2.8902251002158494,
+      "grad_norm": 0.03713101148605347,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 28119
+    },
+    {
+      "epoch": 2.890533456675917,
+      "grad_norm": 0.062431566417217255,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 28122
+    },
+    {
+      "epoch": 2.890841813135985,
+      "grad_norm": 0.052230529487133026,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 28125
+    },
+    {
+      "epoch": 2.8911501695960533,
+      "grad_norm": 0.05514863133430481,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 28128
+    },
+    {
+      "epoch": 2.891458526056121,
+      "grad_norm": 0.04317644238471985,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 28131
+    },
+    {
+      "epoch": 2.8917668825161886,
+      "grad_norm": 0.05251277983188629,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 28134
+    },
+    {
+      "epoch": 2.8920752389762567,
+      "grad_norm": 0.10419400781393051,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 28137
+    },
+    {
+      "epoch": 2.8923835954363244,
+      "grad_norm": 0.08806058764457703,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 28140
+    },
+    {
+      "epoch": 2.892691951896392,
+      "grad_norm": 0.059875234961509705,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 28143
+    },
+    {
+      "epoch": 2.89300030835646,
+      "grad_norm": 0.036619991064071655,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 28146
+    },
+    {
+      "epoch": 2.893308664816528,
+      "grad_norm": 0.034454282373189926,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 28149
+    },
+    {
+      "epoch": 2.8936170212765955,
+      "grad_norm": 0.05209527164697647,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 28152
+    },
+    {
+      "epoch": 2.8939253777366636,
+      "grad_norm": 0.11549924314022064,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 28155
+    },
+    {
+      "epoch": 2.8942337341967317,
+      "grad_norm": 0.06887582689523697,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 28158
+    },
+    {
+      "epoch": 2.8945420906567993,
+      "grad_norm": 0.046488743275403976,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 28161
+    },
+    {
+      "epoch": 2.894850447116867,
+      "grad_norm": 0.05086890980601311,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 28164
+    },
+    {
+      "epoch": 2.895158803576935,
+      "grad_norm": 0.03719864413142204,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 28167
+    },
+    {
+      "epoch": 2.8954671600370028,
+      "grad_norm": 0.059325605630874634,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 28170
+    },
+    {
+      "epoch": 2.8957755164970704,
+      "grad_norm": 0.088272824883461,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 28173
+    },
+    {
+      "epoch": 2.8960838729571385,
+      "grad_norm": 0.0874638557434082,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 28176
+    },
+    {
+      "epoch": 2.896392229417206,
+      "grad_norm": 0.059453610330820084,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 28179
+    },
+    {
+      "epoch": 2.896700585877274,
+      "grad_norm": 0.05221414193511009,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 28182
+    },
+    {
+      "epoch": 2.897008942337342,
+      "grad_norm": 0.03308747336268425,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 28185
+    },
+    {
+      "epoch": 2.89731729879741,
+      "grad_norm": 0.07506965100765228,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 28188
+    },
+    {
+      "epoch": 2.8976256552574777,
+      "grad_norm": 0.06189137324690819,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 28191
+    },
+    {
+      "epoch": 2.8979340117175454,
+      "grad_norm": 0.07855790853500366,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 28194
+    },
+    {
+      "epoch": 2.8982423681776135,
+      "grad_norm": 0.08880770951509476,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 28197
+    },
+    {
+      "epoch": 2.898550724637681,
+      "grad_norm": 0.05860179290175438,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 28200
+    },
+    {
+      "epoch": 2.898859081097749,
+      "grad_norm": 0.06037219241261482,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 28203
+    },
+    {
+      "epoch": 2.899167437557817,
+      "grad_norm": 0.16910460591316223,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 28206
+    },
+    {
+      "epoch": 2.8994757940178846,
+      "grad_norm": 0.05912697687745094,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 28209
+    },
+    {
+      "epoch": 2.8997841504779522,
+      "grad_norm": 0.04245199263095856,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 28212
+    },
+    {
+      "epoch": 2.9000925069380203,
+      "grad_norm": 0.03307129442691803,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 28215
+    },
+    {
+      "epoch": 2.9004008633980884,
+      "grad_norm": 0.03456464782357216,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 28218
+    },
+    {
+      "epoch": 2.900709219858156,
+      "grad_norm": 0.04525744169950485,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 28221
+    },
+    {
+      "epoch": 2.9010175763182238,
+      "grad_norm": 0.046930085867643356,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 28224
+    },
+    {
+      "epoch": 2.901325932778292,
+      "grad_norm": 0.07545941323041916,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 28227
+    },
+    {
+      "epoch": 2.9016342892383595,
+      "grad_norm": 0.045901909470558167,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 28230
+    },
+    {
+      "epoch": 2.901942645698427,
+      "grad_norm": 0.06182632967829704,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 28233
+    },
+    {
+      "epoch": 2.9022510021584953,
+      "grad_norm": 0.1126553937792778,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 28236
+    },
+    {
+      "epoch": 2.902559358618563,
+      "grad_norm": 0.05637908726930618,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 28239
+    },
+    {
+      "epoch": 2.902867715078631,
+      "grad_norm": 0.03390496224164963,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 28242
+    },
+    {
+      "epoch": 2.9031760715386987,
+      "grad_norm": 0.1719658523797989,
+      "learning_rate": 0.01,
+      "loss": 2.0446,
+      "step": 28245
+    },
+    {
+      "epoch": 2.903484427998767,
+      "grad_norm": 0.054201435297727585,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 28248
+    },
+    {
+      "epoch": 2.9037927844588345,
+      "grad_norm": 0.05339137092232704,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 28251
+    },
+    {
+      "epoch": 2.904101140918902,
+      "grad_norm": 0.0490683875977993,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 28254
+    },
+    {
+      "epoch": 2.9044094973789703,
+      "grad_norm": 0.048101890832185745,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 28257
+    },
+    {
+      "epoch": 2.904717853839038,
+      "grad_norm": 0.034970205277204514,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 28260
+    },
+    {
+      "epoch": 2.9050262102991056,
+      "grad_norm": 0.05500125139951706,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 28263
+    },
+    {
+      "epoch": 2.9053345667591737,
+      "grad_norm": 0.0717696025967598,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 28266
+    },
+    {
+      "epoch": 2.9056429232192413,
+      "grad_norm": 0.08134905993938446,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 28269
+    },
+    {
+      "epoch": 2.9059512796793094,
+      "grad_norm": 0.13240204751491547,
+      "learning_rate": 0.01,
+      "loss": 2.0268,
+      "step": 28272
+    },
+    {
+      "epoch": 2.906259636139377,
+      "grad_norm": 0.04060851037502289,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 28275
+    },
+    {
+      "epoch": 2.906567992599445,
+      "grad_norm": 0.04560050368309021,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 28278
+    },
+    {
+      "epoch": 2.906876349059513,
+      "grad_norm": 0.051935113966464996,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 28281
+    },
+    {
+      "epoch": 2.9071847055195805,
+      "grad_norm": 0.04502344876527786,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 28284
+    },
+    {
+      "epoch": 2.9074930619796486,
+      "grad_norm": 0.04561367630958557,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 28287
+    },
+    {
+      "epoch": 2.9078014184397163,
+      "grad_norm": 0.05686529353260994,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 28290
+    },
+    {
+      "epoch": 2.908109774899784,
+      "grad_norm": 0.06489894539117813,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 28293
+    },
+    {
+      "epoch": 2.908418131359852,
+      "grad_norm": 0.08427495509386063,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 28296
+    },
+    {
+      "epoch": 2.9087264878199197,
+      "grad_norm": 0.08298421651124954,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 28299
+    },
+    {
+      "epoch": 2.909034844279988,
+      "grad_norm": 0.035743072628974915,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 28302
+    },
+    {
+      "epoch": 2.9093432007400555,
+      "grad_norm": 0.11177317798137665,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 28305
+    },
+    {
+      "epoch": 2.9096515572001236,
+      "grad_norm": 0.10962017625570297,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 28308
+    },
+    {
+      "epoch": 2.9099599136601912,
+      "grad_norm": 0.033734966069459915,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 28311
+    },
+    {
+      "epoch": 2.910268270120259,
+      "grad_norm": 0.050082772970199585,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 28314
+    },
+    {
+      "epoch": 2.910576626580327,
+      "grad_norm": 0.04048197343945503,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 28317
+    },
+    {
+      "epoch": 2.9108849830403947,
+      "grad_norm": 0.039657291024923325,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 28320
+    },
+    {
+      "epoch": 2.9111933395004623,
+      "grad_norm": 0.0639658197760582,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 28323
+    },
+    {
+      "epoch": 2.9115016959605304,
+      "grad_norm": 0.09306667000055313,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 28326
+    },
+    {
+      "epoch": 2.911810052420598,
+      "grad_norm": 0.06027163937687874,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 28329
+    },
+    {
+      "epoch": 2.912118408880666,
+      "grad_norm": 0.08570882678031921,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 28332
+    },
+    {
+      "epoch": 2.912426765340734,
+      "grad_norm": 0.07936012744903564,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 28335
+    },
+    {
+      "epoch": 2.912735121800802,
+      "grad_norm": 0.06479290872812271,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 28338
+    },
+    {
+      "epoch": 2.9130434782608696,
+      "grad_norm": 0.06973425298929214,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 28341
+    },
+    {
+      "epoch": 2.9133518347209373,
+      "grad_norm": 0.05557083338499069,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 28344
+    },
+    {
+      "epoch": 2.9136601911810054,
+      "grad_norm": 0.04901493340730667,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 28347
+    },
+    {
+      "epoch": 2.913968547641073,
+      "grad_norm": 0.1036488264799118,
+      "learning_rate": 0.01,
+      "loss": 2.0486,
+      "step": 28350
+    },
+    {
+      "epoch": 2.9142769041011407,
+      "grad_norm": 0.04318247362971306,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 28353
+    },
+    {
+      "epoch": 2.914585260561209,
+      "grad_norm": 0.0404491052031517,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 28356
+    },
+    {
+      "epoch": 2.9148936170212765,
+      "grad_norm": 0.03727111592888832,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 28359
+    },
+    {
+      "epoch": 2.9152019734813446,
+      "grad_norm": 0.04736052453517914,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 28362
+    },
+    {
+      "epoch": 2.9155103299414122,
+      "grad_norm": 0.13712109625339508,
+      "learning_rate": 0.01,
+      "loss": 2.0434,
+      "step": 28365
+    },
+    {
+      "epoch": 2.9158186864014803,
+      "grad_norm": 0.046428218483924866,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 28368
+    },
+    {
+      "epoch": 2.916127042861548,
+      "grad_norm": 0.036403872072696686,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 28371
+    },
+    {
+      "epoch": 2.9164353993216157,
+      "grad_norm": 0.04359270632266998,
+      "learning_rate": 0.01,
+      "loss": 2.0386,
+      "step": 28374
+    },
+    {
+      "epoch": 2.9167437557816838,
+      "grad_norm": 0.06155150756239891,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 28377
+    },
+    {
+      "epoch": 2.9170521122417514,
+      "grad_norm": 0.035659730434417725,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 28380
+    },
+    {
+      "epoch": 2.917360468701819,
+      "grad_norm": 0.09453250467777252,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 28383
+    },
+    {
+      "epoch": 2.917668825161887,
+      "grad_norm": 0.062337253242731094,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 28386
+    },
+    {
+      "epoch": 2.917977181621955,
+      "grad_norm": 0.05488646402955055,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 28389
+    },
+    {
+      "epoch": 2.918285538082023,
+      "grad_norm": 0.07092556357383728,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 28392
+    },
+    {
+      "epoch": 2.9185938945420906,
+      "grad_norm": 0.10335639864206314,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 28395
+    },
+    {
+      "epoch": 2.9189022510021587,
+      "grad_norm": 0.1187339797616005,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 28398
+    },
+    {
+      "epoch": 2.9192106074622264,
+      "grad_norm": 0.09254027903079987,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 28401
+    },
+    {
+      "epoch": 2.919518963922294,
+      "grad_norm": 0.04120078682899475,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 28404
+    },
+    {
+      "epoch": 2.919827320382362,
+      "grad_norm": 0.04366682097315788,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 28407
+    },
+    {
+      "epoch": 2.92013567684243,
+      "grad_norm": 0.03700088709592819,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 28410
+    },
+    {
+      "epoch": 2.9204440333024975,
+      "grad_norm": 0.03409232571721077,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 28413
+    },
+    {
+      "epoch": 2.9207523897625656,
+      "grad_norm": 0.04503092169761658,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 28416
+    },
+    {
+      "epoch": 2.9210607462226332,
+      "grad_norm": 0.04988139122724533,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 28419
+    },
+    {
+      "epoch": 2.9213691026827013,
+      "grad_norm": 0.05121064558625221,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 28422
+    },
+    {
+      "epoch": 2.921677459142769,
+      "grad_norm": 0.03916258364915848,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 28425
+    },
+    {
+      "epoch": 2.921985815602837,
+      "grad_norm": 0.1349031776189804,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 28428
+    },
+    {
+      "epoch": 2.9222941720629048,
+      "grad_norm": 0.12042077630758286,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 28431
+    },
+    {
+      "epoch": 2.9226025285229724,
+      "grad_norm": 0.08157286792993546,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 28434
+    },
+    {
+      "epoch": 2.9229108849830405,
+      "grad_norm": 0.07750184834003448,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 28437
+    },
+    {
+      "epoch": 2.923219241443108,
+      "grad_norm": 0.07270103693008423,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 28440
+    },
+    {
+      "epoch": 2.923527597903176,
+      "grad_norm": 0.05140276625752449,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 28443
+    },
+    {
+      "epoch": 2.923835954363244,
+      "grad_norm": 0.07426592707633972,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 28446
+    },
+    {
+      "epoch": 2.9241443108233116,
+      "grad_norm": 0.05136018618941307,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 28449
+    },
+    {
+      "epoch": 2.9244526672833797,
+      "grad_norm": 0.03354816511273384,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 28452
+    },
+    {
+      "epoch": 2.9247610237434474,
+      "grad_norm": 0.03671132028102875,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 28455
+    },
+    {
+      "epoch": 2.9250693802035155,
+      "grad_norm": 0.05378049612045288,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 28458
+    },
+    {
+      "epoch": 2.925377736663583,
+      "grad_norm": 0.04772641509771347,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 28461
+    },
+    {
+      "epoch": 2.925686093123651,
+      "grad_norm": 0.05942932888865471,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 28464
+    },
+    {
+      "epoch": 2.925994449583719,
+      "grad_norm": 0.059020113199949265,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 28467
+    },
+    {
+      "epoch": 2.9263028060437866,
+      "grad_norm": 0.19283726811408997,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 28470
+    },
+    {
+      "epoch": 2.9266111625038542,
+      "grad_norm": 0.09231256693601608,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 28473
+    },
+    {
+      "epoch": 2.9269195189639223,
+      "grad_norm": 0.058674681931734085,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 28476
+    },
+    {
+      "epoch": 2.92722787542399,
+      "grad_norm": 0.05014891177415848,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 28479
+    },
+    {
+      "epoch": 2.927536231884058,
+      "grad_norm": 0.060597196221351624,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 28482
+    },
+    {
+      "epoch": 2.9278445883441258,
+      "grad_norm": 0.08379890769720078,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 28485
+    },
+    {
+      "epoch": 2.928152944804194,
+      "grad_norm": 0.04832527041435242,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 28488
+    },
+    {
+      "epoch": 2.9284613012642615,
+      "grad_norm": 0.06709366291761398,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 28491
+    },
+    {
+      "epoch": 2.928769657724329,
+      "grad_norm": 0.044588811695575714,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 28494
+    },
+    {
+      "epoch": 2.9290780141843973,
+      "grad_norm": 0.05735667049884796,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 28497
+    },
+    {
+      "epoch": 2.929386370644465,
+      "grad_norm": 0.041256386786699295,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 28500
+    },
+    {
+      "epoch": 2.9296947271045326,
+      "grad_norm": 0.03460060432553291,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 28503
+    },
+    {
+      "epoch": 2.9300030835646007,
+      "grad_norm": 0.05080440267920494,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 28506
+    },
+    {
+      "epoch": 2.9303114400246684,
+      "grad_norm": 0.1281091421842575,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 28509
+    },
+    {
+      "epoch": 2.9306197964847365,
+      "grad_norm": 0.05667036399245262,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 28512
+    },
+    {
+      "epoch": 2.930928152944804,
+      "grad_norm": 0.10026687383651733,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 28515
+    },
+    {
+      "epoch": 2.9312365094048722,
+      "grad_norm": 0.04924483224749565,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 28518
+    },
+    {
+      "epoch": 2.93154486586494,
+      "grad_norm": 0.04905860871076584,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 28521
+    },
+    {
+      "epoch": 2.9318532223250076,
+      "grad_norm": 0.0525185652077198,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 28524
+    },
+    {
+      "epoch": 2.9321615787850757,
+      "grad_norm": 0.04196888953447342,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 28527
+    },
+    {
+      "epoch": 2.9324699352451433,
+      "grad_norm": 0.04660086706280708,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 28530
+    },
+    {
+      "epoch": 2.932778291705211,
+      "grad_norm": 0.05741078406572342,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 28533
+    },
+    {
+      "epoch": 2.933086648165279,
+      "grad_norm": 0.052953120321035385,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 28536
+    },
+    {
+      "epoch": 2.9333950046253467,
+      "grad_norm": 0.04325610026717186,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 28539
+    },
+    {
+      "epoch": 2.933703361085415,
+      "grad_norm": 0.04387751594185829,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 28542
+    },
+    {
+      "epoch": 2.9340117175454825,
+      "grad_norm": 0.03476174548268318,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 28545
+    },
+    {
+      "epoch": 2.9343200740055506,
+      "grad_norm": 0.03839149698615074,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 28548
+    },
+    {
+      "epoch": 2.9346284304656183,
+      "grad_norm": 0.0841054692864418,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 28551
+    },
+    {
+      "epoch": 2.934936786925686,
+      "grad_norm": 0.10850472748279572,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 28554
+    },
+    {
+      "epoch": 2.935245143385754,
+      "grad_norm": 0.12255658954381943,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 28557
+    },
+    {
+      "epoch": 2.9355534998458217,
+      "grad_norm": 0.05664276331663132,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 28560
+    },
+    {
+      "epoch": 2.9358618563058894,
+      "grad_norm": 0.11724577099084854,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 28563
+    },
+    {
+      "epoch": 2.9361702127659575,
+      "grad_norm": 0.06823346763849258,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 28566
+    },
+    {
+      "epoch": 2.936478569226025,
+      "grad_norm": 0.035576723515987396,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 28569
+    },
+    {
+      "epoch": 2.9367869256860932,
+      "grad_norm": 0.03186320886015892,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 28572
+    },
+    {
+      "epoch": 2.937095282146161,
+      "grad_norm": 0.03750181198120117,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 28575
+    },
+    {
+      "epoch": 2.937403638606229,
+      "grad_norm": 0.05272996798157692,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 28578
+    },
+    {
+      "epoch": 2.9377119950662967,
+      "grad_norm": 0.033289846032857895,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 28581
+    },
+    {
+      "epoch": 2.9380203515263643,
+      "grad_norm": 0.11037155240774155,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 28584
+    },
+    {
+      "epoch": 2.9383287079864324,
+      "grad_norm": 0.10620691627264023,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 28587
+    },
+    {
+      "epoch": 2.9386370644465,
+      "grad_norm": 0.052820686250925064,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 28590
+    },
+    {
+      "epoch": 2.9389454209065677,
+      "grad_norm": 0.057419613003730774,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 28593
+    },
+    {
+      "epoch": 2.939253777366636,
+      "grad_norm": 0.04066954553127289,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 28596
+    },
+    {
+      "epoch": 2.9395621338267035,
+      "grad_norm": 0.04638439416885376,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 28599
+    },
+    {
+      "epoch": 2.9398704902867716,
+      "grad_norm": 0.11629427969455719,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 28602
+    },
+    {
+      "epoch": 2.9401788467468393,
+      "grad_norm": 0.05921720713376999,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 28605
+    },
+    {
+      "epoch": 2.9404872032069074,
+      "grad_norm": 0.05282594636082649,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 28608
+    },
+    {
+      "epoch": 2.940795559666975,
+      "grad_norm": 0.04680659994482994,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 28611
+    },
+    {
+      "epoch": 2.9411039161270427,
+      "grad_norm": 0.11889224499464035,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 28614
+    },
+    {
+      "epoch": 2.941412272587111,
+      "grad_norm": 0.03500881418585777,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 28617
+    },
+    {
+      "epoch": 2.9417206290471785,
+      "grad_norm": 0.08639416098594666,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 28620
+    },
+    {
+      "epoch": 2.942028985507246,
+      "grad_norm": 0.03335639461874962,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 28623
+    },
+    {
+      "epoch": 2.9423373419673142,
+      "grad_norm": 0.045041028410196304,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 28626
+    },
+    {
+      "epoch": 2.9426456984273823,
+      "grad_norm": 0.047020211815834045,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 28629
+    },
+    {
+      "epoch": 2.94295405488745,
+      "grad_norm": 0.059858404099941254,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 28632
+    },
+    {
+      "epoch": 2.9432624113475176,
+      "grad_norm": 0.053345970809459686,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 28635
+    },
+    {
+      "epoch": 2.9435707678075858,
+      "grad_norm": 0.05528027564287186,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 28638
+    },
+    {
+      "epoch": 2.9438791242676534,
+      "grad_norm": 0.09895586222410202,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 28641
+    },
+    {
+      "epoch": 2.944187480727721,
+      "grad_norm": 0.11827319115400314,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 28644
+    },
+    {
+      "epoch": 2.944495837187789,
+      "grad_norm": 0.05013230815529823,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 28647
+    },
+    {
+      "epoch": 2.944804193647857,
+      "grad_norm": 0.0422658771276474,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 28650
+    },
+    {
+      "epoch": 2.9451125501079245,
+      "grad_norm": 0.04022778570652008,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 28653
+    },
+    {
+      "epoch": 2.9454209065679926,
+      "grad_norm": 0.043504439294338226,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 28656
+    },
+    {
+      "epoch": 2.9457292630280607,
+      "grad_norm": 0.06587618589401245,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 28659
+    },
+    {
+      "epoch": 2.9460376194881284,
+      "grad_norm": 0.10226722061634064,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 28662
+    },
+    {
+      "epoch": 2.946345975948196,
+      "grad_norm": 0.047040197998285294,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 28665
+    },
+    {
+      "epoch": 2.946654332408264,
+      "grad_norm": 0.0919933021068573,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 28668
+    },
+    {
+      "epoch": 2.946962688868332,
+      "grad_norm": 0.05978929251432419,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 28671
+    },
+    {
+      "epoch": 2.9472710453283995,
+      "grad_norm": 0.05814214050769806,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 28674
+    },
+    {
+      "epoch": 2.9475794017884676,
+      "grad_norm": 0.09039480984210968,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 28677
+    },
+    {
+      "epoch": 2.947887758248535,
+      "grad_norm": 0.1607222557067871,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 28680
+    },
+    {
+      "epoch": 2.948196114708603,
+      "grad_norm": 0.04198214039206505,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 28683
+    },
+    {
+      "epoch": 2.948504471168671,
+      "grad_norm": 0.06442588567733765,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 28686
+    },
+    {
+      "epoch": 2.948812827628739,
+      "grad_norm": 0.11082901805639267,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 28689
+    },
+    {
+      "epoch": 2.9491211840888067,
+      "grad_norm": 0.03669416159391403,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 28692
+    },
+    {
+      "epoch": 2.9494295405488744,
+      "grad_norm": 0.06837104260921478,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 28695
+    },
+    {
+      "epoch": 2.9497378970089425,
+      "grad_norm": 0.0466892383992672,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 28698
+    },
+    {
+      "epoch": 2.95004625346901,
+      "grad_norm": 0.06962305307388306,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 28701
+    },
+    {
+      "epoch": 2.950354609929078,
+      "grad_norm": 0.06136500835418701,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 28704
+    },
+    {
+      "epoch": 2.950662966389146,
+      "grad_norm": 0.050067611038684845,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 28707
+    },
+    {
+      "epoch": 2.9509713228492136,
+      "grad_norm": 0.0527566559612751,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 28710
+    },
+    {
+      "epoch": 2.9512796793092813,
+      "grad_norm": 0.12275776267051697,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 28713
+    },
+    {
+      "epoch": 2.9515880357693494,
+      "grad_norm": 0.052756138145923615,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 28716
+    },
+    {
+      "epoch": 2.9518963922294175,
+      "grad_norm": 0.038834694772958755,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 28719
+    },
+    {
+      "epoch": 2.952204748689485,
+      "grad_norm": 0.07736565917730331,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 28722
+    },
+    {
+      "epoch": 2.952513105149553,
+      "grad_norm": 0.10177972167730331,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 28725
+    },
+    {
+      "epoch": 2.952821461609621,
+      "grad_norm": 0.082985520362854,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 28728
+    },
+    {
+      "epoch": 2.9531298180696886,
+      "grad_norm": 0.05385474115610123,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 28731
+    },
+    {
+      "epoch": 2.953438174529756,
+      "grad_norm": 0.0803777202963829,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 28734
+    },
+    {
+      "epoch": 2.9537465309898243,
+      "grad_norm": 0.06875913590192795,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 28737
+    },
+    {
+      "epoch": 2.954054887449892,
+      "grad_norm": 0.09149770438671112,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 28740
+    },
+    {
+      "epoch": 2.95436324390996,
+      "grad_norm": 0.1182807981967926,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 28743
+    },
+    {
+      "epoch": 2.9546716003700277,
+      "grad_norm": 0.044717706739902496,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 28746
+    },
+    {
+      "epoch": 2.954979956830096,
+      "grad_norm": 0.04887622967362404,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 28749
+    },
+    {
+      "epoch": 2.9552883132901635,
+      "grad_norm": 0.041407834738492966,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 28752
+    },
+    {
+      "epoch": 2.955596669750231,
+      "grad_norm": 0.05831639841198921,
+      "learning_rate": 0.01,
+      "loss": 2.0485,
+      "step": 28755
+    },
+    {
+      "epoch": 2.9559050262102993,
+      "grad_norm": 0.08279699832201004,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 28758
+    },
+    {
+      "epoch": 2.956213382670367,
+      "grad_norm": 0.08988698571920395,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 28761
+    },
+    {
+      "epoch": 2.9565217391304346,
+      "grad_norm": 0.0645131915807724,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 28764
+    },
+    {
+      "epoch": 2.9568300955905027,
+      "grad_norm": 0.08262491226196289,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 28767
+    },
+    {
+      "epoch": 2.9571384520505704,
+      "grad_norm": 0.03773394227027893,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 28770
+    },
+    {
+      "epoch": 2.9574468085106385,
+      "grad_norm": 0.08012068271636963,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 28773
+    },
+    {
+      "epoch": 2.957755164970706,
+      "grad_norm": 0.04637681320309639,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 28776
+    },
+    {
+      "epoch": 2.9580635214307742,
+      "grad_norm": 0.031988177448511124,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 28779
+    },
+    {
+      "epoch": 2.958371877890842,
+      "grad_norm": 0.08325552940368652,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 28782
+    },
+    {
+      "epoch": 2.9586802343509095,
+      "grad_norm": 0.06131797283887863,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 28785
+    },
+    {
+      "epoch": 2.9589885908109776,
+      "grad_norm": 0.04146185144782066,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 28788
+    },
+    {
+      "epoch": 2.9592969472710453,
+      "grad_norm": 0.04004296660423279,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 28791
+    },
+    {
+      "epoch": 2.959605303731113,
+      "grad_norm": 0.07771418243646622,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 28794
+    },
+    {
+      "epoch": 2.959913660191181,
+      "grad_norm": 0.048570699989795685,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 28797
+    },
+    {
+      "epoch": 2.9602220166512487,
+      "grad_norm": 0.0664374902844429,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 28800
+    },
+    {
+      "epoch": 2.960530373111317,
+      "grad_norm": 0.09744110703468323,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 28803
+    },
+    {
+      "epoch": 2.9608387295713845,
+      "grad_norm": 0.03430997580289841,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 28806
+    },
+    {
+      "epoch": 2.9611470860314526,
+      "grad_norm": 0.036999545991420746,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 28809
+    },
+    {
+      "epoch": 2.9614554424915203,
+      "grad_norm": 0.05290836840867996,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 28812
+    },
+    {
+      "epoch": 2.961763798951588,
+      "grad_norm": 0.061555683612823486,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 28815
+    },
+    {
+      "epoch": 2.962072155411656,
+      "grad_norm": 0.06276807934045792,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 28818
+    },
+    {
+      "epoch": 2.9623805118717237,
+      "grad_norm": 0.05685890465974808,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 28821
+    },
+    {
+      "epoch": 2.9626888683317913,
+      "grad_norm": 0.09975235909223557,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 28824
+    },
+    {
+      "epoch": 2.9629972247918595,
+      "grad_norm": 0.04205232113599777,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 28827
+    },
+    {
+      "epoch": 2.963305581251927,
+      "grad_norm": 0.07952512800693512,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 28830
+    },
+    {
+      "epoch": 2.963613937711995,
+      "grad_norm": 0.04118579626083374,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 28833
+    },
+    {
+      "epoch": 2.963922294172063,
+      "grad_norm": 0.10875571519136429,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 28836
+    },
+    {
+      "epoch": 2.964230650632131,
+      "grad_norm": 0.16663865745067596,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 28839
+    },
+    {
+      "epoch": 2.9645390070921986,
+      "grad_norm": 0.08370403200387955,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 28842
+    },
+    {
+      "epoch": 2.9648473635522663,
+      "grad_norm": 0.07209211587905884,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 28845
+    },
+    {
+      "epoch": 2.9651557200123344,
+      "grad_norm": 0.05848775431513786,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 28848
+    },
+    {
+      "epoch": 2.965464076472402,
+      "grad_norm": 0.02791527472436428,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 28851
+    },
+    {
+      "epoch": 2.9657724329324697,
+      "grad_norm": 0.03520442545413971,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 28854
+    },
+    {
+      "epoch": 2.966080789392538,
+      "grad_norm": 0.03414986655116081,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 28857
+    },
+    {
+      "epoch": 2.9663891458526055,
+      "grad_norm": 0.0448157899081707,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 28860
+    },
+    {
+      "epoch": 2.9666975023126736,
+      "grad_norm": 0.09282497316598892,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 28863
+    },
+    {
+      "epoch": 2.9670058587727413,
+      "grad_norm": 0.05999337136745453,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 28866
+    },
+    {
+      "epoch": 2.9673142152328094,
+      "grad_norm": 0.038877278566360474,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 28869
+    },
+    {
+      "epoch": 2.967622571692877,
+      "grad_norm": 0.04760310798883438,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 28872
+    },
+    {
+      "epoch": 2.9679309281529447,
+      "grad_norm": 0.049317944794893265,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 28875
+    },
+    {
+      "epoch": 2.968239284613013,
+      "grad_norm": 0.038830939680337906,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 28878
+    },
+    {
+      "epoch": 2.9685476410730804,
+      "grad_norm": 0.18675923347473145,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 28881
+    },
+    {
+      "epoch": 2.968855997533148,
+      "grad_norm": 0.04269779101014137,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 28884
+    },
+    {
+      "epoch": 2.969164353993216,
+      "grad_norm": 0.04740308225154877,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 28887
+    },
+    {
+      "epoch": 2.969472710453284,
+      "grad_norm": 0.058595623821020126,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 28890
+    },
+    {
+      "epoch": 2.969781066913352,
+      "grad_norm": 0.06136814132332802,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 28893
+    },
+    {
+      "epoch": 2.9700894233734196,
+      "grad_norm": 0.059679534286260605,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 28896
+    },
+    {
+      "epoch": 2.9703977798334877,
+      "grad_norm": 0.044182952493429184,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 28899
+    },
+    {
+      "epoch": 2.9707061362935554,
+      "grad_norm": 0.04152587056159973,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 28902
+    },
+    {
+      "epoch": 2.971014492753623,
+      "grad_norm": 0.0424608550965786,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 28905
+    },
+    {
+      "epoch": 2.971322849213691,
+      "grad_norm": 0.040096018463373184,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 28908
+    },
+    {
+      "epoch": 2.971631205673759,
+      "grad_norm": 0.03288499265909195,
+      "learning_rate": 0.01,
+      "loss": 2.0416,
+      "step": 28911
+    },
+    {
+      "epoch": 2.9719395621338265,
+      "grad_norm": 0.05150043964385986,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 28914
+    },
+    {
+      "epoch": 2.9722479185938946,
+      "grad_norm": 0.05535360425710678,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 28917
+    },
+    {
+      "epoch": 2.9725562750539622,
+      "grad_norm": 0.09103085100650787,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 28920
+    },
+    {
+      "epoch": 2.9728646315140304,
+      "grad_norm": 0.060518983751535416,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 28923
+    },
+    {
+      "epoch": 2.973172987974098,
+      "grad_norm": 0.15921367704868317,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 28926
+    },
+    {
+      "epoch": 2.973481344434166,
+      "grad_norm": 0.04451125115156174,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 28929
+    },
+    {
+      "epoch": 2.973789700894234,
+      "grad_norm": 0.06099553778767586,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 28932
+    },
+    {
+      "epoch": 2.9740980573543014,
+      "grad_norm": 0.07864733040332794,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 28935
+    },
+    {
+      "epoch": 2.9744064138143695,
+      "grad_norm": 0.04279434680938721,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 28938
+    },
+    {
+      "epoch": 2.974714770274437,
+      "grad_norm": 0.06299113482236862,
+      "learning_rate": 0.01,
+      "loss": 2.0379,
+      "step": 28941
+    },
+    {
+      "epoch": 2.975023126734505,
+      "grad_norm": 0.04803795740008354,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 28944
+    },
+    {
+      "epoch": 2.975331483194573,
+      "grad_norm": 0.04646730050444603,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 28947
+    },
+    {
+      "epoch": 2.9756398396546406,
+      "grad_norm": 0.05352159962058067,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 28950
+    },
+    {
+      "epoch": 2.9759481961147087,
+      "grad_norm": 0.07281997799873352,
+      "learning_rate": 0.01,
+      "loss": 2.0328,
+      "step": 28953
+    },
+    {
+      "epoch": 2.9762565525747764,
+      "grad_norm": 0.050049133598804474,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 28956
+    },
+    {
+      "epoch": 2.9765649090348445,
+      "grad_norm": 0.039795007556676865,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 28959
+    },
+    {
+      "epoch": 2.976873265494912,
+      "grad_norm": 0.09085065871477127,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 28962
+    },
+    {
+      "epoch": 2.97718162195498,
+      "grad_norm": 0.1592961698770523,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 28965
+    },
+    {
+      "epoch": 2.977489978415048,
+      "grad_norm": 0.10092988610267639,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 28968
+    },
+    {
+      "epoch": 2.9777983348751156,
+      "grad_norm": 0.046051934361457825,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 28971
+    },
+    {
+      "epoch": 2.9781066913351832,
+      "grad_norm": 0.04267173632979393,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 28974
+    },
+    {
+      "epoch": 2.9784150477952513,
+      "grad_norm": 0.032134927809238434,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 28977
+    },
+    {
+      "epoch": 2.978723404255319,
+      "grad_norm": 0.055023569613695145,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 28980
+    },
+    {
+      "epoch": 2.979031760715387,
+      "grad_norm": 0.07177849858999252,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 28983
+    },
+    {
+      "epoch": 2.9793401171754548,
+      "grad_norm": 0.05890415981411934,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 28986
+    },
+    {
+      "epoch": 2.979648473635523,
+      "grad_norm": 0.04443821310997009,
+      "learning_rate": 0.01,
+      "loss": 2.042,
+      "step": 28989
+    },
+    {
+      "epoch": 2.9799568300955905,
+      "grad_norm": 0.04153164103627205,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 28992
+    },
+    {
+      "epoch": 2.980265186555658,
+      "grad_norm": 0.07844668626785278,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 28995
+    },
+    {
+      "epoch": 2.9805735430157263,
+      "grad_norm": 0.09027374535799026,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 28998
+    },
+    {
+      "epoch": 2.980881899475794,
+      "grad_norm": 0.04559837654232979,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 29001
+    },
+    {
+      "epoch": 2.9811902559358616,
+      "grad_norm": 0.03771069645881653,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 29004
+    },
+    {
+      "epoch": 2.9814986123959297,
+      "grad_norm": 0.09008541703224182,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 29007
+    },
+    {
+      "epoch": 2.9818069688559974,
+      "grad_norm": 0.060089852660894394,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 29010
+    },
+    {
+      "epoch": 2.9821153253160655,
+      "grad_norm": 0.11291385442018509,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 29013
+    },
+    {
+      "epoch": 2.982423681776133,
+      "grad_norm": 0.09720003604888916,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 29016
+    },
+    {
+      "epoch": 2.9827320382362013,
+      "grad_norm": 0.060459867119789124,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 29019
+    },
+    {
+      "epoch": 2.983040394696269,
+      "grad_norm": 0.07619535177946091,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 29022
+    },
+    {
+      "epoch": 2.9833487511563366,
+      "grad_norm": 0.05402089282870293,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 29025
+    },
+    {
+      "epoch": 2.9836571076164047,
+      "grad_norm": 0.05162883922457695,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 29028
+    },
+    {
+      "epoch": 2.9839654640764723,
+      "grad_norm": 0.037369467318058014,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 29031
+    },
+    {
+      "epoch": 2.98427382053654,
+      "grad_norm": 0.05483279377222061,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 29034
+    },
+    {
+      "epoch": 2.984582176996608,
+      "grad_norm": 0.050847407430410385,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 29037
+    },
+    {
+      "epoch": 2.9848905334566758,
+      "grad_norm": 0.07075628638267517,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 29040
+    },
+    {
+      "epoch": 2.985198889916744,
+      "grad_norm": 0.05580511689186096,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 29043
+    },
+    {
+      "epoch": 2.9855072463768115,
+      "grad_norm": 0.043689336627721786,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 29046
+    },
+    {
+      "epoch": 2.9858156028368796,
+      "grad_norm": 0.07577245682477951,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 29049
+    },
+    {
+      "epoch": 2.9861239592969473,
+      "grad_norm": 0.07917779684066772,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 29052
+    },
+    {
+      "epoch": 2.986432315757015,
+      "grad_norm": 0.07137954980134964,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 29055
+    },
+    {
+      "epoch": 2.986740672217083,
+      "grad_norm": 0.09875428676605225,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 29058
+    },
+    {
+      "epoch": 2.9870490286771507,
+      "grad_norm": 0.059274058789014816,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 29061
+    },
+    {
+      "epoch": 2.9873573851372184,
+      "grad_norm": 0.04235726222395897,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 29064
+    },
+    {
+      "epoch": 2.9876657415972865,
+      "grad_norm": 0.1677253097295761,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 29067
+    },
+    {
+      "epoch": 2.987974098057354,
+      "grad_norm": 0.05953408405184746,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 29070
+    },
+    {
+      "epoch": 2.9882824545174222,
+      "grad_norm": 0.031471315771341324,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 29073
+    },
+    {
+      "epoch": 2.98859081097749,
+      "grad_norm": 0.0490463487803936,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 29076
+    },
+    {
+      "epoch": 2.988899167437558,
+      "grad_norm": 0.0685456171631813,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 29079
+    },
+    {
+      "epoch": 2.9892075238976257,
+      "grad_norm": 0.04472583159804344,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 29082
+    },
+    {
+      "epoch": 2.9895158803576933,
+      "grad_norm": 0.03528788685798645,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 29085
+    },
+    {
+      "epoch": 2.9898242368177614,
+      "grad_norm": 0.04196497052907944,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 29088
+    },
+    {
+      "epoch": 2.990132593277829,
+      "grad_norm": 0.1347953975200653,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 29091
+    },
+    {
+      "epoch": 2.9904409497378968,
+      "grad_norm": 0.03483536094427109,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 29094
+    },
+    {
+      "epoch": 2.990749306197965,
+      "grad_norm": 0.04203316941857338,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 29097
+    },
+    {
+      "epoch": 2.9910576626580325,
+      "grad_norm": 0.039628706872463226,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 29100
+    },
+    {
+      "epoch": 2.9913660191181006,
+      "grad_norm": 0.04524881765246391,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 29103
+    },
+    {
+      "epoch": 2.9916743755781683,
+      "grad_norm": 0.0704786404967308,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 29106
+    },
+    {
+      "epoch": 2.9919827320382364,
+      "grad_norm": 0.06169109791517258,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 29109
+    },
+    {
+      "epoch": 2.992291088498304,
+      "grad_norm": 0.05929429456591606,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 29112
+    },
+    {
+      "epoch": 2.9925994449583717,
+      "grad_norm": 0.06046414375305176,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 29115
+    },
+    {
+      "epoch": 2.99290780141844,
+      "grad_norm": 0.03632686287164688,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 29118
+    },
+    {
+      "epoch": 2.9932161578785075,
+      "grad_norm": 0.0416223518550396,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 29121
+    },
+    {
+      "epoch": 2.993524514338575,
+      "grad_norm": 0.033993568271398544,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 29124
+    },
+    {
+      "epoch": 2.9938328707986432,
+      "grad_norm": 0.11413303017616272,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 29127
+    },
+    {
+      "epoch": 2.9941412272587113,
+      "grad_norm": 0.04512523114681244,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 29130
+    },
+    {
+      "epoch": 2.994449583718779,
+      "grad_norm": 0.08577805757522583,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 29133
+    },
+    {
+      "epoch": 2.9947579401788467,
+      "grad_norm": 0.09573300927877426,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 29136
+    },
+    {
+      "epoch": 2.9950662966389148,
+      "grad_norm": 0.04170147702097893,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 29139
+    },
+    {
+      "epoch": 2.9953746530989824,
+      "grad_norm": 0.04233024641871452,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 29142
+    },
+    {
+      "epoch": 2.99568300955905,
+      "grad_norm": 0.05406009405851364,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 29145
+    },
+    {
+      "epoch": 2.995991366019118,
+      "grad_norm": 0.037997808307409286,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 29148
+    },
+    {
+      "epoch": 2.996299722479186,
+      "grad_norm": 0.08010167628526688,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 29151
+    },
+    {
+      "epoch": 2.9966080789392535,
+      "grad_norm": 0.05372076854109764,
+      "learning_rate": 0.01,
+      "loss": 2.0406,
+      "step": 29154
+    },
+    {
+      "epoch": 2.9969164353993216,
+      "grad_norm": 0.04186830669641495,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 29157
+    },
+    {
+      "epoch": 2.9972247918593897,
+      "grad_norm": 0.0803917944431305,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 29160
+    },
+    {
+      "epoch": 2.9975331483194574,
+      "grad_norm": 0.06086145341396332,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 29163
+    },
+    {
+      "epoch": 2.997841504779525,
+      "grad_norm": 0.1197366788983345,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 29166
+    },
+    {
+      "epoch": 2.998149861239593,
+      "grad_norm": 0.09424477070569992,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 29169
+    },
+    {
+      "epoch": 2.998458217699661,
+      "grad_norm": 0.05025864019989967,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 29172
+    },
+    {
+      "epoch": 2.9987665741597285,
+      "grad_norm": 0.0840819776058197,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 29175
+    },
+    {
+      "epoch": 2.9990749306197966,
+      "grad_norm": 0.03415597230195999,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 29178
+    },
+    {
+      "epoch": 2.9993832870798642,
+      "grad_norm": 0.09066252410411835,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 29181
+    },
+    {
+      "epoch": 2.999691643539932,
+      "grad_norm": 0.05019732564687729,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 29184
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.09318925440311432,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 29187
+    },
+    {
+      "epoch": 2.999383477188656,
+      "grad_norm": 0.08568185567855835,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 29190
+    },
+    {
+      "epoch": 2.9996917385943282,
+      "grad_norm": 0.05479726567864418,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 29193
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.0812709629535675,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 29196
+    },
+    {
+      "epoch": 3.000308261405672,
+      "grad_norm": 0.06423875689506531,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 29199
+    },
+    {
+      "epoch": 3.000616522811344,
+      "grad_norm": 0.06900475919246674,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 29202
+    },
+    {
+      "epoch": 3.000924784217016,
+      "grad_norm": 0.09736461192369461,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 29205
+    },
+    {
+      "epoch": 3.001233045622688,
+      "grad_norm": 0.10431299358606339,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 29208
+    },
+    {
+      "epoch": 3.00154130702836,
+      "grad_norm": 0.06412489712238312,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 29211
+    },
+    {
+      "epoch": 3.0018495684340323,
+      "grad_norm": 0.07305043190717697,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 29214
+    },
+    {
+      "epoch": 3.002157829839704,
+      "grad_norm": 0.07680006325244904,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 29217
+    },
+    {
+      "epoch": 3.0024660912453762,
+      "grad_norm": 0.04416871443390846,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 29220
+    },
+    {
+      "epoch": 3.002774352651048,
+      "grad_norm": 0.05895330011844635,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 29223
+    },
+    {
+      "epoch": 3.00308261405672,
+      "grad_norm": 0.16763944923877716,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 29226
+    },
+    {
+      "epoch": 3.003390875462392,
+      "grad_norm": 0.04152580350637436,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 29229
+    },
+    {
+      "epoch": 3.003699136868064,
+      "grad_norm": 0.05650210753083229,
+      "learning_rate": 0.01,
+      "loss": 2.0373,
+      "step": 29232
+    },
+    {
+      "epoch": 3.0040073982737363,
+      "grad_norm": 0.10183783620595932,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 29235
+    },
+    {
+      "epoch": 3.004315659679408,
+      "grad_norm": 0.11545984447002411,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 29238
+    },
+    {
+      "epoch": 3.0046239210850803,
+      "grad_norm": 0.07768990844488144,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 29241
+    },
+    {
+      "epoch": 3.004932182490752,
+      "grad_norm": 0.06256531924009323,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 29244
+    },
+    {
+      "epoch": 3.005240443896424,
+      "grad_norm": 0.041665416210889816,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 29247
+    },
+    {
+      "epoch": 3.0055487053020964,
+      "grad_norm": 0.051996082067489624,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 29250
+    },
+    {
+      "epoch": 3.005856966707768,
+      "grad_norm": 0.032815441489219666,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 29253
+    },
+    {
+      "epoch": 3.0061652281134403,
+      "grad_norm": 0.060963522642850876,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 29256
+    },
+    {
+      "epoch": 3.006473489519112,
+      "grad_norm": 0.11818858236074448,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 29259
+    },
+    {
+      "epoch": 3.0067817509247843,
+      "grad_norm": 0.08653881400823593,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 29262
+    },
+    {
+      "epoch": 3.007090012330456,
+      "grad_norm": 0.09629751741886139,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 29265
+    },
+    {
+      "epoch": 3.0073982737361282,
+      "grad_norm": 0.04238956794142723,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 29268
+    },
+    {
+      "epoch": 3.0077065351418004,
+      "grad_norm": 0.040573202073574066,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 29271
+    },
+    {
+      "epoch": 3.008014796547472,
+      "grad_norm": 0.08061878383159637,
+      "learning_rate": 0.01,
+      "loss": 2.0562,
+      "step": 29274
+    },
+    {
+      "epoch": 3.0083230579531444,
+      "grad_norm": 0.06944199651479721,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 29277
+    },
+    {
+      "epoch": 3.008631319358816,
+      "grad_norm": 0.06629909574985504,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 29280
+    },
+    {
+      "epoch": 3.0089395807644883,
+      "grad_norm": 0.06654530018568039,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 29283
+    },
+    {
+      "epoch": 3.0092478421701605,
+      "grad_norm": 0.08806800842285156,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 29286
+    },
+    {
+      "epoch": 3.0095561035758323,
+      "grad_norm": 0.12367472797632217,
+      "learning_rate": 0.01,
+      "loss": 2.0382,
+      "step": 29289
+    },
+    {
+      "epoch": 3.0098643649815044,
+      "grad_norm": 0.052121471613645554,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 29292
+    },
+    {
+      "epoch": 3.010172626387176,
+      "grad_norm": 0.0409090481698513,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 29295
+    },
+    {
+      "epoch": 3.0104808877928484,
+      "grad_norm": 0.09956279397010803,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 29298
+    },
+    {
+      "epoch": 3.01078914919852,
+      "grad_norm": 0.04090806096792221,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 29301
+    },
+    {
+      "epoch": 3.0110974106041923,
+      "grad_norm": 0.06308458000421524,
+      "learning_rate": 0.01,
+      "loss": 2.0347,
+      "step": 29304
+    },
+    {
+      "epoch": 3.0114056720098645,
+      "grad_norm": 0.05140318349003792,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 29307
+    },
+    {
+      "epoch": 3.0117139334155363,
+      "grad_norm": 0.0383441224694252,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 29310
+    },
+    {
+      "epoch": 3.0120221948212085,
+      "grad_norm": 0.06803669035434723,
+      "learning_rate": 0.01,
+      "loss": 2.0442,
+      "step": 29313
+    },
+    {
+      "epoch": 3.0123304562268802,
+      "grad_norm": 0.042336028069257736,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 29316
+    },
+    {
+      "epoch": 3.0126387176325524,
+      "grad_norm": 0.052575305104255676,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 29319
+    },
+    {
+      "epoch": 3.0129469790382246,
+      "grad_norm": 0.04428831860423088,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 29322
+    },
+    {
+      "epoch": 3.0132552404438964,
+      "grad_norm": 0.03720409423112869,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 29325
+    },
+    {
+      "epoch": 3.0135635018495686,
+      "grad_norm": 0.03491399809718132,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 29328
+    },
+    {
+      "epoch": 3.0138717632552403,
+      "grad_norm": 0.11868512630462646,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 29331
+    },
+    {
+      "epoch": 3.0141800246609125,
+      "grad_norm": 0.05709204822778702,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 29334
+    },
+    {
+      "epoch": 3.0144882860665843,
+      "grad_norm": 0.09501231461763382,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 29337
+    },
+    {
+      "epoch": 3.0147965474722564,
+      "grad_norm": 0.04431547597050667,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 29340
+    },
+    {
+      "epoch": 3.0151048088779286,
+      "grad_norm": 0.07580556720495224,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 29343
+    },
+    {
+      "epoch": 3.0154130702836004,
+      "grad_norm": 0.05567536503076553,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 29346
+    },
+    {
+      "epoch": 3.0157213316892726,
+      "grad_norm": 0.03637940436601639,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 29349
+    },
+    {
+      "epoch": 3.0160295930949443,
+      "grad_norm": 0.07379139959812164,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 29352
+    },
+    {
+      "epoch": 3.0163378545006165,
+      "grad_norm": 0.08311998844146729,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 29355
+    },
+    {
+      "epoch": 3.0166461159062887,
+      "grad_norm": 0.11170487850904465,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 29358
+    },
+    {
+      "epoch": 3.0169543773119605,
+      "grad_norm": 0.055827848613262177,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 29361
+    },
+    {
+      "epoch": 3.0172626387176327,
+      "grad_norm": 0.06052641570568085,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 29364
+    },
+    {
+      "epoch": 3.0175709001233044,
+      "grad_norm": 0.03150554001331329,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 29367
+    },
+    {
+      "epoch": 3.0178791615289766,
+      "grad_norm": 0.06298622488975525,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 29370
+    },
+    {
+      "epoch": 3.0181874229346484,
+      "grad_norm": 0.09742710739374161,
+      "learning_rate": 0.01,
+      "loss": 2.0411,
+      "step": 29373
+    },
+    {
+      "epoch": 3.0184956843403206,
+      "grad_norm": 0.07219108939170837,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 29376
+    },
+    {
+      "epoch": 3.0188039457459928,
+      "grad_norm": 0.05558139458298683,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 29379
+    },
+    {
+      "epoch": 3.0191122071516645,
+      "grad_norm": 0.05763263627886772,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 29382
+    },
+    {
+      "epoch": 3.0194204685573367,
+      "grad_norm": 0.04832174628973007,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 29385
+    },
+    {
+      "epoch": 3.0197287299630085,
+      "grad_norm": 0.03823337331414223,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 29388
+    },
+    {
+      "epoch": 3.0200369913686806,
+      "grad_norm": 0.05253903195261955,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 29391
+    },
+    {
+      "epoch": 3.020345252774353,
+      "grad_norm": 0.10889255255460739,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 29394
+    },
+    {
+      "epoch": 3.0206535141800246,
+      "grad_norm": 0.04247021675109863,
+      "learning_rate": 0.01,
+      "loss": 2.0466,
+      "step": 29397
+    },
+    {
+      "epoch": 3.020961775585697,
+      "grad_norm": 0.09347319602966309,
+      "learning_rate": 0.01,
+      "loss": 2.037,
+      "step": 29400
+    },
+    {
+      "epoch": 3.0212700369913685,
+      "grad_norm": 0.05651739612221718,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 29403
+    },
+    {
+      "epoch": 3.0215782983970407,
+      "grad_norm": 0.0666181743144989,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 29406
+    },
+    {
+      "epoch": 3.021886559802713,
+      "grad_norm": 0.053186848759651184,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 29409
+    },
+    {
+      "epoch": 3.0221948212083847,
+      "grad_norm": 0.05201537534594536,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 29412
+    },
+    {
+      "epoch": 3.022503082614057,
+      "grad_norm": 0.02726483717560768,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 29415
+    },
+    {
+      "epoch": 3.0228113440197286,
+      "grad_norm": 0.10047302395105362,
+      "learning_rate": 0.01,
+      "loss": 2.0336,
+      "step": 29418
+    },
+    {
+      "epoch": 3.023119605425401,
+      "grad_norm": 0.05461571365594864,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 29421
+    },
+    {
+      "epoch": 3.0234278668310726,
+      "grad_norm": 0.08373844623565674,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 29424
+    },
+    {
+      "epoch": 3.0237361282367448,
+      "grad_norm": 0.045885663479566574,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 29427
+    },
+    {
+      "epoch": 3.024044389642417,
+      "grad_norm": 0.054790932685136795,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 29430
+    },
+    {
+      "epoch": 3.0243526510480887,
+      "grad_norm": 0.04917832091450691,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 29433
+    },
+    {
+      "epoch": 3.024660912453761,
+      "grad_norm": 0.09375031292438507,
+      "learning_rate": 0.01,
+      "loss": 2.0482,
+      "step": 29436
+    },
+    {
+      "epoch": 3.0249691738594326,
+      "grad_norm": 0.051234230399131775,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 29439
+    },
+    {
+      "epoch": 3.025277435265105,
+      "grad_norm": 0.057380858808755875,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 29442
+    },
+    {
+      "epoch": 3.025585696670777,
+      "grad_norm": 0.060605574399232864,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 29445
+    },
+    {
+      "epoch": 3.025893958076449,
+      "grad_norm": 0.09125442057847977,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 29448
+    },
+    {
+      "epoch": 3.026202219482121,
+      "grad_norm": 0.04081615433096886,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 29451
+    },
+    {
+      "epoch": 3.0265104808877927,
+      "grad_norm": 0.05347365140914917,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 29454
+    },
+    {
+      "epoch": 3.026818742293465,
+      "grad_norm": 0.04458871856331825,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 29457
+    },
+    {
+      "epoch": 3.0271270036991367,
+      "grad_norm": 0.04996645078063011,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 29460
+    },
+    {
+      "epoch": 3.027435265104809,
+      "grad_norm": 0.051296427845954895,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 29463
+    },
+    {
+      "epoch": 3.027743526510481,
+      "grad_norm": 0.05448664352297783,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 29466
+    },
+    {
+      "epoch": 3.028051787916153,
+      "grad_norm": 0.0587022639811039,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 29469
+    },
+    {
+      "epoch": 3.028360049321825,
+      "grad_norm": 0.0711282268166542,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 29472
+    },
+    {
+      "epoch": 3.0286683107274968,
+      "grad_norm": 0.14440664649009705,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 29475
+    },
+    {
+      "epoch": 3.028976572133169,
+      "grad_norm": 0.06147081404924393,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 29478
+    },
+    {
+      "epoch": 3.029284833538841,
+      "grad_norm": 0.0842541828751564,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 29481
+    },
+    {
+      "epoch": 3.029593094944513,
+      "grad_norm": 0.04915475845336914,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 29484
+    },
+    {
+      "epoch": 3.029901356350185,
+      "grad_norm": 0.04254012182354927,
+      "learning_rate": 0.01,
+      "loss": 2.0406,
+      "step": 29487
+    },
+    {
+      "epoch": 3.030209617755857,
+      "grad_norm": 0.03716140240430832,
+      "learning_rate": 0.01,
+      "loss": 2.0477,
+      "step": 29490
+    },
+    {
+      "epoch": 3.030517879161529,
+      "grad_norm": 0.09156777709722519,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 29493
+    },
+    {
+      "epoch": 3.030826140567201,
+      "grad_norm": 0.05401970446109772,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 29496
+    },
+    {
+      "epoch": 3.031134401972873,
+      "grad_norm": 0.11564016342163086,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 29499
+    },
+    {
+      "epoch": 3.031442663378545,
+      "grad_norm": 0.12813927233219147,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 29502
+    },
+    {
+      "epoch": 3.031750924784217,
+      "grad_norm": 0.04971994087100029,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 29505
+    },
+    {
+      "epoch": 3.032059186189889,
+      "grad_norm": 0.037013307213783264,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 29508
+    },
+    {
+      "epoch": 3.032367447595561,
+      "grad_norm": 0.048204224556684494,
+      "learning_rate": 0.01,
+      "loss": 2.0635,
+      "step": 29511
+    },
+    {
+      "epoch": 3.032675709001233,
+      "grad_norm": 0.034393493086099625,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 29514
+    },
+    {
+      "epoch": 3.0329839704069053,
+      "grad_norm": 0.041031621396541595,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 29517
+    },
+    {
+      "epoch": 3.033292231812577,
+      "grad_norm": 0.049610815942287445,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 29520
+    },
+    {
+      "epoch": 3.033600493218249,
+      "grad_norm": 0.07062069326639175,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 29523
+    },
+    {
+      "epoch": 3.033908754623921,
+      "grad_norm": 0.040892720222473145,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 29526
+    },
+    {
+      "epoch": 3.034217016029593,
+      "grad_norm": 0.036804720759391785,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 29529
+    },
+    {
+      "epoch": 3.034525277435265,
+      "grad_norm": 0.06749982386827469,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 29532
+    },
+    {
+      "epoch": 3.034833538840937,
+      "grad_norm": 0.07996654510498047,
+      "learning_rate": 0.01,
+      "loss": 2.0587,
+      "step": 29535
+    },
+    {
+      "epoch": 3.0351418002466093,
+      "grad_norm": 0.05621659383177757,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 29538
+    },
+    {
+      "epoch": 3.035450061652281,
+      "grad_norm": 0.04464147612452507,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 29541
+    },
+    {
+      "epoch": 3.0357583230579532,
+      "grad_norm": 0.04619302973151207,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 29544
+    },
+    {
+      "epoch": 3.036066584463625,
+      "grad_norm": 0.09262579679489136,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 29547
+    },
+    {
+      "epoch": 3.036374845869297,
+      "grad_norm": 0.07954932749271393,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 29550
+    },
+    {
+      "epoch": 3.0366831072749694,
+      "grad_norm": 0.08746150135993958,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 29553
+    },
+    {
+      "epoch": 3.036991368680641,
+      "grad_norm": 0.04415113106369972,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 29556
+    },
+    {
+      "epoch": 3.0372996300863133,
+      "grad_norm": 0.08307299762964249,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 29559
+    },
+    {
+      "epoch": 3.037607891491985,
+      "grad_norm": 0.05882325395941734,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 29562
+    },
+    {
+      "epoch": 3.0379161528976573,
+      "grad_norm": 0.08524999022483826,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 29565
+    },
+    {
+      "epoch": 3.038224414303329,
+      "grad_norm": 0.07542447000741959,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 29568
+    },
+    {
+      "epoch": 3.038532675709001,
+      "grad_norm": 0.06394769251346588,
+      "learning_rate": 0.01,
+      "loss": 2.0551,
+      "step": 29571
+    },
+    {
+      "epoch": 3.0388409371146734,
+      "grad_norm": 0.07811316847801208,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 29574
+    },
+    {
+      "epoch": 3.039149198520345,
+      "grad_norm": 0.0849740207195282,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 29577
+    },
+    {
+      "epoch": 3.0394574599260173,
+      "grad_norm": 0.05540047585964203,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 29580
+    },
+    {
+      "epoch": 3.039765721331689,
+      "grad_norm": 0.044082462787628174,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 29583
+    },
+    {
+      "epoch": 3.0400739827373613,
+      "grad_norm": 0.11179853975772858,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 29586
+    },
+    {
+      "epoch": 3.0403822441430335,
+      "grad_norm": 0.10160455852746964,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 29589
+    },
+    {
+      "epoch": 3.0406905055487052,
+      "grad_norm": 0.08072539418935776,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 29592
+    },
+    {
+      "epoch": 3.0409987669543774,
+      "grad_norm": 0.039637934416532516,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 29595
+    },
+    {
+      "epoch": 3.041307028360049,
+      "grad_norm": 0.05847008153796196,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 29598
+    },
+    {
+      "epoch": 3.0416152897657214,
+      "grad_norm": 0.04309094697237015,
+      "learning_rate": 0.01,
+      "loss": 2.0445,
+      "step": 29601
+    },
+    {
+      "epoch": 3.041923551171393,
+      "grad_norm": 0.04230069741606712,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 29604
+    },
+    {
+      "epoch": 3.0422318125770653,
+      "grad_norm": 0.08122226595878601,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 29607
+    },
+    {
+      "epoch": 3.0425400739827375,
+      "grad_norm": 0.05134423449635506,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 29610
+    },
+    {
+      "epoch": 3.0428483353884093,
+      "grad_norm": 0.06343290954828262,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 29613
+    },
+    {
+      "epoch": 3.0431565967940815,
+      "grad_norm": 0.04415202513337135,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 29616
+    },
+    {
+      "epoch": 3.043464858199753,
+      "grad_norm": 0.09160872548818588,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 29619
+    },
+    {
+      "epoch": 3.0437731196054254,
+      "grad_norm": 0.0800345167517662,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 29622
+    },
+    {
+      "epoch": 3.0440813810110976,
+      "grad_norm": 0.11805953085422516,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 29625
+    },
+    {
+      "epoch": 3.0443896424167693,
+      "grad_norm": 0.036926135420799255,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 29628
+    },
+    {
+      "epoch": 3.0446979038224415,
+      "grad_norm": 0.04666229337453842,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 29631
+    },
+    {
+      "epoch": 3.0450061652281133,
+      "grad_norm": 0.152203768491745,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 29634
+    },
+    {
+      "epoch": 3.0453144266337855,
+      "grad_norm": 0.07508762180805206,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 29637
+    },
+    {
+      "epoch": 3.0456226880394572,
+      "grad_norm": 0.07548423856496811,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 29640
+    },
+    {
+      "epoch": 3.0459309494451294,
+      "grad_norm": 0.03427097201347351,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 29643
+    },
+    {
+      "epoch": 3.0462392108508016,
+      "grad_norm": 0.0399865061044693,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 29646
+    },
+    {
+      "epoch": 3.0465474722564734,
+      "grad_norm": 0.03335277736186981,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 29649
+    },
+    {
+      "epoch": 3.0468557336621456,
+      "grad_norm": 0.0520477369427681,
+      "learning_rate": 0.01,
+      "loss": 2.0389,
+      "step": 29652
+    },
+    {
+      "epoch": 3.0471639950678173,
+      "grad_norm": 0.08565925061702728,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 29655
+    },
+    {
+      "epoch": 3.0474722564734895,
+      "grad_norm": 0.062498703598976135,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 29658
+    },
+    {
+      "epoch": 3.0477805178791617,
+      "grad_norm": 0.08171094208955765,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 29661
+    },
+    {
+      "epoch": 3.0480887792848335,
+      "grad_norm": 0.06803334504365921,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 29664
+    },
+    {
+      "epoch": 3.0483970406905057,
+      "grad_norm": 0.113568514585495,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 29667
+    },
+    {
+      "epoch": 3.0487053020961774,
+      "grad_norm": 0.049806151539087296,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 29670
+    },
+    {
+      "epoch": 3.0490135635018496,
+      "grad_norm": 0.05908326804637909,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 29673
+    },
+    {
+      "epoch": 3.049321824907522,
+      "grad_norm": 0.0421968549489975,
+      "learning_rate": 0.01,
+      "loss": 2.038,
+      "step": 29676
+    },
+    {
+      "epoch": 3.0496300863131935,
+      "grad_norm": 0.04743592441082001,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 29679
+    },
+    {
+      "epoch": 3.0499383477188657,
+      "grad_norm": 0.11457180231809616,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 29682
+    },
+    {
+      "epoch": 3.0502466091245375,
+      "grad_norm": 0.052697159349918365,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 29685
+    },
+    {
+      "epoch": 3.0505548705302097,
+      "grad_norm": 0.140494704246521,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 29688
+    },
+    {
+      "epoch": 3.0508631319358814,
+      "grad_norm": 0.03548673912882805,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 29691
+    },
+    {
+      "epoch": 3.0511713933415536,
+      "grad_norm": 0.03223096579313278,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 29694
+    },
+    {
+      "epoch": 3.051479654747226,
+      "grad_norm": 0.040596138685941696,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 29697
+    },
+    {
+      "epoch": 3.0517879161528976,
+      "grad_norm": 0.09559677541255951,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 29700
+    },
+    {
+      "epoch": 3.0520961775585698,
+      "grad_norm": 0.10038384050130844,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 29703
+    },
+    {
+      "epoch": 3.0524044389642415,
+      "grad_norm": 0.06066306680440903,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 29706
+    },
+    {
+      "epoch": 3.0527127003699137,
+      "grad_norm": 0.08278308808803558,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 29709
+    },
+    {
+      "epoch": 3.053020961775586,
+      "grad_norm": 0.05971555784344673,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 29712
+    },
+    {
+      "epoch": 3.0533292231812577,
+      "grad_norm": 0.036974966526031494,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 29715
+    },
+    {
+      "epoch": 3.05363748458693,
+      "grad_norm": 0.0310160294175148,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 29718
+    },
+    {
+      "epoch": 3.0539457459926016,
+      "grad_norm": 0.03587731346487999,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 29721
+    },
+    {
+      "epoch": 3.054254007398274,
+      "grad_norm": 0.05497679486870766,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 29724
+    },
+    {
+      "epoch": 3.0545622688039455,
+      "grad_norm": 0.06713774055242538,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 29727
+    },
+    {
+      "epoch": 3.0548705302096177,
+      "grad_norm": 0.034498222172260284,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 29730
+    },
+    {
+      "epoch": 3.05517879161529,
+      "grad_norm": 0.12036336213350296,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 29733
+    },
+    {
+      "epoch": 3.0554870530209617,
+      "grad_norm": 0.08315537869930267,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 29736
+    },
+    {
+      "epoch": 3.055795314426634,
+      "grad_norm": 0.041693978011608124,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 29739
+    },
+    {
+      "epoch": 3.0561035758323056,
+      "grad_norm": 0.08889281749725342,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 29742
+    },
+    {
+      "epoch": 3.056411837237978,
+      "grad_norm": 0.07068068534135818,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 29745
+    },
+    {
+      "epoch": 3.05672009864365,
+      "grad_norm": 0.09744048863649368,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 29748
+    },
+    {
+      "epoch": 3.0570283600493218,
+      "grad_norm": 0.04529158025979996,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 29751
+    },
+    {
+      "epoch": 3.057336621454994,
+      "grad_norm": 0.041666992008686066,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 29754
+    },
+    {
+      "epoch": 3.0576448828606657,
+      "grad_norm": 0.07549773156642914,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 29757
+    },
+    {
+      "epoch": 3.057953144266338,
+      "grad_norm": 0.046318188309669495,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 29760
+    },
+    {
+      "epoch": 3.0582614056720097,
+      "grad_norm": 0.05954563990235329,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 29763
+    },
+    {
+      "epoch": 3.058569667077682,
+      "grad_norm": 0.03786987066268921,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 29766
+    },
+    {
+      "epoch": 3.058877928483354,
+      "grad_norm": 0.04249919578433037,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 29769
+    },
+    {
+      "epoch": 3.059186189889026,
+      "grad_norm": 0.09349595755338669,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 29772
+    },
+    {
+      "epoch": 3.059494451294698,
+      "grad_norm": 0.05637031048536301,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 29775
+    },
+    {
+      "epoch": 3.0598027127003697,
+      "grad_norm": 0.09952693432569504,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 29778
+    },
+    {
+      "epoch": 3.060110974106042,
+      "grad_norm": 0.10634247213602066,
+      "learning_rate": 0.01,
+      "loss": 2.0425,
+      "step": 29781
+    },
+    {
+      "epoch": 3.060419235511714,
+      "grad_norm": 0.0774141326546669,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 29784
+    },
+    {
+      "epoch": 3.060727496917386,
+      "grad_norm": 0.04012331739068031,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 29787
+    },
+    {
+      "epoch": 3.061035758323058,
+      "grad_norm": 0.04289887100458145,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 29790
+    },
+    {
+      "epoch": 3.06134401972873,
+      "grad_norm": 0.03678658604621887,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 29793
+    },
+    {
+      "epoch": 3.061652281134402,
+      "grad_norm": 0.034676194190979004,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 29796
+    },
+    {
+      "epoch": 3.0619605425400738,
+      "grad_norm": 0.08951381593942642,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 29799
+    },
+    {
+      "epoch": 3.062268803945746,
+      "grad_norm": 0.053729988634586334,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 29802
+    },
+    {
+      "epoch": 3.062577065351418,
+      "grad_norm": 0.08025490492582321,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 29805
+    },
+    {
+      "epoch": 3.06288532675709,
+      "grad_norm": 0.11099457740783691,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 29808
+    },
+    {
+      "epoch": 3.063193588162762,
+      "grad_norm": 0.13354651629924774,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 29811
+    },
+    {
+      "epoch": 3.063501849568434,
+      "grad_norm": 0.07378706336021423,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 29814
+    },
+    {
+      "epoch": 3.063810110974106,
+      "grad_norm": 0.04699387028813362,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 29817
+    },
+    {
+      "epoch": 3.0641183723797782,
+      "grad_norm": 0.08287378400564194,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 29820
+    },
+    {
+      "epoch": 3.06442663378545,
+      "grad_norm": 0.04004284739494324,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 29823
+    },
+    {
+      "epoch": 3.064734895191122,
+      "grad_norm": 0.05801365152001381,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 29826
+    },
+    {
+      "epoch": 3.065043156596794,
+      "grad_norm": 0.05196039378643036,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 29829
+    },
+    {
+      "epoch": 3.065351418002466,
+      "grad_norm": 0.04543706774711609,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 29832
+    },
+    {
+      "epoch": 3.065659679408138,
+      "grad_norm": 0.05413155257701874,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 29835
+    },
+    {
+      "epoch": 3.06596794081381,
+      "grad_norm": 0.05158795788884163,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 29838
+    },
+    {
+      "epoch": 3.0662762022194823,
+      "grad_norm": 0.05547649413347244,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 29841
+    },
+    {
+      "epoch": 3.066584463625154,
+      "grad_norm": 0.04968711733818054,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 29844
+    },
+    {
+      "epoch": 3.066892725030826,
+      "grad_norm": 0.05625564232468605,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 29847
+    },
+    {
+      "epoch": 3.067200986436498,
+      "grad_norm": 0.05007552728056908,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 29850
+    },
+    {
+      "epoch": 3.06750924784217,
+      "grad_norm": 0.08748306334018707,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 29853
+    },
+    {
+      "epoch": 3.0678175092478424,
+      "grad_norm": 0.08939902484416962,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 29856
+    },
+    {
+      "epoch": 3.068125770653514,
+      "grad_norm": 0.04412224516272545,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 29859
+    },
+    {
+      "epoch": 3.0684340320591863,
+      "grad_norm": 0.04574199765920639,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 29862
+    },
+    {
+      "epoch": 3.068742293464858,
+      "grad_norm": 0.059090469032526016,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 29865
+    },
+    {
+      "epoch": 3.0690505548705302,
+      "grad_norm": 0.048365235328674316,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 29868
+    },
+    {
+      "epoch": 3.0693588162762024,
+      "grad_norm": 0.05044705048203468,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 29871
+    },
+    {
+      "epoch": 3.069667077681874,
+      "grad_norm": 0.06108652427792549,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 29874
+    },
+    {
+      "epoch": 3.0699753390875464,
+      "grad_norm": 0.04400194063782692,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 29877
+    },
+    {
+      "epoch": 3.070283600493218,
+      "grad_norm": 0.047024693340063095,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 29880
+    },
+    {
+      "epoch": 3.0705918618988903,
+      "grad_norm": 0.06958126276731491,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 29883
+    },
+    {
+      "epoch": 3.070900123304562,
+      "grad_norm": 0.07789502292871475,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 29886
+    },
+    {
+      "epoch": 3.0712083847102343,
+      "grad_norm": 0.06438666582107544,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 29889
+    },
+    {
+      "epoch": 3.0715166461159065,
+      "grad_norm": 0.1446814239025116,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 29892
+    },
+    {
+      "epoch": 3.071824907521578,
+      "grad_norm": 0.055988240987062454,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 29895
+    },
+    {
+      "epoch": 3.0721331689272504,
+      "grad_norm": 0.04055717960000038,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 29898
+    },
+    {
+      "epoch": 3.072441430332922,
+      "grad_norm": 0.11918565630912781,
+      "learning_rate": 0.01,
+      "loss": 2.0436,
+      "step": 29901
+    },
+    {
+      "epoch": 3.0727496917385944,
+      "grad_norm": 0.04726850986480713,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 29904
+    },
+    {
+      "epoch": 3.0730579531442666,
+      "grad_norm": 0.0526423342525959,
+      "learning_rate": 0.01,
+      "loss": 2.0541,
+      "step": 29907
+    },
+    {
+      "epoch": 3.0733662145499383,
+      "grad_norm": 0.048728957772254944,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 29910
+    },
+    {
+      "epoch": 3.0736744759556105,
+      "grad_norm": 0.036348532885313034,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 29913
+    },
+    {
+      "epoch": 3.0739827373612822,
+      "grad_norm": 0.08039457350969315,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 29916
+    },
+    {
+      "epoch": 3.0742909987669544,
+      "grad_norm": 0.05370686575770378,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 29919
+    },
+    {
+      "epoch": 3.074599260172626,
+      "grad_norm": 0.08430914580821991,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 29922
+    },
+    {
+      "epoch": 3.0749075215782984,
+      "grad_norm": 0.04675029218196869,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 29925
+    },
+    {
+      "epoch": 3.0752157829839706,
+      "grad_norm": 0.0455501526594162,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 29928
+    },
+    {
+      "epoch": 3.0755240443896423,
+      "grad_norm": 0.04330252856016159,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 29931
+    },
+    {
+      "epoch": 3.0758323057953145,
+      "grad_norm": 0.04302576184272766,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 29934
+    },
+    {
+      "epoch": 3.0761405672009863,
+      "grad_norm": 0.07246873527765274,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 29937
+    },
+    {
+      "epoch": 3.0764488286066585,
+      "grad_norm": 0.12140212953090668,
+      "learning_rate": 0.01,
+      "loss": 2.0384,
+      "step": 29940
+    },
+    {
+      "epoch": 3.0767570900123307,
+      "grad_norm": 0.050940465182065964,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 29943
+    },
+    {
+      "epoch": 3.0770653514180024,
+      "grad_norm": 0.08281772583723068,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 29946
+    },
+    {
+      "epoch": 3.0773736128236746,
+      "grad_norm": 0.05485936999320984,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 29949
+    },
+    {
+      "epoch": 3.0776818742293464,
+      "grad_norm": 0.0354488380253315,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 29952
+    },
+    {
+      "epoch": 3.0779901356350186,
+      "grad_norm": 0.04454338923096657,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 29955
+    },
+    {
+      "epoch": 3.0782983970406903,
+      "grad_norm": 0.07980217784643173,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 29958
+    },
+    {
+      "epoch": 3.0786066584463625,
+      "grad_norm": 0.07601354271173477,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 29961
+    },
+    {
+      "epoch": 3.0789149198520347,
+      "grad_norm": 0.036425743252038956,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 29964
+    },
+    {
+      "epoch": 3.0792231812577064,
+      "grad_norm": 0.05909854918718338,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 29967
+    },
+    {
+      "epoch": 3.0795314426633786,
+      "grad_norm": 0.038156334310770035,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 29970
+    },
+    {
+      "epoch": 3.0798397040690504,
+      "grad_norm": 0.03241376578807831,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 29973
+    },
+    {
+      "epoch": 3.0801479654747226,
+      "grad_norm": 0.039934635162353516,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 29976
+    },
+    {
+      "epoch": 3.0804562268803948,
+      "grad_norm": 0.05287212133407593,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 29979
+    },
+    {
+      "epoch": 3.0807644882860665,
+      "grad_norm": 0.0807357057929039,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 29982
+    },
+    {
+      "epoch": 3.0810727496917387,
+      "grad_norm": 0.07264538109302521,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 29985
+    },
+    {
+      "epoch": 3.0813810110974105,
+      "grad_norm": 0.06328427046537399,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 29988
+    },
+    {
+      "epoch": 3.0816892725030827,
+      "grad_norm": 0.05093936249613762,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 29991
+    },
+    {
+      "epoch": 3.0819975339087544,
+      "grad_norm": 0.035449884831905365,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 29994
+    },
+    {
+      "epoch": 3.0823057953144266,
+      "grad_norm": 0.13781675696372986,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 29997
+    },
+    {
+      "epoch": 3.082614056720099,
+      "grad_norm": 0.06310711055994034,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 30000
+    },
+    {
+      "epoch": 3.0829223181257706,
+      "grad_norm": 0.059962496161460876,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 30003
+    },
+    {
+      "epoch": 3.0832305795314427,
+      "grad_norm": 0.07414010167121887,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 30006
+    },
+    {
+      "epoch": 3.0835388409371145,
+      "grad_norm": 0.05147621035575867,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 30009
+    },
+    {
+      "epoch": 3.0838471023427867,
+      "grad_norm": 0.08669129014015198,
+      "learning_rate": 0.01,
+      "loss": 2.029,
+      "step": 30012
+    },
+    {
+      "epoch": 3.084155363748459,
+      "grad_norm": 0.03791889548301697,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 30015
+    },
+    {
+      "epoch": 3.0844636251541306,
+      "grad_norm": 0.08246924728155136,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 30018
+    },
+    {
+      "epoch": 3.084771886559803,
+      "grad_norm": 0.11454186588525772,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 30021
+    },
+    {
+      "epoch": 3.0850801479654746,
+      "grad_norm": 0.07639762759208679,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 30024
+    },
+    {
+      "epoch": 3.085388409371147,
+      "grad_norm": 0.05283968895673752,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 30027
+    },
+    {
+      "epoch": 3.0856966707768185,
+      "grad_norm": 0.041770000010728836,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 30030
+    },
+    {
+      "epoch": 3.0860049321824907,
+      "grad_norm": 0.034824103116989136,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 30033
+    },
+    {
+      "epoch": 3.086313193588163,
+      "grad_norm": 0.043972861021757126,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 30036
+    },
+    {
+      "epoch": 3.0866214549938347,
+      "grad_norm": 0.03452165424823761,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 30039
+    },
+    {
+      "epoch": 3.086929716399507,
+      "grad_norm": 0.07175113260746002,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 30042
+    },
+    {
+      "epoch": 3.0872379778051786,
+      "grad_norm": 0.0780046209692955,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 30045
+    },
+    {
+      "epoch": 3.087546239210851,
+      "grad_norm": 0.06838949769735336,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 30048
+    },
+    {
+      "epoch": 3.087854500616523,
+      "grad_norm": 0.114894799888134,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 30051
+    },
+    {
+      "epoch": 3.0881627620221948,
+      "grad_norm": 0.10075607150793076,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 30054
+    },
+    {
+      "epoch": 3.088471023427867,
+      "grad_norm": 0.04926849529147148,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 30057
+    },
+    {
+      "epoch": 3.0887792848335387,
+      "grad_norm": 0.04143848270177841,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 30060
+    },
+    {
+      "epoch": 3.089087546239211,
+      "grad_norm": 0.04664148762822151,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 30063
+    },
+    {
+      "epoch": 3.089395807644883,
+      "grad_norm": 0.04764292761683464,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 30066
+    },
+    {
+      "epoch": 3.089704069050555,
+      "grad_norm": 0.06317167729139328,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 30069
+    },
+    {
+      "epoch": 3.090012330456227,
+      "grad_norm": 0.07983100414276123,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 30072
+    },
+    {
+      "epoch": 3.090320591861899,
+      "grad_norm": 0.03481268137693405,
+      "learning_rate": 0.01,
+      "loss": 2.0441,
+      "step": 30075
+    },
+    {
+      "epoch": 3.090628853267571,
+      "grad_norm": 0.042781632393598557,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 30078
+    },
+    {
+      "epoch": 3.0909371146732427,
+      "grad_norm": 0.05203581228852272,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 30081
+    },
+    {
+      "epoch": 3.091245376078915,
+      "grad_norm": 0.07452182471752167,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 30084
+    },
+    {
+      "epoch": 3.091553637484587,
+      "grad_norm": 0.05319884046912193,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 30087
+    },
+    {
+      "epoch": 3.091861898890259,
+      "grad_norm": 0.11648225039243698,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 30090
+    },
+    {
+      "epoch": 3.092170160295931,
+      "grad_norm": 0.07870937138795853,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 30093
+    },
+    {
+      "epoch": 3.092478421701603,
+      "grad_norm": 0.04574506729841232,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 30096
+    },
+    {
+      "epoch": 3.092786683107275,
+      "grad_norm": 0.07913927733898163,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 30099
+    },
+    {
+      "epoch": 3.0930949445129468,
+      "grad_norm": 0.08093220740556717,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 30102
+    },
+    {
+      "epoch": 3.093403205918619,
+      "grad_norm": 0.07089852541685104,
+      "learning_rate": 0.01,
+      "loss": 2.0292,
+      "step": 30105
+    },
+    {
+      "epoch": 3.093711467324291,
+      "grad_norm": 0.06293002516031265,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 30108
+    },
+    {
+      "epoch": 3.094019728729963,
+      "grad_norm": 0.057953476905822754,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 30111
+    },
+    {
+      "epoch": 3.094327990135635,
+      "grad_norm": 0.06989496946334839,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 30114
+    },
+    {
+      "epoch": 3.094636251541307,
+      "grad_norm": 0.04050515964627266,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 30117
+    },
+    {
+      "epoch": 3.094944512946979,
+      "grad_norm": 0.1112913116812706,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 30120
+    },
+    {
+      "epoch": 3.0952527743526512,
+      "grad_norm": 0.05135345458984375,
+      "learning_rate": 0.01,
+      "loss": 2.044,
+      "step": 30123
+    },
+    {
+      "epoch": 3.095561035758323,
+      "grad_norm": 0.1411667764186859,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 30126
+    },
+    {
+      "epoch": 3.095869297163995,
+      "grad_norm": 0.08077210187911987,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 30129
+    },
+    {
+      "epoch": 3.096177558569667,
+      "grad_norm": 0.10934364050626755,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 30132
+    },
+    {
+      "epoch": 3.096485819975339,
+      "grad_norm": 0.07186675071716309,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 30135
+    },
+    {
+      "epoch": 3.0967940813810113,
+      "grad_norm": 0.11918327957391739,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 30138
+    },
+    {
+      "epoch": 3.097102342786683,
+      "grad_norm": 0.052199963480234146,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 30141
+    },
+    {
+      "epoch": 3.0974106041923553,
+      "grad_norm": 0.06796009093523026,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 30144
+    },
+    {
+      "epoch": 3.097718865598027,
+      "grad_norm": 0.055949617177248,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 30147
+    },
+    {
+      "epoch": 3.098027127003699,
+      "grad_norm": 0.06256923824548721,
+      "learning_rate": 0.01,
+      "loss": 2.0299,
+      "step": 30150
+    },
+    {
+      "epoch": 3.098335388409371,
+      "grad_norm": 0.04565596580505371,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 30153
+    },
+    {
+      "epoch": 3.098643649815043,
+      "grad_norm": 0.051202937960624695,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 30156
+    },
+    {
+      "epoch": 3.0989519112207153,
+      "grad_norm": 0.09557224065065384,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 30159
+    },
+    {
+      "epoch": 3.099260172626387,
+      "grad_norm": 0.07491281628608704,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 30162
+    },
+    {
+      "epoch": 3.0995684340320593,
+      "grad_norm": 0.06766588240861893,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 30165
+    },
+    {
+      "epoch": 3.099876695437731,
+      "grad_norm": 0.036496859043836594,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 30168
+    },
+    {
+      "epoch": 3.1001849568434032,
+      "grad_norm": 0.09757262468338013,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 30171
+    },
+    {
+      "epoch": 3.1004932182490754,
+      "grad_norm": 0.046625055372714996,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 30174
+    },
+    {
+      "epoch": 3.100801479654747,
+      "grad_norm": 0.07655061781406403,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 30177
+    },
+    {
+      "epoch": 3.1011097410604194,
+      "grad_norm": 0.04577295854687691,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 30180
+    },
+    {
+      "epoch": 3.101418002466091,
+      "grad_norm": 0.04528047516942024,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 30183
+    },
+    {
+      "epoch": 3.1017262638717633,
+      "grad_norm": 0.04630058631300926,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 30186
+    },
+    {
+      "epoch": 3.102034525277435,
+      "grad_norm": 0.03936396911740303,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 30189
+    },
+    {
+      "epoch": 3.1023427866831073,
+      "grad_norm": 0.05879136547446251,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 30192
+    },
+    {
+      "epoch": 3.1026510480887795,
+      "grad_norm": 0.061699915677309036,
+      "learning_rate": 0.01,
+      "loss": 2.0418,
+      "step": 30195
+    },
+    {
+      "epoch": 3.102959309494451,
+      "grad_norm": 0.05961352214217186,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 30198
+    },
+    {
+      "epoch": 3.1032675709001234,
+      "grad_norm": 0.10209197551012039,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 30201
+    },
+    {
+      "epoch": 3.103575832305795,
+      "grad_norm": 0.10693737864494324,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 30204
+    },
+    {
+      "epoch": 3.1038840937114673,
+      "grad_norm": 0.11183932423591614,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 30207
+    },
+    {
+      "epoch": 3.1041923551171395,
+      "grad_norm": 0.07532890886068344,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 30210
+    },
+    {
+      "epoch": 3.1045006165228113,
+      "grad_norm": 0.04888713359832764,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 30213
+    },
+    {
+      "epoch": 3.1048088779284835,
+      "grad_norm": 0.0533674955368042,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 30216
+    },
+    {
+      "epoch": 3.1051171393341552,
+      "grad_norm": 0.044245727360248566,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 30219
+    },
+    {
+      "epoch": 3.1054254007398274,
+      "grad_norm": 0.032582368701696396,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 30222
+    },
+    {
+      "epoch": 3.105733662145499,
+      "grad_norm": 0.10102632641792297,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 30225
+    },
+    {
+      "epoch": 3.1060419235511714,
+      "grad_norm": 0.03793931379914284,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 30228
+    },
+    {
+      "epoch": 3.1063501849568436,
+      "grad_norm": 0.04896242544054985,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 30231
+    },
+    {
+      "epoch": 3.1066584463625153,
+      "grad_norm": 0.057475607842206955,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 30234
+    },
+    {
+      "epoch": 3.1069667077681875,
+      "grad_norm": 0.037703339010477066,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 30237
+    },
+    {
+      "epoch": 3.1072749691738593,
+      "grad_norm": 0.04800937697291374,
+      "learning_rate": 0.01,
+      "loss": 2.043,
+      "step": 30240
+    },
+    {
+      "epoch": 3.1075832305795315,
+      "grad_norm": 0.042761534452438354,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 30243
+    },
+    {
+      "epoch": 3.1078914919852036,
+      "grad_norm": 0.0864332839846611,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 30246
+    },
+    {
+      "epoch": 3.1081997533908754,
+      "grad_norm": 0.10482830554246902,
+      "learning_rate": 0.01,
+      "loss": 2.0601,
+      "step": 30249
+    },
+    {
+      "epoch": 3.1085080147965476,
+      "grad_norm": 0.10628984868526459,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 30252
+    },
+    {
+      "epoch": 3.1088162762022193,
+      "grad_norm": 0.04853438585996628,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 30255
+    },
+    {
+      "epoch": 3.1091245376078915,
+      "grad_norm": 0.05274122208356857,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 30258
+    },
+    {
+      "epoch": 3.1094327990135637,
+      "grad_norm": 0.03757349029183388,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 30261
+    },
+    {
+      "epoch": 3.1097410604192355,
+      "grad_norm": 0.04505928233265877,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 30264
+    },
+    {
+      "epoch": 3.1100493218249077,
+      "grad_norm": 0.08804783225059509,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 30267
+    },
+    {
+      "epoch": 3.1103575832305794,
+      "grad_norm": 0.11545568704605103,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 30270
+    },
+    {
+      "epoch": 3.1106658446362516,
+      "grad_norm": 0.04955466091632843,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 30273
+    },
+    {
+      "epoch": 3.1109741060419234,
+      "grad_norm": 0.04930401220917702,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 30276
+    },
+    {
+      "epoch": 3.1112823674475956,
+      "grad_norm": 0.05701540783047676,
+      "learning_rate": 0.01,
+      "loss": 2.046,
+      "step": 30279
+    },
+    {
+      "epoch": 3.1115906288532678,
+      "grad_norm": 0.07442028075456619,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 30282
+    },
+    {
+      "epoch": 3.1118988902589395,
+      "grad_norm": 0.06538541615009308,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 30285
+    },
+    {
+      "epoch": 3.1122071516646117,
+      "grad_norm": 0.0826522633433342,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 30288
+    },
+    {
+      "epoch": 3.1125154130702835,
+      "grad_norm": 0.1353752762079239,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 30291
+    },
+    {
+      "epoch": 3.1128236744759556,
+      "grad_norm": 0.06406208872795105,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 30294
+    },
+    {
+      "epoch": 3.1131319358816274,
+      "grad_norm": 0.036653418093919754,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 30297
+    },
+    {
+      "epoch": 3.1134401972872996,
+      "grad_norm": 0.0839834213256836,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 30300
+    },
+    {
+      "epoch": 3.113748458692972,
+      "grad_norm": 0.04629015177488327,
+      "learning_rate": 0.01,
+      "loss": 2.0488,
+      "step": 30303
+    },
+    {
+      "epoch": 3.1140567200986435,
+      "grad_norm": 0.08630986511707306,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 30306
+    },
+    {
+      "epoch": 3.1143649815043157,
+      "grad_norm": 0.07894504070281982,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 30309
+    },
+    {
+      "epoch": 3.1146732429099875,
+      "grad_norm": 0.09203556925058365,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 30312
+    },
+    {
+      "epoch": 3.1149815043156597,
+      "grad_norm": 0.061055563390254974,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 30315
+    },
+    {
+      "epoch": 3.115289765721332,
+      "grad_norm": 0.0711871087551117,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 30318
+    },
+    {
+      "epoch": 3.1155980271270036,
+      "grad_norm": 0.07315775007009506,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 30321
+    },
+    {
+      "epoch": 3.115906288532676,
+      "grad_norm": 0.1050182655453682,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 30324
+    },
+    {
+      "epoch": 3.1162145499383476,
+      "grad_norm": 0.04959526285529137,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 30327
+    },
+    {
+      "epoch": 3.1165228113440198,
+      "grad_norm": 0.061066534370183945,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 30330
+    },
+    {
+      "epoch": 3.116831072749692,
+      "grad_norm": 0.057984329760074615,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 30333
+    },
+    {
+      "epoch": 3.1171393341553637,
+      "grad_norm": 0.07337040454149246,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 30336
+    },
+    {
+      "epoch": 3.117447595561036,
+      "grad_norm": 0.07801029831171036,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 30339
+    },
+    {
+      "epoch": 3.1177558569667077,
+      "grad_norm": 0.06625111401081085,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 30342
+    },
+    {
+      "epoch": 3.11806411837238,
+      "grad_norm": 0.11416434496641159,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 30345
+    },
+    {
+      "epoch": 3.1183723797780516,
+      "grad_norm": 0.05046262592077255,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 30348
+    },
+    {
+      "epoch": 3.118680641183724,
+      "grad_norm": 0.030621282756328583,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 30351
+    },
+    {
+      "epoch": 3.118988902589396,
+      "grad_norm": 0.03274908661842346,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 30354
+    },
+    {
+      "epoch": 3.1192971639950677,
+      "grad_norm": 0.03673490136861801,
+      "learning_rate": 0.01,
+      "loss": 2.0383,
+      "step": 30357
+    },
+    {
+      "epoch": 3.11960542540074,
+      "grad_norm": 0.07161369919776917,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 30360
+    },
+    {
+      "epoch": 3.1199136868064117,
+      "grad_norm": 0.04244636744260788,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 30363
+    },
+    {
+      "epoch": 3.120221948212084,
+      "grad_norm": 0.05357150360941887,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 30366
+    },
+    {
+      "epoch": 3.120530209617756,
+      "grad_norm": 0.06689538061618805,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 30369
+    },
+    {
+      "epoch": 3.120838471023428,
+      "grad_norm": 0.08160898089408875,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 30372
+    },
+    {
+      "epoch": 3.1211467324291,
+      "grad_norm": 0.04037999361753464,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 30375
+    },
+    {
+      "epoch": 3.1214549938347718,
+      "grad_norm": 0.09855981171131134,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 30378
+    },
+    {
+      "epoch": 3.121763255240444,
+      "grad_norm": 0.04645252600312233,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 30381
+    },
+    {
+      "epoch": 3.1220715166461157,
+      "grad_norm": 0.10605467110872269,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 30384
+    },
+    {
+      "epoch": 3.122379778051788,
+      "grad_norm": 0.11321547627449036,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 30387
+    },
+    {
+      "epoch": 3.12268803945746,
+      "grad_norm": 0.07283324748277664,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 30390
+    },
+    {
+      "epoch": 3.122996300863132,
+      "grad_norm": 0.07217562943696976,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 30393
+    },
+    {
+      "epoch": 3.123304562268804,
+      "grad_norm": 0.036078888922929764,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 30396
+    },
+    {
+      "epoch": 3.123612823674476,
+      "grad_norm": 0.03730477765202522,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 30399
+    },
+    {
+      "epoch": 3.123921085080148,
+      "grad_norm": 0.04707048460841179,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 30402
+    },
+    {
+      "epoch": 3.12422934648582,
+      "grad_norm": 0.0571308508515358,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 30405
+    },
+    {
+      "epoch": 3.124537607891492,
+      "grad_norm": 0.05167640373110771,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 30408
+    },
+    {
+      "epoch": 3.124845869297164,
+      "grad_norm": 0.06759858876466751,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 30411
+    },
+    {
+      "epoch": 3.125154130702836,
+      "grad_norm": 0.031637318432331085,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 30414
+    },
+    {
+      "epoch": 3.125462392108508,
+      "grad_norm": 0.0991082713007927,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 30417
+    },
+    {
+      "epoch": 3.12577065351418,
+      "grad_norm": 0.06524378806352615,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 30420
+    },
+    {
+      "epoch": 3.126078914919852,
+      "grad_norm": 0.05751586705446243,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 30423
+    },
+    {
+      "epoch": 3.126387176325524,
+      "grad_norm": 0.03776116296648979,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 30426
+    },
+    {
+      "epoch": 3.126695437731196,
+      "grad_norm": 0.0619078204035759,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 30429
+    },
+    {
+      "epoch": 3.127003699136868,
+      "grad_norm": 0.0511760450899601,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 30432
+    },
+    {
+      "epoch": 3.12731196054254,
+      "grad_norm": 0.053964611142873764,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 30435
+    },
+    {
+      "epoch": 3.127620221948212,
+      "grad_norm": 0.11773833632469177,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 30438
+    },
+    {
+      "epoch": 3.1279284833538843,
+      "grad_norm": 0.04505350813269615,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 30441
+    },
+    {
+      "epoch": 3.128236744759556,
+      "grad_norm": 0.10077274590730667,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 30444
+    },
+    {
+      "epoch": 3.1285450061652282,
+      "grad_norm": 0.05291616916656494,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 30447
+    },
+    {
+      "epoch": 3.1288532675709,
+      "grad_norm": 0.049340371042490005,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 30450
+    },
+    {
+      "epoch": 3.129161528976572,
+      "grad_norm": 0.06910324841737747,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 30453
+    },
+    {
+      "epoch": 3.1294697903822444,
+      "grad_norm": 0.07307056337594986,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 30456
+    },
+    {
+      "epoch": 3.129778051787916,
+      "grad_norm": 0.06828980147838593,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 30459
+    },
+    {
+      "epoch": 3.1300863131935883,
+      "grad_norm": 0.0480424240231514,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 30462
+    },
+    {
+      "epoch": 3.13039457459926,
+      "grad_norm": 0.10072044283151627,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 30465
+    },
+    {
+      "epoch": 3.1307028360049323,
+      "grad_norm": 0.04987982660531998,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 30468
+    },
+    {
+      "epoch": 3.131011097410604,
+      "grad_norm": 0.10036738961935043,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 30471
+    },
+    {
+      "epoch": 3.131319358816276,
+      "grad_norm": 0.06445086747407913,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 30474
+    },
+    {
+      "epoch": 3.1316276202219484,
+      "grad_norm": 0.10858535021543503,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 30477
+    },
+    {
+      "epoch": 3.13193588162762,
+      "grad_norm": 0.041025932878255844,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 30480
+    },
+    {
+      "epoch": 3.1322441430332923,
+      "grad_norm": 0.05287109315395355,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 30483
+    },
+    {
+      "epoch": 3.132552404438964,
+      "grad_norm": 0.09659206867218018,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 30486
+    },
+    {
+      "epoch": 3.1328606658446363,
+      "grad_norm": 0.07235360145568848,
+      "learning_rate": 0.01,
+      "loss": 2.0521,
+      "step": 30489
+    },
+    {
+      "epoch": 3.133168927250308,
+      "grad_norm": 0.042338863015174866,
+      "learning_rate": 0.01,
+      "loss": 2.0403,
+      "step": 30492
+    },
+    {
+      "epoch": 3.1334771886559802,
+      "grad_norm": 0.045676395297050476,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 30495
+    },
+    {
+      "epoch": 3.1337854500616524,
+      "grad_norm": 0.04957246780395508,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 30498
+    },
+    {
+      "epoch": 3.134093711467324,
+      "grad_norm": 0.047321684658527374,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 30501
+    },
+    {
+      "epoch": 3.1344019728729964,
+      "grad_norm": 0.09219278395175934,
+      "learning_rate": 0.01,
+      "loss": 2.0326,
+      "step": 30504
+    },
+    {
+      "epoch": 3.134710234278668,
+      "grad_norm": 0.07666533440351486,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 30507
+    },
+    {
+      "epoch": 3.1350184956843403,
+      "grad_norm": 0.06832748651504517,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 30510
+    },
+    {
+      "epoch": 3.1353267570900125,
+      "grad_norm": 0.05527606979012489,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 30513
+    },
+    {
+      "epoch": 3.1356350184956843,
+      "grad_norm": 0.09272732585668564,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 30516
+    },
+    {
+      "epoch": 3.1359432799013565,
+      "grad_norm": 0.02908925898373127,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 30519
+    },
+    {
+      "epoch": 3.136251541307028,
+      "grad_norm": 0.04611453413963318,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 30522
+    },
+    {
+      "epoch": 3.1365598027127004,
+      "grad_norm": 0.05230382829904556,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 30525
+    },
+    {
+      "epoch": 3.1368680641183726,
+      "grad_norm": 0.09377764165401459,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 30528
+    },
+    {
+      "epoch": 3.1371763255240444,
+      "grad_norm": 0.049708276987075806,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 30531
+    },
+    {
+      "epoch": 3.1374845869297165,
+      "grad_norm": 0.05086242035031319,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 30534
+    },
+    {
+      "epoch": 3.1377928483353883,
+      "grad_norm": 0.05085150897502899,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 30537
+    },
+    {
+      "epoch": 3.1381011097410605,
+      "grad_norm": 0.05230359733104706,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 30540
+    },
+    {
+      "epoch": 3.1384093711467322,
+      "grad_norm": 0.07910077273845673,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 30543
+    },
+    {
+      "epoch": 3.1387176325524044,
+      "grad_norm": 0.044358085840940475,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 30546
+    },
+    {
+      "epoch": 3.1390258939580766,
+      "grad_norm": 0.07269710302352905,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 30549
+    },
+    {
+      "epoch": 3.1393341553637484,
+      "grad_norm": 0.09917914122343063,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 30552
+    },
+    {
+      "epoch": 3.1396424167694206,
+      "grad_norm": 0.0755099207162857,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 30555
+    },
+    {
+      "epoch": 3.1399506781750923,
+      "grad_norm": 0.07911227643489838,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 30558
+    },
+    {
+      "epoch": 3.1402589395807645,
+      "grad_norm": 0.04883533716201782,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 30561
+    },
+    {
+      "epoch": 3.1405672009864363,
+      "grad_norm": 0.0375591404736042,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 30564
+    },
+    {
+      "epoch": 3.1408754623921085,
+      "grad_norm": 0.08654747903347015,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 30567
+    },
+    {
+      "epoch": 3.1411837237977807,
+      "grad_norm": 0.07025197148323059,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 30570
+    },
+    {
+      "epoch": 3.1414919852034524,
+      "grad_norm": 0.11750215291976929,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 30573
+    },
+    {
+      "epoch": 3.1418002466091246,
+      "grad_norm": 0.037444472312927246,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 30576
+    },
+    {
+      "epoch": 3.1421085080147964,
+      "grad_norm": 0.044617678970098495,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 30579
+    },
+    {
+      "epoch": 3.1424167694204685,
+      "grad_norm": 0.06604386866092682,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 30582
+    },
+    {
+      "epoch": 3.1427250308261407,
+      "grad_norm": 0.09958125650882721,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 30585
+    },
+    {
+      "epoch": 3.1430332922318125,
+      "grad_norm": 0.05710573121905327,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 30588
+    },
+    {
+      "epoch": 3.1433415536374847,
+      "grad_norm": 0.05984263867139816,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 30591
+    },
+    {
+      "epoch": 3.1436498150431564,
+      "grad_norm": 0.04073350876569748,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 30594
+    },
+    {
+      "epoch": 3.1439580764488286,
+      "grad_norm": 0.07050776481628418,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 30597
+    },
+    {
+      "epoch": 3.144266337854501,
+      "grad_norm": 0.03400518372654915,
+      "learning_rate": 0.01,
+      "loss": 2.0404,
+      "step": 30600
+    },
+    {
+      "epoch": 3.1445745992601726,
+      "grad_norm": 0.06751801073551178,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 30603
+    },
+    {
+      "epoch": 3.1448828606658448,
+      "grad_norm": 0.06015434488654137,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 30606
+    },
+    {
+      "epoch": 3.1451911220715165,
+      "grad_norm": 0.11231853067874908,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 30609
+    },
+    {
+      "epoch": 3.1454993834771887,
+      "grad_norm": 0.047932934015989304,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 30612
+    },
+    {
+      "epoch": 3.1458076448828605,
+      "grad_norm": 0.04414699971675873,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 30615
+    },
+    {
+      "epoch": 3.1461159062885327,
+      "grad_norm": 0.06486550718545914,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 30618
+    },
+    {
+      "epoch": 3.146424167694205,
+      "grad_norm": 0.056324832141399384,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 30621
+    },
+    {
+      "epoch": 3.1467324290998766,
+      "grad_norm": 0.030867785215377808,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 30624
+    },
+    {
+      "epoch": 3.147040690505549,
+      "grad_norm": 0.050489556044340134,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 30627
+    },
+    {
+      "epoch": 3.1473489519112205,
+      "grad_norm": 0.04749634861946106,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 30630
+    },
+    {
+      "epoch": 3.1476572133168927,
+      "grad_norm": 0.05312662571668625,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 30633
+    },
+    {
+      "epoch": 3.147965474722565,
+      "grad_norm": 0.1234770268201828,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 30636
+    },
+    {
+      "epoch": 3.1482737361282367,
+      "grad_norm": 0.1423310786485672,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 30639
+    },
+    {
+      "epoch": 3.148581997533909,
+      "grad_norm": 0.08930431306362152,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 30642
+    },
+    {
+      "epoch": 3.1488902589395806,
+      "grad_norm": 0.03335024416446686,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 30645
+    },
+    {
+      "epoch": 3.149198520345253,
+      "grad_norm": 0.07785200327634811,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 30648
+    },
+    {
+      "epoch": 3.1495067817509246,
+      "grad_norm": 0.047731272876262665,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 30651
+    },
+    {
+      "epoch": 3.1498150431565968,
+      "grad_norm": 0.10476487874984741,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 30654
+    },
+    {
+      "epoch": 3.150123304562269,
+      "grad_norm": 0.05200812220573425,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 30657
+    },
+    {
+      "epoch": 3.1504315659679407,
+      "grad_norm": 0.06658156216144562,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 30660
+    },
+    {
+      "epoch": 3.150739827373613,
+      "grad_norm": 0.06515184789896011,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 30663
+    },
+    {
+      "epoch": 3.1510480887792847,
+      "grad_norm": 0.03604321926832199,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 30666
+    },
+    {
+      "epoch": 3.151356350184957,
+      "grad_norm": 0.10496728122234344,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 30669
+    },
+    {
+      "epoch": 3.151664611590629,
+      "grad_norm": 0.05922207236289978,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 30672
+    },
+    {
+      "epoch": 3.151972872996301,
+      "grad_norm": 0.03718853369355202,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 30675
+    },
+    {
+      "epoch": 3.152281134401973,
+      "grad_norm": 0.04691898077726364,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 30678
+    },
+    {
+      "epoch": 3.1525893958076447,
+      "grad_norm": 0.048930030316114426,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 30681
+    },
+    {
+      "epoch": 3.152897657213317,
+      "grad_norm": 0.028419634327292442,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 30684
+    },
+    {
+      "epoch": 3.1532059186189887,
+      "grad_norm": 0.05914349481463432,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 30687
+    },
+    {
+      "epoch": 3.153514180024661,
+      "grad_norm": 0.05946047231554985,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 30690
+    },
+    {
+      "epoch": 3.153822441430333,
+      "grad_norm": 0.06951475888490677,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 30693
+    },
+    {
+      "epoch": 3.154130702836005,
+      "grad_norm": 0.055885329842567444,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 30696
+    },
+    {
+      "epoch": 3.154438964241677,
+      "grad_norm": 0.043436676263809204,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 30699
+    },
+    {
+      "epoch": 3.1547472256473488,
+      "grad_norm": 0.04978411644697189,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 30702
+    },
+    {
+      "epoch": 3.155055487053021,
+      "grad_norm": 0.11427465081214905,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 30705
+    },
+    {
+      "epoch": 3.155363748458693,
+      "grad_norm": 0.045963071286678314,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 30708
+    },
+    {
+      "epoch": 3.155672009864365,
+      "grad_norm": 0.03811359032988548,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 30711
+    },
+    {
+      "epoch": 3.155980271270037,
+      "grad_norm": 0.05763382837176323,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 30714
+    },
+    {
+      "epoch": 3.156288532675709,
+      "grad_norm": 0.045626237988471985,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 30717
+    },
+    {
+      "epoch": 3.156596794081381,
+      "grad_norm": 0.037544216960668564,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 30720
+    },
+    {
+      "epoch": 3.1569050554870532,
+      "grad_norm": 0.14238816499710083,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 30723
+    },
+    {
+      "epoch": 3.157213316892725,
+      "grad_norm": 0.048318054527044296,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 30726
+    },
+    {
+      "epoch": 3.157521578298397,
+      "grad_norm": 0.041052673012018204,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 30729
+    },
+    {
+      "epoch": 3.157829839704069,
+      "grad_norm": 0.0370815135538578,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 30732
+    },
+    {
+      "epoch": 3.158138101109741,
+      "grad_norm": 0.07156988233327866,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 30735
+    },
+    {
+      "epoch": 3.158446362515413,
+      "grad_norm": 0.0757046490907669,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 30738
+    },
+    {
+      "epoch": 3.158754623921085,
+      "grad_norm": 0.0674174353480339,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 30741
+    },
+    {
+      "epoch": 3.1590628853267573,
+      "grad_norm": 0.05094176158308983,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 30744
+    },
+    {
+      "epoch": 3.159371146732429,
+      "grad_norm": 0.04377339780330658,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 30747
+    },
+    {
+      "epoch": 3.159679408138101,
+      "grad_norm": 0.04788428172469139,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 30750
+    },
+    {
+      "epoch": 3.159987669543773,
+      "grad_norm": 0.08424562215805054,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 30753
+    },
+    {
+      "epoch": 3.160295930949445,
+      "grad_norm": 0.0742245689034462,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 30756
+    },
+    {
+      "epoch": 3.160604192355117,
+      "grad_norm": 0.127692312002182,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 30759
+    },
+    {
+      "epoch": 3.160912453760789,
+      "grad_norm": 0.06112000718712807,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 30762
+    },
+    {
+      "epoch": 3.1612207151664613,
+      "grad_norm": 0.04676929488778114,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 30765
+    },
+    {
+      "epoch": 3.161528976572133,
+      "grad_norm": 0.07944846153259277,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 30768
+    },
+    {
+      "epoch": 3.1618372379778052,
+      "grad_norm": 0.05377965793013573,
+      "learning_rate": 0.01,
+      "loss": 2.0424,
+      "step": 30771
+    },
+    {
+      "epoch": 3.162145499383477,
+      "grad_norm": 0.04961777105927467,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 30774
+    },
+    {
+      "epoch": 3.162453760789149,
+      "grad_norm": 0.043640993535518646,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 30777
+    },
+    {
+      "epoch": 3.1627620221948214,
+      "grad_norm": 0.08145361393690109,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 30780
+    },
+    {
+      "epoch": 3.163070283600493,
+      "grad_norm": 0.03488341346383095,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 30783
+    },
+    {
+      "epoch": 3.1633785450061653,
+      "grad_norm": 0.054735999554395676,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 30786
+    },
+    {
+      "epoch": 3.163686806411837,
+      "grad_norm": 0.05064836144447327,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 30789
+    },
+    {
+      "epoch": 3.1639950678175093,
+      "grad_norm": 0.05693964287638664,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 30792
+    },
+    {
+      "epoch": 3.1643033292231815,
+      "grad_norm": 0.08762123435735703,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 30795
+    },
+    {
+      "epoch": 3.164611590628853,
+      "grad_norm": 0.08652741461992264,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 30798
+    },
+    {
+      "epoch": 3.1649198520345254,
+      "grad_norm": 0.15153749287128448,
+      "learning_rate": 0.01,
+      "loss": 2.0326,
+      "step": 30801
+    },
+    {
+      "epoch": 3.165228113440197,
+      "grad_norm": 0.09822986274957657,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 30804
+    },
+    {
+      "epoch": 3.1655363748458694,
+      "grad_norm": 0.07334254682064056,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 30807
+    },
+    {
+      "epoch": 3.165844636251541,
+      "grad_norm": 0.042224787175655365,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 30810
+    },
+    {
+      "epoch": 3.1661528976572133,
+      "grad_norm": 0.047276921570301056,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 30813
+    },
+    {
+      "epoch": 3.1664611590628855,
+      "grad_norm": 0.112834133207798,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 30816
+    },
+    {
+      "epoch": 3.1667694204685573,
+      "grad_norm": 0.0851617380976677,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 30819
+    },
+    {
+      "epoch": 3.1670776818742294,
+      "grad_norm": 0.046194661408662796,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 30822
+    },
+    {
+      "epoch": 3.167385943279901,
+      "grad_norm": 0.056417178362607956,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 30825
+    },
+    {
+      "epoch": 3.1676942046855734,
+      "grad_norm": 0.06759685277938843,
+      "learning_rate": 0.01,
+      "loss": 2.0281,
+      "step": 30828
+    },
+    {
+      "epoch": 3.1680024660912456,
+      "grad_norm": 0.08961043506860733,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 30831
+    },
+    {
+      "epoch": 3.1683107274969173,
+      "grad_norm": 0.03684352710843086,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 30834
+    },
+    {
+      "epoch": 3.1686189889025895,
+      "grad_norm": 0.1388491988182068,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 30837
+    },
+    {
+      "epoch": 3.1689272503082613,
+      "grad_norm": 0.03769170120358467,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 30840
+    },
+    {
+      "epoch": 3.1692355117139335,
+      "grad_norm": 0.040609996765851974,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 30843
+    },
+    {
+      "epoch": 3.1695437731196052,
+      "grad_norm": 0.04916587471961975,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 30846
+    },
+    {
+      "epoch": 3.1698520345252774,
+      "grad_norm": 0.042920999228954315,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 30849
+    },
+    {
+      "epoch": 3.1701602959309496,
+      "grad_norm": 0.05777138099074364,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 30852
+    },
+    {
+      "epoch": 3.1704685573366214,
+      "grad_norm": 0.06128811836242676,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 30855
+    },
+    {
+      "epoch": 3.1707768187422936,
+      "grad_norm": 0.08042391389608383,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 30858
+    },
+    {
+      "epoch": 3.1710850801479653,
+      "grad_norm": 0.036383189260959625,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 30861
+    },
+    {
+      "epoch": 3.1713933415536375,
+      "grad_norm": 0.05068094655871391,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 30864
+    },
+    {
+      "epoch": 3.1717016029593097,
+      "grad_norm": 0.09005576372146606,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 30867
+    },
+    {
+      "epoch": 3.1720098643649814,
+      "grad_norm": 0.0589495450258255,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 30870
+    },
+    {
+      "epoch": 3.1723181257706536,
+      "grad_norm": 0.04160517826676369,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 30873
+    },
+    {
+      "epoch": 3.1726263871763254,
+      "grad_norm": 0.044718582183122635,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 30876
+    },
+    {
+      "epoch": 3.1729346485819976,
+      "grad_norm": 0.09011929482221603,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 30879
+    },
+    {
+      "epoch": 3.1732429099876693,
+      "grad_norm": 0.0597953200340271,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 30882
+    },
+    {
+      "epoch": 3.1735511713933415,
+      "grad_norm": 0.06141204759478569,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 30885
+    },
+    {
+      "epoch": 3.1738594327990137,
+      "grad_norm": 0.09013784676790237,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 30888
+    },
+    {
+      "epoch": 3.1741676942046855,
+      "grad_norm": 0.08281320333480835,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 30891
+    },
+    {
+      "epoch": 3.1744759556103577,
+      "grad_norm": 0.09314266592264175,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 30894
+    },
+    {
+      "epoch": 3.1747842170160294,
+      "grad_norm": 0.04410851374268532,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 30897
+    },
+    {
+      "epoch": 3.1750924784217016,
+      "grad_norm": 0.12074366211891174,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 30900
+    },
+    {
+      "epoch": 3.175400739827374,
+      "grad_norm": 0.06861037015914917,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 30903
+    },
+    {
+      "epoch": 3.1757090012330456,
+      "grad_norm": 0.038134749978780746,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 30906
+    },
+    {
+      "epoch": 3.1760172626387178,
+      "grad_norm": 0.11108820140361786,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 30909
+    },
+    {
+      "epoch": 3.1763255240443895,
+      "grad_norm": 0.047687213867902756,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 30912
+    },
+    {
+      "epoch": 3.1766337854500617,
+      "grad_norm": 0.05983440950512886,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 30915
+    },
+    {
+      "epoch": 3.176942046855734,
+      "grad_norm": 0.038930587470531464,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 30918
+    },
+    {
+      "epoch": 3.1772503082614056,
+      "grad_norm": 0.08851155638694763,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 30921
+    },
+    {
+      "epoch": 3.177558569667078,
+      "grad_norm": 0.08430106192827225,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 30924
+    },
+    {
+      "epoch": 3.1778668310727496,
+      "grad_norm": 0.04469950869679451,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 30927
+    },
+    {
+      "epoch": 3.178175092478422,
+      "grad_norm": 0.06740089505910873,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 30930
+    },
+    {
+      "epoch": 3.1784833538840935,
+      "grad_norm": 0.06175517663359642,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 30933
+    },
+    {
+      "epoch": 3.1787916152897657,
+      "grad_norm": 0.047650065273046494,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 30936
+    },
+    {
+      "epoch": 3.179099876695438,
+      "grad_norm": 0.07261566072702408,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 30939
+    },
+    {
+      "epoch": 3.1794081381011097,
+      "grad_norm": 0.1270940601825714,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 30942
+    },
+    {
+      "epoch": 3.179716399506782,
+      "grad_norm": 0.054443880915641785,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 30945
+    },
+    {
+      "epoch": 3.1800246609124536,
+      "grad_norm": 0.06243740767240524,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 30948
+    },
+    {
+      "epoch": 3.180332922318126,
+      "grad_norm": 0.04665446653962135,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 30951
+    },
+    {
+      "epoch": 3.1806411837237976,
+      "grad_norm": 0.0470532663166523,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 30954
+    },
+    {
+      "epoch": 3.1809494451294698,
+      "grad_norm": 0.03944860398769379,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 30957
+    },
+    {
+      "epoch": 3.181257706535142,
+      "grad_norm": 0.04705018922686577,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 30960
+    },
+    {
+      "epoch": 3.1815659679408137,
+      "grad_norm": 0.04603470116853714,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 30963
+    },
+    {
+      "epoch": 3.181874229346486,
+      "grad_norm": 0.04761103168129921,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 30966
+    },
+    {
+      "epoch": 3.1821824907521576,
+      "grad_norm": 0.09109831601381302,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 30969
+    },
+    {
+      "epoch": 3.18249075215783,
+      "grad_norm": 0.07111615687608719,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 30972
+    },
+    {
+      "epoch": 3.182799013563502,
+      "grad_norm": 0.07623612880706787,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 30975
+    },
+    {
+      "epoch": 3.183107274969174,
+      "grad_norm": 0.0768456682562828,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 30978
+    },
+    {
+      "epoch": 3.183415536374846,
+      "grad_norm": 0.07147437334060669,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 30981
+    },
+    {
+      "epoch": 3.1837237977805177,
+      "grad_norm": 0.04732421785593033,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 30984
+    },
+    {
+      "epoch": 3.18403205918619,
+      "grad_norm": 0.04406769201159477,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 30987
+    },
+    {
+      "epoch": 3.184340320591862,
+      "grad_norm": 0.03760458901524544,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 30990
+    },
+    {
+      "epoch": 3.184648581997534,
+      "grad_norm": 0.051892757415771484,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 30993
+    },
+    {
+      "epoch": 3.184956843403206,
+      "grad_norm": 0.03961913660168648,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 30996
+    },
+    {
+      "epoch": 3.185265104808878,
+      "grad_norm": 0.07263363152742386,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 30999
+    },
+    {
+      "epoch": 3.18557336621455,
+      "grad_norm": 0.08618062734603882,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 31002
+    },
+    {
+      "epoch": 3.1858816276202218,
+      "grad_norm": 0.08712664991617203,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 31005
+    },
+    {
+      "epoch": 3.186189889025894,
+      "grad_norm": 0.05723334103822708,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 31008
+    },
+    {
+      "epoch": 3.186498150431566,
+      "grad_norm": 0.06170898675918579,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 31011
+    },
+    {
+      "epoch": 3.186806411837238,
+      "grad_norm": 0.04730013385415077,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 31014
+    },
+    {
+      "epoch": 3.18711467324291,
+      "grad_norm": 0.0381946824491024,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 31017
+    },
+    {
+      "epoch": 3.187422934648582,
+      "grad_norm": 0.09591019153594971,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 31020
+    },
+    {
+      "epoch": 3.187731196054254,
+      "grad_norm": 0.047843087464571,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 31023
+    },
+    {
+      "epoch": 3.188039457459926,
+      "grad_norm": 0.04370959475636482,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 31026
+    },
+    {
+      "epoch": 3.188347718865598,
+      "grad_norm": 0.060692187398672104,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 31029
+    },
+    {
+      "epoch": 3.18865598027127,
+      "grad_norm": 0.05906793847680092,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 31032
+    },
+    {
+      "epoch": 3.188964241676942,
+      "grad_norm": 0.06203675642609596,
+      "learning_rate": 0.01,
+      "loss": 2.0325,
+      "step": 31035
+    },
+    {
+      "epoch": 3.189272503082614,
+      "grad_norm": 0.07943592220544815,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 31038
+    },
+    {
+      "epoch": 3.189580764488286,
+      "grad_norm": 0.08803451061248779,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 31041
+    },
+    {
+      "epoch": 3.189889025893958,
+      "grad_norm": 0.07365550100803375,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 31044
+    },
+    {
+      "epoch": 3.1901972872996303,
+      "grad_norm": 0.0795077532529831,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 31047
+    },
+    {
+      "epoch": 3.190505548705302,
+      "grad_norm": 0.08341194689273834,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 31050
+    },
+    {
+      "epoch": 3.190813810110974,
+      "grad_norm": 0.05842220410704613,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 31053
+    },
+    {
+      "epoch": 3.191122071516646,
+      "grad_norm": 0.09980861842632294,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 31056
+    },
+    {
+      "epoch": 3.191430332922318,
+      "grad_norm": 0.061474163085222244,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 31059
+    },
+    {
+      "epoch": 3.1917385943279903,
+      "grad_norm": 0.06752124428749084,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 31062
+    },
+    {
+      "epoch": 3.192046855733662,
+      "grad_norm": 0.06160835176706314,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 31065
+    },
+    {
+      "epoch": 3.1923551171393343,
+      "grad_norm": 0.09516782313585281,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 31068
+    },
+    {
+      "epoch": 3.192663378545006,
+      "grad_norm": 0.049451183527708054,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 31071
+    },
+    {
+      "epoch": 3.1929716399506782,
+      "grad_norm": 0.08874432742595673,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 31074
+    },
+    {
+      "epoch": 3.19327990135635,
+      "grad_norm": 0.07393426448106766,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 31077
+    },
+    {
+      "epoch": 3.193588162762022,
+      "grad_norm": 0.09064768254756927,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 31080
+    },
+    {
+      "epoch": 3.1938964241676944,
+      "grad_norm": 0.05667688325047493,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 31083
+    },
+    {
+      "epoch": 3.194204685573366,
+      "grad_norm": 0.03858955577015877,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 31086
+    },
+    {
+      "epoch": 3.1945129469790383,
+      "grad_norm": 0.03137395530939102,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 31089
+    },
+    {
+      "epoch": 3.19482120838471,
+      "grad_norm": 0.09366928040981293,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 31092
+    },
+    {
+      "epoch": 3.1951294697903823,
+      "grad_norm": 0.05282336100935936,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 31095
+    },
+    {
+      "epoch": 3.1954377311960545,
+      "grad_norm": 0.07649802416563034,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 31098
+    },
+    {
+      "epoch": 3.195745992601726,
+      "grad_norm": 0.07249470800161362,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 31101
+    },
+    {
+      "epoch": 3.1960542540073984,
+      "grad_norm": 0.12900890409946442,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 31104
+    },
+    {
+      "epoch": 3.19636251541307,
+      "grad_norm": 0.08060257881879807,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 31107
+    },
+    {
+      "epoch": 3.1966707768187423,
+      "grad_norm": 0.07051622122526169,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 31110
+    },
+    {
+      "epoch": 3.1969790382244145,
+      "grad_norm": 0.0567597970366478,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 31113
+    },
+    {
+      "epoch": 3.1972872996300863,
+      "grad_norm": 0.053274448961019516,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 31116
+    },
+    {
+      "epoch": 3.1975955610357585,
+      "grad_norm": 0.0634993240237236,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 31119
+    },
+    {
+      "epoch": 3.1979038224414302,
+      "grad_norm": 0.060209862887859344,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 31122
+    },
+    {
+      "epoch": 3.1982120838471024,
+      "grad_norm": 0.05891990661621094,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 31125
+    },
+    {
+      "epoch": 3.198520345252774,
+      "grad_norm": 0.03743661195039749,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 31128
+    },
+    {
+      "epoch": 3.1988286066584464,
+      "grad_norm": 0.047772981226444244,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 31131
+    },
+    {
+      "epoch": 3.1991368680641186,
+      "grad_norm": 0.09134045243263245,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 31134
+    },
+    {
+      "epoch": 3.1994451294697903,
+      "grad_norm": 0.040920648723840714,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 31137
+    },
+    {
+      "epoch": 3.1997533908754625,
+      "grad_norm": 0.09245988726615906,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 31140
+    },
+    {
+      "epoch": 3.2000616522811343,
+      "grad_norm": 0.06823042035102844,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 31143
+    },
+    {
+      "epoch": 3.2003699136868065,
+      "grad_norm": 0.10324928909540176,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 31146
+    },
+    {
+      "epoch": 3.200678175092478,
+      "grad_norm": 0.09294021129608154,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 31149
+    },
+    {
+      "epoch": 3.2009864364981504,
+      "grad_norm": 0.07790663093328476,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 31152
+    },
+    {
+      "epoch": 3.2012946979038226,
+      "grad_norm": 0.041240107268095016,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 31155
+    },
+    {
+      "epoch": 3.2016029593094943,
+      "grad_norm": 0.042019765824079514,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 31158
+    },
+    {
+      "epoch": 3.2019112207151665,
+      "grad_norm": 0.03546491637825966,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 31161
+    },
+    {
+      "epoch": 3.2022194821208383,
+      "grad_norm": 0.08740135282278061,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 31164
+    },
+    {
+      "epoch": 3.2025277435265105,
+      "grad_norm": 0.06613611429929733,
+      "learning_rate": 0.01,
+      "loss": 2.0345,
+      "step": 31167
+    },
+    {
+      "epoch": 3.2028360049321827,
+      "grad_norm": 0.1089286357164383,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 31170
+    },
+    {
+      "epoch": 3.2031442663378544,
+      "grad_norm": 0.037778157740831375,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 31173
+    },
+    {
+      "epoch": 3.2034525277435266,
+      "grad_norm": 0.07214018702507019,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 31176
+    },
+    {
+      "epoch": 3.2037607891491984,
+      "grad_norm": 0.05182633548974991,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 31179
+    },
+    {
+      "epoch": 3.2040690505548706,
+      "grad_norm": 0.043808627873659134,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 31182
+    },
+    {
+      "epoch": 3.2043773119605428,
+      "grad_norm": 0.03699186071753502,
+      "learning_rate": 0.01,
+      "loss": 2.0391,
+      "step": 31185
+    },
+    {
+      "epoch": 3.2046855733662145,
+      "grad_norm": 0.10249976068735123,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 31188
+    },
+    {
+      "epoch": 3.2049938347718867,
+      "grad_norm": 0.05326079949736595,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 31191
+    },
+    {
+      "epoch": 3.2053020961775585,
+      "grad_norm": 0.09879110008478165,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 31194
+    },
+    {
+      "epoch": 3.2056103575832307,
+      "grad_norm": 0.042158786207437515,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 31197
+    },
+    {
+      "epoch": 3.2059186189889024,
+      "grad_norm": 0.13437750935554504,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 31200
+    },
+    {
+      "epoch": 3.2062268803945746,
+      "grad_norm": 0.07030022144317627,
+      "learning_rate": 0.01,
+      "loss": 2.0499,
+      "step": 31203
+    },
+    {
+      "epoch": 3.206535141800247,
+      "grad_norm": 0.044758979231119156,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 31206
+    },
+    {
+      "epoch": 3.2068434032059185,
+      "grad_norm": 0.030563069507479668,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 31209
+    },
+    {
+      "epoch": 3.2071516646115907,
+      "grad_norm": 0.047487739473581314,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 31212
+    },
+    {
+      "epoch": 3.2074599260172625,
+      "grad_norm": 0.044461995363235474,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 31215
+    },
+    {
+      "epoch": 3.2077681874229347,
+      "grad_norm": 0.16204911470413208,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 31218
+    },
+    {
+      "epoch": 3.2080764488286064,
+      "grad_norm": 0.0470188669860363,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 31221
+    },
+    {
+      "epoch": 3.2083847102342786,
+      "grad_norm": 0.04323815181851387,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 31224
+    },
+    {
+      "epoch": 3.208692971639951,
+      "grad_norm": 0.04388147220015526,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 31227
+    },
+    {
+      "epoch": 3.2090012330456226,
+      "grad_norm": 0.06811438500881195,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 31230
+    },
+    {
+      "epoch": 3.2093094944512948,
+      "grad_norm": 0.09423845261335373,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 31233
+    },
+    {
+      "epoch": 3.2096177558569665,
+      "grad_norm": 0.05107983946800232,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 31236
+    },
+    {
+      "epoch": 3.2099260172626387,
+      "grad_norm": 0.06188793480396271,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 31239
+    },
+    {
+      "epoch": 3.210234278668311,
+      "grad_norm": 0.04658438265323639,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 31242
+    },
+    {
+      "epoch": 3.2105425400739827,
+      "grad_norm": 0.05382300913333893,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 31245
+    },
+    {
+      "epoch": 3.210850801479655,
+      "grad_norm": 0.04271601140499115,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 31248
+    },
+    {
+      "epoch": 3.2111590628853266,
+      "grad_norm": 0.06256501376628876,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 31251
+    },
+    {
+      "epoch": 3.211467324290999,
+      "grad_norm": 0.03523874282836914,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 31254
+    },
+    {
+      "epoch": 3.211775585696671,
+      "grad_norm": 0.18781809508800507,
+      "learning_rate": 0.01,
+      "loss": 2.0421,
+      "step": 31257
+    },
+    {
+      "epoch": 3.2120838471023427,
+      "grad_norm": 0.11113902926445007,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 31260
+    },
+    {
+      "epoch": 3.212392108508015,
+      "grad_norm": 0.11578498035669327,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 31263
+    },
+    {
+      "epoch": 3.2127003699136867,
+      "grad_norm": 0.06620439887046814,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 31266
+    },
+    {
+      "epoch": 3.213008631319359,
+      "grad_norm": 0.04588307812809944,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 31269
+    },
+    {
+      "epoch": 3.2133168927250306,
+      "grad_norm": 0.06933780014514923,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 31272
+    },
+    {
+      "epoch": 3.213625154130703,
+      "grad_norm": 0.03479130566120148,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 31275
+    },
+    {
+      "epoch": 3.213933415536375,
+      "grad_norm": 0.05823906138539314,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 31278
+    },
+    {
+      "epoch": 3.2142416769420468,
+      "grad_norm": 0.044943809509277344,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 31281
+    },
+    {
+      "epoch": 3.214549938347719,
+      "grad_norm": 0.05428524687886238,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 31284
+    },
+    {
+      "epoch": 3.2148581997533907,
+      "grad_norm": 0.07103761285543442,
+      "learning_rate": 0.01,
+      "loss": 2.0401,
+      "step": 31287
+    },
+    {
+      "epoch": 3.215166461159063,
+      "grad_norm": 0.1237725242972374,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 31290
+    },
+    {
+      "epoch": 3.215474722564735,
+      "grad_norm": 0.03638492897152901,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 31293
+    },
+    {
+      "epoch": 3.215782983970407,
+      "grad_norm": 0.05026063695549965,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 31296
+    },
+    {
+      "epoch": 3.216091245376079,
+      "grad_norm": 0.09602409601211548,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 31299
+    },
+    {
+      "epoch": 3.216399506781751,
+      "grad_norm": 0.12711849808692932,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 31302
+    },
+    {
+      "epoch": 3.216707768187423,
+      "grad_norm": 0.051612287759780884,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 31305
+    },
+    {
+      "epoch": 3.2170160295930947,
+      "grad_norm": 0.055910855531692505,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 31308
+    },
+    {
+      "epoch": 3.217324290998767,
+      "grad_norm": 0.053995631635189056,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 31311
+    },
+    {
+      "epoch": 3.217632552404439,
+      "grad_norm": 0.03721768036484718,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 31314
+    },
+    {
+      "epoch": 3.217940813810111,
+      "grad_norm": 0.04018721356987953,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 31317
+    },
+    {
+      "epoch": 3.218249075215783,
+      "grad_norm": 0.06476118415594101,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 31320
+    },
+    {
+      "epoch": 3.218557336621455,
+      "grad_norm": 0.09766072034835815,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 31323
+    },
+    {
+      "epoch": 3.218865598027127,
+      "grad_norm": 0.0436365082859993,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 31326
+    },
+    {
+      "epoch": 3.219173859432799,
+      "grad_norm": 0.04617559164762497,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 31329
+    },
+    {
+      "epoch": 3.219482120838471,
+      "grad_norm": 0.0487680621445179,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 31332
+    },
+    {
+      "epoch": 3.219790382244143,
+      "grad_norm": 0.036623213440179825,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 31335
+    },
+    {
+      "epoch": 3.220098643649815,
+      "grad_norm": 0.04713229089975357,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 31338
+    },
+    {
+      "epoch": 3.220406905055487,
+      "grad_norm": 0.03848060593008995,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 31341
+    },
+    {
+      "epoch": 3.220715166461159,
+      "grad_norm": 0.03674410656094551,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 31344
+    },
+    {
+      "epoch": 3.221023427866831,
+      "grad_norm": 0.09635547548532486,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 31347
+    },
+    {
+      "epoch": 3.2213316892725032,
+      "grad_norm": 0.06311628222465515,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 31350
+    },
+    {
+      "epoch": 3.221639950678175,
+      "grad_norm": 0.05867060646414757,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 31353
+    },
+    {
+      "epoch": 3.221948212083847,
+      "grad_norm": 0.0816822350025177,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 31356
+    },
+    {
+      "epoch": 3.222256473489519,
+      "grad_norm": 0.0825878456234932,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 31359
+    },
+    {
+      "epoch": 3.222564734895191,
+      "grad_norm": 0.06253322958946228,
+      "learning_rate": 0.01,
+      "loss": 2.0461,
+      "step": 31362
+    },
+    {
+      "epoch": 3.2228729963008633,
+      "grad_norm": 0.07637360692024231,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 31365
+    },
+    {
+      "epoch": 3.223181257706535,
+      "grad_norm": 0.081746406853199,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 31368
+    },
+    {
+      "epoch": 3.2234895191122073,
+      "grad_norm": 0.11683212220668793,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 31371
+    },
+    {
+      "epoch": 3.223797780517879,
+      "grad_norm": 0.07254528999328613,
+      "learning_rate": 0.01,
+      "loss": 2.0392,
+      "step": 31374
+    },
+    {
+      "epoch": 3.224106041923551,
+      "grad_norm": 0.05394696444272995,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 31377
+    },
+    {
+      "epoch": 3.2244143033292234,
+      "grad_norm": 0.044905390590429306,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 31380
+    },
+    {
+      "epoch": 3.224722564734895,
+      "grad_norm": 0.07633423805236816,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 31383
+    },
+    {
+      "epoch": 3.2250308261405674,
+      "grad_norm": 0.03892616555094719,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 31386
+    },
+    {
+      "epoch": 3.225339087546239,
+      "grad_norm": 0.03514908254146576,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 31389
+    },
+    {
+      "epoch": 3.2256473489519113,
+      "grad_norm": 0.07658538222312927,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 31392
+    },
+    {
+      "epoch": 3.225955610357583,
+      "grad_norm": 0.1004643440246582,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 31395
+    },
+    {
+      "epoch": 3.2262638717632552,
+      "grad_norm": 0.08478312194347382,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 31398
+    },
+    {
+      "epoch": 3.2265721331689274,
+      "grad_norm": 0.05344710871577263,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 31401
+    },
+    {
+      "epoch": 3.226880394574599,
+      "grad_norm": 0.05862101912498474,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 31404
+    },
+    {
+      "epoch": 3.2271886559802714,
+      "grad_norm": 0.03430565074086189,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 31407
+    },
+    {
+      "epoch": 3.227496917385943,
+      "grad_norm": 0.03822310268878937,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 31410
+    },
+    {
+      "epoch": 3.2278051787916153,
+      "grad_norm": 0.04391561448574066,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 31413
+    },
+    {
+      "epoch": 3.228113440197287,
+      "grad_norm": 0.053595174103975296,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 31416
+    },
+    {
+      "epoch": 3.2284217016029593,
+      "grad_norm": 0.06710030138492584,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 31419
+    },
+    {
+      "epoch": 3.2287299630086315,
+      "grad_norm": 0.19077260792255402,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 31422
+    },
+    {
+      "epoch": 3.229038224414303,
+      "grad_norm": 0.050228483974933624,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 31425
+    },
+    {
+      "epoch": 3.2293464858199754,
+      "grad_norm": 0.047246966511011124,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 31428
+    },
+    {
+      "epoch": 3.229654747225647,
+      "grad_norm": 0.03976801037788391,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 31431
+    },
+    {
+      "epoch": 3.2299630086313194,
+      "grad_norm": 0.052110929042100906,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 31434
+    },
+    {
+      "epoch": 3.2302712700369915,
+      "grad_norm": 0.1090363934636116,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 31437
+    },
+    {
+      "epoch": 3.2305795314426633,
+      "grad_norm": 0.13336141407489777,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 31440
+    },
+    {
+      "epoch": 3.2308877928483355,
+      "grad_norm": 0.11287815868854523,
+      "learning_rate": 0.01,
+      "loss": 2.0384,
+      "step": 31443
+    },
+    {
+      "epoch": 3.2311960542540072,
+      "grad_norm": 0.05775724723935127,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 31446
+    },
+    {
+      "epoch": 3.2315043156596794,
+      "grad_norm": 0.0489221066236496,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 31449
+    },
+    {
+      "epoch": 3.2318125770653516,
+      "grad_norm": 0.10708906501531601,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 31452
+    },
+    {
+      "epoch": 3.2321208384710234,
+      "grad_norm": 0.04961223527789116,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 31455
+    },
+    {
+      "epoch": 3.2324290998766956,
+      "grad_norm": 0.04106178134679794,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 31458
+    },
+    {
+      "epoch": 3.2327373612823673,
+      "grad_norm": 0.051446568220853806,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 31461
+    },
+    {
+      "epoch": 3.2330456226880395,
+      "grad_norm": 0.046735066920518875,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 31464
+    },
+    {
+      "epoch": 3.2333538840937113,
+      "grad_norm": 0.04740822687745094,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 31467
+    },
+    {
+      "epoch": 3.2336621454993835,
+      "grad_norm": 0.03593340888619423,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 31470
+    },
+    {
+      "epoch": 3.2339704069050557,
+      "grad_norm": 0.058127082884311676,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 31473
+    },
+    {
+      "epoch": 3.2342786683107274,
+      "grad_norm": 0.05138585716485977,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 31476
+    },
+    {
+      "epoch": 3.2345869297163996,
+      "grad_norm": 0.04029039293527603,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 31479
+    },
+    {
+      "epoch": 3.2348951911220714,
+      "grad_norm": 0.05703110247850418,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 31482
+    },
+    {
+      "epoch": 3.2352034525277436,
+      "grad_norm": 0.06218143180012703,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 31485
+    },
+    {
+      "epoch": 3.2355117139334153,
+      "grad_norm": 0.07277306169271469,
+      "learning_rate": 0.01,
+      "loss": 2.0413,
+      "step": 31488
+    },
+    {
+      "epoch": 3.2358199753390875,
+      "grad_norm": 0.08520001918077469,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 31491
+    },
+    {
+      "epoch": 3.2361282367447597,
+      "grad_norm": 0.062044426798820496,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 31494
+    },
+    {
+      "epoch": 3.2364364981504314,
+      "grad_norm": 0.062475502490997314,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 31497
+    },
+    {
+      "epoch": 3.2367447595561036,
+      "grad_norm": 0.05683014541864395,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 31500
+    },
+    {
+      "epoch": 3.2370530209617754,
+      "grad_norm": 0.0409809909760952,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 31503
+    },
+    {
+      "epoch": 3.2373612823674476,
+      "grad_norm": 0.05398055911064148,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 31506
+    },
+    {
+      "epoch": 3.2376695437731198,
+      "grad_norm": 0.09552880376577377,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 31509
+    },
+    {
+      "epoch": 3.2379778051787915,
+      "grad_norm": 0.042877551168203354,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 31512
+    },
+    {
+      "epoch": 3.2382860665844637,
+      "grad_norm": 0.03086467832326889,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 31515
+    },
+    {
+      "epoch": 3.2385943279901355,
+      "grad_norm": 0.03575371578335762,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 31518
+    },
+    {
+      "epoch": 3.2389025893958077,
+      "grad_norm": 0.04962699115276337,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 31521
+    },
+    {
+      "epoch": 3.23921085080148,
+      "grad_norm": 0.07324356585741043,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 31524
+    },
+    {
+      "epoch": 3.2395191122071516,
+      "grad_norm": 0.08169112354516983,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 31527
+    },
+    {
+      "epoch": 3.239827373612824,
+      "grad_norm": 0.04129362106323242,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 31530
+    },
+    {
+      "epoch": 3.2401356350184956,
+      "grad_norm": 0.04921339079737663,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 31533
+    },
+    {
+      "epoch": 3.2404438964241677,
+      "grad_norm": 0.048255033791065216,
+      "learning_rate": 0.01,
+      "loss": 2.0486,
+      "step": 31536
+    },
+    {
+      "epoch": 3.2407521578298395,
+      "grad_norm": 0.0420132540166378,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 31539
+    },
+    {
+      "epoch": 3.2410604192355117,
+      "grad_norm": 0.037761542946100235,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 31542
+    },
+    {
+      "epoch": 3.241368680641184,
+      "grad_norm": 0.059902340173721313,
+      "learning_rate": 0.01,
+      "loss": 2.0371,
+      "step": 31545
+    },
+    {
+      "epoch": 3.2416769420468556,
+      "grad_norm": 0.07208621501922607,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 31548
+    },
+    {
+      "epoch": 3.241985203452528,
+      "grad_norm": 0.039220135658979416,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 31551
+    },
+    {
+      "epoch": 3.2422934648581996,
+      "grad_norm": 0.09032812714576721,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 31554
+    },
+    {
+      "epoch": 3.2426017262638718,
+      "grad_norm": 0.04668070003390312,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 31557
+    },
+    {
+      "epoch": 3.242909987669544,
+      "grad_norm": 0.07167590409517288,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 31560
+    },
+    {
+      "epoch": 3.2432182490752157,
+      "grad_norm": 0.09092319756746292,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 31563
+    },
+    {
+      "epoch": 3.243526510480888,
+      "grad_norm": 0.05797998234629631,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 31566
+    },
+    {
+      "epoch": 3.2438347718865597,
+      "grad_norm": 0.05728116258978844,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 31569
+    },
+    {
+      "epoch": 3.244143033292232,
+      "grad_norm": 0.10221456736326218,
+      "learning_rate": 0.01,
+      "loss": 2.0452,
+      "step": 31572
+    },
+    {
+      "epoch": 3.244451294697904,
+      "grad_norm": 0.0785583034157753,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 31575
+    },
+    {
+      "epoch": 3.244759556103576,
+      "grad_norm": 0.06105490028858185,
+      "learning_rate": 0.01,
+      "loss": 2.035,
+      "step": 31578
+    },
+    {
+      "epoch": 3.245067817509248,
+      "grad_norm": 0.031959742307662964,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 31581
+    },
+    {
+      "epoch": 3.2453760789149197,
+      "grad_norm": 0.07373019307851791,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 31584
+    },
+    {
+      "epoch": 3.245684340320592,
+      "grad_norm": 0.05989330634474754,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 31587
+    },
+    {
+      "epoch": 3.2459926017262637,
+      "grad_norm": 0.0927957072854042,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 31590
+    },
+    {
+      "epoch": 3.246300863131936,
+      "grad_norm": 0.048555102199316025,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 31593
+    },
+    {
+      "epoch": 3.246609124537608,
+      "grad_norm": 0.05515364184975624,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 31596
+    },
+    {
+      "epoch": 3.24691738594328,
+      "grad_norm": 0.04753278195858002,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 31599
+    },
+    {
+      "epoch": 3.247225647348952,
+      "grad_norm": 0.06210014224052429,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 31602
+    },
+    {
+      "epoch": 3.2475339087546238,
+      "grad_norm": 0.060913342982530594,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 31605
+    },
+    {
+      "epoch": 3.247842170160296,
+      "grad_norm": 0.08301867544651031,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 31608
+    },
+    {
+      "epoch": 3.2481504315659677,
+      "grad_norm": 0.06884342432022095,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 31611
+    },
+    {
+      "epoch": 3.24845869297164,
+      "grad_norm": 0.046835094690322876,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 31614
+    },
+    {
+      "epoch": 3.248766954377312,
+      "grad_norm": 0.11094243824481964,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 31617
+    },
+    {
+      "epoch": 3.249075215782984,
+      "grad_norm": 0.0603213869035244,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 31620
+    },
+    {
+      "epoch": 3.249383477188656,
+      "grad_norm": 0.07283317297697067,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 31623
+    },
+    {
+      "epoch": 3.249691738594328,
+      "grad_norm": 0.05507282167673111,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 31626
+    },
+    {
+      "epoch": 3.25,
+      "grad_norm": 0.039191924035549164,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 31629
+    },
+    {
+      "epoch": 3.250308261405672,
+      "grad_norm": 0.1467886120080948,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 31632
+    },
+    {
+      "epoch": 3.250616522811344,
+      "grad_norm": 0.06955720484256744,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 31635
+    },
+    {
+      "epoch": 3.250924784217016,
+      "grad_norm": 0.0634884312748909,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 31638
+    },
+    {
+      "epoch": 3.251233045622688,
+      "grad_norm": 0.035188958048820496,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 31641
+    },
+    {
+      "epoch": 3.25154130702836,
+      "grad_norm": 0.07601841539144516,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 31644
+    },
+    {
+      "epoch": 3.2518495684340323,
+      "grad_norm": 0.06929823011159897,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 31647
+    },
+    {
+      "epoch": 3.252157829839704,
+      "grad_norm": 0.06300003081560135,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 31650
+    },
+    {
+      "epoch": 3.2524660912453762,
+      "grad_norm": 0.04881738871335983,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 31653
+    },
+    {
+      "epoch": 3.252774352651048,
+      "grad_norm": 0.04084230959415436,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 31656
+    },
+    {
+      "epoch": 3.25308261405672,
+      "grad_norm": 0.10158465802669525,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 31659
+    },
+    {
+      "epoch": 3.253390875462392,
+      "grad_norm": 0.04703471064567566,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 31662
+    },
+    {
+      "epoch": 3.253699136868064,
+      "grad_norm": 0.11601495742797852,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 31665
+    },
+    {
+      "epoch": 3.2540073982737363,
+      "grad_norm": 0.037081558257341385,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 31668
+    },
+    {
+      "epoch": 3.254315659679408,
+      "grad_norm": 0.03952954337000847,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 31671
+    },
+    {
+      "epoch": 3.2546239210850803,
+      "grad_norm": 0.09437094628810883,
+      "learning_rate": 0.01,
+      "loss": 2.0528,
+      "step": 31674
+    },
+    {
+      "epoch": 3.254932182490752,
+      "grad_norm": 0.04648669436573982,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 31677
+    },
+    {
+      "epoch": 3.255240443896424,
+      "grad_norm": 0.06830704212188721,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 31680
+    },
+    {
+      "epoch": 3.255548705302096,
+      "grad_norm": 0.039575349539518356,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 31683
+    },
+    {
+      "epoch": 3.255856966707768,
+      "grad_norm": 0.046931225806474686,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 31686
+    },
+    {
+      "epoch": 3.2561652281134403,
+      "grad_norm": 0.039921220391988754,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 31689
+    },
+    {
+      "epoch": 3.256473489519112,
+      "grad_norm": 0.1725001186132431,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 31692
+    },
+    {
+      "epoch": 3.2567817509247843,
+      "grad_norm": 0.10149878263473511,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 31695
+    },
+    {
+      "epoch": 3.2570900123304565,
+      "grad_norm": 0.0641789436340332,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 31698
+    },
+    {
+      "epoch": 3.2573982737361282,
+      "grad_norm": 0.09631607681512833,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 31701
+    },
+    {
+      "epoch": 3.2577065351418004,
+      "grad_norm": 0.04941624775528908,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 31704
+    },
+    {
+      "epoch": 3.258014796547472,
+      "grad_norm": 0.03704220801591873,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 31707
+    },
+    {
+      "epoch": 3.2583230579531444,
+      "grad_norm": 0.05716734007000923,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 31710
+    },
+    {
+      "epoch": 3.258631319358816,
+      "grad_norm": 0.04187513142824173,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 31713
+    },
+    {
+      "epoch": 3.2589395807644883,
+      "grad_norm": 0.04240023344755173,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 31716
+    },
+    {
+      "epoch": 3.2592478421701605,
+      "grad_norm": 0.0434168316423893,
+      "learning_rate": 0.01,
+      "loss": 2.0293,
+      "step": 31719
+    },
+    {
+      "epoch": 3.2595561035758323,
+      "grad_norm": 0.10827293992042542,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 31722
+    },
+    {
+      "epoch": 3.2598643649815044,
+      "grad_norm": 0.04294963926076889,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 31725
+    },
+    {
+      "epoch": 3.260172626387176,
+      "grad_norm": 0.11746285110712051,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 31728
+    },
+    {
+      "epoch": 3.2604808877928484,
+      "grad_norm": 0.13203038275241852,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 31731
+    },
+    {
+      "epoch": 3.26078914919852,
+      "grad_norm": 0.12324203550815582,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 31734
+    },
+    {
+      "epoch": 3.2610974106041923,
+      "grad_norm": 0.09455161541700363,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 31737
+    },
+    {
+      "epoch": 3.2614056720098645,
+      "grad_norm": 0.0551525354385376,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 31740
+    },
+    {
+      "epoch": 3.2617139334155363,
+      "grad_norm": 0.04528261721134186,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 31743
+    },
+    {
+      "epoch": 3.2620221948212085,
+      "grad_norm": 0.049497511237859726,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 31746
+    },
+    {
+      "epoch": 3.2623304562268802,
+      "grad_norm": 0.04797462373971939,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 31749
+    },
+    {
+      "epoch": 3.2626387176325524,
+      "grad_norm": 0.047107502818107605,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 31752
+    },
+    {
+      "epoch": 3.262946979038224,
+      "grad_norm": 0.03374217078089714,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 31755
+    },
+    {
+      "epoch": 3.2632552404438964,
+      "grad_norm": 0.05429157614707947,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 31758
+    },
+    {
+      "epoch": 3.2635635018495686,
+      "grad_norm": 0.12066765129566193,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 31761
+    },
+    {
+      "epoch": 3.2638717632552403,
+      "grad_norm": 0.15858452022075653,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 31764
+    },
+    {
+      "epoch": 3.2641800246609125,
+      "grad_norm": 0.05770573392510414,
+      "learning_rate": 0.01,
+      "loss": 2.0513,
+      "step": 31767
+    },
+    {
+      "epoch": 3.2644882860665847,
+      "grad_norm": 0.053706735372543335,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 31770
+    },
+    {
+      "epoch": 3.2647965474722564,
+      "grad_norm": 0.03511708602309227,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 31773
+    },
+    {
+      "epoch": 3.2651048088779286,
+      "grad_norm": 0.036902979016304016,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 31776
+    },
+    {
+      "epoch": 3.2654130702836004,
+      "grad_norm": 0.0450621172785759,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 31779
+    },
+    {
+      "epoch": 3.2657213316892726,
+      "grad_norm": 0.07691927254199982,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 31782
+    },
+    {
+      "epoch": 3.2660295930949443,
+      "grad_norm": 0.09160054475069046,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 31785
+    },
+    {
+      "epoch": 3.2663378545006165,
+      "grad_norm": 0.09052203595638275,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 31788
+    },
+    {
+      "epoch": 3.2666461159062887,
+      "grad_norm": 0.08806884288787842,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 31791
+    },
+    {
+      "epoch": 3.2669543773119605,
+      "grad_norm": 0.05619393661618233,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 31794
+    },
+    {
+      "epoch": 3.2672626387176327,
+      "grad_norm": 0.05563674122095108,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 31797
+    },
+    {
+      "epoch": 3.2675709001233044,
+      "grad_norm": 0.058199040591716766,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 31800
+    },
+    {
+      "epoch": 3.2678791615289766,
+      "grad_norm": 0.06011686101555824,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 31803
+    },
+    {
+      "epoch": 3.2681874229346484,
+      "grad_norm": 0.05391063541173935,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 31806
+    },
+    {
+      "epoch": 3.2684956843403206,
+      "grad_norm": 0.1030382513999939,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 31809
+    },
+    {
+      "epoch": 3.2688039457459928,
+      "grad_norm": 0.03978987783193588,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 31812
+    },
+    {
+      "epoch": 3.2691122071516645,
+      "grad_norm": 0.08974360674619675,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 31815
+    },
+    {
+      "epoch": 3.2694204685573367,
+      "grad_norm": 0.07530324161052704,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 31818
+    },
+    {
+      "epoch": 3.2697287299630085,
+      "grad_norm": 0.04715275764465332,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 31821
+    },
+    {
+      "epoch": 3.2700369913686806,
+      "grad_norm": 0.050625238567590714,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 31824
+    },
+    {
+      "epoch": 3.270345252774353,
+      "grad_norm": 0.05567210912704468,
+      "learning_rate": 0.01,
+      "loss": 2.0554,
+      "step": 31827
+    },
+    {
+      "epoch": 3.2706535141800246,
+      "grad_norm": 0.09813915193080902,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 31830
+    },
+    {
+      "epoch": 3.270961775585697,
+      "grad_norm": 0.08543343842029572,
+      "learning_rate": 0.01,
+      "loss": 2.0405,
+      "step": 31833
+    },
+    {
+      "epoch": 3.2712700369913685,
+      "grad_norm": 0.03340763971209526,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 31836
+    },
+    {
+      "epoch": 3.2715782983970407,
+      "grad_norm": 0.0335637666285038,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 31839
+    },
+    {
+      "epoch": 3.271886559802713,
+      "grad_norm": 0.06983290612697601,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 31842
+    },
+    {
+      "epoch": 3.2721948212083847,
+      "grad_norm": 0.1157732903957367,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 31845
+    },
+    {
+      "epoch": 3.272503082614057,
+      "grad_norm": 0.11864369362592697,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 31848
+    },
+    {
+      "epoch": 3.2728113440197286,
+      "grad_norm": 0.09083148092031479,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 31851
+    },
+    {
+      "epoch": 3.273119605425401,
+      "grad_norm": 0.06563573330640793,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 31854
+    },
+    {
+      "epoch": 3.2734278668310726,
+      "grad_norm": 0.043879635632038116,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 31857
+    },
+    {
+      "epoch": 3.2737361282367448,
+      "grad_norm": 0.03548846393823624,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 31860
+    },
+    {
+      "epoch": 3.274044389642417,
+      "grad_norm": 0.036084435880184174,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 31863
+    },
+    {
+      "epoch": 3.2743526510480887,
+      "grad_norm": 0.03813619166612625,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 31866
+    },
+    {
+      "epoch": 3.274660912453761,
+      "grad_norm": 0.05276734009385109,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 31869
+    },
+    {
+      "epoch": 3.2749691738594326,
+      "grad_norm": 0.07557803392410278,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 31872
+    },
+    {
+      "epoch": 3.275277435265105,
+      "grad_norm": 0.060760460793972015,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 31875
+    },
+    {
+      "epoch": 3.2755856966707766,
+      "grad_norm": 0.11498884111642838,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 31878
+    },
+    {
+      "epoch": 3.275893958076449,
+      "grad_norm": 0.12193593382835388,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 31881
+    },
+    {
+      "epoch": 3.276202219482121,
+      "grad_norm": 0.10637890547513962,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 31884
+    },
+    {
+      "epoch": 3.2765104808877927,
+      "grad_norm": 0.08633279800415039,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 31887
+    },
+    {
+      "epoch": 3.276818742293465,
+      "grad_norm": 0.08105628192424774,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 31890
+    },
+    {
+      "epoch": 3.2771270036991367,
+      "grad_norm": 0.03692932799458504,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 31893
+    },
+    {
+      "epoch": 3.277435265104809,
+      "grad_norm": 0.03064770996570587,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 31896
+    },
+    {
+      "epoch": 3.277743526510481,
+      "grad_norm": 0.038218267261981964,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 31899
+    },
+    {
+      "epoch": 3.278051787916153,
+      "grad_norm": 0.05573924258351326,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 31902
+    },
+    {
+      "epoch": 3.278360049321825,
+      "grad_norm": 0.14293062686920166,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 31905
+    },
+    {
+      "epoch": 3.2786683107274968,
+      "grad_norm": 0.082049660384655,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 31908
+    },
+    {
+      "epoch": 3.278976572133169,
+      "grad_norm": 0.08122528344392776,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 31911
+    },
+    {
+      "epoch": 3.279284833538841,
+      "grad_norm": 0.054671067744493484,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 31914
+    },
+    {
+      "epoch": 3.279593094944513,
+      "grad_norm": 0.11112997680902481,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 31917
+    },
+    {
+      "epoch": 3.279901356350185,
+      "grad_norm": 0.12811101973056793,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 31920
+    },
+    {
+      "epoch": 3.280209617755857,
+      "grad_norm": 0.07637016475200653,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 31923
+    },
+    {
+      "epoch": 3.280517879161529,
+      "grad_norm": 0.03716239705681801,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 31926
+    },
+    {
+      "epoch": 3.280826140567201,
+      "grad_norm": 0.03950963541865349,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 31929
+    },
+    {
+      "epoch": 3.281134401972873,
+      "grad_norm": 0.09012940526008606,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 31932
+    },
+    {
+      "epoch": 3.281442663378545,
+      "grad_norm": 0.04000500217080116,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 31935
+    },
+    {
+      "epoch": 3.281750924784217,
+      "grad_norm": 0.06817129254341125,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 31938
+    },
+    {
+      "epoch": 3.282059186189889,
+      "grad_norm": 0.14285409450531006,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 31941
+    },
+    {
+      "epoch": 3.282367447595561,
+      "grad_norm": 0.03885696083307266,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 31944
+    },
+    {
+      "epoch": 3.282675709001233,
+      "grad_norm": 0.047379713505506516,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 31947
+    },
+    {
+      "epoch": 3.282983970406905,
+      "grad_norm": 0.05476104095578194,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 31950
+    },
+    {
+      "epoch": 3.283292231812577,
+      "grad_norm": 0.03158089146018028,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 31953
+    },
+    {
+      "epoch": 3.283600493218249,
+      "grad_norm": 0.04537857696413994,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 31956
+    },
+    {
+      "epoch": 3.283908754623921,
+      "grad_norm": 0.1525343656539917,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 31959
+    },
+    {
+      "epoch": 3.284217016029593,
+      "grad_norm": 0.11916167289018631,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 31962
+    },
+    {
+      "epoch": 3.2845252774352653,
+      "grad_norm": 0.09215250611305237,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 31965
+    },
+    {
+      "epoch": 3.284833538840937,
+      "grad_norm": 0.053620483726263046,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 31968
+    },
+    {
+      "epoch": 3.2851418002466093,
+      "grad_norm": 0.03575912117958069,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 31971
+    },
+    {
+      "epoch": 3.285450061652281,
+      "grad_norm": 0.09148744493722916,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 31974
+    },
+    {
+      "epoch": 3.2857583230579532,
+      "grad_norm": 0.06433127075433731,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 31977
+    },
+    {
+      "epoch": 3.286066584463625,
+      "grad_norm": 0.04605916887521744,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 31980
+    },
+    {
+      "epoch": 3.286374845869297,
+      "grad_norm": 0.07150716334581375,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 31983
+    },
+    {
+      "epoch": 3.2866831072749694,
+      "grad_norm": 0.04914524033665657,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 31986
+    },
+    {
+      "epoch": 3.286991368680641,
+      "grad_norm": 0.10281821340322495,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 31989
+    },
+    {
+      "epoch": 3.2872996300863133,
+      "grad_norm": 0.04685597866773605,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 31992
+    },
+    {
+      "epoch": 3.287607891491985,
+      "grad_norm": 0.04735150560736656,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 31995
+    },
+    {
+      "epoch": 3.2879161528976573,
+      "grad_norm": 0.07897822558879852,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 31998
+    },
+    {
+      "epoch": 3.288224414303329,
+      "grad_norm": 0.08091110736131668,
+      "learning_rate": 0.01,
+      "loss": 2.0397,
+      "step": 32001
+    },
+    {
+      "epoch": 3.288532675709001,
+      "grad_norm": 0.038828156888484955,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 32004
+    },
+    {
+      "epoch": 3.2888409371146734,
+      "grad_norm": 0.11410044133663177,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 32007
+    },
+    {
+      "epoch": 3.289149198520345,
+      "grad_norm": 0.11741339415311813,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 32010
+    },
+    {
+      "epoch": 3.2894574599260173,
+      "grad_norm": 0.054089032113552094,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 32013
+    },
+    {
+      "epoch": 3.289765721331689,
+      "grad_norm": 0.034770797938108444,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 32016
+    },
+    {
+      "epoch": 3.2900739827373613,
+      "grad_norm": 0.06313812732696533,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 32019
+    },
+    {
+      "epoch": 3.2903822441430335,
+      "grad_norm": 0.05844837799668312,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 32022
+    },
+    {
+      "epoch": 3.2906905055487052,
+      "grad_norm": 0.0374666191637516,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 32025
+    },
+    {
+      "epoch": 3.2909987669543774,
+      "grad_norm": 0.05456427484750748,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 32028
+    },
+    {
+      "epoch": 3.291307028360049,
+      "grad_norm": 0.11318197846412659,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 32031
+    },
+    {
+      "epoch": 3.2916152897657214,
+      "grad_norm": 0.04832073673605919,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 32034
+    },
+    {
+      "epoch": 3.2919235511713936,
+      "grad_norm": 0.07790713757276535,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 32037
+    },
+    {
+      "epoch": 3.2922318125770653,
+      "grad_norm": 0.05794338509440422,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 32040
+    },
+    {
+      "epoch": 3.2925400739827375,
+      "grad_norm": 0.06488461047410965,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 32043
+    },
+    {
+      "epoch": 3.2928483353884093,
+      "grad_norm": 0.10804028809070587,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 32046
+    },
+    {
+      "epoch": 3.2931565967940815,
+      "grad_norm": 0.04927225038409233,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 32049
+    },
+    {
+      "epoch": 3.293464858199753,
+      "grad_norm": 0.039539139717817307,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 32052
+    },
+    {
+      "epoch": 3.2937731196054254,
+      "grad_norm": 0.046698734164237976,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 32055
+    },
+    {
+      "epoch": 3.2940813810110976,
+      "grad_norm": 0.07978003472089767,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 32058
+    },
+    {
+      "epoch": 3.2943896424167693,
+      "grad_norm": 0.05909251049160957,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 32061
+    },
+    {
+      "epoch": 3.2946979038224415,
+      "grad_norm": 0.09814689308404922,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 32064
+    },
+    {
+      "epoch": 3.2950061652281133,
+      "grad_norm": 0.07647134363651276,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 32067
+    },
+    {
+      "epoch": 3.2953144266337855,
+      "grad_norm": 0.09203072637319565,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 32070
+    },
+    {
+      "epoch": 3.2956226880394572,
+      "grad_norm": 0.14764010906219482,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 32073
+    },
+    {
+      "epoch": 3.2959309494451294,
+      "grad_norm": 0.07483326643705368,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 32076
+    },
+    {
+      "epoch": 3.2962392108508016,
+      "grad_norm": 0.03683464601635933,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 32079
+    },
+    {
+      "epoch": 3.2965474722564734,
+      "grad_norm": 0.0658450499176979,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 32082
+    },
+    {
+      "epoch": 3.2968557336621456,
+      "grad_norm": 0.05505736172199249,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 32085
+    },
+    {
+      "epoch": 3.2971639950678173,
+      "grad_norm": 0.07295443117618561,
+      "learning_rate": 0.01,
+      "loss": 2.0444,
+      "step": 32088
+    },
+    {
+      "epoch": 3.2974722564734895,
+      "grad_norm": 0.03865521401166916,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 32091
+    },
+    {
+      "epoch": 3.2977805178791617,
+      "grad_norm": 0.12185568362474442,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 32094
+    },
+    {
+      "epoch": 3.2980887792848335,
+      "grad_norm": 0.04646170511841774,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 32097
+    },
+    {
+      "epoch": 3.2983970406905057,
+      "grad_norm": 0.10177022218704224,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 32100
+    },
+    {
+      "epoch": 3.2987053020961774,
+      "grad_norm": 0.04585393890738487,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 32103
+    },
+    {
+      "epoch": 3.2990135635018496,
+      "grad_norm": 0.1089714989066124,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 32106
+    },
+    {
+      "epoch": 3.299321824907522,
+      "grad_norm": 0.04579438269138336,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 32109
+    },
+    {
+      "epoch": 3.2996300863131935,
+      "grad_norm": 0.0699036568403244,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 32112
+    },
+    {
+      "epoch": 3.2999383477188657,
+      "grad_norm": 0.10070015490055084,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 32115
+    },
+    {
+      "epoch": 3.3002466091245375,
+      "grad_norm": 0.08273176848888397,
+      "learning_rate": 0.01,
+      "loss": 2.0368,
+      "step": 32118
+    },
+    {
+      "epoch": 3.3005548705302097,
+      "grad_norm": 0.08648907393217087,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 32121
+    },
+    {
+      "epoch": 3.3008631319358814,
+      "grad_norm": 0.059946708381175995,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 32124
+    },
+    {
+      "epoch": 3.3011713933415536,
+      "grad_norm": 0.034971099346876144,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 32127
+    },
+    {
+      "epoch": 3.301479654747226,
+      "grad_norm": 0.12256456911563873,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 32130
+    },
+    {
+      "epoch": 3.3017879161528976,
+      "grad_norm": 0.041138794273138046,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 32133
+    },
+    {
+      "epoch": 3.3020961775585698,
+      "grad_norm": 0.04145865887403488,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 32136
+    },
+    {
+      "epoch": 3.3024044389642415,
+      "grad_norm": 0.07052136212587357,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 32139
+    },
+    {
+      "epoch": 3.3027127003699137,
+      "grad_norm": 0.05553466081619263,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 32142
+    },
+    {
+      "epoch": 3.3030209617755855,
+      "grad_norm": 0.038209814578294754,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 32145
+    },
+    {
+      "epoch": 3.3033292231812577,
+      "grad_norm": 0.04882222041487694,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 32148
+    },
+    {
+      "epoch": 3.30363748458693,
+      "grad_norm": 0.11157884448766708,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 32151
+    },
+    {
+      "epoch": 3.3039457459926016,
+      "grad_norm": 0.09312507510185242,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 32154
+    },
+    {
+      "epoch": 3.304254007398274,
+      "grad_norm": 0.05357594043016434,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 32157
+    },
+    {
+      "epoch": 3.304562268803946,
+      "grad_norm": 0.07024755328893661,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 32160
+    },
+    {
+      "epoch": 3.3048705302096177,
+      "grad_norm": 0.0773010402917862,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 32163
+    },
+    {
+      "epoch": 3.30517879161529,
+      "grad_norm": 0.04807543754577637,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 32166
+    },
+    {
+      "epoch": 3.3054870530209617,
+      "grad_norm": 0.03652816265821457,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 32169
+    },
+    {
+      "epoch": 3.305795314426634,
+      "grad_norm": 0.06250651925802231,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 32172
+    },
+    {
+      "epoch": 3.3061035758323056,
+      "grad_norm": 0.08029799908399582,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 32175
+    },
+    {
+      "epoch": 3.306411837237978,
+      "grad_norm": 0.10732737928628922,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 32178
+    },
+    {
+      "epoch": 3.30672009864365,
+      "grad_norm": 0.07458434998989105,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 32181
+    },
+    {
+      "epoch": 3.3070283600493218,
+      "grad_norm": 0.0794452428817749,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 32184
+    },
+    {
+      "epoch": 3.307336621454994,
+      "grad_norm": 0.06960796564817429,
+      "learning_rate": 0.01,
+      "loss": 2.0533,
+      "step": 32187
+    },
+    {
+      "epoch": 3.3076448828606657,
+      "grad_norm": 0.09482841193675995,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 32190
+    },
+    {
+      "epoch": 3.307953144266338,
+      "grad_norm": 0.08358022570610046,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 32193
+    },
+    {
+      "epoch": 3.3082614056720097,
+      "grad_norm": 0.03823438659310341,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 32196
+    },
+    {
+      "epoch": 3.308569667077682,
+      "grad_norm": 0.07413027435541153,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 32199
+    },
+    {
+      "epoch": 3.308877928483354,
+      "grad_norm": 0.03165535256266594,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 32202
+    },
+    {
+      "epoch": 3.309186189889026,
+      "grad_norm": 0.1173541322350502,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 32205
+    },
+    {
+      "epoch": 3.309494451294698,
+      "grad_norm": 0.07542740553617477,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 32208
+    },
+    {
+      "epoch": 3.3098027127003697,
+      "grad_norm": 0.104609914124012,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 32211
+    },
+    {
+      "epoch": 3.310110974106042,
+      "grad_norm": 0.0530179999768734,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 32214
+    },
+    {
+      "epoch": 3.3104192355117137,
+      "grad_norm": 0.070836141705513,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 32217
+    },
+    {
+      "epoch": 3.310727496917386,
+      "grad_norm": 0.04056829586625099,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 32220
+    },
+    {
+      "epoch": 3.311035758323058,
+      "grad_norm": 0.10000187903642654,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 32223
+    },
+    {
+      "epoch": 3.31134401972873,
+      "grad_norm": 0.04653482139110565,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 32226
+    },
+    {
+      "epoch": 3.311652281134402,
+      "grad_norm": 0.10693076252937317,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 32229
+    },
+    {
+      "epoch": 3.311960542540074,
+      "grad_norm": 0.055378034710884094,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 32232
+    },
+    {
+      "epoch": 3.312268803945746,
+      "grad_norm": 0.037242140620946884,
+      "learning_rate": 0.01,
+      "loss": 2.0367,
+      "step": 32235
+    },
+    {
+      "epoch": 3.312577065351418,
+      "grad_norm": 0.04497090354561806,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 32238
+    },
+    {
+      "epoch": 3.31288532675709,
+      "grad_norm": 0.048384904861450195,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 32241
+    },
+    {
+      "epoch": 3.313193588162762,
+      "grad_norm": 0.05588208884000778,
+      "learning_rate": 0.01,
+      "loss": 2.0375,
+      "step": 32244
+    },
+    {
+      "epoch": 3.313501849568434,
+      "grad_norm": 0.05643755942583084,
+      "learning_rate": 0.01,
+      "loss": 2.0384,
+      "step": 32247
+    },
+    {
+      "epoch": 3.313810110974106,
+      "grad_norm": 0.03415621444582939,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 32250
+    },
+    {
+      "epoch": 3.3141183723797782,
+      "grad_norm": 0.03537356108427048,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 32253
+    },
+    {
+      "epoch": 3.31442663378545,
+      "grad_norm": 0.1299961805343628,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 32256
+    },
+    {
+      "epoch": 3.314734895191122,
+      "grad_norm": 0.05058746412396431,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 32259
+    },
+    {
+      "epoch": 3.315043156596794,
+      "grad_norm": 0.06381676346063614,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 32262
+    },
+    {
+      "epoch": 3.315351418002466,
+      "grad_norm": 0.039552103728055954,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 32265
+    },
+    {
+      "epoch": 3.315659679408138,
+      "grad_norm": 0.03872091323137283,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 32268
+    },
+    {
+      "epoch": 3.31596794081381,
+      "grad_norm": 0.04546678811311722,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 32271
+    },
+    {
+      "epoch": 3.3162762022194823,
+      "grad_norm": 0.04541923850774765,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 32274
+    },
+    {
+      "epoch": 3.316584463625154,
+      "grad_norm": 0.04341261461377144,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 32277
+    },
+    {
+      "epoch": 3.316892725030826,
+      "grad_norm": 0.10473504662513733,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 32280
+    },
+    {
+      "epoch": 3.317200986436498,
+      "grad_norm": 0.0668938085436821,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 32283
+    },
+    {
+      "epoch": 3.31750924784217,
+      "grad_norm": 0.12829263508319855,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 32286
+    },
+    {
+      "epoch": 3.3178175092478424,
+      "grad_norm": 0.05443095043301582,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 32289
+    },
+    {
+      "epoch": 3.318125770653514,
+      "grad_norm": 0.06268401443958282,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 32292
+    },
+    {
+      "epoch": 3.3184340320591863,
+      "grad_norm": 0.05700231343507767,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 32295
+    },
+    {
+      "epoch": 3.318742293464858,
+      "grad_norm": 0.07467647641897202,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 32298
+    },
+    {
+      "epoch": 3.3190505548705302,
+      "grad_norm": 0.0675196647644043,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 32301
+    },
+    {
+      "epoch": 3.3193588162762024,
+      "grad_norm": 0.044399552047252655,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 32304
+    },
+    {
+      "epoch": 3.319667077681874,
+      "grad_norm": 0.0451500304043293,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 32307
+    },
+    {
+      "epoch": 3.3199753390875464,
+      "grad_norm": 0.041543807834386826,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 32310
+    },
+    {
+      "epoch": 3.320283600493218,
+      "grad_norm": 0.08818163722753525,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 32313
+    },
+    {
+      "epoch": 3.3205918618988903,
+      "grad_norm": 0.05897468701004982,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 32316
+    },
+    {
+      "epoch": 3.320900123304562,
+      "grad_norm": 0.10485360026359558,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 32319
+    },
+    {
+      "epoch": 3.3212083847102343,
+      "grad_norm": 0.048387110233306885,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 32322
+    },
+    {
+      "epoch": 3.3215166461159065,
+      "grad_norm": 0.07535526156425476,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 32325
+    },
+    {
+      "epoch": 3.321824907521578,
+      "grad_norm": 0.04594804719090462,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 32328
+    },
+    {
+      "epoch": 3.3221331689272504,
+      "grad_norm": 0.03757678344845772,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 32331
+    },
+    {
+      "epoch": 3.322441430332922,
+      "grad_norm": 0.046382974833250046,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 32334
+    },
+    {
+      "epoch": 3.3227496917385944,
+      "grad_norm": 0.03983695060014725,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 32337
+    },
+    {
+      "epoch": 3.323057953144266,
+      "grad_norm": 0.0615588016808033,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 32340
+    },
+    {
+      "epoch": 3.3233662145499383,
+      "grad_norm": 0.10143701732158661,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 32343
+    },
+    {
+      "epoch": 3.3236744759556105,
+      "grad_norm": 0.06885727494955063,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 32346
+    },
+    {
+      "epoch": 3.3239827373612822,
+      "grad_norm": 0.10029948502779007,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 32349
+    },
+    {
+      "epoch": 3.3242909987669544,
+      "grad_norm": 0.08583710342645645,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 32352
+    },
+    {
+      "epoch": 3.3245992601726266,
+      "grad_norm": 0.08596043288707733,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 32355
+    },
+    {
+      "epoch": 3.3249075215782984,
+      "grad_norm": 0.0431043840944767,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 32358
+    },
+    {
+      "epoch": 3.3252157829839706,
+      "grad_norm": 0.044695861637592316,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 32361
+    },
+    {
+      "epoch": 3.3255240443896423,
+      "grad_norm": 0.03210937976837158,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 32364
+    },
+    {
+      "epoch": 3.3258323057953145,
+      "grad_norm": 0.03838266804814339,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 32367
+    },
+    {
+      "epoch": 3.3261405672009863,
+      "grad_norm": 0.05611315369606018,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 32370
+    },
+    {
+      "epoch": 3.3264488286066585,
+      "grad_norm": 0.10025975853204727,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 32373
+    },
+    {
+      "epoch": 3.3267570900123307,
+      "grad_norm": 0.10004039853811264,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 32376
+    },
+    {
+      "epoch": 3.3270653514180024,
+      "grad_norm": 0.0655856728553772,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 32379
+    },
+    {
+      "epoch": 3.3273736128236746,
+      "grad_norm": 0.09608997404575348,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 32382
+    },
+    {
+      "epoch": 3.3276818742293464,
+      "grad_norm": 0.14821460843086243,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 32385
+    },
+    {
+      "epoch": 3.3279901356350186,
+      "grad_norm": 0.05276164412498474,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 32388
+    },
+    {
+      "epoch": 3.3282983970406903,
+      "grad_norm": 0.049289003014564514,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 32391
+    },
+    {
+      "epoch": 3.3286066584463625,
+      "grad_norm": 0.05115703493356705,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 32394
+    },
+    {
+      "epoch": 3.3289149198520347,
+      "grad_norm": 0.04041285440325737,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 32397
+    },
+    {
+      "epoch": 3.3292231812577064,
+      "grad_norm": 0.04861520603299141,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 32400
+    },
+    {
+      "epoch": 3.3295314426633786,
+      "grad_norm": 0.1111973226070404,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 32403
+    },
+    {
+      "epoch": 3.3298397040690504,
+      "grad_norm": 0.061960045248270035,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 32406
+    },
+    {
+      "epoch": 3.3301479654747226,
+      "grad_norm": 0.1108783558011055,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 32409
+    },
+    {
+      "epoch": 3.3304562268803943,
+      "grad_norm": 0.06897444278001785,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 32412
+    },
+    {
+      "epoch": 3.3307644882860665,
+      "grad_norm": 0.046277862042188644,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 32415
+    },
+    {
+      "epoch": 3.3310727496917387,
+      "grad_norm": 0.09898782521486282,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 32418
+    },
+    {
+      "epoch": 3.3313810110974105,
+      "grad_norm": 0.059529174119234085,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 32421
+    },
+    {
+      "epoch": 3.3316892725030827,
+      "grad_norm": 0.08007462322711945,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 32424
+    },
+    {
+      "epoch": 3.331997533908755,
+      "grad_norm": 0.07130561769008636,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 32427
+    },
+    {
+      "epoch": 3.3323057953144266,
+      "grad_norm": 0.04967787116765976,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 32430
+    },
+    {
+      "epoch": 3.332614056720099,
+      "grad_norm": 0.08194708079099655,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 32433
+    },
+    {
+      "epoch": 3.3329223181257706,
+      "grad_norm": 0.056519269943237305,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 32436
+    },
+    {
+      "epoch": 3.3332305795314427,
+      "grad_norm": 0.08086001873016357,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 32439
+    },
+    {
+      "epoch": 3.3335388409371145,
+      "grad_norm": 0.04036881402134895,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 32442
+    },
+    {
+      "epoch": 3.3338471023427867,
+      "grad_norm": 0.04373360425233841,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 32445
+    },
+    {
+      "epoch": 3.334155363748459,
+      "grad_norm": 0.04562424495816231,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 32448
+    },
+    {
+      "epoch": 3.3344636251541306,
+      "grad_norm": 0.2069234549999237,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 32451
+    },
+    {
+      "epoch": 3.334771886559803,
+      "grad_norm": 0.08092590421438217,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 32454
+    },
+    {
+      "epoch": 3.3350801479654746,
+      "grad_norm": 0.059557970613241196,
+      "learning_rate": 0.01,
+      "loss": 2.0523,
+      "step": 32457
+    },
+    {
+      "epoch": 3.335388409371147,
+      "grad_norm": 0.039045874029397964,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 32460
+    },
+    {
+      "epoch": 3.3356966707768185,
+      "grad_norm": 0.03718112036585808,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 32463
+    },
+    {
+      "epoch": 3.3360049321824907,
+      "grad_norm": 0.05162828043103218,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 32466
+    },
+    {
+      "epoch": 3.336313193588163,
+      "grad_norm": 0.05825696140527725,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 32469
+    },
+    {
+      "epoch": 3.3366214549938347,
+      "grad_norm": 0.0458202064037323,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 32472
+    },
+    {
+      "epoch": 3.336929716399507,
+      "grad_norm": 0.07931084930896759,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 32475
+    },
+    {
+      "epoch": 3.3372379778051786,
+      "grad_norm": 0.05673946067690849,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 32478
+    },
+    {
+      "epoch": 3.337546239210851,
+      "grad_norm": 0.05536726489663124,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 32481
+    },
+    {
+      "epoch": 3.337854500616523,
+      "grad_norm": 0.053581688553094864,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 32484
+    },
+    {
+      "epoch": 3.3381627620221948,
+      "grad_norm": 0.057449061423540115,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 32487
+    },
+    {
+      "epoch": 3.338471023427867,
+      "grad_norm": 0.0768120214343071,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 32490
+    },
+    {
+      "epoch": 3.3387792848335387,
+      "grad_norm": 0.1322673261165619,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 32493
+    },
+    {
+      "epoch": 3.339087546239211,
+      "grad_norm": 0.10641443729400635,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 32496
+    },
+    {
+      "epoch": 3.339395807644883,
+      "grad_norm": 0.13880647718906403,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 32499
+    },
+    {
+      "epoch": 3.339704069050555,
+      "grad_norm": 0.04653307795524597,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 32502
+    },
+    {
+      "epoch": 3.340012330456227,
+      "grad_norm": 0.05031618848443031,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 32505
+    },
+    {
+      "epoch": 3.340320591861899,
+      "grad_norm": 0.042637672275304794,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 32508
+    },
+    {
+      "epoch": 3.340628853267571,
+      "grad_norm": 0.03557129204273224,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 32511
+    },
+    {
+      "epoch": 3.3409371146732427,
+      "grad_norm": 0.04271169379353523,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 32514
+    },
+    {
+      "epoch": 3.341245376078915,
+      "grad_norm": 0.06750103831291199,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 32517
+    },
+    {
+      "epoch": 3.341553637484587,
+      "grad_norm": 0.05727256461977959,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 32520
+    },
+    {
+      "epoch": 3.341861898890259,
+      "grad_norm": 0.05737854540348053,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 32523
+    },
+    {
+      "epoch": 3.342170160295931,
+      "grad_norm": 0.04525689780712128,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 32526
+    },
+    {
+      "epoch": 3.342478421701603,
+      "grad_norm": 0.038834329694509506,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 32529
+    },
+    {
+      "epoch": 3.342786683107275,
+      "grad_norm": 0.14812909066677094,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 32532
+    },
+    {
+      "epoch": 3.3430949445129468,
+      "grad_norm": 0.09346423298120499,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 32535
+    },
+    {
+      "epoch": 3.343403205918619,
+      "grad_norm": 0.03221321851015091,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 32538
+    },
+    {
+      "epoch": 3.343711467324291,
+      "grad_norm": 0.05457564815878868,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 32541
+    },
+    {
+      "epoch": 3.344019728729963,
+      "grad_norm": 0.05203322321176529,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 32544
+    },
+    {
+      "epoch": 3.344327990135635,
+      "grad_norm": 0.05318872258067131,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 32547
+    },
+    {
+      "epoch": 3.344636251541307,
+      "grad_norm": 0.034635335206985474,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 32550
+    },
+    {
+      "epoch": 3.344944512946979,
+      "grad_norm": 0.0421120747923851,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 32553
+    },
+    {
+      "epoch": 3.3452527743526512,
+      "grad_norm": 0.08523180335760117,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 32556
+    },
+    {
+      "epoch": 3.345561035758323,
+      "grad_norm": 0.11694061011075974,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 32559
+    },
+    {
+      "epoch": 3.345869297163995,
+      "grad_norm": 0.05000199005007744,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 32562
+    },
+    {
+      "epoch": 3.346177558569667,
+      "grad_norm": 0.0399484746158123,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 32565
+    },
+    {
+      "epoch": 3.346485819975339,
+      "grad_norm": 0.0491316057741642,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 32568
+    },
+    {
+      "epoch": 3.3467940813810113,
+      "grad_norm": 0.042924270033836365,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 32571
+    },
+    {
+      "epoch": 3.347102342786683,
+      "grad_norm": 0.03486446663737297,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 32574
+    },
+    {
+      "epoch": 3.3474106041923553,
+      "grad_norm": 0.031064294278621674,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 32577
+    },
+    {
+      "epoch": 3.347718865598027,
+      "grad_norm": 0.05951589718461037,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 32580
+    },
+    {
+      "epoch": 3.348027127003699,
+      "grad_norm": 0.04387381300330162,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 32583
+    },
+    {
+      "epoch": 3.348335388409371,
+      "grad_norm": 0.05328337103128433,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 32586
+    },
+    {
+      "epoch": 3.348643649815043,
+      "grad_norm": 0.04561325162649155,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 32589
+    },
+    {
+      "epoch": 3.3489519112207153,
+      "grad_norm": 0.047260623425245285,
+      "learning_rate": 0.01,
+      "loss": 2.0387,
+      "step": 32592
+    },
+    {
+      "epoch": 3.349260172626387,
+      "grad_norm": 0.21082252264022827,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 32595
+    },
+    {
+      "epoch": 3.3495684340320593,
+      "grad_norm": 0.08391027897596359,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 32598
+    },
+    {
+      "epoch": 3.349876695437731,
+      "grad_norm": 0.07036472856998444,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 32601
+    },
+    {
+      "epoch": 3.3501849568434032,
+      "grad_norm": 0.03812922164797783,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 32604
+    },
+    {
+      "epoch": 3.350493218249075,
+      "grad_norm": 0.041235774755477905,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 32607
+    },
+    {
+      "epoch": 3.350801479654747,
+      "grad_norm": 0.05174950137734413,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 32610
+    },
+    {
+      "epoch": 3.3511097410604194,
+      "grad_norm": 0.058260124176740646,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 32613
+    },
+    {
+      "epoch": 3.351418002466091,
+      "grad_norm": 0.045651067048311234,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 32616
+    },
+    {
+      "epoch": 3.3517262638717633,
+      "grad_norm": 0.043610829859972,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 32619
+    },
+    {
+      "epoch": 3.3520345252774355,
+      "grad_norm": 0.04924603924155235,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 32622
+    },
+    {
+      "epoch": 3.3523427866831073,
+      "grad_norm": 0.04765019193291664,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 32625
+    },
+    {
+      "epoch": 3.3526510480887795,
+      "grad_norm": 0.04744412377476692,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 32628
+    },
+    {
+      "epoch": 3.352959309494451,
+      "grad_norm": 0.09456950426101685,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 32631
+    },
+    {
+      "epoch": 3.3532675709001234,
+      "grad_norm": 0.17356513440608978,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 32634
+    },
+    {
+      "epoch": 3.353575832305795,
+      "grad_norm": 0.08420834690332413,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 32637
+    },
+    {
+      "epoch": 3.3538840937114673,
+      "grad_norm": 0.09453277289867401,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 32640
+    },
+    {
+      "epoch": 3.3541923551171395,
+      "grad_norm": 0.05444180220365524,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 32643
+    },
+    {
+      "epoch": 3.3545006165228113,
+      "grad_norm": 0.04824339225888252,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 32646
+    },
+    {
+      "epoch": 3.3548088779284835,
+      "grad_norm": 0.06650727242231369,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 32649
+    },
+    {
+      "epoch": 3.3551171393341552,
+      "grad_norm": 0.05119656026363373,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 32652
+    },
+    {
+      "epoch": 3.3554254007398274,
+      "grad_norm": 0.044617343693971634,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 32655
+    },
+    {
+      "epoch": 3.355733662145499,
+      "grad_norm": 0.035579435527324677,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 32658
+    },
+    {
+      "epoch": 3.3560419235511714,
+      "grad_norm": 0.05802566558122635,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 32661
+    },
+    {
+      "epoch": 3.3563501849568436,
+      "grad_norm": 0.050934724509716034,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 32664
+    },
+    {
+      "epoch": 3.3566584463625153,
+      "grad_norm": 0.055400047451257706,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 32667
+    },
+    {
+      "epoch": 3.3569667077681875,
+      "grad_norm": 0.1315484195947647,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 32670
+    },
+    {
+      "epoch": 3.3572749691738593,
+      "grad_norm": 0.13136912882328033,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 32673
+    },
+    {
+      "epoch": 3.3575832305795315,
+      "grad_norm": 0.08574076741933823,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 32676
+    },
+    {
+      "epoch": 3.357891491985203,
+      "grad_norm": 0.04678389057517052,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 32679
+    },
+    {
+      "epoch": 3.3581997533908754,
+      "grad_norm": 0.03356343135237694,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 32682
+    },
+    {
+      "epoch": 3.3585080147965476,
+      "grad_norm": 0.046139661222696304,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 32685
+    },
+    {
+      "epoch": 3.3588162762022193,
+      "grad_norm": 0.03130761533975601,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 32688
+    },
+    {
+      "epoch": 3.3591245376078915,
+      "grad_norm": 0.17764367163181305,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 32691
+    },
+    {
+      "epoch": 3.3594327990135637,
+      "grad_norm": 0.04135056957602501,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 32694
+    },
+    {
+      "epoch": 3.3597410604192355,
+      "grad_norm": 0.11181548237800598,
+      "learning_rate": 0.01,
+      "loss": 2.048,
+      "step": 32697
+    },
+    {
+      "epoch": 3.3600493218249077,
+      "grad_norm": 0.07631994783878326,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 32700
+    },
+    {
+      "epoch": 3.3603575832305794,
+      "grad_norm": 0.03839050978422165,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 32703
+    },
+    {
+      "epoch": 3.3606658446362516,
+      "grad_norm": 0.03893091529607773,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 32706
+    },
+    {
+      "epoch": 3.3609741060419234,
+      "grad_norm": 0.15776588022708893,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 32709
+    },
+    {
+      "epoch": 3.3612823674475956,
+      "grad_norm": 0.125548854470253,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 32712
+    },
+    {
+      "epoch": 3.3615906288532678,
+      "grad_norm": 0.06952936947345734,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 32715
+    },
+    {
+      "epoch": 3.3618988902589395,
+      "grad_norm": 0.03826635703444481,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 32718
+    },
+    {
+      "epoch": 3.3622071516646117,
+      "grad_norm": 0.03977655619382858,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 32721
+    },
+    {
+      "epoch": 3.3625154130702835,
+      "grad_norm": 0.02882550098001957,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 32724
+    },
+    {
+      "epoch": 3.3628236744759556,
+      "grad_norm": 0.03502441197633743,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 32727
+    },
+    {
+      "epoch": 3.3631319358816274,
+      "grad_norm": 0.04370797425508499,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 32730
+    },
+    {
+      "epoch": 3.3634401972872996,
+      "grad_norm": 0.03528802841901779,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 32733
+    },
+    {
+      "epoch": 3.363748458692972,
+      "grad_norm": 0.08671889454126358,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 32736
+    },
+    {
+      "epoch": 3.3640567200986435,
+      "grad_norm": 0.1123836413025856,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 32739
+    },
+    {
+      "epoch": 3.3643649815043157,
+      "grad_norm": 0.061064526438713074,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 32742
+    },
+    {
+      "epoch": 3.3646732429099875,
+      "grad_norm": 0.05037948489189148,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 32745
+    },
+    {
+      "epoch": 3.3649815043156597,
+      "grad_norm": 0.052206844091415405,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 32748
+    },
+    {
+      "epoch": 3.365289765721332,
+      "grad_norm": 0.05795833095908165,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 32751
+    },
+    {
+      "epoch": 3.3655980271270036,
+      "grad_norm": 0.030604414641857147,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 32754
+    },
+    {
+      "epoch": 3.365906288532676,
+      "grad_norm": 0.05441366508603096,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 32757
+    },
+    {
+      "epoch": 3.3662145499383476,
+      "grad_norm": 0.07509131729602814,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 32760
+    },
+    {
+      "epoch": 3.3665228113440198,
+      "grad_norm": 0.046888504177331924,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 32763
+    },
+    {
+      "epoch": 3.366831072749692,
+      "grad_norm": 0.05030560493469238,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 32766
+    },
+    {
+      "epoch": 3.3671393341553637,
+      "grad_norm": 0.0426168255507946,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 32769
+    },
+    {
+      "epoch": 3.367447595561036,
+      "grad_norm": 0.10896468907594681,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 32772
+    },
+    {
+      "epoch": 3.3677558569667077,
+      "grad_norm": 0.11696910113096237,
+      "learning_rate": 0.01,
+      "loss": 2.0353,
+      "step": 32775
+    },
+    {
+      "epoch": 3.36806411837238,
+      "grad_norm": 0.07340724021196365,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 32778
+    },
+    {
+      "epoch": 3.3683723797780516,
+      "grad_norm": 0.037968121469020844,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 32781
+    },
+    {
+      "epoch": 3.368680641183724,
+      "grad_norm": 0.044434670358896255,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 32784
+    },
+    {
+      "epoch": 3.368988902589396,
+      "grad_norm": 0.03823886066675186,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 32787
+    },
+    {
+      "epoch": 3.3692971639950677,
+      "grad_norm": 0.06556801497936249,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 32790
+    },
+    {
+      "epoch": 3.36960542540074,
+      "grad_norm": 0.06128913164138794,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 32793
+    },
+    {
+      "epoch": 3.3699136868064117,
+      "grad_norm": 0.08499012142419815,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 32796
+    },
+    {
+      "epoch": 3.370221948212084,
+      "grad_norm": 0.03410051763057709,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 32799
+    },
+    {
+      "epoch": 3.3705302096177556,
+      "grad_norm": 0.08818015456199646,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 32802
+    },
+    {
+      "epoch": 3.370838471023428,
+      "grad_norm": 0.045091863721609116,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 32805
+    },
+    {
+      "epoch": 3.3711467324291,
+      "grad_norm": 0.10982260853052139,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 32808
+    },
+    {
+      "epoch": 3.3714549938347718,
+      "grad_norm": 0.04633982852101326,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 32811
+    },
+    {
+      "epoch": 3.371763255240444,
+      "grad_norm": 0.04701898992061615,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 32814
+    },
+    {
+      "epoch": 3.372071516646116,
+      "grad_norm": 0.03449505567550659,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 32817
+    },
+    {
+      "epoch": 3.372379778051788,
+      "grad_norm": 0.03621023893356323,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 32820
+    },
+    {
+      "epoch": 3.37268803945746,
+      "grad_norm": 0.04743462800979614,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 32823
+    },
+    {
+      "epoch": 3.372996300863132,
+      "grad_norm": 0.04240218549966812,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 32826
+    },
+    {
+      "epoch": 3.373304562268804,
+      "grad_norm": 0.09400332719087601,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 32829
+    },
+    {
+      "epoch": 3.373612823674476,
+      "grad_norm": 0.07313279062509537,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 32832
+    },
+    {
+      "epoch": 3.373921085080148,
+      "grad_norm": 0.07604516297578812,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 32835
+    },
+    {
+      "epoch": 3.37422934648582,
+      "grad_norm": 0.044236812740564346,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 32838
+    },
+    {
+      "epoch": 3.374537607891492,
+      "grad_norm": 0.051601652055978775,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 32841
+    },
+    {
+      "epoch": 3.374845869297164,
+      "grad_norm": 0.10818912088871002,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 32844
+    },
+    {
+      "epoch": 3.375154130702836,
+      "grad_norm": 0.04563935101032257,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 32847
+    },
+    {
+      "epoch": 3.375462392108508,
+      "grad_norm": 0.053665511310100555,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 32850
+    },
+    {
+      "epoch": 3.37577065351418,
+      "grad_norm": 0.08934652805328369,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 32853
+    },
+    {
+      "epoch": 3.376078914919852,
+      "grad_norm": 0.0752192884683609,
+      "learning_rate": 0.01,
+      "loss": 2.0429,
+      "step": 32856
+    },
+    {
+      "epoch": 3.376387176325524,
+      "grad_norm": 0.05763142928481102,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 32859
+    },
+    {
+      "epoch": 3.376695437731196,
+      "grad_norm": 0.03926476463675499,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 32862
+    },
+    {
+      "epoch": 3.377003699136868,
+      "grad_norm": 0.05735384672880173,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 32865
+    },
+    {
+      "epoch": 3.37731196054254,
+      "grad_norm": 0.18665596842765808,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 32868
+    },
+    {
+      "epoch": 3.377620221948212,
+      "grad_norm": 0.06702150404453278,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 32871
+    },
+    {
+      "epoch": 3.377928483353884,
+      "grad_norm": 0.051258910447359085,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 32874
+    },
+    {
+      "epoch": 3.378236744759556,
+      "grad_norm": 0.05804390087723732,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 32877
+    },
+    {
+      "epoch": 3.3785450061652282,
+      "grad_norm": 0.04688677936792374,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 32880
+    },
+    {
+      "epoch": 3.3788532675709,
+      "grad_norm": 0.057768989354372025,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 32883
+    },
+    {
+      "epoch": 3.379161528976572,
+      "grad_norm": 0.05571329593658447,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 32886
+    },
+    {
+      "epoch": 3.3794697903822444,
+      "grad_norm": 0.04497957229614258,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 32889
+    },
+    {
+      "epoch": 3.379778051787916,
+      "grad_norm": 0.043453045189380646,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 32892
+    },
+    {
+      "epoch": 3.3800863131935883,
+      "grad_norm": 0.045709915459156036,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 32895
+    },
+    {
+      "epoch": 3.38039457459926,
+      "grad_norm": 0.05974254012107849,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 32898
+    },
+    {
+      "epoch": 3.3807028360049323,
+      "grad_norm": 0.16592206060886383,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 32901
+    },
+    {
+      "epoch": 3.381011097410604,
+      "grad_norm": 0.04671747609972954,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 32904
+    },
+    {
+      "epoch": 3.381319358816276,
+      "grad_norm": 0.07180918008089066,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 32907
+    },
+    {
+      "epoch": 3.3816276202219484,
+      "grad_norm": 0.06775587797164917,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 32910
+    },
+    {
+      "epoch": 3.38193588162762,
+      "grad_norm": 0.04381205141544342,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 32913
+    },
+    {
+      "epoch": 3.3822441430332923,
+      "grad_norm": 0.0345197468996048,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 32916
+    },
+    {
+      "epoch": 3.382552404438964,
+      "grad_norm": 0.04965033382177353,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 32919
+    },
+    {
+      "epoch": 3.3828606658446363,
+      "grad_norm": 0.20276527106761932,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 32922
+    },
+    {
+      "epoch": 3.383168927250308,
+      "grad_norm": 0.0702800378203392,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 32925
+    },
+    {
+      "epoch": 3.3834771886559802,
+      "grad_norm": 0.05775219202041626,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 32928
+    },
+    {
+      "epoch": 3.3837854500616524,
+      "grad_norm": 0.0662347599864006,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 32931
+    },
+    {
+      "epoch": 3.384093711467324,
+      "grad_norm": 0.05067736655473709,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 32934
+    },
+    {
+      "epoch": 3.3844019728729964,
+      "grad_norm": 0.057027652859687805,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 32937
+    },
+    {
+      "epoch": 3.384710234278668,
+      "grad_norm": 0.0408274307847023,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 32940
+    },
+    {
+      "epoch": 3.3850184956843403,
+      "grad_norm": 0.049467723816633224,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 32943
+    },
+    {
+      "epoch": 3.3853267570900125,
+      "grad_norm": 0.07573775947093964,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 32946
+    },
+    {
+      "epoch": 3.3856350184956843,
+      "grad_norm": 0.04027678817510605,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 32949
+    },
+    {
+      "epoch": 3.3859432799013565,
+      "grad_norm": 0.09980335086584091,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 32952
+    },
+    {
+      "epoch": 3.386251541307028,
+      "grad_norm": 0.06230602413415909,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 32955
+    },
+    {
+      "epoch": 3.3865598027127004,
+      "grad_norm": 0.09655454754829407,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 32958
+    },
+    {
+      "epoch": 3.3868680641183726,
+      "grad_norm": 0.053587790578603745,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 32961
+    },
+    {
+      "epoch": 3.3871763255240444,
+      "grad_norm": 0.08395679295063019,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 32964
+    },
+    {
+      "epoch": 3.3874845869297165,
+      "grad_norm": 0.08353892713785172,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 32967
+    },
+    {
+      "epoch": 3.3877928483353883,
+      "grad_norm": 0.11298651993274689,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 32970
+    },
+    {
+      "epoch": 3.3881011097410605,
+      "grad_norm": 0.08317071199417114,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 32973
+    },
+    {
+      "epoch": 3.3884093711467322,
+      "grad_norm": 0.07725278288125992,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 32976
+    },
+    {
+      "epoch": 3.3887176325524044,
+      "grad_norm": 0.06264784932136536,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 32979
+    },
+    {
+      "epoch": 3.3890258939580766,
+      "grad_norm": 0.0588025264441967,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 32982
+    },
+    {
+      "epoch": 3.3893341553637484,
+      "grad_norm": 0.033383727073669434,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 32985
+    },
+    {
+      "epoch": 3.3896424167694206,
+      "grad_norm": 0.04963357746601105,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 32988
+    },
+    {
+      "epoch": 3.3899506781750923,
+      "grad_norm": 0.03166192024946213,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 32991
+    },
+    {
+      "epoch": 3.3902589395807645,
+      "grad_norm": 0.0424019880592823,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 32994
+    },
+    {
+      "epoch": 3.3905672009864363,
+      "grad_norm": 0.0549466572701931,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 32997
+    },
+    {
+      "epoch": 3.3908754623921085,
+      "grad_norm": 0.06859169900417328,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 33000
+    },
+    {
+      "epoch": 3.3911837237977807,
+      "grad_norm": 0.05035685375332832,
+      "learning_rate": 0.01,
+      "loss": 2.0422,
+      "step": 33003
+    },
+    {
+      "epoch": 3.3914919852034524,
+      "grad_norm": 0.10227832943201065,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 33006
+    },
+    {
+      "epoch": 3.3918002466091246,
+      "grad_norm": 0.052029043436050415,
+      "learning_rate": 0.01,
+      "loss": 2.0341,
+      "step": 33009
+    },
+    {
+      "epoch": 3.392108508014797,
+      "grad_norm": 0.03569505736231804,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 33012
+    },
+    {
+      "epoch": 3.3924167694204685,
+      "grad_norm": 0.05492673069238663,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 33015
+    },
+    {
+      "epoch": 3.3927250308261407,
+      "grad_norm": 0.10698610544204712,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 33018
+    },
+    {
+      "epoch": 3.3930332922318125,
+      "grad_norm": 0.051218993961811066,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 33021
+    },
+    {
+      "epoch": 3.3933415536374847,
+      "grad_norm": 0.10021807998418808,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 33024
+    },
+    {
+      "epoch": 3.3936498150431564,
+      "grad_norm": 0.043556150048971176,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 33027
+    },
+    {
+      "epoch": 3.3939580764488286,
+      "grad_norm": 0.052555110305547714,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 33030
+    },
+    {
+      "epoch": 3.394266337854501,
+      "grad_norm": 0.1656499058008194,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 33033
+    },
+    {
+      "epoch": 3.3945745992601726,
+      "grad_norm": 0.08107822388410568,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 33036
+    },
+    {
+      "epoch": 3.3948828606658448,
+      "grad_norm": 0.05703389272093773,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 33039
+    },
+    {
+      "epoch": 3.3951911220715165,
+      "grad_norm": 0.060602329671382904,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 33042
+    },
+    {
+      "epoch": 3.3954993834771887,
+      "grad_norm": 0.05680840089917183,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 33045
+    },
+    {
+      "epoch": 3.3958076448828605,
+      "grad_norm": 0.053718626499176025,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 33048
+    },
+    {
+      "epoch": 3.3961159062885327,
+      "grad_norm": 0.04885102063417435,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 33051
+    },
+    {
+      "epoch": 3.396424167694205,
+      "grad_norm": 0.046444397419691086,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 33054
+    },
+    {
+      "epoch": 3.3967324290998766,
+      "grad_norm": 0.03031921572983265,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 33057
+    },
+    {
+      "epoch": 3.397040690505549,
+      "grad_norm": 0.048908550292253494,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 33060
+    },
+    {
+      "epoch": 3.3973489519112205,
+      "grad_norm": 0.06003925949335098,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 33063
+    },
+    {
+      "epoch": 3.3976572133168927,
+      "grad_norm": 0.10929730534553528,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 33066
+    },
+    {
+      "epoch": 3.3979654747225645,
+      "grad_norm": 0.09032581746578217,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 33069
+    },
+    {
+      "epoch": 3.3982737361282367,
+      "grad_norm": 0.043940551578998566,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 33072
+    },
+    {
+      "epoch": 3.398581997533909,
+      "grad_norm": 0.10148674994707108,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 33075
+    },
+    {
+      "epoch": 3.3988902589395806,
+      "grad_norm": 0.04877715930342674,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 33078
+    },
+    {
+      "epoch": 3.399198520345253,
+      "grad_norm": 0.10485149919986725,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 33081
+    },
+    {
+      "epoch": 3.399506781750925,
+      "grad_norm": 0.12222550064325333,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 33084
+    },
+    {
+      "epoch": 3.3998150431565968,
+      "grad_norm": 0.0744134709239006,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 33087
+    },
+    {
+      "epoch": 3.400123304562269,
+      "grad_norm": 0.08050314337015152,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 33090
+    },
+    {
+      "epoch": 3.4004315659679407,
+      "grad_norm": 0.0806950256228447,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 33093
+    },
+    {
+      "epoch": 3.400739827373613,
+      "grad_norm": 0.06369089335203171,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 33096
+    },
+    {
+      "epoch": 3.4010480887792847,
+      "grad_norm": 0.06041014939546585,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 33099
+    },
+    {
+      "epoch": 3.401356350184957,
+      "grad_norm": 0.0476149246096611,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 33102
+    },
+    {
+      "epoch": 3.401664611590629,
+      "grad_norm": 0.05710010603070259,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 33105
+    },
+    {
+      "epoch": 3.401972872996301,
+      "grad_norm": 0.04351675137877464,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 33108
+    },
+    {
+      "epoch": 3.402281134401973,
+      "grad_norm": 0.04341613128781319,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 33111
+    },
+    {
+      "epoch": 3.4025893958076447,
+      "grad_norm": 0.04619600623846054,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 33114
+    },
+    {
+      "epoch": 3.402897657213317,
+      "grad_norm": 0.07674260437488556,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 33117
+    },
+    {
+      "epoch": 3.4032059186189887,
+      "grad_norm": 0.03869005665183067,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 33120
+    },
+    {
+      "epoch": 3.403514180024661,
+      "grad_norm": 0.08710911124944687,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 33123
+    },
+    {
+      "epoch": 3.403822441430333,
+      "grad_norm": 0.09343576431274414,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 33126
+    },
+    {
+      "epoch": 3.404130702836005,
+      "grad_norm": 0.048723649233579636,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 33129
+    },
+    {
+      "epoch": 3.404438964241677,
+      "grad_norm": 0.07655228674411774,
+      "learning_rate": 0.01,
+      "loss": 2.0235,
+      "step": 33132
+    },
+    {
+      "epoch": 3.4047472256473488,
+      "grad_norm": 0.08873139321804047,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 33135
+    },
+    {
+      "epoch": 3.405055487053021,
+      "grad_norm": 0.06514773517847061,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 33138
+    },
+    {
+      "epoch": 3.405363748458693,
+      "grad_norm": 0.06446196138858795,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 33141
+    },
+    {
+      "epoch": 3.405672009864365,
+      "grad_norm": 0.12146038562059402,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 33144
+    },
+    {
+      "epoch": 3.405980271270037,
+      "grad_norm": 0.06639200448989868,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 33147
+    },
+    {
+      "epoch": 3.406288532675709,
+      "grad_norm": 0.05644892901182175,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 33150
+    },
+    {
+      "epoch": 3.406596794081381,
+      "grad_norm": 0.037636831402778625,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 33153
+    },
+    {
+      "epoch": 3.4069050554870532,
+      "grad_norm": 0.031332679092884064,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 33156
+    },
+    {
+      "epoch": 3.407213316892725,
+      "grad_norm": 0.051785338670015335,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 33159
+    },
+    {
+      "epoch": 3.407521578298397,
+      "grad_norm": 0.07761172205209732,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 33162
+    },
+    {
+      "epoch": 3.407829839704069,
+      "grad_norm": 0.09536123275756836,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 33165
+    },
+    {
+      "epoch": 3.408138101109741,
+      "grad_norm": 0.04122615605592728,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 33168
+    },
+    {
+      "epoch": 3.408446362515413,
+      "grad_norm": 0.0360184982419014,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 33171
+    },
+    {
+      "epoch": 3.408754623921085,
+      "grad_norm": 0.03222360834479332,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 33174
+    },
+    {
+      "epoch": 3.4090628853267573,
+      "grad_norm": 0.043042391538619995,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 33177
+    },
+    {
+      "epoch": 3.409371146732429,
+      "grad_norm": 0.06593729555606842,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 33180
+    },
+    {
+      "epoch": 3.409679408138101,
+      "grad_norm": 0.09074060618877411,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 33183
+    },
+    {
+      "epoch": 3.409987669543773,
+      "grad_norm": 0.06270799785852432,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 33186
+    },
+    {
+      "epoch": 3.410295930949445,
+      "grad_norm": 0.08982829004526138,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 33189
+    },
+    {
+      "epoch": 3.410604192355117,
+      "grad_norm": 0.04844099283218384,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 33192
+    },
+    {
+      "epoch": 3.410912453760789,
+      "grad_norm": 0.07782240957021713,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 33195
+    },
+    {
+      "epoch": 3.4112207151664613,
+      "grad_norm": 0.07867666333913803,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 33198
+    },
+    {
+      "epoch": 3.411528976572133,
+      "grad_norm": 0.07254987210035324,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 33201
+    },
+    {
+      "epoch": 3.4118372379778052,
+      "grad_norm": 0.04480341821908951,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 33204
+    },
+    {
+      "epoch": 3.412145499383477,
+      "grad_norm": 0.03986749053001404,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 33207
+    },
+    {
+      "epoch": 3.412453760789149,
+      "grad_norm": 0.04472361132502556,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 33210
+    },
+    {
+      "epoch": 3.4127620221948214,
+      "grad_norm": 0.07955579459667206,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 33213
+    },
+    {
+      "epoch": 3.413070283600493,
+      "grad_norm": 0.056707967072725296,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 33216
+    },
+    {
+      "epoch": 3.4133785450061653,
+      "grad_norm": 0.04070746898651123,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 33219
+    },
+    {
+      "epoch": 3.413686806411837,
+      "grad_norm": 0.03773896023631096,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 33222
+    },
+    {
+      "epoch": 3.4139950678175093,
+      "grad_norm": 0.10554299503564835,
+      "learning_rate": 0.01,
+      "loss": 2.0331,
+      "step": 33225
+    },
+    {
+      "epoch": 3.4143033292231815,
+      "grad_norm": 0.0555710643529892,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 33228
+    },
+    {
+      "epoch": 3.414611590628853,
+      "grad_norm": 0.1139519140124321,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 33231
+    },
+    {
+      "epoch": 3.4149198520345254,
+      "grad_norm": 0.042904384434223175,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 33234
+    },
+    {
+      "epoch": 3.415228113440197,
+      "grad_norm": 0.10528502613306046,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 33237
+    },
+    {
+      "epoch": 3.4155363748458694,
+      "grad_norm": 0.03914659097790718,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 33240
+    },
+    {
+      "epoch": 3.415844636251541,
+      "grad_norm": 0.11122586578130722,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 33243
+    },
+    {
+      "epoch": 3.4161528976572133,
+      "grad_norm": 0.06572670489549637,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 33246
+    },
+    {
+      "epoch": 3.4164611590628855,
+      "grad_norm": 0.05022534728050232,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 33249
+    },
+    {
+      "epoch": 3.4167694204685573,
+      "grad_norm": 0.08149400353431702,
+      "learning_rate": 0.01,
+      "loss": 2.0326,
+      "step": 33252
+    },
+    {
+      "epoch": 3.4170776818742294,
+      "grad_norm": 0.04350002855062485,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 33255
+    },
+    {
+      "epoch": 3.417385943279901,
+      "grad_norm": 0.0445462241768837,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 33258
+    },
+    {
+      "epoch": 3.4176942046855734,
+      "grad_norm": 0.04082934185862541,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 33261
+    },
+    {
+      "epoch": 3.418002466091245,
+      "grad_norm": 0.060355301946401596,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 33264
+    },
+    {
+      "epoch": 3.4183107274969173,
+      "grad_norm": 0.04439264163374901,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 33267
+    },
+    {
+      "epoch": 3.4186189889025895,
+      "grad_norm": 0.11760549992322922,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 33270
+    },
+    {
+      "epoch": 3.4189272503082613,
+      "grad_norm": 0.08547448366880417,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 33273
+    },
+    {
+      "epoch": 3.4192355117139335,
+      "grad_norm": 0.06072860211133957,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 33276
+    },
+    {
+      "epoch": 3.4195437731196057,
+      "grad_norm": 0.03637344762682915,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 33279
+    },
+    {
+      "epoch": 3.4198520345252774,
+      "grad_norm": 0.0440024733543396,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 33282
+    },
+    {
+      "epoch": 3.4201602959309496,
+      "grad_norm": 0.040933944284915924,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 33285
+    },
+    {
+      "epoch": 3.4204685573366214,
+      "grad_norm": 0.10233576595783234,
+      "learning_rate": 0.01,
+      "loss": 2.0433,
+      "step": 33288
+    },
+    {
+      "epoch": 3.4207768187422936,
+      "grad_norm": 0.10494884103536606,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 33291
+    },
+    {
+      "epoch": 3.4210850801479653,
+      "grad_norm": 0.04247460886836052,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 33294
+    },
+    {
+      "epoch": 3.4213933415536375,
+      "grad_norm": 0.07929468154907227,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 33297
+    },
+    {
+      "epoch": 3.4217016029593097,
+      "grad_norm": 0.04947086423635483,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 33300
+    },
+    {
+      "epoch": 3.4220098643649814,
+      "grad_norm": 0.05473649874329567,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 33303
+    },
+    {
+      "epoch": 3.4223181257706536,
+      "grad_norm": 0.06789970397949219,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 33306
+    },
+    {
+      "epoch": 3.4226263871763254,
+      "grad_norm": 0.044122181832790375,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 33309
+    },
+    {
+      "epoch": 3.4229346485819976,
+      "grad_norm": 0.0713338777422905,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 33312
+    },
+    {
+      "epoch": 3.4232429099876693,
+      "grad_norm": 0.09774953871965408,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 33315
+    },
+    {
+      "epoch": 3.4235511713933415,
+      "grad_norm": 0.049434032291173935,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 33318
+    },
+    {
+      "epoch": 3.4238594327990137,
+      "grad_norm": 0.06290262192487717,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 33321
+    },
+    {
+      "epoch": 3.4241676942046855,
+      "grad_norm": 0.042732108384370804,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 33324
+    },
+    {
+      "epoch": 3.4244759556103577,
+      "grad_norm": 0.03461041674017906,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 33327
+    },
+    {
+      "epoch": 3.4247842170160294,
+      "grad_norm": 0.04503572732210159,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 33330
+    },
+    {
+      "epoch": 3.4250924784217016,
+      "grad_norm": 0.04774646461009979,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 33333
+    },
+    {
+      "epoch": 3.4254007398273734,
+      "grad_norm": 0.08576779067516327,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 33336
+    },
+    {
+      "epoch": 3.4257090012330456,
+      "grad_norm": 0.13729599118232727,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 33339
+    },
+    {
+      "epoch": 3.4260172626387178,
+      "grad_norm": 0.060716260224580765,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 33342
+    },
+    {
+      "epoch": 3.4263255240443895,
+      "grad_norm": 0.037777479737997055,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 33345
+    },
+    {
+      "epoch": 3.4266337854500617,
+      "grad_norm": 0.042688485234975815,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 33348
+    },
+    {
+      "epoch": 3.426942046855734,
+      "grad_norm": 0.07009312510490417,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 33351
+    },
+    {
+      "epoch": 3.4272503082614056,
+      "grad_norm": 0.10260313749313354,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 33354
+    },
+    {
+      "epoch": 3.427558569667078,
+      "grad_norm": 0.08466068655252457,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 33357
+    },
+    {
+      "epoch": 3.4278668310727496,
+      "grad_norm": 0.061912521719932556,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 33360
+    },
+    {
+      "epoch": 3.428175092478422,
+      "grad_norm": 0.0470789298415184,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 33363
+    },
+    {
+      "epoch": 3.4284833538840935,
+      "grad_norm": 0.0477573424577713,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 33366
+    },
+    {
+      "epoch": 3.4287916152897657,
+      "grad_norm": 0.03324504569172859,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 33369
+    },
+    {
+      "epoch": 3.429099876695438,
+      "grad_norm": 0.07741666585206985,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 33372
+    },
+    {
+      "epoch": 3.4294081381011097,
+      "grad_norm": 0.05770926922559738,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 33375
+    },
+    {
+      "epoch": 3.429716399506782,
+      "grad_norm": 0.08471731096506119,
+      "learning_rate": 0.01,
+      "loss": 2.0451,
+      "step": 33378
+    },
+    {
+      "epoch": 3.4300246609124536,
+      "grad_norm": 0.04667286202311516,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 33381
+    },
+    {
+      "epoch": 3.430332922318126,
+      "grad_norm": 0.0683809220790863,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 33384
+    },
+    {
+      "epoch": 3.4306411837237976,
+      "grad_norm": 0.07834406197071075,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 33387
+    },
+    {
+      "epoch": 3.4309494451294698,
+      "grad_norm": 0.04956913739442825,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 33390
+    },
+    {
+      "epoch": 3.431257706535142,
+      "grad_norm": 0.05492135509848595,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 33393
+    },
+    {
+      "epoch": 3.4315659679408137,
+      "grad_norm": 0.05343586578965187,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 33396
+    },
+    {
+      "epoch": 3.431874229346486,
+      "grad_norm": 0.04083942621946335,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 33399
+    },
+    {
+      "epoch": 3.4321824907521576,
+      "grad_norm": 0.06474661827087402,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 33402
+    },
+    {
+      "epoch": 3.43249075215783,
+      "grad_norm": 0.09690015017986298,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 33405
+    },
+    {
+      "epoch": 3.432799013563502,
+      "grad_norm": 0.17796829342842102,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 33408
+    },
+    {
+      "epoch": 3.433107274969174,
+      "grad_norm": 0.11173928529024124,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 33411
+    },
+    {
+      "epoch": 3.433415536374846,
+      "grad_norm": 0.045607730746269226,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 33414
+    },
+    {
+      "epoch": 3.4337237977805177,
+      "grad_norm": 0.03982311487197876,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 33417
+    },
+    {
+      "epoch": 3.43403205918619,
+      "grad_norm": 0.03420604392886162,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 33420
+    },
+    {
+      "epoch": 3.434340320591862,
+      "grad_norm": 0.03757992014288902,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 33423
+    },
+    {
+      "epoch": 3.434648581997534,
+      "grad_norm": 0.05664653331041336,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 33426
+    },
+    {
+      "epoch": 3.434956843403206,
+      "grad_norm": 0.05265260115265846,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 33429
+    },
+    {
+      "epoch": 3.435265104808878,
+      "grad_norm": 0.0430876798927784,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 33432
+    },
+    {
+      "epoch": 3.43557336621455,
+      "grad_norm": 0.04210485517978668,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 33435
+    },
+    {
+      "epoch": 3.4358816276202218,
+      "grad_norm": 0.049002841114997864,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 33438
+    },
+    {
+      "epoch": 3.436189889025894,
+      "grad_norm": 0.0786898285150528,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 33441
+    },
+    {
+      "epoch": 3.436498150431566,
+      "grad_norm": 0.04218638688325882,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 33444
+    },
+    {
+      "epoch": 3.436806411837238,
+      "grad_norm": 0.13452041149139404,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 33447
+    },
+    {
+      "epoch": 3.43711467324291,
+      "grad_norm": 0.06728319823741913,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 33450
+    },
+    {
+      "epoch": 3.437422934648582,
+      "grad_norm": 0.09041707217693329,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 33453
+    },
+    {
+      "epoch": 3.437731196054254,
+      "grad_norm": 0.04817497730255127,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 33456
+    },
+    {
+      "epoch": 3.438039457459926,
+      "grad_norm": 0.05033170431852341,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 33459
+    },
+    {
+      "epoch": 3.438347718865598,
+      "grad_norm": 0.12856252491474152,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 33462
+    },
+    {
+      "epoch": 3.43865598027127,
+      "grad_norm": 0.03690528869628906,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 33465
+    },
+    {
+      "epoch": 3.438964241676942,
+      "grad_norm": 0.09053459018468857,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 33468
+    },
+    {
+      "epoch": 3.439272503082614,
+      "grad_norm": 0.11929309368133545,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 33471
+    },
+    {
+      "epoch": 3.4395807644882863,
+      "grad_norm": 0.06987284123897552,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 33474
+    },
+    {
+      "epoch": 3.439889025893958,
+      "grad_norm": 0.06181707605719566,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 33477
+    },
+    {
+      "epoch": 3.4401972872996303,
+      "grad_norm": 0.045914020389318466,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 33480
+    },
+    {
+      "epoch": 3.440505548705302,
+      "grad_norm": 0.04277556762099266,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 33483
+    },
+    {
+      "epoch": 3.440813810110974,
+      "grad_norm": 0.045943450182676315,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 33486
+    },
+    {
+      "epoch": 3.441122071516646,
+      "grad_norm": 0.04440785571932793,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 33489
+    },
+    {
+      "epoch": 3.441430332922318,
+      "grad_norm": 0.12033234536647797,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 33492
+    },
+    {
+      "epoch": 3.4417385943279903,
+      "grad_norm": 0.06069677323102951,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 33495
+    },
+    {
+      "epoch": 3.442046855733662,
+      "grad_norm": 0.08571046590805054,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 33498
+    },
+    {
+      "epoch": 3.4423551171393343,
+      "grad_norm": 0.05251142010092735,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 33501
+    },
+    {
+      "epoch": 3.442663378545006,
+      "grad_norm": 0.09658701717853546,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 33504
+    },
+    {
+      "epoch": 3.4429716399506782,
+      "grad_norm": 0.10625968873500824,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 33507
+    },
+    {
+      "epoch": 3.44327990135635,
+      "grad_norm": 0.061645977199077606,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 33510
+    },
+    {
+      "epoch": 3.443588162762022,
+      "grad_norm": 0.06879527121782303,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 33513
+    },
+    {
+      "epoch": 3.4438964241676944,
+      "grad_norm": 0.07986783981323242,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 33516
+    },
+    {
+      "epoch": 3.444204685573366,
+      "grad_norm": 0.06323929876089096,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 33519
+    },
+    {
+      "epoch": 3.4445129469790383,
+      "grad_norm": 0.08186205476522446,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 33522
+    },
+    {
+      "epoch": 3.44482120838471,
+      "grad_norm": 0.1049259677529335,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 33525
+    },
+    {
+      "epoch": 3.4451294697903823,
+      "grad_norm": 0.12427592277526855,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 33528
+    },
+    {
+      "epoch": 3.445437731196054,
+      "grad_norm": 0.04911283776164055,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 33531
+    },
+    {
+      "epoch": 3.445745992601726,
+      "grad_norm": 0.07451221346855164,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 33534
+    },
+    {
+      "epoch": 3.4460542540073984,
+      "grad_norm": 0.04838255047798157,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 33537
+    },
+    {
+      "epoch": 3.44636251541307,
+      "grad_norm": 0.0435669869184494,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 33540
+    },
+    {
+      "epoch": 3.4466707768187423,
+      "grad_norm": 0.12010036408901215,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 33543
+    },
+    {
+      "epoch": 3.4469790382244145,
+      "grad_norm": 0.04258548840880394,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 33546
+    },
+    {
+      "epoch": 3.4472872996300863,
+      "grad_norm": 0.04736476391553879,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 33549
+    },
+    {
+      "epoch": 3.4475955610357585,
+      "grad_norm": 0.04423545300960541,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 33552
+    },
+    {
+      "epoch": 3.4479038224414302,
+      "grad_norm": 0.07585839927196503,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 33555
+    },
+    {
+      "epoch": 3.4482120838471024,
+      "grad_norm": 0.03559425473213196,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 33558
+    },
+    {
+      "epoch": 3.448520345252774,
+      "grad_norm": 0.058567702770233154,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 33561
+    },
+    {
+      "epoch": 3.4488286066584464,
+      "grad_norm": 0.05344400927424431,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 33564
+    },
+    {
+      "epoch": 3.4491368680641186,
+      "grad_norm": 0.0584418885409832,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 33567
+    },
+    {
+      "epoch": 3.4494451294697903,
+      "grad_norm": 0.14322839677333832,
+      "learning_rate": 0.01,
+      "loss": 2.0362,
+      "step": 33570
+    },
+    {
+      "epoch": 3.4497533908754625,
+      "grad_norm": 0.039136361330747604,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 33573
+    },
+    {
+      "epoch": 3.4500616522811343,
+      "grad_norm": 0.0871317982673645,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 33576
+    },
+    {
+      "epoch": 3.4503699136868065,
+      "grad_norm": 0.07295375317335129,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 33579
+    },
+    {
+      "epoch": 3.450678175092478,
+      "grad_norm": 0.04469291865825653,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 33582
+    },
+    {
+      "epoch": 3.4509864364981504,
+      "grad_norm": 0.06063535064458847,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 33585
+    },
+    {
+      "epoch": 3.4512946979038226,
+      "grad_norm": 0.07924182713031769,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 33588
+    },
+    {
+      "epoch": 3.4516029593094943,
+      "grad_norm": 0.089077427983284,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 33591
+    },
+    {
+      "epoch": 3.4519112207151665,
+      "grad_norm": 0.08197617530822754,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 33594
+    },
+    {
+      "epoch": 3.4522194821208383,
+      "grad_norm": 0.039551399648189545,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 33597
+    },
+    {
+      "epoch": 3.4525277435265105,
+      "grad_norm": 0.08920145779848099,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 33600
+    },
+    {
+      "epoch": 3.4528360049321827,
+      "grad_norm": 0.04301462695002556,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 33603
+    },
+    {
+      "epoch": 3.4531442663378544,
+      "grad_norm": 0.05201060697436333,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 33606
+    },
+    {
+      "epoch": 3.4534525277435266,
+      "grad_norm": 0.05899956077337265,
+      "learning_rate": 0.01,
+      "loss": 2.0318,
+      "step": 33609
+    },
+    {
+      "epoch": 3.4537607891491984,
+      "grad_norm": 0.09653299301862717,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 33612
+    },
+    {
+      "epoch": 3.4540690505548706,
+      "grad_norm": 0.06416913866996765,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 33615
+    },
+    {
+      "epoch": 3.4543773119605428,
+      "grad_norm": 0.07932529598474503,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 33618
+    },
+    {
+      "epoch": 3.4546855733662145,
+      "grad_norm": 0.06251846253871918,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 33621
+    },
+    {
+      "epoch": 3.4549938347718867,
+      "grad_norm": 0.07022767513990402,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 33624
+    },
+    {
+      "epoch": 3.4553020961775585,
+      "grad_norm": 0.08116226643323898,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 33627
+    },
+    {
+      "epoch": 3.4556103575832307,
+      "grad_norm": 0.07052188366651535,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 33630
+    },
+    {
+      "epoch": 3.4559186189889024,
+      "grad_norm": 0.055427566170692444,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 33633
+    },
+    {
+      "epoch": 3.4562268803945746,
+      "grad_norm": 0.049462925642728806,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 33636
+    },
+    {
+      "epoch": 3.456535141800247,
+      "grad_norm": 0.05117397755384445,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 33639
+    },
+    {
+      "epoch": 3.4568434032059185,
+      "grad_norm": 0.14244894683361053,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 33642
+    },
+    {
+      "epoch": 3.4571516646115907,
+      "grad_norm": 0.040531981736421585,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 33645
+    },
+    {
+      "epoch": 3.4574599260172625,
+      "grad_norm": 0.0898265540599823,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 33648
+    },
+    {
+      "epoch": 3.4577681874229347,
+      "grad_norm": 0.06414288282394409,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 33651
+    },
+    {
+      "epoch": 3.4580764488286064,
+      "grad_norm": 0.06362918019294739,
+      "learning_rate": 0.01,
+      "loss": 2.031,
+      "step": 33654
+    },
+    {
+      "epoch": 3.4583847102342786,
+      "grad_norm": 0.0881992056965828,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 33657
+    },
+    {
+      "epoch": 3.458692971639951,
+      "grad_norm": 0.03909778594970703,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 33660
+    },
+    {
+      "epoch": 3.4590012330456226,
+      "grad_norm": 0.06597696989774704,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 33663
+    },
+    {
+      "epoch": 3.4593094944512948,
+      "grad_norm": 0.05599299073219299,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 33666
+    },
+    {
+      "epoch": 3.4596177558569665,
+      "grad_norm": 0.0734795406460762,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 33669
+    },
+    {
+      "epoch": 3.4599260172626387,
+      "grad_norm": 0.07773378491401672,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 33672
+    },
+    {
+      "epoch": 3.460234278668311,
+      "grad_norm": 0.04273884743452072,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 33675
+    },
+    {
+      "epoch": 3.4605425400739827,
+      "grad_norm": 0.08914119005203247,
+      "learning_rate": 0.01,
+      "loss": 2.0455,
+      "step": 33678
+    },
+    {
+      "epoch": 3.460850801479655,
+      "grad_norm": 0.059121765196323395,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 33681
+    },
+    {
+      "epoch": 3.4611590628853266,
+      "grad_norm": 0.0828641727566719,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 33684
+    },
+    {
+      "epoch": 3.461467324290999,
+      "grad_norm": 0.07057880610227585,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 33687
+    },
+    {
+      "epoch": 3.461775585696671,
+      "grad_norm": 0.0789676234126091,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 33690
+    },
+    {
+      "epoch": 3.4620838471023427,
+      "grad_norm": 0.06654086709022522,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 33693
+    },
+    {
+      "epoch": 3.462392108508015,
+      "grad_norm": 0.08804110437631607,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 33696
+    },
+    {
+      "epoch": 3.4627003699136867,
+      "grad_norm": 0.05654985085129738,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 33699
+    },
+    {
+      "epoch": 3.463008631319359,
+      "grad_norm": 0.03474681079387665,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 33702
+    },
+    {
+      "epoch": 3.4633168927250306,
+      "grad_norm": 0.03495550900697708,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 33705
+    },
+    {
+      "epoch": 3.463625154130703,
+      "grad_norm": 0.07207003980875015,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 33708
+    },
+    {
+      "epoch": 3.463933415536375,
+      "grad_norm": 0.10733482986688614,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 33711
+    },
+    {
+      "epoch": 3.4642416769420468,
+      "grad_norm": 0.17830905318260193,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 33714
+    },
+    {
+      "epoch": 3.464549938347719,
+      "grad_norm": 0.10432233661413193,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 33717
+    },
+    {
+      "epoch": 3.4648581997533907,
+      "grad_norm": 0.08804329484701157,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 33720
+    },
+    {
+      "epoch": 3.465166461159063,
+      "grad_norm": 0.09405447542667389,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 33723
+    },
+    {
+      "epoch": 3.4654747225647347,
+      "grad_norm": 0.08153831958770752,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 33726
+    },
+    {
+      "epoch": 3.465782983970407,
+      "grad_norm": 0.04865524545311928,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 33729
+    },
+    {
+      "epoch": 3.466091245376079,
+      "grad_norm": 0.0705978274345398,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 33732
+    },
+    {
+      "epoch": 3.466399506781751,
+      "grad_norm": 0.08283443003892899,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 33735
+    },
+    {
+      "epoch": 3.466707768187423,
+      "grad_norm": 0.059983205050230026,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 33738
+    },
+    {
+      "epoch": 3.467016029593095,
+      "grad_norm": 0.045960474759340286,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 33741
+    },
+    {
+      "epoch": 3.467324290998767,
+      "grad_norm": 0.03882883861660957,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 33744
+    },
+    {
+      "epoch": 3.467632552404439,
+      "grad_norm": 0.07004483044147491,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 33747
+    },
+    {
+      "epoch": 3.467940813810111,
+      "grad_norm": 0.047444459050893784,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 33750
+    },
+    {
+      "epoch": 3.468249075215783,
+      "grad_norm": 0.04262397438287735,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 33753
+    },
+    {
+      "epoch": 3.468557336621455,
+      "grad_norm": 0.05599913001060486,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 33756
+    },
+    {
+      "epoch": 3.468865598027127,
+      "grad_norm": 0.03917532414197922,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 33759
+    },
+    {
+      "epoch": 3.469173859432799,
+      "grad_norm": 0.14925096929073334,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 33762
+    },
+    {
+      "epoch": 3.469482120838471,
+      "grad_norm": 0.09570999443531036,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 33765
+    },
+    {
+      "epoch": 3.469790382244143,
+      "grad_norm": 0.07883327454328537,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 33768
+    },
+    {
+      "epoch": 3.470098643649815,
+      "grad_norm": 0.06480975449085236,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 33771
+    },
+    {
+      "epoch": 3.470406905055487,
+      "grad_norm": 0.054496169090270996,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 33774
+    },
+    {
+      "epoch": 3.470715166461159,
+      "grad_norm": 0.04248659685254097,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 33777
+    },
+    {
+      "epoch": 3.471023427866831,
+      "grad_norm": 0.04117753356695175,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 33780
+    },
+    {
+      "epoch": 3.4713316892725032,
+      "grad_norm": 0.06876087188720703,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 33783
+    },
+    {
+      "epoch": 3.471639950678175,
+      "grad_norm": 0.08680430799722672,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 33786
+    },
+    {
+      "epoch": 3.471948212083847,
+      "grad_norm": 0.07227423787117004,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 33789
+    },
+    {
+      "epoch": 3.472256473489519,
+      "grad_norm": 0.04946593940258026,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 33792
+    },
+    {
+      "epoch": 3.472564734895191,
+      "grad_norm": 0.05215618014335632,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 33795
+    },
+    {
+      "epoch": 3.4728729963008633,
+      "grad_norm": 0.044869258999824524,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 33798
+    },
+    {
+      "epoch": 3.473181257706535,
+      "grad_norm": 0.048952534794807434,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 33801
+    },
+    {
+      "epoch": 3.4734895191122073,
+      "grad_norm": 0.031407974660396576,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 33804
+    },
+    {
+      "epoch": 3.473797780517879,
+      "grad_norm": 0.033864185214042664,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 33807
+    },
+    {
+      "epoch": 3.474106041923551,
+      "grad_norm": 0.08060206472873688,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 33810
+    },
+    {
+      "epoch": 3.4744143033292234,
+      "grad_norm": 0.059409331530332565,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 33813
+    },
+    {
+      "epoch": 3.474722564734895,
+      "grad_norm": 0.04131161794066429,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 33816
+    },
+    {
+      "epoch": 3.4750308261405674,
+      "grad_norm": 0.04256697744131088,
+      "learning_rate": 0.01,
+      "loss": 2.0479,
+      "step": 33819
+    },
+    {
+      "epoch": 3.475339087546239,
+      "grad_norm": 0.08916765451431274,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 33822
+    },
+    {
+      "epoch": 3.4756473489519113,
+      "grad_norm": 0.08474129438400269,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 33825
+    },
+    {
+      "epoch": 3.475955610357583,
+      "grad_norm": 0.05649823695421219,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 33828
+    },
+    {
+      "epoch": 3.4762638717632552,
+      "grad_norm": 0.05935325473546982,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 33831
+    },
+    {
+      "epoch": 3.4765721331689274,
+      "grad_norm": 0.04731081798672676,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 33834
+    },
+    {
+      "epoch": 3.476880394574599,
+      "grad_norm": 0.03790346160531044,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 33837
+    },
+    {
+      "epoch": 3.4771886559802714,
+      "grad_norm": 0.03726886212825775,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 33840
+    },
+    {
+      "epoch": 3.477496917385943,
+      "grad_norm": 0.07210247963666916,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 33843
+    },
+    {
+      "epoch": 3.4778051787916153,
+      "grad_norm": 0.07953071594238281,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 33846
+    },
+    {
+      "epoch": 3.478113440197287,
+      "grad_norm": 0.08004479855298996,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 33849
+    },
+    {
+      "epoch": 3.4784217016029593,
+      "grad_norm": 0.12617255747318268,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 33852
+    },
+    {
+      "epoch": 3.4787299630086315,
+      "grad_norm": 0.08742248266935349,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 33855
+    },
+    {
+      "epoch": 3.479038224414303,
+      "grad_norm": 0.07119353115558624,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 33858
+    },
+    {
+      "epoch": 3.4793464858199754,
+      "grad_norm": 0.06368902325630188,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 33861
+    },
+    {
+      "epoch": 3.479654747225647,
+      "grad_norm": 0.04121660441160202,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 33864
+    },
+    {
+      "epoch": 3.4799630086313194,
+      "grad_norm": 0.1237914115190506,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 33867
+    },
+    {
+      "epoch": 3.4802712700369915,
+      "grad_norm": 0.02918284945189953,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 33870
+    },
+    {
+      "epoch": 3.4805795314426633,
+      "grad_norm": 0.0423772819340229,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 33873
+    },
+    {
+      "epoch": 3.4808877928483355,
+      "grad_norm": 0.05689895898103714,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 33876
+    },
+    {
+      "epoch": 3.4811960542540072,
+      "grad_norm": 0.057178955525159836,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 33879
+    },
+    {
+      "epoch": 3.4815043156596794,
+      "grad_norm": 0.04840772971510887,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 33882
+    },
+    {
+      "epoch": 3.4818125770653516,
+      "grad_norm": 0.0467451848089695,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 33885
+    },
+    {
+      "epoch": 3.4821208384710234,
+      "grad_norm": 0.08849336951971054,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 33888
+    },
+    {
+      "epoch": 3.4824290998766956,
+      "grad_norm": 0.040159258991479874,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 33891
+    },
+    {
+      "epoch": 3.4827373612823673,
+      "grad_norm": 0.03237557038664818,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 33894
+    },
+    {
+      "epoch": 3.4830456226880395,
+      "grad_norm": 0.11882402747869492,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 33897
+    },
+    {
+      "epoch": 3.4833538840937113,
+      "grad_norm": 0.04206301271915436,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 33900
+    },
+    {
+      "epoch": 3.4836621454993835,
+      "grad_norm": 0.08875782787799835,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 33903
+    },
+    {
+      "epoch": 3.4839704069050557,
+      "grad_norm": 0.07771812379360199,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 33906
+    },
+    {
+      "epoch": 3.4842786683107274,
+      "grad_norm": 0.0839293822646141,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 33909
+    },
+    {
+      "epoch": 3.4845869297163996,
+      "grad_norm": 0.051187288016080856,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 33912
+    },
+    {
+      "epoch": 3.4848951911220714,
+      "grad_norm": 0.07497724145650864,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 33915
+    },
+    {
+      "epoch": 3.4852034525277436,
+      "grad_norm": 0.09341751039028168,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 33918
+    },
+    {
+      "epoch": 3.4855117139334153,
+      "grad_norm": 0.17380410432815552,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 33921
+    },
+    {
+      "epoch": 3.4858199753390875,
+      "grad_norm": 0.1407613456249237,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 33924
+    },
+    {
+      "epoch": 3.4861282367447597,
+      "grad_norm": 0.04020331799983978,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 33927
+    },
+    {
+      "epoch": 3.4864364981504314,
+      "grad_norm": 0.07008705288171768,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 33930
+    },
+    {
+      "epoch": 3.4867447595561036,
+      "grad_norm": 0.04310291260480881,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 33933
+    },
+    {
+      "epoch": 3.487053020961776,
+      "grad_norm": 0.03993843123316765,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 33936
+    },
+    {
+      "epoch": 3.4873612823674476,
+      "grad_norm": 0.04465002939105034,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 33939
+    },
+    {
+      "epoch": 3.4876695437731198,
+      "grad_norm": 0.03715788200497627,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 33942
+    },
+    {
+      "epoch": 3.4879778051787915,
+      "grad_norm": 0.06855741888284683,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 33945
+    },
+    {
+      "epoch": 3.4882860665844637,
+      "grad_norm": 0.05004393681883812,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 33948
+    },
+    {
+      "epoch": 3.4885943279901355,
+      "grad_norm": 0.11410415172576904,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 33951
+    },
+    {
+      "epoch": 3.4889025893958077,
+      "grad_norm": 0.09844734519720078,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 33954
+    },
+    {
+      "epoch": 3.48921085080148,
+      "grad_norm": 0.06708649545907974,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 33957
+    },
+    {
+      "epoch": 3.4895191122071516,
+      "grad_norm": 0.10117511451244354,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 33960
+    },
+    {
+      "epoch": 3.489827373612824,
+      "grad_norm": 0.09758836030960083,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 33963
+    },
+    {
+      "epoch": 3.4901356350184956,
+      "grad_norm": 0.05189559981226921,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 33966
+    },
+    {
+      "epoch": 3.4904438964241677,
+      "grad_norm": 0.04805564507842064,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 33969
+    },
+    {
+      "epoch": 3.4907521578298395,
+      "grad_norm": 0.10716047137975693,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 33972
+    },
+    {
+      "epoch": 3.4910604192355117,
+      "grad_norm": 0.0761241689324379,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 33975
+    },
+    {
+      "epoch": 3.491368680641184,
+      "grad_norm": 0.09611841291189194,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 33978
+    },
+    {
+      "epoch": 3.4916769420468556,
+      "grad_norm": 0.05685526877641678,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 33981
+    },
+    {
+      "epoch": 3.491985203452528,
+      "grad_norm": 0.044158194214105606,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 33984
+    },
+    {
+      "epoch": 3.4922934648581996,
+      "grad_norm": 0.0626639872789383,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 33987
+    },
+    {
+      "epoch": 3.4926017262638718,
+      "grad_norm": 0.0559966154396534,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 33990
+    },
+    {
+      "epoch": 3.4929099876695435,
+      "grad_norm": 0.05526448041200638,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 33993
+    },
+    {
+      "epoch": 3.4932182490752157,
+      "grad_norm": 0.07910171896219254,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 33996
+    },
+    {
+      "epoch": 3.493526510480888,
+      "grad_norm": 0.09646596759557724,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 33999
+    },
+    {
+      "epoch": 3.4938347718865597,
+      "grad_norm": 0.06635434925556183,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 34002
+    },
+    {
+      "epoch": 3.494143033292232,
+      "grad_norm": 0.04328935965895653,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 34005
+    },
+    {
+      "epoch": 3.494451294697904,
+      "grad_norm": 0.040060825645923615,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 34008
+    },
+    {
+      "epoch": 3.494759556103576,
+      "grad_norm": 0.0396769754588604,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 34011
+    },
+    {
+      "epoch": 3.495067817509248,
+      "grad_norm": 0.035870879888534546,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 34014
+    },
+    {
+      "epoch": 3.4953760789149197,
+      "grad_norm": 0.07099170982837677,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 34017
+    },
+    {
+      "epoch": 3.495684340320592,
+      "grad_norm": 0.06994140148162842,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 34020
+    },
+    {
+      "epoch": 3.4959926017262637,
+      "grad_norm": 0.03673701733350754,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 34023
+    },
+    {
+      "epoch": 3.496300863131936,
+      "grad_norm": 0.036696165800094604,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 34026
+    },
+    {
+      "epoch": 3.496609124537608,
+      "grad_norm": 0.1089354008436203,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 34029
+    },
+    {
+      "epoch": 3.49691738594328,
+      "grad_norm": 0.04758044332265854,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 34032
+    },
+    {
+      "epoch": 3.497225647348952,
+      "grad_norm": 0.05122329294681549,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 34035
+    },
+    {
+      "epoch": 3.4975339087546238,
+      "grad_norm": 0.04168769717216492,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 34038
+    },
+    {
+      "epoch": 3.497842170160296,
+      "grad_norm": 0.03426910564303398,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 34041
+    },
+    {
+      "epoch": 3.4981504315659677,
+      "grad_norm": 0.036272600293159485,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 34044
+    },
+    {
+      "epoch": 3.49845869297164,
+      "grad_norm": 0.11142997443675995,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 34047
+    },
+    {
+      "epoch": 3.498766954377312,
+      "grad_norm": 0.08410037308931351,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 34050
+    },
+    {
+      "epoch": 3.499075215782984,
+      "grad_norm": 0.10037896782159805,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 34053
+    },
+    {
+      "epoch": 3.499383477188656,
+      "grad_norm": 0.05194736644625664,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 34056
+    },
+    {
+      "epoch": 3.499691738594328,
+      "grad_norm": 0.037128567695617676,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 34059
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.059042248874902725,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 34062
+    },
+    {
+      "epoch": 3.5003082614056718,
+      "grad_norm": 0.06282515078783035,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 34065
+    },
+    {
+      "epoch": 3.500616522811344,
+      "grad_norm": 0.04248201847076416,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 34068
+    },
+    {
+      "epoch": 3.500924784217016,
+      "grad_norm": 0.047459498047828674,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 34071
+    },
+    {
+      "epoch": 3.501233045622688,
+      "grad_norm": 0.04447497799992561,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 34074
+    },
+    {
+      "epoch": 3.50154130702836,
+      "grad_norm": 0.03569423034787178,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 34077
+    },
+    {
+      "epoch": 3.5018495684340323,
+      "grad_norm": 0.055864010006189346,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 34080
+    },
+    {
+      "epoch": 3.502157829839704,
+      "grad_norm": 0.055652111768722534,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 34083
+    },
+    {
+      "epoch": 3.5024660912453762,
+      "grad_norm": 0.1195710226893425,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 34086
+    },
+    {
+      "epoch": 3.502774352651048,
+      "grad_norm": 0.04436422884464264,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 34089
+    },
+    {
+      "epoch": 3.50308261405672,
+      "grad_norm": 0.047165922820568085,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 34092
+    },
+    {
+      "epoch": 3.503390875462392,
+      "grad_norm": 0.07143979519605637,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 34095
+    },
+    {
+      "epoch": 3.503699136868064,
+      "grad_norm": 0.04734091833233833,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 34098
+    },
+    {
+      "epoch": 3.5040073982737363,
+      "grad_norm": 0.05058762803673744,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 34101
+    },
+    {
+      "epoch": 3.504315659679408,
+      "grad_norm": 0.03917768597602844,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 34104
+    },
+    {
+      "epoch": 3.5046239210850803,
+      "grad_norm": 0.06384671479463577,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 34107
+    },
+    {
+      "epoch": 3.504932182490752,
+      "grad_norm": 0.046532079577445984,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 34110
+    },
+    {
+      "epoch": 3.505240443896424,
+      "grad_norm": 0.033960457891225815,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 34113
+    },
+    {
+      "epoch": 3.505548705302096,
+      "grad_norm": 0.061024975031614304,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 34116
+    },
+    {
+      "epoch": 3.505856966707768,
+      "grad_norm": 0.09465770423412323,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 34119
+    },
+    {
+      "epoch": 3.5061652281134403,
+      "grad_norm": 0.08510823547840118,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 34122
+    },
+    {
+      "epoch": 3.506473489519112,
+      "grad_norm": 0.12201209366321564,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 34125
+    },
+    {
+      "epoch": 3.5067817509247843,
+      "grad_norm": 0.08503463119268417,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 34128
+    },
+    {
+      "epoch": 3.5070900123304565,
+      "grad_norm": 0.07231762260198593,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 34131
+    },
+    {
+      "epoch": 3.5073982737361282,
+      "grad_norm": 0.042714521288871765,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 34134
+    },
+    {
+      "epoch": 3.5077065351418,
+      "grad_norm": 0.06119026616215706,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 34137
+    },
+    {
+      "epoch": 3.508014796547472,
+      "grad_norm": 0.047117751091718674,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 34140
+    },
+    {
+      "epoch": 3.5083230579531444,
+      "grad_norm": 0.06318202614784241,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 34143
+    },
+    {
+      "epoch": 3.508631319358816,
+      "grad_norm": 0.05050423741340637,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 34146
+    },
+    {
+      "epoch": 3.5089395807644883,
+      "grad_norm": 0.06105552241206169,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 34149
+    },
+    {
+      "epoch": 3.5092478421701605,
+      "grad_norm": 0.05940508469939232,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 34152
+    },
+    {
+      "epoch": 3.5095561035758323,
+      "grad_norm": 0.05157890543341637,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 34155
+    },
+    {
+      "epoch": 3.5098643649815044,
+      "grad_norm": 0.06597436964511871,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 34158
+    },
+    {
+      "epoch": 3.510172626387176,
+      "grad_norm": 0.10424167662858963,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 34161
+    },
+    {
+      "epoch": 3.5104808877928484,
+      "grad_norm": 0.09572573006153107,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 34164
+    },
+    {
+      "epoch": 3.51078914919852,
+      "grad_norm": 0.07485774159431458,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 34167
+    },
+    {
+      "epoch": 3.5110974106041923,
+      "grad_norm": 0.12162437289953232,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 34170
+    },
+    {
+      "epoch": 3.5114056720098645,
+      "grad_norm": 0.09338829666376114,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 34173
+    },
+    {
+      "epoch": 3.5117139334155363,
+      "grad_norm": 0.06095868721604347,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 34176
+    },
+    {
+      "epoch": 3.5120221948212085,
+      "grad_norm": 0.057355985045433044,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 34179
+    },
+    {
+      "epoch": 3.5123304562268807,
+      "grad_norm": 0.10891444981098175,
+      "learning_rate": 0.01,
+      "loss": 2.0339,
+      "step": 34182
+    },
+    {
+      "epoch": 3.5126387176325524,
+      "grad_norm": 0.10567829012870789,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 34185
+    },
+    {
+      "epoch": 3.512946979038224,
+      "grad_norm": 0.05244187265634537,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 34188
+    },
+    {
+      "epoch": 3.5132552404438964,
+      "grad_norm": 0.046076592057943344,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 34191
+    },
+    {
+      "epoch": 3.5135635018495686,
+      "grad_norm": 0.04405316337943077,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 34194
+    },
+    {
+      "epoch": 3.5138717632552403,
+      "grad_norm": 0.03997344523668289,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 34197
+    },
+    {
+      "epoch": 3.5141800246609125,
+      "grad_norm": 0.03909468650817871,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 34200
+    },
+    {
+      "epoch": 3.5144882860665847,
+      "grad_norm": 0.11657961457967758,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 34203
+    },
+    {
+      "epoch": 3.5147965474722564,
+      "grad_norm": 0.11744063347578049,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 34206
+    },
+    {
+      "epoch": 3.5151048088779286,
+      "grad_norm": 0.039932142943143845,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 34209
+    },
+    {
+      "epoch": 3.5154130702836004,
+      "grad_norm": 0.03299909457564354,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 34212
+    },
+    {
+      "epoch": 3.5157213316892726,
+      "grad_norm": 0.1055513322353363,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 34215
+    },
+    {
+      "epoch": 3.5160295930949443,
+      "grad_norm": 0.05073374882340431,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 34218
+    },
+    {
+      "epoch": 3.5163378545006165,
+      "grad_norm": 0.05423841252923012,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 34221
+    },
+    {
+      "epoch": 3.5166461159062887,
+      "grad_norm": 0.05695211887359619,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 34224
+    },
+    {
+      "epoch": 3.5169543773119605,
+      "grad_norm": 0.04092913866043091,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 34227
+    },
+    {
+      "epoch": 3.5172626387176327,
+      "grad_norm": 0.05083661898970604,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 34230
+    },
+    {
+      "epoch": 3.5175709001233044,
+      "grad_norm": 0.04499472677707672,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 34233
+    },
+    {
+      "epoch": 3.5178791615289766,
+      "grad_norm": 0.09770061075687408,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 34236
+    },
+    {
+      "epoch": 3.5181874229346484,
+      "grad_norm": 0.09951679408550262,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 34239
+    },
+    {
+      "epoch": 3.5184956843403206,
+      "grad_norm": 0.11089123040437698,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 34242
+    },
+    {
+      "epoch": 3.5188039457459928,
+      "grad_norm": 0.05511726066470146,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 34245
+    },
+    {
+      "epoch": 3.5191122071516645,
+      "grad_norm": 0.035139378160238266,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 34248
+    },
+    {
+      "epoch": 3.5194204685573367,
+      "grad_norm": 0.05882834270596504,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 34251
+    },
+    {
+      "epoch": 3.519728729963009,
+      "grad_norm": 0.06780868023633957,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 34254
+    },
+    {
+      "epoch": 3.5200369913686806,
+      "grad_norm": 0.053879499435424805,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 34257
+    },
+    {
+      "epoch": 3.5203452527743524,
+      "grad_norm": 0.045473113656044006,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 34260
+    },
+    {
+      "epoch": 3.5206535141800246,
+      "grad_norm": 0.050460174679756165,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 34263
+    },
+    {
+      "epoch": 3.520961775585697,
+      "grad_norm": 0.040959011763334274,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 34266
+    },
+    {
+      "epoch": 3.5212700369913685,
+      "grad_norm": 0.048418451100587845,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 34269
+    },
+    {
+      "epoch": 3.5215782983970407,
+      "grad_norm": 0.06287720799446106,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 34272
+    },
+    {
+      "epoch": 3.521886559802713,
+      "grad_norm": 0.04529783874750137,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 34275
+    },
+    {
+      "epoch": 3.5221948212083847,
+      "grad_norm": 0.03729906305670738,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 34278
+    },
+    {
+      "epoch": 3.522503082614057,
+      "grad_norm": 0.1012713834643364,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 34281
+    },
+    {
+      "epoch": 3.5228113440197286,
+      "grad_norm": 0.10641775280237198,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 34284
+    },
+    {
+      "epoch": 3.523119605425401,
+      "grad_norm": 0.09289912134408951,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 34287
+    },
+    {
+      "epoch": 3.5234278668310726,
+      "grad_norm": 0.13354292511940002,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 34290
+    },
+    {
+      "epoch": 3.5237361282367448,
+      "grad_norm": 0.08027501404285431,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 34293
+    },
+    {
+      "epoch": 3.524044389642417,
+      "grad_norm": 0.047286976128816605,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 34296
+    },
+    {
+      "epoch": 3.5243526510480887,
+      "grad_norm": 0.042211420834064484,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 34299
+    },
+    {
+      "epoch": 3.524660912453761,
+      "grad_norm": 0.11374162137508392,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 34302
+    },
+    {
+      "epoch": 3.5249691738594326,
+      "grad_norm": 0.05604710429906845,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 34305
+    },
+    {
+      "epoch": 3.525277435265105,
+      "grad_norm": 0.09744080156087875,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 34308
+    },
+    {
+      "epoch": 3.5255856966707766,
+      "grad_norm": 0.08903683722019196,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 34311
+    },
+    {
+      "epoch": 3.525893958076449,
+      "grad_norm": 0.048648543655872345,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 34314
+    },
+    {
+      "epoch": 3.526202219482121,
+      "grad_norm": 0.0710827112197876,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 34317
+    },
+    {
+      "epoch": 3.5265104808877927,
+      "grad_norm": 0.04736728593707085,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 34320
+    },
+    {
+      "epoch": 3.526818742293465,
+      "grad_norm": 0.05635381117463112,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 34323
+    },
+    {
+      "epoch": 3.527127003699137,
+      "grad_norm": 0.06686391681432724,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 34326
+    },
+    {
+      "epoch": 3.527435265104809,
+      "grad_norm": 0.03842944651842117,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 34329
+    },
+    {
+      "epoch": 3.5277435265104806,
+      "grad_norm": 0.042534928768873215,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 34332
+    },
+    {
+      "epoch": 3.528051787916153,
+      "grad_norm": 0.06037287786602974,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 34335
+    },
+    {
+      "epoch": 3.528360049321825,
+      "grad_norm": 0.06533370167016983,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 34338
+    },
+    {
+      "epoch": 3.5286683107274968,
+      "grad_norm": 0.0877017006278038,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 34341
+    },
+    {
+      "epoch": 3.528976572133169,
+      "grad_norm": 0.09864972531795502,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 34344
+    },
+    {
+      "epoch": 3.529284833538841,
+      "grad_norm": 0.049320898950099945,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 34347
+    },
+    {
+      "epoch": 3.529593094944513,
+      "grad_norm": 0.12996013462543488,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 34350
+    },
+    {
+      "epoch": 3.529901356350185,
+      "grad_norm": 0.06548713147640228,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 34353
+    },
+    {
+      "epoch": 3.530209617755857,
+      "grad_norm": 0.045482341200113297,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 34356
+    },
+    {
+      "epoch": 3.530517879161529,
+      "grad_norm": 0.04433637857437134,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 34359
+    },
+    {
+      "epoch": 3.530826140567201,
+      "grad_norm": 0.04450944438576698,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 34362
+    },
+    {
+      "epoch": 3.531134401972873,
+      "grad_norm": 0.07188856601715088,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 34365
+    },
+    {
+      "epoch": 3.531442663378545,
+      "grad_norm": 0.049853190779685974,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 34368
+    },
+    {
+      "epoch": 3.531750924784217,
+      "grad_norm": 0.08662360161542892,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 34371
+    },
+    {
+      "epoch": 3.532059186189889,
+      "grad_norm": 0.07737040519714355,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 34374
+    },
+    {
+      "epoch": 3.532367447595561,
+      "grad_norm": 0.0868213102221489,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 34377
+    },
+    {
+      "epoch": 3.532675709001233,
+      "grad_norm": 0.06288056075572968,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 34380
+    },
+    {
+      "epoch": 3.532983970406905,
+      "grad_norm": 0.05537475645542145,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 34383
+    },
+    {
+      "epoch": 3.533292231812577,
+      "grad_norm": 0.13551141321659088,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 34386
+    },
+    {
+      "epoch": 3.533600493218249,
+      "grad_norm": 0.040236156433820724,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 34389
+    },
+    {
+      "epoch": 3.533908754623921,
+      "grad_norm": 0.06904727220535278,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 34392
+    },
+    {
+      "epoch": 3.534217016029593,
+      "grad_norm": 0.06675262004137039,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 34395
+    },
+    {
+      "epoch": 3.5345252774352653,
+      "grad_norm": 0.05011274665594101,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 34398
+    },
+    {
+      "epoch": 3.534833538840937,
+      "grad_norm": 0.04494976997375488,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 34401
+    },
+    {
+      "epoch": 3.5351418002466093,
+      "grad_norm": 0.04114719480276108,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 34404
+    },
+    {
+      "epoch": 3.535450061652281,
+      "grad_norm": 0.03544189780950546,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 34407
+    },
+    {
+      "epoch": 3.5357583230579532,
+      "grad_norm": 0.10223556309938431,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 34410
+    },
+    {
+      "epoch": 3.536066584463625,
+      "grad_norm": 0.03573578596115112,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 34413
+    },
+    {
+      "epoch": 3.536374845869297,
+      "grad_norm": 0.08199316263198853,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 34416
+    },
+    {
+      "epoch": 3.5366831072749694,
+      "grad_norm": 0.04587673768401146,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 34419
+    },
+    {
+      "epoch": 3.536991368680641,
+      "grad_norm": 0.061022888869047165,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 34422
+    },
+    {
+      "epoch": 3.5372996300863133,
+      "grad_norm": 0.09616536647081375,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 34425
+    },
+    {
+      "epoch": 3.537607891491985,
+      "grad_norm": 0.10521430522203445,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 34428
+    },
+    {
+      "epoch": 3.5379161528976573,
+      "grad_norm": 0.0551154688000679,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 34431
+    },
+    {
+      "epoch": 3.538224414303329,
+      "grad_norm": 0.0405128113925457,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 34434
+    },
+    {
+      "epoch": 3.538532675709001,
+      "grad_norm": 0.035197604447603226,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 34437
+    },
+    {
+      "epoch": 3.5388409371146734,
+      "grad_norm": 0.026394899934530258,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 34440
+    },
+    {
+      "epoch": 3.539149198520345,
+      "grad_norm": 0.09831299632787704,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 34443
+    },
+    {
+      "epoch": 3.5394574599260173,
+      "grad_norm": 0.08986491709947586,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 34446
+    },
+    {
+      "epoch": 3.5397657213316895,
+      "grad_norm": 0.07974385470151901,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 34449
+    },
+    {
+      "epoch": 3.5400739827373613,
+      "grad_norm": 0.11314170062541962,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 34452
+    },
+    {
+      "epoch": 3.540382244143033,
+      "grad_norm": 0.05535215139389038,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 34455
+    },
+    {
+      "epoch": 3.5406905055487052,
+      "grad_norm": 0.0592242069542408,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 34458
+    },
+    {
+      "epoch": 3.5409987669543774,
+      "grad_norm": 0.12061761319637299,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 34461
+    },
+    {
+      "epoch": 3.541307028360049,
+      "grad_norm": 0.11443237960338593,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 34464
+    },
+    {
+      "epoch": 3.5416152897657214,
+      "grad_norm": 0.050069622695446014,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 34467
+    },
+    {
+      "epoch": 3.5419235511713936,
+      "grad_norm": 0.0419803261756897,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 34470
+    },
+    {
+      "epoch": 3.5422318125770653,
+      "grad_norm": 0.03493333235383034,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 34473
+    },
+    {
+      "epoch": 3.5425400739827375,
+      "grad_norm": 0.036889005452394485,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 34476
+    },
+    {
+      "epoch": 3.5428483353884093,
+      "grad_norm": 0.05217687413096428,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 34479
+    },
+    {
+      "epoch": 3.5431565967940815,
+      "grad_norm": 0.050754569470882416,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 34482
+    },
+    {
+      "epoch": 3.543464858199753,
+      "grad_norm": 0.04706338793039322,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 34485
+    },
+    {
+      "epoch": 3.5437731196054254,
+      "grad_norm": 0.07039839029312134,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 34488
+    },
+    {
+      "epoch": 3.5440813810110976,
+      "grad_norm": 0.07069671154022217,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 34491
+    },
+    {
+      "epoch": 3.5443896424167693,
+      "grad_norm": 0.09058693796396255,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 34494
+    },
+    {
+      "epoch": 3.5446979038224415,
+      "grad_norm": 0.037666015326976776,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 34497
+    },
+    {
+      "epoch": 3.5450061652281133,
+      "grad_norm": 0.04604499414563179,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 34500
+    },
+    {
+      "epoch": 3.5453144266337855,
+      "grad_norm": 0.03853873908519745,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 34503
+    },
+    {
+      "epoch": 3.5456226880394572,
+      "grad_norm": 0.07123208791017532,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 34506
+    },
+    {
+      "epoch": 3.5459309494451294,
+      "grad_norm": 0.0936029776930809,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 34509
+    },
+    {
+      "epoch": 3.5462392108508016,
+      "grad_norm": 0.07412207871675491,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 34512
+    },
+    {
+      "epoch": 3.5465474722564734,
+      "grad_norm": 0.07681329548358917,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 34515
+    },
+    {
+      "epoch": 3.5468557336621456,
+      "grad_norm": 0.10403033345937729,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 34518
+    },
+    {
+      "epoch": 3.5471639950678178,
+      "grad_norm": 0.14240513741970062,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 34521
+    },
+    {
+      "epoch": 3.5474722564734895,
+      "grad_norm": 0.14203394949436188,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 34524
+    },
+    {
+      "epoch": 3.5477805178791613,
+      "grad_norm": 0.09036475419998169,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 34527
+    },
+    {
+      "epoch": 3.5480887792848335,
+      "grad_norm": 0.037448760122060776,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 34530
+    },
+    {
+      "epoch": 3.5483970406905057,
+      "grad_norm": 0.05539664253592491,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 34533
+    },
+    {
+      "epoch": 3.5487053020961774,
+      "grad_norm": 0.05857717618346214,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 34536
+    },
+    {
+      "epoch": 3.5490135635018496,
+      "grad_norm": 0.06406868249177933,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 34539
+    },
+    {
+      "epoch": 3.549321824907522,
+      "grad_norm": 0.04234686121344566,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 34542
+    },
+    {
+      "epoch": 3.5496300863131935,
+      "grad_norm": 0.038916390389204025,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 34545
+    },
+    {
+      "epoch": 3.5499383477188657,
+      "grad_norm": 0.039077602326869965,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 34548
+    },
+    {
+      "epoch": 3.5502466091245375,
+      "grad_norm": 0.03960174322128296,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 34551
+    },
+    {
+      "epoch": 3.5505548705302097,
+      "grad_norm": 0.12132997810840607,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 34554
+    },
+    {
+      "epoch": 3.5508631319358814,
+      "grad_norm": 0.09045036137104034,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 34557
+    },
+    {
+      "epoch": 3.5511713933415536,
+      "grad_norm": 0.07062508165836334,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 34560
+    },
+    {
+      "epoch": 3.551479654747226,
+      "grad_norm": 0.08182030916213989,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 34563
+    },
+    {
+      "epoch": 3.5517879161528976,
+      "grad_norm": 0.05917488783597946,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 34566
+    },
+    {
+      "epoch": 3.5520961775585698,
+      "grad_norm": 0.08519969880580902,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 34569
+    },
+    {
+      "epoch": 3.5524044389642415,
+      "grad_norm": 0.04442654922604561,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 34572
+    },
+    {
+      "epoch": 3.5527127003699137,
+      "grad_norm": 0.08265768736600876,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 34575
+    },
+    {
+      "epoch": 3.5530209617755855,
+      "grad_norm": 0.0891944020986557,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 34578
+    },
+    {
+      "epoch": 3.5533292231812577,
+      "grad_norm": 0.0688168928027153,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 34581
+    },
+    {
+      "epoch": 3.55363748458693,
+      "grad_norm": 0.0936504453420639,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 34584
+    },
+    {
+      "epoch": 3.5539457459926016,
+      "grad_norm": 0.10172853618860245,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 34587
+    },
+    {
+      "epoch": 3.554254007398274,
+      "grad_norm": 0.04822350665926933,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 34590
+    },
+    {
+      "epoch": 3.554562268803946,
+      "grad_norm": 0.05222393944859505,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 34593
+    },
+    {
+      "epoch": 3.5548705302096177,
+      "grad_norm": 0.03592358157038689,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 34596
+    },
+    {
+      "epoch": 3.5551787916152895,
+      "grad_norm": 0.03903461620211601,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 34599
+    },
+    {
+      "epoch": 3.5554870530209617,
+      "grad_norm": 0.0467611663043499,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 34602
+    },
+    {
+      "epoch": 3.555795314426634,
+      "grad_norm": 0.12140758335590363,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 34605
+    },
+    {
+      "epoch": 3.5561035758323056,
+      "grad_norm": 0.04155382886528969,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 34608
+    },
+    {
+      "epoch": 3.556411837237978,
+      "grad_norm": 0.039924267679452896,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 34611
+    },
+    {
+      "epoch": 3.55672009864365,
+      "grad_norm": 0.10513463616371155,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 34614
+    },
+    {
+      "epoch": 3.5570283600493218,
+      "grad_norm": 0.08956603705883026,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 34617
+    },
+    {
+      "epoch": 3.557336621454994,
+      "grad_norm": 0.045444682240486145,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 34620
+    },
+    {
+      "epoch": 3.5576448828606657,
+      "grad_norm": 0.05367853119969368,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 34623
+    },
+    {
+      "epoch": 3.557953144266338,
+      "grad_norm": 0.0465155765414238,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 34626
+    },
+    {
+      "epoch": 3.5582614056720097,
+      "grad_norm": 0.06788338720798492,
+      "learning_rate": 0.01,
+      "loss": 2.0526,
+      "step": 34629
+    },
+    {
+      "epoch": 3.558569667077682,
+      "grad_norm": 0.04508093744516373,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 34632
+    },
+    {
+      "epoch": 3.558877928483354,
+      "grad_norm": 0.05827740207314491,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 34635
+    },
+    {
+      "epoch": 3.559186189889026,
+      "grad_norm": 0.09915097057819366,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 34638
+    },
+    {
+      "epoch": 3.559494451294698,
+      "grad_norm": 0.08571092784404755,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 34641
+    },
+    {
+      "epoch": 3.55980271270037,
+      "grad_norm": 0.06480662524700165,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 34644
+    },
+    {
+      "epoch": 3.560110974106042,
+      "grad_norm": 0.055505797266960144,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 34647
+    },
+    {
+      "epoch": 3.5604192355117137,
+      "grad_norm": 0.05562606081366539,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 34650
+    },
+    {
+      "epoch": 3.560727496917386,
+      "grad_norm": 0.05720195546746254,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 34653
+    },
+    {
+      "epoch": 3.561035758323058,
+      "grad_norm": 0.14654351770877838,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 34656
+    },
+    {
+      "epoch": 3.56134401972873,
+      "grad_norm": 0.04594961181282997,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 34659
+    },
+    {
+      "epoch": 3.561652281134402,
+      "grad_norm": 0.052737195044755936,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 34662
+    },
+    {
+      "epoch": 3.561960542540074,
+      "grad_norm": 0.0722641721367836,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 34665
+    },
+    {
+      "epoch": 3.562268803945746,
+      "grad_norm": 0.03973691165447235,
+      "learning_rate": 0.01,
+      "loss": 2.0317,
+      "step": 34668
+    },
+    {
+      "epoch": 3.562577065351418,
+      "grad_norm": 0.06284237653017044,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 34671
+    },
+    {
+      "epoch": 3.56288532675709,
+      "grad_norm": 0.0880017802119255,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 34674
+    },
+    {
+      "epoch": 3.563193588162762,
+      "grad_norm": 0.06454182416200638,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 34677
+    },
+    {
+      "epoch": 3.563501849568434,
+      "grad_norm": 0.07054916024208069,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 34680
+    },
+    {
+      "epoch": 3.563810110974106,
+      "grad_norm": 0.08058945089578629,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 34683
+    },
+    {
+      "epoch": 3.5641183723797782,
+      "grad_norm": 0.08660910278558731,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 34686
+    },
+    {
+      "epoch": 3.56442663378545,
+      "grad_norm": 0.050544124096632004,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 34689
+    },
+    {
+      "epoch": 3.564734895191122,
+      "grad_norm": 0.056985314935445786,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 34692
+    },
+    {
+      "epoch": 3.565043156596794,
+      "grad_norm": 0.08204400539398193,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 34695
+    },
+    {
+      "epoch": 3.565351418002466,
+      "grad_norm": 0.09590046107769012,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 34698
+    },
+    {
+      "epoch": 3.565659679408138,
+      "grad_norm": 0.03453322499990463,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 34701
+    },
+    {
+      "epoch": 3.56596794081381,
+      "grad_norm": 0.03801025450229645,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 34704
+    },
+    {
+      "epoch": 3.5662762022194823,
+      "grad_norm": 0.05342378839850426,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 34707
+    },
+    {
+      "epoch": 3.566584463625154,
+      "grad_norm": 0.06871719658374786,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 34710
+    },
+    {
+      "epoch": 3.566892725030826,
+      "grad_norm": 0.06653191894292831,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 34713
+    },
+    {
+      "epoch": 3.5672009864364984,
+      "grad_norm": 0.10531385242938995,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 34716
+    },
+    {
+      "epoch": 3.56750924784217,
+      "grad_norm": 0.05481969192624092,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 34719
+    },
+    {
+      "epoch": 3.567817509247842,
+      "grad_norm": 0.04904542118310928,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 34722
+    },
+    {
+      "epoch": 3.568125770653514,
+      "grad_norm": 0.034647226333618164,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 34725
+    },
+    {
+      "epoch": 3.5684340320591863,
+      "grad_norm": 0.07234811037778854,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 34728
+    },
+    {
+      "epoch": 3.568742293464858,
+      "grad_norm": 0.10623595118522644,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 34731
+    },
+    {
+      "epoch": 3.5690505548705302,
+      "grad_norm": 0.04955185204744339,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 34734
+    },
+    {
+      "epoch": 3.5693588162762024,
+      "grad_norm": 0.08183744549751282,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 34737
+    },
+    {
+      "epoch": 3.569667077681874,
+      "grad_norm": 0.04063693434000015,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 34740
+    },
+    {
+      "epoch": 3.5699753390875464,
+      "grad_norm": 0.05661273002624512,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 34743
+    },
+    {
+      "epoch": 3.570283600493218,
+      "grad_norm": 0.03300248831510544,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 34746
+    },
+    {
+      "epoch": 3.5705918618988903,
+      "grad_norm": 0.09492766112089157,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 34749
+    },
+    {
+      "epoch": 3.570900123304562,
+      "grad_norm": 0.0505099892616272,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 34752
+    },
+    {
+      "epoch": 3.5712083847102343,
+      "grad_norm": 0.094505175948143,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 34755
+    },
+    {
+      "epoch": 3.5715166461159065,
+      "grad_norm": 0.046727851033210754,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 34758
+    },
+    {
+      "epoch": 3.571824907521578,
+      "grad_norm": 0.06659562885761261,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 34761
+    },
+    {
+      "epoch": 3.5721331689272504,
+      "grad_norm": 0.0828694999217987,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 34764
+    },
+    {
+      "epoch": 3.572441430332922,
+      "grad_norm": 0.12309973686933517,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 34767
+    },
+    {
+      "epoch": 3.5727496917385944,
+      "grad_norm": 0.03831718862056732,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 34770
+    },
+    {
+      "epoch": 3.573057953144266,
+      "grad_norm": 0.049565766006708145,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 34773
+    },
+    {
+      "epoch": 3.5733662145499383,
+      "grad_norm": 0.055259205400943756,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 34776
+    },
+    {
+      "epoch": 3.5736744759556105,
+      "grad_norm": 0.08979593217372894,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 34779
+    },
+    {
+      "epoch": 3.5739827373612822,
+      "grad_norm": 0.05166800692677498,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 34782
+    },
+    {
+      "epoch": 3.5742909987669544,
+      "grad_norm": 0.06904742866754532,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 34785
+    },
+    {
+      "epoch": 3.5745992601726266,
+      "grad_norm": 0.0667278841137886,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 34788
+    },
+    {
+      "epoch": 3.5749075215782984,
+      "grad_norm": 0.03483438491821289,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 34791
+    },
+    {
+      "epoch": 3.57521578298397,
+      "grad_norm": 0.08964372426271439,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 34794
+    },
+    {
+      "epoch": 3.5755240443896423,
+      "grad_norm": 0.09426937997341156,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 34797
+    },
+    {
+      "epoch": 3.5758323057953145,
+      "grad_norm": 0.0574275478720665,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 34800
+    },
+    {
+      "epoch": 3.5761405672009863,
+      "grad_norm": 0.03654215484857559,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 34803
+    },
+    {
+      "epoch": 3.5764488286066585,
+      "grad_norm": 0.05129201337695122,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 34806
+    },
+    {
+      "epoch": 3.5767570900123307,
+      "grad_norm": 0.07386317104101181,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 34809
+    },
+    {
+      "epoch": 3.5770653514180024,
+      "grad_norm": 0.1415167599916458,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 34812
+    },
+    {
+      "epoch": 3.5773736128236746,
+      "grad_norm": 0.04465651139616966,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 34815
+    },
+    {
+      "epoch": 3.5776818742293464,
+      "grad_norm": 0.05919940024614334,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 34818
+    },
+    {
+      "epoch": 3.5779901356350186,
+      "grad_norm": 0.03873635083436966,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 34821
+    },
+    {
+      "epoch": 3.5782983970406903,
+      "grad_norm": 0.05425990745425224,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 34824
+    },
+    {
+      "epoch": 3.5786066584463625,
+      "grad_norm": 0.11823546886444092,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 34827
+    },
+    {
+      "epoch": 3.5789149198520347,
+      "grad_norm": 0.12624318897724152,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 34830
+    },
+    {
+      "epoch": 3.5792231812577064,
+      "grad_norm": 0.05450008437037468,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 34833
+    },
+    {
+      "epoch": 3.5795314426633786,
+      "grad_norm": 0.04888772964477539,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 34836
+    },
+    {
+      "epoch": 3.579839704069051,
+      "grad_norm": 0.033485714346170425,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 34839
+    },
+    {
+      "epoch": 3.5801479654747226,
+      "grad_norm": 0.08570647239685059,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 34842
+    },
+    {
+      "epoch": 3.5804562268803943,
+      "grad_norm": 0.06110506132245064,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 34845
+    },
+    {
+      "epoch": 3.5807644882860665,
+      "grad_norm": 0.08815490454435349,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 34848
+    },
+    {
+      "epoch": 3.5810727496917387,
+      "grad_norm": 0.09731253236532211,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 34851
+    },
+    {
+      "epoch": 3.5813810110974105,
+      "grad_norm": 0.08022457361221313,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 34854
+    },
+    {
+      "epoch": 3.5816892725030827,
+      "grad_norm": 0.042772453278303146,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 34857
+    },
+    {
+      "epoch": 3.581997533908755,
+      "grad_norm": 0.030573785305023193,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 34860
+    },
+    {
+      "epoch": 3.5823057953144266,
+      "grad_norm": 0.05686675012111664,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 34863
+    },
+    {
+      "epoch": 3.582614056720099,
+      "grad_norm": 0.04291863366961479,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 34866
+    },
+    {
+      "epoch": 3.5829223181257706,
+      "grad_norm": 0.0715685561299324,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 34869
+    },
+    {
+      "epoch": 3.5832305795314427,
+      "grad_norm": 0.06964897364377975,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 34872
+    },
+    {
+      "epoch": 3.5835388409371145,
+      "grad_norm": 0.043709173798561096,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 34875
+    },
+    {
+      "epoch": 3.5838471023427867,
+      "grad_norm": 0.06593792140483856,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 34878
+    },
+    {
+      "epoch": 3.584155363748459,
+      "grad_norm": 0.04702428728342056,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 34881
+    },
+    {
+      "epoch": 3.5844636251541306,
+      "grad_norm": 0.08917208015918732,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 34884
+    },
+    {
+      "epoch": 3.584771886559803,
+      "grad_norm": 0.1167001873254776,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 34887
+    },
+    {
+      "epoch": 3.5850801479654746,
+      "grad_norm": 0.07775711268186569,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 34890
+    },
+    {
+      "epoch": 3.585388409371147,
+      "grad_norm": 0.08915986865758896,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 34893
+    },
+    {
+      "epoch": 3.5856966707768185,
+      "grad_norm": 0.045163143426179886,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 34896
+    },
+    {
+      "epoch": 3.5860049321824907,
+      "grad_norm": 0.0571911595761776,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 34899
+    },
+    {
+      "epoch": 3.586313193588163,
+      "grad_norm": 0.0707104504108429,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 34902
+    },
+    {
+      "epoch": 3.5866214549938347,
+      "grad_norm": 0.04241738095879555,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 34905
+    },
+    {
+      "epoch": 3.586929716399507,
+      "grad_norm": 0.045163869857788086,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 34908
+    },
+    {
+      "epoch": 3.587237977805179,
+      "grad_norm": 0.03191215172410011,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 34911
+    },
+    {
+      "epoch": 3.587546239210851,
+      "grad_norm": 0.033771395683288574,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 34914
+    },
+    {
+      "epoch": 3.5878545006165226,
+      "grad_norm": 0.06919904053211212,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 34917
+    },
+    {
+      "epoch": 3.5881627620221948,
+      "grad_norm": 0.05089094117283821,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 34920
+    },
+    {
+      "epoch": 3.588471023427867,
+      "grad_norm": 0.10625668615102768,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 34923
+    },
+    {
+      "epoch": 3.5887792848335387,
+      "grad_norm": 0.05255312845110893,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 34926
+    },
+    {
+      "epoch": 3.589087546239211,
+      "grad_norm": 0.09395511448383331,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 34929
+    },
+    {
+      "epoch": 3.589395807644883,
+      "grad_norm": 0.04209842532873154,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 34932
+    },
+    {
+      "epoch": 3.589704069050555,
+      "grad_norm": 0.0340830534696579,
+      "learning_rate": 0.01,
+      "loss": 2.036,
+      "step": 34935
+    },
+    {
+      "epoch": 3.590012330456227,
+      "grad_norm": 0.041663311421871185,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 34938
+    },
+    {
+      "epoch": 3.590320591861899,
+      "grad_norm": 0.044666144996881485,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 34941
+    },
+    {
+      "epoch": 3.590628853267571,
+      "grad_norm": 0.05369654670357704,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 34944
+    },
+    {
+      "epoch": 3.5909371146732427,
+      "grad_norm": 0.054408960044384,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 34947
+    },
+    {
+      "epoch": 3.591245376078915,
+      "grad_norm": 0.09313920885324478,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 34950
+    },
+    {
+      "epoch": 3.591553637484587,
+      "grad_norm": 0.05207030475139618,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 34953
+    },
+    {
+      "epoch": 3.591861898890259,
+      "grad_norm": 0.06822825968265533,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 34956
+    },
+    {
+      "epoch": 3.592170160295931,
+      "grad_norm": 0.05392748489975929,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 34959
+    },
+    {
+      "epoch": 3.592478421701603,
+      "grad_norm": 0.06300345808267593,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 34962
+    },
+    {
+      "epoch": 3.592786683107275,
+      "grad_norm": 0.09526319801807404,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 34965
+    },
+    {
+      "epoch": 3.5930949445129468,
+      "grad_norm": 0.060867104679346085,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 34968
+    },
+    {
+      "epoch": 3.593403205918619,
+      "grad_norm": 0.09763433039188385,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 34971
+    },
+    {
+      "epoch": 3.593711467324291,
+      "grad_norm": 0.05499181151390076,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 34974
+    },
+    {
+      "epoch": 3.594019728729963,
+      "grad_norm": 0.18828648328781128,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 34977
+    },
+    {
+      "epoch": 3.594327990135635,
+      "grad_norm": 0.1367756426334381,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 34980
+    },
+    {
+      "epoch": 3.5946362515413073,
+      "grad_norm": 0.0752996876835823,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 34983
+    },
+    {
+      "epoch": 3.594944512946979,
+      "grad_norm": 0.07500961422920227,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 34986
+    },
+    {
+      "epoch": 3.595252774352651,
+      "grad_norm": 0.08114928752183914,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 34989
+    },
+    {
+      "epoch": 3.595561035758323,
+      "grad_norm": 0.0735306516289711,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 34992
+    },
+    {
+      "epoch": 3.595869297163995,
+      "grad_norm": 0.04386765882372856,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 34995
+    },
+    {
+      "epoch": 3.596177558569667,
+      "grad_norm": 0.052437882870435715,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 34998
+    },
+    {
+      "epoch": 3.596485819975339,
+      "grad_norm": 0.044234346598386765,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 35001
+    },
+    {
+      "epoch": 3.5967940813810113,
+      "grad_norm": 0.04302847385406494,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 35004
+    },
+    {
+      "epoch": 3.597102342786683,
+      "grad_norm": 0.047404494136571884,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 35007
+    },
+    {
+      "epoch": 3.5974106041923553,
+      "grad_norm": 0.04339216649532318,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 35010
+    },
+    {
+      "epoch": 3.597718865598027,
+      "grad_norm": 0.07121206820011139,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 35013
+    },
+    {
+      "epoch": 3.598027127003699,
+      "grad_norm": 0.06314744055271149,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 35016
+    },
+    {
+      "epoch": 3.598335388409371,
+      "grad_norm": 0.0557650588452816,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 35019
+    },
+    {
+      "epoch": 3.598643649815043,
+      "grad_norm": 0.043066419661045074,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 35022
+    },
+    {
+      "epoch": 3.5989519112207153,
+      "grad_norm": 0.0739695131778717,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 35025
+    },
+    {
+      "epoch": 3.599260172626387,
+      "grad_norm": 0.13933037221431732,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 35028
+    },
+    {
+      "epoch": 3.5995684340320593,
+      "grad_norm": 0.03341522440314293,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 35031
+    },
+    {
+      "epoch": 3.599876695437731,
+      "grad_norm": 0.07127437740564346,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 35034
+    },
+    {
+      "epoch": 3.6001849568434032,
+      "grad_norm": 0.05078571289777756,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 35037
+    },
+    {
+      "epoch": 3.600493218249075,
+      "grad_norm": 0.049321308732032776,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 35040
+    },
+    {
+      "epoch": 3.600801479654747,
+      "grad_norm": 0.04840512201189995,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 35043
+    },
+    {
+      "epoch": 3.6011097410604194,
+      "grad_norm": 0.05541226640343666,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 35046
+    },
+    {
+      "epoch": 3.601418002466091,
+      "grad_norm": 0.03456197306513786,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 35049
+    },
+    {
+      "epoch": 3.6017262638717633,
+      "grad_norm": 0.0342303030192852,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 35052
+    },
+    {
+      "epoch": 3.6020345252774355,
+      "grad_norm": 0.043709222227334976,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 35055
+    },
+    {
+      "epoch": 3.6023427866831073,
+      "grad_norm": 0.05025354400277138,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 35058
+    },
+    {
+      "epoch": 3.6026510480887795,
+      "grad_norm": 0.0903453454375267,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 35061
+    },
+    {
+      "epoch": 3.602959309494451,
+      "grad_norm": 0.029984181746840477,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 35064
+    },
+    {
+      "epoch": 3.6032675709001234,
+      "grad_norm": 0.03956552594900131,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 35067
+    },
+    {
+      "epoch": 3.603575832305795,
+      "grad_norm": 0.0535871721804142,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 35070
+    },
+    {
+      "epoch": 3.6038840937114673,
+      "grad_norm": 0.07621357589960098,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 35073
+    },
+    {
+      "epoch": 3.6041923551171395,
+      "grad_norm": 0.08122174441814423,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 35076
+    },
+    {
+      "epoch": 3.6045006165228113,
+      "grad_norm": 0.04478932544589043,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 35079
+    },
+    {
+      "epoch": 3.6048088779284835,
+      "grad_norm": 0.07432077080011368,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 35082
+    },
+    {
+      "epoch": 3.6051171393341552,
+      "grad_norm": 0.08805304020643234,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 35085
+    },
+    {
+      "epoch": 3.6054254007398274,
+      "grad_norm": 0.15478205680847168,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 35088
+    },
+    {
+      "epoch": 3.605733662145499,
+      "grad_norm": 0.11871577054262161,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 35091
+    },
+    {
+      "epoch": 3.6060419235511714,
+      "grad_norm": 0.06366606056690216,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 35094
+    },
+    {
+      "epoch": 3.6063501849568436,
+      "grad_norm": 0.10421419143676758,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 35097
+    },
+    {
+      "epoch": 3.6066584463625153,
+      "grad_norm": 0.08190574496984482,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 35100
+    },
+    {
+      "epoch": 3.6069667077681875,
+      "grad_norm": 0.03852493688464165,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 35103
+    },
+    {
+      "epoch": 3.6072749691738597,
+      "grad_norm": 0.04792521148920059,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 35106
+    },
+    {
+      "epoch": 3.6075832305795315,
+      "grad_norm": 0.04788130521774292,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 35109
+    },
+    {
+      "epoch": 3.607891491985203,
+      "grad_norm": 0.06820268929004669,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 35112
+    },
+    {
+      "epoch": 3.6081997533908754,
+      "grad_norm": 0.05702923610806465,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 35115
+    },
+    {
+      "epoch": 3.6085080147965476,
+      "grad_norm": 0.10051339119672775,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 35118
+    },
+    {
+      "epoch": 3.6088162762022193,
+      "grad_norm": 0.07067910581827164,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 35121
+    },
+    {
+      "epoch": 3.6091245376078915,
+      "grad_norm": 0.03547900170087814,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 35124
+    },
+    {
+      "epoch": 3.6094327990135637,
+      "grad_norm": 0.062315478920936584,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 35127
+    },
+    {
+      "epoch": 3.6097410604192355,
+      "grad_norm": 0.12956464290618896,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 35130
+    },
+    {
+      "epoch": 3.6100493218249077,
+      "grad_norm": 0.1570357233285904,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 35133
+    },
+    {
+      "epoch": 3.6103575832305794,
+      "grad_norm": 0.06165264546871185,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 35136
+    },
+    {
+      "epoch": 3.6106658446362516,
+      "grad_norm": 0.046166177839040756,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 35139
+    },
+    {
+      "epoch": 3.6109741060419234,
+      "grad_norm": 0.052293550223112106,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 35142
+    },
+    {
+      "epoch": 3.6112823674475956,
+      "grad_norm": 0.046433527022600174,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 35145
+    },
+    {
+      "epoch": 3.6115906288532678,
+      "grad_norm": 0.029988888651132584,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 35148
+    },
+    {
+      "epoch": 3.6118988902589395,
+      "grad_norm": 0.05726629123091698,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 35151
+    },
+    {
+      "epoch": 3.6122071516646117,
+      "grad_norm": 0.054002657532691956,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 35154
+    },
+    {
+      "epoch": 3.6125154130702835,
+      "grad_norm": 0.05368361249566078,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 35157
+    },
+    {
+      "epoch": 3.6128236744759556,
+      "grad_norm": 0.04732634872198105,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 35160
+    },
+    {
+      "epoch": 3.6131319358816274,
+      "grad_norm": 0.05976017564535141,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 35163
+    },
+    {
+      "epoch": 3.6134401972872996,
+      "grad_norm": 0.04517058655619621,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 35166
+    },
+    {
+      "epoch": 3.613748458692972,
+      "grad_norm": 0.12990356981754303,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 35169
+    },
+    {
+      "epoch": 3.6140567200986435,
+      "grad_norm": 0.059477876871824265,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 35172
+    },
+    {
+      "epoch": 3.6143649815043157,
+      "grad_norm": 0.05780341103672981,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 35175
+    },
+    {
+      "epoch": 3.614673242909988,
+      "grad_norm": 0.03971264883875847,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 35178
+    },
+    {
+      "epoch": 3.6149815043156597,
+      "grad_norm": 0.04069376364350319,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 35181
+    },
+    {
+      "epoch": 3.6152897657213314,
+      "grad_norm": 0.049401164054870605,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 35184
+    },
+    {
+      "epoch": 3.6155980271270036,
+      "grad_norm": 0.058909036219120026,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 35187
+    },
+    {
+      "epoch": 3.615906288532676,
+      "grad_norm": 0.1364602893590927,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 35190
+    },
+    {
+      "epoch": 3.6162145499383476,
+      "grad_norm": 0.0892588272690773,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 35193
+    },
+    {
+      "epoch": 3.6165228113440198,
+      "grad_norm": 0.03974326699972153,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 35196
+    },
+    {
+      "epoch": 3.616831072749692,
+      "grad_norm": 0.03898739442229271,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 35199
+    },
+    {
+      "epoch": 3.6171393341553637,
+      "grad_norm": 0.044964905828237534,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 35202
+    },
+    {
+      "epoch": 3.617447595561036,
+      "grad_norm": 0.11181700229644775,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 35205
+    },
+    {
+      "epoch": 3.6177558569667077,
+      "grad_norm": 0.0697384923696518,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 35208
+    },
+    {
+      "epoch": 3.61806411837238,
+      "grad_norm": 0.03923085704445839,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 35211
+    },
+    {
+      "epoch": 3.6183723797780516,
+      "grad_norm": 0.03635834902524948,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 35214
+    },
+    {
+      "epoch": 3.618680641183724,
+      "grad_norm": 0.05013216286897659,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 35217
+    },
+    {
+      "epoch": 3.618988902589396,
+      "grad_norm": 0.03531458228826523,
+      "learning_rate": 0.01,
+      "loss": 2.0256,
+      "step": 35220
+    },
+    {
+      "epoch": 3.6192971639950677,
+      "grad_norm": 0.09228625893592834,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 35223
+    },
+    {
+      "epoch": 3.61960542540074,
+      "grad_norm": 0.06587129831314087,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 35226
+    },
+    {
+      "epoch": 3.6199136868064117,
+      "grad_norm": 0.11610520631074905,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 35229
+    },
+    {
+      "epoch": 3.620221948212084,
+      "grad_norm": 0.05594666674733162,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 35232
+    },
+    {
+      "epoch": 3.6205302096177556,
+      "grad_norm": 0.04040682688355446,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 35235
+    },
+    {
+      "epoch": 3.620838471023428,
+      "grad_norm": 0.038559917360544205,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 35238
+    },
+    {
+      "epoch": 3.6211467324291,
+      "grad_norm": 0.05012970045208931,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 35241
+    },
+    {
+      "epoch": 3.6214549938347718,
+      "grad_norm": 0.09769418835639954,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 35244
+    },
+    {
+      "epoch": 3.621763255240444,
+      "grad_norm": 0.07791107892990112,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 35247
+    },
+    {
+      "epoch": 3.622071516646116,
+      "grad_norm": 0.06556175649166107,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 35250
+    },
+    {
+      "epoch": 3.622379778051788,
+      "grad_norm": 0.04608479142189026,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 35253
+    },
+    {
+      "epoch": 3.6226880394574597,
+      "grad_norm": 0.03681867569684982,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 35256
+    },
+    {
+      "epoch": 3.622996300863132,
+      "grad_norm": 0.05940413847565651,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 35259
+    },
+    {
+      "epoch": 3.623304562268804,
+      "grad_norm": 0.09911686182022095,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 35262
+    },
+    {
+      "epoch": 3.623612823674476,
+      "grad_norm": 0.03739270940423012,
+      "learning_rate": 0.01,
+      "loss": 2.0285,
+      "step": 35265
+    },
+    {
+      "epoch": 3.623921085080148,
+      "grad_norm": 0.11673395335674286,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 35268
+    },
+    {
+      "epoch": 3.62422934648582,
+      "grad_norm": 0.0795954018831253,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 35271
+    },
+    {
+      "epoch": 3.624537607891492,
+      "grad_norm": 0.04180069640278816,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 35274
+    },
+    {
+      "epoch": 3.624845869297164,
+      "grad_norm": 0.09516190737485886,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 35277
+    },
+    {
+      "epoch": 3.625154130702836,
+      "grad_norm": 0.11448989808559418,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 35280
+    },
+    {
+      "epoch": 3.625462392108508,
+      "grad_norm": 0.06945902854204178,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 35283
+    },
+    {
+      "epoch": 3.62577065351418,
+      "grad_norm": 0.039453648030757904,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 35286
+    },
+    {
+      "epoch": 3.626078914919852,
+      "grad_norm": 0.06266641616821289,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 35289
+    },
+    {
+      "epoch": 3.626387176325524,
+      "grad_norm": 0.04908052831888199,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 35292
+    },
+    {
+      "epoch": 3.626695437731196,
+      "grad_norm": 0.06186684966087341,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 35295
+    },
+    {
+      "epoch": 3.627003699136868,
+      "grad_norm": 0.13145780563354492,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 35298
+    },
+    {
+      "epoch": 3.6273119605425403,
+      "grad_norm": 0.05850294232368469,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 35301
+    },
+    {
+      "epoch": 3.627620221948212,
+      "grad_norm": 0.04934421926736832,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 35304
+    },
+    {
+      "epoch": 3.627928483353884,
+      "grad_norm": 0.07607220858335495,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 35307
+    },
+    {
+      "epoch": 3.628236744759556,
+      "grad_norm": 0.046615466475486755,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 35310
+    },
+    {
+      "epoch": 3.6285450061652282,
+      "grad_norm": 0.11176659911870956,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 35313
+    },
+    {
+      "epoch": 3.6288532675709,
+      "grad_norm": 0.07844175398349762,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 35316
+    },
+    {
+      "epoch": 3.629161528976572,
+      "grad_norm": 0.0730755403637886,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 35319
+    },
+    {
+      "epoch": 3.6294697903822444,
+      "grad_norm": 0.08503863960504532,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 35322
+    },
+    {
+      "epoch": 3.629778051787916,
+      "grad_norm": 0.03936958312988281,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 35325
+    },
+    {
+      "epoch": 3.6300863131935883,
+      "grad_norm": 0.11586350947618484,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 35328
+    },
+    {
+      "epoch": 3.63039457459926,
+      "grad_norm": 0.06128397583961487,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 35331
+    },
+    {
+      "epoch": 3.6307028360049323,
+      "grad_norm": 0.04872961342334747,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 35334
+    },
+    {
+      "epoch": 3.631011097410604,
+      "grad_norm": 0.08416412770748138,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 35337
+    },
+    {
+      "epoch": 3.631319358816276,
+      "grad_norm": 0.05844739452004433,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 35340
+    },
+    {
+      "epoch": 3.6316276202219484,
+      "grad_norm": 0.08564214408397675,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 35343
+    },
+    {
+      "epoch": 3.63193588162762,
+      "grad_norm": 0.08769746124744415,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 35346
+    },
+    {
+      "epoch": 3.6322441430332923,
+      "grad_norm": 0.03819827735424042,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 35349
+    },
+    {
+      "epoch": 3.632552404438964,
+      "grad_norm": 0.12311212718486786,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 35352
+    },
+    {
+      "epoch": 3.6328606658446363,
+      "grad_norm": 0.037630997598171234,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 35355
+    },
+    {
+      "epoch": 3.633168927250308,
+      "grad_norm": 0.08739642798900604,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 35358
+    },
+    {
+      "epoch": 3.6334771886559802,
+      "grad_norm": 0.04889992997050285,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 35361
+    },
+    {
+      "epoch": 3.6337854500616524,
+      "grad_norm": 0.035435836762189865,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 35364
+    },
+    {
+      "epoch": 3.634093711467324,
+      "grad_norm": 0.059248197823762894,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 35367
+    },
+    {
+      "epoch": 3.6344019728729964,
+      "grad_norm": 0.08831764757633209,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 35370
+    },
+    {
+      "epoch": 3.6347102342786686,
+      "grad_norm": 0.14940893650054932,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 35373
+    },
+    {
+      "epoch": 3.6350184956843403,
+      "grad_norm": 0.06616901606321335,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 35376
+    },
+    {
+      "epoch": 3.635326757090012,
+      "grad_norm": 0.06391241401433945,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 35379
+    },
+    {
+      "epoch": 3.6356350184956843,
+      "grad_norm": 0.0580880343914032,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 35382
+    },
+    {
+      "epoch": 3.6359432799013565,
+      "grad_norm": 0.057876862585544586,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 35385
+    },
+    {
+      "epoch": 3.636251541307028,
+      "grad_norm": 0.044347915798425674,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 35388
+    },
+    {
+      "epoch": 3.6365598027127004,
+      "grad_norm": 0.0400017574429512,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 35391
+    },
+    {
+      "epoch": 3.6368680641183726,
+      "grad_norm": 0.11369085311889648,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 35394
+    },
+    {
+      "epoch": 3.6371763255240444,
+      "grad_norm": 0.10995481163263321,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 35397
+    },
+    {
+      "epoch": 3.6374845869297165,
+      "grad_norm": 0.05769447609782219,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 35400
+    },
+    {
+      "epoch": 3.6377928483353883,
+      "grad_norm": 0.07872482389211655,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 35403
+    },
+    {
+      "epoch": 3.6381011097410605,
+      "grad_norm": 0.05196002870798111,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 35406
+    },
+    {
+      "epoch": 3.6384093711467322,
+      "grad_norm": 0.0493265725672245,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 35409
+    },
+    {
+      "epoch": 3.6387176325524044,
+      "grad_norm": 0.04237900674343109,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 35412
+    },
+    {
+      "epoch": 3.6390258939580766,
+      "grad_norm": 0.06882897764444351,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 35415
+    },
+    {
+      "epoch": 3.6393341553637484,
+      "grad_norm": 0.058774422854185104,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 35418
+    },
+    {
+      "epoch": 3.6396424167694206,
+      "grad_norm": 0.10032794624567032,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 35421
+    },
+    {
+      "epoch": 3.6399506781750923,
+      "grad_norm": 0.042988426983356476,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 35424
+    },
+    {
+      "epoch": 3.6402589395807645,
+      "grad_norm": 0.055391326546669006,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 35427
+    },
+    {
+      "epoch": 3.6405672009864363,
+      "grad_norm": 0.03977194428443909,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 35430
+    },
+    {
+      "epoch": 3.6408754623921085,
+      "grad_norm": 0.0903034657239914,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 35433
+    },
+    {
+      "epoch": 3.6411837237977807,
+      "grad_norm": 0.04610143229365349,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 35436
+    },
+    {
+      "epoch": 3.6414919852034524,
+      "grad_norm": 0.08295582234859467,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 35439
+    },
+    {
+      "epoch": 3.6418002466091246,
+      "grad_norm": 0.09561655670404434,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 35442
+    },
+    {
+      "epoch": 3.642108508014797,
+      "grad_norm": 0.07968761771917343,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 35445
+    },
+    {
+      "epoch": 3.6424167694204685,
+      "grad_norm": 0.04458128660917282,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 35448
+    },
+    {
+      "epoch": 3.6427250308261403,
+      "grad_norm": 0.12892895936965942,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 35451
+    },
+    {
+      "epoch": 3.6430332922318125,
+      "grad_norm": 0.1063292846083641,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 35454
+    },
+    {
+      "epoch": 3.6433415536374847,
+      "grad_norm": 0.04187742993235588,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 35457
+    },
+    {
+      "epoch": 3.6436498150431564,
+      "grad_norm": 0.06096494942903519,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 35460
+    },
+    {
+      "epoch": 3.6439580764488286,
+      "grad_norm": 0.035430386662483215,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 35463
+    },
+    {
+      "epoch": 3.644266337854501,
+      "grad_norm": 0.039337921887636185,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 35466
+    },
+    {
+      "epoch": 3.6445745992601726,
+      "grad_norm": 0.04446728155016899,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 35469
+    },
+    {
+      "epoch": 3.6448828606658448,
+      "grad_norm": 0.06175538897514343,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 35472
+    },
+    {
+      "epoch": 3.6451911220715165,
+      "grad_norm": 0.1305261105298996,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 35475
+    },
+    {
+      "epoch": 3.6454993834771887,
+      "grad_norm": 0.14111317694187164,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 35478
+    },
+    {
+      "epoch": 3.6458076448828605,
+      "grad_norm": 0.13947373628616333,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 35481
+    },
+    {
+      "epoch": 3.6461159062885327,
+      "grad_norm": 0.057133182883262634,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 35484
+    },
+    {
+      "epoch": 3.646424167694205,
+      "grad_norm": 0.03709038347005844,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 35487
+    },
+    {
+      "epoch": 3.6467324290998766,
+      "grad_norm": 0.04786526411771774,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 35490
+    },
+    {
+      "epoch": 3.647040690505549,
+      "grad_norm": 0.03816988691687584,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 35493
+    },
+    {
+      "epoch": 3.6473489519112205,
+      "grad_norm": 0.05153171718120575,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 35496
+    },
+    {
+      "epoch": 3.6476572133168927,
+      "grad_norm": 0.09072964638471603,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 35499
+    },
+    {
+      "epoch": 3.6479654747225645,
+      "grad_norm": 0.06302040815353394,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 35502
+    },
+    {
+      "epoch": 3.6482737361282367,
+      "grad_norm": 0.04940348491072655,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 35505
+    },
+    {
+      "epoch": 3.648581997533909,
+      "grad_norm": 0.04952861741185188,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 35508
+    },
+    {
+      "epoch": 3.6488902589395806,
+      "grad_norm": 0.03227638080716133,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 35511
+    },
+    {
+      "epoch": 3.649198520345253,
+      "grad_norm": 0.04237101599574089,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 35514
+    },
+    {
+      "epoch": 3.649506781750925,
+      "grad_norm": 0.10213712602853775,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 35517
+    },
+    {
+      "epoch": 3.6498150431565968,
+      "grad_norm": 0.0747971460223198,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 35520
+    },
+    {
+      "epoch": 3.650123304562269,
+      "grad_norm": 0.07060317695140839,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 35523
+    },
+    {
+      "epoch": 3.6504315659679407,
+      "grad_norm": 0.13745035231113434,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 35526
+    },
+    {
+      "epoch": 3.650739827373613,
+      "grad_norm": 0.11544306576251984,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 35529
+    },
+    {
+      "epoch": 3.6510480887792847,
+      "grad_norm": 0.07669739425182343,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 35532
+    },
+    {
+      "epoch": 3.651356350184957,
+      "grad_norm": 0.04519101604819298,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 35535
+    },
+    {
+      "epoch": 3.651664611590629,
+      "grad_norm": 0.04446371644735336,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 35538
+    },
+    {
+      "epoch": 3.651972872996301,
+      "grad_norm": 0.05518916994333267,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 35541
+    },
+    {
+      "epoch": 3.652281134401973,
+      "grad_norm": 0.04405786469578743,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 35544
+    },
+    {
+      "epoch": 3.6525893958076447,
+      "grad_norm": 0.06154394894838333,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 35547
+    },
+    {
+      "epoch": 3.652897657213317,
+      "grad_norm": 0.07919877767562866,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 35550
+    },
+    {
+      "epoch": 3.6532059186189887,
+      "grad_norm": 0.0575590617954731,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 35553
+    },
+    {
+      "epoch": 3.653514180024661,
+      "grad_norm": 0.05680471658706665,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 35556
+    },
+    {
+      "epoch": 3.653822441430333,
+      "grad_norm": 0.0352899394929409,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 35559
+    },
+    {
+      "epoch": 3.654130702836005,
+      "grad_norm": 0.034432023763656616,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 35562
+    },
+    {
+      "epoch": 3.654438964241677,
+      "grad_norm": 0.10298652946949005,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 35565
+    },
+    {
+      "epoch": 3.654747225647349,
+      "grad_norm": 0.03996056690812111,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 35568
+    },
+    {
+      "epoch": 3.655055487053021,
+      "grad_norm": 0.07411230355501175,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 35571
+    },
+    {
+      "epoch": 3.6553637484586927,
+      "grad_norm": 0.04124278202652931,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 35574
+    },
+    {
+      "epoch": 3.655672009864365,
+      "grad_norm": 0.035065338015556335,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 35577
+    },
+    {
+      "epoch": 3.655980271270037,
+      "grad_norm": 0.04023493826389313,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 35580
+    },
+    {
+      "epoch": 3.656288532675709,
+      "grad_norm": 0.0333552286028862,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 35583
+    },
+    {
+      "epoch": 3.656596794081381,
+      "grad_norm": 0.1286098062992096,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 35586
+    },
+    {
+      "epoch": 3.6569050554870532,
+      "grad_norm": 0.061940498650074005,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 35589
+    },
+    {
+      "epoch": 3.657213316892725,
+      "grad_norm": 0.08766448497772217,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 35592
+    },
+    {
+      "epoch": 3.657521578298397,
+      "grad_norm": 0.07218505442142487,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 35595
+    },
+    {
+      "epoch": 3.657829839704069,
+      "grad_norm": 0.11700677126646042,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 35598
+    },
+    {
+      "epoch": 3.658138101109741,
+      "grad_norm": 0.05746941268444061,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 35601
+    },
+    {
+      "epoch": 3.658446362515413,
+      "grad_norm": 0.0517272874712944,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 35604
+    },
+    {
+      "epoch": 3.658754623921085,
+      "grad_norm": 0.04524237662553787,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 35607
+    },
+    {
+      "epoch": 3.6590628853267573,
+      "grad_norm": 0.053268514573574066,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 35610
+    },
+    {
+      "epoch": 3.659371146732429,
+      "grad_norm": 0.06391450762748718,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 35613
+    },
+    {
+      "epoch": 3.659679408138101,
+      "grad_norm": 0.06404808908700943,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 35616
+    },
+    {
+      "epoch": 3.659987669543773,
+      "grad_norm": 0.06734498590230942,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 35619
+    },
+    {
+      "epoch": 3.660295930949445,
+      "grad_norm": 0.056514717638492584,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 35622
+    },
+    {
+      "epoch": 3.660604192355117,
+      "grad_norm": 0.04377966374158859,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 35625
+    },
+    {
+      "epoch": 3.660912453760789,
+      "grad_norm": 0.0527014285326004,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 35628
+    },
+    {
+      "epoch": 3.6612207151664613,
+      "grad_norm": 0.08603314310312271,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 35631
+    },
+    {
+      "epoch": 3.661528976572133,
+      "grad_norm": 0.06637904793024063,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 35634
+    },
+    {
+      "epoch": 3.6618372379778052,
+      "grad_norm": 0.042211759835481644,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 35637
+    },
+    {
+      "epoch": 3.6621454993834774,
+      "grad_norm": 0.08589443564414978,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 35640
+    },
+    {
+      "epoch": 3.662453760789149,
+      "grad_norm": 0.12761181592941284,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 35643
+    },
+    {
+      "epoch": 3.662762022194821,
+      "grad_norm": 0.08702104538679123,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 35646
+    },
+    {
+      "epoch": 3.663070283600493,
+      "grad_norm": 0.1481538712978363,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 35649
+    },
+    {
+      "epoch": 3.6633785450061653,
+      "grad_norm": 0.05475825071334839,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 35652
+    },
+    {
+      "epoch": 3.663686806411837,
+      "grad_norm": 0.0481879860162735,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 35655
+    },
+    {
+      "epoch": 3.6639950678175093,
+      "grad_norm": 0.07099757343530655,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 35658
+    },
+    {
+      "epoch": 3.6643033292231815,
+      "grad_norm": 0.09161633253097534,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 35661
+    },
+    {
+      "epoch": 3.664611590628853,
+      "grad_norm": 0.06675172597169876,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 35664
+    },
+    {
+      "epoch": 3.6649198520345254,
+      "grad_norm": 0.06166239082813263,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 35667
+    },
+    {
+      "epoch": 3.665228113440197,
+      "grad_norm": 0.07947093993425369,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 35670
+    },
+    {
+      "epoch": 3.6655363748458694,
+      "grad_norm": 0.07639001309871674,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 35673
+    },
+    {
+      "epoch": 3.665844636251541,
+      "grad_norm": 0.11109112948179245,
+      "learning_rate": 0.01,
+      "loss": 2.033,
+      "step": 35676
+    },
+    {
+      "epoch": 3.6661528976572133,
+      "grad_norm": 0.0517781600356102,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 35679
+    },
+    {
+      "epoch": 3.6664611590628855,
+      "grad_norm": 0.05615885183215141,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 35682
+    },
+    {
+      "epoch": 3.6667694204685573,
+      "grad_norm": 0.054977256804704666,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 35685
+    },
+    {
+      "epoch": 3.6670776818742294,
+      "grad_norm": 0.08208134025335312,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 35688
+    },
+    {
+      "epoch": 3.667385943279901,
+      "grad_norm": 0.053141169250011444,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 35691
+    },
+    {
+      "epoch": 3.6676942046855734,
+      "grad_norm": 0.0867038443684578,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 35694
+    },
+    {
+      "epoch": 3.668002466091245,
+      "grad_norm": 0.04153072461485863,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 35697
+    },
+    {
+      "epoch": 3.6683107274969173,
+      "grad_norm": 0.050722386687994,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 35700
+    },
+    {
+      "epoch": 3.6686189889025895,
+      "grad_norm": 0.036444198340177536,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 35703
+    },
+    {
+      "epoch": 3.6689272503082613,
+      "grad_norm": 0.06851452589035034,
+      "learning_rate": 0.01,
+      "loss": 2.0365,
+      "step": 35706
+    },
+    {
+      "epoch": 3.6692355117139335,
+      "grad_norm": 0.05442693457007408,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 35709
+    },
+    {
+      "epoch": 3.6695437731196057,
+      "grad_norm": 0.04789496958255768,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 35712
+    },
+    {
+      "epoch": 3.6698520345252774,
+      "grad_norm": 0.05229181796312332,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 35715
+    },
+    {
+      "epoch": 3.670160295930949,
+      "grad_norm": 0.05596926435828209,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 35718
+    },
+    {
+      "epoch": 3.6704685573366214,
+      "grad_norm": 0.04481478035449982,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 35721
+    },
+    {
+      "epoch": 3.6707768187422936,
+      "grad_norm": 0.053352151066064835,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 35724
+    },
+    {
+      "epoch": 3.6710850801479653,
+      "grad_norm": 0.07115049660205841,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 35727
+    },
+    {
+      "epoch": 3.6713933415536375,
+      "grad_norm": 0.06887483596801758,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 35730
+    },
+    {
+      "epoch": 3.6717016029593097,
+      "grad_norm": 0.14638669788837433,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 35733
+    },
+    {
+      "epoch": 3.6720098643649814,
+      "grad_norm": 0.06305021792650223,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 35736
+    },
+    {
+      "epoch": 3.6723181257706536,
+      "grad_norm": 0.041157372295856476,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 35739
+    },
+    {
+      "epoch": 3.6726263871763254,
+      "grad_norm": 0.10821440070867538,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 35742
+    },
+    {
+      "epoch": 3.6729346485819976,
+      "grad_norm": 0.06525052338838577,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 35745
+    },
+    {
+      "epoch": 3.6732429099876693,
+      "grad_norm": 0.03519715368747711,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 35748
+    },
+    {
+      "epoch": 3.6735511713933415,
+      "grad_norm": 0.0548231340944767,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 35751
+    },
+    {
+      "epoch": 3.6738594327990137,
+      "grad_norm": 0.08685827255249023,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 35754
+    },
+    {
+      "epoch": 3.6741676942046855,
+      "grad_norm": 0.05920972675085068,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 35757
+    },
+    {
+      "epoch": 3.6744759556103577,
+      "grad_norm": 0.060545727610588074,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 35760
+    },
+    {
+      "epoch": 3.67478421701603,
+      "grad_norm": 0.05436963587999344,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 35763
+    },
+    {
+      "epoch": 3.6750924784217016,
+      "grad_norm": 0.05975078046321869,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 35766
+    },
+    {
+      "epoch": 3.6754007398273734,
+      "grad_norm": 0.05793645977973938,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 35769
+    },
+    {
+      "epoch": 3.6757090012330456,
+      "grad_norm": 0.07932164520025253,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 35772
+    },
+    {
+      "epoch": 3.6760172626387178,
+      "grad_norm": 0.05182117223739624,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 35775
+    },
+    {
+      "epoch": 3.6763255240443895,
+      "grad_norm": 0.05408613011240959,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 35778
+    },
+    {
+      "epoch": 3.6766337854500617,
+      "grad_norm": 0.0396132618188858,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 35781
+    },
+    {
+      "epoch": 3.676942046855734,
+      "grad_norm": 0.04768325388431549,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 35784
+    },
+    {
+      "epoch": 3.6772503082614056,
+      "grad_norm": 0.09452294558286667,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 35787
+    },
+    {
+      "epoch": 3.677558569667078,
+      "grad_norm": 0.09846623986959457,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 35790
+    },
+    {
+      "epoch": 3.6778668310727496,
+      "grad_norm": 0.1176767498254776,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 35793
+    },
+    {
+      "epoch": 3.678175092478422,
+      "grad_norm": 0.07070811092853546,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 35796
+    },
+    {
+      "epoch": 3.6784833538840935,
+      "grad_norm": 0.04359079524874687,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 35799
+    },
+    {
+      "epoch": 3.6787916152897657,
+      "grad_norm": 0.05718059092760086,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 35802
+    },
+    {
+      "epoch": 3.679099876695438,
+      "grad_norm": 0.03475908935070038,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 35805
+    },
+    {
+      "epoch": 3.6794081381011097,
+      "grad_norm": 0.03629058972001076,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 35808
+    },
+    {
+      "epoch": 3.679716399506782,
+      "grad_norm": 0.11145780235528946,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 35811
+    },
+    {
+      "epoch": 3.6800246609124536,
+      "grad_norm": 0.07155486941337585,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 35814
+    },
+    {
+      "epoch": 3.680332922318126,
+      "grad_norm": 0.05105472728610039,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 35817
+    },
+    {
+      "epoch": 3.6806411837237976,
+      "grad_norm": 0.04478003829717636,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 35820
+    },
+    {
+      "epoch": 3.6809494451294698,
+      "grad_norm": 0.03695627674460411,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 35823
+    },
+    {
+      "epoch": 3.681257706535142,
+      "grad_norm": 0.04222894459962845,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 35826
+    },
+    {
+      "epoch": 3.6815659679408137,
+      "grad_norm": 0.04362770542502403,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 35829
+    },
+    {
+      "epoch": 3.681874229346486,
+      "grad_norm": 0.07785011827945709,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 35832
+    },
+    {
+      "epoch": 3.682182490752158,
+      "grad_norm": 0.08382470905780792,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 35835
+    },
+    {
+      "epoch": 3.68249075215783,
+      "grad_norm": 0.09897731244564056,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 35838
+    },
+    {
+      "epoch": 3.6827990135635016,
+      "grad_norm": 0.05617908388376236,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 35841
+    },
+    {
+      "epoch": 3.683107274969174,
+      "grad_norm": 0.07543769478797913,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 35844
+    },
+    {
+      "epoch": 3.683415536374846,
+      "grad_norm": 0.04036831855773926,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 35847
+    },
+    {
+      "epoch": 3.6837237977805177,
+      "grad_norm": 0.03254294767975807,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 35850
+    },
+    {
+      "epoch": 3.68403205918619,
+      "grad_norm": 0.08003180474042892,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 35853
+    },
+    {
+      "epoch": 3.684340320591862,
+      "grad_norm": 0.0821150541305542,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 35856
+    },
+    {
+      "epoch": 3.684648581997534,
+      "grad_norm": 0.15571752190589905,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 35859
+    },
+    {
+      "epoch": 3.684956843403206,
+      "grad_norm": 0.07748742401599884,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 35862
+    },
+    {
+      "epoch": 3.685265104808878,
+      "grad_norm": 0.08278176933526993,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 35865
+    },
+    {
+      "epoch": 3.68557336621455,
+      "grad_norm": 0.08014661818742752,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 35868
+    },
+    {
+      "epoch": 3.6858816276202218,
+      "grad_norm": 0.03511490300297737,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 35871
+    },
+    {
+      "epoch": 3.686189889025894,
+      "grad_norm": 0.04790462180972099,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 35874
+    },
+    {
+      "epoch": 3.686498150431566,
+      "grad_norm": 0.05917114019393921,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 35877
+    },
+    {
+      "epoch": 3.686806411837238,
+      "grad_norm": 0.09026920050382614,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 35880
+    },
+    {
+      "epoch": 3.68711467324291,
+      "grad_norm": 0.07264431565999985,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 35883
+    },
+    {
+      "epoch": 3.687422934648582,
+      "grad_norm": 0.03798295557498932,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 35886
+    },
+    {
+      "epoch": 3.687731196054254,
+      "grad_norm": 0.1147773265838623,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 35889
+    },
+    {
+      "epoch": 3.688039457459926,
+      "grad_norm": 0.05281820893287659,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 35892
+    },
+    {
+      "epoch": 3.688347718865598,
+      "grad_norm": 0.0617707259953022,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 35895
+    },
+    {
+      "epoch": 3.68865598027127,
+      "grad_norm": 0.05084836110472679,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 35898
+    },
+    {
+      "epoch": 3.688964241676942,
+      "grad_norm": 0.04952919855713844,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 35901
+    },
+    {
+      "epoch": 3.689272503082614,
+      "grad_norm": 0.08321710675954819,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 35904
+    },
+    {
+      "epoch": 3.6895807644882863,
+      "grad_norm": 0.04074413329362869,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 35907
+    },
+    {
+      "epoch": 3.689889025893958,
+      "grad_norm": 0.11404154449701309,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 35910
+    },
+    {
+      "epoch": 3.69019728729963,
+      "grad_norm": 0.07283198833465576,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 35913
+    },
+    {
+      "epoch": 3.690505548705302,
+      "grad_norm": 0.0645599216222763,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 35916
+    },
+    {
+      "epoch": 3.690813810110974,
+      "grad_norm": 0.0537913516163826,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 35919
+    },
+    {
+      "epoch": 3.691122071516646,
+      "grad_norm": 0.04984259232878685,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 35922
+    },
+    {
+      "epoch": 3.691430332922318,
+      "grad_norm": 0.05349741503596306,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 35925
+    },
+    {
+      "epoch": 3.6917385943279903,
+      "grad_norm": 0.05882362276315689,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 35928
+    },
+    {
+      "epoch": 3.692046855733662,
+      "grad_norm": 0.0809783861041069,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 35931
+    },
+    {
+      "epoch": 3.6923551171393343,
+      "grad_norm": 0.07075916230678558,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 35934
+    },
+    {
+      "epoch": 3.692663378545006,
+      "grad_norm": 0.058204181492328644,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 35937
+    },
+    {
+      "epoch": 3.6929716399506782,
+      "grad_norm": 0.10149111598730087,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 35940
+    },
+    {
+      "epoch": 3.69327990135635,
+      "grad_norm": 0.0722658559679985,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 35943
+    },
+    {
+      "epoch": 3.693588162762022,
+      "grad_norm": 0.08981306105852127,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 35946
+    },
+    {
+      "epoch": 3.6938964241676944,
+      "grad_norm": 0.04798738285899162,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 35949
+    },
+    {
+      "epoch": 3.694204685573366,
+      "grad_norm": 0.0338914729654789,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 35952
+    },
+    {
+      "epoch": 3.6945129469790383,
+      "grad_norm": 0.04813714697957039,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 35955
+    },
+    {
+      "epoch": 3.6948212083847105,
+      "grad_norm": 0.11086717247962952,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 35958
+    },
+    {
+      "epoch": 3.6951294697903823,
+      "grad_norm": 0.09754761308431625,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 35961
+    },
+    {
+      "epoch": 3.695437731196054,
+      "grad_norm": 0.07381515204906464,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 35964
+    },
+    {
+      "epoch": 3.695745992601726,
+      "grad_norm": 0.0501638762652874,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 35967
+    },
+    {
+      "epoch": 3.6960542540073984,
+      "grad_norm": 0.04152151197195053,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 35970
+    },
+    {
+      "epoch": 3.69636251541307,
+      "grad_norm": 0.06455028802156448,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 35973
+    },
+    {
+      "epoch": 3.6966707768187423,
+      "grad_norm": 0.07925435900688171,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 35976
+    },
+    {
+      "epoch": 3.6969790382244145,
+      "grad_norm": 0.17305727303028107,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 35979
+    },
+    {
+      "epoch": 3.6972872996300863,
+      "grad_norm": 0.04664710536599159,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 35982
+    },
+    {
+      "epoch": 3.6975955610357585,
+      "grad_norm": 0.03971162810921669,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 35985
+    },
+    {
+      "epoch": 3.6979038224414302,
+      "grad_norm": 0.04888729378581047,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 35988
+    },
+    {
+      "epoch": 3.6982120838471024,
+      "grad_norm": 0.03428930416703224,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 35991
+    },
+    {
+      "epoch": 3.698520345252774,
+      "grad_norm": 0.04609334468841553,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 35994
+    },
+    {
+      "epoch": 3.6988286066584464,
+      "grad_norm": 0.06719902157783508,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 35997
+    },
+    {
+      "epoch": 3.6991368680641186,
+      "grad_norm": 0.0409255288541317,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 36000
+    },
+    {
+      "epoch": 3.6994451294697903,
+      "grad_norm": 0.03683490306138992,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 36003
+    },
+    {
+      "epoch": 3.6997533908754625,
+      "grad_norm": 0.050057847052812576,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 36006
+    },
+    {
+      "epoch": 3.7000616522811343,
+      "grad_norm": 0.06051032990217209,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 36009
+    },
+    {
+      "epoch": 3.7003699136868065,
+      "grad_norm": 0.141363725066185,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 36012
+    },
+    {
+      "epoch": 3.700678175092478,
+      "grad_norm": 0.04017074033617973,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 36015
+    },
+    {
+      "epoch": 3.7009864364981504,
+      "grad_norm": 0.05374070256948471,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 36018
+    },
+    {
+      "epoch": 3.7012946979038226,
+      "grad_norm": 0.07037216424942017,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 36021
+    },
+    {
+      "epoch": 3.7016029593094943,
+      "grad_norm": 0.06400209665298462,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 36024
+    },
+    {
+      "epoch": 3.7019112207151665,
+      "grad_norm": 0.04186910763382912,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 36027
+    },
+    {
+      "epoch": 3.7022194821208387,
+      "grad_norm": 0.05100405216217041,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 36030
+    },
+    {
+      "epoch": 3.7025277435265105,
+      "grad_norm": 0.06786265224218369,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 36033
+    },
+    {
+      "epoch": 3.7028360049321822,
+      "grad_norm": 0.058080319315195084,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 36036
+    },
+    {
+      "epoch": 3.7031442663378544,
+      "grad_norm": 0.09919019788503647,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 36039
+    },
+    {
+      "epoch": 3.7034525277435266,
+      "grad_norm": 0.08252627402544022,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 36042
+    },
+    {
+      "epoch": 3.7037607891491984,
+      "grad_norm": 0.06026385724544525,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 36045
+    },
+    {
+      "epoch": 3.7040690505548706,
+      "grad_norm": 0.08052903413772583,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 36048
+    },
+    {
+      "epoch": 3.7043773119605428,
+      "grad_norm": 0.0790569856762886,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 36051
+    },
+    {
+      "epoch": 3.7046855733662145,
+      "grad_norm": 0.04464380070567131,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 36054
+    },
+    {
+      "epoch": 3.7049938347718867,
+      "grad_norm": 0.06567423790693283,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 36057
+    },
+    {
+      "epoch": 3.7053020961775585,
+      "grad_norm": 0.06602831929922104,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 36060
+    },
+    {
+      "epoch": 3.7056103575832307,
+      "grad_norm": 0.06838707625865936,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 36063
+    },
+    {
+      "epoch": 3.7059186189889024,
+      "grad_norm": 0.09807326644659042,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 36066
+    },
+    {
+      "epoch": 3.7062268803945746,
+      "grad_norm": 0.06948495656251907,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 36069
+    },
+    {
+      "epoch": 3.706535141800247,
+      "grad_norm": 0.07893040776252747,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 36072
+    },
+    {
+      "epoch": 3.7068434032059185,
+      "grad_norm": 0.04618433490395546,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 36075
+    },
+    {
+      "epoch": 3.7071516646115907,
+      "grad_norm": 0.04784570261836052,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 36078
+    },
+    {
+      "epoch": 3.7074599260172625,
+      "grad_norm": 0.03364891558885574,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 36081
+    },
+    {
+      "epoch": 3.7077681874229347,
+      "grad_norm": 0.03379710018634796,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 36084
+    },
+    {
+      "epoch": 3.7080764488286064,
+      "grad_norm": 0.03952077031135559,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 36087
+    },
+    {
+      "epoch": 3.7083847102342786,
+      "grad_norm": 0.07781558483839035,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 36090
+    },
+    {
+      "epoch": 3.708692971639951,
+      "grad_norm": 0.04583312198519707,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 36093
+    },
+    {
+      "epoch": 3.7090012330456226,
+      "grad_norm": 0.08673562854528427,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 36096
+    },
+    {
+      "epoch": 3.7093094944512948,
+      "grad_norm": 0.05132952705025673,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 36099
+    },
+    {
+      "epoch": 3.709617755856967,
+      "grad_norm": 0.09407106041908264,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 36102
+    },
+    {
+      "epoch": 3.7099260172626387,
+      "grad_norm": 0.052429962903261185,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 36105
+    },
+    {
+      "epoch": 3.7102342786683105,
+      "grad_norm": 0.10452061146497726,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 36108
+    },
+    {
+      "epoch": 3.7105425400739827,
+      "grad_norm": 0.03275587409734726,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 36111
+    },
+    {
+      "epoch": 3.710850801479655,
+      "grad_norm": 0.10667680203914642,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 36114
+    },
+    {
+      "epoch": 3.7111590628853266,
+      "grad_norm": 0.08240865170955658,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 36117
+    },
+    {
+      "epoch": 3.711467324290999,
+      "grad_norm": 0.0850924402475357,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 36120
+    },
+    {
+      "epoch": 3.711775585696671,
+      "grad_norm": 0.03906584531068802,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 36123
+    },
+    {
+      "epoch": 3.7120838471023427,
+      "grad_norm": 0.04797567054629326,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 36126
+    },
+    {
+      "epoch": 3.712392108508015,
+      "grad_norm": 0.042494967579841614,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 36129
+    },
+    {
+      "epoch": 3.7127003699136867,
+      "grad_norm": 0.05491645634174347,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 36132
+    },
+    {
+      "epoch": 3.713008631319359,
+      "grad_norm": 0.10729935020208359,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 36135
+    },
+    {
+      "epoch": 3.7133168927250306,
+      "grad_norm": 0.043170586228370667,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 36138
+    },
+    {
+      "epoch": 3.713625154130703,
+      "grad_norm": 0.08662062138319016,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 36141
+    },
+    {
+      "epoch": 3.713933415536375,
+      "grad_norm": 0.11379045993089676,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 36144
+    },
+    {
+      "epoch": 3.7142416769420468,
+      "grad_norm": 0.03698570281267166,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 36147
+    },
+    {
+      "epoch": 3.714549938347719,
+      "grad_norm": 0.11952532082796097,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 36150
+    },
+    {
+      "epoch": 3.7148581997533907,
+      "grad_norm": 0.06563457101583481,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 36153
+    },
+    {
+      "epoch": 3.715166461159063,
+      "grad_norm": 0.07150832563638687,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 36156
+    },
+    {
+      "epoch": 3.7154747225647347,
+      "grad_norm": 0.04657771810889244,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 36159
+    },
+    {
+      "epoch": 3.715782983970407,
+      "grad_norm": 0.055245641618967056,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 36162
+    },
+    {
+      "epoch": 3.716091245376079,
+      "grad_norm": 0.040487829595804214,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 36165
+    },
+    {
+      "epoch": 3.716399506781751,
+      "grad_norm": 0.12351846694946289,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 36168
+    },
+    {
+      "epoch": 3.716707768187423,
+      "grad_norm": 0.053484536707401276,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 36171
+    },
+    {
+      "epoch": 3.717016029593095,
+      "grad_norm": 0.06141964718699455,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 36174
+    },
+    {
+      "epoch": 3.717324290998767,
+      "grad_norm": 0.09509648382663727,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 36177
+    },
+    {
+      "epoch": 3.717632552404439,
+      "grad_norm": 0.0969184935092926,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 36180
+    },
+    {
+      "epoch": 3.717940813810111,
+      "grad_norm": 0.0884992703795433,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 36183
+    },
+    {
+      "epoch": 3.718249075215783,
+      "grad_norm": 0.07382161915302277,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 36186
+    },
+    {
+      "epoch": 3.718557336621455,
+      "grad_norm": 0.07776398956775665,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 36189
+    },
+    {
+      "epoch": 3.718865598027127,
+      "grad_norm": 0.07049771398305893,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 36192
+    },
+    {
+      "epoch": 3.719173859432799,
+      "grad_norm": 0.08669892698526382,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 36195
+    },
+    {
+      "epoch": 3.719482120838471,
+      "grad_norm": 0.05756537616252899,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 36198
+    },
+    {
+      "epoch": 3.719790382244143,
+      "grad_norm": 0.06954919546842575,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 36201
+    },
+    {
+      "epoch": 3.720098643649815,
+      "grad_norm": 0.09297992289066315,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 36204
+    },
+    {
+      "epoch": 3.720406905055487,
+      "grad_norm": 0.07941876351833344,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 36207
+    },
+    {
+      "epoch": 3.720715166461159,
+      "grad_norm": 0.0698881521821022,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 36210
+    },
+    {
+      "epoch": 3.721023427866831,
+      "grad_norm": 0.09939584881067276,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 36213
+    },
+    {
+      "epoch": 3.7213316892725032,
+      "grad_norm": 0.059318870306015015,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 36216
+    },
+    {
+      "epoch": 3.721639950678175,
+      "grad_norm": 0.06797734647989273,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 36219
+    },
+    {
+      "epoch": 3.721948212083847,
+      "grad_norm": 0.03932074084877968,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 36222
+    },
+    {
+      "epoch": 3.7222564734895194,
+      "grad_norm": 0.07458118349313736,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 36225
+    },
+    {
+      "epoch": 3.722564734895191,
+      "grad_norm": 0.09383490681648254,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 36228
+    },
+    {
+      "epoch": 3.722872996300863,
+      "grad_norm": 0.06527485698461533,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 36231
+    },
+    {
+      "epoch": 3.723181257706535,
+      "grad_norm": 0.050437018275260925,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 36234
+    },
+    {
+      "epoch": 3.7234895191122073,
+      "grad_norm": 0.04380796477198601,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 36237
+    },
+    {
+      "epoch": 3.723797780517879,
+      "grad_norm": 0.04857274517416954,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 36240
+    },
+    {
+      "epoch": 3.724106041923551,
+      "grad_norm": 0.032346438616514206,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 36243
+    },
+    {
+      "epoch": 3.7244143033292234,
+      "grad_norm": 0.045984454452991486,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 36246
+    },
+    {
+      "epoch": 3.724722564734895,
+      "grad_norm": 0.058199312537908554,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 36249
+    },
+    {
+      "epoch": 3.7250308261405674,
+      "grad_norm": 0.0391390286386013,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 36252
+    },
+    {
+      "epoch": 3.725339087546239,
+      "grad_norm": 0.058478716760873795,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 36255
+    },
+    {
+      "epoch": 3.7256473489519113,
+      "grad_norm": 0.10977184772491455,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 36258
+    },
+    {
+      "epoch": 3.725955610357583,
+      "grad_norm": 0.06433524191379547,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 36261
+    },
+    {
+      "epoch": 3.7262638717632552,
+      "grad_norm": 0.07871260493993759,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 36264
+    },
+    {
+      "epoch": 3.7265721331689274,
+      "grad_norm": 0.09282384812831879,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 36267
+    },
+    {
+      "epoch": 3.726880394574599,
+      "grad_norm": 0.03165189549326897,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 36270
+    },
+    {
+      "epoch": 3.7271886559802714,
+      "grad_norm": 0.0789475068449974,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 36273
+    },
+    {
+      "epoch": 3.727496917385943,
+      "grad_norm": 0.13641710579395294,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 36276
+    },
+    {
+      "epoch": 3.7278051787916153,
+      "grad_norm": 0.08470667898654938,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 36279
+    },
+    {
+      "epoch": 3.728113440197287,
+      "grad_norm": 0.09110618382692337,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 36282
+    },
+    {
+      "epoch": 3.7284217016029593,
+      "grad_norm": 0.079217828810215,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 36285
+    },
+    {
+      "epoch": 3.7287299630086315,
+      "grad_norm": 0.03365809842944145,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 36288
+    },
+    {
+      "epoch": 3.729038224414303,
+      "grad_norm": 0.0691695362329483,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 36291
+    },
+    {
+      "epoch": 3.7293464858199754,
+      "grad_norm": 0.03766563534736633,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 36294
+    },
+    {
+      "epoch": 3.7296547472256476,
+      "grad_norm": 0.13410314917564392,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 36297
+    },
+    {
+      "epoch": 3.7299630086313194,
+      "grad_norm": 0.08052795380353928,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 36300
+    },
+    {
+      "epoch": 3.730271270036991,
+      "grad_norm": 0.04205624386668205,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 36303
+    },
+    {
+      "epoch": 3.7305795314426633,
+      "grad_norm": 0.055557433515787125,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 36306
+    },
+    {
+      "epoch": 3.7308877928483355,
+      "grad_norm": 0.08415018022060394,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 36309
+    },
+    {
+      "epoch": 3.7311960542540072,
+      "grad_norm": 0.057384416460990906,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 36312
+    },
+    {
+      "epoch": 3.7315043156596794,
+      "grad_norm": 0.046833690255880356,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 36315
+    },
+    {
+      "epoch": 3.7318125770653516,
+      "grad_norm": 0.0527094230055809,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 36318
+    },
+    {
+      "epoch": 3.7321208384710234,
+      "grad_norm": 0.04863157868385315,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 36321
+    },
+    {
+      "epoch": 3.7324290998766956,
+      "grad_norm": 0.05578998848795891,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 36324
+    },
+    {
+      "epoch": 3.7327373612823673,
+      "grad_norm": 0.08895915746688843,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 36327
+    },
+    {
+      "epoch": 3.7330456226880395,
+      "grad_norm": 0.14340227842330933,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 36330
+    },
+    {
+      "epoch": 3.7333538840937113,
+      "grad_norm": 0.074753038585186,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 36333
+    },
+    {
+      "epoch": 3.7336621454993835,
+      "grad_norm": 0.05069069564342499,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 36336
+    },
+    {
+      "epoch": 3.7339704069050557,
+      "grad_norm": 0.0417485274374485,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 36339
+    },
+    {
+      "epoch": 3.7342786683107274,
+      "grad_norm": 0.04727747291326523,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 36342
+    },
+    {
+      "epoch": 3.7345869297163996,
+      "grad_norm": 0.056336645036935806,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 36345
+    },
+    {
+      "epoch": 3.7348951911220714,
+      "grad_norm": 0.09889603406190872,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 36348
+    },
+    {
+      "epoch": 3.7352034525277436,
+      "grad_norm": 0.06653422117233276,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 36351
+    },
+    {
+      "epoch": 3.7355117139334153,
+      "grad_norm": 0.05115421861410141,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 36354
+    },
+    {
+      "epoch": 3.7358199753390875,
+      "grad_norm": 0.04250407963991165,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 36357
+    },
+    {
+      "epoch": 3.7361282367447597,
+      "grad_norm": 0.037854380905628204,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 36360
+    },
+    {
+      "epoch": 3.7364364981504314,
+      "grad_norm": 0.06039261072874069,
+      "learning_rate": 0.01,
+      "loss": 2.0364,
+      "step": 36363
+    },
+    {
+      "epoch": 3.7367447595561036,
+      "grad_norm": 0.09946703165769577,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 36366
+    },
+    {
+      "epoch": 3.737053020961776,
+      "grad_norm": 0.10038801282644272,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 36369
+    },
+    {
+      "epoch": 3.7373612823674476,
+      "grad_norm": 0.07845285534858704,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 36372
+    },
+    {
+      "epoch": 3.7376695437731193,
+      "grad_norm": 0.08079241961240768,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 36375
+    },
+    {
+      "epoch": 3.7379778051787915,
+      "grad_norm": 0.04414941743016243,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 36378
+    },
+    {
+      "epoch": 3.7382860665844637,
+      "grad_norm": 0.14685457944869995,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 36381
+    },
+    {
+      "epoch": 3.7385943279901355,
+      "grad_norm": 0.04633990302681923,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 36384
+    },
+    {
+      "epoch": 3.7389025893958077,
+      "grad_norm": 0.04103631526231766,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 36387
+    },
+    {
+      "epoch": 3.73921085080148,
+      "grad_norm": 0.03546123206615448,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 36390
+    },
+    {
+      "epoch": 3.7395191122071516,
+      "grad_norm": 0.08792685717344284,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 36393
+    },
+    {
+      "epoch": 3.739827373612824,
+      "grad_norm": 0.09209249913692474,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 36396
+    },
+    {
+      "epoch": 3.7401356350184956,
+      "grad_norm": 0.0480416901409626,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 36399
+    },
+    {
+      "epoch": 3.7404438964241677,
+      "grad_norm": 0.03383079916238785,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 36402
+    },
+    {
+      "epoch": 3.7407521578298395,
+      "grad_norm": 0.06483932584524155,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 36405
+    },
+    {
+      "epoch": 3.7410604192355117,
+      "grad_norm": 0.04043160006403923,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 36408
+    },
+    {
+      "epoch": 3.741368680641184,
+      "grad_norm": 0.047389864921569824,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 36411
+    },
+    {
+      "epoch": 3.7416769420468556,
+      "grad_norm": 0.04405448958277702,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 36414
+    },
+    {
+      "epoch": 3.741985203452528,
+      "grad_norm": 0.03635013848543167,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 36417
+    },
+    {
+      "epoch": 3.7422934648582,
+      "grad_norm": 0.06296937167644501,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 36420
+    },
+    {
+      "epoch": 3.7426017262638718,
+      "grad_norm": 0.08138510584831238,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 36423
+    },
+    {
+      "epoch": 3.7429099876695435,
+      "grad_norm": 0.05814410001039505,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 36426
+    },
+    {
+      "epoch": 3.7432182490752157,
+      "grad_norm": 0.09043169766664505,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 36429
+    },
+    {
+      "epoch": 3.743526510480888,
+      "grad_norm": 0.038527343422174454,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 36432
+    },
+    {
+      "epoch": 3.7438347718865597,
+      "grad_norm": 0.03342151269316673,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 36435
+    },
+    {
+      "epoch": 3.744143033292232,
+      "grad_norm": 0.054572802037000656,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 36438
+    },
+    {
+      "epoch": 3.744451294697904,
+      "grad_norm": 0.10896478593349457,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 36441
+    },
+    {
+      "epoch": 3.744759556103576,
+      "grad_norm": 0.04996712505817413,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 36444
+    },
+    {
+      "epoch": 3.745067817509248,
+      "grad_norm": 0.06720001995563507,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 36447
+    },
+    {
+      "epoch": 3.7453760789149197,
+      "grad_norm": 0.07731924951076508,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 36450
+    },
+    {
+      "epoch": 3.745684340320592,
+      "grad_norm": 0.07545910775661469,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 36453
+    },
+    {
+      "epoch": 3.7459926017262637,
+      "grad_norm": 0.07236587256193161,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 36456
+    },
+    {
+      "epoch": 3.746300863131936,
+      "grad_norm": 0.05105200409889221,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 36459
+    },
+    {
+      "epoch": 3.746609124537608,
+      "grad_norm": 0.06345131993293762,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 36462
+    },
+    {
+      "epoch": 3.74691738594328,
+      "grad_norm": 0.08864062279462814,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 36465
+    },
+    {
+      "epoch": 3.747225647348952,
+      "grad_norm": 0.06058958172798157,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 36468
+    },
+    {
+      "epoch": 3.7475339087546238,
+      "grad_norm": 0.08054932206869125,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 36471
+    },
+    {
+      "epoch": 3.747842170160296,
+      "grad_norm": 0.04538315162062645,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 36474
+    },
+    {
+      "epoch": 3.7481504315659677,
+      "grad_norm": 0.11694948375225067,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 36477
+    },
+    {
+      "epoch": 3.74845869297164,
+      "grad_norm": 0.037599921226501465,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 36480
+    },
+    {
+      "epoch": 3.748766954377312,
+      "grad_norm": 0.062208350747823715,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 36483
+    },
+    {
+      "epoch": 3.749075215782984,
+      "grad_norm": 0.04657081514596939,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 36486
+    },
+    {
+      "epoch": 3.749383477188656,
+      "grad_norm": 0.11601486057043076,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 36489
+    },
+    {
+      "epoch": 3.7496917385943282,
+      "grad_norm": 0.04120590165257454,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 36492
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 0.07783524692058563,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 36495
+    },
+    {
+      "epoch": 3.7503082614056718,
+      "grad_norm": 0.04452613368630409,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 36498
+    },
+    {
+      "epoch": 3.750616522811344,
+      "grad_norm": 0.10393857210874557,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 36501
+    },
+    {
+      "epoch": 3.750924784217016,
+      "grad_norm": 0.11419197171926498,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 36504
+    },
+    {
+      "epoch": 3.751233045622688,
+      "grad_norm": 0.0752970427274704,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 36507
+    },
+    {
+      "epoch": 3.75154130702836,
+      "grad_norm": 0.06409469246864319,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 36510
+    },
+    {
+      "epoch": 3.7518495684340323,
+      "grad_norm": 0.053437262773513794,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 36513
+    },
+    {
+      "epoch": 3.752157829839704,
+      "grad_norm": 0.04102127254009247,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 36516
+    },
+    {
+      "epoch": 3.7524660912453762,
+      "grad_norm": 0.03595980256795883,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 36519
+    },
+    {
+      "epoch": 3.752774352651048,
+      "grad_norm": 0.10266067087650299,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 36522
+    },
+    {
+      "epoch": 3.75308261405672,
+      "grad_norm": 0.07302891463041306,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 36525
+    },
+    {
+      "epoch": 3.753390875462392,
+      "grad_norm": 0.08043140172958374,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 36528
+    },
+    {
+      "epoch": 3.753699136868064,
+      "grad_norm": 0.0765608549118042,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 36531
+    },
+    {
+      "epoch": 3.7540073982737363,
+      "grad_norm": 0.040770966559648514,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 36534
+    },
+    {
+      "epoch": 3.754315659679408,
+      "grad_norm": 0.09125746041536331,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 36537
+    },
+    {
+      "epoch": 3.7546239210850803,
+      "grad_norm": 0.10734240710735321,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 36540
+    },
+    {
+      "epoch": 3.754932182490752,
+      "grad_norm": 0.05876481533050537,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 36543
+    },
+    {
+      "epoch": 3.755240443896424,
+      "grad_norm": 0.04223402962088585,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 36546
+    },
+    {
+      "epoch": 3.755548705302096,
+      "grad_norm": 0.03419182077050209,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 36549
+    },
+    {
+      "epoch": 3.755856966707768,
+      "grad_norm": 0.08762037754058838,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 36552
+    },
+    {
+      "epoch": 3.7561652281134403,
+      "grad_norm": 0.035800088196992874,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 36555
+    },
+    {
+      "epoch": 3.756473489519112,
+      "grad_norm": 0.07538183778524399,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 36558
+    },
+    {
+      "epoch": 3.7567817509247843,
+      "grad_norm": 0.05067085847258568,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 36561
+    },
+    {
+      "epoch": 3.7570900123304565,
+      "grad_norm": 0.07041851431131363,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 36564
+    },
+    {
+      "epoch": 3.7573982737361282,
+      "grad_norm": 0.06819632649421692,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 36567
+    },
+    {
+      "epoch": 3.7577065351418,
+      "grad_norm": 0.11158425360918045,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 36570
+    },
+    {
+      "epoch": 3.758014796547472,
+      "grad_norm": 0.09598013758659363,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 36573
+    },
+    {
+      "epoch": 3.7583230579531444,
+      "grad_norm": 0.0501101016998291,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 36576
+    },
+    {
+      "epoch": 3.758631319358816,
+      "grad_norm": 0.03867189958691597,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 36579
+    },
+    {
+      "epoch": 3.7589395807644883,
+      "grad_norm": 0.03836961090564728,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 36582
+    },
+    {
+      "epoch": 3.7592478421701605,
+      "grad_norm": 0.04135824367403984,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 36585
+    },
+    {
+      "epoch": 3.7595561035758323,
+      "grad_norm": 0.0922931507229805,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 36588
+    },
+    {
+      "epoch": 3.7598643649815044,
+      "grad_norm": 0.048077963292598724,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 36591
+    },
+    {
+      "epoch": 3.760172626387176,
+      "grad_norm": 0.04263912886381149,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 36594
+    },
+    {
+      "epoch": 3.7604808877928484,
+      "grad_norm": 0.05825957655906677,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 36597
+    },
+    {
+      "epoch": 3.76078914919852,
+      "grad_norm": 0.04579320177435875,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 36600
+    },
+    {
+      "epoch": 3.7610974106041923,
+      "grad_norm": 0.07970302551984787,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 36603
+    },
+    {
+      "epoch": 3.7614056720098645,
+      "grad_norm": 0.1091640293598175,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 36606
+    },
+    {
+      "epoch": 3.7617139334155363,
+      "grad_norm": 0.07577066123485565,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 36609
+    },
+    {
+      "epoch": 3.7620221948212085,
+      "grad_norm": 0.07901261001825333,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 36612
+    },
+    {
+      "epoch": 3.7623304562268807,
+      "grad_norm": 0.040557824075222015,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 36615
+    },
+    {
+      "epoch": 3.7626387176325524,
+      "grad_norm": 0.03163420036435127,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 36618
+    },
+    {
+      "epoch": 3.762946979038224,
+      "grad_norm": 0.0960499569773674,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 36621
+    },
+    {
+      "epoch": 3.7632552404438964,
+      "grad_norm": 0.05283837392926216,
+      "learning_rate": 0.01,
+      "loss": 2.0351,
+      "step": 36624
+    },
+    {
+      "epoch": 3.7635635018495686,
+      "grad_norm": 0.03764748573303223,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 36627
+    },
+    {
+      "epoch": 3.7638717632552403,
+      "grad_norm": 0.07017308473587036,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 36630
+    },
+    {
+      "epoch": 3.7641800246609125,
+      "grad_norm": 0.06506321579217911,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 36633
+    },
+    {
+      "epoch": 3.7644882860665847,
+      "grad_norm": 0.040651753544807434,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 36636
+    },
+    {
+      "epoch": 3.7647965474722564,
+      "grad_norm": 0.0897100567817688,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 36639
+    },
+    {
+      "epoch": 3.7651048088779286,
+      "grad_norm": 0.1067187637090683,
+      "learning_rate": 0.01,
+      "loss": 2.0369,
+      "step": 36642
+    },
+    {
+      "epoch": 3.7654130702836004,
+      "grad_norm": 0.08220155537128448,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 36645
+    },
+    {
+      "epoch": 3.7657213316892726,
+      "grad_norm": 0.0830981582403183,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 36648
+    },
+    {
+      "epoch": 3.7660295930949443,
+      "grad_norm": 0.04124632850289345,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 36651
+    },
+    {
+      "epoch": 3.7663378545006165,
+      "grad_norm": 0.03570066764950752,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 36654
+    },
+    {
+      "epoch": 3.7666461159062887,
+      "grad_norm": 0.04715826362371445,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 36657
+    },
+    {
+      "epoch": 3.7669543773119605,
+      "grad_norm": 0.05129298195242882,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 36660
+    },
+    {
+      "epoch": 3.7672626387176327,
+      "grad_norm": 0.07690376043319702,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 36663
+    },
+    {
+      "epoch": 3.7675709001233044,
+      "grad_norm": 0.06316812336444855,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 36666
+    },
+    {
+      "epoch": 3.7678791615289766,
+      "grad_norm": 0.045724622905254364,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 36669
+    },
+    {
+      "epoch": 3.7681874229346484,
+      "grad_norm": 0.12661319971084595,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 36672
+    },
+    {
+      "epoch": 3.7684956843403206,
+      "grad_norm": 0.03898628428578377,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 36675
+    },
+    {
+      "epoch": 3.7688039457459928,
+      "grad_norm": 0.04297169670462608,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 36678
+    },
+    {
+      "epoch": 3.7691122071516645,
+      "grad_norm": 0.047098271548748016,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 36681
+    },
+    {
+      "epoch": 3.7694204685573367,
+      "grad_norm": 0.0880102887749672,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 36684
+    },
+    {
+      "epoch": 3.769728729963009,
+      "grad_norm": 0.09598886221647263,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 36687
+    },
+    {
+      "epoch": 3.7700369913686806,
+      "grad_norm": 0.07920796424150467,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 36690
+    },
+    {
+      "epoch": 3.7703452527743524,
+      "grad_norm": 0.09643390029668808,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 36693
+    },
+    {
+      "epoch": 3.7706535141800246,
+      "grad_norm": 0.07026758790016174,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 36696
+    },
+    {
+      "epoch": 3.770961775585697,
+      "grad_norm": 0.04533863440155983,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 36699
+    },
+    {
+      "epoch": 3.7712700369913685,
+      "grad_norm": 0.03511887788772583,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 36702
+    },
+    {
+      "epoch": 3.7715782983970407,
+      "grad_norm": 0.03846128657460213,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 36705
+    },
+    {
+      "epoch": 3.771886559802713,
+      "grad_norm": 0.07523466646671295,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 36708
+    },
+    {
+      "epoch": 3.7721948212083847,
+      "grad_norm": 0.09649614989757538,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 36711
+    },
+    {
+      "epoch": 3.772503082614057,
+      "grad_norm": 0.04845314472913742,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 36714
+    },
+    {
+      "epoch": 3.7728113440197286,
+      "grad_norm": 0.037477899342775345,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 36717
+    },
+    {
+      "epoch": 3.773119605425401,
+      "grad_norm": 0.08308611810207367,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 36720
+    },
+    {
+      "epoch": 3.7734278668310726,
+      "grad_norm": 0.048357848078012466,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 36723
+    },
+    {
+      "epoch": 3.7737361282367448,
+      "grad_norm": 0.04424552246928215,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 36726
+    },
+    {
+      "epoch": 3.774044389642417,
+      "grad_norm": 0.04857276752591133,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 36729
+    },
+    {
+      "epoch": 3.7743526510480887,
+      "grad_norm": 0.039389774203300476,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 36732
+    },
+    {
+      "epoch": 3.774660912453761,
+      "grad_norm": 0.04370071366429329,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 36735
+    },
+    {
+      "epoch": 3.7749691738594326,
+      "grad_norm": 0.05456184223294258,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 36738
+    },
+    {
+      "epoch": 3.775277435265105,
+      "grad_norm": 0.0888199731707573,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 36741
+    },
+    {
+      "epoch": 3.7755856966707766,
+      "grad_norm": 0.19867762923240662,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 36744
+    },
+    {
+      "epoch": 3.775893958076449,
+      "grad_norm": 0.17659060657024384,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 36747
+    },
+    {
+      "epoch": 3.776202219482121,
+      "grad_norm": 0.0629454255104065,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 36750
+    },
+    {
+      "epoch": 3.7765104808877927,
+      "grad_norm": 0.05801904946565628,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 36753
+    },
+    {
+      "epoch": 3.776818742293465,
+      "grad_norm": 0.04907959699630737,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 36756
+    },
+    {
+      "epoch": 3.777127003699137,
+      "grad_norm": 0.08606990426778793,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 36759
+    },
+    {
+      "epoch": 3.777435265104809,
+      "grad_norm": 0.03293627128005028,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 36762
+    },
+    {
+      "epoch": 3.7777435265104806,
+      "grad_norm": 0.033272456377744675,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 36765
+    },
+    {
+      "epoch": 3.778051787916153,
+      "grad_norm": 0.10520119965076447,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 36768
+    },
+    {
+      "epoch": 3.778360049321825,
+      "grad_norm": 0.03949661925435066,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 36771
+    },
+    {
+      "epoch": 3.7786683107274968,
+      "grad_norm": 0.10781146585941315,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 36774
+    },
+    {
+      "epoch": 3.778976572133169,
+      "grad_norm": 0.1707312911748886,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 36777
+    },
+    {
+      "epoch": 3.779284833538841,
+      "grad_norm": 0.10689792037010193,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 36780
+    },
+    {
+      "epoch": 3.779593094944513,
+      "grad_norm": 0.06377099454402924,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 36783
+    },
+    {
+      "epoch": 3.779901356350185,
+      "grad_norm": 0.050422243773937225,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 36786
+    },
+    {
+      "epoch": 3.780209617755857,
+      "grad_norm": 0.041165612637996674,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 36789
+    },
+    {
+      "epoch": 3.780517879161529,
+      "grad_norm": 0.07220548391342163,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 36792
+    },
+    {
+      "epoch": 3.780826140567201,
+      "grad_norm": 0.05860710144042969,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 36795
+    },
+    {
+      "epoch": 3.781134401972873,
+      "grad_norm": 0.03842824697494507,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 36798
+    },
+    {
+      "epoch": 3.781442663378545,
+      "grad_norm": 0.04346943646669388,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 36801
+    },
+    {
+      "epoch": 3.781750924784217,
+      "grad_norm": 0.16297590732574463,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 36804
+    },
+    {
+      "epoch": 3.782059186189889,
+      "grad_norm": 0.06720095127820969,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 36807
+    },
+    {
+      "epoch": 3.782367447595561,
+      "grad_norm": 0.04820172116160393,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 36810
+    },
+    {
+      "epoch": 3.782675709001233,
+      "grad_norm": 0.0431947223842144,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 36813
+    },
+    {
+      "epoch": 3.782983970406905,
+      "grad_norm": 0.0517662838101387,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 36816
+    },
+    {
+      "epoch": 3.783292231812577,
+      "grad_norm": 0.06249597296118736,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 36819
+    },
+    {
+      "epoch": 3.783600493218249,
+      "grad_norm": 0.04001644626259804,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 36822
+    },
+    {
+      "epoch": 3.783908754623921,
+      "grad_norm": 0.0713401585817337,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 36825
+    },
+    {
+      "epoch": 3.784217016029593,
+      "grad_norm": 0.08583984524011612,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 36828
+    },
+    {
+      "epoch": 3.7845252774352653,
+      "grad_norm": 0.05482606962323189,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 36831
+    },
+    {
+      "epoch": 3.784833538840937,
+      "grad_norm": 0.04373779147863388,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 36834
+    },
+    {
+      "epoch": 3.7851418002466093,
+      "grad_norm": 0.09752769768238068,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 36837
+    },
+    {
+      "epoch": 3.785450061652281,
+      "grad_norm": 0.07635717839002609,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 36840
+    },
+    {
+      "epoch": 3.7857583230579532,
+      "grad_norm": 0.07021810859441757,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 36843
+    },
+    {
+      "epoch": 3.786066584463625,
+      "grad_norm": 0.0541268065571785,
+      "learning_rate": 0.01,
+      "loss": 2.0252,
+      "step": 36846
+    },
+    {
+      "epoch": 3.786374845869297,
+      "grad_norm": 0.07205335795879364,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 36849
+    },
+    {
+      "epoch": 3.7866831072749694,
+      "grad_norm": 0.057571567595005035,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 36852
+    },
+    {
+      "epoch": 3.786991368680641,
+      "grad_norm": 0.11135399341583252,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 36855
+    },
+    {
+      "epoch": 3.7872996300863133,
+      "grad_norm": 0.0698394924402237,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 36858
+    },
+    {
+      "epoch": 3.787607891491985,
+      "grad_norm": 0.09506375342607498,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 36861
+    },
+    {
+      "epoch": 3.7879161528976573,
+      "grad_norm": 0.08534973114728928,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 36864
+    },
+    {
+      "epoch": 3.788224414303329,
+      "grad_norm": 0.08360368013381958,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 36867
+    },
+    {
+      "epoch": 3.788532675709001,
+      "grad_norm": 0.08635003864765167,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 36870
+    },
+    {
+      "epoch": 3.7888409371146734,
+      "grad_norm": 0.033921390771865845,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 36873
+    },
+    {
+      "epoch": 3.789149198520345,
+      "grad_norm": 0.04312542825937271,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 36876
+    },
+    {
+      "epoch": 3.7894574599260173,
+      "grad_norm": 0.06157702952623367,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 36879
+    },
+    {
+      "epoch": 3.7897657213316895,
+      "grad_norm": 0.05039060115814209,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 36882
+    },
+    {
+      "epoch": 3.7900739827373613,
+      "grad_norm": 0.05873194709420204,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 36885
+    },
+    {
+      "epoch": 3.790382244143033,
+      "grad_norm": 0.07444643974304199,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 36888
+    },
+    {
+      "epoch": 3.7906905055487052,
+      "grad_norm": 0.06560081243515015,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 36891
+    },
+    {
+      "epoch": 3.7909987669543774,
+      "grad_norm": 0.08456238359212875,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 36894
+    },
+    {
+      "epoch": 3.791307028360049,
+      "grad_norm": 0.09048344194889069,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 36897
+    },
+    {
+      "epoch": 3.7916152897657214,
+      "grad_norm": 0.07640919834375381,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 36900
+    },
+    {
+      "epoch": 3.7919235511713936,
+      "grad_norm": 0.0968911275267601,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 36903
+    },
+    {
+      "epoch": 3.7922318125770653,
+      "grad_norm": 0.08743193000555038,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 36906
+    },
+    {
+      "epoch": 3.7925400739827375,
+      "grad_norm": 0.056414127349853516,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 36909
+    },
+    {
+      "epoch": 3.7928483353884093,
+      "grad_norm": 0.04174700006842613,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 36912
+    },
+    {
+      "epoch": 3.7931565967940815,
+      "grad_norm": 0.03753121197223663,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 36915
+    },
+    {
+      "epoch": 3.793464858199753,
+      "grad_norm": 0.04957683011889458,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 36918
+    },
+    {
+      "epoch": 3.7937731196054254,
+      "grad_norm": 0.051355455070734024,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 36921
+    },
+    {
+      "epoch": 3.7940813810110976,
+      "grad_norm": 0.0345148891210556,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 36924
+    },
+    {
+      "epoch": 3.7943896424167693,
+      "grad_norm": 0.12389633059501648,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 36927
+    },
+    {
+      "epoch": 3.7946979038224415,
+      "grad_norm": 0.05504097789525986,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 36930
+    },
+    {
+      "epoch": 3.7950061652281133,
+      "grad_norm": 0.044859953224658966,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 36933
+    },
+    {
+      "epoch": 3.7953144266337855,
+      "grad_norm": 0.06867794692516327,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 36936
+    },
+    {
+      "epoch": 3.7956226880394572,
+      "grad_norm": 0.10308001190423965,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 36939
+    },
+    {
+      "epoch": 3.7959309494451294,
+      "grad_norm": 0.07946982234716415,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 36942
+    },
+    {
+      "epoch": 3.7962392108508016,
+      "grad_norm": 0.058571770787239075,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 36945
+    },
+    {
+      "epoch": 3.7965474722564734,
+      "grad_norm": 0.07778012007474899,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 36948
+    },
+    {
+      "epoch": 3.7968557336621456,
+      "grad_norm": 0.06965510547161102,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 36951
+    },
+    {
+      "epoch": 3.7971639950678178,
+      "grad_norm": 0.06443614512681961,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 36954
+    },
+    {
+      "epoch": 3.7974722564734895,
+      "grad_norm": 0.04750034958124161,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 36957
+    },
+    {
+      "epoch": 3.7977805178791613,
+      "grad_norm": 0.039454251527786255,
+      "learning_rate": 0.01,
+      "loss": 2.0343,
+      "step": 36960
+    },
+    {
+      "epoch": 3.7980887792848335,
+      "grad_norm": 0.07771621644496918,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 36963
+    },
+    {
+      "epoch": 3.7983970406905057,
+      "grad_norm": 0.06523919850587845,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 36966
+    },
+    {
+      "epoch": 3.7987053020961774,
+      "grad_norm": 0.055610157549381256,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 36969
+    },
+    {
+      "epoch": 3.7990135635018496,
+      "grad_norm": 0.08819667249917984,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 36972
+    },
+    {
+      "epoch": 3.799321824907522,
+      "grad_norm": 0.05418279394507408,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 36975
+    },
+    {
+      "epoch": 3.7996300863131935,
+      "grad_norm": 0.06648479402065277,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 36978
+    },
+    {
+      "epoch": 3.7999383477188657,
+      "grad_norm": 0.04629239812493324,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 36981
+    },
+    {
+      "epoch": 3.8002466091245375,
+      "grad_norm": 0.042264122515916824,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 36984
+    },
+    {
+      "epoch": 3.8005548705302097,
+      "grad_norm": 0.10186997801065445,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 36987
+    },
+    {
+      "epoch": 3.8008631319358814,
+      "grad_norm": 0.09278517216444016,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 36990
+    },
+    {
+      "epoch": 3.8011713933415536,
+      "grad_norm": 0.07084151357412338,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 36993
+    },
+    {
+      "epoch": 3.801479654747226,
+      "grad_norm": 0.07990908622741699,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 36996
+    },
+    {
+      "epoch": 3.8017879161528976,
+      "grad_norm": 0.039643771946430206,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 36999
+    },
+    {
+      "epoch": 3.8020961775585698,
+      "grad_norm": 0.03678774833679199,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 37002
+    },
+    {
+      "epoch": 3.8024044389642415,
+      "grad_norm": 0.040611330419778824,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 37005
+    },
+    {
+      "epoch": 3.8027127003699137,
+      "grad_norm": 0.03812083601951599,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 37008
+    },
+    {
+      "epoch": 3.8030209617755855,
+      "grad_norm": 0.07582142949104309,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 37011
+    },
+    {
+      "epoch": 3.8033292231812577,
+      "grad_norm": 0.14520719647407532,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 37014
+    },
+    {
+      "epoch": 3.80363748458693,
+      "grad_norm": 0.04497974365949631,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 37017
+    },
+    {
+      "epoch": 3.8039457459926016,
+      "grad_norm": 0.04178651422262192,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 37020
+    },
+    {
+      "epoch": 3.804254007398274,
+      "grad_norm": 0.047346848994493484,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 37023
+    },
+    {
+      "epoch": 3.804562268803946,
+      "grad_norm": 0.04626936465501785,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 37026
+    },
+    {
+      "epoch": 3.8048705302096177,
+      "grad_norm": 0.055330242961645126,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 37029
+    },
+    {
+      "epoch": 3.8051787916152895,
+      "grad_norm": 0.040033504366874695,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 37032
+    },
+    {
+      "epoch": 3.8054870530209617,
+      "grad_norm": 0.03897986188530922,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 37035
+    },
+    {
+      "epoch": 3.805795314426634,
+      "grad_norm": 0.07525118440389633,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 37038
+    },
+    {
+      "epoch": 3.8061035758323056,
+      "grad_norm": 0.06256567686796188,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 37041
+    },
+    {
+      "epoch": 3.806411837237978,
+      "grad_norm": 0.05028248950839043,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 37044
+    },
+    {
+      "epoch": 3.80672009864365,
+      "grad_norm": 0.03530338406562805,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 37047
+    },
+    {
+      "epoch": 3.8070283600493218,
+      "grad_norm": 0.044229693710803986,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 37050
+    },
+    {
+      "epoch": 3.807336621454994,
+      "grad_norm": 0.1117880642414093,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 37053
+    },
+    {
+      "epoch": 3.8076448828606657,
+      "grad_norm": 0.04328616335988045,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 37056
+    },
+    {
+      "epoch": 3.807953144266338,
+      "grad_norm": 0.05948890000581741,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 37059
+    },
+    {
+      "epoch": 3.8082614056720097,
+      "grad_norm": 0.043167680501937866,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 37062
+    },
+    {
+      "epoch": 3.808569667077682,
+      "grad_norm": 0.04852902144193649,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 37065
+    },
+    {
+      "epoch": 3.808877928483354,
+      "grad_norm": 0.06212899461388588,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 37068
+    },
+    {
+      "epoch": 3.809186189889026,
+      "grad_norm": 0.07403160631656647,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 37071
+    },
+    {
+      "epoch": 3.809494451294698,
+      "grad_norm": 0.07734464108943939,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 37074
+    },
+    {
+      "epoch": 3.80980271270037,
+      "grad_norm": 0.05057765170931816,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 37077
+    },
+    {
+      "epoch": 3.810110974106042,
+      "grad_norm": 0.11746016144752502,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 37080
+    },
+    {
+      "epoch": 3.8104192355117137,
+      "grad_norm": 0.10531330853700638,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 37083
+    },
+    {
+      "epoch": 3.810727496917386,
+      "grad_norm": 0.11436725407838821,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 37086
+    },
+    {
+      "epoch": 3.811035758323058,
+      "grad_norm": 0.04365989565849304,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 37089
+    },
+    {
+      "epoch": 3.81134401972873,
+      "grad_norm": 0.04271963611245155,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 37092
+    },
+    {
+      "epoch": 3.811652281134402,
+      "grad_norm": 0.03649172931909561,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 37095
+    },
+    {
+      "epoch": 3.811960542540074,
+      "grad_norm": 0.054175905883312225,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 37098
+    },
+    {
+      "epoch": 3.812268803945746,
+      "grad_norm": 0.041479047387838364,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 37101
+    },
+    {
+      "epoch": 3.812577065351418,
+      "grad_norm": 0.03695906326174736,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 37104
+    },
+    {
+      "epoch": 3.81288532675709,
+      "grad_norm": 0.05167945846915245,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 37107
+    },
+    {
+      "epoch": 3.813193588162762,
+      "grad_norm": 0.0706062912940979,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 37110
+    },
+    {
+      "epoch": 3.813501849568434,
+      "grad_norm": 0.04282987117767334,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 37113
+    },
+    {
+      "epoch": 3.813810110974106,
+      "grad_norm": 0.048825137317180634,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 37116
+    },
+    {
+      "epoch": 3.8141183723797782,
+      "grad_norm": 0.13310399651527405,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 37119
+    },
+    {
+      "epoch": 3.81442663378545,
+      "grad_norm": 0.05691874772310257,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 37122
+    },
+    {
+      "epoch": 3.814734895191122,
+      "grad_norm": 0.07542920112609863,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 37125
+    },
+    {
+      "epoch": 3.815043156596794,
+      "grad_norm": 0.09146812558174133,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 37128
+    },
+    {
+      "epoch": 3.815351418002466,
+      "grad_norm": 0.04742105305194855,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 37131
+    },
+    {
+      "epoch": 3.815659679408138,
+      "grad_norm": 0.03387041017413139,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 37134
+    },
+    {
+      "epoch": 3.81596794081381,
+      "grad_norm": 0.036746036261320114,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 37137
+    },
+    {
+      "epoch": 3.8162762022194823,
+      "grad_norm": 0.06593815982341766,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 37140
+    },
+    {
+      "epoch": 3.816584463625154,
+      "grad_norm": 0.10538439452648163,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 37143
+    },
+    {
+      "epoch": 3.816892725030826,
+      "grad_norm": 0.07849781960248947,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 37146
+    },
+    {
+      "epoch": 3.8172009864364984,
+      "grad_norm": 0.10899082571268082,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 37149
+    },
+    {
+      "epoch": 3.81750924784217,
+      "grad_norm": 0.05171935632824898,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 37152
+    },
+    {
+      "epoch": 3.817817509247842,
+      "grad_norm": 0.06028240546584129,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 37155
+    },
+    {
+      "epoch": 3.818125770653514,
+      "grad_norm": 0.0440855398774147,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 37158
+    },
+    {
+      "epoch": 3.8184340320591863,
+      "grad_norm": 0.05764901638031006,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 37161
+    },
+    {
+      "epoch": 3.818742293464858,
+      "grad_norm": 0.042115770280361176,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 37164
+    },
+    {
+      "epoch": 3.8190505548705302,
+      "grad_norm": 0.0819876417517662,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 37167
+    },
+    {
+      "epoch": 3.8193588162762024,
+      "grad_norm": 0.09432919323444366,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 37170
+    },
+    {
+      "epoch": 3.819667077681874,
+      "grad_norm": 0.1223209798336029,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 37173
+    },
+    {
+      "epoch": 3.8199753390875464,
+      "grad_norm": 0.08615586161613464,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 37176
+    },
+    {
+      "epoch": 3.820283600493218,
+      "grad_norm": 0.0847785696387291,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 37179
+    },
+    {
+      "epoch": 3.8205918618988903,
+      "grad_norm": 0.059272028505802155,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 37182
+    },
+    {
+      "epoch": 3.820900123304562,
+      "grad_norm": 0.042237572371959686,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 37185
+    },
+    {
+      "epoch": 3.8212083847102343,
+      "grad_norm": 0.04662346839904785,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 37188
+    },
+    {
+      "epoch": 3.8215166461159065,
+      "grad_norm": 0.04566279426217079,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 37191
+    },
+    {
+      "epoch": 3.821824907521578,
+      "grad_norm": 0.07829894125461578,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 37194
+    },
+    {
+      "epoch": 3.8221331689272504,
+      "grad_norm": 0.09386474639177322,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 37197
+    },
+    {
+      "epoch": 3.822441430332922,
+      "grad_norm": 0.04232482239603996,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 37200
+    },
+    {
+      "epoch": 3.8227496917385944,
+      "grad_norm": 0.09978868067264557,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 37203
+    },
+    {
+      "epoch": 3.823057953144266,
+      "grad_norm": 0.04246772453188896,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 37206
+    },
+    {
+      "epoch": 3.8233662145499383,
+      "grad_norm": 0.13856659829616547,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 37209
+    },
+    {
+      "epoch": 3.8236744759556105,
+      "grad_norm": 0.05659123882651329,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 37212
+    },
+    {
+      "epoch": 3.8239827373612822,
+      "grad_norm": 0.05944579839706421,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 37215
+    },
+    {
+      "epoch": 3.8242909987669544,
+      "grad_norm": 0.040483999997377396,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 37218
+    },
+    {
+      "epoch": 3.8245992601726266,
+      "grad_norm": 0.05651898682117462,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 37221
+    },
+    {
+      "epoch": 3.8249075215782984,
+      "grad_norm": 0.04841217026114464,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 37224
+    },
+    {
+      "epoch": 3.82521578298397,
+      "grad_norm": 0.05571025237441063,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 37227
+    },
+    {
+      "epoch": 3.8255240443896423,
+      "grad_norm": 0.07893476635217667,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 37230
+    },
+    {
+      "epoch": 3.8258323057953145,
+      "grad_norm": 0.07275935262441635,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 37233
+    },
+    {
+      "epoch": 3.8261405672009863,
+      "grad_norm": 0.14288173615932465,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 37236
+    },
+    {
+      "epoch": 3.8264488286066585,
+      "grad_norm": 0.18501073122024536,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 37239
+    },
+    {
+      "epoch": 3.8267570900123307,
+      "grad_norm": 0.12003073841333389,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 37242
+    },
+    {
+      "epoch": 3.8270653514180024,
+      "grad_norm": 0.05952821299433708,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 37245
+    },
+    {
+      "epoch": 3.8273736128236746,
+      "grad_norm": 0.052555590867996216,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 37248
+    },
+    {
+      "epoch": 3.8276818742293464,
+      "grad_norm": 0.05527700483798981,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 37251
+    },
+    {
+      "epoch": 3.8279901356350186,
+      "grad_norm": 0.043385427445173264,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 37254
+    },
+    {
+      "epoch": 3.8282983970406903,
+      "grad_norm": 0.05360834673047066,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 37257
+    },
+    {
+      "epoch": 3.8286066584463625,
+      "grad_norm": 0.04542825371026993,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 37260
+    },
+    {
+      "epoch": 3.8289149198520347,
+      "grad_norm": 0.04607833921909332,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 37263
+    },
+    {
+      "epoch": 3.8292231812577064,
+      "grad_norm": 0.059813372790813446,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 37266
+    },
+    {
+      "epoch": 3.8295314426633786,
+      "grad_norm": 0.11444689333438873,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 37269
+    },
+    {
+      "epoch": 3.829839704069051,
+      "grad_norm": 0.07510776817798615,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 37272
+    },
+    {
+      "epoch": 3.8301479654747226,
+      "grad_norm": 0.051781926304101944,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 37275
+    },
+    {
+      "epoch": 3.8304562268803943,
+      "grad_norm": 0.051285672932863235,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 37278
+    },
+    {
+      "epoch": 3.8307644882860665,
+      "grad_norm": 0.0464506596326828,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 37281
+    },
+    {
+      "epoch": 3.8310727496917387,
+      "grad_norm": 0.03997113183140755,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 37284
+    },
+    {
+      "epoch": 3.8313810110974105,
+      "grad_norm": 0.03474081680178642,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 37287
+    },
+    {
+      "epoch": 3.8316892725030827,
+      "grad_norm": 0.13113395869731903,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 37290
+    },
+    {
+      "epoch": 3.831997533908755,
+      "grad_norm": 0.03894534707069397,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 37293
+    },
+    {
+      "epoch": 3.8323057953144266,
+      "grad_norm": 0.030656691640615463,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 37296
+    },
+    {
+      "epoch": 3.832614056720099,
+      "grad_norm": 0.08632262796163559,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 37299
+    },
+    {
+      "epoch": 3.8329223181257706,
+      "grad_norm": 0.09762530028820038,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 37302
+    },
+    {
+      "epoch": 3.8332305795314427,
+      "grad_norm": 0.07323916256427765,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 37305
+    },
+    {
+      "epoch": 3.8335388409371145,
+      "grad_norm": 0.08117776364088058,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 37308
+    },
+    {
+      "epoch": 3.8338471023427867,
+      "grad_norm": 0.07396585494279861,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 37311
+    },
+    {
+      "epoch": 3.834155363748459,
+      "grad_norm": 0.07192173600196838,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 37314
+    },
+    {
+      "epoch": 3.8344636251541306,
+      "grad_norm": 0.06731805950403214,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 37317
+    },
+    {
+      "epoch": 3.834771886559803,
+      "grad_norm": 0.06851818412542343,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 37320
+    },
+    {
+      "epoch": 3.8350801479654746,
+      "grad_norm": 0.07433861494064331,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 37323
+    },
+    {
+      "epoch": 3.835388409371147,
+      "grad_norm": 0.13405515253543854,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 37326
+    },
+    {
+      "epoch": 3.8356966707768185,
+      "grad_norm": 0.0912710577249527,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 37329
+    },
+    {
+      "epoch": 3.8360049321824907,
+      "grad_norm": 0.05791820213198662,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 37332
+    },
+    {
+      "epoch": 3.836313193588163,
+      "grad_norm": 0.06457755714654922,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 37335
+    },
+    {
+      "epoch": 3.8366214549938347,
+      "grad_norm": 0.06486820429563522,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 37338
+    },
+    {
+      "epoch": 3.836929716399507,
+      "grad_norm": 0.04546729475259781,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 37341
+    },
+    {
+      "epoch": 3.837237977805179,
+      "grad_norm": 0.058978717774152756,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 37344
+    },
+    {
+      "epoch": 3.837546239210851,
+      "grad_norm": 0.09566085040569305,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 37347
+    },
+    {
+      "epoch": 3.8378545006165226,
+      "grad_norm": 0.038889624178409576,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 37350
+    },
+    {
+      "epoch": 3.8381627620221948,
+      "grad_norm": 0.09589283168315887,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 37353
+    },
+    {
+      "epoch": 3.838471023427867,
+      "grad_norm": 0.04719104990363121,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 37356
+    },
+    {
+      "epoch": 3.8387792848335387,
+      "grad_norm": 0.09488753974437714,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 37359
+    },
+    {
+      "epoch": 3.839087546239211,
+      "grad_norm": 0.09091513603925705,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 37362
+    },
+    {
+      "epoch": 3.839395807644883,
+      "grad_norm": 0.06685637682676315,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 37365
+    },
+    {
+      "epoch": 3.839704069050555,
+      "grad_norm": 0.0734042376279831,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 37368
+    },
+    {
+      "epoch": 3.840012330456227,
+      "grad_norm": 0.04996969550848007,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 37371
+    },
+    {
+      "epoch": 3.840320591861899,
+      "grad_norm": 0.07901585847139359,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 37374
+    },
+    {
+      "epoch": 3.840628853267571,
+      "grad_norm": 0.06289222836494446,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 37377
+    },
+    {
+      "epoch": 3.8409371146732427,
+      "grad_norm": 0.07179414480924606,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 37380
+    },
+    {
+      "epoch": 3.841245376078915,
+      "grad_norm": 0.05835287645459175,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 37383
+    },
+    {
+      "epoch": 3.841553637484587,
+      "grad_norm": 0.07789372652769089,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 37386
+    },
+    {
+      "epoch": 3.841861898890259,
+      "grad_norm": 0.10866318643093109,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 37389
+    },
+    {
+      "epoch": 3.842170160295931,
+      "grad_norm": 0.12253516167402267,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 37392
+    },
+    {
+      "epoch": 3.842478421701603,
+      "grad_norm": 0.1773093193769455,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 37395
+    },
+    {
+      "epoch": 3.842786683107275,
+      "grad_norm": 0.12998591363430023,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 37398
+    },
+    {
+      "epoch": 3.8430949445129468,
+      "grad_norm": 0.07139603793621063,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 37401
+    },
+    {
+      "epoch": 3.843403205918619,
+      "grad_norm": 0.05447724089026451,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 37404
+    },
+    {
+      "epoch": 3.843711467324291,
+      "grad_norm": 0.053450725972652435,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 37407
+    },
+    {
+      "epoch": 3.844019728729963,
+      "grad_norm": 0.042635634541511536,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 37410
+    },
+    {
+      "epoch": 3.844327990135635,
+      "grad_norm": 0.06024963781237602,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 37413
+    },
+    {
+      "epoch": 3.8446362515413073,
+      "grad_norm": 0.04149017482995987,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 37416
+    },
+    {
+      "epoch": 3.844944512946979,
+      "grad_norm": 0.043965183198451996,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 37419
+    },
+    {
+      "epoch": 3.845252774352651,
+      "grad_norm": 0.06676291674375534,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 37422
+    },
+    {
+      "epoch": 3.845561035758323,
+      "grad_norm": 0.1391642689704895,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 37425
+    },
+    {
+      "epoch": 3.845869297163995,
+      "grad_norm": 0.12824542820453644,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 37428
+    },
+    {
+      "epoch": 3.846177558569667,
+      "grad_norm": 0.038818515837192535,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 37431
+    },
+    {
+      "epoch": 3.846485819975339,
+      "grad_norm": 0.06105605140328407,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 37434
+    },
+    {
+      "epoch": 3.8467940813810113,
+      "grad_norm": 0.05692486837506294,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 37437
+    },
+    {
+      "epoch": 3.847102342786683,
+      "grad_norm": 0.03328600525856018,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 37440
+    },
+    {
+      "epoch": 3.8474106041923553,
+      "grad_norm": 0.05563362315297127,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 37443
+    },
+    {
+      "epoch": 3.847718865598027,
+      "grad_norm": 0.04523022845387459,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 37446
+    },
+    {
+      "epoch": 3.848027127003699,
+      "grad_norm": 0.12618042528629303,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 37449
+    },
+    {
+      "epoch": 3.848335388409371,
+      "grad_norm": 0.10030481964349747,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 37452
+    },
+    {
+      "epoch": 3.848643649815043,
+      "grad_norm": 0.09524470567703247,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 37455
+    },
+    {
+      "epoch": 3.8489519112207153,
+      "grad_norm": 0.06369511038064957,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 37458
+    },
+    {
+      "epoch": 3.849260172626387,
+      "grad_norm": 0.052571844309568405,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 37461
+    },
+    {
+      "epoch": 3.8495684340320593,
+      "grad_norm": 0.040871210396289825,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 37464
+    },
+    {
+      "epoch": 3.849876695437731,
+      "grad_norm": 0.0393734946846962,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 37467
+    },
+    {
+      "epoch": 3.8501849568434032,
+      "grad_norm": 0.10289833694696426,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 37470
+    },
+    {
+      "epoch": 3.850493218249075,
+      "grad_norm": 0.053732771426439285,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 37473
+    },
+    {
+      "epoch": 3.850801479654747,
+      "grad_norm": 0.035672031342983246,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 37476
+    },
+    {
+      "epoch": 3.8511097410604194,
+      "grad_norm": 0.07263844460248947,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 37479
+    },
+    {
+      "epoch": 3.851418002466091,
+      "grad_norm": 0.06151323765516281,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 37482
+    },
+    {
+      "epoch": 3.8517262638717633,
+      "grad_norm": 0.08033137768507004,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 37485
+    },
+    {
+      "epoch": 3.8520345252774355,
+      "grad_norm": 0.09900759160518646,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 37488
+    },
+    {
+      "epoch": 3.8523427866831073,
+      "grad_norm": 0.061891715973615646,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 37491
+    },
+    {
+      "epoch": 3.8526510480887795,
+      "grad_norm": 0.039978478103876114,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 37494
+    },
+    {
+      "epoch": 3.852959309494451,
+      "grad_norm": 0.04366837814450264,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 37497
+    },
+    {
+      "epoch": 3.8532675709001234,
+      "grad_norm": 0.03704690560698509,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 37500
+    },
+    {
+      "epoch": 3.853575832305795,
+      "grad_norm": 0.03964201733469963,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 37503
+    },
+    {
+      "epoch": 3.8538840937114673,
+      "grad_norm": 0.060866422951221466,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 37506
+    },
+    {
+      "epoch": 3.8541923551171395,
+      "grad_norm": 0.1373164802789688,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 37509
+    },
+    {
+      "epoch": 3.8545006165228113,
+      "grad_norm": 0.12396371364593506,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 37512
+    },
+    {
+      "epoch": 3.8548088779284835,
+      "grad_norm": 0.05732661485671997,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 37515
+    },
+    {
+      "epoch": 3.8551171393341552,
+      "grad_norm": 0.072254478931427,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 37518
+    },
+    {
+      "epoch": 3.8554254007398274,
+      "grad_norm": 0.05103025957942009,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 37521
+    },
+    {
+      "epoch": 3.855733662145499,
+      "grad_norm": 0.06759181618690491,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 37524
+    },
+    {
+      "epoch": 3.8560419235511714,
+      "grad_norm": 0.08190746605396271,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 37527
+    },
+    {
+      "epoch": 3.8563501849568436,
+      "grad_norm": 0.052758537232875824,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 37530
+    },
+    {
+      "epoch": 3.8566584463625153,
+      "grad_norm": 0.06971180438995361,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 37533
+    },
+    {
+      "epoch": 3.8569667077681875,
+      "grad_norm": 0.051832135766744614,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 37536
+    },
+    {
+      "epoch": 3.8572749691738597,
+      "grad_norm": 0.11838405579328537,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 37539
+    },
+    {
+      "epoch": 3.8575832305795315,
+      "grad_norm": 0.0657329186797142,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 37542
+    },
+    {
+      "epoch": 3.857891491985203,
+      "grad_norm": 0.042532552033662796,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 37545
+    },
+    {
+      "epoch": 3.8581997533908754,
+      "grad_norm": 0.029605695977807045,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 37548
+    },
+    {
+      "epoch": 3.8585080147965476,
+      "grad_norm": 0.045208025723695755,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 37551
+    },
+    {
+      "epoch": 3.8588162762022193,
+      "grad_norm": 0.061939138919115067,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 37554
+    },
+    {
+      "epoch": 3.8591245376078915,
+      "grad_norm": 0.08816343545913696,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 37557
+    },
+    {
+      "epoch": 3.8594327990135637,
+      "grad_norm": 0.06432296335697174,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 37560
+    },
+    {
+      "epoch": 3.8597410604192355,
+      "grad_norm": 0.07691509276628494,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 37563
+    },
+    {
+      "epoch": 3.8600493218249077,
+      "grad_norm": 0.06328088790178299,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 37566
+    },
+    {
+      "epoch": 3.8603575832305794,
+      "grad_norm": 0.10897444188594818,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 37569
+    },
+    {
+      "epoch": 3.8606658446362516,
+      "grad_norm": 0.07461336255073547,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 37572
+    },
+    {
+      "epoch": 3.8609741060419234,
+      "grad_norm": 0.08543375879526138,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 37575
+    },
+    {
+      "epoch": 3.8612823674475956,
+      "grad_norm": 0.09546118974685669,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 37578
+    },
+    {
+      "epoch": 3.8615906288532678,
+      "grad_norm": 0.0741833969950676,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 37581
+    },
+    {
+      "epoch": 3.8618988902589395,
+      "grad_norm": 0.06818868219852448,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 37584
+    },
+    {
+      "epoch": 3.8622071516646117,
+      "grad_norm": 0.05515025556087494,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 37587
+    },
+    {
+      "epoch": 3.8625154130702835,
+      "grad_norm": 0.035819582641124725,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 37590
+    },
+    {
+      "epoch": 3.8628236744759556,
+      "grad_norm": 0.03145124390721321,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 37593
+    },
+    {
+      "epoch": 3.8631319358816274,
+      "grad_norm": 0.036893073469400406,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 37596
+    },
+    {
+      "epoch": 3.8634401972872996,
+      "grad_norm": 0.03417917340993881,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 37599
+    },
+    {
+      "epoch": 3.863748458692972,
+      "grad_norm": 0.06035599485039711,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 37602
+    },
+    {
+      "epoch": 3.8640567200986435,
+      "grad_norm": 0.12924471497535706,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 37605
+    },
+    {
+      "epoch": 3.8643649815043157,
+      "grad_norm": 0.1035354882478714,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 37608
+    },
+    {
+      "epoch": 3.864673242909988,
+      "grad_norm": 0.08066008985042572,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 37611
+    },
+    {
+      "epoch": 3.8649815043156597,
+      "grad_norm": 0.09418363124132156,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 37614
+    },
+    {
+      "epoch": 3.8652897657213314,
+      "grad_norm": 0.04805073142051697,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 37617
+    },
+    {
+      "epoch": 3.8655980271270036,
+      "grad_norm": 0.04753410443663597,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 37620
+    },
+    {
+      "epoch": 3.865906288532676,
+      "grad_norm": 0.0458386205136776,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 37623
+    },
+    {
+      "epoch": 3.8662145499383476,
+      "grad_norm": 0.1318642795085907,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 37626
+    },
+    {
+      "epoch": 3.8665228113440198,
+      "grad_norm": 0.10072299838066101,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 37629
+    },
+    {
+      "epoch": 3.866831072749692,
+      "grad_norm": 0.06701991707086563,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 37632
+    },
+    {
+      "epoch": 3.8671393341553637,
+      "grad_norm": 0.055372897535562515,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 37635
+    },
+    {
+      "epoch": 3.867447595561036,
+      "grad_norm": 0.061906322836875916,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 37638
+    },
+    {
+      "epoch": 3.8677558569667077,
+      "grad_norm": 0.05329981446266174,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 37641
+    },
+    {
+      "epoch": 3.86806411837238,
+      "grad_norm": 0.05068975314497948,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 37644
+    },
+    {
+      "epoch": 3.8683723797780516,
+      "grad_norm": 0.09255755692720413,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 37647
+    },
+    {
+      "epoch": 3.868680641183724,
+      "grad_norm": 0.08617328852415085,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 37650
+    },
+    {
+      "epoch": 3.868988902589396,
+      "grad_norm": 0.044413745403289795,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 37653
+    },
+    {
+      "epoch": 3.8692971639950677,
+      "grad_norm": 0.086729995906353,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 37656
+    },
+    {
+      "epoch": 3.86960542540074,
+      "grad_norm": 0.07304412871599197,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 37659
+    },
+    {
+      "epoch": 3.8699136868064117,
+      "grad_norm": 0.06820013374090195,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 37662
+    },
+    {
+      "epoch": 3.870221948212084,
+      "grad_norm": 0.06025752052664757,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 37665
+    },
+    {
+      "epoch": 3.8705302096177556,
+      "grad_norm": 0.0649590715765953,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 37668
+    },
+    {
+      "epoch": 3.870838471023428,
+      "grad_norm": 0.03553159162402153,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 37671
+    },
+    {
+      "epoch": 3.8711467324291,
+      "grad_norm": 0.06017104908823967,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 37674
+    },
+    {
+      "epoch": 3.8714549938347718,
+      "grad_norm": 0.05896780639886856,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 37677
+    },
+    {
+      "epoch": 3.871763255240444,
+      "grad_norm": 0.03493823856115341,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 37680
+    },
+    {
+      "epoch": 3.872071516646116,
+      "grad_norm": 0.11454503238201141,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 37683
+    },
+    {
+      "epoch": 3.872379778051788,
+      "grad_norm": 0.03603978455066681,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 37686
+    },
+    {
+      "epoch": 3.8726880394574597,
+      "grad_norm": 0.03927883505821228,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 37689
+    },
+    {
+      "epoch": 3.872996300863132,
+      "grad_norm": 0.06592239439487457,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 37692
+    },
+    {
+      "epoch": 3.873304562268804,
+      "grad_norm": 0.04402392357587814,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 37695
+    },
+    {
+      "epoch": 3.873612823674476,
+      "grad_norm": 0.04484686255455017,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 37698
+    },
+    {
+      "epoch": 3.873921085080148,
+      "grad_norm": 0.03577352687716484,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 37701
+    },
+    {
+      "epoch": 3.87422934648582,
+      "grad_norm": 0.03952937200665474,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 37704
+    },
+    {
+      "epoch": 3.874537607891492,
+      "grad_norm": 0.048469673842191696,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 37707
+    },
+    {
+      "epoch": 3.874845869297164,
+      "grad_norm": 0.04531377553939819,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 37710
+    },
+    {
+      "epoch": 3.875154130702836,
+      "grad_norm": 0.11962947994470596,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 37713
+    },
+    {
+      "epoch": 3.875462392108508,
+      "grad_norm": 0.03379445523023605,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 37716
+    },
+    {
+      "epoch": 3.87577065351418,
+      "grad_norm": 0.036722879856824875,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 37719
+    },
+    {
+      "epoch": 3.876078914919852,
+      "grad_norm": 0.03215174376964569,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 37722
+    },
+    {
+      "epoch": 3.876387176325524,
+      "grad_norm": 0.04591568931937218,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 37725
+    },
+    {
+      "epoch": 3.876695437731196,
+      "grad_norm": 0.04655212163925171,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 37728
+    },
+    {
+      "epoch": 3.877003699136868,
+      "grad_norm": 0.053555794060230255,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 37731
+    },
+    {
+      "epoch": 3.8773119605425403,
+      "grad_norm": 0.05492212250828743,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 37734
+    },
+    {
+      "epoch": 3.877620221948212,
+      "grad_norm": 0.07766193896532059,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 37737
+    },
+    {
+      "epoch": 3.877928483353884,
+      "grad_norm": 0.040087949484586716,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 37740
+    },
+    {
+      "epoch": 3.878236744759556,
+      "grad_norm": 0.07583244144916534,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 37743
+    },
+    {
+      "epoch": 3.8785450061652282,
+      "grad_norm": 0.0815076008439064,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 37746
+    },
+    {
+      "epoch": 3.8788532675709,
+      "grad_norm": 0.09940320998430252,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 37749
+    },
+    {
+      "epoch": 3.879161528976572,
+      "grad_norm": 0.07148890197277069,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 37752
+    },
+    {
+      "epoch": 3.8794697903822444,
+      "grad_norm": 0.0479053258895874,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 37755
+    },
+    {
+      "epoch": 3.879778051787916,
+      "grad_norm": 0.0394553542137146,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 37758
+    },
+    {
+      "epoch": 3.8800863131935883,
+      "grad_norm": 0.04193533584475517,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 37761
+    },
+    {
+      "epoch": 3.88039457459926,
+      "grad_norm": 0.09164579212665558,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 37764
+    },
+    {
+      "epoch": 3.8807028360049323,
+      "grad_norm": 0.06852073222398758,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 37767
+    },
+    {
+      "epoch": 3.881011097410604,
+      "grad_norm": 0.049102578312158585,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 37770
+    },
+    {
+      "epoch": 3.881319358816276,
+      "grad_norm": 0.12974314391613007,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 37773
+    },
+    {
+      "epoch": 3.8816276202219484,
+      "grad_norm": 0.07722889631986618,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 37776
+    },
+    {
+      "epoch": 3.88193588162762,
+      "grad_norm": 0.07344157248735428,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 37779
+    },
+    {
+      "epoch": 3.8822441430332923,
+      "grad_norm": 0.035780053585767746,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 37782
+    },
+    {
+      "epoch": 3.882552404438964,
+      "grad_norm": 0.04180228337645531,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 37785
+    },
+    {
+      "epoch": 3.8828606658446363,
+      "grad_norm": 0.06659300625324249,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 37788
+    },
+    {
+      "epoch": 3.883168927250308,
+      "grad_norm": 0.08275042474269867,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 37791
+    },
+    {
+      "epoch": 3.8834771886559802,
+      "grad_norm": 0.12271406501531601,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 37794
+    },
+    {
+      "epoch": 3.8837854500616524,
+      "grad_norm": 0.14413659274578094,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 37797
+    },
+    {
+      "epoch": 3.884093711467324,
+      "grad_norm": 0.0712043046951294,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 37800
+    },
+    {
+      "epoch": 3.8844019728729964,
+      "grad_norm": 0.06497600674629211,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 37803
+    },
+    {
+      "epoch": 3.8847102342786686,
+      "grad_norm": 0.053064875304698944,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 37806
+    },
+    {
+      "epoch": 3.8850184956843403,
+      "grad_norm": 0.04865000769495964,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 37809
+    },
+    {
+      "epoch": 3.885326757090012,
+      "grad_norm": 0.03862098976969719,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 37812
+    },
+    {
+      "epoch": 3.8856350184956843,
+      "grad_norm": 0.07821226119995117,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 37815
+    },
+    {
+      "epoch": 3.8859432799013565,
+      "grad_norm": 0.04477280378341675,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 37818
+    },
+    {
+      "epoch": 3.886251541307028,
+      "grad_norm": 0.11052905023097992,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 37821
+    },
+    {
+      "epoch": 3.8865598027127004,
+      "grad_norm": 0.07528085261583328,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 37824
+    },
+    {
+      "epoch": 3.8868680641183726,
+      "grad_norm": 0.09329832345247269,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 37827
+    },
+    {
+      "epoch": 3.8871763255240444,
+      "grad_norm": 0.03858843818306923,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 37830
+    },
+    {
+      "epoch": 3.8874845869297165,
+      "grad_norm": 0.05364726856350899,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 37833
+    },
+    {
+      "epoch": 3.8877928483353883,
+      "grad_norm": 0.036494240164756775,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 37836
+    },
+    {
+      "epoch": 3.8881011097410605,
+      "grad_norm": 0.033809784799814224,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 37839
+    },
+    {
+      "epoch": 3.8884093711467322,
+      "grad_norm": 0.08358091861009598,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 37842
+    },
+    {
+      "epoch": 3.8887176325524044,
+      "grad_norm": 0.04314170405268669,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 37845
+    },
+    {
+      "epoch": 3.8890258939580766,
+      "grad_norm": 0.06448940187692642,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 37848
+    },
+    {
+      "epoch": 3.8893341553637484,
+      "grad_norm": 0.04236144572496414,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 37851
+    },
+    {
+      "epoch": 3.8896424167694206,
+      "grad_norm": 0.08682555705308914,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 37854
+    },
+    {
+      "epoch": 3.8899506781750923,
+      "grad_norm": 0.1599910706281662,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 37857
+    },
+    {
+      "epoch": 3.8902589395807645,
+      "grad_norm": 0.04853571951389313,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 37860
+    },
+    {
+      "epoch": 3.8905672009864363,
+      "grad_norm": 0.054008036851882935,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 37863
+    },
+    {
+      "epoch": 3.8908754623921085,
+      "grad_norm": 0.0459190271794796,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 37866
+    },
+    {
+      "epoch": 3.8911837237977807,
+      "grad_norm": 0.047940943390131,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 37869
+    },
+    {
+      "epoch": 3.8914919852034524,
+      "grad_norm": 0.041892241686582565,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 37872
+    },
+    {
+      "epoch": 3.8918002466091246,
+      "grad_norm": 0.037861257791519165,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 37875
+    },
+    {
+      "epoch": 3.892108508014797,
+      "grad_norm": 0.1314014196395874,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 37878
+    },
+    {
+      "epoch": 3.8924167694204685,
+      "grad_norm": 0.053526077419519424,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 37881
+    },
+    {
+      "epoch": 3.8927250308261403,
+      "grad_norm": 0.044471751898527145,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 37884
+    },
+    {
+      "epoch": 3.8930332922318125,
+      "grad_norm": 0.10089153051376343,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 37887
+    },
+    {
+      "epoch": 3.8933415536374847,
+      "grad_norm": 0.09028996527194977,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 37890
+    },
+    {
+      "epoch": 3.8936498150431564,
+      "grad_norm": 0.06257162988185883,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 37893
+    },
+    {
+      "epoch": 3.8939580764488286,
+      "grad_norm": 0.056669414043426514,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 37896
+    },
+    {
+      "epoch": 3.894266337854501,
+      "grad_norm": 0.09135915338993073,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 37899
+    },
+    {
+      "epoch": 3.8945745992601726,
+      "grad_norm": 0.055741071701049805,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 37902
+    },
+    {
+      "epoch": 3.8948828606658448,
+      "grad_norm": 0.06624908000230789,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 37905
+    },
+    {
+      "epoch": 3.8951911220715165,
+      "grad_norm": 0.030004529282450676,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 37908
+    },
+    {
+      "epoch": 3.8954993834771887,
+      "grad_norm": 0.04271325841546059,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 37911
+    },
+    {
+      "epoch": 3.8958076448828605,
+      "grad_norm": 0.04701056331396103,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 37914
+    },
+    {
+      "epoch": 3.8961159062885327,
+      "grad_norm": 0.04298432916402817,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 37917
+    },
+    {
+      "epoch": 3.896424167694205,
+      "grad_norm": 0.04561863839626312,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 37920
+    },
+    {
+      "epoch": 3.8967324290998766,
+      "grad_norm": 0.12965039908885956,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 37923
+    },
+    {
+      "epoch": 3.897040690505549,
+      "grad_norm": 0.1109326183795929,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 37926
+    },
+    {
+      "epoch": 3.8973489519112205,
+      "grad_norm": 0.06224660202860832,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 37929
+    },
+    {
+      "epoch": 3.8976572133168927,
+      "grad_norm": 0.06318973749876022,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 37932
+    },
+    {
+      "epoch": 3.8979654747225645,
+      "grad_norm": 0.06799352914094925,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 37935
+    },
+    {
+      "epoch": 3.8982737361282367,
+      "grad_norm": 0.07071252167224884,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 37938
+    },
+    {
+      "epoch": 3.898581997533909,
+      "grad_norm": 0.03879907727241516,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 37941
+    },
+    {
+      "epoch": 3.8988902589395806,
+      "grad_norm": 0.048606619238853455,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 37944
+    },
+    {
+      "epoch": 3.899198520345253,
+      "grad_norm": 0.09814203530550003,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 37947
+    },
+    {
+      "epoch": 3.899506781750925,
+      "grad_norm": 0.04557656869292259,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 37950
+    },
+    {
+      "epoch": 3.8998150431565968,
+      "grad_norm": 0.04122527688741684,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 37953
+    },
+    {
+      "epoch": 3.900123304562269,
+      "grad_norm": 0.060118671506643295,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 37956
+    },
+    {
+      "epoch": 3.9004315659679407,
+      "grad_norm": 0.14660358428955078,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 37959
+    },
+    {
+      "epoch": 3.900739827373613,
+      "grad_norm": 0.09819690883159637,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 37962
+    },
+    {
+      "epoch": 3.9010480887792847,
+      "grad_norm": 0.041503068059682846,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 37965
+    },
+    {
+      "epoch": 3.901356350184957,
+      "grad_norm": 0.04021459445357323,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 37968
+    },
+    {
+      "epoch": 3.901664611590629,
+      "grad_norm": 0.04789562150835991,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 37971
+    },
+    {
+      "epoch": 3.901972872996301,
+      "grad_norm": 0.03949809446930885,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 37974
+    },
+    {
+      "epoch": 3.902281134401973,
+      "grad_norm": 0.041829485446214676,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 37977
+    },
+    {
+      "epoch": 3.9025893958076447,
+      "grad_norm": 0.13577204942703247,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 37980
+    },
+    {
+      "epoch": 3.902897657213317,
+      "grad_norm": 0.03821146488189697,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 37983
+    },
+    {
+      "epoch": 3.9032059186189887,
+      "grad_norm": 0.07382892072200775,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 37986
+    },
+    {
+      "epoch": 3.903514180024661,
+      "grad_norm": 0.04287475720047951,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 37989
+    },
+    {
+      "epoch": 3.903822441430333,
+      "grad_norm": 0.08588143438100815,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 37992
+    },
+    {
+      "epoch": 3.904130702836005,
+      "grad_norm": 0.060327284038066864,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 37995
+    },
+    {
+      "epoch": 3.904438964241677,
+      "grad_norm": 0.050726667046546936,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 37998
+    },
+    {
+      "epoch": 3.904747225647349,
+      "grad_norm": 0.0876336544752121,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 38001
+    },
+    {
+      "epoch": 3.905055487053021,
+      "grad_norm": 0.05690234154462814,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 38004
+    },
+    {
+      "epoch": 3.9053637484586927,
+      "grad_norm": 0.07977151870727539,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 38007
+    },
+    {
+      "epoch": 3.905672009864365,
+      "grad_norm": 0.038005661219358444,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 38010
+    },
+    {
+      "epoch": 3.905980271270037,
+      "grad_norm": 0.1033119261264801,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 38013
+    },
+    {
+      "epoch": 3.906288532675709,
+      "grad_norm": 0.05024554207921028,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 38016
+    },
+    {
+      "epoch": 3.906596794081381,
+      "grad_norm": 0.07625278830528259,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 38019
+    },
+    {
+      "epoch": 3.9069050554870532,
+      "grad_norm": 0.10337291657924652,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 38022
+    },
+    {
+      "epoch": 3.907213316892725,
+      "grad_norm": 0.04415413364768028,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 38025
+    },
+    {
+      "epoch": 3.907521578298397,
+      "grad_norm": 0.0347541943192482,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 38028
+    },
+    {
+      "epoch": 3.907829839704069,
+      "grad_norm": 0.036209288984537125,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 38031
+    },
+    {
+      "epoch": 3.908138101109741,
+      "grad_norm": 0.0959850400686264,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 38034
+    },
+    {
+      "epoch": 3.908446362515413,
+      "grad_norm": 0.05654274299740791,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 38037
+    },
+    {
+      "epoch": 3.908754623921085,
+      "grad_norm": 0.11333990842103958,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 38040
+    },
+    {
+      "epoch": 3.9090628853267573,
+      "grad_norm": 0.05329067260026932,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 38043
+    },
+    {
+      "epoch": 3.909371146732429,
+      "grad_norm": 0.03489474579691887,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 38046
+    },
+    {
+      "epoch": 3.909679408138101,
+      "grad_norm": 0.044765155762434006,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 38049
+    },
+    {
+      "epoch": 3.909987669543773,
+      "grad_norm": 0.11689729988574982,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 38052
+    },
+    {
+      "epoch": 3.910295930949445,
+      "grad_norm": 0.04785189777612686,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 38055
+    },
+    {
+      "epoch": 3.910604192355117,
+      "grad_norm": 0.040157750248909,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 38058
+    },
+    {
+      "epoch": 3.910912453760789,
+      "grad_norm": 0.04107039049267769,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 38061
+    },
+    {
+      "epoch": 3.9112207151664613,
+      "grad_norm": 0.1298598051071167,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 38064
+    },
+    {
+      "epoch": 3.911528976572133,
+      "grad_norm": 0.04832058027386665,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 38067
+    },
+    {
+      "epoch": 3.9118372379778052,
+      "grad_norm": 0.035566527396440506,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 38070
+    },
+    {
+      "epoch": 3.9121454993834774,
+      "grad_norm": 0.12645745277404785,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 38073
+    },
+    {
+      "epoch": 3.912453760789149,
+      "grad_norm": 0.13044588267803192,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 38076
+    },
+    {
+      "epoch": 3.912762022194821,
+      "grad_norm": 0.0751858800649643,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 38079
+    },
+    {
+      "epoch": 3.913070283600493,
+      "grad_norm": 0.06417164206504822,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 38082
+    },
+    {
+      "epoch": 3.9133785450061653,
+      "grad_norm": 0.03895511105656624,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 38085
+    },
+    {
+      "epoch": 3.913686806411837,
+      "grad_norm": 0.03581606224179268,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 38088
+    },
+    {
+      "epoch": 3.9139950678175093,
+      "grad_norm": 0.02936650812625885,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 38091
+    },
+    {
+      "epoch": 3.9143033292231815,
+      "grad_norm": 0.036661721765995026,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 38094
+    },
+    {
+      "epoch": 3.914611590628853,
+      "grad_norm": 0.08295129984617233,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 38097
+    },
+    {
+      "epoch": 3.9149198520345254,
+      "grad_norm": 0.09887045621871948,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 38100
+    },
+    {
+      "epoch": 3.915228113440197,
+      "grad_norm": 0.0821511521935463,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 38103
+    },
+    {
+      "epoch": 3.9155363748458694,
+      "grad_norm": 0.0894683301448822,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 38106
+    },
+    {
+      "epoch": 3.915844636251541,
+      "grad_norm": 0.043715961277484894,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 38109
+    },
+    {
+      "epoch": 3.9161528976572133,
+      "grad_norm": 0.06341227144002914,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 38112
+    },
+    {
+      "epoch": 3.9164611590628855,
+      "grad_norm": 0.10574996471405029,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 38115
+    },
+    {
+      "epoch": 3.9167694204685573,
+      "grad_norm": 0.07099676132202148,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 38118
+    },
+    {
+      "epoch": 3.9170776818742294,
+      "grad_norm": 0.051802802830934525,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 38121
+    },
+    {
+      "epoch": 3.917385943279901,
+      "grad_norm": 0.059143371880054474,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 38124
+    },
+    {
+      "epoch": 3.9176942046855734,
+      "grad_norm": 0.07478468120098114,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 38127
+    },
+    {
+      "epoch": 3.918002466091245,
+      "grad_norm": 0.05269934609532356,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 38130
+    },
+    {
+      "epoch": 3.9183107274969173,
+      "grad_norm": 0.03774020075798035,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 38133
+    },
+    {
+      "epoch": 3.9186189889025895,
+      "grad_norm": 0.06620101630687714,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 38136
+    },
+    {
+      "epoch": 3.9189272503082613,
+      "grad_norm": 0.13302361965179443,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 38139
+    },
+    {
+      "epoch": 3.9192355117139335,
+      "grad_norm": 0.05435695871710777,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 38142
+    },
+    {
+      "epoch": 3.9195437731196057,
+      "grad_norm": 0.041471030563116074,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 38145
+    },
+    {
+      "epoch": 3.9198520345252774,
+      "grad_norm": 0.032130394130945206,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 38148
+    },
+    {
+      "epoch": 3.920160295930949,
+      "grad_norm": 0.032780006527900696,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 38151
+    },
+    {
+      "epoch": 3.9204685573366214,
+      "grad_norm": 0.03699677437543869,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 38154
+    },
+    {
+      "epoch": 3.9207768187422936,
+      "grad_norm": 0.04473430663347244,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 38157
+    },
+    {
+      "epoch": 3.9210850801479653,
+      "grad_norm": 0.06131662428379059,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 38160
+    },
+    {
+      "epoch": 3.9213933415536375,
+      "grad_norm": 0.10085607320070267,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 38163
+    },
+    {
+      "epoch": 3.9217016029593097,
+      "grad_norm": 0.11734067648649216,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 38166
+    },
+    {
+      "epoch": 3.9220098643649814,
+      "grad_norm": 0.09287893027067184,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 38169
+    },
+    {
+      "epoch": 3.9223181257706536,
+      "grad_norm": 0.058617495000362396,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 38172
+    },
+    {
+      "epoch": 3.9226263871763254,
+      "grad_norm": 0.0597362145781517,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 38175
+    },
+    {
+      "epoch": 3.9229346485819976,
+      "grad_norm": 0.046750448644161224,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 38178
+    },
+    {
+      "epoch": 3.9232429099876693,
+      "grad_norm": 0.03863513097167015,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 38181
+    },
+    {
+      "epoch": 3.9235511713933415,
+      "grad_norm": 0.03276417404413223,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 38184
+    },
+    {
+      "epoch": 3.9238594327990137,
+      "grad_norm": 0.03270519897341728,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 38187
+    },
+    {
+      "epoch": 3.9241676942046855,
+      "grad_norm": 0.05820414423942566,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 38190
+    },
+    {
+      "epoch": 3.9244759556103577,
+      "grad_norm": 0.0589318610727787,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 38193
+    },
+    {
+      "epoch": 3.92478421701603,
+      "grad_norm": 0.05013816058635712,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 38196
+    },
+    {
+      "epoch": 3.9250924784217016,
+      "grad_norm": 0.10934565216302872,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 38199
+    },
+    {
+      "epoch": 3.9254007398273734,
+      "grad_norm": 0.08114335685968399,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 38202
+    },
+    {
+      "epoch": 3.9257090012330456,
+      "grad_norm": 0.056287068873643875,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 38205
+    },
+    {
+      "epoch": 3.9260172626387178,
+      "grad_norm": 0.04509197920560837,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 38208
+    },
+    {
+      "epoch": 3.9263255240443895,
+      "grad_norm": 0.04409080743789673,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 38211
+    },
+    {
+      "epoch": 3.9266337854500617,
+      "grad_norm": 0.044174257665872574,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 38214
+    },
+    {
+      "epoch": 3.926942046855734,
+      "grad_norm": 0.05122366175055504,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 38217
+    },
+    {
+      "epoch": 3.9272503082614056,
+      "grad_norm": 0.12582433223724365,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 38220
+    },
+    {
+      "epoch": 3.927558569667078,
+      "grad_norm": 0.060516782104969025,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 38223
+    },
+    {
+      "epoch": 3.9278668310727496,
+      "grad_norm": 0.04210484027862549,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 38226
+    },
+    {
+      "epoch": 3.928175092478422,
+      "grad_norm": 0.043509434908628464,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 38229
+    },
+    {
+      "epoch": 3.9284833538840935,
+      "grad_norm": 0.03601152077317238,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 38232
+    },
+    {
+      "epoch": 3.9287916152897657,
+      "grad_norm": 0.04590906575322151,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 38235
+    },
+    {
+      "epoch": 3.929099876695438,
+      "grad_norm": 0.06336984038352966,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 38238
+    },
+    {
+      "epoch": 3.9294081381011097,
+      "grad_norm": 0.07014600187540054,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 38241
+    },
+    {
+      "epoch": 3.929716399506782,
+      "grad_norm": 0.07760234922170639,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 38244
+    },
+    {
+      "epoch": 3.9300246609124536,
+      "grad_norm": 0.1088213101029396,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 38247
+    },
+    {
+      "epoch": 3.930332922318126,
+      "grad_norm": 0.04306039586663246,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 38250
+    },
+    {
+      "epoch": 3.9306411837237976,
+      "grad_norm": 0.057269759476184845,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 38253
+    },
+    {
+      "epoch": 3.9309494451294698,
+      "grad_norm": 0.04819104075431824,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 38256
+    },
+    {
+      "epoch": 3.931257706535142,
+      "grad_norm": 0.038784999400377274,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 38259
+    },
+    {
+      "epoch": 3.9315659679408137,
+      "grad_norm": 0.08599941432476044,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 38262
+    },
+    {
+      "epoch": 3.931874229346486,
+      "grad_norm": 0.10680285841226578,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 38265
+    },
+    {
+      "epoch": 3.932182490752158,
+      "grad_norm": 0.09683788567781448,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 38268
+    },
+    {
+      "epoch": 3.93249075215783,
+      "grad_norm": 0.040799640119075775,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 38271
+    },
+    {
+      "epoch": 3.9327990135635016,
+      "grad_norm": 0.08265798538923264,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 38274
+    },
+    {
+      "epoch": 3.933107274969174,
+      "grad_norm": 0.0628572627902031,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 38277
+    },
+    {
+      "epoch": 3.933415536374846,
+      "grad_norm": 0.06475923955440521,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 38280
+    },
+    {
+      "epoch": 3.9337237977805177,
+      "grad_norm": 0.043740466237068176,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 38283
+    },
+    {
+      "epoch": 3.93403205918619,
+      "grad_norm": 0.029720323160290718,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 38286
+    },
+    {
+      "epoch": 3.934340320591862,
+      "grad_norm": 0.050941772758960724,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 38289
+    },
+    {
+      "epoch": 3.934648581997534,
+      "grad_norm": 0.1068761870265007,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 38292
+    },
+    {
+      "epoch": 3.934956843403206,
+      "grad_norm": 0.06410616636276245,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 38295
+    },
+    {
+      "epoch": 3.935265104808878,
+      "grad_norm": 0.10598262399435043,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 38298
+    },
+    {
+      "epoch": 3.93557336621455,
+      "grad_norm": 0.05296116694808006,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 38301
+    },
+    {
+      "epoch": 3.9358816276202218,
+      "grad_norm": 0.04025321081280708,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 38304
+    },
+    {
+      "epoch": 3.936189889025894,
+      "grad_norm": 0.11896882951259613,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 38307
+    },
+    {
+      "epoch": 3.936498150431566,
+      "grad_norm": 0.07203146070241928,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 38310
+    },
+    {
+      "epoch": 3.936806411837238,
+      "grad_norm": 0.05813155323266983,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 38313
+    },
+    {
+      "epoch": 3.93711467324291,
+      "grad_norm": 0.0439312644302845,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 38316
+    },
+    {
+      "epoch": 3.937422934648582,
+      "grad_norm": 0.05411090329289436,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 38319
+    },
+    {
+      "epoch": 3.937731196054254,
+      "grad_norm": 0.035287659615278244,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 38322
+    },
+    {
+      "epoch": 3.938039457459926,
+      "grad_norm": 0.08188273012638092,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 38325
+    },
+    {
+      "epoch": 3.938347718865598,
+      "grad_norm": 0.06879545003175735,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 38328
+    },
+    {
+      "epoch": 3.93865598027127,
+      "grad_norm": 0.06965189427137375,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 38331
+    },
+    {
+      "epoch": 3.938964241676942,
+      "grad_norm": 0.07698633521795273,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 38334
+    },
+    {
+      "epoch": 3.939272503082614,
+      "grad_norm": 0.06989213079214096,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 38337
+    },
+    {
+      "epoch": 3.9395807644882863,
+      "grad_norm": 0.09224730730056763,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 38340
+    },
+    {
+      "epoch": 3.939889025893958,
+      "grad_norm": 0.15504223108291626,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 38343
+    },
+    {
+      "epoch": 3.94019728729963,
+      "grad_norm": 0.06935428082942963,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 38346
+    },
+    {
+      "epoch": 3.940505548705302,
+      "grad_norm": 0.051906321197748184,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 38349
+    },
+    {
+      "epoch": 3.940813810110974,
+      "grad_norm": 0.059517163783311844,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 38352
+    },
+    {
+      "epoch": 3.941122071516646,
+      "grad_norm": 0.031905319541692734,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 38355
+    },
+    {
+      "epoch": 3.941430332922318,
+      "grad_norm": 0.06473187357187271,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 38358
+    },
+    {
+      "epoch": 3.9417385943279903,
+      "grad_norm": 0.0756862536072731,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 38361
+    },
+    {
+      "epoch": 3.942046855733662,
+      "grad_norm": 0.06513907015323639,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 38364
+    },
+    {
+      "epoch": 3.9423551171393343,
+      "grad_norm": 0.0908990353345871,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 38367
+    },
+    {
+      "epoch": 3.942663378545006,
+      "grad_norm": 0.06427323073148727,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 38370
+    },
+    {
+      "epoch": 3.9429716399506782,
+      "grad_norm": 0.09153769910335541,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 38373
+    },
+    {
+      "epoch": 3.94327990135635,
+      "grad_norm": 0.0451352633535862,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 38376
+    },
+    {
+      "epoch": 3.943588162762022,
+      "grad_norm": 0.06844813376665115,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 38379
+    },
+    {
+      "epoch": 3.9438964241676944,
+      "grad_norm": 0.07901204377412796,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 38382
+    },
+    {
+      "epoch": 3.944204685573366,
+      "grad_norm": 0.08094186335802078,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 38385
+    },
+    {
+      "epoch": 3.9445129469790383,
+      "grad_norm": 0.09053739905357361,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 38388
+    },
+    {
+      "epoch": 3.9448212083847105,
+      "grad_norm": 0.04751008749008179,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 38391
+    },
+    {
+      "epoch": 3.9451294697903823,
+      "grad_norm": 0.042345285415649414,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 38394
+    },
+    {
+      "epoch": 3.945437731196054,
+      "grad_norm": 0.1027415469288826,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 38397
+    },
+    {
+      "epoch": 3.945745992601726,
+      "grad_norm": 0.12780915200710297,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 38400
+    },
+    {
+      "epoch": 3.9460542540073984,
+      "grad_norm": 0.0731390118598938,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 38403
+    },
+    {
+      "epoch": 3.94636251541307,
+      "grad_norm": 0.07617928087711334,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 38406
+    },
+    {
+      "epoch": 3.9466707768187423,
+      "grad_norm": 0.049755457788705826,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 38409
+    },
+    {
+      "epoch": 3.9469790382244145,
+      "grad_norm": 0.043812233954668045,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 38412
+    },
+    {
+      "epoch": 3.9472872996300863,
+      "grad_norm": 0.07829032093286514,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 38415
+    },
+    {
+      "epoch": 3.9475955610357585,
+      "grad_norm": 0.056485142558813095,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 38418
+    },
+    {
+      "epoch": 3.9479038224414302,
+      "grad_norm": 0.09569665789604187,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 38421
+    },
+    {
+      "epoch": 3.9482120838471024,
+      "grad_norm": 0.07466506212949753,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 38424
+    },
+    {
+      "epoch": 3.948520345252774,
+      "grad_norm": 0.08420269191265106,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 38427
+    },
+    {
+      "epoch": 3.9488286066584464,
+      "grad_norm": 0.05188721418380737,
+      "learning_rate": 0.01,
+      "loss": 2.0409,
+      "step": 38430
+    },
+    {
+      "epoch": 3.9491368680641186,
+      "grad_norm": 0.042890697717666626,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 38433
+    },
+    {
+      "epoch": 3.9494451294697903,
+      "grad_norm": 0.05752531811594963,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 38436
+    },
+    {
+      "epoch": 3.9497533908754625,
+      "grad_norm": 0.0403323695063591,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 38439
+    },
+    {
+      "epoch": 3.9500616522811343,
+      "grad_norm": 0.035122547298669815,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 38442
+    },
+    {
+      "epoch": 3.9503699136868065,
+      "grad_norm": 0.04063395410776138,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 38445
+    },
+    {
+      "epoch": 3.950678175092478,
+      "grad_norm": 0.05300283804535866,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 38448
+    },
+    {
+      "epoch": 3.9509864364981504,
+      "grad_norm": 0.06941550225019455,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 38451
+    },
+    {
+      "epoch": 3.9512946979038226,
+      "grad_norm": 0.04268152639269829,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 38454
+    },
+    {
+      "epoch": 3.9516029593094943,
+      "grad_norm": 0.058871544897556305,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 38457
+    },
+    {
+      "epoch": 3.9519112207151665,
+      "grad_norm": 0.0643804594874382,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 38460
+    },
+    {
+      "epoch": 3.9522194821208387,
+      "grad_norm": 0.04509326070547104,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 38463
+    },
+    {
+      "epoch": 3.9525277435265105,
+      "grad_norm": 0.04596889764070511,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 38466
+    },
+    {
+      "epoch": 3.9528360049321822,
+      "grad_norm": 0.08639726787805557,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 38469
+    },
+    {
+      "epoch": 3.9531442663378544,
+      "grad_norm": 0.08404930680990219,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 38472
+    },
+    {
+      "epoch": 3.9534525277435266,
+      "grad_norm": 0.09206783026456833,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 38475
+    },
+    {
+      "epoch": 3.9537607891491984,
+      "grad_norm": 0.041316352784633636,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 38478
+    },
+    {
+      "epoch": 3.9540690505548706,
+      "grad_norm": 0.07000883668661118,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 38481
+    },
+    {
+      "epoch": 3.9543773119605428,
+      "grad_norm": 0.048716410994529724,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 38484
+    },
+    {
+      "epoch": 3.9546855733662145,
+      "grad_norm": 0.05251453444361687,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 38487
+    },
+    {
+      "epoch": 3.9549938347718867,
+      "grad_norm": 0.04517770931124687,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 38490
+    },
+    {
+      "epoch": 3.9553020961775585,
+      "grad_norm": 0.07227790355682373,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 38493
+    },
+    {
+      "epoch": 3.9556103575832307,
+      "grad_norm": 0.09953956305980682,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 38496
+    },
+    {
+      "epoch": 3.9559186189889024,
+      "grad_norm": 0.06194831430912018,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 38499
+    },
+    {
+      "epoch": 3.9562268803945746,
+      "grad_norm": 0.10566883534193039,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 38502
+    },
+    {
+      "epoch": 3.956535141800247,
+      "grad_norm": 0.04950516298413277,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 38505
+    },
+    {
+      "epoch": 3.9568434032059185,
+      "grad_norm": 0.04657790809869766,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 38508
+    },
+    {
+      "epoch": 3.9571516646115907,
+      "grad_norm": 0.05642826855182648,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 38511
+    },
+    {
+      "epoch": 3.9574599260172625,
+      "grad_norm": 0.049003373831510544,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 38514
+    },
+    {
+      "epoch": 3.9577681874229347,
+      "grad_norm": 0.05019281804561615,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 38517
+    },
+    {
+      "epoch": 3.9580764488286064,
+      "grad_norm": 0.05060233548283577,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 38520
+    },
+    {
+      "epoch": 3.9583847102342786,
+      "grad_norm": 0.051873739808797836,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 38523
+    },
+    {
+      "epoch": 3.958692971639951,
+      "grad_norm": 0.11860267072916031,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 38526
+    },
+    {
+      "epoch": 3.9590012330456226,
+      "grad_norm": 0.03894282132387161,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 38529
+    },
+    {
+      "epoch": 3.9593094944512948,
+      "grad_norm": 0.0839400663971901,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 38532
+    },
+    {
+      "epoch": 3.959617755856967,
+      "grad_norm": 0.10270547866821289,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 38535
+    },
+    {
+      "epoch": 3.9599260172626387,
+      "grad_norm": 0.07309349626302719,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 38538
+    },
+    {
+      "epoch": 3.9602342786683105,
+      "grad_norm": 0.03441464155912399,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 38541
+    },
+    {
+      "epoch": 3.9605425400739827,
+      "grad_norm": 0.04434891417622566,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 38544
+    },
+    {
+      "epoch": 3.960850801479655,
+      "grad_norm": 0.030151626095175743,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 38547
+    },
+    {
+      "epoch": 3.9611590628853266,
+      "grad_norm": 0.03249426558613777,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 38550
+    },
+    {
+      "epoch": 3.961467324290999,
+      "grad_norm": 0.06774075329303741,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 38553
+    },
+    {
+      "epoch": 3.961775585696671,
+      "grad_norm": 0.1510474681854248,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 38556
+    },
+    {
+      "epoch": 3.9620838471023427,
+      "grad_norm": 0.080832839012146,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 38559
+    },
+    {
+      "epoch": 3.962392108508015,
+      "grad_norm": 0.08914501219987869,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 38562
+    },
+    {
+      "epoch": 3.9627003699136867,
+      "grad_norm": 0.10509749501943588,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 38565
+    },
+    {
+      "epoch": 3.963008631319359,
+      "grad_norm": 0.0980307012796402,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 38568
+    },
+    {
+      "epoch": 3.9633168927250306,
+      "grad_norm": 0.04917595908045769,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 38571
+    },
+    {
+      "epoch": 3.963625154130703,
+      "grad_norm": 0.04090237617492676,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 38574
+    },
+    {
+      "epoch": 3.963933415536375,
+      "grad_norm": 0.06782566010951996,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 38577
+    },
+    {
+      "epoch": 3.9642416769420468,
+      "grad_norm": 0.04495246335864067,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 38580
+    },
+    {
+      "epoch": 3.964549938347719,
+      "grad_norm": 0.11113790422677994,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 38583
+    },
+    {
+      "epoch": 3.9648581997533907,
+      "grad_norm": 0.09196839481592178,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 38586
+    },
+    {
+      "epoch": 3.965166461159063,
+      "grad_norm": 0.07392636686563492,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 38589
+    },
+    {
+      "epoch": 3.9654747225647347,
+      "grad_norm": 0.0521097257733345,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 38592
+    },
+    {
+      "epoch": 3.965782983970407,
+      "grad_norm": 0.05052189901471138,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 38595
+    },
+    {
+      "epoch": 3.966091245376079,
+      "grad_norm": 0.08592119067907333,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 38598
+    },
+    {
+      "epoch": 3.966399506781751,
+      "grad_norm": 0.08535821735858917,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 38601
+    },
+    {
+      "epoch": 3.966707768187423,
+      "grad_norm": 0.06658685207366943,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 38604
+    },
+    {
+      "epoch": 3.967016029593095,
+      "grad_norm": 0.04116528108716011,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 38607
+    },
+    {
+      "epoch": 3.967324290998767,
+      "grad_norm": 0.04124008119106293,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 38610
+    },
+    {
+      "epoch": 3.967632552404439,
+      "grad_norm": 0.047569263726472855,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 38613
+    },
+    {
+      "epoch": 3.967940813810111,
+      "grad_norm": 0.04800506308674812,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 38616
+    },
+    {
+      "epoch": 3.968249075215783,
+      "grad_norm": 0.06227673590183258,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 38619
+    },
+    {
+      "epoch": 3.968557336621455,
+      "grad_norm": 0.036228880286216736,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 38622
+    },
+    {
+      "epoch": 3.968865598027127,
+      "grad_norm": 0.03745630383491516,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 38625
+    },
+    {
+      "epoch": 3.969173859432799,
+      "grad_norm": 0.08907367289066315,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 38628
+    },
+    {
+      "epoch": 3.969482120838471,
+      "grad_norm": 0.054506488144397736,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 38631
+    },
+    {
+      "epoch": 3.969790382244143,
+      "grad_norm": 0.04185614362359047,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 38634
+    },
+    {
+      "epoch": 3.970098643649815,
+      "grad_norm": 0.04917627200484276,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 38637
+    },
+    {
+      "epoch": 3.970406905055487,
+      "grad_norm": 0.040258195251226425,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 38640
+    },
+    {
+      "epoch": 3.970715166461159,
+      "grad_norm": 0.08127589523792267,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 38643
+    },
+    {
+      "epoch": 3.971023427866831,
+      "grad_norm": 0.07298692315816879,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 38646
+    },
+    {
+      "epoch": 3.9713316892725032,
+      "grad_norm": 0.09801699966192245,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 38649
+    },
+    {
+      "epoch": 3.971639950678175,
+      "grad_norm": 0.10016069561243057,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 38652
+    },
+    {
+      "epoch": 3.971948212083847,
+      "grad_norm": 0.038826216012239456,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 38655
+    },
+    {
+      "epoch": 3.9722564734895194,
+      "grad_norm": 0.04124876856803894,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 38658
+    },
+    {
+      "epoch": 3.972564734895191,
+      "grad_norm": 0.03699329122900963,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 38661
+    },
+    {
+      "epoch": 3.972872996300863,
+      "grad_norm": 0.03506563603878021,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 38664
+    },
+    {
+      "epoch": 3.973181257706535,
+      "grad_norm": 0.08274342864751816,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 38667
+    },
+    {
+      "epoch": 3.9734895191122073,
+      "grad_norm": 0.10225984454154968,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 38670
+    },
+    {
+      "epoch": 3.973797780517879,
+      "grad_norm": 0.08367688208818436,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 38673
+    },
+    {
+      "epoch": 3.974106041923551,
+      "grad_norm": 0.05682690069079399,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 38676
+    },
+    {
+      "epoch": 3.9744143033292234,
+      "grad_norm": 0.03980601951479912,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 38679
+    },
+    {
+      "epoch": 3.974722564734895,
+      "grad_norm": 0.03310983628034592,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 38682
+    },
+    {
+      "epoch": 3.9750308261405674,
+      "grad_norm": 0.05037367716431618,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 38685
+    },
+    {
+      "epoch": 3.975339087546239,
+      "grad_norm": 0.09068721532821655,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 38688
+    },
+    {
+      "epoch": 3.9756473489519113,
+      "grad_norm": 0.10251244902610779,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 38691
+    },
+    {
+      "epoch": 3.975955610357583,
+      "grad_norm": 0.04860818758606911,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 38694
+    },
+    {
+      "epoch": 3.9762638717632552,
+      "grad_norm": 0.0460125096142292,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 38697
+    },
+    {
+      "epoch": 3.9765721331689274,
+      "grad_norm": 0.03295229375362396,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 38700
+    },
+    {
+      "epoch": 3.976880394574599,
+      "grad_norm": 0.0434846356511116,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 38703
+    },
+    {
+      "epoch": 3.9771886559802714,
+      "grad_norm": 0.04077495262026787,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 38706
+    },
+    {
+      "epoch": 3.977496917385943,
+      "grad_norm": 0.08263571560382843,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 38709
+    },
+    {
+      "epoch": 3.9778051787916153,
+      "grad_norm": 0.1033141016960144,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 38712
+    },
+    {
+      "epoch": 3.978113440197287,
+      "grad_norm": 0.12253855168819427,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 38715
+    },
+    {
+      "epoch": 3.9784217016029593,
+      "grad_norm": 0.08476614207029343,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 38718
+    },
+    {
+      "epoch": 3.9787299630086315,
+      "grad_norm": 0.08248502016067505,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 38721
+    },
+    {
+      "epoch": 3.979038224414303,
+      "grad_norm": 0.06338318437337875,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 38724
+    },
+    {
+      "epoch": 3.9793464858199754,
+      "grad_norm": 0.061125461012125015,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 38727
+    },
+    {
+      "epoch": 3.9796547472256476,
+      "grad_norm": 0.04191330447793007,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 38730
+    },
+    {
+      "epoch": 3.9799630086313194,
+      "grad_norm": 0.04181262478232384,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 38733
+    },
+    {
+      "epoch": 3.980271270036991,
+      "grad_norm": 0.054846856743097305,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 38736
+    },
+    {
+      "epoch": 3.9805795314426633,
+      "grad_norm": 0.1322845071554184,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 38739
+    },
+    {
+      "epoch": 3.9808877928483355,
+      "grad_norm": 0.06795060634613037,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 38742
+    },
+    {
+      "epoch": 3.9811960542540072,
+      "grad_norm": 0.04729272425174713,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 38745
+    },
+    {
+      "epoch": 3.9815043156596794,
+      "grad_norm": 0.05951160192489624,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 38748
+    },
+    {
+      "epoch": 3.9818125770653516,
+      "grad_norm": 0.14003396034240723,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 38751
+    },
+    {
+      "epoch": 3.9821208384710234,
+      "grad_norm": 0.04996626079082489,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 38754
+    },
+    {
+      "epoch": 3.9824290998766956,
+      "grad_norm": 0.08134348690509796,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 38757
+    },
+    {
+      "epoch": 3.9827373612823673,
+      "grad_norm": 0.04592986777424812,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 38760
+    },
+    {
+      "epoch": 3.9830456226880395,
+      "grad_norm": 0.06769111007452011,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 38763
+    },
+    {
+      "epoch": 3.9833538840937113,
+      "grad_norm": 0.04501201957464218,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 38766
+    },
+    {
+      "epoch": 3.9836621454993835,
+      "grad_norm": 0.07144643366336823,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 38769
+    },
+    {
+      "epoch": 3.9839704069050557,
+      "grad_norm": 0.05670906975865364,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 38772
+    },
+    {
+      "epoch": 3.9842786683107274,
+      "grad_norm": 0.03870624676346779,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 38775
+    },
+    {
+      "epoch": 3.9845869297163996,
+      "grad_norm": 0.08053532242774963,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 38778
+    },
+    {
+      "epoch": 3.9848951911220714,
+      "grad_norm": 0.03753774240612984,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 38781
+    },
+    {
+      "epoch": 3.9852034525277436,
+      "grad_norm": 0.04568708315491676,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 38784
+    },
+    {
+      "epoch": 3.9855117139334153,
+      "grad_norm": 0.03626095503568649,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 38787
+    },
+    {
+      "epoch": 3.9858199753390875,
+      "grad_norm": 0.04620308801531792,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 38790
+    },
+    {
+      "epoch": 3.9861282367447597,
+      "grad_norm": 0.05767418071627617,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 38793
+    },
+    {
+      "epoch": 3.9864364981504314,
+      "grad_norm": 0.04969481751322746,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 38796
+    },
+    {
+      "epoch": 3.9867447595561036,
+      "grad_norm": 0.0532878078520298,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 38799
+    },
+    {
+      "epoch": 3.987053020961776,
+      "grad_norm": 0.07379619777202606,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 38802
+    },
+    {
+      "epoch": 3.9873612823674476,
+      "grad_norm": 0.07538430392742157,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 38805
+    },
+    {
+      "epoch": 3.9876695437731193,
+      "grad_norm": 0.0686052069067955,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 38808
+    },
+    {
+      "epoch": 3.9879778051787915,
+      "grad_norm": 0.08726216852664948,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 38811
+    },
+    {
+      "epoch": 3.9882860665844637,
+      "grad_norm": 0.08535965532064438,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 38814
+    },
+    {
+      "epoch": 3.9885943279901355,
+      "grad_norm": 0.046852223575115204,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 38817
+    },
+    {
+      "epoch": 3.9889025893958077,
+      "grad_norm": 0.06200144812464714,
+      "learning_rate": 0.01,
+      "loss": 2.0423,
+      "step": 38820
+    },
+    {
+      "epoch": 3.98921085080148,
+      "grad_norm": 0.03675130382180214,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 38823
+    },
+    {
+      "epoch": 3.9895191122071516,
+      "grad_norm": 0.054221536964178085,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 38826
+    },
+    {
+      "epoch": 3.989827373612824,
+      "grad_norm": 0.0411151684820652,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 38829
+    },
+    {
+      "epoch": 3.9901356350184956,
+      "grad_norm": 0.03962259367108345,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 38832
+    },
+    {
+      "epoch": 3.9904438964241677,
+      "grad_norm": 0.04097359627485275,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 38835
+    },
+    {
+      "epoch": 3.9907521578298395,
+      "grad_norm": 0.13092494010925293,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 38838
+    },
+    {
+      "epoch": 3.9910604192355117,
+      "grad_norm": 0.03308350592851639,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 38841
+    },
+    {
+      "epoch": 3.991368680641184,
+      "grad_norm": 0.06447092443704605,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 38844
+    },
+    {
+      "epoch": 3.9916769420468556,
+      "grad_norm": 0.06456363946199417,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 38847
+    },
+    {
+      "epoch": 3.991985203452528,
+      "grad_norm": 0.04208895191550255,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 38850
+    },
+    {
+      "epoch": 3.9922934648582,
+      "grad_norm": 0.03307265415787697,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 38853
+    },
+    {
+      "epoch": 3.9926017262638718,
+      "grad_norm": 0.06227843463420868,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 38856
+    },
+    {
+      "epoch": 3.9929099876695435,
+      "grad_norm": 0.1568191796541214,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 38859
+    },
+    {
+      "epoch": 3.9932182490752157,
+      "grad_norm": 0.05974143370985985,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 38862
+    },
+    {
+      "epoch": 3.993526510480888,
+      "grad_norm": 0.07882939279079437,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 38865
+    },
+    {
+      "epoch": 3.9938347718865597,
+      "grad_norm": 0.053341448307037354,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 38868
+    },
+    {
+      "epoch": 3.994143033292232,
+      "grad_norm": 0.0673101395368576,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 38871
+    },
+    {
+      "epoch": 3.994451294697904,
+      "grad_norm": 0.06340447813272476,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 38874
+    },
+    {
+      "epoch": 3.994759556103576,
+      "grad_norm": 0.06743529438972473,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 38877
+    },
+    {
+      "epoch": 3.995067817509248,
+      "grad_norm": 0.044702883809804916,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 38880
+    },
+    {
+      "epoch": 3.9953760789149197,
+      "grad_norm": 0.038102682679891586,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 38883
+    },
+    {
+      "epoch": 3.995684340320592,
+      "grad_norm": 0.0471016988158226,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 38886
+    },
+    {
+      "epoch": 3.9959926017262637,
+      "grad_norm": 0.10289987921714783,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 38889
+    },
+    {
+      "epoch": 3.996300863131936,
+      "grad_norm": 0.08012085407972336,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 38892
+    },
+    {
+      "epoch": 3.996609124537608,
+      "grad_norm": 0.07357537001371384,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 38895
+    },
+    {
+      "epoch": 3.99691738594328,
+      "grad_norm": 0.08909714221954346,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 38898
+    },
+    {
+      "epoch": 3.997225647348952,
+      "grad_norm": 0.044727593660354614,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 38901
+    },
+    {
+      "epoch": 3.9975339087546238,
+      "grad_norm": 0.039593808352947235,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 38904
+    },
+    {
+      "epoch": 3.997842170160296,
+      "grad_norm": 0.03478686884045601,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 38907
+    },
+    {
+      "epoch": 3.9981504315659677,
+      "grad_norm": 0.09729648381471634,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 38910
+    },
+    {
+      "epoch": 3.99845869297164,
+      "grad_norm": 0.12499833852052689,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 38913
+    },
+    {
+      "epoch": 3.998766954377312,
+      "grad_norm": 0.10060276091098785,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 38916
+    },
+    {
+      "epoch": 3.999075215782984,
+      "grad_norm": 0.060861632227897644,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 38919
+    },
+    {
+      "epoch": 3.999383477188656,
+      "grad_norm": 0.058074526488780975,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 38922
+    },
+    {
+      "epoch": 3.9996917385943282,
+      "grad_norm": 0.08413925021886826,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 38925
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.06637567281723022,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 38928
+    },
+    {
+      "epoch": 3.9937423061140747,
+      "grad_norm": 0.04972624033689499,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 38931
+    },
+    {
+      "epoch": 3.9940500615510874,
+      "grad_norm": 0.04883033037185669,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 38934
+    },
+    {
+      "epoch": 3.9943578169881,
+      "grad_norm": 0.03551949933171272,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 38937
+    },
+    {
+      "epoch": 3.994665572425113,
+      "grad_norm": 0.03538177162408829,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 38940
+    },
+    {
+      "epoch": 3.994973327862126,
+      "grad_norm": 0.13709349930286407,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 38943
+    },
+    {
+      "epoch": 3.9952810832991386,
+      "grad_norm": 0.05646511912345886,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 38946
+    },
+    {
+      "epoch": 3.995588838736151,
+      "grad_norm": 0.05141003429889679,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 38949
+    },
+    {
+      "epoch": 3.9958965941731637,
+      "grad_norm": 0.1064891591668129,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 38952
+    },
+    {
+      "epoch": 3.9962043496101765,
+      "grad_norm": 0.058718711137771606,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 38955
+    },
+    {
+      "epoch": 3.9965121050471892,
+      "grad_norm": 0.05866394191980362,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 38958
+    },
+    {
+      "epoch": 3.996819860484202,
+      "grad_norm": 0.036083538085222244,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 38961
+    },
+    {
+      "epoch": 3.9971276159212143,
+      "grad_norm": 0.05041573569178581,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 38964
+    },
+    {
+      "epoch": 3.997435371358227,
+      "grad_norm": 0.05070210620760918,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 38967
+    },
+    {
+      "epoch": 3.99774312679524,
+      "grad_norm": 0.06114819273352623,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 38970
+    },
+    {
+      "epoch": 3.9980508822322527,
+      "grad_norm": 0.05506499856710434,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 38973
+    },
+    {
+      "epoch": 3.9983586376692655,
+      "grad_norm": 0.06224251538515091,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 38976
+    },
+    {
+      "epoch": 3.9986663931062782,
+      "grad_norm": 0.05054394155740738,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 38979
+    },
+    {
+      "epoch": 3.998974148543291,
+      "grad_norm": 0.03862088546156883,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 38982
+    },
+    {
+      "epoch": 3.999281903980304,
+      "grad_norm": 0.04840533435344696,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 38985
+    },
+    {
+      "epoch": 3.9995896594173166,
+      "grad_norm": 0.05625883862376213,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 38988
+    },
+    {
+      "epoch": 3.9998974148543294,
+      "grad_norm": 0.04202372580766678,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 38991
+    },
+    {
+      "epoch": 4.000205170291342,
+      "grad_norm": 0.03988848999142647,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 38994
+    },
+    {
+      "epoch": 4.000512925728355,
+      "grad_norm": 0.08701229840517044,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 38997
+    },
+    {
+      "epoch": 4.000820681165367,
+      "grad_norm": 0.13237404823303223,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 39000
+    },
+    {
+      "epoch": 4.00112843660238,
+      "grad_norm": 0.0758156031370163,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 39003
+    },
+    {
+      "epoch": 4.001436192039392,
+      "grad_norm": 0.07759084552526474,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 39006
+    },
+    {
+      "epoch": 4.001743947476405,
+      "grad_norm": 0.08195871859788895,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 39009
+    },
+    {
+      "epoch": 4.002051702913418,
+      "grad_norm": 0.048866480588912964,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 39012
+    },
+    {
+      "epoch": 4.002359458350431,
+      "grad_norm": 0.04807426407933235,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 39015
+    },
+    {
+      "epoch": 4.0026672137874435,
+      "grad_norm": 0.07215370982885361,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 39018
+    },
+    {
+      "epoch": 4.002974969224456,
+      "grad_norm": 0.05422010272741318,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 39021
+    },
+    {
+      "epoch": 4.003282724661469,
+      "grad_norm": 0.06720750778913498,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 39024
+    },
+    {
+      "epoch": 4.003590480098482,
+      "grad_norm": 0.04077715426683426,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 39027
+    },
+    {
+      "epoch": 4.003898235535495,
+      "grad_norm": 0.05866453796625137,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 39030
+    },
+    {
+      "epoch": 4.004205990972507,
+      "grad_norm": 0.03861930966377258,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 39033
+    },
+    {
+      "epoch": 4.00451374640952,
+      "grad_norm": 0.03546518459916115,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 39036
+    },
+    {
+      "epoch": 4.004821501846533,
+      "grad_norm": 0.034643933176994324,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 39039
+    },
+    {
+      "epoch": 4.005129257283546,
+      "grad_norm": 0.0831715315580368,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 39042
+    },
+    {
+      "epoch": 4.005437012720558,
+      "grad_norm": 0.06925207376480103,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 39045
+    },
+    {
+      "epoch": 4.00574476815757,
+      "grad_norm": 0.08311621099710464,
+      "learning_rate": 0.01,
+      "loss": 2.0402,
+      "step": 39048
+    },
+    {
+      "epoch": 4.006052523594583,
+      "grad_norm": 0.08028697222471237,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 39051
+    },
+    {
+      "epoch": 4.006360279031596,
+      "grad_norm": 0.1259499341249466,
+      "learning_rate": 0.01,
+      "loss": 2.028,
+      "step": 39054
+    },
+    {
+      "epoch": 4.006668034468609,
+      "grad_norm": 0.06442322582006454,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 39057
+    },
+    {
+      "epoch": 4.0069757899056215,
+      "grad_norm": 0.08281309902667999,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 39060
+    },
+    {
+      "epoch": 4.007283545342634,
+      "grad_norm": 0.04970608651638031,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 39063
+    },
+    {
+      "epoch": 4.007591300779647,
+      "grad_norm": 0.04361598566174507,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 39066
+    },
+    {
+      "epoch": 4.00789905621666,
+      "grad_norm": 0.04321054369211197,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 39069
+    },
+    {
+      "epoch": 4.008206811653673,
+      "grad_norm": 0.05904306843876839,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 39072
+    },
+    {
+      "epoch": 4.008514567090685,
+      "grad_norm": 0.0728040412068367,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 39075
+    },
+    {
+      "epoch": 4.008822322527698,
+      "grad_norm": 0.09545755386352539,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 39078
+    },
+    {
+      "epoch": 4.009130077964711,
+      "grad_norm": 0.10636181384325027,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 39081
+    },
+    {
+      "epoch": 4.009437833401724,
+      "grad_norm": 0.08768682181835175,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 39084
+    },
+    {
+      "epoch": 4.0097455888387366,
+      "grad_norm": 0.04949640482664108,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 39087
+    },
+    {
+      "epoch": 4.010053344275748,
+      "grad_norm": 0.0413932166993618,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 39090
+    },
+    {
+      "epoch": 4.010361099712761,
+      "grad_norm": 0.02756902389228344,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 39093
+    },
+    {
+      "epoch": 4.010668855149774,
+      "grad_norm": 0.08485575020313263,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 39096
+    },
+    {
+      "epoch": 4.010976610586787,
+      "grad_norm": 0.0896422490477562,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 39099
+    },
+    {
+      "epoch": 4.0112843660238,
+      "grad_norm": 0.09696487337350845,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 39102
+    },
+    {
+      "epoch": 4.011592121460812,
+      "grad_norm": 0.059906743466854095,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 39105
+    },
+    {
+      "epoch": 4.011899876897825,
+      "grad_norm": 0.12596943974494934,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 39108
+    },
+    {
+      "epoch": 4.012207632334838,
+      "grad_norm": 0.08249086886644363,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 39111
+    },
+    {
+      "epoch": 4.012515387771851,
+      "grad_norm": 0.1073683500289917,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 39114
+    },
+    {
+      "epoch": 4.0128231432088635,
+      "grad_norm": 0.050193801522254944,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 39117
+    },
+    {
+      "epoch": 4.013130898645876,
+      "grad_norm": 0.04790155962109566,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 39120
+    },
+    {
+      "epoch": 4.013438654082889,
+      "grad_norm": 0.1130373477935791,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 39123
+    },
+    {
+      "epoch": 4.013746409519902,
+      "grad_norm": 0.16575901210308075,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 39126
+    },
+    {
+      "epoch": 4.014054164956915,
+      "grad_norm": 0.03896254301071167,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 39129
+    },
+    {
+      "epoch": 4.014361920393927,
+      "grad_norm": 0.033224210143089294,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 39132
+    },
+    {
+      "epoch": 4.014669675830939,
+      "grad_norm": 0.04422049596905708,
+      "learning_rate": 0.01,
+      "loss": 2.0278,
+      "step": 39135
+    },
+    {
+      "epoch": 4.014977431267952,
+      "grad_norm": 0.046339549124240875,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 39138
+    },
+    {
+      "epoch": 4.015285186704965,
+      "grad_norm": 0.04534962773323059,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 39141
+    },
+    {
+      "epoch": 4.015592942141978,
+      "grad_norm": 0.0634172111749649,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 39144
+    },
+    {
+      "epoch": 4.01590069757899,
+      "grad_norm": 0.0435175783932209,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 39147
+    },
+    {
+      "epoch": 4.016208453016003,
+      "grad_norm": 0.04143211990594864,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 39150
+    },
+    {
+      "epoch": 4.016516208453016,
+      "grad_norm": 0.06689900159835815,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 39153
+    },
+    {
+      "epoch": 4.016823963890029,
+      "grad_norm": 0.19841893017292023,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 39156
+    },
+    {
+      "epoch": 4.0171317193270415,
+      "grad_norm": 0.09940902143716812,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 39159
+    },
+    {
+      "epoch": 4.017439474764054,
+      "grad_norm": 0.0738014280796051,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 39162
+    },
+    {
+      "epoch": 4.017747230201067,
+      "grad_norm": 0.06441953033208847,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 39165
+    },
+    {
+      "epoch": 4.01805498563808,
+      "grad_norm": 0.03884090855717659,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 39168
+    },
+    {
+      "epoch": 4.018362741075093,
+      "grad_norm": 0.036479830741882324,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 39171
+    },
+    {
+      "epoch": 4.018670496512105,
+      "grad_norm": 0.04188670963048935,
+      "learning_rate": 0.01,
+      "loss": 2.0239,
+      "step": 39174
+    },
+    {
+      "epoch": 4.018978251949118,
+      "grad_norm": 0.07043974846601486,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 39177
+    },
+    {
+      "epoch": 4.01928600738613,
+      "grad_norm": 0.07807689160108566,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 39180
+    },
+    {
+      "epoch": 4.019593762823143,
+      "grad_norm": 0.05508783459663391,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 39183
+    },
+    {
+      "epoch": 4.019901518260156,
+      "grad_norm": 0.040311504155397415,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 39186
+    },
+    {
+      "epoch": 4.020209273697168,
+      "grad_norm": 0.03907238692045212,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 39189
+    },
+    {
+      "epoch": 4.020517029134181,
+      "grad_norm": 0.1371290385723114,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 39192
+    },
+    {
+      "epoch": 4.020824784571194,
+      "grad_norm": 0.05499713122844696,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 39195
+    },
+    {
+      "epoch": 4.021132540008207,
+      "grad_norm": 0.049838222563266754,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 39198
+    },
+    {
+      "epoch": 4.0214402954452195,
+      "grad_norm": 0.04565451666712761,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 39201
+    },
+    {
+      "epoch": 4.021748050882232,
+      "grad_norm": 0.06939809769392014,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 39204
+    },
+    {
+      "epoch": 4.022055806319245,
+      "grad_norm": 0.036337222903966904,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 39207
+    },
+    {
+      "epoch": 4.022363561756258,
+      "grad_norm": 0.11887694150209427,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 39210
+    },
+    {
+      "epoch": 4.022671317193271,
+      "grad_norm": 0.05074974149465561,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 39213
+    },
+    {
+      "epoch": 4.022979072630283,
+      "grad_norm": 0.04023103788495064,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 39216
+    },
+    {
+      "epoch": 4.023286828067296,
+      "grad_norm": 0.09158769249916077,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 39219
+    },
+    {
+      "epoch": 4.023594583504309,
+      "grad_norm": 0.06678687036037445,
+      "learning_rate": 0.01,
+      "loss": 2.0338,
+      "step": 39222
+    },
+    {
+      "epoch": 4.023902338941321,
+      "grad_norm": 0.055356625467538834,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 39225
+    },
+    {
+      "epoch": 4.024210094378334,
+      "grad_norm": 0.051128923892974854,
+      "learning_rate": 0.01,
+      "loss": 2.0291,
+      "step": 39228
+    },
+    {
+      "epoch": 4.024517849815346,
+      "grad_norm": 0.08770928531885147,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 39231
+    },
+    {
+      "epoch": 4.024825605252359,
+      "grad_norm": 0.06450860947370529,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 39234
+    },
+    {
+      "epoch": 4.025133360689372,
+      "grad_norm": 0.03998043015599251,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 39237
+    },
+    {
+      "epoch": 4.025441116126385,
+      "grad_norm": 0.036666858941316605,
+      "learning_rate": 0.01,
+      "loss": 2.0451,
+      "step": 39240
+    },
+    {
+      "epoch": 4.025748871563398,
+      "grad_norm": 0.03850167244672775,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 39243
+    },
+    {
+      "epoch": 4.02605662700041,
+      "grad_norm": 0.03165037930011749,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 39246
+    },
+    {
+      "epoch": 4.026364382437423,
+      "grad_norm": 0.03471562638878822,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 39249
+    },
+    {
+      "epoch": 4.026672137874436,
+      "grad_norm": 0.10216906666755676,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 39252
+    },
+    {
+      "epoch": 4.026979893311449,
+      "grad_norm": 0.13467098772525787,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 39255
+    },
+    {
+      "epoch": 4.0272876487484615,
+      "grad_norm": 0.05322883278131485,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 39258
+    },
+    {
+      "epoch": 4.027595404185474,
+      "grad_norm": 0.0329531729221344,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 39261
+    },
+    {
+      "epoch": 4.027903159622487,
+      "grad_norm": 0.04283340275287628,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 39264
+    },
+    {
+      "epoch": 4.0282109150595,
+      "grad_norm": 0.09303436428308487,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 39267
+    },
+    {
+      "epoch": 4.028518670496512,
+      "grad_norm": 0.04981641098856926,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 39270
+    },
+    {
+      "epoch": 4.0288264259335245,
+      "grad_norm": 0.05858089402318001,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 39273
+    },
+    {
+      "epoch": 4.029134181370537,
+      "grad_norm": 0.05271727591753006,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 39276
+    },
+    {
+      "epoch": 4.02944193680755,
+      "grad_norm": 0.0625569149851799,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 39279
+    },
+    {
+      "epoch": 4.029749692244563,
+      "grad_norm": 0.06783869862556458,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 39282
+    },
+    {
+      "epoch": 4.030057447681576,
+      "grad_norm": 0.05175579711794853,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 39285
+    },
+    {
+      "epoch": 4.030365203118588,
+      "grad_norm": 0.12327279895544052,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 39288
+    },
+    {
+      "epoch": 4.030672958555601,
+      "grad_norm": 0.09904535114765167,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 39291
+    },
+    {
+      "epoch": 4.030980713992614,
+      "grad_norm": 0.06502281874418259,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 39294
+    },
+    {
+      "epoch": 4.031288469429627,
+      "grad_norm": 0.05546945706009865,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 39297
+    },
+    {
+      "epoch": 4.0315962248666395,
+      "grad_norm": 0.03637825697660446,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 39300
+    },
+    {
+      "epoch": 4.031903980303652,
+      "grad_norm": 0.03696468472480774,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 39303
+    },
+    {
+      "epoch": 4.032211735740665,
+      "grad_norm": 0.11666533350944519,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 39306
+    },
+    {
+      "epoch": 4.032519491177678,
+      "grad_norm": 0.10067463666200638,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 39309
+    },
+    {
+      "epoch": 4.032827246614691,
+      "grad_norm": 0.0598643533885479,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 39312
+    },
+    {
+      "epoch": 4.0331350020517025,
+      "grad_norm": 0.057127151638269424,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 39315
+    },
+    {
+      "epoch": 4.033442757488715,
+      "grad_norm": 0.05786604434251785,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 39318
+    },
+    {
+      "epoch": 4.033750512925728,
+      "grad_norm": 0.040237389504909515,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 39321
+    },
+    {
+      "epoch": 4.034058268362741,
+      "grad_norm": 0.09420931339263916,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 39324
+    },
+    {
+      "epoch": 4.034366023799754,
+      "grad_norm": 0.11208292096853256,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 39327
+    },
+    {
+      "epoch": 4.034673779236766,
+      "grad_norm": 0.07000657171010971,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 39330
+    },
+    {
+      "epoch": 4.034981534673779,
+      "grad_norm": 0.05524434149265289,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 39333
+    },
+    {
+      "epoch": 4.035289290110792,
+      "grad_norm": 0.03453601896762848,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 39336
+    },
+    {
+      "epoch": 4.035597045547805,
+      "grad_norm": 0.046687569469213486,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 39339
+    },
+    {
+      "epoch": 4.0359048009848175,
+      "grad_norm": 0.04349486157298088,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 39342
+    },
+    {
+      "epoch": 4.03621255642183,
+      "grad_norm": 0.08332299441099167,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 39345
+    },
+    {
+      "epoch": 4.036520311858843,
+      "grad_norm": 0.07960904389619827,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 39348
+    },
+    {
+      "epoch": 4.036828067295856,
+      "grad_norm": 0.06175240874290466,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 39351
+    },
+    {
+      "epoch": 4.037135822732869,
+      "grad_norm": 0.03728936240077019,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 39354
+    },
+    {
+      "epoch": 4.037443578169881,
+      "grad_norm": 0.046342454850673676,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 39357
+    },
+    {
+      "epoch": 4.037751333606893,
+      "grad_norm": 0.059486132115125656,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 39360
+    },
+    {
+      "epoch": 4.038059089043906,
+      "grad_norm": 0.09025575965642929,
+      "learning_rate": 0.01,
+      "loss": 2.0306,
+      "step": 39363
+    },
+    {
+      "epoch": 4.038366844480919,
+      "grad_norm": 0.0599956177175045,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 39366
+    },
+    {
+      "epoch": 4.038674599917932,
+      "grad_norm": 0.06386277079582214,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 39369
+    },
+    {
+      "epoch": 4.038982355354944,
+      "grad_norm": 0.10979870706796646,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 39372
+    },
+    {
+      "epoch": 4.039290110791957,
+      "grad_norm": 0.13358891010284424,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 39375
+    },
+    {
+      "epoch": 4.03959786622897,
+      "grad_norm": 0.07506405562162399,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 39378
+    },
+    {
+      "epoch": 4.039905621665983,
+      "grad_norm": 0.03969530388712883,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 39381
+    },
+    {
+      "epoch": 4.0402133771029956,
+      "grad_norm": 0.06700876355171204,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 39384
+    },
+    {
+      "epoch": 4.040521132540008,
+      "grad_norm": 0.06389278918504715,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 39387
+    },
+    {
+      "epoch": 4.040828887977021,
+      "grad_norm": 0.04753594473004341,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 39390
+    },
+    {
+      "epoch": 4.041136643414034,
+      "grad_norm": 0.0517578050494194,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 39393
+    },
+    {
+      "epoch": 4.041444398851047,
+      "grad_norm": 0.043198052793741226,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 39396
+    },
+    {
+      "epoch": 4.0417521542880595,
+      "grad_norm": 0.04130061715841293,
+      "learning_rate": 0.01,
+      "loss": 2.0426,
+      "step": 39399
+    },
+    {
+      "epoch": 4.042059909725072,
+      "grad_norm": 0.037592917680740356,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 39402
+    },
+    {
+      "epoch": 4.042367665162084,
+      "grad_norm": 0.07257888466119766,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 39405
+    },
+    {
+      "epoch": 4.042675420599097,
+      "grad_norm": 0.08611748367547989,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 39408
+    },
+    {
+      "epoch": 4.04298317603611,
+      "grad_norm": 0.11179140955209732,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 39411
+    },
+    {
+      "epoch": 4.0432909314731225,
+      "grad_norm": 0.09140360355377197,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 39414
+    },
+    {
+      "epoch": 4.043598686910135,
+      "grad_norm": 0.040217410773038864,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 39417
+    },
+    {
+      "epoch": 4.043906442347148,
+      "grad_norm": 0.04423901066184044,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 39420
+    },
+    {
+      "epoch": 4.044214197784161,
+      "grad_norm": 0.047180719673633575,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 39423
+    },
+    {
+      "epoch": 4.044521953221174,
+      "grad_norm": 0.03836076334118843,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 39426
+    },
+    {
+      "epoch": 4.044829708658186,
+      "grad_norm": 0.08214177936315536,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 39429
+    },
+    {
+      "epoch": 4.045137464095199,
+      "grad_norm": 0.04551496356725693,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 39432
+    },
+    {
+      "epoch": 4.045445219532212,
+      "grad_norm": 0.08608072996139526,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 39435
+    },
+    {
+      "epoch": 4.045752974969225,
+      "grad_norm": 0.03998774662613869,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 39438
+    },
+    {
+      "epoch": 4.0460607304062375,
+      "grad_norm": 0.06219779700040817,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 39441
+    },
+    {
+      "epoch": 4.04636848584325,
+      "grad_norm": 0.12329383194446564,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 39444
+    },
+    {
+      "epoch": 4.046676241280263,
+      "grad_norm": 0.10584307461977005,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 39447
+    },
+    {
+      "epoch": 4.046983996717275,
+      "grad_norm": 0.06407187879085541,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 39450
+    },
+    {
+      "epoch": 4.047291752154288,
+      "grad_norm": 0.057736970484256744,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 39453
+    },
+    {
+      "epoch": 4.0475995075913005,
+      "grad_norm": 0.062264811247587204,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 39456
+    },
+    {
+      "epoch": 4.047907263028313,
+      "grad_norm": 0.0482734851539135,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 39459
+    },
+    {
+      "epoch": 4.048215018465326,
+      "grad_norm": 0.03942275047302246,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 39462
+    },
+    {
+      "epoch": 4.048522773902339,
+      "grad_norm": 0.08987545222043991,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 39465
+    },
+    {
+      "epoch": 4.048830529339352,
+      "grad_norm": 0.08133430033922195,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 39468
+    },
+    {
+      "epoch": 4.049138284776364,
+      "grad_norm": 0.03588191419839859,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 39471
+    },
+    {
+      "epoch": 4.049446040213377,
+      "grad_norm": 0.058000437915325165,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 39474
+    },
+    {
+      "epoch": 4.04975379565039,
+      "grad_norm": 0.08807062357664108,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 39477
+    },
+    {
+      "epoch": 4.050061551087403,
+      "grad_norm": 0.05153276026248932,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 39480
+    },
+    {
+      "epoch": 4.0503693065244155,
+      "grad_norm": 0.08411730825901031,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 39483
+    },
+    {
+      "epoch": 4.050677061961428,
+      "grad_norm": 0.09271445870399475,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 39486
+    },
+    {
+      "epoch": 4.050984817398441,
+      "grad_norm": 0.07173626124858856,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 39489
+    },
+    {
+      "epoch": 4.051292572835454,
+      "grad_norm": 0.058220986276865005,
+      "learning_rate": 0.01,
+      "loss": 2.0296,
+      "step": 39492
+    },
+    {
+      "epoch": 4.051600328272466,
+      "grad_norm": 0.09059619903564453,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 39495
+    },
+    {
+      "epoch": 4.0519080837094785,
+      "grad_norm": 0.11159291118383408,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 39498
+    },
+    {
+      "epoch": 4.052215839146491,
+      "grad_norm": 0.03904344514012337,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 39501
+    },
+    {
+      "epoch": 4.052523594583504,
+      "grad_norm": 0.12682397663593292,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 39504
+    },
+    {
+      "epoch": 4.052831350020517,
+      "grad_norm": 0.06277670711278915,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 39507
+    },
+    {
+      "epoch": 4.05313910545753,
+      "grad_norm": 0.03622843325138092,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 39510
+    },
+    {
+      "epoch": 4.053446860894542,
+      "grad_norm": 0.056741926819086075,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 39513
+    },
+    {
+      "epoch": 4.053754616331555,
+      "grad_norm": 0.0654342994093895,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 39516
+    },
+    {
+      "epoch": 4.054062371768568,
+      "grad_norm": 0.05300883948802948,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 39519
+    },
+    {
+      "epoch": 4.054370127205581,
+      "grad_norm": 0.1103016659617424,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 39522
+    },
+    {
+      "epoch": 4.0546778826425935,
+      "grad_norm": 0.10315825045108795,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 39525
+    },
+    {
+      "epoch": 4.054985638079606,
+      "grad_norm": 0.11007767170667648,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 39528
+    },
+    {
+      "epoch": 4.055293393516619,
+      "grad_norm": 0.053403954952955246,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 39531
+    },
+    {
+      "epoch": 4.055601148953632,
+      "grad_norm": 0.05486089363694191,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 39534
+    },
+    {
+      "epoch": 4.055908904390645,
+      "grad_norm": 0.05680430680513382,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 39537
+    },
+    {
+      "epoch": 4.056216659827657,
+      "grad_norm": 0.07202226668596268,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 39540
+    },
+    {
+      "epoch": 4.056524415264669,
+      "grad_norm": 0.050710614770650864,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 39543
+    },
+    {
+      "epoch": 4.056832170701682,
+      "grad_norm": 0.05018226429820061,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 39546
+    },
+    {
+      "epoch": 4.057139926138695,
+      "grad_norm": 0.04539674147963524,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 39549
+    },
+    {
+      "epoch": 4.057447681575708,
+      "grad_norm": 0.0405915230512619,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 39552
+    },
+    {
+      "epoch": 4.0577554370127205,
+      "grad_norm": 0.04220306873321533,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 39555
+    },
+    {
+      "epoch": 4.058063192449733,
+      "grad_norm": 0.06802038848400116,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 39558
+    },
+    {
+      "epoch": 4.058370947886746,
+      "grad_norm": 0.16860038042068481,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 39561
+    },
+    {
+      "epoch": 4.058678703323759,
+      "grad_norm": 0.13039590418338776,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 39564
+    },
+    {
+      "epoch": 4.058986458760772,
+      "grad_norm": 0.05722019821405411,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 39567
+    },
+    {
+      "epoch": 4.059294214197784,
+      "grad_norm": 0.05984622612595558,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 39570
+    },
+    {
+      "epoch": 4.059601969634797,
+      "grad_norm": 0.04732242226600647,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 39573
+    },
+    {
+      "epoch": 4.05990972507181,
+      "grad_norm": 0.04328930750489235,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 39576
+    },
+    {
+      "epoch": 4.060217480508823,
+      "grad_norm": 0.06630559265613556,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 39579
+    },
+    {
+      "epoch": 4.0605252359458355,
+      "grad_norm": 0.06271739304065704,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 39582
+    },
+    {
+      "epoch": 4.060832991382847,
+      "grad_norm": 0.05598871782422066,
+      "learning_rate": 0.01,
+      "loss": 2.0322,
+      "step": 39585
+    },
+    {
+      "epoch": 4.06114074681986,
+      "grad_norm": 0.04648594930768013,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 39588
+    },
+    {
+      "epoch": 4.061448502256873,
+      "grad_norm": 0.06579215824604034,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 39591
+    },
+    {
+      "epoch": 4.061756257693886,
+      "grad_norm": 0.05206866189837456,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 39594
+    },
+    {
+      "epoch": 4.0620640131308985,
+      "grad_norm": 0.06146420165896416,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 39597
+    },
+    {
+      "epoch": 4.062371768567911,
+      "grad_norm": 0.09550274908542633,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 39600
+    },
+    {
+      "epoch": 4.062679524004924,
+      "grad_norm": 0.041216351091861725,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 39603
+    },
+    {
+      "epoch": 4.062987279441937,
+      "grad_norm": 0.050457730889320374,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 39606
+    },
+    {
+      "epoch": 4.06329503487895,
+      "grad_norm": 0.03148980066180229,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 39609
+    },
+    {
+      "epoch": 4.063602790315962,
+      "grad_norm": 0.03415573388338089,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 39612
+    },
+    {
+      "epoch": 4.063910545752975,
+      "grad_norm": 0.038664404302835464,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 39615
+    },
+    {
+      "epoch": 4.064218301189988,
+      "grad_norm": 0.10088005661964417,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 39618
+    },
+    {
+      "epoch": 4.064526056627001,
+      "grad_norm": 0.05986418575048447,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 39621
+    },
+    {
+      "epoch": 4.0648338120640135,
+      "grad_norm": 0.04586298391222954,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 39624
+    },
+    {
+      "epoch": 4.065141567501026,
+      "grad_norm": 0.09720548987388611,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 39627
+    },
+    {
+      "epoch": 4.065449322938038,
+      "grad_norm": 0.1310441493988037,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 39630
+    },
+    {
+      "epoch": 4.065757078375051,
+      "grad_norm": 0.0445309616625309,
+      "learning_rate": 0.01,
+      "loss": 2.0295,
+      "step": 39633
+    },
+    {
+      "epoch": 4.066064833812064,
+      "grad_norm": 0.05492424964904785,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 39636
+    },
+    {
+      "epoch": 4.0663725892490765,
+      "grad_norm": 0.09137557446956635,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 39639
+    },
+    {
+      "epoch": 4.066680344686089,
+      "grad_norm": 0.07394784688949585,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 39642
+    },
+    {
+      "epoch": 4.066988100123102,
+      "grad_norm": 0.0890335887670517,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 39645
+    },
+    {
+      "epoch": 4.067295855560115,
+      "grad_norm": 0.06005888432264328,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 39648
+    },
+    {
+      "epoch": 4.067603610997128,
+      "grad_norm": 0.08164118975400925,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 39651
+    },
+    {
+      "epoch": 4.06791136643414,
+      "grad_norm": 0.041462332010269165,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 39654
+    },
+    {
+      "epoch": 4.068219121871153,
+      "grad_norm": 0.14200016856193542,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 39657
+    },
+    {
+      "epoch": 4.068526877308166,
+      "grad_norm": 0.04974464327096939,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 39660
+    },
+    {
+      "epoch": 4.068834632745179,
+      "grad_norm": 0.05781686678528786,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 39663
+    },
+    {
+      "epoch": 4.0691423881821915,
+      "grad_norm": 0.050336726009845734,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 39666
+    },
+    {
+      "epoch": 4.069450143619204,
+      "grad_norm": 0.0692603662610054,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 39669
+    },
+    {
+      "epoch": 4.069757899056217,
+      "grad_norm": 0.052262064069509506,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 39672
+    },
+    {
+      "epoch": 4.070065654493229,
+      "grad_norm": 0.0743674710392952,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 39675
+    },
+    {
+      "epoch": 4.070373409930242,
+      "grad_norm": 0.03609304502606392,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 39678
+    },
+    {
+      "epoch": 4.0706811653672546,
+      "grad_norm": 0.05369256064295769,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 39681
+    },
+    {
+      "epoch": 4.070988920804267,
+      "grad_norm": 0.04112347960472107,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 39684
+    },
+    {
+      "epoch": 4.07129667624128,
+      "grad_norm": 0.05583459138870239,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 39687
+    },
+    {
+      "epoch": 4.071604431678293,
+      "grad_norm": 0.06682361662387848,
+      "learning_rate": 0.01,
+      "loss": 2.0344,
+      "step": 39690
+    },
+    {
+      "epoch": 4.071912187115306,
+      "grad_norm": 0.08554589748382568,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 39693
+    },
+    {
+      "epoch": 4.0722199425523185,
+      "grad_norm": 0.07087216526269913,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 39696
+    },
+    {
+      "epoch": 4.072527697989331,
+      "grad_norm": 0.06562238186597824,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 39699
+    },
+    {
+      "epoch": 4.072835453426344,
+      "grad_norm": 0.036247964948415756,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 39702
+    },
+    {
+      "epoch": 4.073143208863357,
+      "grad_norm": 0.08515971153974533,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 39705
+    },
+    {
+      "epoch": 4.07345096430037,
+      "grad_norm": 0.07769300043582916,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 39708
+    },
+    {
+      "epoch": 4.073758719737382,
+      "grad_norm": 0.0872216522693634,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 39711
+    },
+    {
+      "epoch": 4.074066475174395,
+      "grad_norm": 0.06590529531240463,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 39714
+    },
+    {
+      "epoch": 4.074374230611408,
+      "grad_norm": 0.059110917150974274,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 39717
+    },
+    {
+      "epoch": 4.07468198604842,
+      "grad_norm": 0.04299427196383476,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 39720
+    },
+    {
+      "epoch": 4.074989741485433,
+      "grad_norm": 0.10769655555486679,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 39723
+    },
+    {
+      "epoch": 4.075297496922445,
+      "grad_norm": 0.08587364107370377,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 39726
+    },
+    {
+      "epoch": 4.075605252359458,
+      "grad_norm": 0.08979383856058121,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 39729
+    },
+    {
+      "epoch": 4.075913007796471,
+      "grad_norm": 0.12115131318569183,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 39732
+    },
+    {
+      "epoch": 4.076220763233484,
+      "grad_norm": 0.1939251571893692,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 39735
+    },
+    {
+      "epoch": 4.0765285186704965,
+      "grad_norm": 0.05781332403421402,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 39738
+    },
+    {
+      "epoch": 4.076836274107509,
+      "grad_norm": 0.03456675633788109,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 39741
+    },
+    {
+      "epoch": 4.077144029544522,
+      "grad_norm": 0.040457479655742645,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 39744
+    },
+    {
+      "epoch": 4.077451784981535,
+      "grad_norm": 0.06412683427333832,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 39747
+    },
+    {
+      "epoch": 4.077759540418548,
+      "grad_norm": 0.05500893294811249,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 39750
+    },
+    {
+      "epoch": 4.07806729585556,
+      "grad_norm": 0.05995875224471092,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 39753
+    },
+    {
+      "epoch": 4.078375051292573,
+      "grad_norm": 0.1099732369184494,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 39756
+    },
+    {
+      "epoch": 4.078682806729586,
+      "grad_norm": 0.11458323895931244,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 39759
+    },
+    {
+      "epoch": 4.078990562166599,
+      "grad_norm": 0.07978501915931702,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 39762
+    },
+    {
+      "epoch": 4.079298317603611,
+      "grad_norm": 0.04269685968756676,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 39765
+    },
+    {
+      "epoch": 4.079606073040623,
+      "grad_norm": 0.051762815564870834,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 39768
+    },
+    {
+      "epoch": 4.079913828477636,
+      "grad_norm": 0.030949685722589493,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 39771
+    },
+    {
+      "epoch": 4.080221583914649,
+      "grad_norm": 0.1207941323518753,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 39774
+    },
+    {
+      "epoch": 4.080529339351662,
+      "grad_norm": 0.12475111335515976,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 39777
+    },
+    {
+      "epoch": 4.0808370947886745,
+      "grad_norm": 0.10487036406993866,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 39780
+    },
+    {
+      "epoch": 4.081144850225687,
+      "grad_norm": 0.088489830493927,
+      "learning_rate": 0.01,
+      "loss": 2.0399,
+      "step": 39783
+    },
+    {
+      "epoch": 4.0814526056627,
+      "grad_norm": 0.06538698077201843,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 39786
+    },
+    {
+      "epoch": 4.081760361099713,
+      "grad_norm": 0.05029508098959923,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 39789
+    },
+    {
+      "epoch": 4.082068116536726,
+      "grad_norm": 0.04905729368329048,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 39792
+    },
+    {
+      "epoch": 4.082375871973738,
+      "grad_norm": 0.060302965342998505,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 39795
+    },
+    {
+      "epoch": 4.082683627410751,
+      "grad_norm": 0.06065473333001137,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 39798
+    },
+    {
+      "epoch": 4.082991382847764,
+      "grad_norm": 0.04160602018237114,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 39801
+    },
+    {
+      "epoch": 4.083299138284777,
+      "grad_norm": 0.04311453923583031,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 39804
+    },
+    {
+      "epoch": 4.0836068937217895,
+      "grad_norm": 0.03628389909863472,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 39807
+    },
+    {
+      "epoch": 4.083914649158801,
+      "grad_norm": 0.09799020737409592,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 39810
+    },
+    {
+      "epoch": 4.084222404595814,
+      "grad_norm": 0.07026764005422592,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 39813
+    },
+    {
+      "epoch": 4.084530160032827,
+      "grad_norm": 0.11306377500295639,
+      "learning_rate": 0.01,
+      "loss": 2.0241,
+      "step": 39816
+    },
+    {
+      "epoch": 4.08483791546984,
+      "grad_norm": 0.13824355602264404,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 39819
+    },
+    {
+      "epoch": 4.0851456709068525,
+      "grad_norm": 0.07312388718128204,
+      "learning_rate": 0.01,
+      "loss": 2.0323,
+      "step": 39822
+    },
+    {
+      "epoch": 4.085453426343865,
+      "grad_norm": 0.06378597021102905,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 39825
+    },
+    {
+      "epoch": 4.085761181780878,
+      "grad_norm": 0.04329591616988182,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 39828
+    },
+    {
+      "epoch": 4.086068937217891,
+      "grad_norm": 0.05057251825928688,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 39831
+    },
+    {
+      "epoch": 4.086376692654904,
+      "grad_norm": 0.047054242342710495,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 39834
+    },
+    {
+      "epoch": 4.086684448091916,
+      "grad_norm": 0.06537993997335434,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 39837
+    },
+    {
+      "epoch": 4.086992203528929,
+      "grad_norm": 0.08527065068483353,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 39840
+    },
+    {
+      "epoch": 4.087299958965942,
+      "grad_norm": 0.05401263013482094,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 39843
+    },
+    {
+      "epoch": 4.087607714402955,
+      "grad_norm": 0.04258432239294052,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 39846
+    },
+    {
+      "epoch": 4.087915469839968,
+      "grad_norm": 0.08178570866584778,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 39849
+    },
+    {
+      "epoch": 4.08822322527698,
+      "grad_norm": 0.06067502498626709,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 39852
+    },
+    {
+      "epoch": 4.088530980713992,
+      "grad_norm": 0.04289801046252251,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 39855
+    },
+    {
+      "epoch": 4.088838736151005,
+      "grad_norm": 0.10091704875230789,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 39858
+    },
+    {
+      "epoch": 4.089146491588018,
+      "grad_norm": 0.05211932212114334,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 39861
+    },
+    {
+      "epoch": 4.089454247025031,
+      "grad_norm": 0.08183594793081284,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 39864
+    },
+    {
+      "epoch": 4.089762002462043,
+      "grad_norm": 0.13209478557109833,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 39867
+    },
+    {
+      "epoch": 4.090069757899056,
+      "grad_norm": 0.08384159207344055,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 39870
+    },
+    {
+      "epoch": 4.090377513336069,
+      "grad_norm": 0.10875227302312851,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 39873
+    },
+    {
+      "epoch": 4.090685268773082,
+      "grad_norm": 0.09350070357322693,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 39876
+    },
+    {
+      "epoch": 4.0909930242100945,
+      "grad_norm": 0.0466117262840271,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 39879
+    },
+    {
+      "epoch": 4.091300779647107,
+      "grad_norm": 0.09274325519800186,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 39882
+    },
+    {
+      "epoch": 4.09160853508412,
+      "grad_norm": 0.05788525938987732,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 39885
+    },
+    {
+      "epoch": 4.091916290521133,
+      "grad_norm": 0.07033047825098038,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 39888
+    },
+    {
+      "epoch": 4.092224045958146,
+      "grad_norm": 0.07227031141519547,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 39891
+    },
+    {
+      "epoch": 4.092531801395158,
+      "grad_norm": 0.08519124984741211,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 39894
+    },
+    {
+      "epoch": 4.09283955683217,
+      "grad_norm": 0.038923412561416626,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 39897
+    },
+    {
+      "epoch": 4.093147312269183,
+      "grad_norm": 0.03778720274567604,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 39900
+    },
+    {
+      "epoch": 4.093455067706196,
+      "grad_norm": 0.046422503888607025,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 39903
+    },
+    {
+      "epoch": 4.093762823143209,
+      "grad_norm": 0.13946807384490967,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 39906
+    },
+    {
+      "epoch": 4.094070578580221,
+      "grad_norm": 0.07924194633960724,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 39909
+    },
+    {
+      "epoch": 4.094378334017234,
+      "grad_norm": 0.04913933575153351,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 39912
+    },
+    {
+      "epoch": 4.094686089454247,
+      "grad_norm": 0.03534455969929695,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 39915
+    },
+    {
+      "epoch": 4.09499384489126,
+      "grad_norm": 0.03871719539165497,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 39918
+    },
+    {
+      "epoch": 4.0953016003282725,
+      "grad_norm": 0.05986125394701958,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 39921
+    },
+    {
+      "epoch": 4.095609355765285,
+      "grad_norm": 0.06470568478107452,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 39924
+    },
+    {
+      "epoch": 4.095917111202298,
+      "grad_norm": 0.05725434049963951,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 39927
+    },
+    {
+      "epoch": 4.096224866639311,
+      "grad_norm": 0.05102056264877319,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 39930
+    },
+    {
+      "epoch": 4.096532622076324,
+      "grad_norm": 0.04540586844086647,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 39933
+    },
+    {
+      "epoch": 4.096840377513336,
+      "grad_norm": 0.08318594843149185,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 39936
+    },
+    {
+      "epoch": 4.097148132950349,
+      "grad_norm": 0.0833197757601738,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 39939
+    },
+    {
+      "epoch": 4.097455888387361,
+      "grad_norm": 0.06058152765035629,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 39942
+    },
+    {
+      "epoch": 4.097763643824374,
+      "grad_norm": 0.09669670462608337,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 39945
+    },
+    {
+      "epoch": 4.098071399261387,
+      "grad_norm": 0.06633847951889038,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 39948
+    },
+    {
+      "epoch": 4.098379154698399,
+      "grad_norm": 0.1021406352519989,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 39951
+    },
+    {
+      "epoch": 4.098686910135412,
+      "grad_norm": 0.047858573496341705,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 39954
+    },
+    {
+      "epoch": 4.098994665572425,
+      "grad_norm": 0.04331503435969353,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 39957
+    },
+    {
+      "epoch": 4.099302421009438,
+      "grad_norm": 0.0399680957198143,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 39960
+    },
+    {
+      "epoch": 4.0996101764464505,
+      "grad_norm": 0.09427163749933243,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 39963
+    },
+    {
+      "epoch": 4.099917931883463,
+      "grad_norm": 0.06199577823281288,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 39966
+    },
+    {
+      "epoch": 4.100225687320476,
+      "grad_norm": 0.04445228725671768,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 39969
+    },
+    {
+      "epoch": 4.100533442757489,
+      "grad_norm": 0.04458901658654213,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 39972
+    },
+    {
+      "epoch": 4.100841198194502,
+      "grad_norm": 0.05720875784754753,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 39975
+    },
+    {
+      "epoch": 4.101148953631514,
+      "grad_norm": 0.0806422233581543,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 39978
+    },
+    {
+      "epoch": 4.101456709068527,
+      "grad_norm": 0.07191216200590134,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 39981
+    },
+    {
+      "epoch": 4.10176446450554,
+      "grad_norm": 0.04921800270676613,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 39984
+    },
+    {
+      "epoch": 4.102072219942552,
+      "grad_norm": 0.09171293675899506,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 39987
+    },
+    {
+      "epoch": 4.102379975379565,
+      "grad_norm": 0.08604893833398819,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 39990
+    },
+    {
+      "epoch": 4.1026877308165774,
+      "grad_norm": 0.0489855632185936,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 39993
+    },
+    {
+      "epoch": 4.10299548625359,
+      "grad_norm": 0.09026241302490234,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 39996
+    },
+    {
+      "epoch": 4.103303241690603,
+      "grad_norm": 0.03529585152864456,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 39999
+    },
+    {
+      "epoch": 4.103610997127616,
+      "grad_norm": 0.10228559374809265,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 40002
+    },
+    {
+      "epoch": 4.103918752564629,
+      "grad_norm": 0.04634955897927284,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 40005
+    },
+    {
+      "epoch": 4.104226508001641,
+      "grad_norm": 0.059635523706674576,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 40008
+    },
+    {
+      "epoch": 4.104534263438654,
+      "grad_norm": 0.07889647036790848,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 40011
+    },
+    {
+      "epoch": 4.104842018875667,
+      "grad_norm": 0.056453004479408264,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 40014
+    },
+    {
+      "epoch": 4.10514977431268,
+      "grad_norm": 0.038360368460416794,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 40017
+    },
+    {
+      "epoch": 4.1054575297496925,
+      "grad_norm": 0.08849389106035233,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 40020
+    },
+    {
+      "epoch": 4.105765285186705,
+      "grad_norm": 0.06063401326537132,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 40023
+    },
+    {
+      "epoch": 4.106073040623718,
+      "grad_norm": 0.08102670311927795,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 40026
+    },
+    {
+      "epoch": 4.106380796060731,
+      "grad_norm": 0.12359301745891571,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 40029
+    },
+    {
+      "epoch": 4.106688551497743,
+      "grad_norm": 0.04933517053723335,
+      "learning_rate": 0.01,
+      "loss": 2.0297,
+      "step": 40032
+    },
+    {
+      "epoch": 4.1069963069347555,
+      "grad_norm": 0.06319262832403183,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 40035
+    },
+    {
+      "epoch": 4.107304062371768,
+      "grad_norm": 0.1100512370467186,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 40038
+    },
+    {
+      "epoch": 4.107611817808781,
+      "grad_norm": 0.04997371882200241,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 40041
+    },
+    {
+      "epoch": 4.107919573245794,
+      "grad_norm": 0.034478992223739624,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 40044
+    },
+    {
+      "epoch": 4.108227328682807,
+      "grad_norm": 0.07332266122102737,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 40047
+    },
+    {
+      "epoch": 4.108535084119819,
+      "grad_norm": 0.07530300319194794,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 40050
+    },
+    {
+      "epoch": 4.108842839556832,
+      "grad_norm": 0.05256710946559906,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 40053
+    },
+    {
+      "epoch": 4.109150594993845,
+      "grad_norm": 0.0342395044863224,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 40056
+    },
+    {
+      "epoch": 4.109458350430858,
+      "grad_norm": 0.05822393298149109,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 40059
+    },
+    {
+      "epoch": 4.1097661058678705,
+      "grad_norm": 0.11184973269701004,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 40062
+    },
+    {
+      "epoch": 4.110073861304883,
+      "grad_norm": 0.10424233227968216,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 40065
+    },
+    {
+      "epoch": 4.110381616741896,
+      "grad_norm": 0.07766477763652802,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 40068
+    },
+    {
+      "epoch": 4.110689372178909,
+      "grad_norm": 0.05194033682346344,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 40071
+    },
+    {
+      "epoch": 4.110997127615922,
+      "grad_norm": 0.059180594980716705,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 40074
+    },
+    {
+      "epoch": 4.1113048830529335,
+      "grad_norm": 0.056441277265548706,
+      "learning_rate": 0.01,
+      "loss": 2.0308,
+      "step": 40077
+    },
+    {
+      "epoch": 4.111612638489946,
+      "grad_norm": 0.10722373425960541,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 40080
+    },
+    {
+      "epoch": 4.111920393926959,
+      "grad_norm": 0.05328166112303734,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 40083
+    },
+    {
+      "epoch": 4.112228149363972,
+      "grad_norm": 0.0920194610953331,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 40086
+    },
+    {
+      "epoch": 4.112535904800985,
+      "grad_norm": 0.09387510269880295,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 40089
+    },
+    {
+      "epoch": 4.112843660237997,
+      "grad_norm": 0.05598600208759308,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 40092
+    },
+    {
+      "epoch": 4.11315141567501,
+      "grad_norm": 0.07474517822265625,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 40095
+    },
+    {
+      "epoch": 4.113459171112023,
+      "grad_norm": 0.05041923001408577,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 40098
+    },
+    {
+      "epoch": 4.113766926549036,
+      "grad_norm": 0.09958838671445847,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 40101
+    },
+    {
+      "epoch": 4.1140746819860485,
+      "grad_norm": 0.07080104202032089,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 40104
+    },
+    {
+      "epoch": 4.114382437423061,
+      "grad_norm": 0.04018507897853851,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 40107
+    },
+    {
+      "epoch": 4.114690192860074,
+      "grad_norm": 0.034507866948843,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 40110
+    },
+    {
+      "epoch": 4.114997948297087,
+      "grad_norm": 0.09373819082975388,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 40113
+    },
+    {
+      "epoch": 4.1153057037341,
+      "grad_norm": 0.09473507106304169,
+      "learning_rate": 0.01,
+      "loss": 2.0374,
+      "step": 40116
+    },
+    {
+      "epoch": 4.115613459171112,
+      "grad_norm": 0.06728008389472961,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 40119
+    },
+    {
+      "epoch": 4.115921214608124,
+      "grad_norm": 0.03551316261291504,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 40122
+    },
+    {
+      "epoch": 4.116228970045137,
+      "grad_norm": 0.048464663326740265,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 40125
+    },
+    {
+      "epoch": 4.11653672548215,
+      "grad_norm": 0.06069257855415344,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 40128
+    },
+    {
+      "epoch": 4.116844480919163,
+      "grad_norm": 0.04196292161941528,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 40131
+    },
+    {
+      "epoch": 4.117152236356175,
+      "grad_norm": 0.056623879820108414,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 40134
+    },
+    {
+      "epoch": 4.117459991793188,
+      "grad_norm": 0.04838422313332558,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 40137
+    },
+    {
+      "epoch": 4.117767747230201,
+      "grad_norm": 0.05446647107601166,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 40140
+    },
+    {
+      "epoch": 4.118075502667214,
+      "grad_norm": 0.04145834967494011,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 40143
+    },
+    {
+      "epoch": 4.118383258104227,
+      "grad_norm": 0.07565242797136307,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 40146
+    },
+    {
+      "epoch": 4.118691013541239,
+      "grad_norm": 0.16024605929851532,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 40149
+    },
+    {
+      "epoch": 4.118998768978252,
+      "grad_norm": 0.10673293471336365,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 40152
+    },
+    {
+      "epoch": 4.119306524415265,
+      "grad_norm": 0.04822041094303131,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 40155
+    },
+    {
+      "epoch": 4.119614279852278,
+      "grad_norm": 0.05074264481663704,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 40158
+    },
+    {
+      "epoch": 4.1199220352892905,
+      "grad_norm": 0.0467289499938488,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 40161
+    },
+    {
+      "epoch": 4.120229790726303,
+      "grad_norm": 0.06745675206184387,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 40164
+    },
+    {
+      "epoch": 4.120537546163315,
+      "grad_norm": 0.05463431030511856,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 40167
+    },
+    {
+      "epoch": 4.120845301600328,
+      "grad_norm": 0.057621899992227554,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 40170
+    },
+    {
+      "epoch": 4.121153057037341,
+      "grad_norm": 0.03955928236246109,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 40173
+    },
+    {
+      "epoch": 4.1214608124743535,
+      "grad_norm": 0.09852177649736404,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 40176
+    },
+    {
+      "epoch": 4.121768567911366,
+      "grad_norm": 0.12283693999052048,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 40179
+    },
+    {
+      "epoch": 4.122076323348379,
+      "grad_norm": 0.046991802752017975,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 40182
+    },
+    {
+      "epoch": 4.122384078785392,
+      "grad_norm": 0.032669685781002045,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 40185
+    },
+    {
+      "epoch": 4.122691834222405,
+      "grad_norm": 0.029325902462005615,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 40188
+    },
+    {
+      "epoch": 4.122999589659417,
+      "grad_norm": 0.05086469277739525,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 40191
+    },
+    {
+      "epoch": 4.12330734509643,
+      "grad_norm": 0.05128968507051468,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 40194
+    },
+    {
+      "epoch": 4.123615100533443,
+      "grad_norm": 0.06649978458881378,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 40197
+    },
+    {
+      "epoch": 4.123922855970456,
+      "grad_norm": 0.07910530269145966,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 40200
+    },
+    {
+      "epoch": 4.1242306114074685,
+      "grad_norm": 0.07179386913776398,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 40203
+    },
+    {
+      "epoch": 4.124538366844481,
+      "grad_norm": 0.04028356075286865,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 40206
+    },
+    {
+      "epoch": 4.124846122281494,
+      "grad_norm": 0.14858920872211456,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 40209
+    },
+    {
+      "epoch": 4.125153877718506,
+      "grad_norm": 0.0865345299243927,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 40212
+    },
+    {
+      "epoch": 4.125461633155519,
+      "grad_norm": 0.04412895813584328,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 40215
+    },
+    {
+      "epoch": 4.1257693885925315,
+      "grad_norm": 0.0482075996696949,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 40218
+    },
+    {
+      "epoch": 4.126077144029544,
+      "grad_norm": 0.0381084606051445,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 40221
+    },
+    {
+      "epoch": 4.126384899466557,
+      "grad_norm": 0.03551500290632248,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 40224
+    },
+    {
+      "epoch": 4.12669265490357,
+      "grad_norm": 0.029879622161388397,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 40227
+    },
+    {
+      "epoch": 4.127000410340583,
+      "grad_norm": 0.04461509361863136,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 40230
+    },
+    {
+      "epoch": 4.127308165777595,
+      "grad_norm": 0.08034146577119827,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 40233
+    },
+    {
+      "epoch": 4.127615921214608,
+      "grad_norm": 0.03752557560801506,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 40236
+    },
+    {
+      "epoch": 4.127923676651621,
+      "grad_norm": 0.09391690045595169,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 40239
+    },
+    {
+      "epoch": 4.128231432088634,
+      "grad_norm": 0.10475057363510132,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 40242
+    },
+    {
+      "epoch": 4.1285391875256465,
+      "grad_norm": 0.05833762139081955,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 40245
+    },
+    {
+      "epoch": 4.128846942962659,
+      "grad_norm": 0.07527395337820053,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 40248
+    },
+    {
+      "epoch": 4.129154698399672,
+      "grad_norm": 0.037238337099552155,
+      "learning_rate": 0.01,
+      "loss": 2.0569,
+      "step": 40251
+    },
+    {
+      "epoch": 4.129462453836685,
+      "grad_norm": 0.08547326922416687,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 40254
+    },
+    {
+      "epoch": 4.129770209273697,
+      "grad_norm": 0.04352400079369545,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 40257
+    },
+    {
+      "epoch": 4.1300779647107095,
+      "grad_norm": 0.03715146332979202,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 40260
+    },
+    {
+      "epoch": 4.130385720147722,
+      "grad_norm": 0.03312180936336517,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 40263
+    },
+    {
+      "epoch": 4.130693475584735,
+      "grad_norm": 0.14868606626987457,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 40266
+    },
+    {
+      "epoch": 4.131001231021748,
+      "grad_norm": 0.0412895530462265,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 40269
+    },
+    {
+      "epoch": 4.131308986458761,
+      "grad_norm": 0.059335898607969284,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 40272
+    },
+    {
+      "epoch": 4.131616741895773,
+      "grad_norm": 0.08401087671518326,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 40275
+    },
+    {
+      "epoch": 4.131924497332786,
+      "grad_norm": 0.09615478664636612,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 40278
+    },
+    {
+      "epoch": 4.132232252769799,
+      "grad_norm": 0.06387398391962051,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 40281
+    },
+    {
+      "epoch": 4.132540008206812,
+      "grad_norm": 0.07762457430362701,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 40284
+    },
+    {
+      "epoch": 4.132847763643825,
+      "grad_norm": 0.08931776881217957,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 40287
+    },
+    {
+      "epoch": 4.133155519080837,
+      "grad_norm": 0.03763844072818756,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 40290
+    },
+    {
+      "epoch": 4.13346327451785,
+      "grad_norm": 0.06352074444293976,
+      "learning_rate": 0.01,
+      "loss": 2.0431,
+      "step": 40293
+    },
+    {
+      "epoch": 4.133771029954863,
+      "grad_norm": 0.04778370261192322,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 40296
+    },
+    {
+      "epoch": 4.134078785391876,
+      "grad_norm": 0.06776099652051926,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 40299
+    },
+    {
+      "epoch": 4.134386540828888,
+      "grad_norm": 0.072423554956913,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 40302
+    },
+    {
+      "epoch": 4.1346942962659,
+      "grad_norm": 0.10224346071481705,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 40305
+    },
+    {
+      "epoch": 4.135002051702913,
+      "grad_norm": 0.045054350048303604,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 40308
+    },
+    {
+      "epoch": 4.135309807139926,
+      "grad_norm": 0.10343952476978302,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 40311
+    },
+    {
+      "epoch": 4.135617562576939,
+      "grad_norm": 0.12315492331981659,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 40314
+    },
+    {
+      "epoch": 4.1359253180139515,
+      "grad_norm": 0.06620458513498306,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 40317
+    },
+    {
+      "epoch": 4.136233073450964,
+      "grad_norm": 0.06378698348999023,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 40320
+    },
+    {
+      "epoch": 4.136540828887977,
+      "grad_norm": 0.06669528782367706,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 40323
+    },
+    {
+      "epoch": 4.13684858432499,
+      "grad_norm": 0.07347302883863449,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 40326
+    },
+    {
+      "epoch": 4.137156339762003,
+      "grad_norm": 0.06786450743675232,
+      "learning_rate": 0.01,
+      "loss": 2.0327,
+      "step": 40329
+    },
+    {
+      "epoch": 4.137464095199015,
+      "grad_norm": 0.10124550014734268,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 40332
+    },
+    {
+      "epoch": 4.137771850636028,
+      "grad_norm": 0.07563018798828125,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 40335
+    },
+    {
+      "epoch": 4.138079606073041,
+      "grad_norm": 0.08920261263847351,
+      "learning_rate": 0.01,
+      "loss": 2.032,
+      "step": 40338
+    },
+    {
+      "epoch": 4.138387361510054,
+      "grad_norm": 0.07678209245204926,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 40341
+    },
+    {
+      "epoch": 4.1386951169470665,
+      "grad_norm": 0.03569749370217323,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 40344
+    },
+    {
+      "epoch": 4.139002872384078,
+      "grad_norm": 0.046265408396720886,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 40347
+    },
+    {
+      "epoch": 4.139310627821091,
+      "grad_norm": 0.04766137897968292,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 40350
+    },
+    {
+      "epoch": 4.139618383258104,
+      "grad_norm": 0.09092257916927338,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 40353
+    },
+    {
+      "epoch": 4.139926138695117,
+      "grad_norm": 0.05511738359928131,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 40356
+    },
+    {
+      "epoch": 4.1402338941321295,
+      "grad_norm": 0.04944039136171341,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 40359
+    },
+    {
+      "epoch": 4.140541649569142,
+      "grad_norm": 0.04905549809336662,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 40362
+    },
+    {
+      "epoch": 4.140849405006155,
+      "grad_norm": 0.07323266565799713,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 40365
+    },
+    {
+      "epoch": 4.141157160443168,
+      "grad_norm": 0.07263099402189255,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 40368
+    },
+    {
+      "epoch": 4.141464915880181,
+      "grad_norm": 0.05902627855539322,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 40371
+    },
+    {
+      "epoch": 4.141772671317193,
+      "grad_norm": 0.0597088448703289,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 40374
+    },
+    {
+      "epoch": 4.142080426754206,
+      "grad_norm": 0.04829247295856476,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 40377
+    },
+    {
+      "epoch": 4.142388182191219,
+      "grad_norm": 0.10093509405851364,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 40380
+    },
+    {
+      "epoch": 4.142695937628232,
+      "grad_norm": 0.06805883347988129,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 40383
+    },
+    {
+      "epoch": 4.1430036930652445,
+      "grad_norm": 0.05700362101197243,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 40386
+    },
+    {
+      "epoch": 4.143311448502257,
+      "grad_norm": 0.10143036395311356,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 40389
+    },
+    {
+      "epoch": 4.143619203939269,
+      "grad_norm": 0.0752461776137352,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 40392
+    },
+    {
+      "epoch": 4.143926959376282,
+      "grad_norm": 0.10534628480672836,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 40395
+    },
+    {
+      "epoch": 4.144234714813295,
+      "grad_norm": 0.042690735310316086,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 40398
+    },
+    {
+      "epoch": 4.1445424702503075,
+      "grad_norm": 0.04265071824193001,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 40401
+    },
+    {
+      "epoch": 4.14485022568732,
+      "grad_norm": 0.04449966922402382,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 40404
+    },
+    {
+      "epoch": 4.145157981124333,
+      "grad_norm": 0.06278623640537262,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 40407
+    },
+    {
+      "epoch": 4.145465736561346,
+      "grad_norm": 0.09633596986532211,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 40410
+    },
+    {
+      "epoch": 4.145773491998359,
+      "grad_norm": 0.06465546041727066,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 40413
+    },
+    {
+      "epoch": 4.146081247435371,
+      "grad_norm": 0.07929039746522903,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 40416
+    },
+    {
+      "epoch": 4.146389002872384,
+      "grad_norm": 0.04692156985402107,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 40419
+    },
+    {
+      "epoch": 4.146696758309397,
+      "grad_norm": 0.09933728724718094,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 40422
+    },
+    {
+      "epoch": 4.14700451374641,
+      "grad_norm": 0.07595301419496536,
+      "learning_rate": 0.01,
+      "loss": 2.0262,
+      "step": 40425
+    },
+    {
+      "epoch": 4.1473122691834226,
+      "grad_norm": 0.06487558037042618,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 40428
+    },
+    {
+      "epoch": 4.147620024620435,
+      "grad_norm": 0.06612923741340637,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 40431
+    },
+    {
+      "epoch": 4.147927780057448,
+      "grad_norm": 0.08177967369556427,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 40434
+    },
+    {
+      "epoch": 4.14823553549446,
+      "grad_norm": 0.12479262799024582,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 40437
+    },
+    {
+      "epoch": 4.148543290931473,
+      "grad_norm": 0.053598951548337936,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 40440
+    },
+    {
+      "epoch": 4.148851046368486,
+      "grad_norm": 0.05604923143982887,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 40443
+    },
+    {
+      "epoch": 4.149158801805498,
+      "grad_norm": 0.044350240379571915,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 40446
+    },
+    {
+      "epoch": 4.149466557242511,
+      "grad_norm": 0.03988419473171234,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 40449
+    },
+    {
+      "epoch": 4.149774312679524,
+      "grad_norm": 0.03739091381430626,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 40452
+    },
+    {
+      "epoch": 4.150082068116537,
+      "grad_norm": 0.0570409819483757,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 40455
+    },
+    {
+      "epoch": 4.1503898235535495,
+      "grad_norm": 0.11358576267957687,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 40458
+    },
+    {
+      "epoch": 4.150697578990562,
+      "grad_norm": 0.08568841218948364,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 40461
+    },
+    {
+      "epoch": 4.151005334427575,
+      "grad_norm": 0.10045678168535233,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 40464
+    },
+    {
+      "epoch": 4.151313089864588,
+      "grad_norm": 0.0720805749297142,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 40467
+    },
+    {
+      "epoch": 4.151620845301601,
+      "grad_norm": 0.136969655752182,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 40470
+    },
+    {
+      "epoch": 4.151928600738613,
+      "grad_norm": 0.1128631979227066,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 40473
+    },
+    {
+      "epoch": 4.152236356175626,
+      "grad_norm": 0.06798527389764786,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 40476
+    },
+    {
+      "epoch": 4.152544111612638,
+      "grad_norm": 0.04990516975522041,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 40479
+    },
+    {
+      "epoch": 4.152851867049651,
+      "grad_norm": 0.0540931336581707,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 40482
+    },
+    {
+      "epoch": 4.153159622486664,
+      "grad_norm": 0.06866136193275452,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 40485
+    },
+    {
+      "epoch": 4.153467377923676,
+      "grad_norm": 0.05682525411248207,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 40488
+    },
+    {
+      "epoch": 4.153775133360689,
+      "grad_norm": 0.07948753237724304,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 40491
+    },
+    {
+      "epoch": 4.154082888797702,
+      "grad_norm": 0.10209072381258011,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 40494
+    },
+    {
+      "epoch": 4.154390644234715,
+      "grad_norm": 0.04499402642250061,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 40497
+    },
+    {
+      "epoch": 4.1546983996717275,
+      "grad_norm": 0.03882760554552078,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 40500
+    },
+    {
+      "epoch": 4.15500615510874,
+      "grad_norm": 0.1203235313296318,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 40503
+    },
+    {
+      "epoch": 4.155313910545753,
+      "grad_norm": 0.10118594765663147,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 40506
+    },
+    {
+      "epoch": 4.155621665982766,
+      "grad_norm": 0.057894591242074966,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 40509
+    },
+    {
+      "epoch": 4.155929421419779,
+      "grad_norm": 0.09510418772697449,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 40512
+    },
+    {
+      "epoch": 4.156237176856791,
+      "grad_norm": 0.06964008510112762,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 40515
+    },
+    {
+      "epoch": 4.156544932293804,
+      "grad_norm": 0.04545210301876068,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 40518
+    },
+    {
+      "epoch": 4.156852687730817,
+      "grad_norm": 0.03616971895098686,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 40521
+    },
+    {
+      "epoch": 4.157160443167829,
+      "grad_norm": 0.038192588835954666,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 40524
+    },
+    {
+      "epoch": 4.157468198604842,
+      "grad_norm": 0.046076469123363495,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 40527
+    },
+    {
+      "epoch": 4.157775954041854,
+      "grad_norm": 0.08816591650247574,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 40530
+    },
+    {
+      "epoch": 4.158083709478867,
+      "grad_norm": 0.10513858497142792,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 40533
+    },
+    {
+      "epoch": 4.15839146491588,
+      "grad_norm": 0.05273423343896866,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 40536
+    },
+    {
+      "epoch": 4.158699220352893,
+      "grad_norm": 0.11179885268211365,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 40539
+    },
+    {
+      "epoch": 4.1590069757899055,
+      "grad_norm": 0.049401313066482544,
+      "learning_rate": 0.01,
+      "loss": 2.0208,
+      "step": 40542
+    },
+    {
+      "epoch": 4.159314731226918,
+      "grad_norm": 0.04798784479498863,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 40545
+    },
+    {
+      "epoch": 4.159622486663931,
+      "grad_norm": 0.04317957162857056,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 40548
+    },
+    {
+      "epoch": 4.159930242100944,
+      "grad_norm": 0.031536102294921875,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 40551
+    },
+    {
+      "epoch": 4.160237997537957,
+      "grad_norm": 0.09016025066375732,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 40554
+    },
+    {
+      "epoch": 4.160545752974969,
+      "grad_norm": 0.048575472086668015,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 40557
+    },
+    {
+      "epoch": 4.160853508411982,
+      "grad_norm": 0.1236988827586174,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 40560
+    },
+    {
+      "epoch": 4.161161263848995,
+      "grad_norm": 0.13077518343925476,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 40563
+    },
+    {
+      "epoch": 4.161469019286008,
+      "grad_norm": 0.10104335844516754,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 40566
+    },
+    {
+      "epoch": 4.16177677472302,
+      "grad_norm": 0.08807270973920822,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 40569
+    },
+    {
+      "epoch": 4.162084530160032,
+      "grad_norm": 0.05569664016366005,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 40572
+    },
+    {
+      "epoch": 4.162392285597045,
+      "grad_norm": 0.04396286606788635,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 40575
+    },
+    {
+      "epoch": 4.162700041034058,
+      "grad_norm": 0.041279278695583344,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 40578
+    },
+    {
+      "epoch": 4.163007796471071,
+      "grad_norm": 0.05588890612125397,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 40581
+    },
+    {
+      "epoch": 4.163315551908084,
+      "grad_norm": 0.03629462048411369,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 40584
+    },
+    {
+      "epoch": 4.163623307345096,
+      "grad_norm": 0.05996616184711456,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 40587
+    },
+    {
+      "epoch": 4.163931062782109,
+      "grad_norm": 0.0385168232023716,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 40590
+    },
+    {
+      "epoch": 4.164238818219122,
+      "grad_norm": 0.04890631511807442,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 40593
+    },
+    {
+      "epoch": 4.164546573656135,
+      "grad_norm": 0.06312946230173111,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 40596
+    },
+    {
+      "epoch": 4.1648543290931475,
+      "grad_norm": 0.05250309780240059,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 40599
+    },
+    {
+      "epoch": 4.16516208453016,
+      "grad_norm": 0.05581989511847496,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 40602
+    },
+    {
+      "epoch": 4.165469839967173,
+      "grad_norm": 0.033916786313056946,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 40605
+    },
+    {
+      "epoch": 4.165777595404186,
+      "grad_norm": 0.11886032670736313,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 40608
+    },
+    {
+      "epoch": 4.166085350841199,
+      "grad_norm": 0.06368937343358994,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 40611
+    },
+    {
+      "epoch": 4.1663931062782105,
+      "grad_norm": 0.04808172211050987,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 40614
+    },
+    {
+      "epoch": 4.166700861715223,
+      "grad_norm": 0.03308833763003349,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 40617
+    },
+    {
+      "epoch": 4.167008617152236,
+      "grad_norm": 0.041934408247470856,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 40620
+    },
+    {
+      "epoch": 4.167316372589249,
+      "grad_norm": 0.0798778086900711,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 40623
+    },
+    {
+      "epoch": 4.167624128026262,
+      "grad_norm": 0.055231209844350815,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 40626
+    },
+    {
+      "epoch": 4.167931883463274,
+      "grad_norm": 0.04620687663555145,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 40629
+    },
+    {
+      "epoch": 4.168239638900287,
+      "grad_norm": 0.09074298292398453,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 40632
+    },
+    {
+      "epoch": 4.1685473943373,
+      "grad_norm": 0.09173654764890671,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 40635
+    },
+    {
+      "epoch": 4.168855149774313,
+      "grad_norm": 0.052037082612514496,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 40638
+    },
+    {
+      "epoch": 4.1691629052113255,
+      "grad_norm": 0.07793330401182175,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 40641
+    },
+    {
+      "epoch": 4.169470660648338,
+      "grad_norm": 0.12099308520555496,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 40644
+    },
+    {
+      "epoch": 4.169778416085351,
+      "grad_norm": 0.08884571492671967,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 40647
+    },
+    {
+      "epoch": 4.170086171522364,
+      "grad_norm": 0.07165110856294632,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 40650
+    },
+    {
+      "epoch": 4.170393926959377,
+      "grad_norm": 0.10293866693973541,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 40653
+    },
+    {
+      "epoch": 4.170701682396389,
+      "grad_norm": 0.09834318608045578,
+      "learning_rate": 0.01,
+      "loss": 2.0313,
+      "step": 40656
+    },
+    {
+      "epoch": 4.171009437833401,
+      "grad_norm": 0.047723252326250076,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 40659
+    },
+    {
+      "epoch": 4.171317193270414,
+      "grad_norm": 0.05522621423006058,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 40662
+    },
+    {
+      "epoch": 4.171624948707427,
+      "grad_norm": 0.03939025476574898,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 40665
+    },
+    {
+      "epoch": 4.17193270414444,
+      "grad_norm": 0.03266485407948494,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 40668
+    },
+    {
+      "epoch": 4.172240459581452,
+      "grad_norm": 0.033200979232788086,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 40671
+    },
+    {
+      "epoch": 4.172548215018465,
+      "grad_norm": 0.07104014605283737,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 40674
+    },
+    {
+      "epoch": 4.172855970455478,
+      "grad_norm": 0.0823940560221672,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 40677
+    },
+    {
+      "epoch": 4.173163725892491,
+      "grad_norm": 0.062255583703517914,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 40680
+    },
+    {
+      "epoch": 4.1734714813295035,
+      "grad_norm": 0.13120077550411224,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 40683
+    },
+    {
+      "epoch": 4.173779236766516,
+      "grad_norm": 0.07655888795852661,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 40686
+    },
+    {
+      "epoch": 4.174086992203529,
+      "grad_norm": 0.07436554878950119,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 40689
+    },
+    {
+      "epoch": 4.174394747640542,
+      "grad_norm": 0.052044034004211426,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 40692
+    },
+    {
+      "epoch": 4.174702503077555,
+      "grad_norm": 0.0385587215423584,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 40695
+    },
+    {
+      "epoch": 4.175010258514567,
+      "grad_norm": 0.04062971845269203,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 40698
+    },
+    {
+      "epoch": 4.17531801395158,
+      "grad_norm": 0.05078468099236488,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 40701
+    },
+    {
+      "epoch": 4.175625769388592,
+      "grad_norm": 0.05820399895310402,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 40704
+    },
+    {
+      "epoch": 4.175933524825605,
+      "grad_norm": 0.05630197748541832,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 40707
+    },
+    {
+      "epoch": 4.176241280262618,
+      "grad_norm": 0.047048747539520264,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 40710
+    },
+    {
+      "epoch": 4.17654903569963,
+      "grad_norm": 0.09604250639677048,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 40713
+    },
+    {
+      "epoch": 4.176856791136643,
+      "grad_norm": 0.06629175692796707,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 40716
+    },
+    {
+      "epoch": 4.177164546573656,
+      "grad_norm": 0.1041264608502388,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 40719
+    },
+    {
+      "epoch": 4.177472302010669,
+      "grad_norm": 0.10279016941785812,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 40722
+    },
+    {
+      "epoch": 4.1777800574476815,
+      "grad_norm": 0.047386229038238525,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 40725
+    },
+    {
+      "epoch": 4.178087812884694,
+      "grad_norm": 0.03940087929368019,
+      "learning_rate": 0.01,
+      "loss": 2.0454,
+      "step": 40728
+    },
+    {
+      "epoch": 4.178395568321707,
+      "grad_norm": 0.1065608337521553,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 40731
+    },
+    {
+      "epoch": 4.17870332375872,
+      "grad_norm": 0.03307456895709038,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 40734
+    },
+    {
+      "epoch": 4.179011079195733,
+      "grad_norm": 0.044952236115932465,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 40737
+    },
+    {
+      "epoch": 4.1793188346327454,
+      "grad_norm": 0.04699942097067833,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 40740
+    },
+    {
+      "epoch": 4.179626590069758,
+      "grad_norm": 0.04086962714791298,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 40743
+    },
+    {
+      "epoch": 4.179934345506771,
+      "grad_norm": 0.04364610090851784,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 40746
+    },
+    {
+      "epoch": 4.180242100943783,
+      "grad_norm": 0.13256701827049255,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 40749
+    },
+    {
+      "epoch": 4.180549856380796,
+      "grad_norm": 0.1418839991092682,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 40752
+    },
+    {
+      "epoch": 4.1808576118178085,
+      "grad_norm": 0.10980424284934998,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 40755
+    },
+    {
+      "epoch": 4.181165367254821,
+      "grad_norm": 0.11712696403265,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 40758
+    },
+    {
+      "epoch": 4.181473122691834,
+      "grad_norm": 0.04676150158047676,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 40761
+    },
+    {
+      "epoch": 4.181780878128847,
+      "grad_norm": 0.06010957807302475,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 40764
+    },
+    {
+      "epoch": 4.18208863356586,
+      "grad_norm": 0.04058792442083359,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 40767
+    },
+    {
+      "epoch": 4.182396389002872,
+      "grad_norm": 0.05905113369226456,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 40770
+    },
+    {
+      "epoch": 4.182704144439885,
+      "grad_norm": 0.031559672206640244,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 40773
+    },
+    {
+      "epoch": 4.183011899876898,
+      "grad_norm": 0.03369845449924469,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 40776
+    },
+    {
+      "epoch": 4.183319655313911,
+      "grad_norm": 0.03323826193809509,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 40779
+    },
+    {
+      "epoch": 4.1836274107509235,
+      "grad_norm": 0.09272177517414093,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 40782
+    },
+    {
+      "epoch": 4.183935166187936,
+      "grad_norm": 0.12024440616369247,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 40785
+    },
+    {
+      "epoch": 4.184242921624949,
+      "grad_norm": 0.06225709244608879,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 40788
+    },
+    {
+      "epoch": 4.184550677061962,
+      "grad_norm": 0.07221471518278122,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 40791
+    },
+    {
+      "epoch": 4.184858432498974,
+      "grad_norm": 0.05865592509508133,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 40794
+    },
+    {
+      "epoch": 4.1851661879359865,
+      "grad_norm": 0.05323261767625809,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 40797
+    },
+    {
+      "epoch": 4.185473943372999,
+      "grad_norm": 0.0738186463713646,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 40800
+    },
+    {
+      "epoch": 4.185781698810012,
+      "grad_norm": 0.05065621808171272,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 40803
+    },
+    {
+      "epoch": 4.186089454247025,
+      "grad_norm": 0.07014346867799759,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 40806
+    },
+    {
+      "epoch": 4.186397209684038,
+      "grad_norm": 0.09202650934457779,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 40809
+    },
+    {
+      "epoch": 4.18670496512105,
+      "grad_norm": 0.07297877967357635,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 40812
+    },
+    {
+      "epoch": 4.187012720558063,
+      "grad_norm": 0.11443314701318741,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 40815
+    },
+    {
+      "epoch": 4.187320475995076,
+      "grad_norm": 0.10481325536966324,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 40818
+    },
+    {
+      "epoch": 4.187628231432089,
+      "grad_norm": 0.03423392400145531,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 40821
+    },
+    {
+      "epoch": 4.1879359868691015,
+      "grad_norm": 0.03760581091046333,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 40824
+    },
+    {
+      "epoch": 4.188243742306114,
+      "grad_norm": 0.04355360567569733,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 40827
+    },
+    {
+      "epoch": 4.188551497743127,
+      "grad_norm": 0.07982810586690903,
+      "learning_rate": 0.01,
+      "loss": 2.0474,
+      "step": 40830
+    },
+    {
+      "epoch": 4.18885925318014,
+      "grad_norm": 0.17527154088020325,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 40833
+    },
+    {
+      "epoch": 4.189167008617153,
+      "grad_norm": 0.09842000156641006,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 40836
+    },
+    {
+      "epoch": 4.1894747640541645,
+      "grad_norm": 0.04522009938955307,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 40839
+    },
+    {
+      "epoch": 4.189782519491177,
+      "grad_norm": 0.03711741417646408,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 40842
+    },
+    {
+      "epoch": 4.19009027492819,
+      "grad_norm": 0.06014026701450348,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 40845
+    },
+    {
+      "epoch": 4.190398030365203,
+      "grad_norm": 0.06271639466285706,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 40848
+    },
+    {
+      "epoch": 4.190705785802216,
+      "grad_norm": 0.09300049394369125,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 40851
+    },
+    {
+      "epoch": 4.191013541239228,
+      "grad_norm": 0.05301731079816818,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 40854
+    },
+    {
+      "epoch": 4.191321296676241,
+      "grad_norm": 0.031089186668395996,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 40857
+    },
+    {
+      "epoch": 4.191629052113254,
+      "grad_norm": 0.04351799562573433,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 40860
+    },
+    {
+      "epoch": 4.191936807550267,
+      "grad_norm": 0.03952482342720032,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 40863
+    },
+    {
+      "epoch": 4.1922445629872795,
+      "grad_norm": 0.09664411842823029,
+      "learning_rate": 0.01,
+      "loss": 2.0385,
+      "step": 40866
+    },
+    {
+      "epoch": 4.192552318424292,
+      "grad_norm": 0.11290398985147476,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 40869
+    },
+    {
+      "epoch": 4.192860073861305,
+      "grad_norm": 0.03346327319741249,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 40872
+    },
+    {
+      "epoch": 4.193167829298318,
+      "grad_norm": 0.07427312433719635,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 40875
+    },
+    {
+      "epoch": 4.193475584735331,
+      "grad_norm": 0.053705740720033646,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 40878
+    },
+    {
+      "epoch": 4.193783340172343,
+      "grad_norm": 0.0523017942905426,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 40881
+    },
+    {
+      "epoch": 4.194091095609355,
+      "grad_norm": 0.056644510477781296,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 40884
+    },
+    {
+      "epoch": 4.194398851046368,
+      "grad_norm": 0.03145689144730568,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 40887
+    },
+    {
+      "epoch": 4.194706606483381,
+      "grad_norm": 0.05200685188174248,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 40890
+    },
+    {
+      "epoch": 4.195014361920394,
+      "grad_norm": 0.06475922465324402,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 40893
+    },
+    {
+      "epoch": 4.1953221173574065,
+      "grad_norm": 0.0805899053812027,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 40896
+    },
+    {
+      "epoch": 4.195629872794419,
+      "grad_norm": 0.09969832003116608,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 40899
+    },
+    {
+      "epoch": 4.195937628231432,
+      "grad_norm": 0.0634358674287796,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 40902
+    },
+    {
+      "epoch": 4.196245383668445,
+      "grad_norm": 0.09012358635663986,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 40905
+    },
+    {
+      "epoch": 4.196553139105458,
+      "grad_norm": 0.06891899555921555,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 40908
+    },
+    {
+      "epoch": 4.19686089454247,
+      "grad_norm": 0.07446330040693283,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 40911
+    },
+    {
+      "epoch": 4.197168649979483,
+      "grad_norm": 0.03457944840192795,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 40914
+    },
+    {
+      "epoch": 4.197476405416496,
+      "grad_norm": 0.03754154592752457,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 40917
+    },
+    {
+      "epoch": 4.197784160853509,
+      "grad_norm": 0.0626758560538292,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 40920
+    },
+    {
+      "epoch": 4.1980919162905215,
+      "grad_norm": 0.0431663803756237,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 40923
+    },
+    {
+      "epoch": 4.198399671727534,
+      "grad_norm": 0.059978071600198746,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 40926
+    },
+    {
+      "epoch": 4.198707427164546,
+      "grad_norm": 0.045393504202365875,
+      "learning_rate": 0.01,
+      "loss": 2.0355,
+      "step": 40929
+    },
+    {
+      "epoch": 4.199015182601559,
+      "grad_norm": 0.1481604129076004,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 40932
+    },
+    {
+      "epoch": 4.199322938038572,
+      "grad_norm": 0.08962929248809814,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 40935
+    },
+    {
+      "epoch": 4.1996306934755845,
+      "grad_norm": 0.10029556602239609,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 40938
+    },
+    {
+      "epoch": 4.199938448912597,
+      "grad_norm": 0.060972318053245544,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 40941
+    },
+    {
+      "epoch": 4.20024620434961,
+      "grad_norm": 0.08175275474786758,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 40944
+    },
+    {
+      "epoch": 4.200553959786623,
+      "grad_norm": 0.07742544263601303,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 40947
+    },
+    {
+      "epoch": 4.200861715223636,
+      "grad_norm": 0.06145945191383362,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 40950
+    },
+    {
+      "epoch": 4.201169470660648,
+      "grad_norm": 0.03785976395010948,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 40953
+    },
+    {
+      "epoch": 4.201477226097661,
+      "grad_norm": 0.049219708889722824,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 40956
+    },
+    {
+      "epoch": 4.201784981534674,
+      "grad_norm": 0.05865636467933655,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 40959
+    },
+    {
+      "epoch": 4.202092736971687,
+      "grad_norm": 0.04115782678127289,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 40962
+    },
+    {
+      "epoch": 4.2024004924086995,
+      "grad_norm": 0.07156416773796082,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 40965
+    },
+    {
+      "epoch": 4.202708247845712,
+      "grad_norm": 0.11241970211267471,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 40968
+    },
+    {
+      "epoch": 4.203016003282725,
+      "grad_norm": 0.052014194428920746,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 40971
+    },
+    {
+      "epoch": 4.203323758719737,
+      "grad_norm": 0.03571850433945656,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 40974
+    },
+    {
+      "epoch": 4.20363151415675,
+      "grad_norm": 0.03782523795962334,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 40977
+    },
+    {
+      "epoch": 4.2039392695937625,
+      "grad_norm": 0.05531090870499611,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 40980
+    },
+    {
+      "epoch": 4.204247025030775,
+      "grad_norm": 0.04620610550045967,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 40983
+    },
+    {
+      "epoch": 4.204554780467788,
+      "grad_norm": 0.034513216465711594,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 40986
+    },
+    {
+      "epoch": 4.204862535904801,
+      "grad_norm": 0.12451375275850296,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 40989
+    },
+    {
+      "epoch": 4.205170291341814,
+      "grad_norm": 0.038147494196891785,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 40992
+    },
+    {
+      "epoch": 4.205478046778826,
+      "grad_norm": 0.0428970605134964,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 40995
+    },
+    {
+      "epoch": 4.205785802215839,
+      "grad_norm": 0.04861719533801079,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 40998
+    },
+    {
+      "epoch": 4.206093557652852,
+      "grad_norm": 0.05677000805735588,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 41001
+    },
+    {
+      "epoch": 4.206401313089865,
+      "grad_norm": 0.040025245398283005,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 41004
+    },
+    {
+      "epoch": 4.2067090685268775,
+      "grad_norm": 0.09284701943397522,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 41007
+    },
+    {
+      "epoch": 4.20701682396389,
+      "grad_norm": 0.06293929368257523,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 41010
+    },
+    {
+      "epoch": 4.207324579400903,
+      "grad_norm": 0.10422204434871674,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 41013
+    },
+    {
+      "epoch": 4.207632334837916,
+      "grad_norm": 0.05504370108246803,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 41016
+    },
+    {
+      "epoch": 4.207940090274928,
+      "grad_norm": 0.03916056081652641,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 41019
+    },
+    {
+      "epoch": 4.2082478457119405,
+      "grad_norm": 0.05535780265927315,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 41022
+    },
+    {
+      "epoch": 4.208555601148953,
+      "grad_norm": 0.11614353209733963,
+      "learning_rate": 0.01,
+      "loss": 2.0394,
+      "step": 41025
+    },
+    {
+      "epoch": 4.208863356585966,
+      "grad_norm": 0.054821841418743134,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 41028
+    },
+    {
+      "epoch": 4.209171112022979,
+      "grad_norm": 0.0761130154132843,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 41031
+    },
+    {
+      "epoch": 4.209478867459992,
+      "grad_norm": 0.05529718101024628,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 41034
+    },
+    {
+      "epoch": 4.2097866228970044,
+      "grad_norm": 0.05739063769578934,
+      "learning_rate": 0.01,
+      "loss": 2.0408,
+      "step": 41037
+    },
+    {
+      "epoch": 4.210094378334017,
+      "grad_norm": 0.03629623353481293,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 41040
+    },
+    {
+      "epoch": 4.21040213377103,
+      "grad_norm": 0.09579025954008102,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 41043
+    },
+    {
+      "epoch": 4.210709889208043,
+      "grad_norm": 0.09709680080413818,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 41046
+    },
+    {
+      "epoch": 4.211017644645056,
+      "grad_norm": 0.06710288673639297,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 41049
+    },
+    {
+      "epoch": 4.211325400082068,
+      "grad_norm": 0.17285798490047455,
+      "learning_rate": 0.01,
+      "loss": 2.0525,
+      "step": 41052
+    },
+    {
+      "epoch": 4.211633155519081,
+      "grad_norm": 0.13886108994483948,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 41055
+    },
+    {
+      "epoch": 4.211940910956094,
+      "grad_norm": 0.08146315068006516,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 41058
+    },
+    {
+      "epoch": 4.212248666393107,
+      "grad_norm": 0.041313815861940384,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 41061
+    },
+    {
+      "epoch": 4.212556421830119,
+      "grad_norm": 0.07856039702892303,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 41064
+    },
+    {
+      "epoch": 4.212864177267131,
+      "grad_norm": 0.03942018374800682,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 41067
+    },
+    {
+      "epoch": 4.213171932704144,
+      "grad_norm": 0.06405628472566605,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 41070
+    },
+    {
+      "epoch": 4.213479688141157,
+      "grad_norm": 0.11233191937208176,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 41073
+    },
+    {
+      "epoch": 4.21378744357817,
+      "grad_norm": 0.061680734157562256,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 41076
+    },
+    {
+      "epoch": 4.2140951990151825,
+      "grad_norm": 0.049544453620910645,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 41079
+    },
+    {
+      "epoch": 4.214402954452195,
+      "grad_norm": 0.03959604352712631,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 41082
+    },
+    {
+      "epoch": 4.214710709889208,
+      "grad_norm": 0.03029470518231392,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 41085
+    },
+    {
+      "epoch": 4.215018465326221,
+      "grad_norm": 0.04742530733346939,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 41088
+    },
+    {
+      "epoch": 4.215326220763234,
+      "grad_norm": 0.11508273333311081,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 41091
+    },
+    {
+      "epoch": 4.215633976200246,
+      "grad_norm": 0.16011863946914673,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 41094
+    },
+    {
+      "epoch": 4.215941731637259,
+      "grad_norm": 0.0831206887960434,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 41097
+    },
+    {
+      "epoch": 4.216249487074272,
+      "grad_norm": 0.05117999389767647,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 41100
+    },
+    {
+      "epoch": 4.216557242511285,
+      "grad_norm": 0.08137553930282593,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 41103
+    },
+    {
+      "epoch": 4.2168649979482975,
+      "grad_norm": 0.04084109142422676,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 41106
+    },
+    {
+      "epoch": 4.217172753385309,
+      "grad_norm": 0.11708495765924454,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 41109
+    },
+    {
+      "epoch": 4.217480508822322,
+      "grad_norm": 0.036096177995204926,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 41112
+    },
+    {
+      "epoch": 4.217788264259335,
+      "grad_norm": 0.03205428272485733,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 41115
+    },
+    {
+      "epoch": 4.218096019696348,
+      "grad_norm": 0.05120716243982315,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 41118
+    },
+    {
+      "epoch": 4.2184037751333605,
+      "grad_norm": 0.08021282404661179,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 41121
+    },
+    {
+      "epoch": 4.218711530570373,
+      "grad_norm": 0.05003154277801514,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 41124
+    },
+    {
+      "epoch": 4.219019286007386,
+      "grad_norm": 0.0867641344666481,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 41127
+    },
+    {
+      "epoch": 4.219327041444399,
+      "grad_norm": 0.0697675347328186,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 41130
+    },
+    {
+      "epoch": 4.219634796881412,
+      "grad_norm": 0.06696760654449463,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 41133
+    },
+    {
+      "epoch": 4.219942552318424,
+      "grad_norm": 0.09599725902080536,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 41136
+    },
+    {
+      "epoch": 4.220250307755437,
+      "grad_norm": 0.04233787953853607,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 41139
+    },
+    {
+      "epoch": 4.22055806319245,
+      "grad_norm": 0.08935974538326263,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 41142
+    },
+    {
+      "epoch": 4.220865818629463,
+      "grad_norm": 0.049136627465486526,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 41145
+    },
+    {
+      "epoch": 4.2211735740664755,
+      "grad_norm": 0.0316915288567543,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 41148
+    },
+    {
+      "epoch": 4.221481329503488,
+      "grad_norm": 0.04271787405014038,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 41151
+    },
+    {
+      "epoch": 4.2217890849405,
+      "grad_norm": 0.03403954207897186,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 41154
+    },
+    {
+      "epoch": 4.222096840377513,
+      "grad_norm": 0.07457596808671951,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 41157
+    },
+    {
+      "epoch": 4.222404595814526,
+      "grad_norm": 0.06875099986791611,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 41160
+    },
+    {
+      "epoch": 4.2227123512515385,
+      "grad_norm": 0.05186415836215019,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 41163
+    },
+    {
+      "epoch": 4.223020106688551,
+      "grad_norm": 0.051673002541065216,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 41166
+    },
+    {
+      "epoch": 4.223327862125564,
+      "grad_norm": 0.03935972973704338,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 41169
+    },
+    {
+      "epoch": 4.223635617562577,
+      "grad_norm": 0.07746358215808868,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 41172
+    },
+    {
+      "epoch": 4.22394337299959,
+      "grad_norm": 0.14719370007514954,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 41175
+    },
+    {
+      "epoch": 4.224251128436602,
+      "grad_norm": 0.08811825513839722,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 41178
+    },
+    {
+      "epoch": 4.224558883873615,
+      "grad_norm": 0.04378504678606987,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 41181
+    },
+    {
+      "epoch": 4.224866639310628,
+      "grad_norm": 0.03978777304291725,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 41184
+    },
+    {
+      "epoch": 4.225174394747641,
+      "grad_norm": 0.03860542178153992,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 41187
+    },
+    {
+      "epoch": 4.225482150184654,
+      "grad_norm": 0.06839700043201447,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 41190
+    },
+    {
+      "epoch": 4.225789905621666,
+      "grad_norm": 0.09107422828674316,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 41193
+    },
+    {
+      "epoch": 4.226097661058679,
+      "grad_norm": 0.08443713933229446,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 41196
+    },
+    {
+      "epoch": 4.226405416495691,
+      "grad_norm": 0.07428136467933655,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 41199
+    },
+    {
+      "epoch": 4.226713171932704,
+      "grad_norm": 0.07102616876363754,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 41202
+    },
+    {
+      "epoch": 4.227020927369717,
+      "grad_norm": 0.06131945922970772,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 41205
+    },
+    {
+      "epoch": 4.227328682806729,
+      "grad_norm": 0.07921816408634186,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 41208
+    },
+    {
+      "epoch": 4.227636438243742,
+      "grad_norm": 0.05274220556020737,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 41211
+    },
+    {
+      "epoch": 4.227944193680755,
+      "grad_norm": 0.12561160326004028,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 41214
+    },
+    {
+      "epoch": 4.228251949117768,
+      "grad_norm": 0.053055427968502045,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 41217
+    },
+    {
+      "epoch": 4.2285597045547805,
+      "grad_norm": 0.04999428242444992,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 41220
+    },
+    {
+      "epoch": 4.228867459991793,
+      "grad_norm": 0.06127345934510231,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 41223
+    },
+    {
+      "epoch": 4.229175215428806,
+      "grad_norm": 0.05634259432554245,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 41226
+    },
+    {
+      "epoch": 4.229482970865819,
+      "grad_norm": 0.030200546607375145,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 41229
+    },
+    {
+      "epoch": 4.229790726302832,
+      "grad_norm": 0.12201087921857834,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 41232
+    },
+    {
+      "epoch": 4.230098481739844,
+      "grad_norm": 0.07482577860355377,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 41235
+    },
+    {
+      "epoch": 4.230406237176857,
+      "grad_norm": 0.10311067849397659,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 41238
+    },
+    {
+      "epoch": 4.23071399261387,
+      "grad_norm": 0.09040312469005585,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 41241
+    },
+    {
+      "epoch": 4.231021748050882,
+      "grad_norm": 0.06239837408065796,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 41244
+    },
+    {
+      "epoch": 4.231329503487895,
+      "grad_norm": 0.033743564039468765,
+      "learning_rate": 0.01,
+      "loss": 2.0273,
+      "step": 41247
+    },
+    {
+      "epoch": 4.231637258924907,
+      "grad_norm": 0.046916600316762924,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 41250
+    },
+    {
+      "epoch": 4.23194501436192,
+      "grad_norm": 0.06271935999393463,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 41253
+    },
+    {
+      "epoch": 4.232252769798933,
+      "grad_norm": 0.05794825777411461,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 41256
+    },
+    {
+      "epoch": 4.232560525235946,
+      "grad_norm": 0.0886342003941536,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 41259
+    },
+    {
+      "epoch": 4.2328682806729585,
+      "grad_norm": 0.06492005288600922,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 41262
+    },
+    {
+      "epoch": 4.233176036109971,
+      "grad_norm": 0.1275957226753235,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 41265
+    },
+    {
+      "epoch": 4.233483791546984,
+      "grad_norm": 0.08362898975610733,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 41268
+    },
+    {
+      "epoch": 4.233791546983997,
+      "grad_norm": 0.0909474641084671,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 41271
+    },
+    {
+      "epoch": 4.23409930242101,
+      "grad_norm": 0.04275381565093994,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 41274
+    },
+    {
+      "epoch": 4.234407057858022,
+      "grad_norm": 0.04790336266160011,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 41277
+    },
+    {
+      "epoch": 4.234714813295035,
+      "grad_norm": 0.03371073678135872,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 41280
+    },
+    {
+      "epoch": 4.235022568732048,
+      "grad_norm": 0.05681711062788963,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 41283
+    },
+    {
+      "epoch": 4.235330324169061,
+      "grad_norm": 0.040656253695487976,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 41286
+    },
+    {
+      "epoch": 4.235638079606073,
+      "grad_norm": 0.04097260907292366,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 41289
+    },
+    {
+      "epoch": 4.235945835043085,
+      "grad_norm": 0.042979929596185684,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 41292
+    },
+    {
+      "epoch": 4.236253590480098,
+      "grad_norm": 0.03832445293664932,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 41295
+    },
+    {
+      "epoch": 4.236561345917111,
+      "grad_norm": 0.04340042173862457,
+      "learning_rate": 0.01,
+      "loss": 2.0301,
+      "step": 41298
+    },
+    {
+      "epoch": 4.236869101354124,
+      "grad_norm": 0.10501760244369507,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 41301
+    },
+    {
+      "epoch": 4.2371768567911365,
+      "grad_norm": 0.09777186065912247,
+      "learning_rate": 0.01,
+      "loss": 2.0283,
+      "step": 41304
+    },
+    {
+      "epoch": 4.237484612228149,
+      "grad_norm": 0.06624646484851837,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 41307
+    },
+    {
+      "epoch": 4.237792367665162,
+      "grad_norm": 0.055278148502111435,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 41310
+    },
+    {
+      "epoch": 4.238100123102175,
+      "grad_norm": 0.11712317168712616,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 41313
+    },
+    {
+      "epoch": 4.238407878539188,
+      "grad_norm": 0.06384153664112091,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 41316
+    },
+    {
+      "epoch": 4.2387156339762,
+      "grad_norm": 0.0486673079431057,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 41319
+    },
+    {
+      "epoch": 4.239023389413213,
+      "grad_norm": 0.04492638260126114,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 41322
+    },
+    {
+      "epoch": 4.239331144850226,
+      "grad_norm": 0.034185852855443954,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 41325
+    },
+    {
+      "epoch": 4.239638900287239,
+      "grad_norm": 0.04562927410006523,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 41328
+    },
+    {
+      "epoch": 4.239946655724252,
+      "grad_norm": 0.1411084234714508,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 41331
+    },
+    {
+      "epoch": 4.2402544111612634,
+      "grad_norm": 0.17035670578479767,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 41334
+    },
+    {
+      "epoch": 4.240562166598276,
+      "grad_norm": 0.06952272355556488,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 41337
+    },
+    {
+      "epoch": 4.240869922035289,
+      "grad_norm": 0.08368392288684845,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 41340
+    },
+    {
+      "epoch": 4.241177677472302,
+      "grad_norm": 0.04599005728960037,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 41343
+    },
+    {
+      "epoch": 4.241485432909315,
+      "grad_norm": 0.045579466968774796,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 41346
+    },
+    {
+      "epoch": 4.241793188346327,
+      "grad_norm": 0.049636196345090866,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 41349
+    },
+    {
+      "epoch": 4.24210094378334,
+      "grad_norm": 0.047992490231990814,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 41352
+    },
+    {
+      "epoch": 4.242408699220353,
+      "grad_norm": 0.05155181884765625,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 41355
+    },
+    {
+      "epoch": 4.242716454657366,
+      "grad_norm": 0.04805142432451248,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 41358
+    },
+    {
+      "epoch": 4.2430242100943785,
+      "grad_norm": 0.09783121943473816,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 41361
+    },
+    {
+      "epoch": 4.243331965531391,
+      "grad_norm": 0.05536942929029465,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 41364
+    },
+    {
+      "epoch": 4.243639720968404,
+      "grad_norm": 0.1335456669330597,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 41367
+    },
+    {
+      "epoch": 4.243947476405417,
+      "grad_norm": 0.05100298300385475,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 41370
+    },
+    {
+      "epoch": 4.24425523184243,
+      "grad_norm": 0.04158430173993111,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 41373
+    },
+    {
+      "epoch": 4.244562987279442,
+      "grad_norm": 0.03490443900227547,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 41376
+    },
+    {
+      "epoch": 4.244870742716454,
+      "grad_norm": 0.03529435768723488,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 41379
+    },
+    {
+      "epoch": 4.245178498153467,
+      "grad_norm": 0.03825982287526131,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 41382
+    },
+    {
+      "epoch": 4.24548625359048,
+      "grad_norm": 0.11641440540552139,
+      "learning_rate": 0.01,
+      "loss": 2.0487,
+      "step": 41385
+    },
+    {
+      "epoch": 4.245794009027493,
+      "grad_norm": 0.043827321380376816,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 41388
+    },
+    {
+      "epoch": 4.246101764464505,
+      "grad_norm": 0.060956139117479324,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 41391
+    },
+    {
+      "epoch": 4.246409519901518,
+      "grad_norm": 0.11849243938922882,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 41394
+    },
+    {
+      "epoch": 4.246717275338531,
+      "grad_norm": 0.08758600056171417,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 41397
+    },
+    {
+      "epoch": 4.247025030775544,
+      "grad_norm": 0.0848364606499672,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 41400
+    },
+    {
+      "epoch": 4.2473327862125565,
+      "grad_norm": 0.08009137958288193,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 41403
+    },
+    {
+      "epoch": 4.247640541649569,
+      "grad_norm": 0.06123640015721321,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 41406
+    },
+    {
+      "epoch": 4.247948297086582,
+      "grad_norm": 0.042272213846445084,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 41409
+    },
+    {
+      "epoch": 4.248256052523595,
+      "grad_norm": 0.051934532821178436,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 41412
+    },
+    {
+      "epoch": 4.248563807960608,
+      "grad_norm": 0.04881110414862633,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 41415
+    },
+    {
+      "epoch": 4.24887156339762,
+      "grad_norm": 0.03183252364397049,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 41418
+    },
+    {
+      "epoch": 4.249179318834633,
+      "grad_norm": 0.1036367192864418,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 41421
+    },
+    {
+      "epoch": 4.249487074271645,
+      "grad_norm": 0.08779038488864899,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 41424
+    },
+    {
+      "epoch": 4.249794829708658,
+      "grad_norm": 0.07684400677680969,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 41427
+    },
+    {
+      "epoch": 4.250102585145671,
+      "grad_norm": 0.08273176103830338,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 41430
+    },
+    {
+      "epoch": 4.250410340582683,
+      "grad_norm": 0.04036150500178337,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 41433
+    },
+    {
+      "epoch": 4.250718096019696,
+      "grad_norm": 0.04532729461789131,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 41436
+    },
+    {
+      "epoch": 4.251025851456709,
+      "grad_norm": 0.045257970690727234,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 41439
+    },
+    {
+      "epoch": 4.251333606893722,
+      "grad_norm": 0.11603814363479614,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 41442
+    },
+    {
+      "epoch": 4.2516413623307345,
+      "grad_norm": 0.07269345223903656,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 41445
+    },
+    {
+      "epoch": 4.251949117767747,
+      "grad_norm": 0.06193891167640686,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 41448
+    },
+    {
+      "epoch": 4.25225687320476,
+      "grad_norm": 0.09230732172727585,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 41451
+    },
+    {
+      "epoch": 4.252564628641773,
+      "grad_norm": 0.04278898611664772,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 41454
+    },
+    {
+      "epoch": 4.252872384078786,
+      "grad_norm": 0.04253445938229561,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 41457
+    },
+    {
+      "epoch": 4.253180139515798,
+      "grad_norm": 0.07245481014251709,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 41460
+    },
+    {
+      "epoch": 4.253487894952811,
+      "grad_norm": 0.10126817971467972,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 41463
+    },
+    {
+      "epoch": 4.253795650389824,
+      "grad_norm": 0.04528803005814552,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 41466
+    },
+    {
+      "epoch": 4.254103405826836,
+      "grad_norm": 0.04987543821334839,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 41469
+    },
+    {
+      "epoch": 4.254411161263849,
+      "grad_norm": 0.05521645396947861,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 41472
+    },
+    {
+      "epoch": 4.254718916700861,
+      "grad_norm": 0.06709643453359604,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 41475
+    },
+    {
+      "epoch": 4.255026672137874,
+      "grad_norm": 0.07900220155715942,
+      "learning_rate": 0.01,
+      "loss": 2.0332,
+      "step": 41478
+    },
+    {
+      "epoch": 4.255334427574887,
+      "grad_norm": 0.06128143519163132,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 41481
+    },
+    {
+      "epoch": 4.2556421830119,
+      "grad_norm": 0.044023312628269196,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 41484
+    },
+    {
+      "epoch": 4.255949938448913,
+      "grad_norm": 0.034501735121011734,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 41487
+    },
+    {
+      "epoch": 4.256257693885925,
+      "grad_norm": 0.08301688730716705,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 41490
+    },
+    {
+      "epoch": 4.256565449322938,
+      "grad_norm": 0.05008229240775108,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 41493
+    },
+    {
+      "epoch": 4.256873204759951,
+      "grad_norm": 0.08899793028831482,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 41496
+    },
+    {
+      "epoch": 4.257180960196964,
+      "grad_norm": 0.055265627801418304,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 41499
+    },
+    {
+      "epoch": 4.2574887156339765,
+      "grad_norm": 0.04912308603525162,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 41502
+    },
+    {
+      "epoch": 4.257796471070989,
+      "grad_norm": 0.057400181889534,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 41505
+    },
+    {
+      "epoch": 4.258104226508002,
+      "grad_norm": 0.13875073194503784,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 41508
+    },
+    {
+      "epoch": 4.258411981945015,
+      "grad_norm": 0.10987017303705215,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 41511
+    },
+    {
+      "epoch": 4.258719737382027,
+      "grad_norm": 0.06881477683782578,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 41514
+    },
+    {
+      "epoch": 4.2590274928190395,
+      "grad_norm": 0.06515093892812729,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 41517
+    },
+    {
+      "epoch": 4.259335248256052,
+      "grad_norm": 0.05030984431505203,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 41520
+    },
+    {
+      "epoch": 4.259643003693065,
+      "grad_norm": 0.061900872737169266,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 41523
+    },
+    {
+      "epoch": 4.259950759130078,
+      "grad_norm": 0.052526991814374924,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 41526
+    },
+    {
+      "epoch": 4.260258514567091,
+      "grad_norm": 0.07355368137359619,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 41529
+    },
+    {
+      "epoch": 4.260566270004103,
+      "grad_norm": 0.09191373735666275,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 41532
+    },
+    {
+      "epoch": 4.260874025441116,
+      "grad_norm": 0.04627171531319618,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 41535
+    },
+    {
+      "epoch": 4.261181780878129,
+      "grad_norm": 0.04639091342687607,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 41538
+    },
+    {
+      "epoch": 4.261489536315142,
+      "grad_norm": 0.0914679765701294,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 41541
+    },
+    {
+      "epoch": 4.2617972917521545,
+      "grad_norm": 0.08489304035902023,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 41544
+    },
+    {
+      "epoch": 4.262105047189167,
+      "grad_norm": 0.03301496058702469,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 41547
+    },
+    {
+      "epoch": 4.26241280262618,
+      "grad_norm": 0.08343908190727234,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 41550
+    },
+    {
+      "epoch": 4.262720558063193,
+      "grad_norm": 0.05923202261328697,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 41553
+    },
+    {
+      "epoch": 4.263028313500206,
+      "grad_norm": 0.044505976140499115,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 41556
+    },
+    {
+      "epoch": 4.2633360689372175,
+      "grad_norm": 0.05136754736304283,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 41559
+    },
+    {
+      "epoch": 4.26364382437423,
+      "grad_norm": 0.0830690786242485,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 41562
+    },
+    {
+      "epoch": 4.263951579811243,
+      "grad_norm": 0.11335346102714539,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 41565
+    },
+    {
+      "epoch": 4.264259335248256,
+      "grad_norm": 0.0629742443561554,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 41568
+    },
+    {
+      "epoch": 4.264567090685269,
+      "grad_norm": 0.09297124296426773,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 41571
+    },
+    {
+      "epoch": 4.264874846122281,
+      "grad_norm": 0.05171617120504379,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 41574
+    },
+    {
+      "epoch": 4.265182601559294,
+      "grad_norm": 0.09433134645223618,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 41577
+    },
+    {
+      "epoch": 4.265490356996307,
+      "grad_norm": 0.058073755353689194,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 41580
+    },
+    {
+      "epoch": 4.26579811243332,
+      "grad_norm": 0.037740156054496765,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 41583
+    },
+    {
+      "epoch": 4.2661058678703325,
+      "grad_norm": 0.12522292137145996,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 41586
+    },
+    {
+      "epoch": 4.266413623307345,
+      "grad_norm": 0.07263893634080887,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 41589
+    },
+    {
+      "epoch": 4.266721378744358,
+      "grad_norm": 0.08800274133682251,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 41592
+    },
+    {
+      "epoch": 4.267029134181371,
+      "grad_norm": 0.04018472880125046,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 41595
+    },
+    {
+      "epoch": 4.267336889618384,
+      "grad_norm": 0.04245147481560707,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 41598
+    },
+    {
+      "epoch": 4.267644645055396,
+      "grad_norm": 0.03626253455877304,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 41601
+    },
+    {
+      "epoch": 4.267952400492408,
+      "grad_norm": 0.1148434430360794,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 41604
+    },
+    {
+      "epoch": 4.268260155929421,
+      "grad_norm": 0.03992197662591934,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 41607
+    },
+    {
+      "epoch": 4.268567911366434,
+      "grad_norm": 0.07781147211790085,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 41610
+    },
+    {
+      "epoch": 4.268875666803447,
+      "grad_norm": 0.08838597685098648,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 41613
+    },
+    {
+      "epoch": 4.269183422240459,
+      "grad_norm": 0.07930974662303925,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 41616
+    },
+    {
+      "epoch": 4.269491177677472,
+      "grad_norm": 0.08094513416290283,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 41619
+    },
+    {
+      "epoch": 4.269798933114485,
+      "grad_norm": 0.06766554713249207,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 41622
+    },
+    {
+      "epoch": 4.270106688551498,
+      "grad_norm": 0.07384651899337769,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 41625
+    },
+    {
+      "epoch": 4.2704144439885106,
+      "grad_norm": 0.043780550360679626,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 41628
+    },
+    {
+      "epoch": 4.270722199425523,
+      "grad_norm": 0.06705281883478165,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 41631
+    },
+    {
+      "epoch": 4.271029954862536,
+      "grad_norm": 0.051532238721847534,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 41634
+    },
+    {
+      "epoch": 4.271337710299549,
+      "grad_norm": 0.09668219834566116,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 41637
+    },
+    {
+      "epoch": 4.271645465736562,
+      "grad_norm": 0.0774909183382988,
+      "learning_rate": 0.01,
+      "loss": 2.0307,
+      "step": 41640
+    },
+    {
+      "epoch": 4.2719532211735745,
+      "grad_norm": 0.11848165839910507,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 41643
+    },
+    {
+      "epoch": 4.272260976610587,
+      "grad_norm": 0.09055308997631073,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 41646
+    },
+    {
+      "epoch": 4.272568732047599,
+      "grad_norm": 0.06557660549879074,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 41649
+    },
+    {
+      "epoch": 4.272876487484612,
+      "grad_norm": 0.055074457079172134,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 41652
+    },
+    {
+      "epoch": 4.273184242921625,
+      "grad_norm": 0.07604243606328964,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 41655
+    },
+    {
+      "epoch": 4.2734919983586375,
+      "grad_norm": 0.07098755240440369,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 41658
+    },
+    {
+      "epoch": 4.27379975379565,
+      "grad_norm": 0.11538244038820267,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 41661
+    },
+    {
+      "epoch": 4.274107509232663,
+      "grad_norm": 0.03773383051156998,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 41664
+    },
+    {
+      "epoch": 4.274415264669676,
+      "grad_norm": 0.03163829445838928,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 41667
+    },
+    {
+      "epoch": 4.274723020106689,
+      "grad_norm": 0.0678187757730484,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 41670
+    },
+    {
+      "epoch": 4.275030775543701,
+      "grad_norm": 0.0774695947766304,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 41673
+    },
+    {
+      "epoch": 4.275338530980714,
+      "grad_norm": 0.05830325931310654,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 41676
+    },
+    {
+      "epoch": 4.275646286417727,
+      "grad_norm": 0.11642345786094666,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 41679
+    },
+    {
+      "epoch": 4.27595404185474,
+      "grad_norm": 0.12688972055912018,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 41682
+    },
+    {
+      "epoch": 4.2762617972917525,
+      "grad_norm": 0.11521381884813309,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 41685
+    },
+    {
+      "epoch": 4.276569552728765,
+      "grad_norm": 0.1203092560172081,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 41688
+    },
+    {
+      "epoch": 4.276877308165778,
+      "grad_norm": 0.09710691124200821,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 41691
+    },
+    {
+      "epoch": 4.27718506360279,
+      "grad_norm": 0.07945617288351059,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 41694
+    },
+    {
+      "epoch": 4.277492819039803,
+      "grad_norm": 0.07407965511083603,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 41697
+    },
+    {
+      "epoch": 4.2778005744768155,
+      "grad_norm": 0.06321244686841965,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 41700
+    },
+    {
+      "epoch": 4.278108329913828,
+      "grad_norm": 0.056950200349092484,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 41703
+    },
+    {
+      "epoch": 4.278416085350841,
+      "grad_norm": 0.039905134588479996,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 41706
+    },
+    {
+      "epoch": 4.278723840787854,
+      "grad_norm": 0.05050415173172951,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 41709
+    },
+    {
+      "epoch": 4.279031596224867,
+      "grad_norm": 0.05510304123163223,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 41712
+    },
+    {
+      "epoch": 4.279339351661879,
+      "grad_norm": 0.05586516857147217,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 41715
+    },
+    {
+      "epoch": 4.279647107098892,
+      "grad_norm": 0.05404721572995186,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 41718
+    },
+    {
+      "epoch": 4.279954862535905,
+      "grad_norm": 0.04340701550245285,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 41721
+    },
+    {
+      "epoch": 4.280262617972918,
+      "grad_norm": 0.053679756820201874,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 41724
+    },
+    {
+      "epoch": 4.2805703734099305,
+      "grad_norm": 0.1286962479352951,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 41727
+    },
+    {
+      "epoch": 4.280878128846943,
+      "grad_norm": 0.07245181500911713,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 41730
+    },
+    {
+      "epoch": 4.281185884283956,
+      "grad_norm": 0.08518802374601364,
+      "learning_rate": 0.01,
+      "loss": 2.0305,
+      "step": 41733
+    },
+    {
+      "epoch": 4.281493639720969,
+      "grad_norm": 0.07493653893470764,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 41736
+    },
+    {
+      "epoch": 4.281801395157981,
+      "grad_norm": 0.05031900852918625,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 41739
+    },
+    {
+      "epoch": 4.2821091505949935,
+      "grad_norm": 0.07814056426286697,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 41742
+    },
+    {
+      "epoch": 4.282416906032006,
+      "grad_norm": 0.047507043927907944,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 41745
+    },
+    {
+      "epoch": 4.282724661469019,
+      "grad_norm": 0.06547687947750092,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 41748
+    },
+    {
+      "epoch": 4.283032416906032,
+      "grad_norm": 0.059194277971982956,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 41751
+    },
+    {
+      "epoch": 4.283340172343045,
+      "grad_norm": 0.04526950791478157,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 41754
+    },
+    {
+      "epoch": 4.283647927780057,
+      "grad_norm": 0.031040268018841743,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 41757
+    },
+    {
+      "epoch": 4.28395568321707,
+      "grad_norm": 0.09988913685083389,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 41760
+    },
+    {
+      "epoch": 4.284263438654083,
+      "grad_norm": 0.04122784361243248,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 41763
+    },
+    {
+      "epoch": 4.284571194091096,
+      "grad_norm": 0.055508848279714584,
+      "learning_rate": 0.01,
+      "loss": 2.0457,
+      "step": 41766
+    },
+    {
+      "epoch": 4.2848789495281085,
+      "grad_norm": 0.10151500999927521,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 41769
+    },
+    {
+      "epoch": 4.285186704965121,
+      "grad_norm": 0.041592396795749664,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 41772
+    },
+    {
+      "epoch": 4.285494460402134,
+      "grad_norm": 0.06108968332409859,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 41775
+    },
+    {
+      "epoch": 4.285802215839147,
+      "grad_norm": 0.06336821615695953,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 41778
+    },
+    {
+      "epoch": 4.28610997127616,
+      "grad_norm": 0.03348274528980255,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 41781
+    },
+    {
+      "epoch": 4.286417726713172,
+      "grad_norm": 0.04698178544640541,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 41784
+    },
+    {
+      "epoch": 4.286725482150184,
+      "grad_norm": 0.04394696652889252,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 41787
+    },
+    {
+      "epoch": 4.287033237587197,
+      "grad_norm": 0.15938106179237366,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 41790
+    },
+    {
+      "epoch": 4.28734099302421,
+      "grad_norm": 0.08870629966259003,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 41793
+    },
+    {
+      "epoch": 4.287648748461223,
+      "grad_norm": 0.051104262471199036,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 41796
+    },
+    {
+      "epoch": 4.2879565038982355,
+      "grad_norm": 0.03226442262530327,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 41799
+    },
+    {
+      "epoch": 4.288264259335248,
+      "grad_norm": 0.033240124583244324,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 41802
+    },
+    {
+      "epoch": 4.288572014772261,
+      "grad_norm": 0.04799087345600128,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 41805
+    },
+    {
+      "epoch": 4.288879770209274,
+      "grad_norm": 0.04805191978812218,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 41808
+    },
+    {
+      "epoch": 4.289187525646287,
+      "grad_norm": 0.10758639872074127,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 41811
+    },
+    {
+      "epoch": 4.289495281083299,
+      "grad_norm": 0.06141864135861397,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 41814
+    },
+    {
+      "epoch": 4.289803036520312,
+      "grad_norm": 0.057809922844171524,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 41817
+    },
+    {
+      "epoch": 4.290110791957325,
+      "grad_norm": 0.06831464916467667,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 41820
+    },
+    {
+      "epoch": 4.290418547394338,
+      "grad_norm": 0.08271254599094391,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 41823
+    },
+    {
+      "epoch": 4.2907263028313505,
+      "grad_norm": 0.04244139790534973,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 41826
+    },
+    {
+      "epoch": 4.291034058268362,
+      "grad_norm": 0.04482168331742287,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 41829
+    },
+    {
+      "epoch": 4.291341813705375,
+      "grad_norm": 0.04742707312107086,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 41832
+    },
+    {
+      "epoch": 4.291649569142388,
+      "grad_norm": 0.12863384187221527,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 41835
+    },
+    {
+      "epoch": 4.291957324579401,
+      "grad_norm": 0.12212485820055008,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 41838
+    },
+    {
+      "epoch": 4.2922650800164135,
+      "grad_norm": 0.0465460866689682,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 41841
+    },
+    {
+      "epoch": 4.292572835453426,
+      "grad_norm": 0.052605509757995605,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 41844
+    },
+    {
+      "epoch": 4.292880590890439,
+      "grad_norm": 0.046292662620544434,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 41847
+    },
+    {
+      "epoch": 4.293188346327452,
+      "grad_norm": 0.04217211529612541,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 41850
+    },
+    {
+      "epoch": 4.293496101764465,
+      "grad_norm": 0.08655425906181335,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 41853
+    },
+    {
+      "epoch": 4.293803857201477,
+      "grad_norm": 0.0645185261964798,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 41856
+    },
+    {
+      "epoch": 4.29411161263849,
+      "grad_norm": 0.06806469708681107,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 41859
+    },
+    {
+      "epoch": 4.294419368075503,
+      "grad_norm": 0.05178931728005409,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 41862
+    },
+    {
+      "epoch": 4.294727123512516,
+      "grad_norm": 0.0440845787525177,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 41865
+    },
+    {
+      "epoch": 4.2950348789495285,
+      "grad_norm": 0.04800264537334442,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 41868
+    },
+    {
+      "epoch": 4.295342634386541,
+      "grad_norm": 0.03780617192387581,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 41871
+    },
+    {
+      "epoch": 4.295650389823553,
+      "grad_norm": 0.0536942295730114,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 41874
+    },
+    {
+      "epoch": 4.295958145260566,
+      "grad_norm": 0.09988999366760254,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 41877
+    },
+    {
+      "epoch": 4.296265900697579,
+      "grad_norm": 0.06863691657781601,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 41880
+    },
+    {
+      "epoch": 4.2965736561345915,
+      "grad_norm": 0.04894992709159851,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 41883
+    },
+    {
+      "epoch": 4.296881411571604,
+      "grad_norm": 0.041689950972795486,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 41886
+    },
+    {
+      "epoch": 4.297189167008617,
+      "grad_norm": 0.10208067297935486,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 41889
+    },
+    {
+      "epoch": 4.29749692244563,
+      "grad_norm": 0.030840028077363968,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 41892
+    },
+    {
+      "epoch": 4.297804677882643,
+      "grad_norm": 0.04242958873510361,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 41895
+    },
+    {
+      "epoch": 4.298112433319655,
+      "grad_norm": 0.11578787118196487,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 41898
+    },
+    {
+      "epoch": 4.298420188756668,
+      "grad_norm": 0.059238482266664505,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 41901
+    },
+    {
+      "epoch": 4.298727944193681,
+      "grad_norm": 0.03906838968396187,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 41904
+    },
+    {
+      "epoch": 4.299035699630694,
+      "grad_norm": 0.06461112201213837,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 41907
+    },
+    {
+      "epoch": 4.2993434550677065,
+      "grad_norm": 0.1491340696811676,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 41910
+    },
+    {
+      "epoch": 4.299651210504719,
+      "grad_norm": 0.06621528416872025,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 41913
+    },
+    {
+      "epoch": 4.299958965941732,
+      "grad_norm": 0.06322132796049118,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 41916
+    },
+    {
+      "epoch": 4.300266721378744,
+      "grad_norm": 0.04899013042449951,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 41919
+    },
+    {
+      "epoch": 4.300574476815757,
+      "grad_norm": 0.04941810667514801,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 41922
+    },
+    {
+      "epoch": 4.3008822322527696,
+      "grad_norm": 0.04341074824333191,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 41925
+    },
+    {
+      "epoch": 4.301189987689782,
+      "grad_norm": 0.08827680349349976,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 41928
+    },
+    {
+      "epoch": 4.301497743126795,
+      "grad_norm": 0.05221998319029808,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 41931
+    },
+    {
+      "epoch": 4.301805498563808,
+      "grad_norm": 0.10630286484956741,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 41934
+    },
+    {
+      "epoch": 4.302113254000821,
+      "grad_norm": 0.043958306312561035,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 41937
+    },
+    {
+      "epoch": 4.3024210094378335,
+      "grad_norm": 0.057621411979198456,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 41940
+    },
+    {
+      "epoch": 4.302728764874846,
+      "grad_norm": 0.09467203170061111,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 41943
+    },
+    {
+      "epoch": 4.303036520311859,
+      "grad_norm": 0.06720858812332153,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 41946
+    },
+    {
+      "epoch": 4.303344275748872,
+      "grad_norm": 0.06861304491758347,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 41949
+    },
+    {
+      "epoch": 4.303652031185885,
+      "grad_norm": 0.038296621292829514,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 41952
+    },
+    {
+      "epoch": 4.303959786622897,
+      "grad_norm": 0.05808268114924431,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 41955
+    },
+    {
+      "epoch": 4.30426754205991,
+      "grad_norm": 0.05521778017282486,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 41958
+    },
+    {
+      "epoch": 4.304575297496923,
+      "grad_norm": 0.05509946867823601,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 41961
+    },
+    {
+      "epoch": 4.304883052933935,
+      "grad_norm": 0.04729253426194191,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 41964
+    },
+    {
+      "epoch": 4.305190808370948,
+      "grad_norm": 0.06389732658863068,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 41967
+    },
+    {
+      "epoch": 4.30549856380796,
+      "grad_norm": 0.04609325900673866,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 41970
+    },
+    {
+      "epoch": 4.305806319244973,
+      "grad_norm": 0.06793837994337082,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 41973
+    },
+    {
+      "epoch": 4.306114074681986,
+      "grad_norm": 0.0791841521859169,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 41976
+    },
+    {
+      "epoch": 4.306421830118999,
+      "grad_norm": 0.07721048593521118,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 41979
+    },
+    {
+      "epoch": 4.3067295855560115,
+      "grad_norm": 0.04971477389335632,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 41982
+    },
+    {
+      "epoch": 4.307037340993024,
+      "grad_norm": 0.11230115592479706,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 41985
+    },
+    {
+      "epoch": 4.307345096430037,
+      "grad_norm": 0.08187992870807648,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 41988
+    },
+    {
+      "epoch": 4.30765285186705,
+      "grad_norm": 0.07407639175653458,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 41991
+    },
+    {
+      "epoch": 4.307960607304063,
+      "grad_norm": 0.051022280007600784,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 41994
+    },
+    {
+      "epoch": 4.308268362741075,
+      "grad_norm": 0.041618864983320236,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 41997
+    },
+    {
+      "epoch": 4.308576118178088,
+      "grad_norm": 0.0507926307618618,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 42000
+    },
+    {
+      "epoch": 4.308883873615101,
+      "grad_norm": 0.07467113435268402,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 42003
+    },
+    {
+      "epoch": 4.309191629052114,
+      "grad_norm": 0.06563274562358856,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 42006
+    },
+    {
+      "epoch": 4.309499384489126,
+      "grad_norm": 0.05920867621898651,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 42009
+    },
+    {
+      "epoch": 4.309807139926138,
+      "grad_norm": 0.12882298231124878,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 42012
+    },
+    {
+      "epoch": 4.310114895363151,
+      "grad_norm": 0.07217466831207275,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 42015
+    },
+    {
+      "epoch": 4.310422650800164,
+      "grad_norm": 0.06030949205160141,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 42018
+    },
+    {
+      "epoch": 4.310730406237177,
+      "grad_norm": 0.04477029666304588,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 42021
+    },
+    {
+      "epoch": 4.3110381616741895,
+      "grad_norm": 0.042469825595617294,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 42024
+    },
+    {
+      "epoch": 4.311345917111202,
+      "grad_norm": 0.061194293200969696,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 42027
+    },
+    {
+      "epoch": 4.311653672548215,
+      "grad_norm": 0.07249957323074341,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 42030
+    },
+    {
+      "epoch": 4.311961427985228,
+      "grad_norm": 0.1172809898853302,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 42033
+    },
+    {
+      "epoch": 4.312269183422241,
+      "grad_norm": 0.05327514931559563,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 42036
+    },
+    {
+      "epoch": 4.312576938859253,
+      "grad_norm": 0.06902623176574707,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 42039
+    },
+    {
+      "epoch": 4.312884694296266,
+      "grad_norm": 0.039753060787916183,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 42042
+    },
+    {
+      "epoch": 4.313192449733279,
+      "grad_norm": 0.047289080917835236,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 42045
+    },
+    {
+      "epoch": 4.313500205170292,
+      "grad_norm": 0.11222794651985168,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 42048
+    },
+    {
+      "epoch": 4.313807960607304,
+      "grad_norm": 0.09836508333683014,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 42051
+    },
+    {
+      "epoch": 4.314115716044316,
+      "grad_norm": 0.06988558173179626,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 42054
+    },
+    {
+      "epoch": 4.314423471481329,
+      "grad_norm": 0.05915610492229462,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 42057
+    },
+    {
+      "epoch": 4.314731226918342,
+      "grad_norm": 0.14300785958766937,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 42060
+    },
+    {
+      "epoch": 4.315038982355355,
+      "grad_norm": 0.055672645568847656,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 42063
+    },
+    {
+      "epoch": 4.3153467377923675,
+      "grad_norm": 0.07753617316484451,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 42066
+    },
+    {
+      "epoch": 4.31565449322938,
+      "grad_norm": 0.04788988083600998,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 42069
+    },
+    {
+      "epoch": 4.315962248666393,
+      "grad_norm": 0.04761343449354172,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 42072
+    },
+    {
+      "epoch": 4.316270004103406,
+      "grad_norm": 0.051400624215602875,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 42075
+    },
+    {
+      "epoch": 4.316577759540419,
+      "grad_norm": 0.050774481147527695,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 42078
+    },
+    {
+      "epoch": 4.3168855149774314,
+      "grad_norm": 0.08724766224622726,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 42081
+    },
+    {
+      "epoch": 4.317193270414444,
+      "grad_norm": 0.07128842920064926,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 42084
+    },
+    {
+      "epoch": 4.317501025851457,
+      "grad_norm": 0.16834615170955658,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 42087
+    },
+    {
+      "epoch": 4.31780878128847,
+      "grad_norm": 0.07254631817340851,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 42090
+    },
+    {
+      "epoch": 4.318116536725482,
+      "grad_norm": 0.0355360321700573,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 42093
+    },
+    {
+      "epoch": 4.3184242921624945,
+      "grad_norm": 0.03480659797787666,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 42096
+    },
+    {
+      "epoch": 4.318732047599507,
+      "grad_norm": 0.04848271980881691,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 42099
+    },
+    {
+      "epoch": 4.31903980303652,
+      "grad_norm": 0.047225676476955414,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 42102
+    },
+    {
+      "epoch": 4.319347558473533,
+      "grad_norm": 0.07466583698987961,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 42105
+    },
+    {
+      "epoch": 4.319655313910546,
+      "grad_norm": 0.03580273687839508,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 42108
+    },
+    {
+      "epoch": 4.319963069347558,
+      "grad_norm": 0.0513874776661396,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 42111
+    },
+    {
+      "epoch": 4.320270824784571,
+      "grad_norm": 0.08838716894388199,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 42114
+    },
+    {
+      "epoch": 4.320578580221584,
+      "grad_norm": 0.048829689621925354,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 42117
+    },
+    {
+      "epoch": 4.320886335658597,
+      "grad_norm": 0.0572056770324707,
+      "learning_rate": 0.01,
+      "loss": 2.0366,
+      "step": 42120
+    },
+    {
+      "epoch": 4.3211940910956095,
+      "grad_norm": 0.0953284278512001,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 42123
+    },
+    {
+      "epoch": 4.321501846532622,
+      "grad_norm": 0.08970024436712265,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 42126
+    },
+    {
+      "epoch": 4.321809601969635,
+      "grad_norm": 0.03790782392024994,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 42129
+    },
+    {
+      "epoch": 4.322117357406648,
+      "grad_norm": 0.11770451813936234,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 42132
+    },
+    {
+      "epoch": 4.322425112843661,
+      "grad_norm": 0.08634433895349503,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 42135
+    },
+    {
+      "epoch": 4.3227328682806725,
+      "grad_norm": 0.08693015575408936,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 42138
+    },
+    {
+      "epoch": 4.323040623717685,
+      "grad_norm": 0.07767558842897415,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 42141
+    },
+    {
+      "epoch": 4.323348379154698,
+      "grad_norm": 0.03783591464161873,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 42144
+    },
+    {
+      "epoch": 4.323656134591711,
+      "grad_norm": 0.036055248230695724,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 42147
+    },
+    {
+      "epoch": 4.323963890028724,
+      "grad_norm": 0.056252893060445786,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 42150
+    },
+    {
+      "epoch": 4.324271645465736,
+      "grad_norm": 0.07847966253757477,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 42153
+    },
+    {
+      "epoch": 4.324579400902749,
+      "grad_norm": 0.07340630888938904,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 42156
+    },
+    {
+      "epoch": 4.324887156339762,
+      "grad_norm": 0.0631684884428978,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 42159
+    },
+    {
+      "epoch": 4.325194911776775,
+      "grad_norm": 0.03661903738975525,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 42162
+    },
+    {
+      "epoch": 4.3255026672137875,
+      "grad_norm": 0.06492697447538376,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 42165
+    },
+    {
+      "epoch": 4.3258104226508,
+      "grad_norm": 0.07288751751184464,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 42168
+    },
+    {
+      "epoch": 4.326118178087813,
+      "grad_norm": 0.04615149274468422,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 42171
+    },
+    {
+      "epoch": 4.326425933524826,
+      "grad_norm": 0.06735312938690186,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 42174
+    },
+    {
+      "epoch": 4.326733688961839,
+      "grad_norm": 0.05109792947769165,
+      "learning_rate": 0.01,
+      "loss": 2.0274,
+      "step": 42177
+    },
+    {
+      "epoch": 4.327041444398851,
+      "grad_norm": 0.047006141394376755,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 42180
+    },
+    {
+      "epoch": 4.327349199835863,
+      "grad_norm": 0.11803962290287018,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 42183
+    },
+    {
+      "epoch": 4.327656955272876,
+      "grad_norm": 0.06282762438058853,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 42186
+    },
+    {
+      "epoch": 4.327964710709889,
+      "grad_norm": 0.07451837509870529,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 42189
+    },
+    {
+      "epoch": 4.328272466146902,
+      "grad_norm": 0.05453307554125786,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 42192
+    },
+    {
+      "epoch": 4.328580221583914,
+      "grad_norm": 0.03320490941405296,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 42195
+    },
+    {
+      "epoch": 4.328887977020927,
+      "grad_norm": 0.03890087082982063,
+      "learning_rate": 0.01,
+      "loss": 2.0277,
+      "step": 42198
+    },
+    {
+      "epoch": 4.32919573245794,
+      "grad_norm": 0.04567558318376541,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 42201
+    },
+    {
+      "epoch": 4.329503487894953,
+      "grad_norm": 0.11574912816286087,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 42204
+    },
+    {
+      "epoch": 4.3298112433319655,
+      "grad_norm": 0.1619519144296646,
+      "learning_rate": 0.01,
+      "loss": 2.0264,
+      "step": 42207
+    },
+    {
+      "epoch": 4.330118998768978,
+      "grad_norm": 0.07246614992618561,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 42210
+    },
+    {
+      "epoch": 4.330426754205991,
+      "grad_norm": 0.06830492615699768,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 42213
+    },
+    {
+      "epoch": 4.330734509643004,
+      "grad_norm": 0.03900605067610741,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 42216
+    },
+    {
+      "epoch": 4.331042265080017,
+      "grad_norm": 0.04035233333706856,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 42219
+    },
+    {
+      "epoch": 4.331350020517029,
+      "grad_norm": 0.08681115508079529,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 42222
+    },
+    {
+      "epoch": 4.331657775954042,
+      "grad_norm": 0.10521326214075089,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 42225
+    },
+    {
+      "epoch": 4.331965531391054,
+      "grad_norm": 0.05561526492238045,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 42228
+    },
+    {
+      "epoch": 4.332273286828067,
+      "grad_norm": 0.10138184577226639,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 42231
+    },
+    {
+      "epoch": 4.33258104226508,
+      "grad_norm": 0.05151226371526718,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 42234
+    },
+    {
+      "epoch": 4.3328887977020925,
+      "grad_norm": 0.04308367893099785,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 42237
+    },
+    {
+      "epoch": 4.333196553139105,
+      "grad_norm": 0.09749255329370499,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 42240
+    },
+    {
+      "epoch": 4.333504308576118,
+      "grad_norm": 0.078557088971138,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 42243
+    },
+    {
+      "epoch": 4.333812064013131,
+      "grad_norm": 0.05035284161567688,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 42246
+    },
+    {
+      "epoch": 4.334119819450144,
+      "grad_norm": 0.08703385293483734,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 42249
+    },
+    {
+      "epoch": 4.334427574887156,
+      "grad_norm": 0.06751354783773422,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 42252
+    },
+    {
+      "epoch": 4.334735330324169,
+      "grad_norm": 0.06264498084783554,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 42255
+    },
+    {
+      "epoch": 4.335043085761182,
+      "grad_norm": 0.08927709609270096,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 42258
+    },
+    {
+      "epoch": 4.335350841198195,
+      "grad_norm": 0.0882975161075592,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 42261
+    },
+    {
+      "epoch": 4.3356585966352075,
+      "grad_norm": 0.08568523824214935,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 42264
+    },
+    {
+      "epoch": 4.33596635207222,
+      "grad_norm": 0.10143941640853882,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 42267
+    },
+    {
+      "epoch": 4.336274107509233,
+      "grad_norm": 0.04442453384399414,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 42270
+    },
+    {
+      "epoch": 4.336581862946245,
+      "grad_norm": 0.04202549159526825,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 42273
+    },
+    {
+      "epoch": 4.336889618383258,
+      "grad_norm": 0.07024986296892166,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 42276
+    },
+    {
+      "epoch": 4.3371973738202705,
+      "grad_norm": 0.0714881494641304,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 42279
+    },
+    {
+      "epoch": 4.337505129257283,
+      "grad_norm": 0.05677810311317444,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 42282
+    },
+    {
+      "epoch": 4.337812884694296,
+      "grad_norm": 0.08475086092948914,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 42285
+    },
+    {
+      "epoch": 4.338120640131309,
+      "grad_norm": 0.05561790242791176,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 42288
+    },
+    {
+      "epoch": 4.338428395568322,
+      "grad_norm": 0.0845925584435463,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 42291
+    },
+    {
+      "epoch": 4.338736151005334,
+      "grad_norm": 0.07267442345619202,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 42294
+    },
+    {
+      "epoch": 4.339043906442347,
+      "grad_norm": 0.06486202031373978,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 42297
+    },
+    {
+      "epoch": 4.33935166187936,
+      "grad_norm": 0.08498860895633698,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 42300
+    },
+    {
+      "epoch": 4.339659417316373,
+      "grad_norm": 0.05737947300076485,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 42303
+    },
+    {
+      "epoch": 4.3399671727533855,
+      "grad_norm": 0.1146705150604248,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 42306
+    },
+    {
+      "epoch": 4.340274928190398,
+      "grad_norm": 0.06821808218955994,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 42309
+    },
+    {
+      "epoch": 4.340582683627411,
+      "grad_norm": 0.06554093956947327,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 42312
+    },
+    {
+      "epoch": 4.340890439064424,
+      "grad_norm": 0.055730611085891724,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 42315
+    },
+    {
+      "epoch": 4.341198194501436,
+      "grad_norm": 0.12713190913200378,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 42318
+    },
+    {
+      "epoch": 4.3415059499384485,
+      "grad_norm": 0.04995013028383255,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 42321
+    },
+    {
+      "epoch": 4.341813705375461,
+      "grad_norm": 0.05785641819238663,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 42324
+    },
+    {
+      "epoch": 4.342121460812474,
+      "grad_norm": 0.06933962553739548,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 42327
+    },
+    {
+      "epoch": 4.342429216249487,
+      "grad_norm": 0.08787462115287781,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 42330
+    },
+    {
+      "epoch": 4.3427369716865,
+      "grad_norm": 0.03759448602795601,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 42333
+    },
+    {
+      "epoch": 4.343044727123512,
+      "grad_norm": 0.08787883818149567,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 42336
+    },
+    {
+      "epoch": 4.343352482560525,
+      "grad_norm": 0.07424912601709366,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 42339
+    },
+    {
+      "epoch": 4.343660237997538,
+      "grad_norm": 0.08268291503190994,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 42342
+    },
+    {
+      "epoch": 4.343967993434551,
+      "grad_norm": 0.09832719713449478,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 42345
+    },
+    {
+      "epoch": 4.3442757488715635,
+      "grad_norm": 0.08465267717838287,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 42348
+    },
+    {
+      "epoch": 4.344583504308576,
+      "grad_norm": 0.08722463995218277,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 42351
+    },
+    {
+      "epoch": 4.344891259745589,
+      "grad_norm": 0.03291288763284683,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 42354
+    },
+    {
+      "epoch": 4.345199015182602,
+      "grad_norm": 0.05389249697327614,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 42357
+    },
+    {
+      "epoch": 4.345506770619615,
+      "grad_norm": 0.057934414595365524,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 42360
+    },
+    {
+      "epoch": 4.3458145260566265,
+      "grad_norm": 0.09527049958705902,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 42363
+    },
+    {
+      "epoch": 4.346122281493639,
+      "grad_norm": 0.09653705358505249,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 42366
+    },
+    {
+      "epoch": 4.346430036930652,
+      "grad_norm": 0.050867971032857895,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 42369
+    },
+    {
+      "epoch": 4.346737792367665,
+      "grad_norm": 0.17523349821567535,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 42372
+    },
+    {
+      "epoch": 4.347045547804678,
+      "grad_norm": 0.04011973738670349,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 42375
+    },
+    {
+      "epoch": 4.3473533032416904,
+      "grad_norm": 0.0476989708840847,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 42378
+    },
+    {
+      "epoch": 4.347661058678703,
+      "grad_norm": 0.062353190034627914,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 42381
+    },
+    {
+      "epoch": 4.347968814115716,
+      "grad_norm": 0.03900708258152008,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 42384
+    },
+    {
+      "epoch": 4.348276569552729,
+      "grad_norm": 0.05116381496191025,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 42387
+    },
+    {
+      "epoch": 4.348584324989742,
+      "grad_norm": 0.0740213617682457,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 42390
+    },
+    {
+      "epoch": 4.348892080426754,
+      "grad_norm": 0.04419323056936264,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 42393
+    },
+    {
+      "epoch": 4.349199835863767,
+      "grad_norm": 0.09228520840406418,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 42396
+    },
+    {
+      "epoch": 4.34950759130078,
+      "grad_norm": 0.06969837844371796,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 42399
+    },
+    {
+      "epoch": 4.349815346737793,
+      "grad_norm": 0.06493277847766876,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 42402
+    },
+    {
+      "epoch": 4.3501231021748055,
+      "grad_norm": 0.10470300912857056,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 42405
+    },
+    {
+      "epoch": 4.350430857611817,
+      "grad_norm": 0.055797114968299866,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 42408
+    },
+    {
+      "epoch": 4.35073861304883,
+      "grad_norm": 0.08295798301696777,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 42411
+    },
+    {
+      "epoch": 4.351046368485843,
+      "grad_norm": 0.03319832682609558,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 42414
+    },
+    {
+      "epoch": 4.351354123922856,
+      "grad_norm": 0.08333670347929001,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 42417
+    },
+    {
+      "epoch": 4.3516618793598685,
+      "grad_norm": 0.15636205673217773,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 42420
+    },
+    {
+      "epoch": 4.351969634796881,
+      "grad_norm": 0.04145883768796921,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 42423
+    },
+    {
+      "epoch": 4.352277390233894,
+      "grad_norm": 0.06319641321897507,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 42426
+    },
+    {
+      "epoch": 4.352585145670907,
+      "grad_norm": 0.10287382453680038,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 42429
+    },
+    {
+      "epoch": 4.35289290110792,
+      "grad_norm": 0.05073247849941254,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 42432
+    },
+    {
+      "epoch": 4.353200656544932,
+      "grad_norm": 0.05621569603681564,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 42435
+    },
+    {
+      "epoch": 4.353508411981945,
+      "grad_norm": 0.044255949556827545,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 42438
+    },
+    {
+      "epoch": 4.353816167418958,
+      "grad_norm": 0.09396500885486603,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 42441
+    },
+    {
+      "epoch": 4.354123922855971,
+      "grad_norm": 0.04211452975869179,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 42444
+    },
+    {
+      "epoch": 4.3544316782929835,
+      "grad_norm": 0.09612414985895157,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 42447
+    },
+    {
+      "epoch": 4.354739433729996,
+      "grad_norm": 0.07747933268547058,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 42450
+    },
+    {
+      "epoch": 4.355047189167008,
+      "grad_norm": 0.06796756386756897,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 42453
+    },
+    {
+      "epoch": 4.355354944604021,
+      "grad_norm": 0.052594736218452454,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 42456
+    },
+    {
+      "epoch": 4.355662700041034,
+      "grad_norm": 0.09094807505607605,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 42459
+    },
+    {
+      "epoch": 4.3559704554780465,
+      "grad_norm": 0.06846702843904495,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 42462
+    },
+    {
+      "epoch": 4.356278210915059,
+      "grad_norm": 0.059900566935539246,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 42465
+    },
+    {
+      "epoch": 4.356585966352072,
+      "grad_norm": 0.07737037539482117,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 42468
+    },
+    {
+      "epoch": 4.356893721789085,
+      "grad_norm": 0.10885298252105713,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 42471
+    },
+    {
+      "epoch": 4.357201477226098,
+      "grad_norm": 0.12740680575370789,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 42474
+    },
+    {
+      "epoch": 4.35750923266311,
+      "grad_norm": 0.05705080181360245,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 42477
+    },
+    {
+      "epoch": 4.357816988100123,
+      "grad_norm": 0.047689832746982574,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 42480
+    },
+    {
+      "epoch": 4.358124743537136,
+      "grad_norm": 0.06346622854471207,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 42483
+    },
+    {
+      "epoch": 4.358432498974149,
+      "grad_norm": 0.08776500821113586,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 42486
+    },
+    {
+      "epoch": 4.3587402544111615,
+      "grad_norm": 0.08766961097717285,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 42489
+    },
+    {
+      "epoch": 4.359048009848174,
+      "grad_norm": 0.045849695801734924,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 42492
+    },
+    {
+      "epoch": 4.359355765285187,
+      "grad_norm": 0.10003980994224548,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 42495
+    },
+    {
+      "epoch": 4.359663520722199,
+      "grad_norm": 0.05734704062342644,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 42498
+    },
+    {
+      "epoch": 4.359971276159212,
+      "grad_norm": 0.05776417255401611,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 42501
+    },
+    {
+      "epoch": 4.3602790315962245,
+      "grad_norm": 0.03467059135437012,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 42504
+    },
+    {
+      "epoch": 4.360586787033237,
+      "grad_norm": 0.03380267322063446,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 42507
+    },
+    {
+      "epoch": 4.36089454247025,
+      "grad_norm": 0.06621702015399933,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 42510
+    },
+    {
+      "epoch": 4.361202297907263,
+      "grad_norm": 0.05189868062734604,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 42513
+    },
+    {
+      "epoch": 4.361510053344276,
+      "grad_norm": 0.05479085072875023,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 42516
+    },
+    {
+      "epoch": 4.361817808781288,
+      "grad_norm": 0.07013875991106033,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 42519
+    },
+    {
+      "epoch": 4.362125564218301,
+      "grad_norm": 0.07816500216722488,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 42522
+    },
+    {
+      "epoch": 4.362433319655314,
+      "grad_norm": 0.05446416139602661,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 42525
+    },
+    {
+      "epoch": 4.362741075092327,
+      "grad_norm": 0.05260547250509262,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 42528
+    },
+    {
+      "epoch": 4.36304883052934,
+      "grad_norm": 0.042641572654247284,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 42531
+    },
+    {
+      "epoch": 4.363356585966352,
+      "grad_norm": 0.05067918077111244,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 42534
+    },
+    {
+      "epoch": 4.363664341403365,
+      "grad_norm": 0.05734257400035858,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 42537
+    },
+    {
+      "epoch": 4.363972096840378,
+      "grad_norm": 0.12982626259326935,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 42540
+    },
+    {
+      "epoch": 4.36427985227739,
+      "grad_norm": 0.06049492582678795,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 42543
+    },
+    {
+      "epoch": 4.364587607714403,
+      "grad_norm": 0.03683345392346382,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 42546
+    },
+    {
+      "epoch": 4.364895363151415,
+      "grad_norm": 0.055864546447992325,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 42549
+    },
+    {
+      "epoch": 4.365203118588428,
+      "grad_norm": 0.046534840017557144,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 42552
+    },
+    {
+      "epoch": 4.365510874025441,
+      "grad_norm": 0.07829025387763977,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 42555
+    },
+    {
+      "epoch": 4.365818629462454,
+      "grad_norm": 0.03704720363020897,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 42558
+    },
+    {
+      "epoch": 4.3661263848994665,
+      "grad_norm": 0.03933021053671837,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 42561
+    },
+    {
+      "epoch": 4.366434140336479,
+      "grad_norm": 0.05160917341709137,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 42564
+    },
+    {
+      "epoch": 4.366741895773492,
+      "grad_norm": 0.16336973011493683,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 42567
+    },
+    {
+      "epoch": 4.367049651210505,
+      "grad_norm": 0.15664827823638916,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 42570
+    },
+    {
+      "epoch": 4.367357406647518,
+      "grad_norm": 0.06535400450229645,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 42573
+    },
+    {
+      "epoch": 4.36766516208453,
+      "grad_norm": 0.10573652386665344,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 42576
+    },
+    {
+      "epoch": 4.367972917521543,
+      "grad_norm": 0.06016778200864792,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 42579
+    },
+    {
+      "epoch": 4.368280672958556,
+      "grad_norm": 0.07716617733240128,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 42582
+    },
+    {
+      "epoch": 4.368588428395569,
+      "grad_norm": 0.0795513242483139,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 42585
+    },
+    {
+      "epoch": 4.368896183832581,
+      "grad_norm": 0.042184434831142426,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 42588
+    },
+    {
+      "epoch": 4.369203939269593,
+      "grad_norm": 0.10784424096345901,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 42591
+    },
+    {
+      "epoch": 4.369511694706606,
+      "grad_norm": 0.0906398668885231,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 42594
+    },
+    {
+      "epoch": 4.369819450143619,
+      "grad_norm": 0.07714588940143585,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 42597
+    },
+    {
+      "epoch": 4.370127205580632,
+      "grad_norm": 0.07262132316827774,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 42600
+    },
+    {
+      "epoch": 4.3704349610176445,
+      "grad_norm": 0.04687703400850296,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 42603
+    },
+    {
+      "epoch": 4.370742716454657,
+      "grad_norm": 0.03540663421154022,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 42606
+    },
+    {
+      "epoch": 4.37105047189167,
+      "grad_norm": 0.0655878484249115,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 42609
+    },
+    {
+      "epoch": 4.371358227328683,
+      "grad_norm": 0.14327290654182434,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 42612
+    },
+    {
+      "epoch": 4.371665982765696,
+      "grad_norm": 0.051352307200431824,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 42615
+    },
+    {
+      "epoch": 4.371973738202708,
+      "grad_norm": 0.03427935391664505,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 42618
+    },
+    {
+      "epoch": 4.372281493639721,
+      "grad_norm": 0.06677401810884476,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 42621
+    },
+    {
+      "epoch": 4.372589249076734,
+      "grad_norm": 0.0740409791469574,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 42624
+    },
+    {
+      "epoch": 4.372897004513747,
+      "grad_norm": 0.07356753200292587,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 42627
+    },
+    {
+      "epoch": 4.3732047599507595,
+      "grad_norm": 0.0342116579413414,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 42630
+    },
+    {
+      "epoch": 4.373512515387771,
+      "grad_norm": 0.04127725213766098,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 42633
+    },
+    {
+      "epoch": 4.373820270824784,
+      "grad_norm": 0.05259817838668823,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 42636
+    },
+    {
+      "epoch": 4.374128026261797,
+      "grad_norm": 0.041953880339860916,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 42639
+    },
+    {
+      "epoch": 4.37443578169881,
+      "grad_norm": 0.042254798114299774,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 42642
+    },
+    {
+      "epoch": 4.3747435371358225,
+      "grad_norm": 0.0722968801856041,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 42645
+    },
+    {
+      "epoch": 4.375051292572835,
+      "grad_norm": 0.12318527698516846,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 42648
+    },
+    {
+      "epoch": 4.375359048009848,
+      "grad_norm": 0.048684168606996536,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 42651
+    },
+    {
+      "epoch": 4.375666803446861,
+      "grad_norm": 0.09488745033740997,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 42654
+    },
+    {
+      "epoch": 4.375974558883874,
+      "grad_norm": 0.06261547654867172,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 42657
+    },
+    {
+      "epoch": 4.376282314320886,
+      "grad_norm": 0.03739078342914581,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 42660
+    },
+    {
+      "epoch": 4.376590069757899,
+      "grad_norm": 0.03686247393488884,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 42663
+    },
+    {
+      "epoch": 4.376897825194912,
+      "grad_norm": 0.038870710879564285,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 42666
+    },
+    {
+      "epoch": 4.377205580631925,
+      "grad_norm": 0.04153933376073837,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 42669
+    },
+    {
+      "epoch": 4.3775133360689376,
+      "grad_norm": 0.07403891533613205,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 42672
+    },
+    {
+      "epoch": 4.37782109150595,
+      "grad_norm": 0.12554992735385895,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 42675
+    },
+    {
+      "epoch": 4.378128846942962,
+      "grad_norm": 0.07833071053028107,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 42678
+    },
+    {
+      "epoch": 4.378436602379975,
+      "grad_norm": 0.05616781488060951,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 42681
+    },
+    {
+      "epoch": 4.378744357816988,
+      "grad_norm": 0.034405291080474854,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 42684
+    },
+    {
+      "epoch": 4.379052113254001,
+      "grad_norm": 0.03216303512454033,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 42687
+    },
+    {
+      "epoch": 4.379359868691013,
+      "grad_norm": 0.03294415399432182,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 42690
+    },
+    {
+      "epoch": 4.379667624128026,
+      "grad_norm": 0.07378646731376648,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 42693
+    },
+    {
+      "epoch": 4.379975379565039,
+      "grad_norm": 0.0922391265630722,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 42696
+    },
+    {
+      "epoch": 4.380283135002052,
+      "grad_norm": 0.06231202930212021,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 42699
+    },
+    {
+      "epoch": 4.3805908904390645,
+      "grad_norm": 0.07775771617889404,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 42702
+    },
+    {
+      "epoch": 4.380898645876077,
+      "grad_norm": 0.041460320353507996,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 42705
+    },
+    {
+      "epoch": 4.38120640131309,
+      "grad_norm": 0.10348492860794067,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 42708
+    },
+    {
+      "epoch": 4.381514156750103,
+      "grad_norm": 0.06981760263442993,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 42711
+    },
+    {
+      "epoch": 4.381821912187116,
+      "grad_norm": 0.05024630203843117,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 42714
+    },
+    {
+      "epoch": 4.382129667624128,
+      "grad_norm": 0.09696882963180542,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 42717
+    },
+    {
+      "epoch": 4.382437423061141,
+      "grad_norm": 0.08955802023410797,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 42720
+    },
+    {
+      "epoch": 4.382745178498153,
+      "grad_norm": 0.053882379084825516,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 42723
+    },
+    {
+      "epoch": 4.383052933935166,
+      "grad_norm": 0.09641711413860321,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 42726
+    },
+    {
+      "epoch": 4.383360689372179,
+      "grad_norm": 0.03148825094103813,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 42729
+    },
+    {
+      "epoch": 4.383668444809191,
+      "grad_norm": 0.13528583943843842,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 42732
+    },
+    {
+      "epoch": 4.383976200246204,
+      "grad_norm": 0.08340415358543396,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 42735
+    },
+    {
+      "epoch": 4.384283955683217,
+      "grad_norm": 0.04938989877700806,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 42738
+    },
+    {
+      "epoch": 4.38459171112023,
+      "grad_norm": 0.03983978554606438,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 42741
+    },
+    {
+      "epoch": 4.3848994665572425,
+      "grad_norm": 0.04310479387640953,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 42744
+    },
+    {
+      "epoch": 4.385207221994255,
+      "grad_norm": 0.047569889575242996,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 42747
+    },
+    {
+      "epoch": 4.385514977431268,
+      "grad_norm": 0.04927491769194603,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 42750
+    },
+    {
+      "epoch": 4.385822732868281,
+      "grad_norm": 0.17855384945869446,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 42753
+    },
+    {
+      "epoch": 4.386130488305294,
+      "grad_norm": 0.06133484095335007,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 42756
+    },
+    {
+      "epoch": 4.386438243742306,
+      "grad_norm": 0.0671703889966011,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 42759
+    },
+    {
+      "epoch": 4.386745999179319,
+      "grad_norm": 0.06793912500143051,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 42762
+    },
+    {
+      "epoch": 4.387053754616332,
+      "grad_norm": 0.042816221714019775,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 42765
+    },
+    {
+      "epoch": 4.387361510053344,
+      "grad_norm": 0.035412371158599854,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 42768
+    },
+    {
+      "epoch": 4.387669265490357,
+      "grad_norm": 0.05142438784241676,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 42771
+    },
+    {
+      "epoch": 4.387977020927369,
+      "grad_norm": 0.04674524813890457,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 42774
+    },
+    {
+      "epoch": 4.388284776364382,
+      "grad_norm": 0.2127211093902588,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 42777
+    },
+    {
+      "epoch": 4.388592531801395,
+      "grad_norm": 0.09352391213178635,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 42780
+    },
+    {
+      "epoch": 4.388900287238408,
+      "grad_norm": 0.06070404127240181,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 42783
+    },
+    {
+      "epoch": 4.3892080426754205,
+      "grad_norm": 0.0433930978178978,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 42786
+    },
+    {
+      "epoch": 4.389515798112433,
+      "grad_norm": 0.03960421308875084,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 42789
+    },
+    {
+      "epoch": 4.389823553549446,
+      "grad_norm": 0.043349895626306534,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 42792
+    },
+    {
+      "epoch": 4.390131308986459,
+      "grad_norm": 0.04139290004968643,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 42795
+    },
+    {
+      "epoch": 4.390439064423472,
+      "grad_norm": 0.08798687905073166,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 42798
+    },
+    {
+      "epoch": 4.390746819860484,
+      "grad_norm": 0.061553046107292175,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 42801
+    },
+    {
+      "epoch": 4.391054575297497,
+      "grad_norm": 0.04880871623754501,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 42804
+    },
+    {
+      "epoch": 4.39136233073451,
+      "grad_norm": 0.06029786914587021,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 42807
+    },
+    {
+      "epoch": 4.391670086171523,
+      "grad_norm": 0.047968216240406036,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 42810
+    },
+    {
+      "epoch": 4.391977841608535,
+      "grad_norm": 0.11133129149675369,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 42813
+    },
+    {
+      "epoch": 4.392285597045547,
+      "grad_norm": 0.09831759333610535,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 42816
+    },
+    {
+      "epoch": 4.39259335248256,
+      "grad_norm": 0.0829746425151825,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 42819
+    },
+    {
+      "epoch": 4.392901107919573,
+      "grad_norm": 0.04381052777171135,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 42822
+    },
+    {
+      "epoch": 4.393208863356586,
+      "grad_norm": 0.06539449840784073,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 42825
+    },
+    {
+      "epoch": 4.393516618793599,
+      "grad_norm": 0.03784729540348053,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 42828
+    },
+    {
+      "epoch": 4.393824374230611,
+      "grad_norm": 0.10152629017829895,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 42831
+    },
+    {
+      "epoch": 4.394132129667624,
+      "grad_norm": 0.043561842292547226,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 42834
+    },
+    {
+      "epoch": 4.394439885104637,
+      "grad_norm": 0.06225054711103439,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 42837
+    },
+    {
+      "epoch": 4.39474764054165,
+      "grad_norm": 0.08047157526016235,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 42840
+    },
+    {
+      "epoch": 4.3950553959786625,
+      "grad_norm": 0.09267039597034454,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 42843
+    },
+    {
+      "epoch": 4.395363151415675,
+      "grad_norm": 0.10524188727140427,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 42846
+    },
+    {
+      "epoch": 4.395670906852688,
+      "grad_norm": 0.0726306214928627,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 42849
+    },
+    {
+      "epoch": 4.395978662289701,
+      "grad_norm": 0.036908458918333054,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 42852
+    },
+    {
+      "epoch": 4.396286417726714,
+      "grad_norm": 0.052610307931900024,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 42855
+    },
+    {
+      "epoch": 4.3965941731637255,
+      "grad_norm": 0.04229322075843811,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 42858
+    },
+    {
+      "epoch": 4.396901928600738,
+      "grad_norm": 0.1478923261165619,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 42861
+    },
+    {
+      "epoch": 4.397209684037751,
+      "grad_norm": 0.08942049741744995,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 42864
+    },
+    {
+      "epoch": 4.397517439474764,
+      "grad_norm": 0.051714204251766205,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 42867
+    },
+    {
+      "epoch": 4.397825194911777,
+      "grad_norm": 0.03621753305196762,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 42870
+    },
+    {
+      "epoch": 4.398132950348789,
+      "grad_norm": 0.03929129242897034,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 42873
+    },
+    {
+      "epoch": 4.398440705785802,
+      "grad_norm": 0.04855189844965935,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 42876
+    },
+    {
+      "epoch": 4.398748461222815,
+      "grad_norm": 0.060284826904535294,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 42879
+    },
+    {
+      "epoch": 4.399056216659828,
+      "grad_norm": 0.07179313898086548,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 42882
+    },
+    {
+      "epoch": 4.3993639720968405,
+      "grad_norm": 0.10332684218883514,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 42885
+    },
+    {
+      "epoch": 4.399671727533853,
+      "grad_norm": 0.13461638987064362,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 42888
+    },
+    {
+      "epoch": 4.399979482970866,
+      "grad_norm": 0.08879181742668152,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 42891
+    },
+    {
+      "epoch": 4.400287238407879,
+      "grad_norm": 0.064743272960186,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 42894
+    },
+    {
+      "epoch": 4.400594993844892,
+      "grad_norm": 0.04650283232331276,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 42897
+    },
+    {
+      "epoch": 4.400902749281904,
+      "grad_norm": 0.05072256550192833,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 42900
+    },
+    {
+      "epoch": 4.401210504718916,
+      "grad_norm": 0.05984727293252945,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 42903
+    },
+    {
+      "epoch": 4.401518260155929,
+      "grad_norm": 0.0773845687508583,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 42906
+    },
+    {
+      "epoch": 4.401826015592942,
+      "grad_norm": 0.06231288984417915,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 42909
+    },
+    {
+      "epoch": 4.402133771029955,
+      "grad_norm": 0.07276583462953568,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 42912
+    },
+    {
+      "epoch": 4.402441526466967,
+      "grad_norm": 0.043372802436351776,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 42915
+    },
+    {
+      "epoch": 4.40274928190398,
+      "grad_norm": 0.05217559635639191,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 42918
+    },
+    {
+      "epoch": 4.403057037340993,
+      "grad_norm": 0.058887772262096405,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 42921
+    },
+    {
+      "epoch": 4.403364792778006,
+      "grad_norm": 0.04377232864499092,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 42924
+    },
+    {
+      "epoch": 4.4036725482150185,
+      "grad_norm": 0.032925624400377274,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 42927
+    },
+    {
+      "epoch": 4.403980303652031,
+      "grad_norm": 0.1263015866279602,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 42930
+    },
+    {
+      "epoch": 4.404288059089044,
+      "grad_norm": 0.046885982155799866,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 42933
+    },
+    {
+      "epoch": 4.404595814526057,
+      "grad_norm": 0.10321955382823944,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 42936
+    },
+    {
+      "epoch": 4.40490356996307,
+      "grad_norm": 0.06488531827926636,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 42939
+    },
+    {
+      "epoch": 4.405211325400082,
+      "grad_norm": 0.09423347562551498,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 42942
+    },
+    {
+      "epoch": 4.405519080837095,
+      "grad_norm": 0.06422553211450577,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 42945
+    },
+    {
+      "epoch": 4.405826836274107,
+      "grad_norm": 0.05872100591659546,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 42948
+    },
+    {
+      "epoch": 4.40613459171112,
+      "grad_norm": 0.043210845440626144,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 42951
+    },
+    {
+      "epoch": 4.406442347148133,
+      "grad_norm": 0.041857898235321045,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 42954
+    },
+    {
+      "epoch": 4.406750102585145,
+      "grad_norm": 0.03347545117139816,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 42957
+    },
+    {
+      "epoch": 4.407057858022158,
+      "grad_norm": 0.0377437062561512,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 42960
+    },
+    {
+      "epoch": 4.407365613459171,
+      "grad_norm": 0.07432477176189423,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 42963
+    },
+    {
+      "epoch": 4.407673368896184,
+      "grad_norm": 0.038358092308044434,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 42966
+    },
+    {
+      "epoch": 4.4079811243331966,
+      "grad_norm": 0.09125792980194092,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 42969
+    },
+    {
+      "epoch": 4.408288879770209,
+      "grad_norm": 0.11572171747684479,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 42972
+    },
+    {
+      "epoch": 4.408596635207222,
+      "grad_norm": 0.05868324264883995,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 42975
+    },
+    {
+      "epoch": 4.408904390644235,
+      "grad_norm": 0.0679447203874588,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 42978
+    },
+    {
+      "epoch": 4.409212146081248,
+      "grad_norm": 0.050446171313524246,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 42981
+    },
+    {
+      "epoch": 4.4095199015182605,
+      "grad_norm": 0.05003255605697632,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 42984
+    },
+    {
+      "epoch": 4.409827656955273,
+      "grad_norm": 0.05460989847779274,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 42987
+    },
+    {
+      "epoch": 4.410135412392286,
+      "grad_norm": 0.04039241746068001,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 42990
+    },
+    {
+      "epoch": 4.410443167829298,
+      "grad_norm": 0.04114096984267235,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 42993
+    },
+    {
+      "epoch": 4.410750923266311,
+      "grad_norm": 0.1269209384918213,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 42996
+    },
+    {
+      "epoch": 4.4110586787033235,
+      "grad_norm": 0.059638604521751404,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 42999
+    },
+    {
+      "epoch": 4.411366434140336,
+      "grad_norm": 0.09658347070217133,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 43002
+    },
+    {
+      "epoch": 4.411674189577349,
+      "grad_norm": 0.05749305710196495,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 43005
+    },
+    {
+      "epoch": 4.411981945014362,
+      "grad_norm": 0.04522128403186798,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 43008
+    },
+    {
+      "epoch": 4.412289700451375,
+      "grad_norm": 0.034170012921094894,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 43011
+    },
+    {
+      "epoch": 4.412597455888387,
+      "grad_norm": 0.04310225695371628,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 43014
+    },
+    {
+      "epoch": 4.4129052113254,
+      "grad_norm": 0.15212641656398773,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 43017
+    },
+    {
+      "epoch": 4.413212966762413,
+      "grad_norm": 0.12978756427764893,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 43020
+    },
+    {
+      "epoch": 4.413520722199426,
+      "grad_norm": 0.07762716710567474,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 43023
+    },
+    {
+      "epoch": 4.4138284776364385,
+      "grad_norm": 0.049494557082653046,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 43026
+    },
+    {
+      "epoch": 4.414136233073451,
+      "grad_norm": 0.060781434178352356,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 43029
+    },
+    {
+      "epoch": 4.414443988510464,
+      "grad_norm": 0.03154407814145088,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 43032
+    },
+    {
+      "epoch": 4.414751743947477,
+      "grad_norm": 0.048733990639448166,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 43035
+    },
+    {
+      "epoch": 4.415059499384489,
+      "grad_norm": 0.04283260181546211,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 43038
+    },
+    {
+      "epoch": 4.4153672548215015,
+      "grad_norm": 0.06798703223466873,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 43041
+    },
+    {
+      "epoch": 4.415675010258514,
+      "grad_norm": 0.10504204034805298,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 43044
+    },
+    {
+      "epoch": 4.415982765695527,
+      "grad_norm": 0.1270654797554016,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 43047
+    },
+    {
+      "epoch": 4.41629052113254,
+      "grad_norm": 0.06907479465007782,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 43050
+    },
+    {
+      "epoch": 4.416598276569553,
+      "grad_norm": 0.034706976264715195,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 43053
+    },
+    {
+      "epoch": 4.416906032006565,
+      "grad_norm": 0.033550575375556946,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 43056
+    },
+    {
+      "epoch": 4.417213787443578,
+      "grad_norm": 0.03790535405278206,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 43059
+    },
+    {
+      "epoch": 4.417521542880591,
+      "grad_norm": 0.04516111686825752,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 43062
+    },
+    {
+      "epoch": 4.417829298317604,
+      "grad_norm": 0.07906030863523483,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 43065
+    },
+    {
+      "epoch": 4.4181370537546165,
+      "grad_norm": 0.11393613368272781,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 43068
+    },
+    {
+      "epoch": 4.418444809191629,
+      "grad_norm": 0.04494154825806618,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 43071
+    },
+    {
+      "epoch": 4.418752564628642,
+      "grad_norm": 0.04806877672672272,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 43074
+    },
+    {
+      "epoch": 4.419060320065655,
+      "grad_norm": 0.09511521458625793,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 43077
+    },
+    {
+      "epoch": 4.419368075502668,
+      "grad_norm": 0.036587249487638474,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 43080
+    },
+    {
+      "epoch": 4.4196758309396795,
+      "grad_norm": 0.03894515708088875,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 43083
+    },
+    {
+      "epoch": 4.419983586376692,
+      "grad_norm": 0.06960659474134445,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 43086
+    },
+    {
+      "epoch": 4.420291341813705,
+      "grad_norm": 0.04635101556777954,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 43089
+    },
+    {
+      "epoch": 4.420599097250718,
+      "grad_norm": 0.09964662045240402,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 43092
+    },
+    {
+      "epoch": 4.420906852687731,
+      "grad_norm": 0.10087699443101883,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 43095
+    },
+    {
+      "epoch": 4.421214608124743,
+      "grad_norm": 0.09748293459415436,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 43098
+    },
+    {
+      "epoch": 4.421522363561756,
+      "grad_norm": 0.04654933884739876,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 43101
+    },
+    {
+      "epoch": 4.421830118998769,
+      "grad_norm": 0.05607482045888901,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 43104
+    },
+    {
+      "epoch": 4.422137874435782,
+      "grad_norm": 0.0513363815844059,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 43107
+    },
+    {
+      "epoch": 4.4224456298727945,
+      "grad_norm": 0.048932064324617386,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 43110
+    },
+    {
+      "epoch": 4.422753385309807,
+      "grad_norm": 0.05465288087725639,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 43113
+    },
+    {
+      "epoch": 4.42306114074682,
+      "grad_norm": 0.04400217905640602,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 43116
+    },
+    {
+      "epoch": 4.423368896183833,
+      "grad_norm": 0.03483066335320473,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 43119
+    },
+    {
+      "epoch": 4.423676651620846,
+      "grad_norm": 0.04909312352538109,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 43122
+    },
+    {
+      "epoch": 4.4239844070578584,
+      "grad_norm": 0.09361930191516876,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 43125
+    },
+    {
+      "epoch": 4.42429216249487,
+      "grad_norm": 0.06980263441801071,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 43128
+    },
+    {
+      "epoch": 4.424599917931883,
+      "grad_norm": 0.06628268212080002,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 43131
+    },
+    {
+      "epoch": 4.424907673368896,
+      "grad_norm": 0.10180728137493134,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 43134
+    },
+    {
+      "epoch": 4.425215428805909,
+      "grad_norm": 0.10624035447835922,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 43137
+    },
+    {
+      "epoch": 4.4255231842429215,
+      "grad_norm": 0.08775760233402252,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 43140
+    },
+    {
+      "epoch": 4.425830939679934,
+      "grad_norm": 0.06270244717597961,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 43143
+    },
+    {
+      "epoch": 4.426138695116947,
+      "grad_norm": 0.06912422925233841,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 43146
+    },
+    {
+      "epoch": 4.42644645055396,
+      "grad_norm": 0.05379674583673477,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 43149
+    },
+    {
+      "epoch": 4.426754205990973,
+      "grad_norm": 0.041072819381952286,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 43152
+    },
+    {
+      "epoch": 4.427061961427985,
+      "grad_norm": 0.05191728100180626,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 43155
+    },
+    {
+      "epoch": 4.427369716864998,
+      "grad_norm": 0.03773737698793411,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 43158
+    },
+    {
+      "epoch": 4.427677472302011,
+      "grad_norm": 0.10148506611585617,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 43161
+    },
+    {
+      "epoch": 4.427985227739024,
+      "grad_norm": 0.071526899933815,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 43164
+    },
+    {
+      "epoch": 4.4282929831760365,
+      "grad_norm": 0.07906235009431839,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 43167
+    },
+    {
+      "epoch": 4.428600738613049,
+      "grad_norm": 0.18193936347961426,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 43170
+    },
+    {
+      "epoch": 4.428908494050061,
+      "grad_norm": 0.1734820157289505,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 43173
+    },
+    {
+      "epoch": 4.429216249487074,
+      "grad_norm": 0.09781987965106964,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 43176
+    },
+    {
+      "epoch": 4.429524004924087,
+      "grad_norm": 0.06472992151975632,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 43179
+    },
+    {
+      "epoch": 4.4298317603610995,
+      "grad_norm": 0.04425714537501335,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 43182
+    },
+    {
+      "epoch": 4.430139515798112,
+      "grad_norm": 0.04678405821323395,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 43185
+    },
+    {
+      "epoch": 4.430447271235125,
+      "grad_norm": 0.04311143234372139,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 43188
+    },
+    {
+      "epoch": 4.430755026672138,
+      "grad_norm": 0.060729045420885086,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 43191
+    },
+    {
+      "epoch": 4.431062782109151,
+      "grad_norm": 0.07931876182556152,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 43194
+    },
+    {
+      "epoch": 4.431370537546163,
+      "grad_norm": 0.048263415694236755,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 43197
+    },
+    {
+      "epoch": 4.431678292983176,
+      "grad_norm": 0.05372344329953194,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 43200
+    },
+    {
+      "epoch": 4.431986048420189,
+      "grad_norm": 0.06506495922803879,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 43203
+    },
+    {
+      "epoch": 4.432293803857202,
+      "grad_norm": 0.03484785556793213,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 43206
+    },
+    {
+      "epoch": 4.4326015592942145,
+      "grad_norm": 0.18487481772899628,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 43209
+    },
+    {
+      "epoch": 4.432909314731227,
+      "grad_norm": 0.11683176457881927,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 43212
+    },
+    {
+      "epoch": 4.43321707016824,
+      "grad_norm": 0.08598039299249649,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 43215
+    },
+    {
+      "epoch": 4.433524825605252,
+      "grad_norm": 0.0889374166727066,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 43218
+    },
+    {
+      "epoch": 4.433832581042265,
+      "grad_norm": 0.061477046459913254,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 43221
+    },
+    {
+      "epoch": 4.4341403364792775,
+      "grad_norm": 0.046285588294267654,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 43224
+    },
+    {
+      "epoch": 4.43444809191629,
+      "grad_norm": 0.06310747563838959,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 43227
+    },
+    {
+      "epoch": 4.434755847353303,
+      "grad_norm": 0.04792320355772972,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 43230
+    },
+    {
+      "epoch": 4.435063602790316,
+      "grad_norm": 0.03359830752015114,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 43233
+    },
+    {
+      "epoch": 4.435371358227329,
+      "grad_norm": 0.04137792810797691,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 43236
+    },
+    {
+      "epoch": 4.435679113664341,
+      "grad_norm": 0.04756288602948189,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 43239
+    },
+    {
+      "epoch": 4.435986869101354,
+      "grad_norm": 0.054949600249528885,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 43242
+    },
+    {
+      "epoch": 4.436294624538367,
+      "grad_norm": 0.14858773350715637,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 43245
+    },
+    {
+      "epoch": 4.43660237997538,
+      "grad_norm": 0.08089875429868698,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 43248
+    },
+    {
+      "epoch": 4.4369101354123925,
+      "grad_norm": 0.0340421199798584,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 43251
+    },
+    {
+      "epoch": 4.437217890849405,
+      "grad_norm": 0.04666365310549736,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 43254
+    },
+    {
+      "epoch": 4.437525646286418,
+      "grad_norm": 0.04860776662826538,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 43257
+    },
+    {
+      "epoch": 4.437833401723431,
+      "grad_norm": 0.04814871773123741,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 43260
+    },
+    {
+      "epoch": 4.438141157160443,
+      "grad_norm": 0.07159804552793503,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 43263
+    },
+    {
+      "epoch": 4.4384489125974556,
+      "grad_norm": 0.08109670132398605,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 43266
+    },
+    {
+      "epoch": 4.438756668034468,
+      "grad_norm": 0.10254354774951935,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 43269
+    },
+    {
+      "epoch": 4.439064423471481,
+      "grad_norm": 0.10418622195720673,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 43272
+    },
+    {
+      "epoch": 4.439372178908494,
+      "grad_norm": 0.04432861879467964,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 43275
+    },
+    {
+      "epoch": 4.439679934345507,
+      "grad_norm": 0.05000505968928337,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 43278
+    },
+    {
+      "epoch": 4.4399876897825195,
+      "grad_norm": 0.0718022808432579,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 43281
+    },
+    {
+      "epoch": 4.440295445219532,
+      "grad_norm": 0.038088515400886536,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 43284
+    },
+    {
+      "epoch": 4.440603200656545,
+      "grad_norm": 0.04395698755979538,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 43287
+    },
+    {
+      "epoch": 4.440910956093558,
+      "grad_norm": 0.03977638855576515,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 43290
+    },
+    {
+      "epoch": 4.441218711530571,
+      "grad_norm": 0.11403058469295502,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 43293
+    },
+    {
+      "epoch": 4.441526466967583,
+      "grad_norm": 0.03809293359518051,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 43296
+    },
+    {
+      "epoch": 4.441834222404596,
+      "grad_norm": 0.04494985193014145,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 43299
+    },
+    {
+      "epoch": 4.442141977841609,
+      "grad_norm": 0.11755307763814926,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 43302
+    },
+    {
+      "epoch": 4.442449733278622,
+      "grad_norm": 0.046333327889442444,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 43305
+    },
+    {
+      "epoch": 4.442757488715634,
+      "grad_norm": 0.05719512701034546,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 43308
+    },
+    {
+      "epoch": 4.443065244152646,
+      "grad_norm": 0.0707264393568039,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 43311
+    },
+    {
+      "epoch": 4.443372999589659,
+      "grad_norm": 0.043551988899707794,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 43314
+    },
+    {
+      "epoch": 4.443680755026672,
+      "grad_norm": 0.11364515870809555,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 43317
+    },
+    {
+      "epoch": 4.443988510463685,
+      "grad_norm": 0.15514688193798065,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 43320
+    },
+    {
+      "epoch": 4.4442962659006975,
+      "grad_norm": 0.10261968523263931,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 43323
+    },
+    {
+      "epoch": 4.44460402133771,
+      "grad_norm": 0.09683386236429214,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 43326
+    },
+    {
+      "epoch": 4.444911776774723,
+      "grad_norm": 0.055726949125528336,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 43329
+    },
+    {
+      "epoch": 4.445219532211736,
+      "grad_norm": 0.053106311708688736,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 43332
+    },
+    {
+      "epoch": 4.445527287648749,
+      "grad_norm": 0.043625880032777786,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 43335
+    },
+    {
+      "epoch": 4.445835043085761,
+      "grad_norm": 0.05958040803670883,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 43338
+    },
+    {
+      "epoch": 4.446142798522774,
+      "grad_norm": 0.043522778898477554,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 43341
+    },
+    {
+      "epoch": 4.446450553959787,
+      "grad_norm": 0.038740675896406174,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 43344
+    },
+    {
+      "epoch": 4.4467583093968,
+      "grad_norm": 0.043132517486810684,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 43347
+    },
+    {
+      "epoch": 4.4470660648338125,
+      "grad_norm": 0.07173844426870346,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 43350
+    },
+    {
+      "epoch": 4.447373820270824,
+      "grad_norm": 0.06787187606096268,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 43353
+    },
+    {
+      "epoch": 4.447681575707837,
+      "grad_norm": 0.1125207170844078,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 43356
+    },
+    {
+      "epoch": 4.44798933114485,
+      "grad_norm": 0.037585411220788956,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 43359
+    },
+    {
+      "epoch": 4.448297086581863,
+      "grad_norm": 0.09652412682771683,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 43362
+    },
+    {
+      "epoch": 4.4486048420188755,
+      "grad_norm": 0.0548916831612587,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 43365
+    },
+    {
+      "epoch": 4.448912597455888,
+      "grad_norm": 0.07648541778326035,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 43368
+    },
+    {
+      "epoch": 4.449220352892901,
+      "grad_norm": 0.07968877255916595,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 43371
+    },
+    {
+      "epoch": 4.449528108329914,
+      "grad_norm": 0.046214036643505096,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 43374
+    },
+    {
+      "epoch": 4.449835863766927,
+      "grad_norm": 0.03589135408401489,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 43377
+    },
+    {
+      "epoch": 4.450143619203939,
+      "grad_norm": 0.1217019185423851,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 43380
+    },
+    {
+      "epoch": 4.450451374640952,
+      "grad_norm": 0.14711309969425201,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 43383
+    },
+    {
+      "epoch": 4.450759130077965,
+      "grad_norm": 0.15004196763038635,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 43386
+    },
+    {
+      "epoch": 4.451066885514978,
+      "grad_norm": 0.143090158700943,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 43389
+    },
+    {
+      "epoch": 4.4513746409519905,
+      "grad_norm": 0.06957674771547318,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 43392
+    },
+    {
+      "epoch": 4.451682396389003,
+      "grad_norm": 0.10823699831962585,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 43395
+    },
+    {
+      "epoch": 4.451990151826015,
+      "grad_norm": 0.0608796700835228,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 43398
+    },
+    {
+      "epoch": 4.452297907263028,
+      "grad_norm": 0.038806330412626266,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 43401
+    },
+    {
+      "epoch": 4.452605662700041,
+      "grad_norm": 0.053961463272571564,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 43404
+    },
+    {
+      "epoch": 4.4529134181370535,
+      "grad_norm": 0.03717347979545593,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 43407
+    },
+    {
+      "epoch": 4.453221173574066,
+      "grad_norm": 0.04289384186267853,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 43410
+    },
+    {
+      "epoch": 4.453528929011079,
+      "grad_norm": 0.08649279922246933,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 43413
+    },
+    {
+      "epoch": 4.453836684448092,
+      "grad_norm": 0.1384621560573578,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 43416
+    },
+    {
+      "epoch": 4.454144439885105,
+      "grad_norm": 0.09169614315032959,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 43419
+    },
+    {
+      "epoch": 4.4544521953221174,
+      "grad_norm": 0.07004016637802124,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 43422
+    },
+    {
+      "epoch": 4.45475995075913,
+      "grad_norm": 0.0574488528072834,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 43425
+    },
+    {
+      "epoch": 4.455067706196143,
+      "grad_norm": 0.048521172255277634,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 43428
+    },
+    {
+      "epoch": 4.455375461633156,
+      "grad_norm": 0.11481918394565582,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 43431
+    },
+    {
+      "epoch": 4.455683217070169,
+      "grad_norm": 0.09158429503440857,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 43434
+    },
+    {
+      "epoch": 4.455990972507181,
+      "grad_norm": 0.039962366223335266,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 43437
+    },
+    {
+      "epoch": 4.456298727944194,
+      "grad_norm": 0.04112045839428902,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 43440
+    },
+    {
+      "epoch": 4.456606483381206,
+      "grad_norm": 0.057623397558927536,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 43443
+    },
+    {
+      "epoch": 4.456914238818219,
+      "grad_norm": 0.0683390349149704,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 43446
+    },
+    {
+      "epoch": 4.457221994255232,
+      "grad_norm": 0.09766959398984909,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 43449
+    },
+    {
+      "epoch": 4.457529749692244,
+      "grad_norm": 0.041423216462135315,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 43452
+    },
+    {
+      "epoch": 4.457837505129257,
+      "grad_norm": 0.08813819289207458,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 43455
+    },
+    {
+      "epoch": 4.45814526056627,
+      "grad_norm": 0.06834255903959274,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 43458
+    },
+    {
+      "epoch": 4.458453016003283,
+      "grad_norm": 0.08782140165567398,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 43461
+    },
+    {
+      "epoch": 4.4587607714402955,
+      "grad_norm": 0.08695351332426071,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 43464
+    },
+    {
+      "epoch": 4.459068526877308,
+      "grad_norm": 0.09302899986505508,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 43467
+    },
+    {
+      "epoch": 4.459376282314321,
+      "grad_norm": 0.06920462846755981,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 43470
+    },
+    {
+      "epoch": 4.459684037751334,
+      "grad_norm": 0.04809924215078354,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 43473
+    },
+    {
+      "epoch": 4.459991793188347,
+      "grad_norm": 0.06642390787601471,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 43476
+    },
+    {
+      "epoch": 4.460299548625359,
+      "grad_norm": 0.04986566677689552,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 43479
+    },
+    {
+      "epoch": 4.460607304062372,
+      "grad_norm": 0.061649248003959656,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 43482
+    },
+    {
+      "epoch": 4.460915059499385,
+      "grad_norm": 0.049117859452962875,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 43485
+    },
+    {
+      "epoch": 4.461222814936397,
+      "grad_norm": 0.049163367599248886,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 43488
+    },
+    {
+      "epoch": 4.46153057037341,
+      "grad_norm": 0.036282382905483246,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 43491
+    },
+    {
+      "epoch": 4.461838325810422,
+      "grad_norm": 0.10942814499139786,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 43494
+    },
+    {
+      "epoch": 4.462146081247435,
+      "grad_norm": 0.06589425355195999,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 43497
+    },
+    {
+      "epoch": 4.462453836684448,
+      "grad_norm": 0.0967121571302414,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 43500
+    },
+    {
+      "epoch": 4.462761592121461,
+      "grad_norm": 0.0673070177435875,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 43503
+    },
+    {
+      "epoch": 4.4630693475584735,
+      "grad_norm": 0.09439177066087723,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 43506
+    },
+    {
+      "epoch": 4.463377102995486,
+      "grad_norm": 0.06875397264957428,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 43509
+    },
+    {
+      "epoch": 4.463684858432499,
+      "grad_norm": 0.0774875059723854,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 43512
+    },
+    {
+      "epoch": 4.463992613869512,
+      "grad_norm": 0.07138268649578094,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 43515
+    },
+    {
+      "epoch": 4.464300369306525,
+      "grad_norm": 0.03174857050180435,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 43518
+    },
+    {
+      "epoch": 4.464608124743537,
+      "grad_norm": 0.11445671319961548,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 43521
+    },
+    {
+      "epoch": 4.46491588018055,
+      "grad_norm": 0.041139326989650726,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 43524
+    },
+    {
+      "epoch": 4.465223635617563,
+      "grad_norm": 0.043187763541936874,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 43527
+    },
+    {
+      "epoch": 4.465531391054576,
+      "grad_norm": 0.09367438405752182,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 43530
+    },
+    {
+      "epoch": 4.465839146491588,
+      "grad_norm": 0.07866624742746353,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 43533
+    },
+    {
+      "epoch": 4.4661469019286,
+      "grad_norm": 0.08221796900033951,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 43536
+    },
+    {
+      "epoch": 4.466454657365613,
+      "grad_norm": 0.07687985897064209,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 43539
+    },
+    {
+      "epoch": 4.466762412802626,
+      "grad_norm": 0.042440030723810196,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 43542
+    },
+    {
+      "epoch": 4.467070168239639,
+      "grad_norm": 0.03422633558511734,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 43545
+    },
+    {
+      "epoch": 4.4673779236766515,
+      "grad_norm": 0.05047105997800827,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 43548
+    },
+    {
+      "epoch": 4.467685679113664,
+      "grad_norm": 0.0817263275384903,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 43551
+    },
+    {
+      "epoch": 4.467993434550677,
+      "grad_norm": 0.07055595517158508,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 43554
+    },
+    {
+      "epoch": 4.46830118998769,
+      "grad_norm": 0.12304790318012238,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 43557
+    },
+    {
+      "epoch": 4.468608945424703,
+      "grad_norm": 0.057413533329963684,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 43560
+    },
+    {
+      "epoch": 4.468916700861715,
+      "grad_norm": 0.1333860456943512,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 43563
+    },
+    {
+      "epoch": 4.469224456298728,
+      "grad_norm": 0.04758629575371742,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 43566
+    },
+    {
+      "epoch": 4.469532211735741,
+      "grad_norm": 0.0403926707804203,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 43569
+    },
+    {
+      "epoch": 4.469839967172754,
+      "grad_norm": 0.05051876977086067,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 43572
+    },
+    {
+      "epoch": 4.470147722609767,
+      "grad_norm": 0.12117879837751389,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 43575
+    },
+    {
+      "epoch": 4.4704554780467785,
+      "grad_norm": 0.0969235822558403,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 43578
+    },
+    {
+      "epoch": 4.470763233483791,
+      "grad_norm": 0.10259240120649338,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 43581
+    },
+    {
+      "epoch": 4.471070988920804,
+      "grad_norm": 0.056327935308218,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 43584
+    },
+    {
+      "epoch": 4.471378744357817,
+      "grad_norm": 0.09567862004041672,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 43587
+    },
+    {
+      "epoch": 4.47168649979483,
+      "grad_norm": 0.0687853991985321,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 43590
+    },
+    {
+      "epoch": 4.471994255231842,
+      "grad_norm": 0.03400299698114395,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 43593
+    },
+    {
+      "epoch": 4.472302010668855,
+      "grad_norm": 0.05291087552905083,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 43596
+    },
+    {
+      "epoch": 4.472609766105868,
+      "grad_norm": 0.09208089858293533,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 43599
+    },
+    {
+      "epoch": 4.472917521542881,
+      "grad_norm": 0.08073403686285019,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 43602
+    },
+    {
+      "epoch": 4.4732252769798935,
+      "grad_norm": 0.04130372405052185,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 43605
+    },
+    {
+      "epoch": 4.473533032416906,
+      "grad_norm": 0.06413860619068146,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 43608
+    },
+    {
+      "epoch": 4.473840787853919,
+      "grad_norm": 0.10064514726400375,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 43611
+    },
+    {
+      "epoch": 4.474148543290932,
+      "grad_norm": 0.04522353783249855,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 43614
+    },
+    {
+      "epoch": 4.474456298727945,
+      "grad_norm": 0.03851405903697014,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 43617
+    },
+    {
+      "epoch": 4.474764054164957,
+      "grad_norm": 0.03857644274830818,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 43620
+    },
+    {
+      "epoch": 4.475071809601969,
+      "grad_norm": 0.07934223115444183,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 43623
+    },
+    {
+      "epoch": 4.475379565038982,
+      "grad_norm": 0.034822918474674225,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 43626
+    },
+    {
+      "epoch": 4.475687320475995,
+      "grad_norm": 0.07185374945402145,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 43629
+    },
+    {
+      "epoch": 4.475995075913008,
+      "grad_norm": 0.03542499616742134,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 43632
+    },
+    {
+      "epoch": 4.47630283135002,
+      "grad_norm": 0.034001853317022324,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 43635
+    },
+    {
+      "epoch": 4.476610586787033,
+      "grad_norm": 0.04940172657370567,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 43638
+    },
+    {
+      "epoch": 4.476918342224046,
+      "grad_norm": 0.04434162750840187,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 43641
+    },
+    {
+      "epoch": 4.477226097661059,
+      "grad_norm": 0.08915964514017105,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 43644
+    },
+    {
+      "epoch": 4.4775338530980715,
+      "grad_norm": 0.05530424416065216,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 43647
+    },
+    {
+      "epoch": 4.477841608535084,
+      "grad_norm": 0.06746593117713928,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 43650
+    },
+    {
+      "epoch": 4.478149363972097,
+      "grad_norm": 0.05016857758164406,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 43653
+    },
+    {
+      "epoch": 4.47845711940911,
+      "grad_norm": 0.04478998854756355,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 43656
+    },
+    {
+      "epoch": 4.478764874846123,
+      "grad_norm": 0.04121081531047821,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 43659
+    },
+    {
+      "epoch": 4.479072630283135,
+      "grad_norm": 0.0476783849298954,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 43662
+    },
+    {
+      "epoch": 4.479380385720148,
+      "grad_norm": 0.15095461905002594,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 43665
+    },
+    {
+      "epoch": 4.47968814115716,
+      "grad_norm": 0.06961385160684586,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 43668
+    },
+    {
+      "epoch": 4.479995896594173,
+      "grad_norm": 0.040564242750406265,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 43671
+    },
+    {
+      "epoch": 4.480303652031186,
+      "grad_norm": 0.07302725315093994,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 43674
+    },
+    {
+      "epoch": 4.480611407468198,
+      "grad_norm": 0.12494261562824249,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 43677
+    },
+    {
+      "epoch": 4.480919162905211,
+      "grad_norm": 0.06394487619400024,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 43680
+    },
+    {
+      "epoch": 4.481226918342224,
+      "grad_norm": 0.0740828663110733,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 43683
+    },
+    {
+      "epoch": 4.481534673779237,
+      "grad_norm": 0.06996148079633713,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 43686
+    },
+    {
+      "epoch": 4.4818424292162495,
+      "grad_norm": 0.041247058659791946,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 43689
+    },
+    {
+      "epoch": 4.482150184653262,
+      "grad_norm": 0.04092543199658394,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 43692
+    },
+    {
+      "epoch": 4.482457940090275,
+      "grad_norm": 0.05542224645614624,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 43695
+    },
+    {
+      "epoch": 4.482765695527288,
+      "grad_norm": 0.0780789703130722,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 43698
+    },
+    {
+      "epoch": 4.483073450964301,
+      "grad_norm": 0.06363578885793686,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 43701
+    },
+    {
+      "epoch": 4.483381206401313,
+      "grad_norm": 0.12707744538784027,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 43704
+    },
+    {
+      "epoch": 4.483688961838326,
+      "grad_norm": 0.14165590703487396,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 43707
+    },
+    {
+      "epoch": 4.483996717275339,
+      "grad_norm": 0.08245092630386353,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 43710
+    },
+    {
+      "epoch": 4.484304472712351,
+      "grad_norm": 0.048774946480989456,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 43713
+    },
+    {
+      "epoch": 4.484612228149364,
+      "grad_norm": 0.05654727295041084,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 43716
+    },
+    {
+      "epoch": 4.4849199835863764,
+      "grad_norm": 0.046862781047821045,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 43719
+    },
+    {
+      "epoch": 4.485227739023389,
+      "grad_norm": 0.06660104542970657,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 43722
+    },
+    {
+      "epoch": 4.485535494460402,
+      "grad_norm": 0.05910905450582504,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 43725
+    },
+    {
+      "epoch": 4.485843249897415,
+      "grad_norm": 0.05047675967216492,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 43728
+    },
+    {
+      "epoch": 4.486151005334428,
+      "grad_norm": 0.03190279006958008,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 43731
+    },
+    {
+      "epoch": 4.48645876077144,
+      "grad_norm": 0.04960602894425392,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 43734
+    },
+    {
+      "epoch": 4.486766516208453,
+      "grad_norm": 0.10073108971118927,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 43737
+    },
+    {
+      "epoch": 4.487074271645466,
+      "grad_norm": 0.051219914108514786,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 43740
+    },
+    {
+      "epoch": 4.487382027082479,
+      "grad_norm": 0.051645878702402115,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 43743
+    },
+    {
+      "epoch": 4.4876897825194915,
+      "grad_norm": 0.08252865076065063,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 43746
+    },
+    {
+      "epoch": 4.487997537956504,
+      "grad_norm": 0.06515108048915863,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 43749
+    },
+    {
+      "epoch": 4.488305293393517,
+      "grad_norm": 0.038027096539735794,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 43752
+    },
+    {
+      "epoch": 4.48861304883053,
+      "grad_norm": 0.06444455683231354,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 43755
+    },
+    {
+      "epoch": 4.488920804267542,
+      "grad_norm": 0.03724157065153122,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 43758
+    },
+    {
+      "epoch": 4.4892285597045545,
+      "grad_norm": 0.04831653833389282,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 43761
+    },
+    {
+      "epoch": 4.489536315141567,
+      "grad_norm": 0.062199629843235016,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 43764
+    },
+    {
+      "epoch": 4.48984407057858,
+      "grad_norm": 0.047005314379930496,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 43767
+    },
+    {
+      "epoch": 4.490151826015593,
+      "grad_norm": 0.1316227912902832,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 43770
+    },
+    {
+      "epoch": 4.490459581452606,
+      "grad_norm": 0.05741405114531517,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 43773
+    },
+    {
+      "epoch": 4.490767336889618,
+      "grad_norm": 0.10407622158527374,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 43776
+    },
+    {
+      "epoch": 4.491075092326631,
+      "grad_norm": 0.05979277566075325,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 43779
+    },
+    {
+      "epoch": 4.491382847763644,
+      "grad_norm": 0.033753860741853714,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 43782
+    },
+    {
+      "epoch": 4.491690603200657,
+      "grad_norm": 0.03481614217162132,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 43785
+    },
+    {
+      "epoch": 4.4919983586376695,
+      "grad_norm": 0.05884877219796181,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 43788
+    },
+    {
+      "epoch": 4.492306114074682,
+      "grad_norm": 0.10501875728368759,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 43791
+    },
+    {
+      "epoch": 4.492613869511695,
+      "grad_norm": 0.07319078594446182,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 43794
+    },
+    {
+      "epoch": 4.492921624948708,
+      "grad_norm": 0.04055263102054596,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 43797
+    },
+    {
+      "epoch": 4.493229380385721,
+      "grad_norm": 0.1268715113401413,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 43800
+    },
+    {
+      "epoch": 4.4935371358227325,
+      "grad_norm": 0.0518607497215271,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 43803
+    },
+    {
+      "epoch": 4.493844891259745,
+      "grad_norm": 0.08158674836158752,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 43806
+    },
+    {
+      "epoch": 4.494152646696758,
+      "grad_norm": 0.07762011885643005,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 43809
+    },
+    {
+      "epoch": 4.494460402133771,
+      "grad_norm": 0.09235764294862747,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 43812
+    },
+    {
+      "epoch": 4.494768157570784,
+      "grad_norm": 0.050333961844444275,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 43815
+    },
+    {
+      "epoch": 4.495075913007796,
+      "grad_norm": 0.0546598955988884,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 43818
+    },
+    {
+      "epoch": 4.495383668444809,
+      "grad_norm": 0.07909304648637772,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 43821
+    },
+    {
+      "epoch": 4.495691423881822,
+      "grad_norm": 0.10644279420375824,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 43824
+    },
+    {
+      "epoch": 4.495999179318835,
+      "grad_norm": 0.06348330527544022,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 43827
+    },
+    {
+      "epoch": 4.4963069347558475,
+      "grad_norm": 0.060767900198698044,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 43830
+    },
+    {
+      "epoch": 4.49661469019286,
+      "grad_norm": 0.050787363201379776,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 43833
+    },
+    {
+      "epoch": 4.496922445629873,
+      "grad_norm": 0.0362878292798996,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 43836
+    },
+    {
+      "epoch": 4.497230201066886,
+      "grad_norm": 0.07702784240245819,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 43839
+    },
+    {
+      "epoch": 4.497537956503899,
+      "grad_norm": 0.05007346719503403,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 43842
+    },
+    {
+      "epoch": 4.4978457119409105,
+      "grad_norm": 0.04794207960367203,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 43845
+    },
+    {
+      "epoch": 4.498153467377923,
+      "grad_norm": 0.04176037013530731,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 43848
+    },
+    {
+      "epoch": 4.498461222814936,
+      "grad_norm": 0.06070488691329956,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 43851
+    },
+    {
+      "epoch": 4.498768978251949,
+      "grad_norm": 0.0750289261341095,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 43854
+    },
+    {
+      "epoch": 4.499076733688962,
+      "grad_norm": 0.1204972118139267,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 43857
+    },
+    {
+      "epoch": 4.499384489125974,
+      "grad_norm": 0.03909352049231529,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 43860
+    },
+    {
+      "epoch": 4.499692244562987,
+      "grad_norm": 0.060046713799238205,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 43863
+    },
+    {
+      "epoch": 4.5,
+      "grad_norm": 0.03170366585254669,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 43866
+    },
+    {
+      "epoch": 4.500307755437013,
+      "grad_norm": 0.07817841321229935,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 43869
+    },
+    {
+      "epoch": 4.500615510874026,
+      "grad_norm": 0.06809035688638687,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 43872
+    },
+    {
+      "epoch": 4.500923266311038,
+      "grad_norm": 0.0353827029466629,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 43875
+    },
+    {
+      "epoch": 4.501231021748051,
+      "grad_norm": 0.05174241214990616,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 43878
+    },
+    {
+      "epoch": 4.501538777185064,
+      "grad_norm": 0.05049923434853554,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 43881
+    },
+    {
+      "epoch": 4.501846532622077,
+      "grad_norm": 0.04014773294329643,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 43884
+    },
+    {
+      "epoch": 4.502154288059089,
+      "grad_norm": 0.08103401213884354,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 43887
+    },
+    {
+      "epoch": 4.502462043496102,
+      "grad_norm": 0.04270040616393089,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 43890
+    },
+    {
+      "epoch": 4.502769798933114,
+      "grad_norm": 0.05039724335074425,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 43893
+    },
+    {
+      "epoch": 4.503077554370127,
+      "grad_norm": 0.10098660737276077,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 43896
+    },
+    {
+      "epoch": 4.50338530980714,
+      "grad_norm": 0.08645161241292953,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 43899
+    },
+    {
+      "epoch": 4.5036930652441525,
+      "grad_norm": 0.06836593151092529,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 43902
+    },
+    {
+      "epoch": 4.504000820681165,
+      "grad_norm": 0.09137406945228577,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 43905
+    },
+    {
+      "epoch": 4.504308576118178,
+      "grad_norm": 0.04432569816708565,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 43908
+    },
+    {
+      "epoch": 4.504616331555191,
+      "grad_norm": 0.03743622079491615,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 43911
+    },
+    {
+      "epoch": 4.504924086992204,
+      "grad_norm": 0.09280388057231903,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 43914
+    },
+    {
+      "epoch": 4.505231842429216,
+      "grad_norm": 0.06827981770038605,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 43917
+    },
+    {
+      "epoch": 4.505539597866229,
+      "grad_norm": 0.13798165321350098,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 43920
+    },
+    {
+      "epoch": 4.505847353303242,
+      "grad_norm": 0.046148303896188736,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 43923
+    },
+    {
+      "epoch": 4.506155108740255,
+      "grad_norm": 0.06119319424033165,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 43926
+    },
+    {
+      "epoch": 4.5064628641772675,
+      "grad_norm": 0.04381044954061508,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 43929
+    },
+    {
+      "epoch": 4.506770619614279,
+      "grad_norm": 0.043389271944761276,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 43932
+    },
+    {
+      "epoch": 4.507078375051293,
+      "grad_norm": 0.0379410944879055,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 43935
+    },
+    {
+      "epoch": 4.507386130488305,
+      "grad_norm": 0.043701719492673874,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 43938
+    },
+    {
+      "epoch": 4.507693885925318,
+      "grad_norm": 0.11455623805522919,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 43941
+    },
+    {
+      "epoch": 4.5080016413623305,
+      "grad_norm": 0.06367077678442001,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 43944
+    },
+    {
+      "epoch": 4.508309396799343,
+      "grad_norm": 0.12104804813861847,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 43947
+    },
+    {
+      "epoch": 4.508617152236356,
+      "grad_norm": 0.08374746888875961,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 43950
+    },
+    {
+      "epoch": 4.508924907673369,
+      "grad_norm": 0.12107446789741516,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 43953
+    },
+    {
+      "epoch": 4.509232663110382,
+      "grad_norm": 0.043210193514823914,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 43956
+    },
+    {
+      "epoch": 4.509540418547394,
+      "grad_norm": 0.05984468385577202,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 43959
+    },
+    {
+      "epoch": 4.509848173984407,
+      "grad_norm": 0.04827868938446045,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 43962
+    },
+    {
+      "epoch": 4.51015592942142,
+      "grad_norm": 0.04763716831803322,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 43965
+    },
+    {
+      "epoch": 4.510463684858433,
+      "grad_norm": 0.045888885855674744,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 43968
+    },
+    {
+      "epoch": 4.5107714402954455,
+      "grad_norm": 0.06434401869773865,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 43971
+    },
+    {
+      "epoch": 4.511079195732458,
+      "grad_norm": 0.09480862319469452,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 43974
+    },
+    {
+      "epoch": 4.51138695116947,
+      "grad_norm": 0.10739507526159286,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 43977
+    },
+    {
+      "epoch": 4.511694706606484,
+      "grad_norm": 0.06446696072816849,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 43980
+    },
+    {
+      "epoch": 4.512002462043496,
+      "grad_norm": 0.0567287839949131,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 43983
+    },
+    {
+      "epoch": 4.5123102174805085,
+      "grad_norm": 0.0359378419816494,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 43986
+    },
+    {
+      "epoch": 4.512617972917521,
+      "grad_norm": 0.052227430045604706,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 43989
+    },
+    {
+      "epoch": 4.512925728354534,
+      "grad_norm": 0.058354392647743225,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 43992
+    },
+    {
+      "epoch": 4.513233483791547,
+      "grad_norm": 0.06100015714764595,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 43995
+    },
+    {
+      "epoch": 4.51354123922856,
+      "grad_norm": 0.0857095718383789,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 43998
+    },
+    {
+      "epoch": 4.513848994665572,
+      "grad_norm": 0.11230607330799103,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 44001
+    },
+    {
+      "epoch": 4.514156750102585,
+      "grad_norm": 0.082217276096344,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 44004
+    },
+    {
+      "epoch": 4.514464505539598,
+      "grad_norm": 0.07966562360525131,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 44007
+    },
+    {
+      "epoch": 4.514772260976611,
+      "grad_norm": 0.0608987994492054,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 44010
+    },
+    {
+      "epoch": 4.5150800164136236,
+      "grad_norm": 0.029933054000139236,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 44013
+    },
+    {
+      "epoch": 4.515387771850636,
+      "grad_norm": 0.04696916788816452,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 44016
+    },
+    {
+      "epoch": 4.515695527287649,
+      "grad_norm": 0.0494072362780571,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 44019
+    },
+    {
+      "epoch": 4.516003282724661,
+      "grad_norm": 0.06069015711545944,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 44022
+    },
+    {
+      "epoch": 4.516311038161675,
+      "grad_norm": 0.05014580115675926,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 44025
+    },
+    {
+      "epoch": 4.516618793598687,
+      "grad_norm": 0.04753846675157547,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 44028
+    },
+    {
+      "epoch": 4.516926549035699,
+      "grad_norm": 0.053366005420684814,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 44031
+    },
+    {
+      "epoch": 4.517234304472712,
+      "grad_norm": 0.09559368342161179,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 44034
+    },
+    {
+      "epoch": 4.517542059909725,
+      "grad_norm": 0.12470517307519913,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 44037
+    },
+    {
+      "epoch": 4.517849815346738,
+      "grad_norm": 0.0809641033411026,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 44040
+    },
+    {
+      "epoch": 4.5181575707837505,
+      "grad_norm": 0.04899699613451958,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 44043
+    },
+    {
+      "epoch": 4.518465326220763,
+      "grad_norm": 0.06368394196033478,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 44046
+    },
+    {
+      "epoch": 4.518773081657776,
+      "grad_norm": 0.06713195890188217,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 44049
+    },
+    {
+      "epoch": 4.519080837094789,
+      "grad_norm": 0.12240014225244522,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 44052
+    },
+    {
+      "epoch": 4.519388592531802,
+      "grad_norm": 0.159604012966156,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 44055
+    },
+    {
+      "epoch": 4.519696347968814,
+      "grad_norm": 0.09968844056129456,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 44058
+    },
+    {
+      "epoch": 4.520004103405827,
+      "grad_norm": 0.1151421070098877,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 44061
+    },
+    {
+      "epoch": 4.52031185884284,
+      "grad_norm": 0.117409847676754,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 44064
+    },
+    {
+      "epoch": 4.520619614279852,
+      "grad_norm": 0.05031849071383476,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 44067
+    },
+    {
+      "epoch": 4.5209273697168655,
+      "grad_norm": 0.04156704619526863,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 44070
+    },
+    {
+      "epoch": 4.521235125153877,
+      "grad_norm": 0.04768043011426926,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 44073
+    },
+    {
+      "epoch": 4.52154288059089,
+      "grad_norm": 0.04492770880460739,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 44076
+    },
+    {
+      "epoch": 4.521850636027903,
+      "grad_norm": 0.0356624610722065,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 44079
+    },
+    {
+      "epoch": 4.522158391464916,
+      "grad_norm": 0.06383637338876724,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 44082
+    },
+    {
+      "epoch": 4.5224661469019285,
+      "grad_norm": 0.09323472529649734,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 44085
+    },
+    {
+      "epoch": 4.522773902338941,
+      "grad_norm": 0.08748291432857513,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 44088
+    },
+    {
+      "epoch": 4.523081657775954,
+      "grad_norm": 0.12068930268287659,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 44091
+    },
+    {
+      "epoch": 4.523389413212967,
+      "grad_norm": 0.12926380336284637,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 44094
+    },
+    {
+      "epoch": 4.52369716864998,
+      "grad_norm": 0.053785938769578934,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 44097
+    },
+    {
+      "epoch": 4.524004924086992,
+      "grad_norm": 0.056706249713897705,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 44100
+    },
+    {
+      "epoch": 4.524312679524005,
+      "grad_norm": 0.040739428251981735,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 44103
+    },
+    {
+      "epoch": 4.524620434961018,
+      "grad_norm": 0.07002007216215134,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 44106
+    },
+    {
+      "epoch": 4.524928190398031,
+      "grad_norm": 0.08624965697526932,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 44109
+    },
+    {
+      "epoch": 4.525235945835043,
+      "grad_norm": 0.05654805898666382,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 44112
+    },
+    {
+      "epoch": 4.525543701272056,
+      "grad_norm": 0.0629354789853096,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 44115
+    },
+    {
+      "epoch": 4.525851456709068,
+      "grad_norm": 0.10447383671998978,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 44118
+    },
+    {
+      "epoch": 4.526159212146081,
+      "grad_norm": 0.08281921595335007,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 44121
+    },
+    {
+      "epoch": 4.526466967583094,
+      "grad_norm": 0.10570526868104935,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 44124
+    },
+    {
+      "epoch": 4.5267747230201065,
+      "grad_norm": 0.040542326867580414,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 44127
+    },
+    {
+      "epoch": 4.527082478457119,
+      "grad_norm": 0.060009557753801346,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 44130
+    },
+    {
+      "epoch": 4.527390233894132,
+      "grad_norm": 0.09265922009944916,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 44133
+    },
+    {
+      "epoch": 4.527697989331145,
+      "grad_norm": 0.0457131452858448,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 44136
+    },
+    {
+      "epoch": 4.528005744768158,
+      "grad_norm": 0.038852475583553314,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 44139
+    },
+    {
+      "epoch": 4.52831350020517,
+      "grad_norm": 0.03757132962346077,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 44142
+    },
+    {
+      "epoch": 4.528621255642183,
+      "grad_norm": 0.04073426127433777,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 44145
+    },
+    {
+      "epoch": 4.528929011079196,
+      "grad_norm": 0.05840800702571869,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 44148
+    },
+    {
+      "epoch": 4.529236766516209,
+      "grad_norm": 0.06951642036437988,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 44151
+    },
+    {
+      "epoch": 4.5295445219532215,
+      "grad_norm": 0.062115710228681564,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 44154
+    },
+    {
+      "epoch": 4.529852277390233,
+      "grad_norm": 0.0439622737467289,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 44157
+    },
+    {
+      "epoch": 4.530160032827247,
+      "grad_norm": 0.06429772078990936,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 44160
+    },
+    {
+      "epoch": 4.530467788264259,
+      "grad_norm": 0.1072159856557846,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 44163
+    },
+    {
+      "epoch": 4.530775543701272,
+      "grad_norm": 0.08709017932415009,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 44166
+    },
+    {
+      "epoch": 4.531083299138285,
+      "grad_norm": 0.07494400441646576,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 44169
+    },
+    {
+      "epoch": 4.531391054575297,
+      "grad_norm": 0.09518608450889587,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 44172
+    },
+    {
+      "epoch": 4.53169881001231,
+      "grad_norm": 0.04578256234526634,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 44175
+    },
+    {
+      "epoch": 4.532006565449323,
+      "grad_norm": 0.03707175329327583,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 44178
+    },
+    {
+      "epoch": 4.532314320886336,
+      "grad_norm": 0.042071472853422165,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 44181
+    },
+    {
+      "epoch": 4.5326220763233485,
+      "grad_norm": 0.05591168627142906,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 44184
+    },
+    {
+      "epoch": 4.532929831760361,
+      "grad_norm": 0.04541923105716705,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 44187
+    },
+    {
+      "epoch": 4.533237587197374,
+      "grad_norm": 0.07019765675067902,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 44190
+    },
+    {
+      "epoch": 4.533545342634387,
+      "grad_norm": 0.04877987504005432,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 44193
+    },
+    {
+      "epoch": 4.5338530980714,
+      "grad_norm": 0.04220633581280708,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 44196
+    },
+    {
+      "epoch": 4.534160853508412,
+      "grad_norm": 0.17436882853507996,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 44199
+    },
+    {
+      "epoch": 4.534468608945424,
+      "grad_norm": 0.03670203685760498,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 44202
+    },
+    {
+      "epoch": 4.534776364382438,
+      "grad_norm": 0.04544707387685776,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 44205
+    },
+    {
+      "epoch": 4.53508411981945,
+      "grad_norm": 0.03833407908678055,
+      "learning_rate": 0.01,
+      "loss": 2.0196,
+      "step": 44208
+    },
+    {
+      "epoch": 4.535391875256463,
+      "grad_norm": 0.03395809605717659,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 44211
+    },
+    {
+      "epoch": 4.535699630693475,
+      "grad_norm": 0.04812290519475937,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 44214
+    },
+    {
+      "epoch": 4.536007386130488,
+      "grad_norm": 0.04580771178007126,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 44217
+    },
+    {
+      "epoch": 4.536315141567501,
+      "grad_norm": 0.04926560819149017,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 44220
+    },
+    {
+      "epoch": 4.536622897004514,
+      "grad_norm": 0.08950170129537582,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 44223
+    },
+    {
+      "epoch": 4.5369306524415265,
+      "grad_norm": 0.04416525736451149,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 44226
+    },
+    {
+      "epoch": 4.537238407878539,
+      "grad_norm": 0.04912101849913597,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 44229
+    },
+    {
+      "epoch": 4.537546163315552,
+      "grad_norm": 0.039088208228349686,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 44232
+    },
+    {
+      "epoch": 4.537853918752565,
+      "grad_norm": 0.03769220411777496,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 44235
+    },
+    {
+      "epoch": 4.538161674189578,
+      "grad_norm": 0.14437700808048248,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 44238
+    },
+    {
+      "epoch": 4.53846942962659,
+      "grad_norm": 0.09406454116106033,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 44241
+    },
+    {
+      "epoch": 4.538777185063603,
+      "grad_norm": 0.0553220696747303,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 44244
+    },
+    {
+      "epoch": 4.539084940500615,
+      "grad_norm": 0.04257743060588837,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 44247
+    },
+    {
+      "epoch": 4.539392695937629,
+      "grad_norm": 0.05212993547320366,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 44250
+    },
+    {
+      "epoch": 4.539700451374641,
+      "grad_norm": 0.050963662564754486,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 44253
+    },
+    {
+      "epoch": 4.540008206811653,
+      "grad_norm": 0.04554356262087822,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 44256
+    },
+    {
+      "epoch": 4.540315962248666,
+      "grad_norm": 0.11272426694631577,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 44259
+    },
+    {
+      "epoch": 4.540623717685679,
+      "grad_norm": 0.06912017613649368,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 44262
+    },
+    {
+      "epoch": 4.540931473122692,
+      "grad_norm": 0.057526472955942154,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 44265
+    },
+    {
+      "epoch": 4.5412392285597045,
+      "grad_norm": 0.035935260355472565,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 44268
+    },
+    {
+      "epoch": 4.541546983996717,
+      "grad_norm": 0.06651175767183304,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 44271
+    },
+    {
+      "epoch": 4.54185473943373,
+      "grad_norm": 0.11073578149080276,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 44274
+    },
+    {
+      "epoch": 4.542162494870743,
+      "grad_norm": 0.08459115028381348,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 44277
+    },
+    {
+      "epoch": 4.542470250307756,
+      "grad_norm": 0.04882695525884628,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 44280
+    },
+    {
+      "epoch": 4.542778005744768,
+      "grad_norm": 0.0500728003680706,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 44283
+    },
+    {
+      "epoch": 4.543085761181781,
+      "grad_norm": 0.03797118738293648,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 44286
+    },
+    {
+      "epoch": 4.543393516618794,
+      "grad_norm": 0.046217989176511765,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 44289
+    },
+    {
+      "epoch": 4.543701272055806,
+      "grad_norm": 0.057793501764535904,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 44292
+    },
+    {
+      "epoch": 4.544009027492819,
+      "grad_norm": 0.03469870239496231,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 44295
+    },
+    {
+      "epoch": 4.544316782929831,
+      "grad_norm": 0.11007340252399445,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 44298
+    },
+    {
+      "epoch": 4.544624538366844,
+      "grad_norm": 0.07559779286384583,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 44301
+    },
+    {
+      "epoch": 4.544932293803857,
+      "grad_norm": 0.10915376991033554,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 44304
+    },
+    {
+      "epoch": 4.54524004924087,
+      "grad_norm": 0.07973102480173111,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 44307
+    },
+    {
+      "epoch": 4.5455478046778826,
+      "grad_norm": 0.06999395787715912,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 44310
+    },
+    {
+      "epoch": 4.545855560114895,
+      "grad_norm": 0.05275473743677139,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 44313
+    },
+    {
+      "epoch": 4.546163315551908,
+      "grad_norm": 0.054669059813022614,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 44316
+    },
+    {
+      "epoch": 4.546471070988921,
+      "grad_norm": 0.046913836151361465,
+      "learning_rate": 0.01,
+      "loss": 2.041,
+      "step": 44319
+    },
+    {
+      "epoch": 4.546778826425934,
+      "grad_norm": 0.043503716588020325,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 44322
+    },
+    {
+      "epoch": 4.5470865818629465,
+      "grad_norm": 0.058872975409030914,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 44325
+    },
+    {
+      "epoch": 4.547394337299959,
+      "grad_norm": 0.08830869197845459,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 44328
+    },
+    {
+      "epoch": 4.547702092736972,
+      "grad_norm": 0.0850677490234375,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 44331
+    },
+    {
+      "epoch": 4.548009848173985,
+      "grad_norm": 0.058975327759981155,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 44334
+    },
+    {
+      "epoch": 4.548317603610997,
+      "grad_norm": 0.04586830735206604,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 44337
+    },
+    {
+      "epoch": 4.5486253590480095,
+      "grad_norm": 0.046169981360435486,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 44340
+    },
+    {
+      "epoch": 4.548933114485022,
+      "grad_norm": 0.06343907862901688,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 44343
+    },
+    {
+      "epoch": 4.549240869922035,
+      "grad_norm": 0.03225269168615341,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 44346
+    },
+    {
+      "epoch": 4.549548625359048,
+      "grad_norm": 0.0852704718708992,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 44349
+    },
+    {
+      "epoch": 4.549856380796061,
+      "grad_norm": 0.058122653514146805,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 44352
+    },
+    {
+      "epoch": 4.550164136233073,
+      "grad_norm": 0.08065648376941681,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 44355
+    },
+    {
+      "epoch": 4.550471891670086,
+      "grad_norm": 0.04599373787641525,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 44358
+    },
+    {
+      "epoch": 4.550779647107099,
+      "grad_norm": 0.03361526504158974,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 44361
+    },
+    {
+      "epoch": 4.551087402544112,
+      "grad_norm": 0.04159310460090637,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 44364
+    },
+    {
+      "epoch": 4.5513951579811245,
+      "grad_norm": 0.06349869817495346,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 44367
+    },
+    {
+      "epoch": 4.551702913418137,
+      "grad_norm": 0.08402734994888306,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 44370
+    },
+    {
+      "epoch": 4.55201066885515,
+      "grad_norm": 0.0695672482252121,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 44373
+    },
+    {
+      "epoch": 4.552318424292163,
+      "grad_norm": 0.038909196853637695,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 44376
+    },
+    {
+      "epoch": 4.552626179729176,
+      "grad_norm": 0.032846856862306595,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 44379
+    },
+    {
+      "epoch": 4.5529339351661875,
+      "grad_norm": 0.03906947746872902,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 44382
+    },
+    {
+      "epoch": 4.5532416906032,
+      "grad_norm": 0.06816279888153076,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 44385
+    },
+    {
+      "epoch": 4.553549446040213,
+      "grad_norm": 0.13728150725364685,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 44388
+    },
+    {
+      "epoch": 4.553857201477226,
+      "grad_norm": 0.11365284025669098,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 44391
+    },
+    {
+      "epoch": 4.554164956914239,
+      "grad_norm": 0.058936670422554016,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 44394
+    },
+    {
+      "epoch": 4.554472712351251,
+      "grad_norm": 0.044918566942214966,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 44397
+    },
+    {
+      "epoch": 4.554780467788264,
+      "grad_norm": 0.04904935136437416,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 44400
+    },
+    {
+      "epoch": 4.555088223225277,
+      "grad_norm": 0.04912208393216133,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 44403
+    },
+    {
+      "epoch": 4.55539597866229,
+      "grad_norm": 0.0992012619972229,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 44406
+    },
+    {
+      "epoch": 4.5557037340993025,
+      "grad_norm": 0.05080341920256615,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 44409
+    },
+    {
+      "epoch": 4.556011489536315,
+      "grad_norm": 0.10849186778068542,
+      "learning_rate": 0.01,
+      "loss": 2.0349,
+      "step": 44412
+    },
+    {
+      "epoch": 4.556319244973328,
+      "grad_norm": 0.04914950951933861,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 44415
+    },
+    {
+      "epoch": 4.556627000410341,
+      "grad_norm": 0.040149789303541183,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 44418
+    },
+    {
+      "epoch": 4.556934755847354,
+      "grad_norm": 0.04807689040899277,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 44421
+    },
+    {
+      "epoch": 4.557242511284366,
+      "grad_norm": 0.03420715779066086,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 44424
+    },
+    {
+      "epoch": 4.557550266721378,
+      "grad_norm": 0.033837419003248215,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 44427
+    },
+    {
+      "epoch": 4.557858022158391,
+      "grad_norm": 0.05432615429162979,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 44430
+    },
+    {
+      "epoch": 4.558165777595404,
+      "grad_norm": 0.09718727320432663,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 44433
+    },
+    {
+      "epoch": 4.558473533032417,
+      "grad_norm": 0.09546732157468796,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 44436
+    },
+    {
+      "epoch": 4.558781288469429,
+      "grad_norm": 0.04132693633437157,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 44439
+    },
+    {
+      "epoch": 4.559089043906442,
+      "grad_norm": 0.07769662886857986,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 44442
+    },
+    {
+      "epoch": 4.559396799343455,
+      "grad_norm": 0.08077646046876907,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 44445
+    },
+    {
+      "epoch": 4.559704554780468,
+      "grad_norm": 0.0781002938747406,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 44448
+    },
+    {
+      "epoch": 4.5600123102174805,
+      "grad_norm": 0.057897116988897324,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 44451
+    },
+    {
+      "epoch": 4.560320065654493,
+      "grad_norm": 0.06867967545986176,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 44454
+    },
+    {
+      "epoch": 4.560627821091506,
+      "grad_norm": 0.06109149381518364,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 44457
+    },
+    {
+      "epoch": 4.560935576528519,
+      "grad_norm": 0.038964226841926575,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 44460
+    },
+    {
+      "epoch": 4.561243331965532,
+      "grad_norm": 0.0609222911298275,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 44463
+    },
+    {
+      "epoch": 4.5615510874025444,
+      "grad_norm": 0.09428851306438446,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 44466
+    },
+    {
+      "epoch": 4.561858842839557,
+      "grad_norm": 0.09745607525110245,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 44469
+    },
+    {
+      "epoch": 4.562166598276569,
+      "grad_norm": 0.09413721412420273,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 44472
+    },
+    {
+      "epoch": 4.562474353713582,
+      "grad_norm": 0.0537935309112072,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 44475
+    },
+    {
+      "epoch": 4.562782109150595,
+      "grad_norm": 0.03104214183986187,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 44478
+    },
+    {
+      "epoch": 4.5630898645876075,
+      "grad_norm": 0.028320681303739548,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 44481
+    },
+    {
+      "epoch": 4.56339762002462,
+      "grad_norm": 0.09736626595258713,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 44484
+    },
+    {
+      "epoch": 4.563705375461633,
+      "grad_norm": 0.03548412024974823,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 44487
+    },
+    {
+      "epoch": 4.564013130898646,
+      "grad_norm": 0.11299438029527664,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 44490
+    },
+    {
+      "epoch": 4.564320886335659,
+      "grad_norm": 0.07769119739532471,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 44493
+    },
+    {
+      "epoch": 4.564628641772671,
+      "grad_norm": 0.04660061001777649,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 44496
+    },
+    {
+      "epoch": 4.564936397209684,
+      "grad_norm": 0.045153431594371796,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 44499
+    },
+    {
+      "epoch": 4.565244152646697,
+      "grad_norm": 0.040483929216861725,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 44502
+    },
+    {
+      "epoch": 4.56555190808371,
+      "grad_norm": 0.04947364702820778,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 44505
+    },
+    {
+      "epoch": 4.5658596635207225,
+      "grad_norm": 0.0688960924744606,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 44508
+    },
+    {
+      "epoch": 4.566167418957735,
+      "grad_norm": 0.11255665123462677,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 44511
+    },
+    {
+      "epoch": 4.566475174394748,
+      "grad_norm": 0.07215286791324615,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 44514
+    },
+    {
+      "epoch": 4.56678292983176,
+      "grad_norm": 0.07593537867069244,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 44517
+    },
+    {
+      "epoch": 4.567090685268773,
+      "grad_norm": 0.053777799010276794,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 44520
+    },
+    {
+      "epoch": 4.5673984407057855,
+      "grad_norm": 0.04630263149738312,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 44523
+    },
+    {
+      "epoch": 4.567706196142798,
+      "grad_norm": 0.1870114654302597,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 44526
+    },
+    {
+      "epoch": 4.568013951579811,
+      "grad_norm": 0.04411546140909195,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 44529
+    },
+    {
+      "epoch": 4.568321707016824,
+      "grad_norm": 0.04254088178277016,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 44532
+    },
+    {
+      "epoch": 4.568629462453837,
+      "grad_norm": 0.048210930079221725,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 44535
+    },
+    {
+      "epoch": 4.568937217890849,
+      "grad_norm": 0.046062175184488297,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 44538
+    },
+    {
+      "epoch": 4.569244973327862,
+      "grad_norm": 0.04656999930739403,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 44541
+    },
+    {
+      "epoch": 4.569552728764875,
+      "grad_norm": 0.04990769177675247,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 44544
+    },
+    {
+      "epoch": 4.569860484201888,
+      "grad_norm": 0.06637271493673325,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 44547
+    },
+    {
+      "epoch": 4.5701682396389005,
+      "grad_norm": 0.0446067713201046,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 44550
+    },
+    {
+      "epoch": 4.570475995075913,
+      "grad_norm": 0.044182851910591125,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 44553
+    },
+    {
+      "epoch": 4.570783750512926,
+      "grad_norm": 0.03427280858159065,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 44556
+    },
+    {
+      "epoch": 4.571091505949939,
+      "grad_norm": 0.0435144379734993,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 44559
+    },
+    {
+      "epoch": 4.571399261386951,
+      "grad_norm": 0.046480659395456314,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 44562
+    },
+    {
+      "epoch": 4.5717070168239635,
+      "grad_norm": 0.047410015016794205,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 44565
+    },
+    {
+      "epoch": 4.572014772260976,
+      "grad_norm": 0.0613541379570961,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 44568
+    },
+    {
+      "epoch": 4.572322527697989,
+      "grad_norm": 0.04605888947844505,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 44571
+    },
+    {
+      "epoch": 4.572630283135002,
+      "grad_norm": 0.08377258479595184,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 44574
+    },
+    {
+      "epoch": 4.572938038572015,
+      "grad_norm": 0.09191520512104034,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 44577
+    },
+    {
+      "epoch": 4.573245794009027,
+      "grad_norm": 0.11382446438074112,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 44580
+    },
+    {
+      "epoch": 4.57355354944604,
+      "grad_norm": 0.05597241222858429,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 44583
+    },
+    {
+      "epoch": 4.573861304883053,
+      "grad_norm": 0.044153302907943726,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 44586
+    },
+    {
+      "epoch": 4.574169060320066,
+      "grad_norm": 0.0449075847864151,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 44589
+    },
+    {
+      "epoch": 4.5744768157570785,
+      "grad_norm": 0.06252507120370865,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 44592
+    },
+    {
+      "epoch": 4.574784571194091,
+      "grad_norm": 0.047335173934698105,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 44595
+    },
+    {
+      "epoch": 4.575092326631104,
+      "grad_norm": 0.046538159251213074,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 44598
+    },
+    {
+      "epoch": 4.575400082068117,
+      "grad_norm": 0.043892424553632736,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 44601
+    },
+    {
+      "epoch": 4.57570783750513,
+      "grad_norm": 0.14133551716804504,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 44604
+    },
+    {
+      "epoch": 4.5760155929421416,
+      "grad_norm": 0.06392577290534973,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 44607
+    },
+    {
+      "epoch": 4.576323348379154,
+      "grad_norm": 0.07599660754203796,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 44610
+    },
+    {
+      "epoch": 4.576631103816167,
+      "grad_norm": 0.0688132718205452,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 44613
+    },
+    {
+      "epoch": 4.57693885925318,
+      "grad_norm": 0.05262705683708191,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 44616
+    },
+    {
+      "epoch": 4.577246614690193,
+      "grad_norm": 0.04527880623936653,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 44619
+    },
+    {
+      "epoch": 4.5775543701272055,
+      "grad_norm": 0.05383728817105293,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 44622
+    },
+    {
+      "epoch": 4.577862125564218,
+      "grad_norm": 0.07166964560747147,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 44625
+    },
+    {
+      "epoch": 4.578169881001231,
+      "grad_norm": 0.04192749038338661,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 44628
+    },
+    {
+      "epoch": 4.578477636438244,
+      "grad_norm": 0.06395778805017471,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 44631
+    },
+    {
+      "epoch": 4.578785391875257,
+      "grad_norm": 0.05522435903549194,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 44634
+    },
+    {
+      "epoch": 4.579093147312269,
+      "grad_norm": 0.0383891724050045,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 44637
+    },
+    {
+      "epoch": 4.579400902749282,
+      "grad_norm": 0.11526113003492355,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 44640
+    },
+    {
+      "epoch": 4.579708658186295,
+      "grad_norm": 0.04987471178174019,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 44643
+    },
+    {
+      "epoch": 4.580016413623308,
+      "grad_norm": 0.055467016994953156,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 44646
+    },
+    {
+      "epoch": 4.5803241690603205,
+      "grad_norm": 0.08823923766613007,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 44649
+    },
+    {
+      "epoch": 4.580631924497332,
+      "grad_norm": 0.06108856573700905,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 44652
+    },
+    {
+      "epoch": 4.580939679934345,
+      "grad_norm": 0.046790335327386856,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 44655
+    },
+    {
+      "epoch": 4.581247435371358,
+      "grad_norm": 0.04152047634124756,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 44658
+    },
+    {
+      "epoch": 4.581555190808371,
+      "grad_norm": 0.06719448417425156,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 44661
+    },
+    {
+      "epoch": 4.5818629462453835,
+      "grad_norm": 0.14338631927967072,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 44664
+    },
+    {
+      "epoch": 4.582170701682396,
+      "grad_norm": 0.06660101562738419,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 44667
+    },
+    {
+      "epoch": 4.582478457119409,
+      "grad_norm": 0.0552043579518795,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 44670
+    },
+    {
+      "epoch": 4.582786212556422,
+      "grad_norm": 0.08016394078731537,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 44673
+    },
+    {
+      "epoch": 4.583093967993435,
+      "grad_norm": 0.05147711932659149,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 44676
+    },
+    {
+      "epoch": 4.583401723430447,
+      "grad_norm": 0.044893477112054825,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 44679
+    },
+    {
+      "epoch": 4.58370947886746,
+      "grad_norm": 0.1312035769224167,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 44682
+    },
+    {
+      "epoch": 4.584017234304473,
+      "grad_norm": 0.0376642569899559,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 44685
+    },
+    {
+      "epoch": 4.584324989741486,
+      "grad_norm": 0.042604606598615646,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 44688
+    },
+    {
+      "epoch": 4.5846327451784985,
+      "grad_norm": 0.04277000203728676,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 44691
+    },
+    {
+      "epoch": 4.584940500615511,
+      "grad_norm": 0.07913113385438919,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 44694
+    },
+    {
+      "epoch": 4.585248256052523,
+      "grad_norm": 0.08000600337982178,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 44697
+    },
+    {
+      "epoch": 4.585556011489536,
+      "grad_norm": 0.05239587649703026,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 44700
+    },
+    {
+      "epoch": 4.585863766926549,
+      "grad_norm": 0.04813678562641144,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 44703
+    },
+    {
+      "epoch": 4.5861715223635615,
+      "grad_norm": 0.11590283364057541,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 44706
+    },
+    {
+      "epoch": 4.586479277800574,
+      "grad_norm": 0.19828790426254272,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 44709
+    },
+    {
+      "epoch": 4.586787033237587,
+      "grad_norm": 0.14478129148483276,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 44712
+    },
+    {
+      "epoch": 4.5870947886746,
+      "grad_norm": 0.042516618967056274,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 44715
+    },
+    {
+      "epoch": 4.587402544111613,
+      "grad_norm": 0.04690965265035629,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 44718
+    },
+    {
+      "epoch": 4.587710299548625,
+      "grad_norm": 0.035743579268455505,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 44721
+    },
+    {
+      "epoch": 4.588018054985638,
+      "grad_norm": 0.041939638555049896,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 44724
+    },
+    {
+      "epoch": 4.588325810422651,
+      "grad_norm": 0.039647046476602554,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 44727
+    },
+    {
+      "epoch": 4.588633565859664,
+      "grad_norm": 0.04442642256617546,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 44730
+    },
+    {
+      "epoch": 4.5889413212966765,
+      "grad_norm": 0.08688386529684067,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 44733
+    },
+    {
+      "epoch": 4.589249076733689,
+      "grad_norm": 0.03309828042984009,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 44736
+    },
+    {
+      "epoch": 4.589556832170702,
+      "grad_norm": 0.1094844713807106,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 44739
+    },
+    {
+      "epoch": 4.589864587607714,
+      "grad_norm": 0.04613770171999931,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 44742
+    },
+    {
+      "epoch": 4.590172343044727,
+      "grad_norm": 0.10529785603284836,
+      "learning_rate": 0.01,
+      "loss": 2.0468,
+      "step": 44745
+    },
+    {
+      "epoch": 4.5904800984817395,
+      "grad_norm": 0.09901240468025208,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 44748
+    },
+    {
+      "epoch": 4.590787853918752,
+      "grad_norm": 0.08486896753311157,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 44751
+    },
+    {
+      "epoch": 4.591095609355765,
+      "grad_norm": 0.05477188900113106,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 44754
+    },
+    {
+      "epoch": 4.591403364792778,
+      "grad_norm": 0.03555532172322273,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 44757
+    },
+    {
+      "epoch": 4.591711120229791,
+      "grad_norm": 0.0637480840086937,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 44760
+    },
+    {
+      "epoch": 4.5920188756668034,
+      "grad_norm": 0.09844323992729187,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 44763
+    },
+    {
+      "epoch": 4.592326631103816,
+      "grad_norm": 0.06749750673770905,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 44766
+    },
+    {
+      "epoch": 4.592634386540829,
+      "grad_norm": 0.0942719504237175,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 44769
+    },
+    {
+      "epoch": 4.592942141977842,
+      "grad_norm": 0.043663688004016876,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 44772
+    },
+    {
+      "epoch": 4.593249897414855,
+      "grad_norm": 0.1031889095902443,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 44775
+    },
+    {
+      "epoch": 4.593557652851867,
+      "grad_norm": 0.060684684664011,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 44778
+    },
+    {
+      "epoch": 4.59386540828888,
+      "grad_norm": 0.08744547516107559,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 44781
+    },
+    {
+      "epoch": 4.594173163725893,
+      "grad_norm": 0.05547311529517174,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 44784
+    },
+    {
+      "epoch": 4.594480919162905,
+      "grad_norm": 0.05069347843527794,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 44787
+    },
+    {
+      "epoch": 4.594788674599918,
+      "grad_norm": 0.11832018941640854,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 44790
+    },
+    {
+      "epoch": 4.59509643003693,
+      "grad_norm": 0.09115731716156006,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 44793
+    },
+    {
+      "epoch": 4.595404185473943,
+      "grad_norm": 0.03262989968061447,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 44796
+    },
+    {
+      "epoch": 4.595711940910956,
+      "grad_norm": 0.08880288898944855,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 44799
+    },
+    {
+      "epoch": 4.596019696347969,
+      "grad_norm": 0.059965986758470535,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 44802
+    },
+    {
+      "epoch": 4.5963274517849815,
+      "grad_norm": 0.040046051144599915,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 44805
+    },
+    {
+      "epoch": 4.596635207221994,
+      "grad_norm": 0.05327828601002693,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 44808
+    },
+    {
+      "epoch": 4.596942962659007,
+      "grad_norm": 0.0532514713704586,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 44811
+    },
+    {
+      "epoch": 4.59725071809602,
+      "grad_norm": 0.0502878837287426,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 44814
+    },
+    {
+      "epoch": 4.597558473533033,
+      "grad_norm": 0.05360707640647888,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 44817
+    },
+    {
+      "epoch": 4.597866228970045,
+      "grad_norm": 0.08648170530796051,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 44820
+    },
+    {
+      "epoch": 4.598173984407058,
+      "grad_norm": 0.056780677288770676,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 44823
+    },
+    {
+      "epoch": 4.598481739844071,
+      "grad_norm": 0.032956380397081375,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 44826
+    },
+    {
+      "epoch": 4.598789495281084,
+      "grad_norm": 0.04322326183319092,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 44829
+    },
+    {
+      "epoch": 4.599097250718096,
+      "grad_norm": 0.10878471285104752,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 44832
+    },
+    {
+      "epoch": 4.599405006155108,
+      "grad_norm": 0.041027724742889404,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 44835
+    },
+    {
+      "epoch": 4.599712761592121,
+      "grad_norm": 0.09541984647512436,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 44838
+    },
+    {
+      "epoch": 4.600020517029134,
+      "grad_norm": 0.0648130476474762,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 44841
+    },
+    {
+      "epoch": 4.600328272466147,
+      "grad_norm": 0.10848550498485565,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 44844
+    },
+    {
+      "epoch": 4.6006360279031595,
+      "grad_norm": 0.062398817390203476,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 44847
+    },
+    {
+      "epoch": 4.600943783340172,
+      "grad_norm": 0.05789678171277046,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 44850
+    },
+    {
+      "epoch": 4.601251538777185,
+      "grad_norm": 0.03198247775435448,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 44853
+    },
+    {
+      "epoch": 4.601559294214198,
+      "grad_norm": 0.05636049434542656,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 44856
+    },
+    {
+      "epoch": 4.601867049651211,
+      "grad_norm": 0.07471849024295807,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 44859
+    },
+    {
+      "epoch": 4.602174805088223,
+      "grad_norm": 0.09839174151420593,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 44862
+    },
+    {
+      "epoch": 4.602482560525236,
+      "grad_norm": 0.08253321796655655,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 44865
+    },
+    {
+      "epoch": 4.602790315962249,
+      "grad_norm": 0.06502492725849152,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 44868
+    },
+    {
+      "epoch": 4.603098071399262,
+      "grad_norm": 0.04798365756869316,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 44871
+    },
+    {
+      "epoch": 4.6034058268362745,
+      "grad_norm": 0.04243507608771324,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 44874
+    },
+    {
+      "epoch": 4.603713582273286,
+      "grad_norm": 0.03410327062010765,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 44877
+    },
+    {
+      "epoch": 4.604021337710299,
+      "grad_norm": 0.09597765654325485,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 44880
+    },
+    {
+      "epoch": 4.604329093147312,
+      "grad_norm": 0.06745591014623642,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 44883
+    },
+    {
+      "epoch": 4.604636848584325,
+      "grad_norm": 0.10422816127538681,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 44886
+    },
+    {
+      "epoch": 4.6049446040213375,
+      "grad_norm": 0.05743027105927467,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 44889
+    },
+    {
+      "epoch": 4.60525235945835,
+      "grad_norm": 0.04410535469651222,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 44892
+    },
+    {
+      "epoch": 4.605560114895363,
+      "grad_norm": 0.07853147387504578,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 44895
+    },
+    {
+      "epoch": 4.605867870332376,
+      "grad_norm": 0.056344617158174515,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 44898
+    },
+    {
+      "epoch": 4.606175625769389,
+      "grad_norm": 0.045788802206516266,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 44901
+    },
+    {
+      "epoch": 4.606483381206401,
+      "grad_norm": 0.07213722914457321,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 44904
+    },
+    {
+      "epoch": 4.606791136643414,
+      "grad_norm": 0.08182168006896973,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 44907
+    },
+    {
+      "epoch": 4.607098892080427,
+      "grad_norm": 0.06299345195293427,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 44910
+    },
+    {
+      "epoch": 4.60740664751744,
+      "grad_norm": 0.046955596655607224,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 44913
+    },
+    {
+      "epoch": 4.607714402954453,
+      "grad_norm": 0.05844559147953987,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 44916
+    },
+    {
+      "epoch": 4.608022158391465,
+      "grad_norm": 0.04892998933792114,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 44919
+    },
+    {
+      "epoch": 4.608329913828477,
+      "grad_norm": 0.040492888540029526,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 44922
+    },
+    {
+      "epoch": 4.60863766926549,
+      "grad_norm": 0.043649546802043915,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 44925
+    },
+    {
+      "epoch": 4.608945424702503,
+      "grad_norm": 0.19346511363983154,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 44928
+    },
+    {
+      "epoch": 4.609253180139516,
+      "grad_norm": 0.0724545568227768,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 44931
+    },
+    {
+      "epoch": 4.609560935576528,
+      "grad_norm": 0.048261623829603195,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 44934
+    },
+    {
+      "epoch": 4.609868691013541,
+      "grad_norm": 0.06115753576159477,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 44937
+    },
+    {
+      "epoch": 4.610176446450554,
+      "grad_norm": 0.05957823619246483,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 44940
+    },
+    {
+      "epoch": 4.610484201887567,
+      "grad_norm": 0.08261608332395554,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 44943
+    },
+    {
+      "epoch": 4.6107919573245795,
+      "grad_norm": 0.04762962833046913,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 44946
+    },
+    {
+      "epoch": 4.611099712761592,
+      "grad_norm": 0.06040159985423088,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 44949
+    },
+    {
+      "epoch": 4.611407468198605,
+      "grad_norm": 0.06975025683641434,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 44952
+    },
+    {
+      "epoch": 4.611715223635618,
+      "grad_norm": 0.06275621056556702,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 44955
+    },
+    {
+      "epoch": 4.612022979072631,
+      "grad_norm": 0.04177596792578697,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 44958
+    },
+    {
+      "epoch": 4.612330734509643,
+      "grad_norm": 0.05913246050477028,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 44961
+    },
+    {
+      "epoch": 4.612638489946656,
+      "grad_norm": 0.05337538942694664,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 44964
+    },
+    {
+      "epoch": 4.612946245383668,
+      "grad_norm": 0.04543198645114899,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 44967
+    },
+    {
+      "epoch": 4.613254000820681,
+      "grad_norm": 0.05056144669651985,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 44970
+    },
+    {
+      "epoch": 4.613561756257694,
+      "grad_norm": 0.2109542340040207,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 44973
+    },
+    {
+      "epoch": 4.613869511694706,
+      "grad_norm": 0.11741621792316437,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 44976
+    },
+    {
+      "epoch": 4.614177267131719,
+      "grad_norm": 0.06218573451042175,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 44979
+    },
+    {
+      "epoch": 4.614485022568732,
+      "grad_norm": 0.04318585991859436,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 44982
+    },
+    {
+      "epoch": 4.614792778005745,
+      "grad_norm": 0.05122276023030281,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 44985
+    },
+    {
+      "epoch": 4.6151005334427575,
+      "grad_norm": 0.06790675967931747,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 44988
+    },
+    {
+      "epoch": 4.61540828887977,
+      "grad_norm": 0.06424128264188766,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 44991
+    },
+    {
+      "epoch": 4.615716044316783,
+      "grad_norm": 0.055609673261642456,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 44994
+    },
+    {
+      "epoch": 4.616023799753796,
+      "grad_norm": 0.04479989409446716,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 44997
+    },
+    {
+      "epoch": 4.616331555190809,
+      "grad_norm": 0.046773526817560196,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 45000
+    },
+    {
+      "epoch": 4.616639310627821,
+      "grad_norm": 0.044277340173721313,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 45003
+    },
+    {
+      "epoch": 4.616947066064834,
+      "grad_norm": 0.06604879349470139,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 45006
+    },
+    {
+      "epoch": 4.617254821501847,
+      "grad_norm": 0.07454725354909897,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 45009
+    },
+    {
+      "epoch": 4.617562576938859,
+      "grad_norm": 0.09837469458580017,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 45012
+    },
+    {
+      "epoch": 4.617870332375872,
+      "grad_norm": 0.12640568614006042,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 45015
+    },
+    {
+      "epoch": 4.618178087812884,
+      "grad_norm": 0.06357913464307785,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 45018
+    },
+    {
+      "epoch": 4.618485843249897,
+      "grad_norm": 0.05319363251328468,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 45021
+    },
+    {
+      "epoch": 4.61879359868691,
+      "grad_norm": 0.05755499005317688,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 45024
+    },
+    {
+      "epoch": 4.619101354123923,
+      "grad_norm": 0.03921099379658699,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 45027
+    },
+    {
+      "epoch": 4.6194091095609355,
+      "grad_norm": 0.062433432787656784,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 45030
+    },
+    {
+      "epoch": 4.619716864997948,
+      "grad_norm": 0.048542365431785583,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 45033
+    },
+    {
+      "epoch": 4.620024620434961,
+      "grad_norm": 0.12152877449989319,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 45036
+    },
+    {
+      "epoch": 4.620332375871974,
+      "grad_norm": 0.052054520696401596,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 45039
+    },
+    {
+      "epoch": 4.620640131308987,
+      "grad_norm": 0.11336644738912582,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 45042
+    },
+    {
+      "epoch": 4.620947886745999,
+      "grad_norm": 0.05044134706258774,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 45045
+    },
+    {
+      "epoch": 4.621255642183012,
+      "grad_norm": 0.06728419661521912,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 45048
+    },
+    {
+      "epoch": 4.621563397620025,
+      "grad_norm": 0.0826968103647232,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 45051
+    },
+    {
+      "epoch": 4.621871153057038,
+      "grad_norm": 0.07135532796382904,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 45054
+    },
+    {
+      "epoch": 4.62217890849405,
+      "grad_norm": 0.05112279951572418,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 45057
+    },
+    {
+      "epoch": 4.6224866639310624,
+      "grad_norm": 0.11058257520198822,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 45060
+    },
+    {
+      "epoch": 4.622794419368075,
+      "grad_norm": 0.12000041455030441,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 45063
+    },
+    {
+      "epoch": 4.623102174805088,
+      "grad_norm": 0.06073766201734543,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 45066
+    },
+    {
+      "epoch": 4.623409930242101,
+      "grad_norm": 0.055321842432022095,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 45069
+    },
+    {
+      "epoch": 4.623717685679114,
+      "grad_norm": 0.04220004379749298,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 45072
+    },
+    {
+      "epoch": 4.624025441116126,
+      "grad_norm": 0.03700163960456848,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 45075
+    },
+    {
+      "epoch": 4.624333196553139,
+      "grad_norm": 0.108041912317276,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 45078
+    },
+    {
+      "epoch": 4.624640951990152,
+      "grad_norm": 0.04682699963450432,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 45081
+    },
+    {
+      "epoch": 4.624948707427165,
+      "grad_norm": 0.07901785522699356,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 45084
+    },
+    {
+      "epoch": 4.6252564628641775,
+      "grad_norm": 0.07119090855121613,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 45087
+    },
+    {
+      "epoch": 4.62556421830119,
+      "grad_norm": 0.08904542028903961,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 45090
+    },
+    {
+      "epoch": 4.625871973738203,
+      "grad_norm": 0.06656675785779953,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 45093
+    },
+    {
+      "epoch": 4.626179729175216,
+      "grad_norm": 0.06560304015874863,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 45096
+    },
+    {
+      "epoch": 4.626487484612229,
+      "grad_norm": 0.05664901062846184,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 45099
+    },
+    {
+      "epoch": 4.6267952400492405,
+      "grad_norm": 0.0491751953959465,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 45102
+    },
+    {
+      "epoch": 4.627102995486253,
+      "grad_norm": 0.11787789314985275,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 45105
+    },
+    {
+      "epoch": 4.627410750923266,
+      "grad_norm": 0.04068392515182495,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 45108
+    },
+    {
+      "epoch": 4.627718506360279,
+      "grad_norm": 0.03958519548177719,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 45111
+    },
+    {
+      "epoch": 4.628026261797292,
+      "grad_norm": 0.05667821317911148,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 45114
+    },
+    {
+      "epoch": 4.628334017234304,
+      "grad_norm": 0.04660872370004654,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 45117
+    },
+    {
+      "epoch": 4.628641772671317,
+      "grad_norm": 0.05565042048692703,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 45120
+    },
+    {
+      "epoch": 4.62894952810833,
+      "grad_norm": 0.05158378556370735,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 45123
+    },
+    {
+      "epoch": 4.629257283545343,
+      "grad_norm": 0.04500264301896095,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 45126
+    },
+    {
+      "epoch": 4.6295650389823555,
+      "grad_norm": 0.03781789168715477,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 45129
+    },
+    {
+      "epoch": 4.629872794419368,
+      "grad_norm": 0.09032827615737915,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 45132
+    },
+    {
+      "epoch": 4.630180549856381,
+      "grad_norm": 0.09161376953125,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 45135
+    },
+    {
+      "epoch": 4.630488305293394,
+      "grad_norm": 0.07650554925203323,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 45138
+    },
+    {
+      "epoch": 4.630796060730407,
+      "grad_norm": 0.07098524272441864,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 45141
+    },
+    {
+      "epoch": 4.631103816167419,
+      "grad_norm": 0.053486138582229614,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 45144
+    },
+    {
+      "epoch": 4.631411571604431,
+      "grad_norm": 0.06696584820747375,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 45147
+    },
+    {
+      "epoch": 4.631719327041444,
+      "grad_norm": 0.08330658078193665,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 45150
+    },
+    {
+      "epoch": 4.632027082478457,
+      "grad_norm": 0.05537624657154083,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 45153
+    },
+    {
+      "epoch": 4.63233483791547,
+      "grad_norm": 0.07297627627849579,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 45156
+    },
+    {
+      "epoch": 4.632642593352482,
+      "grad_norm": 0.07492797076702118,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 45159
+    },
+    {
+      "epoch": 4.632950348789495,
+      "grad_norm": 0.06871726363897324,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 45162
+    },
+    {
+      "epoch": 4.633258104226508,
+      "grad_norm": 0.04638645797967911,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 45165
+    },
+    {
+      "epoch": 4.633565859663521,
+      "grad_norm": 0.10735375434160233,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 45168
+    },
+    {
+      "epoch": 4.6338736151005335,
+      "grad_norm": 0.1550443023443222,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 45171
+    },
+    {
+      "epoch": 4.634181370537546,
+      "grad_norm": 0.14652806520462036,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 45174
+    },
+    {
+      "epoch": 4.634489125974559,
+      "grad_norm": 0.06344394385814667,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 45177
+    },
+    {
+      "epoch": 4.634796881411572,
+      "grad_norm": 0.03594981133937836,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 45180
+    },
+    {
+      "epoch": 4.635104636848585,
+      "grad_norm": 0.034205902367830276,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 45183
+    },
+    {
+      "epoch": 4.635412392285597,
+      "grad_norm": 0.041966572403907776,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 45186
+    },
+    {
+      "epoch": 4.63572014772261,
+      "grad_norm": 0.05192418769001961,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 45189
+    },
+    {
+      "epoch": 4.636027903159622,
+      "grad_norm": 0.04771101847290993,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 45192
+    },
+    {
+      "epoch": 4.636335658596635,
+      "grad_norm": 0.05367758497595787,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 45195
+    },
+    {
+      "epoch": 4.636643414033648,
+      "grad_norm": 0.08973217755556107,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 45198
+    },
+    {
+      "epoch": 4.63695116947066,
+      "grad_norm": 0.06793297082185745,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 45201
+    },
+    {
+      "epoch": 4.637258924907673,
+      "grad_norm": 0.06299318373203278,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 45204
+    },
+    {
+      "epoch": 4.637566680344686,
+      "grad_norm": 0.106409452855587,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 45207
+    },
+    {
+      "epoch": 4.637874435781699,
+      "grad_norm": 0.044573575258255005,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 45210
+    },
+    {
+      "epoch": 4.638182191218712,
+      "grad_norm": 0.04855922982096672,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 45213
+    },
+    {
+      "epoch": 4.638489946655724,
+      "grad_norm": 0.05310840159654617,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 45216
+    },
+    {
+      "epoch": 4.638797702092737,
+      "grad_norm": 0.1408100426197052,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 45219
+    },
+    {
+      "epoch": 4.63910545752975,
+      "grad_norm": 0.07140195369720459,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 45222
+    },
+    {
+      "epoch": 4.639413212966763,
+      "grad_norm": 0.0864189863204956,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 45225
+    },
+    {
+      "epoch": 4.6397209684037755,
+      "grad_norm": 0.0877571702003479,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 45228
+    },
+    {
+      "epoch": 4.640028723840788,
+      "grad_norm": 0.06890790909528732,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 45231
+    },
+    {
+      "epoch": 4.640336479277801,
+      "grad_norm": 0.04007607325911522,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 45234
+    },
+    {
+      "epoch": 4.640644234714813,
+      "grad_norm": 0.09311459213495255,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 45237
+    },
+    {
+      "epoch": 4.640951990151826,
+      "grad_norm": 0.0721447765827179,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 45240
+    },
+    {
+      "epoch": 4.6412597455888385,
+      "grad_norm": 0.07306293398141861,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 45243
+    },
+    {
+      "epoch": 4.641567501025851,
+      "grad_norm": 0.044194191694259644,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 45246
+    },
+    {
+      "epoch": 4.641875256462864,
+      "grad_norm": 0.07482686638832092,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 45249
+    },
+    {
+      "epoch": 4.642183011899877,
+      "grad_norm": 0.05982416495680809,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 45252
+    },
+    {
+      "epoch": 4.64249076733689,
+      "grad_norm": 0.06082857400178909,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 45255
+    },
+    {
+      "epoch": 4.642798522773902,
+      "grad_norm": 0.12249299138784409,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 45258
+    },
+    {
+      "epoch": 4.643106278210915,
+      "grad_norm": 0.04660319164395332,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 45261
+    },
+    {
+      "epoch": 4.643414033647928,
+      "grad_norm": 0.13198639452457428,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 45264
+    },
+    {
+      "epoch": 4.643721789084941,
+      "grad_norm": 0.11743400990962982,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 45267
+    },
+    {
+      "epoch": 4.6440295445219535,
+      "grad_norm": 0.08904967457056046,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 45270
+    },
+    {
+      "epoch": 4.644337299958966,
+      "grad_norm": 0.051379457116127014,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 45273
+    },
+    {
+      "epoch": 4.644645055395979,
+      "grad_norm": 0.053696874529123306,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 45276
+    },
+    {
+      "epoch": 4.644952810832992,
+      "grad_norm": 0.03924264386296272,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 45279
+    },
+    {
+      "epoch": 4.645260566270004,
+      "grad_norm": 0.04208714887499809,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 45282
+    },
+    {
+      "epoch": 4.6455683217070165,
+      "grad_norm": 0.11654502153396606,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 45285
+    },
+    {
+      "epoch": 4.645876077144029,
+      "grad_norm": 0.06177401542663574,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 45288
+    },
+    {
+      "epoch": 4.646183832581042,
+      "grad_norm": 0.04343295469880104,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 45291
+    },
+    {
+      "epoch": 4.646491588018055,
+      "grad_norm": 0.036727067083120346,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 45294
+    },
+    {
+      "epoch": 4.646799343455068,
+      "grad_norm": 0.06670525670051575,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 45297
+    },
+    {
+      "epoch": 4.64710709889208,
+      "grad_norm": 0.08610829710960388,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 45300
+    },
+    {
+      "epoch": 4.647414854329093,
+      "grad_norm": 0.04402744024991989,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 45303
+    },
+    {
+      "epoch": 4.647722609766106,
+      "grad_norm": 0.09110675007104874,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 45306
+    },
+    {
+      "epoch": 4.648030365203119,
+      "grad_norm": 0.12931224703788757,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 45309
+    },
+    {
+      "epoch": 4.6483381206401315,
+      "grad_norm": 0.03825344517827034,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 45312
+    },
+    {
+      "epoch": 4.648645876077144,
+      "grad_norm": 0.04855644330382347,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 45315
+    },
+    {
+      "epoch": 4.648953631514157,
+      "grad_norm": 0.08731474727392197,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 45318
+    },
+    {
+      "epoch": 4.649261386951169,
+      "grad_norm": 0.06591195613145828,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 45321
+    },
+    {
+      "epoch": 4.649569142388183,
+      "grad_norm": 0.051368821412324905,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 45324
+    },
+    {
+      "epoch": 4.6498768978251945,
+      "grad_norm": 0.09959837794303894,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 45327
+    },
+    {
+      "epoch": 4.650184653262207,
+      "grad_norm": 0.08669820427894592,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 45330
+    },
+    {
+      "epoch": 4.65049240869922,
+      "grad_norm": 0.03989667445421219,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 45333
+    },
+    {
+      "epoch": 4.650800164136233,
+      "grad_norm": 0.033241745084524155,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 45336
+    },
+    {
+      "epoch": 4.651107919573246,
+      "grad_norm": 0.03395982086658478,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 45339
+    },
+    {
+      "epoch": 4.651415675010258,
+      "grad_norm": 0.038975682109594345,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 45342
+    },
+    {
+      "epoch": 4.651723430447271,
+      "grad_norm": 0.04904354363679886,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 45345
+    },
+    {
+      "epoch": 4.652031185884284,
+      "grad_norm": 0.11245663464069366,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 45348
+    },
+    {
+      "epoch": 4.652338941321297,
+      "grad_norm": 0.11784044653177261,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 45351
+    },
+    {
+      "epoch": 4.6526466967583096,
+      "grad_norm": 0.054707154631614685,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 45354
+    },
+    {
+      "epoch": 4.652954452195322,
+      "grad_norm": 0.07143942266702652,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 45357
+    },
+    {
+      "epoch": 4.653262207632335,
+      "grad_norm": 0.05629519745707512,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 45360
+    },
+    {
+      "epoch": 4.653569963069348,
+      "grad_norm": 0.03609561547636986,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 45363
+    },
+    {
+      "epoch": 4.65387771850636,
+      "grad_norm": 0.049254827201366425,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 45366
+    },
+    {
+      "epoch": 4.6541854739433735,
+      "grad_norm": 0.049836140125989914,
+      "learning_rate": 0.01,
+      "loss": 2.0302,
+      "step": 45369
+    },
+    {
+      "epoch": 4.654493229380385,
+      "grad_norm": 0.043429117649793625,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 45372
+    },
+    {
+      "epoch": 4.654800984817398,
+      "grad_norm": 0.09842117130756378,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 45375
+    },
+    {
+      "epoch": 4.655108740254411,
+      "grad_norm": 0.06700047105550766,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 45378
+    },
+    {
+      "epoch": 4.655416495691424,
+      "grad_norm": 0.0717921331524849,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 45381
+    },
+    {
+      "epoch": 4.6557242511284365,
+      "grad_norm": 0.11065293103456497,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 45384
+    },
+    {
+      "epoch": 4.656032006565449,
+      "grad_norm": 0.07875839620828629,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 45387
+    },
+    {
+      "epoch": 4.656339762002462,
+      "grad_norm": 0.07618141919374466,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 45390
+    },
+    {
+      "epoch": 4.656647517439475,
+      "grad_norm": 0.05607747659087181,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 45393
+    },
+    {
+      "epoch": 4.656955272876488,
+      "grad_norm": 0.08270974457263947,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 45396
+    },
+    {
+      "epoch": 4.6572630283135,
+      "grad_norm": 0.10751419514417648,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 45399
+    },
+    {
+      "epoch": 4.657570783750513,
+      "grad_norm": 0.07406135648488998,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 45402
+    },
+    {
+      "epoch": 4.657878539187526,
+      "grad_norm": 0.06641979515552521,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 45405
+    },
+    {
+      "epoch": 4.658186294624539,
+      "grad_norm": 0.04510427638888359,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 45408
+    },
+    {
+      "epoch": 4.658494050061551,
+      "grad_norm": 0.0381624698638916,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 45411
+    },
+    {
+      "epoch": 4.658801805498564,
+      "grad_norm": 0.08912669867277145,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 45414
+    },
+    {
+      "epoch": 4.659109560935576,
+      "grad_norm": 0.04501614347100258,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 45417
+    },
+    {
+      "epoch": 4.659417316372589,
+      "grad_norm": 0.10741306841373444,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 45420
+    },
+    {
+      "epoch": 4.659725071809602,
+      "grad_norm": 0.05976792797446251,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 45423
+    },
+    {
+      "epoch": 4.6600328272466145,
+      "grad_norm": 0.09859906136989594,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 45426
+    },
+    {
+      "epoch": 4.660340582683627,
+      "grad_norm": 0.050599753856658936,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 45429
+    },
+    {
+      "epoch": 4.66064833812064,
+      "grad_norm": 0.04848974943161011,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 45432
+    },
+    {
+      "epoch": 4.660956093557653,
+      "grad_norm": 0.07970554381608963,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 45435
+    },
+    {
+      "epoch": 4.661263848994666,
+      "grad_norm": 0.06040119752287865,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 45438
+    },
+    {
+      "epoch": 4.661571604431678,
+      "grad_norm": 0.05686302110552788,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 45441
+    },
+    {
+      "epoch": 4.661879359868691,
+      "grad_norm": 0.0757538229227066,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 45444
+    },
+    {
+      "epoch": 4.662187115305704,
+      "grad_norm": 0.07825610786676407,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 45447
+    },
+    {
+      "epoch": 4.662494870742717,
+      "grad_norm": 0.048520684242248535,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 45450
+    },
+    {
+      "epoch": 4.6628026261797295,
+      "grad_norm": 0.046711694449186325,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 45453
+    },
+    {
+      "epoch": 4.663110381616741,
+      "grad_norm": 0.058854326605796814,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 45456
+    },
+    {
+      "epoch": 4.663418137053755,
+      "grad_norm": 0.04749085754156113,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 45459
+    },
+    {
+      "epoch": 4.663725892490767,
+      "grad_norm": 0.05710771679878235,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 45462
+    },
+    {
+      "epoch": 4.66403364792778,
+      "grad_norm": 0.10407797247171402,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 45465
+    },
+    {
+      "epoch": 4.6643414033647925,
+      "grad_norm": 0.07726982980966568,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 45468
+    },
+    {
+      "epoch": 4.664649158801805,
+      "grad_norm": 0.11478970944881439,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 45471
+    },
+    {
+      "epoch": 4.664956914238818,
+      "grad_norm": 0.1078820750117302,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 45474
+    },
+    {
+      "epoch": 4.665264669675831,
+      "grad_norm": 0.06113171949982643,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 45477
+    },
+    {
+      "epoch": 4.665572425112844,
+      "grad_norm": 0.04639894515275955,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 45480
+    },
+    {
+      "epoch": 4.665880180549856,
+      "grad_norm": 0.04308944195508957,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 45483
+    },
+    {
+      "epoch": 4.666187935986869,
+      "grad_norm": 0.036519356071949005,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 45486
+    },
+    {
+      "epoch": 4.666495691423882,
+      "grad_norm": 0.04407155141234398,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 45489
+    },
+    {
+      "epoch": 4.666803446860895,
+      "grad_norm": 0.03955180197954178,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 45492
+    },
+    {
+      "epoch": 4.6671112022979075,
+      "grad_norm": 0.06779368221759796,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 45495
+    },
+    {
+      "epoch": 4.66741895773492,
+      "grad_norm": 0.11751384288072586,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 45498
+    },
+    {
+      "epoch": 4.667726713171932,
+      "grad_norm": 0.05809053033590317,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 45501
+    },
+    {
+      "epoch": 4.668034468608946,
+      "grad_norm": 0.07685140520334244,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 45504
+    },
+    {
+      "epoch": 4.668342224045958,
+      "grad_norm": 0.055087387561798096,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 45507
+    },
+    {
+      "epoch": 4.668649979482971,
+      "grad_norm": 0.05234852805733681,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 45510
+    },
+    {
+      "epoch": 4.668957734919983,
+      "grad_norm": 0.048037078231573105,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 45513
+    },
+    {
+      "epoch": 4.669265490356996,
+      "grad_norm": 0.03744732588529587,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 45516
+    },
+    {
+      "epoch": 4.669573245794009,
+      "grad_norm": 0.053839731961488724,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 45519
+    },
+    {
+      "epoch": 4.669881001231022,
+      "grad_norm": 0.06782069802284241,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 45522
+    },
+    {
+      "epoch": 4.6701887566680345,
+      "grad_norm": 0.05053957551717758,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 45525
+    },
+    {
+      "epoch": 4.670496512105047,
+      "grad_norm": 0.07486964762210846,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 45528
+    },
+    {
+      "epoch": 4.67080426754206,
+      "grad_norm": 0.07367604970932007,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 45531
+    },
+    {
+      "epoch": 4.671112022979073,
+      "grad_norm": 0.06390465050935745,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 45534
+    },
+    {
+      "epoch": 4.671419778416086,
+      "grad_norm": 0.1271253079175949,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 45537
+    },
+    {
+      "epoch": 4.671727533853098,
+      "grad_norm": 0.08515705168247223,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 45540
+    },
+    {
+      "epoch": 4.672035289290111,
+      "grad_norm": 0.03478894755244255,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 45543
+    },
+    {
+      "epoch": 4.672343044727123,
+      "grad_norm": 0.051871661096811295,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 45546
+    },
+    {
+      "epoch": 4.672650800164137,
+      "grad_norm": 0.0924333781003952,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 45549
+    },
+    {
+      "epoch": 4.672958555601149,
+      "grad_norm": 0.13955476880073547,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 45552
+    },
+    {
+      "epoch": 4.673266311038161,
+      "grad_norm": 0.07893053442239761,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 45555
+    },
+    {
+      "epoch": 4.673574066475174,
+      "grad_norm": 0.048040106892585754,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 45558
+    },
+    {
+      "epoch": 4.673881821912187,
+      "grad_norm": 0.05462590605020523,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 45561
+    },
+    {
+      "epoch": 4.6741895773492,
+      "grad_norm": 0.07224661111831665,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 45564
+    },
+    {
+      "epoch": 4.6744973327862125,
+      "grad_norm": 0.09311232715845108,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 45567
+    },
+    {
+      "epoch": 4.674805088223225,
+      "grad_norm": 0.051444876939058304,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 45570
+    },
+    {
+      "epoch": 4.675112843660238,
+      "grad_norm": 0.04749058559536934,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 45573
+    },
+    {
+      "epoch": 4.675420599097251,
+      "grad_norm": 0.04723438620567322,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 45576
+    },
+    {
+      "epoch": 4.675728354534264,
+      "grad_norm": 0.10198043286800385,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 45579
+    },
+    {
+      "epoch": 4.676036109971276,
+      "grad_norm": 0.07877392321825027,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 45582
+    },
+    {
+      "epoch": 4.676343865408289,
+      "grad_norm": 0.10339850187301636,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 45585
+    },
+    {
+      "epoch": 4.676651620845302,
+      "grad_norm": 0.07544226199388504,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 45588
+    },
+    {
+      "epoch": 4.676959376282314,
+      "grad_norm": 0.06043079495429993,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 45591
+    },
+    {
+      "epoch": 4.6772671317193275,
+      "grad_norm": 0.06057659909129143,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 45594
+    },
+    {
+      "epoch": 4.677574887156339,
+      "grad_norm": 0.08405612409114838,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 45597
+    },
+    {
+      "epoch": 4.677882642593352,
+      "grad_norm": 0.048926811665296555,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 45600
+    },
+    {
+      "epoch": 4.678190398030365,
+      "grad_norm": 0.05887090042233467,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 45603
+    },
+    {
+      "epoch": 4.678498153467378,
+      "grad_norm": 0.04116439446806908,
+      "learning_rate": 0.01,
+      "loss": 2.0309,
+      "step": 45606
+    },
+    {
+      "epoch": 4.6788059089043905,
+      "grad_norm": 0.04675902798771858,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 45609
+    },
+    {
+      "epoch": 4.679113664341403,
+      "grad_norm": 0.036510296165943146,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 45612
+    },
+    {
+      "epoch": 4.679421419778416,
+      "grad_norm": 0.06395171582698822,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 45615
+    },
+    {
+      "epoch": 4.679729175215429,
+      "grad_norm": 0.09183214604854584,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 45618
+    },
+    {
+      "epoch": 4.680036930652442,
+      "grad_norm": 0.03406717628240585,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 45621
+    },
+    {
+      "epoch": 4.680344686089454,
+      "grad_norm": 0.11709492653608322,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 45624
+    },
+    {
+      "epoch": 4.680652441526467,
+      "grad_norm": 0.054468922317028046,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 45627
+    },
+    {
+      "epoch": 4.68096019696348,
+      "grad_norm": 0.07991788536310196,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 45630
+    },
+    {
+      "epoch": 4.681267952400493,
+      "grad_norm": 0.046350929886102676,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 45633
+    },
+    {
+      "epoch": 4.681575707837505,
+      "grad_norm": 0.09740296006202698,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 45636
+    },
+    {
+      "epoch": 4.681883463274518,
+      "grad_norm": 0.0619819350540638,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 45639
+    },
+    {
+      "epoch": 4.68219121871153,
+      "grad_norm": 0.04457273334264755,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 45642
+    },
+    {
+      "epoch": 4.682498974148543,
+      "grad_norm": 0.07806507498025894,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 45645
+    },
+    {
+      "epoch": 4.682806729585556,
+      "grad_norm": 0.07045695930719376,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 45648
+    },
+    {
+      "epoch": 4.6831144850225686,
+      "grad_norm": 0.0896361917257309,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 45651
+    },
+    {
+      "epoch": 4.683422240459581,
+      "grad_norm": 0.04496191442012787,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 45654
+    },
+    {
+      "epoch": 4.683729995896594,
+      "grad_norm": 0.041538260877132416,
+      "learning_rate": 0.01,
+      "loss": 2.0357,
+      "step": 45657
+    },
+    {
+      "epoch": 4.684037751333607,
+      "grad_norm": 0.07717595249414444,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 45660
+    },
+    {
+      "epoch": 4.68434550677062,
+      "grad_norm": 0.046603985130786896,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 45663
+    },
+    {
+      "epoch": 4.6846532622076325,
+      "grad_norm": 0.0826171487569809,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 45666
+    },
+    {
+      "epoch": 4.684961017644645,
+      "grad_norm": 0.09356506168842316,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 45669
+    },
+    {
+      "epoch": 4.685268773081658,
+      "grad_norm": 0.04961970075964928,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 45672
+    },
+    {
+      "epoch": 4.685576528518671,
+      "grad_norm": 0.08724430203437805,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 45675
+    },
+    {
+      "epoch": 4.685884283955684,
+      "grad_norm": 0.093470498919487,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 45678
+    },
+    {
+      "epoch": 4.6861920393926955,
+      "grad_norm": 0.04148003086447716,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 45681
+    },
+    {
+      "epoch": 4.686499794829709,
+      "grad_norm": 0.04073407128453255,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 45684
+    },
+    {
+      "epoch": 4.686807550266721,
+      "grad_norm": 0.06185337156057358,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 45687
+    },
+    {
+      "epoch": 4.687115305703734,
+      "grad_norm": 0.06323441118001938,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 45690
+    },
+    {
+      "epoch": 4.687423061140747,
+      "grad_norm": 0.06550868600606918,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 45693
+    },
+    {
+      "epoch": 4.687730816577759,
+      "grad_norm": 0.10702592879533768,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 45696
+    },
+    {
+      "epoch": 4.688038572014772,
+      "grad_norm": 0.07939399778842926,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 45699
+    },
+    {
+      "epoch": 4.688346327451785,
+      "grad_norm": 0.09626048058271408,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 45702
+    },
+    {
+      "epoch": 4.688654082888798,
+      "grad_norm": 0.10478439927101135,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 45705
+    },
+    {
+      "epoch": 4.6889618383258105,
+      "grad_norm": 0.06496943533420563,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 45708
+    },
+    {
+      "epoch": 4.689269593762823,
+      "grad_norm": 0.060445625334978104,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 45711
+    },
+    {
+      "epoch": 4.689577349199836,
+      "grad_norm": 0.050236016511917114,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 45714
+    },
+    {
+      "epoch": 4.689885104636849,
+      "grad_norm": 0.07051600515842438,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 45717
+    },
+    {
+      "epoch": 4.690192860073862,
+      "grad_norm": 0.14171123504638672,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 45720
+    },
+    {
+      "epoch": 4.690500615510874,
+      "grad_norm": 0.04662410169839859,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 45723
+    },
+    {
+      "epoch": 4.690808370947886,
+      "grad_norm": 0.04155685007572174,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 45726
+    },
+    {
+      "epoch": 4.6911161263849,
+      "grad_norm": 0.03519884869456291,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 45729
+    },
+    {
+      "epoch": 4.691423881821912,
+      "grad_norm": 0.04860999807715416,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 45732
+    },
+    {
+      "epoch": 4.691731637258925,
+      "grad_norm": 0.05452680215239525,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 45735
+    },
+    {
+      "epoch": 4.692039392695937,
+      "grad_norm": 0.08102243393659592,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 45738
+    },
+    {
+      "epoch": 4.69234714813295,
+      "grad_norm": 0.04494756832718849,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 45741
+    },
+    {
+      "epoch": 4.692654903569963,
+      "grad_norm": 0.05493905395269394,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 45744
+    },
+    {
+      "epoch": 4.692962659006976,
+      "grad_norm": 0.037139080464839935,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 45747
+    },
+    {
+      "epoch": 4.6932704144439885,
+      "grad_norm": 0.04770129173994064,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 45750
+    },
+    {
+      "epoch": 4.693578169881001,
+      "grad_norm": 0.1386919915676117,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 45753
+    },
+    {
+      "epoch": 4.693885925318014,
+      "grad_norm": 0.038857076317071915,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 45756
+    },
+    {
+      "epoch": 4.694193680755027,
+      "grad_norm": 0.08089780062437057,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 45759
+    },
+    {
+      "epoch": 4.69450143619204,
+      "grad_norm": 0.05583566427230835,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 45762
+    },
+    {
+      "epoch": 4.694809191629052,
+      "grad_norm": 0.06422879546880722,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 45765
+    },
+    {
+      "epoch": 4.695116947066065,
+      "grad_norm": 0.037838250398635864,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 45768
+    },
+    {
+      "epoch": 4.695424702503077,
+      "grad_norm": 0.04485329985618591,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 45771
+    },
+    {
+      "epoch": 4.695732457940091,
+      "grad_norm": 0.13175508379936218,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 45774
+    },
+    {
+      "epoch": 4.696040213377103,
+      "grad_norm": 0.04778193309903145,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 45777
+    },
+    {
+      "epoch": 4.696347968814115,
+      "grad_norm": 0.07214421778917313,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 45780
+    },
+    {
+      "epoch": 4.696655724251128,
+      "grad_norm": 0.04276994615793228,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 45783
+    },
+    {
+      "epoch": 4.696963479688141,
+      "grad_norm": 0.049977049231529236,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 45786
+    },
+    {
+      "epoch": 4.697271235125154,
+      "grad_norm": 0.031859464943408966,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 45789
+    },
+    {
+      "epoch": 4.6975789905621665,
+      "grad_norm": 0.05537641420960426,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 45792
+    },
+    {
+      "epoch": 4.697886745999179,
+      "grad_norm": 0.07403066009283066,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 45795
+    },
+    {
+      "epoch": 4.698194501436192,
+      "grad_norm": 0.037336938083171844,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 45798
+    },
+    {
+      "epoch": 4.698502256873205,
+      "grad_norm": 0.04439583793282509,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 45801
+    },
+    {
+      "epoch": 4.698810012310218,
+      "grad_norm": 0.05648301541805267,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 45804
+    },
+    {
+      "epoch": 4.6991177677472304,
+      "grad_norm": 0.07558471709489822,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 45807
+    },
+    {
+      "epoch": 4.699425523184243,
+      "grad_norm": 0.04831695184111595,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 45810
+    },
+    {
+      "epoch": 4.699733278621256,
+      "grad_norm": 0.05198276415467262,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 45813
+    },
+    {
+      "epoch": 4.700041034058268,
+      "grad_norm": 0.05408303439617157,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 45816
+    },
+    {
+      "epoch": 4.700348789495282,
+      "grad_norm": 0.107550710439682,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 45819
+    },
+    {
+      "epoch": 4.7006565449322935,
+      "grad_norm": 0.05953781679272652,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 45822
+    },
+    {
+      "epoch": 4.700964300369306,
+      "grad_norm": 0.09626014530658722,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 45825
+    },
+    {
+      "epoch": 4.701272055806319,
+      "grad_norm": 0.08343878388404846,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 45828
+    },
+    {
+      "epoch": 4.701579811243332,
+      "grad_norm": 0.037152498960494995,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 45831
+    },
+    {
+      "epoch": 4.701887566680345,
+      "grad_norm": 0.10978374630212784,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 45834
+    },
+    {
+      "epoch": 4.702195322117357,
+      "grad_norm": 0.04754413291811943,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 45837
+    },
+    {
+      "epoch": 4.70250307755437,
+      "grad_norm": 0.04514869302511215,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 45840
+    },
+    {
+      "epoch": 4.702810832991383,
+      "grad_norm": 0.08105579018592834,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 45843
+    },
+    {
+      "epoch": 4.703118588428396,
+      "grad_norm": 0.032461777329444885,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 45846
+    },
+    {
+      "epoch": 4.7034263438654085,
+      "grad_norm": 0.13264963030815125,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 45849
+    },
+    {
+      "epoch": 4.703734099302421,
+      "grad_norm": 0.06816978752613068,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 45852
+    },
+    {
+      "epoch": 4.704041854739434,
+      "grad_norm": 0.09646057337522507,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 45855
+    },
+    {
+      "epoch": 4.704349610176447,
+      "grad_norm": 0.06418583542108536,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 45858
+    },
+    {
+      "epoch": 4.704657365613459,
+      "grad_norm": 0.03923073783516884,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 45861
+    },
+    {
+      "epoch": 4.704965121050472,
+      "grad_norm": 0.03935366868972778,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 45864
+    },
+    {
+      "epoch": 4.705272876487484,
+      "grad_norm": 0.0409211590886116,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 45867
+    },
+    {
+      "epoch": 4.705580631924497,
+      "grad_norm": 0.02996712550520897,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 45870
+    },
+    {
+      "epoch": 4.70588838736151,
+      "grad_norm": 0.03149070590734482,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 45873
+    },
+    {
+      "epoch": 4.706196142798523,
+      "grad_norm": 0.11619815230369568,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 45876
+    },
+    {
+      "epoch": 4.706503898235535,
+      "grad_norm": 0.08959172666072845,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 45879
+    },
+    {
+      "epoch": 4.706811653672548,
+      "grad_norm": 0.03898904472589493,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 45882
+    },
+    {
+      "epoch": 4.707119409109561,
+      "grad_norm": 0.05690839886665344,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 45885
+    },
+    {
+      "epoch": 4.707427164546574,
+      "grad_norm": 0.08014564961194992,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 45888
+    },
+    {
+      "epoch": 4.7077349199835865,
+      "grad_norm": 0.07317539304494858,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 45891
+    },
+    {
+      "epoch": 4.708042675420599,
+      "grad_norm": 0.040355369448661804,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 45894
+    },
+    {
+      "epoch": 4.708350430857612,
+      "grad_norm": 0.04205753654241562,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 45897
+    },
+    {
+      "epoch": 4.708658186294625,
+      "grad_norm": 0.07082203030586243,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 45900
+    },
+    {
+      "epoch": 4.708965941731638,
+      "grad_norm": 0.037901826202869415,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 45903
+    },
+    {
+      "epoch": 4.7092736971686495,
+      "grad_norm": 0.04728815332055092,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 45906
+    },
+    {
+      "epoch": 4.709581452605663,
+      "grad_norm": 0.048279549926519394,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 45909
+    },
+    {
+      "epoch": 4.709889208042675,
+      "grad_norm": 0.06365431100130081,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 45912
+    },
+    {
+      "epoch": 4.710196963479688,
+      "grad_norm": 0.07750722020864487,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 45915
+    },
+    {
+      "epoch": 4.710504718916701,
+      "grad_norm": 0.07337852567434311,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 45918
+    },
+    {
+      "epoch": 4.710812474353713,
+      "grad_norm": 0.055369194597005844,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 45921
+    },
+    {
+      "epoch": 4.711120229790726,
+      "grad_norm": 0.07653316110372543,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 45924
+    },
+    {
+      "epoch": 4.711427985227739,
+      "grad_norm": 0.13082928955554962,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 45927
+    },
+    {
+      "epoch": 4.711735740664752,
+      "grad_norm": 0.07909176498651505,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 45930
+    },
+    {
+      "epoch": 4.7120434961017645,
+      "grad_norm": 0.055445730686187744,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 45933
+    },
+    {
+      "epoch": 4.712351251538777,
+      "grad_norm": 0.08689797669649124,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 45936
+    },
+    {
+      "epoch": 4.71265900697579,
+      "grad_norm": 0.051975540816783905,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 45939
+    },
+    {
+      "epoch": 4.712966762412803,
+      "grad_norm": 0.1104854941368103,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 45942
+    },
+    {
+      "epoch": 4.713274517849816,
+      "grad_norm": 0.05927729606628418,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 45945
+    },
+    {
+      "epoch": 4.713582273286828,
+      "grad_norm": 0.051379501819610596,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 45948
+    },
+    {
+      "epoch": 4.71389002872384,
+      "grad_norm": 0.03545750305056572,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 45951
+    },
+    {
+      "epoch": 4.714197784160854,
+      "grad_norm": 0.044772762805223465,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 45954
+    },
+    {
+      "epoch": 4.714505539597866,
+      "grad_norm": 0.04172508046030998,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 45957
+    },
+    {
+      "epoch": 4.714813295034879,
+      "grad_norm": 0.04818372428417206,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 45960
+    },
+    {
+      "epoch": 4.7151210504718915,
+      "grad_norm": 0.10572110861539841,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 45963
+    },
+    {
+      "epoch": 4.715428805908904,
+      "grad_norm": 0.04968751594424248,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 45966
+    },
+    {
+      "epoch": 4.715736561345917,
+      "grad_norm": 0.08244633674621582,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 45969
+    },
+    {
+      "epoch": 4.71604431678293,
+      "grad_norm": 0.0786314308643341,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 45972
+    },
+    {
+      "epoch": 4.716352072219943,
+      "grad_norm": 0.05768943205475807,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 45975
+    },
+    {
+      "epoch": 4.716659827656955,
+      "grad_norm": 0.04455199092626572,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 45978
+    },
+    {
+      "epoch": 4.716967583093968,
+      "grad_norm": 0.05535956844687462,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 45981
+    },
+    {
+      "epoch": 4.717275338530981,
+      "grad_norm": 0.03555374965071678,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 45984
+    },
+    {
+      "epoch": 4.717583093967994,
+      "grad_norm": 0.06458409130573273,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 45987
+    },
+    {
+      "epoch": 4.7178908494050065,
+      "grad_norm": 0.05734160915017128,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 45990
+    },
+    {
+      "epoch": 4.718198604842019,
+      "grad_norm": 0.06240745261311531,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 45993
+    },
+    {
+      "epoch": 4.718506360279031,
+      "grad_norm": 0.08987503498792648,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 45996
+    },
+    {
+      "epoch": 4.718814115716045,
+      "grad_norm": 0.06078198924660683,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 45999
+    },
+    {
+      "epoch": 4.719121871153057,
+      "grad_norm": 0.12024971842765808,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 46002
+    },
+    {
+      "epoch": 4.7194296265900695,
+      "grad_norm": 0.06620301306247711,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 46005
+    },
+    {
+      "epoch": 4.719737382027082,
+      "grad_norm": 0.0450078584253788,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 46008
+    },
+    {
+      "epoch": 4.720045137464095,
+      "grad_norm": 0.04509684816002846,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 46011
+    },
+    {
+      "epoch": 4.720352892901108,
+      "grad_norm": 0.06029663607478142,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 46014
+    },
+    {
+      "epoch": 4.720660648338121,
+      "grad_norm": 0.06620631366968155,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 46017
+    },
+    {
+      "epoch": 4.720968403775133,
+      "grad_norm": 0.09920935332775116,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 46020
+    },
+    {
+      "epoch": 4.721276159212146,
+      "grad_norm": 0.04000372067093849,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 46023
+    },
+    {
+      "epoch": 4.721583914649159,
+      "grad_norm": 0.10262927412986755,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 46026
+    },
+    {
+      "epoch": 4.721891670086172,
+      "grad_norm": 0.07029026746749878,
+      "learning_rate": 0.01,
+      "loss": 2.0492,
+      "step": 46029
+    },
+    {
+      "epoch": 4.7221994255231845,
+      "grad_norm": 0.03733726218342781,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 46032
+    },
+    {
+      "epoch": 4.722507180960197,
+      "grad_norm": 0.16273632645606995,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 46035
+    },
+    {
+      "epoch": 4.72281493639721,
+      "grad_norm": 0.052774641662836075,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 46038
+    },
+    {
+      "epoch": 4.723122691834222,
+      "grad_norm": 0.04785846173763275,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 46041
+    },
+    {
+      "epoch": 4.723430447271236,
+      "grad_norm": 0.046762675046920776,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 46044
+    },
+    {
+      "epoch": 4.7237382027082475,
+      "grad_norm": 0.055082205682992935,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 46047
+    },
+    {
+      "epoch": 4.72404595814526,
+      "grad_norm": 0.12330979853868484,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 46050
+    },
+    {
+      "epoch": 4.724353713582273,
+      "grad_norm": 0.0426286906003952,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 46053
+    },
+    {
+      "epoch": 4.724661469019286,
+      "grad_norm": 0.09888054430484772,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 46056
+    },
+    {
+      "epoch": 4.724969224456299,
+      "grad_norm": 0.09335997700691223,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 46059
+    },
+    {
+      "epoch": 4.725276979893311,
+      "grad_norm": 0.06959028542041779,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 46062
+    },
+    {
+      "epoch": 4.725584735330324,
+      "grad_norm": 0.06538756936788559,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 46065
+    },
+    {
+      "epoch": 4.725892490767337,
+      "grad_norm": 0.0652519017457962,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 46068
+    },
+    {
+      "epoch": 4.72620024620435,
+      "grad_norm": 0.049920666962862015,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 46071
+    },
+    {
+      "epoch": 4.7265080016413625,
+      "grad_norm": 0.0796656534075737,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 46074
+    },
+    {
+      "epoch": 4.726815757078375,
+      "grad_norm": 0.08785120397806168,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 46077
+    },
+    {
+      "epoch": 4.727123512515388,
+      "grad_norm": 0.05050429701805115,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 46080
+    },
+    {
+      "epoch": 4.727431267952401,
+      "grad_norm": 0.039991263300180435,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 46083
+    },
+    {
+      "epoch": 4.727739023389413,
+      "grad_norm": 0.04895581305027008,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 46086
+    },
+    {
+      "epoch": 4.728046778826426,
+      "grad_norm": 0.13111530244350433,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 46089
+    },
+    {
+      "epoch": 4.728354534263438,
+      "grad_norm": 0.07493706792593002,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 46092
+    },
+    {
+      "epoch": 4.728662289700451,
+      "grad_norm": 0.06003058701753616,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 46095
+    },
+    {
+      "epoch": 4.728970045137464,
+      "grad_norm": 0.05581682547926903,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 46098
+    },
+    {
+      "epoch": 4.729277800574477,
+      "grad_norm": 0.0741606056690216,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 46101
+    },
+    {
+      "epoch": 4.7295855560114894,
+      "grad_norm": 0.05753305181860924,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 46104
+    },
+    {
+      "epoch": 4.729893311448502,
+      "grad_norm": 0.06790949404239655,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 46107
+    },
+    {
+      "epoch": 4.730201066885515,
+      "grad_norm": 0.05599404126405716,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 46110
+    },
+    {
+      "epoch": 4.730508822322528,
+      "grad_norm": 0.06556524336338043,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 46113
+    },
+    {
+      "epoch": 4.730816577759541,
+      "grad_norm": 0.043739136308431625,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 46116
+    },
+    {
+      "epoch": 4.731124333196553,
+      "grad_norm": 0.05598102882504463,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 46119
+    },
+    {
+      "epoch": 4.731432088633566,
+      "grad_norm": 0.0398273728787899,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 46122
+    },
+    {
+      "epoch": 4.731739844070579,
+      "grad_norm": 0.06281284987926483,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 46125
+    },
+    {
+      "epoch": 4.732047599507592,
+      "grad_norm": 0.0575578436255455,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 46128
+    },
+    {
+      "epoch": 4.732355354944604,
+      "grad_norm": 0.044283825904130936,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 46131
+    },
+    {
+      "epoch": 4.732663110381616,
+      "grad_norm": 0.11150834709405899,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 46134
+    },
+    {
+      "epoch": 4.732970865818629,
+      "grad_norm": 0.11505670100450516,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 46137
+    },
+    {
+      "epoch": 4.733278621255642,
+      "grad_norm": 0.10406883805990219,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 46140
+    },
+    {
+      "epoch": 4.733586376692655,
+      "grad_norm": 0.07568677514791489,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 46143
+    },
+    {
+      "epoch": 4.7338941321296675,
+      "grad_norm": 0.05774052441120148,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 46146
+    },
+    {
+      "epoch": 4.73420188756668,
+      "grad_norm": 0.05138164013624191,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 46149
+    },
+    {
+      "epoch": 4.734509643003693,
+      "grad_norm": 0.09138276427984238,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 46152
+    },
+    {
+      "epoch": 4.734817398440706,
+      "grad_norm": 0.04819070175290108,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 46155
+    },
+    {
+      "epoch": 4.735125153877719,
+      "grad_norm": 0.06231243535876274,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 46158
+    },
+    {
+      "epoch": 4.735432909314731,
+      "grad_norm": 0.05212223157286644,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 46161
+    },
+    {
+      "epoch": 4.735740664751744,
+      "grad_norm": 0.05681807920336723,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 46164
+    },
+    {
+      "epoch": 4.736048420188757,
+      "grad_norm": 0.039965178817510605,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 46167
+    },
+    {
+      "epoch": 4.73635617562577,
+      "grad_norm": 0.0815226286649704,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 46170
+    },
+    {
+      "epoch": 4.7366639310627825,
+      "grad_norm": 0.121616430580616,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 46173
+    },
+    {
+      "epoch": 4.736971686499794,
+      "grad_norm": 0.06908087432384491,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 46176
+    },
+    {
+      "epoch": 4.737279441936807,
+      "grad_norm": 0.043230995535850525,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 46179
+    },
+    {
+      "epoch": 4.73758719737382,
+      "grad_norm": 0.10230391472578049,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 46182
+    },
+    {
+      "epoch": 4.737894952810833,
+      "grad_norm": 0.06521748751401901,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 46185
+    },
+    {
+      "epoch": 4.7382027082478455,
+      "grad_norm": 0.052706390619277954,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 46188
+    },
+    {
+      "epoch": 4.738510463684858,
+      "grad_norm": 0.0332234725356102,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 46191
+    },
+    {
+      "epoch": 4.738818219121871,
+      "grad_norm": 0.061153993010520935,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 46194
+    },
+    {
+      "epoch": 4.739125974558884,
+      "grad_norm": 0.12053243815898895,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 46197
+    },
+    {
+      "epoch": 4.739433729995897,
+      "grad_norm": 0.09798255562782288,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 46200
+    },
+    {
+      "epoch": 4.739741485432909,
+      "grad_norm": 0.05450673773884773,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 46203
+    },
+    {
+      "epoch": 4.740049240869922,
+      "grad_norm": 0.03913121297955513,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 46206
+    },
+    {
+      "epoch": 4.740356996306935,
+      "grad_norm": 0.036051902920007706,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 46209
+    },
+    {
+      "epoch": 4.740664751743948,
+      "grad_norm": 0.04099172353744507,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 46212
+    },
+    {
+      "epoch": 4.7409725071809605,
+      "grad_norm": 0.03864535689353943,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 46215
+    },
+    {
+      "epoch": 4.741280262617973,
+      "grad_norm": 0.055634405463933945,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 46218
+    },
+    {
+      "epoch": 4.741588018054985,
+      "grad_norm": 0.06536100059747696,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 46221
+    },
+    {
+      "epoch": 4.741895773491998,
+      "grad_norm": 0.05309274420142174,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 46224
+    },
+    {
+      "epoch": 4.742203528929011,
+      "grad_norm": 0.09953348338603973,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 46227
+    },
+    {
+      "epoch": 4.7425112843660235,
+      "grad_norm": 0.04084203764796257,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 46230
+    },
+    {
+      "epoch": 4.742819039803036,
+      "grad_norm": 0.12028990685939789,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 46233
+    },
+    {
+      "epoch": 4.743126795240049,
+      "grad_norm": 0.04335862770676613,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 46236
+    },
+    {
+      "epoch": 4.743434550677062,
+      "grad_norm": 0.06697961688041687,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 46239
+    },
+    {
+      "epoch": 4.743742306114075,
+      "grad_norm": 0.047941289842128754,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 46242
+    },
+    {
+      "epoch": 4.744050061551087,
+      "grad_norm": 0.03778136894106865,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 46245
+    },
+    {
+      "epoch": 4.7443578169881,
+      "grad_norm": 0.04735583811998367,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 46248
+    },
+    {
+      "epoch": 4.744665572425113,
+      "grad_norm": 0.04471844062209129,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 46251
+    },
+    {
+      "epoch": 4.744973327862126,
+      "grad_norm": 0.03621865063905716,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 46254
+    },
+    {
+      "epoch": 4.745281083299139,
+      "grad_norm": 0.12407182902097702,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 46257
+    },
+    {
+      "epoch": 4.745588838736151,
+      "grad_norm": 0.05654411017894745,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 46260
+    },
+    {
+      "epoch": 4.745896594173164,
+      "grad_norm": 0.08955933898687363,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 46263
+    },
+    {
+      "epoch": 4.746204349610176,
+      "grad_norm": 0.08366284519433975,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 46266
+    },
+    {
+      "epoch": 4.746512105047189,
+      "grad_norm": 0.056272462010383606,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 46269
+    },
+    {
+      "epoch": 4.746819860484202,
+      "grad_norm": 0.05544830113649368,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 46272
+    },
+    {
+      "epoch": 4.747127615921214,
+      "grad_norm": 0.04018571227788925,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 46275
+    },
+    {
+      "epoch": 4.747435371358227,
+      "grad_norm": 0.07131030410528183,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 46278
+    },
+    {
+      "epoch": 4.74774312679524,
+      "grad_norm": 0.03746544569730759,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 46281
+    },
+    {
+      "epoch": 4.748050882232253,
+      "grad_norm": 0.09419632703065872,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 46284
+    },
+    {
+      "epoch": 4.7483586376692655,
+      "grad_norm": 0.05771811679005623,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 46287
+    },
+    {
+      "epoch": 4.748666393106278,
+      "grad_norm": 0.0985054075717926,
+      "learning_rate": 0.01,
+      "loss": 2.0356,
+      "step": 46290
+    },
+    {
+      "epoch": 4.748974148543291,
+      "grad_norm": 0.06566286832094193,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 46293
+    },
+    {
+      "epoch": 4.749281903980304,
+      "grad_norm": 0.043330125510692596,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 46296
+    },
+    {
+      "epoch": 4.749589659417317,
+      "grad_norm": 0.05344987288117409,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 46299
+    },
+    {
+      "epoch": 4.749897414854329,
+      "grad_norm": 0.032919514924287796,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 46302
+    },
+    {
+      "epoch": 4.750205170291342,
+      "grad_norm": 0.08051367104053497,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 46305
+    },
+    {
+      "epoch": 4.750512925728355,
+      "grad_norm": 0.09481201320886612,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 46308
+    },
+    {
+      "epoch": 4.750820681165367,
+      "grad_norm": 0.1356784701347351,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 46311
+    },
+    {
+      "epoch": 4.75112843660238,
+      "grad_norm": 0.12291105091571808,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 46314
+    },
+    {
+      "epoch": 4.751436192039392,
+      "grad_norm": 0.050581369549036026,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 46317
+    },
+    {
+      "epoch": 4.751743947476405,
+      "grad_norm": 0.08423294126987457,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 46320
+    },
+    {
+      "epoch": 4.752051702913418,
+      "grad_norm": 0.0642426609992981,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 46323
+    },
+    {
+      "epoch": 4.752359458350431,
+      "grad_norm": 0.042584337294101715,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 46326
+    },
+    {
+      "epoch": 4.7526672137874435,
+      "grad_norm": 0.04110237583518028,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 46329
+    },
+    {
+      "epoch": 4.752974969224456,
+      "grad_norm": 0.034602370113134384,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 46332
+    },
+    {
+      "epoch": 4.753282724661469,
+      "grad_norm": 0.05274072289466858,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 46335
+    },
+    {
+      "epoch": 4.753590480098482,
+      "grad_norm": 0.09157788008451462,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 46338
+    },
+    {
+      "epoch": 4.753898235535495,
+      "grad_norm": 0.04491696506738663,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 46341
+    },
+    {
+      "epoch": 4.754205990972507,
+      "grad_norm": 0.044836509972810745,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 46344
+    },
+    {
+      "epoch": 4.75451374640952,
+      "grad_norm": 0.08086413145065308,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 46347
+    },
+    {
+      "epoch": 4.754821501846533,
+      "grad_norm": 0.04024931788444519,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 46350
+    },
+    {
+      "epoch": 4.755129257283546,
+      "grad_norm": 0.053472552448511124,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 46353
+    },
+    {
+      "epoch": 4.755437012720558,
+      "grad_norm": 0.06579295545816422,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 46356
+    },
+    {
+      "epoch": 4.75574476815757,
+      "grad_norm": 0.07256436347961426,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 46359
+    },
+    {
+      "epoch": 4.756052523594583,
+      "grad_norm": 0.1356937736272812,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 46362
+    },
+    {
+      "epoch": 4.756360279031596,
+      "grad_norm": 0.06238779425621033,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 46365
+    },
+    {
+      "epoch": 4.756668034468609,
+      "grad_norm": 0.03617888316512108,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 46368
+    },
+    {
+      "epoch": 4.7569757899056215,
+      "grad_norm": 0.039169710129499435,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 46371
+    },
+    {
+      "epoch": 4.757283545342634,
+      "grad_norm": 0.03998740762472153,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 46374
+    },
+    {
+      "epoch": 4.757591300779647,
+      "grad_norm": 0.09922688454389572,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 46377
+    },
+    {
+      "epoch": 4.75789905621666,
+      "grad_norm": 0.07870779931545258,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 46380
+    },
+    {
+      "epoch": 4.758206811653673,
+      "grad_norm": 0.09917730838060379,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 46383
+    },
+    {
+      "epoch": 4.758514567090685,
+      "grad_norm": 0.14981189370155334,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 46386
+    },
+    {
+      "epoch": 4.758822322527698,
+      "grad_norm": 0.08356380462646484,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 46389
+    },
+    {
+      "epoch": 4.759130077964711,
+      "grad_norm": 0.05980211868882179,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 46392
+    },
+    {
+      "epoch": 4.759437833401724,
+      "grad_norm": 0.05891801416873932,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 46395
+    },
+    {
+      "epoch": 4.7597455888387366,
+      "grad_norm": 0.08687376976013184,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 46398
+    },
+    {
+      "epoch": 4.760053344275748,
+      "grad_norm": 0.03482063114643097,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 46401
+    },
+    {
+      "epoch": 4.760361099712761,
+      "grad_norm": 0.04206204041838646,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 46404
+    },
+    {
+      "epoch": 4.760668855149774,
+      "grad_norm": 0.051311641931533813,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 46407
+    },
+    {
+      "epoch": 4.760976610586787,
+      "grad_norm": 0.06586962938308716,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 46410
+    },
+    {
+      "epoch": 4.7612843660238,
+      "grad_norm": 0.058405712246894836,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 46413
+    },
+    {
+      "epoch": 4.761592121460812,
+      "grad_norm": 0.060229748487472534,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 46416
+    },
+    {
+      "epoch": 4.761899876897825,
+      "grad_norm": 0.08985790610313416,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 46419
+    },
+    {
+      "epoch": 4.762207632334838,
+      "grad_norm": 0.13349880278110504,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 46422
+    },
+    {
+      "epoch": 4.762515387771851,
+      "grad_norm": 0.05167052149772644,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 46425
+    },
+    {
+      "epoch": 4.7628231432088635,
+      "grad_norm": 0.03675553575158119,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 46428
+    },
+    {
+      "epoch": 4.763130898645876,
+      "grad_norm": 0.06192685663700104,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 46431
+    },
+    {
+      "epoch": 4.763438654082889,
+      "grad_norm": 0.06537027657032013,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 46434
+    },
+    {
+      "epoch": 4.763746409519902,
+      "grad_norm": 0.07077626138925552,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 46437
+    },
+    {
+      "epoch": 4.764054164956915,
+      "grad_norm": 0.0640728622674942,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 46440
+    },
+    {
+      "epoch": 4.764361920393927,
+      "grad_norm": 0.04746294394135475,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 46443
+    },
+    {
+      "epoch": 4.764669675830939,
+      "grad_norm": 0.04115734621882439,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 46446
+    },
+    {
+      "epoch": 4.764977431267952,
+      "grad_norm": 0.11647975444793701,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 46449
+    },
+    {
+      "epoch": 4.765285186704965,
+      "grad_norm": 0.0611908994615078,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 46452
+    },
+    {
+      "epoch": 4.765592942141978,
+      "grad_norm": 0.11524607241153717,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 46455
+    },
+    {
+      "epoch": 4.76590069757899,
+      "grad_norm": 0.10703855752944946,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 46458
+    },
+    {
+      "epoch": 4.766208453016003,
+      "grad_norm": 0.052681829780340195,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 46461
+    },
+    {
+      "epoch": 4.766516208453016,
+      "grad_norm": 0.04313669726252556,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 46464
+    },
+    {
+      "epoch": 4.766823963890029,
+      "grad_norm": 0.06806657463312149,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 46467
+    },
+    {
+      "epoch": 4.7671317193270415,
+      "grad_norm": 0.07249857485294342,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 46470
+    },
+    {
+      "epoch": 4.767439474764054,
+      "grad_norm": 0.07816947996616364,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 46473
+    },
+    {
+      "epoch": 4.767747230201067,
+      "grad_norm": 0.05120289698243141,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 46476
+    },
+    {
+      "epoch": 4.76805498563808,
+      "grad_norm": 0.09938623756170273,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 46479
+    },
+    {
+      "epoch": 4.768362741075093,
+      "grad_norm": 0.060353558510541916,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 46482
+    },
+    {
+      "epoch": 4.768670496512105,
+      "grad_norm": 0.10951244086027145,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 46485
+    },
+    {
+      "epoch": 4.768978251949118,
+      "grad_norm": 0.04746172949671745,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 46488
+    },
+    {
+      "epoch": 4.76928600738613,
+      "grad_norm": 0.07137439399957657,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 46491
+    },
+    {
+      "epoch": 4.769593762823143,
+      "grad_norm": 0.13574503362178802,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 46494
+    },
+    {
+      "epoch": 4.769901518260156,
+      "grad_norm": 0.047539595514535904,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 46497
+    },
+    {
+      "epoch": 4.770209273697168,
+      "grad_norm": 0.0612449012696743,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 46500
+    },
+    {
+      "epoch": 4.770517029134181,
+      "grad_norm": 0.05257963016629219,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 46503
+    },
+    {
+      "epoch": 4.770824784571194,
+      "grad_norm": 0.05487014353275299,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 46506
+    },
+    {
+      "epoch": 4.771132540008207,
+      "grad_norm": 0.03313500061631203,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 46509
+    },
+    {
+      "epoch": 4.7714402954452195,
+      "grad_norm": 0.04244723170995712,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 46512
+    },
+    {
+      "epoch": 4.771748050882232,
+      "grad_norm": 0.06064971908926964,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 46515
+    },
+    {
+      "epoch": 4.772055806319245,
+      "grad_norm": 0.04784930869936943,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 46518
+    },
+    {
+      "epoch": 4.772363561756258,
+      "grad_norm": 0.04266538843512535,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 46521
+    },
+    {
+      "epoch": 4.772671317193271,
+      "grad_norm": 0.09486623853445053,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 46524
+    },
+    {
+      "epoch": 4.772979072630283,
+      "grad_norm": 0.05113883316516876,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 46527
+    },
+    {
+      "epoch": 4.773286828067296,
+      "grad_norm": 0.11440654844045639,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 46530
+    },
+    {
+      "epoch": 4.773594583504309,
+      "grad_norm": 0.07778608053922653,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 46533
+    },
+    {
+      "epoch": 4.773902338941321,
+      "grad_norm": 0.042305488139390945,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 46536
+    },
+    {
+      "epoch": 4.774210094378334,
+      "grad_norm": 0.06056047976016998,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 46539
+    },
+    {
+      "epoch": 4.774517849815346,
+      "grad_norm": 0.03576405718922615,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 46542
+    },
+    {
+      "epoch": 4.774825605252359,
+      "grad_norm": 0.04558248445391655,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 46545
+    },
+    {
+      "epoch": 4.775133360689372,
+      "grad_norm": 0.04870672523975372,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 46548
+    },
+    {
+      "epoch": 4.775441116126385,
+      "grad_norm": 0.04591087996959686,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 46551
+    },
+    {
+      "epoch": 4.775748871563398,
+      "grad_norm": 0.08307138830423355,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 46554
+    },
+    {
+      "epoch": 4.77605662700041,
+      "grad_norm": 0.0701267346739769,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 46557
+    },
+    {
+      "epoch": 4.776364382437423,
+      "grad_norm": 0.13456737995147705,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 46560
+    },
+    {
+      "epoch": 4.776672137874436,
+      "grad_norm": 0.06605277210474014,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 46563
+    },
+    {
+      "epoch": 4.776979893311449,
+      "grad_norm": 0.05360471457242966,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 46566
+    },
+    {
+      "epoch": 4.7772876487484615,
+      "grad_norm": 0.04942501336336136,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 46569
+    },
+    {
+      "epoch": 4.777595404185474,
+      "grad_norm": 0.09876788407564163,
+      "learning_rate": 0.01,
+      "loss": 2.0284,
+      "step": 46572
+    },
+    {
+      "epoch": 4.777903159622487,
+      "grad_norm": 0.07075868546962738,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 46575
+    },
+    {
+      "epoch": 4.7782109150595,
+      "grad_norm": 0.06624365597963333,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 46578
+    },
+    {
+      "epoch": 4.778518670496512,
+      "grad_norm": 0.04903232306241989,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 46581
+    },
+    {
+      "epoch": 4.7788264259335245,
+      "grad_norm": 0.06817879527807236,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 46584
+    },
+    {
+      "epoch": 4.779134181370537,
+      "grad_norm": 0.10077866911888123,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 46587
+    },
+    {
+      "epoch": 4.77944193680755,
+      "grad_norm": 0.05322563648223877,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 46590
+    },
+    {
+      "epoch": 4.779749692244563,
+      "grad_norm": 0.09559939801692963,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 46593
+    },
+    {
+      "epoch": 4.780057447681576,
+      "grad_norm": 0.06942281872034073,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 46596
+    },
+    {
+      "epoch": 4.780365203118588,
+      "grad_norm": 0.1250401884317398,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 46599
+    },
+    {
+      "epoch": 4.780672958555601,
+      "grad_norm": 0.09563729166984558,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 46602
+    },
+    {
+      "epoch": 4.780980713992614,
+      "grad_norm": 0.08675848692655563,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 46605
+    },
+    {
+      "epoch": 4.781288469429627,
+      "grad_norm": 0.0791327953338623,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 46608
+    },
+    {
+      "epoch": 4.7815962248666395,
+      "grad_norm": 0.10748487710952759,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 46611
+    },
+    {
+      "epoch": 4.781903980303652,
+      "grad_norm": 0.056079965084791183,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 46614
+    },
+    {
+      "epoch": 4.782211735740665,
+      "grad_norm": 0.04614640399813652,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 46617
+    },
+    {
+      "epoch": 4.782519491177678,
+      "grad_norm": 0.07398012280464172,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 46620
+    },
+    {
+      "epoch": 4.782827246614691,
+      "grad_norm": 0.07945587486028671,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 46623
+    },
+    {
+      "epoch": 4.7831350020517025,
+      "grad_norm": 0.040440626442432404,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 46626
+    },
+    {
+      "epoch": 4.783442757488715,
+      "grad_norm": 0.1381184458732605,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 46629
+    },
+    {
+      "epoch": 4.783750512925728,
+      "grad_norm": 0.06996659189462662,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 46632
+    },
+    {
+      "epoch": 4.784058268362741,
+      "grad_norm": 0.03792927786707878,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 46635
+    },
+    {
+      "epoch": 4.784366023799754,
+      "grad_norm": 0.03911750763654709,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 46638
+    },
+    {
+      "epoch": 4.784673779236766,
+      "grad_norm": 0.04588170349597931,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 46641
+    },
+    {
+      "epoch": 4.784981534673779,
+      "grad_norm": 0.10618551820516586,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 46644
+    },
+    {
+      "epoch": 4.785289290110792,
+      "grad_norm": 0.07402194291353226,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 46647
+    },
+    {
+      "epoch": 4.785597045547805,
+      "grad_norm": 0.05611973628401756,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 46650
+    },
+    {
+      "epoch": 4.7859048009848175,
+      "grad_norm": 0.062334805727005005,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 46653
+    },
+    {
+      "epoch": 4.78621255642183,
+      "grad_norm": 0.05972002074122429,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 46656
+    },
+    {
+      "epoch": 4.786520311858843,
+      "grad_norm": 0.10354367643594742,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 46659
+    },
+    {
+      "epoch": 4.786828067295856,
+      "grad_norm": 0.07653992623090744,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 46662
+    },
+    {
+      "epoch": 4.787135822732869,
+      "grad_norm": 0.0790034830570221,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 46665
+    },
+    {
+      "epoch": 4.787443578169881,
+      "grad_norm": 0.11524873226881027,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 46668
+    },
+    {
+      "epoch": 4.787751333606893,
+      "grad_norm": 0.11828222125768661,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 46671
+    },
+    {
+      "epoch": 4.788059089043906,
+      "grad_norm": 0.06129033491015434,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 46674
+    },
+    {
+      "epoch": 4.788366844480919,
+      "grad_norm": 0.060850802809000015,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 46677
+    },
+    {
+      "epoch": 4.788674599917932,
+      "grad_norm": 0.0488838255405426,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 46680
+    },
+    {
+      "epoch": 4.788982355354944,
+      "grad_norm": 0.03288913145661354,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 46683
+    },
+    {
+      "epoch": 4.789290110791957,
+      "grad_norm": 0.038961250334978104,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 46686
+    },
+    {
+      "epoch": 4.78959786622897,
+      "grad_norm": 0.10202857851982117,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 46689
+    },
+    {
+      "epoch": 4.789905621665983,
+      "grad_norm": 0.05087895691394806,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 46692
+    },
+    {
+      "epoch": 4.7902133771029956,
+      "grad_norm": 0.07954477518796921,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 46695
+    },
+    {
+      "epoch": 4.790521132540008,
+      "grad_norm": 0.11283519119024277,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 46698
+    },
+    {
+      "epoch": 4.790828887977021,
+      "grad_norm": 0.04189791902899742,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 46701
+    },
+    {
+      "epoch": 4.791136643414034,
+      "grad_norm": 0.09559596329927444,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 46704
+    },
+    {
+      "epoch": 4.791444398851047,
+      "grad_norm": 0.07935875654220581,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 46707
+    },
+    {
+      "epoch": 4.7917521542880595,
+      "grad_norm": 0.09595377743244171,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 46710
+    },
+    {
+      "epoch": 4.792059909725072,
+      "grad_norm": 0.06680227816104889,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 46713
+    },
+    {
+      "epoch": 4.792367665162084,
+      "grad_norm": 0.07122410833835602,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 46716
+    },
+    {
+      "epoch": 4.792675420599097,
+      "grad_norm": 0.0937454104423523,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 46719
+    },
+    {
+      "epoch": 4.79298317603611,
+      "grad_norm": 0.08886270225048065,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 46722
+    },
+    {
+      "epoch": 4.7932909314731225,
+      "grad_norm": 0.04537273570895195,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 46725
+    },
+    {
+      "epoch": 4.793598686910135,
+      "grad_norm": 0.10327073931694031,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 46728
+    },
+    {
+      "epoch": 4.793906442347148,
+      "grad_norm": 0.08701428025960922,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 46731
+    },
+    {
+      "epoch": 4.794214197784161,
+      "grad_norm": 0.06731964647769928,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 46734
+    },
+    {
+      "epoch": 4.794521953221174,
+      "grad_norm": 0.15847548842430115,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 46737
+    },
+    {
+      "epoch": 4.794829708658186,
+      "grad_norm": 0.04891221970319748,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 46740
+    },
+    {
+      "epoch": 4.795137464095199,
+      "grad_norm": 0.040448226034641266,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 46743
+    },
+    {
+      "epoch": 4.795445219532212,
+      "grad_norm": 0.04388444870710373,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 46746
+    },
+    {
+      "epoch": 4.795752974969225,
+      "grad_norm": 0.05044842138886452,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 46749
+    },
+    {
+      "epoch": 4.7960607304062375,
+      "grad_norm": 0.17745061218738556,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 46752
+    },
+    {
+      "epoch": 4.79636848584325,
+      "grad_norm": 0.07918280363082886,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 46755
+    },
+    {
+      "epoch": 4.796676241280263,
+      "grad_norm": 0.0841284915804863,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 46758
+    },
+    {
+      "epoch": 4.796983996717275,
+      "grad_norm": 0.07425282150506973,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 46761
+    },
+    {
+      "epoch": 4.797291752154288,
+      "grad_norm": 0.11900149285793304,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 46764
+    },
+    {
+      "epoch": 4.7975995075913005,
+      "grad_norm": 0.054541297256946564,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 46767
+    },
+    {
+      "epoch": 4.797907263028313,
+      "grad_norm": 0.05434846505522728,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 46770
+    },
+    {
+      "epoch": 4.798215018465326,
+      "grad_norm": 0.045795127749443054,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 46773
+    },
+    {
+      "epoch": 4.798522773902339,
+      "grad_norm": 0.044646721333265305,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 46776
+    },
+    {
+      "epoch": 4.798830529339352,
+      "grad_norm": 0.06140962243080139,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 46779
+    },
+    {
+      "epoch": 4.799138284776364,
+      "grad_norm": 0.1806284338235855,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 46782
+    },
+    {
+      "epoch": 4.799446040213377,
+      "grad_norm": 0.1726996749639511,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 46785
+    },
+    {
+      "epoch": 4.79975379565039,
+      "grad_norm": 0.093462273478508,
+      "learning_rate": 0.01,
+      "loss": 2.0224,
+      "step": 46788
+    },
+    {
+      "epoch": 4.800061551087403,
+      "grad_norm": 0.09162432700395584,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 46791
+    },
+    {
+      "epoch": 4.8003693065244155,
+      "grad_norm": 0.05243920534849167,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 46794
+    },
+    {
+      "epoch": 4.800677061961428,
+      "grad_norm": 0.05156363546848297,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 46797
+    },
+    {
+      "epoch": 4.800984817398441,
+      "grad_norm": 0.03927348554134369,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 46800
+    },
+    {
+      "epoch": 4.801292572835454,
+      "grad_norm": 0.04184743016958237,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 46803
+    },
+    {
+      "epoch": 4.801600328272466,
+      "grad_norm": 0.10456445813179016,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 46806
+    },
+    {
+      "epoch": 4.8019080837094785,
+      "grad_norm": 0.14495427906513214,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 46809
+    },
+    {
+      "epoch": 4.802215839146491,
+      "grad_norm": 0.06935403496026993,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 46812
+    },
+    {
+      "epoch": 4.802523594583504,
+      "grad_norm": 0.07718656957149506,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 46815
+    },
+    {
+      "epoch": 4.802831350020517,
+      "grad_norm": 0.04619847983121872,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 46818
+    },
+    {
+      "epoch": 4.80313910545753,
+      "grad_norm": 0.04528527334332466,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 46821
+    },
+    {
+      "epoch": 4.803446860894542,
+      "grad_norm": 0.0626635029911995,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 46824
+    },
+    {
+      "epoch": 4.803754616331555,
+      "grad_norm": 0.055482156574726105,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 46827
+    },
+    {
+      "epoch": 4.804062371768568,
+      "grad_norm": 0.04317907243967056,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 46830
+    },
+    {
+      "epoch": 4.804370127205581,
+      "grad_norm": 0.12392894178628922,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 46833
+    },
+    {
+      "epoch": 4.8046778826425935,
+      "grad_norm": 0.07023516297340393,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 46836
+    },
+    {
+      "epoch": 4.804985638079606,
+      "grad_norm": 0.10643202811479568,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 46839
+    },
+    {
+      "epoch": 4.805293393516619,
+      "grad_norm": 0.07375111430883408,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 46842
+    },
+    {
+      "epoch": 4.805601148953632,
+      "grad_norm": 0.039533890783786774,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 46845
+    },
+    {
+      "epoch": 4.805908904390645,
+      "grad_norm": 0.03812634199857712,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 46848
+    },
+    {
+      "epoch": 4.806216659827657,
+      "grad_norm": 0.049066439270973206,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 46851
+    },
+    {
+      "epoch": 4.806524415264669,
+      "grad_norm": 0.17093585431575775,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 46854
+    },
+    {
+      "epoch": 4.806832170701682,
+      "grad_norm": 0.06310732662677765,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 46857
+    },
+    {
+      "epoch": 4.807139926138695,
+      "grad_norm": 0.04781056195497513,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 46860
+    },
+    {
+      "epoch": 4.807447681575708,
+      "grad_norm": 0.04082764312624931,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 46863
+    },
+    {
+      "epoch": 4.8077554370127205,
+      "grad_norm": 0.06627890467643738,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 46866
+    },
+    {
+      "epoch": 4.808063192449733,
+      "grad_norm": 0.07753533869981766,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 46869
+    },
+    {
+      "epoch": 4.808370947886746,
+      "grad_norm": 0.05382109060883522,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 46872
+    },
+    {
+      "epoch": 4.808678703323759,
+      "grad_norm": 0.06153309345245361,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 46875
+    },
+    {
+      "epoch": 4.808986458760772,
+      "grad_norm": 0.0524064376950264,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 46878
+    },
+    {
+      "epoch": 4.809294214197784,
+      "grad_norm": 0.04740193486213684,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 46881
+    },
+    {
+      "epoch": 4.809601969634797,
+      "grad_norm": 0.04750111699104309,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 46884
+    },
+    {
+      "epoch": 4.80990972507181,
+      "grad_norm": 0.1953573226928711,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 46887
+    },
+    {
+      "epoch": 4.810217480508823,
+      "grad_norm": 0.05784987658262253,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 46890
+    },
+    {
+      "epoch": 4.8105252359458355,
+      "grad_norm": 0.06495459377765656,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 46893
+    },
+    {
+      "epoch": 4.810832991382847,
+      "grad_norm": 0.04319414868950844,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 46896
+    },
+    {
+      "epoch": 4.81114074681986,
+      "grad_norm": 0.05057989060878754,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 46899
+    },
+    {
+      "epoch": 4.811448502256873,
+      "grad_norm": 0.04836462065577507,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 46902
+    },
+    {
+      "epoch": 4.811756257693886,
+      "grad_norm": 0.06928513944149017,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 46905
+    },
+    {
+      "epoch": 4.8120640131308985,
+      "grad_norm": 0.04418247938156128,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 46908
+    },
+    {
+      "epoch": 4.812371768567911,
+      "grad_norm": 0.05600956827402115,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 46911
+    },
+    {
+      "epoch": 4.812679524004924,
+      "grad_norm": 0.18245935440063477,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 46914
+    },
+    {
+      "epoch": 4.812987279441937,
+      "grad_norm": 0.05147803574800491,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 46917
+    },
+    {
+      "epoch": 4.81329503487895,
+      "grad_norm": 0.049182165414094925,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 46920
+    },
+    {
+      "epoch": 4.813602790315962,
+      "grad_norm": 0.03926403820514679,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 46923
+    },
+    {
+      "epoch": 4.813910545752975,
+      "grad_norm": 0.04399009421467781,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 46926
+    },
+    {
+      "epoch": 4.814218301189988,
+      "grad_norm": 0.05575468763709068,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 46929
+    },
+    {
+      "epoch": 4.814526056627001,
+      "grad_norm": 0.056770700961351395,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 46932
+    },
+    {
+      "epoch": 4.8148338120640135,
+      "grad_norm": 0.0790528804063797,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 46935
+    },
+    {
+      "epoch": 4.815141567501026,
+      "grad_norm": 0.07976264506578445,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 46938
+    },
+    {
+      "epoch": 4.815449322938038,
+      "grad_norm": 0.058160681277513504,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 46941
+    },
+    {
+      "epoch": 4.815757078375051,
+      "grad_norm": 0.05894159525632858,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 46944
+    },
+    {
+      "epoch": 4.816064833812064,
+      "grad_norm": 0.061383627355098724,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 46947
+    },
+    {
+      "epoch": 4.8163725892490765,
+      "grad_norm": 0.06993051618337631,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 46950
+    },
+    {
+      "epoch": 4.816680344686089,
+      "grad_norm": 0.1546536386013031,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 46953
+    },
+    {
+      "epoch": 4.816988100123102,
+      "grad_norm": 0.11143519729375839,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 46956
+    },
+    {
+      "epoch": 4.817295855560115,
+      "grad_norm": 0.054721418768167496,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 46959
+    },
+    {
+      "epoch": 4.817603610997128,
+      "grad_norm": 0.04508695378899574,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 46962
+    },
+    {
+      "epoch": 4.81791136643414,
+      "grad_norm": 0.06738116592168808,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 46965
+    },
+    {
+      "epoch": 4.818219121871153,
+      "grad_norm": 0.05306596681475639,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 46968
+    },
+    {
+      "epoch": 4.818526877308166,
+      "grad_norm": 0.05283980816602707,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 46971
+    },
+    {
+      "epoch": 4.818834632745179,
+      "grad_norm": 0.04085366427898407,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 46974
+    },
+    {
+      "epoch": 4.8191423881821915,
+      "grad_norm": 0.0549020953476429,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 46977
+    },
+    {
+      "epoch": 4.819450143619204,
+      "grad_norm": 0.08194736391305923,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 46980
+    },
+    {
+      "epoch": 4.819757899056217,
+      "grad_norm": 0.07541119307279587,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 46983
+    },
+    {
+      "epoch": 4.820065654493229,
+      "grad_norm": 0.05164722725749016,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 46986
+    },
+    {
+      "epoch": 4.820373409930242,
+      "grad_norm": 0.03965797275304794,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 46989
+    },
+    {
+      "epoch": 4.8206811653672546,
+      "grad_norm": 0.05798366665840149,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 46992
+    },
+    {
+      "epoch": 4.820988920804267,
+      "grad_norm": 0.09747203439474106,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 46995
+    },
+    {
+      "epoch": 4.82129667624128,
+      "grad_norm": 0.037640564143657684,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 46998
+    },
+    {
+      "epoch": 4.821604431678293,
+      "grad_norm": 0.06714722514152527,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 47001
+    },
+    {
+      "epoch": 4.821912187115306,
+      "grad_norm": 0.06069006398320198,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 47004
+    },
+    {
+      "epoch": 4.8222199425523185,
+      "grad_norm": 0.04241282492876053,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 47007
+    },
+    {
+      "epoch": 4.822527697989331,
+      "grad_norm": 0.07319297641515732,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 47010
+    },
+    {
+      "epoch": 4.822835453426344,
+      "grad_norm": 0.11949057132005692,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 47013
+    },
+    {
+      "epoch": 4.823143208863357,
+      "grad_norm": 0.10720200836658478,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 47016
+    },
+    {
+      "epoch": 4.82345096430037,
+      "grad_norm": 0.05312662571668625,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 47019
+    },
+    {
+      "epoch": 4.823758719737382,
+      "grad_norm": 0.05288184434175491,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 47022
+    },
+    {
+      "epoch": 4.824066475174395,
+      "grad_norm": 0.04024795815348625,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 47025
+    },
+    {
+      "epoch": 4.824374230611408,
+      "grad_norm": 0.03714558109641075,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 47028
+    },
+    {
+      "epoch": 4.82468198604842,
+      "grad_norm": 0.043919868767261505,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 47031
+    },
+    {
+      "epoch": 4.824989741485433,
+      "grad_norm": 0.09199999272823334,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 47034
+    },
+    {
+      "epoch": 4.825297496922445,
+      "grad_norm": 0.09651891142129898,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 47037
+    },
+    {
+      "epoch": 4.825605252359458,
+      "grad_norm": 0.11211739480495453,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 47040
+    },
+    {
+      "epoch": 4.825913007796471,
+      "grad_norm": 0.11191114038228989,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 47043
+    },
+    {
+      "epoch": 4.826220763233484,
+      "grad_norm": 0.06726837903261185,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 47046
+    },
+    {
+      "epoch": 4.8265285186704965,
+      "grad_norm": 0.04758633300662041,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 47049
+    },
+    {
+      "epoch": 4.826836274107509,
+      "grad_norm": 0.07138072699308395,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 47052
+    },
+    {
+      "epoch": 4.827144029544522,
+      "grad_norm": 0.08691728860139847,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 47055
+    },
+    {
+      "epoch": 4.827451784981535,
+      "grad_norm": 0.060696471482515335,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 47058
+    },
+    {
+      "epoch": 4.827759540418548,
+      "grad_norm": 0.04225600138306618,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 47061
+    },
+    {
+      "epoch": 4.82806729585556,
+      "grad_norm": 0.03796116262674332,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 47064
+    },
+    {
+      "epoch": 4.828375051292573,
+      "grad_norm": 0.056367259472608566,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 47067
+    },
+    {
+      "epoch": 4.828682806729586,
+      "grad_norm": 0.10544770210981369,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 47070
+    },
+    {
+      "epoch": 4.828990562166599,
+      "grad_norm": 0.06839180737733841,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 47073
+    },
+    {
+      "epoch": 4.829298317603611,
+      "grad_norm": 0.08350997418165207,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 47076
+    },
+    {
+      "epoch": 4.829606073040623,
+      "grad_norm": 0.05540642887353897,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 47079
+    },
+    {
+      "epoch": 4.829913828477636,
+      "grad_norm": 0.08671069890260696,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 47082
+    },
+    {
+      "epoch": 4.830221583914649,
+      "grad_norm": 0.1410573422908783,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 47085
+    },
+    {
+      "epoch": 4.830529339351662,
+      "grad_norm": 0.06324710696935654,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 47088
+    },
+    {
+      "epoch": 4.8308370947886745,
+      "grad_norm": 0.04952717572450638,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 47091
+    },
+    {
+      "epoch": 4.831144850225687,
+      "grad_norm": 0.058474645018577576,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 47094
+    },
+    {
+      "epoch": 4.8314526056627,
+      "grad_norm": 0.08904005587100983,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 47097
+    },
+    {
+      "epoch": 4.831760361099713,
+      "grad_norm": 0.03303196281194687,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 47100
+    },
+    {
+      "epoch": 4.832068116536726,
+      "grad_norm": 0.0696646049618721,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 47103
+    },
+    {
+      "epoch": 4.832375871973738,
+      "grad_norm": 0.09613518416881561,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 47106
+    },
+    {
+      "epoch": 4.832683627410751,
+      "grad_norm": 0.08955409377813339,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 47109
+    },
+    {
+      "epoch": 4.832991382847764,
+      "grad_norm": 0.07022068649530411,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 47112
+    },
+    {
+      "epoch": 4.833299138284777,
+      "grad_norm": 0.05557604506611824,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 47115
+    },
+    {
+      "epoch": 4.8336068937217895,
+      "grad_norm": 0.04961394891142845,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 47118
+    },
+    {
+      "epoch": 4.833914649158801,
+      "grad_norm": 0.07381154596805573,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 47121
+    },
+    {
+      "epoch": 4.834222404595814,
+      "grad_norm": 0.08874998986721039,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 47124
+    },
+    {
+      "epoch": 4.834530160032827,
+      "grad_norm": 0.12152114510536194,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 47127
+    },
+    {
+      "epoch": 4.83483791546984,
+      "grad_norm": 0.18640320003032684,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 47130
+    },
+    {
+      "epoch": 4.8351456709068525,
+      "grad_norm": 0.13211971521377563,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 47133
+    },
+    {
+      "epoch": 4.835453426343865,
+      "grad_norm": 0.0825662836432457,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 47136
+    },
+    {
+      "epoch": 4.835761181780878,
+      "grad_norm": 0.05962991341948509,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 47139
+    },
+    {
+      "epoch": 4.836068937217891,
+      "grad_norm": 0.045787274837493896,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 47142
+    },
+    {
+      "epoch": 4.836376692654904,
+      "grad_norm": 0.05166277661919594,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 47145
+    },
+    {
+      "epoch": 4.836684448091916,
+      "grad_norm": 0.07144337892532349,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 47148
+    },
+    {
+      "epoch": 4.836992203528929,
+      "grad_norm": 0.03732317313551903,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 47151
+    },
+    {
+      "epoch": 4.837299958965942,
+      "grad_norm": 0.06613879650831223,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 47154
+    },
+    {
+      "epoch": 4.837607714402955,
+      "grad_norm": 0.05564868822693825,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 47157
+    },
+    {
+      "epoch": 4.837915469839967,
+      "grad_norm": 0.03259768709540367,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 47160
+    },
+    {
+      "epoch": 4.83822322527698,
+      "grad_norm": 0.11291679739952087,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 47163
+    },
+    {
+      "epoch": 4.838530980713992,
+      "grad_norm": 0.06787018477916718,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 47166
+    },
+    {
+      "epoch": 4.838838736151005,
+      "grad_norm": 0.10512536764144897,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 47169
+    },
+    {
+      "epoch": 4.839146491588018,
+      "grad_norm": 0.10330035537481308,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 47172
+    },
+    {
+      "epoch": 4.839454247025031,
+      "grad_norm": 0.06286308169364929,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 47175
+    },
+    {
+      "epoch": 4.839762002462043,
+      "grad_norm": 0.04793789982795715,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 47178
+    },
+    {
+      "epoch": 4.840069757899056,
+      "grad_norm": 0.043382007628679276,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 47181
+    },
+    {
+      "epoch": 4.840377513336069,
+      "grad_norm": 0.08022630959749222,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 47184
+    },
+    {
+      "epoch": 4.840685268773082,
+      "grad_norm": 0.09300918877124786,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 47187
+    },
+    {
+      "epoch": 4.8409930242100945,
+      "grad_norm": 0.05550304800271988,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 47190
+    },
+    {
+      "epoch": 4.841300779647107,
+      "grad_norm": 0.0407077856361866,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 47193
+    },
+    {
+      "epoch": 4.84160853508412,
+      "grad_norm": 0.04916580393910408,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 47196
+    },
+    {
+      "epoch": 4.841916290521133,
+      "grad_norm": 0.090577132999897,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 47199
+    },
+    {
+      "epoch": 4.842224045958146,
+      "grad_norm": 0.11230761557817459,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 47202
+    },
+    {
+      "epoch": 4.8425318013951575,
+      "grad_norm": 0.11779794096946716,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 47205
+    },
+    {
+      "epoch": 4.842839556832171,
+      "grad_norm": 0.04648299142718315,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 47208
+    },
+    {
+      "epoch": 4.843147312269183,
+      "grad_norm": 0.04901035502552986,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 47211
+    },
+    {
+      "epoch": 4.843455067706196,
+      "grad_norm": 0.05582377314567566,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 47214
+    },
+    {
+      "epoch": 4.843762823143209,
+      "grad_norm": 0.08325490355491638,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 47217
+    },
+    {
+      "epoch": 4.844070578580221,
+      "grad_norm": 0.09663175791501999,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 47220
+    },
+    {
+      "epoch": 4.844378334017234,
+      "grad_norm": 0.067304328083992,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 47223
+    },
+    {
+      "epoch": 4.844686089454247,
+      "grad_norm": 0.09804333001375198,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 47226
+    },
+    {
+      "epoch": 4.84499384489126,
+      "grad_norm": 0.07931843400001526,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 47229
+    },
+    {
+      "epoch": 4.8453016003282725,
+      "grad_norm": 0.07244410365819931,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 47232
+    },
+    {
+      "epoch": 4.845609355765285,
+      "grad_norm": 0.08068135380744934,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 47235
+    },
+    {
+      "epoch": 4.845917111202298,
+      "grad_norm": 0.062296342104673386,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 47238
+    },
+    {
+      "epoch": 4.846224866639311,
+      "grad_norm": 0.09988021850585938,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 47241
+    },
+    {
+      "epoch": 4.846532622076324,
+      "grad_norm": 0.12119688838720322,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 47244
+    },
+    {
+      "epoch": 4.846840377513336,
+      "grad_norm": 0.06407175213098526,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 47247
+    },
+    {
+      "epoch": 4.847148132950348,
+      "grad_norm": 0.05600906163454056,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 47250
+    },
+    {
+      "epoch": 4.847455888387362,
+      "grad_norm": 0.04448190703988075,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 47253
+    },
+    {
+      "epoch": 4.847763643824374,
+      "grad_norm": 0.046519339084625244,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 47256
+    },
+    {
+      "epoch": 4.848071399261387,
+      "grad_norm": 0.07584928721189499,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 47259
+    },
+    {
+      "epoch": 4.848379154698399,
+      "grad_norm": 0.06511086225509644,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 47262
+    },
+    {
+      "epoch": 4.848686910135412,
+      "grad_norm": 0.10978051275014877,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 47265
+    },
+    {
+      "epoch": 4.848994665572425,
+      "grad_norm": 0.10694348812103271,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 47268
+    },
+    {
+      "epoch": 4.849302421009438,
+      "grad_norm": 0.11489532142877579,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 47271
+    },
+    {
+      "epoch": 4.8496101764464505,
+      "grad_norm": 0.061532165855169296,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 47274
+    },
+    {
+      "epoch": 4.849917931883463,
+      "grad_norm": 0.034115977585315704,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 47277
+    },
+    {
+      "epoch": 4.850225687320476,
+      "grad_norm": 0.07114948332309723,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 47280
+    },
+    {
+      "epoch": 4.850533442757489,
+      "grad_norm": 0.03593161702156067,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 47283
+    },
+    {
+      "epoch": 4.850841198194502,
+      "grad_norm": 0.09017783403396606,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 47286
+    },
+    {
+      "epoch": 4.851148953631514,
+      "grad_norm": 0.045291002839803696,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 47289
+    },
+    {
+      "epoch": 4.851456709068527,
+      "grad_norm": 0.05328730493783951,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 47292
+    },
+    {
+      "epoch": 4.851764464505539,
+      "grad_norm": 0.08284175395965576,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 47295
+    },
+    {
+      "epoch": 4.852072219942553,
+      "grad_norm": 0.04549727216362953,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 47298
+    },
+    {
+      "epoch": 4.852379975379565,
+      "grad_norm": 0.09103915095329285,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 47301
+    },
+    {
+      "epoch": 4.8526877308165774,
+      "grad_norm": 0.09698940068483353,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 47304
+    },
+    {
+      "epoch": 4.85299548625359,
+      "grad_norm": 0.12437165528535843,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 47307
+    },
+    {
+      "epoch": 4.853303241690603,
+      "grad_norm": 0.0723615437746048,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 47310
+    },
+    {
+      "epoch": 4.853610997127616,
+      "grad_norm": 0.06390126794576645,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 47313
+    },
+    {
+      "epoch": 4.853918752564629,
+      "grad_norm": 0.04488823935389519,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 47316
+    },
+    {
+      "epoch": 4.854226508001641,
+      "grad_norm": 0.036022037267684937,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 47319
+    },
+    {
+      "epoch": 4.854534263438654,
+      "grad_norm": 0.06746529787778854,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 47322
+    },
+    {
+      "epoch": 4.854842018875667,
+      "grad_norm": 0.06267426908016205,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 47325
+    },
+    {
+      "epoch": 4.85514977431268,
+      "grad_norm": 0.08596473187208176,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 47328
+    },
+    {
+      "epoch": 4.8554575297496925,
+      "grad_norm": 0.04593675211071968,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 47331
+    },
+    {
+      "epoch": 4.855765285186705,
+      "grad_norm": 0.08707760274410248,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 47334
+    },
+    {
+      "epoch": 4.856073040623718,
+      "grad_norm": 0.05224407836794853,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 47337
+    },
+    {
+      "epoch": 4.85638079606073,
+      "grad_norm": 0.05910166725516319,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 47340
+    },
+    {
+      "epoch": 4.856688551497744,
+      "grad_norm": 0.14354678988456726,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 47343
+    },
+    {
+      "epoch": 4.8569963069347555,
+      "grad_norm": 0.11897499859333038,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 47346
+    },
+    {
+      "epoch": 4.857304062371768,
+      "grad_norm": 0.06904298812150955,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 47349
+    },
+    {
+      "epoch": 4.857611817808781,
+      "grad_norm": 0.04855037108063698,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 47352
+    },
+    {
+      "epoch": 4.857919573245794,
+      "grad_norm": 0.042485300451517105,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 47355
+    },
+    {
+      "epoch": 4.858227328682807,
+      "grad_norm": 0.052240677177906036,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 47358
+    },
+    {
+      "epoch": 4.858535084119819,
+      "grad_norm": 0.09882494062185287,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 47361
+    },
+    {
+      "epoch": 4.858842839556832,
+      "grad_norm": 0.059417568147182465,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 47364
+    },
+    {
+      "epoch": 4.859150594993845,
+      "grad_norm": 0.042187973856925964,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 47367
+    },
+    {
+      "epoch": 4.859458350430858,
+      "grad_norm": 0.11045199632644653,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 47370
+    },
+    {
+      "epoch": 4.8597661058678705,
+      "grad_norm": 0.06902570277452469,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 47373
+    },
+    {
+      "epoch": 4.860073861304883,
+      "grad_norm": 0.1256779134273529,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 47376
+    },
+    {
+      "epoch": 4.860381616741896,
+      "grad_norm": 0.0714133083820343,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 47379
+    },
+    {
+      "epoch": 4.860689372178909,
+      "grad_norm": 0.050711020827293396,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 47382
+    },
+    {
+      "epoch": 4.860997127615921,
+      "grad_norm": 0.04845008999109268,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 47385
+    },
+    {
+      "epoch": 4.861304883052934,
+      "grad_norm": 0.04254496470093727,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 47388
+    },
+    {
+      "epoch": 4.861612638489946,
+      "grad_norm": 0.03248690441250801,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 47391
+    },
+    {
+      "epoch": 4.861920393926959,
+      "grad_norm": 0.05999647080898285,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 47394
+    },
+    {
+      "epoch": 4.862228149363972,
+      "grad_norm": 0.09282340854406357,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 47397
+    },
+    {
+      "epoch": 4.862535904800985,
+      "grad_norm": 0.04029770568013191,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 47400
+    },
+    {
+      "epoch": 4.862843660237997,
+      "grad_norm": 0.030764177441596985,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 47403
+    },
+    {
+      "epoch": 4.86315141567501,
+      "grad_norm": 0.04076581448316574,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 47406
+    },
+    {
+      "epoch": 4.863459171112023,
+      "grad_norm": 0.0759870857000351,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 47409
+    },
+    {
+      "epoch": 4.863766926549036,
+      "grad_norm": 0.041376739740371704,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 47412
+    },
+    {
+      "epoch": 4.8640746819860485,
+      "grad_norm": 0.13923729956150055,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 47415
+    },
+    {
+      "epoch": 4.864382437423061,
+      "grad_norm": 0.06702505052089691,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 47418
+    },
+    {
+      "epoch": 4.864690192860074,
+      "grad_norm": 0.04206367954611778,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 47421
+    },
+    {
+      "epoch": 4.864997948297087,
+      "grad_norm": 0.043106552213430405,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 47424
+    },
+    {
+      "epoch": 4.8653057037341,
+      "grad_norm": 0.04751205816864967,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 47427
+    },
+    {
+      "epoch": 4.8656134591711115,
+      "grad_norm": 0.11286492645740509,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 47430
+    },
+    {
+      "epoch": 4.865921214608125,
+      "grad_norm": 0.04195297136902809,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 47433
+    },
+    {
+      "epoch": 4.866228970045137,
+      "grad_norm": 0.03481072559952736,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 47436
+    },
+    {
+      "epoch": 4.86653672548215,
+      "grad_norm": 0.0647101029753685,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 47439
+    },
+    {
+      "epoch": 4.866844480919163,
+      "grad_norm": 0.09517384320497513,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 47442
+    },
+    {
+      "epoch": 4.867152236356175,
+      "grad_norm": 0.06879560649394989,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 47445
+    },
+    {
+      "epoch": 4.867459991793188,
+      "grad_norm": 0.08635231107473373,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 47448
+    },
+    {
+      "epoch": 4.867767747230201,
+      "grad_norm": 0.04220673069357872,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 47451
+    },
+    {
+      "epoch": 4.868075502667214,
+      "grad_norm": 0.07045330852270126,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 47454
+    },
+    {
+      "epoch": 4.868383258104227,
+      "grad_norm": 0.04260876774787903,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 47457
+    },
+    {
+      "epoch": 4.868691013541239,
+      "grad_norm": 0.040992897003889084,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 47460
+    },
+    {
+      "epoch": 4.868998768978252,
+      "grad_norm": 0.04472249746322632,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 47463
+    },
+    {
+      "epoch": 4.869306524415265,
+      "grad_norm": 0.03990897163748741,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 47466
+    },
+    {
+      "epoch": 4.869614279852278,
+      "grad_norm": 0.10024430602788925,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 47469
+    },
+    {
+      "epoch": 4.8699220352892905,
+      "grad_norm": 0.08980470895767212,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 47472
+    },
+    {
+      "epoch": 4.870229790726302,
+      "grad_norm": 0.05399052053689957,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 47475
+    },
+    {
+      "epoch": 4.870537546163316,
+      "grad_norm": 0.03817398473620415,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 47478
+    },
+    {
+      "epoch": 4.870845301600328,
+      "grad_norm": 0.0670650526881218,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 47481
+    },
+    {
+      "epoch": 4.871153057037341,
+      "grad_norm": 0.0896904319524765,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 47484
+    },
+    {
+      "epoch": 4.8714608124743535,
+      "grad_norm": 0.10440392047166824,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 47487
+    },
+    {
+      "epoch": 4.871768567911366,
+      "grad_norm": 0.0894385576248169,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 47490
+    },
+    {
+      "epoch": 4.872076323348379,
+      "grad_norm": 0.05117709934711456,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 47493
+    },
+    {
+      "epoch": 4.872384078785392,
+      "grad_norm": 0.03292097896337509,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 47496
+    },
+    {
+      "epoch": 4.872691834222405,
+      "grad_norm": 0.032028477638959885,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 47499
+    },
+    {
+      "epoch": 4.872999589659417,
+      "grad_norm": 0.046977583318948746,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 47502
+    },
+    {
+      "epoch": 4.87330734509643,
+      "grad_norm": 0.11463116854429245,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 47505
+    },
+    {
+      "epoch": 4.873615100533443,
+      "grad_norm": 0.031687479466199875,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 47508
+    },
+    {
+      "epoch": 4.873922855970456,
+      "grad_norm": 0.04342150315642357,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 47511
+    },
+    {
+      "epoch": 4.8742306114074685,
+      "grad_norm": 0.03970487043261528,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 47514
+    },
+    {
+      "epoch": 4.874538366844481,
+      "grad_norm": 0.05668526887893677,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 47517
+    },
+    {
+      "epoch": 4.874846122281493,
+      "grad_norm": 0.05608886480331421,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 47520
+    },
+    {
+      "epoch": 4.875153877718507,
+      "grad_norm": 0.13083675503730774,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 47523
+    },
+    {
+      "epoch": 4.875461633155519,
+      "grad_norm": 0.040449466556310654,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 47526
+    },
+    {
+      "epoch": 4.8757693885925315,
+      "grad_norm": 0.08093597739934921,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 47529
+    },
+    {
+      "epoch": 4.876077144029544,
+      "grad_norm": 0.08471735566854477,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 47532
+    },
+    {
+      "epoch": 4.876384899466557,
+      "grad_norm": 0.05568789690732956,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 47535
+    },
+    {
+      "epoch": 4.87669265490357,
+      "grad_norm": 0.07529760152101517,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 47538
+    },
+    {
+      "epoch": 4.877000410340583,
+      "grad_norm": 0.06487952917814255,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 47541
+    },
+    {
+      "epoch": 4.877308165777595,
+      "grad_norm": 0.049183983355760574,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 47544
+    },
+    {
+      "epoch": 4.877615921214608,
+      "grad_norm": 0.08050165325403214,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 47547
+    },
+    {
+      "epoch": 4.877923676651621,
+      "grad_norm": 0.061584021896123886,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 47550
+    },
+    {
+      "epoch": 4.878231432088634,
+      "grad_norm": 0.052319157868623734,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 47553
+    },
+    {
+      "epoch": 4.8785391875256465,
+      "grad_norm": 0.059711821377277374,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 47556
+    },
+    {
+      "epoch": 4.878846942962659,
+      "grad_norm": 0.10813643783330917,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 47559
+    },
+    {
+      "epoch": 4.879154698399672,
+      "grad_norm": 0.0562468096613884,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 47562
+    },
+    {
+      "epoch": 4.879462453836684,
+      "grad_norm": 0.14491213858127594,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 47565
+    },
+    {
+      "epoch": 4.879770209273698,
+      "grad_norm": 0.06594674289226532,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 47568
+    },
+    {
+      "epoch": 4.8800779647107095,
+      "grad_norm": 0.05883457511663437,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 47571
+    },
+    {
+      "epoch": 4.880385720147722,
+      "grad_norm": 0.049259252846241,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 47574
+    },
+    {
+      "epoch": 4.880693475584735,
+      "grad_norm": 0.0817684531211853,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 47577
+    },
+    {
+      "epoch": 4.881001231021748,
+      "grad_norm": 0.07413051277399063,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 47580
+    },
+    {
+      "epoch": 4.881308986458761,
+      "grad_norm": 0.03806111216545105,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 47583
+    },
+    {
+      "epoch": 4.881616741895773,
+      "grad_norm": 0.11986687034368515,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 47586
+    },
+    {
+      "epoch": 4.881924497332786,
+      "grad_norm": 0.09377092123031616,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 47589
+    },
+    {
+      "epoch": 4.882232252769799,
+      "grad_norm": 0.05205828696489334,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 47592
+    },
+    {
+      "epoch": 4.882540008206812,
+      "grad_norm": 0.07014745473861694,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 47595
+    },
+    {
+      "epoch": 4.882847763643825,
+      "grad_norm": 0.04630004242062569,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 47598
+    },
+    {
+      "epoch": 4.883155519080837,
+      "grad_norm": 0.06072268262505531,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 47601
+    },
+    {
+      "epoch": 4.88346327451785,
+      "grad_norm": 0.04583202302455902,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 47604
+    },
+    {
+      "epoch": 4.883771029954863,
+      "grad_norm": 0.10760218650102615,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 47607
+    },
+    {
+      "epoch": 4.884078785391875,
+      "grad_norm": 0.09803519397974014,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 47610
+    },
+    {
+      "epoch": 4.8843865408288885,
+      "grad_norm": 0.05660640448331833,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 47613
+    },
+    {
+      "epoch": 4.8846942962659,
+      "grad_norm": 0.07300207018852234,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 47616
+    },
+    {
+      "epoch": 4.885002051702913,
+      "grad_norm": 0.05942653492093086,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 47619
+    },
+    {
+      "epoch": 4.885309807139926,
+      "grad_norm": 0.0775621309876442,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 47622
+    },
+    {
+      "epoch": 4.885617562576939,
+      "grad_norm": 0.08601850271224976,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 47625
+    },
+    {
+      "epoch": 4.8859253180139515,
+      "grad_norm": 0.07847361266613007,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 47628
+    },
+    {
+      "epoch": 4.886233073450964,
+      "grad_norm": 0.06876290589570999,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 47631
+    },
+    {
+      "epoch": 4.886540828887977,
+      "grad_norm": 0.06279505789279938,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 47634
+    },
+    {
+      "epoch": 4.88684858432499,
+      "grad_norm": 0.0461944080889225,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 47637
+    },
+    {
+      "epoch": 4.887156339762003,
+      "grad_norm": 0.08709289878606796,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 47640
+    },
+    {
+      "epoch": 4.887464095199015,
+      "grad_norm": 0.07311128079891205,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 47643
+    },
+    {
+      "epoch": 4.887771850636028,
+      "grad_norm": 0.11558615416288376,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 47646
+    },
+    {
+      "epoch": 4.888079606073041,
+      "grad_norm": 0.044442903250455856,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 47649
+    },
+    {
+      "epoch": 4.888387361510054,
+      "grad_norm": 0.03592165187001228,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 47652
+    },
+    {
+      "epoch": 4.888695116947066,
+      "grad_norm": 0.053503166884183884,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 47655
+    },
+    {
+      "epoch": 4.889002872384079,
+      "grad_norm": 0.05185321345925331,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 47658
+    },
+    {
+      "epoch": 4.889310627821091,
+      "grad_norm": 0.10305842012166977,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 47661
+    },
+    {
+      "epoch": 4.889618383258104,
+      "grad_norm": 0.04605614393949509,
+      "learning_rate": 0.01,
+      "loss": 2.0359,
+      "step": 47664
+    },
+    {
+      "epoch": 4.889926138695117,
+      "grad_norm": 0.08601140230894089,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 47667
+    },
+    {
+      "epoch": 4.8902338941321295,
+      "grad_norm": 0.039195116609334946,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 47670
+    },
+    {
+      "epoch": 4.890541649569142,
+      "grad_norm": 0.04058361053466797,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 47673
+    },
+    {
+      "epoch": 4.890849405006155,
+      "grad_norm": 0.049590058624744415,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 47676
+    },
+    {
+      "epoch": 4.891157160443168,
+      "grad_norm": 0.05253410339355469,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 47679
+    },
+    {
+      "epoch": 4.891464915880181,
+      "grad_norm": 0.05858082324266434,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 47682
+    },
+    {
+      "epoch": 4.891772671317193,
+      "grad_norm": 0.05094842612743378,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 47685
+    },
+    {
+      "epoch": 4.892080426754206,
+      "grad_norm": 0.0539892241358757,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 47688
+    },
+    {
+      "epoch": 4.892388182191219,
+      "grad_norm": 0.15945689380168915,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 47691
+    },
+    {
+      "epoch": 4.892695937628232,
+      "grad_norm": 0.03433748334646225,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 47694
+    },
+    {
+      "epoch": 4.8930036930652445,
+      "grad_norm": 0.047357238829135895,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 47697
+    },
+    {
+      "epoch": 4.893311448502256,
+      "grad_norm": 0.05974509194493294,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 47700
+    },
+    {
+      "epoch": 4.89361920393927,
+      "grad_norm": 0.05918588861823082,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 47703
+    },
+    {
+      "epoch": 4.893926959376282,
+      "grad_norm": 0.04418300464749336,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 47706
+    },
+    {
+      "epoch": 4.894234714813295,
+      "grad_norm": 0.034721486270427704,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 47709
+    },
+    {
+      "epoch": 4.8945424702503075,
+      "grad_norm": 0.08422865718603134,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 47712
+    },
+    {
+      "epoch": 4.89485022568732,
+      "grad_norm": 0.12269464135169983,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 47715
+    },
+    {
+      "epoch": 4.895157981124333,
+      "grad_norm": 0.06587913632392883,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 47718
+    },
+    {
+      "epoch": 4.895465736561346,
+      "grad_norm": 0.12484870105981827,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 47721
+    },
+    {
+      "epoch": 4.895773491998359,
+      "grad_norm": 0.07100055366754532,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 47724
+    },
+    {
+      "epoch": 4.896081247435371,
+      "grad_norm": 0.03272338956594467,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 47727
+    },
+    {
+      "epoch": 4.896389002872384,
+      "grad_norm": 0.03650757670402527,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 47730
+    },
+    {
+      "epoch": 4.896696758309397,
+      "grad_norm": 0.04600701853632927,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 47733
+    },
+    {
+      "epoch": 4.89700451374641,
+      "grad_norm": 0.06520526856184006,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 47736
+    },
+    {
+      "epoch": 4.8973122691834226,
+      "grad_norm": 0.04096482694149017,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 47739
+    },
+    {
+      "epoch": 4.897620024620435,
+      "grad_norm": 0.03720017522573471,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 47742
+    },
+    {
+      "epoch": 4.897927780057447,
+      "grad_norm": 0.04412877559661865,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 47745
+    },
+    {
+      "epoch": 4.898235535494461,
+      "grad_norm": 0.03968506306409836,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 47748
+    },
+    {
+      "epoch": 4.898543290931473,
+      "grad_norm": 0.10775657743215561,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 47751
+    },
+    {
+      "epoch": 4.898851046368486,
+      "grad_norm": 0.05303318426012993,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 47754
+    },
+    {
+      "epoch": 4.899158801805498,
+      "grad_norm": 0.0775865837931633,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 47757
+    },
+    {
+      "epoch": 4.899466557242511,
+      "grad_norm": 0.10708294063806534,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 47760
+    },
+    {
+      "epoch": 4.899774312679524,
+      "grad_norm": 0.05531914532184601,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 47763
+    },
+    {
+      "epoch": 4.900082068116537,
+      "grad_norm": 0.057492613792419434,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 47766
+    },
+    {
+      "epoch": 4.9003898235535495,
+      "grad_norm": 0.04654363915324211,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 47769
+    },
+    {
+      "epoch": 4.900697578990562,
+      "grad_norm": 0.060792889446020126,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 47772
+    },
+    {
+      "epoch": 4.901005334427575,
+      "grad_norm": 0.053037889301776886,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 47775
+    },
+    {
+      "epoch": 4.901313089864588,
+      "grad_norm": 0.07746932655572891,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 47778
+    },
+    {
+      "epoch": 4.901620845301601,
+      "grad_norm": 0.07753589004278183,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 47781
+    },
+    {
+      "epoch": 4.901928600738613,
+      "grad_norm": 0.07140833884477615,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 47784
+    },
+    {
+      "epoch": 4.902236356175626,
+      "grad_norm": 0.12805962562561035,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 47787
+    },
+    {
+      "epoch": 4.902544111612638,
+      "grad_norm": 0.10444001108407974,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 47790
+    },
+    {
+      "epoch": 4.902851867049652,
+      "grad_norm": 0.13313624262809753,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 47793
+    },
+    {
+      "epoch": 4.903159622486664,
+      "grad_norm": 0.0864720493555069,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 47796
+    },
+    {
+      "epoch": 4.903467377923676,
+      "grad_norm": 0.05329997465014458,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 47799
+    },
+    {
+      "epoch": 4.903775133360689,
+      "grad_norm": 0.053774427622556686,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 47802
+    },
+    {
+      "epoch": 4.904082888797702,
+      "grad_norm": 0.06760123372077942,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 47805
+    },
+    {
+      "epoch": 4.904390644234715,
+      "grad_norm": 0.06961317360401154,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 47808
+    },
+    {
+      "epoch": 4.9046983996717275,
+      "grad_norm": 0.05297344923019409,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 47811
+    },
+    {
+      "epoch": 4.90500615510874,
+      "grad_norm": 0.06930317729711533,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 47814
+    },
+    {
+      "epoch": 4.905313910545753,
+      "grad_norm": 0.039554011076688766,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 47817
+    },
+    {
+      "epoch": 4.905621665982766,
+      "grad_norm": 0.03862786293029785,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 47820
+    },
+    {
+      "epoch": 4.905929421419779,
+      "grad_norm": 0.03559514880180359,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 47823
+    },
+    {
+      "epoch": 4.906237176856791,
+      "grad_norm": 0.11802364140748978,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 47826
+    },
+    {
+      "epoch": 4.906544932293804,
+      "grad_norm": 0.049168869853019714,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 47829
+    },
+    {
+      "epoch": 4.906852687730817,
+      "grad_norm": 0.07636560499668121,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 47832
+    },
+    {
+      "epoch": 4.907160443167829,
+      "grad_norm": 0.08972848206758499,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 47835
+    },
+    {
+      "epoch": 4.9074681986048425,
+      "grad_norm": 0.06916697323322296,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 47838
+    },
+    {
+      "epoch": 4.907775954041854,
+      "grad_norm": 0.058684833347797394,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 47841
+    },
+    {
+      "epoch": 4.908083709478867,
+      "grad_norm": 0.045184120535850525,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 47844
+    },
+    {
+      "epoch": 4.90839146491588,
+      "grad_norm": 0.11755534261465073,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 47847
+    },
+    {
+      "epoch": 4.908699220352893,
+      "grad_norm": 0.07570980489253998,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 47850
+    },
+    {
+      "epoch": 4.9090069757899055,
+      "grad_norm": 0.050079140812158585,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 47853
+    },
+    {
+      "epoch": 4.909314731226918,
+      "grad_norm": 0.12019573897123337,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 47856
+    },
+    {
+      "epoch": 4.909622486663931,
+      "grad_norm": 0.10437455028295517,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 47859
+    },
+    {
+      "epoch": 4.909930242100944,
+      "grad_norm": 0.05818986892700195,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 47862
+    },
+    {
+      "epoch": 4.910237997537957,
+      "grad_norm": 0.04473674297332764,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 47865
+    },
+    {
+      "epoch": 4.910545752974969,
+      "grad_norm": 0.07120556384325027,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 47868
+    },
+    {
+      "epoch": 4.910853508411982,
+      "grad_norm": 0.0661742091178894,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 47871
+    },
+    {
+      "epoch": 4.911161263848995,
+      "grad_norm": 0.07108204811811447,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 47874
+    },
+    {
+      "epoch": 4.911469019286008,
+      "grad_norm": 0.09054633975028992,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 47877
+    },
+    {
+      "epoch": 4.91177677472302,
+      "grad_norm": 0.13030128180980682,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 47880
+    },
+    {
+      "epoch": 4.912084530160033,
+      "grad_norm": 0.15652155876159668,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 47883
+    },
+    {
+      "epoch": 4.912392285597045,
+      "grad_norm": 0.10536278784275055,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 47886
+    },
+    {
+      "epoch": 4.912700041034058,
+      "grad_norm": 0.0899362787604332,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 47889
+    },
+    {
+      "epoch": 4.913007796471071,
+      "grad_norm": 0.057359661906957626,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 47892
+    },
+    {
+      "epoch": 4.913315551908084,
+      "grad_norm": 0.04800909757614136,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 47895
+    },
+    {
+      "epoch": 4.913623307345096,
+      "grad_norm": 0.04786820337176323,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 47898
+    },
+    {
+      "epoch": 4.913931062782109,
+      "grad_norm": 0.04169714078307152,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 47901
+    },
+    {
+      "epoch": 4.914238818219122,
+      "grad_norm": 0.04943560063838959,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 47904
+    },
+    {
+      "epoch": 4.914546573656135,
+      "grad_norm": 0.06505610793828964,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 47907
+    },
+    {
+      "epoch": 4.9148543290931475,
+      "grad_norm": 0.057665206491947174,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 47910
+    },
+    {
+      "epoch": 4.91516208453016,
+      "grad_norm": 0.060052540153265,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 47913
+    },
+    {
+      "epoch": 4.915469839967173,
+      "grad_norm": 0.062361933290958405,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 47916
+    },
+    {
+      "epoch": 4.915777595404186,
+      "grad_norm": 0.15210957825183868,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 47919
+    },
+    {
+      "epoch": 4.916085350841199,
+      "grad_norm": 0.05437038466334343,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 47922
+    },
+    {
+      "epoch": 4.9163931062782105,
+      "grad_norm": 0.07012643665075302,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 47925
+    },
+    {
+      "epoch": 4.916700861715223,
+      "grad_norm": 0.056515853852033615,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 47928
+    },
+    {
+      "epoch": 4.917008617152236,
+      "grad_norm": 0.13166914880275726,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 47931
+    },
+    {
+      "epoch": 4.917316372589249,
+      "grad_norm": 0.05462309718132019,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 47934
+    },
+    {
+      "epoch": 4.917624128026262,
+      "grad_norm": 0.0725654810667038,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 47937
+    },
+    {
+      "epoch": 4.917931883463274,
+      "grad_norm": 0.09769755601882935,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 47940
+    },
+    {
+      "epoch": 4.918239638900287,
+      "grad_norm": 0.10543838888406754,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 47943
+    },
+    {
+      "epoch": 4.9185473943373,
+      "grad_norm": 0.08627963066101074,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 47946
+    },
+    {
+      "epoch": 4.918855149774313,
+      "grad_norm": 0.05162891000509262,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 47949
+    },
+    {
+      "epoch": 4.9191629052113255,
+      "grad_norm": 0.040748897939920425,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 47952
+    },
+    {
+      "epoch": 4.919470660648338,
+      "grad_norm": 0.042143408209085464,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 47955
+    },
+    {
+      "epoch": 4.919778416085351,
+      "grad_norm": 0.035993028432130814,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 47958
+    },
+    {
+      "epoch": 4.920086171522364,
+      "grad_norm": 0.13255073130130768,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 47961
+    },
+    {
+      "epoch": 4.920393926959377,
+      "grad_norm": 0.06527701765298843,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 47964
+    },
+    {
+      "epoch": 4.920701682396389,
+      "grad_norm": 0.08764603734016418,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 47967
+    },
+    {
+      "epoch": 4.921009437833401,
+      "grad_norm": 0.11476396769285202,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 47970
+    },
+    {
+      "epoch": 4.921317193270414,
+      "grad_norm": 0.11724267154932022,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 47973
+    },
+    {
+      "epoch": 4.921624948707427,
+      "grad_norm": 0.06371249258518219,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 47976
+    },
+    {
+      "epoch": 4.92193270414444,
+      "grad_norm": 0.05650569498538971,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 47979
+    },
+    {
+      "epoch": 4.922240459581452,
+      "grad_norm": 0.05101975426077843,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 47982
+    },
+    {
+      "epoch": 4.922548215018465,
+      "grad_norm": 0.06948293745517731,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 47985
+    },
+    {
+      "epoch": 4.922855970455478,
+      "grad_norm": 0.05776602029800415,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 47988
+    },
+    {
+      "epoch": 4.923163725892491,
+      "grad_norm": 0.03927835077047348,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 47991
+    },
+    {
+      "epoch": 4.9234714813295035,
+      "grad_norm": 0.07943009585142136,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 47994
+    },
+    {
+      "epoch": 4.923779236766516,
+      "grad_norm": 0.07352134585380554,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 47997
+    },
+    {
+      "epoch": 4.924086992203529,
+      "grad_norm": 0.07295040786266327,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 48000
+    },
+    {
+      "epoch": 4.924394747640542,
+      "grad_norm": 0.08415599912405014,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 48003
+    },
+    {
+      "epoch": 4.924702503077555,
+      "grad_norm": 0.0890984758734703,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 48006
+    },
+    {
+      "epoch": 4.925010258514567,
+      "grad_norm": 0.12055069953203201,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 48009
+    },
+    {
+      "epoch": 4.92531801395158,
+      "grad_norm": 0.0676504373550415,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 48012
+    },
+    {
+      "epoch": 4.925625769388592,
+      "grad_norm": 0.047612790018320084,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 48015
+    },
+    {
+      "epoch": 4.925933524825605,
+      "grad_norm": 0.06470039486885071,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 48018
+    },
+    {
+      "epoch": 4.926241280262618,
+      "grad_norm": 0.07510625571012497,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 48021
+    },
+    {
+      "epoch": 4.92654903569963,
+      "grad_norm": 0.07793190330266953,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 48024
+    },
+    {
+      "epoch": 4.926856791136643,
+      "grad_norm": 0.08655183017253876,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 48027
+    },
+    {
+      "epoch": 4.927164546573656,
+      "grad_norm": 0.07441940903663635,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 48030
+    },
+    {
+      "epoch": 4.927472302010669,
+      "grad_norm": 0.04783869534730911,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 48033
+    },
+    {
+      "epoch": 4.9277800574476815,
+      "grad_norm": 0.13127107918262482,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 48036
+    },
+    {
+      "epoch": 4.928087812884694,
+      "grad_norm": 0.05267849192023277,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 48039
+    },
+    {
+      "epoch": 4.928395568321707,
+      "grad_norm": 0.04276135936379433,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 48042
+    },
+    {
+      "epoch": 4.92870332375872,
+      "grad_norm": 0.04237223416566849,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 48045
+    },
+    {
+      "epoch": 4.929011079195733,
+      "grad_norm": 0.052018001675605774,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 48048
+    },
+    {
+      "epoch": 4.9293188346327454,
+      "grad_norm": 0.05408487468957901,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 48051
+    },
+    {
+      "epoch": 4.929626590069758,
+      "grad_norm": 0.07300473004579544,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 48054
+    },
+    {
+      "epoch": 4.929934345506771,
+      "grad_norm": 0.08607280254364014,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 48057
+    },
+    {
+      "epoch": 4.930242100943783,
+      "grad_norm": 0.061927784234285355,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 48060
+    },
+    {
+      "epoch": 4.930549856380796,
+      "grad_norm": 0.08371556550264359,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 48063
+    },
+    {
+      "epoch": 4.9308576118178085,
+      "grad_norm": 0.03664041683077812,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 48066
+    },
+    {
+      "epoch": 4.931165367254821,
+      "grad_norm": 0.06964648514986038,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 48069
+    },
+    {
+      "epoch": 4.931473122691834,
+      "grad_norm": 0.07257051765918732,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 48072
+    },
+    {
+      "epoch": 4.931780878128847,
+      "grad_norm": 0.06781255453824997,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 48075
+    },
+    {
+      "epoch": 4.93208863356586,
+      "grad_norm": 0.04258018732070923,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 48078
+    },
+    {
+      "epoch": 4.932396389002872,
+      "grad_norm": 0.05434480682015419,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 48081
+    },
+    {
+      "epoch": 4.932704144439885,
+      "grad_norm": 0.03577110916376114,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 48084
+    },
+    {
+      "epoch": 4.933011899876898,
+      "grad_norm": 0.11260990798473358,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 48087
+    },
+    {
+      "epoch": 4.933319655313911,
+      "grad_norm": 0.09070918709039688,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 48090
+    },
+    {
+      "epoch": 4.9336274107509235,
+      "grad_norm": 0.06896468997001648,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 48093
+    },
+    {
+      "epoch": 4.933935166187936,
+      "grad_norm": 0.050933387130498886,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 48096
+    },
+    {
+      "epoch": 4.934242921624949,
+      "grad_norm": 0.04397554323077202,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 48099
+    },
+    {
+      "epoch": 4.934550677061962,
+      "grad_norm": 0.03253142163157463,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 48102
+    },
+    {
+      "epoch": 4.934858432498974,
+      "grad_norm": 0.051716264337301254,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 48105
+    },
+    {
+      "epoch": 4.9351661879359865,
+      "grad_norm": 0.04921911656856537,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 48108
+    },
+    {
+      "epoch": 4.935473943372999,
+      "grad_norm": 0.05059951916337013,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 48111
+    },
+    {
+      "epoch": 4.935781698810012,
+      "grad_norm": 0.11451517045497894,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 48114
+    },
+    {
+      "epoch": 4.936089454247025,
+      "grad_norm": 0.06287554651498795,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 48117
+    },
+    {
+      "epoch": 4.936397209684038,
+      "grad_norm": 0.11319955438375473,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 48120
+    },
+    {
+      "epoch": 4.93670496512105,
+      "grad_norm": 0.04149606078863144,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 48123
+    },
+    {
+      "epoch": 4.937012720558063,
+      "grad_norm": 0.044562436640262604,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 48126
+    },
+    {
+      "epoch": 4.937320475995076,
+      "grad_norm": 0.04073583707213402,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 48129
+    },
+    {
+      "epoch": 4.937628231432089,
+      "grad_norm": 0.05489436164498329,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 48132
+    },
+    {
+      "epoch": 4.9379359868691015,
+      "grad_norm": 0.07751783728599548,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 48135
+    },
+    {
+      "epoch": 4.938243742306114,
+      "grad_norm": 0.06630382686853409,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 48138
+    },
+    {
+      "epoch": 4.938551497743127,
+      "grad_norm": 0.07446899265050888,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 48141
+    },
+    {
+      "epoch": 4.93885925318014,
+      "grad_norm": 0.0776534453034401,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 48144
+    },
+    {
+      "epoch": 4.939167008617153,
+      "grad_norm": 0.07835227996110916,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 48147
+    },
+    {
+      "epoch": 4.9394747640541645,
+      "grad_norm": 0.0798632800579071,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 48150
+    },
+    {
+      "epoch": 4.939782519491177,
+      "grad_norm": 0.044298071414232254,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 48153
+    },
+    {
+      "epoch": 4.94009027492819,
+      "grad_norm": 0.10383112728595734,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 48156
+    },
+    {
+      "epoch": 4.940398030365203,
+      "grad_norm": 0.12793728709220886,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 48159
+    },
+    {
+      "epoch": 4.940705785802216,
+      "grad_norm": 0.038207195699214935,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 48162
+    },
+    {
+      "epoch": 4.941013541239228,
+      "grad_norm": 0.0840001255273819,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 48165
+    },
+    {
+      "epoch": 4.941321296676241,
+      "grad_norm": 0.11788827925920486,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 48168
+    },
+    {
+      "epoch": 4.941629052113254,
+      "grad_norm": 0.05462497100234032,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 48171
+    },
+    {
+      "epoch": 4.941936807550267,
+      "grad_norm": 0.0785188302397728,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 48174
+    },
+    {
+      "epoch": 4.9422445629872795,
+      "grad_norm": 0.09669160097837448,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 48177
+    },
+    {
+      "epoch": 4.942552318424292,
+      "grad_norm": 0.0880751982331276,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 48180
+    },
+    {
+      "epoch": 4.942860073861305,
+      "grad_norm": 0.04356632009148598,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 48183
+    },
+    {
+      "epoch": 4.943167829298318,
+      "grad_norm": 0.055611494928598404,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 48186
+    },
+    {
+      "epoch": 4.943475584735331,
+      "grad_norm": 0.05191848427057266,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 48189
+    },
+    {
+      "epoch": 4.943783340172343,
+      "grad_norm": 0.06415262818336487,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 48192
+    },
+    {
+      "epoch": 4.944091095609355,
+      "grad_norm": 0.04284198582172394,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 48195
+    },
+    {
+      "epoch": 4.944398851046368,
+      "grad_norm": 0.03310967609286308,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 48198
+    },
+    {
+      "epoch": 4.944706606483381,
+      "grad_norm": 0.11449988186359406,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 48201
+    },
+    {
+      "epoch": 4.945014361920394,
+      "grad_norm": 0.0641254335641861,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 48204
+    },
+    {
+      "epoch": 4.9453221173574065,
+      "grad_norm": 0.08665329962968826,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 48207
+    },
+    {
+      "epoch": 4.945629872794419,
+      "grad_norm": 0.03855408355593681,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 48210
+    },
+    {
+      "epoch": 4.945937628231432,
+      "grad_norm": 0.03156152740120888,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 48213
+    },
+    {
+      "epoch": 4.946245383668445,
+      "grad_norm": 0.052971091121435165,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 48216
+    },
+    {
+      "epoch": 4.946553139105458,
+      "grad_norm": 0.04167775437235832,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 48219
+    },
+    {
+      "epoch": 4.94686089454247,
+      "grad_norm": 0.051760535687208176,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 48222
+    },
+    {
+      "epoch": 4.947168649979483,
+      "grad_norm": 0.0969996377825737,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 48225
+    },
+    {
+      "epoch": 4.947476405416496,
+      "grad_norm": 0.05064645782113075,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 48228
+    },
+    {
+      "epoch": 4.947784160853509,
+      "grad_norm": 0.04742693901062012,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 48231
+    },
+    {
+      "epoch": 4.9480919162905215,
+      "grad_norm": 0.03507422283291817,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 48234
+    },
+    {
+      "epoch": 4.948399671727534,
+      "grad_norm": 0.03717927634716034,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 48237
+    },
+    {
+      "epoch": 4.948707427164546,
+      "grad_norm": 0.11404886096715927,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 48240
+    },
+    {
+      "epoch": 4.949015182601559,
+      "grad_norm": 0.05842389166355133,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 48243
+    },
+    {
+      "epoch": 4.949322938038572,
+      "grad_norm": 0.0430951751768589,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 48246
+    },
+    {
+      "epoch": 4.9496306934755845,
+      "grad_norm": 0.053793564438819885,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 48249
+    },
+    {
+      "epoch": 4.949938448912597,
+      "grad_norm": 0.04643286019563675,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 48252
+    },
+    {
+      "epoch": 4.95024620434961,
+      "grad_norm": 0.0575956366956234,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 48255
+    },
+    {
+      "epoch": 4.950553959786623,
+      "grad_norm": 0.05439090356230736,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 48258
+    },
+    {
+      "epoch": 4.950861715223636,
+      "grad_norm": 0.05502607673406601,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 48261
+    },
+    {
+      "epoch": 4.951169470660648,
+      "grad_norm": 0.07984888553619385,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 48264
+    },
+    {
+      "epoch": 4.951477226097661,
+      "grad_norm": 0.041320640593767166,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 48267
+    },
+    {
+      "epoch": 4.951784981534674,
+      "grad_norm": 0.12548021972179413,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 48270
+    },
+    {
+      "epoch": 4.952092736971687,
+      "grad_norm": 0.035638876259326935,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 48273
+    },
+    {
+      "epoch": 4.9524004924086995,
+      "grad_norm": 0.13796448707580566,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 48276
+    },
+    {
+      "epoch": 4.952708247845712,
+      "grad_norm": 0.08978889882564545,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 48279
+    },
+    {
+      "epoch": 4.953016003282725,
+      "grad_norm": 0.0677703395485878,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 48282
+    },
+    {
+      "epoch": 4.953323758719737,
+      "grad_norm": 0.04120895639061928,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 48285
+    },
+    {
+      "epoch": 4.95363151415675,
+      "grad_norm": 0.043558813631534576,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 48288
+    },
+    {
+      "epoch": 4.9539392695937625,
+      "grad_norm": 0.03488008677959442,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 48291
+    },
+    {
+      "epoch": 4.954247025030775,
+      "grad_norm": 0.06290942430496216,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 48294
+    },
+    {
+      "epoch": 4.954554780467788,
+      "grad_norm": 0.05662504583597183,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 48297
+    },
+    {
+      "epoch": 4.954862535904801,
+      "grad_norm": 0.04810362681746483,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 48300
+    },
+    {
+      "epoch": 4.955170291341814,
+      "grad_norm": 0.06158094108104706,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 48303
+    },
+    {
+      "epoch": 4.955478046778826,
+      "grad_norm": 0.1016242504119873,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 48306
+    },
+    {
+      "epoch": 4.955785802215839,
+      "grad_norm": 0.05129433050751686,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 48309
+    },
+    {
+      "epoch": 4.956093557652852,
+      "grad_norm": 0.09010904282331467,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 48312
+    },
+    {
+      "epoch": 4.956401313089865,
+      "grad_norm": 0.09995526820421219,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 48315
+    },
+    {
+      "epoch": 4.9567090685268775,
+      "grad_norm": 0.0642913207411766,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 48318
+    },
+    {
+      "epoch": 4.95701682396389,
+      "grad_norm": 0.03862082585692406,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 48321
+    },
+    {
+      "epoch": 4.957324579400903,
+      "grad_norm": 0.11760932207107544,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 48324
+    },
+    {
+      "epoch": 4.957632334837916,
+      "grad_norm": 0.09624608606100082,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 48327
+    },
+    {
+      "epoch": 4.957940090274928,
+      "grad_norm": 0.06814394891262054,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 48330
+    },
+    {
+      "epoch": 4.9582478457119405,
+      "grad_norm": 0.09379391372203827,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 48333
+    },
+    {
+      "epoch": 4.958555601148953,
+      "grad_norm": 0.04310177266597748,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 48336
+    },
+    {
+      "epoch": 4.958863356585966,
+      "grad_norm": 0.048953697085380554,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 48339
+    },
+    {
+      "epoch": 4.959171112022979,
+      "grad_norm": 0.03562283515930176,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 48342
+    },
+    {
+      "epoch": 4.959478867459992,
+      "grad_norm": 0.040618475526571274,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 48345
+    },
+    {
+      "epoch": 4.9597866228970044,
+      "grad_norm": 0.11639663577079773,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 48348
+    },
+    {
+      "epoch": 4.960094378334017,
+      "grad_norm": 0.07946206629276276,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 48351
+    },
+    {
+      "epoch": 4.96040213377103,
+      "grad_norm": 0.05358874425292015,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 48354
+    },
+    {
+      "epoch": 4.960709889208043,
+      "grad_norm": 0.07157052308320999,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 48357
+    },
+    {
+      "epoch": 4.961017644645056,
+      "grad_norm": 0.038184091448783875,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 48360
+    },
+    {
+      "epoch": 4.961325400082068,
+      "grad_norm": 0.04140111058950424,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 48363
+    },
+    {
+      "epoch": 4.961633155519081,
+      "grad_norm": 0.08905436098575592,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 48366
+    },
+    {
+      "epoch": 4.961940910956094,
+      "grad_norm": 0.07200294733047485,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 48369
+    },
+    {
+      "epoch": 4.962248666393107,
+      "grad_norm": 0.0786944180727005,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 48372
+    },
+    {
+      "epoch": 4.962556421830119,
+      "grad_norm": 0.08718118071556091,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 48375
+    },
+    {
+      "epoch": 4.962864177267131,
+      "grad_norm": 0.10729999840259552,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 48378
+    },
+    {
+      "epoch": 4.963171932704144,
+      "grad_norm": 0.05224734544754028,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 48381
+    },
+    {
+      "epoch": 4.963479688141157,
+      "grad_norm": 0.10347335785627365,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 48384
+    },
+    {
+      "epoch": 4.96378744357817,
+      "grad_norm": 0.09103485941886902,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 48387
+    },
+    {
+      "epoch": 4.9640951990151825,
+      "grad_norm": 0.05988696217536926,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 48390
+    },
+    {
+      "epoch": 4.964402954452195,
+      "grad_norm": 0.05333305522799492,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 48393
+    },
+    {
+      "epoch": 4.964710709889208,
+      "grad_norm": 0.039692364633083344,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 48396
+    },
+    {
+      "epoch": 4.965018465326221,
+      "grad_norm": 0.048610180616378784,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 48399
+    },
+    {
+      "epoch": 4.965326220763234,
+      "grad_norm": 0.06438957899808884,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 48402
+    },
+    {
+      "epoch": 4.965633976200246,
+      "grad_norm": 0.06306184828281403,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 48405
+    },
+    {
+      "epoch": 4.965941731637259,
+      "grad_norm": 0.0924021527171135,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 48408
+    },
+    {
+      "epoch": 4.966249487074272,
+      "grad_norm": 0.06868547946214676,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 48411
+    },
+    {
+      "epoch": 4.966557242511285,
+      "grad_norm": 0.04776471480727196,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 48414
+    },
+    {
+      "epoch": 4.9668649979482975,
+      "grad_norm": 0.0417228639125824,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 48417
+    },
+    {
+      "epoch": 4.967172753385309,
+      "grad_norm": 0.03331043943762779,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 48420
+    },
+    {
+      "epoch": 4.967480508822322,
+      "grad_norm": 0.07828322052955627,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 48423
+    },
+    {
+      "epoch": 4.967788264259335,
+      "grad_norm": 0.11830344796180725,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 48426
+    },
+    {
+      "epoch": 4.968096019696348,
+      "grad_norm": 0.13753825426101685,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 48429
+    },
+    {
+      "epoch": 4.9684037751333605,
+      "grad_norm": 0.10867384821176529,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 48432
+    },
+    {
+      "epoch": 4.968711530570373,
+      "grad_norm": 0.14065410196781158,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 48435
+    },
+    {
+      "epoch": 4.969019286007386,
+      "grad_norm": 0.1064314916729927,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 48438
+    },
+    {
+      "epoch": 4.969327041444399,
+      "grad_norm": 0.08995570242404938,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 48441
+    },
+    {
+      "epoch": 4.969634796881412,
+      "grad_norm": 0.05942520126700401,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 48444
+    },
+    {
+      "epoch": 4.969942552318424,
+      "grad_norm": 0.03557314723730087,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 48447
+    },
+    {
+      "epoch": 4.970250307755437,
+      "grad_norm": 0.058727700263261795,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 48450
+    },
+    {
+      "epoch": 4.97055806319245,
+      "grad_norm": 0.07610124349594116,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 48453
+    },
+    {
+      "epoch": 4.970865818629463,
+      "grad_norm": 0.062314316630363464,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 48456
+    },
+    {
+      "epoch": 4.9711735740664755,
+      "grad_norm": 0.060933034867048264,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 48459
+    },
+    {
+      "epoch": 4.971481329503488,
+      "grad_norm": 0.11259069293737411,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 48462
+    },
+    {
+      "epoch": 4.9717890849405,
+      "grad_norm": 0.039150118827819824,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 48465
+    },
+    {
+      "epoch": 4.972096840377513,
+      "grad_norm": 0.06994688510894775,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 48468
+    },
+    {
+      "epoch": 4.972404595814526,
+      "grad_norm": 0.08845566213130951,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 48471
+    },
+    {
+      "epoch": 4.9727123512515385,
+      "grad_norm": 0.0835181474685669,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 48474
+    },
+    {
+      "epoch": 4.973020106688551,
+      "grad_norm": 0.059876278042793274,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 48477
+    },
+    {
+      "epoch": 4.973327862125564,
+      "grad_norm": 0.08387959748506546,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 48480
+    },
+    {
+      "epoch": 4.973635617562577,
+      "grad_norm": 0.04472680762410164,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 48483
+    },
+    {
+      "epoch": 4.97394337299959,
+      "grad_norm": 0.04980779439210892,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 48486
+    },
+    {
+      "epoch": 4.974251128436602,
+      "grad_norm": 0.05865331366658211,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 48489
+    },
+    {
+      "epoch": 4.974558883873615,
+      "grad_norm": 0.07731412351131439,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 48492
+    },
+    {
+      "epoch": 4.974866639310628,
+      "grad_norm": 0.0759957805275917,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 48495
+    },
+    {
+      "epoch": 4.975174394747641,
+      "grad_norm": 0.06392304599285126,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 48498
+    },
+    {
+      "epoch": 4.975482150184654,
+      "grad_norm": 0.04608240723609924,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 48501
+    },
+    {
+      "epoch": 4.975789905621666,
+      "grad_norm": 0.04515860974788666,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 48504
+    },
+    {
+      "epoch": 4.976097661058679,
+      "grad_norm": 0.049364540725946426,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 48507
+    },
+    {
+      "epoch": 4.976405416495691,
+      "grad_norm": 0.05034893751144409,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 48510
+    },
+    {
+      "epoch": 4.976713171932704,
+      "grad_norm": 0.05965251848101616,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 48513
+    },
+    {
+      "epoch": 4.977020927369717,
+      "grad_norm": 0.03976500406861305,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 48516
+    },
+    {
+      "epoch": 4.977328682806729,
+      "grad_norm": 0.0873221829533577,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 48519
+    },
+    {
+      "epoch": 4.977636438243742,
+      "grad_norm": 0.10230670869350433,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 48522
+    },
+    {
+      "epoch": 4.977944193680755,
+      "grad_norm": 0.1398303061723709,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 48525
+    },
+    {
+      "epoch": 4.978251949117768,
+      "grad_norm": 0.047815751284360886,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 48528
+    },
+    {
+      "epoch": 4.9785597045547805,
+      "grad_norm": 0.05280955508351326,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 48531
+    },
+    {
+      "epoch": 4.978867459991793,
+      "grad_norm": 0.05164310708642006,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 48534
+    },
+    {
+      "epoch": 4.979175215428806,
+      "grad_norm": 0.04212404415011406,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 48537
+    },
+    {
+      "epoch": 4.979482970865819,
+      "grad_norm": 0.03191149979829788,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 48540
+    },
+    {
+      "epoch": 4.979790726302832,
+      "grad_norm": 0.09330854564905167,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 48543
+    },
+    {
+      "epoch": 4.980098481739844,
+      "grad_norm": 0.03602724149823189,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 48546
+    },
+    {
+      "epoch": 4.980406237176857,
+      "grad_norm": 0.1307518631219864,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 48549
+    },
+    {
+      "epoch": 4.98071399261387,
+      "grad_norm": 0.17725148797035217,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 48552
+    },
+    {
+      "epoch": 4.981021748050882,
+      "grad_norm": 0.13484884798526764,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 48555
+    },
+    {
+      "epoch": 4.981329503487895,
+      "grad_norm": 0.15105341374874115,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 48558
+    },
+    {
+      "epoch": 4.981637258924907,
+      "grad_norm": 0.07108210027217865,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 48561
+    },
+    {
+      "epoch": 4.98194501436192,
+      "grad_norm": 0.04044497013092041,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 48564
+    },
+    {
+      "epoch": 4.982252769798933,
+      "grad_norm": 0.03189781680703163,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 48567
+    },
+    {
+      "epoch": 4.982560525235946,
+      "grad_norm": 0.12914758920669556,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 48570
+    },
+    {
+      "epoch": 4.9828682806729585,
+      "grad_norm": 0.09454645961523056,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 48573
+    },
+    {
+      "epoch": 4.983176036109971,
+      "grad_norm": 0.09286819398403168,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 48576
+    },
+    {
+      "epoch": 4.983483791546984,
+      "grad_norm": 0.05915534123778343,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 48579
+    },
+    {
+      "epoch": 4.983791546983997,
+      "grad_norm": 0.09016136080026627,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 48582
+    },
+    {
+      "epoch": 4.98409930242101,
+      "grad_norm": 0.06749111413955688,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 48585
+    },
+    {
+      "epoch": 4.984407057858022,
+      "grad_norm": 0.05699833482503891,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 48588
+    },
+    {
+      "epoch": 4.984714813295035,
+      "grad_norm": 0.06643623113632202,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 48591
+    },
+    {
+      "epoch": 4.985022568732048,
+      "grad_norm": 0.05193551629781723,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 48594
+    },
+    {
+      "epoch": 4.985330324169061,
+      "grad_norm": 0.03996053710579872,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 48597
+    },
+    {
+      "epoch": 4.985638079606073,
+      "grad_norm": 0.09908989816904068,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 48600
+    },
+    {
+      "epoch": 4.985945835043085,
+      "grad_norm": 0.09618446230888367,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 48603
+    },
+    {
+      "epoch": 4.986253590480098,
+      "grad_norm": 0.04452410712838173,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 48606
+    },
+    {
+      "epoch": 4.986561345917111,
+      "grad_norm": 0.08481396734714508,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 48609
+    },
+    {
+      "epoch": 4.986869101354124,
+      "grad_norm": 0.07361941784620285,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 48612
+    },
+    {
+      "epoch": 4.9871768567911365,
+      "grad_norm": 0.04092632979154587,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 48615
+    },
+    {
+      "epoch": 4.987484612228149,
+      "grad_norm": 0.05964810773730278,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 48618
+    },
+    {
+      "epoch": 4.987792367665162,
+      "grad_norm": 0.04937126114964485,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 48621
+    },
+    {
+      "epoch": 4.988100123102175,
+      "grad_norm": 0.061311025172472,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 48624
+    },
+    {
+      "epoch": 4.988407878539188,
+      "grad_norm": 0.11813945323228836,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 48627
+    },
+    {
+      "epoch": 4.9887156339762,
+      "grad_norm": 0.13528530299663544,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 48630
+    },
+    {
+      "epoch": 4.989023389413213,
+      "grad_norm": 0.13620121777057648,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 48633
+    },
+    {
+      "epoch": 4.989331144850226,
+      "grad_norm": 0.050104204565286636,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 48636
+    },
+    {
+      "epoch": 4.989638900287239,
+      "grad_norm": 0.04701809585094452,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 48639
+    },
+    {
+      "epoch": 4.989946655724252,
+      "grad_norm": 0.05600928142666817,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 48642
+    },
+    {
+      "epoch": 4.9902544111612634,
+      "grad_norm": 0.03713737428188324,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 48645
+    },
+    {
+      "epoch": 4.990562166598276,
+      "grad_norm": 0.10586056858301163,
+      "learning_rate": 0.01,
+      "loss": 2.023,
+      "step": 48648
+    },
+    {
+      "epoch": 4.990869922035289,
+      "grad_norm": 0.07154802978038788,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 48651
+    },
+    {
+      "epoch": 4.991177677472302,
+      "grad_norm": 0.046645209193229675,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 48654
+    },
+    {
+      "epoch": 4.991485432909315,
+      "grad_norm": 0.10580138862133026,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 48657
+    },
+    {
+      "epoch": 4.991793188346327,
+      "grad_norm": 0.05583944171667099,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 48660
+    },
+    {
+      "epoch": 4.99210094378334,
+      "grad_norm": 0.04817153140902519,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 48663
+    },
+    {
+      "epoch": 4.992408699220353,
+      "grad_norm": 0.1129242405295372,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 48666
+    },
+    {
+      "epoch": 4.992716454657366,
+      "grad_norm": 0.09947514533996582,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 48669
+    },
+    {
+      "epoch": 4.9930242100943785,
+      "grad_norm": 0.07797490060329437,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 48672
+    },
+    {
+      "epoch": 4.993331965531391,
+      "grad_norm": 0.04402044415473938,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 48675
+    },
+    {
+      "epoch": 4.993639720968404,
+      "grad_norm": 0.05189018324017525,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 48678
+    },
+    {
+      "epoch": 4.993947476405417,
+      "grad_norm": 0.04376624524593353,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 48681
+    },
+    {
+      "epoch": 4.99425523184243,
+      "grad_norm": 0.06388211250305176,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 48684
+    },
+    {
+      "epoch": 4.994562987279442,
+      "grad_norm": 0.05990754812955856,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 48687
+    },
+    {
+      "epoch": 4.994870742716454,
+      "grad_norm": 0.06319117546081543,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 48690
+    },
+    {
+      "epoch": 4.995178498153467,
+      "grad_norm": 0.06176729500293732,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 48693
+    },
+    {
+      "epoch": 4.99548625359048,
+      "grad_norm": 0.06712348759174347,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 48696
+    },
+    {
+      "epoch": 4.995794009027493,
+      "grad_norm": 0.046728748828172684,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 48699
+    },
+    {
+      "epoch": 4.996101764464505,
+      "grad_norm": 0.04008499160408974,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 48702
+    },
+    {
+      "epoch": 4.996409519901518,
+      "grad_norm": 0.04931486025452614,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 48705
+    },
+    {
+      "epoch": 4.996717275338531,
+      "grad_norm": 0.09341800212860107,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 48708
+    },
+    {
+      "epoch": 4.997025030775544,
+      "grad_norm": 0.17475301027297974,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 48711
+    },
+    {
+      "epoch": 4.9973327862125565,
+      "grad_norm": 0.10119811445474625,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 48714
+    },
+    {
+      "epoch": 4.997640541649569,
+      "grad_norm": 0.04519075155258179,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 48717
+    },
+    {
+      "epoch": 4.997948297086582,
+      "grad_norm": 0.04394150897860527,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 48720
+    },
+    {
+      "epoch": 4.998256052523595,
+      "grad_norm": 0.03471103683114052,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 48723
+    },
+    {
+      "epoch": 4.998563807960608,
+      "grad_norm": 0.06422135978937149,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 48726
+    },
+    {
+      "epoch": 4.99887156339762,
+      "grad_norm": 0.10134468972682953,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 48729
+    },
+    {
+      "epoch": 4.999179318834633,
+      "grad_norm": 0.07451833784580231,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 48732
+    },
+    {
+      "epoch": 4.999487074271645,
+      "grad_norm": 0.12009834498167038,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 48735
+    },
+    {
+      "epoch": 4.999794829708658,
+      "grad_norm": 0.18911142647266388,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 48738
+    },
+    {
+      "epoch": 5.005751258087707,
+      "grad_norm": 0.11327698826789856,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 48741
+    },
+    {
+      "epoch": 5.006059361199548,
+      "grad_norm": 0.05276428908109665,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 48744
+    },
+    {
+      "epoch": 5.006367464311389,
+      "grad_norm": 0.04349682852625847,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 48747
+    },
+    {
+      "epoch": 5.006675567423231,
+      "grad_norm": 0.043818019330501556,
+      "learning_rate": 0.01,
+      "loss": 2.0334,
+      "step": 48750
+    },
+    {
+      "epoch": 5.006983670535073,
+      "grad_norm": 0.04884595423936844,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 48753
+    },
+    {
+      "epoch": 5.007291773646914,
+      "grad_norm": 0.04434294253587723,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 48756
+    },
+    {
+      "epoch": 5.007599876758755,
+      "grad_norm": 0.07619164884090424,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 48759
+    },
+    {
+      "epoch": 5.0079079798705965,
+      "grad_norm": 0.04731893911957741,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 48762
+    },
+    {
+      "epoch": 5.008216082982438,
+      "grad_norm": 0.03700774163007736,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 48765
+    },
+    {
+      "epoch": 5.00852418609428,
+      "grad_norm": 0.03112631104886532,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 48768
+    },
+    {
+      "epoch": 5.008832289206121,
+      "grad_norm": 0.12492241710424423,
+      "learning_rate": 0.01,
+      "loss": 2.0255,
+      "step": 48771
+    },
+    {
+      "epoch": 5.0091403923179625,
+      "grad_norm": 0.03778549283742905,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 48774
+    },
+    {
+      "epoch": 5.009448495429804,
+      "grad_norm": 0.06593509018421173,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 48777
+    },
+    {
+      "epoch": 5.009756598541645,
+      "grad_norm": 0.04298333451151848,
+      "learning_rate": 0.01,
+      "loss": 2.03,
+      "step": 48780
+    },
+    {
+      "epoch": 5.010064701653486,
+      "grad_norm": 0.04566636681556702,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 48783
+    },
+    {
+      "epoch": 5.010372804765328,
+      "grad_norm": 0.045355070382356644,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 48786
+    },
+    {
+      "epoch": 5.01068090787717,
+      "grad_norm": 0.10494118183851242,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 48789
+    },
+    {
+      "epoch": 5.010989010989011,
+      "grad_norm": 0.07352183014154434,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 48792
+    },
+    {
+      "epoch": 5.011297114100852,
+      "grad_norm": 0.07270955294370651,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 48795
+    },
+    {
+      "epoch": 5.0116052172126935,
+      "grad_norm": 0.04416406527161598,
+      "learning_rate": 0.01,
+      "loss": 2.0346,
+      "step": 48798
+    },
+    {
+      "epoch": 5.011913320324536,
+      "grad_norm": 0.08084416389465332,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 48801
+    },
+    {
+      "epoch": 5.012221423436377,
+      "grad_norm": 0.04440615326166153,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 48804
+    },
+    {
+      "epoch": 5.012529526548218,
+      "grad_norm": 0.0649610310792923,
+      "learning_rate": 0.01,
+      "loss": 2.0398,
+      "step": 48807
+    },
+    {
+      "epoch": 5.012837629660059,
+      "grad_norm": 0.08533710986375809,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 48810
+    },
+    {
+      "epoch": 5.013145732771901,
+      "grad_norm": 0.06310707330703735,
+      "learning_rate": 0.01,
+      "loss": 2.0566,
+      "step": 48813
+    },
+    {
+      "epoch": 5.013453835883743,
+      "grad_norm": 0.07096098363399506,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 48816
+    },
+    {
+      "epoch": 5.013761938995584,
+      "grad_norm": 0.08358955383300781,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 48819
+    },
+    {
+      "epoch": 5.014070042107425,
+      "grad_norm": 0.07774472236633301,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 48822
+    },
+    {
+      "epoch": 5.014378145219267,
+      "grad_norm": 0.090924933552742,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 48825
+    },
+    {
+      "epoch": 5.014686248331108,
+      "grad_norm": 0.08830771595239639,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 48828
+    },
+    {
+      "epoch": 5.014994351442949,
+      "grad_norm": 0.09647586941719055,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 48831
+    },
+    {
+      "epoch": 5.015302454554791,
+      "grad_norm": 0.09413495659828186,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 48834
+    },
+    {
+      "epoch": 5.0156105576666326,
+      "grad_norm": 0.046874430030584335,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 48837
+    },
+    {
+      "epoch": 5.015918660778474,
+      "grad_norm": 0.050309523940086365,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 48840
+    },
+    {
+      "epoch": 5.016226763890315,
+      "grad_norm": 0.045271504670381546,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 48843
+    },
+    {
+      "epoch": 5.016534867002156,
+      "grad_norm": 0.03297847509384155,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 48846
+    },
+    {
+      "epoch": 5.0168429701139985,
+      "grad_norm": 0.08260242640972137,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 48849
+    },
+    {
+      "epoch": 5.01715107322584,
+      "grad_norm": 0.04784693941473961,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 48852
+    },
+    {
+      "epoch": 5.017459176337681,
+      "grad_norm": 0.07972519099712372,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 48855
+    },
+    {
+      "epoch": 5.017767279449522,
+      "grad_norm": 0.09882769733667374,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 48858
+    },
+    {
+      "epoch": 5.0180753825613635,
+      "grad_norm": 0.059475596994161606,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 48861
+    },
+    {
+      "epoch": 5.018383485673206,
+      "grad_norm": 0.03795355185866356,
+      "learning_rate": 0.01,
+      "loss": 2.0319,
+      "step": 48864
+    },
+    {
+      "epoch": 5.018691588785047,
+      "grad_norm": 0.04603972285985947,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 48867
+    },
+    {
+      "epoch": 5.018999691896888,
+      "grad_norm": 0.0678548589348793,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 48870
+    },
+    {
+      "epoch": 5.0193077950087295,
+      "grad_norm": 0.0728834941983223,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 48873
+    },
+    {
+      "epoch": 5.019615898120571,
+      "grad_norm": 0.08255651593208313,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 48876
+    },
+    {
+      "epoch": 5.019924001232413,
+      "grad_norm": 0.08427122980356216,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 48879
+    },
+    {
+      "epoch": 5.020232104344254,
+      "grad_norm": 0.06279109418392181,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 48882
+    },
+    {
+      "epoch": 5.020540207456095,
+      "grad_norm": 0.12473344802856445,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 48885
+    },
+    {
+      "epoch": 5.020848310567937,
+      "grad_norm": 0.05066891387104988,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 48888
+    },
+    {
+      "epoch": 5.021156413679778,
+      "grad_norm": 0.038826942443847656,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 48891
+    },
+    {
+      "epoch": 5.021464516791619,
+      "grad_norm": 0.058017902076244354,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 48894
+    },
+    {
+      "epoch": 5.021772619903461,
+      "grad_norm": 0.09492091834545135,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 48897
+    },
+    {
+      "epoch": 5.022080723015303,
+      "grad_norm": 0.037246495485305786,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 48900
+    },
+    {
+      "epoch": 5.022388826127144,
+      "grad_norm": 0.06708169728517532,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 48903
+    },
+    {
+      "epoch": 5.022696929238985,
+      "grad_norm": 0.07195274531841278,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 48906
+    },
+    {
+      "epoch": 5.023005032350826,
+      "grad_norm": 0.04685629904270172,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 48909
+    },
+    {
+      "epoch": 5.0233131354626686,
+      "grad_norm": 0.08163397759199142,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 48912
+    },
+    {
+      "epoch": 5.02362123857451,
+      "grad_norm": 0.05403751879930496,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 48915
+    },
+    {
+      "epoch": 5.023929341686351,
+      "grad_norm": 0.038569968193769455,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 48918
+    },
+    {
+      "epoch": 5.024237444798192,
+      "grad_norm": 0.055535174906253815,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 48921
+    },
+    {
+      "epoch": 5.024545547910034,
+      "grad_norm": 0.05628294125199318,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 48924
+    },
+    {
+      "epoch": 5.024853651021876,
+      "grad_norm": 0.06412038207054138,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 48927
+    },
+    {
+      "epoch": 5.025161754133717,
+      "grad_norm": 0.09966779500246048,
+      "learning_rate": 0.01,
+      "loss": 2.0417,
+      "step": 48930
+    },
+    {
+      "epoch": 5.025469857245558,
+      "grad_norm": 0.059476420283317566,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 48933
+    },
+    {
+      "epoch": 5.0257779603573995,
+      "grad_norm": 0.04914550483226776,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 48936
+    },
+    {
+      "epoch": 5.026086063469241,
+      "grad_norm": 0.03358187898993492,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 48939
+    },
+    {
+      "epoch": 5.026394166581082,
+      "grad_norm": 0.0671076700091362,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 48942
+    },
+    {
+      "epoch": 5.026702269692924,
+      "grad_norm": 0.14911647140979767,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 48945
+    },
+    {
+      "epoch": 5.0270103728047655,
+      "grad_norm": 0.05344945192337036,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 48948
+    },
+    {
+      "epoch": 5.027318475916607,
+      "grad_norm": 0.05113920569419861,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 48951
+    },
+    {
+      "epoch": 5.027626579028448,
+      "grad_norm": 0.052601687610149384,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 48954
+    },
+    {
+      "epoch": 5.027934682140289,
+      "grad_norm": 0.04715648666024208,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 48957
+    },
+    {
+      "epoch": 5.028242785252131,
+      "grad_norm": 0.0414542555809021,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 48960
+    },
+    {
+      "epoch": 5.028550888363973,
+      "grad_norm": 0.057792216539382935,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 48963
+    },
+    {
+      "epoch": 5.028858991475814,
+      "grad_norm": 0.12371774017810822,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 48966
+    },
+    {
+      "epoch": 5.029167094587655,
+      "grad_norm": 0.053585126996040344,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 48969
+    },
+    {
+      "epoch": 5.0294751976994965,
+      "grad_norm": 0.051625724881887436,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 48972
+    },
+    {
+      "epoch": 5.029783300811339,
+      "grad_norm": 0.03650549426674843,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 48975
+    },
+    {
+      "epoch": 5.03009140392318,
+      "grad_norm": 0.039793796837329865,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 48978
+    },
+    {
+      "epoch": 5.030399507035021,
+      "grad_norm": 0.03223228082060814,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 48981
+    },
+    {
+      "epoch": 5.030707610146862,
+      "grad_norm": 0.06604496389627457,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 48984
+    },
+    {
+      "epoch": 5.031015713258704,
+      "grad_norm": 0.13230498135089874,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 48987
+    },
+    {
+      "epoch": 5.031323816370545,
+      "grad_norm": 0.07130693644285202,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 48990
+    },
+    {
+      "epoch": 5.031631919482387,
+      "grad_norm": 0.13743533194065094,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 48993
+    },
+    {
+      "epoch": 5.031940022594228,
+      "grad_norm": 0.08371279388666153,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 48996
+    },
+    {
+      "epoch": 5.03224812570607,
+      "grad_norm": 0.06169065833091736,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 48999
+    },
+    {
+      "epoch": 5.032556228817911,
+      "grad_norm": 0.05723833665251732,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 49002
+    },
+    {
+      "epoch": 5.032864331929752,
+      "grad_norm": 0.043473754078149796,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 49005
+    },
+    {
+      "epoch": 5.033172435041594,
+      "grad_norm": 0.08324360102415085,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 49008
+    },
+    {
+      "epoch": 5.0334805381534355,
+      "grad_norm": 0.04197626933455467,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 49011
+    },
+    {
+      "epoch": 5.033788641265277,
+      "grad_norm": 0.07882298529148102,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 49014
+    },
+    {
+      "epoch": 5.034096744377118,
+      "grad_norm": 0.08073476701974869,
+      "learning_rate": 0.01,
+      "loss": 2.0272,
+      "step": 49017
+    },
+    {
+      "epoch": 5.034404847488959,
+      "grad_norm": 0.0678163543343544,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 49020
+    },
+    {
+      "epoch": 5.0347129506008015,
+      "grad_norm": 0.04524783417582512,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 49023
+    },
+    {
+      "epoch": 5.035021053712643,
+      "grad_norm": 0.06805024296045303,
+      "learning_rate": 0.01,
+      "loss": 2.0321,
+      "step": 49026
+    },
+    {
+      "epoch": 5.035329156824484,
+      "grad_norm": 0.12917374074459076,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 49029
+    },
+    {
+      "epoch": 5.035637259936325,
+      "grad_norm": 0.04843614995479584,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 49032
+    },
+    {
+      "epoch": 5.0359453630481665,
+      "grad_norm": 0.09850560128688812,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 49035
+    },
+    {
+      "epoch": 5.036253466160008,
+      "grad_norm": 0.04039955511689186,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 49038
+    },
+    {
+      "epoch": 5.03656156927185,
+      "grad_norm": 0.06758705526590347,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 49041
+    },
+    {
+      "epoch": 5.036869672383691,
+      "grad_norm": 0.03571697697043419,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 49044
+    },
+    {
+      "epoch": 5.0371777754955325,
+      "grad_norm": 0.034528639167547226,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 49047
+    },
+    {
+      "epoch": 5.037485878607374,
+      "grad_norm": 0.05735669657588005,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 49050
+    },
+    {
+      "epoch": 5.037793981719215,
+      "grad_norm": 0.11180947721004486,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 49053
+    },
+    {
+      "epoch": 5.038102084831057,
+      "grad_norm": 0.08823433518409729,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 49056
+    },
+    {
+      "epoch": 5.038410187942898,
+      "grad_norm": 0.06959319114685059,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 49059
+    },
+    {
+      "epoch": 5.03871829105474,
+      "grad_norm": 0.03767579421401024,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 49062
+    },
+    {
+      "epoch": 5.039026394166581,
+      "grad_norm": 0.032121919095516205,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 49065
+    },
+    {
+      "epoch": 5.039334497278422,
+      "grad_norm": 0.09328833967447281,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 49068
+    },
+    {
+      "epoch": 5.039642600390264,
+      "grad_norm": 0.05128837004303932,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 49071
+    },
+    {
+      "epoch": 5.039950703502106,
+      "grad_norm": 0.04546624422073364,
+      "learning_rate": 0.01,
+      "loss": 2.0395,
+      "step": 49074
+    },
+    {
+      "epoch": 5.040258806613947,
+      "grad_norm": 0.03980513662099838,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 49077
+    },
+    {
+      "epoch": 5.040566909725788,
+      "grad_norm": 0.04204264655709267,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 49080
+    },
+    {
+      "epoch": 5.040875012837629,
+      "grad_norm": 0.04164162278175354,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 49083
+    },
+    {
+      "epoch": 5.041183115949471,
+      "grad_norm": 0.0698886588215828,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 49086
+    },
+    {
+      "epoch": 5.041491219061313,
+      "grad_norm": 0.11362603306770325,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 49089
+    },
+    {
+      "epoch": 5.041799322173154,
+      "grad_norm": 0.08820103108882904,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 49092
+    },
+    {
+      "epoch": 5.042107425284995,
+      "grad_norm": 0.06023460626602173,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 49095
+    },
+    {
+      "epoch": 5.042415528396837,
+      "grad_norm": 0.03523271903395653,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 49098
+    },
+    {
+      "epoch": 5.042723631508678,
+      "grad_norm": 0.062407344579696655,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 49101
+    },
+    {
+      "epoch": 5.04303173462052,
+      "grad_norm": 0.04747510328888893,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 49104
+    },
+    {
+      "epoch": 5.043339837732361,
+      "grad_norm": 0.04824723303318024,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 49107
+    },
+    {
+      "epoch": 5.0436479408442025,
+      "grad_norm": 0.1672361046075821,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 49110
+    },
+    {
+      "epoch": 5.043956043956044,
+      "grad_norm": 0.09675043821334839,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 49113
+    },
+    {
+      "epoch": 5.044264147067885,
+      "grad_norm": 0.06804441660642624,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 49116
+    },
+    {
+      "epoch": 5.044572250179727,
+      "grad_norm": 0.05074039101600647,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 49119
+    },
+    {
+      "epoch": 5.0448803532915685,
+      "grad_norm": 0.049977827817201614,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 49122
+    },
+    {
+      "epoch": 5.04518845640341,
+      "grad_norm": 0.05549508333206177,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 49125
+    },
+    {
+      "epoch": 5.045496559515251,
+      "grad_norm": 0.03386814519762993,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 49128
+    },
+    {
+      "epoch": 5.045804662627092,
+      "grad_norm": 0.04851256310939789,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 49131
+    },
+    {
+      "epoch": 5.0461127657389335,
+      "grad_norm": 0.0848245769739151,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 49134
+    },
+    {
+      "epoch": 5.046420868850776,
+      "grad_norm": 0.03692524507641792,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 49137
+    },
+    {
+      "epoch": 5.046728971962617,
+      "grad_norm": 0.0406675823032856,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 49140
+    },
+    {
+      "epoch": 5.047037075074458,
+      "grad_norm": 0.07141982764005661,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 49143
+    },
+    {
+      "epoch": 5.0473451781862995,
+      "grad_norm": 0.0453253835439682,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 49146
+    },
+    {
+      "epoch": 5.047653281298141,
+      "grad_norm": 0.04967833310365677,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 49149
+    },
+    {
+      "epoch": 5.047961384409983,
+      "grad_norm": 0.09721168875694275,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 49152
+    },
+    {
+      "epoch": 5.048269487521824,
+      "grad_norm": 0.03806902468204498,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 49155
+    },
+    {
+      "epoch": 5.048577590633665,
+      "grad_norm": 0.048705510795116425,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 49158
+    },
+    {
+      "epoch": 5.048885693745507,
+      "grad_norm": 0.04676016420125961,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 49161
+    },
+    {
+      "epoch": 5.049193796857348,
+      "grad_norm": 0.1444026082754135,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 49164
+    },
+    {
+      "epoch": 5.04950189996919,
+      "grad_norm": 0.07510826736688614,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 49167
+    },
+    {
+      "epoch": 5.049810003081031,
+      "grad_norm": 0.05828002095222473,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 49170
+    },
+    {
+      "epoch": 5.050118106192873,
+      "grad_norm": 0.04933730885386467,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 49173
+    },
+    {
+      "epoch": 5.050426209304714,
+      "grad_norm": 0.07392150163650513,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 49176
+    },
+    {
+      "epoch": 5.050734312416555,
+      "grad_norm": 0.03829526528716087,
+      "learning_rate": 0.01,
+      "loss": 2.0377,
+      "step": 49179
+    },
+    {
+      "epoch": 5.051042415528396,
+      "grad_norm": 0.06644531339406967,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 49182
+    },
+    {
+      "epoch": 5.0513505186402385,
+      "grad_norm": 0.045286018401384354,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 49185
+    },
+    {
+      "epoch": 5.05165862175208,
+      "grad_norm": 0.1067107766866684,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 49188
+    },
+    {
+      "epoch": 5.051966724863921,
+      "grad_norm": 0.09251461923122406,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 49191
+    },
+    {
+      "epoch": 5.052274827975762,
+      "grad_norm": 0.051101312041282654,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 49194
+    },
+    {
+      "epoch": 5.052582931087604,
+      "grad_norm": 0.060298655182123184,
+      "learning_rate": 0.01,
+      "loss": 2.0393,
+      "step": 49197
+    },
+    {
+      "epoch": 5.052891034199446,
+      "grad_norm": 0.11112777143716812,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 49200
+    },
+    {
+      "epoch": 5.053199137311287,
+      "grad_norm": 0.05832698941230774,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 49203
+    },
+    {
+      "epoch": 5.053507240423128,
+      "grad_norm": 0.04670676961541176,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 49206
+    },
+    {
+      "epoch": 5.0538153435349695,
+      "grad_norm": 0.03849385678768158,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 49209
+    },
+    {
+      "epoch": 5.054123446646811,
+      "grad_norm": 0.043539416044950485,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 49212
+    },
+    {
+      "epoch": 5.054431549758653,
+      "grad_norm": 0.04201329126954079,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 49215
+    },
+    {
+      "epoch": 5.054739652870494,
+      "grad_norm": 0.046697720885276794,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 49218
+    },
+    {
+      "epoch": 5.0550477559823355,
+      "grad_norm": 0.13588139414787292,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 49221
+    },
+    {
+      "epoch": 5.055355859094177,
+      "grad_norm": 0.06000100448727608,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 49224
+    },
+    {
+      "epoch": 5.055663962206018,
+      "grad_norm": 0.05746195465326309,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 49227
+    },
+    {
+      "epoch": 5.05597206531786,
+      "grad_norm": 0.05890364944934845,
+      "learning_rate": 0.01,
+      "loss": 2.0246,
+      "step": 49230
+    },
+    {
+      "epoch": 5.056280168429701,
+      "grad_norm": 0.05617443472146988,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 49233
+    },
+    {
+      "epoch": 5.056588271541543,
+      "grad_norm": 0.035529427230358124,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 49236
+    },
+    {
+      "epoch": 5.056896374653384,
+      "grad_norm": 0.08901556581258774,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 49239
+    },
+    {
+      "epoch": 5.057204477765225,
+      "grad_norm": 0.07534974068403244,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 49242
+    },
+    {
+      "epoch": 5.0575125808770665,
+      "grad_norm": 0.09517022967338562,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 49245
+    },
+    {
+      "epoch": 5.057820683988909,
+      "grad_norm": 0.0671476274728775,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 49248
+    },
+    {
+      "epoch": 5.05812878710075,
+      "grad_norm": 0.08913878351449966,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 49251
+    },
+    {
+      "epoch": 5.058436890212591,
+      "grad_norm": 0.06179509684443474,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 49254
+    },
+    {
+      "epoch": 5.058744993324432,
+      "grad_norm": 0.055043041706085205,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 49257
+    },
+    {
+      "epoch": 5.059053096436274,
+      "grad_norm": 0.05528896301984787,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 49260
+    },
+    {
+      "epoch": 5.059361199548116,
+      "grad_norm": 0.1153421550989151,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 49263
+    },
+    {
+      "epoch": 5.059669302659957,
+      "grad_norm": 0.11637650430202484,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 49266
+    },
+    {
+      "epoch": 5.059977405771798,
+      "grad_norm": 0.04912794381380081,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 49269
+    },
+    {
+      "epoch": 5.06028550888364,
+      "grad_norm": 0.030755288898944855,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 49272
+    },
+    {
+      "epoch": 5.060593611995481,
+      "grad_norm": 0.03253510594367981,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 49275
+    },
+    {
+      "epoch": 5.060901715107323,
+      "grad_norm": 0.0353284515440464,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 49278
+    },
+    {
+      "epoch": 5.061209818219164,
+      "grad_norm": 0.04621535912156105,
+      "learning_rate": 0.01,
+      "loss": 2.0276,
+      "step": 49281
+    },
+    {
+      "epoch": 5.0615179213310055,
+      "grad_norm": 0.08098949491977692,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 49284
+    },
+    {
+      "epoch": 5.061826024442847,
+      "grad_norm": 0.07662132382392883,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 49287
+    },
+    {
+      "epoch": 5.062134127554688,
+      "grad_norm": 0.06088561937212944,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 49290
+    },
+    {
+      "epoch": 5.062442230666529,
+      "grad_norm": 0.09895024448633194,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 49293
+    },
+    {
+      "epoch": 5.0627503337783715,
+      "grad_norm": 0.05712589621543884,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 49296
+    },
+    {
+      "epoch": 5.063058436890213,
+      "grad_norm": 0.03906116634607315,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 49299
+    },
+    {
+      "epoch": 5.063366540002054,
+      "grad_norm": 0.04972919449210167,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 49302
+    },
+    {
+      "epoch": 5.063674643113895,
+      "grad_norm": 0.048737164586782455,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 49305
+    },
+    {
+      "epoch": 5.0639827462257365,
+      "grad_norm": 0.06541066616773605,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 49308
+    },
+    {
+      "epoch": 5.064290849337579,
+      "grad_norm": 0.048164594918489456,
+      "learning_rate": 0.01,
+      "loss": 2.0352,
+      "step": 49311
+    },
+    {
+      "epoch": 5.06459895244942,
+      "grad_norm": 0.09063941240310669,
+      "learning_rate": 0.01,
+      "loss": 2.0265,
+      "step": 49314
+    },
+    {
+      "epoch": 5.064907055561261,
+      "grad_norm": 0.04573111608624458,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 49317
+    },
+    {
+      "epoch": 5.0652151586731025,
+      "grad_norm": 0.033861320465803146,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 49320
+    },
+    {
+      "epoch": 5.065523261784944,
+      "grad_norm": 0.03708826005458832,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 49323
+    },
+    {
+      "epoch": 5.065831364896786,
+      "grad_norm": 0.10216791927814484,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 49326
+    },
+    {
+      "epoch": 5.066139468008627,
+      "grad_norm": 0.06112508475780487,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 49329
+    },
+    {
+      "epoch": 5.066447571120468,
+      "grad_norm": 0.06171619892120361,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 49332
+    },
+    {
+      "epoch": 5.06675567423231,
+      "grad_norm": 0.03746636584401131,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 49335
+    },
+    {
+      "epoch": 5.067063777344151,
+      "grad_norm": 0.06251370906829834,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 49338
+    },
+    {
+      "epoch": 5.067371880455992,
+      "grad_norm": 0.03557705506682396,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 49341
+    },
+    {
+      "epoch": 5.067679983567834,
+      "grad_norm": 0.036395229399204254,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 49344
+    },
+    {
+      "epoch": 5.067988086679676,
+      "grad_norm": 0.12709848582744598,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 49347
+    },
+    {
+      "epoch": 5.068296189791517,
+      "grad_norm": 0.05967814847826958,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 49350
+    },
+    {
+      "epoch": 5.068604292903358,
+      "grad_norm": 0.09794972836971283,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 49353
+    },
+    {
+      "epoch": 5.068912396015199,
+      "grad_norm": 0.06263583153486252,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 49356
+    },
+    {
+      "epoch": 5.0692204991270415,
+      "grad_norm": 0.08471337705850601,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 49359
+    },
+    {
+      "epoch": 5.069528602238883,
+      "grad_norm": 0.06739859282970428,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 49362
+    },
+    {
+      "epoch": 5.069836705350724,
+      "grad_norm": 0.09328118711709976,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 49365
+    },
+    {
+      "epoch": 5.070144808462565,
+      "grad_norm": 0.09012078493833542,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 49368
+    },
+    {
+      "epoch": 5.070452911574407,
+      "grad_norm": 0.03964925929903984,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 49371
+    },
+    {
+      "epoch": 5.070761014686249,
+      "grad_norm": 0.12138742208480835,
+      "learning_rate": 0.01,
+      "loss": 2.0361,
+      "step": 49374
+    },
+    {
+      "epoch": 5.07106911779809,
+      "grad_norm": 0.04422127828001976,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 49377
+    },
+    {
+      "epoch": 5.071377220909931,
+      "grad_norm": 0.038298092782497406,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 49380
+    },
+    {
+      "epoch": 5.0716853240217725,
+      "grad_norm": 0.04425305128097534,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 49383
+    },
+    {
+      "epoch": 5.071993427133614,
+      "grad_norm": 0.08276382088661194,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 49386
+    },
+    {
+      "epoch": 5.072301530245455,
+      "grad_norm": 0.06378073245286942,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 49389
+    },
+    {
+      "epoch": 5.072609633357297,
+      "grad_norm": 0.12052054703235626,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 49392
+    },
+    {
+      "epoch": 5.0729177364691385,
+      "grad_norm": 0.04634268954396248,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 49395
+    },
+    {
+      "epoch": 5.07322583958098,
+      "grad_norm": 0.055755455046892166,
+      "learning_rate": 0.01,
+      "loss": 2.0214,
+      "step": 49398
+    },
+    {
+      "epoch": 5.073533942692821,
+      "grad_norm": 0.04300512373447418,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 49401
+    },
+    {
+      "epoch": 5.073842045804662,
+      "grad_norm": 0.08284097909927368,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 49404
+    },
+    {
+      "epoch": 5.074150148916504,
+      "grad_norm": 0.046609655022621155,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 49407
+    },
+    {
+      "epoch": 5.074458252028346,
+      "grad_norm": 0.06387645751237869,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 49410
+    },
+    {
+      "epoch": 5.074766355140187,
+      "grad_norm": 0.0506816990673542,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 49413
+    },
+    {
+      "epoch": 5.075074458252028,
+      "grad_norm": 0.03865443170070648,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 49416
+    },
+    {
+      "epoch": 5.0753825613638694,
+      "grad_norm": 0.07702656835317612,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 49419
+    },
+    {
+      "epoch": 5.075690664475712,
+      "grad_norm": 0.12181714177131653,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 49422
+    },
+    {
+      "epoch": 5.075998767587553,
+      "grad_norm": 0.06351039558649063,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 49425
+    },
+    {
+      "epoch": 5.076306870699394,
+      "grad_norm": 0.09312722831964493,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 49428
+    },
+    {
+      "epoch": 5.076614973811235,
+      "grad_norm": 0.06879006326198578,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 49431
+    },
+    {
+      "epoch": 5.076923076923077,
+      "grad_norm": 0.06043083220720291,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 49434
+    },
+    {
+      "epoch": 5.077231180034918,
+      "grad_norm": 0.049164239317178726,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 49437
+    },
+    {
+      "epoch": 5.07753928314676,
+      "grad_norm": 0.04470280185341835,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 49440
+    },
+    {
+      "epoch": 5.077847386258601,
+      "grad_norm": 0.04513731226325035,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 49443
+    },
+    {
+      "epoch": 5.078155489370443,
+      "grad_norm": 0.07849404215812683,
+      "learning_rate": 0.01,
+      "loss": 2.0244,
+      "step": 49446
+    },
+    {
+      "epoch": 5.078463592482284,
+      "grad_norm": 0.04768620431423187,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 49449
+    },
+    {
+      "epoch": 5.078771695594125,
+      "grad_norm": 0.09680715203285217,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 49452
+    },
+    {
+      "epoch": 5.079079798705967,
+      "grad_norm": 0.07119087129831314,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 49455
+    },
+    {
+      "epoch": 5.0793879018178085,
+      "grad_norm": 0.11560992896556854,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 49458
+    },
+    {
+      "epoch": 5.07969600492965,
+      "grad_norm": 0.039443932473659515,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 49461
+    },
+    {
+      "epoch": 5.080004108041491,
+      "grad_norm": 0.040348734706640244,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 49464
+    },
+    {
+      "epoch": 5.080312211153332,
+      "grad_norm": 0.10496348142623901,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 49467
+    },
+    {
+      "epoch": 5.0806203142651745,
+      "grad_norm": 0.10493961721658707,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 49470
+    },
+    {
+      "epoch": 5.080928417377016,
+      "grad_norm": 0.08104293048381805,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 49473
+    },
+    {
+      "epoch": 5.081236520488857,
+      "grad_norm": 0.07609682530164719,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 49476
+    },
+    {
+      "epoch": 5.081544623600698,
+      "grad_norm": 0.03504324331879616,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 49479
+    },
+    {
+      "epoch": 5.0818527267125395,
+      "grad_norm": 0.03438768908381462,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 49482
+    },
+    {
+      "epoch": 5.082160829824382,
+      "grad_norm": 0.0372585728764534,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 49485
+    },
+    {
+      "epoch": 5.082468932936223,
+      "grad_norm": 0.06997973471879959,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 49488
+    },
+    {
+      "epoch": 5.082777036048064,
+      "grad_norm": 0.07300589978694916,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 49491
+    },
+    {
+      "epoch": 5.0830851391599055,
+      "grad_norm": 0.053376756608486176,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 49494
+    },
+    {
+      "epoch": 5.083393242271747,
+      "grad_norm": 0.060256477445364,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 49497
+    },
+    {
+      "epoch": 5.083701345383588,
+      "grad_norm": 0.0293456818908453,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 49500
+    },
+    {
+      "epoch": 5.08400944849543,
+      "grad_norm": 0.08368721604347229,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 49503
+    },
+    {
+      "epoch": 5.084317551607271,
+      "grad_norm": 0.15514978766441345,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 49506
+    },
+    {
+      "epoch": 5.084625654719113,
+      "grad_norm": 0.07931728661060333,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 49509
+    },
+    {
+      "epoch": 5.084933757830954,
+      "grad_norm": 0.038475409150123596,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 49512
+    },
+    {
+      "epoch": 5.085241860942795,
+      "grad_norm": 0.03167513385415077,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 49515
+    },
+    {
+      "epoch": 5.085549964054637,
+      "grad_norm": 0.053547751158475876,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 49518
+    },
+    {
+      "epoch": 5.085858067166479,
+      "grad_norm": 0.06962644308805466,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 49521
+    },
+    {
+      "epoch": 5.08616617027832,
+      "grad_norm": 0.06724515557289124,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 49524
+    },
+    {
+      "epoch": 5.086474273390161,
+      "grad_norm": 0.06945919245481491,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 49527
+    },
+    {
+      "epoch": 5.086782376502002,
+      "grad_norm": 0.03792745992541313,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 49530
+    },
+    {
+      "epoch": 5.0870904796138445,
+      "grad_norm": 0.05056298151612282,
+      "learning_rate": 0.01,
+      "loss": 2.0414,
+      "step": 49533
+    },
+    {
+      "epoch": 5.087398582725686,
+      "grad_norm": 0.07531926780939102,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 49536
+    },
+    {
+      "epoch": 5.087706685837527,
+      "grad_norm": 0.0834566205739975,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 49539
+    },
+    {
+      "epoch": 5.088014788949368,
+      "grad_norm": 0.038478609174489975,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 49542
+    },
+    {
+      "epoch": 5.08832289206121,
+      "grad_norm": 0.06002604216337204,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 49545
+    },
+    {
+      "epoch": 5.088630995173051,
+      "grad_norm": 0.046000853180885315,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 49548
+    },
+    {
+      "epoch": 5.088939098284893,
+      "grad_norm": 0.06248699501156807,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 49551
+    },
+    {
+      "epoch": 5.089247201396734,
+      "grad_norm": 0.10761575400829315,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 49554
+    },
+    {
+      "epoch": 5.0895553045085755,
+      "grad_norm": 0.09329602122306824,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 49557
+    },
+    {
+      "epoch": 5.089863407620417,
+      "grad_norm": 0.08233852684497833,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 49560
+    },
+    {
+      "epoch": 5.090171510732258,
+      "grad_norm": 0.06591898202896118,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 49563
+    },
+    {
+      "epoch": 5.0904796138441,
+      "grad_norm": 0.11147011071443558,
+      "learning_rate": 0.01,
+      "loss": 2.0415,
+      "step": 49566
+    },
+    {
+      "epoch": 5.0907877169559415,
+      "grad_norm": 0.10501214861869812,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 49569
+    },
+    {
+      "epoch": 5.091095820067783,
+      "grad_norm": 0.04929099977016449,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 49572
+    },
+    {
+      "epoch": 5.091403923179624,
+      "grad_norm": 0.09390784054994583,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 49575
+    },
+    {
+      "epoch": 5.091712026291465,
+      "grad_norm": 0.08599641174077988,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 49578
+    },
+    {
+      "epoch": 5.092020129403307,
+      "grad_norm": 0.04285844415426254,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 49581
+    },
+    {
+      "epoch": 5.092328232515149,
+      "grad_norm": 0.034273531287908554,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 49584
+    },
+    {
+      "epoch": 5.09263633562699,
+      "grad_norm": 0.04617328196763992,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 49587
+    },
+    {
+      "epoch": 5.092944438738831,
+      "grad_norm": 0.08739766478538513,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 49590
+    },
+    {
+      "epoch": 5.093252541850672,
+      "grad_norm": 0.06263922899961472,
+      "learning_rate": 0.01,
+      "loss": 2.0396,
+      "step": 49593
+    },
+    {
+      "epoch": 5.093560644962514,
+      "grad_norm": 0.06044527515769005,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 49596
+    },
+    {
+      "epoch": 5.093868748074356,
+      "grad_norm": 0.046116676181554794,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 49599
+    },
+    {
+      "epoch": 5.094176851186197,
+      "grad_norm": 0.1035202220082283,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 49602
+    },
+    {
+      "epoch": 5.094484954298038,
+      "grad_norm": 0.03816954419016838,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 49605
+    },
+    {
+      "epoch": 5.09479305740988,
+      "grad_norm": 0.04604346677660942,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 49608
+    },
+    {
+      "epoch": 5.095101160521721,
+      "grad_norm": 0.05297078937292099,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 49611
+    },
+    {
+      "epoch": 5.095409263633563,
+      "grad_norm": 0.03836045414209366,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 49614
+    },
+    {
+      "epoch": 5.095717366745404,
+      "grad_norm": 0.047086507081985474,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 49617
+    },
+    {
+      "epoch": 5.096025469857246,
+      "grad_norm": 0.04657857492566109,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 49620
+    },
+    {
+      "epoch": 5.096333572969087,
+      "grad_norm": 0.08411470055580139,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 49623
+    },
+    {
+      "epoch": 5.096641676080928,
+      "grad_norm": 0.16819079220294952,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 49626
+    },
+    {
+      "epoch": 5.09694977919277,
+      "grad_norm": 0.04950143024325371,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 49629
+    },
+    {
+      "epoch": 5.0972578823046115,
+      "grad_norm": 0.039262961596250534,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 49632
+    },
+    {
+      "epoch": 5.097565985416453,
+      "grad_norm": 0.03746853023767471,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 49635
+    },
+    {
+      "epoch": 5.097874088528294,
+      "grad_norm": 0.052741505205631256,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 49638
+    },
+    {
+      "epoch": 5.098182191640135,
+      "grad_norm": 0.1192871630191803,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 49641
+    },
+    {
+      "epoch": 5.098490294751977,
+      "grad_norm": 0.054899461567401886,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 49644
+    },
+    {
+      "epoch": 5.098798397863819,
+      "grad_norm": 0.05850033089518547,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 49647
+    },
+    {
+      "epoch": 5.09910650097566,
+      "grad_norm": 0.08758574724197388,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 49650
+    },
+    {
+      "epoch": 5.099414604087501,
+      "grad_norm": 0.09481152147054672,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 49653
+    },
+    {
+      "epoch": 5.0997227071993425,
+      "grad_norm": 0.12264327704906464,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 49656
+    },
+    {
+      "epoch": 5.100030810311184,
+      "grad_norm": 0.06704699993133545,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 49659
+    },
+    {
+      "epoch": 5.100338913423026,
+      "grad_norm": 0.10752804577350616,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 49662
+    },
+    {
+      "epoch": 5.100647016534867,
+      "grad_norm": 0.09217878431081772,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 49665
+    },
+    {
+      "epoch": 5.1009551196467084,
+      "grad_norm": 0.04479275271296501,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 49668
+    },
+    {
+      "epoch": 5.10126322275855,
+      "grad_norm": 0.06177330017089844,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 49671
+    },
+    {
+      "epoch": 5.101571325870391,
+      "grad_norm": 0.09911325573921204,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 49674
+    },
+    {
+      "epoch": 5.101879428982233,
+      "grad_norm": 0.06496170163154602,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 49677
+    },
+    {
+      "epoch": 5.102187532094074,
+      "grad_norm": 0.04296518489718437,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 49680
+    },
+    {
+      "epoch": 5.102495635205916,
+      "grad_norm": 0.05492605268955231,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 49683
+    },
+    {
+      "epoch": 5.102803738317757,
+      "grad_norm": 0.033720582723617554,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 49686
+    },
+    {
+      "epoch": 5.103111841429598,
+      "grad_norm": 0.10172949731349945,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 49689
+    },
+    {
+      "epoch": 5.103419944541439,
+      "grad_norm": 0.05326984077692032,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 49692
+    },
+    {
+      "epoch": 5.103728047653282,
+      "grad_norm": 0.10597484558820724,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 49695
+    },
+    {
+      "epoch": 5.104036150765123,
+      "grad_norm": 0.03910377249121666,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 49698
+    },
+    {
+      "epoch": 5.104344253876964,
+      "grad_norm": 0.04232342913746834,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 49701
+    },
+    {
+      "epoch": 5.104652356988805,
+      "grad_norm": 0.11343611031770706,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 49704
+    },
+    {
+      "epoch": 5.104960460100647,
+      "grad_norm": 0.06593618541955948,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 49707
+    },
+    {
+      "epoch": 5.105268563212489,
+      "grad_norm": 0.07576625049114227,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 49710
+    },
+    {
+      "epoch": 5.10557666632433,
+      "grad_norm": 0.10935617238283157,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 49713
+    },
+    {
+      "epoch": 5.105884769436171,
+      "grad_norm": 0.04176800325512886,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 49716
+    },
+    {
+      "epoch": 5.106192872548013,
+      "grad_norm": 0.05533302202820778,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 49719
+    },
+    {
+      "epoch": 5.106500975659854,
+      "grad_norm": 0.05446555092930794,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 49722
+    },
+    {
+      "epoch": 5.106809078771696,
+      "grad_norm": 0.05760158970952034,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 49725
+    },
+    {
+      "epoch": 5.107117181883537,
+      "grad_norm": 0.11345206201076508,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 49728
+    },
+    {
+      "epoch": 5.1074252849953785,
+      "grad_norm": 0.09869284927845001,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 49731
+    },
+    {
+      "epoch": 5.10773338810722,
+      "grad_norm": 0.050909820944070816,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 49734
+    },
+    {
+      "epoch": 5.108041491219061,
+      "grad_norm": 0.03889784961938858,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 49737
+    },
+    {
+      "epoch": 5.108349594330903,
+      "grad_norm": 0.04064812883734703,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 49740
+    },
+    {
+      "epoch": 5.1086576974427444,
+      "grad_norm": 0.13409452140331268,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 49743
+    },
+    {
+      "epoch": 5.108965800554586,
+      "grad_norm": 0.0955972746014595,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 49746
+    },
+    {
+      "epoch": 5.109273903666427,
+      "grad_norm": 0.11944996565580368,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 49749
+    },
+    {
+      "epoch": 5.109582006778268,
+      "grad_norm": 0.06599771976470947,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 49752
+    },
+    {
+      "epoch": 5.1098901098901095,
+      "grad_norm": 0.05203530192375183,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 49755
+    },
+    {
+      "epoch": 5.110198213001952,
+      "grad_norm": 0.052099455147981644,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 49758
+    },
+    {
+      "epoch": 5.110506316113793,
+      "grad_norm": 0.06005854532122612,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 49761
+    },
+    {
+      "epoch": 5.110814419225634,
+      "grad_norm": 0.0429551899433136,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 49764
+    },
+    {
+      "epoch": 5.111122522337475,
+      "grad_norm": 0.07173555344343185,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 49767
+    },
+    {
+      "epoch": 5.111430625449317,
+      "grad_norm": 0.05251854285597801,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 49770
+    },
+    {
+      "epoch": 5.111738728561159,
+      "grad_norm": 0.06687440723180771,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 49773
+    },
+    {
+      "epoch": 5.112046831673,
+      "grad_norm": 0.0841129869222641,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 49776
+    },
+    {
+      "epoch": 5.112354934784841,
+      "grad_norm": 0.086825892329216,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 49779
+    },
+    {
+      "epoch": 5.112663037896683,
+      "grad_norm": 0.059246089309453964,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 49782
+    },
+    {
+      "epoch": 5.112971141008524,
+      "grad_norm": 0.061836544424295425,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 49785
+    },
+    {
+      "epoch": 5.113279244120366,
+      "grad_norm": 0.0878891870379448,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 49788
+    },
+    {
+      "epoch": 5.113587347232207,
+      "grad_norm": 0.061963435262441635,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 49791
+    },
+    {
+      "epoch": 5.113895450344049,
+      "grad_norm": 0.07002797722816467,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 49794
+    },
+    {
+      "epoch": 5.11420355345589,
+      "grad_norm": 0.07307753711938858,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 49797
+    },
+    {
+      "epoch": 5.114511656567731,
+      "grad_norm": 0.05582696199417114,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 49800
+    },
+    {
+      "epoch": 5.114819759679572,
+      "grad_norm": 0.11906569451093674,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 49803
+    },
+    {
+      "epoch": 5.1151278627914145,
+      "grad_norm": 0.041671089828014374,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 49806
+    },
+    {
+      "epoch": 5.115435965903256,
+      "grad_norm": 0.040786996483802795,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 49809
+    },
+    {
+      "epoch": 5.115744069015097,
+      "grad_norm": 0.06962720304727554,
+      "learning_rate": 0.01,
+      "loss": 2.0245,
+      "step": 49812
+    },
+    {
+      "epoch": 5.116052172126938,
+      "grad_norm": 0.05079250782728195,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 49815
+    },
+    {
+      "epoch": 5.11636027523878,
+      "grad_norm": 0.03422404080629349,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 49818
+    },
+    {
+      "epoch": 5.116668378350622,
+      "grad_norm": 0.13034658133983612,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 49821
+    },
+    {
+      "epoch": 5.116976481462463,
+      "grad_norm": 0.05432206392288208,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 49824
+    },
+    {
+      "epoch": 5.117284584574304,
+      "grad_norm": 0.11153416335582733,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 49827
+    },
+    {
+      "epoch": 5.1175926876861455,
+      "grad_norm": 0.04324459657073021,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 49830
+    },
+    {
+      "epoch": 5.117900790797987,
+      "grad_norm": 0.035336676985025406,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 49833
+    },
+    {
+      "epoch": 5.118208893909829,
+      "grad_norm": 0.07241083681583405,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 49836
+    },
+    {
+      "epoch": 5.11851699702167,
+      "grad_norm": 0.05331496521830559,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 49839
+    },
+    {
+      "epoch": 5.118825100133511,
+      "grad_norm": 0.04856008291244507,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 49842
+    },
+    {
+      "epoch": 5.119133203245353,
+      "grad_norm": 0.07900402694940567,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 49845
+    },
+    {
+      "epoch": 5.119441306357194,
+      "grad_norm": 0.04300111532211304,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 49848
+    },
+    {
+      "epoch": 5.119749409469035,
+      "grad_norm": 0.09242533892393112,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 49851
+    },
+    {
+      "epoch": 5.120057512580877,
+      "grad_norm": 0.047889694571495056,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 49854
+    },
+    {
+      "epoch": 5.120365615692719,
+      "grad_norm": 0.04471924528479576,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 49857
+    },
+    {
+      "epoch": 5.12067371880456,
+      "grad_norm": 0.11666672676801682,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 49860
+    },
+    {
+      "epoch": 5.120981821916401,
+      "grad_norm": 0.0470040999352932,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 49863
+    },
+    {
+      "epoch": 5.121289925028242,
+      "grad_norm": 0.10009215027093887,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 49866
+    },
+    {
+      "epoch": 5.121598028140085,
+      "grad_norm": 0.06280893087387085,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 49869
+    },
+    {
+      "epoch": 5.121906131251926,
+      "grad_norm": 0.046029094606637955,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 49872
+    },
+    {
+      "epoch": 5.122214234363767,
+      "grad_norm": 0.11030067503452301,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 49875
+    },
+    {
+      "epoch": 5.122522337475608,
+      "grad_norm": 0.033092547208070755,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 49878
+    },
+    {
+      "epoch": 5.12283044058745,
+      "grad_norm": 0.051649268716573715,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 49881
+    },
+    {
+      "epoch": 5.123138543699292,
+      "grad_norm": 0.06344709545373917,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 49884
+    },
+    {
+      "epoch": 5.123446646811133,
+      "grad_norm": 0.08509237319231033,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 49887
+    },
+    {
+      "epoch": 5.123754749922974,
+      "grad_norm": 0.04539303854107857,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 49890
+    },
+    {
+      "epoch": 5.124062853034816,
+      "grad_norm": 0.03646918386220932,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 49893
+    },
+    {
+      "epoch": 5.124370956146657,
+      "grad_norm": 0.12570123374462128,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 49896
+    },
+    {
+      "epoch": 5.124679059258498,
+      "grad_norm": 0.12240875512361526,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 49899
+    },
+    {
+      "epoch": 5.12498716237034,
+      "grad_norm": 0.10540018230676651,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 49902
+    },
+    {
+      "epoch": 5.1252952654821815,
+      "grad_norm": 0.08514107763767242,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 49905
+    },
+    {
+      "epoch": 5.125603368594023,
+      "grad_norm": 0.04462766274809837,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 49908
+    },
+    {
+      "epoch": 5.125911471705864,
+      "grad_norm": 0.03537189960479736,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 49911
+    },
+    {
+      "epoch": 5.126219574817705,
+      "grad_norm": 0.053130924701690674,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 49914
+    },
+    {
+      "epoch": 5.126527677929547,
+      "grad_norm": 0.08888891339302063,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 49917
+    },
+    {
+      "epoch": 5.126835781041389,
+      "grad_norm": 0.051719069480895996,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 49920
+    },
+    {
+      "epoch": 5.12714388415323,
+      "grad_norm": 0.123109832406044,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 49923
+    },
+    {
+      "epoch": 5.127451987265071,
+      "grad_norm": 0.08993472158908844,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 49926
+    },
+    {
+      "epoch": 5.1277600903769125,
+      "grad_norm": 0.10910573601722717,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 49929
+    },
+    {
+      "epoch": 5.128068193488755,
+      "grad_norm": 0.13778774440288544,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 49932
+    },
+    {
+      "epoch": 5.128376296600596,
+      "grad_norm": 0.08411096781492233,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 49935
+    },
+    {
+      "epoch": 5.128684399712437,
+      "grad_norm": 0.05669613182544708,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 49938
+    },
+    {
+      "epoch": 5.128992502824278,
+      "grad_norm": 0.037695713341236115,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 49941
+    },
+    {
+      "epoch": 5.12930060593612,
+      "grad_norm": 0.07977671176195145,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 49944
+    },
+    {
+      "epoch": 5.129608709047961,
+      "grad_norm": 0.10012632608413696,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 49947
+    },
+    {
+      "epoch": 5.129916812159803,
+      "grad_norm": 0.06524799764156342,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 49950
+    },
+    {
+      "epoch": 5.130224915271644,
+      "grad_norm": 0.1130509302020073,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 49953
+    },
+    {
+      "epoch": 5.130533018383486,
+      "grad_norm": 0.06062379479408264,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 49956
+    },
+    {
+      "epoch": 5.130841121495327,
+      "grad_norm": 0.06229391694068909,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 49959
+    },
+    {
+      "epoch": 5.131149224607168,
+      "grad_norm": 0.04881103336811066,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 49962
+    },
+    {
+      "epoch": 5.13145732771901,
+      "grad_norm": 0.0779888853430748,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 49965
+    },
+    {
+      "epoch": 5.131765430830852,
+      "grad_norm": 0.03707825019955635,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 49968
+    },
+    {
+      "epoch": 5.132073533942693,
+      "grad_norm": 0.12145381420850754,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 49971
+    },
+    {
+      "epoch": 5.132381637054534,
+      "grad_norm": 0.08990494906902313,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 49974
+    },
+    {
+      "epoch": 5.132689740166375,
+      "grad_norm": 0.09638572484254837,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 49977
+    },
+    {
+      "epoch": 5.1329978432782175,
+      "grad_norm": 0.037770826369524,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 49980
+    },
+    {
+      "epoch": 5.133305946390059,
+      "grad_norm": 0.038396887481212616,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 49983
+    },
+    {
+      "epoch": 5.1336140495019,
+      "grad_norm": 0.052246298640966415,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 49986
+    },
+    {
+      "epoch": 5.133922152613741,
+      "grad_norm": 0.043142352253198624,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 49989
+    },
+    {
+      "epoch": 5.134230255725583,
+      "grad_norm": 0.04713413119316101,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 49992
+    },
+    {
+      "epoch": 5.134538358837425,
+      "grad_norm": 0.06687591224908829,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 49995
+    },
+    {
+      "epoch": 5.134846461949266,
+      "grad_norm": 0.057987745851278305,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 49998
+    },
+    {
+      "epoch": 5.135154565061107,
+      "grad_norm": 0.048778288066387177,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 50001
+    },
+    {
+      "epoch": 5.1354626681729485,
+      "grad_norm": 0.04489746689796448,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 50004
+    },
+    {
+      "epoch": 5.13577077128479,
+      "grad_norm": 0.08324793726205826,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 50007
+    },
+    {
+      "epoch": 5.136078874396631,
+      "grad_norm": 0.04505028948187828,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 50010
+    },
+    {
+      "epoch": 5.136386977508473,
+      "grad_norm": 0.07827684283256531,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 50013
+    },
+    {
+      "epoch": 5.136695080620314,
+      "grad_norm": 0.07187260687351227,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 50016
+    },
+    {
+      "epoch": 5.137003183732156,
+      "grad_norm": 0.0374215804040432,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 50019
+    },
+    {
+      "epoch": 5.137311286843997,
+      "grad_norm": 0.06169632822275162,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 50022
+    },
+    {
+      "epoch": 5.137619389955838,
+      "grad_norm": 0.08089695870876312,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 50025
+    },
+    {
+      "epoch": 5.13792749306768,
+      "grad_norm": 0.06311798840761185,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 50028
+    },
+    {
+      "epoch": 5.138235596179522,
+      "grad_norm": 0.0869307816028595,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 50031
+    },
+    {
+      "epoch": 5.138543699291363,
+      "grad_norm": 0.07645547389984131,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 50034
+    },
+    {
+      "epoch": 5.138851802403204,
+      "grad_norm": 0.09578394144773483,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 50037
+    },
+    {
+      "epoch": 5.139159905515045,
+      "grad_norm": 0.09977228194475174,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 50040
+    },
+    {
+      "epoch": 5.139468008626887,
+      "grad_norm": 0.05820827558636665,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 50043
+    },
+    {
+      "epoch": 5.139776111738729,
+      "grad_norm": 0.03890974819660187,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 50046
+    },
+    {
+      "epoch": 5.14008421485057,
+      "grad_norm": 0.10406100749969482,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 50049
+    },
+    {
+      "epoch": 5.140392317962411,
+      "grad_norm": 0.057227544486522675,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 50052
+    },
+    {
+      "epoch": 5.140700421074253,
+      "grad_norm": 0.07142063230276108,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 50055
+    },
+    {
+      "epoch": 5.141008524186094,
+      "grad_norm": 0.047450605779886246,
+      "learning_rate": 0.01,
+      "loss": 2.0203,
+      "step": 50058
+    },
+    {
+      "epoch": 5.141316627297936,
+      "grad_norm": 0.07935485243797302,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 50061
+    },
+    {
+      "epoch": 5.141624730409777,
+      "grad_norm": 0.14235766232013702,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 50064
+    },
+    {
+      "epoch": 5.141932833521619,
+      "grad_norm": 0.054345373064279556,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 50067
+    },
+    {
+      "epoch": 5.14224093663346,
+      "grad_norm": 0.03830769658088684,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 50070
+    },
+    {
+      "epoch": 5.142549039745301,
+      "grad_norm": 0.034983497112989426,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 50073
+    },
+    {
+      "epoch": 5.142857142857143,
+      "grad_norm": 0.05580296739935875,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 50076
+    },
+    {
+      "epoch": 5.1431652459689845,
+      "grad_norm": 0.11556407809257507,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 50079
+    },
+    {
+      "epoch": 5.143473349080826,
+      "grad_norm": 0.05813178792595863,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 50082
+    },
+    {
+      "epoch": 5.143781452192667,
+      "grad_norm": 0.08391161262989044,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 50085
+    },
+    {
+      "epoch": 5.144089555304508,
+      "grad_norm": 0.1391163021326065,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 50088
+    },
+    {
+      "epoch": 5.14439765841635,
+      "grad_norm": 0.038378529250621796,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 50091
+    },
+    {
+      "epoch": 5.144705761528192,
+      "grad_norm": 0.03499170020222664,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 50094
+    },
+    {
+      "epoch": 5.145013864640033,
+      "grad_norm": 0.038385018706321716,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 50097
+    },
+    {
+      "epoch": 5.145321967751874,
+      "grad_norm": 0.04664738476276398,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 50100
+    },
+    {
+      "epoch": 5.1456300708637155,
+      "grad_norm": 0.05988583341240883,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 50103
+    },
+    {
+      "epoch": 5.145938173975557,
+      "grad_norm": 0.05822984501719475,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 50106
+    },
+    {
+      "epoch": 5.146246277087399,
+      "grad_norm": 0.1176968440413475,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 50109
+    },
+    {
+      "epoch": 5.14655438019924,
+      "grad_norm": 0.04409262165427208,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 50112
+    },
+    {
+      "epoch": 5.146862483311081,
+      "grad_norm": 0.18404054641723633,
+      "learning_rate": 0.01,
+      "loss": 2.0388,
+      "step": 50115
+    },
+    {
+      "epoch": 5.147170586422923,
+      "grad_norm": 0.07370147854089737,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 50118
+    },
+    {
+      "epoch": 5.147478689534764,
+      "grad_norm": 0.05898935720324516,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 50121
+    },
+    {
+      "epoch": 5.147786792646606,
+      "grad_norm": 0.04993924871087074,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 50124
+    },
+    {
+      "epoch": 5.148094895758447,
+      "grad_norm": 0.07191821932792664,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 50127
+    },
+    {
+      "epoch": 5.148402998870289,
+      "grad_norm": 0.06233205646276474,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 50130
+    },
+    {
+      "epoch": 5.14871110198213,
+      "grad_norm": 0.10599711537361145,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 50133
+    },
+    {
+      "epoch": 5.149019205093971,
+      "grad_norm": 0.05540407449007034,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 50136
+    },
+    {
+      "epoch": 5.149327308205813,
+      "grad_norm": 0.0512741319835186,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 50139
+    },
+    {
+      "epoch": 5.149635411317655,
+      "grad_norm": 0.060490820556879044,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 50142
+    },
+    {
+      "epoch": 5.149943514429496,
+      "grad_norm": 0.03850322216749191,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 50145
+    },
+    {
+      "epoch": 5.150251617541337,
+      "grad_norm": 0.12944872677326202,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 50148
+    },
+    {
+      "epoch": 5.150559720653178,
+      "grad_norm": 0.1282481700181961,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 50151
+    },
+    {
+      "epoch": 5.15086782376502,
+      "grad_norm": 0.09446661174297333,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 50154
+    },
+    {
+      "epoch": 5.151175926876862,
+      "grad_norm": 0.04041890799999237,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 50157
+    },
+    {
+      "epoch": 5.151484029988703,
+      "grad_norm": 0.056447647511959076,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 50160
+    },
+    {
+      "epoch": 5.151792133100544,
+      "grad_norm": 0.0449320413172245,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 50163
+    },
+    {
+      "epoch": 5.1521002362123856,
+      "grad_norm": 0.07914337515830994,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 50166
+    },
+    {
+      "epoch": 5.152408339324227,
+      "grad_norm": 0.06510215252637863,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 50169
+    },
+    {
+      "epoch": 5.152716442436069,
+      "grad_norm": 0.04413073882460594,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 50172
+    },
+    {
+      "epoch": 5.15302454554791,
+      "grad_norm": 0.045415036380290985,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 50175
+    },
+    {
+      "epoch": 5.1533326486597515,
+      "grad_norm": 0.04759620502591133,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 50178
+    },
+    {
+      "epoch": 5.153640751771593,
+      "grad_norm": 0.08537039160728455,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 50181
+    },
+    {
+      "epoch": 5.153948854883434,
+      "grad_norm": 0.10971151292324066,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 50184
+    },
+    {
+      "epoch": 5.154256957995276,
+      "grad_norm": 0.04773986339569092,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 50187
+    },
+    {
+      "epoch": 5.154565061107117,
+      "grad_norm": 0.13717158138751984,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 50190
+    },
+    {
+      "epoch": 5.154873164218959,
+      "grad_norm": 0.05396975204348564,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 50193
+    },
+    {
+      "epoch": 5.1551812673308,
+      "grad_norm": 0.03611796349287033,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 50196
+    },
+    {
+      "epoch": 5.155489370442641,
+      "grad_norm": 0.0544486828148365,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 50199
+    },
+    {
+      "epoch": 5.1557974735544825,
+      "grad_norm": 0.1198575496673584,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 50202
+    },
+    {
+      "epoch": 5.156105576666325,
+      "grad_norm": 0.04922753944993019,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 50205
+    },
+    {
+      "epoch": 5.156413679778166,
+      "grad_norm": 0.07132868468761444,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 50208
+    },
+    {
+      "epoch": 5.156721782890007,
+      "grad_norm": 0.06232021376490593,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 50211
+    },
+    {
+      "epoch": 5.157029886001848,
+      "grad_norm": 0.0492694191634655,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 50214
+    },
+    {
+      "epoch": 5.15733798911369,
+      "grad_norm": 0.040781404823064804,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 50217
+    },
+    {
+      "epoch": 5.157646092225532,
+      "grad_norm": 0.04086026921868324,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 50220
+    },
+    {
+      "epoch": 5.157954195337373,
+      "grad_norm": 0.04154505580663681,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 50223
+    },
+    {
+      "epoch": 5.158262298449214,
+      "grad_norm": 0.043042492121458054,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 50226
+    },
+    {
+      "epoch": 5.158570401561056,
+      "grad_norm": 0.10211813449859619,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 50229
+    },
+    {
+      "epoch": 5.158878504672897,
+      "grad_norm": 0.0644875019788742,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 50232
+    },
+    {
+      "epoch": 5.159186607784739,
+      "grad_norm": 0.09346074610948563,
+      "learning_rate": 0.01,
+      "loss": 2.0304,
+      "step": 50235
+    },
+    {
+      "epoch": 5.15949471089658,
+      "grad_norm": 0.04961559548974037,
+      "learning_rate": 0.01,
+      "loss": 2.0267,
+      "step": 50238
+    },
+    {
+      "epoch": 5.159802814008422,
+      "grad_norm": 0.07135666906833649,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 50241
+    },
+    {
+      "epoch": 5.160110917120263,
+      "grad_norm": 0.0363953560590744,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 50244
+    },
+    {
+      "epoch": 5.160419020232104,
+      "grad_norm": 0.05364762246608734,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 50247
+    },
+    {
+      "epoch": 5.160727123343945,
+      "grad_norm": 0.07405871897935867,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 50250
+    },
+    {
+      "epoch": 5.1610352264557875,
+      "grad_norm": 0.06045752391219139,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 50253
+    },
+    {
+      "epoch": 5.161343329567629,
+      "grad_norm": 0.05497819930315018,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 50256
+    },
+    {
+      "epoch": 5.16165143267947,
+      "grad_norm": 0.12490664422512054,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 50259
+    },
+    {
+      "epoch": 5.161959535791311,
+      "grad_norm": 0.04628274589776993,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 50262
+    },
+    {
+      "epoch": 5.1622676389031525,
+      "grad_norm": 0.06884883344173431,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 50265
+    },
+    {
+      "epoch": 5.162575742014995,
+      "grad_norm": 0.057228926569223404,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 50268
+    },
+    {
+      "epoch": 5.162883845126836,
+      "grad_norm": 0.05467236414551735,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 50271
+    },
+    {
+      "epoch": 5.163191948238677,
+      "grad_norm": 0.10743111371994019,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 50274
+    },
+    {
+      "epoch": 5.1635000513505185,
+      "grad_norm": 0.06426141411066055,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 50277
+    },
+    {
+      "epoch": 5.16380815446236,
+      "grad_norm": 0.09047043323516846,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 50280
+    },
+    {
+      "epoch": 5.164116257574202,
+      "grad_norm": 0.032664790749549866,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 50283
+    },
+    {
+      "epoch": 5.164424360686043,
+      "grad_norm": 0.05022372305393219,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 50286
+    },
+    {
+      "epoch": 5.164732463797884,
+      "grad_norm": 0.06359392404556274,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 50289
+    },
+    {
+      "epoch": 5.165040566909726,
+      "grad_norm": 0.0480986088514328,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 50292
+    },
+    {
+      "epoch": 5.165348670021567,
+      "grad_norm": 0.04164107143878937,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 50295
+    },
+    {
+      "epoch": 5.165656773133408,
+      "grad_norm": 0.06567779183387756,
+      "learning_rate": 0.01,
+      "loss": 2.0243,
+      "step": 50298
+    },
+    {
+      "epoch": 5.16596487624525,
+      "grad_norm": 0.11087372153997421,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 50301
+    },
+    {
+      "epoch": 5.166272979357092,
+      "grad_norm": 0.056734826415777206,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 50304
+    },
+    {
+      "epoch": 5.166581082468933,
+      "grad_norm": 0.06245841085910797,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 50307
+    },
+    {
+      "epoch": 5.166889185580774,
+      "grad_norm": 0.07361923158168793,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 50310
+    },
+    {
+      "epoch": 5.167197288692615,
+      "grad_norm": 0.08085532486438751,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 50313
+    },
+    {
+      "epoch": 5.167505391804458,
+      "grad_norm": 0.08500105887651443,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 50316
+    },
+    {
+      "epoch": 5.167813494916299,
+      "grad_norm": 0.07736580818891525,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 50319
+    },
+    {
+      "epoch": 5.16812159802814,
+      "grad_norm": 0.08762416243553162,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 50322
+    },
+    {
+      "epoch": 5.168429701139981,
+      "grad_norm": 0.05399163439869881,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 50325
+    },
+    {
+      "epoch": 5.168737804251823,
+      "grad_norm": 0.05431235954165459,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 50328
+    },
+    {
+      "epoch": 5.169045907363665,
+      "grad_norm": 0.04496511071920395,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 50331
+    },
+    {
+      "epoch": 5.169354010475506,
+      "grad_norm": 0.04711058363318443,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 50334
+    },
+    {
+      "epoch": 5.169662113587347,
+      "grad_norm": 0.08989371359348297,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 50337
+    },
+    {
+      "epoch": 5.1699702166991885,
+      "grad_norm": 0.048945628106594086,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 50340
+    },
+    {
+      "epoch": 5.17027831981103,
+      "grad_norm": 0.09270982444286346,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 50343
+    },
+    {
+      "epoch": 5.170586422922872,
+      "grad_norm": 0.05704176053404808,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 50346
+    },
+    {
+      "epoch": 5.170894526034713,
+      "grad_norm": 0.03455890342593193,
+      "learning_rate": 0.01,
+      "loss": 2.0412,
+      "step": 50349
+    },
+    {
+      "epoch": 5.1712026291465545,
+      "grad_norm": 0.041682351380586624,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 50352
+    },
+    {
+      "epoch": 5.171510732258396,
+      "grad_norm": 0.04951677471399307,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 50355
+    },
+    {
+      "epoch": 5.171818835370237,
+      "grad_norm": 0.056010644882917404,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 50358
+    },
+    {
+      "epoch": 5.172126938482078,
+      "grad_norm": 0.1273152083158493,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 50361
+    },
+    {
+      "epoch": 5.17243504159392,
+      "grad_norm": 0.08742735534906387,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 50364
+    },
+    {
+      "epoch": 5.172743144705762,
+      "grad_norm": 0.03796623274683952,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 50367
+    },
+    {
+      "epoch": 5.173051247817603,
+      "grad_norm": 0.035885389894247055,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 50370
+    },
+    {
+      "epoch": 5.173359350929444,
+      "grad_norm": 0.040873028337955475,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 50373
+    },
+    {
+      "epoch": 5.1736674540412855,
+      "grad_norm": 0.03645748645067215,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 50376
+    },
+    {
+      "epoch": 5.173975557153128,
+      "grad_norm": 0.08969543129205704,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 50379
+    },
+    {
+      "epoch": 5.174283660264969,
+      "grad_norm": 0.05043506622314453,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 50382
+    },
+    {
+      "epoch": 5.17459176337681,
+      "grad_norm": 0.053546857088804245,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 50385
+    },
+    {
+      "epoch": 5.174899866488651,
+      "grad_norm": 0.03222036734223366,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 50388
+    },
+    {
+      "epoch": 5.175207969600493,
+      "grad_norm": 0.07949094474315643,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 50391
+    },
+    {
+      "epoch": 5.175516072712334,
+      "grad_norm": 0.0858977660536766,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 50394
+    },
+    {
+      "epoch": 5.175824175824176,
+      "grad_norm": 0.10471560060977936,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 50397
+    },
+    {
+      "epoch": 5.176132278936017,
+      "grad_norm": 0.10742645710706711,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 50400
+    },
+    {
+      "epoch": 5.176440382047859,
+      "grad_norm": 0.035415954887866974,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 50403
+    },
+    {
+      "epoch": 5.1767484851597,
+      "grad_norm": 0.043476980179548264,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 50406
+    },
+    {
+      "epoch": 5.177056588271541,
+      "grad_norm": 0.05659857392311096,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 50409
+    },
+    {
+      "epoch": 5.177364691383383,
+      "grad_norm": 0.09809272736310959,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 50412
+    },
+    {
+      "epoch": 5.1776727944952246,
+      "grad_norm": 0.0645933598279953,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 50415
+    },
+    {
+      "epoch": 5.177980897607066,
+      "grad_norm": 0.11137600988149643,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 50418
+    },
+    {
+      "epoch": 5.178289000718907,
+      "grad_norm": 0.03614633530378342,
+      "learning_rate": 0.01,
+      "loss": 2.0087,
+      "step": 50421
+    },
+    {
+      "epoch": 5.178597103830748,
+      "grad_norm": 0.09580224752426147,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 50424
+    },
+    {
+      "epoch": 5.1789052069425905,
+      "grad_norm": 0.0874679684638977,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 50427
+    },
+    {
+      "epoch": 5.179213310054432,
+      "grad_norm": 0.06637660413980484,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 50430
+    },
+    {
+      "epoch": 5.179521413166273,
+      "grad_norm": 0.049818407744169235,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 50433
+    },
+    {
+      "epoch": 5.179829516278114,
+      "grad_norm": 0.035875897854566574,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 50436
+    },
+    {
+      "epoch": 5.1801376193899555,
+      "grad_norm": 0.12956075370311737,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 50439
+    },
+    {
+      "epoch": 5.180445722501798,
+      "grad_norm": 0.0568457767367363,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 50442
+    },
+    {
+      "epoch": 5.180753825613639,
+      "grad_norm": 0.06275864690542221,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 50445
+    },
+    {
+      "epoch": 5.18106192872548,
+      "grad_norm": 0.07609914243221283,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 50448
+    },
+    {
+      "epoch": 5.1813700318373215,
+      "grad_norm": 0.0721612498164177,
+      "learning_rate": 0.01,
+      "loss": 2.0329,
+      "step": 50451
+    },
+    {
+      "epoch": 5.181678134949163,
+      "grad_norm": 0.08733054995536804,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 50454
+    },
+    {
+      "epoch": 5.181986238061004,
+      "grad_norm": 0.10601712018251419,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 50457
+    },
+    {
+      "epoch": 5.182294341172846,
+      "grad_norm": 0.17671173810958862,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 50460
+    },
+    {
+      "epoch": 5.182602444284687,
+      "grad_norm": 0.08600035309791565,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 50463
+    },
+    {
+      "epoch": 5.182910547396529,
+      "grad_norm": 0.05488591641187668,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 50466
+    },
+    {
+      "epoch": 5.18321865050837,
+      "grad_norm": 0.08030954003334045,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 50469
+    },
+    {
+      "epoch": 5.183526753620211,
+      "grad_norm": 0.05973450466990471,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 50472
+    },
+    {
+      "epoch": 5.183834856732053,
+      "grad_norm": 0.07898557931184769,
+      "learning_rate": 0.01,
+      "loss": 2.0174,
+      "step": 50475
+    },
+    {
+      "epoch": 5.184142959843895,
+      "grad_norm": 0.10037455707788467,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 50478
+    },
+    {
+      "epoch": 5.184451062955736,
+      "grad_norm": 0.08369861543178558,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 50481
+    },
+    {
+      "epoch": 5.184759166067577,
+      "grad_norm": 0.06350437551736832,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 50484
+    },
+    {
+      "epoch": 5.185067269179418,
+      "grad_norm": 0.05332079157233238,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 50487
+    },
+    {
+      "epoch": 5.1853753722912606,
+      "grad_norm": 0.09419017285108566,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 50490
+    },
+    {
+      "epoch": 5.185683475403102,
+      "grad_norm": 0.05123918130993843,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 50493
+    },
+    {
+      "epoch": 5.185991578514943,
+      "grad_norm": 0.08905530720949173,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 50496
+    },
+    {
+      "epoch": 5.186299681626784,
+      "grad_norm": 0.04132084175944328,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 50499
+    },
+    {
+      "epoch": 5.186607784738626,
+      "grad_norm": 0.04010608792304993,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 50502
+    },
+    {
+      "epoch": 5.186915887850467,
+      "grad_norm": 0.035268135368824005,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 50505
+    },
+    {
+      "epoch": 5.187223990962309,
+      "grad_norm": 0.07214079797267914,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 50508
+    },
+    {
+      "epoch": 5.18753209407415,
+      "grad_norm": 0.0641101822257042,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 50511
+    },
+    {
+      "epoch": 5.1878401971859915,
+      "grad_norm": 0.05329526588320732,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 50514
+    },
+    {
+      "epoch": 5.188148300297833,
+      "grad_norm": 0.10318466275930405,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 50517
+    },
+    {
+      "epoch": 5.188456403409674,
+      "grad_norm": 0.04416116699576378,
+      "learning_rate": 0.01,
+      "loss": 2.0333,
+      "step": 50520
+    },
+    {
+      "epoch": 5.188764506521516,
+      "grad_norm": 0.03857633098959923,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 50523
+    },
+    {
+      "epoch": 5.1890726096333575,
+      "grad_norm": 0.054197683930397034,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 50526
+    },
+    {
+      "epoch": 5.189380712745199,
+      "grad_norm": 0.045404430478811264,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 50529
+    },
+    {
+      "epoch": 5.18968881585704,
+      "grad_norm": 0.10853458195924759,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 50532
+    },
+    {
+      "epoch": 5.189996918968881,
+      "grad_norm": 0.043089404702186584,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 50535
+    },
+    {
+      "epoch": 5.190305022080723,
+      "grad_norm": 0.05289897695183754,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 50538
+    },
+    {
+      "epoch": 5.190613125192565,
+      "grad_norm": 0.06284762918949127,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 50541
+    },
+    {
+      "epoch": 5.190921228304406,
+      "grad_norm": 0.0615723617374897,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 50544
+    },
+    {
+      "epoch": 5.191229331416247,
+      "grad_norm": 0.10425807535648346,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 50547
+    },
+    {
+      "epoch": 5.1915374345280885,
+      "grad_norm": 0.09661048650741577,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 50550
+    },
+    {
+      "epoch": 5.19184553763993,
+      "grad_norm": 0.059544507414102554,
+      "learning_rate": 0.01,
+      "loss": 2.0324,
+      "step": 50553
+    },
+    {
+      "epoch": 5.192153640751772,
+      "grad_norm": 0.08903210610151291,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 50556
+    },
+    {
+      "epoch": 5.192461743863613,
+      "grad_norm": 0.04461904987692833,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 50559
+    },
+    {
+      "epoch": 5.192769846975454,
+      "grad_norm": 0.07844073325395584,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 50562
+    },
+    {
+      "epoch": 5.193077950087296,
+      "grad_norm": 0.08520335704088211,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 50565
+    },
+    {
+      "epoch": 5.193386053199137,
+      "grad_norm": 0.13090673089027405,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 50568
+    },
+    {
+      "epoch": 5.193694156310979,
+      "grad_norm": 0.08202611654996872,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 50571
+    },
+    {
+      "epoch": 5.19400225942282,
+      "grad_norm": 0.05638153478503227,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 50574
+    },
+    {
+      "epoch": 5.194310362534662,
+      "grad_norm": 0.03813661262392998,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 50577
+    },
+    {
+      "epoch": 5.194618465646503,
+      "grad_norm": 0.13044418394565582,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 50580
+    },
+    {
+      "epoch": 5.194926568758344,
+      "grad_norm": 0.04875580221414566,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 50583
+    },
+    {
+      "epoch": 5.195234671870186,
+      "grad_norm": 0.03675463795661926,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 50586
+    },
+    {
+      "epoch": 5.1955427749820275,
+      "grad_norm": 0.050565268844366074,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 50589
+    },
+    {
+      "epoch": 5.195850878093869,
+      "grad_norm": 0.08294785022735596,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 50592
+    },
+    {
+      "epoch": 5.19615898120571,
+      "grad_norm": 0.04059421271085739,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 50595
+    },
+    {
+      "epoch": 5.196467084317551,
+      "grad_norm": 0.044956889003515244,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 50598
+    },
+    {
+      "epoch": 5.1967751874293935,
+      "grad_norm": 0.11942701786756516,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 50601
+    },
+    {
+      "epoch": 5.197083290541235,
+      "grad_norm": 0.04948386549949646,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 50604
+    },
+    {
+      "epoch": 5.197391393653076,
+      "grad_norm": 0.08146567642688751,
+      "learning_rate": 0.01,
+      "loss": 2.0315,
+      "step": 50607
+    },
+    {
+      "epoch": 5.197699496764917,
+      "grad_norm": 0.0451958030462265,
+      "learning_rate": 0.01,
+      "loss": 2.0138,
+      "step": 50610
+    },
+    {
+      "epoch": 5.1980075998767585,
+      "grad_norm": 0.036100223660469055,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 50613
+    },
+    {
+      "epoch": 5.1983157029886,
+      "grad_norm": 0.035324469208717346,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 50616
+    },
+    {
+      "epoch": 5.198623806100442,
+      "grad_norm": 0.05005744844675064,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 50619
+    },
+    {
+      "epoch": 5.198931909212283,
+      "grad_norm": 0.08080115169286728,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 50622
+    },
+    {
+      "epoch": 5.1992400123241245,
+      "grad_norm": 0.07723166048526764,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 50625
+    },
+    {
+      "epoch": 5.199548115435966,
+      "grad_norm": 0.11351467669010162,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 50628
+    },
+    {
+      "epoch": 5.199856218547807,
+      "grad_norm": 0.03940621018409729,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 50631
+    },
+    {
+      "epoch": 5.200164321659649,
+      "grad_norm": 0.06582506000995636,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 50634
+    },
+    {
+      "epoch": 5.20047242477149,
+      "grad_norm": 0.030670402571558952,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 50637
+    },
+    {
+      "epoch": 5.200780527883332,
+      "grad_norm": 0.06636448204517365,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 50640
+    },
+    {
+      "epoch": 5.201088630995173,
+      "grad_norm": 0.06079031527042389,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 50643
+    },
+    {
+      "epoch": 5.201396734107014,
+      "grad_norm": 0.1049150750041008,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 50646
+    },
+    {
+      "epoch": 5.2017048372188555,
+      "grad_norm": 0.08547158539295197,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 50649
+    },
+    {
+      "epoch": 5.202012940330698,
+      "grad_norm": 0.06914935261011124,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 50652
+    },
+    {
+      "epoch": 5.202321043442539,
+      "grad_norm": 0.05997157096862793,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 50655
+    },
+    {
+      "epoch": 5.20262914655438,
+      "grad_norm": 0.052482035011053085,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 50658
+    },
+    {
+      "epoch": 5.202937249666221,
+      "grad_norm": 0.12078742682933807,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 50661
+    },
+    {
+      "epoch": 5.203245352778063,
+      "grad_norm": 0.08411680161952972,
+      "learning_rate": 0.01,
+      "loss": 2.0155,
+      "step": 50664
+    },
+    {
+      "epoch": 5.203553455889905,
+      "grad_norm": 0.06186600774526596,
+      "learning_rate": 0.01,
+      "loss": 2.022,
+      "step": 50667
+    },
+    {
+      "epoch": 5.203861559001746,
+      "grad_norm": 0.09473706781864166,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 50670
+    },
+    {
+      "epoch": 5.204169662113587,
+      "grad_norm": 0.05376753211021423,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 50673
+    },
+    {
+      "epoch": 5.204477765225429,
+      "grad_norm": 0.08271234482526779,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 50676
+    },
+    {
+      "epoch": 5.20478586833727,
+      "grad_norm": 0.07704611122608185,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 50679
+    },
+    {
+      "epoch": 5.205093971449112,
+      "grad_norm": 0.0716482475399971,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 50682
+    },
+    {
+      "epoch": 5.205402074560953,
+      "grad_norm": 0.07560107111930847,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 50685
+    },
+    {
+      "epoch": 5.2057101776727945,
+      "grad_norm": 0.07398194819688797,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 50688
+    },
+    {
+      "epoch": 5.206018280784636,
+      "grad_norm": 0.09211590141057968,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 50691
+    },
+    {
+      "epoch": 5.206326383896477,
+      "grad_norm": 0.0511045902967453,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 50694
+    },
+    {
+      "epoch": 5.206634487008319,
+      "grad_norm": 0.046343546360731125,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 50697
+    },
+    {
+      "epoch": 5.2069425901201605,
+      "grad_norm": 0.0472230389714241,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 50700
+    },
+    {
+      "epoch": 5.207250693232002,
+      "grad_norm": 0.09529436379671097,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 50703
+    },
+    {
+      "epoch": 5.207558796343843,
+      "grad_norm": 0.15323568880558014,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 50706
+    },
+    {
+      "epoch": 5.207866899455684,
+      "grad_norm": 0.1495465189218521,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 50709
+    },
+    {
+      "epoch": 5.2081750025675255,
+      "grad_norm": 0.09416273981332779,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 50712
+    },
+    {
+      "epoch": 5.208483105679368,
+      "grad_norm": 0.05051533132791519,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 50715
+    },
+    {
+      "epoch": 5.208791208791209,
+      "grad_norm": 0.03743917867541313,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 50718
+    },
+    {
+      "epoch": 5.20909931190305,
+      "grad_norm": 0.04133860394358635,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 50721
+    },
+    {
+      "epoch": 5.2094074150148915,
+      "grad_norm": 0.05624905601143837,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 50724
+    },
+    {
+      "epoch": 5.209715518126733,
+      "grad_norm": 0.04526040330529213,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 50727
+    },
+    {
+      "epoch": 5.210023621238575,
+      "grad_norm": 0.06159577891230583,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 50730
+    },
+    {
+      "epoch": 5.210331724350416,
+      "grad_norm": 0.10257083177566528,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 50733
+    },
+    {
+      "epoch": 5.210639827462257,
+      "grad_norm": 0.0896536186337471,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 50736
+    },
+    {
+      "epoch": 5.210947930574099,
+      "grad_norm": 0.0499761626124382,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 50739
+    },
+    {
+      "epoch": 5.21125603368594,
+      "grad_norm": 0.044219888746738434,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 50742
+    },
+    {
+      "epoch": 5.211564136797782,
+      "grad_norm": 0.056992996484041214,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 50745
+    },
+    {
+      "epoch": 5.211872239909623,
+      "grad_norm": 0.10208120197057724,
+      "learning_rate": 0.01,
+      "loss": 2.014,
+      "step": 50748
+    },
+    {
+      "epoch": 5.212180343021465,
+      "grad_norm": 0.07243528962135315,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 50751
+    },
+    {
+      "epoch": 5.212488446133306,
+      "grad_norm": 0.04413043335080147,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 50754
+    },
+    {
+      "epoch": 5.212796549245147,
+      "grad_norm": 0.09615755081176758,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 50757
+    },
+    {
+      "epoch": 5.213104652356988,
+      "grad_norm": 0.04845009744167328,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 50760
+    },
+    {
+      "epoch": 5.2134127554688305,
+      "grad_norm": 0.12175197154283524,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 50763
+    },
+    {
+      "epoch": 5.213720858580672,
+      "grad_norm": 0.045076884329319,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 50766
+    },
+    {
+      "epoch": 5.214028961692513,
+      "grad_norm": 0.07600415498018265,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 50769
+    },
+    {
+      "epoch": 5.214337064804354,
+      "grad_norm": 0.06227222830057144,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 50772
+    },
+    {
+      "epoch": 5.214645167916196,
+      "grad_norm": 0.10495235025882721,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 50775
+    },
+    {
+      "epoch": 5.214953271028038,
+      "grad_norm": 0.08388978242874146,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 50778
+    },
+    {
+      "epoch": 5.215261374139879,
+      "grad_norm": 0.07591544091701508,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 50781
+    },
+    {
+      "epoch": 5.21556947725172,
+      "grad_norm": 0.07411694526672363,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 50784
+    },
+    {
+      "epoch": 5.2158775803635615,
+      "grad_norm": 0.07496411353349686,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 50787
+    },
+    {
+      "epoch": 5.216185683475403,
+      "grad_norm": 0.07005130499601364,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 50790
+    },
+    {
+      "epoch": 5.216493786587245,
+      "grad_norm": 0.05165226384997368,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 50793
+    },
+    {
+      "epoch": 5.216801889699086,
+      "grad_norm": 0.09063982218503952,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 50796
+    },
+    {
+      "epoch": 5.2171099928109275,
+      "grad_norm": 0.03189365193247795,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 50799
+    },
+    {
+      "epoch": 5.217418095922769,
+      "grad_norm": 0.07051954418420792,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 50802
+    },
+    {
+      "epoch": 5.21772619903461,
+      "grad_norm": 0.045339349657297134,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 50805
+    },
+    {
+      "epoch": 5.218034302146451,
+      "grad_norm": 0.041392091661691666,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 50808
+    },
+    {
+      "epoch": 5.218342405258293,
+      "grad_norm": 0.05115804076194763,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 50811
+    },
+    {
+      "epoch": 5.218650508370135,
+      "grad_norm": 0.0681786835193634,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 50814
+    },
+    {
+      "epoch": 5.218958611481976,
+      "grad_norm": 0.057782597839832306,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 50817
+    },
+    {
+      "epoch": 5.219266714593817,
+      "grad_norm": 0.04151969403028488,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 50820
+    },
+    {
+      "epoch": 5.2195748177056585,
+      "grad_norm": 0.06815563142299652,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 50823
+    },
+    {
+      "epoch": 5.219882920817501,
+      "grad_norm": 0.1696784943342209,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 50826
+    },
+    {
+      "epoch": 5.220191023929342,
+      "grad_norm": 0.04126200079917908,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 50829
+    },
+    {
+      "epoch": 5.220499127041183,
+      "grad_norm": 0.03278988599777222,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 50832
+    },
+    {
+      "epoch": 5.220807230153024,
+      "grad_norm": 0.0632409155368805,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 50835
+    },
+    {
+      "epoch": 5.221115333264866,
+      "grad_norm": 0.06449749320745468,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 50838
+    },
+    {
+      "epoch": 5.221423436376708,
+      "grad_norm": 0.04530004784464836,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 50841
+    },
+    {
+      "epoch": 5.221731539488549,
+      "grad_norm": 0.06646733731031418,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 50844
+    },
+    {
+      "epoch": 5.22203964260039,
+      "grad_norm": 0.040766507387161255,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 50847
+    },
+    {
+      "epoch": 5.222347745712232,
+      "grad_norm": 0.04910712689161301,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 50850
+    },
+    {
+      "epoch": 5.222655848824073,
+      "grad_norm": 0.03790167346596718,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 50853
+    },
+    {
+      "epoch": 5.222963951935915,
+      "grad_norm": 0.04654322564601898,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 50856
+    },
+    {
+      "epoch": 5.223272055047756,
+      "grad_norm": 0.05322917550802231,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 50859
+    },
+    {
+      "epoch": 5.2235801581595975,
+      "grad_norm": 0.10331053286790848,
+      "learning_rate": 0.01,
+      "loss": 2.0289,
+      "step": 50862
+    },
+    {
+      "epoch": 5.223888261271439,
+      "grad_norm": 0.14793960750102997,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 50865
+    },
+    {
+      "epoch": 5.22419636438328,
+      "grad_norm": 0.06507808715105057,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 50868
+    },
+    {
+      "epoch": 5.224504467495121,
+      "grad_norm": 0.06138547509908676,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 50871
+    },
+    {
+      "epoch": 5.2248125706069635,
+      "grad_norm": 0.04522119089961052,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 50874
+    },
+    {
+      "epoch": 5.225120673718805,
+      "grad_norm": 0.044236768037080765,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 50877
+    },
+    {
+      "epoch": 5.225428776830646,
+      "grad_norm": 0.05313803628087044,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 50880
+    },
+    {
+      "epoch": 5.225736879942487,
+      "grad_norm": 0.06283990293741226,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 50883
+    },
+    {
+      "epoch": 5.2260449830543285,
+      "grad_norm": 0.05976197123527527,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 50886
+    },
+    {
+      "epoch": 5.226353086166171,
+      "grad_norm": 0.10755223780870438,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 50889
+    },
+    {
+      "epoch": 5.226661189278012,
+      "grad_norm": 0.040908023715019226,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 50892
+    },
+    {
+      "epoch": 5.226969292389853,
+      "grad_norm": 0.08019308000802994,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 50895
+    },
+    {
+      "epoch": 5.2272773955016945,
+      "grad_norm": 0.04088309779763222,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 50898
+    },
+    {
+      "epoch": 5.227585498613536,
+      "grad_norm": 0.06042852997779846,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 50901
+    },
+    {
+      "epoch": 5.227893601725377,
+      "grad_norm": 0.06619790196418762,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 50904
+    },
+    {
+      "epoch": 5.228201704837219,
+      "grad_norm": 0.0903916209936142,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 50907
+    },
+    {
+      "epoch": 5.22850980794906,
+      "grad_norm": 0.06611988693475723,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 50910
+    },
+    {
+      "epoch": 5.228817911060902,
+      "grad_norm": 0.07255574315786362,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 50913
+    },
+    {
+      "epoch": 5.229126014172743,
+      "grad_norm": 0.07492070645093918,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 50916
+    },
+    {
+      "epoch": 5.229434117284584,
+      "grad_norm": 0.07850334793329239,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 50919
+    },
+    {
+      "epoch": 5.229742220396426,
+      "grad_norm": 0.0485299751162529,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 50922
+    },
+    {
+      "epoch": 5.230050323508268,
+      "grad_norm": 0.09580907970666885,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 50925
+    },
+    {
+      "epoch": 5.230358426620109,
+      "grad_norm": 0.10689478367567062,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 50928
+    },
+    {
+      "epoch": 5.23066652973195,
+      "grad_norm": 0.08192954957485199,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 50931
+    },
+    {
+      "epoch": 5.230974632843791,
+      "grad_norm": 0.06330616772174835,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 50934
+    },
+    {
+      "epoch": 5.2312827359556335,
+      "grad_norm": 0.041825070977211,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 50937
+    },
+    {
+      "epoch": 5.231590839067475,
+      "grad_norm": 0.04370388388633728,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 50940
+    },
+    {
+      "epoch": 5.231898942179316,
+      "grad_norm": 0.039774972945451736,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 50943
+    },
+    {
+      "epoch": 5.232207045291157,
+      "grad_norm": 0.06065038964152336,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 50946
+    },
+    {
+      "epoch": 5.232515148402999,
+      "grad_norm": 0.1340397149324417,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 50949
+    },
+    {
+      "epoch": 5.232823251514841,
+      "grad_norm": 0.05149194970726967,
+      "learning_rate": 0.01,
+      "loss": 2.0259,
+      "step": 50952
+    },
+    {
+      "epoch": 5.233131354626682,
+      "grad_norm": 0.043973732739686966,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 50955
+    },
+    {
+      "epoch": 5.233439457738523,
+      "grad_norm": 0.034024860709905624,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 50958
+    },
+    {
+      "epoch": 5.2337475608503645,
+      "grad_norm": 0.03558748587965965,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 50961
+    },
+    {
+      "epoch": 5.234055663962206,
+      "grad_norm": 0.03661567345261574,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 50964
+    },
+    {
+      "epoch": 5.234363767074047,
+      "grad_norm": 0.11346925795078278,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 50967
+    },
+    {
+      "epoch": 5.234671870185889,
+      "grad_norm": 0.05934759974479675,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 50970
+    },
+    {
+      "epoch": 5.2349799732977305,
+      "grad_norm": 0.08264946192502975,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 50973
+    },
+    {
+      "epoch": 5.235288076409572,
+      "grad_norm": 0.0470314584672451,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 50976
+    },
+    {
+      "epoch": 5.235596179521413,
+      "grad_norm": 0.04239676892757416,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 50979
+    },
+    {
+      "epoch": 5.235904282633254,
+      "grad_norm": 0.09247265011072159,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 50982
+    },
+    {
+      "epoch": 5.236212385745096,
+      "grad_norm": 0.05270032957196236,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 50985
+    },
+    {
+      "epoch": 5.236520488856938,
+      "grad_norm": 0.0351402647793293,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 50988
+    },
+    {
+      "epoch": 5.236828591968779,
+      "grad_norm": 0.04401613771915436,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 50991
+    },
+    {
+      "epoch": 5.23713669508062,
+      "grad_norm": 0.0353761650621891,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 50994
+    },
+    {
+      "epoch": 5.2374447981924614,
+      "grad_norm": 0.13582314550876617,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 50997
+    },
+    {
+      "epoch": 5.237752901304304,
+      "grad_norm": 0.06481800973415375,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 51000
+    },
+    {
+      "epoch": 5.238061004416145,
+      "grad_norm": 0.06499173492193222,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 51003
+    },
+    {
+      "epoch": 5.238369107527986,
+      "grad_norm": 0.059027299284935,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 51006
+    },
+    {
+      "epoch": 5.238677210639827,
+      "grad_norm": 0.11211716383695602,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 51009
+    },
+    {
+      "epoch": 5.238985313751669,
+      "grad_norm": 0.041850294917821884,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 51012
+    },
+    {
+      "epoch": 5.23929341686351,
+      "grad_norm": 0.11037513613700867,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 51015
+    },
+    {
+      "epoch": 5.239601519975352,
+      "grad_norm": 0.06720243394374847,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 51018
+    },
+    {
+      "epoch": 5.239909623087193,
+      "grad_norm": 0.05992870405316353,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 51021
+    },
+    {
+      "epoch": 5.240217726199035,
+      "grad_norm": 0.046941906213760376,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 51024
+    },
+    {
+      "epoch": 5.240525829310876,
+      "grad_norm": 0.0415794812142849,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 51027
+    },
+    {
+      "epoch": 5.240833932422717,
+      "grad_norm": 0.046740952879190445,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 51030
+    },
+    {
+      "epoch": 5.241142035534559,
+      "grad_norm": 0.04328366369009018,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 51033
+    },
+    {
+      "epoch": 5.2414501386464005,
+      "grad_norm": 0.10686604678630829,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 51036
+    },
+    {
+      "epoch": 5.241758241758242,
+      "grad_norm": 0.0445748046040535,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 51039
+    },
+    {
+      "epoch": 5.242066344870083,
+      "grad_norm": 0.04299461841583252,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 51042
+    },
+    {
+      "epoch": 5.242374447981924,
+      "grad_norm": 0.06391250342130661,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 51045
+    },
+    {
+      "epoch": 5.2426825510937665,
+      "grad_norm": 0.03528610244393349,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 51048
+    },
+    {
+      "epoch": 5.242990654205608,
+      "grad_norm": 0.030670564621686935,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 51051
+    },
+    {
+      "epoch": 5.243298757317449,
+      "grad_norm": 0.11291905492544174,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 51054
+    },
+    {
+      "epoch": 5.24360686042929,
+      "grad_norm": 0.08508855104446411,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 51057
+    },
+    {
+      "epoch": 5.2439149635411315,
+      "grad_norm": 0.1319185048341751,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 51060
+    },
+    {
+      "epoch": 5.244223066652973,
+      "grad_norm": 0.08218291401863098,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 51063
+    },
+    {
+      "epoch": 5.244531169764815,
+      "grad_norm": 0.06311830878257751,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 51066
+    },
+    {
+      "epoch": 5.244839272876656,
+      "grad_norm": 0.03945956006646156,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 51069
+    },
+    {
+      "epoch": 5.2451473759884975,
+      "grad_norm": 0.045635320246219635,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 51072
+    },
+    {
+      "epoch": 5.245455479100339,
+      "grad_norm": 0.047721873968839645,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 51075
+    },
+    {
+      "epoch": 5.24576358221218,
+      "grad_norm": 0.04936512932181358,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 51078
+    },
+    {
+      "epoch": 5.246071685324022,
+      "grad_norm": 0.05856813117861748,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 51081
+    },
+    {
+      "epoch": 5.246379788435863,
+      "grad_norm": 0.05068356543779373,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 51084
+    },
+    {
+      "epoch": 5.246687891547705,
+      "grad_norm": 0.05900732800364494,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 51087
+    },
+    {
+      "epoch": 5.246995994659546,
+      "grad_norm": 0.12454086542129517,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 51090
+    },
+    {
+      "epoch": 5.247304097771387,
+      "grad_norm": 0.037892140448093414,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 51093
+    },
+    {
+      "epoch": 5.247612200883229,
+      "grad_norm": 0.07839351892471313,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 51096
+    },
+    {
+      "epoch": 5.247920303995071,
+      "grad_norm": 0.04203120246529579,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 51099
+    },
+    {
+      "epoch": 5.248228407106912,
+      "grad_norm": 0.07562438398599625,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 51102
+    },
+    {
+      "epoch": 5.248536510218753,
+      "grad_norm": 0.044604718685150146,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 51105
+    },
+    {
+      "epoch": 5.248844613330594,
+      "grad_norm": 0.09438357502222061,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 51108
+    },
+    {
+      "epoch": 5.249152716442436,
+      "grad_norm": 0.04773309826850891,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 51111
+    },
+    {
+      "epoch": 5.249460819554278,
+      "grad_norm": 0.033143848180770874,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 51114
+    },
+    {
+      "epoch": 5.249768922666119,
+      "grad_norm": 0.0772877037525177,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 51117
+    },
+    {
+      "epoch": 5.25007702577796,
+      "grad_norm": 0.054028406739234924,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 51120
+    },
+    {
+      "epoch": 5.250385128889802,
+      "grad_norm": 0.07667548954486847,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 51123
+    },
+    {
+      "epoch": 5.250693232001643,
+      "grad_norm": 0.06274887174367905,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 51126
+    },
+    {
+      "epoch": 5.251001335113485,
+      "grad_norm": 0.09405583888292313,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 51129
+    },
+    {
+      "epoch": 5.251309438225326,
+      "grad_norm": 0.06297732889652252,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 51132
+    },
+    {
+      "epoch": 5.2516175413371675,
+      "grad_norm": 0.08258040994405746,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 51135
+    },
+    {
+      "epoch": 5.251925644449009,
+      "grad_norm": 0.11019886285066605,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 51138
+    },
+    {
+      "epoch": 5.25223374756085,
+      "grad_norm": 0.06206972151994705,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 51141
+    },
+    {
+      "epoch": 5.252541850672692,
+      "grad_norm": 0.045681048184633255,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 51144
+    },
+    {
+      "epoch": 5.2528499537845335,
+      "grad_norm": 0.041622232645750046,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 51147
+    },
+    {
+      "epoch": 5.253158056896375,
+      "grad_norm": 0.05589982122182846,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 51150
+    },
+    {
+      "epoch": 5.253466160008216,
+      "grad_norm": 0.06394509226083755,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 51153
+    },
+    {
+      "epoch": 5.253774263120057,
+      "grad_norm": 0.05354580655694008,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 51156
+    },
+    {
+      "epoch": 5.2540823662318985,
+      "grad_norm": 0.06941652297973633,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 51159
+    },
+    {
+      "epoch": 5.254390469343741,
+      "grad_norm": 0.0771641731262207,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 51162
+    },
+    {
+      "epoch": 5.254698572455582,
+      "grad_norm": 0.08898761868476868,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 51165
+    },
+    {
+      "epoch": 5.255006675567423,
+      "grad_norm": 0.03489289805293083,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 51168
+    },
+    {
+      "epoch": 5.255314778679264,
+      "grad_norm": 0.0783892497420311,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 51171
+    },
+    {
+      "epoch": 5.255622881791106,
+      "grad_norm": 0.08177445828914642,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 51174
+    },
+    {
+      "epoch": 5.255930984902948,
+      "grad_norm": 0.04709051921963692,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 51177
+    },
+    {
+      "epoch": 5.256239088014789,
+      "grad_norm": 0.13084626197814941,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 51180
+    },
+    {
+      "epoch": 5.25654719112663,
+      "grad_norm": 0.04432738944888115,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 51183
+    },
+    {
+      "epoch": 5.256855294238472,
+      "grad_norm": 0.053161200135946274,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 51186
+    },
+    {
+      "epoch": 5.257163397350313,
+      "grad_norm": 0.039357706904411316,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 51189
+    },
+    {
+      "epoch": 5.257471500462155,
+      "grad_norm": 0.052901457995176315,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 51192
+    },
+    {
+      "epoch": 5.257779603573996,
+      "grad_norm": 0.11097205430269241,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 51195
+    },
+    {
+      "epoch": 5.258087706685838,
+      "grad_norm": 0.07301553338766098,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 51198
+    },
+    {
+      "epoch": 5.258395809797679,
+      "grad_norm": 0.17759805917739868,
+      "learning_rate": 0.01,
+      "loss": 2.0266,
+      "step": 51201
+    },
+    {
+      "epoch": 5.25870391290952,
+      "grad_norm": 0.07599959522485733,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 51204
+    },
+    {
+      "epoch": 5.259012016021362,
+      "grad_norm": 0.13010208308696747,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 51207
+    },
+    {
+      "epoch": 5.2593201191332035,
+      "grad_norm": 0.07105369120836258,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 51210
+    },
+    {
+      "epoch": 5.259628222245045,
+      "grad_norm": 0.16369540989398956,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 51213
+    },
+    {
+      "epoch": 5.259936325356886,
+      "grad_norm": 0.11588908731937408,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 51216
+    },
+    {
+      "epoch": 5.260244428468727,
+      "grad_norm": 0.045902661979198456,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 51219
+    },
+    {
+      "epoch": 5.260552531580569,
+      "grad_norm": 0.04714805632829666,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 51222
+    },
+    {
+      "epoch": 5.260860634692411,
+      "grad_norm": 0.05099105089902878,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 51225
+    },
+    {
+      "epoch": 5.261168737804252,
+      "grad_norm": 0.050430044531822205,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 51228
+    },
+    {
+      "epoch": 5.261476840916093,
+      "grad_norm": 0.03998453542590141,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 51231
+    },
+    {
+      "epoch": 5.2617849440279345,
+      "grad_norm": 0.05709048733115196,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 51234
+    },
+    {
+      "epoch": 5.262093047139776,
+      "grad_norm": 0.04667646437883377,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 51237
+    },
+    {
+      "epoch": 5.262401150251618,
+      "grad_norm": 0.0413067601621151,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 51240
+    },
+    {
+      "epoch": 5.262709253363459,
+      "grad_norm": 0.037300482392311096,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 51243
+    },
+    {
+      "epoch": 5.2630173564753004,
+      "grad_norm": 0.10188855230808258,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 51246
+    },
+    {
+      "epoch": 5.263325459587142,
+      "grad_norm": 0.060019396245479584,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 51249
+    },
+    {
+      "epoch": 5.263633562698983,
+      "grad_norm": 0.07587499171495438,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 51252
+    },
+    {
+      "epoch": 5.263941665810824,
+      "grad_norm": 0.17544515430927277,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 51255
+    },
+    {
+      "epoch": 5.264249768922666,
+      "grad_norm": 0.046111736446619034,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 51258
+    },
+    {
+      "epoch": 5.264557872034508,
+      "grad_norm": 0.03697605058550835,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 51261
+    },
+    {
+      "epoch": 5.264865975146349,
+      "grad_norm": 0.04376983270049095,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 51264
+    },
+    {
+      "epoch": 5.26517407825819,
+      "grad_norm": 0.06879153102636337,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 51267
+    },
+    {
+      "epoch": 5.265482181370031,
+      "grad_norm": 0.05067580193281174,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 51270
+    },
+    {
+      "epoch": 5.265790284481874,
+      "grad_norm": 0.04876921325922012,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 51273
+    },
+    {
+      "epoch": 5.266098387593715,
+      "grad_norm": 0.08174460381269455,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 51276
+    },
+    {
+      "epoch": 5.266406490705556,
+      "grad_norm": 0.08350025117397308,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 51279
+    },
+    {
+      "epoch": 5.266714593817397,
+      "grad_norm": 0.10765060782432556,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 51282
+    },
+    {
+      "epoch": 5.267022696929239,
+      "grad_norm": 0.09596231579780579,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 51285
+    },
+    {
+      "epoch": 5.267330800041081,
+      "grad_norm": 0.03738878667354584,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 51288
+    },
+    {
+      "epoch": 5.267638903152922,
+      "grad_norm": 0.03415136784315109,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 51291
+    },
+    {
+      "epoch": 5.267947006264763,
+      "grad_norm": 0.04650856927037239,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 51294
+    },
+    {
+      "epoch": 5.268255109376605,
+      "grad_norm": 0.06548796594142914,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 51297
+    },
+    {
+      "epoch": 5.268563212488446,
+      "grad_norm": 0.05420374125242233,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 51300
+    },
+    {
+      "epoch": 5.268871315600288,
+      "grad_norm": 0.10747869312763214,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 51303
+    },
+    {
+      "epoch": 5.269179418712129,
+      "grad_norm": 0.03331341966986656,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 51306
+    },
+    {
+      "epoch": 5.2694875218239705,
+      "grad_norm": 0.10144448280334473,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 51309
+    },
+    {
+      "epoch": 5.269795624935812,
+      "grad_norm": 0.08695587515830994,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 51312
+    },
+    {
+      "epoch": 5.270103728047653,
+      "grad_norm": 0.053817540407180786,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 51315
+    },
+    {
+      "epoch": 5.270411831159494,
+      "grad_norm": 0.04269054904580116,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 51318
+    },
+    {
+      "epoch": 5.2707199342713364,
+      "grad_norm": 0.048786550760269165,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 51321
+    },
+    {
+      "epoch": 5.271028037383178,
+      "grad_norm": 0.0634552389383316,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 51324
+    },
+    {
+      "epoch": 5.271336140495019,
+      "grad_norm": 0.14295388758182526,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 51327
+    },
+    {
+      "epoch": 5.27164424360686,
+      "grad_norm": 0.05080576241016388,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 51330
+    },
+    {
+      "epoch": 5.2719523467187015,
+      "grad_norm": 0.06417543441057205,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 51333
+    },
+    {
+      "epoch": 5.272260449830544,
+      "grad_norm": 0.06565088778734207,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 51336
+    },
+    {
+      "epoch": 5.272568552942385,
+      "grad_norm": 0.028631992638111115,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 51339
+    },
+    {
+      "epoch": 5.272876656054226,
+      "grad_norm": 0.10242610424757004,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 51342
+    },
+    {
+      "epoch": 5.273184759166067,
+      "grad_norm": 0.11979396641254425,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 51345
+    },
+    {
+      "epoch": 5.273492862277909,
+      "grad_norm": 0.07476961612701416,
+      "learning_rate": 0.01,
+      "loss": 2.0282,
+      "step": 51348
+    },
+    {
+      "epoch": 5.273800965389751,
+      "grad_norm": 0.05603867769241333,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 51351
+    },
+    {
+      "epoch": 5.274109068501592,
+      "grad_norm": 0.04897085204720497,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 51354
+    },
+    {
+      "epoch": 5.274417171613433,
+      "grad_norm": 0.03439117968082428,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 51357
+    },
+    {
+      "epoch": 5.274725274725275,
+      "grad_norm": 0.08314791321754456,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 51360
+    },
+    {
+      "epoch": 5.275033377837116,
+      "grad_norm": 0.0685926079750061,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 51363
+    },
+    {
+      "epoch": 5.275341480948958,
+      "grad_norm": 0.088614821434021,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 51366
+    },
+    {
+      "epoch": 5.275649584060799,
+      "grad_norm": 0.03873438388109207,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 51369
+    },
+    {
+      "epoch": 5.275957687172641,
+      "grad_norm": 0.11569857597351074,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 51372
+    },
+    {
+      "epoch": 5.276265790284482,
+      "grad_norm": 0.11536852270364761,
+      "learning_rate": 0.01,
+      "loss": 2.024,
+      "step": 51375
+    },
+    {
+      "epoch": 5.276573893396323,
+      "grad_norm": 0.040931250900030136,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 51378
+    },
+    {
+      "epoch": 5.276881996508164,
+      "grad_norm": 0.13397638499736786,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 51381
+    },
+    {
+      "epoch": 5.2771900996200065,
+      "grad_norm": 0.045118771493434906,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 51384
+    },
+    {
+      "epoch": 5.277498202731848,
+      "grad_norm": 0.04626978933811188,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 51387
+    },
+    {
+      "epoch": 5.277806305843689,
+      "grad_norm": 0.03898259624838829,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 51390
+    },
+    {
+      "epoch": 5.27811440895553,
+      "grad_norm": 0.14309214055538177,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 51393
+    },
+    {
+      "epoch": 5.278422512067372,
+      "grad_norm": 0.11283177882432938,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 51396
+    },
+    {
+      "epoch": 5.278730615179214,
+      "grad_norm": 0.07250397652387619,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 51399
+    },
+    {
+      "epoch": 5.279038718291055,
+      "grad_norm": 0.04467407613992691,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 51402
+    },
+    {
+      "epoch": 5.279346821402896,
+      "grad_norm": 0.03758449852466583,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 51405
+    },
+    {
+      "epoch": 5.2796549245147375,
+      "grad_norm": 0.04290134087204933,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 51408
+    },
+    {
+      "epoch": 5.279963027626579,
+      "grad_norm": 0.21099735796451569,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 51411
+    },
+    {
+      "epoch": 5.28027113073842,
+      "grad_norm": 0.10676965117454529,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 51414
+    },
+    {
+      "epoch": 5.280579233850262,
+      "grad_norm": 0.11726178228855133,
+      "learning_rate": 0.01,
+      "loss": 2.0066,
+      "step": 51417
+    },
+    {
+      "epoch": 5.280887336962103,
+      "grad_norm": 0.06197341904044151,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 51420
+    },
+    {
+      "epoch": 5.281195440073945,
+      "grad_norm": 0.04718432575464249,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 51423
+    },
+    {
+      "epoch": 5.281503543185786,
+      "grad_norm": 0.049280427396297455,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 51426
+    },
+    {
+      "epoch": 5.281811646297627,
+      "grad_norm": 0.036890413612127304,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 51429
+    },
+    {
+      "epoch": 5.282119749409469,
+      "grad_norm": 0.03962325677275658,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 51432
+    },
+    {
+      "epoch": 5.282427852521311,
+      "grad_norm": 0.03786443546414375,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 51435
+    },
+    {
+      "epoch": 5.282735955633152,
+      "grad_norm": 0.06800255179405212,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 51438
+    },
+    {
+      "epoch": 5.283044058744993,
+      "grad_norm": 0.1254369616508484,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 51441
+    },
+    {
+      "epoch": 5.283352161856834,
+      "grad_norm": 0.10241379588842392,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 51444
+    },
+    {
+      "epoch": 5.283660264968677,
+      "grad_norm": 0.07036352902650833,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 51447
+    },
+    {
+      "epoch": 5.283968368080518,
+      "grad_norm": 0.04033416137099266,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 51450
+    },
+    {
+      "epoch": 5.284276471192359,
+      "grad_norm": 0.03396594151854515,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 51453
+    },
+    {
+      "epoch": 5.2845845743042,
+      "grad_norm": 0.038950271904468536,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 51456
+    },
+    {
+      "epoch": 5.284892677416042,
+      "grad_norm": 0.12410634011030197,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 51459
+    },
+    {
+      "epoch": 5.285200780527884,
+      "grad_norm": 0.08871324360370636,
+      "learning_rate": 0.01,
+      "loss": 2.0303,
+      "step": 51462
+    },
+    {
+      "epoch": 5.285508883639725,
+      "grad_norm": 0.06051253527402878,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 51465
+    },
+    {
+      "epoch": 5.285816986751566,
+      "grad_norm": 0.08062247186899185,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 51468
+    },
+    {
+      "epoch": 5.286125089863408,
+      "grad_norm": 0.09288395196199417,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 51471
+    },
+    {
+      "epoch": 5.286433192975249,
+      "grad_norm": 0.11983957886695862,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 51474
+    },
+    {
+      "epoch": 5.28674129608709,
+      "grad_norm": 0.12828132510185242,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 51477
+    },
+    {
+      "epoch": 5.287049399198932,
+      "grad_norm": 0.09051026403903961,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 51480
+    },
+    {
+      "epoch": 5.2873575023107735,
+      "grad_norm": 0.08622095733880997,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 51483
+    },
+    {
+      "epoch": 5.287665605422615,
+      "grad_norm": 0.0727437287569046,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 51486
+    },
+    {
+      "epoch": 5.287973708534456,
+      "grad_norm": 0.052949920296669006,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 51489
+    },
+    {
+      "epoch": 5.288281811646297,
+      "grad_norm": 0.04807107150554657,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 51492
+    },
+    {
+      "epoch": 5.288589914758139,
+      "grad_norm": 0.12251218408346176,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 51495
+    },
+    {
+      "epoch": 5.288898017869981,
+      "grad_norm": 0.07985778152942657,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 51498
+    },
+    {
+      "epoch": 5.289206120981822,
+      "grad_norm": 0.0659778043627739,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 51501
+    },
+    {
+      "epoch": 5.289514224093663,
+      "grad_norm": 0.0646664947271347,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 51504
+    },
+    {
+      "epoch": 5.2898223272055045,
+      "grad_norm": 0.05297090485692024,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 51507
+    },
+    {
+      "epoch": 5.290130430317346,
+      "grad_norm": 0.039955638349056244,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 51510
+    },
+    {
+      "epoch": 5.290438533429188,
+      "grad_norm": 0.03601599857211113,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 51513
+    },
+    {
+      "epoch": 5.290746636541029,
+      "grad_norm": 0.03554227575659752,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 51516
+    },
+    {
+      "epoch": 5.29105473965287,
+      "grad_norm": 0.0429302453994751,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 51519
+    },
+    {
+      "epoch": 5.291362842764712,
+      "grad_norm": 0.07092410326004028,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 51522
+    },
+    {
+      "epoch": 5.291670945876553,
+      "grad_norm": 0.09126897156238556,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 51525
+    },
+    {
+      "epoch": 5.291979048988395,
+      "grad_norm": 0.08033829182386398,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 51528
+    },
+    {
+      "epoch": 5.292287152100236,
+      "grad_norm": 0.0636729821562767,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 51531
+    },
+    {
+      "epoch": 5.292595255212078,
+      "grad_norm": 0.05228522792458534,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 51534
+    },
+    {
+      "epoch": 5.292903358323919,
+      "grad_norm": 0.04770840331912041,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 51537
+    },
+    {
+      "epoch": 5.29321146143576,
+      "grad_norm": 0.06592493504285812,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 51540
+    },
+    {
+      "epoch": 5.293519564547602,
+      "grad_norm": 0.043730318546295166,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 51543
+    },
+    {
+      "epoch": 5.293827667659444,
+      "grad_norm": 0.11756035685539246,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 51546
+    },
+    {
+      "epoch": 5.294135770771285,
+      "grad_norm": 0.05374009907245636,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 51549
+    },
+    {
+      "epoch": 5.294443873883126,
+      "grad_norm": 0.03395754098892212,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 51552
+    },
+    {
+      "epoch": 5.294751976994967,
+      "grad_norm": 0.04565683752298355,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 51555
+    },
+    {
+      "epoch": 5.2950600801068095,
+      "grad_norm": 0.06022532284259796,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 51558
+    },
+    {
+      "epoch": 5.295368183218651,
+      "grad_norm": 0.03321307152509689,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 51561
+    },
+    {
+      "epoch": 5.295676286330492,
+      "grad_norm": 0.09450728446245193,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 51564
+    },
+    {
+      "epoch": 5.295984389442333,
+      "grad_norm": 0.07322457432746887,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 51567
+    },
+    {
+      "epoch": 5.296292492554175,
+      "grad_norm": 0.07152074575424194,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 51570
+    },
+    {
+      "epoch": 5.296600595666016,
+      "grad_norm": 0.09634838998317719,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 51573
+    },
+    {
+      "epoch": 5.296908698777858,
+      "grad_norm": 0.055339280515909195,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 51576
+    },
+    {
+      "epoch": 5.297216801889699,
+      "grad_norm": 0.03655093163251877,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 51579
+    },
+    {
+      "epoch": 5.2975249050015405,
+      "grad_norm": 0.04170520231127739,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 51582
+    },
+    {
+      "epoch": 5.297833008113382,
+      "grad_norm": 0.04089641198515892,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 51585
+    },
+    {
+      "epoch": 5.298141111225223,
+      "grad_norm": 0.03209071606397629,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 51588
+    },
+    {
+      "epoch": 5.298449214337065,
+      "grad_norm": 0.040264058858156204,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 51591
+    },
+    {
+      "epoch": 5.298757317448906,
+      "grad_norm": 0.040750179439783096,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 51594
+    },
+    {
+      "epoch": 5.299065420560748,
+      "grad_norm": 0.09608472883701324,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 51597
+    },
+    {
+      "epoch": 5.299373523672589,
+      "grad_norm": 0.0965270921587944,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 51600
+    },
+    {
+      "epoch": 5.29968162678443,
+      "grad_norm": 0.07110472768545151,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 51603
+    },
+    {
+      "epoch": 5.2999897298962715,
+      "grad_norm": 0.08404947817325592,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 51606
+    },
+    {
+      "epoch": 5.300297833008114,
+      "grad_norm": 0.04731042683124542,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 51609
+    },
+    {
+      "epoch": 5.300605936119955,
+      "grad_norm": 0.045290857553482056,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 51612
+    },
+    {
+      "epoch": 5.300914039231796,
+      "grad_norm": 0.04302288219332695,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 51615
+    },
+    {
+      "epoch": 5.301222142343637,
+      "grad_norm": 0.04258449375629425,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 51618
+    },
+    {
+      "epoch": 5.301530245455479,
+      "grad_norm": 0.04786079004406929,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 51621
+    },
+    {
+      "epoch": 5.301838348567321,
+      "grad_norm": 0.14586381614208221,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 51624
+    },
+    {
+      "epoch": 5.302146451679162,
+      "grad_norm": 0.06271131336688995,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 51627
+    },
+    {
+      "epoch": 5.302454554791003,
+      "grad_norm": 0.05278899893164635,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 51630
+    },
+    {
+      "epoch": 5.302762657902845,
+      "grad_norm": 0.04671400785446167,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 51633
+    },
+    {
+      "epoch": 5.303070761014686,
+      "grad_norm": 0.07106892019510269,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 51636
+    },
+    {
+      "epoch": 5.303378864126528,
+      "grad_norm": 0.031581420451402664,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 51639
+    },
+    {
+      "epoch": 5.303686967238369,
+      "grad_norm": 0.04924008622765541,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 51642
+    },
+    {
+      "epoch": 5.303995070350211,
+      "grad_norm": 0.11483835428953171,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 51645
+    },
+    {
+      "epoch": 5.304303173462052,
+      "grad_norm": 0.07446176558732986,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 51648
+    },
+    {
+      "epoch": 5.304611276573893,
+      "grad_norm": 0.051111143082380295,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 51651
+    },
+    {
+      "epoch": 5.304919379685735,
+      "grad_norm": 0.040741320699453354,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 51654
+    },
+    {
+      "epoch": 5.3052274827975765,
+      "grad_norm": 0.06887371838092804,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 51657
+    },
+    {
+      "epoch": 5.305535585909418,
+      "grad_norm": 0.09351345151662827,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 51660
+    },
+    {
+      "epoch": 5.305843689021259,
+      "grad_norm": 0.0568513348698616,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 51663
+    },
+    {
+      "epoch": 5.3061517921331,
+      "grad_norm": 0.05117219313979149,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 51666
+    },
+    {
+      "epoch": 5.3064598952449415,
+      "grad_norm": 0.04374157264828682,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 51669
+    },
+    {
+      "epoch": 5.306767998356784,
+      "grad_norm": 0.04442832991480827,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 51672
+    },
+    {
+      "epoch": 5.307076101468625,
+      "grad_norm": 0.0560653991997242,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 51675
+    },
+    {
+      "epoch": 5.307384204580466,
+      "grad_norm": 0.11248762905597687,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 51678
+    },
+    {
+      "epoch": 5.3076923076923075,
+      "grad_norm": 0.05383045971393585,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 51681
+    },
+    {
+      "epoch": 5.308000410804149,
+      "grad_norm": 0.11476562172174454,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 51684
+    },
+    {
+      "epoch": 5.308308513915991,
+      "grad_norm": 0.04139787703752518,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 51687
+    },
+    {
+      "epoch": 5.308616617027832,
+      "grad_norm": 0.03532332554459572,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 51690
+    },
+    {
+      "epoch": 5.308924720139673,
+      "grad_norm": 0.04809233918786049,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 51693
+    },
+    {
+      "epoch": 5.309232823251515,
+      "grad_norm": 0.0827130451798439,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 51696
+    },
+    {
+      "epoch": 5.309540926363356,
+      "grad_norm": 0.05010407045483589,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 51699
+    },
+    {
+      "epoch": 5.309849029475198,
+      "grad_norm": 0.04733329638838768,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 51702
+    },
+    {
+      "epoch": 5.310157132587039,
+      "grad_norm": 0.03515147417783737,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 51705
+    },
+    {
+      "epoch": 5.310465235698881,
+      "grad_norm": 0.12705236673355103,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 51708
+    },
+    {
+      "epoch": 5.310773338810722,
+      "grad_norm": 0.04811834171414375,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 51711
+    },
+    {
+      "epoch": 5.311081441922563,
+      "grad_norm": 0.04282277077436447,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 51714
+    },
+    {
+      "epoch": 5.311389545034405,
+      "grad_norm": 0.039072856307029724,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 51717
+    },
+    {
+      "epoch": 5.311697648146247,
+      "grad_norm": 0.06571515649557114,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 51720
+    },
+    {
+      "epoch": 5.312005751258088,
+      "grad_norm": 0.10408081859350204,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 51723
+    },
+    {
+      "epoch": 5.312313854369929,
+      "grad_norm": 0.12248563021421432,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 51726
+    },
+    {
+      "epoch": 5.31262195748177,
+      "grad_norm": 0.1178579330444336,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 51729
+    },
+    {
+      "epoch": 5.312930060593612,
+      "grad_norm": 0.042379625141620636,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 51732
+    },
+    {
+      "epoch": 5.313238163705454,
+      "grad_norm": 0.05351175367832184,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 51735
+    },
+    {
+      "epoch": 5.313546266817295,
+      "grad_norm": 0.03492636978626251,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 51738
+    },
+    {
+      "epoch": 5.313854369929136,
+      "grad_norm": 0.037354279309511185,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 51741
+    },
+    {
+      "epoch": 5.3141624730409776,
+      "grad_norm": 0.0453021377325058,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 51744
+    },
+    {
+      "epoch": 5.314470576152819,
+      "grad_norm": 0.040230412036180496,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 51747
+    },
+    {
+      "epoch": 5.314778679264661,
+      "grad_norm": 0.1004524752497673,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 51750
+    },
+    {
+      "epoch": 5.315086782376502,
+      "grad_norm": 0.08435871452093124,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 51753
+    },
+    {
+      "epoch": 5.3153948854883435,
+      "grad_norm": 0.04728049412369728,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 51756
+    },
+    {
+      "epoch": 5.315702988600185,
+      "grad_norm": 0.043835487216711044,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 51759
+    },
+    {
+      "epoch": 5.316011091712026,
+      "grad_norm": 0.03761078417301178,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 51762
+    },
+    {
+      "epoch": 5.316319194823867,
+      "grad_norm": 0.11318635195493698,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 51765
+    },
+    {
+      "epoch": 5.316627297935709,
+      "grad_norm": 0.06980354338884354,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 51768
+    },
+    {
+      "epoch": 5.316935401047551,
+      "grad_norm": 0.09948752075433731,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 51771
+    },
+    {
+      "epoch": 5.317243504159392,
+      "grad_norm": 0.07391460239887238,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 51774
+    },
+    {
+      "epoch": 5.317551607271233,
+      "grad_norm": 0.06508208811283112,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 51777
+    },
+    {
+      "epoch": 5.3178597103830745,
+      "grad_norm": 0.0692305937409401,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 51780
+    },
+    {
+      "epoch": 5.318167813494917,
+      "grad_norm": 0.07129498571157455,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 51783
+    },
+    {
+      "epoch": 5.318475916606758,
+      "grad_norm": 0.05130002647638321,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 51786
+    },
+    {
+      "epoch": 5.318784019718599,
+      "grad_norm": 0.07163992524147034,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 51789
+    },
+    {
+      "epoch": 5.31909212283044,
+      "grad_norm": 0.09687676280736923,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 51792
+    },
+    {
+      "epoch": 5.319400225942282,
+      "grad_norm": 0.08195064216852188,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 51795
+    },
+    {
+      "epoch": 5.319708329054124,
+      "grad_norm": 0.05641806125640869,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 51798
+    },
+    {
+      "epoch": 5.320016432165965,
+      "grad_norm": 0.06832972913980484,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 51801
+    },
+    {
+      "epoch": 5.320324535277806,
+      "grad_norm": 0.06402765214443207,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 51804
+    },
+    {
+      "epoch": 5.320632638389648,
+      "grad_norm": 0.045856647193431854,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 51807
+    },
+    {
+      "epoch": 5.320940741501489,
+      "grad_norm": 0.10627341270446777,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 51810
+    },
+    {
+      "epoch": 5.321248844613331,
+      "grad_norm": 0.07530563324689865,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 51813
+    },
+    {
+      "epoch": 5.321556947725172,
+      "grad_norm": 0.06049705296754837,
+      "learning_rate": 0.01,
+      "loss": 2.0227,
+      "step": 51816
+    },
+    {
+      "epoch": 5.3218650508370136,
+      "grad_norm": 0.05264494568109512,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 51819
+    },
+    {
+      "epoch": 5.322173153948855,
+      "grad_norm": 0.03524477034807205,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 51822
+    },
+    {
+      "epoch": 5.322481257060696,
+      "grad_norm": 0.03644990921020508,
+      "learning_rate": 0.01,
+      "loss": 2.0254,
+      "step": 51825
+    },
+    {
+      "epoch": 5.322789360172537,
+      "grad_norm": 0.10562828928232193,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 51828
+    },
+    {
+      "epoch": 5.3230974632843795,
+      "grad_norm": 0.06203228980302811,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 51831
+    },
+    {
+      "epoch": 5.323405566396221,
+      "grad_norm": 0.06565918028354645,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 51834
+    },
+    {
+      "epoch": 5.323713669508062,
+      "grad_norm": 0.03533428907394409,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 51837
+    },
+    {
+      "epoch": 5.324021772619903,
+      "grad_norm": 0.054743893444538116,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 51840
+    },
+    {
+      "epoch": 5.3243298757317445,
+      "grad_norm": 0.04253426194190979,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 51843
+    },
+    {
+      "epoch": 5.324637978843587,
+      "grad_norm": 0.05230151116847992,
+      "learning_rate": 0.01,
+      "loss": 2.0335,
+      "step": 51846
+    },
+    {
+      "epoch": 5.324946081955428,
+      "grad_norm": 0.08154330402612686,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 51849
+    },
+    {
+      "epoch": 5.325254185067269,
+      "grad_norm": 0.0635039433836937,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 51852
+    },
+    {
+      "epoch": 5.3255622881791105,
+      "grad_norm": 0.07905185222625732,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 51855
+    },
+    {
+      "epoch": 5.325870391290952,
+      "grad_norm": 0.04034522920846939,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 51858
+    },
+    {
+      "epoch": 5.326178494402793,
+      "grad_norm": 0.045876212418079376,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 51861
+    },
+    {
+      "epoch": 5.326486597514635,
+      "grad_norm": 0.03529779985547066,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 51864
+    },
+    {
+      "epoch": 5.326794700626476,
+      "grad_norm": 0.04534045606851578,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 51867
+    },
+    {
+      "epoch": 5.327102803738318,
+      "grad_norm": 0.126200869679451,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 51870
+    },
+    {
+      "epoch": 5.327410906850159,
+      "grad_norm": 0.052106015384197235,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 51873
+    },
+    {
+      "epoch": 5.327719009962,
+      "grad_norm": 0.048572082072496414,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 51876
+    },
+    {
+      "epoch": 5.328027113073842,
+      "grad_norm": 0.057247593998909,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 51879
+    },
+    {
+      "epoch": 5.328335216185684,
+      "grad_norm": 0.05806770548224449,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 51882
+    },
+    {
+      "epoch": 5.328643319297525,
+      "grad_norm": 0.13235436379909515,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 51885
+    },
+    {
+      "epoch": 5.328951422409366,
+      "grad_norm": 0.05209680646657944,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 51888
+    },
+    {
+      "epoch": 5.329259525521207,
+      "grad_norm": 0.04446623846888542,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 51891
+    },
+    {
+      "epoch": 5.32956762863305,
+      "grad_norm": 0.0674331933259964,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 51894
+    },
+    {
+      "epoch": 5.329875731744891,
+      "grad_norm": 0.11401736736297607,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 51897
+    },
+    {
+      "epoch": 5.330183834856732,
+      "grad_norm": 0.08108768612146378,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 51900
+    },
+    {
+      "epoch": 5.330491937968573,
+      "grad_norm": 0.06601007282733917,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 51903
+    },
+    {
+      "epoch": 5.330800041080415,
+      "grad_norm": 0.05304531753063202,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 51906
+    },
+    {
+      "epoch": 5.331108144192257,
+      "grad_norm": 0.04129822179675102,
+      "learning_rate": 0.01,
+      "loss": 2.0316,
+      "step": 51909
+    },
+    {
+      "epoch": 5.331416247304098,
+      "grad_norm": 0.06261187046766281,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 51912
+    },
+    {
+      "epoch": 5.331724350415939,
+      "grad_norm": 0.06908193230628967,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 51915
+    },
+    {
+      "epoch": 5.3320324535277805,
+      "grad_norm": 0.10063374042510986,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 51918
+    },
+    {
+      "epoch": 5.332340556639622,
+      "grad_norm": 0.11441531032323837,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 51921
+    },
+    {
+      "epoch": 5.332648659751463,
+      "grad_norm": 0.058185406029224396,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 51924
+    },
+    {
+      "epoch": 5.332956762863305,
+      "grad_norm": 0.041803937405347824,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 51927
+    },
+    {
+      "epoch": 5.3332648659751465,
+      "grad_norm": 0.12652553617954254,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 51930
+    },
+    {
+      "epoch": 5.333572969086988,
+      "grad_norm": 0.04237103462219238,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 51933
+    },
+    {
+      "epoch": 5.333881072198829,
+      "grad_norm": 0.052531465888023376,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 51936
+    },
+    {
+      "epoch": 5.33418917531067,
+      "grad_norm": 0.0361291877925396,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 51939
+    },
+    {
+      "epoch": 5.334497278422512,
+      "grad_norm": 0.17653311789035797,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 51942
+    },
+    {
+      "epoch": 5.334805381534354,
+      "grad_norm": 0.04208230972290039,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 51945
+    },
+    {
+      "epoch": 5.335113484646195,
+      "grad_norm": 0.05036168918013573,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 51948
+    },
+    {
+      "epoch": 5.335421587758036,
+      "grad_norm": 0.032406702637672424,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 51951
+    },
+    {
+      "epoch": 5.3357296908698775,
+      "grad_norm": 0.03707049414515495,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 51954
+    },
+    {
+      "epoch": 5.33603779398172,
+      "grad_norm": 0.10224796086549759,
+      "learning_rate": 0.01,
+      "loss": 2.0212,
+      "step": 51957
+    },
+    {
+      "epoch": 5.336345897093561,
+      "grad_norm": 0.07522287964820862,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 51960
+    },
+    {
+      "epoch": 5.336654000205402,
+      "grad_norm": 0.08081628382205963,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 51963
+    },
+    {
+      "epoch": 5.336962103317243,
+      "grad_norm": 0.03779719024896622,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 51966
+    },
+    {
+      "epoch": 5.337270206429085,
+      "grad_norm": 0.05938563495874405,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 51969
+    },
+    {
+      "epoch": 5.337578309540927,
+      "grad_norm": 0.058824703097343445,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 51972
+    },
+    {
+      "epoch": 5.337886412652768,
+      "grad_norm": 0.09814310073852539,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 51975
+    },
+    {
+      "epoch": 5.338194515764609,
+      "grad_norm": 0.038881540298461914,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 51978
+    },
+    {
+      "epoch": 5.338502618876451,
+      "grad_norm": 0.06858659535646439,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 51981
+    },
+    {
+      "epoch": 5.338810721988292,
+      "grad_norm": 0.05999214947223663,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 51984
+    },
+    {
+      "epoch": 5.339118825100133,
+      "grad_norm": 0.13660067319869995,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 51987
+    },
+    {
+      "epoch": 5.339426928211975,
+      "grad_norm": 0.039648644626140594,
+      "learning_rate": 0.01,
+      "loss": 2.0287,
+      "step": 51990
+    },
+    {
+      "epoch": 5.3397350313238165,
+      "grad_norm": 0.04098828509449959,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 51993
+    },
+    {
+      "epoch": 5.340043134435658,
+      "grad_norm": 0.07291056960821152,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 51996
+    },
+    {
+      "epoch": 5.340351237547499,
+      "grad_norm": 0.07260385155677795,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 51999
+    },
+    {
+      "epoch": 5.34065934065934,
+      "grad_norm": 0.08830993622541428,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 52002
+    },
+    {
+      "epoch": 5.3409674437711825,
+      "grad_norm": 0.053973227739334106,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 52005
+    },
+    {
+      "epoch": 5.341275546883024,
+      "grad_norm": 0.06421860307455063,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 52008
+    },
+    {
+      "epoch": 5.341583649994865,
+      "grad_norm": 0.06905265897512436,
+      "learning_rate": 0.01,
+      "loss": 2.0258,
+      "step": 52011
+    },
+    {
+      "epoch": 5.341891753106706,
+      "grad_norm": 0.1201087012887001,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 52014
+    },
+    {
+      "epoch": 5.3421998562185475,
+      "grad_norm": 0.0491718128323555,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 52017
+    },
+    {
+      "epoch": 5.342507959330389,
+      "grad_norm": 0.04123944044113159,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 52020
+    },
+    {
+      "epoch": 5.342816062442231,
+      "grad_norm": 0.059747952967882156,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 52023
+    },
+    {
+      "epoch": 5.343124165554072,
+      "grad_norm": 0.07639419287443161,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 52026
+    },
+    {
+      "epoch": 5.3434322686659135,
+      "grad_norm": 0.04553646221756935,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 52029
+    },
+    {
+      "epoch": 5.343740371777755,
+      "grad_norm": 0.05324438959360123,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 52032
+    },
+    {
+      "epoch": 5.344048474889596,
+      "grad_norm": 0.07225726544857025,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 52035
+    },
+    {
+      "epoch": 5.344356578001438,
+      "grad_norm": 0.10456754267215729,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 52038
+    },
+    {
+      "epoch": 5.344664681113279,
+      "grad_norm": 0.06349442899227142,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 52041
+    },
+    {
+      "epoch": 5.344972784225121,
+      "grad_norm": 0.08539248257875443,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 52044
+    },
+    {
+      "epoch": 5.345280887336962,
+      "grad_norm": 0.04877861216664314,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 52047
+    },
+    {
+      "epoch": 5.345588990448803,
+      "grad_norm": 0.0842355415225029,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 52050
+    },
+    {
+      "epoch": 5.345897093560645,
+      "grad_norm": 0.03320346772670746,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 52053
+    },
+    {
+      "epoch": 5.346205196672487,
+      "grad_norm": 0.09604234993457794,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 52056
+    },
+    {
+      "epoch": 5.346513299784328,
+      "grad_norm": 0.04435954615473747,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 52059
+    },
+    {
+      "epoch": 5.346821402896169,
+      "grad_norm": 0.07951738685369492,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 52062
+    },
+    {
+      "epoch": 5.34712950600801,
+      "grad_norm": 0.05883636698126793,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 52065
+    },
+    {
+      "epoch": 5.3474376091198526,
+      "grad_norm": 0.08984318375587463,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 52068
+    },
+    {
+      "epoch": 5.347745712231694,
+      "grad_norm": 0.11542551219463348,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 52071
+    },
+    {
+      "epoch": 5.348053815343535,
+      "grad_norm": 0.07631759345531464,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 52074
+    },
+    {
+      "epoch": 5.348361918455376,
+      "grad_norm": 0.05458938702940941,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 52077
+    },
+    {
+      "epoch": 5.348670021567218,
+      "grad_norm": 0.03696439042687416,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 52080
+    },
+    {
+      "epoch": 5.348978124679059,
+      "grad_norm": 0.05871397256851196,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 52083
+    },
+    {
+      "epoch": 5.349286227790901,
+      "grad_norm": 0.046202853322029114,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 52086
+    },
+    {
+      "epoch": 5.349594330902742,
+      "grad_norm": 0.035825759172439575,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 52089
+    },
+    {
+      "epoch": 5.3499024340145835,
+      "grad_norm": 0.12834282219409943,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 52092
+    },
+    {
+      "epoch": 5.350210537126425,
+      "grad_norm": 0.05964813381433487,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 52095
+    },
+    {
+      "epoch": 5.350518640238266,
+      "grad_norm": 0.0718245729804039,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 52098
+    },
+    {
+      "epoch": 5.350826743350108,
+      "grad_norm": 0.043250247836112976,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 52101
+    },
+    {
+      "epoch": 5.3511348464619495,
+      "grad_norm": 0.0491093210875988,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 52104
+    },
+    {
+      "epoch": 5.351442949573791,
+      "grad_norm": 0.06597597897052765,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 52107
+    },
+    {
+      "epoch": 5.351751052685632,
+      "grad_norm": 0.1260969340801239,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 52110
+    },
+    {
+      "epoch": 5.352059155797473,
+      "grad_norm": 0.04004066437482834,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 52113
+    },
+    {
+      "epoch": 5.3523672589093145,
+      "grad_norm": 0.03897303342819214,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 52116
+    },
+    {
+      "epoch": 5.352675362021157,
+      "grad_norm": 0.05720753222703934,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 52119
+    },
+    {
+      "epoch": 5.352983465132998,
+      "grad_norm": 0.09778839349746704,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 52122
+    },
+    {
+      "epoch": 5.353291568244839,
+      "grad_norm": 0.10825862735509872,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 52125
+    },
+    {
+      "epoch": 5.3535996713566805,
+      "grad_norm": 0.06417248398065567,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 52128
+    },
+    {
+      "epoch": 5.353907774468522,
+      "grad_norm": 0.03844156488776207,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 52131
+    },
+    {
+      "epoch": 5.354215877580364,
+      "grad_norm": 0.05346846207976341,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 52134
+    },
+    {
+      "epoch": 5.354523980692205,
+      "grad_norm": 0.07128198444843292,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 52137
+    },
+    {
+      "epoch": 5.354832083804046,
+      "grad_norm": 0.03174986317753792,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 52140
+    },
+    {
+      "epoch": 5.355140186915888,
+      "grad_norm": 0.044908635318279266,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 52143
+    },
+    {
+      "epoch": 5.355448290027729,
+      "grad_norm": 0.14737063646316528,
+      "learning_rate": 0.01,
+      "loss": 2.0202,
+      "step": 52146
+    },
+    {
+      "epoch": 5.355756393139571,
+      "grad_norm": 0.08280967175960541,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 52149
+    },
+    {
+      "epoch": 5.356064496251412,
+      "grad_norm": 0.037819184362888336,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 52152
+    },
+    {
+      "epoch": 5.356372599363254,
+      "grad_norm": 0.03838292881846428,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 52155
+    },
+    {
+      "epoch": 5.356680702475095,
+      "grad_norm": 0.04593195393681526,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 52158
+    },
+    {
+      "epoch": 5.356988805586936,
+      "grad_norm": 0.12287592142820358,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 52161
+    },
+    {
+      "epoch": 5.357296908698778,
+      "grad_norm": 0.05081895366311073,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 52164
+    },
+    {
+      "epoch": 5.3576050118106195,
+      "grad_norm": 0.15700340270996094,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 52167
+    },
+    {
+      "epoch": 5.357913114922461,
+      "grad_norm": 0.049683380872011185,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 52170
+    },
+    {
+      "epoch": 5.358221218034302,
+      "grad_norm": 0.03398337587714195,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 52173
+    },
+    {
+      "epoch": 5.358529321146143,
+      "grad_norm": 0.04002157971262932,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 52176
+    },
+    {
+      "epoch": 5.358837424257985,
+      "grad_norm": 0.037759970873594284,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 52179
+    },
+    {
+      "epoch": 5.359145527369827,
+      "grad_norm": 0.04797700420022011,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 52182
+    },
+    {
+      "epoch": 5.359453630481668,
+      "grad_norm": 0.06852797418832779,
+      "learning_rate": 0.01,
+      "loss": 2.0288,
+      "step": 52185
+    },
+    {
+      "epoch": 5.359761733593509,
+      "grad_norm": 0.04743816703557968,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 52188
+    },
+    {
+      "epoch": 5.3600698367053505,
+      "grad_norm": 0.05316634103655815,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 52191
+    },
+    {
+      "epoch": 5.360377939817192,
+      "grad_norm": 0.07262757420539856,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 52194
+    },
+    {
+      "epoch": 5.360686042929034,
+      "grad_norm": 0.13863062858581543,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 52197
+    },
+    {
+      "epoch": 5.360994146040875,
+      "grad_norm": 0.0368843711912632,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 52200
+    },
+    {
+      "epoch": 5.3613022491527165,
+      "grad_norm": 0.05429408326745033,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 52203
+    },
+    {
+      "epoch": 5.361610352264558,
+      "grad_norm": 0.04265204817056656,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 52206
+    },
+    {
+      "epoch": 5.361918455376399,
+      "grad_norm": 0.04122883826494217,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 52209
+    },
+    {
+      "epoch": 5.362226558488241,
+      "grad_norm": 0.04695272073149681,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 52212
+    },
+    {
+      "epoch": 5.362534661600082,
+      "grad_norm": 0.04153517633676529,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 52215
+    },
+    {
+      "epoch": 5.362842764711924,
+      "grad_norm": 0.09016060084104538,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 52218
+    },
+    {
+      "epoch": 5.363150867823765,
+      "grad_norm": 0.06228369474411011,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 52221
+    },
+    {
+      "epoch": 5.363458970935606,
+      "grad_norm": 0.09304657578468323,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 52224
+    },
+    {
+      "epoch": 5.363767074047448,
+      "grad_norm": 0.07483693957328796,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 52227
+    },
+    {
+      "epoch": 5.36407517715929,
+      "grad_norm": 0.08632933348417282,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 52230
+    },
+    {
+      "epoch": 5.364383280271131,
+      "grad_norm": 0.08443045616149902,
+      "learning_rate": 0.01,
+      "loss": 2.0161,
+      "step": 52233
+    },
+    {
+      "epoch": 5.364691383382972,
+      "grad_norm": 0.05415327101945877,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 52236
+    },
+    {
+      "epoch": 5.364999486494813,
+      "grad_norm": 0.048056915402412415,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 52239
+    },
+    {
+      "epoch": 5.365307589606655,
+      "grad_norm": 0.03430894389748573,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 52242
+    },
+    {
+      "epoch": 5.365615692718497,
+      "grad_norm": 0.10147974640130997,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 52245
+    },
+    {
+      "epoch": 5.365923795830338,
+      "grad_norm": 0.09672293066978455,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 52248
+    },
+    {
+      "epoch": 5.366231898942179,
+      "grad_norm": 0.08904405683279037,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 52251
+    },
+    {
+      "epoch": 5.366540002054021,
+      "grad_norm": 0.05163189023733139,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 52254
+    },
+    {
+      "epoch": 5.366848105165862,
+      "grad_norm": 0.03694668784737587,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 52257
+    },
+    {
+      "epoch": 5.367156208277704,
+      "grad_norm": 0.036716192960739136,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 52260
+    },
+    {
+      "epoch": 5.367464311389545,
+      "grad_norm": 0.046369098126888275,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 52263
+    },
+    {
+      "epoch": 5.3677724145013865,
+      "grad_norm": 0.06647578626871109,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 52266
+    },
+    {
+      "epoch": 5.368080517613228,
+      "grad_norm": 0.11546402424573898,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 52269
+    },
+    {
+      "epoch": 5.368388620725069,
+      "grad_norm": 0.08381645381450653,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 52272
+    },
+    {
+      "epoch": 5.36869672383691,
+      "grad_norm": 0.09374178946018219,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 52275
+    },
+    {
+      "epoch": 5.3690048269487525,
+      "grad_norm": 0.13528801500797272,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 52278
+    },
+    {
+      "epoch": 5.369312930060594,
+      "grad_norm": 0.09537555277347565,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 52281
+    },
+    {
+      "epoch": 5.369621033172435,
+      "grad_norm": 0.06615101546049118,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 52284
+    },
+    {
+      "epoch": 5.369929136284276,
+      "grad_norm": 0.03951498121023178,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 52287
+    },
+    {
+      "epoch": 5.3702372393961175,
+      "grad_norm": 0.04514235630631447,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 52290
+    },
+    {
+      "epoch": 5.37054534250796,
+      "grad_norm": 0.05101289227604866,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 52293
+    },
+    {
+      "epoch": 5.370853445619801,
+      "grad_norm": 0.10146338492631912,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 52296
+    },
+    {
+      "epoch": 5.371161548731642,
+      "grad_norm": 0.06167877838015556,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 52299
+    },
+    {
+      "epoch": 5.3714696518434835,
+      "grad_norm": 0.10302039980888367,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 52302
+    },
+    {
+      "epoch": 5.371777754955325,
+      "grad_norm": 0.07716668397188187,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 52305
+    },
+    {
+      "epoch": 5.372085858067167,
+      "grad_norm": 0.08410248160362244,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 52308
+    },
+    {
+      "epoch": 5.372393961179008,
+      "grad_norm": 0.05326802283525467,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 52311
+    },
+    {
+      "epoch": 5.372702064290849,
+      "grad_norm": 0.08413115888834,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 52314
+    },
+    {
+      "epoch": 5.373010167402691,
+      "grad_norm": 0.053719133138656616,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 52317
+    },
+    {
+      "epoch": 5.373318270514532,
+      "grad_norm": 0.07752490788698196,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 52320
+    },
+    {
+      "epoch": 5.373626373626374,
+      "grad_norm": 0.0719638466835022,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 52323
+    },
+    {
+      "epoch": 5.373934476738215,
+      "grad_norm": 0.0800856202840805,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 52326
+    },
+    {
+      "epoch": 5.374242579850057,
+      "grad_norm": 0.0889526829123497,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 52329
+    },
+    {
+      "epoch": 5.374550682961898,
+      "grad_norm": 0.08525931090116501,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 52332
+    },
+    {
+      "epoch": 5.374858786073739,
+      "grad_norm": 0.038573917001485825,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 52335
+    },
+    {
+      "epoch": 5.37516688918558,
+      "grad_norm": 0.03686724230647087,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 52338
+    },
+    {
+      "epoch": 5.3754749922974225,
+      "grad_norm": 0.042029861360788345,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 52341
+    },
+    {
+      "epoch": 5.375783095409264,
+      "grad_norm": 0.08990278095006943,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 52344
+    },
+    {
+      "epoch": 5.376091198521105,
+      "grad_norm": 0.09406303614377975,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 52347
+    },
+    {
+      "epoch": 5.376399301632946,
+      "grad_norm": 0.1285853236913681,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 52350
+    },
+    {
+      "epoch": 5.376707404744788,
+      "grad_norm": 0.06205267086625099,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 52353
+    },
+    {
+      "epoch": 5.37701550785663,
+      "grad_norm": 0.09469316154718399,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 52356
+    },
+    {
+      "epoch": 5.377323610968471,
+      "grad_norm": 0.08087198436260223,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 52359
+    },
+    {
+      "epoch": 5.377631714080312,
+      "grad_norm": 0.10511235147714615,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 52362
+    },
+    {
+      "epoch": 5.3779398171921535,
+      "grad_norm": 0.07102752476930618,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 52365
+    },
+    {
+      "epoch": 5.378247920303995,
+      "grad_norm": 0.04023940861225128,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 52368
+    },
+    {
+      "epoch": 5.378556023415836,
+      "grad_norm": 0.06366323679685593,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 52371
+    },
+    {
+      "epoch": 5.378864126527678,
+      "grad_norm": 0.045444704592227936,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 52374
+    },
+    {
+      "epoch": 5.3791722296395195,
+      "grad_norm": 0.07034427672624588,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 52377
+    },
+    {
+      "epoch": 5.379480332751361,
+      "grad_norm": 0.04844088852405548,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 52380
+    },
+    {
+      "epoch": 5.379788435863202,
+      "grad_norm": 0.05628515034914017,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 52383
+    },
+    {
+      "epoch": 5.380096538975043,
+      "grad_norm": 0.10569090396165848,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 52386
+    },
+    {
+      "epoch": 5.380404642086885,
+      "grad_norm": 0.14752137660980225,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 52389
+    },
+    {
+      "epoch": 5.380712745198727,
+      "grad_norm": 0.09316124767065048,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 52392
+    },
+    {
+      "epoch": 5.381020848310568,
+      "grad_norm": 0.0789172351360321,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 52395
+    },
+    {
+      "epoch": 5.381328951422409,
+      "grad_norm": 0.04636065661907196,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 52398
+    },
+    {
+      "epoch": 5.3816370545342505,
+      "grad_norm": 0.051946815103292465,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 52401
+    },
+    {
+      "epoch": 5.381945157646093,
+      "grad_norm": 0.0423317514359951,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 52404
+    },
+    {
+      "epoch": 5.382253260757934,
+      "grad_norm": 0.03624933212995529,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 52407
+    },
+    {
+      "epoch": 5.382561363869775,
+      "grad_norm": 0.0415869876742363,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 52410
+    },
+    {
+      "epoch": 5.382869466981616,
+      "grad_norm": 0.1304771602153778,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 52413
+    },
+    {
+      "epoch": 5.383177570093458,
+      "grad_norm": 0.05917195603251457,
+      "learning_rate": 0.01,
+      "loss": 2.0216,
+      "step": 52416
+    },
+    {
+      "epoch": 5.3834856732053,
+      "grad_norm": 0.04846251383423805,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 52419
+    },
+    {
+      "epoch": 5.383793776317141,
+      "grad_norm": 0.03447849303483963,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 52422
+    },
+    {
+      "epoch": 5.384101879428982,
+      "grad_norm": 0.042299896478652954,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 52425
+    },
+    {
+      "epoch": 5.384409982540824,
+      "grad_norm": 0.053316473960876465,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 52428
+    },
+    {
+      "epoch": 5.384718085652665,
+      "grad_norm": 0.05238528177142143,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 52431
+    },
+    {
+      "epoch": 5.385026188764506,
+      "grad_norm": 0.05088873207569122,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 52434
+    },
+    {
+      "epoch": 5.385334291876348,
+      "grad_norm": 0.03723147138953209,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 52437
+    },
+    {
+      "epoch": 5.3856423949881895,
+      "grad_norm": 0.04240123927593231,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 52440
+    },
+    {
+      "epoch": 5.385950498100031,
+      "grad_norm": 0.04458059370517731,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 52443
+    },
+    {
+      "epoch": 5.386258601211872,
+      "grad_norm": 0.20143818855285645,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 52446
+    },
+    {
+      "epoch": 5.386566704323713,
+      "grad_norm": 0.05507873743772507,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 52449
+    },
+    {
+      "epoch": 5.3868748074355555,
+      "grad_norm": 0.040351588279008865,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 52452
+    },
+    {
+      "epoch": 5.387182910547397,
+      "grad_norm": 0.044757384806871414,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 52455
+    },
+    {
+      "epoch": 5.387491013659238,
+      "grad_norm": 0.05201313644647598,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 52458
+    },
+    {
+      "epoch": 5.387799116771079,
+      "grad_norm": 0.053258076310157776,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 52461
+    },
+    {
+      "epoch": 5.3881072198829205,
+      "grad_norm": 0.08738038688898087,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 52464
+    },
+    {
+      "epoch": 5.388415322994762,
+      "grad_norm": 0.06154690682888031,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 52467
+    },
+    {
+      "epoch": 5.388723426106604,
+      "grad_norm": 0.045472290366888046,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 52470
+    },
+    {
+      "epoch": 5.389031529218445,
+      "grad_norm": 0.05027802661061287,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 52473
+    },
+    {
+      "epoch": 5.3893396323302865,
+      "grad_norm": 0.06531531363725662,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 52476
+    },
+    {
+      "epoch": 5.389647735442128,
+      "grad_norm": 0.0642043873667717,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 52479
+    },
+    {
+      "epoch": 5.389955838553969,
+      "grad_norm": 0.03432301804423332,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 52482
+    },
+    {
+      "epoch": 5.390263941665811,
+      "grad_norm": 0.14385442435741425,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 52485
+    },
+    {
+      "epoch": 5.390572044777652,
+      "grad_norm": 0.05068526789546013,
+      "learning_rate": 0.01,
+      "loss": 2.0122,
+      "step": 52488
+    },
+    {
+      "epoch": 5.390880147889494,
+      "grad_norm": 0.0718868151307106,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 52491
+    },
+    {
+      "epoch": 5.391188251001335,
+      "grad_norm": 0.06522323191165924,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 52494
+    },
+    {
+      "epoch": 5.391496354113176,
+      "grad_norm": 0.03200114890933037,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 52497
+    },
+    {
+      "epoch": 5.391804457225018,
+      "grad_norm": 0.11630173027515411,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 52500
+    },
+    {
+      "epoch": 5.39211256033686,
+      "grad_norm": 0.07612688839435577,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 52503
+    },
+    {
+      "epoch": 5.392420663448701,
+      "grad_norm": 0.09313171356916428,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 52506
+    },
+    {
+      "epoch": 5.392728766560542,
+      "grad_norm": 0.07034334540367126,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 52509
+    },
+    {
+      "epoch": 5.393036869672383,
+      "grad_norm": 0.07270947843790054,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 52512
+    },
+    {
+      "epoch": 5.3933449727842255,
+      "grad_norm": 0.051841702312231064,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 52515
+    },
+    {
+      "epoch": 5.393653075896067,
+      "grad_norm": 0.04291679337620735,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 52518
+    },
+    {
+      "epoch": 5.393961179007908,
+      "grad_norm": 0.03652627766132355,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 52521
+    },
+    {
+      "epoch": 5.394269282119749,
+      "grad_norm": 0.03804773837327957,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 52524
+    },
+    {
+      "epoch": 5.394577385231591,
+      "grad_norm": 0.05173708498477936,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 52527
+    },
+    {
+      "epoch": 5.394885488343432,
+      "grad_norm": 0.07177083939313889,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 52530
+    },
+    {
+      "epoch": 5.395193591455274,
+      "grad_norm": 0.08862923830747604,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 52533
+    },
+    {
+      "epoch": 5.395501694567115,
+      "grad_norm": 0.09389739483594894,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 52536
+    },
+    {
+      "epoch": 5.3958097976789565,
+      "grad_norm": 0.03884744644165039,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 52539
+    },
+    {
+      "epoch": 5.396117900790798,
+      "grad_norm": 0.04176723584532738,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 52542
+    },
+    {
+      "epoch": 5.396426003902639,
+      "grad_norm": 0.04131205379962921,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 52545
+    },
+    {
+      "epoch": 5.396734107014481,
+      "grad_norm": 0.03634188696742058,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 52548
+    },
+    {
+      "epoch": 5.3970422101263225,
+      "grad_norm": 0.09134528040885925,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 52551
+    },
+    {
+      "epoch": 5.397350313238164,
+      "grad_norm": 0.07570212334394455,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 52554
+    },
+    {
+      "epoch": 5.397658416350005,
+      "grad_norm": 0.11560734361410141,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 52557
+    },
+    {
+      "epoch": 5.397966519461846,
+      "grad_norm": 0.03760908916592598,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 52560
+    },
+    {
+      "epoch": 5.398274622573688,
+      "grad_norm": 0.05205194652080536,
+      "learning_rate": 0.01,
+      "loss": 2.0133,
+      "step": 52563
+    },
+    {
+      "epoch": 5.39858272568553,
+      "grad_norm": 0.07257603853940964,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 52566
+    },
+    {
+      "epoch": 5.398890828797371,
+      "grad_norm": 0.07266250997781754,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 52569
+    },
+    {
+      "epoch": 5.399198931909212,
+      "grad_norm": 0.0367962047457695,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 52572
+    },
+    {
+      "epoch": 5.3995070350210534,
+      "grad_norm": 0.05613946169614792,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 52575
+    },
+    {
+      "epoch": 5.399815138132896,
+      "grad_norm": 0.10363534092903137,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 52578
+    },
+    {
+      "epoch": 5.400123241244737,
+      "grad_norm": 0.09217333793640137,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 52581
+    },
+    {
+      "epoch": 5.400431344356578,
+      "grad_norm": 0.0713183805346489,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 52584
+    },
+    {
+      "epoch": 5.400739447468419,
+      "grad_norm": 0.0670512244105339,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 52587
+    },
+    {
+      "epoch": 5.401047550580261,
+      "grad_norm": 0.10012122988700867,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 52590
+    },
+    {
+      "epoch": 5.401355653692102,
+      "grad_norm": 0.047828782349824905,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 52593
+    },
+    {
+      "epoch": 5.401663756803944,
+      "grad_norm": 0.09758254140615463,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 52596
+    },
+    {
+      "epoch": 5.401971859915785,
+      "grad_norm": 0.044294700026512146,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 52599
+    },
+    {
+      "epoch": 5.402279963027627,
+      "grad_norm": 0.04579133540391922,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 52602
+    },
+    {
+      "epoch": 5.402588066139468,
+      "grad_norm": 0.04279174283146858,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 52605
+    },
+    {
+      "epoch": 5.402896169251309,
+      "grad_norm": 0.08676068484783173,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 52608
+    },
+    {
+      "epoch": 5.403204272363151,
+      "grad_norm": 0.09505254775285721,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 52611
+    },
+    {
+      "epoch": 5.4035123754749925,
+      "grad_norm": 0.06927873194217682,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 52614
+    },
+    {
+      "epoch": 5.403820478586834,
+      "grad_norm": 0.09732890874147415,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 52617
+    },
+    {
+      "epoch": 5.404128581698675,
+      "grad_norm": 0.03702055662870407,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 52620
+    },
+    {
+      "epoch": 5.404436684810516,
+      "grad_norm": 0.08636049926280975,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 52623
+    },
+    {
+      "epoch": 5.404744787922358,
+      "grad_norm": 0.07209763675928116,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 52626
+    },
+    {
+      "epoch": 5.4050528910342,
+      "grad_norm": 0.05202876031398773,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 52629
+    },
+    {
+      "epoch": 5.405360994146041,
+      "grad_norm": 0.0709710493683815,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 52632
+    },
+    {
+      "epoch": 5.405669097257882,
+      "grad_norm": 0.03506077453494072,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 52635
+    },
+    {
+      "epoch": 5.4059772003697235,
+      "grad_norm": 0.04027107357978821,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 52638
+    },
+    {
+      "epoch": 5.406285303481565,
+      "grad_norm": 0.07768195122480392,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 52641
+    },
+    {
+      "epoch": 5.406593406593407,
+      "grad_norm": 0.13171711564064026,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 52644
+    },
+    {
+      "epoch": 5.406901509705248,
+      "grad_norm": 0.13514290750026703,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 52647
+    },
+    {
+      "epoch": 5.4072096128170895,
+      "grad_norm": 0.05603281781077385,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 52650
+    },
+    {
+      "epoch": 5.407517715928931,
+      "grad_norm": 0.07667485624551773,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 52653
+    },
+    {
+      "epoch": 5.407825819040772,
+      "grad_norm": 0.06463679671287537,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 52656
+    },
+    {
+      "epoch": 5.408133922152614,
+      "grad_norm": 0.06655814498662949,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 52659
+    },
+    {
+      "epoch": 5.408442025264455,
+      "grad_norm": 0.1041889637708664,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 52662
+    },
+    {
+      "epoch": 5.408750128376297,
+      "grad_norm": 0.06591400504112244,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 52665
+    },
+    {
+      "epoch": 5.409058231488138,
+      "grad_norm": 0.033431414514780045,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 52668
+    },
+    {
+      "epoch": 5.409366334599979,
+      "grad_norm": 0.04565748572349548,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 52671
+    },
+    {
+      "epoch": 5.409674437711821,
+      "grad_norm": 0.04200183227658272,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 52674
+    },
+    {
+      "epoch": 5.409982540823663,
+      "grad_norm": 0.03807186335325241,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 52677
+    },
+    {
+      "epoch": 5.410290643935504,
+      "grad_norm": 0.09112047404050827,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 52680
+    },
+    {
+      "epoch": 5.410598747047345,
+      "grad_norm": 0.10211756825447083,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 52683
+    },
+    {
+      "epoch": 5.410906850159186,
+      "grad_norm": 0.1345423012971878,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 52686
+    },
+    {
+      "epoch": 5.411214953271028,
+      "grad_norm": 0.07363910228013992,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 52689
+    },
+    {
+      "epoch": 5.41152305638287,
+      "grad_norm": 0.08829092979431152,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 52692
+    },
+    {
+      "epoch": 5.411831159494711,
+      "grad_norm": 0.06006823107600212,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 52695
+    },
+    {
+      "epoch": 5.412139262606552,
+      "grad_norm": 0.06111016124486923,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 52698
+    },
+    {
+      "epoch": 5.412447365718394,
+      "grad_norm": 0.057088688015937805,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 52701
+    },
+    {
+      "epoch": 5.412755468830235,
+      "grad_norm": 0.04059537127614021,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 52704
+    },
+    {
+      "epoch": 5.413063571942077,
+      "grad_norm": 0.2159087210893631,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 52707
+    },
+    {
+      "epoch": 5.413371675053918,
+      "grad_norm": 0.07151810079813004,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 52710
+    },
+    {
+      "epoch": 5.4136797781657595,
+      "grad_norm": 0.09309381246566772,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 52713
+    },
+    {
+      "epoch": 5.413987881277601,
+      "grad_norm": 0.039323315024375916,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 52716
+    },
+    {
+      "epoch": 5.414295984389442,
+      "grad_norm": 0.09738846868276596,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 52719
+    },
+    {
+      "epoch": 5.414604087501283,
+      "grad_norm": 0.07482850551605225,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 52722
+    },
+    {
+      "epoch": 5.4149121906131255,
+      "grad_norm": 0.05960662290453911,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 52725
+    },
+    {
+      "epoch": 5.415220293724967,
+      "grad_norm": 0.061568450182676315,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 52728
+    },
+    {
+      "epoch": 5.415528396836808,
+      "grad_norm": 0.05249075964093208,
+      "learning_rate": 0.01,
+      "loss": 2.0298,
+      "step": 52731
+    },
+    {
+      "epoch": 5.415836499948649,
+      "grad_norm": 0.04260426387190819,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 52734
+    },
+    {
+      "epoch": 5.4161446030604905,
+      "grad_norm": 0.035422783344984055,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 52737
+    },
+    {
+      "epoch": 5.416452706172333,
+      "grad_norm": 0.03902021422982216,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 52740
+    },
+    {
+      "epoch": 5.416760809284174,
+      "grad_norm": 0.04080606997013092,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 52743
+    },
+    {
+      "epoch": 5.417068912396015,
+      "grad_norm": 0.04044007509946823,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 52746
+    },
+    {
+      "epoch": 5.417377015507856,
+      "grad_norm": 0.04506811872124672,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 52749
+    },
+    {
+      "epoch": 5.417685118619698,
+      "grad_norm": 0.053388748317956924,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 52752
+    },
+    {
+      "epoch": 5.41799322173154,
+      "grad_norm": 0.09019593149423599,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 52755
+    },
+    {
+      "epoch": 5.418301324843381,
+      "grad_norm": 0.12063855677843094,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 52758
+    },
+    {
+      "epoch": 5.418609427955222,
+      "grad_norm": 0.07498898357152939,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 52761
+    },
+    {
+      "epoch": 5.418917531067064,
+      "grad_norm": 0.07565269619226456,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 52764
+    },
+    {
+      "epoch": 5.419225634178905,
+      "grad_norm": 0.037328194826841354,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 52767
+    },
+    {
+      "epoch": 5.419533737290747,
+      "grad_norm": 0.059995539486408234,
+      "learning_rate": 0.01,
+      "loss": 2.0149,
+      "step": 52770
+    },
+    {
+      "epoch": 5.419841840402588,
+      "grad_norm": 0.18053491413593292,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 52773
+    },
+    {
+      "epoch": 5.42014994351443,
+      "grad_norm": 0.1632491648197174,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 52776
+    },
+    {
+      "epoch": 5.420458046626271,
+      "grad_norm": 0.06863465160131454,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 52779
+    },
+    {
+      "epoch": 5.420766149738112,
+      "grad_norm": 0.05202522501349449,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 52782
+    },
+    {
+      "epoch": 5.421074252849953,
+      "grad_norm": 0.039888009428977966,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 52785
+    },
+    {
+      "epoch": 5.4213823559617955,
+      "grad_norm": 0.04122290015220642,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 52788
+    },
+    {
+      "epoch": 5.421690459073637,
+      "grad_norm": 0.03863810375332832,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 52791
+    },
+    {
+      "epoch": 5.421998562185478,
+      "grad_norm": 0.04215069115161896,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 52794
+    },
+    {
+      "epoch": 5.422306665297319,
+      "grad_norm": 0.09779973328113556,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 52797
+    },
+    {
+      "epoch": 5.422614768409161,
+      "grad_norm": 0.0620935894548893,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 52800
+    },
+    {
+      "epoch": 5.422922871521003,
+      "grad_norm": 0.05332612618803978,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 52803
+    },
+    {
+      "epoch": 5.423230974632844,
+      "grad_norm": 0.04671257734298706,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 52806
+    },
+    {
+      "epoch": 5.423539077744685,
+      "grad_norm": 0.03246668353676796,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 52809
+    },
+    {
+      "epoch": 5.4238471808565265,
+      "grad_norm": 0.04667939245700836,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 52812
+    },
+    {
+      "epoch": 5.424155283968368,
+      "grad_norm": 0.08633271604776382,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 52815
+    },
+    {
+      "epoch": 5.42446338708021,
+      "grad_norm": 0.07084902375936508,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 52818
+    },
+    {
+      "epoch": 5.424771490192051,
+      "grad_norm": 0.05388319492340088,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 52821
+    },
+    {
+      "epoch": 5.4250795933038924,
+      "grad_norm": 0.05652158334851265,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 52824
+    },
+    {
+      "epoch": 5.425387696415734,
+      "grad_norm": 0.09929019957780838,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 52827
+    },
+    {
+      "epoch": 5.425695799527575,
+      "grad_norm": 0.1725517362356186,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 52830
+    },
+    {
+      "epoch": 5.426003902639417,
+      "grad_norm": 0.06725597381591797,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 52833
+    },
+    {
+      "epoch": 5.426312005751258,
+      "grad_norm": 0.03839759901165962,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 52836
+    },
+    {
+      "epoch": 5.4266201088631,
+      "grad_norm": 0.03678586333990097,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 52839
+    },
+    {
+      "epoch": 5.426928211974941,
+      "grad_norm": 0.04450935125350952,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 52842
+    },
+    {
+      "epoch": 5.427236315086782,
+      "grad_norm": 0.05059666931629181,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 52845
+    },
+    {
+      "epoch": 5.427544418198623,
+      "grad_norm": 0.05548638850450516,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 52848
+    },
+    {
+      "epoch": 5.427852521310466,
+      "grad_norm": 0.03727211058139801,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 52851
+    },
+    {
+      "epoch": 5.428160624422307,
+      "grad_norm": 0.1027202233672142,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 52854
+    },
+    {
+      "epoch": 5.428468727534148,
+      "grad_norm": 0.07482005655765533,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 52857
+    },
+    {
+      "epoch": 5.428776830645989,
+      "grad_norm": 0.06325170397758484,
+      "learning_rate": 0.01,
+      "loss": 2.0173,
+      "step": 52860
+    },
+    {
+      "epoch": 5.429084933757831,
+      "grad_norm": 0.11125578731298447,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 52863
+    },
+    {
+      "epoch": 5.429393036869673,
+      "grad_norm": 0.07175584882497787,
+      "learning_rate": 0.01,
+      "loss": 2.0186,
+      "step": 52866
+    },
+    {
+      "epoch": 5.429701139981514,
+      "grad_norm": 0.03667625039815903,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 52869
+    },
+    {
+      "epoch": 5.430009243093355,
+      "grad_norm": 0.0515107661485672,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 52872
+    },
+    {
+      "epoch": 5.430317346205197,
+      "grad_norm": 0.09832940995693207,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 52875
+    },
+    {
+      "epoch": 5.430625449317038,
+      "grad_norm": 0.06182995066046715,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 52878
+    },
+    {
+      "epoch": 5.430933552428879,
+      "grad_norm": 0.1432095468044281,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 52881
+    },
+    {
+      "epoch": 5.431241655540721,
+      "grad_norm": 0.11832471191883087,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 52884
+    },
+    {
+      "epoch": 5.4315497586525625,
+      "grad_norm": 0.07653117924928665,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 52887
+    },
+    {
+      "epoch": 5.431857861764404,
+      "grad_norm": 0.06316263228654861,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 52890
+    },
+    {
+      "epoch": 5.432165964876245,
+      "grad_norm": 0.058272287249565125,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 52893
+    },
+    {
+      "epoch": 5.432474067988086,
+      "grad_norm": 0.06886950880289078,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 52896
+    },
+    {
+      "epoch": 5.4327821710999284,
+      "grad_norm": 0.053393036127090454,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 52899
+    },
+    {
+      "epoch": 5.43309027421177,
+      "grad_norm": 0.06651714444160461,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 52902
+    },
+    {
+      "epoch": 5.433398377323611,
+      "grad_norm": 0.04366718977689743,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 52905
+    },
+    {
+      "epoch": 5.433706480435452,
+      "grad_norm": 0.040242016315460205,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 52908
+    },
+    {
+      "epoch": 5.4340145835472935,
+      "grad_norm": 0.0598304383456707,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 52911
+    },
+    {
+      "epoch": 5.434322686659136,
+      "grad_norm": 0.17316143214702606,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 52914
+    },
+    {
+      "epoch": 5.434630789770977,
+      "grad_norm": 0.13133604824543,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 52917
+    },
+    {
+      "epoch": 5.434938892882818,
+      "grad_norm": 0.0751945972442627,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 52920
+    },
+    {
+      "epoch": 5.435246995994659,
+      "grad_norm": 0.06649640202522278,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 52923
+    },
+    {
+      "epoch": 5.435555099106501,
+      "grad_norm": 0.13051864504814148,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 52926
+    },
+    {
+      "epoch": 5.435863202218343,
+      "grad_norm": 0.06402359157800674,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 52929
+    },
+    {
+      "epoch": 5.436171305330184,
+      "grad_norm": 0.06441336125135422,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 52932
+    },
+    {
+      "epoch": 5.436479408442025,
+      "grad_norm": 0.09019657224416733,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 52935
+    },
+    {
+      "epoch": 5.436787511553867,
+      "grad_norm": 0.04703563451766968,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 52938
+    },
+    {
+      "epoch": 5.437095614665708,
+      "grad_norm": 0.03909967467188835,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 52941
+    },
+    {
+      "epoch": 5.437403717777549,
+      "grad_norm": 0.05746940150856972,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 52944
+    },
+    {
+      "epoch": 5.437711820889391,
+      "grad_norm": 0.11890576034784317,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 52947
+    },
+    {
+      "epoch": 5.438019924001233,
+      "grad_norm": 0.07725408673286438,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 52950
+    },
+    {
+      "epoch": 5.438328027113074,
+      "grad_norm": 0.05112382769584656,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 52953
+    },
+    {
+      "epoch": 5.438636130224915,
+      "grad_norm": 0.08786381781101227,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 52956
+    },
+    {
+      "epoch": 5.438944233336756,
+      "grad_norm": 0.07547960430383682,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 52959
+    },
+    {
+      "epoch": 5.4392523364485985,
+      "grad_norm": 0.06332848221063614,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 52962
+    },
+    {
+      "epoch": 5.43956043956044,
+      "grad_norm": 0.09106019139289856,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 52965
+    },
+    {
+      "epoch": 5.439868542672281,
+      "grad_norm": 0.1387149691581726,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 52968
+    },
+    {
+      "epoch": 5.440176645784122,
+      "grad_norm": 0.07359499484300613,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 52971
+    },
+    {
+      "epoch": 5.440484748895964,
+      "grad_norm": 0.06554149091243744,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 52974
+    },
+    {
+      "epoch": 5.440792852007805,
+      "grad_norm": 0.08161085098981857,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 52977
+    },
+    {
+      "epoch": 5.441100955119647,
+      "grad_norm": 0.07675690948963165,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 52980
+    },
+    {
+      "epoch": 5.441409058231488,
+      "grad_norm": 0.04832153394818306,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 52983
+    },
+    {
+      "epoch": 5.4417171613433295,
+      "grad_norm": 0.03355338051915169,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 52986
+    },
+    {
+      "epoch": 5.442025264455171,
+      "grad_norm": 0.03371288999915123,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 52989
+    },
+    {
+      "epoch": 5.442333367567012,
+      "grad_norm": 0.11185171455144882,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 52992
+    },
+    {
+      "epoch": 5.442641470678854,
+      "grad_norm": 0.059647444635629654,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 52995
+    },
+    {
+      "epoch": 5.442949573790695,
+      "grad_norm": 0.03346579894423485,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 52998
+    },
+    {
+      "epoch": 5.443257676902537,
+      "grad_norm": 0.0569952167570591,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 53001
+    },
+    {
+      "epoch": 5.443565780014378,
+      "grad_norm": 0.07441236078739166,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 53004
+    },
+    {
+      "epoch": 5.443873883126219,
+      "grad_norm": 0.11744049191474915,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 53007
+    },
+    {
+      "epoch": 5.444181986238061,
+      "grad_norm": 0.04918253421783447,
+      "learning_rate": 0.01,
+      "loss": 2.0148,
+      "step": 53010
+    },
+    {
+      "epoch": 5.444490089349903,
+      "grad_norm": 0.03746315464377403,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 53013
+    },
+    {
+      "epoch": 5.444798192461744,
+      "grad_norm": 0.055012766271829605,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 53016
+    },
+    {
+      "epoch": 5.445106295573585,
+      "grad_norm": 0.06424736231565475,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 53019
+    },
+    {
+      "epoch": 5.445414398685426,
+      "grad_norm": 0.09382134675979614,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 53022
+    },
+    {
+      "epoch": 5.445722501797269,
+      "grad_norm": 0.048184242099523544,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 53025
+    },
+    {
+      "epoch": 5.44603060490911,
+      "grad_norm": 0.05549463629722595,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 53028
+    },
+    {
+      "epoch": 5.446338708020951,
+      "grad_norm": 0.10756690055131912,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 53031
+    },
+    {
+      "epoch": 5.446646811132792,
+      "grad_norm": 0.03808826580643654,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 53034
+    },
+    {
+      "epoch": 5.446954914244634,
+      "grad_norm": 0.050678886473178864,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 53037
+    },
+    {
+      "epoch": 5.447263017356475,
+      "grad_norm": 0.03573575243353844,
+      "learning_rate": 0.01,
+      "loss": 2.0233,
+      "step": 53040
+    },
+    {
+      "epoch": 5.447571120468317,
+      "grad_norm": 0.05302232876420021,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 53043
+    },
+    {
+      "epoch": 5.447879223580158,
+      "grad_norm": 0.03617763891816139,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 53046
+    },
+    {
+      "epoch": 5.448187326692,
+      "grad_norm": 0.039550527930259705,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 53049
+    },
+    {
+      "epoch": 5.448495429803841,
+      "grad_norm": 0.041170112788677216,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 53052
+    },
+    {
+      "epoch": 5.448803532915682,
+      "grad_norm": 0.04932108521461487,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 53055
+    },
+    {
+      "epoch": 5.449111636027524,
+      "grad_norm": 0.03305281326174736,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 53058
+    },
+    {
+      "epoch": 5.4494197391393655,
+      "grad_norm": 0.04628564044833183,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 53061
+    },
+    {
+      "epoch": 5.449727842251207,
+      "grad_norm": 0.04561088606715202,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 53064
+    },
+    {
+      "epoch": 5.450035945363048,
+      "grad_norm": 0.11011772602796555,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 53067
+    },
+    {
+      "epoch": 5.450344048474889,
+      "grad_norm": 0.06649364531040192,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 53070
+    },
+    {
+      "epoch": 5.450652151586731,
+      "grad_norm": 0.1190907210111618,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 53073
+    },
+    {
+      "epoch": 5.450960254698573,
+      "grad_norm": 0.03680254891514778,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 53076
+    },
+    {
+      "epoch": 5.451268357810414,
+      "grad_norm": 0.04634016007184982,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 53079
+    },
+    {
+      "epoch": 5.451576460922255,
+      "grad_norm": 0.04958662390708923,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 53082
+    },
+    {
+      "epoch": 5.4518845640340965,
+      "grad_norm": 0.04827871546149254,
+      "learning_rate": 0.01,
+      "loss": 2.0159,
+      "step": 53085
+    },
+    {
+      "epoch": 5.452192667145939,
+      "grad_norm": 0.10182766616344452,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 53088
+    },
+    {
+      "epoch": 5.45250077025778,
+      "grad_norm": 0.08067356050014496,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 53091
+    },
+    {
+      "epoch": 5.452808873369621,
+      "grad_norm": 0.06457395106554031,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 53094
+    },
+    {
+      "epoch": 5.453116976481462,
+      "grad_norm": 0.09059132635593414,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 53097
+    },
+    {
+      "epoch": 5.453425079593304,
+      "grad_norm": 0.05704977363348007,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 53100
+    },
+    {
+      "epoch": 5.453733182705145,
+      "grad_norm": 0.09887251257896423,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 53103
+    },
+    {
+      "epoch": 5.454041285816987,
+      "grad_norm": 0.03840193152427673,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 53106
+    },
+    {
+      "epoch": 5.454349388928828,
+      "grad_norm": 0.08743079751729965,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 53109
+    },
+    {
+      "epoch": 5.45465749204067,
+      "grad_norm": 0.054274316877126694,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 53112
+    },
+    {
+      "epoch": 5.454965595152511,
+      "grad_norm": 0.033519454300403595,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 53115
+    },
+    {
+      "epoch": 5.455273698264352,
+      "grad_norm": 0.1241462305188179,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 53118
+    },
+    {
+      "epoch": 5.455581801376194,
+      "grad_norm": 0.11901737749576569,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 53121
+    },
+    {
+      "epoch": 5.455889904488036,
+      "grad_norm": 0.06983425468206406,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 53124
+    },
+    {
+      "epoch": 5.456198007599877,
+      "grad_norm": 0.07202869653701782,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 53127
+    },
+    {
+      "epoch": 5.456506110711718,
+      "grad_norm": 0.049013327807188034,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 53130
+    },
+    {
+      "epoch": 5.456814213823559,
+      "grad_norm": 0.04855305328965187,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 53133
+    },
+    {
+      "epoch": 5.457122316935401,
+      "grad_norm": 0.05479707196354866,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 53136
+    },
+    {
+      "epoch": 5.457430420047243,
+      "grad_norm": 0.04120011255145073,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 53139
+    },
+    {
+      "epoch": 5.457738523159084,
+      "grad_norm": 0.11232541501522064,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 53142
+    },
+    {
+      "epoch": 5.458046626270925,
+      "grad_norm": 0.0826067104935646,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 53145
+    },
+    {
+      "epoch": 5.458354729382767,
+      "grad_norm": 0.0358428992331028,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 53148
+    },
+    {
+      "epoch": 5.458662832494608,
+      "grad_norm": 0.06724508851766586,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 53151
+    },
+    {
+      "epoch": 5.45897093560645,
+      "grad_norm": 0.05402735620737076,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 53154
+    },
+    {
+      "epoch": 5.459279038718291,
+      "grad_norm": 0.10397180914878845,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 53157
+    },
+    {
+      "epoch": 5.4595871418301325,
+      "grad_norm": 0.16697724163532257,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 53160
+    },
+    {
+      "epoch": 5.459895244941974,
+      "grad_norm": 0.12541674077510834,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 53163
+    },
+    {
+      "epoch": 5.460203348053815,
+      "grad_norm": 0.11177428066730499,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 53166
+    },
+    {
+      "epoch": 5.460511451165657,
+      "grad_norm": 0.12203194200992584,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 53169
+    },
+    {
+      "epoch": 5.460819554277498,
+      "grad_norm": 0.03482053428888321,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 53172
+    },
+    {
+      "epoch": 5.46112765738934,
+      "grad_norm": 0.03225795179605484,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 53175
+    },
+    {
+      "epoch": 5.461435760501181,
+      "grad_norm": 0.03760567680001259,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 53178
+    },
+    {
+      "epoch": 5.461743863613022,
+      "grad_norm": 0.12543334066867828,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 53181
+    },
+    {
+      "epoch": 5.462051966724864,
+      "grad_norm": 0.04533065855503082,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 53184
+    },
+    {
+      "epoch": 5.462360069836706,
+      "grad_norm": 0.07519727945327759,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 53187
+    },
+    {
+      "epoch": 5.462668172948547,
+      "grad_norm": 0.11682818084955215,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 53190
+    },
+    {
+      "epoch": 5.462976276060388,
+      "grad_norm": 0.043735865503549576,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 53193
+    },
+    {
+      "epoch": 5.463284379172229,
+      "grad_norm": 0.09902381896972656,
+      "learning_rate": 0.01,
+      "loss": 2.0089,
+      "step": 53196
+    },
+    {
+      "epoch": 5.463592482284071,
+      "grad_norm": 0.05974416807293892,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 53199
+    },
+    {
+      "epoch": 5.463900585395913,
+      "grad_norm": 0.07005894929170609,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 53202
+    },
+    {
+      "epoch": 5.464208688507754,
+      "grad_norm": 0.055365189909935,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 53205
+    },
+    {
+      "epoch": 5.464516791619595,
+      "grad_norm": 0.07333406060934067,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 53208
+    },
+    {
+      "epoch": 5.464824894731437,
+      "grad_norm": 0.03612162545323372,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 53211
+    },
+    {
+      "epoch": 5.465132997843278,
+      "grad_norm": 0.036687567830085754,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 53214
+    },
+    {
+      "epoch": 5.46544110095512,
+      "grad_norm": 0.0842447429895401,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 53217
+    },
+    {
+      "epoch": 5.465749204066961,
+      "grad_norm": 0.06415588408708572,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 53220
+    },
+    {
+      "epoch": 5.466057307178803,
+      "grad_norm": 0.0578574612736702,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 53223
+    },
+    {
+      "epoch": 5.466365410290644,
+      "grad_norm": 0.04436985403299332,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 53226
+    },
+    {
+      "epoch": 5.466673513402485,
+      "grad_norm": 0.1035226359963417,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 53229
+    },
+    {
+      "epoch": 5.466981616514326,
+      "grad_norm": 0.04867973551154137,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 53232
+    },
+    {
+      "epoch": 5.4672897196261685,
+      "grad_norm": 0.0710483267903328,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 53235
+    },
+    {
+      "epoch": 5.46759782273801,
+      "grad_norm": 0.052401233464479446,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 53238
+    },
+    {
+      "epoch": 5.467905925849851,
+      "grad_norm": 0.036160893738269806,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 53241
+    },
+    {
+      "epoch": 5.468214028961692,
+      "grad_norm": 0.04797692224383354,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 53244
+    },
+    {
+      "epoch": 5.4685221320735335,
+      "grad_norm": 0.037899672985076904,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 53247
+    },
+    {
+      "epoch": 5.468830235185376,
+      "grad_norm": 0.054781220853328705,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 53250
+    },
+    {
+      "epoch": 5.469138338297217,
+      "grad_norm": 0.09595558792352676,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 53253
+    },
+    {
+      "epoch": 5.469446441409058,
+      "grad_norm": 0.0511094331741333,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 53256
+    },
+    {
+      "epoch": 5.4697545445208995,
+      "grad_norm": 0.05115760117769241,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 53259
+    },
+    {
+      "epoch": 5.470062647632741,
+      "grad_norm": 0.05593256279826164,
+      "learning_rate": 0.01,
+      "loss": 2.019,
+      "step": 53262
+    },
+    {
+      "epoch": 5.470370750744583,
+      "grad_norm": 0.048911042511463165,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 53265
+    },
+    {
+      "epoch": 5.470678853856424,
+      "grad_norm": 0.05227271467447281,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 53268
+    },
+    {
+      "epoch": 5.470986956968265,
+      "grad_norm": 0.11259990185499191,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 53271
+    },
+    {
+      "epoch": 5.471295060080107,
+      "grad_norm": 0.04125455394387245,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 53274
+    },
+    {
+      "epoch": 5.471603163191948,
+      "grad_norm": 0.11485335975885391,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 53277
+    },
+    {
+      "epoch": 5.47191126630379,
+      "grad_norm": 0.07149787247180939,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 53280
+    },
+    {
+      "epoch": 5.472219369415631,
+      "grad_norm": 0.05775618925690651,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 53283
+    },
+    {
+      "epoch": 5.472527472527473,
+      "grad_norm": 0.04515690356492996,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 53286
+    },
+    {
+      "epoch": 5.472835575639314,
+      "grad_norm": 0.0433974415063858,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 53289
+    },
+    {
+      "epoch": 5.473143678751155,
+      "grad_norm": 0.03472794219851494,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 53292
+    },
+    {
+      "epoch": 5.473451781862996,
+      "grad_norm": 0.04648016393184662,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 53295
+    },
+    {
+      "epoch": 5.473759884974839,
+      "grad_norm": 0.048154812306165695,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 53298
+    },
+    {
+      "epoch": 5.47406798808668,
+      "grad_norm": 0.040329910814762115,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 53301
+    },
+    {
+      "epoch": 5.474376091198521,
+      "grad_norm": 0.10805533081293106,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 53304
+    },
+    {
+      "epoch": 5.474684194310362,
+      "grad_norm": 0.04139288142323494,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 53307
+    },
+    {
+      "epoch": 5.474992297422204,
+      "grad_norm": 0.08323890715837479,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 53310
+    },
+    {
+      "epoch": 5.475300400534046,
+      "grad_norm": 0.07098772376775742,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 53313
+    },
+    {
+      "epoch": 5.475608503645887,
+      "grad_norm": 0.09589933604001999,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 53316
+    },
+    {
+      "epoch": 5.475916606757728,
+      "grad_norm": 0.09487750381231308,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 53319
+    },
+    {
+      "epoch": 5.4762247098695696,
+      "grad_norm": 0.04062903672456741,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 53322
+    },
+    {
+      "epoch": 5.476532812981411,
+      "grad_norm": 0.036831121891736984,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 53325
+    },
+    {
+      "epoch": 5.476840916093252,
+      "grad_norm": 0.05887876823544502,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 53328
+    },
+    {
+      "epoch": 5.477149019205094,
+      "grad_norm": 0.07680708169937134,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 53331
+    },
+    {
+      "epoch": 5.4774571223169355,
+      "grad_norm": 0.06624645739793777,
+      "learning_rate": 0.01,
+      "loss": 2.0242,
+      "step": 53334
+    },
+    {
+      "epoch": 5.477765225428777,
+      "grad_norm": 0.051530107855796814,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 53337
+    },
+    {
+      "epoch": 5.478073328540618,
+      "grad_norm": 0.046162448823451996,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 53340
+    },
+    {
+      "epoch": 5.478381431652459,
+      "grad_norm": 0.038336243480443954,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 53343
+    },
+    {
+      "epoch": 5.478689534764301,
+      "grad_norm": 0.03402607515454292,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 53346
+    },
+    {
+      "epoch": 5.478997637876143,
+      "grad_norm": 0.10580262541770935,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 53349
+    },
+    {
+      "epoch": 5.479305740987984,
+      "grad_norm": 0.13053545355796814,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 53352
+    },
+    {
+      "epoch": 5.479613844099825,
+      "grad_norm": 0.050016142427921295,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 53355
+    },
+    {
+      "epoch": 5.4799219472116665,
+      "grad_norm": 0.03820059821009636,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 53358
+    },
+    {
+      "epoch": 5.480230050323509,
+      "grad_norm": 0.04353965446352959,
+      "learning_rate": 0.01,
+      "loss": 2.0286,
+      "step": 53361
+    },
+    {
+      "epoch": 5.48053815343535,
+      "grad_norm": 0.04177290201187134,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 53364
+    },
+    {
+      "epoch": 5.480846256547191,
+      "grad_norm": 0.03511528670787811,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 53367
+    },
+    {
+      "epoch": 5.481154359659032,
+      "grad_norm": 0.10579612106084824,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 53370
+    },
+    {
+      "epoch": 5.481462462770874,
+      "grad_norm": 0.08428878337144852,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 53373
+    },
+    {
+      "epoch": 5.481770565882716,
+      "grad_norm": 0.045665886253118515,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 53376
+    },
+    {
+      "epoch": 5.482078668994557,
+      "grad_norm": 0.10956547409296036,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 53379
+    },
+    {
+      "epoch": 5.482386772106398,
+      "grad_norm": 0.07283762842416763,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 53382
+    },
+    {
+      "epoch": 5.48269487521824,
+      "grad_norm": 0.11426263302564621,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 53385
+    },
+    {
+      "epoch": 5.483002978330081,
+      "grad_norm": 0.06644517928361893,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 53388
+    },
+    {
+      "epoch": 5.483311081441922,
+      "grad_norm": 0.06233721226453781,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 53391
+    },
+    {
+      "epoch": 5.483619184553764,
+      "grad_norm": 0.06707890331745148,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 53394
+    },
+    {
+      "epoch": 5.4839272876656056,
+      "grad_norm": 0.06242217868566513,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 53397
+    },
+    {
+      "epoch": 5.484235390777447,
+      "grad_norm": 0.03913586214184761,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 53400
+    },
+    {
+      "epoch": 5.484543493889288,
+      "grad_norm": 0.10017285495996475,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 53403
+    },
+    {
+      "epoch": 5.484851597001129,
+      "grad_norm": 0.07849404960870743,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 53406
+    },
+    {
+      "epoch": 5.4851597001129715,
+      "grad_norm": 0.06911784410476685,
+      "learning_rate": 0.01,
+      "loss": 2.0207,
+      "step": 53409
+    },
+    {
+      "epoch": 5.485467803224813,
+      "grad_norm": 0.07752948254346848,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 53412
+    },
+    {
+      "epoch": 5.485775906336654,
+      "grad_norm": 0.0855301171541214,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 53415
+    },
+    {
+      "epoch": 5.486084009448495,
+      "grad_norm": 0.039153728634119034,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 53418
+    },
+    {
+      "epoch": 5.4863921125603365,
+      "grad_norm": 0.09871362149715424,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 53421
+    },
+    {
+      "epoch": 5.486700215672179,
+      "grad_norm": 0.05920962989330292,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 53424
+    },
+    {
+      "epoch": 5.48700831878402,
+      "grad_norm": 0.04103608801960945,
+      "learning_rate": 0.01,
+      "loss": 2.0363,
+      "step": 53427
+    },
+    {
+      "epoch": 5.487316421895861,
+      "grad_norm": 0.08447948098182678,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 53430
+    },
+    {
+      "epoch": 5.4876245250077025,
+      "grad_norm": 0.055456507951021194,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 53433
+    },
+    {
+      "epoch": 5.487932628119544,
+      "grad_norm": 0.05828576907515526,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 53436
+    },
+    {
+      "epoch": 5.488240731231386,
+      "grad_norm": 0.09117809683084488,
+      "learning_rate": 0.01,
+      "loss": 2.0269,
+      "step": 53439
+    },
+    {
+      "epoch": 5.488548834343227,
+      "grad_norm": 0.08137334883213043,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 53442
+    },
+    {
+      "epoch": 5.488856937455068,
+      "grad_norm": 0.08830280601978302,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 53445
+    },
+    {
+      "epoch": 5.48916504056691,
+      "grad_norm": 0.04386366158723831,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 53448
+    },
+    {
+      "epoch": 5.489473143678751,
+      "grad_norm": 0.0500885471701622,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 53451
+    },
+    {
+      "epoch": 5.489781246790592,
+      "grad_norm": 0.044326718896627426,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 53454
+    },
+    {
+      "epoch": 5.490089349902434,
+      "grad_norm": 0.1106194257736206,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 53457
+    },
+    {
+      "epoch": 5.490397453014276,
+      "grad_norm": 0.05222434550523758,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 53460
+    },
+    {
+      "epoch": 5.490705556126117,
+      "grad_norm": 0.08016736060380936,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 53463
+    },
+    {
+      "epoch": 5.491013659237958,
+      "grad_norm": 0.06921573728322983,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 53466
+    },
+    {
+      "epoch": 5.491321762349799,
+      "grad_norm": 0.10140072554349899,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 53469
+    },
+    {
+      "epoch": 5.491629865461642,
+      "grad_norm": 0.05930350720882416,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 53472
+    },
+    {
+      "epoch": 5.491937968573483,
+      "grad_norm": 0.054523758590221405,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 53475
+    },
+    {
+      "epoch": 5.492246071685324,
+      "grad_norm": 0.06033066287636757,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 53478
+    },
+    {
+      "epoch": 5.492554174797165,
+      "grad_norm": 0.11593367904424667,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 53481
+    },
+    {
+      "epoch": 5.492862277909007,
+      "grad_norm": 0.054470766335725784,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 53484
+    },
+    {
+      "epoch": 5.493170381020848,
+      "grad_norm": 0.05710428208112717,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 53487
+    },
+    {
+      "epoch": 5.49347848413269,
+      "grad_norm": 0.08135402947664261,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 53490
+    },
+    {
+      "epoch": 5.493786587244531,
+      "grad_norm": 0.13751734793186188,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 53493
+    },
+    {
+      "epoch": 5.4940946903563725,
+      "grad_norm": 0.07379119843244553,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 53496
+    },
+    {
+      "epoch": 5.494402793468214,
+      "grad_norm": 0.0595991387963295,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 53499
+    },
+    {
+      "epoch": 5.494710896580055,
+      "grad_norm": 0.05081510916352272,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 53502
+    },
+    {
+      "epoch": 5.495018999691897,
+      "grad_norm": 0.040366947650909424,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 53505
+    },
+    {
+      "epoch": 5.4953271028037385,
+      "grad_norm": 0.10154601186513901,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 53508
+    },
+    {
+      "epoch": 5.49563520591558,
+      "grad_norm": 0.0562141053378582,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 53511
+    },
+    {
+      "epoch": 5.495943309027421,
+      "grad_norm": 0.07368075847625732,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 53514
+    },
+    {
+      "epoch": 5.496251412139262,
+      "grad_norm": 0.04702366143465042,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 53517
+    },
+    {
+      "epoch": 5.496559515251104,
+      "grad_norm": 0.05341365188360214,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 53520
+    },
+    {
+      "epoch": 5.496867618362946,
+      "grad_norm": 0.10820145905017853,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 53523
+    },
+    {
+      "epoch": 5.497175721474787,
+      "grad_norm": 0.053429488092660904,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 53526
+    },
+    {
+      "epoch": 5.497483824586628,
+      "grad_norm": 0.1272108554840088,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 53529
+    },
+    {
+      "epoch": 5.4977919276984695,
+      "grad_norm": 0.043057285249233246,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 53532
+    },
+    {
+      "epoch": 5.498100030810312,
+      "grad_norm": 0.055451132357120514,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 53535
+    },
+    {
+      "epoch": 5.498408133922153,
+      "grad_norm": 0.11000920832157135,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 53538
+    },
+    {
+      "epoch": 5.498716237033994,
+      "grad_norm": 0.043610602617263794,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 53541
+    },
+    {
+      "epoch": 5.499024340145835,
+      "grad_norm": 0.12951324880123138,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 53544
+    },
+    {
+      "epoch": 5.499332443257677,
+      "grad_norm": 0.1155344694852829,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 53547
+    },
+    {
+      "epoch": 5.499640546369518,
+      "grad_norm": 0.04202111065387726,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 53550
+    },
+    {
+      "epoch": 5.49994864948136,
+      "grad_norm": 0.05797869712114334,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 53553
+    },
+    {
+      "epoch": 5.500256752593201,
+      "grad_norm": 0.055514171719551086,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 53556
+    },
+    {
+      "epoch": 5.500564855705043,
+      "grad_norm": 0.032696306705474854,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 53559
+    },
+    {
+      "epoch": 5.500872958816884,
+      "grad_norm": 0.07791826128959656,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 53562
+    },
+    {
+      "epoch": 5.501181061928725,
+      "grad_norm": 0.12042553722858429,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 53565
+    },
+    {
+      "epoch": 5.501489165040567,
+      "grad_norm": 0.08734118938446045,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 53568
+    },
+    {
+      "epoch": 5.5017972681524085,
+      "grad_norm": 0.1427556276321411,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 53571
+    },
+    {
+      "epoch": 5.50210537126425,
+      "grad_norm": 0.05691046267747879,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 53574
+    },
+    {
+      "epoch": 5.502413474376091,
+      "grad_norm": 0.05428704619407654,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 53577
+    },
+    {
+      "epoch": 5.502721577487932,
+      "grad_norm": 0.07390681654214859,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 53580
+    },
+    {
+      "epoch": 5.503029680599774,
+      "grad_norm": 0.04650009423494339,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 53583
+    },
+    {
+      "epoch": 5.503337783711616,
+      "grad_norm": 0.06623617559671402,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 53586
+    },
+    {
+      "epoch": 5.503645886823457,
+      "grad_norm": 0.08254354447126389,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 53589
+    },
+    {
+      "epoch": 5.503953989935298,
+      "grad_norm": 0.07407466322183609,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 53592
+    },
+    {
+      "epoch": 5.5042620930471395,
+      "grad_norm": 0.05553009733557701,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 53595
+    },
+    {
+      "epoch": 5.504570196158982,
+      "grad_norm": 0.068050317466259,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 53598
+    },
+    {
+      "epoch": 5.504878299270823,
+      "grad_norm": 0.036538973450660706,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 53601
+    },
+    {
+      "epoch": 5.505186402382664,
+      "grad_norm": 0.04083050787448883,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 53604
+    },
+    {
+      "epoch": 5.5054945054945055,
+      "grad_norm": 0.03994308039546013,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 53607
+    },
+    {
+      "epoch": 5.505802608606347,
+      "grad_norm": 0.039935458451509476,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 53610
+    },
+    {
+      "epoch": 5.506110711718188,
+      "grad_norm": 0.09762649983167648,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 53613
+    },
+    {
+      "epoch": 5.50641881483003,
+      "grad_norm": 0.058254171162843704,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 53616
+    },
+    {
+      "epoch": 5.506726917941871,
+      "grad_norm": 0.11991842836141586,
+      "learning_rate": 0.01,
+      "loss": 2.0198,
+      "step": 53619
+    },
+    {
+      "epoch": 5.507035021053713,
+      "grad_norm": 0.04026370868086815,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 53622
+    },
+    {
+      "epoch": 5.507343124165554,
+      "grad_norm": 0.03855542093515396,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 53625
+    },
+    {
+      "epoch": 5.507651227277395,
+      "grad_norm": 0.04439062252640724,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 53628
+    },
+    {
+      "epoch": 5.507959330389237,
+      "grad_norm": 0.09283201396465302,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 53631
+    },
+    {
+      "epoch": 5.508267433501079,
+      "grad_norm": 0.05495656281709671,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 53634
+    },
+    {
+      "epoch": 5.50857553661292,
+      "grad_norm": 0.08464661240577698,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 53637
+    },
+    {
+      "epoch": 5.508883639724761,
+      "grad_norm": 0.06856382638216019,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 53640
+    },
+    {
+      "epoch": 5.509191742836602,
+      "grad_norm": 0.10367941111326218,
+      "learning_rate": 0.01,
+      "loss": 2.0157,
+      "step": 53643
+    },
+    {
+      "epoch": 5.509499845948444,
+      "grad_norm": 0.07614094018936157,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 53646
+    },
+    {
+      "epoch": 5.509807949060286,
+      "grad_norm": 0.07301559299230576,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 53649
+    },
+    {
+      "epoch": 5.510116052172127,
+      "grad_norm": 0.03928094357252121,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 53652
+    },
+    {
+      "epoch": 5.510424155283968,
+      "grad_norm": 0.04542843624949455,
+      "learning_rate": 0.01,
+      "loss": 2.0183,
+      "step": 53655
+    },
+    {
+      "epoch": 5.51073225839581,
+      "grad_norm": 0.08662543445825577,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 53658
+    },
+    {
+      "epoch": 5.511040361507651,
+      "grad_norm": 0.045004189014434814,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 53661
+    },
+    {
+      "epoch": 5.511348464619493,
+      "grad_norm": 0.03661379590630531,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 53664
+    },
+    {
+      "epoch": 5.511656567731334,
+      "grad_norm": 0.05117429420351982,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 53667
+    },
+    {
+      "epoch": 5.5119646708431755,
+      "grad_norm": 0.06222992762923241,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 53670
+    },
+    {
+      "epoch": 5.512272773955017,
+      "grad_norm": 0.04274081811308861,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 53673
+    },
+    {
+      "epoch": 5.512580877066858,
+      "grad_norm": 0.09774081408977509,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 53676
+    },
+    {
+      "epoch": 5.512888980178699,
+      "grad_norm": 0.15694968402385712,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 53679
+    },
+    {
+      "epoch": 5.5131970832905415,
+      "grad_norm": 0.06906448304653168,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 53682
+    },
+    {
+      "epoch": 5.513505186402383,
+      "grad_norm": 0.04059962183237076,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 53685
+    },
+    {
+      "epoch": 5.513813289514224,
+      "grad_norm": 0.03401316702365875,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 53688
+    },
+    {
+      "epoch": 5.514121392626065,
+      "grad_norm": 0.03932545334100723,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 53691
+    },
+    {
+      "epoch": 5.514429495737907,
+      "grad_norm": 0.04314441978931427,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 53694
+    },
+    {
+      "epoch": 5.514737598849749,
+      "grad_norm": 0.04282282292842865,
+      "learning_rate": 0.01,
+      "loss": 2.0311,
+      "step": 53697
+    },
+    {
+      "epoch": 5.51504570196159,
+      "grad_norm": 0.07264575362205505,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 53700
+    },
+    {
+      "epoch": 5.515353805073431,
+      "grad_norm": 0.031450774520635605,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 53703
+    },
+    {
+      "epoch": 5.5156619081852725,
+      "grad_norm": 0.04431367665529251,
+      "learning_rate": 0.01,
+      "loss": 2.0218,
+      "step": 53706
+    },
+    {
+      "epoch": 5.515970011297114,
+      "grad_norm": 0.059152550995349884,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 53709
+    },
+    {
+      "epoch": 5.516278114408956,
+      "grad_norm": 0.05627863109111786,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 53712
+    },
+    {
+      "epoch": 5.516586217520797,
+      "grad_norm": 0.12007083743810654,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 53715
+    },
+    {
+      "epoch": 5.516894320632638,
+      "grad_norm": 0.04498304799199104,
+      "learning_rate": 0.01,
+      "loss": 2.0171,
+      "step": 53718
+    },
+    {
+      "epoch": 5.51720242374448,
+      "grad_norm": 0.05469832196831703,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 53721
+    },
+    {
+      "epoch": 5.517510526856321,
+      "grad_norm": 0.12083622813224792,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 53724
+    },
+    {
+      "epoch": 5.517818629968163,
+      "grad_norm": 0.05532558634877205,
+      "learning_rate": 0.01,
+      "loss": 2.0263,
+      "step": 53727
+    },
+    {
+      "epoch": 5.518126733080004,
+      "grad_norm": 0.0655357614159584,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 53730
+    },
+    {
+      "epoch": 5.518434836191846,
+      "grad_norm": 0.03844601288437843,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 53733
+    },
+    {
+      "epoch": 5.518742939303687,
+      "grad_norm": 0.08736705034971237,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 53736
+    },
+    {
+      "epoch": 5.519051042415528,
+      "grad_norm": 0.03798610344529152,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 53739
+    },
+    {
+      "epoch": 5.519359145527369,
+      "grad_norm": 0.08266885578632355,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 53742
+    },
+    {
+      "epoch": 5.5196672486392115,
+      "grad_norm": 0.12414561212062836,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 53745
+    },
+    {
+      "epoch": 5.519975351751053,
+      "grad_norm": 0.07735760509967804,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 53748
+    },
+    {
+      "epoch": 5.520283454862894,
+      "grad_norm": 0.07708865404129028,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 53751
+    },
+    {
+      "epoch": 5.520591557974735,
+      "grad_norm": 0.08443256467580795,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 53754
+    },
+    {
+      "epoch": 5.520899661086577,
+      "grad_norm": 0.061041250824928284,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 53757
+    },
+    {
+      "epoch": 5.521207764198419,
+      "grad_norm": 0.06206076219677925,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 53760
+    },
+    {
+      "epoch": 5.52151586731026,
+      "grad_norm": 0.07099471986293793,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 53763
+    },
+    {
+      "epoch": 5.521823970422101,
+      "grad_norm": 0.08134863525629044,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 53766
+    },
+    {
+      "epoch": 5.5221320735339425,
+      "grad_norm": 0.08037443459033966,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 53769
+    },
+    {
+      "epoch": 5.522440176645784,
+      "grad_norm": 0.06751944124698639,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 53772
+    },
+    {
+      "epoch": 5.522748279757625,
+      "grad_norm": 0.04069173336029053,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 53775
+    },
+    {
+      "epoch": 5.523056382869467,
+      "grad_norm": 0.04320038482546806,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 53778
+    },
+    {
+      "epoch": 5.5233644859813085,
+      "grad_norm": 0.11609040200710297,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 53781
+    },
+    {
+      "epoch": 5.52367258909315,
+      "grad_norm": 0.050336502492427826,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 53784
+    },
+    {
+      "epoch": 5.523980692204991,
+      "grad_norm": 0.039348434656858444,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 53787
+    },
+    {
+      "epoch": 5.524288795316833,
+      "grad_norm": 0.08109190315008163,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 53790
+    },
+    {
+      "epoch": 5.524596898428674,
+      "grad_norm": 0.07181521505117416,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 53793
+    },
+    {
+      "epoch": 5.524905001540516,
+      "grad_norm": 0.04678622633218765,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 53796
+    },
+    {
+      "epoch": 5.525213104652357,
+      "grad_norm": 0.079826720058918,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 53799
+    },
+    {
+      "epoch": 5.525521207764198,
+      "grad_norm": 0.08414063602685928,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 53802
+    },
+    {
+      "epoch": 5.5258293108760395,
+      "grad_norm": 0.04829956591129303,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 53805
+    },
+    {
+      "epoch": 5.526137413987882,
+      "grad_norm": 0.0862230584025383,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 53808
+    },
+    {
+      "epoch": 5.526445517099723,
+      "grad_norm": 0.06035618856549263,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 53811
+    },
+    {
+      "epoch": 5.526753620211564,
+      "grad_norm": 0.04363265261054039,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 53814
+    },
+    {
+      "epoch": 5.527061723323405,
+      "grad_norm": 0.06226656585931778,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 53817
+    },
+    {
+      "epoch": 5.527369826435247,
+      "grad_norm": 0.10821603238582611,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 53820
+    },
+    {
+      "epoch": 5.527677929547089,
+      "grad_norm": 0.06734979152679443,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 53823
+    },
+    {
+      "epoch": 5.52798603265893,
+      "grad_norm": 0.07929663360118866,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 53826
+    },
+    {
+      "epoch": 5.528294135770771,
+      "grad_norm": 0.11255302280187607,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 53829
+    },
+    {
+      "epoch": 5.528602238882613,
+      "grad_norm": 0.06797608733177185,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 53832
+    },
+    {
+      "epoch": 5.528910341994454,
+      "grad_norm": 0.07606469094753265,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 53835
+    },
+    {
+      "epoch": 5.529218445106295,
+      "grad_norm": 0.04698120057582855,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 53838
+    },
+    {
+      "epoch": 5.529526548218137,
+      "grad_norm": 0.07087966799736023,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 53841
+    },
+    {
+      "epoch": 5.5298346513299785,
+      "grad_norm": 0.048692237585783005,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 53844
+    },
+    {
+      "epoch": 5.53014275444182,
+      "grad_norm": 0.04359065368771553,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 53847
+    },
+    {
+      "epoch": 5.530450857553661,
+      "grad_norm": 0.05424369126558304,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 53850
+    },
+    {
+      "epoch": 5.530758960665503,
+      "grad_norm": 0.13464900851249695,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 53853
+    },
+    {
+      "epoch": 5.5310670637773445,
+      "grad_norm": 0.038927361369132996,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 53856
+    },
+    {
+      "epoch": 5.531375166889186,
+      "grad_norm": 0.040312107652425766,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 53859
+    },
+    {
+      "epoch": 5.531683270001027,
+      "grad_norm": 0.05727381631731987,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 53862
+    },
+    {
+      "epoch": 5.531991373112868,
+      "grad_norm": 0.06750021874904633,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 53865
+    },
+    {
+      "epoch": 5.5322994762247095,
+      "grad_norm": 0.08220100402832031,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 53868
+    },
+    {
+      "epoch": 5.532607579336552,
+      "grad_norm": 0.05215999484062195,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 53871
+    },
+    {
+      "epoch": 5.532915682448393,
+      "grad_norm": 0.06510480493307114,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 53874
+    },
+    {
+      "epoch": 5.533223785560234,
+      "grad_norm": 0.04641730710864067,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 53877
+    },
+    {
+      "epoch": 5.5335318886720755,
+      "grad_norm": 0.04414188116788864,
+      "learning_rate": 0.01,
+      "loss": 2.0314,
+      "step": 53880
+    },
+    {
+      "epoch": 5.533839991783917,
+      "grad_norm": 0.041879378259181976,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 53883
+    },
+    {
+      "epoch": 5.534148094895759,
+      "grad_norm": 0.039413318037986755,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 53886
+    },
+    {
+      "epoch": 5.5344561980076,
+      "grad_norm": 0.0784526839852333,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 53889
+    },
+    {
+      "epoch": 5.534764301119441,
+      "grad_norm": 0.1049879789352417,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 53892
+    },
+    {
+      "epoch": 5.535072404231283,
+      "grad_norm": 0.15476803481578827,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 53895
+    },
+    {
+      "epoch": 5.535380507343124,
+      "grad_norm": 0.1224508136510849,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 53898
+    },
+    {
+      "epoch": 5.535688610454965,
+      "grad_norm": 0.060283955186605453,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 53901
+    },
+    {
+      "epoch": 5.535996713566807,
+      "grad_norm": 0.05532943457365036,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 53904
+    },
+    {
+      "epoch": 5.536304816678649,
+      "grad_norm": 0.04745487496256828,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 53907
+    },
+    {
+      "epoch": 5.53661291979049,
+      "grad_norm": 0.05770549178123474,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 53910
+    },
+    {
+      "epoch": 5.536921022902331,
+      "grad_norm": 0.05675048008561134,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 53913
+    },
+    {
+      "epoch": 5.537229126014172,
+      "grad_norm": 0.05538811534643173,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 53916
+    },
+    {
+      "epoch": 5.5375372291260145,
+      "grad_norm": 0.05474154278635979,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 53919
+    },
+    {
+      "epoch": 5.537845332237856,
+      "grad_norm": 0.1373576670885086,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 53922
+    },
+    {
+      "epoch": 5.538153435349697,
+      "grad_norm": 0.04938695207238197,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 53925
+    },
+    {
+      "epoch": 5.538461538461538,
+      "grad_norm": 0.08616656810045242,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 53928
+    },
+    {
+      "epoch": 5.53876964157338,
+      "grad_norm": 0.0794534832239151,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 53931
+    },
+    {
+      "epoch": 5.539077744685221,
+      "grad_norm": 0.07034434378147125,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 53934
+    },
+    {
+      "epoch": 5.539385847797063,
+      "grad_norm": 0.09252175688743591,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 53937
+    },
+    {
+      "epoch": 5.539693950908904,
+      "grad_norm": 0.05770866200327873,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 53940
+    },
+    {
+      "epoch": 5.5400020540207455,
+      "grad_norm": 0.06952711939811707,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 53943
+    },
+    {
+      "epoch": 5.540310157132587,
+      "grad_norm": 0.06332490593194962,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 53946
+    },
+    {
+      "epoch": 5.540618260244429,
+      "grad_norm": 0.03297451511025429,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 53949
+    },
+    {
+      "epoch": 5.54092636335627,
+      "grad_norm": 0.03710363432765007,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 53952
+    },
+    {
+      "epoch": 5.5412344664681115,
+      "grad_norm": 0.05471167713403702,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 53955
+    },
+    {
+      "epoch": 5.541542569579953,
+      "grad_norm": 0.0861392468214035,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 53958
+    },
+    {
+      "epoch": 5.541850672691794,
+      "grad_norm": 0.08588635921478271,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 53961
+    },
+    {
+      "epoch": 5.542158775803635,
+      "grad_norm": 0.09388336539268494,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 53964
+    },
+    {
+      "epoch": 5.542466878915477,
+      "grad_norm": 0.1397831290960312,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 53967
+    },
+    {
+      "epoch": 5.542774982027319,
+      "grad_norm": 0.10498721152544022,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 53970
+    },
+    {
+      "epoch": 5.54308308513916,
+      "grad_norm": 0.0574629008769989,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 53973
+    },
+    {
+      "epoch": 5.543391188251001,
+      "grad_norm": 0.03409456089138985,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 53976
+    },
+    {
+      "epoch": 5.5436992913628425,
+      "grad_norm": 0.03830345720052719,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 53979
+    },
+    {
+      "epoch": 5.544007394474685,
+      "grad_norm": 0.0545862540602684,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 53982
+    },
+    {
+      "epoch": 5.544315497586526,
+      "grad_norm": 0.047009505331516266,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 53985
+    },
+    {
+      "epoch": 5.544623600698367,
+      "grad_norm": 0.0546221099793911,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 53988
+    },
+    {
+      "epoch": 5.544931703810208,
+      "grad_norm": 0.05153699591755867,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 53991
+    },
+    {
+      "epoch": 5.54523980692205,
+      "grad_norm": 0.1104668453335762,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 53994
+    },
+    {
+      "epoch": 5.545547910033891,
+      "grad_norm": 0.09211790561676025,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 53997
+    },
+    {
+      "epoch": 5.545856013145733,
+      "grad_norm": 0.03746125102043152,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 54000
+    },
+    {
+      "epoch": 5.546164116257574,
+      "grad_norm": 0.05226179584860802,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 54003
+    },
+    {
+      "epoch": 5.546472219369416,
+      "grad_norm": 0.08278732746839523,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 54006
+    },
+    {
+      "epoch": 5.546780322481257,
+      "grad_norm": 0.08003208786249161,
+      "learning_rate": 0.01,
+      "loss": 2.0169,
+      "step": 54009
+    },
+    {
+      "epoch": 5.547088425593098,
+      "grad_norm": 0.09529578685760498,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 54012
+    },
+    {
+      "epoch": 5.54739652870494,
+      "grad_norm": 0.08121462911367416,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 54015
+    },
+    {
+      "epoch": 5.5477046318167815,
+      "grad_norm": 0.07625698298215866,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 54018
+    },
+    {
+      "epoch": 5.548012734928623,
+      "grad_norm": 0.11102813482284546,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 54021
+    },
+    {
+      "epoch": 5.548320838040464,
+      "grad_norm": 0.05425316095352173,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 54024
+    },
+    {
+      "epoch": 5.548628941152305,
+      "grad_norm": 0.050913844257593155,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 54027
+    },
+    {
+      "epoch": 5.548937044264147,
+      "grad_norm": 0.049932949244976044,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 54030
+    },
+    {
+      "epoch": 5.549245147375989,
+      "grad_norm": 0.058598097413778305,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 54033
+    },
+    {
+      "epoch": 5.54955325048783,
+      "grad_norm": 0.06767188012599945,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 54036
+    },
+    {
+      "epoch": 5.549861353599671,
+      "grad_norm": 0.0709863007068634,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 54039
+    },
+    {
+      "epoch": 5.5501694567115125,
+      "grad_norm": 0.09273546189069748,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 54042
+    },
+    {
+      "epoch": 5.550477559823355,
+      "grad_norm": 0.04681755229830742,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 54045
+    },
+    {
+      "epoch": 5.550785662935196,
+      "grad_norm": 0.08744305372238159,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 54048
+    },
+    {
+      "epoch": 5.551093766047037,
+      "grad_norm": 0.03497626259922981,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 54051
+    },
+    {
+      "epoch": 5.5514018691588785,
+      "grad_norm": 0.05988140404224396,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 54054
+    },
+    {
+      "epoch": 5.55170997227072,
+      "grad_norm": 0.0464041493833065,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 54057
+    },
+    {
+      "epoch": 5.552018075382561,
+      "grad_norm": 0.051982734352350235,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 54060
+    },
+    {
+      "epoch": 5.552326178494403,
+      "grad_norm": 0.045320626348257065,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 54063
+    },
+    {
+      "epoch": 5.552634281606244,
+      "grad_norm": 0.04230513423681259,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 54066
+    },
+    {
+      "epoch": 5.552942384718086,
+      "grad_norm": 0.0393347293138504,
+      "learning_rate": 0.01,
+      "loss": 2.0179,
+      "step": 54069
+    },
+    {
+      "epoch": 5.553250487829927,
+      "grad_norm": 0.1013195738196373,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 54072
+    },
+    {
+      "epoch": 5.553558590941768,
+      "grad_norm": 0.08008227497339249,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 54075
+    },
+    {
+      "epoch": 5.55386669405361,
+      "grad_norm": 0.0729672759771347,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 54078
+    },
+    {
+      "epoch": 5.554174797165452,
+      "grad_norm": 0.0624358095228672,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 54081
+    },
+    {
+      "epoch": 5.554482900277293,
+      "grad_norm": 0.03352997824549675,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 54084
+    },
+    {
+      "epoch": 5.554791003389134,
+      "grad_norm": 0.03241400793194771,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 54087
+    },
+    {
+      "epoch": 5.555099106500975,
+      "grad_norm": 0.033829499036073685,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 54090
+    },
+    {
+      "epoch": 5.555407209612817,
+      "grad_norm": 0.03398764133453369,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 54093
+    },
+    {
+      "epoch": 5.555715312724659,
+      "grad_norm": 0.04630080237984657,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 54096
+    },
+    {
+      "epoch": 5.5560234158365,
+      "grad_norm": 0.08986920863389969,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 54099
+    },
+    {
+      "epoch": 5.556331518948341,
+      "grad_norm": 0.07276762276887894,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 54102
+    },
+    {
+      "epoch": 5.556639622060183,
+      "grad_norm": 0.07180680334568024,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 54105
+    },
+    {
+      "epoch": 5.556947725172025,
+      "grad_norm": 0.0470985509455204,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 54108
+    },
+    {
+      "epoch": 5.557255828283866,
+      "grad_norm": 0.07641714811325073,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 54111
+    },
+    {
+      "epoch": 5.557563931395707,
+      "grad_norm": 0.11065167188644409,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 54114
+    },
+    {
+      "epoch": 5.5578720345075485,
+      "grad_norm": 0.10439474880695343,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 54117
+    },
+    {
+      "epoch": 5.55818013761939,
+      "grad_norm": 0.044907502830028534,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 54120
+    },
+    {
+      "epoch": 5.558488240731231,
+      "grad_norm": 0.052572451531887054,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 54123
+    },
+    {
+      "epoch": 5.558796343843073,
+      "grad_norm": 0.03218472748994827,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 54126
+    },
+    {
+      "epoch": 5.5591044469549145,
+      "grad_norm": 0.04737766459584236,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 54129
+    },
+    {
+      "epoch": 5.559412550066756,
+      "grad_norm": 0.045241083949804306,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 54132
+    },
+    {
+      "epoch": 5.559720653178597,
+      "grad_norm": 0.08022741228342056,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 54135
+    },
+    {
+      "epoch": 5.560028756290438,
+      "grad_norm": 0.1389225870370865,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 54138
+    },
+    {
+      "epoch": 5.56033685940228,
+      "grad_norm": 0.1731427162885666,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 54141
+    },
+    {
+      "epoch": 5.560644962514122,
+      "grad_norm": 0.13785654306411743,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 54144
+    },
+    {
+      "epoch": 5.560953065625963,
+      "grad_norm": 0.08374970406293869,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 54147
+    },
+    {
+      "epoch": 5.561261168737804,
+      "grad_norm": 0.038511909544467926,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 54150
+    },
+    {
+      "epoch": 5.5615692718496454,
+      "grad_norm": 0.04264768585562706,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 54153
+    },
+    {
+      "epoch": 5.561877374961487,
+      "grad_norm": 0.04369993880391121,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 54156
+    },
+    {
+      "epoch": 5.562185478073329,
+      "grad_norm": 0.04195026680827141,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 54159
+    },
+    {
+      "epoch": 5.56249358118517,
+      "grad_norm": 0.04474294185638428,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 54162
+    },
+    {
+      "epoch": 5.562801684297011,
+      "grad_norm": 0.049389854073524475,
+      "learning_rate": 0.01,
+      "loss": 2.027,
+      "step": 54165
+    },
+    {
+      "epoch": 5.563109787408853,
+      "grad_norm": 0.06065073981881142,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 54168
+    },
+    {
+      "epoch": 5.563417890520694,
+      "grad_norm": 0.1028941348195076,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 54171
+    },
+    {
+      "epoch": 5.563725993632536,
+      "grad_norm": 0.05602087453007698,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 54174
+    },
+    {
+      "epoch": 5.564034096744377,
+      "grad_norm": 0.09422741830348969,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 54177
+    },
+    {
+      "epoch": 5.564342199856219,
+      "grad_norm": 0.08502856642007828,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 54180
+    },
+    {
+      "epoch": 5.56465030296806,
+      "grad_norm": 0.1169920489192009,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 54183
+    },
+    {
+      "epoch": 5.564958406079901,
+      "grad_norm": 0.08334633708000183,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 54186
+    },
+    {
+      "epoch": 5.565266509191742,
+      "grad_norm": 0.04524173587560654,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 54189
+    },
+    {
+      "epoch": 5.5655746123035845,
+      "grad_norm": 0.04702974855899811,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 54192
+    },
+    {
+      "epoch": 5.565882715415426,
+      "grad_norm": 0.052152637392282486,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 54195
+    },
+    {
+      "epoch": 5.566190818527267,
+      "grad_norm": 0.07976489514112473,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 54198
+    },
+    {
+      "epoch": 5.566498921639108,
+      "grad_norm": 0.10382959991693497,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 54201
+    },
+    {
+      "epoch": 5.5668070247509505,
+      "grad_norm": 0.062233816832304,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 54204
+    },
+    {
+      "epoch": 5.567115127862792,
+      "grad_norm": 0.0779416561126709,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 54207
+    },
+    {
+      "epoch": 5.567423230974633,
+      "grad_norm": 0.05779222771525383,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 54210
+    },
+    {
+      "epoch": 5.567731334086474,
+      "grad_norm": 0.035859446972608566,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 54213
+    },
+    {
+      "epoch": 5.5680394371983155,
+      "grad_norm": 0.03528286889195442,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 54216
+    },
+    {
+      "epoch": 5.568347540310157,
+      "grad_norm": 0.03673080727458,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 54219
+    },
+    {
+      "epoch": 5.568655643421999,
+      "grad_norm": 0.10049016028642654,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 54222
+    },
+    {
+      "epoch": 5.56896374653384,
+      "grad_norm": 0.09414393454790115,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 54225
+    },
+    {
+      "epoch": 5.5692718496456814,
+      "grad_norm": 0.09574013203382492,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 54228
+    },
+    {
+      "epoch": 5.569579952757523,
+      "grad_norm": 0.06100035086274147,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 54231
+    },
+    {
+      "epoch": 5.569888055869364,
+      "grad_norm": 0.03801640495657921,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 54234
+    },
+    {
+      "epoch": 5.570196158981206,
+      "grad_norm": 0.04721992835402489,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 54237
+    },
+    {
+      "epoch": 5.570504262093047,
+      "grad_norm": 0.14547844231128693,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 54240
+    },
+    {
+      "epoch": 5.570812365204889,
+      "grad_norm": 0.0706636980175972,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 54243
+    },
+    {
+      "epoch": 5.57112046831673,
+      "grad_norm": 0.05795181170105934,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 54246
+    },
+    {
+      "epoch": 5.571428571428571,
+      "grad_norm": 0.04111074283719063,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 54249
+    },
+    {
+      "epoch": 5.571736674540412,
+      "grad_norm": 0.07275404036045074,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 54252
+    },
+    {
+      "epoch": 5.572044777652255,
+      "grad_norm": 0.06297045946121216,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 54255
+    },
+    {
+      "epoch": 5.572352880764096,
+      "grad_norm": 0.03748214617371559,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 54258
+    },
+    {
+      "epoch": 5.572660983875937,
+      "grad_norm": 0.0806899219751358,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 54261
+    },
+    {
+      "epoch": 5.572969086987778,
+      "grad_norm": 0.07261806726455688,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 54264
+    },
+    {
+      "epoch": 5.57327719009962,
+      "grad_norm": 0.09828619658946991,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 54267
+    },
+    {
+      "epoch": 5.573585293211462,
+      "grad_norm": 0.057755280286073685,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 54270
+    },
+    {
+      "epoch": 5.573893396323303,
+      "grad_norm": 0.053999438881874084,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 54273
+    },
+    {
+      "epoch": 5.574201499435144,
+      "grad_norm": 0.058348219841718674,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 54276
+    },
+    {
+      "epoch": 5.574509602546986,
+      "grad_norm": 0.0748370811343193,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 54279
+    },
+    {
+      "epoch": 5.574817705658827,
+      "grad_norm": 0.12268301099538803,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 54282
+    },
+    {
+      "epoch": 5.575125808770668,
+      "grad_norm": 0.04664819315075874,
+      "learning_rate": 0.01,
+      "loss": 2.0232,
+      "step": 54285
+    },
+    {
+      "epoch": 5.57543391188251,
+      "grad_norm": 0.061234939843416214,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 54288
+    },
+    {
+      "epoch": 5.5757420149943515,
+      "grad_norm": 0.034011028707027435,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 54291
+    },
+    {
+      "epoch": 5.576050118106193,
+      "grad_norm": 0.03644438087940216,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 54294
+    },
+    {
+      "epoch": 5.576358221218034,
+      "grad_norm": 0.03703255578875542,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 54297
+    },
+    {
+      "epoch": 5.576666324329876,
+      "grad_norm": 0.06192943826317787,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 54300
+    },
+    {
+      "epoch": 5.5769744274417175,
+      "grad_norm": 0.07393354922533035,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 54303
+    },
+    {
+      "epoch": 5.577282530553559,
+      "grad_norm": 0.10186754167079926,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 54306
+    },
+    {
+      "epoch": 5.5775906336654,
+      "grad_norm": 0.05970088392496109,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 54309
+    },
+    {
+      "epoch": 5.577898736777241,
+      "grad_norm": 0.07863925397396088,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 54312
+    },
+    {
+      "epoch": 5.5782068398890825,
+      "grad_norm": 0.08788790553808212,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 54315
+    },
+    {
+      "epoch": 5.578514943000925,
+      "grad_norm": 0.04963497817516327,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 54318
+    },
+    {
+      "epoch": 5.578823046112766,
+      "grad_norm": 0.06898415833711624,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 54321
+    },
+    {
+      "epoch": 5.579131149224607,
+      "grad_norm": 0.07146736234426498,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 54324
+    },
+    {
+      "epoch": 5.579439252336448,
+      "grad_norm": 0.08930405974388123,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 54327
+    },
+    {
+      "epoch": 5.57974735544829,
+      "grad_norm": 0.11615147441625595,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 54330
+    },
+    {
+      "epoch": 5.580055458560132,
+      "grad_norm": 0.12027255445718765,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 54333
+    },
+    {
+      "epoch": 5.580363561671973,
+      "grad_norm": 0.09193901717662811,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 54336
+    },
+    {
+      "epoch": 5.580671664783814,
+      "grad_norm": 0.0674210712313652,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 54339
+    },
+    {
+      "epoch": 5.580979767895656,
+      "grad_norm": 0.05408472940325737,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 54342
+    },
+    {
+      "epoch": 5.581287871007497,
+      "grad_norm": 0.04289040341973305,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 54345
+    },
+    {
+      "epoch": 5.581595974119338,
+      "grad_norm": 0.04341943934559822,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 54348
+    },
+    {
+      "epoch": 5.58190407723118,
+      "grad_norm": 0.03630705177783966,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 54351
+    },
+    {
+      "epoch": 5.582212180343022,
+      "grad_norm": 0.10383928567171097,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 54354
+    },
+    {
+      "epoch": 5.582520283454863,
+      "grad_norm": 0.042623065412044525,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 54357
+    },
+    {
+      "epoch": 5.582828386566704,
+      "grad_norm": 0.07011984288692474,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 54360
+    },
+    {
+      "epoch": 5.583136489678545,
+      "grad_norm": 0.09444596618413925,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 54363
+    },
+    {
+      "epoch": 5.5834445927903875,
+      "grad_norm": 0.0737951472401619,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 54366
+    },
+    {
+      "epoch": 5.583752695902229,
+      "grad_norm": 0.06040029227733612,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 54369
+    },
+    {
+      "epoch": 5.58406079901407,
+      "grad_norm": 0.059084903448820114,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 54372
+    },
+    {
+      "epoch": 5.584368902125911,
+      "grad_norm": 0.043141599744558334,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 54375
+    },
+    {
+      "epoch": 5.584677005237753,
+      "grad_norm": 0.07285004109144211,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 54378
+    },
+    {
+      "epoch": 5.584985108349595,
+      "grad_norm": 0.05245014652609825,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 54381
+    },
+    {
+      "epoch": 5.585293211461436,
+      "grad_norm": 0.06724348664283752,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 54384
+    },
+    {
+      "epoch": 5.585601314573277,
+      "grad_norm": 0.0695512667298317,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 54387
+    },
+    {
+      "epoch": 5.5859094176851185,
+      "grad_norm": 0.08221160620450974,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 54390
+    },
+    {
+      "epoch": 5.58621752079696,
+      "grad_norm": 0.04641614854335785,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 54393
+    },
+    {
+      "epoch": 5.586525623908802,
+      "grad_norm": 0.04592394083738327,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 54396
+    },
+    {
+      "epoch": 5.586833727020643,
+      "grad_norm": 0.04568556696176529,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 54399
+    },
+    {
+      "epoch": 5.587141830132484,
+      "grad_norm": 0.05011072754859924,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 54402
+    },
+    {
+      "epoch": 5.587449933244326,
+      "grad_norm": 0.05919931083917618,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 54405
+    },
+    {
+      "epoch": 5.587758036356167,
+      "grad_norm": 0.1045934408903122,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 54408
+    },
+    {
+      "epoch": 5.588066139468008,
+      "grad_norm": 0.06998443603515625,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 54411
+    },
+    {
+      "epoch": 5.58837424257985,
+      "grad_norm": 0.07034524530172348,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 54414
+    },
+    {
+      "epoch": 5.588682345691692,
+      "grad_norm": 0.10572990775108337,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 54417
+    },
+    {
+      "epoch": 5.588990448803533,
+      "grad_norm": 0.04444350302219391,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 54420
+    },
+    {
+      "epoch": 5.589298551915374,
+      "grad_norm": 0.06670062988996506,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 54423
+    },
+    {
+      "epoch": 5.589606655027215,
+      "grad_norm": 0.08078444004058838,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 54426
+    },
+    {
+      "epoch": 5.589914758139058,
+      "grad_norm": 0.06824928522109985,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 54429
+    },
+    {
+      "epoch": 5.590222861250899,
+      "grad_norm": 0.04053177684545517,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 54432
+    },
+    {
+      "epoch": 5.59053096436274,
+      "grad_norm": 0.07821489870548248,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 54435
+    },
+    {
+      "epoch": 5.590839067474581,
+      "grad_norm": 0.09564534574747086,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 54438
+    },
+    {
+      "epoch": 5.591147170586423,
+      "grad_norm": 0.06303589791059494,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 54441
+    },
+    {
+      "epoch": 5.591455273698264,
+      "grad_norm": 0.08767011016607285,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 54444
+    },
+    {
+      "epoch": 5.591763376810106,
+      "grad_norm": 0.06719005852937698,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 54447
+    },
+    {
+      "epoch": 5.592071479921947,
+      "grad_norm": 0.0828474760055542,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 54450
+    },
+    {
+      "epoch": 5.592379583033789,
+      "grad_norm": 0.07154206931591034,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 54453
+    },
+    {
+      "epoch": 5.59268768614563,
+      "grad_norm": 0.0818270891904831,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 54456
+    },
+    {
+      "epoch": 5.592995789257472,
+      "grad_norm": 0.046949781477451324,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 54459
+    },
+    {
+      "epoch": 5.593303892369313,
+      "grad_norm": 0.17068155109882355,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 54462
+    },
+    {
+      "epoch": 5.5936119954811545,
+      "grad_norm": 0.14092612266540527,
+      "learning_rate": 0.01,
+      "loss": 2.0142,
+      "step": 54465
+    },
+    {
+      "epoch": 5.593920098592996,
+      "grad_norm": 0.12458490580320358,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 54468
+    },
+    {
+      "epoch": 5.594228201704837,
+      "grad_norm": 0.07861830294132233,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 54471
+    },
+    {
+      "epoch": 5.594536304816678,
+      "grad_norm": 0.0819336324930191,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 54474
+    },
+    {
+      "epoch": 5.5948444079285204,
+      "grad_norm": 0.04734665900468826,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 54477
+    },
+    {
+      "epoch": 5.595152511040362,
+      "grad_norm": 0.04812568798661232,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 54480
+    },
+    {
+      "epoch": 5.595460614152203,
+      "grad_norm": 0.03981874883174896,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 54483
+    },
+    {
+      "epoch": 5.595768717264044,
+      "grad_norm": 0.034884098917245865,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 54486
+    },
+    {
+      "epoch": 5.5960768203758855,
+      "grad_norm": 0.04170473292469978,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 54489
+    },
+    {
+      "epoch": 5.596384923487728,
+      "grad_norm": 0.06037883087992668,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 54492
+    },
+    {
+      "epoch": 5.596693026599569,
+      "grad_norm": 0.06271529197692871,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 54495
+    },
+    {
+      "epoch": 5.59700112971141,
+      "grad_norm": 0.0646272599697113,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 54498
+    },
+    {
+      "epoch": 5.597309232823251,
+      "grad_norm": 0.05499435216188431,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 54501
+    },
+    {
+      "epoch": 5.597617335935093,
+      "grad_norm": 0.06327464431524277,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 54504
+    },
+    {
+      "epoch": 5.597925439046934,
+      "grad_norm": 0.07843179255723953,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 54507
+    },
+    {
+      "epoch": 5.598233542158776,
+      "grad_norm": 0.13546060025691986,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 54510
+    },
+    {
+      "epoch": 5.598541645270617,
+      "grad_norm": 0.05903356522321701,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 54513
+    },
+    {
+      "epoch": 5.598849748382459,
+      "grad_norm": 0.10746420919895172,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 54516
+    },
+    {
+      "epoch": 5.5991578514943,
+      "grad_norm": 0.07332249730825424,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 54519
+    },
+    {
+      "epoch": 5.599465954606141,
+      "grad_norm": 0.05419987067580223,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 54522
+    },
+    {
+      "epoch": 5.599774057717983,
+      "grad_norm": 0.06261706352233887,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 54525
+    },
+    {
+      "epoch": 5.600082160829825,
+      "grad_norm": 0.06889716535806656,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 54528
+    },
+    {
+      "epoch": 5.600390263941666,
+      "grad_norm": 0.07229785621166229,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 54531
+    },
+    {
+      "epoch": 5.600698367053507,
+      "grad_norm": 0.07027054578065872,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 54534
+    },
+    {
+      "epoch": 5.601006470165348,
+      "grad_norm": 0.0769757479429245,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 54537
+    },
+    {
+      "epoch": 5.60131457327719,
+      "grad_norm": 0.07835566997528076,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 54540
+    },
+    {
+      "epoch": 5.601622676389032,
+      "grad_norm": 0.059996653348207474,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 54543
+    },
+    {
+      "epoch": 5.601930779500873,
+      "grad_norm": 0.0982513576745987,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 54546
+    },
+    {
+      "epoch": 5.602238882612714,
+      "grad_norm": 0.08362521976232529,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 54549
+    },
+    {
+      "epoch": 5.602546985724556,
+      "grad_norm": 0.0593801811337471,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 54552
+    },
+    {
+      "epoch": 5.602855088836398,
+      "grad_norm": 0.09806974232196808,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 54555
+    },
+    {
+      "epoch": 5.603163191948239,
+      "grad_norm": 0.041621893644332886,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 54558
+    },
+    {
+      "epoch": 5.60347129506008,
+      "grad_norm": 0.04981600120663643,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 54561
+    },
+    {
+      "epoch": 5.6037793981719215,
+      "grad_norm": 0.08302900940179825,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 54564
+    },
+    {
+      "epoch": 5.604087501283763,
+      "grad_norm": 0.03515220060944557,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 54567
+    },
+    {
+      "epoch": 5.604395604395604,
+      "grad_norm": 0.0783744752407074,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 54570
+    },
+    {
+      "epoch": 5.604703707507446,
+      "grad_norm": 0.03960327431559563,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 54573
+    },
+    {
+      "epoch": 5.605011810619287,
+      "grad_norm": 0.04865705221891403,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 54576
+    },
+    {
+      "epoch": 5.605319913731129,
+      "grad_norm": 0.03959539160132408,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 54579
+    },
+    {
+      "epoch": 5.60562801684297,
+      "grad_norm": 0.08320534974336624,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 54582
+    },
+    {
+      "epoch": 5.605936119954811,
+      "grad_norm": 0.038179684430360794,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 54585
+    },
+    {
+      "epoch": 5.606244223066653,
+      "grad_norm": 0.04269682243466377,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 54588
+    },
+    {
+      "epoch": 5.606552326178495,
+      "grad_norm": 0.08585328608751297,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 54591
+    },
+    {
+      "epoch": 5.606860429290336,
+      "grad_norm": 0.04968829080462456,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 54594
+    },
+    {
+      "epoch": 5.607168532402177,
+      "grad_norm": 0.04485391452908516,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 54597
+    },
+    {
+      "epoch": 5.607476635514018,
+      "grad_norm": 0.03415564075112343,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 54600
+    },
+    {
+      "epoch": 5.60778473862586,
+      "grad_norm": 0.10459756851196289,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 54603
+    },
+    {
+      "epoch": 5.608092841737702,
+      "grad_norm": 0.04644683003425598,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 54606
+    },
+    {
+      "epoch": 5.608400944849543,
+      "grad_norm": 0.09354628622531891,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 54609
+    },
+    {
+      "epoch": 5.608709047961384,
+      "grad_norm": 0.07625284790992737,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 54612
+    },
+    {
+      "epoch": 5.609017151073226,
+      "grad_norm": 0.07845180481672287,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 54615
+    },
+    {
+      "epoch": 5.609325254185067,
+      "grad_norm": 0.060051124542951584,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 54618
+    },
+    {
+      "epoch": 5.609633357296909,
+      "grad_norm": 0.07250028848648071,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 54621
+    },
+    {
+      "epoch": 5.60994146040875,
+      "grad_norm": 0.10221341997385025,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 54624
+    },
+    {
+      "epoch": 5.610249563520592,
+      "grad_norm": 0.03648064285516739,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 54627
+    },
+    {
+      "epoch": 5.610557666632433,
+      "grad_norm": 0.04965173453092575,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 54630
+    },
+    {
+      "epoch": 5.610865769744274,
+      "grad_norm": 0.04213662073016167,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 54633
+    },
+    {
+      "epoch": 5.611173872856115,
+      "grad_norm": 0.11422538757324219,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 54636
+    },
+    {
+      "epoch": 5.6114819759679575,
+      "grad_norm": 0.054801687598228455,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 54639
+    },
+    {
+      "epoch": 5.611790079079799,
+      "grad_norm": 0.06683303415775299,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 54642
+    },
+    {
+      "epoch": 5.61209818219164,
+      "grad_norm": 0.04719147831201553,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 54645
+    },
+    {
+      "epoch": 5.612406285303481,
+      "grad_norm": 0.040852706879377365,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 54648
+    },
+    {
+      "epoch": 5.612714388415323,
+      "grad_norm": 0.038343146443367004,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 54651
+    },
+    {
+      "epoch": 5.613022491527165,
+      "grad_norm": 0.03426671028137207,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 54654
+    },
+    {
+      "epoch": 5.613330594639006,
+      "grad_norm": 0.08030149340629578,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 54657
+    },
+    {
+      "epoch": 5.613638697750847,
+      "grad_norm": 0.04974358528852463,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 54660
+    },
+    {
+      "epoch": 5.6139468008626885,
+      "grad_norm": 0.04733881726861,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 54663
+    },
+    {
+      "epoch": 5.61425490397453,
+      "grad_norm": 0.04850970208644867,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 54666
+    },
+    {
+      "epoch": 5.614563007086372,
+      "grad_norm": 0.08880652487277985,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 54669
+    },
+    {
+      "epoch": 5.614871110198213,
+      "grad_norm": 0.07070966809988022,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 54672
+    },
+    {
+      "epoch": 5.615179213310054,
+      "grad_norm": 0.043020039796829224,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 54675
+    },
+    {
+      "epoch": 5.615487316421896,
+      "grad_norm": 0.06512738764286041,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 54678
+    },
+    {
+      "epoch": 5.615795419533737,
+      "grad_norm": 0.09249335527420044,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 54681
+    },
+    {
+      "epoch": 5.616103522645579,
+      "grad_norm": 0.12002009153366089,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 54684
+    },
+    {
+      "epoch": 5.61641162575742,
+      "grad_norm": 0.046711284667253494,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 54687
+    },
+    {
+      "epoch": 5.616719728869262,
+      "grad_norm": 0.0453648641705513,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 54690
+    },
+    {
+      "epoch": 5.617027831981103,
+      "grad_norm": 0.06286431849002838,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 54693
+    },
+    {
+      "epoch": 5.617335935092944,
+      "grad_norm": 0.10770335048437119,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 54696
+    },
+    {
+      "epoch": 5.617644038204785,
+      "grad_norm": 0.038174375891685486,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 54699
+    },
+    {
+      "epoch": 5.617952141316628,
+      "grad_norm": 0.0749521404504776,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 54702
+    },
+    {
+      "epoch": 5.618260244428469,
+      "grad_norm": 0.08033082634210587,
+      "learning_rate": 0.01,
+      "loss": 2.0188,
+      "step": 54705
+    },
+    {
+      "epoch": 5.61856834754031,
+      "grad_norm": 0.06879189610481262,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 54708
+    },
+    {
+      "epoch": 5.618876450652151,
+      "grad_norm": 0.06201532483100891,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 54711
+    },
+    {
+      "epoch": 5.6191845537639935,
+      "grad_norm": 0.0482785739004612,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 54714
+    },
+    {
+      "epoch": 5.619492656875835,
+      "grad_norm": 0.08434705436229706,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 54717
+    },
+    {
+      "epoch": 5.619800759987676,
+      "grad_norm": 0.06971675902605057,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 54720
+    },
+    {
+      "epoch": 5.620108863099517,
+      "grad_norm": 0.05243263766169548,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 54723
+    },
+    {
+      "epoch": 5.620416966211359,
+      "grad_norm": 0.05071935057640076,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 54726
+    },
+    {
+      "epoch": 5.6207250693232,
+      "grad_norm": 0.07799027860164642,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 54729
+    },
+    {
+      "epoch": 5.621033172435042,
+      "grad_norm": 0.05585414171218872,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 54732
+    },
+    {
+      "epoch": 5.621341275546883,
+      "grad_norm": 0.04890783876180649,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 54735
+    },
+    {
+      "epoch": 5.6216493786587245,
+      "grad_norm": 0.035397402942180634,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 54738
+    },
+    {
+      "epoch": 5.621957481770566,
+      "grad_norm": 0.04959969222545624,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 54741
+    },
+    {
+      "epoch": 5.622265584882407,
+      "grad_norm": 0.1236480176448822,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 54744
+    },
+    {
+      "epoch": 5.622573687994249,
+      "grad_norm": 0.05167923495173454,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 54747
+    },
+    {
+      "epoch": 5.62288179110609,
+      "grad_norm": 0.08985866606235504,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 54750
+    },
+    {
+      "epoch": 5.623189894217932,
+      "grad_norm": 0.051048051565885544,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 54753
+    },
+    {
+      "epoch": 5.623497997329773,
+      "grad_norm": 0.034315209835767746,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 54756
+    },
+    {
+      "epoch": 5.623806100441614,
+      "grad_norm": 0.034064918756484985,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 54759
+    },
+    {
+      "epoch": 5.6241142035534555,
+      "grad_norm": 0.04847462847828865,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 54762
+    },
+    {
+      "epoch": 5.624422306665298,
+      "grad_norm": 0.036340802907943726,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 54765
+    },
+    {
+      "epoch": 5.624730409777139,
+      "grad_norm": 0.07472583651542664,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 54768
+    },
+    {
+      "epoch": 5.62503851288898,
+      "grad_norm": 0.05728251487016678,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 54771
+    },
+    {
+      "epoch": 5.625346616000821,
+      "grad_norm": 0.08100686967372894,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 54774
+    },
+    {
+      "epoch": 5.625654719112663,
+      "grad_norm": 0.03416838496923447,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 54777
+    },
+    {
+      "epoch": 5.625962822224505,
+      "grad_norm": 0.08783119171857834,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 54780
+    },
+    {
+      "epoch": 5.626270925336346,
+      "grad_norm": 0.06878135353326797,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 54783
+    },
+    {
+      "epoch": 5.626579028448187,
+      "grad_norm": 0.08717949688434601,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 54786
+    },
+    {
+      "epoch": 5.626887131560029,
+      "grad_norm": 0.06441211700439453,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 54789
+    },
+    {
+      "epoch": 5.62719523467187,
+      "grad_norm": 0.0610957108438015,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 54792
+    },
+    {
+      "epoch": 5.627503337783711,
+      "grad_norm": 0.04509379714727402,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 54795
+    },
+    {
+      "epoch": 5.627811440895553,
+      "grad_norm": 0.05279795825481415,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 54798
+    },
+    {
+      "epoch": 5.628119544007395,
+      "grad_norm": 0.03723832964897156,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 54801
+    },
+    {
+      "epoch": 5.628427647119236,
+      "grad_norm": 0.10692798346281052,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 54804
+    },
+    {
+      "epoch": 5.628735750231077,
+      "grad_norm": 0.09552669525146484,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 54807
+    },
+    {
+      "epoch": 5.629043853342919,
+      "grad_norm": 0.11247537285089493,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 54810
+    },
+    {
+      "epoch": 5.6293519564547605,
+      "grad_norm": 0.08334960043430328,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 54813
+    },
+    {
+      "epoch": 5.629660059566602,
+      "grad_norm": 0.03950768709182739,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 54816
+    },
+    {
+      "epoch": 5.629968162678443,
+      "grad_norm": 0.032677773386240005,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 54819
+    },
+    {
+      "epoch": 5.630276265790284,
+      "grad_norm": 0.0373060442507267,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 54822
+    },
+    {
+      "epoch": 5.6305843689021255,
+      "grad_norm": 0.07568337768316269,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 54825
+    },
+    {
+      "epoch": 5.630892472013968,
+      "grad_norm": 0.08441541343927383,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 54828
+    },
+    {
+      "epoch": 5.631200575125809,
+      "grad_norm": 0.06486140936613083,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 54831
+    },
+    {
+      "epoch": 5.63150867823765,
+      "grad_norm": 0.06372927129268646,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 54834
+    },
+    {
+      "epoch": 5.6318167813494915,
+      "grad_norm": 0.06529858708381653,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 54837
+    },
+    {
+      "epoch": 5.632124884461333,
+      "grad_norm": 0.08682098984718323,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 54840
+    },
+    {
+      "epoch": 5.632432987573175,
+      "grad_norm": 0.05896419286727905,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 54843
+    },
+    {
+      "epoch": 5.632741090685016,
+      "grad_norm": 0.06567796319723129,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 54846
+    },
+    {
+      "epoch": 5.633049193796857,
+      "grad_norm": 0.06519652158021927,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 54849
+    },
+    {
+      "epoch": 5.633357296908699,
+      "grad_norm": 0.09215736389160156,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 54852
+    },
+    {
+      "epoch": 5.63366540002054,
+      "grad_norm": 0.08821207284927368,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 54855
+    },
+    {
+      "epoch": 5.633973503132381,
+      "grad_norm": 0.06266074627637863,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 54858
+    },
+    {
+      "epoch": 5.634281606244223,
+      "grad_norm": 0.11765746027231216,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 54861
+    },
+    {
+      "epoch": 5.634589709356065,
+      "grad_norm": 0.07203897088766098,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 54864
+    },
+    {
+      "epoch": 5.634897812467906,
+      "grad_norm": 0.05088105797767639,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 54867
+    },
+    {
+      "epoch": 5.635205915579747,
+      "grad_norm": 0.04583646357059479,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 54870
+    },
+    {
+      "epoch": 5.635514018691588,
+      "grad_norm": 0.03866413235664368,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 54873
+    },
+    {
+      "epoch": 5.635822121803431,
+      "grad_norm": 0.02919906936585903,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 54876
+    },
+    {
+      "epoch": 5.636130224915272,
+      "grad_norm": 0.04621945321559906,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 54879
+    },
+    {
+      "epoch": 5.636438328027113,
+      "grad_norm": 0.131577268242836,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 54882
+    },
+    {
+      "epoch": 5.636746431138954,
+      "grad_norm": 0.13305455446243286,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 54885
+    },
+    {
+      "epoch": 5.637054534250796,
+      "grad_norm": 0.046413157135248184,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 54888
+    },
+    {
+      "epoch": 5.637362637362637,
+      "grad_norm": 0.0828058049082756,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 54891
+    },
+    {
+      "epoch": 5.637670740474479,
+      "grad_norm": 0.09685652703046799,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 54894
+    },
+    {
+      "epoch": 5.63797884358632,
+      "grad_norm": 0.055401384830474854,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 54897
+    },
+    {
+      "epoch": 5.6382869466981616,
+      "grad_norm": 0.04287857934832573,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 54900
+    },
+    {
+      "epoch": 5.638595049810003,
+      "grad_norm": 0.09044301509857178,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 54903
+    },
+    {
+      "epoch": 5.638903152921845,
+      "grad_norm": 0.07684798538684845,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 54906
+    },
+    {
+      "epoch": 5.639211256033686,
+      "grad_norm": 0.05879480764269829,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 54909
+    },
+    {
+      "epoch": 5.6395193591455275,
+      "grad_norm": 0.04824329912662506,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 54912
+    },
+    {
+      "epoch": 5.639827462257369,
+      "grad_norm": 0.0385722778737545,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 54915
+    },
+    {
+      "epoch": 5.64013556536921,
+      "grad_norm": 0.10394332557916641,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 54918
+    },
+    {
+      "epoch": 5.640443668481051,
+      "grad_norm": 0.08453499525785446,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 54921
+    },
+    {
+      "epoch": 5.640751771592893,
+      "grad_norm": 0.037400417029857635,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 54924
+    },
+    {
+      "epoch": 5.641059874704735,
+      "grad_norm": 0.04908459633588791,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 54927
+    },
+    {
+      "epoch": 5.641367977816576,
+      "grad_norm": 0.06965642422437668,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 54930
+    },
+    {
+      "epoch": 5.641676080928417,
+      "grad_norm": 0.11287185549736023,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 54933
+    },
+    {
+      "epoch": 5.6419841840402585,
+      "grad_norm": 0.039775166660547256,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 54936
+    },
+    {
+      "epoch": 5.642292287152101,
+      "grad_norm": 0.05167205631732941,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 54939
+    },
+    {
+      "epoch": 5.642600390263942,
+      "grad_norm": 0.09172987937927246,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 54942
+    },
+    {
+      "epoch": 5.642908493375783,
+      "grad_norm": 0.11361383646726608,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 54945
+    },
+    {
+      "epoch": 5.643216596487624,
+      "grad_norm": 0.07098819315433502,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 54948
+    },
+    {
+      "epoch": 5.643524699599466,
+      "grad_norm": 0.06430887430906296,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 54951
+    },
+    {
+      "epoch": 5.643832802711307,
+      "grad_norm": 0.04281394183635712,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 54954
+    },
+    {
+      "epoch": 5.644140905823149,
+      "grad_norm": 0.04618782550096512,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 54957
+    },
+    {
+      "epoch": 5.64444900893499,
+      "grad_norm": 0.049203407019376755,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 54960
+    },
+    {
+      "epoch": 5.644757112046832,
+      "grad_norm": 0.07782324403524399,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 54963
+    },
+    {
+      "epoch": 5.645065215158673,
+      "grad_norm": 0.11196355521678925,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 54966
+    },
+    {
+      "epoch": 5.645373318270515,
+      "grad_norm": 0.08459745347499847,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 54969
+    },
+    {
+      "epoch": 5.645681421382356,
+      "grad_norm": 0.05610903725028038,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 54972
+    },
+    {
+      "epoch": 5.6459895244941976,
+      "grad_norm": 0.07638850063085556,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 54975
+    },
+    {
+      "epoch": 5.646297627606039,
+      "grad_norm": 0.0996907651424408,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 54978
+    },
+    {
+      "epoch": 5.64660573071788,
+      "grad_norm": 0.05218489095568657,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 54981
+    },
+    {
+      "epoch": 5.646913833829721,
+      "grad_norm": 0.05683809146285057,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 54984
+    },
+    {
+      "epoch": 5.6472219369415635,
+      "grad_norm": 0.08494491875171661,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 54987
+    },
+    {
+      "epoch": 5.647530040053405,
+      "grad_norm": 0.09330440312623978,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 54990
+    },
+    {
+      "epoch": 5.647838143165246,
+      "grad_norm": 0.1456758677959442,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 54993
+    },
+    {
+      "epoch": 5.648146246277087,
+      "grad_norm": 0.13807424902915955,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 54996
+    },
+    {
+      "epoch": 5.6484543493889285,
+      "grad_norm": 0.0649813711643219,
+      "learning_rate": 0.01,
+      "loss": 2.017,
+      "step": 54999
+    },
+    {
+      "epoch": 5.648762452500771,
+      "grad_norm": 0.0742340162396431,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 55002
+    },
+    {
+      "epoch": 5.649070555612612,
+      "grad_norm": 0.09716660529375076,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 55005
+    },
+    {
+      "epoch": 5.649378658724453,
+      "grad_norm": 0.056228429079055786,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 55008
+    },
+    {
+      "epoch": 5.6496867618362945,
+      "grad_norm": 0.045122385025024414,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 55011
+    },
+    {
+      "epoch": 5.649994864948136,
+      "grad_norm": 0.0388353131711483,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 55014
+    },
+    {
+      "epoch": 5.650302968059977,
+      "grad_norm": 0.03766791522502899,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 55017
+    },
+    {
+      "epoch": 5.650611071171819,
+      "grad_norm": 0.037102095782756805,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 55020
+    },
+    {
+      "epoch": 5.65091917428366,
+      "grad_norm": 0.03547612577676773,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 55023
+    },
+    {
+      "epoch": 5.651227277395502,
+      "grad_norm": 0.14113888144493103,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 55026
+    },
+    {
+      "epoch": 5.651535380507343,
+      "grad_norm": 0.13051262497901917,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 55029
+    },
+    {
+      "epoch": 5.651843483619184,
+      "grad_norm": 0.050375718623399734,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 55032
+    },
+    {
+      "epoch": 5.652151586731026,
+      "grad_norm": 0.05951124429702759,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 55035
+    },
+    {
+      "epoch": 5.652459689842868,
+      "grad_norm": 0.0422302782535553,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 55038
+    },
+    {
+      "epoch": 5.652767792954709,
+      "grad_norm": 0.04383467882871628,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 55041
+    },
+    {
+      "epoch": 5.65307589606655,
+      "grad_norm": 0.06659442186355591,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 55044
+    },
+    {
+      "epoch": 5.653383999178391,
+      "grad_norm": 0.03979937732219696,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 55047
+    },
+    {
+      "epoch": 5.653692102290233,
+      "grad_norm": 0.07855933159589767,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 55050
+    },
+    {
+      "epoch": 5.654000205402075,
+      "grad_norm": 0.06732465326786041,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 55053
+    },
+    {
+      "epoch": 5.654308308513916,
+      "grad_norm": 0.057772446423769,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 55056
+    },
+    {
+      "epoch": 5.654616411625757,
+      "grad_norm": 0.08292173594236374,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 55059
+    },
+    {
+      "epoch": 5.654924514737599,
+      "grad_norm": 0.0935162603855133,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 55062
+    },
+    {
+      "epoch": 5.655232617849441,
+      "grad_norm": 0.06377442181110382,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 55065
+    },
+    {
+      "epoch": 5.655540720961282,
+      "grad_norm": 0.08845983445644379,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 55068
+    },
+    {
+      "epoch": 5.655848824073123,
+      "grad_norm": 0.06194892153143883,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 55071
+    },
+    {
+      "epoch": 5.6561569271849645,
+      "grad_norm": 0.06892132014036179,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 55074
+    },
+    {
+      "epoch": 5.656465030296806,
+      "grad_norm": 0.04041128605604172,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 55077
+    },
+    {
+      "epoch": 5.656773133408647,
+      "grad_norm": 0.09989805519580841,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 55080
+    },
+    {
+      "epoch": 5.657081236520489,
+      "grad_norm": 0.0866503044962883,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 55083
+    },
+    {
+      "epoch": 5.6573893396323305,
+      "grad_norm": 0.043805480003356934,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 55086
+    },
+    {
+      "epoch": 5.657697442744172,
+      "grad_norm": 0.038269441574811935,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 55089
+    },
+    {
+      "epoch": 5.658005545856013,
+      "grad_norm": 0.04652794450521469,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 55092
+    },
+    {
+      "epoch": 5.658313648967854,
+      "grad_norm": 0.09699436277151108,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 55095
+    },
+    {
+      "epoch": 5.658621752079696,
+      "grad_norm": 0.07722003012895584,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 55098
+    },
+    {
+      "epoch": 5.658929855191538,
+      "grad_norm": 0.08195403218269348,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 55101
+    },
+    {
+      "epoch": 5.659237958303379,
+      "grad_norm": 0.04434705525636673,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 55104
+    },
+    {
+      "epoch": 5.65954606141522,
+      "grad_norm": 0.03978164121508598,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 55107
+    },
+    {
+      "epoch": 5.6598541645270615,
+      "grad_norm": 0.1201346144080162,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 55110
+    },
+    {
+      "epoch": 5.660162267638903,
+      "grad_norm": 0.17597338557243347,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 55113
+    },
+    {
+      "epoch": 5.660470370750745,
+      "grad_norm": 0.13311070203781128,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 55116
+    },
+    {
+      "epoch": 5.660778473862586,
+      "grad_norm": 0.11461780220270157,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 55119
+    },
+    {
+      "epoch": 5.661086576974427,
+      "grad_norm": 0.04778537154197693,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 55122
+    },
+    {
+      "epoch": 5.661394680086269,
+      "grad_norm": 0.03900016471743584,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 55125
+    },
+    {
+      "epoch": 5.66170278319811,
+      "grad_norm": 0.03934657946228981,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 55128
+    },
+    {
+      "epoch": 5.662010886309952,
+      "grad_norm": 0.07235165685415268,
+      "learning_rate": 0.01,
+      "loss": 2.0177,
+      "step": 55131
+    },
+    {
+      "epoch": 5.662318989421793,
+      "grad_norm": 0.09168880432844162,
+      "learning_rate": 0.01,
+      "loss": 2.0312,
+      "step": 55134
+    },
+    {
+      "epoch": 5.662627092533635,
+      "grad_norm": 0.058992091566324234,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 55137
+    },
+    {
+      "epoch": 5.662935195645476,
+      "grad_norm": 0.044471897184848785,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 55140
+    },
+    {
+      "epoch": 5.663243298757317,
+      "grad_norm": 0.04644366726279259,
+      "learning_rate": 0.01,
+      "loss": 2.018,
+      "step": 55143
+    },
+    {
+      "epoch": 5.663551401869158,
+      "grad_norm": 0.04064124450087547,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 55146
+    },
+    {
+      "epoch": 5.6638595049810005,
+      "grad_norm": 0.0546630322933197,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 55149
+    },
+    {
+      "epoch": 5.664167608092842,
+      "grad_norm": 0.07874899357557297,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 55152
+    },
+    {
+      "epoch": 5.664475711204683,
+      "grad_norm": 0.10178565979003906,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 55155
+    },
+    {
+      "epoch": 5.664783814316524,
+      "grad_norm": 0.10570824146270752,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 55158
+    },
+    {
+      "epoch": 5.6650919174283665,
+      "grad_norm": 0.046325813978910446,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 55161
+    },
+    {
+      "epoch": 5.665400020540208,
+      "grad_norm": 0.05771588161587715,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 55164
+    },
+    {
+      "epoch": 5.665708123652049,
+      "grad_norm": 0.0589786060154438,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 55167
+    },
+    {
+      "epoch": 5.66601622676389,
+      "grad_norm": 0.043214842677116394,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 55170
+    },
+    {
+      "epoch": 5.6663243298757315,
+      "grad_norm": 0.09702472388744354,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 55173
+    },
+    {
+      "epoch": 5.666632432987573,
+      "grad_norm": 0.04949367418885231,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 55176
+    },
+    {
+      "epoch": 5.666940536099415,
+      "grad_norm": 0.10156387835741043,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 55179
+    },
+    {
+      "epoch": 5.667248639211256,
+      "grad_norm": 0.04134589806199074,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 55182
+    },
+    {
+      "epoch": 5.6675567423230975,
+      "grad_norm": 0.09493035078048706,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 55185
+    },
+    {
+      "epoch": 5.667864845434939,
+      "grad_norm": 0.06814124435186386,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 55188
+    },
+    {
+      "epoch": 5.66817294854678,
+      "grad_norm": 0.05078943446278572,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 55191
+    },
+    {
+      "epoch": 5.668481051658622,
+      "grad_norm": 0.04670390486717224,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 55194
+    },
+    {
+      "epoch": 5.668789154770463,
+      "grad_norm": 0.04085162654519081,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 55197
+    },
+    {
+      "epoch": 5.669097257882305,
+      "grad_norm": 0.05365972965955734,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 55200
+    },
+    {
+      "epoch": 5.669405360994146,
+      "grad_norm": 0.06176181137561798,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 55203
+    },
+    {
+      "epoch": 5.669713464105987,
+      "grad_norm": 0.08370962738990784,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 55206
+    },
+    {
+      "epoch": 5.6700215672178285,
+      "grad_norm": 0.06180374324321747,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 55209
+    },
+    {
+      "epoch": 5.670329670329671,
+      "grad_norm": 0.04040297865867615,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 55212
+    },
+    {
+      "epoch": 5.670637773441512,
+      "grad_norm": 0.041469234973192215,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 55215
+    },
+    {
+      "epoch": 5.670945876553353,
+      "grad_norm": 0.04767376556992531,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 55218
+    },
+    {
+      "epoch": 5.671253979665194,
+      "grad_norm": 0.127590611577034,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 55221
+    },
+    {
+      "epoch": 5.671562082777036,
+      "grad_norm": 0.09273551404476166,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 55224
+    },
+    {
+      "epoch": 5.671870185888878,
+      "grad_norm": 0.06734216213226318,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 55227
+    },
+    {
+      "epoch": 5.672178289000719,
+      "grad_norm": 0.04960544407367706,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 55230
+    },
+    {
+      "epoch": 5.67248639211256,
+      "grad_norm": 0.08085974305868149,
+      "learning_rate": 0.01,
+      "loss": 2.0236,
+      "step": 55233
+    },
+    {
+      "epoch": 5.672794495224402,
+      "grad_norm": 0.09541311115026474,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 55236
+    },
+    {
+      "epoch": 5.673102598336243,
+      "grad_norm": 0.04482199624180794,
+      "learning_rate": 0.01,
+      "loss": 2.0372,
+      "step": 55239
+    },
+    {
+      "epoch": 5.673410701448085,
+      "grad_norm": 0.0415610708296299,
+      "learning_rate": 0.01,
+      "loss": 2.0211,
+      "step": 55242
+    },
+    {
+      "epoch": 5.673718804559926,
+      "grad_norm": 0.08066660910844803,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 55245
+    },
+    {
+      "epoch": 5.6740269076717675,
+      "grad_norm": 0.06819882988929749,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 55248
+    },
+    {
+      "epoch": 5.674335010783609,
+      "grad_norm": 0.10055188089609146,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 55251
+    },
+    {
+      "epoch": 5.67464311389545,
+      "grad_norm": 0.05193830654025078,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 55254
+    },
+    {
+      "epoch": 5.674951217007292,
+      "grad_norm": 0.09766846150159836,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 55257
+    },
+    {
+      "epoch": 5.6752593201191335,
+      "grad_norm": 0.041506506502628326,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 55260
+    },
+    {
+      "epoch": 5.675567423230975,
+      "grad_norm": 0.04589756205677986,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 55263
+    },
+    {
+      "epoch": 5.675875526342816,
+      "grad_norm": 0.05841269716620445,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 55266
+    },
+    {
+      "epoch": 5.676183629454657,
+      "grad_norm": 0.08855876326560974,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 55269
+    },
+    {
+      "epoch": 5.6764917325664985,
+      "grad_norm": 0.055619269609451294,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 55272
+    },
+    {
+      "epoch": 5.676799835678341,
+      "grad_norm": 0.04398469999432564,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 55275
+    },
+    {
+      "epoch": 5.677107938790182,
+      "grad_norm": 0.073618583381176,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 55278
+    },
+    {
+      "epoch": 5.677416041902023,
+      "grad_norm": 0.03379726782441139,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 55281
+    },
+    {
+      "epoch": 5.6777241450138645,
+      "grad_norm": 0.040897171944379807,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 55284
+    },
+    {
+      "epoch": 5.678032248125706,
+      "grad_norm": 0.09364822506904602,
+      "learning_rate": 0.01,
+      "loss": 2.02,
+      "step": 55287
+    },
+    {
+      "epoch": 5.678340351237548,
+      "grad_norm": 0.05096183717250824,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 55290
+    },
+    {
+      "epoch": 5.678648454349389,
+      "grad_norm": 0.03920522332191467,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 55293
+    },
+    {
+      "epoch": 5.67895655746123,
+      "grad_norm": 0.12465393543243408,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 55296
+    },
+    {
+      "epoch": 5.679264660573072,
+      "grad_norm": 0.0732463076710701,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 55299
+    },
+    {
+      "epoch": 5.679572763684913,
+      "grad_norm": 0.048661403357982635,
+      "learning_rate": 0.01,
+      "loss": 2.0204,
+      "step": 55302
+    },
+    {
+      "epoch": 5.679880866796754,
+      "grad_norm": 0.072269506752491,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 55305
+    },
+    {
+      "epoch": 5.680188969908596,
+      "grad_norm": 0.09489680081605911,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 55308
+    },
+    {
+      "epoch": 5.680497073020438,
+      "grad_norm": 0.05124731361865997,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 55311
+    },
+    {
+      "epoch": 5.680805176132279,
+      "grad_norm": 0.05603638291358948,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 55314
+    },
+    {
+      "epoch": 5.68111327924412,
+      "grad_norm": 0.045195356011390686,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 55317
+    },
+    {
+      "epoch": 5.681421382355962,
+      "grad_norm": 0.08664342761039734,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 55320
+    },
+    {
+      "epoch": 5.6817294854678035,
+      "grad_norm": 0.06681100279092789,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 55323
+    },
+    {
+      "epoch": 5.682037588579645,
+      "grad_norm": 0.040817707777023315,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 55326
+    },
+    {
+      "epoch": 5.682345691691486,
+      "grad_norm": 0.04150600731372833,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 55329
+    },
+    {
+      "epoch": 5.682653794803327,
+      "grad_norm": 0.05804332345724106,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 55332
+    },
+    {
+      "epoch": 5.682961897915169,
+      "grad_norm": 0.06679341942071915,
+      "learning_rate": 0.01,
+      "loss": 2.0132,
+      "step": 55335
+    },
+    {
+      "epoch": 5.683270001027011,
+      "grad_norm": 0.1082872524857521,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 55338
+    },
+    {
+      "epoch": 5.683578104138852,
+      "grad_norm": 0.039868131279945374,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 55341
+    },
+    {
+      "epoch": 5.683886207250693,
+      "grad_norm": 0.056761614978313446,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 55344
+    },
+    {
+      "epoch": 5.6841943103625345,
+      "grad_norm": 0.05743727087974548,
+      "learning_rate": 0.01,
+      "loss": 2.0146,
+      "step": 55347
+    },
+    {
+      "epoch": 5.684502413474376,
+      "grad_norm": 0.07240551710128784,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 55350
+    },
+    {
+      "epoch": 5.684810516586218,
+      "grad_norm": 0.04794316738843918,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 55353
+    },
+    {
+      "epoch": 5.685118619698059,
+      "grad_norm": 0.03990946710109711,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 55356
+    },
+    {
+      "epoch": 5.6854267228099005,
+      "grad_norm": 0.030655885115265846,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 55359
+    },
+    {
+      "epoch": 5.685734825921742,
+      "grad_norm": 0.05635961890220642,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 55362
+    },
+    {
+      "epoch": 5.686042929033583,
+      "grad_norm": 0.058308038860559464,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 55365
+    },
+    {
+      "epoch": 5.686351032145424,
+      "grad_norm": 0.08183693885803223,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 55368
+    },
+    {
+      "epoch": 5.686659135257266,
+      "grad_norm": 0.14002621173858643,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 55371
+    },
+    {
+      "epoch": 5.686967238369108,
+      "grad_norm": 0.06427164375782013,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 55374
+    },
+    {
+      "epoch": 5.687275341480949,
+      "grad_norm": 0.03860678896307945,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 55377
+    },
+    {
+      "epoch": 5.68758344459279,
+      "grad_norm": 0.035588864237070084,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 55380
+    },
+    {
+      "epoch": 5.6878915477046315,
+      "grad_norm": 0.035896506160497665,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 55383
+    },
+    {
+      "epoch": 5.688199650816474,
+      "grad_norm": 0.042555298656225204,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 55386
+    },
+    {
+      "epoch": 5.688507753928315,
+      "grad_norm": 0.061510197818279266,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 55389
+    },
+    {
+      "epoch": 5.688815857040156,
+      "grad_norm": 0.058596979826688766,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 55392
+    },
+    {
+      "epoch": 5.689123960151997,
+      "grad_norm": 0.14717184007167816,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 55395
+    },
+    {
+      "epoch": 5.689432063263839,
+      "grad_norm": 0.1496891975402832,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 55398
+    },
+    {
+      "epoch": 5.68974016637568,
+      "grad_norm": 0.09148060530424118,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 55401
+    },
+    {
+      "epoch": 5.690048269487522,
+      "grad_norm": 0.06617487967014313,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 55404
+    },
+    {
+      "epoch": 5.690356372599363,
+      "grad_norm": 0.0447409562766552,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 55407
+    },
+    {
+      "epoch": 5.690664475711205,
+      "grad_norm": 0.07360169291496277,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 55410
+    },
+    {
+      "epoch": 5.690972578823046,
+      "grad_norm": 0.03748020529747009,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 55413
+    },
+    {
+      "epoch": 5.691280681934888,
+      "grad_norm": 0.0812792256474495,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 55416
+    },
+    {
+      "epoch": 5.691588785046729,
+      "grad_norm": 0.06878095865249634,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 55419
+    },
+    {
+      "epoch": 5.6918968881585705,
+      "grad_norm": 0.10537760704755783,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 55422
+    },
+    {
+      "epoch": 5.692204991270412,
+      "grad_norm": 0.04603094980120659,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 55425
+    },
+    {
+      "epoch": 5.692513094382253,
+      "grad_norm": 0.09780580550432205,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 55428
+    },
+    {
+      "epoch": 5.692821197494094,
+      "grad_norm": 0.07139319181442261,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 55431
+    },
+    {
+      "epoch": 5.6931293006059365,
+      "grad_norm": 0.10430671274662018,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 55434
+    },
+    {
+      "epoch": 5.693437403717778,
+      "grad_norm": 0.07455434650182724,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 55437
+    },
+    {
+      "epoch": 5.693745506829619,
+      "grad_norm": 0.06788843870162964,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 55440
+    },
+    {
+      "epoch": 5.69405360994146,
+      "grad_norm": 0.07325445860624313,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 55443
+    },
+    {
+      "epoch": 5.6943617130533015,
+      "grad_norm": 0.06702747941017151,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 55446
+    },
+    {
+      "epoch": 5.694669816165144,
+      "grad_norm": 0.05118415504693985,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 55449
+    },
+    {
+      "epoch": 5.694977919276985,
+      "grad_norm": 0.08692149817943573,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 55452
+    },
+    {
+      "epoch": 5.695286022388826,
+      "grad_norm": 0.11190303415060043,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 55455
+    },
+    {
+      "epoch": 5.6955941255006675,
+      "grad_norm": 0.048586875200271606,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 55458
+    },
+    {
+      "epoch": 5.695902228612509,
+      "grad_norm": 0.04205205664038658,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 55461
+    },
+    {
+      "epoch": 5.69621033172435,
+      "grad_norm": 0.06149724870920181,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 55464
+    },
+    {
+      "epoch": 5.696518434836192,
+      "grad_norm": 0.0992293581366539,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 55467
+    },
+    {
+      "epoch": 5.696826537948033,
+      "grad_norm": 0.13116061687469482,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 55470
+    },
+    {
+      "epoch": 5.697134641059875,
+      "grad_norm": 0.06159098818898201,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 55473
+    },
+    {
+      "epoch": 5.697442744171716,
+      "grad_norm": 0.03453617915511131,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 55476
+    },
+    {
+      "epoch": 5.697750847283557,
+      "grad_norm": 0.09355533123016357,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 55479
+    },
+    {
+      "epoch": 5.698058950395399,
+      "grad_norm": 0.059095028787851334,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 55482
+    },
+    {
+      "epoch": 5.698367053507241,
+      "grad_norm": 0.0825907438993454,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 55485
+    },
+    {
+      "epoch": 5.698675156619082,
+      "grad_norm": 0.07268752157688141,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 55488
+    },
+    {
+      "epoch": 5.698983259730923,
+      "grad_norm": 0.07047848403453827,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 55491
+    },
+    {
+      "epoch": 5.699291362842764,
+      "grad_norm": 0.09214555472135544,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 55494
+    },
+    {
+      "epoch": 5.699599465954606,
+      "grad_norm": 0.11323852092027664,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 55497
+    },
+    {
+      "epoch": 5.699907569066448,
+      "grad_norm": 0.06520719081163406,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 55500
+    },
+    {
+      "epoch": 5.700215672178289,
+      "grad_norm": 0.04669101536273956,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 55503
+    },
+    {
+      "epoch": 5.70052377529013,
+      "grad_norm": 0.04730800911784172,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 55506
+    },
+    {
+      "epoch": 5.700831878401972,
+      "grad_norm": 0.043729908764362335,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 55509
+    },
+    {
+      "epoch": 5.701139981513814,
+      "grad_norm": 0.03472939133644104,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 55512
+    },
+    {
+      "epoch": 5.701448084625655,
+      "grad_norm": 0.050825025886297226,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 55515
+    },
+    {
+      "epoch": 5.701756187737496,
+      "grad_norm": 0.06947685033082962,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 55518
+    },
+    {
+      "epoch": 5.7020642908493375,
+      "grad_norm": 0.036018531769514084,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 55521
+    },
+    {
+      "epoch": 5.702372393961179,
+      "grad_norm": 0.10103366523981094,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 55524
+    },
+    {
+      "epoch": 5.70268049707302,
+      "grad_norm": 0.07310350984334946,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 55527
+    },
+    {
+      "epoch": 5.702988600184862,
+      "grad_norm": 0.13540640473365784,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 55530
+    },
+    {
+      "epoch": 5.7032967032967035,
+      "grad_norm": 0.0949084460735321,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 55533
+    },
+    {
+      "epoch": 5.703604806408545,
+      "grad_norm": 0.059144023805856705,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 55536
+    },
+    {
+      "epoch": 5.703912909520386,
+      "grad_norm": 0.0527474619448185,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 55539
+    },
+    {
+      "epoch": 5.704221012632227,
+      "grad_norm": 0.06749939173460007,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 55542
+    },
+    {
+      "epoch": 5.704529115744069,
+      "grad_norm": 0.03750371187925339,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 55545
+    },
+    {
+      "epoch": 5.704837218855911,
+      "grad_norm": 0.04007263481616974,
+      "learning_rate": 0.01,
+      "loss": 2.0147,
+      "step": 55548
+    },
+    {
+      "epoch": 5.705145321967752,
+      "grad_norm": 0.1166180819272995,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 55551
+    },
+    {
+      "epoch": 5.705453425079593,
+      "grad_norm": 0.09550142288208008,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 55554
+    },
+    {
+      "epoch": 5.7057615281914345,
+      "grad_norm": 0.03766748309135437,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 55557
+    },
+    {
+      "epoch": 5.706069631303276,
+      "grad_norm": 0.05305793881416321,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 55560
+    },
+    {
+      "epoch": 5.706377734415118,
+      "grad_norm": 0.042423125356435776,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 55563
+    },
+    {
+      "epoch": 5.706685837526959,
+      "grad_norm": 0.08636484295129776,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 55566
+    },
+    {
+      "epoch": 5.7069939406388,
+      "grad_norm": 0.05162766948342323,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 55569
+    },
+    {
+      "epoch": 5.707302043750642,
+      "grad_norm": 0.08721368759870529,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 55572
+    },
+    {
+      "epoch": 5.707610146862484,
+      "grad_norm": 0.043255776166915894,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 55575
+    },
+    {
+      "epoch": 5.707918249974325,
+      "grad_norm": 0.0371641106903553,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 55578
+    },
+    {
+      "epoch": 5.708226353086166,
+      "grad_norm": 0.03695603460073471,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 55581
+    },
+    {
+      "epoch": 5.708534456198008,
+      "grad_norm": 0.07745962589979172,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 55584
+    },
+    {
+      "epoch": 5.708842559309849,
+      "grad_norm": 0.10786707699298859,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 55587
+    },
+    {
+      "epoch": 5.70915066242169,
+      "grad_norm": 0.11932874470949173,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 55590
+    },
+    {
+      "epoch": 5.709458765533532,
+      "grad_norm": 0.09362657368183136,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 55593
+    },
+    {
+      "epoch": 5.7097668686453735,
+      "grad_norm": 0.049765367060899734,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 55596
+    },
+    {
+      "epoch": 5.710074971757215,
+      "grad_norm": 0.0707634910941124,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 55599
+    },
+    {
+      "epoch": 5.710383074869056,
+      "grad_norm": 0.05576471611857414,
+      "learning_rate": 0.01,
+      "loss": 2.0247,
+      "step": 55602
+    },
+    {
+      "epoch": 5.710691177980897,
+      "grad_norm": 0.036324393004179,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 55605
+    },
+    {
+      "epoch": 5.7109992810927395,
+      "grad_norm": 0.043876856565475464,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 55608
+    },
+    {
+      "epoch": 5.711307384204581,
+      "grad_norm": 0.04931717738509178,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 55611
+    },
+    {
+      "epoch": 5.711615487316422,
+      "grad_norm": 0.10611497610807419,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 55614
+    },
+    {
+      "epoch": 5.711923590428263,
+      "grad_norm": 0.04808138310909271,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 55617
+    },
+    {
+      "epoch": 5.7122316935401045,
+      "grad_norm": 0.05550553277134895,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 55620
+    },
+    {
+      "epoch": 5.712539796651946,
+      "grad_norm": 0.07001514732837677,
+      "learning_rate": 0.01,
+      "loss": 2.0225,
+      "step": 55623
+    },
+    {
+      "epoch": 5.712847899763788,
+      "grad_norm": 0.04201444238424301,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 55626
+    },
+    {
+      "epoch": 5.713156002875629,
+      "grad_norm": 0.04204641655087471,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 55629
+    },
+    {
+      "epoch": 5.7134641059874705,
+      "grad_norm": 0.07260296493768692,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 55632
+    },
+    {
+      "epoch": 5.713772209099312,
+      "grad_norm": 0.04765797406435013,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 55635
+    },
+    {
+      "epoch": 5.714080312211153,
+      "grad_norm": 0.10189679265022278,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 55638
+    },
+    {
+      "epoch": 5.714388415322995,
+      "grad_norm": 0.08928123861551285,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 55641
+    },
+    {
+      "epoch": 5.714696518434836,
+      "grad_norm": 0.041944585740566254,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 55644
+    },
+    {
+      "epoch": 5.715004621546678,
+      "grad_norm": 0.062304090708494186,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 55647
+    },
+    {
+      "epoch": 5.715312724658519,
+      "grad_norm": 0.03752179071307182,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 55650
+    },
+    {
+      "epoch": 5.71562082777036,
+      "grad_norm": 0.05441533774137497,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 55653
+    },
+    {
+      "epoch": 5.715928930882201,
+      "grad_norm": 0.0389556922018528,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 55656
+    },
+    {
+      "epoch": 5.716237033994044,
+      "grad_norm": 0.046292494982481,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 55659
+    },
+    {
+      "epoch": 5.716545137105885,
+      "grad_norm": 0.0998992770910263,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 55662
+    },
+    {
+      "epoch": 5.716853240217726,
+      "grad_norm": 0.07048474252223969,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 55665
+    },
+    {
+      "epoch": 5.717161343329567,
+      "grad_norm": 0.03293071687221527,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 55668
+    },
+    {
+      "epoch": 5.7174694464414095,
+      "grad_norm": 0.040962520986795425,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 55671
+    },
+    {
+      "epoch": 5.717777549553251,
+      "grad_norm": 0.08064155280590057,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 55674
+    },
+    {
+      "epoch": 5.718085652665092,
+      "grad_norm": 0.10333026945590973,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 55677
+    },
+    {
+      "epoch": 5.718393755776933,
+      "grad_norm": 0.06801532953977585,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 55680
+    },
+    {
+      "epoch": 5.718701858888775,
+      "grad_norm": 0.051722411066293716,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 55683
+    },
+    {
+      "epoch": 5.719009962000616,
+      "grad_norm": 0.08710015565156937,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 55686
+    },
+    {
+      "epoch": 5.719318065112458,
+      "grad_norm": 0.06947915256023407,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 55689
+    },
+    {
+      "epoch": 5.719626168224299,
+      "grad_norm": 0.08588126301765442,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 55692
+    },
+    {
+      "epoch": 5.7199342713361405,
+      "grad_norm": 0.1062588170170784,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 55695
+    },
+    {
+      "epoch": 5.720242374447982,
+      "grad_norm": 0.044331252574920654,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 55698
+    },
+    {
+      "epoch": 5.720550477559823,
+      "grad_norm": 0.06031282618641853,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 55701
+    },
+    {
+      "epoch": 5.720858580671665,
+      "grad_norm": 0.0506250225007534,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 55704
+    },
+    {
+      "epoch": 5.7211666837835065,
+      "grad_norm": 0.04892846941947937,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 55707
+    },
+    {
+      "epoch": 5.721474786895348,
+      "grad_norm": 0.0958351120352745,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 55710
+    },
+    {
+      "epoch": 5.721782890007189,
+      "grad_norm": 0.03931442275643349,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 55713
+    },
+    {
+      "epoch": 5.72209099311903,
+      "grad_norm": 0.10494615882635117,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 55716
+    },
+    {
+      "epoch": 5.7223990962308715,
+      "grad_norm": 0.06329034268856049,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 55719
+    },
+    {
+      "epoch": 5.722707199342714,
+      "grad_norm": 0.08237000554800034,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 55722
+    },
+    {
+      "epoch": 5.723015302454555,
+      "grad_norm": 0.063106007874012,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 55725
+    },
+    {
+      "epoch": 5.723323405566396,
+      "grad_norm": 0.04435192793607712,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 55728
+    },
+    {
+      "epoch": 5.7236315086782374,
+      "grad_norm": 0.09624417126178741,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 55731
+    },
+    {
+      "epoch": 5.723939611790079,
+      "grad_norm": 0.04826802760362625,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 55734
+    },
+    {
+      "epoch": 5.724247714901921,
+      "grad_norm": 0.13376633822917938,
+      "learning_rate": 0.01,
+      "loss": 2.0249,
+      "step": 55737
+    },
+    {
+      "epoch": 5.724555818013762,
+      "grad_norm": 0.10685109347105026,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 55740
+    },
+    {
+      "epoch": 5.724863921125603,
+      "grad_norm": 0.03832371160387993,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 55743
+    },
+    {
+      "epoch": 5.725172024237445,
+      "grad_norm": 0.06955388933420181,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 55746
+    },
+    {
+      "epoch": 5.725480127349286,
+      "grad_norm": 0.10767482221126556,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 55749
+    },
+    {
+      "epoch": 5.725788230461127,
+      "grad_norm": 0.09542632102966309,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 55752
+    },
+    {
+      "epoch": 5.726096333572969,
+      "grad_norm": 0.07101954519748688,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 55755
+    },
+    {
+      "epoch": 5.726404436684811,
+      "grad_norm": 0.04699917882680893,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 55758
+    },
+    {
+      "epoch": 5.726712539796652,
+      "grad_norm": 0.03901418298482895,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 55761
+    },
+    {
+      "epoch": 5.727020642908493,
+      "grad_norm": 0.03572557494044304,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 55764
+    },
+    {
+      "epoch": 5.727328746020335,
+      "grad_norm": 0.048515480011701584,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 55767
+    },
+    {
+      "epoch": 5.7276368491321765,
+      "grad_norm": 0.09653262794017792,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 55770
+    },
+    {
+      "epoch": 5.727944952244018,
+      "grad_norm": 0.05700475722551346,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 55773
+    },
+    {
+      "epoch": 5.728253055355859,
+      "grad_norm": 0.1171969622373581,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 55776
+    },
+    {
+      "epoch": 5.7285611584677,
+      "grad_norm": 0.043651919811964035,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 55779
+    },
+    {
+      "epoch": 5.728869261579542,
+      "grad_norm": 0.034597914665937424,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 55782
+    },
+    {
+      "epoch": 5.729177364691384,
+      "grad_norm": 0.036805570125579834,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 55785
+    },
+    {
+      "epoch": 5.729485467803225,
+      "grad_norm": 0.0409085676074028,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 55788
+    },
+    {
+      "epoch": 5.729793570915066,
+      "grad_norm": 0.041818439960479736,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 55791
+    },
+    {
+      "epoch": 5.7301016740269075,
+      "grad_norm": 0.04998002573847771,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 55794
+    },
+    {
+      "epoch": 5.730409777138749,
+      "grad_norm": 0.13136835396289825,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 55797
+    },
+    {
+      "epoch": 5.730717880250591,
+      "grad_norm": 0.09088631719350815,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 55800
+    },
+    {
+      "epoch": 5.731025983362432,
+      "grad_norm": 0.04242292046546936,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 55803
+    },
+    {
+      "epoch": 5.7313340864742734,
+      "grad_norm": 0.06529027223587036,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 55806
+    },
+    {
+      "epoch": 5.731642189586115,
+      "grad_norm": 0.07886287569999695,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 55809
+    },
+    {
+      "epoch": 5.731950292697956,
+      "grad_norm": 0.05292735993862152,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 55812
+    },
+    {
+      "epoch": 5.732258395809797,
+      "grad_norm": 0.08382521569728851,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 55815
+    },
+    {
+      "epoch": 5.732566498921639,
+      "grad_norm": 0.16346225142478943,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 55818
+    },
+    {
+      "epoch": 5.732874602033481,
+      "grad_norm": 0.1087159663438797,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 55821
+    },
+    {
+      "epoch": 5.733182705145322,
+      "grad_norm": 0.08878868818283081,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 55824
+    },
+    {
+      "epoch": 5.733490808257163,
+      "grad_norm": 0.05677541717886925,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 55827
+    },
+    {
+      "epoch": 5.733798911369005,
+      "grad_norm": 0.047212038189172745,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 55830
+    },
+    {
+      "epoch": 5.734107014480847,
+      "grad_norm": 0.04530366137623787,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 55833
+    },
+    {
+      "epoch": 5.734415117592688,
+      "grad_norm": 0.03656543791294098,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 55836
+    },
+    {
+      "epoch": 5.734723220704529,
+      "grad_norm": 0.05110878124833107,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 55839
+    },
+    {
+      "epoch": 5.73503132381637,
+      "grad_norm": 0.046113792806863785,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 55842
+    },
+    {
+      "epoch": 5.735339426928212,
+      "grad_norm": 0.05249856784939766,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 55845
+    },
+    {
+      "epoch": 5.735647530040054,
+      "grad_norm": 0.10378940403461456,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 55848
+    },
+    {
+      "epoch": 5.735955633151895,
+      "grad_norm": 0.04738283157348633,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 55851
+    },
+    {
+      "epoch": 5.736263736263736,
+      "grad_norm": 0.042795851826667786,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 55854
+    },
+    {
+      "epoch": 5.736571839375578,
+      "grad_norm": 0.060027096420526505,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 55857
+    },
+    {
+      "epoch": 5.736879942487419,
+      "grad_norm": 0.060275256633758545,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 55860
+    },
+    {
+      "epoch": 5.737188045599261,
+      "grad_norm": 0.038434866815805435,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 55863
+    },
+    {
+      "epoch": 5.737496148711102,
+      "grad_norm": 0.06112179905176163,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 55866
+    },
+    {
+      "epoch": 5.7378042518229435,
+      "grad_norm": 0.057865045964717865,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 55869
+    },
+    {
+      "epoch": 5.738112354934785,
+      "grad_norm": 0.10420432686805725,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 55872
+    },
+    {
+      "epoch": 5.738420458046626,
+      "grad_norm": 0.10990259796380997,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 55875
+    },
+    {
+      "epoch": 5.738728561158467,
+      "grad_norm": 0.047898877412080765,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 55878
+    },
+    {
+      "epoch": 5.7390366642703095,
+      "grad_norm": 0.09271835535764694,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 55881
+    },
+    {
+      "epoch": 5.739344767382151,
+      "grad_norm": 0.046148382127285004,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 55884
+    },
+    {
+      "epoch": 5.739652870493992,
+      "grad_norm": 0.09337476640939713,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 55887
+    },
+    {
+      "epoch": 5.739960973605833,
+      "grad_norm": 0.0676882416009903,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 55890
+    },
+    {
+      "epoch": 5.7402690767176745,
+      "grad_norm": 0.07760494947433472,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 55893
+    },
+    {
+      "epoch": 5.740577179829517,
+      "grad_norm": 0.0753534585237503,
+      "learning_rate": 0.01,
+      "loss": 2.0221,
+      "step": 55896
+    },
+    {
+      "epoch": 5.740885282941358,
+      "grad_norm": 0.07207735627889633,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 55899
+    },
+    {
+      "epoch": 5.741193386053199,
+      "grad_norm": 0.10786162316799164,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 55902
+    },
+    {
+      "epoch": 5.74150148916504,
+      "grad_norm": 0.05810957029461861,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 55905
+    },
+    {
+      "epoch": 5.741809592276882,
+      "grad_norm": 0.043440163135528564,
+      "learning_rate": 0.01,
+      "loss": 2.025,
+      "step": 55908
+    },
+    {
+      "epoch": 5.742117695388723,
+      "grad_norm": 0.10091213136911392,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 55911
+    },
+    {
+      "epoch": 5.742425798500565,
+      "grad_norm": 0.05655033513903618,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 55914
+    },
+    {
+      "epoch": 5.742733901612406,
+      "grad_norm": 0.04754583537578583,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 55917
+    },
+    {
+      "epoch": 5.743042004724248,
+      "grad_norm": 0.04805220663547516,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 55920
+    },
+    {
+      "epoch": 5.743350107836089,
+      "grad_norm": 0.03961215168237686,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 55923
+    },
+    {
+      "epoch": 5.743658210947931,
+      "grad_norm": 0.0900992751121521,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 55926
+    },
+    {
+      "epoch": 5.743966314059772,
+      "grad_norm": 0.06700705736875534,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 55929
+    },
+    {
+      "epoch": 5.744274417171614,
+      "grad_norm": 0.06921757012605667,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 55932
+    },
+    {
+      "epoch": 5.744582520283455,
+      "grad_norm": 0.08982661366462708,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 55935
+    },
+    {
+      "epoch": 5.744890623395296,
+      "grad_norm": 0.059035494923591614,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 55938
+    },
+    {
+      "epoch": 5.745198726507137,
+      "grad_norm": 0.11368583887815475,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 55941
+    },
+    {
+      "epoch": 5.7455068296189795,
+      "grad_norm": 0.08658309280872345,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 55944
+    },
+    {
+      "epoch": 5.745814932730821,
+      "grad_norm": 0.06159745901823044,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 55947
+    },
+    {
+      "epoch": 5.746123035842662,
+      "grad_norm": 0.09689722955226898,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 55950
+    },
+    {
+      "epoch": 5.746431138954503,
+      "grad_norm": 0.04213634133338928,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 55953
+    },
+    {
+      "epoch": 5.746739242066345,
+      "grad_norm": 0.09022536873817444,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 55956
+    },
+    {
+      "epoch": 5.747047345178187,
+      "grad_norm": 0.047155823558568954,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 55959
+    },
+    {
+      "epoch": 5.747355448290028,
+      "grad_norm": 0.12048459053039551,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 55962
+    },
+    {
+      "epoch": 5.747663551401869,
+      "grad_norm": 0.14343701303005219,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 55965
+    },
+    {
+      "epoch": 5.7479716545137105,
+      "grad_norm": 0.10892770439386368,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 55968
+    },
+    {
+      "epoch": 5.748279757625552,
+      "grad_norm": 0.08472274988889694,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 55971
+    },
+    {
+      "epoch": 5.748587860737393,
+      "grad_norm": 0.06257399171590805,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 55974
+    },
+    {
+      "epoch": 5.748895963849235,
+      "grad_norm": 0.050180837512016296,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 55977
+    },
+    {
+      "epoch": 5.749204066961076,
+      "grad_norm": 0.04516899958252907,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 55980
+    },
+    {
+      "epoch": 5.749512170072918,
+      "grad_norm": 0.06078992411494255,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 55983
+    },
+    {
+      "epoch": 5.749820273184759,
+      "grad_norm": 0.060303375124931335,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 55986
+    },
+    {
+      "epoch": 5.7501283762966,
+      "grad_norm": 0.046474192291498184,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 55989
+    },
+    {
+      "epoch": 5.750436479408442,
+      "grad_norm": 0.042475223541259766,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 55992
+    },
+    {
+      "epoch": 5.750744582520284,
+      "grad_norm": 0.10338003933429718,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 55995
+    },
+    {
+      "epoch": 5.751052685632125,
+      "grad_norm": 0.07736215740442276,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 55998
+    },
+    {
+      "epoch": 5.751360788743966,
+      "grad_norm": 0.12122607231140137,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 56001
+    },
+    {
+      "epoch": 5.751668891855807,
+      "grad_norm": 0.12788768112659454,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 56004
+    },
+    {
+      "epoch": 5.751976994967649,
+      "grad_norm": 0.1270526796579361,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 56007
+    },
+    {
+      "epoch": 5.752285098079491,
+      "grad_norm": 0.07083453238010406,
+      "learning_rate": 0.01,
+      "loss": 2.0271,
+      "step": 56010
+    },
+    {
+      "epoch": 5.752593201191332,
+      "grad_norm": 0.0539688877761364,
+      "learning_rate": 0.01,
+      "loss": 2.0156,
+      "step": 56013
+    },
+    {
+      "epoch": 5.752901304303173,
+      "grad_norm": 0.05941673740744591,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 56016
+    },
+    {
+      "epoch": 5.753209407415015,
+      "grad_norm": 0.058209896087646484,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 56019
+    },
+    {
+      "epoch": 5.753517510526857,
+      "grad_norm": 0.046941760927438736,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 56022
+    },
+    {
+      "epoch": 5.753825613638698,
+      "grad_norm": 0.05372985079884529,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 56025
+    },
+    {
+      "epoch": 5.754133716750539,
+      "grad_norm": 0.04523862898349762,
+      "learning_rate": 0.01,
+      "loss": 2.0294,
+      "step": 56028
+    },
+    {
+      "epoch": 5.754441819862381,
+      "grad_norm": 0.03816952556371689,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 56031
+    },
+    {
+      "epoch": 5.754749922974222,
+      "grad_norm": 0.06410220265388489,
+      "learning_rate": 0.01,
+      "loss": 2.0222,
+      "step": 56034
+    },
+    {
+      "epoch": 5.755058026086063,
+      "grad_norm": 0.03941243886947632,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 56037
+    },
+    {
+      "epoch": 5.755366129197905,
+      "grad_norm": 0.054606303572654724,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 56040
+    },
+    {
+      "epoch": 5.7556742323097465,
+      "grad_norm": 0.12037497013807297,
+      "learning_rate": 0.01,
+      "loss": 2.013,
+      "step": 56043
+    },
+    {
+      "epoch": 5.755982335421588,
+      "grad_norm": 0.09681439399719238,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 56046
+    },
+    {
+      "epoch": 5.756290438533429,
+      "grad_norm": 0.09768577665090561,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 56049
+    },
+    {
+      "epoch": 5.75659854164527,
+      "grad_norm": 0.08952365070581436,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 56052
+    },
+    {
+      "epoch": 5.7569066447571124,
+      "grad_norm": 0.09528356790542603,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 56055
+    },
+    {
+      "epoch": 5.757214747868954,
+      "grad_norm": 0.05678083375096321,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 56058
+    },
+    {
+      "epoch": 5.757522850980795,
+      "grad_norm": 0.040836770087480545,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 56061
+    },
+    {
+      "epoch": 5.757830954092636,
+      "grad_norm": 0.053255595266819,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 56064
+    },
+    {
+      "epoch": 5.7581390572044775,
+      "grad_norm": 0.053145814687013626,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 56067
+    },
+    {
+      "epoch": 5.758447160316319,
+      "grad_norm": 0.04844813421368599,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 56070
+    },
+    {
+      "epoch": 5.758755263428161,
+      "grad_norm": 0.08748742938041687,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 56073
+    },
+    {
+      "epoch": 5.759063366540002,
+      "grad_norm": 0.03808058798313141,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 56076
+    },
+    {
+      "epoch": 5.759371469651843,
+      "grad_norm": 0.0475618913769722,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 56079
+    },
+    {
+      "epoch": 5.759679572763685,
+      "grad_norm": 0.04134085774421692,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 56082
+    },
+    {
+      "epoch": 5.759987675875526,
+      "grad_norm": 0.035581640899181366,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 56085
+    },
+    {
+      "epoch": 5.760295778987368,
+      "grad_norm": 0.1027594655752182,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 56088
+    },
+    {
+      "epoch": 5.760603882099209,
+      "grad_norm": 0.06380899995565414,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 56091
+    },
+    {
+      "epoch": 5.760911985211051,
+      "grad_norm": 0.04780165106058121,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 56094
+    },
+    {
+      "epoch": 5.761220088322892,
+      "grad_norm": 0.0501483753323555,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 56097
+    },
+    {
+      "epoch": 5.761528191434733,
+      "grad_norm": 0.06381183862686157,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 56100
+    },
+    {
+      "epoch": 5.761836294546574,
+      "grad_norm": 0.08505979180335999,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 56103
+    },
+    {
+      "epoch": 5.762144397658417,
+      "grad_norm": 0.09470894932746887,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 56106
+    },
+    {
+      "epoch": 5.762452500770258,
+      "grad_norm": 0.08021756261587143,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 56109
+    },
+    {
+      "epoch": 5.762760603882099,
+      "grad_norm": 0.05404616892337799,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 56112
+    },
+    {
+      "epoch": 5.76306870699394,
+      "grad_norm": 0.05812733992934227,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 56115
+    },
+    {
+      "epoch": 5.7633768101057825,
+      "grad_norm": 0.045452944934368134,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 56118
+    },
+    {
+      "epoch": 5.763684913217624,
+      "grad_norm": 0.06332234293222427,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 56121
+    },
+    {
+      "epoch": 5.763993016329465,
+      "grad_norm": 0.0571395680308342,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 56124
+    },
+    {
+      "epoch": 5.764301119441306,
+      "grad_norm": 0.057966724038124084,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 56127
+    },
+    {
+      "epoch": 5.764609222553148,
+      "grad_norm": 0.03700026124715805,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 56130
+    },
+    {
+      "epoch": 5.764917325664989,
+      "grad_norm": 0.06366675347089767,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 56133
+    },
+    {
+      "epoch": 5.765225428776831,
+      "grad_norm": 0.12095164507627487,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 56136
+    },
+    {
+      "epoch": 5.765533531888672,
+      "grad_norm": 0.05511580780148506,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 56139
+    },
+    {
+      "epoch": 5.7658416350005135,
+      "grad_norm": 0.06662758439779282,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 56142
+    },
+    {
+      "epoch": 5.766149738112355,
+      "grad_norm": 0.05345858260989189,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 56145
+    },
+    {
+      "epoch": 5.766457841224196,
+      "grad_norm": 0.05073786526918411,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 56148
+    },
+    {
+      "epoch": 5.766765944336038,
+      "grad_norm": 0.041545137763023376,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 56151
+    },
+    {
+      "epoch": 5.767074047447879,
+      "grad_norm": 0.058929163962602615,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 56154
+    },
+    {
+      "epoch": 5.767382150559721,
+      "grad_norm": 0.0575728714466095,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 56157
+    },
+    {
+      "epoch": 5.767690253671562,
+      "grad_norm": 0.05252945050597191,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 56160
+    },
+    {
+      "epoch": 5.767998356783403,
+      "grad_norm": 0.09407012909650803,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 56163
+    },
+    {
+      "epoch": 5.7683064598952445,
+      "grad_norm": 0.07041514664888382,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 56166
+    },
+    {
+      "epoch": 5.768614563007087,
+      "grad_norm": 0.0976720005273819,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 56169
+    },
+    {
+      "epoch": 5.768922666118928,
+      "grad_norm": 0.06705368310213089,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 56172
+    },
+    {
+      "epoch": 5.769230769230769,
+      "grad_norm": 0.044996488839387894,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 56175
+    },
+    {
+      "epoch": 5.76953887234261,
+      "grad_norm": 0.05325739085674286,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 56178
+    },
+    {
+      "epoch": 5.769846975454453,
+      "grad_norm": 0.06506424397230148,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 56181
+    },
+    {
+      "epoch": 5.770155078566294,
+      "grad_norm": 0.05719895660877228,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 56184
+    },
+    {
+      "epoch": 5.770463181678135,
+      "grad_norm": 0.057186469435691833,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 56187
+    },
+    {
+      "epoch": 5.770771284789976,
+      "grad_norm": 0.06783724576234818,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 56190
+    },
+    {
+      "epoch": 5.771079387901818,
+      "grad_norm": 0.10789873450994492,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 56193
+    },
+    {
+      "epoch": 5.771387491013659,
+      "grad_norm": 0.07688331604003906,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 56196
+    },
+    {
+      "epoch": 5.771695594125501,
+      "grad_norm": 0.056692346930503845,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 56199
+    },
+    {
+      "epoch": 5.772003697237342,
+      "grad_norm": 0.04593949392437935,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 56202
+    },
+    {
+      "epoch": 5.772311800349184,
+      "grad_norm": 0.050233110785484314,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 56205
+    },
+    {
+      "epoch": 5.772619903461025,
+      "grad_norm": 0.049515530467033386,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 56208
+    },
+    {
+      "epoch": 5.772928006572866,
+      "grad_norm": 0.10692589730024338,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 56211
+    },
+    {
+      "epoch": 5.773236109684708,
+      "grad_norm": 0.054260801523923874,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 56214
+    },
+    {
+      "epoch": 5.7735442127965495,
+      "grad_norm": 0.042156677693128586,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 56217
+    },
+    {
+      "epoch": 5.773852315908391,
+      "grad_norm": 0.06891094893217087,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 56220
+    },
+    {
+      "epoch": 5.774160419020232,
+      "grad_norm": 0.09000008553266525,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 56223
+    },
+    {
+      "epoch": 5.774468522132073,
+      "grad_norm": 0.051672134548425674,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 56226
+    },
+    {
+      "epoch": 5.7747766252439146,
+      "grad_norm": 0.04013100638985634,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 56229
+    },
+    {
+      "epoch": 5.775084728355757,
+      "grad_norm": 0.051951587200164795,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 56232
+    },
+    {
+      "epoch": 5.775392831467598,
+      "grad_norm": 0.038350678980350494,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 56235
+    },
+    {
+      "epoch": 5.775700934579439,
+      "grad_norm": 0.10026026517152786,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 56238
+    },
+    {
+      "epoch": 5.7760090376912805,
+      "grad_norm": 0.07415781170129776,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 56241
+    },
+    {
+      "epoch": 5.776317140803122,
+      "grad_norm": 0.04173508659005165,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 56244
+    },
+    {
+      "epoch": 5.776625243914964,
+      "grad_norm": 0.1886957734823227,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 56247
+    },
+    {
+      "epoch": 5.776933347026805,
+      "grad_norm": 0.13357476890087128,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 56250
+    },
+    {
+      "epoch": 5.777241450138646,
+      "grad_norm": 0.08402860164642334,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 56253
+    },
+    {
+      "epoch": 5.777549553250488,
+      "grad_norm": 0.08017169684171677,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 56256
+    },
+    {
+      "epoch": 5.777857656362329,
+      "grad_norm": 0.08612015098333359,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 56259
+    },
+    {
+      "epoch": 5.77816575947417,
+      "grad_norm": 0.06540670245885849,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 56262
+    },
+    {
+      "epoch": 5.778473862586012,
+      "grad_norm": 0.03119773231446743,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 56265
+    },
+    {
+      "epoch": 5.778781965697854,
+      "grad_norm": 0.12924209237098694,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 56268
+    },
+    {
+      "epoch": 5.779090068809695,
+      "grad_norm": 0.07823926955461502,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 56271
+    },
+    {
+      "epoch": 5.779398171921536,
+      "grad_norm": 0.05518370866775513,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 56274
+    },
+    {
+      "epoch": 5.779706275033378,
+      "grad_norm": 0.041604407131671906,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 56277
+    },
+    {
+      "epoch": 5.78001437814522,
+      "grad_norm": 0.05646267905831337,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 56280
+    },
+    {
+      "epoch": 5.780322481257061,
+      "grad_norm": 0.051092468202114105,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 56283
+    },
+    {
+      "epoch": 5.780630584368902,
+      "grad_norm": 0.05034413933753967,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 56286
+    },
+    {
+      "epoch": 5.780938687480743,
+      "grad_norm": 0.03945804387331009,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 56289
+    },
+    {
+      "epoch": 5.781246790592585,
+      "grad_norm": 0.046318311244249344,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 56292
+    },
+    {
+      "epoch": 5.781554893704427,
+      "grad_norm": 0.0614444762468338,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 56295
+    },
+    {
+      "epoch": 5.781862996816268,
+      "grad_norm": 0.06555134057998657,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 56298
+    },
+    {
+      "epoch": 5.782171099928109,
+      "grad_norm": 0.08905301988124847,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 56301
+    },
+    {
+      "epoch": 5.7824792030399506,
+      "grad_norm": 0.06253067404031754,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 56304
+    },
+    {
+      "epoch": 5.782787306151792,
+      "grad_norm": 0.05287610739469528,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 56307
+    },
+    {
+      "epoch": 5.783095409263634,
+      "grad_norm": 0.11111927032470703,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 56310
+    },
+    {
+      "epoch": 5.783403512375475,
+      "grad_norm": 0.05433667451143265,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 56313
+    },
+    {
+      "epoch": 5.7837116154873165,
+      "grad_norm": 0.03661932423710823,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 56316
+    },
+    {
+      "epoch": 5.784019718599158,
+      "grad_norm": 0.06176665052771568,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 56319
+    },
+    {
+      "epoch": 5.784327821710999,
+      "grad_norm": 0.05721355602145195,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 56322
+    },
+    {
+      "epoch": 5.78463592482284,
+      "grad_norm": 0.043899018317461014,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 56325
+    },
+    {
+      "epoch": 5.784944027934682,
+      "grad_norm": 0.17083556950092316,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 56328
+    },
+    {
+      "epoch": 5.785252131046524,
+      "grad_norm": 0.12060192972421646,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 56331
+    },
+    {
+      "epoch": 5.785560234158365,
+      "grad_norm": 0.1456676870584488,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 56334
+    },
+    {
+      "epoch": 5.785868337270206,
+      "grad_norm": 0.08020723611116409,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 56337
+    },
+    {
+      "epoch": 5.7861764403820475,
+      "grad_norm": 0.058028947561979294,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 56340
+    },
+    {
+      "epoch": 5.78648454349389,
+      "grad_norm": 0.031094344332814217,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 56343
+    },
+    {
+      "epoch": 5.786792646605731,
+      "grad_norm": 0.03784453868865967,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 56346
+    },
+    {
+      "epoch": 5.787100749717572,
+      "grad_norm": 0.03983352333307266,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 56349
+    },
+    {
+      "epoch": 5.787408852829413,
+      "grad_norm": 0.048995792865753174,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 56352
+    },
+    {
+      "epoch": 5.787716955941255,
+      "grad_norm": 0.040493566542863846,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 56355
+    },
+    {
+      "epoch": 5.788025059053096,
+      "grad_norm": 0.04488976299762726,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 56358
+    },
+    {
+      "epoch": 5.788333162164938,
+      "grad_norm": 0.1611676812171936,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 56361
+    },
+    {
+      "epoch": 5.788641265276779,
+      "grad_norm": 0.079713374376297,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 56364
+    },
+    {
+      "epoch": 5.788949368388621,
+      "grad_norm": 0.06308376044034958,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 56367
+    },
+    {
+      "epoch": 5.789257471500462,
+      "grad_norm": 0.049073606729507446,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 56370
+    },
+    {
+      "epoch": 5.789565574612304,
+      "grad_norm": 0.06402777135372162,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 56373
+    },
+    {
+      "epoch": 5.789873677724145,
+      "grad_norm": 0.05663265287876129,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 56376
+    },
+    {
+      "epoch": 5.790181780835987,
+      "grad_norm": 0.05829964950680733,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 56379
+    },
+    {
+      "epoch": 5.790489883947828,
+      "grad_norm": 0.05013945326209068,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 56382
+    },
+    {
+      "epoch": 5.790797987059669,
+      "grad_norm": 0.04025455191731453,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 56385
+    },
+    {
+      "epoch": 5.79110609017151,
+      "grad_norm": 0.05371120572090149,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 56388
+    },
+    {
+      "epoch": 5.7914141932833525,
+      "grad_norm": 0.03760494664311409,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 56391
+    },
+    {
+      "epoch": 5.791722296395194,
+      "grad_norm": 0.05783433839678764,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 56394
+    },
+    {
+      "epoch": 5.792030399507035,
+      "grad_norm": 0.04827668145298958,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 56397
+    },
+    {
+      "epoch": 5.792338502618876,
+      "grad_norm": 0.055127400904893875,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 56400
+    },
+    {
+      "epoch": 5.7926466057307175,
+      "grad_norm": 0.07991458475589752,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 56403
+    },
+    {
+      "epoch": 5.79295470884256,
+      "grad_norm": 0.06287828832864761,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 56406
+    },
+    {
+      "epoch": 5.793262811954401,
+      "grad_norm": 0.07629155367612839,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 56409
+    },
+    {
+      "epoch": 5.793570915066242,
+      "grad_norm": 0.10665415227413177,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 56412
+    },
+    {
+      "epoch": 5.7938790181780835,
+      "grad_norm": 0.048113856464624405,
+      "learning_rate": 0.01,
+      "loss": 2.0162,
+      "step": 56415
+    },
+    {
+      "epoch": 5.794187121289925,
+      "grad_norm": 0.12309219688177109,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 56418
+    },
+    {
+      "epoch": 5.794495224401766,
+      "grad_norm": 0.08479398488998413,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 56421
+    },
+    {
+      "epoch": 5.794803327513608,
+      "grad_norm": 0.06182289868593216,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 56424
+    },
+    {
+      "epoch": 5.795111430625449,
+      "grad_norm": 0.042767733335494995,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 56427
+    },
+    {
+      "epoch": 5.795419533737291,
+      "grad_norm": 0.07279690355062485,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 56430
+    },
+    {
+      "epoch": 5.795727636849132,
+      "grad_norm": 0.05660317465662956,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 56433
+    },
+    {
+      "epoch": 5.796035739960974,
+      "grad_norm": 0.048105914145708084,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 56436
+    },
+    {
+      "epoch": 5.796343843072815,
+      "grad_norm": 0.09529261291027069,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 56439
+    },
+    {
+      "epoch": 5.796651946184657,
+      "grad_norm": 0.09696771949529648,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 56442
+    },
+    {
+      "epoch": 5.796960049296498,
+      "grad_norm": 0.10225138068199158,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 56445
+    },
+    {
+      "epoch": 5.797268152408339,
+      "grad_norm": 0.09067768603563309,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 56448
+    },
+    {
+      "epoch": 5.79757625552018,
+      "grad_norm": 0.03427765890955925,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 56451
+    },
+    {
+      "epoch": 5.797884358632023,
+      "grad_norm": 0.03189976140856743,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 56454
+    },
+    {
+      "epoch": 5.798192461743864,
+      "grad_norm": 0.04521431401371956,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 56457
+    },
+    {
+      "epoch": 5.798500564855705,
+      "grad_norm": 0.05165833979845047,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 56460
+    },
+    {
+      "epoch": 5.798808667967546,
+      "grad_norm": 0.051866378635168076,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 56463
+    },
+    {
+      "epoch": 5.799116771079388,
+      "grad_norm": 0.03956654295325279,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 56466
+    },
+    {
+      "epoch": 5.79942487419123,
+      "grad_norm": 0.04093864560127258,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 56469
+    },
+    {
+      "epoch": 5.799732977303071,
+      "grad_norm": 0.05415187031030655,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 56472
+    },
+    {
+      "epoch": 5.800041080414912,
+      "grad_norm": 0.057444531470537186,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 56475
+    },
+    {
+      "epoch": 5.8003491835267535,
+      "grad_norm": 0.08681999891996384,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 56478
+    },
+    {
+      "epoch": 5.800657286638595,
+      "grad_norm": 0.08110462129116058,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 56481
+    },
+    {
+      "epoch": 5.800965389750436,
+      "grad_norm": 0.04393799602985382,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 56484
+    },
+    {
+      "epoch": 5.801273492862278,
+      "grad_norm": 0.05482124909758568,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 56487
+    },
+    {
+      "epoch": 5.8015815959741195,
+      "grad_norm": 0.11982918530702591,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 56490
+    },
+    {
+      "epoch": 5.801889699085961,
+      "grad_norm": 0.10209926217794418,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 56493
+    },
+    {
+      "epoch": 5.802197802197802,
+      "grad_norm": 0.0397653728723526,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 56496
+    },
+    {
+      "epoch": 5.802505905309643,
+      "grad_norm": 0.09022632986307144,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 56499
+    },
+    {
+      "epoch": 5.802814008421485,
+      "grad_norm": 0.07979937642812729,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 56502
+    },
+    {
+      "epoch": 5.803122111533327,
+      "grad_norm": 0.05430705472826958,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 56505
+    },
+    {
+      "epoch": 5.803430214645168,
+      "grad_norm": 0.05586136132478714,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 56508
+    },
+    {
+      "epoch": 5.803738317757009,
+      "grad_norm": 0.05488377809524536,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 56511
+    },
+    {
+      "epoch": 5.8040464208688505,
+      "grad_norm": 0.04581581428647041,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 56514
+    },
+    {
+      "epoch": 5.804354523980692,
+      "grad_norm": 0.11193082481622696,
+      "learning_rate": 0.01,
+      "loss": 2.0275,
+      "step": 56517
+    },
+    {
+      "epoch": 5.804662627092534,
+      "grad_norm": 0.0932486429810524,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 56520
+    },
+    {
+      "epoch": 5.804970730204375,
+      "grad_norm": 0.079062819480896,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 56523
+    },
+    {
+      "epoch": 5.805278833316216,
+      "grad_norm": 0.04556051269173622,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 56526
+    },
+    {
+      "epoch": 5.805586936428058,
+      "grad_norm": 0.04513680562376976,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 56529
+    },
+    {
+      "epoch": 5.8058950395399,
+      "grad_norm": 0.06065935268998146,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 56532
+    },
+    {
+      "epoch": 5.806203142651741,
+      "grad_norm": 0.03341379761695862,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 56535
+    },
+    {
+      "epoch": 5.806511245763582,
+      "grad_norm": 0.13032658398151398,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 56538
+    },
+    {
+      "epoch": 5.806819348875424,
+      "grad_norm": 0.049995262175798416,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 56541
+    },
+    {
+      "epoch": 5.807127451987265,
+      "grad_norm": 0.10843599587678909,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 56544
+    },
+    {
+      "epoch": 5.807435555099106,
+      "grad_norm": 0.12207616865634918,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 56547
+    },
+    {
+      "epoch": 5.807743658210948,
+      "grad_norm": 0.18218643963336945,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 56550
+    },
+    {
+      "epoch": 5.8080517613227896,
+      "grad_norm": 0.1526477336883545,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 56553
+    },
+    {
+      "epoch": 5.808359864434631,
+      "grad_norm": 0.08327028155326843,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 56556
+    },
+    {
+      "epoch": 5.808667967546472,
+      "grad_norm": 0.05534267798066139,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 56559
+    },
+    {
+      "epoch": 5.808976070658313,
+      "grad_norm": 0.04754161089658737,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 56562
+    },
+    {
+      "epoch": 5.8092841737701555,
+      "grad_norm": 0.039216890931129456,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 56565
+    },
+    {
+      "epoch": 5.809592276881997,
+      "grad_norm": 0.05179159343242645,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 56568
+    },
+    {
+      "epoch": 5.809900379993838,
+      "grad_norm": 0.03787977248430252,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 56571
+    },
+    {
+      "epoch": 5.810208483105679,
+      "grad_norm": 0.06589755415916443,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 56574
+    },
+    {
+      "epoch": 5.8105165862175205,
+      "grad_norm": 0.04956568777561188,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 56577
+    },
+    {
+      "epoch": 5.810824689329362,
+      "grad_norm": 0.039915554225444794,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 56580
+    },
+    {
+      "epoch": 5.811132792441204,
+      "grad_norm": 0.10185667872428894,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 56583
+    },
+    {
+      "epoch": 5.811440895553045,
+      "grad_norm": 0.048244234174489975,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 56586
+    },
+    {
+      "epoch": 5.8117489986648865,
+      "grad_norm": 0.07163000851869583,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 56589
+    },
+    {
+      "epoch": 5.812057101776728,
+      "grad_norm": 0.04289057105779648,
+      "learning_rate": 0.01,
+      "loss": 2.0326,
+      "step": 56592
+    },
+    {
+      "epoch": 5.812365204888569,
+      "grad_norm": 0.06520239263772964,
+      "learning_rate": 0.01,
+      "loss": 2.0223,
+      "step": 56595
+    },
+    {
+      "epoch": 5.812673308000411,
+      "grad_norm": 0.038391079753637314,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 56598
+    },
+    {
+      "epoch": 5.812981411112252,
+      "grad_norm": 0.039008256047964096,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 56601
+    },
+    {
+      "epoch": 5.813289514224094,
+      "grad_norm": 0.1321951448917389,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 56604
+    },
+    {
+      "epoch": 5.813597617335935,
+      "grad_norm": 0.058711450546979904,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 56607
+    },
+    {
+      "epoch": 5.813905720447776,
+      "grad_norm": 0.10861786454916,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 56610
+    },
+    {
+      "epoch": 5.8142138235596175,
+      "grad_norm": 0.09143928438425064,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 56613
+    },
+    {
+      "epoch": 5.81452192667146,
+      "grad_norm": 0.051334600895643234,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 56616
+    },
+    {
+      "epoch": 5.814830029783301,
+      "grad_norm": 0.0675191655755043,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 56619
+    },
+    {
+      "epoch": 5.815138132895142,
+      "grad_norm": 0.038013167679309845,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 56622
+    },
+    {
+      "epoch": 5.815446236006983,
+      "grad_norm": 0.043695393949747086,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 56625
+    },
+    {
+      "epoch": 5.8157543391188256,
+      "grad_norm": 0.034250471740961075,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 56628
+    },
+    {
+      "epoch": 5.816062442230667,
+      "grad_norm": 0.11236503720283508,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 56631
+    },
+    {
+      "epoch": 5.816370545342508,
+      "grad_norm": 0.053564272820949554,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 56634
+    },
+    {
+      "epoch": 5.816678648454349,
+      "grad_norm": 0.08416426926851273,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 56637
+    },
+    {
+      "epoch": 5.816986751566191,
+      "grad_norm": 0.047465983778238297,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 56640
+    },
+    {
+      "epoch": 5.817294854678032,
+      "grad_norm": 0.12391756474971771,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 56643
+    },
+    {
+      "epoch": 5.817602957789874,
+      "grad_norm": 0.12705251574516296,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 56646
+    },
+    {
+      "epoch": 5.817911060901715,
+      "grad_norm": 0.09919688105583191,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 56649
+    },
+    {
+      "epoch": 5.8182191640135565,
+      "grad_norm": 0.08405104279518127,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 56652
+    },
+    {
+      "epoch": 5.818527267125398,
+      "grad_norm": 0.04591039568185806,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 56655
+    },
+    {
+      "epoch": 5.818835370237239,
+      "grad_norm": 0.04161922633647919,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 56658
+    },
+    {
+      "epoch": 5.819143473349081,
+      "grad_norm": 0.05382911115884781,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 56661
+    },
+    {
+      "epoch": 5.8194515764609225,
+      "grad_norm": 0.05096543952822685,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 56664
+    },
+    {
+      "epoch": 5.819759679572764,
+      "grad_norm": 0.03849412500858307,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 56667
+    },
+    {
+      "epoch": 5.820067782684605,
+      "grad_norm": 0.04661410301923752,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 56670
+    },
+    {
+      "epoch": 5.820375885796446,
+      "grad_norm": 0.11310985684394836,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 56673
+    },
+    {
+      "epoch": 5.8206839889082875,
+      "grad_norm": 0.04283592104911804,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 56676
+    },
+    {
+      "epoch": 5.82099209202013,
+      "grad_norm": 0.09948618710041046,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 56679
+    },
+    {
+      "epoch": 5.821300195131971,
+      "grad_norm": 0.07223562151193619,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 56682
+    },
+    {
+      "epoch": 5.821608298243812,
+      "grad_norm": 0.067604660987854,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 56685
+    },
+    {
+      "epoch": 5.8219164013556535,
+      "grad_norm": 0.03564516827464104,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 56688
+    },
+    {
+      "epoch": 5.822224504467496,
+      "grad_norm": 0.046253006905317307,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 56691
+    },
+    {
+      "epoch": 5.822532607579337,
+      "grad_norm": 0.05620354413986206,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 56694
+    },
+    {
+      "epoch": 5.822840710691178,
+      "grad_norm": 0.110667385160923,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 56697
+    },
+    {
+      "epoch": 5.823148813803019,
+      "grad_norm": 0.05062146484851837,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 56700
+    },
+    {
+      "epoch": 5.823456916914861,
+      "grad_norm": 0.10704591870307922,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 56703
+    },
+    {
+      "epoch": 5.823765020026702,
+      "grad_norm": 0.05372390151023865,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 56706
+    },
+    {
+      "epoch": 5.824073123138544,
+      "grad_norm": 0.05827971547842026,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 56709
+    },
+    {
+      "epoch": 5.824381226250385,
+      "grad_norm": 0.04202426224946976,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 56712
+    },
+    {
+      "epoch": 5.824689329362227,
+      "grad_norm": 0.08493483811616898,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 56715
+    },
+    {
+      "epoch": 5.824997432474068,
+      "grad_norm": 0.043885089457035065,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 56718
+    },
+    {
+      "epoch": 5.825305535585909,
+      "grad_norm": 0.04681296646595001,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 56721
+    },
+    {
+      "epoch": 5.825613638697751,
+      "grad_norm": 0.04231657460331917,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 56724
+    },
+    {
+      "epoch": 5.8259217418095925,
+      "grad_norm": 0.04888215288519859,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 56727
+    },
+    {
+      "epoch": 5.826229844921434,
+      "grad_norm": 0.0635887086391449,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 56730
+    },
+    {
+      "epoch": 5.826537948033275,
+      "grad_norm": 0.056875888258218765,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 56733
+    },
+    {
+      "epoch": 5.826846051145116,
+      "grad_norm": 0.05551552772521973,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 56736
+    },
+    {
+      "epoch": 5.827154154256958,
+      "grad_norm": 0.10465902090072632,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 56739
+    },
+    {
+      "epoch": 5.8274622573688,
+      "grad_norm": 0.08767648041248322,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 56742
+    },
+    {
+      "epoch": 5.827770360480641,
+      "grad_norm": 0.04951823502779007,
+      "learning_rate": 0.01,
+      "loss": 2.0045,
+      "step": 56745
+    },
+    {
+      "epoch": 5.828078463592482,
+      "grad_norm": 0.04926539584994316,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 56748
+    },
+    {
+      "epoch": 5.8283865667043235,
+      "grad_norm": 0.05217811465263367,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 56751
+    },
+    {
+      "epoch": 5.828694669816165,
+      "grad_norm": 0.0460977777838707,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 56754
+    },
+    {
+      "epoch": 5.829002772928007,
+      "grad_norm": 0.046701837331056595,
+      "learning_rate": 0.01,
+      "loss": 2.0079,
+      "step": 56757
+    },
+    {
+      "epoch": 5.829310876039848,
+      "grad_norm": 0.07417173683643341,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 56760
+    },
+    {
+      "epoch": 5.8296189791516895,
+      "grad_norm": 0.09357844293117523,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 56763
+    },
+    {
+      "epoch": 5.829927082263531,
+      "grad_norm": 0.05171707645058632,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 56766
+    },
+    {
+      "epoch": 5.830235185375372,
+      "grad_norm": 0.08236930519342422,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 56769
+    },
+    {
+      "epoch": 5.830543288487213,
+      "grad_norm": 0.041667647659778595,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 56772
+    },
+    {
+      "epoch": 5.830851391599055,
+      "grad_norm": 0.055335305631160736,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 56775
+    },
+    {
+      "epoch": 5.831159494710897,
+      "grad_norm": 0.03667658939957619,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 56778
+    },
+    {
+      "epoch": 5.831467597822738,
+      "grad_norm": 0.05994969233870506,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 56781
+    },
+    {
+      "epoch": 5.831775700934579,
+      "grad_norm": 0.1395941525697708,
+      "learning_rate": 0.01,
+      "loss": 2.0184,
+      "step": 56784
+    },
+    {
+      "epoch": 5.832083804046421,
+      "grad_norm": 0.0712098479270935,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 56787
+    },
+    {
+      "epoch": 5.832391907158263,
+      "grad_norm": 0.08360549062490463,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 56790
+    },
+    {
+      "epoch": 5.832700010270104,
+      "grad_norm": 0.07638189941644669,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 56793
+    },
+    {
+      "epoch": 5.833008113381945,
+      "grad_norm": 0.07885719835758209,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 56796
+    },
+    {
+      "epoch": 5.833316216493786,
+      "grad_norm": 0.055790308862924576,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 56799
+    },
+    {
+      "epoch": 5.833624319605628,
+      "grad_norm": 0.03926543518900871,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 56802
+    },
+    {
+      "epoch": 5.83393242271747,
+      "grad_norm": 0.10085275024175644,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 56805
+    },
+    {
+      "epoch": 5.834240525829311,
+      "grad_norm": 0.07995092123746872,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 56808
+    },
+    {
+      "epoch": 5.834548628941152,
+      "grad_norm": 0.09439895302057266,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 56811
+    },
+    {
+      "epoch": 5.834856732052994,
+      "grad_norm": 0.04961158335208893,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 56814
+    },
+    {
+      "epoch": 5.835164835164835,
+      "grad_norm": 0.052938688546419144,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 56817
+    },
+    {
+      "epoch": 5.835472938276677,
+      "grad_norm": 0.04960361868143082,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 56820
+    },
+    {
+      "epoch": 5.835781041388518,
+      "grad_norm": 0.08586288243532181,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 56823
+    },
+    {
+      "epoch": 5.8360891445003595,
+      "grad_norm": 0.07194332033395767,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 56826
+    },
+    {
+      "epoch": 5.836397247612201,
+      "grad_norm": 0.062126416712999344,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 56829
+    },
+    {
+      "epoch": 5.836705350724042,
+      "grad_norm": 0.06656119972467422,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 56832
+    },
+    {
+      "epoch": 5.837013453835883,
+      "grad_norm": 0.0884518176317215,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 56835
+    },
+    {
+      "epoch": 5.8373215569477255,
+      "grad_norm": 0.04115528613328934,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 56838
+    },
+    {
+      "epoch": 5.837629660059567,
+      "grad_norm": 0.05185645818710327,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 56841
+    },
+    {
+      "epoch": 5.837937763171408,
+      "grad_norm": 0.11587214469909668,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 56844
+    },
+    {
+      "epoch": 5.838245866283249,
+      "grad_norm": 0.13516643643379211,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 56847
+    },
+    {
+      "epoch": 5.8385539693950905,
+      "grad_norm": 0.0748513862490654,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 56850
+    },
+    {
+      "epoch": 5.838862072506933,
+      "grad_norm": 0.12254352867603302,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 56853
+    },
+    {
+      "epoch": 5.839170175618774,
+      "grad_norm": 0.07876627147197723,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 56856
+    },
+    {
+      "epoch": 5.839478278730615,
+      "grad_norm": 0.07364428788423538,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 56859
+    },
+    {
+      "epoch": 5.8397863818424565,
+      "grad_norm": 0.0350789949297905,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 56862
+    },
+    {
+      "epoch": 5.840094484954298,
+      "grad_norm": 0.046593520790338516,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 56865
+    },
+    {
+      "epoch": 5.840402588066139,
+      "grad_norm": 0.05111907422542572,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 56868
+    },
+    {
+      "epoch": 5.840710691177981,
+      "grad_norm": 0.046573907136917114,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 56871
+    },
+    {
+      "epoch": 5.841018794289822,
+      "grad_norm": 0.05139172077178955,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 56874
+    },
+    {
+      "epoch": 5.841326897401664,
+      "grad_norm": 0.06465848535299301,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 56877
+    },
+    {
+      "epoch": 5.841635000513505,
+      "grad_norm": 0.046135857701301575,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 56880
+    },
+    {
+      "epoch": 5.841943103625347,
+      "grad_norm": 0.039065003395080566,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 56883
+    },
+    {
+      "epoch": 5.842251206737188,
+      "grad_norm": 0.05646162107586861,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 56886
+    },
+    {
+      "epoch": 5.84255930984903,
+      "grad_norm": 0.1448718160390854,
+      "learning_rate": 0.01,
+      "loss": 2.0192,
+      "step": 56889
+    },
+    {
+      "epoch": 5.842867412960871,
+      "grad_norm": 0.12302438169717789,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 56892
+    },
+    {
+      "epoch": 5.843175516072712,
+      "grad_norm": 0.0701448991894722,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 56895
+    },
+    {
+      "epoch": 5.843483619184553,
+      "grad_norm": 0.08622895181179047,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 56898
+    },
+    {
+      "epoch": 5.8437917222963955,
+      "grad_norm": 0.04280500486493111,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 56901
+    },
+    {
+      "epoch": 5.844099825408237,
+      "grad_norm": 0.04177187755703926,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 56904
+    },
+    {
+      "epoch": 5.844407928520078,
+      "grad_norm": 0.04321052134037018,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 56907
+    },
+    {
+      "epoch": 5.844716031631919,
+      "grad_norm": 0.10391934216022491,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 56910
+    },
+    {
+      "epoch": 5.845024134743761,
+      "grad_norm": 0.07634170353412628,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 56913
+    },
+    {
+      "epoch": 5.845332237855603,
+      "grad_norm": 0.05854891613125801,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 56916
+    },
+    {
+      "epoch": 5.845640340967444,
+      "grad_norm": 0.08058078587055206,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 56919
+    },
+    {
+      "epoch": 5.845948444079285,
+      "grad_norm": 0.10716883093118668,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 56922
+    },
+    {
+      "epoch": 5.8462565471911265,
+      "grad_norm": 0.1634785681962967,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 56925
+    },
+    {
+      "epoch": 5.846564650302968,
+      "grad_norm": 0.08859071135520935,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 56928
+    },
+    {
+      "epoch": 5.846872753414809,
+      "grad_norm": 0.05146167427301407,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 56931
+    },
+    {
+      "epoch": 5.847180856526651,
+      "grad_norm": 0.057162970304489136,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 56934
+    },
+    {
+      "epoch": 5.8474889596384925,
+      "grad_norm": 0.042328424751758575,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 56937
+    },
+    {
+      "epoch": 5.847797062750334,
+      "grad_norm": 0.05657609552145004,
+      "learning_rate": 0.01,
+      "loss": 2.0205,
+      "step": 56940
+    },
+    {
+      "epoch": 5.848105165862175,
+      "grad_norm": 0.11977431923151016,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 56943
+    },
+    {
+      "epoch": 5.848413268974016,
+      "grad_norm": 0.13686580955982208,
+      "learning_rate": 0.01,
+      "loss": 2.0215,
+      "step": 56946
+    },
+    {
+      "epoch": 5.848721372085858,
+      "grad_norm": 0.06063876673579216,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 56949
+    },
+    {
+      "epoch": 5.8490294751977,
+      "grad_norm": 0.05211206525564194,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 56952
+    },
+    {
+      "epoch": 5.849337578309541,
+      "grad_norm": 0.03639880567789078,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 56955
+    },
+    {
+      "epoch": 5.849645681421382,
+      "grad_norm": 0.07671289145946503,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 56958
+    },
+    {
+      "epoch": 5.8499537845332235,
+      "grad_norm": 0.03963657468557358,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 56961
+    },
+    {
+      "epoch": 5.850261887645065,
+      "grad_norm": 0.040136873722076416,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 56964
+    },
+    {
+      "epoch": 5.850569990756907,
+      "grad_norm": 0.06622539460659027,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 56967
+    },
+    {
+      "epoch": 5.850878093868748,
+      "grad_norm": 0.04892749339342117,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 56970
+    },
+    {
+      "epoch": 5.851186196980589,
+      "grad_norm": 0.10630463808774948,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 56973
+    },
+    {
+      "epoch": 5.851494300092431,
+      "grad_norm": 0.08189810067415237,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 56976
+    },
+    {
+      "epoch": 5.851802403204273,
+      "grad_norm": 0.04350016266107559,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 56979
+    },
+    {
+      "epoch": 5.852110506316114,
+      "grad_norm": 0.04086649417877197,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 56982
+    },
+    {
+      "epoch": 5.852418609427955,
+      "grad_norm": 0.0460704006254673,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 56985
+    },
+    {
+      "epoch": 5.852726712539797,
+      "grad_norm": 0.04429091140627861,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 56988
+    },
+    {
+      "epoch": 5.853034815651638,
+      "grad_norm": 0.08944038301706314,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 56991
+    },
+    {
+      "epoch": 5.853342918763479,
+      "grad_norm": 0.039852675050497055,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 56994
+    },
+    {
+      "epoch": 5.853651021875321,
+      "grad_norm": 0.051225848495960236,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 56997
+    },
+    {
+      "epoch": 5.8539591249871625,
+      "grad_norm": 0.07715211063623428,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 57000
+    },
+    {
+      "epoch": 5.854267228099004,
+      "grad_norm": 0.05754586309194565,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 57003
+    },
+    {
+      "epoch": 5.854575331210845,
+      "grad_norm": 0.0871339812874794,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 57006
+    },
+    {
+      "epoch": 5.854883434322686,
+      "grad_norm": 0.03794780746102333,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 57009
+    },
+    {
+      "epoch": 5.8551915374345285,
+      "grad_norm": 0.039838340133428574,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 57012
+    },
+    {
+      "epoch": 5.85549964054637,
+      "grad_norm": 0.03378939628601074,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 57015
+    },
+    {
+      "epoch": 5.855807743658211,
+      "grad_norm": 0.033859383314847946,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 57018
+    },
+    {
+      "epoch": 5.856115846770052,
+      "grad_norm": 0.08148138970136642,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 57021
+    },
+    {
+      "epoch": 5.8564239498818935,
+      "grad_norm": 0.11949529498815536,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 57024
+    },
+    {
+      "epoch": 5.856732052993735,
+      "grad_norm": 0.06349018216133118,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 57027
+    },
+    {
+      "epoch": 5.857040156105577,
+      "grad_norm": 0.03922737017273903,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 57030
+    },
+    {
+      "epoch": 5.857348259217418,
+      "grad_norm": 0.04704468697309494,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 57033
+    },
+    {
+      "epoch": 5.8576563623292595,
+      "grad_norm": 0.06949195265769958,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 57036
+    },
+    {
+      "epoch": 5.857964465441101,
+      "grad_norm": 0.08585426956415176,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 57039
+    },
+    {
+      "epoch": 5.858272568552943,
+      "grad_norm": 0.09531115740537643,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 57042
+    },
+    {
+      "epoch": 5.858580671664784,
+      "grad_norm": 0.07545875012874603,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 57045
+    },
+    {
+      "epoch": 5.858888774776625,
+      "grad_norm": 0.11838539689779282,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 57048
+    },
+    {
+      "epoch": 5.859196877888467,
+      "grad_norm": 0.1177034080028534,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 57051
+    },
+    {
+      "epoch": 5.859504981000308,
+      "grad_norm": 0.08327648043632507,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 57054
+    },
+    {
+      "epoch": 5.859813084112149,
+      "grad_norm": 0.05028563365340233,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 57057
+    },
+    {
+      "epoch": 5.860121187223991,
+      "grad_norm": 0.03819550573825836,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 57060
+    },
+    {
+      "epoch": 5.860429290335833,
+      "grad_norm": 0.06301325559616089,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 57063
+    },
+    {
+      "epoch": 5.860737393447674,
+      "grad_norm": 0.09482091665267944,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 57066
+    },
+    {
+      "epoch": 5.861045496559515,
+      "grad_norm": 0.07003244757652283,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 57069
+    },
+    {
+      "epoch": 5.861353599671356,
+      "grad_norm": 0.04322872310876846,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 57072
+    },
+    {
+      "epoch": 5.8616617027831985,
+      "grad_norm": 0.04223179072141647,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 57075
+    },
+    {
+      "epoch": 5.86196980589504,
+      "grad_norm": 0.09732475876808167,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 57078
+    },
+    {
+      "epoch": 5.862277909006881,
+      "grad_norm": 0.08102111518383026,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 57081
+    },
+    {
+      "epoch": 5.862586012118722,
+      "grad_norm": 0.06564348191022873,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 57084
+    },
+    {
+      "epoch": 5.862894115230564,
+      "grad_norm": 0.11364365369081497,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 57087
+    },
+    {
+      "epoch": 5.863202218342405,
+      "grad_norm": 0.05005546286702156,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 57090
+    },
+    {
+      "epoch": 5.863510321454247,
+      "grad_norm": 0.10924969613552094,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 57093
+    },
+    {
+      "epoch": 5.863818424566088,
+      "grad_norm": 0.06546615809202194,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 57096
+    },
+    {
+      "epoch": 5.8641265276779295,
+      "grad_norm": 0.04817098379135132,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 57099
+    },
+    {
+      "epoch": 5.864434630789771,
+      "grad_norm": 0.038462646305561066,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 57102
+    },
+    {
+      "epoch": 5.864742733901612,
+      "grad_norm": 0.03544708341360092,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 57105
+    },
+    {
+      "epoch": 5.865050837013454,
+      "grad_norm": 0.032861895859241486,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 57108
+    },
+    {
+      "epoch": 5.8653589401252955,
+      "grad_norm": 0.13162890076637268,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 57111
+    },
+    {
+      "epoch": 5.865667043237137,
+      "grad_norm": 0.13489802181720734,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 57114
+    },
+    {
+      "epoch": 5.865975146348978,
+      "grad_norm": 0.08409550786018372,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 57117
+    },
+    {
+      "epoch": 5.866283249460819,
+      "grad_norm": 0.04150281846523285,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 57120
+    },
+    {
+      "epoch": 5.8665913525726605,
+      "grad_norm": 0.04543888941407204,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 57123
+    },
+    {
+      "epoch": 5.866899455684503,
+      "grad_norm": 0.0386471152305603,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 57126
+    },
+    {
+      "epoch": 5.867207558796344,
+      "grad_norm": 0.029136119410395622,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 57129
+    },
+    {
+      "epoch": 5.867515661908185,
+      "grad_norm": 0.031866688281297684,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 57132
+    },
+    {
+      "epoch": 5.8678237650200264,
+      "grad_norm": 0.05178746208548546,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 57135
+    },
+    {
+      "epoch": 5.868131868131869,
+      "grad_norm": 0.10231690853834152,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 57138
+    },
+    {
+      "epoch": 5.86843997124371,
+      "grad_norm": 0.06103064864873886,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 57141
+    },
+    {
+      "epoch": 5.868748074355551,
+      "grad_norm": 0.05619410425424576,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 57144
+    },
+    {
+      "epoch": 5.869056177467392,
+      "grad_norm": 0.07958339899778366,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 57147
+    },
+    {
+      "epoch": 5.869364280579234,
+      "grad_norm": 0.12638811767101288,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 57150
+    },
+    {
+      "epoch": 5.869672383691075,
+      "grad_norm": 0.05043957009911537,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 57153
+    },
+    {
+      "epoch": 5.869980486802917,
+      "grad_norm": 0.049426887184381485,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 57156
+    },
+    {
+      "epoch": 5.870288589914758,
+      "grad_norm": 0.058896906673908234,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 57159
+    },
+    {
+      "epoch": 5.8705966930266,
+      "grad_norm": 0.06517619639635086,
+      "learning_rate": 0.01,
+      "loss": 2.0153,
+      "step": 57162
+    },
+    {
+      "epoch": 5.870904796138441,
+      "grad_norm": 0.04372745007276535,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 57165
+    },
+    {
+      "epoch": 5.871212899250282,
+      "grad_norm": 0.11192513257265091,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 57168
+    },
+    {
+      "epoch": 5.871521002362124,
+      "grad_norm": 0.04966152831912041,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 57171
+    },
+    {
+      "epoch": 5.8718291054739655,
+      "grad_norm": 0.04251588135957718,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 57174
+    },
+    {
+      "epoch": 5.872137208585807,
+      "grad_norm": 0.03627980872988701,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 57177
+    },
+    {
+      "epoch": 5.872445311697648,
+      "grad_norm": 0.035204388201236725,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 57180
+    },
+    {
+      "epoch": 5.872753414809489,
+      "grad_norm": 0.097038634121418,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 57183
+    },
+    {
+      "epoch": 5.873061517921331,
+      "grad_norm": 0.0762251615524292,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 57186
+    },
+    {
+      "epoch": 5.873369621033173,
+      "grad_norm": 0.12268881499767303,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 57189
+    },
+    {
+      "epoch": 5.873677724145014,
+      "grad_norm": 0.04830612614750862,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 57192
+    },
+    {
+      "epoch": 5.873985827256855,
+      "grad_norm": 0.03913014009594917,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 57195
+    },
+    {
+      "epoch": 5.8742939303686965,
+      "grad_norm": 0.043585240840911865,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 57198
+    },
+    {
+      "epoch": 5.874602033480538,
+      "grad_norm": 0.07954023778438568,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 57201
+    },
+    {
+      "epoch": 5.87491013659238,
+      "grad_norm": 0.08643771708011627,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 57204
+    },
+    {
+      "epoch": 5.875218239704221,
+      "grad_norm": 0.0801578015089035,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 57207
+    },
+    {
+      "epoch": 5.8755263428160625,
+      "grad_norm": 0.07038013637065887,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 57210
+    },
+    {
+      "epoch": 5.875834445927904,
+      "grad_norm": 0.09337194263935089,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 57213
+    },
+    {
+      "epoch": 5.876142549039745,
+      "grad_norm": 0.07053009420633316,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 57216
+    },
+    {
+      "epoch": 5.876450652151586,
+      "grad_norm": 0.0576632060110569,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 57219
+    },
+    {
+      "epoch": 5.876758755263428,
+      "grad_norm": 0.0459139384329319,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 57222
+    },
+    {
+      "epoch": 5.87706685837527,
+      "grad_norm": 0.0351143553853035,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 57225
+    },
+    {
+      "epoch": 5.877374961487111,
+      "grad_norm": 0.07681302726268768,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 57228
+    },
+    {
+      "epoch": 5.877683064598952,
+      "grad_norm": 0.08413039892911911,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 57231
+    },
+    {
+      "epoch": 5.877991167710794,
+      "grad_norm": 0.07289810478687286,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 57234
+    },
+    {
+      "epoch": 5.878299270822636,
+      "grad_norm": 0.10013864189386368,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 57237
+    },
+    {
+      "epoch": 5.878607373934477,
+      "grad_norm": 0.042369335889816284,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 57240
+    },
+    {
+      "epoch": 5.878915477046318,
+      "grad_norm": 0.06880860030651093,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 57243
+    },
+    {
+      "epoch": 5.879223580158159,
+      "grad_norm": 0.07166635245084763,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 57246
+    },
+    {
+      "epoch": 5.879531683270001,
+      "grad_norm": 0.07346632331609726,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 57249
+    },
+    {
+      "epoch": 5.879839786381843,
+      "grad_norm": 0.039783868938684464,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 57252
+    },
+    {
+      "epoch": 5.880147889493684,
+      "grad_norm": 0.03261881321668625,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 57255
+    },
+    {
+      "epoch": 5.880455992605525,
+      "grad_norm": 0.05967556685209274,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 57258
+    },
+    {
+      "epoch": 5.880764095717367,
+      "grad_norm": 0.09463750571012497,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 57261
+    },
+    {
+      "epoch": 5.881072198829208,
+      "grad_norm": 0.11261694133281708,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 57264
+    },
+    {
+      "epoch": 5.88138030194105,
+      "grad_norm": 0.04502738267183304,
+      "learning_rate": 0.01,
+      "loss": 2.0151,
+      "step": 57267
+    },
+    {
+      "epoch": 5.881688405052891,
+      "grad_norm": 0.048553161323070526,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 57270
+    },
+    {
+      "epoch": 5.8819965081647325,
+      "grad_norm": 0.04814717173576355,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 57273
+    },
+    {
+      "epoch": 5.882304611276574,
+      "grad_norm": 0.11577591300010681,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 57276
+    },
+    {
+      "epoch": 5.882612714388415,
+      "grad_norm": 0.07953160256147385,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 57279
+    },
+    {
+      "epoch": 5.882920817500256,
+      "grad_norm": 0.0455656498670578,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 57282
+    },
+    {
+      "epoch": 5.8832289206120985,
+      "grad_norm": 0.035418231040239334,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 57285
+    },
+    {
+      "epoch": 5.88353702372394,
+      "grad_norm": 0.03444494679570198,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 57288
+    },
+    {
+      "epoch": 5.883845126835781,
+      "grad_norm": 0.03543773293495178,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 57291
+    },
+    {
+      "epoch": 5.884153229947622,
+      "grad_norm": 0.032163105905056,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 57294
+    },
+    {
+      "epoch": 5.884461333059464,
+      "grad_norm": 0.051954787224531174,
+      "learning_rate": 0.01,
+      "loss": 2.0193,
+      "step": 57297
+    },
+    {
+      "epoch": 5.884769436171306,
+      "grad_norm": 0.046241723001003265,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 57300
+    },
+    {
+      "epoch": 5.885077539283147,
+      "grad_norm": 0.08583594113588333,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 57303
+    },
+    {
+      "epoch": 5.885385642394988,
+      "grad_norm": 0.10830538719892502,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 57306
+    },
+    {
+      "epoch": 5.885693745506829,
+      "grad_norm": 0.04670783504843712,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 57309
+    },
+    {
+      "epoch": 5.886001848618671,
+      "grad_norm": 0.08331948518753052,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 57312
+    },
+    {
+      "epoch": 5.886309951730513,
+      "grad_norm": 0.12991750240325928,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 57315
+    },
+    {
+      "epoch": 5.886618054842354,
+      "grad_norm": 0.03781532123684883,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 57318
+    },
+    {
+      "epoch": 5.886926157954195,
+      "grad_norm": 0.03549522906541824,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 57321
+    },
+    {
+      "epoch": 5.887234261066037,
+      "grad_norm": 0.03147748485207558,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 57324
+    },
+    {
+      "epoch": 5.887542364177878,
+      "grad_norm": 0.06375817954540253,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 57327
+    },
+    {
+      "epoch": 5.88785046728972,
+      "grad_norm": 0.06203543394804001,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 57330
+    },
+    {
+      "epoch": 5.888158570401561,
+      "grad_norm": 0.07817850261926651,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 57333
+    },
+    {
+      "epoch": 5.888466673513403,
+      "grad_norm": 0.06806226819753647,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 57336
+    },
+    {
+      "epoch": 5.888774776625244,
+      "grad_norm": 0.06197969987988472,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 57339
+    },
+    {
+      "epoch": 5.889082879737085,
+      "grad_norm": 0.10083005577325821,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 57342
+    },
+    {
+      "epoch": 5.889390982848926,
+      "grad_norm": 0.04016329348087311,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 57345
+    },
+    {
+      "epoch": 5.8896990859607685,
+      "grad_norm": 0.047332972288131714,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 57348
+    },
+    {
+      "epoch": 5.89000718907261,
+      "grad_norm": 0.042285602539777756,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 57351
+    },
+    {
+      "epoch": 5.890315292184451,
+      "grad_norm": 0.0776951014995575,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 57354
+    },
+    {
+      "epoch": 5.890623395296292,
+      "grad_norm": 0.1287294626235962,
+      "learning_rate": 0.01,
+      "loss": 2.0342,
+      "step": 57357
+    },
+    {
+      "epoch": 5.890931498408134,
+      "grad_norm": 0.08230342715978622,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 57360
+    },
+    {
+      "epoch": 5.891239601519976,
+      "grad_norm": 0.03847470134496689,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 57363
+    },
+    {
+      "epoch": 5.891547704631817,
+      "grad_norm": 0.07836028188467026,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 57366
+    },
+    {
+      "epoch": 5.891855807743658,
+      "grad_norm": 0.03260621055960655,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 57369
+    },
+    {
+      "epoch": 5.8921639108554995,
+      "grad_norm": 0.036644842475652695,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 57372
+    },
+    {
+      "epoch": 5.892472013967341,
+      "grad_norm": 0.03415127098560333,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 57375
+    },
+    {
+      "epoch": 5.892780117079182,
+      "grad_norm": 0.05682433769106865,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 57378
+    },
+    {
+      "epoch": 5.893088220191024,
+      "grad_norm": 0.09808420389890671,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 57381
+    },
+    {
+      "epoch": 5.8933963233028654,
+      "grad_norm": 0.05321237072348595,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 57384
+    },
+    {
+      "epoch": 5.893704426414707,
+      "grad_norm": 0.03606530278921127,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 57387
+    },
+    {
+      "epoch": 5.894012529526548,
+      "grad_norm": 0.04051011800765991,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 57390
+    },
+    {
+      "epoch": 5.89432063263839,
+      "grad_norm": 0.039596255868673325,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 57393
+    },
+    {
+      "epoch": 5.894628735750231,
+      "grad_norm": 0.0865618884563446,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 57396
+    },
+    {
+      "epoch": 5.894936838862073,
+      "grad_norm": 0.05438371002674103,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 57399
+    },
+    {
+      "epoch": 5.895244941973914,
+      "grad_norm": 0.14930184185504913,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 57402
+    },
+    {
+      "epoch": 5.895553045085755,
+      "grad_norm": 0.05614696815609932,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 57405
+    },
+    {
+      "epoch": 5.895861148197596,
+      "grad_norm": 0.06839166581630707,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 57408
+    },
+    {
+      "epoch": 5.896169251309439,
+      "grad_norm": 0.0660998523235321,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 57411
+    },
+    {
+      "epoch": 5.89647735442128,
+      "grad_norm": 0.07487353682518005,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 57414
+    },
+    {
+      "epoch": 5.896785457533121,
+      "grad_norm": 0.05647150054574013,
+      "learning_rate": 0.01,
+      "loss": 2.0115,
+      "step": 57417
+    },
+    {
+      "epoch": 5.897093560644962,
+      "grad_norm": 0.04570586606860161,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 57420
+    },
+    {
+      "epoch": 5.897401663756804,
+      "grad_norm": 0.10942738503217697,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 57423
+    },
+    {
+      "epoch": 5.897709766868646,
+      "grad_norm": 0.08452852070331573,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 57426
+    },
+    {
+      "epoch": 5.898017869980487,
+      "grad_norm": 0.07079064846038818,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 57429
+    },
+    {
+      "epoch": 5.898325973092328,
+      "grad_norm": 0.09797564148902893,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 57432
+    },
+    {
+      "epoch": 5.89863407620417,
+      "grad_norm": 0.08445065468549728,
+      "learning_rate": 0.01,
+      "loss": 2.0102,
+      "step": 57435
+    },
+    {
+      "epoch": 5.898942179316011,
+      "grad_norm": 0.08496986329555511,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 57438
+    },
+    {
+      "epoch": 5.899250282427852,
+      "grad_norm": 0.04799206554889679,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 57441
+    },
+    {
+      "epoch": 5.899558385539694,
+      "grad_norm": 0.047865018248558044,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 57444
+    },
+    {
+      "epoch": 5.8998664886515355,
+      "grad_norm": 0.040415599942207336,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 57447
+    },
+    {
+      "epoch": 5.900174591763377,
+      "grad_norm": 0.05303875729441643,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 57450
+    },
+    {
+      "epoch": 5.900482694875218,
+      "grad_norm": 0.13320137560367584,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 57453
+    },
+    {
+      "epoch": 5.900790797987059,
+      "grad_norm": 0.05218903720378876,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 57456
+    },
+    {
+      "epoch": 5.9010989010989015,
+      "grad_norm": 0.06745684146881104,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 57459
+    },
+    {
+      "epoch": 5.901407004210743,
+      "grad_norm": 0.04268602281808853,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 57462
+    },
+    {
+      "epoch": 5.901715107322584,
+      "grad_norm": 0.034592967480421066,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 57465
+    },
+    {
+      "epoch": 5.902023210434425,
+      "grad_norm": 0.11330728232860565,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 57468
+    },
+    {
+      "epoch": 5.9023313135462665,
+      "grad_norm": 0.035573434084653854,
+      "learning_rate": 0.01,
+      "loss": 2.0189,
+      "step": 57471
+    },
+    {
+      "epoch": 5.902639416658108,
+      "grad_norm": 0.09724743664264679,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 57474
+    },
+    {
+      "epoch": 5.90294751976995,
+      "grad_norm": 0.06769842654466629,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 57477
+    },
+    {
+      "epoch": 5.903255622881791,
+      "grad_norm": 0.07262661308050156,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 57480
+    },
+    {
+      "epoch": 5.903563725993632,
+      "grad_norm": 0.08442901074886322,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 57483
+    },
+    {
+      "epoch": 5.903871829105474,
+      "grad_norm": 0.07244334369897842,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 57486
+    },
+    {
+      "epoch": 5.904179932217316,
+      "grad_norm": 0.0898541584610939,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 57489
+    },
+    {
+      "epoch": 5.904488035329157,
+      "grad_norm": 0.043548524379730225,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 57492
+    },
+    {
+      "epoch": 5.904796138440998,
+      "grad_norm": 0.035105109214782715,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 57495
+    },
+    {
+      "epoch": 5.90510424155284,
+      "grad_norm": 0.06654632091522217,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 57498
+    },
+    {
+      "epoch": 5.905412344664681,
+      "grad_norm": 0.09614937752485275,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 57501
+    },
+    {
+      "epoch": 5.905720447776522,
+      "grad_norm": 0.07232387363910675,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 57504
+    },
+    {
+      "epoch": 5.906028550888364,
+      "grad_norm": 0.06795191764831543,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 57507
+    },
+    {
+      "epoch": 5.906336654000206,
+      "grad_norm": 0.1627775877714157,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 57510
+    },
+    {
+      "epoch": 5.906644757112047,
+      "grad_norm": 0.043800655752420425,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 57513
+    },
+    {
+      "epoch": 5.906952860223888,
+      "grad_norm": 0.09019804745912552,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 57516
+    },
+    {
+      "epoch": 5.907260963335729,
+      "grad_norm": 0.03738325461745262,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 57519
+    },
+    {
+      "epoch": 5.9075690664475715,
+      "grad_norm": 0.05333522707223892,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 57522
+    },
+    {
+      "epoch": 5.907877169559413,
+      "grad_norm": 0.03560793399810791,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 57525
+    },
+    {
+      "epoch": 5.908185272671254,
+      "grad_norm": 0.045784592628479004,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 57528
+    },
+    {
+      "epoch": 5.908493375783095,
+      "grad_norm": 0.05486133694648743,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 57531
+    },
+    {
+      "epoch": 5.908801478894937,
+      "grad_norm": 0.06346391886472702,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 57534
+    },
+    {
+      "epoch": 5.909109582006778,
+      "grad_norm": 0.06552404165267944,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 57537
+    },
+    {
+      "epoch": 5.90941768511862,
+      "grad_norm": 0.07176794856786728,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 57540
+    },
+    {
+      "epoch": 5.909725788230461,
+      "grad_norm": 0.05690211430191994,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 57543
+    },
+    {
+      "epoch": 5.9100338913423025,
+      "grad_norm": 0.07518581300973892,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 57546
+    },
+    {
+      "epoch": 5.910341994454144,
+      "grad_norm": 0.09749139845371246,
+      "learning_rate": 0.01,
+      "loss": 2.0104,
+      "step": 57549
+    },
+    {
+      "epoch": 5.910650097565985,
+      "grad_norm": 0.06265246123075485,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 57552
+    },
+    {
+      "epoch": 5.910958200677827,
+      "grad_norm": 0.09425773471593857,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 57555
+    },
+    {
+      "epoch": 5.911266303789668,
+      "grad_norm": 0.07862033694982529,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 57558
+    },
+    {
+      "epoch": 5.91157440690151,
+      "grad_norm": 0.052102018147706985,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 57561
+    },
+    {
+      "epoch": 5.911882510013351,
+      "grad_norm": 0.04333946481347084,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 57564
+    },
+    {
+      "epoch": 5.912190613125192,
+      "grad_norm": 0.03312124311923981,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 57567
+    },
+    {
+      "epoch": 5.912498716237034,
+      "grad_norm": 0.11217939108610153,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 57570
+    },
+    {
+      "epoch": 5.912806819348876,
+      "grad_norm": 0.050783589482307434,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 57573
+    },
+    {
+      "epoch": 5.913114922460717,
+      "grad_norm": 0.05247243866324425,
+      "learning_rate": 0.01,
+      "loss": 2.0166,
+      "step": 57576
+    },
+    {
+      "epoch": 5.913423025572558,
+      "grad_norm": 0.03641197457909584,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 57579
+    },
+    {
+      "epoch": 5.913731128684399,
+      "grad_norm": 0.03802553936839104,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 57582
+    },
+    {
+      "epoch": 5.914039231796242,
+      "grad_norm": 0.04474913701415062,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 57585
+    },
+    {
+      "epoch": 5.914347334908083,
+      "grad_norm": 0.050698988139629364,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 57588
+    },
+    {
+      "epoch": 5.914655438019924,
+      "grad_norm": 0.12217648327350616,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 57591
+    },
+    {
+      "epoch": 5.914963541131765,
+      "grad_norm": 0.10421942919492722,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 57594
+    },
+    {
+      "epoch": 5.915271644243607,
+      "grad_norm": 0.09191686660051346,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 57597
+    },
+    {
+      "epoch": 5.915579747355448,
+      "grad_norm": 0.07071512192487717,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 57600
+    },
+    {
+      "epoch": 5.91588785046729,
+      "grad_norm": 0.0706767663359642,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 57603
+    },
+    {
+      "epoch": 5.916195953579131,
+      "grad_norm": 0.04327024519443512,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 57606
+    },
+    {
+      "epoch": 5.916504056690973,
+      "grad_norm": 0.04743794724345207,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 57609
+    },
+    {
+      "epoch": 5.916812159802814,
+      "grad_norm": 0.04772097244858742,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 57612
+    },
+    {
+      "epoch": 5.917120262914655,
+      "grad_norm": 0.0430108904838562,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 57615
+    },
+    {
+      "epoch": 5.917428366026497,
+      "grad_norm": 0.07793119549751282,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 57618
+    },
+    {
+      "epoch": 5.9177364691383385,
+      "grad_norm": 0.10654459148645401,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 57621
+    },
+    {
+      "epoch": 5.91804457225018,
+      "grad_norm": 0.03443089500069618,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 57624
+    },
+    {
+      "epoch": 5.918352675362021,
+      "grad_norm": 0.07121492922306061,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 57627
+    },
+    {
+      "epoch": 5.918660778473862,
+      "grad_norm": 0.04436872899532318,
+      "learning_rate": 0.01,
+      "loss": 2.0261,
+      "step": 57630
+    },
+    {
+      "epoch": 5.918968881585704,
+      "grad_norm": 0.08435078710317612,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 57633
+    },
+    {
+      "epoch": 5.919276984697546,
+      "grad_norm": 0.1054021418094635,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 57636
+    },
+    {
+      "epoch": 5.919585087809387,
+      "grad_norm": 0.05516025796532631,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 57639
+    },
+    {
+      "epoch": 5.919893190921228,
+      "grad_norm": 0.067764513194561,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 57642
+    },
+    {
+      "epoch": 5.9202012940330695,
+      "grad_norm": 0.0520036555826664,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 57645
+    },
+    {
+      "epoch": 5.920509397144912,
+      "grad_norm": 0.071280837059021,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 57648
+    },
+    {
+      "epoch": 5.920817500256753,
+      "grad_norm": 0.04528915137052536,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 57651
+    },
+    {
+      "epoch": 5.921125603368594,
+      "grad_norm": 0.0340840145945549,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 57654
+    },
+    {
+      "epoch": 5.921433706480435,
+      "grad_norm": 0.033173661679029465,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 57657
+    },
+    {
+      "epoch": 5.921741809592277,
+      "grad_norm": 0.13926701247692108,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 57660
+    },
+    {
+      "epoch": 5.922049912704118,
+      "grad_norm": 0.07138355076313019,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 57663
+    },
+    {
+      "epoch": 5.92235801581596,
+      "grad_norm": 0.0728163793683052,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 57666
+    },
+    {
+      "epoch": 5.922666118927801,
+      "grad_norm": 0.08973317593336105,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 57669
+    },
+    {
+      "epoch": 5.922974222039643,
+      "grad_norm": 0.0479208379983902,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 57672
+    },
+    {
+      "epoch": 5.923282325151484,
+      "grad_norm": 0.05372065678238869,
+      "learning_rate": 0.01,
+      "loss": 2.0279,
+      "step": 57675
+    },
+    {
+      "epoch": 5.923590428263325,
+      "grad_norm": 0.048417530953884125,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 57678
+    },
+    {
+      "epoch": 5.923898531375167,
+      "grad_norm": 0.03113439865410328,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 57681
+    },
+    {
+      "epoch": 5.924206634487009,
+      "grad_norm": 0.0359586738049984,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 57684
+    },
+    {
+      "epoch": 5.92451473759885,
+      "grad_norm": 0.12019523233175278,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 57687
+    },
+    {
+      "epoch": 5.924822840710691,
+      "grad_norm": 0.0717984288930893,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 57690
+    },
+    {
+      "epoch": 5.925130943822532,
+      "grad_norm": 0.06465215235948563,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 57693
+    },
+    {
+      "epoch": 5.925439046934374,
+      "grad_norm": 0.1053556352853775,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 57696
+    },
+    {
+      "epoch": 5.925747150046216,
+      "grad_norm": 0.04565432667732239,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 57699
+    },
+    {
+      "epoch": 5.926055253158057,
+      "grad_norm": 0.05233050882816315,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 57702
+    },
+    {
+      "epoch": 5.926363356269898,
+      "grad_norm": 0.05413827672600746,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 57705
+    },
+    {
+      "epoch": 5.92667145938174,
+      "grad_norm": 0.0694475993514061,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 57708
+    },
+    {
+      "epoch": 5.926979562493581,
+      "grad_norm": 0.03973362222313881,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 57711
+    },
+    {
+      "epoch": 5.927287665605423,
+      "grad_norm": 0.048481449484825134,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 57714
+    },
+    {
+      "epoch": 5.927595768717264,
+      "grad_norm": 0.04115751013159752,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 57717
+    },
+    {
+      "epoch": 5.9279038718291055,
+      "grad_norm": 0.056814514100551605,
+      "learning_rate": 0.01,
+      "loss": 2.0123,
+      "step": 57720
+    },
+    {
+      "epoch": 5.928211974940947,
+      "grad_norm": 0.07510741055011749,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 57723
+    },
+    {
+      "epoch": 5.928520078052788,
+      "grad_norm": 0.10000273585319519,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 57726
+    },
+    {
+      "epoch": 5.928828181164629,
+      "grad_norm": 0.10124478489160538,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 57729
+    },
+    {
+      "epoch": 5.929136284276471,
+      "grad_norm": 0.1368652582168579,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 57732
+    },
+    {
+      "epoch": 5.929444387388313,
+      "grad_norm": 0.06806531548500061,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 57735
+    },
+    {
+      "epoch": 5.929752490500154,
+      "grad_norm": 0.03443896025419235,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 57738
+    },
+    {
+      "epoch": 5.930060593611995,
+      "grad_norm": 0.03277003765106201,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 57741
+    },
+    {
+      "epoch": 5.930368696723837,
+      "grad_norm": 0.04082879796624184,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 57744
+    },
+    {
+      "epoch": 5.930676799835679,
+      "grad_norm": 0.11657889932394028,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 57747
+    },
+    {
+      "epoch": 5.93098490294752,
+      "grad_norm": 0.08914856612682343,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 57750
+    },
+    {
+      "epoch": 5.931293006059361,
+      "grad_norm": 0.06309764087200165,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 57753
+    },
+    {
+      "epoch": 5.931601109171202,
+      "grad_norm": 0.051464032381772995,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 57756
+    },
+    {
+      "epoch": 5.931909212283044,
+      "grad_norm": 0.037711381912231445,
+      "learning_rate": 0.01,
+      "loss": 2.0257,
+      "step": 57759
+    },
+    {
+      "epoch": 5.932217315394886,
+      "grad_norm": 0.06184235215187073,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 57762
+    },
+    {
+      "epoch": 5.932525418506727,
+      "grad_norm": 0.07253403961658478,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 57765
+    },
+    {
+      "epoch": 5.932833521618568,
+      "grad_norm": 0.0968317911028862,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 57768
+    },
+    {
+      "epoch": 5.93314162473041,
+      "grad_norm": 0.0691918134689331,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 57771
+    },
+    {
+      "epoch": 5.933449727842251,
+      "grad_norm": 0.0434565395116806,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 57774
+    },
+    {
+      "epoch": 5.933757830954093,
+      "grad_norm": 0.04627044126391411,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 57777
+    },
+    {
+      "epoch": 5.934065934065934,
+      "grad_norm": 0.04456416517496109,
+      "learning_rate": 0.01,
+      "loss": 2.0219,
+      "step": 57780
+    },
+    {
+      "epoch": 5.934374037177776,
+      "grad_norm": 0.038884878158569336,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 57783
+    },
+    {
+      "epoch": 5.934682140289617,
+      "grad_norm": 0.03564108535647392,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 57786
+    },
+    {
+      "epoch": 5.934990243401458,
+      "grad_norm": 0.06104537844657898,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 57789
+    },
+    {
+      "epoch": 5.935298346513299,
+      "grad_norm": 0.06845796853303909,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 57792
+    },
+    {
+      "epoch": 5.9356064496251415,
+      "grad_norm": 0.08483695238828659,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 57795
+    },
+    {
+      "epoch": 5.935914552736983,
+      "grad_norm": 0.05989287048578262,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 57798
+    },
+    {
+      "epoch": 5.936222655848824,
+      "grad_norm": 0.04331030324101448,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 57801
+    },
+    {
+      "epoch": 5.936530758960665,
+      "grad_norm": 0.05447021499276161,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 57804
+    },
+    {
+      "epoch": 5.9368388620725066,
+      "grad_norm": 0.10276515781879425,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 57807
+    },
+    {
+      "epoch": 5.937146965184349,
+      "grad_norm": 0.041528135538101196,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 57810
+    },
+    {
+      "epoch": 5.93745506829619,
+      "grad_norm": 0.09760654717683792,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 57813
+    },
+    {
+      "epoch": 5.937763171408031,
+      "grad_norm": 0.07332238554954529,
+      "learning_rate": 0.01,
+      "loss": 2.0096,
+      "step": 57816
+    },
+    {
+      "epoch": 5.9380712745198725,
+      "grad_norm": 0.07201950252056122,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 57819
+    },
+    {
+      "epoch": 5.938379377631714,
+      "grad_norm": 0.09542597830295563,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 57822
+    },
+    {
+      "epoch": 5.938687480743555,
+      "grad_norm": 0.08886115252971649,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 57825
+    },
+    {
+      "epoch": 5.938995583855397,
+      "grad_norm": 0.0604422464966774,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 57828
+    },
+    {
+      "epoch": 5.939303686967238,
+      "grad_norm": 0.04549078643321991,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 57831
+    },
+    {
+      "epoch": 5.93961179007908,
+      "grad_norm": 0.09901918470859528,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 57834
+    },
+    {
+      "epoch": 5.939919893190921,
+      "grad_norm": 0.10208103060722351,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 57837
+    },
+    {
+      "epoch": 5.940227996302763,
+      "grad_norm": 0.08842794597148895,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 57840
+    },
+    {
+      "epoch": 5.940536099414604,
+      "grad_norm": 0.06819964945316315,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 57843
+    },
+    {
+      "epoch": 5.940844202526446,
+      "grad_norm": 0.07434326410293579,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 57846
+    },
+    {
+      "epoch": 5.941152305638287,
+      "grad_norm": 0.06985678523778915,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 57849
+    },
+    {
+      "epoch": 5.941460408750128,
+      "grad_norm": 0.04155593365430832,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 57852
+    },
+    {
+      "epoch": 5.941768511861969,
+      "grad_norm": 0.07389968633651733,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 57855
+    },
+    {
+      "epoch": 5.942076614973812,
+      "grad_norm": 0.11060785502195358,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 57858
+    },
+    {
+      "epoch": 5.942384718085653,
+      "grad_norm": 0.035507991909980774,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 57861
+    },
+    {
+      "epoch": 5.942692821197494,
+      "grad_norm": 0.07544751465320587,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 57864
+    },
+    {
+      "epoch": 5.943000924309335,
+      "grad_norm": 0.03793545812368393,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 57867
+    },
+    {
+      "epoch": 5.943309027421177,
+      "grad_norm": 0.07497958838939667,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 57870
+    },
+    {
+      "epoch": 5.943617130533019,
+      "grad_norm": 0.0712079107761383,
+      "learning_rate": 0.01,
+      "loss": 2.0135,
+      "step": 57873
+    },
+    {
+      "epoch": 5.94392523364486,
+      "grad_norm": 0.03715150058269501,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 57876
+    },
+    {
+      "epoch": 5.944233336756701,
+      "grad_norm": 0.0650571882724762,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 57879
+    },
+    {
+      "epoch": 5.9445414398685426,
+      "grad_norm": 0.0977357029914856,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 57882
+    },
+    {
+      "epoch": 5.944849542980384,
+      "grad_norm": 0.06483075022697449,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 57885
+    },
+    {
+      "epoch": 5.945157646092225,
+      "grad_norm": 0.09345372766256332,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 57888
+    },
+    {
+      "epoch": 5.945465749204067,
+      "grad_norm": 0.06418720632791519,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 57891
+    },
+    {
+      "epoch": 5.9457738523159085,
+      "grad_norm": 0.05608039349317551,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 57894
+    },
+    {
+      "epoch": 5.94608195542775,
+      "grad_norm": 0.04540601745247841,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 57897
+    },
+    {
+      "epoch": 5.946390058539591,
+      "grad_norm": 0.07881978154182434,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 57900
+    },
+    {
+      "epoch": 5.946698161651433,
+      "grad_norm": 0.05478779971599579,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 57903
+    },
+    {
+      "epoch": 5.947006264763274,
+      "grad_norm": 0.06574473530054092,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 57906
+    },
+    {
+      "epoch": 5.947314367875116,
+      "grad_norm": 0.06080929934978485,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 57909
+    },
+    {
+      "epoch": 5.947622470986957,
+      "grad_norm": 0.06462790071964264,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 57912
+    },
+    {
+      "epoch": 5.947930574098798,
+      "grad_norm": 0.06746607273817062,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 57915
+    },
+    {
+      "epoch": 5.9482386772106395,
+      "grad_norm": 0.048096004873514175,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 57918
+    },
+    {
+      "epoch": 5.948546780322482,
+      "grad_norm": 0.038138337433338165,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 57921
+    },
+    {
+      "epoch": 5.948854883434323,
+      "grad_norm": 0.05787679925560951,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 57924
+    },
+    {
+      "epoch": 5.949162986546164,
+      "grad_norm": 0.21429073810577393,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 57927
+    },
+    {
+      "epoch": 5.949471089658005,
+      "grad_norm": 0.103017657995224,
+      "learning_rate": 0.01,
+      "loss": 2.0201,
+      "step": 57930
+    },
+    {
+      "epoch": 5.949779192769847,
+      "grad_norm": 0.08076374232769012,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 57933
+    },
+    {
+      "epoch": 5.950087295881689,
+      "grad_norm": 0.06803920865058899,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 57936
+    },
+    {
+      "epoch": 5.95039539899353,
+      "grad_norm": 0.0801796242594719,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 57939
+    },
+    {
+      "epoch": 5.950703502105371,
+      "grad_norm": 0.0715690553188324,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 57942
+    },
+    {
+      "epoch": 5.951011605217213,
+      "grad_norm": 0.051856063306331635,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 57945
+    },
+    {
+      "epoch": 5.951319708329054,
+      "grad_norm": 0.03726186603307724,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 57948
+    },
+    {
+      "epoch": 5.951627811440895,
+      "grad_norm": 0.04276345297694206,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 57951
+    },
+    {
+      "epoch": 5.951935914552737,
+      "grad_norm": 0.06446687877178192,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 57954
+    },
+    {
+      "epoch": 5.952244017664579,
+      "grad_norm": 0.05513986572623253,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 57957
+    },
+    {
+      "epoch": 5.95255212077642,
+      "grad_norm": 0.06750188022851944,
+      "learning_rate": 0.01,
+      "loss": 2.0185,
+      "step": 57960
+    },
+    {
+      "epoch": 5.952860223888261,
+      "grad_norm": 0.048399876803159714,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 57963
+    },
+    {
+      "epoch": 5.953168327000102,
+      "grad_norm": 0.04391476884484291,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 57966
+    },
+    {
+      "epoch": 5.9534764301119445,
+      "grad_norm": 0.05262988805770874,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 57969
+    },
+    {
+      "epoch": 5.953784533223786,
+      "grad_norm": 0.1819307655096054,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 57972
+    },
+    {
+      "epoch": 5.954092636335627,
+      "grad_norm": 0.04456859081983566,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 57975
+    },
+    {
+      "epoch": 5.954400739447468,
+      "grad_norm": 0.11903820186853409,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 57978
+    },
+    {
+      "epoch": 5.9547088425593095,
+      "grad_norm": 0.08669394254684448,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 57981
+    },
+    {
+      "epoch": 5.955016945671151,
+      "grad_norm": 0.05538726970553398,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 57984
+    },
+    {
+      "epoch": 5.955325048782993,
+      "grad_norm": 0.06540273129940033,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 57987
+    },
+    {
+      "epoch": 5.955633151894834,
+      "grad_norm": 0.05096886307001114,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 57990
+    },
+    {
+      "epoch": 5.9559412550066755,
+      "grad_norm": 0.0463346429169178,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 57993
+    },
+    {
+      "epoch": 5.956249358118517,
+      "grad_norm": 0.12739206850528717,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 57996
+    },
+    {
+      "epoch": 5.956557461230359,
+      "grad_norm": 0.03725777193903923,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 57999
+    },
+    {
+      "epoch": 5.9568655643422,
+      "grad_norm": 0.041961897164583206,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 58002
+    },
+    {
+      "epoch": 5.957173667454041,
+      "grad_norm": 0.039089135825634,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 58005
+    },
+    {
+      "epoch": 5.957481770565883,
+      "grad_norm": 0.03455056622624397,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 58008
+    },
+    {
+      "epoch": 5.957789873677724,
+      "grad_norm": 0.10427875816822052,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 58011
+    },
+    {
+      "epoch": 5.958097976789565,
+      "grad_norm": 0.11531616002321243,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 58014
+    },
+    {
+      "epoch": 5.958406079901407,
+      "grad_norm": 0.05732328072190285,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 58017
+    },
+    {
+      "epoch": 5.958714183013249,
+      "grad_norm": 0.033681951463222504,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 58020
+    },
+    {
+      "epoch": 5.95902228612509,
+      "grad_norm": 0.03731759265065193,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 58023
+    },
+    {
+      "epoch": 5.959330389236931,
+      "grad_norm": 0.03403918072581291,
+      "learning_rate": 0.01,
+      "loss": 2.0195,
+      "step": 58026
+    },
+    {
+      "epoch": 5.959638492348772,
+      "grad_norm": 0.10777036845684052,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 58029
+    },
+    {
+      "epoch": 5.959946595460615,
+      "grad_norm": 0.03863784670829773,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 58032
+    },
+    {
+      "epoch": 5.960254698572456,
+      "grad_norm": 0.11702859401702881,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 58035
+    },
+    {
+      "epoch": 5.960562801684297,
+      "grad_norm": 0.04561970755457878,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 58038
+    },
+    {
+      "epoch": 5.960870904796138,
+      "grad_norm": 0.07471676170825958,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 58041
+    },
+    {
+      "epoch": 5.96117900790798,
+      "grad_norm": 0.06594023108482361,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 58044
+    },
+    {
+      "epoch": 5.961487111019821,
+      "grad_norm": 0.10957533866167068,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 58047
+    },
+    {
+      "epoch": 5.961795214131663,
+      "grad_norm": 0.06384073197841644,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 58050
+    },
+    {
+      "epoch": 5.962103317243504,
+      "grad_norm": 0.04451090097427368,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 58053
+    },
+    {
+      "epoch": 5.9624114203553455,
+      "grad_norm": 0.036886006593704224,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 58056
+    },
+    {
+      "epoch": 5.962719523467187,
+      "grad_norm": 0.06101905182003975,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 58059
+    },
+    {
+      "epoch": 5.963027626579028,
+      "grad_norm": 0.037433281540870667,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 58062
+    },
+    {
+      "epoch": 5.96333572969087,
+      "grad_norm": 0.04497874528169632,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 58065
+    },
+    {
+      "epoch": 5.9636438328027115,
+      "grad_norm": 0.10465840995311737,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 58068
+    },
+    {
+      "epoch": 5.963951935914553,
+      "grad_norm": 0.08285420387983322,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 58071
+    },
+    {
+      "epoch": 5.964260039026394,
+      "grad_norm": 0.0607406347990036,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 58074
+    },
+    {
+      "epoch": 5.964568142138235,
+      "grad_norm": 0.07265600562095642,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 58077
+    },
+    {
+      "epoch": 5.9648762452500765,
+      "grad_norm": 0.06835401058197021,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 58080
+    },
+    {
+      "epoch": 5.965184348361919,
+      "grad_norm": 0.08795753121376038,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 58083
+    },
+    {
+      "epoch": 5.96549245147376,
+      "grad_norm": 0.06189946457743645,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 58086
+    },
+    {
+      "epoch": 5.965800554585601,
+      "grad_norm": 0.05601467564702034,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 58089
+    },
+    {
+      "epoch": 5.9661086576974425,
+      "grad_norm": 0.07447000592947006,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 58092
+    },
+    {
+      "epoch": 5.966416760809285,
+      "grad_norm": 0.08373255282640457,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 58095
+    },
+    {
+      "epoch": 5.966724863921126,
+      "grad_norm": 0.09241122007369995,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 58098
+    },
+    {
+      "epoch": 5.967032967032967,
+      "grad_norm": 0.0638846680521965,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 58101
+    },
+    {
+      "epoch": 5.967341070144808,
+      "grad_norm": 0.0532197505235672,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 58104
+    },
+    {
+      "epoch": 5.96764917325665,
+      "grad_norm": 0.10008466243743896,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 58107
+    },
+    {
+      "epoch": 5.967957276368491,
+      "grad_norm": 0.047970160841941833,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 58110
+    },
+    {
+      "epoch": 5.968265379480333,
+      "grad_norm": 0.046584948897361755,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 58113
+    },
+    {
+      "epoch": 5.968573482592174,
+      "grad_norm": 0.04055574908852577,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 58116
+    },
+    {
+      "epoch": 5.968881585704016,
+      "grad_norm": 0.11462211608886719,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 58119
+    },
+    {
+      "epoch": 5.969189688815857,
+      "grad_norm": 0.07639250159263611,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 58122
+    },
+    {
+      "epoch": 5.969497791927698,
+      "grad_norm": 0.08450743556022644,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 58125
+    },
+    {
+      "epoch": 5.96980589503954,
+      "grad_norm": 0.06165710464119911,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 58128
+    },
+    {
+      "epoch": 5.9701139981513816,
+      "grad_norm": 0.042419277131557465,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 58131
+    },
+    {
+      "epoch": 5.970422101263223,
+      "grad_norm": 0.04378311708569527,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 58134
+    },
+    {
+      "epoch": 5.970730204375064,
+      "grad_norm": 0.05443147197365761,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 58137
+    },
+    {
+      "epoch": 5.971038307486905,
+      "grad_norm": 0.05718831345438957,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 58140
+    },
+    {
+      "epoch": 5.971346410598747,
+      "grad_norm": 0.0620989128947258,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 58143
+    },
+    {
+      "epoch": 5.971654513710589,
+      "grad_norm": 0.11351976543664932,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 58146
+    },
+    {
+      "epoch": 5.97196261682243,
+      "grad_norm": 0.05350454896688461,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 58149
+    },
+    {
+      "epoch": 5.972270719934271,
+      "grad_norm": 0.08652366697788239,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 58152
+    },
+    {
+      "epoch": 5.9725788230461125,
+      "grad_norm": 0.054177410900592804,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 58155
+    },
+    {
+      "epoch": 5.972886926157955,
+      "grad_norm": 0.047728173434734344,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 58158
+    },
+    {
+      "epoch": 5.973195029269796,
+      "grad_norm": 0.048977095633745193,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 58161
+    },
+    {
+      "epoch": 5.973503132381637,
+      "grad_norm": 0.042776644229888916,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 58164
+    },
+    {
+      "epoch": 5.9738112354934785,
+      "grad_norm": 0.09739521145820618,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 58167
+    },
+    {
+      "epoch": 5.97411933860532,
+      "grad_norm": 0.0636570006608963,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 58170
+    },
+    {
+      "epoch": 5.974427441717161,
+      "grad_norm": 0.09212375432252884,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 58173
+    },
+    {
+      "epoch": 5.974735544829003,
+      "grad_norm": 0.06589896231889725,
+      "learning_rate": 0.01,
+      "loss": 2.0172,
+      "step": 58176
+    },
+    {
+      "epoch": 5.975043647940844,
+      "grad_norm": 0.04317329451441765,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 58179
+    },
+    {
+      "epoch": 5.975351751052686,
+      "grad_norm": 0.04963945969939232,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 58182
+    },
+    {
+      "epoch": 5.975659854164527,
+      "grad_norm": 0.054626185446977615,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 58185
+    },
+    {
+      "epoch": 5.975967957276368,
+      "grad_norm": 0.0750926285982132,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 58188
+    },
+    {
+      "epoch": 5.97627606038821,
+      "grad_norm": 0.09490969032049179,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 58191
+    },
+    {
+      "epoch": 5.976584163500052,
+      "grad_norm": 0.19652721285820007,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 58194
+    },
+    {
+      "epoch": 5.976892266611893,
+      "grad_norm": 0.13692377507686615,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 58197
+    },
+    {
+      "epoch": 5.977200369723734,
+      "grad_norm": 0.10185639560222626,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 58200
+    },
+    {
+      "epoch": 5.977508472835575,
+      "grad_norm": 0.08201511949300766,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 58203
+    },
+    {
+      "epoch": 5.977816575947417,
+      "grad_norm": 0.06355354934930801,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 58206
+    },
+    {
+      "epoch": 5.978124679059259,
+      "grad_norm": 0.07959393411874771,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 58209
+    },
+    {
+      "epoch": 5.9784327821711,
+      "grad_norm": 0.08127464354038239,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 58212
+    },
+    {
+      "epoch": 5.978740885282941,
+      "grad_norm": 0.052127279341220856,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 58215
+    },
+    {
+      "epoch": 5.979048988394783,
+      "grad_norm": 0.05060075968503952,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 58218
+    },
+    {
+      "epoch": 5.979357091506624,
+      "grad_norm": 0.0653509721159935,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 58221
+    },
+    {
+      "epoch": 5.979665194618466,
+      "grad_norm": 0.04588339477777481,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 58224
+    },
+    {
+      "epoch": 5.979973297730307,
+      "grad_norm": 0.027341017499566078,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 58227
+    },
+    {
+      "epoch": 5.9802814008421485,
+      "grad_norm": 0.039760638028383255,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 58230
+    },
+    {
+      "epoch": 5.98058950395399,
+      "grad_norm": 0.07517565786838531,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 58233
+    },
+    {
+      "epoch": 5.980897607065831,
+      "grad_norm": 0.07364608347415924,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 58236
+    },
+    {
+      "epoch": 5.981205710177672,
+      "grad_norm": 0.057517606765031815,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 58239
+    },
+    {
+      "epoch": 5.9815138132895145,
+      "grad_norm": 0.032561108469963074,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 58242
+    },
+    {
+      "epoch": 5.981821916401356,
+      "grad_norm": 0.08549618721008301,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 58245
+    },
+    {
+      "epoch": 5.982130019513197,
+      "grad_norm": 0.17904026806354523,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 58248
+    },
+    {
+      "epoch": 5.982438122625038,
+      "grad_norm": 0.10775711387395859,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 58251
+    },
+    {
+      "epoch": 5.98274622573688,
+      "grad_norm": 0.07369556277990341,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 58254
+    },
+    {
+      "epoch": 5.983054328848722,
+      "grad_norm": 0.059407759457826614,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 58257
+    },
+    {
+      "epoch": 5.983362431960563,
+      "grad_norm": 0.05741313472390175,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 58260
+    },
+    {
+      "epoch": 5.983670535072404,
+      "grad_norm": 0.07792560756206512,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 58263
+    },
+    {
+      "epoch": 5.9839786381842455,
+      "grad_norm": 0.058633413165807724,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 58266
+    },
+    {
+      "epoch": 5.984286741296087,
+      "grad_norm": 0.04207305610179901,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 58269
+    },
+    {
+      "epoch": 5.984594844407929,
+      "grad_norm": 0.039717648178339005,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 58272
+    },
+    {
+      "epoch": 5.98490294751977,
+      "grad_norm": 0.03505009412765503,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 58275
+    },
+    {
+      "epoch": 5.985211050631611,
+      "grad_norm": 0.03421800956130028,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 58278
+    },
+    {
+      "epoch": 5.985519153743453,
+      "grad_norm": 0.11582440882921219,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 58281
+    },
+    {
+      "epoch": 5.985827256855294,
+      "grad_norm": 0.058088093996047974,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 58284
+    },
+    {
+      "epoch": 5.986135359967136,
+      "grad_norm": 0.08388987928628922,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 58287
+    },
+    {
+      "epoch": 5.986443463078977,
+      "grad_norm": 0.07332730293273926,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 58290
+    },
+    {
+      "epoch": 5.986751566190819,
+      "grad_norm": 0.03843742609024048,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 58293
+    },
+    {
+      "epoch": 5.98705966930266,
+      "grad_norm": 0.07863465696573257,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 58296
+    },
+    {
+      "epoch": 5.987367772414501,
+      "grad_norm": 0.08023706078529358,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 58299
+    },
+    {
+      "epoch": 5.987675875526342,
+      "grad_norm": 0.037095215171575546,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 58302
+    },
+    {
+      "epoch": 5.9879839786381845,
+      "grad_norm": 0.04752547666430473,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 58305
+    },
+    {
+      "epoch": 5.988292081750026,
+      "grad_norm": 0.053975239396095276,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 58308
+    },
+    {
+      "epoch": 5.988600184861867,
+      "grad_norm": 0.08200733363628387,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 58311
+    },
+    {
+      "epoch": 5.988908287973708,
+      "grad_norm": 0.056134093552827835,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 58314
+    },
+    {
+      "epoch": 5.98921639108555,
+      "grad_norm": 0.04401000216603279,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 58317
+    },
+    {
+      "epoch": 5.989524494197392,
+      "grad_norm": 0.06809469312429428,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 58320
+    },
+    {
+      "epoch": 5.989832597309233,
+      "grad_norm": 0.06399431079626083,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 58323
+    },
+    {
+      "epoch": 5.990140700421074,
+      "grad_norm": 0.1221606507897377,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 58326
+    },
+    {
+      "epoch": 5.9904488035329155,
+      "grad_norm": 0.054819442331790924,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 58329
+    },
+    {
+      "epoch": 5.990756906644757,
+      "grad_norm": 0.053588200360536575,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 58332
+    },
+    {
+      "epoch": 5.991065009756598,
+      "grad_norm": 0.060072384774684906,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 58335
+    },
+    {
+      "epoch": 5.99137311286844,
+      "grad_norm": 0.09491874277591705,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 58338
+    },
+    {
+      "epoch": 5.9916812159802815,
+      "grad_norm": 0.054036714136600494,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 58341
+    },
+    {
+      "epoch": 5.991989319092123,
+      "grad_norm": 0.04162592813372612,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 58344
+    },
+    {
+      "epoch": 5.992297422203964,
+      "grad_norm": 0.05602291598916054,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 58347
+    },
+    {
+      "epoch": 5.992605525315806,
+      "grad_norm": 0.038415491580963135,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 58350
+    },
+    {
+      "epoch": 5.992913628427647,
+      "grad_norm": 0.07123242318630219,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 58353
+    },
+    {
+      "epoch": 5.993221731539489,
+      "grad_norm": 0.06751185655593872,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 58356
+    },
+    {
+      "epoch": 5.99352983465133,
+      "grad_norm": 0.07169979065656662,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 58359
+    },
+    {
+      "epoch": 5.993837937763171,
+      "grad_norm": 0.13775870203971863,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 58362
+    },
+    {
+      "epoch": 5.9941460408750125,
+      "grad_norm": 0.08033827692270279,
+      "learning_rate": 0.01,
+      "loss": 2.0378,
+      "step": 58365
+    },
+    {
+      "epoch": 5.994454143986855,
+      "grad_norm": 0.07793736457824707,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 58368
+    },
+    {
+      "epoch": 5.994762247098696,
+      "grad_norm": 0.06328029185533524,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 58371
+    },
+    {
+      "epoch": 5.995070350210537,
+      "grad_norm": 0.05584852769970894,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 58374
+    },
+    {
+      "epoch": 5.995378453322378,
+      "grad_norm": 0.04354814067482948,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 58377
+    },
+    {
+      "epoch": 5.99568655643422,
+      "grad_norm": 0.05028558522462845,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 58380
+    },
+    {
+      "epoch": 5.995994659546062,
+      "grad_norm": 0.053422022610902786,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 58383
+    },
+    {
+      "epoch": 5.996302762657903,
+      "grad_norm": 0.03941678628325462,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 58386
+    },
+    {
+      "epoch": 5.996610865769744,
+      "grad_norm": 0.03764568641781807,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 58389
+    },
+    {
+      "epoch": 5.996918968881586,
+      "grad_norm": 0.044926904141902924,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 58392
+    },
+    {
+      "epoch": 5.997227071993427,
+      "grad_norm": 0.046521514654159546,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 58395
+    },
+    {
+      "epoch": 5.997535175105268,
+      "grad_norm": 0.12329057604074478,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 58398
+    },
+    {
+      "epoch": 5.99784327821711,
+      "grad_norm": 0.045972906053066254,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 58401
+    },
+    {
+      "epoch": 5.9981513813289515,
+      "grad_norm": 0.1293979287147522,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 58404
+    },
+    {
+      "epoch": 5.998459484440793,
+      "grad_norm": 0.06918042153120041,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 58407
+    },
+    {
+      "epoch": 5.998767587552634,
+      "grad_norm": 0.04279787465929985,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 58410
+    },
+    {
+      "epoch": 5.999075690664475,
+      "grad_norm": 0.09146596491336823,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 58413
+    },
+    {
+      "epoch": 5.9993837937763175,
+      "grad_norm": 0.06269899010658264,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 58416
+    },
+    {
+      "epoch": 5.999691896888159,
+      "grad_norm": 0.04407677426934242,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 58419
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.04100476950407028,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 58422
+    },
+    {
+      "epoch": 5.999691928527419,
+      "grad_norm": 0.05454897880554199,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 58425
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.12488140910863876,
+      "learning_rate": 0.01,
+      "loss": 2.0163,
+      "step": 58428
+    },
+    {
+      "epoch": 6.000308071472582,
+      "grad_norm": 0.08063201606273651,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 58431
+    },
+    {
+      "epoch": 6.000616142945163,
+      "grad_norm": 0.08693194389343262,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 58434
+    },
+    {
+      "epoch": 6.000924214417745,
+      "grad_norm": 0.048637568950653076,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 58437
+    },
+    {
+      "epoch": 6.001232285890326,
+      "grad_norm": 0.047587309032678604,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 58440
+    },
+    {
+      "epoch": 6.0015403573629085,
+      "grad_norm": 0.04123762995004654,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 58443
+    },
+    {
+      "epoch": 6.00184842883549,
+      "grad_norm": 0.038054220378398895,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 58446
+    },
+    {
+      "epoch": 6.002156500308072,
+      "grad_norm": 0.07207085937261581,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 58449
+    },
+    {
+      "epoch": 6.002464571780653,
+      "grad_norm": 0.09471093118190765,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 58452
+    },
+    {
+      "epoch": 6.002772643253235,
+      "grad_norm": 0.050923608243465424,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 58455
+    },
+    {
+      "epoch": 6.003080714725816,
+      "grad_norm": 0.09074869751930237,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 58458
+    },
+    {
+      "epoch": 6.003388786198398,
+      "grad_norm": 0.12536507844924927,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 58461
+    },
+    {
+      "epoch": 6.003696857670979,
+      "grad_norm": 0.04264422506093979,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 58464
+    },
+    {
+      "epoch": 6.004004929143561,
+      "grad_norm": 0.03600556403398514,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 58467
+    },
+    {
+      "epoch": 6.004313000616143,
+      "grad_norm": 0.030723579227924347,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 58470
+    },
+    {
+      "epoch": 6.0046210720887245,
+      "grad_norm": 0.042884331196546555,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 58473
+    },
+    {
+      "epoch": 6.004929143561307,
+      "grad_norm": 0.07880793511867523,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 58476
+    },
+    {
+      "epoch": 6.005237215033888,
+      "grad_norm": 0.08945189416408539,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 58479
+    },
+    {
+      "epoch": 6.00554528650647,
+      "grad_norm": 0.07887895405292511,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 58482
+    },
+    {
+      "epoch": 6.005853357979051,
+      "grad_norm": 0.13316544890403748,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 58485
+    },
+    {
+      "epoch": 6.006161429451633,
+      "grad_norm": 0.06402456760406494,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 58488
+    },
+    {
+      "epoch": 6.006469500924214,
+      "grad_norm": 0.03985295444726944,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 58491
+    },
+    {
+      "epoch": 6.006777572396796,
+      "grad_norm": 0.03584379702806473,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 58494
+    },
+    {
+      "epoch": 6.007085643869377,
+      "grad_norm": 0.03731578588485718,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 58497
+    },
+    {
+      "epoch": 6.007393715341959,
+      "grad_norm": 0.04280545935034752,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 58500
+    },
+    {
+      "epoch": 6.007701786814541,
+      "grad_norm": 0.037182942032814026,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 58503
+    },
+    {
+      "epoch": 6.008009858287123,
+      "grad_norm": 0.04180056229233742,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 58506
+    },
+    {
+      "epoch": 6.008317929759705,
+      "grad_norm": 0.05659356340765953,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 58509
+    },
+    {
+      "epoch": 6.008626001232286,
+      "grad_norm": 0.04323246702551842,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 58512
+    },
+    {
+      "epoch": 6.008934072704868,
+      "grad_norm": 0.15815724432468414,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 58515
+    },
+    {
+      "epoch": 6.009242144177449,
+      "grad_norm": 0.041637711226940155,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 58518
+    },
+    {
+      "epoch": 6.009550215650031,
+      "grad_norm": 0.09015413373708725,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 58521
+    },
+    {
+      "epoch": 6.009858287122612,
+      "grad_norm": 0.10606445372104645,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 58524
+    },
+    {
+      "epoch": 6.010166358595194,
+      "grad_norm": 0.06839009374380112,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 58527
+    },
+    {
+      "epoch": 6.0104744300677755,
+      "grad_norm": 0.051956940442323685,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 58530
+    },
+    {
+      "epoch": 6.0107825015403575,
+      "grad_norm": 0.034211624413728714,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 58533
+    },
+    {
+      "epoch": 6.011090573012939,
+      "grad_norm": 0.04351538419723511,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 58536
+    },
+    {
+      "epoch": 6.011398644485521,
+      "grad_norm": 0.07221609354019165,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 58539
+    },
+    {
+      "epoch": 6.011706715958102,
+      "grad_norm": 0.11168903857469559,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 58542
+    },
+    {
+      "epoch": 6.012014787430684,
+      "grad_norm": 0.06332577764987946,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 58545
+    },
+    {
+      "epoch": 6.012322858903265,
+      "grad_norm": 0.0567488819360733,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 58548
+    },
+    {
+      "epoch": 6.012630930375847,
+      "grad_norm": 0.04636390134692192,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 58551
+    },
+    {
+      "epoch": 6.012939001848429,
+      "grad_norm": 0.040273915976285934,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 58554
+    },
+    {
+      "epoch": 6.01324707332101,
+      "grad_norm": 0.09343009442090988,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 58557
+    },
+    {
+      "epoch": 6.013555144793592,
+      "grad_norm": 0.07643181830644608,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 58560
+    },
+    {
+      "epoch": 6.013863216266174,
+      "grad_norm": 0.1327466368675232,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 58563
+    },
+    {
+      "epoch": 6.014171287738756,
+      "grad_norm": 0.1656189262866974,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 58566
+    },
+    {
+      "epoch": 6.014479359211337,
+      "grad_norm": 0.07180788367986679,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 58569
+    },
+    {
+      "epoch": 6.014787430683919,
+      "grad_norm": 0.04216361045837402,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 58572
+    },
+    {
+      "epoch": 6.0150955021565,
+      "grad_norm": 0.04374970495700836,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 58575
+    },
+    {
+      "epoch": 6.015403573629082,
+      "grad_norm": 0.038325972855091095,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 58578
+    },
+    {
+      "epoch": 6.015711645101663,
+      "grad_norm": 0.03523325175046921,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 58581
+    },
+    {
+      "epoch": 6.016019716574245,
+      "grad_norm": 0.03328186646103859,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 58584
+    },
+    {
+      "epoch": 6.016327788046826,
+      "grad_norm": 0.09052954614162445,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 58587
+    },
+    {
+      "epoch": 6.0166358595194085,
+      "grad_norm": 0.08605097979307175,
+      "learning_rate": 0.01,
+      "loss": 2.0121,
+      "step": 58590
+    },
+    {
+      "epoch": 6.0169439309919905,
+      "grad_norm": 0.05200384184718132,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 58593
+    },
+    {
+      "epoch": 6.017252002464572,
+      "grad_norm": 0.07615986466407776,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 58596
+    },
+    {
+      "epoch": 6.017560073937154,
+      "grad_norm": 0.043676577508449554,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 58599
+    },
+    {
+      "epoch": 6.017868145409735,
+      "grad_norm": 0.058520250022411346,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 58602
+    },
+    {
+      "epoch": 6.018176216882317,
+      "grad_norm": 0.07133082300424576,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 58605
+    },
+    {
+      "epoch": 6.018484288354898,
+      "grad_norm": 0.08711257576942444,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 58608
+    },
+    {
+      "epoch": 6.01879235982748,
+      "grad_norm": 0.0483718067407608,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 58611
+    },
+    {
+      "epoch": 6.019100431300061,
+      "grad_norm": 0.05014586076140404,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 58614
+    },
+    {
+      "epoch": 6.019408502772643,
+      "grad_norm": 0.10905557870864868,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 58617
+    },
+    {
+      "epoch": 6.0197165742452245,
+      "grad_norm": 0.03817577287554741,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 58620
+    },
+    {
+      "epoch": 6.020024645717807,
+      "grad_norm": 0.0783611536026001,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 58623
+    },
+    {
+      "epoch": 6.020332717190388,
+      "grad_norm": 0.0832376778125763,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 58626
+    },
+    {
+      "epoch": 6.02064078866297,
+      "grad_norm": 0.05190072953701019,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 58629
+    },
+    {
+      "epoch": 6.020948860135552,
+      "grad_norm": 0.09829907864332199,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 58632
+    },
+    {
+      "epoch": 6.021256931608133,
+      "grad_norm": 0.08857165277004242,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 58635
+    },
+    {
+      "epoch": 6.021565003080715,
+      "grad_norm": 0.07858851552009583,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 58638
+    },
+    {
+      "epoch": 6.021873074553296,
+      "grad_norm": 0.04164360463619232,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 58641
+    },
+    {
+      "epoch": 6.022181146025878,
+      "grad_norm": 0.037627752870321274,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 58644
+    },
+    {
+      "epoch": 6.0224892174984594,
+      "grad_norm": 0.03442845493555069,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 58647
+    },
+    {
+      "epoch": 6.0227972889710415,
+      "grad_norm": 0.04693920910358429,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 58650
+    },
+    {
+      "epoch": 6.023105360443623,
+      "grad_norm": 0.07563184201717377,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 58653
+    },
+    {
+      "epoch": 6.023413431916205,
+      "grad_norm": 0.08100397139787674,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 58656
+    },
+    {
+      "epoch": 6.023721503388786,
+      "grad_norm": 0.12132440507411957,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 58659
+    },
+    {
+      "epoch": 6.024029574861368,
+      "grad_norm": 0.04983309283852577,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 58662
+    },
+    {
+      "epoch": 6.024337646333949,
+      "grad_norm": 0.06472596526145935,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 58665
+    },
+    {
+      "epoch": 6.024645717806531,
+      "grad_norm": 0.034389570355415344,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 58668
+    },
+    {
+      "epoch": 6.024953789279113,
+      "grad_norm": 0.10626552999019623,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 58671
+    },
+    {
+      "epoch": 6.025261860751694,
+      "grad_norm": 0.0675535798072815,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 58674
+    },
+    {
+      "epoch": 6.025569932224276,
+      "grad_norm": 0.04010651633143425,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 58677
+    },
+    {
+      "epoch": 6.0258780036968576,
+      "grad_norm": 0.034451622515916824,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 58680
+    },
+    {
+      "epoch": 6.02618607516944,
+      "grad_norm": 0.06580517441034317,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 58683
+    },
+    {
+      "epoch": 6.026494146642021,
+      "grad_norm": 0.05853043496608734,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 58686
+    },
+    {
+      "epoch": 6.026802218114603,
+      "grad_norm": 0.04377220198512077,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 58689
+    },
+    {
+      "epoch": 6.027110289587184,
+      "grad_norm": 0.03288479149341583,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 58692
+    },
+    {
+      "epoch": 6.027418361059766,
+      "grad_norm": 0.04060712829232216,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 58695
+    },
+    {
+      "epoch": 6.027726432532347,
+      "grad_norm": 0.11332166939973831,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 58698
+    },
+    {
+      "epoch": 6.028034504004929,
+      "grad_norm": 0.13282176852226257,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 58701
+    },
+    {
+      "epoch": 6.02834257547751,
+      "grad_norm": 0.08038832992315292,
+      "learning_rate": 0.01,
+      "loss": 2.0108,
+      "step": 58704
+    },
+    {
+      "epoch": 6.0286506469500925,
+      "grad_norm": 0.05667036771774292,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 58707
+    },
+    {
+      "epoch": 6.0289587184226745,
+      "grad_norm": 0.06444176286458969,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 58710
+    },
+    {
+      "epoch": 6.029266789895256,
+      "grad_norm": 0.08207248896360397,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 58713
+    },
+    {
+      "epoch": 6.029574861367838,
+      "grad_norm": 0.057919155806303024,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 58716
+    },
+    {
+      "epoch": 6.029882932840419,
+      "grad_norm": 0.0630873367190361,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 58719
+    },
+    {
+      "epoch": 6.030191004313001,
+      "grad_norm": 0.14096465706825256,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 58722
+    },
+    {
+      "epoch": 6.030499075785582,
+      "grad_norm": 0.06392169743776321,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 58725
+    },
+    {
+      "epoch": 6.030807147258164,
+      "grad_norm": 0.051594968885183334,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 58728
+    },
+    {
+      "epoch": 6.031115218730745,
+      "grad_norm": 0.05645729973912239,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 58731
+    },
+    {
+      "epoch": 6.031423290203327,
+      "grad_norm": 0.056138813495635986,
+      "learning_rate": 0.01,
+      "loss": 2.016,
+      "step": 58734
+    },
+    {
+      "epoch": 6.0317313616759085,
+      "grad_norm": 0.08171719312667847,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 58737
+    },
+    {
+      "epoch": 6.032039433148491,
+      "grad_norm": 0.0954081118106842,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 58740
+    },
+    {
+      "epoch": 6.032347504621072,
+      "grad_norm": 0.09134764224290848,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 58743
+    },
+    {
+      "epoch": 6.032655576093654,
+      "grad_norm": 0.05880044028162956,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 58746
+    },
+    {
+      "epoch": 6.032963647566235,
+      "grad_norm": 0.06801950186491013,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 58749
+    },
+    {
+      "epoch": 6.033271719038817,
+      "grad_norm": 0.038541264832019806,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 58752
+    },
+    {
+      "epoch": 6.033579790511399,
+      "grad_norm": 0.03735314682126045,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 58755
+    },
+    {
+      "epoch": 6.03388786198398,
+      "grad_norm": 0.10068950057029724,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 58758
+    },
+    {
+      "epoch": 6.034195933456562,
+      "grad_norm": 0.04904381185770035,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 58761
+    },
+    {
+      "epoch": 6.034504004929143,
+      "grad_norm": 0.05660933256149292,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 58764
+    },
+    {
+      "epoch": 6.0348120764017255,
+      "grad_norm": 0.0632736012339592,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 58767
+    },
+    {
+      "epoch": 6.035120147874307,
+      "grad_norm": 0.06228591129183769,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 58770
+    },
+    {
+      "epoch": 6.035428219346889,
+      "grad_norm": 0.03460165485739708,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 58773
+    },
+    {
+      "epoch": 6.03573629081947,
+      "grad_norm": 0.04163900017738342,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 58776
+    },
+    {
+      "epoch": 6.036044362292052,
+      "grad_norm": 0.07276707142591476,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 58779
+    },
+    {
+      "epoch": 6.036352433764633,
+      "grad_norm": 0.07738452404737473,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 58782
+    },
+    {
+      "epoch": 6.036660505237215,
+      "grad_norm": 0.1065252274274826,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 58785
+    },
+    {
+      "epoch": 6.036968576709796,
+      "grad_norm": 0.05357460305094719,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 58788
+    },
+    {
+      "epoch": 6.037276648182378,
+      "grad_norm": 0.03857807815074921,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 58791
+    },
+    {
+      "epoch": 6.03758471965496,
+      "grad_norm": 0.029968131333589554,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 58794
+    },
+    {
+      "epoch": 6.0378927911275415,
+      "grad_norm": 0.0720914974808693,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 58797
+    },
+    {
+      "epoch": 6.038200862600124,
+      "grad_norm": 0.12587358057498932,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 58800
+    },
+    {
+      "epoch": 6.038508934072705,
+      "grad_norm": 0.07049199193716049,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 58803
+    },
+    {
+      "epoch": 6.038817005545287,
+      "grad_norm": 0.05485573783516884,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 58806
+    },
+    {
+      "epoch": 6.039125077017868,
+      "grad_norm": 0.05481892451643944,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 58809
+    },
+    {
+      "epoch": 6.03943314849045,
+      "grad_norm": 0.07906284928321838,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 58812
+    },
+    {
+      "epoch": 6.039741219963031,
+      "grad_norm": 0.10519934445619583,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 58815
+    },
+    {
+      "epoch": 6.040049291435613,
+      "grad_norm": 0.06294261664152145,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 58818
+    },
+    {
+      "epoch": 6.040357362908194,
+      "grad_norm": 0.0904410257935524,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 58821
+    },
+    {
+      "epoch": 6.040665434380776,
+      "grad_norm": 0.053544074296951294,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 58824
+    },
+    {
+      "epoch": 6.040973505853358,
+      "grad_norm": 0.07926991581916809,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 58827
+    },
+    {
+      "epoch": 6.04128157732594,
+      "grad_norm": 0.09197907149791718,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 58830
+    },
+    {
+      "epoch": 6.041589648798522,
+      "grad_norm": 0.05957088991999626,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 58833
+    },
+    {
+      "epoch": 6.041897720271103,
+      "grad_norm": 0.1431286334991455,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 58836
+    },
+    {
+      "epoch": 6.042205791743685,
+      "grad_norm": 0.06428467482328415,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 58839
+    },
+    {
+      "epoch": 6.042513863216266,
+      "grad_norm": 0.07960224896669388,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 58842
+    },
+    {
+      "epoch": 6.042821934688848,
+      "grad_norm": 0.040846168994903564,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 58845
+    },
+    {
+      "epoch": 6.043130006161429,
+      "grad_norm": 0.04430864006280899,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 58848
+    },
+    {
+      "epoch": 6.043438077634011,
+      "grad_norm": 0.04534313082695007,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 58851
+    },
+    {
+      "epoch": 6.0437461491065925,
+      "grad_norm": 0.04214495047926903,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 58854
+    },
+    {
+      "epoch": 6.0440542205791745,
+      "grad_norm": 0.036912716925144196,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 58857
+    },
+    {
+      "epoch": 6.044362292051756,
+      "grad_norm": 0.05208496004343033,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 58860
+    },
+    {
+      "epoch": 6.044670363524338,
+      "grad_norm": 0.10304973274469376,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 58863
+    },
+    {
+      "epoch": 6.044978434996919,
+      "grad_norm": 0.14296580851078033,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 58866
+    },
+    {
+      "epoch": 6.045286506469501,
+      "grad_norm": 0.037938348948955536,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 58869
+    },
+    {
+      "epoch": 6.045594577942083,
+      "grad_norm": 0.054850075393915176,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 58872
+    },
+    {
+      "epoch": 6.045902649414664,
+      "grad_norm": 0.0785508081316948,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 58875
+    },
+    {
+      "epoch": 6.046210720887246,
+      "grad_norm": 0.07267174124717712,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 58878
+    },
+    {
+      "epoch": 6.046518792359827,
+      "grad_norm": 0.04367669299244881,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 58881
+    },
+    {
+      "epoch": 6.046826863832409,
+      "grad_norm": 0.046264924108982086,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 58884
+    },
+    {
+      "epoch": 6.047134935304991,
+      "grad_norm": 0.08904755860567093,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 58887
+    },
+    {
+      "epoch": 6.047443006777573,
+      "grad_norm": 0.06151941046118736,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 58890
+    },
+    {
+      "epoch": 6.047751078250154,
+      "grad_norm": 0.09928801655769348,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 58893
+    },
+    {
+      "epoch": 6.048059149722736,
+      "grad_norm": 0.04402837157249451,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 58896
+    },
+    {
+      "epoch": 6.048367221195317,
+      "grad_norm": 0.08484535664319992,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 58899
+    },
+    {
+      "epoch": 6.048675292667899,
+      "grad_norm": 0.07044728100299835,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 58902
+    },
+    {
+      "epoch": 6.04898336414048,
+      "grad_norm": 0.05778975039720535,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 58905
+    },
+    {
+      "epoch": 6.049291435613062,
+      "grad_norm": 0.08870556205511093,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 58908
+    },
+    {
+      "epoch": 6.049599507085643,
+      "grad_norm": 0.06904757022857666,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 58911
+    },
+    {
+      "epoch": 6.0499075785582255,
+      "grad_norm": 0.07772015035152435,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 58914
+    },
+    {
+      "epoch": 6.0502156500308075,
+      "grad_norm": 0.05567057430744171,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 58917
+    },
+    {
+      "epoch": 6.050523721503389,
+      "grad_norm": 0.06966744363307953,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 58920
+    },
+    {
+      "epoch": 6.050831792975971,
+      "grad_norm": 0.05111026018857956,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 58923
+    },
+    {
+      "epoch": 6.051139864448552,
+      "grad_norm": 0.07328280061483383,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 58926
+    },
+    {
+      "epoch": 6.051447935921134,
+      "grad_norm": 0.06918063014745712,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 58929
+    },
+    {
+      "epoch": 6.051756007393715,
+      "grad_norm": 0.09920765459537506,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 58932
+    },
+    {
+      "epoch": 6.052064078866297,
+      "grad_norm": 0.057603057473897934,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 58935
+    },
+    {
+      "epoch": 6.052372150338878,
+      "grad_norm": 0.0861181914806366,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 58938
+    },
+    {
+      "epoch": 6.05268022181146,
+      "grad_norm": 0.1477275937795639,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 58941
+    },
+    {
+      "epoch": 6.0529882932840415,
+      "grad_norm": 0.04588627070188522,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 58944
+    },
+    {
+      "epoch": 6.053296364756624,
+      "grad_norm": 0.09523765742778778,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 58947
+    },
+    {
+      "epoch": 6.053604436229205,
+      "grad_norm": 0.05901789665222168,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 58950
+    },
+    {
+      "epoch": 6.053912507701787,
+      "grad_norm": 0.0464630164206028,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 58953
+    },
+    {
+      "epoch": 6.054220579174369,
+      "grad_norm": 0.06684694439172745,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 58956
+    },
+    {
+      "epoch": 6.05452865064695,
+      "grad_norm": 0.062367189675569534,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 58959
+    },
+    {
+      "epoch": 6.054836722119532,
+      "grad_norm": 0.050568412989377975,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 58962
+    },
+    {
+      "epoch": 6.055144793592113,
+      "grad_norm": 0.05513317883014679,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 58965
+    },
+    {
+      "epoch": 6.055452865064695,
+      "grad_norm": 0.05048702657222748,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 58968
+    },
+    {
+      "epoch": 6.055760936537276,
+      "grad_norm": 0.04989680275321007,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 58971
+    },
+    {
+      "epoch": 6.0560690080098585,
+      "grad_norm": 0.03208424523472786,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 58974
+    },
+    {
+      "epoch": 6.05637707948244,
+      "grad_norm": 0.11022038012742996,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 58977
+    },
+    {
+      "epoch": 6.056685150955022,
+      "grad_norm": 0.10186657309532166,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 58980
+    },
+    {
+      "epoch": 6.056993222427603,
+      "grad_norm": 0.04729219898581505,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 58983
+    },
+    {
+      "epoch": 6.057301293900185,
+      "grad_norm": 0.08434822410345078,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 58986
+    },
+    {
+      "epoch": 6.057609365372766,
+      "grad_norm": 0.05614599958062172,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 58989
+    },
+    {
+      "epoch": 6.057917436845348,
+      "grad_norm": 0.06520257890224457,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 58992
+    },
+    {
+      "epoch": 6.05822550831793,
+      "grad_norm": 0.035195183008909225,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 58995
+    },
+    {
+      "epoch": 6.058533579790511,
+      "grad_norm": 0.08378977328538895,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 58998
+    },
+    {
+      "epoch": 6.058841651263093,
+      "grad_norm": 0.05082874745130539,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 59001
+    },
+    {
+      "epoch": 6.0591497227356745,
+      "grad_norm": 0.10246657580137253,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 59004
+    },
+    {
+      "epoch": 6.059457794208257,
+      "grad_norm": 0.0681087076663971,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 59007
+    },
+    {
+      "epoch": 6.059765865680838,
+      "grad_norm": 0.10320194065570831,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 59010
+    },
+    {
+      "epoch": 6.06007393715342,
+      "grad_norm": 0.04554738476872444,
+      "learning_rate": 0.01,
+      "loss": 2.0199,
+      "step": 59013
+    },
+    {
+      "epoch": 6.060382008626001,
+      "grad_norm": 0.11251513659954071,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 59016
+    },
+    {
+      "epoch": 6.060690080098583,
+      "grad_norm": 0.042101431638002396,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 59019
+    },
+    {
+      "epoch": 6.060998151571164,
+      "grad_norm": 0.05242328718304634,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 59022
+    },
+    {
+      "epoch": 6.061306223043746,
+      "grad_norm": 0.07696748524904251,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 59025
+    },
+    {
+      "epoch": 6.061614294516327,
+      "grad_norm": 0.07448645681142807,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 59028
+    },
+    {
+      "epoch": 6.061922365988909,
+      "grad_norm": 0.05332387611269951,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 59031
+    },
+    {
+      "epoch": 6.0622304374614915,
+      "grad_norm": 0.046122957020998,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 59034
+    },
+    {
+      "epoch": 6.062538508934073,
+      "grad_norm": 0.06084294244647026,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 59037
+    },
+    {
+      "epoch": 6.062846580406655,
+      "grad_norm": 0.09956135600805283,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 59040
+    },
+    {
+      "epoch": 6.063154651879236,
+      "grad_norm": 0.05223178490996361,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 59043
+    },
+    {
+      "epoch": 6.063462723351818,
+      "grad_norm": 0.08285564929246902,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 59046
+    },
+    {
+      "epoch": 6.063770794824399,
+      "grad_norm": 0.05346845090389252,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 59049
+    },
+    {
+      "epoch": 6.064078866296981,
+      "grad_norm": 0.08269158750772476,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 59052
+    },
+    {
+      "epoch": 6.064386937769562,
+      "grad_norm": 0.07285866141319275,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 59055
+    },
+    {
+      "epoch": 6.064695009242144,
+      "grad_norm": 0.07736973464488983,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 59058
+    },
+    {
+      "epoch": 6.0650030807147255,
+      "grad_norm": 0.0628119483590126,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 59061
+    },
+    {
+      "epoch": 6.0653111521873075,
+      "grad_norm": 0.044494885951280594,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 59064
+    },
+    {
+      "epoch": 6.065619223659889,
+      "grad_norm": 0.1270086020231247,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 59067
+    },
+    {
+      "epoch": 6.065927295132471,
+      "grad_norm": 0.045628707855939865,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 59070
+    },
+    {
+      "epoch": 6.066235366605053,
+      "grad_norm": 0.07165969908237457,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 59073
+    },
+    {
+      "epoch": 6.066543438077634,
+      "grad_norm": 0.047646794468164444,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 59076
+    },
+    {
+      "epoch": 6.066851509550216,
+      "grad_norm": 0.05064648762345314,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 59079
+    },
+    {
+      "epoch": 6.067159581022797,
+      "grad_norm": 0.07075773924589157,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 59082
+    },
+    {
+      "epoch": 6.067467652495379,
+      "grad_norm": 0.0734126940369606,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 59085
+    },
+    {
+      "epoch": 6.06777572396796,
+      "grad_norm": 0.07513909041881561,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 59088
+    },
+    {
+      "epoch": 6.068083795440542,
+      "grad_norm": 0.04252862557768822,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 59091
+    },
+    {
+      "epoch": 6.068391866913124,
+      "grad_norm": 0.06020990386605263,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 59094
+    },
+    {
+      "epoch": 6.068699938385706,
+      "grad_norm": 0.048400718718767166,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 59097
+    },
+    {
+      "epoch": 6.069008009858287,
+      "grad_norm": 0.10299364477396011,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 59100
+    },
+    {
+      "epoch": 6.069316081330869,
+      "grad_norm": 0.07474274933338165,
+      "learning_rate": 0.01,
+      "loss": 2.0238,
+      "step": 59103
+    },
+    {
+      "epoch": 6.06962415280345,
+      "grad_norm": 0.069256491959095,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 59106
+    },
+    {
+      "epoch": 6.069932224276032,
+      "grad_norm": 0.07016000151634216,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 59109
+    },
+    {
+      "epoch": 6.070240295748613,
+      "grad_norm": 0.10577663034200668,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 59112
+    },
+    {
+      "epoch": 6.070548367221195,
+      "grad_norm": 0.10170361399650574,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 59115
+    },
+    {
+      "epoch": 6.070856438693777,
+      "grad_norm": 0.03669926896691322,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 59118
+    },
+    {
+      "epoch": 6.0711645101663585,
+      "grad_norm": 0.04145694151520729,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 59121
+    },
+    {
+      "epoch": 6.0714725816389405,
+      "grad_norm": 0.052021101117134094,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 59124
+    },
+    {
+      "epoch": 6.071780653111522,
+      "grad_norm": 0.10580527782440186,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 59127
+    },
+    {
+      "epoch": 6.072088724584104,
+      "grad_norm": 0.05877790227532387,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 59130
+    },
+    {
+      "epoch": 6.072396796056685,
+      "grad_norm": 0.047968316823244095,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 59133
+    },
+    {
+      "epoch": 6.072704867529267,
+      "grad_norm": 0.05399814620614052,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 59136
+    },
+    {
+      "epoch": 6.073012939001848,
+      "grad_norm": 0.0343160405755043,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 59139
+    },
+    {
+      "epoch": 6.07332101047443,
+      "grad_norm": 0.032716646790504456,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 59142
+    },
+    {
+      "epoch": 6.073629081947011,
+      "grad_norm": 0.09958682209253311,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 59145
+    },
+    {
+      "epoch": 6.073937153419593,
+      "grad_norm": 0.14549368619918823,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 59148
+    },
+    {
+      "epoch": 6.0742452248921746,
+      "grad_norm": 0.12808917462825775,
+      "learning_rate": 0.01,
+      "loss": 2.0228,
+      "step": 59151
+    },
+    {
+      "epoch": 6.074553296364757,
+      "grad_norm": 0.06274955719709396,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 59154
+    },
+    {
+      "epoch": 6.074861367837339,
+      "grad_norm": 0.06658618897199631,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 59157
+    },
+    {
+      "epoch": 6.07516943930992,
+      "grad_norm": 0.07997333258390427,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 59160
+    },
+    {
+      "epoch": 6.075477510782502,
+      "grad_norm": 0.040492795407772064,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 59163
+    },
+    {
+      "epoch": 6.075785582255083,
+      "grad_norm": 0.03949406370520592,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 59166
+    },
+    {
+      "epoch": 6.076093653727665,
+      "grad_norm": 0.0330752469599247,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 59169
+    },
+    {
+      "epoch": 6.076401725200246,
+      "grad_norm": 0.03734998777508736,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 59172
+    },
+    {
+      "epoch": 6.076709796672828,
+      "grad_norm": 0.04145140200853348,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 59175
+    },
+    {
+      "epoch": 6.0770178681454095,
+      "grad_norm": 0.04934060946106911,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 59178
+    },
+    {
+      "epoch": 6.0773259396179915,
+      "grad_norm": 0.09385284036397934,
+      "learning_rate": 0.01,
+      "loss": 2.0134,
+      "step": 59181
+    },
+    {
+      "epoch": 6.077634011090573,
+      "grad_norm": 0.14405569434165955,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 59184
+    },
+    {
+      "epoch": 6.077942082563155,
+      "grad_norm": 0.12107555568218231,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 59187
+    },
+    {
+      "epoch": 6.078250154035736,
+      "grad_norm": 0.05842713639140129,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 59190
+    },
+    {
+      "epoch": 6.078558225508318,
+      "grad_norm": 0.049650490283966064,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 59193
+    },
+    {
+      "epoch": 6.0788662969809,
+      "grad_norm": 0.06532774120569229,
+      "learning_rate": 0.01,
+      "loss": 2.0231,
+      "step": 59196
+    },
+    {
+      "epoch": 6.079174368453481,
+      "grad_norm": 0.03809603303670883,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 59199
+    },
+    {
+      "epoch": 6.079482439926063,
+      "grad_norm": 0.07845079898834229,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 59202
+    },
+    {
+      "epoch": 6.079790511398644,
+      "grad_norm": 0.15672674775123596,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 59205
+    },
+    {
+      "epoch": 6.080098582871226,
+      "grad_norm": 0.11329050362110138,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 59208
+    },
+    {
+      "epoch": 6.080406654343808,
+      "grad_norm": 0.08986864238977432,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 59211
+    },
+    {
+      "epoch": 6.08071472581639,
+      "grad_norm": 0.040678489953279495,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 59214
+    },
+    {
+      "epoch": 6.081022797288971,
+      "grad_norm": 0.09467331320047379,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 59217
+    },
+    {
+      "epoch": 6.081330868761553,
+      "grad_norm": 0.0797513797879219,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 59220
+    },
+    {
+      "epoch": 6.081638940234134,
+      "grad_norm": 0.04497670754790306,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 59223
+    },
+    {
+      "epoch": 6.081947011706716,
+      "grad_norm": 0.08867593109607697,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 59226
+    },
+    {
+      "epoch": 6.082255083179297,
+      "grad_norm": 0.1013980433344841,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 59229
+    },
+    {
+      "epoch": 6.082563154651879,
+      "grad_norm": 0.05733179301023483,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 59232
+    },
+    {
+      "epoch": 6.082871226124461,
+      "grad_norm": 0.09696310758590698,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 59235
+    },
+    {
+      "epoch": 6.0831792975970425,
+      "grad_norm": 0.07615906745195389,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 59238
+    },
+    {
+      "epoch": 6.0834873690696245,
+      "grad_norm": 0.08034536242485046,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 59241
+    },
+    {
+      "epoch": 6.083795440542206,
+      "grad_norm": 0.04975961148738861,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 59244
+    },
+    {
+      "epoch": 6.084103512014788,
+      "grad_norm": 0.09680989384651184,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 59247
+    },
+    {
+      "epoch": 6.084411583487369,
+      "grad_norm": 0.06167134270071983,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 59250
+    },
+    {
+      "epoch": 6.084719654959951,
+      "grad_norm": 0.08627012372016907,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 59253
+    },
+    {
+      "epoch": 6.085027726432532,
+      "grad_norm": 0.053778212517499924,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 59256
+    },
+    {
+      "epoch": 6.085335797905114,
+      "grad_norm": 0.10851852595806122,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 59259
+    },
+    {
+      "epoch": 6.085643869377695,
+      "grad_norm": 0.08398763090372086,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 59262
+    },
+    {
+      "epoch": 6.085951940850277,
+      "grad_norm": 0.09030786901712418,
+      "learning_rate": 0.01,
+      "loss": 2.009,
+      "step": 59265
+    },
+    {
+      "epoch": 6.0862600123228585,
+      "grad_norm": 0.08578209578990936,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 59268
+    },
+    {
+      "epoch": 6.086568083795441,
+      "grad_norm": 0.06893924623727798,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 59271
+    },
+    {
+      "epoch": 6.086876155268023,
+      "grad_norm": 0.044449809938669205,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 59274
+    },
+    {
+      "epoch": 6.087184226740604,
+      "grad_norm": 0.03283555805683136,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 59277
+    },
+    {
+      "epoch": 6.087492298213186,
+      "grad_norm": 0.05007379502058029,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 59280
+    },
+    {
+      "epoch": 6.087800369685767,
+      "grad_norm": 0.05644026771187782,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 59283
+    },
+    {
+      "epoch": 6.088108441158349,
+      "grad_norm": 0.06911731511354446,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 59286
+    },
+    {
+      "epoch": 6.08841651263093,
+      "grad_norm": 0.06676710397005081,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 59289
+    },
+    {
+      "epoch": 6.088724584103512,
+      "grad_norm": 0.03128058835864067,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 59292
+    },
+    {
+      "epoch": 6.089032655576093,
+      "grad_norm": 0.05113793537020683,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 59295
+    },
+    {
+      "epoch": 6.0893407270486755,
+      "grad_norm": 0.10867103934288025,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 59298
+    },
+    {
+      "epoch": 6.089648798521257,
+      "grad_norm": 0.10439618676900864,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 59301
+    },
+    {
+      "epoch": 6.089956869993839,
+      "grad_norm": 0.06279850751161575,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 59304
+    },
+    {
+      "epoch": 6.09026494146642,
+      "grad_norm": 0.07557038217782974,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 59307
+    },
+    {
+      "epoch": 6.090573012939002,
+      "grad_norm": 0.06895217299461365,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 59310
+    },
+    {
+      "epoch": 6.090881084411583,
+      "grad_norm": 0.058836501091718674,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 59313
+    },
+    {
+      "epoch": 6.091189155884165,
+      "grad_norm": 0.10100754350423813,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 59316
+    },
+    {
+      "epoch": 6.091497227356747,
+      "grad_norm": 0.03705455735325813,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 59319
+    },
+    {
+      "epoch": 6.091805298829328,
+      "grad_norm": 0.09770546853542328,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 59322
+    },
+    {
+      "epoch": 6.09211337030191,
+      "grad_norm": 0.07368703931570053,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 59325
+    },
+    {
+      "epoch": 6.0924214417744915,
+      "grad_norm": 0.07088816910982132,
+      "learning_rate": 0.01,
+      "loss": 2.0124,
+      "step": 59328
+    },
+    {
+      "epoch": 6.092729513247074,
+      "grad_norm": 0.05165349692106247,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 59331
+    },
+    {
+      "epoch": 6.093037584719655,
+      "grad_norm": 0.08270641416311264,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 59334
+    },
+    {
+      "epoch": 6.093345656192237,
+      "grad_norm": 0.07178719341754913,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 59337
+    },
+    {
+      "epoch": 6.093653727664818,
+      "grad_norm": 0.07591982930898666,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 59340
+    },
+    {
+      "epoch": 6.0939617991374,
+      "grad_norm": 0.08362317830324173,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 59343
+    },
+    {
+      "epoch": 6.094269870609981,
+      "grad_norm": 0.04838939011096954,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 59346
+    },
+    {
+      "epoch": 6.094577942082563,
+      "grad_norm": 0.049368396401405334,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 59349
+    },
+    {
+      "epoch": 6.094886013555144,
+      "grad_norm": 0.05581490695476532,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 59352
+    },
+    {
+      "epoch": 6.095194085027726,
+      "grad_norm": 0.04965098947286606,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 59355
+    },
+    {
+      "epoch": 6.0955021565003085,
+      "grad_norm": 0.03643220290541649,
+      "learning_rate": 0.01,
+      "loss": 2.0234,
+      "step": 59358
+    },
+    {
+      "epoch": 6.09581022797289,
+      "grad_norm": 0.08060404658317566,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 59361
+    },
+    {
+      "epoch": 6.096118299445472,
+      "grad_norm": 0.08821488916873932,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 59364
+    },
+    {
+      "epoch": 6.096426370918053,
+      "grad_norm": 0.04556925222277641,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 59367
+    },
+    {
+      "epoch": 6.096734442390635,
+      "grad_norm": 0.07568595558404922,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 59370
+    },
+    {
+      "epoch": 6.097042513863216,
+      "grad_norm": 0.06441906094551086,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 59373
+    },
+    {
+      "epoch": 6.097350585335798,
+      "grad_norm": 0.04911012202501297,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 59376
+    },
+    {
+      "epoch": 6.097658656808379,
+      "grad_norm": 0.10038016736507416,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 59379
+    },
+    {
+      "epoch": 6.097966728280961,
+      "grad_norm": 0.08682964742183685,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 59382
+    },
+    {
+      "epoch": 6.0982747997535425,
+      "grad_norm": 0.07904212176799774,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 59385
+    },
+    {
+      "epoch": 6.0985828712261245,
+      "grad_norm": 0.07314693182706833,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 59388
+    },
+    {
+      "epoch": 6.098890942698706,
+      "grad_norm": 0.04866940528154373,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 59391
+    },
+    {
+      "epoch": 6.099199014171288,
+      "grad_norm": 0.040898289531469345,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 59394
+    },
+    {
+      "epoch": 6.09950708564387,
+      "grad_norm": 0.04597887024283409,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 59397
+    },
+    {
+      "epoch": 6.099815157116451,
+      "grad_norm": 0.04504725709557533,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 59400
+    },
+    {
+      "epoch": 6.100123228589033,
+      "grad_norm": 0.032154761254787445,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 59403
+    },
+    {
+      "epoch": 6.100431300061614,
+      "grad_norm": 0.06787566095590591,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 59406
+    },
+    {
+      "epoch": 6.100739371534196,
+      "grad_norm": 0.07191510498523712,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 59409
+    },
+    {
+      "epoch": 6.101047443006777,
+      "grad_norm": 0.08066849410533905,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 59412
+    },
+    {
+      "epoch": 6.101355514479359,
+      "grad_norm": 0.05546758696436882,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 59415
+    },
+    {
+      "epoch": 6.101663585951941,
+      "grad_norm": 0.059205375611782074,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 59418
+    },
+    {
+      "epoch": 6.101971657424523,
+      "grad_norm": 0.03468778356909752,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 59421
+    },
+    {
+      "epoch": 6.102279728897104,
+      "grad_norm": 0.06100203096866608,
+      "learning_rate": 0.01,
+      "loss": 2.0237,
+      "step": 59424
+    },
+    {
+      "epoch": 6.102587800369686,
+      "grad_norm": 0.04229642450809479,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 59427
+    },
+    {
+      "epoch": 6.102895871842267,
+      "grad_norm": 0.09364788234233856,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 59430
+    },
+    {
+      "epoch": 6.103203943314849,
+      "grad_norm": 0.0706750899553299,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 59433
+    },
+    {
+      "epoch": 6.103512014787431,
+      "grad_norm": 0.1148315817117691,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 59436
+    },
+    {
+      "epoch": 6.103820086260012,
+      "grad_norm": 0.057128921151161194,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 59439
+    },
+    {
+      "epoch": 6.104128157732594,
+      "grad_norm": 0.054126545786857605,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 59442
+    },
+    {
+      "epoch": 6.1044362292051755,
+      "grad_norm": 0.050927091389894485,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 59445
+    },
+    {
+      "epoch": 6.1047443006777575,
+      "grad_norm": 0.05484678968787193,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 59448
+    },
+    {
+      "epoch": 6.105052372150339,
+      "grad_norm": 0.06659942865371704,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 59451
+    },
+    {
+      "epoch": 6.105360443622921,
+      "grad_norm": 0.06512758880853653,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 59454
+    },
+    {
+      "epoch": 6.105668515095502,
+      "grad_norm": 0.04978008568286896,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 59457
+    },
+    {
+      "epoch": 6.105976586568084,
+      "grad_norm": 0.03648846223950386,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 59460
+    },
+    {
+      "epoch": 6.106284658040665,
+      "grad_norm": 0.049807704985141754,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 59463
+    },
+    {
+      "epoch": 6.106592729513247,
+      "grad_norm": 0.04505782201886177,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 59466
+    },
+    {
+      "epoch": 6.106900800985828,
+      "grad_norm": 0.05675986409187317,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 59469
+    },
+    {
+      "epoch": 6.10720887245841,
+      "grad_norm": 0.06363193690776825,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 59472
+    },
+    {
+      "epoch": 6.107516943930992,
+      "grad_norm": 0.038001857697963715,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 59475
+    },
+    {
+      "epoch": 6.107825015403574,
+      "grad_norm": 0.034627631306648254,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 59478
+    },
+    {
+      "epoch": 6.108133086876156,
+      "grad_norm": 0.0814884752035141,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 59481
+    },
+    {
+      "epoch": 6.108441158348737,
+      "grad_norm": 0.0862545445561409,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 59484
+    },
+    {
+      "epoch": 6.108749229821319,
+      "grad_norm": 0.05416722968220711,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 59487
+    },
+    {
+      "epoch": 6.1090573012939,
+      "grad_norm": 0.04082367196679115,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 59490
+    },
+    {
+      "epoch": 6.109365372766482,
+      "grad_norm": 0.0981779396533966,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 59493
+    },
+    {
+      "epoch": 6.109673444239063,
+      "grad_norm": 0.03428742289543152,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 59496
+    },
+    {
+      "epoch": 6.109981515711645,
+      "grad_norm": 0.07878245413303375,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 59499
+    },
+    {
+      "epoch": 6.110289587184226,
+      "grad_norm": 0.1018940880894661,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 59502
+    },
+    {
+      "epoch": 6.1105976586568085,
+      "grad_norm": 0.03978511691093445,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 59505
+    },
+    {
+      "epoch": 6.11090573012939,
+      "grad_norm": 0.05915730074048042,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 59508
+    },
+    {
+      "epoch": 6.111213801601972,
+      "grad_norm": 0.0455721952021122,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 59511
+    },
+    {
+      "epoch": 6.111521873074553,
+      "grad_norm": 0.050054021179676056,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 59514
+    },
+    {
+      "epoch": 6.111829944547135,
+      "grad_norm": 0.0369170643389225,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 59517
+    },
+    {
+      "epoch": 6.112138016019717,
+      "grad_norm": 0.044435255229473114,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 59520
+    },
+    {
+      "epoch": 6.112446087492298,
+      "grad_norm": 0.04968203604221344,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 59523
+    },
+    {
+      "epoch": 6.11275415896488,
+      "grad_norm": 0.04474148899316788,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 59526
+    },
+    {
+      "epoch": 6.113062230437461,
+      "grad_norm": 0.03932742401957512,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 59529
+    },
+    {
+      "epoch": 6.113370301910043,
+      "grad_norm": 0.10069919377565384,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 59532
+    },
+    {
+      "epoch": 6.1136783733826245,
+      "grad_norm": 0.08742933720350266,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 59535
+    },
+    {
+      "epoch": 6.113986444855207,
+      "grad_norm": 0.15486708283424377,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 59538
+    },
+    {
+      "epoch": 6.114294516327788,
+      "grad_norm": 0.05859851837158203,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 59541
+    },
+    {
+      "epoch": 6.11460258780037,
+      "grad_norm": 0.04623505100607872,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 59544
+    },
+    {
+      "epoch": 6.114910659272951,
+      "grad_norm": 0.037663958966732025,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 59547
+    },
+    {
+      "epoch": 6.115218730745533,
+      "grad_norm": 0.03428487107157707,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 59550
+    },
+    {
+      "epoch": 6.115526802218114,
+      "grad_norm": 0.03549535945057869,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 59553
+    },
+    {
+      "epoch": 6.115834873690696,
+      "grad_norm": 0.0357876755297184,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 59556
+    },
+    {
+      "epoch": 6.116142945163278,
+      "grad_norm": 0.05952336639165878,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 59559
+    },
+    {
+      "epoch": 6.116451016635859,
+      "grad_norm": 0.06661629676818848,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 59562
+    },
+    {
+      "epoch": 6.1167590881084415,
+      "grad_norm": 0.037326935678720474,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 59565
+    },
+    {
+      "epoch": 6.117067159581023,
+      "grad_norm": 0.08297394216060638,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 59568
+    },
+    {
+      "epoch": 6.117375231053605,
+      "grad_norm": 0.08954669535160065,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 59571
+    },
+    {
+      "epoch": 6.117683302526186,
+      "grad_norm": 0.06428781896829605,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 59574
+    },
+    {
+      "epoch": 6.117991373998768,
+      "grad_norm": 0.0963866114616394,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 59577
+    },
+    {
+      "epoch": 6.118299445471349,
+      "grad_norm": 0.07301589101552963,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 59580
+    },
+    {
+      "epoch": 6.118607516943931,
+      "grad_norm": 0.05073026567697525,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 59583
+    },
+    {
+      "epoch": 6.118915588416512,
+      "grad_norm": 0.04244513809680939,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 59586
+    },
+    {
+      "epoch": 6.119223659889094,
+      "grad_norm": 0.0322258435189724,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 59589
+    },
+    {
+      "epoch": 6.1195317313616755,
+      "grad_norm": 0.040786877274513245,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 59592
+    },
+    {
+      "epoch": 6.1198398028342575,
+      "grad_norm": 0.1062447801232338,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 59595
+    },
+    {
+      "epoch": 6.12014787430684,
+      "grad_norm": 0.07283103466033936,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 59598
+    },
+    {
+      "epoch": 6.120455945779421,
+      "grad_norm": 0.04273154214024544,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 59601
+    },
+    {
+      "epoch": 6.120764017252003,
+      "grad_norm": 0.0657401755452156,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 59604
+    },
+    {
+      "epoch": 6.121072088724584,
+      "grad_norm": 0.03795840963721275,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 59607
+    },
+    {
+      "epoch": 6.121380160197166,
+      "grad_norm": 0.03355659544467926,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 59610
+    },
+    {
+      "epoch": 6.121688231669747,
+      "grad_norm": 0.13272467255592346,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 59613
+    },
+    {
+      "epoch": 6.121996303142329,
+      "grad_norm": 0.07595691084861755,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 59616
+    },
+    {
+      "epoch": 6.12230437461491,
+      "grad_norm": 0.0829743966460228,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 59619
+    },
+    {
+      "epoch": 6.122612446087492,
+      "grad_norm": 0.05950823426246643,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 59622
+    },
+    {
+      "epoch": 6.122920517560074,
+      "grad_norm": 0.06740408390760422,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 59625
+    },
+    {
+      "epoch": 6.123228589032656,
+      "grad_norm": 0.03415486961603165,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 59628
+    },
+    {
+      "epoch": 6.123536660505237,
+      "grad_norm": 0.06124688684940338,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 59631
+    },
+    {
+      "epoch": 6.123844731977819,
+      "grad_norm": 0.06151755899190903,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 59634
+    },
+    {
+      "epoch": 6.124152803450401,
+      "grad_norm": 0.12702985107898712,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 59637
+    },
+    {
+      "epoch": 6.124460874922982,
+      "grad_norm": 0.08845409005880356,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 59640
+    },
+    {
+      "epoch": 6.124768946395564,
+      "grad_norm": 0.061156027019023895,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 59643
+    },
+    {
+      "epoch": 6.125077017868145,
+      "grad_norm": 0.050899405032396317,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 59646
+    },
+    {
+      "epoch": 6.125385089340727,
+      "grad_norm": 0.066854327917099,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 59649
+    },
+    {
+      "epoch": 6.1256931608133085,
+      "grad_norm": 0.06697458028793335,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 59652
+    },
+    {
+      "epoch": 6.1260012322858906,
+      "grad_norm": 0.05130474641919136,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 59655
+    },
+    {
+      "epoch": 6.126309303758472,
+      "grad_norm": 0.04654531553387642,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 59658
+    },
+    {
+      "epoch": 6.126617375231054,
+      "grad_norm": 0.036213308572769165,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 59661
+    },
+    {
+      "epoch": 6.126925446703635,
+      "grad_norm": 0.052255984395742416,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 59664
+    },
+    {
+      "epoch": 6.127233518176217,
+      "grad_norm": 0.07511945068836212,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 59667
+    },
+    {
+      "epoch": 6.127541589648798,
+      "grad_norm": 0.09632497280836105,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 59670
+    },
+    {
+      "epoch": 6.12784966112138,
+      "grad_norm": 0.059645794332027435,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 59673
+    },
+    {
+      "epoch": 6.128157732593962,
+      "grad_norm": 0.08466697484254837,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 59676
+    },
+    {
+      "epoch": 6.128465804066543,
+      "grad_norm": 0.06881053000688553,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 59679
+    },
+    {
+      "epoch": 6.1287738755391254,
+      "grad_norm": 0.04392458125948906,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 59682
+    },
+    {
+      "epoch": 6.129081947011707,
+      "grad_norm": 0.046918466687202454,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 59685
+    },
+    {
+      "epoch": 6.129390018484289,
+      "grad_norm": 0.13943582773208618,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 59688
+    },
+    {
+      "epoch": 6.12969808995687,
+      "grad_norm": 0.032962556928396225,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 59691
+    },
+    {
+      "epoch": 6.130006161429452,
+      "grad_norm": 0.04881718382239342,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 59694
+    },
+    {
+      "epoch": 6.130314232902033,
+      "grad_norm": 0.049873992800712585,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 59697
+    },
+    {
+      "epoch": 6.130622304374615,
+      "grad_norm": 0.11788579076528549,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 59700
+    },
+    {
+      "epoch": 6.130930375847196,
+      "grad_norm": 0.06456585973501205,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 59703
+    },
+    {
+      "epoch": 6.131238447319778,
+      "grad_norm": 0.04507961496710777,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 59706
+    },
+    {
+      "epoch": 6.1315465187923595,
+      "grad_norm": 0.04093556106090546,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 59709
+    },
+    {
+      "epoch": 6.1318545902649415,
+      "grad_norm": 0.034374650567770004,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 59712
+    },
+    {
+      "epoch": 6.132162661737523,
+      "grad_norm": 0.09779324382543564,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 59715
+    },
+    {
+      "epoch": 6.132470733210105,
+      "grad_norm": 0.03912382200360298,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 59718
+    },
+    {
+      "epoch": 6.132778804682687,
+      "grad_norm": 0.046537820249795914,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 59721
+    },
+    {
+      "epoch": 6.133086876155268,
+      "grad_norm": 0.04848983883857727,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 59724
+    },
+    {
+      "epoch": 6.13339494762785,
+      "grad_norm": 0.03337480127811432,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 59727
+    },
+    {
+      "epoch": 6.133703019100431,
+      "grad_norm": 0.06988492608070374,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 59730
+    },
+    {
+      "epoch": 6.134011090573013,
+      "grad_norm": 0.1184145137667656,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 59733
+    },
+    {
+      "epoch": 6.134319162045594,
+      "grad_norm": 0.1065261960029602,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 59736
+    },
+    {
+      "epoch": 6.134627233518176,
+      "grad_norm": 0.03989608585834503,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 59739
+    },
+    {
+      "epoch": 6.134935304990758,
+      "grad_norm": 0.05950331315398216,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 59742
+    },
+    {
+      "epoch": 6.13524337646334,
+      "grad_norm": 0.044078536331653595,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 59745
+    },
+    {
+      "epoch": 6.135551447935921,
+      "grad_norm": 0.06196853518486023,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 59748
+    },
+    {
+      "epoch": 6.135859519408503,
+      "grad_norm": 0.05450844019651413,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 59751
+    },
+    {
+      "epoch": 6.136167590881084,
+      "grad_norm": 0.05260879918932915,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 59754
+    },
+    {
+      "epoch": 6.136475662353666,
+      "grad_norm": 0.04335511103272438,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 59757
+    },
+    {
+      "epoch": 6.136783733826248,
+      "grad_norm": 0.05219843611121178,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 59760
+    },
+    {
+      "epoch": 6.137091805298829,
+      "grad_norm": 0.06641986966133118,
+      "learning_rate": 0.01,
+      "loss": 2.021,
+      "step": 59763
+    },
+    {
+      "epoch": 6.137399876771411,
+      "grad_norm": 0.10496216267347336,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 59766
+    },
+    {
+      "epoch": 6.1377079482439925,
+      "grad_norm": 0.1305655688047409,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 59769
+    },
+    {
+      "epoch": 6.1380160197165745,
+      "grad_norm": 0.06683500856161118,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 59772
+    },
+    {
+      "epoch": 6.138324091189156,
+      "grad_norm": 0.06884671747684479,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 59775
+    },
+    {
+      "epoch": 6.138632162661738,
+      "grad_norm": 0.04899144545197487,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 59778
+    },
+    {
+      "epoch": 6.138940234134319,
+      "grad_norm": 0.061058346182107925,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 59781
+    },
+    {
+      "epoch": 6.139248305606901,
+      "grad_norm": 0.03729567304253578,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 59784
+    },
+    {
+      "epoch": 6.139556377079482,
+      "grad_norm": 0.1232062354683876,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 59787
+    },
+    {
+      "epoch": 6.139864448552064,
+      "grad_norm": 0.07398378103971481,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 59790
+    },
+    {
+      "epoch": 6.140172520024645,
+      "grad_norm": 0.06523609161376953,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 59793
+    },
+    {
+      "epoch": 6.140480591497227,
+      "grad_norm": 0.04748447984457016,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 59796
+    },
+    {
+      "epoch": 6.140788662969809,
+      "grad_norm": 0.10762014240026474,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 59799
+    },
+    {
+      "epoch": 6.141096734442391,
+      "grad_norm": 0.05152611434459686,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 59802
+    },
+    {
+      "epoch": 6.141404805914973,
+      "grad_norm": 0.03151378408074379,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 59805
+    },
+    {
+      "epoch": 6.141712877387554,
+      "grad_norm": 0.045448195189237595,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 59808
+    },
+    {
+      "epoch": 6.142020948860136,
+      "grad_norm": 0.03704848885536194,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 59811
+    },
+    {
+      "epoch": 6.142329020332717,
+      "grad_norm": 0.0592067614197731,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 59814
+    },
+    {
+      "epoch": 6.142637091805299,
+      "grad_norm": 0.11523483693599701,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 59817
+    },
+    {
+      "epoch": 6.14294516327788,
+      "grad_norm": 0.14963838458061218,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 59820
+    },
+    {
+      "epoch": 6.143253234750462,
+      "grad_norm": 0.08962684869766235,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 59823
+    },
+    {
+      "epoch": 6.143561306223043,
+      "grad_norm": 0.06388752162456512,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 59826
+    },
+    {
+      "epoch": 6.1438693776956255,
+      "grad_norm": 0.06632399559020996,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 59829
+    },
+    {
+      "epoch": 6.144177449168207,
+      "grad_norm": 0.08517513424158096,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 59832
+    },
+    {
+      "epoch": 6.144485520640789,
+      "grad_norm": 0.05397310480475426,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 59835
+    },
+    {
+      "epoch": 6.144793592113371,
+      "grad_norm": 0.04470159485936165,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 59838
+    },
+    {
+      "epoch": 6.145101663585952,
+      "grad_norm": 0.048731885850429535,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 59841
+    },
+    {
+      "epoch": 6.145409735058534,
+      "grad_norm": 0.13363894820213318,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 59844
+    },
+    {
+      "epoch": 6.145717806531115,
+      "grad_norm": 0.04807397723197937,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 59847
+    },
+    {
+      "epoch": 6.146025878003697,
+      "grad_norm": 0.05982908979058266,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 59850
+    },
+    {
+      "epoch": 6.146333949476278,
+      "grad_norm": 0.0730501338839531,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 59853
+    },
+    {
+      "epoch": 6.14664202094886,
+      "grad_norm": 0.04355951398611069,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 59856
+    },
+    {
+      "epoch": 6.1469500924214415,
+      "grad_norm": 0.10257246345281601,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 59859
+    },
+    {
+      "epoch": 6.147258163894024,
+      "grad_norm": 0.0933048278093338,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 59862
+    },
+    {
+      "epoch": 6.147566235366605,
+      "grad_norm": 0.03633146733045578,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 59865
+    },
+    {
+      "epoch": 6.147874306839187,
+      "grad_norm": 0.07809139788150787,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 59868
+    },
+    {
+      "epoch": 6.148182378311768,
+      "grad_norm": 0.06432943791151047,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 59871
+    },
+    {
+      "epoch": 6.14849044978435,
+      "grad_norm": 0.04523482546210289,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 59874
+    },
+    {
+      "epoch": 6.148798521256932,
+      "grad_norm": 0.04328586533665657,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 59877
+    },
+    {
+      "epoch": 6.149106592729513,
+      "grad_norm": 0.04336148872971535,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 59880
+    },
+    {
+      "epoch": 6.149414664202095,
+      "grad_norm": 0.11612840741872787,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 59883
+    },
+    {
+      "epoch": 6.149722735674676,
+      "grad_norm": 0.040377210825681686,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 59886
+    },
+    {
+      "epoch": 6.1500308071472585,
+      "grad_norm": 0.05374842882156372,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 59889
+    },
+    {
+      "epoch": 6.15033887861984,
+      "grad_norm": 0.042278483510017395,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 59892
+    },
+    {
+      "epoch": 6.150646950092422,
+      "grad_norm": 0.06057736650109291,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 59895
+    },
+    {
+      "epoch": 6.150955021565003,
+      "grad_norm": 0.07426716387271881,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 59898
+    },
+    {
+      "epoch": 6.151263093037585,
+      "grad_norm": 0.06555135548114777,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 59901
+    },
+    {
+      "epoch": 6.151571164510166,
+      "grad_norm": 0.031795963644981384,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 59904
+    },
+    {
+      "epoch": 6.151879235982748,
+      "grad_norm": 0.0928960070014,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 59907
+    },
+    {
+      "epoch": 6.152187307455329,
+      "grad_norm": 0.05229020118713379,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 59910
+    },
+    {
+      "epoch": 6.152495378927911,
+      "grad_norm": 0.05363921821117401,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 59913
+    },
+    {
+      "epoch": 6.1528034504004925,
+      "grad_norm": 0.05340690538287163,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 59916
+    },
+    {
+      "epoch": 6.1531115218730745,
+      "grad_norm": 0.04054640233516693,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 59919
+    },
+    {
+      "epoch": 6.153419593345657,
+      "grad_norm": 0.05909667909145355,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 59922
+    },
+    {
+      "epoch": 6.153727664818238,
+      "grad_norm": 0.04531288146972656,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 59925
+    },
+    {
+      "epoch": 6.15403573629082,
+      "grad_norm": 0.041250940412282944,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 59928
+    },
+    {
+      "epoch": 6.154343807763401,
+      "grad_norm": 0.042447373270988464,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 59931
+    },
+    {
+      "epoch": 6.154651879235983,
+      "grad_norm": 0.06417303532361984,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 59934
+    },
+    {
+      "epoch": 6.154959950708564,
+      "grad_norm": 0.12572647631168365,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 59937
+    },
+    {
+      "epoch": 6.155268022181146,
+      "grad_norm": 0.06711404025554657,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 59940
+    },
+    {
+      "epoch": 6.155576093653727,
+      "grad_norm": 0.09349125623703003,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 59943
+    },
+    {
+      "epoch": 6.155884165126309,
+      "grad_norm": 0.046295300126075745,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 59946
+    },
+    {
+      "epoch": 6.156192236598891,
+      "grad_norm": 0.08443499356508255,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 59949
+    },
+    {
+      "epoch": 6.156500308071473,
+      "grad_norm": 0.048439864069223404,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 59952
+    },
+    {
+      "epoch": 6.156808379544054,
+      "grad_norm": 0.10180726647377014,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 59955
+    },
+    {
+      "epoch": 6.157116451016636,
+      "grad_norm": 0.03144579380750656,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 59958
+    },
+    {
+      "epoch": 6.157424522489218,
+      "grad_norm": 0.0712040439248085,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 59961
+    },
+    {
+      "epoch": 6.157732593961799,
+      "grad_norm": 0.054967526346445084,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 59964
+    },
+    {
+      "epoch": 6.158040665434381,
+      "grad_norm": 0.04168044403195381,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 59967
+    },
+    {
+      "epoch": 6.158348736906962,
+      "grad_norm": 0.05962495878338814,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 59970
+    },
+    {
+      "epoch": 6.158656808379544,
+      "grad_norm": 0.11579214036464691,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 59973
+    },
+    {
+      "epoch": 6.1589648798521255,
+      "grad_norm": 0.03569880872964859,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 59976
+    },
+    {
+      "epoch": 6.1592729513247075,
+      "grad_norm": 0.09522563964128494,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 59979
+    },
+    {
+      "epoch": 6.159581022797289,
+      "grad_norm": 0.07732608169317245,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 59982
+    },
+    {
+      "epoch": 6.159889094269871,
+      "grad_norm": 0.057550933212041855,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 59985
+    },
+    {
+      "epoch": 6.160197165742452,
+      "grad_norm": 0.0942176878452301,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 59988
+    },
+    {
+      "epoch": 6.160505237215034,
+      "grad_norm": 0.0959196537733078,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 59991
+    },
+    {
+      "epoch": 6.160813308687615,
+      "grad_norm": 0.07278607040643692,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 59994
+    },
+    {
+      "epoch": 6.161121380160197,
+      "grad_norm": 0.05867347866296768,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 59997
+    },
+    {
+      "epoch": 6.161429451632779,
+      "grad_norm": 0.03255612030625343,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 60000
+    },
+    {
+      "epoch": 6.16173752310536,
+      "grad_norm": 0.056017834693193436,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 60003
+    },
+    {
+      "epoch": 6.162045594577942,
+      "grad_norm": 0.0943060889840126,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 60006
+    },
+    {
+      "epoch": 6.162353666050524,
+      "grad_norm": 0.08472935855388641,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 60009
+    },
+    {
+      "epoch": 6.162661737523106,
+      "grad_norm": 0.053891006857156754,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 60012
+    },
+    {
+      "epoch": 6.162969808995687,
+      "grad_norm": 0.042883194983005524,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 60015
+    },
+    {
+      "epoch": 6.163277880468269,
+      "grad_norm": 0.05750950425863266,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 60018
+    },
+    {
+      "epoch": 6.16358595194085,
+      "grad_norm": 0.07917001843452454,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 60021
+    },
+    {
+      "epoch": 6.163894023413432,
+      "grad_norm": 0.06416572630405426,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 60024
+    },
+    {
+      "epoch": 6.164202094886013,
+      "grad_norm": 0.06587263941764832,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 60027
+    },
+    {
+      "epoch": 6.164510166358595,
+      "grad_norm": 0.09511373192071915,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 60030
+    },
+    {
+      "epoch": 6.164818237831176,
+      "grad_norm": 0.11690583825111389,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 60033
+    },
+    {
+      "epoch": 6.1651263093037585,
+      "grad_norm": 0.05916045978665352,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 60036
+    },
+    {
+      "epoch": 6.1654343807763405,
+      "grad_norm": 0.07679677754640579,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 60039
+    },
+    {
+      "epoch": 6.165742452248922,
+      "grad_norm": 0.05868346244096756,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 60042
+    },
+    {
+      "epoch": 6.166050523721504,
+      "grad_norm": 0.04336436092853546,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 60045
+    },
+    {
+      "epoch": 6.166358595194085,
+      "grad_norm": 0.053842004388570786,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 60048
+    },
+    {
+      "epoch": 6.166666666666667,
+      "grad_norm": 0.05783376097679138,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 60051
+    },
+    {
+      "epoch": 6.166974738139248,
+      "grad_norm": 0.12786927819252014,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 60054
+    },
+    {
+      "epoch": 6.16728280961183,
+      "grad_norm": 0.10869529098272324,
+      "learning_rate": 0.01,
+      "loss": 2.0226,
+      "step": 60057
+    },
+    {
+      "epoch": 6.167590881084411,
+      "grad_norm": 0.09978947043418884,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 60060
+    },
+    {
+      "epoch": 6.167898952556993,
+      "grad_norm": 0.06388422101736069,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 60063
+    },
+    {
+      "epoch": 6.1682070240295745,
+      "grad_norm": 0.05348409339785576,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 60066
+    },
+    {
+      "epoch": 6.168515095502157,
+      "grad_norm": 0.047684285789728165,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 60069
+    },
+    {
+      "epoch": 6.168823166974738,
+      "grad_norm": 0.031933482736349106,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 60072
+    },
+    {
+      "epoch": 6.16913123844732,
+      "grad_norm": 0.06409978866577148,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 60075
+    },
+    {
+      "epoch": 6.169439309919902,
+      "grad_norm": 0.07085554301738739,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 60078
+    },
+    {
+      "epoch": 6.169747381392483,
+      "grad_norm": 0.04958326742053032,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 60081
+    },
+    {
+      "epoch": 6.170055452865065,
+      "grad_norm": 0.037209220230579376,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 60084
+    },
+    {
+      "epoch": 6.170363524337646,
+      "grad_norm": 0.044104501605033875,
+      "learning_rate": 0.01,
+      "loss": 2.0076,
+      "step": 60087
+    },
+    {
+      "epoch": 6.170671595810228,
+      "grad_norm": 0.06651319563388824,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 60090
+    },
+    {
+      "epoch": 6.1709796672828094,
+      "grad_norm": 0.15120342373847961,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 60093
+    },
+    {
+      "epoch": 6.1712877387553915,
+      "grad_norm": 0.17715202271938324,
+      "learning_rate": 0.01,
+      "loss": 2.015,
+      "step": 60096
+    },
+    {
+      "epoch": 6.171595810227973,
+      "grad_norm": 0.07036132365465164,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 60099
+    },
+    {
+      "epoch": 6.171903881700555,
+      "grad_norm": 0.05515659227967262,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 60102
+    },
+    {
+      "epoch": 6.172211953173136,
+      "grad_norm": 0.054766733199357986,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 60105
+    },
+    {
+      "epoch": 6.172520024645718,
+      "grad_norm": 0.06652894616127014,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 60108
+    },
+    {
+      "epoch": 6.172828096118299,
+      "grad_norm": 0.05079234391450882,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 60111
+    },
+    {
+      "epoch": 6.173136167590881,
+      "grad_norm": 0.04592439532279968,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 60114
+    },
+    {
+      "epoch": 6.173444239063462,
+      "grad_norm": 0.03100336715579033,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 60117
+    },
+    {
+      "epoch": 6.173752310536044,
+      "grad_norm": 0.03455538675189018,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 60120
+    },
+    {
+      "epoch": 6.174060382008626,
+      "grad_norm": 0.04403119906783104,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 60123
+    },
+    {
+      "epoch": 6.1743684534812076,
+      "grad_norm": 0.1155620738863945,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 60126
+    },
+    {
+      "epoch": 6.17467652495379,
+      "grad_norm": 0.06684102863073349,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 60129
+    },
+    {
+      "epoch": 6.174984596426371,
+      "grad_norm": 0.04837580397725105,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 60132
+    },
+    {
+      "epoch": 6.175292667898953,
+      "grad_norm": 0.11638902872800827,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 60135
+    },
+    {
+      "epoch": 6.175600739371534,
+      "grad_norm": 0.06774953007698059,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 60138
+    },
+    {
+      "epoch": 6.175908810844116,
+      "grad_norm": 0.05518386512994766,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 60141
+    },
+    {
+      "epoch": 6.176216882316697,
+      "grad_norm": 0.03791549429297447,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 60144
+    },
+    {
+      "epoch": 6.176524953789279,
+      "grad_norm": 0.0599411316215992,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 60147
+    },
+    {
+      "epoch": 6.17683302526186,
+      "grad_norm": 0.08578135818243027,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 60150
+    },
+    {
+      "epoch": 6.1771410967344424,
+      "grad_norm": 0.04651505872607231,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 60153
+    },
+    {
+      "epoch": 6.177449168207024,
+      "grad_norm": 0.09119927883148193,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 60156
+    },
+    {
+      "epoch": 6.177757239679606,
+      "grad_norm": 0.12429966032505035,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 60159
+    },
+    {
+      "epoch": 6.178065311152188,
+      "grad_norm": 0.0350540392100811,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 60162
+    },
+    {
+      "epoch": 6.178373382624769,
+      "grad_norm": 0.06904268264770508,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 60165
+    },
+    {
+      "epoch": 6.178681454097351,
+      "grad_norm": 0.06134537607431412,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 60168
+    },
+    {
+      "epoch": 6.178989525569932,
+      "grad_norm": 0.07607719302177429,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 60171
+    },
+    {
+      "epoch": 6.179297597042514,
+      "grad_norm": 0.1023942083120346,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 60174
+    },
+    {
+      "epoch": 6.179605668515095,
+      "grad_norm": 0.04248664528131485,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 60177
+    },
+    {
+      "epoch": 6.179913739987677,
+      "grad_norm": 0.036357518285512924,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 60180
+    },
+    {
+      "epoch": 6.1802218114602585,
+      "grad_norm": 0.04085879400372505,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 60183
+    },
+    {
+      "epoch": 6.180529882932841,
+      "grad_norm": 0.05432041734457016,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 60186
+    },
+    {
+      "epoch": 6.180837954405422,
+      "grad_norm": 0.0373106487095356,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 60189
+    },
+    {
+      "epoch": 6.181146025878004,
+      "grad_norm": 0.13600775599479675,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 60192
+    },
+    {
+      "epoch": 6.181454097350585,
+      "grad_norm": 0.10217521339654922,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 60195
+    },
+    {
+      "epoch": 6.181762168823167,
+      "grad_norm": 0.11078688502311707,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 60198
+    },
+    {
+      "epoch": 6.182070240295749,
+      "grad_norm": 0.11216065287590027,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 60201
+    },
+    {
+      "epoch": 6.18237831176833,
+      "grad_norm": 0.07498262822628021,
+      "learning_rate": 0.01,
+      "loss": 2.004,
+      "step": 60204
+    },
+    {
+      "epoch": 6.182686383240912,
+      "grad_norm": 0.0481102392077446,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 60207
+    },
+    {
+      "epoch": 6.182994454713493,
+      "grad_norm": 0.06563237309455872,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 60210
+    },
+    {
+      "epoch": 6.1833025261860755,
+      "grad_norm": 0.065886951982975,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 60213
+    },
+    {
+      "epoch": 6.183610597658657,
+      "grad_norm": 0.06081174686551094,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 60216
+    },
+    {
+      "epoch": 6.183918669131239,
+      "grad_norm": 0.040813907980918884,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 60219
+    },
+    {
+      "epoch": 6.18422674060382,
+      "grad_norm": 0.08460288494825363,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 60222
+    },
+    {
+      "epoch": 6.184534812076402,
+      "grad_norm": 0.1325504183769226,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 60225
+    },
+    {
+      "epoch": 6.184842883548983,
+      "grad_norm": 0.047543518245220184,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 60228
+    },
+    {
+      "epoch": 6.185150955021565,
+      "grad_norm": 0.09578222036361694,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 60231
+    },
+    {
+      "epoch": 6.185459026494146,
+      "grad_norm": 0.08167849481105804,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 60234
+    },
+    {
+      "epoch": 6.185767097966728,
+      "grad_norm": 0.04946918785572052,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 60237
+    },
+    {
+      "epoch": 6.1860751694393095,
+      "grad_norm": 0.08602156490087509,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 60240
+    },
+    {
+      "epoch": 6.1863832409118915,
+      "grad_norm": 0.07393812388181686,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 60243
+    },
+    {
+      "epoch": 6.186691312384474,
+      "grad_norm": 0.04238250479102135,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 60246
+    },
+    {
+      "epoch": 6.186999383857055,
+      "grad_norm": 0.05332363396883011,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 60249
+    },
+    {
+      "epoch": 6.187307455329637,
+      "grad_norm": 0.048810552805662155,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 60252
+    },
+    {
+      "epoch": 6.187615526802218,
+      "grad_norm": 0.04000997915863991,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 60255
+    },
+    {
+      "epoch": 6.1879235982748,
+      "grad_norm": 0.07165291905403137,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 60258
+    },
+    {
+      "epoch": 6.188231669747381,
+      "grad_norm": 0.04843559488654137,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 60261
+    },
+    {
+      "epoch": 6.188539741219963,
+      "grad_norm": 0.048593342304229736,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 60264
+    },
+    {
+      "epoch": 6.188847812692544,
+      "grad_norm": 0.05250190943479538,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 60267
+    },
+    {
+      "epoch": 6.189155884165126,
+      "grad_norm": 0.09123876690864563,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 60270
+    },
+    {
+      "epoch": 6.189463955637708,
+      "grad_norm": 0.11671872437000275,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 60273
+    },
+    {
+      "epoch": 6.18977202711029,
+      "grad_norm": 0.048420678824186325,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 60276
+    },
+    {
+      "epoch": 6.190080098582872,
+      "grad_norm": 0.11455632746219635,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 60279
+    },
+    {
+      "epoch": 6.190388170055453,
+      "grad_norm": 0.08793000131845474,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 60282
+    },
+    {
+      "epoch": 6.190696241528035,
+      "grad_norm": 0.0460558645427227,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 60285
+    },
+    {
+      "epoch": 6.191004313000616,
+      "grad_norm": 0.0427497997879982,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 60288
+    },
+    {
+      "epoch": 6.191312384473198,
+      "grad_norm": 0.04136178269982338,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 60291
+    },
+    {
+      "epoch": 6.191620455945779,
+      "grad_norm": 0.06737589836120605,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 60294
+    },
+    {
+      "epoch": 6.191928527418361,
+      "grad_norm": 0.05001933127641678,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 60297
+    },
+    {
+      "epoch": 6.1922365988909425,
+      "grad_norm": 0.13523389399051666,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 60300
+    },
+    {
+      "epoch": 6.1925446703635245,
+      "grad_norm": 0.09118752926588058,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 60303
+    },
+    {
+      "epoch": 6.192852741836106,
+      "grad_norm": 0.061937130987644196,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 60306
+    },
+    {
+      "epoch": 6.193160813308688,
+      "grad_norm": 0.09417203068733215,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 60309
+    },
+    {
+      "epoch": 6.193468884781269,
+      "grad_norm": 0.1443474441766739,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 60312
+    },
+    {
+      "epoch": 6.193776956253851,
+      "grad_norm": 0.12889619171619415,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 60315
+    },
+    {
+      "epoch": 6.194085027726432,
+      "grad_norm": 0.08010325580835342,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 60318
+    },
+    {
+      "epoch": 6.194393099199014,
+      "grad_norm": 0.0863533690571785,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 60321
+    },
+    {
+      "epoch": 6.194701170671596,
+      "grad_norm": 0.0709909126162529,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 60324
+    },
+    {
+      "epoch": 6.195009242144177,
+      "grad_norm": 0.04236047342419624,
+      "learning_rate": 0.01,
+      "loss": 2.0178,
+      "step": 60327
+    },
+    {
+      "epoch": 6.195317313616759,
+      "grad_norm": 0.06927076727151871,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 60330
+    },
+    {
+      "epoch": 6.195625385089341,
+      "grad_norm": 0.07250656932592392,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 60333
+    },
+    {
+      "epoch": 6.195933456561923,
+      "grad_norm": 0.06588046252727509,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 60336
+    },
+    {
+      "epoch": 6.196241528034504,
+      "grad_norm": 0.07093591243028641,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 60339
+    },
+    {
+      "epoch": 6.196549599507086,
+      "grad_norm": 0.10864640772342682,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 60342
+    },
+    {
+      "epoch": 6.196857670979667,
+      "grad_norm": 0.12769301235675812,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 60345
+    },
+    {
+      "epoch": 6.197165742452249,
+      "grad_norm": 0.03568786382675171,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 60348
+    },
+    {
+      "epoch": 6.19747381392483,
+      "grad_norm": 0.08632007986307144,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 60351
+    },
+    {
+      "epoch": 6.197781885397412,
+      "grad_norm": 0.047804027795791626,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 60354
+    },
+    {
+      "epoch": 6.198089956869993,
+      "grad_norm": 0.08226168155670166,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 60357
+    },
+    {
+      "epoch": 6.1983980283425755,
+      "grad_norm": 0.06796471774578094,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 60360
+    },
+    {
+      "epoch": 6.1987060998151575,
+      "grad_norm": 0.05324438586831093,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 60363
+    },
+    {
+      "epoch": 6.199014171287739,
+      "grad_norm": 0.05389709025621414,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 60366
+    },
+    {
+      "epoch": 6.199322242760321,
+      "grad_norm": 0.04287771135568619,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 60369
+    },
+    {
+      "epoch": 6.199630314232902,
+      "grad_norm": 0.06289256364107132,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 60372
+    },
+    {
+      "epoch": 6.199938385705484,
+      "grad_norm": 0.04109280928969383,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 60375
+    },
+    {
+      "epoch": 6.200246457178065,
+      "grad_norm": 0.13262084126472473,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 60378
+    },
+    {
+      "epoch": 6.200554528650647,
+      "grad_norm": 0.05487149953842163,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 60381
+    },
+    {
+      "epoch": 6.200862600123228,
+      "grad_norm": 0.08410999178886414,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 60384
+    },
+    {
+      "epoch": 6.20117067159581,
+      "grad_norm": 0.09868727624416351,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 60387
+    },
+    {
+      "epoch": 6.2014787430683915,
+      "grad_norm": 0.09402374923229218,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 60390
+    },
+    {
+      "epoch": 6.201786814540974,
+      "grad_norm": 0.09398900717496872,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 60393
+    },
+    {
+      "epoch": 6.202094886013555,
+      "grad_norm": 0.07040904462337494,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 60396
+    },
+    {
+      "epoch": 6.202402957486137,
+      "grad_norm": 0.09048958867788315,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 60399
+    },
+    {
+      "epoch": 6.202711028958719,
+      "grad_norm": 0.08249911665916443,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 60402
+    },
+    {
+      "epoch": 6.2030191004313,
+      "grad_norm": 0.1375819742679596,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 60405
+    },
+    {
+      "epoch": 6.203327171903882,
+      "grad_norm": 0.12005623430013657,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 60408
+    },
+    {
+      "epoch": 6.203635243376463,
+      "grad_norm": 0.08359742164611816,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 60411
+    },
+    {
+      "epoch": 6.203943314849045,
+      "grad_norm": 0.08574346452951431,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 60414
+    },
+    {
+      "epoch": 6.204251386321626,
+      "grad_norm": 0.04601282626390457,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 60417
+    },
+    {
+      "epoch": 6.2045594577942085,
+      "grad_norm": 0.04463723674416542,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 60420
+    },
+    {
+      "epoch": 6.20486752926679,
+      "grad_norm": 0.05424230918288231,
+      "learning_rate": 0.01,
+      "loss": 2.0251,
+      "step": 60423
+    },
+    {
+      "epoch": 6.205175600739372,
+      "grad_norm": 0.0748244971036911,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 60426
+    },
+    {
+      "epoch": 6.205483672211953,
+      "grad_norm": 0.05594291165471077,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 60429
+    },
+    {
+      "epoch": 6.205791743684535,
+      "grad_norm": 0.03947063907980919,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 60432
+    },
+    {
+      "epoch": 6.206099815157116,
+      "grad_norm": 0.037693090736866,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 60435
+    },
+    {
+      "epoch": 6.206407886629698,
+      "grad_norm": 0.09147128462791443,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 60438
+    },
+    {
+      "epoch": 6.206715958102279,
+      "grad_norm": 0.03525812551379204,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 60441
+    },
+    {
+      "epoch": 6.207024029574861,
+      "grad_norm": 0.0658460482954979,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 60444
+    },
+    {
+      "epoch": 6.207332101047443,
+      "grad_norm": 0.10629269480705261,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 60447
+    },
+    {
+      "epoch": 6.2076401725200245,
+      "grad_norm": 0.07760986685752869,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 60450
+    },
+    {
+      "epoch": 6.207948243992607,
+      "grad_norm": 0.0387595109641552,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 60453
+    },
+    {
+      "epoch": 6.208256315465188,
+      "grad_norm": 0.04037671163678169,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 60456
+    },
+    {
+      "epoch": 6.20856438693777,
+      "grad_norm": 0.049021679908037186,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 60459
+    },
+    {
+      "epoch": 6.208872458410351,
+      "grad_norm": 0.04115486890077591,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 60462
+    },
+    {
+      "epoch": 6.209180529882933,
+      "grad_norm": 0.06775692850351334,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 60465
+    },
+    {
+      "epoch": 6.209488601355514,
+      "grad_norm": 0.04244436323642731,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 60468
+    },
+    {
+      "epoch": 6.209796672828096,
+      "grad_norm": 0.046518050134181976,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 60471
+    },
+    {
+      "epoch": 6.210104744300677,
+      "grad_norm": 0.058811433613300323,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 60474
+    },
+    {
+      "epoch": 6.210412815773259,
+      "grad_norm": 0.08617477864027023,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 60477
+    },
+    {
+      "epoch": 6.2107208872458415,
+      "grad_norm": 0.1398414820432663,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 60480
+    },
+    {
+      "epoch": 6.211028958718423,
+      "grad_norm": 0.04411006346344948,
+      "learning_rate": 0.01,
+      "loss": 2.0152,
+      "step": 60483
+    },
+    {
+      "epoch": 6.211337030191005,
+      "grad_norm": 0.05274950712919235,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 60486
+    },
+    {
+      "epoch": 6.211645101663586,
+      "grad_norm": 0.03863611817359924,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 60489
+    },
+    {
+      "epoch": 6.211953173136168,
+      "grad_norm": 0.04664864018559456,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 60492
+    },
+    {
+      "epoch": 6.212261244608749,
+      "grad_norm": 0.03843964263796806,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 60495
+    },
+    {
+      "epoch": 6.212569316081331,
+      "grad_norm": 0.02867240644991398,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 60498
+    },
+    {
+      "epoch": 6.212877387553912,
+      "grad_norm": 0.03859257698059082,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 60501
+    },
+    {
+      "epoch": 6.213185459026494,
+      "grad_norm": 0.04639716073870659,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 60504
+    },
+    {
+      "epoch": 6.2134935304990755,
+      "grad_norm": 0.05346795171499252,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 60507
+    },
+    {
+      "epoch": 6.2138016019716575,
+      "grad_norm": 0.09336802363395691,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 60510
+    },
+    {
+      "epoch": 6.214109673444239,
+      "grad_norm": 0.10028377920389175,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 60513
+    },
+    {
+      "epoch": 6.214417744916821,
+      "grad_norm": 0.04280233010649681,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 60516
+    },
+    {
+      "epoch": 6.214725816389402,
+      "grad_norm": 0.07395056635141373,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 60519
+    },
+    {
+      "epoch": 6.215033887861984,
+      "grad_norm": 0.08924415707588196,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 60522
+    },
+    {
+      "epoch": 6.215341959334566,
+      "grad_norm": 0.06368307024240494,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 60525
+    },
+    {
+      "epoch": 6.215650030807147,
+      "grad_norm": 0.07443444430828094,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 60528
+    },
+    {
+      "epoch": 6.215958102279729,
+      "grad_norm": 0.06193174049258232,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 60531
+    },
+    {
+      "epoch": 6.21626617375231,
+      "grad_norm": 0.05765648931264877,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 60534
+    },
+    {
+      "epoch": 6.216574245224892,
+      "grad_norm": 0.07838603854179382,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 60537
+    },
+    {
+      "epoch": 6.216882316697474,
+      "grad_norm": 0.0865786075592041,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 60540
+    },
+    {
+      "epoch": 6.217190388170056,
+      "grad_norm": 0.06937874108552933,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 60543
+    },
+    {
+      "epoch": 6.217498459642637,
+      "grad_norm": 0.04611770063638687,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 60546
+    },
+    {
+      "epoch": 6.217806531115219,
+      "grad_norm": 0.10261213779449463,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 60549
+    },
+    {
+      "epoch": 6.2181146025878,
+      "grad_norm": 0.06451880186796188,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 60552
+    },
+    {
+      "epoch": 6.218422674060382,
+      "grad_norm": 0.06459330767393112,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 60555
+    },
+    {
+      "epoch": 6.218730745532963,
+      "grad_norm": 0.06935889273881912,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 60558
+    },
+    {
+      "epoch": 6.219038817005545,
+      "grad_norm": 0.07407599687576294,
+      "learning_rate": 0.01,
+      "loss": 2.0052,
+      "step": 60561
+    },
+    {
+      "epoch": 6.219346888478127,
+      "grad_norm": 0.09501077234745026,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 60564
+    },
+    {
+      "epoch": 6.2196549599507085,
+      "grad_norm": 0.0854761153459549,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 60567
+    },
+    {
+      "epoch": 6.2199630314232905,
+      "grad_norm": 0.052260614931583405,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 60570
+    },
+    {
+      "epoch": 6.220271102895872,
+      "grad_norm": 0.03612583130598068,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 60573
+    },
+    {
+      "epoch": 6.220579174368454,
+      "grad_norm": 0.04795749485492706,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 60576
+    },
+    {
+      "epoch": 6.220887245841035,
+      "grad_norm": 0.055898912250995636,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 60579
+    },
+    {
+      "epoch": 6.221195317313617,
+      "grad_norm": 0.056687548756599426,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 60582
+    },
+    {
+      "epoch": 6.221503388786198,
+      "grad_norm": 0.052614741027355194,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 60585
+    },
+    {
+      "epoch": 6.22181146025878,
+      "grad_norm": 0.08860262483358383,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 60588
+    },
+    {
+      "epoch": 6.222119531731361,
+      "grad_norm": 0.10606038570404053,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 60591
+    },
+    {
+      "epoch": 6.222427603203943,
+      "grad_norm": 0.05913405120372772,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 60594
+    },
+    {
+      "epoch": 6.2227356746765246,
+      "grad_norm": 0.05625823140144348,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 60597
+    },
+    {
+      "epoch": 6.223043746149107,
+      "grad_norm": 0.10503005236387253,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 60600
+    },
+    {
+      "epoch": 6.223351817621689,
+      "grad_norm": 0.06931479275226593,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 60603
+    },
+    {
+      "epoch": 6.22365988909427,
+      "grad_norm": 0.07577981054782867,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 60606
+    },
+    {
+      "epoch": 6.223967960566852,
+      "grad_norm": 0.04593481123447418,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 60609
+    },
+    {
+      "epoch": 6.224276032039433,
+      "grad_norm": 0.040800996124744415,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 60612
+    },
+    {
+      "epoch": 6.224584103512015,
+      "grad_norm": 0.035023946315050125,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 60615
+    },
+    {
+      "epoch": 6.224892174984596,
+      "grad_norm": 0.10616744309663773,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 60618
+    },
+    {
+      "epoch": 6.225200246457178,
+      "grad_norm": 0.06860567629337311,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 60621
+    },
+    {
+      "epoch": 6.2255083179297594,
+      "grad_norm": 0.06851069629192352,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 60624
+    },
+    {
+      "epoch": 6.2258163894023415,
+      "grad_norm": 0.06767469644546509,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 60627
+    },
+    {
+      "epoch": 6.226124460874923,
+      "grad_norm": 0.06542889773845673,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 60630
+    },
+    {
+      "epoch": 6.226432532347505,
+      "grad_norm": 0.0505717433989048,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 60633
+    },
+    {
+      "epoch": 6.226740603820086,
+      "grad_norm": 0.04307612031698227,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 60636
+    },
+    {
+      "epoch": 6.227048675292668,
+      "grad_norm": 0.12549397349357605,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 60639
+    },
+    {
+      "epoch": 6.227356746765249,
+      "grad_norm": 0.10073670744895935,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 60642
+    },
+    {
+      "epoch": 6.227664818237831,
+      "grad_norm": 0.042758628726005554,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 60645
+    },
+    {
+      "epoch": 6.227972889710413,
+      "grad_norm": 0.0379779078066349,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 60648
+    },
+    {
+      "epoch": 6.228280961182994,
+      "grad_norm": 0.05374911054968834,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 60651
+    },
+    {
+      "epoch": 6.228589032655576,
+      "grad_norm": 0.0438760407269001,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 60654
+    },
+    {
+      "epoch": 6.228897104128158,
+      "grad_norm": 0.05077781528234482,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 60657
+    },
+    {
+      "epoch": 6.22920517560074,
+      "grad_norm": 0.05097432807087898,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 60660
+    },
+    {
+      "epoch": 6.229513247073321,
+      "grad_norm": 0.061216987669467926,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 60663
+    },
+    {
+      "epoch": 6.229821318545903,
+      "grad_norm": 0.056796032935380936,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 60666
+    },
+    {
+      "epoch": 6.230129390018484,
+      "grad_norm": 0.04003208130598068,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 60669
+    },
+    {
+      "epoch": 6.230437461491066,
+      "grad_norm": 0.07271415740251541,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 60672
+    },
+    {
+      "epoch": 6.230745532963647,
+      "grad_norm": 0.07599439471960068,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 60675
+    },
+    {
+      "epoch": 6.231053604436229,
+      "grad_norm": 0.40395355224609375,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 60678
+    },
+    {
+      "epoch": 6.231361675908811,
+      "grad_norm": 0.10945183783769608,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 60681
+    },
+    {
+      "epoch": 6.2316697473813925,
+      "grad_norm": 0.10694218426942825,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 60684
+    },
+    {
+      "epoch": 6.2319778188539745,
+      "grad_norm": 0.05942070484161377,
+      "learning_rate": 0.01,
+      "loss": 2.0023,
+      "step": 60687
+    },
+    {
+      "epoch": 6.232285890326556,
+      "grad_norm": 0.039438262581825256,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 60690
+    },
+    {
+      "epoch": 6.232593961799138,
+      "grad_norm": 0.042470790445804596,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 60693
+    },
+    {
+      "epoch": 6.232902033271719,
+      "grad_norm": 0.05736652761697769,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 60696
+    },
+    {
+      "epoch": 6.233210104744301,
+      "grad_norm": 0.03797580674290657,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 60699
+    },
+    {
+      "epoch": 6.233518176216882,
+      "grad_norm": 0.042780667543411255,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 60702
+    },
+    {
+      "epoch": 6.233826247689464,
+      "grad_norm": 0.04742565006017685,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 60705
+    },
+    {
+      "epoch": 6.234134319162045,
+      "grad_norm": 0.10806404799222946,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 60708
+    },
+    {
+      "epoch": 6.234442390634627,
+      "grad_norm": 0.09392410516738892,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 60711
+    },
+    {
+      "epoch": 6.2347504621072085,
+      "grad_norm": 0.06125201657414436,
+      "learning_rate": 0.01,
+      "loss": 1.996,
+      "step": 60714
+    },
+    {
+      "epoch": 6.235058533579791,
+      "grad_norm": 0.08396472036838531,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 60717
+    },
+    {
+      "epoch": 6.235366605052372,
+      "grad_norm": 0.05900922417640686,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 60720
+    },
+    {
+      "epoch": 6.235674676524954,
+      "grad_norm": 0.10509319603443146,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 60723
+    },
+    {
+      "epoch": 6.235982747997536,
+      "grad_norm": 0.13138900697231293,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 60726
+    },
+    {
+      "epoch": 6.236290819470117,
+      "grad_norm": 0.05676103010773659,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 60729
+    },
+    {
+      "epoch": 6.236598890942699,
+      "grad_norm": 0.04155607149004936,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 60732
+    },
+    {
+      "epoch": 6.23690696241528,
+      "grad_norm": 0.03284106031060219,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 60735
+    },
+    {
+      "epoch": 6.237215033887862,
+      "grad_norm": 0.03673785179853439,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 60738
+    },
+    {
+      "epoch": 6.237523105360443,
+      "grad_norm": 0.04972091317176819,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 60741
+    },
+    {
+      "epoch": 6.2378311768330255,
+      "grad_norm": 0.06821681559085846,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 60744
+    },
+    {
+      "epoch": 6.238139248305607,
+      "grad_norm": 0.14047065377235413,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 60747
+    },
+    {
+      "epoch": 6.238447319778189,
+      "grad_norm": 0.09423762559890747,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 60750
+    },
+    {
+      "epoch": 6.23875539125077,
+      "grad_norm": 0.08889120817184448,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 60753
+    },
+    {
+      "epoch": 6.239063462723352,
+      "grad_norm": 0.07696396857500076,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 60756
+    },
+    {
+      "epoch": 6.239371534195933,
+      "grad_norm": 0.13203272223472595,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 60759
+    },
+    {
+      "epoch": 6.239679605668515,
+      "grad_norm": 0.07637298852205276,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 60762
+    },
+    {
+      "epoch": 6.239987677141097,
+      "grad_norm": 0.035956770181655884,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 60765
+    },
+    {
+      "epoch": 6.240295748613678,
+      "grad_norm": 0.051339924335479736,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 60768
+    },
+    {
+      "epoch": 6.24060382008626,
+      "grad_norm": 0.05336092412471771,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 60771
+    },
+    {
+      "epoch": 6.2409118915588415,
+      "grad_norm": 0.061884574592113495,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 60774
+    },
+    {
+      "epoch": 6.241219963031424,
+      "grad_norm": 0.0763407051563263,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 60777
+    },
+    {
+      "epoch": 6.241528034504005,
+      "grad_norm": 0.06376196444034576,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 60780
+    },
+    {
+      "epoch": 6.241836105976587,
+      "grad_norm": 0.04951758310198784,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 60783
+    },
+    {
+      "epoch": 6.242144177449168,
+      "grad_norm": 0.03801729902625084,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 60786
+    },
+    {
+      "epoch": 6.24245224892175,
+      "grad_norm": 0.05235125496983528,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 60789
+    },
+    {
+      "epoch": 6.242760320394331,
+      "grad_norm": 0.09875135123729706,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 60792
+    },
+    {
+      "epoch": 6.243068391866913,
+      "grad_norm": 0.07917656749486923,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 60795
+    },
+    {
+      "epoch": 6.243376463339494,
+      "grad_norm": 0.09031876921653748,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 60798
+    },
+    {
+      "epoch": 6.243684534812076,
+      "grad_norm": 0.03974262252449989,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 60801
+    },
+    {
+      "epoch": 6.2439926062846585,
+      "grad_norm": 0.10321113467216492,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 60804
+    },
+    {
+      "epoch": 6.24430067775724,
+      "grad_norm": 0.041257478296756744,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 60807
+    },
+    {
+      "epoch": 6.244608749229822,
+      "grad_norm": 0.054699040949344635,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 60810
+    },
+    {
+      "epoch": 6.244916820702403,
+      "grad_norm": 0.06867776066064835,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 60813
+    },
+    {
+      "epoch": 6.245224892174985,
+      "grad_norm": 0.04628169536590576,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 60816
+    },
+    {
+      "epoch": 6.245532963647566,
+      "grad_norm": 0.04721110686659813,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 60819
+    },
+    {
+      "epoch": 6.245841035120148,
+      "grad_norm": 0.1265910118818283,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 60822
+    },
+    {
+      "epoch": 6.246149106592729,
+      "grad_norm": 0.09225190430879593,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 60825
+    },
+    {
+      "epoch": 6.246457178065311,
+      "grad_norm": 0.06333420425653458,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 60828
+    },
+    {
+      "epoch": 6.2467652495378925,
+      "grad_norm": 0.06048038601875305,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 60831
+    },
+    {
+      "epoch": 6.2470733210104745,
+      "grad_norm": 0.032611507922410965,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 60834
+    },
+    {
+      "epoch": 6.247381392483056,
+      "grad_norm": 0.038066837936639786,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 60837
+    },
+    {
+      "epoch": 6.247689463955638,
+      "grad_norm": 0.0952901542186737,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 60840
+    },
+    {
+      "epoch": 6.247997535428219,
+      "grad_norm": 0.051439084112644196,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 60843
+    },
+    {
+      "epoch": 6.248305606900801,
+      "grad_norm": 0.12137739360332489,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 60846
+    },
+    {
+      "epoch": 6.248613678373383,
+      "grad_norm": 0.04107406362891197,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 60849
+    },
+    {
+      "epoch": 6.248921749845964,
+      "grad_norm": 0.053470056504011154,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 60852
+    },
+    {
+      "epoch": 6.249229821318546,
+      "grad_norm": 0.10296319425106049,
+      "learning_rate": 0.01,
+      "loss": 2.0136,
+      "step": 60855
+    },
+    {
+      "epoch": 6.249537892791127,
+      "grad_norm": 0.05375886335968971,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 60858
+    },
+    {
+      "epoch": 6.249845964263709,
+      "grad_norm": 0.04653402417898178,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 60861
+    },
+    {
+      "epoch": 6.250154035736291,
+      "grad_norm": 0.04048166796565056,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 60864
+    },
+    {
+      "epoch": 6.250462107208873,
+      "grad_norm": 0.04350602254271507,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 60867
+    },
+    {
+      "epoch": 6.250770178681454,
+      "grad_norm": 0.06153450533747673,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 60870
+    },
+    {
+      "epoch": 6.251078250154036,
+      "grad_norm": 0.05672883987426758,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 60873
+    },
+    {
+      "epoch": 6.251386321626617,
+      "grad_norm": 0.04813450202345848,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 60876
+    },
+    {
+      "epoch": 6.251694393099199,
+      "grad_norm": 0.03838599845767021,
+      "learning_rate": 0.01,
+      "loss": 2.0072,
+      "step": 60879
+    },
+    {
+      "epoch": 6.252002464571781,
+      "grad_norm": 0.04607195779681206,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 60882
+    },
+    {
+      "epoch": 6.252310536044362,
+      "grad_norm": 0.06564761698246002,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 60885
+    },
+    {
+      "epoch": 6.252618607516944,
+      "grad_norm": 0.0656748041510582,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 60888
+    },
+    {
+      "epoch": 6.2529266789895255,
+      "grad_norm": 0.09336625039577484,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 60891
+    },
+    {
+      "epoch": 6.2532347504621075,
+      "grad_norm": 0.08100081235170364,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 60894
+    },
+    {
+      "epoch": 6.253542821934689,
+      "grad_norm": 0.06392189860343933,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 60897
+    },
+    {
+      "epoch": 6.253850893407271,
+      "grad_norm": 0.11884801834821701,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 60900
+    },
+    {
+      "epoch": 6.254158964879852,
+      "grad_norm": 0.08046597987413406,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 60903
+    },
+    {
+      "epoch": 6.254467036352434,
+      "grad_norm": 0.03278736025094986,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 60906
+    },
+    {
+      "epoch": 6.254775107825015,
+      "grad_norm": 0.043066561222076416,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 60909
+    },
+    {
+      "epoch": 6.255083179297597,
+      "grad_norm": 0.08042460680007935,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 60912
+    },
+    {
+      "epoch": 6.255391250770178,
+      "grad_norm": 0.05132247880101204,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 60915
+    },
+    {
+      "epoch": 6.25569932224276,
+      "grad_norm": 0.04719965532422066,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 60918
+    },
+    {
+      "epoch": 6.2560073937153415,
+      "grad_norm": 0.08247886598110199,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 60921
+    },
+    {
+      "epoch": 6.256315465187924,
+      "grad_norm": 0.0856601819396019,
+      "learning_rate": 0.01,
+      "loss": 2.0051,
+      "step": 60924
+    },
+    {
+      "epoch": 6.256623536660506,
+      "grad_norm": 0.06547616422176361,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 60927
+    },
+    {
+      "epoch": 6.256931608133087,
+      "grad_norm": 0.04583069309592247,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 60930
+    },
+    {
+      "epoch": 6.257239679605669,
+      "grad_norm": 0.1286034882068634,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 60933
+    },
+    {
+      "epoch": 6.25754775107825,
+      "grad_norm": 0.07549592107534409,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 60936
+    },
+    {
+      "epoch": 6.257855822550832,
+      "grad_norm": 0.06334639340639114,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 60939
+    },
+    {
+      "epoch": 6.258163894023413,
+      "grad_norm": 0.050806958228349686,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 60942
+    },
+    {
+      "epoch": 6.258471965495995,
+      "grad_norm": 0.03487928584218025,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 60945
+    },
+    {
+      "epoch": 6.258780036968576,
+      "grad_norm": 0.03992629423737526,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 60948
+    },
+    {
+      "epoch": 6.2590881084411585,
+      "grad_norm": 0.06908503919839859,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 60951
+    },
+    {
+      "epoch": 6.25939617991374,
+      "grad_norm": 0.10641307383775711,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 60954
+    },
+    {
+      "epoch": 6.259704251386322,
+      "grad_norm": 0.05809535086154938,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 60957
+    },
+    {
+      "epoch": 6.260012322858903,
+      "grad_norm": 0.1139754205942154,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 60960
+    },
+    {
+      "epoch": 6.260320394331485,
+      "grad_norm": 0.1501069813966751,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 60963
+    },
+    {
+      "epoch": 6.260628465804067,
+      "grad_norm": 0.10279611498117447,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 60966
+    },
+    {
+      "epoch": 6.260936537276648,
+      "grad_norm": 0.04817177355289459,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 60969
+    },
+    {
+      "epoch": 6.26124460874923,
+      "grad_norm": 0.036743566393852234,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 60972
+    },
+    {
+      "epoch": 6.261552680221811,
+      "grad_norm": 0.041490793228149414,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 60975
+    },
+    {
+      "epoch": 6.261860751694393,
+      "grad_norm": 0.034102845937013626,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 60978
+    },
+    {
+      "epoch": 6.2621688231669745,
+      "grad_norm": 0.04929140582680702,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 60981
+    },
+    {
+      "epoch": 6.262476894639557,
+      "grad_norm": 0.1550266444683075,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 60984
+    },
+    {
+      "epoch": 6.262784966112138,
+      "grad_norm": 0.1096281111240387,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 60987
+    },
+    {
+      "epoch": 6.26309303758472,
+      "grad_norm": 0.057930875569581985,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 60990
+    },
+    {
+      "epoch": 6.263401109057301,
+      "grad_norm": 0.04393153637647629,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 60993
+    },
+    {
+      "epoch": 6.263709180529883,
+      "grad_norm": 0.039023157209157944,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 60996
+    },
+    {
+      "epoch": 6.264017252002464,
+      "grad_norm": 0.05385271832346916,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 60999
+    },
+    {
+      "epoch": 6.264325323475046,
+      "grad_norm": 0.11487401276826859,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 61002
+    },
+    {
+      "epoch": 6.264633394947628,
+      "grad_norm": 0.12395527213811874,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 61005
+    },
+    {
+      "epoch": 6.264941466420209,
+      "grad_norm": 0.05898820608854294,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 61008
+    },
+    {
+      "epoch": 6.2652495378927915,
+      "grad_norm": 0.06743215769529343,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 61011
+    },
+    {
+      "epoch": 6.265557609365373,
+      "grad_norm": 0.04019298404455185,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 61014
+    },
+    {
+      "epoch": 6.265865680837955,
+      "grad_norm": 0.053353726863861084,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 61017
+    },
+    {
+      "epoch": 6.266173752310536,
+      "grad_norm": 0.05660318583250046,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 61020
+    },
+    {
+      "epoch": 6.266481823783118,
+      "grad_norm": 0.08507215976715088,
+      "learning_rate": 0.01,
+      "loss": 2.0229,
+      "step": 61023
+    },
+    {
+      "epoch": 6.266789895255699,
+      "grad_norm": 0.10339481383562088,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 61026
+    },
+    {
+      "epoch": 6.267097966728281,
+      "grad_norm": 0.06297086924314499,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 61029
+    },
+    {
+      "epoch": 6.267406038200862,
+      "grad_norm": 0.03891312703490257,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 61032
+    },
+    {
+      "epoch": 6.267714109673444,
+      "grad_norm": 0.05895547196269035,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 61035
+    },
+    {
+      "epoch": 6.2680221811460255,
+      "grad_norm": 0.058469437062740326,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 61038
+    },
+    {
+      "epoch": 6.2683302526186075,
+      "grad_norm": 0.04402352124452591,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 61041
+    },
+    {
+      "epoch": 6.268638324091189,
+      "grad_norm": 0.07434606552124023,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 61044
+    },
+    {
+      "epoch": 6.268946395563771,
+      "grad_norm": 0.07005775719881058,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 61047
+    },
+    {
+      "epoch": 6.269254467036353,
+      "grad_norm": 0.08956246078014374,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 61050
+    },
+    {
+      "epoch": 6.269562538508934,
+      "grad_norm": 0.04520926997065544,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 61053
+    },
+    {
+      "epoch": 6.269870609981516,
+      "grad_norm": 0.09920626133680344,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 61056
+    },
+    {
+      "epoch": 6.270178681454097,
+      "grad_norm": 0.10885010659694672,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 61059
+    },
+    {
+      "epoch": 6.270486752926679,
+      "grad_norm": 0.0662059485912323,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 61062
+    },
+    {
+      "epoch": 6.27079482439926,
+      "grad_norm": 0.09873577207326889,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 61065
+    },
+    {
+      "epoch": 6.271102895871842,
+      "grad_norm": 0.0979217141866684,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 61068
+    },
+    {
+      "epoch": 6.271410967344424,
+      "grad_norm": 0.06409544497728348,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 61071
+    },
+    {
+      "epoch": 6.271719038817006,
+      "grad_norm": 0.05898788571357727,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 61074
+    },
+    {
+      "epoch": 6.272027110289587,
+      "grad_norm": 0.048941727727651596,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 61077
+    },
+    {
+      "epoch": 6.272335181762169,
+      "grad_norm": 0.04335403069853783,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 61080
+    },
+    {
+      "epoch": 6.272643253234751,
+      "grad_norm": 0.08291465789079666,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 61083
+    },
+    {
+      "epoch": 6.272951324707332,
+      "grad_norm": 0.09848134219646454,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 61086
+    },
+    {
+      "epoch": 6.273259396179914,
+      "grad_norm": 0.10949241369962692,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 61089
+    },
+    {
+      "epoch": 6.273567467652495,
+      "grad_norm": 0.07039642333984375,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 61092
+    },
+    {
+      "epoch": 6.273875539125077,
+      "grad_norm": 0.0475379079580307,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 61095
+    },
+    {
+      "epoch": 6.2741836105976585,
+      "grad_norm": 0.07531355321407318,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 61098
+    },
+    {
+      "epoch": 6.2744916820702406,
+      "grad_norm": 0.04126337915658951,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 61101
+    },
+    {
+      "epoch": 6.274799753542822,
+      "grad_norm": 0.03969768062233925,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 61104
+    },
+    {
+      "epoch": 6.275107825015404,
+      "grad_norm": 0.06768359243869781,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 61107
+    },
+    {
+      "epoch": 6.275415896487985,
+      "grad_norm": 0.10613830387592316,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 61110
+    },
+    {
+      "epoch": 6.275723967960567,
+      "grad_norm": 0.09580209851264954,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 61113
+    },
+    {
+      "epoch": 6.276032039433148,
+      "grad_norm": 0.06177694723010063,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 61116
+    },
+    {
+      "epoch": 6.27634011090573,
+      "grad_norm": 0.10451437532901764,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 61119
+    },
+    {
+      "epoch": 6.276648182378311,
+      "grad_norm": 0.1352027803659439,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 61122
+    },
+    {
+      "epoch": 6.276956253850893,
+      "grad_norm": 0.0892416313290596,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 61125
+    },
+    {
+      "epoch": 6.2772643253234754,
+      "grad_norm": 0.04253193363547325,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 61128
+    },
+    {
+      "epoch": 6.277572396796057,
+      "grad_norm": 0.03616737946867943,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 61131
+    },
+    {
+      "epoch": 6.277880468268639,
+      "grad_norm": 0.09631824493408203,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 61134
+    },
+    {
+      "epoch": 6.27818853974122,
+      "grad_norm": 0.08627616614103317,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 61137
+    },
+    {
+      "epoch": 6.278496611213802,
+      "grad_norm": 0.07595238834619522,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 61140
+    },
+    {
+      "epoch": 6.278804682686383,
+      "grad_norm": 0.05115518346428871,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 61143
+    },
+    {
+      "epoch": 6.279112754158965,
+      "grad_norm": 0.07910732924938202,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 61146
+    },
+    {
+      "epoch": 6.279420825631546,
+      "grad_norm": 0.07935528457164764,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 61149
+    },
+    {
+      "epoch": 6.279728897104128,
+      "grad_norm": 0.09080187976360321,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 61152
+    },
+    {
+      "epoch": 6.2800369685767095,
+      "grad_norm": 0.07743295282125473,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 61155
+    },
+    {
+      "epoch": 6.2803450400492915,
+      "grad_norm": 0.07669807970523834,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 61158
+    },
+    {
+      "epoch": 6.280653111521873,
+      "grad_norm": 0.09190505743026733,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 61161
+    },
+    {
+      "epoch": 6.280961182994455,
+      "grad_norm": 0.052665840834379196,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 61164
+    },
+    {
+      "epoch": 6.281269254467037,
+      "grad_norm": 0.05133817344903946,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 61167
+    },
+    {
+      "epoch": 6.281577325939618,
+      "grad_norm": 0.038431648164987564,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 61170
+    },
+    {
+      "epoch": 6.2818853974122,
+      "grad_norm": 0.057378388941287994,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 61173
+    },
+    {
+      "epoch": 6.282193468884781,
+      "grad_norm": 0.08355475962162018,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 61176
+    },
+    {
+      "epoch": 6.282501540357363,
+      "grad_norm": 0.07666066288948059,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 61179
+    },
+    {
+      "epoch": 6.282809611829944,
+      "grad_norm": 0.12239743769168854,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 61182
+    },
+    {
+      "epoch": 6.283117683302526,
+      "grad_norm": 0.10445159673690796,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 61185
+    },
+    {
+      "epoch": 6.283425754775108,
+      "grad_norm": 0.04401460289955139,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 61188
+    },
+    {
+      "epoch": 6.28373382624769,
+      "grad_norm": 0.08247647434473038,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 61191
+    },
+    {
+      "epoch": 6.284041897720271,
+      "grad_norm": 0.06200701370835304,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 61194
+    },
+    {
+      "epoch": 6.284349969192853,
+      "grad_norm": 0.04233565926551819,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 61197
+    },
+    {
+      "epoch": 6.284658040665434,
+      "grad_norm": 0.09187982976436615,
+      "learning_rate": 0.01,
+      "loss": 1.9991,
+      "step": 61200
+    },
+    {
+      "epoch": 6.284966112138016,
+      "grad_norm": 0.03373527526855469,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 61203
+    },
+    {
+      "epoch": 6.285274183610598,
+      "grad_norm": 0.09450080990791321,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 61206
+    },
+    {
+      "epoch": 6.285582255083179,
+      "grad_norm": 0.048632752150297165,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 61209
+    },
+    {
+      "epoch": 6.285890326555761,
+      "grad_norm": 0.11564646661281586,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 61212
+    },
+    {
+      "epoch": 6.2861983980283425,
+      "grad_norm": 0.04203884303569794,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 61215
+    },
+    {
+      "epoch": 6.2865064695009245,
+      "grad_norm": 0.08716481178998947,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 61218
+    },
+    {
+      "epoch": 6.286814540973506,
+      "grad_norm": 0.07940181344747543,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 61221
+    },
+    {
+      "epoch": 6.287122612446088,
+      "grad_norm": 0.06720132380723953,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 61224
+    },
+    {
+      "epoch": 6.287430683918669,
+      "grad_norm": 0.05566677451133728,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 61227
+    },
+    {
+      "epoch": 6.287738755391251,
+      "grad_norm": 0.09876339137554169,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 61230
+    },
+    {
+      "epoch": 6.288046826863832,
+      "grad_norm": 0.054778262972831726,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 61233
+    },
+    {
+      "epoch": 6.288354898336414,
+      "grad_norm": 0.06283058226108551,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 61236
+    },
+    {
+      "epoch": 6.288662969808995,
+      "grad_norm": 0.054917361587285995,
+      "learning_rate": 0.01,
+      "loss": 2.0248,
+      "step": 61239
+    },
+    {
+      "epoch": 6.288971041281577,
+      "grad_norm": 0.10483105480670929,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 61242
+    },
+    {
+      "epoch": 6.2892791127541585,
+      "grad_norm": 0.07183388620615005,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 61245
+    },
+    {
+      "epoch": 6.289587184226741,
+      "grad_norm": 0.12913282215595245,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 61248
+    },
+    {
+      "epoch": 6.289895255699323,
+      "grad_norm": 0.1015811562538147,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 61251
+    },
+    {
+      "epoch": 6.290203327171904,
+      "grad_norm": 0.09693120419979095,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 61254
+    },
+    {
+      "epoch": 6.290511398644486,
+      "grad_norm": 0.05746886506676674,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 61257
+    },
+    {
+      "epoch": 6.290819470117067,
+      "grad_norm": 0.07819090038537979,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 61260
+    },
+    {
+      "epoch": 6.291127541589649,
+      "grad_norm": 0.039794620126485825,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 61263
+    },
+    {
+      "epoch": 6.29143561306223,
+      "grad_norm": 0.06474246084690094,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 61266
+    },
+    {
+      "epoch": 6.291743684534812,
+      "grad_norm": 0.05018305405974388,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 61269
+    },
+    {
+      "epoch": 6.292051756007393,
+      "grad_norm": 0.10312799364328384,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 61272
+    },
+    {
+      "epoch": 6.2923598274799755,
+      "grad_norm": 0.0430091917514801,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 61275
+    },
+    {
+      "epoch": 6.292667898952557,
+      "grad_norm": 0.08931942284107208,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 61278
+    },
+    {
+      "epoch": 6.292975970425139,
+      "grad_norm": 0.09583033621311188,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 61281
+    },
+    {
+      "epoch": 6.293284041897721,
+      "grad_norm": 0.048897985368967056,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 61284
+    },
+    {
+      "epoch": 6.293592113370302,
+      "grad_norm": 0.07638606429100037,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 61287
+    },
+    {
+      "epoch": 6.293900184842884,
+      "grad_norm": 0.07323414832353592,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 61290
+    },
+    {
+      "epoch": 6.294208256315465,
+      "grad_norm": 0.08963891863822937,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 61293
+    },
+    {
+      "epoch": 6.294516327788047,
+      "grad_norm": 0.047504641115665436,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 61296
+    },
+    {
+      "epoch": 6.294824399260628,
+      "grad_norm": 0.04849178344011307,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 61299
+    },
+    {
+      "epoch": 6.29513247073321,
+      "grad_norm": 0.06779339909553528,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 61302
+    },
+    {
+      "epoch": 6.2954405422057915,
+      "grad_norm": 0.03595037758350372,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 61305
+    },
+    {
+      "epoch": 6.295748613678374,
+      "grad_norm": 0.04232453927397728,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 61308
+    },
+    {
+      "epoch": 6.296056685150955,
+      "grad_norm": 0.03607799485325813,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 61311
+    },
+    {
+      "epoch": 6.296364756623537,
+      "grad_norm": 0.07026591151952744,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 61314
+    },
+    {
+      "epoch": 6.296672828096118,
+      "grad_norm": 0.09154605120420456,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 61317
+    },
+    {
+      "epoch": 6.2969808995687,
+      "grad_norm": 0.0732119083404541,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 61320
+    },
+    {
+      "epoch": 6.297288971041281,
+      "grad_norm": 0.12241560965776443,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 61323
+    },
+    {
+      "epoch": 6.297597042513863,
+      "grad_norm": 0.04552261903882027,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 61326
+    },
+    {
+      "epoch": 6.297905113986445,
+      "grad_norm": 0.10219983011484146,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 61329
+    },
+    {
+      "epoch": 6.298213185459026,
+      "grad_norm": 0.09614024311304092,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 61332
+    },
+    {
+      "epoch": 6.2985212569316085,
+      "grad_norm": 0.049760278314352036,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 61335
+    },
+    {
+      "epoch": 6.29882932840419,
+      "grad_norm": 0.11046284437179565,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 61338
+    },
+    {
+      "epoch": 6.299137399876772,
+      "grad_norm": 0.045198407024145126,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 61341
+    },
+    {
+      "epoch": 6.299445471349353,
+      "grad_norm": 0.0499146468937397,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 61344
+    },
+    {
+      "epoch": 6.299753542821935,
+      "grad_norm": 0.0700070783495903,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 61347
+    },
+    {
+      "epoch": 6.300061614294516,
+      "grad_norm": 0.042327769100666046,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 61350
+    },
+    {
+      "epoch": 6.300369685767098,
+      "grad_norm": 0.042714498937129974,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 61353
+    },
+    {
+      "epoch": 6.300677757239679,
+      "grad_norm": 0.06567323952913284,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 61356
+    },
+    {
+      "epoch": 6.300985828712261,
+      "grad_norm": 0.07993308454751968,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 61359
+    },
+    {
+      "epoch": 6.3012939001848425,
+      "grad_norm": 0.08032864332199097,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 61362
+    },
+    {
+      "epoch": 6.3016019716574245,
+      "grad_norm": 0.05398751050233841,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 61365
+    },
+    {
+      "epoch": 6.301910043130006,
+      "grad_norm": 0.04342971369624138,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 61368
+    },
+    {
+      "epoch": 6.302218114602588,
+      "grad_norm": 0.0811370387673378,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 61371
+    },
+    {
+      "epoch": 6.30252618607517,
+      "grad_norm": 0.09355277568101883,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 61374
+    },
+    {
+      "epoch": 6.302834257547751,
+      "grad_norm": 0.06557541340589523,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 61377
+    },
+    {
+      "epoch": 6.303142329020333,
+      "grad_norm": 0.038711193948984146,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 61380
+    },
+    {
+      "epoch": 6.303450400492914,
+      "grad_norm": 0.04032447189092636,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 61383
+    },
+    {
+      "epoch": 6.303758471965496,
+      "grad_norm": 0.03779391944408417,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 61386
+    },
+    {
+      "epoch": 6.304066543438077,
+      "grad_norm": 0.054389867931604385,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 61389
+    },
+    {
+      "epoch": 6.304374614910659,
+      "grad_norm": 0.0877804309129715,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 61392
+    },
+    {
+      "epoch": 6.304682686383241,
+      "grad_norm": 0.06545720249414444,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 61395
+    },
+    {
+      "epoch": 6.304990757855823,
+      "grad_norm": 0.10159642994403839,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 61398
+    },
+    {
+      "epoch": 6.305298829328404,
+      "grad_norm": 0.045314956456422806,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 61401
+    },
+    {
+      "epoch": 6.305606900800986,
+      "grad_norm": 0.1006140187382698,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 61404
+    },
+    {
+      "epoch": 6.305914972273568,
+      "grad_norm": 0.09133684635162354,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 61407
+    },
+    {
+      "epoch": 6.306223043746149,
+      "grad_norm": 0.11910711973905563,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 61410
+    },
+    {
+      "epoch": 6.306531115218731,
+      "grad_norm": 0.11194338649511337,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 61413
+    },
+    {
+      "epoch": 6.306839186691312,
+      "grad_norm": 0.050022371113300323,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 61416
+    },
+    {
+      "epoch": 6.307147258163894,
+      "grad_norm": 0.039385128766298294,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 61419
+    },
+    {
+      "epoch": 6.3074553296364755,
+      "grad_norm": 0.048881761729717255,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 61422
+    },
+    {
+      "epoch": 6.3077634011090575,
+      "grad_norm": 0.062211450189352036,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 61425
+    },
+    {
+      "epoch": 6.308071472581639,
+      "grad_norm": 0.047393426299095154,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 61428
+    },
+    {
+      "epoch": 6.308379544054221,
+      "grad_norm": 0.03558918833732605,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 61431
+    },
+    {
+      "epoch": 6.308687615526802,
+      "grad_norm": 0.05444757267832756,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 61434
+    },
+    {
+      "epoch": 6.308995686999384,
+      "grad_norm": 0.0579729862511158,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 61437
+    },
+    {
+      "epoch": 6.309303758471965,
+      "grad_norm": 0.10818339139223099,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 61440
+    },
+    {
+      "epoch": 6.309611829944547,
+      "grad_norm": 0.053433943539857864,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 61443
+    },
+    {
+      "epoch": 6.309919901417128,
+      "grad_norm": 0.09320402890443802,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 61446
+    },
+    {
+      "epoch": 6.31022797288971,
+      "grad_norm": 0.10474387556314468,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 61449
+    },
+    {
+      "epoch": 6.310536044362292,
+      "grad_norm": 0.059648603200912476,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 61452
+    },
+    {
+      "epoch": 6.310844115834874,
+      "grad_norm": 0.05086411163210869,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 61455
+    },
+    {
+      "epoch": 6.311152187307456,
+      "grad_norm": 0.05620935931801796,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 61458
+    },
+    {
+      "epoch": 6.311460258780037,
+      "grad_norm": 0.07844830304384232,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 61461
+    },
+    {
+      "epoch": 6.311768330252619,
+      "grad_norm": 0.1557673066854477,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 61464
+    },
+    {
+      "epoch": 6.3120764017252,
+      "grad_norm": 0.09628250449895859,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 61467
+    },
+    {
+      "epoch": 6.312384473197782,
+      "grad_norm": 0.045699939131736755,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 61470
+    },
+    {
+      "epoch": 6.312692544670363,
+      "grad_norm": 0.033145416527986526,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 61473
+    },
+    {
+      "epoch": 6.313000616142945,
+      "grad_norm": 0.03600894287228584,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 61476
+    },
+    {
+      "epoch": 6.313308687615526,
+      "grad_norm": 0.09642411768436432,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 61479
+    },
+    {
+      "epoch": 6.3136167590881085,
+      "grad_norm": 0.047719743102788925,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 61482
+    },
+    {
+      "epoch": 6.3139248305606905,
+      "grad_norm": 0.0641685277223587,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 61485
+    },
+    {
+      "epoch": 6.314232902033272,
+      "grad_norm": 0.06864326447248459,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 61488
+    },
+    {
+      "epoch": 6.314540973505854,
+      "grad_norm": 0.062220584601163864,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 61491
+    },
+    {
+      "epoch": 6.314849044978435,
+      "grad_norm": 0.0497710257768631,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 61494
+    },
+    {
+      "epoch": 6.315157116451017,
+      "grad_norm": 0.06263673305511475,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 61497
+    },
+    {
+      "epoch": 6.315465187923598,
+      "grad_norm": 0.1437608301639557,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 61500
+    },
+    {
+      "epoch": 6.31577325939618,
+      "grad_norm": 0.1427847445011139,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 61503
+    },
+    {
+      "epoch": 6.316081330868761,
+      "grad_norm": 0.06643825769424438,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 61506
+    },
+    {
+      "epoch": 6.316389402341343,
+      "grad_norm": 0.05207568779587746,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 61509
+    },
+    {
+      "epoch": 6.3166974738139245,
+      "grad_norm": 0.05009368434548378,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 61512
+    },
+    {
+      "epoch": 6.317005545286507,
+      "grad_norm": 0.060909830033779144,
+      "learning_rate": 0.01,
+      "loss": 2.0055,
+      "step": 61515
+    },
+    {
+      "epoch": 6.317313616759088,
+      "grad_norm": 0.050222769379615784,
+      "learning_rate": 0.01,
+      "loss": 2.0128,
+      "step": 61518
+    },
+    {
+      "epoch": 6.31762168823167,
+      "grad_norm": 0.03374829515814781,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 61521
+    },
+    {
+      "epoch": 6.317929759704251,
+      "grad_norm": 0.07113096117973328,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 61524
+    },
+    {
+      "epoch": 6.318237831176833,
+      "grad_norm": 0.061376191675662994,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 61527
+    },
+    {
+      "epoch": 6.318545902649415,
+      "grad_norm": 0.07323316484689713,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 61530
+    },
+    {
+      "epoch": 6.318853974121996,
+      "grad_norm": 0.07050851732492447,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 61533
+    },
+    {
+      "epoch": 6.319162045594578,
+      "grad_norm": 0.04889465495944023,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 61536
+    },
+    {
+      "epoch": 6.319470117067159,
+      "grad_norm": 0.10975257307291031,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 61539
+    },
+    {
+      "epoch": 6.3197781885397415,
+      "grad_norm": 0.06909262388944626,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 61542
+    },
+    {
+      "epoch": 6.320086260012323,
+      "grad_norm": 0.0708390325307846,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 61545
+    },
+    {
+      "epoch": 6.320394331484905,
+      "grad_norm": 0.04191325604915619,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 61548
+    },
+    {
+      "epoch": 6.320702402957486,
+      "grad_norm": 0.04027121514081955,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 61551
+    },
+    {
+      "epoch": 6.321010474430068,
+      "grad_norm": 0.03705594688653946,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 61554
+    },
+    {
+      "epoch": 6.321318545902649,
+      "grad_norm": 0.04737750440835953,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 61557
+    },
+    {
+      "epoch": 6.321626617375231,
+      "grad_norm": 0.1080993041396141,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 61560
+    },
+    {
+      "epoch": 6.321934688847812,
+      "grad_norm": 0.08178503066301346,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 61563
+    },
+    {
+      "epoch": 6.322242760320394,
+      "grad_norm": 0.09055802971124649,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 61566
+    },
+    {
+      "epoch": 6.3225508317929755,
+      "grad_norm": 0.03342537209391594,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 61569
+    },
+    {
+      "epoch": 6.3228589032655576,
+      "grad_norm": 0.12082437425851822,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 61572
+    },
+    {
+      "epoch": 6.32316697473814,
+      "grad_norm": 0.05053761973977089,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 61575
+    },
+    {
+      "epoch": 6.323475046210721,
+      "grad_norm": 0.0372505746781826,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 61578
+    },
+    {
+      "epoch": 6.323783117683303,
+      "grad_norm": 0.0640554130077362,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 61581
+    },
+    {
+      "epoch": 6.324091189155884,
+      "grad_norm": 0.04472392797470093,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 61584
+    },
+    {
+      "epoch": 6.324399260628466,
+      "grad_norm": 0.09325999766588211,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 61587
+    },
+    {
+      "epoch": 6.324707332101047,
+      "grad_norm": 0.07012490928173065,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 61590
+    },
+    {
+      "epoch": 6.325015403573629,
+      "grad_norm": 0.13324911892414093,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 61593
+    },
+    {
+      "epoch": 6.32532347504621,
+      "grad_norm": 0.074961818754673,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 61596
+    },
+    {
+      "epoch": 6.3256315465187924,
+      "grad_norm": 0.06711915135383606,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 61599
+    },
+    {
+      "epoch": 6.325939617991374,
+      "grad_norm": 0.12803415954113007,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 61602
+    },
+    {
+      "epoch": 6.326247689463956,
+      "grad_norm": 0.07260297238826752,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 61605
+    },
+    {
+      "epoch": 6.326555760936538,
+      "grad_norm": 0.16781297326087952,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 61608
+    },
+    {
+      "epoch": 6.326863832409119,
+      "grad_norm": 0.05119356885552406,
+      "learning_rate": 0.01,
+      "loss": 2.0217,
+      "step": 61611
+    },
+    {
+      "epoch": 6.327171903881701,
+      "grad_norm": 0.05765308439731598,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 61614
+    },
+    {
+      "epoch": 6.327479975354282,
+      "grad_norm": 0.039318110793828964,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 61617
+    },
+    {
+      "epoch": 6.327788046826864,
+      "grad_norm": 0.04655304551124573,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 61620
+    },
+    {
+      "epoch": 6.328096118299445,
+      "grad_norm": 0.08642113953828812,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 61623
+    },
+    {
+      "epoch": 6.328404189772027,
+      "grad_norm": 0.04844063147902489,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 61626
+    },
+    {
+      "epoch": 6.3287122612446085,
+      "grad_norm": 0.058742884546518326,
+      "learning_rate": 0.01,
+      "loss": 2.0026,
+      "step": 61629
+    },
+    {
+      "epoch": 6.3290203327171906,
+      "grad_norm": 0.036662664264440536,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 61632
+    },
+    {
+      "epoch": 6.329328404189772,
+      "grad_norm": 0.08772391080856323,
+      "learning_rate": 0.01,
+      "loss": 2.0158,
+      "step": 61635
+    },
+    {
+      "epoch": 6.329636475662354,
+      "grad_norm": 0.06502344459295273,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 61638
+    },
+    {
+      "epoch": 6.329944547134935,
+      "grad_norm": 0.06459985673427582,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 61641
+    },
+    {
+      "epoch": 6.330252618607517,
+      "grad_norm": 0.03769855946302414,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 61644
+    },
+    {
+      "epoch": 6.330560690080098,
+      "grad_norm": 0.046935372054576874,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 61647
+    },
+    {
+      "epoch": 6.33086876155268,
+      "grad_norm": 0.04765419661998749,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 61650
+    },
+    {
+      "epoch": 6.331176833025262,
+      "grad_norm": 0.07804597169160843,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 61653
+    },
+    {
+      "epoch": 6.331484904497843,
+      "grad_norm": 0.04642481356859207,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 61656
+    },
+    {
+      "epoch": 6.3317929759704255,
+      "grad_norm": 0.11623187363147736,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 61659
+    },
+    {
+      "epoch": 6.332101047443007,
+      "grad_norm": 0.08578827977180481,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 61662
+    },
+    {
+      "epoch": 6.332409118915589,
+      "grad_norm": 0.05141110345721245,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 61665
+    },
+    {
+      "epoch": 6.33271719038817,
+      "grad_norm": 0.03601311147212982,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 61668
+    },
+    {
+      "epoch": 6.333025261860752,
+      "grad_norm": 0.06763188540935516,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 61671
+    },
+    {
+      "epoch": 6.333333333333333,
+      "grad_norm": 0.08627673238515854,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 61674
+    },
+    {
+      "epoch": 6.333641404805915,
+      "grad_norm": 0.06787147372961044,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 61677
+    },
+    {
+      "epoch": 6.333949476278496,
+      "grad_norm": 0.08557649701833725,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 61680
+    },
+    {
+      "epoch": 6.334257547751078,
+      "grad_norm": 0.08303692191839218,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 61683
+    },
+    {
+      "epoch": 6.33456561922366,
+      "grad_norm": 0.06721941381692886,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 61686
+    },
+    {
+      "epoch": 6.3348736906962415,
+      "grad_norm": 0.11784909665584564,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 61689
+    },
+    {
+      "epoch": 6.335181762168824,
+      "grad_norm": 0.04680579900741577,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 61692
+    },
+    {
+      "epoch": 6.335489833641405,
+      "grad_norm": 0.10561893880367279,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 61695
+    },
+    {
+      "epoch": 6.335797905113987,
+      "grad_norm": 0.10209949314594269,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 61698
+    },
+    {
+      "epoch": 6.336105976586568,
+      "grad_norm": 0.0697985514998436,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 61701
+    },
+    {
+      "epoch": 6.33641404805915,
+      "grad_norm": 0.04770383983850479,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 61704
+    },
+    {
+      "epoch": 6.336722119531731,
+      "grad_norm": 0.03596745431423187,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 61707
+    },
+    {
+      "epoch": 6.337030191004313,
+      "grad_norm": 0.05980326980352402,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 61710
+    },
+    {
+      "epoch": 6.337338262476894,
+      "grad_norm": 0.04213361069560051,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 61713
+    },
+    {
+      "epoch": 6.337646333949476,
+      "grad_norm": 0.12276975810527802,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 61716
+    },
+    {
+      "epoch": 6.337954405422058,
+      "grad_norm": 0.04725951328873634,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 61719
+    },
+    {
+      "epoch": 6.33826247689464,
+      "grad_norm": 0.08252322673797607,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 61722
+    },
+    {
+      "epoch": 6.338570548367221,
+      "grad_norm": 0.12866702675819397,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 61725
+    },
+    {
+      "epoch": 6.338878619839803,
+      "grad_norm": 0.11567038297653198,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 61728
+    },
+    {
+      "epoch": 6.339186691312385,
+      "grad_norm": 0.048908136785030365,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 61731
+    },
+    {
+      "epoch": 6.339494762784966,
+      "grad_norm": 0.04868915677070618,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 61734
+    },
+    {
+      "epoch": 6.339802834257548,
+      "grad_norm": 0.03802793473005295,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 61737
+    },
+    {
+      "epoch": 6.340110905730129,
+      "grad_norm": 0.040892910212278366,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 61740
+    },
+    {
+      "epoch": 6.340418977202711,
+      "grad_norm": 0.057897284626960754,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 61743
+    },
+    {
+      "epoch": 6.3407270486752925,
+      "grad_norm": 0.034404922276735306,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 61746
+    },
+    {
+      "epoch": 6.3410351201478745,
+      "grad_norm": 0.05493491142988205,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 61749
+    },
+    {
+      "epoch": 6.341343191620456,
+      "grad_norm": 0.09743682295084,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 61752
+    },
+    {
+      "epoch": 6.341651263093038,
+      "grad_norm": 0.06471186131238937,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 61755
+    },
+    {
+      "epoch": 6.341959334565619,
+      "grad_norm": 0.07594270259141922,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 61758
+    },
+    {
+      "epoch": 6.342267406038201,
+      "grad_norm": 0.05741607025265694,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 61761
+    },
+    {
+      "epoch": 6.342575477510782,
+      "grad_norm": 0.07117846608161926,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 61764
+    },
+    {
+      "epoch": 6.342883548983364,
+      "grad_norm": 0.08182870596647263,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 61767
+    },
+    {
+      "epoch": 6.343191620455945,
+      "grad_norm": 0.06772911548614502,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 61770
+    },
+    {
+      "epoch": 6.343499691928527,
+      "grad_norm": 0.04902360588312149,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 61773
+    },
+    {
+      "epoch": 6.343807763401109,
+      "grad_norm": 0.12090668827295303,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 61776
+    },
+    {
+      "epoch": 6.344115834873691,
+      "grad_norm": 0.04338901489973068,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 61779
+    },
+    {
+      "epoch": 6.344423906346273,
+      "grad_norm": 0.07592718303203583,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 61782
+    },
+    {
+      "epoch": 6.344731977818854,
+      "grad_norm": 0.0540313795208931,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 61785
+    },
+    {
+      "epoch": 6.345040049291436,
+      "grad_norm": 0.10975626856088638,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 61788
+    },
+    {
+      "epoch": 6.345348120764017,
+      "grad_norm": 0.1409357637166977,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 61791
+    },
+    {
+      "epoch": 6.345656192236599,
+      "grad_norm": 0.10727324336767197,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 61794
+    },
+    {
+      "epoch": 6.34596426370918,
+      "grad_norm": 0.08473125845193863,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 61797
+    },
+    {
+      "epoch": 6.346272335181762,
+      "grad_norm": 0.06947775185108185,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 61800
+    },
+    {
+      "epoch": 6.346580406654343,
+      "grad_norm": 0.06887286901473999,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 61803
+    },
+    {
+      "epoch": 6.3468884781269255,
+      "grad_norm": 0.08204744011163712,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 61806
+    },
+    {
+      "epoch": 6.3471965495995075,
+      "grad_norm": 0.06321869790554047,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 61809
+    },
+    {
+      "epoch": 6.347504621072089,
+      "grad_norm": 0.06997507065534592,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 61812
+    },
+    {
+      "epoch": 6.347812692544671,
+      "grad_norm": 0.04468904808163643,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 61815
+    },
+    {
+      "epoch": 6.348120764017252,
+      "grad_norm": 0.07359863072633743,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 61818
+    },
+    {
+      "epoch": 6.348428835489834,
+      "grad_norm": 0.08013535290956497,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 61821
+    },
+    {
+      "epoch": 6.348736906962415,
+      "grad_norm": 0.08246944099664688,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 61824
+    },
+    {
+      "epoch": 6.349044978434997,
+      "grad_norm": 0.05869217962026596,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 61827
+    },
+    {
+      "epoch": 6.349353049907578,
+      "grad_norm": 0.04672044888138771,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 61830
+    },
+    {
+      "epoch": 6.34966112138016,
+      "grad_norm": 0.03837782144546509,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 61833
+    },
+    {
+      "epoch": 6.3499691928527415,
+      "grad_norm": 0.04574459418654442,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 61836
+    },
+    {
+      "epoch": 6.350277264325324,
+      "grad_norm": 0.03845152631402016,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 61839
+    },
+    {
+      "epoch": 6.350585335797905,
+      "grad_norm": 0.03869105130434036,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 61842
+    },
+    {
+      "epoch": 6.350893407270487,
+      "grad_norm": 0.08753649145364761,
+      "learning_rate": 0.01,
+      "loss": 2.0112,
+      "step": 61845
+    },
+    {
+      "epoch": 6.351201478743068,
+      "grad_norm": 0.05175444483757019,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 61848
+    },
+    {
+      "epoch": 6.35150955021565,
+      "grad_norm": 0.08007702976465225,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 61851
+    },
+    {
+      "epoch": 6.351817621688232,
+      "grad_norm": 0.08175135403871536,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 61854
+    },
+    {
+      "epoch": 6.352125693160813,
+      "grad_norm": 0.042124420404434204,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 61857
+    },
+    {
+      "epoch": 6.352433764633395,
+      "grad_norm": 0.030145036056637764,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 61860
+    },
+    {
+      "epoch": 6.352741836105976,
+      "grad_norm": 0.05578388646245003,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 61863
+    },
+    {
+      "epoch": 6.3530499075785585,
+      "grad_norm": 0.06224137172102928,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 61866
+    },
+    {
+      "epoch": 6.35335797905114,
+      "grad_norm": 0.03723108768463135,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 61869
+    },
+    {
+      "epoch": 6.353666050523722,
+      "grad_norm": 0.05681099742650986,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 61872
+    },
+    {
+      "epoch": 6.353974121996303,
+      "grad_norm": 0.06745577603578568,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 61875
+    },
+    {
+      "epoch": 6.354282193468885,
+      "grad_norm": 0.17442718148231506,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 61878
+    },
+    {
+      "epoch": 6.354590264941466,
+      "grad_norm": 0.07263945788145065,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 61881
+    },
+    {
+      "epoch": 6.354898336414048,
+      "grad_norm": 0.047694019973278046,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 61884
+    },
+    {
+      "epoch": 6.35520640788663,
+      "grad_norm": 0.03988689184188843,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 61887
+    },
+    {
+      "epoch": 6.355514479359211,
+      "grad_norm": 0.06448500603437424,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 61890
+    },
+    {
+      "epoch": 6.355822550831793,
+      "grad_norm": 0.09949947893619537,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 61893
+    },
+    {
+      "epoch": 6.3561306223043745,
+      "grad_norm": 0.04593118652701378,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 61896
+    },
+    {
+      "epoch": 6.356438693776957,
+      "grad_norm": 0.05074120685458183,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 61899
+    },
+    {
+      "epoch": 6.356746765249538,
+      "grad_norm": 0.044762320816516876,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 61902
+    },
+    {
+      "epoch": 6.35705483672212,
+      "grad_norm": 0.04705975204706192,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 61905
+    },
+    {
+      "epoch": 6.357362908194701,
+      "grad_norm": 0.03957728296518326,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 61908
+    },
+    {
+      "epoch": 6.357670979667283,
+      "grad_norm": 0.050351545214653015,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 61911
+    },
+    {
+      "epoch": 6.357979051139864,
+      "grad_norm": 0.05305488035082817,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 61914
+    },
+    {
+      "epoch": 6.358287122612446,
+      "grad_norm": 0.03815659508109093,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 61917
+    },
+    {
+      "epoch": 6.358595194085027,
+      "grad_norm": 0.09699538350105286,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 61920
+    },
+    {
+      "epoch": 6.358903265557609,
+      "grad_norm": 0.13865162432193756,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 61923
+    },
+    {
+      "epoch": 6.359211337030191,
+      "grad_norm": 0.0713433250784874,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 61926
+    },
+    {
+      "epoch": 6.359519408502773,
+      "grad_norm": 0.06301635503768921,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 61929
+    },
+    {
+      "epoch": 6.359827479975355,
+      "grad_norm": 0.06763561815023422,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 61932
+    },
+    {
+      "epoch": 6.360135551447936,
+      "grad_norm": 0.051800407469272614,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 61935
+    },
+    {
+      "epoch": 6.360443622920518,
+      "grad_norm": 0.07054926455020905,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 61938
+    },
+    {
+      "epoch": 6.360751694393099,
+      "grad_norm": 0.09424702078104019,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 61941
+    },
+    {
+      "epoch": 6.361059765865681,
+      "grad_norm": 0.06715767085552216,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 61944
+    },
+    {
+      "epoch": 6.361367837338262,
+      "grad_norm": 0.052323173731565475,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 61947
+    },
+    {
+      "epoch": 6.361675908810844,
+      "grad_norm": 0.04951905459165573,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 61950
+    },
+    {
+      "epoch": 6.3619839802834255,
+      "grad_norm": 0.056527990847826004,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 61953
+    },
+    {
+      "epoch": 6.3622920517560075,
+      "grad_norm": 0.04317193478345871,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 61956
+    },
+    {
+      "epoch": 6.362600123228589,
+      "grad_norm": 0.07142667472362518,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 61959
+    },
+    {
+      "epoch": 6.362908194701171,
+      "grad_norm": 0.087591253221035,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 61962
+    },
+    {
+      "epoch": 6.363216266173752,
+      "grad_norm": 0.044100482016801834,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 61965
+    },
+    {
+      "epoch": 6.363524337646334,
+      "grad_norm": 0.05616932362318039,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 61968
+    },
+    {
+      "epoch": 6.363832409118915,
+      "grad_norm": 0.05864831432700157,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 61971
+    },
+    {
+      "epoch": 6.364140480591497,
+      "grad_norm": 0.04249054566025734,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 61974
+    },
+    {
+      "epoch": 6.364448552064079,
+      "grad_norm": 0.047203633934259415,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 61977
+    },
+    {
+      "epoch": 6.36475662353666,
+      "grad_norm": 0.0450938381254673,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 61980
+    },
+    {
+      "epoch": 6.365064695009242,
+      "grad_norm": 0.13708913326263428,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 61983
+    },
+    {
+      "epoch": 6.365372766481824,
+      "grad_norm": 0.0903947725892067,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 61986
+    },
+    {
+      "epoch": 6.365680837954406,
+      "grad_norm": 0.039598993957042694,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 61989
+    },
+    {
+      "epoch": 6.365988909426987,
+      "grad_norm": 0.06196282058954239,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 61992
+    },
+    {
+      "epoch": 6.366296980899569,
+      "grad_norm": 0.058830760419368744,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 61995
+    },
+    {
+      "epoch": 6.36660505237215,
+      "grad_norm": 0.03394271805882454,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 61998
+    },
+    {
+      "epoch": 6.366913123844732,
+      "grad_norm": 0.11069590598344803,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 62001
+    },
+    {
+      "epoch": 6.367221195317313,
+      "grad_norm": 0.03749784827232361,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 62004
+    },
+    {
+      "epoch": 6.367529266789895,
+      "grad_norm": 0.06352918595075607,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 62007
+    },
+    {
+      "epoch": 6.367837338262477,
+      "grad_norm": 0.037357673048973083,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 62010
+    },
+    {
+      "epoch": 6.3681454097350585,
+      "grad_norm": 0.09099667519330978,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 62013
+    },
+    {
+      "epoch": 6.3684534812076405,
+      "grad_norm": 0.054853178560733795,
+      "learning_rate": 0.01,
+      "loss": 2.0031,
+      "step": 62016
+    },
+    {
+      "epoch": 6.368761552680222,
+      "grad_norm": 0.04815756157040596,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 62019
+    },
+    {
+      "epoch": 6.369069624152804,
+      "grad_norm": 0.12704138457775116,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 62022
+    },
+    {
+      "epoch": 6.369377695625385,
+      "grad_norm": 0.1324865221977234,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 62025
+    },
+    {
+      "epoch": 6.369685767097967,
+      "grad_norm": 0.10938073694705963,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 62028
+    },
+    {
+      "epoch": 6.369993838570548,
+      "grad_norm": 0.055503327399492264,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 62031
+    },
+    {
+      "epoch": 6.37030191004313,
+      "grad_norm": 0.049480196088552475,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 62034
+    },
+    {
+      "epoch": 6.370609981515711,
+      "grad_norm": 0.04294247552752495,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 62037
+    },
+    {
+      "epoch": 6.370918052988293,
+      "grad_norm": 0.03694196790456772,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 62040
+    },
+    {
+      "epoch": 6.3712261244608746,
+      "grad_norm": 0.08281165361404419,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 62043
+    },
+    {
+      "epoch": 6.371534195933457,
+      "grad_norm": 0.05012732744216919,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 62046
+    },
+    {
+      "epoch": 6.371842267406038,
+      "grad_norm": 0.06646348536014557,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 62049
+    },
+    {
+      "epoch": 6.37215033887862,
+      "grad_norm": 0.046320248395204544,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 62052
+    },
+    {
+      "epoch": 6.372458410351202,
+      "grad_norm": 0.13739866018295288,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 62055
+    },
+    {
+      "epoch": 6.372766481823783,
+      "grad_norm": 0.10860442370176315,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 62058
+    },
+    {
+      "epoch": 6.373074553296365,
+      "grad_norm": 0.1861836165189743,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 62061
+    },
+    {
+      "epoch": 6.373382624768946,
+      "grad_norm": 0.07088860869407654,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 62064
+    },
+    {
+      "epoch": 6.373690696241528,
+      "grad_norm": 0.055799081921577454,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 62067
+    },
+    {
+      "epoch": 6.3739987677141094,
+      "grad_norm": 0.08463507890701294,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 62070
+    },
+    {
+      "epoch": 6.3743068391866915,
+      "grad_norm": 0.07785413414239883,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 62073
+    },
+    {
+      "epoch": 6.374614910659273,
+      "grad_norm": 0.042249925434589386,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 62076
+    },
+    {
+      "epoch": 6.374922982131855,
+      "grad_norm": 0.08121603727340698,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 62079
+    },
+    {
+      "epoch": 6.375231053604436,
+      "grad_norm": 0.08757340908050537,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 62082
+    },
+    {
+      "epoch": 6.375539125077018,
+      "grad_norm": 0.11189188808202744,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 62085
+    },
+    {
+      "epoch": 6.3758471965496,
+      "grad_norm": 0.04559013247489929,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 62088
+    },
+    {
+      "epoch": 6.376155268022181,
+      "grad_norm": 0.044948089867830276,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 62091
+    },
+    {
+      "epoch": 6.376463339494763,
+      "grad_norm": 0.030356265604496002,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 62094
+    },
+    {
+      "epoch": 6.376771410967344,
+      "grad_norm": 0.12627916038036346,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 62097
+    },
+    {
+      "epoch": 6.377079482439926,
+      "grad_norm": 0.046097807586193085,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 62100
+    },
+    {
+      "epoch": 6.377387553912508,
+      "grad_norm": 0.057985857129096985,
+      "learning_rate": 0.01,
+      "loss": 2.0091,
+      "step": 62103
+    },
+    {
+      "epoch": 6.37769562538509,
+      "grad_norm": 0.07194691151380539,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 62106
+    },
+    {
+      "epoch": 6.378003696857671,
+      "grad_norm": 0.07052842527627945,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 62109
+    },
+    {
+      "epoch": 6.378311768330253,
+      "grad_norm": 0.0420130155980587,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 62112
+    },
+    {
+      "epoch": 6.378619839802834,
+      "grad_norm": 0.08693109452724457,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 62115
+    },
+    {
+      "epoch": 6.378927911275416,
+      "grad_norm": 0.05671975761651993,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 62118
+    },
+    {
+      "epoch": 6.379235982747997,
+      "grad_norm": 0.04355829954147339,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 62121
+    },
+    {
+      "epoch": 6.379544054220579,
+      "grad_norm": 0.045782145112752914,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 62124
+    },
+    {
+      "epoch": 6.37985212569316,
+      "grad_norm": 0.07497083395719528,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 62127
+    },
+    {
+      "epoch": 6.3801601971657425,
+      "grad_norm": 0.067107655107975,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 62130
+    },
+    {
+      "epoch": 6.3804682686383245,
+      "grad_norm": 0.10962393879890442,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 62133
+    },
+    {
+      "epoch": 6.380776340110906,
+      "grad_norm": 0.128241628408432,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 62136
+    },
+    {
+      "epoch": 6.381084411583488,
+      "grad_norm": 0.03978749364614487,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 62139
+    },
+    {
+      "epoch": 6.381392483056069,
+      "grad_norm": 0.0986032783985138,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 62142
+    },
+    {
+      "epoch": 6.381700554528651,
+      "grad_norm": 0.05384555831551552,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 62145
+    },
+    {
+      "epoch": 6.382008626001232,
+      "grad_norm": 0.08981155604124069,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 62148
+    },
+    {
+      "epoch": 6.382316697473814,
+      "grad_norm": 0.14535114169120789,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 62151
+    },
+    {
+      "epoch": 6.382624768946395,
+      "grad_norm": 0.11009112000465393,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 62154
+    },
+    {
+      "epoch": 6.382932840418977,
+      "grad_norm": 0.07238960266113281,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 62157
+    },
+    {
+      "epoch": 6.3832409118915585,
+      "grad_norm": 0.0916430875658989,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 62160
+    },
+    {
+      "epoch": 6.383548983364141,
+      "grad_norm": 0.0457562655210495,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 62163
+    },
+    {
+      "epoch": 6.383857054836722,
+      "grad_norm": 0.09100457280874252,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 62166
+    },
+    {
+      "epoch": 6.384165126309304,
+      "grad_norm": 0.06494874507188797,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 62169
+    },
+    {
+      "epoch": 6.384473197781885,
+      "grad_norm": 0.06534988433122635,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 62172
+    },
+    {
+      "epoch": 6.384781269254467,
+      "grad_norm": 0.04221831634640694,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 62175
+    },
+    {
+      "epoch": 6.385089340727049,
+      "grad_norm": 0.1350170373916626,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 62178
+    },
+    {
+      "epoch": 6.38539741219963,
+      "grad_norm": 0.10296313464641571,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 62181
+    },
+    {
+      "epoch": 6.385705483672212,
+      "grad_norm": 0.052958227694034576,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 62184
+    },
+    {
+      "epoch": 6.386013555144793,
+      "grad_norm": 0.046673450618982315,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 62187
+    },
+    {
+      "epoch": 6.3863216266173755,
+      "grad_norm": 0.052400704473257065,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 62190
+    },
+    {
+      "epoch": 6.386629698089957,
+      "grad_norm": 0.05483454838395119,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 62193
+    },
+    {
+      "epoch": 6.386937769562539,
+      "grad_norm": 0.04824971407651901,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 62196
+    },
+    {
+      "epoch": 6.38724584103512,
+      "grad_norm": 0.04102877154946327,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 62199
+    },
+    {
+      "epoch": 6.387553912507702,
+      "grad_norm": 0.052934035658836365,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 62202
+    },
+    {
+      "epoch": 6.387861983980283,
+      "grad_norm": 0.13166899979114532,
+      "learning_rate": 0.01,
+      "loss": 2.0194,
+      "step": 62205
+    },
+    {
+      "epoch": 6.388170055452865,
+      "grad_norm": 0.06457889080047607,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 62208
+    },
+    {
+      "epoch": 6.388478126925447,
+      "grad_norm": 0.061608802527189255,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 62211
+    },
+    {
+      "epoch": 6.388786198398028,
+      "grad_norm": 0.03722333163022995,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 62214
+    },
+    {
+      "epoch": 6.38909426987061,
+      "grad_norm": 0.03712205961346626,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 62217
+    },
+    {
+      "epoch": 6.3894023413431915,
+      "grad_norm": 0.10517171770334244,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 62220
+    },
+    {
+      "epoch": 6.389710412815774,
+      "grad_norm": 0.07701227068901062,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 62223
+    },
+    {
+      "epoch": 6.390018484288355,
+      "grad_norm": 0.13547784090042114,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 62226
+    },
+    {
+      "epoch": 6.390326555760937,
+      "grad_norm": 0.11633336544036865,
+      "learning_rate": 0.01,
+      "loss": 2.0137,
+      "step": 62229
+    },
+    {
+      "epoch": 6.390634627233518,
+      "grad_norm": 0.09636779129505157,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 62232
+    },
+    {
+      "epoch": 6.3909426987061,
+      "grad_norm": 0.08367501944303513,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 62235
+    },
+    {
+      "epoch": 6.391250770178681,
+      "grad_norm": 0.06181328743696213,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 62238
+    },
+    {
+      "epoch": 6.391558841651263,
+      "grad_norm": 0.08050753176212311,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 62241
+    },
+    {
+      "epoch": 6.391866913123844,
+      "grad_norm": 0.04931900277733803,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 62244
+    },
+    {
+      "epoch": 6.392174984596426,
+      "grad_norm": 0.050407491624355316,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 62247
+    },
+    {
+      "epoch": 6.392483056069008,
+      "grad_norm": 0.03501970320940018,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 62250
+    },
+    {
+      "epoch": 6.39279112754159,
+      "grad_norm": 0.04515289515256882,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 62253
+    },
+    {
+      "epoch": 6.393099199014172,
+      "grad_norm": 0.07897801697254181,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 62256
+    },
+    {
+      "epoch": 6.393407270486753,
+      "grad_norm": 0.09098079800605774,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 62259
+    },
+    {
+      "epoch": 6.393715341959335,
+      "grad_norm": 0.046846143901348114,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 62262
+    },
+    {
+      "epoch": 6.394023413431916,
+      "grad_norm": 0.08435478061437607,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 62265
+    },
+    {
+      "epoch": 6.394331484904498,
+      "grad_norm": 0.04902444779872894,
+      "learning_rate": 0.01,
+      "loss": 2.0083,
+      "step": 62268
+    },
+    {
+      "epoch": 6.394639556377079,
+      "grad_norm": 0.053175996989011765,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 62271
+    },
+    {
+      "epoch": 6.394947627849661,
+      "grad_norm": 0.04319942370057106,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 62274
+    },
+    {
+      "epoch": 6.3952556993222425,
+      "grad_norm": 0.056976351886987686,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 62277
+    },
+    {
+      "epoch": 6.3955637707948245,
+      "grad_norm": 0.12992046773433685,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 62280
+    },
+    {
+      "epoch": 6.395871842267406,
+      "grad_norm": 0.0788559541106224,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 62283
+    },
+    {
+      "epoch": 6.396179913739988,
+      "grad_norm": 0.039425671100616455,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 62286
+    },
+    {
+      "epoch": 6.39648798521257,
+      "grad_norm": 0.0928255170583725,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 62289
+    },
+    {
+      "epoch": 6.396796056685151,
+      "grad_norm": 0.06437569856643677,
+      "learning_rate": 0.01,
+      "loss": 2.011,
+      "step": 62292
+    },
+    {
+      "epoch": 6.397104128157733,
+      "grad_norm": 0.05675747990608215,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 62295
+    },
+    {
+      "epoch": 6.397412199630314,
+      "grad_norm": 0.06161409988999367,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 62298
+    },
+    {
+      "epoch": 6.397720271102896,
+      "grad_norm": 0.035643287003040314,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 62301
+    },
+    {
+      "epoch": 6.398028342575477,
+      "grad_norm": 0.06932775676250458,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 62304
+    },
+    {
+      "epoch": 6.398336414048059,
+      "grad_norm": 0.1191667765378952,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 62307
+    },
+    {
+      "epoch": 6.398644485520641,
+      "grad_norm": 0.06702014803886414,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 62310
+    },
+    {
+      "epoch": 6.398952556993223,
+      "grad_norm": 0.045102011412382126,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 62313
+    },
+    {
+      "epoch": 6.399260628465804,
+      "grad_norm": 0.042483970522880554,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 62316
+    },
+    {
+      "epoch": 6.399568699938386,
+      "grad_norm": 0.035098928958177567,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 62319
+    },
+    {
+      "epoch": 6.399876771410967,
+      "grad_norm": 0.05023285374045372,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 62322
+    },
+    {
+      "epoch": 6.400184842883549,
+      "grad_norm": 0.14267301559448242,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 62325
+    },
+    {
+      "epoch": 6.40049291435613,
+      "grad_norm": 0.12871447205543518,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 62328
+    },
+    {
+      "epoch": 6.400800985828712,
+      "grad_norm": 0.18785615265369415,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 62331
+    },
+    {
+      "epoch": 6.401109057301294,
+      "grad_norm": 0.14859537780284882,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 62334
+    },
+    {
+      "epoch": 6.4014171287738755,
+      "grad_norm": 0.10787920653820038,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 62337
+    },
+    {
+      "epoch": 6.4017252002464575,
+      "grad_norm": 0.0459236241877079,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 62340
+    },
+    {
+      "epoch": 6.402033271719039,
+      "grad_norm": 0.04476577043533325,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 62343
+    },
+    {
+      "epoch": 6.402341343191621,
+      "grad_norm": 0.04856594651937485,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 62346
+    },
+    {
+      "epoch": 6.402649414664202,
+      "grad_norm": 0.047260161489248276,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 62349
+    },
+    {
+      "epoch": 6.402957486136784,
+      "grad_norm": 0.05097677931189537,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 62352
+    },
+    {
+      "epoch": 6.403265557609365,
+      "grad_norm": 0.10697360336780548,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 62355
+    },
+    {
+      "epoch": 6.403573629081947,
+      "grad_norm": 0.11435877531766891,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 62358
+    },
+    {
+      "epoch": 6.403881700554528,
+      "grad_norm": 0.036153070628643036,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 62361
+    },
+    {
+      "epoch": 6.40418977202711,
+      "grad_norm": 0.07002195715904236,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 62364
+    },
+    {
+      "epoch": 6.4044978434996915,
+      "grad_norm": 0.05492027476429939,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 62367
+    },
+    {
+      "epoch": 6.404805914972274,
+      "grad_norm": 0.06763515621423721,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 62370
+    },
+    {
+      "epoch": 6.405113986444855,
+      "grad_norm": 0.03954880312085152,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 62373
+    },
+    {
+      "epoch": 6.405422057917437,
+      "grad_norm": 0.06005239859223366,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 62376
+    },
+    {
+      "epoch": 6.405730129390019,
+      "grad_norm": 0.06014340743422508,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 62379
+    },
+    {
+      "epoch": 6.4060382008626,
+      "grad_norm": 0.05861683562397957,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 62382
+    },
+    {
+      "epoch": 6.406346272335182,
+      "grad_norm": 0.07609166949987411,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 62385
+    },
+    {
+      "epoch": 6.406654343807763,
+      "grad_norm": 0.08934800326824188,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 62388
+    },
+    {
+      "epoch": 6.406962415280345,
+      "grad_norm": 0.14556893706321716,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 62391
+    },
+    {
+      "epoch": 6.407270486752926,
+      "grad_norm": 0.07740515470504761,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 62394
+    },
+    {
+      "epoch": 6.4075785582255085,
+      "grad_norm": 0.04366164281964302,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 62397
+    },
+    {
+      "epoch": 6.40788662969809,
+      "grad_norm": 0.04770512506365776,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 62400
+    },
+    {
+      "epoch": 6.408194701170672,
+      "grad_norm": 0.04657142981886864,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 62403
+    },
+    {
+      "epoch": 6.408502772643253,
+      "grad_norm": 0.04207182675600052,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 62406
+    },
+    {
+      "epoch": 6.408810844115835,
+      "grad_norm": 0.05520906671881676,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 62409
+    },
+    {
+      "epoch": 6.409118915588417,
+      "grad_norm": 0.07524003833532333,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 62412
+    },
+    {
+      "epoch": 6.409426987060998,
+      "grad_norm": 0.04558379203081131,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 62415
+    },
+    {
+      "epoch": 6.40973505853358,
+      "grad_norm": 0.09892844408750534,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 62418
+    },
+    {
+      "epoch": 6.410043130006161,
+      "grad_norm": 0.07591421902179718,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 62421
+    },
+    {
+      "epoch": 6.410351201478743,
+      "grad_norm": 0.09021826833486557,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 62424
+    },
+    {
+      "epoch": 6.4106592729513245,
+      "grad_norm": 0.043195899575948715,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 62427
+    },
+    {
+      "epoch": 6.410967344423907,
+      "grad_norm": 0.10521116852760315,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 62430
+    },
+    {
+      "epoch": 6.411275415896488,
+      "grad_norm": 0.061183419078588486,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 62433
+    },
+    {
+      "epoch": 6.41158348736907,
+      "grad_norm": 0.0777936577796936,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 62436
+    },
+    {
+      "epoch": 6.411891558841651,
+      "grad_norm": 0.07783685624599457,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 62439
+    },
+    {
+      "epoch": 6.412199630314233,
+      "grad_norm": 0.10137835890054703,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 62442
+    },
+    {
+      "epoch": 6.412507701786814,
+      "grad_norm": 0.07583697885274887,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 62445
+    },
+    {
+      "epoch": 6.412815773259396,
+      "grad_norm": 0.05925040692090988,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 62448
+    },
+    {
+      "epoch": 6.413123844731977,
+      "grad_norm": 0.07420961558818817,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 62451
+    },
+    {
+      "epoch": 6.413431916204559,
+      "grad_norm": 0.07555720955133438,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 62454
+    },
+    {
+      "epoch": 6.4137399876771415,
+      "grad_norm": 0.05991441011428833,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 62457
+    },
+    {
+      "epoch": 6.414048059149723,
+      "grad_norm": 0.04959188401699066,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 62460
+    },
+    {
+      "epoch": 6.414356130622305,
+      "grad_norm": 0.09466046094894409,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 62463
+    },
+    {
+      "epoch": 6.414664202094886,
+      "grad_norm": 0.08828117698431015,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 62466
+    },
+    {
+      "epoch": 6.414972273567468,
+      "grad_norm": 0.0796947106719017,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 62469
+    },
+    {
+      "epoch": 6.415280345040049,
+      "grad_norm": 0.09187500178813934,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 62472
+    },
+    {
+      "epoch": 6.415588416512631,
+      "grad_norm": 0.032242316752672195,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 62475
+    },
+    {
+      "epoch": 6.415896487985212,
+      "grad_norm": 0.08650948852300644,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 62478
+    },
+    {
+      "epoch": 6.416204559457794,
+      "grad_norm": 0.07760648429393768,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 62481
+    },
+    {
+      "epoch": 6.4165126309303755,
+      "grad_norm": 0.10039196908473969,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 62484
+    },
+    {
+      "epoch": 6.4168207024029575,
+      "grad_norm": 0.06620651483535767,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 62487
+    },
+    {
+      "epoch": 6.41712877387554,
+      "grad_norm": 0.07534004002809525,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 62490
+    },
+    {
+      "epoch": 6.417436845348121,
+      "grad_norm": 0.04290451481938362,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 62493
+    },
+    {
+      "epoch": 6.417744916820703,
+      "grad_norm": 0.040604498237371445,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 62496
+    },
+    {
+      "epoch": 6.418052988293284,
+      "grad_norm": 0.04257834702730179,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 62499
+    },
+    {
+      "epoch": 6.418361059765866,
+      "grad_norm": 0.06738926470279694,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 62502
+    },
+    {
+      "epoch": 6.418669131238447,
+      "grad_norm": 0.05635688826441765,
+      "learning_rate": 0.01,
+      "loss": 2.0168,
+      "step": 62505
+    },
+    {
+      "epoch": 6.418977202711029,
+      "grad_norm": 0.10578261315822601,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 62508
+    },
+    {
+      "epoch": 6.41928527418361,
+      "grad_norm": 0.08642464131116867,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 62511
+    },
+    {
+      "epoch": 6.419593345656192,
+      "grad_norm": 0.036675188690423965,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 62514
+    },
+    {
+      "epoch": 6.419901417128774,
+      "grad_norm": 0.14191675186157227,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 62517
+    },
+    {
+      "epoch": 6.420209488601356,
+      "grad_norm": 0.13353928923606873,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 62520
+    },
+    {
+      "epoch": 6.420517560073937,
+      "grad_norm": 0.06307082623243332,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 62523
+    },
+    {
+      "epoch": 6.420825631546519,
+      "grad_norm": 0.04877189174294472,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 62526
+    },
+    {
+      "epoch": 6.4211337030191,
+      "grad_norm": 0.07062195241451263,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 62529
+    },
+    {
+      "epoch": 6.421441774491682,
+      "grad_norm": 0.03799920529127121,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 62532
+    },
+    {
+      "epoch": 6.421749845964264,
+      "grad_norm": 0.0635172426700592,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 62535
+    },
+    {
+      "epoch": 6.422057917436845,
+      "grad_norm": 0.04074868932366371,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 62538
+    },
+    {
+      "epoch": 6.422365988909427,
+      "grad_norm": 0.1227545514702797,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 62541
+    },
+    {
+      "epoch": 6.4226740603820085,
+      "grad_norm": 0.10859929770231247,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 62544
+    },
+    {
+      "epoch": 6.4229821318545905,
+      "grad_norm": 0.10589033365249634,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 62547
+    },
+    {
+      "epoch": 6.423290203327172,
+      "grad_norm": 0.1324671059846878,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 62550
+    },
+    {
+      "epoch": 6.423598274799754,
+      "grad_norm": 0.0601799450814724,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 62553
+    },
+    {
+      "epoch": 6.423906346272335,
+      "grad_norm": 0.05349354445934296,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 62556
+    },
+    {
+      "epoch": 6.424214417744917,
+      "grad_norm": 0.03839118406176567,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 62559
+    },
+    {
+      "epoch": 6.424522489217498,
+      "grad_norm": 0.035764630883932114,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 62562
+    },
+    {
+      "epoch": 6.42483056069008,
+      "grad_norm": 0.05545911565423012,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 62565
+    },
+    {
+      "epoch": 6.425138632162661,
+      "grad_norm": 0.044913556426763535,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 62568
+    },
+    {
+      "epoch": 6.425446703635243,
+      "grad_norm": 0.040859974920749664,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 62571
+    },
+    {
+      "epoch": 6.425754775107825,
+      "grad_norm": 0.04757828265428543,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 62574
+    },
+    {
+      "epoch": 6.426062846580407,
+      "grad_norm": 0.0439058393239975,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 62577
+    },
+    {
+      "epoch": 6.426370918052989,
+      "grad_norm": 0.059357352554798126,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 62580
+    },
+    {
+      "epoch": 6.42667898952557,
+      "grad_norm": 0.1129942387342453,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 62583
+    },
+    {
+      "epoch": 6.426987060998152,
+      "grad_norm": 0.09056062996387482,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 62586
+    },
+    {
+      "epoch": 6.427295132470733,
+      "grad_norm": 0.05224711075425148,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 62589
+    },
+    {
+      "epoch": 6.427603203943315,
+      "grad_norm": 0.03806770220398903,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 62592
+    },
+    {
+      "epoch": 6.427911275415896,
+      "grad_norm": 0.045875824987888336,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 62595
+    },
+    {
+      "epoch": 6.428219346888478,
+      "grad_norm": 0.054107557982206345,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 62598
+    },
+    {
+      "epoch": 6.4285274183610595,
+      "grad_norm": 0.049330100417137146,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 62601
+    },
+    {
+      "epoch": 6.4288354898336415,
+      "grad_norm": 0.09636044502258301,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 62604
+    },
+    {
+      "epoch": 6.429143561306223,
+      "grad_norm": 0.06452757120132446,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 62607
+    },
+    {
+      "epoch": 6.429451632778805,
+      "grad_norm": 0.04598628729581833,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 62610
+    },
+    {
+      "epoch": 6.429759704251387,
+      "grad_norm": 0.06763578951358795,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 62613
+    },
+    {
+      "epoch": 6.430067775723968,
+      "grad_norm": 0.12232638895511627,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 62616
+    },
+    {
+      "epoch": 6.43037584719655,
+      "grad_norm": 0.09073688089847565,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 62619
+    },
+    {
+      "epoch": 6.430683918669131,
+      "grad_norm": 0.08724263310432434,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 62622
+    },
+    {
+      "epoch": 6.430991990141713,
+      "grad_norm": 0.11307045817375183,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 62625
+    },
+    {
+      "epoch": 6.431300061614294,
+      "grad_norm": 0.037230703979730606,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 62628
+    },
+    {
+      "epoch": 6.431608133086876,
+      "grad_norm": 0.06643498688936234,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 62631
+    },
+    {
+      "epoch": 6.431916204559458,
+      "grad_norm": 0.04486044496297836,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 62634
+    },
+    {
+      "epoch": 6.43222427603204,
+      "grad_norm": 0.03463894501328468,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 62637
+    },
+    {
+      "epoch": 6.432532347504621,
+      "grad_norm": 0.04060268774628639,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 62640
+    },
+    {
+      "epoch": 6.432840418977203,
+      "grad_norm": 0.11157824844121933,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 62643
+    },
+    {
+      "epoch": 6.433148490449784,
+      "grad_norm": 0.05599267780780792,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 62646
+    },
+    {
+      "epoch": 6.433456561922366,
+      "grad_norm": 0.05528303235769272,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 62649
+    },
+    {
+      "epoch": 6.433764633394947,
+      "grad_norm": 0.053959596902132034,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 62652
+    },
+    {
+      "epoch": 6.434072704867529,
+      "grad_norm": 0.06588397920131683,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 62655
+    },
+    {
+      "epoch": 6.434380776340111,
+      "grad_norm": 0.058823537081480026,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 62658
+    },
+    {
+      "epoch": 6.4346888478126925,
+      "grad_norm": 0.1689031720161438,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 62661
+    },
+    {
+      "epoch": 6.4349969192852745,
+      "grad_norm": 0.09005407243967056,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 62664
+    },
+    {
+      "epoch": 6.435304990757856,
+      "grad_norm": 0.04335148632526398,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 62667
+    },
+    {
+      "epoch": 6.435613062230438,
+      "grad_norm": 0.04634424299001694,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 62670
+    },
+    {
+      "epoch": 6.435921133703019,
+      "grad_norm": 0.03891894966363907,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 62673
+    },
+    {
+      "epoch": 6.436229205175601,
+      "grad_norm": 0.05635913461446762,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 62676
+    },
+    {
+      "epoch": 6.436537276648182,
+      "grad_norm": 0.05616259574890137,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 62679
+    },
+    {
+      "epoch": 6.436845348120764,
+      "grad_norm": 0.1745036244392395,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 62682
+    },
+    {
+      "epoch": 6.437153419593345,
+      "grad_norm": 0.09336113184690475,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 62685
+    },
+    {
+      "epoch": 6.437461491065927,
+      "grad_norm": 0.06508792191743851,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 62688
+    },
+    {
+      "epoch": 6.437769562538509,
+      "grad_norm": 0.08041112869977951,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 62691
+    },
+    {
+      "epoch": 6.438077634011091,
+      "grad_norm": 0.050097182393074036,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 62694
+    },
+    {
+      "epoch": 6.438385705483673,
+      "grad_norm": 0.09112852811813354,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 62697
+    },
+    {
+      "epoch": 6.438693776956254,
+      "grad_norm": 0.06812035292387009,
+      "learning_rate": 0.01,
+      "loss": 2.0099,
+      "step": 62700
+    },
+    {
+      "epoch": 6.439001848428836,
+      "grad_norm": 0.04868394508957863,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 62703
+    },
+    {
+      "epoch": 6.439309919901417,
+      "grad_norm": 0.03731069713830948,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 62706
+    },
+    {
+      "epoch": 6.439617991373999,
+      "grad_norm": 0.03705144673585892,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 62709
+    },
+    {
+      "epoch": 6.43992606284658,
+      "grad_norm": 0.06617160141468048,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 62712
+    },
+    {
+      "epoch": 6.440234134319162,
+      "grad_norm": 0.09520690143108368,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 62715
+    },
+    {
+      "epoch": 6.440542205791743,
+      "grad_norm": 0.10066824406385422,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 62718
+    },
+    {
+      "epoch": 6.4408502772643255,
+      "grad_norm": 0.04232907295227051,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 62721
+    },
+    {
+      "epoch": 6.441158348736907,
+      "grad_norm": 0.08188414573669434,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 62724
+    },
+    {
+      "epoch": 6.441466420209489,
+      "grad_norm": 0.11537371575832367,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 62727
+    },
+    {
+      "epoch": 6.44177449168207,
+      "grad_norm": 0.03267863765358925,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 62730
+    },
+    {
+      "epoch": 6.442082563154652,
+      "grad_norm": 0.0837605893611908,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 62733
+    },
+    {
+      "epoch": 6.442390634627234,
+      "grad_norm": 0.05545349791646004,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 62736
+    },
+    {
+      "epoch": 6.442698706099815,
+      "grad_norm": 0.06512763351202011,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 62739
+    },
+    {
+      "epoch": 6.443006777572397,
+      "grad_norm": 0.051384586840867996,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 62742
+    },
+    {
+      "epoch": 6.443314849044978,
+      "grad_norm": 0.035269416868686676,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 62745
+    },
+    {
+      "epoch": 6.44362292051756,
+      "grad_norm": 0.11074446141719818,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 62748
+    },
+    {
+      "epoch": 6.4439309919901415,
+      "grad_norm": 0.05051399767398834,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 62751
+    },
+    {
+      "epoch": 6.444239063462724,
+      "grad_norm": 0.07275965064764023,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 62754
+    },
+    {
+      "epoch": 6.444547134935305,
+      "grad_norm": 0.07027794420719147,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 62757
+    },
+    {
+      "epoch": 6.444855206407887,
+      "grad_norm": 0.11397121101617813,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 62760
+    },
+    {
+      "epoch": 6.445163277880468,
+      "grad_norm": 0.08914832770824432,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 62763
+    },
+    {
+      "epoch": 6.44547134935305,
+      "grad_norm": 0.1111316904425621,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 62766
+    },
+    {
+      "epoch": 6.445779420825631,
+      "grad_norm": 0.11033376306295395,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 62769
+    },
+    {
+      "epoch": 6.446087492298213,
+      "grad_norm": 0.07109701633453369,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 62772
+    },
+    {
+      "epoch": 6.446395563770794,
+      "grad_norm": 0.0448211170732975,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 62775
+    },
+    {
+      "epoch": 6.446703635243376,
+      "grad_norm": 0.04059508815407753,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 62778
+    },
+    {
+      "epoch": 6.4470117067159585,
+      "grad_norm": 0.04148499295115471,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 62781
+    },
+    {
+      "epoch": 6.44731977818854,
+      "grad_norm": 0.04664915055036545,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 62784
+    },
+    {
+      "epoch": 6.447627849661122,
+      "grad_norm": 0.03414386510848999,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 62787
+    },
+    {
+      "epoch": 6.447935921133703,
+      "grad_norm": 0.047360535711050034,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 62790
+    },
+    {
+      "epoch": 6.448243992606285,
+      "grad_norm": 0.11773015558719635,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 62793
+    },
+    {
+      "epoch": 6.448552064078866,
+      "grad_norm": 0.08103271573781967,
+      "learning_rate": 0.01,
+      "loss": 2.0067,
+      "step": 62796
+    },
+    {
+      "epoch": 6.448860135551448,
+      "grad_norm": 0.07534775882959366,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 62799
+    },
+    {
+      "epoch": 6.449168207024029,
+      "grad_norm": 0.0421251580119133,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 62802
+    },
+    {
+      "epoch": 6.449476278496611,
+      "grad_norm": 0.06340447813272476,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 62805
+    },
+    {
+      "epoch": 6.4497843499691925,
+      "grad_norm": 0.06287892907857895,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 62808
+    },
+    {
+      "epoch": 6.4500924214417745,
+      "grad_norm": 0.11696802824735641,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 62811
+    },
+    {
+      "epoch": 6.450400492914357,
+      "grad_norm": 0.061545319855213165,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 62814
+    },
+    {
+      "epoch": 6.450708564386938,
+      "grad_norm": 0.04134809225797653,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 62817
+    },
+    {
+      "epoch": 6.45101663585952,
+      "grad_norm": 0.11149853467941284,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 62820
+    },
+    {
+      "epoch": 6.451324707332101,
+      "grad_norm": 0.06323488801717758,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 62823
+    },
+    {
+      "epoch": 6.451632778804683,
+      "grad_norm": 0.0836939662694931,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 62826
+    },
+    {
+      "epoch": 6.451940850277264,
+      "grad_norm": 0.04251860827207565,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 62829
+    },
+    {
+      "epoch": 6.452248921749846,
+      "grad_norm": 0.08365978300571442,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 62832
+    },
+    {
+      "epoch": 6.452556993222427,
+      "grad_norm": 0.07985960692167282,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 62835
+    },
+    {
+      "epoch": 6.452865064695009,
+      "grad_norm": 0.0690624862909317,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 62838
+    },
+    {
+      "epoch": 6.453173136167591,
+      "grad_norm": 0.07743382453918457,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 62841
+    },
+    {
+      "epoch": 6.453481207640173,
+      "grad_norm": 0.06049206852912903,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 62844
+    },
+    {
+      "epoch": 6.453789279112754,
+      "grad_norm": 0.049416352063417435,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 62847
+    },
+    {
+      "epoch": 6.454097350585336,
+      "grad_norm": 0.03889904171228409,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 62850
+    },
+    {
+      "epoch": 6.454405422057917,
+      "grad_norm": 0.03183984011411667,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 62853
+    },
+    {
+      "epoch": 6.454713493530499,
+      "grad_norm": 0.03685016930103302,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 62856
+    },
+    {
+      "epoch": 6.455021565003081,
+      "grad_norm": 0.05258958786725998,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 62859
+    },
+    {
+      "epoch": 6.455329636475662,
+      "grad_norm": 0.12214828282594681,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 62862
+    },
+    {
+      "epoch": 6.455637707948244,
+      "grad_norm": 0.04400373622775078,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 62865
+    },
+    {
+      "epoch": 6.4559457794208255,
+      "grad_norm": 0.09445548057556152,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 62868
+    },
+    {
+      "epoch": 6.4562538508934075,
+      "grad_norm": 0.05584699660539627,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 62871
+    },
+    {
+      "epoch": 6.456561922365989,
+      "grad_norm": 0.041519295424222946,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 62874
+    },
+    {
+      "epoch": 6.456869993838571,
+      "grad_norm": 0.08750978112220764,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 62877
+    },
+    {
+      "epoch": 6.457178065311152,
+      "grad_norm": 0.0746956318616867,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 62880
+    },
+    {
+      "epoch": 6.457486136783734,
+      "grad_norm": 0.05346643924713135,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 62883
+    },
+    {
+      "epoch": 6.457794208256315,
+      "grad_norm": 0.033672984689474106,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 62886
+    },
+    {
+      "epoch": 6.458102279728897,
+      "grad_norm": 0.07131464779376984,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 62889
+    },
+    {
+      "epoch": 6.458410351201478,
+      "grad_norm": 0.08488958328962326,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 62892
+    },
+    {
+      "epoch": 6.45871842267406,
+      "grad_norm": 0.041019588708877563,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 62895
+    },
+    {
+      "epoch": 6.459026494146642,
+      "grad_norm": 0.07616984844207764,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 62898
+    },
+    {
+      "epoch": 6.459334565619224,
+      "grad_norm": 0.071281798183918,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 62901
+    },
+    {
+      "epoch": 6.459642637091806,
+      "grad_norm": 0.04009273648262024,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 62904
+    },
+    {
+      "epoch": 6.459950708564387,
+      "grad_norm": 0.059646125882864,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 62907
+    },
+    {
+      "epoch": 6.460258780036969,
+      "grad_norm": 0.08733568340539932,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 62910
+    },
+    {
+      "epoch": 6.46056685150955,
+      "grad_norm": 0.06603758037090302,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 62913
+    },
+    {
+      "epoch": 6.460874922982132,
+      "grad_norm": 0.036482322961091995,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 62916
+    },
+    {
+      "epoch": 6.461182994454713,
+      "grad_norm": 0.08736536651849747,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 62919
+    },
+    {
+      "epoch": 6.461491065927295,
+      "grad_norm": 0.10915714502334595,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 62922
+    },
+    {
+      "epoch": 6.461799137399876,
+      "grad_norm": 0.07954758405685425,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 62925
+    },
+    {
+      "epoch": 6.4621072088724585,
+      "grad_norm": 0.05602665990591049,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 62928
+    },
+    {
+      "epoch": 6.46241528034504,
+      "grad_norm": 0.09531257301568985,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 62931
+    },
+    {
+      "epoch": 6.462723351817622,
+      "grad_norm": 0.044863101094961166,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 62934
+    },
+    {
+      "epoch": 6.463031423290204,
+      "grad_norm": 0.0783008560538292,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 62937
+    },
+    {
+      "epoch": 6.463339494762785,
+      "grad_norm": 0.05667195841670036,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 62940
+    },
+    {
+      "epoch": 6.463647566235367,
+      "grad_norm": 0.13006943464279175,
+      "learning_rate": 0.01,
+      "loss": 2.0073,
+      "step": 62943
+    },
+    {
+      "epoch": 6.463955637707948,
+      "grad_norm": 0.08932863920927048,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 62946
+    },
+    {
+      "epoch": 6.46426370918053,
+      "grad_norm": 0.04788200557231903,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 62949
+    },
+    {
+      "epoch": 6.464571780653111,
+      "grad_norm": 0.03904096782207489,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 62952
+    },
+    {
+      "epoch": 6.464879852125693,
+      "grad_norm": 0.06443644315004349,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 62955
+    },
+    {
+      "epoch": 6.4651879235982745,
+      "grad_norm": 0.08434759825468063,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 62958
+    },
+    {
+      "epoch": 6.465495995070857,
+      "grad_norm": 0.03871188312768936,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 62961
+    },
+    {
+      "epoch": 6.465804066543438,
+      "grad_norm": 0.11748526990413666,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 62964
+    },
+    {
+      "epoch": 6.46611213801602,
+      "grad_norm": 0.05763693153858185,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 62967
+    },
+    {
+      "epoch": 6.466420209488601,
+      "grad_norm": 0.0419640988111496,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 62970
+    },
+    {
+      "epoch": 6.466728280961183,
+      "grad_norm": 0.061243150383234024,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 62973
+    },
+    {
+      "epoch": 6.467036352433764,
+      "grad_norm": 0.10403522849082947,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 62976
+    },
+    {
+      "epoch": 6.467344423906346,
+      "grad_norm": 0.14000679552555084,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 62979
+    },
+    {
+      "epoch": 6.467652495378928,
+      "grad_norm": 0.08773007243871689,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 62982
+    },
+    {
+      "epoch": 6.467960566851509,
+      "grad_norm": 0.04810422658920288,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 62985
+    },
+    {
+      "epoch": 6.4682686383240915,
+      "grad_norm": 0.05809152126312256,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 62988
+    },
+    {
+      "epoch": 6.468576709796673,
+      "grad_norm": 0.05598244071006775,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 62991
+    },
+    {
+      "epoch": 6.468884781269255,
+      "grad_norm": 0.07979332655668259,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 62994
+    },
+    {
+      "epoch": 6.469192852741836,
+      "grad_norm": 0.044398579746484756,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 62997
+    },
+    {
+      "epoch": 6.469500924214418,
+      "grad_norm": 0.04663238674402237,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 63000
+    },
+    {
+      "epoch": 6.469808995686999,
+      "grad_norm": 0.04528966173529625,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 63003
+    },
+    {
+      "epoch": 6.470117067159581,
+      "grad_norm": 0.06028036028146744,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 63006
+    },
+    {
+      "epoch": 6.470425138632162,
+      "grad_norm": 0.094419464468956,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 63009
+    },
+    {
+      "epoch": 6.470733210104744,
+      "grad_norm": 0.18547490239143372,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 63012
+    },
+    {
+      "epoch": 6.471041281577326,
+      "grad_norm": 0.09084778279066086,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 63015
+    },
+    {
+      "epoch": 6.4713493530499075,
+      "grad_norm": 0.07517223805189133,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 63018
+    },
+    {
+      "epoch": 6.47165742452249,
+      "grad_norm": 0.041310783475637436,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 63021
+    },
+    {
+      "epoch": 6.471965495995071,
+      "grad_norm": 0.04006296768784523,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 63024
+    },
+    {
+      "epoch": 6.472273567467653,
+      "grad_norm": 0.03506629914045334,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 63027
+    },
+    {
+      "epoch": 6.472581638940234,
+      "grad_norm": 0.04409859701991081,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 63030
+    },
+    {
+      "epoch": 6.472889710412816,
+      "grad_norm": 0.054022327065467834,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 63033
+    },
+    {
+      "epoch": 6.473197781885397,
+      "grad_norm": 0.06177505478262901,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 63036
+    },
+    {
+      "epoch": 6.473505853357979,
+      "grad_norm": 0.10999149829149246,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 63039
+    },
+    {
+      "epoch": 6.47381392483056,
+      "grad_norm": 0.130665123462677,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 63042
+    },
+    {
+      "epoch": 6.4741219963031424,
+      "grad_norm": 0.15091292560100555,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 63045
+    },
+    {
+      "epoch": 6.474430067775724,
+      "grad_norm": 0.1041717529296875,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 63048
+    },
+    {
+      "epoch": 6.474738139248306,
+      "grad_norm": 0.05246599763631821,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 63051
+    },
+    {
+      "epoch": 6.475046210720887,
+      "grad_norm": 0.04646655172109604,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 63054
+    },
+    {
+      "epoch": 6.475354282193469,
+      "grad_norm": 0.047972384840250015,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 63057
+    },
+    {
+      "epoch": 6.475662353666051,
+      "grad_norm": 0.057053469121456146,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 63060
+    },
+    {
+      "epoch": 6.475970425138632,
+      "grad_norm": 0.044478513300418854,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 63063
+    },
+    {
+      "epoch": 6.476278496611214,
+      "grad_norm": 0.13227659463882446,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 63066
+    },
+    {
+      "epoch": 6.476586568083795,
+      "grad_norm": 0.09201903641223907,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 63069
+    },
+    {
+      "epoch": 6.476894639556377,
+      "grad_norm": 0.07452027499675751,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 63072
+    },
+    {
+      "epoch": 6.4772027110289585,
+      "grad_norm": 0.04616081714630127,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 63075
+    },
+    {
+      "epoch": 6.4775107825015406,
+      "grad_norm": 0.047700148075819016,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 63078
+    },
+    {
+      "epoch": 6.477818853974122,
+      "grad_norm": 0.0366378016769886,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 63081
+    },
+    {
+      "epoch": 6.478126925446704,
+      "grad_norm": 0.037750110030174255,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 63084
+    },
+    {
+      "epoch": 6.478434996919285,
+      "grad_norm": 0.05896897614002228,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 63087
+    },
+    {
+      "epoch": 6.478743068391867,
+      "grad_norm": 0.13248635828495026,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 63090
+    },
+    {
+      "epoch": 6.479051139864448,
+      "grad_norm": 0.14373722672462463,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 63093
+    },
+    {
+      "epoch": 6.47935921133703,
+      "grad_norm": 0.07670046389102936,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 63096
+    },
+    {
+      "epoch": 6.479667282809612,
+      "grad_norm": 0.05786823108792305,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 63099
+    },
+    {
+      "epoch": 6.479975354282193,
+      "grad_norm": 0.10580223053693771,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 63102
+    },
+    {
+      "epoch": 6.4802834257547755,
+      "grad_norm": 0.09750565141439438,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 63105
+    },
+    {
+      "epoch": 6.480591497227357,
+      "grad_norm": 0.07120301574468613,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 63108
+    },
+    {
+      "epoch": 6.480899568699939,
+      "grad_norm": 0.046966131776571274,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 63111
+    },
+    {
+      "epoch": 6.48120764017252,
+      "grad_norm": 0.04642505198717117,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 63114
+    },
+    {
+      "epoch": 6.481515711645102,
+      "grad_norm": 0.0937868282198906,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 63117
+    },
+    {
+      "epoch": 6.481823783117683,
+      "grad_norm": 0.11612875759601593,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 63120
+    },
+    {
+      "epoch": 6.482131854590265,
+      "grad_norm": 0.042997948825359344,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 63123
+    },
+    {
+      "epoch": 6.482439926062846,
+      "grad_norm": 0.10609213262796402,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 63126
+    },
+    {
+      "epoch": 6.482747997535428,
+      "grad_norm": 0.09225320816040039,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 63129
+    },
+    {
+      "epoch": 6.4830560690080095,
+      "grad_norm": 0.053099583834409714,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 63132
+    },
+    {
+      "epoch": 6.4833641404805915,
+      "grad_norm": 0.04756970703601837,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 63135
+    },
+    {
+      "epoch": 6.483672211953174,
+      "grad_norm": 0.04079375043511391,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 63138
+    },
+    {
+      "epoch": 6.483980283425755,
+      "grad_norm": 0.08000772446393967,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 63141
+    },
+    {
+      "epoch": 6.484288354898337,
+      "grad_norm": 0.05502250790596008,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 63144
+    },
+    {
+      "epoch": 6.484596426370918,
+      "grad_norm": 0.07607629895210266,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 63147
+    },
+    {
+      "epoch": 6.4849044978435,
+      "grad_norm": 0.0457836389541626,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 63150
+    },
+    {
+      "epoch": 6.485212569316081,
+      "grad_norm": 0.04571022093296051,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 63153
+    },
+    {
+      "epoch": 6.485520640788663,
+      "grad_norm": 0.054583076387643814,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 63156
+    },
+    {
+      "epoch": 6.485828712261244,
+      "grad_norm": 0.08559385687112808,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 63159
+    },
+    {
+      "epoch": 6.486136783733826,
+      "grad_norm": 0.06536659598350525,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 63162
+    },
+    {
+      "epoch": 6.486444855206408,
+      "grad_norm": 0.031179916113615036,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 63165
+    },
+    {
+      "epoch": 6.48675292667899,
+      "grad_norm": 0.038442451506853104,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 63168
+    },
+    {
+      "epoch": 6.487060998151571,
+      "grad_norm": 0.05742299556732178,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 63171
+    },
+    {
+      "epoch": 6.487369069624153,
+      "grad_norm": 0.07068540900945663,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 63174
+    },
+    {
+      "epoch": 6.487677141096734,
+      "grad_norm": 0.041148263961076736,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 63177
+    },
+    {
+      "epoch": 6.487985212569316,
+      "grad_norm": 0.08058618754148483,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 63180
+    },
+    {
+      "epoch": 6.488293284041898,
+      "grad_norm": 0.10617338865995407,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 63183
+    },
+    {
+      "epoch": 6.488601355514479,
+      "grad_norm": 0.13301149010658264,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 63186
+    },
+    {
+      "epoch": 6.488909426987061,
+      "grad_norm": 0.12188655138015747,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 63189
+    },
+    {
+      "epoch": 6.4892174984596425,
+      "grad_norm": 0.07612743228673935,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 63192
+    },
+    {
+      "epoch": 6.4895255699322245,
+      "grad_norm": 0.08802825957536697,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 63195
+    },
+    {
+      "epoch": 6.489833641404806,
+      "grad_norm": 0.05692920833826065,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 63198
+    },
+    {
+      "epoch": 6.490141712877388,
+      "grad_norm": 0.0412018783390522,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 63201
+    },
+    {
+      "epoch": 6.490449784349969,
+      "grad_norm": 0.0954689234495163,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 63204
+    },
+    {
+      "epoch": 6.490757855822551,
+      "grad_norm": 0.11650601774454117,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 63207
+    },
+    {
+      "epoch": 6.491065927295132,
+      "grad_norm": 0.06705356389284134,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 63210
+    },
+    {
+      "epoch": 6.491373998767714,
+      "grad_norm": 0.04871569201350212,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 63213
+    },
+    {
+      "epoch": 6.491682070240296,
+      "grad_norm": 0.041996728628873825,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 63216
+    },
+    {
+      "epoch": 6.491990141712877,
+      "grad_norm": 0.03821207210421562,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 63219
+    },
+    {
+      "epoch": 6.492298213185459,
+      "grad_norm": 0.03589041158556938,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 63222
+    },
+    {
+      "epoch": 6.492606284658041,
+      "grad_norm": 0.05628860369324684,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 63225
+    },
+    {
+      "epoch": 6.492914356130623,
+      "grad_norm": 0.10156149417161942,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 63228
+    },
+    {
+      "epoch": 6.493222427603204,
+      "grad_norm": 0.1743369847536087,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 63231
+    },
+    {
+      "epoch": 6.493530499075786,
+      "grad_norm": 0.13694548606872559,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 63234
+    },
+    {
+      "epoch": 6.493838570548367,
+      "grad_norm": 0.12643152475357056,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 63237
+    },
+    {
+      "epoch": 6.494146642020949,
+      "grad_norm": 0.14127200841903687,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 63240
+    },
+    {
+      "epoch": 6.49445471349353,
+      "grad_norm": 0.05180880054831505,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 63243
+    },
+    {
+      "epoch": 6.494762784966112,
+      "grad_norm": 0.050564274191856384,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 63246
+    },
+    {
+      "epoch": 6.495070856438693,
+      "grad_norm": 0.09520803391933441,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 63249
+    },
+    {
+      "epoch": 6.4953789279112755,
+      "grad_norm": 0.09522721916437149,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 63252
+    },
+    {
+      "epoch": 6.495686999383857,
+      "grad_norm": 0.1086997389793396,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 63255
+    },
+    {
+      "epoch": 6.495995070856439,
+      "grad_norm": 0.08121590316295624,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 63258
+    },
+    {
+      "epoch": 6.496303142329021,
+      "grad_norm": 0.06703486293554306,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 63261
+    },
+    {
+      "epoch": 6.496611213801602,
+      "grad_norm": 0.04283798858523369,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 63264
+    },
+    {
+      "epoch": 6.496919285274184,
+      "grad_norm": 0.04943424090743065,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 63267
+    },
+    {
+      "epoch": 6.497227356746765,
+      "grad_norm": 0.0578305721282959,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 63270
+    },
+    {
+      "epoch": 6.497535428219347,
+      "grad_norm": 0.0359860323369503,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 63273
+    },
+    {
+      "epoch": 6.497843499691928,
+      "grad_norm": 0.04893532767891884,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 63276
+    },
+    {
+      "epoch": 6.49815157116451,
+      "grad_norm": 0.04525068402290344,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 63279
+    },
+    {
+      "epoch": 6.4984596426370915,
+      "grad_norm": 0.04810551553964615,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 63282
+    },
+    {
+      "epoch": 6.498767714109674,
+      "grad_norm": 0.05285782366991043,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 63285
+    },
+    {
+      "epoch": 6.499075785582255,
+      "grad_norm": 0.032118529081344604,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 63288
+    },
+    {
+      "epoch": 6.499383857054837,
+      "grad_norm": 0.049875080585479736,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 63291
+    },
+    {
+      "epoch": 6.499691928527418,
+      "grad_norm": 0.04658648371696472,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 63294
+    },
+    {
+      "epoch": 6.5,
+      "grad_norm": 0.05129887908697128,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 63297
+    },
+    {
+      "epoch": 6.500308071472581,
+      "grad_norm": 0.07968533784151077,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 63300
+    },
+    {
+      "epoch": 6.500616142945163,
+      "grad_norm": 0.08152198046445847,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 63303
+    },
+    {
+      "epoch": 6.500924214417745,
+      "grad_norm": 0.07405789196491241,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 63306
+    },
+    {
+      "epoch": 6.501232285890326,
+      "grad_norm": 0.07860677689313889,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 63309
+    },
+    {
+      "epoch": 6.5015403573629085,
+      "grad_norm": 0.051370490342378616,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 63312
+    },
+    {
+      "epoch": 6.50184842883549,
+      "grad_norm": 0.04912220686674118,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 63315
+    },
+    {
+      "epoch": 6.502156500308072,
+      "grad_norm": 0.042900361120700836,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 63318
+    },
+    {
+      "epoch": 6.502464571780653,
+      "grad_norm": 0.16261926293373108,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 63321
+    },
+    {
+      "epoch": 6.502772643253235,
+      "grad_norm": 0.11455561220645905,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 63324
+    },
+    {
+      "epoch": 6.503080714725816,
+      "grad_norm": 0.04585309699177742,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 63327
+    },
+    {
+      "epoch": 6.503388786198398,
+      "grad_norm": 0.03483370691537857,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 63330
+    },
+    {
+      "epoch": 6.503696857670979,
+      "grad_norm": 0.08230151981115341,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 63333
+    },
+    {
+      "epoch": 6.504004929143561,
+      "grad_norm": 0.06978236138820648,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 63336
+    },
+    {
+      "epoch": 6.504313000616143,
+      "grad_norm": 0.03723878785967827,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 63339
+    },
+    {
+      "epoch": 6.5046210720887245,
+      "grad_norm": 0.0846422016620636,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 63342
+    },
+    {
+      "epoch": 6.504929143561307,
+      "grad_norm": 0.06443771719932556,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 63345
+    },
+    {
+      "epoch": 6.505237215033888,
+      "grad_norm": 0.07399289309978485,
+      "learning_rate": 0.01,
+      "loss": 2.0061,
+      "step": 63348
+    },
+    {
+      "epoch": 6.50554528650647,
+      "grad_norm": 0.08382997661828995,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 63351
+    },
+    {
+      "epoch": 6.505853357979051,
+      "grad_norm": 0.0767715722322464,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 63354
+    },
+    {
+      "epoch": 6.506161429451633,
+      "grad_norm": 0.07842829078435898,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 63357
+    },
+    {
+      "epoch": 6.506469500924214,
+      "grad_norm": 0.04764385148882866,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 63360
+    },
+    {
+      "epoch": 6.506777572396796,
+      "grad_norm": 0.08968043327331543,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 63363
+    },
+    {
+      "epoch": 6.507085643869377,
+      "grad_norm": 0.07326337695121765,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 63366
+    },
+    {
+      "epoch": 6.507393715341959,
+      "grad_norm": 0.04753720387816429,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 63369
+    },
+    {
+      "epoch": 6.507701786814541,
+      "grad_norm": 0.05614113807678223,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 63372
+    },
+    {
+      "epoch": 6.508009858287123,
+      "grad_norm": 0.04377448558807373,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 63375
+    },
+    {
+      "epoch": 6.508317929759704,
+      "grad_norm": 0.03820153325796127,
+      "learning_rate": 0.01,
+      "loss": 2.0105,
+      "step": 63378
+    },
+    {
+      "epoch": 6.508626001232286,
+      "grad_norm": 0.03955162316560745,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 63381
+    },
+    {
+      "epoch": 6.508934072704868,
+      "grad_norm": 0.03919532150030136,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 63384
+    },
+    {
+      "epoch": 6.509242144177449,
+      "grad_norm": 0.13084079325199127,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 63387
+    },
+    {
+      "epoch": 6.509550215650031,
+      "grad_norm": 0.05490676313638687,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 63390
+    },
+    {
+      "epoch": 6.509858287122612,
+      "grad_norm": 0.048263441771268845,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 63393
+    },
+    {
+      "epoch": 6.510166358595194,
+      "grad_norm": 0.07833252102136612,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 63396
+    },
+    {
+      "epoch": 6.5104744300677755,
+      "grad_norm": 0.06575249880552292,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 63399
+    },
+    {
+      "epoch": 6.5107825015403575,
+      "grad_norm": 0.1300118863582611,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 63402
+    },
+    {
+      "epoch": 6.511090573012939,
+      "grad_norm": 0.05606196075677872,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 63405
+    },
+    {
+      "epoch": 6.511398644485521,
+      "grad_norm": 0.04275618493556976,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 63408
+    },
+    {
+      "epoch": 6.511706715958102,
+      "grad_norm": 0.11378994584083557,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 63411
+    },
+    {
+      "epoch": 6.512014787430684,
+      "grad_norm": 0.04268041253089905,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 63414
+    },
+    {
+      "epoch": 6.512322858903266,
+      "grad_norm": 0.18747419118881226,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 63417
+    },
+    {
+      "epoch": 6.512630930375847,
+      "grad_norm": 0.10926712304353714,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 63420
+    },
+    {
+      "epoch": 6.512939001848429,
+      "grad_norm": 0.12267261743545532,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 63423
+    },
+    {
+      "epoch": 6.51324707332101,
+      "grad_norm": 0.06755910813808441,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 63426
+    },
+    {
+      "epoch": 6.513555144793592,
+      "grad_norm": 0.06156865134835243,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 63429
+    },
+    {
+      "epoch": 6.513863216266174,
+      "grad_norm": 0.0420864075422287,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 63432
+    },
+    {
+      "epoch": 6.514171287738756,
+      "grad_norm": 0.037585970014333725,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 63435
+    },
+    {
+      "epoch": 6.514479359211337,
+      "grad_norm": 0.042937878519296646,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 63438
+    },
+    {
+      "epoch": 6.514787430683919,
+      "grad_norm": 0.04278058186173439,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 63441
+    },
+    {
+      "epoch": 6.5150955021565,
+      "grad_norm": 0.053777556866407394,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 63444
+    },
+    {
+      "epoch": 6.515403573629082,
+      "grad_norm": 0.11828344315290451,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 63447
+    },
+    {
+      "epoch": 6.515711645101663,
+      "grad_norm": 0.07692524790763855,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 63450
+    },
+    {
+      "epoch": 6.516019716574245,
+      "grad_norm": 0.036424048244953156,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 63453
+    },
+    {
+      "epoch": 6.516327788046826,
+      "grad_norm": 0.051881253719329834,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 63456
+    },
+    {
+      "epoch": 6.5166358595194085,
+      "grad_norm": 0.09392795711755753,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 63459
+    },
+    {
+      "epoch": 6.5169439309919905,
+      "grad_norm": 0.05411149188876152,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 63462
+    },
+    {
+      "epoch": 6.517252002464572,
+      "grad_norm": 0.07104165852069855,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 63465
+    },
+    {
+      "epoch": 6.517560073937154,
+      "grad_norm": 0.0653427243232727,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 63468
+    },
+    {
+      "epoch": 6.517868145409735,
+      "grad_norm": 0.07316904515028,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 63471
+    },
+    {
+      "epoch": 6.518176216882317,
+      "grad_norm": 0.08225788921117783,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 63474
+    },
+    {
+      "epoch": 6.518484288354898,
+      "grad_norm": 0.07449998706579208,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 63477
+    },
+    {
+      "epoch": 6.51879235982748,
+      "grad_norm": 0.07791251689195633,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 63480
+    },
+    {
+      "epoch": 6.519100431300061,
+      "grad_norm": 0.1233309879899025,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 63483
+    },
+    {
+      "epoch": 6.519408502772643,
+      "grad_norm": 0.041390251368284225,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 63486
+    },
+    {
+      "epoch": 6.5197165742452245,
+      "grad_norm": 0.039635319262742996,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 63489
+    },
+    {
+      "epoch": 6.520024645717807,
+      "grad_norm": 0.03680063784122467,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 63492
+    },
+    {
+      "epoch": 6.520332717190389,
+      "grad_norm": 0.08327323198318481,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 63495
+    },
+    {
+      "epoch": 6.52064078866297,
+      "grad_norm": 0.10368441045284271,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 63498
+    },
+    {
+      "epoch": 6.520948860135551,
+      "grad_norm": 0.05991170555353165,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 63501
+    },
+    {
+      "epoch": 6.521256931608133,
+      "grad_norm": 0.05868203192949295,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 63504
+    },
+    {
+      "epoch": 6.521565003080715,
+      "grad_norm": 0.04409966990351677,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 63507
+    },
+    {
+      "epoch": 6.521873074553296,
+      "grad_norm": 0.04939546808600426,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 63510
+    },
+    {
+      "epoch": 6.522181146025878,
+      "grad_norm": 0.041482388973236084,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 63513
+    },
+    {
+      "epoch": 6.5224892174984594,
+      "grad_norm": 0.03503831848502159,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 63516
+    },
+    {
+      "epoch": 6.5227972889710415,
+      "grad_norm": 0.07293230295181274,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 63519
+    },
+    {
+      "epoch": 6.523105360443623,
+      "grad_norm": 0.13121885061264038,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 63522
+    },
+    {
+      "epoch": 6.523413431916205,
+      "grad_norm": 0.05674733221530914,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 63525
+    },
+    {
+      "epoch": 6.523721503388786,
+      "grad_norm": 0.13403059542179108,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 63528
+    },
+    {
+      "epoch": 6.524029574861368,
+      "grad_norm": 0.05892037972807884,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 63531
+    },
+    {
+      "epoch": 6.524337646333949,
+      "grad_norm": 0.05163764953613281,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 63534
+    },
+    {
+      "epoch": 6.524645717806531,
+      "grad_norm": 0.10264305770397186,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 63537
+    },
+    {
+      "epoch": 6.524953789279113,
+      "grad_norm": 0.060135189443826675,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 63540
+    },
+    {
+      "epoch": 6.525261860751694,
+      "grad_norm": 0.033466801047325134,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 63543
+    },
+    {
+      "epoch": 6.525569932224276,
+      "grad_norm": 0.05220318213105202,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 63546
+    },
+    {
+      "epoch": 6.5258780036968576,
+      "grad_norm": 0.04468940570950508,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 63549
+    },
+    {
+      "epoch": 6.52618607516944,
+      "grad_norm": 0.07253078371286392,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 63552
+    },
+    {
+      "epoch": 6.526494146642021,
+      "grad_norm": 0.039232298731803894,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 63555
+    },
+    {
+      "epoch": 6.526802218114603,
+      "grad_norm": 0.08601675182580948,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 63558
+    },
+    {
+      "epoch": 6.527110289587184,
+      "grad_norm": 0.11496642231941223,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 63561
+    },
+    {
+      "epoch": 6.527418361059766,
+      "grad_norm": 0.21819603443145752,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 63564
+    },
+    {
+      "epoch": 6.527726432532347,
+      "grad_norm": 0.15641893446445465,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 63567
+    },
+    {
+      "epoch": 6.528034504004929,
+      "grad_norm": 0.07390442490577698,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 63570
+    },
+    {
+      "epoch": 6.52834257547751,
+      "grad_norm": 0.03637871518731117,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 63573
+    },
+    {
+      "epoch": 6.5286506469500925,
+      "grad_norm": 0.05536114424467087,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 63576
+    },
+    {
+      "epoch": 6.528958718422674,
+      "grad_norm": 0.08773189783096313,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 63579
+    },
+    {
+      "epoch": 6.529266789895256,
+      "grad_norm": 0.05206519737839699,
+      "learning_rate": 0.01,
+      "loss": 2.0209,
+      "step": 63582
+    },
+    {
+      "epoch": 6.529574861367838,
+      "grad_norm": 0.12326787412166595,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 63585
+    },
+    {
+      "epoch": 6.529882932840419,
+      "grad_norm": 0.10255524516105652,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 63588
+    },
+    {
+      "epoch": 6.530191004313001,
+      "grad_norm": 0.08496958762407303,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 63591
+    },
+    {
+      "epoch": 6.530499075785582,
+      "grad_norm": 0.040596868842840195,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 63594
+    },
+    {
+      "epoch": 6.530807147258164,
+      "grad_norm": 0.033690448850393295,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 63597
+    },
+    {
+      "epoch": 6.531115218730745,
+      "grad_norm": 0.0766686350107193,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 63600
+    },
+    {
+      "epoch": 6.531423290203327,
+      "grad_norm": 0.12199605256319046,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 63603
+    },
+    {
+      "epoch": 6.5317313616759085,
+      "grad_norm": 0.03897617757320404,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 63606
+    },
+    {
+      "epoch": 6.532039433148491,
+      "grad_norm": 0.05026644468307495,
+      "learning_rate": 0.01,
+      "loss": 2.0126,
+      "step": 63609
+    },
+    {
+      "epoch": 6.532347504621072,
+      "grad_norm": 0.03886674344539642,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 63612
+    },
+    {
+      "epoch": 6.532655576093654,
+      "grad_norm": 0.09314893931150436,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 63615
+    },
+    {
+      "epoch": 6.532963647566236,
+      "grad_norm": 0.048525359481573105,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 63618
+    },
+    {
+      "epoch": 6.533271719038817,
+      "grad_norm": 0.027339305728673935,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 63621
+    },
+    {
+      "epoch": 6.533579790511399,
+      "grad_norm": 0.0443040207028389,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 63624
+    },
+    {
+      "epoch": 6.53388786198398,
+      "grad_norm": 0.06850685924291611,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 63627
+    },
+    {
+      "epoch": 6.534195933456562,
+      "grad_norm": 0.06264297664165497,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 63630
+    },
+    {
+      "epoch": 6.534504004929143,
+      "grad_norm": 0.06047286093235016,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 63633
+    },
+    {
+      "epoch": 6.5348120764017255,
+      "grad_norm": 0.08621931076049805,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 63636
+    },
+    {
+      "epoch": 6.535120147874307,
+      "grad_norm": 0.06598355621099472,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 63639
+    },
+    {
+      "epoch": 6.535428219346889,
+      "grad_norm": 0.04652687907218933,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 63642
+    },
+    {
+      "epoch": 6.53573629081947,
+      "grad_norm": 0.04016298055648804,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 63645
+    },
+    {
+      "epoch": 6.536044362292052,
+      "grad_norm": 0.039485424757003784,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 63648
+    },
+    {
+      "epoch": 6.536352433764633,
+      "grad_norm": 0.03557172790169716,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 63651
+    },
+    {
+      "epoch": 6.536660505237215,
+      "grad_norm": 0.0416797399520874,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 63654
+    },
+    {
+      "epoch": 6.536968576709796,
+      "grad_norm": 0.09189657121896744,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 63657
+    },
+    {
+      "epoch": 6.537276648182378,
+      "grad_norm": 0.04484305903315544,
+      "learning_rate": 0.01,
+      "loss": 2.0131,
+      "step": 63660
+    },
+    {
+      "epoch": 6.53758471965496,
+      "grad_norm": 0.059794455766677856,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 63663
+    },
+    {
+      "epoch": 6.5378927911275415,
+      "grad_norm": 0.05551350116729736,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 63666
+    },
+    {
+      "epoch": 6.538200862600124,
+      "grad_norm": 0.19956345856189728,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 63669
+    },
+    {
+      "epoch": 6.538508934072705,
+      "grad_norm": 0.0885189101099968,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 63672
+    },
+    {
+      "epoch": 6.538817005545287,
+      "grad_norm": 0.05539345368742943,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 63675
+    },
+    {
+      "epoch": 6.539125077017868,
+      "grad_norm": 0.037228211760520935,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 63678
+    },
+    {
+      "epoch": 6.53943314849045,
+      "grad_norm": 0.03209219127893448,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 63681
+    },
+    {
+      "epoch": 6.539741219963031,
+      "grad_norm": 0.029509764164686203,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 63684
+    },
+    {
+      "epoch": 6.540049291435613,
+      "grad_norm": 0.052676524966955185,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 63687
+    },
+    {
+      "epoch": 6.540357362908194,
+      "grad_norm": 0.0744934156537056,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 63690
+    },
+    {
+      "epoch": 6.540665434380776,
+      "grad_norm": 0.05411761999130249,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 63693
+    },
+    {
+      "epoch": 6.5409735058533585,
+      "grad_norm": 0.04553327336907387,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 63696
+    },
+    {
+      "epoch": 6.54128157732594,
+      "grad_norm": 0.16345827281475067,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 63699
+    },
+    {
+      "epoch": 6.541589648798521,
+      "grad_norm": 0.05779232829809189,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 63702
+    },
+    {
+      "epoch": 6.541897720271103,
+      "grad_norm": 0.032575078308582306,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 63705
+    },
+    {
+      "epoch": 6.542205791743685,
+      "grad_norm": 0.043234266340732574,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 63708
+    },
+    {
+      "epoch": 6.542513863216266,
+      "grad_norm": 0.050926946103572845,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 63711
+    },
+    {
+      "epoch": 6.542821934688848,
+      "grad_norm": 0.040199290961027145,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 63714
+    },
+    {
+      "epoch": 6.543130006161429,
+      "grad_norm": 0.0373854786157608,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 63717
+    },
+    {
+      "epoch": 6.543438077634011,
+      "grad_norm": 0.04420515522360802,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 63720
+    },
+    {
+      "epoch": 6.5437461491065925,
+      "grad_norm": 0.10335779935121536,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 63723
+    },
+    {
+      "epoch": 6.5440542205791745,
+      "grad_norm": 0.08750100433826447,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 63726
+    },
+    {
+      "epoch": 6.544362292051756,
+      "grad_norm": 0.08968067169189453,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 63729
+    },
+    {
+      "epoch": 6.544670363524338,
+      "grad_norm": 0.05931568518280983,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 63732
+    },
+    {
+      "epoch": 6.544978434996919,
+      "grad_norm": 0.04726015031337738,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 63735
+    },
+    {
+      "epoch": 6.545286506469501,
+      "grad_norm": 0.048892538994550705,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 63738
+    },
+    {
+      "epoch": 6.545594577942083,
+      "grad_norm": 0.042674802243709564,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 63741
+    },
+    {
+      "epoch": 6.545902649414664,
+      "grad_norm": 0.11424809694290161,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 63744
+    },
+    {
+      "epoch": 6.546210720887246,
+      "grad_norm": 0.037352122366428375,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 63747
+    },
+    {
+      "epoch": 6.546518792359827,
+      "grad_norm": 0.04247456043958664,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 63750
+    },
+    {
+      "epoch": 6.546826863832409,
+      "grad_norm": 0.05194523185491562,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 63753
+    },
+    {
+      "epoch": 6.547134935304991,
+      "grad_norm": 0.055400025099515915,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 63756
+    },
+    {
+      "epoch": 6.547443006777573,
+      "grad_norm": 0.06295850872993469,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 63759
+    },
+    {
+      "epoch": 6.547751078250154,
+      "grad_norm": 0.05281165987253189,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 63762
+    },
+    {
+      "epoch": 6.548059149722736,
+      "grad_norm": 0.04838143289089203,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 63765
+    },
+    {
+      "epoch": 6.548367221195317,
+      "grad_norm": 0.03333796560764313,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 63768
+    },
+    {
+      "epoch": 6.548675292667899,
+      "grad_norm": 0.029952581971883774,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 63771
+    },
+    {
+      "epoch": 6.54898336414048,
+      "grad_norm": 0.03036186844110489,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 63774
+    },
+    {
+      "epoch": 6.549291435613062,
+      "grad_norm": 0.06050482019782066,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 63777
+    },
+    {
+      "epoch": 6.549599507085643,
+      "grad_norm": 0.08855880796909332,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 63780
+    },
+    {
+      "epoch": 6.5499075785582255,
+      "grad_norm": 0.08679243922233582,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 63783
+    },
+    {
+      "epoch": 6.5502156500308075,
+      "grad_norm": 0.04234171286225319,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 63786
+    },
+    {
+      "epoch": 6.550523721503389,
+      "grad_norm": 0.029922012239694595,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 63789
+    },
+    {
+      "epoch": 6.550831792975971,
+      "grad_norm": 0.11024188250303268,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 63792
+    },
+    {
+      "epoch": 6.551139864448552,
+      "grad_norm": 0.03447732329368591,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 63795
+    },
+    {
+      "epoch": 6.551447935921134,
+      "grad_norm": 0.04188188910484314,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 63798
+    },
+    {
+      "epoch": 6.551756007393715,
+      "grad_norm": 0.040945570915937424,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 63801
+    },
+    {
+      "epoch": 6.552064078866297,
+      "grad_norm": 0.07144046574831009,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 63804
+    },
+    {
+      "epoch": 6.552372150338878,
+      "grad_norm": 0.043573856353759766,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 63807
+    },
+    {
+      "epoch": 6.55268022181146,
+      "grad_norm": 0.04076509550213814,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 63810
+    },
+    {
+      "epoch": 6.5529882932840415,
+      "grad_norm": 0.042045023292303085,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 63813
+    },
+    {
+      "epoch": 6.553296364756624,
+      "grad_norm": 0.047036029398441315,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 63816
+    },
+    {
+      "epoch": 6.553604436229206,
+      "grad_norm": 0.08206555247306824,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 63819
+    },
+    {
+      "epoch": 6.553912507701787,
+      "grad_norm": 0.04344995692372322,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 63822
+    },
+    {
+      "epoch": 6.554220579174368,
+      "grad_norm": 0.05620454624295235,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 63825
+    },
+    {
+      "epoch": 6.55452865064695,
+      "grad_norm": 0.042846135795116425,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 63828
+    },
+    {
+      "epoch": 6.554836722119532,
+      "grad_norm": 0.04379443824291229,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 63831
+    },
+    {
+      "epoch": 6.555144793592113,
+      "grad_norm": 0.050486281514167786,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 63834
+    },
+    {
+      "epoch": 6.555452865064695,
+      "grad_norm": 0.10757488012313843,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 63837
+    },
+    {
+      "epoch": 6.555760936537276,
+      "grad_norm": 0.09059060364961624,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 63840
+    },
+    {
+      "epoch": 6.5560690080098585,
+      "grad_norm": 0.055471502244472504,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 63843
+    },
+    {
+      "epoch": 6.55637707948244,
+      "grad_norm": 0.10934463888406754,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 63846
+    },
+    {
+      "epoch": 6.556685150955022,
+      "grad_norm": 0.08888571709394455,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 63849
+    },
+    {
+      "epoch": 6.556993222427603,
+      "grad_norm": 0.11083342134952545,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 63852
+    },
+    {
+      "epoch": 6.557301293900185,
+      "grad_norm": 0.07405796647071838,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 63855
+    },
+    {
+      "epoch": 6.557609365372766,
+      "grad_norm": 0.08451070636510849,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 63858
+    },
+    {
+      "epoch": 6.557917436845348,
+      "grad_norm": 0.05954553931951523,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 63861
+    },
+    {
+      "epoch": 6.55822550831793,
+      "grad_norm": 0.06008210405707359,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 63864
+    },
+    {
+      "epoch": 6.558533579790511,
+      "grad_norm": 0.0861048474907875,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 63867
+    },
+    {
+      "epoch": 6.558841651263093,
+      "grad_norm": 0.06240614131093025,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 63870
+    },
+    {
+      "epoch": 6.5591497227356745,
+      "grad_norm": 0.05393391102552414,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 63873
+    },
+    {
+      "epoch": 6.559457794208257,
+      "grad_norm": 0.08556719869375229,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 63876
+    },
+    {
+      "epoch": 6.559765865680838,
+      "grad_norm": 0.045306041836738586,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 63879
+    },
+    {
+      "epoch": 6.56007393715342,
+      "grad_norm": 0.11380963772535324,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 63882
+    },
+    {
+      "epoch": 6.560382008626001,
+      "grad_norm": 0.04248273745179176,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 63885
+    },
+    {
+      "epoch": 6.560690080098583,
+      "grad_norm": 0.040910713374614716,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 63888
+    },
+    {
+      "epoch": 6.560998151571164,
+      "grad_norm": 0.046450424939394,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 63891
+    },
+    {
+      "epoch": 6.561306223043746,
+      "grad_norm": 0.07835663855075836,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 63894
+    },
+    {
+      "epoch": 6.561614294516328,
+      "grad_norm": 0.04756823554635048,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 63897
+    },
+    {
+      "epoch": 6.561922365988909,
+      "grad_norm": 0.10655516386032104,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 63900
+    },
+    {
+      "epoch": 6.562230437461491,
+      "grad_norm": 0.08743315935134888,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 63903
+    },
+    {
+      "epoch": 6.562538508934073,
+      "grad_norm": 0.05087851360440254,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 63906
+    },
+    {
+      "epoch": 6.562846580406655,
+      "grad_norm": 0.03635932877659798,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 63909
+    },
+    {
+      "epoch": 6.563154651879236,
+      "grad_norm": 0.09845136106014252,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 63912
+    },
+    {
+      "epoch": 6.563462723351818,
+      "grad_norm": 0.03554631769657135,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 63915
+    },
+    {
+      "epoch": 6.563770794824399,
+      "grad_norm": 0.053068485110998154,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 63918
+    },
+    {
+      "epoch": 6.564078866296981,
+      "grad_norm": 0.09752979129552841,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 63921
+    },
+    {
+      "epoch": 6.564386937769562,
+      "grad_norm": 0.10045387595891953,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 63924
+    },
+    {
+      "epoch": 6.564695009242144,
+      "grad_norm": 0.1082577034831047,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 63927
+    },
+    {
+      "epoch": 6.5650030807147255,
+      "grad_norm": 0.09065879136323929,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 63930
+    },
+    {
+      "epoch": 6.5653111521873075,
+      "grad_norm": 0.06496050953865051,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 63933
+    },
+    {
+      "epoch": 6.565619223659889,
+      "grad_norm": 0.07092882692813873,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 63936
+    },
+    {
+      "epoch": 6.565927295132471,
+      "grad_norm": 0.048295971006155014,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 63939
+    },
+    {
+      "epoch": 6.566235366605053,
+      "grad_norm": 0.045929424464702606,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 63942
+    },
+    {
+      "epoch": 6.566543438077634,
+      "grad_norm": 0.03663928061723709,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 63945
+    },
+    {
+      "epoch": 6.566851509550216,
+      "grad_norm": 0.03723525255918503,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 63948
+    },
+    {
+      "epoch": 6.567159581022797,
+      "grad_norm": 0.05147719010710716,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 63951
+    },
+    {
+      "epoch": 6.567467652495379,
+      "grad_norm": 0.04248051717877388,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 63954
+    },
+    {
+      "epoch": 6.56777572396796,
+      "grad_norm": 0.054685961455106735,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 63957
+    },
+    {
+      "epoch": 6.568083795440542,
+      "grad_norm": 0.07126769423484802,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 63960
+    },
+    {
+      "epoch": 6.568391866913124,
+      "grad_norm": 0.0660753920674324,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 63963
+    },
+    {
+      "epoch": 6.568699938385706,
+      "grad_norm": 0.04418491572141647,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 63966
+    },
+    {
+      "epoch": 6.569008009858287,
+      "grad_norm": 0.03849758952856064,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 63969
+    },
+    {
+      "epoch": 6.569316081330869,
+      "grad_norm": 0.0360284261405468,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 63972
+    },
+    {
+      "epoch": 6.56962415280345,
+      "grad_norm": 0.04783300310373306,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 63975
+    },
+    {
+      "epoch": 6.569932224276032,
+      "grad_norm": 0.104099340736866,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 63978
+    },
+    {
+      "epoch": 6.570240295748613,
+      "grad_norm": 0.06214939430356026,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 63981
+    },
+    {
+      "epoch": 6.570548367221195,
+      "grad_norm": 0.03583189472556114,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 63984
+    },
+    {
+      "epoch": 6.570856438693777,
+      "grad_norm": 0.03806734457612038,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 63987
+    },
+    {
+      "epoch": 6.5711645101663585,
+      "grad_norm": 0.037077397108078,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 63990
+    },
+    {
+      "epoch": 6.5714725816389405,
+      "grad_norm": 0.1366073340177536,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 63993
+    },
+    {
+      "epoch": 6.571780653111522,
+      "grad_norm": 0.06739377975463867,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 63996
+    },
+    {
+      "epoch": 6.572088724584104,
+      "grad_norm": 0.041534364223480225,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 63999
+    },
+    {
+      "epoch": 6.572396796056685,
+      "grad_norm": 0.04053931310772896,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 64002
+    },
+    {
+      "epoch": 6.572704867529267,
+      "grad_norm": 0.04500468447804451,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 64005
+    },
+    {
+      "epoch": 6.573012939001848,
+      "grad_norm": 0.07406989485025406,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 64008
+    },
+    {
+      "epoch": 6.57332101047443,
+      "grad_norm": 0.05247531831264496,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 64011
+    },
+    {
+      "epoch": 6.573629081947011,
+      "grad_norm": 0.117593914270401,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 64014
+    },
+    {
+      "epoch": 6.573937153419593,
+      "grad_norm": 0.07242528349161148,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 64017
+    },
+    {
+      "epoch": 6.574245224892175,
+      "grad_norm": 0.05359083414077759,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 64020
+    },
+    {
+      "epoch": 6.574553296364757,
+      "grad_norm": 0.036466315388679504,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 64023
+    },
+    {
+      "epoch": 6.574861367837338,
+      "grad_norm": 0.03925064578652382,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 64026
+    },
+    {
+      "epoch": 6.57516943930992,
+      "grad_norm": 0.05542049929499626,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 64029
+    },
+    {
+      "epoch": 6.575477510782502,
+      "grad_norm": 0.06237075477838516,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 64032
+    },
+    {
+      "epoch": 6.575785582255083,
+      "grad_norm": 0.04179569333791733,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 64035
+    },
+    {
+      "epoch": 6.576093653727665,
+      "grad_norm": 0.11507870256900787,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 64038
+    },
+    {
+      "epoch": 6.576401725200246,
+      "grad_norm": 0.06069584935903549,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 64041
+    },
+    {
+      "epoch": 6.576709796672828,
+      "grad_norm": 0.041366275399923325,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 64044
+    },
+    {
+      "epoch": 6.5770178681454095,
+      "grad_norm": 0.06424157321453094,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 64047
+    },
+    {
+      "epoch": 6.5773259396179915,
+      "grad_norm": 0.06879021972417831,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 64050
+    },
+    {
+      "epoch": 6.577634011090573,
+      "grad_norm": 0.043172743171453476,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 64053
+    },
+    {
+      "epoch": 6.577942082563155,
+      "grad_norm": 0.05824809893965721,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 64056
+    },
+    {
+      "epoch": 6.578250154035736,
+      "grad_norm": 0.045327652245759964,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 64059
+    },
+    {
+      "epoch": 6.578558225508318,
+      "grad_norm": 0.11322353035211563,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 64062
+    },
+    {
+      "epoch": 6.5788662969809,
+      "grad_norm": 0.06283524632453918,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 64065
+    },
+    {
+      "epoch": 6.579174368453481,
+      "grad_norm": 0.06951749324798584,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 64068
+    },
+    {
+      "epoch": 6.579482439926063,
+      "grad_norm": 0.1091168001294136,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 64071
+    },
+    {
+      "epoch": 6.579790511398644,
+      "grad_norm": 0.0779736191034317,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 64074
+    },
+    {
+      "epoch": 6.580098582871226,
+      "grad_norm": 0.06814195215702057,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 64077
+    },
+    {
+      "epoch": 6.580406654343808,
+      "grad_norm": 0.06565315276384354,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 64080
+    },
+    {
+      "epoch": 6.58071472581639,
+      "grad_norm": 0.08950956165790558,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 64083
+    },
+    {
+      "epoch": 6.581022797288971,
+      "grad_norm": 0.06202126666903496,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 64086
+    },
+    {
+      "epoch": 6.581330868761553,
+      "grad_norm": 0.10947906970977783,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 64089
+    },
+    {
+      "epoch": 6.581638940234134,
+      "grad_norm": 0.051477983593940735,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 64092
+    },
+    {
+      "epoch": 6.581947011706716,
+      "grad_norm": 0.05183662101626396,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 64095
+    },
+    {
+      "epoch": 6.582255083179298,
+      "grad_norm": 0.04651165008544922,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 64098
+    },
+    {
+      "epoch": 6.582563154651879,
+      "grad_norm": 0.035913266241550446,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 64101
+    },
+    {
+      "epoch": 6.58287122612446,
+      "grad_norm": 0.0991610661149025,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 64104
+    },
+    {
+      "epoch": 6.5831792975970425,
+      "grad_norm": 0.04391265660524368,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 64107
+    },
+    {
+      "epoch": 6.5834873690696245,
+      "grad_norm": 0.14331793785095215,
+      "learning_rate": 0.01,
+      "loss": 2.026,
+      "step": 64110
+    },
+    {
+      "epoch": 6.583795440542206,
+      "grad_norm": 0.10682855546474457,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 64113
+    },
+    {
+      "epoch": 6.584103512014788,
+      "grad_norm": 0.07142464816570282,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 64116
+    },
+    {
+      "epoch": 6.584411583487369,
+      "grad_norm": 0.05298778787255287,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 64119
+    },
+    {
+      "epoch": 6.584719654959951,
+      "grad_norm": 0.10031582415103912,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 64122
+    },
+    {
+      "epoch": 6.585027726432532,
+      "grad_norm": 0.03978649899363518,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 64125
+    },
+    {
+      "epoch": 6.585335797905114,
+      "grad_norm": 0.05913195759057999,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 64128
+    },
+    {
+      "epoch": 6.585643869377695,
+      "grad_norm": 0.03559190407395363,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 64131
+    },
+    {
+      "epoch": 6.585951940850277,
+      "grad_norm": 0.09079554677009583,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 64134
+    },
+    {
+      "epoch": 6.5862600123228585,
+      "grad_norm": 0.04525060951709747,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 64137
+    },
+    {
+      "epoch": 6.586568083795441,
+      "grad_norm": 0.03372851014137268,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 64140
+    },
+    {
+      "epoch": 6.586876155268023,
+      "grad_norm": 0.08057595044374466,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 64143
+    },
+    {
+      "epoch": 6.587184226740604,
+      "grad_norm": 0.09445854276418686,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 64146
+    },
+    {
+      "epoch": 6.587492298213186,
+      "grad_norm": 0.07656941562891006,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 64149
+    },
+    {
+      "epoch": 6.587800369685767,
+      "grad_norm": 0.08286568522453308,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 64152
+    },
+    {
+      "epoch": 6.588108441158349,
+      "grad_norm": 0.11229046434164047,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 64155
+    },
+    {
+      "epoch": 6.58841651263093,
+      "grad_norm": 0.06434221565723419,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 64158
+    },
+    {
+      "epoch": 6.588724584103512,
+      "grad_norm": 0.08923232555389404,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 64161
+    },
+    {
+      "epoch": 6.589032655576093,
+      "grad_norm": 0.07555590569972992,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 64164
+    },
+    {
+      "epoch": 6.5893407270486755,
+      "grad_norm": 0.05860813334584236,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 64167
+    },
+    {
+      "epoch": 6.589648798521257,
+      "grad_norm": 0.03441310673952103,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 64170
+    },
+    {
+      "epoch": 6.589956869993839,
+      "grad_norm": 0.0361168198287487,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 64173
+    },
+    {
+      "epoch": 6.59026494146642,
+      "grad_norm": 0.04222318157553673,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 64176
+    },
+    {
+      "epoch": 6.590573012939002,
+      "grad_norm": 0.12981876730918884,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 64179
+    },
+    {
+      "epoch": 6.590881084411583,
+      "grad_norm": 0.05409705266356468,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 64182
+    },
+    {
+      "epoch": 6.591189155884165,
+      "grad_norm": 0.053002189844846725,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 64185
+    },
+    {
+      "epoch": 6.591497227356747,
+      "grad_norm": 0.061368245631456375,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 64188
+    },
+    {
+      "epoch": 6.591805298829328,
+      "grad_norm": 0.10517262667417526,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 64191
+    },
+    {
+      "epoch": 6.59211337030191,
+      "grad_norm": 0.06446947157382965,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 64194
+    },
+    {
+      "epoch": 6.5924214417744915,
+      "grad_norm": 0.06423608213663101,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 64197
+    },
+    {
+      "epoch": 6.592729513247074,
+      "grad_norm": 0.034221209585666656,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 64200
+    },
+    {
+      "epoch": 6.593037584719655,
+      "grad_norm": 0.12078302353620529,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 64203
+    },
+    {
+      "epoch": 6.593345656192237,
+      "grad_norm": 0.08189702033996582,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 64206
+    },
+    {
+      "epoch": 6.593653727664818,
+      "grad_norm": 0.0505688339471817,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 64209
+    },
+    {
+      "epoch": 6.5939617991374,
+      "grad_norm": 0.05289214104413986,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 64212
+    },
+    {
+      "epoch": 6.594269870609981,
+      "grad_norm": 0.0513903982937336,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 64215
+    },
+    {
+      "epoch": 6.594577942082563,
+      "grad_norm": 0.06634233146905899,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 64218
+    },
+    {
+      "epoch": 6.594886013555145,
+      "grad_norm": 0.0847700983285904,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 64221
+    },
+    {
+      "epoch": 6.595194085027726,
+      "grad_norm": 0.08042684942483902,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 64224
+    },
+    {
+      "epoch": 6.595502156500308,
+      "grad_norm": 0.046448756009340286,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 64227
+    },
+    {
+      "epoch": 6.59581022797289,
+      "grad_norm": 0.09152430295944214,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 64230
+    },
+    {
+      "epoch": 6.596118299445472,
+      "grad_norm": 0.05537572130560875,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 64233
+    },
+    {
+      "epoch": 6.596426370918053,
+      "grad_norm": 0.040232930332422256,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 64236
+    },
+    {
+      "epoch": 6.596734442390635,
+      "grad_norm": 0.040378376841545105,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 64239
+    },
+    {
+      "epoch": 6.597042513863216,
+      "grad_norm": 0.036219045519828796,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 64242
+    },
+    {
+      "epoch": 6.597350585335798,
+      "grad_norm": 0.0671059712767601,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 64245
+    },
+    {
+      "epoch": 6.597658656808379,
+      "grad_norm": 0.05195783078670502,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 64248
+    },
+    {
+      "epoch": 6.597966728280961,
+      "grad_norm": 0.05707523226737976,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 64251
+    },
+    {
+      "epoch": 6.5982747997535425,
+      "grad_norm": 0.10035475343465805,
+      "learning_rate": 0.01,
+      "loss": 2.0062,
+      "step": 64254
+    },
+    {
+      "epoch": 6.5985828712261245,
+      "grad_norm": 0.08237627893686295,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 64257
+    },
+    {
+      "epoch": 6.598890942698706,
+      "grad_norm": 0.06940212100744247,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 64260
+    },
+    {
+      "epoch": 6.599199014171288,
+      "grad_norm": 0.08880186080932617,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 64263
+    },
+    {
+      "epoch": 6.59950708564387,
+      "grad_norm": 0.07859902083873749,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 64266
+    },
+    {
+      "epoch": 6.599815157116451,
+      "grad_norm": 0.10869091004133224,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 64269
+    },
+    {
+      "epoch": 6.600123228589033,
+      "grad_norm": 0.09375559538602829,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 64272
+    },
+    {
+      "epoch": 6.600431300061614,
+      "grad_norm": 0.047256410121917725,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 64275
+    },
+    {
+      "epoch": 6.600739371534196,
+      "grad_norm": 0.04523398354649544,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 64278
+    },
+    {
+      "epoch": 6.601047443006777,
+      "grad_norm": 0.03981922194361687,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 64281
+    },
+    {
+      "epoch": 6.601355514479359,
+      "grad_norm": 0.05818277969956398,
+      "learning_rate": 0.01,
+      "loss": 2.002,
+      "step": 64284
+    },
+    {
+      "epoch": 6.601663585951941,
+      "grad_norm": 0.04212537035346031,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 64287
+    },
+    {
+      "epoch": 6.601971657424523,
+      "grad_norm": 0.033235009759664536,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 64290
+    },
+    {
+      "epoch": 6.602279728897104,
+      "grad_norm": 0.0338917002081871,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 64293
+    },
+    {
+      "epoch": 6.602587800369686,
+      "grad_norm": 0.07970410585403442,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 64296
+    },
+    {
+      "epoch": 6.602895871842268,
+      "grad_norm": 0.05640435963869095,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 64299
+    },
+    {
+      "epoch": 6.603203943314849,
+      "grad_norm": 0.03606804460287094,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 64302
+    },
+    {
+      "epoch": 6.60351201478743,
+      "grad_norm": 0.04679587855935097,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 64305
+    },
+    {
+      "epoch": 6.603820086260012,
+      "grad_norm": 0.10526128858327866,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 64308
+    },
+    {
+      "epoch": 6.604128157732594,
+      "grad_norm": 0.1200772076845169,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 64311
+    },
+    {
+      "epoch": 6.6044362292051755,
+      "grad_norm": 0.1391524225473404,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 64314
+    },
+    {
+      "epoch": 6.6047443006777575,
+      "grad_norm": 0.08838554471731186,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 64317
+    },
+    {
+      "epoch": 6.605052372150339,
+      "grad_norm": 0.07253210991621017,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 64320
+    },
+    {
+      "epoch": 6.605360443622921,
+      "grad_norm": 0.04417060315608978,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 64323
+    },
+    {
+      "epoch": 6.605668515095502,
+      "grad_norm": 0.04443688318133354,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 64326
+    },
+    {
+      "epoch": 6.605976586568084,
+      "grad_norm": 0.05711906775832176,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 64329
+    },
+    {
+      "epoch": 6.606284658040665,
+      "grad_norm": 0.05172707885503769,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 64332
+    },
+    {
+      "epoch": 6.606592729513247,
+      "grad_norm": 0.08967714011669159,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 64335
+    },
+    {
+      "epoch": 6.606900800985828,
+      "grad_norm": 0.03300314024090767,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 64338
+    },
+    {
+      "epoch": 6.60720887245841,
+      "grad_norm": 0.052630916237831116,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 64341
+    },
+    {
+      "epoch": 6.607516943930992,
+      "grad_norm": 0.041368499398231506,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 64344
+    },
+    {
+      "epoch": 6.607825015403574,
+      "grad_norm": 0.04549229145050049,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 64347
+    },
+    {
+      "epoch": 6.608133086876156,
+      "grad_norm": 0.057475391775369644,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 64350
+    },
+    {
+      "epoch": 6.608441158348737,
+      "grad_norm": 0.056770164519548416,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 64353
+    },
+    {
+      "epoch": 6.608749229821319,
+      "grad_norm": 0.12769392132759094,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 64356
+    },
+    {
+      "epoch": 6.6090573012939,
+      "grad_norm": 0.05759736895561218,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 64359
+    },
+    {
+      "epoch": 6.609365372766482,
+      "grad_norm": 0.10004235059022903,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 64362
+    },
+    {
+      "epoch": 6.609673444239063,
+      "grad_norm": 0.0365755669772625,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 64365
+    },
+    {
+      "epoch": 6.609981515711645,
+      "grad_norm": 0.07999947667121887,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 64368
+    },
+    {
+      "epoch": 6.610289587184226,
+      "grad_norm": 0.055819109082221985,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 64371
+    },
+    {
+      "epoch": 6.6105976586568085,
+      "grad_norm": 0.046389635652303696,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 64374
+    },
+    {
+      "epoch": 6.61090573012939,
+      "grad_norm": 0.04378519952297211,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 64377
+    },
+    {
+      "epoch": 6.611213801601972,
+      "grad_norm": 0.04519166797399521,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 64380
+    },
+    {
+      "epoch": 6.611521873074553,
+      "grad_norm": 0.037725985050201416,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 64383
+    },
+    {
+      "epoch": 6.611829944547135,
+      "grad_norm": 0.03793521970510483,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 64386
+    },
+    {
+      "epoch": 6.612138016019717,
+      "grad_norm": 0.04742049798369408,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 64389
+    },
+    {
+      "epoch": 6.612446087492298,
+      "grad_norm": 0.059588026255369186,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 64392
+    },
+    {
+      "epoch": 6.61275415896488,
+      "grad_norm": 0.06134435534477234,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 64395
+    },
+    {
+      "epoch": 6.613062230437461,
+      "grad_norm": 0.11348873376846313,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 64398
+    },
+    {
+      "epoch": 6.613370301910043,
+      "grad_norm": 0.03893901780247688,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 64401
+    },
+    {
+      "epoch": 6.6136783733826245,
+      "grad_norm": 0.11874943226575851,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 64404
+    },
+    {
+      "epoch": 6.613986444855207,
+      "grad_norm": 0.08608514815568924,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 64407
+    },
+    {
+      "epoch": 6.614294516327788,
+      "grad_norm": 0.04932364821434021,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 64410
+    },
+    {
+      "epoch": 6.61460258780037,
+      "grad_norm": 0.04308156296610832,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 64413
+    },
+    {
+      "epoch": 6.614910659272951,
+      "grad_norm": 0.050084516406059265,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 64416
+    },
+    {
+      "epoch": 6.615218730745533,
+      "grad_norm": 0.054838646203279495,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 64419
+    },
+    {
+      "epoch": 6.615526802218115,
+      "grad_norm": 0.08248993754386902,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 64422
+    },
+    {
+      "epoch": 6.615834873690696,
+      "grad_norm": 0.09142714738845825,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 64425
+    },
+    {
+      "epoch": 6.616142945163277,
+      "grad_norm": 0.07664018124341965,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 64428
+    },
+    {
+      "epoch": 6.616451016635859,
+      "grad_norm": 0.10052567720413208,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 64431
+    },
+    {
+      "epoch": 6.6167590881084415,
+      "grad_norm": 0.04676009342074394,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 64434
+    },
+    {
+      "epoch": 6.617067159581023,
+      "grad_norm": 0.05421235412359238,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 64437
+    },
+    {
+      "epoch": 6.617375231053605,
+      "grad_norm": 0.05721297487616539,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 64440
+    },
+    {
+      "epoch": 6.617683302526186,
+      "grad_norm": 0.05209130793809891,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 64443
+    },
+    {
+      "epoch": 6.617991373998768,
+      "grad_norm": 0.032787423580884933,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 64446
+    },
+    {
+      "epoch": 6.618299445471349,
+      "grad_norm": 0.11715228855609894,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 64449
+    },
+    {
+      "epoch": 6.618607516943931,
+      "grad_norm": 0.044540733098983765,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 64452
+    },
+    {
+      "epoch": 6.618915588416512,
+      "grad_norm": 0.06670498847961426,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 64455
+    },
+    {
+      "epoch": 6.619223659889094,
+      "grad_norm": 0.09549476206302643,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 64458
+    },
+    {
+      "epoch": 6.6195317313616755,
+      "grad_norm": 0.06052359193563461,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 64461
+    },
+    {
+      "epoch": 6.6198398028342575,
+      "grad_norm": 0.04301074892282486,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 64464
+    },
+    {
+      "epoch": 6.62014787430684,
+      "grad_norm": 0.06434419006109238,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 64467
+    },
+    {
+      "epoch": 6.620455945779421,
+      "grad_norm": 0.051222655922174454,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 64470
+    },
+    {
+      "epoch": 6.620764017252003,
+      "grad_norm": 0.06115943565964699,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 64473
+    },
+    {
+      "epoch": 6.621072088724584,
+      "grad_norm": 0.07994543761014938,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 64476
+    },
+    {
+      "epoch": 6.621380160197166,
+      "grad_norm": 0.06953758746385574,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 64479
+    },
+    {
+      "epoch": 6.621688231669747,
+      "grad_norm": 0.12298289686441422,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 64482
+    },
+    {
+      "epoch": 6.621996303142329,
+      "grad_norm": 0.05092281848192215,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 64485
+    },
+    {
+      "epoch": 6.62230437461491,
+      "grad_norm": 0.043478354811668396,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 64488
+    },
+    {
+      "epoch": 6.622612446087492,
+      "grad_norm": 0.043610502034425735,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 64491
+    },
+    {
+      "epoch": 6.622920517560074,
+      "grad_norm": 0.052348487079143524,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 64494
+    },
+    {
+      "epoch": 6.623228589032656,
+      "grad_norm": 0.05436374992132187,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 64497
+    },
+    {
+      "epoch": 6.623536660505238,
+      "grad_norm": 0.057916752994060516,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 64500
+    },
+    {
+      "epoch": 6.623844731977819,
+      "grad_norm": 0.047713443636894226,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 64503
+    },
+    {
+      "epoch": 6.6241528034504,
+      "grad_norm": 0.03382120653986931,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 64506
+    },
+    {
+      "epoch": 6.624460874922982,
+      "grad_norm": 0.0990116074681282,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 64509
+    },
+    {
+      "epoch": 6.624768946395564,
+      "grad_norm": 0.0787537470459938,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 64512
+    },
+    {
+      "epoch": 6.625077017868145,
+      "grad_norm": 0.07667285948991776,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 64515
+    },
+    {
+      "epoch": 6.625385089340727,
+      "grad_norm": 0.04837285354733467,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 64518
+    },
+    {
+      "epoch": 6.6256931608133085,
+      "grad_norm": 0.04143200442194939,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 64521
+    },
+    {
+      "epoch": 6.6260012322858906,
+      "grad_norm": 0.033114057034254074,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 64524
+    },
+    {
+      "epoch": 6.626309303758472,
+      "grad_norm": 0.0412205308675766,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 64527
+    },
+    {
+      "epoch": 6.626617375231054,
+      "grad_norm": 0.10429148375988007,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 64530
+    },
+    {
+      "epoch": 6.626925446703635,
+      "grad_norm": 0.05905142053961754,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 64533
+    },
+    {
+      "epoch": 6.627233518176217,
+      "grad_norm": 0.05937071144580841,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 64536
+    },
+    {
+      "epoch": 6.627541589648798,
+      "grad_norm": 0.10505139082670212,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 64539
+    },
+    {
+      "epoch": 6.62784966112138,
+      "grad_norm": 0.08412894606590271,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 64542
+    },
+    {
+      "epoch": 6.628157732593962,
+      "grad_norm": 0.07351752370595932,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 64545
+    },
+    {
+      "epoch": 6.628465804066543,
+      "grad_norm": 0.07745961844921112,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 64548
+    },
+    {
+      "epoch": 6.6287738755391254,
+      "grad_norm": 0.07289955765008926,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 64551
+    },
+    {
+      "epoch": 6.629081947011707,
+      "grad_norm": 0.04405929520726204,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 64554
+    },
+    {
+      "epoch": 6.629390018484289,
+      "grad_norm": 0.0912909060716629,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 64557
+    },
+    {
+      "epoch": 6.62969808995687,
+      "grad_norm": 0.042689189314842224,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 64560
+    },
+    {
+      "epoch": 6.630006161429452,
+      "grad_norm": 0.14025220274925232,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 64563
+    },
+    {
+      "epoch": 6.630314232902033,
+      "grad_norm": 0.1245136708021164,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 64566
+    },
+    {
+      "epoch": 6.630622304374615,
+      "grad_norm": 0.09421566873788834,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 64569
+    },
+    {
+      "epoch": 6.630930375847196,
+      "grad_norm": 0.0803411453962326,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 64572
+    },
+    {
+      "epoch": 6.631238447319778,
+      "grad_norm": 0.06703267991542816,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 64575
+    },
+    {
+      "epoch": 6.6315465187923595,
+      "grad_norm": 0.1008155420422554,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 64578
+    },
+    {
+      "epoch": 6.6318545902649415,
+      "grad_norm": 0.08870470523834229,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 64581
+    },
+    {
+      "epoch": 6.632162661737523,
+      "grad_norm": 0.06180182099342346,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 64584
+    },
+    {
+      "epoch": 6.632470733210105,
+      "grad_norm": 0.06106216087937355,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 64587
+    },
+    {
+      "epoch": 6.632778804682687,
+      "grad_norm": 0.05340215936303139,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 64590
+    },
+    {
+      "epoch": 6.633086876155268,
+      "grad_norm": 0.04080619663000107,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 64593
+    },
+    {
+      "epoch": 6.63339494762785,
+      "grad_norm": 0.031813837587833405,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 64596
+    },
+    {
+      "epoch": 6.633703019100431,
+      "grad_norm": 0.0589885450899601,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 64599
+    },
+    {
+      "epoch": 6.634011090573013,
+      "grad_norm": 0.14397123456001282,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 64602
+    },
+    {
+      "epoch": 6.634319162045594,
+      "grad_norm": 0.058227211236953735,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 64605
+    },
+    {
+      "epoch": 6.634627233518176,
+      "grad_norm": 0.05160403624176979,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 64608
+    },
+    {
+      "epoch": 6.634935304990758,
+      "grad_norm": 0.052490971982479095,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 64611
+    },
+    {
+      "epoch": 6.63524337646334,
+      "grad_norm": 0.06203007325530052,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 64614
+    },
+    {
+      "epoch": 6.635551447935921,
+      "grad_norm": 0.06790245324373245,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 64617
+    },
+    {
+      "epoch": 6.635859519408503,
+      "grad_norm": 0.0575333833694458,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 64620
+    },
+    {
+      "epoch": 6.636167590881085,
+      "grad_norm": 0.05253339931368828,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 64623
+    },
+    {
+      "epoch": 6.636475662353666,
+      "grad_norm": 0.0605839341878891,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 64626
+    },
+    {
+      "epoch": 6.636783733826247,
+      "grad_norm": 0.046164620667696,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 64629
+    },
+    {
+      "epoch": 6.637091805298829,
+      "grad_norm": 0.04549845680594444,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 64632
+    },
+    {
+      "epoch": 6.637399876771411,
+      "grad_norm": 0.04419662430882454,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 64635
+    },
+    {
+      "epoch": 6.6377079482439925,
+      "grad_norm": 0.04874144867062569,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 64638
+    },
+    {
+      "epoch": 6.6380160197165745,
+      "grad_norm": 0.06364091485738754,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 64641
+    },
+    {
+      "epoch": 6.638324091189156,
+      "grad_norm": 0.11188903450965881,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 64644
+    },
+    {
+      "epoch": 6.638632162661738,
+      "grad_norm": 0.03762582689523697,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 64647
+    },
+    {
+      "epoch": 6.638940234134319,
+      "grad_norm": 0.03571620211005211,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 64650
+    },
+    {
+      "epoch": 6.639248305606901,
+      "grad_norm": 0.05100245773792267,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 64653
+    },
+    {
+      "epoch": 6.639556377079482,
+      "grad_norm": 0.0773586705327034,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 64656
+    },
+    {
+      "epoch": 6.639864448552064,
+      "grad_norm": 0.056971583515405655,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 64659
+    },
+    {
+      "epoch": 6.640172520024645,
+      "grad_norm": 0.04947191849350929,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 64662
+    },
+    {
+      "epoch": 6.640480591497227,
+      "grad_norm": 0.039240311831235886,
+      "learning_rate": 0.01,
+      "loss": 2.0069,
+      "step": 64665
+    },
+    {
+      "epoch": 6.640788662969809,
+      "grad_norm": 0.03866692632436752,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 64668
+    },
+    {
+      "epoch": 6.641096734442391,
+      "grad_norm": 0.05008673667907715,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 64671
+    },
+    {
+      "epoch": 6.641404805914973,
+      "grad_norm": 0.12828943133354187,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 64674
+    },
+    {
+      "epoch": 6.641712877387554,
+      "grad_norm": 0.04267043247818947,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 64677
+    },
+    {
+      "epoch": 6.642020948860136,
+      "grad_norm": 0.09645269066095352,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 64680
+    },
+    {
+      "epoch": 6.642329020332717,
+      "grad_norm": 0.053970299661159515,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 64683
+    },
+    {
+      "epoch": 6.642637091805299,
+      "grad_norm": 0.15396493673324585,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 64686
+    },
+    {
+      "epoch": 6.64294516327788,
+      "grad_norm": 0.06772658973932266,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 64689
+    },
+    {
+      "epoch": 6.643253234750462,
+      "grad_norm": 0.05688326433300972,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 64692
+    },
+    {
+      "epoch": 6.643561306223043,
+      "grad_norm": 0.05814405530691147,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 64695
+    },
+    {
+      "epoch": 6.6438693776956255,
+      "grad_norm": 0.033603183925151825,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 64698
+    },
+    {
+      "epoch": 6.6441774491682075,
+      "grad_norm": 0.06284506618976593,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 64701
+    },
+    {
+      "epoch": 6.644485520640789,
+      "grad_norm": 0.03463011607527733,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 64704
+    },
+    {
+      "epoch": 6.64479359211337,
+      "grad_norm": 0.039407879114151,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 64707
+    },
+    {
+      "epoch": 6.645101663585952,
+      "grad_norm": 0.034094471484422684,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 64710
+    },
+    {
+      "epoch": 6.645409735058534,
+      "grad_norm": 0.15402834117412567,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 64713
+    },
+    {
+      "epoch": 6.645717806531115,
+      "grad_norm": 0.055694807320833206,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 64716
+    },
+    {
+      "epoch": 6.646025878003697,
+      "grad_norm": 0.07708931714296341,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 64719
+    },
+    {
+      "epoch": 6.646333949476278,
+      "grad_norm": 0.07107258588075638,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 64722
+    },
+    {
+      "epoch": 6.64664202094886,
+      "grad_norm": 0.06758569926023483,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 64725
+    },
+    {
+      "epoch": 6.6469500924214415,
+      "grad_norm": 0.07093843817710876,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 64728
+    },
+    {
+      "epoch": 6.647258163894024,
+      "grad_norm": 0.06615183502435684,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 64731
+    },
+    {
+      "epoch": 6.647566235366605,
+      "grad_norm": 0.06096789985895157,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 64734
+    },
+    {
+      "epoch": 6.647874306839187,
+      "grad_norm": 0.03493250906467438,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 64737
+    },
+    {
+      "epoch": 6.648182378311768,
+      "grad_norm": 0.058653466403484344,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 64740
+    },
+    {
+      "epoch": 6.64849044978435,
+      "grad_norm": 0.1340709775686264,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 64743
+    },
+    {
+      "epoch": 6.648798521256932,
+      "grad_norm": 0.0948764905333519,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 64746
+    },
+    {
+      "epoch": 6.649106592729513,
+      "grad_norm": 0.03758051246404648,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 64749
+    },
+    {
+      "epoch": 6.649414664202095,
+      "grad_norm": 0.05022870749235153,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 64752
+    },
+    {
+      "epoch": 6.649722735674676,
+      "grad_norm": 0.04376206919550896,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 64755
+    },
+    {
+      "epoch": 6.6500308071472585,
+      "grad_norm": 0.05255085602402687,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 64758
+    },
+    {
+      "epoch": 6.65033887861984,
+      "grad_norm": 0.04554371163249016,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 64761
+    },
+    {
+      "epoch": 6.650646950092422,
+      "grad_norm": 0.046166516840457916,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 64764
+    },
+    {
+      "epoch": 6.650955021565003,
+      "grad_norm": 0.06942164152860641,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 64767
+    },
+    {
+      "epoch": 6.651263093037585,
+      "grad_norm": 0.07894229888916016,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 64770
+    },
+    {
+      "epoch": 6.651571164510166,
+      "grad_norm": 0.10424064844846725,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 64773
+    },
+    {
+      "epoch": 6.651879235982748,
+      "grad_norm": 0.1305040717124939,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 64776
+    },
+    {
+      "epoch": 6.652187307455329,
+      "grad_norm": 0.0960015282034874,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 64779
+    },
+    {
+      "epoch": 6.652495378927911,
+      "grad_norm": 0.051948267966508865,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 64782
+    },
+    {
+      "epoch": 6.6528034504004925,
+      "grad_norm": 0.049625445157289505,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 64785
+    },
+    {
+      "epoch": 6.6531115218730745,
+      "grad_norm": 0.03437644988298416,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 64788
+    },
+    {
+      "epoch": 6.653419593345657,
+      "grad_norm": 0.04177222028374672,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 64791
+    },
+    {
+      "epoch": 6.653727664818238,
+      "grad_norm": 0.12556210160255432,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 64794
+    },
+    {
+      "epoch": 6.65403573629082,
+      "grad_norm": 0.06504391878843307,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 64797
+    },
+    {
+      "epoch": 6.654343807763401,
+      "grad_norm": 0.0797589123249054,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 64800
+    },
+    {
+      "epoch": 6.654651879235983,
+      "grad_norm": 0.05979294329881668,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 64803
+    },
+    {
+      "epoch": 6.654959950708564,
+      "grad_norm": 0.07763978838920593,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 64806
+    },
+    {
+      "epoch": 6.655268022181146,
+      "grad_norm": 0.06913294643163681,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 64809
+    },
+    {
+      "epoch": 6.655576093653727,
+      "grad_norm": 0.0825156420469284,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 64812
+    },
+    {
+      "epoch": 6.655884165126309,
+      "grad_norm": 0.04779151454567909,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 64815
+    },
+    {
+      "epoch": 6.656192236598891,
+      "grad_norm": 0.09039946645498276,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 64818
+    },
+    {
+      "epoch": 6.656500308071473,
+      "grad_norm": 0.03368096798658371,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 64821
+    },
+    {
+      "epoch": 6.656808379544055,
+      "grad_norm": 0.11677176505327225,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 64824
+    },
+    {
+      "epoch": 6.657116451016636,
+      "grad_norm": 0.09988972544670105,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 64827
+    },
+    {
+      "epoch": 6.657424522489217,
+      "grad_norm": 0.1377715915441513,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 64830
+    },
+    {
+      "epoch": 6.657732593961799,
+      "grad_norm": 0.04986416921019554,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 64833
+    },
+    {
+      "epoch": 6.658040665434381,
+      "grad_norm": 0.06681905686855316,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 64836
+    },
+    {
+      "epoch": 6.658348736906962,
+      "grad_norm": 0.05175319314002991,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 64839
+    },
+    {
+      "epoch": 6.658656808379544,
+      "grad_norm": 0.048531901091337204,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 64842
+    },
+    {
+      "epoch": 6.6589648798521255,
+      "grad_norm": 0.06149359792470932,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 64845
+    },
+    {
+      "epoch": 6.6592729513247075,
+      "grad_norm": 0.0352664515376091,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 64848
+    },
+    {
+      "epoch": 6.659581022797289,
+      "grad_norm": 0.0510052926838398,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 64851
+    },
+    {
+      "epoch": 6.659889094269871,
+      "grad_norm": 0.06455834954977036,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 64854
+    },
+    {
+      "epoch": 6.660197165742452,
+      "grad_norm": 0.13268320262432098,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 64857
+    },
+    {
+      "epoch": 6.660505237215034,
+      "grad_norm": 0.0314185805618763,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 64860
+    },
+    {
+      "epoch": 6.660813308687615,
+      "grad_norm": 0.06962891668081284,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 64863
+    },
+    {
+      "epoch": 6.661121380160197,
+      "grad_norm": 0.11769289523363113,
+      "learning_rate": 0.01,
+      "loss": 2.0032,
+      "step": 64866
+    },
+    {
+      "epoch": 6.661429451632779,
+      "grad_norm": 0.03778669983148575,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 64869
+    },
+    {
+      "epoch": 6.66173752310536,
+      "grad_norm": 0.09534257650375366,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 64872
+    },
+    {
+      "epoch": 6.662045594577942,
+      "grad_norm": 0.03727418929338455,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 64875
+    },
+    {
+      "epoch": 6.662353666050524,
+      "grad_norm": 0.04617612436413765,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 64878
+    },
+    {
+      "epoch": 6.662661737523106,
+      "grad_norm": 0.043121397495269775,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 64881
+    },
+    {
+      "epoch": 6.662969808995687,
+      "grad_norm": 0.03990964964032173,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 64884
+    },
+    {
+      "epoch": 6.663277880468269,
+      "grad_norm": 0.03709634020924568,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 64887
+    },
+    {
+      "epoch": 6.66358595194085,
+      "grad_norm": 0.042194269597530365,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 64890
+    },
+    {
+      "epoch": 6.663894023413432,
+      "grad_norm": 0.08765631169080734,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 64893
+    },
+    {
+      "epoch": 6.664202094886013,
+      "grad_norm": 0.10358686000108719,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 64896
+    },
+    {
+      "epoch": 6.664510166358595,
+      "grad_norm": 0.05577368661761284,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 64899
+    },
+    {
+      "epoch": 6.664818237831177,
+      "grad_norm": 0.07221733778715134,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 64902
+    },
+    {
+      "epoch": 6.6651263093037585,
+      "grad_norm": 0.08421463519334793,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 64905
+    },
+    {
+      "epoch": 6.66543438077634,
+      "grad_norm": 0.04022625833749771,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 64908
+    },
+    {
+      "epoch": 6.665742452248922,
+      "grad_norm": 0.10746564716100693,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 64911
+    },
+    {
+      "epoch": 6.666050523721504,
+      "grad_norm": 0.10090082883834839,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 64914
+    },
+    {
+      "epoch": 6.666358595194085,
+      "grad_norm": 0.03953251615166664,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 64917
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.09518130123615265,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 64920
+    },
+    {
+      "epoch": 6.666974738139248,
+      "grad_norm": 0.06792232394218445,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 64923
+    },
+    {
+      "epoch": 6.66728280961183,
+      "grad_norm": 0.08679928630590439,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 64926
+    },
+    {
+      "epoch": 6.667590881084411,
+      "grad_norm": 0.07284796237945557,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 64929
+    },
+    {
+      "epoch": 6.667898952556993,
+      "grad_norm": 0.07962015271186829,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 64932
+    },
+    {
+      "epoch": 6.6682070240295745,
+      "grad_norm": 0.04905517399311066,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 64935
+    },
+    {
+      "epoch": 6.668515095502157,
+      "grad_norm": 0.1008431613445282,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 64938
+    },
+    {
+      "epoch": 6.668823166974738,
+      "grad_norm": 0.07995069026947021,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 64941
+    },
+    {
+      "epoch": 6.66913123844732,
+      "grad_norm": 0.08693604916334152,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 64944
+    },
+    {
+      "epoch": 6.669439309919902,
+      "grad_norm": 0.07200144976377487,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 64947
+    },
+    {
+      "epoch": 6.669747381392483,
+      "grad_norm": 0.10411632061004639,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 64950
+    },
+    {
+      "epoch": 6.670055452865065,
+      "grad_norm": 0.03982757031917572,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 64953
+    },
+    {
+      "epoch": 6.670363524337646,
+      "grad_norm": 0.03512009605765343,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 64956
+    },
+    {
+      "epoch": 6.670671595810228,
+      "grad_norm": 0.056927431374788284,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 64959
+    },
+    {
+      "epoch": 6.6709796672828094,
+      "grad_norm": 0.04020826891064644,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 64962
+    },
+    {
+      "epoch": 6.6712877387553915,
+      "grad_norm": 0.051895178854465485,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 64965
+    },
+    {
+      "epoch": 6.671595810227973,
+      "grad_norm": 0.1494154930114746,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 64968
+    },
+    {
+      "epoch": 6.671903881700555,
+      "grad_norm": 0.06913480907678604,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 64971
+    },
+    {
+      "epoch": 6.672211953173136,
+      "grad_norm": 0.0441024973988533,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 64974
+    },
+    {
+      "epoch": 6.672520024645718,
+      "grad_norm": 0.053763311356306076,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 64977
+    },
+    {
+      "epoch": 6.672828096118299,
+      "grad_norm": 0.08884178847074509,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 64980
+    },
+    {
+      "epoch": 6.673136167590881,
+      "grad_norm": 0.06425255537033081,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 64983
+    },
+    {
+      "epoch": 6.673444239063462,
+      "grad_norm": 0.07111632078886032,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 64986
+    },
+    {
+      "epoch": 6.673752310536044,
+      "grad_norm": 0.06380520761013031,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 64989
+    },
+    {
+      "epoch": 6.674060382008626,
+      "grad_norm": 0.06785110384225845,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 64992
+    },
+    {
+      "epoch": 6.6743684534812076,
+      "grad_norm": 0.13397693634033203,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 64995
+    },
+    {
+      "epoch": 6.67467652495379,
+      "grad_norm": 0.06838545948266983,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 64998
+    },
+    {
+      "epoch": 6.674984596426371,
+      "grad_norm": 0.05458493158221245,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 65001
+    },
+    {
+      "epoch": 6.675292667898953,
+      "grad_norm": 0.04406043142080307,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 65004
+    },
+    {
+      "epoch": 6.675600739371534,
+      "grad_norm": 0.03164122626185417,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 65007
+    },
+    {
+      "epoch": 6.675908810844116,
+      "grad_norm": 0.03512512892484665,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 65010
+    },
+    {
+      "epoch": 6.676216882316697,
+      "grad_norm": 0.03886500746011734,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 65013
+    },
+    {
+      "epoch": 6.676524953789279,
+      "grad_norm": 0.04235444217920303,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 65016
+    },
+    {
+      "epoch": 6.67683302526186,
+      "grad_norm": 0.11992576718330383,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 65019
+    },
+    {
+      "epoch": 6.6771410967344424,
+      "grad_norm": 0.07870931923389435,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 65022
+    },
+    {
+      "epoch": 6.6774491682070245,
+      "grad_norm": 0.11495670676231384,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 65025
+    },
+    {
+      "epoch": 6.677757239679606,
+      "grad_norm": 0.10441329330205917,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 65028
+    },
+    {
+      "epoch": 6.678065311152187,
+      "grad_norm": 0.08615846931934357,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 65031
+    },
+    {
+      "epoch": 6.678373382624769,
+      "grad_norm": 0.053645677864551544,
+      "learning_rate": 0.01,
+      "loss": 2.0039,
+      "step": 65034
+    },
+    {
+      "epoch": 6.678681454097351,
+      "grad_norm": 0.0375412255525589,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 65037
+    },
+    {
+      "epoch": 6.678989525569932,
+      "grad_norm": 0.03644879162311554,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 65040
+    },
+    {
+      "epoch": 6.679297597042514,
+      "grad_norm": 0.09399533271789551,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 65043
+    },
+    {
+      "epoch": 6.679605668515095,
+      "grad_norm": 0.06882072985172272,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 65046
+    },
+    {
+      "epoch": 6.679913739987677,
+      "grad_norm": 0.10664650052785873,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 65049
+    },
+    {
+      "epoch": 6.6802218114602585,
+      "grad_norm": 0.044063057750463486,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 65052
+    },
+    {
+      "epoch": 6.680529882932841,
+      "grad_norm": 0.037052053958177567,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 65055
+    },
+    {
+      "epoch": 6.680837954405422,
+      "grad_norm": 0.038574136793613434,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 65058
+    },
+    {
+      "epoch": 6.681146025878004,
+      "grad_norm": 0.06184682250022888,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 65061
+    },
+    {
+      "epoch": 6.681454097350585,
+      "grad_norm": 0.09441886097192764,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 65064
+    },
+    {
+      "epoch": 6.681762168823167,
+      "grad_norm": 0.07284927368164062,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 65067
+    },
+    {
+      "epoch": 6.682070240295749,
+      "grad_norm": 0.07610774785280228,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 65070
+    },
+    {
+      "epoch": 6.68237831176833,
+      "grad_norm": 0.04308845102787018,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 65073
+    },
+    {
+      "epoch": 6.682686383240912,
+      "grad_norm": 0.08398102223873138,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 65076
+    },
+    {
+      "epoch": 6.682994454713493,
+      "grad_norm": 0.11537694185972214,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 65079
+    },
+    {
+      "epoch": 6.6833025261860755,
+      "grad_norm": 0.03176436200737953,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 65082
+    },
+    {
+      "epoch": 6.683610597658657,
+      "grad_norm": 0.09571046382188797,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 65085
+    },
+    {
+      "epoch": 6.683918669131239,
+      "grad_norm": 0.04681390896439552,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 65088
+    },
+    {
+      "epoch": 6.68422674060382,
+      "grad_norm": 0.26577845215797424,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 65091
+    },
+    {
+      "epoch": 6.684534812076402,
+      "grad_norm": 0.12212441116571426,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 65094
+    },
+    {
+      "epoch": 6.684842883548983,
+      "grad_norm": 0.4320748448371887,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 65097
+    },
+    {
+      "epoch": 6.685150955021565,
+      "grad_norm": 0.04880547523498535,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 65100
+    },
+    {
+      "epoch": 6.685459026494147,
+      "grad_norm": 0.08876339346170425,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 65103
+    },
+    {
+      "epoch": 6.685767097966728,
+      "grad_norm": 0.06511343270540237,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 65106
+    },
+    {
+      "epoch": 6.6860751694393095,
+      "grad_norm": 0.050562430173158646,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 65109
+    },
+    {
+      "epoch": 6.6863832409118915,
+      "grad_norm": 0.04717005044221878,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 65112
+    },
+    {
+      "epoch": 6.686691312384474,
+      "grad_norm": 0.08370763063430786,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 65115
+    },
+    {
+      "epoch": 6.686999383857055,
+      "grad_norm": 0.0477861724793911,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 65118
+    },
+    {
+      "epoch": 6.687307455329637,
+      "grad_norm": 0.02870395965874195,
+      "learning_rate": 0.01,
+      "loss": 2.0043,
+      "step": 65121
+    },
+    {
+      "epoch": 6.687615526802218,
+      "grad_norm": 0.04413042590022087,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 65124
+    },
+    {
+      "epoch": 6.6879235982748,
+      "grad_norm": 0.041404690593481064,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 65127
+    },
+    {
+      "epoch": 6.688231669747381,
+      "grad_norm": 0.04159929230809212,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 65130
+    },
+    {
+      "epoch": 6.688539741219963,
+      "grad_norm": 0.06357663124799728,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 65133
+    },
+    {
+      "epoch": 6.688847812692544,
+      "grad_norm": 0.03610898181796074,
+      "learning_rate": 0.01,
+      "loss": 2.0024,
+      "step": 65136
+    },
+    {
+      "epoch": 6.689155884165126,
+      "grad_norm": 0.038069941103458405,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 65139
+    },
+    {
+      "epoch": 6.689463955637708,
+      "grad_norm": 0.05543750151991844,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 65142
+    },
+    {
+      "epoch": 6.68977202711029,
+      "grad_norm": 0.10848159343004227,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 65145
+    },
+    {
+      "epoch": 6.690080098582872,
+      "grad_norm": 0.1446017026901245,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 65148
+    },
+    {
+      "epoch": 6.690388170055453,
+      "grad_norm": 0.10384636372327805,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 65151
+    },
+    {
+      "epoch": 6.690696241528035,
+      "grad_norm": 0.07247021794319153,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 65154
+    },
+    {
+      "epoch": 6.691004313000616,
+      "grad_norm": 0.04569610580801964,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 65157
+    },
+    {
+      "epoch": 6.691312384473198,
+      "grad_norm": 0.039695657789707184,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 65160
+    },
+    {
+      "epoch": 6.691620455945779,
+      "grad_norm": 0.053198885172605515,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 65163
+    },
+    {
+      "epoch": 6.691928527418361,
+      "grad_norm": 0.042604029178619385,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 65166
+    },
+    {
+      "epoch": 6.6922365988909425,
+      "grad_norm": 0.031200909987092018,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 65169
+    },
+    {
+      "epoch": 6.6925446703635245,
+      "grad_norm": 0.10486329346895218,
+      "learning_rate": 0.01,
+      "loss": 1.9919,
+      "step": 65172
+    },
+    {
+      "epoch": 6.692852741836106,
+      "grad_norm": 0.046825818717479706,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 65175
+    },
+    {
+      "epoch": 6.693160813308688,
+      "grad_norm": 0.1281237006187439,
+      "learning_rate": 0.01,
+      "loss": 2.0059,
+      "step": 65178
+    },
+    {
+      "epoch": 6.693468884781269,
+      "grad_norm": 0.09412495791912079,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 65181
+    },
+    {
+      "epoch": 6.693776956253851,
+      "grad_norm": 0.08914454281330109,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 65184
+    },
+    {
+      "epoch": 6.694085027726432,
+      "grad_norm": 0.07769876718521118,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 65187
+    },
+    {
+      "epoch": 6.694393099199014,
+      "grad_norm": 0.10803245007991791,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 65190
+    },
+    {
+      "epoch": 6.694701170671596,
+      "grad_norm": 0.07634082436561584,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 65193
+    },
+    {
+      "epoch": 6.695009242144177,
+      "grad_norm": 0.033576395362615585,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 65196
+    },
+    {
+      "epoch": 6.695317313616759,
+      "grad_norm": 0.04042442888021469,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 65199
+    },
+    {
+      "epoch": 6.695625385089341,
+      "grad_norm": 0.04007275775074959,
+      "learning_rate": 0.01,
+      "loss": 2.0092,
+      "step": 65202
+    },
+    {
+      "epoch": 6.695933456561923,
+      "grad_norm": 0.036564283072948456,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 65205
+    },
+    {
+      "epoch": 6.696241528034504,
+      "grad_norm": 0.03179538995027542,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 65208
+    },
+    {
+      "epoch": 6.696549599507086,
+      "grad_norm": 0.050880614668130875,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 65211
+    },
+    {
+      "epoch": 6.696857670979667,
+      "grad_norm": 0.07803148031234741,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 65214
+    },
+    {
+      "epoch": 6.697165742452249,
+      "grad_norm": 0.11088874191045761,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 65217
+    },
+    {
+      "epoch": 6.69747381392483,
+      "grad_norm": 0.05180731415748596,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 65220
+    },
+    {
+      "epoch": 6.697781885397412,
+      "grad_norm": 0.09849876910448074,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 65223
+    },
+    {
+      "epoch": 6.698089956869994,
+      "grad_norm": 0.07928299903869629,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 65226
+    },
+    {
+      "epoch": 6.6983980283425755,
+      "grad_norm": 0.058122437447309494,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 65229
+    },
+    {
+      "epoch": 6.698706099815157,
+      "grad_norm": 0.059113189578056335,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 65232
+    },
+    {
+      "epoch": 6.699014171287739,
+      "grad_norm": 0.03366167098283768,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 65235
+    },
+    {
+      "epoch": 6.699322242760321,
+      "grad_norm": 0.036344975233078,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 65238
+    },
+    {
+      "epoch": 6.699630314232902,
+      "grad_norm": 0.10664808005094528,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 65241
+    },
+    {
+      "epoch": 6.699938385705484,
+      "grad_norm": 0.09889326244592667,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 65244
+    },
+    {
+      "epoch": 6.700246457178065,
+      "grad_norm": 0.05460183694958687,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 65247
+    },
+    {
+      "epoch": 6.700554528650647,
+      "grad_norm": 0.08168340474367142,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 65250
+    },
+    {
+      "epoch": 6.700862600123228,
+      "grad_norm": 0.0726526603102684,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 65253
+    },
+    {
+      "epoch": 6.70117067159581,
+      "grad_norm": 0.0469973124563694,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 65256
+    },
+    {
+      "epoch": 6.7014787430683915,
+      "grad_norm": 0.05402790755033493,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 65259
+    },
+    {
+      "epoch": 6.701786814540974,
+      "grad_norm": 0.03723495826125145,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 65262
+    },
+    {
+      "epoch": 6.702094886013555,
+      "grad_norm": 0.1069960966706276,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 65265
+    },
+    {
+      "epoch": 6.702402957486137,
+      "grad_norm": 0.0806579738855362,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 65268
+    },
+    {
+      "epoch": 6.702711028958719,
+      "grad_norm": 0.05724218115210533,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 65271
+    },
+    {
+      "epoch": 6.7030191004313,
+      "grad_norm": 0.10128419101238251,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 65274
+    },
+    {
+      "epoch": 6.703327171903882,
+      "grad_norm": 0.1476512998342514,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 65277
+    },
+    {
+      "epoch": 6.703635243376463,
+      "grad_norm": 0.10202914476394653,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 65280
+    },
+    {
+      "epoch": 6.703943314849045,
+      "grad_norm": 0.042887743562459946,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 65283
+    },
+    {
+      "epoch": 6.704251386321626,
+      "grad_norm": 0.0640069991350174,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 65286
+    },
+    {
+      "epoch": 6.7045594577942085,
+      "grad_norm": 0.042582917958498,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 65289
+    },
+    {
+      "epoch": 6.70486752926679,
+      "grad_norm": 0.0463312529027462,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 65292
+    },
+    {
+      "epoch": 6.705175600739372,
+      "grad_norm": 0.05490421503782272,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 65295
+    },
+    {
+      "epoch": 6.705483672211953,
+      "grad_norm": 0.07986756414175034,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 65298
+    },
+    {
+      "epoch": 6.705791743684535,
+      "grad_norm": 0.06604810804128647,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 65301
+    },
+    {
+      "epoch": 6.706099815157117,
+      "grad_norm": 0.036626990884542465,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 65304
+    },
+    {
+      "epoch": 6.706407886629698,
+      "grad_norm": 0.03825300931930542,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 65307
+    },
+    {
+      "epoch": 6.706715958102279,
+      "grad_norm": 0.07136386632919312,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 65310
+    },
+    {
+      "epoch": 6.707024029574861,
+      "grad_norm": 0.059436917304992676,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 65313
+    },
+    {
+      "epoch": 6.707332101047443,
+      "grad_norm": 0.06798838824033737,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 65316
+    },
+    {
+      "epoch": 6.7076401725200245,
+      "grad_norm": 0.05605984479188919,
+      "learning_rate": 0.01,
+      "loss": 2.0175,
+      "step": 65319
+    },
+    {
+      "epoch": 6.707948243992607,
+      "grad_norm": 0.039418842643499374,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 65322
+    },
+    {
+      "epoch": 6.708256315465188,
+      "grad_norm": 0.037476614117622375,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 65325
+    },
+    {
+      "epoch": 6.70856438693777,
+      "grad_norm": 0.03878027945756912,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 65328
+    },
+    {
+      "epoch": 6.708872458410351,
+      "grad_norm": 0.14439933001995087,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 65331
+    },
+    {
+      "epoch": 6.709180529882933,
+      "grad_norm": 0.1026519313454628,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 65334
+    },
+    {
+      "epoch": 6.709488601355514,
+      "grad_norm": 0.059400398284196854,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 65337
+    },
+    {
+      "epoch": 6.709796672828096,
+      "grad_norm": 0.03667070344090462,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 65340
+    },
+    {
+      "epoch": 6.710104744300677,
+      "grad_norm": 0.03821050748229027,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 65343
+    },
+    {
+      "epoch": 6.710412815773259,
+      "grad_norm": 0.034569595009088516,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 65346
+    },
+    {
+      "epoch": 6.7107208872458415,
+      "grad_norm": 0.0894671380519867,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 65349
+    },
+    {
+      "epoch": 6.711028958718423,
+      "grad_norm": 0.1035616546869278,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 65352
+    },
+    {
+      "epoch": 6.711337030191005,
+      "grad_norm": 0.07528524845838547,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 65355
+    },
+    {
+      "epoch": 6.711645101663586,
+      "grad_norm": 0.073489710688591,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 65358
+    },
+    {
+      "epoch": 6.711953173136168,
+      "grad_norm": 0.047937504947185516,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 65361
+    },
+    {
+      "epoch": 6.712261244608749,
+      "grad_norm": 0.049860548228025436,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 65364
+    },
+    {
+      "epoch": 6.712569316081331,
+      "grad_norm": 0.04179536923766136,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 65367
+    },
+    {
+      "epoch": 6.712877387553912,
+      "grad_norm": 0.03599520027637482,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 65370
+    },
+    {
+      "epoch": 6.713185459026494,
+      "grad_norm": 0.03743865713477135,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 65373
+    },
+    {
+      "epoch": 6.7134935304990755,
+      "grad_norm": 0.04530680924654007,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 65376
+    },
+    {
+      "epoch": 6.7138016019716575,
+      "grad_norm": 0.05956900492310524,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 65379
+    },
+    {
+      "epoch": 6.714109673444239,
+      "grad_norm": 0.03879319131374359,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 65382
+    },
+    {
+      "epoch": 6.714417744916821,
+      "grad_norm": 0.08729801326990128,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 65385
+    },
+    {
+      "epoch": 6.714725816389402,
+      "grad_norm": 0.06801248341798782,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 65388
+    },
+    {
+      "epoch": 6.715033887861984,
+      "grad_norm": 0.10621631890535355,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 65391
+    },
+    {
+      "epoch": 6.715341959334566,
+      "grad_norm": 0.06591764837503433,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 65394
+    },
+    {
+      "epoch": 6.715650030807147,
+      "grad_norm": 0.08169190585613251,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 65397
+    },
+    {
+      "epoch": 6.715958102279729,
+      "grad_norm": 0.06591983139514923,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 65400
+    },
+    {
+      "epoch": 6.71626617375231,
+      "grad_norm": 0.102086141705513,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 65403
+    },
+    {
+      "epoch": 6.716574245224892,
+      "grad_norm": 0.13108035922050476,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 65406
+    },
+    {
+      "epoch": 6.716882316697474,
+      "grad_norm": 0.09894704073667526,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 65409
+    },
+    {
+      "epoch": 6.717190388170056,
+      "grad_norm": 0.05907077714800835,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 65412
+    },
+    {
+      "epoch": 6.717498459642637,
+      "grad_norm": 0.044414445757865906,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 65415
+    },
+    {
+      "epoch": 6.717806531115219,
+      "grad_norm": 0.04454610496759415,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 65418
+    },
+    {
+      "epoch": 6.7181146025878,
+      "grad_norm": 0.044069502502679825,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 65421
+    },
+    {
+      "epoch": 6.718422674060382,
+      "grad_norm": 0.05330649018287659,
+      "learning_rate": 0.01,
+      "loss": 2.0213,
+      "step": 65424
+    },
+    {
+      "epoch": 6.718730745532964,
+      "grad_norm": 0.13220176100730896,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 65427
+    },
+    {
+      "epoch": 6.719038817005545,
+      "grad_norm": 0.05000099912285805,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 65430
+    },
+    {
+      "epoch": 6.7193468884781264,
+      "grad_norm": 0.053126510232686996,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 65433
+    },
+    {
+      "epoch": 6.7196549599507085,
+      "grad_norm": 0.050792306661605835,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 65436
+    },
+    {
+      "epoch": 6.7199630314232905,
+      "grad_norm": 0.05356533080339432,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 65439
+    },
+    {
+      "epoch": 6.720271102895872,
+      "grad_norm": 0.03830750659108162,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 65442
+    },
+    {
+      "epoch": 6.720579174368454,
+      "grad_norm": 0.0419539138674736,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 65445
+    },
+    {
+      "epoch": 6.720887245841035,
+      "grad_norm": 0.04633840546011925,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 65448
+    },
+    {
+      "epoch": 6.721195317313617,
+      "grad_norm": 0.048049312084913254,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 65451
+    },
+    {
+      "epoch": 6.721503388786198,
+      "grad_norm": 0.031329698860645294,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 65454
+    },
+    {
+      "epoch": 6.72181146025878,
+      "grad_norm": 0.03230907768011093,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 65457
+    },
+    {
+      "epoch": 6.722119531731361,
+      "grad_norm": 0.09823235124349594,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 65460
+    },
+    {
+      "epoch": 6.722427603203943,
+      "grad_norm": 0.11478835344314575,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 65463
+    },
+    {
+      "epoch": 6.7227356746765246,
+      "grad_norm": 0.040770623832941055,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 65466
+    },
+    {
+      "epoch": 6.723043746149107,
+      "grad_norm": 0.11709921061992645,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 65469
+    },
+    {
+      "epoch": 6.723351817621689,
+      "grad_norm": 0.041359901428222656,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 65472
+    },
+    {
+      "epoch": 6.72365988909427,
+      "grad_norm": 0.03594958409667015,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 65475
+    },
+    {
+      "epoch": 6.723967960566852,
+      "grad_norm": 0.03344079852104187,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 65478
+    },
+    {
+      "epoch": 6.724276032039433,
+      "grad_norm": 0.03400930017232895,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 65481
+    },
+    {
+      "epoch": 6.724584103512015,
+      "grad_norm": 0.11725469678640366,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 65484
+    },
+    {
+      "epoch": 6.724892174984596,
+      "grad_norm": 0.08199802041053772,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 65487
+    },
+    {
+      "epoch": 6.725200246457178,
+      "grad_norm": 0.04940160736441612,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 65490
+    },
+    {
+      "epoch": 6.7255083179297594,
+      "grad_norm": 0.08786390721797943,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 65493
+    },
+    {
+      "epoch": 6.7258163894023415,
+      "grad_norm": 0.04642069712281227,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 65496
+    },
+    {
+      "epoch": 6.726124460874923,
+      "grad_norm": 0.0684671550989151,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 65499
+    },
+    {
+      "epoch": 6.726432532347505,
+      "grad_norm": 0.05767513066530228,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 65502
+    },
+    {
+      "epoch": 6.726740603820086,
+      "grad_norm": 0.04825710505247116,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 65505
+    },
+    {
+      "epoch": 6.727048675292668,
+      "grad_norm": 0.08460171520709991,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 65508
+    },
+    {
+      "epoch": 6.727356746765249,
+      "grad_norm": 0.08406694233417511,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 65511
+    },
+    {
+      "epoch": 6.727664818237831,
+      "grad_norm": 0.09590575098991394,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 65514
+    },
+    {
+      "epoch": 6.727972889710413,
+      "grad_norm": 0.05631608888506889,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 65517
+    },
+    {
+      "epoch": 6.728280961182994,
+      "grad_norm": 0.04405519738793373,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 65520
+    },
+    {
+      "epoch": 6.728589032655576,
+      "grad_norm": 0.04463804140686989,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 65523
+    },
+    {
+      "epoch": 6.728897104128158,
+      "grad_norm": 0.048253465443849564,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 65526
+    },
+    {
+      "epoch": 6.72920517560074,
+      "grad_norm": 0.058578286319971085,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 65529
+    },
+    {
+      "epoch": 6.729513247073321,
+      "grad_norm": 0.0791950449347496,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 65532
+    },
+    {
+      "epoch": 6.729821318545903,
+      "grad_norm": 0.112459696829319,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 65535
+    },
+    {
+      "epoch": 6.730129390018484,
+      "grad_norm": 0.03688877448439598,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 65538
+    },
+    {
+      "epoch": 6.730437461491066,
+      "grad_norm": 0.041237879544496536,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 65541
+    },
+    {
+      "epoch": 6.730745532963647,
+      "grad_norm": 0.09671095013618469,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 65544
+    },
+    {
+      "epoch": 6.731053604436229,
+      "grad_norm": 0.05881296843290329,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 65547
+    },
+    {
+      "epoch": 6.731361675908811,
+      "grad_norm": 0.07665944844484329,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 65550
+    },
+    {
+      "epoch": 6.7316697473813925,
+      "grad_norm": 0.12369941920042038,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 65553
+    },
+    {
+      "epoch": 6.7319778188539745,
+      "grad_norm": 0.03289101645350456,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 65556
+    },
+    {
+      "epoch": 6.732285890326556,
+      "grad_norm": 0.07767914235591888,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 65559
+    },
+    {
+      "epoch": 6.732593961799138,
+      "grad_norm": 0.16727425158023834,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 65562
+    },
+    {
+      "epoch": 6.732902033271719,
+      "grad_norm": 0.13580100238323212,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 65565
+    },
+    {
+      "epoch": 6.733210104744301,
+      "grad_norm": 0.06935961544513702,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 65568
+    },
+    {
+      "epoch": 6.733518176216882,
+      "grad_norm": 0.07026053220033646,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 65571
+    },
+    {
+      "epoch": 6.733826247689464,
+      "grad_norm": 0.08974307775497437,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 65574
+    },
+    {
+      "epoch": 6.734134319162045,
+      "grad_norm": 0.04675333574414253,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 65577
+    },
+    {
+      "epoch": 6.734442390634627,
+      "grad_norm": 0.04853864386677742,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 65580
+    },
+    {
+      "epoch": 6.7347504621072085,
+      "grad_norm": 0.03581572324037552,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 65583
+    },
+    {
+      "epoch": 6.735058533579791,
+      "grad_norm": 0.03860647231340408,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 65586
+    },
+    {
+      "epoch": 6.735366605052372,
+      "grad_norm": 0.0573217049241066,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 65589
+    },
+    {
+      "epoch": 6.735674676524954,
+      "grad_norm": 0.09368444234132767,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 65592
+    },
+    {
+      "epoch": 6.735982747997536,
+      "grad_norm": 0.21956096589565277,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 65595
+    },
+    {
+      "epoch": 6.736290819470117,
+      "grad_norm": 0.12725447118282318,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 65598
+    },
+    {
+      "epoch": 6.736598890942699,
+      "grad_norm": 0.06952683627605438,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 65601
+    },
+    {
+      "epoch": 6.73690696241528,
+      "grad_norm": 0.03885588422417641,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 65604
+    },
+    {
+      "epoch": 6.737215033887862,
+      "grad_norm": 0.06260085850954056,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 65607
+    },
+    {
+      "epoch": 6.737523105360443,
+      "grad_norm": 0.045172885060310364,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 65610
+    },
+    {
+      "epoch": 6.7378311768330255,
+      "grad_norm": 0.05928365886211395,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 65613
+    },
+    {
+      "epoch": 6.738139248305607,
+      "grad_norm": 0.039838556200265884,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 65616
+    },
+    {
+      "epoch": 6.738447319778189,
+      "grad_norm": 0.035764627158641815,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 65619
+    },
+    {
+      "epoch": 6.73875539125077,
+      "grad_norm": 0.05831705033779144,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 65622
+    },
+    {
+      "epoch": 6.739063462723352,
+      "grad_norm": 0.07349859923124313,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 65625
+    },
+    {
+      "epoch": 6.739371534195934,
+      "grad_norm": 0.05156675726175308,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 65628
+    },
+    {
+      "epoch": 6.739679605668515,
+      "grad_norm": 0.09530221670866013,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 65631
+    },
+    {
+      "epoch": 6.739987677141096,
+      "grad_norm": 0.04756055027246475,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 65634
+    },
+    {
+      "epoch": 6.740295748613678,
+      "grad_norm": 0.038086969405412674,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 65637
+    },
+    {
+      "epoch": 6.74060382008626,
+      "grad_norm": 0.06435366719961166,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 65640
+    },
+    {
+      "epoch": 6.7409118915588415,
+      "grad_norm": 0.12931565940380096,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 65643
+    },
+    {
+      "epoch": 6.741219963031424,
+      "grad_norm": 0.05568262189626694,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 65646
+    },
+    {
+      "epoch": 6.741528034504005,
+      "grad_norm": 0.06151677295565605,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 65649
+    },
+    {
+      "epoch": 6.741836105976587,
+      "grad_norm": 0.05449217930436134,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 65652
+    },
+    {
+      "epoch": 6.742144177449168,
+      "grad_norm": 0.05131798982620239,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 65655
+    },
+    {
+      "epoch": 6.74245224892175,
+      "grad_norm": 0.04320504143834114,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 65658
+    },
+    {
+      "epoch": 6.742760320394331,
+      "grad_norm": 0.06800030171871185,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 65661
+    },
+    {
+      "epoch": 6.743068391866913,
+      "grad_norm": 0.04916686192154884,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 65664
+    },
+    {
+      "epoch": 6.743376463339494,
+      "grad_norm": 0.10272326320409775,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 65667
+    },
+    {
+      "epoch": 6.743684534812076,
+      "grad_norm": 0.03326984494924545,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 65670
+    },
+    {
+      "epoch": 6.7439926062846585,
+      "grad_norm": 0.03591003641486168,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 65673
+    },
+    {
+      "epoch": 6.74430067775724,
+      "grad_norm": 0.0768413096666336,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 65676
+    },
+    {
+      "epoch": 6.744608749229822,
+      "grad_norm": 0.04805091395974159,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 65679
+    },
+    {
+      "epoch": 6.744916820702403,
+      "grad_norm": 0.060281701385974884,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 65682
+    },
+    {
+      "epoch": 6.745224892174985,
+      "grad_norm": 0.08178987354040146,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 65685
+    },
+    {
+      "epoch": 6.745532963647566,
+      "grad_norm": 0.047426290810108185,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 65688
+    },
+    {
+      "epoch": 6.745841035120148,
+      "grad_norm": 0.06165017560124397,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 65691
+    },
+    {
+      "epoch": 6.746149106592729,
+      "grad_norm": 0.0794510692358017,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 65694
+    },
+    {
+      "epoch": 6.746457178065311,
+      "grad_norm": 0.1278935670852661,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 65697
+    },
+    {
+      "epoch": 6.7467652495378925,
+      "grad_norm": 0.09933564066886902,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 65700
+    },
+    {
+      "epoch": 6.7470733210104745,
+      "grad_norm": 0.06581506133079529,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 65703
+    },
+    {
+      "epoch": 6.747381392483056,
+      "grad_norm": 0.07437628507614136,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 65706
+    },
+    {
+      "epoch": 6.747689463955638,
+      "grad_norm": 0.06629917025566101,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 65709
+    },
+    {
+      "epoch": 6.747997535428219,
+      "grad_norm": 0.04644669219851494,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 65712
+    },
+    {
+      "epoch": 6.748305606900801,
+      "grad_norm": 0.042347557842731476,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 65715
+    },
+    {
+      "epoch": 6.748613678373383,
+      "grad_norm": 0.05482964962720871,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 65718
+    },
+    {
+      "epoch": 6.748921749845964,
+      "grad_norm": 0.07373082637786865,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 65721
+    },
+    {
+      "epoch": 6.749229821318546,
+      "grad_norm": 0.08168603479862213,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 65724
+    },
+    {
+      "epoch": 6.749537892791127,
+      "grad_norm": 0.09342950582504272,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 65727
+    },
+    {
+      "epoch": 6.749845964263709,
+      "grad_norm": 0.05923538655042648,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 65730
+    },
+    {
+      "epoch": 6.750154035736291,
+      "grad_norm": 0.09214897453784943,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 65733
+    },
+    {
+      "epoch": 6.750462107208873,
+      "grad_norm": 0.034419383853673935,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 65736
+    },
+    {
+      "epoch": 6.750770178681454,
+      "grad_norm": 0.05979537591338158,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 65739
+    },
+    {
+      "epoch": 6.751078250154036,
+      "grad_norm": 0.11106430739164352,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 65742
+    },
+    {
+      "epoch": 6.751386321626617,
+      "grad_norm": 0.04116882011294365,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 65745
+    },
+    {
+      "epoch": 6.751694393099199,
+      "grad_norm": 0.0603414922952652,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 65748
+    },
+    {
+      "epoch": 6.752002464571781,
+      "grad_norm": 0.08985871076583862,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 65751
+    },
+    {
+      "epoch": 6.752310536044362,
+      "grad_norm": 0.058961328119039536,
+      "learning_rate": 0.01,
+      "loss": 2.0145,
+      "step": 65754
+    },
+    {
+      "epoch": 6.752618607516944,
+      "grad_norm": 0.0800008699297905,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 65757
+    },
+    {
+      "epoch": 6.7529266789895255,
+      "grad_norm": 0.06763100624084473,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 65760
+    },
+    {
+      "epoch": 6.7532347504621075,
+      "grad_norm": 0.08112933486700058,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 65763
+    },
+    {
+      "epoch": 6.753542821934689,
+      "grad_norm": 0.03964497894048691,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 65766
+    },
+    {
+      "epoch": 6.753850893407271,
+      "grad_norm": 0.1308111995458603,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 65769
+    },
+    {
+      "epoch": 6.754158964879852,
+      "grad_norm": 0.09832756221294403,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 65772
+    },
+    {
+      "epoch": 6.754467036352434,
+      "grad_norm": 0.08136483281850815,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 65775
+    },
+    {
+      "epoch": 6.754775107825015,
+      "grad_norm": 0.0976012796163559,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 65778
+    },
+    {
+      "epoch": 6.755083179297597,
+      "grad_norm": 0.14576563239097595,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 65781
+    },
+    {
+      "epoch": 6.755391250770178,
+      "grad_norm": 0.09006581455469131,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 65784
+    },
+    {
+      "epoch": 6.75569932224276,
+      "grad_norm": 0.07345472276210785,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 65787
+    },
+    {
+      "epoch": 6.7560073937153415,
+      "grad_norm": 0.06007661297917366,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 65790
+    },
+    {
+      "epoch": 6.756315465187924,
+      "grad_norm": 0.07360505312681198,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 65793
+    },
+    {
+      "epoch": 6.756623536660506,
+      "grad_norm": 0.07598039507865906,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 65796
+    },
+    {
+      "epoch": 6.756931608133087,
+      "grad_norm": 0.09206894040107727,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 65799
+    },
+    {
+      "epoch": 6.757239679605669,
+      "grad_norm": 0.0724322646856308,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 65802
+    },
+    {
+      "epoch": 6.75754775107825,
+      "grad_norm": 0.06573556363582611,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 65805
+    },
+    {
+      "epoch": 6.757855822550832,
+      "grad_norm": 0.04363197833299637,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 65808
+    },
+    {
+      "epoch": 6.758163894023413,
+      "grad_norm": 0.03666981682181358,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 65811
+    },
+    {
+      "epoch": 6.758471965495995,
+      "grad_norm": 0.05076993256807327,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 65814
+    },
+    {
+      "epoch": 6.758780036968576,
+      "grad_norm": 0.07672358304262161,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 65817
+    },
+    {
+      "epoch": 6.7590881084411585,
+      "grad_norm": 0.1728837937116623,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 65820
+    },
+    {
+      "epoch": 6.75939617991374,
+      "grad_norm": 0.14630280435085297,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 65823
+    },
+    {
+      "epoch": 6.759704251386322,
+      "grad_norm": 0.08959884941577911,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 65826
+    },
+    {
+      "epoch": 6.760012322858904,
+      "grad_norm": 0.05487535893917084,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 65829
+    },
+    {
+      "epoch": 6.760320394331485,
+      "grad_norm": 0.05097530409693718,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 65832
+    },
+    {
+      "epoch": 6.760628465804066,
+      "grad_norm": 0.04289190098643303,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 65835
+    },
+    {
+      "epoch": 6.760936537276648,
+      "grad_norm": 0.054402586072683334,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 65838
+    },
+    {
+      "epoch": 6.76124460874923,
+      "grad_norm": 0.046257637441158295,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 65841
+    },
+    {
+      "epoch": 6.761552680221811,
+      "grad_norm": 0.033798422664403915,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 65844
+    },
+    {
+      "epoch": 6.761860751694393,
+      "grad_norm": 0.03898288682103157,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 65847
+    },
+    {
+      "epoch": 6.7621688231669745,
+      "grad_norm": 0.061388492584228516,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 65850
+    },
+    {
+      "epoch": 6.762476894639557,
+      "grad_norm": 0.03913845494389534,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 65853
+    },
+    {
+      "epoch": 6.762784966112138,
+      "grad_norm": 0.06032055988907814,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 65856
+    },
+    {
+      "epoch": 6.76309303758472,
+      "grad_norm": 0.11745115369558334,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 65859
+    },
+    {
+      "epoch": 6.763401109057301,
+      "grad_norm": 0.10399042069911957,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 65862
+    },
+    {
+      "epoch": 6.763709180529883,
+      "grad_norm": 0.06745419651269913,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 65865
+    },
+    {
+      "epoch": 6.764017252002464,
+      "grad_norm": 0.036637041717767715,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 65868
+    },
+    {
+      "epoch": 6.764325323475046,
+      "grad_norm": 0.03688213601708412,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 65871
+    },
+    {
+      "epoch": 6.764633394947628,
+      "grad_norm": 0.041792191565036774,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 65874
+    },
+    {
+      "epoch": 6.764941466420209,
+      "grad_norm": 0.03558392450213432,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 65877
+    },
+    {
+      "epoch": 6.7652495378927915,
+      "grad_norm": 0.037666697055101395,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 65880
+    },
+    {
+      "epoch": 6.765557609365373,
+      "grad_norm": 0.03376712277531624,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 65883
+    },
+    {
+      "epoch": 6.765865680837955,
+      "grad_norm": 0.09008733928203583,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 65886
+    },
+    {
+      "epoch": 6.766173752310536,
+      "grad_norm": 0.10737526416778564,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 65889
+    },
+    {
+      "epoch": 6.766481823783118,
+      "grad_norm": 0.12039290368556976,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 65892
+    },
+    {
+      "epoch": 6.766789895255699,
+      "grad_norm": 0.10525926202535629,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 65895
+    },
+    {
+      "epoch": 6.767097966728281,
+      "grad_norm": 0.11887960135936737,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 65898
+    },
+    {
+      "epoch": 6.767406038200862,
+      "grad_norm": 0.06034785136580467,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 65901
+    },
+    {
+      "epoch": 6.767714109673444,
+      "grad_norm": 0.04543771967291832,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 65904
+    },
+    {
+      "epoch": 6.7680221811460255,
+      "grad_norm": 0.04256156459450722,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 65907
+    },
+    {
+      "epoch": 6.7683302526186075,
+      "grad_norm": 0.04168063402175903,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 65910
+    },
+    {
+      "epoch": 6.768638324091189,
+      "grad_norm": 0.05196443200111389,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 65913
+    },
+    {
+      "epoch": 6.768946395563771,
+      "grad_norm": 0.03945057466626167,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 65916
+    },
+    {
+      "epoch": 6.769254467036353,
+      "grad_norm": 0.09867379069328308,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 65919
+    },
+    {
+      "epoch": 6.769562538508934,
+      "grad_norm": 0.03700730949640274,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 65922
+    },
+    {
+      "epoch": 6.769870609981516,
+      "grad_norm": 0.05173708498477936,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 65925
+    },
+    {
+      "epoch": 6.770178681454097,
+      "grad_norm": 0.04833926633000374,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 65928
+    },
+    {
+      "epoch": 6.770486752926679,
+      "grad_norm": 0.05311394855380058,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 65931
+    },
+    {
+      "epoch": 6.77079482439926,
+      "grad_norm": 0.13476112484931946,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 65934
+    },
+    {
+      "epoch": 6.771102895871842,
+      "grad_norm": 0.05324669927358627,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 65937
+    },
+    {
+      "epoch": 6.771410967344424,
+      "grad_norm": 0.050452303141355515,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 65940
+    },
+    {
+      "epoch": 6.771719038817006,
+      "grad_norm": 0.03241961449384689,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 65943
+    },
+    {
+      "epoch": 6.772027110289587,
+      "grad_norm": 0.1082826629281044,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 65946
+    },
+    {
+      "epoch": 6.772335181762169,
+      "grad_norm": 0.08889704197645187,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 65949
+    },
+    {
+      "epoch": 6.772643253234751,
+      "grad_norm": 0.08153937757015228,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 65952
+    },
+    {
+      "epoch": 6.772951324707332,
+      "grad_norm": 0.07284180819988251,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 65955
+    },
+    {
+      "epoch": 6.773259396179914,
+      "grad_norm": 0.049150798469781876,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 65958
+    },
+    {
+      "epoch": 6.773567467652495,
+      "grad_norm": 0.10383135825395584,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 65961
+    },
+    {
+      "epoch": 6.773875539125077,
+      "grad_norm": 0.05065099522471428,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 65964
+    },
+    {
+      "epoch": 6.7741836105976585,
+      "grad_norm": 0.05559004843235016,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 65967
+    },
+    {
+      "epoch": 6.7744916820702406,
+      "grad_norm": 0.050398439168930054,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 65970
+    },
+    {
+      "epoch": 6.774799753542822,
+      "grad_norm": 0.05198364332318306,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 65973
+    },
+    {
+      "epoch": 6.775107825015404,
+      "grad_norm": 0.06122811511158943,
+      "learning_rate": 0.01,
+      "loss": 1.9957,
+      "step": 65976
+    },
+    {
+      "epoch": 6.775415896487985,
+      "grad_norm": 0.09441248327493668,
+      "learning_rate": 0.01,
+      "loss": 2.0144,
+      "step": 65979
+    },
+    {
+      "epoch": 6.775723967960567,
+      "grad_norm": 0.06502705812454224,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 65982
+    },
+    {
+      "epoch": 6.776032039433148,
+      "grad_norm": 0.07678141444921494,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 65985
+    },
+    {
+      "epoch": 6.77634011090573,
+      "grad_norm": 0.05069658160209656,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 65988
+    },
+    {
+      "epoch": 6.776648182378311,
+      "grad_norm": 0.11829594522714615,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 65991
+    },
+    {
+      "epoch": 6.776956253850893,
+      "grad_norm": 0.036905381828546524,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 65994
+    },
+    {
+      "epoch": 6.7772643253234754,
+      "grad_norm": 0.04046548530459404,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 65997
+    },
+    {
+      "epoch": 6.777572396796057,
+      "grad_norm": 0.04027654603123665,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 66000
+    },
+    {
+      "epoch": 6.777880468268639,
+      "grad_norm": 0.044706303626298904,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 66003
+    },
+    {
+      "epoch": 6.77818853974122,
+      "grad_norm": 0.0753062292933464,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 66006
+    },
+    {
+      "epoch": 6.778496611213802,
+      "grad_norm": 0.10951106995344162,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 66009
+    },
+    {
+      "epoch": 6.778804682686383,
+      "grad_norm": 0.053536590188741684,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 66012
+    },
+    {
+      "epoch": 6.779112754158965,
+      "grad_norm": 0.060393862426280975,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 66015
+    },
+    {
+      "epoch": 6.779420825631546,
+      "grad_norm": 0.049091193825006485,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 66018
+    },
+    {
+      "epoch": 6.779728897104128,
+      "grad_norm": 0.0649518296122551,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 66021
+    },
+    {
+      "epoch": 6.7800369685767095,
+      "grad_norm": 0.039447683840990067,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 66024
+    },
+    {
+      "epoch": 6.7803450400492915,
+      "grad_norm": 0.04264757037162781,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 66027
+    },
+    {
+      "epoch": 6.7806531115218736,
+      "grad_norm": 0.06174059584736824,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 66030
+    },
+    {
+      "epoch": 6.780961182994455,
+      "grad_norm": 0.05155632272362709,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 66033
+    },
+    {
+      "epoch": 6.781269254467036,
+      "grad_norm": 0.08351369947195053,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 66036
+    },
+    {
+      "epoch": 6.781577325939618,
+      "grad_norm": 0.09187694638967514,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 66039
+    },
+    {
+      "epoch": 6.7818853974122,
+      "grad_norm": 0.21938036382198334,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 66042
+    },
+    {
+      "epoch": 6.782193468884781,
+      "grad_norm": 0.051359206438064575,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 66045
+    },
+    {
+      "epoch": 6.782501540357363,
+      "grad_norm": 0.08765391260385513,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 66048
+    },
+    {
+      "epoch": 6.782809611829944,
+      "grad_norm": 0.03909270092844963,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 66051
+    },
+    {
+      "epoch": 6.783117683302526,
+      "grad_norm": 0.06215377524495125,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 66054
+    },
+    {
+      "epoch": 6.783425754775108,
+      "grad_norm": 0.05099578574299812,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 66057
+    },
+    {
+      "epoch": 6.78373382624769,
+      "grad_norm": 0.037330977618694305,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 66060
+    },
+    {
+      "epoch": 6.784041897720271,
+      "grad_norm": 0.034515704959630966,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 66063
+    },
+    {
+      "epoch": 6.784349969192853,
+      "grad_norm": 0.07824211567640305,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 66066
+    },
+    {
+      "epoch": 6.784658040665434,
+      "grad_norm": 0.13664306700229645,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 66069
+    },
+    {
+      "epoch": 6.784966112138016,
+      "grad_norm": 0.11008740961551666,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 66072
+    },
+    {
+      "epoch": 6.785274183610598,
+      "grad_norm": 0.07172096520662308,
+      "learning_rate": 0.01,
+      "loss": 2.0085,
+      "step": 66075
+    },
+    {
+      "epoch": 6.785582255083179,
+      "grad_norm": 0.08354049175977707,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 66078
+    },
+    {
+      "epoch": 6.785890326555761,
+      "grad_norm": 0.04367983713746071,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 66081
+    },
+    {
+      "epoch": 6.7861983980283425,
+      "grad_norm": 0.03163396567106247,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 66084
+    },
+    {
+      "epoch": 6.7865064695009245,
+      "grad_norm": 0.0645727813243866,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 66087
+    },
+    {
+      "epoch": 6.786814540973506,
+      "grad_norm": 0.04862186312675476,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 66090
+    },
+    {
+      "epoch": 6.787122612446088,
+      "grad_norm": 0.035945162177085876,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 66093
+    },
+    {
+      "epoch": 6.787430683918669,
+      "grad_norm": 0.10152018815279007,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 66096
+    },
+    {
+      "epoch": 6.787738755391251,
+      "grad_norm": 0.047067925333976746,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 66099
+    },
+    {
+      "epoch": 6.788046826863832,
+      "grad_norm": 0.0677446648478508,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 66102
+    },
+    {
+      "epoch": 6.788354898336414,
+      "grad_norm": 0.03530653938651085,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 66105
+    },
+    {
+      "epoch": 6.788662969808995,
+      "grad_norm": 0.04466121271252632,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 66108
+    },
+    {
+      "epoch": 6.788971041281577,
+      "grad_norm": 0.036109503358602524,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 66111
+    },
+    {
+      "epoch": 6.7892791127541585,
+      "grad_norm": 0.03432103618979454,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 66114
+    },
+    {
+      "epoch": 6.789587184226741,
+      "grad_norm": 0.06858955323696136,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 66117
+    },
+    {
+      "epoch": 6.789895255699323,
+      "grad_norm": 0.053574852645397186,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 66120
+    },
+    {
+      "epoch": 6.790203327171904,
+      "grad_norm": 0.0781317800283432,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 66123
+    },
+    {
+      "epoch": 6.790511398644486,
+      "grad_norm": 0.0939275249838829,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 66126
+    },
+    {
+      "epoch": 6.790819470117067,
+      "grad_norm": 0.1132221594452858,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 66129
+    },
+    {
+      "epoch": 6.791127541589649,
+      "grad_norm": 0.07932759076356888,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 66132
+    },
+    {
+      "epoch": 6.79143561306223,
+      "grad_norm": 0.04189478978514671,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 66135
+    },
+    {
+      "epoch": 6.791743684534812,
+      "grad_norm": 0.06015237420797348,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 66138
+    },
+    {
+      "epoch": 6.792051756007393,
+      "grad_norm": 0.05187036469578743,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 66141
+    },
+    {
+      "epoch": 6.7923598274799755,
+      "grad_norm": 0.07512509822845459,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 66144
+    },
+    {
+      "epoch": 6.792667898952557,
+      "grad_norm": 0.10434945672750473,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 66147
+    },
+    {
+      "epoch": 6.792975970425139,
+      "grad_norm": 0.07873903959989548,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 66150
+    },
+    {
+      "epoch": 6.793284041897721,
+      "grad_norm": 0.04768504202365875,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 66153
+    },
+    {
+      "epoch": 6.793592113370302,
+      "grad_norm": 0.09988034516572952,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 66156
+    },
+    {
+      "epoch": 6.793900184842883,
+      "grad_norm": 0.07220810651779175,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 66159
+    },
+    {
+      "epoch": 6.794208256315465,
+      "grad_norm": 0.036455634981393814,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 66162
+    },
+    {
+      "epoch": 6.794516327788047,
+      "grad_norm": 0.04097495973110199,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 66165
+    },
+    {
+      "epoch": 6.794824399260628,
+      "grad_norm": 0.03743165358901024,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 66168
+    },
+    {
+      "epoch": 6.79513247073321,
+      "grad_norm": 0.14310762286186218,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 66171
+    },
+    {
+      "epoch": 6.7954405422057915,
+      "grad_norm": 0.067170649766922,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 66174
+    },
+    {
+      "epoch": 6.795748613678374,
+      "grad_norm": 0.04540235176682472,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 66177
+    },
+    {
+      "epoch": 6.796056685150955,
+      "grad_norm": 0.035935305058956146,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 66180
+    },
+    {
+      "epoch": 6.796364756623537,
+      "grad_norm": 0.10116109251976013,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 66183
+    },
+    {
+      "epoch": 6.796672828096118,
+      "grad_norm": 0.09814930707216263,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 66186
+    },
+    {
+      "epoch": 6.7969808995687,
+      "grad_norm": 0.04029448702931404,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 66189
+    },
+    {
+      "epoch": 6.797288971041281,
+      "grad_norm": 0.07453613728284836,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 66192
+    },
+    {
+      "epoch": 6.797597042513863,
+      "grad_norm": 0.07255508750677109,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 66195
+    },
+    {
+      "epoch": 6.797905113986445,
+      "grad_norm": 0.07982856780290604,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 66198
+    },
+    {
+      "epoch": 6.798213185459026,
+      "grad_norm": 0.09988022595643997,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 66201
+    },
+    {
+      "epoch": 6.7985212569316085,
+      "grad_norm": 0.09353185445070267,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 66204
+    },
+    {
+      "epoch": 6.79882932840419,
+      "grad_norm": 0.04059029743075371,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 66207
+    },
+    {
+      "epoch": 6.799137399876772,
+      "grad_norm": 0.04064788296818733,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 66210
+    },
+    {
+      "epoch": 6.799445471349353,
+      "grad_norm": 0.052321575582027435,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 66213
+    },
+    {
+      "epoch": 6.799753542821935,
+      "grad_norm": 0.13621580600738525,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 66216
+    },
+    {
+      "epoch": 6.800061614294516,
+      "grad_norm": 0.04013410210609436,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 66219
+    },
+    {
+      "epoch": 6.800369685767098,
+      "grad_norm": 0.04535072669386864,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 66222
+    },
+    {
+      "epoch": 6.800677757239679,
+      "grad_norm": 0.05595749244093895,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 66225
+    },
+    {
+      "epoch": 6.800985828712261,
+      "grad_norm": 0.03648608550429344,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 66228
+    },
+    {
+      "epoch": 6.801293900184843,
+      "grad_norm": 0.039576612412929535,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 66231
+    },
+    {
+      "epoch": 6.8016019716574245,
+      "grad_norm": 0.0474928542971611,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 66234
+    },
+    {
+      "epoch": 6.801910043130006,
+      "grad_norm": 0.04102933779358864,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 66237
+    },
+    {
+      "epoch": 6.802218114602588,
+      "grad_norm": 0.03818250820040703,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 66240
+    },
+    {
+      "epoch": 6.80252618607517,
+      "grad_norm": 0.038981273770332336,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 66243
+    },
+    {
+      "epoch": 6.802834257547751,
+      "grad_norm": 0.12116497755050659,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 66246
+    },
+    {
+      "epoch": 6.803142329020333,
+      "grad_norm": 0.05318591371178627,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 66249
+    },
+    {
+      "epoch": 6.803450400492914,
+      "grad_norm": 0.05970082804560661,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 66252
+    },
+    {
+      "epoch": 6.803758471965496,
+      "grad_norm": 0.06618655472993851,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 66255
+    },
+    {
+      "epoch": 6.804066543438077,
+      "grad_norm": 0.05062811076641083,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 66258
+    },
+    {
+      "epoch": 6.804374614910659,
+      "grad_norm": 0.03643089160323143,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 66261
+    },
+    {
+      "epoch": 6.804682686383241,
+      "grad_norm": 0.051748260855674744,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 66264
+    },
+    {
+      "epoch": 6.804990757855823,
+      "grad_norm": 0.029979856684803963,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 66267
+    },
+    {
+      "epoch": 6.805298829328404,
+      "grad_norm": 0.03129233047366142,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 66270
+    },
+    {
+      "epoch": 6.805606900800986,
+      "grad_norm": 0.06324151903390884,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 66273
+    },
+    {
+      "epoch": 6.805914972273568,
+      "grad_norm": 0.13628403842449188,
+      "learning_rate": 0.01,
+      "loss": 2.0176,
+      "step": 66276
+    },
+    {
+      "epoch": 6.806223043746149,
+      "grad_norm": 0.05685223639011383,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 66279
+    },
+    {
+      "epoch": 6.806531115218731,
+      "grad_norm": 0.0537252239882946,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 66282
+    },
+    {
+      "epoch": 6.806839186691312,
+      "grad_norm": 0.11015389859676361,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 66285
+    },
+    {
+      "epoch": 6.807147258163894,
+      "grad_norm": 0.04667497053742409,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 66288
+    },
+    {
+      "epoch": 6.8074553296364755,
+      "grad_norm": 0.03570316731929779,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 66291
+    },
+    {
+      "epoch": 6.8077634011090575,
+      "grad_norm": 0.0808587595820427,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 66294
+    },
+    {
+      "epoch": 6.808071472581639,
+      "grad_norm": 0.10974206030368805,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 66297
+    },
+    {
+      "epoch": 6.808379544054221,
+      "grad_norm": 0.10623728483915329,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 66300
+    },
+    {
+      "epoch": 6.808687615526802,
+      "grad_norm": 0.07686350494623184,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 66303
+    },
+    {
+      "epoch": 6.808995686999384,
+      "grad_norm": 0.0518915168941021,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 66306
+    },
+    {
+      "epoch": 6.809303758471965,
+      "grad_norm": 0.05631876736879349,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 66309
+    },
+    {
+      "epoch": 6.809611829944547,
+      "grad_norm": 0.07675952464342117,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 66312
+    },
+    {
+      "epoch": 6.809919901417128,
+      "grad_norm": 0.05211762338876724,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 66315
+    },
+    {
+      "epoch": 6.81022797288971,
+      "grad_norm": 0.07859452068805695,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 66318
+    },
+    {
+      "epoch": 6.810536044362292,
+      "grad_norm": 0.05693339928984642,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 66321
+    },
+    {
+      "epoch": 6.810844115834874,
+      "grad_norm": 0.03533701226115227,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 66324
+    },
+    {
+      "epoch": 6.811152187307456,
+      "grad_norm": 0.10251706093549728,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 66327
+    },
+    {
+      "epoch": 6.811460258780037,
+      "grad_norm": 0.06078473851084709,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 66330
+    },
+    {
+      "epoch": 6.811768330252619,
+      "grad_norm": 0.0667276531457901,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 66333
+    },
+    {
+      "epoch": 6.8120764017252,
+      "grad_norm": 0.08299112319946289,
+      "learning_rate": 0.01,
+      "loss": 2.012,
+      "step": 66336
+    },
+    {
+      "epoch": 6.812384473197782,
+      "grad_norm": 0.10200154036283493,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 66339
+    },
+    {
+      "epoch": 6.812692544670363,
+      "grad_norm": 0.06807447969913483,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 66342
+    },
+    {
+      "epoch": 6.813000616142945,
+      "grad_norm": 0.12666809558868408,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 66345
+    },
+    {
+      "epoch": 6.813308687615526,
+      "grad_norm": 0.07474800199270248,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 66348
+    },
+    {
+      "epoch": 6.8136167590881085,
+      "grad_norm": 0.045581359416246414,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 66351
+    },
+    {
+      "epoch": 6.8139248305606905,
+      "grad_norm": 0.03591908887028694,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 66354
+    },
+    {
+      "epoch": 6.814232902033272,
+      "grad_norm": 0.0317840613424778,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 66357
+    },
+    {
+      "epoch": 6.814540973505853,
+      "grad_norm": 0.12169979512691498,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 66360
+    },
+    {
+      "epoch": 6.814849044978435,
+      "grad_norm": 0.159201517701149,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 66363
+    },
+    {
+      "epoch": 6.815157116451017,
+      "grad_norm": 0.0912163183093071,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 66366
+    },
+    {
+      "epoch": 6.815465187923598,
+      "grad_norm": 0.05623574182391167,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 66369
+    },
+    {
+      "epoch": 6.81577325939618,
+      "grad_norm": 0.0320771262049675,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 66372
+    },
+    {
+      "epoch": 6.816081330868761,
+      "grad_norm": 0.033067572861909866,
+      "learning_rate": 0.01,
+      "loss": 2.0206,
+      "step": 66375
+    },
+    {
+      "epoch": 6.816389402341343,
+      "grad_norm": 0.043655119836330414,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 66378
+    },
+    {
+      "epoch": 6.8166974738139245,
+      "grad_norm": 0.03852735459804535,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 66381
+    },
+    {
+      "epoch": 6.817005545286507,
+      "grad_norm": 0.06447228044271469,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 66384
+    },
+    {
+      "epoch": 6.817313616759088,
+      "grad_norm": 0.1129051074385643,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 66387
+    },
+    {
+      "epoch": 6.81762168823167,
+      "grad_norm": 0.09710655361413956,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 66390
+    },
+    {
+      "epoch": 6.817929759704251,
+      "grad_norm": 0.13679730892181396,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 66393
+    },
+    {
+      "epoch": 6.818237831176833,
+      "grad_norm": 0.07610119134187698,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 66396
+    },
+    {
+      "epoch": 6.818545902649415,
+      "grad_norm": 0.06364092975854874,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 66399
+    },
+    {
+      "epoch": 6.818853974121996,
+      "grad_norm": 0.06607314199209213,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 66402
+    },
+    {
+      "epoch": 6.819162045594578,
+      "grad_norm": 0.06625553220510483,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 66405
+    },
+    {
+      "epoch": 6.819470117067159,
+      "grad_norm": 0.052945904433727264,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 66408
+    },
+    {
+      "epoch": 6.8197781885397415,
+      "grad_norm": 0.0977964773774147,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 66411
+    },
+    {
+      "epoch": 6.820086260012323,
+      "grad_norm": 0.09723694622516632,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 66414
+    },
+    {
+      "epoch": 6.820394331484905,
+      "grad_norm": 0.05858434736728668,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 66417
+    },
+    {
+      "epoch": 6.820702402957486,
+      "grad_norm": 0.09421772509813309,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 66420
+    },
+    {
+      "epoch": 6.821010474430068,
+      "grad_norm": 0.040733322501182556,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 66423
+    },
+    {
+      "epoch": 6.821318545902649,
+      "grad_norm": 0.0649617463350296,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 66426
+    },
+    {
+      "epoch": 6.821626617375231,
+      "grad_norm": 0.10224436968564987,
+      "learning_rate": 0.01,
+      "loss": 2.0167,
+      "step": 66429
+    },
+    {
+      "epoch": 6.821934688847813,
+      "grad_norm": 0.07691013067960739,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 66432
+    },
+    {
+      "epoch": 6.822242760320394,
+      "grad_norm": 0.1012914851307869,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 66435
+    },
+    {
+      "epoch": 6.8225508317929755,
+      "grad_norm": 0.04432602971792221,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 66438
+    },
+    {
+      "epoch": 6.8228589032655576,
+      "grad_norm": 0.0863911360502243,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 66441
+    },
+    {
+      "epoch": 6.82316697473814,
+      "grad_norm": 0.0747685432434082,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 66444
+    },
+    {
+      "epoch": 6.823475046210721,
+      "grad_norm": 0.04725578799843788,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 66447
+    },
+    {
+      "epoch": 6.823783117683303,
+      "grad_norm": 0.1259828805923462,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 66450
+    },
+    {
+      "epoch": 6.824091189155884,
+      "grad_norm": 0.05945250764489174,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 66453
+    },
+    {
+      "epoch": 6.824399260628466,
+      "grad_norm": 0.03467526659369469,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 66456
+    },
+    {
+      "epoch": 6.824707332101047,
+      "grad_norm": 0.04494983330368996,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 66459
+    },
+    {
+      "epoch": 6.825015403573629,
+      "grad_norm": 0.042065590620040894,
+      "learning_rate": 0.01,
+      "loss": 2.0022,
+      "step": 66462
+    },
+    {
+      "epoch": 6.82532347504621,
+      "grad_norm": 0.1278277486562729,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 66465
+    },
+    {
+      "epoch": 6.8256315465187924,
+      "grad_norm": 0.04210862144827843,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 66468
+    },
+    {
+      "epoch": 6.825939617991374,
+      "grad_norm": 0.10816308856010437,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 66471
+    },
+    {
+      "epoch": 6.826247689463956,
+      "grad_norm": 0.04535592719912529,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 66474
+    },
+    {
+      "epoch": 6.826555760936538,
+      "grad_norm": 0.06771806627511978,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 66477
+    },
+    {
+      "epoch": 6.826863832409119,
+      "grad_norm": 0.13082286715507507,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 66480
+    },
+    {
+      "epoch": 6.827171903881701,
+      "grad_norm": 0.06389307230710983,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 66483
+    },
+    {
+      "epoch": 6.827479975354282,
+      "grad_norm": 0.04055839404463768,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 66486
+    },
+    {
+      "epoch": 6.827788046826864,
+      "grad_norm": 0.03493233397603035,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 66489
+    },
+    {
+      "epoch": 6.828096118299445,
+      "grad_norm": 0.03460715711116791,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 66492
+    },
+    {
+      "epoch": 6.828404189772027,
+      "grad_norm": 0.04448126628994942,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 66495
+    },
+    {
+      "epoch": 6.8287122612446085,
+      "grad_norm": 0.14231470227241516,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 66498
+    },
+    {
+      "epoch": 6.8290203327171906,
+      "grad_norm": 0.13960252702236176,
+      "learning_rate": 0.01,
+      "loss": 2.0018,
+      "step": 66501
+    },
+    {
+      "epoch": 6.829328404189772,
+      "grad_norm": 0.053078074008226395,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 66504
+    },
+    {
+      "epoch": 6.829636475662354,
+      "grad_norm": 0.048882726579904556,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 66507
+    },
+    {
+      "epoch": 6.829944547134935,
+      "grad_norm": 0.052697841078042984,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 66510
+    },
+    {
+      "epoch": 6.830252618607517,
+      "grad_norm": 0.07299689203500748,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 66513
+    },
+    {
+      "epoch": 6.830560690080098,
+      "grad_norm": 0.06163341924548149,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 66516
+    },
+    {
+      "epoch": 6.83086876155268,
+      "grad_norm": 0.11083027720451355,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 66519
+    },
+    {
+      "epoch": 6.831176833025262,
+      "grad_norm": 0.07942959666252136,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 66522
+    },
+    {
+      "epoch": 6.831484904497843,
+      "grad_norm": 0.048363711684942245,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 66525
+    },
+    {
+      "epoch": 6.8317929759704255,
+      "grad_norm": 0.03769001364707947,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 66528
+    },
+    {
+      "epoch": 6.832101047443007,
+      "grad_norm": 0.03747798129916191,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 66531
+    },
+    {
+      "epoch": 6.832409118915589,
+      "grad_norm": 0.05700261518359184,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 66534
+    },
+    {
+      "epoch": 6.83271719038817,
+      "grad_norm": 0.08550623804330826,
+      "learning_rate": 0.01,
+      "loss": 2.0129,
+      "step": 66537
+    },
+    {
+      "epoch": 6.833025261860752,
+      "grad_norm": 0.05213450267910957,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 66540
+    },
+    {
+      "epoch": 6.833333333333333,
+      "grad_norm": 0.12226033955812454,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 66543
+    },
+    {
+      "epoch": 6.833641404805915,
+      "grad_norm": 0.06343870609998703,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 66546
+    },
+    {
+      "epoch": 6.833949476278496,
+      "grad_norm": 0.06879778206348419,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 66549
+    },
+    {
+      "epoch": 6.834257547751078,
+      "grad_norm": 0.04970945045351982,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 66552
+    },
+    {
+      "epoch": 6.83456561922366,
+      "grad_norm": 0.04559726640582085,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 66555
+    },
+    {
+      "epoch": 6.8348736906962415,
+      "grad_norm": 0.08180249482393265,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 66558
+    },
+    {
+      "epoch": 6.835181762168823,
+      "grad_norm": 0.044129811227321625,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 66561
+    },
+    {
+      "epoch": 6.835489833641405,
+      "grad_norm": 0.038416411727666855,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 66564
+    },
+    {
+      "epoch": 6.835797905113987,
+      "grad_norm": 0.06428057700395584,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 66567
+    },
+    {
+      "epoch": 6.836105976586568,
+      "grad_norm": 0.07858021557331085,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 66570
+    },
+    {
+      "epoch": 6.83641404805915,
+      "grad_norm": 0.07772648334503174,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 66573
+    },
+    {
+      "epoch": 6.836722119531731,
+      "grad_norm": 0.03566915541887283,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 66576
+    },
+    {
+      "epoch": 6.837030191004313,
+      "grad_norm": 0.04193243384361267,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 66579
+    },
+    {
+      "epoch": 6.837338262476894,
+      "grad_norm": 0.058167651295661926,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 66582
+    },
+    {
+      "epoch": 6.837646333949476,
+      "grad_norm": 0.12235833704471588,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 66585
+    },
+    {
+      "epoch": 6.837954405422058,
+      "grad_norm": 0.052113085985183716,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 66588
+    },
+    {
+      "epoch": 6.83826247689464,
+      "grad_norm": 0.0758160799741745,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 66591
+    },
+    {
+      "epoch": 6.838570548367221,
+      "grad_norm": 0.05963435024023056,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 66594
+    },
+    {
+      "epoch": 6.838878619839803,
+      "grad_norm": 0.040900785475969315,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 66597
+    },
+    {
+      "epoch": 6.839186691312385,
+      "grad_norm": 0.042031388729810715,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 66600
+    },
+    {
+      "epoch": 6.839494762784966,
+      "grad_norm": 0.056323789060115814,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 66603
+    },
+    {
+      "epoch": 6.839802834257548,
+      "grad_norm": 0.07060620188713074,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 66606
+    },
+    {
+      "epoch": 6.840110905730129,
+      "grad_norm": 0.05406404659152031,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 66609
+    },
+    {
+      "epoch": 6.840418977202711,
+      "grad_norm": 0.06324061751365662,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 66612
+    },
+    {
+      "epoch": 6.8407270486752925,
+      "grad_norm": 0.04893625155091286,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 66615
+    },
+    {
+      "epoch": 6.8410351201478745,
+      "grad_norm": 0.062072839587926865,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 66618
+    },
+    {
+      "epoch": 6.841343191620456,
+      "grad_norm": 0.12954573333263397,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 66621
+    },
+    {
+      "epoch": 6.841651263093038,
+      "grad_norm": 0.06360910087823868,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 66624
+    },
+    {
+      "epoch": 6.841959334565619,
+      "grad_norm": 0.0654262900352478,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 66627
+    },
+    {
+      "epoch": 6.842267406038201,
+      "grad_norm": 0.12571407854557037,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 66630
+    },
+    {
+      "epoch": 6.842575477510783,
+      "grad_norm": 0.09581385552883148,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 66633
+    },
+    {
+      "epoch": 6.842883548983364,
+      "grad_norm": 0.08289510756731033,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 66636
+    },
+    {
+      "epoch": 6.843191620455945,
+      "grad_norm": 0.043533485382795334,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 66639
+    },
+    {
+      "epoch": 6.843499691928527,
+      "grad_norm": 0.04672224447131157,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 66642
+    },
+    {
+      "epoch": 6.843807763401109,
+      "grad_norm": 0.03264494985342026,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 66645
+    },
+    {
+      "epoch": 6.844115834873691,
+      "grad_norm": 0.10711422562599182,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 66648
+    },
+    {
+      "epoch": 6.844423906346273,
+      "grad_norm": 0.08725102990865707,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 66651
+    },
+    {
+      "epoch": 6.844731977818854,
+      "grad_norm": 0.03802330791950226,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 66654
+    },
+    {
+      "epoch": 6.845040049291436,
+      "grad_norm": 0.0540679395198822,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 66657
+    },
+    {
+      "epoch": 6.845348120764017,
+      "grad_norm": 0.10413230210542679,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 66660
+    },
+    {
+      "epoch": 6.845656192236599,
+      "grad_norm": 0.07961270213127136,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 66663
+    },
+    {
+      "epoch": 6.84596426370918,
+      "grad_norm": 0.06600063294172287,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 66666
+    },
+    {
+      "epoch": 6.846272335181762,
+      "grad_norm": 0.10893017053604126,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 66669
+    },
+    {
+      "epoch": 6.846580406654343,
+      "grad_norm": 0.07856502383947372,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 66672
+    },
+    {
+      "epoch": 6.8468884781269255,
+      "grad_norm": 0.06285273283720016,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 66675
+    },
+    {
+      "epoch": 6.8471965495995075,
+      "grad_norm": 0.07361593842506409,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 66678
+    },
+    {
+      "epoch": 6.847504621072089,
+      "grad_norm": 0.07996716350317001,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 66681
+    },
+    {
+      "epoch": 6.847812692544671,
+      "grad_norm": 0.05737863853573799,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 66684
+    },
+    {
+      "epoch": 6.848120764017252,
+      "grad_norm": 0.05021853744983673,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 66687
+    },
+    {
+      "epoch": 6.848428835489834,
+      "grad_norm": 0.033845316618680954,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 66690
+    },
+    {
+      "epoch": 6.848736906962415,
+      "grad_norm": 0.10472722351551056,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 66693
+    },
+    {
+      "epoch": 6.849044978434997,
+      "grad_norm": 0.07523515820503235,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 66696
+    },
+    {
+      "epoch": 6.849353049907578,
+      "grad_norm": 0.08648163080215454,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 66699
+    },
+    {
+      "epoch": 6.84966112138016,
+      "grad_norm": 0.05042044818401337,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 66702
+    },
+    {
+      "epoch": 6.8499691928527415,
+      "grad_norm": 0.04480702057480812,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 66705
+    },
+    {
+      "epoch": 6.850277264325324,
+      "grad_norm": 0.07494408637285233,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 66708
+    },
+    {
+      "epoch": 6.850585335797905,
+      "grad_norm": 0.061382777988910675,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 66711
+    },
+    {
+      "epoch": 6.850893407270487,
+      "grad_norm": 0.04542558267712593,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 66714
+    },
+    {
+      "epoch": 6.851201478743068,
+      "grad_norm": 0.039626192301511765,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 66717
+    },
+    {
+      "epoch": 6.85150955021565,
+      "grad_norm": 0.04274001717567444,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 66720
+    },
+    {
+      "epoch": 6.851817621688232,
+      "grad_norm": 0.16487199068069458,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 66723
+    },
+    {
+      "epoch": 6.852125693160813,
+      "grad_norm": 0.04666552320122719,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 66726
+    },
+    {
+      "epoch": 6.852433764633395,
+      "grad_norm": 0.04287717863917351,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 66729
+    },
+    {
+      "epoch": 6.852741836105976,
+      "grad_norm": 0.04304524138569832,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 66732
+    },
+    {
+      "epoch": 6.8530499075785585,
+      "grad_norm": 0.06935624033212662,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 66735
+    },
+    {
+      "epoch": 6.85335797905114,
+      "grad_norm": 0.06261073797941208,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 66738
+    },
+    {
+      "epoch": 6.853666050523722,
+      "grad_norm": 0.07173455506563187,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 66741
+    },
+    {
+      "epoch": 6.853974121996303,
+      "grad_norm": 0.08861742913722992,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 66744
+    },
+    {
+      "epoch": 6.854282193468885,
+      "grad_norm": 0.05243033543229103,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 66747
+    },
+    {
+      "epoch": 6.854590264941466,
+      "grad_norm": 0.03281958028674126,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 66750
+    },
+    {
+      "epoch": 6.854898336414048,
+      "grad_norm": 0.10798471421003342,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 66753
+    },
+    {
+      "epoch": 6.85520640788663,
+      "grad_norm": 0.10174170881509781,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 66756
+    },
+    {
+      "epoch": 6.855514479359211,
+      "grad_norm": 0.03573278710246086,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 66759
+    },
+    {
+      "epoch": 6.8558225508317925,
+      "grad_norm": 0.12209411710500717,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 66762
+    },
+    {
+      "epoch": 6.8561306223043745,
+      "grad_norm": 0.06323892623186111,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 66765
+    },
+    {
+      "epoch": 6.856438693776957,
+      "grad_norm": 0.03758067637681961,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 66768
+    },
+    {
+      "epoch": 6.856746765249538,
+      "grad_norm": 0.0766826942563057,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 66771
+    },
+    {
+      "epoch": 6.85705483672212,
+      "grad_norm": 0.08893147110939026,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 66774
+    },
+    {
+      "epoch": 6.857362908194701,
+      "grad_norm": 0.06650066375732422,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 66777
+    },
+    {
+      "epoch": 6.857670979667283,
+      "grad_norm": 0.04757289960980415,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 66780
+    },
+    {
+      "epoch": 6.857979051139864,
+      "grad_norm": 0.0641031265258789,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 66783
+    },
+    {
+      "epoch": 6.858287122612446,
+      "grad_norm": 0.04178297147154808,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 66786
+    },
+    {
+      "epoch": 6.858595194085027,
+      "grad_norm": 0.042551252990961075,
+      "learning_rate": 0.01,
+      "loss": 2.0181,
+      "step": 66789
+    },
+    {
+      "epoch": 6.858903265557609,
+      "grad_norm": 0.05980202183127403,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 66792
+    },
+    {
+      "epoch": 6.859211337030191,
+      "grad_norm": 0.11070612818002701,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 66795
+    },
+    {
+      "epoch": 6.859519408502773,
+      "grad_norm": 0.03422870859503746,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 66798
+    },
+    {
+      "epoch": 6.859827479975355,
+      "grad_norm": 0.04661295935511589,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 66801
+    },
+    {
+      "epoch": 6.860135551447936,
+      "grad_norm": 0.04450797662138939,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 66804
+    },
+    {
+      "epoch": 6.860443622920518,
+      "grad_norm": 0.03987280651926994,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 66807
+    },
+    {
+      "epoch": 6.860751694393099,
+      "grad_norm": 0.0804753229022026,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 66810
+    },
+    {
+      "epoch": 6.861059765865681,
+      "grad_norm": 0.09321844577789307,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 66813
+    },
+    {
+      "epoch": 6.861367837338262,
+      "grad_norm": 0.128792867064476,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 66816
+    },
+    {
+      "epoch": 6.861675908810844,
+      "grad_norm": 0.0815480500459671,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 66819
+    },
+    {
+      "epoch": 6.8619839802834255,
+      "grad_norm": 0.047995518893003464,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 66822
+    },
+    {
+      "epoch": 6.8622920517560075,
+      "grad_norm": 0.03637409210205078,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 66825
+    },
+    {
+      "epoch": 6.862600123228589,
+      "grad_norm": 0.03809746354818344,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 66828
+    },
+    {
+      "epoch": 6.862908194701171,
+      "grad_norm": 0.03449690714478493,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 66831
+    },
+    {
+      "epoch": 6.863216266173753,
+      "grad_norm": 0.09826438128948212,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 66834
+    },
+    {
+      "epoch": 6.863524337646334,
+      "grad_norm": 0.09669643640518188,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 66837
+    },
+    {
+      "epoch": 6.863832409118915,
+      "grad_norm": 0.03975251317024231,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 66840
+    },
+    {
+      "epoch": 6.864140480591497,
+      "grad_norm": 0.045158423483371735,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 66843
+    },
+    {
+      "epoch": 6.864448552064079,
+      "grad_norm": 0.08310145139694214,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 66846
+    },
+    {
+      "epoch": 6.86475662353666,
+      "grad_norm": 0.068028524518013,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 66849
+    },
+    {
+      "epoch": 6.865064695009242,
+      "grad_norm": 0.02985006757080555,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 66852
+    },
+    {
+      "epoch": 6.865372766481824,
+      "grad_norm": 0.03493596985936165,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 66855
+    },
+    {
+      "epoch": 6.865680837954406,
+      "grad_norm": 0.04760652035474777,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 66858
+    },
+    {
+      "epoch": 6.865988909426987,
+      "grad_norm": 0.08393906056880951,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 66861
+    },
+    {
+      "epoch": 6.866296980899569,
+      "grad_norm": 0.05745869129896164,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 66864
+    },
+    {
+      "epoch": 6.86660505237215,
+      "grad_norm": 0.04294672608375549,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 66867
+    },
+    {
+      "epoch": 6.866913123844732,
+      "grad_norm": 0.052304986864328384,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 66870
+    },
+    {
+      "epoch": 6.867221195317313,
+      "grad_norm": 0.04665624350309372,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 66873
+    },
+    {
+      "epoch": 6.867529266789895,
+      "grad_norm": 0.037743669003248215,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 66876
+    },
+    {
+      "epoch": 6.867837338262477,
+      "grad_norm": 0.19228719174861908,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 66879
+    },
+    {
+      "epoch": 6.8681454097350585,
+      "grad_norm": 0.044259946793317795,
+      "learning_rate": 0.01,
+      "loss": 2.0154,
+      "step": 66882
+    },
+    {
+      "epoch": 6.8684534812076405,
+      "grad_norm": 0.03942949324846268,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 66885
+    },
+    {
+      "epoch": 6.868761552680222,
+      "grad_norm": 0.03812559321522713,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 66888
+    },
+    {
+      "epoch": 6.869069624152804,
+      "grad_norm": 0.03519313409924507,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 66891
+    },
+    {
+      "epoch": 6.869377695625385,
+      "grad_norm": 0.04644708335399628,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 66894
+    },
+    {
+      "epoch": 6.869685767097967,
+      "grad_norm": 0.059009701013565063,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 66897
+    },
+    {
+      "epoch": 6.869993838570548,
+      "grad_norm": 0.08703511953353882,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 66900
+    },
+    {
+      "epoch": 6.87030191004313,
+      "grad_norm": 0.06761835515499115,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 66903
+    },
+    {
+      "epoch": 6.870609981515711,
+      "grad_norm": 0.12400759011507034,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 66906
+    },
+    {
+      "epoch": 6.870918052988293,
+      "grad_norm": 0.13963590562343597,
+      "learning_rate": 0.01,
+      "loss": 1.9987,
+      "step": 66909
+    },
+    {
+      "epoch": 6.8712261244608746,
+      "grad_norm": 0.1004619374871254,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 66912
+    },
+    {
+      "epoch": 6.871534195933457,
+      "grad_norm": 0.05865470692515373,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 66915
+    },
+    {
+      "epoch": 6.871842267406038,
+      "grad_norm": 0.06006789952516556,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 66918
+    },
+    {
+      "epoch": 6.87215033887862,
+      "grad_norm": 0.03859030827879906,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 66921
+    },
+    {
+      "epoch": 6.872458410351202,
+      "grad_norm": 0.03140386939048767,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 66924
+    },
+    {
+      "epoch": 6.872766481823783,
+      "grad_norm": 0.030816826969385147,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 66927
+    },
+    {
+      "epoch": 6.873074553296365,
+      "grad_norm": 0.031165439635515213,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 66930
+    },
+    {
+      "epoch": 6.873382624768946,
+      "grad_norm": 0.1848057508468628,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 66933
+    },
+    {
+      "epoch": 6.873690696241528,
+      "grad_norm": 0.11120212823152542,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 66936
+    },
+    {
+      "epoch": 6.8739987677141094,
+      "grad_norm": 0.051300667226314545,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 66939
+    },
+    {
+      "epoch": 6.8743068391866915,
+      "grad_norm": 0.03566049784421921,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 66942
+    },
+    {
+      "epoch": 6.874614910659273,
+      "grad_norm": 0.0652051493525505,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 66945
+    },
+    {
+      "epoch": 6.874922982131855,
+      "grad_norm": 0.06834497302770615,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 66948
+    },
+    {
+      "epoch": 6.875231053604436,
+      "grad_norm": 0.058571867644786835,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 66951
+    },
+    {
+      "epoch": 6.875539125077018,
+      "grad_norm": 0.05274929851293564,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 66954
+    },
+    {
+      "epoch": 6.8758471965496,
+      "grad_norm": 0.044645778834819794,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 66957
+    },
+    {
+      "epoch": 6.876155268022181,
+      "grad_norm": 0.05121855065226555,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 66960
+    },
+    {
+      "epoch": 6.876463339494762,
+      "grad_norm": 0.0632731169462204,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 66963
+    },
+    {
+      "epoch": 6.876771410967344,
+      "grad_norm": 0.0810575783252716,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 66966
+    },
+    {
+      "epoch": 6.877079482439926,
+      "grad_norm": 0.06958875060081482,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 66969
+    },
+    {
+      "epoch": 6.877387553912508,
+      "grad_norm": 0.06824204325675964,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 66972
+    },
+    {
+      "epoch": 6.87769562538509,
+      "grad_norm": 0.05225484445691109,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 66975
+    },
+    {
+      "epoch": 6.878003696857671,
+      "grad_norm": 0.05065394937992096,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 66978
+    },
+    {
+      "epoch": 6.878311768330253,
+      "grad_norm": 0.13061989843845367,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 66981
+    },
+    {
+      "epoch": 6.878619839802834,
+      "grad_norm": 0.12182870507240295,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 66984
+    },
+    {
+      "epoch": 6.878927911275416,
+      "grad_norm": 0.048450104892253876,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 66987
+    },
+    {
+      "epoch": 6.879235982747997,
+      "grad_norm": 0.055096838623285294,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 66990
+    },
+    {
+      "epoch": 6.879544054220579,
+      "grad_norm": 0.05567490682005882,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 66993
+    },
+    {
+      "epoch": 6.87985212569316,
+      "grad_norm": 0.05405956879258156,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 66996
+    },
+    {
+      "epoch": 6.8801601971657425,
+      "grad_norm": 0.08444008976221085,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 66999
+    },
+    {
+      "epoch": 6.8804682686383245,
+      "grad_norm": 0.08067743480205536,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 67002
+    },
+    {
+      "epoch": 6.880776340110906,
+      "grad_norm": 0.042782243341207504,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 67005
+    },
+    {
+      "epoch": 6.881084411583488,
+      "grad_norm": 0.036419086158275604,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 67008
+    },
+    {
+      "epoch": 6.881392483056069,
+      "grad_norm": 0.05873832479119301,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 67011
+    },
+    {
+      "epoch": 6.881700554528651,
+      "grad_norm": 0.06187519058585167,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 67014
+    },
+    {
+      "epoch": 6.882008626001232,
+      "grad_norm": 0.049339089542627335,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 67017
+    },
+    {
+      "epoch": 6.882316697473814,
+      "grad_norm": 0.037161026149988174,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 67020
+    },
+    {
+      "epoch": 6.882624768946395,
+      "grad_norm": 0.03601270541548729,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 67023
+    },
+    {
+      "epoch": 6.882932840418977,
+      "grad_norm": 0.07066098600625992,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 67026
+    },
+    {
+      "epoch": 6.8832409118915585,
+      "grad_norm": 0.1363425999879837,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 67029
+    },
+    {
+      "epoch": 6.883548983364141,
+      "grad_norm": 0.060069505125284195,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 67032
+    },
+    {
+      "epoch": 6.883857054836723,
+      "grad_norm": 0.08887060731649399,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 67035
+    },
+    {
+      "epoch": 6.884165126309304,
+      "grad_norm": 0.03954588621854782,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 67038
+    },
+    {
+      "epoch": 6.884473197781885,
+      "grad_norm": 0.04933086037635803,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 67041
+    },
+    {
+      "epoch": 6.884781269254467,
+      "grad_norm": 0.05626452714204788,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 67044
+    },
+    {
+      "epoch": 6.885089340727049,
+      "grad_norm": 0.04893886297941208,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 67047
+    },
+    {
+      "epoch": 6.88539741219963,
+      "grad_norm": 0.10105928778648376,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 67050
+    },
+    {
+      "epoch": 6.885705483672212,
+      "grad_norm": 0.08832778036594391,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 67053
+    },
+    {
+      "epoch": 6.886013555144793,
+      "grad_norm": 0.059674110263586044,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 67056
+    },
+    {
+      "epoch": 6.8863216266173755,
+      "grad_norm": 0.06233122944831848,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 67059
+    },
+    {
+      "epoch": 6.886629698089957,
+      "grad_norm": 0.04024443402886391,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 67062
+    },
+    {
+      "epoch": 6.886937769562539,
+      "grad_norm": 0.03950473666191101,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 67065
+    },
+    {
+      "epoch": 6.88724584103512,
+      "grad_norm": 0.11595027148723602,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 67068
+    },
+    {
+      "epoch": 6.887553912507702,
+      "grad_norm": 0.10208969563245773,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 67071
+    },
+    {
+      "epoch": 6.887861983980283,
+      "grad_norm": 0.1006581112742424,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 67074
+    },
+    {
+      "epoch": 6.888170055452865,
+      "grad_norm": 0.06794067472219467,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 67077
+    },
+    {
+      "epoch": 6.888478126925447,
+      "grad_norm": 0.06897623836994171,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 67080
+    },
+    {
+      "epoch": 6.888786198398028,
+      "grad_norm": 0.07438699901103973,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 67083
+    },
+    {
+      "epoch": 6.88909426987061,
+      "grad_norm": 0.057675886899232864,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 67086
+    },
+    {
+      "epoch": 6.8894023413431915,
+      "grad_norm": 0.0770813375711441,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 67089
+    },
+    {
+      "epoch": 6.889710412815774,
+      "grad_norm": 0.07523828744888306,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 67092
+    },
+    {
+      "epoch": 6.890018484288355,
+      "grad_norm": 0.11752041429281235,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 67095
+    },
+    {
+      "epoch": 6.890326555760937,
+      "grad_norm": 0.07361220568418503,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 67098
+    },
+    {
+      "epoch": 6.890634627233518,
+      "grad_norm": 0.1421179324388504,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 67101
+    },
+    {
+      "epoch": 6.8909426987061,
+      "grad_norm": 0.10113172978162766,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 67104
+    },
+    {
+      "epoch": 6.891250770178681,
+      "grad_norm": 0.07775873690843582,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 67107
+    },
+    {
+      "epoch": 6.891558841651263,
+      "grad_norm": 0.06705240905284882,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 67110
+    },
+    {
+      "epoch": 6.891866913123844,
+      "grad_norm": 0.09598524123430252,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 67113
+    },
+    {
+      "epoch": 6.892174984596426,
+      "grad_norm": 0.04524253308773041,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 67116
+    },
+    {
+      "epoch": 6.892483056069008,
+      "grad_norm": 0.041274599730968475,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 67119
+    },
+    {
+      "epoch": 6.89279112754159,
+      "grad_norm": 0.09016481041908264,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 67122
+    },
+    {
+      "epoch": 6.893099199014172,
+      "grad_norm": 0.06743042171001434,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 67125
+    },
+    {
+      "epoch": 6.893407270486753,
+      "grad_norm": 0.08920291066169739,
+      "learning_rate": 0.01,
+      "loss": 2.0093,
+      "step": 67128
+    },
+    {
+      "epoch": 6.893715341959335,
+      "grad_norm": 0.08669997751712799,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 67131
+    },
+    {
+      "epoch": 6.894023413431916,
+      "grad_norm": 0.03727005049586296,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 67134
+    },
+    {
+      "epoch": 6.894331484904498,
+      "grad_norm": 0.047707851976156235,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 67137
+    },
+    {
+      "epoch": 6.894639556377079,
+      "grad_norm": 0.03828176483511925,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 67140
+    },
+    {
+      "epoch": 6.894947627849661,
+      "grad_norm": 0.04043647646903992,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 67143
+    },
+    {
+      "epoch": 6.8952556993222425,
+      "grad_norm": 0.03885728865861893,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 67146
+    },
+    {
+      "epoch": 6.8955637707948245,
+      "grad_norm": 0.041276995092630386,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 67149
+    },
+    {
+      "epoch": 6.895871842267406,
+      "grad_norm": 0.04943333566188812,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 67152
+    },
+    {
+      "epoch": 6.896179913739988,
+      "grad_norm": 0.04476524516940117,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 67155
+    },
+    {
+      "epoch": 6.89648798521257,
+      "grad_norm": 0.10550431907176971,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 67158
+    },
+    {
+      "epoch": 6.896796056685151,
+      "grad_norm": 0.09191994369029999,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 67161
+    },
+    {
+      "epoch": 6.897104128157732,
+      "grad_norm": 0.05506186559796333,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 67164
+    },
+    {
+      "epoch": 6.897412199630314,
+      "grad_norm": 0.06780190020799637,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 67167
+    },
+    {
+      "epoch": 6.897720271102896,
+      "grad_norm": 0.04398869350552559,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 67170
+    },
+    {
+      "epoch": 6.898028342575477,
+      "grad_norm": 0.05016426742076874,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 67173
+    },
+    {
+      "epoch": 6.898336414048059,
+      "grad_norm": 0.03760296478867531,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 67176
+    },
+    {
+      "epoch": 6.898644485520641,
+      "grad_norm": 0.06771603971719742,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 67179
+    },
+    {
+      "epoch": 6.898952556993223,
+      "grad_norm": 0.07435780018568039,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 67182
+    },
+    {
+      "epoch": 6.899260628465804,
+      "grad_norm": 0.05153966695070267,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 67185
+    },
+    {
+      "epoch": 6.899568699938386,
+      "grad_norm": 0.06906291097402573,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 67188
+    },
+    {
+      "epoch": 6.899876771410967,
+      "grad_norm": 0.08245383203029633,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 67191
+    },
+    {
+      "epoch": 6.900184842883549,
+      "grad_norm": 0.06429754942655563,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 67194
+    },
+    {
+      "epoch": 6.90049291435613,
+      "grad_norm": 0.03510384261608124,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 67197
+    },
+    {
+      "epoch": 6.900800985828712,
+      "grad_norm": 0.063795305788517,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 67200
+    },
+    {
+      "epoch": 6.901109057301294,
+      "grad_norm": 0.07294676452875137,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 67203
+    },
+    {
+      "epoch": 6.9014171287738755,
+      "grad_norm": 0.04569892957806587,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 67206
+    },
+    {
+      "epoch": 6.9017252002464575,
+      "grad_norm": 0.0556999109685421,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 67209
+    },
+    {
+      "epoch": 6.902033271719039,
+      "grad_norm": 0.038515105843544006,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 67212
+    },
+    {
+      "epoch": 6.902341343191621,
+      "grad_norm": 0.08272943645715714,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 67215
+    },
+    {
+      "epoch": 6.902649414664202,
+      "grad_norm": 0.08796203136444092,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 67218
+    },
+    {
+      "epoch": 6.902957486136784,
+      "grad_norm": 0.0925317257642746,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 67221
+    },
+    {
+      "epoch": 6.903265557609365,
+      "grad_norm": 0.04561033099889755,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 67224
+    },
+    {
+      "epoch": 6.903573629081947,
+      "grad_norm": 0.10068716108798981,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 67227
+    },
+    {
+      "epoch": 6.903881700554528,
+      "grad_norm": 0.06739984452724457,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 67230
+    },
+    {
+      "epoch": 6.90418977202711,
+      "grad_norm": 0.05066872388124466,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 67233
+    },
+    {
+      "epoch": 6.904497843499692,
+      "grad_norm": 0.047137267887592316,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 67236
+    },
+    {
+      "epoch": 6.904805914972274,
+      "grad_norm": 0.10280490666627884,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 67239
+    },
+    {
+      "epoch": 6.905113986444855,
+      "grad_norm": 0.08560338616371155,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 67242
+    },
+    {
+      "epoch": 6.905422057917437,
+      "grad_norm": 0.13022232055664062,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 67245
+    },
+    {
+      "epoch": 6.905730129390019,
+      "grad_norm": 0.04727041721343994,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 67248
+    },
+    {
+      "epoch": 6.9060382008626,
+      "grad_norm": 0.04549260064959526,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 67251
+    },
+    {
+      "epoch": 6.906346272335182,
+      "grad_norm": 0.14286646246910095,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 67254
+    },
+    {
+      "epoch": 6.906654343807763,
+      "grad_norm": 0.04126777872443199,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 67257
+    },
+    {
+      "epoch": 6.906962415280345,
+      "grad_norm": 0.04667791351675987,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 67260
+    },
+    {
+      "epoch": 6.907270486752926,
+      "grad_norm": 0.11388275027275085,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 67263
+    },
+    {
+      "epoch": 6.9075785582255085,
+      "grad_norm": 0.06364165991544724,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 67266
+    },
+    {
+      "epoch": 6.90788662969809,
+      "grad_norm": 0.09224580973386765,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 67269
+    },
+    {
+      "epoch": 6.908194701170672,
+      "grad_norm": 0.07214430719614029,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 67272
+    },
+    {
+      "epoch": 6.908502772643253,
+      "grad_norm": 0.04083709791302681,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 67275
+    },
+    {
+      "epoch": 6.908810844115835,
+      "grad_norm": 0.034071508795022964,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 67278
+    },
+    {
+      "epoch": 6.909118915588417,
+      "grad_norm": 0.06297708302736282,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 67281
+    },
+    {
+      "epoch": 6.909426987060998,
+      "grad_norm": 0.0424063615500927,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 67284
+    },
+    {
+      "epoch": 6.90973505853358,
+      "grad_norm": 0.04424364119768143,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 67287
+    },
+    {
+      "epoch": 6.910043130006161,
+      "grad_norm": 0.03503073751926422,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 67290
+    },
+    {
+      "epoch": 6.910351201478743,
+      "grad_norm": 0.04922670125961304,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 67293
+    },
+    {
+      "epoch": 6.9106592729513245,
+      "grad_norm": 0.0473727248609066,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 67296
+    },
+    {
+      "epoch": 6.910967344423907,
+      "grad_norm": 0.09901537001132965,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 67299
+    },
+    {
+      "epoch": 6.911275415896488,
+      "grad_norm": 0.08178123086690903,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 67302
+    },
+    {
+      "epoch": 6.91158348736907,
+      "grad_norm": 0.06637603044509888,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 67305
+    },
+    {
+      "epoch": 6.911891558841651,
+      "grad_norm": 0.043472740799188614,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 67308
+    },
+    {
+      "epoch": 6.912199630314233,
+      "grad_norm": 0.03270704671740532,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 67311
+    },
+    {
+      "epoch": 6.912507701786814,
+      "grad_norm": 0.08558196574449539,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 67314
+    },
+    {
+      "epoch": 6.912815773259396,
+      "grad_norm": 0.09473107755184174,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 67317
+    },
+    {
+      "epoch": 6.913123844731977,
+      "grad_norm": 0.041321370750665665,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 67320
+    },
+    {
+      "epoch": 6.913431916204559,
+      "grad_norm": 0.10107487440109253,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 67323
+    },
+    {
+      "epoch": 6.9137399876771415,
+      "grad_norm": 0.13234496116638184,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 67326
+    },
+    {
+      "epoch": 6.914048059149723,
+      "grad_norm": 0.05052580311894417,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 67329
+    },
+    {
+      "epoch": 6.914356130622305,
+      "grad_norm": 0.03741452470421791,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 67332
+    },
+    {
+      "epoch": 6.914664202094886,
+      "grad_norm": 0.03609738498926163,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 67335
+    },
+    {
+      "epoch": 6.914972273567468,
+      "grad_norm": 0.045461494475603104,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 67338
+    },
+    {
+      "epoch": 6.915280345040049,
+      "grad_norm": 0.07647386193275452,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 67341
+    },
+    {
+      "epoch": 6.915588416512631,
+      "grad_norm": 0.0546477772295475,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 67344
+    },
+    {
+      "epoch": 6.915896487985212,
+      "grad_norm": 0.05500864237546921,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 67347
+    },
+    {
+      "epoch": 6.916204559457794,
+      "grad_norm": 0.06956373900175095,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 67350
+    },
+    {
+      "epoch": 6.9165126309303755,
+      "grad_norm": 0.13182616233825684,
+      "learning_rate": 0.01,
+      "loss": 1.9938,
+      "step": 67353
+    },
+    {
+      "epoch": 6.9168207024029575,
+      "grad_norm": 0.06322868913412094,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 67356
+    },
+    {
+      "epoch": 6.91712877387554,
+      "grad_norm": 0.053960807621479034,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 67359
+    },
+    {
+      "epoch": 6.917436845348121,
+      "grad_norm": 0.03774489462375641,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 67362
+    },
+    {
+      "epoch": 6.917744916820702,
+      "grad_norm": 0.04947468638420105,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 67365
+    },
+    {
+      "epoch": 6.918052988293284,
+      "grad_norm": 0.09014926850795746,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 67368
+    },
+    {
+      "epoch": 6.918361059765866,
+      "grad_norm": 0.047003794461488724,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 67371
+    },
+    {
+      "epoch": 6.918669131238447,
+      "grad_norm": 0.04293885454535484,
+      "learning_rate": 0.01,
+      "loss": 2.0253,
+      "step": 67374
+    },
+    {
+      "epoch": 6.918977202711029,
+      "grad_norm": 0.08349283039569855,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 67377
+    },
+    {
+      "epoch": 6.91928527418361,
+      "grad_norm": 0.10643094033002853,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 67380
+    },
+    {
+      "epoch": 6.919593345656192,
+      "grad_norm": 0.05152636021375656,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 67383
+    },
+    {
+      "epoch": 6.919901417128774,
+      "grad_norm": 0.045483458787202835,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 67386
+    },
+    {
+      "epoch": 6.920209488601356,
+      "grad_norm": 0.10970417410135269,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 67389
+    },
+    {
+      "epoch": 6.920517560073937,
+      "grad_norm": 0.0620252899825573,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 67392
+    },
+    {
+      "epoch": 6.920825631546519,
+      "grad_norm": 0.042169239372015,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 67395
+    },
+    {
+      "epoch": 6.9211337030191,
+      "grad_norm": 0.07429879903793335,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 67398
+    },
+    {
+      "epoch": 6.921441774491682,
+      "grad_norm": 0.16096588969230652,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 67401
+    },
+    {
+      "epoch": 6.921749845964264,
+      "grad_norm": 0.06262468546628952,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 67404
+    },
+    {
+      "epoch": 6.922057917436845,
+      "grad_norm": 0.08196594566106796,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 67407
+    },
+    {
+      "epoch": 6.922365988909427,
+      "grad_norm": 0.07014669477939606,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 67410
+    },
+    {
+      "epoch": 6.9226740603820085,
+      "grad_norm": 0.048560310155153275,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 67413
+    },
+    {
+      "epoch": 6.9229821318545905,
+      "grad_norm": 0.053055111318826675,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 67416
+    },
+    {
+      "epoch": 6.923290203327172,
+      "grad_norm": 0.0406053327023983,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 67419
+    },
+    {
+      "epoch": 6.923598274799754,
+      "grad_norm": 0.05388790741562843,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 67422
+    },
+    {
+      "epoch": 6.923906346272335,
+      "grad_norm": 0.03949485719203949,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 67425
+    },
+    {
+      "epoch": 6.924214417744917,
+      "grad_norm": 0.04466044530272484,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 67428
+    },
+    {
+      "epoch": 6.924522489217498,
+      "grad_norm": 0.08420266211032867,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 67431
+    },
+    {
+      "epoch": 6.92483056069008,
+      "grad_norm": 0.06232386827468872,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 67434
+    },
+    {
+      "epoch": 6.925138632162662,
+      "grad_norm": 0.10322209447622299,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 67437
+    },
+    {
+      "epoch": 6.925446703635243,
+      "grad_norm": 0.04155166074633598,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 67440
+    },
+    {
+      "epoch": 6.925754775107825,
+      "grad_norm": 0.08774439245462418,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 67443
+    },
+    {
+      "epoch": 6.926062846580407,
+      "grad_norm": 0.08442433178424835,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 67446
+    },
+    {
+      "epoch": 6.926370918052989,
+      "grad_norm": 0.0743742361664772,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 67449
+    },
+    {
+      "epoch": 6.92667898952557,
+      "grad_norm": 0.09326713532209396,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 67452
+    },
+    {
+      "epoch": 6.926987060998152,
+      "grad_norm": 0.05275671184062958,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 67455
+    },
+    {
+      "epoch": 6.927295132470733,
+      "grad_norm": 0.0862322598695755,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 67458
+    },
+    {
+      "epoch": 6.927603203943315,
+      "grad_norm": 0.06256649643182755,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 67461
+    },
+    {
+      "epoch": 6.927911275415896,
+      "grad_norm": 0.08660640567541122,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 67464
+    },
+    {
+      "epoch": 6.928219346888478,
+      "grad_norm": 0.05134911462664604,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 67467
+    },
+    {
+      "epoch": 6.9285274183610595,
+      "grad_norm": 0.09808807820081711,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 67470
+    },
+    {
+      "epoch": 6.9288354898336415,
+      "grad_norm": 0.04677712544798851,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 67473
+    },
+    {
+      "epoch": 6.929143561306223,
+      "grad_norm": 0.040743302553892136,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 67476
+    },
+    {
+      "epoch": 6.929451632778805,
+      "grad_norm": 0.04844764620065689,
+      "learning_rate": 0.01,
+      "loss": 2.0035,
+      "step": 67479
+    },
+    {
+      "epoch": 6.929759704251387,
+      "grad_norm": 0.042815107852220535,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 67482
+    },
+    {
+      "epoch": 6.930067775723968,
+      "grad_norm": 0.040781863033771515,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 67485
+    },
+    {
+      "epoch": 6.93037584719655,
+      "grad_norm": 0.07171075791120529,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 67488
+    },
+    {
+      "epoch": 6.930683918669131,
+      "grad_norm": 0.03687571734189987,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 67491
+    },
+    {
+      "epoch": 6.930991990141713,
+      "grad_norm": 0.04326467588543892,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 67494
+    },
+    {
+      "epoch": 6.931300061614294,
+      "grad_norm": 0.24775917828083038,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 67497
+    },
+    {
+      "epoch": 6.931608133086876,
+      "grad_norm": 0.09830349683761597,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 67500
+    },
+    {
+      "epoch": 6.931916204559458,
+      "grad_norm": 0.05593564361333847,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 67503
+    },
+    {
+      "epoch": 6.93222427603204,
+      "grad_norm": 0.05278097838163376,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 67506
+    },
+    {
+      "epoch": 6.932532347504621,
+      "grad_norm": 0.06709492206573486,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 67509
+    },
+    {
+      "epoch": 6.932840418977203,
+      "grad_norm": 0.08978651463985443,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 67512
+    },
+    {
+      "epoch": 6.933148490449784,
+      "grad_norm": 0.10322022438049316,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 67515
+    },
+    {
+      "epoch": 6.933456561922366,
+      "grad_norm": 0.08842363208532333,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 67518
+    },
+    {
+      "epoch": 6.933764633394947,
+      "grad_norm": 0.1034180223941803,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 67521
+    },
+    {
+      "epoch": 6.934072704867529,
+      "grad_norm": 0.05395280942320824,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 67524
+    },
+    {
+      "epoch": 6.934380776340111,
+      "grad_norm": 0.04033922404050827,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 67527
+    },
+    {
+      "epoch": 6.9346888478126925,
+      "grad_norm": 0.10645033419132233,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 67530
+    },
+    {
+      "epoch": 6.9349969192852745,
+      "grad_norm": 0.03000570461153984,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 67533
+    },
+    {
+      "epoch": 6.935304990757856,
+      "grad_norm": 0.06463263183832169,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 67536
+    },
+    {
+      "epoch": 6.935613062230438,
+      "grad_norm": 0.0651232898235321,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 67539
+    },
+    {
+      "epoch": 6.935921133703019,
+      "grad_norm": 0.048303090035915375,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 67542
+    },
+    {
+      "epoch": 6.936229205175601,
+      "grad_norm": 0.09571239352226257,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 67545
+    },
+    {
+      "epoch": 6.936537276648182,
+      "grad_norm": 0.04778117686510086,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 67548
+    },
+    {
+      "epoch": 6.936845348120764,
+      "grad_norm": 0.16268859803676605,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 67551
+    },
+    {
+      "epoch": 6.937153419593345,
+      "grad_norm": 0.08549009263515472,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 67554
+    },
+    {
+      "epoch": 6.937461491065927,
+      "grad_norm": 0.0720728412270546,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 67557
+    },
+    {
+      "epoch": 6.937769562538509,
+      "grad_norm": 0.04524467885494232,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 67560
+    },
+    {
+      "epoch": 6.938077634011091,
+      "grad_norm": 0.0630386620759964,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 67563
+    },
+    {
+      "epoch": 6.938385705483672,
+      "grad_norm": 0.07228325307369232,
+      "learning_rate": 0.01,
+      "loss": 2.0101,
+      "step": 67566
+    },
+    {
+      "epoch": 6.938693776956254,
+      "grad_norm": 0.0827326700091362,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 67569
+    },
+    {
+      "epoch": 6.939001848428836,
+      "grad_norm": 0.0640362948179245,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 67572
+    },
+    {
+      "epoch": 6.939309919901417,
+      "grad_norm": 0.0734630599617958,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 67575
+    },
+    {
+      "epoch": 6.939617991373999,
+      "grad_norm": 0.09056416898965836,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 67578
+    },
+    {
+      "epoch": 6.93992606284658,
+      "grad_norm": 0.04308030381798744,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 67581
+    },
+    {
+      "epoch": 6.940234134319162,
+      "grad_norm": 0.038212817162275314,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 67584
+    },
+    {
+      "epoch": 6.940542205791743,
+      "grad_norm": 0.06669415533542633,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 67587
+    },
+    {
+      "epoch": 6.9408502772643255,
+      "grad_norm": 0.10573708266019821,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 67590
+    },
+    {
+      "epoch": 6.941158348736907,
+      "grad_norm": 0.13396821916103363,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 67593
+    },
+    {
+      "epoch": 6.941466420209489,
+      "grad_norm": 0.0890778973698616,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 67596
+    },
+    {
+      "epoch": 6.94177449168207,
+      "grad_norm": 0.06729544699192047,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 67599
+    },
+    {
+      "epoch": 6.942082563154652,
+      "grad_norm": 0.04166124016046524,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 67602
+    },
+    {
+      "epoch": 6.942390634627234,
+      "grad_norm": 0.06778619438409805,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 67605
+    },
+    {
+      "epoch": 6.942698706099815,
+      "grad_norm": 0.052519191056489944,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 67608
+    },
+    {
+      "epoch": 6.943006777572397,
+      "grad_norm": 0.051862701773643494,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 67611
+    },
+    {
+      "epoch": 6.943314849044978,
+      "grad_norm": 0.04436833783984184,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 67614
+    },
+    {
+      "epoch": 6.94362292051756,
+      "grad_norm": 0.08768297731876373,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 67617
+    },
+    {
+      "epoch": 6.9439309919901415,
+      "grad_norm": 0.10748183727264404,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 67620
+    },
+    {
+      "epoch": 6.944239063462724,
+      "grad_norm": 0.06303614377975464,
+      "learning_rate": 0.01,
+      "loss": 2.0053,
+      "step": 67623
+    },
+    {
+      "epoch": 6.944547134935305,
+      "grad_norm": 0.0495496429502964,
+      "learning_rate": 0.01,
+      "loss": 2.0075,
+      "step": 67626
+    },
+    {
+      "epoch": 6.944855206407887,
+      "grad_norm": 0.07309143245220184,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 67629
+    },
+    {
+      "epoch": 6.945163277880468,
+      "grad_norm": 0.09533344954252243,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 67632
+    },
+    {
+      "epoch": 6.94547134935305,
+      "grad_norm": 0.07221339643001556,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 67635
+    },
+    {
+      "epoch": 6.945779420825632,
+      "grad_norm": 0.0772852674126625,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 67638
+    },
+    {
+      "epoch": 6.946087492298213,
+      "grad_norm": 0.03566101938486099,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 67641
+    },
+    {
+      "epoch": 6.946395563770794,
+      "grad_norm": 0.0641225203871727,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 67644
+    },
+    {
+      "epoch": 6.946703635243376,
+      "grad_norm": 0.11845903843641281,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 67647
+    },
+    {
+      "epoch": 6.9470117067159585,
+      "grad_norm": 0.1297493278980255,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 67650
+    },
+    {
+      "epoch": 6.94731977818854,
+      "grad_norm": 0.0657927542924881,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 67653
+    },
+    {
+      "epoch": 6.947627849661122,
+      "grad_norm": 0.0618324875831604,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 67656
+    },
+    {
+      "epoch": 6.947935921133703,
+      "grad_norm": 0.04335208609700203,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 67659
+    },
+    {
+      "epoch": 6.948243992606285,
+      "grad_norm": 0.03921616077423096,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 67662
+    },
+    {
+      "epoch": 6.948552064078866,
+      "grad_norm": 0.05289353057742119,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 67665
+    },
+    {
+      "epoch": 6.948860135551448,
+      "grad_norm": 0.037309400737285614,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 67668
+    },
+    {
+      "epoch": 6.949168207024029,
+      "grad_norm": 0.07012271881103516,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 67671
+    },
+    {
+      "epoch": 6.949476278496611,
+      "grad_norm": 0.08038291335105896,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 67674
+    },
+    {
+      "epoch": 6.9497843499691925,
+      "grad_norm": 0.08565990626811981,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 67677
+    },
+    {
+      "epoch": 6.9500924214417745,
+      "grad_norm": 0.13325315713882446,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 67680
+    },
+    {
+      "epoch": 6.950400492914357,
+      "grad_norm": 0.06681843101978302,
+      "learning_rate": 0.01,
+      "loss": 1.9903,
+      "step": 67683
+    },
+    {
+      "epoch": 6.950708564386938,
+      "grad_norm": 0.051343731582164764,
+      "learning_rate": 0.01,
+      "loss": 2.0182,
+      "step": 67686
+    },
+    {
+      "epoch": 6.95101663585952,
+      "grad_norm": 0.031108930706977844,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 67689
+    },
+    {
+      "epoch": 6.951324707332101,
+      "grad_norm": 0.06201615929603577,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 67692
+    },
+    {
+      "epoch": 6.951632778804683,
+      "grad_norm": 0.059839654713869095,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 67695
+    },
+    {
+      "epoch": 6.951940850277264,
+      "grad_norm": 0.04882393404841423,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 67698
+    },
+    {
+      "epoch": 6.952248921749846,
+      "grad_norm": 0.031460147351026535,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 67701
+    },
+    {
+      "epoch": 6.952556993222427,
+      "grad_norm": 0.09777851402759552,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 67704
+    },
+    {
+      "epoch": 6.952865064695009,
+      "grad_norm": 0.0367160402238369,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 67707
+    },
+    {
+      "epoch": 6.953173136167591,
+      "grad_norm": 0.032731834799051285,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 67710
+    },
+    {
+      "epoch": 6.953481207640173,
+      "grad_norm": 0.07224103063344955,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 67713
+    },
+    {
+      "epoch": 6.953789279112754,
+      "grad_norm": 0.062362559139728546,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 67716
+    },
+    {
+      "epoch": 6.954097350585336,
+      "grad_norm": 0.05006725341081619,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 67719
+    },
+    {
+      "epoch": 6.954405422057917,
+      "grad_norm": 0.03857605531811714,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 67722
+    },
+    {
+      "epoch": 6.954713493530499,
+      "grad_norm": 0.10484008491039276,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 67725
+    },
+    {
+      "epoch": 6.955021565003081,
+      "grad_norm": 0.09993112832307816,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 67728
+    },
+    {
+      "epoch": 6.955329636475662,
+      "grad_norm": 0.08906539529561996,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 67731
+    },
+    {
+      "epoch": 6.955637707948244,
+      "grad_norm": 0.04950569570064545,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 67734
+    },
+    {
+      "epoch": 6.9559457794208255,
+      "grad_norm": 0.04009339585900307,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 67737
+    },
+    {
+      "epoch": 6.9562538508934075,
+      "grad_norm": 0.03402652591466904,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 67740
+    },
+    {
+      "epoch": 6.956561922365989,
+      "grad_norm": 0.036231786012649536,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 67743
+    },
+    {
+      "epoch": 6.956869993838571,
+      "grad_norm": 0.06914620101451874,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 67746
+    },
+    {
+      "epoch": 6.957178065311152,
+      "grad_norm": 0.12120147049427032,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 67749
+    },
+    {
+      "epoch": 6.957486136783734,
+      "grad_norm": 0.04155363887548447,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 67752
+    },
+    {
+      "epoch": 6.957794208256315,
+      "grad_norm": 0.03857012465596199,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 67755
+    },
+    {
+      "epoch": 6.958102279728897,
+      "grad_norm": 0.05268177390098572,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 67758
+    },
+    {
+      "epoch": 6.958410351201479,
+      "grad_norm": 0.14224505424499512,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 67761
+    },
+    {
+      "epoch": 6.95871842267406,
+      "grad_norm": 0.10331138968467712,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 67764
+    },
+    {
+      "epoch": 6.9590264941466415,
+      "grad_norm": 0.13701403141021729,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 67767
+    },
+    {
+      "epoch": 6.959334565619224,
+      "grad_norm": 0.09614556282758713,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 67770
+    },
+    {
+      "epoch": 6.959642637091806,
+      "grad_norm": 0.07385848462581635,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 67773
+    },
+    {
+      "epoch": 6.959950708564387,
+      "grad_norm": 0.05891263857483864,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 67776
+    },
+    {
+      "epoch": 6.960258780036969,
+      "grad_norm": 0.06532695889472961,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 67779
+    },
+    {
+      "epoch": 6.96056685150955,
+      "grad_norm": 0.048335302621126175,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 67782
+    },
+    {
+      "epoch": 6.960874922982132,
+      "grad_norm": 0.0490678995847702,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 67785
+    },
+    {
+      "epoch": 6.961182994454713,
+      "grad_norm": 0.057322729378938675,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 67788
+    },
+    {
+      "epoch": 6.961491065927295,
+      "grad_norm": 0.09200582653284073,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 67791
+    },
+    {
+      "epoch": 6.961799137399876,
+      "grad_norm": 0.08140811324119568,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 67794
+    },
+    {
+      "epoch": 6.9621072088724585,
+      "grad_norm": 0.08386159688234329,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 67797
+    },
+    {
+      "epoch": 6.96241528034504,
+      "grad_norm": 0.09306125342845917,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 67800
+    },
+    {
+      "epoch": 6.962723351817622,
+      "grad_norm": 0.0702492967247963,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 67803
+    },
+    {
+      "epoch": 6.963031423290204,
+      "grad_norm": 0.07399389147758484,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 67806
+    },
+    {
+      "epoch": 6.963339494762785,
+      "grad_norm": 0.05269570276141167,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 67809
+    },
+    {
+      "epoch": 6.963647566235367,
+      "grad_norm": 0.0840226262807846,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 67812
+    },
+    {
+      "epoch": 6.963955637707948,
+      "grad_norm": 0.06911315023899078,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 67815
+    },
+    {
+      "epoch": 6.96426370918053,
+      "grad_norm": 0.04625274986028671,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 67818
+    },
+    {
+      "epoch": 6.964571780653111,
+      "grad_norm": 0.033663198351860046,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 67821
+    },
+    {
+      "epoch": 6.964879852125693,
+      "grad_norm": 0.035202909260988235,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 67824
+    },
+    {
+      "epoch": 6.9651879235982745,
+      "grad_norm": 0.04282594099640846,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 67827
+    },
+    {
+      "epoch": 6.965495995070857,
+      "grad_norm": 0.07154384255409241,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 67830
+    },
+    {
+      "epoch": 6.965804066543438,
+      "grad_norm": 0.09028030931949615,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 67833
+    },
+    {
+      "epoch": 6.96611213801602,
+      "grad_norm": 0.046586085110902786,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 67836
+    },
+    {
+      "epoch": 6.966420209488601,
+      "grad_norm": 0.03928957134485245,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 67839
+    },
+    {
+      "epoch": 6.966728280961183,
+      "grad_norm": 0.04739971458911896,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 67842
+    },
+    {
+      "epoch": 6.967036352433764,
+      "grad_norm": 0.06463371962308884,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 67845
+    },
+    {
+      "epoch": 6.967344423906346,
+      "grad_norm": 0.03283145651221275,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 67848
+    },
+    {
+      "epoch": 6.967652495378928,
+      "grad_norm": 0.15586607158184052,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 67851
+    },
+    {
+      "epoch": 6.967960566851509,
+      "grad_norm": 0.08082199841737747,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 67854
+    },
+    {
+      "epoch": 6.9682686383240915,
+      "grad_norm": 0.06927336007356644,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 67857
+    },
+    {
+      "epoch": 6.968576709796673,
+      "grad_norm": 0.03947534039616585,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 67860
+    },
+    {
+      "epoch": 6.968884781269255,
+      "grad_norm": 0.059715401381254196,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 67863
+    },
+    {
+      "epoch": 6.969192852741836,
+      "grad_norm": 0.056624263525009155,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 67866
+    },
+    {
+      "epoch": 6.969500924214418,
+      "grad_norm": 0.0887855812907219,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 67869
+    },
+    {
+      "epoch": 6.969808995686999,
+      "grad_norm": 0.08432299643754959,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 67872
+    },
+    {
+      "epoch": 6.970117067159581,
+      "grad_norm": 0.0985739529132843,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 67875
+    },
+    {
+      "epoch": 6.970425138632162,
+      "grad_norm": 0.0615275539457798,
+      "learning_rate": 0.01,
+      "loss": 2.0046,
+      "step": 67878
+    },
+    {
+      "epoch": 6.970733210104744,
+      "grad_norm": 0.04872935637831688,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 67881
+    },
+    {
+      "epoch": 6.971041281577326,
+      "grad_norm": 0.04057471454143524,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 67884
+    },
+    {
+      "epoch": 6.9713493530499075,
+      "grad_norm": 0.09079626202583313,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 67887
+    },
+    {
+      "epoch": 6.97165742452249,
+      "grad_norm": 0.12390229851007462,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 67890
+    },
+    {
+      "epoch": 6.971965495995071,
+      "grad_norm": 0.08013787120580673,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 67893
+    },
+    {
+      "epoch": 6.972273567467653,
+      "grad_norm": 0.10543009638786316,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 67896
+    },
+    {
+      "epoch": 6.972581638940234,
+      "grad_norm": 0.06962577998638153,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 67899
+    },
+    {
+      "epoch": 6.972889710412816,
+      "grad_norm": 0.05179775133728981,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 67902
+    },
+    {
+      "epoch": 6.973197781885397,
+      "grad_norm": 0.04999608173966408,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 67905
+    },
+    {
+      "epoch": 6.973505853357979,
+      "grad_norm": 0.037668853998184204,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 67908
+    },
+    {
+      "epoch": 6.97381392483056,
+      "grad_norm": 0.04385687783360481,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 67911
+    },
+    {
+      "epoch": 6.9741219963031424,
+      "grad_norm": 0.03812088444828987,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 67914
+    },
+    {
+      "epoch": 6.974430067775724,
+      "grad_norm": 0.08592873066663742,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 67917
+    },
+    {
+      "epoch": 6.974738139248306,
+      "grad_norm": 0.07441151887178421,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 67920
+    },
+    {
+      "epoch": 6.975046210720887,
+      "grad_norm": 0.055950410664081573,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 67923
+    },
+    {
+      "epoch": 6.975354282193469,
+      "grad_norm": 0.07224567234516144,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 67926
+    },
+    {
+      "epoch": 6.975662353666051,
+      "grad_norm": 0.17487280070781708,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 67929
+    },
+    {
+      "epoch": 6.975970425138632,
+      "grad_norm": 0.040765196084976196,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 67932
+    },
+    {
+      "epoch": 6.976278496611214,
+      "grad_norm": 0.05561240762472153,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 67935
+    },
+    {
+      "epoch": 6.976586568083795,
+      "grad_norm": 0.07443460822105408,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 67938
+    },
+    {
+      "epoch": 6.976894639556377,
+      "grad_norm": 0.0669572651386261,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 67941
+    },
+    {
+      "epoch": 6.9772027110289585,
+      "grad_norm": 0.05161336436867714,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 67944
+    },
+    {
+      "epoch": 6.9775107825015406,
+      "grad_norm": 0.0417034886777401,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 67947
+    },
+    {
+      "epoch": 6.977818853974122,
+      "grad_norm": 0.06474210321903229,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 67950
+    },
+    {
+      "epoch": 6.978126925446704,
+      "grad_norm": 0.07746025174856186,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 67953
+    },
+    {
+      "epoch": 6.978434996919285,
+      "grad_norm": 0.045474573969841,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 67956
+    },
+    {
+      "epoch": 6.978743068391867,
+      "grad_norm": 0.03205155208706856,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 67959
+    },
+    {
+      "epoch": 6.979051139864449,
+      "grad_norm": 0.03780858963727951,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 67962
+    },
+    {
+      "epoch": 6.97935921133703,
+      "grad_norm": 0.04689191281795502,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 67965
+    },
+    {
+      "epoch": 6.979667282809611,
+      "grad_norm": 0.05307697504758835,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 67968
+    },
+    {
+      "epoch": 6.979975354282193,
+      "grad_norm": 0.19176854193210602,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 67971
+    },
+    {
+      "epoch": 6.9802834257547755,
+      "grad_norm": 0.13577581942081451,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 67974
+    },
+    {
+      "epoch": 6.980591497227357,
+      "grad_norm": 0.10115823894739151,
+      "learning_rate": 0.01,
+      "loss": 2.0068,
+      "step": 67977
+    },
+    {
+      "epoch": 6.980899568699939,
+      "grad_norm": 0.08148372173309326,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 67980
+    },
+    {
+      "epoch": 6.98120764017252,
+      "grad_norm": 0.05191081017255783,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 67983
+    },
+    {
+      "epoch": 6.981515711645102,
+      "grad_norm": 0.06080775707960129,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 67986
+    },
+    {
+      "epoch": 6.981823783117683,
+      "grad_norm": 0.07841724157333374,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 67989
+    },
+    {
+      "epoch": 6.982131854590265,
+      "grad_norm": 0.04652271047234535,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 67992
+    },
+    {
+      "epoch": 6.982439926062846,
+      "grad_norm": 0.10853028297424316,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 67995
+    },
+    {
+      "epoch": 6.982747997535428,
+      "grad_norm": 0.03634324297308922,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 67998
+    },
+    {
+      "epoch": 6.9830560690080095,
+      "grad_norm": 0.13538099825382233,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 68001
+    },
+    {
+      "epoch": 6.9833641404805915,
+      "grad_norm": 0.06273335963487625,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 68004
+    },
+    {
+      "epoch": 6.983672211953174,
+      "grad_norm": 0.08126869797706604,
+      "learning_rate": 0.01,
+      "loss": 2.0036,
+      "step": 68007
+    },
+    {
+      "epoch": 6.983980283425755,
+      "grad_norm": 0.05156498774886131,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 68010
+    },
+    {
+      "epoch": 6.984288354898337,
+      "grad_norm": 0.05055593326687813,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 68013
+    },
+    {
+      "epoch": 6.984596426370918,
+      "grad_norm": 0.04435109347105026,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 68016
+    },
+    {
+      "epoch": 6.9849044978435,
+      "grad_norm": 0.13580620288848877,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 68019
+    },
+    {
+      "epoch": 6.985212569316081,
+      "grad_norm": 0.03387708216905594,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 68022
+    },
+    {
+      "epoch": 6.985520640788663,
+      "grad_norm": 0.09359924495220184,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 68025
+    },
+    {
+      "epoch": 6.985828712261244,
+      "grad_norm": 0.09586784243583679,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 68028
+    },
+    {
+      "epoch": 6.986136783733826,
+      "grad_norm": 0.08071450144052505,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 68031
+    },
+    {
+      "epoch": 6.986444855206408,
+      "grad_norm": 0.04609205946326256,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 68034
+    },
+    {
+      "epoch": 6.98675292667899,
+      "grad_norm": 0.05660049244761467,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 68037
+    },
+    {
+      "epoch": 6.987060998151571,
+      "grad_norm": 0.04224297031760216,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 68040
+    },
+    {
+      "epoch": 6.987369069624153,
+      "grad_norm": 0.056567490100860596,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 68043
+    },
+    {
+      "epoch": 6.987677141096734,
+      "grad_norm": 0.10390833020210266,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 68046
+    },
+    {
+      "epoch": 6.987985212569316,
+      "grad_norm": 0.04906808212399483,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 68049
+    },
+    {
+      "epoch": 6.988293284041898,
+      "grad_norm": 0.0763496458530426,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 68052
+    },
+    {
+      "epoch": 6.988601355514479,
+      "grad_norm": 0.042168740183115005,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 68055
+    },
+    {
+      "epoch": 6.988909426987061,
+      "grad_norm": 0.044075168669223785,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 68058
+    },
+    {
+      "epoch": 6.9892174984596425,
+      "grad_norm": 0.06443364918231964,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 68061
+    },
+    {
+      "epoch": 6.9895255699322245,
+      "grad_norm": 0.06181558594107628,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 68064
+    },
+    {
+      "epoch": 6.989833641404806,
+      "grad_norm": 0.061378926038742065,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 68067
+    },
+    {
+      "epoch": 6.990141712877388,
+      "grad_norm": 0.11718502640724182,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 68070
+    },
+    {
+      "epoch": 6.990449784349969,
+      "grad_norm": 0.10844454169273376,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 68073
+    },
+    {
+      "epoch": 6.990757855822551,
+      "grad_norm": 0.04362885281443596,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 68076
+    },
+    {
+      "epoch": 6.991065927295132,
+      "grad_norm": 0.04131380096077919,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 68079
+    },
+    {
+      "epoch": 6.991373998767714,
+      "grad_norm": 0.03312958776950836,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 68082
+    },
+    {
+      "epoch": 6.991682070240296,
+      "grad_norm": 0.05648859217762947,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 68085
+    },
+    {
+      "epoch": 6.991990141712877,
+      "grad_norm": 0.07927577197551727,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 68088
+    },
+    {
+      "epoch": 6.992298213185459,
+      "grad_norm": 0.081300288438797,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 68091
+    },
+    {
+      "epoch": 6.992606284658041,
+      "grad_norm": 0.08239664137363434,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 68094
+    },
+    {
+      "epoch": 6.992914356130623,
+      "grad_norm": 0.14173032343387604,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 68097
+    },
+    {
+      "epoch": 6.993222427603204,
+      "grad_norm": 0.039897531270980835,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 68100
+    },
+    {
+      "epoch": 6.993530499075786,
+      "grad_norm": 0.08170229941606522,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 68103
+    },
+    {
+      "epoch": 6.993838570548367,
+      "grad_norm": 0.09108921140432358,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 68106
+    },
+    {
+      "epoch": 6.994146642020949,
+      "grad_norm": 0.048272643238306046,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 68109
+    },
+    {
+      "epoch": 6.99445471349353,
+      "grad_norm": 0.06137422099709511,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 68112
+    },
+    {
+      "epoch": 6.994762784966112,
+      "grad_norm": 0.060723815113306046,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 68115
+    },
+    {
+      "epoch": 6.995070856438693,
+      "grad_norm": 0.11094491183757782,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 68118
+    },
+    {
+      "epoch": 6.9953789279112755,
+      "grad_norm": 0.041866280138492584,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 68121
+    },
+    {
+      "epoch": 6.995686999383857,
+      "grad_norm": 0.032954033464193344,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 68124
+    },
+    {
+      "epoch": 6.995995070856439,
+      "grad_norm": 0.034947719424963,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 68127
+    },
+    {
+      "epoch": 6.996303142329021,
+      "grad_norm": 0.03938392922282219,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 68130
+    },
+    {
+      "epoch": 6.996611213801602,
+      "grad_norm": 0.1045796275138855,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 68133
+    },
+    {
+      "epoch": 6.996919285274184,
+      "grad_norm": 0.10855524986982346,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 68136
+    },
+    {
+      "epoch": 6.997227356746765,
+      "grad_norm": 0.08628056198358536,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 68139
+    },
+    {
+      "epoch": 6.997535428219347,
+      "grad_norm": 0.0505918487906456,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 68142
+    },
+    {
+      "epoch": 6.997843499691928,
+      "grad_norm": 0.04251931607723236,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 68145
+    },
+    {
+      "epoch": 6.99815157116451,
+      "grad_norm": 0.05464445427060127,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 68148
+    },
+    {
+      "epoch": 6.9984596426370915,
+      "grad_norm": 0.04259735345840454,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 68151
+    },
+    {
+      "epoch": 6.998767714109674,
+      "grad_norm": 0.06472709774971008,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 68154
+    },
+    {
+      "epoch": 6.999075785582255,
+      "grad_norm": 0.0695829764008522,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 68157
+    },
+    {
+      "epoch": 6.999383857054837,
+      "grad_norm": 0.09822755306959152,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 68160
+    },
+    {
+      "epoch": 6.999691928527419,
+      "grad_norm": 0.10002230852842331,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 68163
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.05069267004728317,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 68166
+    },
+    {
+      "epoch": 7.00606372045221,
+      "grad_norm": 0.054282162338495255,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 68169
+    },
+    {
+      "epoch": 7.006372045220966,
+      "grad_norm": 0.06988398730754852,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 68172
+    },
+    {
+      "epoch": 7.006680369989723,
+      "grad_norm": 0.06562650203704834,
+      "learning_rate": 0.01,
+      "loss": 2.0001,
+      "step": 68175
+    },
+    {
+      "epoch": 7.006988694758479,
+      "grad_norm": 0.04521455615758896,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 68178
+    },
+    {
+      "epoch": 7.007297019527235,
+      "grad_norm": 0.05735592916607857,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 68181
+    },
+    {
+      "epoch": 7.007605344295992,
+      "grad_norm": 0.08022990822792053,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 68184
+    },
+    {
+      "epoch": 7.007913669064748,
+      "grad_norm": 0.052689116448163986,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 68187
+    },
+    {
+      "epoch": 7.008221993833504,
+      "grad_norm": 0.03922291100025177,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 68190
+    },
+    {
+      "epoch": 7.008530318602261,
+      "grad_norm": 0.11853530257940292,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 68193
+    },
+    {
+      "epoch": 7.008838643371018,
+      "grad_norm": 0.07033178955316544,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 68196
+    },
+    {
+      "epoch": 7.009146968139774,
+      "grad_norm": 0.049129668623209,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 68199
+    },
+    {
+      "epoch": 7.00945529290853,
+      "grad_norm": 0.044174324721097946,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 68202
+    },
+    {
+      "epoch": 7.009763617677287,
+      "grad_norm": 0.05447569862008095,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 68205
+    },
+    {
+      "epoch": 7.010071942446043,
+      "grad_norm": 0.06834445148706436,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 68208
+    },
+    {
+      "epoch": 7.010380267214799,
+      "grad_norm": 0.056646499782800674,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 68211
+    },
+    {
+      "epoch": 7.010688591983556,
+      "grad_norm": 0.08008157461881638,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 68214
+    },
+    {
+      "epoch": 7.010996916752313,
+      "grad_norm": 0.07106824219226837,
+      "learning_rate": 0.01,
+      "loss": 1.9977,
+      "step": 68217
+    },
+    {
+      "epoch": 7.011305241521069,
+      "grad_norm": 0.05258387327194214,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 68220
+    },
+    {
+      "epoch": 7.011613566289825,
+      "grad_norm": 0.08909857273101807,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 68223
+    },
+    {
+      "epoch": 7.011921891058582,
+      "grad_norm": 0.05345101282000542,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 68226
+    },
+    {
+      "epoch": 7.012230215827338,
+      "grad_norm": 0.04160739853978157,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 68229
+    },
+    {
+      "epoch": 7.012538540596094,
+      "grad_norm": 0.03168926015496254,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 68232
+    },
+    {
+      "epoch": 7.012846865364851,
+      "grad_norm": 0.035804539918899536,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 68235
+    },
+    {
+      "epoch": 7.013155190133608,
+      "grad_norm": 0.04234134405851364,
+      "learning_rate": 0.01,
+      "loss": 2.0013,
+      "step": 68238
+    },
+    {
+      "epoch": 7.013463514902364,
+      "grad_norm": 0.06941762566566467,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 68241
+    },
+    {
+      "epoch": 7.01377183967112,
+      "grad_norm": 0.07563450187444687,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 68244
+    },
+    {
+      "epoch": 7.014080164439877,
+      "grad_norm": 0.07446388155221939,
+      "learning_rate": 0.01,
+      "loss": 2.0143,
+      "step": 68247
+    },
+    {
+      "epoch": 7.014388489208633,
+      "grad_norm": 0.040372107177972794,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 68250
+    },
+    {
+      "epoch": 7.014696813977389,
+      "grad_norm": 0.12671950459480286,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 68253
+    },
+    {
+      "epoch": 7.015005138746146,
+      "grad_norm": 0.06948719173669815,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 68256
+    },
+    {
+      "epoch": 7.015313463514902,
+      "grad_norm": 0.0657772570848465,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 68259
+    },
+    {
+      "epoch": 7.015621788283659,
+      "grad_norm": 0.07090414315462112,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 68262
+    },
+    {
+      "epoch": 7.015930113052415,
+      "grad_norm": 0.0721154510974884,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 68265
+    },
+    {
+      "epoch": 7.016238437821172,
+      "grad_norm": 0.05126158893108368,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 68268
+    },
+    {
+      "epoch": 7.016546762589928,
+      "grad_norm": 0.07218746840953827,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 68271
+    },
+    {
+      "epoch": 7.016855087358684,
+      "grad_norm": 0.08373812586069107,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 68274
+    },
+    {
+      "epoch": 7.017163412127441,
+      "grad_norm": 0.07843849062919617,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 68277
+    },
+    {
+      "epoch": 7.017471736896197,
+      "grad_norm": 0.042954906821250916,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 68280
+    },
+    {
+      "epoch": 7.017780061664954,
+      "grad_norm": 0.04174837842583656,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 68283
+    },
+    {
+      "epoch": 7.0180883864337105,
+      "grad_norm": 0.04753096401691437,
+      "learning_rate": 0.01,
+      "loss": 2.0111,
+      "step": 68286
+    },
+    {
+      "epoch": 7.018396711202467,
+      "grad_norm": 0.05489262193441391,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 68289
+    },
+    {
+      "epoch": 7.018705035971223,
+      "grad_norm": 0.09776978939771652,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 68292
+    },
+    {
+      "epoch": 7.019013360739979,
+      "grad_norm": 0.05086861178278923,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 68295
+    },
+    {
+      "epoch": 7.019321685508736,
+      "grad_norm": 0.07187198102474213,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 68298
+    },
+    {
+      "epoch": 7.019630010277492,
+      "grad_norm": 0.10095050185918808,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 68301
+    },
+    {
+      "epoch": 7.019938335046248,
+      "grad_norm": 0.06792483478784561,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 68304
+    },
+    {
+      "epoch": 7.0202466598150055,
+      "grad_norm": 0.050041310489177704,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 68307
+    },
+    {
+      "epoch": 7.020554984583762,
+      "grad_norm": 0.04710831120610237,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 68310
+    },
+    {
+      "epoch": 7.020863309352518,
+      "grad_norm": 0.056352101266384125,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 68313
+    },
+    {
+      "epoch": 7.021171634121274,
+      "grad_norm": 0.08362201601266861,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 68316
+    },
+    {
+      "epoch": 7.021479958890031,
+      "grad_norm": 0.03935181722044945,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 68319
+    },
+    {
+      "epoch": 7.021788283658787,
+      "grad_norm": 0.141304612159729,
+      "learning_rate": 0.01,
+      "loss": 2.0116,
+      "step": 68322
+    },
+    {
+      "epoch": 7.022096608427543,
+      "grad_norm": 0.041724372655153275,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 68325
+    },
+    {
+      "epoch": 7.0224049331963005,
+      "grad_norm": 0.04254898428916931,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 68328
+    },
+    {
+      "epoch": 7.022713257965057,
+      "grad_norm": 0.06883671879768372,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 68331
+    },
+    {
+      "epoch": 7.023021582733813,
+      "grad_norm": 0.05049630254507065,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 68334
+    },
+    {
+      "epoch": 7.0233299075025695,
+      "grad_norm": 0.047389786690473557,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 68337
+    },
+    {
+      "epoch": 7.023638232271326,
+      "grad_norm": 0.14553938806056976,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 68340
+    },
+    {
+      "epoch": 7.023946557040082,
+      "grad_norm": 0.0530347004532814,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 68343
+    },
+    {
+      "epoch": 7.024254881808838,
+      "grad_norm": 0.07391063868999481,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 68346
+    },
+    {
+      "epoch": 7.024563206577595,
+      "grad_norm": 0.082502581179142,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 68349
+    },
+    {
+      "epoch": 7.024871531346352,
+      "grad_norm": 0.09995315968990326,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 68352
+    },
+    {
+      "epoch": 7.025179856115108,
+      "grad_norm": 0.0911843478679657,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 68355
+    },
+    {
+      "epoch": 7.0254881808838645,
+      "grad_norm": 0.04006153717637062,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 68358
+    },
+    {
+      "epoch": 7.025796505652621,
+      "grad_norm": 0.09338364750146866,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 68361
+    },
+    {
+      "epoch": 7.026104830421377,
+      "grad_norm": 0.03738764673471451,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 68364
+    },
+    {
+      "epoch": 7.026413155190133,
+      "grad_norm": 0.048774126917123795,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 68367
+    },
+    {
+      "epoch": 7.02672147995889,
+      "grad_norm": 0.08035308867692947,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 68370
+    },
+    {
+      "epoch": 7.027029804727646,
+      "grad_norm": 0.054234910756349564,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 68373
+    },
+    {
+      "epoch": 7.027338129496403,
+      "grad_norm": 0.08153880387544632,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 68376
+    },
+    {
+      "epoch": 7.0276464542651595,
+      "grad_norm": 0.047218646854162216,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 68379
+    },
+    {
+      "epoch": 7.027954779033916,
+      "grad_norm": 0.061120592057704926,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 68382
+    },
+    {
+      "epoch": 7.028263103802672,
+      "grad_norm": 0.08370841294527054,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 68385
+    },
+    {
+      "epoch": 7.0285714285714285,
+      "grad_norm": 0.0927995815873146,
+      "learning_rate": 0.01,
+      "loss": 2.0014,
+      "step": 68388
+    },
+    {
+      "epoch": 7.028879753340185,
+      "grad_norm": 0.040699124336242676,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 68391
+    },
+    {
+      "epoch": 7.029188078108941,
+      "grad_norm": 0.12133225053548813,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 68394
+    },
+    {
+      "epoch": 7.029496402877698,
+      "grad_norm": 0.0616045743227005,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 68397
+    },
+    {
+      "epoch": 7.029804727646455,
+      "grad_norm": 0.0576595813035965,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 68400
+    },
+    {
+      "epoch": 7.030113052415211,
+      "grad_norm": 0.11460481584072113,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 68403
+    },
+    {
+      "epoch": 7.030421377183967,
+      "grad_norm": 0.046943407505750656,
+      "learning_rate": 0.01,
+      "loss": 2.0127,
+      "step": 68406
+    },
+    {
+      "epoch": 7.0307297019527235,
+      "grad_norm": 0.034080930054187775,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 68409
+    },
+    {
+      "epoch": 7.03103802672148,
+      "grad_norm": 0.09172356873750687,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 68412
+    },
+    {
+      "epoch": 7.031346351490236,
+      "grad_norm": 0.06134629622101784,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 68415
+    },
+    {
+      "epoch": 7.031654676258992,
+      "grad_norm": 0.038659993559122086,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 68418
+    },
+    {
+      "epoch": 7.03196300102775,
+      "grad_norm": 0.03268113732337952,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 68421
+    },
+    {
+      "epoch": 7.032271325796506,
+      "grad_norm": 0.05008908361196518,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 68424
+    },
+    {
+      "epoch": 7.032579650565262,
+      "grad_norm": 0.05112696439027786,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 68427
+    },
+    {
+      "epoch": 7.0328879753340185,
+      "grad_norm": 0.03850901499390602,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 68430
+    },
+    {
+      "epoch": 7.033196300102775,
+      "grad_norm": 0.05724974349141121,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 68433
+    },
+    {
+      "epoch": 7.033504624871531,
+      "grad_norm": 0.056213781237602234,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 68436
+    },
+    {
+      "epoch": 7.0338129496402875,
+      "grad_norm": 0.03625565767288208,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 68439
+    },
+    {
+      "epoch": 7.034121274409044,
+      "grad_norm": 0.040962837636470795,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 68442
+    },
+    {
+      "epoch": 7.034429599177801,
+      "grad_norm": 0.0518268421292305,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 68445
+    },
+    {
+      "epoch": 7.034737923946557,
+      "grad_norm": 0.07236559689044952,
+      "learning_rate": 0.01,
+      "loss": 2.0084,
+      "step": 68448
+    },
+    {
+      "epoch": 7.035046248715314,
+      "grad_norm": 0.05604173243045807,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 68451
+    },
+    {
+      "epoch": 7.03535457348407,
+      "grad_norm": 0.0653105154633522,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 68454
+    },
+    {
+      "epoch": 7.035662898252826,
+      "grad_norm": 0.1655261069536209,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 68457
+    },
+    {
+      "epoch": 7.0359712230215825,
+      "grad_norm": 0.04352293908596039,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 68460
+    },
+    {
+      "epoch": 7.036279547790339,
+      "grad_norm": 0.04556136578321457,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 68463
+    },
+    {
+      "epoch": 7.036587872559096,
+      "grad_norm": 0.10971470177173615,
+      "learning_rate": 0.01,
+      "loss": 2.008,
+      "step": 68466
+    },
+    {
+      "epoch": 7.036896197327852,
+      "grad_norm": 0.05627235397696495,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 68469
+    },
+    {
+      "epoch": 7.037204522096609,
+      "grad_norm": 0.03856800124049187,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 68472
+    },
+    {
+      "epoch": 7.037512846865365,
+      "grad_norm": 0.032653167843818665,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 68475
+    },
+    {
+      "epoch": 7.037821171634121,
+      "grad_norm": 0.03872492164373398,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 68478
+    },
+    {
+      "epoch": 7.0381294964028775,
+      "grad_norm": 0.08332209289073944,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 68481
+    },
+    {
+      "epoch": 7.038437821171634,
+      "grad_norm": 0.11315266042947769,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 68484
+    },
+    {
+      "epoch": 7.03874614594039,
+      "grad_norm": 0.058779433369636536,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 68487
+    },
+    {
+      "epoch": 7.039054470709147,
+      "grad_norm": 0.05338497459888458,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 68490
+    },
+    {
+      "epoch": 7.039362795477904,
+      "grad_norm": 0.05961502715945244,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 68493
+    },
+    {
+      "epoch": 7.03967112024666,
+      "grad_norm": 0.10511081665754318,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 68496
+    },
+    {
+      "epoch": 7.039979445015416,
+      "grad_norm": 0.060618769377470016,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 68499
+    },
+    {
+      "epoch": 7.040287769784173,
+      "grad_norm": 0.05174575001001358,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 68502
+    },
+    {
+      "epoch": 7.040596094552929,
+      "grad_norm": 0.038627609610557556,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 68505
+    },
+    {
+      "epoch": 7.040904419321685,
+      "grad_norm": 0.04214971512556076,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 68508
+    },
+    {
+      "epoch": 7.0412127440904415,
+      "grad_norm": 0.08534090965986252,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 68511
+    },
+    {
+      "epoch": 7.041521068859199,
+      "grad_norm": 0.047119926661252975,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 68514
+    },
+    {
+      "epoch": 7.041829393627955,
+      "grad_norm": 0.05264575034379959,
+      "learning_rate": 0.01,
+      "loss": 2.003,
+      "step": 68517
+    },
+    {
+      "epoch": 7.042137718396711,
+      "grad_norm": 0.06278250366449356,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 68520
+    },
+    {
+      "epoch": 7.042446043165468,
+      "grad_norm": 0.09359404444694519,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 68523
+    },
+    {
+      "epoch": 7.042754367934224,
+      "grad_norm": 0.043020278215408325,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 68526
+    },
+    {
+      "epoch": 7.04306269270298,
+      "grad_norm": 0.04135395586490631,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 68529
+    },
+    {
+      "epoch": 7.0433710174717366,
+      "grad_norm": 0.09166526049375534,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 68532
+    },
+    {
+      "epoch": 7.043679342240494,
+      "grad_norm": 0.0364113412797451,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 68535
+    },
+    {
+      "epoch": 7.04398766700925,
+      "grad_norm": 0.04818427935242653,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 68538
+    },
+    {
+      "epoch": 7.044295991778006,
+      "grad_norm": 0.09592068195343018,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 68541
+    },
+    {
+      "epoch": 7.044604316546763,
+      "grad_norm": 0.07666312903165817,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 68544
+    },
+    {
+      "epoch": 7.044912641315519,
+      "grad_norm": 0.053152818232774734,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 68547
+    },
+    {
+      "epoch": 7.045220966084275,
+      "grad_norm": 0.04496999457478523,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 68550
+    },
+    {
+      "epoch": 7.045529290853032,
+      "grad_norm": 0.07192201167345047,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 68553
+    },
+    {
+      "epoch": 7.045837615621788,
+      "grad_norm": 0.07535843551158905,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 68556
+    },
+    {
+      "epoch": 7.046145940390545,
+      "grad_norm": 0.04308157041668892,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 68559
+    },
+    {
+      "epoch": 7.046454265159301,
+      "grad_norm": 0.15922586619853973,
+      "learning_rate": 0.01,
+      "loss": 2.0141,
+      "step": 68562
+    },
+    {
+      "epoch": 7.046762589928058,
+      "grad_norm": 0.06239364296197891,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 68565
+    },
+    {
+      "epoch": 7.047070914696814,
+      "grad_norm": 0.1485653668642044,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 68568
+    },
+    {
+      "epoch": 7.04737923946557,
+      "grad_norm": 0.07881268113851547,
+      "learning_rate": 0.01,
+      "loss": 2.0164,
+      "step": 68571
+    },
+    {
+      "epoch": 7.047687564234327,
+      "grad_norm": 0.037833794951438904,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 68574
+    },
+    {
+      "epoch": 7.047995889003083,
+      "grad_norm": 0.06409921497106552,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 68577
+    },
+    {
+      "epoch": 7.048304213771839,
+      "grad_norm": 0.041993625462055206,
+      "learning_rate": 0.01,
+      "loss": 2.0,
+      "step": 68580
+    },
+    {
+      "epoch": 7.048612538540596,
+      "grad_norm": 0.05115489289164543,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 68583
+    },
+    {
+      "epoch": 7.048920863309353,
+      "grad_norm": 0.037257734686136246,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 68586
+    },
+    {
+      "epoch": 7.049229188078109,
+      "grad_norm": 0.035398274660110474,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 68589
+    },
+    {
+      "epoch": 7.049537512846865,
+      "grad_norm": 0.05111207067966461,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 68592
+    },
+    {
+      "epoch": 7.049845837615622,
+      "grad_norm": 0.07091052085161209,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 68595
+    },
+    {
+      "epoch": 7.050154162384378,
+      "grad_norm": 0.06888426840305328,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 68598
+    },
+    {
+      "epoch": 7.050462487153134,
+      "grad_norm": 0.10825680941343307,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 68601
+    },
+    {
+      "epoch": 7.0507708119218915,
+      "grad_norm": 0.13681887090206146,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 68604
+    },
+    {
+      "epoch": 7.051079136690648,
+      "grad_norm": 0.09142705798149109,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 68607
+    },
+    {
+      "epoch": 7.051387461459404,
+      "grad_norm": 0.06074422225356102,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 68610
+    },
+    {
+      "epoch": 7.05169578622816,
+      "grad_norm": 0.05737810209393501,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 68613
+    },
+    {
+      "epoch": 7.052004110996917,
+      "grad_norm": 0.04696836322546005,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 68616
+    },
+    {
+      "epoch": 7.052312435765673,
+      "grad_norm": 0.06998541951179504,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 68619
+    },
+    {
+      "epoch": 7.052620760534429,
+      "grad_norm": 0.042207296937704086,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 68622
+    },
+    {
+      "epoch": 7.052929085303186,
+      "grad_norm": 0.044253282248973846,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 68625
+    },
+    {
+      "epoch": 7.053237410071943,
+      "grad_norm": 0.03673321008682251,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 68628
+    },
+    {
+      "epoch": 7.053545734840699,
+      "grad_norm": 0.06506076455116272,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 68631
+    },
+    {
+      "epoch": 7.053854059609455,
+      "grad_norm": 0.08981776982545853,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 68634
+    },
+    {
+      "epoch": 7.054162384378212,
+      "grad_norm": 0.04000147059559822,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 68637
+    },
+    {
+      "epoch": 7.054470709146968,
+      "grad_norm": 0.1273726224899292,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 68640
+    },
+    {
+      "epoch": 7.054779033915724,
+      "grad_norm": 0.16758222877979279,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 68643
+    },
+    {
+      "epoch": 7.055087358684481,
+      "grad_norm": 0.14335897564888,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 68646
+    },
+    {
+      "epoch": 7.055395683453237,
+      "grad_norm": 0.08894363045692444,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 68649
+    },
+    {
+      "epoch": 7.055704008221994,
+      "grad_norm": 0.06796260178089142,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 68652
+    },
+    {
+      "epoch": 7.0560123329907505,
+      "grad_norm": 0.09469815343618393,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 68655
+    },
+    {
+      "epoch": 7.056320657759507,
+      "grad_norm": 0.07564645260572433,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 68658
+    },
+    {
+      "epoch": 7.056628982528263,
+      "grad_norm": 0.03675295040011406,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 68661
+    },
+    {
+      "epoch": 7.056937307297019,
+      "grad_norm": 0.03465329483151436,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 68664
+    },
+    {
+      "epoch": 7.057245632065776,
+      "grad_norm": 0.05159129574894905,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 68667
+    },
+    {
+      "epoch": 7.057553956834532,
+      "grad_norm": 0.10650437325239182,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 68670
+    },
+    {
+      "epoch": 7.057862281603289,
+      "grad_norm": 0.07661871612071991,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 68673
+    },
+    {
+      "epoch": 7.0581706063720455,
+      "grad_norm": 0.03891372308135033,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 68676
+    },
+    {
+      "epoch": 7.058478931140802,
+      "grad_norm": 0.07626038044691086,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 68679
+    },
+    {
+      "epoch": 7.058787255909558,
+      "grad_norm": 0.05628202483057976,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 68682
+    },
+    {
+      "epoch": 7.059095580678314,
+      "grad_norm": 0.05379130318760872,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 68685
+    },
+    {
+      "epoch": 7.059403905447071,
+      "grad_norm": 0.04126124083995819,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 68688
+    },
+    {
+      "epoch": 7.059712230215827,
+      "grad_norm": 0.04331153631210327,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 68691
+    },
+    {
+      "epoch": 7.060020554984583,
+      "grad_norm": 0.0378563366830349,
+      "learning_rate": 0.01,
+      "loss": 2.0034,
+      "step": 68694
+    },
+    {
+      "epoch": 7.060328879753341,
+      "grad_norm": 0.04773402959108353,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 68697
+    },
+    {
+      "epoch": 7.060637204522097,
+      "grad_norm": 0.0653529167175293,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 68700
+    },
+    {
+      "epoch": 7.060945529290853,
+      "grad_norm": 0.08799976110458374,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 68703
+    },
+    {
+      "epoch": 7.0612538540596095,
+      "grad_norm": 0.04789255931973457,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 68706
+    },
+    {
+      "epoch": 7.061562178828366,
+      "grad_norm": 0.06388677656650543,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 68709
+    },
+    {
+      "epoch": 7.061870503597122,
+      "grad_norm": 0.03625022992491722,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 68712
+    },
+    {
+      "epoch": 7.062178828365878,
+      "grad_norm": 0.0345403291285038,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 68715
+    },
+    {
+      "epoch": 7.062487153134635,
+      "grad_norm": 0.03319702669978142,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 68718
+    },
+    {
+      "epoch": 7.062795477903392,
+      "grad_norm": 0.05995936691761017,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 68721
+    },
+    {
+      "epoch": 7.063103802672148,
+      "grad_norm": 0.04327778145670891,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 68724
+    },
+    {
+      "epoch": 7.0634121274409045,
+      "grad_norm": 0.15124131739139557,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 68727
+    },
+    {
+      "epoch": 7.063720452209661,
+      "grad_norm": 0.08264999836683273,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 68730
+    },
+    {
+      "epoch": 7.064028776978417,
+      "grad_norm": 0.0442788302898407,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 68733
+    },
+    {
+      "epoch": 7.064337101747173,
+      "grad_norm": 0.04176313802599907,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 68736
+    },
+    {
+      "epoch": 7.06464542651593,
+      "grad_norm": 0.04693405702710152,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 68739
+    },
+    {
+      "epoch": 7.064953751284687,
+      "grad_norm": 0.0443805493414402,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 68742
+    },
+    {
+      "epoch": 7.065262076053443,
+      "grad_norm": 0.086354561150074,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 68745
+    },
+    {
+      "epoch": 7.0655704008222,
+      "grad_norm": 0.05768636614084244,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 68748
+    },
+    {
+      "epoch": 7.065878725590956,
+      "grad_norm": 0.08465427905321121,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 68751
+    },
+    {
+      "epoch": 7.066187050359712,
+      "grad_norm": 0.08750718086957932,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 68754
+    },
+    {
+      "epoch": 7.0664953751284685,
+      "grad_norm": 0.06704618036746979,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 68757
+    },
+    {
+      "epoch": 7.066803699897225,
+      "grad_norm": 0.12646348774433136,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 68760
+    },
+    {
+      "epoch": 7.067112024665981,
+      "grad_norm": 0.07180266082286835,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 68763
+    },
+    {
+      "epoch": 7.067420349434738,
+      "grad_norm": 0.04616543650627136,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 68766
+    },
+    {
+      "epoch": 7.067728674203495,
+      "grad_norm": 0.044225241988897324,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 68769
+    },
+    {
+      "epoch": 7.068036998972251,
+      "grad_norm": 0.03617358207702637,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 68772
+    },
+    {
+      "epoch": 7.068345323741007,
+      "grad_norm": 0.09863235056400299,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 68775
+    },
+    {
+      "epoch": 7.0686536485097635,
+      "grad_norm": 0.06617226451635361,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 68778
+    },
+    {
+      "epoch": 7.06896197327852,
+      "grad_norm": 0.12394807487726212,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 68781
+    },
+    {
+      "epoch": 7.069270298047276,
+      "grad_norm": 0.060029856860637665,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 68784
+    },
+    {
+      "epoch": 7.069578622816033,
+      "grad_norm": 0.045732829719781876,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 68787
+    },
+    {
+      "epoch": 7.06988694758479,
+      "grad_norm": 0.06220393627882004,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 68790
+    },
+    {
+      "epoch": 7.070195272353546,
+      "grad_norm": 0.04120052605867386,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 68793
+    },
+    {
+      "epoch": 7.070503597122302,
+      "grad_norm": 0.049835655838251114,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 68796
+    },
+    {
+      "epoch": 7.070811921891059,
+      "grad_norm": 0.03683771193027496,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 68799
+    },
+    {
+      "epoch": 7.071120246659815,
+      "grad_norm": 0.03647764027118683,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 68802
+    },
+    {
+      "epoch": 7.071428571428571,
+      "grad_norm": 0.16000273823738098,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 68805
+    },
+    {
+      "epoch": 7.0717368961973275,
+      "grad_norm": 0.03855353966355324,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 68808
+    },
+    {
+      "epoch": 7.072045220966085,
+      "grad_norm": 0.040406130254268646,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 68811
+    },
+    {
+      "epoch": 7.072353545734841,
+      "grad_norm": 0.03346218913793564,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 68814
+    },
+    {
+      "epoch": 7.072661870503597,
+      "grad_norm": 0.11391355842351913,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 68817
+    },
+    {
+      "epoch": 7.072970195272354,
+      "grad_norm": 0.060539521276950836,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 68820
+    },
+    {
+      "epoch": 7.07327852004111,
+      "grad_norm": 0.1784716248512268,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 68823
+    },
+    {
+      "epoch": 7.073586844809866,
+      "grad_norm": 0.04299479350447655,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 68826
+    },
+    {
+      "epoch": 7.0738951695786225,
+      "grad_norm": 0.05543602257966995,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 68829
+    },
+    {
+      "epoch": 7.074203494347379,
+      "grad_norm": 0.039931852370500565,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 68832
+    },
+    {
+      "epoch": 7.074511819116136,
+      "grad_norm": 0.044154778122901917,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 68835
+    },
+    {
+      "epoch": 7.074820143884892,
+      "grad_norm": 0.037736304104328156,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 68838
+    },
+    {
+      "epoch": 7.075128468653649,
+      "grad_norm": 0.06381729245185852,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 68841
+    },
+    {
+      "epoch": 7.075436793422405,
+      "grad_norm": 0.03960365802049637,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 68844
+    },
+    {
+      "epoch": 7.075745118191161,
+      "grad_norm": 0.11026263236999512,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 68847
+    },
+    {
+      "epoch": 7.076053442959918,
+      "grad_norm": 0.09295713901519775,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 68850
+    },
+    {
+      "epoch": 7.076361767728674,
+      "grad_norm": 0.07637894153594971,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 68853
+    },
+    {
+      "epoch": 7.076670092497431,
+      "grad_norm": 0.062266815453767776,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 68856
+    },
+    {
+      "epoch": 7.076978417266187,
+      "grad_norm": 0.09450382739305496,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 68859
+    },
+    {
+      "epoch": 7.077286742034944,
+      "grad_norm": 0.07935396581888199,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 68862
+    },
+    {
+      "epoch": 7.0775950668037,
+      "grad_norm": 0.08313512057065964,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 68865
+    },
+    {
+      "epoch": 7.077903391572456,
+      "grad_norm": 0.044498976320028305,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 68868
+    },
+    {
+      "epoch": 7.078211716341213,
+      "grad_norm": 0.054316215217113495,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 68871
+    },
+    {
+      "epoch": 7.078520041109969,
+      "grad_norm": 0.07995834946632385,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 68874
+    },
+    {
+      "epoch": 7.078828365878725,
+      "grad_norm": 0.04581683129072189,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 68877
+    },
+    {
+      "epoch": 7.079136690647482,
+      "grad_norm": 0.03488059714436531,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 68880
+    },
+    {
+      "epoch": 7.079445015416239,
+      "grad_norm": 0.0278830174356699,
+      "learning_rate": 0.01,
+      "loss": 2.007,
+      "step": 68883
+    },
+    {
+      "epoch": 7.079753340184995,
+      "grad_norm": 0.03560703247785568,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 68886
+    },
+    {
+      "epoch": 7.080061664953751,
+      "grad_norm": 0.1039506047964096,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 68889
+    },
+    {
+      "epoch": 7.080369989722508,
+      "grad_norm": 0.09519243985414505,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 68892
+    },
+    {
+      "epoch": 7.080678314491264,
+      "grad_norm": 0.039108019322156906,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 68895
+    },
+    {
+      "epoch": 7.08098663926002,
+      "grad_norm": 0.08773428946733475,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 68898
+    },
+    {
+      "epoch": 7.081294964028777,
+      "grad_norm": 0.03855564817786217,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 68901
+    },
+    {
+      "epoch": 7.081603288797534,
+      "grad_norm": 0.06529016047716141,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 68904
+    },
+    {
+      "epoch": 7.08191161356629,
+      "grad_norm": 0.07360291481018066,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 68907
+    },
+    {
+      "epoch": 7.082219938335046,
+      "grad_norm": 0.10342194885015488,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 68910
+    },
+    {
+      "epoch": 7.082528263103803,
+      "grad_norm": 0.04935605451464653,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 68913
+    },
+    {
+      "epoch": 7.082836587872559,
+      "grad_norm": 0.08817791938781738,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 68916
+    },
+    {
+      "epoch": 7.083144912641315,
+      "grad_norm": 0.052056267857551575,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 68919
+    },
+    {
+      "epoch": 7.083453237410072,
+      "grad_norm": 0.04656489938497543,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 68922
+    },
+    {
+      "epoch": 7.083761562178829,
+      "grad_norm": 0.06415385752916336,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 68925
+    },
+    {
+      "epoch": 7.084069886947585,
+      "grad_norm": 0.07505769282579422,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 68928
+    },
+    {
+      "epoch": 7.084378211716341,
+      "grad_norm": 0.12667891383171082,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 68931
+    },
+    {
+      "epoch": 7.084686536485098,
+      "grad_norm": 0.1055910661816597,
+      "learning_rate": 0.01,
+      "loss": 2.0074,
+      "step": 68934
+    },
+    {
+      "epoch": 7.084994861253854,
+      "grad_norm": 0.06105726957321167,
+      "learning_rate": 0.01,
+      "loss": 1.9979,
+      "step": 68937
+    },
+    {
+      "epoch": 7.08530318602261,
+      "grad_norm": 0.04314691945910454,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 68940
+    },
+    {
+      "epoch": 7.085611510791367,
+      "grad_norm": 0.039434608072042465,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 68943
+    },
+    {
+      "epoch": 7.085919835560123,
+      "grad_norm": 0.04623565822839737,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 68946
+    },
+    {
+      "epoch": 7.08622816032888,
+      "grad_norm": 0.03403662517666817,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 68949
+    },
+    {
+      "epoch": 7.0865364850976365,
+      "grad_norm": 0.044097188860177994,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 68952
+    },
+    {
+      "epoch": 7.086844809866393,
+      "grad_norm": 0.1629336178302765,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 68955
+    },
+    {
+      "epoch": 7.087153134635149,
+      "grad_norm": 0.061776354908943176,
+      "learning_rate": 0.01,
+      "loss": 2.0016,
+      "step": 68958
+    },
+    {
+      "epoch": 7.087461459403905,
+      "grad_norm": 0.053984384983778,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 68961
+    },
+    {
+      "epoch": 7.087769784172662,
+      "grad_norm": 0.05198853835463524,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 68964
+    },
+    {
+      "epoch": 7.088078108941418,
+      "grad_norm": 0.030461978167295456,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 68967
+    },
+    {
+      "epoch": 7.088386433710174,
+      "grad_norm": 0.12187229096889496,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 68970
+    },
+    {
+      "epoch": 7.0886947584789315,
+      "grad_norm": 0.046721212565898895,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 68973
+    },
+    {
+      "epoch": 7.089003083247688,
+      "grad_norm": 0.113240085542202,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 68976
+    },
+    {
+      "epoch": 7.089311408016444,
+      "grad_norm": 0.17302365601062775,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 68979
+    },
+    {
+      "epoch": 7.0896197327852,
+      "grad_norm": 0.1341373175382614,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 68982
+    },
+    {
+      "epoch": 7.089928057553957,
+      "grad_norm": 0.04806041717529297,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 68985
+    },
+    {
+      "epoch": 7.090236382322713,
+      "grad_norm": 0.053426358848810196,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 68988
+    },
+    {
+      "epoch": 7.090544707091469,
+      "grad_norm": 0.05874931067228317,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 68991
+    },
+    {
+      "epoch": 7.0908530318602265,
+      "grad_norm": 0.04069295898079872,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 68994
+    },
+    {
+      "epoch": 7.091161356628983,
+      "grad_norm": 0.04333820566534996,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 68997
+    },
+    {
+      "epoch": 7.091469681397739,
+      "grad_norm": 0.08710654824972153,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 69000
+    },
+    {
+      "epoch": 7.0917780061664955,
+      "grad_norm": 0.05234698951244354,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 69003
+    },
+    {
+      "epoch": 7.092086330935252,
+      "grad_norm": 0.10434626042842865,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 69006
+    },
+    {
+      "epoch": 7.092394655704008,
+      "grad_norm": 0.10928107053041458,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 69009
+    },
+    {
+      "epoch": 7.092702980472764,
+      "grad_norm": 0.04639718681573868,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 69012
+    },
+    {
+      "epoch": 7.093011305241521,
+      "grad_norm": 0.04324621334671974,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 69015
+    },
+    {
+      "epoch": 7.093319630010278,
+      "grad_norm": 0.04631352052092552,
+      "learning_rate": 0.01,
+      "loss": 2.0098,
+      "step": 69018
+    },
+    {
+      "epoch": 7.093627954779034,
+      "grad_norm": 0.09556543827056885,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 69021
+    },
+    {
+      "epoch": 7.0939362795477905,
+      "grad_norm": 0.15257610380649567,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 69024
+    },
+    {
+      "epoch": 7.094244604316547,
+      "grad_norm": 0.060181524604558945,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 69027
+    },
+    {
+      "epoch": 7.094552929085303,
+      "grad_norm": 0.04848707467317581,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 69030
+    },
+    {
+      "epoch": 7.094861253854059,
+      "grad_norm": 0.05945300683379173,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 69033
+    },
+    {
+      "epoch": 7.095169578622816,
+      "grad_norm": 0.08706066012382507,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 69036
+    },
+    {
+      "epoch": 7.095477903391572,
+      "grad_norm": 0.12813836336135864,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 69039
+    },
+    {
+      "epoch": 7.095786228160329,
+      "grad_norm": 0.08248843252658844,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 69042
+    },
+    {
+      "epoch": 7.0960945529290855,
+      "grad_norm": 0.05092019960284233,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 69045
+    },
+    {
+      "epoch": 7.096402877697842,
+      "grad_norm": 0.04098663106560707,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 69048
+    },
+    {
+      "epoch": 7.096711202466598,
+      "grad_norm": 0.04048779979348183,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 69051
+    },
+    {
+      "epoch": 7.0970195272353545,
+      "grad_norm": 0.06100444123148918,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 69054
+    },
+    {
+      "epoch": 7.097327852004111,
+      "grad_norm": 0.047660503536462784,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 69057
+    },
+    {
+      "epoch": 7.097636176772867,
+      "grad_norm": 0.04160270839929581,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 69060
+    },
+    {
+      "epoch": 7.097944501541624,
+      "grad_norm": 0.03557828441262245,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 69063
+    },
+    {
+      "epoch": 7.098252826310381,
+      "grad_norm": 0.06803721934556961,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 69066
+    },
+    {
+      "epoch": 7.098561151079137,
+      "grad_norm": 0.10636363923549652,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 69069
+    },
+    {
+      "epoch": 7.098869475847893,
+      "grad_norm": 0.0738653913140297,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 69072
+    },
+    {
+      "epoch": 7.0991778006166495,
+      "grad_norm": 0.0479535274207592,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 69075
+    },
+    {
+      "epoch": 7.099486125385406,
+      "grad_norm": 0.06804942339658737,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 69078
+    },
+    {
+      "epoch": 7.099794450154162,
+      "grad_norm": 0.06756333261728287,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 69081
+    },
+    {
+      "epoch": 7.100102774922918,
+      "grad_norm": 0.047195300459861755,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 69084
+    },
+    {
+      "epoch": 7.100411099691676,
+      "grad_norm": 0.03911405801773071,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 69087
+    },
+    {
+      "epoch": 7.100719424460432,
+      "grad_norm": 0.08936751633882523,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 69090
+    },
+    {
+      "epoch": 7.101027749229188,
+      "grad_norm": 0.040167100727558136,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 69093
+    },
+    {
+      "epoch": 7.1013360739979445,
+      "grad_norm": 0.03756394237279892,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 69096
+    },
+    {
+      "epoch": 7.101644398766701,
+      "grad_norm": 0.03320080041885376,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 69099
+    },
+    {
+      "epoch": 7.101952723535457,
+      "grad_norm": 0.07805169373750687,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 69102
+    },
+    {
+      "epoch": 7.1022610483042135,
+      "grad_norm": 0.06461729109287262,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 69105
+    },
+    {
+      "epoch": 7.10256937307297,
+      "grad_norm": 0.03744588792324066,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 69108
+    },
+    {
+      "epoch": 7.102877697841727,
+      "grad_norm": 0.03426771238446236,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 69111
+    },
+    {
+      "epoch": 7.103186022610483,
+      "grad_norm": 0.055179059505462646,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 69114
+    },
+    {
+      "epoch": 7.10349434737924,
+      "grad_norm": 0.0941012054681778,
+      "learning_rate": 0.01,
+      "loss": 1.9946,
+      "step": 69117
+    },
+    {
+      "epoch": 7.103802672147996,
+      "grad_norm": 0.04403389245271683,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 69120
+    },
+    {
+      "epoch": 7.104110996916752,
+      "grad_norm": 0.06421434134244919,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 69123
+    },
+    {
+      "epoch": 7.1044193216855085,
+      "grad_norm": 0.06467723846435547,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 69126
+    },
+    {
+      "epoch": 7.104727646454265,
+      "grad_norm": 0.05050961673259735,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 69129
+    },
+    {
+      "epoch": 7.105035971223022,
+      "grad_norm": 0.05761819705367088,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 69132
+    },
+    {
+      "epoch": 7.105344295991778,
+      "grad_norm": 0.04577011242508888,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 69135
+    },
+    {
+      "epoch": 7.105652620760535,
+      "grad_norm": 0.06268461793661118,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 69138
+    },
+    {
+      "epoch": 7.105960945529291,
+      "grad_norm": 0.039437875151634216,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 69141
+    },
+    {
+      "epoch": 7.106269270298047,
+      "grad_norm": 0.09671974182128906,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 69144
+    },
+    {
+      "epoch": 7.1065775950668035,
+      "grad_norm": 0.08435185253620148,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 69147
+    },
+    {
+      "epoch": 7.10688591983556,
+      "grad_norm": 0.11098583042621613,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 69150
+    },
+    {
+      "epoch": 7.107194244604316,
+      "grad_norm": 0.04098542779684067,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 69153
+    },
+    {
+      "epoch": 7.107502569373073,
+      "grad_norm": 0.0462697297334671,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 69156
+    },
+    {
+      "epoch": 7.10781089414183,
+      "grad_norm": 0.03672792762517929,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 69159
+    },
+    {
+      "epoch": 7.108119218910586,
+      "grad_norm": 0.06978480517864227,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 69162
+    },
+    {
+      "epoch": 7.108427543679342,
+      "grad_norm": 0.04515951871871948,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 69165
+    },
+    {
+      "epoch": 7.108735868448099,
+      "grad_norm": 0.03993092477321625,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 69168
+    },
+    {
+      "epoch": 7.109044193216855,
+      "grad_norm": 0.0673045888543129,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 69171
+    },
+    {
+      "epoch": 7.109352517985611,
+      "grad_norm": 0.05873231217265129,
+      "learning_rate": 0.01,
+      "loss": 2.0119,
+      "step": 69174
+    },
+    {
+      "epoch": 7.1096608427543675,
+      "grad_norm": 0.053032733500003815,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 69177
+    },
+    {
+      "epoch": 7.109969167523125,
+      "grad_norm": 0.0582476407289505,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 69180
+    },
+    {
+      "epoch": 7.110277492291881,
+      "grad_norm": 0.08659794926643372,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 69183
+    },
+    {
+      "epoch": 7.110585817060637,
+      "grad_norm": 0.0926288440823555,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 69186
+    },
+    {
+      "epoch": 7.110894141829394,
+      "grad_norm": 0.07831533998250961,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 69189
+    },
+    {
+      "epoch": 7.11120246659815,
+      "grad_norm": 0.04761062562465668,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 69192
+    },
+    {
+      "epoch": 7.111510791366906,
+      "grad_norm": 0.055868249386548996,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 69195
+    },
+    {
+      "epoch": 7.1118191161356625,
+      "grad_norm": 0.04533650353550911,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 69198
+    },
+    {
+      "epoch": 7.11212744090442,
+      "grad_norm": 0.10463204234838486,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 69201
+    },
+    {
+      "epoch": 7.112435765673176,
+      "grad_norm": 0.07214853912591934,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 69204
+    },
+    {
+      "epoch": 7.112744090441932,
+      "grad_norm": 0.059007078409194946,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 69207
+    },
+    {
+      "epoch": 7.113052415210689,
+      "grad_norm": 0.054935675114393234,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 69210
+    },
+    {
+      "epoch": 7.113360739979445,
+      "grad_norm": 0.05916939303278923,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 69213
+    },
+    {
+      "epoch": 7.113669064748201,
+      "grad_norm": 0.08450639992952347,
+      "learning_rate": 0.01,
+      "loss": 2.0054,
+      "step": 69216
+    },
+    {
+      "epoch": 7.113977389516958,
+      "grad_norm": 0.06881245970726013,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 69219
+    },
+    {
+      "epoch": 7.114285714285714,
+      "grad_norm": 0.07292987406253815,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 69222
+    },
+    {
+      "epoch": 7.114594039054471,
+      "grad_norm": 0.0532345175743103,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 69225
+    },
+    {
+      "epoch": 7.114902363823227,
+      "grad_norm": 0.051116786897182465,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 69228
+    },
+    {
+      "epoch": 7.115210688591984,
+      "grad_norm": 0.0622846744954586,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 69231
+    },
+    {
+      "epoch": 7.11551901336074,
+      "grad_norm": 0.052440278232097626,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 69234
+    },
+    {
+      "epoch": 7.115827338129496,
+      "grad_norm": 0.03851577267050743,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 69237
+    },
+    {
+      "epoch": 7.116135662898253,
+      "grad_norm": 0.10518687963485718,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 69240
+    },
+    {
+      "epoch": 7.116443987667009,
+      "grad_norm": 0.08948950469493866,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 69243
+    },
+    {
+      "epoch": 7.116752312435766,
+      "grad_norm": 0.13956782221794128,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 69246
+    },
+    {
+      "epoch": 7.117060637204522,
+      "grad_norm": 0.06303557753562927,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 69249
+    },
+    {
+      "epoch": 7.117368961973279,
+      "grad_norm": 0.056266095489263535,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 69252
+    },
+    {
+      "epoch": 7.117677286742035,
+      "grad_norm": 0.07958390563726425,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 69255
+    },
+    {
+      "epoch": 7.117985611510791,
+      "grad_norm": 0.0753171443939209,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 69258
+    },
+    {
+      "epoch": 7.118293936279548,
+      "grad_norm": 0.05444815382361412,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 69261
+    },
+    {
+      "epoch": 7.118602261048304,
+      "grad_norm": 0.061939787119627,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 69264
+    },
+    {
+      "epoch": 7.11891058581706,
+      "grad_norm": 0.118798166513443,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 69267
+    },
+    {
+      "epoch": 7.1192189105858175,
+      "grad_norm": 0.048886317759752274,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 69270
+    },
+    {
+      "epoch": 7.119527235354574,
+      "grad_norm": 0.05422835797071457,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 69273
+    },
+    {
+      "epoch": 7.11983556012333,
+      "grad_norm": 0.05022811144590378,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 69276
+    },
+    {
+      "epoch": 7.120143884892086,
+      "grad_norm": 0.07943117618560791,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 69279
+    },
+    {
+      "epoch": 7.120452209660843,
+      "grad_norm": 0.03822775557637215,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 69282
+    },
+    {
+      "epoch": 7.120760534429599,
+      "grad_norm": 0.06427093595266342,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 69285
+    },
+    {
+      "epoch": 7.121068859198355,
+      "grad_norm": 0.046098969876766205,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 69288
+    },
+    {
+      "epoch": 7.121377183967112,
+      "grad_norm": 0.08187385648488998,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 69291
+    },
+    {
+      "epoch": 7.121685508735869,
+      "grad_norm": 0.0525503009557724,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 69294
+    },
+    {
+      "epoch": 7.121993833504625,
+      "grad_norm": 0.03687702864408493,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 69297
+    },
+    {
+      "epoch": 7.122302158273381,
+      "grad_norm": 0.06423792988061905,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 69300
+    },
+    {
+      "epoch": 7.122610483042138,
+      "grad_norm": 0.12554574012756348,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 69303
+    },
+    {
+      "epoch": 7.122918807810894,
+      "grad_norm": 0.06923548877239227,
+      "learning_rate": 0.01,
+      "loss": 2.0027,
+      "step": 69306
+    },
+    {
+      "epoch": 7.12322713257965,
+      "grad_norm": 0.0461004301905632,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 69309
+    },
+    {
+      "epoch": 7.123535457348407,
+      "grad_norm": 0.04713965207338333,
+      "learning_rate": 0.01,
+      "loss": 2.0065,
+      "step": 69312
+    },
+    {
+      "epoch": 7.123843782117164,
+      "grad_norm": 0.043988946825265884,
+      "learning_rate": 0.01,
+      "loss": 2.0037,
+      "step": 69315
+    },
+    {
+      "epoch": 7.12415210688592,
+      "grad_norm": 0.03573235124349594,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 69318
+    },
+    {
+      "epoch": 7.1244604316546765,
+      "grad_norm": 0.03654714301228523,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 69321
+    },
+    {
+      "epoch": 7.124768756423433,
+      "grad_norm": 0.11765187233686447,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 69324
+    },
+    {
+      "epoch": 7.125077081192189,
+      "grad_norm": 0.06880181282758713,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 69327
+    },
+    {
+      "epoch": 7.125385405960945,
+      "grad_norm": 0.056520819664001465,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 69330
+    },
+    {
+      "epoch": 7.125693730729702,
+      "grad_norm": 0.04125680774450302,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 69333
+    },
+    {
+      "epoch": 7.126002055498458,
+      "grad_norm": 0.04342076554894447,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 69336
+    },
+    {
+      "epoch": 7.126310380267215,
+      "grad_norm": 0.03851790353655815,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 69339
+    },
+    {
+      "epoch": 7.1266187050359715,
+      "grad_norm": 0.03577910736203194,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 69342
+    },
+    {
+      "epoch": 7.126927029804728,
+      "grad_norm": 0.03813933581113815,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 69345
+    },
+    {
+      "epoch": 7.127235354573484,
+      "grad_norm": 0.05139464884996414,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 69348
+    },
+    {
+      "epoch": 7.12754367934224,
+      "grad_norm": 0.1345084309577942,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 69351
+    },
+    {
+      "epoch": 7.127852004110997,
+      "grad_norm": 0.04550566524267197,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 69354
+    },
+    {
+      "epoch": 7.128160328879753,
+      "grad_norm": 0.056097351014614105,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 69357
+    },
+    {
+      "epoch": 7.128468653648509,
+      "grad_norm": 0.03391636162996292,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 69360
+    },
+    {
+      "epoch": 7.1287769784172665,
+      "grad_norm": 0.04839693009853363,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 69363
+    },
+    {
+      "epoch": 7.129085303186023,
+      "grad_norm": 0.053586047142744064,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 69366
+    },
+    {
+      "epoch": 7.129393627954779,
+      "grad_norm": 0.07650946080684662,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 69369
+    },
+    {
+      "epoch": 7.1297019527235355,
+      "grad_norm": 0.12655211985111237,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 69372
+    },
+    {
+      "epoch": 7.130010277492292,
+      "grad_norm": 0.04835767298936844,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 69375
+    },
+    {
+      "epoch": 7.130318602261048,
+      "grad_norm": 0.10156697034835815,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 69378
+    },
+    {
+      "epoch": 7.130626927029804,
+      "grad_norm": 0.04780452698469162,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 69381
+    },
+    {
+      "epoch": 7.130935251798562,
+      "grad_norm": 0.06014325097203255,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 69384
+    },
+    {
+      "epoch": 7.131243576567318,
+      "grad_norm": 0.0386577844619751,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 69387
+    },
+    {
+      "epoch": 7.131551901336074,
+      "grad_norm": 0.06388622522354126,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 69390
+    },
+    {
+      "epoch": 7.1318602261048305,
+      "grad_norm": 0.08713452517986298,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 69393
+    },
+    {
+      "epoch": 7.132168550873587,
+      "grad_norm": 0.12167992442846298,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 69396
+    },
+    {
+      "epoch": 7.132476875642343,
+      "grad_norm": 0.12622815370559692,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 69399
+    },
+    {
+      "epoch": 7.132785200411099,
+      "grad_norm": 0.052541423588991165,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 69402
+    },
+    {
+      "epoch": 7.133093525179856,
+      "grad_norm": 0.035066355019807816,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 69405
+    },
+    {
+      "epoch": 7.133401849948613,
+      "grad_norm": 0.047106850892305374,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 69408
+    },
+    {
+      "epoch": 7.133710174717369,
+      "grad_norm": 0.12262306362390518,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 69411
+    },
+    {
+      "epoch": 7.1340184994861255,
+      "grad_norm": 0.0389021597802639,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 69414
+    },
+    {
+      "epoch": 7.134326824254882,
+      "grad_norm": 0.03964044153690338,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 69417
+    },
+    {
+      "epoch": 7.134635149023638,
+      "grad_norm": 0.03052535466849804,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 69420
+    },
+    {
+      "epoch": 7.1349434737923945,
+      "grad_norm": 0.036022480577230453,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 69423
+    },
+    {
+      "epoch": 7.135251798561151,
+      "grad_norm": 0.055815573781728745,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 69426
+    },
+    {
+      "epoch": 7.135560123329907,
+      "grad_norm": 0.05197738856077194,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 69429
+    },
+    {
+      "epoch": 7.135868448098664,
+      "grad_norm": 0.1406523585319519,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 69432
+    },
+    {
+      "epoch": 7.136176772867421,
+      "grad_norm": 0.056158799678087234,
+      "learning_rate": 0.01,
+      "loss": 1.9983,
+      "step": 69435
+    },
+    {
+      "epoch": 7.136485097636177,
+      "grad_norm": 0.10209375619888306,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 69438
+    },
+    {
+      "epoch": 7.136793422404933,
+      "grad_norm": 0.12295509874820709,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 69441
+    },
+    {
+      "epoch": 7.1371017471736895,
+      "grad_norm": 0.09482801705598831,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 69444
+    },
+    {
+      "epoch": 7.137410071942446,
+      "grad_norm": 0.043474819511175156,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 69447
+    },
+    {
+      "epoch": 7.137718396711202,
+      "grad_norm": 0.03609083592891693,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 69450
+    },
+    {
+      "epoch": 7.138026721479959,
+      "grad_norm": 0.0358879528939724,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 69453
+    },
+    {
+      "epoch": 7.138335046248716,
+      "grad_norm": 0.04629027098417282,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 69456
+    },
+    {
+      "epoch": 7.138643371017472,
+      "grad_norm": 0.04923400655388832,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 69459
+    },
+    {
+      "epoch": 7.138951695786228,
+      "grad_norm": 0.05828343704342842,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 69462
+    },
+    {
+      "epoch": 7.1392600205549845,
+      "grad_norm": 0.059361234307289124,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 69465
+    },
+    {
+      "epoch": 7.139568345323741,
+      "grad_norm": 0.04384483024477959,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 69468
+    },
+    {
+      "epoch": 7.139876670092497,
+      "grad_norm": 0.03761192038655281,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 69471
+    },
+    {
+      "epoch": 7.1401849948612535,
+      "grad_norm": 0.14516139030456543,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 69474
+    },
+    {
+      "epoch": 7.140493319630011,
+      "grad_norm": 0.12505464255809784,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 69477
+    },
+    {
+      "epoch": 7.140801644398767,
+      "grad_norm": 0.06923521310091019,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 69480
+    },
+    {
+      "epoch": 7.141109969167523,
+      "grad_norm": 0.03950193524360657,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 69483
+    },
+    {
+      "epoch": 7.14141829393628,
+      "grad_norm": 0.05078789219260216,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 69486
+    },
+    {
+      "epoch": 7.141726618705036,
+      "grad_norm": 0.03560269996523857,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 69489
+    },
+    {
+      "epoch": 7.142034943473792,
+      "grad_norm": 0.050856418907642365,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 69492
+    },
+    {
+      "epoch": 7.1423432682425485,
+      "grad_norm": 0.06913231313228607,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 69495
+    },
+    {
+      "epoch": 7.142651593011305,
+      "grad_norm": 0.11764489114284515,
+      "learning_rate": 0.01,
+      "loss": 2.0103,
+      "step": 69498
+    },
+    {
+      "epoch": 7.142959917780062,
+      "grad_norm": 0.062085945159196854,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 69501
+    },
+    {
+      "epoch": 7.143268242548818,
+      "grad_norm": 0.029841940850019455,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 69504
+    },
+    {
+      "epoch": 7.143576567317575,
+      "grad_norm": 0.05607277899980545,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 69507
+    },
+    {
+      "epoch": 7.143884892086331,
+      "grad_norm": 0.03662040829658508,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 69510
+    },
+    {
+      "epoch": 7.144193216855087,
+      "grad_norm": 0.03297117352485657,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 69513
+    },
+    {
+      "epoch": 7.1445015416238435,
+      "grad_norm": 0.044707778841257095,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 69516
+    },
+    {
+      "epoch": 7.1448098663926,
+      "grad_norm": 0.10864292085170746,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 69519
+    },
+    {
+      "epoch": 7.145118191161357,
+      "grad_norm": 0.07812972366809845,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 69522
+    },
+    {
+      "epoch": 7.145426515930113,
+      "grad_norm": 0.03777284547686577,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 69525
+    },
+    {
+      "epoch": 7.14573484069887,
+      "grad_norm": 0.04257459193468094,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 69528
+    },
+    {
+      "epoch": 7.146043165467626,
+      "grad_norm": 0.04293552413582802,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 69531
+    },
+    {
+      "epoch": 7.146351490236382,
+      "grad_norm": 0.06645122170448303,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 69534
+    },
+    {
+      "epoch": 7.146659815005139,
+      "grad_norm": 0.04492383822798729,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 69537
+    },
+    {
+      "epoch": 7.146968139773895,
+      "grad_norm": 0.03826465830206871,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 69540
+    },
+    {
+      "epoch": 7.147276464542651,
+      "grad_norm": 0.09625104069709778,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 69543
+    },
+    {
+      "epoch": 7.147584789311408,
+      "grad_norm": 0.08541668951511383,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 69546
+    },
+    {
+      "epoch": 7.147893114080165,
+      "grad_norm": 0.05660329386591911,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 69549
+    },
+    {
+      "epoch": 7.148201438848921,
+      "grad_norm": 0.04568130522966385,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 69552
+    },
+    {
+      "epoch": 7.148509763617677,
+      "grad_norm": 0.08119482547044754,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 69555
+    },
+    {
+      "epoch": 7.148818088386434,
+      "grad_norm": 0.05496470630168915,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 69558
+    },
+    {
+      "epoch": 7.14912641315519,
+      "grad_norm": 0.09700389951467514,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 69561
+    },
+    {
+      "epoch": 7.149434737923946,
+      "grad_norm": 0.0743723064661026,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 69564
+    },
+    {
+      "epoch": 7.149743062692703,
+      "grad_norm": 0.08218356221914291,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 69567
+    },
+    {
+      "epoch": 7.15005138746146,
+      "grad_norm": 0.04347396269440651,
+      "learning_rate": 0.01,
+      "loss": 2.0071,
+      "step": 69570
+    },
+    {
+      "epoch": 7.150359712230216,
+      "grad_norm": 0.03787241131067276,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 69573
+    },
+    {
+      "epoch": 7.150668036998972,
+      "grad_norm": 0.05633244663476944,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 69576
+    },
+    {
+      "epoch": 7.150976361767729,
+      "grad_norm": 0.09029576182365417,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 69579
+    },
+    {
+      "epoch": 7.151284686536485,
+      "grad_norm": 0.12735773622989655,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 69582
+    },
+    {
+      "epoch": 7.151593011305241,
+      "grad_norm": 0.06273090839385986,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 69585
+    },
+    {
+      "epoch": 7.151901336073998,
+      "grad_norm": 0.052082959562540054,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 69588
+    },
+    {
+      "epoch": 7.152209660842755,
+      "grad_norm": 0.08231306821107864,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 69591
+    },
+    {
+      "epoch": 7.152517985611511,
+      "grad_norm": 0.07055219262838364,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 69594
+    },
+    {
+      "epoch": 7.152826310380267,
+      "grad_norm": 0.07429744303226471,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 69597
+    },
+    {
+      "epoch": 7.153134635149024,
+      "grad_norm": 0.11131955683231354,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 69600
+    },
+    {
+      "epoch": 7.15344295991778,
+      "grad_norm": 0.1321670562028885,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 69603
+    },
+    {
+      "epoch": 7.153751284686536,
+      "grad_norm": 0.07883578538894653,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 69606
+    },
+    {
+      "epoch": 7.154059609455293,
+      "grad_norm": 0.06068054959177971,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 69609
+    },
+    {
+      "epoch": 7.154367934224049,
+      "grad_norm": 0.04163357615470886,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 69612
+    },
+    {
+      "epoch": 7.154676258992806,
+      "grad_norm": 0.04159896448254585,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 69615
+    },
+    {
+      "epoch": 7.154984583761562,
+      "grad_norm": 0.053873077034950256,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 69618
+    },
+    {
+      "epoch": 7.155292908530319,
+      "grad_norm": 0.11469054222106934,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 69621
+    },
+    {
+      "epoch": 7.155601233299075,
+      "grad_norm": 0.06450911611318588,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 69624
+    },
+    {
+      "epoch": 7.155909558067831,
+      "grad_norm": 0.07107888162136078,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 69627
+    },
+    {
+      "epoch": 7.156217882836588,
+      "grad_norm": 0.05037635937333107,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 69630
+    },
+    {
+      "epoch": 7.156526207605344,
+      "grad_norm": 0.04990808293223381,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 69633
+    },
+    {
+      "epoch": 7.1568345323741,
+      "grad_norm": 0.10353230684995651,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 69636
+    },
+    {
+      "epoch": 7.1571428571428575,
+      "grad_norm": 0.05986601114273071,
+      "learning_rate": 0.01,
+      "loss": 2.0041,
+      "step": 69639
+    },
+    {
+      "epoch": 7.157451181911614,
+      "grad_norm": 0.03945085406303406,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 69642
+    },
+    {
+      "epoch": 7.15775950668037,
+      "grad_norm": 0.08378833532333374,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 69645
+    },
+    {
+      "epoch": 7.158067831449126,
+      "grad_norm": 0.07518744468688965,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 69648
+    },
+    {
+      "epoch": 7.158376156217883,
+      "grad_norm": 0.06334148347377777,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 69651
+    },
+    {
+      "epoch": 7.158684480986639,
+      "grad_norm": 0.08733922243118286,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 69654
+    },
+    {
+      "epoch": 7.158992805755395,
+      "grad_norm": 0.051821425557136536,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 69657
+    },
+    {
+      "epoch": 7.1593011305241525,
+      "grad_norm": 0.05659188702702522,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 69660
+    },
+    {
+      "epoch": 7.159609455292909,
+      "grad_norm": 0.07240952551364899,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 69663
+    },
+    {
+      "epoch": 7.159917780061665,
+      "grad_norm": 0.045318614691495895,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 69666
+    },
+    {
+      "epoch": 7.160226104830421,
+      "grad_norm": 0.08784624189138412,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 69669
+    },
+    {
+      "epoch": 7.160534429599178,
+      "grad_norm": 0.11292146146297455,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 69672
+    },
+    {
+      "epoch": 7.160842754367934,
+      "grad_norm": 0.22784489393234253,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 69675
+    },
+    {
+      "epoch": 7.16115107913669,
+      "grad_norm": 0.07886439561843872,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 69678
+    },
+    {
+      "epoch": 7.161459403905447,
+      "grad_norm": 0.15344084799289703,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 69681
+    },
+    {
+      "epoch": 7.161767728674204,
+      "grad_norm": 0.07547879219055176,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 69684
+    },
+    {
+      "epoch": 7.16207605344296,
+      "grad_norm": 0.051672521978616714,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 69687
+    },
+    {
+      "epoch": 7.1623843782117165,
+      "grad_norm": 0.03164055570960045,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 69690
+    },
+    {
+      "epoch": 7.162692702980473,
+      "grad_norm": 0.03657243773341179,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 69693
+    },
+    {
+      "epoch": 7.163001027749229,
+      "grad_norm": 0.0893481895327568,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 69696
+    },
+    {
+      "epoch": 7.163309352517985,
+      "grad_norm": 0.043061092495918274,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 69699
+    },
+    {
+      "epoch": 7.163617677286742,
+      "grad_norm": 0.04253191873431206,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 69702
+    },
+    {
+      "epoch": 7.163926002055499,
+      "grad_norm": 0.035961344838142395,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 69705
+    },
+    {
+      "epoch": 7.164234326824255,
+      "grad_norm": 0.03632194176316261,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 69708
+    },
+    {
+      "epoch": 7.1645426515930115,
+      "grad_norm": 0.038655634969472885,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 69711
+    },
+    {
+      "epoch": 7.164850976361768,
+      "grad_norm": 0.06551500409841537,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 69714
+    },
+    {
+      "epoch": 7.165159301130524,
+      "grad_norm": 0.06754742562770844,
+      "learning_rate": 0.01,
+      "loss": 2.0004,
+      "step": 69717
+    },
+    {
+      "epoch": 7.16546762589928,
+      "grad_norm": 0.04923485219478607,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 69720
+    },
+    {
+      "epoch": 7.165775950668037,
+      "grad_norm": 0.13406577706336975,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 69723
+    },
+    {
+      "epoch": 7.166084275436793,
+      "grad_norm": 0.07451289147138596,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 69726
+    },
+    {
+      "epoch": 7.16639260020555,
+      "grad_norm": 0.03220764547586441,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 69729
+    },
+    {
+      "epoch": 7.1667009249743066,
+      "grad_norm": 0.041585735976696014,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 69732
+    },
+    {
+      "epoch": 7.167009249743063,
+      "grad_norm": 0.030779125168919563,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 69735
+    },
+    {
+      "epoch": 7.167317574511819,
+      "grad_norm": 0.05135290324687958,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 69738
+    },
+    {
+      "epoch": 7.1676258992805755,
+      "grad_norm": 0.12039119750261307,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 69741
+    },
+    {
+      "epoch": 7.167934224049332,
+      "grad_norm": 0.06299499422311783,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 69744
+    },
+    {
+      "epoch": 7.168242548818088,
+      "grad_norm": 0.09064209461212158,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 69747
+    },
+    {
+      "epoch": 7.168550873586844,
+      "grad_norm": 0.08681541681289673,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 69750
+    },
+    {
+      "epoch": 7.168859198355602,
+      "grad_norm": 0.04912255331873894,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 69753
+    },
+    {
+      "epoch": 7.169167523124358,
+      "grad_norm": 0.07424995303153992,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 69756
+    },
+    {
+      "epoch": 7.169475847893114,
+      "grad_norm": 0.0543096549808979,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 69759
+    },
+    {
+      "epoch": 7.1697841726618705,
+      "grad_norm": 0.05928770452737808,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 69762
+    },
+    {
+      "epoch": 7.170092497430627,
+      "grad_norm": 0.07718564569950104,
+      "learning_rate": 0.01,
+      "loss": 2.0063,
+      "step": 69765
+    },
+    {
+      "epoch": 7.170400822199383,
+      "grad_norm": 0.040126901119947433,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 69768
+    },
+    {
+      "epoch": 7.170709146968139,
+      "grad_norm": 0.057264335453510284,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 69771
+    },
+    {
+      "epoch": 7.171017471736896,
+      "grad_norm": 0.05051598697900772,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 69774
+    },
+    {
+      "epoch": 7.171325796505653,
+      "grad_norm": 0.06816336512565613,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 69777
+    },
+    {
+      "epoch": 7.171634121274409,
+      "grad_norm": 0.19789281487464905,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 69780
+    },
+    {
+      "epoch": 7.1719424460431656,
+      "grad_norm": 0.11976084858179092,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 69783
+    },
+    {
+      "epoch": 7.172250770811922,
+      "grad_norm": 0.15244784951210022,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 69786
+    },
+    {
+      "epoch": 7.172559095580678,
+      "grad_norm": 0.11733053624629974,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 69789
+    },
+    {
+      "epoch": 7.1728674203494345,
+      "grad_norm": 0.04050086438655853,
+      "learning_rate": 0.01,
+      "loss": 1.9928,
+      "step": 69792
+    },
+    {
+      "epoch": 7.173175745118191,
+      "grad_norm": 0.045436665415763855,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 69795
+    },
+    {
+      "epoch": 7.173484069886948,
+      "grad_norm": 0.08359482884407043,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 69798
+    },
+    {
+      "epoch": 7.173792394655704,
+      "grad_norm": 0.04188248887658119,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 69801
+    },
+    {
+      "epoch": 7.174100719424461,
+      "grad_norm": 0.07575372606515884,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 69804
+    },
+    {
+      "epoch": 7.174409044193217,
+      "grad_norm": 0.09911274164915085,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 69807
+    },
+    {
+      "epoch": 7.174717368961973,
+      "grad_norm": 0.040851060301065445,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 69810
+    },
+    {
+      "epoch": 7.1750256937307295,
+      "grad_norm": 0.042334407567977905,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 69813
+    },
+    {
+      "epoch": 7.175334018499486,
+      "grad_norm": 0.03406887874007225,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 69816
+    },
+    {
+      "epoch": 7.175642343268242,
+      "grad_norm": 0.06487525999546051,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 69819
+    },
+    {
+      "epoch": 7.175950668036999,
+      "grad_norm": 0.09128468483686447,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 69822
+    },
+    {
+      "epoch": 7.176258992805756,
+      "grad_norm": 0.06886399537324905,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 69825
+    },
+    {
+      "epoch": 7.176567317574512,
+      "grad_norm": 0.09585999697446823,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 69828
+    },
+    {
+      "epoch": 7.176875642343268,
+      "grad_norm": 0.10458575189113617,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 69831
+    },
+    {
+      "epoch": 7.1771839671120246,
+      "grad_norm": 0.08773664385080338,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 69834
+    },
+    {
+      "epoch": 7.177492291880781,
+      "grad_norm": 0.0375094898045063,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 69837
+    },
+    {
+      "epoch": 7.177800616649537,
+      "grad_norm": 0.04263398051261902,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 69840
+    },
+    {
+      "epoch": 7.178108941418294,
+      "grad_norm": 0.04100068286061287,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 69843
+    },
+    {
+      "epoch": 7.178417266187051,
+      "grad_norm": 0.11466946452856064,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 69846
+    },
+    {
+      "epoch": 7.178725590955807,
+      "grad_norm": 0.0899171456694603,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 69849
+    },
+    {
+      "epoch": 7.179033915724563,
+      "grad_norm": 0.061083417385816574,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 69852
+    },
+    {
+      "epoch": 7.17934224049332,
+      "grad_norm": 0.1451992243528366,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 69855
+    },
+    {
+      "epoch": 7.179650565262076,
+      "grad_norm": 0.051237862557172775,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 69858
+    },
+    {
+      "epoch": 7.179958890030832,
+      "grad_norm": 0.07895320653915405,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 69861
+    },
+    {
+      "epoch": 7.1802672147995885,
+      "grad_norm": 0.05484063923358917,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 69864
+    },
+    {
+      "epoch": 7.180575539568346,
+      "grad_norm": 0.06715114414691925,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 69867
+    },
+    {
+      "epoch": 7.180883864337102,
+      "grad_norm": 0.070882648229599,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 69870
+    },
+    {
+      "epoch": 7.181192189105858,
+      "grad_norm": 0.06570753455162048,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 69873
+    },
+    {
+      "epoch": 7.181500513874615,
+      "grad_norm": 0.10089487582445145,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 69876
+    },
+    {
+      "epoch": 7.181808838643371,
+      "grad_norm": 0.062209602445364,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 69879
+    },
+    {
+      "epoch": 7.182117163412127,
+      "grad_norm": 0.08328171074390411,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 69882
+    },
+    {
+      "epoch": 7.1824254881808836,
+      "grad_norm": 0.0742257609963417,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 69885
+    },
+    {
+      "epoch": 7.18273381294964,
+      "grad_norm": 0.05598544701933861,
+      "learning_rate": 0.01,
+      "loss": 2.0187,
+      "step": 69888
+    },
+    {
+      "epoch": 7.183042137718397,
+      "grad_norm": 0.07855668663978577,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 69891
+    },
+    {
+      "epoch": 7.183350462487153,
+      "grad_norm": 0.09616530686616898,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 69894
+    },
+    {
+      "epoch": 7.18365878725591,
+      "grad_norm": 0.044885601848363876,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 69897
+    },
+    {
+      "epoch": 7.183967112024666,
+      "grad_norm": 0.08826860785484314,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 69900
+    },
+    {
+      "epoch": 7.184275436793422,
+      "grad_norm": 0.07852128893136978,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 69903
+    },
+    {
+      "epoch": 7.184583761562179,
+      "grad_norm": 0.094597227871418,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 69906
+    },
+    {
+      "epoch": 7.184892086330935,
+      "grad_norm": 0.04735616222023964,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 69909
+    },
+    {
+      "epoch": 7.185200411099692,
+      "grad_norm": 0.05992793291807175,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 69912
+    },
+    {
+      "epoch": 7.185508735868448,
+      "grad_norm": 0.08416339755058289,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 69915
+    },
+    {
+      "epoch": 7.185817060637205,
+      "grad_norm": 0.059381093829870224,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 69918
+    },
+    {
+      "epoch": 7.186125385405961,
+      "grad_norm": 0.037824418395757675,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 69921
+    },
+    {
+      "epoch": 7.186433710174717,
+      "grad_norm": 0.04952680319547653,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 69924
+    },
+    {
+      "epoch": 7.186742034943474,
+      "grad_norm": 0.07231353968381882,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 69927
+    },
+    {
+      "epoch": 7.18705035971223,
+      "grad_norm": 0.11462277173995972,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 69930
+    },
+    {
+      "epoch": 7.187358684480986,
+      "grad_norm": 0.07132146507501602,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 69933
+    },
+    {
+      "epoch": 7.187667009249743,
+      "grad_norm": 0.0858379453420639,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 69936
+    },
+    {
+      "epoch": 7.1879753340185,
+      "grad_norm": 0.045474838465452194,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 69939
+    },
+    {
+      "epoch": 7.188283658787256,
+      "grad_norm": 0.12445644289255142,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 69942
+    },
+    {
+      "epoch": 7.188591983556012,
+      "grad_norm": 0.044513870030641556,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 69945
+    },
+    {
+      "epoch": 7.188900308324769,
+      "grad_norm": 0.03778283670544624,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 69948
+    },
+    {
+      "epoch": 7.189208633093525,
+      "grad_norm": 0.044365961104631424,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 69951
+    },
+    {
+      "epoch": 7.189516957862281,
+      "grad_norm": 0.03456363081932068,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 69954
+    },
+    {
+      "epoch": 7.189825282631038,
+      "grad_norm": 0.0453643873333931,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 69957
+    },
+    {
+      "epoch": 7.190133607399795,
+      "grad_norm": 0.07771015167236328,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 69960
+    },
+    {
+      "epoch": 7.190441932168551,
+      "grad_norm": 0.0756489485502243,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 69963
+    },
+    {
+      "epoch": 7.190750256937307,
+      "grad_norm": 0.12961985170841217,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 69966
+    },
+    {
+      "epoch": 7.191058581706064,
+      "grad_norm": 0.07470633834600449,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 69969
+    },
+    {
+      "epoch": 7.19136690647482,
+      "grad_norm": 0.039709094911813736,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 69972
+    },
+    {
+      "epoch": 7.191675231243576,
+      "grad_norm": 0.0402153842151165,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 69975
+    },
+    {
+      "epoch": 7.191983556012333,
+      "grad_norm": 0.04119057580828667,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 69978
+    },
+    {
+      "epoch": 7.19229188078109,
+      "grad_norm": 0.036542341113090515,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 69981
+    },
+    {
+      "epoch": 7.192600205549846,
+      "grad_norm": 0.03711630031466484,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 69984
+    },
+    {
+      "epoch": 7.1929085303186024,
+      "grad_norm": 0.050209950655698776,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 69987
+    },
+    {
+      "epoch": 7.193216855087359,
+      "grad_norm": 0.06897444278001785,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 69990
+    },
+    {
+      "epoch": 7.193525179856115,
+      "grad_norm": 0.10195379704236984,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 69993
+    },
+    {
+      "epoch": 7.193833504624871,
+      "grad_norm": 0.08220812678337097,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 69996
+    },
+    {
+      "epoch": 7.194141829393628,
+      "grad_norm": 0.08391463756561279,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 69999
+    },
+    {
+      "epoch": 7.194450154162384,
+      "grad_norm": 0.03341416269540787,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 70002
+    },
+    {
+      "epoch": 7.194758478931141,
+      "grad_norm": 0.07957329601049423,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 70005
+    },
+    {
+      "epoch": 7.1950668036998975,
+      "grad_norm": 0.040082935243844986,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 70008
+    },
+    {
+      "epoch": 7.195375128468654,
+      "grad_norm": 0.07456467300653458,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 70011
+    },
+    {
+      "epoch": 7.19568345323741,
+      "grad_norm": 0.046207722276449203,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 70014
+    },
+    {
+      "epoch": 7.195991778006166,
+      "grad_norm": 0.06496291607618332,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 70017
+    },
+    {
+      "epoch": 7.196300102774923,
+      "grad_norm": 0.047517258673906326,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 70020
+    },
+    {
+      "epoch": 7.196608427543679,
+      "grad_norm": 0.033581919968128204,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 70023
+    },
+    {
+      "epoch": 7.196916752312436,
+      "grad_norm": 0.08832699060440063,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 70026
+    },
+    {
+      "epoch": 7.1972250770811925,
+      "grad_norm": 0.12826929986476898,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 70029
+    },
+    {
+      "epoch": 7.197533401849949,
+      "grad_norm": 0.11776229739189148,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 70032
+    },
+    {
+      "epoch": 7.197841726618705,
+      "grad_norm": 0.08085273951292038,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 70035
+    },
+    {
+      "epoch": 7.1981500513874614,
+      "grad_norm": 0.06540251523256302,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 70038
+    },
+    {
+      "epoch": 7.198458376156218,
+      "grad_norm": 0.05728829279541969,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 70041
+    },
+    {
+      "epoch": 7.198766700924974,
+      "grad_norm": 0.036182671785354614,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 70044
+    },
+    {
+      "epoch": 7.19907502569373,
+      "grad_norm": 0.03832792863249779,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 70047
+    },
+    {
+      "epoch": 7.199383350462488,
+      "grad_norm": 0.06086043268442154,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 70050
+    },
+    {
+      "epoch": 7.199691675231244,
+      "grad_norm": 0.02882802113890648,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 70053
+    },
+    {
+      "epoch": 7.2,
+      "grad_norm": 0.07022470235824585,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 70056
+    },
+    {
+      "epoch": 7.2003083247687565,
+      "grad_norm": 0.061370596289634705,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 70059
+    },
+    {
+      "epoch": 7.200616649537513,
+      "grad_norm": 0.1273171752691269,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 70062
+    },
+    {
+      "epoch": 7.200924974306269,
+      "grad_norm": 0.11996814608573914,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 70065
+    },
+    {
+      "epoch": 7.201233299075025,
+      "grad_norm": 0.0758858323097229,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 70068
+    },
+    {
+      "epoch": 7.201541623843782,
+      "grad_norm": 0.04552829638123512,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 70071
+    },
+    {
+      "epoch": 7.201849948612539,
+      "grad_norm": 0.04942410811781883,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 70074
+    },
+    {
+      "epoch": 7.202158273381295,
+      "grad_norm": 0.10216709226369858,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 70077
+    },
+    {
+      "epoch": 7.2024665981500515,
+      "grad_norm": 0.05046083778142929,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 70080
+    },
+    {
+      "epoch": 7.202774922918808,
+      "grad_norm": 0.09301755577325821,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 70083
+    },
+    {
+      "epoch": 7.203083247687564,
+      "grad_norm": 0.06181971728801727,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 70086
+    },
+    {
+      "epoch": 7.2033915724563204,
+      "grad_norm": 0.07836408168077469,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 70089
+    },
+    {
+      "epoch": 7.203699897225077,
+      "grad_norm": 0.04453438147902489,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 70092
+    },
+    {
+      "epoch": 7.204008221993833,
+      "grad_norm": 0.11325328052043915,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 70095
+    },
+    {
+      "epoch": 7.20431654676259,
+      "grad_norm": 0.17005731165409088,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 70098
+    },
+    {
+      "epoch": 7.204624871531347,
+      "grad_norm": 0.11921730637550354,
+      "learning_rate": 0.01,
+      "loss": 2.0139,
+      "step": 70101
+    },
+    {
+      "epoch": 7.204933196300103,
+      "grad_norm": 0.07009397447109222,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 70104
+    },
+    {
+      "epoch": 7.205241521068859,
+      "grad_norm": 0.037264637649059296,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 70107
+    },
+    {
+      "epoch": 7.2055498458376155,
+      "grad_norm": 0.11778846383094788,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 70110
+    },
+    {
+      "epoch": 7.205858170606372,
+      "grad_norm": 0.13234436511993408,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 70113
+    },
+    {
+      "epoch": 7.206166495375128,
+      "grad_norm": 0.08017081022262573,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 70116
+    },
+    {
+      "epoch": 7.206474820143885,
+      "grad_norm": 0.10627610236406326,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 70119
+    },
+    {
+      "epoch": 7.206783144912642,
+      "grad_norm": 0.111640565097332,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 70122
+    },
+    {
+      "epoch": 7.207091469681398,
+      "grad_norm": 0.049196891486644745,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 70125
+    },
+    {
+      "epoch": 7.207399794450154,
+      "grad_norm": 0.041943904012441635,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 70128
+    },
+    {
+      "epoch": 7.2077081192189105,
+      "grad_norm": 0.06802201271057129,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 70131
+    },
+    {
+      "epoch": 7.208016443987667,
+      "grad_norm": 0.043165501207113266,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 70134
+    },
+    {
+      "epoch": 7.208324768756423,
+      "grad_norm": 0.06273946911096573,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 70137
+    },
+    {
+      "epoch": 7.2086330935251794,
+      "grad_norm": 0.07074591517448425,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 70140
+    },
+    {
+      "epoch": 7.208941418293937,
+      "grad_norm": 0.0768829733133316,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 70143
+    },
+    {
+      "epoch": 7.209249743062693,
+      "grad_norm": 0.039911992847919464,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 70146
+    },
+    {
+      "epoch": 7.209558067831449,
+      "grad_norm": 0.06030252203345299,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 70149
+    },
+    {
+      "epoch": 7.209866392600206,
+      "grad_norm": 0.03984086588025093,
+      "learning_rate": 0.01,
+      "loss": 1.994,
+      "step": 70152
+    },
+    {
+      "epoch": 7.210174717368962,
+      "grad_norm": 0.032670505344867706,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 70155
+    },
+    {
+      "epoch": 7.210483042137718,
+      "grad_norm": 0.10686333477497101,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 70158
+    },
+    {
+      "epoch": 7.2107913669064745,
+      "grad_norm": 0.04879242554306984,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 70161
+    },
+    {
+      "epoch": 7.211099691675232,
+      "grad_norm": 0.07059837132692337,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 70164
+    },
+    {
+      "epoch": 7.211408016443988,
+      "grad_norm": 0.09764829277992249,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 70167
+    },
+    {
+      "epoch": 7.211716341212744,
+      "grad_norm": 0.05652858689427376,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 70170
+    },
+    {
+      "epoch": 7.212024665981501,
+      "grad_norm": 0.051845990121364594,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 70173
+    },
+    {
+      "epoch": 7.212332990750257,
+      "grad_norm": 0.06887342035770416,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 70176
+    },
+    {
+      "epoch": 7.212641315519013,
+      "grad_norm": 0.05305744335055351,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 70179
+    },
+    {
+      "epoch": 7.2129496402877695,
+      "grad_norm": 0.05928622931241989,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 70182
+    },
+    {
+      "epoch": 7.213257965056526,
+      "grad_norm": 0.05189640820026398,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 70185
+    },
+    {
+      "epoch": 7.213566289825283,
+      "grad_norm": 0.050485290586948395,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 70188
+    },
+    {
+      "epoch": 7.213874614594039,
+      "grad_norm": 0.05905221402645111,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 70191
+    },
+    {
+      "epoch": 7.214182939362796,
+      "grad_norm": 0.06555161625146866,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 70194
+    },
+    {
+      "epoch": 7.214491264131552,
+      "grad_norm": 0.03727792948484421,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 70197
+    },
+    {
+      "epoch": 7.214799588900308,
+      "grad_norm": 0.032580602914094925,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 70200
+    },
+    {
+      "epoch": 7.215107913669065,
+      "grad_norm": 0.042042069137096405,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 70203
+    },
+    {
+      "epoch": 7.215416238437821,
+      "grad_norm": 0.1405077874660492,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 70206
+    },
+    {
+      "epoch": 7.215724563206577,
+      "grad_norm": 0.08418957144021988,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 70209
+    },
+    {
+      "epoch": 7.216032887975334,
+      "grad_norm": 0.09414199739694595,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 70212
+    },
+    {
+      "epoch": 7.216341212744091,
+      "grad_norm": 0.053635187447071075,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 70215
+    },
+    {
+      "epoch": 7.216649537512847,
+      "grad_norm": 0.04404306784272194,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 70218
+    },
+    {
+      "epoch": 7.216957862281603,
+      "grad_norm": 0.0472695529460907,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 70221
+    },
+    {
+      "epoch": 7.21726618705036,
+      "grad_norm": 0.04150666296482086,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 70224
+    },
+    {
+      "epoch": 7.217574511819116,
+      "grad_norm": 0.06759631633758545,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 70227
+    },
+    {
+      "epoch": 7.217882836587872,
+      "grad_norm": 0.12022452801465988,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 70230
+    },
+    {
+      "epoch": 7.2181911613566285,
+      "grad_norm": 0.10294222086668015,
+      "learning_rate": 0.01,
+      "loss": 2.0086,
+      "step": 70233
+    },
+    {
+      "epoch": 7.218499486125386,
+      "grad_norm": 0.036887966096401215,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 70236
+    },
+    {
+      "epoch": 7.218807810894142,
+      "grad_norm": 0.03965602442622185,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 70239
+    },
+    {
+      "epoch": 7.219116135662898,
+      "grad_norm": 0.033555611968040466,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 70242
+    },
+    {
+      "epoch": 7.219424460431655,
+      "grad_norm": 0.03445450961589813,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 70245
+    },
+    {
+      "epoch": 7.219732785200411,
+      "grad_norm": 0.04350513964891434,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 70248
+    },
+    {
+      "epoch": 7.220041109969167,
+      "grad_norm": 0.06522413343191147,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 70251
+    },
+    {
+      "epoch": 7.220349434737924,
+      "grad_norm": 0.06339959055185318,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 70254
+    },
+    {
+      "epoch": 7.220657759506681,
+      "grad_norm": 0.04636206105351448,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 70257
+    },
+    {
+      "epoch": 7.220966084275437,
+      "grad_norm": 0.055801764130592346,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 70260
+    },
+    {
+      "epoch": 7.221274409044193,
+      "grad_norm": 0.03851722180843353,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 70263
+    },
+    {
+      "epoch": 7.22158273381295,
+      "grad_norm": 0.06473366171121597,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 70266
+    },
+    {
+      "epoch": 7.221891058581706,
+      "grad_norm": 0.08311999589204788,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 70269
+    },
+    {
+      "epoch": 7.222199383350462,
+      "grad_norm": 0.07522797584533691,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 70272
+    },
+    {
+      "epoch": 7.222507708119219,
+      "grad_norm": 0.09197179228067398,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 70275
+    },
+    {
+      "epoch": 7.222816032887975,
+      "grad_norm": 0.08302085846662521,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 70278
+    },
+    {
+      "epoch": 7.223124357656732,
+      "grad_norm": 0.07616004347801208,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 70281
+    },
+    {
+      "epoch": 7.223432682425488,
+      "grad_norm": 0.07368069142103195,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 70284
+    },
+    {
+      "epoch": 7.223741007194245,
+      "grad_norm": 0.07486431300640106,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 70287
+    },
+    {
+      "epoch": 7.224049331963001,
+      "grad_norm": 0.045419856905937195,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 70290
+    },
+    {
+      "epoch": 7.224357656731757,
+      "grad_norm": 0.05169800668954849,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 70293
+    },
+    {
+      "epoch": 7.224665981500514,
+      "grad_norm": 0.06700783222913742,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 70296
+    },
+    {
+      "epoch": 7.22497430626927,
+      "grad_norm": 0.0943138375878334,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 70299
+    },
+    {
+      "epoch": 7.225282631038027,
+      "grad_norm": 0.07699436694383621,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 70302
+    },
+    {
+      "epoch": 7.2255909558067835,
+      "grad_norm": 0.058129556477069855,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 70305
+    },
+    {
+      "epoch": 7.22589928057554,
+      "grad_norm": 0.11801168322563171,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 70308
+    },
+    {
+      "epoch": 7.226207605344296,
+      "grad_norm": 0.046479515731334686,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 70311
+    },
+    {
+      "epoch": 7.226515930113052,
+      "grad_norm": 0.0361471064388752,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 70314
+    },
+    {
+      "epoch": 7.226824254881809,
+      "grad_norm": 0.12745732069015503,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 70317
+    },
+    {
+      "epoch": 7.227132579650565,
+      "grad_norm": 0.11508181691169739,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 70320
+    },
+    {
+      "epoch": 7.227440904419321,
+      "grad_norm": 0.09497086703777313,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 70323
+    },
+    {
+      "epoch": 7.2277492291880785,
+      "grad_norm": 0.045153483748435974,
+      "learning_rate": 0.01,
+      "loss": 1.9945,
+      "step": 70326
+    },
+    {
+      "epoch": 7.228057553956835,
+      "grad_norm": 0.03023081086575985,
+      "learning_rate": 0.01,
+      "loss": 2.0191,
+      "step": 70329
+    },
+    {
+      "epoch": 7.228365878725591,
+      "grad_norm": 0.036539237946271896,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 70332
+    },
+    {
+      "epoch": 7.228674203494347,
+      "grad_norm": 0.05517994239926338,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 70335
+    },
+    {
+      "epoch": 7.228982528263104,
+      "grad_norm": 0.08023694902658463,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 70338
+    },
+    {
+      "epoch": 7.22929085303186,
+      "grad_norm": 0.04393969848752022,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 70341
+    },
+    {
+      "epoch": 7.229599177800616,
+      "grad_norm": 0.036692652851343155,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 70344
+    },
+    {
+      "epoch": 7.229907502569373,
+      "grad_norm": 0.03756039962172508,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 70347
+    },
+    {
+      "epoch": 7.23021582733813,
+      "grad_norm": 0.046152982860803604,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 70350
+    },
+    {
+      "epoch": 7.230524152106886,
+      "grad_norm": 0.24146318435668945,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 70353
+    },
+    {
+      "epoch": 7.2308324768756425,
+      "grad_norm": 0.14046429097652435,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 70356
+    },
+    {
+      "epoch": 7.231140801644399,
+      "grad_norm": 0.10172725468873978,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 70359
+    },
+    {
+      "epoch": 7.231449126413155,
+      "grad_norm": 0.08112365752458572,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 70362
+    },
+    {
+      "epoch": 7.231757451181911,
+      "grad_norm": 0.036505427211523056,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 70365
+    },
+    {
+      "epoch": 7.232065775950668,
+      "grad_norm": 0.06573676317930222,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 70368
+    },
+    {
+      "epoch": 7.232374100719425,
+      "grad_norm": 0.04601527750492096,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 70371
+    },
+    {
+      "epoch": 7.232682425488181,
+      "grad_norm": 0.04128945618867874,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 70374
+    },
+    {
+      "epoch": 7.2329907502569375,
+      "grad_norm": 0.03756236284971237,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 70377
+    },
+    {
+      "epoch": 7.233299075025694,
+      "grad_norm": 0.045076120644807816,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 70380
+    },
+    {
+      "epoch": 7.23360739979445,
+      "grad_norm": 0.07399538904428482,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 70383
+    },
+    {
+      "epoch": 7.233915724563206,
+      "grad_norm": 0.10196070373058319,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 70386
+    },
+    {
+      "epoch": 7.234224049331963,
+      "grad_norm": 0.06493962556123734,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 70389
+    },
+    {
+      "epoch": 7.234532374100719,
+      "grad_norm": 0.0905206948518753,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 70392
+    },
+    {
+      "epoch": 7.234840698869476,
+      "grad_norm": 0.051535073667764664,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 70395
+    },
+    {
+      "epoch": 7.2351490236382325,
+      "grad_norm": 0.03794541209936142,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 70398
+    },
+    {
+      "epoch": 7.235457348406989,
+      "grad_norm": 0.039028409868478775,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 70401
+    },
+    {
+      "epoch": 7.235765673175745,
+      "grad_norm": 0.03177224472165108,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 70404
+    },
+    {
+      "epoch": 7.2360739979445015,
+      "grad_norm": 0.03279658779501915,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 70407
+    },
+    {
+      "epoch": 7.236382322713258,
+      "grad_norm": 0.11790499091148376,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 70410
+    },
+    {
+      "epoch": 7.236690647482014,
+      "grad_norm": 0.08712024986743927,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 70413
+    },
+    {
+      "epoch": 7.23699897225077,
+      "grad_norm": 0.043823301792144775,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 70416
+    },
+    {
+      "epoch": 7.237307297019528,
+      "grad_norm": 0.03984406962990761,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 70419
+    },
+    {
+      "epoch": 7.237615621788284,
+      "grad_norm": 0.03345489874482155,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 70422
+    },
+    {
+      "epoch": 7.23792394655704,
+      "grad_norm": 0.030647002160549164,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 70425
+    },
+    {
+      "epoch": 7.2382322713257965,
+      "grad_norm": 0.061715736985206604,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 70428
+    },
+    {
+      "epoch": 7.238540596094553,
+      "grad_norm": 0.0880758985877037,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 70431
+    },
+    {
+      "epoch": 7.238848920863309,
+      "grad_norm": 0.051462333649396896,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 70434
+    },
+    {
+      "epoch": 7.239157245632065,
+      "grad_norm": 0.055362194776535034,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 70437
+    },
+    {
+      "epoch": 7.239465570400823,
+      "grad_norm": 0.04368078336119652,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 70440
+    },
+    {
+      "epoch": 7.239773895169579,
+      "grad_norm": 0.09169706702232361,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 70443
+    },
+    {
+      "epoch": 7.240082219938335,
+      "grad_norm": 0.03903011977672577,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 70446
+    },
+    {
+      "epoch": 7.2403905447070915,
+      "grad_norm": 0.19656112790107727,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 70449
+    },
+    {
+      "epoch": 7.240698869475848,
+      "grad_norm": 0.07515250146389008,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 70452
+    },
+    {
+      "epoch": 7.241007194244604,
+      "grad_norm": 0.0853821262717247,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 70455
+    },
+    {
+      "epoch": 7.2413155190133605,
+      "grad_norm": 0.06227338686585426,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 70458
+    },
+    {
+      "epoch": 7.241623843782117,
+      "grad_norm": 0.047402843832969666,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 70461
+    },
+    {
+      "epoch": 7.241932168550874,
+      "grad_norm": 0.03917578607797623,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 70464
+    },
+    {
+      "epoch": 7.24224049331963,
+      "grad_norm": 0.055436741560697556,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 70467
+    },
+    {
+      "epoch": 7.242548818088387,
+      "grad_norm": 0.05304566025733948,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 70470
+    },
+    {
+      "epoch": 7.242857142857143,
+      "grad_norm": 0.04701549932360649,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 70473
+    },
+    {
+      "epoch": 7.243165467625899,
+      "grad_norm": 0.045375335961580276,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 70476
+    },
+    {
+      "epoch": 7.2434737923946555,
+      "grad_norm": 0.09745540469884872,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 70479
+    },
+    {
+      "epoch": 7.243782117163412,
+      "grad_norm": 0.10383448749780655,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 70482
+    },
+    {
+      "epoch": 7.244090441932169,
+      "grad_norm": 0.08810384571552277,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 70485
+    },
+    {
+      "epoch": 7.244398766700925,
+      "grad_norm": 0.05690396949648857,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 70488
+    },
+    {
+      "epoch": 7.244707091469682,
+      "grad_norm": 0.03899385407567024,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 70491
+    },
+    {
+      "epoch": 7.245015416238438,
+      "grad_norm": 0.050195615738630295,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 70494
+    },
+    {
+      "epoch": 7.245323741007194,
+      "grad_norm": 0.046150173991918564,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 70497
+    },
+    {
+      "epoch": 7.2456320657759505,
+      "grad_norm": 0.052966147661209106,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 70500
+    },
+    {
+      "epoch": 7.245940390544707,
+      "grad_norm": 0.10062668472528458,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 70503
+    },
+    {
+      "epoch": 7.246248715313463,
+      "grad_norm": 0.07878834009170532,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 70506
+    },
+    {
+      "epoch": 7.24655704008222,
+      "grad_norm": 0.05327494442462921,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 70509
+    },
+    {
+      "epoch": 7.246865364850977,
+      "grad_norm": 0.03351713344454765,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 70512
+    },
+    {
+      "epoch": 7.247173689619733,
+      "grad_norm": 0.037406474351882935,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 70515
+    },
+    {
+      "epoch": 7.247482014388489,
+      "grad_norm": 0.11417467892169952,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 70518
+    },
+    {
+      "epoch": 7.247790339157246,
+      "grad_norm": 0.10274399816989899,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 70521
+    },
+    {
+      "epoch": 7.248098663926002,
+      "grad_norm": 0.03881362825632095,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 70524
+    },
+    {
+      "epoch": 7.248406988694758,
+      "grad_norm": 0.051997456699609756,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 70527
+    },
+    {
+      "epoch": 7.2487153134635145,
+      "grad_norm": 0.06575778871774673,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 70530
+    },
+    {
+      "epoch": 7.249023638232272,
+      "grad_norm": 0.04608520120382309,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 70533
+    },
+    {
+      "epoch": 7.249331963001028,
+      "grad_norm": 0.04164005443453789,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 70536
+    },
+    {
+      "epoch": 7.249640287769784,
+      "grad_norm": 0.05026416853070259,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 70539
+    },
+    {
+      "epoch": 7.249948612538541,
+      "grad_norm": 0.12604720890522003,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 70542
+    },
+    {
+      "epoch": 7.250256937307297,
+      "grad_norm": 0.04925797879695892,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 70545
+    },
+    {
+      "epoch": 7.250565262076053,
+      "grad_norm": 0.031429026275873184,
+      "learning_rate": 0.01,
+      "loss": 2.0033,
+      "step": 70548
+    },
+    {
+      "epoch": 7.2508735868448095,
+      "grad_norm": 0.10795203596353531,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 70551
+    },
+    {
+      "epoch": 7.251181911613566,
+      "grad_norm": 0.06839077174663544,
+      "learning_rate": 0.01,
+      "loss": 1.9978,
+      "step": 70554
+    },
+    {
+      "epoch": 7.251490236382323,
+      "grad_norm": 0.09800900518894196,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 70557
+    },
+    {
+      "epoch": 7.251798561151079,
+      "grad_norm": 0.06234534829854965,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 70560
+    },
+    {
+      "epoch": 7.252106885919836,
+      "grad_norm": 0.03908616676926613,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 70563
+    },
+    {
+      "epoch": 7.252415210688592,
+      "grad_norm": 0.06500250101089478,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 70566
+    },
+    {
+      "epoch": 7.252723535457348,
+      "grad_norm": 0.050453029572963715,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 70569
+    },
+    {
+      "epoch": 7.253031860226105,
+      "grad_norm": 0.10915414243936539,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 70572
+    },
+    {
+      "epoch": 7.253340184994861,
+      "grad_norm": 0.04887484759092331,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 70575
+    },
+    {
+      "epoch": 7.253648509763618,
+      "grad_norm": 0.07314929366111755,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 70578
+    },
+    {
+      "epoch": 7.253956834532374,
+      "grad_norm": 0.05896662175655365,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 70581
+    },
+    {
+      "epoch": 7.254265159301131,
+      "grad_norm": 0.04080936685204506,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 70584
+    },
+    {
+      "epoch": 7.254573484069887,
+      "grad_norm": 0.058053355664014816,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 70587
+    },
+    {
+      "epoch": 7.254881808838643,
+      "grad_norm": 0.04267947003245354,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 70590
+    },
+    {
+      "epoch": 7.2551901336074,
+      "grad_norm": 0.04386641085147858,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 70593
+    },
+    {
+      "epoch": 7.255498458376156,
+      "grad_norm": 0.06609844416379929,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 70596
+    },
+    {
+      "epoch": 7.255806783144912,
+      "grad_norm": 0.0940227136015892,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 70599
+    },
+    {
+      "epoch": 7.256115107913669,
+      "grad_norm": 0.03769245743751526,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 70602
+    },
+    {
+      "epoch": 7.256423432682426,
+      "grad_norm": 0.0913299098610878,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 70605
+    },
+    {
+      "epoch": 7.256731757451182,
+      "grad_norm": 0.10391309857368469,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 70608
+    },
+    {
+      "epoch": 7.257040082219938,
+      "grad_norm": 0.10451003164052963,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 70611
+    },
+    {
+      "epoch": 7.257348406988695,
+      "grad_norm": 0.059426721185445786,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 70614
+    },
+    {
+      "epoch": 7.257656731757451,
+      "grad_norm": 0.1078886017203331,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 70617
+    },
+    {
+      "epoch": 7.257965056526207,
+      "grad_norm": 0.13962066173553467,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 70620
+    },
+    {
+      "epoch": 7.2582733812949645,
+      "grad_norm": 0.058368947356939316,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 70623
+    },
+    {
+      "epoch": 7.258581706063721,
+      "grad_norm": 0.06157883629202843,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 70626
+    },
+    {
+      "epoch": 7.258890030832477,
+      "grad_norm": 0.06946786493062973,
+      "learning_rate": 0.01,
+      "loss": 2.0012,
+      "step": 70629
+    },
+    {
+      "epoch": 7.259198355601233,
+      "grad_norm": 0.049864597618579865,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 70632
+    },
+    {
+      "epoch": 7.25950668036999,
+      "grad_norm": 0.046192020177841187,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 70635
+    },
+    {
+      "epoch": 7.259815005138746,
+      "grad_norm": 0.03985748812556267,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 70638
+    },
+    {
+      "epoch": 7.260123329907502,
+      "grad_norm": 0.05892154574394226,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 70641
+    },
+    {
+      "epoch": 7.260431654676259,
+      "grad_norm": 0.04130382463335991,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 70644
+    },
+    {
+      "epoch": 7.260739979445016,
+      "grad_norm": 0.0541166253387928,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 70647
+    },
+    {
+      "epoch": 7.261048304213772,
+      "grad_norm": 0.039703115820884705,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 70650
+    },
+    {
+      "epoch": 7.261356628982528,
+      "grad_norm": 0.10397301614284515,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 70653
+    },
+    {
+      "epoch": 7.261664953751285,
+      "grad_norm": 0.09001152962446213,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 70656
+    },
+    {
+      "epoch": 7.261973278520041,
+      "grad_norm": 0.05959209427237511,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 70659
+    },
+    {
+      "epoch": 7.262281603288797,
+      "grad_norm": 0.06122048199176788,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 70662
+    },
+    {
+      "epoch": 7.262589928057554,
+      "grad_norm": 0.03712654113769531,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 70665
+    },
+    {
+      "epoch": 7.262898252826311,
+      "grad_norm": 0.06078263372182846,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 70668
+    },
+    {
+      "epoch": 7.263206577595067,
+      "grad_norm": 0.07234025001525879,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 70671
+    },
+    {
+      "epoch": 7.2635149023638235,
+      "grad_norm": 0.10953123867511749,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 70674
+    },
+    {
+      "epoch": 7.26382322713258,
+      "grad_norm": 0.10636293888092041,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 70677
+    },
+    {
+      "epoch": 7.264131551901336,
+      "grad_norm": 0.054197005927562714,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 70680
+    },
+    {
+      "epoch": 7.264439876670092,
+      "grad_norm": 0.0320538766682148,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 70683
+    },
+    {
+      "epoch": 7.264748201438849,
+      "grad_norm": 0.03399870917201042,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 70686
+    },
+    {
+      "epoch": 7.265056526207605,
+      "grad_norm": 0.115841805934906,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 70689
+    },
+    {
+      "epoch": 7.265364850976361,
+      "grad_norm": 0.04310959577560425,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 70692
+    },
+    {
+      "epoch": 7.2656731757451185,
+      "grad_norm": 0.16276821494102478,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 70695
+    },
+    {
+      "epoch": 7.265981500513875,
+      "grad_norm": 0.08462589234113693,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 70698
+    },
+    {
+      "epoch": 7.266289825282631,
+      "grad_norm": 0.05889437720179558,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 70701
+    },
+    {
+      "epoch": 7.266598150051387,
+      "grad_norm": 0.0893205776810646,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 70704
+    },
+    {
+      "epoch": 7.266906474820144,
+      "grad_norm": 0.03911794722080231,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 70707
+    },
+    {
+      "epoch": 7.2672147995889,
+      "grad_norm": 0.043647050857543945,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 70710
+    },
+    {
+      "epoch": 7.267523124357656,
+      "grad_norm": 0.03468010574579239,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 70713
+    },
+    {
+      "epoch": 7.2678314491264135,
+      "grad_norm": 0.0826035663485527,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 70716
+    },
+    {
+      "epoch": 7.26813977389517,
+      "grad_norm": 0.1150084063410759,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 70719
+    },
+    {
+      "epoch": 7.268448098663926,
+      "grad_norm": 0.06612154096364975,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 70722
+    },
+    {
+      "epoch": 7.2687564234326825,
+      "grad_norm": 0.09974373131990433,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 70725
+    },
+    {
+      "epoch": 7.269064748201439,
+      "grad_norm": 0.03690455108880997,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 70728
+    },
+    {
+      "epoch": 7.269373072970195,
+      "grad_norm": 0.0326729491353035,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 70731
+    },
+    {
+      "epoch": 7.269681397738951,
+      "grad_norm": 0.11430519074201584,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 70734
+    },
+    {
+      "epoch": 7.269989722507708,
+      "grad_norm": 0.043645069003105164,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 70737
+    },
+    {
+      "epoch": 7.270298047276465,
+      "grad_norm": 0.08154550194740295,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 70740
+    },
+    {
+      "epoch": 7.270606372045221,
+      "grad_norm": 0.07893747091293335,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 70743
+    },
+    {
+      "epoch": 7.2709146968139775,
+      "grad_norm": 0.05688686668872833,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 70746
+    },
+    {
+      "epoch": 7.271223021582734,
+      "grad_norm": 0.05141865462064743,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 70749
+    },
+    {
+      "epoch": 7.27153134635149,
+      "grad_norm": 0.041718002408742905,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 70752
+    },
+    {
+      "epoch": 7.271839671120246,
+      "grad_norm": 0.045142821967601776,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 70755
+    },
+    {
+      "epoch": 7.272147995889003,
+      "grad_norm": 0.06518824398517609,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 70758
+    },
+    {
+      "epoch": 7.27245632065776,
+      "grad_norm": 0.0417245589196682,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 70761
+    },
+    {
+      "epoch": 7.272764645426516,
+      "grad_norm": 0.06344173848628998,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 70764
+    },
+    {
+      "epoch": 7.2730729701952725,
+      "grad_norm": 0.1255732625722885,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 70767
+    },
+    {
+      "epoch": 7.273381294964029,
+      "grad_norm": 0.08490406721830368,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 70770
+    },
+    {
+      "epoch": 7.273689619732785,
+      "grad_norm": 0.0501803420484066,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 70773
+    },
+    {
+      "epoch": 7.2739979445015415,
+      "grad_norm": 0.04829498007893562,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 70776
+    },
+    {
+      "epoch": 7.274306269270298,
+      "grad_norm": 0.036724966019392014,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 70779
+    },
+    {
+      "epoch": 7.274614594039054,
+      "grad_norm": 0.03728116676211357,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 70782
+    },
+    {
+      "epoch": 7.274922918807811,
+      "grad_norm": 0.03553638234734535,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 70785
+    },
+    {
+      "epoch": 7.275231243576568,
+      "grad_norm": 0.04956892877817154,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 70788
+    },
+    {
+      "epoch": 7.275539568345324,
+      "grad_norm": 0.06011252477765083,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 70791
+    },
+    {
+      "epoch": 7.27584789311408,
+      "grad_norm": 0.05820993334054947,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 70794
+    },
+    {
+      "epoch": 7.2761562178828365,
+      "grad_norm": 0.11845695972442627,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 70797
+    },
+    {
+      "epoch": 7.276464542651593,
+      "grad_norm": 0.05176176875829697,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 70800
+    },
+    {
+      "epoch": 7.276772867420349,
+      "grad_norm": 0.043042611330747604,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 70803
+    },
+    {
+      "epoch": 7.277081192189106,
+      "grad_norm": 0.07894362509250641,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 70806
+    },
+    {
+      "epoch": 7.277389516957863,
+      "grad_norm": 0.04929111897945404,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 70809
+    },
+    {
+      "epoch": 7.277697841726619,
+      "grad_norm": 0.09846871346235275,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 70812
+    },
+    {
+      "epoch": 7.278006166495375,
+      "grad_norm": 0.07354690134525299,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 70815
+    },
+    {
+      "epoch": 7.2783144912641315,
+      "grad_norm": 0.1623416393995285,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 70818
+    },
+    {
+      "epoch": 7.278622816032888,
+      "grad_norm": 0.11214367300271988,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 70821
+    },
+    {
+      "epoch": 7.278931140801644,
+      "grad_norm": 0.06578688323497772,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 70824
+    },
+    {
+      "epoch": 7.2792394655704005,
+      "grad_norm": 0.05367565527558327,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 70827
+    },
+    {
+      "epoch": 7.279547790339157,
+      "grad_norm": 0.039414871484041214,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 70830
+    },
+    {
+      "epoch": 7.279856115107914,
+      "grad_norm": 0.08670930564403534,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 70833
+    },
+    {
+      "epoch": 7.28016443987667,
+      "grad_norm": 0.04982636123895645,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 70836
+    },
+    {
+      "epoch": 7.280472764645427,
+      "grad_norm": 0.04024531692266464,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 70839
+    },
+    {
+      "epoch": 7.280781089414183,
+      "grad_norm": 0.05240694060921669,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 70842
+    },
+    {
+      "epoch": 7.281089414182939,
+      "grad_norm": 0.047303877770900726,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 70845
+    },
+    {
+      "epoch": 7.2813977389516955,
+      "grad_norm": 0.14105463027954102,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 70848
+    },
+    {
+      "epoch": 7.281706063720452,
+      "grad_norm": 0.05714312568306923,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 70851
+    },
+    {
+      "epoch": 7.282014388489209,
+      "grad_norm": 0.10342295467853546,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 70854
+    },
+    {
+      "epoch": 7.282322713257965,
+      "grad_norm": 0.12087305635213852,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 70857
+    },
+    {
+      "epoch": 7.282631038026722,
+      "grad_norm": 0.05759386718273163,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 70860
+    },
+    {
+      "epoch": 7.282939362795478,
+      "grad_norm": 0.04616248235106468,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 70863
+    },
+    {
+      "epoch": 7.283247687564234,
+      "grad_norm": 0.04186839610338211,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 70866
+    },
+    {
+      "epoch": 7.2835560123329905,
+      "grad_norm": 0.04863681644201279,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 70869
+    },
+    {
+      "epoch": 7.283864337101747,
+      "grad_norm": 0.050969015806913376,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 70872
+    },
+    {
+      "epoch": 7.284172661870503,
+      "grad_norm": 0.053215015679597855,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 70875
+    },
+    {
+      "epoch": 7.28448098663926,
+      "grad_norm": 0.10984474420547485,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 70878
+    },
+    {
+      "epoch": 7.284789311408017,
+      "grad_norm": 0.09052685648202896,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 70881
+    },
+    {
+      "epoch": 7.285097636176773,
+      "grad_norm": 0.09505495429039001,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 70884
+    },
+    {
+      "epoch": 7.285405960945529,
+      "grad_norm": 0.03760739415884018,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 70887
+    },
+    {
+      "epoch": 7.285714285714286,
+      "grad_norm": 0.03946032375097275,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 70890
+    },
+    {
+      "epoch": 7.286022610483042,
+      "grad_norm": 0.12385135889053345,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 70893
+    },
+    {
+      "epoch": 7.286330935251798,
+      "grad_norm": 0.11148396134376526,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 70896
+    },
+    {
+      "epoch": 7.286639260020555,
+      "grad_norm": 0.07253854721784592,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 70899
+    },
+    {
+      "epoch": 7.286947584789312,
+      "grad_norm": 0.048602595925331116,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 70902
+    },
+    {
+      "epoch": 7.287255909558068,
+      "grad_norm": 0.05885033681988716,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 70905
+    },
+    {
+      "epoch": 7.287564234326824,
+      "grad_norm": 0.057307545095682144,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 70908
+    },
+    {
+      "epoch": 7.287872559095581,
+      "grad_norm": 0.08410783857107162,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 70911
+    },
+    {
+      "epoch": 7.288180883864337,
+      "grad_norm": 0.07850154489278793,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 70914
+    },
+    {
+      "epoch": 7.288489208633093,
+      "grad_norm": 0.07075075805187225,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 70917
+    },
+    {
+      "epoch": 7.2887975334018495,
+      "grad_norm": 0.07684222608804703,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 70920
+    },
+    {
+      "epoch": 7.289105858170607,
+      "grad_norm": 0.10370174795389175,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 70923
+    },
+    {
+      "epoch": 7.289414182939363,
+      "grad_norm": 0.15722966194152832,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 70926
+    },
+    {
+      "epoch": 7.289722507708119,
+      "grad_norm": 0.08678031712770462,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 70929
+    },
+    {
+      "epoch": 7.290030832476876,
+      "grad_norm": 0.046776097267866135,
+      "learning_rate": 0.01,
+      "loss": 2.0017,
+      "step": 70932
+    },
+    {
+      "epoch": 7.290339157245632,
+      "grad_norm": 0.09871851652860641,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 70935
+    },
+    {
+      "epoch": 7.290647482014388,
+      "grad_norm": 0.03779609501361847,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 70938
+    },
+    {
+      "epoch": 7.290955806783145,
+      "grad_norm": 0.10300016403198242,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 70941
+    },
+    {
+      "epoch": 7.291264131551902,
+      "grad_norm": 0.051939114928245544,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 70944
+    },
+    {
+      "epoch": 7.291572456320658,
+      "grad_norm": 0.08400366455316544,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 70947
+    },
+    {
+      "epoch": 7.291880781089414,
+      "grad_norm": 0.04677712917327881,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 70950
+    },
+    {
+      "epoch": 7.292189105858171,
+      "grad_norm": 0.07676515728235245,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 70953
+    },
+    {
+      "epoch": 7.292497430626927,
+      "grad_norm": 0.0831826701760292,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 70956
+    },
+    {
+      "epoch": 7.292805755395683,
+      "grad_norm": 0.10186062008142471,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 70959
+    },
+    {
+      "epoch": 7.29311408016444,
+      "grad_norm": 0.04252918064594269,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 70962
+    },
+    {
+      "epoch": 7.293422404933196,
+      "grad_norm": 0.09480611979961395,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 70965
+    },
+    {
+      "epoch": 7.293730729701953,
+      "grad_norm": 0.0868213027715683,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 70968
+    },
+    {
+      "epoch": 7.294039054470709,
+      "grad_norm": 0.0904143899679184,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 70971
+    },
+    {
+      "epoch": 7.294347379239466,
+      "grad_norm": 0.05634519085288048,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 70974
+    },
+    {
+      "epoch": 7.294655704008222,
+      "grad_norm": 0.09237989783287048,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 70977
+    },
+    {
+      "epoch": 7.294964028776978,
+      "grad_norm": 0.07044427841901779,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 70980
+    },
+    {
+      "epoch": 7.295272353545735,
+      "grad_norm": 0.04916350916028023,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 70983
+    },
+    {
+      "epoch": 7.295580678314491,
+      "grad_norm": 0.037802912294864655,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 70986
+    },
+    {
+      "epoch": 7.295889003083247,
+      "grad_norm": 0.04316788166761398,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 70989
+    },
+    {
+      "epoch": 7.2961973278520045,
+      "grad_norm": 0.05099441111087799,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 70992
+    },
+    {
+      "epoch": 7.296505652620761,
+      "grad_norm": 0.09012050926685333,
+      "learning_rate": 0.01,
+      "loss": 2.0042,
+      "step": 70995
+    },
+    {
+      "epoch": 7.296813977389517,
+      "grad_norm": 0.07649296522140503,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 70998
+    },
+    {
+      "epoch": 7.297122302158273,
+      "grad_norm": 0.06650659441947937,
+      "learning_rate": 0.01,
+      "loss": 2.0077,
+      "step": 71001
+    },
+    {
+      "epoch": 7.29743062692703,
+      "grad_norm": 0.06413618475198746,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 71004
+    },
+    {
+      "epoch": 7.297738951695786,
+      "grad_norm": 0.039578627794981,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 71007
+    },
+    {
+      "epoch": 7.298047276464542,
+      "grad_norm": 0.07007674872875214,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 71010
+    },
+    {
+      "epoch": 7.298355601233299,
+      "grad_norm": 0.05441688373684883,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 71013
+    },
+    {
+      "epoch": 7.298663926002056,
+      "grad_norm": 0.04143676534295082,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 71016
+    },
+    {
+      "epoch": 7.298972250770812,
+      "grad_norm": 0.12231336534023285,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 71019
+    },
+    {
+      "epoch": 7.299280575539568,
+      "grad_norm": 0.03967202827334404,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 71022
+    },
+    {
+      "epoch": 7.299588900308325,
+      "grad_norm": 0.08297855406999588,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 71025
+    },
+    {
+      "epoch": 7.299897225077081,
+      "grad_norm": 0.08017514646053314,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 71028
+    },
+    {
+      "epoch": 7.300205549845837,
+      "grad_norm": 0.06581740826368332,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 71031
+    },
+    {
+      "epoch": 7.300513874614594,
+      "grad_norm": 0.10993166267871857,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 71034
+    },
+    {
+      "epoch": 7.300822199383351,
+      "grad_norm": 0.04322252422571182,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 71037
+    },
+    {
+      "epoch": 7.301130524152107,
+      "grad_norm": 0.04396703466773033,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 71040
+    },
+    {
+      "epoch": 7.3014388489208635,
+      "grad_norm": 0.042626939713954926,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 71043
+    },
+    {
+      "epoch": 7.30174717368962,
+      "grad_norm": 0.03424423187971115,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 71046
+    },
+    {
+      "epoch": 7.302055498458376,
+      "grad_norm": 0.11951079219579697,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 71049
+    },
+    {
+      "epoch": 7.302363823227132,
+      "grad_norm": 0.07407492399215698,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 71052
+    },
+    {
+      "epoch": 7.302672147995889,
+      "grad_norm": 0.0694723054766655,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 71055
+    },
+    {
+      "epoch": 7.302980472764645,
+      "grad_norm": 0.042367156594991684,
+      "learning_rate": 0.01,
+      "loss": 1.999,
+      "step": 71058
+    },
+    {
+      "epoch": 7.303288797533402,
+      "grad_norm": 0.052406225353479385,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 71061
+    },
+    {
+      "epoch": 7.3035971223021585,
+      "grad_norm": 0.060722094029188156,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 71064
+    },
+    {
+      "epoch": 7.303905447070915,
+      "grad_norm": 0.0947566032409668,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 71067
+    },
+    {
+      "epoch": 7.304213771839671,
+      "grad_norm": 0.14418601989746094,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 71070
+    },
+    {
+      "epoch": 7.304522096608427,
+      "grad_norm": 0.0899231880903244,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 71073
+    },
+    {
+      "epoch": 7.304830421377184,
+      "grad_norm": 0.057492516934871674,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 71076
+    },
+    {
+      "epoch": 7.30513874614594,
+      "grad_norm": 0.042988330125808716,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 71079
+    },
+    {
+      "epoch": 7.305447070914697,
+      "grad_norm": 0.06894852221012115,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 71082
+    },
+    {
+      "epoch": 7.305755395683454,
+      "grad_norm": 0.047955472022295,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 71085
+    },
+    {
+      "epoch": 7.30606372045221,
+      "grad_norm": 0.0502055399119854,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 71088
+    },
+    {
+      "epoch": 7.306372045220966,
+      "grad_norm": 0.042237646877765656,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 71091
+    },
+    {
+      "epoch": 7.3066803699897225,
+      "grad_norm": 0.07256586849689484,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 71094
+    },
+    {
+      "epoch": 7.306988694758479,
+      "grad_norm": 0.12676836550235748,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 71097
+    },
+    {
+      "epoch": 7.307297019527235,
+      "grad_norm": 0.10870865732431412,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 71100
+    },
+    {
+      "epoch": 7.307605344295991,
+      "grad_norm": 0.10361792892217636,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 71103
+    },
+    {
+      "epoch": 7.307913669064749,
+      "grad_norm": 0.05035272613167763,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 71106
+    },
+    {
+      "epoch": 7.308221993833505,
+      "grad_norm": 0.09772878140211105,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 71109
+    },
+    {
+      "epoch": 7.308530318602261,
+      "grad_norm": 0.05700725317001343,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 71112
+    },
+    {
+      "epoch": 7.3088386433710175,
+      "grad_norm": 0.04437270760536194,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 71115
+    },
+    {
+      "epoch": 7.309146968139774,
+      "grad_norm": 0.0368645042181015,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 71118
+    },
+    {
+      "epoch": 7.30945529290853,
+      "grad_norm": 0.0390106737613678,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 71121
+    },
+    {
+      "epoch": 7.309763617677286,
+      "grad_norm": 0.18610909581184387,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 71124
+    },
+    {
+      "epoch": 7.310071942446044,
+      "grad_norm": 0.16700612008571625,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 71127
+    },
+    {
+      "epoch": 7.3103802672148,
+      "grad_norm": 0.15059486031532288,
+      "learning_rate": 0.01,
+      "loss": 2.0078,
+      "step": 71130
+    },
+    {
+      "epoch": 7.310688591983556,
+      "grad_norm": 0.08873619139194489,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 71133
+    },
+    {
+      "epoch": 7.310996916752313,
+      "grad_norm": 0.05713324993848801,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 71136
+    },
+    {
+      "epoch": 7.311305241521069,
+      "grad_norm": 0.04134386405348778,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 71139
+    },
+    {
+      "epoch": 7.311613566289825,
+      "grad_norm": 0.03700621798634529,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 71142
+    },
+    {
+      "epoch": 7.3119218910585815,
+      "grad_norm": 0.04931655153632164,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 71145
+    },
+    {
+      "epoch": 7.312230215827338,
+      "grad_norm": 0.03286673501133919,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 71148
+    },
+    {
+      "epoch": 7.312538540596094,
+      "grad_norm": 0.04770418256521225,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 71151
+    },
+    {
+      "epoch": 7.312846865364851,
+      "grad_norm": 0.04040040448307991,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 71154
+    },
+    {
+      "epoch": 7.313155190133608,
+      "grad_norm": 0.041751354932785034,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 71157
+    },
+    {
+      "epoch": 7.313463514902364,
+      "grad_norm": 0.05756094679236412,
+      "learning_rate": 0.01,
+      "loss": 2.0109,
+      "step": 71160
+    },
+    {
+      "epoch": 7.31377183967112,
+      "grad_norm": 0.0506497286260128,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 71163
+    },
+    {
+      "epoch": 7.3140801644398765,
+      "grad_norm": 0.11589113622903824,
+      "learning_rate": 0.01,
+      "loss": 1.9893,
+      "step": 71166
+    },
+    {
+      "epoch": 7.314388489208633,
+      "grad_norm": 0.042656585574150085,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 71169
+    },
+    {
+      "epoch": 7.314696813977389,
+      "grad_norm": 0.09102977067232132,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 71172
+    },
+    {
+      "epoch": 7.315005138746146,
+      "grad_norm": 0.07668044418096542,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 71175
+    },
+    {
+      "epoch": 7.315313463514903,
+      "grad_norm": 0.1044536828994751,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 71178
+    },
+    {
+      "epoch": 7.315621788283659,
+      "grad_norm": 0.03361400589346886,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 71181
+    },
+    {
+      "epoch": 7.315930113052415,
+      "grad_norm": 0.036341775208711624,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 71184
+    },
+    {
+      "epoch": 7.316238437821172,
+      "grad_norm": 0.06732170283794403,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 71187
+    },
+    {
+      "epoch": 7.316546762589928,
+      "grad_norm": 0.056870389729738235,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 71190
+    },
+    {
+      "epoch": 7.316855087358684,
+      "grad_norm": 0.05577719956636429,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 71193
+    },
+    {
+      "epoch": 7.3171634121274405,
+      "grad_norm": 0.045024048537015915,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 71196
+    },
+    {
+      "epoch": 7.317471736896198,
+      "grad_norm": 0.05428456515073776,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 71199
+    },
+    {
+      "epoch": 7.317780061664954,
+      "grad_norm": 0.0677822083234787,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 71202
+    },
+    {
+      "epoch": 7.31808838643371,
+      "grad_norm": 0.10919659584760666,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 71205
+    },
+    {
+      "epoch": 7.318396711202467,
+      "grad_norm": 0.0439334474503994,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 71208
+    },
+    {
+      "epoch": 7.318705035971223,
+      "grad_norm": 0.04475986957550049,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 71211
+    },
+    {
+      "epoch": 7.319013360739979,
+      "grad_norm": 0.04004666581749916,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 71214
+    },
+    {
+      "epoch": 7.3193216855087355,
+      "grad_norm": 0.03669333830475807,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 71217
+    },
+    {
+      "epoch": 7.319630010277493,
+      "grad_norm": 0.06024492159485817,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 71220
+    },
+    {
+      "epoch": 7.319938335046249,
+      "grad_norm": 0.052151869982481,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 71223
+    },
+    {
+      "epoch": 7.320246659815005,
+      "grad_norm": 0.04867462441325188,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 71226
+    },
+    {
+      "epoch": 7.320554984583762,
+      "grad_norm": 0.13954968750476837,
+      "learning_rate": 0.01,
+      "loss": 2.0095,
+      "step": 71229
+    },
+    {
+      "epoch": 7.320863309352518,
+      "grad_norm": 0.07975666970014572,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 71232
+    },
+    {
+      "epoch": 7.321171634121274,
+      "grad_norm": 0.06927415728569031,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 71235
+    },
+    {
+      "epoch": 7.321479958890031,
+      "grad_norm": 0.05637050047516823,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 71238
+    },
+    {
+      "epoch": 7.321788283658787,
+      "grad_norm": 0.037037819623947144,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 71241
+    },
+    {
+      "epoch": 7.322096608427544,
+      "grad_norm": 0.03341764956712723,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 71244
+    },
+    {
+      "epoch": 7.3224049331963,
+      "grad_norm": 0.11337383091449738,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 71247
+    },
+    {
+      "epoch": 7.322713257965057,
+      "grad_norm": 0.05702521279454231,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 71250
+    },
+    {
+      "epoch": 7.323021582733813,
+      "grad_norm": 0.06875944882631302,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 71253
+    },
+    {
+      "epoch": 7.323329907502569,
+      "grad_norm": 0.09147006273269653,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 71256
+    },
+    {
+      "epoch": 7.323638232271326,
+      "grad_norm": 0.07072997838258743,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 71259
+    },
+    {
+      "epoch": 7.323946557040082,
+      "grad_norm": 0.08873667567968369,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 71262
+    },
+    {
+      "epoch": 7.324254881808839,
+      "grad_norm": 0.06994672864675522,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 71265
+    },
+    {
+      "epoch": 7.324563206577595,
+      "grad_norm": 0.06470712274312973,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 71268
+    },
+    {
+      "epoch": 7.324871531346352,
+      "grad_norm": 0.07046987116336823,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 71271
+    },
+    {
+      "epoch": 7.325179856115108,
+      "grad_norm": 0.07113350182771683,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 71274
+    },
+    {
+      "epoch": 7.325488180883864,
+      "grad_norm": 0.07272438704967499,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 71277
+    },
+    {
+      "epoch": 7.325796505652621,
+      "grad_norm": 0.14653034508228302,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 71280
+    },
+    {
+      "epoch": 7.326104830421377,
+      "grad_norm": 0.0946449339389801,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 71283
+    },
+    {
+      "epoch": 7.326413155190133,
+      "grad_norm": 0.041181258857250214,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 71286
+    },
+    {
+      "epoch": 7.32672147995889,
+      "grad_norm": 0.049023959785699844,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 71289
+    },
+    {
+      "epoch": 7.327029804727647,
+      "grad_norm": 0.03693239763379097,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 71292
+    },
+    {
+      "epoch": 7.327338129496403,
+      "grad_norm": 0.03846334293484688,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 71295
+    },
+    {
+      "epoch": 7.327646454265159,
+      "grad_norm": 0.061339206993579865,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 71298
+    },
+    {
+      "epoch": 7.327954779033916,
+      "grad_norm": 0.056017693132162094,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 71301
+    },
+    {
+      "epoch": 7.328263103802672,
+      "grad_norm": 0.046218231320381165,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 71304
+    },
+    {
+      "epoch": 7.328571428571428,
+      "grad_norm": 0.04835920408368111,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 71307
+    },
+    {
+      "epoch": 7.328879753340185,
+      "grad_norm": 0.035025544464588165,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 71310
+    },
+    {
+      "epoch": 7.329188078108942,
+      "grad_norm": 0.048764798790216446,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 71313
+    },
+    {
+      "epoch": 7.329496402877698,
+      "grad_norm": 0.05903776362538338,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 71316
+    },
+    {
+      "epoch": 7.329804727646454,
+      "grad_norm": 0.05478388071060181,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 71319
+    },
+    {
+      "epoch": 7.330113052415211,
+      "grad_norm": 0.1303192377090454,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 71322
+    },
+    {
+      "epoch": 7.330421377183967,
+      "grad_norm": 0.10789795964956284,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 71325
+    },
+    {
+      "epoch": 7.330729701952723,
+      "grad_norm": 0.07266957312822342,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 71328
+    },
+    {
+      "epoch": 7.33103802672148,
+      "grad_norm": 0.043115101754665375,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 71331
+    },
+    {
+      "epoch": 7.331346351490236,
+      "grad_norm": 0.04297550767660141,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 71334
+    },
+    {
+      "epoch": 7.331654676258993,
+      "grad_norm": 0.1441577821969986,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 71337
+    },
+    {
+      "epoch": 7.3319630010277494,
+      "grad_norm": 0.04849548637866974,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 71340
+    },
+    {
+      "epoch": 7.332271325796506,
+      "grad_norm": 0.03402062505483627,
+      "learning_rate": 0.01,
+      "loss": 1.998,
+      "step": 71343
+    },
+    {
+      "epoch": 7.332579650565262,
+      "grad_norm": 0.03585872799158096,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 71346
+    },
+    {
+      "epoch": 7.332887975334018,
+      "grad_norm": 0.07403067499399185,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 71349
+    },
+    {
+      "epoch": 7.333196300102775,
+      "grad_norm": 0.05098595842719078,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 71352
+    },
+    {
+      "epoch": 7.333504624871531,
+      "grad_norm": 0.06518220156431198,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 71355
+    },
+    {
+      "epoch": 7.333812949640288,
+      "grad_norm": 0.04573628678917885,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 71358
+    },
+    {
+      "epoch": 7.3341212744090445,
+      "grad_norm": 0.07890237867832184,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 71361
+    },
+    {
+      "epoch": 7.334429599177801,
+      "grad_norm": 0.09591496735811234,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 71364
+    },
+    {
+      "epoch": 7.334737923946557,
+      "grad_norm": 0.13394559919834137,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 71367
+    },
+    {
+      "epoch": 7.335046248715313,
+      "grad_norm": 0.10109490901231766,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 71370
+    },
+    {
+      "epoch": 7.33535457348407,
+      "grad_norm": 0.09080103784799576,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 71373
+    },
+    {
+      "epoch": 7.335662898252826,
+      "grad_norm": 0.09327443689107895,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 71376
+    },
+    {
+      "epoch": 7.335971223021582,
+      "grad_norm": 0.05303800478577614,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 71379
+    },
+    {
+      "epoch": 7.3362795477903395,
+      "grad_norm": 0.04416405409574509,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 71382
+    },
+    {
+      "epoch": 7.336587872559096,
+      "grad_norm": 0.031716201454401016,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 71385
+    },
+    {
+      "epoch": 7.336896197327852,
+      "grad_norm": 0.045393042266368866,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 71388
+    },
+    {
+      "epoch": 7.3372045220966084,
+      "grad_norm": 0.09052672982215881,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 71391
+    },
+    {
+      "epoch": 7.337512846865365,
+      "grad_norm": 0.046372365206480026,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 71394
+    },
+    {
+      "epoch": 7.337821171634121,
+      "grad_norm": 0.037272438406944275,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 71397
+    },
+    {
+      "epoch": 7.338129496402877,
+      "grad_norm": 0.10809680074453354,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 71400
+    },
+    {
+      "epoch": 7.338437821171635,
+      "grad_norm": 0.15810643136501312,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 71403
+    },
+    {
+      "epoch": 7.338746145940391,
+      "grad_norm": 0.13274040818214417,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 71406
+    },
+    {
+      "epoch": 7.339054470709147,
+      "grad_norm": 0.09903276711702347,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 71409
+    },
+    {
+      "epoch": 7.3393627954779035,
+      "grad_norm": 0.06785092502832413,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 71412
+    },
+    {
+      "epoch": 7.33967112024666,
+      "grad_norm": 0.044940706342458725,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 71415
+    },
+    {
+      "epoch": 7.339979445015416,
+      "grad_norm": 0.04073977470397949,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 71418
+    },
+    {
+      "epoch": 7.340287769784172,
+      "grad_norm": 0.03847421333193779,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 71421
+    },
+    {
+      "epoch": 7.340596094552929,
+      "grad_norm": 0.04725421220064163,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 71424
+    },
+    {
+      "epoch": 7.340904419321686,
+      "grad_norm": 0.05868101492524147,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 71427
+    },
+    {
+      "epoch": 7.341212744090442,
+      "grad_norm": 0.06789273023605347,
+      "learning_rate": 0.01,
+      "loss": 1.9981,
+      "step": 71430
+    },
+    {
+      "epoch": 7.3415210688591985,
+      "grad_norm": 0.09006938338279724,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 71433
+    },
+    {
+      "epoch": 7.341829393627955,
+      "grad_norm": 0.05754098296165466,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 71436
+    },
+    {
+      "epoch": 7.342137718396711,
+      "grad_norm": 0.03706538304686546,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 71439
+    },
+    {
+      "epoch": 7.3424460431654675,
+      "grad_norm": 0.04427531361579895,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 71442
+    },
+    {
+      "epoch": 7.342754367934224,
+      "grad_norm": 0.0534258633852005,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 71445
+    },
+    {
+      "epoch": 7.34306269270298,
+      "grad_norm": 0.0816885381937027,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 71448
+    },
+    {
+      "epoch": 7.343371017471737,
+      "grad_norm": 0.09698469191789627,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 71451
+    },
+    {
+      "epoch": 7.343679342240494,
+      "grad_norm": 0.043008897453546524,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 71454
+    },
+    {
+      "epoch": 7.34398766700925,
+      "grad_norm": 0.1080782413482666,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 71457
+    },
+    {
+      "epoch": 7.344295991778006,
+      "grad_norm": 0.06114106625318527,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 71460
+    },
+    {
+      "epoch": 7.3446043165467625,
+      "grad_norm": 0.08944262564182281,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 71463
+    },
+    {
+      "epoch": 7.344912641315519,
+      "grad_norm": 0.08879029750823975,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 71466
+    },
+    {
+      "epoch": 7.345220966084275,
+      "grad_norm": 0.09932418167591095,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 71469
+    },
+    {
+      "epoch": 7.345529290853031,
+      "grad_norm": 0.050661344081163406,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 71472
+    },
+    {
+      "epoch": 7.345837615621789,
+      "grad_norm": 0.04675713926553726,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 71475
+    },
+    {
+      "epoch": 7.346145940390545,
+      "grad_norm": 0.04077441617846489,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 71478
+    },
+    {
+      "epoch": 7.346454265159301,
+      "grad_norm": 0.04659559577703476,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 71481
+    },
+    {
+      "epoch": 7.3467625899280575,
+      "grad_norm": 0.09812557697296143,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 71484
+    },
+    {
+      "epoch": 7.347070914696814,
+      "grad_norm": 0.03201153874397278,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 71487
+    },
+    {
+      "epoch": 7.34737923946557,
+      "grad_norm": 0.12122839689254761,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 71490
+    },
+    {
+      "epoch": 7.3476875642343265,
+      "grad_norm": 0.11808911710977554,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 71493
+    },
+    {
+      "epoch": 7.347995889003084,
+      "grad_norm": 0.06771431118249893,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 71496
+    },
+    {
+      "epoch": 7.34830421377184,
+      "grad_norm": 0.07350712269544601,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 71499
+    },
+    {
+      "epoch": 7.348612538540596,
+      "grad_norm": 0.043973151594400406,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 71502
+    },
+    {
+      "epoch": 7.348920863309353,
+      "grad_norm": 0.03686361759901047,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 71505
+    },
+    {
+      "epoch": 7.349229188078109,
+      "grad_norm": 0.07905039191246033,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 71508
+    },
+    {
+      "epoch": 7.349537512846865,
+      "grad_norm": 0.05678734928369522,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 71511
+    },
+    {
+      "epoch": 7.3498458376156215,
+      "grad_norm": 0.10984398424625397,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 71514
+    },
+    {
+      "epoch": 7.350154162384378,
+      "grad_norm": 0.11405175179243088,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 71517
+    },
+    {
+      "epoch": 7.350462487153135,
+      "grad_norm": 0.04801741987466812,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 71520
+    },
+    {
+      "epoch": 7.350770811921891,
+      "grad_norm": 0.04515586793422699,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 71523
+    },
+    {
+      "epoch": 7.351079136690648,
+      "grad_norm": 0.06455957144498825,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 71526
+    },
+    {
+      "epoch": 7.351387461459404,
+      "grad_norm": 0.032415423542261124,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 71529
+    },
+    {
+      "epoch": 7.35169578622816,
+      "grad_norm": 0.08034674823284149,
+      "learning_rate": 0.01,
+      "loss": 1.9948,
+      "step": 71532
+    },
+    {
+      "epoch": 7.3520041109969165,
+      "grad_norm": 0.08528207242488861,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 71535
+    },
+    {
+      "epoch": 7.352312435765673,
+      "grad_norm": 0.04610508680343628,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 71538
+    },
+    {
+      "epoch": 7.35262076053443,
+      "grad_norm": 0.09876418858766556,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 71541
+    },
+    {
+      "epoch": 7.352929085303186,
+      "grad_norm": 0.04421771690249443,
+      "learning_rate": 0.01,
+      "loss": 1.9937,
+      "step": 71544
+    },
+    {
+      "epoch": 7.353237410071943,
+      "grad_norm": 0.09357340633869171,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 71547
+    },
+    {
+      "epoch": 7.353545734840699,
+      "grad_norm": 0.07575297355651855,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 71550
+    },
+    {
+      "epoch": 7.353854059609455,
+      "grad_norm": 0.07043871283531189,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 71553
+    },
+    {
+      "epoch": 7.354162384378212,
+      "grad_norm": 0.08759008347988129,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 71556
+    },
+    {
+      "epoch": 7.354470709146968,
+      "grad_norm": 0.07198178768157959,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 71559
+    },
+    {
+      "epoch": 7.354779033915724,
+      "grad_norm": 0.039580024778842926,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 71562
+    },
+    {
+      "epoch": 7.355087358684481,
+      "grad_norm": 0.13564977049827576,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 71565
+    },
+    {
+      "epoch": 7.355395683453238,
+      "grad_norm": 0.07854889333248138,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 71568
+    },
+    {
+      "epoch": 7.355704008221994,
+      "grad_norm": 0.057042330503463745,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 71571
+    },
+    {
+      "epoch": 7.35601233299075,
+      "grad_norm": 0.15226583182811737,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 71574
+    },
+    {
+      "epoch": 7.356320657759507,
+      "grad_norm": 0.07464998960494995,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 71577
+    },
+    {
+      "epoch": 7.356628982528263,
+      "grad_norm": 0.04508376121520996,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 71580
+    },
+    {
+      "epoch": 7.356937307297019,
+      "grad_norm": 0.03513191640377045,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 71583
+    },
+    {
+      "epoch": 7.357245632065776,
+      "grad_norm": 0.048798494040966034,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 71586
+    },
+    {
+      "epoch": 7.357553956834533,
+      "grad_norm": 0.053385645151138306,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 71589
+    },
+    {
+      "epoch": 7.357862281603289,
+      "grad_norm": 0.0522787980735302,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 71592
+    },
+    {
+      "epoch": 7.358170606372045,
+      "grad_norm": 0.07104373723268509,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 71595
+    },
+    {
+      "epoch": 7.358478931140802,
+      "grad_norm": 0.08457489311695099,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 71598
+    },
+    {
+      "epoch": 7.358787255909558,
+      "grad_norm": 0.07561776041984558,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 71601
+    },
+    {
+      "epoch": 7.359095580678314,
+      "grad_norm": 0.07273251563310623,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 71604
+    },
+    {
+      "epoch": 7.359403905447071,
+      "grad_norm": 0.14364686608314514,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 71607
+    },
+    {
+      "epoch": 7.359712230215827,
+      "grad_norm": 0.06388013809919357,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 71610
+    },
+    {
+      "epoch": 7.360020554984584,
+      "grad_norm": 0.03216857835650444,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 71613
+    },
+    {
+      "epoch": 7.36032887975334,
+      "grad_norm": 0.05028613656759262,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 71616
+    },
+    {
+      "epoch": 7.360637204522097,
+      "grad_norm": 0.06463950127363205,
+      "learning_rate": 0.01,
+      "loss": 2.0118,
+      "step": 71619
+    },
+    {
+      "epoch": 7.360945529290853,
+      "grad_norm": 0.05385703966021538,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 71622
+    },
+    {
+      "epoch": 7.361253854059609,
+      "grad_norm": 0.05802121013402939,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 71625
+    },
+    {
+      "epoch": 7.361562178828366,
+      "grad_norm": 0.045689478516578674,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 71628
+    },
+    {
+      "epoch": 7.361870503597122,
+      "grad_norm": 0.039995331317186356,
+      "learning_rate": 0.01,
+      "loss": 1.9995,
+      "step": 71631
+    },
+    {
+      "epoch": 7.362178828365879,
+      "grad_norm": 0.11462247371673584,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 71634
+    },
+    {
+      "epoch": 7.362487153134635,
+      "grad_norm": 0.0777439996600151,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 71637
+    },
+    {
+      "epoch": 7.362795477903392,
+      "grad_norm": 0.045238759368658066,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 71640
+    },
+    {
+      "epoch": 7.363103802672148,
+      "grad_norm": 0.02978505939245224,
+      "learning_rate": 0.01,
+      "loss": 2.0006,
+      "step": 71643
+    },
+    {
+      "epoch": 7.363412127440904,
+      "grad_norm": 0.058491241186857224,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 71646
+    },
+    {
+      "epoch": 7.363720452209661,
+      "grad_norm": 0.037735093384981155,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 71649
+    },
+    {
+      "epoch": 7.364028776978417,
+      "grad_norm": 0.030737249180674553,
+      "learning_rate": 0.01,
+      "loss": 1.9954,
+      "step": 71652
+    },
+    {
+      "epoch": 7.364337101747173,
+      "grad_norm": 0.03551234304904938,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 71655
+    },
+    {
+      "epoch": 7.3646454265159305,
+      "grad_norm": 0.03984438255429268,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 71658
+    },
+    {
+      "epoch": 7.364953751284687,
+      "grad_norm": 0.16038914024829865,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 71661
+    },
+    {
+      "epoch": 7.365262076053443,
+      "grad_norm": 0.04318477585911751,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 71664
+    },
+    {
+      "epoch": 7.365570400822199,
+      "grad_norm": 0.04927264153957367,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 71667
+    },
+    {
+      "epoch": 7.365878725590956,
+      "grad_norm": 0.052578482776880264,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 71670
+    },
+    {
+      "epoch": 7.366187050359712,
+      "grad_norm": 0.04304051399230957,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 71673
+    },
+    {
+      "epoch": 7.366495375128468,
+      "grad_norm": 0.11516401916742325,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 71676
+    },
+    {
+      "epoch": 7.3668036998972255,
+      "grad_norm": 0.07391729950904846,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 71679
+    },
+    {
+      "epoch": 7.367112024665982,
+      "grad_norm": 0.04147100821137428,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 71682
+    },
+    {
+      "epoch": 7.367420349434738,
+      "grad_norm": 0.0550803616642952,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 71685
+    },
+    {
+      "epoch": 7.367728674203494,
+      "grad_norm": 0.06658782064914703,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 71688
+    },
+    {
+      "epoch": 7.368036998972251,
+      "grad_norm": 0.07493577152490616,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 71691
+    },
+    {
+      "epoch": 7.368345323741007,
+      "grad_norm": 0.06494047492742538,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 71694
+    },
+    {
+      "epoch": 7.368653648509763,
+      "grad_norm": 0.044212259352207184,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 71697
+    },
+    {
+      "epoch": 7.36896197327852,
+      "grad_norm": 0.04893089085817337,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 71700
+    },
+    {
+      "epoch": 7.369270298047277,
+      "grad_norm": 0.06412500143051147,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 71703
+    },
+    {
+      "epoch": 7.369578622816033,
+      "grad_norm": 0.11510922014713287,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 71706
+    },
+    {
+      "epoch": 7.3698869475847895,
+      "grad_norm": 0.06492109596729279,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 71709
+    },
+    {
+      "epoch": 7.370195272353546,
+      "grad_norm": 0.07777494192123413,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 71712
+    },
+    {
+      "epoch": 7.370503597122302,
+      "grad_norm": 0.06416970491409302,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 71715
+    },
+    {
+      "epoch": 7.370811921891058,
+      "grad_norm": 0.11529435962438583,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 71718
+    },
+    {
+      "epoch": 7.371120246659815,
+      "grad_norm": 0.033961158245801926,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 71721
+    },
+    {
+      "epoch": 7.371428571428572,
+      "grad_norm": 0.09541164338588715,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 71724
+    },
+    {
+      "epoch": 7.371736896197328,
+      "grad_norm": 0.05110830441117287,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 71727
+    },
+    {
+      "epoch": 7.3720452209660845,
+      "grad_norm": 0.08816475421190262,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 71730
+    },
+    {
+      "epoch": 7.372353545734841,
+      "grad_norm": 0.11616508662700653,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 71733
+    },
+    {
+      "epoch": 7.372661870503597,
+      "grad_norm": 0.06593319773674011,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 71736
+    },
+    {
+      "epoch": 7.372970195272353,
+      "grad_norm": 0.08161871880292892,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 71739
+    },
+    {
+      "epoch": 7.37327852004111,
+      "grad_norm": 0.04027361795306206,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 71742
+    },
+    {
+      "epoch": 7.373586844809866,
+      "grad_norm": 0.04333282262086868,
+      "learning_rate": 0.01,
+      "loss": 1.9982,
+      "step": 71745
+    },
+    {
+      "epoch": 7.373895169578623,
+      "grad_norm": 0.04270986467599869,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 71748
+    },
+    {
+      "epoch": 7.3742034943473795,
+      "grad_norm": 0.09752412140369415,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 71751
+    },
+    {
+      "epoch": 7.374511819116136,
+      "grad_norm": 0.06835508346557617,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 71754
+    },
+    {
+      "epoch": 7.374820143884892,
+      "grad_norm": 0.07079548388719559,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 71757
+    },
+    {
+      "epoch": 7.3751284686536485,
+      "grad_norm": 0.06948207318782806,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 71760
+    },
+    {
+      "epoch": 7.375436793422405,
+      "grad_norm": 0.03838725760579109,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 71763
+    },
+    {
+      "epoch": 7.375745118191161,
+      "grad_norm": 0.07975951582193375,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 71766
+    },
+    {
+      "epoch": 7.376053442959917,
+      "grad_norm": 0.06829675287008286,
+      "learning_rate": 0.01,
+      "loss": 2.0081,
+      "step": 71769
+    },
+    {
+      "epoch": 7.376361767728675,
+      "grad_norm": 0.0858384519815445,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 71772
+    },
+    {
+      "epoch": 7.376670092497431,
+      "grad_norm": 0.037788160145282745,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 71775
+    },
+    {
+      "epoch": 7.376978417266187,
+      "grad_norm": 0.040428534150123596,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 71778
+    },
+    {
+      "epoch": 7.3772867420349435,
+      "grad_norm": 0.041471801698207855,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 71781
+    },
+    {
+      "epoch": 7.3775950668037,
+      "grad_norm": 0.06048744544386864,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 71784
+    },
+    {
+      "epoch": 7.377903391572456,
+      "grad_norm": 0.0954107865691185,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 71787
+    },
+    {
+      "epoch": 7.378211716341212,
+      "grad_norm": 0.11208479851484299,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 71790
+    },
+    {
+      "epoch": 7.378520041109969,
+      "grad_norm": 0.037479761987924576,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 71793
+    },
+    {
+      "epoch": 7.378828365878726,
+      "grad_norm": 0.047237321734428406,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 71796
+    },
+    {
+      "epoch": 7.379136690647482,
+      "grad_norm": 0.05881095305085182,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 71799
+    },
+    {
+      "epoch": 7.3794450154162385,
+      "grad_norm": 0.054790887981653214,
+      "learning_rate": 0.01,
+      "loss": 2.01,
+      "step": 71802
+    },
+    {
+      "epoch": 7.379753340184995,
+      "grad_norm": 0.0610164999961853,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 71805
+    },
+    {
+      "epoch": 7.380061664953751,
+      "grad_norm": 0.04166460409760475,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 71808
+    },
+    {
+      "epoch": 7.3803699897225075,
+      "grad_norm": 0.058832067996263504,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 71811
+    },
+    {
+      "epoch": 7.380678314491264,
+      "grad_norm": 0.04992865025997162,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 71814
+    },
+    {
+      "epoch": 7.380986639260021,
+      "grad_norm": 0.11985593289136887,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 71817
+    },
+    {
+      "epoch": 7.381294964028777,
+      "grad_norm": 0.048121824860572815,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 71820
+    },
+    {
+      "epoch": 7.381603288797534,
+      "grad_norm": 0.0449887290596962,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 71823
+    },
+    {
+      "epoch": 7.38191161356629,
+      "grad_norm": 0.08304189890623093,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 71826
+    },
+    {
+      "epoch": 7.382219938335046,
+      "grad_norm": 0.06549306958913803,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 71829
+    },
+    {
+      "epoch": 7.3825282631038025,
+      "grad_norm": 0.05262019485235214,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 71832
+    },
+    {
+      "epoch": 7.382836587872559,
+      "grad_norm": 0.04085802286863327,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 71835
+    },
+    {
+      "epoch": 7.383144912641315,
+      "grad_norm": 0.03672410920262337,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 71838
+    },
+    {
+      "epoch": 7.383453237410072,
+      "grad_norm": 0.04606030136346817,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 71841
+    },
+    {
+      "epoch": 7.383761562178829,
+      "grad_norm": 0.11922565847635269,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 71844
+    },
+    {
+      "epoch": 7.384069886947585,
+      "grad_norm": 0.060678258538246155,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 71847
+    },
+    {
+      "epoch": 7.384378211716341,
+      "grad_norm": 0.11298535019159317,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 71850
+    },
+    {
+      "epoch": 7.3846865364850975,
+      "grad_norm": 0.03260503709316254,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 71853
+    },
+    {
+      "epoch": 7.384994861253854,
+      "grad_norm": 0.044581152498722076,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 71856
+    },
+    {
+      "epoch": 7.38530318602261,
+      "grad_norm": 0.044959817081689835,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 71859
+    },
+    {
+      "epoch": 7.385611510791367,
+      "grad_norm": 0.04784368351101875,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 71862
+    },
+    {
+      "epoch": 7.385919835560124,
+      "grad_norm": 0.08526644110679626,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 71865
+    },
+    {
+      "epoch": 7.38622816032888,
+      "grad_norm": 0.057515472173690796,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 71868
+    },
+    {
+      "epoch": 7.386536485097636,
+      "grad_norm": 0.04409664124250412,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 71871
+    },
+    {
+      "epoch": 7.386844809866393,
+      "grad_norm": 0.04344470053911209,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 71874
+    },
+    {
+      "epoch": 7.387153134635149,
+      "grad_norm": 0.03933620825409889,
+      "learning_rate": 0.01,
+      "loss": 1.9926,
+      "step": 71877
+    },
+    {
+      "epoch": 7.387461459403905,
+      "grad_norm": 0.1177898719906807,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 71880
+    },
+    {
+      "epoch": 7.3877697841726615,
+      "grad_norm": 0.045402709394693375,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 71883
+    },
+    {
+      "epoch": 7.388078108941419,
+      "grad_norm": 0.05361394211649895,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 71886
+    },
+    {
+      "epoch": 7.388386433710175,
+      "grad_norm": 0.05513688176870346,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 71889
+    },
+    {
+      "epoch": 7.388694758478931,
+      "grad_norm": 0.03191964328289032,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 71892
+    },
+    {
+      "epoch": 7.389003083247688,
+      "grad_norm": 0.05432026833295822,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 71895
+    },
+    {
+      "epoch": 7.389311408016444,
+      "grad_norm": 0.06853076070547104,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 71898
+    },
+    {
+      "epoch": 7.3896197327852,
+      "grad_norm": 0.07504719495773315,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 71901
+    },
+    {
+      "epoch": 7.3899280575539565,
+      "grad_norm": 0.08959703892469406,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 71904
+    },
+    {
+      "epoch": 7.390236382322713,
+      "grad_norm": 0.04118792340159416,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 71907
+    },
+    {
+      "epoch": 7.39054470709147,
+      "grad_norm": 0.06344828754663467,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 71910
+    },
+    {
+      "epoch": 7.390853031860226,
+      "grad_norm": 0.03989424929022789,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 71913
+    },
+    {
+      "epoch": 7.391161356628983,
+      "grad_norm": 0.07987693697214127,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 71916
+    },
+    {
+      "epoch": 7.391469681397739,
+      "grad_norm": 0.09878303110599518,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 71919
+    },
+    {
+      "epoch": 7.391778006166495,
+      "grad_norm": 0.15211869776248932,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 71922
+    },
+    {
+      "epoch": 7.392086330935252,
+      "grad_norm": 0.09302935749292374,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 71925
+    },
+    {
+      "epoch": 7.392394655704008,
+      "grad_norm": 0.08965945243835449,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 71928
+    },
+    {
+      "epoch": 7.392702980472764,
+      "grad_norm": 0.059964247047901154,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 71931
+    },
+    {
+      "epoch": 7.393011305241521,
+      "grad_norm": 0.03914433345198631,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 71934
+    },
+    {
+      "epoch": 7.393319630010278,
+      "grad_norm": 0.033087339252233505,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 71937
+    },
+    {
+      "epoch": 7.393627954779034,
+      "grad_norm": 0.03930596262216568,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 71940
+    },
+    {
+      "epoch": 7.39393627954779,
+      "grad_norm": 0.1557454764842987,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 71943
+    },
+    {
+      "epoch": 7.394244604316547,
+      "grad_norm": 0.12098176777362823,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 71946
+    },
+    {
+      "epoch": 7.394552929085303,
+      "grad_norm": 0.062015559524297714,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 71949
+    },
+    {
+      "epoch": 7.394861253854059,
+      "grad_norm": 0.051067329943180084,
+      "learning_rate": 0.01,
+      "loss": 2.0197,
+      "step": 71952
+    },
+    {
+      "epoch": 7.395169578622816,
+      "grad_norm": 0.05703721195459366,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 71955
+    },
+    {
+      "epoch": 7.395477903391573,
+      "grad_norm": 0.05166958272457123,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 71958
+    },
+    {
+      "epoch": 7.395786228160329,
+      "grad_norm": 0.06948833912611008,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 71961
+    },
+    {
+      "epoch": 7.396094552929085,
+      "grad_norm": 0.09943950176239014,
+      "learning_rate": 0.01,
+      "loss": 2.0107,
+      "step": 71964
+    },
+    {
+      "epoch": 7.396402877697842,
+      "grad_norm": 0.04081091284751892,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 71967
+    },
+    {
+      "epoch": 7.396711202466598,
+      "grad_norm": 0.08712594956159592,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 71970
+    },
+    {
+      "epoch": 7.397019527235354,
+      "grad_norm": 0.10174566507339478,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 71973
+    },
+    {
+      "epoch": 7.397327852004111,
+      "grad_norm": 0.08403330296278,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 71976
+    },
+    {
+      "epoch": 7.397636176772868,
+      "grad_norm": 0.07419370114803314,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 71979
+    },
+    {
+      "epoch": 7.397944501541624,
+      "grad_norm": 0.08721747249364853,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 71982
+    },
+    {
+      "epoch": 7.39825282631038,
+      "grad_norm": 0.07544389367103577,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 71985
+    },
+    {
+      "epoch": 7.398561151079137,
+      "grad_norm": 0.03575440123677254,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 71988
+    },
+    {
+      "epoch": 7.398869475847893,
+      "grad_norm": 0.04670301452279091,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 71991
+    },
+    {
+      "epoch": 7.399177800616649,
+      "grad_norm": 0.04096400365233421,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 71994
+    },
+    {
+      "epoch": 7.399486125385406,
+      "grad_norm": 0.04938502982258797,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 71997
+    },
+    {
+      "epoch": 7.399794450154163,
+      "grad_norm": 0.13924731314182281,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 72000
+    },
+    {
+      "epoch": 7.400102774922919,
+      "grad_norm": 0.10529632121324539,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 72003
+    },
+    {
+      "epoch": 7.400411099691675,
+      "grad_norm": 0.06417892873287201,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 72006
+    },
+    {
+      "epoch": 7.400719424460432,
+      "grad_norm": 0.04323699697852135,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 72009
+    },
+    {
+      "epoch": 7.401027749229188,
+      "grad_norm": 0.04034925252199173,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 72012
+    },
+    {
+      "epoch": 7.401336073997944,
+      "grad_norm": 0.12776517868041992,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 72015
+    },
+    {
+      "epoch": 7.401644398766701,
+      "grad_norm": 0.12993237376213074,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 72018
+    },
+    {
+      "epoch": 7.401952723535457,
+      "grad_norm": 0.09225061535835266,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 72021
+    },
+    {
+      "epoch": 7.402261048304214,
+      "grad_norm": 0.0604284442961216,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 72024
+    },
+    {
+      "epoch": 7.4025693730729705,
+      "grad_norm": 0.042524028569459915,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 72027
+    },
+    {
+      "epoch": 7.402877697841727,
+      "grad_norm": 0.0651375949382782,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 72030
+    },
+    {
+      "epoch": 7.403186022610483,
+      "grad_norm": 0.04578625038266182,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 72033
+    },
+    {
+      "epoch": 7.403494347379239,
+      "grad_norm": 0.05088494345545769,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 72036
+    },
+    {
+      "epoch": 7.403802672147996,
+      "grad_norm": 0.04542406648397446,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 72039
+    },
+    {
+      "epoch": 7.404110996916752,
+      "grad_norm": 0.035000525414943695,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 72042
+    },
+    {
+      "epoch": 7.404419321685509,
+      "grad_norm": 0.10675634443759918,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 72045
+    },
+    {
+      "epoch": 7.4047276464542655,
+      "grad_norm": 0.071378692984581,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 72048
+    },
+    {
+      "epoch": 7.405035971223022,
+      "grad_norm": 0.047410156577825546,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 72051
+    },
+    {
+      "epoch": 7.405344295991778,
+      "grad_norm": 0.04333693906664848,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 72054
+    },
+    {
+      "epoch": 7.405652620760534,
+      "grad_norm": 0.038545072078704834,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 72057
+    },
+    {
+      "epoch": 7.405960945529291,
+      "grad_norm": 0.16361141204833984,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 72060
+    },
+    {
+      "epoch": 7.406269270298047,
+      "grad_norm": 0.11350101232528687,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 72063
+    },
+    {
+      "epoch": 7.406577595066803,
+      "grad_norm": 0.06721406430006027,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 72066
+    },
+    {
+      "epoch": 7.40688591983556,
+      "grad_norm": 0.061075255274772644,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 72069
+    },
+    {
+      "epoch": 7.407194244604317,
+      "grad_norm": 0.048585545271635056,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 72072
+    },
+    {
+      "epoch": 7.407502569373073,
+      "grad_norm": 0.04021257162094116,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 72075
+    },
+    {
+      "epoch": 7.4078108941418295,
+      "grad_norm": 0.054019588977098465,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 72078
+    },
+    {
+      "epoch": 7.408119218910586,
+      "grad_norm": 0.05107961595058441,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 72081
+    },
+    {
+      "epoch": 7.408427543679342,
+      "grad_norm": 0.10263427346944809,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 72084
+    },
+    {
+      "epoch": 7.408735868448098,
+      "grad_norm": 0.12283501029014587,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 72087
+    },
+    {
+      "epoch": 7.409044193216855,
+      "grad_norm": 0.058549631386995316,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 72090
+    },
+    {
+      "epoch": 7.409352517985612,
+      "grad_norm": 0.05259247124195099,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 72093
+    },
+    {
+      "epoch": 7.409660842754368,
+      "grad_norm": 0.04460529983043671,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 72096
+    },
+    {
+      "epoch": 7.4099691675231245,
+      "grad_norm": 0.0474291630089283,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 72099
+    },
+    {
+      "epoch": 7.410277492291881,
+      "grad_norm": 0.049115218222141266,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 72102
+    },
+    {
+      "epoch": 7.410585817060637,
+      "grad_norm": 0.06523936986923218,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 72105
+    },
+    {
+      "epoch": 7.410894141829393,
+      "grad_norm": 0.059646543115377426,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 72108
+    },
+    {
+      "epoch": 7.41120246659815,
+      "grad_norm": 0.04670571908354759,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 72111
+    },
+    {
+      "epoch": 7.411510791366906,
+      "grad_norm": 0.1110573336482048,
+      "learning_rate": 0.01,
+      "loss": 1.9958,
+      "step": 72114
+    },
+    {
+      "epoch": 7.411819116135663,
+      "grad_norm": 0.04579848423600197,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 72117
+    },
+    {
+      "epoch": 7.4121274409044196,
+      "grad_norm": 0.12220747023820877,
+      "learning_rate": 0.01,
+      "loss": 2.0113,
+      "step": 72120
+    },
+    {
+      "epoch": 7.412435765673176,
+      "grad_norm": 0.08545442670583725,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 72123
+    },
+    {
+      "epoch": 7.412744090441932,
+      "grad_norm": 0.05775977298617363,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 72126
+    },
+    {
+      "epoch": 7.4130524152106885,
+      "grad_norm": 0.03229491040110588,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 72129
+    },
+    {
+      "epoch": 7.413360739979445,
+      "grad_norm": 0.03648471459746361,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 72132
+    },
+    {
+      "epoch": 7.413669064748201,
+      "grad_norm": 0.06855278462171555,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 72135
+    },
+    {
+      "epoch": 7.413977389516958,
+      "grad_norm": 0.12804359197616577,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 72138
+    },
+    {
+      "epoch": 7.414285714285715,
+      "grad_norm": 0.06988925486803055,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 72141
+    },
+    {
+      "epoch": 7.414594039054471,
+      "grad_norm": 0.08092012256383896,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 72144
+    },
+    {
+      "epoch": 7.414902363823227,
+      "grad_norm": 0.062308866530656815,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 72147
+    },
+    {
+      "epoch": 7.4152106885919835,
+      "grad_norm": 0.1187475249171257,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 72150
+    },
+    {
+      "epoch": 7.41551901336074,
+      "grad_norm": 0.038580019026994705,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 72153
+    },
+    {
+      "epoch": 7.415827338129496,
+      "grad_norm": 0.06444518268108368,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 72156
+    },
+    {
+      "epoch": 7.416135662898252,
+      "grad_norm": 0.09006829559803009,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 72159
+    },
+    {
+      "epoch": 7.41644398766701,
+      "grad_norm": 0.1387302577495575,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 72162
+    },
+    {
+      "epoch": 7.416752312435766,
+      "grad_norm": 0.12567099928855896,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 72165
+    },
+    {
+      "epoch": 7.417060637204522,
+      "grad_norm": 0.046082064509391785,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 72168
+    },
+    {
+      "epoch": 7.4173689619732786,
+      "grad_norm": 0.044009629637002945,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 72171
+    },
+    {
+      "epoch": 7.417677286742035,
+      "grad_norm": 0.07774317264556885,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 72174
+    },
+    {
+      "epoch": 7.417985611510791,
+      "grad_norm": 0.11599218100309372,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 72177
+    },
+    {
+      "epoch": 7.4182939362795475,
+      "grad_norm": 0.03975779935717583,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 72180
+    },
+    {
+      "epoch": 7.418602261048305,
+      "grad_norm": 0.06963899731636047,
+      "learning_rate": 0.01,
+      "loss": 1.9975,
+      "step": 72183
+    },
+    {
+      "epoch": 7.418910585817061,
+      "grad_norm": 0.0915340930223465,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 72186
+    },
+    {
+      "epoch": 7.419218910585817,
+      "grad_norm": 0.057447649538517,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 72189
+    },
+    {
+      "epoch": 7.419527235354574,
+      "grad_norm": 0.037418052554130554,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 72192
+    },
+    {
+      "epoch": 7.41983556012333,
+      "grad_norm": 0.03583716228604317,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 72195
+    },
+    {
+      "epoch": 7.420143884892086,
+      "grad_norm": 0.04073040559887886,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 72198
+    },
+    {
+      "epoch": 7.4204522096608425,
+      "grad_norm": 0.1194196417927742,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 72201
+    },
+    {
+      "epoch": 7.420760534429599,
+      "grad_norm": 0.07829037308692932,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 72204
+    },
+    {
+      "epoch": 7.421068859198356,
+      "grad_norm": 0.05158103257417679,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 72207
+    },
+    {
+      "epoch": 7.421377183967112,
+      "grad_norm": 0.13578635454177856,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 72210
+    },
+    {
+      "epoch": 7.421685508735869,
+      "grad_norm": 0.07069911062717438,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 72213
+    },
+    {
+      "epoch": 7.421993833504625,
+      "grad_norm": 0.04424047842621803,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 72216
+    },
+    {
+      "epoch": 7.422302158273381,
+      "grad_norm": 0.03345247730612755,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 72219
+    },
+    {
+      "epoch": 7.4226104830421376,
+      "grad_norm": 0.04876269772648811,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 72222
+    },
+    {
+      "epoch": 7.422918807810894,
+      "grad_norm": 0.0414804071187973,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 72225
+    },
+    {
+      "epoch": 7.42322713257965,
+      "grad_norm": 0.10436029732227325,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 72228
+    },
+    {
+      "epoch": 7.423535457348407,
+      "grad_norm": 0.08055151253938675,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 72231
+    },
+    {
+      "epoch": 7.423843782117164,
+      "grad_norm": 0.11014366894960403,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 72234
+    },
+    {
+      "epoch": 7.42415210688592,
+      "grad_norm": 0.06515113264322281,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 72237
+    },
+    {
+      "epoch": 7.424460431654676,
+      "grad_norm": 0.04275600612163544,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 72240
+    },
+    {
+      "epoch": 7.424768756423433,
+      "grad_norm": 0.1309288591146469,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 72243
+    },
+    {
+      "epoch": 7.425077081192189,
+      "grad_norm": 0.04589364305138588,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 72246
+    },
+    {
+      "epoch": 7.425385405960945,
+      "grad_norm": 0.06936043500900269,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 72249
+    },
+    {
+      "epoch": 7.4256937307297015,
+      "grad_norm": 0.06822799891233444,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 72252
+    },
+    {
+      "epoch": 7.426002055498459,
+      "grad_norm": 0.04731842502951622,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 72255
+    },
+    {
+      "epoch": 7.426310380267215,
+      "grad_norm": 0.14833036065101624,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 72258
+    },
+    {
+      "epoch": 7.426618705035971,
+      "grad_norm": 0.05417739227414131,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 72261
+    },
+    {
+      "epoch": 7.426927029804728,
+      "grad_norm": 0.06367053091526031,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 72264
+    },
+    {
+      "epoch": 7.427235354573484,
+      "grad_norm": 0.03428191319108009,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 72267
+    },
+    {
+      "epoch": 7.42754367934224,
+      "grad_norm": 0.044105369597673416,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 72270
+    },
+    {
+      "epoch": 7.4278520041109966,
+      "grad_norm": 0.08640552312135696,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 72273
+    },
+    {
+      "epoch": 7.428160328879754,
+      "grad_norm": 0.13555385172367096,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 72276
+    },
+    {
+      "epoch": 7.42846865364851,
+      "grad_norm": 0.06566069275140762,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 72279
+    },
+    {
+      "epoch": 7.428776978417266,
+      "grad_norm": 0.1009216234087944,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 72282
+    },
+    {
+      "epoch": 7.429085303186023,
+      "grad_norm": 0.06078832969069481,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 72285
+    },
+    {
+      "epoch": 7.429393627954779,
+      "grad_norm": 0.04854779317975044,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 72288
+    },
+    {
+      "epoch": 7.429701952723535,
+      "grad_norm": 0.06042282283306122,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 72291
+    },
+    {
+      "epoch": 7.430010277492292,
+      "grad_norm": 0.07528941333293915,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 72294
+    },
+    {
+      "epoch": 7.430318602261048,
+      "grad_norm": 0.048977117985486984,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 72297
+    },
+    {
+      "epoch": 7.430626927029805,
+      "grad_norm": 0.03058001957833767,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 72300
+    },
+    {
+      "epoch": 7.430935251798561,
+      "grad_norm": 0.031676169484853745,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 72303
+    },
+    {
+      "epoch": 7.431243576567318,
+      "grad_norm": 0.04474344104528427,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 72306
+    },
+    {
+      "epoch": 7.431551901336074,
+      "grad_norm": 0.04538916051387787,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 72309
+    },
+    {
+      "epoch": 7.43186022610483,
+      "grad_norm": 0.0947703942656517,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 72312
+    },
+    {
+      "epoch": 7.432168550873587,
+      "grad_norm": 0.05523570254445076,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 72315
+    },
+    {
+      "epoch": 7.432476875642343,
+      "grad_norm": 0.09945077449083328,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 72318
+    },
+    {
+      "epoch": 7.4327852004111,
+      "grad_norm": 0.11858788877725601,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 72321
+    },
+    {
+      "epoch": 7.433093525179856,
+      "grad_norm": 0.056699126958847046,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 72324
+    },
+    {
+      "epoch": 7.433401849948613,
+      "grad_norm": 0.03996944800019264,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 72327
+    },
+    {
+      "epoch": 7.433710174717369,
+      "grad_norm": 0.04307657852768898,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 72330
+    },
+    {
+      "epoch": 7.434018499486125,
+      "grad_norm": 0.03870538994669914,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 72333
+    },
+    {
+      "epoch": 7.434326824254882,
+      "grad_norm": 0.04004659503698349,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 72336
+    },
+    {
+      "epoch": 7.434635149023638,
+      "grad_norm": 0.10981098562479019,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 72339
+    },
+    {
+      "epoch": 7.434943473792394,
+      "grad_norm": 0.06930761784315109,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 72342
+    },
+    {
+      "epoch": 7.4352517985611515,
+      "grad_norm": 0.09295953810214996,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 72345
+    },
+    {
+      "epoch": 7.435560123329908,
+      "grad_norm": 0.07768440246582031,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 72348
+    },
+    {
+      "epoch": 7.435868448098664,
+      "grad_norm": 0.059423208236694336,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 72351
+    },
+    {
+      "epoch": 7.43617677286742,
+      "grad_norm": 0.06254580616950989,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 72354
+    },
+    {
+      "epoch": 7.436485097636177,
+      "grad_norm": 0.05663672834634781,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 72357
+    },
+    {
+      "epoch": 7.436793422404933,
+      "grad_norm": 0.09405243396759033,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 72360
+    },
+    {
+      "epoch": 7.437101747173689,
+      "grad_norm": 0.04204673692584038,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 72363
+    },
+    {
+      "epoch": 7.437410071942446,
+      "grad_norm": 0.04828297346830368,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 72366
+    },
+    {
+      "epoch": 7.437718396711203,
+      "grad_norm": 0.12901097536087036,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 72369
+    },
+    {
+      "epoch": 7.438026721479959,
+      "grad_norm": 0.04471548646688461,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 72372
+    },
+    {
+      "epoch": 7.438335046248715,
+      "grad_norm": 0.10098598897457123,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 72375
+    },
+    {
+      "epoch": 7.438643371017472,
+      "grad_norm": 0.06057806685566902,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 72378
+    },
+    {
+      "epoch": 7.438951695786228,
+      "grad_norm": 0.07276615500450134,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 72381
+    },
+    {
+      "epoch": 7.439260020554984,
+      "grad_norm": 0.06687156856060028,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 72384
+    },
+    {
+      "epoch": 7.439568345323741,
+      "grad_norm": 0.10728184133768082,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 72387
+    },
+    {
+      "epoch": 7.439876670092497,
+      "grad_norm": 0.10064877569675446,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 72390
+    },
+    {
+      "epoch": 7.440184994861254,
+      "grad_norm": 0.07631184905767441,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 72393
+    },
+    {
+      "epoch": 7.4404933196300105,
+      "grad_norm": 0.07702022790908813,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 72396
+    },
+    {
+      "epoch": 7.440801644398767,
+      "grad_norm": 0.0484774112701416,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 72399
+    },
+    {
+      "epoch": 7.441109969167523,
+      "grad_norm": 0.03002420999109745,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 72402
+    },
+    {
+      "epoch": 7.441418293936279,
+      "grad_norm": 0.03068717196583748,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 72405
+    },
+    {
+      "epoch": 7.441726618705036,
+      "grad_norm": 0.03828945755958557,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 72408
+    },
+    {
+      "epoch": 7.442034943473792,
+      "grad_norm": 0.11569713056087494,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 72411
+    },
+    {
+      "epoch": 7.442343268242549,
+      "grad_norm": 0.06883276998996735,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 72414
+    },
+    {
+      "epoch": 7.4426515930113055,
+      "grad_norm": 0.12915529310703278,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 72417
+    },
+    {
+      "epoch": 7.442959917780062,
+      "grad_norm": 0.05600148066878319,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 72420
+    },
+    {
+      "epoch": 7.443268242548818,
+      "grad_norm": 0.061895426362752914,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 72423
+    },
+    {
+      "epoch": 7.443576567317574,
+      "grad_norm": 0.04677951708436012,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 72426
+    },
+    {
+      "epoch": 7.443884892086331,
+      "grad_norm": 0.09104373306035995,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 72429
+    },
+    {
+      "epoch": 7.444193216855087,
+      "grad_norm": 0.04269783943891525,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 72432
+    },
+    {
+      "epoch": 7.444501541623843,
+      "grad_norm": 0.04557257890701294,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 72435
+    },
+    {
+      "epoch": 7.444809866392601,
+      "grad_norm": 0.045879703015089035,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 72438
+    },
+    {
+      "epoch": 7.445118191161357,
+      "grad_norm": 0.09514186531305313,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 72441
+    },
+    {
+      "epoch": 7.445426515930113,
+      "grad_norm": 0.07008050382137299,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 72444
+    },
+    {
+      "epoch": 7.4457348406988695,
+      "grad_norm": 0.08302075415849686,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 72447
+    },
+    {
+      "epoch": 7.446043165467626,
+      "grad_norm": 0.052289433777332306,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 72450
+    },
+    {
+      "epoch": 7.446351490236382,
+      "grad_norm": 0.04208613932132721,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 72453
+    },
+    {
+      "epoch": 7.446659815005138,
+      "grad_norm": 0.0372624509036541,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 72456
+    },
+    {
+      "epoch": 7.446968139773896,
+      "grad_norm": 0.035864707082509995,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 72459
+    },
+    {
+      "epoch": 7.447276464542652,
+      "grad_norm": 0.03874954208731651,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 72462
+    },
+    {
+      "epoch": 7.447584789311408,
+      "grad_norm": 0.09163645654916763,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 72465
+    },
+    {
+      "epoch": 7.4478931140801645,
+      "grad_norm": 0.067960225045681,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 72468
+    },
+    {
+      "epoch": 7.448201438848921,
+      "grad_norm": 0.17044830322265625,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 72471
+    },
+    {
+      "epoch": 7.448509763617677,
+      "grad_norm": 0.0597783625125885,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 72474
+    },
+    {
+      "epoch": 7.4488180883864334,
+      "grad_norm": 0.08422503620386124,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 72477
+    },
+    {
+      "epoch": 7.44912641315519,
+      "grad_norm": 0.09853143244981766,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 72480
+    },
+    {
+      "epoch": 7.449434737923947,
+      "grad_norm": 0.09409277886152267,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 72483
+    },
+    {
+      "epoch": 7.449743062692703,
+      "grad_norm": 0.04450713470578194,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 72486
+    },
+    {
+      "epoch": 7.45005138746146,
+      "grad_norm": 0.11108853667974472,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 72489
+    },
+    {
+      "epoch": 7.450359712230216,
+      "grad_norm": 0.09322323650121689,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 72492
+    },
+    {
+      "epoch": 7.450668036998972,
+      "grad_norm": 0.11880034953355789,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 72495
+    },
+    {
+      "epoch": 7.4509763617677285,
+      "grad_norm": 0.06539753824472427,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 72498
+    },
+    {
+      "epoch": 7.451284686536485,
+      "grad_norm": 0.04002269357442856,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 72501
+    },
+    {
+      "epoch": 7.451593011305242,
+      "grad_norm": 0.0480516254901886,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 72504
+    },
+    {
+      "epoch": 7.451901336073998,
+      "grad_norm": 0.09114290028810501,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 72507
+    },
+    {
+      "epoch": 7.452209660842755,
+      "grad_norm": 0.057204026728868484,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 72510
+    },
+    {
+      "epoch": 7.452517985611511,
+      "grad_norm": 0.05074182525277138,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 72513
+    },
+    {
+      "epoch": 7.452826310380267,
+      "grad_norm": 0.09631665050983429,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 72516
+    },
+    {
+      "epoch": 7.4531346351490235,
+      "grad_norm": 0.04635360836982727,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 72519
+    },
+    {
+      "epoch": 7.45344295991778,
+      "grad_norm": 0.09677247703075409,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 72522
+    },
+    {
+      "epoch": 7.453751284686536,
+      "grad_norm": 0.05190379172563553,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 72525
+    },
+    {
+      "epoch": 7.4540596094552924,
+      "grad_norm": 0.08189938962459564,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 72528
+    },
+    {
+      "epoch": 7.45436793422405,
+      "grad_norm": 0.03724074363708496,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 72531
+    },
+    {
+      "epoch": 7.454676258992806,
+      "grad_norm": 0.07144693285226822,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 72534
+    },
+    {
+      "epoch": 7.454984583761562,
+      "grad_norm": 0.12171963602304459,
+      "learning_rate": 0.01,
+      "loss": 2.0007,
+      "step": 72537
+    },
+    {
+      "epoch": 7.455292908530319,
+      "grad_norm": 0.09564460813999176,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 72540
+    },
+    {
+      "epoch": 7.455601233299075,
+      "grad_norm": 0.04494168609380722,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 72543
+    },
+    {
+      "epoch": 7.455909558067831,
+      "grad_norm": 0.049825269728899,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 72546
+    },
+    {
+      "epoch": 7.4562178828365875,
+      "grad_norm": 0.09712143987417221,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 72549
+    },
+    {
+      "epoch": 7.456526207605345,
+      "grad_norm": 0.10748656839132309,
+      "learning_rate": 0.01,
+      "loss": 2.0011,
+      "step": 72552
+    },
+    {
+      "epoch": 7.456834532374101,
+      "grad_norm": 0.06327345967292786,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 72555
+    },
+    {
+      "epoch": 7.457142857142857,
+      "grad_norm": 0.11034796386957169,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 72558
+    },
+    {
+      "epoch": 7.457451181911614,
+      "grad_norm": 0.11681102961301804,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 72561
+    },
+    {
+      "epoch": 7.45775950668037,
+      "grad_norm": 0.11021427065134048,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 72564
+    },
+    {
+      "epoch": 7.458067831449126,
+      "grad_norm": 0.0419340580701828,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 72567
+    },
+    {
+      "epoch": 7.4583761562178825,
+      "grad_norm": 0.04036851227283478,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 72570
+    },
+    {
+      "epoch": 7.458684480986639,
+      "grad_norm": 0.049127209931612015,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 72573
+    },
+    {
+      "epoch": 7.458992805755396,
+      "grad_norm": 0.04860240966081619,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 72576
+    },
+    {
+      "epoch": 7.459301130524152,
+      "grad_norm": 0.08606428653001785,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 72579
+    },
+    {
+      "epoch": 7.459609455292909,
+      "grad_norm": 0.06198833882808685,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 72582
+    },
+    {
+      "epoch": 7.459917780061665,
+      "grad_norm": 0.049000997096300125,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 72585
+    },
+    {
+      "epoch": 7.460226104830421,
+      "grad_norm": 0.061215490102767944,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 72588
+    },
+    {
+      "epoch": 7.460534429599178,
+      "grad_norm": 0.048497434705495834,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 72591
+    },
+    {
+      "epoch": 7.460842754367934,
+      "grad_norm": 0.04738443344831467,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 72594
+    },
+    {
+      "epoch": 7.461151079136691,
+      "grad_norm": 0.10871072858572006,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 72597
+    },
+    {
+      "epoch": 7.461459403905447,
+      "grad_norm": 0.08442126959562302,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 72600
+    },
+    {
+      "epoch": 7.461767728674204,
+      "grad_norm": 0.11305920034646988,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 72603
+    },
+    {
+      "epoch": 7.46207605344296,
+      "grad_norm": 0.14034248888492584,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 72606
+    },
+    {
+      "epoch": 7.462384378211716,
+      "grad_norm": 0.05142485722899437,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 72609
+    },
+    {
+      "epoch": 7.462692702980473,
+      "grad_norm": 0.03771292045712471,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 72612
+    },
+    {
+      "epoch": 7.463001027749229,
+      "grad_norm": 0.040064070373773575,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 72615
+    },
+    {
+      "epoch": 7.463309352517985,
+      "grad_norm": 0.041471678763628006,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 72618
+    },
+    {
+      "epoch": 7.463617677286742,
+      "grad_norm": 0.047812748700380325,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 72621
+    },
+    {
+      "epoch": 7.463926002055499,
+      "grad_norm": 0.0737353190779686,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 72624
+    },
+    {
+      "epoch": 7.464234326824255,
+      "grad_norm": 0.05118251591920853,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 72627
+    },
+    {
+      "epoch": 7.464542651593011,
+      "grad_norm": 0.15912941098213196,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 72630
+    },
+    {
+      "epoch": 7.464850976361768,
+      "grad_norm": 0.05329582840204239,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 72633
+    },
+    {
+      "epoch": 7.465159301130524,
+      "grad_norm": 0.08995845913887024,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 72636
+    },
+    {
+      "epoch": 7.46546762589928,
+      "grad_norm": 0.0392749048769474,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 72639
+    },
+    {
+      "epoch": 7.4657759506680375,
+      "grad_norm": 0.031700219959020615,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 72642
+    },
+    {
+      "epoch": 7.466084275436794,
+      "grad_norm": 0.11098748445510864,
+      "learning_rate": 0.01,
+      "loss": 2.0008,
+      "step": 72645
+    },
+    {
+      "epoch": 7.46639260020555,
+      "grad_norm": 0.06115696206688881,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 72648
+    },
+    {
+      "epoch": 7.466700924974306,
+      "grad_norm": 0.0435514971613884,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 72651
+    },
+    {
+      "epoch": 7.467009249743063,
+      "grad_norm": 0.04764613136649132,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 72654
+    },
+    {
+      "epoch": 7.467317574511819,
+      "grad_norm": 0.04776312783360481,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 72657
+    },
+    {
+      "epoch": 7.467625899280575,
+      "grad_norm": 0.06918791681528091,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 72660
+    },
+    {
+      "epoch": 7.467934224049332,
+      "grad_norm": 0.05166517570614815,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 72663
+    },
+    {
+      "epoch": 7.468242548818089,
+      "grad_norm": 0.04008159786462784,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 72666
+    },
+    {
+      "epoch": 7.468550873586845,
+      "grad_norm": 0.046165283769369125,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 72669
+    },
+    {
+      "epoch": 7.468859198355601,
+      "grad_norm": 0.051842544227838516,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 72672
+    },
+    {
+      "epoch": 7.469167523124358,
+      "grad_norm": 0.04817887768149376,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 72675
+    },
+    {
+      "epoch": 7.469475847893114,
+      "grad_norm": 0.09956707060337067,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 72678
+    },
+    {
+      "epoch": 7.46978417266187,
+      "grad_norm": 0.07913441956043243,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 72681
+    },
+    {
+      "epoch": 7.470092497430627,
+      "grad_norm": 0.07743464410305023,
+      "learning_rate": 0.01,
+      "loss": 1.9883,
+      "step": 72684
+    },
+    {
+      "epoch": 7.470400822199383,
+      "grad_norm": 0.06038682535290718,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 72687
+    },
+    {
+      "epoch": 7.47070914696814,
+      "grad_norm": 0.056036096066236496,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 72690
+    },
+    {
+      "epoch": 7.4710174717368965,
+      "grad_norm": 0.0786886215209961,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 72693
+    },
+    {
+      "epoch": 7.471325796505653,
+      "grad_norm": 0.054459333419799805,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 72696
+    },
+    {
+      "epoch": 7.471634121274409,
+      "grad_norm": 0.05316760390996933,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 72699
+    },
+    {
+      "epoch": 7.471942446043165,
+      "grad_norm": 0.04993414506316185,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 72702
+    },
+    {
+      "epoch": 7.472250770811922,
+      "grad_norm": 0.08382388204336166,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 72705
+    },
+    {
+      "epoch": 7.472559095580678,
+      "grad_norm": 0.05314520001411438,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 72708
+    },
+    {
+      "epoch": 7.472867420349434,
+      "grad_norm": 0.061621155589818954,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 72711
+    },
+    {
+      "epoch": 7.4731757451181915,
+      "grad_norm": 0.12471956759691238,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 72714
+    },
+    {
+      "epoch": 7.473484069886948,
+      "grad_norm": 0.05979030579328537,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 72717
+    },
+    {
+      "epoch": 7.473792394655704,
+      "grad_norm": 0.08850116282701492,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 72720
+    },
+    {
+      "epoch": 7.47410071942446,
+      "grad_norm": 0.12835745513439178,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 72723
+    },
+    {
+      "epoch": 7.474409044193217,
+      "grad_norm": 0.04156319797039032,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 72726
+    },
+    {
+      "epoch": 7.474717368961973,
+      "grad_norm": 0.05407378822565079,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 72729
+    },
+    {
+      "epoch": 7.475025693730729,
+      "grad_norm": 0.0464872308075428,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 72732
+    },
+    {
+      "epoch": 7.4753340184994865,
+      "grad_norm": 0.05535002052783966,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 72735
+    },
+    {
+      "epoch": 7.475642343268243,
+      "grad_norm": 0.04521692544221878,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 72738
+    },
+    {
+      "epoch": 7.475950668036999,
+      "grad_norm": 0.11313310265541077,
+      "learning_rate": 0.01,
+      "loss": 2.0125,
+      "step": 72741
+    },
+    {
+      "epoch": 7.4762589928057555,
+      "grad_norm": 0.04605510085821152,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 72744
+    },
+    {
+      "epoch": 7.476567317574512,
+      "grad_norm": 0.0692354366183281,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 72747
+    },
+    {
+      "epoch": 7.476875642343268,
+      "grad_norm": 0.0635964423418045,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 72750
+    },
+    {
+      "epoch": 7.477183967112024,
+      "grad_norm": 0.10048150271177292,
+      "learning_rate": 0.01,
+      "loss": 2.0048,
+      "step": 72753
+    },
+    {
+      "epoch": 7.477492291880781,
+      "grad_norm": 0.056747522205114365,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 72756
+    },
+    {
+      "epoch": 7.477800616649538,
+      "grad_norm": 0.0338391549885273,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 72759
+    },
+    {
+      "epoch": 7.478108941418294,
+      "grad_norm": 0.09825774282217026,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 72762
+    },
+    {
+      "epoch": 7.4784172661870505,
+      "grad_norm": 0.07101613283157349,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 72765
+    },
+    {
+      "epoch": 7.478725590955807,
+      "grad_norm": 0.06692194938659668,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 72768
+    },
+    {
+      "epoch": 7.479033915724563,
+      "grad_norm": 0.07290485501289368,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 72771
+    },
+    {
+      "epoch": 7.479342240493319,
+      "grad_norm": 0.042110659182071686,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 72774
+    },
+    {
+      "epoch": 7.479650565262076,
+      "grad_norm": 0.08670128881931305,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 72777
+    },
+    {
+      "epoch": 7.479958890030833,
+      "grad_norm": 0.050455573946237564,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 72780
+    },
+    {
+      "epoch": 7.480267214799589,
+      "grad_norm": 0.13073371350765228,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 72783
+    },
+    {
+      "epoch": 7.4805755395683455,
+      "grad_norm": 0.11058662831783295,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 72786
+    },
+    {
+      "epoch": 7.480883864337102,
+      "grad_norm": 0.05915442109107971,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 72789
+    },
+    {
+      "epoch": 7.481192189105858,
+      "grad_norm": 0.05119236186146736,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 72792
+    },
+    {
+      "epoch": 7.4815005138746145,
+      "grad_norm": 0.07420118898153305,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 72795
+    },
+    {
+      "epoch": 7.481808838643371,
+      "grad_norm": 0.061627451330423355,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 72798
+    },
+    {
+      "epoch": 7.482117163412127,
+      "grad_norm": 0.04703347012400627,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 72801
+    },
+    {
+      "epoch": 7.482425488180884,
+      "grad_norm": 0.038575369864702225,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 72804
+    },
+    {
+      "epoch": 7.482733812949641,
+      "grad_norm": 0.07879947870969772,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 72807
+    },
+    {
+      "epoch": 7.483042137718397,
+      "grad_norm": 0.0844467431306839,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 72810
+    },
+    {
+      "epoch": 7.483350462487153,
+      "grad_norm": 0.06974819302558899,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 72813
+    },
+    {
+      "epoch": 7.4836587872559095,
+      "grad_norm": 0.060370590537786484,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 72816
+    },
+    {
+      "epoch": 7.483967112024666,
+      "grad_norm": 0.02921084314584732,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 72819
+    },
+    {
+      "epoch": 7.484275436793422,
+      "grad_norm": 0.04770229756832123,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 72822
+    },
+    {
+      "epoch": 7.484583761562178,
+      "grad_norm": 0.08941125124692917,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 72825
+    },
+    {
+      "epoch": 7.484892086330936,
+      "grad_norm": 0.03461920842528343,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 72828
+    },
+    {
+      "epoch": 7.485200411099692,
+      "grad_norm": 0.10770180821418762,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 72831
+    },
+    {
+      "epoch": 7.485508735868448,
+      "grad_norm": 0.08093126863241196,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 72834
+    },
+    {
+      "epoch": 7.4858170606372045,
+      "grad_norm": 0.08929312229156494,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 72837
+    },
+    {
+      "epoch": 7.486125385405961,
+      "grad_norm": 0.11654918640851974,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 72840
+    },
+    {
+      "epoch": 7.486433710174717,
+      "grad_norm": 0.04198765754699707,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 72843
+    },
+    {
+      "epoch": 7.4867420349434735,
+      "grad_norm": 0.041191764175891876,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 72846
+    },
+    {
+      "epoch": 7.48705035971223,
+      "grad_norm": 0.042416587471961975,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 72849
+    },
+    {
+      "epoch": 7.487358684480987,
+      "grad_norm": 0.033558644354343414,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 72852
+    },
+    {
+      "epoch": 7.487667009249743,
+      "grad_norm": 0.04876953363418579,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 72855
+    },
+    {
+      "epoch": 7.4879753340185,
+      "grad_norm": 0.11887954920530319,
+      "learning_rate": 0.01,
+      "loss": 1.9885,
+      "step": 72858
+    },
+    {
+      "epoch": 7.488283658787256,
+      "grad_norm": 0.09340287744998932,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 72861
+    },
+    {
+      "epoch": 7.488591983556012,
+      "grad_norm": 0.0659438893198967,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 72864
+    },
+    {
+      "epoch": 7.4889003083247685,
+      "grad_norm": 0.05727607384324074,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 72867
+    },
+    {
+      "epoch": 7.489208633093525,
+      "grad_norm": 0.051362354308366776,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 72870
+    },
+    {
+      "epoch": 7.489516957862282,
+      "grad_norm": 0.07566554844379425,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 72873
+    },
+    {
+      "epoch": 7.489825282631038,
+      "grad_norm": 0.052347857505083084,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 72876
+    },
+    {
+      "epoch": 7.490133607399795,
+      "grad_norm": 0.03814227133989334,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 72879
+    },
+    {
+      "epoch": 7.490441932168551,
+      "grad_norm": 0.03464813157916069,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 72882
+    },
+    {
+      "epoch": 7.490750256937307,
+      "grad_norm": 0.12053199857473373,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 72885
+    },
+    {
+      "epoch": 7.4910585817060635,
+      "grad_norm": 0.04944891110062599,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 72888
+    },
+    {
+      "epoch": 7.49136690647482,
+      "grad_norm": 0.12440962344408035,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 72891
+    },
+    {
+      "epoch": 7.491675231243576,
+      "grad_norm": 0.07274243980646133,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 72894
+    },
+    {
+      "epoch": 7.491983556012333,
+      "grad_norm": 0.041211891919374466,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 72897
+    },
+    {
+      "epoch": 7.49229188078109,
+      "grad_norm": 0.07639609277248383,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 72900
+    },
+    {
+      "epoch": 7.492600205549846,
+      "grad_norm": 0.05690251663327217,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 72903
+    },
+    {
+      "epoch": 7.492908530318602,
+      "grad_norm": 0.04439922422170639,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 72906
+    },
+    {
+      "epoch": 7.493216855087359,
+      "grad_norm": 0.03292795643210411,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 72909
+    },
+    {
+      "epoch": 7.493525179856115,
+      "grad_norm": 0.03560163453221321,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 72912
+    },
+    {
+      "epoch": 7.493833504624871,
+      "grad_norm": 0.09666794538497925,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 72915
+    },
+    {
+      "epoch": 7.494141829393628,
+      "grad_norm": 0.0834769606590271,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 72918
+    },
+    {
+      "epoch": 7.494450154162385,
+      "grad_norm": 0.04032580181956291,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 72921
+    },
+    {
+      "epoch": 7.494758478931141,
+      "grad_norm": 0.1358065903186798,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 72924
+    },
+    {
+      "epoch": 7.495066803699897,
+      "grad_norm": 0.07956533133983612,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 72927
+    },
+    {
+      "epoch": 7.495375128468654,
+      "grad_norm": 0.09689958393573761,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 72930
+    },
+    {
+      "epoch": 7.49568345323741,
+      "grad_norm": 0.09367521107196808,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 72933
+    },
+    {
+      "epoch": 7.495991778006166,
+      "grad_norm": 0.054803263396024704,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 72936
+    },
+    {
+      "epoch": 7.4963001027749225,
+      "grad_norm": 0.04972374439239502,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 72939
+    },
+    {
+      "epoch": 7.49660842754368,
+      "grad_norm": 0.044981930404901505,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 72942
+    },
+    {
+      "epoch": 7.496916752312436,
+      "grad_norm": 0.03693649172782898,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 72945
+    },
+    {
+      "epoch": 7.497225077081192,
+      "grad_norm": 0.12305953353643417,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 72948
+    },
+    {
+      "epoch": 7.497533401849949,
+      "grad_norm": 0.10410351306200027,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 72951
+    },
+    {
+      "epoch": 7.497841726618705,
+      "grad_norm": 0.04615464434027672,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 72954
+    },
+    {
+      "epoch": 7.498150051387461,
+      "grad_norm": 0.06190275400876999,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 72957
+    },
+    {
+      "epoch": 7.498458376156218,
+      "grad_norm": 0.03822600841522217,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 72960
+    },
+    {
+      "epoch": 7.498766700924975,
+      "grad_norm": 0.06215745583176613,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 72963
+    },
+    {
+      "epoch": 7.499075025693731,
+      "grad_norm": 0.03885975107550621,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 72966
+    },
+    {
+      "epoch": 7.499383350462487,
+      "grad_norm": 0.04294813051819801,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 72969
+    },
+    {
+      "epoch": 7.499691675231244,
+      "grad_norm": 0.04972092807292938,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 72972
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 0.11483730375766754,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 72975
+    },
+    {
+      "epoch": 7.500308324768756,
+      "grad_norm": 0.09318023920059204,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 72978
+    },
+    {
+      "epoch": 7.500616649537513,
+      "grad_norm": 0.08225825428962708,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 72981
+    },
+    {
+      "epoch": 7.500924974306269,
+      "grad_norm": 0.06477591395378113,
+      "learning_rate": 0.01,
+      "loss": 2.0082,
+      "step": 72984
+    },
+    {
+      "epoch": 7.501233299075025,
+      "grad_norm": 0.03535725921392441,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 72987
+    },
+    {
+      "epoch": 7.501541623843782,
+      "grad_norm": 0.05080735310912132,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 72990
+    },
+    {
+      "epoch": 7.501849948612539,
+      "grad_norm": 0.0634625032544136,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 72993
+    },
+    {
+      "epoch": 7.502158273381295,
+      "grad_norm": 0.04936002939939499,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 72996
+    },
+    {
+      "epoch": 7.502466598150051,
+      "grad_norm": 0.055662546306848526,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 72999
+    },
+    {
+      "epoch": 7.502774922918808,
+      "grad_norm": 0.08707310259342194,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 73002
+    },
+    {
+      "epoch": 7.503083247687564,
+      "grad_norm": 0.0861874595284462,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 73005
+    },
+    {
+      "epoch": 7.503391572456321,
+      "grad_norm": 0.04461105167865753,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 73008
+    },
+    {
+      "epoch": 7.5036998972250775,
+      "grad_norm": 0.03987470641732216,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 73011
+    },
+    {
+      "epoch": 7.504008221993834,
+      "grad_norm": 0.03856007382273674,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 73014
+    },
+    {
+      "epoch": 7.50431654676259,
+      "grad_norm": 0.12175683677196503,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 73017
+    },
+    {
+      "epoch": 7.504624871531346,
+      "grad_norm": 0.04544226825237274,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 73020
+    },
+    {
+      "epoch": 7.504933196300103,
+      "grad_norm": 0.10112664103507996,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 73023
+    },
+    {
+      "epoch": 7.505241521068859,
+      "grad_norm": 0.0891900509595871,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 73026
+    },
+    {
+      "epoch": 7.505549845837615,
+      "grad_norm": 0.03948020190000534,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 73029
+    },
+    {
+      "epoch": 7.505858170606372,
+      "grad_norm": 0.041939374059438705,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 73032
+    },
+    {
+      "epoch": 7.506166495375129,
+      "grad_norm": 0.037538815289735794,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 73035
+    },
+    {
+      "epoch": 7.506474820143885,
+      "grad_norm": 0.03573056682944298,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 73038
+    },
+    {
+      "epoch": 7.506783144912641,
+      "grad_norm": 0.0640433058142662,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 73041
+    },
+    {
+      "epoch": 7.507091469681398,
+      "grad_norm": 0.1231018453836441,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 73044
+    },
+    {
+      "epoch": 7.507399794450154,
+      "grad_norm": 0.13998620212078094,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 73047
+    },
+    {
+      "epoch": 7.50770811921891,
+      "grad_norm": 0.05272144079208374,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 73050
+    },
+    {
+      "epoch": 7.508016443987667,
+      "grad_norm": 0.06140724569559097,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 73053
+    },
+    {
+      "epoch": 7.508324768756424,
+      "grad_norm": 0.0687750056385994,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 73056
+    },
+    {
+      "epoch": 7.50863309352518,
+      "grad_norm": 0.06672102957963943,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 73059
+    },
+    {
+      "epoch": 7.5089414182939365,
+      "grad_norm": 0.11446451395750046,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 73062
+    },
+    {
+      "epoch": 7.509249743062693,
+      "grad_norm": 0.05556824430823326,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 73065
+    },
+    {
+      "epoch": 7.509558067831449,
+      "grad_norm": 0.035088442265987396,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 73068
+    },
+    {
+      "epoch": 7.509866392600205,
+      "grad_norm": 0.06371704488992691,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 73071
+    },
+    {
+      "epoch": 7.510174717368962,
+      "grad_norm": 0.052248984575271606,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 73074
+    },
+    {
+      "epoch": 7.510483042137718,
+      "grad_norm": 0.04099287465214729,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 73077
+    },
+    {
+      "epoch": 7.510791366906475,
+      "grad_norm": 0.09975527226924896,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 73080
+    },
+    {
+      "epoch": 7.5110996916752315,
+      "grad_norm": 0.07624788582324982,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 73083
+    },
+    {
+      "epoch": 7.511408016443988,
+      "grad_norm": 0.07405281066894531,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 73086
+    },
+    {
+      "epoch": 7.511716341212744,
+      "grad_norm": 0.041836485266685486,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 73089
+    },
+    {
+      "epoch": 7.5120246659815,
+      "grad_norm": 0.043866775929927826,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 73092
+    },
+    {
+      "epoch": 7.512332990750257,
+      "grad_norm": 0.032251179218292236,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 73095
+    },
+    {
+      "epoch": 7.512641315519013,
+      "grad_norm": 0.11554509401321411,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 73098
+    },
+    {
+      "epoch": 7.51294964028777,
+      "grad_norm": 0.06754744052886963,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 73101
+    },
+    {
+      "epoch": 7.5132579650565265,
+      "grad_norm": 0.09699345380067825,
+      "learning_rate": 0.01,
+      "loss": 2.0038,
+      "step": 73104
+    },
+    {
+      "epoch": 7.513566289825283,
+      "grad_norm": 0.08807633817195892,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 73107
+    },
+    {
+      "epoch": 7.513874614594039,
+      "grad_norm": 0.037080902606248856,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 73110
+    },
+    {
+      "epoch": 7.5141829393627955,
+      "grad_norm": 0.06916693598031998,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 73113
+    },
+    {
+      "epoch": 7.514491264131552,
+      "grad_norm": 0.07650911808013916,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 73116
+    },
+    {
+      "epoch": 7.514799588900308,
+      "grad_norm": 0.09734641015529633,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 73119
+    },
+    {
+      "epoch": 7.515107913669064,
+      "grad_norm": 0.11928999423980713,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 73122
+    },
+    {
+      "epoch": 7.515416238437821,
+      "grad_norm": 0.07260621339082718,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 73125
+    },
+    {
+      "epoch": 7.515724563206578,
+      "grad_norm": 0.046437349170446396,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 73128
+    },
+    {
+      "epoch": 7.516032887975334,
+      "grad_norm": 0.06214399263262749,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 73131
+    },
+    {
+      "epoch": 7.5163412127440905,
+      "grad_norm": 0.051302261650562286,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 73134
+    },
+    {
+      "epoch": 7.516649537512847,
+      "grad_norm": 0.041365306824445724,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 73137
+    },
+    {
+      "epoch": 7.516957862281603,
+      "grad_norm": 0.044153954833745956,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 73140
+    },
+    {
+      "epoch": 7.517266187050359,
+      "grad_norm": 0.10011675208806992,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 73143
+    },
+    {
+      "epoch": 7.517574511819117,
+      "grad_norm": 0.05302093178033829,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 73146
+    },
+    {
+      "epoch": 7.517882836587873,
+      "grad_norm": 0.0799306333065033,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 73149
+    },
+    {
+      "epoch": 7.518191161356629,
+      "grad_norm": 0.0636538714170456,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 73152
+    },
+    {
+      "epoch": 7.5184994861253855,
+      "grad_norm": 0.05640498548746109,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 73155
+    },
+    {
+      "epoch": 7.518807810894142,
+      "grad_norm": 0.09565453976392746,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 73158
+    },
+    {
+      "epoch": 7.519116135662898,
+      "grad_norm": 0.04038292542099953,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 73161
+    },
+    {
+      "epoch": 7.5194244604316545,
+      "grad_norm": 0.06211680546402931,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 73164
+    },
+    {
+      "epoch": 7.519732785200411,
+      "grad_norm": 0.13773837685585022,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 73167
+    },
+    {
+      "epoch": 7.520041109969167,
+      "grad_norm": 0.04517769068479538,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 73170
+    },
+    {
+      "epoch": 7.520349434737924,
+      "grad_norm": 0.06648048013448715,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 73173
+    },
+    {
+      "epoch": 7.520657759506681,
+      "grad_norm": 0.05798725783824921,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 73176
+    },
+    {
+      "epoch": 7.520966084275437,
+      "grad_norm": 0.07782495766878128,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 73179
+    },
+    {
+      "epoch": 7.521274409044193,
+      "grad_norm": 0.06339671462774277,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 73182
+    },
+    {
+      "epoch": 7.5215827338129495,
+      "grad_norm": 0.04342902451753616,
+      "learning_rate": 0.01,
+      "loss": 2.006,
+      "step": 73185
+    },
+    {
+      "epoch": 7.521891058581706,
+      "grad_norm": 0.054694198071956635,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 73188
+    },
+    {
+      "epoch": 7.522199383350462,
+      "grad_norm": 0.058501482009887695,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 73191
+    },
+    {
+      "epoch": 7.522507708119219,
+      "grad_norm": 0.06266117095947266,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 73194
+    },
+    {
+      "epoch": 7.522816032887976,
+      "grad_norm": 0.07241412997245789,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 73197
+    },
+    {
+      "epoch": 7.523124357656732,
+      "grad_norm": 0.06054093688726425,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 73200
+    },
+    {
+      "epoch": 7.523432682425488,
+      "grad_norm": 0.03412376344203949,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 73203
+    },
+    {
+      "epoch": 7.5237410071942445,
+      "grad_norm": 0.05688127875328064,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 73206
+    },
+    {
+      "epoch": 7.524049331963001,
+      "grad_norm": 0.048515647649765015,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 73209
+    },
+    {
+      "epoch": 7.524357656731757,
+      "grad_norm": 0.06496395915746689,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 73212
+    },
+    {
+      "epoch": 7.5246659815005135,
+      "grad_norm": 0.10364048182964325,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 73215
+    },
+    {
+      "epoch": 7.524974306269271,
+      "grad_norm": 0.15099607408046722,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 73218
+    },
+    {
+      "epoch": 7.525282631038027,
+      "grad_norm": 0.059038564562797546,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 73221
+    },
+    {
+      "epoch": 7.525590955806783,
+      "grad_norm": 0.056402787566185,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 73224
+    },
+    {
+      "epoch": 7.52589928057554,
+      "grad_norm": 0.07999878376722336,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 73227
+    },
+    {
+      "epoch": 7.526207605344296,
+      "grad_norm": 0.1127920150756836,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 73230
+    },
+    {
+      "epoch": 7.526515930113052,
+      "grad_norm": 0.058007560670375824,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 73233
+    },
+    {
+      "epoch": 7.5268242548818085,
+      "grad_norm": 0.039497777819633484,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 73236
+    },
+    {
+      "epoch": 7.527132579650566,
+      "grad_norm": 0.033727843314409256,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 73239
+    },
+    {
+      "epoch": 7.527440904419322,
+      "grad_norm": 0.04165032133460045,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 73242
+    },
+    {
+      "epoch": 7.527749229188078,
+      "grad_norm": 0.06078481674194336,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 73245
+    },
+    {
+      "epoch": 7.528057553956835,
+      "grad_norm": 0.09847116470336914,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 73248
+    },
+    {
+      "epoch": 7.528365878725591,
+      "grad_norm": 0.1251547634601593,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 73251
+    },
+    {
+      "epoch": 7.528674203494347,
+      "grad_norm": 0.09058269113302231,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 73254
+    },
+    {
+      "epoch": 7.5289825282631035,
+      "grad_norm": 0.10906793177127838,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 73257
+    },
+    {
+      "epoch": 7.52929085303186,
+      "grad_norm": 0.03927643597126007,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 73260
+    },
+    {
+      "epoch": 7.529599177800616,
+      "grad_norm": 0.041548848152160645,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 73263
+    },
+    {
+      "epoch": 7.529907502569373,
+      "grad_norm": 0.03454883396625519,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 73266
+    },
+    {
+      "epoch": 7.53021582733813,
+      "grad_norm": 0.03971594572067261,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 73269
+    },
+    {
+      "epoch": 7.530524152106886,
+      "grad_norm": 0.15191693603992462,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 73272
+    },
+    {
+      "epoch": 7.530832476875642,
+      "grad_norm": 0.051092084497213364,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 73275
+    },
+    {
+      "epoch": 7.531140801644399,
+      "grad_norm": 0.06430649757385254,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 73278
+    },
+    {
+      "epoch": 7.531449126413155,
+      "grad_norm": 0.06463275849819183,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 73281
+    },
+    {
+      "epoch": 7.531757451181912,
+      "grad_norm": 0.037970978766679764,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 73284
+    },
+    {
+      "epoch": 7.532065775950668,
+      "grad_norm": 0.052109669893980026,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 73287
+    },
+    {
+      "epoch": 7.532374100719425,
+      "grad_norm": 0.04990661144256592,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 73290
+    },
+    {
+      "epoch": 7.532682425488181,
+      "grad_norm": 0.08009975403547287,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 73293
+    },
+    {
+      "epoch": 7.532990750256937,
+      "grad_norm": 0.05050569027662277,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 73296
+    },
+    {
+      "epoch": 7.533299075025694,
+      "grad_norm": 0.07872025668621063,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 73299
+    },
+    {
+      "epoch": 7.53360739979445,
+      "grad_norm": 0.04549863189458847,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 73302
+    },
+    {
+      "epoch": 7.533915724563206,
+      "grad_norm": 0.10252397507429123,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 73305
+    },
+    {
+      "epoch": 7.5342240493319625,
+      "grad_norm": 0.06999849528074265,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 73308
+    },
+    {
+      "epoch": 7.53453237410072,
+      "grad_norm": 0.062004949897527695,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 73311
+    },
+    {
+      "epoch": 7.534840698869476,
+      "grad_norm": 0.10107864439487457,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 73314
+    },
+    {
+      "epoch": 7.535149023638232,
+      "grad_norm": 0.042918670922517776,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 73317
+    },
+    {
+      "epoch": 7.535457348406989,
+      "grad_norm": 0.11060534417629242,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 73320
+    },
+    {
+      "epoch": 7.535765673175745,
+      "grad_norm": 0.033137235790491104,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 73323
+    },
+    {
+      "epoch": 7.536073997944501,
+      "grad_norm": 0.07771358639001846,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 73326
+    },
+    {
+      "epoch": 7.536382322713258,
+      "grad_norm": 0.05748758465051651,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 73329
+    },
+    {
+      "epoch": 7.536690647482015,
+      "grad_norm": 0.06228995695710182,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 73332
+    },
+    {
+      "epoch": 7.536998972250771,
+      "grad_norm": 0.06400372087955475,
+      "learning_rate": 0.01,
+      "loss": 1.9905,
+      "step": 73335
+    },
+    {
+      "epoch": 7.537307297019527,
+      "grad_norm": 0.11260011047124863,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 73338
+    },
+    {
+      "epoch": 7.537615621788284,
+      "grad_norm": 0.052524223923683167,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 73341
+    },
+    {
+      "epoch": 7.53792394655704,
+      "grad_norm": 0.07686839252710342,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 73344
+    },
+    {
+      "epoch": 7.538232271325796,
+      "grad_norm": 0.11364465206861496,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 73347
+    },
+    {
+      "epoch": 7.538540596094553,
+      "grad_norm": 0.06682565808296204,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 73350
+    },
+    {
+      "epoch": 7.538848920863309,
+      "grad_norm": 0.04542539268732071,
+      "learning_rate": 0.01,
+      "loss": 1.9923,
+      "step": 73353
+    },
+    {
+      "epoch": 7.539157245632066,
+      "grad_norm": 0.0473509319126606,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 73356
+    },
+    {
+      "epoch": 7.539465570400822,
+      "grad_norm": 0.04453996196389198,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 73359
+    },
+    {
+      "epoch": 7.539773895169579,
+      "grad_norm": 0.03908911719918251,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 73362
+    },
+    {
+      "epoch": 7.540082219938335,
+      "grad_norm": 0.051413651555776596,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 73365
+    },
+    {
+      "epoch": 7.540390544707091,
+      "grad_norm": 0.053165409713983536,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 73368
+    },
+    {
+      "epoch": 7.540698869475848,
+      "grad_norm": 0.049723848700523376,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 73371
+    },
+    {
+      "epoch": 7.541007194244604,
+      "grad_norm": 0.06980393826961517,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 73374
+    },
+    {
+      "epoch": 7.541315519013361,
+      "grad_norm": 0.09864058345556259,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 73377
+    },
+    {
+      "epoch": 7.5416238437821175,
+      "grad_norm": 0.0965089350938797,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 73380
+    },
+    {
+      "epoch": 7.541932168550874,
+      "grad_norm": 0.09559361636638641,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 73383
+    },
+    {
+      "epoch": 7.54224049331963,
+      "grad_norm": 0.03276180103421211,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 73386
+    },
+    {
+      "epoch": 7.542548818088386,
+      "grad_norm": 0.07006944715976715,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 73389
+    },
+    {
+      "epoch": 7.542857142857143,
+      "grad_norm": 0.09948286414146423,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 73392
+    },
+    {
+      "epoch": 7.543165467625899,
+      "grad_norm": 0.09028782695531845,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 73395
+    },
+    {
+      "epoch": 7.543473792394655,
+      "grad_norm": 0.06747119873762131,
+      "learning_rate": 0.01,
+      "loss": 1.9929,
+      "step": 73398
+    },
+    {
+      "epoch": 7.5437821171634125,
+      "grad_norm": 0.09048298001289368,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 73401
+    },
+    {
+      "epoch": 7.544090441932169,
+      "grad_norm": 0.03283631429076195,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 73404
+    },
+    {
+      "epoch": 7.544398766700925,
+      "grad_norm": 0.04949327930808067,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 73407
+    },
+    {
+      "epoch": 7.544707091469681,
+      "grad_norm": 0.07692143321037292,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 73410
+    },
+    {
+      "epoch": 7.545015416238438,
+      "grad_norm": 0.039769385010004044,
+      "learning_rate": 0.01,
+      "loss": 2.0088,
+      "step": 73413
+    },
+    {
+      "epoch": 7.545323741007194,
+      "grad_norm": 0.05932982638478279,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 73416
+    },
+    {
+      "epoch": 7.54563206577595,
+      "grad_norm": 0.06630944460630417,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 73419
+    },
+    {
+      "epoch": 7.5459403905447076,
+      "grad_norm": 0.05466363951563835,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 73422
+    },
+    {
+      "epoch": 7.546248715313464,
+      "grad_norm": 0.04922345280647278,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 73425
+    },
+    {
+      "epoch": 7.54655704008222,
+      "grad_norm": 0.056301772594451904,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 73428
+    },
+    {
+      "epoch": 7.5468653648509765,
+      "grad_norm": 0.05948075279593468,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 73431
+    },
+    {
+      "epoch": 7.547173689619733,
+      "grad_norm": 0.07484395802021027,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 73434
+    },
+    {
+      "epoch": 7.547482014388489,
+      "grad_norm": 0.055150918662548065,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 73437
+    },
+    {
+      "epoch": 7.547790339157245,
+      "grad_norm": 0.0486358217895031,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 73440
+    },
+    {
+      "epoch": 7.548098663926002,
+      "grad_norm": 0.058294083923101425,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 73443
+    },
+    {
+      "epoch": 7.548406988694758,
+      "grad_norm": 0.034051090478897095,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 73446
+    },
+    {
+      "epoch": 7.548715313463515,
+      "grad_norm": 0.04121851176023483,
+      "learning_rate": 0.01,
+      "loss": 1.9947,
+      "step": 73449
+    },
+    {
+      "epoch": 7.5490236382322715,
+      "grad_norm": 0.05448736622929573,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 73452
+    },
+    {
+      "epoch": 7.549331963001028,
+      "grad_norm": 0.12663137912750244,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 73455
+    },
+    {
+      "epoch": 7.549640287769784,
+      "grad_norm": 0.07261858880519867,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 73458
+    },
+    {
+      "epoch": 7.54994861253854,
+      "grad_norm": 0.04712305963039398,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 73461
+    },
+    {
+      "epoch": 7.550256937307297,
+      "grad_norm": 0.1148359552025795,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 73464
+    },
+    {
+      "epoch": 7.550565262076054,
+      "grad_norm": 0.03637329116463661,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 73467
+    },
+    {
+      "epoch": 7.55087358684481,
+      "grad_norm": 0.03120165877044201,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 73470
+    },
+    {
+      "epoch": 7.5511819116135666,
+      "grad_norm": 0.03387007117271423,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 73473
+    },
+    {
+      "epoch": 7.551490236382323,
+      "grad_norm": 0.10267225652933121,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 73476
+    },
+    {
+      "epoch": 7.551798561151079,
+      "grad_norm": 0.0743681788444519,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 73479
+    },
+    {
+      "epoch": 7.5521068859198355,
+      "grad_norm": 0.041468244045972824,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 73482
+    },
+    {
+      "epoch": 7.552415210688592,
+      "grad_norm": 0.03812147676944733,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 73485
+    },
+    {
+      "epoch": 7.552723535457348,
+      "grad_norm": 0.046897489577531815,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 73488
+    },
+    {
+      "epoch": 7.553031860226104,
+      "grad_norm": 0.043389540165662766,
+      "learning_rate": 0.01,
+      "loss": 1.9916,
+      "step": 73491
+    },
+    {
+      "epoch": 7.553340184994862,
+      "grad_norm": 0.04722335562109947,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 73494
+    },
+    {
+      "epoch": 7.553648509763618,
+      "grad_norm": 0.03955325484275818,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 73497
+    },
+    {
+      "epoch": 7.553956834532374,
+      "grad_norm": 0.11962294578552246,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 73500
+    },
+    {
+      "epoch": 7.5542651593011305,
+      "grad_norm": 0.03522321209311485,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 73503
+    },
+    {
+      "epoch": 7.554573484069887,
+      "grad_norm": 0.13331198692321777,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 73506
+    },
+    {
+      "epoch": 7.554881808838643,
+      "grad_norm": 0.05670026317238808,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 73509
+    },
+    {
+      "epoch": 7.555190133607399,
+      "grad_norm": 0.05296124890446663,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 73512
+    },
+    {
+      "epoch": 7.555498458376157,
+      "grad_norm": 0.06895041465759277,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 73515
+    },
+    {
+      "epoch": 7.555806783144913,
+      "grad_norm": 0.06345079094171524,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 73518
+    },
+    {
+      "epoch": 7.556115107913669,
+      "grad_norm": 0.055405255407094955,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 73521
+    },
+    {
+      "epoch": 7.556423432682426,
+      "grad_norm": 0.053103215992450714,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 73524
+    },
+    {
+      "epoch": 7.556731757451182,
+      "grad_norm": 0.04727816581726074,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 73527
+    },
+    {
+      "epoch": 7.557040082219938,
+      "grad_norm": 0.059660717844963074,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 73530
+    },
+    {
+      "epoch": 7.5573484069886945,
+      "grad_norm": 0.05450676754117012,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 73533
+    },
+    {
+      "epoch": 7.557656731757451,
+      "grad_norm": 0.06650180369615555,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 73536
+    },
+    {
+      "epoch": 7.557965056526208,
+      "grad_norm": 0.08162131160497665,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 73539
+    },
+    {
+      "epoch": 7.558273381294964,
+      "grad_norm": 0.06207094341516495,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 73542
+    },
+    {
+      "epoch": 7.558581706063721,
+      "grad_norm": 0.07756184041500092,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 73545
+    },
+    {
+      "epoch": 7.558890030832477,
+      "grad_norm": 0.036559395492076874,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 73548
+    },
+    {
+      "epoch": 7.559198355601233,
+      "grad_norm": 0.03526364639401436,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 73551
+    },
+    {
+      "epoch": 7.5595066803699895,
+      "grad_norm": 0.05811254307627678,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 73554
+    },
+    {
+      "epoch": 7.559815005138746,
+      "grad_norm": 0.10756142437458038,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 73557
+    },
+    {
+      "epoch": 7.560123329907503,
+      "grad_norm": 0.06948500126600266,
+      "learning_rate": 0.01,
+      "loss": 2.0019,
+      "step": 73560
+    },
+    {
+      "epoch": 7.560431654676259,
+      "grad_norm": 0.04926348850131035,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 73563
+    },
+    {
+      "epoch": 7.560739979445016,
+      "grad_norm": 0.039930544793605804,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 73566
+    },
+    {
+      "epoch": 7.561048304213772,
+      "grad_norm": 0.03368406742811203,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 73569
+    },
+    {
+      "epoch": 7.561356628982528,
+      "grad_norm": 0.05603760853409767,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 73572
+    },
+    {
+      "epoch": 7.561664953751285,
+      "grad_norm": 0.1923237293958664,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 73575
+    },
+    {
+      "epoch": 7.561973278520041,
+      "grad_norm": 0.09853487461805344,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 73578
+    },
+    {
+      "epoch": 7.562281603288797,
+      "grad_norm": 0.0466923862695694,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 73581
+    },
+    {
+      "epoch": 7.5625899280575535,
+      "grad_norm": 0.06811875104904175,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 73584
+    },
+    {
+      "epoch": 7.562898252826311,
+      "grad_norm": 0.07578512281179428,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 73587
+    },
+    {
+      "epoch": 7.563206577595067,
+      "grad_norm": 0.056927092373371124,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 73590
+    },
+    {
+      "epoch": 7.563514902363823,
+      "grad_norm": 0.04470854252576828,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 73593
+    },
+    {
+      "epoch": 7.56382322713258,
+      "grad_norm": 0.0404474139213562,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 73596
+    },
+    {
+      "epoch": 7.564131551901336,
+      "grad_norm": 0.042724210768938065,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 73599
+    },
+    {
+      "epoch": 7.564439876670092,
+      "grad_norm": 0.05680299550294876,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 73602
+    },
+    {
+      "epoch": 7.564748201438849,
+      "grad_norm": 0.10340821743011475,
+      "learning_rate": 0.01,
+      "loss": 1.9967,
+      "step": 73605
+    },
+    {
+      "epoch": 7.565056526207606,
+      "grad_norm": 0.1016075536608696,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 73608
+    },
+    {
+      "epoch": 7.565364850976362,
+      "grad_norm": 0.0473574735224247,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 73611
+    },
+    {
+      "epoch": 7.565673175745118,
+      "grad_norm": 0.06154468655586243,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 73614
+    },
+    {
+      "epoch": 7.565981500513875,
+      "grad_norm": 0.04448488727211952,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 73617
+    },
+    {
+      "epoch": 7.566289825282631,
+      "grad_norm": 0.039563100785017014,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 73620
+    },
+    {
+      "epoch": 7.566598150051387,
+      "grad_norm": 0.038476377725601196,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 73623
+    },
+    {
+      "epoch": 7.566906474820144,
+      "grad_norm": 0.04110477492213249,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 73626
+    },
+    {
+      "epoch": 7.5672147995889,
+      "grad_norm": 0.057609956711530685,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 73629
+    },
+    {
+      "epoch": 7.567523124357657,
+      "grad_norm": 0.03673383593559265,
+      "learning_rate": 0.01,
+      "loss": 1.9838,
+      "step": 73632
+    },
+    {
+      "epoch": 7.567831449126413,
+      "grad_norm": 0.05421116575598717,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 73635
+    },
+    {
+      "epoch": 7.56813977389517,
+      "grad_norm": 0.09983667731285095,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 73638
+    },
+    {
+      "epoch": 7.568448098663926,
+      "grad_norm": 0.06927699595689774,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 73641
+    },
+    {
+      "epoch": 7.568756423432682,
+      "grad_norm": 0.10717348009347916,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 73644
+    },
+    {
+      "epoch": 7.569064748201439,
+      "grad_norm": 0.11684148758649826,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 73647
+    },
+    {
+      "epoch": 7.569373072970195,
+      "grad_norm": 0.07459786534309387,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 73650
+    },
+    {
+      "epoch": 7.569681397738952,
+      "grad_norm": 0.07895784080028534,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 73653
+    },
+    {
+      "epoch": 7.569989722507708,
+      "grad_norm": 0.03824291005730629,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 73656
+    },
+    {
+      "epoch": 7.570298047276465,
+      "grad_norm": 0.037501122802495956,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 73659
+    },
+    {
+      "epoch": 7.570606372045221,
+      "grad_norm": 0.04228674992918968,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 73662
+    },
+    {
+      "epoch": 7.570914696813977,
+      "grad_norm": 0.04469291865825653,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 73665
+    },
+    {
+      "epoch": 7.571223021582734,
+      "grad_norm": 0.053544167429208755,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 73668
+    },
+    {
+      "epoch": 7.57153134635149,
+      "grad_norm": 0.052348144352436066,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 73671
+    },
+    {
+      "epoch": 7.571839671120246,
+      "grad_norm": 0.12371430546045303,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 73674
+    },
+    {
+      "epoch": 7.5721479958890034,
+      "grad_norm": 0.048446524888277054,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 73677
+    },
+    {
+      "epoch": 7.57245632065776,
+      "grad_norm": 0.049632761627435684,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 73680
+    },
+    {
+      "epoch": 7.572764645426516,
+      "grad_norm": 0.053510334342718124,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 73683
+    },
+    {
+      "epoch": 7.573072970195272,
+      "grad_norm": 0.0622355155646801,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 73686
+    },
+    {
+      "epoch": 7.573381294964029,
+      "grad_norm": 0.046275582164525986,
+      "learning_rate": 0.01,
+      "loss": 1.9884,
+      "step": 73689
+    },
+    {
+      "epoch": 7.573689619732785,
+      "grad_norm": 0.07243076711893082,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 73692
+    },
+    {
+      "epoch": 7.573997944501541,
+      "grad_norm": 0.07151640206575394,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 73695
+    },
+    {
+      "epoch": 7.5743062692702985,
+      "grad_norm": 0.07176624238491058,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 73698
+    },
+    {
+      "epoch": 7.574614594039055,
+      "grad_norm": 0.06733313947916031,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 73701
+    },
+    {
+      "epoch": 7.574922918807811,
+      "grad_norm": 0.07993333786725998,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 73704
+    },
+    {
+      "epoch": 7.575231243576567,
+      "grad_norm": 0.11807093769311905,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 73707
+    },
+    {
+      "epoch": 7.575539568345324,
+      "grad_norm": 0.07152784615755081,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 73710
+    },
+    {
+      "epoch": 7.57584789311408,
+      "grad_norm": 0.05039028450846672,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 73713
+    },
+    {
+      "epoch": 7.576156217882836,
+      "grad_norm": 0.0637911856174469,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 73716
+    },
+    {
+      "epoch": 7.576464542651593,
+      "grad_norm": 0.04242941737174988,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 73719
+    },
+    {
+      "epoch": 7.576772867420349,
+      "grad_norm": 0.053346943110227585,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 73722
+    },
+    {
+      "epoch": 7.577081192189106,
+      "grad_norm": 0.03864128142595291,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 73725
+    },
+    {
+      "epoch": 7.5773895169578624,
+      "grad_norm": 0.052774764597415924,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 73728
+    },
+    {
+      "epoch": 7.577697841726619,
+      "grad_norm": 0.036650512367486954,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 73731
+    },
+    {
+      "epoch": 7.578006166495375,
+      "grad_norm": 0.10396769642829895,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 73734
+    },
+    {
+      "epoch": 7.578314491264131,
+      "grad_norm": 0.0873318612575531,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 73737
+    },
+    {
+      "epoch": 7.578622816032888,
+      "grad_norm": 0.044649526476860046,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 73740
+    },
+    {
+      "epoch": 7.578931140801645,
+      "grad_norm": 0.04690408334136009,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 73743
+    },
+    {
+      "epoch": 7.579239465570401,
+      "grad_norm": 0.03519474342465401,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 73746
+    },
+    {
+      "epoch": 7.5795477903391575,
+      "grad_norm": 0.05188727751374245,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 73749
+    },
+    {
+      "epoch": 7.579856115107914,
+      "grad_norm": 0.04967820271849632,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 73752
+    },
+    {
+      "epoch": 7.58016443987667,
+      "grad_norm": 0.03920558840036392,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 73755
+    },
+    {
+      "epoch": 7.580472764645426,
+      "grad_norm": 0.0493493489921093,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 73758
+    },
+    {
+      "epoch": 7.580781089414183,
+      "grad_norm": 0.03327930346131325,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 73761
+    },
+    {
+      "epoch": 7.581089414182939,
+      "grad_norm": 0.041295699775218964,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 73764
+    },
+    {
+      "epoch": 7.581397738951695,
+      "grad_norm": 0.10000699013471603,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 73767
+    },
+    {
+      "epoch": 7.5817060637204525,
+      "grad_norm": 0.06557929515838623,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 73770
+    },
+    {
+      "epoch": 7.582014388489209,
+      "grad_norm": 0.06633325666189194,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 73773
+    },
+    {
+      "epoch": 7.582322713257965,
+      "grad_norm": 0.08267974108457565,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 73776
+    },
+    {
+      "epoch": 7.5826310380267214,
+      "grad_norm": 0.0816752165555954,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 73779
+    },
+    {
+      "epoch": 7.582939362795478,
+      "grad_norm": 0.03547079861164093,
+      "learning_rate": 0.01,
+      "loss": 1.997,
+      "step": 73782
+    },
+    {
+      "epoch": 7.583247687564234,
+      "grad_norm": 0.09666965156793594,
+      "learning_rate": 0.01,
+      "loss": 2.0044,
+      "step": 73785
+    },
+    {
+      "epoch": 7.58355601233299,
+      "grad_norm": 0.056526027619838715,
+      "learning_rate": 0.01,
+      "loss": 2.0003,
+      "step": 73788
+    },
+    {
+      "epoch": 7.583864337101748,
+      "grad_norm": 0.04913655295968056,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 73791
+    },
+    {
+      "epoch": 7.584172661870504,
+      "grad_norm": 0.10376237332820892,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 73794
+    },
+    {
+      "epoch": 7.58448098663926,
+      "grad_norm": 0.09046986699104309,
+      "learning_rate": 0.01,
+      "loss": 1.9953,
+      "step": 73797
+    },
+    {
+      "epoch": 7.5847893114080165,
+      "grad_norm": 0.03161385655403137,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 73800
+    },
+    {
+      "epoch": 7.585097636176773,
+      "grad_norm": 0.0820087417960167,
+      "learning_rate": 0.01,
+      "loss": 1.9976,
+      "step": 73803
+    },
+    {
+      "epoch": 7.585405960945529,
+      "grad_norm": 0.04556608945131302,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 73806
+    },
+    {
+      "epoch": 7.585714285714285,
+      "grad_norm": 0.03845854103565216,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 73809
+    },
+    {
+      "epoch": 7.586022610483042,
+      "grad_norm": 0.04036925733089447,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 73812
+    },
+    {
+      "epoch": 7.586330935251799,
+      "grad_norm": 0.12400572001934052,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 73815
+    },
+    {
+      "epoch": 7.586639260020555,
+      "grad_norm": 0.11692599207162857,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 73818
+    },
+    {
+      "epoch": 7.5869475847893115,
+      "grad_norm": 0.10816165804862976,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 73821
+    },
+    {
+      "epoch": 7.587255909558068,
+      "grad_norm": 0.08663230389356613,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 73824
+    },
+    {
+      "epoch": 7.587564234326824,
+      "grad_norm": 0.10679950565099716,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 73827
+    },
+    {
+      "epoch": 7.5878725590955804,
+      "grad_norm": 0.08458245545625687,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 73830
+    },
+    {
+      "epoch": 7.588180883864337,
+      "grad_norm": 0.053197022527456284,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 73833
+    },
+    {
+      "epoch": 7.588489208633094,
+      "grad_norm": 0.033534977585077286,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 73836
+    },
+    {
+      "epoch": 7.58879753340185,
+      "grad_norm": 0.11700337380170822,
+      "learning_rate": 0.01,
+      "loss": 2.0056,
+      "step": 73839
+    },
+    {
+      "epoch": 7.589105858170607,
+      "grad_norm": 0.26823991537094116,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 73842
+    },
+    {
+      "epoch": 7.589414182939363,
+      "grad_norm": 0.04270010069012642,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 73845
+    },
+    {
+      "epoch": 7.589722507708119,
+      "grad_norm": 0.07512800395488739,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 73848
+    },
+    {
+      "epoch": 7.5900308324768755,
+      "grad_norm": 0.049686167389154434,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 73851
+    },
+    {
+      "epoch": 7.590339157245632,
+      "grad_norm": 0.036004502326250076,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 73854
+    },
+    {
+      "epoch": 7.590647482014388,
+      "grad_norm": 0.03266850858926773,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 73857
+    },
+    {
+      "epoch": 7.590955806783145,
+      "grad_norm": 0.041615623980760574,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 73860
+    },
+    {
+      "epoch": 7.591264131551902,
+      "grad_norm": 0.07116426527500153,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 73863
+    },
+    {
+      "epoch": 7.591572456320658,
+      "grad_norm": 0.10572681576013565,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 73866
+    },
+    {
+      "epoch": 7.591880781089414,
+      "grad_norm": 0.06464870274066925,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 73869
+    },
+    {
+      "epoch": 7.5921891058581705,
+      "grad_norm": 0.062195464968681335,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 73872
+    },
+    {
+      "epoch": 7.592497430626927,
+      "grad_norm": 0.05104728415608406,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 73875
+    },
+    {
+      "epoch": 7.592805755395683,
+      "grad_norm": 0.039663366973400116,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 73878
+    },
+    {
+      "epoch": 7.59311408016444,
+      "grad_norm": 0.030902177095413208,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 73881
+    },
+    {
+      "epoch": 7.593422404933197,
+      "grad_norm": 0.09647796303033829,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 73884
+    },
+    {
+      "epoch": 7.593730729701953,
+      "grad_norm": 0.08927876502275467,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 73887
+    },
+    {
+      "epoch": 7.594039054470709,
+      "grad_norm": 0.07941711694002151,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 73890
+    },
+    {
+      "epoch": 7.594347379239466,
+      "grad_norm": 0.067719966173172,
+      "learning_rate": 0.01,
+      "loss": 1.9934,
+      "step": 73893
+    },
+    {
+      "epoch": 7.594655704008222,
+      "grad_norm": 0.07280530035495758,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 73896
+    },
+    {
+      "epoch": 7.594964028776978,
+      "grad_norm": 0.04862016811966896,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 73899
+    },
+    {
+      "epoch": 7.5952723535457345,
+      "grad_norm": 0.03787263110280037,
+      "learning_rate": 0.01,
+      "loss": 2.0097,
+      "step": 73902
+    },
+    {
+      "epoch": 7.595580678314491,
+      "grad_norm": 0.05039701610803604,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 73905
+    },
+    {
+      "epoch": 7.595889003083248,
+      "grad_norm": 0.1198352724313736,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 73908
+    },
+    {
+      "epoch": 7.596197327852004,
+      "grad_norm": 0.07588653266429901,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 73911
+    },
+    {
+      "epoch": 7.596505652620761,
+      "grad_norm": 0.13404078781604767,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 73914
+    },
+    {
+      "epoch": 7.596813977389517,
+      "grad_norm": 0.14311884343624115,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 73917
+    },
+    {
+      "epoch": 7.597122302158273,
+      "grad_norm": 0.07449166476726532,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 73920
+    },
+    {
+      "epoch": 7.5974306269270295,
+      "grad_norm": 0.06578455120325089,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 73923
+    },
+    {
+      "epoch": 7.597738951695787,
+      "grad_norm": 0.04347372427582741,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 73926
+    },
+    {
+      "epoch": 7.598047276464543,
+      "grad_norm": 0.09917577356100082,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 73929
+    },
+    {
+      "epoch": 7.598355601233299,
+      "grad_norm": 0.11529657244682312,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 73932
+    },
+    {
+      "epoch": 7.598663926002056,
+      "grad_norm": 0.05575413629412651,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 73935
+    },
+    {
+      "epoch": 7.598972250770812,
+      "grad_norm": 0.039321184158325195,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 73938
+    },
+    {
+      "epoch": 7.599280575539568,
+      "grad_norm": 0.038657065480947495,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 73941
+    },
+    {
+      "epoch": 7.599588900308325,
+      "grad_norm": 0.11320240050554276,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 73944
+    },
+    {
+      "epoch": 7.599897225077081,
+      "grad_norm": 0.04929359257221222,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 73947
+    },
+    {
+      "epoch": 7.600205549845837,
+      "grad_norm": 0.06260153651237488,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 73950
+    },
+    {
+      "epoch": 7.600513874614594,
+      "grad_norm": 0.06550951302051544,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 73953
+    },
+    {
+      "epoch": 7.600822199383351,
+      "grad_norm": 0.056575268507003784,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 73956
+    },
+    {
+      "epoch": 7.601130524152107,
+      "grad_norm": 0.06484346836805344,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 73959
+    },
+    {
+      "epoch": 7.601438848920863,
+      "grad_norm": 0.06656456738710403,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 73962
+    },
+    {
+      "epoch": 7.60174717368962,
+      "grad_norm": 0.08706523478031158,
+      "learning_rate": 0.01,
+      "loss": 2.0047,
+      "step": 73965
+    },
+    {
+      "epoch": 7.602055498458376,
+      "grad_norm": 0.08907262235879898,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 73968
+    },
+    {
+      "epoch": 7.602363823227132,
+      "grad_norm": 0.08218540251255035,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 73971
+    },
+    {
+      "epoch": 7.602672147995889,
+      "grad_norm": 0.04430277273058891,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 73974
+    },
+    {
+      "epoch": 7.602980472764646,
+      "grad_norm": 0.04185359552502632,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 73977
+    },
+    {
+      "epoch": 7.603288797533402,
+      "grad_norm": 0.14391767978668213,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 73980
+    },
+    {
+      "epoch": 7.603597122302158,
+      "grad_norm": 0.04302511736750603,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 73983
+    },
+    {
+      "epoch": 7.603905447070915,
+      "grad_norm": 0.09394198656082153,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 73986
+    },
+    {
+      "epoch": 7.604213771839671,
+      "grad_norm": 0.04285489022731781,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 73989
+    },
+    {
+      "epoch": 7.604522096608427,
+      "grad_norm": 0.05482661724090576,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 73992
+    },
+    {
+      "epoch": 7.604830421377184,
+      "grad_norm": 0.08892566710710526,
+      "learning_rate": 0.01,
+      "loss": 1.9877,
+      "step": 73995
+    },
+    {
+      "epoch": 7.605138746145941,
+      "grad_norm": 0.04641614109277725,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 73998
+    },
+    {
+      "epoch": 7.605447070914697,
+      "grad_norm": 0.06017431989312172,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 74001
+    },
+    {
+      "epoch": 7.605755395683453,
+      "grad_norm": 0.11227381974458694,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 74004
+    },
+    {
+      "epoch": 7.60606372045221,
+      "grad_norm": 0.05176539719104767,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 74007
+    },
+    {
+      "epoch": 7.606372045220966,
+      "grad_norm": 0.08115067332983017,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 74010
+    },
+    {
+      "epoch": 7.606680369989722,
+      "grad_norm": 0.08569370210170746,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 74013
+    },
+    {
+      "epoch": 7.606988694758479,
+      "grad_norm": 0.037027645856142044,
+      "learning_rate": 0.01,
+      "loss": 1.9993,
+      "step": 74016
+    },
+    {
+      "epoch": 7.607297019527236,
+      "grad_norm": 0.049683477729558945,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 74019
+    },
+    {
+      "epoch": 7.607605344295992,
+      "grad_norm": 0.040880728513002396,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 74022
+    },
+    {
+      "epoch": 7.607913669064748,
+      "grad_norm": 0.07329095155000687,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 74025
+    },
+    {
+      "epoch": 7.608221993833505,
+      "grad_norm": 0.12572570145130157,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 74028
+    },
+    {
+      "epoch": 7.608530318602261,
+      "grad_norm": 0.045191969722509384,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 74031
+    },
+    {
+      "epoch": 7.608838643371017,
+      "grad_norm": 0.05481408163905144,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 74034
+    },
+    {
+      "epoch": 7.609146968139774,
+      "grad_norm": 0.07012124359607697,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 74037
+    },
+    {
+      "epoch": 7.60945529290853,
+      "grad_norm": 0.08710244297981262,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 74040
+    },
+    {
+      "epoch": 7.609763617677286,
+      "grad_norm": 0.06888248026371002,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 74043
+    },
+    {
+      "epoch": 7.6100719424460435,
+      "grad_norm": 0.09116239845752716,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 74046
+    },
+    {
+      "epoch": 7.6103802672148,
+      "grad_norm": 0.14548470079898834,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 74049
+    },
+    {
+      "epoch": 7.610688591983556,
+      "grad_norm": 0.030288826674222946,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 74052
+    },
+    {
+      "epoch": 7.610996916752312,
+      "grad_norm": 0.03586267679929733,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 74055
+    },
+    {
+      "epoch": 7.611305241521069,
+      "grad_norm": 0.03944223001599312,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 74058
+    },
+    {
+      "epoch": 7.611613566289825,
+      "grad_norm": 0.06248027831315994,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 74061
+    },
+    {
+      "epoch": 7.611921891058582,
+      "grad_norm": 0.06302045285701752,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 74064
+    },
+    {
+      "epoch": 7.6122302158273385,
+      "grad_norm": 0.06776783615350723,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 74067
+    },
+    {
+      "epoch": 7.612538540596095,
+      "grad_norm": 0.11014634370803833,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 74070
+    },
+    {
+      "epoch": 7.612846865364851,
+      "grad_norm": 0.23355001211166382,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 74073
+    },
+    {
+      "epoch": 7.613155190133607,
+      "grad_norm": 0.17155928909778595,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 74076
+    },
+    {
+      "epoch": 7.613463514902364,
+      "grad_norm": 0.08190061151981354,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 74079
+    },
+    {
+      "epoch": 7.61377183967112,
+      "grad_norm": 0.047722749412059784,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 74082
+    },
+    {
+      "epoch": 7.614080164439876,
+      "grad_norm": 0.03172357380390167,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 74085
+    },
+    {
+      "epoch": 7.614388489208633,
+      "grad_norm": 0.044041916728019714,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 74088
+    },
+    {
+      "epoch": 7.61469681397739,
+      "grad_norm": 0.07261842489242554,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 74091
+    },
+    {
+      "epoch": 7.615005138746146,
+      "grad_norm": 0.04767949879169464,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 74094
+    },
+    {
+      "epoch": 7.6153134635149025,
+      "grad_norm": 0.059191979467868805,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 74097
+    },
+    {
+      "epoch": 7.615621788283659,
+      "grad_norm": 0.044217757880687714,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 74100
+    },
+    {
+      "epoch": 7.615930113052415,
+      "grad_norm": 0.048439063131809235,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 74103
+    },
+    {
+      "epoch": 7.616238437821171,
+      "grad_norm": 0.0705018863081932,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 74106
+    },
+    {
+      "epoch": 7.616546762589928,
+      "grad_norm": 0.07772061973810196,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 74109
+    },
+    {
+      "epoch": 7.616855087358685,
+      "grad_norm": 0.1228966936469078,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 74112
+    },
+    {
+      "epoch": 7.617163412127441,
+      "grad_norm": 0.056798454374074936,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 74115
+    },
+    {
+      "epoch": 7.6174717368961975,
+      "grad_norm": 0.07300012558698654,
+      "learning_rate": 0.01,
+      "loss": 1.9922,
+      "step": 74118
+    },
+    {
+      "epoch": 7.617780061664954,
+      "grad_norm": 0.034426249563694,
+      "learning_rate": 0.01,
+      "loss": 2.0117,
+      "step": 74121
+    },
+    {
+      "epoch": 7.61808838643371,
+      "grad_norm": 0.03159654513001442,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 74124
+    },
+    {
+      "epoch": 7.618396711202466,
+      "grad_norm": 0.04546064883470535,
+      "learning_rate": 0.01,
+      "loss": 1.9969,
+      "step": 74127
+    },
+    {
+      "epoch": 7.618705035971223,
+      "grad_norm": 0.0788118839263916,
+      "learning_rate": 0.01,
+      "loss": 2.0015,
+      "step": 74130
+    },
+    {
+      "epoch": 7.619013360739979,
+      "grad_norm": 0.06187133863568306,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 74133
+    },
+    {
+      "epoch": 7.619321685508736,
+      "grad_norm": 0.11151747405529022,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 74136
+    },
+    {
+      "epoch": 7.6196300102774925,
+      "grad_norm": 0.08039221912622452,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 74139
+    },
+    {
+      "epoch": 7.619938335046249,
+      "grad_norm": 0.08473746478557587,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 74142
+    },
+    {
+      "epoch": 7.620246659815005,
+      "grad_norm": 0.05223767086863518,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 74145
+    },
+    {
+      "epoch": 7.6205549845837615,
+      "grad_norm": 0.038708411157131195,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 74148
+    },
+    {
+      "epoch": 7.620863309352518,
+      "grad_norm": 0.04193318262696266,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 74151
+    },
+    {
+      "epoch": 7.621171634121274,
+      "grad_norm": 0.036744605749845505,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 74154
+    },
+    {
+      "epoch": 7.621479958890031,
+      "grad_norm": 0.04393354803323746,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 74157
+    },
+    {
+      "epoch": 7.621788283658788,
+      "grad_norm": 0.04185178875923157,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 74160
+    },
+    {
+      "epoch": 7.622096608427544,
+      "grad_norm": 0.09790702164173126,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 74163
+    },
+    {
+      "epoch": 7.6224049331963,
+      "grad_norm": 0.05819176137447357,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 74166
+    },
+    {
+      "epoch": 7.6227132579650565,
+      "grad_norm": 0.08030474185943604,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 74169
+    },
+    {
+      "epoch": 7.623021582733813,
+      "grad_norm": 0.1077614575624466,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 74172
+    },
+    {
+      "epoch": 7.623329907502569,
+      "grad_norm": 0.06347894668579102,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 74175
+    },
+    {
+      "epoch": 7.623638232271325,
+      "grad_norm": 0.08716027438640594,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 74178
+    },
+    {
+      "epoch": 7.623946557040082,
+      "grad_norm": 0.04236210137605667,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 74181
+    },
+    {
+      "epoch": 7.624254881808839,
+      "grad_norm": 0.05839502438902855,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 74184
+    },
+    {
+      "epoch": 7.624563206577595,
+      "grad_norm": 0.09577804803848267,
+      "learning_rate": 0.01,
+      "loss": 1.9956,
+      "step": 74187
+    },
+    {
+      "epoch": 7.6248715313463515,
+      "grad_norm": 0.1194511279463768,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 74190
+    },
+    {
+      "epoch": 7.625179856115108,
+      "grad_norm": 0.09263192117214203,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 74193
+    },
+    {
+      "epoch": 7.625488180883864,
+      "grad_norm": 0.042605165392160416,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 74196
+    },
+    {
+      "epoch": 7.6257965056526205,
+      "grad_norm": 0.05061962082982063,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 74199
+    },
+    {
+      "epoch": 7.626104830421378,
+      "grad_norm": 0.03659835457801819,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 74202
+    },
+    {
+      "epoch": 7.626413155190134,
+      "grad_norm": 0.03055495396256447,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 74205
+    },
+    {
+      "epoch": 7.62672147995889,
+      "grad_norm": 0.10416147112846375,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 74208
+    },
+    {
+      "epoch": 7.627029804727647,
+      "grad_norm": 0.07998328655958176,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 74211
+    },
+    {
+      "epoch": 7.627338129496403,
+      "grad_norm": 0.0935954824090004,
+      "learning_rate": 0.01,
+      "loss": 1.995,
+      "step": 74214
+    },
+    {
+      "epoch": 7.627646454265159,
+      "grad_norm": 0.07641303539276123,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 74217
+    },
+    {
+      "epoch": 7.6279547790339155,
+      "grad_norm": 0.0807437002658844,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 74220
+    },
+    {
+      "epoch": 7.628263103802672,
+      "grad_norm": 0.1031765341758728,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 74223
+    },
+    {
+      "epoch": 7.628571428571428,
+      "grad_norm": 0.06459169834852219,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 74226
+    },
+    {
+      "epoch": 7.628879753340185,
+      "grad_norm": 0.04952657222747803,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 74229
+    },
+    {
+      "epoch": 7.629188078108942,
+      "grad_norm": 0.03405149653553963,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 74232
+    },
+    {
+      "epoch": 7.629496402877698,
+      "grad_norm": 0.04447784274816513,
+      "learning_rate": 0.01,
+      "loss": 1.9852,
+      "step": 74235
+    },
+    {
+      "epoch": 7.629804727646454,
+      "grad_norm": 0.07677244395017624,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 74238
+    },
+    {
+      "epoch": 7.6301130524152105,
+      "grad_norm": 0.1031339168548584,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 74241
+    },
+    {
+      "epoch": 7.630421377183967,
+      "grad_norm": 0.1357228308916092,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 74244
+    },
+    {
+      "epoch": 7.630729701952723,
+      "grad_norm": 0.14237457513809204,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 74247
+    },
+    {
+      "epoch": 7.63103802672148,
+      "grad_norm": 0.11732343584299088,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 74250
+    },
+    {
+      "epoch": 7.631346351490237,
+      "grad_norm": 0.056218016892671585,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 74253
+    },
+    {
+      "epoch": 7.631654676258993,
+      "grad_norm": 0.0473707839846611,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 74256
+    },
+    {
+      "epoch": 7.631963001027749,
+      "grad_norm": 0.06250836700201035,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 74259
+    },
+    {
+      "epoch": 7.632271325796506,
+      "grad_norm": 0.039323173463344574,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 74262
+    },
+    {
+      "epoch": 7.632579650565262,
+      "grad_norm": 0.1005803793668747,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 74265
+    },
+    {
+      "epoch": 7.632887975334018,
+      "grad_norm": 0.08147422969341278,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 74268
+    },
+    {
+      "epoch": 7.6331963001027745,
+      "grad_norm": 0.07019870728254318,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 74271
+    },
+    {
+      "epoch": 7.633504624871532,
+      "grad_norm": 0.08461744338274002,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 74274
+    },
+    {
+      "epoch": 7.633812949640288,
+      "grad_norm": 0.129719540476799,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 74277
+    },
+    {
+      "epoch": 7.634121274409044,
+      "grad_norm": 0.17371834814548492,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 74280
+    },
+    {
+      "epoch": 7.634429599177801,
+      "grad_norm": 0.0878586545586586,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 74283
+    },
+    {
+      "epoch": 7.634737923946557,
+      "grad_norm": 0.04784208908677101,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 74286
+    },
+    {
+      "epoch": 7.635046248715313,
+      "grad_norm": 0.05528968945145607,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 74289
+    },
+    {
+      "epoch": 7.6353545734840695,
+      "grad_norm": 0.04713103175163269,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 74292
+    },
+    {
+      "epoch": 7.635662898252827,
+      "grad_norm": 0.13385649025440216,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 74295
+    },
+    {
+      "epoch": 7.635971223021583,
+      "grad_norm": 0.09349873661994934,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 74298
+    },
+    {
+      "epoch": 7.636279547790339,
+      "grad_norm": 0.07165966928005219,
+      "learning_rate": 0.01,
+      "loss": 1.9984,
+      "step": 74301
+    },
+    {
+      "epoch": 7.636587872559096,
+      "grad_norm": 0.04691112041473389,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 74304
+    },
+    {
+      "epoch": 7.636896197327852,
+      "grad_norm": 0.0479462556540966,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 74307
+    },
+    {
+      "epoch": 7.637204522096608,
+      "grad_norm": 0.07629293203353882,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 74310
+    },
+    {
+      "epoch": 7.637512846865365,
+      "grad_norm": 0.09805203974246979,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 74313
+    },
+    {
+      "epoch": 7.637821171634121,
+      "grad_norm": 0.07098491489887238,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 74316
+    },
+    {
+      "epoch": 7.638129496402878,
+      "grad_norm": 0.06553947925567627,
+      "learning_rate": 0.01,
+      "loss": 1.9971,
+      "step": 74319
+    },
+    {
+      "epoch": 7.638437821171634,
+      "grad_norm": 0.04797985777258873,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 74322
+    },
+    {
+      "epoch": 7.638746145940391,
+      "grad_norm": 0.05064231529831886,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 74325
+    },
+    {
+      "epoch": 7.639054470709147,
+      "grad_norm": 0.07695363461971283,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 74328
+    },
+    {
+      "epoch": 7.639362795477903,
+      "grad_norm": 0.09473993629217148,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 74331
+    },
+    {
+      "epoch": 7.63967112024666,
+      "grad_norm": 0.05327700823545456,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 74334
+    },
+    {
+      "epoch": 7.639979445015416,
+      "grad_norm": 0.07472631335258484,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 74337
+    },
+    {
+      "epoch": 7.640287769784173,
+      "grad_norm": 0.04138251394033432,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 74340
+    },
+    {
+      "epoch": 7.640596094552929,
+      "grad_norm": 0.08596426993608475,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 74343
+    },
+    {
+      "epoch": 7.640904419321686,
+      "grad_norm": 0.08237265050411224,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 74346
+    },
+    {
+      "epoch": 7.641212744090442,
+      "grad_norm": 0.09412717074155807,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 74349
+    },
+    {
+      "epoch": 7.641521068859198,
+      "grad_norm": 0.07472441345453262,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 74352
+    },
+    {
+      "epoch": 7.641829393627955,
+      "grad_norm": 0.06872202455997467,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 74355
+    },
+    {
+      "epoch": 7.642137718396711,
+      "grad_norm": 0.05549430847167969,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 74358
+    },
+    {
+      "epoch": 7.642446043165467,
+      "grad_norm": 0.08854615688323975,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 74361
+    },
+    {
+      "epoch": 7.642754367934224,
+      "grad_norm": 0.12863200902938843,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 74364
+    },
+    {
+      "epoch": 7.643062692702981,
+      "grad_norm": 0.08491957187652588,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 74367
+    },
+    {
+      "epoch": 7.643371017471737,
+      "grad_norm": 0.04691247269511223,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 74370
+    },
+    {
+      "epoch": 7.643679342240493,
+      "grad_norm": 0.05789770558476448,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 74373
+    },
+    {
+      "epoch": 7.64398766700925,
+      "grad_norm": 0.051950305700302124,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 74376
+    },
+    {
+      "epoch": 7.644295991778006,
+      "grad_norm": 0.12850762903690338,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 74379
+    },
+    {
+      "epoch": 7.644604316546762,
+      "grad_norm": 0.07465063780546188,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 74382
+    },
+    {
+      "epoch": 7.6449126413155195,
+      "grad_norm": 0.06719266623258591,
+      "learning_rate": 0.01,
+      "loss": 1.9874,
+      "step": 74385
+    },
+    {
+      "epoch": 7.645220966084276,
+      "grad_norm": 0.06720881164073944,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 74388
+    },
+    {
+      "epoch": 7.645529290853032,
+      "grad_norm": 0.08389756828546524,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 74391
+    },
+    {
+      "epoch": 7.645837615621788,
+      "grad_norm": 0.14512526988983154,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 74394
+    },
+    {
+      "epoch": 7.646145940390545,
+      "grad_norm": 0.07856389135122299,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 74397
+    },
+    {
+      "epoch": 7.646454265159301,
+      "grad_norm": 0.06200588494539261,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 74400
+    },
+    {
+      "epoch": 7.646762589928057,
+      "grad_norm": 0.06547322124242783,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 74403
+    },
+    {
+      "epoch": 7.647070914696814,
+      "grad_norm": 0.04517245292663574,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 74406
+    },
+    {
+      "epoch": 7.64737923946557,
+      "grad_norm": 0.039127618074417114,
+      "learning_rate": 0.01,
+      "loss": 1.9918,
+      "step": 74409
+    },
+    {
+      "epoch": 7.647687564234327,
+      "grad_norm": 0.12178776413202286,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 74412
+    },
+    {
+      "epoch": 7.6479958890030835,
+      "grad_norm": 0.03933233022689819,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 74415
+    },
+    {
+      "epoch": 7.64830421377184,
+      "grad_norm": 0.10145100206136703,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 74418
+    },
+    {
+      "epoch": 7.648612538540596,
+      "grad_norm": 0.0637749508023262,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 74421
+    },
+    {
+      "epoch": 7.648920863309352,
+      "grad_norm": 0.051577746868133545,
+      "learning_rate": 0.01,
+      "loss": 1.9965,
+      "step": 74424
+    },
+    {
+      "epoch": 7.649229188078109,
+      "grad_norm": 0.08104600012302399,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 74427
+    },
+    {
+      "epoch": 7.649537512846865,
+      "grad_norm": 0.09299812465906143,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 74430
+    },
+    {
+      "epoch": 7.649845837615622,
+      "grad_norm": 0.06638593971729279,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 74433
+    },
+    {
+      "epoch": 7.6501541623843785,
+      "grad_norm": 0.0563388355076313,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 74436
+    },
+    {
+      "epoch": 7.650462487153135,
+      "grad_norm": 0.06957016885280609,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 74439
+    },
+    {
+      "epoch": 7.650770811921891,
+      "grad_norm": 0.034762538969516754,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 74442
+    },
+    {
+      "epoch": 7.651079136690647,
+      "grad_norm": 0.043936941772699356,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 74445
+    },
+    {
+      "epoch": 7.651387461459404,
+      "grad_norm": 0.03923318162560463,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 74448
+    },
+    {
+      "epoch": 7.65169578622816,
+      "grad_norm": 0.07552552223205566,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 74451
+    },
+    {
+      "epoch": 7.652004110996916,
+      "grad_norm": 0.05596005916595459,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 74454
+    },
+    {
+      "epoch": 7.6523124357656735,
+      "grad_norm": 0.056482747197151184,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 74457
+    },
+    {
+      "epoch": 7.65262076053443,
+      "grad_norm": 0.06143519654870033,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 74460
+    },
+    {
+      "epoch": 7.652929085303186,
+      "grad_norm": 0.07248186320066452,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 74463
+    },
+    {
+      "epoch": 7.6532374100719425,
+      "grad_norm": 0.14804747700691223,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 74466
+    },
+    {
+      "epoch": 7.653545734840699,
+      "grad_norm": 0.03471624106168747,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 74469
+    },
+    {
+      "epoch": 7.653854059609455,
+      "grad_norm": 0.09821716696023941,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 74472
+    },
+    {
+      "epoch": 7.654162384378211,
+      "grad_norm": 0.09262007474899292,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 74475
+    },
+    {
+      "epoch": 7.654470709146969,
+      "grad_norm": 0.04531139135360718,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 74478
+    },
+    {
+      "epoch": 7.654779033915725,
+      "grad_norm": 0.04309883713722229,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 74481
+    },
+    {
+      "epoch": 7.655087358684481,
+      "grad_norm": 0.05365990102291107,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 74484
+    },
+    {
+      "epoch": 7.6553956834532375,
+      "grad_norm": 0.03971986845135689,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 74487
+    },
+    {
+      "epoch": 7.655704008221994,
+      "grad_norm": 0.05418335273861885,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 74490
+    },
+    {
+      "epoch": 7.65601233299075,
+      "grad_norm": 0.043865058571100235,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 74493
+    },
+    {
+      "epoch": 7.656320657759506,
+      "grad_norm": 0.17038394510746002,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 74496
+    },
+    {
+      "epoch": 7.656628982528263,
+      "grad_norm": 0.07129910588264465,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 74499
+    },
+    {
+      "epoch": 7.656937307297019,
+      "grad_norm": 0.0684434175491333,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 74502
+    },
+    {
+      "epoch": 7.657245632065776,
+      "grad_norm": 0.04037448391318321,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 74505
+    },
+    {
+      "epoch": 7.6575539568345325,
+      "grad_norm": 0.12347517162561417,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 74508
+    },
+    {
+      "epoch": 7.657862281603289,
+      "grad_norm": 0.061841484159231186,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 74511
+    },
+    {
+      "epoch": 7.658170606372045,
+      "grad_norm": 0.10890454053878784,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 74514
+    },
+    {
+      "epoch": 7.6584789311408015,
+      "grad_norm": 0.06274345517158508,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 74517
+    },
+    {
+      "epoch": 7.658787255909558,
+      "grad_norm": 0.07236039638519287,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 74520
+    },
+    {
+      "epoch": 7.659095580678315,
+      "grad_norm": 0.09525978565216064,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 74523
+    },
+    {
+      "epoch": 7.659403905447071,
+      "grad_norm": 0.056358352303504944,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 74526
+    },
+    {
+      "epoch": 7.659712230215828,
+      "grad_norm": 0.07835443317890167,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 74529
+    },
+    {
+      "epoch": 7.660020554984584,
+      "grad_norm": 0.05551443248987198,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 74532
+    },
+    {
+      "epoch": 7.66032887975334,
+      "grad_norm": 0.05699507147073746,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 74535
+    },
+    {
+      "epoch": 7.6606372045220965,
+      "grad_norm": 0.05638470873236656,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 74538
+    },
+    {
+      "epoch": 7.660945529290853,
+      "grad_norm": 0.09323201328516006,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 74541
+    },
+    {
+      "epoch": 7.661253854059609,
+      "grad_norm": 0.08503230661153793,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 74544
+    },
+    {
+      "epoch": 7.661562178828365,
+      "grad_norm": 0.07198848575353622,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 74547
+    },
+    {
+      "epoch": 7.661870503597123,
+      "grad_norm": 0.07523935288190842,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 74550
+    },
+    {
+      "epoch": 7.662178828365879,
+      "grad_norm": 0.07453587651252747,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 74553
+    },
+    {
+      "epoch": 7.662487153134635,
+      "grad_norm": 0.060776762664318085,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 74556
+    },
+    {
+      "epoch": 7.6627954779033916,
+      "grad_norm": 0.06872498989105225,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 74559
+    },
+    {
+      "epoch": 7.663103802672148,
+      "grad_norm": 0.06788360327482224,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 74562
+    },
+    {
+      "epoch": 7.663412127440904,
+      "grad_norm": 0.04364737123250961,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 74565
+    },
+    {
+      "epoch": 7.6637204522096605,
+      "grad_norm": 0.04367050155997276,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 74568
+    },
+    {
+      "epoch": 7.664028776978418,
+      "grad_norm": 0.053800318390131,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 74571
+    },
+    {
+      "epoch": 7.664337101747174,
+      "grad_norm": 0.049637794494628906,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 74574
+    },
+    {
+      "epoch": 7.66464542651593,
+      "grad_norm": 0.04785891994833946,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 74577
+    },
+    {
+      "epoch": 7.664953751284687,
+      "grad_norm": 0.07477859407663345,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 74580
+    },
+    {
+      "epoch": 7.665262076053443,
+      "grad_norm": 0.06930319964885712,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 74583
+    },
+    {
+      "epoch": 7.665570400822199,
+      "grad_norm": 0.09280212223529816,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 74586
+    },
+    {
+      "epoch": 7.6658787255909555,
+      "grad_norm": 0.07756971567869186,
+      "learning_rate": 0.01,
+      "loss": 1.9962,
+      "step": 74589
+    },
+    {
+      "epoch": 7.666187050359712,
+      "grad_norm": 0.04834478721022606,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 74592
+    },
+    {
+      "epoch": 7.666495375128469,
+      "grad_norm": 0.06786291301250458,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 74595
+    },
+    {
+      "epoch": 7.666803699897225,
+      "grad_norm": 0.05437411740422249,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 74598
+    },
+    {
+      "epoch": 7.667112024665982,
+      "grad_norm": 0.06262870132923126,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 74601
+    },
+    {
+      "epoch": 7.667420349434738,
+      "grad_norm": 0.03245949372649193,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 74604
+    },
+    {
+      "epoch": 7.667728674203494,
+      "grad_norm": 0.029952844604849815,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 74607
+    },
+    {
+      "epoch": 7.6680369989722506,
+      "grad_norm": 0.05287361145019531,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 74610
+    },
+    {
+      "epoch": 7.668345323741007,
+      "grad_norm": 0.049092501401901245,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 74613
+    },
+    {
+      "epoch": 7.668653648509764,
+      "grad_norm": 0.1970340609550476,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 74616
+    },
+    {
+      "epoch": 7.66896197327852,
+      "grad_norm": 0.06653330475091934,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 74619
+    },
+    {
+      "epoch": 7.669270298047277,
+      "grad_norm": 0.08722051233053207,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 74622
+    },
+    {
+      "epoch": 7.669578622816033,
+      "grad_norm": 0.06388158351182938,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 74625
+    },
+    {
+      "epoch": 7.669886947584789,
+      "grad_norm": 0.04464726522564888,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 74628
+    },
+    {
+      "epoch": 7.670195272353546,
+      "grad_norm": 0.03846354782581329,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 74631
+    },
+    {
+      "epoch": 7.670503597122302,
+      "grad_norm": 0.038826361298561096,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 74634
+    },
+    {
+      "epoch": 7.670811921891058,
+      "grad_norm": 0.052090950310230255,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 74637
+    },
+    {
+      "epoch": 7.6711202466598145,
+      "grad_norm": 0.07527358084917068,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 74640
+    },
+    {
+      "epoch": 7.671428571428572,
+      "grad_norm": 0.05919142812490463,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 74643
+    },
+    {
+      "epoch": 7.671736896197328,
+      "grad_norm": 0.03949635475873947,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 74646
+    },
+    {
+      "epoch": 7.672045220966084,
+      "grad_norm": 0.12237614393234253,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 74649
+    },
+    {
+      "epoch": 7.672353545734841,
+      "grad_norm": 0.10136833041906357,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 74652
+    },
+    {
+      "epoch": 7.672661870503597,
+      "grad_norm": 0.11704662442207336,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 74655
+    },
+    {
+      "epoch": 7.672970195272353,
+      "grad_norm": 0.06047737970948219,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 74658
+    },
+    {
+      "epoch": 7.67327852004111,
+      "grad_norm": 0.05044262483716011,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 74661
+    },
+    {
+      "epoch": 7.673586844809867,
+      "grad_norm": 0.042892586439847946,
+      "learning_rate": 0.01,
+      "loss": 1.9935,
+      "step": 74664
+    },
+    {
+      "epoch": 7.673895169578623,
+      "grad_norm": 0.04377809539437294,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 74667
+    },
+    {
+      "epoch": 7.674203494347379,
+      "grad_norm": 0.06385743618011475,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 74670
+    },
+    {
+      "epoch": 7.674511819116136,
+      "grad_norm": 0.04909441992640495,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 74673
+    },
+    {
+      "epoch": 7.674820143884892,
+      "grad_norm": 0.06172732263803482,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 74676
+    },
+    {
+      "epoch": 7.675128468653648,
+      "grad_norm": 0.06348180025815964,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 74679
+    },
+    {
+      "epoch": 7.675436793422405,
+      "grad_norm": 0.12463052570819855,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 74682
+    },
+    {
+      "epoch": 7.675745118191161,
+      "grad_norm": 0.04291809722781181,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 74685
+    },
+    {
+      "epoch": 7.676053442959918,
+      "grad_norm": 0.0535232275724411,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 74688
+    },
+    {
+      "epoch": 7.676361767728674,
+      "grad_norm": 0.0402233712375164,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 74691
+    },
+    {
+      "epoch": 7.676670092497431,
+      "grad_norm": 0.047401294112205505,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 74694
+    },
+    {
+      "epoch": 7.676978417266187,
+      "grad_norm": 0.04413991421461105,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 74697
+    },
+    {
+      "epoch": 7.677286742034943,
+      "grad_norm": 0.037016935646533966,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 74700
+    },
+    {
+      "epoch": 7.6775950668037,
+      "grad_norm": 0.041594166308641434,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 74703
+    },
+    {
+      "epoch": 7.677903391572456,
+      "grad_norm": 0.12671832740306854,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 74706
+    },
+    {
+      "epoch": 7.678211716341213,
+      "grad_norm": 0.04342658072710037,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 74709
+    },
+    {
+      "epoch": 7.678520041109969,
+      "grad_norm": 0.05605921521782875,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 74712
+    },
+    {
+      "epoch": 7.678828365878726,
+      "grad_norm": 0.1270131766796112,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 74715
+    },
+    {
+      "epoch": 7.679136690647482,
+      "grad_norm": 0.07328061014413834,
+      "learning_rate": 0.01,
+      "loss": 1.9882,
+      "step": 74718
+    },
+    {
+      "epoch": 7.679445015416238,
+      "grad_norm": 0.09672074764966965,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 74721
+    },
+    {
+      "epoch": 7.679753340184995,
+      "grad_norm": 0.05370582267642021,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 74724
+    },
+    {
+      "epoch": 7.680061664953751,
+      "grad_norm": 0.059833183884620667,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 74727
+    },
+    {
+      "epoch": 7.680369989722507,
+      "grad_norm": 0.044636938720941544,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 74730
+    },
+    {
+      "epoch": 7.6806783144912645,
+      "grad_norm": 0.0414838008582592,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 74733
+    },
+    {
+      "epoch": 7.680986639260021,
+      "grad_norm": 0.06097675487399101,
+      "learning_rate": 0.01,
+      "loss": 2.0028,
+      "step": 74736
+    },
+    {
+      "epoch": 7.681294964028777,
+      "grad_norm": 0.1072980985045433,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 74739
+    },
+    {
+      "epoch": 7.681603288797533,
+      "grad_norm": 0.0738445371389389,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 74742
+    },
+    {
+      "epoch": 7.68191161356629,
+      "grad_norm": 0.06711167842149734,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 74745
+    },
+    {
+      "epoch": 7.682219938335046,
+      "grad_norm": 0.07318748533725739,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 74748
+    },
+    {
+      "epoch": 7.682528263103802,
+      "grad_norm": 0.07076602429151535,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 74751
+    },
+    {
+      "epoch": 7.6828365878725595,
+      "grad_norm": 0.05887584760785103,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 74754
+    },
+    {
+      "epoch": 7.683144912641316,
+      "grad_norm": 0.1313849240541458,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 74757
+    },
+    {
+      "epoch": 7.683453237410072,
+      "grad_norm": 0.04983710125088692,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 74760
+    },
+    {
+      "epoch": 7.683761562178828,
+      "grad_norm": 0.051432423293590546,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 74763
+    },
+    {
+      "epoch": 7.684069886947585,
+      "grad_norm": 0.09883440285921097,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 74766
+    },
+    {
+      "epoch": 7.684378211716341,
+      "grad_norm": 0.07337962836027145,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 74769
+    },
+    {
+      "epoch": 7.684686536485097,
+      "grad_norm": 0.05677385628223419,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 74772
+    },
+    {
+      "epoch": 7.684994861253854,
+      "grad_norm": 0.04124249145388603,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 74775
+    },
+    {
+      "epoch": 7.685303186022611,
+      "grad_norm": 0.03997712954878807,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 74778
+    },
+    {
+      "epoch": 7.685611510791367,
+      "grad_norm": 0.04726220294833183,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 74781
+    },
+    {
+      "epoch": 7.6859198355601235,
+      "grad_norm": 0.04512409493327141,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 74784
+    },
+    {
+      "epoch": 7.68622816032888,
+      "grad_norm": 0.09864774346351624,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 74787
+    },
+    {
+      "epoch": 7.686536485097636,
+      "grad_norm": 0.062484268099069595,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 74790
+    },
+    {
+      "epoch": 7.686844809866392,
+      "grad_norm": 0.09673088043928146,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 74793
+    },
+    {
+      "epoch": 7.687153134635149,
+      "grad_norm": 0.05277227610349655,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 74796
+    },
+    {
+      "epoch": 7.687461459403906,
+      "grad_norm": 0.06944403797388077,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 74799
+    },
+    {
+      "epoch": 7.687769784172662,
+      "grad_norm": 0.07019193470478058,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 74802
+    },
+    {
+      "epoch": 7.6880781089414185,
+      "grad_norm": 0.12948362529277802,
+      "learning_rate": 0.01,
+      "loss": 1.9872,
+      "step": 74805
+    },
+    {
+      "epoch": 7.688386433710175,
+      "grad_norm": 0.06813304126262665,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 74808
+    },
+    {
+      "epoch": 7.688694758478931,
+      "grad_norm": 0.0443788580596447,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 74811
+    },
+    {
+      "epoch": 7.689003083247687,
+      "grad_norm": 0.04802471399307251,
+      "learning_rate": 0.01,
+      "loss": 1.9944,
+      "step": 74814
+    },
+    {
+      "epoch": 7.689311408016444,
+      "grad_norm": 0.05055609717965126,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 74817
+    },
+    {
+      "epoch": 7.6896197327852,
+      "grad_norm": 0.031212560832500458,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 74820
+    },
+    {
+      "epoch": 7.689928057553956,
+      "grad_norm": 0.11334410309791565,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 74823
+    },
+    {
+      "epoch": 7.690236382322714,
+      "grad_norm": 0.059831589460372925,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 74826
+    },
+    {
+      "epoch": 7.69054470709147,
+      "grad_norm": 0.10885782539844513,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 74829
+    },
+    {
+      "epoch": 7.690853031860226,
+      "grad_norm": 0.088487409055233,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 74832
+    },
+    {
+      "epoch": 7.6911613566289825,
+      "grad_norm": 0.050246331840753555,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 74835
+    },
+    {
+      "epoch": 7.691469681397739,
+      "grad_norm": 0.035412561148405075,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 74838
+    },
+    {
+      "epoch": 7.691778006166495,
+      "grad_norm": 0.08675045520067215,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 74841
+    },
+    {
+      "epoch": 7.692086330935252,
+      "grad_norm": 0.1179322674870491,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 74844
+    },
+    {
+      "epoch": 7.692394655704009,
+      "grad_norm": 0.060479529201984406,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 74847
+    },
+    {
+      "epoch": 7.692702980472765,
+      "grad_norm": 0.03623998537659645,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 74850
+    },
+    {
+      "epoch": 7.693011305241521,
+      "grad_norm": 0.039840538054704666,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 74853
+    },
+    {
+      "epoch": 7.6933196300102775,
+      "grad_norm": 0.0399545356631279,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 74856
+    },
+    {
+      "epoch": 7.693627954779034,
+      "grad_norm": 0.13020242750644684,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 74859
+    },
+    {
+      "epoch": 7.69393627954779,
+      "grad_norm": 0.05399748310446739,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 74862
+    },
+    {
+      "epoch": 7.694244604316546,
+      "grad_norm": 0.043653350323438644,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 74865
+    },
+    {
+      "epoch": 7.694552929085303,
+      "grad_norm": 0.04083268716931343,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 74868
+    },
+    {
+      "epoch": 7.69486125385406,
+      "grad_norm": 0.04305655509233475,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 74871
+    },
+    {
+      "epoch": 7.695169578622816,
+      "grad_norm": 0.06946787238121033,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 74874
+    },
+    {
+      "epoch": 7.695477903391573,
+      "grad_norm": 0.11347746849060059,
+      "learning_rate": 0.01,
+      "loss": 1.9921,
+      "step": 74877
+    },
+    {
+      "epoch": 7.695786228160329,
+      "grad_norm": 0.14138512313365936,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 74880
+    },
+    {
+      "epoch": 7.696094552929085,
+      "grad_norm": 0.05391167849302292,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 74883
+    },
+    {
+      "epoch": 7.6964028776978415,
+      "grad_norm": 0.03958969935774803,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 74886
+    },
+    {
+      "epoch": 7.696711202466598,
+      "grad_norm": 0.04602275788784027,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 74889
+    },
+    {
+      "epoch": 7.697019527235355,
+      "grad_norm": 0.05577864497900009,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 74892
+    },
+    {
+      "epoch": 7.697327852004111,
+      "grad_norm": 0.08632632344961166,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 74895
+    },
+    {
+      "epoch": 7.697636176772868,
+      "grad_norm": 0.06367254257202148,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 74898
+    },
+    {
+      "epoch": 7.697944501541624,
+      "grad_norm": 0.07105473428964615,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 74901
+    },
+    {
+      "epoch": 7.69825282631038,
+      "grad_norm": 0.053121648728847504,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 74904
+    },
+    {
+      "epoch": 7.6985611510791365,
+      "grad_norm": 0.056938666850328445,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 74907
+    },
+    {
+      "epoch": 7.698869475847893,
+      "grad_norm": 0.06268006563186646,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 74910
+    },
+    {
+      "epoch": 7.699177800616649,
+      "grad_norm": 0.06493799388408661,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 74913
+    },
+    {
+      "epoch": 7.699486125385406,
+      "grad_norm": 0.05832339823246002,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 74916
+    },
+    {
+      "epoch": 7.699794450154163,
+      "grad_norm": 0.10625136643648148,
+      "learning_rate": 0.01,
+      "loss": 1.9989,
+      "step": 74919
+    },
+    {
+      "epoch": 7.700102774922919,
+      "grad_norm": 0.04736369848251343,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 74922
+    },
+    {
+      "epoch": 7.700411099691675,
+      "grad_norm": 0.03734734281897545,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 74925
+    },
+    {
+      "epoch": 7.700719424460432,
+      "grad_norm": 0.05115366354584694,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 74928
+    },
+    {
+      "epoch": 7.701027749229188,
+      "grad_norm": 0.03465920314192772,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 74931
+    },
+    {
+      "epoch": 7.701336073997944,
+      "grad_norm": 0.14209450781345367,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 74934
+    },
+    {
+      "epoch": 7.701644398766701,
+      "grad_norm": 0.07537394762039185,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 74937
+    },
+    {
+      "epoch": 7.701952723535458,
+      "grad_norm": 0.03767615184187889,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 74940
+    },
+    {
+      "epoch": 7.702261048304214,
+      "grad_norm": 0.034113120287656784,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 74943
+    },
+    {
+      "epoch": 7.70256937307297,
+      "grad_norm": 0.03793106600642204,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 74946
+    },
+    {
+      "epoch": 7.702877697841727,
+      "grad_norm": 0.041996173560619354,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 74949
+    },
+    {
+      "epoch": 7.703186022610483,
+      "grad_norm": 0.042373474687337875,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 74952
+    },
+    {
+      "epoch": 7.703494347379239,
+      "grad_norm": 0.05560959875583649,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 74955
+    },
+    {
+      "epoch": 7.7038026721479955,
+      "grad_norm": 0.052527736872434616,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 74958
+    },
+    {
+      "epoch": 7.704110996916752,
+      "grad_norm": 0.1509440392255783,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 74961
+    },
+    {
+      "epoch": 7.704419321685509,
+      "grad_norm": 0.07441670447587967,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 74964
+    },
+    {
+      "epoch": 7.704727646454265,
+      "grad_norm": 0.09107119590044022,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 74967
+    },
+    {
+      "epoch": 7.705035971223022,
+      "grad_norm": 0.4417809247970581,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 74970
+    },
+    {
+      "epoch": 7.705344295991778,
+      "grad_norm": 0.049799490720033646,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 74973
+    },
+    {
+      "epoch": 7.705652620760534,
+      "grad_norm": 0.05132484808564186,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 74976
+    },
+    {
+      "epoch": 7.705960945529291,
+      "grad_norm": 0.04304105415940285,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 74979
+    },
+    {
+      "epoch": 7.706269270298048,
+      "grad_norm": 0.05429000407457352,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 74982
+    },
+    {
+      "epoch": 7.706577595066804,
+      "grad_norm": 0.035690829157829285,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 74985
+    },
+    {
+      "epoch": 7.70688591983556,
+      "grad_norm": 0.037559330463409424,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 74988
+    },
+    {
+      "epoch": 7.707194244604317,
+      "grad_norm": 0.03172672912478447,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 74991
+    },
+    {
+      "epoch": 7.707502569373073,
+      "grad_norm": 0.03989896923303604,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 74994
+    },
+    {
+      "epoch": 7.707810894141829,
+      "grad_norm": 0.06976346671581268,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 74997
+    },
+    {
+      "epoch": 7.708119218910586,
+      "grad_norm": 0.0848923996090889,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 75000
+    },
+    {
+      "epoch": 7.708427543679342,
+      "grad_norm": 0.07228533923625946,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 75003
+    },
+    {
+      "epoch": 7.708735868448098,
+      "grad_norm": 0.050755999982357025,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 75006
+    },
+    {
+      "epoch": 7.709044193216855,
+      "grad_norm": 0.137029767036438,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 75009
+    },
+    {
+      "epoch": 7.709352517985612,
+      "grad_norm": 0.061711635440588,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 75012
+    },
+    {
+      "epoch": 7.709660842754368,
+      "grad_norm": 0.047601111233234406,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 75015
+    },
+    {
+      "epoch": 7.709969167523124,
+      "grad_norm": 0.052504342049360275,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 75018
+    },
+    {
+      "epoch": 7.710277492291881,
+      "grad_norm": 0.10551578551530838,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 75021
+    },
+    {
+      "epoch": 7.710585817060637,
+      "grad_norm": 0.07759327441453934,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 75024
+    },
+    {
+      "epoch": 7.710894141829393,
+      "grad_norm": 0.05110660567879677,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 75027
+    },
+    {
+      "epoch": 7.7112024665981505,
+      "grad_norm": 0.040343545377254486,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 75030
+    },
+    {
+      "epoch": 7.711510791366907,
+      "grad_norm": 0.04899616539478302,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 75033
+    },
+    {
+      "epoch": 7.711819116135663,
+      "grad_norm": 0.072836734354496,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 75036
+    },
+    {
+      "epoch": 7.712127440904419,
+      "grad_norm": 0.05975237488746643,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 75039
+    },
+    {
+      "epoch": 7.712435765673176,
+      "grad_norm": 0.0680779218673706,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 75042
+    },
+    {
+      "epoch": 7.712744090441932,
+      "grad_norm": 0.04388720169663429,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 75045
+    },
+    {
+      "epoch": 7.713052415210688,
+      "grad_norm": 0.04224342107772827,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 75048
+    },
+    {
+      "epoch": 7.713360739979445,
+      "grad_norm": 0.05068013072013855,
+      "learning_rate": 0.01,
+      "loss": 2.0021,
+      "step": 75051
+    },
+    {
+      "epoch": 7.713669064748202,
+      "grad_norm": 0.08357524126768112,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 75054
+    },
+    {
+      "epoch": 7.713977389516958,
+      "grad_norm": 0.04642781615257263,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 75057
+    },
+    {
+      "epoch": 7.714285714285714,
+      "grad_norm": 0.039685651659965515,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 75060
+    },
+    {
+      "epoch": 7.714594039054471,
+      "grad_norm": 0.05703577399253845,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 75063
+    },
+    {
+      "epoch": 7.714902363823227,
+      "grad_norm": 0.06070883944630623,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 75066
+    },
+    {
+      "epoch": 7.715210688591983,
+      "grad_norm": 0.12112396210432053,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 75069
+    },
+    {
+      "epoch": 7.71551901336074,
+      "grad_norm": 0.04316992312669754,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 75072
+    },
+    {
+      "epoch": 7.715827338129497,
+      "grad_norm": 0.09497283399105072,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 75075
+    },
+    {
+      "epoch": 7.716135662898253,
+      "grad_norm": 0.1215139627456665,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 75078
+    },
+    {
+      "epoch": 7.7164439876670095,
+      "grad_norm": 0.11401921510696411,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 75081
+    },
+    {
+      "epoch": 7.716752312435766,
+      "grad_norm": 0.034442540258169174,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 75084
+    },
+    {
+      "epoch": 7.717060637204522,
+      "grad_norm": 0.0502055361866951,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 75087
+    },
+    {
+      "epoch": 7.717368961973278,
+      "grad_norm": 0.03723570331931114,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 75090
+    },
+    {
+      "epoch": 7.717677286742035,
+      "grad_norm": 0.03539155796170235,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 75093
+    },
+    {
+      "epoch": 7.717985611510791,
+      "grad_norm": 0.052659958600997925,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 75096
+    },
+    {
+      "epoch": 7.718293936279547,
+      "grad_norm": 0.10651390254497528,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 75099
+    },
+    {
+      "epoch": 7.7186022610483045,
+      "grad_norm": 0.09155957400798798,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 75102
+    },
+    {
+      "epoch": 7.718910585817061,
+      "grad_norm": 0.07794260233640671,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 75105
+    },
+    {
+      "epoch": 7.719218910585817,
+      "grad_norm": 0.10518870502710342,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 75108
+    },
+    {
+      "epoch": 7.719527235354573,
+      "grad_norm": 0.05979525297880173,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 75111
+    },
+    {
+      "epoch": 7.71983556012333,
+      "grad_norm": 0.10988526046276093,
+      "learning_rate": 0.01,
+      "loss": 1.9961,
+      "step": 75114
+    },
+    {
+      "epoch": 7.720143884892086,
+      "grad_norm": 0.0853358581662178,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 75117
+    },
+    {
+      "epoch": 7.720452209660843,
+      "grad_norm": 0.046666599810123444,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 75120
+    },
+    {
+      "epoch": 7.7207605344295995,
+      "grad_norm": 0.05268367752432823,
+      "learning_rate": 0.01,
+      "loss": 1.9879,
+      "step": 75123
+    },
+    {
+      "epoch": 7.721068859198356,
+      "grad_norm": 0.04782354086637497,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 75126
+    },
+    {
+      "epoch": 7.721377183967112,
+      "grad_norm": 0.04205673933029175,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 75129
+    },
+    {
+      "epoch": 7.7216855087358685,
+      "grad_norm": 0.0986146479845047,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 75132
+    },
+    {
+      "epoch": 7.721993833504625,
+      "grad_norm": 0.05076294392347336,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 75135
+    },
+    {
+      "epoch": 7.722302158273381,
+      "grad_norm": 0.08255535364151001,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 75138
+    },
+    {
+      "epoch": 7.722610483042137,
+      "grad_norm": 0.08000553399324417,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 75141
+    },
+    {
+      "epoch": 7.722918807810894,
+      "grad_norm": 0.03682593256235123,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 75144
+    },
+    {
+      "epoch": 7.723227132579651,
+      "grad_norm": 0.09714195877313614,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 75147
+    },
+    {
+      "epoch": 7.723535457348407,
+      "grad_norm": 0.05460646376013756,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 75150
+    },
+    {
+      "epoch": 7.7238437821171635,
+      "grad_norm": 0.03487299382686615,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 75153
+    },
+    {
+      "epoch": 7.72415210688592,
+      "grad_norm": 0.06751782447099686,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 75156
+    },
+    {
+      "epoch": 7.724460431654676,
+      "grad_norm": 0.10416395217180252,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 75159
+    },
+    {
+      "epoch": 7.724768756423432,
+      "grad_norm": 0.12070560455322266,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 75162
+    },
+    {
+      "epoch": 7.725077081192189,
+      "grad_norm": 0.0539083257317543,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 75165
+    },
+    {
+      "epoch": 7.725385405960946,
+      "grad_norm": 0.08535733073949814,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 75168
+    },
+    {
+      "epoch": 7.725693730729702,
+      "grad_norm": 0.047993749380111694,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 75171
+    },
+    {
+      "epoch": 7.7260020554984585,
+      "grad_norm": 0.06361022591590881,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 75174
+    },
+    {
+      "epoch": 7.726310380267215,
+      "grad_norm": 0.049908868968486786,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 75177
+    },
+    {
+      "epoch": 7.726618705035971,
+      "grad_norm": 0.09875412285327911,
+      "learning_rate": 0.01,
+      "loss": 2.001,
+      "step": 75180
+    },
+    {
+      "epoch": 7.7269270298047275,
+      "grad_norm": 0.0869210883975029,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 75183
+    },
+    {
+      "epoch": 7.727235354573484,
+      "grad_norm": 0.0573507621884346,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 75186
+    },
+    {
+      "epoch": 7.72754367934224,
+      "grad_norm": 0.03936505317687988,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 75189
+    },
+    {
+      "epoch": 7.727852004110997,
+      "grad_norm": 0.0410919226706028,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 75192
+    },
+    {
+      "epoch": 7.728160328879754,
+      "grad_norm": 0.082386814057827,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 75195
+    },
+    {
+      "epoch": 7.72846865364851,
+      "grad_norm": 0.04592809081077576,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 75198
+    },
+    {
+      "epoch": 7.728776978417266,
+      "grad_norm": 0.03361072763800621,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 75201
+    },
+    {
+      "epoch": 7.7290853031860225,
+      "grad_norm": 0.08248721063137054,
+      "learning_rate": 0.01,
+      "loss": 1.992,
+      "step": 75204
+    },
+    {
+      "epoch": 7.729393627954779,
+      "grad_norm": 0.05343814939260483,
+      "learning_rate": 0.01,
+      "loss": 1.9968,
+      "step": 75207
+    },
+    {
+      "epoch": 7.729701952723535,
+      "grad_norm": 0.10746193677186966,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 75210
+    },
+    {
+      "epoch": 7.730010277492292,
+      "grad_norm": 0.07758577913045883,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 75213
+    },
+    {
+      "epoch": 7.730318602261049,
+      "grad_norm": 0.05135120823979378,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 75216
+    },
+    {
+      "epoch": 7.730626927029805,
+      "grad_norm": 0.10420482605695724,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 75219
+    },
+    {
+      "epoch": 7.730935251798561,
+      "grad_norm": 0.06070501729846001,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 75222
+    },
+    {
+      "epoch": 7.7312435765673175,
+      "grad_norm": 0.06866266578435898,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 75225
+    },
+    {
+      "epoch": 7.731551901336074,
+      "grad_norm": 0.07661131769418716,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 75228
+    },
+    {
+      "epoch": 7.73186022610483,
+      "grad_norm": 0.12407959252595901,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 75231
+    },
+    {
+      "epoch": 7.7321685508735865,
+      "grad_norm": 0.11292631924152374,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 75234
+    },
+    {
+      "epoch": 7.732476875642344,
+      "grad_norm": 0.10384037345647812,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 75237
+    },
+    {
+      "epoch": 7.7327852004111,
+      "grad_norm": 0.04758400097489357,
+      "learning_rate": 0.01,
+      "loss": 1.9988,
+      "step": 75240
+    },
+    {
+      "epoch": 7.733093525179856,
+      "grad_norm": 0.0348513089120388,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 75243
+    },
+    {
+      "epoch": 7.733401849948613,
+      "grad_norm": 0.03560272976756096,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 75246
+    },
+    {
+      "epoch": 7.733710174717369,
+      "grad_norm": 0.038762979209423065,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 75249
+    },
+    {
+      "epoch": 7.734018499486125,
+      "grad_norm": 0.06412766128778458,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 75252
+    },
+    {
+      "epoch": 7.7343268242548815,
+      "grad_norm": 0.09387977421283722,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 75255
+    },
+    {
+      "epoch": 7.734635149023639,
+      "grad_norm": 0.0382736399769783,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 75258
+    },
+    {
+      "epoch": 7.734943473792395,
+      "grad_norm": 0.061939794570207596,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 75261
+    },
+    {
+      "epoch": 7.735251798561151,
+      "grad_norm": 0.10443083941936493,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 75264
+    },
+    {
+      "epoch": 7.735560123329908,
+      "grad_norm": 0.10148416459560394,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 75267
+    },
+    {
+      "epoch": 7.735868448098664,
+      "grad_norm": 0.06240933761000633,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 75270
+    },
+    {
+      "epoch": 7.73617677286742,
+      "grad_norm": 0.062156736850738525,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 75273
+    },
+    {
+      "epoch": 7.7364850976361765,
+      "grad_norm": 0.07212886959314346,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 75276
+    },
+    {
+      "epoch": 7.736793422404933,
+      "grad_norm": 0.06498091667890549,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 75279
+    },
+    {
+      "epoch": 7.737101747173689,
+      "grad_norm": 0.07634437829256058,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 75282
+    },
+    {
+      "epoch": 7.737410071942446,
+      "grad_norm": 0.09608074277639389,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 75285
+    },
+    {
+      "epoch": 7.737718396711203,
+      "grad_norm": 0.07703720033168793,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 75288
+    },
+    {
+      "epoch": 7.738026721479959,
+      "grad_norm": 0.07131288200616837,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 75291
+    },
+    {
+      "epoch": 7.738335046248715,
+      "grad_norm": 0.08612216264009476,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 75294
+    },
+    {
+      "epoch": 7.738643371017472,
+      "grad_norm": 0.19164220988750458,
+      "learning_rate": 0.01,
+      "loss": 1.9955,
+      "step": 75297
+    },
+    {
+      "epoch": 7.738951695786228,
+      "grad_norm": 0.06035240739583969,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 75300
+    },
+    {
+      "epoch": 7.739260020554985,
+      "grad_norm": 0.04050587862730026,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 75303
+    },
+    {
+      "epoch": 7.739568345323741,
+      "grad_norm": 0.03798283636569977,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 75306
+    },
+    {
+      "epoch": 7.739876670092498,
+      "grad_norm": 0.06570883095264435,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 75309
+    },
+    {
+      "epoch": 7.740184994861254,
+      "grad_norm": 0.05782955512404442,
+      "learning_rate": 0.01,
+      "loss": 1.9966,
+      "step": 75312
+    },
+    {
+      "epoch": 7.74049331963001,
+      "grad_norm": 0.047876227647066116,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 75315
+    },
+    {
+      "epoch": 7.740801644398767,
+      "grad_norm": 0.06209714338183403,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 75318
+    },
+    {
+      "epoch": 7.741109969167523,
+      "grad_norm": 0.04943614453077316,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 75321
+    },
+    {
+      "epoch": 7.741418293936279,
+      "grad_norm": 0.05134053900837898,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 75324
+    },
+    {
+      "epoch": 7.7417266187050355,
+      "grad_norm": 0.03939725086092949,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 75327
+    },
+    {
+      "epoch": 7.742034943473793,
+      "grad_norm": 0.06609564274549484,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 75330
+    },
+    {
+      "epoch": 7.742343268242549,
+      "grad_norm": 0.06503685563802719,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 75333
+    },
+    {
+      "epoch": 7.742651593011305,
+      "grad_norm": 0.16153855621814728,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 75336
+    },
+    {
+      "epoch": 7.742959917780062,
+      "grad_norm": 0.07932616025209427,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 75339
+    },
+    {
+      "epoch": 7.743268242548818,
+      "grad_norm": 0.12574706971645355,
+      "learning_rate": 0.01,
+      "loss": 1.9941,
+      "step": 75342
+    },
+    {
+      "epoch": 7.743576567317574,
+      "grad_norm": 0.0709739476442337,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 75345
+    },
+    {
+      "epoch": 7.743884892086331,
+      "grad_norm": 0.05811988189816475,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 75348
+    },
+    {
+      "epoch": 7.744193216855088,
+      "grad_norm": 0.05921448767185211,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 75351
+    },
+    {
+      "epoch": 7.744501541623844,
+      "grad_norm": 0.0461372472345829,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 75354
+    },
+    {
+      "epoch": 7.7448098663926,
+      "grad_norm": 0.05297720059752464,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 75357
+    },
+    {
+      "epoch": 7.745118191161357,
+      "grad_norm": 0.05457683280110359,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 75360
+    },
+    {
+      "epoch": 7.745426515930113,
+      "grad_norm": 0.05880630761384964,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 75363
+    },
+    {
+      "epoch": 7.745734840698869,
+      "grad_norm": 0.04461346194148064,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 75366
+    },
+    {
+      "epoch": 7.746043165467626,
+      "grad_norm": 0.036604154855012894,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 75369
+    },
+    {
+      "epoch": 7.746351490236382,
+      "grad_norm": 0.1032840833067894,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 75372
+    },
+    {
+      "epoch": 7.746659815005139,
+      "grad_norm": 0.060783445835113525,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 75375
+    },
+    {
+      "epoch": 7.746968139773895,
+      "grad_norm": 0.09891743212938309,
+      "learning_rate": 0.01,
+      "loss": 1.9895,
+      "step": 75378
+    },
+    {
+      "epoch": 7.747276464542652,
+      "grad_norm": 0.06625793129205704,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 75381
+    },
+    {
+      "epoch": 7.747584789311408,
+      "grad_norm": 0.0494624525308609,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 75384
+    },
+    {
+      "epoch": 7.747893114080164,
+      "grad_norm": 0.06007739156484604,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 75387
+    },
+    {
+      "epoch": 7.748201438848921,
+      "grad_norm": 0.051491837948560715,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 75390
+    },
+    {
+      "epoch": 7.748509763617677,
+      "grad_norm": 0.04195559397339821,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 75393
+    },
+    {
+      "epoch": 7.748818088386434,
+      "grad_norm": 0.05081287398934364,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 75396
+    },
+    {
+      "epoch": 7.7491264131551905,
+      "grad_norm": 0.06100494787096977,
+      "learning_rate": 0.01,
+      "loss": 1.9906,
+      "step": 75399
+    },
+    {
+      "epoch": 7.749434737923947,
+      "grad_norm": 0.18386486172676086,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 75402
+    },
+    {
+      "epoch": 7.749743062692703,
+      "grad_norm": 0.06177636235952377,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 75405
+    },
+    {
+      "epoch": 7.750051387461459,
+      "grad_norm": 0.0422503836452961,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 75408
+    },
+    {
+      "epoch": 7.750359712230216,
+      "grad_norm": 0.0480877049267292,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 75411
+    },
+    {
+      "epoch": 7.750668036998972,
+      "grad_norm": 0.04990262910723686,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 75414
+    },
+    {
+      "epoch": 7.750976361767728,
+      "grad_norm": 0.034894928336143494,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 75417
+    },
+    {
+      "epoch": 7.751284686536485,
+      "grad_norm": 0.05810752511024475,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 75420
+    },
+    {
+      "epoch": 7.751593011305242,
+      "grad_norm": 0.04341167211532593,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 75423
+    },
+    {
+      "epoch": 7.751901336073998,
+      "grad_norm": 0.08866117149591446,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 75426
+    },
+    {
+      "epoch": 7.752209660842754,
+      "grad_norm": 0.13528284430503845,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 75429
+    },
+    {
+      "epoch": 7.752517985611511,
+      "grad_norm": 0.05988186597824097,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 75432
+    },
+    {
+      "epoch": 7.752826310380267,
+      "grad_norm": 0.12178362905979156,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 75435
+    },
+    {
+      "epoch": 7.753134635149023,
+      "grad_norm": 0.08513082563877106,
+      "learning_rate": 0.01,
+      "loss": 1.9998,
+      "step": 75438
+    },
+    {
+      "epoch": 7.7534429599177805,
+      "grad_norm": 0.05664602294564247,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 75441
+    },
+    {
+      "epoch": 7.753751284686537,
+      "grad_norm": 0.059373270720243454,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 75444
+    },
+    {
+      "epoch": 7.754059609455293,
+      "grad_norm": 0.06777229905128479,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 75447
+    },
+    {
+      "epoch": 7.7543679342240495,
+      "grad_norm": 0.06595178693532944,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 75450
+    },
+    {
+      "epoch": 7.754676258992806,
+      "grad_norm": 0.03551295027136803,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 75453
+    },
+    {
+      "epoch": 7.754984583761562,
+      "grad_norm": 0.13232779502868652,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 75456
+    },
+    {
+      "epoch": 7.755292908530318,
+      "grad_norm": 0.11032851040363312,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 75459
+    },
+    {
+      "epoch": 7.755601233299075,
+      "grad_norm": 0.09487207978963852,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 75462
+    },
+    {
+      "epoch": 7.755909558067831,
+      "grad_norm": 0.07793836295604706,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 75465
+    },
+    {
+      "epoch": 7.756217882836588,
+      "grad_norm": 0.0555163212120533,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 75468
+    },
+    {
+      "epoch": 7.7565262076053445,
+      "grad_norm": 0.04495738446712494,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 75471
+    },
+    {
+      "epoch": 7.756834532374101,
+      "grad_norm": 0.03979022800922394,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 75474
+    },
+    {
+      "epoch": 7.757142857142857,
+      "grad_norm": 0.054774973541498184,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 75477
+    },
+    {
+      "epoch": 7.757451181911613,
+      "grad_norm": 0.07539385557174683,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 75480
+    },
+    {
+      "epoch": 7.75775950668037,
+      "grad_norm": 0.05782768502831459,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 75483
+    },
+    {
+      "epoch": 7.758067831449126,
+      "grad_norm": 0.035946596413850784,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 75486
+    },
+    {
+      "epoch": 7.758376156217883,
+      "grad_norm": 0.04099863022565842,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 75489
+    },
+    {
+      "epoch": 7.7586844809866395,
+      "grad_norm": 0.10557963699102402,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 75492
+    },
+    {
+      "epoch": 7.758992805755396,
+      "grad_norm": 0.061207111924886703,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 75495
+    },
+    {
+      "epoch": 7.759301130524152,
+      "grad_norm": 0.08292508125305176,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 75498
+    },
+    {
+      "epoch": 7.7596094552929085,
+      "grad_norm": 0.04073723405599594,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 75501
+    },
+    {
+      "epoch": 7.759917780061665,
+      "grad_norm": 0.05526462569832802,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 75504
+    },
+    {
+      "epoch": 7.760226104830421,
+      "grad_norm": 0.06078262999653816,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 75507
+    },
+    {
+      "epoch": 7.760534429599177,
+      "grad_norm": 0.07975959032773972,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 75510
+    },
+    {
+      "epoch": 7.760842754367935,
+      "grad_norm": 0.06245322525501251,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 75513
+    },
+    {
+      "epoch": 7.761151079136691,
+      "grad_norm": 0.10303689539432526,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 75516
+    },
+    {
+      "epoch": 7.761459403905447,
+      "grad_norm": 0.1304595023393631,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 75519
+    },
+    {
+      "epoch": 7.7617677286742035,
+      "grad_norm": 0.04025387763977051,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 75522
+    },
+    {
+      "epoch": 7.76207605344296,
+      "grad_norm": 0.0802154690027237,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 75525
+    },
+    {
+      "epoch": 7.762384378211716,
+      "grad_norm": 0.09343558549880981,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 75528
+    },
+    {
+      "epoch": 7.762692702980472,
+      "grad_norm": 0.07421892136335373,
+      "learning_rate": 0.01,
+      "loss": 1.9904,
+      "step": 75531
+    },
+    {
+      "epoch": 7.76300102774923,
+      "grad_norm": 0.03761820122599602,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 75534
+    },
+    {
+      "epoch": 7.763309352517986,
+      "grad_norm": 0.02898077666759491,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 75537
+    },
+    {
+      "epoch": 7.763617677286742,
+      "grad_norm": 0.060065463185310364,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 75540
+    },
+    {
+      "epoch": 7.7639260020554985,
+      "grad_norm": 0.08949063718318939,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 75543
+    },
+    {
+      "epoch": 7.764234326824255,
+      "grad_norm": 0.05697142332792282,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 75546
+    },
+    {
+      "epoch": 7.764542651593011,
+      "grad_norm": 0.03366275504231453,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 75549
+    },
+    {
+      "epoch": 7.7648509763617675,
+      "grad_norm": 0.11962536722421646,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 75552
+    },
+    {
+      "epoch": 7.765159301130524,
+      "grad_norm": 0.049452245235443115,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 75555
+    },
+    {
+      "epoch": 7.76546762589928,
+      "grad_norm": 0.10120616108179092,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 75558
+    },
+    {
+      "epoch": 7.765775950668037,
+      "grad_norm": 0.049389541149139404,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 75561
+    },
+    {
+      "epoch": 7.766084275436794,
+      "grad_norm": 0.07128654420375824,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 75564
+    },
+    {
+      "epoch": 7.76639260020555,
+      "grad_norm": 0.05818384140729904,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 75567
+    },
+    {
+      "epoch": 7.766700924974306,
+      "grad_norm": 0.03975757211446762,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 75570
+    },
+    {
+      "epoch": 7.7670092497430625,
+      "grad_norm": 0.046733129769563675,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 75573
+    },
+    {
+      "epoch": 7.767317574511819,
+      "grad_norm": 0.06592223793268204,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 75576
+    },
+    {
+      "epoch": 7.767625899280576,
+      "grad_norm": 0.09464466571807861,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 75579
+    },
+    {
+      "epoch": 7.767934224049332,
+      "grad_norm": 0.07415366172790527,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 75582
+    },
+    {
+      "epoch": 7.768242548818089,
+      "grad_norm": 0.06457046419382095,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 75585
+    },
+    {
+      "epoch": 7.768550873586845,
+      "grad_norm": 0.10259808599948883,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 75588
+    },
+    {
+      "epoch": 7.768859198355601,
+      "grad_norm": 0.186797097325325,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 75591
+    },
+    {
+      "epoch": 7.7691675231243575,
+      "grad_norm": 0.13249987363815308,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 75594
+    },
+    {
+      "epoch": 7.769475847893114,
+      "grad_norm": 0.07284609973430634,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 75597
+    },
+    {
+      "epoch": 7.76978417266187,
+      "grad_norm": 0.05419408157467842,
+      "learning_rate": 0.01,
+      "loss": 2.0002,
+      "step": 75600
+    },
+    {
+      "epoch": 7.7700924974306265,
+      "grad_norm": 0.05547893047332764,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 75603
+    },
+    {
+      "epoch": 7.770400822199384,
+      "grad_norm": 0.04928140714764595,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 75606
+    },
+    {
+      "epoch": 7.77070914696814,
+      "grad_norm": 0.04238557815551758,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 75609
+    },
+    {
+      "epoch": 7.771017471736896,
+      "grad_norm": 0.08550449460744858,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 75612
+    },
+    {
+      "epoch": 7.771325796505653,
+      "grad_norm": 0.07773346453905106,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 75615
+    },
+    {
+      "epoch": 7.771634121274409,
+      "grad_norm": 0.06905487179756165,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 75618
+    },
+    {
+      "epoch": 7.771942446043165,
+      "grad_norm": 0.07177229970693588,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 75621
+    },
+    {
+      "epoch": 7.7722507708119215,
+      "grad_norm": 0.04951430857181549,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 75624
+    },
+    {
+      "epoch": 7.772559095580679,
+      "grad_norm": 0.06549627333879471,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 75627
+    },
+    {
+      "epoch": 7.772867420349435,
+      "grad_norm": 0.07106509059667587,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 75630
+    },
+    {
+      "epoch": 7.773175745118191,
+      "grad_norm": 0.09129591286182404,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 75633
+    },
+    {
+      "epoch": 7.773484069886948,
+      "grad_norm": 0.04063316062092781,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 75636
+    },
+    {
+      "epoch": 7.773792394655704,
+      "grad_norm": 0.13061417639255524,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 75639
+    },
+    {
+      "epoch": 7.77410071942446,
+      "grad_norm": 0.0752875953912735,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 75642
+    },
+    {
+      "epoch": 7.7744090441932165,
+      "grad_norm": 0.03646187111735344,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 75645
+    },
+    {
+      "epoch": 7.774717368961973,
+      "grad_norm": 0.03636857867240906,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 75648
+    },
+    {
+      "epoch": 7.77502569373073,
+      "grad_norm": 0.04543618857860565,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 75651
+    },
+    {
+      "epoch": 7.775334018499486,
+      "grad_norm": 0.048112813383340836,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 75654
+    },
+    {
+      "epoch": 7.775642343268243,
+      "grad_norm": 0.054791998118162155,
+      "learning_rate": 0.01,
+      "loss": 2.0064,
+      "step": 75657
+    },
+    {
+      "epoch": 7.775950668036999,
+      "grad_norm": 0.052178870886564255,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 75660
+    },
+    {
+      "epoch": 7.776258992805755,
+      "grad_norm": 0.05024852603673935,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 75663
+    },
+    {
+      "epoch": 7.776567317574512,
+      "grad_norm": 0.03811731934547424,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 75666
+    },
+    {
+      "epoch": 7.776875642343268,
+      "grad_norm": 0.07078245282173157,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 75669
+    },
+    {
+      "epoch": 7.777183967112025,
+      "grad_norm": 0.07220859825611115,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 75672
+    },
+    {
+      "epoch": 7.777492291880781,
+      "grad_norm": 0.07006082683801651,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 75675
+    },
+    {
+      "epoch": 7.777800616649538,
+      "grad_norm": 0.10420866310596466,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 75678
+    },
+    {
+      "epoch": 7.778108941418294,
+      "grad_norm": 0.08756904304027557,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 75681
+    },
+    {
+      "epoch": 7.77841726618705,
+      "grad_norm": 0.05425836890935898,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 75684
+    },
+    {
+      "epoch": 7.778725590955807,
+      "grad_norm": 0.044821374118328094,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 75687
+    },
+    {
+      "epoch": 7.779033915724563,
+      "grad_norm": 0.059608519077301025,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 75690
+    },
+    {
+      "epoch": 7.779342240493319,
+      "grad_norm": 0.0818125456571579,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 75693
+    },
+    {
+      "epoch": 7.779650565262076,
+      "grad_norm": 0.09120543301105499,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 75696
+    },
+    {
+      "epoch": 7.779958890030833,
+      "grad_norm": 0.12951593101024628,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 75699
+    },
+    {
+      "epoch": 7.780267214799589,
+      "grad_norm": 0.12418895214796066,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 75702
+    },
+    {
+      "epoch": 7.780575539568345,
+      "grad_norm": 0.1442076563835144,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 75705
+    },
+    {
+      "epoch": 7.780883864337102,
+      "grad_norm": 0.1782861202955246,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 75708
+    },
+    {
+      "epoch": 7.781192189105858,
+      "grad_norm": 0.13022981584072113,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 75711
+    },
+    {
+      "epoch": 7.781500513874614,
+      "grad_norm": 0.05073774605989456,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 75714
+    },
+    {
+      "epoch": 7.7818088386433715,
+      "grad_norm": 0.09594649821519852,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 75717
+    },
+    {
+      "epoch": 7.782117163412128,
+      "grad_norm": 0.04567958042025566,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 75720
+    },
+    {
+      "epoch": 7.782425488180884,
+      "grad_norm": 0.0413205660879612,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 75723
+    },
+    {
+      "epoch": 7.78273381294964,
+      "grad_norm": 0.046227239072322845,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 75726
+    },
+    {
+      "epoch": 7.783042137718397,
+      "grad_norm": 0.052021145820617676,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 75729
+    },
+    {
+      "epoch": 7.783350462487153,
+      "grad_norm": 0.05464644357562065,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 75732
+    },
+    {
+      "epoch": 7.783658787255909,
+      "grad_norm": 0.07341251522302628,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 75735
+    },
+    {
+      "epoch": 7.783967112024666,
+      "grad_norm": 0.03740450739860535,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 75738
+    },
+    {
+      "epoch": 7.784275436793422,
+      "grad_norm": 0.051203422248363495,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 75741
+    },
+    {
+      "epoch": 7.784583761562179,
+      "grad_norm": 0.071673683822155,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 75744
+    },
+    {
+      "epoch": 7.784892086330935,
+      "grad_norm": 0.1421179473400116,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 75747
+    },
+    {
+      "epoch": 7.785200411099692,
+      "grad_norm": 0.054482489824295044,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 75750
+    },
+    {
+      "epoch": 7.785508735868448,
+      "grad_norm": 0.03319859504699707,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 75753
+    },
+    {
+      "epoch": 7.785817060637204,
+      "grad_norm": 0.08589328080415726,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 75756
+    },
+    {
+      "epoch": 7.786125385405961,
+      "grad_norm": 0.056132521480321884,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 75759
+    },
+    {
+      "epoch": 7.786433710174718,
+      "grad_norm": 0.10983091592788696,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 75762
+    },
+    {
+      "epoch": 7.786742034943474,
+      "grad_norm": 0.1317470520734787,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 75765
+    },
+    {
+      "epoch": 7.7870503597122305,
+      "grad_norm": 0.1627342402935028,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 75768
+    },
+    {
+      "epoch": 7.787358684480987,
+      "grad_norm": 0.13085448741912842,
+      "learning_rate": 0.01,
+      "loss": 1.982,
+      "step": 75771
+    },
+    {
+      "epoch": 7.787667009249743,
+      "grad_norm": 0.08374191075563431,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 75774
+    },
+    {
+      "epoch": 7.787975334018499,
+      "grad_norm": 0.05581342428922653,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 75777
+    },
+    {
+      "epoch": 7.788283658787256,
+      "grad_norm": 0.07613394409418106,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 75780
+    },
+    {
+      "epoch": 7.788591983556012,
+      "grad_norm": 0.07574499398469925,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 75783
+    },
+    {
+      "epoch": 7.788900308324768,
+      "grad_norm": 0.05884881317615509,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 75786
+    },
+    {
+      "epoch": 7.7892086330935255,
+      "grad_norm": 0.05409419536590576,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 75789
+    },
+    {
+      "epoch": 7.789516957862282,
+      "grad_norm": 0.06512802094221115,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 75792
+    },
+    {
+      "epoch": 7.789825282631038,
+      "grad_norm": 0.04650108888745308,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 75795
+    },
+    {
+      "epoch": 7.790133607399794,
+      "grad_norm": 0.06513743847608566,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 75798
+    },
+    {
+      "epoch": 7.790441932168551,
+      "grad_norm": 0.07151713222265244,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 75801
+    },
+    {
+      "epoch": 7.790750256937307,
+      "grad_norm": 0.09430654346942902,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 75804
+    },
+    {
+      "epoch": 7.791058581706063,
+      "grad_norm": 0.10549556463956833,
+      "learning_rate": 0.01,
+      "loss": 1.9892,
+      "step": 75807
+    },
+    {
+      "epoch": 7.7913669064748206,
+      "grad_norm": 0.11038587987422943,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 75810
+    },
+    {
+      "epoch": 7.791675231243577,
+      "grad_norm": 0.06415724009275436,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 75813
+    },
+    {
+      "epoch": 7.791983556012333,
+      "grad_norm": 0.04490191116929054,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 75816
+    },
+    {
+      "epoch": 7.7922918807810895,
+      "grad_norm": 0.06832468509674072,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 75819
+    },
+    {
+      "epoch": 7.792600205549846,
+      "grad_norm": 0.07226554304361343,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 75822
+    },
+    {
+      "epoch": 7.792908530318602,
+      "grad_norm": 0.05115282163023949,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 75825
+    },
+    {
+      "epoch": 7.793216855087358,
+      "grad_norm": 0.05715862661600113,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 75828
+    },
+    {
+      "epoch": 7.793525179856115,
+      "grad_norm": 0.03245139494538307,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 75831
+    },
+    {
+      "epoch": 7.793833504624872,
+      "grad_norm": 0.036928169429302216,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 75834
+    },
+    {
+      "epoch": 7.794141829393628,
+      "grad_norm": 0.056494537740945816,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 75837
+    },
+    {
+      "epoch": 7.7944501541623845,
+      "grad_norm": 0.12836699187755585,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 75840
+    },
+    {
+      "epoch": 7.794758478931141,
+      "grad_norm": 0.11228854954242706,
+      "learning_rate": 0.01,
+      "loss": 1.9997,
+      "step": 75843
+    },
+    {
+      "epoch": 7.795066803699897,
+      "grad_norm": 0.13982988893985748,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 75846
+    },
+    {
+      "epoch": 7.795375128468653,
+      "grad_norm": 0.05313684791326523,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 75849
+    },
+    {
+      "epoch": 7.79568345323741,
+      "grad_norm": 0.038301192224025726,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 75852
+    },
+    {
+      "epoch": 7.795991778006167,
+      "grad_norm": 0.030965665355324745,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 75855
+    },
+    {
+      "epoch": 7.796300102774923,
+      "grad_norm": 0.05267125740647316,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 75858
+    },
+    {
+      "epoch": 7.7966084275436796,
+      "grad_norm": 0.11764229089021683,
+      "learning_rate": 0.01,
+      "loss": 2.0029,
+      "step": 75861
+    },
+    {
+      "epoch": 7.796916752312436,
+      "grad_norm": 0.12397600710391998,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 75864
+    },
+    {
+      "epoch": 7.797225077081192,
+      "grad_norm": 0.15025153756141663,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 75867
+    },
+    {
+      "epoch": 7.7975334018499485,
+      "grad_norm": 0.09809085726737976,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 75870
+    },
+    {
+      "epoch": 7.797841726618705,
+      "grad_norm": 0.054064489901065826,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 75873
+    },
+    {
+      "epoch": 7.798150051387461,
+      "grad_norm": 0.03437699377536774,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 75876
+    },
+    {
+      "epoch": 7.798458376156217,
+      "grad_norm": 0.03941173478960991,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 75879
+    },
+    {
+      "epoch": 7.798766700924975,
+      "grad_norm": 0.058382656425237656,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 75882
+    },
+    {
+      "epoch": 7.799075025693731,
+      "grad_norm": 0.03915715217590332,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 75885
+    },
+    {
+      "epoch": 7.799383350462487,
+      "grad_norm": 0.03458473086357117,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 75888
+    },
+    {
+      "epoch": 7.7996916752312435,
+      "grad_norm": 0.03535837680101395,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 75891
+    },
+    {
+      "epoch": 7.8,
+      "grad_norm": 0.10532993078231812,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 75894
+    },
+    {
+      "epoch": 7.800308324768756,
+      "grad_norm": 0.0679871067404747,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 75897
+    },
+    {
+      "epoch": 7.800616649537513,
+      "grad_norm": 0.09095168858766556,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 75900
+    },
+    {
+      "epoch": 7.80092497430627,
+      "grad_norm": 0.04139483720064163,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 75903
+    },
+    {
+      "epoch": 7.801233299075026,
+      "grad_norm": 0.034134142100811005,
+      "learning_rate": 0.01,
+      "loss": 1.9854,
+      "step": 75906
+    },
+    {
+      "epoch": 7.801541623843782,
+      "grad_norm": 0.03619580343365669,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 75909
+    },
+    {
+      "epoch": 7.8018499486125386,
+      "grad_norm": 0.0656135305762291,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 75912
+    },
+    {
+      "epoch": 7.802158273381295,
+      "grad_norm": 0.09268335998058319,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 75915
+    },
+    {
+      "epoch": 7.802466598150051,
+      "grad_norm": 0.041520409286022186,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 75918
+    },
+    {
+      "epoch": 7.8027749229188075,
+      "grad_norm": 0.05438603460788727,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 75921
+    },
+    {
+      "epoch": 7.803083247687564,
+      "grad_norm": 0.05482487007975578,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 75924
+    },
+    {
+      "epoch": 7.803391572456321,
+      "grad_norm": 0.08617548644542694,
+      "learning_rate": 0.01,
+      "loss": 1.9959,
+      "step": 75927
+    },
+    {
+      "epoch": 7.803699897225077,
+      "grad_norm": 0.08135716617107391,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 75930
+    },
+    {
+      "epoch": 7.804008221993834,
+      "grad_norm": 0.0932932123541832,
+      "learning_rate": 0.01,
+      "loss": 1.9868,
+      "step": 75933
+    },
+    {
+      "epoch": 7.80431654676259,
+      "grad_norm": 0.06797019392251968,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 75936
+    },
+    {
+      "epoch": 7.804624871531346,
+      "grad_norm": 0.09875380992889404,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 75939
+    },
+    {
+      "epoch": 7.8049331963001025,
+      "grad_norm": 0.07438071072101593,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 75942
+    },
+    {
+      "epoch": 7.805241521068859,
+      "grad_norm": 0.06156305968761444,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 75945
+    },
+    {
+      "epoch": 7.805549845837616,
+      "grad_norm": 0.07288916409015656,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 75948
+    },
+    {
+      "epoch": 7.805858170606372,
+      "grad_norm": 0.06853283941745758,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 75951
+    },
+    {
+      "epoch": 7.806166495375129,
+      "grad_norm": 0.07273688167333603,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 75954
+    },
+    {
+      "epoch": 7.806474820143885,
+      "grad_norm": 0.0601770393550396,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 75957
+    },
+    {
+      "epoch": 7.806783144912641,
+      "grad_norm": 0.07322997599840164,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 75960
+    },
+    {
+      "epoch": 7.8070914696813976,
+      "grad_norm": 0.08790867030620575,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 75963
+    },
+    {
+      "epoch": 7.807399794450154,
+      "grad_norm": 0.07412376254796982,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 75966
+    },
+    {
+      "epoch": 7.80770811921891,
+      "grad_norm": 0.06878551095724106,
+      "learning_rate": 0.01,
+      "loss": 1.9917,
+      "step": 75969
+    },
+    {
+      "epoch": 7.808016443987667,
+      "grad_norm": 0.11822834610939026,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 75972
+    },
+    {
+      "epoch": 7.808324768756424,
+      "grad_norm": 0.039995722472667694,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 75975
+    },
+    {
+      "epoch": 7.80863309352518,
+      "grad_norm": 0.10060998797416687,
+      "learning_rate": 0.01,
+      "loss": 1.9936,
+      "step": 75978
+    },
+    {
+      "epoch": 7.808941418293936,
+      "grad_norm": 0.0810547024011612,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 75981
+    },
+    {
+      "epoch": 7.809249743062693,
+      "grad_norm": 0.10982684791088104,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 75984
+    },
+    {
+      "epoch": 7.809558067831449,
+      "grad_norm": 0.05619985982775688,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 75987
+    },
+    {
+      "epoch": 7.809866392600205,
+      "grad_norm": 0.05136675387620926,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 75990
+    },
+    {
+      "epoch": 7.810174717368962,
+      "grad_norm": 0.03846390172839165,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 75993
+    },
+    {
+      "epoch": 7.810483042137719,
+      "grad_norm": 0.03186337649822235,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 75996
+    },
+    {
+      "epoch": 7.810791366906475,
+      "grad_norm": 0.104001484811306,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 75999
+    },
+    {
+      "epoch": 7.811099691675231,
+      "grad_norm": 0.14364612102508545,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 76002
+    },
+    {
+      "epoch": 7.811408016443988,
+      "grad_norm": 0.09179578721523285,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 76005
+    },
+    {
+      "epoch": 7.811716341212744,
+      "grad_norm": 0.043222784996032715,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 76008
+    },
+    {
+      "epoch": 7.8120246659815,
+      "grad_norm": 0.05643289536237717,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 76011
+    },
+    {
+      "epoch": 7.812332990750257,
+      "grad_norm": 0.09506022185087204,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 76014
+    },
+    {
+      "epoch": 7.812641315519013,
+      "grad_norm": 0.03661196306347847,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 76017
+    },
+    {
+      "epoch": 7.81294964028777,
+      "grad_norm": 0.08643520623445511,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 76020
+    },
+    {
+      "epoch": 7.813257965056526,
+      "grad_norm": 0.06595692783594131,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 76023
+    },
+    {
+      "epoch": 7.813566289825283,
+      "grad_norm": 0.0512213371694088,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 76026
+    },
+    {
+      "epoch": 7.813874614594039,
+      "grad_norm": 0.05945082753896713,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 76029
+    },
+    {
+      "epoch": 7.814182939362795,
+      "grad_norm": 0.1069507747888565,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 76032
+    },
+    {
+      "epoch": 7.814491264131552,
+      "grad_norm": 0.06452130526304245,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 76035
+    },
+    {
+      "epoch": 7.814799588900309,
+      "grad_norm": 0.138530895113945,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 76038
+    },
+    {
+      "epoch": 7.815107913669065,
+      "grad_norm": 0.10860975086688995,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 76041
+    },
+    {
+      "epoch": 7.815416238437821,
+      "grad_norm": 0.10083730518817902,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 76044
+    },
+    {
+      "epoch": 7.815724563206578,
+      "grad_norm": 0.08879086375236511,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 76047
+    },
+    {
+      "epoch": 7.816032887975334,
+      "grad_norm": 0.06405922770500183,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 76050
+    },
+    {
+      "epoch": 7.81634121274409,
+      "grad_norm": 0.06605300307273865,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 76053
+    },
+    {
+      "epoch": 7.816649537512847,
+      "grad_norm": 0.08002083748579025,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 76056
+    },
+    {
+      "epoch": 7.816957862281603,
+      "grad_norm": 0.058482032269239426,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 76059
+    },
+    {
+      "epoch": 7.817266187050359,
+      "grad_norm": 0.10967635363340378,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 76062
+    },
+    {
+      "epoch": 7.8175745118191164,
+      "grad_norm": 0.04840927943587303,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 76065
+    },
+    {
+      "epoch": 7.817882836587873,
+      "grad_norm": 0.08987026661634445,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 76068
+    },
+    {
+      "epoch": 7.818191161356629,
+      "grad_norm": 0.07145990431308746,
+      "learning_rate": 0.01,
+      "loss": 1.9992,
+      "step": 76071
+    },
+    {
+      "epoch": 7.818499486125385,
+      "grad_norm": 0.07257866114377975,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 76074
+    },
+    {
+      "epoch": 7.818807810894142,
+      "grad_norm": 0.043422192335128784,
+      "learning_rate": 0.01,
+      "loss": 1.9808,
+      "step": 76077
+    },
+    {
+      "epoch": 7.819116135662898,
+      "grad_norm": 0.05529724061489105,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 76080
+    },
+    {
+      "epoch": 7.819424460431654,
+      "grad_norm": 0.09055875241756439,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 76083
+    },
+    {
+      "epoch": 7.8197327852004115,
+      "grad_norm": 0.14238907396793365,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 76086
+    },
+    {
+      "epoch": 7.820041109969168,
+      "grad_norm": 0.07417774200439453,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 76089
+    },
+    {
+      "epoch": 7.820349434737924,
+      "grad_norm": 0.05844178423285484,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 76092
+    },
+    {
+      "epoch": 7.82065775950668,
+      "grad_norm": 0.07558295875787735,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 76095
+    },
+    {
+      "epoch": 7.820966084275437,
+      "grad_norm": 0.07094567269086838,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 76098
+    },
+    {
+      "epoch": 7.821274409044193,
+      "grad_norm": 0.052053775638341904,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 76101
+    },
+    {
+      "epoch": 7.821582733812949,
+      "grad_norm": 0.03560258448123932,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 76104
+    },
+    {
+      "epoch": 7.821891058581706,
+      "grad_norm": 0.05725477635860443,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 76107
+    },
+    {
+      "epoch": 7.822199383350463,
+      "grad_norm": 0.06399683654308319,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 76110
+    },
+    {
+      "epoch": 7.822507708119219,
+      "grad_norm": 0.09520284831523895,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 76113
+    },
+    {
+      "epoch": 7.8228160328879754,
+      "grad_norm": 0.07167236506938934,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 76116
+    },
+    {
+      "epoch": 7.823124357656732,
+      "grad_norm": 0.08711408823728561,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 76119
+    },
+    {
+      "epoch": 7.823432682425488,
+      "grad_norm": 0.03872615098953247,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 76122
+    },
+    {
+      "epoch": 7.823741007194244,
+      "grad_norm": 0.0413740873336792,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 76125
+    },
+    {
+      "epoch": 7.824049331963001,
+      "grad_norm": 0.06173890829086304,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 76128
+    },
+    {
+      "epoch": 7.824357656731758,
+      "grad_norm": 0.03490239009261131,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 76131
+    },
+    {
+      "epoch": 7.824665981500514,
+      "grad_norm": 0.06848657131195068,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 76134
+    },
+    {
+      "epoch": 7.8249743062692705,
+      "grad_norm": 0.09866426885128021,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 76137
+    },
+    {
+      "epoch": 7.825282631038027,
+      "grad_norm": 0.08280821889638901,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 76140
+    },
+    {
+      "epoch": 7.825590955806783,
+      "grad_norm": 0.0681089535355568,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 76143
+    },
+    {
+      "epoch": 7.825899280575539,
+      "grad_norm": 0.062307052314281464,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 76146
+    },
+    {
+      "epoch": 7.826207605344296,
+      "grad_norm": 0.0920134112238884,
+      "learning_rate": 0.01,
+      "loss": 1.9972,
+      "step": 76149
+    },
+    {
+      "epoch": 7.826515930113052,
+      "grad_norm": 0.07431496679782867,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 76152
+    },
+    {
+      "epoch": 7.826824254881809,
+      "grad_norm": 0.11818663030862808,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 76155
+    },
+    {
+      "epoch": 7.8271325796505655,
+      "grad_norm": 0.09648428857326508,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 76158
+    },
+    {
+      "epoch": 7.827440904419322,
+      "grad_norm": 0.052823755890131,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 76161
+    },
+    {
+      "epoch": 7.827749229188078,
+      "grad_norm": 0.048545703291893005,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 76164
+    },
+    {
+      "epoch": 7.8280575539568344,
+      "grad_norm": 0.03413026034832001,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 76167
+    },
+    {
+      "epoch": 7.828365878725591,
+      "grad_norm": 0.050574932247400284,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 76170
+    },
+    {
+      "epoch": 7.828674203494347,
+      "grad_norm": 0.04505613073706627,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 76173
+    },
+    {
+      "epoch": 7.828982528263104,
+      "grad_norm": 0.05546938627958298,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 76176
+    },
+    {
+      "epoch": 7.829290853031861,
+      "grad_norm": 0.10264269262552261,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 76179
+    },
+    {
+      "epoch": 7.829599177800617,
+      "grad_norm": 0.044220950454473495,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 76182
+    },
+    {
+      "epoch": 7.829907502569373,
+      "grad_norm": 0.042216695845127106,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 76185
+    },
+    {
+      "epoch": 7.8302158273381295,
+      "grad_norm": 0.04472136124968529,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 76188
+    },
+    {
+      "epoch": 7.830524152106886,
+      "grad_norm": 0.04147589951753616,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 76191
+    },
+    {
+      "epoch": 7.830832476875642,
+      "grad_norm": 0.07456754893064499,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 76194
+    },
+    {
+      "epoch": 7.831140801644398,
+      "grad_norm": 0.09007889777421951,
+      "learning_rate": 0.01,
+      "loss": 1.9913,
+      "step": 76197
+    },
+    {
+      "epoch": 7.831449126413155,
+      "grad_norm": 0.07880180329084396,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 76200
+    },
+    {
+      "epoch": 7.831757451181912,
+      "grad_norm": 0.03387047350406647,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 76203
+    },
+    {
+      "epoch": 7.832065775950668,
+      "grad_norm": 0.03345126286149025,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 76206
+    },
+    {
+      "epoch": 7.8323741007194245,
+      "grad_norm": 0.041953541338443756,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 76209
+    },
+    {
+      "epoch": 7.832682425488181,
+      "grad_norm": 0.04632129892706871,
+      "learning_rate": 0.01,
+      "loss": 1.9862,
+      "step": 76212
+    },
+    {
+      "epoch": 7.832990750256937,
+      "grad_norm": 0.04222474619746208,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 76215
+    },
+    {
+      "epoch": 7.8332990750256934,
+      "grad_norm": 0.07548588514328003,
+      "learning_rate": 0.01,
+      "loss": 1.9887,
+      "step": 76218
+    },
+    {
+      "epoch": 7.833607399794451,
+      "grad_norm": 0.09007594734430313,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 76221
+    },
+    {
+      "epoch": 7.833915724563207,
+      "grad_norm": 0.08240062743425369,
+      "learning_rate": 0.01,
+      "loss": 2.0005,
+      "step": 76224
+    },
+    {
+      "epoch": 7.834224049331963,
+      "grad_norm": 0.12393136322498322,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 76227
+    },
+    {
+      "epoch": 7.83453237410072,
+      "grad_norm": 0.05861087515950203,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 76230
+    },
+    {
+      "epoch": 7.834840698869476,
+      "grad_norm": 0.0426303967833519,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 76233
+    },
+    {
+      "epoch": 7.835149023638232,
+      "grad_norm": 0.06229201331734657,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 76236
+    },
+    {
+      "epoch": 7.8354573484069885,
+      "grad_norm": 0.07277770340442657,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 76239
+    },
+    {
+      "epoch": 7.835765673175745,
+      "grad_norm": 0.06426133960485458,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 76242
+    },
+    {
+      "epoch": 7.836073997944501,
+      "grad_norm": 0.05311770737171173,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 76245
+    },
+    {
+      "epoch": 7.836382322713258,
+      "grad_norm": 0.05998200178146362,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 76248
+    },
+    {
+      "epoch": 7.836690647482015,
+      "grad_norm": 0.07519685477018356,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 76251
+    },
+    {
+      "epoch": 7.836998972250771,
+      "grad_norm": 0.12155578285455704,
+      "learning_rate": 0.01,
+      "loss": 2.0025,
+      "step": 76254
+    },
+    {
+      "epoch": 7.837307297019527,
+      "grad_norm": 0.11642058193683624,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 76257
+    },
+    {
+      "epoch": 7.8376156217882835,
+      "grad_norm": 0.06032479554414749,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 76260
+    },
+    {
+      "epoch": 7.83792394655704,
+      "grad_norm": 0.05517754331231117,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 76263
+    },
+    {
+      "epoch": 7.838232271325796,
+      "grad_norm": 0.029733365401625633,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 76266
+    },
+    {
+      "epoch": 7.838540596094553,
+      "grad_norm": 0.10677066445350647,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 76269
+    },
+    {
+      "epoch": 7.83884892086331,
+      "grad_norm": 0.07923242449760437,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 76272
+    },
+    {
+      "epoch": 7.839157245632066,
+      "grad_norm": 0.05212559178471565,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 76275
+    },
+    {
+      "epoch": 7.839465570400822,
+      "grad_norm": 0.08050580322742462,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 76278
+    },
+    {
+      "epoch": 7.839773895169579,
+      "grad_norm": 0.08338946104049683,
+      "learning_rate": 0.01,
+      "loss": 2.0165,
+      "step": 76281
+    },
+    {
+      "epoch": 7.840082219938335,
+      "grad_norm": 0.0553726889193058,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 76284
+    },
+    {
+      "epoch": 7.840390544707091,
+      "grad_norm": 0.06254331022500992,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 76287
+    },
+    {
+      "epoch": 7.8406988694758475,
+      "grad_norm": 0.058102309703826904,
+      "learning_rate": 0.01,
+      "loss": 1.9898,
+      "step": 76290
+    },
+    {
+      "epoch": 7.841007194244605,
+      "grad_norm": 0.04690523445606232,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 76293
+    },
+    {
+      "epoch": 7.841315519013361,
+      "grad_norm": 0.036311473697423935,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 76296
+    },
+    {
+      "epoch": 7.841623843782117,
+      "grad_norm": 0.03946409747004509,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 76299
+    },
+    {
+      "epoch": 7.841932168550874,
+      "grad_norm": 0.04926838353276253,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 76302
+    },
+    {
+      "epoch": 7.84224049331963,
+      "grad_norm": 0.1399640291929245,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 76305
+    },
+    {
+      "epoch": 7.842548818088386,
+      "grad_norm": 0.1891847401857376,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 76308
+    },
+    {
+      "epoch": 7.8428571428571425,
+      "grad_norm": 0.13039416074752808,
+      "learning_rate": 0.01,
+      "loss": 1.9849,
+      "step": 76311
+    },
+    {
+      "epoch": 7.8431654676259,
+      "grad_norm": 0.0651208832859993,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 76314
+    },
+    {
+      "epoch": 7.843473792394656,
+      "grad_norm": 0.038830872625112534,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 76317
+    },
+    {
+      "epoch": 7.843782117163412,
+      "grad_norm": 0.04478251934051514,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 76320
+    },
+    {
+      "epoch": 7.844090441932169,
+      "grad_norm": 0.03909466788172722,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 76323
+    },
+    {
+      "epoch": 7.844398766700925,
+      "grad_norm": 0.045009203255176544,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 76326
+    },
+    {
+      "epoch": 7.844707091469681,
+      "grad_norm": 0.042739205062389374,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 76329
+    },
+    {
+      "epoch": 7.845015416238438,
+      "grad_norm": 0.04345414787530899,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 76332
+    },
+    {
+      "epoch": 7.845323741007194,
+      "grad_norm": 0.056563850492239,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 76335
+    },
+    {
+      "epoch": 7.84563206577595,
+      "grad_norm": 0.04494136944413185,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 76338
+    },
+    {
+      "epoch": 7.845940390544707,
+      "grad_norm": 0.04633985832333565,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 76341
+    },
+    {
+      "epoch": 7.846248715313464,
+      "grad_norm": 0.13986924290657043,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 76344
+    },
+    {
+      "epoch": 7.84655704008222,
+      "grad_norm": 0.11160435527563095,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 76347
+    },
+    {
+      "epoch": 7.846865364850976,
+      "grad_norm": 0.04715929180383682,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 76350
+    },
+    {
+      "epoch": 7.847173689619733,
+      "grad_norm": 0.03196023032069206,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 76353
+    },
+    {
+      "epoch": 7.847482014388489,
+      "grad_norm": 0.05613362416625023,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 76356
+    },
+    {
+      "epoch": 7.847790339157246,
+      "grad_norm": 0.06527100503444672,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 76359
+    },
+    {
+      "epoch": 7.848098663926002,
+      "grad_norm": 0.030473025515675545,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 76362
+    },
+    {
+      "epoch": 7.848406988694759,
+      "grad_norm": 0.0393981859087944,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 76365
+    },
+    {
+      "epoch": 7.848715313463515,
+      "grad_norm": 0.03515271842479706,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 76368
+    },
+    {
+      "epoch": 7.849023638232271,
+      "grad_norm": 0.03752501681447029,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 76371
+    },
+    {
+      "epoch": 7.849331963001028,
+      "grad_norm": 0.05749157443642616,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 76374
+    },
+    {
+      "epoch": 7.849640287769784,
+      "grad_norm": 0.1413191258907318,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 76377
+    },
+    {
+      "epoch": 7.84994861253854,
+      "grad_norm": 0.12259876728057861,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 76380
+    },
+    {
+      "epoch": 7.850256937307297,
+      "grad_norm": 0.04768449068069458,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 76383
+    },
+    {
+      "epoch": 7.850565262076054,
+      "grad_norm": 0.07419627904891968,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 76386
+    },
+    {
+      "epoch": 7.85087358684481,
+      "grad_norm": 0.05266867205500603,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 76389
+    },
+    {
+      "epoch": 7.851181911613566,
+      "grad_norm": 0.05332281067967415,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 76392
+    },
+    {
+      "epoch": 7.851490236382323,
+      "grad_norm": 0.04726416617631912,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 76395
+    },
+    {
+      "epoch": 7.851798561151079,
+      "grad_norm": 0.0462283231317997,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 76398
+    },
+    {
+      "epoch": 7.852106885919835,
+      "grad_norm": 0.038481879979372025,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 76401
+    },
+    {
+      "epoch": 7.852415210688592,
+      "grad_norm": 0.06793560832738876,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 76404
+    },
+    {
+      "epoch": 7.852723535457349,
+      "grad_norm": 0.10262496024370193,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 76407
+    },
+    {
+      "epoch": 7.853031860226105,
+      "grad_norm": 0.07564928382635117,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 76410
+    },
+    {
+      "epoch": 7.853340184994861,
+      "grad_norm": 0.09007153660058975,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 76413
+    },
+    {
+      "epoch": 7.853648509763618,
+      "grad_norm": 0.05348995327949524,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 76416
+    },
+    {
+      "epoch": 7.853956834532374,
+      "grad_norm": 0.04418191686272621,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 76419
+    },
+    {
+      "epoch": 7.85426515930113,
+      "grad_norm": 0.038563381880521774,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 76422
+    },
+    {
+      "epoch": 7.854573484069887,
+      "grad_norm": 0.12509912252426147,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 76425
+    },
+    {
+      "epoch": 7.854881808838643,
+      "grad_norm": 0.11487328261137009,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 76428
+    },
+    {
+      "epoch": 7.8551901336074,
+      "grad_norm": 0.04599343240261078,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 76431
+    },
+    {
+      "epoch": 7.8554984583761565,
+      "grad_norm": 0.0848371610045433,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 76434
+    },
+    {
+      "epoch": 7.855806783144913,
+      "grad_norm": 0.06545133888721466,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 76437
+    },
+    {
+      "epoch": 7.856115107913669,
+      "grad_norm": 0.09184720367193222,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 76440
+    },
+    {
+      "epoch": 7.856423432682425,
+      "grad_norm": 0.04043002799153328,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 76443
+    },
+    {
+      "epoch": 7.856731757451182,
+      "grad_norm": 0.08890711516141891,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 76446
+    },
+    {
+      "epoch": 7.857040082219938,
+      "grad_norm": 0.13198690116405487,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 76449
+    },
+    {
+      "epoch": 7.857348406988695,
+      "grad_norm": 0.19590161740779877,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 76452
+    },
+    {
+      "epoch": 7.8576567317574515,
+      "grad_norm": 0.1467210203409195,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 76455
+    },
+    {
+      "epoch": 7.857965056526208,
+      "grad_norm": 0.06122981756925583,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 76458
+    },
+    {
+      "epoch": 7.858273381294964,
+      "grad_norm": 0.04069874808192253,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 76461
+    },
+    {
+      "epoch": 7.85858170606372,
+      "grad_norm": 0.05008160322904587,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 76464
+    },
+    {
+      "epoch": 7.858890030832477,
+      "grad_norm": 0.04874841496348381,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 76467
+    },
+    {
+      "epoch": 7.859198355601233,
+      "grad_norm": 0.0698808953166008,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 76470
+    },
+    {
+      "epoch": 7.859506680369989,
+      "grad_norm": 0.0393466018140316,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 76473
+    },
+    {
+      "epoch": 7.859815005138746,
+      "grad_norm": 0.04882080852985382,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 76476
+    },
+    {
+      "epoch": 7.860123329907503,
+      "grad_norm": 0.03266339749097824,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 76479
+    },
+    {
+      "epoch": 7.860431654676259,
+      "grad_norm": 0.028406213968992233,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 76482
+    },
+    {
+      "epoch": 7.8607399794450155,
+      "grad_norm": 0.10809627175331116,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 76485
+    },
+    {
+      "epoch": 7.861048304213772,
+      "grad_norm": 0.045726872980594635,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 76488
+    },
+    {
+      "epoch": 7.861356628982528,
+      "grad_norm": 0.06965349614620209,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 76491
+    },
+    {
+      "epoch": 7.861664953751284,
+      "grad_norm": 0.04185190051794052,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 76494
+    },
+    {
+      "epoch": 7.861973278520042,
+      "grad_norm": 0.09308112412691116,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 76497
+    },
+    {
+      "epoch": 7.862281603288798,
+      "grad_norm": 0.07587264478206635,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 76500
+    },
+    {
+      "epoch": 7.862589928057554,
+      "grad_norm": 0.05060400441288948,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 76503
+    },
+    {
+      "epoch": 7.8628982528263105,
+      "grad_norm": 0.03997211903333664,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 76506
+    },
+    {
+      "epoch": 7.863206577595067,
+      "grad_norm": 0.04034551605582237,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 76509
+    },
+    {
+      "epoch": 7.863514902363823,
+      "grad_norm": 0.03622760251164436,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 76512
+    },
+    {
+      "epoch": 7.863823227132579,
+      "grad_norm": 0.04714718833565712,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 76515
+    },
+    {
+      "epoch": 7.864131551901336,
+      "grad_norm": 0.14129970967769623,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 76518
+    },
+    {
+      "epoch": 7.864439876670092,
+      "grad_norm": 0.04471175745129585,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 76521
+    },
+    {
+      "epoch": 7.864748201438849,
+      "grad_norm": 0.11593741178512573,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 76524
+    },
+    {
+      "epoch": 7.8650565262076055,
+      "grad_norm": 0.03648509830236435,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 76527
+    },
+    {
+      "epoch": 7.865364850976362,
+      "grad_norm": 0.04938739538192749,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 76530
+    },
+    {
+      "epoch": 7.865673175745118,
+      "grad_norm": 0.04273061454296112,
+      "learning_rate": 0.01,
+      "loss": 1.9756,
+      "step": 76533
+    },
+    {
+      "epoch": 7.8659815005138745,
+      "grad_norm": 0.042451903223991394,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 76536
+    },
+    {
+      "epoch": 7.866289825282631,
+      "grad_norm": 0.04022755101323128,
+      "learning_rate": 0.01,
+      "loss": 2.0009,
+      "step": 76539
+    },
+    {
+      "epoch": 7.866598150051388,
+      "grad_norm": 0.09318958967924118,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 76542
+    },
+    {
+      "epoch": 7.866906474820144,
+      "grad_norm": 0.07233930379152298,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 76545
+    },
+    {
+      "epoch": 7.867214799588901,
+      "grad_norm": 0.06357026845216751,
+      "learning_rate": 0.01,
+      "loss": 1.9933,
+      "step": 76548
+    },
+    {
+      "epoch": 7.867523124357657,
+      "grad_norm": 0.04960564523935318,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 76551
+    },
+    {
+      "epoch": 7.867831449126413,
+      "grad_norm": 0.039174117147922516,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 76554
+    },
+    {
+      "epoch": 7.8681397738951695,
+      "grad_norm": 0.049112312495708466,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 76557
+    },
+    {
+      "epoch": 7.868448098663926,
+      "grad_norm": 0.08262576162815094,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 76560
+    },
+    {
+      "epoch": 7.868756423432682,
+      "grad_norm": 0.09073341637849808,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 76563
+    },
+    {
+      "epoch": 7.869064748201438,
+      "grad_norm": 0.10602176189422607,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 76566
+    },
+    {
+      "epoch": 7.869373072970196,
+      "grad_norm": 0.179390549659729,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 76569
+    },
+    {
+      "epoch": 7.869681397738952,
+      "grad_norm": 0.10436874628067017,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 76572
+    },
+    {
+      "epoch": 7.869989722507708,
+      "grad_norm": 0.06214969605207443,
+      "learning_rate": 0.01,
+      "loss": 1.9924,
+      "step": 76575
+    },
+    {
+      "epoch": 7.8702980472764645,
+      "grad_norm": 0.04790744557976723,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 76578
+    },
+    {
+      "epoch": 7.870606372045221,
+      "grad_norm": 0.127052441239357,
+      "learning_rate": 0.01,
+      "loss": 1.9836,
+      "step": 76581
+    },
+    {
+      "epoch": 7.870914696813977,
+      "grad_norm": 0.06101028993725777,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 76584
+    },
+    {
+      "epoch": 7.8712230215827335,
+      "grad_norm": 0.04569672420620918,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 76587
+    },
+    {
+      "epoch": 7.871531346351491,
+      "grad_norm": 0.04835071787238121,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 76590
+    },
+    {
+      "epoch": 7.871839671120247,
+      "grad_norm": 0.048351310193538666,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 76593
+    },
+    {
+      "epoch": 7.872147995889003,
+      "grad_norm": 0.035411905497312546,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 76596
+    },
+    {
+      "epoch": 7.87245632065776,
+      "grad_norm": 0.09007140249013901,
+      "learning_rate": 0.01,
+      "loss": 1.9871,
+      "step": 76599
+    },
+    {
+      "epoch": 7.872764645426516,
+      "grad_norm": 0.04166480526328087,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 76602
+    },
+    {
+      "epoch": 7.873072970195272,
+      "grad_norm": 0.11124366521835327,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 76605
+    },
+    {
+      "epoch": 7.8733812949640285,
+      "grad_norm": 0.07696343213319778,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 76608
+    },
+    {
+      "epoch": 7.873689619732785,
+      "grad_norm": 0.05737284570932388,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 76611
+    },
+    {
+      "epoch": 7.873997944501542,
+      "grad_norm": 0.05126175656914711,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 76614
+    },
+    {
+      "epoch": 7.874306269270298,
+      "grad_norm": 0.04633242264389992,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 76617
+    },
+    {
+      "epoch": 7.874614594039055,
+      "grad_norm": 0.08935544639825821,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 76620
+    },
+    {
+      "epoch": 7.874922918807811,
+      "grad_norm": 0.06309745460748672,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 76623
+    },
+    {
+      "epoch": 7.875231243576567,
+      "grad_norm": 0.10125695914030075,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 76626
+    },
+    {
+      "epoch": 7.8755395683453235,
+      "grad_norm": 0.1011071428656578,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 76629
+    },
+    {
+      "epoch": 7.87584789311408,
+      "grad_norm": 0.07101625204086304,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 76632
+    },
+    {
+      "epoch": 7.876156217882837,
+      "grad_norm": 0.07854854315519333,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 76635
+    },
+    {
+      "epoch": 7.876464542651593,
+      "grad_norm": 0.05637693777680397,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 76638
+    },
+    {
+      "epoch": 7.87677286742035,
+      "grad_norm": 0.07590261101722717,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 76641
+    },
+    {
+      "epoch": 7.877081192189106,
+      "grad_norm": 0.08829659968614578,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 76644
+    },
+    {
+      "epoch": 7.877389516957862,
+      "grad_norm": 0.060618992894887924,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 76647
+    },
+    {
+      "epoch": 7.877697841726619,
+      "grad_norm": 0.048382773995399475,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 76650
+    },
+    {
+      "epoch": 7.878006166495375,
+      "grad_norm": 0.10038956254720688,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 76653
+    },
+    {
+      "epoch": 7.878314491264131,
+      "grad_norm": 0.10625267773866653,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 76656
+    },
+    {
+      "epoch": 7.8786228160328875,
+      "grad_norm": 0.03938886523246765,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 76659
+    },
+    {
+      "epoch": 7.878931140801645,
+      "grad_norm": 0.07031627744436264,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 76662
+    },
+    {
+      "epoch": 7.879239465570401,
+      "grad_norm": 0.04160325601696968,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 76665
+    },
+    {
+      "epoch": 7.879547790339157,
+      "grad_norm": 0.04842138662934303,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 76668
+    },
+    {
+      "epoch": 7.879856115107914,
+      "grad_norm": 0.04438458010554314,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 76671
+    },
+    {
+      "epoch": 7.88016443987667,
+      "grad_norm": 0.13517126441001892,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 76674
+    },
+    {
+      "epoch": 7.880472764645426,
+      "grad_norm": 0.04812932014465332,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 76677
+    },
+    {
+      "epoch": 7.880781089414183,
+      "grad_norm": 0.04208145663142204,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 76680
+    },
+    {
+      "epoch": 7.88108941418294,
+      "grad_norm": 0.04132638871669769,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 76683
+    },
+    {
+      "epoch": 7.881397738951696,
+      "grad_norm": 0.08753607422113419,
+      "learning_rate": 0.01,
+      "loss": 1.9897,
+      "step": 76686
+    },
+    {
+      "epoch": 7.881706063720452,
+      "grad_norm": 0.05934629589319229,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 76689
+    },
+    {
+      "epoch": 7.882014388489209,
+      "grad_norm": 0.1018429771065712,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 76692
+    },
+    {
+      "epoch": 7.882322713257965,
+      "grad_norm": 0.09443504363298416,
+      "learning_rate": 0.01,
+      "loss": 1.9822,
+      "step": 76695
+    },
+    {
+      "epoch": 7.882631038026721,
+      "grad_norm": 0.06363807618618011,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 76698
+    },
+    {
+      "epoch": 7.882939362795478,
+      "grad_norm": 0.03805544599890709,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 76701
+    },
+    {
+      "epoch": 7.883247687564234,
+      "grad_norm": 0.08899611979722977,
+      "learning_rate": 0.01,
+      "loss": 1.9915,
+      "step": 76704
+    },
+    {
+      "epoch": 7.883556012332991,
+      "grad_norm": 0.03682880103588104,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 76707
+    },
+    {
+      "epoch": 7.883864337101747,
+      "grad_norm": 0.10958962887525558,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 76710
+    },
+    {
+      "epoch": 7.884172661870504,
+      "grad_norm": 0.10766680538654327,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 76713
+    },
+    {
+      "epoch": 7.88448098663926,
+      "grad_norm": 0.0537521131336689,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 76716
+    },
+    {
+      "epoch": 7.884789311408016,
+      "grad_norm": 0.05160677060484886,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 76719
+    },
+    {
+      "epoch": 7.885097636176773,
+      "grad_norm": 0.05430131033062935,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 76722
+    },
+    {
+      "epoch": 7.885405960945529,
+      "grad_norm": 0.04517725110054016,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 76725
+    },
+    {
+      "epoch": 7.885714285714286,
+      "grad_norm": 0.03742944449186325,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 76728
+    },
+    {
+      "epoch": 7.886022610483042,
+      "grad_norm": 0.03228388726711273,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 76731
+    },
+    {
+      "epoch": 7.886330935251799,
+      "grad_norm": 0.1236288994550705,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 76734
+    },
+    {
+      "epoch": 7.886639260020555,
+      "grad_norm": 0.06338070333003998,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 76737
+    },
+    {
+      "epoch": 7.886947584789311,
+      "grad_norm": 0.08285798132419586,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 76740
+    },
+    {
+      "epoch": 7.887255909558068,
+      "grad_norm": 0.07970725744962692,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 76743
+    },
+    {
+      "epoch": 7.887564234326824,
+      "grad_norm": 0.12220615893602371,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 76746
+    },
+    {
+      "epoch": 7.88787255909558,
+      "grad_norm": 0.04591503366827965,
+      "learning_rate": 0.01,
+      "loss": 1.9973,
+      "step": 76749
+    },
+    {
+      "epoch": 7.8881808838643375,
+      "grad_norm": 0.03333088383078575,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 76752
+    },
+    {
+      "epoch": 7.888489208633094,
+      "grad_norm": 0.03739148750901222,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 76755
+    },
+    {
+      "epoch": 7.88879753340185,
+      "grad_norm": 0.048116832971572876,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 76758
+    },
+    {
+      "epoch": 7.889105858170606,
+      "grad_norm": 0.051414813846349716,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 76761
+    },
+    {
+      "epoch": 7.889414182939363,
+      "grad_norm": 0.04409851133823395,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 76764
+    },
+    {
+      "epoch": 7.889722507708119,
+      "grad_norm": 0.09132188558578491,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 76767
+    },
+    {
+      "epoch": 7.890030832476875,
+      "grad_norm": 0.10048239678144455,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 76770
+    },
+    {
+      "epoch": 7.8903391572456325,
+      "grad_norm": 0.1443302035331726,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 76773
+    },
+    {
+      "epoch": 7.890647482014389,
+      "grad_norm": 0.10798861086368561,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 76776
+    },
+    {
+      "epoch": 7.890955806783145,
+      "grad_norm": 0.08225159347057343,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 76779
+    },
+    {
+      "epoch": 7.891264131551901,
+      "grad_norm": 0.0331072062253952,
+      "learning_rate": 0.01,
+      "loss": 1.9865,
+      "step": 76782
+    },
+    {
+      "epoch": 7.891572456320658,
+      "grad_norm": 0.03549625724554062,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 76785
+    },
+    {
+      "epoch": 7.891880781089414,
+      "grad_norm": 0.041428498923778534,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 76788
+    },
+    {
+      "epoch": 7.89218910585817,
+      "grad_norm": 0.059432294219732285,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 76791
+    },
+    {
+      "epoch": 7.892497430626927,
+      "grad_norm": 0.11326347291469574,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 76794
+    },
+    {
+      "epoch": 7.892805755395683,
+      "grad_norm": 0.07109850645065308,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 76797
+    },
+    {
+      "epoch": 7.89311408016444,
+      "grad_norm": 0.0451069213449955,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 76800
+    },
+    {
+      "epoch": 7.8934224049331965,
+      "grad_norm": 0.03823963180184364,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 76803
+    },
+    {
+      "epoch": 7.893730729701953,
+      "grad_norm": 0.07655234634876251,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 76806
+    },
+    {
+      "epoch": 7.894039054470709,
+      "grad_norm": 0.06917836517095566,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 76809
+    },
+    {
+      "epoch": 7.894347379239465,
+      "grad_norm": 0.03960380703210831,
+      "learning_rate": 0.01,
+      "loss": 1.9825,
+      "step": 76812
+    },
+    {
+      "epoch": 7.894655704008222,
+      "grad_norm": 0.04303770512342453,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 76815
+    },
+    {
+      "epoch": 7.894964028776979,
+      "grad_norm": 0.0763452872633934,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 76818
+    },
+    {
+      "epoch": 7.895272353545735,
+      "grad_norm": 0.0780813917517662,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 76821
+    },
+    {
+      "epoch": 7.8955806783144915,
+      "grad_norm": 0.14030003547668457,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 76824
+    },
+    {
+      "epoch": 7.895889003083248,
+      "grad_norm": 0.06355911493301392,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 76827
+    },
+    {
+      "epoch": 7.896197327852004,
+      "grad_norm": 0.06618300080299377,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 76830
+    },
+    {
+      "epoch": 7.89650565262076,
+      "grad_norm": 0.07435224950313568,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 76833
+    },
+    {
+      "epoch": 7.896813977389517,
+      "grad_norm": 0.06644349545240402,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 76836
+    },
+    {
+      "epoch": 7.897122302158273,
+      "grad_norm": 0.03589347004890442,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 76839
+    },
+    {
+      "epoch": 7.897430626927029,
+      "grad_norm": 0.10725891590118408,
+      "learning_rate": 0.01,
+      "loss": 1.993,
+      "step": 76842
+    },
+    {
+      "epoch": 7.8977389516957865,
+      "grad_norm": 0.03418595716357231,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 76845
+    },
+    {
+      "epoch": 7.898047276464543,
+      "grad_norm": 0.04659489914774895,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 76848
+    },
+    {
+      "epoch": 7.898355601233299,
+      "grad_norm": 0.11705581843852997,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 76851
+    },
+    {
+      "epoch": 7.8986639260020555,
+      "grad_norm": 0.16697385907173157,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 76854
+    },
+    {
+      "epoch": 7.898972250770812,
+      "grad_norm": 0.18981826305389404,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 76857
+    },
+    {
+      "epoch": 7.899280575539568,
+      "grad_norm": 0.12204299122095108,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 76860
+    },
+    {
+      "epoch": 7.899588900308324,
+      "grad_norm": 0.08213921636343002,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 76863
+    },
+    {
+      "epoch": 7.899897225077082,
+      "grad_norm": 0.05336681753396988,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 76866
+    },
+    {
+      "epoch": 7.900205549845838,
+      "grad_norm": 0.034992218017578125,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 76869
+    },
+    {
+      "epoch": 7.900513874614594,
+      "grad_norm": 0.052847620099782944,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 76872
+    },
+    {
+      "epoch": 7.9008221993833505,
+      "grad_norm": 0.050509583204984665,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 76875
+    },
+    {
+      "epoch": 7.901130524152107,
+      "grad_norm": 0.03660079464316368,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 76878
+    },
+    {
+      "epoch": 7.901438848920863,
+      "grad_norm": 0.04175078123807907,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 76881
+    },
+    {
+      "epoch": 7.901747173689619,
+      "grad_norm": 0.07008788734674454,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 76884
+    },
+    {
+      "epoch": 7.902055498458376,
+      "grad_norm": 0.1135629415512085,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 76887
+    },
+    {
+      "epoch": 7.902363823227133,
+      "grad_norm": 0.14856696128845215,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 76890
+    },
+    {
+      "epoch": 7.902672147995889,
+      "grad_norm": 0.07151875644922256,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 76893
+    },
+    {
+      "epoch": 7.9029804727646455,
+      "grad_norm": 0.05615755915641785,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 76896
+    },
+    {
+      "epoch": 7.903288797533402,
+      "grad_norm": 0.045580677688121796,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 76899
+    },
+    {
+      "epoch": 7.903597122302158,
+      "grad_norm": 0.09109973162412643,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 76902
+    },
+    {
+      "epoch": 7.9039054470709145,
+      "grad_norm": 0.08682844787836075,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 76905
+    },
+    {
+      "epoch": 7.904213771839671,
+      "grad_norm": 0.03720599412918091,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 76908
+    },
+    {
+      "epoch": 7.904522096608428,
+      "grad_norm": 0.03967565670609474,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 76911
+    },
+    {
+      "epoch": 7.904830421377184,
+      "grad_norm": 0.04664463922381401,
+      "learning_rate": 0.01,
+      "loss": 2.0114,
+      "step": 76914
+    },
+    {
+      "epoch": 7.905138746145941,
+      "grad_norm": 0.04584002122282982,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 76917
+    },
+    {
+      "epoch": 7.905447070914697,
+      "grad_norm": 0.03752794861793518,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 76920
+    },
+    {
+      "epoch": 7.905755395683453,
+      "grad_norm": 0.04139968752861023,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 76923
+    },
+    {
+      "epoch": 7.9060637204522095,
+      "grad_norm": 0.03457412123680115,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 76926
+    },
+    {
+      "epoch": 7.906372045220966,
+      "grad_norm": 0.08590247482061386,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 76929
+    },
+    {
+      "epoch": 7.906680369989722,
+      "grad_norm": 0.04614677280187607,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 76932
+    },
+    {
+      "epoch": 7.906988694758479,
+      "grad_norm": 0.054161593317985535,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 76935
+    },
+    {
+      "epoch": 7.907297019527236,
+      "grad_norm": 0.07664169371128082,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 76938
+    },
+    {
+      "epoch": 7.907605344295992,
+      "grad_norm": 0.059912554919719696,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 76941
+    },
+    {
+      "epoch": 7.907913669064748,
+      "grad_norm": 0.08653954416513443,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 76944
+    },
+    {
+      "epoch": 7.9082219938335045,
+      "grad_norm": 0.09183872491121292,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 76947
+    },
+    {
+      "epoch": 7.908530318602261,
+      "grad_norm": 0.05259134992957115,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 76950
+    },
+    {
+      "epoch": 7.908838643371017,
+      "grad_norm": 0.08326982706785202,
+      "learning_rate": 0.01,
+      "loss": 2.0106,
+      "step": 76953
+    },
+    {
+      "epoch": 7.909146968139774,
+      "grad_norm": 0.09940369427204132,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 76956
+    },
+    {
+      "epoch": 7.909455292908531,
+      "grad_norm": 0.04686363413929939,
+      "learning_rate": 0.01,
+      "loss": 1.9952,
+      "step": 76959
+    },
+    {
+      "epoch": 7.909763617677287,
+      "grad_norm": 0.09306104481220245,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 76962
+    },
+    {
+      "epoch": 7.910071942446043,
+      "grad_norm": 0.07574539631605148,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 76965
+    },
+    {
+      "epoch": 7.9103802672148,
+      "grad_norm": 0.11940892785787582,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 76968
+    },
+    {
+      "epoch": 7.910688591983556,
+      "grad_norm": 0.07245193421840668,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 76971
+    },
+    {
+      "epoch": 7.910996916752312,
+      "grad_norm": 0.050414931029081345,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 76974
+    },
+    {
+      "epoch": 7.9113052415210685,
+      "grad_norm": 0.043788183480501175,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 76977
+    },
+    {
+      "epoch": 7.911613566289825,
+      "grad_norm": 0.07120411843061447,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 76980
+    },
+    {
+      "epoch": 7.911921891058582,
+      "grad_norm": 0.04601344093680382,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 76983
+    },
+    {
+      "epoch": 7.912230215827338,
+      "grad_norm": 0.053321413695812225,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 76986
+    },
+    {
+      "epoch": 7.912538540596095,
+      "grad_norm": 0.09367197006940842,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 76989
+    },
+    {
+      "epoch": 7.912846865364851,
+      "grad_norm": 0.10960140079259872,
+      "learning_rate": 0.01,
+      "loss": 2.0057,
+      "step": 76992
+    },
+    {
+      "epoch": 7.913155190133607,
+      "grad_norm": 0.08764875680208206,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 76995
+    },
+    {
+      "epoch": 7.9134635149023635,
+      "grad_norm": 0.07808122038841248,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 76998
+    },
+    {
+      "epoch": 7.913771839671121,
+      "grad_norm": 0.059402767568826675,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 77001
+    },
+    {
+      "epoch": 7.914080164439877,
+      "grad_norm": 0.09941643476486206,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 77004
+    },
+    {
+      "epoch": 7.914388489208633,
+      "grad_norm": 0.142394557595253,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 77007
+    },
+    {
+      "epoch": 7.91469681397739,
+      "grad_norm": 0.05479384586215019,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 77010
+    },
+    {
+      "epoch": 7.915005138746146,
+      "grad_norm": 0.04836162552237511,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 77013
+    },
+    {
+      "epoch": 7.915313463514902,
+      "grad_norm": 0.07146017253398895,
+      "learning_rate": 0.01,
+      "loss": 1.991,
+      "step": 77016
+    },
+    {
+      "epoch": 7.915621788283659,
+      "grad_norm": 0.06672468036413193,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 77019
+    },
+    {
+      "epoch": 7.915930113052415,
+      "grad_norm": 0.050683509558439255,
+      "learning_rate": 0.01,
+      "loss": 1.9801,
+      "step": 77022
+    },
+    {
+      "epoch": 7.916238437821171,
+      "grad_norm": 0.04118918254971504,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 77025
+    },
+    {
+      "epoch": 7.916546762589928,
+      "grad_norm": 0.11386791616678238,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 77028
+    },
+    {
+      "epoch": 7.916855087358685,
+      "grad_norm": 0.04421692341566086,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 77031
+    },
+    {
+      "epoch": 7.917163412127441,
+      "grad_norm": 0.03897789120674133,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 77034
+    },
+    {
+      "epoch": 7.917471736896197,
+      "grad_norm": 0.07048334926366806,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 77037
+    },
+    {
+      "epoch": 7.917780061664954,
+      "grad_norm": 0.07954574376344681,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 77040
+    },
+    {
+      "epoch": 7.91808838643371,
+      "grad_norm": 0.07385550439357758,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 77043
+    },
+    {
+      "epoch": 7.918396711202466,
+      "grad_norm": 0.099297434091568,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 77046
+    },
+    {
+      "epoch": 7.918705035971223,
+      "grad_norm": 0.10075324028730392,
+      "learning_rate": 0.01,
+      "loss": 1.9802,
+      "step": 77049
+    },
+    {
+      "epoch": 7.91901336073998,
+      "grad_norm": 0.04904831945896149,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 77052
+    },
+    {
+      "epoch": 7.919321685508736,
+      "grad_norm": 0.042833998799324036,
+      "learning_rate": 0.01,
+      "loss": 1.9832,
+      "step": 77055
+    },
+    {
+      "epoch": 7.919630010277492,
+      "grad_norm": 0.03839438781142235,
+      "learning_rate": 0.01,
+      "loss": 1.9994,
+      "step": 77058
+    },
+    {
+      "epoch": 7.919938335046249,
+      "grad_norm": 0.05418895557522774,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 77061
+    },
+    {
+      "epoch": 7.920246659815005,
+      "grad_norm": 0.03642089292407036,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 77064
+    },
+    {
+      "epoch": 7.920554984583761,
+      "grad_norm": 0.10740983486175537,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 77067
+    },
+    {
+      "epoch": 7.920863309352518,
+      "grad_norm": 0.0470796562731266,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 77070
+    },
+    {
+      "epoch": 7.921171634121275,
+      "grad_norm": 0.0717134177684784,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 77073
+    },
+    {
+      "epoch": 7.921479958890031,
+      "grad_norm": 0.09272985905408859,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 77076
+    },
+    {
+      "epoch": 7.921788283658787,
+      "grad_norm": 0.06581149250268936,
+      "learning_rate": 0.01,
+      "loss": 2.005,
+      "step": 77079
+    },
+    {
+      "epoch": 7.922096608427544,
+      "grad_norm": 0.10736197978258133,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 77082
+    },
+    {
+      "epoch": 7.9224049331963,
+      "grad_norm": 0.0577731616795063,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 77085
+    },
+    {
+      "epoch": 7.922713257965056,
+      "grad_norm": 0.05568409711122513,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 77088
+    },
+    {
+      "epoch": 7.923021582733813,
+      "grad_norm": 0.047936148941516876,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 77091
+    },
+    {
+      "epoch": 7.92332990750257,
+      "grad_norm": 0.04763433337211609,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 77094
+    },
+    {
+      "epoch": 7.923638232271326,
+      "grad_norm": 0.08284518867731094,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 77097
+    },
+    {
+      "epoch": 7.923946557040082,
+      "grad_norm": 0.06552954018115997,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 77100
+    },
+    {
+      "epoch": 7.924254881808839,
+      "grad_norm": 0.07731842994689941,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 77103
+    },
+    {
+      "epoch": 7.924563206577595,
+      "grad_norm": 0.07882971316576004,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 77106
+    },
+    {
+      "epoch": 7.924871531346351,
+      "grad_norm": 0.06781595945358276,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 77109
+    },
+    {
+      "epoch": 7.925179856115108,
+      "grad_norm": 0.07871492207050323,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 77112
+    },
+    {
+      "epoch": 7.925488180883864,
+      "grad_norm": 0.09492309391498566,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 77115
+    },
+    {
+      "epoch": 7.92579650565262,
+      "grad_norm": 0.07608136534690857,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 77118
+    },
+    {
+      "epoch": 7.9261048304213775,
+      "grad_norm": 0.14081300795078278,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 77121
+    },
+    {
+      "epoch": 7.926413155190134,
+      "grad_norm": 0.10385705530643463,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 77124
+    },
+    {
+      "epoch": 7.92672147995889,
+      "grad_norm": 0.059850629419088364,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 77127
+    },
+    {
+      "epoch": 7.927029804727646,
+      "grad_norm": 0.03840718790888786,
+      "learning_rate": 0.01,
+      "loss": 1.9974,
+      "step": 77130
+    },
+    {
+      "epoch": 7.927338129496403,
+      "grad_norm": 0.05099863559007645,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 77133
+    },
+    {
+      "epoch": 7.927646454265159,
+      "grad_norm": 0.10900937765836716,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 77136
+    },
+    {
+      "epoch": 7.927954779033916,
+      "grad_norm": 0.05620981380343437,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 77139
+    },
+    {
+      "epoch": 7.9282631038026725,
+      "grad_norm": 0.07825091481208801,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 77142
+    },
+    {
+      "epoch": 7.928571428571429,
+      "grad_norm": 0.047291044145822525,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 77145
+    },
+    {
+      "epoch": 7.928879753340185,
+      "grad_norm": 0.08942444622516632,
+      "learning_rate": 0.01,
+      "loss": 1.9908,
+      "step": 77148
+    },
+    {
+      "epoch": 7.929188078108941,
+      "grad_norm": 0.0645388588309288,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 77151
+    },
+    {
+      "epoch": 7.929496402877698,
+      "grad_norm": 0.04266830161213875,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 77154
+    },
+    {
+      "epoch": 7.929804727646454,
+      "grad_norm": 0.04416031017899513,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 77157
+    },
+    {
+      "epoch": 7.93011305241521,
+      "grad_norm": 0.05279887467622757,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 77160
+    },
+    {
+      "epoch": 7.930421377183967,
+      "grad_norm": 0.0377926304936409,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 77163
+    },
+    {
+      "epoch": 7.930729701952724,
+      "grad_norm": 0.03909745067358017,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 77166
+    },
+    {
+      "epoch": 7.93103802672148,
+      "grad_norm": 0.05067327246069908,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 77169
+    },
+    {
+      "epoch": 7.9313463514902365,
+      "grad_norm": 0.11055053770542145,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 77172
+    },
+    {
+      "epoch": 7.931654676258993,
+      "grad_norm": 0.13083787262439728,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 77175
+    },
+    {
+      "epoch": 7.931963001027749,
+      "grad_norm": 0.05712626501917839,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 77178
+    },
+    {
+      "epoch": 7.932271325796505,
+      "grad_norm": 0.034157317131757736,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 77181
+    },
+    {
+      "epoch": 7.932579650565262,
+      "grad_norm": 0.040653664618730545,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 77184
+    },
+    {
+      "epoch": 7.932887975334019,
+      "grad_norm": 0.05120181292295456,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 77187
+    },
+    {
+      "epoch": 7.933196300102775,
+      "grad_norm": 0.04877161234617233,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 77190
+    },
+    {
+      "epoch": 7.9335046248715315,
+      "grad_norm": 0.038983944803476334,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 77193
+    },
+    {
+      "epoch": 7.933812949640288,
+      "grad_norm": 0.10702808201313019,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 77196
+    },
+    {
+      "epoch": 7.934121274409044,
+      "grad_norm": 0.13693293929100037,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 77199
+    },
+    {
+      "epoch": 7.9344295991778,
+      "grad_norm": 0.06795570999383926,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 77202
+    },
+    {
+      "epoch": 7.934737923946557,
+      "grad_norm": 0.05547187477350235,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 77205
+    },
+    {
+      "epoch": 7.935046248715313,
+      "grad_norm": 0.042718227952718735,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 77208
+    },
+    {
+      "epoch": 7.93535457348407,
+      "grad_norm": 0.03317321464419365,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 77211
+    },
+    {
+      "epoch": 7.935662898252827,
+      "grad_norm": 0.03335217386484146,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 77214
+    },
+    {
+      "epoch": 7.935971223021583,
+      "grad_norm": 0.04254760593175888,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 77217
+    },
+    {
+      "epoch": 7.936279547790339,
+      "grad_norm": 0.0777120515704155,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 77220
+    },
+    {
+      "epoch": 7.9365878725590955,
+      "grad_norm": 0.10709066689014435,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 77223
+    },
+    {
+      "epoch": 7.936896197327852,
+      "grad_norm": 0.07401084899902344,
+      "learning_rate": 0.01,
+      "loss": 1.9964,
+      "step": 77226
+    },
+    {
+      "epoch": 7.937204522096608,
+      "grad_norm": 0.03645917400717735,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 77229
+    },
+    {
+      "epoch": 7.937512846865365,
+      "grad_norm": 0.09955336153507233,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 77232
+    },
+    {
+      "epoch": 7.937821171634122,
+      "grad_norm": 0.0894830971956253,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 77235
+    },
+    {
+      "epoch": 7.938129496402878,
+      "grad_norm": 0.06080229952931404,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 77238
+    },
+    {
+      "epoch": 7.938437821171634,
+      "grad_norm": 0.04642382264137268,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 77241
+    },
+    {
+      "epoch": 7.9387461459403905,
+      "grad_norm": 0.036709267646074295,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 77244
+    },
+    {
+      "epoch": 7.939054470709147,
+      "grad_norm": 0.041632648557424545,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 77247
+    },
+    {
+      "epoch": 7.939362795477903,
+      "grad_norm": 0.03514658287167549,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 77250
+    },
+    {
+      "epoch": 7.939671120246659,
+      "grad_norm": 0.028975483030080795,
+      "learning_rate": 0.01,
+      "loss": 1.987,
+      "step": 77253
+    },
+    {
+      "epoch": 7.939979445015416,
+      "grad_norm": 0.04937351867556572,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 77256
+    },
+    {
+      "epoch": 7.940287769784173,
+      "grad_norm": 0.07647955417633057,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 77259
+    },
+    {
+      "epoch": 7.940596094552929,
+      "grad_norm": 0.054201941937208176,
+      "learning_rate": 0.01,
+      "loss": 1.9985,
+      "step": 77262
+    },
+    {
+      "epoch": 7.940904419321686,
+      "grad_norm": 0.07279737293720245,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 77265
+    },
+    {
+      "epoch": 7.941212744090442,
+      "grad_norm": 0.08488020300865173,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 77268
+    },
+    {
+      "epoch": 7.941521068859198,
+      "grad_norm": 0.11421249061822891,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 77271
+    },
+    {
+      "epoch": 7.9418293936279545,
+      "grad_norm": 0.042113203555345535,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 77274
+    },
+    {
+      "epoch": 7.942137718396712,
+      "grad_norm": 0.06829381734132767,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 77277
+    },
+    {
+      "epoch": 7.942446043165468,
+      "grad_norm": 0.09846700727939606,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 77280
+    },
+    {
+      "epoch": 7.942754367934224,
+      "grad_norm": 0.03628145530819893,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 77283
+    },
+    {
+      "epoch": 7.943062692702981,
+      "grad_norm": 0.15278568863868713,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 77286
+    },
+    {
+      "epoch": 7.943371017471737,
+      "grad_norm": 0.0523282065987587,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 77289
+    },
+    {
+      "epoch": 7.943679342240493,
+      "grad_norm": 0.033178601413965225,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 77292
+    },
+    {
+      "epoch": 7.9439876670092495,
+      "grad_norm": 0.045855265110731125,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 77295
+    },
+    {
+      "epoch": 7.944295991778006,
+      "grad_norm": 0.043520331382751465,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 77298
+    },
+    {
+      "epoch": 7.944604316546762,
+      "grad_norm": 0.04189394786953926,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 77301
+    },
+    {
+      "epoch": 7.944912641315519,
+      "grad_norm": 0.04767363518476486,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 77304
+    },
+    {
+      "epoch": 7.945220966084276,
+      "grad_norm": 0.08458670228719711,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 77307
+    },
+    {
+      "epoch": 7.945529290853032,
+      "grad_norm": 0.0853220671415329,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 77310
+    },
+    {
+      "epoch": 7.945837615621788,
+      "grad_norm": 0.06229695677757263,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 77313
+    },
+    {
+      "epoch": 7.946145940390545,
+      "grad_norm": 0.09782623499631882,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 77316
+    },
+    {
+      "epoch": 7.946454265159301,
+      "grad_norm": 0.09582273662090302,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 77319
+    },
+    {
+      "epoch": 7.946762589928057,
+      "grad_norm": 0.057305045425891876,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 77322
+    },
+    {
+      "epoch": 7.947070914696814,
+      "grad_norm": 0.12745541334152222,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 77325
+    },
+    {
+      "epoch": 7.947379239465571,
+      "grad_norm": 0.044055260717868805,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 77328
+    },
+    {
+      "epoch": 7.947687564234327,
+      "grad_norm": 0.09511730819940567,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 77331
+    },
+    {
+      "epoch": 7.947995889003083,
+      "grad_norm": 0.05790237337350845,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 77334
+    },
+    {
+      "epoch": 7.94830421377184,
+      "grad_norm": 0.11035138368606567,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 77337
+    },
+    {
+      "epoch": 7.948612538540596,
+      "grad_norm": 0.04935318976640701,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 77340
+    },
+    {
+      "epoch": 7.948920863309352,
+      "grad_norm": 0.10183247923851013,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 77343
+    },
+    {
+      "epoch": 7.9492291880781085,
+      "grad_norm": 0.06453783065080643,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 77346
+    },
+    {
+      "epoch": 7.949537512846866,
+      "grad_norm": 0.05836249515414238,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 77349
+    },
+    {
+      "epoch": 7.949845837615622,
+      "grad_norm": 0.05511557310819626,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 77352
+    },
+    {
+      "epoch": 7.950154162384378,
+      "grad_norm": 0.09671493619680405,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 77355
+    },
+    {
+      "epoch": 7.950462487153135,
+      "grad_norm": 0.05204296484589577,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 77358
+    },
+    {
+      "epoch": 7.950770811921891,
+      "grad_norm": 0.09407757967710495,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 77361
+    },
+    {
+      "epoch": 7.951079136690647,
+      "grad_norm": 0.08344008028507233,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 77364
+    },
+    {
+      "epoch": 7.951387461459404,
+      "grad_norm": 0.059847306460142136,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 77367
+    },
+    {
+      "epoch": 7.951695786228161,
+      "grad_norm": 0.05866989120841026,
+      "learning_rate": 0.01,
+      "loss": 1.9939,
+      "step": 77370
+    },
+    {
+      "epoch": 7.952004110996917,
+      "grad_norm": 0.053589146584272385,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 77373
+    },
+    {
+      "epoch": 7.952312435765673,
+      "grad_norm": 0.10758215188980103,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 77376
+    },
+    {
+      "epoch": 7.95262076053443,
+      "grad_norm": 0.033645015209913254,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 77379
+    },
+    {
+      "epoch": 7.952929085303186,
+      "grad_norm": 0.050443585962057114,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 77382
+    },
+    {
+      "epoch": 7.953237410071942,
+      "grad_norm": 0.07028111815452576,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 77385
+    },
+    {
+      "epoch": 7.953545734840699,
+      "grad_norm": 0.05370425805449486,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 77388
+    },
+    {
+      "epoch": 7.953854059609455,
+      "grad_norm": 0.08151637017726898,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 77391
+    },
+    {
+      "epoch": 7.954162384378212,
+      "grad_norm": 0.046112194657325745,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 77394
+    },
+    {
+      "epoch": 7.954470709146968,
+      "grad_norm": 0.09836447238922119,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 77397
+    },
+    {
+      "epoch": 7.954779033915725,
+      "grad_norm": 0.056767839938402176,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 77400
+    },
+    {
+      "epoch": 7.955087358684481,
+      "grad_norm": 0.08596283942461014,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 77403
+    },
+    {
+      "epoch": 7.955395683453237,
+      "grad_norm": 0.03643207252025604,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 77406
+    },
+    {
+      "epoch": 7.955704008221994,
+      "grad_norm": 0.13480544090270996,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 77409
+    },
+    {
+      "epoch": 7.95601233299075,
+      "grad_norm": 0.058044008910655975,
+      "learning_rate": 0.01,
+      "loss": 1.9799,
+      "step": 77412
+    },
+    {
+      "epoch": 7.956320657759507,
+      "grad_norm": 0.06414663046598434,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 77415
+    },
+    {
+      "epoch": 7.9566289825282634,
+      "grad_norm": 0.09021684527397156,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 77418
+    },
+    {
+      "epoch": 7.95693730729702,
+      "grad_norm": 0.08076009154319763,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 77421
+    },
+    {
+      "epoch": 7.957245632065776,
+      "grad_norm": 0.051470015197992325,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 77424
+    },
+    {
+      "epoch": 7.957553956834532,
+      "grad_norm": 0.10929223895072937,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 77427
+    },
+    {
+      "epoch": 7.957862281603289,
+      "grad_norm": 0.08006475865840912,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 77430
+    },
+    {
+      "epoch": 7.958170606372045,
+      "grad_norm": 0.07928687334060669,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 77433
+    },
+    {
+      "epoch": 7.958478931140801,
+      "grad_norm": 0.07244311273097992,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 77436
+    },
+    {
+      "epoch": 7.958787255909558,
+      "grad_norm": 0.0400693379342556,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 77439
+    },
+    {
+      "epoch": 7.959095580678315,
+      "grad_norm": 0.03099592588841915,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 77442
+    },
+    {
+      "epoch": 7.959403905447071,
+      "grad_norm": 0.05279327183961868,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 77445
+    },
+    {
+      "epoch": 7.959712230215827,
+      "grad_norm": 0.11631178855895996,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 77448
+    },
+    {
+      "epoch": 7.960020554984584,
+      "grad_norm": 0.06946063786745071,
+      "learning_rate": 0.01,
+      "loss": 1.9858,
+      "step": 77451
+    },
+    {
+      "epoch": 7.96032887975334,
+      "grad_norm": 0.0573054701089859,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 77454
+    },
+    {
+      "epoch": 7.960637204522096,
+      "grad_norm": 0.08694712072610855,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 77457
+    },
+    {
+      "epoch": 7.9609455292908535,
+      "grad_norm": 0.046338554471731186,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 77460
+    },
+    {
+      "epoch": 7.96125385405961,
+      "grad_norm": 0.04105210676789284,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 77463
+    },
+    {
+      "epoch": 7.961562178828366,
+      "grad_norm": 0.03583085164427757,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 77466
+    },
+    {
+      "epoch": 7.9618705035971225,
+      "grad_norm": 0.08635620027780533,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 77469
+    },
+    {
+      "epoch": 7.962178828365879,
+      "grad_norm": 0.0696592703461647,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 77472
+    },
+    {
+      "epoch": 7.962487153134635,
+      "grad_norm": 0.0656665712594986,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 77475
+    },
+    {
+      "epoch": 7.962795477903391,
+      "grad_norm": 0.05063663423061371,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 77478
+    },
+    {
+      "epoch": 7.963103802672148,
+      "grad_norm": 0.08557248115539551,
+      "learning_rate": 0.01,
+      "loss": 1.9875,
+      "step": 77481
+    },
+    {
+      "epoch": 7.963412127440904,
+      "grad_norm": 0.047050029039382935,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 77484
+    },
+    {
+      "epoch": 7.963720452209661,
+      "grad_norm": 0.03459413722157478,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 77487
+    },
+    {
+      "epoch": 7.9640287769784175,
+      "grad_norm": 0.08979766815900803,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 77490
+    },
+    {
+      "epoch": 7.964337101747174,
+      "grad_norm": 0.05879922956228256,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 77493
+    },
+    {
+      "epoch": 7.96464542651593,
+      "grad_norm": 0.04225922003388405,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 77496
+    },
+    {
+      "epoch": 7.964953751284686,
+      "grad_norm": 0.043802402913570404,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 77499
+    },
+    {
+      "epoch": 7.965262076053443,
+      "grad_norm": 0.05574416741728783,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 77502
+    },
+    {
+      "epoch": 7.965570400822199,
+      "grad_norm": 0.22006060183048248,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 77505
+    },
+    {
+      "epoch": 7.965878725590956,
+      "grad_norm": 0.18854233622550964,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 77508
+    },
+    {
+      "epoch": 7.9661870503597125,
+      "grad_norm": 0.11298780888319016,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 77511
+    },
+    {
+      "epoch": 7.966495375128469,
+      "grad_norm": 0.0345136821269989,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 77514
+    },
+    {
+      "epoch": 7.966803699897225,
+      "grad_norm": 0.06418821960687637,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 77517
+    },
+    {
+      "epoch": 7.9671120246659815,
+      "grad_norm": 0.04282977804541588,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 77520
+    },
+    {
+      "epoch": 7.967420349434738,
+      "grad_norm": 0.07152153551578522,
+      "learning_rate": 0.01,
+      "loss": 1.9839,
+      "step": 77523
+    },
+    {
+      "epoch": 7.967728674203494,
+      "grad_norm": 0.07487302273511887,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 77526
+    },
+    {
+      "epoch": 7.96803699897225,
+      "grad_norm": 0.038246192038059235,
+      "learning_rate": 0.01,
+      "loss": 1.9902,
+      "step": 77529
+    },
+    {
+      "epoch": 7.968345323741008,
+      "grad_norm": 0.10243113338947296,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 77532
+    },
+    {
+      "epoch": 7.968653648509764,
+      "grad_norm": 0.05641162022948265,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 77535
+    },
+    {
+      "epoch": 7.96896197327852,
+      "grad_norm": 0.060813263058662415,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 77538
+    },
+    {
+      "epoch": 7.9692702980472765,
+      "grad_norm": 0.07178650796413422,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 77541
+    },
+    {
+      "epoch": 7.969578622816033,
+      "grad_norm": 0.11889466643333435,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 77544
+    },
+    {
+      "epoch": 7.969886947584789,
+      "grad_norm": 0.048543453216552734,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 77547
+    },
+    {
+      "epoch": 7.970195272353545,
+      "grad_norm": 0.09376556426286697,
+      "learning_rate": 0.01,
+      "loss": 1.985,
+      "step": 77550
+    },
+    {
+      "epoch": 7.970503597122303,
+      "grad_norm": 0.12909746170043945,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 77553
+    },
+    {
+      "epoch": 7.970811921891059,
+      "grad_norm": 0.1283407062292099,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 77556
+    },
+    {
+      "epoch": 7.971120246659815,
+      "grad_norm": 0.05502600222826004,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 77559
+    },
+    {
+      "epoch": 7.9714285714285715,
+      "grad_norm": 0.036075152456760406,
+      "learning_rate": 0.01,
+      "loss": 1.9823,
+      "step": 77562
+    },
+    {
+      "epoch": 7.971736896197328,
+      "grad_norm": 0.03715295344591141,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 77565
+    },
+    {
+      "epoch": 7.972045220966084,
+      "grad_norm": 0.10578956454992294,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 77568
+    },
+    {
+      "epoch": 7.9723535457348405,
+      "grad_norm": 0.07572545111179352,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 77571
+    },
+    {
+      "epoch": 7.972661870503597,
+      "grad_norm": 0.04459410533308983,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 77574
+    },
+    {
+      "epoch": 7.972970195272353,
+      "grad_norm": 0.047775302082300186,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 77577
+    },
+    {
+      "epoch": 7.97327852004111,
+      "grad_norm": 0.03661422058939934,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 77580
+    },
+    {
+      "epoch": 7.973586844809867,
+      "grad_norm": 0.06153013929724693,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 77583
+    },
+    {
+      "epoch": 7.973895169578623,
+      "grad_norm": 0.06503260880708694,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 77586
+    },
+    {
+      "epoch": 7.974203494347379,
+      "grad_norm": 0.07872997224330902,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 77589
+    },
+    {
+      "epoch": 7.9745118191161355,
+      "grad_norm": 0.09111315757036209,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 77592
+    },
+    {
+      "epoch": 7.974820143884892,
+      "grad_norm": 0.04510453715920448,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 77595
+    },
+    {
+      "epoch": 7.975128468653649,
+      "grad_norm": 0.1190980076789856,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 77598
+    },
+    {
+      "epoch": 7.975436793422405,
+      "grad_norm": 0.08375338464975357,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 77601
+    },
+    {
+      "epoch": 7.975745118191162,
+      "grad_norm": 0.041891466826200485,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 77604
+    },
+    {
+      "epoch": 7.976053442959918,
+      "grad_norm": 0.04583370313048363,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 77607
+    },
+    {
+      "epoch": 7.976361767728674,
+      "grad_norm": 0.09042174369096756,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 77610
+    },
+    {
+      "epoch": 7.9766700924974305,
+      "grad_norm": 0.05806156247854233,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 77613
+    },
+    {
+      "epoch": 7.976978417266187,
+      "grad_norm": 0.04089076817035675,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 77616
+    },
+    {
+      "epoch": 7.977286742034943,
+      "grad_norm": 0.049693990498781204,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 77619
+    },
+    {
+      "epoch": 7.9775950668036995,
+      "grad_norm": 0.07871042937040329,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 77622
+    },
+    {
+      "epoch": 7.977903391572457,
+      "grad_norm": 0.11597729474306107,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 77625
+    },
+    {
+      "epoch": 7.978211716341213,
+      "grad_norm": 0.040133338421583176,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 77628
+    },
+    {
+      "epoch": 7.978520041109969,
+      "grad_norm": 0.08938458561897278,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 77631
+    },
+    {
+      "epoch": 7.978828365878726,
+      "grad_norm": 0.0434061661362648,
+      "learning_rate": 0.01,
+      "loss": 1.9925,
+      "step": 77634
+    },
+    {
+      "epoch": 7.979136690647482,
+      "grad_norm": 0.137980654835701,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 77637
+    },
+    {
+      "epoch": 7.979445015416238,
+      "grad_norm": 0.07370985299348831,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 77640
+    },
+    {
+      "epoch": 7.9797533401849945,
+      "grad_norm": 0.11431551724672318,
+      "learning_rate": 0.01,
+      "loss": 1.9951,
+      "step": 77643
+    },
+    {
+      "epoch": 7.980061664953752,
+      "grad_norm": 0.1513805091381073,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 77646
+    },
+    {
+      "epoch": 7.980369989722508,
+      "grad_norm": 0.1414353847503662,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 77649
+    },
+    {
+      "epoch": 7.980678314491264,
+      "grad_norm": 0.0823885053396225,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 77652
+    },
+    {
+      "epoch": 7.980986639260021,
+      "grad_norm": 0.058041948825120926,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 77655
+    },
+    {
+      "epoch": 7.981294964028777,
+      "grad_norm": 0.04310723394155502,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 77658
+    },
+    {
+      "epoch": 7.981603288797533,
+      "grad_norm": 0.041882872581481934,
+      "learning_rate": 0.01,
+      "loss": 1.9996,
+      "step": 77661
+    },
+    {
+      "epoch": 7.9819116135662895,
+      "grad_norm": 0.05595613643527031,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 77664
+    },
+    {
+      "epoch": 7.982219938335046,
+      "grad_norm": 0.0313473679125309,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 77667
+    },
+    {
+      "epoch": 7.982528263103803,
+      "grad_norm": 0.059281859546899796,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 77670
+    },
+    {
+      "epoch": 7.982836587872559,
+      "grad_norm": 0.03586166724562645,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 77673
+    },
+    {
+      "epoch": 7.983144912641316,
+      "grad_norm": 0.04249844327569008,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 77676
+    },
+    {
+      "epoch": 7.983453237410072,
+      "grad_norm": 0.05303768441081047,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 77679
+    },
+    {
+      "epoch": 7.983761562178828,
+      "grad_norm": 0.05120926350355148,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 77682
+    },
+    {
+      "epoch": 7.984069886947585,
+      "grad_norm": 0.11158134043216705,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 77685
+    },
+    {
+      "epoch": 7.984378211716341,
+      "grad_norm": 0.03699832409620285,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 77688
+    },
+    {
+      "epoch": 7.984686536485098,
+      "grad_norm": 0.03826748579740524,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 77691
+    },
+    {
+      "epoch": 7.984994861253854,
+      "grad_norm": 0.03865990415215492,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 77694
+    },
+    {
+      "epoch": 7.985303186022611,
+      "grad_norm": 0.07432328909635544,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 77697
+    },
+    {
+      "epoch": 7.985611510791367,
+      "grad_norm": 0.11026125401258469,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 77700
+    },
+    {
+      "epoch": 7.985919835560123,
+      "grad_norm": 0.08051371574401855,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 77703
+    },
+    {
+      "epoch": 7.98622816032888,
+      "grad_norm": 0.04520987346768379,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 77706
+    },
+    {
+      "epoch": 7.986536485097636,
+      "grad_norm": 0.08161762356758118,
+      "learning_rate": 0.01,
+      "loss": 1.9942,
+      "step": 77709
+    },
+    {
+      "epoch": 7.986844809866392,
+      "grad_norm": 0.08353172987699509,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 77712
+    },
+    {
+      "epoch": 7.9871531346351485,
+      "grad_norm": 0.05255158990621567,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 77715
+    },
+    {
+      "epoch": 7.987461459403906,
+      "grad_norm": 0.06637035310268402,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 77718
+    },
+    {
+      "epoch": 7.987769784172662,
+      "grad_norm": 0.05454893782734871,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 77721
+    },
+    {
+      "epoch": 7.988078108941418,
+      "grad_norm": 0.06507664918899536,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 77724
+    },
+    {
+      "epoch": 7.988386433710175,
+      "grad_norm": 0.03984587639570236,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 77727
+    },
+    {
+      "epoch": 7.988694758478931,
+      "grad_norm": 0.03732791915535927,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 77730
+    },
+    {
+      "epoch": 7.989003083247687,
+      "grad_norm": 0.05173516273498535,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 77733
+    },
+    {
+      "epoch": 7.9893114080164445,
+      "grad_norm": 0.11875148862600327,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 77736
+    },
+    {
+      "epoch": 7.989619732785201,
+      "grad_norm": 0.0765930786728859,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 77739
+    },
+    {
+      "epoch": 7.989928057553957,
+      "grad_norm": 0.10672271251678467,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 77742
+    },
+    {
+      "epoch": 7.990236382322713,
+      "grad_norm": 0.08630343526601791,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 77745
+    },
+    {
+      "epoch": 7.99054470709147,
+      "grad_norm": 0.07723753899335861,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 77748
+    },
+    {
+      "epoch": 7.990853031860226,
+      "grad_norm": 0.05617719888687134,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 77751
+    },
+    {
+      "epoch": 7.991161356628982,
+      "grad_norm": 0.03311438858509064,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 77754
+    },
+    {
+      "epoch": 7.991469681397739,
+      "grad_norm": 0.03674701601266861,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 77757
+    },
+    {
+      "epoch": 7.991778006166495,
+      "grad_norm": 0.046452175825834274,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 77760
+    },
+    {
+      "epoch": 7.992086330935252,
+      "grad_norm": 0.12231157720088959,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 77763
+    },
+    {
+      "epoch": 7.992394655704008,
+      "grad_norm": 0.03398605436086655,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 77766
+    },
+    {
+      "epoch": 7.992702980472765,
+      "grad_norm": 0.0747060477733612,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 77769
+    },
+    {
+      "epoch": 7.993011305241521,
+      "grad_norm": 0.051723454147577286,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 77772
+    },
+    {
+      "epoch": 7.993319630010277,
+      "grad_norm": 0.045055799186229706,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 77775
+    },
+    {
+      "epoch": 7.993627954779034,
+      "grad_norm": 0.06425558030605316,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 77778
+    },
+    {
+      "epoch": 7.99393627954779,
+      "grad_norm": 0.04054182767868042,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 77781
+    },
+    {
+      "epoch": 7.994244604316547,
+      "grad_norm": 0.034561723470687866,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 77784
+    },
+    {
+      "epoch": 7.9945529290853035,
+      "grad_norm": 0.034934625029563904,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 77787
+    },
+    {
+      "epoch": 7.99486125385406,
+      "grad_norm": 0.04506867378950119,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 77790
+    },
+    {
+      "epoch": 7.995169578622816,
+      "grad_norm": 0.06831274926662445,
+      "learning_rate": 0.01,
+      "loss": 1.9826,
+      "step": 77793
+    },
+    {
+      "epoch": 7.995477903391572,
+      "grad_norm": 0.06778151541948318,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 77796
+    },
+    {
+      "epoch": 7.995786228160329,
+      "grad_norm": 0.1303955465555191,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 77799
+    },
+    {
+      "epoch": 7.996094552929085,
+      "grad_norm": 0.04436461254954338,
+      "learning_rate": 0.01,
+      "loss": 1.9891,
+      "step": 77802
+    },
+    {
+      "epoch": 7.996402877697841,
+      "grad_norm": 0.08111438155174255,
+      "learning_rate": 0.01,
+      "loss": 1.9764,
+      "step": 77805
+    },
+    {
+      "epoch": 7.9967112024665985,
+      "grad_norm": 0.050543639808893204,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 77808
+    },
+    {
+      "epoch": 7.997019527235355,
+      "grad_norm": 0.09472209960222244,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 77811
+    },
+    {
+      "epoch": 7.997327852004111,
+      "grad_norm": 0.05899638310074806,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 77814
+    },
+    {
+      "epoch": 7.997636176772867,
+      "grad_norm": 0.03349081799387932,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 77817
+    },
+    {
+      "epoch": 7.997944501541624,
+      "grad_norm": 0.03433087468147278,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 77820
+    },
+    {
+      "epoch": 7.99825282631038,
+      "grad_norm": 0.035451170057058334,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 77823
+    },
+    {
+      "epoch": 7.998561151079136,
+      "grad_norm": 0.07916496694087982,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 77826
+    },
+    {
+      "epoch": 7.9988694758478935,
+      "grad_norm": 0.08633740246295929,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 77829
+    },
+    {
+      "epoch": 7.99917780061665,
+      "grad_norm": 0.06353379040956497,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 77832
+    },
+    {
+      "epoch": 7.999486125385406,
+      "grad_norm": 0.040657125413417816,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 77835
+    },
+    {
+      "epoch": 7.9997944501541625,
+      "grad_norm": 0.09802965819835663,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 77838
+    },
+    {
+      "epoch": 8.003392967304134,
+      "grad_norm": 0.08189691603183746,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 77841
+    },
+    {
+      "epoch": 8.003701418877236,
+      "grad_norm": 0.04771219938993454,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 77844
+    },
+    {
+      "epoch": 8.00400987045034,
+      "grad_norm": 0.045312631875276566,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 77847
+    },
+    {
+      "epoch": 8.004318322023442,
+      "grad_norm": 0.077674999833107,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 77850
+    },
+    {
+      "epoch": 8.004626773596545,
+      "grad_norm": 0.03592463582754135,
+      "learning_rate": 0.01,
+      "loss": 2.0094,
+      "step": 77853
+    },
+    {
+      "epoch": 8.004935225169648,
+      "grad_norm": 0.034409284591674805,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 77856
+    },
+    {
+      "epoch": 8.00524367674275,
+      "grad_norm": 0.08009333908557892,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 77859
+    },
+    {
+      "epoch": 8.005552128315854,
+      "grad_norm": 0.06267620623111725,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 77862
+    },
+    {
+      "epoch": 8.005860579888957,
+      "grad_norm": 0.06999866664409637,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 77865
+    },
+    {
+      "epoch": 8.006169031462061,
+      "grad_norm": 0.04267257824540138,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 77868
+    },
+    {
+      "epoch": 8.006477483035164,
+      "grad_norm": 0.13487477600574493,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 77871
+    },
+    {
+      "epoch": 8.006785934608267,
+      "grad_norm": 0.1191130056977272,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 77874
+    },
+    {
+      "epoch": 8.00709438618137,
+      "grad_norm": 0.10439812391996384,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 77877
+    },
+    {
+      "epoch": 8.007402837754473,
+      "grad_norm": 0.08259162306785583,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 77880
+    },
+    {
+      "epoch": 8.007711289327576,
+      "grad_norm": 0.05473419278860092,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 77883
+    },
+    {
+      "epoch": 8.008019740900679,
+      "grad_norm": 0.06642327457666397,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 77886
+    },
+    {
+      "epoch": 8.008328192473781,
+      "grad_norm": 0.06296679377555847,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 77889
+    },
+    {
+      "epoch": 8.008636644046884,
+      "grad_norm": 0.03993913158774376,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 77892
+    },
+    {
+      "epoch": 8.008945095619987,
+      "grad_norm": 0.03741374984383583,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 77895
+    },
+    {
+      "epoch": 8.00925354719309,
+      "grad_norm": 0.03153381496667862,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 77898
+    },
+    {
+      "epoch": 8.009561998766193,
+      "grad_norm": 0.04553218185901642,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 77901
+    },
+    {
+      "epoch": 8.009870450339296,
+      "grad_norm": 0.10947732627391815,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 77904
+    },
+    {
+      "epoch": 8.0101789019124,
+      "grad_norm": 0.0484643280506134,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 77907
+    },
+    {
+      "epoch": 8.010487353485503,
+      "grad_norm": 0.05445571616292,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 77910
+    },
+    {
+      "epoch": 8.010795805058606,
+      "grad_norm": 0.061891742050647736,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 77913
+    },
+    {
+      "epoch": 8.01110425663171,
+      "grad_norm": 0.04123906418681145,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 77916
+    },
+    {
+      "epoch": 8.011412708204812,
+      "grad_norm": 0.03290227800607681,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 77919
+    },
+    {
+      "epoch": 8.011721159777915,
+      "grad_norm": 0.06250923126935959,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 77922
+    },
+    {
+      "epoch": 8.012029611351018,
+      "grad_norm": 0.05352925509214401,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 77925
+    },
+    {
+      "epoch": 8.01233806292412,
+      "grad_norm": 0.1590663343667984,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 77928
+    },
+    {
+      "epoch": 8.012646514497224,
+      "grad_norm": 0.09142278879880905,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 77931
+    },
+    {
+      "epoch": 8.012954966070327,
+      "grad_norm": 0.1245221421122551,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 77934
+    },
+    {
+      "epoch": 8.01326341764343,
+      "grad_norm": 0.06170954182744026,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 77937
+    },
+    {
+      "epoch": 8.013571869216532,
+      "grad_norm": 0.042779482901096344,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 77940
+    },
+    {
+      "epoch": 8.013880320789635,
+      "grad_norm": 0.030806196853518486,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 77943
+    },
+    {
+      "epoch": 8.01418877236274,
+      "grad_norm": 0.038949958980083466,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 77946
+    },
+    {
+      "epoch": 8.014497223935843,
+      "grad_norm": 0.049347564578056335,
+      "learning_rate": 0.01,
+      "loss": 1.9765,
+      "step": 77949
+    },
+    {
+      "epoch": 8.014805675508946,
+      "grad_norm": 0.06526634842157364,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 77952
+    },
+    {
+      "epoch": 8.015114127082049,
+      "grad_norm": 0.09283331781625748,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 77955
+    },
+    {
+      "epoch": 8.015422578655151,
+      "grad_norm": 0.13813799619674683,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 77958
+    },
+    {
+      "epoch": 8.015731030228254,
+      "grad_norm": 0.06759168952703476,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 77961
+    },
+    {
+      "epoch": 8.016039481801357,
+      "grad_norm": 0.029820196330547333,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 77964
+    },
+    {
+      "epoch": 8.01634793337446,
+      "grad_norm": 0.042072828859090805,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 77967
+    },
+    {
+      "epoch": 8.016656384947563,
+      "grad_norm": 0.030679654330015182,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 77970
+    },
+    {
+      "epoch": 8.016964836520666,
+      "grad_norm": 0.054546356201171875,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 77973
+    },
+    {
+      "epoch": 8.017273288093769,
+      "grad_norm": 0.12222815304994583,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 77976
+    },
+    {
+      "epoch": 8.017581739666872,
+      "grad_norm": 0.071864552795887,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 77979
+    },
+    {
+      "epoch": 8.017890191239974,
+      "grad_norm": 0.08329971879720688,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 77982
+    },
+    {
+      "epoch": 8.01819864281308,
+      "grad_norm": 0.051458850502967834,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 77985
+    },
+    {
+      "epoch": 8.018507094386182,
+      "grad_norm": 0.02989008091390133,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 77988
+    },
+    {
+      "epoch": 8.018815545959285,
+      "grad_norm": 0.03450706601142883,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 77991
+    },
+    {
+      "epoch": 8.019123997532388,
+      "grad_norm": 0.13265404105186462,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 77994
+    },
+    {
+      "epoch": 8.01943244910549,
+      "grad_norm": 0.04526760056614876,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 77997
+    },
+    {
+      "epoch": 8.019740900678594,
+      "grad_norm": 0.0469353124499321,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 78000
+    },
+    {
+      "epoch": 8.020049352251696,
+      "grad_norm": 0.1211186870932579,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 78003
+    },
+    {
+      "epoch": 8.0203578038248,
+      "grad_norm": 0.07616788893938065,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 78006
+    },
+    {
+      "epoch": 8.020666255397902,
+      "grad_norm": 0.06855116039514542,
+      "learning_rate": 0.01,
+      "loss": 1.9846,
+      "step": 78009
+    },
+    {
+      "epoch": 8.020974706971005,
+      "grad_norm": 0.04688597843050957,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 78012
+    },
+    {
+      "epoch": 8.021283158544108,
+      "grad_norm": 0.05328085646033287,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 78015
+    },
+    {
+      "epoch": 8.021591610117211,
+      "grad_norm": 0.051051221787929535,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 78018
+    },
+    {
+      "epoch": 8.021900061690314,
+      "grad_norm": 0.11757440865039825,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 78021
+    },
+    {
+      "epoch": 8.022208513263418,
+      "grad_norm": 0.04119103401899338,
+      "learning_rate": 0.01,
+      "loss": 1.9864,
+      "step": 78024
+    },
+    {
+      "epoch": 8.022516964836521,
+      "grad_norm": 0.03476836904883385,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 78027
+    },
+    {
+      "epoch": 8.022825416409624,
+      "grad_norm": 0.035968925803899765,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 78030
+    },
+    {
+      "epoch": 8.023133867982727,
+      "grad_norm": 0.12972131371498108,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 78033
+    },
+    {
+      "epoch": 8.02344231955583,
+      "grad_norm": 0.06801027059555054,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 78036
+    },
+    {
+      "epoch": 8.023750771128933,
+      "grad_norm": 0.04971321299672127,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 78039
+    },
+    {
+      "epoch": 8.024059222702036,
+      "grad_norm": 0.035642463713884354,
+      "learning_rate": 0.01,
+      "loss": 1.9866,
+      "step": 78042
+    },
+    {
+      "epoch": 8.024367674275139,
+      "grad_norm": 0.09247110038995743,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 78045
+    },
+    {
+      "epoch": 8.024676125848242,
+      "grad_norm": 0.10186565667390823,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 78048
+    },
+    {
+      "epoch": 8.024984577421344,
+      "grad_norm": 0.040058303624391556,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 78051
+    },
+    {
+      "epoch": 8.025293028994447,
+      "grad_norm": 0.09354520589113235,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 78054
+    },
+    {
+      "epoch": 8.02560148056755,
+      "grad_norm": 0.07421064376831055,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 78057
+    },
+    {
+      "epoch": 8.025909932140653,
+      "grad_norm": 0.057933758944272995,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 78060
+    },
+    {
+      "epoch": 8.026218383713758,
+      "grad_norm": 0.05190437287092209,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 78063
+    },
+    {
+      "epoch": 8.02652683528686,
+      "grad_norm": 0.06811270117759705,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 78066
+    },
+    {
+      "epoch": 8.026835286859964,
+      "grad_norm": 0.0652383342385292,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 78069
+    },
+    {
+      "epoch": 8.027143738433066,
+      "grad_norm": 0.04355868324637413,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 78072
+    },
+    {
+      "epoch": 8.02745219000617,
+      "grad_norm": 0.04373233765363693,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 78075
+    },
+    {
+      "epoch": 8.027760641579272,
+      "grad_norm": 0.04739624261856079,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 78078
+    },
+    {
+      "epoch": 8.028069093152375,
+      "grad_norm": 0.10096476972103119,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 78081
+    },
+    {
+      "epoch": 8.028377544725478,
+      "grad_norm": 0.04511374235153198,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 78084
+    },
+    {
+      "epoch": 8.02868599629858,
+      "grad_norm": 0.1182047426700592,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 78087
+    },
+    {
+      "epoch": 8.028994447871684,
+      "grad_norm": 0.044908009469509125,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 78090
+    },
+    {
+      "epoch": 8.029302899444787,
+      "grad_norm": 0.03893660008907318,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 78093
+    },
+    {
+      "epoch": 8.02961135101789,
+      "grad_norm": 0.08515772223472595,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 78096
+    },
+    {
+      "epoch": 8.029919802590992,
+      "grad_norm": 0.09454181045293808,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 78099
+    },
+    {
+      "epoch": 8.030228254164097,
+      "grad_norm": 0.050406310707330704,
+      "learning_rate": 0.01,
+      "loss": 1.9842,
+      "step": 78102
+    },
+    {
+      "epoch": 8.0305367057372,
+      "grad_norm": 0.1353389024734497,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 78105
+    },
+    {
+      "epoch": 8.030845157310303,
+      "grad_norm": 0.14368806779384613,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 78108
+    },
+    {
+      "epoch": 8.031153608883406,
+      "grad_norm": 0.11506987363100052,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 78111
+    },
+    {
+      "epoch": 8.031462060456509,
+      "grad_norm": 0.04621575027704239,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 78114
+    },
+    {
+      "epoch": 8.031770512029611,
+      "grad_norm": 0.06082366406917572,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 78117
+    },
+    {
+      "epoch": 8.032078963602714,
+      "grad_norm": 0.053143613040447235,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 78120
+    },
+    {
+      "epoch": 8.032387415175817,
+      "grad_norm": 0.04185790941119194,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 78123
+    },
+    {
+      "epoch": 8.03269586674892,
+      "grad_norm": 0.06397274881601334,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 78126
+    },
+    {
+      "epoch": 8.033004318322023,
+      "grad_norm": 0.1405619978904724,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 78129
+    },
+    {
+      "epoch": 8.033312769895126,
+      "grad_norm": 0.04655090346932411,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 78132
+    },
+    {
+      "epoch": 8.033621221468229,
+      "grad_norm": 0.041024383157491684,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 78135
+    },
+    {
+      "epoch": 8.033929673041332,
+      "grad_norm": 0.04685423895716667,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 78138
+    },
+    {
+      "epoch": 8.034238124614436,
+      "grad_norm": 0.044466495513916016,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 78141
+    },
+    {
+      "epoch": 8.03454657618754,
+      "grad_norm": 0.048852045089006424,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 78144
+    },
+    {
+      "epoch": 8.034855027760642,
+      "grad_norm": 0.03688976168632507,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 78147
+    },
+    {
+      "epoch": 8.035163479333745,
+      "grad_norm": 0.05261220410466194,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 78150
+    },
+    {
+      "epoch": 8.035471930906848,
+      "grad_norm": 0.09963783621788025,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 78153
+    },
+    {
+      "epoch": 8.03578038247995,
+      "grad_norm": 0.035708215087652206,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 78156
+    },
+    {
+      "epoch": 8.036088834053054,
+      "grad_norm": 0.031529273837804794,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 78159
+    },
+    {
+      "epoch": 8.036397285626157,
+      "grad_norm": 0.045165106654167175,
+      "learning_rate": 0.01,
+      "loss": 1.9857,
+      "step": 78162
+    },
+    {
+      "epoch": 8.03670573719926,
+      "grad_norm": 0.10784061253070831,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 78165
+    },
+    {
+      "epoch": 8.037014188772362,
+      "grad_norm": 0.04860498756170273,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 78168
+    },
+    {
+      "epoch": 8.037322640345465,
+      "grad_norm": 0.055530451238155365,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 78171
+    },
+    {
+      "epoch": 8.037631091918568,
+      "grad_norm": 0.03332025185227394,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 78174
+    },
+    {
+      "epoch": 8.037939543491671,
+      "grad_norm": 0.04176173731684685,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 78177
+    },
+    {
+      "epoch": 8.038247995064776,
+      "grad_norm": 0.04813525825738907,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 78180
+    },
+    {
+      "epoch": 8.038556446637878,
+      "grad_norm": 0.06635365635156631,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 78183
+    },
+    {
+      "epoch": 8.038864898210981,
+      "grad_norm": 0.12882399559020996,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 78186
+    },
+    {
+      "epoch": 8.039173349784084,
+      "grad_norm": 0.03530122712254524,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 78189
+    },
+    {
+      "epoch": 8.039481801357187,
+      "grad_norm": 0.11424420028924942,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 78192
+    },
+    {
+      "epoch": 8.03979025293029,
+      "grad_norm": 0.05186377838253975,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 78195
+    },
+    {
+      "epoch": 8.040098704503393,
+      "grad_norm": 0.09932134300470352,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 78198
+    },
+    {
+      "epoch": 8.040407156076496,
+      "grad_norm": 0.05855415016412735,
+      "learning_rate": 0.01,
+      "loss": 1.9889,
+      "step": 78201
+    },
+    {
+      "epoch": 8.040715607649599,
+      "grad_norm": 0.041826751083135605,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 78204
+    },
+    {
+      "epoch": 8.041024059222702,
+      "grad_norm": 0.04150779917836189,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 78207
+    },
+    {
+      "epoch": 8.041332510795804,
+      "grad_norm": 0.04734393209218979,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 78210
+    },
+    {
+      "epoch": 8.041640962368907,
+      "grad_norm": 0.03574753925204277,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 78213
+    },
+    {
+      "epoch": 8.04194941394201,
+      "grad_norm": 0.05483570322394371,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 78216
+    },
+    {
+      "epoch": 8.042257865515115,
+      "grad_norm": 0.10191001743078232,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 78219
+    },
+    {
+      "epoch": 8.042566317088218,
+      "grad_norm": 0.0770745575428009,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 78222
+    },
+    {
+      "epoch": 8.04287476866132,
+      "grad_norm": 0.04721737280488014,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 78225
+    },
+    {
+      "epoch": 8.043183220234424,
+      "grad_norm": 0.06755993515253067,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 78228
+    },
+    {
+      "epoch": 8.043491671807526,
+      "grad_norm": 0.07830438762903214,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 78231
+    },
+    {
+      "epoch": 8.04380012338063,
+      "grad_norm": 0.08949405699968338,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 78234
+    },
+    {
+      "epoch": 8.044108574953732,
+      "grad_norm": 0.05038638040423393,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 78237
+    },
+    {
+      "epoch": 8.044417026526835,
+      "grad_norm": 0.07417938113212585,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 78240
+    },
+    {
+      "epoch": 8.044725478099938,
+      "grad_norm": 0.07824010401964188,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 78243
+    },
+    {
+      "epoch": 8.04503392967304,
+      "grad_norm": 0.06778912246227264,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 78246
+    },
+    {
+      "epoch": 8.045342381246144,
+      "grad_norm": 0.09098850935697556,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 78249
+    },
+    {
+      "epoch": 8.045650832819247,
+      "grad_norm": 0.07562565058469772,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 78252
+    },
+    {
+      "epoch": 8.04595928439235,
+      "grad_norm": 0.06573687493801117,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 78255
+    },
+    {
+      "epoch": 8.046267735965454,
+      "grad_norm": 0.08350115269422531,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 78258
+    },
+    {
+      "epoch": 8.046576187538557,
+      "grad_norm": 0.06631845980882645,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 78261
+    },
+    {
+      "epoch": 8.04688463911166,
+      "grad_norm": 0.10218732804059982,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 78264
+    },
+    {
+      "epoch": 8.047193090684763,
+      "grad_norm": 0.08082437515258789,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 78267
+    },
+    {
+      "epoch": 8.047501542257866,
+      "grad_norm": 0.07313011586666107,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 78270
+    },
+    {
+      "epoch": 8.047809993830969,
+      "grad_norm": 0.07694101333618164,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 78273
+    },
+    {
+      "epoch": 8.048118445404072,
+      "grad_norm": 0.09288178384304047,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 78276
+    },
+    {
+      "epoch": 8.048426896977174,
+      "grad_norm": 0.13423343002796173,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 78279
+    },
+    {
+      "epoch": 8.048735348550277,
+      "grad_norm": 0.06640274822711945,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 78282
+    },
+    {
+      "epoch": 8.04904380012338,
+      "grad_norm": 0.042986899614334106,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 78285
+    },
+    {
+      "epoch": 8.049352251696483,
+      "grad_norm": 0.08272531628608704,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 78288
+    },
+    {
+      "epoch": 8.049660703269586,
+      "grad_norm": 0.06874362379312515,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 78291
+    },
+    {
+      "epoch": 8.049969154842689,
+      "grad_norm": 0.0621911846101284,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 78294
+    },
+    {
+      "epoch": 8.050277606415793,
+      "grad_norm": 0.10276035964488983,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 78297
+    },
+    {
+      "epoch": 8.050586057988896,
+      "grad_norm": 0.0899696797132492,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 78300
+    },
+    {
+      "epoch": 8.050894509562,
+      "grad_norm": 0.05405474826693535,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 78303
+    },
+    {
+      "epoch": 8.051202961135102,
+      "grad_norm": 0.10780783742666245,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 78306
+    },
+    {
+      "epoch": 8.051511412708205,
+      "grad_norm": 0.057042498141527176,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 78309
+    },
+    {
+      "epoch": 8.051819864281308,
+      "grad_norm": 0.08696413785219193,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 78312
+    },
+    {
+      "epoch": 8.05212831585441,
+      "grad_norm": 0.05428167060017586,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 78315
+    },
+    {
+      "epoch": 8.052436767427514,
+      "grad_norm": 0.03312308341264725,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 78318
+    },
+    {
+      "epoch": 8.052745219000617,
+      "grad_norm": 0.054687369614839554,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 78321
+    },
+    {
+      "epoch": 8.05305367057372,
+      "grad_norm": 0.078495554625988,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 78324
+    },
+    {
+      "epoch": 8.053362122146822,
+      "grad_norm": 0.0804872140288353,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 78327
+    },
+    {
+      "epoch": 8.053670573719925,
+      "grad_norm": 0.0818592831492424,
+      "learning_rate": 0.01,
+      "loss": 1.9901,
+      "step": 78330
+    },
+    {
+      "epoch": 8.053979025293028,
+      "grad_norm": 0.0550164058804512,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 78333
+    },
+    {
+      "epoch": 8.054287476866133,
+      "grad_norm": 0.04894454777240753,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 78336
+    },
+    {
+      "epoch": 8.054595928439236,
+      "grad_norm": 0.04496561363339424,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 78339
+    },
+    {
+      "epoch": 8.054904380012339,
+      "grad_norm": 0.04322918504476547,
+      "learning_rate": 0.01,
+      "loss": 1.9787,
+      "step": 78342
+    },
+    {
+      "epoch": 8.055212831585441,
+      "grad_norm": 0.05272132530808449,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 78345
+    },
+    {
+      "epoch": 8.055521283158544,
+      "grad_norm": 0.045108698308467865,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 78348
+    },
+    {
+      "epoch": 8.055829734731647,
+      "grad_norm": 0.10814124345779419,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 78351
+    },
+    {
+      "epoch": 8.05613818630475,
+      "grad_norm": 0.04985149949789047,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 78354
+    },
+    {
+      "epoch": 8.056446637877853,
+      "grad_norm": 0.10839637368917465,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 78357
+    },
+    {
+      "epoch": 8.056755089450956,
+      "grad_norm": 0.0739506408572197,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 78360
+    },
+    {
+      "epoch": 8.057063541024059,
+      "grad_norm": 0.0455130897462368,
+      "learning_rate": 0.01,
+      "loss": 1.9867,
+      "step": 78363
+    },
+    {
+      "epoch": 8.057371992597162,
+      "grad_norm": 0.04794808104634285,
+      "learning_rate": 0.01,
+      "loss": 1.9912,
+      "step": 78366
+    },
+    {
+      "epoch": 8.057680444170265,
+      "grad_norm": 0.10262172669172287,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 78369
+    },
+    {
+      "epoch": 8.057988895743367,
+      "grad_norm": 0.1068585216999054,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 78372
+    },
+    {
+      "epoch": 8.058297347316472,
+      "grad_norm": 0.0809110701084137,
+      "learning_rate": 0.01,
+      "loss": 1.9853,
+      "step": 78375
+    },
+    {
+      "epoch": 8.058605798889575,
+      "grad_norm": 0.03217419609427452,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 78378
+    },
+    {
+      "epoch": 8.058914250462678,
+      "grad_norm": 0.07601505517959595,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 78381
+    },
+    {
+      "epoch": 8.05922270203578,
+      "grad_norm": 0.1304755061864853,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 78384
+    },
+    {
+      "epoch": 8.059531153608884,
+      "grad_norm": 0.06895633786916733,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 78387
+    },
+    {
+      "epoch": 8.059839605181986,
+      "grad_norm": 0.07370155304670334,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 78390
+    },
+    {
+      "epoch": 8.06014805675509,
+      "grad_norm": 0.07934858649969101,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 78393
+    },
+    {
+      "epoch": 8.060456508328192,
+      "grad_norm": 0.16500116884708405,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 78396
+    },
+    {
+      "epoch": 8.060764959901295,
+      "grad_norm": 0.036092065274715424,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 78399
+    },
+    {
+      "epoch": 8.061073411474398,
+      "grad_norm": 0.03540043160319328,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 78402
+    },
+    {
+      "epoch": 8.061381863047501,
+      "grad_norm": 0.050339192152023315,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 78405
+    },
+    {
+      "epoch": 8.061690314620604,
+      "grad_norm": 0.05511642247438431,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 78408
+    },
+    {
+      "epoch": 8.061998766193707,
+      "grad_norm": 0.033187735825777054,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 78411
+    },
+    {
+      "epoch": 8.062307217766811,
+      "grad_norm": 0.06421135365962982,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 78414
+    },
+    {
+      "epoch": 8.062615669339914,
+      "grad_norm": 0.042910970747470856,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 78417
+    },
+    {
+      "epoch": 8.062924120913017,
+      "grad_norm": 0.06334998458623886,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 78420
+    },
+    {
+      "epoch": 8.06323257248612,
+      "grad_norm": 0.0897008404135704,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 78423
+    },
+    {
+      "epoch": 8.063541024059223,
+      "grad_norm": 0.07113679498434067,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 78426
+    },
+    {
+      "epoch": 8.063849475632326,
+      "grad_norm": 0.08476056903600693,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 78429
+    },
+    {
+      "epoch": 8.064157927205429,
+      "grad_norm": 0.05988757312297821,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 78432
+    },
+    {
+      "epoch": 8.064466378778532,
+      "grad_norm": 0.13881082832813263,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 78435
+    },
+    {
+      "epoch": 8.064774830351634,
+      "grad_norm": 0.05506844446063042,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 78438
+    },
+    {
+      "epoch": 8.065083281924737,
+      "grad_norm": 0.057986658066511154,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 78441
+    },
+    {
+      "epoch": 8.06539173349784,
+      "grad_norm": 0.06731247901916504,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 78444
+    },
+    {
+      "epoch": 8.065700185070943,
+      "grad_norm": 0.10083737224340439,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 78447
+    },
+    {
+      "epoch": 8.066008636644046,
+      "grad_norm": 0.11750660091638565,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 78450
+    },
+    {
+      "epoch": 8.06631708821715,
+      "grad_norm": 0.07133502513170242,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 78453
+    },
+    {
+      "epoch": 8.066625539790254,
+      "grad_norm": 0.055653370916843414,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 78456
+    },
+    {
+      "epoch": 8.066933991363356,
+      "grad_norm": 0.04406362399458885,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 78459
+    },
+    {
+      "epoch": 8.06724244293646,
+      "grad_norm": 0.07277071475982666,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 78462
+    },
+    {
+      "epoch": 8.067550894509562,
+      "grad_norm": 0.09624621272087097,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 78465
+    },
+    {
+      "epoch": 8.067859346082665,
+      "grad_norm": 0.12320689857006073,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 78468
+    },
+    {
+      "epoch": 8.068167797655768,
+      "grad_norm": 0.05010239779949188,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 78471
+    },
+    {
+      "epoch": 8.06847624922887,
+      "grad_norm": 0.09163736552000046,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 78474
+    },
+    {
+      "epoch": 8.068784700801974,
+      "grad_norm": 0.07241861522197723,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 78477
+    },
+    {
+      "epoch": 8.069093152375077,
+      "grad_norm": 0.09269708395004272,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 78480
+    },
+    {
+      "epoch": 8.06940160394818,
+      "grad_norm": 0.12642695009708405,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 78483
+    },
+    {
+      "epoch": 8.069710055521282,
+      "grad_norm": 0.051001306623220444,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 78486
+    },
+    {
+      "epoch": 8.070018507094387,
+      "grad_norm": 0.036773428320884705,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 78489
+    },
+    {
+      "epoch": 8.07032695866749,
+      "grad_norm": 0.0340329185128212,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 78492
+    },
+    {
+      "epoch": 8.070635410240593,
+      "grad_norm": 0.05808752402663231,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 78495
+    },
+    {
+      "epoch": 8.070943861813696,
+      "grad_norm": 0.07179629802703857,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 78498
+    },
+    {
+      "epoch": 8.071252313386799,
+      "grad_norm": 0.0458885133266449,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 78501
+    },
+    {
+      "epoch": 8.071560764959901,
+      "grad_norm": 0.04442944377660751,
+      "learning_rate": 0.01,
+      "loss": 1.9907,
+      "step": 78504
+    },
+    {
+      "epoch": 8.071869216533004,
+      "grad_norm": 0.08146031200885773,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 78507
+    },
+    {
+      "epoch": 8.072177668106107,
+      "grad_norm": 0.08753294497728348,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 78510
+    },
+    {
+      "epoch": 8.07248611967921,
+      "grad_norm": 0.12875844538211823,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 78513
+    },
+    {
+      "epoch": 8.072794571252313,
+      "grad_norm": 0.11097821593284607,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 78516
+    },
+    {
+      "epoch": 8.073103022825416,
+      "grad_norm": 0.03981398791074753,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 78519
+    },
+    {
+      "epoch": 8.073411474398519,
+      "grad_norm": 0.047386813908815384,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 78522
+    },
+    {
+      "epoch": 8.073719925971622,
+      "grad_norm": 0.058222249150276184,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 78525
+    },
+    {
+      "epoch": 8.074028377544726,
+      "grad_norm": 0.04991047456860542,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 78528
+    },
+    {
+      "epoch": 8.07433682911783,
+      "grad_norm": 0.08375823497772217,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 78531
+    },
+    {
+      "epoch": 8.074645280690932,
+      "grad_norm": 0.07699902355670929,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 78534
+    },
+    {
+      "epoch": 8.074953732264035,
+      "grad_norm": 0.10210588574409485,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 78537
+    },
+    {
+      "epoch": 8.075262183837138,
+      "grad_norm": 0.04255238175392151,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 78540
+    },
+    {
+      "epoch": 8.07557063541024,
+      "grad_norm": 0.059122033417224884,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 78543
+    },
+    {
+      "epoch": 8.075879086983344,
+      "grad_norm": 0.11363697797060013,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 78546
+    },
+    {
+      "epoch": 8.076187538556447,
+      "grad_norm": 0.11338911950588226,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 78549
+    },
+    {
+      "epoch": 8.07649599012955,
+      "grad_norm": 0.049438465386629105,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 78552
+    },
+    {
+      "epoch": 8.076804441702652,
+      "grad_norm": 0.040645066648721695,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 78555
+    },
+    {
+      "epoch": 8.077112893275755,
+      "grad_norm": 0.1113462969660759,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 78558
+    },
+    {
+      "epoch": 8.077421344848858,
+      "grad_norm": 0.07110290974378586,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 78561
+    },
+    {
+      "epoch": 8.077729796421961,
+      "grad_norm": 0.12881991267204285,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 78564
+    },
+    {
+      "epoch": 8.078038247995066,
+      "grad_norm": 0.048802804201841354,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 78567
+    },
+    {
+      "epoch": 8.078346699568169,
+      "grad_norm": 0.049757737666368484,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 78570
+    },
+    {
+      "epoch": 8.078655151141271,
+      "grad_norm": 0.1158534437417984,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 78573
+    },
+    {
+      "epoch": 8.078963602714374,
+      "grad_norm": 0.037280384451150894,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 78576
+    },
+    {
+      "epoch": 8.079272054287477,
+      "grad_norm": 0.07610873132944107,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 78579
+    },
+    {
+      "epoch": 8.07958050586058,
+      "grad_norm": 0.09582018852233887,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 78582
+    },
+    {
+      "epoch": 8.079888957433683,
+      "grad_norm": 0.0594031997025013,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 78585
+    },
+    {
+      "epoch": 8.080197409006786,
+      "grad_norm": 0.10252749919891357,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 78588
+    },
+    {
+      "epoch": 8.080505860579889,
+      "grad_norm": 0.05171642452478409,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 78591
+    },
+    {
+      "epoch": 8.080814312152992,
+      "grad_norm": 0.044461604207754135,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 78594
+    },
+    {
+      "epoch": 8.081122763726095,
+      "grad_norm": 0.04300977289676666,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 78597
+    },
+    {
+      "epoch": 8.081431215299197,
+      "grad_norm": 0.04183531552553177,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 78600
+    },
+    {
+      "epoch": 8.0817396668723,
+      "grad_norm": 0.14074847102165222,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 78603
+    },
+    {
+      "epoch": 8.082048118445405,
+      "grad_norm": 0.08665565401315689,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 78606
+    },
+    {
+      "epoch": 8.082356570018508,
+      "grad_norm": 0.10652801394462585,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 78609
+    },
+    {
+      "epoch": 8.08266502159161,
+      "grad_norm": 0.08588683605194092,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 78612
+    },
+    {
+      "epoch": 8.082973473164714,
+      "grad_norm": 0.05880355089902878,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 78615
+    },
+    {
+      "epoch": 8.083281924737816,
+      "grad_norm": 0.052834395319223404,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 78618
+    },
+    {
+      "epoch": 8.08359037631092,
+      "grad_norm": 0.03865375742316246,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 78621
+    },
+    {
+      "epoch": 8.083898827884022,
+      "grad_norm": 0.04939945414662361,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 78624
+    },
+    {
+      "epoch": 8.084207279457125,
+      "grad_norm": 0.10530546307563782,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 78627
+    },
+    {
+      "epoch": 8.084515731030228,
+      "grad_norm": 0.07206293940544128,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 78630
+    },
+    {
+      "epoch": 8.084824182603331,
+      "grad_norm": 0.05856684222817421,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 78633
+    },
+    {
+      "epoch": 8.085132634176434,
+      "grad_norm": 0.07027263939380646,
+      "learning_rate": 0.01,
+      "loss": 1.9876,
+      "step": 78636
+    },
+    {
+      "epoch": 8.085441085749537,
+      "grad_norm": 0.09340980648994446,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 78639
+    },
+    {
+      "epoch": 8.08574953732264,
+      "grad_norm": 0.06842325627803802,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 78642
+    },
+    {
+      "epoch": 8.086057988895744,
+      "grad_norm": 0.06957598775625229,
+      "learning_rate": 0.01,
+      "loss": 1.9856,
+      "step": 78645
+    },
+    {
+      "epoch": 8.086366440468847,
+      "grad_norm": 0.09373968094587326,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 78648
+    },
+    {
+      "epoch": 8.08667489204195,
+      "grad_norm": 0.05674611032009125,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 78651
+    },
+    {
+      "epoch": 8.086983343615053,
+      "grad_norm": 0.0915849506855011,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 78654
+    },
+    {
+      "epoch": 8.087291795188156,
+      "grad_norm": 0.08001093566417694,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 78657
+    },
+    {
+      "epoch": 8.087600246761259,
+      "grad_norm": 0.09341204166412354,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 78660
+    },
+    {
+      "epoch": 8.087908698334362,
+      "grad_norm": 0.04382408410310745,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 78663
+    },
+    {
+      "epoch": 8.088217149907464,
+      "grad_norm": 0.038650207221508026,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 78666
+    },
+    {
+      "epoch": 8.088525601480567,
+      "grad_norm": 0.05687751621007919,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 78669
+    },
+    {
+      "epoch": 8.08883405305367,
+      "grad_norm": 0.07734004408121109,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 78672
+    },
+    {
+      "epoch": 8.089142504626773,
+      "grad_norm": 0.08115539699792862,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 78675
+    },
+    {
+      "epoch": 8.089450956199876,
+      "grad_norm": 0.07764489203691483,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 78678
+    },
+    {
+      "epoch": 8.089759407772979,
+      "grad_norm": 0.04523799568414688,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 78681
+    },
+    {
+      "epoch": 8.090067859346084,
+      "grad_norm": 0.06825567036867142,
+      "learning_rate": 0.01,
+      "loss": 1.9818,
+      "step": 78684
+    },
+    {
+      "epoch": 8.090376310919186,
+      "grad_norm": 0.10438218712806702,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 78687
+    },
+    {
+      "epoch": 8.09068476249229,
+      "grad_norm": 0.0568692572414875,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 78690
+    },
+    {
+      "epoch": 8.090993214065392,
+      "grad_norm": 0.0484088696539402,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 78693
+    },
+    {
+      "epoch": 8.091301665638495,
+      "grad_norm": 0.09627815335988998,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 78696
+    },
+    {
+      "epoch": 8.091610117211598,
+      "grad_norm": 0.06979174166917801,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 78699
+    },
+    {
+      "epoch": 8.0919185687847,
+      "grad_norm": 0.07845932990312576,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 78702
+    },
+    {
+      "epoch": 8.092227020357804,
+      "grad_norm": 0.0819554328918457,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 78705
+    },
+    {
+      "epoch": 8.092535471930907,
+      "grad_norm": 0.07853356748819351,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 78708
+    },
+    {
+      "epoch": 8.09284392350401,
+      "grad_norm": 0.04697944596409798,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 78711
+    },
+    {
+      "epoch": 8.093152375077112,
+      "grad_norm": 0.15374042093753815,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 78714
+    },
+    {
+      "epoch": 8.093460826650215,
+      "grad_norm": 0.05507707968354225,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 78717
+    },
+    {
+      "epoch": 8.093769278223318,
+      "grad_norm": 0.08202649652957916,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 78720
+    },
+    {
+      "epoch": 8.094077729796423,
+      "grad_norm": 0.06541413813829422,
+      "learning_rate": 0.01,
+      "loss": 1.9949,
+      "step": 78723
+    },
+    {
+      "epoch": 8.094386181369526,
+      "grad_norm": 0.08683803677558899,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 78726
+    },
+    {
+      "epoch": 8.094694632942629,
+      "grad_norm": 0.11512414366006851,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 78729
+    },
+    {
+      "epoch": 8.095003084515731,
+      "grad_norm": 0.08850602805614471,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 78732
+    },
+    {
+      "epoch": 8.095311536088834,
+      "grad_norm": 0.13835574686527252,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 78735
+    },
+    {
+      "epoch": 8.095619987661937,
+      "grad_norm": 0.080533467233181,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 78738
+    },
+    {
+      "epoch": 8.09592843923504,
+      "grad_norm": 0.07228211313486099,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 78741
+    },
+    {
+      "epoch": 8.096236890808143,
+      "grad_norm": 0.05860430374741554,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 78744
+    },
+    {
+      "epoch": 8.096545342381246,
+      "grad_norm": 0.03812957927584648,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 78747
+    },
+    {
+      "epoch": 8.096853793954349,
+      "grad_norm": 0.04736247658729553,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 78750
+    },
+    {
+      "epoch": 8.097162245527452,
+      "grad_norm": 0.0691232830286026,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 78753
+    },
+    {
+      "epoch": 8.097470697100555,
+      "grad_norm": 0.09025582671165466,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 78756
+    },
+    {
+      "epoch": 8.097779148673657,
+      "grad_norm": 0.06315673142671585,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 78759
+    },
+    {
+      "epoch": 8.098087600246762,
+      "grad_norm": 0.04398050159215927,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 78762
+    },
+    {
+      "epoch": 8.098396051819865,
+      "grad_norm": 0.04201388731598854,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 78765
+    },
+    {
+      "epoch": 8.098704503392968,
+      "grad_norm": 0.11394036561250687,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 78768
+    },
+    {
+      "epoch": 8.09901295496607,
+      "grad_norm": 0.05638299137353897,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 78771
+    },
+    {
+      "epoch": 8.099321406539174,
+      "grad_norm": 0.04001401364803314,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 78774
+    },
+    {
+      "epoch": 8.099629858112277,
+      "grad_norm": 0.13935402035713196,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 78777
+    },
+    {
+      "epoch": 8.09993830968538,
+      "grad_norm": 0.0622687041759491,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 78780
+    },
+    {
+      "epoch": 8.100246761258482,
+      "grad_norm": 0.04048505052924156,
+      "learning_rate": 0.01,
+      "loss": 1.9869,
+      "step": 78783
+    },
+    {
+      "epoch": 8.100555212831585,
+      "grad_norm": 0.04119844362139702,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 78786
+    },
+    {
+      "epoch": 8.100863664404688,
+      "grad_norm": 0.03986690193414688,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 78789
+    },
+    {
+      "epoch": 8.101172115977791,
+      "grad_norm": 0.07594916224479675,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 78792
+    },
+    {
+      "epoch": 8.101480567550894,
+      "grad_norm": 0.0828123390674591,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 78795
+    },
+    {
+      "epoch": 8.101789019123997,
+      "grad_norm": 0.05890440568327904,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 78798
+    },
+    {
+      "epoch": 8.102097470697101,
+      "grad_norm": 0.09964809566736221,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 78801
+    },
+    {
+      "epoch": 8.102405922270204,
+      "grad_norm": 0.06926414370536804,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 78804
+    },
+    {
+      "epoch": 8.102714373843307,
+      "grad_norm": 0.0764726996421814,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 78807
+    },
+    {
+      "epoch": 8.10302282541641,
+      "grad_norm": 0.08960816264152527,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 78810
+    },
+    {
+      "epoch": 8.103331276989513,
+      "grad_norm": 0.05448700487613678,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 78813
+    },
+    {
+      "epoch": 8.103639728562616,
+      "grad_norm": 0.060038816183805466,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 78816
+    },
+    {
+      "epoch": 8.103948180135719,
+      "grad_norm": 0.06182781606912613,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 78819
+    },
+    {
+      "epoch": 8.104256631708822,
+      "grad_norm": 0.050959113985300064,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 78822
+    },
+    {
+      "epoch": 8.104565083281924,
+      "grad_norm": 0.042466700077056885,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 78825
+    },
+    {
+      "epoch": 8.104873534855027,
+      "grad_norm": 0.04001423716545105,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 78828
+    },
+    {
+      "epoch": 8.10518198642813,
+      "grad_norm": 0.04415873810648918,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 78831
+    },
+    {
+      "epoch": 8.105490438001233,
+      "grad_norm": 0.09015651047229767,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 78834
+    },
+    {
+      "epoch": 8.105798889574336,
+      "grad_norm": 0.1599554866552353,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 78837
+    },
+    {
+      "epoch": 8.10610734114744,
+      "grad_norm": 0.09218296408653259,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 78840
+    },
+    {
+      "epoch": 8.106415792720544,
+      "grad_norm": 0.07291611284017563,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 78843
+    },
+    {
+      "epoch": 8.106724244293646,
+      "grad_norm": 0.042586617171764374,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 78846
+    },
+    {
+      "epoch": 8.10703269586675,
+      "grad_norm": 0.03527746722102165,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 78849
+    },
+    {
+      "epoch": 8.107341147439852,
+      "grad_norm": 0.04984118789434433,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 78852
+    },
+    {
+      "epoch": 8.107649599012955,
+      "grad_norm": 0.08641780912876129,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 78855
+    },
+    {
+      "epoch": 8.107958050586058,
+      "grad_norm": 0.05808859318494797,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 78858
+    },
+    {
+      "epoch": 8.108266502159161,
+      "grad_norm": 0.06159026548266411,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 78861
+    },
+    {
+      "epoch": 8.108574953732264,
+      "grad_norm": 0.08137316256761551,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 78864
+    },
+    {
+      "epoch": 8.108883405305367,
+      "grad_norm": 0.06507211178541183,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 78867
+    },
+    {
+      "epoch": 8.10919185687847,
+      "grad_norm": 0.04363485053181648,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 78870
+    },
+    {
+      "epoch": 8.109500308451572,
+      "grad_norm": 0.1976436972618103,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 78873
+    },
+    {
+      "epoch": 8.109808760024675,
+      "grad_norm": 0.18802042305469513,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 78876
+    },
+    {
+      "epoch": 8.11011721159778,
+      "grad_norm": 0.11292927712202072,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 78879
+    },
+    {
+      "epoch": 8.110425663170883,
+      "grad_norm": 0.06154080107808113,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 78882
+    },
+    {
+      "epoch": 8.110734114743986,
+      "grad_norm": 0.04565787315368652,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 78885
+    },
+    {
+      "epoch": 8.111042566317089,
+      "grad_norm": 0.07542978227138519,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 78888
+    },
+    {
+      "epoch": 8.111351017890192,
+      "grad_norm": 0.11255155503749847,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 78891
+    },
+    {
+      "epoch": 8.111659469463294,
+      "grad_norm": 0.09924767911434174,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 78894
+    },
+    {
+      "epoch": 8.111967921036397,
+      "grad_norm": 0.06724300235509872,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 78897
+    },
+    {
+      "epoch": 8.1122763726095,
+      "grad_norm": 0.061787478625774384,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 78900
+    },
+    {
+      "epoch": 8.112584824182603,
+      "grad_norm": 0.06443925201892853,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 78903
+    },
+    {
+      "epoch": 8.112893275755706,
+      "grad_norm": 0.042934153228998184,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 78906
+    },
+    {
+      "epoch": 8.113201727328809,
+      "grad_norm": 0.04675191268324852,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 78909
+    },
+    {
+      "epoch": 8.113510178901912,
+      "grad_norm": 0.15414629876613617,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 78912
+    },
+    {
+      "epoch": 8.113818630475015,
+      "grad_norm": 0.06636694073677063,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 78915
+    },
+    {
+      "epoch": 8.11412708204812,
+      "grad_norm": 0.08664028346538544,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 78918
+    },
+    {
+      "epoch": 8.114435533621222,
+      "grad_norm": 0.04274681210517883,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 78921
+    },
+    {
+      "epoch": 8.114743985194325,
+      "grad_norm": 0.05404650419950485,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 78924
+    },
+    {
+      "epoch": 8.115052436767428,
+      "grad_norm": 0.04403231665492058,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 78927
+    },
+    {
+      "epoch": 8.11536088834053,
+      "grad_norm": 0.06800958514213562,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 78930
+    },
+    {
+      "epoch": 8.115669339913634,
+      "grad_norm": 0.06787417083978653,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 78933
+    },
+    {
+      "epoch": 8.115977791486737,
+      "grad_norm": 0.046553414314985275,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 78936
+    },
+    {
+      "epoch": 8.11628624305984,
+      "grad_norm": 0.051033198833465576,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 78939
+    },
+    {
+      "epoch": 8.116594694632942,
+      "grad_norm": 0.05013803020119667,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 78942
+    },
+    {
+      "epoch": 8.116903146206045,
+      "grad_norm": 0.06531897187232971,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 78945
+    },
+    {
+      "epoch": 8.117211597779148,
+      "grad_norm": 0.03636271879076958,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 78948
+    },
+    {
+      "epoch": 8.117520049352251,
+      "grad_norm": 0.04407816380262375,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 78951
+    },
+    {
+      "epoch": 8.117828500925354,
+      "grad_norm": 0.049220457673072815,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 78954
+    },
+    {
+      "epoch": 8.118136952498459,
+      "grad_norm": 0.061168961226940155,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 78957
+    },
+    {
+      "epoch": 8.118445404071561,
+      "grad_norm": 0.10130732506513596,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 78960
+    },
+    {
+      "epoch": 8.118753855644664,
+      "grad_norm": 0.1906558871269226,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 78963
+    },
+    {
+      "epoch": 8.119062307217767,
+      "grad_norm": 0.08722063153982162,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 78966
+    },
+    {
+      "epoch": 8.11937075879087,
+      "grad_norm": 0.06511518359184265,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 78969
+    },
+    {
+      "epoch": 8.119679210363973,
+      "grad_norm": 0.04923033341765404,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 78972
+    },
+    {
+      "epoch": 8.119987661937076,
+      "grad_norm": 0.048516470938920975,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 78975
+    },
+    {
+      "epoch": 8.120296113510179,
+      "grad_norm": 0.04488534480333328,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 78978
+    },
+    {
+      "epoch": 8.120604565083282,
+      "grad_norm": 0.04538138210773468,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 78981
+    },
+    {
+      "epoch": 8.120913016656385,
+      "grad_norm": 0.07497114688158035,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 78984
+    },
+    {
+      "epoch": 8.121221468229487,
+      "grad_norm": 0.10374471545219421,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 78987
+    },
+    {
+      "epoch": 8.12152991980259,
+      "grad_norm": 0.13686977326869965,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 78990
+    },
+    {
+      "epoch": 8.121838371375693,
+      "grad_norm": 0.06521309167146683,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 78993
+    },
+    {
+      "epoch": 8.122146822948798,
+      "grad_norm": 0.03203187882900238,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 78996
+    },
+    {
+      "epoch": 8.1224552745219,
+      "grad_norm": 0.034699197858572006,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 78999
+    },
+    {
+      "epoch": 8.122763726095004,
+      "grad_norm": 0.1084802970290184,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 79002
+    },
+    {
+      "epoch": 8.123072177668107,
+      "grad_norm": 0.07200446724891663,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 79005
+    },
+    {
+      "epoch": 8.12338062924121,
+      "grad_norm": 0.057980168610811234,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 79008
+    },
+    {
+      "epoch": 8.123689080814312,
+      "grad_norm": 0.03710618615150452,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 79011
+    },
+    {
+      "epoch": 8.123997532387415,
+      "grad_norm": 0.04603053256869316,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 79014
+    },
+    {
+      "epoch": 8.124305983960518,
+      "grad_norm": 0.088069386780262,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 79017
+    },
+    {
+      "epoch": 8.124614435533621,
+      "grad_norm": 0.04834933578968048,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 79020
+    },
+    {
+      "epoch": 8.124922887106724,
+      "grad_norm": 0.04151962324976921,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 79023
+    },
+    {
+      "epoch": 8.125231338679827,
+      "grad_norm": 0.051632124930620193,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 79026
+    },
+    {
+      "epoch": 8.12553979025293,
+      "grad_norm": 0.143802672624588,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 79029
+    },
+    {
+      "epoch": 8.125848241826032,
+      "grad_norm": 0.04302970692515373,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 79032
+    },
+    {
+      "epoch": 8.126156693399137,
+      "grad_norm": 0.1300458461046219,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 79035
+    },
+    {
+      "epoch": 8.12646514497224,
+      "grad_norm": 0.06043313816189766,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 79038
+    },
+    {
+      "epoch": 8.126773596545343,
+      "grad_norm": 0.03919702395796776,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 79041
+    },
+    {
+      "epoch": 8.127082048118446,
+      "grad_norm": 0.0503271222114563,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 79044
+    },
+    {
+      "epoch": 8.127390499691549,
+      "grad_norm": 0.035495515912771225,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 79047
+    },
+    {
+      "epoch": 8.127698951264652,
+      "grad_norm": 0.03414744511246681,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 79050
+    },
+    {
+      "epoch": 8.128007402837754,
+      "grad_norm": 0.08046834170818329,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 79053
+    },
+    {
+      "epoch": 8.128315854410857,
+      "grad_norm": 0.03690198436379433,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 79056
+    },
+    {
+      "epoch": 8.12862430598396,
+      "grad_norm": 0.06119079142808914,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 79059
+    },
+    {
+      "epoch": 8.128932757557063,
+      "grad_norm": 0.0938824862241745,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 79062
+    },
+    {
+      "epoch": 8.129241209130166,
+      "grad_norm": 0.09503504633903503,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 79065
+    },
+    {
+      "epoch": 8.129549660703269,
+      "grad_norm": 0.06319358944892883,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 79068
+    },
+    {
+      "epoch": 8.129858112276372,
+      "grad_norm": 0.08098098635673523,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 79071
+    },
+    {
+      "epoch": 8.130166563849476,
+      "grad_norm": 0.06519079208374023,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 79074
+    },
+    {
+      "epoch": 8.13047501542258,
+      "grad_norm": 0.13367730379104614,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 79077
+    },
+    {
+      "epoch": 8.130783466995682,
+      "grad_norm": 0.09786729514598846,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 79080
+    },
+    {
+      "epoch": 8.131091918568785,
+      "grad_norm": 0.11255167424678802,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 79083
+    },
+    {
+      "epoch": 8.131400370141888,
+      "grad_norm": 0.05492749437689781,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 79086
+    },
+    {
+      "epoch": 8.13170882171499,
+      "grad_norm": 0.03756438195705414,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 79089
+    },
+    {
+      "epoch": 8.132017273288094,
+      "grad_norm": 0.03793524205684662,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 79092
+    },
+    {
+      "epoch": 8.132325724861197,
+      "grad_norm": 0.03778066486120224,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 79095
+    },
+    {
+      "epoch": 8.1326341764343,
+      "grad_norm": 0.034888532012701035,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 79098
+    },
+    {
+      "epoch": 8.132942628007402,
+      "grad_norm": 0.0687558501958847,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 79101
+    },
+    {
+      "epoch": 8.133251079580505,
+      "grad_norm": 0.1305745244026184,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 79104
+    },
+    {
+      "epoch": 8.133559531153608,
+      "grad_norm": 0.06419795006513596,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 79107
+    },
+    {
+      "epoch": 8.133867982726711,
+      "grad_norm": 0.03223300725221634,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 79110
+    },
+    {
+      "epoch": 8.134176434299816,
+      "grad_norm": 0.03920907899737358,
+      "learning_rate": 0.01,
+      "loss": 1.9834,
+      "step": 79113
+    },
+    {
+      "epoch": 8.134484885872919,
+      "grad_norm": 0.05031749606132507,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 79116
+    },
+    {
+      "epoch": 8.134793337446022,
+      "grad_norm": 0.08466044068336487,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 79119
+    },
+    {
+      "epoch": 8.135101789019124,
+      "grad_norm": 0.06535778194665909,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 79122
+    },
+    {
+      "epoch": 8.135410240592227,
+      "grad_norm": 0.037378523498773575,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 79125
+    },
+    {
+      "epoch": 8.13571869216533,
+      "grad_norm": 0.11165131628513336,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 79128
+    },
+    {
+      "epoch": 8.136027143738433,
+      "grad_norm": 0.07167794555425644,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 79131
+    },
+    {
+      "epoch": 8.136335595311536,
+      "grad_norm": 0.10369572788476944,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 79134
+    },
+    {
+      "epoch": 8.136644046884639,
+      "grad_norm": 0.051352955400943756,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 79137
+    },
+    {
+      "epoch": 8.136952498457742,
+      "grad_norm": 0.11947130411863327,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 79140
+    },
+    {
+      "epoch": 8.137260950030845,
+      "grad_norm": 0.07348339259624481,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 79143
+    },
+    {
+      "epoch": 8.137569401603947,
+      "grad_norm": 0.145231232047081,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 79146
+    },
+    {
+      "epoch": 8.13787785317705,
+      "grad_norm": 0.053605154156684875,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 79149
+    },
+    {
+      "epoch": 8.138186304750155,
+      "grad_norm": 0.04041539132595062,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 79152
+    },
+    {
+      "epoch": 8.138494756323258,
+      "grad_norm": 0.06164342164993286,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 79155
+    },
+    {
+      "epoch": 8.13880320789636,
+      "grad_norm": 0.04916723445057869,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 79158
+    },
+    {
+      "epoch": 8.139111659469464,
+      "grad_norm": 0.04106106981635094,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 79161
+    },
+    {
+      "epoch": 8.139420111042567,
+      "grad_norm": 0.03548407554626465,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 79164
+    },
+    {
+      "epoch": 8.13972856261567,
+      "grad_norm": 0.05280381441116333,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 79167
+    },
+    {
+      "epoch": 8.140037014188772,
+      "grad_norm": 0.10448241233825684,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 79170
+    },
+    {
+      "epoch": 8.140345465761875,
+      "grad_norm": 0.05269154906272888,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 79173
+    },
+    {
+      "epoch": 8.140653917334978,
+      "grad_norm": 0.14911656081676483,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 79176
+    },
+    {
+      "epoch": 8.140962368908081,
+      "grad_norm": 0.051300134509801865,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 79179
+    },
+    {
+      "epoch": 8.141270820481184,
+      "grad_norm": 0.03888317197561264,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 79182
+    },
+    {
+      "epoch": 8.141579272054287,
+      "grad_norm": 0.04629917070269585,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 79185
+    },
+    {
+      "epoch": 8.14188772362739,
+      "grad_norm": 0.03899998590350151,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 79188
+    },
+    {
+      "epoch": 8.142196175200494,
+      "grad_norm": 0.04058917611837387,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 79191
+    },
+    {
+      "epoch": 8.142504626773597,
+      "grad_norm": 0.03686654940247536,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 79194
+    },
+    {
+      "epoch": 8.1428130783467,
+      "grad_norm": 0.048519935458898544,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 79197
+    },
+    {
+      "epoch": 8.143121529919803,
+      "grad_norm": 0.037117861211299896,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 79200
+    },
+    {
+      "epoch": 8.143429981492906,
+      "grad_norm": 0.10335171222686768,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 79203
+    },
+    {
+      "epoch": 8.143738433066009,
+      "grad_norm": 0.07381299138069153,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 79206
+    },
+    {
+      "epoch": 8.144046884639112,
+      "grad_norm": 0.06755456328392029,
+      "learning_rate": 0.01,
+      "loss": 1.9778,
+      "step": 79209
+    },
+    {
+      "epoch": 8.144355336212215,
+      "grad_norm": 0.05779864639043808,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 79212
+    },
+    {
+      "epoch": 8.144663787785317,
+      "grad_norm": 0.09358736872673035,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 79215
+    },
+    {
+      "epoch": 8.14497223935842,
+      "grad_norm": 0.04860676825046539,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 79218
+    },
+    {
+      "epoch": 8.145280690931523,
+      "grad_norm": 0.032770249992609024,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 79221
+    },
+    {
+      "epoch": 8.145589142504626,
+      "grad_norm": 0.04577682912349701,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 79224
+    },
+    {
+      "epoch": 8.145897594077729,
+      "grad_norm": 0.0953214019536972,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 79227
+    },
+    {
+      "epoch": 8.146206045650834,
+      "grad_norm": 0.07593677937984467,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 79230
+    },
+    {
+      "epoch": 8.146514497223937,
+      "grad_norm": 0.11113808304071426,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 79233
+    },
+    {
+      "epoch": 8.14682294879704,
+      "grad_norm": 0.13596338033676147,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 79236
+    },
+    {
+      "epoch": 8.147131400370142,
+      "grad_norm": 0.1248130351305008,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 79239
+    },
+    {
+      "epoch": 8.147439851943245,
+      "grad_norm": 0.09058087319135666,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 79242
+    },
+    {
+      "epoch": 8.147748303516348,
+      "grad_norm": 0.08766442537307739,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 79245
+    },
+    {
+      "epoch": 8.148056755089451,
+      "grad_norm": 0.04625086486339569,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 79248
+    },
+    {
+      "epoch": 8.148365206662554,
+      "grad_norm": 0.04851190373301506,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 79251
+    },
+    {
+      "epoch": 8.148673658235657,
+      "grad_norm": 0.0595279186964035,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 79254
+    },
+    {
+      "epoch": 8.14898210980876,
+      "grad_norm": 0.05664240941405296,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 79257
+    },
+    {
+      "epoch": 8.149290561381862,
+      "grad_norm": 0.06132050231099129,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 79260
+    },
+    {
+      "epoch": 8.149599012954965,
+      "grad_norm": 0.051271144300699234,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 79263
+    },
+    {
+      "epoch": 8.149907464528068,
+      "grad_norm": 0.07151178270578384,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 79266
+    },
+    {
+      "epoch": 8.150215916101173,
+      "grad_norm": 0.051999252289533615,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 79269
+    },
+    {
+      "epoch": 8.150524367674276,
+      "grad_norm": 0.05515766888856888,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 79272
+    },
+    {
+      "epoch": 8.150832819247379,
+      "grad_norm": 0.07011020928621292,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 79275
+    },
+    {
+      "epoch": 8.151141270820482,
+      "grad_norm": 0.049128126353025436,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 79278
+    },
+    {
+      "epoch": 8.151449722393584,
+      "grad_norm": 0.04374023899435997,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 79281
+    },
+    {
+      "epoch": 8.151758173966687,
+      "grad_norm": 0.10206493735313416,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 79284
+    },
+    {
+      "epoch": 8.15206662553979,
+      "grad_norm": 0.06682003289461136,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 79287
+    },
+    {
+      "epoch": 8.152375077112893,
+      "grad_norm": 0.08957073092460632,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 79290
+    },
+    {
+      "epoch": 8.152683528685996,
+      "grad_norm": 0.08002035319805145,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 79293
+    },
+    {
+      "epoch": 8.152991980259099,
+      "grad_norm": 0.09390459209680557,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 79296
+    },
+    {
+      "epoch": 8.153300431832202,
+      "grad_norm": 0.06374812871217728,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 79299
+    },
+    {
+      "epoch": 8.153608883405305,
+      "grad_norm": 0.07658891379833221,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 79302
+    },
+    {
+      "epoch": 8.153917334978408,
+      "grad_norm": 0.11377871036529541,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 79305
+    },
+    {
+      "epoch": 8.154225786551512,
+      "grad_norm": 0.049400169402360916,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 79308
+    },
+    {
+      "epoch": 8.154534238124615,
+      "grad_norm": 0.08506350219249725,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 79311
+    },
+    {
+      "epoch": 8.154842689697718,
+      "grad_norm": 0.06659626960754395,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 79314
+    },
+    {
+      "epoch": 8.15515114127082,
+      "grad_norm": 0.037887092679739,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 79317
+    },
+    {
+      "epoch": 8.155459592843924,
+      "grad_norm": 0.05094122514128685,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 79320
+    },
+    {
+      "epoch": 8.155768044417027,
+      "grad_norm": 0.05743343010544777,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 79323
+    },
+    {
+      "epoch": 8.15607649599013,
+      "grad_norm": 0.07642214000225067,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 79326
+    },
+    {
+      "epoch": 8.156384947563232,
+      "grad_norm": 0.07716453075408936,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 79329
+    },
+    {
+      "epoch": 8.156693399136335,
+      "grad_norm": 0.04788043722510338,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 79332
+    },
+    {
+      "epoch": 8.157001850709438,
+      "grad_norm": 0.037050921469926834,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 79335
+    },
+    {
+      "epoch": 8.157310302282541,
+      "grad_norm": 0.062088605016469955,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 79338
+    },
+    {
+      "epoch": 8.157618753855644,
+      "grad_norm": 0.042694903910160065,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 79341
+    },
+    {
+      "epoch": 8.157927205428747,
+      "grad_norm": 0.03817545622587204,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 79344
+    },
+    {
+      "epoch": 8.158235657001851,
+      "grad_norm": 0.04757656529545784,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 79347
+    },
+    {
+      "epoch": 8.158544108574954,
+      "grad_norm": 0.07190608233213425,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 79350
+    },
+    {
+      "epoch": 8.158852560148057,
+      "grad_norm": 0.12211509048938751,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 79353
+    },
+    {
+      "epoch": 8.15916101172116,
+      "grad_norm": 0.05178214982151985,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 79356
+    },
+    {
+      "epoch": 8.159469463294263,
+      "grad_norm": 0.04701949656009674,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 79359
+    },
+    {
+      "epoch": 8.159777914867366,
+      "grad_norm": 0.10068555921316147,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 79362
+    },
+    {
+      "epoch": 8.160086366440469,
+      "grad_norm": 0.1051318496465683,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 79365
+    },
+    {
+      "epoch": 8.160394818013572,
+      "grad_norm": 0.07793916016817093,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 79368
+    },
+    {
+      "epoch": 8.160703269586675,
+      "grad_norm": 0.061303503811359406,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 79371
+    },
+    {
+      "epoch": 8.161011721159777,
+      "grad_norm": 0.05692371726036072,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 79374
+    },
+    {
+      "epoch": 8.16132017273288,
+      "grad_norm": 0.061426080763339996,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 79377
+    },
+    {
+      "epoch": 8.161628624305983,
+      "grad_norm": 0.04750361666083336,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 79380
+    },
+    {
+      "epoch": 8.161937075879086,
+      "grad_norm": 0.03101014718413353,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 79383
+    },
+    {
+      "epoch": 8.16224552745219,
+      "grad_norm": 0.043352603912353516,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 79386
+    },
+    {
+      "epoch": 8.162553979025294,
+      "grad_norm": 0.11773629486560822,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 79389
+    },
+    {
+      "epoch": 8.162862430598397,
+      "grad_norm": 0.08607412129640579,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 79392
+    },
+    {
+      "epoch": 8.1631708821715,
+      "grad_norm": 0.06435645371675491,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 79395
+    },
+    {
+      "epoch": 8.163479333744602,
+      "grad_norm": 0.035051826387643814,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 79398
+    },
+    {
+      "epoch": 8.163787785317705,
+      "grad_norm": 0.0367707759141922,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 79401
+    },
+    {
+      "epoch": 8.164096236890808,
+      "grad_norm": 0.04549092799425125,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 79404
+    },
+    {
+      "epoch": 8.164404688463911,
+      "grad_norm": 0.050185445696115494,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 79407
+    },
+    {
+      "epoch": 8.164713140037014,
+      "grad_norm": 0.053017307072877884,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 79410
+    },
+    {
+      "epoch": 8.165021591610117,
+      "grad_norm": 0.03889346867799759,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 79413
+    },
+    {
+      "epoch": 8.16533004318322,
+      "grad_norm": 0.09231189638376236,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 79416
+    },
+    {
+      "epoch": 8.165638494756323,
+      "grad_norm": 0.06801720708608627,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 79419
+    },
+    {
+      "epoch": 8.165946946329425,
+      "grad_norm": 0.08188890665769577,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 79422
+    },
+    {
+      "epoch": 8.16625539790253,
+      "grad_norm": 0.07723134011030197,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 79425
+    },
+    {
+      "epoch": 8.166563849475633,
+      "grad_norm": 0.05977725237607956,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 79428
+    },
+    {
+      "epoch": 8.166872301048736,
+      "grad_norm": 0.04500715434551239,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 79431
+    },
+    {
+      "epoch": 8.167180752621839,
+      "grad_norm": 0.044927868992090225,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 79434
+    },
+    {
+      "epoch": 8.167489204194942,
+      "grad_norm": 0.05496992543339729,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 79437
+    },
+    {
+      "epoch": 8.167797655768045,
+      "grad_norm": 0.0723772794008255,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 79440
+    },
+    {
+      "epoch": 8.168106107341147,
+      "grad_norm": 0.0771140456199646,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 79443
+    },
+    {
+      "epoch": 8.16841455891425,
+      "grad_norm": 0.041013602167367935,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 79446
+    },
+    {
+      "epoch": 8.168723010487353,
+      "grad_norm": 0.035580825060606,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 79449
+    },
+    {
+      "epoch": 8.169031462060456,
+      "grad_norm": 0.06367741525173187,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 79452
+    },
+    {
+      "epoch": 8.169339913633559,
+      "grad_norm": 0.08021467179059982,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 79455
+    },
+    {
+      "epoch": 8.169648365206662,
+      "grad_norm": 0.08360808342695236,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 79458
+    },
+    {
+      "epoch": 8.169956816779765,
+      "grad_norm": 0.0339963473379612,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 79461
+    },
+    {
+      "epoch": 8.17026526835287,
+      "grad_norm": 0.07280755043029785,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 79464
+    },
+    {
+      "epoch": 8.170573719925972,
+      "grad_norm": 0.05875898897647858,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 79467
+    },
+    {
+      "epoch": 8.170882171499075,
+      "grad_norm": 0.05665907263755798,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 79470
+    },
+    {
+      "epoch": 8.171190623072178,
+      "grad_norm": 0.06836891919374466,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 79473
+    },
+    {
+      "epoch": 8.171499074645281,
+      "grad_norm": 0.05284962058067322,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 79476
+    },
+    {
+      "epoch": 8.171807526218384,
+      "grad_norm": 0.05243814364075661,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 79479
+    },
+    {
+      "epoch": 8.172115977791487,
+      "grad_norm": 0.11004475504159927,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 79482
+    },
+    {
+      "epoch": 8.17242442936459,
+      "grad_norm": 0.0900677889585495,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 79485
+    },
+    {
+      "epoch": 8.172732880937692,
+      "grad_norm": 0.07300006598234177,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 79488
+    },
+    {
+      "epoch": 8.173041332510795,
+      "grad_norm": 0.04999767243862152,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 79491
+    },
+    {
+      "epoch": 8.173349784083898,
+      "grad_norm": 0.06433633714914322,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 79494
+    },
+    {
+      "epoch": 8.173658235657001,
+      "grad_norm": 0.08774617314338684,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 79497
+    },
+    {
+      "epoch": 8.173966687230104,
+      "grad_norm": 0.06742536276578903,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 79500
+    },
+    {
+      "epoch": 8.174275138803209,
+      "grad_norm": 0.0422503799200058,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 79503
+    },
+    {
+      "epoch": 8.174583590376312,
+      "grad_norm": 0.15565431118011475,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 79506
+    },
+    {
+      "epoch": 8.174892041949414,
+      "grad_norm": 0.12713783979415894,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 79509
+    },
+    {
+      "epoch": 8.175200493522517,
+      "grad_norm": 0.08080834895372391,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 79512
+    },
+    {
+      "epoch": 8.17550894509562,
+      "grad_norm": 0.0670383870601654,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 79515
+    },
+    {
+      "epoch": 8.175817396668723,
+      "grad_norm": 0.04987994581460953,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 79518
+    },
+    {
+      "epoch": 8.176125848241826,
+      "grad_norm": 0.06311977654695511,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 79521
+    },
+    {
+      "epoch": 8.176434299814929,
+      "grad_norm": 0.05733686313033104,
+      "learning_rate": 0.01,
+      "loss": 1.9776,
+      "step": 79524
+    },
+    {
+      "epoch": 8.176742751388032,
+      "grad_norm": 0.03165985271334648,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 79527
+    },
+    {
+      "epoch": 8.177051202961135,
+      "grad_norm": 0.1158774122595787,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 79530
+    },
+    {
+      "epoch": 8.177359654534238,
+      "grad_norm": 0.06680413335561752,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 79533
+    },
+    {
+      "epoch": 8.17766810610734,
+      "grad_norm": 0.08957581222057343,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 79536
+    },
+    {
+      "epoch": 8.177976557680445,
+      "grad_norm": 0.04930703341960907,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 79539
+    },
+    {
+      "epoch": 8.178285009253548,
+      "grad_norm": 0.040650609880685806,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 79542
+    },
+    {
+      "epoch": 8.17859346082665,
+      "grad_norm": 0.04556789621710777,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 79545
+    },
+    {
+      "epoch": 8.178901912399754,
+      "grad_norm": 0.12077522277832031,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 79548
+    },
+    {
+      "epoch": 8.179210363972857,
+      "grad_norm": 0.041626036167144775,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 79551
+    },
+    {
+      "epoch": 8.17951881554596,
+      "grad_norm": 0.06800248473882675,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 79554
+    },
+    {
+      "epoch": 8.179827267119062,
+      "grad_norm": 0.07558157294988632,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 79557
+    },
+    {
+      "epoch": 8.180135718692165,
+      "grad_norm": 0.0845264345407486,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 79560
+    },
+    {
+      "epoch": 8.180444170265268,
+      "grad_norm": 0.07842035591602325,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 79563
+    },
+    {
+      "epoch": 8.180752621838371,
+      "grad_norm": 0.07811528444290161,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 79566
+    },
+    {
+      "epoch": 8.181061073411474,
+      "grad_norm": 0.07147359102964401,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 79569
+    },
+    {
+      "epoch": 8.181369524984577,
+      "grad_norm": 0.08217006921768188,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 79572
+    },
+    {
+      "epoch": 8.18167797655768,
+      "grad_norm": 0.06407132744789124,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 79575
+    },
+    {
+      "epoch": 8.181986428130784,
+      "grad_norm": 0.07353561371564865,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 79578
+    },
+    {
+      "epoch": 8.182294879703887,
+      "grad_norm": 0.11215024441480637,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 79581
+    },
+    {
+      "epoch": 8.18260333127699,
+      "grad_norm": 0.042454659938812256,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 79584
+    },
+    {
+      "epoch": 8.182911782850093,
+      "grad_norm": 0.06562569737434387,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 79587
+    },
+    {
+      "epoch": 8.183220234423196,
+      "grad_norm": 0.05847795680165291,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 79590
+    },
+    {
+      "epoch": 8.183528685996299,
+      "grad_norm": 0.03575105592608452,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 79593
+    },
+    {
+      "epoch": 8.183837137569402,
+      "grad_norm": 0.04204864054918289,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 79596
+    },
+    {
+      "epoch": 8.184145589142505,
+      "grad_norm": 0.13108733296394348,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 79599
+    },
+    {
+      "epoch": 8.184454040715607,
+      "grad_norm": 0.05569970980286598,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 79602
+    },
+    {
+      "epoch": 8.18476249228871,
+      "grad_norm": 0.07396795600652695,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 79605
+    },
+    {
+      "epoch": 8.185070943861813,
+      "grad_norm": 0.05299641191959381,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 79608
+    },
+    {
+      "epoch": 8.185379395434916,
+      "grad_norm": 0.04353434219956398,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 79611
+    },
+    {
+      "epoch": 8.185687847008019,
+      "grad_norm": 0.032626885920763016,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 79614
+    },
+    {
+      "epoch": 8.185996298581124,
+      "grad_norm": 0.06502965092658997,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 79617
+    },
+    {
+      "epoch": 8.186304750154227,
+      "grad_norm": 0.15308474004268646,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 79620
+    },
+    {
+      "epoch": 8.18661320172733,
+      "grad_norm": 0.13402137160301208,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 79623
+    },
+    {
+      "epoch": 8.186921653300432,
+      "grad_norm": 0.07767334580421448,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 79626
+    },
+    {
+      "epoch": 8.187230104873535,
+      "grad_norm": 0.0863209217786789,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 79629
+    },
+    {
+      "epoch": 8.187538556446638,
+      "grad_norm": 0.07084184139966965,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 79632
+    },
+    {
+      "epoch": 8.187847008019741,
+      "grad_norm": 0.06534894555807114,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 79635
+    },
+    {
+      "epoch": 8.188155459592844,
+      "grad_norm": 0.03883112967014313,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 79638
+    },
+    {
+      "epoch": 8.188463911165947,
+      "grad_norm": 0.08439522236585617,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 79641
+    },
+    {
+      "epoch": 8.18877236273905,
+      "grad_norm": 0.07321714609861374,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 79644
+    },
+    {
+      "epoch": 8.189080814312153,
+      "grad_norm": 0.04626638814806938,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 79647
+    },
+    {
+      "epoch": 8.189389265885255,
+      "grad_norm": 0.10366689413785934,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 79650
+    },
+    {
+      "epoch": 8.189697717458358,
+      "grad_norm": 0.08205170929431915,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 79653
+    },
+    {
+      "epoch": 8.190006169031463,
+      "grad_norm": 0.06450674682855606,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 79656
+    },
+    {
+      "epoch": 8.190314620604566,
+      "grad_norm": 0.05880254507064819,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 79659
+    },
+    {
+      "epoch": 8.190623072177669,
+      "grad_norm": 0.07828585803508759,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 79662
+    },
+    {
+      "epoch": 8.190931523750772,
+      "grad_norm": 0.06465885043144226,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 79665
+    },
+    {
+      "epoch": 8.191239975323874,
+      "grad_norm": 0.09105885773897171,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 79668
+    },
+    {
+      "epoch": 8.191548426896977,
+      "grad_norm": 0.0810224637389183,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 79671
+    },
+    {
+      "epoch": 8.19185687847008,
+      "grad_norm": 0.04941980540752411,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 79674
+    },
+    {
+      "epoch": 8.192165330043183,
+      "grad_norm": 0.06221986562013626,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 79677
+    },
+    {
+      "epoch": 8.192473781616286,
+      "grad_norm": 0.0800907164812088,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 79680
+    },
+    {
+      "epoch": 8.192782233189389,
+      "grad_norm": 0.06656412035226822,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 79683
+    },
+    {
+      "epoch": 8.193090684762492,
+      "grad_norm": 0.05629335716366768,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 79686
+    },
+    {
+      "epoch": 8.193399136335595,
+      "grad_norm": 0.1350521743297577,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 79689
+    },
+    {
+      "epoch": 8.193707587908698,
+      "grad_norm": 0.05548480898141861,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 79692
+    },
+    {
+      "epoch": 8.194016039481802,
+      "grad_norm": 0.038812119513750076,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 79695
+    },
+    {
+      "epoch": 8.194324491054905,
+      "grad_norm": 0.09969807416200638,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 79698
+    },
+    {
+      "epoch": 8.194632942628008,
+      "grad_norm": 0.10124362260103226,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 79701
+    },
+    {
+      "epoch": 8.194941394201111,
+      "grad_norm": 0.048608239740133286,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 79704
+    },
+    {
+      "epoch": 8.195249845774214,
+      "grad_norm": 0.03750385716557503,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 79707
+    },
+    {
+      "epoch": 8.195558297347317,
+      "grad_norm": 0.07153517007827759,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 79710
+    },
+    {
+      "epoch": 8.19586674892042,
+      "grad_norm": 0.07376372069120407,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 79713
+    },
+    {
+      "epoch": 8.196175200493522,
+      "grad_norm": 0.08296847343444824,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 79716
+    },
+    {
+      "epoch": 8.196483652066625,
+      "grad_norm": 0.12598127126693726,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 79719
+    },
+    {
+      "epoch": 8.196792103639728,
+      "grad_norm": 0.06051617115736008,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 79722
+    },
+    {
+      "epoch": 8.197100555212831,
+      "grad_norm": 0.05080085247755051,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 79725
+    },
+    {
+      "epoch": 8.197409006785934,
+      "grad_norm": 0.04279420152306557,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 79728
+    },
+    {
+      "epoch": 8.197717458359037,
+      "grad_norm": 0.03769794479012489,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 79731
+    },
+    {
+      "epoch": 8.198025909932142,
+      "grad_norm": 0.03623786196112633,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 79734
+    },
+    {
+      "epoch": 8.198334361505244,
+      "grad_norm": 0.059139419347047806,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 79737
+    },
+    {
+      "epoch": 8.198642813078347,
+      "grad_norm": 0.051768649369478226,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 79740
+    },
+    {
+      "epoch": 8.19895126465145,
+      "grad_norm": 0.054058823734521866,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 79743
+    },
+    {
+      "epoch": 8.199259716224553,
+      "grad_norm": 0.06856348365545273,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 79746
+    },
+    {
+      "epoch": 8.199568167797656,
+      "grad_norm": 0.0483131967484951,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 79749
+    },
+    {
+      "epoch": 8.199876619370759,
+      "grad_norm": 0.0742148756980896,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 79752
+    },
+    {
+      "epoch": 8.200185070943862,
+      "grad_norm": 0.05363794043660164,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 79755
+    },
+    {
+      "epoch": 8.200493522516965,
+      "grad_norm": 0.07250255346298218,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 79758
+    },
+    {
+      "epoch": 8.200801974090068,
+      "grad_norm": 0.10672403872013092,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 79761
+    },
+    {
+      "epoch": 8.20111042566317,
+      "grad_norm": 0.08459288626909256,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 79764
+    },
+    {
+      "epoch": 8.201418877236273,
+      "grad_norm": 0.06241921707987785,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 79767
+    },
+    {
+      "epoch": 8.201727328809376,
+      "grad_norm": 0.047745924443006516,
+      "learning_rate": 0.01,
+      "loss": 1.9878,
+      "step": 79770
+    },
+    {
+      "epoch": 8.20203578038248,
+      "grad_norm": 0.11685614287853241,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 79773
+    },
+    {
+      "epoch": 8.202344231955584,
+      "grad_norm": 0.03775307908654213,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 79776
+    },
+    {
+      "epoch": 8.202652683528687,
+      "grad_norm": 0.04868243262171745,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 79779
+    },
+    {
+      "epoch": 8.20296113510179,
+      "grad_norm": 0.055632028728723526,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 79782
+    },
+    {
+      "epoch": 8.203269586674892,
+      "grad_norm": 0.05138279125094414,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 79785
+    },
+    {
+      "epoch": 8.203578038247995,
+      "grad_norm": 0.032587356865406036,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 79788
+    },
+    {
+      "epoch": 8.203886489821098,
+      "grad_norm": 0.08916875720024109,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 79791
+    },
+    {
+      "epoch": 8.204194941394201,
+      "grad_norm": 0.08575033396482468,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 79794
+    },
+    {
+      "epoch": 8.204503392967304,
+      "grad_norm": 0.06784272938966751,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 79797
+    },
+    {
+      "epoch": 8.204811844540407,
+      "grad_norm": 0.10347246378660202,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 79800
+    },
+    {
+      "epoch": 8.20512029611351,
+      "grad_norm": 0.04444975405931473,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 79803
+    },
+    {
+      "epoch": 8.205428747686613,
+      "grad_norm": 0.10432004928588867,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 79806
+    },
+    {
+      "epoch": 8.205737199259715,
+      "grad_norm": 0.1217341423034668,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 79809
+    },
+    {
+      "epoch": 8.20604565083282,
+      "grad_norm": 0.09438272565603256,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 79812
+    },
+    {
+      "epoch": 8.206354102405923,
+      "grad_norm": 0.08129420131444931,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 79815
+    },
+    {
+      "epoch": 8.206662553979026,
+      "grad_norm": 0.09659760445356369,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 79818
+    },
+    {
+      "epoch": 8.206971005552129,
+      "grad_norm": 0.06486459076404572,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 79821
+    },
+    {
+      "epoch": 8.207279457125232,
+      "grad_norm": 0.057728540152311325,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 79824
+    },
+    {
+      "epoch": 8.207587908698335,
+      "grad_norm": 0.08968479186296463,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 79827
+    },
+    {
+      "epoch": 8.207896360271437,
+      "grad_norm": 0.05536945164203644,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 79830
+    },
+    {
+      "epoch": 8.20820481184454,
+      "grad_norm": 0.03714819252490997,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 79833
+    },
+    {
+      "epoch": 8.208513263417643,
+      "grad_norm": 0.04826650768518448,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 79836
+    },
+    {
+      "epoch": 8.208821714990746,
+      "grad_norm": 0.10562440007925034,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 79839
+    },
+    {
+      "epoch": 8.209130166563849,
+      "grad_norm": 0.12888909876346588,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 79842
+    },
+    {
+      "epoch": 8.209438618136952,
+      "grad_norm": 0.04846480116248131,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 79845
+    },
+    {
+      "epoch": 8.209747069710055,
+      "grad_norm": 0.041596248745918274,
+      "learning_rate": 0.01,
+      "loss": 1.9798,
+      "step": 79848
+    },
+    {
+      "epoch": 8.21005552128316,
+      "grad_norm": 0.05451950803399086,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 79851
+    },
+    {
+      "epoch": 8.210363972856262,
+      "grad_norm": 0.046558499336242676,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 79854
+    },
+    {
+      "epoch": 8.210672424429365,
+      "grad_norm": 0.06177009269595146,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 79857
+    },
+    {
+      "epoch": 8.210980876002468,
+      "grad_norm": 0.09283863753080368,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 79860
+    },
+    {
+      "epoch": 8.211289327575571,
+      "grad_norm": 0.05722719803452492,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 79863
+    },
+    {
+      "epoch": 8.211597779148674,
+      "grad_norm": 0.08050020784139633,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 79866
+    },
+    {
+      "epoch": 8.211906230721777,
+      "grad_norm": 0.0464584156870842,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 79869
+    },
+    {
+      "epoch": 8.21221468229488,
+      "grad_norm": 0.03943070396780968,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 79872
+    },
+    {
+      "epoch": 8.212523133867982,
+      "grad_norm": 0.1355932652950287,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 79875
+    },
+    {
+      "epoch": 8.212831585441085,
+      "grad_norm": 0.060039740055799484,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 79878
+    },
+    {
+      "epoch": 8.213140037014188,
+      "grad_norm": 0.05428614467382431,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 79881
+    },
+    {
+      "epoch": 8.213448488587291,
+      "grad_norm": 0.04141784459352493,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 79884
+    },
+    {
+      "epoch": 8.213756940160394,
+      "grad_norm": 0.044172655791044235,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 79887
+    },
+    {
+      "epoch": 8.214065391733499,
+      "grad_norm": 0.10583905875682831,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 79890
+    },
+    {
+      "epoch": 8.214373843306602,
+      "grad_norm": 0.12462706863880157,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 79893
+    },
+    {
+      "epoch": 8.214682294879704,
+      "grad_norm": 0.14810222387313843,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 79896
+    },
+    {
+      "epoch": 8.214990746452807,
+      "grad_norm": 0.07370836287736893,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 79899
+    },
+    {
+      "epoch": 8.21529919802591,
+      "grad_norm": 0.047002315521240234,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 79902
+    },
+    {
+      "epoch": 8.215607649599013,
+      "grad_norm": 0.03757517784833908,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 79905
+    },
+    {
+      "epoch": 8.215916101172116,
+      "grad_norm": 0.03486863523721695,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 79908
+    },
+    {
+      "epoch": 8.216224552745219,
+      "grad_norm": 0.045810382813215256,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 79911
+    },
+    {
+      "epoch": 8.216533004318322,
+      "grad_norm": 0.03867745399475098,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 79914
+    },
+    {
+      "epoch": 8.216841455891425,
+      "grad_norm": 0.04912863299250603,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 79917
+    },
+    {
+      "epoch": 8.217149907464528,
+      "grad_norm": 0.04494732245802879,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 79920
+    },
+    {
+      "epoch": 8.21745835903763,
+      "grad_norm": 0.042188648134469986,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 79923
+    },
+    {
+      "epoch": 8.217766810610733,
+      "grad_norm": 0.04156626760959625,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 79926
+    },
+    {
+      "epoch": 8.218075262183838,
+      "grad_norm": 0.08161705732345581,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 79929
+    },
+    {
+      "epoch": 8.21838371375694,
+      "grad_norm": 0.0440400093793869,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 79932
+    },
+    {
+      "epoch": 8.218692165330044,
+      "grad_norm": 0.04624645411968231,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 79935
+    },
+    {
+      "epoch": 8.219000616903147,
+      "grad_norm": 0.05811585113406181,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 79938
+    },
+    {
+      "epoch": 8.21930906847625,
+      "grad_norm": 0.11906293034553528,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 79941
+    },
+    {
+      "epoch": 8.219617520049352,
+      "grad_norm": 0.13277171552181244,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 79944
+    },
+    {
+      "epoch": 8.219925971622455,
+      "grad_norm": 0.04586183652281761,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 79947
+    },
+    {
+      "epoch": 8.220234423195558,
+      "grad_norm": 0.03742174431681633,
+      "learning_rate": 0.01,
+      "loss": 1.9767,
+      "step": 79950
+    },
+    {
+      "epoch": 8.220542874768661,
+      "grad_norm": 0.05153629928827286,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 79953
+    },
+    {
+      "epoch": 8.220851326341764,
+      "grad_norm": 0.07394517213106155,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 79956
+    },
+    {
+      "epoch": 8.221159777914867,
+      "grad_norm": 0.05437902361154556,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 79959
+    },
+    {
+      "epoch": 8.22146822948797,
+      "grad_norm": 0.0376947782933712,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 79962
+    },
+    {
+      "epoch": 8.221776681061073,
+      "grad_norm": 0.06275657564401627,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 79965
+    },
+    {
+      "epoch": 8.222085132634177,
+      "grad_norm": 0.04809385538101196,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 79968
+    },
+    {
+      "epoch": 8.22239358420728,
+      "grad_norm": 0.04617860168218613,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 79971
+    },
+    {
+      "epoch": 8.222702035780383,
+      "grad_norm": 0.040207915008068085,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 79974
+    },
+    {
+      "epoch": 8.223010487353486,
+      "grad_norm": 0.047990135848522186,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 79977
+    },
+    {
+      "epoch": 8.223318938926589,
+      "grad_norm": 0.12925958633422852,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 79980
+    },
+    {
+      "epoch": 8.223627390499692,
+      "grad_norm": 0.0795065239071846,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 79983
+    },
+    {
+      "epoch": 8.223935842072795,
+      "grad_norm": 0.0691371038556099,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 79986
+    },
+    {
+      "epoch": 8.224244293645897,
+      "grad_norm": 0.05801672115921974,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 79989
+    },
+    {
+      "epoch": 8.224552745219,
+      "grad_norm": 0.03973524644970894,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 79992
+    },
+    {
+      "epoch": 8.224861196792103,
+      "grad_norm": 0.03368094190955162,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 79995
+    },
+    {
+      "epoch": 8.225169648365206,
+      "grad_norm": 0.03496599942445755,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 79998
+    },
+    {
+      "epoch": 8.225478099938309,
+      "grad_norm": 0.04070889577269554,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 80001
+    },
+    {
+      "epoch": 8.225786551511412,
+      "grad_norm": 0.05320926755666733,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 80004
+    },
+    {
+      "epoch": 8.226095003084517,
+      "grad_norm": 0.10373469442129135,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 80007
+    },
+    {
+      "epoch": 8.22640345465762,
+      "grad_norm": 0.12059690803289413,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 80010
+    },
+    {
+      "epoch": 8.226711906230722,
+      "grad_norm": 0.1084703654050827,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 80013
+    },
+    {
+      "epoch": 8.227020357803825,
+      "grad_norm": 0.04436837136745453,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 80016
+    },
+    {
+      "epoch": 8.227328809376928,
+      "grad_norm": 0.03714198246598244,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 80019
+    },
+    {
+      "epoch": 8.227637260950031,
+      "grad_norm": 0.03939532861113548,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 80022
+    },
+    {
+      "epoch": 8.227945712523134,
+      "grad_norm": 0.05460543558001518,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 80025
+    },
+    {
+      "epoch": 8.228254164096237,
+      "grad_norm": 0.10572293400764465,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 80028
+    },
+    {
+      "epoch": 8.22856261566934,
+      "grad_norm": 0.052371978759765625,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 80031
+    },
+    {
+      "epoch": 8.228871067242443,
+      "grad_norm": 0.03213518485426903,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 80034
+    },
+    {
+      "epoch": 8.229179518815545,
+      "grad_norm": 0.10267605632543564,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 80037
+    },
+    {
+      "epoch": 8.229487970388648,
+      "grad_norm": 0.11686369776725769,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 80040
+    },
+    {
+      "epoch": 8.229796421961751,
+      "grad_norm": 0.07059639692306519,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 80043
+    },
+    {
+      "epoch": 8.230104873534856,
+      "grad_norm": 0.048195354640483856,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 80046
+    },
+    {
+      "epoch": 8.230413325107959,
+      "grad_norm": 0.04270948842167854,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 80049
+    },
+    {
+      "epoch": 8.230721776681062,
+      "grad_norm": 0.04734241962432861,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 80052
+    },
+    {
+      "epoch": 8.231030228254165,
+      "grad_norm": 0.05041227117180824,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 80055
+    },
+    {
+      "epoch": 8.231338679827267,
+      "grad_norm": 0.060459088534116745,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 80058
+    },
+    {
+      "epoch": 8.23164713140037,
+      "grad_norm": 0.1127137765288353,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 80061
+    },
+    {
+      "epoch": 8.231955582973473,
+      "grad_norm": 0.11185494065284729,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 80064
+    },
+    {
+      "epoch": 8.232264034546576,
+      "grad_norm": 0.09597989171743393,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 80067
+    },
+    {
+      "epoch": 8.232572486119679,
+      "grad_norm": 0.04449654370546341,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 80070
+    },
+    {
+      "epoch": 8.232880937692782,
+      "grad_norm": 0.12597239017486572,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 80073
+    },
+    {
+      "epoch": 8.233189389265885,
+      "grad_norm": 0.045494552701711655,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 80076
+    },
+    {
+      "epoch": 8.233497840838988,
+      "grad_norm": 0.04805656895041466,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 80079
+    },
+    {
+      "epoch": 8.23380629241209,
+      "grad_norm": 0.06371653079986572,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 80082
+    },
+    {
+      "epoch": 8.234114743985195,
+      "grad_norm": 0.10116062313318253,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 80085
+    },
+    {
+      "epoch": 8.234423195558298,
+      "grad_norm": 0.08505774289369583,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 80088
+    },
+    {
+      "epoch": 8.234731647131401,
+      "grad_norm": 0.04425555467605591,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 80091
+    },
+    {
+      "epoch": 8.235040098704504,
+      "grad_norm": 0.06315706670284271,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 80094
+    },
+    {
+      "epoch": 8.235348550277607,
+      "grad_norm": 0.05656632408499718,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 80097
+    },
+    {
+      "epoch": 8.23565700185071,
+      "grad_norm": 0.10018274188041687,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 80100
+    },
+    {
+      "epoch": 8.235965453423812,
+      "grad_norm": 0.04145848751068115,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 80103
+    },
+    {
+      "epoch": 8.236273904996915,
+      "grad_norm": 0.0376301109790802,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 80106
+    },
+    {
+      "epoch": 8.236582356570018,
+      "grad_norm": 0.08260329812765121,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 80109
+    },
+    {
+      "epoch": 8.236890808143121,
+      "grad_norm": 0.03497771918773651,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 80112
+    },
+    {
+      "epoch": 8.237199259716224,
+      "grad_norm": 0.07940883189439774,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 80115
+    },
+    {
+      "epoch": 8.237507711289327,
+      "grad_norm": 0.06283021718263626,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 80118
+    },
+    {
+      "epoch": 8.23781616286243,
+      "grad_norm": 0.09119091182947159,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 80121
+    },
+    {
+      "epoch": 8.238124614435534,
+      "grad_norm": 0.11459033936262131,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 80124
+    },
+    {
+      "epoch": 8.238433066008637,
+      "grad_norm": 0.1344277411699295,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 80127
+    },
+    {
+      "epoch": 8.23874151758174,
+      "grad_norm": 0.08821120858192444,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 80130
+    },
+    {
+      "epoch": 8.239049969154843,
+      "grad_norm": 0.04385138675570488,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 80133
+    },
+    {
+      "epoch": 8.239358420727946,
+      "grad_norm": 0.04020407050848007,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 80136
+    },
+    {
+      "epoch": 8.239666872301049,
+      "grad_norm": 0.049893226474523544,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 80139
+    },
+    {
+      "epoch": 8.239975323874152,
+      "grad_norm": 0.11456778645515442,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 80142
+    },
+    {
+      "epoch": 8.240283775447255,
+      "grad_norm": 0.04332318529486656,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 80145
+    },
+    {
+      "epoch": 8.240592227020358,
+      "grad_norm": 0.03990904986858368,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 80148
+    },
+    {
+      "epoch": 8.24090067859346,
+      "grad_norm": 0.08683262765407562,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 80151
+    },
+    {
+      "epoch": 8.241209130166563,
+      "grad_norm": 0.034948088228702545,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 80154
+    },
+    {
+      "epoch": 8.241517581739666,
+      "grad_norm": 0.03932984918355942,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 80157
+    },
+    {
+      "epoch": 8.241826033312769,
+      "grad_norm": 0.11009229719638824,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 80160
+    },
+    {
+      "epoch": 8.242134484885874,
+      "grad_norm": 0.07181216031312943,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 80163
+    },
+    {
+      "epoch": 8.242442936458977,
+      "grad_norm": 0.03774251788854599,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 80166
+    },
+    {
+      "epoch": 8.24275138803208,
+      "grad_norm": 0.048288244754076004,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 80169
+    },
+    {
+      "epoch": 8.243059839605182,
+      "grad_norm": 0.05333415046334267,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 80172
+    },
+    {
+      "epoch": 8.243368291178285,
+      "grad_norm": 0.13225023448467255,
+      "learning_rate": 0.01,
+      "loss": 1.9806,
+      "step": 80175
+    },
+    {
+      "epoch": 8.243676742751388,
+      "grad_norm": 0.09060022234916687,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 80178
+    },
+    {
+      "epoch": 8.243985194324491,
+      "grad_norm": 0.039559829980134964,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 80181
+    },
+    {
+      "epoch": 8.244293645897594,
+      "grad_norm": 0.041622404009103775,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 80184
+    },
+    {
+      "epoch": 8.244602097470697,
+      "grad_norm": 0.11407449096441269,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 80187
+    },
+    {
+      "epoch": 8.2449105490438,
+      "grad_norm": 0.053344253450632095,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 80190
+    },
+    {
+      "epoch": 8.245219000616903,
+      "grad_norm": 0.08898445218801498,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 80193
+    },
+    {
+      "epoch": 8.245527452190005,
+      "grad_norm": 0.08817441761493683,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 80196
+    },
+    {
+      "epoch": 8.245835903763108,
+      "grad_norm": 0.052410975098609924,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 80199
+    },
+    {
+      "epoch": 8.246144355336213,
+      "grad_norm": 0.04144204780459404,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 80202
+    },
+    {
+      "epoch": 8.246452806909316,
+      "grad_norm": 0.095216304063797,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 80205
+    },
+    {
+      "epoch": 8.246761258482419,
+      "grad_norm": 0.05521360784769058,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 80208
+    },
+    {
+      "epoch": 8.247069710055522,
+      "grad_norm": 0.08642126619815826,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 80211
+    },
+    {
+      "epoch": 8.247378161628625,
+      "grad_norm": 0.06482682377099991,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 80214
+    },
+    {
+      "epoch": 8.247686613201727,
+      "grad_norm": 0.04611750319600105,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 80217
+    },
+    {
+      "epoch": 8.24799506477483,
+      "grad_norm": 0.04172108694911003,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 80220
+    },
+    {
+      "epoch": 8.248303516347933,
+      "grad_norm": 0.038534946739673615,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 80223
+    },
+    {
+      "epoch": 8.248611967921036,
+      "grad_norm": 0.0295123141258955,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 80226
+    },
+    {
+      "epoch": 8.248920419494139,
+      "grad_norm": 0.05776858329772949,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 80229
+    },
+    {
+      "epoch": 8.249228871067242,
+      "grad_norm": 0.21042542159557343,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 80232
+    },
+    {
+      "epoch": 8.249537322640345,
+      "grad_norm": 0.06884560734033585,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 80235
+    },
+    {
+      "epoch": 8.249845774213448,
+      "grad_norm": 0.03800230845808983,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 80238
+    },
+    {
+      "epoch": 8.250154225786552,
+      "grad_norm": 0.06354160606861115,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 80241
+    },
+    {
+      "epoch": 8.250462677359655,
+      "grad_norm": 0.0691051259636879,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 80244
+    },
+    {
+      "epoch": 8.250771128932758,
+      "grad_norm": 0.0348472073674202,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 80247
+    },
+    {
+      "epoch": 8.251079580505861,
+      "grad_norm": 0.03314311057329178,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 80250
+    },
+    {
+      "epoch": 8.251388032078964,
+      "grad_norm": 0.03092731535434723,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 80253
+    },
+    {
+      "epoch": 8.251696483652067,
+      "grad_norm": 0.10996025800704956,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 80256
+    },
+    {
+      "epoch": 8.25200493522517,
+      "grad_norm": 0.05737313628196716,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 80259
+    },
+    {
+      "epoch": 8.252313386798273,
+      "grad_norm": 0.06444061547517776,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 80262
+    },
+    {
+      "epoch": 8.252621838371375,
+      "grad_norm": 0.09499012678861618,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 80265
+    },
+    {
+      "epoch": 8.252930289944478,
+      "grad_norm": 0.10432090610265732,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 80268
+    },
+    {
+      "epoch": 8.253238741517581,
+      "grad_norm": 0.06488598883152008,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 80271
+    },
+    {
+      "epoch": 8.253547193090684,
+      "grad_norm": 0.06193705275654793,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 80274
+    },
+    {
+      "epoch": 8.253855644663787,
+      "grad_norm": 0.0635480210185051,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 80277
+    },
+    {
+      "epoch": 8.254164096236892,
+      "grad_norm": 0.04530000314116478,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 80280
+    },
+    {
+      "epoch": 8.254472547809995,
+      "grad_norm": 0.07301802933216095,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 80283
+    },
+    {
+      "epoch": 8.254780999383097,
+      "grad_norm": 0.0911862924695015,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 80286
+    },
+    {
+      "epoch": 8.2550894509562,
+      "grad_norm": 0.12006412446498871,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 80289
+    },
+    {
+      "epoch": 8.255397902529303,
+      "grad_norm": 0.04310046136379242,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 80292
+    },
+    {
+      "epoch": 8.255706354102406,
+      "grad_norm": 0.055970869958400726,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 80295
+    },
+    {
+      "epoch": 8.256014805675509,
+      "grad_norm": 0.08021442592144012,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 80298
+    },
+    {
+      "epoch": 8.256323257248612,
+      "grad_norm": 0.05120861530303955,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 80301
+    },
+    {
+      "epoch": 8.256631708821715,
+      "grad_norm": 0.09437873214483261,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 80304
+    },
+    {
+      "epoch": 8.256940160394818,
+      "grad_norm": 0.04031134024262428,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 80307
+    },
+    {
+      "epoch": 8.25724861196792,
+      "grad_norm": 0.10186951607465744,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 80310
+    },
+    {
+      "epoch": 8.257557063541023,
+      "grad_norm": 0.05123843625187874,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 80313
+    },
+    {
+      "epoch": 8.257865515114126,
+      "grad_norm": 0.03436708450317383,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 80316
+    },
+    {
+      "epoch": 8.258173966687231,
+      "grad_norm": 0.045604050159454346,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 80319
+    },
+    {
+      "epoch": 8.258482418260334,
+      "grad_norm": 0.04059969261288643,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 80322
+    },
+    {
+      "epoch": 8.258790869833437,
+      "grad_norm": 0.05603836104273796,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 80325
+    },
+    {
+      "epoch": 8.25909932140654,
+      "grad_norm": 0.041143469512462616,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 80328
+    },
+    {
+      "epoch": 8.259407772979642,
+      "grad_norm": 0.051349326968193054,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 80331
+    },
+    {
+      "epoch": 8.259716224552745,
+      "grad_norm": 0.07487104833126068,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 80334
+    },
+    {
+      "epoch": 8.260024676125848,
+      "grad_norm": 0.09332002699375153,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 80337
+    },
+    {
+      "epoch": 8.260333127698951,
+      "grad_norm": 0.11811631172895432,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 80340
+    },
+    {
+      "epoch": 8.260641579272054,
+      "grad_norm": 0.041397687047719955,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 80343
+    },
+    {
+      "epoch": 8.260950030845157,
+      "grad_norm": 0.0496373251080513,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 80346
+    },
+    {
+      "epoch": 8.26125848241826,
+      "grad_norm": 0.0499790795147419,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 80349
+    },
+    {
+      "epoch": 8.261566933991363,
+      "grad_norm": 0.13844771683216095,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 80352
+    },
+    {
+      "epoch": 8.261875385564466,
+      "grad_norm": 0.06328003853559494,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 80355
+    },
+    {
+      "epoch": 8.26218383713757,
+      "grad_norm": 0.05448591709136963,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 80358
+    },
+    {
+      "epoch": 8.262492288710673,
+      "grad_norm": 0.034886013716459274,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 80361
+    },
+    {
+      "epoch": 8.262800740283776,
+      "grad_norm": 0.12212526053190231,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 80364
+    },
+    {
+      "epoch": 8.263109191856879,
+      "grad_norm": 0.04155818372964859,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 80367
+    },
+    {
+      "epoch": 8.263417643429982,
+      "grad_norm": 0.0539507232606411,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 80370
+    },
+    {
+      "epoch": 8.263726095003085,
+      "grad_norm": 0.04046839475631714,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 80373
+    },
+    {
+      "epoch": 8.264034546576188,
+      "grad_norm": 0.04962043836712837,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 80376
+    },
+    {
+      "epoch": 8.26434299814929,
+      "grad_norm": 0.043131254613399506,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 80379
+    },
+    {
+      "epoch": 8.264651449722393,
+      "grad_norm": 0.0805496796965599,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 80382
+    },
+    {
+      "epoch": 8.264959901295496,
+      "grad_norm": 0.08264986425638199,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 80385
+    },
+    {
+      "epoch": 8.265268352868599,
+      "grad_norm": 0.08037829399108887,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 80388
+    },
+    {
+      "epoch": 8.265576804441702,
+      "grad_norm": 0.06496891379356384,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 80391
+    },
+    {
+      "epoch": 8.265885256014805,
+      "grad_norm": 0.049148254096508026,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 80394
+    },
+    {
+      "epoch": 8.26619370758791,
+      "grad_norm": 0.0740509107708931,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 80397
+    },
+    {
+      "epoch": 8.266502159161012,
+      "grad_norm": 0.035503897815942764,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 80400
+    },
+    {
+      "epoch": 8.266810610734115,
+      "grad_norm": 0.043681301176548004,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 80403
+    },
+    {
+      "epoch": 8.267119062307218,
+      "grad_norm": 0.04133985564112663,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 80406
+    },
+    {
+      "epoch": 8.267427513880321,
+      "grad_norm": 0.09284191578626633,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 80409
+    },
+    {
+      "epoch": 8.267735965453424,
+      "grad_norm": 0.0532713383436203,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 80412
+    },
+    {
+      "epoch": 8.268044417026527,
+      "grad_norm": 0.09650903940200806,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 80415
+    },
+    {
+      "epoch": 8.26835286859963,
+      "grad_norm": 0.04948548227548599,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 80418
+    },
+    {
+      "epoch": 8.268661320172733,
+      "grad_norm": 0.029745401814579964,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 80421
+    },
+    {
+      "epoch": 8.268969771745835,
+      "grad_norm": 0.04314611479640007,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 80424
+    },
+    {
+      "epoch": 8.269278223318938,
+      "grad_norm": 0.035734035074710846,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 80427
+    },
+    {
+      "epoch": 8.269586674892041,
+      "grad_norm": 0.06398891657590866,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 80430
+    },
+    {
+      "epoch": 8.269895126465144,
+      "grad_norm": 0.08245398104190826,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 80433
+    },
+    {
+      "epoch": 8.270203578038249,
+      "grad_norm": 0.0568409189581871,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 80436
+    },
+    {
+      "epoch": 8.270512029611352,
+      "grad_norm": 0.05230523645877838,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 80439
+    },
+    {
+      "epoch": 8.270820481184455,
+      "grad_norm": 0.09985850751399994,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 80442
+    },
+    {
+      "epoch": 8.271128932757557,
+      "grad_norm": 0.0723845511674881,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 80445
+    },
+    {
+      "epoch": 8.27143738433066,
+      "grad_norm": 0.1009666696190834,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 80448
+    },
+    {
+      "epoch": 8.271745835903763,
+      "grad_norm": 0.03674089536070824,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 80451
+    },
+    {
+      "epoch": 8.272054287476866,
+      "grad_norm": 0.0843566283583641,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 80454
+    },
+    {
+      "epoch": 8.272362739049969,
+      "grad_norm": 0.04326341301202774,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 80457
+    },
+    {
+      "epoch": 8.272671190623072,
+      "grad_norm": 0.07571718096733093,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 80460
+    },
+    {
+      "epoch": 8.272979642196175,
+      "grad_norm": 0.04678694158792496,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 80463
+    },
+    {
+      "epoch": 8.273288093769278,
+      "grad_norm": 0.10458555072546005,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 80466
+    },
+    {
+      "epoch": 8.27359654534238,
+      "grad_norm": 0.08447905629873276,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 80469
+    },
+    {
+      "epoch": 8.273904996915483,
+      "grad_norm": 0.07039906084537506,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 80472
+    },
+    {
+      "epoch": 8.274213448488588,
+      "grad_norm": 0.05202379822731018,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 80475
+    },
+    {
+      "epoch": 8.274521900061691,
+      "grad_norm": 0.08551952987909317,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 80478
+    },
+    {
+      "epoch": 8.274830351634794,
+      "grad_norm": 0.10451101511716843,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 80481
+    },
+    {
+      "epoch": 8.275138803207897,
+      "grad_norm": 0.13095597922801971,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 80484
+    },
+    {
+      "epoch": 8.275447254781,
+      "grad_norm": 0.10141288489103317,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 80487
+    },
+    {
+      "epoch": 8.275755706354103,
+      "grad_norm": 0.09214049577713013,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 80490
+    },
+    {
+      "epoch": 8.276064157927205,
+      "grad_norm": 0.042113929986953735,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 80493
+    },
+    {
+      "epoch": 8.276372609500308,
+      "grad_norm": 0.051108475774526596,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 80496
+    },
+    {
+      "epoch": 8.276681061073411,
+      "grad_norm": 0.04188302531838417,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 80499
+    },
+    {
+      "epoch": 8.276989512646514,
+      "grad_norm": 0.0321899838745594,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 80502
+    },
+    {
+      "epoch": 8.277297964219617,
+      "grad_norm": 0.038304295390844345,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 80505
+    },
+    {
+      "epoch": 8.27760641579272,
+      "grad_norm": 0.04976971819996834,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 80508
+    },
+    {
+      "epoch": 8.277914867365823,
+      "grad_norm": 0.10897812247276306,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 80511
+    },
+    {
+      "epoch": 8.278223318938927,
+      "grad_norm": 0.04283289983868599,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 80514
+    },
+    {
+      "epoch": 8.27853177051203,
+      "grad_norm": 0.0434601865708828,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 80517
+    },
+    {
+      "epoch": 8.278840222085133,
+      "grad_norm": 0.1142946258187294,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 80520
+    },
+    {
+      "epoch": 8.279148673658236,
+      "grad_norm": 0.12008026242256165,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 80523
+    },
+    {
+      "epoch": 8.279457125231339,
+      "grad_norm": 0.04869767278432846,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 80526
+    },
+    {
+      "epoch": 8.279765576804442,
+      "grad_norm": 0.037496909499168396,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 80529
+    },
+    {
+      "epoch": 8.280074028377545,
+      "grad_norm": 0.041276488453149796,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 80532
+    },
+    {
+      "epoch": 8.280382479950648,
+      "grad_norm": 0.054701920598745346,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 80535
+    },
+    {
+      "epoch": 8.28069093152375,
+      "grad_norm": 0.06538514047861099,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 80538
+    },
+    {
+      "epoch": 8.280999383096853,
+      "grad_norm": 0.05490681156516075,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 80541
+    },
+    {
+      "epoch": 8.281307834669956,
+      "grad_norm": 0.03307463601231575,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 80544
+    },
+    {
+      "epoch": 8.28161628624306,
+      "grad_norm": 0.12409739941358566,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 80547
+    },
+    {
+      "epoch": 8.281924737816162,
+      "grad_norm": 0.0904029905796051,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 80550
+    },
+    {
+      "epoch": 8.282233189389267,
+      "grad_norm": 0.09008060395717621,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 80553
+    },
+    {
+      "epoch": 8.28254164096237,
+      "grad_norm": 0.06393894553184509,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 80556
+    },
+    {
+      "epoch": 8.282850092535472,
+      "grad_norm": 0.0343107245862484,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 80559
+    },
+    {
+      "epoch": 8.283158544108575,
+      "grad_norm": 0.04057645797729492,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 80562
+    },
+    {
+      "epoch": 8.283466995681678,
+      "grad_norm": 0.04213458299636841,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 80565
+    },
+    {
+      "epoch": 8.283775447254781,
+      "grad_norm": 0.03605170175433159,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 80568
+    },
+    {
+      "epoch": 8.284083898827884,
+      "grad_norm": 0.04358351603150368,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 80571
+    },
+    {
+      "epoch": 8.284392350400987,
+      "grad_norm": 0.04204123094677925,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 80574
+    },
+    {
+      "epoch": 8.28470080197409,
+      "grad_norm": 0.037242695689201355,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 80577
+    },
+    {
+      "epoch": 8.285009253547193,
+      "grad_norm": 0.09867773950099945,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 80580
+    },
+    {
+      "epoch": 8.285317705120296,
+      "grad_norm": 0.08788815140724182,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 80583
+    },
+    {
+      "epoch": 8.285626156693398,
+      "grad_norm": 0.0887393206357956,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 80586
+    },
+    {
+      "epoch": 8.285934608266501,
+      "grad_norm": 0.04635678976774216,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 80589
+    },
+    {
+      "epoch": 8.286243059839606,
+      "grad_norm": 0.09219451993703842,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 80592
+    },
+    {
+      "epoch": 8.286551511412709,
+      "grad_norm": 0.08263669162988663,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 80595
+    },
+    {
+      "epoch": 8.286859962985812,
+      "grad_norm": 0.03986113891005516,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 80598
+    },
+    {
+      "epoch": 8.287168414558915,
+      "grad_norm": 0.04161595553159714,
+      "learning_rate": 0.01,
+      "loss": 1.9827,
+      "step": 80601
+    },
+    {
+      "epoch": 8.287476866132018,
+      "grad_norm": 0.03629259765148163,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 80604
+    },
+    {
+      "epoch": 8.28778531770512,
+      "grad_norm": 0.05570584908127785,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 80607
+    },
+    {
+      "epoch": 8.288093769278223,
+      "grad_norm": 0.04991071671247482,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 80610
+    },
+    {
+      "epoch": 8.288402220851326,
+      "grad_norm": 0.11207103729248047,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 80613
+    },
+    {
+      "epoch": 8.288710672424429,
+      "grad_norm": 0.04575880244374275,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 80616
+    },
+    {
+      "epoch": 8.289019123997532,
+      "grad_norm": 0.046922992914915085,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 80619
+    },
+    {
+      "epoch": 8.289327575570635,
+      "grad_norm": 0.04037918150424957,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 80622
+    },
+    {
+      "epoch": 8.289636027143738,
+      "grad_norm": 0.06104828417301178,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 80625
+    },
+    {
+      "epoch": 8.28994447871684,
+      "grad_norm": 0.05090169981122017,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 80628
+    },
+    {
+      "epoch": 8.290252930289945,
+      "grad_norm": 0.061494361609220505,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 80631
+    },
+    {
+      "epoch": 8.290561381863048,
+      "grad_norm": 0.04942132532596588,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 80634
+    },
+    {
+      "epoch": 8.290869833436151,
+      "grad_norm": 0.038736648857593536,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 80637
+    },
+    {
+      "epoch": 8.291178285009254,
+      "grad_norm": 0.03663397207856178,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 80640
+    },
+    {
+      "epoch": 8.291486736582357,
+      "grad_norm": 0.10741380602121353,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 80643
+    },
+    {
+      "epoch": 8.29179518815546,
+      "grad_norm": 0.13612329959869385,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 80646
+    },
+    {
+      "epoch": 8.292103639728563,
+      "grad_norm": 0.06880326569080353,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 80649
+    },
+    {
+      "epoch": 8.292412091301665,
+      "grad_norm": 0.0510687455534935,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 80652
+    },
+    {
+      "epoch": 8.292720542874768,
+      "grad_norm": 0.05295844376087189,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 80655
+    },
+    {
+      "epoch": 8.293028994447871,
+      "grad_norm": 0.04496245086193085,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 80658
+    },
+    {
+      "epoch": 8.293337446020974,
+      "grad_norm": 0.07872740179300308,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 80661
+    },
+    {
+      "epoch": 8.293645897594077,
+      "grad_norm": 0.07173445075750351,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 80664
+    },
+    {
+      "epoch": 8.29395434916718,
+      "grad_norm": 0.08514854311943054,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 80667
+    },
+    {
+      "epoch": 8.294262800740285,
+      "grad_norm": 0.07339558005332947,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 80670
+    },
+    {
+      "epoch": 8.294571252313387,
+      "grad_norm": 0.07977834343910217,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 80673
+    },
+    {
+      "epoch": 8.29487970388649,
+      "grad_norm": 0.1017339676618576,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 80676
+    },
+    {
+      "epoch": 8.295188155459593,
+      "grad_norm": 0.038830697536468506,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 80679
+    },
+    {
+      "epoch": 8.295496607032696,
+      "grad_norm": 0.10399825125932693,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 80682
+    },
+    {
+      "epoch": 8.295805058605799,
+      "grad_norm": 0.0402609184384346,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 80685
+    },
+    {
+      "epoch": 8.296113510178902,
+      "grad_norm": 0.034392885863780975,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 80688
+    },
+    {
+      "epoch": 8.296421961752005,
+      "grad_norm": 0.1786748617887497,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 80691
+    },
+    {
+      "epoch": 8.296730413325108,
+      "grad_norm": 0.08792231976985931,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 80694
+    },
+    {
+      "epoch": 8.29703886489821,
+      "grad_norm": 0.10741207748651505,
+      "learning_rate": 0.01,
+      "loss": 1.9861,
+      "step": 80697
+    },
+    {
+      "epoch": 8.297347316471313,
+      "grad_norm": 0.08831042796373367,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 80700
+    },
+    {
+      "epoch": 8.297655768044416,
+      "grad_norm": 0.09200239181518555,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 80703
+    },
+    {
+      "epoch": 8.29796421961752,
+      "grad_norm": 0.05381232500076294,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 80706
+    },
+    {
+      "epoch": 8.298272671190624,
+      "grad_norm": 0.032259222120046616,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 80709
+    },
+    {
+      "epoch": 8.298581122763727,
+      "grad_norm": 0.049715928733348846,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 80712
+    },
+    {
+      "epoch": 8.29888957433683,
+      "grad_norm": 0.0569734163582325,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 80715
+    },
+    {
+      "epoch": 8.299198025909932,
+      "grad_norm": 0.04328460991382599,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 80718
+    },
+    {
+      "epoch": 8.299506477483035,
+      "grad_norm": 0.09637049585580826,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 80721
+    },
+    {
+      "epoch": 8.299814929056138,
+      "grad_norm": 0.0515410453081131,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 80724
+    },
+    {
+      "epoch": 8.300123380629241,
+      "grad_norm": 0.041060756891965866,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 80727
+    },
+    {
+      "epoch": 8.300431832202344,
+      "grad_norm": 0.07743452489376068,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 80730
+    },
+    {
+      "epoch": 8.300740283775447,
+      "grad_norm": 0.08073935657739639,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 80733
+    },
+    {
+      "epoch": 8.30104873534855,
+      "grad_norm": 0.13999415934085846,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 80736
+    },
+    {
+      "epoch": 8.301357186921653,
+      "grad_norm": 0.0980861559510231,
+      "learning_rate": 0.01,
+      "loss": 1.9986,
+      "step": 80739
+    },
+    {
+      "epoch": 8.301665638494756,
+      "grad_norm": 0.09149259328842163,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 80742
+    },
+    {
+      "epoch": 8.301974090067858,
+      "grad_norm": 0.05871947482228279,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 80745
+    },
+    {
+      "epoch": 8.302282541640963,
+      "grad_norm": 0.045800916850566864,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 80748
+    },
+    {
+      "epoch": 8.302590993214066,
+      "grad_norm": 0.0665321871638298,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 80751
+    },
+    {
+      "epoch": 8.302899444787169,
+      "grad_norm": 0.08737419545650482,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 80754
+    },
+    {
+      "epoch": 8.303207896360272,
+      "grad_norm": 0.05846291407942772,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 80757
+    },
+    {
+      "epoch": 8.303516347933375,
+      "grad_norm": 0.058516647666692734,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 80760
+    },
+    {
+      "epoch": 8.303824799506478,
+      "grad_norm": 0.05862552672624588,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 80763
+    },
+    {
+      "epoch": 8.30413325107958,
+      "grad_norm": 0.04875510185956955,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 80766
+    },
+    {
+      "epoch": 8.304441702652683,
+      "grad_norm": 0.10074342787265778,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 80769
+    },
+    {
+      "epoch": 8.304750154225786,
+      "grad_norm": 0.03849528357386589,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 80772
+    },
+    {
+      "epoch": 8.305058605798889,
+      "grad_norm": 0.1587829291820526,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 80775
+    },
+    {
+      "epoch": 8.305367057371992,
+      "grad_norm": 0.06458421051502228,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 80778
+    },
+    {
+      "epoch": 8.305675508945095,
+      "grad_norm": 0.07363933324813843,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 80781
+    },
+    {
+      "epoch": 8.305983960518198,
+      "grad_norm": 0.03997383266687393,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 80784
+    },
+    {
+      "epoch": 8.306292412091302,
+      "grad_norm": 0.037976641207933426,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 80787
+    },
+    {
+      "epoch": 8.306600863664405,
+      "grad_norm": 0.039313819259405136,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 80790
+    },
+    {
+      "epoch": 8.306909315237508,
+      "grad_norm": 0.1089361310005188,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 80793
+    },
+    {
+      "epoch": 8.307217766810611,
+      "grad_norm": 0.15032267570495605,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 80796
+    },
+    {
+      "epoch": 8.307526218383714,
+      "grad_norm": 0.06436207890510559,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 80799
+    },
+    {
+      "epoch": 8.307834669956817,
+      "grad_norm": 0.03206050768494606,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 80802
+    },
+    {
+      "epoch": 8.30814312152992,
+      "grad_norm": 0.040079426020383835,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 80805
+    },
+    {
+      "epoch": 8.308451573103023,
+      "grad_norm": 0.04091683030128479,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 80808
+    },
+    {
+      "epoch": 8.308760024676126,
+      "grad_norm": 0.07872221618890762,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 80811
+    },
+    {
+      "epoch": 8.309068476249228,
+      "grad_norm": 0.10001574456691742,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 80814
+    },
+    {
+      "epoch": 8.309376927822331,
+      "grad_norm": 0.07447927445173264,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 80817
+    },
+    {
+      "epoch": 8.309685379395434,
+      "grad_norm": 0.048718079924583435,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 80820
+    },
+    {
+      "epoch": 8.309993830968537,
+      "grad_norm": 0.08882563561201096,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 80823
+    },
+    {
+      "epoch": 8.310302282541642,
+      "grad_norm": 0.04461067542433739,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 80826
+    },
+    {
+      "epoch": 8.310610734114745,
+      "grad_norm": 0.09042662382125854,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 80829
+    },
+    {
+      "epoch": 8.310919185687847,
+      "grad_norm": 0.07167694717645645,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 80832
+    },
+    {
+      "epoch": 8.31122763726095,
+      "grad_norm": 0.06814853847026825,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 80835
+    },
+    {
+      "epoch": 8.311536088834053,
+      "grad_norm": 0.09978099912405014,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 80838
+    },
+    {
+      "epoch": 8.311844540407156,
+      "grad_norm": 0.08152987062931061,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 80841
+    },
+    {
+      "epoch": 8.312152991980259,
+      "grad_norm": 0.03953191637992859,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 80844
+    },
+    {
+      "epoch": 8.312461443553362,
+      "grad_norm": 0.0646815299987793,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 80847
+    },
+    {
+      "epoch": 8.312769895126465,
+      "grad_norm": 0.10371677577495575,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 80850
+    },
+    {
+      "epoch": 8.313078346699568,
+      "grad_norm": 0.1298215538263321,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 80853
+    },
+    {
+      "epoch": 8.31338679827267,
+      "grad_norm": 0.05871061235666275,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 80856
+    },
+    {
+      "epoch": 8.313695249845773,
+      "grad_norm": 0.03924513980746269,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 80859
+    },
+    {
+      "epoch": 8.314003701418876,
+      "grad_norm": 0.05203021690249443,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 80862
+    },
+    {
+      "epoch": 8.314312152991981,
+      "grad_norm": 0.07967104017734528,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 80865
+    },
+    {
+      "epoch": 8.314620604565084,
+      "grad_norm": 0.09806268662214279,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 80868
+    },
+    {
+      "epoch": 8.314929056138187,
+      "grad_norm": 0.08093216270208359,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 80871
+    },
+    {
+      "epoch": 8.31523750771129,
+      "grad_norm": 0.05892954394221306,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 80874
+    },
+    {
+      "epoch": 8.315545959284393,
+      "grad_norm": 0.04190627112984657,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 80877
+    },
+    {
+      "epoch": 8.315854410857495,
+      "grad_norm": 0.09374337643384933,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 80880
+    },
+    {
+      "epoch": 8.316162862430598,
+      "grad_norm": 0.09757273644208908,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 80883
+    },
+    {
+      "epoch": 8.316471314003701,
+      "grad_norm": 0.07102153450250626,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 80886
+    },
+    {
+      "epoch": 8.316779765576804,
+      "grad_norm": 0.08393047004938126,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 80889
+    },
+    {
+      "epoch": 8.317088217149907,
+      "grad_norm": 0.09592314064502716,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 80892
+    },
+    {
+      "epoch": 8.31739666872301,
+      "grad_norm": 0.05190633237361908,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 80895
+    },
+    {
+      "epoch": 8.317705120296113,
+      "grad_norm": 0.041763097047805786,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 80898
+    },
+    {
+      "epoch": 8.318013571869216,
+      "grad_norm": 0.050488606095314026,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 80901
+    },
+    {
+      "epoch": 8.31832202344232,
+      "grad_norm": 0.03751280531287193,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 80904
+    },
+    {
+      "epoch": 8.318630475015423,
+      "grad_norm": 0.04075468331575394,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 80907
+    },
+    {
+      "epoch": 8.318938926588526,
+      "grad_norm": 0.05772188678383827,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 80910
+    },
+    {
+      "epoch": 8.319247378161629,
+      "grad_norm": 0.06489258259534836,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 80913
+    },
+    {
+      "epoch": 8.319555829734732,
+      "grad_norm": 0.050094038248062134,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 80916
+    },
+    {
+      "epoch": 8.319864281307835,
+      "grad_norm": 0.08120666444301605,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 80919
+    },
+    {
+      "epoch": 8.320172732880938,
+      "grad_norm": 0.12392567843198776,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 80922
+    },
+    {
+      "epoch": 8.32048118445404,
+      "grad_norm": 0.0636238157749176,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 80925
+    },
+    {
+      "epoch": 8.320789636027143,
+      "grad_norm": 0.051416803151369095,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 80928
+    },
+    {
+      "epoch": 8.321098087600246,
+      "grad_norm": 0.0705709159374237,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 80931
+    },
+    {
+      "epoch": 8.32140653917335,
+      "grad_norm": 0.1259060502052307,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 80934
+    },
+    {
+      "epoch": 8.321714990746452,
+      "grad_norm": 0.06882835924625397,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 80937
+    },
+    {
+      "epoch": 8.322023442319555,
+      "grad_norm": 0.06935711205005646,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 80940
+    },
+    {
+      "epoch": 8.32233189389266,
+      "grad_norm": 0.07351617515087128,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 80943
+    },
+    {
+      "epoch": 8.322640345465762,
+      "grad_norm": 0.04996279254555702,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 80946
+    },
+    {
+      "epoch": 8.322948797038865,
+      "grad_norm": 0.08058802038431168,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 80949
+    },
+    {
+      "epoch": 8.323257248611968,
+      "grad_norm": 0.09268704801797867,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 80952
+    },
+    {
+      "epoch": 8.323565700185071,
+      "grad_norm": 0.14823900163173676,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 80955
+    },
+    {
+      "epoch": 8.323874151758174,
+      "grad_norm": 0.11526891589164734,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 80958
+    },
+    {
+      "epoch": 8.324182603331277,
+      "grad_norm": 0.10285145789384842,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 80961
+    },
+    {
+      "epoch": 8.32449105490438,
+      "grad_norm": 0.04557880014181137,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 80964
+    },
+    {
+      "epoch": 8.324799506477483,
+      "grad_norm": 0.039869774132966995,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 80967
+    },
+    {
+      "epoch": 8.325107958050586,
+      "grad_norm": 0.03244692087173462,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 80970
+    },
+    {
+      "epoch": 8.325416409623688,
+      "grad_norm": 0.040278058499097824,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 80973
+    },
+    {
+      "epoch": 8.325724861196791,
+      "grad_norm": 0.0789649486541748,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 80976
+    },
+    {
+      "epoch": 8.326033312769894,
+      "grad_norm": 0.06306403130292892,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 80979
+    },
+    {
+      "epoch": 8.326341764342999,
+      "grad_norm": 0.04320656880736351,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 80982
+    },
+    {
+      "epoch": 8.326650215916102,
+      "grad_norm": 0.03742117062211037,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 80985
+    },
+    {
+      "epoch": 8.326958667489205,
+      "grad_norm": 0.08791963756084442,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 80988
+    },
+    {
+      "epoch": 8.327267119062308,
+      "grad_norm": 0.05794864147901535,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 80991
+    },
+    {
+      "epoch": 8.32757557063541,
+      "grad_norm": 0.09958462417125702,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 80994
+    },
+    {
+      "epoch": 8.327884022208513,
+      "grad_norm": 0.0908295288681984,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 80997
+    },
+    {
+      "epoch": 8.328192473781616,
+      "grad_norm": 0.05362537130713463,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 81000
+    },
+    {
+      "epoch": 8.328500925354719,
+      "grad_norm": 0.03491285815834999,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 81003
+    },
+    {
+      "epoch": 8.328809376927822,
+      "grad_norm": 0.04607011377811432,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 81006
+    },
+    {
+      "epoch": 8.329117828500925,
+      "grad_norm": 0.036862924695014954,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 81009
+    },
+    {
+      "epoch": 8.329426280074028,
+      "grad_norm": 0.03782851994037628,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 81012
+    },
+    {
+      "epoch": 8.32973473164713,
+      "grad_norm": 0.10079167783260345,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 81015
+    },
+    {
+      "epoch": 8.330043183220234,
+      "grad_norm": 0.06869961321353912,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 81018
+    },
+    {
+      "epoch": 8.330351634793338,
+      "grad_norm": 0.09453833103179932,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 81021
+    },
+    {
+      "epoch": 8.330660086366441,
+      "grad_norm": 0.06898962706327438,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 81024
+    },
+    {
+      "epoch": 8.330968537939544,
+      "grad_norm": 0.0684114545583725,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 81027
+    },
+    {
+      "epoch": 8.331276989512647,
+      "grad_norm": 0.07283013314008713,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 81030
+    },
+    {
+      "epoch": 8.33158544108575,
+      "grad_norm": 0.059807684272527695,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 81033
+    },
+    {
+      "epoch": 8.331893892658853,
+      "grad_norm": 0.06721007823944092,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 81036
+    },
+    {
+      "epoch": 8.332202344231955,
+      "grad_norm": 0.0669078379869461,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 81039
+    },
+    {
+      "epoch": 8.332510795805058,
+      "grad_norm": 0.07649901509284973,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 81042
+    },
+    {
+      "epoch": 8.332819247378161,
+      "grad_norm": 0.07991660386323929,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 81045
+    },
+    {
+      "epoch": 8.333127698951264,
+      "grad_norm": 0.0629889965057373,
+      "learning_rate": 0.01,
+      "loss": 1.9894,
+      "step": 81048
+    },
+    {
+      "epoch": 8.333436150524367,
+      "grad_norm": 0.036793168634176254,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 81051
+    },
+    {
+      "epoch": 8.33374460209747,
+      "grad_norm": 0.032679762691259384,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 81054
+    },
+    {
+      "epoch": 8.334053053670573,
+      "grad_norm": 0.08353456854820251,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 81057
+    },
+    {
+      "epoch": 8.334361505243677,
+      "grad_norm": 0.06313187628984451,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 81060
+    },
+    {
+      "epoch": 8.33466995681678,
+      "grad_norm": 0.03921021521091461,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 81063
+    },
+    {
+      "epoch": 8.334978408389883,
+      "grad_norm": 0.044407907873392105,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 81066
+    },
+    {
+      "epoch": 8.335286859962986,
+      "grad_norm": 0.0652817115187645,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 81069
+    },
+    {
+      "epoch": 8.335595311536089,
+      "grad_norm": 0.09266695380210876,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 81072
+    },
+    {
+      "epoch": 8.335903763109192,
+      "grad_norm": 0.07401798665523529,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 81075
+    },
+    {
+      "epoch": 8.336212214682295,
+      "grad_norm": 0.031432922929525375,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 81078
+    },
+    {
+      "epoch": 8.336520666255398,
+      "grad_norm": 0.06749407202005386,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 81081
+    },
+    {
+      "epoch": 8.3368291178285,
+      "grad_norm": 0.052030496299266815,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 81084
+    },
+    {
+      "epoch": 8.337137569401603,
+      "grad_norm": 0.034096091985702515,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 81087
+    },
+    {
+      "epoch": 8.337446020974706,
+      "grad_norm": 0.04452771693468094,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 81090
+    },
+    {
+      "epoch": 8.33775447254781,
+      "grad_norm": 0.11472178995609283,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 81093
+    },
+    {
+      "epoch": 8.338062924120914,
+      "grad_norm": 0.09258271008729935,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 81096
+    },
+    {
+      "epoch": 8.338371375694017,
+      "grad_norm": 0.05785766988992691,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 81099
+    },
+    {
+      "epoch": 8.33867982726712,
+      "grad_norm": 0.03770722821354866,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 81102
+    },
+    {
+      "epoch": 8.338988278840223,
+      "grad_norm": 0.038187529891729355,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 81105
+    },
+    {
+      "epoch": 8.339296730413325,
+      "grad_norm": 0.044281262904405594,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 81108
+    },
+    {
+      "epoch": 8.339605181986428,
+      "grad_norm": 0.050860438495874405,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 81111
+    },
+    {
+      "epoch": 8.339913633559531,
+      "grad_norm": 0.08445332944393158,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 81114
+    },
+    {
+      "epoch": 8.340222085132634,
+      "grad_norm": 0.11284556984901428,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 81117
+    },
+    {
+      "epoch": 8.340530536705737,
+      "grad_norm": 0.05600573495030403,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 81120
+    },
+    {
+      "epoch": 8.34083898827884,
+      "grad_norm": 0.042293425649404526,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 81123
+    },
+    {
+      "epoch": 8.341147439851943,
+      "grad_norm": 0.03872934356331825,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 81126
+    },
+    {
+      "epoch": 8.341455891425046,
+      "grad_norm": 0.043065737932920456,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 81129
+    },
+    {
+      "epoch": 8.341764342998149,
+      "grad_norm": 0.04543273523449898,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 81132
+    },
+    {
+      "epoch": 8.342072794571253,
+      "grad_norm": 0.15710321068763733,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 81135
+    },
+    {
+      "epoch": 8.342381246144356,
+      "grad_norm": 0.045610420405864716,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 81138
+    },
+    {
+      "epoch": 8.342689697717459,
+      "grad_norm": 0.04882073402404785,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 81141
+    },
+    {
+      "epoch": 8.342998149290562,
+      "grad_norm": 0.037526652216911316,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 81144
+    },
+    {
+      "epoch": 8.343306600863665,
+      "grad_norm": 0.03276045620441437,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 81147
+    },
+    {
+      "epoch": 8.343615052436768,
+      "grad_norm": 0.04296638444066048,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 81150
+    },
+    {
+      "epoch": 8.34392350400987,
+      "grad_norm": 0.07781503349542618,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 81153
+    },
+    {
+      "epoch": 8.344231955582973,
+      "grad_norm": 0.039542727172374725,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 81156
+    },
+    {
+      "epoch": 8.344540407156076,
+      "grad_norm": 0.045888807624578476,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 81159
+    },
+    {
+      "epoch": 8.34484885872918,
+      "grad_norm": 0.05557486042380333,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 81162
+    },
+    {
+      "epoch": 8.345157310302282,
+      "grad_norm": 0.06092909723520279,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 81165
+    },
+    {
+      "epoch": 8.345465761875385,
+      "grad_norm": 0.17519673705101013,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 81168
+    },
+    {
+      "epoch": 8.345774213448488,
+      "grad_norm": 0.09101748466491699,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 81171
+    },
+    {
+      "epoch": 8.346082665021592,
+      "grad_norm": 0.05962994322180748,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 81174
+    },
+    {
+      "epoch": 8.346391116594695,
+      "grad_norm": 0.0337313674390316,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 81177
+    },
+    {
+      "epoch": 8.346699568167798,
+      "grad_norm": 0.06669741123914719,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 81180
+    },
+    {
+      "epoch": 8.347008019740901,
+      "grad_norm": 0.09194620698690414,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 81183
+    },
+    {
+      "epoch": 8.347316471314004,
+      "grad_norm": 0.04323070868849754,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 81186
+    },
+    {
+      "epoch": 8.347624922887107,
+      "grad_norm": 0.0381295420229435,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 81189
+    },
+    {
+      "epoch": 8.34793337446021,
+      "grad_norm": 0.03433126211166382,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 81192
+    },
+    {
+      "epoch": 8.348241826033313,
+      "grad_norm": 0.03897920623421669,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 81195
+    },
+    {
+      "epoch": 8.348550277606416,
+      "grad_norm": 0.04720534011721611,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 81198
+    },
+    {
+      "epoch": 8.348858729179518,
+      "grad_norm": 0.036591820418834686,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 81201
+    },
+    {
+      "epoch": 8.349167180752621,
+      "grad_norm": 0.06843716651201248,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 81204
+    },
+    {
+      "epoch": 8.349475632325724,
+      "grad_norm": 0.05791595205664635,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 81207
+    },
+    {
+      "epoch": 8.349784083898827,
+      "grad_norm": 0.11137792468070984,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 81210
+    },
+    {
+      "epoch": 8.350092535471932,
+      "grad_norm": 0.08609547466039658,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 81213
+    },
+    {
+      "epoch": 8.350400987045035,
+      "grad_norm": 0.1130346804857254,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 81216
+    },
+    {
+      "epoch": 8.350709438618138,
+      "grad_norm": 0.06280580908060074,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 81219
+    },
+    {
+      "epoch": 8.35101789019124,
+      "grad_norm": 0.059064436703920364,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 81222
+    },
+    {
+      "epoch": 8.351326341764343,
+      "grad_norm": 0.03772406652569771,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 81225
+    },
+    {
+      "epoch": 8.351634793337446,
+      "grad_norm": 0.06599714607000351,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 81228
+    },
+    {
+      "epoch": 8.351943244910549,
+      "grad_norm": 0.08209977298974991,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 81231
+    },
+    {
+      "epoch": 8.352251696483652,
+      "grad_norm": 0.14253711700439453,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 81234
+    },
+    {
+      "epoch": 8.352560148056755,
+      "grad_norm": 0.06577881425619125,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 81237
+    },
+    {
+      "epoch": 8.352868599629858,
+      "grad_norm": 0.07357854396104813,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 81240
+    },
+    {
+      "epoch": 8.35317705120296,
+      "grad_norm": 0.05654516443610191,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 81243
+    },
+    {
+      "epoch": 8.353485502776063,
+      "grad_norm": 0.04166090860962868,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 81246
+    },
+    {
+      "epoch": 8.353793954349166,
+      "grad_norm": 0.03527102991938591,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 81249
+    },
+    {
+      "epoch": 8.354102405922271,
+      "grad_norm": 0.04787033051252365,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 81252
+    },
+    {
+      "epoch": 8.354410857495374,
+      "grad_norm": 0.1353912353515625,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 81255
+    },
+    {
+      "epoch": 8.354719309068477,
+      "grad_norm": 0.049617063254117966,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 81258
+    },
+    {
+      "epoch": 8.35502776064158,
+      "grad_norm": 0.0665043517947197,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 81261
+    },
+    {
+      "epoch": 8.355336212214683,
+      "grad_norm": 0.05647074803709984,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 81264
+    },
+    {
+      "epoch": 8.355644663787785,
+      "grad_norm": 0.0868644118309021,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 81267
+    },
+    {
+      "epoch": 8.355953115360888,
+      "grad_norm": 0.12009144574403763,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 81270
+    },
+    {
+      "epoch": 8.356261566933991,
+      "grad_norm": 0.09910716116428375,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 81273
+    },
+    {
+      "epoch": 8.356570018507094,
+      "grad_norm": 0.08180820196866989,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 81276
+    },
+    {
+      "epoch": 8.356878470080197,
+      "grad_norm": 0.04328274726867676,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 81279
+    },
+    {
+      "epoch": 8.3571869216533,
+      "grad_norm": 0.04688665270805359,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 81282
+    },
+    {
+      "epoch": 8.357495373226403,
+      "grad_norm": 0.05602161958813667,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 81285
+    },
+    {
+      "epoch": 8.357803824799506,
+      "grad_norm": 0.04227988049387932,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 81288
+    },
+    {
+      "epoch": 8.35811227637261,
+      "grad_norm": 0.10519243031740189,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 81291
+    },
+    {
+      "epoch": 8.358420727945713,
+      "grad_norm": 0.051333289593458176,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 81294
+    },
+    {
+      "epoch": 8.358729179518816,
+      "grad_norm": 0.08162673562765121,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 81297
+    },
+    {
+      "epoch": 8.359037631091919,
+      "grad_norm": 0.04508831351995468,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 81300
+    },
+    {
+      "epoch": 8.359346082665022,
+      "grad_norm": 0.062075745314359665,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 81303
+    },
+    {
+      "epoch": 8.359654534238125,
+      "grad_norm": 0.055787600576877594,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 81306
+    },
+    {
+      "epoch": 8.359962985811228,
+      "grad_norm": 0.05872397869825363,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 81309
+    },
+    {
+      "epoch": 8.36027143738433,
+      "grad_norm": 0.048803798854351044,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 81312
+    },
+    {
+      "epoch": 8.360579888957433,
+      "grad_norm": 0.03489650413393974,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 81315
+    },
+    {
+      "epoch": 8.360888340530536,
+      "grad_norm": 0.050747793167829514,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 81318
+    },
+    {
+      "epoch": 8.36119679210364,
+      "grad_norm": 0.08788271993398666,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 81321
+    },
+    {
+      "epoch": 8.361505243676742,
+      "grad_norm": 0.11580553650856018,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 81324
+    },
+    {
+      "epoch": 8.361813695249845,
+      "grad_norm": 0.05664614588022232,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 81327
+    },
+    {
+      "epoch": 8.36212214682295,
+      "grad_norm": 0.07168247550725937,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 81330
+    },
+    {
+      "epoch": 8.362430598396053,
+      "grad_norm": 0.0754905566573143,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 81333
+    },
+    {
+      "epoch": 8.362739049969155,
+      "grad_norm": 0.055534813553094864,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 81336
+    },
+    {
+      "epoch": 8.363047501542258,
+      "grad_norm": 0.13377831876277924,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 81339
+    },
+    {
+      "epoch": 8.363355953115361,
+      "grad_norm": 0.06023402139544487,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 81342
+    },
+    {
+      "epoch": 8.363664404688464,
+      "grad_norm": 0.05767045170068741,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 81345
+    },
+    {
+      "epoch": 8.363972856261567,
+      "grad_norm": 0.059322405606508255,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 81348
+    },
+    {
+      "epoch": 8.36428130783467,
+      "grad_norm": 0.042639970779418945,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 81351
+    },
+    {
+      "epoch": 8.364589759407773,
+      "grad_norm": 0.05120370537042618,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 81354
+    },
+    {
+      "epoch": 8.364898210980876,
+      "grad_norm": 0.04822765290737152,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 81357
+    },
+    {
+      "epoch": 8.365206662553978,
+      "grad_norm": 0.05606779083609581,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 81360
+    },
+    {
+      "epoch": 8.365515114127081,
+      "grad_norm": 0.05448019132018089,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 81363
+    },
+    {
+      "epoch": 8.365823565700184,
+      "grad_norm": 0.036996424198150635,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 81366
+    },
+    {
+      "epoch": 8.366132017273289,
+      "grad_norm": 0.03727582097053528,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 81369
+    },
+    {
+      "epoch": 8.366440468846392,
+      "grad_norm": 0.04183301329612732,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 81372
+    },
+    {
+      "epoch": 8.366748920419495,
+      "grad_norm": 0.06489592790603638,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 81375
+    },
+    {
+      "epoch": 8.367057371992598,
+      "grad_norm": 0.19914110004901886,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 81378
+    },
+    {
+      "epoch": 8.3673658235657,
+      "grad_norm": 0.10219461470842361,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 81381
+    },
+    {
+      "epoch": 8.367674275138803,
+      "grad_norm": 0.0821671262383461,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 81384
+    },
+    {
+      "epoch": 8.367982726711906,
+      "grad_norm": 0.08505228161811829,
+      "learning_rate": 0.01,
+      "loss": 2.0058,
+      "step": 81387
+    },
+    {
+      "epoch": 8.36829117828501,
+      "grad_norm": 0.07406182587146759,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 81390
+    },
+    {
+      "epoch": 8.368599629858112,
+      "grad_norm": 0.029416732490062714,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 81393
+    },
+    {
+      "epoch": 8.368908081431215,
+      "grad_norm": 0.03504988178610802,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 81396
+    },
+    {
+      "epoch": 8.369216533004318,
+      "grad_norm": 0.03661525994539261,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 81399
+    },
+    {
+      "epoch": 8.36952498457742,
+      "grad_norm": 0.04680737107992172,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 81402
+    },
+    {
+      "epoch": 8.369833436150524,
+      "grad_norm": 0.0412558875977993,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 81405
+    },
+    {
+      "epoch": 8.370141887723628,
+      "grad_norm": 0.08087299019098282,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 81408
+    },
+    {
+      "epoch": 8.370450339296731,
+      "grad_norm": 0.07752052694559097,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 81411
+    },
+    {
+      "epoch": 8.370758790869834,
+      "grad_norm": 0.08101408183574677,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 81414
+    },
+    {
+      "epoch": 8.371067242442937,
+      "grad_norm": 0.09059862792491913,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 81417
+    },
+    {
+      "epoch": 8.37137569401604,
+      "grad_norm": 0.05472108721733093,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 81420
+    },
+    {
+      "epoch": 8.371684145589143,
+      "grad_norm": 0.05902374908328056,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 81423
+    },
+    {
+      "epoch": 8.371992597162246,
+      "grad_norm": 0.16040004789829254,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 81426
+    },
+    {
+      "epoch": 8.372301048735348,
+      "grad_norm": 0.0819641575217247,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 81429
+    },
+    {
+      "epoch": 8.372609500308451,
+      "grad_norm": 0.06774242967367172,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 81432
+    },
+    {
+      "epoch": 8.372917951881554,
+      "grad_norm": 0.06819859147071838,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 81435
+    },
+    {
+      "epoch": 8.373226403454657,
+      "grad_norm": 0.06440560519695282,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 81438
+    },
+    {
+      "epoch": 8.37353485502776,
+      "grad_norm": 0.05694984272122383,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 81441
+    },
+    {
+      "epoch": 8.373843306600863,
+      "grad_norm": 0.060925718396902084,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 81444
+    },
+    {
+      "epoch": 8.374151758173968,
+      "grad_norm": 0.07545589655637741,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 81447
+    },
+    {
+      "epoch": 8.37446020974707,
+      "grad_norm": 0.05031956359744072,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 81450
+    },
+    {
+      "epoch": 8.374768661320173,
+      "grad_norm": 0.05578349158167839,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 81453
+    },
+    {
+      "epoch": 8.375077112893276,
+      "grad_norm": 0.04404652491211891,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 81456
+    },
+    {
+      "epoch": 8.375385564466379,
+      "grad_norm": 0.05373278632760048,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 81459
+    },
+    {
+      "epoch": 8.375694016039482,
+      "grad_norm": 0.10714561492204666,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 81462
+    },
+    {
+      "epoch": 8.376002467612585,
+      "grad_norm": 0.062353793531656265,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 81465
+    },
+    {
+      "epoch": 8.376310919185688,
+      "grad_norm": 0.10784858465194702,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 81468
+    },
+    {
+      "epoch": 8.37661937075879,
+      "grad_norm": 0.17207425832748413,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 81471
+    },
+    {
+      "epoch": 8.376927822331893,
+      "grad_norm": 0.06604339182376862,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 81474
+    },
+    {
+      "epoch": 8.377236273904996,
+      "grad_norm": 0.04885290563106537,
+      "learning_rate": 0.01,
+      "loss": 1.9859,
+      "step": 81477
+    },
+    {
+      "epoch": 8.3775447254781,
+      "grad_norm": 0.03585289046168327,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 81480
+    },
+    {
+      "epoch": 8.377853177051202,
+      "grad_norm": 0.055384352803230286,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 81483
+    },
+    {
+      "epoch": 8.378161628624307,
+      "grad_norm": 0.055960532277822495,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 81486
+    },
+    {
+      "epoch": 8.37847008019741,
+      "grad_norm": 0.04403486102819443,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 81489
+    },
+    {
+      "epoch": 8.378778531770513,
+      "grad_norm": 0.057822804898023605,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 81492
+    },
+    {
+      "epoch": 8.379086983343615,
+      "grad_norm": 0.12054527550935745,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 81495
+    },
+    {
+      "epoch": 8.379395434916718,
+      "grad_norm": 0.04777431860566139,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 81498
+    },
+    {
+      "epoch": 8.379703886489821,
+      "grad_norm": 0.10239491611719131,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 81501
+    },
+    {
+      "epoch": 8.380012338062924,
+      "grad_norm": 0.04464276507496834,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 81504
+    },
+    {
+      "epoch": 8.380320789636027,
+      "grad_norm": 0.09886369854211807,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 81507
+    },
+    {
+      "epoch": 8.38062924120913,
+      "grad_norm": 0.054858770221471786,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 81510
+    },
+    {
+      "epoch": 8.380937692782233,
+      "grad_norm": 0.05009688064455986,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 81513
+    },
+    {
+      "epoch": 8.381246144355336,
+      "grad_norm": 0.11478911340236664,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 81516
+    },
+    {
+      "epoch": 8.381554595928439,
+      "grad_norm": 0.09376154839992523,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 81519
+    },
+    {
+      "epoch": 8.381863047501541,
+      "grad_norm": 0.049762289971113205,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 81522
+    },
+    {
+      "epoch": 8.382171499074646,
+      "grad_norm": 0.09793833643198013,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 81525
+    },
+    {
+      "epoch": 8.382479950647749,
+      "grad_norm": 0.03599982336163521,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 81528
+    },
+    {
+      "epoch": 8.382788402220852,
+      "grad_norm": 0.06792202591896057,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 81531
+    },
+    {
+      "epoch": 8.383096853793955,
+      "grad_norm": 0.051511526107788086,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 81534
+    },
+    {
+      "epoch": 8.383405305367058,
+      "grad_norm": 0.04394154995679855,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 81537
+    },
+    {
+      "epoch": 8.38371375694016,
+      "grad_norm": 0.05700390040874481,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 81540
+    },
+    {
+      "epoch": 8.384022208513263,
+      "grad_norm": 0.03976348415017128,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 81543
+    },
+    {
+      "epoch": 8.384330660086366,
+      "grad_norm": 0.06435202807188034,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 81546
+    },
+    {
+      "epoch": 8.38463911165947,
+      "grad_norm": 0.13295422494411469,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 81549
+    },
+    {
+      "epoch": 8.384947563232572,
+      "grad_norm": 0.06909388303756714,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 81552
+    },
+    {
+      "epoch": 8.385256014805675,
+      "grad_norm": 0.13283444941043854,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 81555
+    },
+    {
+      "epoch": 8.385564466378778,
+      "grad_norm": 0.07221397757530212,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 81558
+    },
+    {
+      "epoch": 8.38587291795188,
+      "grad_norm": 0.05165419727563858,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 81561
+    },
+    {
+      "epoch": 8.386181369524985,
+      "grad_norm": 0.04133574664592743,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 81564
+    },
+    {
+      "epoch": 8.386489821098088,
+      "grad_norm": 0.05390239879488945,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 81567
+    },
+    {
+      "epoch": 8.386798272671191,
+      "grad_norm": 0.043544042855501175,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 81570
+    },
+    {
+      "epoch": 8.387106724244294,
+      "grad_norm": 0.07669848203659058,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 81573
+    },
+    {
+      "epoch": 8.387415175817397,
+      "grad_norm": 0.06485964357852936,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 81576
+    },
+    {
+      "epoch": 8.3877236273905,
+      "grad_norm": 0.09705982357263565,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 81579
+    },
+    {
+      "epoch": 8.388032078963603,
+      "grad_norm": 0.04892526939511299,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 81582
+    },
+    {
+      "epoch": 8.388340530536706,
+      "grad_norm": 0.041197143495082855,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 81585
+    },
+    {
+      "epoch": 8.388648982109808,
+      "grad_norm": 0.06936054676771164,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 81588
+    },
+    {
+      "epoch": 8.388957433682911,
+      "grad_norm": 0.09122376143932343,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 81591
+    },
+    {
+      "epoch": 8.389265885256014,
+      "grad_norm": 0.05180079862475395,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 81594
+    },
+    {
+      "epoch": 8.389574336829117,
+      "grad_norm": 0.055592991411685944,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 81597
+    },
+    {
+      "epoch": 8.38988278840222,
+      "grad_norm": 0.049899131059646606,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 81600
+    },
+    {
+      "epoch": 8.390191239975325,
+      "grad_norm": 0.04033705219626427,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 81603
+    },
+    {
+      "epoch": 8.390499691548428,
+      "grad_norm": 0.03256101533770561,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 81606
+    },
+    {
+      "epoch": 8.39080814312153,
+      "grad_norm": 0.12379596382379532,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 81609
+    },
+    {
+      "epoch": 8.391116594694633,
+      "grad_norm": 0.058602914214134216,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 81612
+    },
+    {
+      "epoch": 8.391425046267736,
+      "grad_norm": 0.03897072747349739,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 81615
+    },
+    {
+      "epoch": 8.391733497840839,
+      "grad_norm": 0.05257393419742584,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 81618
+    },
+    {
+      "epoch": 8.392041949413942,
+      "grad_norm": 0.03705655410885811,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 81621
+    },
+    {
+      "epoch": 8.392350400987045,
+      "grad_norm": 0.03860629349946976,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 81624
+    },
+    {
+      "epoch": 8.392658852560148,
+      "grad_norm": 0.05735693871974945,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 81627
+    },
+    {
+      "epoch": 8.39296730413325,
+      "grad_norm": 0.0472671315073967,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 81630
+    },
+    {
+      "epoch": 8.393275755706354,
+      "grad_norm": 0.04570858180522919,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 81633
+    },
+    {
+      "epoch": 8.393584207279456,
+      "grad_norm": 0.12013877183198929,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 81636
+    },
+    {
+      "epoch": 8.39389265885256,
+      "grad_norm": 0.1255820095539093,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 81639
+    },
+    {
+      "epoch": 8.394201110425664,
+      "grad_norm": 0.045648593455553055,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 81642
+    },
+    {
+      "epoch": 8.394509561998767,
+      "grad_norm": 0.03392338007688522,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 81645
+    },
+    {
+      "epoch": 8.39481801357187,
+      "grad_norm": 0.03878732770681381,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 81648
+    },
+    {
+      "epoch": 8.395126465144973,
+      "grad_norm": 0.05559149011969566,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 81651
+    },
+    {
+      "epoch": 8.395434916718076,
+      "grad_norm": 0.044413961470127106,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 81654
+    },
+    {
+      "epoch": 8.395743368291178,
+      "grad_norm": 0.0495944581925869,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 81657
+    },
+    {
+      "epoch": 8.396051819864281,
+      "grad_norm": 0.0679146870970726,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 81660
+    },
+    {
+      "epoch": 8.396360271437384,
+      "grad_norm": 0.10142147541046143,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 81663
+    },
+    {
+      "epoch": 8.396668723010487,
+      "grad_norm": 0.056064870208501816,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 81666
+    },
+    {
+      "epoch": 8.39697717458359,
+      "grad_norm": 0.09618430584669113,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 81669
+    },
+    {
+      "epoch": 8.397285626156693,
+      "grad_norm": 0.04978022351861,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 81672
+    },
+    {
+      "epoch": 8.397594077729796,
+      "grad_norm": 0.05624723434448242,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 81675
+    },
+    {
+      "epoch": 8.397902529302899,
+      "grad_norm": 0.09562240540981293,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 81678
+    },
+    {
+      "epoch": 8.398210980876003,
+      "grad_norm": 0.045868344604969025,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 81681
+    },
+    {
+      "epoch": 8.398519432449106,
+      "grad_norm": 0.07864541560411453,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 81684
+    },
+    {
+      "epoch": 8.398827884022209,
+      "grad_norm": 0.106157585978508,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 81687
+    },
+    {
+      "epoch": 8.399136335595312,
+      "grad_norm": 0.057111166417598724,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 81690
+    },
+    {
+      "epoch": 8.399444787168415,
+      "grad_norm": 0.10949339717626572,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 81693
+    },
+    {
+      "epoch": 8.399753238741518,
+      "grad_norm": 0.05869073048233986,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 81696
+    },
+    {
+      "epoch": 8.40006169031462,
+      "grad_norm": 0.033000875264406204,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 81699
+    },
+    {
+      "epoch": 8.400370141887723,
+      "grad_norm": 0.04507824033498764,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 81702
+    },
+    {
+      "epoch": 8.400678593460826,
+      "grad_norm": 0.04142510145902634,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 81705
+    },
+    {
+      "epoch": 8.40098704503393,
+      "grad_norm": 0.0351688452064991,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 81708
+    },
+    {
+      "epoch": 8.401295496607032,
+      "grad_norm": 0.039388950914144516,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 81711
+    },
+    {
+      "epoch": 8.401603948180135,
+      "grad_norm": 0.0492679700255394,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 81714
+    },
+    {
+      "epoch": 8.401912399753238,
+      "grad_norm": 0.11415845900774002,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 81717
+    },
+    {
+      "epoch": 8.402220851326343,
+      "grad_norm": 0.06980131566524506,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 81720
+    },
+    {
+      "epoch": 8.402529302899445,
+      "grad_norm": 0.10816799849271774,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 81723
+    },
+    {
+      "epoch": 8.402837754472548,
+      "grad_norm": 0.042601246386766434,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 81726
+    },
+    {
+      "epoch": 8.403146206045651,
+      "grad_norm": 0.040994688868522644,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 81729
+    },
+    {
+      "epoch": 8.403454657618754,
+      "grad_norm": 0.03924047201871872,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 81732
+    },
+    {
+      "epoch": 8.403763109191857,
+      "grad_norm": 0.045216597616672516,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 81735
+    },
+    {
+      "epoch": 8.40407156076496,
+      "grad_norm": 0.05718405172228813,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 81738
+    },
+    {
+      "epoch": 8.404380012338063,
+      "grad_norm": 0.06921820342540741,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 81741
+    },
+    {
+      "epoch": 8.404688463911166,
+      "grad_norm": 0.07314276695251465,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 81744
+    },
+    {
+      "epoch": 8.404996915484269,
+      "grad_norm": 0.09398096799850464,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 81747
+    },
+    {
+      "epoch": 8.405305367057371,
+      "grad_norm": 0.0631139725446701,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 81750
+    },
+    {
+      "epoch": 8.405613818630474,
+      "grad_norm": 0.07777784019708633,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 81753
+    },
+    {
+      "epoch": 8.405922270203577,
+      "grad_norm": 0.05013573169708252,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 81756
+    },
+    {
+      "epoch": 8.406230721776682,
+      "grad_norm": 0.12393694370985031,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 81759
+    },
+    {
+      "epoch": 8.406539173349785,
+      "grad_norm": 0.04600420221686363,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 81762
+    },
+    {
+      "epoch": 8.406847624922888,
+      "grad_norm": 0.09453027695417404,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 81765
+    },
+    {
+      "epoch": 8.40715607649599,
+      "grad_norm": 0.08567167818546295,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 81768
+    },
+    {
+      "epoch": 8.407464528069093,
+      "grad_norm": 0.06742667406797409,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 81771
+    },
+    {
+      "epoch": 8.407772979642196,
+      "grad_norm": 0.03893236815929413,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 81774
+    },
+    {
+      "epoch": 8.4080814312153,
+      "grad_norm": 0.04947619512677193,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 81777
+    },
+    {
+      "epoch": 8.408389882788402,
+      "grad_norm": 0.03255341574549675,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 81780
+    },
+    {
+      "epoch": 8.408698334361505,
+      "grad_norm": 0.04050638526678085,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 81783
+    },
+    {
+      "epoch": 8.409006785934608,
+      "grad_norm": 0.08423347026109695,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 81786
+    },
+    {
+      "epoch": 8.40931523750771,
+      "grad_norm": 0.0701289102435112,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 81789
+    },
+    {
+      "epoch": 8.409623689080814,
+      "grad_norm": 0.09731067717075348,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 81792
+    },
+    {
+      "epoch": 8.409932140653916,
+      "grad_norm": 0.07500086724758148,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 81795
+    },
+    {
+      "epoch": 8.410240592227021,
+      "grad_norm": 0.07082207500934601,
+      "learning_rate": 0.01,
+      "loss": 1.988,
+      "step": 81798
+    },
+    {
+      "epoch": 8.410549043800124,
+      "grad_norm": 0.08339577168226242,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 81801
+    },
+    {
+      "epoch": 8.410857495373227,
+      "grad_norm": 0.04280835762619972,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 81804
+    },
+    {
+      "epoch": 8.41116594694633,
+      "grad_norm": 0.07703588902950287,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 81807
+    },
+    {
+      "epoch": 8.411474398519433,
+      "grad_norm": 0.06267133355140686,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 81810
+    },
+    {
+      "epoch": 8.411782850092536,
+      "grad_norm": 0.07680845260620117,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 81813
+    },
+    {
+      "epoch": 8.412091301665638,
+      "grad_norm": 0.08015266060829163,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 81816
+    },
+    {
+      "epoch": 8.412399753238741,
+      "grad_norm": 0.06529653072357178,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 81819
+    },
+    {
+      "epoch": 8.412708204811844,
+      "grad_norm": 0.042141351848840714,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 81822
+    },
+    {
+      "epoch": 8.413016656384947,
+      "grad_norm": 0.04604843631386757,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 81825
+    },
+    {
+      "epoch": 8.41332510795805,
+      "grad_norm": 0.08277709037065506,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 81828
+    },
+    {
+      "epoch": 8.413633559531153,
+      "grad_norm": 0.038434263318777084,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 81831
+    },
+    {
+      "epoch": 8.413942011104258,
+      "grad_norm": 0.05674101784825325,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 81834
+    },
+    {
+      "epoch": 8.41425046267736,
+      "grad_norm": 0.037434451282024384,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 81837
+    },
+    {
+      "epoch": 8.414558914250463,
+      "grad_norm": 0.14585204422473907,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 81840
+    },
+    {
+      "epoch": 8.414867365823566,
+      "grad_norm": 0.12854383885860443,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 81843
+    },
+    {
+      "epoch": 8.415175817396669,
+      "grad_norm": 0.09817156195640564,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 81846
+    },
+    {
+      "epoch": 8.415484268969772,
+      "grad_norm": 0.07401823252439499,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 81849
+    },
+    {
+      "epoch": 8.415792720542875,
+      "grad_norm": 0.038704339414834976,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 81852
+    },
+    {
+      "epoch": 8.416101172115978,
+      "grad_norm": 0.0486832819879055,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 81855
+    },
+    {
+      "epoch": 8.41640962368908,
+      "grad_norm": 0.04595676064491272,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 81858
+    },
+    {
+      "epoch": 8.416718075262184,
+      "grad_norm": 0.05181482061743736,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 81861
+    },
+    {
+      "epoch": 8.417026526835286,
+      "grad_norm": 0.05815410986542702,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 81864
+    },
+    {
+      "epoch": 8.41733497840839,
+      "grad_norm": 0.08507470041513443,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 81867
+    },
+    {
+      "epoch": 8.417643429981492,
+      "grad_norm": 0.056439630687236786,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 81870
+    },
+    {
+      "epoch": 8.417951881554597,
+      "grad_norm": 0.037034884095191956,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 81873
+    },
+    {
+      "epoch": 8.4182603331277,
+      "grad_norm": 0.0958690196275711,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 81876
+    },
+    {
+      "epoch": 8.418568784700803,
+      "grad_norm": 0.1169692724943161,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 81879
+    },
+    {
+      "epoch": 8.418877236273905,
+      "grad_norm": 0.06900906562805176,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 81882
+    },
+    {
+      "epoch": 8.419185687847008,
+      "grad_norm": 0.09297920763492584,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 81885
+    },
+    {
+      "epoch": 8.419494139420111,
+      "grad_norm": 0.08288362622261047,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 81888
+    },
+    {
+      "epoch": 8.419802590993214,
+      "grad_norm": 0.09441418945789337,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 81891
+    },
+    {
+      "epoch": 8.420111042566317,
+      "grad_norm": 0.058979861438274384,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 81894
+    },
+    {
+      "epoch": 8.42041949413942,
+      "grad_norm": 0.053389038890600204,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 81897
+    },
+    {
+      "epoch": 8.420727945712523,
+      "grad_norm": 0.05778560787439346,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 81900
+    },
+    {
+      "epoch": 8.421036397285626,
+      "grad_norm": 0.0623948872089386,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 81903
+    },
+    {
+      "epoch": 8.421344848858729,
+      "grad_norm": 0.04240475967526436,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 81906
+    },
+    {
+      "epoch": 8.421653300431831,
+      "grad_norm": 0.07669229060411453,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 81909
+    },
+    {
+      "epoch": 8.421961752004936,
+      "grad_norm": 0.11650708317756653,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 81912
+    },
+    {
+      "epoch": 8.422270203578039,
+      "grad_norm": 0.13154225051403046,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 81915
+    },
+    {
+      "epoch": 8.422578655151142,
+      "grad_norm": 0.07503214478492737,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 81918
+    },
+    {
+      "epoch": 8.422887106724245,
+      "grad_norm": 0.04421877861022949,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 81921
+    },
+    {
+      "epoch": 8.423195558297348,
+      "grad_norm": 0.03446371480822563,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 81924
+    },
+    {
+      "epoch": 8.42350400987045,
+      "grad_norm": 0.03708928823471069,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 81927
+    },
+    {
+      "epoch": 8.423812461443553,
+      "grad_norm": 0.032508306205272675,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 81930
+    },
+    {
+      "epoch": 8.424120913016656,
+      "grad_norm": 0.05352804809808731,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 81933
+    },
+    {
+      "epoch": 8.42442936458976,
+      "grad_norm": 0.15011733770370483,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 81936
+    },
+    {
+      "epoch": 8.424737816162862,
+      "grad_norm": 0.07135441154241562,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 81939
+    },
+    {
+      "epoch": 8.425046267735965,
+      "grad_norm": 0.05304455757141113,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 81942
+    },
+    {
+      "epoch": 8.425354719309068,
+      "grad_norm": 0.0310453400015831,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 81945
+    },
+    {
+      "epoch": 8.42566317088217,
+      "grad_norm": 0.03331723436713219,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 81948
+    },
+    {
+      "epoch": 8.425971622455275,
+      "grad_norm": 0.13005290925502777,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 81951
+    },
+    {
+      "epoch": 8.426280074028378,
+      "grad_norm": 0.09037820249795914,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 81954
+    },
+    {
+      "epoch": 8.426588525601481,
+      "grad_norm": 0.05044369027018547,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 81957
+    },
+    {
+      "epoch": 8.426896977174584,
+      "grad_norm": 0.06943196058273315,
+      "learning_rate": 0.01,
+      "loss": 1.989,
+      "step": 81960
+    },
+    {
+      "epoch": 8.427205428747687,
+      "grad_norm": 0.07625371217727661,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 81963
+    },
+    {
+      "epoch": 8.42751388032079,
+      "grad_norm": 0.046040359884500504,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 81966
+    },
+    {
+      "epoch": 8.427822331893893,
+      "grad_norm": 0.03187291696667671,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 81969
+    },
+    {
+      "epoch": 8.428130783466996,
+      "grad_norm": 0.030660714954137802,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 81972
+    },
+    {
+      "epoch": 8.428439235040099,
+      "grad_norm": 0.05950603634119034,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 81975
+    },
+    {
+      "epoch": 8.428747686613201,
+      "grad_norm": 0.13979393243789673,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 81978
+    },
+    {
+      "epoch": 8.429056138186304,
+      "grad_norm": 0.10096840560436249,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 81981
+    },
+    {
+      "epoch": 8.429364589759407,
+      "grad_norm": 0.070526123046875,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 81984
+    },
+    {
+      "epoch": 8.42967304133251,
+      "grad_norm": 0.047776173800230026,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 81987
+    },
+    {
+      "epoch": 8.429981492905615,
+      "grad_norm": 0.03126342594623566,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 81990
+    },
+    {
+      "epoch": 8.430289944478718,
+      "grad_norm": 0.1021898165345192,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 81993
+    },
+    {
+      "epoch": 8.43059839605182,
+      "grad_norm": 0.0435447096824646,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 81996
+    },
+    {
+      "epoch": 8.430906847624923,
+      "grad_norm": 0.09668678045272827,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 81999
+    },
+    {
+      "epoch": 8.431215299198026,
+      "grad_norm": 0.03794991225004196,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 82002
+    },
+    {
+      "epoch": 8.43152375077113,
+      "grad_norm": 0.044686999171972275,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 82005
+    },
+    {
+      "epoch": 8.431832202344232,
+      "grad_norm": 0.03441571816802025,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 82008
+    },
+    {
+      "epoch": 8.432140653917335,
+      "grad_norm": 0.0457640215754509,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 82011
+    },
+    {
+      "epoch": 8.432449105490438,
+      "grad_norm": 0.06535539031028748,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 82014
+    },
+    {
+      "epoch": 8.43275755706354,
+      "grad_norm": 0.06897538155317307,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 82017
+    },
+    {
+      "epoch": 8.433066008636644,
+      "grad_norm": 0.07700546830892563,
+      "learning_rate": 0.01,
+      "loss": 1.9769,
+      "step": 82020
+    },
+    {
+      "epoch": 8.433374460209746,
+      "grad_norm": 0.07638740539550781,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 82023
+    },
+    {
+      "epoch": 8.43368291178285,
+      "grad_norm": 0.08059564232826233,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 82026
+    },
+    {
+      "epoch": 8.433991363355954,
+      "grad_norm": 0.09341084212064743,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 82029
+    },
+    {
+      "epoch": 8.434299814929057,
+      "grad_norm": 0.05106695368885994,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 82032
+    },
+    {
+      "epoch": 8.43460826650216,
+      "grad_norm": 0.05583924800157547,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 82035
+    },
+    {
+      "epoch": 8.434916718075263,
+      "grad_norm": 0.05295472964644432,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 82038
+    },
+    {
+      "epoch": 8.435225169648366,
+      "grad_norm": 0.12497080117464066,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 82041
+    },
+    {
+      "epoch": 8.435533621221468,
+      "grad_norm": 0.08089610189199448,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 82044
+    },
+    {
+      "epoch": 8.435842072794571,
+      "grad_norm": 0.0737653374671936,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 82047
+    },
+    {
+      "epoch": 8.436150524367674,
+      "grad_norm": 0.04853172227740288,
+      "learning_rate": 0.01,
+      "loss": 1.99,
+      "step": 82050
+    },
+    {
+      "epoch": 8.436458975940777,
+      "grad_norm": 0.032473742961883545,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 82053
+    },
+    {
+      "epoch": 8.43676742751388,
+      "grad_norm": 0.03237362578511238,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 82056
+    },
+    {
+      "epoch": 8.437075879086983,
+      "grad_norm": 0.03169579803943634,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 82059
+    },
+    {
+      "epoch": 8.437384330660086,
+      "grad_norm": 0.08158145844936371,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 82062
+    },
+    {
+      "epoch": 8.437692782233189,
+      "grad_norm": 0.11348774284124374,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 82065
+    },
+    {
+      "epoch": 8.438001233806293,
+      "grad_norm": 0.052066583186388016,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 82068
+    },
+    {
+      "epoch": 8.438309685379396,
+      "grad_norm": 0.05296801030635834,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 82071
+    },
+    {
+      "epoch": 8.438618136952499,
+      "grad_norm": 0.07596609741449356,
+      "learning_rate": 0.01,
+      "loss": 1.9749,
+      "step": 82074
+    },
+    {
+      "epoch": 8.438926588525602,
+      "grad_norm": 0.057574667036533356,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 82077
+    },
+    {
+      "epoch": 8.439235040098705,
+      "grad_norm": 0.037525586783885956,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 82080
+    },
+    {
+      "epoch": 8.439543491671808,
+      "grad_norm": 0.05512101203203201,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 82083
+    },
+    {
+      "epoch": 8.43985194324491,
+      "grad_norm": 0.09372948855161667,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 82086
+    },
+    {
+      "epoch": 8.440160394818014,
+      "grad_norm": 0.08818025887012482,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 82089
+    },
+    {
+      "epoch": 8.440468846391116,
+      "grad_norm": 0.06324193626642227,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 82092
+    },
+    {
+      "epoch": 8.44077729796422,
+      "grad_norm": 0.0913136675953865,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 82095
+    },
+    {
+      "epoch": 8.441085749537322,
+      "grad_norm": 0.05476130172610283,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 82098
+    },
+    {
+      "epoch": 8.441394201110425,
+      "grad_norm": 0.12479494512081146,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 82101
+    },
+    {
+      "epoch": 8.441702652683528,
+      "grad_norm": 0.07040181756019592,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 82104
+    },
+    {
+      "epoch": 8.442011104256633,
+      "grad_norm": 0.04941793903708458,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 82107
+    },
+    {
+      "epoch": 8.442319555829735,
+      "grad_norm": 0.03700859844684601,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 82110
+    },
+    {
+      "epoch": 8.442628007402838,
+      "grad_norm": 0.04570907726883888,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 82113
+    },
+    {
+      "epoch": 8.442936458975941,
+      "grad_norm": 0.03453392907977104,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 82116
+    },
+    {
+      "epoch": 8.443244910549044,
+      "grad_norm": 0.05497143417596817,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 82119
+    },
+    {
+      "epoch": 8.443553362122147,
+      "grad_norm": 0.13390731811523438,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 82122
+    },
+    {
+      "epoch": 8.44386181369525,
+      "grad_norm": 0.08949720859527588,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 82125
+    },
+    {
+      "epoch": 8.444170265268353,
+      "grad_norm": 0.08513578027486801,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 82128
+    },
+    {
+      "epoch": 8.444478716841456,
+      "grad_norm": 0.06023092195391655,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 82131
+    },
+    {
+      "epoch": 8.444787168414559,
+      "grad_norm": 0.03851139172911644,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 82134
+    },
+    {
+      "epoch": 8.445095619987661,
+      "grad_norm": 0.04166555404663086,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 82137
+    },
+    {
+      "epoch": 8.445404071560764,
+      "grad_norm": 0.09604350477457047,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 82140
+    },
+    {
+      "epoch": 8.445712523133867,
+      "grad_norm": 0.10311726480722427,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 82143
+    },
+    {
+      "epoch": 8.446020974706972,
+      "grad_norm": 0.07740923017263412,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 82146
+    },
+    {
+      "epoch": 8.446329426280075,
+      "grad_norm": 0.08379843086004257,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 82149
+    },
+    {
+      "epoch": 8.446637877853178,
+      "grad_norm": 0.11364433914422989,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 82152
+    },
+    {
+      "epoch": 8.44694632942628,
+      "grad_norm": 0.05542511120438576,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 82155
+    },
+    {
+      "epoch": 8.447254780999383,
+      "grad_norm": 0.09093154221773148,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 82158
+    },
+    {
+      "epoch": 8.447563232572486,
+      "grad_norm": 0.034466832876205444,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 82161
+    },
+    {
+      "epoch": 8.44787168414559,
+      "grad_norm": 0.03495443984866142,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 82164
+    },
+    {
+      "epoch": 8.448180135718692,
+      "grad_norm": 0.09456860274076462,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 82167
+    },
+    {
+      "epoch": 8.448488587291795,
+      "grad_norm": 0.04304103925824165,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 82170
+    },
+    {
+      "epoch": 8.448797038864898,
+      "grad_norm": 0.06458231806755066,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 82173
+    },
+    {
+      "epoch": 8.449105490438,
+      "grad_norm": 0.0449569895863533,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 82176
+    },
+    {
+      "epoch": 8.449413942011104,
+      "grad_norm": 0.12885241210460663,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 82179
+    },
+    {
+      "epoch": 8.449722393584207,
+      "grad_norm": 0.044088032096624374,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 82182
+    },
+    {
+      "epoch": 8.450030845157311,
+      "grad_norm": 0.09161775559186935,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 82185
+    },
+    {
+      "epoch": 8.450339296730414,
+      "grad_norm": 0.03646433353424072,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 82188
+    },
+    {
+      "epoch": 8.450647748303517,
+      "grad_norm": 0.12289903312921524,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 82191
+    },
+    {
+      "epoch": 8.45095619987662,
+      "grad_norm": 0.06749970465898514,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 82194
+    },
+    {
+      "epoch": 8.451264651449723,
+      "grad_norm": 0.11313161998987198,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 82197
+    },
+    {
+      "epoch": 8.451573103022826,
+      "grad_norm": 0.04492225497961044,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 82200
+    },
+    {
+      "epoch": 8.451881554595928,
+      "grad_norm": 0.042263831943273544,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 82203
+    },
+    {
+      "epoch": 8.452190006169031,
+      "grad_norm": 0.029859155416488647,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 82206
+    },
+    {
+      "epoch": 8.452498457742134,
+      "grad_norm": 0.0463617742061615,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 82209
+    },
+    {
+      "epoch": 8.452806909315237,
+      "grad_norm": 0.1206277385354042,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 82212
+    },
+    {
+      "epoch": 8.45311536088834,
+      "grad_norm": 0.05199189484119415,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 82215
+    },
+    {
+      "epoch": 8.453423812461443,
+      "grad_norm": 0.08049193769693375,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 82218
+    },
+    {
+      "epoch": 8.453732264034546,
+      "grad_norm": 0.07479532808065414,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 82221
+    },
+    {
+      "epoch": 8.45404071560765,
+      "grad_norm": 0.0769563689827919,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 82224
+    },
+    {
+      "epoch": 8.454349167180753,
+      "grad_norm": 0.09309972077608109,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 82227
+    },
+    {
+      "epoch": 8.454657618753856,
+      "grad_norm": 0.040779951959848404,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 82230
+    },
+    {
+      "epoch": 8.45496607032696,
+      "grad_norm": 0.060668691992759705,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 82233
+    },
+    {
+      "epoch": 8.455274521900062,
+      "grad_norm": 0.09466727823019028,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 82236
+    },
+    {
+      "epoch": 8.455582973473165,
+      "grad_norm": 0.08183979243040085,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 82239
+    },
+    {
+      "epoch": 8.455891425046268,
+      "grad_norm": 0.10821831226348877,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 82242
+    },
+    {
+      "epoch": 8.45619987661937,
+      "grad_norm": 0.07753495126962662,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 82245
+    },
+    {
+      "epoch": 8.456508328192474,
+      "grad_norm": 0.06566669046878815,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 82248
+    },
+    {
+      "epoch": 8.456816779765576,
+      "grad_norm": 0.043311357498168945,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 82251
+    },
+    {
+      "epoch": 8.45712523133868,
+      "grad_norm": 0.06685193628072739,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 82254
+    },
+    {
+      "epoch": 8.457433682911782,
+      "grad_norm": 0.050073493272066116,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 82257
+    },
+    {
+      "epoch": 8.457742134484885,
+      "grad_norm": 0.045595601201057434,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 82260
+    },
+    {
+      "epoch": 8.45805058605799,
+      "grad_norm": 0.11144230514764786,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 82263
+    },
+    {
+      "epoch": 8.458359037631093,
+      "grad_norm": 0.09206417948007584,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 82266
+    },
+    {
+      "epoch": 8.458667489204196,
+      "grad_norm": 0.03726569563150406,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 82269
+    },
+    {
+      "epoch": 8.458975940777298,
+      "grad_norm": 0.039987582713365555,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 82272
+    },
+    {
+      "epoch": 8.459284392350401,
+      "grad_norm": 0.09079570323228836,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 82275
+    },
+    {
+      "epoch": 8.459592843923504,
+      "grad_norm": 0.0509127639234066,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 82278
+    },
+    {
+      "epoch": 8.459901295496607,
+      "grad_norm": 0.09340399503707886,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 82281
+    },
+    {
+      "epoch": 8.46020974706971,
+      "grad_norm": 0.12762415409088135,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 82284
+    },
+    {
+      "epoch": 8.460518198642813,
+      "grad_norm": 0.12151426821947098,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 82287
+    },
+    {
+      "epoch": 8.460826650215916,
+      "grad_norm": 0.04861297830939293,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 82290
+    },
+    {
+      "epoch": 8.461135101789019,
+      "grad_norm": 0.08445765823125839,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 82293
+    },
+    {
+      "epoch": 8.461443553362122,
+      "grad_norm": 0.06112569943070412,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 82296
+    },
+    {
+      "epoch": 8.461752004935224,
+      "grad_norm": 0.04099706560373306,
+      "learning_rate": 0.01,
+      "loss": 1.9777,
+      "step": 82299
+    },
+    {
+      "epoch": 8.462060456508329,
+      "grad_norm": 0.049014266580343246,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 82302
+    },
+    {
+      "epoch": 8.462368908081432,
+      "grad_norm": 0.05799461528658867,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 82305
+    },
+    {
+      "epoch": 8.462677359654535,
+      "grad_norm": 0.08514149487018585,
+      "learning_rate": 0.01,
+      "loss": 1.9999,
+      "step": 82308
+    },
+    {
+      "epoch": 8.462985811227638,
+      "grad_norm": 0.07080193608999252,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 82311
+    },
+    {
+      "epoch": 8.46329426280074,
+      "grad_norm": 0.08238495886325836,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 82314
+    },
+    {
+      "epoch": 8.463602714373843,
+      "grad_norm": 0.05924540013074875,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 82317
+    },
+    {
+      "epoch": 8.463911165946946,
+      "grad_norm": 0.0707138329744339,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 82320
+    },
+    {
+      "epoch": 8.46421961752005,
+      "grad_norm": 0.12265108525753021,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 82323
+    },
+    {
+      "epoch": 8.464528069093152,
+      "grad_norm": 0.03225354850292206,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 82326
+    },
+    {
+      "epoch": 8.464836520666255,
+      "grad_norm": 0.034755658358335495,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 82329
+    },
+    {
+      "epoch": 8.465144972239358,
+      "grad_norm": 0.05887124314904213,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 82332
+    },
+    {
+      "epoch": 8.46545342381246,
+      "grad_norm": 0.09077783674001694,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 82335
+    },
+    {
+      "epoch": 8.465761875385564,
+      "grad_norm": 0.06020389124751091,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 82338
+    },
+    {
+      "epoch": 8.466070326958668,
+      "grad_norm": 0.09305040538311005,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 82341
+    },
+    {
+      "epoch": 8.466378778531771,
+      "grad_norm": 0.06462978571653366,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 82344
+    },
+    {
+      "epoch": 8.466687230104874,
+      "grad_norm": 0.0358070433139801,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 82347
+    },
+    {
+      "epoch": 8.466995681677977,
+      "grad_norm": 0.038024164736270905,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 82350
+    },
+    {
+      "epoch": 8.46730413325108,
+      "grad_norm": 0.06127464398741722,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 82353
+    },
+    {
+      "epoch": 8.467612584824183,
+      "grad_norm": 0.051408492028713226,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 82356
+    },
+    {
+      "epoch": 8.467921036397286,
+      "grad_norm": 0.08558293431997299,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 82359
+    },
+    {
+      "epoch": 8.468229487970389,
+      "grad_norm": 0.06267976760864258,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 82362
+    },
+    {
+      "epoch": 8.468537939543491,
+      "grad_norm": 0.12160573154687881,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 82365
+    },
+    {
+      "epoch": 8.468846391116594,
+      "grad_norm": 0.05924427509307861,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 82368
+    },
+    {
+      "epoch": 8.469154842689697,
+      "grad_norm": 0.06296840310096741,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 82371
+    },
+    {
+      "epoch": 8.4694632942628,
+      "grad_norm": 0.031919799745082855,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 82374
+    },
+    {
+      "epoch": 8.469771745835903,
+      "grad_norm": 0.03885785862803459,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 82377
+    },
+    {
+      "epoch": 8.470080197409008,
+      "grad_norm": 0.0466669462621212,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 82380
+    },
+    {
+      "epoch": 8.47038864898211,
+      "grad_norm": 0.04599282145500183,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 82383
+    },
+    {
+      "epoch": 8.470697100555213,
+      "grad_norm": 0.040090885013341904,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 82386
+    },
+    {
+      "epoch": 8.471005552128316,
+      "grad_norm": 0.08816850185394287,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 82389
+    },
+    {
+      "epoch": 8.47131400370142,
+      "grad_norm": 0.0854404866695404,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 82392
+    },
+    {
+      "epoch": 8.471622455274522,
+      "grad_norm": 0.045315563678741455,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 82395
+    },
+    {
+      "epoch": 8.471930906847625,
+      "grad_norm": 0.049634288996458054,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 82398
+    },
+    {
+      "epoch": 8.472239358420728,
+      "grad_norm": 0.058090757578611374,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 82401
+    },
+    {
+      "epoch": 8.47254780999383,
+      "grad_norm": 0.047061413526535034,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 82404
+    },
+    {
+      "epoch": 8.472856261566934,
+      "grad_norm": 0.04962011054158211,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 82407
+    },
+    {
+      "epoch": 8.473164713140036,
+      "grad_norm": 0.09965474903583527,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 82410
+    },
+    {
+      "epoch": 8.47347316471314,
+      "grad_norm": 0.05239838361740112,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 82413
+    },
+    {
+      "epoch": 8.473781616286242,
+      "grad_norm": 0.04054271802306175,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 82416
+    },
+    {
+      "epoch": 8.474090067859347,
+      "grad_norm": 0.04137061908841133,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 82419
+    },
+    {
+      "epoch": 8.47439851943245,
+      "grad_norm": 0.037401482462882996,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 82422
+    },
+    {
+      "epoch": 8.474706971005553,
+      "grad_norm": 0.08365823328495026,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 82425
+    },
+    {
+      "epoch": 8.475015422578656,
+      "grad_norm": 0.03423890843987465,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 82428
+    },
+    {
+      "epoch": 8.475323874151758,
+      "grad_norm": 0.04772539809346199,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 82431
+    },
+    {
+      "epoch": 8.475632325724861,
+      "grad_norm": 0.049579694867134094,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 82434
+    },
+    {
+      "epoch": 8.475940777297964,
+      "grad_norm": 0.06134011968970299,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 82437
+    },
+    {
+      "epoch": 8.476249228871067,
+      "grad_norm": 0.0950068011879921,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 82440
+    },
+    {
+      "epoch": 8.47655768044417,
+      "grad_norm": 0.07112467288970947,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 82443
+    },
+    {
+      "epoch": 8.476866132017273,
+      "grad_norm": 0.07818299531936646,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 82446
+    },
+    {
+      "epoch": 8.477174583590376,
+      "grad_norm": 0.09234732389450073,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 82449
+    },
+    {
+      "epoch": 8.477483035163479,
+      "grad_norm": 0.09724172949790955,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 82452
+    },
+    {
+      "epoch": 8.477791486736582,
+      "grad_norm": 0.16661600768566132,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 82455
+    },
+    {
+      "epoch": 8.478099938309686,
+      "grad_norm": 0.08794001489877701,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 82458
+    },
+    {
+      "epoch": 8.478408389882789,
+      "grad_norm": 0.03663558512926102,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 82461
+    },
+    {
+      "epoch": 8.478716841455892,
+      "grad_norm": 0.05669325217604637,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 82464
+    },
+    {
+      "epoch": 8.479025293028995,
+      "grad_norm": 0.10977914929389954,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 82467
+    },
+    {
+      "epoch": 8.479333744602098,
+      "grad_norm": 0.08488729596138,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 82470
+    },
+    {
+      "epoch": 8.4796421961752,
+      "grad_norm": 0.06477433443069458,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 82473
+    },
+    {
+      "epoch": 8.479950647748304,
+      "grad_norm": 0.07088177651166916,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 82476
+    },
+    {
+      "epoch": 8.480259099321406,
+      "grad_norm": 0.052466023713350296,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 82479
+    },
+    {
+      "epoch": 8.48056755089451,
+      "grad_norm": 0.039188675582408905,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 82482
+    },
+    {
+      "epoch": 8.480876002467612,
+      "grad_norm": 0.05050719901919365,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 82485
+    },
+    {
+      "epoch": 8.481184454040715,
+      "grad_norm": 0.04674654081463814,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 82488
+    },
+    {
+      "epoch": 8.481492905613818,
+      "grad_norm": 0.04279822111129761,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 82491
+    },
+    {
+      "epoch": 8.48180135718692,
+      "grad_norm": 0.1351974904537201,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 82494
+    },
+    {
+      "epoch": 8.482109808760026,
+      "grad_norm": 0.06532531976699829,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 82497
+    },
+    {
+      "epoch": 8.482418260333128,
+      "grad_norm": 0.10240834206342697,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 82500
+    },
+    {
+      "epoch": 8.482726711906231,
+      "grad_norm": 0.04963400214910507,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 82503
+    },
+    {
+      "epoch": 8.483035163479334,
+      "grad_norm": 0.07292640954256058,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 82506
+    },
+    {
+      "epoch": 8.483343615052437,
+      "grad_norm": 0.09584075212478638,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 82509
+    },
+    {
+      "epoch": 8.48365206662554,
+      "grad_norm": 0.051588594913482666,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 82512
+    },
+    {
+      "epoch": 8.483960518198643,
+      "grad_norm": 0.0447857603430748,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 82515
+    },
+    {
+      "epoch": 8.484268969771746,
+      "grad_norm": 0.04466879740357399,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 82518
+    },
+    {
+      "epoch": 8.484577421344849,
+      "grad_norm": 0.041589345782995224,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 82521
+    },
+    {
+      "epoch": 8.484885872917951,
+      "grad_norm": 0.1081608310341835,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 82524
+    },
+    {
+      "epoch": 8.485194324491054,
+      "grad_norm": 0.13699102401733398,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 82527
+    },
+    {
+      "epoch": 8.485502776064157,
+      "grad_norm": 0.0641539990901947,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 82530
+    },
+    {
+      "epoch": 8.48581122763726,
+      "grad_norm": 0.04673508554697037,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 82533
+    },
+    {
+      "epoch": 8.486119679210365,
+      "grad_norm": 0.04663087800145149,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 82536
+    },
+    {
+      "epoch": 8.486428130783468,
+      "grad_norm": 0.040559981018304825,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 82539
+    },
+    {
+      "epoch": 8.48673658235657,
+      "grad_norm": 0.03951507806777954,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 82542
+    },
+    {
+      "epoch": 8.487045033929673,
+      "grad_norm": 0.04637393727898598,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 82545
+    },
+    {
+      "epoch": 8.487353485502776,
+      "grad_norm": 0.09550820291042328,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 82548
+    },
+    {
+      "epoch": 8.48766193707588,
+      "grad_norm": 0.043463677167892456,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 82551
+    },
+    {
+      "epoch": 8.487970388648982,
+      "grad_norm": 0.04329133778810501,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 82554
+    },
+    {
+      "epoch": 8.488278840222085,
+      "grad_norm": 0.044975396245718,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 82557
+    },
+    {
+      "epoch": 8.488587291795188,
+      "grad_norm": 0.04665372148156166,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 82560
+    },
+    {
+      "epoch": 8.48889574336829,
+      "grad_norm": 0.06570030748844147,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 82563
+    },
+    {
+      "epoch": 8.489204194941394,
+      "grad_norm": 0.07015156745910645,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 82566
+    },
+    {
+      "epoch": 8.489512646514497,
+      "grad_norm": 0.11700818687677383,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 82569
+    },
+    {
+      "epoch": 8.4898210980876,
+      "grad_norm": 0.04150941222906113,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 82572
+    },
+    {
+      "epoch": 8.490129549660704,
+      "grad_norm": 0.057442814111709595,
+      "learning_rate": 0.01,
+      "loss": 1.9041,
+      "step": 82575
+    },
+    {
+      "epoch": 8.490438001233807,
+      "grad_norm": 0.05241570249199867,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 82578
+    },
+    {
+      "epoch": 8.49074645280691,
+      "grad_norm": 0.04771868884563446,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 82581
+    },
+    {
+      "epoch": 8.491054904380013,
+      "grad_norm": 0.038936082273721695,
+      "learning_rate": 0.01,
+      "loss": 1.9809,
+      "step": 82584
+    },
+    {
+      "epoch": 8.491363355953116,
+      "grad_norm": 0.05392014607787132,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 82587
+    },
+    {
+      "epoch": 8.491671807526219,
+      "grad_norm": 0.04587749391794205,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 82590
+    },
+    {
+      "epoch": 8.491980259099321,
+      "grad_norm": 0.0473514162003994,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 82593
+    },
+    {
+      "epoch": 8.492288710672424,
+      "grad_norm": 0.04507661983370781,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 82596
+    },
+    {
+      "epoch": 8.492597162245527,
+      "grad_norm": 0.06868678331375122,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 82599
+    },
+    {
+      "epoch": 8.49290561381863,
+      "grad_norm": 0.0536494106054306,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 82602
+    },
+    {
+      "epoch": 8.493214065391733,
+      "grad_norm": 0.06346127390861511,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 82605
+    },
+    {
+      "epoch": 8.493522516964836,
+      "grad_norm": 0.1364736407995224,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 82608
+    },
+    {
+      "epoch": 8.493830968537939,
+      "grad_norm": 0.1090131625533104,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 82611
+    },
+    {
+      "epoch": 8.494139420111043,
+      "grad_norm": 0.06526965647935867,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 82614
+    },
+    {
+      "epoch": 8.494447871684146,
+      "grad_norm": 0.03388083726167679,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 82617
+    },
+    {
+      "epoch": 8.49475632325725,
+      "grad_norm": 0.044526632875204086,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 82620
+    },
+    {
+      "epoch": 8.495064774830352,
+      "grad_norm": 0.05937231332063675,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 82623
+    },
+    {
+      "epoch": 8.495373226403455,
+      "grad_norm": 0.0789630115032196,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 82626
+    },
+    {
+      "epoch": 8.495681677976558,
+      "grad_norm": 0.10676635056734085,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 82629
+    },
+    {
+      "epoch": 8.49599012954966,
+      "grad_norm": 0.06259648501873016,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 82632
+    },
+    {
+      "epoch": 8.496298581122764,
+      "grad_norm": 0.10849929600954056,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 82635
+    },
+    {
+      "epoch": 8.496607032695866,
+      "grad_norm": 0.044495515525341034,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 82638
+    },
+    {
+      "epoch": 8.49691548426897,
+      "grad_norm": 0.09883397072553635,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 82641
+    },
+    {
+      "epoch": 8.497223935842072,
+      "grad_norm": 0.06384971737861633,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 82644
+    },
+    {
+      "epoch": 8.497532387415175,
+      "grad_norm": 0.09279752522706985,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 82647
+    },
+    {
+      "epoch": 8.497840838988278,
+      "grad_norm": 0.053611353039741516,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 82650
+    },
+    {
+      "epoch": 8.498149290561383,
+      "grad_norm": 0.05061632767319679,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 82653
+    },
+    {
+      "epoch": 8.498457742134486,
+      "grad_norm": 0.04062135890126228,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 82656
+    },
+    {
+      "epoch": 8.498766193707588,
+      "grad_norm": 0.05774087458848953,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 82659
+    },
+    {
+      "epoch": 8.499074645280691,
+      "grad_norm": 0.1062593087553978,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 82662
+    },
+    {
+      "epoch": 8.499383096853794,
+      "grad_norm": 0.058600399643182755,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 82665
+    },
+    {
+      "epoch": 8.499691548426897,
+      "grad_norm": 0.03446796536445618,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 82668
+    },
+    {
+      "epoch": 8.5,
+      "grad_norm": 0.05771895498037338,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 82671
+    },
+    {
+      "epoch": 8.500308451573103,
+      "grad_norm": 0.047667182981967926,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 82674
+    },
+    {
+      "epoch": 8.500616903146206,
+      "grad_norm": 0.05757983401417732,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 82677
+    },
+    {
+      "epoch": 8.500925354719309,
+      "grad_norm": 0.0631047859787941,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 82680
+    },
+    {
+      "epoch": 8.501233806292412,
+      "grad_norm": 0.10003430396318436,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 82683
+    },
+    {
+      "epoch": 8.501542257865514,
+      "grad_norm": 0.04452775418758392,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 82686
+    },
+    {
+      "epoch": 8.501850709438617,
+      "grad_norm": 0.07562670856714249,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 82689
+    },
+    {
+      "epoch": 8.502159161011722,
+      "grad_norm": 0.08108774572610855,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 82692
+    },
+    {
+      "epoch": 8.502467612584825,
+      "grad_norm": 0.07505381852388382,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 82695
+    },
+    {
+      "epoch": 8.502776064157928,
+      "grad_norm": 0.09649732708930969,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 82698
+    },
+    {
+      "epoch": 8.50308451573103,
+      "grad_norm": 0.05110400915145874,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 82701
+    },
+    {
+      "epoch": 8.503392967304134,
+      "grad_norm": 0.11231815814971924,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 82704
+    },
+    {
+      "epoch": 8.503701418877236,
+      "grad_norm": 0.033437538892030716,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 82707
+    },
+    {
+      "epoch": 8.50400987045034,
+      "grad_norm": 0.09465573728084564,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 82710
+    },
+    {
+      "epoch": 8.504318322023442,
+      "grad_norm": 0.047320276498794556,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 82713
+    },
+    {
+      "epoch": 8.504626773596545,
+      "grad_norm": 0.05426005274057388,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 82716
+    },
+    {
+      "epoch": 8.504935225169648,
+      "grad_norm": 0.1052740216255188,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 82719
+    },
+    {
+      "epoch": 8.50524367674275,
+      "grad_norm": 0.12666532397270203,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 82722
+    },
+    {
+      "epoch": 8.505552128315854,
+      "grad_norm": 0.08487075567245483,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 82725
+    },
+    {
+      "epoch": 8.505860579888957,
+      "grad_norm": 0.05700911208987236,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 82728
+    },
+    {
+      "epoch": 8.506169031462061,
+      "grad_norm": 0.033154360949993134,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 82731
+    },
+    {
+      "epoch": 8.506477483035164,
+      "grad_norm": 0.032864026725292206,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 82734
+    },
+    {
+      "epoch": 8.506785934608267,
+      "grad_norm": 0.056912586092948914,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 82737
+    },
+    {
+      "epoch": 8.50709438618137,
+      "grad_norm": 0.07113908231258392,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 82740
+    },
+    {
+      "epoch": 8.507402837754473,
+      "grad_norm": 0.04963776096701622,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 82743
+    },
+    {
+      "epoch": 8.507711289327576,
+      "grad_norm": 0.03446223959326744,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 82746
+    },
+    {
+      "epoch": 8.508019740900679,
+      "grad_norm": 0.1191178485751152,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 82749
+    },
+    {
+      "epoch": 8.508328192473781,
+      "grad_norm": 0.09850679337978363,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 82752
+    },
+    {
+      "epoch": 8.508636644046884,
+      "grad_norm": 0.04549625515937805,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 82755
+    },
+    {
+      "epoch": 8.508945095619987,
+      "grad_norm": 0.03925361484289169,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 82758
+    },
+    {
+      "epoch": 8.50925354719309,
+      "grad_norm": 0.05189771577715874,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 82761
+    },
+    {
+      "epoch": 8.509561998766193,
+      "grad_norm": 0.04734313488006592,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 82764
+    },
+    {
+      "epoch": 8.509870450339296,
+      "grad_norm": 0.05462914705276489,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 82767
+    },
+    {
+      "epoch": 8.5101789019124,
+      "grad_norm": 0.047157566994428635,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 82770
+    },
+    {
+      "epoch": 8.510487353485503,
+      "grad_norm": 0.03989399969577789,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 82773
+    },
+    {
+      "epoch": 8.510795805058606,
+      "grad_norm": 0.13090354204177856,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 82776
+    },
+    {
+      "epoch": 8.51110425663171,
+      "grad_norm": 0.04144233837723732,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 82779
+    },
+    {
+      "epoch": 8.511412708204812,
+      "grad_norm": 0.06514573842287064,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 82782
+    },
+    {
+      "epoch": 8.511721159777915,
+      "grad_norm": 0.0639931708574295,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 82785
+    },
+    {
+      "epoch": 8.512029611351018,
+      "grad_norm": 0.04106876254081726,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 82788
+    },
+    {
+      "epoch": 8.51233806292412,
+      "grad_norm": 0.045197829604148865,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 82791
+    },
+    {
+      "epoch": 8.512646514497224,
+      "grad_norm": 0.04337150603532791,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 82794
+    },
+    {
+      "epoch": 8.512954966070327,
+      "grad_norm": 0.04576044902205467,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 82797
+    },
+    {
+      "epoch": 8.51326341764343,
+      "grad_norm": 0.06863435357809067,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 82800
+    },
+    {
+      "epoch": 8.513571869216532,
+      "grad_norm": 0.09685313701629639,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 82803
+    },
+    {
+      "epoch": 8.513880320789635,
+      "grad_norm": 0.09746305644512177,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 82806
+    },
+    {
+      "epoch": 8.51418877236274,
+      "grad_norm": 0.039198584854602814,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 82809
+    },
+    {
+      "epoch": 8.514497223935843,
+      "grad_norm": 0.03824876993894577,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 82812
+    },
+    {
+      "epoch": 8.514805675508946,
+      "grad_norm": 0.07905596494674683,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 82815
+    },
+    {
+      "epoch": 8.515114127082049,
+      "grad_norm": 0.1133846566081047,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 82818
+    },
+    {
+      "epoch": 8.515422578655151,
+      "grad_norm": 0.09092739969491959,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 82821
+    },
+    {
+      "epoch": 8.515731030228254,
+      "grad_norm": 0.06377000361680984,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 82824
+    },
+    {
+      "epoch": 8.516039481801357,
+      "grad_norm": 0.033967819064855576,
+      "learning_rate": 0.01,
+      "loss": 1.977,
+      "step": 82827
+    },
+    {
+      "epoch": 8.51634793337446,
+      "grad_norm": 0.03668487071990967,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 82830
+    },
+    {
+      "epoch": 8.516656384947563,
+      "grad_norm": 0.0694398507475853,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 82833
+    },
+    {
+      "epoch": 8.516964836520666,
+      "grad_norm": 0.10825932025909424,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 82836
+    },
+    {
+      "epoch": 8.517273288093769,
+      "grad_norm": 0.12376853823661804,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 82839
+    },
+    {
+      "epoch": 8.517581739666872,
+      "grad_norm": 0.04124746099114418,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 82842
+    },
+    {
+      "epoch": 8.517890191239974,
+      "grad_norm": 0.04353870078921318,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 82845
+    },
+    {
+      "epoch": 8.51819864281308,
+      "grad_norm": 0.04228542745113373,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 82848
+    },
+    {
+      "epoch": 8.518507094386182,
+      "grad_norm": 0.05087719112634659,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 82851
+    },
+    {
+      "epoch": 8.518815545959285,
+      "grad_norm": 0.04548172652721405,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 82854
+    },
+    {
+      "epoch": 8.519123997532388,
+      "grad_norm": 0.04723157361149788,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 82857
+    },
+    {
+      "epoch": 8.51943244910549,
+      "grad_norm": 0.1239178255200386,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 82860
+    },
+    {
+      "epoch": 8.519740900678594,
+      "grad_norm": 0.06999621540307999,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 82863
+    },
+    {
+      "epoch": 8.520049352251696,
+      "grad_norm": 0.07094637304544449,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 82866
+    },
+    {
+      "epoch": 8.5203578038248,
+      "grad_norm": 0.047610655426979065,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 82869
+    },
+    {
+      "epoch": 8.520666255397902,
+      "grad_norm": 0.06207083538174629,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 82872
+    },
+    {
+      "epoch": 8.520974706971005,
+      "grad_norm": 0.04032672569155693,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 82875
+    },
+    {
+      "epoch": 8.521283158544108,
+      "grad_norm": 0.08010223507881165,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 82878
+    },
+    {
+      "epoch": 8.521591610117211,
+      "grad_norm": 0.08639922738075256,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 82881
+    },
+    {
+      "epoch": 8.521900061690314,
+      "grad_norm": 0.05493670329451561,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 82884
+    },
+    {
+      "epoch": 8.522208513263418,
+      "grad_norm": 0.0760105550289154,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 82887
+    },
+    {
+      "epoch": 8.522516964836521,
+      "grad_norm": 0.04124101251363754,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 82890
+    },
+    {
+      "epoch": 8.522825416409624,
+      "grad_norm": 0.04088043421506882,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 82893
+    },
+    {
+      "epoch": 8.523133867982727,
+      "grad_norm": 0.05277232453227043,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 82896
+    },
+    {
+      "epoch": 8.52344231955583,
+      "grad_norm": 0.04177923500537872,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 82899
+    },
+    {
+      "epoch": 8.523750771128933,
+      "grad_norm": 0.04745470732450485,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 82902
+    },
+    {
+      "epoch": 8.524059222702036,
+      "grad_norm": 0.16573688387870789,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 82905
+    },
+    {
+      "epoch": 8.524367674275139,
+      "grad_norm": 0.046021923422813416,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 82908
+    },
+    {
+      "epoch": 8.524676125848242,
+      "grad_norm": 0.08142309635877609,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 82911
+    },
+    {
+      "epoch": 8.524984577421344,
+      "grad_norm": 0.05172238498926163,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 82914
+    },
+    {
+      "epoch": 8.525293028994447,
+      "grad_norm": 0.04434951767325401,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 82917
+    },
+    {
+      "epoch": 8.52560148056755,
+      "grad_norm": 0.0400237962603569,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 82920
+    },
+    {
+      "epoch": 8.525909932140653,
+      "grad_norm": 0.07618872076272964,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 82923
+    },
+    {
+      "epoch": 8.526218383713758,
+      "grad_norm": 0.05563430115580559,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 82926
+    },
+    {
+      "epoch": 8.52652683528686,
+      "grad_norm": 0.12297258526086807,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 82929
+    },
+    {
+      "epoch": 8.526835286859964,
+      "grad_norm": 0.04254676029086113,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 82932
+    },
+    {
+      "epoch": 8.527143738433066,
+      "grad_norm": 0.04348786920309067,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 82935
+    },
+    {
+      "epoch": 8.52745219000617,
+      "grad_norm": 0.0981455147266388,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 82938
+    },
+    {
+      "epoch": 8.527760641579272,
+      "grad_norm": 0.06708617508411407,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 82941
+    },
+    {
+      "epoch": 8.528069093152375,
+      "grad_norm": 0.0691312775015831,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 82944
+    },
+    {
+      "epoch": 8.528377544725478,
+      "grad_norm": 0.040481530129909515,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 82947
+    },
+    {
+      "epoch": 8.52868599629858,
+      "grad_norm": 0.05544349551200867,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 82950
+    },
+    {
+      "epoch": 8.528994447871684,
+      "grad_norm": 0.06614125519990921,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 82953
+    },
+    {
+      "epoch": 8.529302899444787,
+      "grad_norm": 0.10472485423088074,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 82956
+    },
+    {
+      "epoch": 8.52961135101789,
+      "grad_norm": 0.11792192608118057,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 82959
+    },
+    {
+      "epoch": 8.529919802590992,
+      "grad_norm": 0.03551974147558212,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 82962
+    },
+    {
+      "epoch": 8.530228254164097,
+      "grad_norm": 0.10112152248620987,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 82965
+    },
+    {
+      "epoch": 8.5305367057372,
+      "grad_norm": 0.10229824483394623,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 82968
+    },
+    {
+      "epoch": 8.530845157310303,
+      "grad_norm": 0.04930291324853897,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 82971
+    },
+    {
+      "epoch": 8.531153608883406,
+      "grad_norm": 0.08808545768260956,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 82974
+    },
+    {
+      "epoch": 8.531462060456509,
+      "grad_norm": 0.05333753675222397,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 82977
+    },
+    {
+      "epoch": 8.531770512029611,
+      "grad_norm": 0.039773065596818924,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 82980
+    },
+    {
+      "epoch": 8.532078963602714,
+      "grad_norm": 0.04518011584877968,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 82983
+    },
+    {
+      "epoch": 8.532387415175817,
+      "grad_norm": 0.038609880954027176,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 82986
+    },
+    {
+      "epoch": 8.53269586674892,
+      "grad_norm": 0.054003313183784485,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 82989
+    },
+    {
+      "epoch": 8.533004318322023,
+      "grad_norm": 0.08430000394582748,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 82992
+    },
+    {
+      "epoch": 8.533312769895126,
+      "grad_norm": 0.10585802793502808,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 82995
+    },
+    {
+      "epoch": 8.533621221468229,
+      "grad_norm": 0.05213607847690582,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 82998
+    },
+    {
+      "epoch": 8.533929673041332,
+      "grad_norm": 0.04785900563001633,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 83001
+    },
+    {
+      "epoch": 8.534238124614436,
+      "grad_norm": 0.041036851704120636,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 83004
+    },
+    {
+      "epoch": 8.53454657618754,
+      "grad_norm": 0.03290323168039322,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 83007
+    },
+    {
+      "epoch": 8.534855027760642,
+      "grad_norm": 0.036765702068805695,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 83010
+    },
+    {
+      "epoch": 8.535163479333745,
+      "grad_norm": 0.10406450927257538,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 83013
+    },
+    {
+      "epoch": 8.535471930906848,
+      "grad_norm": 0.040695950388908386,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 83016
+    },
+    {
+      "epoch": 8.53578038247995,
+      "grad_norm": 0.047379184514284134,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 83019
+    },
+    {
+      "epoch": 8.536088834053054,
+      "grad_norm": 0.05943119153380394,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 83022
+    },
+    {
+      "epoch": 8.536397285626157,
+      "grad_norm": 0.055041078478097916,
+      "learning_rate": 0.01,
+      "loss": 1.9828,
+      "step": 83025
+    },
+    {
+      "epoch": 8.53670573719926,
+      "grad_norm": 0.041810180991888046,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 83028
+    },
+    {
+      "epoch": 8.537014188772362,
+      "grad_norm": 0.10506521165370941,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 83031
+    },
+    {
+      "epoch": 8.537322640345465,
+      "grad_norm": 0.058530621230602264,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 83034
+    },
+    {
+      "epoch": 8.537631091918568,
+      "grad_norm": 0.03754234313964844,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 83037
+    },
+    {
+      "epoch": 8.537939543491671,
+      "grad_norm": 0.03643619269132614,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 83040
+    },
+    {
+      "epoch": 8.538247995064776,
+      "grad_norm": 0.03979889675974846,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 83043
+    },
+    {
+      "epoch": 8.538556446637878,
+      "grad_norm": 0.11619020998477936,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 83046
+    },
+    {
+      "epoch": 8.538864898210981,
+      "grad_norm": 0.06696553528308868,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 83049
+    },
+    {
+      "epoch": 8.539173349784084,
+      "grad_norm": 0.05368531122803688,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 83052
+    },
+    {
+      "epoch": 8.539481801357187,
+      "grad_norm": 0.13212686777114868,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 83055
+    },
+    {
+      "epoch": 8.53979025293029,
+      "grad_norm": 0.06416644155979156,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 83058
+    },
+    {
+      "epoch": 8.540098704503393,
+      "grad_norm": 0.08319517225027084,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 83061
+    },
+    {
+      "epoch": 8.540407156076496,
+      "grad_norm": 0.05614497885107994,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 83064
+    },
+    {
+      "epoch": 8.540715607649599,
+      "grad_norm": 0.048746507614851,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 83067
+    },
+    {
+      "epoch": 8.541024059222702,
+      "grad_norm": 0.0499294176697731,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 83070
+    },
+    {
+      "epoch": 8.541332510795804,
+      "grad_norm": 0.041048891842365265,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 83073
+    },
+    {
+      "epoch": 8.541640962368907,
+      "grad_norm": 0.038432419300079346,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 83076
+    },
+    {
+      "epoch": 8.54194941394201,
+      "grad_norm": 0.03735635429620743,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 83079
+    },
+    {
+      "epoch": 8.542257865515115,
+      "grad_norm": 0.11857511103153229,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 83082
+    },
+    {
+      "epoch": 8.542566317088218,
+      "grad_norm": 0.0405680276453495,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 83085
+    },
+    {
+      "epoch": 8.54287476866132,
+      "grad_norm": 0.09402373433113098,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 83088
+    },
+    {
+      "epoch": 8.543183220234424,
+      "grad_norm": 0.14297622442245483,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 83091
+    },
+    {
+      "epoch": 8.543491671807526,
+      "grad_norm": 0.0900854766368866,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 83094
+    },
+    {
+      "epoch": 8.54380012338063,
+      "grad_norm": 0.057596318423748016,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 83097
+    },
+    {
+      "epoch": 8.544108574953732,
+      "grad_norm": 0.04268134385347366,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 83100
+    },
+    {
+      "epoch": 8.544417026526835,
+      "grad_norm": 0.05697793513536453,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 83103
+    },
+    {
+      "epoch": 8.544725478099938,
+      "grad_norm": 0.055229440331459045,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 83106
+    },
+    {
+      "epoch": 8.54503392967304,
+      "grad_norm": 0.04192070662975311,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 83109
+    },
+    {
+      "epoch": 8.545342381246144,
+      "grad_norm": 0.14123336970806122,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 83112
+    },
+    {
+      "epoch": 8.545650832819247,
+      "grad_norm": 0.12341219931840897,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 83115
+    },
+    {
+      "epoch": 8.54595928439235,
+      "grad_norm": 0.06125897914171219,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 83118
+    },
+    {
+      "epoch": 8.546267735965454,
+      "grad_norm": 0.04034245386719704,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 83121
+    },
+    {
+      "epoch": 8.546576187538557,
+      "grad_norm": 0.03154856339097023,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 83124
+    },
+    {
+      "epoch": 8.54688463911166,
+      "grad_norm": 0.0465167835354805,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 83127
+    },
+    {
+      "epoch": 8.547193090684763,
+      "grad_norm": 0.07463836669921875,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 83130
+    },
+    {
+      "epoch": 8.547501542257866,
+      "grad_norm": 0.04135550931096077,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 83133
+    },
+    {
+      "epoch": 8.547809993830969,
+      "grad_norm": 0.07107899338006973,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 83136
+    },
+    {
+      "epoch": 8.548118445404072,
+      "grad_norm": 0.08107513189315796,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 83139
+    },
+    {
+      "epoch": 8.548426896977174,
+      "grad_norm": 0.045669108629226685,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 83142
+    },
+    {
+      "epoch": 8.548735348550277,
+      "grad_norm": 0.04469345510005951,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 83145
+    },
+    {
+      "epoch": 8.54904380012338,
+      "grad_norm": 0.0745619535446167,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 83148
+    },
+    {
+      "epoch": 8.549352251696483,
+      "grad_norm": 0.06670184433460236,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 83151
+    },
+    {
+      "epoch": 8.549660703269586,
+      "grad_norm": 0.04800862818956375,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 83154
+    },
+    {
+      "epoch": 8.549969154842689,
+      "grad_norm": 0.034935347735881805,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 83157
+    },
+    {
+      "epoch": 8.550277606415793,
+      "grad_norm": 0.08265478909015656,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 83160
+    },
+    {
+      "epoch": 8.550586057988896,
+      "grad_norm": 0.11869093775749207,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 83163
+    },
+    {
+      "epoch": 8.550894509562,
+      "grad_norm": 0.0789896696805954,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 83166
+    },
+    {
+      "epoch": 8.551202961135102,
+      "grad_norm": 0.059261299669742584,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 83169
+    },
+    {
+      "epoch": 8.551511412708205,
+      "grad_norm": 0.03951359540224075,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 83172
+    },
+    {
+      "epoch": 8.551819864281308,
+      "grad_norm": 0.03497397154569626,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 83175
+    },
+    {
+      "epoch": 8.55212831585441,
+      "grad_norm": 0.04117455706000328,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 83178
+    },
+    {
+      "epoch": 8.552436767427514,
+      "grad_norm": 0.039526063948869705,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 83181
+    },
+    {
+      "epoch": 8.552745219000617,
+      "grad_norm": 0.09101678431034088,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 83184
+    },
+    {
+      "epoch": 8.55305367057372,
+      "grad_norm": 0.06288325041532516,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 83187
+    },
+    {
+      "epoch": 8.553362122146822,
+      "grad_norm": 0.08381551504135132,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 83190
+    },
+    {
+      "epoch": 8.553670573719925,
+      "grad_norm": 0.054703015834093094,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 83193
+    },
+    {
+      "epoch": 8.553979025293028,
+      "grad_norm": 0.12512286007404327,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 83196
+    },
+    {
+      "epoch": 8.554287476866133,
+      "grad_norm": 0.06523016840219498,
+      "learning_rate": 0.01,
+      "loss": 1.9963,
+      "step": 83199
+    },
+    {
+      "epoch": 8.554595928439236,
+      "grad_norm": 0.038854289799928665,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 83202
+    },
+    {
+      "epoch": 8.554904380012339,
+      "grad_norm": 0.042029645293951035,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 83205
+    },
+    {
+      "epoch": 8.555212831585441,
+      "grad_norm": 0.04491233825683594,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 83208
+    },
+    {
+      "epoch": 8.555521283158544,
+      "grad_norm": 0.06444685161113739,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 83211
+    },
+    {
+      "epoch": 8.555829734731647,
+      "grad_norm": 0.09291630983352661,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 83214
+    },
+    {
+      "epoch": 8.55613818630475,
+      "grad_norm": 0.10955164581537247,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 83217
+    },
+    {
+      "epoch": 8.556446637877853,
+      "grad_norm": 0.06862539798021317,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 83220
+    },
+    {
+      "epoch": 8.556755089450956,
+      "grad_norm": 0.05166905000805855,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 83223
+    },
+    {
+      "epoch": 8.557063541024059,
+      "grad_norm": 0.11506062000989914,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 83226
+    },
+    {
+      "epoch": 8.557371992597162,
+      "grad_norm": 0.042790256440639496,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 83229
+    },
+    {
+      "epoch": 8.557680444170265,
+      "grad_norm": 0.06458380818367004,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 83232
+    },
+    {
+      "epoch": 8.557988895743367,
+      "grad_norm": 0.1059490218758583,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 83235
+    },
+    {
+      "epoch": 8.558297347316472,
+      "grad_norm": 0.09995119273662567,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 83238
+    },
+    {
+      "epoch": 8.558605798889575,
+      "grad_norm": 0.06877456605434418,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 83241
+    },
+    {
+      "epoch": 8.558914250462678,
+      "grad_norm": 0.054978057742118835,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 83244
+    },
+    {
+      "epoch": 8.55922270203578,
+      "grad_norm": 0.040848322212696075,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 83247
+    },
+    {
+      "epoch": 8.559531153608884,
+      "grad_norm": 0.07616148144006729,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 83250
+    },
+    {
+      "epoch": 8.559839605181986,
+      "grad_norm": 0.05610670521855354,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 83253
+    },
+    {
+      "epoch": 8.56014805675509,
+      "grad_norm": 0.06030314415693283,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 83256
+    },
+    {
+      "epoch": 8.560456508328192,
+      "grad_norm": 0.037397000938653946,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 83259
+    },
+    {
+      "epoch": 8.560764959901295,
+      "grad_norm": 0.08328816294670105,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 83262
+    },
+    {
+      "epoch": 8.561073411474398,
+      "grad_norm": 0.04315868392586708,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 83265
+    },
+    {
+      "epoch": 8.561381863047501,
+      "grad_norm": 0.09421800822019577,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 83268
+    },
+    {
+      "epoch": 8.561690314620604,
+      "grad_norm": 0.0682055726647377,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 83271
+    },
+    {
+      "epoch": 8.561998766193707,
+      "grad_norm": 0.10536344349384308,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 83274
+    },
+    {
+      "epoch": 8.562307217766811,
+      "grad_norm": 0.06057734787464142,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 83277
+    },
+    {
+      "epoch": 8.562615669339914,
+      "grad_norm": 0.07082143425941467,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 83280
+    },
+    {
+      "epoch": 8.562924120913017,
+      "grad_norm": 0.061276625841856,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 83283
+    },
+    {
+      "epoch": 8.56323257248612,
+      "grad_norm": 0.06699836999177933,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 83286
+    },
+    {
+      "epoch": 8.563541024059223,
+      "grad_norm": 0.055695511400699615,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 83289
+    },
+    {
+      "epoch": 8.563849475632326,
+      "grad_norm": 0.04054784029722214,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 83292
+    },
+    {
+      "epoch": 8.564157927205429,
+      "grad_norm": 0.11725812405347824,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 83295
+    },
+    {
+      "epoch": 8.564466378778532,
+      "grad_norm": 0.034822978079319,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 83298
+    },
+    {
+      "epoch": 8.564774830351634,
+      "grad_norm": 0.09544063359498978,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 83301
+    },
+    {
+      "epoch": 8.565083281924737,
+      "grad_norm": 0.058509718626737595,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 83304
+    },
+    {
+      "epoch": 8.56539173349784,
+      "grad_norm": 0.06134885177016258,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 83307
+    },
+    {
+      "epoch": 8.565700185070943,
+      "grad_norm": 0.07492076605558395,
+      "learning_rate": 0.01,
+      "loss": 1.984,
+      "step": 83310
+    },
+    {
+      "epoch": 8.566008636644046,
+      "grad_norm": 0.10118688642978668,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 83313
+    },
+    {
+      "epoch": 8.56631708821715,
+      "grad_norm": 0.18619944155216217,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 83316
+    },
+    {
+      "epoch": 8.566625539790254,
+      "grad_norm": 0.13170650601387024,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 83319
+    },
+    {
+      "epoch": 8.566933991363356,
+      "grad_norm": 0.05305859446525574,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 83322
+    },
+    {
+      "epoch": 8.56724244293646,
+      "grad_norm": 0.04763732850551605,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 83325
+    },
+    {
+      "epoch": 8.567550894509562,
+      "grad_norm": 0.05493495613336563,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 83328
+    },
+    {
+      "epoch": 8.567859346082665,
+      "grad_norm": 0.07272114604711533,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 83331
+    },
+    {
+      "epoch": 8.568167797655768,
+      "grad_norm": 0.07231943309307098,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 83334
+    },
+    {
+      "epoch": 8.56847624922887,
+      "grad_norm": 0.04583733528852463,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 83337
+    },
+    {
+      "epoch": 8.568784700801974,
+      "grad_norm": 0.04909888654947281,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 83340
+    },
+    {
+      "epoch": 8.569093152375077,
+      "grad_norm": 0.05668563023209572,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 83343
+    },
+    {
+      "epoch": 8.56940160394818,
+      "grad_norm": 0.09962517023086548,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 83346
+    },
+    {
+      "epoch": 8.569710055521282,
+      "grad_norm": 0.0734582468867302,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 83349
+    },
+    {
+      "epoch": 8.570018507094385,
+      "grad_norm": 0.0973704606294632,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 83352
+    },
+    {
+      "epoch": 8.57032695866749,
+      "grad_norm": 0.03622818738222122,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 83355
+    },
+    {
+      "epoch": 8.570635410240593,
+      "grad_norm": 0.041889727115631104,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 83358
+    },
+    {
+      "epoch": 8.570943861813696,
+      "grad_norm": 0.0979967936873436,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 83361
+    },
+    {
+      "epoch": 8.571252313386799,
+      "grad_norm": 0.07424527406692505,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 83364
+    },
+    {
+      "epoch": 8.571560764959901,
+      "grad_norm": 0.05315127596259117,
+      "learning_rate": 0.01,
+      "loss": 1.8978,
+      "step": 83367
+    },
+    {
+      "epoch": 8.571869216533004,
+      "grad_norm": 0.0411335751414299,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 83370
+    },
+    {
+      "epoch": 8.572177668106107,
+      "grad_norm": 0.12241435796022415,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 83373
+    },
+    {
+      "epoch": 8.57248611967921,
+      "grad_norm": 0.060931794345378876,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 83376
+    },
+    {
+      "epoch": 8.572794571252313,
+      "grad_norm": 0.04587867110967636,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 83379
+    },
+    {
+      "epoch": 8.573103022825416,
+      "grad_norm": 0.047924742102622986,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 83382
+    },
+    {
+      "epoch": 8.573411474398519,
+      "grad_norm": 0.036105889827013016,
+      "learning_rate": 0.01,
+      "loss": 1.9755,
+      "step": 83385
+    },
+    {
+      "epoch": 8.573719925971622,
+      "grad_norm": 0.05183165520429611,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 83388
+    },
+    {
+      "epoch": 8.574028377544725,
+      "grad_norm": 0.05435674265027046,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 83391
+    },
+    {
+      "epoch": 8.57433682911783,
+      "grad_norm": 0.04219867289066315,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 83394
+    },
+    {
+      "epoch": 8.574645280690932,
+      "grad_norm": 0.052681926637887955,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 83397
+    },
+    {
+      "epoch": 8.574953732264035,
+      "grad_norm": 0.1265089213848114,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 83400
+    },
+    {
+      "epoch": 8.575262183837138,
+      "grad_norm": 0.12898126244544983,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 83403
+    },
+    {
+      "epoch": 8.57557063541024,
+      "grad_norm": 0.08191855251789093,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 83406
+    },
+    {
+      "epoch": 8.575879086983344,
+      "grad_norm": 0.04627726227045059,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 83409
+    },
+    {
+      "epoch": 8.576187538556447,
+      "grad_norm": 0.054639171808958054,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 83412
+    },
+    {
+      "epoch": 8.57649599012955,
+      "grad_norm": 0.043029773980379105,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 83415
+    },
+    {
+      "epoch": 8.576804441702652,
+      "grad_norm": 0.029653949663043022,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 83418
+    },
+    {
+      "epoch": 8.577112893275755,
+      "grad_norm": 0.058871399611234665,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 83421
+    },
+    {
+      "epoch": 8.577421344848858,
+      "grad_norm": 0.09742128849029541,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 83424
+    },
+    {
+      "epoch": 8.577729796421961,
+      "grad_norm": 0.10765515267848969,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 83427
+    },
+    {
+      "epoch": 8.578038247995064,
+      "grad_norm": 0.0624842531979084,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 83430
+    },
+    {
+      "epoch": 8.578346699568169,
+      "grad_norm": 0.04315463826060295,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 83433
+    },
+    {
+      "epoch": 8.578655151141271,
+      "grad_norm": 0.040104761719703674,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 83436
+    },
+    {
+      "epoch": 8.578963602714374,
+      "grad_norm": 0.08021046966314316,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 83439
+    },
+    {
+      "epoch": 8.579272054287477,
+      "grad_norm": 0.03588419780135155,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 83442
+    },
+    {
+      "epoch": 8.57958050586058,
+      "grad_norm": 0.035717666149139404,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 83445
+    },
+    {
+      "epoch": 8.579888957433683,
+      "grad_norm": 0.05417096987366676,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 83448
+    },
+    {
+      "epoch": 8.580197409006786,
+      "grad_norm": 0.06687122583389282,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 83451
+    },
+    {
+      "epoch": 8.580505860579889,
+      "grad_norm": 0.1007574051618576,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 83454
+    },
+    {
+      "epoch": 8.580814312152992,
+      "grad_norm": 0.03898779675364494,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 83457
+    },
+    {
+      "epoch": 8.581122763726095,
+      "grad_norm": 0.12776483595371246,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 83460
+    },
+    {
+      "epoch": 8.581431215299197,
+      "grad_norm": 0.04682155326008797,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 83463
+    },
+    {
+      "epoch": 8.5817396668723,
+      "grad_norm": 0.06401663273572922,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 83466
+    },
+    {
+      "epoch": 8.582048118445403,
+      "grad_norm": 0.05969379097223282,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 83469
+    },
+    {
+      "epoch": 8.582356570018508,
+      "grad_norm": 0.04335704445838928,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 83472
+    },
+    {
+      "epoch": 8.58266502159161,
+      "grad_norm": 0.039102621376514435,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 83475
+    },
+    {
+      "epoch": 8.582973473164714,
+      "grad_norm": 0.0432644747197628,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 83478
+    },
+    {
+      "epoch": 8.583281924737816,
+      "grad_norm": 0.11703160405158997,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 83481
+    },
+    {
+      "epoch": 8.58359037631092,
+      "grad_norm": 0.09045902639627457,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 83484
+    },
+    {
+      "epoch": 8.583898827884022,
+      "grad_norm": 0.08998353034257889,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 83487
+    },
+    {
+      "epoch": 8.584207279457125,
+      "grad_norm": 0.08874877542257309,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 83490
+    },
+    {
+      "epoch": 8.584515731030228,
+      "grad_norm": 0.06457845866680145,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 83493
+    },
+    {
+      "epoch": 8.584824182603331,
+      "grad_norm": 0.0741107240319252,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 83496
+    },
+    {
+      "epoch": 8.585132634176434,
+      "grad_norm": 0.04416229575872421,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 83499
+    },
+    {
+      "epoch": 8.585441085749537,
+      "grad_norm": 0.04339141771197319,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 83502
+    },
+    {
+      "epoch": 8.58574953732264,
+      "grad_norm": 0.04124277085065842,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 83505
+    },
+    {
+      "epoch": 8.586057988895742,
+      "grad_norm": 0.05257042124867439,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 83508
+    },
+    {
+      "epoch": 8.586366440468847,
+      "grad_norm": 0.061114367097616196,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 83511
+    },
+    {
+      "epoch": 8.58667489204195,
+      "grad_norm": 0.09313886612653732,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 83514
+    },
+    {
+      "epoch": 8.586983343615053,
+      "grad_norm": 0.08771451562643051,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 83517
+    },
+    {
+      "epoch": 8.587291795188156,
+      "grad_norm": 0.04571617394685745,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 83520
+    },
+    {
+      "epoch": 8.587600246761259,
+      "grad_norm": 0.033631812781095505,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 83523
+    },
+    {
+      "epoch": 8.587908698334362,
+      "grad_norm": 0.05788915976881981,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 83526
+    },
+    {
+      "epoch": 8.588217149907464,
+      "grad_norm": 0.09849164634943008,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 83529
+    },
+    {
+      "epoch": 8.588525601480567,
+      "grad_norm": 0.06190909817814827,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 83532
+    },
+    {
+      "epoch": 8.58883405305367,
+      "grad_norm": 0.05096215009689331,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 83535
+    },
+    {
+      "epoch": 8.589142504626773,
+      "grad_norm": 0.10473909974098206,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 83538
+    },
+    {
+      "epoch": 8.589450956199876,
+      "grad_norm": 0.10345686972141266,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 83541
+    },
+    {
+      "epoch": 8.589759407772979,
+      "grad_norm": 0.060924358665943146,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 83544
+    },
+    {
+      "epoch": 8.590067859346082,
+      "grad_norm": 0.06810735166072845,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 83547
+    },
+    {
+      "epoch": 8.590376310919186,
+      "grad_norm": 0.11324567347764969,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 83550
+    },
+    {
+      "epoch": 8.59068476249229,
+      "grad_norm": 0.04607374966144562,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 83553
+    },
+    {
+      "epoch": 8.590993214065392,
+      "grad_norm": 0.07352105528116226,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 83556
+    },
+    {
+      "epoch": 8.591301665638495,
+      "grad_norm": 0.060928743332624435,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 83559
+    },
+    {
+      "epoch": 8.591610117211598,
+      "grad_norm": 0.03746924549341202,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 83562
+    },
+    {
+      "epoch": 8.5919185687847,
+      "grad_norm": 0.06220705434679985,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 83565
+    },
+    {
+      "epoch": 8.592227020357804,
+      "grad_norm": 0.07664959877729416,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 83568
+    },
+    {
+      "epoch": 8.592535471930907,
+      "grad_norm": 0.09033758193254471,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 83571
+    },
+    {
+      "epoch": 8.59284392350401,
+      "grad_norm": 0.034112896770238876,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 83574
+    },
+    {
+      "epoch": 8.593152375077112,
+      "grad_norm": 0.04212299734354019,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 83577
+    },
+    {
+      "epoch": 8.593460826650215,
+      "grad_norm": 0.08151619881391525,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 83580
+    },
+    {
+      "epoch": 8.593769278223318,
+      "grad_norm": 0.05353982374072075,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 83583
+    },
+    {
+      "epoch": 8.594077729796421,
+      "grad_norm": 0.057196274399757385,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 83586
+    },
+    {
+      "epoch": 8.594386181369526,
+      "grad_norm": 0.03714732453227043,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 83589
+    },
+    {
+      "epoch": 8.594694632942629,
+      "grad_norm": 0.03743036463856697,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 83592
+    },
+    {
+      "epoch": 8.595003084515731,
+      "grad_norm": 0.0360550694167614,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 83595
+    },
+    {
+      "epoch": 8.595311536088834,
+      "grad_norm": 0.1245717853307724,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 83598
+    },
+    {
+      "epoch": 8.595619987661937,
+      "grad_norm": 0.049492839723825455,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 83601
+    },
+    {
+      "epoch": 8.59592843923504,
+      "grad_norm": 0.06890092045068741,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 83604
+    },
+    {
+      "epoch": 8.596236890808143,
+      "grad_norm": 0.049979422241449356,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 83607
+    },
+    {
+      "epoch": 8.596545342381246,
+      "grad_norm": 0.08141881227493286,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 83610
+    },
+    {
+      "epoch": 8.596853793954349,
+      "grad_norm": 0.06307000666856766,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 83613
+    },
+    {
+      "epoch": 8.597162245527452,
+      "grad_norm": 0.057910919189453125,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 83616
+    },
+    {
+      "epoch": 8.597470697100555,
+      "grad_norm": 0.03516214340925217,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 83619
+    },
+    {
+      "epoch": 8.597779148673657,
+      "grad_norm": 0.04284656420350075,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 83622
+    },
+    {
+      "epoch": 8.59808760024676,
+      "grad_norm": 0.14242172241210938,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 83625
+    },
+    {
+      "epoch": 8.598396051819865,
+      "grad_norm": 0.04614776000380516,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 83628
+    },
+    {
+      "epoch": 8.598704503392968,
+      "grad_norm": 0.03917064145207405,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 83631
+    },
+    {
+      "epoch": 8.59901295496607,
+      "grad_norm": 0.037492744624614716,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 83634
+    },
+    {
+      "epoch": 8.599321406539174,
+      "grad_norm": 0.04139791429042816,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 83637
+    },
+    {
+      "epoch": 8.599629858112277,
+      "grad_norm": 0.05112408474087715,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 83640
+    },
+    {
+      "epoch": 8.59993830968538,
+      "grad_norm": 0.034215446561574936,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 83643
+    },
+    {
+      "epoch": 8.600246761258482,
+      "grad_norm": 0.06597255170345306,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 83646
+    },
+    {
+      "epoch": 8.600555212831585,
+      "grad_norm": 0.05571454018354416,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 83649
+    },
+    {
+      "epoch": 8.600863664404688,
+      "grad_norm": 0.04197344928979874,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 83652
+    },
+    {
+      "epoch": 8.601172115977791,
+      "grad_norm": 0.04194719344377518,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 83655
+    },
+    {
+      "epoch": 8.601480567550894,
+      "grad_norm": 0.032722242176532745,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 83658
+    },
+    {
+      "epoch": 8.601789019123997,
+      "grad_norm": 0.051183219999074936,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 83661
+    },
+    {
+      "epoch": 8.6020974706971,
+      "grad_norm": 0.06459864228963852,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 83664
+    },
+    {
+      "epoch": 8.602405922270204,
+      "grad_norm": 0.07737734168767929,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 83667
+    },
+    {
+      "epoch": 8.602714373843307,
+      "grad_norm": 0.09721986949443817,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 83670
+    },
+    {
+      "epoch": 8.60302282541641,
+      "grad_norm": 0.03549128770828247,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 83673
+    },
+    {
+      "epoch": 8.603331276989513,
+      "grad_norm": 0.040530189871788025,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 83676
+    },
+    {
+      "epoch": 8.603639728562616,
+      "grad_norm": 0.05348721519112587,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 83679
+    },
+    {
+      "epoch": 8.603948180135719,
+      "grad_norm": 0.052718210965394974,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 83682
+    },
+    {
+      "epoch": 8.604256631708822,
+      "grad_norm": 0.06333748251199722,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 83685
+    },
+    {
+      "epoch": 8.604565083281924,
+      "grad_norm": 0.14412350952625275,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 83688
+    },
+    {
+      "epoch": 8.604873534855027,
+      "grad_norm": 0.11650783568620682,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 83691
+    },
+    {
+      "epoch": 8.60518198642813,
+      "grad_norm": 0.12182766944169998,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 83694
+    },
+    {
+      "epoch": 8.605490438001233,
+      "grad_norm": 0.04255716875195503,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 83697
+    },
+    {
+      "epoch": 8.605798889574336,
+      "grad_norm": 0.03763425350189209,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 83700
+    },
+    {
+      "epoch": 8.606107341147439,
+      "grad_norm": 0.03447338193655014,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 83703
+    },
+    {
+      "epoch": 8.606415792720544,
+      "grad_norm": 0.04566109552979469,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 83706
+    },
+    {
+      "epoch": 8.606724244293646,
+      "grad_norm": 0.03781317174434662,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 83709
+    },
+    {
+      "epoch": 8.60703269586675,
+      "grad_norm": 0.11389817297458649,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 83712
+    },
+    {
+      "epoch": 8.607341147439852,
+      "grad_norm": 0.058238059282302856,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 83715
+    },
+    {
+      "epoch": 8.607649599012955,
+      "grad_norm": 0.04476723074913025,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 83718
+    },
+    {
+      "epoch": 8.607958050586058,
+      "grad_norm": 0.03965180367231369,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 83721
+    },
+    {
+      "epoch": 8.608266502159161,
+      "grad_norm": 0.052368249744176865,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 83724
+    },
+    {
+      "epoch": 8.608574953732264,
+      "grad_norm": 0.039367351680994034,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 83727
+    },
+    {
+      "epoch": 8.608883405305367,
+      "grad_norm": 0.03978506475687027,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 83730
+    },
+    {
+      "epoch": 8.60919185687847,
+      "grad_norm": 0.03550339117646217,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 83733
+    },
+    {
+      "epoch": 8.609500308451572,
+      "grad_norm": 0.06137849763035774,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 83736
+    },
+    {
+      "epoch": 8.609808760024675,
+      "grad_norm": 0.10555483400821686,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 83739
+    },
+    {
+      "epoch": 8.610117211597778,
+      "grad_norm": 0.056359391659498215,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 83742
+    },
+    {
+      "epoch": 8.610425663170883,
+      "grad_norm": 0.039710041135549545,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 83745
+    },
+    {
+      "epoch": 8.610734114743986,
+      "grad_norm": 0.03399188816547394,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 83748
+    },
+    {
+      "epoch": 8.611042566317089,
+      "grad_norm": 0.05377564951777458,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 83751
+    },
+    {
+      "epoch": 8.611351017890192,
+      "grad_norm": 0.13132654130458832,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 83754
+    },
+    {
+      "epoch": 8.611659469463294,
+      "grad_norm": 0.0361807644367218,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 83757
+    },
+    {
+      "epoch": 8.611967921036397,
+      "grad_norm": 0.1197436973452568,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 83760
+    },
+    {
+      "epoch": 8.6122763726095,
+      "grad_norm": 0.06834107637405396,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 83763
+    },
+    {
+      "epoch": 8.612584824182603,
+      "grad_norm": 0.055377859622240067,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 83766
+    },
+    {
+      "epoch": 8.612893275755706,
+      "grad_norm": 0.03706175461411476,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 83769
+    },
+    {
+      "epoch": 8.613201727328809,
+      "grad_norm": 0.058211736381053925,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 83772
+    },
+    {
+      "epoch": 8.613510178901912,
+      "grad_norm": 0.09024802595376968,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 83775
+    },
+    {
+      "epoch": 8.613818630475015,
+      "grad_norm": 0.046823352575302124,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 83778
+    },
+    {
+      "epoch": 8.61412708204812,
+      "grad_norm": 0.049351755529642105,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 83781
+    },
+    {
+      "epoch": 8.614435533621222,
+      "grad_norm": 0.04153472185134888,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 83784
+    },
+    {
+      "epoch": 8.614743985194325,
+      "grad_norm": 0.08837446570396423,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 83787
+    },
+    {
+      "epoch": 8.615052436767428,
+      "grad_norm": 0.09479968249797821,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 83790
+    },
+    {
+      "epoch": 8.61536088834053,
+      "grad_norm": 0.07434171438217163,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 83793
+    },
+    {
+      "epoch": 8.615669339913634,
+      "grad_norm": 0.04010748118162155,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 83796
+    },
+    {
+      "epoch": 8.615977791486737,
+      "grad_norm": 0.05582146719098091,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 83799
+    },
+    {
+      "epoch": 8.61628624305984,
+      "grad_norm": 0.057304516434669495,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 83802
+    },
+    {
+      "epoch": 8.616594694632942,
+      "grad_norm": 0.046966370195150375,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 83805
+    },
+    {
+      "epoch": 8.616903146206045,
+      "grad_norm": 0.11336352676153183,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 83808
+    },
+    {
+      "epoch": 8.617211597779148,
+      "grad_norm": 0.051966287195682526,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 83811
+    },
+    {
+      "epoch": 8.617520049352251,
+      "grad_norm": 0.03408442437648773,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 83814
+    },
+    {
+      "epoch": 8.617828500925354,
+      "grad_norm": 0.04126676917076111,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 83817
+    },
+    {
+      "epoch": 8.618136952498459,
+      "grad_norm": 0.03017881140112877,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 83820
+    },
+    {
+      "epoch": 8.618445404071561,
+      "grad_norm": 0.13004688918590546,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 83823
+    },
+    {
+      "epoch": 8.618753855644664,
+      "grad_norm": 0.06664896011352539,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 83826
+    },
+    {
+      "epoch": 8.619062307217767,
+      "grad_norm": 0.0494823157787323,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 83829
+    },
+    {
+      "epoch": 8.61937075879087,
+      "grad_norm": 0.0706791952252388,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 83832
+    },
+    {
+      "epoch": 8.619679210363973,
+      "grad_norm": 0.08942440897226334,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 83835
+    },
+    {
+      "epoch": 8.619987661937076,
+      "grad_norm": 0.04947560653090477,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 83838
+    },
+    {
+      "epoch": 8.620296113510179,
+      "grad_norm": 0.05541004240512848,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 83841
+    },
+    {
+      "epoch": 8.620604565083282,
+      "grad_norm": 0.0484788678586483,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 83844
+    },
+    {
+      "epoch": 8.620913016656385,
+      "grad_norm": 0.04355737939476967,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 83847
+    },
+    {
+      "epoch": 8.621221468229487,
+      "grad_norm": 0.1241140142083168,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 83850
+    },
+    {
+      "epoch": 8.62152991980259,
+      "grad_norm": 0.04977406561374664,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 83853
+    },
+    {
+      "epoch": 8.621838371375693,
+      "grad_norm": 0.10179470479488373,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 83856
+    },
+    {
+      "epoch": 8.622146822948798,
+      "grad_norm": 0.05046776682138443,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 83859
+    },
+    {
+      "epoch": 8.6224552745219,
+      "grad_norm": 0.06996992230415344,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 83862
+    },
+    {
+      "epoch": 8.622763726095004,
+      "grad_norm": 0.04119875654578209,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 83865
+    },
+    {
+      "epoch": 8.623072177668107,
+      "grad_norm": 0.04418569058179855,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 83868
+    },
+    {
+      "epoch": 8.62338062924121,
+      "grad_norm": 0.04294402897357941,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 83871
+    },
+    {
+      "epoch": 8.623689080814312,
+      "grad_norm": 0.054850008338689804,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 83874
+    },
+    {
+      "epoch": 8.623997532387415,
+      "grad_norm": 0.053541239351034164,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 83877
+    },
+    {
+      "epoch": 8.624305983960518,
+      "grad_norm": 0.03481874614953995,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 83880
+    },
+    {
+      "epoch": 8.624614435533621,
+      "grad_norm": 0.05405457317829132,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 83883
+    },
+    {
+      "epoch": 8.624922887106724,
+      "grad_norm": 0.07214061915874481,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 83886
+    },
+    {
+      "epoch": 8.625231338679827,
+      "grad_norm": 0.10415913164615631,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 83889
+    },
+    {
+      "epoch": 8.62553979025293,
+      "grad_norm": 0.055225618183612823,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 83892
+    },
+    {
+      "epoch": 8.625848241826032,
+      "grad_norm": 0.05259310081601143,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 83895
+    },
+    {
+      "epoch": 8.626156693399137,
+      "grad_norm": 0.03452230989933014,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 83898
+    },
+    {
+      "epoch": 8.62646514497224,
+      "grad_norm": 0.042962346225976944,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 83901
+    },
+    {
+      "epoch": 8.626773596545343,
+      "grad_norm": 0.0799313634634018,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 83904
+    },
+    {
+      "epoch": 8.627082048118446,
+      "grad_norm": 0.0853738859295845,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 83907
+    },
+    {
+      "epoch": 8.627390499691549,
+      "grad_norm": 0.084406279027462,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 83910
+    },
+    {
+      "epoch": 8.627698951264652,
+      "grad_norm": 0.07429136335849762,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 83913
+    },
+    {
+      "epoch": 8.628007402837754,
+      "grad_norm": 0.10301503539085388,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 83916
+    },
+    {
+      "epoch": 8.628315854410857,
+      "grad_norm": 0.06669825315475464,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 83919
+    },
+    {
+      "epoch": 8.62862430598396,
+      "grad_norm": 0.04260312020778656,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 83922
+    },
+    {
+      "epoch": 8.628932757557063,
+      "grad_norm": 0.09391593933105469,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 83925
+    },
+    {
+      "epoch": 8.629241209130166,
+      "grad_norm": 0.06853640079498291,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 83928
+    },
+    {
+      "epoch": 8.629549660703269,
+      "grad_norm": 0.03541896492242813,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 83931
+    },
+    {
+      "epoch": 8.629858112276372,
+      "grad_norm": 0.07495969533920288,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 83934
+    },
+    {
+      "epoch": 8.630166563849476,
+      "grad_norm": 0.08340419828891754,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 83937
+    },
+    {
+      "epoch": 8.63047501542258,
+      "grad_norm": 0.0702739953994751,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 83940
+    },
+    {
+      "epoch": 8.630783466995682,
+      "grad_norm": 0.07156652212142944,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 83943
+    },
+    {
+      "epoch": 8.631091918568785,
+      "grad_norm": 0.11016754060983658,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 83946
+    },
+    {
+      "epoch": 8.631400370141888,
+      "grad_norm": 0.07731759548187256,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 83949
+    },
+    {
+      "epoch": 8.63170882171499,
+      "grad_norm": 0.0376402772963047,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 83952
+    },
+    {
+      "epoch": 8.632017273288094,
+      "grad_norm": 0.04498516023159027,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 83955
+    },
+    {
+      "epoch": 8.632325724861197,
+      "grad_norm": 0.07725208252668381,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 83958
+    },
+    {
+      "epoch": 8.6326341764343,
+      "grad_norm": 0.17572909593582153,
+      "learning_rate": 0.01,
+      "loss": 1.9886,
+      "step": 83961
+    },
+    {
+      "epoch": 8.632942628007402,
+      "grad_norm": 0.058839064091444016,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 83964
+    },
+    {
+      "epoch": 8.633251079580505,
+      "grad_norm": 0.04480128362774849,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 83967
+    },
+    {
+      "epoch": 8.633559531153608,
+      "grad_norm": 0.04104230925440788,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 83970
+    },
+    {
+      "epoch": 8.633867982726711,
+      "grad_norm": 0.045199841260910034,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 83973
+    },
+    {
+      "epoch": 8.634176434299816,
+      "grad_norm": 0.03659757226705551,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 83976
+    },
+    {
+      "epoch": 8.634484885872919,
+      "grad_norm": 0.05661317706108093,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 83979
+    },
+    {
+      "epoch": 8.634793337446022,
+      "grad_norm": 0.04499388858675957,
+      "learning_rate": 0.01,
+      "loss": 1.9843,
+      "step": 83982
+    },
+    {
+      "epoch": 8.635101789019124,
+      "grad_norm": 0.15299367904663086,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 83985
+    },
+    {
+      "epoch": 8.635410240592227,
+      "grad_norm": 0.041990868747234344,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 83988
+    },
+    {
+      "epoch": 8.63571869216533,
+      "grad_norm": 0.03629782050848007,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 83991
+    },
+    {
+      "epoch": 8.636027143738433,
+      "grad_norm": 0.05006648600101471,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 83994
+    },
+    {
+      "epoch": 8.636335595311536,
+      "grad_norm": 0.043421901762485504,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 83997
+    },
+    {
+      "epoch": 8.636644046884639,
+      "grad_norm": 0.07872863858938217,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 84000
+    },
+    {
+      "epoch": 8.636952498457742,
+      "grad_norm": 0.06140751764178276,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 84003
+    },
+    {
+      "epoch": 8.637260950030845,
+      "grad_norm": 0.04020686075091362,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 84006
+    },
+    {
+      "epoch": 8.637569401603947,
+      "grad_norm": 0.07382489740848541,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 84009
+    },
+    {
+      "epoch": 8.637877853177052,
+      "grad_norm": 0.09471908211708069,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 84012
+    },
+    {
+      "epoch": 8.638186304750155,
+      "grad_norm": 0.13220688700675964,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 84015
+    },
+    {
+      "epoch": 8.638494756323258,
+      "grad_norm": 0.10175812989473343,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 84018
+    },
+    {
+      "epoch": 8.63880320789636,
+      "grad_norm": 0.043288156390190125,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 84021
+    },
+    {
+      "epoch": 8.639111659469464,
+      "grad_norm": 0.05520573630928993,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 84024
+    },
+    {
+      "epoch": 8.639420111042567,
+      "grad_norm": 0.046032682061195374,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 84027
+    },
+    {
+      "epoch": 8.63972856261567,
+      "grad_norm": 0.0387708954513073,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 84030
+    },
+    {
+      "epoch": 8.640037014188772,
+      "grad_norm": 0.06580737978219986,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 84033
+    },
+    {
+      "epoch": 8.640345465761875,
+      "grad_norm": 0.038871776312589645,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 84036
+    },
+    {
+      "epoch": 8.640653917334978,
+      "grad_norm": 0.033866509795188904,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 84039
+    },
+    {
+      "epoch": 8.640962368908081,
+      "grad_norm": 0.08168267458677292,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 84042
+    },
+    {
+      "epoch": 8.641270820481184,
+      "grad_norm": 0.06837775558233261,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 84045
+    },
+    {
+      "epoch": 8.641579272054287,
+      "grad_norm": 0.15111026167869568,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 84048
+    },
+    {
+      "epoch": 8.641887723627391,
+      "grad_norm": 0.08160342276096344,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 84051
+    },
+    {
+      "epoch": 8.642196175200494,
+      "grad_norm": 0.05196947604417801,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 84054
+    },
+    {
+      "epoch": 8.642504626773597,
+      "grad_norm": 0.04552631825208664,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 84057
+    },
+    {
+      "epoch": 8.6428130783467,
+      "grad_norm": 0.04029174521565437,
+      "learning_rate": 0.01,
+      "loss": 1.976,
+      "step": 84060
+    },
+    {
+      "epoch": 8.643121529919803,
+      "grad_norm": 0.04168844595551491,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 84063
+    },
+    {
+      "epoch": 8.643429981492906,
+      "grad_norm": 0.0722026526927948,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 84066
+    },
+    {
+      "epoch": 8.643738433066009,
+      "grad_norm": 0.04248455539345741,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 84069
+    },
+    {
+      "epoch": 8.644046884639112,
+      "grad_norm": 0.0467931292951107,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 84072
+    },
+    {
+      "epoch": 8.644355336212215,
+      "grad_norm": 0.1612360030412674,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 84075
+    },
+    {
+      "epoch": 8.644663787785317,
+      "grad_norm": 0.1361725628376007,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 84078
+    },
+    {
+      "epoch": 8.64497223935842,
+      "grad_norm": 0.058937378227710724,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 84081
+    },
+    {
+      "epoch": 8.645280690931523,
+      "grad_norm": 0.061618443578481674,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 84084
+    },
+    {
+      "epoch": 8.645589142504626,
+      "grad_norm": 0.039914749562740326,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 84087
+    },
+    {
+      "epoch": 8.64589759407773,
+      "grad_norm": 0.04626841843128204,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 84090
+    },
+    {
+      "epoch": 8.646206045650834,
+      "grad_norm": 0.049574919044971466,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 84093
+    },
+    {
+      "epoch": 8.646514497223937,
+      "grad_norm": 0.03658651188015938,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 84096
+    },
+    {
+      "epoch": 8.64682294879704,
+      "grad_norm": 0.035471998155117035,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 84099
+    },
+    {
+      "epoch": 8.647131400370142,
+      "grad_norm": 0.04843059927225113,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 84102
+    },
+    {
+      "epoch": 8.647439851943245,
+      "grad_norm": 0.09598714858293533,
+      "learning_rate": 0.01,
+      "loss": 1.9927,
+      "step": 84105
+    },
+    {
+      "epoch": 8.647748303516348,
+      "grad_norm": 0.1329140067100525,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 84108
+    },
+    {
+      "epoch": 8.648056755089451,
+      "grad_norm": 0.04247164726257324,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 84111
+    },
+    {
+      "epoch": 8.648365206662554,
+      "grad_norm": 0.044990718364715576,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 84114
+    },
+    {
+      "epoch": 8.648673658235657,
+      "grad_norm": 0.05021393671631813,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 84117
+    },
+    {
+      "epoch": 8.64898210980876,
+      "grad_norm": 0.048792239278554916,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 84120
+    },
+    {
+      "epoch": 8.649290561381862,
+      "grad_norm": 0.05867401510477066,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 84123
+    },
+    {
+      "epoch": 8.649599012954965,
+      "grad_norm": 0.04143434017896652,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 84126
+    },
+    {
+      "epoch": 8.64990746452807,
+      "grad_norm": 0.040858276188373566,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 84129
+    },
+    {
+      "epoch": 8.650215916101173,
+      "grad_norm": 0.057876672595739365,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 84132
+    },
+    {
+      "epoch": 8.650524367674276,
+      "grad_norm": 0.12112187594175339,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 84135
+    },
+    {
+      "epoch": 8.650832819247379,
+      "grad_norm": 0.07901202887296677,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 84138
+    },
+    {
+      "epoch": 8.651141270820482,
+      "grad_norm": 0.14830529689788818,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 84141
+    },
+    {
+      "epoch": 8.651449722393584,
+      "grad_norm": 0.10633882135152817,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 84144
+    },
+    {
+      "epoch": 8.651758173966687,
+      "grad_norm": 0.07728983461856842,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 84147
+    },
+    {
+      "epoch": 8.65206662553979,
+      "grad_norm": 0.03409942612051964,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 84150
+    },
+    {
+      "epoch": 8.652375077112893,
+      "grad_norm": 0.035978421568870544,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 84153
+    },
+    {
+      "epoch": 8.652683528685996,
+      "grad_norm": 0.03137461096048355,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 84156
+    },
+    {
+      "epoch": 8.652991980259099,
+      "grad_norm": 0.05230855941772461,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 84159
+    },
+    {
+      "epoch": 8.653300431832202,
+      "grad_norm": 0.040534112602472305,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 84162
+    },
+    {
+      "epoch": 8.653608883405305,
+      "grad_norm": 0.08531369268894196,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 84165
+    },
+    {
+      "epoch": 8.65391733497841,
+      "grad_norm": 0.05055827647447586,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 84168
+    },
+    {
+      "epoch": 8.654225786551512,
+      "grad_norm": 0.13753488659858704,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 84171
+    },
+    {
+      "epoch": 8.654534238124615,
+      "grad_norm": 0.11392176151275635,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 84174
+    },
+    {
+      "epoch": 8.654842689697718,
+      "grad_norm": 0.0441092848777771,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 84177
+    },
+    {
+      "epoch": 8.65515114127082,
+      "grad_norm": 0.05333095043897629,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 84180
+    },
+    {
+      "epoch": 8.655459592843924,
+      "grad_norm": 0.08408643305301666,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 84183
+    },
+    {
+      "epoch": 8.655768044417027,
+      "grad_norm": 0.06018614023923874,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 84186
+    },
+    {
+      "epoch": 8.65607649599013,
+      "grad_norm": 0.049890197813510895,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 84189
+    },
+    {
+      "epoch": 8.656384947563232,
+      "grad_norm": 0.06057172268629074,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 84192
+    },
+    {
+      "epoch": 8.656693399136335,
+      "grad_norm": 0.06060672923922539,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 84195
+    },
+    {
+      "epoch": 8.657001850709438,
+      "grad_norm": 0.06296031922101974,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 84198
+    },
+    {
+      "epoch": 8.657310302282541,
+      "grad_norm": 0.05551283806562424,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 84201
+    },
+    {
+      "epoch": 8.657618753855644,
+      "grad_norm": 0.044948600232601166,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 84204
+    },
+    {
+      "epoch": 8.657927205428749,
+      "grad_norm": 0.04862402006983757,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 84207
+    },
+    {
+      "epoch": 8.658235657001851,
+      "grad_norm": 0.11310559511184692,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 84210
+    },
+    {
+      "epoch": 8.658544108574954,
+      "grad_norm": 0.07889222353696823,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 84213
+    },
+    {
+      "epoch": 8.658852560148057,
+      "grad_norm": 0.12257291376590729,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 84216
+    },
+    {
+      "epoch": 8.65916101172116,
+      "grad_norm": 0.05437285453081131,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 84219
+    },
+    {
+      "epoch": 8.659469463294263,
+      "grad_norm": 0.0486476868391037,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 84222
+    },
+    {
+      "epoch": 8.659777914867366,
+      "grad_norm": 0.04230161011219025,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 84225
+    },
+    {
+      "epoch": 8.660086366440469,
+      "grad_norm": 0.05395093187689781,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 84228
+    },
+    {
+      "epoch": 8.660394818013572,
+      "grad_norm": 0.050544533878564835,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 84231
+    },
+    {
+      "epoch": 8.660703269586675,
+      "grad_norm": 0.0582561120390892,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 84234
+    },
+    {
+      "epoch": 8.661011721159777,
+      "grad_norm": 0.1040964350104332,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 84237
+    },
+    {
+      "epoch": 8.66132017273288,
+      "grad_norm": 0.03669446334242821,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 84240
+    },
+    {
+      "epoch": 8.661628624305983,
+      "grad_norm": 0.11392202973365784,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 84243
+    },
+    {
+      "epoch": 8.661937075879088,
+      "grad_norm": 0.03846215084195137,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 84246
+    },
+    {
+      "epoch": 8.66224552745219,
+      "grad_norm": 0.09960336238145828,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 84249
+    },
+    {
+      "epoch": 8.662553979025294,
+      "grad_norm": 0.07459977269172668,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 84252
+    },
+    {
+      "epoch": 8.662862430598397,
+      "grad_norm": 0.03963596746325493,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 84255
+    },
+    {
+      "epoch": 8.6631708821715,
+      "grad_norm": 0.09526780992746353,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 84258
+    },
+    {
+      "epoch": 8.663479333744602,
+      "grad_norm": 0.13918493688106537,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 84261
+    },
+    {
+      "epoch": 8.663787785317705,
+      "grad_norm": 0.0965191200375557,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 84264
+    },
+    {
+      "epoch": 8.664096236890808,
+      "grad_norm": 0.07372299581766129,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 84267
+    },
+    {
+      "epoch": 8.664404688463911,
+      "grad_norm": 0.07126685976982117,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 84270
+    },
+    {
+      "epoch": 8.664713140037014,
+      "grad_norm": 0.04940878972411156,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 84273
+    },
+    {
+      "epoch": 8.665021591610117,
+      "grad_norm": 0.041048500686883926,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 84276
+    },
+    {
+      "epoch": 8.66533004318322,
+      "grad_norm": 0.05010208114981651,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 84279
+    },
+    {
+      "epoch": 8.665638494756323,
+      "grad_norm": 0.035412829369306564,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 84282
+    },
+    {
+      "epoch": 8.665946946329427,
+      "grad_norm": 0.040003497153520584,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 84285
+    },
+    {
+      "epoch": 8.66625539790253,
+      "grad_norm": 0.09944424033164978,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 84288
+    },
+    {
+      "epoch": 8.666563849475633,
+      "grad_norm": 0.11967995762825012,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 84291
+    },
+    {
+      "epoch": 8.666872301048736,
+      "grad_norm": 0.0617777556180954,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 84294
+    },
+    {
+      "epoch": 8.667180752621839,
+      "grad_norm": 0.07158997654914856,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 84297
+    },
+    {
+      "epoch": 8.667489204194942,
+      "grad_norm": 0.08579113334417343,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 84300
+    },
+    {
+      "epoch": 8.667797655768045,
+      "grad_norm": 0.06357930600643158,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 84303
+    },
+    {
+      "epoch": 8.668106107341147,
+      "grad_norm": 0.06668613851070404,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 84306
+    },
+    {
+      "epoch": 8.66841455891425,
+      "grad_norm": 0.06983932852745056,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 84309
+    },
+    {
+      "epoch": 8.668723010487353,
+      "grad_norm": 0.07324071228504181,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 84312
+    },
+    {
+      "epoch": 8.669031462060456,
+      "grad_norm": 0.12372716516256332,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 84315
+    },
+    {
+      "epoch": 8.669339913633559,
+      "grad_norm": 0.0913889929652214,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 84318
+    },
+    {
+      "epoch": 8.669648365206662,
+      "grad_norm": 0.08444677293300629,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 84321
+    },
+    {
+      "epoch": 8.669956816779766,
+      "grad_norm": 0.06621287018060684,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 84324
+    },
+    {
+      "epoch": 8.67026526835287,
+      "grad_norm": 0.05836626887321472,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 84327
+    },
+    {
+      "epoch": 8.670573719925972,
+      "grad_norm": 0.04178153723478317,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 84330
+    },
+    {
+      "epoch": 8.670882171499075,
+      "grad_norm": 0.04391157999634743,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 84333
+    },
+    {
+      "epoch": 8.671190623072178,
+      "grad_norm": 0.041439518332481384,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 84336
+    },
+    {
+      "epoch": 8.671499074645281,
+      "grad_norm": 0.04575076699256897,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 84339
+    },
+    {
+      "epoch": 8.671807526218384,
+      "grad_norm": 0.12655547261238098,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 84342
+    },
+    {
+      "epoch": 8.672115977791487,
+      "grad_norm": 0.0857333168387413,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 84345
+    },
+    {
+      "epoch": 8.67242442936459,
+      "grad_norm": 0.07120339572429657,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 84348
+    },
+    {
+      "epoch": 8.672732880937692,
+      "grad_norm": 0.08852653950452805,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 84351
+    },
+    {
+      "epoch": 8.673041332510795,
+      "grad_norm": 0.11633346229791641,
+      "learning_rate": 0.01,
+      "loss": 1.9932,
+      "step": 84354
+    },
+    {
+      "epoch": 8.673349784083898,
+      "grad_norm": 0.07619529217481613,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 84357
+    },
+    {
+      "epoch": 8.673658235657001,
+      "grad_norm": 0.06116296350955963,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 84360
+    },
+    {
+      "epoch": 8.673966687230106,
+      "grad_norm": 0.08434920012950897,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 84363
+    },
+    {
+      "epoch": 8.674275138803209,
+      "grad_norm": 0.08315224945545197,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 84366
+    },
+    {
+      "epoch": 8.674583590376312,
+      "grad_norm": 0.07332847267389297,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 84369
+    },
+    {
+      "epoch": 8.674892041949414,
+      "grad_norm": 0.04000057280063629,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 84372
+    },
+    {
+      "epoch": 8.675200493522517,
+      "grad_norm": 0.11484423279762268,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 84375
+    },
+    {
+      "epoch": 8.67550894509562,
+      "grad_norm": 0.09171456098556519,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 84378
+    },
+    {
+      "epoch": 8.675817396668723,
+      "grad_norm": 0.0660143718123436,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 84381
+    },
+    {
+      "epoch": 8.676125848241826,
+      "grad_norm": 0.03662162646651268,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 84384
+    },
+    {
+      "epoch": 8.676434299814929,
+      "grad_norm": 0.044858962297439575,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 84387
+    },
+    {
+      "epoch": 8.676742751388032,
+      "grad_norm": 0.05560542270541191,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 84390
+    },
+    {
+      "epoch": 8.677051202961135,
+      "grad_norm": 0.053774718195199966,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 84393
+    },
+    {
+      "epoch": 8.677359654534238,
+      "grad_norm": 0.11037126928567886,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 84396
+    },
+    {
+      "epoch": 8.67766810610734,
+      "grad_norm": 0.06227453798055649,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 84399
+    },
+    {
+      "epoch": 8.677976557680445,
+      "grad_norm": 0.09894663840532303,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 84402
+    },
+    {
+      "epoch": 8.678285009253548,
+      "grad_norm": 0.06907567381858826,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 84405
+    },
+    {
+      "epoch": 8.67859346082665,
+      "grad_norm": 0.05042900890111923,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 84408
+    },
+    {
+      "epoch": 8.678901912399754,
+      "grad_norm": 0.08946631848812103,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 84411
+    },
+    {
+      "epoch": 8.679210363972857,
+      "grad_norm": 0.08894028514623642,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 84414
+    },
+    {
+      "epoch": 8.67951881554596,
+      "grad_norm": 0.0353781059384346,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 84417
+    },
+    {
+      "epoch": 8.679827267119062,
+      "grad_norm": 0.031145529821515083,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 84420
+    },
+    {
+      "epoch": 8.680135718692165,
+      "grad_norm": 0.03298868238925934,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 84423
+    },
+    {
+      "epoch": 8.680444170265268,
+      "grad_norm": 0.09564276784658432,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 84426
+    },
+    {
+      "epoch": 8.680752621838371,
+      "grad_norm": 0.030355164781212807,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 84429
+    },
+    {
+      "epoch": 8.681061073411474,
+      "grad_norm": 0.04145561531186104,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 84432
+    },
+    {
+      "epoch": 8.681369524984577,
+      "grad_norm": 0.06485619395971298,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 84435
+    },
+    {
+      "epoch": 8.68167797655768,
+      "grad_norm": 0.0540875568985939,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 84438
+    },
+    {
+      "epoch": 8.681986428130784,
+      "grad_norm": 0.039081066846847534,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 84441
+    },
+    {
+      "epoch": 8.682294879703887,
+      "grad_norm": 0.03886258602142334,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 84444
+    },
+    {
+      "epoch": 8.68260333127699,
+      "grad_norm": 0.0608995221555233,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 84447
+    },
+    {
+      "epoch": 8.682911782850093,
+      "grad_norm": 0.12530915439128876,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 84450
+    },
+    {
+      "epoch": 8.683220234423196,
+      "grad_norm": 0.08348006755113602,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 84453
+    },
+    {
+      "epoch": 8.683528685996299,
+      "grad_norm": 0.0368838831782341,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 84456
+    },
+    {
+      "epoch": 8.683837137569402,
+      "grad_norm": 0.09236108511686325,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 84459
+    },
+    {
+      "epoch": 8.684145589142505,
+      "grad_norm": 0.052283886820077896,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 84462
+    },
+    {
+      "epoch": 8.684454040715607,
+      "grad_norm": 0.10161884874105453,
+      "learning_rate": 0.01,
+      "loss": 1.9785,
+      "step": 84465
+    },
+    {
+      "epoch": 8.68476249228871,
+      "grad_norm": 0.05157725512981415,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 84468
+    },
+    {
+      "epoch": 8.685070943861813,
+      "grad_norm": 0.04975030943751335,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 84471
+    },
+    {
+      "epoch": 8.685379395434916,
+      "grad_norm": 0.11820569634437561,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 84474
+    },
+    {
+      "epoch": 8.685687847008019,
+      "grad_norm": 0.07684563845396042,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 84477
+    },
+    {
+      "epoch": 8.685996298581124,
+      "grad_norm": 0.07410179823637009,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 84480
+    },
+    {
+      "epoch": 8.686304750154227,
+      "grad_norm": 0.0489877387881279,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 84483
+    },
+    {
+      "epoch": 8.68661320172733,
+      "grad_norm": 0.03942607715725899,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 84486
+    },
+    {
+      "epoch": 8.686921653300432,
+      "grad_norm": 0.03467392548918724,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 84489
+    },
+    {
+      "epoch": 8.687230104873535,
+      "grad_norm": 0.05109754577279091,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 84492
+    },
+    {
+      "epoch": 8.687538556446638,
+      "grad_norm": 0.06021739915013313,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 84495
+    },
+    {
+      "epoch": 8.687847008019741,
+      "grad_norm": 0.1043761596083641,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 84498
+    },
+    {
+      "epoch": 8.688155459592844,
+      "grad_norm": 0.047195639461278915,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 84501
+    },
+    {
+      "epoch": 8.688463911165947,
+      "grad_norm": 0.11019480228424072,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 84504
+    },
+    {
+      "epoch": 8.68877236273905,
+      "grad_norm": 0.054665159434080124,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 84507
+    },
+    {
+      "epoch": 8.689080814312153,
+      "grad_norm": 0.11282628029584885,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 84510
+    },
+    {
+      "epoch": 8.689389265885255,
+      "grad_norm": 0.059272971004247665,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 84513
+    },
+    {
+      "epoch": 8.689697717458358,
+      "grad_norm": 0.04295983165502548,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 84516
+    },
+    {
+      "epoch": 8.690006169031463,
+      "grad_norm": 0.045819997787475586,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 84519
+    },
+    {
+      "epoch": 8.690314620604566,
+      "grad_norm": 0.08226123452186584,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 84522
+    },
+    {
+      "epoch": 8.690623072177669,
+      "grad_norm": 0.08438362181186676,
+      "learning_rate": 0.01,
+      "loss": 1.9911,
+      "step": 84525
+    },
+    {
+      "epoch": 8.690931523750772,
+      "grad_norm": 0.1893005669116974,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 84528
+    },
+    {
+      "epoch": 8.691239975323874,
+      "grad_norm": 0.07184705138206482,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 84531
+    },
+    {
+      "epoch": 8.691548426896977,
+      "grad_norm": 0.04775119200348854,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 84534
+    },
+    {
+      "epoch": 8.69185687847008,
+      "grad_norm": 0.04076853394508362,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 84537
+    },
+    {
+      "epoch": 8.692165330043183,
+      "grad_norm": 0.035510528832674026,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 84540
+    },
+    {
+      "epoch": 8.692473781616286,
+      "grad_norm": 0.039787620306015015,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 84543
+    },
+    {
+      "epoch": 8.692782233189389,
+      "grad_norm": 0.035327181220054626,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 84546
+    },
+    {
+      "epoch": 8.693090684762492,
+      "grad_norm": 0.06223301589488983,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 84549
+    },
+    {
+      "epoch": 8.693399136335595,
+      "grad_norm": 0.2152005136013031,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 84552
+    },
+    {
+      "epoch": 8.693707587908698,
+      "grad_norm": 0.10525162518024445,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 84555
+    },
+    {
+      "epoch": 8.694016039481802,
+      "grad_norm": 0.09987024962902069,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 84558
+    },
+    {
+      "epoch": 8.694324491054905,
+      "grad_norm": 0.06907907873392105,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 84561
+    },
+    {
+      "epoch": 8.694632942628008,
+      "grad_norm": 0.05267520993947983,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 84564
+    },
+    {
+      "epoch": 8.694941394201111,
+      "grad_norm": 0.03861707076430321,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 84567
+    },
+    {
+      "epoch": 8.695249845774214,
+      "grad_norm": 0.057318273931741714,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 84570
+    },
+    {
+      "epoch": 8.695558297347317,
+      "grad_norm": 0.05354475975036621,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 84573
+    },
+    {
+      "epoch": 8.69586674892042,
+      "grad_norm": 0.05339827388525009,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 84576
+    },
+    {
+      "epoch": 8.696175200493522,
+      "grad_norm": 0.050137948244810104,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 84579
+    },
+    {
+      "epoch": 8.696483652066625,
+      "grad_norm": 0.033642902970314026,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 84582
+    },
+    {
+      "epoch": 8.696792103639728,
+      "grad_norm": 0.03329618647694588,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 84585
+    },
+    {
+      "epoch": 8.697100555212831,
+      "grad_norm": 0.09975992143154144,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 84588
+    },
+    {
+      "epoch": 8.697409006785934,
+      "grad_norm": 0.15377679467201233,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 84591
+    },
+    {
+      "epoch": 8.697717458359037,
+      "grad_norm": 0.09404116123914719,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 84594
+    },
+    {
+      "epoch": 8.698025909932142,
+      "grad_norm": 0.05135189741849899,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 84597
+    },
+    {
+      "epoch": 8.698334361505244,
+      "grad_norm": 0.04255730286240578,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 84600
+    },
+    {
+      "epoch": 8.698642813078347,
+      "grad_norm": 0.04291402921080589,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 84603
+    },
+    {
+      "epoch": 8.69895126465145,
+      "grad_norm": 0.05834383890032768,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 84606
+    },
+    {
+      "epoch": 8.699259716224553,
+      "grad_norm": 0.05079890415072441,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 84609
+    },
+    {
+      "epoch": 8.699568167797656,
+      "grad_norm": 0.04097910597920418,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 84612
+    },
+    {
+      "epoch": 8.699876619370759,
+      "grad_norm": 0.03598971664905548,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 84615
+    },
+    {
+      "epoch": 8.700185070943862,
+      "grad_norm": 0.04535605013370514,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 84618
+    },
+    {
+      "epoch": 8.700493522516965,
+      "grad_norm": 0.07048530131578445,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 84621
+    },
+    {
+      "epoch": 8.700801974090068,
+      "grad_norm": 0.0713009163737297,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 84624
+    },
+    {
+      "epoch": 8.70111042566317,
+      "grad_norm": 0.10725833475589752,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 84627
+    },
+    {
+      "epoch": 8.701418877236273,
+      "grad_norm": 0.061335306614637375,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 84630
+    },
+    {
+      "epoch": 8.701727328809376,
+      "grad_norm": 0.09121435880661011,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 84633
+    },
+    {
+      "epoch": 8.70203578038248,
+      "grad_norm": 0.039087846875190735,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 84636
+    },
+    {
+      "epoch": 8.702344231955584,
+      "grad_norm": 0.06266528367996216,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 84639
+    },
+    {
+      "epoch": 8.702652683528687,
+      "grad_norm": 0.093431256711483,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 84642
+    },
+    {
+      "epoch": 8.70296113510179,
+      "grad_norm": 0.05956881865859032,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 84645
+    },
+    {
+      "epoch": 8.703269586674892,
+      "grad_norm": 0.08706646412611008,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 84648
+    },
+    {
+      "epoch": 8.703578038247995,
+      "grad_norm": 0.07080619037151337,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 84651
+    },
+    {
+      "epoch": 8.703886489821098,
+      "grad_norm": 0.05497487634420395,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 84654
+    },
+    {
+      "epoch": 8.704194941394201,
+      "grad_norm": 0.054549477994441986,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 84657
+    },
+    {
+      "epoch": 8.704503392967304,
+      "grad_norm": 0.04388706758618355,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 84660
+    },
+    {
+      "epoch": 8.704811844540407,
+      "grad_norm": 0.04360177367925644,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 84663
+    },
+    {
+      "epoch": 8.70512029611351,
+      "grad_norm": 0.07289383560419083,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 84666
+    },
+    {
+      "epoch": 8.705428747686613,
+      "grad_norm": 0.07846248149871826,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 84669
+    },
+    {
+      "epoch": 8.705737199259715,
+      "grad_norm": 0.10681965947151184,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 84672
+    },
+    {
+      "epoch": 8.70604565083282,
+      "grad_norm": 0.16123338043689728,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 84675
+    },
+    {
+      "epoch": 8.706354102405923,
+      "grad_norm": 0.12931574881076813,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 84678
+    },
+    {
+      "epoch": 8.706662553979026,
+      "grad_norm": 0.05700478330254555,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 84681
+    },
+    {
+      "epoch": 8.706971005552129,
+      "grad_norm": 0.041084181517362595,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 84684
+    },
+    {
+      "epoch": 8.707279457125232,
+      "grad_norm": 0.03249974176287651,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 84687
+    },
+    {
+      "epoch": 8.707587908698335,
+      "grad_norm": 0.03458573669195175,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 84690
+    },
+    {
+      "epoch": 8.707896360271437,
+      "grad_norm": 0.037099480628967285,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 84693
+    },
+    {
+      "epoch": 8.70820481184454,
+      "grad_norm": 0.057972583919763565,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 84696
+    },
+    {
+      "epoch": 8.708513263417643,
+      "grad_norm": 0.0855930969119072,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 84699
+    },
+    {
+      "epoch": 8.708821714990746,
+      "grad_norm": 0.10864733159542084,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 84702
+    },
+    {
+      "epoch": 8.709130166563849,
+      "grad_norm": 0.07841694355010986,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 84705
+    },
+    {
+      "epoch": 8.709438618136952,
+      "grad_norm": 0.08079460263252258,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 84708
+    },
+    {
+      "epoch": 8.709747069710055,
+      "grad_norm": 0.03550470247864723,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 84711
+    },
+    {
+      "epoch": 8.71005552128316,
+      "grad_norm": 0.057293109595775604,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 84714
+    },
+    {
+      "epoch": 8.710363972856262,
+      "grad_norm": 0.0852893516421318,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 84717
+    },
+    {
+      "epoch": 8.710672424429365,
+      "grad_norm": 0.06920026242733002,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 84720
+    },
+    {
+      "epoch": 8.710980876002468,
+      "grad_norm": 0.07233172655105591,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 84723
+    },
+    {
+      "epoch": 8.711289327575571,
+      "grad_norm": 0.04527530446648598,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 84726
+    },
+    {
+      "epoch": 8.711597779148674,
+      "grad_norm": 0.10784745216369629,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 84729
+    },
+    {
+      "epoch": 8.711906230721777,
+      "grad_norm": 0.047310832887887955,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 84732
+    },
+    {
+      "epoch": 8.71221468229488,
+      "grad_norm": 0.09746813774108887,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 84735
+    },
+    {
+      "epoch": 8.712523133867982,
+      "grad_norm": 0.04717058315873146,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 84738
+    },
+    {
+      "epoch": 8.712831585441085,
+      "grad_norm": 0.07984232157468796,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 84741
+    },
+    {
+      "epoch": 8.713140037014188,
+      "grad_norm": 0.07670383900403976,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 84744
+    },
+    {
+      "epoch": 8.713448488587291,
+      "grad_norm": 0.07070758193731308,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 84747
+    },
+    {
+      "epoch": 8.713756940160394,
+      "grad_norm": 0.03936431184411049,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 84750
+    },
+    {
+      "epoch": 8.714065391733499,
+      "grad_norm": 0.04538992792367935,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 84753
+    },
+    {
+      "epoch": 8.714373843306602,
+      "grad_norm": 0.07445311546325684,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 84756
+    },
+    {
+      "epoch": 8.714682294879704,
+      "grad_norm": 0.06671564280986786,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 84759
+    },
+    {
+      "epoch": 8.714990746452807,
+      "grad_norm": 0.06232886016368866,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 84762
+    },
+    {
+      "epoch": 8.71529919802591,
+      "grad_norm": 0.06302858144044876,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 84765
+    },
+    {
+      "epoch": 8.715607649599013,
+      "grad_norm": 0.03369203582406044,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 84768
+    },
+    {
+      "epoch": 8.715916101172116,
+      "grad_norm": 0.04007703810930252,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 84771
+    },
+    {
+      "epoch": 8.716224552745219,
+      "grad_norm": 0.06994391232728958,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 84774
+    },
+    {
+      "epoch": 8.716533004318322,
+      "grad_norm": 0.039773739874362946,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 84777
+    },
+    {
+      "epoch": 8.716841455891425,
+      "grad_norm": 0.189493790268898,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 84780
+    },
+    {
+      "epoch": 8.717149907464528,
+      "grad_norm": 0.09057370573282242,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 84783
+    },
+    {
+      "epoch": 8.71745835903763,
+      "grad_norm": 0.12711824476718903,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 84786
+    },
+    {
+      "epoch": 8.717766810610733,
+      "grad_norm": 0.0868084728717804,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 84789
+    },
+    {
+      "epoch": 8.718075262183838,
+      "grad_norm": 0.04147063195705414,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 84792
+    },
+    {
+      "epoch": 8.71838371375694,
+      "grad_norm": 0.04532619193196297,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 84795
+    },
+    {
+      "epoch": 8.718692165330044,
+      "grad_norm": 0.07199708372354507,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 84798
+    },
+    {
+      "epoch": 8.719000616903147,
+      "grad_norm": 0.060223303735256195,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 84801
+    },
+    {
+      "epoch": 8.71930906847625,
+      "grad_norm": 0.03616713732481003,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 84804
+    },
+    {
+      "epoch": 8.719617520049352,
+      "grad_norm": 0.08611498028039932,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 84807
+    },
+    {
+      "epoch": 8.719925971622455,
+      "grad_norm": 0.04345080256462097,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 84810
+    },
+    {
+      "epoch": 8.720234423195558,
+      "grad_norm": 0.07039283215999603,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 84813
+    },
+    {
+      "epoch": 8.720542874768661,
+      "grad_norm": 0.03262734040617943,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 84816
+    },
+    {
+      "epoch": 8.720851326341764,
+      "grad_norm": 0.0588054358959198,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 84819
+    },
+    {
+      "epoch": 8.721159777914867,
+      "grad_norm": 0.17544658482074738,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 84822
+    },
+    {
+      "epoch": 8.72146822948797,
+      "grad_norm": 0.054100871086120605,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 84825
+    },
+    {
+      "epoch": 8.721776681061073,
+      "grad_norm": 0.07093433290719986,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 84828
+    },
+    {
+      "epoch": 8.722085132634177,
+      "grad_norm": 0.041165079921483994,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 84831
+    },
+    {
+      "epoch": 8.72239358420728,
+      "grad_norm": 0.03833869844675064,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 84834
+    },
+    {
+      "epoch": 8.722702035780383,
+      "grad_norm": 0.03908578306436539,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 84837
+    },
+    {
+      "epoch": 8.723010487353486,
+      "grad_norm": 0.05425461381673813,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 84840
+    },
+    {
+      "epoch": 8.723318938926589,
+      "grad_norm": 0.06133231148123741,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 84843
+    },
+    {
+      "epoch": 8.723627390499692,
+      "grad_norm": 0.08673300594091415,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 84846
+    },
+    {
+      "epoch": 8.723935842072795,
+      "grad_norm": 0.051887791603803635,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 84849
+    },
+    {
+      "epoch": 8.724244293645897,
+      "grad_norm": 0.05313627049326897,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 84852
+    },
+    {
+      "epoch": 8.724552745219,
+      "grad_norm": 0.03683719411492348,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 84855
+    },
+    {
+      "epoch": 8.724861196792103,
+      "grad_norm": 0.14222852885723114,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 84858
+    },
+    {
+      "epoch": 8.725169648365206,
+      "grad_norm": 0.06968168914318085,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 84861
+    },
+    {
+      "epoch": 8.725478099938309,
+      "grad_norm": 0.08058342337608337,
+      "learning_rate": 0.01,
+      "loss": 1.9819,
+      "step": 84864
+    },
+    {
+      "epoch": 8.725786551511412,
+      "grad_norm": 0.05098986625671387,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 84867
+    },
+    {
+      "epoch": 8.726095003084517,
+      "grad_norm": 0.07216066867113113,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 84870
+    },
+    {
+      "epoch": 8.72640345465762,
+      "grad_norm": 0.05139782279729843,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 84873
+    },
+    {
+      "epoch": 8.726711906230722,
+      "grad_norm": 0.08893761783838272,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 84876
+    },
+    {
+      "epoch": 8.727020357803825,
+      "grad_norm": 0.12131284922361374,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 84879
+    },
+    {
+      "epoch": 8.727328809376928,
+      "grad_norm": 0.11023629456758499,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 84882
+    },
+    {
+      "epoch": 8.727637260950031,
+      "grad_norm": 0.05484713986515999,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 84885
+    },
+    {
+      "epoch": 8.727945712523134,
+      "grad_norm": 0.0429035983979702,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 84888
+    },
+    {
+      "epoch": 8.728254164096237,
+      "grad_norm": 0.0505860261619091,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 84891
+    },
+    {
+      "epoch": 8.72856261566934,
+      "grad_norm": 0.06414588540792465,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 84894
+    },
+    {
+      "epoch": 8.728871067242443,
+      "grad_norm": 0.050940901041030884,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 84897
+    },
+    {
+      "epoch": 8.729179518815545,
+      "grad_norm": 0.10562390089035034,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 84900
+    },
+    {
+      "epoch": 8.729487970388648,
+      "grad_norm": 0.06857279688119888,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 84903
+    },
+    {
+      "epoch": 8.729796421961751,
+      "grad_norm": 0.07322374731302261,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 84906
+    },
+    {
+      "epoch": 8.730104873534856,
+      "grad_norm": 0.09952128678560257,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 84909
+    },
+    {
+      "epoch": 8.730413325107959,
+      "grad_norm": 0.04598178341984749,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 84912
+    },
+    {
+      "epoch": 8.730721776681062,
+      "grad_norm": 0.08658500760793686,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 84915
+    },
+    {
+      "epoch": 8.731030228254165,
+      "grad_norm": 0.06861251592636108,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 84918
+    },
+    {
+      "epoch": 8.731338679827267,
+      "grad_norm": 0.07586270570755005,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 84921
+    },
+    {
+      "epoch": 8.73164713140037,
+      "grad_norm": 0.08280608057975769,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 84924
+    },
+    {
+      "epoch": 8.731955582973473,
+      "grad_norm": 0.04060471057891846,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 84927
+    },
+    {
+      "epoch": 8.732264034546576,
+      "grad_norm": 0.0505189448595047,
+      "learning_rate": 0.01,
+      "loss": 1.9731,
+      "step": 84930
+    },
+    {
+      "epoch": 8.732572486119679,
+      "grad_norm": 0.04092942178249359,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 84933
+    },
+    {
+      "epoch": 8.732880937692782,
+      "grad_norm": 0.04049207270145416,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 84936
+    },
+    {
+      "epoch": 8.733189389265885,
+      "grad_norm": 0.0332498773932457,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 84939
+    },
+    {
+      "epoch": 8.733497840838988,
+      "grad_norm": 0.04378689453005791,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 84942
+    },
+    {
+      "epoch": 8.73380629241209,
+      "grad_norm": 0.05257491394877434,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 84945
+    },
+    {
+      "epoch": 8.734114743985195,
+      "grad_norm": 0.07151827961206436,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 84948
+    },
+    {
+      "epoch": 8.734423195558298,
+      "grad_norm": 0.05070538818836212,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 84951
+    },
+    {
+      "epoch": 8.734731647131401,
+      "grad_norm": 0.13645638525485992,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 84954
+    },
+    {
+      "epoch": 8.735040098704504,
+      "grad_norm": 0.15855565667152405,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 84957
+    },
+    {
+      "epoch": 8.735348550277607,
+      "grad_norm": 0.10551951825618744,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 84960
+    },
+    {
+      "epoch": 8.73565700185071,
+      "grad_norm": 0.10616222769021988,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 84963
+    },
+    {
+      "epoch": 8.735965453423812,
+      "grad_norm": 0.05103769153356552,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 84966
+    },
+    {
+      "epoch": 8.736273904996915,
+      "grad_norm": 0.042684778571128845,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 84969
+    },
+    {
+      "epoch": 8.736582356570018,
+      "grad_norm": 0.04346473887562752,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 84972
+    },
+    {
+      "epoch": 8.736890808143121,
+      "grad_norm": 0.04694516584277153,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 84975
+    },
+    {
+      "epoch": 8.737199259716224,
+      "grad_norm": 0.03958739712834358,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 84978
+    },
+    {
+      "epoch": 8.737507711289327,
+      "grad_norm": 0.03674643859267235,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 84981
+    },
+    {
+      "epoch": 8.73781616286243,
+      "grad_norm": 0.08573244512081146,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 84984
+    },
+    {
+      "epoch": 8.738124614435534,
+      "grad_norm": 0.07804060727357864,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 84987
+    },
+    {
+      "epoch": 8.738433066008637,
+      "grad_norm": 0.06877141445875168,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 84990
+    },
+    {
+      "epoch": 8.73874151758174,
+      "grad_norm": 0.07265405356884003,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 84993
+    },
+    {
+      "epoch": 8.739049969154843,
+      "grad_norm": 0.09696147590875626,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 84996
+    },
+    {
+      "epoch": 8.739358420727946,
+      "grad_norm": 0.0650443360209465,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 84999
+    },
+    {
+      "epoch": 8.739666872301049,
+      "grad_norm": 0.054474495351314545,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 85002
+    },
+    {
+      "epoch": 8.739975323874152,
+      "grad_norm": 0.03832100331783295,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 85005
+    },
+    {
+      "epoch": 8.740283775447255,
+      "grad_norm": 0.027986668050289154,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 85008
+    },
+    {
+      "epoch": 8.740592227020358,
+      "grad_norm": 0.032918307930231094,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 85011
+    },
+    {
+      "epoch": 8.74090067859346,
+      "grad_norm": 0.07551144808530807,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 85014
+    },
+    {
+      "epoch": 8.741209130166563,
+      "grad_norm": 0.07230864465236664,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 85017
+    },
+    {
+      "epoch": 8.741517581739666,
+      "grad_norm": 0.08824942260980606,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 85020
+    },
+    {
+      "epoch": 8.741826033312769,
+      "grad_norm": 0.07013983279466629,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 85023
+    },
+    {
+      "epoch": 8.742134484885874,
+      "grad_norm": 0.10559581965208054,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 85026
+    },
+    {
+      "epoch": 8.742442936458977,
+      "grad_norm": 0.05553511530160904,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 85029
+    },
+    {
+      "epoch": 8.74275138803208,
+      "grad_norm": 0.051644206047058105,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 85032
+    },
+    {
+      "epoch": 8.743059839605182,
+      "grad_norm": 0.07303252071142197,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 85035
+    },
+    {
+      "epoch": 8.743368291178285,
+      "grad_norm": 0.04157792776823044,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 85038
+    },
+    {
+      "epoch": 8.743676742751388,
+      "grad_norm": 0.09174187481403351,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 85041
+    },
+    {
+      "epoch": 8.743985194324491,
+      "grad_norm": 0.07028010487556458,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 85044
+    },
+    {
+      "epoch": 8.744293645897594,
+      "grad_norm": 0.07988093048334122,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 85047
+    },
+    {
+      "epoch": 8.744602097470697,
+      "grad_norm": 0.04523743689060211,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 85050
+    },
+    {
+      "epoch": 8.7449105490438,
+      "grad_norm": 0.07089216262102127,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 85053
+    },
+    {
+      "epoch": 8.745219000616903,
+      "grad_norm": 0.06905648857355118,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 85056
+    },
+    {
+      "epoch": 8.745527452190005,
+      "grad_norm": 0.08611857891082764,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 85059
+    },
+    {
+      "epoch": 8.745835903763108,
+      "grad_norm": 0.03267739713191986,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 85062
+    },
+    {
+      "epoch": 8.746144355336213,
+      "grad_norm": 0.10321325063705444,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 85065
+    },
+    {
+      "epoch": 8.746452806909316,
+      "grad_norm": 0.10448356717824936,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 85068
+    },
+    {
+      "epoch": 8.746761258482419,
+      "grad_norm": 0.10813847929239273,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 85071
+    },
+    {
+      "epoch": 8.747069710055522,
+      "grad_norm": 0.07072289288043976,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 85074
+    },
+    {
+      "epoch": 8.747378161628625,
+      "grad_norm": 0.07409197837114334,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 85077
+    },
+    {
+      "epoch": 8.747686613201727,
+      "grad_norm": 0.06067260727286339,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 85080
+    },
+    {
+      "epoch": 8.74799506477483,
+      "grad_norm": 0.06234930828213692,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 85083
+    },
+    {
+      "epoch": 8.748303516347933,
+      "grad_norm": 0.08919490873813629,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 85086
+    },
+    {
+      "epoch": 8.748611967921036,
+      "grad_norm": 0.0363055095076561,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 85089
+    },
+    {
+      "epoch": 8.748920419494139,
+      "grad_norm": 0.0357215516269207,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 85092
+    },
+    {
+      "epoch": 8.749228871067242,
+      "grad_norm": 0.06636667251586914,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 85095
+    },
+    {
+      "epoch": 8.749537322640345,
+      "grad_norm": 0.06666699051856995,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 85098
+    },
+    {
+      "epoch": 8.749845774213448,
+      "grad_norm": 0.06280805170536041,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 85101
+    },
+    {
+      "epoch": 8.750154225786552,
+      "grad_norm": 0.05007186904549599,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 85104
+    },
+    {
+      "epoch": 8.750462677359655,
+      "grad_norm": 0.05247753486037254,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 85107
+    },
+    {
+      "epoch": 8.750771128932758,
+      "grad_norm": 0.06049021705985069,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 85110
+    },
+    {
+      "epoch": 8.751079580505861,
+      "grad_norm": 0.042631614953279495,
+      "learning_rate": 0.01,
+      "loss": 1.9899,
+      "step": 85113
+    },
+    {
+      "epoch": 8.751388032078964,
+      "grad_norm": 0.04226498678326607,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 85116
+    },
+    {
+      "epoch": 8.751696483652067,
+      "grad_norm": 0.10684603452682495,
+      "learning_rate": 0.01,
+      "loss": 1.9712,
+      "step": 85119
+    },
+    {
+      "epoch": 8.75200493522517,
+      "grad_norm": 0.1066361740231514,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 85122
+    },
+    {
+      "epoch": 8.752313386798273,
+      "grad_norm": 0.058565832674503326,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 85125
+    },
+    {
+      "epoch": 8.752621838371375,
+      "grad_norm": 0.039739158004522324,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 85128
+    },
+    {
+      "epoch": 8.752930289944478,
+      "grad_norm": 0.03576759994029999,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 85131
+    },
+    {
+      "epoch": 8.753238741517581,
+      "grad_norm": 0.09221669286489487,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 85134
+    },
+    {
+      "epoch": 8.753547193090684,
+      "grad_norm": 0.04816079139709473,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 85137
+    },
+    {
+      "epoch": 8.753855644663787,
+      "grad_norm": 0.03829638287425041,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 85140
+    },
+    {
+      "epoch": 8.754164096236892,
+      "grad_norm": 0.04541938751935959,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 85143
+    },
+    {
+      "epoch": 8.754472547809995,
+      "grad_norm": 0.03273008018732071,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 85146
+    },
+    {
+      "epoch": 8.754780999383097,
+      "grad_norm": 0.06781888753175735,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 85149
+    },
+    {
+      "epoch": 8.7550894509562,
+      "grad_norm": 0.08772855252027512,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 85152
+    },
+    {
+      "epoch": 8.755397902529303,
+      "grad_norm": 0.05312326177954674,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 85155
+    },
+    {
+      "epoch": 8.755706354102406,
+      "grad_norm": 0.13668350875377655,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 85158
+    },
+    {
+      "epoch": 8.756014805675509,
+      "grad_norm": 0.1036950945854187,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 85161
+    },
+    {
+      "epoch": 8.756323257248612,
+      "grad_norm": 0.03806331753730774,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 85164
+    },
+    {
+      "epoch": 8.756631708821715,
+      "grad_norm": 0.0354275219142437,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 85167
+    },
+    {
+      "epoch": 8.756940160394818,
+      "grad_norm": 0.0587974451482296,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 85170
+    },
+    {
+      "epoch": 8.75724861196792,
+      "grad_norm": 0.09037166833877563,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 85173
+    },
+    {
+      "epoch": 8.757557063541023,
+      "grad_norm": 0.045411787927150726,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 85176
+    },
+    {
+      "epoch": 8.757865515114126,
+      "grad_norm": 0.08137772977352142,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 85179
+    },
+    {
+      "epoch": 8.758173966687231,
+      "grad_norm": 0.08651778101921082,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 85182
+    },
+    {
+      "epoch": 8.758482418260334,
+      "grad_norm": 0.06294746696949005,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 85185
+    },
+    {
+      "epoch": 8.758790869833437,
+      "grad_norm": 0.11187928915023804,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 85188
+    },
+    {
+      "epoch": 8.75909932140654,
+      "grad_norm": 0.062168415635824203,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 85191
+    },
+    {
+      "epoch": 8.759407772979642,
+      "grad_norm": 0.03791145980358124,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 85194
+    },
+    {
+      "epoch": 8.759716224552745,
+      "grad_norm": 0.04850441589951515,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 85197
+    },
+    {
+      "epoch": 8.760024676125848,
+      "grad_norm": 0.04435930401086807,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 85200
+    },
+    {
+      "epoch": 8.760333127698951,
+      "grad_norm": 0.11299605667591095,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 85203
+    },
+    {
+      "epoch": 8.760641579272054,
+      "grad_norm": 0.04934825748205185,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 85206
+    },
+    {
+      "epoch": 8.760950030845157,
+      "grad_norm": 0.09066156297922134,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 85209
+    },
+    {
+      "epoch": 8.76125848241826,
+      "grad_norm": 0.07747044414281845,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 85212
+    },
+    {
+      "epoch": 8.761566933991363,
+      "grad_norm": 0.04040999710559845,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 85215
+    },
+    {
+      "epoch": 8.761875385564466,
+      "grad_norm": 0.07093887031078339,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 85218
+    },
+    {
+      "epoch": 8.76218383713757,
+      "grad_norm": 0.05974037945270538,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 85221
+    },
+    {
+      "epoch": 8.762492288710673,
+      "grad_norm": 0.1216970756649971,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 85224
+    },
+    {
+      "epoch": 8.762800740283776,
+      "grad_norm": 0.12822993099689484,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 85227
+    },
+    {
+      "epoch": 8.763109191856879,
+      "grad_norm": 0.10270580649375916,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 85230
+    },
+    {
+      "epoch": 8.763417643429982,
+      "grad_norm": 0.06854567676782608,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 85233
+    },
+    {
+      "epoch": 8.763726095003085,
+      "grad_norm": 0.03591831400990486,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 85236
+    },
+    {
+      "epoch": 8.764034546576188,
+      "grad_norm": 0.0472213476896286,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 85239
+    },
+    {
+      "epoch": 8.76434299814929,
+      "grad_norm": 0.10494640469551086,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 85242
+    },
+    {
+      "epoch": 8.764651449722393,
+      "grad_norm": 0.055173519998788834,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 85245
+    },
+    {
+      "epoch": 8.764959901295496,
+      "grad_norm": 0.05719782039523125,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 85248
+    },
+    {
+      "epoch": 8.765268352868599,
+      "grad_norm": 0.048283543437719345,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 85251
+    },
+    {
+      "epoch": 8.765576804441702,
+      "grad_norm": 0.05473318323493004,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 85254
+    },
+    {
+      "epoch": 8.765885256014805,
+      "grad_norm": 0.037164732813835144,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 85257
+    },
+    {
+      "epoch": 8.76619370758791,
+      "grad_norm": 0.039799343794584274,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 85260
+    },
+    {
+      "epoch": 8.766502159161012,
+      "grad_norm": 0.13168968260288239,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 85263
+    },
+    {
+      "epoch": 8.766810610734115,
+      "grad_norm": 0.04143999144434929,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 85266
+    },
+    {
+      "epoch": 8.767119062307218,
+      "grad_norm": 0.052542950958013535,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 85269
+    },
+    {
+      "epoch": 8.767427513880321,
+      "grad_norm": 0.04957433044910431,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 85272
+    },
+    {
+      "epoch": 8.767735965453424,
+      "grad_norm": 0.07333894819021225,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 85275
+    },
+    {
+      "epoch": 8.768044417026527,
+      "grad_norm": 0.04119938984513283,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 85278
+    },
+    {
+      "epoch": 8.76835286859963,
+      "grad_norm": 0.03327082470059395,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 85281
+    },
+    {
+      "epoch": 8.768661320172733,
+      "grad_norm": 0.04693789780139923,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 85284
+    },
+    {
+      "epoch": 8.768969771745835,
+      "grad_norm": 0.06504803150892258,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 85287
+    },
+    {
+      "epoch": 8.769278223318938,
+      "grad_norm": 0.04598322883248329,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 85290
+    },
+    {
+      "epoch": 8.769586674892041,
+      "grad_norm": 0.03206741437315941,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 85293
+    },
+    {
+      "epoch": 8.769895126465144,
+      "grad_norm": 0.05733088403940201,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 85296
+    },
+    {
+      "epoch": 8.770203578038249,
+      "grad_norm": 0.07618236541748047,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 85299
+    },
+    {
+      "epoch": 8.770512029611352,
+      "grad_norm": 0.05973696708679199,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 85302
+    },
+    {
+      "epoch": 8.770820481184455,
+      "grad_norm": 0.05072012543678284,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 85305
+    },
+    {
+      "epoch": 8.771128932757557,
+      "grad_norm": 0.1093197762966156,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 85308
+    },
+    {
+      "epoch": 8.77143738433066,
+      "grad_norm": 0.04258941113948822,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 85311
+    },
+    {
+      "epoch": 8.771745835903763,
+      "grad_norm": 0.04937519505620003,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 85314
+    },
+    {
+      "epoch": 8.772054287476866,
+      "grad_norm": 0.043309785425662994,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 85317
+    },
+    {
+      "epoch": 8.772362739049969,
+      "grad_norm": 0.044104646891355515,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 85320
+    },
+    {
+      "epoch": 8.772671190623072,
+      "grad_norm": 0.05477212369441986,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 85323
+    },
+    {
+      "epoch": 8.772979642196175,
+      "grad_norm": 0.05495447665452957,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 85326
+    },
+    {
+      "epoch": 8.773288093769278,
+      "grad_norm": 0.10645133256912231,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 85329
+    },
+    {
+      "epoch": 8.77359654534238,
+      "grad_norm": 0.07867071777582169,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 85332
+    },
+    {
+      "epoch": 8.773904996915483,
+      "grad_norm": 0.11647599190473557,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 85335
+    },
+    {
+      "epoch": 8.774213448488588,
+      "grad_norm": 0.03819744288921356,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 85338
+    },
+    {
+      "epoch": 8.774521900061691,
+      "grad_norm": 0.08988303691148758,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 85341
+    },
+    {
+      "epoch": 8.774830351634794,
+      "grad_norm": 0.0678759217262268,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 85344
+    },
+    {
+      "epoch": 8.775138803207897,
+      "grad_norm": 0.08218473941087723,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 85347
+    },
+    {
+      "epoch": 8.775447254781,
+      "grad_norm": 0.07374904304742813,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 85350
+    },
+    {
+      "epoch": 8.775755706354103,
+      "grad_norm": 0.07110534608364105,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 85353
+    },
+    {
+      "epoch": 8.776064157927205,
+      "grad_norm": 0.0576554536819458,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 85356
+    },
+    {
+      "epoch": 8.776372609500308,
+      "grad_norm": 0.08372887223958969,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 85359
+    },
+    {
+      "epoch": 8.776681061073411,
+      "grad_norm": 0.06293521821498871,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 85362
+    },
+    {
+      "epoch": 8.776989512646514,
+      "grad_norm": 0.047408126294612885,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 85365
+    },
+    {
+      "epoch": 8.777297964219617,
+      "grad_norm": 0.07918205112218857,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 85368
+    },
+    {
+      "epoch": 8.77760641579272,
+      "grad_norm": 0.04142564535140991,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 85371
+    },
+    {
+      "epoch": 8.777914867365823,
+      "grad_norm": 0.031209861859679222,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 85374
+    },
+    {
+      "epoch": 8.778223318938927,
+      "grad_norm": 0.048874612897634506,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 85377
+    },
+    {
+      "epoch": 8.77853177051203,
+      "grad_norm": 0.04210768640041351,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 85380
+    },
+    {
+      "epoch": 8.778840222085133,
+      "grad_norm": 0.11695470660924911,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 85383
+    },
+    {
+      "epoch": 8.779148673658236,
+      "grad_norm": 0.08034269511699677,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 85386
+    },
+    {
+      "epoch": 8.779457125231339,
+      "grad_norm": 0.1190868392586708,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 85389
+    },
+    {
+      "epoch": 8.779765576804442,
+      "grad_norm": 0.0867137685418129,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 85392
+    },
+    {
+      "epoch": 8.780074028377545,
+      "grad_norm": 0.08815090358257294,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 85395
+    },
+    {
+      "epoch": 8.780382479950648,
+      "grad_norm": 0.04542379826307297,
+      "learning_rate": 0.01,
+      "loss": 1.9873,
+      "step": 85398
+    },
+    {
+      "epoch": 8.78069093152375,
+      "grad_norm": 0.05640220269560814,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 85401
+    },
+    {
+      "epoch": 8.780999383096853,
+      "grad_norm": 0.08306989073753357,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 85404
+    },
+    {
+      "epoch": 8.781307834669956,
+      "grad_norm": 0.11904260516166687,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 85407
+    },
+    {
+      "epoch": 8.78161628624306,
+      "grad_norm": 0.11178072541952133,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 85410
+    },
+    {
+      "epoch": 8.781924737816162,
+      "grad_norm": 0.09281167387962341,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 85413
+    },
+    {
+      "epoch": 8.782233189389267,
+      "grad_norm": 0.055970724672079086,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 85416
+    },
+    {
+      "epoch": 8.78254164096237,
+      "grad_norm": 0.05929992347955704,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 85419
+    },
+    {
+      "epoch": 8.782850092535472,
+      "grad_norm": 0.05237572267651558,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 85422
+    },
+    {
+      "epoch": 8.783158544108575,
+      "grad_norm": 0.08916623145341873,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 85425
+    },
+    {
+      "epoch": 8.783466995681678,
+      "grad_norm": 0.058517616242170334,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 85428
+    },
+    {
+      "epoch": 8.783775447254781,
+      "grad_norm": 0.04481247439980507,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 85431
+    },
+    {
+      "epoch": 8.784083898827884,
+      "grad_norm": 0.04817540943622589,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 85434
+    },
+    {
+      "epoch": 8.784392350400987,
+      "grad_norm": 0.0524100698530674,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 85437
+    },
+    {
+      "epoch": 8.78470080197409,
+      "grad_norm": 0.1673521101474762,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 85440
+    },
+    {
+      "epoch": 8.785009253547193,
+      "grad_norm": 0.058675531297922134,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 85443
+    },
+    {
+      "epoch": 8.785317705120296,
+      "grad_norm": 0.0696008950471878,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 85446
+    },
+    {
+      "epoch": 8.785626156693398,
+      "grad_norm": 0.05638658627867699,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 85449
+    },
+    {
+      "epoch": 8.785934608266501,
+      "grad_norm": 0.0473102405667305,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 85452
+    },
+    {
+      "epoch": 8.786243059839606,
+      "grad_norm": 0.04515746235847473,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 85455
+    },
+    {
+      "epoch": 8.786551511412709,
+      "grad_norm": 0.043432414531707764,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 85458
+    },
+    {
+      "epoch": 8.786859962985812,
+      "grad_norm": 0.0580633319914341,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 85461
+    },
+    {
+      "epoch": 8.787168414558915,
+      "grad_norm": 0.07870703935623169,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 85464
+    },
+    {
+      "epoch": 8.787476866132018,
+      "grad_norm": 0.08649545162916183,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 85467
+    },
+    {
+      "epoch": 8.78778531770512,
+      "grad_norm": 0.06479387730360031,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 85470
+    },
+    {
+      "epoch": 8.788093769278223,
+      "grad_norm": 0.07489529997110367,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 85473
+    },
+    {
+      "epoch": 8.788402220851326,
+      "grad_norm": 0.08207201212644577,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 85476
+    },
+    {
+      "epoch": 8.788710672424429,
+      "grad_norm": 0.07542536407709122,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 85479
+    },
+    {
+      "epoch": 8.789019123997532,
+      "grad_norm": 0.0705515518784523,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 85482
+    },
+    {
+      "epoch": 8.789327575570635,
+      "grad_norm": 0.05835072696208954,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 85485
+    },
+    {
+      "epoch": 8.789636027143738,
+      "grad_norm": 0.05363980308175087,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 85488
+    },
+    {
+      "epoch": 8.78994447871684,
+      "grad_norm": 0.10389215499162674,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 85491
+    },
+    {
+      "epoch": 8.790252930289945,
+      "grad_norm": 0.0551920086145401,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 85494
+    },
+    {
+      "epoch": 8.790561381863048,
+      "grad_norm": 0.054502177983522415,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 85497
+    },
+    {
+      "epoch": 8.790869833436151,
+      "grad_norm": 0.04676192253828049,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 85500
+    },
+    {
+      "epoch": 8.791178285009254,
+      "grad_norm": 0.0815708339214325,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 85503
+    },
+    {
+      "epoch": 8.791486736582357,
+      "grad_norm": 0.07840388268232346,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 85506
+    },
+    {
+      "epoch": 8.79179518815546,
+      "grad_norm": 0.08336468786001205,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 85509
+    },
+    {
+      "epoch": 8.792103639728563,
+      "grad_norm": 0.1331731379032135,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 85512
+    },
+    {
+      "epoch": 8.792412091301665,
+      "grad_norm": 0.10310441255569458,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 85515
+    },
+    {
+      "epoch": 8.792720542874768,
+      "grad_norm": 0.03826957195997238,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 85518
+    },
+    {
+      "epoch": 8.793028994447871,
+      "grad_norm": 0.04112190008163452,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 85521
+    },
+    {
+      "epoch": 8.793337446020974,
+      "grad_norm": 0.04075111821293831,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 85524
+    },
+    {
+      "epoch": 8.793645897594077,
+      "grad_norm": 0.03421452268958092,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 85527
+    },
+    {
+      "epoch": 8.79395434916718,
+      "grad_norm": 0.1482706516981125,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 85530
+    },
+    {
+      "epoch": 8.794262800740285,
+      "grad_norm": 0.11172833293676376,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 85533
+    },
+    {
+      "epoch": 8.794571252313387,
+      "grad_norm": 0.07505451142787933,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 85536
+    },
+    {
+      "epoch": 8.79487970388649,
+      "grad_norm": 0.07401387393474579,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 85539
+    },
+    {
+      "epoch": 8.795188155459593,
+      "grad_norm": 0.06274157762527466,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 85542
+    },
+    {
+      "epoch": 8.795496607032696,
+      "grad_norm": 0.05745621770620346,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 85545
+    },
+    {
+      "epoch": 8.795805058605799,
+      "grad_norm": 0.04131782427430153,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 85548
+    },
+    {
+      "epoch": 8.796113510178902,
+      "grad_norm": 0.04630117118358612,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 85551
+    },
+    {
+      "epoch": 8.796421961752005,
+      "grad_norm": 0.07542744278907776,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 85554
+    },
+    {
+      "epoch": 8.796730413325108,
+      "grad_norm": 0.10365857928991318,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 85557
+    },
+    {
+      "epoch": 8.79703886489821,
+      "grad_norm": 0.11497282981872559,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 85560
+    },
+    {
+      "epoch": 8.797347316471313,
+      "grad_norm": 0.05868663638830185,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 85563
+    },
+    {
+      "epoch": 8.797655768044416,
+      "grad_norm": 0.04206192493438721,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 85566
+    },
+    {
+      "epoch": 8.79796421961752,
+      "grad_norm": 0.038855090737342834,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 85569
+    },
+    {
+      "epoch": 8.798272671190624,
+      "grad_norm": 0.0406762957572937,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 85572
+    },
+    {
+      "epoch": 8.798581122763727,
+      "grad_norm": 0.12336095422506332,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 85575
+    },
+    {
+      "epoch": 8.79888957433683,
+      "grad_norm": 0.05633760988712311,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 85578
+    },
+    {
+      "epoch": 8.799198025909932,
+      "grad_norm": 0.05526827648282051,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 85581
+    },
+    {
+      "epoch": 8.799506477483035,
+      "grad_norm": 0.0719655379652977,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 85584
+    },
+    {
+      "epoch": 8.799814929056138,
+      "grad_norm": 0.15878325700759888,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 85587
+    },
+    {
+      "epoch": 8.800123380629241,
+      "grad_norm": 0.07367381453514099,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 85590
+    },
+    {
+      "epoch": 8.800431832202344,
+      "grad_norm": 0.08002124726772308,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 85593
+    },
+    {
+      "epoch": 8.800740283775447,
+      "grad_norm": 0.06393203884363174,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 85596
+    },
+    {
+      "epoch": 8.80104873534855,
+      "grad_norm": 0.03970704600214958,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 85599
+    },
+    {
+      "epoch": 8.801357186921653,
+      "grad_norm": 0.052820831537246704,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 85602
+    },
+    {
+      "epoch": 8.801665638494756,
+      "grad_norm": 0.06489507108926773,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 85605
+    },
+    {
+      "epoch": 8.801974090067858,
+      "grad_norm": 0.06538572162389755,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 85608
+    },
+    {
+      "epoch": 8.802282541640963,
+      "grad_norm": 0.041309911757707596,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 85611
+    },
+    {
+      "epoch": 8.802590993214066,
+      "grad_norm": 0.11032170802354813,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 85614
+    },
+    {
+      "epoch": 8.802899444787169,
+      "grad_norm": 0.03958607837557793,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 85617
+    },
+    {
+      "epoch": 8.803207896360272,
+      "grad_norm": 0.06811465322971344,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 85620
+    },
+    {
+      "epoch": 8.803516347933375,
+      "grad_norm": 0.0968390479683876,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 85623
+    },
+    {
+      "epoch": 8.803824799506478,
+      "grad_norm": 0.06796689331531525,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 85626
+    },
+    {
+      "epoch": 8.80413325107958,
+      "grad_norm": 0.05755221098661423,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 85629
+    },
+    {
+      "epoch": 8.804441702652683,
+      "grad_norm": 0.12345452606678009,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 85632
+    },
+    {
+      "epoch": 8.804750154225786,
+      "grad_norm": 0.07588624209165573,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 85635
+    },
+    {
+      "epoch": 8.805058605798889,
+      "grad_norm": 0.04973727837204933,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 85638
+    },
+    {
+      "epoch": 8.805367057371992,
+      "grad_norm": 0.03331485763192177,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 85641
+    },
+    {
+      "epoch": 8.805675508945095,
+      "grad_norm": 0.03431696817278862,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 85644
+    },
+    {
+      "epoch": 8.805983960518198,
+      "grad_norm": 0.045221637934446335,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 85647
+    },
+    {
+      "epoch": 8.806292412091302,
+      "grad_norm": 0.0935794860124588,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 85650
+    },
+    {
+      "epoch": 8.806600863664405,
+      "grad_norm": 0.05588880181312561,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 85653
+    },
+    {
+      "epoch": 8.806909315237508,
+      "grad_norm": 0.039516277611255646,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 85656
+    },
+    {
+      "epoch": 8.807217766810611,
+      "grad_norm": 0.03812456503510475,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 85659
+    },
+    {
+      "epoch": 8.807526218383714,
+      "grad_norm": 0.04115702211856842,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 85662
+    },
+    {
+      "epoch": 8.807834669956817,
+      "grad_norm": 0.141036257147789,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 85665
+    },
+    {
+      "epoch": 8.80814312152992,
+      "grad_norm": 0.05860263481736183,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 85668
+    },
+    {
+      "epoch": 8.808451573103023,
+      "grad_norm": 0.05528310313820839,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 85671
+    },
+    {
+      "epoch": 8.808760024676126,
+      "grad_norm": 0.03591969981789589,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 85674
+    },
+    {
+      "epoch": 8.809068476249228,
+      "grad_norm": 0.13484379649162292,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 85677
+    },
+    {
+      "epoch": 8.809376927822331,
+      "grad_norm": 0.08338242024183273,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 85680
+    },
+    {
+      "epoch": 8.809685379395434,
+      "grad_norm": 0.061400432139635086,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 85683
+    },
+    {
+      "epoch": 8.809993830968537,
+      "grad_norm": 0.029880136251449585,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 85686
+    },
+    {
+      "epoch": 8.810302282541642,
+      "grad_norm": 0.05177440494298935,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 85689
+    },
+    {
+      "epoch": 8.810610734114745,
+      "grad_norm": 0.08280415832996368,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 85692
+    },
+    {
+      "epoch": 8.810919185687847,
+      "grad_norm": 0.09497891366481781,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 85695
+    },
+    {
+      "epoch": 8.81122763726095,
+      "grad_norm": 0.0856609046459198,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 85698
+    },
+    {
+      "epoch": 8.811536088834053,
+      "grad_norm": 0.05290984734892845,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 85701
+    },
+    {
+      "epoch": 8.811844540407156,
+      "grad_norm": 0.04517572000622749,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 85704
+    },
+    {
+      "epoch": 8.812152991980259,
+      "grad_norm": 0.040704116225242615,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 85707
+    },
+    {
+      "epoch": 8.812461443553362,
+      "grad_norm": 0.03796176612377167,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 85710
+    },
+    {
+      "epoch": 8.812769895126465,
+      "grad_norm": 0.11047817766666412,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 85713
+    },
+    {
+      "epoch": 8.813078346699568,
+      "grad_norm": 0.09995049983263016,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 85716
+    },
+    {
+      "epoch": 8.81338679827267,
+      "grad_norm": 0.037516675889492035,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 85719
+    },
+    {
+      "epoch": 8.813695249845773,
+      "grad_norm": 0.08218607306480408,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 85722
+    },
+    {
+      "epoch": 8.814003701418876,
+      "grad_norm": 0.09442456066608429,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 85725
+    },
+    {
+      "epoch": 8.814312152991981,
+      "grad_norm": 0.05343955382704735,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 85728
+    },
+    {
+      "epoch": 8.814620604565084,
+      "grad_norm": 0.0756995677947998,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 85731
+    },
+    {
+      "epoch": 8.814929056138187,
+      "grad_norm": 0.07240238785743713,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 85734
+    },
+    {
+      "epoch": 8.81523750771129,
+      "grad_norm": 0.058714382350444794,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 85737
+    },
+    {
+      "epoch": 8.815545959284393,
+      "grad_norm": 0.08831341564655304,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 85740
+    },
+    {
+      "epoch": 8.815854410857495,
+      "grad_norm": 0.05908043310046196,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 85743
+    },
+    {
+      "epoch": 8.816162862430598,
+      "grad_norm": 0.04405365511775017,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 85746
+    },
+    {
+      "epoch": 8.816471314003701,
+      "grad_norm": 0.04712482914328575,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 85749
+    },
+    {
+      "epoch": 8.816779765576804,
+      "grad_norm": 0.10971397906541824,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 85752
+    },
+    {
+      "epoch": 8.817088217149907,
+      "grad_norm": 0.06402042508125305,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 85755
+    },
+    {
+      "epoch": 8.81739666872301,
+      "grad_norm": 0.09109006822109222,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 85758
+    },
+    {
+      "epoch": 8.817705120296113,
+      "grad_norm": 0.09684780985116959,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 85761
+    },
+    {
+      "epoch": 8.818013571869216,
+      "grad_norm": 0.0380178838968277,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 85764
+    },
+    {
+      "epoch": 8.81832202344232,
+      "grad_norm": 0.07201346755027771,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 85767
+    },
+    {
+      "epoch": 8.818630475015423,
+      "grad_norm": 0.05873304605484009,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 85770
+    },
+    {
+      "epoch": 8.818938926588526,
+      "grad_norm": 0.06054871901869774,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 85773
+    },
+    {
+      "epoch": 8.819247378161629,
+      "grad_norm": 0.06931430101394653,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 85776
+    },
+    {
+      "epoch": 8.819555829734732,
+      "grad_norm": 0.08289149403572083,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 85779
+    },
+    {
+      "epoch": 8.819864281307835,
+      "grad_norm": 0.07705207169055939,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 85782
+    },
+    {
+      "epoch": 8.820172732880938,
+      "grad_norm": 0.1018168181180954,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 85785
+    },
+    {
+      "epoch": 8.82048118445404,
+      "grad_norm": 0.061625886708498,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 85788
+    },
+    {
+      "epoch": 8.820789636027143,
+      "grad_norm": 0.050136588513851166,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 85791
+    },
+    {
+      "epoch": 8.821098087600246,
+      "grad_norm": 0.040379468351602554,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 85794
+    },
+    {
+      "epoch": 8.82140653917335,
+      "grad_norm": 0.032702744007110596,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 85797
+    },
+    {
+      "epoch": 8.821714990746452,
+      "grad_norm": 0.04039518162608147,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 85800
+    },
+    {
+      "epoch": 8.822023442319555,
+      "grad_norm": 0.13541679084300995,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 85803
+    },
+    {
+      "epoch": 8.82233189389266,
+      "grad_norm": 0.11711031198501587,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 85806
+    },
+    {
+      "epoch": 8.822640345465762,
+      "grad_norm": 0.13755975663661957,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 85809
+    },
+    {
+      "epoch": 8.822948797038865,
+      "grad_norm": 0.10734470188617706,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 85812
+    },
+    {
+      "epoch": 8.823257248611968,
+      "grad_norm": 0.112040676176548,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 85815
+    },
+    {
+      "epoch": 8.823565700185071,
+      "grad_norm": 0.07922054082155228,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 85818
+    },
+    {
+      "epoch": 8.823874151758174,
+      "grad_norm": 0.04796600341796875,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 85821
+    },
+    {
+      "epoch": 8.824182603331277,
+      "grad_norm": 0.05426159128546715,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 85824
+    },
+    {
+      "epoch": 8.82449105490438,
+      "grad_norm": 0.04682827368378639,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 85827
+    },
+    {
+      "epoch": 8.824799506477483,
+      "grad_norm": 0.08252523839473724,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 85830
+    },
+    {
+      "epoch": 8.825107958050586,
+      "grad_norm": 0.04207620769739151,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 85833
+    },
+    {
+      "epoch": 8.825416409623688,
+      "grad_norm": 0.0742509737610817,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 85836
+    },
+    {
+      "epoch": 8.825724861196791,
+      "grad_norm": 0.03609723225235939,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 85839
+    },
+    {
+      "epoch": 8.826033312769894,
+      "grad_norm": 0.0418362058699131,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 85842
+    },
+    {
+      "epoch": 8.826341764342999,
+      "grad_norm": 0.05531947314739227,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 85845
+    },
+    {
+      "epoch": 8.826650215916102,
+      "grad_norm": 0.05285338684916496,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 85848
+    },
+    {
+      "epoch": 8.826958667489205,
+      "grad_norm": 0.1251527965068817,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 85851
+    },
+    {
+      "epoch": 8.827267119062308,
+      "grad_norm": 0.038710471242666245,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 85854
+    },
+    {
+      "epoch": 8.82757557063541,
+      "grad_norm": 0.04924651235342026,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 85857
+    },
+    {
+      "epoch": 8.827884022208513,
+      "grad_norm": 0.07448245584964752,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 85860
+    },
+    {
+      "epoch": 8.828192473781616,
+      "grad_norm": 0.03801583871245384,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 85863
+    },
+    {
+      "epoch": 8.828500925354719,
+      "grad_norm": 0.04570269584655762,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 85866
+    },
+    {
+      "epoch": 8.828809376927822,
+      "grad_norm": 0.14020396769046783,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 85869
+    },
+    {
+      "epoch": 8.829117828500925,
+      "grad_norm": 0.07285384088754654,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 85872
+    },
+    {
+      "epoch": 8.829426280074028,
+      "grad_norm": 0.05504117161035538,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 85875
+    },
+    {
+      "epoch": 8.82973473164713,
+      "grad_norm": 0.05914220213890076,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 85878
+    },
+    {
+      "epoch": 8.830043183220234,
+      "grad_norm": 0.04905472695827484,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 85881
+    },
+    {
+      "epoch": 8.830351634793338,
+      "grad_norm": 0.04060604050755501,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 85884
+    },
+    {
+      "epoch": 8.830660086366441,
+      "grad_norm": 0.034003157168626785,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 85887
+    },
+    {
+      "epoch": 8.830968537939544,
+      "grad_norm": 0.06942160427570343,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 85890
+    },
+    {
+      "epoch": 8.831276989512647,
+      "grad_norm": 0.14421160519123077,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 85893
+    },
+    {
+      "epoch": 8.83158544108575,
+      "grad_norm": 0.049919433891773224,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 85896
+    },
+    {
+      "epoch": 8.831893892658853,
+      "grad_norm": 0.03900736942887306,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 85899
+    },
+    {
+      "epoch": 8.832202344231955,
+      "grad_norm": 0.046023812144994736,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 85902
+    },
+    {
+      "epoch": 8.832510795805058,
+      "grad_norm": 0.09192383289337158,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 85905
+    },
+    {
+      "epoch": 8.832819247378161,
+      "grad_norm": 0.08842380344867706,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 85908
+    },
+    {
+      "epoch": 8.833127698951264,
+      "grad_norm": 0.0487644225358963,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 85911
+    },
+    {
+      "epoch": 8.833436150524367,
+      "grad_norm": 0.0899803414940834,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 85914
+    },
+    {
+      "epoch": 8.83374460209747,
+      "grad_norm": 0.09449628740549088,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 85917
+    },
+    {
+      "epoch": 8.834053053670573,
+      "grad_norm": 0.041804246604442596,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 85920
+    },
+    {
+      "epoch": 8.834361505243677,
+      "grad_norm": 0.03399631753563881,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 85923
+    },
+    {
+      "epoch": 8.83466995681678,
+      "grad_norm": 0.1013641282916069,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 85926
+    },
+    {
+      "epoch": 8.834978408389883,
+      "grad_norm": 0.07109085470438004,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 85929
+    },
+    {
+      "epoch": 8.835286859962986,
+      "grad_norm": 0.0444265753030777,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 85932
+    },
+    {
+      "epoch": 8.835595311536089,
+      "grad_norm": 0.04196979105472565,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 85935
+    },
+    {
+      "epoch": 8.835903763109192,
+      "grad_norm": 0.057135023176670074,
+      "learning_rate": 0.01,
+      "loss": 1.9931,
+      "step": 85938
+    },
+    {
+      "epoch": 8.836212214682295,
+      "grad_norm": 0.048445116728544235,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 85941
+    },
+    {
+      "epoch": 8.836520666255398,
+      "grad_norm": 0.04054427519440651,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 85944
+    },
+    {
+      "epoch": 8.8368291178285,
+      "grad_norm": 0.12011062353849411,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 85947
+    },
+    {
+      "epoch": 8.837137569401603,
+      "grad_norm": 0.13074417412281036,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 85950
+    },
+    {
+      "epoch": 8.837446020974706,
+      "grad_norm": 0.05142034962773323,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 85953
+    },
+    {
+      "epoch": 8.83775447254781,
+      "grad_norm": 0.05493297800421715,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 85956
+    },
+    {
+      "epoch": 8.838062924120912,
+      "grad_norm": 0.037361111491918564,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 85959
+    },
+    {
+      "epoch": 8.838371375694017,
+      "grad_norm": 0.04489835351705551,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 85962
+    },
+    {
+      "epoch": 8.83867982726712,
+      "grad_norm": 0.03907403722405434,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 85965
+    },
+    {
+      "epoch": 8.838988278840223,
+      "grad_norm": 0.08557716757059097,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 85968
+    },
+    {
+      "epoch": 8.839296730413325,
+      "grad_norm": 0.061117153614759445,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 85971
+    },
+    {
+      "epoch": 8.839605181986428,
+      "grad_norm": 0.09186840802431107,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 85974
+    },
+    {
+      "epoch": 8.839913633559531,
+      "grad_norm": 0.049297600984573364,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 85977
+    },
+    {
+      "epoch": 8.840222085132634,
+      "grad_norm": 0.0782669186592102,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 85980
+    },
+    {
+      "epoch": 8.840530536705737,
+      "grad_norm": 0.10112703591585159,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 85983
+    },
+    {
+      "epoch": 8.84083898827884,
+      "grad_norm": 0.0647103562951088,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 85986
+    },
+    {
+      "epoch": 8.841147439851943,
+      "grad_norm": 0.04339386895298958,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 85989
+    },
+    {
+      "epoch": 8.841455891425046,
+      "grad_norm": 0.0516306571662426,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 85992
+    },
+    {
+      "epoch": 8.841764342998149,
+      "grad_norm": 0.09764052927494049,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 85995
+    },
+    {
+      "epoch": 8.842072794571251,
+      "grad_norm": 0.060452841222286224,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 85998
+    },
+    {
+      "epoch": 8.842381246144356,
+      "grad_norm": 0.1076023206114769,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 86001
+    },
+    {
+      "epoch": 8.842689697717459,
+      "grad_norm": 0.056612275540828705,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 86004
+    },
+    {
+      "epoch": 8.842998149290562,
+      "grad_norm": 0.12334515899419785,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 86007
+    },
+    {
+      "epoch": 8.843306600863665,
+      "grad_norm": 0.05345802381634712,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 86010
+    },
+    {
+      "epoch": 8.843615052436768,
+      "grad_norm": 0.05063433572649956,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 86013
+    },
+    {
+      "epoch": 8.84392350400987,
+      "grad_norm": 0.05762813985347748,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 86016
+    },
+    {
+      "epoch": 8.844231955582973,
+      "grad_norm": 0.051242899149656296,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 86019
+    },
+    {
+      "epoch": 8.844540407156076,
+      "grad_norm": 0.045474182814359665,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 86022
+    },
+    {
+      "epoch": 8.84484885872918,
+      "grad_norm": 0.062329549342393875,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 86025
+    },
+    {
+      "epoch": 8.845157310302282,
+      "grad_norm": 0.10545872151851654,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 86028
+    },
+    {
+      "epoch": 8.845465761875385,
+      "grad_norm": 0.06179576739668846,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 86031
+    },
+    {
+      "epoch": 8.845774213448488,
+      "grad_norm": 0.10388704389333725,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 86034
+    },
+    {
+      "epoch": 8.84608266502159,
+      "grad_norm": 0.06724070012569427,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 86037
+    },
+    {
+      "epoch": 8.846391116594695,
+      "grad_norm": 0.06309358775615692,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 86040
+    },
+    {
+      "epoch": 8.846699568167798,
+      "grad_norm": 0.07256193459033966,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 86043
+    },
+    {
+      "epoch": 8.847008019740901,
+      "grad_norm": 0.06528967618942261,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 86046
+    },
+    {
+      "epoch": 8.847316471314004,
+      "grad_norm": 0.03754352405667305,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 86049
+    },
+    {
+      "epoch": 8.847624922887107,
+      "grad_norm": 0.04206917807459831,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 86052
+    },
+    {
+      "epoch": 8.84793337446021,
+      "grad_norm": 0.06092971935868263,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 86055
+    },
+    {
+      "epoch": 8.848241826033313,
+      "grad_norm": 0.055908314883708954,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 86058
+    },
+    {
+      "epoch": 8.848550277606416,
+      "grad_norm": 0.04285154864192009,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 86061
+    },
+    {
+      "epoch": 8.848858729179518,
+      "grad_norm": 0.04486127570271492,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 86064
+    },
+    {
+      "epoch": 8.849167180752621,
+      "grad_norm": 0.12317989021539688,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 86067
+    },
+    {
+      "epoch": 8.849475632325724,
+      "grad_norm": 0.04780368134379387,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 86070
+    },
+    {
+      "epoch": 8.849784083898827,
+      "grad_norm": 0.09250175207853317,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 86073
+    },
+    {
+      "epoch": 8.85009253547193,
+      "grad_norm": 0.05390471965074539,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 86076
+    },
+    {
+      "epoch": 8.850400987045035,
+      "grad_norm": 0.04068141430616379,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 86079
+    },
+    {
+      "epoch": 8.850709438618138,
+      "grad_norm": 0.037879377603530884,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 86082
+    },
+    {
+      "epoch": 8.85101789019124,
+      "grad_norm": 0.03136352822184563,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 86085
+    },
+    {
+      "epoch": 8.851326341764343,
+      "grad_norm": 0.13857915997505188,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 86088
+    },
+    {
+      "epoch": 8.851634793337446,
+      "grad_norm": 0.046125274151563644,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 86091
+    },
+    {
+      "epoch": 8.851943244910549,
+      "grad_norm": 0.11460349708795547,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 86094
+    },
+    {
+      "epoch": 8.852251696483652,
+      "grad_norm": 0.0628170296549797,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 86097
+    },
+    {
+      "epoch": 8.852560148056755,
+      "grad_norm": 0.0907411277294159,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 86100
+    },
+    {
+      "epoch": 8.852868599629858,
+      "grad_norm": 0.0656890794634819,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 86103
+    },
+    {
+      "epoch": 8.85317705120296,
+      "grad_norm": 0.0431518629193306,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 86106
+    },
+    {
+      "epoch": 8.853485502776063,
+      "grad_norm": 0.03522025793790817,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 86109
+    },
+    {
+      "epoch": 8.853793954349166,
+      "grad_norm": 0.09385208040475845,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 86112
+    },
+    {
+      "epoch": 8.85410240592227,
+      "grad_norm": 0.04548000171780586,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 86115
+    },
+    {
+      "epoch": 8.854410857495374,
+      "grad_norm": 0.06799448281526566,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 86118
+    },
+    {
+      "epoch": 8.854719309068477,
+      "grad_norm": 0.041807446628808975,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 86121
+    },
+    {
+      "epoch": 8.85502776064158,
+      "grad_norm": 0.11906972527503967,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 86124
+    },
+    {
+      "epoch": 8.855336212214683,
+      "grad_norm": 0.11152509599924088,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 86127
+    },
+    {
+      "epoch": 8.855644663787785,
+      "grad_norm": 0.12766876816749573,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 86130
+    },
+    {
+      "epoch": 8.855953115360888,
+      "grad_norm": 0.11722945421934128,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 86133
+    },
+    {
+      "epoch": 8.856261566933991,
+      "grad_norm": 0.04626612737774849,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 86136
+    },
+    {
+      "epoch": 8.856570018507094,
+      "grad_norm": 0.040816087275743484,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 86139
+    },
+    {
+      "epoch": 8.856878470080197,
+      "grad_norm": 0.0497615672647953,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 86142
+    },
+    {
+      "epoch": 8.8571869216533,
+      "grad_norm": 0.04109172523021698,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 86145
+    },
+    {
+      "epoch": 8.857495373226403,
+      "grad_norm": 0.10170968621969223,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 86148
+    },
+    {
+      "epoch": 8.857803824799506,
+      "grad_norm": 0.05853275582194328,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 86151
+    },
+    {
+      "epoch": 8.858112276372609,
+      "grad_norm": 0.04526359587907791,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 86154
+    },
+    {
+      "epoch": 8.858420727945713,
+      "grad_norm": 0.041641268879175186,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 86157
+    },
+    {
+      "epoch": 8.858729179518816,
+      "grad_norm": 0.04957803338766098,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 86160
+    },
+    {
+      "epoch": 8.859037631091919,
+      "grad_norm": 0.09154763072729111,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 86163
+    },
+    {
+      "epoch": 8.859346082665022,
+      "grad_norm": 0.12526895105838776,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 86166
+    },
+    {
+      "epoch": 8.859654534238125,
+      "grad_norm": 0.056200604885816574,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 86169
+    },
+    {
+      "epoch": 8.859962985811228,
+      "grad_norm": 0.11231166869401932,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 86172
+    },
+    {
+      "epoch": 8.86027143738433,
+      "grad_norm": 0.11459070444107056,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 86175
+    },
+    {
+      "epoch": 8.860579888957433,
+      "grad_norm": 0.11196603626012802,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 86178
+    },
+    {
+      "epoch": 8.860888340530536,
+      "grad_norm": 0.056678954511880875,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 86181
+    },
+    {
+      "epoch": 8.86119679210364,
+      "grad_norm": 0.049753881990909576,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 86184
+    },
+    {
+      "epoch": 8.861505243676742,
+      "grad_norm": 0.045943088829517365,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 86187
+    },
+    {
+      "epoch": 8.861813695249845,
+      "grad_norm": 0.03190896287560463,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 86190
+    },
+    {
+      "epoch": 8.862122146822948,
+      "grad_norm": 0.034997474402189255,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 86193
+    },
+    {
+      "epoch": 8.862430598396053,
+      "grad_norm": 0.07974711060523987,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 86196
+    },
+    {
+      "epoch": 8.862739049969155,
+      "grad_norm": 0.10291502624750137,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 86199
+    },
+    {
+      "epoch": 8.863047501542258,
+      "grad_norm": 0.12553256750106812,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 86202
+    },
+    {
+      "epoch": 8.863355953115361,
+      "grad_norm": 0.11128824949264526,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 86205
+    },
+    {
+      "epoch": 8.863664404688464,
+      "grad_norm": 0.09532744437456131,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 86208
+    },
+    {
+      "epoch": 8.863972856261567,
+      "grad_norm": 0.06013484299182892,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 86211
+    },
+    {
+      "epoch": 8.86428130783467,
+      "grad_norm": 0.04983198642730713,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 86214
+    },
+    {
+      "epoch": 8.864589759407773,
+      "grad_norm": 0.04304705560207367,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 86217
+    },
+    {
+      "epoch": 8.864898210980876,
+      "grad_norm": 0.04451330006122589,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 86220
+    },
+    {
+      "epoch": 8.865206662553978,
+      "grad_norm": 0.052429914474487305,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 86223
+    },
+    {
+      "epoch": 8.865515114127081,
+      "grad_norm": 0.15532724559307098,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 86226
+    },
+    {
+      "epoch": 8.865823565700184,
+      "grad_norm": 0.06498439610004425,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 86229
+    },
+    {
+      "epoch": 8.866132017273289,
+      "grad_norm": 0.028920544311404228,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 86232
+    },
+    {
+      "epoch": 8.866440468846392,
+      "grad_norm": 0.048817794770002365,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 86235
+    },
+    {
+      "epoch": 8.866748920419495,
+      "grad_norm": 0.08832227438688278,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 86238
+    },
+    {
+      "epoch": 8.867057371992598,
+      "grad_norm": 0.06445325165987015,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 86241
+    },
+    {
+      "epoch": 8.8673658235657,
+      "grad_norm": 0.04837856814265251,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 86244
+    },
+    {
+      "epoch": 8.867674275138803,
+      "grad_norm": 0.056823018938302994,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 86247
+    },
+    {
+      "epoch": 8.867982726711906,
+      "grad_norm": 0.0441332571208477,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 86250
+    },
+    {
+      "epoch": 8.86829117828501,
+      "grad_norm": 0.03880564495921135,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 86253
+    },
+    {
+      "epoch": 8.868599629858112,
+      "grad_norm": 0.09879901260137558,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 86256
+    },
+    {
+      "epoch": 8.868908081431215,
+      "grad_norm": 0.07412566244602203,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 86259
+    },
+    {
+      "epoch": 8.869216533004318,
+      "grad_norm": 0.039541300386190414,
+      "learning_rate": 0.01,
+      "loss": 1.9007,
+      "step": 86262
+    },
+    {
+      "epoch": 8.86952498457742,
+      "grad_norm": 0.043198730796575546,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 86265
+    },
+    {
+      "epoch": 8.869833436150524,
+      "grad_norm": 0.08590435981750488,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 86268
+    },
+    {
+      "epoch": 8.870141887723628,
+      "grad_norm": 0.06476505845785141,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 86271
+    },
+    {
+      "epoch": 8.870450339296731,
+      "grad_norm": 0.10437195748090744,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 86274
+    },
+    {
+      "epoch": 8.870758790869834,
+      "grad_norm": 0.1389506608247757,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 86277
+    },
+    {
+      "epoch": 8.871067242442937,
+      "grad_norm": 0.13642540574073792,
+      "learning_rate": 0.01,
+      "loss": 1.9792,
+      "step": 86280
+    },
+    {
+      "epoch": 8.87137569401604,
+      "grad_norm": 0.11155140399932861,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 86283
+    },
+    {
+      "epoch": 8.871684145589143,
+      "grad_norm": 0.06370694935321808,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 86286
+    },
+    {
+      "epoch": 8.871992597162246,
+      "grad_norm": 0.04209110513329506,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 86289
+    },
+    {
+      "epoch": 8.872301048735348,
+      "grad_norm": 0.047301799058914185,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 86292
+    },
+    {
+      "epoch": 8.872609500308451,
+      "grad_norm": 0.10057863593101501,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 86295
+    },
+    {
+      "epoch": 8.872917951881554,
+      "grad_norm": 0.08219350874423981,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 86298
+    },
+    {
+      "epoch": 8.873226403454657,
+      "grad_norm": 0.08292625844478607,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 86301
+    },
+    {
+      "epoch": 8.87353485502776,
+      "grad_norm": 0.07303151488304138,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 86304
+    },
+    {
+      "epoch": 8.873843306600863,
+      "grad_norm": 0.058230068534612656,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 86307
+    },
+    {
+      "epoch": 8.874151758173968,
+      "grad_norm": 0.042837414890527725,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 86310
+    },
+    {
+      "epoch": 8.87446020974707,
+      "grad_norm": 0.05397916957736015,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 86313
+    },
+    {
+      "epoch": 8.874768661320173,
+      "grad_norm": 0.04088616371154785,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 86316
+    },
+    {
+      "epoch": 8.875077112893276,
+      "grad_norm": 0.06871747970581055,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 86319
+    },
+    {
+      "epoch": 8.875385564466379,
+      "grad_norm": 0.10163968056440353,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 86322
+    },
+    {
+      "epoch": 8.875694016039482,
+      "grad_norm": 0.06452172249555588,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 86325
+    },
+    {
+      "epoch": 8.876002467612585,
+      "grad_norm": 0.07819397747516632,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 86328
+    },
+    {
+      "epoch": 8.876310919185688,
+      "grad_norm": 0.0875793993473053,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 86331
+    },
+    {
+      "epoch": 8.87661937075879,
+      "grad_norm": 0.07515552639961243,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 86334
+    },
+    {
+      "epoch": 8.876927822331893,
+      "grad_norm": 0.03545791283249855,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 86337
+    },
+    {
+      "epoch": 8.877236273904996,
+      "grad_norm": 0.10219214111566544,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 86340
+    },
+    {
+      "epoch": 8.8775447254781,
+      "grad_norm": 0.07150894403457642,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 86343
+    },
+    {
+      "epoch": 8.877853177051202,
+      "grad_norm": 0.07047196477651596,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 86346
+    },
+    {
+      "epoch": 8.878161628624307,
+      "grad_norm": 0.058551233261823654,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 86349
+    },
+    {
+      "epoch": 8.87847008019741,
+      "grad_norm": 0.09620966762304306,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 86352
+    },
+    {
+      "epoch": 8.878778531770513,
+      "grad_norm": 0.043070271611213684,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 86355
+    },
+    {
+      "epoch": 8.879086983343615,
+      "grad_norm": 0.049491941928863525,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 86358
+    },
+    {
+      "epoch": 8.879395434916718,
+      "grad_norm": 0.07890214771032333,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 86361
+    },
+    {
+      "epoch": 8.879703886489821,
+      "grad_norm": 0.11723397672176361,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 86364
+    },
+    {
+      "epoch": 8.880012338062924,
+      "grad_norm": 0.04008464887738228,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 86367
+    },
+    {
+      "epoch": 8.880320789636027,
+      "grad_norm": 0.047027554363012314,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 86370
+    },
+    {
+      "epoch": 8.88062924120913,
+      "grad_norm": 0.0419597327709198,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 86373
+    },
+    {
+      "epoch": 8.880937692782233,
+      "grad_norm": 0.03923768550157547,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 86376
+    },
+    {
+      "epoch": 8.881246144355336,
+      "grad_norm": 0.09833482652902603,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 86379
+    },
+    {
+      "epoch": 8.881554595928439,
+      "grad_norm": 0.06715640425682068,
+      "learning_rate": 0.01,
+      "loss": 1.986,
+      "step": 86382
+    },
+    {
+      "epoch": 8.881863047501541,
+      "grad_norm": 0.07140083611011505,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 86385
+    },
+    {
+      "epoch": 8.882171499074646,
+      "grad_norm": 0.13441288471221924,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 86388
+    },
+    {
+      "epoch": 8.882479950647749,
+      "grad_norm": 0.08413909375667572,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 86391
+    },
+    {
+      "epoch": 8.882788402220852,
+      "grad_norm": 0.08267387747764587,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 86394
+    },
+    {
+      "epoch": 8.883096853793955,
+      "grad_norm": 0.056718725711107254,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 86397
+    },
+    {
+      "epoch": 8.883405305367058,
+      "grad_norm": 0.042766470462083817,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 86400
+    },
+    {
+      "epoch": 8.88371375694016,
+      "grad_norm": 0.08797338604927063,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 86403
+    },
+    {
+      "epoch": 8.884022208513263,
+      "grad_norm": 0.043157946318387985,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 86406
+    },
+    {
+      "epoch": 8.884330660086366,
+      "grad_norm": 0.044488899409770966,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 86409
+    },
+    {
+      "epoch": 8.88463911165947,
+      "grad_norm": 0.06237373128533363,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 86412
+    },
+    {
+      "epoch": 8.884947563232572,
+      "grad_norm": 0.03562946245074272,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 86415
+    },
+    {
+      "epoch": 8.885256014805675,
+      "grad_norm": 0.1371944397687912,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 86418
+    },
+    {
+      "epoch": 8.885564466378778,
+      "grad_norm": 0.0579112283885479,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 86421
+    },
+    {
+      "epoch": 8.88587291795188,
+      "grad_norm": 0.07281753420829773,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 86424
+    },
+    {
+      "epoch": 8.886181369524985,
+      "grad_norm": 0.04341709613800049,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 86427
+    },
+    {
+      "epoch": 8.886489821098088,
+      "grad_norm": 0.0515635646879673,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 86430
+    },
+    {
+      "epoch": 8.886798272671191,
+      "grad_norm": 0.02943897247314453,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 86433
+    },
+    {
+      "epoch": 8.887106724244294,
+      "grad_norm": 0.0403241291642189,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 86436
+    },
+    {
+      "epoch": 8.887415175817397,
+      "grad_norm": 0.049521833658218384,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 86439
+    },
+    {
+      "epoch": 8.8877236273905,
+      "grad_norm": 0.03380339592695236,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 86442
+    },
+    {
+      "epoch": 8.888032078963603,
+      "grad_norm": 0.10890775173902512,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 86445
+    },
+    {
+      "epoch": 8.888340530536706,
+      "grad_norm": 0.08257224410772324,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 86448
+    },
+    {
+      "epoch": 8.888648982109808,
+      "grad_norm": 0.07431552559137344,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 86451
+    },
+    {
+      "epoch": 8.888957433682911,
+      "grad_norm": 0.05777169391512871,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 86454
+    },
+    {
+      "epoch": 8.889265885256014,
+      "grad_norm": 0.03272586315870285,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 86457
+    },
+    {
+      "epoch": 8.889574336829117,
+      "grad_norm": 0.036487750709056854,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 86460
+    },
+    {
+      "epoch": 8.889882788402222,
+      "grad_norm": 0.10916222631931305,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 86463
+    },
+    {
+      "epoch": 8.890191239975325,
+      "grad_norm": 0.039324864745140076,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 86466
+    },
+    {
+      "epoch": 8.890499691548428,
+      "grad_norm": 0.08545426279306412,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 86469
+    },
+    {
+      "epoch": 8.89080814312153,
+      "grad_norm": 0.08365831524133682,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 86472
+    },
+    {
+      "epoch": 8.891116594694633,
+      "grad_norm": 0.051737960427999496,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 86475
+    },
+    {
+      "epoch": 8.891425046267736,
+      "grad_norm": 0.047015320509672165,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 86478
+    },
+    {
+      "epoch": 8.891733497840839,
+      "grad_norm": 0.05336649343371391,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 86481
+    },
+    {
+      "epoch": 8.892041949413942,
+      "grad_norm": 0.045248694717884064,
+      "learning_rate": 0.01,
+      "loss": 1.9817,
+      "step": 86484
+    },
+    {
+      "epoch": 8.892350400987045,
+      "grad_norm": 0.10528840124607086,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 86487
+    },
+    {
+      "epoch": 8.892658852560148,
+      "grad_norm": 0.03929603472352028,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 86490
+    },
+    {
+      "epoch": 8.89296730413325,
+      "grad_norm": 0.10144193470478058,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 86493
+    },
+    {
+      "epoch": 8.893275755706354,
+      "grad_norm": 0.07188859581947327,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 86496
+    },
+    {
+      "epoch": 8.893584207279456,
+      "grad_norm": 0.1337640881538391,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 86499
+    },
+    {
+      "epoch": 8.893892658852561,
+      "grad_norm": 0.03903505951166153,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 86502
+    },
+    {
+      "epoch": 8.894201110425664,
+      "grad_norm": 0.14735199511051178,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 86505
+    },
+    {
+      "epoch": 8.894509561998767,
+      "grad_norm": 0.05615058168768883,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 86508
+    },
+    {
+      "epoch": 8.89481801357187,
+      "grad_norm": 0.13095706701278687,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 86511
+    },
+    {
+      "epoch": 8.895126465144973,
+      "grad_norm": 0.064826600253582,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 86514
+    },
+    {
+      "epoch": 8.895434916718076,
+      "grad_norm": 0.03320586308836937,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 86517
+    },
+    {
+      "epoch": 8.895743368291178,
+      "grad_norm": 0.035892751067876816,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 86520
+    },
+    {
+      "epoch": 8.896051819864281,
+      "grad_norm": 0.06792799383401871,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 86523
+    },
+    {
+      "epoch": 8.896360271437384,
+      "grad_norm": 0.03676078841090202,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 86526
+    },
+    {
+      "epoch": 8.896668723010487,
+      "grad_norm": 0.04583815485239029,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 86529
+    },
+    {
+      "epoch": 8.89697717458359,
+      "grad_norm": 0.0372345894575119,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 86532
+    },
+    {
+      "epoch": 8.897285626156693,
+      "grad_norm": 0.10942641645669937,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 86535
+    },
+    {
+      "epoch": 8.897594077729796,
+      "grad_norm": 0.07769682258367538,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 86538
+    },
+    {
+      "epoch": 8.8979025293029,
+      "grad_norm": 0.04752153158187866,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 86541
+    },
+    {
+      "epoch": 8.898210980876003,
+      "grad_norm": 0.05308716744184494,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 86544
+    },
+    {
+      "epoch": 8.898519432449106,
+      "grad_norm": 0.06798425316810608,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 86547
+    },
+    {
+      "epoch": 8.898827884022209,
+      "grad_norm": 0.04612553119659424,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 86550
+    },
+    {
+      "epoch": 8.899136335595312,
+      "grad_norm": 0.04493824392557144,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 86553
+    },
+    {
+      "epoch": 8.899444787168415,
+      "grad_norm": 0.04563165828585625,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 86556
+    },
+    {
+      "epoch": 8.899753238741518,
+      "grad_norm": 0.08082949370145798,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 86559
+    },
+    {
+      "epoch": 8.90006169031462,
+      "grad_norm": 0.05760066583752632,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 86562
+    },
+    {
+      "epoch": 8.900370141887723,
+      "grad_norm": 0.04055507853627205,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 86565
+    },
+    {
+      "epoch": 8.900678593460826,
+      "grad_norm": 0.12776689231395721,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 86568
+    },
+    {
+      "epoch": 8.90098704503393,
+      "grad_norm": 0.033561255782842636,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 86571
+    },
+    {
+      "epoch": 8.901295496607032,
+      "grad_norm": 0.04030522331595421,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 86574
+    },
+    {
+      "epoch": 8.901603948180135,
+      "grad_norm": 0.041517045348882675,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 86577
+    },
+    {
+      "epoch": 8.90191239975324,
+      "grad_norm": 0.13993878662586212,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 86580
+    },
+    {
+      "epoch": 8.902220851326343,
+      "grad_norm": 0.04944363608956337,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 86583
+    },
+    {
+      "epoch": 8.902529302899445,
+      "grad_norm": 0.0440177246928215,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 86586
+    },
+    {
+      "epoch": 8.902837754472548,
+      "grad_norm": 0.14848758280277252,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 86589
+    },
+    {
+      "epoch": 8.903146206045651,
+      "grad_norm": 0.04407470300793648,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 86592
+    },
+    {
+      "epoch": 8.903454657618754,
+      "grad_norm": 0.04730687290430069,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 86595
+    },
+    {
+      "epoch": 8.903763109191857,
+      "grad_norm": 0.04748192057013512,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 86598
+    },
+    {
+      "epoch": 8.90407156076496,
+      "grad_norm": 0.04898633435368538,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 86601
+    },
+    {
+      "epoch": 8.904380012338063,
+      "grad_norm": 0.040645040571689606,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 86604
+    },
+    {
+      "epoch": 8.904688463911166,
+      "grad_norm": 0.08757542073726654,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 86607
+    },
+    {
+      "epoch": 8.904996915484269,
+      "grad_norm": 0.04103793948888779,
+      "learning_rate": 0.01,
+      "loss": 1.9742,
+      "step": 86610
+    },
+    {
+      "epoch": 8.905305367057371,
+      "grad_norm": 0.09652022272348404,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 86613
+    },
+    {
+      "epoch": 8.905613818630474,
+      "grad_norm": 0.09843424707651138,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 86616
+    },
+    {
+      "epoch": 8.905922270203579,
+      "grad_norm": 0.05483357980847359,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 86619
+    },
+    {
+      "epoch": 8.906230721776682,
+      "grad_norm": 0.04492351412773132,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 86622
+    },
+    {
+      "epoch": 8.906539173349785,
+      "grad_norm": 0.037706341594457626,
+      "learning_rate": 0.01,
+      "loss": 1.9847,
+      "step": 86625
+    },
+    {
+      "epoch": 8.906847624922888,
+      "grad_norm": 0.13539664447307587,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 86628
+    },
+    {
+      "epoch": 8.90715607649599,
+      "grad_norm": 0.038968782871961594,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 86631
+    },
+    {
+      "epoch": 8.907464528069093,
+      "grad_norm": 0.04004844278097153,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 86634
+    },
+    {
+      "epoch": 8.907772979642196,
+      "grad_norm": 0.05966116860508919,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 86637
+    },
+    {
+      "epoch": 8.9080814312153,
+      "grad_norm": 0.032242551445961,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 86640
+    },
+    {
+      "epoch": 8.908389882788402,
+      "grad_norm": 0.04070298373699188,
+      "learning_rate": 0.01,
+      "loss": 1.9888,
+      "step": 86643
+    },
+    {
+      "epoch": 8.908698334361505,
+      "grad_norm": 0.09953150898218155,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 86646
+    },
+    {
+      "epoch": 8.909006785934608,
+      "grad_norm": 0.09204036742448807,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 86649
+    },
+    {
+      "epoch": 8.90931523750771,
+      "grad_norm": 0.1648140400648117,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 86652
+    },
+    {
+      "epoch": 8.909623689080814,
+      "grad_norm": 0.09835021197795868,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 86655
+    },
+    {
+      "epoch": 8.909932140653918,
+      "grad_norm": 0.06145261600613594,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 86658
+    },
+    {
+      "epoch": 8.910240592227021,
+      "grad_norm": 0.060555968433618546,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 86661
+    },
+    {
+      "epoch": 8.910549043800124,
+      "grad_norm": 0.03184669092297554,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 86664
+    },
+    {
+      "epoch": 8.910857495373227,
+      "grad_norm": 0.10093247890472412,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 86667
+    },
+    {
+      "epoch": 8.91116594694633,
+      "grad_norm": 0.041602443903684616,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 86670
+    },
+    {
+      "epoch": 8.911474398519433,
+      "grad_norm": 0.05512705445289612,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 86673
+    },
+    {
+      "epoch": 8.911782850092536,
+      "grad_norm": 0.052319008857011795,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 86676
+    },
+    {
+      "epoch": 8.912091301665638,
+      "grad_norm": 0.09350787103176117,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 86679
+    },
+    {
+      "epoch": 8.912399753238741,
+      "grad_norm": 0.05260681360960007,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 86682
+    },
+    {
+      "epoch": 8.912708204811844,
+      "grad_norm": 0.03641582280397415,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 86685
+    },
+    {
+      "epoch": 8.913016656384947,
+      "grad_norm": 0.12317151576280594,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 86688
+    },
+    {
+      "epoch": 8.91332510795805,
+      "grad_norm": 0.05290086567401886,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 86691
+    },
+    {
+      "epoch": 8.913633559531153,
+      "grad_norm": 0.1165960282087326,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 86694
+    },
+    {
+      "epoch": 8.913942011104258,
+      "grad_norm": 0.03640148788690567,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 86697
+    },
+    {
+      "epoch": 8.91425046267736,
+      "grad_norm": 0.08245276659727097,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 86700
+    },
+    {
+      "epoch": 8.914558914250463,
+      "grad_norm": 0.035647567361593246,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 86703
+    },
+    {
+      "epoch": 8.914867365823566,
+      "grad_norm": 0.055706936866045,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 86706
+    },
+    {
+      "epoch": 8.915175817396669,
+      "grad_norm": 0.1239318773150444,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 86709
+    },
+    {
+      "epoch": 8.915484268969772,
+      "grad_norm": 0.08160992711782455,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 86712
+    },
+    {
+      "epoch": 8.915792720542875,
+      "grad_norm": 0.085784412920475,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 86715
+    },
+    {
+      "epoch": 8.916101172115978,
+      "grad_norm": 0.07544608414173126,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 86718
+    },
+    {
+      "epoch": 8.91640962368908,
+      "grad_norm": 0.05595572292804718,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 86721
+    },
+    {
+      "epoch": 8.916718075262184,
+      "grad_norm": 0.03742340952157974,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 86724
+    },
+    {
+      "epoch": 8.917026526835286,
+      "grad_norm": 0.027727732434868813,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 86727
+    },
+    {
+      "epoch": 8.91733497840839,
+      "grad_norm": 0.0767802819609642,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 86730
+    },
+    {
+      "epoch": 8.917643429981492,
+      "grad_norm": 0.07831424474716187,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 86733
+    },
+    {
+      "epoch": 8.917951881554597,
+      "grad_norm": 0.07590378075838089,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 86736
+    },
+    {
+      "epoch": 8.9182603331277,
+      "grad_norm": 0.05041801184415817,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 86739
+    },
+    {
+      "epoch": 8.918568784700803,
+      "grad_norm": 0.039486199617385864,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 86742
+    },
+    {
+      "epoch": 8.918877236273905,
+      "grad_norm": 0.048161353915929794,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 86745
+    },
+    {
+      "epoch": 8.919185687847008,
+      "grad_norm": 0.04758733883500099,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 86748
+    },
+    {
+      "epoch": 8.919494139420111,
+      "grad_norm": 0.03529585897922516,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 86751
+    },
+    {
+      "epoch": 8.919802590993214,
+      "grad_norm": 0.033544719219207764,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 86754
+    },
+    {
+      "epoch": 8.920111042566317,
+      "grad_norm": 0.030024485662579536,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 86757
+    },
+    {
+      "epoch": 8.92041949413942,
+      "grad_norm": 0.04932880774140358,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 86760
+    },
+    {
+      "epoch": 8.920727945712523,
+      "grad_norm": 0.08779890090227127,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 86763
+    },
+    {
+      "epoch": 8.921036397285626,
+      "grad_norm": 0.04417244344949722,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 86766
+    },
+    {
+      "epoch": 8.921344848858729,
+      "grad_norm": 0.10385674238204956,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 86769
+    },
+    {
+      "epoch": 8.921653300431831,
+      "grad_norm": 0.05979326739907265,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 86772
+    },
+    {
+      "epoch": 8.921961752004936,
+      "grad_norm": 0.0792737677693367,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 86775
+    },
+    {
+      "epoch": 8.922270203578039,
+      "grad_norm": 0.07325568795204163,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 86778
+    },
+    {
+      "epoch": 8.922578655151142,
+      "grad_norm": 0.03263154253363609,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 86781
+    },
+    {
+      "epoch": 8.922887106724245,
+      "grad_norm": 0.04729807749390602,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 86784
+    },
+    {
+      "epoch": 8.923195558297348,
+      "grad_norm": 0.03822244331240654,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 86787
+    },
+    {
+      "epoch": 8.92350400987045,
+      "grad_norm": 0.06542357057332993,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 86790
+    },
+    {
+      "epoch": 8.923812461443553,
+      "grad_norm": 0.057463910430669785,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 86793
+    },
+    {
+      "epoch": 8.924120913016656,
+      "grad_norm": 0.09920434653759003,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 86796
+    },
+    {
+      "epoch": 8.92442936458976,
+      "grad_norm": 0.11771022528409958,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 86799
+    },
+    {
+      "epoch": 8.924737816162862,
+      "grad_norm": 0.06464545428752899,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 86802
+    },
+    {
+      "epoch": 8.925046267735965,
+      "grad_norm": 0.051915016025304794,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 86805
+    },
+    {
+      "epoch": 8.925354719309068,
+      "grad_norm": 0.05679486319422722,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 86808
+    },
+    {
+      "epoch": 8.92566317088217,
+      "grad_norm": 0.0370950922369957,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 86811
+    },
+    {
+      "epoch": 8.925971622455275,
+      "grad_norm": 0.03662877529859543,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 86814
+    },
+    {
+      "epoch": 8.926280074028378,
+      "grad_norm": 0.11973265558481216,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 86817
+    },
+    {
+      "epoch": 8.926588525601481,
+      "grad_norm": 0.051350776106119156,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 86820
+    },
+    {
+      "epoch": 8.926896977174584,
+      "grad_norm": 0.08452027291059494,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 86823
+    },
+    {
+      "epoch": 8.927205428747687,
+      "grad_norm": 0.06621302664279938,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 86826
+    },
+    {
+      "epoch": 8.92751388032079,
+      "grad_norm": 0.09285139292478561,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 86829
+    },
+    {
+      "epoch": 8.927822331893893,
+      "grad_norm": 0.05777567997574806,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 86832
+    },
+    {
+      "epoch": 8.928130783466996,
+      "grad_norm": 0.1285361349582672,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 86835
+    },
+    {
+      "epoch": 8.928439235040099,
+      "grad_norm": 0.03722556307911873,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 86838
+    },
+    {
+      "epoch": 8.928747686613201,
+      "grad_norm": 0.03609219565987587,
+      "learning_rate": 0.01,
+      "loss": 1.9835,
+      "step": 86841
+    },
+    {
+      "epoch": 8.929056138186304,
+      "grad_norm": 0.06217125803232193,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 86844
+    },
+    {
+      "epoch": 8.929364589759407,
+      "grad_norm": 0.10021287202835083,
+      "learning_rate": 0.01,
+      "loss": 1.9727,
+      "step": 86847
+    },
+    {
+      "epoch": 8.92967304133251,
+      "grad_norm": 0.15343636274337769,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 86850
+    },
+    {
+      "epoch": 8.929981492905615,
+      "grad_norm": 0.03606894984841347,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 86853
+    },
+    {
+      "epoch": 8.930289944478718,
+      "grad_norm": 0.034981515258550644,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 86856
+    },
+    {
+      "epoch": 8.93059839605182,
+      "grad_norm": 0.043932538479566574,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 86859
+    },
+    {
+      "epoch": 8.930906847624923,
+      "grad_norm": 0.05214953050017357,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 86862
+    },
+    {
+      "epoch": 8.931215299198026,
+      "grad_norm": 0.06819628179073334,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 86865
+    },
+    {
+      "epoch": 8.93152375077113,
+      "grad_norm": 0.04332321509718895,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 86868
+    },
+    {
+      "epoch": 8.931832202344232,
+      "grad_norm": 0.04354182258248329,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 86871
+    },
+    {
+      "epoch": 8.932140653917335,
+      "grad_norm": 0.03317633643746376,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 86874
+    },
+    {
+      "epoch": 8.932449105490438,
+      "grad_norm": 0.0436597503721714,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 86877
+    },
+    {
+      "epoch": 8.93275755706354,
+      "grad_norm": 0.17182035744190216,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 86880
+    },
+    {
+      "epoch": 8.933066008636644,
+      "grad_norm": 0.04708772152662277,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 86883
+    },
+    {
+      "epoch": 8.933374460209746,
+      "grad_norm": 0.10389839112758636,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 86886
+    },
+    {
+      "epoch": 8.93368291178285,
+      "grad_norm": 0.07319999486207962,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 86889
+    },
+    {
+      "epoch": 8.933991363355954,
+      "grad_norm": 0.05096595361828804,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 86892
+    },
+    {
+      "epoch": 8.934299814929057,
+      "grad_norm": 0.03280860185623169,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 86895
+    },
+    {
+      "epoch": 8.93460826650216,
+      "grad_norm": 0.04681391641497612,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 86898
+    },
+    {
+      "epoch": 8.934916718075263,
+      "grad_norm": 0.07191100716590881,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 86901
+    },
+    {
+      "epoch": 8.935225169648366,
+      "grad_norm": 0.050077635794878006,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 86904
+    },
+    {
+      "epoch": 8.935533621221468,
+      "grad_norm": 0.03628779575228691,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 86907
+    },
+    {
+      "epoch": 8.935842072794571,
+      "grad_norm": 0.08990596234798431,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 86910
+    },
+    {
+      "epoch": 8.936150524367674,
+      "grad_norm": 0.061381399631500244,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 86913
+    },
+    {
+      "epoch": 8.936458975940777,
+      "grad_norm": 0.08864876627922058,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 86916
+    },
+    {
+      "epoch": 8.93676742751388,
+      "grad_norm": 0.06189894676208496,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 86919
+    },
+    {
+      "epoch": 8.937075879086983,
+      "grad_norm": 0.048378266394138336,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 86922
+    },
+    {
+      "epoch": 8.937384330660086,
+      "grad_norm": 0.04020404815673828,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 86925
+    },
+    {
+      "epoch": 8.937692782233189,
+      "grad_norm": 0.11243297904729843,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 86928
+    },
+    {
+      "epoch": 8.938001233806293,
+      "grad_norm": 0.08314108848571777,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 86931
+    },
+    {
+      "epoch": 8.938309685379396,
+      "grad_norm": 0.05910830944776535,
+      "learning_rate": 0.01,
+      "loss": 1.9065,
+      "step": 86934
+    },
+    {
+      "epoch": 8.938618136952499,
+      "grad_norm": 0.03916241601109505,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 86937
+    },
+    {
+      "epoch": 8.938926588525602,
+      "grad_norm": 0.03755800798535347,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 86940
+    },
+    {
+      "epoch": 8.939235040098705,
+      "grad_norm": 0.0976007804274559,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 86943
+    },
+    {
+      "epoch": 8.939543491671808,
+      "grad_norm": 0.1216464415192604,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 86946
+    },
+    {
+      "epoch": 8.93985194324491,
+      "grad_norm": 0.07827116549015045,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 86949
+    },
+    {
+      "epoch": 8.940160394818014,
+      "grad_norm": 0.05447017401456833,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 86952
+    },
+    {
+      "epoch": 8.940468846391116,
+      "grad_norm": 0.04793046787381172,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 86955
+    },
+    {
+      "epoch": 8.94077729796422,
+      "grad_norm": 0.03350021317601204,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 86958
+    },
+    {
+      "epoch": 8.941085749537322,
+      "grad_norm": 0.038865212351083755,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 86961
+    },
+    {
+      "epoch": 8.941394201110425,
+      "grad_norm": 0.10471334308385849,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 86964
+    },
+    {
+      "epoch": 8.941702652683528,
+      "grad_norm": 0.06126569211483002,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 86967
+    },
+    {
+      "epoch": 8.942011104256633,
+      "grad_norm": 0.08202920854091644,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 86970
+    },
+    {
+      "epoch": 8.942319555829735,
+      "grad_norm": 0.10397789627313614,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 86973
+    },
+    {
+      "epoch": 8.942628007402838,
+      "grad_norm": 0.07265036553144455,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 86976
+    },
+    {
+      "epoch": 8.942936458975941,
+      "grad_norm": 0.035652872174978256,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 86979
+    },
+    {
+      "epoch": 8.943244910549044,
+      "grad_norm": 0.10862493515014648,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 86982
+    },
+    {
+      "epoch": 8.943553362122147,
+      "grad_norm": 0.06365188211202621,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 86985
+    },
+    {
+      "epoch": 8.94386181369525,
+      "grad_norm": 0.06253690272569656,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 86988
+    },
+    {
+      "epoch": 8.944170265268353,
+      "grad_norm": 0.035485751926898956,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 86991
+    },
+    {
+      "epoch": 8.944478716841456,
+      "grad_norm": 0.03434903174638748,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 86994
+    },
+    {
+      "epoch": 8.944787168414559,
+      "grad_norm": 0.03491094708442688,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 86997
+    },
+    {
+      "epoch": 8.945095619987661,
+      "grad_norm": 0.13820096850395203,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 87000
+    },
+    {
+      "epoch": 8.945404071560764,
+      "grad_norm": 0.11694854497909546,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 87003
+    },
+    {
+      "epoch": 8.945712523133867,
+      "grad_norm": 0.06137260049581528,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 87006
+    },
+    {
+      "epoch": 8.946020974706972,
+      "grad_norm": 0.041880011558532715,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 87009
+    },
+    {
+      "epoch": 8.946329426280075,
+      "grad_norm": 0.047957465052604675,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 87012
+    },
+    {
+      "epoch": 8.946637877853178,
+      "grad_norm": 0.048485107719898224,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 87015
+    },
+    {
+      "epoch": 8.94694632942628,
+      "grad_norm": 0.0791233703494072,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 87018
+    },
+    {
+      "epoch": 8.947254780999383,
+      "grad_norm": 0.07178815454244614,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 87021
+    },
+    {
+      "epoch": 8.947563232572486,
+      "grad_norm": 0.040599554777145386,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 87024
+    },
+    {
+      "epoch": 8.94787168414559,
+      "grad_norm": 0.05260973051190376,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 87027
+    },
+    {
+      "epoch": 8.948180135718692,
+      "grad_norm": 0.057267699390649796,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 87030
+    },
+    {
+      "epoch": 8.948488587291795,
+      "grad_norm": 0.036324381828308105,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 87033
+    },
+    {
+      "epoch": 8.948797038864898,
+      "grad_norm": 0.044209253042936325,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 87036
+    },
+    {
+      "epoch": 8.949105490438,
+      "grad_norm": 0.10309609025716782,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 87039
+    },
+    {
+      "epoch": 8.949413942011104,
+      "grad_norm": 0.10782811045646667,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 87042
+    },
+    {
+      "epoch": 8.949722393584207,
+      "grad_norm": 0.047176990658044815,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 87045
+    },
+    {
+      "epoch": 8.950030845157311,
+      "grad_norm": 0.05118265375494957,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 87048
+    },
+    {
+      "epoch": 8.950339296730414,
+      "grad_norm": 0.04261369630694389,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 87051
+    },
+    {
+      "epoch": 8.950647748303517,
+      "grad_norm": 0.05787837132811546,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 87054
+    },
+    {
+      "epoch": 8.95095619987662,
+      "grad_norm": 0.038357578217983246,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 87057
+    },
+    {
+      "epoch": 8.951264651449723,
+      "grad_norm": 0.05408007279038429,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 87060
+    },
+    {
+      "epoch": 8.951573103022826,
+      "grad_norm": 0.08756155520677567,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 87063
+    },
+    {
+      "epoch": 8.951881554595928,
+      "grad_norm": 0.11525848507881165,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 87066
+    },
+    {
+      "epoch": 8.952190006169031,
+      "grad_norm": 0.04911641404032707,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 87069
+    },
+    {
+      "epoch": 8.952498457742134,
+      "grad_norm": 0.11142449826002121,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 87072
+    },
+    {
+      "epoch": 8.952806909315237,
+      "grad_norm": 0.057849034667015076,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 87075
+    },
+    {
+      "epoch": 8.95311536088834,
+      "grad_norm": 0.054354969412088394,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 87078
+    },
+    {
+      "epoch": 8.953423812461443,
+      "grad_norm": 0.11387278884649277,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 87081
+    },
+    {
+      "epoch": 8.953732264034546,
+      "grad_norm": 0.043686944991350174,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 87084
+    },
+    {
+      "epoch": 8.95404071560765,
+      "grad_norm": 0.11665847152471542,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 87087
+    },
+    {
+      "epoch": 8.954349167180753,
+      "grad_norm": 0.04742182791233063,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 87090
+    },
+    {
+      "epoch": 8.954657618753856,
+      "grad_norm": 0.06386453658342361,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 87093
+    },
+    {
+      "epoch": 8.95496607032696,
+      "grad_norm": 0.047558024525642395,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 87096
+    },
+    {
+      "epoch": 8.955274521900062,
+      "grad_norm": 0.04645347595214844,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 87099
+    },
+    {
+      "epoch": 8.955582973473165,
+      "grad_norm": 0.0866866484284401,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 87102
+    },
+    {
+      "epoch": 8.955891425046268,
+      "grad_norm": 0.07707598805427551,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 87105
+    },
+    {
+      "epoch": 8.95619987661937,
+      "grad_norm": 0.03881002217531204,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 87108
+    },
+    {
+      "epoch": 8.956508328192474,
+      "grad_norm": 0.04681535065174103,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 87111
+    },
+    {
+      "epoch": 8.956816779765576,
+      "grad_norm": 0.09788013994693756,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 87114
+    },
+    {
+      "epoch": 8.95712523133868,
+      "grad_norm": 0.06216461583971977,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 87117
+    },
+    {
+      "epoch": 8.957433682911782,
+      "grad_norm": 0.06608910113573074,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 87120
+    },
+    {
+      "epoch": 8.957742134484885,
+      "grad_norm": 0.08615975826978683,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 87123
+    },
+    {
+      "epoch": 8.95805058605799,
+      "grad_norm": 0.08540789783000946,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 87126
+    },
+    {
+      "epoch": 8.958359037631093,
+      "grad_norm": 0.10461477190256119,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 87129
+    },
+    {
+      "epoch": 8.958667489204196,
+      "grad_norm": 0.06833938509225845,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 87132
+    },
+    {
+      "epoch": 8.958975940777298,
+      "grad_norm": 0.048409439623355865,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 87135
+    },
+    {
+      "epoch": 8.959284392350401,
+      "grad_norm": 0.030358022078871727,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 87138
+    },
+    {
+      "epoch": 8.959592843923504,
+      "grad_norm": 0.044521231204271317,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 87141
+    },
+    {
+      "epoch": 8.959901295496607,
+      "grad_norm": 0.045875221490859985,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 87144
+    },
+    {
+      "epoch": 8.96020974706971,
+      "grad_norm": 0.05634180083870888,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 87147
+    },
+    {
+      "epoch": 8.960518198642813,
+      "grad_norm": 0.04120972007513046,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 87150
+    },
+    {
+      "epoch": 8.960826650215916,
+      "grad_norm": 0.1378134787082672,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 87153
+    },
+    {
+      "epoch": 8.961135101789019,
+      "grad_norm": 0.09309209138154984,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 87156
+    },
+    {
+      "epoch": 8.961443553362122,
+      "grad_norm": 0.07671626657247543,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 87159
+    },
+    {
+      "epoch": 8.961752004935224,
+      "grad_norm": 0.03704199194908142,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 87162
+    },
+    {
+      "epoch": 8.962060456508329,
+      "grad_norm": 0.05708852410316467,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 87165
+    },
+    {
+      "epoch": 8.962368908081432,
+      "grad_norm": 0.04932823032140732,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 87168
+    },
+    {
+      "epoch": 8.962677359654535,
+      "grad_norm": 0.06841997802257538,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 87171
+    },
+    {
+      "epoch": 8.962985811227638,
+      "grad_norm": 0.09124533832073212,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 87174
+    },
+    {
+      "epoch": 8.96329426280074,
+      "grad_norm": 0.0966772809624672,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 87177
+    },
+    {
+      "epoch": 8.963602714373843,
+      "grad_norm": 0.09721286594867706,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 87180
+    },
+    {
+      "epoch": 8.963911165946946,
+      "grad_norm": 0.09772830456495285,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 87183
+    },
+    {
+      "epoch": 8.96421961752005,
+      "grad_norm": 0.061515890061855316,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 87186
+    },
+    {
+      "epoch": 8.964528069093152,
+      "grad_norm": 0.043670911341905594,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 87189
+    },
+    {
+      "epoch": 8.964836520666255,
+      "grad_norm": 0.04097547382116318,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 87192
+    },
+    {
+      "epoch": 8.965144972239358,
+      "grad_norm": 0.045037586241960526,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 87195
+    },
+    {
+      "epoch": 8.96545342381246,
+      "grad_norm": 0.06821490079164505,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 87198
+    },
+    {
+      "epoch": 8.965761875385564,
+      "grad_norm": 0.0660410225391388,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 87201
+    },
+    {
+      "epoch": 8.966070326958668,
+      "grad_norm": 0.0699450895190239,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 87204
+    },
+    {
+      "epoch": 8.966378778531771,
+      "grad_norm": 0.09934176504611969,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 87207
+    },
+    {
+      "epoch": 8.966687230104874,
+      "grad_norm": 0.047030240297317505,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 87210
+    },
+    {
+      "epoch": 8.966995681677977,
+      "grad_norm": 0.08642113953828812,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 87213
+    },
+    {
+      "epoch": 8.96730413325108,
+      "grad_norm": 0.11990761011838913,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 87216
+    },
+    {
+      "epoch": 8.967612584824183,
+      "grad_norm": 0.05154174938797951,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 87219
+    },
+    {
+      "epoch": 8.967921036397286,
+      "grad_norm": 0.03815053030848503,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 87222
+    },
+    {
+      "epoch": 8.968229487970389,
+      "grad_norm": 0.049095261842012405,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 87225
+    },
+    {
+      "epoch": 8.968537939543491,
+      "grad_norm": 0.08481363207101822,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 87228
+    },
+    {
+      "epoch": 8.968846391116594,
+      "grad_norm": 0.04782922565937042,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 87231
+    },
+    {
+      "epoch": 8.969154842689697,
+      "grad_norm": 0.03404981642961502,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 87234
+    },
+    {
+      "epoch": 8.9694632942628,
+      "grad_norm": 0.04017357528209686,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 87237
+    },
+    {
+      "epoch": 8.969771745835903,
+      "grad_norm": 0.04835909977555275,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 87240
+    },
+    {
+      "epoch": 8.970080197409008,
+      "grad_norm": 0.06477301567792892,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 87243
+    },
+    {
+      "epoch": 8.97038864898211,
+      "grad_norm": 0.09944019466638565,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 87246
+    },
+    {
+      "epoch": 8.970697100555213,
+      "grad_norm": 0.04341105371713638,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 87249
+    },
+    {
+      "epoch": 8.971005552128316,
+      "grad_norm": 0.09491550177335739,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 87252
+    },
+    {
+      "epoch": 8.97131400370142,
+      "grad_norm": 0.08826873451471329,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 87255
+    },
+    {
+      "epoch": 8.971622455274522,
+      "grad_norm": 0.09465787559747696,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 87258
+    },
+    {
+      "epoch": 8.971930906847625,
+      "grad_norm": 0.08723200112581253,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 87261
+    },
+    {
+      "epoch": 8.972239358420728,
+      "grad_norm": 0.11504415422677994,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 87264
+    },
+    {
+      "epoch": 8.97254780999383,
+      "grad_norm": 0.04780331254005432,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 87267
+    },
+    {
+      "epoch": 8.972856261566934,
+      "grad_norm": 0.06980123370885849,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 87270
+    },
+    {
+      "epoch": 8.973164713140036,
+      "grad_norm": 0.05528416857123375,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 87273
+    },
+    {
+      "epoch": 8.97347316471314,
+      "grad_norm": 0.07386885583400726,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 87276
+    },
+    {
+      "epoch": 8.973781616286242,
+      "grad_norm": 0.12215300649404526,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 87279
+    },
+    {
+      "epoch": 8.974090067859347,
+      "grad_norm": 0.06513124704360962,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 87282
+    },
+    {
+      "epoch": 8.97439851943245,
+      "grad_norm": 0.051685601472854614,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 87285
+    },
+    {
+      "epoch": 8.974706971005553,
+      "grad_norm": 0.09058764576911926,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 87288
+    },
+    {
+      "epoch": 8.975015422578656,
+      "grad_norm": 0.0636780709028244,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 87291
+    },
+    {
+      "epoch": 8.975323874151758,
+      "grad_norm": 0.1368209272623062,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 87294
+    },
+    {
+      "epoch": 8.975632325724861,
+      "grad_norm": 0.11455652862787247,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 87297
+    },
+    {
+      "epoch": 8.975940777297964,
+      "grad_norm": 0.10468421131372452,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 87300
+    },
+    {
+      "epoch": 8.976249228871067,
+      "grad_norm": 0.1308230608701706,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 87303
+    },
+    {
+      "epoch": 8.97655768044417,
+      "grad_norm": 0.07051686197519302,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 87306
+    },
+    {
+      "epoch": 8.976866132017273,
+      "grad_norm": 0.0951620489358902,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 87309
+    },
+    {
+      "epoch": 8.977174583590376,
+      "grad_norm": 0.0550762303173542,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 87312
+    },
+    {
+      "epoch": 8.977483035163479,
+      "grad_norm": 0.04169492423534393,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 87315
+    },
+    {
+      "epoch": 8.977791486736582,
+      "grad_norm": 0.035810817033052444,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 87318
+    },
+    {
+      "epoch": 8.978099938309686,
+      "grad_norm": 0.040050823241472244,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 87321
+    },
+    {
+      "epoch": 8.978408389882789,
+      "grad_norm": 0.061197470873594284,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 87324
+    },
+    {
+      "epoch": 8.978716841455892,
+      "grad_norm": 0.06339312344789505,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 87327
+    },
+    {
+      "epoch": 8.979025293028995,
+      "grad_norm": 0.08060112595558167,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 87330
+    },
+    {
+      "epoch": 8.979333744602098,
+      "grad_norm": 0.09748803824186325,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 87333
+    },
+    {
+      "epoch": 8.9796421961752,
+      "grad_norm": 0.03794391080737114,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 87336
+    },
+    {
+      "epoch": 8.979950647748304,
+      "grad_norm": 0.07690727710723877,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 87339
+    },
+    {
+      "epoch": 8.980259099321406,
+      "grad_norm": 0.047774530947208405,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 87342
+    },
+    {
+      "epoch": 8.98056755089451,
+      "grad_norm": 0.05362236499786377,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 87345
+    },
+    {
+      "epoch": 8.980876002467612,
+      "grad_norm": 0.06028943508863449,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 87348
+    },
+    {
+      "epoch": 8.981184454040715,
+      "grad_norm": 0.04904436320066452,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 87351
+    },
+    {
+      "epoch": 8.981492905613818,
+      "grad_norm": 0.04311923310160637,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 87354
+    },
+    {
+      "epoch": 8.98180135718692,
+      "grad_norm": 0.11199583858251572,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 87357
+    },
+    {
+      "epoch": 8.982109808760026,
+      "grad_norm": 0.05089421197772026,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 87360
+    },
+    {
+      "epoch": 8.982418260333128,
+      "grad_norm": 0.12338538467884064,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 87363
+    },
+    {
+      "epoch": 8.982726711906231,
+      "grad_norm": 0.06985056400299072,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 87366
+    },
+    {
+      "epoch": 8.983035163479334,
+      "grad_norm": 0.08937839418649673,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 87369
+    },
+    {
+      "epoch": 8.983343615052437,
+      "grad_norm": 0.047080084681510925,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 87372
+    },
+    {
+      "epoch": 8.98365206662554,
+      "grad_norm": 0.03225898742675781,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 87375
+    },
+    {
+      "epoch": 8.983960518198643,
+      "grad_norm": 0.030636625364422798,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 87378
+    },
+    {
+      "epoch": 8.984268969771746,
+      "grad_norm": 0.08280804008245468,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 87381
+    },
+    {
+      "epoch": 8.984577421344849,
+      "grad_norm": 0.09910699725151062,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 87384
+    },
+    {
+      "epoch": 8.984885872917951,
+      "grad_norm": 0.07833388447761536,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 87387
+    },
+    {
+      "epoch": 8.985194324491054,
+      "grad_norm": 0.03863595798611641,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 87390
+    },
+    {
+      "epoch": 8.985502776064157,
+      "grad_norm": 0.039544396102428436,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 87393
+    },
+    {
+      "epoch": 8.98581122763726,
+      "grad_norm": 0.057196080684661865,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 87396
+    },
+    {
+      "epoch": 8.986119679210365,
+      "grad_norm": 0.06040463596582413,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 87399
+    },
+    {
+      "epoch": 8.986428130783468,
+      "grad_norm": 0.09306348115205765,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 87402
+    },
+    {
+      "epoch": 8.98673658235657,
+      "grad_norm": 0.08759880810976028,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 87405
+    },
+    {
+      "epoch": 8.987045033929673,
+      "grad_norm": 0.06999802589416504,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 87408
+    },
+    {
+      "epoch": 8.987353485502776,
+      "grad_norm": 0.1001400351524353,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 87411
+    },
+    {
+      "epoch": 8.98766193707588,
+      "grad_norm": 0.11107636988162994,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 87414
+    },
+    {
+      "epoch": 8.987970388648982,
+      "grad_norm": 0.09698151051998138,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 87417
+    },
+    {
+      "epoch": 8.988278840222085,
+      "grad_norm": 0.096869096159935,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 87420
+    },
+    {
+      "epoch": 8.988587291795188,
+      "grad_norm": 0.08064346015453339,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 87423
+    },
+    {
+      "epoch": 8.98889574336829,
+      "grad_norm": 0.08183082938194275,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 87426
+    },
+    {
+      "epoch": 8.989204194941394,
+      "grad_norm": 0.0429241918027401,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 87429
+    },
+    {
+      "epoch": 8.989512646514497,
+      "grad_norm": 0.0404486320912838,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 87432
+    },
+    {
+      "epoch": 8.9898210980876,
+      "grad_norm": 0.049001146107912064,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 87435
+    },
+    {
+      "epoch": 8.990129549660704,
+      "grad_norm": 0.05515841022133827,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 87438
+    },
+    {
+      "epoch": 8.990438001233807,
+      "grad_norm": 0.08830389380455017,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 87441
+    },
+    {
+      "epoch": 8.99074645280691,
+      "grad_norm": 0.07468203455209732,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 87444
+    },
+    {
+      "epoch": 8.991054904380013,
+      "grad_norm": 0.0886622965335846,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 87447
+    },
+    {
+      "epoch": 8.991363355953116,
+      "grad_norm": 0.04467757046222687,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 87450
+    },
+    {
+      "epoch": 8.991671807526219,
+      "grad_norm": 0.05850984901189804,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 87453
+    },
+    {
+      "epoch": 8.991980259099321,
+      "grad_norm": 0.07062778621912003,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 87456
+    },
+    {
+      "epoch": 8.992288710672424,
+      "grad_norm": 0.09450197964906693,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 87459
+    },
+    {
+      "epoch": 8.992597162245527,
+      "grad_norm": 0.08685463666915894,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 87462
+    },
+    {
+      "epoch": 8.99290561381863,
+      "grad_norm": 0.06854233890771866,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 87465
+    },
+    {
+      "epoch": 8.993214065391733,
+      "grad_norm": 0.13164113461971283,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 87468
+    },
+    {
+      "epoch": 8.993522516964836,
+      "grad_norm": 0.09722437709569931,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 87471
+    },
+    {
+      "epoch": 8.993830968537939,
+      "grad_norm": 0.06404321640729904,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 87474
+    },
+    {
+      "epoch": 8.994139420111043,
+      "grad_norm": 0.04171566292643547,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 87477
+    },
+    {
+      "epoch": 8.994447871684146,
+      "grad_norm": 0.04546424746513367,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 87480
+    },
+    {
+      "epoch": 8.99475632325725,
+      "grad_norm": 0.051627617329359055,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 87483
+    },
+    {
+      "epoch": 8.995064774830352,
+      "grad_norm": 0.11216254532337189,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 87486
+    },
+    {
+      "epoch": 8.995373226403455,
+      "grad_norm": 0.059223853051662445,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 87489
+    },
+    {
+      "epoch": 8.995681677976558,
+      "grad_norm": 0.03879429027438164,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 87492
+    },
+    {
+      "epoch": 8.99599012954966,
+      "grad_norm": 0.03178856894373894,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 87495
+    },
+    {
+      "epoch": 8.996298581122764,
+      "grad_norm": 0.06816231459379196,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 87498
+    },
+    {
+      "epoch": 8.996607032695866,
+      "grad_norm": 0.07472996413707733,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 87501
+    },
+    {
+      "epoch": 8.99691548426897,
+      "grad_norm": 0.053777918219566345,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 87504
+    },
+    {
+      "epoch": 8.997223935842072,
+      "grad_norm": 0.11232874542474747,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 87507
+    },
+    {
+      "epoch": 8.997532387415175,
+      "grad_norm": 0.08200154453516006,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 87510
+    },
+    {
+      "epoch": 8.997840838988278,
+      "grad_norm": 0.1282903105020523,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 87513
+    },
+    {
+      "epoch": 8.998149290561383,
+      "grad_norm": 0.04237685352563858,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 87516
+    },
+    {
+      "epoch": 8.998457742134486,
+      "grad_norm": 0.03894351050257683,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 87519
+    },
+    {
+      "epoch": 8.998766193707588,
+      "grad_norm": 0.04536869749426842,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 87522
+    },
+    {
+      "epoch": 8.999074645280691,
+      "grad_norm": 0.08715896308422089,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 87525
+    },
+    {
+      "epoch": 8.999383096853794,
+      "grad_norm": 0.05087468773126602,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 87528
+    },
+    {
+      "epoch": 8.999691548426897,
+      "grad_norm": 0.04533283784985542,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 87531
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.06103600934147835,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 87534
+    },
+    {
+      "epoch": 9.037476770596738,
+      "grad_norm": 0.0529954619705677,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 87537
+    },
+    {
+      "epoch": 9.03778649597357,
+      "grad_norm": 0.11809484660625458,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 87540
+    },
+    {
+      "epoch": 9.038096221350402,
+      "grad_norm": 0.04205942526459694,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 87543
+    },
+    {
+      "epoch": 9.038405946727234,
+      "grad_norm": 0.12428398430347443,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 87546
+    },
+    {
+      "epoch": 9.038715672104068,
+      "grad_norm": 0.07727963477373123,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 87549
+    },
+    {
+      "epoch": 9.0390253974809,
+      "grad_norm": 0.033585965633392334,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 87552
+    },
+    {
+      "epoch": 9.039335122857732,
+      "grad_norm": 0.1038275957107544,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 87555
+    },
+    {
+      "epoch": 9.039644848234566,
+      "grad_norm": 0.044634561985731125,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 87558
+    },
+    {
+      "epoch": 9.039954573611398,
+      "grad_norm": 0.035352591425180435,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 87561
+    },
+    {
+      "epoch": 9.04026429898823,
+      "grad_norm": 0.034989580512046814,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 87564
+    },
+    {
+      "epoch": 9.040574024365062,
+      "grad_norm": 0.04315212368965149,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 87567
+    },
+    {
+      "epoch": 9.040883749741896,
+      "grad_norm": 0.0486474372446537,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 87570
+    },
+    {
+      "epoch": 9.041193475118728,
+      "grad_norm": 0.07493503391742706,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 87573
+    },
+    {
+      "epoch": 9.04150320049556,
+      "grad_norm": 0.060698721557855606,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 87576
+    },
+    {
+      "epoch": 9.041812925872392,
+      "grad_norm": 0.05844523757696152,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 87579
+    },
+    {
+      "epoch": 9.042122651249226,
+      "grad_norm": 0.049893446266651154,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 87582
+    },
+    {
+      "epoch": 9.042432376626058,
+      "grad_norm": 0.10238265991210938,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 87585
+    },
+    {
+      "epoch": 9.04274210200289,
+      "grad_norm": 0.04680422320961952,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 87588
+    },
+    {
+      "epoch": 9.043051827379724,
+      "grad_norm": 0.07950558513402939,
+      "learning_rate": 0.01,
+      "loss": 1.9744,
+      "step": 87591
+    },
+    {
+      "epoch": 9.043361552756556,
+      "grad_norm": 0.07196646928787231,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 87594
+    },
+    {
+      "epoch": 9.043671278133388,
+      "grad_norm": 0.06920761615037918,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 87597
+    },
+    {
+      "epoch": 9.04398100351022,
+      "grad_norm": 0.05272555723786354,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 87600
+    },
+    {
+      "epoch": 9.044290728887054,
+      "grad_norm": 0.057334136217832565,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 87603
+    },
+    {
+      "epoch": 9.044600454263886,
+      "grad_norm": 0.037144239991903305,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 87606
+    },
+    {
+      "epoch": 9.044910179640718,
+      "grad_norm": 0.06027961149811745,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 87609
+    },
+    {
+      "epoch": 9.045219905017552,
+      "grad_norm": 0.11373438686132431,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 87612
+    },
+    {
+      "epoch": 9.045529630394384,
+      "grad_norm": 0.11062182486057281,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 87615
+    },
+    {
+      "epoch": 9.045839355771216,
+      "grad_norm": 0.14294542372226715,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 87618
+    },
+    {
+      "epoch": 9.046149081148048,
+      "grad_norm": 0.10985869914293289,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 87621
+    },
+    {
+      "epoch": 9.046458806524882,
+      "grad_norm": 0.0926436111330986,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 87624
+    },
+    {
+      "epoch": 9.046768531901714,
+      "grad_norm": 0.07326491922140121,
+      "learning_rate": 0.01,
+      "loss": 1.9863,
+      "step": 87627
+    },
+    {
+      "epoch": 9.047078257278546,
+      "grad_norm": 0.07650945335626602,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 87630
+    },
+    {
+      "epoch": 9.047387982655378,
+      "grad_norm": 0.06261960417032242,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 87633
+    },
+    {
+      "epoch": 9.047697708032212,
+      "grad_norm": 0.04512837529182434,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 87636
+    },
+    {
+      "epoch": 9.048007433409044,
+      "grad_norm": 0.06115034222602844,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 87639
+    },
+    {
+      "epoch": 9.048317158785876,
+      "grad_norm": 0.05525890737771988,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 87642
+    },
+    {
+      "epoch": 9.04862688416271,
+      "grad_norm": 0.047319334000349045,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 87645
+    },
+    {
+      "epoch": 9.048936609539542,
+      "grad_norm": 0.12764939665794373,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 87648
+    },
+    {
+      "epoch": 9.049246334916374,
+      "grad_norm": 0.04160080850124359,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 87651
+    },
+    {
+      "epoch": 9.049556060293206,
+      "grad_norm": 0.09743288904428482,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 87654
+    },
+    {
+      "epoch": 9.04986578567004,
+      "grad_norm": 0.05857638269662857,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 87657
+    },
+    {
+      "epoch": 9.050175511046872,
+      "grad_norm": 0.08639366179704666,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 87660
+    },
+    {
+      "epoch": 9.050485236423704,
+      "grad_norm": 0.046659234911203384,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 87663
+    },
+    {
+      "epoch": 9.050794961800538,
+      "grad_norm": 0.03477618098258972,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 87666
+    },
+    {
+      "epoch": 9.05110468717737,
+      "grad_norm": 0.0495758019387722,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 87669
+    },
+    {
+      "epoch": 9.051414412554202,
+      "grad_norm": 0.08402826637029648,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 87672
+    },
+    {
+      "epoch": 9.051724137931034,
+      "grad_norm": 0.06008414924144745,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 87675
+    },
+    {
+      "epoch": 9.052033863307868,
+      "grad_norm": 0.05847488343715668,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 87678
+    },
+    {
+      "epoch": 9.0523435886847,
+      "grad_norm": 0.08440254628658295,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 87681
+    },
+    {
+      "epoch": 9.052653314061532,
+      "grad_norm": 0.11293185502290726,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 87684
+    },
+    {
+      "epoch": 9.052963039438364,
+      "grad_norm": 0.0945296362042427,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 87687
+    },
+    {
+      "epoch": 9.053272764815198,
+      "grad_norm": 0.0900648981332779,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 87690
+    },
+    {
+      "epoch": 9.05358249019203,
+      "grad_norm": 0.06403549760580063,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 87693
+    },
+    {
+      "epoch": 9.053892215568862,
+      "grad_norm": 0.07860220223665237,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 87696
+    },
+    {
+      "epoch": 9.054201940945696,
+      "grad_norm": 0.0828719511628151,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 87699
+    },
+    {
+      "epoch": 9.054511666322528,
+      "grad_norm": 0.09248120337724686,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 87702
+    },
+    {
+      "epoch": 9.05482139169936,
+      "grad_norm": 0.04926043376326561,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 87705
+    },
+    {
+      "epoch": 9.055131117076192,
+      "grad_norm": 0.07462218403816223,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 87708
+    },
+    {
+      "epoch": 9.055440842453025,
+      "grad_norm": 0.09336571395397186,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 87711
+    },
+    {
+      "epoch": 9.055750567829858,
+      "grad_norm": 0.08080223947763443,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 87714
+    },
+    {
+      "epoch": 9.05606029320669,
+      "grad_norm": 0.051116976886987686,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 87717
+    },
+    {
+      "epoch": 9.056370018583523,
+      "grad_norm": 0.07592598348855972,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 87720
+    },
+    {
+      "epoch": 9.056679743960355,
+      "grad_norm": 0.08347304910421371,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 87723
+    },
+    {
+      "epoch": 9.056989469337188,
+      "grad_norm": 0.12313603609800339,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 87726
+    },
+    {
+      "epoch": 9.05729919471402,
+      "grad_norm": 0.13923709094524384,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 87729
+    },
+    {
+      "epoch": 9.057608920090853,
+      "grad_norm": 0.09582033008337021,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 87732
+    },
+    {
+      "epoch": 9.057918645467685,
+      "grad_norm": 0.07468133419752121,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 87735
+    },
+    {
+      "epoch": 9.058228370844517,
+      "grad_norm": 0.05262709781527519,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 87738
+    },
+    {
+      "epoch": 9.05853809622135,
+      "grad_norm": 0.05200639367103577,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 87741
+    },
+    {
+      "epoch": 9.058847821598183,
+      "grad_norm": 0.08909532427787781,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 87744
+    },
+    {
+      "epoch": 9.059157546975015,
+      "grad_norm": 0.08905582875013351,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 87747
+    },
+    {
+      "epoch": 9.059467272351847,
+      "grad_norm": 0.25634267926216125,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 87750
+    },
+    {
+      "epoch": 9.059776997728681,
+      "grad_norm": 0.11460110545158386,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 87753
+    },
+    {
+      "epoch": 9.060086723105513,
+      "grad_norm": 0.04081350937485695,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 87756
+    },
+    {
+      "epoch": 9.060396448482345,
+      "grad_norm": 0.06634104251861572,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 87759
+    },
+    {
+      "epoch": 9.060706173859177,
+      "grad_norm": 0.03965061530470848,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 87762
+    },
+    {
+      "epoch": 9.061015899236011,
+      "grad_norm": 0.045786574482917786,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 87765
+    },
+    {
+      "epoch": 9.061325624612843,
+      "grad_norm": 0.0300121009349823,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 87768
+    },
+    {
+      "epoch": 9.061635349989675,
+      "grad_norm": 0.04564841836690903,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 87771
+    },
+    {
+      "epoch": 9.06194507536651,
+      "grad_norm": 0.045970767736434937,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 87774
+    },
+    {
+      "epoch": 9.062254800743341,
+      "grad_norm": 0.04083496704697609,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 87777
+    },
+    {
+      "epoch": 9.062564526120173,
+      "grad_norm": 0.053099412471055984,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 87780
+    },
+    {
+      "epoch": 9.062874251497005,
+      "grad_norm": 0.06963446736335754,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 87783
+    },
+    {
+      "epoch": 9.06318397687384,
+      "grad_norm": 0.07380146533250809,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 87786
+    },
+    {
+      "epoch": 9.063493702250671,
+      "grad_norm": 0.05961961671710014,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 87789
+    },
+    {
+      "epoch": 9.063803427627503,
+      "grad_norm": 0.05864206328988075,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 87792
+    },
+    {
+      "epoch": 9.064113153004335,
+      "grad_norm": 0.1107788160443306,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 87795
+    },
+    {
+      "epoch": 9.06442287838117,
+      "grad_norm": 0.0659669041633606,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 87798
+    },
+    {
+      "epoch": 9.064732603758001,
+      "grad_norm": 0.10495319217443466,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 87801
+    },
+    {
+      "epoch": 9.065042329134833,
+      "grad_norm": 0.04215942695736885,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 87804
+    },
+    {
+      "epoch": 9.065352054511667,
+      "grad_norm": 0.057210393249988556,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 87807
+    },
+    {
+      "epoch": 9.065661779888499,
+      "grad_norm": 0.05553483963012695,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 87810
+    },
+    {
+      "epoch": 9.065971505265331,
+      "grad_norm": 0.05527332052588463,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 87813
+    },
+    {
+      "epoch": 9.066281230642163,
+      "grad_norm": 0.03902657330036163,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 87816
+    },
+    {
+      "epoch": 9.066590956018997,
+      "grad_norm": 0.0412173829972744,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 87819
+    },
+    {
+      "epoch": 9.066900681395829,
+      "grad_norm": 0.04407156631350517,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 87822
+    },
+    {
+      "epoch": 9.067210406772661,
+      "grad_norm": 0.10649632662534714,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 87825
+    },
+    {
+      "epoch": 9.067520132149495,
+      "grad_norm": 0.07599849998950958,
+      "learning_rate": 0.01,
+      "loss": 1.9732,
+      "step": 87828
+    },
+    {
+      "epoch": 9.067829857526327,
+      "grad_norm": 0.13357628881931305,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 87831
+    },
+    {
+      "epoch": 9.068139582903159,
+      "grad_norm": 0.047365784645080566,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 87834
+    },
+    {
+      "epoch": 9.068449308279991,
+      "grad_norm": 0.036827005445957184,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 87837
+    },
+    {
+      "epoch": 9.068759033656825,
+      "grad_norm": 0.0627468079328537,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 87840
+    },
+    {
+      "epoch": 9.069068759033657,
+      "grad_norm": 0.04302553832530975,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 87843
+    },
+    {
+      "epoch": 9.069378484410489,
+      "grad_norm": 0.059402476996183395,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 87846
+    },
+    {
+      "epoch": 9.069688209787321,
+      "grad_norm": 0.1257617026567459,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 87849
+    },
+    {
+      "epoch": 9.069997935164155,
+      "grad_norm": 0.11703430861234665,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 87852
+    },
+    {
+      "epoch": 9.070307660540987,
+      "grad_norm": 0.05478603020310402,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 87855
+    },
+    {
+      "epoch": 9.070617385917819,
+      "grad_norm": 0.08334581553936005,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 87858
+    },
+    {
+      "epoch": 9.070927111294653,
+      "grad_norm": 0.09045934677124023,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 87861
+    },
+    {
+      "epoch": 9.071236836671485,
+      "grad_norm": 0.0530586764216423,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 87864
+    },
+    {
+      "epoch": 9.071546562048317,
+      "grad_norm": 0.051753733307123184,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 87867
+    },
+    {
+      "epoch": 9.071856287425149,
+      "grad_norm": 0.03743633255362511,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 87870
+    },
+    {
+      "epoch": 9.072166012801983,
+      "grad_norm": 0.054411083459854126,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 87873
+    },
+    {
+      "epoch": 9.072475738178815,
+      "grad_norm": 0.11333665996789932,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 87876
+    },
+    {
+      "epoch": 9.072785463555647,
+      "grad_norm": 0.06036665290594101,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 87879
+    },
+    {
+      "epoch": 9.07309518893248,
+      "grad_norm": 0.05209678038954735,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 87882
+    },
+    {
+      "epoch": 9.073404914309313,
+      "grad_norm": 0.0966220274567604,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 87885
+    },
+    {
+      "epoch": 9.073714639686145,
+      "grad_norm": 0.17173926532268524,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 87888
+    },
+    {
+      "epoch": 9.074024365062977,
+      "grad_norm": 0.07872462272644043,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 87891
+    },
+    {
+      "epoch": 9.07433409043981,
+      "grad_norm": 0.07835288345813751,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 87894
+    },
+    {
+      "epoch": 9.074643815816643,
+      "grad_norm": 0.05287022143602371,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 87897
+    },
+    {
+      "epoch": 9.074953541193475,
+      "grad_norm": 0.054281800985336304,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 87900
+    },
+    {
+      "epoch": 9.075263266570307,
+      "grad_norm": 0.07372366636991501,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 87903
+    },
+    {
+      "epoch": 9.07557299194714,
+      "grad_norm": 0.06382538378238678,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 87906
+    },
+    {
+      "epoch": 9.075882717323973,
+      "grad_norm": 0.12408961355686188,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 87909
+    },
+    {
+      "epoch": 9.076192442700805,
+      "grad_norm": 0.07578717172145844,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 87912
+    },
+    {
+      "epoch": 9.076502168077639,
+      "grad_norm": 0.04379868134856224,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 87915
+    },
+    {
+      "epoch": 9.07681189345447,
+      "grad_norm": 0.053197700530290604,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 87918
+    },
+    {
+      "epoch": 9.077121618831303,
+      "grad_norm": 0.038773033767938614,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 87921
+    },
+    {
+      "epoch": 9.077431344208135,
+      "grad_norm": 0.08359051495790482,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 87924
+    },
+    {
+      "epoch": 9.077741069584969,
+      "grad_norm": 0.039631109684705734,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 87927
+    },
+    {
+      "epoch": 9.0780507949618,
+      "grad_norm": 0.0650608092546463,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 87930
+    },
+    {
+      "epoch": 9.078360520338633,
+      "grad_norm": 0.07246024161577225,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 87933
+    },
+    {
+      "epoch": 9.078670245715466,
+      "grad_norm": 0.15074563026428223,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 87936
+    },
+    {
+      "epoch": 9.078979971092298,
+      "grad_norm": 0.052601393312215805,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 87939
+    },
+    {
+      "epoch": 9.07928969646913,
+      "grad_norm": 0.0388791598379612,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 87942
+    },
+    {
+      "epoch": 9.079599421845963,
+      "grad_norm": 0.07762248814105988,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 87945
+    },
+    {
+      "epoch": 9.079909147222796,
+      "grad_norm": 0.09564360976219177,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 87948
+    },
+    {
+      "epoch": 9.080218872599628,
+      "grad_norm": 0.06043870002031326,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 87951
+    },
+    {
+      "epoch": 9.08052859797646,
+      "grad_norm": 0.04472746327519417,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 87954
+    },
+    {
+      "epoch": 9.080838323353293,
+      "grad_norm": 0.10817110538482666,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 87957
+    },
+    {
+      "epoch": 9.081148048730126,
+      "grad_norm": 0.06465522199869156,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 87960
+    },
+    {
+      "epoch": 9.081457774106958,
+      "grad_norm": 0.0649423748254776,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 87963
+    },
+    {
+      "epoch": 9.08176749948379,
+      "grad_norm": 0.05806288123130798,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 87966
+    },
+    {
+      "epoch": 9.082077224860624,
+      "grad_norm": 0.04380207508802414,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 87969
+    },
+    {
+      "epoch": 9.082386950237456,
+      "grad_norm": 0.03688583895564079,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 87972
+    },
+    {
+      "epoch": 9.082696675614288,
+      "grad_norm": 0.035529594868421555,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 87975
+    },
+    {
+      "epoch": 9.08300640099112,
+      "grad_norm": 0.03344225138425827,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 87978
+    },
+    {
+      "epoch": 9.083316126367954,
+      "grad_norm": 0.03736155480146408,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 87981
+    },
+    {
+      "epoch": 9.083625851744786,
+      "grad_norm": 0.07427693903446198,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 87984
+    },
+    {
+      "epoch": 9.083935577121618,
+      "grad_norm": 0.048297081142663956,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 87987
+    },
+    {
+      "epoch": 9.084245302498452,
+      "grad_norm": 0.11602925509214401,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 87990
+    },
+    {
+      "epoch": 9.084555027875284,
+      "grad_norm": 0.10392475128173828,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 87993
+    },
+    {
+      "epoch": 9.084864753252116,
+      "grad_norm": 0.16618256270885468,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 87996
+    },
+    {
+      "epoch": 9.085174478628948,
+      "grad_norm": 0.10360529273748398,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 87999
+    },
+    {
+      "epoch": 9.085484204005782,
+      "grad_norm": 0.09314300864934921,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 88002
+    },
+    {
+      "epoch": 9.085793929382614,
+      "grad_norm": 0.07424509525299072,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 88005
+    },
+    {
+      "epoch": 9.086103654759446,
+      "grad_norm": 0.04636434093117714,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 88008
+    },
+    {
+      "epoch": 9.086413380136278,
+      "grad_norm": 0.038621935993433,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 88011
+    },
+    {
+      "epoch": 9.086723105513112,
+      "grad_norm": 0.04796215519309044,
+      "learning_rate": 0.01,
+      "loss": 1.9909,
+      "step": 88014
+    },
+    {
+      "epoch": 9.087032830889944,
+      "grad_norm": 0.05523217096924782,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 88017
+    },
+    {
+      "epoch": 9.087342556266776,
+      "grad_norm": 0.06570705026388168,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 88020
+    },
+    {
+      "epoch": 9.08765228164361,
+      "grad_norm": 0.0739300474524498,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 88023
+    },
+    {
+      "epoch": 9.087962007020442,
+      "grad_norm": 0.09385763108730316,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 88026
+    },
+    {
+      "epoch": 9.088271732397274,
+      "grad_norm": 0.1467464715242386,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 88029
+    },
+    {
+      "epoch": 9.088581457774106,
+      "grad_norm": 0.10290052741765976,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 88032
+    },
+    {
+      "epoch": 9.08889118315094,
+      "grad_norm": 0.07046860456466675,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 88035
+    },
+    {
+      "epoch": 9.089200908527772,
+      "grad_norm": 0.04915101081132889,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 88038
+    },
+    {
+      "epoch": 9.089510633904604,
+      "grad_norm": 0.052908964455127716,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 88041
+    },
+    {
+      "epoch": 9.089820359281438,
+      "grad_norm": 0.039025407284498215,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 88044
+    },
+    {
+      "epoch": 9.09013008465827,
+      "grad_norm": 0.03908696398139,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 88047
+    },
+    {
+      "epoch": 9.090439810035102,
+      "grad_norm": 0.059479523450136185,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 88050
+    },
+    {
+      "epoch": 9.090749535411934,
+      "grad_norm": 0.08643188327550888,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 88053
+    },
+    {
+      "epoch": 9.091059260788768,
+      "grad_norm": 0.07994077354669571,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 88056
+    },
+    {
+      "epoch": 9.0913689861656,
+      "grad_norm": 0.08260933309793472,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 88059
+    },
+    {
+      "epoch": 9.091678711542432,
+      "grad_norm": 0.08402281254529953,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 88062
+    },
+    {
+      "epoch": 9.091988436919264,
+      "grad_norm": 0.12822997570037842,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 88065
+    },
+    {
+      "epoch": 9.092298162296098,
+      "grad_norm": 0.10046152770519257,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 88068
+    },
+    {
+      "epoch": 9.09260788767293,
+      "grad_norm": 0.07425348460674286,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 88071
+    },
+    {
+      "epoch": 9.092917613049762,
+      "grad_norm": 0.05061847344040871,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 88074
+    },
+    {
+      "epoch": 9.093227338426596,
+      "grad_norm": 0.04643300175666809,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 88077
+    },
+    {
+      "epoch": 9.093537063803428,
+      "grad_norm": 0.040220703929662704,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 88080
+    },
+    {
+      "epoch": 9.09384678918026,
+      "grad_norm": 0.04779227450489998,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 88083
+    },
+    {
+      "epoch": 9.094156514557092,
+      "grad_norm": 0.15425650775432587,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 88086
+    },
+    {
+      "epoch": 9.094466239933926,
+      "grad_norm": 0.05012909695506096,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 88089
+    },
+    {
+      "epoch": 9.094775965310758,
+      "grad_norm": 0.04676985740661621,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 88092
+    },
+    {
+      "epoch": 9.09508569068759,
+      "grad_norm": 0.0426156222820282,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 88095
+    },
+    {
+      "epoch": 9.095395416064424,
+      "grad_norm": 0.04056524485349655,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 88098
+    },
+    {
+      "epoch": 9.095705141441256,
+      "grad_norm": 0.13012756407260895,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 88101
+    },
+    {
+      "epoch": 9.096014866818088,
+      "grad_norm": 0.17354853451251984,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 88104
+    },
+    {
+      "epoch": 9.09632459219492,
+      "grad_norm": 0.12001447379589081,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 88107
+    },
+    {
+      "epoch": 9.096634317571754,
+      "grad_norm": 0.07169081270694733,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 88110
+    },
+    {
+      "epoch": 9.096944042948586,
+      "grad_norm": 0.05362652987241745,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 88113
+    },
+    {
+      "epoch": 9.097253768325418,
+      "grad_norm": 0.048951901495456696,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 88116
+    },
+    {
+      "epoch": 9.09756349370225,
+      "grad_norm": 0.06211349740624428,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 88119
+    },
+    {
+      "epoch": 9.097873219079084,
+      "grad_norm": 0.04690651223063469,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 88122
+    },
+    {
+      "epoch": 9.098182944455916,
+      "grad_norm": 0.05850514769554138,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 88125
+    },
+    {
+      "epoch": 9.098492669832748,
+      "grad_norm": 0.05712956562638283,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 88128
+    },
+    {
+      "epoch": 9.098802395209582,
+      "grad_norm": 0.07003414630889893,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 88131
+    },
+    {
+      "epoch": 9.099112120586414,
+      "grad_norm": 0.06625586748123169,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 88134
+    },
+    {
+      "epoch": 9.099421845963246,
+      "grad_norm": 0.05052918195724487,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 88137
+    },
+    {
+      "epoch": 9.099731571340078,
+      "grad_norm": 0.0572190098464489,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 88140
+    },
+    {
+      "epoch": 9.100041296716912,
+      "grad_norm": 0.07441101968288422,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 88143
+    },
+    {
+      "epoch": 9.100351022093744,
+      "grad_norm": 0.09752295166254044,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 88146
+    },
+    {
+      "epoch": 9.100660747470576,
+      "grad_norm": 0.05231703817844391,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 88149
+    },
+    {
+      "epoch": 9.10097047284741,
+      "grad_norm": 0.09808948636054993,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 88152
+    },
+    {
+      "epoch": 9.101280198224242,
+      "grad_norm": 0.057692382484674454,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 88155
+    },
+    {
+      "epoch": 9.101589923601074,
+      "grad_norm": 0.06832569092512131,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 88158
+    },
+    {
+      "epoch": 9.101899648977906,
+      "grad_norm": 0.05263699218630791,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 88161
+    },
+    {
+      "epoch": 9.10220937435474,
+      "grad_norm": 0.10359843820333481,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 88164
+    },
+    {
+      "epoch": 9.102519099731571,
+      "grad_norm": 0.0683496743440628,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 88167
+    },
+    {
+      "epoch": 9.102828825108404,
+      "grad_norm": 0.04285841062664986,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 88170
+    },
+    {
+      "epoch": 9.103138550485236,
+      "grad_norm": 0.04572022333741188,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 88173
+    },
+    {
+      "epoch": 9.10344827586207,
+      "grad_norm": 0.07577993720769882,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 88176
+    },
+    {
+      "epoch": 9.103758001238901,
+      "grad_norm": 0.060188256204128265,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 88179
+    },
+    {
+      "epoch": 9.104067726615733,
+      "grad_norm": 0.044819220900535583,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 88182
+    },
+    {
+      "epoch": 9.104377451992567,
+      "grad_norm": 0.04718435928225517,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 88185
+    },
+    {
+      "epoch": 9.1046871773694,
+      "grad_norm": 0.04264286905527115,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 88188
+    },
+    {
+      "epoch": 9.104996902746231,
+      "grad_norm": 0.04543020576238632,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 88191
+    },
+    {
+      "epoch": 9.105306628123063,
+      "grad_norm": 0.04506497085094452,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 88194
+    },
+    {
+      "epoch": 9.105616353499897,
+      "grad_norm": 0.04982695356011391,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 88197
+    },
+    {
+      "epoch": 9.10592607887673,
+      "grad_norm": 0.07128612697124481,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 88200
+    },
+    {
+      "epoch": 9.106235804253561,
+      "grad_norm": 0.07622264325618744,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 88203
+    },
+    {
+      "epoch": 9.106545529630395,
+      "grad_norm": 0.04159415885806084,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 88206
+    },
+    {
+      "epoch": 9.106855255007227,
+      "grad_norm": 0.11658751219511032,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 88209
+    },
+    {
+      "epoch": 9.10716498038406,
+      "grad_norm": 0.10910781472921371,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 88212
+    },
+    {
+      "epoch": 9.107474705760891,
+      "grad_norm": 0.05028378963470459,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 88215
+    },
+    {
+      "epoch": 9.107784431137725,
+      "grad_norm": 0.05754128098487854,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 88218
+    },
+    {
+      "epoch": 9.108094156514557,
+      "grad_norm": 0.09947003424167633,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 88221
+    },
+    {
+      "epoch": 9.10840388189139,
+      "grad_norm": 0.046979811042547226,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 88224
+    },
+    {
+      "epoch": 9.108713607268221,
+      "grad_norm": 0.04049091413617134,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 88227
+    },
+    {
+      "epoch": 9.109023332645055,
+      "grad_norm": 0.03384038433432579,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 88230
+    },
+    {
+      "epoch": 9.109333058021887,
+      "grad_norm": 0.06660273671150208,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 88233
+    },
+    {
+      "epoch": 9.10964278339872,
+      "grad_norm": 0.05666836351156235,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 88236
+    },
+    {
+      "epoch": 9.109952508775553,
+      "grad_norm": 0.066832534968853,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 88239
+    },
+    {
+      "epoch": 9.110262234152385,
+      "grad_norm": 0.10214311629533768,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 88242
+    },
+    {
+      "epoch": 9.110571959529217,
+      "grad_norm": 0.06897658109664917,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 88245
+    },
+    {
+      "epoch": 9.11088168490605,
+      "grad_norm": 0.11510921269655228,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 88248
+    },
+    {
+      "epoch": 9.111191410282883,
+      "grad_norm": 0.0748416930437088,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 88251
+    },
+    {
+      "epoch": 9.111501135659715,
+      "grad_norm": 0.03657582029700279,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 88254
+    },
+    {
+      "epoch": 9.111810861036547,
+      "grad_norm": 0.049296047538518906,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 88257
+    },
+    {
+      "epoch": 9.112120586413381,
+      "grad_norm": 0.10844705998897552,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 88260
+    },
+    {
+      "epoch": 9.112430311790213,
+      "grad_norm": 0.0642717257142067,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 88263
+    },
+    {
+      "epoch": 9.112740037167045,
+      "grad_norm": 0.04892707243561745,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 88266
+    },
+    {
+      "epoch": 9.113049762543877,
+      "grad_norm": 0.06426113098859787,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 88269
+    },
+    {
+      "epoch": 9.113359487920711,
+      "grad_norm": 0.03934602066874504,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 88272
+    },
+    {
+      "epoch": 9.113669213297543,
+      "grad_norm": 0.05431235954165459,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 88275
+    },
+    {
+      "epoch": 9.113978938674375,
+      "grad_norm": 0.05585315823554993,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 88278
+    },
+    {
+      "epoch": 9.114288664051207,
+      "grad_norm": 0.03399717062711716,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 88281
+    },
+    {
+      "epoch": 9.114598389428041,
+      "grad_norm": 0.10088639706373215,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 88284
+    },
+    {
+      "epoch": 9.114908114804873,
+      "grad_norm": 0.06731191277503967,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 88287
+    },
+    {
+      "epoch": 9.115217840181705,
+      "grad_norm": 0.039947111159563065,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 88290
+    },
+    {
+      "epoch": 9.115527565558539,
+      "grad_norm": 0.04077018052339554,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 88293
+    },
+    {
+      "epoch": 9.11583729093537,
+      "grad_norm": 0.11213209480047226,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 88296
+    },
+    {
+      "epoch": 9.116147016312203,
+      "grad_norm": 0.0802016407251358,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 88299
+    },
+    {
+      "epoch": 9.116456741689035,
+      "grad_norm": 0.06813075393438339,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 88302
+    },
+    {
+      "epoch": 9.116766467065869,
+      "grad_norm": 0.11700167506933212,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 88305
+    },
+    {
+      "epoch": 9.1170761924427,
+      "grad_norm": 0.07587660849094391,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 88308
+    },
+    {
+      "epoch": 9.117385917819533,
+      "grad_norm": 0.062029797583818436,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 88311
+    },
+    {
+      "epoch": 9.117695643196367,
+      "grad_norm": 0.11622146517038345,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 88314
+    },
+    {
+      "epoch": 9.118005368573199,
+      "grad_norm": 0.04664183780550957,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 88317
+    },
+    {
+      "epoch": 9.11831509395003,
+      "grad_norm": 0.10970574617385864,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 88320
+    },
+    {
+      "epoch": 9.118624819326863,
+      "grad_norm": 0.05169505998492241,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 88323
+    },
+    {
+      "epoch": 9.118934544703697,
+      "grad_norm": 0.03827926144003868,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 88326
+    },
+    {
+      "epoch": 9.119244270080529,
+      "grad_norm": 0.04646414518356323,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 88329
+    },
+    {
+      "epoch": 9.11955399545736,
+      "grad_norm": 0.1304306983947754,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 88332
+    },
+    {
+      "epoch": 9.119863720834193,
+      "grad_norm": 0.11228543519973755,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 88335
+    },
+    {
+      "epoch": 9.120173446211027,
+      "grad_norm": 0.051064543426036835,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 88338
+    },
+    {
+      "epoch": 9.120483171587859,
+      "grad_norm": 0.09616478532552719,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 88341
+    },
+    {
+      "epoch": 9.12079289696469,
+      "grad_norm": 0.11419511586427689,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 88344
+    },
+    {
+      "epoch": 9.121102622341525,
+      "grad_norm": 0.08378113061189651,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 88347
+    },
+    {
+      "epoch": 9.121412347718357,
+      "grad_norm": 0.04137773811817169,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 88350
+    },
+    {
+      "epoch": 9.121722073095189,
+      "grad_norm": 0.055357303470373154,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 88353
+    },
+    {
+      "epoch": 9.12203179847202,
+      "grad_norm": 0.04019000381231308,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 88356
+    },
+    {
+      "epoch": 9.122341523848855,
+      "grad_norm": 0.030802376568317413,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 88359
+    },
+    {
+      "epoch": 9.122651249225687,
+      "grad_norm": 0.08665869385004044,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 88362
+    },
+    {
+      "epoch": 9.122960974602519,
+      "grad_norm": 0.11755997687578201,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 88365
+    },
+    {
+      "epoch": 9.123270699979352,
+      "grad_norm": 0.04438506066799164,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 88368
+    },
+    {
+      "epoch": 9.123580425356185,
+      "grad_norm": 0.05268723517656326,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 88371
+    },
+    {
+      "epoch": 9.123890150733017,
+      "grad_norm": 0.042669329792261124,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 88374
+    },
+    {
+      "epoch": 9.124199876109849,
+      "grad_norm": 0.050564251840114594,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 88377
+    },
+    {
+      "epoch": 9.124509601486682,
+      "grad_norm": 0.06595560163259506,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 88380
+    },
+    {
+      "epoch": 9.124819326863514,
+      "grad_norm": 0.03814883902668953,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 88383
+    },
+    {
+      "epoch": 9.125129052240347,
+      "grad_norm": 0.0743202194571495,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 88386
+    },
+    {
+      "epoch": 9.125438777617179,
+      "grad_norm": 0.16788910329341888,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 88389
+    },
+    {
+      "epoch": 9.125748502994012,
+      "grad_norm": 0.038132194429636,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 88392
+    },
+    {
+      "epoch": 9.126058228370844,
+      "grad_norm": 0.04480666667222977,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 88395
+    },
+    {
+      "epoch": 9.126367953747677,
+      "grad_norm": 0.04021817073225975,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 88398
+    },
+    {
+      "epoch": 9.12667767912451,
+      "grad_norm": 0.036103565245866776,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 88401
+    },
+    {
+      "epoch": 9.126987404501342,
+      "grad_norm": 0.052644163370132446,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 88404
+    },
+    {
+      "epoch": 9.127297129878174,
+      "grad_norm": 0.07091756165027618,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 88407
+    },
+    {
+      "epoch": 9.127606855255006,
+      "grad_norm": 0.0955536738038063,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 88410
+    },
+    {
+      "epoch": 9.12791658063184,
+      "grad_norm": 0.13555645942687988,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 88413
+    },
+    {
+      "epoch": 9.128226306008672,
+      "grad_norm": 0.038274601101875305,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 88416
+    },
+    {
+      "epoch": 9.128536031385504,
+      "grad_norm": 0.04228101298213005,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 88419
+    },
+    {
+      "epoch": 9.128845756762338,
+      "grad_norm": 0.03320973366498947,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 88422
+    },
+    {
+      "epoch": 9.12915548213917,
+      "grad_norm": 0.04140046238899231,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 88425
+    },
+    {
+      "epoch": 9.129465207516002,
+      "grad_norm": 0.04351486638188362,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 88428
+    },
+    {
+      "epoch": 9.129774932892834,
+      "grad_norm": 0.17408426105976105,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 88431
+    },
+    {
+      "epoch": 9.130084658269668,
+      "grad_norm": 0.08010722696781158,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 88434
+    },
+    {
+      "epoch": 9.1303943836465,
+      "grad_norm": 0.06910273432731628,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 88437
+    },
+    {
+      "epoch": 9.130704109023332,
+      "grad_norm": 0.03804326429963112,
+      "learning_rate": 0.01,
+      "loss": 1.9711,
+      "step": 88440
+    },
+    {
+      "epoch": 9.131013834400164,
+      "grad_norm": 0.039677515625953674,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 88443
+    },
+    {
+      "epoch": 9.131323559776998,
+      "grad_norm": 0.047551997005939484,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 88446
+    },
+    {
+      "epoch": 9.13163328515383,
+      "grad_norm": 0.05167323350906372,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 88449
+    },
+    {
+      "epoch": 9.131943010530662,
+      "grad_norm": 0.1845468133687973,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 88452
+    },
+    {
+      "epoch": 9.132252735907496,
+      "grad_norm": 0.1411765217781067,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 88455
+    },
+    {
+      "epoch": 9.132562461284328,
+      "grad_norm": 0.07709158211946487,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 88458
+    },
+    {
+      "epoch": 9.13287218666116,
+      "grad_norm": 0.06372318416833878,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 88461
+    },
+    {
+      "epoch": 9.133181912037992,
+      "grad_norm": 0.040460873395204544,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 88464
+    },
+    {
+      "epoch": 9.133491637414826,
+      "grad_norm": 0.055337656289339066,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 88467
+    },
+    {
+      "epoch": 9.133801362791658,
+      "grad_norm": 0.03703693300485611,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 88470
+    },
+    {
+      "epoch": 9.13411108816849,
+      "grad_norm": 0.05583380162715912,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 88473
+    },
+    {
+      "epoch": 9.134420813545324,
+      "grad_norm": 0.05320487171411514,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 88476
+    },
+    {
+      "epoch": 9.134730538922156,
+      "grad_norm": 0.0624091736972332,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 88479
+    },
+    {
+      "epoch": 9.135040264298988,
+      "grad_norm": 0.14112161099910736,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 88482
+    },
+    {
+      "epoch": 9.13534998967582,
+      "grad_norm": 0.08291780948638916,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 88485
+    },
+    {
+      "epoch": 9.135659715052654,
+      "grad_norm": 0.13060036301612854,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 88488
+    },
+    {
+      "epoch": 9.135969440429486,
+      "grad_norm": 0.08995925635099411,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 88491
+    },
+    {
+      "epoch": 9.136279165806318,
+      "grad_norm": 0.0542483888566494,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 88494
+    },
+    {
+      "epoch": 9.13658889118315,
+      "grad_norm": 0.03333907201886177,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 88497
+    },
+    {
+      "epoch": 9.136898616559984,
+      "grad_norm": 0.04014088585972786,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 88500
+    },
+    {
+      "epoch": 9.137208341936816,
+      "grad_norm": 0.044594038277864456,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 88503
+    },
+    {
+      "epoch": 9.137518067313648,
+      "grad_norm": 0.06447494775056839,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 88506
+    },
+    {
+      "epoch": 9.137827792690482,
+      "grad_norm": 0.050712186843156815,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 88509
+    },
+    {
+      "epoch": 9.138137518067314,
+      "grad_norm": 0.06224539503455162,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 88512
+    },
+    {
+      "epoch": 9.138447243444146,
+      "grad_norm": 0.17767208814620972,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 88515
+    },
+    {
+      "epoch": 9.138756968820978,
+      "grad_norm": 0.06980250775814056,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 88518
+    },
+    {
+      "epoch": 9.139066694197812,
+      "grad_norm": 0.03640732541680336,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 88521
+    },
+    {
+      "epoch": 9.139376419574644,
+      "grad_norm": 0.03801807761192322,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 88524
+    },
+    {
+      "epoch": 9.139686144951476,
+      "grad_norm": 0.03328856825828552,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 88527
+    },
+    {
+      "epoch": 9.13999587032831,
+      "grad_norm": 0.07712339609861374,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 88530
+    },
+    {
+      "epoch": 9.140305595705142,
+      "grad_norm": 0.08324889093637466,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 88533
+    },
+    {
+      "epoch": 9.140615321081974,
+      "grad_norm": 0.0427752248942852,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 88536
+    },
+    {
+      "epoch": 9.140925046458806,
+      "grad_norm": 0.07253149896860123,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 88539
+    },
+    {
+      "epoch": 9.14123477183564,
+      "grad_norm": 0.0548226498067379,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 88542
+    },
+    {
+      "epoch": 9.141544497212472,
+      "grad_norm": 0.059894196689128876,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 88545
+    },
+    {
+      "epoch": 9.141854222589304,
+      "grad_norm": 0.08382166177034378,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 88548
+    },
+    {
+      "epoch": 9.142163947966136,
+      "grad_norm": 0.13770000636577606,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 88551
+    },
+    {
+      "epoch": 9.14247367334297,
+      "grad_norm": 0.14796362817287445,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 88554
+    },
+    {
+      "epoch": 9.142783398719802,
+      "grad_norm": 0.12419676780700684,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 88557
+    },
+    {
+      "epoch": 9.143093124096634,
+      "grad_norm": 0.04314526543021202,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 88560
+    },
+    {
+      "epoch": 9.143402849473468,
+      "grad_norm": 0.056149184703826904,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 88563
+    },
+    {
+      "epoch": 9.1437125748503,
+      "grad_norm": 0.04276338964700699,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 88566
+    },
+    {
+      "epoch": 9.144022300227132,
+      "grad_norm": 0.035918641835451126,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 88569
+    },
+    {
+      "epoch": 9.144332025603964,
+      "grad_norm": 0.037139903753995895,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 88572
+    },
+    {
+      "epoch": 9.144641750980798,
+      "grad_norm": 0.11460118740797043,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 88575
+    },
+    {
+      "epoch": 9.14495147635763,
+      "grad_norm": 0.04918679967522621,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 88578
+    },
+    {
+      "epoch": 9.145261201734462,
+      "grad_norm": 0.03683009743690491,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 88581
+    },
+    {
+      "epoch": 9.145570927111295,
+      "grad_norm": 0.06100183352828026,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 88584
+    },
+    {
+      "epoch": 9.145880652488128,
+      "grad_norm": 0.07831919193267822,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 88587
+    },
+    {
+      "epoch": 9.14619037786496,
+      "grad_norm": 0.04614289477467537,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 88590
+    },
+    {
+      "epoch": 9.146500103241792,
+      "grad_norm": 0.12135063856840134,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 88593
+    },
+    {
+      "epoch": 9.146809828618625,
+      "grad_norm": 0.048958707600831985,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 88596
+    },
+    {
+      "epoch": 9.147119553995458,
+      "grad_norm": 0.04851505160331726,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 88599
+    },
+    {
+      "epoch": 9.14742927937229,
+      "grad_norm": 0.1740291863679886,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 88602
+    },
+    {
+      "epoch": 9.147739004749122,
+      "grad_norm": 0.10942324995994568,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 88605
+    },
+    {
+      "epoch": 9.148048730125955,
+      "grad_norm": 0.04716290906071663,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 88608
+    },
+    {
+      "epoch": 9.148358455502787,
+      "grad_norm": 0.03627236187458038,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 88611
+    },
+    {
+      "epoch": 9.14866818087962,
+      "grad_norm": 0.04512391239404678,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 88614
+    },
+    {
+      "epoch": 9.148977906256453,
+      "grad_norm": 0.055295899510383606,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 88617
+    },
+    {
+      "epoch": 9.149287631633285,
+      "grad_norm": 0.03937353938817978,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 88620
+    },
+    {
+      "epoch": 9.149597357010117,
+      "grad_norm": 0.057175163179636,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 88623
+    },
+    {
+      "epoch": 9.14990708238695,
+      "grad_norm": 0.03652303293347359,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 88626
+    },
+    {
+      "epoch": 9.150216807763783,
+      "grad_norm": 0.05022728443145752,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 88629
+    },
+    {
+      "epoch": 9.150526533140615,
+      "grad_norm": 0.04779180511832237,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 88632
+    },
+    {
+      "epoch": 9.150836258517447,
+      "grad_norm": 0.03462984785437584,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 88635
+    },
+    {
+      "epoch": 9.151145983894281,
+      "grad_norm": 0.15472935140132904,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 88638
+    },
+    {
+      "epoch": 9.151455709271113,
+      "grad_norm": 0.10227237641811371,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 88641
+    },
+    {
+      "epoch": 9.151765434647945,
+      "grad_norm": 0.05621829628944397,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 88644
+    },
+    {
+      "epoch": 9.152075160024777,
+      "grad_norm": 0.048698339611291885,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 88647
+    },
+    {
+      "epoch": 9.152384885401611,
+      "grad_norm": 0.04971965774893761,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 88650
+    },
+    {
+      "epoch": 9.152694610778443,
+      "grad_norm": 0.03834971413016319,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 88653
+    },
+    {
+      "epoch": 9.153004336155275,
+      "grad_norm": 0.03815067186951637,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 88656
+    },
+    {
+      "epoch": 9.153314061532107,
+      "grad_norm": 0.04239659011363983,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 88659
+    },
+    {
+      "epoch": 9.153623786908941,
+      "grad_norm": 0.04531934857368469,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 88662
+    },
+    {
+      "epoch": 9.153933512285773,
+      "grad_norm": 0.06465330719947815,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 88665
+    },
+    {
+      "epoch": 9.154243237662605,
+      "grad_norm": 0.06051948666572571,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 88668
+    },
+    {
+      "epoch": 9.154552963039439,
+      "grad_norm": 0.06463321298360825,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 88671
+    },
+    {
+      "epoch": 9.154862688416271,
+      "grad_norm": 0.06618782877922058,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 88674
+    },
+    {
+      "epoch": 9.155172413793103,
+      "grad_norm": 0.1678209751844406,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 88677
+    },
+    {
+      "epoch": 9.155482139169935,
+      "grad_norm": 0.1151156947016716,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 88680
+    },
+    {
+      "epoch": 9.155791864546769,
+      "grad_norm": 0.04810039699077606,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 88683
+    },
+    {
+      "epoch": 9.156101589923601,
+      "grad_norm": 0.0373799167573452,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 88686
+    },
+    {
+      "epoch": 9.156411315300433,
+      "grad_norm": 0.03355636075139046,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 88689
+    },
+    {
+      "epoch": 9.156721040677267,
+      "grad_norm": 0.03397734463214874,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 88692
+    },
+    {
+      "epoch": 9.157030766054099,
+      "grad_norm": 0.07095854729413986,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 88695
+    },
+    {
+      "epoch": 9.157340491430931,
+      "grad_norm": 0.05653863027691841,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 88698
+    },
+    {
+      "epoch": 9.157650216807763,
+      "grad_norm": 0.10083531588315964,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 88701
+    },
+    {
+      "epoch": 9.157959942184597,
+      "grad_norm": 0.07726927101612091,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 88704
+    },
+    {
+      "epoch": 9.158269667561429,
+      "grad_norm": 0.07324657589197159,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 88707
+    },
+    {
+      "epoch": 9.158579392938261,
+      "grad_norm": 0.07510195672512054,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 88710
+    },
+    {
+      "epoch": 9.158889118315093,
+      "grad_norm": 0.043642688542604446,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 88713
+    },
+    {
+      "epoch": 9.159198843691927,
+      "grad_norm": 0.10319583117961884,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 88716
+    },
+    {
+      "epoch": 9.159508569068759,
+      "grad_norm": 0.039388563483953476,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 88719
+    },
+    {
+      "epoch": 9.159818294445591,
+      "grad_norm": 0.12249758839607239,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 88722
+    },
+    {
+      "epoch": 9.160128019822425,
+      "grad_norm": 0.11489095538854599,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 88725
+    },
+    {
+      "epoch": 9.160437745199257,
+      "grad_norm": 0.04785621538758278,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 88728
+    },
+    {
+      "epoch": 9.160747470576089,
+      "grad_norm": 0.06265294551849365,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 88731
+    },
+    {
+      "epoch": 9.161057195952921,
+      "grad_norm": 0.06334743648767471,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 88734
+    },
+    {
+      "epoch": 9.161366921329755,
+      "grad_norm": 0.08872868865728378,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 88737
+    },
+    {
+      "epoch": 9.161676646706587,
+      "grad_norm": 0.052145641297101974,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 88740
+    },
+    {
+      "epoch": 9.161986372083419,
+      "grad_norm": 0.07859448343515396,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 88743
+    },
+    {
+      "epoch": 9.162296097460253,
+      "grad_norm": 0.0986669659614563,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 88746
+    },
+    {
+      "epoch": 9.162605822837085,
+      "grad_norm": 0.05713833123445511,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 88749
+    },
+    {
+      "epoch": 9.162915548213917,
+      "grad_norm": 0.07410735636949539,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 88752
+    },
+    {
+      "epoch": 9.163225273590749,
+      "grad_norm": 0.07402617484331131,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 88755
+    },
+    {
+      "epoch": 9.163534998967583,
+      "grad_norm": 0.05839969217777252,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 88758
+    },
+    {
+      "epoch": 9.163844724344415,
+      "grad_norm": 0.10944623500108719,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 88761
+    },
+    {
+      "epoch": 9.164154449721247,
+      "grad_norm": 0.13309495151042938,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 88764
+    },
+    {
+      "epoch": 9.164464175098079,
+      "grad_norm": 0.0856231153011322,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 88767
+    },
+    {
+      "epoch": 9.164773900474913,
+      "grad_norm": 0.03998791053891182,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 88770
+    },
+    {
+      "epoch": 9.165083625851745,
+      "grad_norm": 0.12182749807834625,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 88773
+    },
+    {
+      "epoch": 9.165393351228577,
+      "grad_norm": 0.0821884498000145,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 88776
+    },
+    {
+      "epoch": 9.16570307660541,
+      "grad_norm": 0.08010543137788773,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 88779
+    },
+    {
+      "epoch": 9.166012801982243,
+      "grad_norm": 0.04775281995534897,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 88782
+    },
+    {
+      "epoch": 9.166322527359075,
+      "grad_norm": 0.04545445367693901,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 88785
+    },
+    {
+      "epoch": 9.166632252735907,
+      "grad_norm": 0.06078621745109558,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 88788
+    },
+    {
+      "epoch": 9.16694197811274,
+      "grad_norm": 0.062157388776540756,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 88791
+    },
+    {
+      "epoch": 9.167251703489573,
+      "grad_norm": 0.039783451706171036,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 88794
+    },
+    {
+      "epoch": 9.167561428866405,
+      "grad_norm": 0.042033612728118896,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 88797
+    },
+    {
+      "epoch": 9.167871154243239,
+      "grad_norm": 0.08056508004665375,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 88800
+    },
+    {
+      "epoch": 9.16818087962007,
+      "grad_norm": 0.05128775164484978,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 88803
+    },
+    {
+      "epoch": 9.168490604996903,
+      "grad_norm": 0.057393159717321396,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 88806
+    },
+    {
+      "epoch": 9.168800330373735,
+      "grad_norm": 0.03829537332057953,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 88809
+    },
+    {
+      "epoch": 9.169110055750568,
+      "grad_norm": 0.045484233647584915,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 88812
+    },
+    {
+      "epoch": 9.1694197811274,
+      "grad_norm": 0.05106499791145325,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 88815
+    },
+    {
+      "epoch": 9.169729506504233,
+      "grad_norm": 0.060923442244529724,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 88818
+    },
+    {
+      "epoch": 9.170039231881065,
+      "grad_norm": 0.1129612997174263,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 88821
+    },
+    {
+      "epoch": 9.170348957257898,
+      "grad_norm": 0.053068701177835464,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 88824
+    },
+    {
+      "epoch": 9.17065868263473,
+      "grad_norm": 0.12930330634117126,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 88827
+    },
+    {
+      "epoch": 9.170968408011563,
+      "grad_norm": 0.04084591194987297,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 88830
+    },
+    {
+      "epoch": 9.171278133388396,
+      "grad_norm": 0.039442602545022964,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 88833
+    },
+    {
+      "epoch": 9.171587858765228,
+      "grad_norm": 0.04807281494140625,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 88836
+    },
+    {
+      "epoch": 9.17189758414206,
+      "grad_norm": 0.049598827958106995,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 88839
+    },
+    {
+      "epoch": 9.172207309518893,
+      "grad_norm": 0.11387526988983154,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 88842
+    },
+    {
+      "epoch": 9.172517034895726,
+      "grad_norm": 0.08218611776828766,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 88845
+    },
+    {
+      "epoch": 9.172826760272558,
+      "grad_norm": 0.0520508848130703,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 88848
+    },
+    {
+      "epoch": 9.17313648564939,
+      "grad_norm": 0.040659550577402115,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 88851
+    },
+    {
+      "epoch": 9.173446211026224,
+      "grad_norm": 0.07411307096481323,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 88854
+    },
+    {
+      "epoch": 9.173755936403056,
+      "grad_norm": 0.03607996925711632,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 88857
+    },
+    {
+      "epoch": 9.174065661779888,
+      "grad_norm": 0.04786485806107521,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 88860
+    },
+    {
+      "epoch": 9.17437538715672,
+      "grad_norm": 0.0968308299779892,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 88863
+    },
+    {
+      "epoch": 9.174685112533554,
+      "grad_norm": 0.09454333037137985,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 88866
+    },
+    {
+      "epoch": 9.174994837910386,
+      "grad_norm": 0.08687274158000946,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 88869
+    },
+    {
+      "epoch": 9.175304563287218,
+      "grad_norm": 0.03763791546225548,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 88872
+    },
+    {
+      "epoch": 9.17561428866405,
+      "grad_norm": 0.04863421246409416,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 88875
+    },
+    {
+      "epoch": 9.175924014040884,
+      "grad_norm": 0.07117434591054916,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 88878
+    },
+    {
+      "epoch": 9.176233739417716,
+      "grad_norm": 0.05109839886426926,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 88881
+    },
+    {
+      "epoch": 9.176543464794548,
+      "grad_norm": 0.05009722337126732,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 88884
+    },
+    {
+      "epoch": 9.176853190171382,
+      "grad_norm": 0.04079516977071762,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 88887
+    },
+    {
+      "epoch": 9.177162915548214,
+      "grad_norm": 0.04275686293840408,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 88890
+    },
+    {
+      "epoch": 9.177472640925046,
+      "grad_norm": 0.04627576842904091,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 88893
+    },
+    {
+      "epoch": 9.177782366301878,
+      "grad_norm": 0.05262031778693199,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 88896
+    },
+    {
+      "epoch": 9.178092091678712,
+      "grad_norm": 0.10498153418302536,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 88899
+    },
+    {
+      "epoch": 9.178401817055544,
+      "grad_norm": 0.08968192338943481,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 88902
+    },
+    {
+      "epoch": 9.178711542432376,
+      "grad_norm": 0.06271538883447647,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 88905
+    },
+    {
+      "epoch": 9.17902126780921,
+      "grad_norm": 0.07553781569004059,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 88908
+    },
+    {
+      "epoch": 9.179330993186042,
+      "grad_norm": 0.050929512828588486,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 88911
+    },
+    {
+      "epoch": 9.179640718562874,
+      "grad_norm": 0.06154989078640938,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 88914
+    },
+    {
+      "epoch": 9.179950443939706,
+      "grad_norm": 0.0712541937828064,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 88917
+    },
+    {
+      "epoch": 9.18026016931654,
+      "grad_norm": 0.06601454317569733,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 88920
+    },
+    {
+      "epoch": 9.180569894693372,
+      "grad_norm": 0.105747289955616,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 88923
+    },
+    {
+      "epoch": 9.180879620070204,
+      "grad_norm": 0.06339000165462494,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 88926
+    },
+    {
+      "epoch": 9.181189345447036,
+      "grad_norm": 0.0613056980073452,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 88929
+    },
+    {
+      "epoch": 9.18149907082387,
+      "grad_norm": 0.09535111486911774,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 88932
+    },
+    {
+      "epoch": 9.181808796200702,
+      "grad_norm": 0.05504938215017319,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 88935
+    },
+    {
+      "epoch": 9.182118521577534,
+      "grad_norm": 0.04098088666796684,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 88938
+    },
+    {
+      "epoch": 9.182428246954368,
+      "grad_norm": 0.15497301518917084,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 88941
+    },
+    {
+      "epoch": 9.1827379723312,
+      "grad_norm": 0.08422484993934631,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 88944
+    },
+    {
+      "epoch": 9.183047697708032,
+      "grad_norm": 0.07801347225904465,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 88947
+    },
+    {
+      "epoch": 9.183357423084864,
+      "grad_norm": 0.07705824077129364,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 88950
+    },
+    {
+      "epoch": 9.183667148461698,
+      "grad_norm": 0.13578467071056366,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 88953
+    },
+    {
+      "epoch": 9.18397687383853,
+      "grad_norm": 0.1082862839102745,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 88956
+    },
+    {
+      "epoch": 9.184286599215362,
+      "grad_norm": 0.08407740294933319,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 88959
+    },
+    {
+      "epoch": 9.184596324592196,
+      "grad_norm": 0.12279217690229416,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 88962
+    },
+    {
+      "epoch": 9.184906049969028,
+      "grad_norm": 0.14694684743881226,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 88965
+    },
+    {
+      "epoch": 9.18521577534586,
+      "grad_norm": 0.07982742041349411,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 88968
+    },
+    {
+      "epoch": 9.185525500722692,
+      "grad_norm": 0.08249851316213608,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 88971
+    },
+    {
+      "epoch": 9.185835226099526,
+      "grad_norm": 0.04720144718885422,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 88974
+    },
+    {
+      "epoch": 9.186144951476358,
+      "grad_norm": 0.09151851385831833,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 88977
+    },
+    {
+      "epoch": 9.18645467685319,
+      "grad_norm": 0.07975881546735764,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 88980
+    },
+    {
+      "epoch": 9.186764402230022,
+      "grad_norm": 0.05716300383210182,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 88983
+    },
+    {
+      "epoch": 9.187074127606856,
+      "grad_norm": 0.050935614854097366,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 88986
+    },
+    {
+      "epoch": 9.187383852983688,
+      "grad_norm": 0.05949466675519943,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 88989
+    },
+    {
+      "epoch": 9.18769357836052,
+      "grad_norm": 0.10654468089342117,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 88992
+    },
+    {
+      "epoch": 9.188003303737354,
+      "grad_norm": 0.055688001215457916,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 88995
+    },
+    {
+      "epoch": 9.188313029114186,
+      "grad_norm": 0.04441916570067406,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 88998
+    },
+    {
+      "epoch": 9.188622754491018,
+      "grad_norm": 0.18019713461399078,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 89001
+    },
+    {
+      "epoch": 9.18893247986785,
+      "grad_norm": 0.10449805855751038,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 89004
+    },
+    {
+      "epoch": 9.189242205244684,
+      "grad_norm": 0.09775479882955551,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 89007
+    },
+    {
+      "epoch": 9.189551930621516,
+      "grad_norm": 0.05437876284122467,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 89010
+    },
+    {
+      "epoch": 9.189861655998348,
+      "grad_norm": 0.05880746990442276,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 89013
+    },
+    {
+      "epoch": 9.190171381375182,
+      "grad_norm": 0.03767773509025574,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 89016
+    },
+    {
+      "epoch": 9.190481106752014,
+      "grad_norm": 0.04172288253903389,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 89019
+    },
+    {
+      "epoch": 9.190790832128846,
+      "grad_norm": 0.045086201280355453,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 89022
+    },
+    {
+      "epoch": 9.191100557505678,
+      "grad_norm": 0.04238392040133476,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 89025
+    },
+    {
+      "epoch": 9.191410282882511,
+      "grad_norm": 0.08289175480604172,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 89028
+    },
+    {
+      "epoch": 9.191720008259344,
+      "grad_norm": 0.054400261491537094,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 89031
+    },
+    {
+      "epoch": 9.192029733636176,
+      "grad_norm": 0.05191842466592789,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 89034
+    },
+    {
+      "epoch": 9.192339459013008,
+      "grad_norm": 0.048883236944675446,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 89037
+    },
+    {
+      "epoch": 9.192649184389841,
+      "grad_norm": 0.06198425218462944,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 89040
+    },
+    {
+      "epoch": 9.192958909766674,
+      "grad_norm": 0.04245370998978615,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 89043
+    },
+    {
+      "epoch": 9.193268635143506,
+      "grad_norm": 0.04683458060026169,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 89046
+    },
+    {
+      "epoch": 9.19357836052034,
+      "grad_norm": 0.04018238186836243,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 89049
+    },
+    {
+      "epoch": 9.193888085897171,
+      "grad_norm": 0.08557484298944473,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 89052
+    },
+    {
+      "epoch": 9.194197811274003,
+      "grad_norm": 0.1037377119064331,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 89055
+    },
+    {
+      "epoch": 9.194507536650836,
+      "grad_norm": 0.05675356835126877,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 89058
+    },
+    {
+      "epoch": 9.19481726202767,
+      "grad_norm": 0.05811937898397446,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 89061
+    },
+    {
+      "epoch": 9.195126987404501,
+      "grad_norm": 0.11276617646217346,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 89064
+    },
+    {
+      "epoch": 9.195436712781333,
+      "grad_norm": 0.06749601662158966,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 89067
+    },
+    {
+      "epoch": 9.195746438158167,
+      "grad_norm": 0.0429714135825634,
+      "learning_rate": 0.01,
+      "loss": 1.9013,
+      "step": 89070
+    },
+    {
+      "epoch": 9.196056163535,
+      "grad_norm": 0.03318256884813309,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 89073
+    },
+    {
+      "epoch": 9.196365888911831,
+      "grad_norm": 0.04414292052388191,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 89076
+    },
+    {
+      "epoch": 9.196675614288663,
+      "grad_norm": 0.08458261936903,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 89079
+    },
+    {
+      "epoch": 9.196985339665497,
+      "grad_norm": 0.04780903086066246,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 89082
+    },
+    {
+      "epoch": 9.19729506504233,
+      "grad_norm": 0.04293385520577431,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 89085
+    },
+    {
+      "epoch": 9.197604790419161,
+      "grad_norm": 0.04496946558356285,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 89088
+    },
+    {
+      "epoch": 9.197914515795993,
+      "grad_norm": 0.0704391747713089,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 89091
+    },
+    {
+      "epoch": 9.198224241172827,
+      "grad_norm": 0.09724891930818558,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 89094
+    },
+    {
+      "epoch": 9.19853396654966,
+      "grad_norm": 0.050086576491594315,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 89097
+    },
+    {
+      "epoch": 9.198843691926491,
+      "grad_norm": 0.037937577813863754,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 89100
+    },
+    {
+      "epoch": 9.199153417303325,
+      "grad_norm": 0.06578069925308228,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 89103
+    },
+    {
+      "epoch": 9.199463142680157,
+      "grad_norm": 0.13135863840579987,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 89106
+    },
+    {
+      "epoch": 9.19977286805699,
+      "grad_norm": 0.0574171282351017,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 89109
+    },
+    {
+      "epoch": 9.200082593433821,
+      "grad_norm": 0.044688817113637924,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 89112
+    },
+    {
+      "epoch": 9.200392318810655,
+      "grad_norm": 0.033014316111803055,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 89115
+    },
+    {
+      "epoch": 9.200702044187487,
+      "grad_norm": 0.1093095913529396,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 89118
+    },
+    {
+      "epoch": 9.20101176956432,
+      "grad_norm": 0.04808812960982323,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 89121
+    },
+    {
+      "epoch": 9.201321494941153,
+      "grad_norm": 0.11078277975320816,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 89124
+    },
+    {
+      "epoch": 9.201631220317985,
+      "grad_norm": 0.0814390480518341,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 89127
+    },
+    {
+      "epoch": 9.201940945694817,
+      "grad_norm": 0.058532990515232086,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 89130
+    },
+    {
+      "epoch": 9.20225067107165,
+      "grad_norm": 0.03977327421307564,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 89133
+    },
+    {
+      "epoch": 9.202560396448483,
+      "grad_norm": 0.04448265582323074,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 89136
+    },
+    {
+      "epoch": 9.202870121825315,
+      "grad_norm": 0.07655752450227737,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 89139
+    },
+    {
+      "epoch": 9.203179847202147,
+      "grad_norm": 0.04212338849902153,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 89142
+    },
+    {
+      "epoch": 9.20348957257898,
+      "grad_norm": 0.1010117158293724,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 89145
+    },
+    {
+      "epoch": 9.203799297955813,
+      "grad_norm": 0.061698488891124725,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 89148
+    },
+    {
+      "epoch": 9.204109023332645,
+      "grad_norm": 0.09103187173604965,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 89151
+    },
+    {
+      "epoch": 9.204418748709477,
+      "grad_norm": 0.04762488976120949,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 89154
+    },
+    {
+      "epoch": 9.204728474086311,
+      "grad_norm": 0.11716189235448837,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 89157
+    },
+    {
+      "epoch": 9.205038199463143,
+      "grad_norm": 0.05731340870261192,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 89160
+    },
+    {
+      "epoch": 9.205347924839975,
+      "grad_norm": 0.07926695793867111,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 89163
+    },
+    {
+      "epoch": 9.205657650216807,
+      "grad_norm": 0.1536962240934372,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 89166
+    },
+    {
+      "epoch": 9.20596737559364,
+      "grad_norm": 0.09955775737762451,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 89169
+    },
+    {
+      "epoch": 9.206277100970473,
+      "grad_norm": 0.07915548235177994,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 89172
+    },
+    {
+      "epoch": 9.206586826347305,
+      "grad_norm": 0.1031462550163269,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 89175
+    },
+    {
+      "epoch": 9.206896551724139,
+      "grad_norm": 0.05071650445461273,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 89178
+    },
+    {
+      "epoch": 9.20720627710097,
+      "grad_norm": 0.03473865985870361,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 89181
+    },
+    {
+      "epoch": 9.207516002477803,
+      "grad_norm": 0.03383462131023407,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 89184
+    },
+    {
+      "epoch": 9.207825727854635,
+      "grad_norm": 0.044631265103816986,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 89187
+    },
+    {
+      "epoch": 9.208135453231469,
+      "grad_norm": 0.040731918066740036,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 89190
+    },
+    {
+      "epoch": 9.2084451786083,
+      "grad_norm": 0.09363244473934174,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 89193
+    },
+    {
+      "epoch": 9.208754903985133,
+      "grad_norm": 0.06661606580018997,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 89196
+    },
+    {
+      "epoch": 9.209064629361965,
+      "grad_norm": 0.05025598779320717,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 89199
+    },
+    {
+      "epoch": 9.209374354738799,
+      "grad_norm": 0.11716365069150925,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 89202
+    },
+    {
+      "epoch": 9.20968408011563,
+      "grad_norm": 0.13573743402957916,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 89205
+    },
+    {
+      "epoch": 9.209993805492463,
+      "grad_norm": 0.049791369587183,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 89208
+    },
+    {
+      "epoch": 9.210303530869297,
+      "grad_norm": 0.03368228301405907,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 89211
+    },
+    {
+      "epoch": 9.210613256246129,
+      "grad_norm": 0.04646635428071022,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 89214
+    },
+    {
+      "epoch": 9.21092298162296,
+      "grad_norm": 0.08860430121421814,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 89217
+    },
+    {
+      "epoch": 9.211232706999793,
+      "grad_norm": 0.04521026834845543,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 89220
+    },
+    {
+      "epoch": 9.211542432376627,
+      "grad_norm": 0.04164019972085953,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 89223
+    },
+    {
+      "epoch": 9.211852157753459,
+      "grad_norm": 0.09966282546520233,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 89226
+    },
+    {
+      "epoch": 9.21216188313029,
+      "grad_norm": 0.07088325917720795,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 89229
+    },
+    {
+      "epoch": 9.212471608507125,
+      "grad_norm": 0.05550432950258255,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 89232
+    },
+    {
+      "epoch": 9.212781333883957,
+      "grad_norm": 0.041758205741643906,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 89235
+    },
+    {
+      "epoch": 9.213091059260789,
+      "grad_norm": 0.04863423854112625,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 89238
+    },
+    {
+      "epoch": 9.21340078463762,
+      "grad_norm": 0.05925336852669716,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 89241
+    },
+    {
+      "epoch": 9.213710510014455,
+      "grad_norm": 0.03509088233113289,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 89244
+    },
+    {
+      "epoch": 9.214020235391287,
+      "grad_norm": 0.04175510257482529,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 89247
+    },
+    {
+      "epoch": 9.214329960768119,
+      "grad_norm": 0.1422414481639862,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 89250
+    },
+    {
+      "epoch": 9.21463968614495,
+      "grad_norm": 0.039441660046577454,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 89253
+    },
+    {
+      "epoch": 9.214949411521784,
+      "grad_norm": 0.11482972651720047,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 89256
+    },
+    {
+      "epoch": 9.215259136898617,
+      "grad_norm": 0.10223820805549622,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 89259
+    },
+    {
+      "epoch": 9.215568862275449,
+      "grad_norm": 0.08964929729700089,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 89262
+    },
+    {
+      "epoch": 9.215878587652282,
+      "grad_norm": 0.08693050593137741,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 89265
+    },
+    {
+      "epoch": 9.216188313029114,
+      "grad_norm": 0.0800836980342865,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 89268
+    },
+    {
+      "epoch": 9.216498038405947,
+      "grad_norm": 0.04038820415735245,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 89271
+    },
+    {
+      "epoch": 9.216807763782779,
+      "grad_norm": 0.04846915602684021,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 89274
+    },
+    {
+      "epoch": 9.217117489159612,
+      "grad_norm": 0.046729911118745804,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 89277
+    },
+    {
+      "epoch": 9.217427214536444,
+      "grad_norm": 0.04161561653017998,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 89280
+    },
+    {
+      "epoch": 9.217736939913276,
+      "grad_norm": 0.07458865642547607,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 89283
+    },
+    {
+      "epoch": 9.21804666529011,
+      "grad_norm": 0.06652171164751053,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 89286
+    },
+    {
+      "epoch": 9.218356390666942,
+      "grad_norm": 0.07299696654081345,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 89289
+    },
+    {
+      "epoch": 9.218666116043774,
+      "grad_norm": 0.06781180948019028,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 89292
+    },
+    {
+      "epoch": 9.218975841420606,
+      "grad_norm": 0.06078062951564789,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 89295
+    },
+    {
+      "epoch": 9.21928556679744,
+      "grad_norm": 0.04806487634778023,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 89298
+    },
+    {
+      "epoch": 9.219595292174272,
+      "grad_norm": 0.03390311077237129,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 89301
+    },
+    {
+      "epoch": 9.219905017551104,
+      "grad_norm": 0.043576549738645554,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 89304
+    },
+    {
+      "epoch": 9.220214742927936,
+      "grad_norm": 0.058129116892814636,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 89307
+    },
+    {
+      "epoch": 9.22052446830477,
+      "grad_norm": 0.06500256806612015,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 89310
+    },
+    {
+      "epoch": 9.220834193681602,
+      "grad_norm": 0.12124396115541458,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 89313
+    },
+    {
+      "epoch": 9.221143919058434,
+      "grad_norm": 0.04002736508846283,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 89316
+    },
+    {
+      "epoch": 9.221453644435268,
+      "grad_norm": 0.10567335039377213,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 89319
+    },
+    {
+      "epoch": 9.2217633698121,
+      "grad_norm": 0.08138985186815262,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 89322
+    },
+    {
+      "epoch": 9.222073095188932,
+      "grad_norm": 0.056923553347587585,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 89325
+    },
+    {
+      "epoch": 9.222382820565764,
+      "grad_norm": 0.07420485466718674,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 89328
+    },
+    {
+      "epoch": 9.222692545942598,
+      "grad_norm": 0.05312581732869148,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 89331
+    },
+    {
+      "epoch": 9.22300227131943,
+      "grad_norm": 0.054729972034692764,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 89334
+    },
+    {
+      "epoch": 9.223311996696262,
+      "grad_norm": 0.0702265128493309,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 89337
+    },
+    {
+      "epoch": 9.223621722073096,
+      "grad_norm": 0.06925257295370102,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 89340
+    },
+    {
+      "epoch": 9.223931447449928,
+      "grad_norm": 0.03326869383454323,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 89343
+    },
+    {
+      "epoch": 9.22424117282676,
+      "grad_norm": 0.05594922602176666,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 89346
+    },
+    {
+      "epoch": 9.224550898203592,
+      "grad_norm": 0.054577749222517014,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 89349
+    },
+    {
+      "epoch": 9.224860623580426,
+      "grad_norm": 0.09225718677043915,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 89352
+    },
+    {
+      "epoch": 9.225170348957258,
+      "grad_norm": 0.05223730951547623,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 89355
+    },
+    {
+      "epoch": 9.22548007433409,
+      "grad_norm": 0.04215600714087486,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 89358
+    },
+    {
+      "epoch": 9.225789799710922,
+      "grad_norm": 0.03782760351896286,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 89361
+    },
+    {
+      "epoch": 9.226099525087756,
+      "grad_norm": 0.042087994515895844,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 89364
+    },
+    {
+      "epoch": 9.226409250464588,
+      "grad_norm": 0.10988534241914749,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 89367
+    },
+    {
+      "epoch": 9.22671897584142,
+      "grad_norm": 0.07139700651168823,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 89370
+    },
+    {
+      "epoch": 9.227028701218254,
+      "grad_norm": 0.03730588033795357,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 89373
+    },
+    {
+      "epoch": 9.227338426595086,
+      "grad_norm": 0.03242401033639908,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 89376
+    },
+    {
+      "epoch": 9.227648151971918,
+      "grad_norm": 0.08017165213823318,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 89379
+    },
+    {
+      "epoch": 9.22795787734875,
+      "grad_norm": 0.0404607355594635,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 89382
+    },
+    {
+      "epoch": 9.228267602725584,
+      "grad_norm": 0.12543754279613495,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 89385
+    },
+    {
+      "epoch": 9.228577328102416,
+      "grad_norm": 0.0824381411075592,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 89388
+    },
+    {
+      "epoch": 9.228887053479248,
+      "grad_norm": 0.0916644036769867,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 89391
+    },
+    {
+      "epoch": 9.229196778856082,
+      "grad_norm": 0.052297160029411316,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 89394
+    },
+    {
+      "epoch": 9.229506504232914,
+      "grad_norm": 0.043394576758146286,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 89397
+    },
+    {
+      "epoch": 9.229816229609746,
+      "grad_norm": 0.03213174641132355,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 89400
+    },
+    {
+      "epoch": 9.230125954986578,
+      "grad_norm": 0.1214103251695633,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 89403
+    },
+    {
+      "epoch": 9.230435680363412,
+      "grad_norm": 0.05036572739481926,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 89406
+    },
+    {
+      "epoch": 9.230745405740244,
+      "grad_norm": 0.07506284862756729,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 89409
+    },
+    {
+      "epoch": 9.231055131117076,
+      "grad_norm": 0.0447104275226593,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 89412
+    },
+    {
+      "epoch": 9.231364856493908,
+      "grad_norm": 0.031230248510837555,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 89415
+    },
+    {
+      "epoch": 9.231674581870742,
+      "grad_norm": 0.041907474398612976,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 89418
+    },
+    {
+      "epoch": 9.231984307247574,
+      "grad_norm": 0.054957807064056396,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 89421
+    },
+    {
+      "epoch": 9.232294032624406,
+      "grad_norm": 0.13959866762161255,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 89424
+    },
+    {
+      "epoch": 9.23260375800124,
+      "grad_norm": 0.07568775117397308,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 89427
+    },
+    {
+      "epoch": 9.232913483378072,
+      "grad_norm": 0.04290013760328293,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 89430
+    },
+    {
+      "epoch": 9.233223208754904,
+      "grad_norm": 0.08651076257228851,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 89433
+    },
+    {
+      "epoch": 9.233532934131736,
+      "grad_norm": 0.04303092509508133,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 89436
+    },
+    {
+      "epoch": 9.23384265950857,
+      "grad_norm": 0.03936919942498207,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 89439
+    },
+    {
+      "epoch": 9.234152384885402,
+      "grad_norm": 0.03311093524098396,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 89442
+    },
+    {
+      "epoch": 9.234462110262234,
+      "grad_norm": 0.07250311225652695,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 89445
+    },
+    {
+      "epoch": 9.234771835639068,
+      "grad_norm": 0.07465573400259018,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 89448
+    },
+    {
+      "epoch": 9.2350815610159,
+      "grad_norm": 0.04738716781139374,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 89451
+    },
+    {
+      "epoch": 9.235391286392732,
+      "grad_norm": 0.07104123383760452,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 89454
+    },
+    {
+      "epoch": 9.235701011769564,
+      "grad_norm": 0.04550927132368088,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 89457
+    },
+    {
+      "epoch": 9.236010737146398,
+      "grad_norm": 0.08226420730352402,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 89460
+    },
+    {
+      "epoch": 9.23632046252323,
+      "grad_norm": 0.0893448144197464,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 89463
+    },
+    {
+      "epoch": 9.236630187900062,
+      "grad_norm": 0.039117638021707535,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 89466
+    },
+    {
+      "epoch": 9.236939913276894,
+      "grad_norm": 0.03807968646287918,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 89469
+    },
+    {
+      "epoch": 9.237249638653728,
+      "grad_norm": 0.030970802530646324,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 89472
+    },
+    {
+      "epoch": 9.23755936403056,
+      "grad_norm": 0.04818034917116165,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 89475
+    },
+    {
+      "epoch": 9.237869089407392,
+      "grad_norm": 0.07317550480365753,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 89478
+    },
+    {
+      "epoch": 9.238178814784225,
+      "grad_norm": 0.10925126820802689,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 89481
+    },
+    {
+      "epoch": 9.238488540161057,
+      "grad_norm": 0.060674529522657394,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 89484
+    },
+    {
+      "epoch": 9.23879826553789,
+      "grad_norm": 0.1061653196811676,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 89487
+    },
+    {
+      "epoch": 9.239107990914722,
+      "grad_norm": 0.057163987308740616,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 89490
+    },
+    {
+      "epoch": 9.239417716291555,
+      "grad_norm": 0.05087054520845413,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 89493
+    },
+    {
+      "epoch": 9.239727441668387,
+      "grad_norm": 0.03398473933339119,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 89496
+    },
+    {
+      "epoch": 9.24003716704522,
+      "grad_norm": 0.14643120765686035,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 89499
+    },
+    {
+      "epoch": 9.240346892422053,
+      "grad_norm": 0.09905476868152618,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 89502
+    },
+    {
+      "epoch": 9.240656617798885,
+      "grad_norm": 0.09477792680263519,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 89505
+    },
+    {
+      "epoch": 9.240966343175717,
+      "grad_norm": 0.061807479709386826,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 89508
+    },
+    {
+      "epoch": 9.24127606855255,
+      "grad_norm": 0.09526131302118301,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 89511
+    },
+    {
+      "epoch": 9.241585793929383,
+      "grad_norm": 0.05511989817023277,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 89514
+    },
+    {
+      "epoch": 9.241895519306215,
+      "grad_norm": 0.047247253358364105,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 89517
+    },
+    {
+      "epoch": 9.242205244683047,
+      "grad_norm": 0.04097644239664078,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 89520
+    },
+    {
+      "epoch": 9.24251497005988,
+      "grad_norm": 0.041109759360551834,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 89523
+    },
+    {
+      "epoch": 9.242824695436713,
+      "grad_norm": 0.045235078781843185,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 89526
+    },
+    {
+      "epoch": 9.243134420813545,
+      "grad_norm": 0.04100433364510536,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 89529
+    },
+    {
+      "epoch": 9.243444146190377,
+      "grad_norm": 0.07918980717658997,
+      "learning_rate": 0.01,
+      "loss": 1.896,
+      "step": 89532
+    },
+    {
+      "epoch": 9.243753871567211,
+      "grad_norm": 0.06758083403110504,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 89535
+    },
+    {
+      "epoch": 9.244063596944043,
+      "grad_norm": 0.09762846678495407,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 89538
+    },
+    {
+      "epoch": 9.244373322320875,
+      "grad_norm": 0.07526645809412003,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 89541
+    },
+    {
+      "epoch": 9.244683047697707,
+      "grad_norm": 0.048858094960451126,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 89544
+    },
+    {
+      "epoch": 9.244992773074541,
+      "grad_norm": 0.05891689285635948,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 89547
+    },
+    {
+      "epoch": 9.245302498451373,
+      "grad_norm": 0.08054155856370926,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 89550
+    },
+    {
+      "epoch": 9.245612223828205,
+      "grad_norm": 0.10516653954982758,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 89553
+    },
+    {
+      "epoch": 9.245921949205039,
+      "grad_norm": 0.1483267843723297,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 89556
+    },
+    {
+      "epoch": 9.246231674581871,
+      "grad_norm": 0.09309061616659164,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 89559
+    },
+    {
+      "epoch": 9.246541399958703,
+      "grad_norm": 0.041449520736932755,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 89562
+    },
+    {
+      "epoch": 9.246851125335535,
+      "grad_norm": 0.05703721567988396,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 89565
+    },
+    {
+      "epoch": 9.247160850712369,
+      "grad_norm": 0.04464108496904373,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 89568
+    },
+    {
+      "epoch": 9.247470576089201,
+      "grad_norm": 0.04087911546230316,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 89571
+    },
+    {
+      "epoch": 9.247780301466033,
+      "grad_norm": 0.043997958302497864,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 89574
+    },
+    {
+      "epoch": 9.248090026842865,
+      "grad_norm": 0.06421921402215958,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 89577
+    },
+    {
+      "epoch": 9.248399752219699,
+      "grad_norm": 0.06022017076611519,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 89580
+    },
+    {
+      "epoch": 9.248709477596531,
+      "grad_norm": 0.11454983055591583,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 89583
+    },
+    {
+      "epoch": 9.249019202973363,
+      "grad_norm": 0.046563856303691864,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 89586
+    },
+    {
+      "epoch": 9.249328928350197,
+      "grad_norm": 0.049571454524993896,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 89589
+    },
+    {
+      "epoch": 9.249638653727029,
+      "grad_norm": 0.10863058269023895,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 89592
+    },
+    {
+      "epoch": 9.249948379103861,
+      "grad_norm": 0.07477369159460068,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 89595
+    },
+    {
+      "epoch": 9.250258104480693,
+      "grad_norm": 0.0527978241443634,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 89598
+    },
+    {
+      "epoch": 9.250567829857527,
+      "grad_norm": 0.07022270560264587,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 89601
+    },
+    {
+      "epoch": 9.250877555234359,
+      "grad_norm": 0.0500531829893589,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 89604
+    },
+    {
+      "epoch": 9.251187280611191,
+      "grad_norm": 0.038342274725437164,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 89607
+    },
+    {
+      "epoch": 9.251497005988025,
+      "grad_norm": 0.08321253955364227,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 89610
+    },
+    {
+      "epoch": 9.251806731364857,
+      "grad_norm": 0.06742928922176361,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 89613
+    },
+    {
+      "epoch": 9.252116456741689,
+      "grad_norm": 0.16153162717819214,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 89616
+    },
+    {
+      "epoch": 9.252426182118521,
+      "grad_norm": 0.06035587936639786,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 89619
+    },
+    {
+      "epoch": 9.252735907495355,
+      "grad_norm": 0.10703792423009872,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 89622
+    },
+    {
+      "epoch": 9.253045632872187,
+      "grad_norm": 0.08650069683790207,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 89625
+    },
+    {
+      "epoch": 9.253355358249019,
+      "grad_norm": 0.06951060891151428,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 89628
+    },
+    {
+      "epoch": 9.253665083625851,
+      "grad_norm": 0.03464649245142937,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 89631
+    },
+    {
+      "epoch": 9.253974809002685,
+      "grad_norm": 0.04344501346349716,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 89634
+    },
+    {
+      "epoch": 9.254284534379517,
+      "grad_norm": 0.07010316103696823,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 89637
+    },
+    {
+      "epoch": 9.254594259756349,
+      "grad_norm": 0.11068741232156754,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 89640
+    },
+    {
+      "epoch": 9.254903985133183,
+      "grad_norm": 0.1069609746336937,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 89643
+    },
+    {
+      "epoch": 9.255213710510015,
+      "grad_norm": 0.03556428849697113,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 89646
+    },
+    {
+      "epoch": 9.255523435886847,
+      "grad_norm": 0.07954979687929153,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 89649
+    },
+    {
+      "epoch": 9.255833161263679,
+      "grad_norm": 0.12106265872716904,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 89652
+    },
+    {
+      "epoch": 9.256142886640513,
+      "grad_norm": 0.06580674648284912,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 89655
+    },
+    {
+      "epoch": 9.256452612017345,
+      "grad_norm": 0.06531988084316254,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 89658
+    },
+    {
+      "epoch": 9.256762337394177,
+      "grad_norm": 0.09100288897752762,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 89661
+    },
+    {
+      "epoch": 9.257072062771009,
+      "grad_norm": 0.04378080740571022,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 89664
+    },
+    {
+      "epoch": 9.257381788147843,
+      "grad_norm": 0.08889858424663544,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 89667
+    },
+    {
+      "epoch": 9.257691513524675,
+      "grad_norm": 0.09178147464990616,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 89670
+    },
+    {
+      "epoch": 9.258001238901507,
+      "grad_norm": 0.12640753388404846,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 89673
+    },
+    {
+      "epoch": 9.25831096427834,
+      "grad_norm": 0.08799244463443756,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 89676
+    },
+    {
+      "epoch": 9.258620689655173,
+      "grad_norm": 0.07690322399139404,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 89679
+    },
+    {
+      "epoch": 9.258930415032005,
+      "grad_norm": 0.07583325356245041,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 89682
+    },
+    {
+      "epoch": 9.259240140408837,
+      "grad_norm": 0.041603825986385345,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 89685
+    },
+    {
+      "epoch": 9.25954986578567,
+      "grad_norm": 0.14184615015983582,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 89688
+    },
+    {
+      "epoch": 9.259859591162503,
+      "grad_norm": 0.08319629728794098,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 89691
+    },
+    {
+      "epoch": 9.260169316539335,
+      "grad_norm": 0.09747131168842316,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 89694
+    },
+    {
+      "epoch": 9.260479041916168,
+      "grad_norm": 3.834672451019287,
+      "learning_rate": 0.01,
+      "loss": 1.9896,
+      "step": 89697
+    },
+    {
+      "epoch": 9.260788767293,
+      "grad_norm": 0.20725275576114655,
+      "learning_rate": 0.01,
+      "loss": 2.0049,
+      "step": 89700
+    },
+    {
+      "epoch": 9.261098492669833,
+      "grad_norm": 0.16313984990119934,
+      "learning_rate": 0.01,
+      "loss": 1.9805,
+      "step": 89703
+    },
+    {
+      "epoch": 9.261408218046665,
+      "grad_norm": 0.08664103597402573,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 89706
+    },
+    {
+      "epoch": 9.261717943423498,
+      "grad_norm": 0.05102289468050003,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 89709
+    },
+    {
+      "epoch": 9.26202766880033,
+      "grad_norm": 0.05602896586060524,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 89712
+    },
+    {
+      "epoch": 9.262337394177163,
+      "grad_norm": 0.04888107627630234,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 89715
+    },
+    {
+      "epoch": 9.262647119553996,
+      "grad_norm": 0.06098625808954239,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 89718
+    },
+    {
+      "epoch": 9.262956844930828,
+      "grad_norm": 0.03745686635375023,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 89721
+    },
+    {
+      "epoch": 9.26326657030766,
+      "grad_norm": 0.035396233201026917,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 89724
+    },
+    {
+      "epoch": 9.263576295684492,
+      "grad_norm": 0.04478583112359047,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 89727
+    },
+    {
+      "epoch": 9.263886021061326,
+      "grad_norm": 0.05072050169110298,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 89730
+    },
+    {
+      "epoch": 9.264195746438158,
+      "grad_norm": 0.09497334808111191,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 89733
+    },
+    {
+      "epoch": 9.26450547181499,
+      "grad_norm": 0.07028527557849884,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 89736
+    },
+    {
+      "epoch": 9.264815197191822,
+      "grad_norm": 0.11074947565793991,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 89739
+    },
+    {
+      "epoch": 9.265124922568656,
+      "grad_norm": 0.051569730043411255,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 89742
+    },
+    {
+      "epoch": 9.265434647945488,
+      "grad_norm": 0.04874719679355621,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 89745
+    },
+    {
+      "epoch": 9.26574437332232,
+      "grad_norm": 0.04530943185091019,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 89748
+    },
+    {
+      "epoch": 9.266054098699154,
+      "grad_norm": 0.06027248129248619,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 89751
+    },
+    {
+      "epoch": 9.266363824075986,
+      "grad_norm": 0.07296335697174072,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 89754
+    },
+    {
+      "epoch": 9.266673549452818,
+      "grad_norm": 0.06033644825220108,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 89757
+    },
+    {
+      "epoch": 9.26698327482965,
+      "grad_norm": 0.04669598489999771,
+      "learning_rate": 0.01,
+      "loss": 1.8868,
+      "step": 89760
+    },
+    {
+      "epoch": 9.267293000206484,
+      "grad_norm": 0.04095854610204697,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 89763
+    },
+    {
+      "epoch": 9.267602725583316,
+      "grad_norm": 0.045767225325107574,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 89766
+    },
+    {
+      "epoch": 9.267912450960148,
+      "grad_norm": 0.10907436907291412,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 89769
+    },
+    {
+      "epoch": 9.26822217633698,
+      "grad_norm": 0.06417029350996017,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 89772
+    },
+    {
+      "epoch": 9.268531901713814,
+      "grad_norm": 0.049392636865377426,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 89775
+    },
+    {
+      "epoch": 9.268841627090646,
+      "grad_norm": 0.06661033630371094,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 89778
+    },
+    {
+      "epoch": 9.269151352467478,
+      "grad_norm": 0.04778747260570526,
+      "learning_rate": 0.01,
+      "loss": 1.8937,
+      "step": 89781
+    },
+    {
+      "epoch": 9.269461077844312,
+      "grad_norm": 0.06895989179611206,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 89784
+    },
+    {
+      "epoch": 9.269770803221144,
+      "grad_norm": 0.06295426189899445,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 89787
+    },
+    {
+      "epoch": 9.270080528597976,
+      "grad_norm": 0.09221582114696503,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 89790
+    },
+    {
+      "epoch": 9.270390253974808,
+      "grad_norm": 0.04774243384599686,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 89793
+    },
+    {
+      "epoch": 9.270699979351642,
+      "grad_norm": 0.06828971207141876,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 89796
+    },
+    {
+      "epoch": 9.271009704728474,
+      "grad_norm": 0.054797038435935974,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 89799
+    },
+    {
+      "epoch": 9.271319430105306,
+      "grad_norm": 0.03856116533279419,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 89802
+    },
+    {
+      "epoch": 9.27162915548214,
+      "grad_norm": 0.035631731152534485,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 89805
+    },
+    {
+      "epoch": 9.271938880858972,
+      "grad_norm": 0.057642899453639984,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 89808
+    },
+    {
+      "epoch": 9.272248606235804,
+      "grad_norm": 0.08952697366476059,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 89811
+    },
+    {
+      "epoch": 9.272558331612636,
+      "grad_norm": 0.04947827383875847,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 89814
+    },
+    {
+      "epoch": 9.27286805698947,
+      "grad_norm": 0.06767935305833817,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 89817
+    },
+    {
+      "epoch": 9.273177782366302,
+      "grad_norm": 0.050101347267627716,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 89820
+    },
+    {
+      "epoch": 9.273487507743134,
+      "grad_norm": 0.03736945986747742,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 89823
+    },
+    {
+      "epoch": 9.273797233119968,
+      "grad_norm": 0.12808963656425476,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 89826
+    },
+    {
+      "epoch": 9.2741069584968,
+      "grad_norm": 0.058463335037231445,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 89829
+    },
+    {
+      "epoch": 9.274416683873632,
+      "grad_norm": 0.06605935841798782,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 89832
+    },
+    {
+      "epoch": 9.274726409250464,
+      "grad_norm": 0.09979791194200516,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 89835
+    },
+    {
+      "epoch": 9.275036134627298,
+      "grad_norm": 0.03688298538327217,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 89838
+    },
+    {
+      "epoch": 9.27534586000413,
+      "grad_norm": 0.06498592346906662,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 89841
+    },
+    {
+      "epoch": 9.275655585380962,
+      "grad_norm": 0.04934919252991676,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 89844
+    },
+    {
+      "epoch": 9.275965310757794,
+      "grad_norm": 0.0446552038192749,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 89847
+    },
+    {
+      "epoch": 9.276275036134628,
+      "grad_norm": 0.052687738090753555,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 89850
+    },
+    {
+      "epoch": 9.27658476151146,
+      "grad_norm": 0.03818351775407791,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 89853
+    },
+    {
+      "epoch": 9.276894486888292,
+      "grad_norm": 0.042557138949632645,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 89856
+    },
+    {
+      "epoch": 9.277204212265126,
+      "grad_norm": 0.055316612124443054,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 89859
+    },
+    {
+      "epoch": 9.277513937641958,
+      "grad_norm": 0.09695148468017578,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 89862
+    },
+    {
+      "epoch": 9.27782366301879,
+      "grad_norm": 0.09990256279706955,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 89865
+    },
+    {
+      "epoch": 9.278133388395622,
+      "grad_norm": 0.07283806800842285,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 89868
+    },
+    {
+      "epoch": 9.278443113772456,
+      "grad_norm": 0.08778601884841919,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 89871
+    },
+    {
+      "epoch": 9.278752839149288,
+      "grad_norm": 0.038158684968948364,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 89874
+    },
+    {
+      "epoch": 9.27906256452612,
+      "grad_norm": 0.03885229304432869,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 89877
+    },
+    {
+      "epoch": 9.279372289902952,
+      "grad_norm": 0.09096448123455048,
+      "learning_rate": 0.01,
+      "loss": 1.8973,
+      "step": 89880
+    },
+    {
+      "epoch": 9.279682015279786,
+      "grad_norm": 0.04410238564014435,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 89883
+    },
+    {
+      "epoch": 9.279991740656618,
+      "grad_norm": 0.05357813090085983,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 89886
+    },
+    {
+      "epoch": 9.28030146603345,
+      "grad_norm": 0.08851207047700882,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 89889
+    },
+    {
+      "epoch": 9.280611191410284,
+      "grad_norm": 0.1310880333185196,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 89892
+    },
+    {
+      "epoch": 9.280920916787116,
+      "grad_norm": 0.06100614741444588,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 89895
+    },
+    {
+      "epoch": 9.281230642163948,
+      "grad_norm": 0.05535770207643509,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 89898
+    },
+    {
+      "epoch": 9.28154036754078,
+      "grad_norm": 0.05252685025334358,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 89901
+    },
+    {
+      "epoch": 9.281850092917614,
+      "grad_norm": 0.044191114604473114,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 89904
+    },
+    {
+      "epoch": 9.282159818294446,
+      "grad_norm": 0.05596025660634041,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 89907
+    },
+    {
+      "epoch": 9.282469543671278,
+      "grad_norm": 0.08656829595565796,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 89910
+    },
+    {
+      "epoch": 9.282779269048111,
+      "grad_norm": 0.05122004449367523,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 89913
+    },
+    {
+      "epoch": 9.283088994424944,
+      "grad_norm": 0.08028247952461243,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 89916
+    },
+    {
+      "epoch": 9.283398719801776,
+      "grad_norm": 0.06237107142806053,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 89919
+    },
+    {
+      "epoch": 9.283708445178608,
+      "grad_norm": 0.12027272582054138,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 89922
+    },
+    {
+      "epoch": 9.284018170555441,
+      "grad_norm": 0.08568006008863449,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 89925
+    },
+    {
+      "epoch": 9.284327895932273,
+      "grad_norm": 0.053323738276958466,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 89928
+    },
+    {
+      "epoch": 9.284637621309106,
+      "grad_norm": 0.12184597551822662,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 89931
+    },
+    {
+      "epoch": 9.28494734668594,
+      "grad_norm": 0.11333894729614258,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 89934
+    },
+    {
+      "epoch": 9.285257072062771,
+      "grad_norm": 0.06243620440363884,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 89937
+    },
+    {
+      "epoch": 9.285566797439603,
+      "grad_norm": 0.04748154804110527,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 89940
+    },
+    {
+      "epoch": 9.285876522816435,
+      "grad_norm": 0.031249042600393295,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 89943
+    },
+    {
+      "epoch": 9.28618624819327,
+      "grad_norm": 0.036813315004110336,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 89946
+    },
+    {
+      "epoch": 9.286495973570101,
+      "grad_norm": 0.03633863851428032,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 89949
+    },
+    {
+      "epoch": 9.286805698946933,
+      "grad_norm": 0.054639723151922226,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 89952
+    },
+    {
+      "epoch": 9.287115424323765,
+      "grad_norm": 0.09005313366651535,
+      "learning_rate": 0.01,
+      "loss": 1.9005,
+      "step": 89955
+    },
+    {
+      "epoch": 9.2874251497006,
+      "grad_norm": 0.1011664867401123,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 89958
+    },
+    {
+      "epoch": 9.287734875077431,
+      "grad_norm": 0.10966178774833679,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 89961
+    },
+    {
+      "epoch": 9.288044600454263,
+      "grad_norm": 0.05809110030531883,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 89964
+    },
+    {
+      "epoch": 9.288354325831097,
+      "grad_norm": 0.04828932136297226,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 89967
+    },
+    {
+      "epoch": 9.28866405120793,
+      "grad_norm": 0.04604678228497505,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 89970
+    },
+    {
+      "epoch": 9.288973776584761,
+      "grad_norm": 0.05810960382223129,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 89973
+    },
+    {
+      "epoch": 9.289283501961593,
+      "grad_norm": 0.05069134756922722,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 89976
+    },
+    {
+      "epoch": 9.289593227338427,
+      "grad_norm": 0.061285290867090225,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 89979
+    },
+    {
+      "epoch": 9.28990295271526,
+      "grad_norm": 0.11732938140630722,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 89982
+    },
+    {
+      "epoch": 9.290212678092091,
+      "grad_norm": 0.08811193704605103,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 89985
+    },
+    {
+      "epoch": 9.290522403468923,
+      "grad_norm": 0.039724934846162796,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 89988
+    },
+    {
+      "epoch": 9.290832128845757,
+      "grad_norm": 0.049808014184236526,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 89991
+    },
+    {
+      "epoch": 9.29114185422259,
+      "grad_norm": 0.041384246200323105,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 89994
+    },
+    {
+      "epoch": 9.291451579599421,
+      "grad_norm": 0.09925634413957596,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 89997
+    },
+    {
+      "epoch": 9.291761304976255,
+      "grad_norm": 0.04714850336313248,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 90000
+    },
+    {
+      "epoch": 9.292071030353087,
+      "grad_norm": 0.07816148549318314,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 90003
+    },
+    {
+      "epoch": 9.29238075572992,
+      "grad_norm": 0.037051137536764145,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 90006
+    },
+    {
+      "epoch": 9.292690481106751,
+      "grad_norm": 0.03261815384030342,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 90009
+    },
+    {
+      "epoch": 9.293000206483585,
+      "grad_norm": 0.061712704598903656,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 90012
+    },
+    {
+      "epoch": 9.293309931860417,
+      "grad_norm": 0.09313449263572693,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 90015
+    },
+    {
+      "epoch": 9.29361965723725,
+      "grad_norm": 0.06282445043325424,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 90018
+    },
+    {
+      "epoch": 9.293929382614083,
+      "grad_norm": 0.04284929111599922,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 90021
+    },
+    {
+      "epoch": 9.294239107990915,
+      "grad_norm": 0.0422949492931366,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 90024
+    },
+    {
+      "epoch": 9.294548833367747,
+      "grad_norm": 0.058564431965351105,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 90027
+    },
+    {
+      "epoch": 9.29485855874458,
+      "grad_norm": 0.10857044905424118,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 90030
+    },
+    {
+      "epoch": 9.295168284121413,
+      "grad_norm": 0.048352450132369995,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 90033
+    },
+    {
+      "epoch": 9.295478009498245,
+      "grad_norm": 0.03552021086215973,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 90036
+    },
+    {
+      "epoch": 9.295787734875077,
+      "grad_norm": 0.050572823733091354,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 90039
+    },
+    {
+      "epoch": 9.29609746025191,
+      "grad_norm": 0.06984183937311172,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 90042
+    },
+    {
+      "epoch": 9.296407185628743,
+      "grad_norm": 0.06555435806512833,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 90045
+    },
+    {
+      "epoch": 9.296716911005575,
+      "grad_norm": 0.05915297940373421,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 90048
+    },
+    {
+      "epoch": 9.297026636382407,
+      "grad_norm": 0.10481391102075577,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 90051
+    },
+    {
+      "epoch": 9.29733636175924,
+      "grad_norm": 0.05892582982778549,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 90054
+    },
+    {
+      "epoch": 9.297646087136073,
+      "grad_norm": 0.042443741112947464,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 90057
+    },
+    {
+      "epoch": 9.297955812512905,
+      "grad_norm": 0.11886375397443771,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 90060
+    },
+    {
+      "epoch": 9.298265537889737,
+      "grad_norm": 0.0912526547908783,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 90063
+    },
+    {
+      "epoch": 9.29857526326657,
+      "grad_norm": 0.06104797124862671,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 90066
+    },
+    {
+      "epoch": 9.298884988643403,
+      "grad_norm": 0.04334283620119095,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 90069
+    },
+    {
+      "epoch": 9.299194714020235,
+      "grad_norm": 0.0571666918694973,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 90072
+    },
+    {
+      "epoch": 9.299504439397069,
+      "grad_norm": 0.04852711036801338,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 90075
+    },
+    {
+      "epoch": 9.2998141647739,
+      "grad_norm": 0.03939691558480263,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 90078
+    },
+    {
+      "epoch": 9.300123890150733,
+      "grad_norm": 0.07253614813089371,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 90081
+    },
+    {
+      "epoch": 9.300433615527565,
+      "grad_norm": 0.07965242862701416,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 90084
+    },
+    {
+      "epoch": 9.300743340904399,
+      "grad_norm": 0.08113238215446472,
+      "learning_rate": 0.01,
+      "loss": 1.9017,
+      "step": 90087
+    },
+    {
+      "epoch": 9.30105306628123,
+      "grad_norm": 0.048515964299440384,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 90090
+    },
+    {
+      "epoch": 9.301362791658063,
+      "grad_norm": 0.12356071174144745,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 90093
+    },
+    {
+      "epoch": 9.301672517034895,
+      "grad_norm": 0.05115806311368942,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 90096
+    },
+    {
+      "epoch": 9.301982242411729,
+      "grad_norm": 0.07609254121780396,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 90099
+    },
+    {
+      "epoch": 9.30229196778856,
+      "grad_norm": 0.07136466354131699,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 90102
+    },
+    {
+      "epoch": 9.302601693165393,
+      "grad_norm": 0.043960392475128174,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 90105
+    },
+    {
+      "epoch": 9.302911418542227,
+      "grad_norm": 0.034706417471170425,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 90108
+    },
+    {
+      "epoch": 9.303221143919059,
+      "grad_norm": 0.055189475417137146,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 90111
+    },
+    {
+      "epoch": 9.30353086929589,
+      "grad_norm": 0.06000189483165741,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 90114
+    },
+    {
+      "epoch": 9.303840594672723,
+      "grad_norm": 0.04692503809928894,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 90117
+    },
+    {
+      "epoch": 9.304150320049557,
+      "grad_norm": 0.09640234708786011,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 90120
+    },
+    {
+      "epoch": 9.304460045426389,
+      "grad_norm": 0.03152190148830414,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 90123
+    },
+    {
+      "epoch": 9.30476977080322,
+      "grad_norm": 0.03566571697592735,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 90126
+    },
+    {
+      "epoch": 9.305079496180054,
+      "grad_norm": 0.04585658758878708,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 90129
+    },
+    {
+      "epoch": 9.305389221556887,
+      "grad_norm": 0.033661067485809326,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 90132
+    },
+    {
+      "epoch": 9.305698946933719,
+      "grad_norm": 0.07970356196165085,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 90135
+    },
+    {
+      "epoch": 9.30600867231055,
+      "grad_norm": 0.15736587345600128,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 90138
+    },
+    {
+      "epoch": 9.306318397687384,
+      "grad_norm": 0.10455471277236938,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 90141
+    },
+    {
+      "epoch": 9.306628123064216,
+      "grad_norm": 0.08667217195034027,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 90144
+    },
+    {
+      "epoch": 9.306937848441049,
+      "grad_norm": 0.08747159689664841,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 90147
+    },
+    {
+      "epoch": 9.307247573817882,
+      "grad_norm": 0.05152470991015434,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 90150
+    },
+    {
+      "epoch": 9.307557299194714,
+      "grad_norm": 0.041649870574474335,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 90153
+    },
+    {
+      "epoch": 9.307867024571546,
+      "grad_norm": 0.07984058558940887,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 90156
+    },
+    {
+      "epoch": 9.308176749948379,
+      "grad_norm": 0.03657011687755585,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 90159
+    },
+    {
+      "epoch": 9.308486475325212,
+      "grad_norm": 0.052740272134542465,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 90162
+    },
+    {
+      "epoch": 9.308796200702044,
+      "grad_norm": 0.037107206881046295,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 90165
+    },
+    {
+      "epoch": 9.309105926078876,
+      "grad_norm": 0.047946516424417496,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 90168
+    },
+    {
+      "epoch": 9.309415651455708,
+      "grad_norm": 0.036284878849983215,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 90171
+    },
+    {
+      "epoch": 9.309725376832542,
+      "grad_norm": 0.09453091770410538,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 90174
+    },
+    {
+      "epoch": 9.310035102209374,
+      "grad_norm": 0.0442422591149807,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 90177
+    },
+    {
+      "epoch": 9.310344827586206,
+      "grad_norm": 0.08194606751203537,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 90180
+    },
+    {
+      "epoch": 9.31065455296304,
+      "grad_norm": 0.08908070623874664,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 90183
+    },
+    {
+      "epoch": 9.310964278339872,
+      "grad_norm": 0.08012444525957108,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 90186
+    },
+    {
+      "epoch": 9.311274003716704,
+      "grad_norm": 0.05593520402908325,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 90189
+    },
+    {
+      "epoch": 9.311583729093536,
+      "grad_norm": 0.10340272635221481,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 90192
+    },
+    {
+      "epoch": 9.31189345447037,
+      "grad_norm": 0.06450596451759338,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 90195
+    },
+    {
+      "epoch": 9.312203179847202,
+      "grad_norm": 0.06476901471614838,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 90198
+    },
+    {
+      "epoch": 9.312512905224034,
+      "grad_norm": 0.04243314638733864,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 90201
+    },
+    {
+      "epoch": 9.312822630600866,
+      "grad_norm": 0.052618399262428284,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 90204
+    },
+    {
+      "epoch": 9.3131323559777,
+      "grad_norm": 0.049052439630031586,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 90207
+    },
+    {
+      "epoch": 9.313442081354532,
+      "grad_norm": 0.08724159002304077,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 90210
+    },
+    {
+      "epoch": 9.313751806731364,
+      "grad_norm": 0.08713918179273605,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 90213
+    },
+    {
+      "epoch": 9.314061532108198,
+      "grad_norm": 0.0694577768445015,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 90216
+    },
+    {
+      "epoch": 9.31437125748503,
+      "grad_norm": 0.07178552448749542,
+      "learning_rate": 0.01,
+      "loss": 1.9807,
+      "step": 90219
+    },
+    {
+      "epoch": 9.314680982861862,
+      "grad_norm": 0.08677991479635239,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 90222
+    },
+    {
+      "epoch": 9.314990708238694,
+      "grad_norm": 0.054113663733005524,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 90225
+    },
+    {
+      "epoch": 9.315300433615528,
+      "grad_norm": 0.12051310390233994,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 90228
+    },
+    {
+      "epoch": 9.31561015899236,
+      "grad_norm": 0.04537734016776085,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 90231
+    },
+    {
+      "epoch": 9.315919884369192,
+      "grad_norm": 0.04667481780052185,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 90234
+    },
+    {
+      "epoch": 9.316229609746026,
+      "grad_norm": 0.12717698514461517,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 90237
+    },
+    {
+      "epoch": 9.316539335122858,
+      "grad_norm": 0.06177690252661705,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 90240
+    },
+    {
+      "epoch": 9.31684906049969,
+      "grad_norm": 0.16978679597377777,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 90243
+    },
+    {
+      "epoch": 9.317158785876522,
+      "grad_norm": 0.13136741518974304,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 90246
+    },
+    {
+      "epoch": 9.317468511253356,
+      "grad_norm": 0.07961270958185196,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 90249
+    },
+    {
+      "epoch": 9.317778236630188,
+      "grad_norm": 0.0522325336933136,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 90252
+    },
+    {
+      "epoch": 9.31808796200702,
+      "grad_norm": 0.05689280480146408,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 90255
+    },
+    {
+      "epoch": 9.318397687383854,
+      "grad_norm": 0.06072314828634262,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 90258
+    },
+    {
+      "epoch": 9.318707412760686,
+      "grad_norm": 0.045333363115787506,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 90261
+    },
+    {
+      "epoch": 9.319017138137518,
+      "grad_norm": 0.046115368604660034,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 90264
+    },
+    {
+      "epoch": 9.31932686351435,
+      "grad_norm": 0.03999978303909302,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 90267
+    },
+    {
+      "epoch": 9.319636588891184,
+      "grad_norm": 0.04297192022204399,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 90270
+    },
+    {
+      "epoch": 9.319946314268016,
+      "grad_norm": 0.03738616779446602,
+      "learning_rate": 0.01,
+      "loss": 1.9768,
+      "step": 90273
+    },
+    {
+      "epoch": 9.320256039644848,
+      "grad_norm": 0.08931316435337067,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 90276
+    },
+    {
+      "epoch": 9.32056576502168,
+      "grad_norm": 0.19095301628112793,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 90279
+    },
+    {
+      "epoch": 9.320875490398514,
+      "grad_norm": 0.12510459125041962,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 90282
+    },
+    {
+      "epoch": 9.321185215775346,
+      "grad_norm": 0.09745209664106369,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 90285
+    },
+    {
+      "epoch": 9.321494941152178,
+      "grad_norm": 0.08224628865718842,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 90288
+    },
+    {
+      "epoch": 9.321804666529012,
+      "grad_norm": 0.05055355280637741,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 90291
+    },
+    {
+      "epoch": 9.322114391905844,
+      "grad_norm": 0.07061073184013367,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 90294
+    },
+    {
+      "epoch": 9.322424117282676,
+      "grad_norm": 0.06762760132551193,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 90297
+    },
+    {
+      "epoch": 9.322733842659508,
+      "grad_norm": 0.049081481993198395,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 90300
+    },
+    {
+      "epoch": 9.323043568036342,
+      "grad_norm": 0.045287519693374634,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 90303
+    },
+    {
+      "epoch": 9.323353293413174,
+      "grad_norm": 0.03433917835354805,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 90306
+    },
+    {
+      "epoch": 9.323663018790006,
+      "grad_norm": 0.10234330594539642,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 90309
+    },
+    {
+      "epoch": 9.323972744166838,
+      "grad_norm": 0.044973790645599365,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 90312
+    },
+    {
+      "epoch": 9.324282469543672,
+      "grad_norm": 0.10993959754705429,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 90315
+    },
+    {
+      "epoch": 9.324592194920504,
+      "grad_norm": 0.147218719124794,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 90318
+    },
+    {
+      "epoch": 9.324901920297336,
+      "grad_norm": 0.10825378447771072,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 90321
+    },
+    {
+      "epoch": 9.32521164567417,
+      "grad_norm": 0.12275347113609314,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 90324
+    },
+    {
+      "epoch": 9.325521371051002,
+      "grad_norm": 0.06429765373468399,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 90327
+    },
+    {
+      "epoch": 9.325831096427834,
+      "grad_norm": 0.05523232743144035,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 90330
+    },
+    {
+      "epoch": 9.326140821804666,
+      "grad_norm": 0.03777391090989113,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 90333
+    },
+    {
+      "epoch": 9.3264505471815,
+      "grad_norm": 0.04383401945233345,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 90336
+    },
+    {
+      "epoch": 9.326760272558332,
+      "grad_norm": 0.05174276977777481,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 90339
+    },
+    {
+      "epoch": 9.327069997935164,
+      "grad_norm": 0.041363853961229324,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 90342
+    },
+    {
+      "epoch": 9.327379723311997,
+      "grad_norm": 0.10272231698036194,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 90345
+    },
+    {
+      "epoch": 9.32768944868883,
+      "grad_norm": 0.06003765016794205,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 90348
+    },
+    {
+      "epoch": 9.327999174065662,
+      "grad_norm": 0.07973077148199081,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 90351
+    },
+    {
+      "epoch": 9.328308899442494,
+      "grad_norm": 0.11034880578517914,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 90354
+    },
+    {
+      "epoch": 9.328618624819327,
+      "grad_norm": 0.12927527725696564,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 90357
+    },
+    {
+      "epoch": 9.32892835019616,
+      "grad_norm": 0.039803747087717056,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 90360
+    },
+    {
+      "epoch": 9.329238075572992,
+      "grad_norm": 0.04133141040802002,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 90363
+    },
+    {
+      "epoch": 9.329547800949825,
+      "grad_norm": 0.04647175967693329,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 90366
+    },
+    {
+      "epoch": 9.329857526326657,
+      "grad_norm": 0.03702395781874657,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 90369
+    },
+    {
+      "epoch": 9.33016725170349,
+      "grad_norm": 0.04293076694011688,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 90372
+    },
+    {
+      "epoch": 9.330476977080322,
+      "grad_norm": 0.04811659827828407,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 90375
+    },
+    {
+      "epoch": 9.330786702457155,
+      "grad_norm": 0.09386304020881653,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 90378
+    },
+    {
+      "epoch": 9.331096427833987,
+      "grad_norm": 0.05897035449743271,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 90381
+    },
+    {
+      "epoch": 9.33140615321082,
+      "grad_norm": 0.041449174284935,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 90384
+    },
+    {
+      "epoch": 9.331715878587652,
+      "grad_norm": 0.03846067562699318,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 90387
+    },
+    {
+      "epoch": 9.332025603964485,
+      "grad_norm": 0.04674214497208595,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 90390
+    },
+    {
+      "epoch": 9.332335329341317,
+      "grad_norm": 0.06381560117006302,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 90393
+    },
+    {
+      "epoch": 9.33264505471815,
+      "grad_norm": 0.060292668640613556,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 90396
+    },
+    {
+      "epoch": 9.332954780094983,
+      "grad_norm": 0.08171455562114716,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 90399
+    },
+    {
+      "epoch": 9.333264505471815,
+      "grad_norm": 0.07539816200733185,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 90402
+    },
+    {
+      "epoch": 9.333574230848647,
+      "grad_norm": 0.16030851006507874,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 90405
+    },
+    {
+      "epoch": 9.33388395622548,
+      "grad_norm": 0.0498042106628418,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 90408
+    },
+    {
+      "epoch": 9.334193681602313,
+      "grad_norm": 0.035880159586668015,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 90411
+    },
+    {
+      "epoch": 9.334503406979145,
+      "grad_norm": 0.032716989517211914,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 90414
+    },
+    {
+      "epoch": 9.334813132355977,
+      "grad_norm": 0.05374199151992798,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 90417
+    },
+    {
+      "epoch": 9.33512285773281,
+      "grad_norm": 0.0500338152050972,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 90420
+    },
+    {
+      "epoch": 9.335432583109643,
+      "grad_norm": 0.05141221731901169,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 90423
+    },
+    {
+      "epoch": 9.335742308486475,
+      "grad_norm": 0.1170889213681221,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 90426
+    },
+    {
+      "epoch": 9.336052033863307,
+      "grad_norm": 0.08207867294549942,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 90429
+    },
+    {
+      "epoch": 9.336361759240141,
+      "grad_norm": 0.126504048705101,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 90432
+    },
+    {
+      "epoch": 9.336671484616973,
+      "grad_norm": 0.09300342202186584,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 90435
+    },
+    {
+      "epoch": 9.336981209993805,
+      "grad_norm": 0.11592224985361099,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 90438
+    },
+    {
+      "epoch": 9.337290935370637,
+      "grad_norm": 0.07160703837871552,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 90441
+    },
+    {
+      "epoch": 9.337600660747471,
+      "grad_norm": 0.05016907677054405,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 90444
+    },
+    {
+      "epoch": 9.337910386124303,
+      "grad_norm": 0.06592227518558502,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 90447
+    },
+    {
+      "epoch": 9.338220111501135,
+      "grad_norm": 0.06010131537914276,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 90450
+    },
+    {
+      "epoch": 9.338529836877969,
+      "grad_norm": 0.06576816737651825,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 90453
+    },
+    {
+      "epoch": 9.338839562254801,
+      "grad_norm": 0.03570152446627617,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 90456
+    },
+    {
+      "epoch": 9.339149287631633,
+      "grad_norm": 0.036140549927949905,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 90459
+    },
+    {
+      "epoch": 9.339459013008465,
+      "grad_norm": 0.06858191639184952,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 90462
+    },
+    {
+      "epoch": 9.339768738385299,
+      "grad_norm": 0.09158644825220108,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 90465
+    },
+    {
+      "epoch": 9.340078463762131,
+      "grad_norm": 0.11463913321495056,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 90468
+    },
+    {
+      "epoch": 9.340388189138963,
+      "grad_norm": 0.06548649817705154,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 90471
+    },
+    {
+      "epoch": 9.340697914515797,
+      "grad_norm": 0.04369160160422325,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 90474
+    },
+    {
+      "epoch": 9.341007639892629,
+      "grad_norm": 0.058460842818021774,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 90477
+    },
+    {
+      "epoch": 9.341317365269461,
+      "grad_norm": 0.03069923259317875,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 90480
+    },
+    {
+      "epoch": 9.341627090646293,
+      "grad_norm": 0.03194497153162956,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 90483
+    },
+    {
+      "epoch": 9.341936816023127,
+      "grad_norm": 0.12790267169475555,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 90486
+    },
+    {
+      "epoch": 9.342246541399959,
+      "grad_norm": 0.058374807238578796,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 90489
+    },
+    {
+      "epoch": 9.342556266776791,
+      "grad_norm": 0.05721079930663109,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 90492
+    },
+    {
+      "epoch": 9.342865992153623,
+      "grad_norm": 0.060209229588508606,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 90495
+    },
+    {
+      "epoch": 9.343175717530457,
+      "grad_norm": 0.05442395433783531,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 90498
+    },
+    {
+      "epoch": 9.343485442907289,
+      "grad_norm": 0.05215894803404808,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 90501
+    },
+    {
+      "epoch": 9.343795168284121,
+      "grad_norm": 0.10672825574874878,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 90504
+    },
+    {
+      "epoch": 9.344104893660955,
+      "grad_norm": 0.0659959614276886,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 90507
+    },
+    {
+      "epoch": 9.344414619037787,
+      "grad_norm": 0.05211605504155159,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 90510
+    },
+    {
+      "epoch": 9.344724344414619,
+      "grad_norm": 0.05553697422146797,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 90513
+    },
+    {
+      "epoch": 9.345034069791451,
+      "grad_norm": 0.03250126913189888,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 90516
+    },
+    {
+      "epoch": 9.345343795168285,
+      "grad_norm": 0.0480884313583374,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 90519
+    },
+    {
+      "epoch": 9.345653520545117,
+      "grad_norm": 0.09070666134357452,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 90522
+    },
+    {
+      "epoch": 9.345963245921949,
+      "grad_norm": 0.07415267080068588,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 90525
+    },
+    {
+      "epoch": 9.34627297129878,
+      "grad_norm": 0.09671496599912643,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 90528
+    },
+    {
+      "epoch": 9.346582696675615,
+      "grad_norm": 0.11006847023963928,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 90531
+    },
+    {
+      "epoch": 9.346892422052447,
+      "grad_norm": 0.04296498745679855,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 90534
+    },
+    {
+      "epoch": 9.347202147429279,
+      "grad_norm": 0.15365970134735107,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 90537
+    },
+    {
+      "epoch": 9.347511872806113,
+      "grad_norm": 0.09222111850976944,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 90540
+    },
+    {
+      "epoch": 9.347821598182945,
+      "grad_norm": 0.07133835554122925,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 90543
+    },
+    {
+      "epoch": 9.348131323559777,
+      "grad_norm": 0.0642387866973877,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 90546
+    },
+    {
+      "epoch": 9.348441048936609,
+      "grad_norm": 0.040202390402555466,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 90549
+    },
+    {
+      "epoch": 9.348750774313443,
+      "grad_norm": 0.0321086123585701,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 90552
+    },
+    {
+      "epoch": 9.349060499690275,
+      "grad_norm": 0.04373585432767868,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 90555
+    },
+    {
+      "epoch": 9.349370225067107,
+      "grad_norm": 0.0545133501291275,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 90558
+    },
+    {
+      "epoch": 9.34967995044394,
+      "grad_norm": 0.03567107766866684,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 90561
+    },
+    {
+      "epoch": 9.349989675820773,
+      "grad_norm": 0.0539405420422554,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 90564
+    },
+    {
+      "epoch": 9.350299401197605,
+      "grad_norm": 0.13431000709533691,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 90567
+    },
+    {
+      "epoch": 9.350609126574437,
+      "grad_norm": 0.12407714128494263,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 90570
+    },
+    {
+      "epoch": 9.35091885195127,
+      "grad_norm": 0.10649970173835754,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 90573
+    },
+    {
+      "epoch": 9.351228577328103,
+      "grad_norm": 0.04066113010048866,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 90576
+    },
+    {
+      "epoch": 9.351538302704935,
+      "grad_norm": 0.0388568751513958,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 90579
+    },
+    {
+      "epoch": 9.351848028081768,
+      "grad_norm": 0.03337500989437103,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 90582
+    },
+    {
+      "epoch": 9.3521577534586,
+      "grad_norm": 0.04804845526814461,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 90585
+    },
+    {
+      "epoch": 9.352467478835433,
+      "grad_norm": 0.07341484725475311,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 90588
+    },
+    {
+      "epoch": 9.352777204212265,
+      "grad_norm": 0.044978316873311996,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 90591
+    },
+    {
+      "epoch": 9.353086929589098,
+      "grad_norm": 0.10656917840242386,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 90594
+    },
+    {
+      "epoch": 9.35339665496593,
+      "grad_norm": 0.04323936998844147,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 90597
+    },
+    {
+      "epoch": 9.353706380342762,
+      "grad_norm": 0.11398027092218399,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 90600
+    },
+    {
+      "epoch": 9.354016105719595,
+      "grad_norm": 0.09031344950199127,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 90603
+    },
+    {
+      "epoch": 9.354325831096428,
+      "grad_norm": 0.08325620740652084,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 90606
+    },
+    {
+      "epoch": 9.35463555647326,
+      "grad_norm": 0.05016699433326721,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 90609
+    },
+    {
+      "epoch": 9.354945281850092,
+      "grad_norm": 0.039517778903245926,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 90612
+    },
+    {
+      "epoch": 9.355255007226926,
+      "grad_norm": 0.031809695065021515,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 90615
+    },
+    {
+      "epoch": 9.355564732603758,
+      "grad_norm": 0.03748030960559845,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 90618
+    },
+    {
+      "epoch": 9.35587445798059,
+      "grad_norm": 0.04791229963302612,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 90621
+    },
+    {
+      "epoch": 9.356184183357422,
+      "grad_norm": 0.12503692507743835,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 90624
+    },
+    {
+      "epoch": 9.356493908734256,
+      "grad_norm": 0.08294980973005295,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 90627
+    },
+    {
+      "epoch": 9.356803634111088,
+      "grad_norm": 0.04816403612494469,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 90630
+    },
+    {
+      "epoch": 9.35711335948792,
+      "grad_norm": 0.10670629888772964,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 90633
+    },
+    {
+      "epoch": 9.357423084864752,
+      "grad_norm": 0.05907604843378067,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 90636
+    },
+    {
+      "epoch": 9.357732810241586,
+      "grad_norm": 0.03816728666424751,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 90639
+    },
+    {
+      "epoch": 9.358042535618418,
+      "grad_norm": 0.10674931108951569,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 90642
+    },
+    {
+      "epoch": 9.35835226099525,
+      "grad_norm": 0.12507742643356323,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 90645
+    },
+    {
+      "epoch": 9.358661986372084,
+      "grad_norm": 0.14347107708454132,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 90648
+    },
+    {
+      "epoch": 9.358971711748916,
+      "grad_norm": 0.1007409393787384,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 90651
+    },
+    {
+      "epoch": 9.359281437125748,
+      "grad_norm": 0.0874379575252533,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 90654
+    },
+    {
+      "epoch": 9.35959116250258,
+      "grad_norm": 0.06938479095697403,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 90657
+    },
+    {
+      "epoch": 9.359900887879414,
+      "grad_norm": 0.04262612760066986,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 90660
+    },
+    {
+      "epoch": 9.360210613256246,
+      "grad_norm": 0.060341037809848785,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 90663
+    },
+    {
+      "epoch": 9.360520338633078,
+      "grad_norm": 0.04864312708377838,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 90666
+    },
+    {
+      "epoch": 9.360830064009912,
+      "grad_norm": 0.04025941342115402,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 90669
+    },
+    {
+      "epoch": 9.361139789386744,
+      "grad_norm": 0.049005236476659775,
+      "learning_rate": 0.01,
+      "loss": 1.8861,
+      "step": 90672
+    },
+    {
+      "epoch": 9.361449514763576,
+      "grad_norm": 0.04088768735527992,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 90675
+    },
+    {
+      "epoch": 9.361759240140408,
+      "grad_norm": 0.03597968444228172,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 90678
+    },
+    {
+      "epoch": 9.362068965517242,
+      "grad_norm": 0.10069102048873901,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 90681
+    },
+    {
+      "epoch": 9.362378690894074,
+      "grad_norm": 0.07363972812891006,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 90684
+    },
+    {
+      "epoch": 9.362688416270906,
+      "grad_norm": 0.052969079464673996,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 90687
+    },
+    {
+      "epoch": 9.36299814164774,
+      "grad_norm": 0.06485411524772644,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 90690
+    },
+    {
+      "epoch": 9.363307867024572,
+      "grad_norm": 0.08032330125570297,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 90693
+    },
+    {
+      "epoch": 9.363617592401404,
+      "grad_norm": 0.08398518711328506,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 90696
+    },
+    {
+      "epoch": 9.363927317778236,
+      "grad_norm": 0.08669327944517136,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 90699
+    },
+    {
+      "epoch": 9.36423704315507,
+      "grad_norm": 0.04991896077990532,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 90702
+    },
+    {
+      "epoch": 9.364546768531902,
+      "grad_norm": 0.07150976359844208,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 90705
+    },
+    {
+      "epoch": 9.364856493908734,
+      "grad_norm": 0.08911240100860596,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 90708
+    },
+    {
+      "epoch": 9.365166219285566,
+      "grad_norm": 0.03794144466519356,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 90711
+    },
+    {
+      "epoch": 9.3654759446624,
+      "grad_norm": 0.16332855820655823,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 90714
+    },
+    {
+      "epoch": 9.365785670039232,
+      "grad_norm": 0.053398165851831436,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 90717
+    },
+    {
+      "epoch": 9.366095395416064,
+      "grad_norm": 0.10861405730247498,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 90720
+    },
+    {
+      "epoch": 9.366405120792898,
+      "grad_norm": 0.061357125639915466,
+      "learning_rate": 0.01,
+      "loss": 1.8979,
+      "step": 90723
+    },
+    {
+      "epoch": 9.36671484616973,
+      "grad_norm": 0.03730995953083038,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 90726
+    },
+    {
+      "epoch": 9.367024571546562,
+      "grad_norm": 0.057155217975378036,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 90729
+    },
+    {
+      "epoch": 9.367334296923394,
+      "grad_norm": 0.04852285608649254,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 90732
+    },
+    {
+      "epoch": 9.367644022300228,
+      "grad_norm": 0.10618282109498978,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 90735
+    },
+    {
+      "epoch": 9.36795374767706,
+      "grad_norm": 0.0839443951845169,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 90738
+    },
+    {
+      "epoch": 9.368263473053892,
+      "grad_norm": 0.05539103224873543,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 90741
+    },
+    {
+      "epoch": 9.368573198430724,
+      "grad_norm": 0.036319341510534286,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 90744
+    },
+    {
+      "epoch": 9.368882923807558,
+      "grad_norm": 0.033435918390750885,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 90747
+    },
+    {
+      "epoch": 9.36919264918439,
+      "grad_norm": 0.0851496085524559,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 90750
+    },
+    {
+      "epoch": 9.369502374561222,
+      "grad_norm": 0.037316080182790756,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 90753
+    },
+    {
+      "epoch": 9.369812099938056,
+      "grad_norm": 0.13374057412147522,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 90756
+    },
+    {
+      "epoch": 9.370121825314888,
+      "grad_norm": 0.12201882153749466,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 90759
+    },
+    {
+      "epoch": 9.37043155069172,
+      "grad_norm": 0.07799084484577179,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 90762
+    },
+    {
+      "epoch": 9.370741276068552,
+      "grad_norm": 0.044005535542964935,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 90765
+    },
+    {
+      "epoch": 9.371051001445386,
+      "grad_norm": 0.0691835880279541,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 90768
+    },
+    {
+      "epoch": 9.371360726822218,
+      "grad_norm": 0.041414376348257065,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 90771
+    },
+    {
+      "epoch": 9.37167045219905,
+      "grad_norm": 0.05714799463748932,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 90774
+    },
+    {
+      "epoch": 9.371980177575884,
+      "grad_norm": 0.0594872385263443,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 90777
+    },
+    {
+      "epoch": 9.372289902952716,
+      "grad_norm": 0.044780146330595016,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 90780
+    },
+    {
+      "epoch": 9.372599628329548,
+      "grad_norm": 0.039911285042762756,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 90783
+    },
+    {
+      "epoch": 9.37290935370638,
+      "grad_norm": 0.09436476230621338,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 90786
+    },
+    {
+      "epoch": 9.373219079083214,
+      "grad_norm": 0.0415303073823452,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 90789
+    },
+    {
+      "epoch": 9.373528804460046,
+      "grad_norm": 0.03091426193714142,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 90792
+    },
+    {
+      "epoch": 9.373838529836878,
+      "grad_norm": 0.04031349718570709,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 90795
+    },
+    {
+      "epoch": 9.374148255213711,
+      "grad_norm": 0.05020138993859291,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 90798
+    },
+    {
+      "epoch": 9.374457980590543,
+      "grad_norm": 0.07621460407972336,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 90801
+    },
+    {
+      "epoch": 9.374767705967376,
+      "grad_norm": 0.049681663513183594,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 90804
+    },
+    {
+      "epoch": 9.375077431344208,
+      "grad_norm": 0.058287203311920166,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 90807
+    },
+    {
+      "epoch": 9.375387156721041,
+      "grad_norm": 0.11261413246393204,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 90810
+    },
+    {
+      "epoch": 9.375696882097873,
+      "grad_norm": 0.13443779945373535,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 90813
+    },
+    {
+      "epoch": 9.376006607474705,
+      "grad_norm": 0.10898131877183914,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 90816
+    },
+    {
+      "epoch": 9.376316332851538,
+      "grad_norm": 0.06500370800495148,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 90819
+    },
+    {
+      "epoch": 9.376626058228371,
+      "grad_norm": 0.07922185212373734,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 90822
+    },
+    {
+      "epoch": 9.376935783605203,
+      "grad_norm": 0.05019281432032585,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 90825
+    },
+    {
+      "epoch": 9.377245508982035,
+      "grad_norm": 0.049235813319683075,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 90828
+    },
+    {
+      "epoch": 9.37755523435887,
+      "grad_norm": 0.04269546642899513,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 90831
+    },
+    {
+      "epoch": 9.377864959735701,
+      "grad_norm": 0.06699442118406296,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 90834
+    },
+    {
+      "epoch": 9.378174685112533,
+      "grad_norm": 0.034419186413288116,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 90837
+    },
+    {
+      "epoch": 9.378484410489365,
+      "grad_norm": 0.11870981007814407,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 90840
+    },
+    {
+      "epoch": 9.3787941358662,
+      "grad_norm": 0.058509461581707,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 90843
+    },
+    {
+      "epoch": 9.379103861243031,
+      "grad_norm": 0.12032335996627808,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 90846
+    },
+    {
+      "epoch": 9.379413586619863,
+      "grad_norm": 0.05439126119017601,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 90849
+    },
+    {
+      "epoch": 9.379723311996695,
+      "grad_norm": 0.03926823288202286,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 90852
+    },
+    {
+      "epoch": 9.38003303737353,
+      "grad_norm": 0.03611249104142189,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 90855
+    },
+    {
+      "epoch": 9.380342762750361,
+      "grad_norm": 0.045519620180130005,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 90858
+    },
+    {
+      "epoch": 9.380652488127193,
+      "grad_norm": 0.05714515969157219,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 90861
+    },
+    {
+      "epoch": 9.380962213504027,
+      "grad_norm": 0.06265503913164139,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 90864
+    },
+    {
+      "epoch": 9.38127193888086,
+      "grad_norm": 0.15718643367290497,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 90867
+    },
+    {
+      "epoch": 9.381581664257691,
+      "grad_norm": 0.13615955412387848,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 90870
+    },
+    {
+      "epoch": 9.381891389634523,
+      "grad_norm": 0.09134041517972946,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 90873
+    },
+    {
+      "epoch": 9.382201115011357,
+      "grad_norm": 0.09111947566270828,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 90876
+    },
+    {
+      "epoch": 9.38251084038819,
+      "grad_norm": 0.050355955958366394,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 90879
+    },
+    {
+      "epoch": 9.382820565765021,
+      "grad_norm": 0.03346274793148041,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 90882
+    },
+    {
+      "epoch": 9.383130291141855,
+      "grad_norm": 0.10060890763998032,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 90885
+    },
+    {
+      "epoch": 9.383440016518687,
+      "grad_norm": 0.07907859981060028,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 90888
+    },
+    {
+      "epoch": 9.38374974189552,
+      "grad_norm": 0.06177898123860359,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 90891
+    },
+    {
+      "epoch": 9.384059467272351,
+      "grad_norm": 0.033914852887392044,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 90894
+    },
+    {
+      "epoch": 9.384369192649185,
+      "grad_norm": 0.03818202018737793,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 90897
+    },
+    {
+      "epoch": 9.384678918026017,
+      "grad_norm": 0.09121721982955933,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 90900
+    },
+    {
+      "epoch": 9.384988643402849,
+      "grad_norm": 0.089520663022995,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 90903
+    },
+    {
+      "epoch": 9.385298368779681,
+      "grad_norm": 0.06648974865674973,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 90906
+    },
+    {
+      "epoch": 9.385608094156515,
+      "grad_norm": 0.07574710994958878,
+      "learning_rate": 0.01,
+      "loss": 1.8943,
+      "step": 90909
+    },
+    {
+      "epoch": 9.385917819533347,
+      "grad_norm": 0.046774134039878845,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 90912
+    },
+    {
+      "epoch": 9.386227544910179,
+      "grad_norm": 0.03651203215122223,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 90915
+    },
+    {
+      "epoch": 9.386537270287013,
+      "grad_norm": 0.07176890969276428,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 90918
+    },
+    {
+      "epoch": 9.386846995663845,
+      "grad_norm": 0.10504325479269028,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 90921
+    },
+    {
+      "epoch": 9.387156721040677,
+      "grad_norm": 0.04525873437523842,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 90924
+    },
+    {
+      "epoch": 9.387466446417509,
+      "grad_norm": 0.03573272004723549,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 90927
+    },
+    {
+      "epoch": 9.387776171794343,
+      "grad_norm": 0.03916339576244354,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 90930
+    },
+    {
+      "epoch": 9.388085897171175,
+      "grad_norm": 0.05404656380414963,
+      "learning_rate": 0.01,
+      "loss": 1.9022,
+      "step": 90933
+    },
+    {
+      "epoch": 9.388395622548007,
+      "grad_norm": 0.12576231360435486,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 90936
+    },
+    {
+      "epoch": 9.38870534792484,
+      "grad_norm": 0.048078350722789764,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 90939
+    },
+    {
+      "epoch": 9.389015073301673,
+      "grad_norm": 0.102527916431427,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 90942
+    },
+    {
+      "epoch": 9.389324798678505,
+      "grad_norm": 0.058806050568819046,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 90945
+    },
+    {
+      "epoch": 9.389634524055337,
+      "grad_norm": 0.07547862827777863,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 90948
+    },
+    {
+      "epoch": 9.38994424943217,
+      "grad_norm": 0.08734951913356781,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 90951
+    },
+    {
+      "epoch": 9.390253974809003,
+      "grad_norm": 0.08816061913967133,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 90954
+    },
+    {
+      "epoch": 9.390563700185835,
+      "grad_norm": 0.05598800256848335,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 90957
+    },
+    {
+      "epoch": 9.390873425562667,
+      "grad_norm": 0.06532081216573715,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 90960
+    },
+    {
+      "epoch": 9.3911831509395,
+      "grad_norm": 0.07077542692422867,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 90963
+    },
+    {
+      "epoch": 9.391492876316333,
+      "grad_norm": 0.05597352236509323,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 90966
+    },
+    {
+      "epoch": 9.391802601693165,
+      "grad_norm": 0.07413847744464874,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 90969
+    },
+    {
+      "epoch": 9.392112327069999,
+      "grad_norm": 0.037031251937150955,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 90972
+    },
+    {
+      "epoch": 9.39242205244683,
+      "grad_norm": 0.06918266415596008,
+      "learning_rate": 0.01,
+      "loss": 1.9733,
+      "step": 90975
+    },
+    {
+      "epoch": 9.392731777823663,
+      "grad_norm": 0.0500267818570137,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 90978
+    },
+    {
+      "epoch": 9.393041503200495,
+      "grad_norm": 0.03715291619300842,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 90981
+    },
+    {
+      "epoch": 9.393351228577329,
+      "grad_norm": 0.050200916826725006,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 90984
+    },
+    {
+      "epoch": 9.39366095395416,
+      "grad_norm": 0.08894219994544983,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 90987
+    },
+    {
+      "epoch": 9.393970679330993,
+      "grad_norm": 0.08678343892097473,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 90990
+    },
+    {
+      "epoch": 9.394280404707827,
+      "grad_norm": 0.09989691525697708,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 90993
+    },
+    {
+      "epoch": 9.394590130084659,
+      "grad_norm": 0.12368086725473404,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 90996
+    },
+    {
+      "epoch": 9.39489985546149,
+      "grad_norm": 0.04987645894289017,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 90999
+    },
+    {
+      "epoch": 9.395209580838323,
+      "grad_norm": 0.03483951464295387,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 91002
+    },
+    {
+      "epoch": 9.395519306215157,
+      "grad_norm": 0.044155314564704895,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 91005
+    },
+    {
+      "epoch": 9.395829031591989,
+      "grad_norm": 0.04584622010588646,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 91008
+    },
+    {
+      "epoch": 9.39613875696882,
+      "grad_norm": 0.04356890171766281,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 91011
+    },
+    {
+      "epoch": 9.396448482345653,
+      "grad_norm": 0.03459366783499718,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 91014
+    },
+    {
+      "epoch": 9.396758207722486,
+      "grad_norm": 0.0733804851770401,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 91017
+    },
+    {
+      "epoch": 9.397067933099319,
+      "grad_norm": 0.10501904040575027,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 91020
+    },
+    {
+      "epoch": 9.39737765847615,
+      "grad_norm": 0.08252866566181183,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 91023
+    },
+    {
+      "epoch": 9.397687383852984,
+      "grad_norm": 0.11142603307962418,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 91026
+    },
+    {
+      "epoch": 9.397997109229816,
+      "grad_norm": 0.06776826083660126,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 91029
+    },
+    {
+      "epoch": 9.398306834606649,
+      "grad_norm": 0.055962324142456055,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 91032
+    },
+    {
+      "epoch": 9.39861655998348,
+      "grad_norm": 0.10423536598682404,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 91035
+    },
+    {
+      "epoch": 9.398926285360314,
+      "grad_norm": 0.03981839492917061,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 91038
+    },
+    {
+      "epoch": 9.399236010737146,
+      "grad_norm": 0.06059722229838371,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 91041
+    },
+    {
+      "epoch": 9.399545736113978,
+      "grad_norm": 0.10510721802711487,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 91044
+    },
+    {
+      "epoch": 9.399855461490812,
+      "grad_norm": 0.07276123762130737,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 91047
+    },
+    {
+      "epoch": 9.400165186867644,
+      "grad_norm": 0.06601347774267197,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 91050
+    },
+    {
+      "epoch": 9.400474912244476,
+      "grad_norm": 0.11832351237535477,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 91053
+    },
+    {
+      "epoch": 9.400784637621308,
+      "grad_norm": 0.056920114904642105,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 91056
+    },
+    {
+      "epoch": 9.401094362998142,
+      "grad_norm": 0.03529761731624603,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 91059
+    },
+    {
+      "epoch": 9.401404088374974,
+      "grad_norm": 0.04953724890947342,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 91062
+    },
+    {
+      "epoch": 9.401713813751806,
+      "grad_norm": 0.11988922953605652,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 91065
+    },
+    {
+      "epoch": 9.402023539128638,
+      "grad_norm": 0.06321597844362259,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 91068
+    },
+    {
+      "epoch": 9.402333264505472,
+      "grad_norm": 0.05112435296177864,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 91071
+    },
+    {
+      "epoch": 9.402642989882304,
+      "grad_norm": 0.12304303050041199,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 91074
+    },
+    {
+      "epoch": 9.402952715259136,
+      "grad_norm": 0.038611236959695816,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 91077
+    },
+    {
+      "epoch": 9.40326244063597,
+      "grad_norm": 0.04994663596153259,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 91080
+    },
+    {
+      "epoch": 9.403572166012802,
+      "grad_norm": 0.034984659403562546,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 91083
+    },
+    {
+      "epoch": 9.403881891389634,
+      "grad_norm": 0.034549251198768616,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 91086
+    },
+    {
+      "epoch": 9.404191616766466,
+      "grad_norm": 0.09396814554929733,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 91089
+    },
+    {
+      "epoch": 9.4045013421433,
+      "grad_norm": 0.05054985359311104,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 91092
+    },
+    {
+      "epoch": 9.404811067520132,
+      "grad_norm": 0.0785905048251152,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 91095
+    },
+    {
+      "epoch": 9.405120792896964,
+      "grad_norm": 0.0745377391576767,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 91098
+    },
+    {
+      "epoch": 9.405430518273798,
+      "grad_norm": 0.05711502581834793,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 91101
+    },
+    {
+      "epoch": 9.40574024365063,
+      "grad_norm": 0.13965703547000885,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 91104
+    },
+    {
+      "epoch": 9.406049969027462,
+      "grad_norm": 0.09781201183795929,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 91107
+    },
+    {
+      "epoch": 9.406359694404294,
+      "grad_norm": 0.04011009633541107,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 91110
+    },
+    {
+      "epoch": 9.406669419781128,
+      "grad_norm": 0.06505788117647171,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 91113
+    },
+    {
+      "epoch": 9.40697914515796,
+      "grad_norm": 0.08113677054643631,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 91116
+    },
+    {
+      "epoch": 9.407288870534792,
+      "grad_norm": 0.12400487810373306,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 91119
+    },
+    {
+      "epoch": 9.407598595911624,
+      "grad_norm": 0.08019889891147614,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 91122
+    },
+    {
+      "epoch": 9.407908321288458,
+      "grad_norm": 0.04559165611863136,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 91125
+    },
+    {
+      "epoch": 9.40821804666529,
+      "grad_norm": 0.10827832669019699,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 91128
+    },
+    {
+      "epoch": 9.408527772042122,
+      "grad_norm": 0.04035136103630066,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 91131
+    },
+    {
+      "epoch": 9.408837497418956,
+      "grad_norm": 0.03153857961297035,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 91134
+    },
+    {
+      "epoch": 9.409147222795788,
+      "grad_norm": 0.07644420862197876,
+      "learning_rate": 0.01,
+      "loss": 1.8987,
+      "step": 91137
+    },
+    {
+      "epoch": 9.40945694817262,
+      "grad_norm": 0.10326854139566422,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 91140
+    },
+    {
+      "epoch": 9.409766673549452,
+      "grad_norm": 0.10630738735198975,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 91143
+    },
+    {
+      "epoch": 9.410076398926286,
+      "grad_norm": 0.10723258554935455,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 91146
+    },
+    {
+      "epoch": 9.410386124303118,
+      "grad_norm": 0.07178448140621185,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 91149
+    },
+    {
+      "epoch": 9.41069584967995,
+      "grad_norm": 0.039453644305467606,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 91152
+    },
+    {
+      "epoch": 9.411005575056784,
+      "grad_norm": 0.04125725105404854,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 91155
+    },
+    {
+      "epoch": 9.411315300433616,
+      "grad_norm": 0.07123439013957977,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 91158
+    },
+    {
+      "epoch": 9.411625025810448,
+      "grad_norm": 0.07479481399059296,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 91161
+    },
+    {
+      "epoch": 9.41193475118728,
+      "grad_norm": 0.062062572687864304,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 91164
+    },
+    {
+      "epoch": 9.412244476564114,
+      "grad_norm": 0.09150844812393188,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 91167
+    },
+    {
+      "epoch": 9.412554201940946,
+      "grad_norm": 0.06751282513141632,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 91170
+    },
+    {
+      "epoch": 9.412863927317778,
+      "grad_norm": 0.09919881075620651,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 91173
+    },
+    {
+      "epoch": 9.41317365269461,
+      "grad_norm": 0.11017844080924988,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 91176
+    },
+    {
+      "epoch": 9.413483378071444,
+      "grad_norm": 0.07611265033483505,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 91179
+    },
+    {
+      "epoch": 9.413793103448276,
+      "grad_norm": 0.10450519621372223,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 91182
+    },
+    {
+      "epoch": 9.414102828825108,
+      "grad_norm": 0.05414097383618355,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 91185
+    },
+    {
+      "epoch": 9.414412554201942,
+      "grad_norm": 0.04041386395692825,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 91188
+    },
+    {
+      "epoch": 9.414722279578774,
+      "grad_norm": 0.0727272480726242,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 91191
+    },
+    {
+      "epoch": 9.415032004955606,
+      "grad_norm": 0.06471801549196243,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 91194
+    },
+    {
+      "epoch": 9.415341730332438,
+      "grad_norm": 0.04700074717402458,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 91197
+    },
+    {
+      "epoch": 9.415651455709272,
+      "grad_norm": 0.04389588534832001,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 91200
+    },
+    {
+      "epoch": 9.415961181086104,
+      "grad_norm": 0.08153852075338364,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 91203
+    },
+    {
+      "epoch": 9.416270906462936,
+      "grad_norm": 0.048129767179489136,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 91206
+    },
+    {
+      "epoch": 9.41658063183977,
+      "grad_norm": 0.057881567627191544,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 91209
+    },
+    {
+      "epoch": 9.416890357216602,
+      "grad_norm": 0.040281008929014206,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 91212
+    },
+    {
+      "epoch": 9.417200082593434,
+      "grad_norm": 0.06497588753700256,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 91215
+    },
+    {
+      "epoch": 9.417509807970266,
+      "grad_norm": 0.045624736696481705,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 91218
+    },
+    {
+      "epoch": 9.4178195333471,
+      "grad_norm": 0.037451039999723434,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 91221
+    },
+    {
+      "epoch": 9.418129258723932,
+      "grad_norm": 0.13164624571800232,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 91224
+    },
+    {
+      "epoch": 9.418438984100764,
+      "grad_norm": 0.09604440629482269,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 91227
+    },
+    {
+      "epoch": 9.418748709477596,
+      "grad_norm": 0.07141084223985672,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 91230
+    },
+    {
+      "epoch": 9.41905843485443,
+      "grad_norm": 0.11629889160394669,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 91233
+    },
+    {
+      "epoch": 9.419368160231262,
+      "grad_norm": 0.07452220469713211,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 91236
+    },
+    {
+      "epoch": 9.419677885608094,
+      "grad_norm": 0.04817759245634079,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 91239
+    },
+    {
+      "epoch": 9.419987610984927,
+      "grad_norm": 0.053956806659698486,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 91242
+    },
+    {
+      "epoch": 9.42029733636176,
+      "grad_norm": 0.053349558264017105,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 91245
+    },
+    {
+      "epoch": 9.420607061738592,
+      "grad_norm": 0.03342839330434799,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 91248
+    },
+    {
+      "epoch": 9.420916787115424,
+      "grad_norm": 0.07104140520095825,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 91251
+    },
+    {
+      "epoch": 9.421226512492257,
+      "grad_norm": 0.10545811057090759,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 91254
+    },
+    {
+      "epoch": 9.42153623786909,
+      "grad_norm": 0.07372496277093887,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 91257
+    },
+    {
+      "epoch": 9.421845963245921,
+      "grad_norm": 0.06759019941091537,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 91260
+    },
+    {
+      "epoch": 9.422155688622755,
+      "grad_norm": 0.08602011948823929,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 91263
+    },
+    {
+      "epoch": 9.422465413999587,
+      "grad_norm": 0.04749952629208565,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 91266
+    },
+    {
+      "epoch": 9.42277513937642,
+      "grad_norm": 0.04223652184009552,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 91269
+    },
+    {
+      "epoch": 9.423084864753251,
+      "grad_norm": 0.08851899951696396,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 91272
+    },
+    {
+      "epoch": 9.423394590130085,
+      "grad_norm": 0.04346892237663269,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 91275
+    },
+    {
+      "epoch": 9.423704315506917,
+      "grad_norm": 0.042326390743255615,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 91278
+    },
+    {
+      "epoch": 9.42401404088375,
+      "grad_norm": 0.10447720438241959,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 91281
+    },
+    {
+      "epoch": 9.424323766260581,
+      "grad_norm": 0.04779525101184845,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 91284
+    },
+    {
+      "epoch": 9.424633491637415,
+      "grad_norm": 0.05690819025039673,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 91287
+    },
+    {
+      "epoch": 9.424943217014247,
+      "grad_norm": 0.05316159501671791,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 91290
+    },
+    {
+      "epoch": 9.42525294239108,
+      "grad_norm": 0.09380586445331573,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 91293
+    },
+    {
+      "epoch": 9.425562667767913,
+      "grad_norm": 0.04061369597911835,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 91296
+    },
+    {
+      "epoch": 9.425872393144745,
+      "grad_norm": 0.052185557782649994,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 91299
+    },
+    {
+      "epoch": 9.426182118521577,
+      "grad_norm": 0.05126102641224861,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 91302
+    },
+    {
+      "epoch": 9.42649184389841,
+      "grad_norm": 0.037692565470933914,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 91305
+    },
+    {
+      "epoch": 9.426801569275243,
+      "grad_norm": 0.03374462202191353,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 91308
+    },
+    {
+      "epoch": 9.427111294652075,
+      "grad_norm": 0.03585955500602722,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 91311
+    },
+    {
+      "epoch": 9.427421020028907,
+      "grad_norm": 0.10094792395830154,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 91314
+    },
+    {
+      "epoch": 9.427730745405741,
+      "grad_norm": 0.10577898472547531,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 91317
+    },
+    {
+      "epoch": 9.428040470782573,
+      "grad_norm": 0.09183473140001297,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 91320
+    },
+    {
+      "epoch": 9.428350196159405,
+      "grad_norm": 0.05396624654531479,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 91323
+    },
+    {
+      "epoch": 9.428659921536237,
+      "grad_norm": 0.035900574177503586,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 91326
+    },
+    {
+      "epoch": 9.428969646913071,
+      "grad_norm": 0.07058069109916687,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 91329
+    },
+    {
+      "epoch": 9.429279372289903,
+      "grad_norm": 0.11194363981485367,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 91332
+    },
+    {
+      "epoch": 9.429589097666735,
+      "grad_norm": 0.06465640664100647,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 91335
+    },
+    {
+      "epoch": 9.429898823043567,
+      "grad_norm": 0.09346874803304672,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 91338
+    },
+    {
+      "epoch": 9.430208548420401,
+      "grad_norm": 0.05075174570083618,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 91341
+    },
+    {
+      "epoch": 9.430518273797233,
+      "grad_norm": 0.04691362753510475,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 91344
+    },
+    {
+      "epoch": 9.430827999174065,
+      "grad_norm": 0.05307089909911156,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 91347
+    },
+    {
+      "epoch": 9.431137724550899,
+      "grad_norm": 0.04385623335838318,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 91350
+    },
+    {
+      "epoch": 9.431447449927731,
+      "grad_norm": 0.14806324243545532,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 91353
+    },
+    {
+      "epoch": 9.431757175304563,
+      "grad_norm": 0.07075516134500504,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 91356
+    },
+    {
+      "epoch": 9.432066900681395,
+      "grad_norm": 0.03600502386689186,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 91359
+    },
+    {
+      "epoch": 9.432376626058229,
+      "grad_norm": 0.05355400964617729,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 91362
+    },
+    {
+      "epoch": 9.432686351435061,
+      "grad_norm": 0.0357879213988781,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 91365
+    },
+    {
+      "epoch": 9.432996076811893,
+      "grad_norm": 0.03480469807982445,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 91368
+    },
+    {
+      "epoch": 9.433305802188727,
+      "grad_norm": 0.07176405936479568,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 91371
+    },
+    {
+      "epoch": 9.433615527565559,
+      "grad_norm": 0.06469430774450302,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 91374
+    },
+    {
+      "epoch": 9.433925252942391,
+      "grad_norm": 0.06961306929588318,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 91377
+    },
+    {
+      "epoch": 9.434234978319223,
+      "grad_norm": 0.04767259955406189,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 91380
+    },
+    {
+      "epoch": 9.434544703696057,
+      "grad_norm": 0.03745473176240921,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 91383
+    },
+    {
+      "epoch": 9.434854429072889,
+      "grad_norm": 0.054868750274181366,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 91386
+    },
+    {
+      "epoch": 9.435164154449721,
+      "grad_norm": 0.12046508491039276,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 91389
+    },
+    {
+      "epoch": 9.435473879826553,
+      "grad_norm": 0.1015070453286171,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 91392
+    },
+    {
+      "epoch": 9.435783605203387,
+      "grad_norm": 0.04544771835207939,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 91395
+    },
+    {
+      "epoch": 9.436093330580219,
+      "grad_norm": 0.04002662003040314,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 91398
+    },
+    {
+      "epoch": 9.43640305595705,
+      "grad_norm": 0.04304361715912819,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 91401
+    },
+    {
+      "epoch": 9.436712781333885,
+      "grad_norm": 0.03594297543168068,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 91404
+    },
+    {
+      "epoch": 9.437022506710717,
+      "grad_norm": 0.05000785365700722,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 91407
+    },
+    {
+      "epoch": 9.437332232087549,
+      "grad_norm": 0.06557739526033401,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 91410
+    },
+    {
+      "epoch": 9.43764195746438,
+      "grad_norm": 0.041119322180747986,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 91413
+    },
+    {
+      "epoch": 9.437951682841215,
+      "grad_norm": 0.054089631885290146,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 91416
+    },
+    {
+      "epoch": 9.438261408218047,
+      "grad_norm": 0.05677805095911026,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 91419
+    },
+    {
+      "epoch": 9.438571133594879,
+      "grad_norm": 0.032810989767313004,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 91422
+    },
+    {
+      "epoch": 9.438880858971713,
+      "grad_norm": 0.12531718611717224,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 91425
+    },
+    {
+      "epoch": 9.439190584348545,
+      "grad_norm": 0.10076866298913956,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 91428
+    },
+    {
+      "epoch": 9.439500309725377,
+      "grad_norm": 0.07036837190389633,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 91431
+    },
+    {
+      "epoch": 9.439810035102209,
+      "grad_norm": 0.10938290506601334,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 91434
+    },
+    {
+      "epoch": 9.440119760479043,
+      "grad_norm": 0.10378368198871613,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 91437
+    },
+    {
+      "epoch": 9.440429485855875,
+      "grad_norm": 0.07203853875398636,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 91440
+    },
+    {
+      "epoch": 9.440739211232707,
+      "grad_norm": 0.04642992094159126,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 91443
+    },
+    {
+      "epoch": 9.441048936609539,
+      "grad_norm": 0.03274267911911011,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 91446
+    },
+    {
+      "epoch": 9.441358661986373,
+      "grad_norm": 0.03794633969664574,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 91449
+    },
+    {
+      "epoch": 9.441668387363205,
+      "grad_norm": 0.045492980629205704,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 91452
+    },
+    {
+      "epoch": 9.441978112740037,
+      "grad_norm": 0.09170827269554138,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 91455
+    },
+    {
+      "epoch": 9.44228783811687,
+      "grad_norm": 0.07204066962003708,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 91458
+    },
+    {
+      "epoch": 9.442597563493702,
+      "grad_norm": 0.07380224019289017,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 91461
+    },
+    {
+      "epoch": 9.442907288870535,
+      "grad_norm": 0.05300910398364067,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 91464
+    },
+    {
+      "epoch": 9.443217014247367,
+      "grad_norm": 0.046401552855968475,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 91467
+    },
+    {
+      "epoch": 9.4435267396242,
+      "grad_norm": 0.039238277822732925,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 91470
+    },
+    {
+      "epoch": 9.443836465001032,
+      "grad_norm": 0.13839541375637054,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 91473
+    },
+    {
+      "epoch": 9.444146190377865,
+      "grad_norm": 0.06052450090646744,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 91476
+    },
+    {
+      "epoch": 9.444455915754698,
+      "grad_norm": 0.05661856010556221,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 91479
+    },
+    {
+      "epoch": 9.44476564113153,
+      "grad_norm": 0.08223719894886017,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 91482
+    },
+    {
+      "epoch": 9.445075366508362,
+      "grad_norm": 0.04421024024486542,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 91485
+    },
+    {
+      "epoch": 9.445385091885194,
+      "grad_norm": 0.0505010187625885,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 91488
+    },
+    {
+      "epoch": 9.445694817262028,
+      "grad_norm": 0.04484516754746437,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 91491
+    },
+    {
+      "epoch": 9.44600454263886,
+      "grad_norm": 0.05021611973643303,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 91494
+    },
+    {
+      "epoch": 9.446314268015692,
+      "grad_norm": 0.03842268884181976,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 91497
+    },
+    {
+      "epoch": 9.446623993392524,
+      "grad_norm": 0.12103386968374252,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 91500
+    },
+    {
+      "epoch": 9.446933718769358,
+      "grad_norm": 0.054335884749889374,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 91503
+    },
+    {
+      "epoch": 9.44724344414619,
+      "grad_norm": 0.05245980620384216,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 91506
+    },
+    {
+      "epoch": 9.447553169523022,
+      "grad_norm": 0.05430982634425163,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 91509
+    },
+    {
+      "epoch": 9.447862894899856,
+      "grad_norm": 0.04710724577307701,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 91512
+    },
+    {
+      "epoch": 9.448172620276688,
+      "grad_norm": 0.03400622680783272,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 91515
+    },
+    {
+      "epoch": 9.44848234565352,
+      "grad_norm": 0.05350743234157562,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 91518
+    },
+    {
+      "epoch": 9.448792071030352,
+      "grad_norm": 0.05976387858390808,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 91521
+    },
+    {
+      "epoch": 9.449101796407186,
+      "grad_norm": 0.07914312928915024,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 91524
+    },
+    {
+      "epoch": 9.449411521784018,
+      "grad_norm": 0.07157033681869507,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 91527
+    },
+    {
+      "epoch": 9.44972124716085,
+      "grad_norm": 0.05218801274895668,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 91530
+    },
+    {
+      "epoch": 9.450030972537684,
+      "grad_norm": 0.04557066783308983,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 91533
+    },
+    {
+      "epoch": 9.450340697914516,
+      "grad_norm": 0.05901813507080078,
+      "learning_rate": 0.01,
+      "loss": 1.8917,
+      "step": 91536
+    },
+    {
+      "epoch": 9.450650423291348,
+      "grad_norm": 0.10986040532588959,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 91539
+    },
+    {
+      "epoch": 9.45096014866818,
+      "grad_norm": 0.03527398407459259,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 91542
+    },
+    {
+      "epoch": 9.451269874045014,
+      "grad_norm": 0.12370608001947403,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 91545
+    },
+    {
+      "epoch": 9.451579599421846,
+      "grad_norm": 0.06623967736959457,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 91548
+    },
+    {
+      "epoch": 9.451889324798678,
+      "grad_norm": 0.04259668290615082,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 91551
+    },
+    {
+      "epoch": 9.45219905017551,
+      "grad_norm": 0.04617312550544739,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 91554
+    },
+    {
+      "epoch": 9.452508775552344,
+      "grad_norm": 0.044426169246435165,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 91557
+    },
+    {
+      "epoch": 9.452818500929176,
+      "grad_norm": 0.05939152091741562,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 91560
+    },
+    {
+      "epoch": 9.453128226306008,
+      "grad_norm": 0.05148043483495712,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 91563
+    },
+    {
+      "epoch": 9.453437951682842,
+      "grad_norm": 0.051897548139095306,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 91566
+    },
+    {
+      "epoch": 9.453747677059674,
+      "grad_norm": 0.04463929310441017,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 91569
+    },
+    {
+      "epoch": 9.454057402436506,
+      "grad_norm": 0.058007046580314636,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 91572
+    },
+    {
+      "epoch": 9.454367127813338,
+      "grad_norm": 0.10628005117177963,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 91575
+    },
+    {
+      "epoch": 9.454676853190172,
+      "grad_norm": 0.15172232687473297,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 91578
+    },
+    {
+      "epoch": 9.454986578567004,
+      "grad_norm": 0.11548087745904922,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 91581
+    },
+    {
+      "epoch": 9.455296303943836,
+      "grad_norm": 0.05106515437364578,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 91584
+    },
+    {
+      "epoch": 9.45560602932067,
+      "grad_norm": 0.052183981984853745,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 91587
+    },
+    {
+      "epoch": 9.455915754697502,
+      "grad_norm": 0.04708034172654152,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 91590
+    },
+    {
+      "epoch": 9.456225480074334,
+      "grad_norm": 0.04493347555398941,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 91593
+    },
+    {
+      "epoch": 9.456535205451166,
+      "grad_norm": 0.05144806206226349,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 91596
+    },
+    {
+      "epoch": 9.456844930828,
+      "grad_norm": 0.10107263922691345,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 91599
+    },
+    {
+      "epoch": 9.457154656204832,
+      "grad_norm": 0.06296777725219727,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 91602
+    },
+    {
+      "epoch": 9.457464381581664,
+      "grad_norm": 0.08265284448862076,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 91605
+    },
+    {
+      "epoch": 9.457774106958496,
+      "grad_norm": 0.0472281277179718,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 91608
+    },
+    {
+      "epoch": 9.45808383233533,
+      "grad_norm": 0.0438406802713871,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 91611
+    },
+    {
+      "epoch": 9.458393557712162,
+      "grad_norm": 0.03896843641996384,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 91614
+    },
+    {
+      "epoch": 9.458703283088994,
+      "grad_norm": 0.08807415515184402,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 91617
+    },
+    {
+      "epoch": 9.459013008465828,
+      "grad_norm": 0.11590650677680969,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 91620
+    },
+    {
+      "epoch": 9.45932273384266,
+      "grad_norm": 0.06758120656013489,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 91623
+    },
+    {
+      "epoch": 9.459632459219492,
+      "grad_norm": 0.05359683930873871,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 91626
+    },
+    {
+      "epoch": 9.459942184596324,
+      "grad_norm": 0.08736684173345566,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 91629
+    },
+    {
+      "epoch": 9.460251909973158,
+      "grad_norm": 0.06551837921142578,
+      "learning_rate": 0.01,
+      "loss": 1.8991,
+      "step": 91632
+    },
+    {
+      "epoch": 9.46056163534999,
+      "grad_norm": 0.09359174221754074,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 91635
+    },
+    {
+      "epoch": 9.460871360726822,
+      "grad_norm": 0.0556727759540081,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 91638
+    },
+    {
+      "epoch": 9.461181086103656,
+      "grad_norm": 0.04954926669597626,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 91641
+    },
+    {
+      "epoch": 9.461490811480488,
+      "grad_norm": 0.03850363940000534,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 91644
+    },
+    {
+      "epoch": 9.46180053685732,
+      "grad_norm": 0.11323880404233932,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 91647
+    },
+    {
+      "epoch": 9.462110262234152,
+      "grad_norm": 0.1314762681722641,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 91650
+    },
+    {
+      "epoch": 9.462419987610986,
+      "grad_norm": 0.049930673092603683,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 91653
+    },
+    {
+      "epoch": 9.462729712987818,
+      "grad_norm": 0.04937874525785446,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 91656
+    },
+    {
+      "epoch": 9.46303943836465,
+      "grad_norm": 0.03621145337820053,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 91659
+    },
+    {
+      "epoch": 9.463349163741482,
+      "grad_norm": 0.036950428038835526,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 91662
+    },
+    {
+      "epoch": 9.463658889118316,
+      "grad_norm": 0.06989537179470062,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 91665
+    },
+    {
+      "epoch": 9.463968614495148,
+      "grad_norm": 0.06468947231769562,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 91668
+    },
+    {
+      "epoch": 9.46427833987198,
+      "grad_norm": 0.07130585610866547,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 91671
+    },
+    {
+      "epoch": 9.464588065248813,
+      "grad_norm": 0.04903630167245865,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 91674
+    },
+    {
+      "epoch": 9.464897790625646,
+      "grad_norm": 0.0678560733795166,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 91677
+    },
+    {
+      "epoch": 9.465207516002478,
+      "grad_norm": 0.11219093948602676,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 91680
+    },
+    {
+      "epoch": 9.46551724137931,
+      "grad_norm": 0.0979389101266861,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 91683
+    },
+    {
+      "epoch": 9.465826966756143,
+      "grad_norm": 0.07034581154584885,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 91686
+    },
+    {
+      "epoch": 9.466136692132975,
+      "grad_norm": 0.06349025666713715,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 91689
+    },
+    {
+      "epoch": 9.466446417509808,
+      "grad_norm": 0.08993300050497055,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 91692
+    },
+    {
+      "epoch": 9.466756142886641,
+      "grad_norm": 0.039396051317453384,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 91695
+    },
+    {
+      "epoch": 9.467065868263473,
+      "grad_norm": 0.04936233535408974,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 91698
+    },
+    {
+      "epoch": 9.467375593640305,
+      "grad_norm": 0.041177988052368164,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 91701
+    },
+    {
+      "epoch": 9.467685319017138,
+      "grad_norm": 0.10703952610492706,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 91704
+    },
+    {
+      "epoch": 9.467995044393971,
+      "grad_norm": 0.03918042033910751,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 91707
+    },
+    {
+      "epoch": 9.468304769770803,
+      "grad_norm": 0.08904026448726654,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 91710
+    },
+    {
+      "epoch": 9.468614495147635,
+      "grad_norm": 0.1134420707821846,
+      "learning_rate": 0.01,
+      "loss": 1.9013,
+      "step": 91713
+    },
+    {
+      "epoch": 9.468924220524467,
+      "grad_norm": 0.11754108965396881,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 91716
+    },
+    {
+      "epoch": 9.469233945901301,
+      "grad_norm": 0.06243946775794029,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 91719
+    },
+    {
+      "epoch": 9.469543671278133,
+      "grad_norm": 0.05874892324209213,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 91722
+    },
+    {
+      "epoch": 9.469853396654965,
+      "grad_norm": 0.06552620232105255,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 91725
+    },
+    {
+      "epoch": 9.4701631220318,
+      "grad_norm": 0.03773637115955353,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 91728
+    },
+    {
+      "epoch": 9.470472847408631,
+      "grad_norm": 0.03343876451253891,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 91731
+    },
+    {
+      "epoch": 9.470782572785463,
+      "grad_norm": 0.06106181442737579,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 91734
+    },
+    {
+      "epoch": 9.471092298162295,
+      "grad_norm": 0.04692843556404114,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 91737
+    },
+    {
+      "epoch": 9.47140202353913,
+      "grad_norm": 0.046416301280260086,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 91740
+    },
+    {
+      "epoch": 9.471711748915961,
+      "grad_norm": 0.05070651322603226,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 91743
+    },
+    {
+      "epoch": 9.472021474292793,
+      "grad_norm": 0.17885123193264008,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 91746
+    },
+    {
+      "epoch": 9.472331199669627,
+      "grad_norm": 0.05231613665819168,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 91749
+    },
+    {
+      "epoch": 9.47264092504646,
+      "grad_norm": 0.06618047505617142,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 91752
+    },
+    {
+      "epoch": 9.472950650423291,
+      "grad_norm": 0.06575289368629456,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 91755
+    },
+    {
+      "epoch": 9.473260375800123,
+      "grad_norm": 0.10142083466053009,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 91758
+    },
+    {
+      "epoch": 9.473570101176957,
+      "grad_norm": 0.0768078863620758,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 91761
+    },
+    {
+      "epoch": 9.47387982655379,
+      "grad_norm": 0.07403820008039474,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 91764
+    },
+    {
+      "epoch": 9.474189551930621,
+      "grad_norm": 0.05097631365060806,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 91767
+    },
+    {
+      "epoch": 9.474499277307453,
+      "grad_norm": 0.05768020078539848,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 91770
+    },
+    {
+      "epoch": 9.474809002684287,
+      "grad_norm": 0.04441077262163162,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 91773
+    },
+    {
+      "epoch": 9.475118728061119,
+      "grad_norm": 0.04517601802945137,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 91776
+    },
+    {
+      "epoch": 9.475428453437951,
+      "grad_norm": 0.03389676660299301,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 91779
+    },
+    {
+      "epoch": 9.475738178814785,
+      "grad_norm": 0.04144905135035515,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 91782
+    },
+    {
+      "epoch": 9.476047904191617,
+      "grad_norm": 0.03675378859043121,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 91785
+    },
+    {
+      "epoch": 9.476357629568449,
+      "grad_norm": 0.07525095343589783,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 91788
+    },
+    {
+      "epoch": 9.476667354945281,
+      "grad_norm": 0.19068317115306854,
+      "learning_rate": 0.01,
+      "loss": 1.9793,
+      "step": 91791
+    },
+    {
+      "epoch": 9.476977080322115,
+      "grad_norm": 0.10045140981674194,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 91794
+    },
+    {
+      "epoch": 9.477286805698947,
+      "grad_norm": 0.04264092817902565,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 91797
+    },
+    {
+      "epoch": 9.477596531075779,
+      "grad_norm": 0.03995588794350624,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 91800
+    },
+    {
+      "epoch": 9.477906256452613,
+      "grad_norm": 0.036804333329200745,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 91803
+    },
+    {
+      "epoch": 9.478215981829445,
+      "grad_norm": 0.05768147483468056,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 91806
+    },
+    {
+      "epoch": 9.478525707206277,
+      "grad_norm": 0.05145462602376938,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 91809
+    },
+    {
+      "epoch": 9.478835432583109,
+      "grad_norm": 0.10472197830677032,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 91812
+    },
+    {
+      "epoch": 9.479145157959943,
+      "grad_norm": 0.04309898242354393,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 91815
+    },
+    {
+      "epoch": 9.479454883336775,
+      "grad_norm": 0.0850476399064064,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 91818
+    },
+    {
+      "epoch": 9.479764608713607,
+      "grad_norm": 0.07317638397216797,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 91821
+    },
+    {
+      "epoch": 9.480074334090439,
+      "grad_norm": 0.13520605862140656,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 91824
+    },
+    {
+      "epoch": 9.480384059467273,
+      "grad_norm": 0.03584807738661766,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 91827
+    },
+    {
+      "epoch": 9.480693784844105,
+      "grad_norm": 0.05352088063955307,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 91830
+    },
+    {
+      "epoch": 9.481003510220937,
+      "grad_norm": 0.03882770612835884,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 91833
+    },
+    {
+      "epoch": 9.48131323559777,
+      "grad_norm": 0.059367503970861435,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 91836
+    },
+    {
+      "epoch": 9.481622960974603,
+      "grad_norm": 0.06464463472366333,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 91839
+    },
+    {
+      "epoch": 9.481932686351435,
+      "grad_norm": 0.042919889092445374,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 91842
+    },
+    {
+      "epoch": 9.482242411728267,
+      "grad_norm": 0.04149112477898598,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 91845
+    },
+    {
+      "epoch": 9.4825521371051,
+      "grad_norm": 0.03673546761274338,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 91848
+    },
+    {
+      "epoch": 9.482861862481933,
+      "grad_norm": 0.04928651452064514,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 91851
+    },
+    {
+      "epoch": 9.483171587858765,
+      "grad_norm": 0.11838643252849579,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 91854
+    },
+    {
+      "epoch": 9.483481313235599,
+      "grad_norm": 0.08629243820905685,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 91857
+    },
+    {
+      "epoch": 9.48379103861243,
+      "grad_norm": 0.05888286978006363,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 91860
+    },
+    {
+      "epoch": 9.484100763989263,
+      "grad_norm": 0.07269131392240524,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 91863
+    },
+    {
+      "epoch": 9.484410489366095,
+      "grad_norm": 0.049393005669116974,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 91866
+    },
+    {
+      "epoch": 9.484720214742929,
+      "grad_norm": 0.057828810065984726,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 91869
+    },
+    {
+      "epoch": 9.48502994011976,
+      "grad_norm": 0.051319871097803116,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 91872
+    },
+    {
+      "epoch": 9.485339665496593,
+      "grad_norm": 0.03736547380685806,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 91875
+    },
+    {
+      "epoch": 9.485649390873425,
+      "grad_norm": 0.03503834083676338,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 91878
+    },
+    {
+      "epoch": 9.485959116250259,
+      "grad_norm": 0.03310151398181915,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 91881
+    },
+    {
+      "epoch": 9.48626884162709,
+      "grad_norm": 0.10385093092918396,
+      "learning_rate": 0.01,
+      "loss": 1.8883,
+      "step": 91884
+    },
+    {
+      "epoch": 9.486578567003923,
+      "grad_norm": 0.061645470559597015,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 91887
+    },
+    {
+      "epoch": 9.486888292380756,
+      "grad_norm": 0.08191576600074768,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 91890
+    },
+    {
+      "epoch": 9.487198017757589,
+      "grad_norm": 0.06519359350204468,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 91893
+    },
+    {
+      "epoch": 9.48750774313442,
+      "grad_norm": 0.07691865414381027,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 91896
+    },
+    {
+      "epoch": 9.487817468511253,
+      "grad_norm": 0.07537966966629028,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 91899
+    },
+    {
+      "epoch": 9.488127193888086,
+      "grad_norm": 0.12493839114904404,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 91902
+    },
+    {
+      "epoch": 9.488436919264919,
+      "grad_norm": 0.04485747963190079,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 91905
+    },
+    {
+      "epoch": 9.48874664464175,
+      "grad_norm": 0.06224820017814636,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 91908
+    },
+    {
+      "epoch": 9.489056370018584,
+      "grad_norm": 0.03537846729159355,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 91911
+    },
+    {
+      "epoch": 9.489366095395416,
+      "grad_norm": 0.05944729223847389,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 91914
+    },
+    {
+      "epoch": 9.489675820772248,
+      "grad_norm": 0.11124984174966812,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 91917
+    },
+    {
+      "epoch": 9.48998554614908,
+      "grad_norm": 0.06960305571556091,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 91920
+    },
+    {
+      "epoch": 9.490295271525914,
+      "grad_norm": 0.08293911069631577,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 91923
+    },
+    {
+      "epoch": 9.490604996902746,
+      "grad_norm": 0.0760396271944046,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 91926
+    },
+    {
+      "epoch": 9.490914722279578,
+      "grad_norm": 0.08136795461177826,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 91929
+    },
+    {
+      "epoch": 9.49122444765641,
+      "grad_norm": 0.0484798438847065,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 91932
+    },
+    {
+      "epoch": 9.491534173033244,
+      "grad_norm": 0.034965045750141144,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 91935
+    },
+    {
+      "epoch": 9.491843898410076,
+      "grad_norm": 0.037586700171232224,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 91938
+    },
+    {
+      "epoch": 9.492153623786908,
+      "grad_norm": 0.1187121570110321,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 91941
+    },
+    {
+      "epoch": 9.492463349163742,
+      "grad_norm": 0.043870143592357635,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 91944
+    },
+    {
+      "epoch": 9.492773074540574,
+      "grad_norm": 0.057566460222005844,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 91947
+    },
+    {
+      "epoch": 9.493082799917406,
+      "grad_norm": 0.04957624152302742,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 91950
+    },
+    {
+      "epoch": 9.493392525294238,
+      "grad_norm": 0.05048470199108124,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 91953
+    },
+    {
+      "epoch": 9.493702250671072,
+      "grad_norm": 0.046507786959409714,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 91956
+    },
+    {
+      "epoch": 9.494011976047904,
+      "grad_norm": 0.04052937403321266,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 91959
+    },
+    {
+      "epoch": 9.494321701424736,
+      "grad_norm": 0.05633610486984253,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 91962
+    },
+    {
+      "epoch": 9.49463142680157,
+      "grad_norm": 0.11403268575668335,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 91965
+    },
+    {
+      "epoch": 9.494941152178402,
+      "grad_norm": 0.06479247659444809,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 91968
+    },
+    {
+      "epoch": 9.495250877555234,
+      "grad_norm": 0.09049377590417862,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 91971
+    },
+    {
+      "epoch": 9.495560602932066,
+      "grad_norm": 0.031010787934064865,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 91974
+    },
+    {
+      "epoch": 9.4958703283089,
+      "grad_norm": 0.057340823113918304,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 91977
+    },
+    {
+      "epoch": 9.496180053685732,
+      "grad_norm": 0.08693788945674896,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 91980
+    },
+    {
+      "epoch": 9.496489779062564,
+      "grad_norm": 0.07110211253166199,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 91983
+    },
+    {
+      "epoch": 9.496799504439396,
+      "grad_norm": 0.03922433406114578,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 91986
+    },
+    {
+      "epoch": 9.49710922981623,
+      "grad_norm": 0.045608848333358765,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 91989
+    },
+    {
+      "epoch": 9.497418955193062,
+      "grad_norm": 0.05738776549696922,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 91992
+    },
+    {
+      "epoch": 9.497728680569894,
+      "grad_norm": 0.03917551040649414,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 91995
+    },
+    {
+      "epoch": 9.498038405946728,
+      "grad_norm": 0.09609241038560867,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 91998
+    },
+    {
+      "epoch": 9.49834813132356,
+      "grad_norm": 0.08743585646152496,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 92001
+    },
+    {
+      "epoch": 9.498657856700392,
+      "grad_norm": 0.09634451568126678,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 92004
+    },
+    {
+      "epoch": 9.498967582077224,
+      "grad_norm": 0.10952823609113693,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 92007
+    },
+    {
+      "epoch": 9.499277307454058,
+      "grad_norm": 0.05229433625936508,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 92010
+    },
+    {
+      "epoch": 9.49958703283089,
+      "grad_norm": 0.05109557509422302,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 92013
+    },
+    {
+      "epoch": 9.499896758207722,
+      "grad_norm": 0.03826722875237465,
+      "learning_rate": 0.01,
+      "loss": 1.9037,
+      "step": 92016
+    },
+    {
+      "epoch": 9.500206483584556,
+      "grad_norm": 0.04049082472920418,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 92019
+    },
+    {
+      "epoch": 9.500516208961388,
+      "grad_norm": 0.06057979166507721,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 92022
+    },
+    {
+      "epoch": 9.50082593433822,
+      "grad_norm": 0.05316155031323433,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 92025
+    },
+    {
+      "epoch": 9.501135659715052,
+      "grad_norm": 0.12783922255039215,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 92028
+    },
+    {
+      "epoch": 9.501445385091886,
+      "grad_norm": 0.054156944155693054,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 92031
+    },
+    {
+      "epoch": 9.501755110468718,
+      "grad_norm": 0.05763794109225273,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 92034
+    },
+    {
+      "epoch": 9.50206483584555,
+      "grad_norm": 0.10921145230531693,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 92037
+    },
+    {
+      "epoch": 9.502374561222382,
+      "grad_norm": 0.033268216997385025,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 92040
+    },
+    {
+      "epoch": 9.502684286599216,
+      "grad_norm": 0.16365273296833038,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 92043
+    },
+    {
+      "epoch": 9.502994011976048,
+      "grad_norm": 0.1298300176858902,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 92046
+    },
+    {
+      "epoch": 9.50330373735288,
+      "grad_norm": 0.05752571299672127,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 92049
+    },
+    {
+      "epoch": 9.503613462729714,
+      "grad_norm": 0.04481608420610428,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 92052
+    },
+    {
+      "epoch": 9.503923188106546,
+      "grad_norm": 0.04134489968419075,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 92055
+    },
+    {
+      "epoch": 9.504232913483378,
+      "grad_norm": 0.04483073204755783,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 92058
+    },
+    {
+      "epoch": 9.50454263886021,
+      "grad_norm": 0.06324251741170883,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 92061
+    },
+    {
+      "epoch": 9.504852364237044,
+      "grad_norm": 0.047804441303014755,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 92064
+    },
+    {
+      "epoch": 9.505162089613876,
+      "grad_norm": 0.04165074974298477,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 92067
+    },
+    {
+      "epoch": 9.505471814990708,
+      "grad_norm": 0.05379807949066162,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 92070
+    },
+    {
+      "epoch": 9.50578154036754,
+      "grad_norm": 0.06253156065940857,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 92073
+    },
+    {
+      "epoch": 9.506091265744374,
+      "grad_norm": 0.04284733533859253,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 92076
+    },
+    {
+      "epoch": 9.506400991121206,
+      "grad_norm": 0.05847569555044174,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 92079
+    },
+    {
+      "epoch": 9.506710716498038,
+      "grad_norm": 0.049981147050857544,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 92082
+    },
+    {
+      "epoch": 9.507020441874872,
+      "grad_norm": 0.056745145469903946,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 92085
+    },
+    {
+      "epoch": 9.507330167251704,
+      "grad_norm": 0.08482272922992706,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 92088
+    },
+    {
+      "epoch": 9.507639892628536,
+      "grad_norm": 0.036681801080703735,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 92091
+    },
+    {
+      "epoch": 9.507949618005368,
+      "grad_norm": 0.05905177444219589,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 92094
+    },
+    {
+      "epoch": 9.508259343382202,
+      "grad_norm": 0.09597323834896088,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 92097
+    },
+    {
+      "epoch": 9.508569068759034,
+      "grad_norm": 0.07549446821212769,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 92100
+    },
+    {
+      "epoch": 9.508878794135866,
+      "grad_norm": 0.06610140949487686,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 92103
+    },
+    {
+      "epoch": 9.5091885195127,
+      "grad_norm": 0.05344447121024132,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 92106
+    },
+    {
+      "epoch": 9.509498244889532,
+      "grad_norm": 0.04548199102282524,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 92109
+    },
+    {
+      "epoch": 9.509807970266364,
+      "grad_norm": 0.06513753533363342,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 92112
+    },
+    {
+      "epoch": 9.510117695643196,
+      "grad_norm": 0.04577668383717537,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 92115
+    },
+    {
+      "epoch": 9.51042742102003,
+      "grad_norm": 0.1895677149295807,
+      "learning_rate": 0.01,
+      "loss": 1.9041,
+      "step": 92118
+    },
+    {
+      "epoch": 9.510737146396862,
+      "grad_norm": 0.04677163064479828,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 92121
+    },
+    {
+      "epoch": 9.511046871773694,
+      "grad_norm": 0.04040669649839401,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 92124
+    },
+    {
+      "epoch": 9.511356597150527,
+      "grad_norm": 0.034733910113573074,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 92127
+    },
+    {
+      "epoch": 9.51166632252736,
+      "grad_norm": 0.04119290038943291,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 92130
+    },
+    {
+      "epoch": 9.511976047904191,
+      "grad_norm": 0.05976971238851547,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 92133
+    },
+    {
+      "epoch": 9.512285773281024,
+      "grad_norm": 0.099224753677845,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 92136
+    },
+    {
+      "epoch": 9.512595498657857,
+      "grad_norm": 0.1720251590013504,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 92139
+    },
+    {
+      "epoch": 9.51290522403469,
+      "grad_norm": 0.10159769654273987,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 92142
+    },
+    {
+      "epoch": 9.513214949411521,
+      "grad_norm": 0.07902245968580246,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 92145
+    },
+    {
+      "epoch": 9.513524674788354,
+      "grad_norm": 0.05916019156575203,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 92148
+    },
+    {
+      "epoch": 9.513834400165187,
+      "grad_norm": 0.07787574082612991,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 92151
+    },
+    {
+      "epoch": 9.51414412554202,
+      "grad_norm": 0.05786137282848358,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 92154
+    },
+    {
+      "epoch": 9.514453850918851,
+      "grad_norm": 0.06773325055837631,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 92157
+    },
+    {
+      "epoch": 9.514763576295685,
+      "grad_norm": 0.08434304594993591,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 92160
+    },
+    {
+      "epoch": 9.515073301672517,
+      "grad_norm": 0.04950158670544624,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 92163
+    },
+    {
+      "epoch": 9.51538302704935,
+      "grad_norm": 0.06106577441096306,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 92166
+    },
+    {
+      "epoch": 9.515692752426181,
+      "grad_norm": 0.07273181527853012,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 92169
+    },
+    {
+      "epoch": 9.516002477803015,
+      "grad_norm": 0.0501217283308506,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 92172
+    },
+    {
+      "epoch": 9.516312203179847,
+      "grad_norm": 0.09720636904239655,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 92175
+    },
+    {
+      "epoch": 9.51662192855668,
+      "grad_norm": 0.10266958177089691,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 92178
+    },
+    {
+      "epoch": 9.516931653933511,
+      "grad_norm": 0.059654563665390015,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 92181
+    },
+    {
+      "epoch": 9.517241379310345,
+      "grad_norm": 0.029505256563425064,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 92184
+    },
+    {
+      "epoch": 9.517551104687177,
+      "grad_norm": 0.03506319597363472,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 92187
+    },
+    {
+      "epoch": 9.51786083006401,
+      "grad_norm": 0.058983176946640015,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 92190
+    },
+    {
+      "epoch": 9.518170555440843,
+      "grad_norm": 0.08283732086420059,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 92193
+    },
+    {
+      "epoch": 9.518480280817675,
+      "grad_norm": 0.06719473749399185,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 92196
+    },
+    {
+      "epoch": 9.518790006194507,
+      "grad_norm": 0.13410811126232147,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 92199
+    },
+    {
+      "epoch": 9.51909973157134,
+      "grad_norm": 0.04871629923582077,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 92202
+    },
+    {
+      "epoch": 9.519409456948173,
+      "grad_norm": 0.06901410222053528,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 92205
+    },
+    {
+      "epoch": 9.519719182325005,
+      "grad_norm": 0.04040338844060898,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 92208
+    },
+    {
+      "epoch": 9.520028907701837,
+      "grad_norm": 0.04763796925544739,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 92211
+    },
+    {
+      "epoch": 9.520338633078671,
+      "grad_norm": 0.056972626596689224,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 92214
+    },
+    {
+      "epoch": 9.520648358455503,
+      "grad_norm": 0.0520622618496418,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 92217
+    },
+    {
+      "epoch": 9.520958083832335,
+      "grad_norm": 0.11873430758714676,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 92220
+    },
+    {
+      "epoch": 9.521267809209167,
+      "grad_norm": 0.04310165345668793,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 92223
+    },
+    {
+      "epoch": 9.521577534586001,
+      "grad_norm": 0.10814736783504486,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 92226
+    },
+    {
+      "epoch": 9.521887259962833,
+      "grad_norm": 0.0352197028696537,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 92229
+    },
+    {
+      "epoch": 9.522196985339665,
+      "grad_norm": 0.07474922388792038,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 92232
+    },
+    {
+      "epoch": 9.522506710716499,
+      "grad_norm": 0.07361693680286407,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 92235
+    },
+    {
+      "epoch": 9.522816436093331,
+      "grad_norm": 0.07509438693523407,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 92238
+    },
+    {
+      "epoch": 9.523126161470163,
+      "grad_norm": 0.053927212953567505,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 92241
+    },
+    {
+      "epoch": 9.523435886846995,
+      "grad_norm": 0.04635823145508766,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 92244
+    },
+    {
+      "epoch": 9.523745612223829,
+      "grad_norm": 0.06367374211549759,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 92247
+    },
+    {
+      "epoch": 9.524055337600661,
+      "grad_norm": 0.08359774947166443,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 92250
+    },
+    {
+      "epoch": 9.524365062977493,
+      "grad_norm": 0.05048409104347229,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 92253
+    },
+    {
+      "epoch": 9.524674788354325,
+      "grad_norm": 0.050950534641742706,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 92256
+    },
+    {
+      "epoch": 9.524984513731159,
+      "grad_norm": 0.09251030534505844,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 92259
+    },
+    {
+      "epoch": 9.525294239107991,
+      "grad_norm": 0.10388283431529999,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 92262
+    },
+    {
+      "epoch": 9.525603964484823,
+      "grad_norm": 0.06290477514266968,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 92265
+    },
+    {
+      "epoch": 9.525913689861657,
+      "grad_norm": 0.06396323442459106,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 92268
+    },
+    {
+      "epoch": 9.526223415238489,
+      "grad_norm": 0.049475327134132385,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 92271
+    },
+    {
+      "epoch": 9.52653314061532,
+      "grad_norm": 0.04767010733485222,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 92274
+    },
+    {
+      "epoch": 9.526842865992153,
+      "grad_norm": 0.050544627010822296,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 92277
+    },
+    {
+      "epoch": 9.527152591368987,
+      "grad_norm": 0.05860473960638046,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 92280
+    },
+    {
+      "epoch": 9.527462316745819,
+      "grad_norm": 0.053062696009874344,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 92283
+    },
+    {
+      "epoch": 9.52777204212265,
+      "grad_norm": 0.04594952240586281,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 92286
+    },
+    {
+      "epoch": 9.528081767499483,
+      "grad_norm": 0.04546007886528969,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 92289
+    },
+    {
+      "epoch": 9.528391492876317,
+      "grad_norm": 0.13376648724079132,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 92292
+    },
+    {
+      "epoch": 9.528701218253149,
+      "grad_norm": 0.039482396095991135,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 92295
+    },
+    {
+      "epoch": 9.52901094362998,
+      "grad_norm": 0.08189577609300613,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 92298
+    },
+    {
+      "epoch": 9.529320669006815,
+      "grad_norm": 0.043656591325998306,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 92301
+    },
+    {
+      "epoch": 9.529630394383647,
+      "grad_norm": 0.047711532562971115,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 92304
+    },
+    {
+      "epoch": 9.529940119760479,
+      "grad_norm": 0.0450308732688427,
+      "learning_rate": 0.01,
+      "loss": 1.8815,
+      "step": 92307
+    },
+    {
+      "epoch": 9.53024984513731,
+      "grad_norm": 0.038819797337055206,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 92310
+    },
+    {
+      "epoch": 9.530559570514145,
+      "grad_norm": 0.07087231427431107,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 92313
+    },
+    {
+      "epoch": 9.530869295890977,
+      "grad_norm": 0.09496890753507614,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 92316
+    },
+    {
+      "epoch": 9.531179021267809,
+      "grad_norm": 0.11516691744327545,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 92319
+    },
+    {
+      "epoch": 9.531488746644643,
+      "grad_norm": 0.07193636149168015,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 92322
+    },
+    {
+      "epoch": 9.531798472021475,
+      "grad_norm": 0.06717932969331741,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 92325
+    },
+    {
+      "epoch": 9.532108197398307,
+      "grad_norm": 0.06636080890893936,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 92328
+    },
+    {
+      "epoch": 9.532417922775139,
+      "grad_norm": 0.03784407302737236,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 92331
+    },
+    {
+      "epoch": 9.532727648151972,
+      "grad_norm": 0.14190025627613068,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 92334
+    },
+    {
+      "epoch": 9.533037373528805,
+      "grad_norm": 0.0631750226020813,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 92337
+    },
+    {
+      "epoch": 9.533347098905637,
+      "grad_norm": 0.058418259024620056,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 92340
+    },
+    {
+      "epoch": 9.53365682428247,
+      "grad_norm": 0.05124678462743759,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 92343
+    },
+    {
+      "epoch": 9.533966549659302,
+      "grad_norm": 0.040973249822854996,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 92346
+    },
+    {
+      "epoch": 9.534276275036135,
+      "grad_norm": 0.039508648216724396,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 92349
+    },
+    {
+      "epoch": 9.534586000412967,
+      "grad_norm": 0.04920020326972008,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 92352
+    },
+    {
+      "epoch": 9.5348957257898,
+      "grad_norm": 0.06824755668640137,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 92355
+    },
+    {
+      "epoch": 9.535205451166632,
+      "grad_norm": 0.061849042773246765,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 92358
+    },
+    {
+      "epoch": 9.535515176543464,
+      "grad_norm": 0.04916265979409218,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 92361
+    },
+    {
+      "epoch": 9.535824901920297,
+      "grad_norm": 0.054886508733034134,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 92364
+    },
+    {
+      "epoch": 9.53613462729713,
+      "grad_norm": 0.04167830944061279,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 92367
+    },
+    {
+      "epoch": 9.536444352673962,
+      "grad_norm": 0.10808145999908447,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 92370
+    },
+    {
+      "epoch": 9.536754078050794,
+      "grad_norm": 0.08303140848875046,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 92373
+    },
+    {
+      "epoch": 9.537063803427628,
+      "grad_norm": 0.046498388051986694,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 92376
+    },
+    {
+      "epoch": 9.53737352880446,
+      "grad_norm": 0.07408872991800308,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 92379
+    },
+    {
+      "epoch": 9.537683254181292,
+      "grad_norm": 0.03534821420907974,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 92382
+    },
+    {
+      "epoch": 9.537992979558124,
+      "grad_norm": 0.050212305039167404,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 92385
+    },
+    {
+      "epoch": 9.538302704934958,
+      "grad_norm": 0.03994080796837807,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 92388
+    },
+    {
+      "epoch": 9.53861243031179,
+      "grad_norm": 0.04362984001636505,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 92391
+    },
+    {
+      "epoch": 9.538922155688622,
+      "grad_norm": 0.0660531148314476,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 92394
+    },
+    {
+      "epoch": 9.539231881065454,
+      "grad_norm": 0.11909878253936768,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 92397
+    },
+    {
+      "epoch": 9.539541606442288,
+      "grad_norm": 0.06280426681041718,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 92400
+    },
+    {
+      "epoch": 9.53985133181912,
+      "grad_norm": 0.07050773501396179,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 92403
+    },
+    {
+      "epoch": 9.540161057195952,
+      "grad_norm": 0.031119951978325844,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 92406
+    },
+    {
+      "epoch": 9.540470782572786,
+      "grad_norm": 0.04030071943998337,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 92409
+    },
+    {
+      "epoch": 9.540780507949618,
+      "grad_norm": 0.11780594289302826,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 92412
+    },
+    {
+      "epoch": 9.54109023332645,
+      "grad_norm": 0.07259298115968704,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 92415
+    },
+    {
+      "epoch": 9.541399958703282,
+      "grad_norm": 0.07972489297389984,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 92418
+    },
+    {
+      "epoch": 9.541709684080116,
+      "grad_norm": 0.06060254946351051,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 92421
+    },
+    {
+      "epoch": 9.542019409456948,
+      "grad_norm": 0.057979363948106766,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 92424
+    },
+    {
+      "epoch": 9.54232913483378,
+      "grad_norm": 0.042889319360256195,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 92427
+    },
+    {
+      "epoch": 9.542638860210614,
+      "grad_norm": 0.11210845410823822,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 92430
+    },
+    {
+      "epoch": 9.542948585587446,
+      "grad_norm": 0.09843835234642029,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 92433
+    },
+    {
+      "epoch": 9.543258310964278,
+      "grad_norm": 0.03776775673031807,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 92436
+    },
+    {
+      "epoch": 9.54356803634111,
+      "grad_norm": 0.06357332319021225,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 92439
+    },
+    {
+      "epoch": 9.543877761717944,
+      "grad_norm": 0.0598602257668972,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 92442
+    },
+    {
+      "epoch": 9.544187487094776,
+      "grad_norm": 0.04040971025824547,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 92445
+    },
+    {
+      "epoch": 9.544497212471608,
+      "grad_norm": 0.055280424654483795,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 92448
+    },
+    {
+      "epoch": 9.544806937848442,
+      "grad_norm": 0.06608704477548599,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 92451
+    },
+    {
+      "epoch": 9.545116663225274,
+      "grad_norm": 0.04509027302265167,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 92454
+    },
+    {
+      "epoch": 9.545426388602106,
+      "grad_norm": 0.12730321288108826,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 92457
+    },
+    {
+      "epoch": 9.545736113978938,
+      "grad_norm": 0.05638007074594498,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 92460
+    },
+    {
+      "epoch": 9.546045839355772,
+      "grad_norm": 0.0741688534617424,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 92463
+    },
+    {
+      "epoch": 9.546355564732604,
+      "grad_norm": 0.11796123534440994,
+      "learning_rate": 0.01,
+      "loss": 1.9914,
+      "step": 92466
+    },
+    {
+      "epoch": 9.546665290109436,
+      "grad_norm": 0.05692535266280174,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 92469
+    },
+    {
+      "epoch": 9.546975015486268,
+      "grad_norm": 0.060367029160261154,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 92472
+    },
+    {
+      "epoch": 9.547284740863102,
+      "grad_norm": 0.062419138848781586,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 92475
+    },
+    {
+      "epoch": 9.547594466239934,
+      "grad_norm": 0.04204673692584038,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 92478
+    },
+    {
+      "epoch": 9.547904191616766,
+      "grad_norm": 0.07150115817785263,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 92481
+    },
+    {
+      "epoch": 9.5482139169936,
+      "grad_norm": 0.08063635975122452,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 92484
+    },
+    {
+      "epoch": 9.548523642370432,
+      "grad_norm": 0.06927258521318436,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 92487
+    },
+    {
+      "epoch": 9.548833367747264,
+      "grad_norm": 0.08684887737035751,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 92490
+    },
+    {
+      "epoch": 9.549143093124096,
+      "grad_norm": 0.10413134843111038,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 92493
+    },
+    {
+      "epoch": 9.54945281850093,
+      "grad_norm": 0.05053870007395744,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 92496
+    },
+    {
+      "epoch": 9.549762543877762,
+      "grad_norm": 0.1296710968017578,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 92499
+    },
+    {
+      "epoch": 9.550072269254594,
+      "grad_norm": 0.052506376057863235,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 92502
+    },
+    {
+      "epoch": 9.550381994631426,
+      "grad_norm": 0.0511232390999794,
+      "learning_rate": 0.01,
+      "loss": 1.9881,
+      "step": 92505
+    },
+    {
+      "epoch": 9.55069172000826,
+      "grad_norm": 0.05901256203651428,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 92508
+    },
+    {
+      "epoch": 9.551001445385092,
+      "grad_norm": 0.03338488191366196,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 92511
+    },
+    {
+      "epoch": 9.551311170761924,
+      "grad_norm": 0.04231966286897659,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 92514
+    },
+    {
+      "epoch": 9.551620896138758,
+      "grad_norm": 0.061377670615911484,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 92517
+    },
+    {
+      "epoch": 9.55193062151559,
+      "grad_norm": 0.06680765748023987,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 92520
+    },
+    {
+      "epoch": 9.552240346892422,
+      "grad_norm": 0.0845053419470787,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 92523
+    },
+    {
+      "epoch": 9.552550072269254,
+      "grad_norm": 0.06742209941148758,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 92526
+    },
+    {
+      "epoch": 9.552859797646088,
+      "grad_norm": 0.0977376326918602,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 92529
+    },
+    {
+      "epoch": 9.55316952302292,
+      "grad_norm": 0.054616544395685196,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 92532
+    },
+    {
+      "epoch": 9.553479248399752,
+      "grad_norm": 0.09607339650392532,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 92535
+    },
+    {
+      "epoch": 9.553788973776586,
+      "grad_norm": 0.04456086456775665,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 92538
+    },
+    {
+      "epoch": 9.554098699153418,
+      "grad_norm": 0.07154840975999832,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 92541
+    },
+    {
+      "epoch": 9.55440842453025,
+      "grad_norm": 0.08459034562110901,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 92544
+    },
+    {
+      "epoch": 9.554718149907082,
+      "grad_norm": 0.03846704959869385,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 92547
+    },
+    {
+      "epoch": 9.555027875283916,
+      "grad_norm": 0.09594748169183731,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 92550
+    },
+    {
+      "epoch": 9.555337600660748,
+      "grad_norm": 0.0669037476181984,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 92553
+    },
+    {
+      "epoch": 9.55564732603758,
+      "grad_norm": 0.06939956545829773,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 92556
+    },
+    {
+      "epoch": 9.555957051414413,
+      "grad_norm": 0.07697443664073944,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 92559
+    },
+    {
+      "epoch": 9.556266776791245,
+      "grad_norm": 0.13253872096538544,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 92562
+    },
+    {
+      "epoch": 9.556576502168078,
+      "grad_norm": 0.040784794837236404,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 92565
+    },
+    {
+      "epoch": 9.55688622754491,
+      "grad_norm": 0.04093778505921364,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 92568
+    },
+    {
+      "epoch": 9.557195952921743,
+      "grad_norm": 0.045812614262104034,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 92571
+    },
+    {
+      "epoch": 9.557505678298575,
+      "grad_norm": 0.04944195970892906,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 92574
+    },
+    {
+      "epoch": 9.557815403675407,
+      "grad_norm": 0.04963093623518944,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 92577
+    },
+    {
+      "epoch": 9.55812512905224,
+      "grad_norm": 0.11150357872247696,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 92580
+    },
+    {
+      "epoch": 9.558434854429073,
+      "grad_norm": 0.07571767270565033,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 92583
+    },
+    {
+      "epoch": 9.558744579805905,
+      "grad_norm": 0.11754986643791199,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 92586
+    },
+    {
+      "epoch": 9.559054305182737,
+      "grad_norm": 0.05976712331175804,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 92589
+    },
+    {
+      "epoch": 9.559364030559571,
+      "grad_norm": 0.04479925334453583,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 92592
+    },
+    {
+      "epoch": 9.559673755936403,
+      "grad_norm": 0.047420915216207504,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 92595
+    },
+    {
+      "epoch": 9.559983481313235,
+      "grad_norm": 0.08457601070404053,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 92598
+    },
+    {
+      "epoch": 9.560293206690067,
+      "grad_norm": 0.09099945425987244,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 92601
+    },
+    {
+      "epoch": 9.560602932066901,
+      "grad_norm": 0.06987515091896057,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 92604
+    },
+    {
+      "epoch": 9.560912657443733,
+      "grad_norm": 0.05262571573257446,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 92607
+    },
+    {
+      "epoch": 9.561222382820565,
+      "grad_norm": 0.05382346734404564,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 92610
+    },
+    {
+      "epoch": 9.561532108197397,
+      "grad_norm": 0.0843144878745079,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 92613
+    },
+    {
+      "epoch": 9.561841833574231,
+      "grad_norm": 0.09895461797714233,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 92616
+    },
+    {
+      "epoch": 9.562151558951063,
+      "grad_norm": 0.036865390837192535,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 92619
+    },
+    {
+      "epoch": 9.562461284327895,
+      "grad_norm": 0.04547613859176636,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 92622
+    },
+    {
+      "epoch": 9.56277100970473,
+      "grad_norm": 0.04785327613353729,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 92625
+    },
+    {
+      "epoch": 9.563080735081561,
+      "grad_norm": 0.03501299396157265,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 92628
+    },
+    {
+      "epoch": 9.563390460458393,
+      "grad_norm": 0.09908343851566315,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 92631
+    },
+    {
+      "epoch": 9.563700185835225,
+      "grad_norm": 0.04419758915901184,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 92634
+    },
+    {
+      "epoch": 9.56400991121206,
+      "grad_norm": 0.0945008248090744,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 92637
+    },
+    {
+      "epoch": 9.564319636588891,
+      "grad_norm": 0.05049658194184303,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 92640
+    },
+    {
+      "epoch": 9.564629361965723,
+      "grad_norm": 0.044659726321697235,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 92643
+    },
+    {
+      "epoch": 9.564939087342557,
+      "grad_norm": 0.09373857825994492,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 92646
+    },
+    {
+      "epoch": 9.565248812719389,
+      "grad_norm": 0.0912860855460167,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 92649
+    },
+    {
+      "epoch": 9.565558538096221,
+      "grad_norm": 0.07291973382234573,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 92652
+    },
+    {
+      "epoch": 9.565868263473053,
+      "grad_norm": 0.08288957923650742,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 92655
+    },
+    {
+      "epoch": 9.566177988849887,
+      "grad_norm": 0.06711997836828232,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 92658
+    },
+    {
+      "epoch": 9.566487714226719,
+      "grad_norm": 0.048471078276634216,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 92661
+    },
+    {
+      "epoch": 9.566797439603551,
+      "grad_norm": 0.10623566806316376,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 92664
+    },
+    {
+      "epoch": 9.567107164980385,
+      "grad_norm": 0.06134321168065071,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 92667
+    },
+    {
+      "epoch": 9.567416890357217,
+      "grad_norm": 0.05728840082883835,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 92670
+    },
+    {
+      "epoch": 9.567726615734049,
+      "grad_norm": 0.07476432621479034,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 92673
+    },
+    {
+      "epoch": 9.568036341110881,
+      "grad_norm": 0.11206084489822388,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 92676
+    },
+    {
+      "epoch": 9.568346066487715,
+      "grad_norm": 0.05124055594205856,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 92679
+    },
+    {
+      "epoch": 9.568655791864547,
+      "grad_norm": 0.05256558582186699,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 92682
+    },
+    {
+      "epoch": 9.568965517241379,
+      "grad_norm": 0.04224633798003197,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 92685
+    },
+    {
+      "epoch": 9.569275242618211,
+      "grad_norm": 0.035335294902324677,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 92688
+    },
+    {
+      "epoch": 9.569584967995045,
+      "grad_norm": 0.09272423386573792,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 92691
+    },
+    {
+      "epoch": 9.569894693371877,
+      "grad_norm": 0.10086727887392044,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 92694
+    },
+    {
+      "epoch": 9.570204418748709,
+      "grad_norm": 0.0433303527534008,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 92697
+    },
+    {
+      "epoch": 9.570514144125543,
+      "grad_norm": 0.05553479865193367,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 92700
+    },
+    {
+      "epoch": 9.570823869502375,
+      "grad_norm": 0.048300497233867645,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 92703
+    },
+    {
+      "epoch": 9.571133594879207,
+      "grad_norm": 0.04815465584397316,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 92706
+    },
+    {
+      "epoch": 9.571443320256039,
+      "grad_norm": 0.035893216729164124,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 92709
+    },
+    {
+      "epoch": 9.571753045632873,
+      "grad_norm": 0.09880388528108597,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 92712
+    },
+    {
+      "epoch": 9.572062771009705,
+      "grad_norm": 0.08409396559000015,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 92715
+    },
+    {
+      "epoch": 9.572372496386537,
+      "grad_norm": 0.0986364409327507,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 92718
+    },
+    {
+      "epoch": 9.572682221763369,
+      "grad_norm": 0.08425180613994598,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 92721
+    },
+    {
+      "epoch": 9.572991947140203,
+      "grad_norm": 0.11412467062473297,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 92724
+    },
+    {
+      "epoch": 9.573301672517035,
+      "grad_norm": 0.06946521252393723,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 92727
+    },
+    {
+      "epoch": 9.573611397893867,
+      "grad_norm": 0.07573796808719635,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 92730
+    },
+    {
+      "epoch": 9.5739211232707,
+      "grad_norm": 0.060061559081077576,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 92733
+    },
+    {
+      "epoch": 9.574230848647533,
+      "grad_norm": 0.036126162856817245,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 92736
+    },
+    {
+      "epoch": 9.574540574024365,
+      "grad_norm": 0.03753150627017021,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 92739
+    },
+    {
+      "epoch": 9.574850299401197,
+      "grad_norm": 0.03168240562081337,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 92742
+    },
+    {
+      "epoch": 9.57516002477803,
+      "grad_norm": 0.043251775205135345,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 92745
+    },
+    {
+      "epoch": 9.575469750154863,
+      "grad_norm": 0.06953471899032593,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 92748
+    },
+    {
+      "epoch": 9.575779475531695,
+      "grad_norm": 0.1758134961128235,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 92751
+    },
+    {
+      "epoch": 9.576089200908529,
+      "grad_norm": 0.06691476702690125,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 92754
+    },
+    {
+      "epoch": 9.57639892628536,
+      "grad_norm": 0.06858311593532562,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 92757
+    },
+    {
+      "epoch": 9.576708651662193,
+      "grad_norm": 0.04022316634654999,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 92760
+    },
+    {
+      "epoch": 9.577018377039025,
+      "grad_norm": 0.038416821509599686,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 92763
+    },
+    {
+      "epoch": 9.577328102415859,
+      "grad_norm": 0.04924089461565018,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 92766
+    },
+    {
+      "epoch": 9.57763782779269,
+      "grad_norm": 0.12462669610977173,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 92769
+    },
+    {
+      "epoch": 9.577947553169523,
+      "grad_norm": 0.07955029606819153,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 92772
+    },
+    {
+      "epoch": 9.578257278546356,
+      "grad_norm": 0.0630539059638977,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 92775
+    },
+    {
+      "epoch": 9.578567003923188,
+      "grad_norm": 0.04444093257188797,
+      "learning_rate": 0.01,
+      "loss": 1.8976,
+      "step": 92778
+    },
+    {
+      "epoch": 9.57887672930002,
+      "grad_norm": 0.08930259197950363,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 92781
+    },
+    {
+      "epoch": 9.579186454676853,
+      "grad_norm": 0.04263152554631233,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 92784
+    },
+    {
+      "epoch": 9.579496180053686,
+      "grad_norm": 0.06594117730855942,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 92787
+    },
+    {
+      "epoch": 9.579805905430518,
+      "grad_norm": 0.0988270565867424,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 92790
+    },
+    {
+      "epoch": 9.58011563080735,
+      "grad_norm": 0.041310135275125504,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 92793
+    },
+    {
+      "epoch": 9.580425356184183,
+      "grad_norm": 0.051440875977277756,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 92796
+    },
+    {
+      "epoch": 9.580735081561016,
+      "grad_norm": 0.03437655046582222,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 92799
+    },
+    {
+      "epoch": 9.581044806937848,
+      "grad_norm": 0.07797033339738846,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 92802
+    },
+    {
+      "epoch": 9.58135453231468,
+      "grad_norm": 0.07684805989265442,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 92805
+    },
+    {
+      "epoch": 9.581664257691514,
+      "grad_norm": 0.04312015697360039,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 92808
+    },
+    {
+      "epoch": 9.581973983068346,
+      "grad_norm": 0.03924431651830673,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 92811
+    },
+    {
+      "epoch": 9.582283708445178,
+      "grad_norm": 0.06500852853059769,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 92814
+    },
+    {
+      "epoch": 9.58259343382201,
+      "grad_norm": 0.08903056383132935,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 92817
+    },
+    {
+      "epoch": 9.582903159198844,
+      "grad_norm": 0.08870675414800644,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 92820
+    },
+    {
+      "epoch": 9.583212884575676,
+      "grad_norm": 0.03884018585085869,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 92823
+    },
+    {
+      "epoch": 9.583522609952508,
+      "grad_norm": 0.1110977828502655,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 92826
+    },
+    {
+      "epoch": 9.58383233532934,
+      "grad_norm": 0.061049140989780426,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 92829
+    },
+    {
+      "epoch": 9.584142060706174,
+      "grad_norm": 0.0767841562628746,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 92832
+    },
+    {
+      "epoch": 9.584451786083006,
+      "grad_norm": 0.05373747646808624,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 92835
+    },
+    {
+      "epoch": 9.584761511459838,
+      "grad_norm": 0.07027586549520493,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 92838
+    },
+    {
+      "epoch": 9.585071236836672,
+      "grad_norm": 0.035306427627801895,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 92841
+    },
+    {
+      "epoch": 9.585380962213504,
+      "grad_norm": 0.049179621040821075,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 92844
+    },
+    {
+      "epoch": 9.585690687590336,
+      "grad_norm": 0.11390917003154755,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 92847
+    },
+    {
+      "epoch": 9.586000412967168,
+      "grad_norm": 0.11494887620210648,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 92850
+    },
+    {
+      "epoch": 9.586310138344002,
+      "grad_norm": 0.057366497814655304,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 92853
+    },
+    {
+      "epoch": 9.586619863720834,
+      "grad_norm": 0.03873692825436592,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 92856
+    },
+    {
+      "epoch": 9.586929589097666,
+      "grad_norm": 0.0915505513548851,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 92859
+    },
+    {
+      "epoch": 9.5872393144745,
+      "grad_norm": 0.08138495683670044,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 92862
+    },
+    {
+      "epoch": 9.587549039851332,
+      "grad_norm": 0.042756564915180206,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 92865
+    },
+    {
+      "epoch": 9.587858765228164,
+      "grad_norm": 0.038679689168930054,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 92868
+    },
+    {
+      "epoch": 9.588168490604996,
+      "grad_norm": 0.060535311698913574,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 92871
+    },
+    {
+      "epoch": 9.58847821598183,
+      "grad_norm": 0.10231431573629379,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 92874
+    },
+    {
+      "epoch": 9.588787941358662,
+      "grad_norm": 0.04502299055457115,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 92877
+    },
+    {
+      "epoch": 9.589097666735494,
+      "grad_norm": 0.09531942009925842,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 92880
+    },
+    {
+      "epoch": 9.589407392112328,
+      "grad_norm": 0.08416978269815445,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 92883
+    },
+    {
+      "epoch": 9.58971711748916,
+      "grad_norm": 0.10989884287118912,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 92886
+    },
+    {
+      "epoch": 9.590026842865992,
+      "grad_norm": 0.1414986252784729,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 92889
+    },
+    {
+      "epoch": 9.590336568242824,
+      "grad_norm": 0.08130434900522232,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 92892
+    },
+    {
+      "epoch": 9.590646293619658,
+      "grad_norm": 0.0677848532795906,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 92895
+    },
+    {
+      "epoch": 9.59095601899649,
+      "grad_norm": 0.0400870256125927,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 92898
+    },
+    {
+      "epoch": 9.591265744373322,
+      "grad_norm": 0.04095206409692764,
+      "learning_rate": 0.01,
+      "loss": 1.8968,
+      "step": 92901
+    },
+    {
+      "epoch": 9.591575469750154,
+      "grad_norm": 0.041344065219163895,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 92904
+    },
+    {
+      "epoch": 9.591885195126988,
+      "grad_norm": 0.0387251190841198,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 92907
+    },
+    {
+      "epoch": 9.59219492050382,
+      "grad_norm": 0.05318775400519371,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 92910
+    },
+    {
+      "epoch": 9.592504645880652,
+      "grad_norm": 0.11325523257255554,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 92913
+    },
+    {
+      "epoch": 9.592814371257486,
+      "grad_norm": 0.0618145689368248,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 92916
+    },
+    {
+      "epoch": 9.593124096634318,
+      "grad_norm": 0.0990997925400734,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 92919
+    },
+    {
+      "epoch": 9.59343382201115,
+      "grad_norm": 0.046235982328653336,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 92922
+    },
+    {
+      "epoch": 9.593743547387982,
+      "grad_norm": 0.10246602445840836,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 92925
+    },
+    {
+      "epoch": 9.594053272764816,
+      "grad_norm": 0.10803071409463882,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 92928
+    },
+    {
+      "epoch": 9.594362998141648,
+      "grad_norm": 0.04555898904800415,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 92931
+    },
+    {
+      "epoch": 9.59467272351848,
+      "grad_norm": 0.10263539105653763,
+      "learning_rate": 0.01,
+      "loss": 1.9812,
+      "step": 92934
+    },
+    {
+      "epoch": 9.594982448895312,
+      "grad_norm": 0.05140804871916771,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 92937
+    },
+    {
+      "epoch": 9.595292174272146,
+      "grad_norm": 0.053982093930244446,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 92940
+    },
+    {
+      "epoch": 9.595601899648978,
+      "grad_norm": 0.10115093737840652,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 92943
+    },
+    {
+      "epoch": 9.59591162502581,
+      "grad_norm": 0.04465701803565025,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 92946
+    },
+    {
+      "epoch": 9.596221350402644,
+      "grad_norm": 0.04940544068813324,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 92949
+    },
+    {
+      "epoch": 9.596531075779476,
+      "grad_norm": 0.05474409461021423,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 92952
+    },
+    {
+      "epoch": 9.596840801156308,
+      "grad_norm": 0.04154708608984947,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 92955
+    },
+    {
+      "epoch": 9.59715052653314,
+      "grad_norm": 0.14576861262321472,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 92958
+    },
+    {
+      "epoch": 9.597460251909974,
+      "grad_norm": 0.09885968267917633,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 92961
+    },
+    {
+      "epoch": 9.597769977286806,
+      "grad_norm": 0.0651608407497406,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 92964
+    },
+    {
+      "epoch": 9.598079702663638,
+      "grad_norm": 0.06366647034883499,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 92967
+    },
+    {
+      "epoch": 9.598389428040472,
+      "grad_norm": 0.07258301228284836,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 92970
+    },
+    {
+      "epoch": 9.598699153417304,
+      "grad_norm": 0.051206622272729874,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 92973
+    },
+    {
+      "epoch": 9.599008878794136,
+      "grad_norm": 0.06507684290409088,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 92976
+    },
+    {
+      "epoch": 9.599318604170968,
+      "grad_norm": 0.03854987770318985,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 92979
+    },
+    {
+      "epoch": 9.599628329547802,
+      "grad_norm": 0.11893489211797714,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 92982
+    },
+    {
+      "epoch": 9.599938054924634,
+      "grad_norm": 0.06678856909275055,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 92985
+    },
+    {
+      "epoch": 9.600247780301466,
+      "grad_norm": 0.08233048021793365,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 92988
+    },
+    {
+      "epoch": 9.6005575056783,
+      "grad_norm": 0.05037279799580574,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 92991
+    },
+    {
+      "epoch": 9.600867231055132,
+      "grad_norm": 0.08859597146511078,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 92994
+    },
+    {
+      "epoch": 9.601176956431964,
+      "grad_norm": 0.10530859231948853,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 92997
+    },
+    {
+      "epoch": 9.601486681808796,
+      "grad_norm": 0.07208415120840073,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 93000
+    },
+    {
+      "epoch": 9.60179640718563,
+      "grad_norm": 0.0347345806658268,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 93003
+    },
+    {
+      "epoch": 9.602106132562461,
+      "grad_norm": 0.08983615785837173,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 93006
+    },
+    {
+      "epoch": 9.602415857939294,
+      "grad_norm": 0.03540850058197975,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 93009
+    },
+    {
+      "epoch": 9.602725583316126,
+      "grad_norm": 0.03276904672384262,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 93012
+    },
+    {
+      "epoch": 9.60303530869296,
+      "grad_norm": 0.09266610443592072,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 93015
+    },
+    {
+      "epoch": 9.603345034069791,
+      "grad_norm": 0.04843165725469589,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 93018
+    },
+    {
+      "epoch": 9.603654759446624,
+      "grad_norm": 0.05365300923585892,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 93021
+    },
+    {
+      "epoch": 9.603964484823457,
+      "grad_norm": 0.07595492154359818,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 93024
+    },
+    {
+      "epoch": 9.60427421020029,
+      "grad_norm": 0.05023255944252014,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 93027
+    },
+    {
+      "epoch": 9.604583935577121,
+      "grad_norm": 0.07495186477899551,
+      "learning_rate": 0.01,
+      "loss": 1.9738,
+      "step": 93030
+    },
+    {
+      "epoch": 9.604893660953953,
+      "grad_norm": 0.05238397419452667,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 93033
+    },
+    {
+      "epoch": 9.605203386330787,
+      "grad_norm": 0.08827707171440125,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 93036
+    },
+    {
+      "epoch": 9.60551311170762,
+      "grad_norm": 0.13802281022071838,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 93039
+    },
+    {
+      "epoch": 9.605822837084451,
+      "grad_norm": 0.048278581351041794,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 93042
+    },
+    {
+      "epoch": 9.606132562461283,
+      "grad_norm": 0.04927634447813034,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 93045
+    },
+    {
+      "epoch": 9.606442287838117,
+      "grad_norm": 0.045615509152412415,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 93048
+    },
+    {
+      "epoch": 9.60675201321495,
+      "grad_norm": 0.04975781962275505,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 93051
+    },
+    {
+      "epoch": 9.607061738591781,
+      "grad_norm": 0.058548714965581894,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 93054
+    },
+    {
+      "epoch": 9.607371463968615,
+      "grad_norm": 0.06122971326112747,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 93057
+    },
+    {
+      "epoch": 9.607681189345447,
+      "grad_norm": 0.0376429446041584,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 93060
+    },
+    {
+      "epoch": 9.60799091472228,
+      "grad_norm": 0.06916382163763046,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 93063
+    },
+    {
+      "epoch": 9.608300640099111,
+      "grad_norm": 0.09932322800159454,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 93066
+    },
+    {
+      "epoch": 9.608610365475945,
+      "grad_norm": 0.15092147886753082,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 93069
+    },
+    {
+      "epoch": 9.608920090852777,
+      "grad_norm": 0.12672391533851624,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 93072
+    },
+    {
+      "epoch": 9.60922981622961,
+      "grad_norm": 0.05705375596880913,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 93075
+    },
+    {
+      "epoch": 9.609539541606443,
+      "grad_norm": 0.06204779073596001,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 93078
+    },
+    {
+      "epoch": 9.609849266983275,
+      "grad_norm": 0.045914001762866974,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 93081
+    },
+    {
+      "epoch": 9.610158992360107,
+      "grad_norm": 0.03120456449687481,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 93084
+    },
+    {
+      "epoch": 9.61046871773694,
+      "grad_norm": 0.03763057291507721,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 93087
+    },
+    {
+      "epoch": 9.610778443113773,
+      "grad_norm": 0.04199619218707085,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 93090
+    },
+    {
+      "epoch": 9.611088168490605,
+      "grad_norm": 0.03523045405745506,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 93093
+    },
+    {
+      "epoch": 9.611397893867437,
+      "grad_norm": 0.12717922031879425,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 93096
+    },
+    {
+      "epoch": 9.611707619244271,
+      "grad_norm": 0.06919953227043152,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 93099
+    },
+    {
+      "epoch": 9.612017344621103,
+      "grad_norm": 0.05430379509925842,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 93102
+    },
+    {
+      "epoch": 9.612327069997935,
+      "grad_norm": 0.06786497682332993,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 93105
+    },
+    {
+      "epoch": 9.612636795374767,
+      "grad_norm": 0.09560369700193405,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 93108
+    },
+    {
+      "epoch": 9.612946520751601,
+      "grad_norm": 0.046379491686820984,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 93111
+    },
+    {
+      "epoch": 9.613256246128433,
+      "grad_norm": 0.04021504893898964,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 93114
+    },
+    {
+      "epoch": 9.613565971505265,
+      "grad_norm": 0.04521257057785988,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 93117
+    },
+    {
+      "epoch": 9.613875696882097,
+      "grad_norm": 0.07617924362421036,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 93120
+    },
+    {
+      "epoch": 9.614185422258931,
+      "grad_norm": 0.07008475810289383,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 93123
+    },
+    {
+      "epoch": 9.614495147635763,
+      "grad_norm": 0.04604260250926018,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 93126
+    },
+    {
+      "epoch": 9.614804873012595,
+      "grad_norm": 0.035664889961481094,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 93129
+    },
+    {
+      "epoch": 9.615114598389429,
+      "grad_norm": 0.06766800582408905,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 93132
+    },
+    {
+      "epoch": 9.61542432376626,
+      "grad_norm": 0.08869469165802002,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 93135
+    },
+    {
+      "epoch": 9.615734049143093,
+      "grad_norm": 0.056641317903995514,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 93138
+    },
+    {
+      "epoch": 9.616043774519925,
+      "grad_norm": 0.08216816186904907,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 93141
+    },
+    {
+      "epoch": 9.616353499896759,
+      "grad_norm": 0.07369711995124817,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 93144
+    },
+    {
+      "epoch": 9.61666322527359,
+      "grad_norm": 0.08076739311218262,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 93147
+    },
+    {
+      "epoch": 9.616972950650423,
+      "grad_norm": 0.06903894990682602,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 93150
+    },
+    {
+      "epoch": 9.617282676027255,
+      "grad_norm": 0.03939782455563545,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 93153
+    },
+    {
+      "epoch": 9.617592401404089,
+      "grad_norm": 0.053101763129234314,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 93156
+    },
+    {
+      "epoch": 9.61790212678092,
+      "grad_norm": 0.0437733419239521,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 93159
+    },
+    {
+      "epoch": 9.618211852157753,
+      "grad_norm": 0.08929965645074844,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 93162
+    },
+    {
+      "epoch": 9.618521577534587,
+      "grad_norm": 0.05190601572394371,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 93165
+    },
+    {
+      "epoch": 9.618831302911419,
+      "grad_norm": 0.08993231505155563,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 93168
+    },
+    {
+      "epoch": 9.61914102828825,
+      "grad_norm": 0.06623240560293198,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 93171
+    },
+    {
+      "epoch": 9.619450753665083,
+      "grad_norm": 0.07955315709114075,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 93174
+    },
+    {
+      "epoch": 9.619760479041917,
+      "grad_norm": 0.0649283230304718,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 93177
+    },
+    {
+      "epoch": 9.620070204418749,
+      "grad_norm": 0.050466012209653854,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 93180
+    },
+    {
+      "epoch": 9.62037992979558,
+      "grad_norm": 0.04734960198402405,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 93183
+    },
+    {
+      "epoch": 9.620689655172415,
+      "grad_norm": 0.15201205015182495,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 93186
+    },
+    {
+      "epoch": 9.620999380549247,
+      "grad_norm": 0.09963729977607727,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 93189
+    },
+    {
+      "epoch": 9.621309105926079,
+      "grad_norm": 0.09011012315750122,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 93192
+    },
+    {
+      "epoch": 9.62161883130291,
+      "grad_norm": 0.09119027107954025,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 93195
+    },
+    {
+      "epoch": 9.621928556679745,
+      "grad_norm": 0.06065910682082176,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 93198
+    },
+    {
+      "epoch": 9.622238282056577,
+      "grad_norm": 0.06997659057378769,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 93201
+    },
+    {
+      "epoch": 9.622548007433409,
+      "grad_norm": 0.06493129581212997,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 93204
+    },
+    {
+      "epoch": 9.622857732810242,
+      "grad_norm": 0.06256227195262909,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 93207
+    },
+    {
+      "epoch": 9.623167458187075,
+      "grad_norm": 0.055494748055934906,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 93210
+    },
+    {
+      "epoch": 9.623477183563907,
+      "grad_norm": 0.1232457309961319,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 93213
+    },
+    {
+      "epoch": 9.623786908940739,
+      "grad_norm": 0.07807248085737228,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 93216
+    },
+    {
+      "epoch": 9.624096634317572,
+      "grad_norm": 0.06218323856592178,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 93219
+    },
+    {
+      "epoch": 9.624406359694405,
+      "grad_norm": 0.09360960870981216,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 93222
+    },
+    {
+      "epoch": 9.624716085071237,
+      "grad_norm": 0.040567923337221146,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 93225
+    },
+    {
+      "epoch": 9.625025810448069,
+      "grad_norm": 0.06413479149341583,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 93228
+    },
+    {
+      "epoch": 9.625335535824902,
+      "grad_norm": 0.04551082104444504,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 93231
+    },
+    {
+      "epoch": 9.625645261201734,
+      "grad_norm": 0.04174100235104561,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 93234
+    },
+    {
+      "epoch": 9.625954986578567,
+      "grad_norm": 0.06028440222144127,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 93237
+    },
+    {
+      "epoch": 9.6262647119554,
+      "grad_norm": 0.049742285162210464,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 93240
+    },
+    {
+      "epoch": 9.626574437332232,
+      "grad_norm": 0.036308467388153076,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 93243
+    },
+    {
+      "epoch": 9.626884162709064,
+      "grad_norm": 0.11060460656881332,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 93246
+    },
+    {
+      "epoch": 9.627193888085896,
+      "grad_norm": 0.05999557301402092,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 93249
+    },
+    {
+      "epoch": 9.62750361346273,
+      "grad_norm": 0.12154773622751236,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 93252
+    },
+    {
+      "epoch": 9.627813338839562,
+      "grad_norm": 0.10073697566986084,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 93255
+    },
+    {
+      "epoch": 9.628123064216394,
+      "grad_norm": 0.05522200092673302,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 93258
+    },
+    {
+      "epoch": 9.628432789593226,
+      "grad_norm": 0.04138825461268425,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 93261
+    },
+    {
+      "epoch": 9.62874251497006,
+      "grad_norm": 0.04448959603905678,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 93264
+    },
+    {
+      "epoch": 9.629052240346892,
+      "grad_norm": 0.043327491730451584,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 93267
+    },
+    {
+      "epoch": 9.629361965723724,
+      "grad_norm": 0.034704022109508514,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 93270
+    },
+    {
+      "epoch": 9.629671691100558,
+      "grad_norm": 0.0420006588101387,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 93273
+    },
+    {
+      "epoch": 9.62998141647739,
+      "grad_norm": 0.04239601641893387,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 93276
+    },
+    {
+      "epoch": 9.630291141854222,
+      "grad_norm": 0.09182067960500717,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 93279
+    },
+    {
+      "epoch": 9.630600867231054,
+      "grad_norm": 0.07504469156265259,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 93282
+    },
+    {
+      "epoch": 9.630910592607888,
+      "grad_norm": 0.0762309581041336,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 93285
+    },
+    {
+      "epoch": 9.63122031798472,
+      "grad_norm": 0.09295113384723663,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 93288
+    },
+    {
+      "epoch": 9.631530043361552,
+      "grad_norm": 0.06212402880191803,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 93291
+    },
+    {
+      "epoch": 9.631839768738386,
+      "grad_norm": 0.08962065726518631,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 93294
+    },
+    {
+      "epoch": 9.632149494115218,
+      "grad_norm": 0.04338514432311058,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 93297
+    },
+    {
+      "epoch": 9.63245921949205,
+      "grad_norm": 0.05132930725812912,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 93300
+    },
+    {
+      "epoch": 9.632768944868882,
+      "grad_norm": 0.03200831636786461,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 93303
+    },
+    {
+      "epoch": 9.633078670245716,
+      "grad_norm": 0.12237490713596344,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 93306
+    },
+    {
+      "epoch": 9.633388395622548,
+      "grad_norm": 0.05049234256148338,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 93309
+    },
+    {
+      "epoch": 9.63369812099938,
+      "grad_norm": 0.1869441121816635,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 93312
+    },
+    {
+      "epoch": 9.634007846376214,
+      "grad_norm": 0.05273398756980896,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 93315
+    },
+    {
+      "epoch": 9.634317571753046,
+      "grad_norm": 0.04563438892364502,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 93318
+    },
+    {
+      "epoch": 9.634627297129878,
+      "grad_norm": 0.04140450805425644,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 93321
+    },
+    {
+      "epoch": 9.63493702250671,
+      "grad_norm": 0.03873087465763092,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 93324
+    },
+    {
+      "epoch": 9.635246747883544,
+      "grad_norm": 0.03511977195739746,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 93327
+    },
+    {
+      "epoch": 9.635556473260376,
+      "grad_norm": 0.0669543594121933,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 93330
+    },
+    {
+      "epoch": 9.635866198637208,
+      "grad_norm": 0.14799566566944122,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 93333
+    },
+    {
+      "epoch": 9.63617592401404,
+      "grad_norm": 0.07729782164096832,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 93336
+    },
+    {
+      "epoch": 9.636485649390874,
+      "grad_norm": 0.03945566713809967,
+      "learning_rate": 0.01,
+      "loss": 1.905,
+      "step": 93339
+    },
+    {
+      "epoch": 9.636795374767706,
+      "grad_norm": 0.03338122367858887,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 93342
+    },
+    {
+      "epoch": 9.637105100144538,
+      "grad_norm": 0.06058921664953232,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 93345
+    },
+    {
+      "epoch": 9.637414825521372,
+      "grad_norm": 0.07650914788246155,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 93348
+    },
+    {
+      "epoch": 9.637724550898204,
+      "grad_norm": 0.06135985255241394,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 93351
+    },
+    {
+      "epoch": 9.638034276275036,
+      "grad_norm": 0.0720086619257927,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 93354
+    },
+    {
+      "epoch": 9.638344001651868,
+      "grad_norm": 0.12504549324512482,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 93357
+    },
+    {
+      "epoch": 9.638653727028702,
+      "grad_norm": 0.058934710919857025,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 93360
+    },
+    {
+      "epoch": 9.638963452405534,
+      "grad_norm": 0.06218874454498291,
+      "learning_rate": 0.01,
+      "loss": 1.9068,
+      "step": 93363
+    },
+    {
+      "epoch": 9.639273177782366,
+      "grad_norm": 0.06695770472288132,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 93366
+    },
+    {
+      "epoch": 9.639582903159198,
+      "grad_norm": 0.05433553829789162,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 93369
+    },
+    {
+      "epoch": 9.639892628536032,
+      "grad_norm": 0.05624277889728546,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 93372
+    },
+    {
+      "epoch": 9.640202353912864,
+      "grad_norm": 0.03462133929133415,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 93375
+    },
+    {
+      "epoch": 9.640512079289696,
+      "grad_norm": 0.04203572869300842,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 93378
+    },
+    {
+      "epoch": 9.64082180466653,
+      "grad_norm": 0.09270434826612473,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 93381
+    },
+    {
+      "epoch": 9.641131530043362,
+      "grad_norm": 0.04383547976613045,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 93384
+    },
+    {
+      "epoch": 9.641441255420194,
+      "grad_norm": 0.041868455708026886,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 93387
+    },
+    {
+      "epoch": 9.641750980797026,
+      "grad_norm": 0.04283531755208969,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 93390
+    },
+    {
+      "epoch": 9.64206070617386,
+      "grad_norm": 0.049920182675123215,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 93393
+    },
+    {
+      "epoch": 9.642370431550692,
+      "grad_norm": 0.12349256128072739,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 93396
+    },
+    {
+      "epoch": 9.642680156927524,
+      "grad_norm": 0.03577020764350891,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 93399
+    },
+    {
+      "epoch": 9.642989882304358,
+      "grad_norm": 0.03897339478135109,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 93402
+    },
+    {
+      "epoch": 9.64329960768119,
+      "grad_norm": 0.08809322863817215,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 93405
+    },
+    {
+      "epoch": 9.643609333058022,
+      "grad_norm": 0.09816404432058334,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 93408
+    },
+    {
+      "epoch": 9.643919058434854,
+      "grad_norm": 0.07009526342153549,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 93411
+    },
+    {
+      "epoch": 9.644228783811688,
+      "grad_norm": 0.04368029162287712,
+      "learning_rate": 0.01,
+      "loss": 1.8955,
+      "step": 93414
+    },
+    {
+      "epoch": 9.64453850918852,
+      "grad_norm": 0.03799109160900116,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 93417
+    },
+    {
+      "epoch": 9.644848234565352,
+      "grad_norm": 0.045468222349882126,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 93420
+    },
+    {
+      "epoch": 9.645157959942185,
+      "grad_norm": 0.033892299979925156,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 93423
+    },
+    {
+      "epoch": 9.645467685319018,
+      "grad_norm": 0.03280634060502052,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 93426
+    },
+    {
+      "epoch": 9.64577741069585,
+      "grad_norm": 0.032541584223508835,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 93429
+    },
+    {
+      "epoch": 9.646087136072682,
+      "grad_norm": 0.0309646837413311,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 93432
+    },
+    {
+      "epoch": 9.646396861449515,
+      "grad_norm": 0.11828819662332535,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 93435
+    },
+    {
+      "epoch": 9.646706586826348,
+      "grad_norm": 0.10888717323541641,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 93438
+    },
+    {
+      "epoch": 9.64701631220318,
+      "grad_norm": 0.0515931062400341,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 93441
+    },
+    {
+      "epoch": 9.647326037580012,
+      "grad_norm": 0.051003944128751755,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 93444
+    },
+    {
+      "epoch": 9.647635762956845,
+      "grad_norm": 0.06467358022928238,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 93447
+    },
+    {
+      "epoch": 9.647945488333677,
+      "grad_norm": 0.041775960475206375,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 93450
+    },
+    {
+      "epoch": 9.64825521371051,
+      "grad_norm": 0.03359523043036461,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 93453
+    },
+    {
+      "epoch": 9.648564939087343,
+      "grad_norm": 0.05004175379872322,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 93456
+    },
+    {
+      "epoch": 9.648874664464175,
+      "grad_norm": 0.07100548595190048,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 93459
+    },
+    {
+      "epoch": 9.649184389841007,
+      "grad_norm": 0.05144869163632393,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 93462
+    },
+    {
+      "epoch": 9.64949411521784,
+      "grad_norm": 0.1208404004573822,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 93465
+    },
+    {
+      "epoch": 9.649803840594673,
+      "grad_norm": 0.04204929992556572,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 93468
+    },
+    {
+      "epoch": 9.650113565971505,
+      "grad_norm": 0.11732341349124908,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 93471
+    },
+    {
+      "epoch": 9.650423291348337,
+      "grad_norm": 0.03591461852192879,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 93474
+    },
+    {
+      "epoch": 9.65073301672517,
+      "grad_norm": 0.09749674052000046,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 93477
+    },
+    {
+      "epoch": 9.651042742102003,
+      "grad_norm": 0.04847512021660805,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 93480
+    },
+    {
+      "epoch": 9.651352467478835,
+      "grad_norm": 0.06401492655277252,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 93483
+    },
+    {
+      "epoch": 9.651662192855667,
+      "grad_norm": 0.13890819251537323,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 93486
+    },
+    {
+      "epoch": 9.651971918232501,
+      "grad_norm": 0.034448061138391495,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 93489
+    },
+    {
+      "epoch": 9.652281643609333,
+      "grad_norm": 0.11787496507167816,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 93492
+    },
+    {
+      "epoch": 9.652591368986165,
+      "grad_norm": 0.05250073969364166,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 93495
+    },
+    {
+      "epoch": 9.652901094362997,
+      "grad_norm": 0.04947948083281517,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 93498
+    },
+    {
+      "epoch": 9.653210819739831,
+      "grad_norm": 0.04423551633954048,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 93501
+    },
+    {
+      "epoch": 9.653520545116663,
+      "grad_norm": 0.041208140552043915,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 93504
+    },
+    {
+      "epoch": 9.653830270493495,
+      "grad_norm": 0.13026712834835052,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 93507
+    },
+    {
+      "epoch": 9.65413999587033,
+      "grad_norm": 0.07483094930648804,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 93510
+    },
+    {
+      "epoch": 9.654449721247161,
+      "grad_norm": 0.05455175042152405,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 93513
+    },
+    {
+      "epoch": 9.654759446623993,
+      "grad_norm": 0.0686800554394722,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 93516
+    },
+    {
+      "epoch": 9.655069172000825,
+      "grad_norm": 0.0592319555580616,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 93519
+    },
+    {
+      "epoch": 9.655378897377659,
+      "grad_norm": 0.03538976609706879,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 93522
+    },
+    {
+      "epoch": 9.655688622754491,
+      "grad_norm": 0.039937421679496765,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 93525
+    },
+    {
+      "epoch": 9.655998348131323,
+      "grad_norm": 0.04662489891052246,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 93528
+    },
+    {
+      "epoch": 9.656308073508157,
+      "grad_norm": 0.1549932211637497,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 93531
+    },
+    {
+      "epoch": 9.656617798884989,
+      "grad_norm": 0.07844512164592743,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 93534
+    },
+    {
+      "epoch": 9.656927524261821,
+      "grad_norm": 0.05962245911359787,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 93537
+    },
+    {
+      "epoch": 9.657237249638653,
+      "grad_norm": 0.06986347585916519,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 93540
+    },
+    {
+      "epoch": 9.657546975015487,
+      "grad_norm": 0.045444563031196594,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 93543
+    },
+    {
+      "epoch": 9.657856700392319,
+      "grad_norm": 0.042958684265613556,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 93546
+    },
+    {
+      "epoch": 9.658166425769151,
+      "grad_norm": 0.05604341998696327,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 93549
+    },
+    {
+      "epoch": 9.658476151145983,
+      "grad_norm": 0.05136589705944061,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 93552
+    },
+    {
+      "epoch": 9.658785876522817,
+      "grad_norm": 0.06568500399589539,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 93555
+    },
+    {
+      "epoch": 9.659095601899649,
+      "grad_norm": 0.042793937027454376,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 93558
+    },
+    {
+      "epoch": 9.659405327276481,
+      "grad_norm": 0.08512739837169647,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 93561
+    },
+    {
+      "epoch": 9.659715052653315,
+      "grad_norm": 0.059787552803754807,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 93564
+    },
+    {
+      "epoch": 9.660024778030147,
+      "grad_norm": 0.095461905002594,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 93567
+    },
+    {
+      "epoch": 9.660334503406979,
+      "grad_norm": 0.11361975967884064,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 93570
+    },
+    {
+      "epoch": 9.660644228783811,
+      "grad_norm": 0.055556997656822205,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 93573
+    },
+    {
+      "epoch": 9.660953954160645,
+      "grad_norm": 0.043150875717401505,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 93576
+    },
+    {
+      "epoch": 9.661263679537477,
+      "grad_norm": 0.0998530462384224,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 93579
+    },
+    {
+      "epoch": 9.661573404914309,
+      "grad_norm": 0.045185189694166183,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 93582
+    },
+    {
+      "epoch": 9.661883130291141,
+      "grad_norm": 0.14654050767421722,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 93585
+    },
+    {
+      "epoch": 9.662192855667975,
+      "grad_norm": 0.10469841212034225,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 93588
+    },
+    {
+      "epoch": 9.662502581044807,
+      "grad_norm": 0.09240784496068954,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 93591
+    },
+    {
+      "epoch": 9.662812306421639,
+      "grad_norm": 0.07269421219825745,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 93594
+    },
+    {
+      "epoch": 9.663122031798473,
+      "grad_norm": 0.042303770780563354,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 93597
+    },
+    {
+      "epoch": 9.663431757175305,
+      "grad_norm": 0.03580670431256294,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 93600
+    },
+    {
+      "epoch": 9.663741482552137,
+      "grad_norm": 0.03630633279681206,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 93603
+    },
+    {
+      "epoch": 9.664051207928969,
+      "grad_norm": 0.0431305356323719,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 93606
+    },
+    {
+      "epoch": 9.664360933305803,
+      "grad_norm": 0.044870369136333466,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 93609
+    },
+    {
+      "epoch": 9.664670658682635,
+      "grad_norm": 0.12574565410614014,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 93612
+    },
+    {
+      "epoch": 9.664980384059467,
+      "grad_norm": 0.12211044132709503,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 93615
+    },
+    {
+      "epoch": 9.6652901094363,
+      "grad_norm": 0.1176372617483139,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 93618
+    },
+    {
+      "epoch": 9.665599834813133,
+      "grad_norm": 0.0372295081615448,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 93621
+    },
+    {
+      "epoch": 9.665909560189965,
+      "grad_norm": 0.040826380252838135,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 93624
+    },
+    {
+      "epoch": 9.666219285566797,
+      "grad_norm": 0.030447613447904587,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 93627
+    },
+    {
+      "epoch": 9.66652901094363,
+      "grad_norm": 0.04039299115538597,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 93630
+    },
+    {
+      "epoch": 9.666838736320463,
+      "grad_norm": 0.08394279330968857,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 93633
+    },
+    {
+      "epoch": 9.667148461697295,
+      "grad_norm": 0.04779527708888054,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 93636
+    },
+    {
+      "epoch": 9.667458187074129,
+      "grad_norm": 0.05684870854020119,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 93639
+    },
+    {
+      "epoch": 9.66776791245096,
+      "grad_norm": 0.04491376876831055,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 93642
+    },
+    {
+      "epoch": 9.668077637827793,
+      "grad_norm": 0.035552386194467545,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 93645
+    },
+    {
+      "epoch": 9.668387363204625,
+      "grad_norm": 0.03917543217539787,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 93648
+    },
+    {
+      "epoch": 9.668697088581458,
+      "grad_norm": 0.05415418744087219,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 93651
+    },
+    {
+      "epoch": 9.66900681395829,
+      "grad_norm": 0.13435325026512146,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 93654
+    },
+    {
+      "epoch": 9.669316539335123,
+      "grad_norm": 0.13546143472194672,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 93657
+    },
+    {
+      "epoch": 9.669626264711955,
+      "grad_norm": 0.11451159417629242,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 93660
+    },
+    {
+      "epoch": 9.669935990088788,
+      "grad_norm": 0.05074388161301613,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 93663
+    },
+    {
+      "epoch": 9.67024571546562,
+      "grad_norm": 0.07373972982168198,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 93666
+    },
+    {
+      "epoch": 9.670555440842453,
+      "grad_norm": 0.08737852424383163,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 93669
+    },
+    {
+      "epoch": 9.670865166219286,
+      "grad_norm": 0.05184822157025337,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 93672
+    },
+    {
+      "epoch": 9.671174891596118,
+      "grad_norm": 0.08283158391714096,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 93675
+    },
+    {
+      "epoch": 9.67148461697295,
+      "grad_norm": 0.10233674198389053,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 93678
+    },
+    {
+      "epoch": 9.671794342349783,
+      "grad_norm": 0.041018836200237274,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 93681
+    },
+    {
+      "epoch": 9.672104067726616,
+      "grad_norm": 0.10328731685876846,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 93684
+    },
+    {
+      "epoch": 9.672413793103448,
+      "grad_norm": 0.04173845052719116,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 93687
+    },
+    {
+      "epoch": 9.67272351848028,
+      "grad_norm": 0.04470297694206238,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 93690
+    },
+    {
+      "epoch": 9.673033243857112,
+      "grad_norm": 0.04156685620546341,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 93693
+    },
+    {
+      "epoch": 9.673342969233946,
+      "grad_norm": 0.04977422207593918,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 93696
+    },
+    {
+      "epoch": 9.673652694610778,
+      "grad_norm": 0.034027110785245895,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 93699
+    },
+    {
+      "epoch": 9.67396241998761,
+      "grad_norm": 0.04569841921329498,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 93702
+    },
+    {
+      "epoch": 9.674272145364444,
+      "grad_norm": 0.04238857328891754,
+      "learning_rate": 0.01,
+      "loss": 1.8984,
+      "step": 93705
+    },
+    {
+      "epoch": 9.674581870741276,
+      "grad_norm": 0.04386452957987785,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 93708
+    },
+    {
+      "epoch": 9.674891596118108,
+      "grad_norm": 0.043252211064100266,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 93711
+    },
+    {
+      "epoch": 9.67520132149494,
+      "grad_norm": 0.10518099367618561,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 93714
+    },
+    {
+      "epoch": 9.675511046871774,
+      "grad_norm": 0.095424585044384,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 93717
+    },
+    {
+      "epoch": 9.675820772248606,
+      "grad_norm": 0.09081821888685226,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 93720
+    },
+    {
+      "epoch": 9.676130497625438,
+      "grad_norm": 0.17587430775165558,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 93723
+    },
+    {
+      "epoch": 9.676440223002272,
+      "grad_norm": 0.0808170810341835,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 93726
+    },
+    {
+      "epoch": 9.676749948379104,
+      "grad_norm": 0.060308847576379776,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 93729
+    },
+    {
+      "epoch": 9.677059673755936,
+      "grad_norm": 0.03717552870512009,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 93732
+    },
+    {
+      "epoch": 9.677369399132768,
+      "grad_norm": 0.033456869423389435,
+      "learning_rate": 0.01,
+      "loss": 1.9771,
+      "step": 93735
+    },
+    {
+      "epoch": 9.677679124509602,
+      "grad_norm": 0.04936378076672554,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 93738
+    },
+    {
+      "epoch": 9.677988849886434,
+      "grad_norm": 0.05672374367713928,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 93741
+    },
+    {
+      "epoch": 9.678298575263266,
+      "grad_norm": 0.05045805498957634,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 93744
+    },
+    {
+      "epoch": 9.6786083006401,
+      "grad_norm": 0.0464802086353302,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 93747
+    },
+    {
+      "epoch": 9.678918026016932,
+      "grad_norm": 0.09897595643997192,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 93750
+    },
+    {
+      "epoch": 9.679227751393764,
+      "grad_norm": 0.04183182865381241,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 93753
+    },
+    {
+      "epoch": 9.679537476770596,
+      "grad_norm": 0.07497096061706543,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 93756
+    },
+    {
+      "epoch": 9.67984720214743,
+      "grad_norm": 0.09940138459205627,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 93759
+    },
+    {
+      "epoch": 9.680156927524262,
+      "grad_norm": 0.0419076569378376,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 93762
+    },
+    {
+      "epoch": 9.680466652901094,
+      "grad_norm": 0.10052657127380371,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 93765
+    },
+    {
+      "epoch": 9.680776378277926,
+      "grad_norm": 0.12983714044094086,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 93768
+    },
+    {
+      "epoch": 9.68108610365476,
+      "grad_norm": 0.06584013253450394,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 93771
+    },
+    {
+      "epoch": 9.681395829031592,
+      "grad_norm": 0.0655849277973175,
+      "learning_rate": 0.01,
+      "loss": 1.9017,
+      "step": 93774
+    },
+    {
+      "epoch": 9.681705554408424,
+      "grad_norm": 0.05604313686490059,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 93777
+    },
+    {
+      "epoch": 9.682015279785258,
+      "grad_norm": 0.049449238926172256,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 93780
+    },
+    {
+      "epoch": 9.68232500516209,
+      "grad_norm": 0.032564688473939896,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 93783
+    },
+    {
+      "epoch": 9.682634730538922,
+      "grad_norm": 0.033273547887802124,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 93786
+    },
+    {
+      "epoch": 9.682944455915754,
+      "grad_norm": 0.033502086997032166,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 93789
+    },
+    {
+      "epoch": 9.683254181292588,
+      "grad_norm": 0.12671968340873718,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 93792
+    },
+    {
+      "epoch": 9.68356390666942,
+      "grad_norm": 0.045263972133398056,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 93795
+    },
+    {
+      "epoch": 9.683873632046252,
+      "grad_norm": 0.08929682523012161,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 93798
+    },
+    {
+      "epoch": 9.684183357423084,
+      "grad_norm": 0.05344196781516075,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 93801
+    },
+    {
+      "epoch": 9.684493082799918,
+      "grad_norm": 0.04417494311928749,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 93804
+    },
+    {
+      "epoch": 9.68480280817675,
+      "grad_norm": 0.04135802015662193,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 93807
+    },
+    {
+      "epoch": 9.685112533553582,
+      "grad_norm": 0.07783691585063934,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 93810
+    },
+    {
+      "epoch": 9.685422258930416,
+      "grad_norm": 0.08820218592882156,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 93813
+    },
+    {
+      "epoch": 9.685731984307248,
+      "grad_norm": 0.03493189811706543,
+      "learning_rate": 0.01,
+      "loss": 1.9053,
+      "step": 93816
+    },
+    {
+      "epoch": 9.68604170968408,
+      "grad_norm": 0.06502649933099747,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 93819
+    },
+    {
+      "epoch": 9.686351435060912,
+      "grad_norm": 0.08531218767166138,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 93822
+    },
+    {
+      "epoch": 9.686661160437746,
+      "grad_norm": 0.12796832621097565,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 93825
+    },
+    {
+      "epoch": 9.686970885814578,
+      "grad_norm": 0.07860841602087021,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 93828
+    },
+    {
+      "epoch": 9.68728061119141,
+      "grad_norm": 0.04484981298446655,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 93831
+    },
+    {
+      "epoch": 9.687590336568244,
+      "grad_norm": 0.04047702997922897,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 93834
+    },
+    {
+      "epoch": 9.687900061945076,
+      "grad_norm": 0.04358268901705742,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 93837
+    },
+    {
+      "epoch": 9.688209787321908,
+      "grad_norm": 0.12023859471082687,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 93840
+    },
+    {
+      "epoch": 9.68851951269874,
+      "grad_norm": 0.038970865309238434,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 93843
+    },
+    {
+      "epoch": 9.688829238075574,
+      "grad_norm": 0.13150472939014435,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 93846
+    },
+    {
+      "epoch": 9.689138963452406,
+      "grad_norm": 0.035846102982759476,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 93849
+    },
+    {
+      "epoch": 9.689448688829238,
+      "grad_norm": 0.0676359161734581,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 93852
+    },
+    {
+      "epoch": 9.689758414206072,
+      "grad_norm": 0.052405230700969696,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 93855
+    },
+    {
+      "epoch": 9.690068139582904,
+      "grad_norm": 0.054263848811388016,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 93858
+    },
+    {
+      "epoch": 9.690377864959736,
+      "grad_norm": 0.09067951887845993,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 93861
+    },
+    {
+      "epoch": 9.690687590336568,
+      "grad_norm": 0.11018327623605728,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 93864
+    },
+    {
+      "epoch": 9.690997315713402,
+      "grad_norm": 0.03586273640394211,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 93867
+    },
+    {
+      "epoch": 9.691307041090234,
+      "grad_norm": 0.08600044995546341,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 93870
+    },
+    {
+      "epoch": 9.691616766467066,
+      "grad_norm": 0.06296738237142563,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 93873
+    },
+    {
+      "epoch": 9.691926491843898,
+      "grad_norm": 0.10977789759635925,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 93876
+    },
+    {
+      "epoch": 9.692236217220731,
+      "grad_norm": 0.09508870542049408,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 93879
+    },
+    {
+      "epoch": 9.692545942597564,
+      "grad_norm": 0.04756419360637665,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 93882
+    },
+    {
+      "epoch": 9.692855667974396,
+      "grad_norm": 0.049777496606111526,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 93885
+    },
+    {
+      "epoch": 9.69316539335123,
+      "grad_norm": 0.03871287778019905,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 93888
+    },
+    {
+      "epoch": 9.693475118728061,
+      "grad_norm": 0.07852483540773392,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 93891
+    },
+    {
+      "epoch": 9.693784844104893,
+      "grad_norm": 0.07227092236280441,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 93894
+    },
+    {
+      "epoch": 9.694094569481726,
+      "grad_norm": 0.0830816924571991,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 93897
+    },
+    {
+      "epoch": 9.69440429485856,
+      "grad_norm": 0.06940535455942154,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 93900
+    },
+    {
+      "epoch": 9.694714020235391,
+      "grad_norm": 0.07247503846883774,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 93903
+    },
+    {
+      "epoch": 9.695023745612223,
+      "grad_norm": 0.08987094461917877,
+      "learning_rate": 0.01,
+      "loss": 1.9027,
+      "step": 93906
+    },
+    {
+      "epoch": 9.695333470989056,
+      "grad_norm": 0.05902135744690895,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 93909
+    },
+    {
+      "epoch": 9.69564319636589,
+      "grad_norm": 0.08232377469539642,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 93912
+    },
+    {
+      "epoch": 9.695952921742721,
+      "grad_norm": 0.07965484261512756,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 93915
+    },
+    {
+      "epoch": 9.696262647119553,
+      "grad_norm": 0.03626954182982445,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 93918
+    },
+    {
+      "epoch": 9.696572372496387,
+      "grad_norm": 0.11025188863277435,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 93921
+    },
+    {
+      "epoch": 9.69688209787322,
+      "grad_norm": 0.06925832480192184,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 93924
+    },
+    {
+      "epoch": 9.697191823250051,
+      "grad_norm": 0.09669098258018494,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 93927
+    },
+    {
+      "epoch": 9.697501548626883,
+      "grad_norm": 0.06018374115228653,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 93930
+    },
+    {
+      "epoch": 9.697811274003717,
+      "grad_norm": 0.03994041308760643,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 93933
+    },
+    {
+      "epoch": 9.69812099938055,
+      "grad_norm": 0.046831220388412476,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 93936
+    },
+    {
+      "epoch": 9.698430724757381,
+      "grad_norm": 0.07097730785608292,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 93939
+    },
+    {
+      "epoch": 9.698740450134215,
+      "grad_norm": 0.15823347866535187,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 93942
+    },
+    {
+      "epoch": 9.699050175511047,
+      "grad_norm": 0.05094249173998833,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 93945
+    },
+    {
+      "epoch": 9.69935990088788,
+      "grad_norm": 0.11403997242450714,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 93948
+    },
+    {
+      "epoch": 9.699669626264711,
+      "grad_norm": 0.04830975458025932,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 93951
+    },
+    {
+      "epoch": 9.699979351641545,
+      "grad_norm": 0.055670302361249924,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 93954
+    },
+    {
+      "epoch": 9.700289077018377,
+      "grad_norm": 0.06251242011785507,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 93957
+    },
+    {
+      "epoch": 9.70059880239521,
+      "grad_norm": 0.0666978657245636,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 93960
+    },
+    {
+      "epoch": 9.700908527772043,
+      "grad_norm": 0.043735913932323456,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 93963
+    },
+    {
+      "epoch": 9.701218253148875,
+      "grad_norm": 0.03567013889551163,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 93966
+    },
+    {
+      "epoch": 9.701527978525707,
+      "grad_norm": 0.12823991477489471,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 93969
+    },
+    {
+      "epoch": 9.70183770390254,
+      "grad_norm": 0.11092595010995865,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 93972
+    },
+    {
+      "epoch": 9.702147429279373,
+      "grad_norm": 0.07530339807271957,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 93975
+    },
+    {
+      "epoch": 9.702457154656205,
+      "grad_norm": 0.040059320628643036,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 93978
+    },
+    {
+      "epoch": 9.702766880033037,
+      "grad_norm": 0.04048408940434456,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 93981
+    },
+    {
+      "epoch": 9.70307660540987,
+      "grad_norm": 0.038570586591959,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 93984
+    },
+    {
+      "epoch": 9.703386330786703,
+      "grad_norm": 0.04940098896622658,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 93987
+    },
+    {
+      "epoch": 9.703696056163535,
+      "grad_norm": 0.056408241391181946,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 93990
+    },
+    {
+      "epoch": 9.704005781540367,
+      "grad_norm": 0.042289089411497116,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 93993
+    },
+    {
+      "epoch": 9.704315506917201,
+      "grad_norm": 0.040116313844919205,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 93996
+    },
+    {
+      "epoch": 9.704625232294033,
+      "grad_norm": 0.09464330226182938,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 93999
+    },
+    {
+      "epoch": 9.704934957670865,
+      "grad_norm": 0.1511952131986618,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 94002
+    },
+    {
+      "epoch": 9.705244683047697,
+      "grad_norm": 0.09546513110399246,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 94005
+    },
+    {
+      "epoch": 9.70555440842453,
+      "grad_norm": 0.05483248457312584,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 94008
+    },
+    {
+      "epoch": 9.705864133801363,
+      "grad_norm": 0.036017656326293945,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 94011
+    },
+    {
+      "epoch": 9.706173859178195,
+      "grad_norm": 0.04008268564939499,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 94014
+    },
+    {
+      "epoch": 9.706483584555027,
+      "grad_norm": 0.03564837574958801,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 94017
+    },
+    {
+      "epoch": 9.70679330993186,
+      "grad_norm": 0.0589480847120285,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 94020
+    },
+    {
+      "epoch": 9.707103035308693,
+      "grad_norm": 0.07352419942617416,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 94023
+    },
+    {
+      "epoch": 9.707412760685525,
+      "grad_norm": 0.06446439027786255,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 94026
+    },
+    {
+      "epoch": 9.707722486062359,
+      "grad_norm": 0.0859951302409172,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 94029
+    },
+    {
+      "epoch": 9.70803221143919,
+      "grad_norm": 0.1257733255624771,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 94032
+    },
+    {
+      "epoch": 9.708341936816023,
+      "grad_norm": 0.0344783179461956,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 94035
+    },
+    {
+      "epoch": 9.708651662192855,
+      "grad_norm": 0.06627075374126434,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 94038
+    },
+    {
+      "epoch": 9.708961387569689,
+      "grad_norm": 0.0375131331384182,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 94041
+    },
+    {
+      "epoch": 9.70927111294652,
+      "grad_norm": 0.03659706190228462,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 94044
+    },
+    {
+      "epoch": 9.709580838323353,
+      "grad_norm": 0.06533805280923843,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 94047
+    },
+    {
+      "epoch": 9.709890563700187,
+      "grad_norm": 0.09291239827871323,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 94050
+    },
+    {
+      "epoch": 9.710200289077019,
+      "grad_norm": 0.06271101534366608,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 94053
+    },
+    {
+      "epoch": 9.71051001445385,
+      "grad_norm": 0.05989930406212807,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 94056
+    },
+    {
+      "epoch": 9.710819739830683,
+      "grad_norm": 0.10313070565462112,
+      "learning_rate": 0.01,
+      "loss": 1.9821,
+      "step": 94059
+    },
+    {
+      "epoch": 9.711129465207517,
+      "grad_norm": 0.12868840992450714,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 94062
+    },
+    {
+      "epoch": 9.711439190584349,
+      "grad_norm": 0.03902678191661835,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 94065
+    },
+    {
+      "epoch": 9.71174891596118,
+      "grad_norm": 0.03207237645983696,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 94068
+    },
+    {
+      "epoch": 9.712058641338015,
+      "grad_norm": 0.04138148948550224,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 94071
+    },
+    {
+      "epoch": 9.712368366714847,
+      "grad_norm": 0.048512060195207596,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 94074
+    },
+    {
+      "epoch": 9.712678092091679,
+      "grad_norm": 0.03996929153800011,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 94077
+    },
+    {
+      "epoch": 9.71298781746851,
+      "grad_norm": 0.042945004999637604,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 94080
+    },
+    {
+      "epoch": 9.713297542845345,
+      "grad_norm": 0.05019443482160568,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 94083
+    },
+    {
+      "epoch": 9.713607268222177,
+      "grad_norm": 0.059888970106840134,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 94086
+    },
+    {
+      "epoch": 9.713916993599009,
+      "grad_norm": 0.04927165061235428,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 94089
+    },
+    {
+      "epoch": 9.71422671897584,
+      "grad_norm": 0.039205461740493774,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 94092
+    },
+    {
+      "epoch": 9.714536444352674,
+      "grad_norm": 0.04728487879037857,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 94095
+    },
+    {
+      "epoch": 9.714846169729507,
+      "grad_norm": 0.03699108585715294,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 94098
+    },
+    {
+      "epoch": 9.715155895106339,
+      "grad_norm": 0.1421017348766327,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 94101
+    },
+    {
+      "epoch": 9.715465620483172,
+      "grad_norm": 0.041983768343925476,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 94104
+    },
+    {
+      "epoch": 9.715775345860004,
+      "grad_norm": 0.10116805136203766,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 94107
+    },
+    {
+      "epoch": 9.716085071236837,
+      "grad_norm": 0.06465780735015869,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 94110
+    },
+    {
+      "epoch": 9.716394796613669,
+      "grad_norm": 0.06708145141601562,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 94113
+    },
+    {
+      "epoch": 9.716704521990502,
+      "grad_norm": 0.06895887106657028,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 94116
+    },
+    {
+      "epoch": 9.717014247367334,
+      "grad_norm": 0.1283697783946991,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 94119
+    },
+    {
+      "epoch": 9.717323972744166,
+      "grad_norm": 0.044195953756570816,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 94122
+    },
+    {
+      "epoch": 9.717633698120999,
+      "grad_norm": 0.034305259585380554,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 94125
+    },
+    {
+      "epoch": 9.717943423497832,
+      "grad_norm": 0.07333043962717056,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 94128
+    },
+    {
+      "epoch": 9.718253148874664,
+      "grad_norm": 0.04106907546520233,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 94131
+    },
+    {
+      "epoch": 9.718562874251496,
+      "grad_norm": 0.04604107141494751,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 94134
+    },
+    {
+      "epoch": 9.71887259962833,
+      "grad_norm": 0.050639402121305466,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 94137
+    },
+    {
+      "epoch": 9.719182325005162,
+      "grad_norm": 0.06188986077904701,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 94140
+    },
+    {
+      "epoch": 9.719492050381994,
+      "grad_norm": 0.07035218924283981,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 94143
+    },
+    {
+      "epoch": 9.719801775758826,
+      "grad_norm": 0.06794343888759613,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 94146
+    },
+    {
+      "epoch": 9.72011150113566,
+      "grad_norm": 0.08361504226922989,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 94149
+    },
+    {
+      "epoch": 9.720421226512492,
+      "grad_norm": 0.11391738802194595,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 94152
+    },
+    {
+      "epoch": 9.720730951889324,
+      "grad_norm": 0.08625376224517822,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 94155
+    },
+    {
+      "epoch": 9.721040677266158,
+      "grad_norm": 0.1078452318906784,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 94158
+    },
+    {
+      "epoch": 9.72135040264299,
+      "grad_norm": 0.053820785135030746,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 94161
+    },
+    {
+      "epoch": 9.721660128019822,
+      "grad_norm": 0.05952325463294983,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 94164
+    },
+    {
+      "epoch": 9.721969853396654,
+      "grad_norm": 0.05312516167759895,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 94167
+    },
+    {
+      "epoch": 9.722279578773488,
+      "grad_norm": 0.0658746063709259,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 94170
+    },
+    {
+      "epoch": 9.72258930415032,
+      "grad_norm": 0.05180962756276131,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 94173
+    },
+    {
+      "epoch": 9.722899029527152,
+      "grad_norm": 0.053514860570430756,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 94176
+    },
+    {
+      "epoch": 9.723208754903986,
+      "grad_norm": 0.09407076984643936,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 94179
+    },
+    {
+      "epoch": 9.723518480280818,
+      "grad_norm": 0.10801953822374344,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 94182
+    },
+    {
+      "epoch": 9.72382820565765,
+      "grad_norm": 0.04985484853386879,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 94185
+    },
+    {
+      "epoch": 9.724137931034482,
+      "grad_norm": 0.04344812035560608,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 94188
+    },
+    {
+      "epoch": 9.724447656411316,
+      "grad_norm": 0.039459019899368286,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 94191
+    },
+    {
+      "epoch": 9.724757381788148,
+      "grad_norm": 0.03749649599194527,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 94194
+    },
+    {
+      "epoch": 9.72506710716498,
+      "grad_norm": 0.036703795194625854,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 94197
+    },
+    {
+      "epoch": 9.725376832541812,
+      "grad_norm": 0.03291583061218262,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 94200
+    },
+    {
+      "epoch": 9.725686557918646,
+      "grad_norm": 0.1574249416589737,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 94203
+    },
+    {
+      "epoch": 9.725996283295478,
+      "grad_norm": 0.15639059245586395,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 94206
+    },
+    {
+      "epoch": 9.72630600867231,
+      "grad_norm": 0.05640319734811783,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 94209
+    },
+    {
+      "epoch": 9.726615734049144,
+      "grad_norm": 0.069240041077137,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 94212
+    },
+    {
+      "epoch": 9.726925459425976,
+      "grad_norm": 0.08438346534967422,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 94215
+    },
+    {
+      "epoch": 9.727235184802808,
+      "grad_norm": 0.03589455783367157,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 94218
+    },
+    {
+      "epoch": 9.72754491017964,
+      "grad_norm": 0.06589308381080627,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 94221
+    },
+    {
+      "epoch": 9.727854635556474,
+      "grad_norm": 0.04225952550768852,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 94224
+    },
+    {
+      "epoch": 9.728164360933306,
+      "grad_norm": 0.0393117219209671,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 94227
+    },
+    {
+      "epoch": 9.728474086310138,
+      "grad_norm": 0.03414738550782204,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 94230
+    },
+    {
+      "epoch": 9.72878381168697,
+      "grad_norm": 0.043435826897621155,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 94233
+    },
+    {
+      "epoch": 9.729093537063804,
+      "grad_norm": 0.05065900832414627,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 94236
+    },
+    {
+      "epoch": 9.729403262440636,
+      "grad_norm": 0.048157140612602234,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 94239
+    },
+    {
+      "epoch": 9.729712987817468,
+      "grad_norm": 0.04764106124639511,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 94242
+    },
+    {
+      "epoch": 9.730022713194302,
+      "grad_norm": 0.1098858192563057,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 94245
+    },
+    {
+      "epoch": 9.730332438571134,
+      "grad_norm": 0.06745657324790955,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 94248
+    },
+    {
+      "epoch": 9.730642163947966,
+      "grad_norm": 0.11722870171070099,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 94251
+    },
+    {
+      "epoch": 9.730951889324798,
+      "grad_norm": 0.04447784274816513,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 94254
+    },
+    {
+      "epoch": 9.731261614701632,
+      "grad_norm": 0.10149898380041122,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 94257
+    },
+    {
+      "epoch": 9.731571340078464,
+      "grad_norm": 0.0666227862238884,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 94260
+    },
+    {
+      "epoch": 9.731881065455296,
+      "grad_norm": 0.08770105987787247,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 94263
+    },
+    {
+      "epoch": 9.732190790832128,
+      "grad_norm": 0.0607566311955452,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 94266
+    },
+    {
+      "epoch": 9.732500516208962,
+      "grad_norm": 0.07751908153295517,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 94269
+    },
+    {
+      "epoch": 9.732810241585794,
+      "grad_norm": 0.04078710824251175,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 94272
+    },
+    {
+      "epoch": 9.733119966962626,
+      "grad_norm": 0.06762789934873581,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 94275
+    },
+    {
+      "epoch": 9.73342969233946,
+      "grad_norm": 0.0777420923113823,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 94278
+    },
+    {
+      "epoch": 9.733739417716292,
+      "grad_norm": 0.07008164376020432,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 94281
+    },
+    {
+      "epoch": 9.734049143093124,
+      "grad_norm": 0.08447692543268204,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 94284
+    },
+    {
+      "epoch": 9.734358868469958,
+      "grad_norm": 0.06624925881624222,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 94287
+    },
+    {
+      "epoch": 9.73466859384679,
+      "grad_norm": 0.04541005566716194,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 94290
+    },
+    {
+      "epoch": 9.734978319223622,
+      "grad_norm": 0.036812674254179,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 94293
+    },
+    {
+      "epoch": 9.735288044600454,
+      "grad_norm": 0.12216412276029587,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 94296
+    },
+    {
+      "epoch": 9.735597769977288,
+      "grad_norm": 0.06444603949785233,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 94299
+    },
+    {
+      "epoch": 9.73590749535412,
+      "grad_norm": 0.05880026891827583,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 94302
+    },
+    {
+      "epoch": 9.736217220730952,
+      "grad_norm": 0.09832989424467087,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 94305
+    },
+    {
+      "epoch": 9.736526946107784,
+      "grad_norm": 0.04116860777139664,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 94308
+    },
+    {
+      "epoch": 9.736836671484618,
+      "grad_norm": 0.08587076514959335,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 94311
+    },
+    {
+      "epoch": 9.73714639686145,
+      "grad_norm": 0.060674894601106644,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 94314
+    },
+    {
+      "epoch": 9.737456122238282,
+      "grad_norm": 0.10280962288379669,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 94317
+    },
+    {
+      "epoch": 9.737765847615115,
+      "grad_norm": 0.09550301730632782,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 94320
+    },
+    {
+      "epoch": 9.738075572991947,
+      "grad_norm": 0.04915456846356392,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 94323
+    },
+    {
+      "epoch": 9.73838529836878,
+      "grad_norm": 0.03588581085205078,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 94326
+    },
+    {
+      "epoch": 9.738695023745612,
+      "grad_norm": 0.03683014586567879,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 94329
+    },
+    {
+      "epoch": 9.739004749122445,
+      "grad_norm": 0.11679235100746155,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 94332
+    },
+    {
+      "epoch": 9.739314474499277,
+      "grad_norm": 0.0934172049164772,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 94335
+    },
+    {
+      "epoch": 9.73962419987611,
+      "grad_norm": 0.05006689950823784,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 94338
+    },
+    {
+      "epoch": 9.739933925252942,
+      "grad_norm": 0.08405818790197372,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 94341
+    },
+    {
+      "epoch": 9.740243650629775,
+      "grad_norm": 0.06463057547807693,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 94344
+    },
+    {
+      "epoch": 9.740553376006607,
+      "grad_norm": 0.05067126825451851,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 94347
+    },
+    {
+      "epoch": 9.74086310138344,
+      "grad_norm": 0.05063599720597267,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 94350
+    },
+    {
+      "epoch": 9.741172826760273,
+      "grad_norm": 0.056624624878168106,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 94353
+    },
+    {
+      "epoch": 9.741482552137105,
+      "grad_norm": 0.04202519729733467,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 94356
+    },
+    {
+      "epoch": 9.741792277513937,
+      "grad_norm": 0.0348421148955822,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 94359
+    },
+    {
+      "epoch": 9.74210200289077,
+      "grad_norm": 0.04054299741983414,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 94362
+    },
+    {
+      "epoch": 9.742411728267603,
+      "grad_norm": 0.1300070583820343,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 94365
+    },
+    {
+      "epoch": 9.742721453644435,
+      "grad_norm": 0.05025966838002205,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 94368
+    },
+    {
+      "epoch": 9.743031179021267,
+      "grad_norm": 0.10579799860715866,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 94371
+    },
+    {
+      "epoch": 9.7433409043981,
+      "grad_norm": 0.06107889860868454,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 94374
+    },
+    {
+      "epoch": 9.743650629774933,
+      "grad_norm": 0.09993497282266617,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 94377
+    },
+    {
+      "epoch": 9.743960355151765,
+      "grad_norm": 0.04587731882929802,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 94380
+    },
+    {
+      "epoch": 9.744270080528597,
+      "grad_norm": 0.0887177363038063,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 94383
+    },
+    {
+      "epoch": 9.744579805905431,
+      "grad_norm": 0.056814342737197876,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 94386
+    },
+    {
+      "epoch": 9.744889531282263,
+      "grad_norm": 0.05701100826263428,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 94389
+    },
+    {
+      "epoch": 9.745199256659095,
+      "grad_norm": 0.06427867710590363,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 94392
+    },
+    {
+      "epoch": 9.745508982035929,
+      "grad_norm": 0.0418747141957283,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 94395
+    },
+    {
+      "epoch": 9.745818707412761,
+      "grad_norm": 0.04488182067871094,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 94398
+    },
+    {
+      "epoch": 9.746128432789593,
+      "grad_norm": 0.1035093367099762,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 94401
+    },
+    {
+      "epoch": 9.746438158166425,
+      "grad_norm": 0.03518347442150116,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 94404
+    },
+    {
+      "epoch": 9.746747883543259,
+      "grad_norm": 0.05361732468008995,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 94407
+    },
+    {
+      "epoch": 9.747057608920091,
+      "grad_norm": 0.08221562951803207,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 94410
+    },
+    {
+      "epoch": 9.747367334296923,
+      "grad_norm": 0.09609878063201904,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 94413
+    },
+    {
+      "epoch": 9.747677059673755,
+      "grad_norm": 0.03668706491589546,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 94416
+    },
+    {
+      "epoch": 9.747986785050589,
+      "grad_norm": 0.04190170392394066,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 94419
+    },
+    {
+      "epoch": 9.748296510427421,
+      "grad_norm": 0.08498567342758179,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 94422
+    },
+    {
+      "epoch": 9.748606235804253,
+      "grad_norm": 0.07723391801118851,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 94425
+    },
+    {
+      "epoch": 9.748915961181087,
+      "grad_norm": 0.07648925483226776,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 94428
+    },
+    {
+      "epoch": 9.749225686557919,
+      "grad_norm": 0.06334717571735382,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 94431
+    },
+    {
+      "epoch": 9.749535411934751,
+      "grad_norm": 0.07004591822624207,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 94434
+    },
+    {
+      "epoch": 9.749845137311583,
+      "grad_norm": 0.05966423079371452,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 94437
+    },
+    {
+      "epoch": 9.750154862688417,
+      "grad_norm": 0.052176862955093384,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 94440
+    },
+    {
+      "epoch": 9.750464588065249,
+      "grad_norm": 0.04152917116880417,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 94443
+    },
+    {
+      "epoch": 9.750774313442081,
+      "grad_norm": 0.05614443123340607,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 94446
+    },
+    {
+      "epoch": 9.751084038818913,
+      "grad_norm": 0.039038725197315216,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 94449
+    },
+    {
+      "epoch": 9.751393764195747,
+      "grad_norm": 0.09913656860589981,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 94452
+    },
+    {
+      "epoch": 9.751703489572579,
+      "grad_norm": 0.08466944843530655,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 94455
+    },
+    {
+      "epoch": 9.752013214949411,
+      "grad_norm": 0.13703182339668274,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 94458
+    },
+    {
+      "epoch": 9.752322940326245,
+      "grad_norm": 0.13580304384231567,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 94461
+    },
+    {
+      "epoch": 9.752632665703077,
+      "grad_norm": 0.0782763808965683,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 94464
+    },
+    {
+      "epoch": 9.752942391079909,
+      "grad_norm": 0.07290788739919662,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 94467
+    },
+    {
+      "epoch": 9.753252116456741,
+      "grad_norm": 0.085691437125206,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 94470
+    },
+    {
+      "epoch": 9.753561841833575,
+      "grad_norm": 0.05388426035642624,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 94473
+    },
+    {
+      "epoch": 9.753871567210407,
+      "grad_norm": 0.1038847267627716,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 94476
+    },
+    {
+      "epoch": 9.754181292587239,
+      "grad_norm": 0.08317159861326218,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 94479
+    },
+    {
+      "epoch": 9.754491017964071,
+      "grad_norm": 0.05998384952545166,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 94482
+    },
+    {
+      "epoch": 9.754800743340905,
+      "grad_norm": 0.04784899204969406,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 94485
+    },
+    {
+      "epoch": 9.755110468717737,
+      "grad_norm": 0.05339213088154793,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 94488
+    },
+    {
+      "epoch": 9.755420194094569,
+      "grad_norm": 0.030975909903645515,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 94491
+    },
+    {
+      "epoch": 9.755729919471403,
+      "grad_norm": 0.069178506731987,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 94494
+    },
+    {
+      "epoch": 9.756039644848235,
+      "grad_norm": 0.07615378499031067,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 94497
+    },
+    {
+      "epoch": 9.756349370225067,
+      "grad_norm": 0.1228436827659607,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 94500
+    },
+    {
+      "epoch": 9.7566590956019,
+      "grad_norm": 0.046662211418151855,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 94503
+    },
+    {
+      "epoch": 9.756968820978733,
+      "grad_norm": 0.03317369148135185,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 94506
+    },
+    {
+      "epoch": 9.757278546355565,
+      "grad_norm": 0.048495639115571976,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 94509
+    },
+    {
+      "epoch": 9.757588271732397,
+      "grad_norm": 0.04089610278606415,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 94512
+    },
+    {
+      "epoch": 9.75789799710923,
+      "grad_norm": 0.0714641660451889,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 94515
+    },
+    {
+      "epoch": 9.758207722486063,
+      "grad_norm": 0.1012755036354065,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 94518
+    },
+    {
+      "epoch": 9.758517447862895,
+      "grad_norm": 0.09065414220094681,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 94521
+    },
+    {
+      "epoch": 9.758827173239727,
+      "grad_norm": 0.061119962483644485,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 94524
+    },
+    {
+      "epoch": 9.75913689861656,
+      "grad_norm": 0.11485234647989273,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 94527
+    },
+    {
+      "epoch": 9.759446623993393,
+      "grad_norm": 0.050607532262802124,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 94530
+    },
+    {
+      "epoch": 9.759756349370225,
+      "grad_norm": 0.07120644301176071,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 94533
+    },
+    {
+      "epoch": 9.760066074747058,
+      "grad_norm": 0.04502887651324272,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 94536
+    },
+    {
+      "epoch": 9.76037580012389,
+      "grad_norm": 0.07653278112411499,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 94539
+    },
+    {
+      "epoch": 9.760685525500723,
+      "grad_norm": 0.1033986359834671,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 94542
+    },
+    {
+      "epoch": 9.760995250877555,
+      "grad_norm": 0.08466217666864395,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 94545
+    },
+    {
+      "epoch": 9.761304976254388,
+      "grad_norm": 0.037229862064123154,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 94548
+    },
+    {
+      "epoch": 9.76161470163122,
+      "grad_norm": 0.11992764472961426,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 94551
+    },
+    {
+      "epoch": 9.761924427008053,
+      "grad_norm": 0.052539899945259094,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 94554
+    },
+    {
+      "epoch": 9.762234152384885,
+      "grad_norm": 0.08188603818416595,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 94557
+    },
+    {
+      "epoch": 9.762543877761718,
+      "grad_norm": 0.09905125945806503,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 94560
+    },
+    {
+      "epoch": 9.76285360313855,
+      "grad_norm": 0.1362028270959854,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 94563
+    },
+    {
+      "epoch": 9.763163328515382,
+      "grad_norm": 0.07078815251588821,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 94566
+    },
+    {
+      "epoch": 9.763473053892216,
+      "grad_norm": 0.04589496925473213,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 94569
+    },
+    {
+      "epoch": 9.763782779269048,
+      "grad_norm": 0.03925814852118492,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 94572
+    },
+    {
+      "epoch": 9.76409250464588,
+      "grad_norm": 0.08715146780014038,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 94575
+    },
+    {
+      "epoch": 9.764402230022712,
+      "grad_norm": 0.05460555851459503,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 94578
+    },
+    {
+      "epoch": 9.764711955399546,
+      "grad_norm": 0.08715993911027908,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 94581
+    },
+    {
+      "epoch": 9.765021680776378,
+      "grad_norm": 0.07886404544115067,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 94584
+    },
+    {
+      "epoch": 9.76533140615321,
+      "grad_norm": 0.06578825414180756,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 94587
+    },
+    {
+      "epoch": 9.765641131530042,
+      "grad_norm": 0.06801187247037888,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 94590
+    },
+    {
+      "epoch": 9.765950856906876,
+      "grad_norm": 0.1005634069442749,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 94593
+    },
+    {
+      "epoch": 9.766260582283708,
+      "grad_norm": 0.04338005930185318,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 94596
+    },
+    {
+      "epoch": 9.76657030766054,
+      "grad_norm": 0.045400865375995636,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 94599
+    },
+    {
+      "epoch": 9.766880033037374,
+      "grad_norm": 0.19539475440979004,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 94602
+    },
+    {
+      "epoch": 9.767189758414206,
+      "grad_norm": 0.04993902146816254,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 94605
+    },
+    {
+      "epoch": 9.767499483791038,
+      "grad_norm": 0.036195334047079086,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 94608
+    },
+    {
+      "epoch": 9.767809209167872,
+      "grad_norm": 0.052537743002176285,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 94611
+    },
+    {
+      "epoch": 9.768118934544704,
+      "grad_norm": 0.03671710193157196,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 94614
+    },
+    {
+      "epoch": 9.768428659921536,
+      "grad_norm": 0.06318136304616928,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 94617
+    },
+    {
+      "epoch": 9.768738385298368,
+      "grad_norm": 0.05337478220462799,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 94620
+    },
+    {
+      "epoch": 9.769048110675202,
+      "grad_norm": 0.05959227681159973,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 94623
+    },
+    {
+      "epoch": 9.769357836052034,
+      "grad_norm": 0.06474972516298294,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 94626
+    },
+    {
+      "epoch": 9.769667561428866,
+      "grad_norm": 0.13635705411434174,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 94629
+    },
+    {
+      "epoch": 9.769977286805698,
+      "grad_norm": 0.047785691916942596,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 94632
+    },
+    {
+      "epoch": 9.770287012182532,
+      "grad_norm": 0.040347639471292496,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 94635
+    },
+    {
+      "epoch": 9.770596737559364,
+      "grad_norm": 0.05130305141210556,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 94638
+    },
+    {
+      "epoch": 9.770906462936196,
+      "grad_norm": 0.1306859850883484,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 94641
+    },
+    {
+      "epoch": 9.77121618831303,
+      "grad_norm": 0.06589134782552719,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 94644
+    },
+    {
+      "epoch": 9.771525913689862,
+      "grad_norm": 0.06414980441331863,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 94647
+    },
+    {
+      "epoch": 9.771835639066694,
+      "grad_norm": 0.0561249703168869,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 94650
+    },
+    {
+      "epoch": 9.772145364443526,
+      "grad_norm": 0.03154081106185913,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 94653
+    },
+    {
+      "epoch": 9.77245508982036,
+      "grad_norm": 0.044503260403871536,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 94656
+    },
+    {
+      "epoch": 9.772764815197192,
+      "grad_norm": 0.03981982544064522,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 94659
+    },
+    {
+      "epoch": 9.773074540574024,
+      "grad_norm": 0.061454884707927704,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 94662
+    },
+    {
+      "epoch": 9.773384265950856,
+      "grad_norm": 0.1729876846075058,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 94665
+    },
+    {
+      "epoch": 9.77369399132769,
+      "grad_norm": 0.06331760436296463,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 94668
+    },
+    {
+      "epoch": 9.774003716704522,
+      "grad_norm": 0.06349468976259232,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 94671
+    },
+    {
+      "epoch": 9.774313442081354,
+      "grad_norm": 0.08989334106445312,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 94674
+    },
+    {
+      "epoch": 9.774623167458188,
+      "grad_norm": 0.06618794798851013,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 94677
+    },
+    {
+      "epoch": 9.77493289283502,
+      "grad_norm": 0.031386930495500565,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 94680
+    },
+    {
+      "epoch": 9.775242618211852,
+      "grad_norm": 0.036483556032180786,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 94683
+    },
+    {
+      "epoch": 9.775552343588684,
+      "grad_norm": 0.043264761567115784,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 94686
+    },
+    {
+      "epoch": 9.775862068965518,
+      "grad_norm": 0.03680359944701195,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 94689
+    },
+    {
+      "epoch": 9.77617179434235,
+      "grad_norm": 0.06547412276268005,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 94692
+    },
+    {
+      "epoch": 9.776481519719182,
+      "grad_norm": 0.16053377091884613,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 94695
+    },
+    {
+      "epoch": 9.776791245096014,
+      "grad_norm": 0.08217158168554306,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 94698
+    },
+    {
+      "epoch": 9.777100970472848,
+      "grad_norm": 0.14973968267440796,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 94701
+    },
+    {
+      "epoch": 9.77741069584968,
+      "grad_norm": 0.07262899726629257,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 94704
+    },
+    {
+      "epoch": 9.777720421226512,
+      "grad_norm": 0.05001377314329147,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 94707
+    },
+    {
+      "epoch": 9.778030146603346,
+      "grad_norm": 0.06505423784255981,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 94710
+    },
+    {
+      "epoch": 9.778339871980178,
+      "grad_norm": 0.04276656731963158,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 94713
+    },
+    {
+      "epoch": 9.77864959735701,
+      "grad_norm": 0.03513118252158165,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 94716
+    },
+    {
+      "epoch": 9.778959322733842,
+      "grad_norm": 0.0770474448800087,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 94719
+    },
+    {
+      "epoch": 9.779269048110676,
+      "grad_norm": 0.050384245812892914,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 94722
+    },
+    {
+      "epoch": 9.779578773487508,
+      "grad_norm": 0.06375786662101746,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 94725
+    },
+    {
+      "epoch": 9.77988849886434,
+      "grad_norm": 0.16862791776657104,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 94728
+    },
+    {
+      "epoch": 9.780198224241174,
+      "grad_norm": 0.07121732085943222,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 94731
+    },
+    {
+      "epoch": 9.780507949618006,
+      "grad_norm": 0.1161705031991005,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 94734
+    },
+    {
+      "epoch": 9.780817674994838,
+      "grad_norm": 0.08478477597236633,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 94737
+    },
+    {
+      "epoch": 9.78112740037167,
+      "grad_norm": 0.1556144654750824,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 94740
+    },
+    {
+      "epoch": 9.781437125748504,
+      "grad_norm": 0.09649685770273209,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 94743
+    },
+    {
+      "epoch": 9.781746851125336,
+      "grad_norm": 0.04481673240661621,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 94746
+    },
+    {
+      "epoch": 9.782056576502168,
+      "grad_norm": 0.0322875939309597,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 94749
+    },
+    {
+      "epoch": 9.782366301879001,
+      "grad_norm": 0.042956385761499405,
+      "learning_rate": 0.01,
+      "loss": 1.8979,
+      "step": 94752
+    },
+    {
+      "epoch": 9.782676027255834,
+      "grad_norm": 0.06072737276554108,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 94755
+    },
+    {
+      "epoch": 9.782985752632666,
+      "grad_norm": 0.04351195693016052,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 94758
+    },
+    {
+      "epoch": 9.783295478009498,
+      "grad_norm": 0.044757772237062454,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 94761
+    },
+    {
+      "epoch": 9.783605203386331,
+      "grad_norm": 0.10578258335590363,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 94764
+    },
+    {
+      "epoch": 9.783914928763163,
+      "grad_norm": 0.1356111615896225,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 94767
+    },
+    {
+      "epoch": 9.784224654139996,
+      "grad_norm": 0.07396072149276733,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 94770
+    },
+    {
+      "epoch": 9.784534379516828,
+      "grad_norm": 0.08166845142841339,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 94773
+    },
+    {
+      "epoch": 9.784844104893661,
+      "grad_norm": 0.07386428862810135,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 94776
+    },
+    {
+      "epoch": 9.785153830270493,
+      "grad_norm": 0.06322996318340302,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 94779
+    },
+    {
+      "epoch": 9.785463555647326,
+      "grad_norm": 0.042523063719272614,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 94782
+    },
+    {
+      "epoch": 9.78577328102416,
+      "grad_norm": 0.05621153116226196,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 94785
+    },
+    {
+      "epoch": 9.786083006400991,
+      "grad_norm": 0.05215676873922348,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 94788
+    },
+    {
+      "epoch": 9.786392731777823,
+      "grad_norm": 0.05371996760368347,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 94791
+    },
+    {
+      "epoch": 9.786702457154655,
+      "grad_norm": 0.05981604382395744,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 94794
+    },
+    {
+      "epoch": 9.78701218253149,
+      "grad_norm": 0.048599109053611755,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 94797
+    },
+    {
+      "epoch": 9.787321907908321,
+      "grad_norm": 0.045395396649837494,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 94800
+    },
+    {
+      "epoch": 9.787631633285153,
+      "grad_norm": 0.06925131380558014,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 94803
+    },
+    {
+      "epoch": 9.787941358661985,
+      "grad_norm": 0.1601201444864273,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 94806
+    },
+    {
+      "epoch": 9.78825108403882,
+      "grad_norm": 0.41700854897499084,
+      "learning_rate": 0.01,
+      "loss": 2.0459,
+      "step": 94809
+    },
+    {
+      "epoch": 9.788560809415651,
+      "grad_norm": 0.1901201605796814,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 94812
+    },
+    {
+      "epoch": 9.788870534792483,
+      "grad_norm": 0.09801574796438217,
+      "learning_rate": 0.01,
+      "loss": 1.9844,
+      "step": 94815
+    },
+    {
+      "epoch": 9.789180260169317,
+      "grad_norm": 0.08561704307794571,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 94818
+    },
+    {
+      "epoch": 9.78948998554615,
+      "grad_norm": 0.05033617466688156,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 94821
+    },
+    {
+      "epoch": 9.789799710922981,
+      "grad_norm": 0.05378749221563339,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 94824
+    },
+    {
+      "epoch": 9.790109436299813,
+      "grad_norm": 0.05089649185538292,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 94827
+    },
+    {
+      "epoch": 9.790419161676647,
+      "grad_norm": 0.04963233694434166,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 94830
+    },
+    {
+      "epoch": 9.79072888705348,
+      "grad_norm": 0.06947825849056244,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 94833
+    },
+    {
+      "epoch": 9.791038612430311,
+      "grad_norm": 0.09509982913732529,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 94836
+    },
+    {
+      "epoch": 9.791348337807145,
+      "grad_norm": 0.04908059537410736,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 94839
+    },
+    {
+      "epoch": 9.791658063183977,
+      "grad_norm": 0.03284559026360512,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 94842
+    },
+    {
+      "epoch": 9.79196778856081,
+      "grad_norm": 0.04718421399593353,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 94845
+    },
+    {
+      "epoch": 9.792277513937641,
+      "grad_norm": 0.05559903010725975,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 94848
+    },
+    {
+      "epoch": 9.792587239314475,
+      "grad_norm": 0.0743434950709343,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 94851
+    },
+    {
+      "epoch": 9.792896964691307,
+      "grad_norm": 0.07633355259895325,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 94854
+    },
+    {
+      "epoch": 9.79320669006814,
+      "grad_norm": 0.10470660775899887,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 94857
+    },
+    {
+      "epoch": 9.793516415444973,
+      "grad_norm": 0.16582956910133362,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 94860
+    },
+    {
+      "epoch": 9.793826140821805,
+      "grad_norm": 0.04770703613758087,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 94863
+    },
+    {
+      "epoch": 9.794135866198637,
+      "grad_norm": 0.11444547772407532,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 94866
+    },
+    {
+      "epoch": 9.79444559157547,
+      "grad_norm": 0.04240221902728081,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 94869
+    },
+    {
+      "epoch": 9.794755316952303,
+      "grad_norm": 0.05061916634440422,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 94872
+    },
+    {
+      "epoch": 9.795065042329135,
+      "grad_norm": 0.04344547167420387,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 94875
+    },
+    {
+      "epoch": 9.795374767705967,
+      "grad_norm": 0.05513598024845123,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 94878
+    },
+    {
+      "epoch": 9.795684493082799,
+      "grad_norm": 0.08316171914339066,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 94881
+    },
+    {
+      "epoch": 9.795994218459633,
+      "grad_norm": 0.06647349894046783,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 94884
+    },
+    {
+      "epoch": 9.796303943836465,
+      "grad_norm": 0.05216540768742561,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 94887
+    },
+    {
+      "epoch": 9.796613669213297,
+      "grad_norm": 0.05408751964569092,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 94890
+    },
+    {
+      "epoch": 9.79692339459013,
+      "grad_norm": 0.039487335830926895,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 94893
+    },
+    {
+      "epoch": 9.797233119966963,
+      "grad_norm": 0.054186344146728516,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 94896
+    },
+    {
+      "epoch": 9.797542845343795,
+      "grad_norm": 0.03691032901406288,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 94899
+    },
+    {
+      "epoch": 9.797852570720627,
+      "grad_norm": 0.12448958307504654,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 94902
+    },
+    {
+      "epoch": 9.79816229609746,
+      "grad_norm": 0.1222798153758049,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 94905
+    },
+    {
+      "epoch": 9.798472021474293,
+      "grad_norm": 0.04015497863292694,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 94908
+    },
+    {
+      "epoch": 9.798781746851125,
+      "grad_norm": 0.04652899503707886,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 94911
+    },
+    {
+      "epoch": 9.799091472227957,
+      "grad_norm": 0.07459242641925812,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 94914
+    },
+    {
+      "epoch": 9.79940119760479,
+      "grad_norm": 0.05896695703268051,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 94917
+    },
+    {
+      "epoch": 9.799710922981623,
+      "grad_norm": 0.12853103876113892,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 94920
+    },
+    {
+      "epoch": 9.800020648358455,
+      "grad_norm": 0.04624097794294357,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 94923
+    },
+    {
+      "epoch": 9.800330373735289,
+      "grad_norm": 0.09572327882051468,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 94926
+    },
+    {
+      "epoch": 9.80064009911212,
+      "grad_norm": 0.03788885474205017,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 94929
+    },
+    {
+      "epoch": 9.800949824488953,
+      "grad_norm": 0.06076391786336899,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 94932
+    },
+    {
+      "epoch": 9.801259549865785,
+      "grad_norm": 0.0562455952167511,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 94935
+    },
+    {
+      "epoch": 9.801569275242619,
+      "grad_norm": 0.04046221449971199,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 94938
+    },
+    {
+      "epoch": 9.80187900061945,
+      "grad_norm": 0.05040094256401062,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 94941
+    },
+    {
+      "epoch": 9.802188725996283,
+      "grad_norm": 0.042106032371520996,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 94944
+    },
+    {
+      "epoch": 9.802498451373117,
+      "grad_norm": 0.04450448974967003,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 94947
+    },
+    {
+      "epoch": 9.802808176749949,
+      "grad_norm": 0.04702811688184738,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 94950
+    },
+    {
+      "epoch": 9.80311790212678,
+      "grad_norm": 0.05073699355125427,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 94953
+    },
+    {
+      "epoch": 9.803427627503613,
+      "grad_norm": 0.0670652911067009,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 94956
+    },
+    {
+      "epoch": 9.803737352880447,
+      "grad_norm": 0.10583470016717911,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 94959
+    },
+    {
+      "epoch": 9.804047078257279,
+      "grad_norm": 0.11010700464248657,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 94962
+    },
+    {
+      "epoch": 9.80435680363411,
+      "grad_norm": 0.04617714136838913,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 94965
+    },
+    {
+      "epoch": 9.804666529010944,
+      "grad_norm": 0.19462630152702332,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 94968
+    },
+    {
+      "epoch": 9.804976254387777,
+      "grad_norm": 0.04227757081389427,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 94971
+    },
+    {
+      "epoch": 9.805285979764609,
+      "grad_norm": 0.04136215150356293,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 94974
+    },
+    {
+      "epoch": 9.80559570514144,
+      "grad_norm": 0.032104820013046265,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 94977
+    },
+    {
+      "epoch": 9.805905430518274,
+      "grad_norm": 0.03970668092370033,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 94980
+    },
+    {
+      "epoch": 9.806215155895107,
+      "grad_norm": 0.046333905309438705,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 94983
+    },
+    {
+      "epoch": 9.806524881271939,
+      "grad_norm": 0.09408304840326309,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 94986
+    },
+    {
+      "epoch": 9.80683460664877,
+      "grad_norm": 0.061889585107564926,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 94989
+    },
+    {
+      "epoch": 9.807144332025604,
+      "grad_norm": 0.05078985542058945,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 94992
+    },
+    {
+      "epoch": 9.807454057402436,
+      "grad_norm": 0.04673658683896065,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 94995
+    },
+    {
+      "epoch": 9.807763782779269,
+      "grad_norm": 0.04183147847652435,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 94998
+    },
+    {
+      "epoch": 9.808073508156102,
+      "grad_norm": 0.050529371947050095,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 95001
+    },
+    {
+      "epoch": 9.808383233532934,
+      "grad_norm": 0.05828772857785225,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 95004
+    },
+    {
+      "epoch": 9.808692958909766,
+      "grad_norm": 0.11839016526937485,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 95007
+    },
+    {
+      "epoch": 9.809002684286598,
+      "grad_norm": 0.05272934213280678,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 95010
+    },
+    {
+      "epoch": 9.809312409663432,
+      "grad_norm": 0.07402176409959793,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 95013
+    },
+    {
+      "epoch": 9.809622135040264,
+      "grad_norm": 0.0874079018831253,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 95016
+    },
+    {
+      "epoch": 9.809931860417096,
+      "grad_norm": 0.050489626824855804,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 95019
+    },
+    {
+      "epoch": 9.810241585793928,
+      "grad_norm": 0.07709213346242905,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 95022
+    },
+    {
+      "epoch": 9.810551311170762,
+      "grad_norm": 0.07870500534772873,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 95025
+    },
+    {
+      "epoch": 9.810861036547594,
+      "grad_norm": 0.1378456950187683,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 95028
+    },
+    {
+      "epoch": 9.811170761924426,
+      "grad_norm": 0.10447420179843903,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 95031
+    },
+    {
+      "epoch": 9.81148048730126,
+      "grad_norm": 0.08695314824581146,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 95034
+    },
+    {
+      "epoch": 9.811790212678092,
+      "grad_norm": 0.03956886753439903,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 95037
+    },
+    {
+      "epoch": 9.812099938054924,
+      "grad_norm": 0.03908482939004898,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 95040
+    },
+    {
+      "epoch": 9.812409663431756,
+      "grad_norm": 0.04740505293011665,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 95043
+    },
+    {
+      "epoch": 9.81271938880859,
+      "grad_norm": 0.07152485102415085,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 95046
+    },
+    {
+      "epoch": 9.813029114185422,
+      "grad_norm": 0.09586811810731888,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 95049
+    },
+    {
+      "epoch": 9.813338839562254,
+      "grad_norm": 0.14370545744895935,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 95052
+    },
+    {
+      "epoch": 9.813648564939088,
+      "grad_norm": 0.10440492630004883,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 95055
+    },
+    {
+      "epoch": 9.81395829031592,
+      "grad_norm": 0.06501776725053787,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 95058
+    },
+    {
+      "epoch": 9.814268015692752,
+      "grad_norm": 0.06675376743078232,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 95061
+    },
+    {
+      "epoch": 9.814577741069584,
+      "grad_norm": 0.05271616205573082,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 95064
+    },
+    {
+      "epoch": 9.814887466446418,
+      "grad_norm": 0.05236978828907013,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 95067
+    },
+    {
+      "epoch": 9.81519719182325,
+      "grad_norm": 0.03621961921453476,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 95070
+    },
+    {
+      "epoch": 9.815506917200082,
+      "grad_norm": 0.06704110652208328,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 95073
+    },
+    {
+      "epoch": 9.815816642576916,
+      "grad_norm": 0.09148068726062775,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 95076
+    },
+    {
+      "epoch": 9.816126367953748,
+      "grad_norm": 0.0595364049077034,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 95079
+    },
+    {
+      "epoch": 9.81643609333058,
+      "grad_norm": 0.14149333536624908,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 95082
+    },
+    {
+      "epoch": 9.816745818707412,
+      "grad_norm": 0.04914262518286705,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 95085
+    },
+    {
+      "epoch": 9.817055544084246,
+      "grad_norm": 0.04873903468251228,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 95088
+    },
+    {
+      "epoch": 9.817365269461078,
+      "grad_norm": 0.08692160248756409,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 95091
+    },
+    {
+      "epoch": 9.81767499483791,
+      "grad_norm": 0.057351529598236084,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 95094
+    },
+    {
+      "epoch": 9.817984720214742,
+      "grad_norm": 0.0691007673740387,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 95097
+    },
+    {
+      "epoch": 9.818294445591576,
+      "grad_norm": 0.08830489963293076,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 95100
+    },
+    {
+      "epoch": 9.818604170968408,
+      "grad_norm": 0.03572781756520271,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 95103
+    },
+    {
+      "epoch": 9.81891389634524,
+      "grad_norm": 0.09434852749109268,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 95106
+    },
+    {
+      "epoch": 9.819223621722074,
+      "grad_norm": 0.09754703938961029,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 95109
+    },
+    {
+      "epoch": 9.819533347098906,
+      "grad_norm": 0.0586463026702404,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 95112
+    },
+    {
+      "epoch": 9.819843072475738,
+      "grad_norm": 0.046134430915117264,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 95115
+    },
+    {
+      "epoch": 9.82015279785257,
+      "grad_norm": 0.05168531462550163,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 95118
+    },
+    {
+      "epoch": 9.820462523229404,
+      "grad_norm": 0.08128103613853455,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 95121
+    },
+    {
+      "epoch": 9.820772248606236,
+      "grad_norm": 0.06025461107492447,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 95124
+    },
+    {
+      "epoch": 9.821081973983068,
+      "grad_norm": 0.07480666786432266,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 95127
+    },
+    {
+      "epoch": 9.8213916993599,
+      "grad_norm": 0.04217499494552612,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 95130
+    },
+    {
+      "epoch": 9.821701424736734,
+      "grad_norm": 0.12061578780412674,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 95133
+    },
+    {
+      "epoch": 9.822011150113566,
+      "grad_norm": 0.04855797812342644,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 95136
+    },
+    {
+      "epoch": 9.822320875490398,
+      "grad_norm": 0.05785534158349037,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 95139
+    },
+    {
+      "epoch": 9.822630600867232,
+      "grad_norm": 0.09376546740531921,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 95142
+    },
+    {
+      "epoch": 9.822940326244064,
+      "grad_norm": 0.03916393965482712,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 95145
+    },
+    {
+      "epoch": 9.823250051620896,
+      "grad_norm": 0.07211345434188843,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 95148
+    },
+    {
+      "epoch": 9.823559776997728,
+      "grad_norm": 0.05421319231390953,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 95151
+    },
+    {
+      "epoch": 9.823869502374562,
+      "grad_norm": 0.09411376714706421,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 95154
+    },
+    {
+      "epoch": 9.824179227751394,
+      "grad_norm": 0.09898379445075989,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 95157
+    },
+    {
+      "epoch": 9.824488953128226,
+      "grad_norm": 0.05836961045861244,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 95160
+    },
+    {
+      "epoch": 9.82479867850506,
+      "grad_norm": 0.03766075149178505,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 95163
+    },
+    {
+      "epoch": 9.825108403881892,
+      "grad_norm": 0.033823855221271515,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 95166
+    },
+    {
+      "epoch": 9.825418129258724,
+      "grad_norm": 0.04306140914559364,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 95169
+    },
+    {
+      "epoch": 9.825727854635556,
+      "grad_norm": 0.03758104890584946,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 95172
+    },
+    {
+      "epoch": 9.82603758001239,
+      "grad_norm": 0.05285748839378357,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 95175
+    },
+    {
+      "epoch": 9.826347305389222,
+      "grad_norm": 0.04774449020624161,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 95178
+    },
+    {
+      "epoch": 9.826657030766054,
+      "grad_norm": 0.04499258100986481,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 95181
+    },
+    {
+      "epoch": 9.826966756142888,
+      "grad_norm": 0.06346732378005981,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 95184
+    },
+    {
+      "epoch": 9.82727648151972,
+      "grad_norm": 0.15880587697029114,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 95187
+    },
+    {
+      "epoch": 9.827586206896552,
+      "grad_norm": 0.07601436227560043,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 95190
+    },
+    {
+      "epoch": 9.827895932273384,
+      "grad_norm": 0.04568092152476311,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 95193
+    },
+    {
+      "epoch": 9.828205657650217,
+      "grad_norm": 0.052929848432540894,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 95196
+    },
+    {
+      "epoch": 9.82851538302705,
+      "grad_norm": 0.04511851444840431,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 95199
+    },
+    {
+      "epoch": 9.828825108403882,
+      "grad_norm": 0.09903811663389206,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 95202
+    },
+    {
+      "epoch": 9.829134833780714,
+      "grad_norm": 0.062109824270009995,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 95205
+    },
+    {
+      "epoch": 9.829444559157547,
+      "grad_norm": 0.08777434378862381,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 95208
+    },
+    {
+      "epoch": 9.82975428453438,
+      "grad_norm": 0.06251123547554016,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 95211
+    },
+    {
+      "epoch": 9.830064009911212,
+      "grad_norm": 0.09451593458652496,
+      "learning_rate": 0.01,
+      "loss": 1.9007,
+      "step": 95214
+    },
+    {
+      "epoch": 9.830373735288045,
+      "grad_norm": 0.07134056836366653,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 95217
+    },
+    {
+      "epoch": 9.830683460664877,
+      "grad_norm": 0.0673132911324501,
+      "learning_rate": 0.01,
+      "loss": 1.9035,
+      "step": 95220
+    },
+    {
+      "epoch": 9.83099318604171,
+      "grad_norm": 0.13773980736732483,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 95223
+    },
+    {
+      "epoch": 9.831302911418542,
+      "grad_norm": 0.0492391400039196,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 95226
+    },
+    {
+      "epoch": 9.831612636795375,
+      "grad_norm": 0.043367672711610794,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 95229
+    },
+    {
+      "epoch": 9.831922362172207,
+      "grad_norm": 0.03407084569334984,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 95232
+    },
+    {
+      "epoch": 9.83223208754904,
+      "grad_norm": 0.0420520156621933,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 95235
+    },
+    {
+      "epoch": 9.832541812925871,
+      "grad_norm": 0.06530791521072388,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 95238
+    },
+    {
+      "epoch": 9.832851538302705,
+      "grad_norm": 0.061357442289590836,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 95241
+    },
+    {
+      "epoch": 9.833161263679537,
+      "grad_norm": 0.04315570369362831,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 95244
+    },
+    {
+      "epoch": 9.83347098905637,
+      "grad_norm": 0.19494427740573883,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 95247
+    },
+    {
+      "epoch": 9.833780714433203,
+      "grad_norm": 0.0513029620051384,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 95250
+    },
+    {
+      "epoch": 9.834090439810035,
+      "grad_norm": 0.04042904078960419,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 95253
+    },
+    {
+      "epoch": 9.834400165186867,
+      "grad_norm": 0.03216146305203438,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 95256
+    },
+    {
+      "epoch": 9.8347098905637,
+      "grad_norm": 0.030513131991028786,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 95259
+    },
+    {
+      "epoch": 9.835019615940533,
+      "grad_norm": 0.05615859106183052,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 95262
+    },
+    {
+      "epoch": 9.835329341317365,
+      "grad_norm": 0.04247753694653511,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 95265
+    },
+    {
+      "epoch": 9.835639066694197,
+      "grad_norm": 0.03994342312216759,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 95268
+    },
+    {
+      "epoch": 9.835948792071031,
+      "grad_norm": 0.05301568657159805,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 95271
+    },
+    {
+      "epoch": 9.836258517447863,
+      "grad_norm": 0.1442309468984604,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 95274
+    },
+    {
+      "epoch": 9.836568242824695,
+      "grad_norm": 0.12944433093070984,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 95277
+    },
+    {
+      "epoch": 9.836877968201527,
+      "grad_norm": 0.09283921867609024,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 95280
+    },
+    {
+      "epoch": 9.837187693578361,
+      "grad_norm": 0.0523645244538784,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 95283
+    },
+    {
+      "epoch": 9.837497418955193,
+      "grad_norm": 0.0349854975938797,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 95286
+    },
+    {
+      "epoch": 9.837807144332025,
+      "grad_norm": 0.03146263211965561,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 95289
+    },
+    {
+      "epoch": 9.838116869708859,
+      "grad_norm": 0.04970834031701088,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 95292
+    },
+    {
+      "epoch": 9.838426595085691,
+      "grad_norm": 0.061335138976573944,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 95295
+    },
+    {
+      "epoch": 9.838736320462523,
+      "grad_norm": 0.09407352656126022,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 95298
+    },
+    {
+      "epoch": 9.839046045839355,
+      "grad_norm": 0.10056418925523758,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 95301
+    },
+    {
+      "epoch": 9.839355771216189,
+      "grad_norm": 0.13487358391284943,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 95304
+    },
+    {
+      "epoch": 9.839665496593021,
+      "grad_norm": 0.06113987788558006,
+      "learning_rate": 0.01,
+      "loss": 1.9075,
+      "step": 95307
+    },
+    {
+      "epoch": 9.839975221969853,
+      "grad_norm": 0.050525832921266556,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 95310
+    },
+    {
+      "epoch": 9.840284947346685,
+      "grad_norm": 0.034293282777071,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 95313
+    },
+    {
+      "epoch": 9.840594672723519,
+      "grad_norm": 0.052330173552036285,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 95316
+    },
+    {
+      "epoch": 9.840904398100351,
+      "grad_norm": 0.06765829026699066,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 95319
+    },
+    {
+      "epoch": 9.841214123477183,
+      "grad_norm": 0.04363483935594559,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 95322
+    },
+    {
+      "epoch": 9.841523848854017,
+      "grad_norm": 0.10867563635110855,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 95325
+    },
+    {
+      "epoch": 9.841833574230849,
+      "grad_norm": 0.04092460125684738,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 95328
+    },
+    {
+      "epoch": 9.842143299607681,
+      "grad_norm": 0.07562436908483505,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 95331
+    },
+    {
+      "epoch": 9.842453024984513,
+      "grad_norm": 0.08705686777830124,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 95334
+    },
+    {
+      "epoch": 9.842762750361347,
+      "grad_norm": 0.0464167594909668,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 95337
+    },
+    {
+      "epoch": 9.843072475738179,
+      "grad_norm": 0.046409085392951965,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 95340
+    },
+    {
+      "epoch": 9.843382201115011,
+      "grad_norm": 0.04688123241066933,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 95343
+    },
+    {
+      "epoch": 9.843691926491843,
+      "grad_norm": 0.04283418133854866,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 95346
+    },
+    {
+      "epoch": 9.844001651868677,
+      "grad_norm": 0.040036801248788834,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 95349
+    },
+    {
+      "epoch": 9.844311377245509,
+      "grad_norm": 0.060970354825258255,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 95352
+    },
+    {
+      "epoch": 9.844621102622341,
+      "grad_norm": 0.08856649696826935,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 95355
+    },
+    {
+      "epoch": 9.844930827999175,
+      "grad_norm": 0.07604091614484787,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 95358
+    },
+    {
+      "epoch": 9.845240553376007,
+      "grad_norm": 0.045264385640621185,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 95361
+    },
+    {
+      "epoch": 9.845550278752839,
+      "grad_norm": 0.0438438318669796,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 95364
+    },
+    {
+      "epoch": 9.84586000412967,
+      "grad_norm": 0.07063893228769302,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 95367
+    },
+    {
+      "epoch": 9.846169729506505,
+      "grad_norm": 0.05054168403148651,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 95370
+    },
+    {
+      "epoch": 9.846479454883337,
+      "grad_norm": 0.04927874729037285,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 95373
+    },
+    {
+      "epoch": 9.846789180260169,
+      "grad_norm": 0.14182618260383606,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 95376
+    },
+    {
+      "epoch": 9.847098905637003,
+      "grad_norm": 0.10691936314105988,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 95379
+    },
+    {
+      "epoch": 9.847408631013835,
+      "grad_norm": 0.09157336503267288,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 95382
+    },
+    {
+      "epoch": 9.847718356390667,
+      "grad_norm": 0.055040616542100906,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 95385
+    },
+    {
+      "epoch": 9.848028081767499,
+      "grad_norm": 0.035021986812353134,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 95388
+    },
+    {
+      "epoch": 9.848337807144333,
+      "grad_norm": 0.08447353541851044,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 95391
+    },
+    {
+      "epoch": 9.848647532521165,
+      "grad_norm": 0.06649502366781235,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 95394
+    },
+    {
+      "epoch": 9.848957257897997,
+      "grad_norm": 0.07391461730003357,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 95397
+    },
+    {
+      "epoch": 9.84926698327483,
+      "grad_norm": 0.0677117258310318,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 95400
+    },
+    {
+      "epoch": 9.849576708651663,
+      "grad_norm": 0.09915631264448166,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 95403
+    },
+    {
+      "epoch": 9.849886434028495,
+      "grad_norm": 0.12769655883312225,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 95406
+    },
+    {
+      "epoch": 9.850196159405327,
+      "grad_norm": 0.07113716751337051,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 95409
+    },
+    {
+      "epoch": 9.85050588478216,
+      "grad_norm": 0.05216817185282707,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 95412
+    },
+    {
+      "epoch": 9.850815610158993,
+      "grad_norm": 0.07261928915977478,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 95415
+    },
+    {
+      "epoch": 9.851125335535825,
+      "grad_norm": 0.035049788653850555,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 95418
+    },
+    {
+      "epoch": 9.851435060912657,
+      "grad_norm": 0.04018479213118553,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 95421
+    },
+    {
+      "epoch": 9.85174478628949,
+      "grad_norm": 0.08323214203119278,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 95424
+    },
+    {
+      "epoch": 9.852054511666323,
+      "grad_norm": 0.07037657499313354,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 95427
+    },
+    {
+      "epoch": 9.852364237043155,
+      "grad_norm": 0.07828336954116821,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 95430
+    },
+    {
+      "epoch": 9.852673962419988,
+      "grad_norm": 0.07287444919347763,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 95433
+    },
+    {
+      "epoch": 9.85298368779682,
+      "grad_norm": 0.08128449320793152,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 95436
+    },
+    {
+      "epoch": 9.853293413173652,
+      "grad_norm": 0.04227570444345474,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 95439
+    },
+    {
+      "epoch": 9.853603138550485,
+      "grad_norm": 0.11544642597436905,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 95442
+    },
+    {
+      "epoch": 9.853912863927318,
+      "grad_norm": 0.09630688279867172,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 95445
+    },
+    {
+      "epoch": 9.85422258930415,
+      "grad_norm": 0.09273266792297363,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 95448
+    },
+    {
+      "epoch": 9.854532314680982,
+      "grad_norm": 0.049246251583099365,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 95451
+    },
+    {
+      "epoch": 9.854842040057815,
+      "grad_norm": 0.0742521584033966,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 95454
+    },
+    {
+      "epoch": 9.855151765434648,
+      "grad_norm": 0.04551522433757782,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 95457
+    },
+    {
+      "epoch": 9.85546149081148,
+      "grad_norm": 0.1283591091632843,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 95460
+    },
+    {
+      "epoch": 9.855771216188312,
+      "grad_norm": 0.04384034126996994,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 95463
+    },
+    {
+      "epoch": 9.856080941565146,
+      "grad_norm": 0.057369425892829895,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 95466
+    },
+    {
+      "epoch": 9.856390666941978,
+      "grad_norm": 0.10041297972202301,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 95469
+    },
+    {
+      "epoch": 9.85670039231881,
+      "grad_norm": 0.09233054518699646,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 95472
+    },
+    {
+      "epoch": 9.857010117695642,
+      "grad_norm": 0.03711867704987526,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 95475
+    },
+    {
+      "epoch": 9.857319843072476,
+      "grad_norm": 0.04442448914051056,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 95478
+    },
+    {
+      "epoch": 9.857629568449308,
+      "grad_norm": 0.09285016357898712,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 95481
+    },
+    {
+      "epoch": 9.85793929382614,
+      "grad_norm": 0.0895034596323967,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 95484
+    },
+    {
+      "epoch": 9.858249019202974,
+      "grad_norm": 0.05122547596693039,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 95487
+    },
+    {
+      "epoch": 9.858558744579806,
+      "grad_norm": 0.053021177649497986,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 95490
+    },
+    {
+      "epoch": 9.858868469956638,
+      "grad_norm": 0.06684883683919907,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 95493
+    },
+    {
+      "epoch": 9.85917819533347,
+      "grad_norm": 0.06873955577611923,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 95496
+    },
+    {
+      "epoch": 9.859487920710304,
+      "grad_norm": 0.03738557919859886,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 95499
+    },
+    {
+      "epoch": 9.859797646087136,
+      "grad_norm": 0.046948738396167755,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 95502
+    },
+    {
+      "epoch": 9.860107371463968,
+      "grad_norm": 0.05666372552514076,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 95505
+    },
+    {
+      "epoch": 9.860417096840802,
+      "grad_norm": 0.12917053699493408,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 95508
+    },
+    {
+      "epoch": 9.860726822217634,
+      "grad_norm": 0.0822906345129013,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 95511
+    },
+    {
+      "epoch": 9.861036547594466,
+      "grad_norm": 0.05405477061867714,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 95514
+    },
+    {
+      "epoch": 9.861346272971298,
+      "grad_norm": 0.041714996099472046,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 95517
+    },
+    {
+      "epoch": 9.861655998348132,
+      "grad_norm": 0.03900561481714249,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 95520
+    },
+    {
+      "epoch": 9.861965723724964,
+      "grad_norm": 0.04210944101214409,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 95523
+    },
+    {
+      "epoch": 9.862275449101796,
+      "grad_norm": 0.0644078478217125,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 95526
+    },
+    {
+      "epoch": 9.862585174478628,
+      "grad_norm": 0.03967900574207306,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 95529
+    },
+    {
+      "epoch": 9.862894899855462,
+      "grad_norm": 0.12260102480649948,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 95532
+    },
+    {
+      "epoch": 9.863204625232294,
+      "grad_norm": 0.06045837700366974,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 95535
+    },
+    {
+      "epoch": 9.863514350609126,
+      "grad_norm": 0.10581725835800171,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 95538
+    },
+    {
+      "epoch": 9.86382407598596,
+      "grad_norm": 0.04299296438694,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 95541
+    },
+    {
+      "epoch": 9.864133801362792,
+      "grad_norm": 0.0774567574262619,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 95544
+    },
+    {
+      "epoch": 9.864443526739624,
+      "grad_norm": 0.05911951884627342,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 95547
+    },
+    {
+      "epoch": 9.864753252116456,
+      "grad_norm": 0.07533189654350281,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 95550
+    },
+    {
+      "epoch": 9.86506297749329,
+      "grad_norm": 0.035809364169836044,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 95553
+    },
+    {
+      "epoch": 9.865372702870122,
+      "grad_norm": 0.08691052347421646,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 95556
+    },
+    {
+      "epoch": 9.865682428246954,
+      "grad_norm": 0.11932963132858276,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 95559
+    },
+    {
+      "epoch": 9.865992153623786,
+      "grad_norm": 0.0466434583067894,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 95562
+    },
+    {
+      "epoch": 9.86630187900062,
+      "grad_norm": 0.045600853860378265,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 95565
+    },
+    {
+      "epoch": 9.866611604377452,
+      "grad_norm": 0.06262704730033875,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 95568
+    },
+    {
+      "epoch": 9.866921329754284,
+      "grad_norm": 0.05210849270224571,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 95571
+    },
+    {
+      "epoch": 9.867231055131118,
+      "grad_norm": 0.10397256910800934,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 95574
+    },
+    {
+      "epoch": 9.86754078050795,
+      "grad_norm": 0.05808919668197632,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 95577
+    },
+    {
+      "epoch": 9.867850505884782,
+      "grad_norm": 0.07182762771844864,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 95580
+    },
+    {
+      "epoch": 9.868160231261614,
+      "grad_norm": 0.07244507968425751,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 95583
+    },
+    {
+      "epoch": 9.868469956638448,
+      "grad_norm": 0.058532387018203735,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 95586
+    },
+    {
+      "epoch": 9.86877968201528,
+      "grad_norm": 0.0981719046831131,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 95589
+    },
+    {
+      "epoch": 9.869089407392112,
+      "grad_norm": 0.07616667449474335,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 95592
+    },
+    {
+      "epoch": 9.869399132768946,
+      "grad_norm": 0.0804872065782547,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 95595
+    },
+    {
+      "epoch": 9.869708858145778,
+      "grad_norm": 0.05157414451241493,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 95598
+    },
+    {
+      "epoch": 9.87001858352261,
+      "grad_norm": 0.033212315291166306,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 95601
+    },
+    {
+      "epoch": 9.870328308899442,
+      "grad_norm": 0.04200216010212898,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 95604
+    },
+    {
+      "epoch": 9.870638034276276,
+      "grad_norm": 0.042693279683589935,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 95607
+    },
+    {
+      "epoch": 9.870947759653108,
+      "grad_norm": 0.16291595995426178,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 95610
+    },
+    {
+      "epoch": 9.87125748502994,
+      "grad_norm": 0.1320657581090927,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 95613
+    },
+    {
+      "epoch": 9.871567210406774,
+      "grad_norm": 0.05239594727754593,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 95616
+    },
+    {
+      "epoch": 9.871876935783606,
+      "grad_norm": 0.04437854140996933,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 95619
+    },
+    {
+      "epoch": 9.872186661160438,
+      "grad_norm": 0.04155666381120682,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 95622
+    },
+    {
+      "epoch": 9.87249638653727,
+      "grad_norm": 0.08610442280769348,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 95625
+    },
+    {
+      "epoch": 9.872806111914104,
+      "grad_norm": 0.12307862192392349,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 95628
+    },
+    {
+      "epoch": 9.873115837290936,
+      "grad_norm": 0.09790744632482529,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 95631
+    },
+    {
+      "epoch": 9.873425562667768,
+      "grad_norm": 0.061377834528684616,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 95634
+    },
+    {
+      "epoch": 9.8737352880446,
+      "grad_norm": 0.08830194175243378,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 95637
+    },
+    {
+      "epoch": 9.874045013421433,
+      "grad_norm": 0.06755909323692322,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 95640
+    },
+    {
+      "epoch": 9.874354738798266,
+      "grad_norm": 0.03959019482135773,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 95643
+    },
+    {
+      "epoch": 9.874664464175098,
+      "grad_norm": 0.08053558319807053,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 95646
+    },
+    {
+      "epoch": 9.874974189551931,
+      "grad_norm": 0.09366048872470856,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 95649
+    },
+    {
+      "epoch": 9.875283914928763,
+      "grad_norm": 0.06561397761106491,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 95652
+    },
+    {
+      "epoch": 9.875593640305595,
+      "grad_norm": 0.08625023812055588,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 95655
+    },
+    {
+      "epoch": 9.875903365682428,
+      "grad_norm": 0.04743865877389908,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 95658
+    },
+    {
+      "epoch": 9.876213091059261,
+      "grad_norm": 0.12848925590515137,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 95661
+    },
+    {
+      "epoch": 9.876522816436093,
+      "grad_norm": 0.13109102845191956,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 95664
+    },
+    {
+      "epoch": 9.876832541812925,
+      "grad_norm": 0.10117898881435394,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 95667
+    },
+    {
+      "epoch": 9.877142267189758,
+      "grad_norm": 0.041157037019729614,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 95670
+    },
+    {
+      "epoch": 9.877451992566591,
+      "grad_norm": 0.03664329648017883,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 95673
+    },
+    {
+      "epoch": 9.877761717943423,
+      "grad_norm": 0.030358728021383286,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 95676
+    },
+    {
+      "epoch": 9.878071443320255,
+      "grad_norm": 0.045593880116939545,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 95679
+    },
+    {
+      "epoch": 9.87838116869709,
+      "grad_norm": 0.11838401108980179,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 95682
+    },
+    {
+      "epoch": 9.878690894073921,
+      "grad_norm": 0.10561030358076096,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 95685
+    },
+    {
+      "epoch": 9.879000619450753,
+      "grad_norm": 0.04690423980355263,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 95688
+    },
+    {
+      "epoch": 9.879310344827585,
+      "grad_norm": 0.03910710662603378,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 95691
+    },
+    {
+      "epoch": 9.87962007020442,
+      "grad_norm": 0.04057015851140022,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 95694
+    },
+    {
+      "epoch": 9.879929795581251,
+      "grad_norm": 0.11349897086620331,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 95697
+    },
+    {
+      "epoch": 9.880239520958083,
+      "grad_norm": 0.05625935643911362,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 95700
+    },
+    {
+      "epoch": 9.880549246334917,
+      "grad_norm": 0.06775499880313873,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 95703
+    },
+    {
+      "epoch": 9.88085897171175,
+      "grad_norm": 0.053724609315395355,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 95706
+    },
+    {
+      "epoch": 9.881168697088581,
+      "grad_norm": 0.047246869653463364,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 95709
+    },
+    {
+      "epoch": 9.881478422465413,
+      "grad_norm": 0.03901404142379761,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 95712
+    },
+    {
+      "epoch": 9.881788147842247,
+      "grad_norm": 0.048984501510858536,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 95715
+    },
+    {
+      "epoch": 9.88209787321908,
+      "grad_norm": 0.06687568873167038,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 95718
+    },
+    {
+      "epoch": 9.882407598595911,
+      "grad_norm": 0.06769860535860062,
+      "learning_rate": 0.01,
+      "loss": 1.9005,
+      "step": 95721
+    },
+    {
+      "epoch": 9.882717323972745,
+      "grad_norm": 0.08247470110654831,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 95724
+    },
+    {
+      "epoch": 9.883027049349577,
+      "grad_norm": 0.05325501412153244,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 95727
+    },
+    {
+      "epoch": 9.88333677472641,
+      "grad_norm": 0.04076659306883812,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 95730
+    },
+    {
+      "epoch": 9.883646500103241,
+      "grad_norm": 0.05497249215841293,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 95733
+    },
+    {
+      "epoch": 9.883956225480075,
+      "grad_norm": 0.06618578732013702,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 95736
+    },
+    {
+      "epoch": 9.884265950856907,
+      "grad_norm": 0.08831392973661423,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 95739
+    },
+    {
+      "epoch": 9.88457567623374,
+      "grad_norm": 0.07051166146993637,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 95742
+    },
+    {
+      "epoch": 9.884885401610571,
+      "grad_norm": 0.09569734334945679,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 95745
+    },
+    {
+      "epoch": 9.885195126987405,
+      "grad_norm": 0.10191222280263901,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 95748
+    },
+    {
+      "epoch": 9.885504852364237,
+      "grad_norm": 0.054515983909368515,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 95751
+    },
+    {
+      "epoch": 9.885814577741069,
+      "grad_norm": 0.03909364342689514,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 95754
+    },
+    {
+      "epoch": 9.886124303117903,
+      "grad_norm": 0.0464809313416481,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 95757
+    },
+    {
+      "epoch": 9.886434028494735,
+      "grad_norm": 0.03945815935730934,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 95760
+    },
+    {
+      "epoch": 9.886743753871567,
+      "grad_norm": 0.037694625556468964,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 95763
+    },
+    {
+      "epoch": 9.887053479248399,
+      "grad_norm": 0.04368235170841217,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 95766
+    },
+    {
+      "epoch": 9.887363204625233,
+      "grad_norm": 0.09463667869567871,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 95769
+    },
+    {
+      "epoch": 9.887672930002065,
+      "grad_norm": 0.07126586139202118,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 95772
+    },
+    {
+      "epoch": 9.887982655378897,
+      "grad_norm": 0.0477958507835865,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 95775
+    },
+    {
+      "epoch": 9.888292380755729,
+      "grad_norm": 0.04310915991663933,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 95778
+    },
+    {
+      "epoch": 9.888602106132563,
+      "grad_norm": 0.05492854118347168,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 95781
+    },
+    {
+      "epoch": 9.888911831509395,
+      "grad_norm": 0.12847760319709778,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 95784
+    },
+    {
+      "epoch": 9.889221556886227,
+      "grad_norm": 0.04450586438179016,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 95787
+    },
+    {
+      "epoch": 9.88953128226306,
+      "grad_norm": 0.07517846673727036,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 95790
+    },
+    {
+      "epoch": 9.889841007639893,
+      "grad_norm": 0.08548619598150253,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 95793
+    },
+    {
+      "epoch": 9.890150733016725,
+      "grad_norm": 0.037881866097450256,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 95796
+    },
+    {
+      "epoch": 9.890460458393557,
+      "grad_norm": 0.07290767133235931,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 95799
+    },
+    {
+      "epoch": 9.89077018377039,
+      "grad_norm": 0.10962492227554321,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 95802
+    },
+    {
+      "epoch": 9.891079909147223,
+      "grad_norm": 0.07267545163631439,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 95805
+    },
+    {
+      "epoch": 9.891389634524055,
+      "grad_norm": 0.04466653987765312,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 95808
+    },
+    {
+      "epoch": 9.891699359900889,
+      "grad_norm": 0.06062067300081253,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 95811
+    },
+    {
+      "epoch": 9.89200908527772,
+      "grad_norm": 0.05122390761971474,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 95814
+    },
+    {
+      "epoch": 9.892318810654553,
+      "grad_norm": 0.03910289704799652,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 95817
+    },
+    {
+      "epoch": 9.892628536031385,
+      "grad_norm": 0.03563802316784859,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 95820
+    },
+    {
+      "epoch": 9.892938261408219,
+      "grad_norm": 0.10363304615020752,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 95823
+    },
+    {
+      "epoch": 9.89324798678505,
+      "grad_norm": 0.04152027145028114,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 95826
+    },
+    {
+      "epoch": 9.893557712161883,
+      "grad_norm": 0.037295687943696976,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 95829
+    },
+    {
+      "epoch": 9.893867437538717,
+      "grad_norm": 0.1127936989068985,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 95832
+    },
+    {
+      "epoch": 9.894177162915549,
+      "grad_norm": 0.06783781945705414,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 95835
+    },
+    {
+      "epoch": 9.89448688829238,
+      "grad_norm": 0.07427812367677689,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 95838
+    },
+    {
+      "epoch": 9.894796613669213,
+      "grad_norm": 0.07258497923612595,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 95841
+    },
+    {
+      "epoch": 9.895106339046047,
+      "grad_norm": 0.04777050018310547,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 95844
+    },
+    {
+      "epoch": 9.895416064422879,
+      "grad_norm": 0.05332339182496071,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 95847
+    },
+    {
+      "epoch": 9.89572578979971,
+      "grad_norm": 0.07612593472003937,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 95850
+    },
+    {
+      "epoch": 9.896035515176543,
+      "grad_norm": 0.0600535124540329,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 95853
+    },
+    {
+      "epoch": 9.896345240553376,
+      "grad_norm": 0.05714597553014755,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 95856
+    },
+    {
+      "epoch": 9.896654965930209,
+      "grad_norm": 0.051216330379247665,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 95859
+    },
+    {
+      "epoch": 9.89696469130704,
+      "grad_norm": 0.05697392299771309,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 95862
+    },
+    {
+      "epoch": 9.897274416683874,
+      "grad_norm": 0.060521986335515976,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 95865
+    },
+    {
+      "epoch": 9.897584142060706,
+      "grad_norm": 0.04552103206515312,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 95868
+    },
+    {
+      "epoch": 9.897893867437539,
+      "grad_norm": 0.1266215741634369,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 95871
+    },
+    {
+      "epoch": 9.89820359281437,
+      "grad_norm": 0.08632038533687592,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 95874
+    },
+    {
+      "epoch": 9.898513318191204,
+      "grad_norm": 0.11756304651498795,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 95877
+    },
+    {
+      "epoch": 9.898823043568036,
+      "grad_norm": 0.08925943821668625,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 95880
+    },
+    {
+      "epoch": 9.899132768944868,
+      "grad_norm": 0.044926900416612625,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 95883
+    },
+    {
+      "epoch": 9.8994424943217,
+      "grad_norm": 0.05228542163968086,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 95886
+    },
+    {
+      "epoch": 9.899752219698534,
+      "grad_norm": 0.04257636144757271,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 95889
+    },
+    {
+      "epoch": 9.900061945075366,
+      "grad_norm": 0.04858774319291115,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 95892
+    },
+    {
+      "epoch": 9.900371670452198,
+      "grad_norm": 0.08439365029335022,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 95895
+    },
+    {
+      "epoch": 9.900681395829032,
+      "grad_norm": 0.061978358775377274,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 95898
+    },
+    {
+      "epoch": 9.900991121205864,
+      "grad_norm": 0.07678073644638062,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 95901
+    },
+    {
+      "epoch": 9.901300846582696,
+      "grad_norm": 0.08055569231510162,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 95904
+    },
+    {
+      "epoch": 9.901610571959528,
+      "grad_norm": 0.09484147280454636,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 95907
+    },
+    {
+      "epoch": 9.901920297336362,
+      "grad_norm": 0.15379561483860016,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 95910
+    },
+    {
+      "epoch": 9.902230022713194,
+      "grad_norm": 0.11704928427934647,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 95913
+    },
+    {
+      "epoch": 9.902539748090026,
+      "grad_norm": 0.054909687489271164,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 95916
+    },
+    {
+      "epoch": 9.90284947346686,
+      "grad_norm": 0.056868162006139755,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 95919
+    },
+    {
+      "epoch": 9.903159198843692,
+      "grad_norm": 0.05950642749667168,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 95922
+    },
+    {
+      "epoch": 9.903468924220524,
+      "grad_norm": 0.06228267773985863,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 95925
+    },
+    {
+      "epoch": 9.903778649597356,
+      "grad_norm": 0.053820542991161346,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 95928
+    },
+    {
+      "epoch": 9.90408837497419,
+      "grad_norm": 0.03804078698158264,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 95931
+    },
+    {
+      "epoch": 9.904398100351022,
+      "grad_norm": 0.04085548594594002,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 95934
+    },
+    {
+      "epoch": 9.904707825727854,
+      "grad_norm": 0.07046306878328323,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 95937
+    },
+    {
+      "epoch": 9.905017551104688,
+      "grad_norm": 0.07246625423431396,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 95940
+    },
+    {
+      "epoch": 9.90532727648152,
+      "grad_norm": 0.20051436126232147,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 95943
+    },
+    {
+      "epoch": 9.905637001858352,
+      "grad_norm": 0.06060566008090973,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 95946
+    },
+    {
+      "epoch": 9.905946727235184,
+      "grad_norm": 0.03977508842945099,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 95949
+    },
+    {
+      "epoch": 9.906256452612018,
+      "grad_norm": 0.055726293474435806,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 95952
+    },
+    {
+      "epoch": 9.90656617798885,
+      "grad_norm": 0.040252525359392166,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 95955
+    },
+    {
+      "epoch": 9.906875903365682,
+      "grad_norm": 0.03784322366118431,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 95958
+    },
+    {
+      "epoch": 9.907185628742514,
+      "grad_norm": 0.055127233266830444,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 95961
+    },
+    {
+      "epoch": 9.907495354119348,
+      "grad_norm": 0.10798919945955276,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 95964
+    },
+    {
+      "epoch": 9.90780507949618,
+      "grad_norm": 0.05120984837412834,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 95967
+    },
+    {
+      "epoch": 9.908114804873012,
+      "grad_norm": 0.08842770755290985,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 95970
+    },
+    {
+      "epoch": 9.908424530249846,
+      "grad_norm": 0.1395348310470581,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 95973
+    },
+    {
+      "epoch": 9.908734255626678,
+      "grad_norm": 0.15325050055980682,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 95976
+    },
+    {
+      "epoch": 9.90904398100351,
+      "grad_norm": 0.08843661844730377,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 95979
+    },
+    {
+      "epoch": 9.909353706380342,
+      "grad_norm": 0.06873669475317001,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 95982
+    },
+    {
+      "epoch": 9.909663431757176,
+      "grad_norm": 0.04710160940885544,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 95985
+    },
+    {
+      "epoch": 9.909973157134008,
+      "grad_norm": 0.083259716629982,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 95988
+    },
+    {
+      "epoch": 9.91028288251084,
+      "grad_norm": 0.048664819449186325,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 95991
+    },
+    {
+      "epoch": 9.910592607887672,
+      "grad_norm": 0.041180554777383804,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 95994
+    },
+    {
+      "epoch": 9.910902333264506,
+      "grad_norm": 0.06644929200410843,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 95997
+    },
+    {
+      "epoch": 9.911212058641338,
+      "grad_norm": 0.04565279185771942,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 96000
+    },
+    {
+      "epoch": 9.91152178401817,
+      "grad_norm": 0.09502280503511429,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 96003
+    },
+    {
+      "epoch": 9.911831509395004,
+      "grad_norm": 0.056951504200696945,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 96006
+    },
+    {
+      "epoch": 9.912141234771836,
+      "grad_norm": 0.04809977486729622,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 96009
+    },
+    {
+      "epoch": 9.912450960148668,
+      "grad_norm": 0.06078126281499863,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 96012
+    },
+    {
+      "epoch": 9.9127606855255,
+      "grad_norm": 0.054974962025880814,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 96015
+    },
+    {
+      "epoch": 9.913070410902334,
+      "grad_norm": 0.07565493881702423,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 96018
+    },
+    {
+      "epoch": 9.913380136279166,
+      "grad_norm": 0.1033167764544487,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 96021
+    },
+    {
+      "epoch": 9.913689861655998,
+      "grad_norm": 0.07701541483402252,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 96024
+    },
+    {
+      "epoch": 9.913999587032832,
+      "grad_norm": 0.1151469275355339,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 96027
+    },
+    {
+      "epoch": 9.914309312409664,
+      "grad_norm": 0.0574842132627964,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 96030
+    },
+    {
+      "epoch": 9.914619037786496,
+      "grad_norm": 0.09228459000587463,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 96033
+    },
+    {
+      "epoch": 9.914928763163328,
+      "grad_norm": 0.07984200119972229,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 96036
+    },
+    {
+      "epoch": 9.915238488540162,
+      "grad_norm": 0.07425150275230408,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 96039
+    },
+    {
+      "epoch": 9.915548213916994,
+      "grad_norm": 0.09937628358602524,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 96042
+    },
+    {
+      "epoch": 9.915857939293826,
+      "grad_norm": 0.11574780195951462,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 96045
+    },
+    {
+      "epoch": 9.91616766467066,
+      "grad_norm": 0.0494074672460556,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 96048
+    },
+    {
+      "epoch": 9.916477390047492,
+      "grad_norm": 0.04185641556978226,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 96051
+    },
+    {
+      "epoch": 9.916787115424324,
+      "grad_norm": 0.03591124713420868,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 96054
+    },
+    {
+      "epoch": 9.917096840801156,
+      "grad_norm": 0.059095386415719986,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 96057
+    },
+    {
+      "epoch": 9.91740656617799,
+      "grad_norm": 0.12536944448947906,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 96060
+    },
+    {
+      "epoch": 9.917716291554822,
+      "grad_norm": 0.04808446019887924,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 96063
+    },
+    {
+      "epoch": 9.918026016931654,
+      "grad_norm": 0.031612422317266464,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 96066
+    },
+    {
+      "epoch": 9.918335742308486,
+      "grad_norm": 0.035517603158950806,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 96069
+    },
+    {
+      "epoch": 9.91864546768532,
+      "grad_norm": 0.041796356439590454,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 96072
+    },
+    {
+      "epoch": 9.918955193062152,
+      "grad_norm": 0.07575595378875732,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 96075
+    },
+    {
+      "epoch": 9.919264918438984,
+      "grad_norm": 0.1338336169719696,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 96078
+    },
+    {
+      "epoch": 9.919574643815817,
+      "grad_norm": 0.0776827409863472,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 96081
+    },
+    {
+      "epoch": 9.91988436919265,
+      "grad_norm": 0.04113177955150604,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 96084
+    },
+    {
+      "epoch": 9.920194094569482,
+      "grad_norm": 0.04225945472717285,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 96087
+    },
+    {
+      "epoch": 9.920503819946314,
+      "grad_norm": 0.03149348869919777,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 96090
+    },
+    {
+      "epoch": 9.920813545323147,
+      "grad_norm": 0.03339305892586708,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 96093
+    },
+    {
+      "epoch": 9.92112327069998,
+      "grad_norm": 0.06829036772251129,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 96096
+    },
+    {
+      "epoch": 9.921432996076812,
+      "grad_norm": 0.0362083837389946,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 96099
+    },
+    {
+      "epoch": 9.921742721453644,
+      "grad_norm": 0.0873231515288353,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 96102
+    },
+    {
+      "epoch": 9.922052446830477,
+      "grad_norm": 0.1664249300956726,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 96105
+    },
+    {
+      "epoch": 9.92236217220731,
+      "grad_norm": 0.09485960006713867,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 96108
+    },
+    {
+      "epoch": 9.922671897584141,
+      "grad_norm": 0.09327783435583115,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 96111
+    },
+    {
+      "epoch": 9.922981622960975,
+      "grad_norm": 0.0972425639629364,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 96114
+    },
+    {
+      "epoch": 9.923291348337807,
+      "grad_norm": 0.05768739432096481,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 96117
+    },
+    {
+      "epoch": 9.92360107371464,
+      "grad_norm": 0.04108816757798195,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 96120
+    },
+    {
+      "epoch": 9.923910799091471,
+      "grad_norm": 0.056108977645635605,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 96123
+    },
+    {
+      "epoch": 9.924220524468305,
+      "grad_norm": 0.07638213783502579,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 96126
+    },
+    {
+      "epoch": 9.924530249845137,
+      "grad_norm": 0.06884434074163437,
+      "learning_rate": 0.01,
+      "loss": 1.9759,
+      "step": 96129
+    },
+    {
+      "epoch": 9.92483997522197,
+      "grad_norm": 0.05558433011174202,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 96132
+    },
+    {
+      "epoch": 9.925149700598803,
+      "grad_norm": 0.15693187713623047,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 96135
+    },
+    {
+      "epoch": 9.925459425975635,
+      "grad_norm": 0.043495237827301025,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 96138
+    },
+    {
+      "epoch": 9.925769151352467,
+      "grad_norm": 0.05494603142142296,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 96141
+    },
+    {
+      "epoch": 9.9260788767293,
+      "grad_norm": 0.05449117347598076,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 96144
+    },
+    {
+      "epoch": 9.926388602106133,
+      "grad_norm": 0.08051479607820511,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 96147
+    },
+    {
+      "epoch": 9.926698327482965,
+      "grad_norm": 0.034158267080783844,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 96150
+    },
+    {
+      "epoch": 9.927008052859797,
+      "grad_norm": 0.17496581375598907,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 96153
+    },
+    {
+      "epoch": 9.927317778236631,
+      "grad_norm": 0.1495855450630188,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 96156
+    },
+    {
+      "epoch": 9.927627503613463,
+      "grad_norm": 0.06771902740001678,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 96159
+    },
+    {
+      "epoch": 9.927937228990295,
+      "grad_norm": 0.12524180114269257,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 96162
+    },
+    {
+      "epoch": 9.928246954367127,
+      "grad_norm": 0.05685262009501457,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 96165
+    },
+    {
+      "epoch": 9.928556679743961,
+      "grad_norm": 0.039354316890239716,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 96168
+    },
+    {
+      "epoch": 9.928866405120793,
+      "grad_norm": 0.06542901694774628,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 96171
+    },
+    {
+      "epoch": 9.929176130497625,
+      "grad_norm": 0.0561458095908165,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 96174
+    },
+    {
+      "epoch": 9.929485855874457,
+      "grad_norm": 0.06581629812717438,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 96177
+    },
+    {
+      "epoch": 9.929795581251291,
+      "grad_norm": 0.14144189655780792,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 96180
+    },
+    {
+      "epoch": 9.930105306628123,
+      "grad_norm": 0.1060752272605896,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 96183
+    },
+    {
+      "epoch": 9.930415032004955,
+      "grad_norm": 0.05076698586344719,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 96186
+    },
+    {
+      "epoch": 9.930724757381789,
+      "grad_norm": 0.0615849643945694,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 96189
+    },
+    {
+      "epoch": 9.931034482758621,
+      "grad_norm": 0.06269994378089905,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 96192
+    },
+    {
+      "epoch": 9.931344208135453,
+      "grad_norm": 0.10822846740484238,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 96195
+    },
+    {
+      "epoch": 9.931653933512285,
+      "grad_norm": 0.13326896727085114,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 96198
+    },
+    {
+      "epoch": 9.931963658889119,
+      "grad_norm": 0.13338686525821686,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 96201
+    },
+    {
+      "epoch": 9.932273384265951,
+      "grad_norm": 0.054622832685709,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 96204
+    },
+    {
+      "epoch": 9.932583109642783,
+      "grad_norm": 0.046194273978471756,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 96207
+    },
+    {
+      "epoch": 9.932892835019615,
+      "grad_norm": 0.049110498279333115,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 96210
+    },
+    {
+      "epoch": 9.933202560396449,
+      "grad_norm": 0.05172931030392647,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 96213
+    },
+    {
+      "epoch": 9.933512285773281,
+      "grad_norm": 0.044886182993650436,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 96216
+    },
+    {
+      "epoch": 9.933822011150113,
+      "grad_norm": 0.04979904741048813,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 96219
+    },
+    {
+      "epoch": 9.934131736526947,
+      "grad_norm": 0.12607920169830322,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 96222
+    },
+    {
+      "epoch": 9.934441461903779,
+      "grad_norm": 0.05032966658473015,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 96225
+    },
+    {
+      "epoch": 9.934751187280611,
+      "grad_norm": 0.08132116496562958,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 96228
+    },
+    {
+      "epoch": 9.935060912657443,
+      "grad_norm": 0.07728183269500732,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 96231
+    },
+    {
+      "epoch": 9.935370638034277,
+      "grad_norm": 0.07342918962240219,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 96234
+    },
+    {
+      "epoch": 9.935680363411109,
+      "grad_norm": 0.056004371494054794,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 96237
+    },
+    {
+      "epoch": 9.93599008878794,
+      "grad_norm": 0.040742505341768265,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 96240
+    },
+    {
+      "epoch": 9.936299814164775,
+      "grad_norm": 0.1418849527835846,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 96243
+    },
+    {
+      "epoch": 9.936609539541607,
+      "grad_norm": 0.09691282361745834,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 96246
+    },
+    {
+      "epoch": 9.936919264918439,
+      "grad_norm": 0.10721616446971893,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 96249
+    },
+    {
+      "epoch": 9.93722899029527,
+      "grad_norm": 0.07174794375896454,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 96252
+    },
+    {
+      "epoch": 9.937538715672105,
+      "grad_norm": 0.04398215934634209,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 96255
+    },
+    {
+      "epoch": 9.937848441048937,
+      "grad_norm": 0.032949745655059814,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 96258
+    },
+    {
+      "epoch": 9.938158166425769,
+      "grad_norm": 0.036921992897987366,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 96261
+    },
+    {
+      "epoch": 9.938467891802603,
+      "grad_norm": 0.11214341968297958,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 96264
+    },
+    {
+      "epoch": 9.938777617179435,
+      "grad_norm": 0.10016003996133804,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 96267
+    },
+    {
+      "epoch": 9.939087342556267,
+      "grad_norm": 0.05769341066479683,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 96270
+    },
+    {
+      "epoch": 9.939397067933099,
+      "grad_norm": 0.046199750155210495,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 96273
+    },
+    {
+      "epoch": 9.939706793309933,
+      "grad_norm": 0.03568249195814133,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 96276
+    },
+    {
+      "epoch": 9.940016518686765,
+      "grad_norm": 0.031168581917881966,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 96279
+    },
+    {
+      "epoch": 9.940326244063597,
+      "grad_norm": 0.038201022893190384,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 96282
+    },
+    {
+      "epoch": 9.940635969440429,
+      "grad_norm": 0.03458917513489723,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 96285
+    },
+    {
+      "epoch": 9.940945694817263,
+      "grad_norm": 0.10806361585855484,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 96288
+    },
+    {
+      "epoch": 9.941255420194095,
+      "grad_norm": 0.07543551176786423,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 96291
+    },
+    {
+      "epoch": 9.941565145570927,
+      "grad_norm": 0.0484190434217453,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 96294
+    },
+    {
+      "epoch": 9.94187487094776,
+      "grad_norm": 0.08257054537534714,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 96297
+    },
+    {
+      "epoch": 9.942184596324593,
+      "grad_norm": 0.08587981015443802,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 96300
+    },
+    {
+      "epoch": 9.942494321701425,
+      "grad_norm": 0.04479651898145676,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 96303
+    },
+    {
+      "epoch": 9.942804047078257,
+      "grad_norm": 0.07514610141515732,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 96306
+    },
+    {
+      "epoch": 9.94311377245509,
+      "grad_norm": 0.043522909283638,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 96309
+    },
+    {
+      "epoch": 9.943423497831922,
+      "grad_norm": 0.034205563366413116,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 96312
+    },
+    {
+      "epoch": 9.943733223208755,
+      "grad_norm": 0.04049249365925789,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 96315
+    },
+    {
+      "epoch": 9.944042948585587,
+      "grad_norm": 0.05075372755527496,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 96318
+    },
+    {
+      "epoch": 9.94435267396242,
+      "grad_norm": 0.07100080698728561,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 96321
+    },
+    {
+      "epoch": 9.944662399339252,
+      "grad_norm": 0.09260247647762299,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 96324
+    },
+    {
+      "epoch": 9.944972124716084,
+      "grad_norm": 0.06434981524944305,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 96327
+    },
+    {
+      "epoch": 9.945281850092918,
+      "grad_norm": 0.04493270814418793,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 96330
+    },
+    {
+      "epoch": 9.94559157546975,
+      "grad_norm": 0.0432509109377861,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 96333
+    },
+    {
+      "epoch": 9.945901300846582,
+      "grad_norm": 0.0948626771569252,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 96336
+    },
+    {
+      "epoch": 9.946211026223414,
+      "grad_norm": 0.10249466449022293,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 96339
+    },
+    {
+      "epoch": 9.946520751600248,
+      "grad_norm": 0.130013570189476,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 96342
+    },
+    {
+      "epoch": 9.94683047697708,
+      "grad_norm": 0.048307985067367554,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 96345
+    },
+    {
+      "epoch": 9.947140202353912,
+      "grad_norm": 0.08395209163427353,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 96348
+    },
+    {
+      "epoch": 9.947449927730746,
+      "grad_norm": 0.05564103275537491,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 96351
+    },
+    {
+      "epoch": 9.947759653107578,
+      "grad_norm": 0.0359511636197567,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 96354
+    },
+    {
+      "epoch": 9.94806937848441,
+      "grad_norm": 0.04344267398118973,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 96357
+    },
+    {
+      "epoch": 9.948379103861242,
+      "grad_norm": 0.048588648438453674,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 96360
+    },
+    {
+      "epoch": 9.948688829238076,
+      "grad_norm": 0.12489151209592819,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 96363
+    },
+    {
+      "epoch": 9.948998554614908,
+      "grad_norm": 0.07510516047477722,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 96366
+    },
+    {
+      "epoch": 9.94930827999174,
+      "grad_norm": 0.09619615226984024,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 96369
+    },
+    {
+      "epoch": 9.949618005368574,
+      "grad_norm": 0.10213521867990494,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 96372
+    },
+    {
+      "epoch": 9.949927730745406,
+      "grad_norm": 0.07023943960666656,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 96375
+    },
+    {
+      "epoch": 9.950237456122238,
+      "grad_norm": 0.06509052217006683,
+      "learning_rate": 0.01,
+      "loss": 1.9042,
+      "step": 96378
+    },
+    {
+      "epoch": 9.95054718149907,
+      "grad_norm": 0.0720127746462822,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 96381
+    },
+    {
+      "epoch": 9.950856906875904,
+      "grad_norm": 0.0382302924990654,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 96384
+    },
+    {
+      "epoch": 9.951166632252736,
+      "grad_norm": 0.030933093279600143,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 96387
+    },
+    {
+      "epoch": 9.951476357629568,
+      "grad_norm": 0.05807695537805557,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 96390
+    },
+    {
+      "epoch": 9.9517860830064,
+      "grad_norm": 0.08505658805370331,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 96393
+    },
+    {
+      "epoch": 9.952095808383234,
+      "grad_norm": 0.053451940417289734,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 96396
+    },
+    {
+      "epoch": 9.952405533760066,
+      "grad_norm": 0.09265648573637009,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 96399
+    },
+    {
+      "epoch": 9.952715259136898,
+      "grad_norm": 0.10184822976589203,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 96402
+    },
+    {
+      "epoch": 9.953024984513732,
+      "grad_norm": 0.10975353419780731,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 96405
+    },
+    {
+      "epoch": 9.953334709890564,
+      "grad_norm": 0.14230191707611084,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 96408
+    },
+    {
+      "epoch": 9.953644435267396,
+      "grad_norm": 0.10149675607681274,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 96411
+    },
+    {
+      "epoch": 9.953954160644228,
+      "grad_norm": 0.04591508209705353,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 96414
+    },
+    {
+      "epoch": 9.954263886021062,
+      "grad_norm": 0.04944121092557907,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 96417
+    },
+    {
+      "epoch": 9.954573611397894,
+      "grad_norm": 0.046226534992456436,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 96420
+    },
+    {
+      "epoch": 9.954883336774726,
+      "grad_norm": 0.04897132143378258,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 96423
+    },
+    {
+      "epoch": 9.955193062151558,
+      "grad_norm": 0.049237675964832306,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 96426
+    },
+    {
+      "epoch": 9.955502787528392,
+      "grad_norm": 0.10047107934951782,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 96429
+    },
+    {
+      "epoch": 9.955812512905224,
+      "grad_norm": 0.05859467759728432,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 96432
+    },
+    {
+      "epoch": 9.956122238282056,
+      "grad_norm": 0.1209932193160057,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 96435
+    },
+    {
+      "epoch": 9.95643196365889,
+      "grad_norm": 0.06430116295814514,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 96438
+    },
+    {
+      "epoch": 9.956741689035722,
+      "grad_norm": 0.03878641873598099,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 96441
+    },
+    {
+      "epoch": 9.957051414412554,
+      "grad_norm": 0.03463767468929291,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 96444
+    },
+    {
+      "epoch": 9.957361139789386,
+      "grad_norm": 0.1033138632774353,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 96447
+    },
+    {
+      "epoch": 9.95767086516622,
+      "grad_norm": 0.07953305542469025,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 96450
+    },
+    {
+      "epoch": 9.957980590543052,
+      "grad_norm": 0.06403129547834396,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 96453
+    },
+    {
+      "epoch": 9.958290315919884,
+      "grad_norm": 0.06012757867574692,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 96456
+    },
+    {
+      "epoch": 9.958600041296718,
+      "grad_norm": 0.042264413088560104,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 96459
+    },
+    {
+      "epoch": 9.95890976667355,
+      "grad_norm": 0.08021681010723114,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 96462
+    },
+    {
+      "epoch": 9.959219492050382,
+      "grad_norm": 0.06785903871059418,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 96465
+    },
+    {
+      "epoch": 9.959529217427214,
+      "grad_norm": 0.07423307001590729,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 96468
+    },
+    {
+      "epoch": 9.959838942804048,
+      "grad_norm": 0.09907034784555435,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 96471
+    },
+    {
+      "epoch": 9.96014866818088,
+      "grad_norm": 0.04186716303229332,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 96474
+    },
+    {
+      "epoch": 9.960458393557712,
+      "grad_norm": 0.09196318686008453,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 96477
+    },
+    {
+      "epoch": 9.960768118934546,
+      "grad_norm": 0.139618918299675,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 96480
+    },
+    {
+      "epoch": 9.961077844311378,
+      "grad_norm": 0.06983165442943573,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 96483
+    },
+    {
+      "epoch": 9.96138756968821,
+      "grad_norm": 0.05909397080540657,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 96486
+    },
+    {
+      "epoch": 9.961697295065042,
+      "grad_norm": 0.05737200006842613,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 96489
+    },
+    {
+      "epoch": 9.962007020441876,
+      "grad_norm": 0.042182207107543945,
+      "learning_rate": 0.01,
+      "loss": 1.894,
+      "step": 96492
+    },
+    {
+      "epoch": 9.962316745818708,
+      "grad_norm": 0.052041444927453995,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 96495
+    },
+    {
+      "epoch": 9.96262647119554,
+      "grad_norm": 0.05366899445652962,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 96498
+    },
+    {
+      "epoch": 9.962936196572372,
+      "grad_norm": 0.05489904433488846,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 96501
+    },
+    {
+      "epoch": 9.963245921949206,
+      "grad_norm": 0.0658499151468277,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 96504
+    },
+    {
+      "epoch": 9.963555647326038,
+      "grad_norm": 0.05275774002075195,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 96507
+    },
+    {
+      "epoch": 9.96386537270287,
+      "grad_norm": 0.037739306688308716,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 96510
+    },
+    {
+      "epoch": 9.964175098079703,
+      "grad_norm": 0.04795077443122864,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 96513
+    },
+    {
+      "epoch": 9.964484823456536,
+      "grad_norm": 0.059451814740896225,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 96516
+    },
+    {
+      "epoch": 9.964794548833368,
+      "grad_norm": 0.05571671202778816,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 96519
+    },
+    {
+      "epoch": 9.9651042742102,
+      "grad_norm": 0.1349763125181198,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 96522
+    },
+    {
+      "epoch": 9.965413999587033,
+      "grad_norm": 0.11729591339826584,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 96525
+    },
+    {
+      "epoch": 9.965723724963865,
+      "grad_norm": 0.0958571508526802,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 96528
+    },
+    {
+      "epoch": 9.966033450340698,
+      "grad_norm": 0.06040540337562561,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 96531
+    },
+    {
+      "epoch": 9.96634317571753,
+      "grad_norm": 0.04997546970844269,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 96534
+    },
+    {
+      "epoch": 9.966652901094363,
+      "grad_norm": 0.043729156255722046,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 96537
+    },
+    {
+      "epoch": 9.966962626471195,
+      "grad_norm": 0.11820992082357407,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 96540
+    },
+    {
+      "epoch": 9.967272351848028,
+      "grad_norm": 0.04722660034894943,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 96543
+    },
+    {
+      "epoch": 9.967582077224861,
+      "grad_norm": 0.0528867170214653,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 96546
+    },
+    {
+      "epoch": 9.967891802601693,
+      "grad_norm": 0.0687720999121666,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 96549
+    },
+    {
+      "epoch": 9.968201527978525,
+      "grad_norm": 0.06109067052602768,
+      "learning_rate": 0.01,
+      "loss": 1.8936,
+      "step": 96552
+    },
+    {
+      "epoch": 9.968511253355357,
+      "grad_norm": 0.03622574731707573,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 96555
+    },
+    {
+      "epoch": 9.968820978732191,
+      "grad_norm": 0.05403563007712364,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 96558
+    },
+    {
+      "epoch": 9.969130704109023,
+      "grad_norm": 0.05532357841730118,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 96561
+    },
+    {
+      "epoch": 9.969440429485855,
+      "grad_norm": 0.04689712077379227,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 96564
+    },
+    {
+      "epoch": 9.96975015486269,
+      "grad_norm": 0.058481086045503616,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 96567
+    },
+    {
+      "epoch": 9.970059880239521,
+      "grad_norm": 0.12241359055042267,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 96570
+    },
+    {
+      "epoch": 9.970369605616353,
+      "grad_norm": 0.1142013892531395,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 96573
+    },
+    {
+      "epoch": 9.970679330993185,
+      "grad_norm": 0.052777521312236786,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 96576
+    },
+    {
+      "epoch": 9.97098905637002,
+      "grad_norm": 0.1075984537601471,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 96579
+    },
+    {
+      "epoch": 9.971298781746851,
+      "grad_norm": 0.0782780796289444,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 96582
+    },
+    {
+      "epoch": 9.971608507123683,
+      "grad_norm": 0.05531663820147514,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 96585
+    },
+    {
+      "epoch": 9.971918232500517,
+      "grad_norm": 0.09160934388637543,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 96588
+    },
+    {
+      "epoch": 9.97222795787735,
+      "grad_norm": 0.09960715472698212,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 96591
+    },
+    {
+      "epoch": 9.972537683254181,
+      "grad_norm": 0.05900340527296066,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 96594
+    },
+    {
+      "epoch": 9.972847408631013,
+      "grad_norm": 0.04487355425953865,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 96597
+    },
+    {
+      "epoch": 9.973157134007847,
+      "grad_norm": 0.036559127271175385,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 96600
+    },
+    {
+      "epoch": 9.97346685938468,
+      "grad_norm": 0.04605504497885704,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 96603
+    },
+    {
+      "epoch": 9.973776584761511,
+      "grad_norm": 0.04498891904950142,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 96606
+    },
+    {
+      "epoch": 9.974086310138343,
+      "grad_norm": 0.07214579731225967,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 96609
+    },
+    {
+      "epoch": 9.974396035515177,
+      "grad_norm": 0.0739932432770729,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 96612
+    },
+    {
+      "epoch": 9.974705760892009,
+      "grad_norm": 0.051533691585063934,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 96615
+    },
+    {
+      "epoch": 9.975015486268841,
+      "grad_norm": 0.048333242535591125,
+      "learning_rate": 0.01,
+      "loss": 1.9022,
+      "step": 96618
+    },
+    {
+      "epoch": 9.975325211645675,
+      "grad_norm": 0.05780849605798721,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 96621
+    },
+    {
+      "epoch": 9.975634937022507,
+      "grad_norm": 0.050407566130161285,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 96624
+    },
+    {
+      "epoch": 9.975944662399339,
+      "grad_norm": 0.03372873738408089,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 96627
+    },
+    {
+      "epoch": 9.976254387776171,
+      "grad_norm": 0.06596677005290985,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 96630
+    },
+    {
+      "epoch": 9.976564113153005,
+      "grad_norm": 0.05827241763472557,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 96633
+    },
+    {
+      "epoch": 9.976873838529837,
+      "grad_norm": 0.06429224461317062,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 96636
+    },
+    {
+      "epoch": 9.977183563906669,
+      "grad_norm": 0.05854794755578041,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 96639
+    },
+    {
+      "epoch": 9.977493289283501,
+      "grad_norm": 0.0726938396692276,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 96642
+    },
+    {
+      "epoch": 9.977803014660335,
+      "grad_norm": 0.12848325073719025,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 96645
+    },
+    {
+      "epoch": 9.978112740037167,
+      "grad_norm": 0.049772996455430984,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 96648
+    },
+    {
+      "epoch": 9.978422465413999,
+      "grad_norm": 0.0814037024974823,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 96651
+    },
+    {
+      "epoch": 9.978732190790833,
+      "grad_norm": 0.03594604507088661,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 96654
+    },
+    {
+      "epoch": 9.979041916167665,
+      "grad_norm": 0.04974773898720741,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 96657
+    },
+    {
+      "epoch": 9.979351641544497,
+      "grad_norm": 0.07818830758333206,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 96660
+    },
+    {
+      "epoch": 9.979661366921329,
+      "grad_norm": 0.070968858897686,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 96663
+    },
+    {
+      "epoch": 9.979971092298163,
+      "grad_norm": 0.043815940618515015,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 96666
+    },
+    {
+      "epoch": 9.980280817674995,
+      "grad_norm": 0.10502909123897552,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 96669
+    },
+    {
+      "epoch": 9.980590543051827,
+      "grad_norm": 0.09334055334329605,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 96672
+    },
+    {
+      "epoch": 9.98090026842866,
+      "grad_norm": 0.07804276049137115,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 96675
+    },
+    {
+      "epoch": 9.981209993805493,
+      "grad_norm": 0.04819759726524353,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 96678
+    },
+    {
+      "epoch": 9.981519719182325,
+      "grad_norm": 0.09342513978481293,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 96681
+    },
+    {
+      "epoch": 9.981829444559157,
+      "grad_norm": 0.08521047979593277,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 96684
+    },
+    {
+      "epoch": 9.98213916993599,
+      "grad_norm": 0.14616277813911438,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 96687
+    },
+    {
+      "epoch": 9.982448895312823,
+      "grad_norm": 0.1458623856306076,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 96690
+    },
+    {
+      "epoch": 9.982758620689655,
+      "grad_norm": 0.09821382910013199,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 96693
+    },
+    {
+      "epoch": 9.983068346066489,
+      "grad_norm": 0.05744251608848572,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 96696
+    },
+    {
+      "epoch": 9.98337807144332,
+      "grad_norm": 0.04756811261177063,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 96699
+    },
+    {
+      "epoch": 9.983687796820153,
+      "grad_norm": 0.045846566557884216,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 96702
+    },
+    {
+      "epoch": 9.983997522196985,
+      "grad_norm": 0.036973029375076294,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 96705
+    },
+    {
+      "epoch": 9.984307247573819,
+      "grad_norm": 0.03365800529718399,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 96708
+    },
+    {
+      "epoch": 9.98461697295065,
+      "grad_norm": 0.07403205335140228,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 96711
+    },
+    {
+      "epoch": 9.984926698327483,
+      "grad_norm": 0.07294251024723053,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 96714
+    },
+    {
+      "epoch": 9.985236423704315,
+      "grad_norm": 0.055860526859760284,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 96717
+    },
+    {
+      "epoch": 9.985546149081149,
+      "grad_norm": 0.122382752597332,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 96720
+    },
+    {
+      "epoch": 9.98585587445798,
+      "grad_norm": 0.15544384717941284,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 96723
+    },
+    {
+      "epoch": 9.986165599834813,
+      "grad_norm": 0.11838815361261368,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 96726
+    },
+    {
+      "epoch": 9.986475325211646,
+      "grad_norm": 0.048733849078416824,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 96729
+    },
+    {
+      "epoch": 9.986785050588479,
+      "grad_norm": 0.05413370579481125,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 96732
+    },
+    {
+      "epoch": 9.98709477596531,
+      "grad_norm": 0.04110780730843544,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 96735
+    },
+    {
+      "epoch": 9.987404501342143,
+      "grad_norm": 0.045921992510557175,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 96738
+    },
+    {
+      "epoch": 9.987714226718976,
+      "grad_norm": 0.09011565148830414,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 96741
+    },
+    {
+      "epoch": 9.988023952095809,
+      "grad_norm": 0.06843160837888718,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 96744
+    },
+    {
+      "epoch": 9.98833367747264,
+      "grad_norm": 0.042586639523506165,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 96747
+    },
+    {
+      "epoch": 9.988643402849473,
+      "grad_norm": 0.06954126805067062,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 96750
+    },
+    {
+      "epoch": 9.988953128226306,
+      "grad_norm": 0.07393761724233627,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 96753
+    },
+    {
+      "epoch": 9.989262853603138,
+      "grad_norm": 0.10468722879886627,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 96756
+    },
+    {
+      "epoch": 9.98957257897997,
+      "grad_norm": 0.05889539793133736,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 96759
+    },
+    {
+      "epoch": 9.989882304356804,
+      "grad_norm": 0.07115521281957626,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 96762
+    },
+    {
+      "epoch": 9.990192029733636,
+      "grad_norm": 0.07057791203260422,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 96765
+    },
+    {
+      "epoch": 9.990501755110468,
+      "grad_norm": 0.062427669763565063,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 96768
+    },
+    {
+      "epoch": 9.9908114804873,
+      "grad_norm": 0.04834975302219391,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 96771
+    },
+    {
+      "epoch": 9.991121205864134,
+      "grad_norm": 0.03624477609992027,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 96774
+    },
+    {
+      "epoch": 9.991430931240966,
+      "grad_norm": 0.03459803760051727,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 96777
+    },
+    {
+      "epoch": 9.991740656617798,
+      "grad_norm": 0.11813360452651978,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 96780
+    },
+    {
+      "epoch": 9.992050381994632,
+      "grad_norm": 0.07058688253164291,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 96783
+    },
+    {
+      "epoch": 9.992360107371464,
+      "grad_norm": 0.09416121989488602,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 96786
+    },
+    {
+      "epoch": 9.992669832748296,
+      "grad_norm": 0.07006985694169998,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 96789
+    },
+    {
+      "epoch": 9.992979558125128,
+      "grad_norm": 0.0769728347659111,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 96792
+    },
+    {
+      "epoch": 9.993289283501962,
+      "grad_norm": 0.15726791322231293,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 96795
+    },
+    {
+      "epoch": 9.993599008878794,
+      "grad_norm": 0.07349522411823273,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 96798
+    },
+    {
+      "epoch": 9.993908734255626,
+      "grad_norm": 0.04822126403450966,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 96801
+    },
+    {
+      "epoch": 9.99421845963246,
+      "grad_norm": 0.03172096982598305,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 96804
+    },
+    {
+      "epoch": 9.994528185009292,
+      "grad_norm": 0.03926796466112137,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 96807
+    },
+    {
+      "epoch": 9.994837910386124,
+      "grad_norm": 0.057607777416706085,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 96810
+    },
+    {
+      "epoch": 9.995147635762956,
+      "grad_norm": 0.07060913741588593,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 96813
+    },
+    {
+      "epoch": 9.99545736113979,
+      "grad_norm": 0.10870376974344254,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 96816
+    },
+    {
+      "epoch": 9.995767086516622,
+      "grad_norm": 0.05336311459541321,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 96819
+    },
+    {
+      "epoch": 9.996076811893454,
+      "grad_norm": 0.10393037647008896,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 96822
+    },
+    {
+      "epoch": 9.996386537270286,
+      "grad_norm": 0.05877384915947914,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 96825
+    },
+    {
+      "epoch": 9.99669626264712,
+      "grad_norm": 0.07578223943710327,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 96828
+    },
+    {
+      "epoch": 9.997005988023952,
+      "grad_norm": 0.046317122876644135,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 96831
+    },
+    {
+      "epoch": 9.997315713400784,
+      "grad_norm": 0.0316709503531456,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 96834
+    },
+    {
+      "epoch": 9.997625438777618,
+      "grad_norm": 0.04384656623005867,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 96837
+    },
+    {
+      "epoch": 9.99793516415445,
+      "grad_norm": 0.061252620071172714,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 96840
+    },
+    {
+      "epoch": 9.998244889531282,
+      "grad_norm": 0.08033587783575058,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 96843
+    },
+    {
+      "epoch": 9.998554614908114,
+      "grad_norm": 0.09759991616010666,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 96846
+    },
+    {
+      "epoch": 9.998864340284948,
+      "grad_norm": 0.060363709926605225,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 96849
+    },
+    {
+      "epoch": 9.99917406566178,
+      "grad_norm": 0.06367176026105881,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 96852
+    },
+    {
+      "epoch": 9.999483791038612,
+      "grad_norm": 0.08114373683929443,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 96855
+    },
+    {
+      "epoch": 9.999793516415444,
+      "grad_norm": 0.04385897517204285,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 96858
+    },
+    {
+      "epoch": 9.947725172024237,
+      "grad_norm": 0.08046164363622665,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 96861
+    },
+    {
+      "epoch": 9.948033275136078,
+      "grad_norm": 0.0815848708152771,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 96864
+    },
+    {
+      "epoch": 9.948341378247921,
+      "grad_norm": 0.06400562822818756,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 96867
+    },
+    {
+      "epoch": 9.948649481359762,
+      "grad_norm": 0.063548743724823,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 96870
+    },
+    {
+      "epoch": 9.948957584471604,
+      "grad_norm": 0.09831225126981735,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 96873
+    },
+    {
+      "epoch": 9.949265687583445,
+      "grad_norm": 0.03349125012755394,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 96876
+    },
+    {
+      "epoch": 9.949573790695286,
+      "grad_norm": 0.08987690508365631,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 96879
+    },
+    {
+      "epoch": 9.949881893807127,
+      "grad_norm": 0.20651590824127197,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 96882
+    },
+    {
+      "epoch": 9.950189996918969,
+      "grad_norm": 0.09576163440942764,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 96885
+    },
+    {
+      "epoch": 9.95049810003081,
+      "grad_norm": 0.07056623697280884,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 96888
+    },
+    {
+      "epoch": 9.950806203142651,
+      "grad_norm": 0.06381063908338547,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 96891
+    },
+    {
+      "epoch": 9.951114306254492,
+      "grad_norm": 0.04859933257102966,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 96894
+    },
+    {
+      "epoch": 9.951422409366334,
+      "grad_norm": 0.05717281252145767,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 96897
+    },
+    {
+      "epoch": 9.951730512478177,
+      "grad_norm": 0.06399691849946976,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 96900
+    },
+    {
+      "epoch": 9.952038615590018,
+      "grad_norm": 0.04647698998451233,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 96903
+    },
+    {
+      "epoch": 9.95234671870186,
+      "grad_norm": 0.0519287995994091,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 96906
+    },
+    {
+      "epoch": 9.9526548218137,
+      "grad_norm": 0.03411509841680527,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 96909
+    },
+    {
+      "epoch": 9.952962924925542,
+      "grad_norm": 0.03161589428782463,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 96912
+    },
+    {
+      "epoch": 9.953271028037383,
+      "grad_norm": 0.0575750507414341,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 96915
+    },
+    {
+      "epoch": 9.953579131149224,
+      "grad_norm": 0.19468317925930023,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 96918
+    },
+    {
+      "epoch": 9.953887234261066,
+      "grad_norm": 0.1294582337141037,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 96921
+    },
+    {
+      "epoch": 9.954195337372907,
+      "grad_norm": 0.04406146705150604,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 96924
+    },
+    {
+      "epoch": 9.954503440484748,
+      "grad_norm": 0.05637388303875923,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 96927
+    },
+    {
+      "epoch": 9.954811543596591,
+      "grad_norm": 0.03598056361079216,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 96930
+    },
+    {
+      "epoch": 9.955119646708432,
+      "grad_norm": 0.03543659299612045,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 96933
+    },
+    {
+      "epoch": 9.955427749820274,
+      "grad_norm": 0.04690230265259743,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 96936
+    },
+    {
+      "epoch": 9.955735852932115,
+      "grad_norm": 0.06882453709840775,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 96939
+    },
+    {
+      "epoch": 9.956043956043956,
+      "grad_norm": 0.05158957466483116,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 96942
+    },
+    {
+      "epoch": 9.956352059155797,
+      "grad_norm": 0.04015819728374481,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 96945
+    },
+    {
+      "epoch": 9.956660162267639,
+      "grad_norm": 0.0542755052447319,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 96948
+    },
+    {
+      "epoch": 9.95696826537948,
+      "grad_norm": 0.03693319484591484,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 96951
+    },
+    {
+      "epoch": 9.957276368491321,
+      "grad_norm": 0.07938143610954285,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 96954
+    },
+    {
+      "epoch": 9.957584471603163,
+      "grad_norm": 0.09673090279102325,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 96957
+    },
+    {
+      "epoch": 9.957892574715004,
+      "grad_norm": 0.07723116129636765,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 96960
+    },
+    {
+      "epoch": 9.958200677826847,
+      "grad_norm": 0.09069114923477173,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 96963
+    },
+    {
+      "epoch": 9.958508780938688,
+      "grad_norm": 0.0510699637234211,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 96966
+    },
+    {
+      "epoch": 9.95881688405053,
+      "grad_norm": 0.07924008369445801,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 96969
+    },
+    {
+      "epoch": 9.95912498716237,
+      "grad_norm": 0.07602304220199585,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 96972
+    },
+    {
+      "epoch": 9.959433090274212,
+      "grad_norm": 0.06452027708292007,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 96975
+    },
+    {
+      "epoch": 9.959741193386053,
+      "grad_norm": 0.053950998932123184,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 96978
+    },
+    {
+      "epoch": 9.960049296497894,
+      "grad_norm": 0.05036810413002968,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 96981
+    },
+    {
+      "epoch": 9.960357399609736,
+      "grad_norm": 0.05384952202439308,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 96984
+    },
+    {
+      "epoch": 9.960665502721577,
+      "grad_norm": 0.11406214535236359,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 96987
+    },
+    {
+      "epoch": 9.960973605833418,
+      "grad_norm": 0.05955125764012337,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 96990
+    },
+    {
+      "epoch": 9.961281708945261,
+      "grad_norm": 0.09721150994300842,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 96993
+    },
+    {
+      "epoch": 9.961589812057102,
+      "grad_norm": 0.08787155151367188,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 96996
+    },
+    {
+      "epoch": 9.961897915168944,
+      "grad_norm": 0.040302030742168427,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 96999
+    },
+    {
+      "epoch": 9.962206018280785,
+      "grad_norm": 0.05395068973302841,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 97002
+    },
+    {
+      "epoch": 9.962514121392626,
+      "grad_norm": 0.0569995641708374,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 97005
+    },
+    {
+      "epoch": 9.962822224504468,
+      "grad_norm": 0.05147429555654526,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 97008
+    },
+    {
+      "epoch": 9.963130327616309,
+      "grad_norm": 0.1475445181131363,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 97011
+    },
+    {
+      "epoch": 9.96343843072815,
+      "grad_norm": 0.1312050074338913,
+      "learning_rate": 0.01,
+      "loss": 1.9815,
+      "step": 97014
+    },
+    {
+      "epoch": 9.963746533839991,
+      "grad_norm": 0.08530642837285995,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 97017
+    },
+    {
+      "epoch": 9.964054636951833,
+      "grad_norm": 0.049393344670534134,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 97020
+    },
+    {
+      "epoch": 9.964362740063674,
+      "grad_norm": 0.05835723131895065,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 97023
+    },
+    {
+      "epoch": 9.964670843175517,
+      "grad_norm": 0.05166759714484215,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 97026
+    },
+    {
+      "epoch": 9.964978946287358,
+      "grad_norm": 0.045471616089344025,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 97029
+    },
+    {
+      "epoch": 9.9652870493992,
+      "grad_norm": 0.07382349669933319,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 97032
+    },
+    {
+      "epoch": 9.96559515251104,
+      "grad_norm": 0.1263457089662552,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 97035
+    },
+    {
+      "epoch": 9.965903255622882,
+      "grad_norm": 0.04571104049682617,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 97038
+    },
+    {
+      "epoch": 9.966211358734723,
+      "grad_norm": 0.04100766405463219,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 97041
+    },
+    {
+      "epoch": 9.966519461846564,
+      "grad_norm": 0.062699094414711,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 97044
+    },
+    {
+      "epoch": 9.966827564958406,
+      "grad_norm": 0.05958600714802742,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 97047
+    },
+    {
+      "epoch": 9.967135668070247,
+      "grad_norm": 0.037050146609544754,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 97050
+    },
+    {
+      "epoch": 9.967443771182088,
+      "grad_norm": 0.09023641049861908,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 97053
+    },
+    {
+      "epoch": 9.967751874293931,
+      "grad_norm": 0.09687948226928711,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 97056
+    },
+    {
+      "epoch": 9.968059977405773,
+      "grad_norm": 0.08525572717189789,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 97059
+    },
+    {
+      "epoch": 9.968368080517614,
+      "grad_norm": 0.07634325325489044,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 97062
+    },
+    {
+      "epoch": 9.968676183629455,
+      "grad_norm": 0.07752718031406403,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 97065
+    },
+    {
+      "epoch": 9.968984286741296,
+      "grad_norm": 0.09133943915367126,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 97068
+    },
+    {
+      "epoch": 9.969292389853138,
+      "grad_norm": 0.08312353491783142,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 97071
+    },
+    {
+      "epoch": 9.969600492964979,
+      "grad_norm": 0.06459927558898926,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 97074
+    },
+    {
+      "epoch": 9.96990859607682,
+      "grad_norm": 0.06323962658643723,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 97077
+    },
+    {
+      "epoch": 9.970216699188661,
+      "grad_norm": 0.063611701130867,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 97080
+    },
+    {
+      "epoch": 9.970524802300503,
+      "grad_norm": 0.052672822028398514,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 97083
+    },
+    {
+      "epoch": 9.970832905412344,
+      "grad_norm": 0.09678171575069427,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 97086
+    },
+    {
+      "epoch": 9.971141008524185,
+      "grad_norm": 0.131818950176239,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 97089
+    },
+    {
+      "epoch": 9.971449111636028,
+      "grad_norm": 0.05830173194408417,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 97092
+    },
+    {
+      "epoch": 9.97175721474787,
+      "grad_norm": 0.09154964983463287,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 97095
+    },
+    {
+      "epoch": 9.97206531785971,
+      "grad_norm": 0.0545133538544178,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 97098
+    },
+    {
+      "epoch": 9.972373420971552,
+      "grad_norm": 0.03951120749115944,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 97101
+    },
+    {
+      "epoch": 9.972681524083393,
+      "grad_norm": 0.03921626880764961,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 97104
+    },
+    {
+      "epoch": 9.972989627195235,
+      "grad_norm": 0.06999723613262177,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 97107
+    },
+    {
+      "epoch": 9.973297730307076,
+      "grad_norm": 0.06404420733451843,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 97110
+    },
+    {
+      "epoch": 9.973605833418917,
+      "grad_norm": 0.12217292934656143,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 97113
+    },
+    {
+      "epoch": 9.973913936530758,
+      "grad_norm": 0.044860534369945526,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 97116
+    },
+    {
+      "epoch": 9.9742220396426,
+      "grad_norm": 0.05524575710296631,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 97119
+    },
+    {
+      "epoch": 9.974530142754443,
+      "grad_norm": 0.05118962749838829,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 97122
+    },
+    {
+      "epoch": 9.974838245866284,
+      "grad_norm": 0.08901583403348923,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 97125
+    },
+    {
+      "epoch": 9.975146348978125,
+      "grad_norm": 0.08375976234674454,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 97128
+    },
+    {
+      "epoch": 9.975454452089966,
+      "grad_norm": 0.10059249401092529,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 97131
+    },
+    {
+      "epoch": 9.975762555201808,
+      "grad_norm": 0.09325744956731796,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 97134
+    },
+    {
+      "epoch": 9.976070658313649,
+      "grad_norm": 0.06354837119579315,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 97137
+    },
+    {
+      "epoch": 9.97637876142549,
+      "grad_norm": 0.05322406068444252,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 97140
+    },
+    {
+      "epoch": 9.976686864537331,
+      "grad_norm": 0.04118852689862251,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 97143
+    },
+    {
+      "epoch": 9.976994967649173,
+      "grad_norm": 0.0986984446644783,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 97146
+    },
+    {
+      "epoch": 9.977303070761014,
+      "grad_norm": 0.11657184362411499,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 97149
+    },
+    {
+      "epoch": 9.977611173872855,
+      "grad_norm": 0.04773658141493797,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 97152
+    },
+    {
+      "epoch": 9.977919276984698,
+      "grad_norm": 0.08913548290729523,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 97155
+    },
+    {
+      "epoch": 9.97822738009654,
+      "grad_norm": 0.1064072772860527,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 97158
+    },
+    {
+      "epoch": 9.97853548320838,
+      "grad_norm": 0.06742316484451294,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 97161
+    },
+    {
+      "epoch": 9.978843586320222,
+      "grad_norm": 0.057154104113578796,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 97164
+    },
+    {
+      "epoch": 9.979151689432063,
+      "grad_norm": 0.04546293243765831,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 97167
+    },
+    {
+      "epoch": 9.979459792543905,
+      "grad_norm": 0.04663031920790672,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 97170
+    },
+    {
+      "epoch": 9.979767895655746,
+      "grad_norm": 0.08622399717569351,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 97173
+    },
+    {
+      "epoch": 9.980075998767587,
+      "grad_norm": 0.039252907037734985,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 97176
+    },
+    {
+      "epoch": 9.980384101879428,
+      "grad_norm": 0.051296334713697433,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 97179
+    },
+    {
+      "epoch": 9.98069220499127,
+      "grad_norm": 0.04865694418549538,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 97182
+    },
+    {
+      "epoch": 9.981000308103113,
+      "grad_norm": 0.03953705355525017,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 97185
+    },
+    {
+      "epoch": 9.981308411214954,
+      "grad_norm": 0.05082760378718376,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 97188
+    },
+    {
+      "epoch": 9.981616514326795,
+      "grad_norm": 0.057986754924058914,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 97191
+    },
+    {
+      "epoch": 9.981924617438636,
+      "grad_norm": 0.0460967980325222,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 97194
+    },
+    {
+      "epoch": 9.982232720550478,
+      "grad_norm": 0.035832539200782776,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 97197
+    },
+    {
+      "epoch": 9.982540823662319,
+      "grad_norm": 0.055434614419937134,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 97200
+    },
+    {
+      "epoch": 9.98284892677416,
+      "grad_norm": 0.04218105971813202,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 97203
+    },
+    {
+      "epoch": 9.983157029886002,
+      "grad_norm": 0.17435774207115173,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 97206
+    },
+    {
+      "epoch": 9.983465132997843,
+      "grad_norm": 0.13309799134731293,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 97209
+    },
+    {
+      "epoch": 9.983773236109684,
+      "grad_norm": 0.06844424456357956,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 97212
+    },
+    {
+      "epoch": 9.984081339221525,
+      "grad_norm": 0.04057778790593147,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 97215
+    },
+    {
+      "epoch": 9.984389442333368,
+      "grad_norm": 0.07741393893957138,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 97218
+    },
+    {
+      "epoch": 9.98469754544521,
+      "grad_norm": 0.0387113019824028,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 97221
+    },
+    {
+      "epoch": 9.98500564855705,
+      "grad_norm": 0.056523408740758896,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 97224
+    },
+    {
+      "epoch": 9.985313751668892,
+      "grad_norm": 0.09296201169490814,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 97227
+    },
+    {
+      "epoch": 9.985621854780733,
+      "grad_norm": 0.05774252116680145,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 97230
+    },
+    {
+      "epoch": 9.985929957892575,
+      "grad_norm": 0.09069366008043289,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 97233
+    },
+    {
+      "epoch": 9.986238061004416,
+      "grad_norm": 0.08415558934211731,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 97236
+    },
+    {
+      "epoch": 9.986546164116257,
+      "grad_norm": 0.0969925969839096,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 97239
+    },
+    {
+      "epoch": 9.986854267228098,
+      "grad_norm": 0.1274038702249527,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 97242
+    },
+    {
+      "epoch": 9.98716237033994,
+      "grad_norm": 0.05992702394723892,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 97245
+    },
+    {
+      "epoch": 9.987470473451783,
+      "grad_norm": 0.045540135353803635,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 97248
+    },
+    {
+      "epoch": 9.987778576563624,
+      "grad_norm": 0.03293858468532562,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 97251
+    },
+    {
+      "epoch": 9.988086679675465,
+      "grad_norm": 0.03892706707119942,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 97254
+    },
+    {
+      "epoch": 9.988394782787307,
+      "grad_norm": 0.08603250980377197,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 97257
+    },
+    {
+      "epoch": 9.988702885899148,
+      "grad_norm": 0.053922899067401886,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 97260
+    },
+    {
+      "epoch": 9.989010989010989,
+      "grad_norm": 0.04998045787215233,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 97263
+    },
+    {
+      "epoch": 9.98931909212283,
+      "grad_norm": 0.05024062469601631,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 97266
+    },
+    {
+      "epoch": 9.989627195234672,
+      "grad_norm": 0.09126166999340057,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 97269
+    },
+    {
+      "epoch": 9.989935298346513,
+      "grad_norm": 0.03651963546872139,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 97272
+    },
+    {
+      "epoch": 9.990243401458354,
+      "grad_norm": 0.06524238735437393,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 97275
+    },
+    {
+      "epoch": 9.990551504570195,
+      "grad_norm": 0.04085659235715866,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 97278
+    },
+    {
+      "epoch": 9.990859607682038,
+      "grad_norm": 0.044826142489910126,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 97281
+    },
+    {
+      "epoch": 9.99116771079388,
+      "grad_norm": 0.07010653614997864,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 97284
+    },
+    {
+      "epoch": 9.991475813905721,
+      "grad_norm": 0.058311570435762405,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 97287
+    },
+    {
+      "epoch": 9.991783917017562,
+      "grad_norm": 0.09507184475660324,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 97290
+    },
+    {
+      "epoch": 9.992092020129403,
+      "grad_norm": 0.035488102585077286,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 97293
+    },
+    {
+      "epoch": 9.992400123241245,
+      "grad_norm": 0.08211905509233475,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 97296
+    },
+    {
+      "epoch": 9.992708226353086,
+      "grad_norm": 0.1444222629070282,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 97299
+    },
+    {
+      "epoch": 9.993016329464927,
+      "grad_norm": 0.0890636295080185,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 97302
+    },
+    {
+      "epoch": 9.993324432576769,
+      "grad_norm": 0.07359254360198975,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 97305
+    },
+    {
+      "epoch": 9.99363253568861,
+      "grad_norm": 0.050952211022377014,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 97308
+    },
+    {
+      "epoch": 9.993940638800453,
+      "grad_norm": 0.08539391309022903,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 97311
+    },
+    {
+      "epoch": 9.994248741912294,
+      "grad_norm": 0.06802923232316971,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 97314
+    },
+    {
+      "epoch": 9.994556845024135,
+      "grad_norm": 0.25237563252449036,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 97317
+    },
+    {
+      "epoch": 9.994864948135977,
+      "grad_norm": 0.03497587889432907,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 97320
+    },
+    {
+      "epoch": 9.995173051247818,
+      "grad_norm": 0.04146720468997955,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 97323
+    },
+    {
+      "epoch": 9.99548115435966,
+      "grad_norm": 0.06152525544166565,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 97326
+    },
+    {
+      "epoch": 9.9957892574715,
+      "grad_norm": 0.0637136921286583,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 97329
+    },
+    {
+      "epoch": 9.996097360583342,
+      "grad_norm": 0.06904961168766022,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 97332
+    },
+    {
+      "epoch": 9.996405463695183,
+      "grad_norm": 0.11976516991853714,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 97335
+    },
+    {
+      "epoch": 9.996713566807024,
+      "grad_norm": 0.102267324924469,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 97338
+    },
+    {
+      "epoch": 9.997021669918865,
+      "grad_norm": 0.07641902565956116,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 97341
+    },
+    {
+      "epoch": 9.997329773030707,
+      "grad_norm": 0.0694715678691864,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 97344
+    },
+    {
+      "epoch": 9.99763787614255,
+      "grad_norm": 0.10273284465074539,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 97347
+    },
+    {
+      "epoch": 9.997945979254391,
+      "grad_norm": 0.06914559006690979,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 97350
+    },
+    {
+      "epoch": 9.998254082366232,
+      "grad_norm": 0.05837008357048035,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 97353
+    },
+    {
+      "epoch": 9.998562185478074,
+      "grad_norm": 0.0501350574195385,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 97356
+    },
+    {
+      "epoch": 9.998870288589915,
+      "grad_norm": 0.05369944125413895,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 97359
+    },
+    {
+      "epoch": 9.999178391701756,
+      "grad_norm": 0.04862399771809578,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 97362
+    },
+    {
+      "epoch": 9.999486494813597,
+      "grad_norm": 0.04371797293424606,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 97365
+    },
+    {
+      "epoch": 9.999794597925439,
+      "grad_norm": 0.03798222169280052,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 97368
+    },
+    {
+      "epoch": 10.00010270103728,
+      "grad_norm": 0.12691354751586914,
+      "learning_rate": 0.01,
+      "loss": 1.9833,
+      "step": 97371
+    },
+    {
+      "epoch": 10.000410804149121,
+      "grad_norm": 0.03646055608987808,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 97374
+    },
+    {
+      "epoch": 10.000718907260964,
+      "grad_norm": 0.04508265480399132,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 97377
+    },
+    {
+      "epoch": 10.001027010372805,
+      "grad_norm": 0.05073779076337814,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 97380
+    },
+    {
+      "epoch": 10.001335113484647,
+      "grad_norm": 0.04759017750620842,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 97383
+    },
+    {
+      "epoch": 10.001643216596488,
+      "grad_norm": 0.05554679408669472,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 97386
+    },
+    {
+      "epoch": 10.00195131970833,
+      "grad_norm": 0.09978814423084259,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 97389
+    },
+    {
+      "epoch": 10.00225942282017,
+      "grad_norm": 0.10586490482091904,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 97392
+    },
+    {
+      "epoch": 10.002567525932012,
+      "grad_norm": 0.06032977253198624,
+      "learning_rate": 0.01,
+      "loss": 1.9007,
+      "step": 97395
+    },
+    {
+      "epoch": 10.002875629043853,
+      "grad_norm": 0.16167794167995453,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 97398
+    },
+    {
+      "epoch": 10.003183732155694,
+      "grad_norm": 0.05945628881454468,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 97401
+    },
+    {
+      "epoch": 10.003491835267535,
+      "grad_norm": 0.06830620020627975,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 97404
+    },
+    {
+      "epoch": 10.003799938379377,
+      "grad_norm": 0.05241851136088371,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 97407
+    },
+    {
+      "epoch": 10.00410804149122,
+      "grad_norm": 0.04357186704874039,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 97410
+    },
+    {
+      "epoch": 10.004416144603061,
+      "grad_norm": 0.08057430386543274,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 97413
+    },
+    {
+      "epoch": 10.004724247714902,
+      "grad_norm": 0.08658464252948761,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 97416
+    },
+    {
+      "epoch": 10.005032350826744,
+      "grad_norm": 0.06090567260980606,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 97419
+    },
+    {
+      "epoch": 10.005340453938585,
+      "grad_norm": 0.09897079318761826,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 97422
+    },
+    {
+      "epoch": 10.005648557050426,
+      "grad_norm": 0.1031961739063263,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 97425
+    },
+    {
+      "epoch": 10.005956660162267,
+      "grad_norm": 0.0531962588429451,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 97428
+    },
+    {
+      "epoch": 10.006264763274109,
+      "grad_norm": 0.10613752156496048,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 97431
+    },
+    {
+      "epoch": 10.00657286638595,
+      "grad_norm": 0.046731069684028625,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 97434
+    },
+    {
+      "epoch": 10.006880969497791,
+      "grad_norm": 0.11759690195322037,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 97437
+    },
+    {
+      "epoch": 10.007189072609634,
+      "grad_norm": 0.04036640003323555,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 97440
+    },
+    {
+      "epoch": 10.007497175721475,
+      "grad_norm": 0.08520452678203583,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 97443
+    },
+    {
+      "epoch": 10.007805278833317,
+      "grad_norm": 0.03992795944213867,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 97446
+    },
+    {
+      "epoch": 10.008113381945158,
+      "grad_norm": 0.0406220443546772,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 97449
+    },
+    {
+      "epoch": 10.008421485057,
+      "grad_norm": 0.04602857306599617,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 97452
+    },
+    {
+      "epoch": 10.00872958816884,
+      "grad_norm": 0.1277029812335968,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 97455
+    },
+    {
+      "epoch": 10.009037691280682,
+      "grad_norm": 0.12406671047210693,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 97458
+    },
+    {
+      "epoch": 10.009345794392523,
+      "grad_norm": 0.050995584577322006,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 97461
+    },
+    {
+      "epoch": 10.009653897504364,
+      "grad_norm": 0.05945532023906708,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 97464
+    },
+    {
+      "epoch": 10.009962000616206,
+      "grad_norm": 0.047782059758901596,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 97467
+    },
+    {
+      "epoch": 10.010270103728047,
+      "grad_norm": 0.0939398780465126,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 97470
+    },
+    {
+      "epoch": 10.01057820683989,
+      "grad_norm": 0.07378079742193222,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 97473
+    },
+    {
+      "epoch": 10.010886309951731,
+      "grad_norm": 0.09242421388626099,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 97476
+    },
+    {
+      "epoch": 10.011194413063572,
+      "grad_norm": 0.04233480617403984,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 97479
+    },
+    {
+      "epoch": 10.011502516175414,
+      "grad_norm": 0.03884666785597801,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 97482
+    },
+    {
+      "epoch": 10.011810619287255,
+      "grad_norm": 0.044267915189266205,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 97485
+    },
+    {
+      "epoch": 10.012118722399096,
+      "grad_norm": 0.0717068612575531,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 97488
+    },
+    {
+      "epoch": 10.012426825510937,
+      "grad_norm": 0.04711198806762695,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 97491
+    },
+    {
+      "epoch": 10.012734928622779,
+      "grad_norm": 0.044493578374385834,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 97494
+    },
+    {
+      "epoch": 10.01304303173462,
+      "grad_norm": 0.05505421757698059,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 97497
+    },
+    {
+      "epoch": 10.013351134846461,
+      "grad_norm": 0.06965785473585129,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 97500
+    },
+    {
+      "epoch": 10.013659237958304,
+      "grad_norm": 0.04743869602680206,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 97503
+    },
+    {
+      "epoch": 10.013967341070146,
+      "grad_norm": 0.043692149221897125,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 97506
+    },
+    {
+      "epoch": 10.014275444181987,
+      "grad_norm": 0.15378451347351074,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 97509
+    },
+    {
+      "epoch": 10.014583547293828,
+      "grad_norm": 0.11334779113531113,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 97512
+    },
+    {
+      "epoch": 10.01489165040567,
+      "grad_norm": 0.05877596512436867,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 97515
+    },
+    {
+      "epoch": 10.01519975351751,
+      "grad_norm": 0.04958738759160042,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 97518
+    },
+    {
+      "epoch": 10.015507856629352,
+      "grad_norm": 0.038849733769893646,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 97521
+    },
+    {
+      "epoch": 10.015815959741193,
+      "grad_norm": 0.04901808872818947,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 97524
+    },
+    {
+      "epoch": 10.016124062853034,
+      "grad_norm": 0.047270163893699646,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 97527
+    },
+    {
+      "epoch": 10.016432165964876,
+      "grad_norm": 0.11565852165222168,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 97530
+    },
+    {
+      "epoch": 10.016740269076717,
+      "grad_norm": 0.04593334347009659,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 97533
+    },
+    {
+      "epoch": 10.01704837218856,
+      "grad_norm": 0.08739186823368073,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 97536
+    },
+    {
+      "epoch": 10.017356475300401,
+      "grad_norm": 0.10565205663442612,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 97539
+    },
+    {
+      "epoch": 10.017664578412242,
+      "grad_norm": 0.07832454144954681,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 97542
+    },
+    {
+      "epoch": 10.017972681524084,
+      "grad_norm": 0.07256269454956055,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 97545
+    },
+    {
+      "epoch": 10.018280784635925,
+      "grad_norm": 0.05213913321495056,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 97548
+    },
+    {
+      "epoch": 10.018588887747766,
+      "grad_norm": 0.05286802724003792,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 97551
+    },
+    {
+      "epoch": 10.018896990859607,
+      "grad_norm": 0.049530480057001114,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 97554
+    },
+    {
+      "epoch": 10.019205093971449,
+      "grad_norm": 0.04447256028652191,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 97557
+    },
+    {
+      "epoch": 10.01951319708329,
+      "grad_norm": 0.12635476887226105,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 97560
+    },
+    {
+      "epoch": 10.019821300195131,
+      "grad_norm": 0.10345647484064102,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 97563
+    },
+    {
+      "epoch": 10.020129403306973,
+      "grad_norm": 0.0937427282333374,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 97566
+    },
+    {
+      "epoch": 10.020437506418816,
+      "grad_norm": 0.05710608884692192,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 97569
+    },
+    {
+      "epoch": 10.020745609530657,
+      "grad_norm": 0.03611237183213234,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 97572
+    },
+    {
+      "epoch": 10.021053712642498,
+      "grad_norm": 0.03928690403699875,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 97575
+    },
+    {
+      "epoch": 10.02136181575434,
+      "grad_norm": 0.03709922730922699,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 97578
+    },
+    {
+      "epoch": 10.02166991886618,
+      "grad_norm": 0.11630689352750778,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 97581
+    },
+    {
+      "epoch": 10.021978021978022,
+      "grad_norm": 0.16802328824996948,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 97584
+    },
+    {
+      "epoch": 10.022286125089863,
+      "grad_norm": 0.07408058643341064,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 97587
+    },
+    {
+      "epoch": 10.022594228201704,
+      "grad_norm": 0.09721572697162628,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 97590
+    },
+    {
+      "epoch": 10.022902331313546,
+      "grad_norm": 0.06861985474824905,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 97593
+    },
+    {
+      "epoch": 10.023210434425387,
+      "grad_norm": 0.0371435172855854,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 97596
+    },
+    {
+      "epoch": 10.02351853753723,
+      "grad_norm": 0.046418141573667526,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 97599
+    },
+    {
+      "epoch": 10.023826640649071,
+      "grad_norm": 0.11881248652935028,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 97602
+    },
+    {
+      "epoch": 10.024134743760913,
+      "grad_norm": 0.0706331804394722,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 97605
+    },
+    {
+      "epoch": 10.024442846872754,
+      "grad_norm": 0.07528916001319885,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 97608
+    },
+    {
+      "epoch": 10.024750949984595,
+      "grad_norm": 0.09359458833932877,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 97611
+    },
+    {
+      "epoch": 10.025059053096436,
+      "grad_norm": 0.08137216418981552,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 97614
+    },
+    {
+      "epoch": 10.025367156208278,
+      "grad_norm": 0.0373852401971817,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 97617
+    },
+    {
+      "epoch": 10.025675259320119,
+      "grad_norm": 0.09041198343038559,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 97620
+    },
+    {
+      "epoch": 10.02598336243196,
+      "grad_norm": 0.0771503895521164,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 97623
+    },
+    {
+      "epoch": 10.026291465543801,
+      "grad_norm": 0.07588748633861542,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 97626
+    },
+    {
+      "epoch": 10.026599568655643,
+      "grad_norm": 0.03945888206362724,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 97629
+    },
+    {
+      "epoch": 10.026907671767486,
+      "grad_norm": 0.12288627028465271,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 97632
+    },
+    {
+      "epoch": 10.027215774879327,
+      "grad_norm": 0.13935460150241852,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 97635
+    },
+    {
+      "epoch": 10.027523877991168,
+      "grad_norm": 0.07311210036277771,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 97638
+    },
+    {
+      "epoch": 10.02783198110301,
+      "grad_norm": 0.04974691942334175,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 97641
+    },
+    {
+      "epoch": 10.02814008421485,
+      "grad_norm": 0.04890896752476692,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 97644
+    },
+    {
+      "epoch": 10.028448187326692,
+      "grad_norm": 0.06187541037797928,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 97647
+    },
+    {
+      "epoch": 10.028756290438533,
+      "grad_norm": 0.0689026489853859,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 97650
+    },
+    {
+      "epoch": 10.029064393550374,
+      "grad_norm": 0.10951229929924011,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 97653
+    },
+    {
+      "epoch": 10.029372496662216,
+      "grad_norm": 0.10918518900871277,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 97656
+    },
+    {
+      "epoch": 10.029680599774057,
+      "grad_norm": 0.062250666320323944,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 97659
+    },
+    {
+      "epoch": 10.029988702885898,
+      "grad_norm": 0.0697980672121048,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 97662
+    },
+    {
+      "epoch": 10.030296805997741,
+      "grad_norm": 0.07267870008945465,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 97665
+    },
+    {
+      "epoch": 10.030604909109583,
+      "grad_norm": 0.061427708715200424,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 97668
+    },
+    {
+      "epoch": 10.030913012221424,
+      "grad_norm": 0.05644296482205391,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 97671
+    },
+    {
+      "epoch": 10.031221115333265,
+      "grad_norm": 0.0500921756029129,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 97674
+    },
+    {
+      "epoch": 10.031529218445106,
+      "grad_norm": 0.04461059346795082,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 97677
+    },
+    {
+      "epoch": 10.031837321556948,
+      "grad_norm": 0.046942368149757385,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 97680
+    },
+    {
+      "epoch": 10.032145424668789,
+      "grad_norm": 0.06114652007818222,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 97683
+    },
+    {
+      "epoch": 10.03245352778063,
+      "grad_norm": 0.08929812908172607,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 97686
+    },
+    {
+      "epoch": 10.032761630892471,
+      "grad_norm": 0.05619662627577782,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 97689
+    },
+    {
+      "epoch": 10.033069734004313,
+      "grad_norm": 0.157526895403862,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 97692
+    },
+    {
+      "epoch": 10.033377837116156,
+      "grad_norm": 0.15800322592258453,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 97695
+    },
+    {
+      "epoch": 10.033685940227997,
+      "grad_norm": 0.10482832789421082,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 97698
+    },
+    {
+      "epoch": 10.033994043339838,
+      "grad_norm": 0.09328657388687134,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 97701
+    },
+    {
+      "epoch": 10.03430214645168,
+      "grad_norm": 0.035461828112602234,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 97704
+    },
+    {
+      "epoch": 10.03461024956352,
+      "grad_norm": 0.08092792332172394,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 97707
+    },
+    {
+      "epoch": 10.034918352675362,
+      "grad_norm": 0.043945904821157455,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 97710
+    },
+    {
+      "epoch": 10.035226455787203,
+      "grad_norm": 0.042452260851860046,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 97713
+    },
+    {
+      "epoch": 10.035534558899045,
+      "grad_norm": 0.0670391321182251,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 97716
+    },
+    {
+      "epoch": 10.035842662010886,
+      "grad_norm": 0.07876156270503998,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 97719
+    },
+    {
+      "epoch": 10.036150765122727,
+      "grad_norm": 0.0945872887969017,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 97722
+    },
+    {
+      "epoch": 10.036458868234568,
+      "grad_norm": 0.05473790690302849,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 97725
+    },
+    {
+      "epoch": 10.036766971346411,
+      "grad_norm": 0.043251413851976395,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 97728
+    },
+    {
+      "epoch": 10.037075074458253,
+      "grad_norm": 0.03373229131102562,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 97731
+    },
+    {
+      "epoch": 10.037383177570094,
+      "grad_norm": 0.04231463745236397,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 97734
+    },
+    {
+      "epoch": 10.037691280681935,
+      "grad_norm": 0.0730859637260437,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 97737
+    },
+    {
+      "epoch": 10.037999383793776,
+      "grad_norm": 0.07237992435693741,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 97740
+    },
+    {
+      "epoch": 10.038307486905618,
+      "grad_norm": 0.10237434506416321,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 97743
+    },
+    {
+      "epoch": 10.038615590017459,
+      "grad_norm": 0.04624316468834877,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 97746
+    },
+    {
+      "epoch": 10.0389236931293,
+      "grad_norm": 0.05136566981673241,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 97749
+    },
+    {
+      "epoch": 10.039231796241141,
+      "grad_norm": 0.049200087785720825,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 97752
+    },
+    {
+      "epoch": 10.039539899352983,
+      "grad_norm": 0.049420248717069626,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 97755
+    },
+    {
+      "epoch": 10.039848002464826,
+      "grad_norm": 0.04861652851104736,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 97758
+    },
+    {
+      "epoch": 10.040156105576667,
+      "grad_norm": 0.08401332050561905,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 97761
+    },
+    {
+      "epoch": 10.040464208688508,
+      "grad_norm": 0.059221383184194565,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 97764
+    },
+    {
+      "epoch": 10.04077231180035,
+      "grad_norm": 0.04965361952781677,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 97767
+    },
+    {
+      "epoch": 10.04108041491219,
+      "grad_norm": 0.06314288079738617,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 97770
+    },
+    {
+      "epoch": 10.041388518024032,
+      "grad_norm": 0.03867454081773758,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 97773
+    },
+    {
+      "epoch": 10.041696621135873,
+      "grad_norm": 0.03719132021069527,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 97776
+    },
+    {
+      "epoch": 10.042004724247715,
+      "grad_norm": 0.1044577807188034,
+      "learning_rate": 0.01,
+      "loss": 1.9766,
+      "step": 97779
+    },
+    {
+      "epoch": 10.042312827359556,
+      "grad_norm": 0.06323778629302979,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 97782
+    },
+    {
+      "epoch": 10.042620930471397,
+      "grad_norm": 0.07908964902162552,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 97785
+    },
+    {
+      "epoch": 10.042929033583238,
+      "grad_norm": 0.06004498898983002,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 97788
+    },
+    {
+      "epoch": 10.043237136695081,
+      "grad_norm": 0.11528830230236053,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 97791
+    },
+    {
+      "epoch": 10.043545239806923,
+      "grad_norm": 0.05327809602022171,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 97794
+    },
+    {
+      "epoch": 10.043853342918764,
+      "grad_norm": 0.03932127356529236,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 97797
+    },
+    {
+      "epoch": 10.044161446030605,
+      "grad_norm": 0.040303848683834076,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 97800
+    },
+    {
+      "epoch": 10.044469549142446,
+      "grad_norm": 0.04516441002488136,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 97803
+    },
+    {
+      "epoch": 10.044777652254288,
+      "grad_norm": 0.12306668609380722,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 97806
+    },
+    {
+      "epoch": 10.045085755366129,
+      "grad_norm": 0.03442014381289482,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 97809
+    },
+    {
+      "epoch": 10.04539385847797,
+      "grad_norm": 0.0522041916847229,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 97812
+    },
+    {
+      "epoch": 10.045701961589812,
+      "grad_norm": 0.04728509485721588,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 97815
+    },
+    {
+      "epoch": 10.046010064701653,
+      "grad_norm": 0.04077201709151268,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 97818
+    },
+    {
+      "epoch": 10.046318167813494,
+      "grad_norm": 0.05416601523756981,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 97821
+    },
+    {
+      "epoch": 10.046626270925337,
+      "grad_norm": 0.07108070701360703,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 97824
+    },
+    {
+      "epoch": 10.046934374037178,
+      "grad_norm": 0.047428593039512634,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 97827
+    },
+    {
+      "epoch": 10.04724247714902,
+      "grad_norm": 0.05475901812314987,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 97830
+    },
+    {
+      "epoch": 10.047550580260861,
+      "grad_norm": 0.03730958327651024,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 97833
+    },
+    {
+      "epoch": 10.047858683372702,
+      "grad_norm": 0.132392019033432,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 97836
+    },
+    {
+      "epoch": 10.048166786484543,
+      "grad_norm": 0.11565215140581131,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 97839
+    },
+    {
+      "epoch": 10.048474889596385,
+      "grad_norm": 0.11753211915493011,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 97842
+    },
+    {
+      "epoch": 10.048782992708226,
+      "grad_norm": 0.08669470250606537,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 97845
+    },
+    {
+      "epoch": 10.049091095820067,
+      "grad_norm": 0.08564909547567368,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 97848
+    },
+    {
+      "epoch": 10.049399198931908,
+      "grad_norm": 0.06066494807600975,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 97851
+    },
+    {
+      "epoch": 10.049707302043752,
+      "grad_norm": 0.03739757835865021,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 97854
+    },
+    {
+      "epoch": 10.050015405155593,
+      "grad_norm": 0.0431349016726017,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 97857
+    },
+    {
+      "epoch": 10.050323508267434,
+      "grad_norm": 0.10025177896022797,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 97860
+    },
+    {
+      "epoch": 10.050631611379275,
+      "grad_norm": 0.08526377379894257,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 97863
+    },
+    {
+      "epoch": 10.050939714491117,
+      "grad_norm": 0.07417726516723633,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 97866
+    },
+    {
+      "epoch": 10.051247817602958,
+      "grad_norm": 0.04944293573498726,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 97869
+    },
+    {
+      "epoch": 10.051555920714799,
+      "grad_norm": 0.045282039791345596,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 97872
+    },
+    {
+      "epoch": 10.05186402382664,
+      "grad_norm": 0.04437606409192085,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 97875
+    },
+    {
+      "epoch": 10.052172126938482,
+      "grad_norm": 0.08778560161590576,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 97878
+    },
+    {
+      "epoch": 10.052480230050323,
+      "grad_norm": 0.08885132521390915,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 97881
+    },
+    {
+      "epoch": 10.052788333162164,
+      "grad_norm": 0.11218232661485672,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 97884
+    },
+    {
+      "epoch": 10.053096436274007,
+      "grad_norm": 0.09743648767471313,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 97887
+    },
+    {
+      "epoch": 10.053404539385848,
+      "grad_norm": 0.07041345536708832,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 97890
+    },
+    {
+      "epoch": 10.05371264249769,
+      "grad_norm": 0.052324507385492325,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 97893
+    },
+    {
+      "epoch": 10.054020745609531,
+      "grad_norm": 0.04444773122668266,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 97896
+    },
+    {
+      "epoch": 10.054328848721372,
+      "grad_norm": 0.07826077193021774,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 97899
+    },
+    {
+      "epoch": 10.054636951833213,
+      "grad_norm": 0.03888843208551407,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 97902
+    },
+    {
+      "epoch": 10.054945054945055,
+      "grad_norm": 0.04141579940915108,
+      "learning_rate": 0.01,
+      "loss": 1.9754,
+      "step": 97905
+    },
+    {
+      "epoch": 10.055253158056896,
+      "grad_norm": 0.039864491671323776,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 97908
+    },
+    {
+      "epoch": 10.055561261168737,
+      "grad_norm": 0.13369201123714447,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 97911
+    },
+    {
+      "epoch": 10.055869364280579,
+      "grad_norm": 0.053901948034763336,
+      "learning_rate": 0.01,
+      "loss": 1.9841,
+      "step": 97914
+    },
+    {
+      "epoch": 10.05617746739242,
+      "grad_norm": 0.06346002221107483,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 97917
+    },
+    {
+      "epoch": 10.056485570504263,
+      "grad_norm": 0.04997136443853378,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 97920
+    },
+    {
+      "epoch": 10.056793673616104,
+      "grad_norm": 0.036513641476631165,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 97923
+    },
+    {
+      "epoch": 10.057101776727945,
+      "grad_norm": 0.04684044420719147,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 97926
+    },
+    {
+      "epoch": 10.057409879839787,
+      "grad_norm": 0.06806880980730057,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 97929
+    },
+    {
+      "epoch": 10.057717982951628,
+      "grad_norm": 0.0641106516122818,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 97932
+    },
+    {
+      "epoch": 10.05802608606347,
+      "grad_norm": 0.06563923507928848,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 97935
+    },
+    {
+      "epoch": 10.05833418917531,
+      "grad_norm": 0.05737874656915665,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 97938
+    },
+    {
+      "epoch": 10.058642292287152,
+      "grad_norm": 0.07749006897211075,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 97941
+    },
+    {
+      "epoch": 10.058950395398993,
+      "grad_norm": 0.11980284005403519,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 97944
+    },
+    {
+      "epoch": 10.059258498510834,
+      "grad_norm": 0.04357290640473366,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 97947
+    },
+    {
+      "epoch": 10.059566601622677,
+      "grad_norm": 0.05592518299818039,
+      "learning_rate": 0.01,
+      "loss": 1.9747,
+      "step": 97950
+    },
+    {
+      "epoch": 10.059874704734519,
+      "grad_norm": 0.08463305234909058,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 97953
+    },
+    {
+      "epoch": 10.06018280784636,
+      "grad_norm": 0.08054778724908829,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 97956
+    },
+    {
+      "epoch": 10.060490910958201,
+      "grad_norm": 0.08148549497127533,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 97959
+    },
+    {
+      "epoch": 10.060799014070042,
+      "grad_norm": 0.06928577274084091,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 97962
+    },
+    {
+      "epoch": 10.061107117181884,
+      "grad_norm": 0.11390242725610733,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 97965
+    },
+    {
+      "epoch": 10.061415220293725,
+      "grad_norm": 0.03531615436077118,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 97968
+    },
+    {
+      "epoch": 10.061723323405566,
+      "grad_norm": 0.05216292291879654,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 97971
+    },
+    {
+      "epoch": 10.062031426517407,
+      "grad_norm": 0.05882908031344414,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 97974
+    },
+    {
+      "epoch": 10.062339529629249,
+      "grad_norm": 0.13236938416957855,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 97977
+    },
+    {
+      "epoch": 10.06264763274109,
+      "grad_norm": 0.04366321116685867,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 97980
+    },
+    {
+      "epoch": 10.062955735852933,
+      "grad_norm": 0.04574698582291603,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 97983
+    },
+    {
+      "epoch": 10.063263838964774,
+      "grad_norm": 0.10057555884122849,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 97986
+    },
+    {
+      "epoch": 10.063571942076615,
+      "grad_norm": 0.06712167710065842,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 97989
+    },
+    {
+      "epoch": 10.063880045188457,
+      "grad_norm": 0.11221185326576233,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 97992
+    },
+    {
+      "epoch": 10.064188148300298,
+      "grad_norm": 0.03754079341888428,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 97995
+    },
+    {
+      "epoch": 10.06449625141214,
+      "grad_norm": 0.06335268914699554,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 97998
+    },
+    {
+      "epoch": 10.06480435452398,
+      "grad_norm": 0.037058502435684204,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 98001
+    },
+    {
+      "epoch": 10.065112457635822,
+      "grad_norm": 0.043899256736040115,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 98004
+    },
+    {
+      "epoch": 10.065420560747663,
+      "grad_norm": 0.04954110458493233,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 98007
+    },
+    {
+      "epoch": 10.065728663859504,
+      "grad_norm": 0.046918366104364395,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 98010
+    },
+    {
+      "epoch": 10.066036766971347,
+      "grad_norm": 0.053783174604177475,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 98013
+    },
+    {
+      "epoch": 10.066344870083189,
+      "grad_norm": 0.03612091764807701,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 98016
+    },
+    {
+      "epoch": 10.06665297319503,
+      "grad_norm": 0.05525078997015953,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 98019
+    },
+    {
+      "epoch": 10.066961076306871,
+      "grad_norm": 0.041319839656353,
+      "learning_rate": 0.01,
+      "loss": 1.9682,
+      "step": 98022
+    },
+    {
+      "epoch": 10.067269179418712,
+      "grad_norm": 0.13023270666599274,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 98025
+    },
+    {
+      "epoch": 10.067577282530554,
+      "grad_norm": 0.14246028661727905,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 98028
+    },
+    {
+      "epoch": 10.067885385642395,
+      "grad_norm": 0.08004479110240936,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 98031
+    },
+    {
+      "epoch": 10.068193488754236,
+      "grad_norm": 0.06986592710018158,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 98034
+    },
+    {
+      "epoch": 10.068501591866077,
+      "grad_norm": 0.08863560855388641,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 98037
+    },
+    {
+      "epoch": 10.068809694977919,
+      "grad_norm": 0.0571909137070179,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 98040
+    },
+    {
+      "epoch": 10.06911779808976,
+      "grad_norm": 0.043870389461517334,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 98043
+    },
+    {
+      "epoch": 10.069425901201603,
+      "grad_norm": 0.04596741124987602,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 98046
+    },
+    {
+      "epoch": 10.069734004313444,
+      "grad_norm": 0.07051980495452881,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 98049
+    },
+    {
+      "epoch": 10.070042107425285,
+      "grad_norm": 0.11420032382011414,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 98052
+    },
+    {
+      "epoch": 10.070350210537127,
+      "grad_norm": 0.08207684010267258,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 98055
+    },
+    {
+      "epoch": 10.070658313648968,
+      "grad_norm": 0.10102898627519608,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 98058
+    },
+    {
+      "epoch": 10.07096641676081,
+      "grad_norm": 0.09918655455112457,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 98061
+    },
+    {
+      "epoch": 10.07127451987265,
+      "grad_norm": 0.061670657247304916,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 98064
+    },
+    {
+      "epoch": 10.071582622984492,
+      "grad_norm": 0.047955017536878586,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 98067
+    },
+    {
+      "epoch": 10.071890726096333,
+      "grad_norm": 0.040025174617767334,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 98070
+    },
+    {
+      "epoch": 10.072198829208174,
+      "grad_norm": 0.09726526588201523,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 98073
+    },
+    {
+      "epoch": 10.072506932320016,
+      "grad_norm": 0.043067190796136856,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 98076
+    },
+    {
+      "epoch": 10.072815035431859,
+      "grad_norm": 0.10681457817554474,
+      "learning_rate": 0.01,
+      "loss": 1.9803,
+      "step": 98079
+    },
+    {
+      "epoch": 10.0731231385437,
+      "grad_norm": 0.08581650257110596,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 98082
+    },
+    {
+      "epoch": 10.073431241655541,
+      "grad_norm": 0.08076117932796478,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 98085
+    },
+    {
+      "epoch": 10.073739344767382,
+      "grad_norm": 0.04616018384695053,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 98088
+    },
+    {
+      "epoch": 10.074047447879224,
+      "grad_norm": 0.03525041416287422,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 98091
+    },
+    {
+      "epoch": 10.074355550991065,
+      "grad_norm": 0.05245237797498703,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 98094
+    },
+    {
+      "epoch": 10.074663654102906,
+      "grad_norm": 0.09995900094509125,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 98097
+    },
+    {
+      "epoch": 10.074971757214747,
+      "grad_norm": 0.06834854930639267,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 98100
+    },
+    {
+      "epoch": 10.075279860326589,
+      "grad_norm": 0.10982584208250046,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 98103
+    },
+    {
+      "epoch": 10.07558796343843,
+      "grad_norm": 0.07823304831981659,
+      "learning_rate": 0.01,
+      "loss": 1.9024,
+      "step": 98106
+    },
+    {
+      "epoch": 10.075896066550273,
+      "grad_norm": 0.050893526524305344,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 98109
+    },
+    {
+      "epoch": 10.076204169662114,
+      "grad_norm": 0.05059904605150223,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 98112
+    },
+    {
+      "epoch": 10.076512272773956,
+      "grad_norm": 0.039330050349235535,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 98115
+    },
+    {
+      "epoch": 10.076820375885797,
+      "grad_norm": 0.03900914266705513,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 98118
+    },
+    {
+      "epoch": 10.077128478997638,
+      "grad_norm": 0.0831712931394577,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 98121
+    },
+    {
+      "epoch": 10.07743658210948,
+      "grad_norm": 0.06024106591939926,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 98124
+    },
+    {
+      "epoch": 10.07774468522132,
+      "grad_norm": 0.04956630989909172,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 98127
+    },
+    {
+      "epoch": 10.078052788333162,
+      "grad_norm": 0.04718092456459999,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 98130
+    },
+    {
+      "epoch": 10.078360891445003,
+      "grad_norm": 0.04898585006594658,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 98133
+    },
+    {
+      "epoch": 10.078668994556844,
+      "grad_norm": 0.06265314668416977,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 98136
+    },
+    {
+      "epoch": 10.078977097668686,
+      "grad_norm": 0.033283136785030365,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 98139
+    },
+    {
+      "epoch": 10.079285200780529,
+      "grad_norm": 0.08072887361049652,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 98142
+    },
+    {
+      "epoch": 10.07959330389237,
+      "grad_norm": 0.07667970657348633,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 98145
+    },
+    {
+      "epoch": 10.079901407004211,
+      "grad_norm": 0.0799844041466713,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 98148
+    },
+    {
+      "epoch": 10.080209510116052,
+      "grad_norm": 0.05267045274376869,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 98151
+    },
+    {
+      "epoch": 10.080517613227894,
+      "grad_norm": 0.09652364999055862,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 98154
+    },
+    {
+      "epoch": 10.080825716339735,
+      "grad_norm": 0.1304524838924408,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 98157
+    },
+    {
+      "epoch": 10.081133819451576,
+      "grad_norm": 0.10048730671405792,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 98160
+    },
+    {
+      "epoch": 10.081441922563418,
+      "grad_norm": 0.07810193300247192,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 98163
+    },
+    {
+      "epoch": 10.081750025675259,
+      "grad_norm": 0.06694421172142029,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 98166
+    },
+    {
+      "epoch": 10.0820581287871,
+      "grad_norm": 0.056690763682127,
+      "learning_rate": 0.01,
+      "loss": 1.9797,
+      "step": 98169
+    },
+    {
+      "epoch": 10.082366231898941,
+      "grad_norm": 0.04931468889117241,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 98172
+    },
+    {
+      "epoch": 10.082674335010784,
+      "grad_norm": 0.057076919823884964,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 98175
+    },
+    {
+      "epoch": 10.082982438122626,
+      "grad_norm": 0.0563746877014637,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 98178
+    },
+    {
+      "epoch": 10.083290541234467,
+      "grad_norm": 0.05300070717930794,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 98181
+    },
+    {
+      "epoch": 10.083598644346308,
+      "grad_norm": 0.0513119101524353,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 98184
+    },
+    {
+      "epoch": 10.08390674745815,
+      "grad_norm": 0.05145730450749397,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 98187
+    },
+    {
+      "epoch": 10.08421485056999,
+      "grad_norm": 0.05403425544500351,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 98190
+    },
+    {
+      "epoch": 10.084522953681832,
+      "grad_norm": 0.06168578565120697,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 98193
+    },
+    {
+      "epoch": 10.084831056793673,
+      "grad_norm": 0.09515652805566788,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 98196
+    },
+    {
+      "epoch": 10.085139159905514,
+      "grad_norm": 0.056796614080667496,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 98199
+    },
+    {
+      "epoch": 10.085447263017356,
+      "grad_norm": 0.04301804304122925,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 98202
+    },
+    {
+      "epoch": 10.085755366129199,
+      "grad_norm": 0.0407172255218029,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 98205
+    },
+    {
+      "epoch": 10.08606346924104,
+      "grad_norm": 0.03882700577378273,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 98208
+    },
+    {
+      "epoch": 10.086371572352881,
+      "grad_norm": 0.055570971220731735,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 98211
+    },
+    {
+      "epoch": 10.086679675464723,
+      "grad_norm": 0.09834793210029602,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 98214
+    },
+    {
+      "epoch": 10.086987778576564,
+      "grad_norm": 0.09147577732801437,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 98217
+    },
+    {
+      "epoch": 10.087295881688405,
+      "grad_norm": 0.08858241885900497,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 98220
+    },
+    {
+      "epoch": 10.087603984800246,
+      "grad_norm": 0.07103687524795532,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 98223
+    },
+    {
+      "epoch": 10.087912087912088,
+      "grad_norm": 0.05130713805556297,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 98226
+    },
+    {
+      "epoch": 10.088220191023929,
+      "grad_norm": 0.0382094606757164,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 98229
+    },
+    {
+      "epoch": 10.08852829413577,
+      "grad_norm": 0.048191484063863754,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 98232
+    },
+    {
+      "epoch": 10.088836397247611,
+      "grad_norm": 0.0679757371544838,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 98235
+    },
+    {
+      "epoch": 10.089144500359454,
+      "grad_norm": 0.0679628774523735,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 98238
+    },
+    {
+      "epoch": 10.089452603471296,
+      "grad_norm": 0.11466601490974426,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 98241
+    },
+    {
+      "epoch": 10.089760706583137,
+      "grad_norm": 0.05466683208942413,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 98244
+    },
+    {
+      "epoch": 10.090068809694978,
+      "grad_norm": 0.10592523962259293,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 98247
+    },
+    {
+      "epoch": 10.09037691280682,
+      "grad_norm": 0.0421140193939209,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 98250
+    },
+    {
+      "epoch": 10.09068501591866,
+      "grad_norm": 0.0407244972884655,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 98253
+    },
+    {
+      "epoch": 10.090993119030502,
+      "grad_norm": 0.11097019910812378,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 98256
+    },
+    {
+      "epoch": 10.091301222142343,
+      "grad_norm": 0.043895840644836426,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 98259
+    },
+    {
+      "epoch": 10.091609325254185,
+      "grad_norm": 0.04927389323711395,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 98262
+    },
+    {
+      "epoch": 10.091917428366026,
+      "grad_norm": 0.04536551982164383,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 98265
+    },
+    {
+      "epoch": 10.092225531477867,
+      "grad_norm": 0.02990969829261303,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 98268
+    },
+    {
+      "epoch": 10.09253363458971,
+      "grad_norm": 0.04231426119804382,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 98271
+    },
+    {
+      "epoch": 10.092841737701551,
+      "grad_norm": 0.10033588856458664,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 98274
+    },
+    {
+      "epoch": 10.093149840813393,
+      "grad_norm": 0.06602507084608078,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 98277
+    },
+    {
+      "epoch": 10.093457943925234,
+      "grad_norm": 0.10166456550359726,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 98280
+    },
+    {
+      "epoch": 10.093766047037075,
+      "grad_norm": 0.055442363023757935,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 98283
+    },
+    {
+      "epoch": 10.094074150148916,
+      "grad_norm": 0.09489361941814423,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 98286
+    },
+    {
+      "epoch": 10.094382253260758,
+      "grad_norm": 0.10719319432973862,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 98289
+    },
+    {
+      "epoch": 10.094690356372599,
+      "grad_norm": 0.04897269234061241,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 98292
+    },
+    {
+      "epoch": 10.09499845948444,
+      "grad_norm": 0.17686758935451508,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 98295
+    },
+    {
+      "epoch": 10.095306562596281,
+      "grad_norm": 0.1403108537197113,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 98298
+    },
+    {
+      "epoch": 10.095614665708124,
+      "grad_norm": 0.09026072919368744,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 98301
+    },
+    {
+      "epoch": 10.095922768819966,
+      "grad_norm": 0.029938554391264915,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 98304
+    },
+    {
+      "epoch": 10.096230871931807,
+      "grad_norm": 0.03506121039390564,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 98307
+    },
+    {
+      "epoch": 10.096538975043648,
+      "grad_norm": 0.058797042816877365,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 98310
+    },
+    {
+      "epoch": 10.09684707815549,
+      "grad_norm": 0.04589388146996498,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 98313
+    },
+    {
+      "epoch": 10.09715518126733,
+      "grad_norm": 0.04293623939156532,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 98316
+    },
+    {
+      "epoch": 10.097463284379172,
+      "grad_norm": 0.04917249456048012,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 98319
+    },
+    {
+      "epoch": 10.097771387491013,
+      "grad_norm": 0.0648779645562172,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 98322
+    },
+    {
+      "epoch": 10.098079490602855,
+      "grad_norm": 0.08563850075006485,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 98325
+    },
+    {
+      "epoch": 10.098387593714696,
+      "grad_norm": 0.06887205690145493,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 98328
+    },
+    {
+      "epoch": 10.098695696826537,
+      "grad_norm": 0.047499123960733414,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 98331
+    },
+    {
+      "epoch": 10.09900379993838,
+      "grad_norm": 0.13514503836631775,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 98334
+    },
+    {
+      "epoch": 10.099311903050221,
+      "grad_norm": 0.04670017212629318,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 98337
+    },
+    {
+      "epoch": 10.099620006162063,
+      "grad_norm": 0.10404140502214432,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 98340
+    },
+    {
+      "epoch": 10.099928109273904,
+      "grad_norm": 0.05887841805815697,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 98343
+    },
+    {
+      "epoch": 10.100236212385745,
+      "grad_norm": 0.0849578008055687,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 98346
+    },
+    {
+      "epoch": 10.100544315497586,
+      "grad_norm": 0.11289205402135849,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 98349
+    },
+    {
+      "epoch": 10.100852418609428,
+      "grad_norm": 0.03669322282075882,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 98352
+    },
+    {
+      "epoch": 10.101160521721269,
+      "grad_norm": 0.04403113201260567,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 98355
+    },
+    {
+      "epoch": 10.10146862483311,
+      "grad_norm": 0.07399263978004456,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 98358
+    },
+    {
+      "epoch": 10.101776727944952,
+      "grad_norm": 0.03815837949514389,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 98361
+    },
+    {
+      "epoch": 10.102084831056793,
+      "grad_norm": 0.043647266924381256,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 98364
+    },
+    {
+      "epoch": 10.102392934168636,
+      "grad_norm": 0.052343569695949554,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 98367
+    },
+    {
+      "epoch": 10.102701037280477,
+      "grad_norm": 0.043194808065891266,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 98370
+    },
+    {
+      "epoch": 10.103009140392318,
+      "grad_norm": 0.13488149642944336,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 98373
+    },
+    {
+      "epoch": 10.10331724350416,
+      "grad_norm": 0.04257534071803093,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 98376
+    },
+    {
+      "epoch": 10.103625346616,
+      "grad_norm": 0.0840926319360733,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 98379
+    },
+    {
+      "epoch": 10.103933449727842,
+      "grad_norm": 0.07795858383178711,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 98382
+    },
+    {
+      "epoch": 10.104241552839683,
+      "grad_norm": 0.0341707319021225,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 98385
+    },
+    {
+      "epoch": 10.104549655951525,
+      "grad_norm": 0.05060354247689247,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 98388
+    },
+    {
+      "epoch": 10.104857759063366,
+      "grad_norm": 0.06552443653345108,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 98391
+    },
+    {
+      "epoch": 10.105165862175207,
+      "grad_norm": 0.05386805906891823,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 98394
+    },
+    {
+      "epoch": 10.10547396528705,
+      "grad_norm": 0.046048302203416824,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 98397
+    },
+    {
+      "epoch": 10.105782068398891,
+      "grad_norm": 0.039221879094839096,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 98400
+    },
+    {
+      "epoch": 10.106090171510733,
+      "grad_norm": 0.04174118861556053,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 98403
+    },
+    {
+      "epoch": 10.106398274622574,
+      "grad_norm": 0.06665545701980591,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 98406
+    },
+    {
+      "epoch": 10.106706377734415,
+      "grad_norm": 0.12490153312683105,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 98409
+    },
+    {
+      "epoch": 10.107014480846257,
+      "grad_norm": 0.09812960773706436,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 98412
+    },
+    {
+      "epoch": 10.107322583958098,
+      "grad_norm": 0.07532085478305817,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 98415
+    },
+    {
+      "epoch": 10.107630687069939,
+      "grad_norm": 0.06836399435997009,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 98418
+    },
+    {
+      "epoch": 10.10793879018178,
+      "grad_norm": 0.045399703085422516,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 98421
+    },
+    {
+      "epoch": 10.108246893293622,
+      "grad_norm": 0.04112372174859047,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 98424
+    },
+    {
+      "epoch": 10.108554996405463,
+      "grad_norm": 0.07263118773698807,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 98427
+    },
+    {
+      "epoch": 10.108863099517306,
+      "grad_norm": 0.0632065162062645,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 98430
+    },
+    {
+      "epoch": 10.109171202629147,
+      "grad_norm": 0.11991485953330994,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 98433
+    },
+    {
+      "epoch": 10.109479305740988,
+      "grad_norm": 0.06707876920700073,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 98436
+    },
+    {
+      "epoch": 10.10978740885283,
+      "grad_norm": 0.06976279616355896,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 98439
+    },
+    {
+      "epoch": 10.110095511964671,
+      "grad_norm": 0.05575570464134216,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 98442
+    },
+    {
+      "epoch": 10.110403615076512,
+      "grad_norm": 0.03482991084456444,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 98445
+    },
+    {
+      "epoch": 10.110711718188353,
+      "grad_norm": 0.09989995509386063,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 98448
+    },
+    {
+      "epoch": 10.111019821300195,
+      "grad_norm": 0.08258899301290512,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 98451
+    },
+    {
+      "epoch": 10.111327924412036,
+      "grad_norm": 0.0769890621304512,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 98454
+    },
+    {
+      "epoch": 10.111636027523877,
+      "grad_norm": 0.13591155409812927,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 98457
+    },
+    {
+      "epoch": 10.11194413063572,
+      "grad_norm": 0.051655545830726624,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 98460
+    },
+    {
+      "epoch": 10.112252233747562,
+      "grad_norm": 0.08032787591218948,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 98463
+    },
+    {
+      "epoch": 10.112560336859403,
+      "grad_norm": 0.04594959691166878,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 98466
+    },
+    {
+      "epoch": 10.112868439971244,
+      "grad_norm": 0.03724440559744835,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 98469
+    },
+    {
+      "epoch": 10.113176543083085,
+      "grad_norm": 0.04428113251924515,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 98472
+    },
+    {
+      "epoch": 10.113484646194927,
+      "grad_norm": 0.11865317076444626,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 98475
+    },
+    {
+      "epoch": 10.113792749306768,
+      "grad_norm": 0.07116881012916565,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 98478
+    },
+    {
+      "epoch": 10.11410085241861,
+      "grad_norm": 0.07484801113605499,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 98481
+    },
+    {
+      "epoch": 10.11440895553045,
+      "grad_norm": 0.1013493537902832,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 98484
+    },
+    {
+      "epoch": 10.114717058642292,
+      "grad_norm": 0.06404826790094376,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 98487
+    },
+    {
+      "epoch": 10.115025161754133,
+      "grad_norm": 0.05708220228552818,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 98490
+    },
+    {
+      "epoch": 10.115333264865976,
+      "grad_norm": 0.05591821298003197,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 98493
+    },
+    {
+      "epoch": 10.115641367977817,
+      "grad_norm": 0.06925122439861298,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 98496
+    },
+    {
+      "epoch": 10.115949471089658,
+      "grad_norm": 0.0722322165966034,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 98499
+    },
+    {
+      "epoch": 10.1162575742015,
+      "grad_norm": 0.07561221718788147,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 98502
+    },
+    {
+      "epoch": 10.116565677313341,
+      "grad_norm": 0.04920943081378937,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 98505
+    },
+    {
+      "epoch": 10.116873780425182,
+      "grad_norm": 0.03715353459119797,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 98508
+    },
+    {
+      "epoch": 10.117181883537024,
+      "grad_norm": 0.046310905367136,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 98511
+    },
+    {
+      "epoch": 10.117489986648865,
+      "grad_norm": 0.1061708852648735,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 98514
+    },
+    {
+      "epoch": 10.117798089760706,
+      "grad_norm": 0.04743969440460205,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 98517
+    },
+    {
+      "epoch": 10.118106192872547,
+      "grad_norm": 0.10688664019107819,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 98520
+    },
+    {
+      "epoch": 10.118414295984389,
+      "grad_norm": 0.06978907436132431,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 98523
+    },
+    {
+      "epoch": 10.118722399096232,
+      "grad_norm": 0.05398084223270416,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 98526
+    },
+    {
+      "epoch": 10.119030502208073,
+      "grad_norm": 0.05511185899376869,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 98529
+    },
+    {
+      "epoch": 10.119338605319914,
+      "grad_norm": 0.03953128308057785,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 98532
+    },
+    {
+      "epoch": 10.119646708431755,
+      "grad_norm": 0.04132496565580368,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 98535
+    },
+    {
+      "epoch": 10.119954811543597,
+      "grad_norm": 0.06378614157438278,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 98538
+    },
+    {
+      "epoch": 10.120262914655438,
+      "grad_norm": 0.08414656668901443,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 98541
+    },
+    {
+      "epoch": 10.12057101776728,
+      "grad_norm": 0.04202824458479881,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 98544
+    },
+    {
+      "epoch": 10.12087912087912,
+      "grad_norm": 0.03575130179524422,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 98547
+    },
+    {
+      "epoch": 10.121187223990962,
+      "grad_norm": 0.1546347439289093,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 98550
+    },
+    {
+      "epoch": 10.121495327102803,
+      "grad_norm": 0.04151654988527298,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 98553
+    },
+    {
+      "epoch": 10.121803430214646,
+      "grad_norm": 0.0792761966586113,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 98556
+    },
+    {
+      "epoch": 10.122111533326487,
+      "grad_norm": 0.06739489734172821,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 98559
+    },
+    {
+      "epoch": 10.122419636438329,
+      "grad_norm": 0.10332737863063812,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 98562
+    },
+    {
+      "epoch": 10.12272773955017,
+      "grad_norm": 0.14671210944652557,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 98565
+    },
+    {
+      "epoch": 10.123035842662011,
+      "grad_norm": 0.09036523848772049,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 98568
+    },
+    {
+      "epoch": 10.123343945773852,
+      "grad_norm": 0.11435233056545258,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 98571
+    },
+    {
+      "epoch": 10.123652048885694,
+      "grad_norm": 0.0682324543595314,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 98574
+    },
+    {
+      "epoch": 10.123960151997535,
+      "grad_norm": 0.05735497549176216,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 98577
+    },
+    {
+      "epoch": 10.124268255109376,
+      "grad_norm": 0.06683596223592758,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 98580
+    },
+    {
+      "epoch": 10.124576358221217,
+      "grad_norm": 0.059495940804481506,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 98583
+    },
+    {
+      "epoch": 10.124884461333059,
+      "grad_norm": 0.03599031642079353,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 98586
+    },
+    {
+      "epoch": 10.125192564444902,
+      "grad_norm": 0.03614237904548645,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 98589
+    },
+    {
+      "epoch": 10.125500667556743,
+      "grad_norm": 0.0915595218539238,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 98592
+    },
+    {
+      "epoch": 10.125808770668584,
+      "grad_norm": 0.08191665261983871,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 98595
+    },
+    {
+      "epoch": 10.126116873780425,
+      "grad_norm": 0.08690435439348221,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 98598
+    },
+    {
+      "epoch": 10.126424976892267,
+      "grad_norm": 0.08599332720041275,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 98601
+    },
+    {
+      "epoch": 10.126733080004108,
+      "grad_norm": 0.08404678106307983,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 98604
+    },
+    {
+      "epoch": 10.12704118311595,
+      "grad_norm": 0.07725178450345993,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 98607
+    },
+    {
+      "epoch": 10.12734928622779,
+      "grad_norm": 0.08164679259061813,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 98610
+    },
+    {
+      "epoch": 10.127657389339632,
+      "grad_norm": 0.056525520980358124,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 98613
+    },
+    {
+      "epoch": 10.127965492451473,
+      "grad_norm": 0.03929056599736214,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 98616
+    },
+    {
+      "epoch": 10.128273595563314,
+      "grad_norm": 0.0967596173286438,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 98619
+    },
+    {
+      "epoch": 10.128581698675157,
+      "grad_norm": 0.044262755662202835,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 98622
+    },
+    {
+      "epoch": 10.128889801786999,
+      "grad_norm": 0.06301670521497726,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 98625
+    },
+    {
+      "epoch": 10.12919790489884,
+      "grad_norm": 0.05127216875553131,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 98628
+    },
+    {
+      "epoch": 10.129506008010681,
+      "grad_norm": 0.09189940243959427,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 98631
+    },
+    {
+      "epoch": 10.129814111122522,
+      "grad_norm": 0.041789088398218155,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 98634
+    },
+    {
+      "epoch": 10.130122214234364,
+      "grad_norm": 0.10127456486225128,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 98637
+    },
+    {
+      "epoch": 10.130430317346205,
+      "grad_norm": 0.043131157755851746,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 98640
+    },
+    {
+      "epoch": 10.130738420458046,
+      "grad_norm": 0.12438475340604782,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 98643
+    },
+    {
+      "epoch": 10.131046523569887,
+      "grad_norm": 0.04802525416016579,
+      "learning_rate": 0.01,
+      "loss": 1.975,
+      "step": 98646
+    },
+    {
+      "epoch": 10.131354626681729,
+      "grad_norm": 0.0507151298224926,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 98649
+    },
+    {
+      "epoch": 10.131662729793572,
+      "grad_norm": 0.086209736764431,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 98652
+    },
+    {
+      "epoch": 10.131970832905413,
+      "grad_norm": 0.060542210936546326,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 98655
+    },
+    {
+      "epoch": 10.132278936017254,
+      "grad_norm": 0.04687495529651642,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 98658
+    },
+    {
+      "epoch": 10.132587039129096,
+      "grad_norm": 0.05478186532855034,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 98661
+    },
+    {
+      "epoch": 10.132895142240937,
+      "grad_norm": 0.0812118798494339,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 98664
+    },
+    {
+      "epoch": 10.133203245352778,
+      "grad_norm": 0.06534958630800247,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 98667
+    },
+    {
+      "epoch": 10.13351134846462,
+      "grad_norm": 0.11175373941659927,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 98670
+    },
+    {
+      "epoch": 10.13381945157646,
+      "grad_norm": 0.03280024975538254,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 98673
+    },
+    {
+      "epoch": 10.134127554688302,
+      "grad_norm": 0.11052536219358444,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 98676
+    },
+    {
+      "epoch": 10.134435657800143,
+      "grad_norm": 0.07362852990627289,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 98679
+    },
+    {
+      "epoch": 10.134743760911984,
+      "grad_norm": 0.061130788177251816,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 98682
+    },
+    {
+      "epoch": 10.135051864023827,
+      "grad_norm": 0.0477757602930069,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 98685
+    },
+    {
+      "epoch": 10.135359967135669,
+      "grad_norm": 0.03536587953567505,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 98688
+    },
+    {
+      "epoch": 10.13566807024751,
+      "grad_norm": 0.04630674049258232,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 98691
+    },
+    {
+      "epoch": 10.135976173359351,
+      "grad_norm": 0.051892686635255814,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 98694
+    },
+    {
+      "epoch": 10.136284276471192,
+      "grad_norm": 0.11422517150640488,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 98697
+    },
+    {
+      "epoch": 10.136592379583034,
+      "grad_norm": 0.12365932762622833,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 98700
+    },
+    {
+      "epoch": 10.136900482694875,
+      "grad_norm": 0.03768802434206009,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 98703
+    },
+    {
+      "epoch": 10.137208585806716,
+      "grad_norm": 0.07217808812856674,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 98706
+    },
+    {
+      "epoch": 10.137516688918558,
+      "grad_norm": 0.10465411841869354,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 98709
+    },
+    {
+      "epoch": 10.137824792030399,
+      "grad_norm": 0.03864384815096855,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 98712
+    },
+    {
+      "epoch": 10.138132895142242,
+      "grad_norm": 0.051542460918426514,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 98715
+    },
+    {
+      "epoch": 10.138440998254083,
+      "grad_norm": 0.04962679743766785,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 98718
+    },
+    {
+      "epoch": 10.138749101365924,
+      "grad_norm": 0.03136231377720833,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 98721
+    },
+    {
+      "epoch": 10.139057204477766,
+      "grad_norm": 0.0425989143550396,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 98724
+    },
+    {
+      "epoch": 10.139365307589607,
+      "grad_norm": 0.11813602596521378,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 98727
+    },
+    {
+      "epoch": 10.139673410701448,
+      "grad_norm": 0.04321186617016792,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 98730
+    },
+    {
+      "epoch": 10.13998151381329,
+      "grad_norm": 0.06261945515871048,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 98733
+    },
+    {
+      "epoch": 10.14028961692513,
+      "grad_norm": 0.04242541640996933,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 98736
+    },
+    {
+      "epoch": 10.140597720036972,
+      "grad_norm": 0.05050935596227646,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 98739
+    },
+    {
+      "epoch": 10.140905823148813,
+      "grad_norm": 0.11892428994178772,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 98742
+    },
+    {
+      "epoch": 10.141213926260654,
+      "grad_norm": 0.1320609301328659,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 98745
+    },
+    {
+      "epoch": 10.141522029372497,
+      "grad_norm": 0.08089153468608856,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 98748
+    },
+    {
+      "epoch": 10.141830132484339,
+      "grad_norm": 0.08700688183307648,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 98751
+    },
+    {
+      "epoch": 10.14213823559618,
+      "grad_norm": 0.14676186442375183,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 98754
+    },
+    {
+      "epoch": 10.142446338708021,
+      "grad_norm": 0.11584928631782532,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 98757
+    },
+    {
+      "epoch": 10.142754441819863,
+      "grad_norm": 0.09546506404876709,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 98760
+    },
+    {
+      "epoch": 10.143062544931704,
+      "grad_norm": 0.04285535216331482,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 98763
+    },
+    {
+      "epoch": 10.143370648043545,
+      "grad_norm": 0.045634932816028595,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 98766
+    },
+    {
+      "epoch": 10.143678751155386,
+      "grad_norm": 0.05358066409826279,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 98769
+    },
+    {
+      "epoch": 10.143986854267228,
+      "grad_norm": 0.05766141787171364,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 98772
+    },
+    {
+      "epoch": 10.144294957379069,
+      "grad_norm": 0.07127935439348221,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 98775
+    },
+    {
+      "epoch": 10.14460306049091,
+      "grad_norm": 0.05239563062787056,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 98778
+    },
+    {
+      "epoch": 10.144911163602753,
+      "grad_norm": 0.06676071882247925,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 98781
+    },
+    {
+      "epoch": 10.145219266714594,
+      "grad_norm": 0.06993058323860168,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 98784
+    },
+    {
+      "epoch": 10.145527369826436,
+      "grad_norm": 0.09807359427213669,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 98787
+    },
+    {
+      "epoch": 10.145835472938277,
+      "grad_norm": 0.03647478297352791,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 98790
+    },
+    {
+      "epoch": 10.146143576050118,
+      "grad_norm": 0.05585514008998871,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 98793
+    },
+    {
+      "epoch": 10.14645167916196,
+      "grad_norm": 0.06017756089568138,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 98796
+    },
+    {
+      "epoch": 10.1467597822738,
+      "grad_norm": 0.052070360630750656,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 98799
+    },
+    {
+      "epoch": 10.147067885385642,
+      "grad_norm": 0.0319511704146862,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 98802
+    },
+    {
+      "epoch": 10.147375988497483,
+      "grad_norm": 0.03699100762605667,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 98805
+    },
+    {
+      "epoch": 10.147684091609324,
+      "grad_norm": 0.1164286732673645,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 98808
+    },
+    {
+      "epoch": 10.147992194721168,
+      "grad_norm": 0.07725051790475845,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 98811
+    },
+    {
+      "epoch": 10.148300297833009,
+      "grad_norm": 0.04559989646077156,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 98814
+    },
+    {
+      "epoch": 10.14860840094485,
+      "grad_norm": 0.049496665596961975,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 98817
+    },
+    {
+      "epoch": 10.148916504056691,
+      "grad_norm": 0.0559081993997097,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 98820
+    },
+    {
+      "epoch": 10.149224607168533,
+      "grad_norm": 0.05707261338829994,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 98823
+    },
+    {
+      "epoch": 10.149532710280374,
+      "grad_norm": 0.045833077281713486,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 98826
+    },
+    {
+      "epoch": 10.149840813392215,
+      "grad_norm": 0.05008349567651749,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 98829
+    },
+    {
+      "epoch": 10.150148916504056,
+      "grad_norm": 0.11216545850038528,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 98832
+    },
+    {
+      "epoch": 10.150457019615898,
+      "grad_norm": 0.04488324001431465,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 98835
+    },
+    {
+      "epoch": 10.150765122727739,
+      "grad_norm": 0.04801878705620766,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 98838
+    },
+    {
+      "epoch": 10.15107322583958,
+      "grad_norm": 0.08773142844438553,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 98841
+    },
+    {
+      "epoch": 10.151381328951423,
+      "grad_norm": 0.07734539359807968,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 98844
+    },
+    {
+      "epoch": 10.151689432063264,
+      "grad_norm": 0.06627245247364044,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 98847
+    },
+    {
+      "epoch": 10.151997535175106,
+      "grad_norm": 0.09218309819698334,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 98850
+    },
+    {
+      "epoch": 10.152305638286947,
+      "grad_norm": 0.11950929462909698,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 98853
+    },
+    {
+      "epoch": 10.152613741398788,
+      "grad_norm": 0.05259060114622116,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 98856
+    },
+    {
+      "epoch": 10.15292184451063,
+      "grad_norm": 0.045241810381412506,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 98859
+    },
+    {
+      "epoch": 10.15322994762247,
+      "grad_norm": 0.03596232831478119,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 98862
+    },
+    {
+      "epoch": 10.153538050734312,
+      "grad_norm": 0.052145007997751236,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 98865
+    },
+    {
+      "epoch": 10.153846153846153,
+      "grad_norm": 0.10584969818592072,
+      "learning_rate": 0.01,
+      "loss": 1.9811,
+      "step": 98868
+    },
+    {
+      "epoch": 10.154154256957995,
+      "grad_norm": 0.0873233750462532,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 98871
+    },
+    {
+      "epoch": 10.154462360069836,
+      "grad_norm": 0.09436676651239395,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 98874
+    },
+    {
+      "epoch": 10.154770463181679,
+      "grad_norm": 0.05636145547032356,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 98877
+    },
+    {
+      "epoch": 10.15507856629352,
+      "grad_norm": 0.08092790096998215,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 98880
+    },
+    {
+      "epoch": 10.155386669405361,
+      "grad_norm": 0.05163249373435974,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 98883
+    },
+    {
+      "epoch": 10.155694772517203,
+      "grad_norm": 0.11606719344854355,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 98886
+    },
+    {
+      "epoch": 10.156002875629044,
+      "grad_norm": 0.050113923847675323,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 98889
+    },
+    {
+      "epoch": 10.156310978740885,
+      "grad_norm": 0.07689042389392853,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 98892
+    },
+    {
+      "epoch": 10.156619081852726,
+      "grad_norm": 0.07897762209177017,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 98895
+    },
+    {
+      "epoch": 10.156927184964568,
+      "grad_norm": 0.05787033215165138,
+      "learning_rate": 0.01,
+      "loss": 1.9027,
+      "step": 98898
+    },
+    {
+      "epoch": 10.157235288076409,
+      "grad_norm": 0.057969070971012115,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 98901
+    },
+    {
+      "epoch": 10.15754339118825,
+      "grad_norm": 0.10790562629699707,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 98904
+    },
+    {
+      "epoch": 10.157851494300093,
+      "grad_norm": 0.07232168316841125,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 98907
+    },
+    {
+      "epoch": 10.158159597411935,
+      "grad_norm": 0.13184112310409546,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 98910
+    },
+    {
+      "epoch": 10.158467700523776,
+      "grad_norm": 0.08225245773792267,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 98913
+    },
+    {
+      "epoch": 10.158775803635617,
+      "grad_norm": 0.04216809943318367,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 98916
+    },
+    {
+      "epoch": 10.159083906747458,
+      "grad_norm": 0.039220843464136124,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 98919
+    },
+    {
+      "epoch": 10.1593920098593,
+      "grad_norm": 0.04414791241288185,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 98922
+    },
+    {
+      "epoch": 10.15970011297114,
+      "grad_norm": 0.06560777127742767,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 98925
+    },
+    {
+      "epoch": 10.160008216082982,
+      "grad_norm": 0.050211433321237564,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 98928
+    },
+    {
+      "epoch": 10.160316319194823,
+      "grad_norm": 0.041922181844711304,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 98931
+    },
+    {
+      "epoch": 10.160624422306665,
+      "grad_norm": 0.035320624709129333,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 98934
+    },
+    {
+      "epoch": 10.160932525418506,
+      "grad_norm": 0.13377632200717926,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 98937
+    },
+    {
+      "epoch": 10.161240628530349,
+      "grad_norm": 0.13629798591136932,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 98940
+    },
+    {
+      "epoch": 10.16154873164219,
+      "grad_norm": 0.0511309877038002,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 98943
+    },
+    {
+      "epoch": 10.161856834754031,
+      "grad_norm": 0.03774381801486015,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 98946
+    },
+    {
+      "epoch": 10.162164937865873,
+      "grad_norm": 0.04062725231051445,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 98949
+    },
+    {
+      "epoch": 10.162473040977714,
+      "grad_norm": 0.04439100995659828,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 98952
+    },
+    {
+      "epoch": 10.162781144089555,
+      "grad_norm": 0.03251064941287041,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 98955
+    },
+    {
+      "epoch": 10.163089247201397,
+      "grad_norm": 0.07445567101240158,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 98958
+    },
+    {
+      "epoch": 10.163397350313238,
+      "grad_norm": 0.0814160406589508,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 98961
+    },
+    {
+      "epoch": 10.163705453425079,
+      "grad_norm": 0.048640090972185135,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 98964
+    },
+    {
+      "epoch": 10.16401355653692,
+      "grad_norm": 0.044837836176157,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 98967
+    },
+    {
+      "epoch": 10.164321659648763,
+      "grad_norm": 0.049571964889764786,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 98970
+    },
+    {
+      "epoch": 10.164629762760605,
+      "grad_norm": 0.045311830937862396,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 98973
+    },
+    {
+      "epoch": 10.164937865872446,
+      "grad_norm": 0.040575895458459854,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 98976
+    },
+    {
+      "epoch": 10.165245968984287,
+      "grad_norm": 0.07294129580259323,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 98979
+    },
+    {
+      "epoch": 10.165554072096128,
+      "grad_norm": 0.054039083421230316,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 98982
+    },
+    {
+      "epoch": 10.16586217520797,
+      "grad_norm": 0.23129263520240784,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 98985
+    },
+    {
+      "epoch": 10.166170278319811,
+      "grad_norm": 0.08735502511262894,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 98988
+    },
+    {
+      "epoch": 10.166478381431652,
+      "grad_norm": 0.04697715863585472,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 98991
+    },
+    {
+      "epoch": 10.166786484543493,
+      "grad_norm": 0.03498484939336777,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 98994
+    },
+    {
+      "epoch": 10.167094587655335,
+      "grad_norm": 0.045701831579208374,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 98997
+    },
+    {
+      "epoch": 10.167402690767176,
+      "grad_norm": 0.0513944998383522,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 99000
+    },
+    {
+      "epoch": 10.167710793879019,
+      "grad_norm": 0.0901717096567154,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 99003
+    },
+    {
+      "epoch": 10.16801889699086,
+      "grad_norm": 0.08373749256134033,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 99006
+    },
+    {
+      "epoch": 10.168327000102702,
+      "grad_norm": 0.07009940594434738,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 99009
+    },
+    {
+      "epoch": 10.168635103214543,
+      "grad_norm": 0.05948394536972046,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 99012
+    },
+    {
+      "epoch": 10.168943206326384,
+      "grad_norm": 0.0597563236951828,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 99015
+    },
+    {
+      "epoch": 10.169251309438225,
+      "grad_norm": 0.06647255271673203,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 99018
+    },
+    {
+      "epoch": 10.169559412550067,
+      "grad_norm": 0.04476379230618477,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 99021
+    },
+    {
+      "epoch": 10.169867515661908,
+      "grad_norm": 0.043609198182821274,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 99024
+    },
+    {
+      "epoch": 10.170175618773749,
+      "grad_norm": 0.05181267112493515,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 99027
+    },
+    {
+      "epoch": 10.17048372188559,
+      "grad_norm": 0.04127652943134308,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 99030
+    },
+    {
+      "epoch": 10.170791824997432,
+      "grad_norm": 0.04352453723549843,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 99033
+    },
+    {
+      "epoch": 10.171099928109275,
+      "grad_norm": 0.1716119796037674,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 99036
+    },
+    {
+      "epoch": 10.171408031221116,
+      "grad_norm": 0.08144392818212509,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 99039
+    },
+    {
+      "epoch": 10.171716134332957,
+      "grad_norm": 0.08152837306261063,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 99042
+    },
+    {
+      "epoch": 10.172024237444798,
+      "grad_norm": 0.046277958899736404,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 99045
+    },
+    {
+      "epoch": 10.17233234055664,
+      "grad_norm": 0.038435254245996475,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 99048
+    },
+    {
+      "epoch": 10.172640443668481,
+      "grad_norm": 0.038541242480278015,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 99051
+    },
+    {
+      "epoch": 10.172948546780322,
+      "grad_norm": 0.04055872932076454,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 99054
+    },
+    {
+      "epoch": 10.173256649892163,
+      "grad_norm": 0.06344175338745117,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 99057
+    },
+    {
+      "epoch": 10.173564753004005,
+      "grad_norm": 0.06188404932618141,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 99060
+    },
+    {
+      "epoch": 10.173872856115846,
+      "grad_norm": 0.12819097936153412,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 99063
+    },
+    {
+      "epoch": 10.174180959227689,
+      "grad_norm": 0.05753006786108017,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 99066
+    },
+    {
+      "epoch": 10.17448906233953,
+      "grad_norm": 0.15209268033504486,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 99069
+    },
+    {
+      "epoch": 10.174797165451372,
+      "grad_norm": 0.10588809102773666,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 99072
+    },
+    {
+      "epoch": 10.175105268563213,
+      "grad_norm": 0.07004300504922867,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 99075
+    },
+    {
+      "epoch": 10.175413371675054,
+      "grad_norm": 0.0807129442691803,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 99078
+    },
+    {
+      "epoch": 10.175721474786895,
+      "grad_norm": 0.056306447833776474,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 99081
+    },
+    {
+      "epoch": 10.176029577898737,
+      "grad_norm": 0.04143708571791649,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 99084
+    },
+    {
+      "epoch": 10.176337681010578,
+      "grad_norm": 0.03577942028641701,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 99087
+    },
+    {
+      "epoch": 10.17664578412242,
+      "grad_norm": 0.05482630059123039,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 99090
+    },
+    {
+      "epoch": 10.17695388723426,
+      "grad_norm": 0.04353375732898712,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 99093
+    },
+    {
+      "epoch": 10.177261990346102,
+      "grad_norm": 0.05555172637104988,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 99096
+    },
+    {
+      "epoch": 10.177570093457945,
+      "grad_norm": 0.08030430227518082,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 99099
+    },
+    {
+      "epoch": 10.177878196569786,
+      "grad_norm": 0.07171040773391724,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 99102
+    },
+    {
+      "epoch": 10.178186299681627,
+      "grad_norm": 0.15464453399181366,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 99105
+    },
+    {
+      "epoch": 10.178494402793469,
+      "grad_norm": 0.055558331310749054,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 99108
+    },
+    {
+      "epoch": 10.17880250590531,
+      "grad_norm": 0.10895703732967377,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 99111
+    },
+    {
+      "epoch": 10.179110609017151,
+      "grad_norm": 0.04339831322431564,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 99114
+    },
+    {
+      "epoch": 10.179418712128992,
+      "grad_norm": 0.04216046631336212,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 99117
+    },
+    {
+      "epoch": 10.179726815240834,
+      "grad_norm": 0.047386620193719864,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 99120
+    },
+    {
+      "epoch": 10.180034918352675,
+      "grad_norm": 0.04763692989945412,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 99123
+    },
+    {
+      "epoch": 10.180343021464516,
+      "grad_norm": 0.05423538386821747,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 99126
+    },
+    {
+      "epoch": 10.180651124576357,
+      "grad_norm": 0.08128771185874939,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 99129
+    },
+    {
+      "epoch": 10.1809592276882,
+      "grad_norm": 0.08237950503826141,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 99132
+    },
+    {
+      "epoch": 10.181267330800042,
+      "grad_norm": 0.07690861076116562,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 99135
+    },
+    {
+      "epoch": 10.181575433911883,
+      "grad_norm": 0.08734242618083954,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 99138
+    },
+    {
+      "epoch": 10.181883537023724,
+      "grad_norm": 0.0518592968583107,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 99141
+    },
+    {
+      "epoch": 10.182191640135565,
+      "grad_norm": 0.08018185198307037,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 99144
+    },
+    {
+      "epoch": 10.182499743247407,
+      "grad_norm": 0.06972198933362961,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 99147
+    },
+    {
+      "epoch": 10.182807846359248,
+      "grad_norm": 0.06894844025373459,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 99150
+    },
+    {
+      "epoch": 10.18311594947109,
+      "grad_norm": 0.09211789071559906,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 99153
+    },
+    {
+      "epoch": 10.18342405258293,
+      "grad_norm": 0.06418921053409576,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 99156
+    },
+    {
+      "epoch": 10.183732155694772,
+      "grad_norm": 0.12758499383926392,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 99159
+    },
+    {
+      "epoch": 10.184040258806615,
+      "grad_norm": 0.07596320658922195,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 99162
+    },
+    {
+      "epoch": 10.184348361918456,
+      "grad_norm": 0.06082075461745262,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 99165
+    },
+    {
+      "epoch": 10.184656465030297,
+      "grad_norm": 0.05972994118928909,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 99168
+    },
+    {
+      "epoch": 10.184964568142139,
+      "grad_norm": 0.053260814398527145,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 99171
+    },
+    {
+      "epoch": 10.18527267125398,
+      "grad_norm": 0.06778529286384583,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 99174
+    },
+    {
+      "epoch": 10.185580774365821,
+      "grad_norm": 0.06456658989191055,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 99177
+    },
+    {
+      "epoch": 10.185888877477662,
+      "grad_norm": 0.06979983299970627,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 99180
+    },
+    {
+      "epoch": 10.186196980589504,
+      "grad_norm": 0.049924179911613464,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 99183
+    },
+    {
+      "epoch": 10.186505083701345,
+      "grad_norm": 0.031892478466033936,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 99186
+    },
+    {
+      "epoch": 10.186813186813186,
+      "grad_norm": 0.05626950040459633,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 99189
+    },
+    {
+      "epoch": 10.187121289925027,
+      "grad_norm": 0.12525342404842377,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 99192
+    },
+    {
+      "epoch": 10.18742939303687,
+      "grad_norm": 0.10168839991092682,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 99195
+    },
+    {
+      "epoch": 10.187737496148712,
+      "grad_norm": 0.04261062666773796,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 99198
+    },
+    {
+      "epoch": 10.188045599260553,
+      "grad_norm": 0.04495152086019516,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 99201
+    },
+    {
+      "epoch": 10.188353702372394,
+      "grad_norm": 0.038814619183540344,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 99204
+    },
+    {
+      "epoch": 10.188661805484236,
+      "grad_norm": 0.036831244826316833,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 99207
+    },
+    {
+      "epoch": 10.188969908596077,
+      "grad_norm": 0.10675276815891266,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 99210
+    },
+    {
+      "epoch": 10.189278011707918,
+      "grad_norm": 0.09722632169723511,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 99213
+    },
+    {
+      "epoch": 10.18958611481976,
+      "grad_norm": 0.1113773062825203,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 99216
+    },
+    {
+      "epoch": 10.1898942179316,
+      "grad_norm": 0.04317975789308548,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 99219
+    },
+    {
+      "epoch": 10.190202321043442,
+      "grad_norm": 0.05818440020084381,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 99222
+    },
+    {
+      "epoch": 10.190510424155285,
+      "grad_norm": 0.08506343513727188,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 99225
+    },
+    {
+      "epoch": 10.190818527267126,
+      "grad_norm": 0.05146276578307152,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 99228
+    },
+    {
+      "epoch": 10.191126630378967,
+      "grad_norm": 0.05243841931223869,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 99231
+    },
+    {
+      "epoch": 10.191434733490809,
+      "grad_norm": 0.048839930444955826,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 99234
+    },
+    {
+      "epoch": 10.19174283660265,
+      "grad_norm": 0.04465770348906517,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 99237
+    },
+    {
+      "epoch": 10.192050939714491,
+      "grad_norm": 0.11541350930929184,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 99240
+    },
+    {
+      "epoch": 10.192359042826332,
+      "grad_norm": 0.06835901737213135,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 99243
+    },
+    {
+      "epoch": 10.192667145938174,
+      "grad_norm": 0.06168109178543091,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 99246
+    },
+    {
+      "epoch": 10.192975249050015,
+      "grad_norm": 0.13129253685474396,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 99249
+    },
+    {
+      "epoch": 10.193283352161856,
+      "grad_norm": 0.037970367819070816,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 99252
+    },
+    {
+      "epoch": 10.193591455273697,
+      "grad_norm": 0.04477749764919281,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 99255
+    },
+    {
+      "epoch": 10.19389955838554,
+      "grad_norm": 0.039190951734781265,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 99258
+    },
+    {
+      "epoch": 10.194207661497382,
+      "grad_norm": 0.08986736088991165,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 99261
+    },
+    {
+      "epoch": 10.194515764609223,
+      "grad_norm": 0.07922101765871048,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 99264
+    },
+    {
+      "epoch": 10.194823867721064,
+      "grad_norm": 0.13183197379112244,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 99267
+    },
+    {
+      "epoch": 10.195131970832906,
+      "grad_norm": 0.11073755472898483,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 99270
+    },
+    {
+      "epoch": 10.195440073944747,
+      "grad_norm": 0.04988548159599304,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 99273
+    },
+    {
+      "epoch": 10.195748177056588,
+      "grad_norm": 0.03584609180688858,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 99276
+    },
+    {
+      "epoch": 10.19605628016843,
+      "grad_norm": 0.03539060056209564,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 99279
+    },
+    {
+      "epoch": 10.19636438328027,
+      "grad_norm": 0.11993156373500824,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 99282
+    },
+    {
+      "epoch": 10.196672486392112,
+      "grad_norm": 0.03564785420894623,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 99285
+    },
+    {
+      "epoch": 10.196980589503953,
+      "grad_norm": 0.10162950307130814,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 99288
+    },
+    {
+      "epoch": 10.197288692615796,
+      "grad_norm": 0.13340280950069427,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 99291
+    },
+    {
+      "epoch": 10.197596795727637,
+      "grad_norm": 0.11081166565418243,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 99294
+    },
+    {
+      "epoch": 10.197904898839479,
+      "grad_norm": 0.06974168866872787,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 99297
+    },
+    {
+      "epoch": 10.19821300195132,
+      "grad_norm": 0.052043814212083817,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 99300
+    },
+    {
+      "epoch": 10.198521105063161,
+      "grad_norm": 0.060473427176475525,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 99303
+    },
+    {
+      "epoch": 10.198829208175002,
+      "grad_norm": 0.06991663575172424,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 99306
+    },
+    {
+      "epoch": 10.199137311286844,
+      "grad_norm": 0.051819734275341034,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 99309
+    },
+    {
+      "epoch": 10.199445414398685,
+      "grad_norm": 0.044669054448604584,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 99312
+    },
+    {
+      "epoch": 10.199753517510526,
+      "grad_norm": 0.044240985065698624,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 99315
+    },
+    {
+      "epoch": 10.200061620622368,
+      "grad_norm": 0.07338231801986694,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 99318
+    },
+    {
+      "epoch": 10.20036972373421,
+      "grad_norm": 0.07212042063474655,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 99321
+    },
+    {
+      "epoch": 10.200677826846052,
+      "grad_norm": 0.09006025642156601,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 99324
+    },
+    {
+      "epoch": 10.200985929957893,
+      "grad_norm": 0.12026429176330566,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 99327
+    },
+    {
+      "epoch": 10.201294033069734,
+      "grad_norm": 0.07984338700771332,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 99330
+    },
+    {
+      "epoch": 10.201602136181576,
+      "grad_norm": 0.17421205341815948,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 99333
+    },
+    {
+      "epoch": 10.201910239293417,
+      "grad_norm": 0.09257086366415024,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 99336
+    },
+    {
+      "epoch": 10.202218342405258,
+      "grad_norm": 0.04701230302453041,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 99339
+    },
+    {
+      "epoch": 10.2025264455171,
+      "grad_norm": 0.03216308727860451,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 99342
+    },
+    {
+      "epoch": 10.20283454862894,
+      "grad_norm": 0.04874720424413681,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 99345
+    },
+    {
+      "epoch": 10.203142651740782,
+      "grad_norm": 0.04846279323101044,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 99348
+    },
+    {
+      "epoch": 10.203450754852623,
+      "grad_norm": 0.10036344826221466,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 99351
+    },
+    {
+      "epoch": 10.203758857964466,
+      "grad_norm": 0.04600038379430771,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 99354
+    },
+    {
+      "epoch": 10.204066961076308,
+      "grad_norm": 0.12066859006881714,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 99357
+    },
+    {
+      "epoch": 10.204375064188149,
+      "grad_norm": 0.10237548500299454,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 99360
+    },
+    {
+      "epoch": 10.20468316729999,
+      "grad_norm": 0.04999920725822449,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 99363
+    },
+    {
+      "epoch": 10.204991270411831,
+      "grad_norm": 0.054514575749635696,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 99366
+    },
+    {
+      "epoch": 10.205299373523673,
+      "grad_norm": 0.09816866368055344,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 99369
+    },
+    {
+      "epoch": 10.205607476635514,
+      "grad_norm": 0.12557558715343475,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 99372
+    },
+    {
+      "epoch": 10.205915579747355,
+      "grad_norm": 0.061024587601423264,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 99375
+    },
+    {
+      "epoch": 10.206223682859196,
+      "grad_norm": 0.042588815093040466,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 99378
+    },
+    {
+      "epoch": 10.206531785971038,
+      "grad_norm": 0.050229258835315704,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 99381
+    },
+    {
+      "epoch": 10.206839889082879,
+      "grad_norm": 0.043651677668094635,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 99384
+    },
+    {
+      "epoch": 10.207147992194722,
+      "grad_norm": 0.03906402364373207,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 99387
+    },
+    {
+      "epoch": 10.207456095306563,
+      "grad_norm": 0.037026580423116684,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 99390
+    },
+    {
+      "epoch": 10.207764198418404,
+      "grad_norm": 0.11340945959091187,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 99393
+    },
+    {
+      "epoch": 10.208072301530246,
+      "grad_norm": 0.07180839031934738,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 99396
+    },
+    {
+      "epoch": 10.208380404642087,
+      "grad_norm": 0.05614294111728668,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 99399
+    },
+    {
+      "epoch": 10.208688507753928,
+      "grad_norm": 0.0997401773929596,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 99402
+    },
+    {
+      "epoch": 10.20899661086577,
+      "grad_norm": 0.056495551019907,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 99405
+    },
+    {
+      "epoch": 10.20930471397761,
+      "grad_norm": 0.11858171969652176,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 99408
+    },
+    {
+      "epoch": 10.209612817089452,
+      "grad_norm": 0.04259796068072319,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 99411
+    },
+    {
+      "epoch": 10.209920920201293,
+      "grad_norm": 0.09054366499185562,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 99414
+    },
+    {
+      "epoch": 10.210229023313136,
+      "grad_norm": 0.16497641801834106,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 99417
+    },
+    {
+      "epoch": 10.210537126424978,
+      "grad_norm": 0.13353492319583893,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 99420
+    },
+    {
+      "epoch": 10.210845229536819,
+      "grad_norm": 0.06359057128429413,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 99423
+    },
+    {
+      "epoch": 10.21115333264866,
+      "grad_norm": 0.06252306699752808,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 99426
+    },
+    {
+      "epoch": 10.211461435760501,
+      "grad_norm": 0.06305409222841263,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 99429
+    },
+    {
+      "epoch": 10.211769538872343,
+      "grad_norm": 0.039893072098493576,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 99432
+    },
+    {
+      "epoch": 10.212077641984184,
+      "grad_norm": 0.055040694773197174,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 99435
+    },
+    {
+      "epoch": 10.212385745096025,
+      "grad_norm": 0.12885569036006927,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 99438
+    },
+    {
+      "epoch": 10.212693848207866,
+      "grad_norm": 0.04882066696882248,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 99441
+    },
+    {
+      "epoch": 10.213001951319708,
+      "grad_norm": 0.05029818043112755,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 99444
+    },
+    {
+      "epoch": 10.213310054431549,
+      "grad_norm": 0.06048410385847092,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 99447
+    },
+    {
+      "epoch": 10.213618157543392,
+      "grad_norm": 0.039041806012392044,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 99450
+    },
+    {
+      "epoch": 10.213926260655233,
+      "grad_norm": 0.11554940789937973,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 99453
+    },
+    {
+      "epoch": 10.214234363767074,
+      "grad_norm": 0.07602337002754211,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 99456
+    },
+    {
+      "epoch": 10.214542466878916,
+      "grad_norm": 0.05026714876294136,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 99459
+    },
+    {
+      "epoch": 10.214850569990757,
+      "grad_norm": 0.05463573336601257,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 99462
+    },
+    {
+      "epoch": 10.215158673102598,
+      "grad_norm": 0.030814429745078087,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 99465
+    },
+    {
+      "epoch": 10.21546677621444,
+      "grad_norm": 0.12945370376110077,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 99468
+    },
+    {
+      "epoch": 10.21577487932628,
+      "grad_norm": 0.0723688155412674,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 99471
+    },
+    {
+      "epoch": 10.216082982438122,
+      "grad_norm": 0.10440348833799362,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 99474
+    },
+    {
+      "epoch": 10.216391085549963,
+      "grad_norm": 0.04114481061697006,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 99477
+    },
+    {
+      "epoch": 10.216699188661806,
+      "grad_norm": 0.10391386598348618,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 99480
+    },
+    {
+      "epoch": 10.217007291773648,
+      "grad_norm": 0.050123002380132675,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 99483
+    },
+    {
+      "epoch": 10.217315394885489,
+      "grad_norm": 0.0880192220211029,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 99486
+    },
+    {
+      "epoch": 10.21762349799733,
+      "grad_norm": 0.09628675878047943,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 99489
+    },
+    {
+      "epoch": 10.217931601109171,
+      "grad_norm": 0.034183621406555176,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 99492
+    },
+    {
+      "epoch": 10.218239704221013,
+      "grad_norm": 0.04564540833234787,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 99495
+    },
+    {
+      "epoch": 10.218547807332854,
+      "grad_norm": 0.04579702392220497,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 99498
+    },
+    {
+      "epoch": 10.218855910444695,
+      "grad_norm": 0.09346463531255722,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 99501
+    },
+    {
+      "epoch": 10.219164013556536,
+      "grad_norm": 0.03784818574786186,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 99504
+    },
+    {
+      "epoch": 10.219472116668378,
+      "grad_norm": 0.10040204226970673,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 99507
+    },
+    {
+      "epoch": 10.219780219780219,
+      "grad_norm": 0.0887511596083641,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 99510
+    },
+    {
+      "epoch": 10.220088322892062,
+      "grad_norm": 0.06611691415309906,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 99513
+    },
+    {
+      "epoch": 10.220396426003903,
+      "grad_norm": 0.040397051721811295,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 99516
+    },
+    {
+      "epoch": 10.220704529115745,
+      "grad_norm": 0.045187678188085556,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 99519
+    },
+    {
+      "epoch": 10.221012632227586,
+      "grad_norm": 0.07442598044872284,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 99522
+    },
+    {
+      "epoch": 10.221320735339427,
+      "grad_norm": 0.07791969925165176,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 99525
+    },
+    {
+      "epoch": 10.221628838451268,
+      "grad_norm": 0.113225556910038,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 99528
+    },
+    {
+      "epoch": 10.22193694156311,
+      "grad_norm": 0.04235997796058655,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 99531
+    },
+    {
+      "epoch": 10.22224504467495,
+      "grad_norm": 0.093788281083107,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 99534
+    },
+    {
+      "epoch": 10.222553147786792,
+      "grad_norm": 0.04600469395518303,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 99537
+    },
+    {
+      "epoch": 10.222861250898633,
+      "grad_norm": 0.03700911998748779,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 99540
+    },
+    {
+      "epoch": 10.223169354010475,
+      "grad_norm": 0.0421181283891201,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 99543
+    },
+    {
+      "epoch": 10.223477457122318,
+      "grad_norm": 0.037187300622463226,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 99546
+    },
+    {
+      "epoch": 10.223785560234159,
+      "grad_norm": 0.03952307999134064,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 99549
+    },
+    {
+      "epoch": 10.224093663346,
+      "grad_norm": 0.1031050756573677,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 99552
+    },
+    {
+      "epoch": 10.224401766457841,
+      "grad_norm": 0.06861066818237305,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 99555
+    },
+    {
+      "epoch": 10.224709869569683,
+      "grad_norm": 0.10859041661024094,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 99558
+    },
+    {
+      "epoch": 10.225017972681524,
+      "grad_norm": 0.09697724133729935,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 99561
+    },
+    {
+      "epoch": 10.225326075793365,
+      "grad_norm": 0.057506710290908813,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 99564
+    },
+    {
+      "epoch": 10.225634178905207,
+      "grad_norm": 0.04600291699171066,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 99567
+    },
+    {
+      "epoch": 10.225942282017048,
+      "grad_norm": 0.03872712329030037,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 99570
+    },
+    {
+      "epoch": 10.226250385128889,
+      "grad_norm": 0.04338938742876053,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 99573
+    },
+    {
+      "epoch": 10.226558488240732,
+      "grad_norm": 0.04375258460640907,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 99576
+    },
+    {
+      "epoch": 10.226866591352573,
+      "grad_norm": 0.03891351819038391,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 99579
+    },
+    {
+      "epoch": 10.227174694464415,
+      "grad_norm": 0.03821360319852829,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 99582
+    },
+    {
+      "epoch": 10.227482797576256,
+      "grad_norm": 0.06259328871965408,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 99585
+    },
+    {
+      "epoch": 10.227790900688097,
+      "grad_norm": 0.08775690943002701,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 99588
+    },
+    {
+      "epoch": 10.228099003799938,
+      "grad_norm": 0.10065601766109467,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 99591
+    },
+    {
+      "epoch": 10.22840710691178,
+      "grad_norm": 0.052031807601451874,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 99594
+    },
+    {
+      "epoch": 10.228715210023621,
+      "grad_norm": 0.04331258311867714,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 99597
+    },
+    {
+      "epoch": 10.229023313135462,
+      "grad_norm": 0.044273119419813156,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 99600
+    },
+    {
+      "epoch": 10.229331416247303,
+      "grad_norm": 0.0545615628361702,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 99603
+    },
+    {
+      "epoch": 10.229639519359145,
+      "grad_norm": 0.0441010445356369,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 99606
+    },
+    {
+      "epoch": 10.229947622470988,
+      "grad_norm": 0.11924725770950317,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 99609
+    },
+    {
+      "epoch": 10.230255725582829,
+      "grad_norm": 0.045972730964422226,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 99612
+    },
+    {
+      "epoch": 10.23056382869467,
+      "grad_norm": 0.039950691163539886,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 99615
+    },
+    {
+      "epoch": 10.230871931806512,
+      "grad_norm": 0.05845572054386139,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 99618
+    },
+    {
+      "epoch": 10.231180034918353,
+      "grad_norm": 0.09689660370349884,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 99621
+    },
+    {
+      "epoch": 10.231488138030194,
+      "grad_norm": 0.055139146745204926,
+      "learning_rate": 0.01,
+      "loss": 1.8958,
+      "step": 99624
+    },
+    {
+      "epoch": 10.231796241142035,
+      "grad_norm": 0.05283995345234871,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 99627
+    },
+    {
+      "epoch": 10.232104344253877,
+      "grad_norm": 0.043195389211177826,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 99630
+    },
+    {
+      "epoch": 10.232412447365718,
+      "grad_norm": 0.07608004659414291,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 99633
+    },
+    {
+      "epoch": 10.23272055047756,
+      "grad_norm": 0.053913239389657974,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 99636
+    },
+    {
+      "epoch": 10.2330286535894,
+      "grad_norm": 0.038333795964717865,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 99639
+    },
+    {
+      "epoch": 10.233336756701243,
+      "grad_norm": 0.0701310858130455,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 99642
+    },
+    {
+      "epoch": 10.233644859813085,
+      "grad_norm": 0.06686718761920929,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 99645
+    },
+    {
+      "epoch": 10.233952962924926,
+      "grad_norm": 0.05882362276315689,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 99648
+    },
+    {
+      "epoch": 10.234261066036767,
+      "grad_norm": 0.04660504683852196,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 99651
+    },
+    {
+      "epoch": 10.234569169148608,
+      "grad_norm": 0.14900517463684082,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 99654
+    },
+    {
+      "epoch": 10.23487727226045,
+      "grad_norm": 0.0535171777009964,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 99657
+    },
+    {
+      "epoch": 10.235185375372291,
+      "grad_norm": 0.07865377515554428,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 99660
+    },
+    {
+      "epoch": 10.235493478484132,
+      "grad_norm": 0.05888461321592331,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 99663
+    },
+    {
+      "epoch": 10.235801581595974,
+      "grad_norm": 0.05282779783010483,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 99666
+    },
+    {
+      "epoch": 10.236109684707815,
+      "grad_norm": 0.0752410888671875,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 99669
+    },
+    {
+      "epoch": 10.236417787819658,
+      "grad_norm": 0.06479644775390625,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 99672
+    },
+    {
+      "epoch": 10.236725890931499,
+      "grad_norm": 0.052778132259845734,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 99675
+    },
+    {
+      "epoch": 10.23703399404334,
+      "grad_norm": 0.05656125769019127,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 99678
+    },
+    {
+      "epoch": 10.237342097155182,
+      "grad_norm": 0.04665529355406761,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 99681
+    },
+    {
+      "epoch": 10.237650200267023,
+      "grad_norm": 0.035816192626953125,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 99684
+    },
+    {
+      "epoch": 10.237958303378864,
+      "grad_norm": 0.04678123816847801,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 99687
+    },
+    {
+      "epoch": 10.238266406490705,
+      "grad_norm": 0.07228527218103409,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 99690
+    },
+    {
+      "epoch": 10.238574509602547,
+      "grad_norm": 0.15547145903110504,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 99693
+    },
+    {
+      "epoch": 10.238882612714388,
+      "grad_norm": 0.10459847003221512,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 99696
+    },
+    {
+      "epoch": 10.23919071582623,
+      "grad_norm": 0.07936637848615646,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 99699
+    },
+    {
+      "epoch": 10.23949881893807,
+      "grad_norm": 0.058107465505599976,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 99702
+    },
+    {
+      "epoch": 10.239806922049913,
+      "grad_norm": 0.03447607904672623,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 99705
+    },
+    {
+      "epoch": 10.240115025161755,
+      "grad_norm": 0.029857028275728226,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 99708
+    },
+    {
+      "epoch": 10.240423128273596,
+      "grad_norm": 0.037723857909440994,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 99711
+    },
+    {
+      "epoch": 10.240731231385437,
+      "grad_norm": 0.051165997982025146,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 99714
+    },
+    {
+      "epoch": 10.241039334497279,
+      "grad_norm": 0.03829183802008629,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 99717
+    },
+    {
+      "epoch": 10.24134743760912,
+      "grad_norm": 0.03870538994669914,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 99720
+    },
+    {
+      "epoch": 10.241655540720961,
+      "grad_norm": 0.16005821526050568,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 99723
+    },
+    {
+      "epoch": 10.241963643832802,
+      "grad_norm": 0.07645492255687714,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 99726
+    },
+    {
+      "epoch": 10.242271746944644,
+      "grad_norm": 0.0685250461101532,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 99729
+    },
+    {
+      "epoch": 10.242579850056485,
+      "grad_norm": 0.058242980390787125,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 99732
+    },
+    {
+      "epoch": 10.242887953168328,
+      "grad_norm": 0.11625604331493378,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 99735
+    },
+    {
+      "epoch": 10.24319605628017,
+      "grad_norm": 0.07878465950489044,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 99738
+    },
+    {
+      "epoch": 10.24350415939201,
+      "grad_norm": 0.10668841004371643,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 99741
+    },
+    {
+      "epoch": 10.243812262503852,
+      "grad_norm": 0.050039686262607574,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 99744
+    },
+    {
+      "epoch": 10.244120365615693,
+      "grad_norm": 0.06939005851745605,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 99747
+    },
+    {
+      "epoch": 10.244428468727534,
+      "grad_norm": 0.049836236983537674,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 99750
+    },
+    {
+      "epoch": 10.244736571839375,
+      "grad_norm": 0.09250394254922867,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 99753
+    },
+    {
+      "epoch": 10.245044674951217,
+      "grad_norm": 0.051497768610715866,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 99756
+    },
+    {
+      "epoch": 10.245352778063058,
+      "grad_norm": 0.03910185024142265,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 99759
+    },
+    {
+      "epoch": 10.2456608811749,
+      "grad_norm": 0.03284444287419319,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 99762
+    },
+    {
+      "epoch": 10.24596898428674,
+      "grad_norm": 0.0388886034488678,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 99765
+    },
+    {
+      "epoch": 10.246277087398584,
+      "grad_norm": 0.1467835158109665,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 99768
+    },
+    {
+      "epoch": 10.246585190510425,
+      "grad_norm": 0.15012861788272858,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 99771
+    },
+    {
+      "epoch": 10.246893293622266,
+      "grad_norm": 0.06511637568473816,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 99774
+    },
+    {
+      "epoch": 10.247201396734107,
+      "grad_norm": 0.047038592398166656,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 99777
+    },
+    {
+      "epoch": 10.247509499845949,
+      "grad_norm": 0.04511455446481705,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 99780
+    },
+    {
+      "epoch": 10.24781760295779,
+      "grad_norm": 0.06333138048648834,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 99783
+    },
+    {
+      "epoch": 10.248125706069631,
+      "grad_norm": 0.08840259909629822,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 99786
+    },
+    {
+      "epoch": 10.248433809181472,
+      "grad_norm": 0.054976560175418854,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 99789
+    },
+    {
+      "epoch": 10.248741912293314,
+      "grad_norm": 0.03340661898255348,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 99792
+    },
+    {
+      "epoch": 10.249050015405155,
+      "grad_norm": 0.04250464215874672,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 99795
+    },
+    {
+      "epoch": 10.249358118516996,
+      "grad_norm": 0.03535330295562744,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 99798
+    },
+    {
+      "epoch": 10.24966622162884,
+      "grad_norm": 0.09514196217060089,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 99801
+    },
+    {
+      "epoch": 10.24997432474068,
+      "grad_norm": 0.12286601215600967,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 99804
+    },
+    {
+      "epoch": 10.250282427852522,
+      "grad_norm": 0.0697341039776802,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 99807
+    },
+    {
+      "epoch": 10.250590530964363,
+      "grad_norm": 0.04430979862809181,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 99810
+    },
+    {
+      "epoch": 10.250898634076204,
+      "grad_norm": 0.0723920464515686,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 99813
+    },
+    {
+      "epoch": 10.251206737188046,
+      "grad_norm": 0.0418916754424572,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 99816
+    },
+    {
+      "epoch": 10.251514840299887,
+      "grad_norm": 0.04525751248002052,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 99819
+    },
+    {
+      "epoch": 10.251822943411728,
+      "grad_norm": 0.0397968664765358,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 99822
+    },
+    {
+      "epoch": 10.25213104652357,
+      "grad_norm": 0.04624214768409729,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 99825
+    },
+    {
+      "epoch": 10.25243914963541,
+      "grad_norm": 0.053155429661273956,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 99828
+    },
+    {
+      "epoch": 10.252747252747252,
+      "grad_norm": 0.062388960272073746,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 99831
+    },
+    {
+      "epoch": 10.253055355859095,
+      "grad_norm": 0.03869679942727089,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 99834
+    },
+    {
+      "epoch": 10.253363458970936,
+      "grad_norm": 0.08281092345714569,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 99837
+    },
+    {
+      "epoch": 10.253671562082777,
+      "grad_norm": 0.10424105823040009,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 99840
+    },
+    {
+      "epoch": 10.253979665194619,
+      "grad_norm": 0.13120271265506744,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 99843
+    },
+    {
+      "epoch": 10.25428776830646,
+      "grad_norm": 0.06335999816656113,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 99846
+    },
+    {
+      "epoch": 10.254595871418301,
+      "grad_norm": 0.035308219492435455,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 99849
+    },
+    {
+      "epoch": 10.254903974530142,
+      "grad_norm": 0.10058265179395676,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 99852
+    },
+    {
+      "epoch": 10.255212077641984,
+      "grad_norm": 0.05711854621767998,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 99855
+    },
+    {
+      "epoch": 10.255520180753825,
+      "grad_norm": 0.13923723995685577,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 99858
+    },
+    {
+      "epoch": 10.255828283865666,
+      "grad_norm": 0.09265607595443726,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 99861
+    },
+    {
+      "epoch": 10.25613638697751,
+      "grad_norm": 0.05978146195411682,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 99864
+    },
+    {
+      "epoch": 10.25644449008935,
+      "grad_norm": 0.05084634944796562,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 99867
+    },
+    {
+      "epoch": 10.256752593201192,
+      "grad_norm": 0.06485851854085922,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 99870
+    },
+    {
+      "epoch": 10.257060696313033,
+      "grad_norm": 0.05474282428622246,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 99873
+    },
+    {
+      "epoch": 10.257368799424874,
+      "grad_norm": 0.08651530742645264,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 99876
+    },
+    {
+      "epoch": 10.257676902536716,
+      "grad_norm": 0.08198468387126923,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 99879
+    },
+    {
+      "epoch": 10.257985005648557,
+      "grad_norm": 0.0553009919822216,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 99882
+    },
+    {
+      "epoch": 10.258293108760398,
+      "grad_norm": 0.10407020151615143,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 99885
+    },
+    {
+      "epoch": 10.25860121187224,
+      "grad_norm": 0.06234075501561165,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 99888
+    },
+    {
+      "epoch": 10.25890931498408,
+      "grad_norm": 0.03628971427679062,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 99891
+    },
+    {
+      "epoch": 10.259217418095922,
+      "grad_norm": 0.036581892520189285,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 99894
+    },
+    {
+      "epoch": 10.259525521207765,
+      "grad_norm": 0.04579423740506172,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 99897
+    },
+    {
+      "epoch": 10.259833624319606,
+      "grad_norm": 0.03739017993211746,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 99900
+    },
+    {
+      "epoch": 10.260141727431447,
+      "grad_norm": 0.1198686808347702,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 99903
+    },
+    {
+      "epoch": 10.260449830543289,
+      "grad_norm": 0.05373760312795639,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 99906
+    },
+    {
+      "epoch": 10.26075793365513,
+      "grad_norm": 0.06018951162695885,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 99909
+    },
+    {
+      "epoch": 10.261066036766971,
+      "grad_norm": 0.05019623786211014,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 99912
+    },
+    {
+      "epoch": 10.261374139878813,
+      "grad_norm": 0.09850800037384033,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 99915
+    },
+    {
+      "epoch": 10.261682242990654,
+      "grad_norm": 0.04459189996123314,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 99918
+    },
+    {
+      "epoch": 10.261990346102495,
+      "grad_norm": 0.08180344104766846,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 99921
+    },
+    {
+      "epoch": 10.262298449214336,
+      "grad_norm": 0.05907866731286049,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 99924
+    },
+    {
+      "epoch": 10.26260655232618,
+      "grad_norm": 0.11938024312257767,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 99927
+    },
+    {
+      "epoch": 10.26291465543802,
+      "grad_norm": 0.09561539441347122,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 99930
+    },
+    {
+      "epoch": 10.263222758549862,
+      "grad_norm": 0.11872205883264542,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 99933
+    },
+    {
+      "epoch": 10.263530861661703,
+      "grad_norm": 0.10854201763868332,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 99936
+    },
+    {
+      "epoch": 10.263838964773544,
+      "grad_norm": 0.07768164575099945,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 99939
+    },
+    {
+      "epoch": 10.264147067885386,
+      "grad_norm": 0.0876104012131691,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 99942
+    },
+    {
+      "epoch": 10.264455170997227,
+      "grad_norm": 0.05668925493955612,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 99945
+    },
+    {
+      "epoch": 10.264763274109068,
+      "grad_norm": 0.07997167110443115,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 99948
+    },
+    {
+      "epoch": 10.26507137722091,
+      "grad_norm": 0.09238320589065552,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 99951
+    },
+    {
+      "epoch": 10.26537948033275,
+      "grad_norm": 0.04772636666893959,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 99954
+    },
+    {
+      "epoch": 10.265687583444592,
+      "grad_norm": 0.04315643757581711,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 99957
+    },
+    {
+      "epoch": 10.265995686556435,
+      "grad_norm": 0.0615558885037899,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 99960
+    },
+    {
+      "epoch": 10.266303789668276,
+      "grad_norm": 0.1043780967593193,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 99963
+    },
+    {
+      "epoch": 10.266611892780118,
+      "grad_norm": 0.05783291906118393,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 99966
+    },
+    {
+      "epoch": 10.266919995891959,
+      "grad_norm": 0.09446723759174347,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 99969
+    },
+    {
+      "epoch": 10.2672280990038,
+      "grad_norm": 0.08406702429056168,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 99972
+    },
+    {
+      "epoch": 10.267536202115641,
+      "grad_norm": 0.055889952927827835,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 99975
+    },
+    {
+      "epoch": 10.267844305227483,
+      "grad_norm": 0.06571850925683975,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 99978
+    },
+    {
+      "epoch": 10.268152408339324,
+      "grad_norm": 0.09014123678207397,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 99981
+    },
+    {
+      "epoch": 10.268460511451165,
+      "grad_norm": 0.05884508416056633,
+      "learning_rate": 0.01,
+      "loss": 1.8986,
+      "step": 99984
+    },
+    {
+      "epoch": 10.268768614563006,
+      "grad_norm": 0.09543084353208542,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 99987
+    },
+    {
+      "epoch": 10.26907671767485,
+      "grad_norm": 0.048076704144477844,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 99990
+    },
+    {
+      "epoch": 10.26938482078669,
+      "grad_norm": 0.11011002212762833,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 99993
+    },
+    {
+      "epoch": 10.269692923898532,
+      "grad_norm": 0.04269436001777649,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 99996
+    },
+    {
+      "epoch": 10.270001027010373,
+      "grad_norm": 0.09961564093828201,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 99999
+    },
+    {
+      "epoch": 10.270309130122214,
+      "grad_norm": 0.08587975800037384,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 100002
+    },
+    {
+      "epoch": 10.270617233234056,
+      "grad_norm": 0.07682827115058899,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 100005
+    },
+    {
+      "epoch": 10.270925336345897,
+      "grad_norm": 0.0795656368136406,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 100008
+    },
+    {
+      "epoch": 10.271233439457738,
+      "grad_norm": 0.054615724831819534,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 100011
+    },
+    {
+      "epoch": 10.27154154256958,
+      "grad_norm": 0.10448376089334488,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 100014
+    },
+    {
+      "epoch": 10.27184964568142,
+      "grad_norm": 0.04411044716835022,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 100017
+    },
+    {
+      "epoch": 10.272157748793262,
+      "grad_norm": 0.04791110008955002,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 100020
+    },
+    {
+      "epoch": 10.272465851905105,
+      "grad_norm": 0.033793505281209946,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 100023
+    },
+    {
+      "epoch": 10.272773955016946,
+      "grad_norm": 0.03846541419625282,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 100026
+    },
+    {
+      "epoch": 10.273082058128788,
+      "grad_norm": 0.06312572211027145,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 100029
+    },
+    {
+      "epoch": 10.273390161240629,
+      "grad_norm": 0.044390659779310226,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 100032
+    },
+    {
+      "epoch": 10.27369826435247,
+      "grad_norm": 0.09738004207611084,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 100035
+    },
+    {
+      "epoch": 10.274006367464311,
+      "grad_norm": 0.07020962238311768,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 100038
+    },
+    {
+      "epoch": 10.274314470576153,
+      "grad_norm": 0.09719542413949966,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 100041
+    },
+    {
+      "epoch": 10.274622573687994,
+      "grad_norm": 0.04547543823719025,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 100044
+    },
+    {
+      "epoch": 10.274930676799835,
+      "grad_norm": 0.09154791384935379,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 100047
+    },
+    {
+      "epoch": 10.275238779911676,
+      "grad_norm": 0.07742651551961899,
+      "learning_rate": 0.01,
+      "loss": 1.972,
+      "step": 100050
+    },
+    {
+      "epoch": 10.275546883023518,
+      "grad_norm": 0.05887138843536377,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 100053
+    },
+    {
+      "epoch": 10.27585498613536,
+      "grad_norm": 0.09364501386880875,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 100056
+    },
+    {
+      "epoch": 10.276163089247202,
+      "grad_norm": 0.1074671745300293,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 100059
+    },
+    {
+      "epoch": 10.276471192359043,
+      "grad_norm": 0.04482719674706459,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 100062
+    },
+    {
+      "epoch": 10.276779295470885,
+      "grad_norm": 0.0769670158624649,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 100065
+    },
+    {
+      "epoch": 10.277087398582726,
+      "grad_norm": 0.08696067333221436,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 100068
+    },
+    {
+      "epoch": 10.277395501694567,
+      "grad_norm": 0.06060957908630371,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 100071
+    },
+    {
+      "epoch": 10.277703604806408,
+      "grad_norm": 0.09612435847520828,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 100074
+    },
+    {
+      "epoch": 10.27801170791825,
+      "grad_norm": 0.06246870011091232,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 100077
+    },
+    {
+      "epoch": 10.27831981103009,
+      "grad_norm": 0.040454257279634476,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 100080
+    },
+    {
+      "epoch": 10.278627914141932,
+      "grad_norm": 0.056246764957904816,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 100083
+    },
+    {
+      "epoch": 10.278936017253773,
+      "grad_norm": 0.039684563875198364,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 100086
+    },
+    {
+      "epoch": 10.279244120365616,
+      "grad_norm": 0.03487759456038475,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 100089
+    },
+    {
+      "epoch": 10.279552223477458,
+      "grad_norm": 0.04042143002152443,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 100092
+    },
+    {
+      "epoch": 10.279860326589299,
+      "grad_norm": 0.07096949219703674,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 100095
+    },
+    {
+      "epoch": 10.28016842970114,
+      "grad_norm": 0.0847165584564209,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 100098
+    },
+    {
+      "epoch": 10.280476532812981,
+      "grad_norm": 0.04604329913854599,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 100101
+    },
+    {
+      "epoch": 10.280784635924823,
+      "grad_norm": 0.1045752689242363,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 100104
+    },
+    {
+      "epoch": 10.281092739036664,
+      "grad_norm": 0.10349004715681076,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 100107
+    },
+    {
+      "epoch": 10.281400842148505,
+      "grad_norm": 0.05549321696162224,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 100110
+    },
+    {
+      "epoch": 10.281708945260347,
+      "grad_norm": 0.042191196233034134,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 100113
+    },
+    {
+      "epoch": 10.282017048372188,
+      "grad_norm": 0.043062347918748856,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 100116
+    },
+    {
+      "epoch": 10.28232515148403,
+      "grad_norm": 0.13222946226596832,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 100119
+    },
+    {
+      "epoch": 10.282633254595872,
+      "grad_norm": 0.04551191255450249,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 100122
+    },
+    {
+      "epoch": 10.282941357707713,
+      "grad_norm": 0.041608721017837524,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 100125
+    },
+    {
+      "epoch": 10.283249460819555,
+      "grad_norm": 0.03729642927646637,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 100128
+    },
+    {
+      "epoch": 10.283557563931396,
+      "grad_norm": 0.11232533305883408,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 100131
+    },
+    {
+      "epoch": 10.283865667043237,
+      "grad_norm": 0.12017875909805298,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 100134
+    },
+    {
+      "epoch": 10.284173770155078,
+      "grad_norm": 0.08112366497516632,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 100137
+    },
+    {
+      "epoch": 10.28448187326692,
+      "grad_norm": 0.053107235580682755,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 100140
+    },
+    {
+      "epoch": 10.284789976378761,
+      "grad_norm": 0.03232099860906601,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 100143
+    },
+    {
+      "epoch": 10.285098079490602,
+      "grad_norm": 0.034685634076595306,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 100146
+    },
+    {
+      "epoch": 10.285406182602443,
+      "grad_norm": 0.06027882918715477,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 100149
+    },
+    {
+      "epoch": 10.285714285714286,
+      "grad_norm": 0.14968882501125336,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 100152
+    },
+    {
+      "epoch": 10.286022388826128,
+      "grad_norm": 0.07659412175416946,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 100155
+    },
+    {
+      "epoch": 10.286330491937969,
+      "grad_norm": 0.05984310433268547,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 100158
+    },
+    {
+      "epoch": 10.28663859504981,
+      "grad_norm": 0.07672173529863358,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 100161
+    },
+    {
+      "epoch": 10.286946698161652,
+      "grad_norm": 0.04206996411085129,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 100164
+    },
+    {
+      "epoch": 10.287254801273493,
+      "grad_norm": 0.07788781821727753,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 100167
+    },
+    {
+      "epoch": 10.287562904385334,
+      "grad_norm": 0.04337256774306297,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 100170
+    },
+    {
+      "epoch": 10.287871007497175,
+      "grad_norm": 0.03464846685528755,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 100173
+    },
+    {
+      "epoch": 10.288179110609017,
+      "grad_norm": 0.10294554382562637,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 100176
+    },
+    {
+      "epoch": 10.288487213720858,
+      "grad_norm": 0.0609651543200016,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 100179
+    },
+    {
+      "epoch": 10.2887953168327,
+      "grad_norm": 0.0706949234008789,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 100182
+    },
+    {
+      "epoch": 10.289103419944542,
+      "grad_norm": 0.08191784471273422,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 100185
+    },
+    {
+      "epoch": 10.289411523056383,
+      "grad_norm": 0.06882110238075256,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 100188
+    },
+    {
+      "epoch": 10.289719626168225,
+      "grad_norm": 0.0786321684718132,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 100191
+    },
+    {
+      "epoch": 10.290027729280066,
+      "grad_norm": 0.03936995565891266,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 100194
+    },
+    {
+      "epoch": 10.290335832391907,
+      "grad_norm": 0.046122245490550995,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 100197
+    },
+    {
+      "epoch": 10.290643935503748,
+      "grad_norm": 0.039319563657045364,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 100200
+    },
+    {
+      "epoch": 10.29095203861559,
+      "grad_norm": 0.11181558668613434,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 100203
+    },
+    {
+      "epoch": 10.291260141727431,
+      "grad_norm": 0.11319585144519806,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 100206
+    },
+    {
+      "epoch": 10.291568244839272,
+      "grad_norm": 0.1189330592751503,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 100209
+    },
+    {
+      "epoch": 10.291876347951114,
+      "grad_norm": 0.1420219987630844,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 100212
+    },
+    {
+      "epoch": 10.292184451062957,
+      "grad_norm": 0.05568074434995651,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 100215
+    },
+    {
+      "epoch": 10.292492554174798,
+      "grad_norm": 0.0501365102827549,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 100218
+    },
+    {
+      "epoch": 10.292800657286639,
+      "grad_norm": 0.038629259914159775,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 100221
+    },
+    {
+      "epoch": 10.29310876039848,
+      "grad_norm": 0.05056120455265045,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 100224
+    },
+    {
+      "epoch": 10.293416863510322,
+      "grad_norm": 0.04930442199110985,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 100227
+    },
+    {
+      "epoch": 10.293724966622163,
+      "grad_norm": 0.04605105519294739,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 100230
+    },
+    {
+      "epoch": 10.294033069734004,
+      "grad_norm": 0.11034657806158066,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 100233
+    },
+    {
+      "epoch": 10.294341172845845,
+      "grad_norm": 0.048685695976018906,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 100236
+    },
+    {
+      "epoch": 10.294649275957687,
+      "grad_norm": 0.1337871253490448,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 100239
+    },
+    {
+      "epoch": 10.294957379069528,
+      "grad_norm": 0.11180833727121353,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 100242
+    },
+    {
+      "epoch": 10.295265482181371,
+      "grad_norm": 0.05863872915506363,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 100245
+    },
+    {
+      "epoch": 10.295573585293212,
+      "grad_norm": 0.0769515186548233,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 100248
+    },
+    {
+      "epoch": 10.295881688405053,
+      "grad_norm": 0.09128164499998093,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 100251
+    },
+    {
+      "epoch": 10.296189791516895,
+      "grad_norm": 0.08195117115974426,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 100254
+    },
+    {
+      "epoch": 10.296497894628736,
+      "grad_norm": 0.05502847582101822,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 100257
+    },
+    {
+      "epoch": 10.296805997740577,
+      "grad_norm": 0.04186360910534859,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 100260
+    },
+    {
+      "epoch": 10.297114100852419,
+      "grad_norm": 0.03892785683274269,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 100263
+    },
+    {
+      "epoch": 10.29742220396426,
+      "grad_norm": 0.042466506361961365,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 100266
+    },
+    {
+      "epoch": 10.297730307076101,
+      "grad_norm": 0.09952762722969055,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 100269
+    },
+    {
+      "epoch": 10.298038410187942,
+      "grad_norm": 0.09318811446428299,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 100272
+    },
+    {
+      "epoch": 10.298346513299784,
+      "grad_norm": 0.05363912135362625,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 100275
+    },
+    {
+      "epoch": 10.298654616411627,
+      "grad_norm": 0.08193670958280563,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 100278
+    },
+    {
+      "epoch": 10.298962719523468,
+      "grad_norm": 0.03696845844388008,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 100281
+    },
+    {
+      "epoch": 10.29927082263531,
+      "grad_norm": 0.03941279277205467,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 100284
+    },
+    {
+      "epoch": 10.29957892574715,
+      "grad_norm": 0.033926572650671005,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 100287
+    },
+    {
+      "epoch": 10.299887028858992,
+      "grad_norm": 0.052354805171489716,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 100290
+    },
+    {
+      "epoch": 10.300195131970833,
+      "grad_norm": 0.04476560652256012,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 100293
+    },
+    {
+      "epoch": 10.300503235082674,
+      "grad_norm": 0.05641723424196243,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 100296
+    },
+    {
+      "epoch": 10.300811338194515,
+      "grad_norm": 0.035899363458156586,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 100299
+    },
+    {
+      "epoch": 10.301119441306357,
+      "grad_norm": 0.10292194783687592,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 100302
+    },
+    {
+      "epoch": 10.301427544418198,
+      "grad_norm": 0.04253707453608513,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 100305
+    },
+    {
+      "epoch": 10.30173564753004,
+      "grad_norm": 0.04332521930336952,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 100308
+    },
+    {
+      "epoch": 10.302043750641882,
+      "grad_norm": 0.060099441558122635,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 100311
+    },
+    {
+      "epoch": 10.302351853753724,
+      "grad_norm": 0.039917197078466415,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 100314
+    },
+    {
+      "epoch": 10.302659956865565,
+      "grad_norm": 0.05037233605980873,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 100317
+    },
+    {
+      "epoch": 10.302968059977406,
+      "grad_norm": 0.038528043776750565,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 100320
+    },
+    {
+      "epoch": 10.303276163089247,
+      "grad_norm": 0.06853263825178146,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 100323
+    },
+    {
+      "epoch": 10.303584266201089,
+      "grad_norm": 0.04892963916063309,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 100326
+    },
+    {
+      "epoch": 10.30389236931293,
+      "grad_norm": 0.06138693541288376,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 100329
+    },
+    {
+      "epoch": 10.304200472424771,
+      "grad_norm": 0.07353366911411285,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 100332
+    },
+    {
+      "epoch": 10.304508575536612,
+      "grad_norm": 0.11467985063791275,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 100335
+    },
+    {
+      "epoch": 10.304816678648454,
+      "grad_norm": 0.059402890503406525,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 100338
+    },
+    {
+      "epoch": 10.305124781760295,
+      "grad_norm": 0.0429309643805027,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 100341
+    },
+    {
+      "epoch": 10.305432884872138,
+      "grad_norm": 0.04219118505716324,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 100344
+    },
+    {
+      "epoch": 10.30574098798398,
+      "grad_norm": 0.03355926647782326,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 100347
+    },
+    {
+      "epoch": 10.30604909109582,
+      "grad_norm": 0.120177261531353,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 100350
+    },
+    {
+      "epoch": 10.306357194207662,
+      "grad_norm": 0.04763586446642876,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 100353
+    },
+    {
+      "epoch": 10.306665297319503,
+      "grad_norm": 0.07865716516971588,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 100356
+    },
+    {
+      "epoch": 10.306973400431344,
+      "grad_norm": 0.047695744782686234,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 100359
+    },
+    {
+      "epoch": 10.307281503543186,
+      "grad_norm": 0.07048449665307999,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 100362
+    },
+    {
+      "epoch": 10.307589606655027,
+      "grad_norm": 0.05483188107609749,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 100365
+    },
+    {
+      "epoch": 10.307897709766868,
+      "grad_norm": 0.057623494416475296,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 100368
+    },
+    {
+      "epoch": 10.30820581287871,
+      "grad_norm": 0.05770512670278549,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 100371
+    },
+    {
+      "epoch": 10.308513915990552,
+      "grad_norm": 0.04068088158965111,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 100374
+    },
+    {
+      "epoch": 10.308822019102394,
+      "grad_norm": 0.046589791774749756,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 100377
+    },
+    {
+      "epoch": 10.309130122214235,
+      "grad_norm": 0.04968668520450592,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 100380
+    },
+    {
+      "epoch": 10.309438225326076,
+      "grad_norm": 0.12425777316093445,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 100383
+    },
+    {
+      "epoch": 10.309746328437917,
+      "grad_norm": 0.07676790654659271,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 100386
+    },
+    {
+      "epoch": 10.310054431549759,
+      "grad_norm": 0.09203135967254639,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 100389
+    },
+    {
+      "epoch": 10.3103625346616,
+      "grad_norm": 0.057039182633161545,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 100392
+    },
+    {
+      "epoch": 10.310670637773441,
+      "grad_norm": 0.0423281192779541,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 100395
+    },
+    {
+      "epoch": 10.310978740885282,
+      "grad_norm": 0.05860494449734688,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 100398
+    },
+    {
+      "epoch": 10.311286843997124,
+      "grad_norm": 0.06034504249691963,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 100401
+    },
+    {
+      "epoch": 10.311594947108965,
+      "grad_norm": 0.07440521568059921,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 100404
+    },
+    {
+      "epoch": 10.311903050220808,
+      "grad_norm": 0.0757577195763588,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 100407
+    },
+    {
+      "epoch": 10.31221115333265,
+      "grad_norm": 0.06766487658023834,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 100410
+    },
+    {
+      "epoch": 10.31251925644449,
+      "grad_norm": 0.09500867873430252,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 100413
+    },
+    {
+      "epoch": 10.312827359556332,
+      "grad_norm": 0.12139034271240234,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 100416
+    },
+    {
+      "epoch": 10.313135462668173,
+      "grad_norm": 0.09395263344049454,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 100419
+    },
+    {
+      "epoch": 10.313443565780014,
+      "grad_norm": 0.1121523380279541,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 100422
+    },
+    {
+      "epoch": 10.313751668891856,
+      "grad_norm": 0.11462993174791336,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 100425
+    },
+    {
+      "epoch": 10.314059772003697,
+      "grad_norm": 0.03882847726345062,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 100428
+    },
+    {
+      "epoch": 10.314367875115538,
+      "grad_norm": 0.03873373568058014,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 100431
+    },
+    {
+      "epoch": 10.31467597822738,
+      "grad_norm": 0.04255416616797447,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 100434
+    },
+    {
+      "epoch": 10.314984081339222,
+      "grad_norm": 0.05258651077747345,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 100437
+    },
+    {
+      "epoch": 10.315292184451064,
+      "grad_norm": 0.04408138990402222,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 100440
+    },
+    {
+      "epoch": 10.315600287562905,
+      "grad_norm": 0.04116139933466911,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 100443
+    },
+    {
+      "epoch": 10.315908390674746,
+      "grad_norm": 0.06574319303035736,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 100446
+    },
+    {
+      "epoch": 10.316216493786587,
+      "grad_norm": 0.040884729474782944,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 100449
+    },
+    {
+      "epoch": 10.316524596898429,
+      "grad_norm": 0.052256714552640915,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 100452
+    },
+    {
+      "epoch": 10.31683270001027,
+      "grad_norm": 0.05303657799959183,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 100455
+    },
+    {
+      "epoch": 10.317140803122111,
+      "grad_norm": 0.19547231495380402,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 100458
+    },
+    {
+      "epoch": 10.317448906233952,
+      "grad_norm": 0.14729028940200806,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 100461
+    },
+    {
+      "epoch": 10.317757009345794,
+      "grad_norm": 0.04628564044833183,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 100464
+    },
+    {
+      "epoch": 10.318065112457635,
+      "grad_norm": 0.04533575847744942,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 100467
+    },
+    {
+      "epoch": 10.318373215569478,
+      "grad_norm": 0.04745553433895111,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 100470
+    },
+    {
+      "epoch": 10.31868131868132,
+      "grad_norm": 0.04309360310435295,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 100473
+    },
+    {
+      "epoch": 10.31898942179316,
+      "grad_norm": 0.0440213605761528,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 100476
+    },
+    {
+      "epoch": 10.319297524905002,
+      "grad_norm": 0.1071145236492157,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 100479
+    },
+    {
+      "epoch": 10.319605628016843,
+      "grad_norm": 0.06137925758957863,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 100482
+    },
+    {
+      "epoch": 10.319913731128684,
+      "grad_norm": 0.06886938214302063,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 100485
+    },
+    {
+      "epoch": 10.320221834240526,
+      "grad_norm": 0.049702685326337814,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 100488
+    },
+    {
+      "epoch": 10.320529937352367,
+      "grad_norm": 0.11094728857278824,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 100491
+    },
+    {
+      "epoch": 10.320838040464208,
+      "grad_norm": 0.06528878957033157,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 100494
+    },
+    {
+      "epoch": 10.32114614357605,
+      "grad_norm": 0.06091786548495293,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 100497
+    },
+    {
+      "epoch": 10.32145424668789,
+      "grad_norm": 0.09342446178197861,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 100500
+    },
+    {
+      "epoch": 10.321762349799734,
+      "grad_norm": 0.08900013566017151,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 100503
+    },
+    {
+      "epoch": 10.322070452911575,
+      "grad_norm": 0.04888587072491646,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 100506
+    },
+    {
+      "epoch": 10.322378556023416,
+      "grad_norm": 0.10019969940185547,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 100509
+    },
+    {
+      "epoch": 10.322686659135258,
+      "grad_norm": 0.11456365883350372,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 100512
+    },
+    {
+      "epoch": 10.322994762247099,
+      "grad_norm": 0.09369392693042755,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 100515
+    },
+    {
+      "epoch": 10.32330286535894,
+      "grad_norm": 0.08547468483448029,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 100518
+    },
+    {
+      "epoch": 10.323610968470781,
+      "grad_norm": 0.15878820419311523,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 100521
+    },
+    {
+      "epoch": 10.323919071582623,
+      "grad_norm": 0.1548968404531479,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 100524
+    },
+    {
+      "epoch": 10.324227174694464,
+      "grad_norm": 0.12693747878074646,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 100527
+    },
+    {
+      "epoch": 10.324535277806305,
+      "grad_norm": 0.08565516024827957,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 100530
+    },
+    {
+      "epoch": 10.324843380918148,
+      "grad_norm": 0.04825468361377716,
+      "learning_rate": 0.01,
+      "loss": 1.8997,
+      "step": 100533
+    },
+    {
+      "epoch": 10.32515148402999,
+      "grad_norm": 0.04770709201693535,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 100536
+    },
+    {
+      "epoch": 10.32545958714183,
+      "grad_norm": 0.038272783160209656,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 100539
+    },
+    {
+      "epoch": 10.325767690253672,
+      "grad_norm": 0.041678592562675476,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 100542
+    },
+    {
+      "epoch": 10.326075793365513,
+      "grad_norm": 0.03310447931289673,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 100545
+    },
+    {
+      "epoch": 10.326383896477354,
+      "grad_norm": 0.03896121308207512,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 100548
+    },
+    {
+      "epoch": 10.326691999589196,
+      "grad_norm": 0.04841841757297516,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 100551
+    },
+    {
+      "epoch": 10.327000102701037,
+      "grad_norm": 0.08073825389146805,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 100554
+    },
+    {
+      "epoch": 10.327308205812878,
+      "grad_norm": 0.09406491369009018,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 100557
+    },
+    {
+      "epoch": 10.32761630892472,
+      "grad_norm": 0.044824738055467606,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 100560
+    },
+    {
+      "epoch": 10.32792441203656,
+      "grad_norm": 0.08207328617572784,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 100563
+    },
+    {
+      "epoch": 10.328232515148404,
+      "grad_norm": 0.10041683167219162,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 100566
+    },
+    {
+      "epoch": 10.328540618260245,
+      "grad_norm": 0.04280916601419449,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 100569
+    },
+    {
+      "epoch": 10.328848721372086,
+      "grad_norm": 0.047373414039611816,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 100572
+    },
+    {
+      "epoch": 10.329156824483928,
+      "grad_norm": 0.06335277855396271,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 100575
+    },
+    {
+      "epoch": 10.329464927595769,
+      "grad_norm": 0.05805109813809395,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 100578
+    },
+    {
+      "epoch": 10.32977303070761,
+      "grad_norm": 0.11722945421934128,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 100581
+    },
+    {
+      "epoch": 10.330081133819451,
+      "grad_norm": 0.06972774118185043,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 100584
+    },
+    {
+      "epoch": 10.330389236931293,
+      "grad_norm": 0.07877903431653976,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 100587
+    },
+    {
+      "epoch": 10.330697340043134,
+      "grad_norm": 0.08234605193138123,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 100590
+    },
+    {
+      "epoch": 10.331005443154975,
+      "grad_norm": 0.05679229274392128,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 100593
+    },
+    {
+      "epoch": 10.331313546266816,
+      "grad_norm": 0.04508120194077492,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 100596
+    },
+    {
+      "epoch": 10.33162164937866,
+      "grad_norm": 0.08512305468320847,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 100599
+    },
+    {
+      "epoch": 10.3319297524905,
+      "grad_norm": 0.07284285128116608,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 100602
+    },
+    {
+      "epoch": 10.332237855602342,
+      "grad_norm": 0.11658084392547607,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 100605
+    },
+    {
+      "epoch": 10.332545958714183,
+      "grad_norm": 0.14260004460811615,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 100608
+    },
+    {
+      "epoch": 10.332854061826025,
+      "grad_norm": 0.06603503972291946,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 100611
+    },
+    {
+      "epoch": 10.333162164937866,
+      "grad_norm": 0.037755999714136124,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 100614
+    },
+    {
+      "epoch": 10.333470268049707,
+      "grad_norm": 0.04604731872677803,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 100617
+    },
+    {
+      "epoch": 10.333778371161548,
+      "grad_norm": 0.05650022253394127,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 100620
+    },
+    {
+      "epoch": 10.33408647427339,
+      "grad_norm": 0.05804010108113289,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 100623
+    },
+    {
+      "epoch": 10.33439457738523,
+      "grad_norm": 0.06526564806699753,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 100626
+    },
+    {
+      "epoch": 10.334702680497074,
+      "grad_norm": 0.055868957191705704,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 100629
+    },
+    {
+      "epoch": 10.335010783608915,
+      "grad_norm": 0.07749674469232559,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 100632
+    },
+    {
+      "epoch": 10.335318886720756,
+      "grad_norm": 0.08783943951129913,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 100635
+    },
+    {
+      "epoch": 10.335626989832598,
+      "grad_norm": 0.0857509896159172,
+      "learning_rate": 0.01,
+      "loss": 1.9775,
+      "step": 100638
+    },
+    {
+      "epoch": 10.335935092944439,
+      "grad_norm": 0.0713290348649025,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 100641
+    },
+    {
+      "epoch": 10.33624319605628,
+      "grad_norm": 0.08365776389837265,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 100644
+    },
+    {
+      "epoch": 10.336551299168121,
+      "grad_norm": 0.06503187865018845,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 100647
+    },
+    {
+      "epoch": 10.336859402279963,
+      "grad_norm": 0.08043991774320602,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 100650
+    },
+    {
+      "epoch": 10.337167505391804,
+      "grad_norm": 0.06439603865146637,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 100653
+    },
+    {
+      "epoch": 10.337475608503645,
+      "grad_norm": 0.1296110451221466,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 100656
+    },
+    {
+      "epoch": 10.337783711615486,
+      "grad_norm": 0.1248815581202507,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 100659
+    },
+    {
+      "epoch": 10.33809181472733,
+      "grad_norm": 0.09833067655563354,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 100662
+    },
+    {
+      "epoch": 10.33839991783917,
+      "grad_norm": 0.056371044367551804,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 100665
+    },
+    {
+      "epoch": 10.338708020951012,
+      "grad_norm": 0.07650956511497498,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 100668
+    },
+    {
+      "epoch": 10.339016124062853,
+      "grad_norm": 0.049769867211580276,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 100671
+    },
+    {
+      "epoch": 10.339324227174695,
+      "grad_norm": 0.03484850376844406,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 100674
+    },
+    {
+      "epoch": 10.339632330286536,
+      "grad_norm": 0.07614227384328842,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 100677
+    },
+    {
+      "epoch": 10.339940433398377,
+      "grad_norm": 0.08095826953649521,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 100680
+    },
+    {
+      "epoch": 10.340248536510218,
+      "grad_norm": 0.05176486819982529,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 100683
+    },
+    {
+      "epoch": 10.34055663962206,
+      "grad_norm": 0.0889786034822464,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 100686
+    },
+    {
+      "epoch": 10.3408647427339,
+      "grad_norm": 0.07184679061174393,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 100689
+    },
+    {
+      "epoch": 10.341172845845744,
+      "grad_norm": 0.057422127574682236,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 100692
+    },
+    {
+      "epoch": 10.341480948957585,
+      "grad_norm": 0.11470889300107956,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 100695
+    },
+    {
+      "epoch": 10.341789052069426,
+      "grad_norm": 0.03806557506322861,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 100698
+    },
+    {
+      "epoch": 10.342097155181268,
+      "grad_norm": 0.14512214064598083,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 100701
+    },
+    {
+      "epoch": 10.342405258293109,
+      "grad_norm": 0.058107271790504456,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 100704
+    },
+    {
+      "epoch": 10.34271336140495,
+      "grad_norm": 0.040821373462677,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 100707
+    },
+    {
+      "epoch": 10.343021464516791,
+      "grad_norm": 0.048679567873477936,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 100710
+    },
+    {
+      "epoch": 10.343329567628633,
+      "grad_norm": 0.03365056589245796,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 100713
+    },
+    {
+      "epoch": 10.343637670740474,
+      "grad_norm": 0.13085153698921204,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 100716
+    },
+    {
+      "epoch": 10.343945773852315,
+      "grad_norm": 0.037281155586242676,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 100719
+    },
+    {
+      "epoch": 10.344253876964157,
+      "grad_norm": 0.04350685700774193,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 100722
+    },
+    {
+      "epoch": 10.344561980076,
+      "grad_norm": 0.03874243050813675,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 100725
+    },
+    {
+      "epoch": 10.34487008318784,
+      "grad_norm": 0.052796006202697754,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 100728
+    },
+    {
+      "epoch": 10.345178186299682,
+      "grad_norm": 0.04441223666071892,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 100731
+    },
+    {
+      "epoch": 10.345486289411523,
+      "grad_norm": 0.04818743094801903,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 100734
+    },
+    {
+      "epoch": 10.345794392523365,
+      "grad_norm": 0.07991688698530197,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 100737
+    },
+    {
+      "epoch": 10.346102495635206,
+      "grad_norm": 0.07785684615373611,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 100740
+    },
+    {
+      "epoch": 10.346410598747047,
+      "grad_norm": 0.05483667552471161,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 100743
+    },
+    {
+      "epoch": 10.346718701858888,
+      "grad_norm": 0.05420774593949318,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 100746
+    },
+    {
+      "epoch": 10.34702680497073,
+      "grad_norm": 0.03596429526805878,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 100749
+    },
+    {
+      "epoch": 10.347334908082571,
+      "grad_norm": 0.04498706012964249,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 100752
+    },
+    {
+      "epoch": 10.347643011194412,
+      "grad_norm": 0.038987092673778534,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 100755
+    },
+    {
+      "epoch": 10.347951114306255,
+      "grad_norm": 0.13664734363555908,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 100758
+    },
+    {
+      "epoch": 10.348259217418097,
+      "grad_norm": 0.0331043042242527,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 100761
+    },
+    {
+      "epoch": 10.348567320529938,
+      "grad_norm": 0.1004151701927185,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 100764
+    },
+    {
+      "epoch": 10.348875423641779,
+      "grad_norm": 0.10532857477664948,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 100767
+    },
+    {
+      "epoch": 10.34918352675362,
+      "grad_norm": 0.05065859481692314,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 100770
+    },
+    {
+      "epoch": 10.349491629865462,
+      "grad_norm": 0.06383136659860611,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 100773
+    },
+    {
+      "epoch": 10.349799732977303,
+      "grad_norm": 0.08385428041219711,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 100776
+    },
+    {
+      "epoch": 10.350107836089144,
+      "grad_norm": 0.06286201626062393,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 100779
+    },
+    {
+      "epoch": 10.350415939200985,
+      "grad_norm": 0.056411102414131165,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 100782
+    },
+    {
+      "epoch": 10.350724042312827,
+      "grad_norm": 0.0546356700360775,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 100785
+    },
+    {
+      "epoch": 10.351032145424668,
+      "grad_norm": 0.04161200299859047,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 100788
+    },
+    {
+      "epoch": 10.351340248536511,
+      "grad_norm": 0.1038103774189949,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 100791
+    },
+    {
+      "epoch": 10.351648351648352,
+      "grad_norm": 0.038900673389434814,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 100794
+    },
+    {
+      "epoch": 10.351956454760193,
+      "grad_norm": 0.04477495327591896,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 100797
+    },
+    {
+      "epoch": 10.352264557872035,
+      "grad_norm": 0.034626856446266174,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 100800
+    },
+    {
+      "epoch": 10.352572660983876,
+      "grad_norm": 0.13096442818641663,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 100803
+    },
+    {
+      "epoch": 10.352880764095717,
+      "grad_norm": 0.05774092674255371,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 100806
+    },
+    {
+      "epoch": 10.353188867207558,
+      "grad_norm": 0.12401199340820312,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 100809
+    },
+    {
+      "epoch": 10.3534969703194,
+      "grad_norm": 0.09031431376934052,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 100812
+    },
+    {
+      "epoch": 10.353805073431241,
+      "grad_norm": 0.052858997136354446,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 100815
+    },
+    {
+      "epoch": 10.354113176543082,
+      "grad_norm": 0.059029802680015564,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 100818
+    },
+    {
+      "epoch": 10.354421279654925,
+      "grad_norm": 0.041640978306531906,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 100821
+    },
+    {
+      "epoch": 10.354729382766767,
+      "grad_norm": 0.1006421223282814,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 100824
+    },
+    {
+      "epoch": 10.355037485878608,
+      "grad_norm": 0.0704476535320282,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 100827
+    },
+    {
+      "epoch": 10.355345588990449,
+      "grad_norm": 0.03361036628484726,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 100830
+    },
+    {
+      "epoch": 10.35565369210229,
+      "grad_norm": 0.052998024970293045,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 100833
+    },
+    {
+      "epoch": 10.355961795214132,
+      "grad_norm": 0.06729929149150848,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 100836
+    },
+    {
+      "epoch": 10.356269898325973,
+      "grad_norm": 0.03766258805990219,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 100839
+    },
+    {
+      "epoch": 10.356578001437814,
+      "grad_norm": 0.13822922110557556,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 100842
+    },
+    {
+      "epoch": 10.356886104549655,
+      "grad_norm": 0.09319800138473511,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 100845
+    },
+    {
+      "epoch": 10.357194207661497,
+      "grad_norm": 0.07348558306694031,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 100848
+    },
+    {
+      "epoch": 10.357502310773338,
+      "grad_norm": 0.14174668490886688,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 100851
+    },
+    {
+      "epoch": 10.357810413885181,
+      "grad_norm": 0.05898958817124367,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 100854
+    },
+    {
+      "epoch": 10.358118516997022,
+      "grad_norm": 0.08681494742631912,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 100857
+    },
+    {
+      "epoch": 10.358426620108864,
+      "grad_norm": 0.036777254194021225,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 100860
+    },
+    {
+      "epoch": 10.358734723220705,
+      "grad_norm": 0.06975427269935608,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 100863
+    },
+    {
+      "epoch": 10.359042826332546,
+      "grad_norm": 0.06511854380369186,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 100866
+    },
+    {
+      "epoch": 10.359350929444387,
+      "grad_norm": 0.057764261960983276,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 100869
+    },
+    {
+      "epoch": 10.359659032556229,
+      "grad_norm": 0.08371473848819733,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 100872
+    },
+    {
+      "epoch": 10.35996713566807,
+      "grad_norm": 0.0854247510433197,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 100875
+    },
+    {
+      "epoch": 10.360275238779911,
+      "grad_norm": 0.04530758038163185,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 100878
+    },
+    {
+      "epoch": 10.360583341891752,
+      "grad_norm": 0.03615698590874672,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 100881
+    },
+    {
+      "epoch": 10.360891445003595,
+      "grad_norm": 0.06091855838894844,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 100884
+    },
+    {
+      "epoch": 10.361199548115437,
+      "grad_norm": 0.04644336923956871,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 100887
+    },
+    {
+      "epoch": 10.361507651227278,
+      "grad_norm": 0.043100859969854355,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 100890
+    },
+    {
+      "epoch": 10.36181575433912,
+      "grad_norm": 0.04365437850356102,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 100893
+    },
+    {
+      "epoch": 10.36212385745096,
+      "grad_norm": 0.11414897441864014,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 100896
+    },
+    {
+      "epoch": 10.362431960562802,
+      "grad_norm": 0.05801675096154213,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 100899
+    },
+    {
+      "epoch": 10.362740063674643,
+      "grad_norm": 0.10355260968208313,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 100902
+    },
+    {
+      "epoch": 10.363048166786484,
+      "grad_norm": 0.049150459468364716,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 100905
+    },
+    {
+      "epoch": 10.363356269898325,
+      "grad_norm": 0.08133083581924438,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 100908
+    },
+    {
+      "epoch": 10.363664373010167,
+      "grad_norm": 0.0906137004494667,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 100911
+    },
+    {
+      "epoch": 10.363972476122008,
+      "grad_norm": 0.06084192171692848,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 100914
+    },
+    {
+      "epoch": 10.364280579233851,
+      "grad_norm": 0.0443035326898098,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 100917
+    },
+    {
+      "epoch": 10.364588682345692,
+      "grad_norm": 0.03251441940665245,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 100920
+    },
+    {
+      "epoch": 10.364896785457534,
+      "grad_norm": 0.056791745126247406,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 100923
+    },
+    {
+      "epoch": 10.365204888569375,
+      "grad_norm": 0.11314947158098221,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 100926
+    },
+    {
+      "epoch": 10.365512991681216,
+      "grad_norm": 0.09841377288103104,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 100929
+    },
+    {
+      "epoch": 10.365821094793057,
+      "grad_norm": 0.060636650770902634,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 100932
+    },
+    {
+      "epoch": 10.366129197904899,
+      "grad_norm": 0.05676575005054474,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 100935
+    },
+    {
+      "epoch": 10.36643730101674,
+      "grad_norm": 0.1176658570766449,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 100938
+    },
+    {
+      "epoch": 10.366745404128581,
+      "grad_norm": 0.033294521272182465,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 100941
+    },
+    {
+      "epoch": 10.367053507240422,
+      "grad_norm": 0.061117228120565414,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 100944
+    },
+    {
+      "epoch": 10.367361610352265,
+      "grad_norm": 0.04445219412446022,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 100947
+    },
+    {
+      "epoch": 10.367669713464107,
+      "grad_norm": 0.06714378297328949,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 100950
+    },
+    {
+      "epoch": 10.367977816575948,
+      "grad_norm": 0.08990859985351562,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 100953
+    },
+    {
+      "epoch": 10.36828591968779,
+      "grad_norm": 0.1147884726524353,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 100956
+    },
+    {
+      "epoch": 10.36859402279963,
+      "grad_norm": 0.07784424722194672,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 100959
+    },
+    {
+      "epoch": 10.368902125911472,
+      "grad_norm": 0.06763900071382523,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 100962
+    },
+    {
+      "epoch": 10.369210229023313,
+      "grad_norm": 0.047785356640815735,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 100965
+    },
+    {
+      "epoch": 10.369518332135154,
+      "grad_norm": 0.044531069695949554,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 100968
+    },
+    {
+      "epoch": 10.369826435246996,
+      "grad_norm": 0.06693824380636215,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 100971
+    },
+    {
+      "epoch": 10.370134538358837,
+      "grad_norm": 0.058132026344537735,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 100974
+    },
+    {
+      "epoch": 10.370442641470678,
+      "grad_norm": 0.09861744195222855,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 100977
+    },
+    {
+      "epoch": 10.370750744582521,
+      "grad_norm": 0.04031767696142197,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 100980
+    },
+    {
+      "epoch": 10.371058847694362,
+      "grad_norm": 0.10672684013843536,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 100983
+    },
+    {
+      "epoch": 10.371366950806204,
+      "grad_norm": 0.04467643052339554,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 100986
+    },
+    {
+      "epoch": 10.371675053918045,
+      "grad_norm": 0.09104414284229279,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 100989
+    },
+    {
+      "epoch": 10.371983157029886,
+      "grad_norm": 0.05075051635503769,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 100992
+    },
+    {
+      "epoch": 10.372291260141727,
+      "grad_norm": 0.11250297725200653,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 100995
+    },
+    {
+      "epoch": 10.372599363253569,
+      "grad_norm": 0.03872716799378395,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 100998
+    },
+    {
+      "epoch": 10.37290746636541,
+      "grad_norm": 0.08845558017492294,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 101001
+    },
+    {
+      "epoch": 10.373215569477251,
+      "grad_norm": 0.07507075369358063,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 101004
+    },
+    {
+      "epoch": 10.373523672589092,
+      "grad_norm": 0.0837487205862999,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 101007
+    },
+    {
+      "epoch": 10.373831775700934,
+      "grad_norm": 0.06820300966501236,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 101010
+    },
+    {
+      "epoch": 10.374139878812777,
+      "grad_norm": 0.11727225035429001,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 101013
+    },
+    {
+      "epoch": 10.374447981924618,
+      "grad_norm": 0.06376228481531143,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 101016
+    },
+    {
+      "epoch": 10.37475608503646,
+      "grad_norm": 0.035101041197776794,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 101019
+    },
+    {
+      "epoch": 10.3750641881483,
+      "grad_norm": 0.032660722732543945,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 101022
+    },
+    {
+      "epoch": 10.375372291260142,
+      "grad_norm": 0.04537061229348183,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 101025
+    },
+    {
+      "epoch": 10.375680394371983,
+      "grad_norm": 0.04433763399720192,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 101028
+    },
+    {
+      "epoch": 10.375988497483824,
+      "grad_norm": 0.0950620248913765,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 101031
+    },
+    {
+      "epoch": 10.376296600595666,
+      "grad_norm": 0.07553868740797043,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 101034
+    },
+    {
+      "epoch": 10.376604703707507,
+      "grad_norm": 0.05983571335673332,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 101037
+    },
+    {
+      "epoch": 10.376912806819348,
+      "grad_norm": 0.08115866035223007,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 101040
+    },
+    {
+      "epoch": 10.37722090993119,
+      "grad_norm": 0.12146302312612534,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 101043
+    },
+    {
+      "epoch": 10.377529013043032,
+      "grad_norm": 0.06241605430841446,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 101046
+    },
+    {
+      "epoch": 10.377837116154874,
+      "grad_norm": 0.03604963421821594,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 101049
+    },
+    {
+      "epoch": 10.378145219266715,
+      "grad_norm": 0.0537908673286438,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 101052
+    },
+    {
+      "epoch": 10.378453322378556,
+      "grad_norm": 0.044751640409231186,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 101055
+    },
+    {
+      "epoch": 10.378761425490397,
+      "grad_norm": 0.0636962354183197,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 101058
+    },
+    {
+      "epoch": 10.379069528602239,
+      "grad_norm": 0.11618872731924057,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 101061
+    },
+    {
+      "epoch": 10.37937763171408,
+      "grad_norm": 0.1530059278011322,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 101064
+    },
+    {
+      "epoch": 10.379685734825921,
+      "grad_norm": 0.05721871927380562,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 101067
+    },
+    {
+      "epoch": 10.379993837937763,
+      "grad_norm": 0.07816772162914276,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 101070
+    },
+    {
+      "epoch": 10.380301941049604,
+      "grad_norm": 0.04495503008365631,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 101073
+    },
+    {
+      "epoch": 10.380610044161447,
+      "grad_norm": 0.07357476651668549,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 101076
+    },
+    {
+      "epoch": 10.380918147273288,
+      "grad_norm": 0.049054451286792755,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 101079
+    },
+    {
+      "epoch": 10.38122625038513,
+      "grad_norm": 0.20468761026859283,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 101082
+    },
+    {
+      "epoch": 10.38153435349697,
+      "grad_norm": 0.12951242923736572,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 101085
+    },
+    {
+      "epoch": 10.381842456608812,
+      "grad_norm": 0.05691198632121086,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 101088
+    },
+    {
+      "epoch": 10.382150559720653,
+      "grad_norm": 0.09454532712697983,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 101091
+    },
+    {
+      "epoch": 10.382458662832494,
+      "grad_norm": 0.046739835292100906,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 101094
+    },
+    {
+      "epoch": 10.382766765944336,
+      "grad_norm": 0.04217229038476944,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 101097
+    },
+    {
+      "epoch": 10.383074869056177,
+      "grad_norm": 0.0381905771791935,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 101100
+    },
+    {
+      "epoch": 10.383382972168018,
+      "grad_norm": 0.03863144665956497,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 101103
+    },
+    {
+      "epoch": 10.38369107527986,
+      "grad_norm": 0.068651482462883,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 101106
+    },
+    {
+      "epoch": 10.383999178391703,
+      "grad_norm": 0.062461916357278824,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 101109
+    },
+    {
+      "epoch": 10.384307281503544,
+      "grad_norm": 0.0718960091471672,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 101112
+    },
+    {
+      "epoch": 10.384615384615385,
+      "grad_norm": 0.044950731098651886,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 101115
+    },
+    {
+      "epoch": 10.384923487727226,
+      "grad_norm": 0.0483747161924839,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 101118
+    },
+    {
+      "epoch": 10.385231590839068,
+      "grad_norm": 0.05185241624712944,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 101121
+    },
+    {
+      "epoch": 10.385539693950909,
+      "grad_norm": 0.03757419437170029,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 101124
+    },
+    {
+      "epoch": 10.38584779706275,
+      "grad_norm": 0.12432057410478592,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 101127
+    },
+    {
+      "epoch": 10.386155900174591,
+      "grad_norm": 0.0804310292005539,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 101130
+    },
+    {
+      "epoch": 10.386464003286433,
+      "grad_norm": 0.09767717868089676,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 101133
+    },
+    {
+      "epoch": 10.386772106398274,
+      "grad_norm": 0.05865232273936272,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 101136
+    },
+    {
+      "epoch": 10.387080209510117,
+      "grad_norm": 0.04162360355257988,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 101139
+    },
+    {
+      "epoch": 10.387388312621958,
+      "grad_norm": 0.12615089118480682,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 101142
+    },
+    {
+      "epoch": 10.3876964157338,
+      "grad_norm": 0.04601338505744934,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 101145
+    },
+    {
+      "epoch": 10.38800451884564,
+      "grad_norm": 0.0515436977148056,
+      "learning_rate": 0.01,
+      "loss": 1.9737,
+      "step": 101148
+    },
+    {
+      "epoch": 10.388312621957482,
+      "grad_norm": 0.05919606238603592,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 101151
+    },
+    {
+      "epoch": 10.388620725069323,
+      "grad_norm": 0.2011437565088272,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 101154
+    },
+    {
+      "epoch": 10.388928828181164,
+      "grad_norm": 0.13947412371635437,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 101157
+    },
+    {
+      "epoch": 10.389236931293006,
+      "grad_norm": 0.10039437562227249,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 101160
+    },
+    {
+      "epoch": 10.389545034404847,
+      "grad_norm": 0.052422210574150085,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 101163
+    },
+    {
+      "epoch": 10.389853137516688,
+      "grad_norm": 0.03887460008263588,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 101166
+    },
+    {
+      "epoch": 10.39016124062853,
+      "grad_norm": 0.06007706746459007,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 101169
+    },
+    {
+      "epoch": 10.390469343740373,
+      "grad_norm": 0.05732855945825577,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 101172
+    },
+    {
+      "epoch": 10.390777446852214,
+      "grad_norm": 0.0792255848646164,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 101175
+    },
+    {
+      "epoch": 10.391085549964055,
+      "grad_norm": 0.06666497141122818,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 101178
+    },
+    {
+      "epoch": 10.391393653075896,
+      "grad_norm": 0.06329864263534546,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 101181
+    },
+    {
+      "epoch": 10.391701756187738,
+      "grad_norm": 0.17151835560798645,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 101184
+    },
+    {
+      "epoch": 10.392009859299579,
+      "grad_norm": 0.11637787520885468,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 101187
+    },
+    {
+      "epoch": 10.39231796241142,
+      "grad_norm": 0.06190858781337738,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 101190
+    },
+    {
+      "epoch": 10.392626065523261,
+      "grad_norm": 0.042335715144872665,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 101193
+    },
+    {
+      "epoch": 10.392934168635103,
+      "grad_norm": 0.03620636835694313,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 101196
+    },
+    {
+      "epoch": 10.393242271746944,
+      "grad_norm": 0.03169773519039154,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 101199
+    },
+    {
+      "epoch": 10.393550374858787,
+      "grad_norm": 0.11039429903030396,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 101202
+    },
+    {
+      "epoch": 10.393858477970628,
+      "grad_norm": 0.07459361106157303,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 101205
+    },
+    {
+      "epoch": 10.39416658108247,
+      "grad_norm": 0.08071883022785187,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 101208
+    },
+    {
+      "epoch": 10.39447468419431,
+      "grad_norm": 0.08011837303638458,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 101211
+    },
+    {
+      "epoch": 10.394782787306152,
+      "grad_norm": 0.08391046524047852,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 101214
+    },
+    {
+      "epoch": 10.395090890417993,
+      "grad_norm": 0.07405078411102295,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 101217
+    },
+    {
+      "epoch": 10.395398993529835,
+      "grad_norm": 0.10181564837694168,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 101220
+    },
+    {
+      "epoch": 10.395707096641676,
+      "grad_norm": 0.0628264769911766,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 101223
+    },
+    {
+      "epoch": 10.396015199753517,
+      "grad_norm": 0.06534316390752792,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 101226
+    },
+    {
+      "epoch": 10.396323302865358,
+      "grad_norm": 0.07027024030685425,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 101229
+    },
+    {
+      "epoch": 10.3966314059772,
+      "grad_norm": 0.09791549295186996,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 101232
+    },
+    {
+      "epoch": 10.396939509089043,
+      "grad_norm": 0.1024041548371315,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 101235
+    },
+    {
+      "epoch": 10.397247612200884,
+      "grad_norm": 0.06439261138439178,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 101238
+    },
+    {
+      "epoch": 10.397555715312725,
+      "grad_norm": 0.04745156317949295,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 101241
+    },
+    {
+      "epoch": 10.397863818424566,
+      "grad_norm": 0.0973968431353569,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 101244
+    },
+    {
+      "epoch": 10.398171921536408,
+      "grad_norm": 0.0683659166097641,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 101247
+    },
+    {
+      "epoch": 10.398480024648249,
+      "grad_norm": 0.04603632539510727,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 101250
+    },
+    {
+      "epoch": 10.39878812776009,
+      "grad_norm": 0.11405742913484573,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 101253
+    },
+    {
+      "epoch": 10.399096230871931,
+      "grad_norm": 0.046128056943416595,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 101256
+    },
+    {
+      "epoch": 10.399404333983773,
+      "grad_norm": 0.04291325807571411,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 101259
+    },
+    {
+      "epoch": 10.399712437095614,
+      "grad_norm": 0.029826214537024498,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 101262
+    },
+    {
+      "epoch": 10.400020540207455,
+      "grad_norm": 0.09401890635490417,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 101265
+    },
+    {
+      "epoch": 10.400328643319298,
+      "grad_norm": 0.04324827343225479,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 101268
+    },
+    {
+      "epoch": 10.40063674643114,
+      "grad_norm": 0.05711056664586067,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 101271
+    },
+    {
+      "epoch": 10.40094484954298,
+      "grad_norm": 0.07818640768527985,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 101274
+    },
+    {
+      "epoch": 10.401252952654822,
+      "grad_norm": 0.05406804382801056,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 101277
+    },
+    {
+      "epoch": 10.401561055766663,
+      "grad_norm": 0.051603805273771286,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 101280
+    },
+    {
+      "epoch": 10.401869158878505,
+      "grad_norm": 0.11336014419794083,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 101283
+    },
+    {
+      "epoch": 10.402177261990346,
+      "grad_norm": 0.07639746367931366,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 101286
+    },
+    {
+      "epoch": 10.402485365102187,
+      "grad_norm": 0.08636012673377991,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 101289
+    },
+    {
+      "epoch": 10.402793468214028,
+      "grad_norm": 0.04350695013999939,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 101292
+    },
+    {
+      "epoch": 10.40310157132587,
+      "grad_norm": 0.030775388702750206,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 101295
+    },
+    {
+      "epoch": 10.403409674437711,
+      "grad_norm": 0.05844002217054367,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 101298
+    },
+    {
+      "epoch": 10.403717777549554,
+      "grad_norm": 0.0463801771402359,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 101301
+    },
+    {
+      "epoch": 10.404025880661395,
+      "grad_norm": 0.03980204463005066,
+      "learning_rate": 0.01,
+      "loss": 1.9772,
+      "step": 101304
+    },
+    {
+      "epoch": 10.404333983773236,
+      "grad_norm": 0.043364375829696655,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 101307
+    },
+    {
+      "epoch": 10.404642086885078,
+      "grad_norm": 0.1211235523223877,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 101310
+    },
+    {
+      "epoch": 10.404950189996919,
+      "grad_norm": 0.044283825904130936,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 101313
+    },
+    {
+      "epoch": 10.40525829310876,
+      "grad_norm": 0.06269537657499313,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 101316
+    },
+    {
+      "epoch": 10.405566396220602,
+      "grad_norm": 0.11686865240335464,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 101319
+    },
+    {
+      "epoch": 10.405874499332443,
+      "grad_norm": 0.04533189907670021,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 101322
+    },
+    {
+      "epoch": 10.406182602444284,
+      "grad_norm": 0.06254197657108307,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 101325
+    },
+    {
+      "epoch": 10.406490705556125,
+      "grad_norm": 0.06646399945020676,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 101328
+    },
+    {
+      "epoch": 10.406798808667968,
+      "grad_norm": 0.057041414082050323,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 101331
+    },
+    {
+      "epoch": 10.40710691177981,
+      "grad_norm": 0.10227653384208679,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 101334
+    },
+    {
+      "epoch": 10.40741501489165,
+      "grad_norm": 0.07010546326637268,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 101337
+    },
+    {
+      "epoch": 10.407723118003492,
+      "grad_norm": 0.10337510704994202,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 101340
+    },
+    {
+      "epoch": 10.408031221115333,
+      "grad_norm": 0.05713025480508804,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 101343
+    },
+    {
+      "epoch": 10.408339324227175,
+      "grad_norm": 0.08766386657953262,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 101346
+    },
+    {
+      "epoch": 10.408647427339016,
+      "grad_norm": 0.057308826595544815,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 101349
+    },
+    {
+      "epoch": 10.408955530450857,
+      "grad_norm": 0.11045978218317032,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 101352
+    },
+    {
+      "epoch": 10.409263633562698,
+      "grad_norm": 0.04463179409503937,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 101355
+    },
+    {
+      "epoch": 10.40957173667454,
+      "grad_norm": 0.06062712520360947,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 101358
+    },
+    {
+      "epoch": 10.409879839786381,
+      "grad_norm": 0.04317406192421913,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 101361
+    },
+    {
+      "epoch": 10.410187942898224,
+      "grad_norm": 0.07476109266281128,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 101364
+    },
+    {
+      "epoch": 10.410496046010065,
+      "grad_norm": 0.040464941412210464,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 101367
+    },
+    {
+      "epoch": 10.410804149121907,
+      "grad_norm": 0.0591270849108696,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 101370
+    },
+    {
+      "epoch": 10.411112252233748,
+      "grad_norm": 0.03704477474093437,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 101373
+    },
+    {
+      "epoch": 10.411420355345589,
+      "grad_norm": 0.04234340786933899,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 101376
+    },
+    {
+      "epoch": 10.41172845845743,
+      "grad_norm": 0.08466865122318268,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 101379
+    },
+    {
+      "epoch": 10.412036561569272,
+      "grad_norm": 0.06604104489088058,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 101382
+    },
+    {
+      "epoch": 10.412344664681113,
+      "grad_norm": 0.07191555202007294,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 101385
+    },
+    {
+      "epoch": 10.412652767792954,
+      "grad_norm": 0.05433400720357895,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 101388
+    },
+    {
+      "epoch": 10.412960870904795,
+      "grad_norm": 0.08139751106500626,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 101391
+    },
+    {
+      "epoch": 10.413268974016638,
+      "grad_norm": 0.08413826674222946,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 101394
+    },
+    {
+      "epoch": 10.41357707712848,
+      "grad_norm": 0.038020603358745575,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 101397
+    },
+    {
+      "epoch": 10.413885180240321,
+      "grad_norm": 0.11257899552583694,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 101400
+    },
+    {
+      "epoch": 10.414193283352162,
+      "grad_norm": 0.05421124026179314,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 101403
+    },
+    {
+      "epoch": 10.414501386464003,
+      "grad_norm": 0.08173589408397675,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 101406
+    },
+    {
+      "epoch": 10.414809489575845,
+      "grad_norm": 0.09774955362081528,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 101409
+    },
+    {
+      "epoch": 10.415117592687686,
+      "grad_norm": 0.06760906428098679,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 101412
+    },
+    {
+      "epoch": 10.415425695799527,
+      "grad_norm": 0.06532621383666992,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 101415
+    },
+    {
+      "epoch": 10.415733798911369,
+      "grad_norm": 0.051986172795295715,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 101418
+    },
+    {
+      "epoch": 10.41604190202321,
+      "grad_norm": 0.04217454046010971,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 101421
+    },
+    {
+      "epoch": 10.416350005135051,
+      "grad_norm": 0.04921566694974899,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 101424
+    },
+    {
+      "epoch": 10.416658108246894,
+      "grad_norm": 0.09551133960485458,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 101427
+    },
+    {
+      "epoch": 10.416966211358735,
+      "grad_norm": 0.07457684725522995,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 101430
+    },
+    {
+      "epoch": 10.417274314470577,
+      "grad_norm": 0.05960552766919136,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 101433
+    },
+    {
+      "epoch": 10.417582417582418,
+      "grad_norm": 0.04713088646531105,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 101436
+    },
+    {
+      "epoch": 10.41789052069426,
+      "grad_norm": 0.08014922589063644,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 101439
+    },
+    {
+      "epoch": 10.4181986238061,
+      "grad_norm": 0.08689352869987488,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 101442
+    },
+    {
+      "epoch": 10.418506726917942,
+      "grad_norm": 0.04182841256260872,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 101445
+    },
+    {
+      "epoch": 10.418814830029783,
+      "grad_norm": 0.0788508802652359,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 101448
+    },
+    {
+      "epoch": 10.419122933141624,
+      "grad_norm": 0.06277182698249817,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 101451
+    },
+    {
+      "epoch": 10.419431036253465,
+      "grad_norm": 0.04268977791070938,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 101454
+    },
+    {
+      "epoch": 10.419739139365308,
+      "grad_norm": 0.08060577511787415,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 101457
+    },
+    {
+      "epoch": 10.42004724247715,
+      "grad_norm": 0.05500905588269234,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 101460
+    },
+    {
+      "epoch": 10.420355345588991,
+      "grad_norm": 0.09350892156362534,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 101463
+    },
+    {
+      "epoch": 10.420663448700832,
+      "grad_norm": 0.10302212089300156,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 101466
+    },
+    {
+      "epoch": 10.420971551812674,
+      "grad_norm": 0.0546325221657753,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 101469
+    },
+    {
+      "epoch": 10.421279654924515,
+      "grad_norm": 0.10181728005409241,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 101472
+    },
+    {
+      "epoch": 10.421587758036356,
+      "grad_norm": 0.1607157438993454,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 101475
+    },
+    {
+      "epoch": 10.421895861148197,
+      "grad_norm": 0.0734848827123642,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 101478
+    },
+    {
+      "epoch": 10.422203964260039,
+      "grad_norm": 0.05575236305594444,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 101481
+    },
+    {
+      "epoch": 10.42251206737188,
+      "grad_norm": 0.03977865353226662,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 101484
+    },
+    {
+      "epoch": 10.422820170483721,
+      "grad_norm": 0.03199474513530731,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 101487
+    },
+    {
+      "epoch": 10.423128273595564,
+      "grad_norm": 0.03404001519083977,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 101490
+    },
+    {
+      "epoch": 10.423436376707405,
+      "grad_norm": 0.055631235241889954,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 101493
+    },
+    {
+      "epoch": 10.423744479819247,
+      "grad_norm": 0.03958062827587128,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 101496
+    },
+    {
+      "epoch": 10.424052582931088,
+      "grad_norm": 0.057350341230630875,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 101499
+    },
+    {
+      "epoch": 10.42436068604293,
+      "grad_norm": 0.041089460253715515,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 101502
+    },
+    {
+      "epoch": 10.42466878915477,
+      "grad_norm": 0.09128007292747498,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 101505
+    },
+    {
+      "epoch": 10.424976892266612,
+      "grad_norm": 0.03688080236315727,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 101508
+    },
+    {
+      "epoch": 10.425284995378453,
+      "grad_norm": 0.04330899193882942,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 101511
+    },
+    {
+      "epoch": 10.425593098490294,
+      "grad_norm": 0.0362963005900383,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 101514
+    },
+    {
+      "epoch": 10.425901201602136,
+      "grad_norm": 0.08184661716222763,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 101517
+    },
+    {
+      "epoch": 10.426209304713977,
+      "grad_norm": 0.07895220816135406,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 101520
+    },
+    {
+      "epoch": 10.42651740782582,
+      "grad_norm": 0.047668907791376114,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 101523
+    },
+    {
+      "epoch": 10.426825510937661,
+      "grad_norm": 0.049812547862529755,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 101526
+    },
+    {
+      "epoch": 10.427133614049502,
+      "grad_norm": 0.04061616584658623,
+      "learning_rate": 0.01,
+      "loss": 1.9689,
+      "step": 101529
+    },
+    {
+      "epoch": 10.427441717161344,
+      "grad_norm": 0.03955051302909851,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 101532
+    },
+    {
+      "epoch": 10.427749820273185,
+      "grad_norm": 0.05154630169272423,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 101535
+    },
+    {
+      "epoch": 10.428057923385026,
+      "grad_norm": 0.134513258934021,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 101538
+    },
+    {
+      "epoch": 10.428366026496867,
+      "grad_norm": 0.054661769419908524,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 101541
+    },
+    {
+      "epoch": 10.428674129608709,
+      "grad_norm": 0.04197604954242706,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 101544
+    },
+    {
+      "epoch": 10.42898223272055,
+      "grad_norm": 0.04818270727992058,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 101547
+    },
+    {
+      "epoch": 10.429290335832391,
+      "grad_norm": 0.07613711804151535,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 101550
+    },
+    {
+      "epoch": 10.429598438944232,
+      "grad_norm": 0.093401700258255,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 101553
+    },
+    {
+      "epoch": 10.429906542056075,
+      "grad_norm": 0.08218391984701157,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 101556
+    },
+    {
+      "epoch": 10.430214645167917,
+      "grad_norm": 0.09376956522464752,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 101559
+    },
+    {
+      "epoch": 10.430522748279758,
+      "grad_norm": 0.06790527701377869,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 101562
+    },
+    {
+      "epoch": 10.4308308513916,
+      "grad_norm": 0.04368535429239273,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 101565
+    },
+    {
+      "epoch": 10.43113895450344,
+      "grad_norm": 0.03572575002908707,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 101568
+    },
+    {
+      "epoch": 10.431447057615282,
+      "grad_norm": 0.10959416627883911,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 101571
+    },
+    {
+      "epoch": 10.431755160727123,
+      "grad_norm": 0.10492794215679169,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 101574
+    },
+    {
+      "epoch": 10.432063263838964,
+      "grad_norm": 0.034721072763204575,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 101577
+    },
+    {
+      "epoch": 10.432371366950806,
+      "grad_norm": 0.08810894191265106,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 101580
+    },
+    {
+      "epoch": 10.432679470062647,
+      "grad_norm": 0.058836497366428375,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 101583
+    },
+    {
+      "epoch": 10.43298757317449,
+      "grad_norm": 0.0684552937746048,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 101586
+    },
+    {
+      "epoch": 10.433295676286331,
+      "grad_norm": 0.04458843544125557,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 101589
+    },
+    {
+      "epoch": 10.433603779398172,
+      "grad_norm": 0.03802128881216049,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 101592
+    },
+    {
+      "epoch": 10.433911882510014,
+      "grad_norm": 0.04631553217768669,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 101595
+    },
+    {
+      "epoch": 10.434219985621855,
+      "grad_norm": 0.0350140780210495,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 101598
+    },
+    {
+      "epoch": 10.434528088733696,
+      "grad_norm": 0.06783036887645721,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 101601
+    },
+    {
+      "epoch": 10.434836191845537,
+      "grad_norm": 0.10419753193855286,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 101604
+    },
+    {
+      "epoch": 10.435144294957379,
+      "grad_norm": 0.057775016874074936,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 101607
+    },
+    {
+      "epoch": 10.43545239806922,
+      "grad_norm": 0.034822724759578705,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 101610
+    },
+    {
+      "epoch": 10.435760501181061,
+      "grad_norm": 0.1210988312959671,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 101613
+    },
+    {
+      "epoch": 10.436068604292903,
+      "grad_norm": 0.08582430332899094,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 101616
+    },
+    {
+      "epoch": 10.436376707404746,
+      "grad_norm": 0.037638887763023376,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 101619
+    },
+    {
+      "epoch": 10.436684810516587,
+      "grad_norm": 0.149775430560112,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 101622
+    },
+    {
+      "epoch": 10.436992913628428,
+      "grad_norm": 0.036241546273231506,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 101625
+    },
+    {
+      "epoch": 10.43730101674027,
+      "grad_norm": 0.051932793110609055,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 101628
+    },
+    {
+      "epoch": 10.43760911985211,
+      "grad_norm": 0.05552282929420471,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 101631
+    },
+    {
+      "epoch": 10.437917222963952,
+      "grad_norm": 0.05709502846002579,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 101634
+    },
+    {
+      "epoch": 10.438225326075793,
+      "grad_norm": 0.055031366646289825,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 101637
+    },
+    {
+      "epoch": 10.438533429187634,
+      "grad_norm": 0.04440178722143173,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 101640
+    },
+    {
+      "epoch": 10.438841532299476,
+      "grad_norm": 0.033032674342393875,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 101643
+    },
+    {
+      "epoch": 10.439149635411317,
+      "grad_norm": 0.06382546573877335,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 101646
+    },
+    {
+      "epoch": 10.43945773852316,
+      "grad_norm": 0.15189075469970703,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 101649
+    },
+    {
+      "epoch": 10.439765841635001,
+      "grad_norm": 0.035068389028310776,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 101652
+    },
+    {
+      "epoch": 10.440073944746842,
+      "grad_norm": 0.1286683976650238,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 101655
+    },
+    {
+      "epoch": 10.440382047858684,
+      "grad_norm": 0.12208466976881027,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 101658
+    },
+    {
+      "epoch": 10.440690150970525,
+      "grad_norm": 0.08962827175855637,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 101661
+    },
+    {
+      "epoch": 10.440998254082366,
+      "grad_norm": 0.05762847512960434,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 101664
+    },
+    {
+      "epoch": 10.441306357194208,
+      "grad_norm": 0.032505180686712265,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 101667
+    },
+    {
+      "epoch": 10.441614460306049,
+      "grad_norm": 0.04459118843078613,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 101670
+    },
+    {
+      "epoch": 10.44192256341789,
+      "grad_norm": 0.04378930851817131,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 101673
+    },
+    {
+      "epoch": 10.442230666529731,
+      "grad_norm": 0.036537352949380875,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 101676
+    },
+    {
+      "epoch": 10.442538769641573,
+      "grad_norm": 0.042318109422922134,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 101679
+    },
+    {
+      "epoch": 10.442846872753416,
+      "grad_norm": 0.08487118780612946,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 101682
+    },
+    {
+      "epoch": 10.443154975865257,
+      "grad_norm": 0.05887780711054802,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 101685
+    },
+    {
+      "epoch": 10.443463078977098,
+      "grad_norm": 0.09170451015233994,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 101688
+    },
+    {
+      "epoch": 10.44377118208894,
+      "grad_norm": 0.07705742865800858,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 101691
+    },
+    {
+      "epoch": 10.44407928520078,
+      "grad_norm": 0.0543515719473362,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 101694
+    },
+    {
+      "epoch": 10.444387388312622,
+      "grad_norm": 0.03692963346838951,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 101697
+    },
+    {
+      "epoch": 10.444695491424463,
+      "grad_norm": 0.04865596815943718,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 101700
+    },
+    {
+      "epoch": 10.445003594536304,
+      "grad_norm": 0.12262431532144547,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 101703
+    },
+    {
+      "epoch": 10.445311697648146,
+      "grad_norm": 0.04820642247796059,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 101706
+    },
+    {
+      "epoch": 10.445619800759987,
+      "grad_norm": 0.10539522767066956,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 101709
+    },
+    {
+      "epoch": 10.44592790387183,
+      "grad_norm": 0.0626974105834961,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 101712
+    },
+    {
+      "epoch": 10.446236006983671,
+      "grad_norm": 0.0897459164261818,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 101715
+    },
+    {
+      "epoch": 10.446544110095513,
+      "grad_norm": 0.049581412225961685,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 101718
+    },
+    {
+      "epoch": 10.446852213207354,
+      "grad_norm": 0.06107807531952858,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 101721
+    },
+    {
+      "epoch": 10.447160316319195,
+      "grad_norm": 0.051195114850997925,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 101724
+    },
+    {
+      "epoch": 10.447468419431036,
+      "grad_norm": 0.09900028258562088,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 101727
+    },
+    {
+      "epoch": 10.447776522542878,
+      "grad_norm": 0.07909564673900604,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 101730
+    },
+    {
+      "epoch": 10.448084625654719,
+      "grad_norm": 0.06978385150432587,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 101733
+    },
+    {
+      "epoch": 10.44839272876656,
+      "grad_norm": 0.07636460661888123,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 101736
+    },
+    {
+      "epoch": 10.448700831878401,
+      "grad_norm": 0.0727100670337677,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 101739
+    },
+    {
+      "epoch": 10.449008934990243,
+      "grad_norm": 0.06627271324396133,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 101742
+    },
+    {
+      "epoch": 10.449317038102086,
+      "grad_norm": 0.07906875014305115,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 101745
+    },
+    {
+      "epoch": 10.449625141213927,
+      "grad_norm": 0.08305513858795166,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 101748
+    },
+    {
+      "epoch": 10.449933244325768,
+      "grad_norm": 0.05429422855377197,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 101751
+    },
+    {
+      "epoch": 10.45024134743761,
+      "grad_norm": 0.10072019696235657,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 101754
+    },
+    {
+      "epoch": 10.45054945054945,
+      "grad_norm": 0.053927551954984665,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 101757
+    },
+    {
+      "epoch": 10.450857553661292,
+      "grad_norm": 0.08865250647068024,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 101760
+    },
+    {
+      "epoch": 10.451165656773133,
+      "grad_norm": 0.08605709671974182,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 101763
+    },
+    {
+      "epoch": 10.451473759884975,
+      "grad_norm": 0.06711055338382721,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 101766
+    },
+    {
+      "epoch": 10.451781862996816,
+      "grad_norm": 0.06331571936607361,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 101769
+    },
+    {
+      "epoch": 10.452089966108657,
+      "grad_norm": 0.08312960714101791,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 101772
+    },
+    {
+      "epoch": 10.452398069220498,
+      "grad_norm": 0.03663955256342888,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 101775
+    },
+    {
+      "epoch": 10.452706172332341,
+      "grad_norm": 0.036953091621398926,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 101778
+    },
+    {
+      "epoch": 10.453014275444183,
+      "grad_norm": 0.12208108603954315,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 101781
+    },
+    {
+      "epoch": 10.453322378556024,
+      "grad_norm": 0.06827596575021744,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 101784
+    },
+    {
+      "epoch": 10.453630481667865,
+      "grad_norm": 0.061575282365083694,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 101787
+    },
+    {
+      "epoch": 10.453938584779706,
+      "grad_norm": 0.04751850664615631,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 101790
+    },
+    {
+      "epoch": 10.454246687891548,
+      "grad_norm": 0.1076505035161972,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 101793
+    },
+    {
+      "epoch": 10.454554791003389,
+      "grad_norm": 0.04396897554397583,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 101796
+    },
+    {
+      "epoch": 10.45486289411523,
+      "grad_norm": 0.03534824028611183,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 101799
+    },
+    {
+      "epoch": 10.455170997227071,
+      "grad_norm": 0.05526803433895111,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 101802
+    },
+    {
+      "epoch": 10.455479100338913,
+      "grad_norm": 0.049421172589063644,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 101805
+    },
+    {
+      "epoch": 10.455787203450754,
+      "grad_norm": 0.06431230902671814,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 101808
+    },
+    {
+      "epoch": 10.456095306562597,
+      "grad_norm": 0.08008962869644165,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 101811
+    },
+    {
+      "epoch": 10.456403409674438,
+      "grad_norm": 0.05697495862841606,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 101814
+    },
+    {
+      "epoch": 10.45671151278628,
+      "grad_norm": 0.11419016867876053,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 101817
+    },
+    {
+      "epoch": 10.45701961589812,
+      "grad_norm": 0.07900203764438629,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 101820
+    },
+    {
+      "epoch": 10.457327719009962,
+      "grad_norm": 0.08858697861433029,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 101823
+    },
+    {
+      "epoch": 10.457635822121803,
+      "grad_norm": 0.051916588097810745,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 101826
+    },
+    {
+      "epoch": 10.457943925233645,
+      "grad_norm": 0.03870150074362755,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 101829
+    },
+    {
+      "epoch": 10.458252028345486,
+      "grad_norm": 0.03717373311519623,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 101832
+    },
+    {
+      "epoch": 10.458560131457327,
+      "grad_norm": 0.12929514050483704,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 101835
+    },
+    {
+      "epoch": 10.458868234569168,
+      "grad_norm": 0.11549792438745499,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 101838
+    },
+    {
+      "epoch": 10.459176337681011,
+      "grad_norm": 0.08777003735303879,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 101841
+    },
+    {
+      "epoch": 10.459484440792853,
+      "grad_norm": 0.063051737844944,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 101844
+    },
+    {
+      "epoch": 10.459792543904694,
+      "grad_norm": 0.04587935656309128,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 101847
+    },
+    {
+      "epoch": 10.460100647016535,
+      "grad_norm": 0.09982280433177948,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 101850
+    },
+    {
+      "epoch": 10.460408750128376,
+      "grad_norm": 0.09921959787607193,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 101853
+    },
+    {
+      "epoch": 10.460716853240218,
+      "grad_norm": 0.038731105625629425,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 101856
+    },
+    {
+      "epoch": 10.461024956352059,
+      "grad_norm": 0.07030004262924194,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 101859
+    },
+    {
+      "epoch": 10.4613330594639,
+      "grad_norm": 0.052664730697870255,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 101862
+    },
+    {
+      "epoch": 10.461641162575742,
+      "grad_norm": 0.11606428027153015,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 101865
+    },
+    {
+      "epoch": 10.461949265687583,
+      "grad_norm": 0.05401783809065819,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 101868
+    },
+    {
+      "epoch": 10.462257368799424,
+      "grad_norm": 0.03583219647407532,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 101871
+    },
+    {
+      "epoch": 10.462565471911267,
+      "grad_norm": 0.07720806449651718,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 101874
+    },
+    {
+      "epoch": 10.462873575023108,
+      "grad_norm": 0.054585494101047516,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 101877
+    },
+    {
+      "epoch": 10.46318167813495,
+      "grad_norm": 0.06677903980016708,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 101880
+    },
+    {
+      "epoch": 10.46348978124679,
+      "grad_norm": 0.03949226066470146,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 101883
+    },
+    {
+      "epoch": 10.463797884358632,
+      "grad_norm": 0.03900604695081711,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 101886
+    },
+    {
+      "epoch": 10.464105987470473,
+      "grad_norm": 0.08953657001256943,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 101889
+    },
+    {
+      "epoch": 10.464414090582315,
+      "grad_norm": 0.09946227818727493,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 101892
+    },
+    {
+      "epoch": 10.464722193694156,
+      "grad_norm": 0.036433037370443344,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 101895
+    },
+    {
+      "epoch": 10.465030296805997,
+      "grad_norm": 0.09698110073804855,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 101898
+    },
+    {
+      "epoch": 10.465338399917838,
+      "grad_norm": 0.04699701443314552,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 101901
+    },
+    {
+      "epoch": 10.465646503029681,
+      "grad_norm": 0.03788859769701958,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 101904
+    },
+    {
+      "epoch": 10.465954606141523,
+      "grad_norm": 0.09247510880231857,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 101907
+    },
+    {
+      "epoch": 10.466262709253364,
+      "grad_norm": 0.09411049634218216,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 101910
+    },
+    {
+      "epoch": 10.466570812365205,
+      "grad_norm": 0.12619060277938843,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 101913
+    },
+    {
+      "epoch": 10.466878915477047,
+      "grad_norm": 0.17733120918273926,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 101916
+    },
+    {
+      "epoch": 10.467187018588888,
+      "grad_norm": 0.10386303067207336,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 101919
+    },
+    {
+      "epoch": 10.467495121700729,
+      "grad_norm": 0.035836637020111084,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 101922
+    },
+    {
+      "epoch": 10.46780322481257,
+      "grad_norm": 0.050569795072078705,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 101925
+    },
+    {
+      "epoch": 10.468111327924412,
+      "grad_norm": 0.04953821375966072,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 101928
+    },
+    {
+      "epoch": 10.468419431036253,
+      "grad_norm": 0.05620476230978966,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 101931
+    },
+    {
+      "epoch": 10.468727534148094,
+      "grad_norm": 0.05553643777966499,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 101934
+    },
+    {
+      "epoch": 10.469035637259937,
+      "grad_norm": 0.04494090378284454,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 101937
+    },
+    {
+      "epoch": 10.469343740371778,
+      "grad_norm": 0.09270351380109787,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 101940
+    },
+    {
+      "epoch": 10.46965184348362,
+      "grad_norm": 0.04817723482847214,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 101943
+    },
+    {
+      "epoch": 10.469959946595461,
+      "grad_norm": 0.09144774824380875,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 101946
+    },
+    {
+      "epoch": 10.470268049707302,
+      "grad_norm": 0.048304855823516846,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 101949
+    },
+    {
+      "epoch": 10.470576152819143,
+      "grad_norm": 0.08919715136289597,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 101952
+    },
+    {
+      "epoch": 10.470884255930985,
+      "grad_norm": 0.07856735587120056,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 101955
+    },
+    {
+      "epoch": 10.471192359042826,
+      "grad_norm": 0.06114446744322777,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 101958
+    },
+    {
+      "epoch": 10.471500462154667,
+      "grad_norm": 0.04188235104084015,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 101961
+    },
+    {
+      "epoch": 10.471808565266508,
+      "grad_norm": 0.04990050569176674,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 101964
+    },
+    {
+      "epoch": 10.472116668378352,
+      "grad_norm": 0.08834641426801682,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 101967
+    },
+    {
+      "epoch": 10.472424771490193,
+      "grad_norm": 0.09818597882986069,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 101970
+    },
+    {
+      "epoch": 10.472732874602034,
+      "grad_norm": 0.048872750252485275,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 101973
+    },
+    {
+      "epoch": 10.473040977713875,
+      "grad_norm": 0.03154008090496063,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 101976
+    },
+    {
+      "epoch": 10.473349080825717,
+      "grad_norm": 0.0439290851354599,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 101979
+    },
+    {
+      "epoch": 10.473657183937558,
+      "grad_norm": 0.0607743076980114,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 101982
+    },
+    {
+      "epoch": 10.473965287049399,
+      "grad_norm": 0.06029101461172104,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 101985
+    },
+    {
+      "epoch": 10.47427339016124,
+      "grad_norm": 0.06238150969147682,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 101988
+    },
+    {
+      "epoch": 10.474581493273082,
+      "grad_norm": 0.09172660112380981,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 101991
+    },
+    {
+      "epoch": 10.474889596384923,
+      "grad_norm": 0.04956614598631859,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 101994
+    },
+    {
+      "epoch": 10.475197699496764,
+      "grad_norm": 0.08379212021827698,
+      "learning_rate": 0.01,
+      "loss": 1.9786,
+      "step": 101997
+    },
+    {
+      "epoch": 10.475505802608607,
+      "grad_norm": 0.04099489748477936,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 102000
+    },
+    {
+      "epoch": 10.475813905720448,
+      "grad_norm": 0.0487617626786232,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 102003
+    },
+    {
+      "epoch": 10.47612200883229,
+      "grad_norm": 0.039811212569475174,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 102006
+    },
+    {
+      "epoch": 10.476430111944131,
+      "grad_norm": 0.044633690267801285,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 102009
+    },
+    {
+      "epoch": 10.476738215055972,
+      "grad_norm": 0.05821068957448006,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 102012
+    },
+    {
+      "epoch": 10.477046318167814,
+      "grad_norm": 0.05007709562778473,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 102015
+    },
+    {
+      "epoch": 10.477354421279655,
+      "grad_norm": 0.04428219422698021,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 102018
+    },
+    {
+      "epoch": 10.477662524391496,
+      "grad_norm": 0.0576540008187294,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 102021
+    },
+    {
+      "epoch": 10.477970627503337,
+      "grad_norm": 0.032921202480793,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 102024
+    },
+    {
+      "epoch": 10.478278730615179,
+      "grad_norm": 0.06370113790035248,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 102027
+    },
+    {
+      "epoch": 10.47858683372702,
+      "grad_norm": 0.11116142570972443,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 102030
+    },
+    {
+      "epoch": 10.478894936838863,
+      "grad_norm": 0.10910597443580627,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 102033
+    },
+    {
+      "epoch": 10.479203039950704,
+      "grad_norm": 0.06832683831453323,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 102036
+    },
+    {
+      "epoch": 10.479511143062545,
+      "grad_norm": 0.036917220801115036,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 102039
+    },
+    {
+      "epoch": 10.479819246174387,
+      "grad_norm": 0.0481659471988678,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 102042
+    },
+    {
+      "epoch": 10.480127349286228,
+      "grad_norm": 0.043729156255722046,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 102045
+    },
+    {
+      "epoch": 10.48043545239807,
+      "grad_norm": 0.046008024364709854,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 102048
+    },
+    {
+      "epoch": 10.48074355550991,
+      "grad_norm": 0.09844762831926346,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 102051
+    },
+    {
+      "epoch": 10.481051658621752,
+      "grad_norm": 0.0938354879617691,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 102054
+    },
+    {
+      "epoch": 10.481359761733593,
+      "grad_norm": 0.07487953454256058,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 102057
+    },
+    {
+      "epoch": 10.481667864845434,
+      "grad_norm": 0.06973642110824585,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 102060
+    },
+    {
+      "epoch": 10.481975967957275,
+      "grad_norm": 0.04177610203623772,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 102063
+    },
+    {
+      "epoch": 10.482284071069119,
+      "grad_norm": 0.042998723685741425,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 102066
+    },
+    {
+      "epoch": 10.48259217418096,
+      "grad_norm": 0.04426990821957588,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 102069
+    },
+    {
+      "epoch": 10.482900277292801,
+      "grad_norm": 0.10405926406383514,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 102072
+    },
+    {
+      "epoch": 10.483208380404642,
+      "grad_norm": 0.06819596141576767,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 102075
+    },
+    {
+      "epoch": 10.483516483516484,
+      "grad_norm": 0.07998234778642654,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 102078
+    },
+    {
+      "epoch": 10.483824586628325,
+      "grad_norm": 0.06735380738973618,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 102081
+    },
+    {
+      "epoch": 10.484132689740166,
+      "grad_norm": 0.1081424206495285,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 102084
+    },
+    {
+      "epoch": 10.484440792852007,
+      "grad_norm": 0.05134006217122078,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 102087
+    },
+    {
+      "epoch": 10.484748895963849,
+      "grad_norm": 0.0856146588921547,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 102090
+    },
+    {
+      "epoch": 10.48505699907569,
+      "grad_norm": 0.19568829238414764,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 102093
+    },
+    {
+      "epoch": 10.485365102187533,
+      "grad_norm": 0.1644512414932251,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 102096
+    },
+    {
+      "epoch": 10.485673205299374,
+      "grad_norm": 0.1425795555114746,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 102099
+    },
+    {
+      "epoch": 10.485981308411215,
+      "grad_norm": 0.07443909347057343,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 102102
+    },
+    {
+      "epoch": 10.486289411523057,
+      "grad_norm": 0.09227898716926575,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 102105
+    },
+    {
+      "epoch": 10.486597514634898,
+      "grad_norm": 0.07184717059135437,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 102108
+    },
+    {
+      "epoch": 10.48690561774674,
+      "grad_norm": 0.06727568805217743,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 102111
+    },
+    {
+      "epoch": 10.48721372085858,
+      "grad_norm": 0.06257406622171402,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 102114
+    },
+    {
+      "epoch": 10.487521823970422,
+      "grad_norm": 0.057932376861572266,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 102117
+    },
+    {
+      "epoch": 10.487829927082263,
+      "grad_norm": 0.04455142468214035,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 102120
+    },
+    {
+      "epoch": 10.488138030194104,
+      "grad_norm": 0.0342220775783062,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 102123
+    },
+    {
+      "epoch": 10.488446133305946,
+      "grad_norm": 0.08886317163705826,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 102126
+    },
+    {
+      "epoch": 10.488754236417789,
+      "grad_norm": 0.05419706553220749,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 102129
+    },
+    {
+      "epoch": 10.48906233952963,
+      "grad_norm": 0.10033708810806274,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 102132
+    },
+    {
+      "epoch": 10.489370442641471,
+      "grad_norm": 0.12334150075912476,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 102135
+    },
+    {
+      "epoch": 10.489678545753312,
+      "grad_norm": 0.061110034584999084,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 102138
+    },
+    {
+      "epoch": 10.489986648865154,
+      "grad_norm": 0.10176794975996017,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 102141
+    },
+    {
+      "epoch": 10.490294751976995,
+      "grad_norm": 0.07273289561271667,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 102144
+    },
+    {
+      "epoch": 10.490602855088836,
+      "grad_norm": 0.055310558527708054,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 102147
+    },
+    {
+      "epoch": 10.490910958200677,
+      "grad_norm": 0.10140859335660934,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 102150
+    },
+    {
+      "epoch": 10.491219061312519,
+      "grad_norm": 0.051482249051332474,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 102153
+    },
+    {
+      "epoch": 10.49152716442436,
+      "grad_norm": 0.08374358713626862,
+      "learning_rate": 0.01,
+      "loss": 1.9065,
+      "step": 102156
+    },
+    {
+      "epoch": 10.491835267536203,
+      "grad_norm": 0.050834473222494125,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 102159
+    },
+    {
+      "epoch": 10.492143370648044,
+      "grad_norm": 0.10886992514133453,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 102162
+    },
+    {
+      "epoch": 10.492451473759886,
+      "grad_norm": 0.06643300503492355,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 102165
+    },
+    {
+      "epoch": 10.492759576871727,
+      "grad_norm": 0.050651658326387405,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 102168
+    },
+    {
+      "epoch": 10.493067679983568,
+      "grad_norm": 0.04459983855485916,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 102171
+    },
+    {
+      "epoch": 10.49337578309541,
+      "grad_norm": 0.0398244671523571,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 102174
+    },
+    {
+      "epoch": 10.49368388620725,
+      "grad_norm": 0.06146949157118797,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 102177
+    },
+    {
+      "epoch": 10.493991989319092,
+      "grad_norm": 0.06933523714542389,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 102180
+    },
+    {
+      "epoch": 10.494300092430933,
+      "grad_norm": 0.08496151119470596,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 102183
+    },
+    {
+      "epoch": 10.494608195542774,
+      "grad_norm": 0.06656196713447571,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 102186
+    },
+    {
+      "epoch": 10.494916298654616,
+      "grad_norm": 0.057561956346035004,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 102189
+    },
+    {
+      "epoch": 10.495224401766459,
+      "grad_norm": 0.03765184432268143,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 102192
+    },
+    {
+      "epoch": 10.4955325048783,
+      "grad_norm": 0.055984582751989365,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 102195
+    },
+    {
+      "epoch": 10.495840607990141,
+      "grad_norm": 0.06702415645122528,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 102198
+    },
+    {
+      "epoch": 10.496148711101982,
+      "grad_norm": 0.04814969003200531,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 102201
+    },
+    {
+      "epoch": 10.496456814213824,
+      "grad_norm": 0.03827301785349846,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 102204
+    },
+    {
+      "epoch": 10.496764917325665,
+      "grad_norm": 0.04301337152719498,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 102207
+    },
+    {
+      "epoch": 10.497073020437506,
+      "grad_norm": 0.04337724670767784,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 102210
+    },
+    {
+      "epoch": 10.497381123549347,
+      "grad_norm": 0.05614280700683594,
+      "learning_rate": 0.01,
+      "loss": 1.9709,
+      "step": 102213
+    },
+    {
+      "epoch": 10.497689226661189,
+      "grad_norm": 0.10362491011619568,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 102216
+    },
+    {
+      "epoch": 10.49799732977303,
+      "grad_norm": 0.05749976262450218,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 102219
+    },
+    {
+      "epoch": 10.498305432884871,
+      "grad_norm": 0.05132078006863594,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 102222
+    },
+    {
+      "epoch": 10.498613535996714,
+      "grad_norm": 0.08391030132770538,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 102225
+    },
+    {
+      "epoch": 10.498921639108556,
+      "grad_norm": 0.08488337695598602,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 102228
+    },
+    {
+      "epoch": 10.499229742220397,
+      "grad_norm": 0.07492703944444656,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 102231
+    },
+    {
+      "epoch": 10.499537845332238,
+      "grad_norm": 0.0787576213479042,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 102234
+    },
+    {
+      "epoch": 10.49984594844408,
+      "grad_norm": 0.0703982412815094,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 102237
+    },
+    {
+      "epoch": 10.50015405155592,
+      "grad_norm": 0.08891353011131287,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 102240
+    },
+    {
+      "epoch": 10.500462154667762,
+      "grad_norm": 0.12258651852607727,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 102243
+    },
+    {
+      "epoch": 10.500770257779603,
+      "grad_norm": 0.04051290079951286,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 102246
+    },
+    {
+      "epoch": 10.501078360891444,
+      "grad_norm": 0.07038848102092743,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 102249
+    },
+    {
+      "epoch": 10.501386464003286,
+      "grad_norm": 0.059617362916469574,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 102252
+    },
+    {
+      "epoch": 10.501694567115127,
+      "grad_norm": 0.06750741600990295,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 102255
+    },
+    {
+      "epoch": 10.50200267022697,
+      "grad_norm": 0.03963152691721916,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 102258
+    },
+    {
+      "epoch": 10.502310773338811,
+      "grad_norm": 0.05788528546690941,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 102261
+    },
+    {
+      "epoch": 10.502618876450653,
+      "grad_norm": 0.0345892570912838,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 102264
+    },
+    {
+      "epoch": 10.502926979562494,
+      "grad_norm": 0.04929973930120468,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 102267
+    },
+    {
+      "epoch": 10.503235082674335,
+      "grad_norm": 0.03237371891736984,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 102270
+    },
+    {
+      "epoch": 10.503543185786176,
+      "grad_norm": 0.12326578050851822,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 102273
+    },
+    {
+      "epoch": 10.503851288898018,
+      "grad_norm": 0.050269100815057755,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 102276
+    },
+    {
+      "epoch": 10.504159392009859,
+      "grad_norm": 0.08985825628042221,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 102279
+    },
+    {
+      "epoch": 10.5044674951217,
+      "grad_norm": 0.07561146467924118,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 102282
+    },
+    {
+      "epoch": 10.504775598233541,
+      "grad_norm": 0.10341837257146835,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 102285
+    },
+    {
+      "epoch": 10.505083701345384,
+      "grad_norm": 0.06874041259288788,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 102288
+    },
+    {
+      "epoch": 10.505391804457226,
+      "grad_norm": 0.05333573743700981,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 102291
+    },
+    {
+      "epoch": 10.505699907569067,
+      "grad_norm": 0.05619170144200325,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 102294
+    },
+    {
+      "epoch": 10.506008010680908,
+      "grad_norm": 0.06462277472019196,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 102297
+    },
+    {
+      "epoch": 10.50631611379275,
+      "grad_norm": 0.05119762942194939,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 102300
+    },
+    {
+      "epoch": 10.50662421690459,
+      "grad_norm": 0.10214772075414658,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 102303
+    },
+    {
+      "epoch": 10.506932320016432,
+      "grad_norm": 0.06918077915906906,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 102306
+    },
+    {
+      "epoch": 10.507240423128273,
+      "grad_norm": 0.06975413113832474,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 102309
+    },
+    {
+      "epoch": 10.507548526240114,
+      "grad_norm": 0.06599710136651993,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 102312
+    },
+    {
+      "epoch": 10.507856629351956,
+      "grad_norm": 0.08133608102798462,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 102315
+    },
+    {
+      "epoch": 10.508164732463797,
+      "grad_norm": 0.0877838209271431,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 102318
+    },
+    {
+      "epoch": 10.50847283557564,
+      "grad_norm": 0.09431187063455582,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 102321
+    },
+    {
+      "epoch": 10.508780938687481,
+      "grad_norm": 0.04923112317919731,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 102324
+    },
+    {
+      "epoch": 10.509089041799323,
+      "grad_norm": 0.10650596022605896,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 102327
+    },
+    {
+      "epoch": 10.509397144911164,
+      "grad_norm": 0.03952709957957268,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 102330
+    },
+    {
+      "epoch": 10.509705248023005,
+      "grad_norm": 0.05486638471484184,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 102333
+    },
+    {
+      "epoch": 10.510013351134846,
+      "grad_norm": 0.10510875284671783,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 102336
+    },
+    {
+      "epoch": 10.510321454246688,
+      "grad_norm": 0.08444318920373917,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 102339
+    },
+    {
+      "epoch": 10.510629557358529,
+      "grad_norm": 0.1074652373790741,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 102342
+    },
+    {
+      "epoch": 10.51093766047037,
+      "grad_norm": 0.08538160473108292,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 102345
+    },
+    {
+      "epoch": 10.511245763582211,
+      "grad_norm": 0.04091689735651016,
+      "learning_rate": 0.01,
+      "loss": 1.8937,
+      "step": 102348
+    },
+    {
+      "epoch": 10.511553866694054,
+      "grad_norm": 0.05738305673003197,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 102351
+    },
+    {
+      "epoch": 10.511861969805896,
+      "grad_norm": 0.0682215765118599,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 102354
+    },
+    {
+      "epoch": 10.512170072917737,
+      "grad_norm": 0.05720492824912071,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 102357
+    },
+    {
+      "epoch": 10.512478176029578,
+      "grad_norm": 0.04315927252173424,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 102360
+    },
+    {
+      "epoch": 10.51278627914142,
+      "grad_norm": 0.049331653863191605,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 102363
+    },
+    {
+      "epoch": 10.51309438225326,
+      "grad_norm": 0.03660670667886734,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 102366
+    },
+    {
+      "epoch": 10.513402485365102,
+      "grad_norm": 0.04299522191286087,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 102369
+    },
+    {
+      "epoch": 10.513710588476943,
+      "grad_norm": 0.1256340593099594,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 102372
+    },
+    {
+      "epoch": 10.514018691588785,
+      "grad_norm": 0.13235409557819366,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 102375
+    },
+    {
+      "epoch": 10.514326794700626,
+      "grad_norm": 0.05709142982959747,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 102378
+    },
+    {
+      "epoch": 10.514634897812467,
+      "grad_norm": 0.07259286940097809,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 102381
+    },
+    {
+      "epoch": 10.51494300092431,
+      "grad_norm": 0.047436416149139404,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 102384
+    },
+    {
+      "epoch": 10.515251104036151,
+      "grad_norm": 0.03879249840974808,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 102387
+    },
+    {
+      "epoch": 10.515559207147993,
+      "grad_norm": 0.050321247428655624,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 102390
+    },
+    {
+      "epoch": 10.515867310259834,
+      "grad_norm": 0.06300079822540283,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 102393
+    },
+    {
+      "epoch": 10.516175413371675,
+      "grad_norm": 0.0426388643682003,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 102396
+    },
+    {
+      "epoch": 10.516483516483516,
+      "grad_norm": 0.04434233158826828,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 102399
+    },
+    {
+      "epoch": 10.516791619595358,
+      "grad_norm": 0.06005145609378815,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 102402
+    },
+    {
+      "epoch": 10.517099722707199,
+      "grad_norm": 0.10316505283117294,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 102405
+    },
+    {
+      "epoch": 10.51740782581904,
+      "grad_norm": 0.08546032756567001,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 102408
+    },
+    {
+      "epoch": 10.517715928930881,
+      "grad_norm": 0.06833061575889587,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 102411
+    },
+    {
+      "epoch": 10.518024032042725,
+      "grad_norm": 0.07763389497995377,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 102414
+    },
+    {
+      "epoch": 10.518332135154566,
+      "grad_norm": 0.13648200035095215,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 102417
+    },
+    {
+      "epoch": 10.518640238266407,
+      "grad_norm": 0.0665363073348999,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 102420
+    },
+    {
+      "epoch": 10.518948341378248,
+      "grad_norm": 0.07532433420419693,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 102423
+    },
+    {
+      "epoch": 10.51925644449009,
+      "grad_norm": 0.05814901739358902,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 102426
+    },
+    {
+      "epoch": 10.51956454760193,
+      "grad_norm": 0.06840399652719498,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 102429
+    },
+    {
+      "epoch": 10.519872650713772,
+      "grad_norm": 0.059958960860967636,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 102432
+    },
+    {
+      "epoch": 10.520180753825613,
+      "grad_norm": 0.041250091046094894,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 102435
+    },
+    {
+      "epoch": 10.520488856937455,
+      "grad_norm": 0.03929082676768303,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 102438
+    },
+    {
+      "epoch": 10.520796960049296,
+      "grad_norm": 0.09511935710906982,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 102441
+    },
+    {
+      "epoch": 10.521105063161137,
+      "grad_norm": 0.1492692083120346,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 102444
+    },
+    {
+      "epoch": 10.52141316627298,
+      "grad_norm": 0.12860162556171417,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 102447
+    },
+    {
+      "epoch": 10.521721269384821,
+      "grad_norm": 0.06535381823778152,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 102450
+    },
+    {
+      "epoch": 10.522029372496663,
+      "grad_norm": 0.03897741809487343,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 102453
+    },
+    {
+      "epoch": 10.522337475608504,
+      "grad_norm": 0.03247740864753723,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 102456
+    },
+    {
+      "epoch": 10.522645578720345,
+      "grad_norm": 0.05719994381070137,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 102459
+    },
+    {
+      "epoch": 10.522953681832186,
+      "grad_norm": 0.0653923973441124,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 102462
+    },
+    {
+      "epoch": 10.523261784944028,
+      "grad_norm": 0.051299069076776505,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 102465
+    },
+    {
+      "epoch": 10.523569888055869,
+      "grad_norm": 0.04528390243649483,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 102468
+    },
+    {
+      "epoch": 10.52387799116771,
+      "grad_norm": 0.03322979062795639,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 102471
+    },
+    {
+      "epoch": 10.524186094279552,
+      "grad_norm": 0.05223941057920456,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 102474
+    },
+    {
+      "epoch": 10.524494197391395,
+      "grad_norm": 0.04564807564020157,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 102477
+    },
+    {
+      "epoch": 10.524802300503236,
+      "grad_norm": 0.07500727474689484,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 102480
+    },
+    {
+      "epoch": 10.525110403615077,
+      "grad_norm": 0.12562263011932373,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 102483
+    },
+    {
+      "epoch": 10.525418506726918,
+      "grad_norm": 0.09475285559892654,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 102486
+    },
+    {
+      "epoch": 10.52572660983876,
+      "grad_norm": 0.10476790368556976,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 102489
+    },
+    {
+      "epoch": 10.526034712950601,
+      "grad_norm": 0.03594356030225754,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 102492
+    },
+    {
+      "epoch": 10.526342816062442,
+      "grad_norm": 0.03649017587304115,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 102495
+    },
+    {
+      "epoch": 10.526650919174283,
+      "grad_norm": 0.10896097868680954,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 102498
+    },
+    {
+      "epoch": 10.526959022286125,
+      "grad_norm": 0.06398046016693115,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 102501
+    },
+    {
+      "epoch": 10.527267125397966,
+      "grad_norm": 0.03608668968081474,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 102504
+    },
+    {
+      "epoch": 10.527575228509807,
+      "grad_norm": 0.03560887649655342,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 102507
+    },
+    {
+      "epoch": 10.527883331621648,
+      "grad_norm": 0.07564496248960495,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 102510
+    },
+    {
+      "epoch": 10.528191434733492,
+      "grad_norm": 0.07473336160182953,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 102513
+    },
+    {
+      "epoch": 10.528499537845333,
+      "grad_norm": 0.09763288497924805,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 102516
+    },
+    {
+      "epoch": 10.528807640957174,
+      "grad_norm": 0.09780536592006683,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 102519
+    },
+    {
+      "epoch": 10.529115744069015,
+      "grad_norm": 0.11341667920351028,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 102522
+    },
+    {
+      "epoch": 10.529423847180857,
+      "grad_norm": 0.04304501786828041,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 102525
+    },
+    {
+      "epoch": 10.529731950292698,
+      "grad_norm": 0.038570135831832886,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 102528
+    },
+    {
+      "epoch": 10.530040053404539,
+      "grad_norm": 0.08291329443454742,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 102531
+    },
+    {
+      "epoch": 10.53034815651638,
+      "grad_norm": 0.06479398161172867,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 102534
+    },
+    {
+      "epoch": 10.530656259628222,
+      "grad_norm": 0.10525523871183395,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 102537
+    },
+    {
+      "epoch": 10.530964362740063,
+      "grad_norm": 0.06369463354349136,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 102540
+    },
+    {
+      "epoch": 10.531272465851906,
+      "grad_norm": 0.04059941694140434,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 102543
+    },
+    {
+      "epoch": 10.531580568963747,
+      "grad_norm": 0.07877116650342941,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 102546
+    },
+    {
+      "epoch": 10.531888672075588,
+      "grad_norm": 0.06322876363992691,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 102549
+    },
+    {
+      "epoch": 10.53219677518743,
+      "grad_norm": 0.09018049389123917,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 102552
+    },
+    {
+      "epoch": 10.532504878299271,
+      "grad_norm": 0.05731950327754021,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 102555
+    },
+    {
+      "epoch": 10.532812981411112,
+      "grad_norm": 0.055027980357408524,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 102558
+    },
+    {
+      "epoch": 10.533121084522953,
+      "grad_norm": 0.044410835951566696,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 102561
+    },
+    {
+      "epoch": 10.533429187634795,
+      "grad_norm": 0.07688567042350769,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 102564
+    },
+    {
+      "epoch": 10.533737290746636,
+      "grad_norm": 0.11015530675649643,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 102567
+    },
+    {
+      "epoch": 10.534045393858477,
+      "grad_norm": 0.06602126359939575,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 102570
+    },
+    {
+      "epoch": 10.534353496970319,
+      "grad_norm": 0.04021180421113968,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 102573
+    },
+    {
+      "epoch": 10.534661600082162,
+      "grad_norm": 0.039868198335170746,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 102576
+    },
+    {
+      "epoch": 10.534969703194003,
+      "grad_norm": 0.06292813271284103,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 102579
+    },
+    {
+      "epoch": 10.535277806305844,
+      "grad_norm": 0.034579694271087646,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 102582
+    },
+    {
+      "epoch": 10.535585909417685,
+      "grad_norm": 0.036958981305360794,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 102585
+    },
+    {
+      "epoch": 10.535894012529527,
+      "grad_norm": 0.04140496253967285,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 102588
+    },
+    {
+      "epoch": 10.536202115641368,
+      "grad_norm": 0.12174369394779205,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 102591
+    },
+    {
+      "epoch": 10.53651021875321,
+      "grad_norm": 0.033348649740219116,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 102594
+    },
+    {
+      "epoch": 10.53681832186505,
+      "grad_norm": 0.03767925873398781,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 102597
+    },
+    {
+      "epoch": 10.537126424976892,
+      "grad_norm": 0.044103484600782394,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 102600
+    },
+    {
+      "epoch": 10.537434528088733,
+      "grad_norm": 0.05621744319796562,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 102603
+    },
+    {
+      "epoch": 10.537742631200576,
+      "grad_norm": 0.05371059104800224,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 102606
+    },
+    {
+      "epoch": 10.538050734312417,
+      "grad_norm": 0.07096249610185623,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 102609
+    },
+    {
+      "epoch": 10.538358837424258,
+      "grad_norm": 0.04149243235588074,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 102612
+    },
+    {
+      "epoch": 10.5386669405361,
+      "grad_norm": 0.04006730020046234,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 102615
+    },
+    {
+      "epoch": 10.538975043647941,
+      "grad_norm": 0.07221797108650208,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 102618
+    },
+    {
+      "epoch": 10.539283146759782,
+      "grad_norm": 0.20590007305145264,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 102621
+    },
+    {
+      "epoch": 10.539591249871624,
+      "grad_norm": 0.08784258365631104,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 102624
+    },
+    {
+      "epoch": 10.539899352983465,
+      "grad_norm": 0.07793600857257843,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 102627
+    },
+    {
+      "epoch": 10.540207456095306,
+      "grad_norm": 0.08263092488050461,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 102630
+    },
+    {
+      "epoch": 10.540515559207147,
+      "grad_norm": 0.051660630851984024,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 102633
+    },
+    {
+      "epoch": 10.540823662318989,
+      "grad_norm": 0.03541038930416107,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 102636
+    },
+    {
+      "epoch": 10.541131765430832,
+      "grad_norm": 0.04703596606850624,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 102639
+    },
+    {
+      "epoch": 10.541439868542673,
+      "grad_norm": 0.04202953726053238,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 102642
+    },
+    {
+      "epoch": 10.541747971654514,
+      "grad_norm": 0.05148185044527054,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 102645
+    },
+    {
+      "epoch": 10.542056074766355,
+      "grad_norm": 0.04630628600716591,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 102648
+    },
+    {
+      "epoch": 10.542364177878197,
+      "grad_norm": 0.058917704969644547,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 102651
+    },
+    {
+      "epoch": 10.542672280990038,
+      "grad_norm": 0.050363462418317795,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 102654
+    },
+    {
+      "epoch": 10.54298038410188,
+      "grad_norm": 0.07656794041395187,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 102657
+    },
+    {
+      "epoch": 10.54328848721372,
+      "grad_norm": 0.09826540946960449,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 102660
+    },
+    {
+      "epoch": 10.543596590325562,
+      "grad_norm": 0.09988114982843399,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 102663
+    },
+    {
+      "epoch": 10.543904693437403,
+      "grad_norm": 0.09750048071146011,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 102666
+    },
+    {
+      "epoch": 10.544212796549246,
+      "grad_norm": 0.0635206550359726,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 102669
+    },
+    {
+      "epoch": 10.544520899661087,
+      "grad_norm": 0.05383175611495972,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 102672
+    },
+    {
+      "epoch": 10.544829002772929,
+      "grad_norm": 0.037810489535331726,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 102675
+    },
+    {
+      "epoch": 10.54513710588477,
+      "grad_norm": 0.04186122491955757,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 102678
+    },
+    {
+      "epoch": 10.545445208996611,
+      "grad_norm": 0.05252018943428993,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 102681
+    },
+    {
+      "epoch": 10.545753312108452,
+      "grad_norm": 0.13041871786117554,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 102684
+    },
+    {
+      "epoch": 10.546061415220294,
+      "grad_norm": 0.11638039350509644,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 102687
+    },
+    {
+      "epoch": 10.546369518332135,
+      "grad_norm": 0.052283596247434616,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 102690
+    },
+    {
+      "epoch": 10.546677621443976,
+      "grad_norm": 0.05509389564394951,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 102693
+    },
+    {
+      "epoch": 10.546985724555817,
+      "grad_norm": 0.08381091803312302,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 102696
+    },
+    {
+      "epoch": 10.547293827667659,
+      "grad_norm": 0.05065052583813667,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 102699
+    },
+    {
+      "epoch": 10.547601930779502,
+      "grad_norm": 0.035547565668821335,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 102702
+    },
+    {
+      "epoch": 10.547910033891343,
+      "grad_norm": 0.051477137953042984,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 102705
+    },
+    {
+      "epoch": 10.548218137003184,
+      "grad_norm": 0.07911818474531174,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 102708
+    },
+    {
+      "epoch": 10.548526240115025,
+      "grad_norm": 0.0840601772069931,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 102711
+    },
+    {
+      "epoch": 10.548834343226867,
+      "grad_norm": 0.05134335905313492,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 102714
+    },
+    {
+      "epoch": 10.549142446338708,
+      "grad_norm": 0.039568942040205,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 102717
+    },
+    {
+      "epoch": 10.54945054945055,
+      "grad_norm": 0.13316182792186737,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 102720
+    },
+    {
+      "epoch": 10.54975865256239,
+      "grad_norm": 0.04146268963813782,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 102723
+    },
+    {
+      "epoch": 10.550066755674232,
+      "grad_norm": 0.06591855734586716,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 102726
+    },
+    {
+      "epoch": 10.550374858786073,
+      "grad_norm": 0.05796660855412483,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 102729
+    },
+    {
+      "epoch": 10.550682961897916,
+      "grad_norm": 0.08629636466503143,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 102732
+    },
+    {
+      "epoch": 10.550991065009757,
+      "grad_norm": 0.0613933801651001,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 102735
+    },
+    {
+      "epoch": 10.551299168121599,
+      "grad_norm": 0.038255706429481506,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 102738
+    },
+    {
+      "epoch": 10.55160727123344,
+      "grad_norm": 0.0523841418325901,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 102741
+    },
+    {
+      "epoch": 10.551915374345281,
+      "grad_norm": 0.05940907076001167,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 102744
+    },
+    {
+      "epoch": 10.552223477457122,
+      "grad_norm": 0.043213821947574615,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 102747
+    },
+    {
+      "epoch": 10.552531580568964,
+      "grad_norm": 0.13153788447380066,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 102750
+    },
+    {
+      "epoch": 10.552839683680805,
+      "grad_norm": 0.047780994325876236,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 102753
+    },
+    {
+      "epoch": 10.553147786792646,
+      "grad_norm": 0.0988832414150238,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 102756
+    },
+    {
+      "epoch": 10.553455889904487,
+      "grad_norm": 0.06222406029701233,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 102759
+    },
+    {
+      "epoch": 10.553763993016329,
+      "grad_norm": 0.052710965275764465,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 102762
+    },
+    {
+      "epoch": 10.55407209612817,
+      "grad_norm": 0.04257237911224365,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 102765
+    },
+    {
+      "epoch": 10.554380199240013,
+      "grad_norm": 0.06545194983482361,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 102768
+    },
+    {
+      "epoch": 10.554688302351854,
+      "grad_norm": 0.13155023753643036,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 102771
+    },
+    {
+      "epoch": 10.554996405463696,
+      "grad_norm": 0.06403207033872604,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 102774
+    },
+    {
+      "epoch": 10.555304508575537,
+      "grad_norm": 0.0546175017952919,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 102777
+    },
+    {
+      "epoch": 10.555612611687378,
+      "grad_norm": 0.04145312309265137,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 102780
+    },
+    {
+      "epoch": 10.55592071479922,
+      "grad_norm": 0.03439625725150108,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 102783
+    },
+    {
+      "epoch": 10.55622881791106,
+      "grad_norm": 0.0312015563249588,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 102786
+    },
+    {
+      "epoch": 10.556536921022902,
+      "grad_norm": 0.09145013242959976,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 102789
+    },
+    {
+      "epoch": 10.556845024134743,
+      "grad_norm": 0.04190470650792122,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 102792
+    },
+    {
+      "epoch": 10.557153127246584,
+      "grad_norm": 0.07646419107913971,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 102795
+    },
+    {
+      "epoch": 10.557461230358427,
+      "grad_norm": 0.04356531798839569,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 102798
+    },
+    {
+      "epoch": 10.557769333470269,
+      "grad_norm": 0.05221903324127197,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 102801
+    },
+    {
+      "epoch": 10.55807743658211,
+      "grad_norm": 0.04501466825604439,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 102804
+    },
+    {
+      "epoch": 10.558385539693951,
+      "grad_norm": 0.11923180520534515,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 102807
+    },
+    {
+      "epoch": 10.558693642805792,
+      "grad_norm": 0.05566087365150452,
+      "learning_rate": 0.01,
+      "loss": 1.8893,
+      "step": 102810
+    },
+    {
+      "epoch": 10.559001745917634,
+      "grad_norm": 0.10343495011329651,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 102813
+    },
+    {
+      "epoch": 10.559309849029475,
+      "grad_norm": 0.055790793150663376,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 102816
+    },
+    {
+      "epoch": 10.559617952141316,
+      "grad_norm": 0.03885180503129959,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 102819
+    },
+    {
+      "epoch": 10.559926055253158,
+      "grad_norm": 0.0437774695456028,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 102822
+    },
+    {
+      "epoch": 10.560234158364999,
+      "grad_norm": 0.044454824179410934,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 102825
+    },
+    {
+      "epoch": 10.56054226147684,
+      "grad_norm": 0.13780072331428528,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 102828
+    },
+    {
+      "epoch": 10.560850364588683,
+      "grad_norm": 0.04543958604335785,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 102831
+    },
+    {
+      "epoch": 10.561158467700524,
+      "grad_norm": 0.040763672441244125,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 102834
+    },
+    {
+      "epoch": 10.561466570812366,
+      "grad_norm": 0.03325378894805908,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 102837
+    },
+    {
+      "epoch": 10.561774673924207,
+      "grad_norm": 0.0563448891043663,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 102840
+    },
+    {
+      "epoch": 10.562082777036048,
+      "grad_norm": 0.0488412082195282,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 102843
+    },
+    {
+      "epoch": 10.56239088014789,
+      "grad_norm": 0.065616674721241,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 102846
+    },
+    {
+      "epoch": 10.56269898325973,
+      "grad_norm": 0.10472216457128525,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 102849
+    },
+    {
+      "epoch": 10.563007086371572,
+      "grad_norm": 0.08060616999864578,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 102852
+    },
+    {
+      "epoch": 10.563315189483413,
+      "grad_norm": 0.062388233840465546,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 102855
+    },
+    {
+      "epoch": 10.563623292595254,
+      "grad_norm": 0.09207983314990997,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 102858
+    },
+    {
+      "epoch": 10.563931395707097,
+      "grad_norm": 0.06631029397249222,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 102861
+    },
+    {
+      "epoch": 10.564239498818939,
+      "grad_norm": 0.07797393947839737,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 102864
+    },
+    {
+      "epoch": 10.56454760193078,
+      "grad_norm": 0.06673528254032135,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 102867
+    },
+    {
+      "epoch": 10.564855705042621,
+      "grad_norm": 0.09968683868646622,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 102870
+    },
+    {
+      "epoch": 10.565163808154463,
+      "grad_norm": 0.03743911162018776,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 102873
+    },
+    {
+      "epoch": 10.565471911266304,
+      "grad_norm": 0.07518713921308517,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 102876
+    },
+    {
+      "epoch": 10.565780014378145,
+      "grad_norm": 0.035093970596790314,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 102879
+    },
+    {
+      "epoch": 10.566088117489986,
+      "grad_norm": 0.060128387063741684,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 102882
+    },
+    {
+      "epoch": 10.566396220601828,
+      "grad_norm": 0.03313090279698372,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 102885
+    },
+    {
+      "epoch": 10.566704323713669,
+      "grad_norm": 0.03142146021127701,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 102888
+    },
+    {
+      "epoch": 10.56701242682551,
+      "grad_norm": 0.05178171396255493,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 102891
+    },
+    {
+      "epoch": 10.567320529937353,
+      "grad_norm": 0.06281552463769913,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 102894
+    },
+    {
+      "epoch": 10.567628633049194,
+      "grad_norm": 0.10062888264656067,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 102897
+    },
+    {
+      "epoch": 10.567936736161036,
+      "grad_norm": 0.08972020447254181,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 102900
+    },
+    {
+      "epoch": 10.568244839272877,
+      "grad_norm": 0.07290536165237427,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 102903
+    },
+    {
+      "epoch": 10.568552942384718,
+      "grad_norm": 0.10600350797176361,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 102906
+    },
+    {
+      "epoch": 10.56886104549656,
+      "grad_norm": 0.07006838172674179,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 102909
+    },
+    {
+      "epoch": 10.5691691486084,
+      "grad_norm": 0.03484833613038063,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 102912
+    },
+    {
+      "epoch": 10.569477251720242,
+      "grad_norm": 0.09895322471857071,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 102915
+    },
+    {
+      "epoch": 10.569785354832083,
+      "grad_norm": 0.08670248836278915,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 102918
+    },
+    {
+      "epoch": 10.570093457943925,
+      "grad_norm": 0.10926210880279541,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 102921
+    },
+    {
+      "epoch": 10.570401561055768,
+      "grad_norm": 0.05001296475529671,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 102924
+    },
+    {
+      "epoch": 10.570709664167609,
+      "grad_norm": 0.05949382856488228,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 102927
+    },
+    {
+      "epoch": 10.57101776727945,
+      "grad_norm": 0.10504626482725143,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 102930
+    },
+    {
+      "epoch": 10.571325870391291,
+      "grad_norm": 0.037058062851428986,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 102933
+    },
+    {
+      "epoch": 10.571633973503133,
+      "grad_norm": 0.06955143809318542,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 102936
+    },
+    {
+      "epoch": 10.571942076614974,
+      "grad_norm": 0.08500155806541443,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 102939
+    },
+    {
+      "epoch": 10.572250179726815,
+      "grad_norm": 0.05796537175774574,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 102942
+    },
+    {
+      "epoch": 10.572558282838656,
+      "grad_norm": 0.10697682201862335,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 102945
+    },
+    {
+      "epoch": 10.572866385950498,
+      "grad_norm": 0.08648983389139175,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 102948
+    },
+    {
+      "epoch": 10.573174489062339,
+      "grad_norm": 0.0635012835264206,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 102951
+    },
+    {
+      "epoch": 10.57348259217418,
+      "grad_norm": 0.08696910738945007,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 102954
+    },
+    {
+      "epoch": 10.573790695286021,
+      "grad_norm": 0.0655401274561882,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 102957
+    },
+    {
+      "epoch": 10.574098798397864,
+      "grad_norm": 0.03808220475912094,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 102960
+    },
+    {
+      "epoch": 10.574406901509706,
+      "grad_norm": 0.10985107719898224,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 102963
+    },
+    {
+      "epoch": 10.574715004621547,
+      "grad_norm": 0.04316844791173935,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 102966
+    },
+    {
+      "epoch": 10.575023107733388,
+      "grad_norm": 0.07953254133462906,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 102969
+    },
+    {
+      "epoch": 10.57533121084523,
+      "grad_norm": 0.09557713568210602,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 102972
+    },
+    {
+      "epoch": 10.57563931395707,
+      "grad_norm": 0.11650662869215012,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 102975
+    },
+    {
+      "epoch": 10.575947417068912,
+      "grad_norm": 0.056457098573446274,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 102978
+    },
+    {
+      "epoch": 10.576255520180753,
+      "grad_norm": 0.04905838891863823,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 102981
+    },
+    {
+      "epoch": 10.576563623292595,
+      "grad_norm": 0.13630616664886475,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 102984
+    },
+    {
+      "epoch": 10.576871726404438,
+      "grad_norm": 0.04883322864770889,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 102987
+    },
+    {
+      "epoch": 10.577179829516279,
+      "grad_norm": 0.055827610194683075,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 102990
+    },
+    {
+      "epoch": 10.57748793262812,
+      "grad_norm": 0.04752153530716896,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 102993
+    },
+    {
+      "epoch": 10.577796035739961,
+      "grad_norm": 0.03894266113638878,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 102996
+    },
+    {
+      "epoch": 10.578104138851803,
+      "grad_norm": 0.1049915999174118,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 102999
+    },
+    {
+      "epoch": 10.578412241963644,
+      "grad_norm": 0.047973405569791794,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 103002
+    },
+    {
+      "epoch": 10.578720345075485,
+      "grad_norm": 0.03760508447885513,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 103005
+    },
+    {
+      "epoch": 10.579028448187326,
+      "grad_norm": 0.036963608115911484,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 103008
+    },
+    {
+      "epoch": 10.579336551299168,
+      "grad_norm": 0.052010320127010345,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 103011
+    },
+    {
+      "epoch": 10.579644654411009,
+      "grad_norm": 0.054368756711483,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 103014
+    },
+    {
+      "epoch": 10.57995275752285,
+      "grad_norm": 0.04135025665163994,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 103017
+    },
+    {
+      "epoch": 10.580260860634692,
+      "grad_norm": 0.1422293335199356,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 103020
+    },
+    {
+      "epoch": 10.580568963746535,
+      "grad_norm": 0.10143115371465683,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 103023
+    },
+    {
+      "epoch": 10.580877066858376,
+      "grad_norm": 0.03662586957216263,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 103026
+    },
+    {
+      "epoch": 10.581185169970217,
+      "grad_norm": 0.06366636604070663,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 103029
+    },
+    {
+      "epoch": 10.581493273082058,
+      "grad_norm": 0.06303658336400986,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 103032
+    },
+    {
+      "epoch": 10.5818013761939,
+      "grad_norm": 0.07589997351169586,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 103035
+    },
+    {
+      "epoch": 10.58210947930574,
+      "grad_norm": 0.06703688204288483,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 103038
+    },
+    {
+      "epoch": 10.582417582417582,
+      "grad_norm": 0.06799531728029251,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 103041
+    },
+    {
+      "epoch": 10.582725685529423,
+      "grad_norm": 0.0771314799785614,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 103044
+    },
+    {
+      "epoch": 10.583033788641265,
+      "grad_norm": 0.0924944281578064,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 103047
+    },
+    {
+      "epoch": 10.583341891753106,
+      "grad_norm": 0.04260650649666786,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 103050
+    },
+    {
+      "epoch": 10.583649994864949,
+      "grad_norm": 0.07379073649644852,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 103053
+    },
+    {
+      "epoch": 10.58395809797679,
+      "grad_norm": 0.10091949254274368,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 103056
+    },
+    {
+      "epoch": 10.584266201088631,
+      "grad_norm": 0.07707443088293076,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 103059
+    },
+    {
+      "epoch": 10.584574304200473,
+      "grad_norm": 0.07976933568716049,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 103062
+    },
+    {
+      "epoch": 10.584882407312314,
+      "grad_norm": 0.04869668558239937,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 103065
+    },
+    {
+      "epoch": 10.585190510424155,
+      "grad_norm": 0.04817415028810501,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 103068
+    },
+    {
+      "epoch": 10.585498613535997,
+      "grad_norm": 0.03385568782687187,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 103071
+    },
+    {
+      "epoch": 10.585806716647838,
+      "grad_norm": 0.04275520145893097,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 103074
+    },
+    {
+      "epoch": 10.586114819759679,
+      "grad_norm": 0.14263440668582916,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 103077
+    },
+    {
+      "epoch": 10.58642292287152,
+      "grad_norm": 0.05198967084288597,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 103080
+    },
+    {
+      "epoch": 10.586731025983362,
+      "grad_norm": 0.041763633489608765,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 103083
+    },
+    {
+      "epoch": 10.587039129095205,
+      "grad_norm": 0.04080444574356079,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 103086
+    },
+    {
+      "epoch": 10.587347232207046,
+      "grad_norm": 0.08491925895214081,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 103089
+    },
+    {
+      "epoch": 10.587655335318887,
+      "grad_norm": 0.07890848815441132,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 103092
+    },
+    {
+      "epoch": 10.587963438430728,
+      "grad_norm": 0.11974522471427917,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 103095
+    },
+    {
+      "epoch": 10.58827154154257,
+      "grad_norm": 0.1163688525557518,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 103098
+    },
+    {
+      "epoch": 10.588579644654411,
+      "grad_norm": 0.048066623508930206,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 103101
+    },
+    {
+      "epoch": 10.588887747766252,
+      "grad_norm": 0.04997541010379791,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 103104
+    },
+    {
+      "epoch": 10.589195850878093,
+      "grad_norm": 0.06714275479316711,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 103107
+    },
+    {
+      "epoch": 10.589503953989935,
+      "grad_norm": 0.03831176087260246,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 103110
+    },
+    {
+      "epoch": 10.589812057101776,
+      "grad_norm": 0.03693114221096039,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 103113
+    },
+    {
+      "epoch": 10.590120160213619,
+      "grad_norm": 0.03890356793999672,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 103116
+    },
+    {
+      "epoch": 10.59042826332546,
+      "grad_norm": 0.11209408938884735,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 103119
+    },
+    {
+      "epoch": 10.590736366437302,
+      "grad_norm": 0.07923496514558792,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 103122
+    },
+    {
+      "epoch": 10.591044469549143,
+      "grad_norm": 0.04634235054254532,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 103125
+    },
+    {
+      "epoch": 10.591352572660984,
+      "grad_norm": 0.05020580068230629,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 103128
+    },
+    {
+      "epoch": 10.591660675772825,
+      "grad_norm": 0.03846663236618042,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 103131
+    },
+    {
+      "epoch": 10.591968778884667,
+      "grad_norm": 0.03346196934580803,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 103134
+    },
+    {
+      "epoch": 10.592276881996508,
+      "grad_norm": 0.0380299910902977,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 103137
+    },
+    {
+      "epoch": 10.59258498510835,
+      "grad_norm": 0.06727387756109238,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 103140
+    },
+    {
+      "epoch": 10.59289308822019,
+      "grad_norm": 0.06813313812017441,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 103143
+    },
+    {
+      "epoch": 10.593201191332032,
+      "grad_norm": 0.1453511118888855,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 103146
+    },
+    {
+      "epoch": 10.593509294443875,
+      "grad_norm": 0.04510387405753136,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 103149
+    },
+    {
+      "epoch": 10.593817397555716,
+      "grad_norm": 0.03431488573551178,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 103152
+    },
+    {
+      "epoch": 10.594125500667557,
+      "grad_norm": 0.0411214642226696,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 103155
+    },
+    {
+      "epoch": 10.594433603779398,
+      "grad_norm": 0.054067108780145645,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 103158
+    },
+    {
+      "epoch": 10.59474170689124,
+      "grad_norm": 0.050103701651096344,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 103161
+    },
+    {
+      "epoch": 10.595049810003081,
+      "grad_norm": 0.04060819745063782,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 103164
+    },
+    {
+      "epoch": 10.595357913114922,
+      "grad_norm": 0.06645625084638596,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 103167
+    },
+    {
+      "epoch": 10.595666016226764,
+      "grad_norm": 0.08807400614023209,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 103170
+    },
+    {
+      "epoch": 10.595974119338605,
+      "grad_norm": 0.1098518967628479,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 103173
+    },
+    {
+      "epoch": 10.596282222450446,
+      "grad_norm": 0.048754602670669556,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 103176
+    },
+    {
+      "epoch": 10.596590325562289,
+      "grad_norm": 0.0500849112868309,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 103179
+    },
+    {
+      "epoch": 10.59689842867413,
+      "grad_norm": 0.0513179674744606,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 103182
+    },
+    {
+      "epoch": 10.597206531785972,
+      "grad_norm": 0.0895581841468811,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 103185
+    },
+    {
+      "epoch": 10.597514634897813,
+      "grad_norm": 0.08084497600793839,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 103188
+    },
+    {
+      "epoch": 10.597822738009654,
+      "grad_norm": 0.04788369685411453,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 103191
+    },
+    {
+      "epoch": 10.598130841121495,
+      "grad_norm": 0.07950782030820847,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 103194
+    },
+    {
+      "epoch": 10.598438944233337,
+      "grad_norm": 0.07396837323904037,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 103197
+    },
+    {
+      "epoch": 10.598747047345178,
+      "grad_norm": 0.08950217068195343,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 103200
+    },
+    {
+      "epoch": 10.59905515045702,
+      "grad_norm": 0.04552943632006645,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 103203
+    },
+    {
+      "epoch": 10.59936325356886,
+      "grad_norm": 0.09012352675199509,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 103206
+    },
+    {
+      "epoch": 10.599671356680702,
+      "grad_norm": 0.05487809702754021,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 103209
+    },
+    {
+      "epoch": 10.599979459792543,
+      "grad_norm": 0.11796051263809204,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 103212
+    },
+    {
+      "epoch": 10.600287562904386,
+      "grad_norm": 0.07048312574625015,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 103215
+    },
+    {
+      "epoch": 10.600595666016227,
+      "grad_norm": 0.04323925822973251,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 103218
+    },
+    {
+      "epoch": 10.600903769128069,
+      "grad_norm": 0.08954882621765137,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 103221
+    },
+    {
+      "epoch": 10.60121187223991,
+      "grad_norm": 0.08253049850463867,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 103224
+    },
+    {
+      "epoch": 10.601519975351751,
+      "grad_norm": 0.08243921399116516,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 103227
+    },
+    {
+      "epoch": 10.601828078463592,
+      "grad_norm": 0.05085128918290138,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 103230
+    },
+    {
+      "epoch": 10.602136181575434,
+      "grad_norm": 0.1342676877975464,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 103233
+    },
+    {
+      "epoch": 10.602444284687275,
+      "grad_norm": 0.05520642176270485,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 103236
+    },
+    {
+      "epoch": 10.602752387799116,
+      "grad_norm": 0.04326225444674492,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 103239
+    },
+    {
+      "epoch": 10.603060490910957,
+      "grad_norm": 0.09352917969226837,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 103242
+    },
+    {
+      "epoch": 10.6033685940228,
+      "grad_norm": 0.048500899225473404,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 103245
+    },
+    {
+      "epoch": 10.603676697134642,
+      "grad_norm": 0.05939725786447525,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 103248
+    },
+    {
+      "epoch": 10.603984800246483,
+      "grad_norm": 0.06792570650577545,
+      "learning_rate": 0.01,
+      "loss": 1.9698,
+      "step": 103251
+    },
+    {
+      "epoch": 10.604292903358324,
+      "grad_norm": 0.08179488033056259,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 103254
+    },
+    {
+      "epoch": 10.604601006470165,
+      "grad_norm": 0.0700092688202858,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 103257
+    },
+    {
+      "epoch": 10.604909109582007,
+      "grad_norm": 0.07969602197408676,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 103260
+    },
+    {
+      "epoch": 10.605217212693848,
+      "grad_norm": 0.05254344269633293,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 103263
+    },
+    {
+      "epoch": 10.60552531580569,
+      "grad_norm": 0.051222894340753555,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 103266
+    },
+    {
+      "epoch": 10.60583341891753,
+      "grad_norm": 0.1603354960680008,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 103269
+    },
+    {
+      "epoch": 10.606141522029372,
+      "grad_norm": 0.0578870065510273,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 103272
+    },
+    {
+      "epoch": 10.606449625141213,
+      "grad_norm": 0.05055926367640495,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 103275
+    },
+    {
+      "epoch": 10.606757728253056,
+      "grad_norm": 0.07149633020162582,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 103278
+    },
+    {
+      "epoch": 10.607065831364897,
+      "grad_norm": 0.0640505850315094,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 103281
+    },
+    {
+      "epoch": 10.607373934476739,
+      "grad_norm": 0.050150126218795776,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 103284
+    },
+    {
+      "epoch": 10.60768203758858,
+      "grad_norm": 0.04435747489333153,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 103287
+    },
+    {
+      "epoch": 10.607990140700421,
+      "grad_norm": 0.0679631382226944,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 103290
+    },
+    {
+      "epoch": 10.608298243812262,
+      "grad_norm": 0.09328020364046097,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 103293
+    },
+    {
+      "epoch": 10.608606346924104,
+      "grad_norm": 0.07357189059257507,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 103296
+    },
+    {
+      "epoch": 10.608914450035945,
+      "grad_norm": 0.059523552656173706,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 103299
+    },
+    {
+      "epoch": 10.609222553147786,
+      "grad_norm": 0.050263259559869766,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 103302
+    },
+    {
+      "epoch": 10.609530656259627,
+      "grad_norm": 0.19661392271518707,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 103305
+    },
+    {
+      "epoch": 10.60983875937147,
+      "grad_norm": 0.061137836426496506,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 103308
+    },
+    {
+      "epoch": 10.610146862483312,
+      "grad_norm": 0.0863509327173233,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 103311
+    },
+    {
+      "epoch": 10.610454965595153,
+      "grad_norm": 0.07706782966852188,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 103314
+    },
+    {
+      "epoch": 10.610763068706994,
+      "grad_norm": 0.052820686250925064,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 103317
+    },
+    {
+      "epoch": 10.611071171818836,
+      "grad_norm": 0.0664602667093277,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 103320
+    },
+    {
+      "epoch": 10.611379274930677,
+      "grad_norm": 0.05431080237030983,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 103323
+    },
+    {
+      "epoch": 10.611687378042518,
+      "grad_norm": 0.03953436389565468,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 103326
+    },
+    {
+      "epoch": 10.61199548115436,
+      "grad_norm": 0.06307961046695709,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 103329
+    },
+    {
+      "epoch": 10.6123035842662,
+      "grad_norm": 0.03332679718732834,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 103332
+    },
+    {
+      "epoch": 10.612611687378042,
+      "grad_norm": 0.038423825055360794,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 103335
+    },
+    {
+      "epoch": 10.612919790489883,
+      "grad_norm": 0.06862691789865494,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 103338
+    },
+    {
+      "epoch": 10.613227893601726,
+      "grad_norm": 0.1888892948627472,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 103341
+    },
+    {
+      "epoch": 10.613535996713567,
+      "grad_norm": 0.15006403625011444,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 103344
+    },
+    {
+      "epoch": 10.613844099825409,
+      "grad_norm": 0.06723996996879578,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 103347
+    },
+    {
+      "epoch": 10.61415220293725,
+      "grad_norm": 0.04298614710569382,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 103350
+    },
+    {
+      "epoch": 10.614460306049091,
+      "grad_norm": 0.032380230724811554,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 103353
+    },
+    {
+      "epoch": 10.614768409160932,
+      "grad_norm": 0.03456073999404907,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 103356
+    },
+    {
+      "epoch": 10.615076512272774,
+      "grad_norm": 0.06206822022795677,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 103359
+    },
+    {
+      "epoch": 10.615384615384615,
+      "grad_norm": 0.0719355046749115,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 103362
+    },
+    {
+      "epoch": 10.615692718496456,
+      "grad_norm": 0.03969092667102814,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 103365
+    },
+    {
+      "epoch": 10.616000821608297,
+      "grad_norm": 0.04106650501489639,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 103368
+    },
+    {
+      "epoch": 10.61630892472014,
+      "grad_norm": 0.11985891312360764,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 103371
+    },
+    {
+      "epoch": 10.616617027831982,
+      "grad_norm": 0.03696669265627861,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 103374
+    },
+    {
+      "epoch": 10.616925130943823,
+      "grad_norm": 0.13408684730529785,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 103377
+    },
+    {
+      "epoch": 10.617233234055664,
+      "grad_norm": 0.1047915667295456,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 103380
+    },
+    {
+      "epoch": 10.617541337167506,
+      "grad_norm": 0.04256188124418259,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 103383
+    },
+    {
+      "epoch": 10.617849440279347,
+      "grad_norm": 0.04784403368830681,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 103386
+    },
+    {
+      "epoch": 10.618157543391188,
+      "grad_norm": 0.04611273109912872,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 103389
+    },
+    {
+      "epoch": 10.61846564650303,
+      "grad_norm": 0.04264547675848007,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 103392
+    },
+    {
+      "epoch": 10.61877374961487,
+      "grad_norm": 0.061344727873802185,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 103395
+    },
+    {
+      "epoch": 10.619081852726712,
+      "grad_norm": 0.10193429887294769,
+      "learning_rate": 0.01,
+      "loss": 1.9824,
+      "step": 103398
+    },
+    {
+      "epoch": 10.619389955838553,
+      "grad_norm": 0.13179463148117065,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 103401
+    },
+    {
+      "epoch": 10.619698058950396,
+      "grad_norm": 0.05853363499045372,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 103404
+    },
+    {
+      "epoch": 10.620006162062237,
+      "grad_norm": 0.04775145649909973,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 103407
+    },
+    {
+      "epoch": 10.620314265174079,
+      "grad_norm": 0.041010648012161255,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 103410
+    },
+    {
+      "epoch": 10.62062236828592,
+      "grad_norm": 0.044196419417858124,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 103413
+    },
+    {
+      "epoch": 10.620930471397761,
+      "grad_norm": 0.0341707244515419,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 103416
+    },
+    {
+      "epoch": 10.621238574509603,
+      "grad_norm": 0.041209764778614044,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 103419
+    },
+    {
+      "epoch": 10.621546677621444,
+      "grad_norm": 0.0913657546043396,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 103422
+    },
+    {
+      "epoch": 10.621854780733285,
+      "grad_norm": 0.10722614079713821,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 103425
+    },
+    {
+      "epoch": 10.622162883845126,
+      "grad_norm": 0.1794777810573578,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 103428
+    },
+    {
+      "epoch": 10.622470986956968,
+      "grad_norm": 0.14904111623764038,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 103431
+    },
+    {
+      "epoch": 10.62277909006881,
+      "grad_norm": 0.06658944487571716,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 103434
+    },
+    {
+      "epoch": 10.623087193180652,
+      "grad_norm": 0.03310980647802353,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 103437
+    },
+    {
+      "epoch": 10.623395296292493,
+      "grad_norm": 0.039458371698856354,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 103440
+    },
+    {
+      "epoch": 10.623703399404334,
+      "grad_norm": 0.065035380423069,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 103443
+    },
+    {
+      "epoch": 10.624011502516176,
+      "grad_norm": 0.04462461546063423,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 103446
+    },
+    {
+      "epoch": 10.624319605628017,
+      "grad_norm": 0.0657215565443039,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 103449
+    },
+    {
+      "epoch": 10.624627708739858,
+      "grad_norm": 0.042614828795194626,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 103452
+    },
+    {
+      "epoch": 10.6249358118517,
+      "grad_norm": 0.05409550666809082,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 103455
+    },
+    {
+      "epoch": 10.62524391496354,
+      "grad_norm": 0.07609287649393082,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 103458
+    },
+    {
+      "epoch": 10.625552018075382,
+      "grad_norm": 0.04622379690408707,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 103461
+    },
+    {
+      "epoch": 10.625860121187223,
+      "grad_norm": 0.049494169652462006,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 103464
+    },
+    {
+      "epoch": 10.626168224299064,
+      "grad_norm": 0.058355800807476044,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 103467
+    },
+    {
+      "epoch": 10.626476327410908,
+      "grad_norm": 0.10112365335226059,
+      "learning_rate": 0.01,
+      "loss": 1.9763,
+      "step": 103470
+    },
+    {
+      "epoch": 10.626784430522749,
+      "grad_norm": 0.11377458274364471,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 103473
+    },
+    {
+      "epoch": 10.62709253363459,
+      "grad_norm": 0.06197943165898323,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 103476
+    },
+    {
+      "epoch": 10.627400636746431,
+      "grad_norm": 0.04241367056965828,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 103479
+    },
+    {
+      "epoch": 10.627708739858273,
+      "grad_norm": 0.04215838760137558,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 103482
+    },
+    {
+      "epoch": 10.628016842970114,
+      "grad_norm": 0.03457578644156456,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 103485
+    },
+    {
+      "epoch": 10.628324946081955,
+      "grad_norm": 0.13142448663711548,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 103488
+    },
+    {
+      "epoch": 10.628633049193796,
+      "grad_norm": 0.05435696244239807,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 103491
+    },
+    {
+      "epoch": 10.628941152305638,
+      "grad_norm": 0.04535049572587013,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 103494
+    },
+    {
+      "epoch": 10.629249255417479,
+      "grad_norm": 0.06671206653118134,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 103497
+    },
+    {
+      "epoch": 10.629557358529322,
+      "grad_norm": 0.04529844969511032,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 103500
+    },
+    {
+      "epoch": 10.629865461641163,
+      "grad_norm": 0.05721516162157059,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 103503
+    },
+    {
+      "epoch": 10.630173564753004,
+      "grad_norm": 0.04245228320360184,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 103506
+    },
+    {
+      "epoch": 10.630481667864846,
+      "grad_norm": 0.06019366532564163,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 103509
+    },
+    {
+      "epoch": 10.630789770976687,
+      "grad_norm": 0.04783914238214493,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 103512
+    },
+    {
+      "epoch": 10.631097874088528,
+      "grad_norm": 0.04354009032249451,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 103515
+    },
+    {
+      "epoch": 10.63140597720037,
+      "grad_norm": 0.0568438284099102,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 103518
+    },
+    {
+      "epoch": 10.63171408031221,
+      "grad_norm": 0.04983118176460266,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 103521
+    },
+    {
+      "epoch": 10.632022183424052,
+      "grad_norm": 0.06222904846072197,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 103524
+    },
+    {
+      "epoch": 10.632330286535893,
+      "grad_norm": 0.06613916158676147,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 103527
+    },
+    {
+      "epoch": 10.632638389647735,
+      "grad_norm": 0.10853058844804764,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 103530
+    },
+    {
+      "epoch": 10.632946492759578,
+      "grad_norm": 0.07089220732450485,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 103533
+    },
+    {
+      "epoch": 10.633254595871419,
+      "grad_norm": 0.054203297942876816,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 103536
+    },
+    {
+      "epoch": 10.63356269898326,
+      "grad_norm": 0.06964759528636932,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 103539
+    },
+    {
+      "epoch": 10.633870802095101,
+      "grad_norm": 0.07693734765052795,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 103542
+    },
+    {
+      "epoch": 10.634178905206943,
+      "grad_norm": 0.08025120943784714,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 103545
+    },
+    {
+      "epoch": 10.634487008318784,
+      "grad_norm": 0.08096393942832947,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 103548
+    },
+    {
+      "epoch": 10.634795111430625,
+      "grad_norm": 0.0732489600777626,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 103551
+    },
+    {
+      "epoch": 10.635103214542466,
+      "grad_norm": 0.06532766669988632,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 103554
+    },
+    {
+      "epoch": 10.635411317654308,
+      "grad_norm": 0.07934626191854477,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 103557
+    },
+    {
+      "epoch": 10.635719420766149,
+      "grad_norm": 0.04989808052778244,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 103560
+    },
+    {
+      "epoch": 10.636027523877992,
+      "grad_norm": 0.057780925184488297,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 103563
+    },
+    {
+      "epoch": 10.636335626989833,
+      "grad_norm": 0.040844134986400604,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 103566
+    },
+    {
+      "epoch": 10.636643730101675,
+      "grad_norm": 0.04943647235631943,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 103569
+    },
+    {
+      "epoch": 10.636951833213516,
+      "grad_norm": 0.13489899039268494,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 103572
+    },
+    {
+      "epoch": 10.637259936325357,
+      "grad_norm": 0.04333699867129326,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 103575
+    },
+    {
+      "epoch": 10.637568039437198,
+      "grad_norm": 0.08138672262430191,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 103578
+    },
+    {
+      "epoch": 10.63787614254904,
+      "grad_norm": 0.08707044273614883,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 103581
+    },
+    {
+      "epoch": 10.63818424566088,
+      "grad_norm": 0.05804765969514847,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 103584
+    },
+    {
+      "epoch": 10.638492348772722,
+      "grad_norm": 0.04205789417028427,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 103587
+    },
+    {
+      "epoch": 10.638800451884563,
+      "grad_norm": 0.05844965577125549,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 103590
+    },
+    {
+      "epoch": 10.639108554996405,
+      "grad_norm": 0.052618734538555145,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 103593
+    },
+    {
+      "epoch": 10.639416658108248,
+      "grad_norm": 0.04634091630578041,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 103596
+    },
+    {
+      "epoch": 10.639724761220089,
+      "grad_norm": 0.043263375759124756,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 103599
+    },
+    {
+      "epoch": 10.64003286433193,
+      "grad_norm": 0.10971635580062866,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 103602
+    },
+    {
+      "epoch": 10.640340967443771,
+      "grad_norm": 0.053758516907691956,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 103605
+    },
+    {
+      "epoch": 10.640649070555613,
+      "grad_norm": 0.08278075605630875,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 103608
+    },
+    {
+      "epoch": 10.640957173667454,
+      "grad_norm": 0.101168192923069,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 103611
+    },
+    {
+      "epoch": 10.641265276779295,
+      "grad_norm": 0.047223761677742004,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 103614
+    },
+    {
+      "epoch": 10.641573379891136,
+      "grad_norm": 0.04839014261960983,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 103617
+    },
+    {
+      "epoch": 10.641881483002978,
+      "grad_norm": 0.11143355071544647,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 103620
+    },
+    {
+      "epoch": 10.642189586114819,
+      "grad_norm": 0.06389351189136505,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 103623
+    },
+    {
+      "epoch": 10.642497689226662,
+      "grad_norm": 0.08618534356355667,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 103626
+    },
+    {
+      "epoch": 10.642805792338503,
+      "grad_norm": 0.06256888061761856,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 103629
+    },
+    {
+      "epoch": 10.643113895450345,
+      "grad_norm": 0.0647435113787651,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 103632
+    },
+    {
+      "epoch": 10.643421998562186,
+      "grad_norm": 0.04641449451446533,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 103635
+    },
+    {
+      "epoch": 10.643730101674027,
+      "grad_norm": 0.08702532947063446,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 103638
+    },
+    {
+      "epoch": 10.644038204785868,
+      "grad_norm": 0.05984359234571457,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 103641
+    },
+    {
+      "epoch": 10.64434630789771,
+      "grad_norm": 0.07888772338628769,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 103644
+    },
+    {
+      "epoch": 10.644654411009551,
+      "grad_norm": 0.11645808070898056,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 103647
+    },
+    {
+      "epoch": 10.644962514121392,
+      "grad_norm": 0.07762550562620163,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 103650
+    },
+    {
+      "epoch": 10.645270617233233,
+      "grad_norm": 0.06812987476587296,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 103653
+    },
+    {
+      "epoch": 10.645578720345075,
+      "grad_norm": 0.12551312148571014,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 103656
+    },
+    {
+      "epoch": 10.645886823456918,
+      "grad_norm": 0.11314782500267029,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 103659
+    },
+    {
+      "epoch": 10.646194926568759,
+      "grad_norm": 0.05329858884215355,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 103662
+    },
+    {
+      "epoch": 10.6465030296806,
+      "grad_norm": 0.05354480445384979,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 103665
+    },
+    {
+      "epoch": 10.646811132792442,
+      "grad_norm": 0.04435323551297188,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 103668
+    },
+    {
+      "epoch": 10.647119235904283,
+      "grad_norm": 0.05537758395075798,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 103671
+    },
+    {
+      "epoch": 10.647427339016124,
+      "grad_norm": 0.10855624824762344,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 103674
+    },
+    {
+      "epoch": 10.647735442127965,
+      "grad_norm": 0.0933624804019928,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 103677
+    },
+    {
+      "epoch": 10.648043545239807,
+      "grad_norm": 0.046108171343803406,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 103680
+    },
+    {
+      "epoch": 10.648351648351648,
+      "grad_norm": 0.08186739683151245,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 103683
+    },
+    {
+      "epoch": 10.648659751463489,
+      "grad_norm": 0.12513357400894165,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 103686
+    },
+    {
+      "epoch": 10.648967854575332,
+      "grad_norm": 0.08822114765644073,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 103689
+    },
+    {
+      "epoch": 10.649275957687173,
+      "grad_norm": 0.06944266706705093,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 103692
+    },
+    {
+      "epoch": 10.649584060799015,
+      "grad_norm": 0.0754237249493599,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 103695
+    },
+    {
+      "epoch": 10.649892163910856,
+      "grad_norm": 0.11520032584667206,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 103698
+    },
+    {
+      "epoch": 10.650200267022697,
+      "grad_norm": 0.07506398856639862,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 103701
+    },
+    {
+      "epoch": 10.650508370134538,
+      "grad_norm": 0.057485081255435944,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 103704
+    },
+    {
+      "epoch": 10.65081647324638,
+      "grad_norm": 0.0681445300579071,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 103707
+    },
+    {
+      "epoch": 10.651124576358221,
+      "grad_norm": 0.04196304827928543,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 103710
+    },
+    {
+      "epoch": 10.651432679470062,
+      "grad_norm": 0.08239495754241943,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 103713
+    },
+    {
+      "epoch": 10.651740782581903,
+      "grad_norm": 0.057062406092882156,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 103716
+    },
+    {
+      "epoch": 10.652048885693745,
+      "grad_norm": 0.04289278760552406,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 103719
+    },
+    {
+      "epoch": 10.652356988805586,
+      "grad_norm": 0.08105800300836563,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 103722
+    },
+    {
+      "epoch": 10.652665091917429,
+      "grad_norm": 0.05087985843420029,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 103725
+    },
+    {
+      "epoch": 10.65297319502927,
+      "grad_norm": 0.06125885993242264,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 103728
+    },
+    {
+      "epoch": 10.653281298141112,
+      "grad_norm": 0.11369001865386963,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 103731
+    },
+    {
+      "epoch": 10.653589401252953,
+      "grad_norm": 0.09310147911310196,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 103734
+    },
+    {
+      "epoch": 10.653897504364794,
+      "grad_norm": 0.0653354674577713,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 103737
+    },
+    {
+      "epoch": 10.654205607476635,
+      "grad_norm": 0.041785553097724915,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 103740
+    },
+    {
+      "epoch": 10.654513710588477,
+      "grad_norm": 0.03783538565039635,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 103743
+    },
+    {
+      "epoch": 10.654821813700318,
+      "grad_norm": 0.09894753247499466,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 103746
+    },
+    {
+      "epoch": 10.65512991681216,
+      "grad_norm": 0.09766748547554016,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 103749
+    },
+    {
+      "epoch": 10.655438019924,
+      "grad_norm": 0.06367824971675873,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 103752
+    },
+    {
+      "epoch": 10.655746123035843,
+      "grad_norm": 0.10057856142520905,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 103755
+    },
+    {
+      "epoch": 10.656054226147685,
+      "grad_norm": 0.06819339096546173,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 103758
+    },
+    {
+      "epoch": 10.656362329259526,
+      "grad_norm": 0.04021526128053665,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 103761
+    },
+    {
+      "epoch": 10.656670432371367,
+      "grad_norm": 0.050846509635448456,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 103764
+    },
+    {
+      "epoch": 10.656978535483209,
+      "grad_norm": 0.048599909991025925,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 103767
+    },
+    {
+      "epoch": 10.65728663859505,
+      "grad_norm": 0.1137971431016922,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 103770
+    },
+    {
+      "epoch": 10.657594741706891,
+      "grad_norm": 0.08687986433506012,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 103773
+    },
+    {
+      "epoch": 10.657902844818732,
+      "grad_norm": 0.07247614115476608,
+      "learning_rate": 0.01,
+      "loss": 1.9829,
+      "step": 103776
+    },
+    {
+      "epoch": 10.658210947930574,
+      "grad_norm": 0.05305969715118408,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 103779
+    },
+    {
+      "epoch": 10.658519051042415,
+      "grad_norm": 0.043938048183918,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 103782
+    },
+    {
+      "epoch": 10.658827154154256,
+      "grad_norm": 0.058209795504808426,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 103785
+    },
+    {
+      "epoch": 10.6591352572661,
+      "grad_norm": 0.08251186460256577,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 103788
+    },
+    {
+      "epoch": 10.65944336037794,
+      "grad_norm": 0.06727088242769241,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 103791
+    },
+    {
+      "epoch": 10.659751463489782,
+      "grad_norm": 0.06473030149936676,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 103794
+    },
+    {
+      "epoch": 10.660059566601623,
+      "grad_norm": 0.12840333580970764,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 103797
+    },
+    {
+      "epoch": 10.660367669713464,
+      "grad_norm": 0.11226318776607513,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 103800
+    },
+    {
+      "epoch": 10.660675772825305,
+      "grad_norm": 0.04158809781074524,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 103803
+    },
+    {
+      "epoch": 10.660983875937147,
+      "grad_norm": 0.11002946645021439,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 103806
+    },
+    {
+      "epoch": 10.661291979048988,
+      "grad_norm": 0.11193925142288208,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 103809
+    },
+    {
+      "epoch": 10.66160008216083,
+      "grad_norm": 0.04690947011113167,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 103812
+    },
+    {
+      "epoch": 10.66190818527267,
+      "grad_norm": 0.03682129830121994,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 103815
+    },
+    {
+      "epoch": 10.662216288384514,
+      "grad_norm": 0.03982137516140938,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 103818
+    },
+    {
+      "epoch": 10.662524391496355,
+      "grad_norm": 0.05768255144357681,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 103821
+    },
+    {
+      "epoch": 10.662832494608196,
+      "grad_norm": 0.03755057603120804,
+      "learning_rate": 0.01,
+      "loss": 1.9722,
+      "step": 103824
+    },
+    {
+      "epoch": 10.663140597720037,
+      "grad_norm": 0.10326661914587021,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 103827
+    },
+    {
+      "epoch": 10.663448700831879,
+      "grad_norm": 0.06437335908412933,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 103830
+    },
+    {
+      "epoch": 10.66375680394372,
+      "grad_norm": 0.10684889554977417,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 103833
+    },
+    {
+      "epoch": 10.664064907055561,
+      "grad_norm": 0.0675310418009758,
+      "learning_rate": 0.01,
+      "loss": 1.97,
+      "step": 103836
+    },
+    {
+      "epoch": 10.664373010167402,
+      "grad_norm": 0.15718181431293488,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 103839
+    },
+    {
+      "epoch": 10.664681113279244,
+      "grad_norm": 0.08253775537014008,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 103842
+    },
+    {
+      "epoch": 10.664989216391085,
+      "grad_norm": 0.06902614235877991,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 103845
+    },
+    {
+      "epoch": 10.665297319502926,
+      "grad_norm": 0.07362678647041321,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 103848
+    },
+    {
+      "epoch": 10.66560542261477,
+      "grad_norm": 0.08512471616268158,
+      "learning_rate": 0.01,
+      "loss": 1.8981,
+      "step": 103851
+    },
+    {
+      "epoch": 10.66591352572661,
+      "grad_norm": 0.09559623152017593,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 103854
+    },
+    {
+      "epoch": 10.666221628838452,
+      "grad_norm": 0.1007629781961441,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 103857
+    },
+    {
+      "epoch": 10.666529731950293,
+      "grad_norm": 0.03798322752118111,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 103860
+    },
+    {
+      "epoch": 10.666837835062134,
+      "grad_norm": 0.056238915771245956,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 103863
+    },
+    {
+      "epoch": 10.667145938173975,
+      "grad_norm": 0.06413375586271286,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 103866
+    },
+    {
+      "epoch": 10.667454041285817,
+      "grad_norm": 0.0409151092171669,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 103869
+    },
+    {
+      "epoch": 10.667762144397658,
+      "grad_norm": 0.07371127605438232,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 103872
+    },
+    {
+      "epoch": 10.6680702475095,
+      "grad_norm": 0.037266094237565994,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 103875
+    },
+    {
+      "epoch": 10.66837835062134,
+      "grad_norm": 0.04316391795873642,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 103878
+    },
+    {
+      "epoch": 10.668686453733184,
+      "grad_norm": 0.044815342873334885,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 103881
+    },
+    {
+      "epoch": 10.668994556845025,
+      "grad_norm": 0.1637774407863617,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 103884
+    },
+    {
+      "epoch": 10.669302659956866,
+      "grad_norm": 0.08285655081272125,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 103887
+    },
+    {
+      "epoch": 10.669610763068707,
+      "grad_norm": 0.050641678273677826,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 103890
+    },
+    {
+      "epoch": 10.669918866180549,
+      "grad_norm": 0.034874316304922104,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 103893
+    },
+    {
+      "epoch": 10.67022696929239,
+      "grad_norm": 0.03403162956237793,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 103896
+    },
+    {
+      "epoch": 10.670535072404231,
+      "grad_norm": 0.030581431463360786,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 103899
+    },
+    {
+      "epoch": 10.670843175516072,
+      "grad_norm": 0.05116745084524155,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 103902
+    },
+    {
+      "epoch": 10.671151278627914,
+      "grad_norm": 0.04860899969935417,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 103905
+    },
+    {
+      "epoch": 10.671459381739755,
+      "grad_norm": 0.08376903831958771,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 103908
+    },
+    {
+      "epoch": 10.671767484851596,
+      "grad_norm": 0.04416081681847572,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 103911
+    },
+    {
+      "epoch": 10.67207558796344,
+      "grad_norm": 0.07991000264883041,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 103914
+    },
+    {
+      "epoch": 10.67238369107528,
+      "grad_norm": 0.0784083679318428,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 103917
+    },
+    {
+      "epoch": 10.672691794187122,
+      "grad_norm": 0.0480702742934227,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 103920
+    },
+    {
+      "epoch": 10.672999897298963,
+      "grad_norm": 0.03636328503489494,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 103923
+    },
+    {
+      "epoch": 10.673308000410804,
+      "grad_norm": 0.04388698935508728,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 103926
+    },
+    {
+      "epoch": 10.673616103522646,
+      "grad_norm": 0.10632793605327606,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 103929
+    },
+    {
+      "epoch": 10.673924206634487,
+      "grad_norm": 0.09042569249868393,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 103932
+    },
+    {
+      "epoch": 10.674232309746328,
+      "grad_norm": 0.11588452756404877,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 103935
+    },
+    {
+      "epoch": 10.67454041285817,
+      "grad_norm": 0.0780494436621666,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 103938
+    },
+    {
+      "epoch": 10.67484851597001,
+      "grad_norm": 0.05877825245261192,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 103941
+    },
+    {
+      "epoch": 10.675156619081854,
+      "grad_norm": 0.051702238619327545,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 103944
+    },
+    {
+      "epoch": 10.675464722193695,
+      "grad_norm": 0.07100822031497955,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 103947
+    },
+    {
+      "epoch": 10.675772825305536,
+      "grad_norm": 0.042687855660915375,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 103950
+    },
+    {
+      "epoch": 10.676080928417377,
+      "grad_norm": 0.04055164381861687,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 103953
+    },
+    {
+      "epoch": 10.676389031529219,
+      "grad_norm": 0.05162336304783821,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 103956
+    },
+    {
+      "epoch": 10.67669713464106,
+      "grad_norm": 0.0412101112306118,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 103959
+    },
+    {
+      "epoch": 10.677005237752901,
+      "grad_norm": 0.0823889672756195,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 103962
+    },
+    {
+      "epoch": 10.677313340864742,
+      "grad_norm": 0.13297125697135925,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 103965
+    },
+    {
+      "epoch": 10.677621443976584,
+      "grad_norm": 0.04852784797549248,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 103968
+    },
+    {
+      "epoch": 10.677929547088425,
+      "grad_norm": 0.04298737272620201,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 103971
+    },
+    {
+      "epoch": 10.678237650200266,
+      "grad_norm": 0.04336988553404808,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 103974
+    },
+    {
+      "epoch": 10.678545753312108,
+      "grad_norm": 0.09840462356805801,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 103977
+    },
+    {
+      "epoch": 10.67885385642395,
+      "grad_norm": 0.06105434149503708,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 103980
+    },
+    {
+      "epoch": 10.679161959535792,
+      "grad_norm": 0.048671264201402664,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 103983
+    },
+    {
+      "epoch": 10.679470062647633,
+      "grad_norm": 0.05965422838926315,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 103986
+    },
+    {
+      "epoch": 10.679778165759474,
+      "grad_norm": 0.0500924289226532,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 103989
+    },
+    {
+      "epoch": 10.680086268871316,
+      "grad_norm": 0.04120882600545883,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 103992
+    },
+    {
+      "epoch": 10.680394371983157,
+      "grad_norm": 0.07993176579475403,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 103995
+    },
+    {
+      "epoch": 10.680702475094998,
+      "grad_norm": 0.03478511795401573,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 103998
+    },
+    {
+      "epoch": 10.68101057820684,
+      "grad_norm": 0.09511188417673111,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 104001
+    },
+    {
+      "epoch": 10.68131868131868,
+      "grad_norm": 0.07670678943395615,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 104004
+    },
+    {
+      "epoch": 10.681626784430522,
+      "grad_norm": 0.05489913374185562,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 104007
+    },
+    {
+      "epoch": 10.681934887542365,
+      "grad_norm": 0.09096886962652206,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 104010
+    },
+    {
+      "epoch": 10.682242990654206,
+      "grad_norm": 0.05405969172716141,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 104013
+    },
+    {
+      "epoch": 10.682551093766048,
+      "grad_norm": 0.08224764466285706,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 104016
+    },
+    {
+      "epoch": 10.682859196877889,
+      "grad_norm": 0.07674596458673477,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 104019
+    },
+    {
+      "epoch": 10.68316729998973,
+      "grad_norm": 0.05211837589740753,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 104022
+    },
+    {
+      "epoch": 10.683475403101571,
+      "grad_norm": 0.07650485634803772,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 104025
+    },
+    {
+      "epoch": 10.683783506213413,
+      "grad_norm": 0.09514635056257248,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 104028
+    },
+    {
+      "epoch": 10.684091609325254,
+      "grad_norm": 0.049703311175107956,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 104031
+    },
+    {
+      "epoch": 10.684399712437095,
+      "grad_norm": 0.037082523107528687,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 104034
+    },
+    {
+      "epoch": 10.684707815548936,
+      "grad_norm": 0.035258326679468155,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 104037
+    },
+    {
+      "epoch": 10.685015918660778,
+      "grad_norm": 0.12024477869272232,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 104040
+    },
+    {
+      "epoch": 10.68532402177262,
+      "grad_norm": 0.10448408126831055,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 104043
+    },
+    {
+      "epoch": 10.685632124884462,
+      "grad_norm": 0.03678799793124199,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 104046
+    },
+    {
+      "epoch": 10.685940227996303,
+      "grad_norm": 0.08878728747367859,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 104049
+    },
+    {
+      "epoch": 10.686248331108144,
+      "grad_norm": 0.04152219742536545,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 104052
+    },
+    {
+      "epoch": 10.686556434219986,
+      "grad_norm": 0.07425837218761444,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 104055
+    },
+    {
+      "epoch": 10.686864537331827,
+      "grad_norm": 0.03856934979557991,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 104058
+    },
+    {
+      "epoch": 10.687172640443668,
+      "grad_norm": 0.03087785467505455,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 104061
+    },
+    {
+      "epoch": 10.68748074355551,
+      "grad_norm": 0.054895635694265366,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 104064
+    },
+    {
+      "epoch": 10.68778884666735,
+      "grad_norm": 0.11289512366056442,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 104067
+    },
+    {
+      "epoch": 10.688096949779192,
+      "grad_norm": 0.050542768090963364,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 104070
+    },
+    {
+      "epoch": 10.688405052891035,
+      "grad_norm": 0.07830153405666351,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 104073
+    },
+    {
+      "epoch": 10.688713156002876,
+      "grad_norm": 0.03961231932044029,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 104076
+    },
+    {
+      "epoch": 10.689021259114718,
+      "grad_norm": 0.03384247049689293,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 104079
+    },
+    {
+      "epoch": 10.689329362226559,
+      "grad_norm": 0.035751618444919586,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 104082
+    },
+    {
+      "epoch": 10.6896374653384,
+      "grad_norm": 0.04944929853081703,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 104085
+    },
+    {
+      "epoch": 10.689945568450241,
+      "grad_norm": 0.06898265331983566,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 104088
+    },
+    {
+      "epoch": 10.690253671562083,
+      "grad_norm": 0.1251329779624939,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 104091
+    },
+    {
+      "epoch": 10.690561774673924,
+      "grad_norm": 0.05030253902077675,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 104094
+    },
+    {
+      "epoch": 10.690869877785765,
+      "grad_norm": 0.041641298681497574,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 104097
+    },
+    {
+      "epoch": 10.691177980897606,
+      "grad_norm": 0.11983517557382584,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 104100
+    },
+    {
+      "epoch": 10.691486084009448,
+      "grad_norm": 0.04607795923948288,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 104103
+    },
+    {
+      "epoch": 10.69179418712129,
+      "grad_norm": 0.11364562064409256,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 104106
+    },
+    {
+      "epoch": 10.692102290233132,
+      "grad_norm": 0.04407837986946106,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 104109
+    },
+    {
+      "epoch": 10.692410393344973,
+      "grad_norm": 0.035745903849601746,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 104112
+    },
+    {
+      "epoch": 10.692718496456814,
+      "grad_norm": 0.03364009037613869,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 104115
+    },
+    {
+      "epoch": 10.693026599568656,
+      "grad_norm": 0.03567780181765556,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 104118
+    },
+    {
+      "epoch": 10.693334702680497,
+      "grad_norm": 0.07825560867786407,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 104121
+    },
+    {
+      "epoch": 10.693642805792338,
+      "grad_norm": 0.04002043232321739,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 104124
+    },
+    {
+      "epoch": 10.69395090890418,
+      "grad_norm": 0.044410590082407,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 104127
+    },
+    {
+      "epoch": 10.69425901201602,
+      "grad_norm": 0.03323044627904892,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 104130
+    },
+    {
+      "epoch": 10.694567115127862,
+      "grad_norm": 0.0373239666223526,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 104133
+    },
+    {
+      "epoch": 10.694875218239705,
+      "grad_norm": 0.058051690459251404,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 104136
+    },
+    {
+      "epoch": 10.695183321351546,
+      "grad_norm": 0.1667533963918686,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 104139
+    },
+    {
+      "epoch": 10.695491424463388,
+      "grad_norm": 0.046205274760723114,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 104142
+    },
+    {
+      "epoch": 10.695799527575229,
+      "grad_norm": 0.07125911861658096,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 104145
+    },
+    {
+      "epoch": 10.69610763068707,
+      "grad_norm": 0.09456393122673035,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 104148
+    },
+    {
+      "epoch": 10.696415733798911,
+      "grad_norm": 0.10501796752214432,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 104151
+    },
+    {
+      "epoch": 10.696723836910753,
+      "grad_norm": 0.05251304805278778,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 104154
+    },
+    {
+      "epoch": 10.697031940022594,
+      "grad_norm": 0.05828684940934181,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 104157
+    },
+    {
+      "epoch": 10.697340043134435,
+      "grad_norm": 0.048068683594465256,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 104160
+    },
+    {
+      "epoch": 10.697648146246276,
+      "grad_norm": 0.09374748915433884,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 104163
+    },
+    {
+      "epoch": 10.697956249358118,
+      "grad_norm": 0.057782042771577835,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 104166
+    },
+    {
+      "epoch": 10.69826435246996,
+      "grad_norm": 0.052854880690574646,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 104169
+    },
+    {
+      "epoch": 10.698572455581802,
+      "grad_norm": 0.050680410116910934,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 104172
+    },
+    {
+      "epoch": 10.698880558693643,
+      "grad_norm": 0.05988655984401703,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 104175
+    },
+    {
+      "epoch": 10.699188661805485,
+      "grad_norm": 0.04978611692786217,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 104178
+    },
+    {
+      "epoch": 10.699496764917326,
+      "grad_norm": 0.06838058680295944,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 104181
+    },
+    {
+      "epoch": 10.699804868029167,
+      "grad_norm": 0.04077281057834625,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 104184
+    },
+    {
+      "epoch": 10.700112971141008,
+      "grad_norm": 0.050454989075660706,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 104187
+    },
+    {
+      "epoch": 10.70042107425285,
+      "grad_norm": 0.04417501762509346,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 104190
+    },
+    {
+      "epoch": 10.70072917736469,
+      "grad_norm": 0.03660339489579201,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 104193
+    },
+    {
+      "epoch": 10.701037280476532,
+      "grad_norm": 0.03626766800880432,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 104196
+    },
+    {
+      "epoch": 10.701345383588375,
+      "grad_norm": 0.11914758384227753,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 104199
+    },
+    {
+      "epoch": 10.701653486700216,
+      "grad_norm": 0.07309869676828384,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 104202
+    },
+    {
+      "epoch": 10.701961589812058,
+      "grad_norm": 0.1361292004585266,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 104205
+    },
+    {
+      "epoch": 10.702269692923899,
+      "grad_norm": 0.07265239208936691,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 104208
+    },
+    {
+      "epoch": 10.70257779603574,
+      "grad_norm": 0.05486297979950905,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 104211
+    },
+    {
+      "epoch": 10.702885899147581,
+      "grad_norm": 0.05077483132481575,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 104214
+    },
+    {
+      "epoch": 10.703194002259423,
+      "grad_norm": 0.04137976095080376,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 104217
+    },
+    {
+      "epoch": 10.703502105371264,
+      "grad_norm": 0.040473662316799164,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 104220
+    },
+    {
+      "epoch": 10.703810208483105,
+      "grad_norm": 0.04199198633432388,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 104223
+    },
+    {
+      "epoch": 10.704118311594947,
+      "grad_norm": 0.054861754179000854,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 104226
+    },
+    {
+      "epoch": 10.704426414706788,
+      "grad_norm": 0.04389738291501999,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 104229
+    },
+    {
+      "epoch": 10.704734517818629,
+      "grad_norm": 0.10139840841293335,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 104232
+    },
+    {
+      "epoch": 10.705042620930472,
+      "grad_norm": 0.18040791153907776,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 104235
+    },
+    {
+      "epoch": 10.705350724042313,
+      "grad_norm": 0.09351298958063126,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 104238
+    },
+    {
+      "epoch": 10.705658827154155,
+      "grad_norm": 0.04835623502731323,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 104241
+    },
+    {
+      "epoch": 10.705966930265996,
+      "grad_norm": 0.07934877276420593,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 104244
+    },
+    {
+      "epoch": 10.706275033377837,
+      "grad_norm": 0.04809239134192467,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 104247
+    },
+    {
+      "epoch": 10.706583136489678,
+      "grad_norm": 0.09115726500749588,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 104250
+    },
+    {
+      "epoch": 10.70689123960152,
+      "grad_norm": 0.09909907728433609,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 104253
+    },
+    {
+      "epoch": 10.707199342713361,
+      "grad_norm": 0.06753792613744736,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 104256
+    },
+    {
+      "epoch": 10.707507445825202,
+      "grad_norm": 0.047283709049224854,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 104259
+    },
+    {
+      "epoch": 10.707815548937043,
+      "grad_norm": 0.0870928168296814,
+      "learning_rate": 0.01,
+      "loss": 1.8767,
+      "step": 104262
+    },
+    {
+      "epoch": 10.708123652048886,
+      "grad_norm": 0.04126760736107826,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 104265
+    },
+    {
+      "epoch": 10.708431755160728,
+      "grad_norm": 0.052089888602495193,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 104268
+    },
+    {
+      "epoch": 10.708739858272569,
+      "grad_norm": 0.040085259824991226,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 104271
+    },
+    {
+      "epoch": 10.70904796138441,
+      "grad_norm": 0.03780793771147728,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 104274
+    },
+    {
+      "epoch": 10.709356064496252,
+      "grad_norm": 0.03962109982967377,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 104277
+    },
+    {
+      "epoch": 10.709664167608093,
+      "grad_norm": 0.10808218270540237,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 104280
+    },
+    {
+      "epoch": 10.709972270719934,
+      "grad_norm": 0.06844981759786606,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 104283
+    },
+    {
+      "epoch": 10.710280373831775,
+      "grad_norm": 0.08019477128982544,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 104286
+    },
+    {
+      "epoch": 10.710588476943617,
+      "grad_norm": 0.07778795063495636,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 104289
+    },
+    {
+      "epoch": 10.710896580055458,
+      "grad_norm": 0.03524641692638397,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 104292
+    },
+    {
+      "epoch": 10.7112046831673,
+      "grad_norm": 0.11965351551771164,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 104295
+    },
+    {
+      "epoch": 10.711512786279142,
+      "grad_norm": 0.10487991571426392,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 104298
+    },
+    {
+      "epoch": 10.711820889390983,
+      "grad_norm": 0.10835332423448563,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 104301
+    },
+    {
+      "epoch": 10.712128992502825,
+      "grad_norm": 0.058550573885440826,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 104304
+    },
+    {
+      "epoch": 10.712437095614666,
+      "grad_norm": 0.041659001260995865,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 104307
+    },
+    {
+      "epoch": 10.712745198726507,
+      "grad_norm": 0.059879958629608154,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 104310
+    },
+    {
+      "epoch": 10.713053301838348,
+      "grad_norm": 0.03812127932906151,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 104313
+    },
+    {
+      "epoch": 10.71336140495019,
+      "grad_norm": 0.0401906780898571,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 104316
+    },
+    {
+      "epoch": 10.713669508062031,
+      "grad_norm": 0.0772302895784378,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 104319
+    },
+    {
+      "epoch": 10.713977611173872,
+      "grad_norm": 0.045765917748212814,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 104322
+    },
+    {
+      "epoch": 10.714285714285714,
+      "grad_norm": 0.05593492090702057,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 104325
+    },
+    {
+      "epoch": 10.714593817397557,
+      "grad_norm": 0.08135896176099777,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 104328
+    },
+    {
+      "epoch": 10.714901920509398,
+      "grad_norm": 0.06063408404588699,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 104331
+    },
+    {
+      "epoch": 10.715210023621239,
+      "grad_norm": 0.04096072539687157,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 104334
+    },
+    {
+      "epoch": 10.71551812673308,
+      "grad_norm": 0.050055526196956635,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 104337
+    },
+    {
+      "epoch": 10.715826229844922,
+      "grad_norm": 0.04591112583875656,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 104340
+    },
+    {
+      "epoch": 10.716134332956763,
+      "grad_norm": 0.1036655455827713,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 104343
+    },
+    {
+      "epoch": 10.716442436068604,
+      "grad_norm": 0.037715498358011246,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 104346
+    },
+    {
+      "epoch": 10.716750539180445,
+      "grad_norm": 0.047764185816049576,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 104349
+    },
+    {
+      "epoch": 10.717058642292287,
+      "grad_norm": 0.06210208684206009,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 104352
+    },
+    {
+      "epoch": 10.717366745404128,
+      "grad_norm": 0.10889417678117752,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 104355
+    },
+    {
+      "epoch": 10.71767484851597,
+      "grad_norm": 0.05223528668284416,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 104358
+    },
+    {
+      "epoch": 10.717982951627812,
+      "grad_norm": 0.08993866294622421,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 104361
+    },
+    {
+      "epoch": 10.718291054739653,
+      "grad_norm": 0.12946274876594543,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 104364
+    },
+    {
+      "epoch": 10.718599157851495,
+      "grad_norm": 0.05884292721748352,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 104367
+    },
+    {
+      "epoch": 10.718907260963336,
+      "grad_norm": 0.05492502823472023,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 104370
+    },
+    {
+      "epoch": 10.719215364075177,
+      "grad_norm": 0.049445055425167084,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 104373
+    },
+    {
+      "epoch": 10.719523467187019,
+      "grad_norm": 0.039928633719682693,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 104376
+    },
+    {
+      "epoch": 10.71983157029886,
+      "grad_norm": 0.045593008399009705,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 104379
+    },
+    {
+      "epoch": 10.720139673410701,
+      "grad_norm": 0.10593906044960022,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 104382
+    },
+    {
+      "epoch": 10.720447776522542,
+      "grad_norm": 0.07790132611989975,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 104385
+    },
+    {
+      "epoch": 10.720755879634384,
+      "grad_norm": 0.08720636367797852,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 104388
+    },
+    {
+      "epoch": 10.721063982746227,
+      "grad_norm": 0.07232289761304855,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 104391
+    },
+    {
+      "epoch": 10.721372085858068,
+      "grad_norm": 0.044367242604494095,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 104394
+    },
+    {
+      "epoch": 10.72168018896991,
+      "grad_norm": 0.050533805042505264,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 104397
+    },
+    {
+      "epoch": 10.72198829208175,
+      "grad_norm": 0.0687800869345665,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 104400
+    },
+    {
+      "epoch": 10.722296395193592,
+      "grad_norm": 0.09476820379495621,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 104403
+    },
+    {
+      "epoch": 10.722604498305433,
+      "grad_norm": 0.08872959762811661,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 104406
+    },
+    {
+      "epoch": 10.722912601417274,
+      "grad_norm": 0.10332068800926208,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 104409
+    },
+    {
+      "epoch": 10.723220704529115,
+      "grad_norm": 0.05728401243686676,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 104412
+    },
+    {
+      "epoch": 10.723528807640957,
+      "grad_norm": 0.04051383212208748,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 104415
+    },
+    {
+      "epoch": 10.723836910752798,
+      "grad_norm": 0.06004336103796959,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 104418
+    },
+    {
+      "epoch": 10.72414501386464,
+      "grad_norm": 0.03955170512199402,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 104421
+    },
+    {
+      "epoch": 10.724453116976482,
+      "grad_norm": 0.031162617728114128,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 104424
+    },
+    {
+      "epoch": 10.724761220088324,
+      "grad_norm": 0.03177347779273987,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 104427
+    },
+    {
+      "epoch": 10.725069323200165,
+      "grad_norm": 0.04968690127134323,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 104430
+    },
+    {
+      "epoch": 10.725377426312006,
+      "grad_norm": 0.05992182344198227,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 104433
+    },
+    {
+      "epoch": 10.725685529423847,
+      "grad_norm": 0.042911652475595474,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 104436
+    },
+    {
+      "epoch": 10.725993632535689,
+      "grad_norm": 0.07481258362531662,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 104439
+    },
+    {
+      "epoch": 10.72630173564753,
+      "grad_norm": 0.192671537399292,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 104442
+    },
+    {
+      "epoch": 10.726609838759371,
+      "grad_norm": 0.08988282084465027,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 104445
+    },
+    {
+      "epoch": 10.726917941871212,
+      "grad_norm": 0.04226021468639374,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 104448
+    },
+    {
+      "epoch": 10.727226044983054,
+      "grad_norm": 0.04570133611559868,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 104451
+    },
+    {
+      "epoch": 10.727534148094897,
+      "grad_norm": 0.05308179929852486,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 104454
+    },
+    {
+      "epoch": 10.727842251206738,
+      "grad_norm": 0.03369840979576111,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 104457
+    },
+    {
+      "epoch": 10.72815035431858,
+      "grad_norm": 0.05607397481799126,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 104460
+    },
+    {
+      "epoch": 10.72845845743042,
+      "grad_norm": 0.0777030736207962,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 104463
+    },
+    {
+      "epoch": 10.728766560542262,
+      "grad_norm": 0.055943794548511505,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 104466
+    },
+    {
+      "epoch": 10.729074663654103,
+      "grad_norm": 0.03508768975734711,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 104469
+    },
+    {
+      "epoch": 10.729382766765944,
+      "grad_norm": 0.06143694743514061,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 104472
+    },
+    {
+      "epoch": 10.729690869877786,
+      "grad_norm": 0.051577091217041016,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 104475
+    },
+    {
+      "epoch": 10.729998972989627,
+      "grad_norm": 0.048224084079265594,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 104478
+    },
+    {
+      "epoch": 10.730307076101468,
+      "grad_norm": 0.17025841772556305,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 104481
+    },
+    {
+      "epoch": 10.73061517921331,
+      "grad_norm": 0.0435960553586483,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 104484
+    },
+    {
+      "epoch": 10.73092328232515,
+      "grad_norm": 0.035013701766729355,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 104487
+    },
+    {
+      "epoch": 10.731231385436994,
+      "grad_norm": 0.03989393264055252,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 104490
+    },
+    {
+      "epoch": 10.731539488548835,
+      "grad_norm": 0.04265978932380676,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 104493
+    },
+    {
+      "epoch": 10.731847591660676,
+      "grad_norm": 0.04900422692298889,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 104496
+    },
+    {
+      "epoch": 10.732155694772517,
+      "grad_norm": 0.040579188615083694,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 104499
+    },
+    {
+      "epoch": 10.732463797884359,
+      "grad_norm": 0.04971577972173691,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 104502
+    },
+    {
+      "epoch": 10.7327719009962,
+      "grad_norm": 0.04374386742711067,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 104505
+    },
+    {
+      "epoch": 10.733080004108041,
+      "grad_norm": 0.0434911847114563,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 104508
+    },
+    {
+      "epoch": 10.733388107219882,
+      "grad_norm": 0.04322110116481781,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 104511
+    },
+    {
+      "epoch": 10.733696210331724,
+      "grad_norm": 0.08954337239265442,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 104514
+    },
+    {
+      "epoch": 10.734004313443565,
+      "grad_norm": 0.09366462379693985,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 104517
+    },
+    {
+      "epoch": 10.734312416555408,
+      "grad_norm": 0.09955374151468277,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 104520
+    },
+    {
+      "epoch": 10.73462051966725,
+      "grad_norm": 0.08628708124160767,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 104523
+    },
+    {
+      "epoch": 10.73492862277909,
+      "grad_norm": 0.0417771115899086,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 104526
+    },
+    {
+      "epoch": 10.735236725890932,
+      "grad_norm": 0.041715867817401886,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 104529
+    },
+    {
+      "epoch": 10.735544829002773,
+      "grad_norm": 0.043204814195632935,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 104532
+    },
+    {
+      "epoch": 10.735852932114614,
+      "grad_norm": 0.04433635249733925,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 104535
+    },
+    {
+      "epoch": 10.736161035226456,
+      "grad_norm": 0.09020950645208359,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 104538
+    },
+    {
+      "epoch": 10.736469138338297,
+      "grad_norm": 0.09889649599790573,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 104541
+    },
+    {
+      "epoch": 10.736777241450138,
+      "grad_norm": 0.0474759116768837,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 104544
+    },
+    {
+      "epoch": 10.73708534456198,
+      "grad_norm": 0.04423435404896736,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 104547
+    },
+    {
+      "epoch": 10.73739344767382,
+      "grad_norm": 0.07683929055929184,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 104550
+    },
+    {
+      "epoch": 10.737701550785664,
+      "grad_norm": 0.06812088936567307,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 104553
+    },
+    {
+      "epoch": 10.738009653897505,
+      "grad_norm": 0.04490400478243828,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 104556
+    },
+    {
+      "epoch": 10.738317757009346,
+      "grad_norm": 0.03471510857343674,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 104559
+    },
+    {
+      "epoch": 10.738625860121187,
+      "grad_norm": 0.06390637904405594,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 104562
+    },
+    {
+      "epoch": 10.738933963233029,
+      "grad_norm": 0.0859960988163948,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 104565
+    },
+    {
+      "epoch": 10.73924206634487,
+      "grad_norm": 0.052980903536081314,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 104568
+    },
+    {
+      "epoch": 10.739550169456711,
+      "grad_norm": 0.10932338237762451,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 104571
+    },
+    {
+      "epoch": 10.739858272568553,
+      "grad_norm": 0.09292354434728622,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 104574
+    },
+    {
+      "epoch": 10.740166375680394,
+      "grad_norm": 0.06333106756210327,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 104577
+    },
+    {
+      "epoch": 10.740474478792235,
+      "grad_norm": 0.06950338929891586,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 104580
+    },
+    {
+      "epoch": 10.740782581904078,
+      "grad_norm": 0.04017019271850586,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 104583
+    },
+    {
+      "epoch": 10.74109068501592,
+      "grad_norm": 0.09870757907629013,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 104586
+    },
+    {
+      "epoch": 10.74139878812776,
+      "grad_norm": 0.07882300764322281,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 104589
+    },
+    {
+      "epoch": 10.741706891239602,
+      "grad_norm": 0.06481506675481796,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 104592
+    },
+    {
+      "epoch": 10.742014994351443,
+      "grad_norm": 0.05816543847322464,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 104595
+    },
+    {
+      "epoch": 10.742323097463284,
+      "grad_norm": 0.046804074198007584,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 104598
+    },
+    {
+      "epoch": 10.742631200575126,
+      "grad_norm": 0.0676804855465889,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 104601
+    },
+    {
+      "epoch": 10.742939303686967,
+      "grad_norm": 0.06489752978086472,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 104604
+    },
+    {
+      "epoch": 10.743247406798808,
+      "grad_norm": 0.033941950649023056,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 104607
+    },
+    {
+      "epoch": 10.74355550991065,
+      "grad_norm": 0.1338924765586853,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 104610
+    },
+    {
+      "epoch": 10.74386361302249,
+      "grad_norm": 0.04623845964670181,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 104613
+    },
+    {
+      "epoch": 10.744171716134334,
+      "grad_norm": 0.07495556771755219,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 104616
+    },
+    {
+      "epoch": 10.744479819246175,
+      "grad_norm": 0.07099320739507675,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 104619
+    },
+    {
+      "epoch": 10.744787922358016,
+      "grad_norm": 0.03660536929965019,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 104622
+    },
+    {
+      "epoch": 10.745096025469858,
+      "grad_norm": 0.06336542963981628,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 104625
+    },
+    {
+      "epoch": 10.745404128581699,
+      "grad_norm": 0.09526252746582031,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 104628
+    },
+    {
+      "epoch": 10.74571223169354,
+      "grad_norm": 0.04167678952217102,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 104631
+    },
+    {
+      "epoch": 10.746020334805381,
+      "grad_norm": 0.049752864986658096,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 104634
+    },
+    {
+      "epoch": 10.746328437917223,
+      "grad_norm": 0.10626377910375595,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 104637
+    },
+    {
+      "epoch": 10.746636541029064,
+      "grad_norm": 0.08697515726089478,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 104640
+    },
+    {
+      "epoch": 10.746944644140905,
+      "grad_norm": 0.07837111502885818,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 104643
+    },
+    {
+      "epoch": 10.747252747252748,
+      "grad_norm": 0.056924059987068176,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 104646
+    },
+    {
+      "epoch": 10.74756085036459,
+      "grad_norm": 0.05303655564785004,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 104649
+    },
+    {
+      "epoch": 10.74786895347643,
+      "grad_norm": 0.043836649507284164,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 104652
+    },
+    {
+      "epoch": 10.748177056588272,
+      "grad_norm": 0.06553082168102264,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 104655
+    },
+    {
+      "epoch": 10.748485159700113,
+      "grad_norm": 0.05108625441789627,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 104658
+    },
+    {
+      "epoch": 10.748793262811954,
+      "grad_norm": 0.04943982884287834,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 104661
+    },
+    {
+      "epoch": 10.749101365923796,
+      "grad_norm": 0.06580820679664612,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 104664
+    },
+    {
+      "epoch": 10.749409469035637,
+      "grad_norm": 0.09794413298368454,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 104667
+    },
+    {
+      "epoch": 10.749717572147478,
+      "grad_norm": 0.08265780657529831,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 104670
+    },
+    {
+      "epoch": 10.75002567525932,
+      "grad_norm": 0.11228012293577194,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 104673
+    },
+    {
+      "epoch": 10.75033377837116,
+      "grad_norm": 0.054208461195230484,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 104676
+    },
+    {
+      "epoch": 10.750641881483002,
+      "grad_norm": 0.1329595148563385,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 104679
+    },
+    {
+      "epoch": 10.750949984594845,
+      "grad_norm": 0.08030585944652557,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 104682
+    },
+    {
+      "epoch": 10.751258087706686,
+      "grad_norm": 0.05090648680925369,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 104685
+    },
+    {
+      "epoch": 10.751566190818528,
+      "grad_norm": 0.05258609354496002,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 104688
+    },
+    {
+      "epoch": 10.751874293930369,
+      "grad_norm": 0.05537552759051323,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 104691
+    },
+    {
+      "epoch": 10.75218239704221,
+      "grad_norm": 0.043664395809173584,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 104694
+    },
+    {
+      "epoch": 10.752490500154051,
+      "grad_norm": 0.03498004749417305,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 104697
+    },
+    {
+      "epoch": 10.752798603265893,
+      "grad_norm": 0.03816048428416252,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 104700
+    },
+    {
+      "epoch": 10.753106706377734,
+      "grad_norm": 0.11150369048118591,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 104703
+    },
+    {
+      "epoch": 10.753414809489575,
+      "grad_norm": 0.108778215944767,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 104706
+    },
+    {
+      "epoch": 10.753722912601418,
+      "grad_norm": 0.05636920407414436,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 104709
+    },
+    {
+      "epoch": 10.75403101571326,
+      "grad_norm": 0.038366153836250305,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 104712
+    },
+    {
+      "epoch": 10.7543391188251,
+      "grad_norm": 0.07660912722349167,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 104715
+    },
+    {
+      "epoch": 10.754647221936942,
+      "grad_norm": 0.06754130125045776,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 104718
+    },
+    {
+      "epoch": 10.754955325048783,
+      "grad_norm": 0.09265803545713425,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 104721
+    },
+    {
+      "epoch": 10.755263428160625,
+      "grad_norm": 0.06105842813849449,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 104724
+    },
+    {
+      "epoch": 10.755571531272466,
+      "grad_norm": 0.07468662410974503,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 104727
+    },
+    {
+      "epoch": 10.755879634384307,
+      "grad_norm": 0.05258229002356529,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 104730
+    },
+    {
+      "epoch": 10.756187737496148,
+      "grad_norm": 0.0563662014901638,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 104733
+    },
+    {
+      "epoch": 10.75649584060799,
+      "grad_norm": 0.04614382982254028,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 104736
+    },
+    {
+      "epoch": 10.75680394371983,
+      "grad_norm": 0.04521797597408295,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 104739
+    },
+    {
+      "epoch": 10.757112046831672,
+      "grad_norm": 0.05099884420633316,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 104742
+    },
+    {
+      "epoch": 10.757420149943515,
+      "grad_norm": 0.05011932551860809,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 104745
+    },
+    {
+      "epoch": 10.757728253055356,
+      "grad_norm": 0.15515439212322235,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 104748
+    },
+    {
+      "epoch": 10.758036356167198,
+      "grad_norm": 0.04284700006246567,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 104751
+    },
+    {
+      "epoch": 10.758344459279039,
+      "grad_norm": 0.045263636857271194,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 104754
+    },
+    {
+      "epoch": 10.75865256239088,
+      "grad_norm": 0.0417122021317482,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 104757
+    },
+    {
+      "epoch": 10.758960665502721,
+      "grad_norm": 0.06633123755455017,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 104760
+    },
+    {
+      "epoch": 10.759268768614563,
+      "grad_norm": 0.043190255761146545,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 104763
+    },
+    {
+      "epoch": 10.759576871726404,
+      "grad_norm": 0.031169939786195755,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 104766
+    },
+    {
+      "epoch": 10.759884974838245,
+      "grad_norm": 0.03632303699851036,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 104769
+    },
+    {
+      "epoch": 10.760193077950087,
+      "grad_norm": 0.07166427373886108,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 104772
+    },
+    {
+      "epoch": 10.76050118106193,
+      "grad_norm": 0.06929494440555573,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 104775
+    },
+    {
+      "epoch": 10.76080928417377,
+      "grad_norm": 0.05270274356007576,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 104778
+    },
+    {
+      "epoch": 10.761117387285612,
+      "grad_norm": 0.1968749314546585,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 104781
+    },
+    {
+      "epoch": 10.761425490397453,
+      "grad_norm": 0.08403410017490387,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 104784
+    },
+    {
+      "epoch": 10.761733593509295,
+      "grad_norm": 0.04399842768907547,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 104787
+    },
+    {
+      "epoch": 10.762041696621136,
+      "grad_norm": 0.05895080417394638,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 104790
+    },
+    {
+      "epoch": 10.762349799732977,
+      "grad_norm": 0.060437798500061035,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 104793
+    },
+    {
+      "epoch": 10.762657902844818,
+      "grad_norm": 0.06297816336154938,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 104796
+    },
+    {
+      "epoch": 10.76296600595666,
+      "grad_norm": 0.06990213692188263,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 104799
+    },
+    {
+      "epoch": 10.763274109068501,
+      "grad_norm": 0.044904761016368866,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 104802
+    },
+    {
+      "epoch": 10.763582212180342,
+      "grad_norm": 0.05253970995545387,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 104805
+    },
+    {
+      "epoch": 10.763890315292185,
+      "grad_norm": 0.03301846608519554,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 104808
+    },
+    {
+      "epoch": 10.764198418404026,
+      "grad_norm": 0.04922977834939957,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 104811
+    },
+    {
+      "epoch": 10.764506521515868,
+      "grad_norm": 0.03909482806921005,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 104814
+    },
+    {
+      "epoch": 10.764814624627709,
+      "grad_norm": 0.04505687952041626,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 104817
+    },
+    {
+      "epoch": 10.76512272773955,
+      "grad_norm": 0.0761536955833435,
+      "learning_rate": 0.01,
+      "loss": 1.8965,
+      "step": 104820
+    },
+    {
+      "epoch": 10.765430830851392,
+      "grad_norm": 0.08779765665531158,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 104823
+    },
+    {
+      "epoch": 10.765738933963233,
+      "grad_norm": 0.16538570821285248,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 104826
+    },
+    {
+      "epoch": 10.766047037075074,
+      "grad_norm": 0.08162948489189148,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 104829
+    },
+    {
+      "epoch": 10.766355140186915,
+      "grad_norm": 0.07075890153646469,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 104832
+    },
+    {
+      "epoch": 10.766663243298757,
+      "grad_norm": 0.040055204182863235,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 104835
+    },
+    {
+      "epoch": 10.7669713464106,
+      "grad_norm": 0.03459317237138748,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 104838
+    },
+    {
+      "epoch": 10.76727944952244,
+      "grad_norm": 0.06219128519296646,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 104841
+    },
+    {
+      "epoch": 10.767587552634282,
+      "grad_norm": 0.06373794376850128,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 104844
+    },
+    {
+      "epoch": 10.767895655746123,
+      "grad_norm": 0.05703836679458618,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 104847
+    },
+    {
+      "epoch": 10.768203758857965,
+      "grad_norm": 0.03425431251525879,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 104850
+    },
+    {
+      "epoch": 10.768511861969806,
+      "grad_norm": 0.04029223695397377,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 104853
+    },
+    {
+      "epoch": 10.768819965081647,
+      "grad_norm": 0.04060615599155426,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 104856
+    },
+    {
+      "epoch": 10.769128068193488,
+      "grad_norm": 0.03595336154103279,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 104859
+    },
+    {
+      "epoch": 10.76943617130533,
+      "grad_norm": 0.046418748795986176,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 104862
+    },
+    {
+      "epoch": 10.769744274417171,
+      "grad_norm": 0.05077604576945305,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 104865
+    },
+    {
+      "epoch": 10.770052377529012,
+      "grad_norm": 0.1307033896446228,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 104868
+    },
+    {
+      "epoch": 10.770360480640855,
+      "grad_norm": 0.03684141859412193,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 104871
+    },
+    {
+      "epoch": 10.770668583752697,
+      "grad_norm": 0.06678260862827301,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 104874
+    },
+    {
+      "epoch": 10.770976686864538,
+      "grad_norm": 0.037949930876493454,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 104877
+    },
+    {
+      "epoch": 10.771284789976379,
+      "grad_norm": 0.03636316582560539,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 104880
+    },
+    {
+      "epoch": 10.77159289308822,
+      "grad_norm": 0.07280274480581284,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 104883
+    },
+    {
+      "epoch": 10.771900996200062,
+      "grad_norm": 0.09522299468517303,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 104886
+    },
+    {
+      "epoch": 10.772209099311903,
+      "grad_norm": 0.08580028265714645,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 104889
+    },
+    {
+      "epoch": 10.772517202423744,
+      "grad_norm": 0.08504713326692581,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 104892
+    },
+    {
+      "epoch": 10.772825305535585,
+      "grad_norm": 0.05523672699928284,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 104895
+    },
+    {
+      "epoch": 10.773133408647427,
+      "grad_norm": 0.08716922998428345,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 104898
+    },
+    {
+      "epoch": 10.77344151175927,
+      "grad_norm": 0.06506255269050598,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 104901
+    },
+    {
+      "epoch": 10.773749614871111,
+      "grad_norm": 0.11295832693576813,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 104904
+    },
+    {
+      "epoch": 10.774057717982952,
+      "grad_norm": 0.03108265809714794,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 104907
+    },
+    {
+      "epoch": 10.774365821094793,
+      "grad_norm": 0.04045362398028374,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 104910
+    },
+    {
+      "epoch": 10.774673924206635,
+      "grad_norm": 0.09854461252689362,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 104913
+    },
+    {
+      "epoch": 10.774982027318476,
+      "grad_norm": 0.10017503798007965,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 104916
+    },
+    {
+      "epoch": 10.775290130430317,
+      "grad_norm": 0.04605772718787193,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 104919
+    },
+    {
+      "epoch": 10.775598233542159,
+      "grad_norm": 0.04422129690647125,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 104922
+    },
+    {
+      "epoch": 10.775906336654,
+      "grad_norm": 0.03512374684214592,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 104925
+    },
+    {
+      "epoch": 10.776214439765841,
+      "grad_norm": 0.05168703570961952,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 104928
+    },
+    {
+      "epoch": 10.776522542877682,
+      "grad_norm": 0.04373502731323242,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 104931
+    },
+    {
+      "epoch": 10.776830645989524,
+      "grad_norm": 0.1377280056476593,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 104934
+    },
+    {
+      "epoch": 10.777138749101367,
+      "grad_norm": 0.09566672146320343,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 104937
+    },
+    {
+      "epoch": 10.777446852213208,
+      "grad_norm": 0.055807486176490784,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 104940
+    },
+    {
+      "epoch": 10.77775495532505,
+      "grad_norm": 0.037279289215803146,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 104943
+    },
+    {
+      "epoch": 10.77806305843689,
+      "grad_norm": 0.04037223383784294,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 104946
+    },
+    {
+      "epoch": 10.778371161548732,
+      "grad_norm": 0.04375686123967171,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 104949
+    },
+    {
+      "epoch": 10.778679264660573,
+      "grad_norm": 0.10367769747972488,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 104952
+    },
+    {
+      "epoch": 10.778987367772414,
+      "grad_norm": 0.06901142001152039,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 104955
+    },
+    {
+      "epoch": 10.779295470884255,
+      "grad_norm": 0.05770748108625412,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 104958
+    },
+    {
+      "epoch": 10.779603573996097,
+      "grad_norm": 0.054886963218450546,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 104961
+    },
+    {
+      "epoch": 10.779911677107938,
+      "grad_norm": 0.050134312361478806,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 104964
+    },
+    {
+      "epoch": 10.780219780219781,
+      "grad_norm": 0.04675987735390663,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 104967
+    },
+    {
+      "epoch": 10.780527883331622,
+      "grad_norm": 0.09763073921203613,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 104970
+    },
+    {
+      "epoch": 10.780835986443464,
+      "grad_norm": 0.04318242520093918,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 104973
+    },
+    {
+      "epoch": 10.781144089555305,
+      "grad_norm": 0.039554860442876816,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 104976
+    },
+    {
+      "epoch": 10.781452192667146,
+      "grad_norm": 0.04658977687358856,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 104979
+    },
+    {
+      "epoch": 10.781760295778987,
+      "grad_norm": 0.036048904061317444,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 104982
+    },
+    {
+      "epoch": 10.782068398890829,
+      "grad_norm": 0.08908376097679138,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 104985
+    },
+    {
+      "epoch": 10.78237650200267,
+      "grad_norm": 0.11830753833055496,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 104988
+    },
+    {
+      "epoch": 10.782684605114511,
+      "grad_norm": 0.13836215436458588,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 104991
+    },
+    {
+      "epoch": 10.782992708226352,
+      "grad_norm": 0.06564682722091675,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 104994
+    },
+    {
+      "epoch": 10.783300811338194,
+      "grad_norm": 0.10762113332748413,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 104997
+    },
+    {
+      "epoch": 10.783608914450037,
+      "grad_norm": 0.04483092948794365,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 105000
+    },
+    {
+      "epoch": 10.783917017561878,
+      "grad_norm": 0.05532277747988701,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 105003
+    },
+    {
+      "epoch": 10.78422512067372,
+      "grad_norm": 0.04673232138156891,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 105006
+    },
+    {
+      "epoch": 10.78453322378556,
+      "grad_norm": 0.07055231183767319,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 105009
+    },
+    {
+      "epoch": 10.784841326897402,
+      "grad_norm": 0.05611100792884827,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 105012
+    },
+    {
+      "epoch": 10.785149430009243,
+      "grad_norm": 0.1283992975950241,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 105015
+    },
+    {
+      "epoch": 10.785457533121084,
+      "grad_norm": 0.16754893958568573,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 105018
+    },
+    {
+      "epoch": 10.785765636232926,
+      "grad_norm": 0.08787377923727036,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 105021
+    },
+    {
+      "epoch": 10.786073739344767,
+      "grad_norm": 0.08007776737213135,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 105024
+    },
+    {
+      "epoch": 10.786381842456608,
+      "grad_norm": 0.07712419331073761,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 105027
+    },
+    {
+      "epoch": 10.786689945568451,
+      "grad_norm": 0.08709436655044556,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 105030
+    },
+    {
+      "epoch": 10.786998048680292,
+      "grad_norm": 0.06541412323713303,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 105033
+    },
+    {
+      "epoch": 10.787306151792134,
+      "grad_norm": 0.11988494545221329,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 105036
+    },
+    {
+      "epoch": 10.787614254903975,
+      "grad_norm": 0.12408696115016937,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 105039
+    },
+    {
+      "epoch": 10.787922358015816,
+      "grad_norm": 0.05449619144201279,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 105042
+    },
+    {
+      "epoch": 10.788230461127657,
+      "grad_norm": 0.04881265014410019,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 105045
+    },
+    {
+      "epoch": 10.788538564239499,
+      "grad_norm": 0.05132092535495758,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 105048
+    },
+    {
+      "epoch": 10.78884666735134,
+      "grad_norm": 0.06299763172864914,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 105051
+    },
+    {
+      "epoch": 10.789154770463181,
+      "grad_norm": 0.0902947410941124,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 105054
+    },
+    {
+      "epoch": 10.789462873575022,
+      "grad_norm": 0.045308105647563934,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 105057
+    },
+    {
+      "epoch": 10.789770976686864,
+      "grad_norm": 0.09681692719459534,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 105060
+    },
+    {
+      "epoch": 10.790079079798707,
+      "grad_norm": 0.08720479160547256,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 105063
+    },
+    {
+      "epoch": 10.790387182910548,
+      "grad_norm": 0.06638406217098236,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 105066
+    },
+    {
+      "epoch": 10.79069528602239,
+      "grad_norm": 0.09998981654644012,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 105069
+    },
+    {
+      "epoch": 10.79100338913423,
+      "grad_norm": 0.05286186933517456,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 105072
+    },
+    {
+      "epoch": 10.791311492246072,
+      "grad_norm": 0.041253458708524704,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 105075
+    },
+    {
+      "epoch": 10.791619595357913,
+      "grad_norm": 0.0601825937628746,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 105078
+    },
+    {
+      "epoch": 10.791927698469754,
+      "grad_norm": 0.03299073502421379,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 105081
+    },
+    {
+      "epoch": 10.792235801581596,
+      "grad_norm": 0.10112438350915909,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 105084
+    },
+    {
+      "epoch": 10.792543904693437,
+      "grad_norm": 0.0791049674153328,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 105087
+    },
+    {
+      "epoch": 10.792852007805278,
+      "grad_norm": 0.09565617889165878,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 105090
+    },
+    {
+      "epoch": 10.793160110917121,
+      "grad_norm": 0.0910487100481987,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 105093
+    },
+    {
+      "epoch": 10.793468214028962,
+      "grad_norm": 0.10622784495353699,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 105096
+    },
+    {
+      "epoch": 10.793776317140804,
+      "grad_norm": 0.0922565832734108,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 105099
+    },
+    {
+      "epoch": 10.794084420252645,
+      "grad_norm": 0.12033677101135254,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 105102
+    },
+    {
+      "epoch": 10.794392523364486,
+      "grad_norm": 0.10453055799007416,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 105105
+    },
+    {
+      "epoch": 10.794700626476327,
+      "grad_norm": 0.07022013515233994,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 105108
+    },
+    {
+      "epoch": 10.795008729588169,
+      "grad_norm": 0.05361005663871765,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 105111
+    },
+    {
+      "epoch": 10.79531683270001,
+      "grad_norm": 0.034731000661849976,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 105114
+    },
+    {
+      "epoch": 10.795624935811851,
+      "grad_norm": 0.06049367040395737,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 105117
+    },
+    {
+      "epoch": 10.795933038923692,
+      "grad_norm": 0.0643046572804451,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 105120
+    },
+    {
+      "epoch": 10.796241142035534,
+      "grad_norm": 0.041732240468263626,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 105123
+    },
+    {
+      "epoch": 10.796549245147377,
+      "grad_norm": 0.10261611640453339,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 105126
+    },
+    {
+      "epoch": 10.796857348259218,
+      "grad_norm": 0.0781281590461731,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 105129
+    },
+    {
+      "epoch": 10.79716545137106,
+      "grad_norm": 0.04957817494869232,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 105132
+    },
+    {
+      "epoch": 10.7974735544829,
+      "grad_norm": 0.05353771522641182,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 105135
+    },
+    {
+      "epoch": 10.797781657594742,
+      "grad_norm": 0.048365335911512375,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 105138
+    },
+    {
+      "epoch": 10.798089760706583,
+      "grad_norm": 0.0623091422021389,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 105141
+    },
+    {
+      "epoch": 10.798397863818424,
+      "grad_norm": 0.05436871945858002,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 105144
+    },
+    {
+      "epoch": 10.798705966930266,
+      "grad_norm": 0.05013308674097061,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 105147
+    },
+    {
+      "epoch": 10.799014070042107,
+      "grad_norm": 0.04122906178236008,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 105150
+    },
+    {
+      "epoch": 10.799322173153948,
+      "grad_norm": 0.04418342188000679,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 105153
+    },
+    {
+      "epoch": 10.799630276265791,
+      "grad_norm": 0.05578082427382469,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 105156
+    },
+    {
+      "epoch": 10.799938379377632,
+      "grad_norm": 0.1632402539253235,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 105159
+    },
+    {
+      "epoch": 10.800246482489474,
+      "grad_norm": 0.06019952893257141,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 105162
+    },
+    {
+      "epoch": 10.800554585601315,
+      "grad_norm": 0.06759855896234512,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 105165
+    },
+    {
+      "epoch": 10.800862688713156,
+      "grad_norm": 0.03876945376396179,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 105168
+    },
+    {
+      "epoch": 10.801170791824998,
+      "grad_norm": 0.09414643049240112,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 105171
+    },
+    {
+      "epoch": 10.801478894936839,
+      "grad_norm": 0.04190301150083542,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 105174
+    },
+    {
+      "epoch": 10.80178699804868,
+      "grad_norm": 0.05320903658866882,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 105177
+    },
+    {
+      "epoch": 10.802095101160521,
+      "grad_norm": 0.056490443646907806,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 105180
+    },
+    {
+      "epoch": 10.802403204272363,
+      "grad_norm": 0.04092302918434143,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 105183
+    },
+    {
+      "epoch": 10.802711307384204,
+      "grad_norm": 0.040519844740629196,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 105186
+    },
+    {
+      "epoch": 10.803019410496045,
+      "grad_norm": 0.04397949203848839,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 105189
+    },
+    {
+      "epoch": 10.803327513607888,
+      "grad_norm": 0.0402667336165905,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 105192
+    },
+    {
+      "epoch": 10.80363561671973,
+      "grad_norm": 0.05854826420545578,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 105195
+    },
+    {
+      "epoch": 10.80394371983157,
+      "grad_norm": 0.05287005379796028,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 105198
+    },
+    {
+      "epoch": 10.804251822943412,
+      "grad_norm": 0.03999342396855354,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 105201
+    },
+    {
+      "epoch": 10.804559926055253,
+      "grad_norm": 0.10459694266319275,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 105204
+    },
+    {
+      "epoch": 10.804868029167094,
+      "grad_norm": 0.03152311593294144,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 105207
+    },
+    {
+      "epoch": 10.805176132278936,
+      "grad_norm": 0.05782946199178696,
+      "learning_rate": 0.01,
+      "loss": 1.9661,
+      "step": 105210
+    },
+    {
+      "epoch": 10.805484235390777,
+      "grad_norm": 0.05249985307455063,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 105213
+    },
+    {
+      "epoch": 10.805792338502618,
+      "grad_norm": 0.05483485013246536,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 105216
+    },
+    {
+      "epoch": 10.80610044161446,
+      "grad_norm": 0.051782358437776566,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 105219
+    },
+    {
+      "epoch": 10.806408544726303,
+      "grad_norm": 0.09521941095590591,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 105222
+    },
+    {
+      "epoch": 10.806716647838144,
+      "grad_norm": 0.06587459146976471,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 105225
+    },
+    {
+      "epoch": 10.807024750949985,
+      "grad_norm": 0.049619127064943314,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 105228
+    },
+    {
+      "epoch": 10.807332854061826,
+      "grad_norm": 0.113286092877388,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 105231
+    },
+    {
+      "epoch": 10.807640957173668,
+      "grad_norm": 0.09651494026184082,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 105234
+    },
+    {
+      "epoch": 10.807949060285509,
+      "grad_norm": 0.04343407601118088,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 105237
+    },
+    {
+      "epoch": 10.80825716339735,
+      "grad_norm": 0.043445128947496414,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 105240
+    },
+    {
+      "epoch": 10.808565266509191,
+      "grad_norm": 0.05014490336179733,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 105243
+    },
+    {
+      "epoch": 10.808873369621033,
+      "grad_norm": 0.03343608230352402,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 105246
+    },
+    {
+      "epoch": 10.809181472732874,
+      "grad_norm": 0.07791464030742645,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 105249
+    },
+    {
+      "epoch": 10.809489575844715,
+      "grad_norm": 0.05957683175802231,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 105252
+    },
+    {
+      "epoch": 10.809797678956558,
+      "grad_norm": 0.10276157408952713,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 105255
+    },
+    {
+      "epoch": 10.8101057820684,
+      "grad_norm": 0.07973282784223557,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 105258
+    },
+    {
+      "epoch": 10.81041388518024,
+      "grad_norm": 0.08871182054281235,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 105261
+    },
+    {
+      "epoch": 10.810721988292082,
+      "grad_norm": 0.060441359877586365,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 105264
+    },
+    {
+      "epoch": 10.811030091403923,
+      "grad_norm": 0.04853738099336624,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 105267
+    },
+    {
+      "epoch": 10.811338194515764,
+      "grad_norm": 0.0857999175786972,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 105270
+    },
+    {
+      "epoch": 10.811646297627606,
+      "grad_norm": 0.05908980965614319,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 105273
+    },
+    {
+      "epoch": 10.811954400739447,
+      "grad_norm": 0.059544436633586884,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 105276
+    },
+    {
+      "epoch": 10.812262503851288,
+      "grad_norm": 0.050158899277448654,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 105279
+    },
+    {
+      "epoch": 10.81257060696313,
+      "grad_norm": 0.04424465075135231,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 105282
+    },
+    {
+      "epoch": 10.812878710074973,
+      "grad_norm": 0.04916578158736229,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 105285
+    },
+    {
+      "epoch": 10.813186813186814,
+      "grad_norm": 0.11380893737077713,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 105288
+    },
+    {
+      "epoch": 10.813494916298655,
+      "grad_norm": 0.15234796702861786,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 105291
+    },
+    {
+      "epoch": 10.813803019410496,
+      "grad_norm": 0.05364861339330673,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 105294
+    },
+    {
+      "epoch": 10.814111122522338,
+      "grad_norm": 0.05904761701822281,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 105297
+    },
+    {
+      "epoch": 10.814419225634179,
+      "grad_norm": 0.07022589445114136,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 105300
+    },
+    {
+      "epoch": 10.81472732874602,
+      "grad_norm": 0.07958179712295532,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 105303
+    },
+    {
+      "epoch": 10.815035431857861,
+      "grad_norm": 0.08987373113632202,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 105306
+    },
+    {
+      "epoch": 10.815343534969703,
+      "grad_norm": 0.04922526329755783,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 105309
+    },
+    {
+      "epoch": 10.815651638081544,
+      "grad_norm": 0.0366976335644722,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 105312
+    },
+    {
+      "epoch": 10.815959741193385,
+      "grad_norm": 0.04322095960378647,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 105315
+    },
+    {
+      "epoch": 10.816267844305228,
+      "grad_norm": 0.07414691895246506,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 105318
+    },
+    {
+      "epoch": 10.81657594741707,
+      "grad_norm": 0.06225109100341797,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 105321
+    },
+    {
+      "epoch": 10.81688405052891,
+      "grad_norm": 0.07541634887456894,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 105324
+    },
+    {
+      "epoch": 10.817192153640752,
+      "grad_norm": 0.033274658024311066,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 105327
+    },
+    {
+      "epoch": 10.817500256752593,
+      "grad_norm": 0.0380442850291729,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 105330
+    },
+    {
+      "epoch": 10.817808359864435,
+      "grad_norm": 0.15397582948207855,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 105333
+    },
+    {
+      "epoch": 10.818116462976276,
+      "grad_norm": 0.08010608702898026,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 105336
+    },
+    {
+      "epoch": 10.818424566088117,
+      "grad_norm": 0.07294953614473343,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 105339
+    },
+    {
+      "epoch": 10.818732669199958,
+      "grad_norm": 0.042549289762973785,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 105342
+    },
+    {
+      "epoch": 10.8190407723118,
+      "grad_norm": 0.03370673209428787,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 105345
+    },
+    {
+      "epoch": 10.819348875423643,
+      "grad_norm": 0.046656735241413116,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 105348
+    },
+    {
+      "epoch": 10.819656978535484,
+      "grad_norm": 0.0739334300160408,
+      "learning_rate": 0.01,
+      "loss": 1.9814,
+      "step": 105351
+    },
+    {
+      "epoch": 10.819965081647325,
+      "grad_norm": 0.047819964587688446,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 105354
+    },
+    {
+      "epoch": 10.820273184759166,
+      "grad_norm": 0.06027652695775032,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 105357
+    },
+    {
+      "epoch": 10.820581287871008,
+      "grad_norm": 0.1334407478570938,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 105360
+    },
+    {
+      "epoch": 10.820889390982849,
+      "grad_norm": 0.04469862952828407,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 105363
+    },
+    {
+      "epoch": 10.82119749409469,
+      "grad_norm": 0.03328389301896095,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 105366
+    },
+    {
+      "epoch": 10.821505597206531,
+      "grad_norm": 0.0356464758515358,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 105369
+    },
+    {
+      "epoch": 10.821813700318373,
+      "grad_norm": 0.04654991999268532,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 105372
+    },
+    {
+      "epoch": 10.822121803430214,
+      "grad_norm": 0.04388376325368881,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 105375
+    },
+    {
+      "epoch": 10.822429906542055,
+      "grad_norm": 0.04658830538392067,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 105378
+    },
+    {
+      "epoch": 10.822738009653898,
+      "grad_norm": 0.04263482987880707,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 105381
+    },
+    {
+      "epoch": 10.82304611276574,
+      "grad_norm": 0.07428035885095596,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 105384
+    },
+    {
+      "epoch": 10.82335421587758,
+      "grad_norm": 0.10800222307443619,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 105387
+    },
+    {
+      "epoch": 10.823662318989422,
+      "grad_norm": 0.05401541665196419,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 105390
+    },
+    {
+      "epoch": 10.823970422101263,
+      "grad_norm": 0.12794886529445648,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 105393
+    },
+    {
+      "epoch": 10.824278525213105,
+      "grad_norm": 0.0494115948677063,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 105396
+    },
+    {
+      "epoch": 10.824586628324946,
+      "grad_norm": 0.035063013434410095,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 105399
+    },
+    {
+      "epoch": 10.824894731436787,
+      "grad_norm": 0.04617554321885109,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 105402
+    },
+    {
+      "epoch": 10.825202834548628,
+      "grad_norm": 0.060868896543979645,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 105405
+    },
+    {
+      "epoch": 10.82551093766047,
+      "grad_norm": 0.035648807883262634,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 105408
+    },
+    {
+      "epoch": 10.825819040772313,
+      "grad_norm": 0.058884233236312866,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 105411
+    },
+    {
+      "epoch": 10.826127143884154,
+      "grad_norm": 0.16527080535888672,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 105414
+    },
+    {
+      "epoch": 10.826435246995995,
+      "grad_norm": 0.08593026548624039,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 105417
+    },
+    {
+      "epoch": 10.826743350107837,
+      "grad_norm": 0.05153747275471687,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 105420
+    },
+    {
+      "epoch": 10.827051453219678,
+      "grad_norm": 0.046141065657138824,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 105423
+    },
+    {
+      "epoch": 10.827359556331519,
+      "grad_norm": 0.05500772222876549,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 105426
+    },
+    {
+      "epoch": 10.82766765944336,
+      "grad_norm": 0.07828307896852493,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 105429
+    },
+    {
+      "epoch": 10.827975762555202,
+      "grad_norm": 0.0653647780418396,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 105432
+    },
+    {
+      "epoch": 10.828283865667043,
+      "grad_norm": 0.049936968833208084,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 105435
+    },
+    {
+      "epoch": 10.828591968778884,
+      "grad_norm": 0.06601990014314651,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 105438
+    },
+    {
+      "epoch": 10.828900071890725,
+      "grad_norm": 0.10459323972463608,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 105441
+    },
+    {
+      "epoch": 10.829208175002567,
+      "grad_norm": 0.06045229360461235,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 105444
+    },
+    {
+      "epoch": 10.82951627811441,
+      "grad_norm": 0.11680969595909119,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 105447
+    },
+    {
+      "epoch": 10.829824381226251,
+      "grad_norm": 0.04912514239549637,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 105450
+    },
+    {
+      "epoch": 10.830132484338092,
+      "grad_norm": 0.05238822475075722,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 105453
+    },
+    {
+      "epoch": 10.830440587449933,
+      "grad_norm": 0.03835156559944153,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 105456
+    },
+    {
+      "epoch": 10.830748690561775,
+      "grad_norm": 0.03455403074622154,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 105459
+    },
+    {
+      "epoch": 10.831056793673616,
+      "grad_norm": 0.09034903347492218,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 105462
+    },
+    {
+      "epoch": 10.831364896785457,
+      "grad_norm": 0.04471142962574959,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 105465
+    },
+    {
+      "epoch": 10.831672999897298,
+      "grad_norm": 0.04482191801071167,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 105468
+    },
+    {
+      "epoch": 10.83198110300914,
+      "grad_norm": 0.04457875341176987,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 105471
+    },
+    {
+      "epoch": 10.832289206120981,
+      "grad_norm": 0.03215394541621208,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 105474
+    },
+    {
+      "epoch": 10.832597309232824,
+      "grad_norm": 0.034922465682029724,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 105477
+    },
+    {
+      "epoch": 10.832905412344665,
+      "grad_norm": 0.11798029392957687,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 105480
+    },
+    {
+      "epoch": 10.833213515456507,
+      "grad_norm": 0.0991378054022789,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 105483
+    },
+    {
+      "epoch": 10.833521618568348,
+      "grad_norm": 0.09143664687871933,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 105486
+    },
+    {
+      "epoch": 10.833829721680189,
+      "grad_norm": 0.07089017331600189,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 105489
+    },
+    {
+      "epoch": 10.83413782479203,
+      "grad_norm": 0.05932415649294853,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 105492
+    },
+    {
+      "epoch": 10.834445927903872,
+      "grad_norm": 0.07785198092460632,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 105495
+    },
+    {
+      "epoch": 10.834754031015713,
+      "grad_norm": 0.04854836314916611,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 105498
+    },
+    {
+      "epoch": 10.835062134127554,
+      "grad_norm": 0.04837128147482872,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 105501
+    },
+    {
+      "epoch": 10.835370237239395,
+      "grad_norm": 0.055888544768095016,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 105504
+    },
+    {
+      "epoch": 10.835678340351237,
+      "grad_norm": 0.07242245227098465,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 105507
+    },
+    {
+      "epoch": 10.83598644346308,
+      "grad_norm": 0.07105785608291626,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 105510
+    },
+    {
+      "epoch": 10.836294546574921,
+      "grad_norm": 0.08170506358146667,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 105513
+    },
+    {
+      "epoch": 10.836602649686762,
+      "grad_norm": 0.11783326417207718,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 105516
+    },
+    {
+      "epoch": 10.836910752798603,
+      "grad_norm": 0.08829626441001892,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 105519
+    },
+    {
+      "epoch": 10.837218855910445,
+      "grad_norm": 0.08297731727361679,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 105522
+    },
+    {
+      "epoch": 10.837526959022286,
+      "grad_norm": 0.07735975831747055,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 105525
+    },
+    {
+      "epoch": 10.837835062134127,
+      "grad_norm": 0.07283669710159302,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 105528
+    },
+    {
+      "epoch": 10.838143165245969,
+      "grad_norm": 0.08361595869064331,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 105531
+    },
+    {
+      "epoch": 10.83845126835781,
+      "grad_norm": 0.08340589702129364,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 105534
+    },
+    {
+      "epoch": 10.838759371469651,
+      "grad_norm": 0.06210103631019592,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 105537
+    },
+    {
+      "epoch": 10.839067474581494,
+      "grad_norm": 0.04093151167035103,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 105540
+    },
+    {
+      "epoch": 10.839375577693335,
+      "grad_norm": 0.049949757754802704,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 105543
+    },
+    {
+      "epoch": 10.839683680805177,
+      "grad_norm": 0.0919983834028244,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 105546
+    },
+    {
+      "epoch": 10.839991783917018,
+      "grad_norm": 0.04312887787818909,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 105549
+    },
+    {
+      "epoch": 10.84029988702886,
+      "grad_norm": 0.03706655651330948,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 105552
+    },
+    {
+      "epoch": 10.8406079901407,
+      "grad_norm": 0.10449834913015366,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 105555
+    },
+    {
+      "epoch": 10.840916093252542,
+      "grad_norm": 0.053434476256370544,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 105558
+    },
+    {
+      "epoch": 10.841224196364383,
+      "grad_norm": 0.05519875884056091,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 105561
+    },
+    {
+      "epoch": 10.841532299476224,
+      "grad_norm": 0.10106226056814194,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 105564
+    },
+    {
+      "epoch": 10.841840402588065,
+      "grad_norm": 0.1005539521574974,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 105567
+    },
+    {
+      "epoch": 10.842148505699907,
+      "grad_norm": 0.07332777231931686,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 105570
+    },
+    {
+      "epoch": 10.84245660881175,
+      "grad_norm": 0.04059474542737007,
+      "learning_rate": 0.01,
+      "loss": 1.9031,
+      "step": 105573
+    },
+    {
+      "epoch": 10.842764711923591,
+      "grad_norm": 0.039799902588129044,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 105576
+    },
+    {
+      "epoch": 10.843072815035432,
+      "grad_norm": 0.06343264877796173,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 105579
+    },
+    {
+      "epoch": 10.843380918147274,
+      "grad_norm": 0.0510416105389595,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 105582
+    },
+    {
+      "epoch": 10.843689021259115,
+      "grad_norm": 0.033778950572013855,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 105585
+    },
+    {
+      "epoch": 10.843997124370956,
+      "grad_norm": 0.05271367356181145,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 105588
+    },
+    {
+      "epoch": 10.844305227482797,
+      "grad_norm": 0.10913360118865967,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 105591
+    },
+    {
+      "epoch": 10.844613330594639,
+      "grad_norm": 0.10924379527568817,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 105594
+    },
+    {
+      "epoch": 10.84492143370648,
+      "grad_norm": 0.0577227920293808,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 105597
+    },
+    {
+      "epoch": 10.845229536818321,
+      "grad_norm": 0.03460630401968956,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 105600
+    },
+    {
+      "epoch": 10.845537639930164,
+      "grad_norm": 0.05811109021306038,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 105603
+    },
+    {
+      "epoch": 10.845845743042005,
+      "grad_norm": 0.10903934389352798,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 105606
+    },
+    {
+      "epoch": 10.846153846153847,
+      "grad_norm": 0.07660596072673798,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 105609
+    },
+    {
+      "epoch": 10.846461949265688,
+      "grad_norm": 0.08013112097978592,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 105612
+    },
+    {
+      "epoch": 10.84677005237753,
+      "grad_norm": 0.0570279099047184,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 105615
+    },
+    {
+      "epoch": 10.84707815548937,
+      "grad_norm": 0.03606827184557915,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 105618
+    },
+    {
+      "epoch": 10.847386258601212,
+      "grad_norm": 0.06117340549826622,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 105621
+    },
+    {
+      "epoch": 10.847694361713053,
+      "grad_norm": 0.09413411468267441,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 105624
+    },
+    {
+      "epoch": 10.848002464824894,
+      "grad_norm": 0.058760952204465866,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 105627
+    },
+    {
+      "epoch": 10.848310567936736,
+      "grad_norm": 0.11886738240718842,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 105630
+    },
+    {
+      "epoch": 10.848618671048577,
+      "grad_norm": 0.08395174890756607,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 105633
+    },
+    {
+      "epoch": 10.84892677416042,
+      "grad_norm": 0.05414260923862457,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 105636
+    },
+    {
+      "epoch": 10.849234877272261,
+      "grad_norm": 0.08581097424030304,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 105639
+    },
+    {
+      "epoch": 10.849542980384102,
+      "grad_norm": 0.05798320099711418,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 105642
+    },
+    {
+      "epoch": 10.849851083495944,
+      "grad_norm": 0.0710185319185257,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 105645
+    },
+    {
+      "epoch": 10.850159186607785,
+      "grad_norm": 0.08223643898963928,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 105648
+    },
+    {
+      "epoch": 10.850467289719626,
+      "grad_norm": 0.1025887280702591,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 105651
+    },
+    {
+      "epoch": 10.850775392831467,
+      "grad_norm": 0.05991736054420471,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 105654
+    },
+    {
+      "epoch": 10.851083495943309,
+      "grad_norm": 0.08577022701501846,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 105657
+    },
+    {
+      "epoch": 10.85139159905515,
+      "grad_norm": 0.08255545049905777,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 105660
+    },
+    {
+      "epoch": 10.851699702166991,
+      "grad_norm": 0.06794632971286774,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 105663
+    },
+    {
+      "epoch": 10.852007805278834,
+      "grad_norm": 0.06520742923021317,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 105666
+    },
+    {
+      "epoch": 10.852315908390676,
+      "grad_norm": 0.10006118565797806,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 105669
+    },
+    {
+      "epoch": 10.852624011502517,
+      "grad_norm": 0.035613786429166794,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 105672
+    },
+    {
+      "epoch": 10.852932114614358,
+      "grad_norm": 0.09051407128572464,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 105675
+    },
+    {
+      "epoch": 10.8532402177262,
+      "grad_norm": 0.08709491789340973,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 105678
+    },
+    {
+      "epoch": 10.85354832083804,
+      "grad_norm": 0.04722654074430466,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 105681
+    },
+    {
+      "epoch": 10.853856423949882,
+      "grad_norm": 0.0712883397936821,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 105684
+    },
+    {
+      "epoch": 10.854164527061723,
+      "grad_norm": 0.07477840781211853,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 105687
+    },
+    {
+      "epoch": 10.854472630173564,
+      "grad_norm": 0.03935958445072174,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 105690
+    },
+    {
+      "epoch": 10.854780733285406,
+      "grad_norm": 0.04436472803354263,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 105693
+    },
+    {
+      "epoch": 10.855088836397247,
+      "grad_norm": 0.05736646056175232,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 105696
+    },
+    {
+      "epoch": 10.855396939509088,
+      "grad_norm": 0.05523679777979851,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 105699
+    },
+    {
+      "epoch": 10.855705042620931,
+      "grad_norm": 0.04662425071001053,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 105702
+    },
+    {
+      "epoch": 10.856013145732772,
+      "grad_norm": 0.0389157310128212,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 105705
+    },
+    {
+      "epoch": 10.856321248844614,
+      "grad_norm": 0.03747498244047165,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 105708
+    },
+    {
+      "epoch": 10.856629351956455,
+      "grad_norm": 0.04804906249046326,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 105711
+    },
+    {
+      "epoch": 10.856937455068296,
+      "grad_norm": 0.0921889990568161,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 105714
+    },
+    {
+      "epoch": 10.857245558180137,
+      "grad_norm": 0.09470658004283905,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 105717
+    },
+    {
+      "epoch": 10.857553661291979,
+      "grad_norm": 0.061489563435316086,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 105720
+    },
+    {
+      "epoch": 10.85786176440382,
+      "grad_norm": 0.05320308357477188,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 105723
+    },
+    {
+      "epoch": 10.858169867515661,
+      "grad_norm": 0.047891486436128616,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 105726
+    },
+    {
+      "epoch": 10.858477970627503,
+      "grad_norm": 0.03658050298690796,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 105729
+    },
+    {
+      "epoch": 10.858786073739346,
+      "grad_norm": 0.0334072969853878,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 105732
+    },
+    {
+      "epoch": 10.859094176851187,
+      "grad_norm": 0.09168829023838043,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 105735
+    },
+    {
+      "epoch": 10.859402279963028,
+      "grad_norm": 0.04450211301445961,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 105738
+    },
+    {
+      "epoch": 10.85971038307487,
+      "grad_norm": 0.07760917395353317,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 105741
+    },
+    {
+      "epoch": 10.86001848618671,
+      "grad_norm": 0.04388490691781044,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 105744
+    },
+    {
+      "epoch": 10.860326589298552,
+      "grad_norm": 0.10173279792070389,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 105747
+    },
+    {
+      "epoch": 10.860634692410393,
+      "grad_norm": 0.037178948521614075,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 105750
+    },
+    {
+      "epoch": 10.860942795522234,
+      "grad_norm": 0.10235720127820969,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 105753
+    },
+    {
+      "epoch": 10.861250898634076,
+      "grad_norm": 0.04074683040380478,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 105756
+    },
+    {
+      "epoch": 10.861559001745917,
+      "grad_norm": 0.10367844253778458,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 105759
+    },
+    {
+      "epoch": 10.861867104857758,
+      "grad_norm": 0.050666362047195435,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 105762
+    },
+    {
+      "epoch": 10.862175207969601,
+      "grad_norm": 0.11741497367620468,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 105765
+    },
+    {
+      "epoch": 10.862483311081442,
+      "grad_norm": 0.06941113620996475,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 105768
+    },
+    {
+      "epoch": 10.862791414193284,
+      "grad_norm": 0.05297830328345299,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 105771
+    },
+    {
+      "epoch": 10.863099517305125,
+      "grad_norm": 0.0719386413693428,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 105774
+    },
+    {
+      "epoch": 10.863407620416966,
+      "grad_norm": 0.044960249215364456,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 105777
+    },
+    {
+      "epoch": 10.863715723528808,
+      "grad_norm": 0.048919543623924255,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 105780
+    },
+    {
+      "epoch": 10.864023826640649,
+      "grad_norm": 0.0472579151391983,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 105783
+    },
+    {
+      "epoch": 10.86433192975249,
+      "grad_norm": 0.11817888915538788,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 105786
+    },
+    {
+      "epoch": 10.864640032864331,
+      "grad_norm": 0.10530492663383484,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 105789
+    },
+    {
+      "epoch": 10.864948135976173,
+      "grad_norm": 0.05322025716304779,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 105792
+    },
+    {
+      "epoch": 10.865256239088016,
+      "grad_norm": 0.08680053055286407,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 105795
+    },
+    {
+      "epoch": 10.865564342199857,
+      "grad_norm": 0.08292030543088913,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 105798
+    },
+    {
+      "epoch": 10.865872445311698,
+      "grad_norm": 0.07252789288759232,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 105801
+    },
+    {
+      "epoch": 10.86618054842354,
+      "grad_norm": 0.06576648354530334,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 105804
+    },
+    {
+      "epoch": 10.86648865153538,
+      "grad_norm": 0.17087021470069885,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 105807
+    },
+    {
+      "epoch": 10.866796754647222,
+      "grad_norm": 0.08984529972076416,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 105810
+    },
+    {
+      "epoch": 10.867104857759063,
+      "grad_norm": 0.07745181024074554,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 105813
+    },
+    {
+      "epoch": 10.867412960870904,
+      "grad_norm": 0.048002324998378754,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 105816
+    },
+    {
+      "epoch": 10.867721063982746,
+      "grad_norm": 0.04052358493208885,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 105819
+    },
+    {
+      "epoch": 10.868029167094587,
+      "grad_norm": 0.038736648857593536,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 105822
+    },
+    {
+      "epoch": 10.868337270206428,
+      "grad_norm": 0.054375361651182175,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 105825
+    },
+    {
+      "epoch": 10.868645373318271,
+      "grad_norm": 0.07048241794109344,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 105828
+    },
+    {
+      "epoch": 10.868953476430113,
+      "grad_norm": 0.03943666070699692,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 105831
+    },
+    {
+      "epoch": 10.869261579541954,
+      "grad_norm": 0.12087706476449966,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 105834
+    },
+    {
+      "epoch": 10.869569682653795,
+      "grad_norm": 0.055464401841163635,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 105837
+    },
+    {
+      "epoch": 10.869877785765636,
+      "grad_norm": 0.05967879295349121,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 105840
+    },
+    {
+      "epoch": 10.870185888877478,
+      "grad_norm": 0.10653529316186905,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 105843
+    },
+    {
+      "epoch": 10.870493991989319,
+      "grad_norm": 0.03763870522379875,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 105846
+    },
+    {
+      "epoch": 10.87080209510116,
+      "grad_norm": 0.11706111580133438,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 105849
+    },
+    {
+      "epoch": 10.871110198213001,
+      "grad_norm": 0.05440943315625191,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 105852
+    },
+    {
+      "epoch": 10.871418301324843,
+      "grad_norm": 0.04757893830537796,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 105855
+    },
+    {
+      "epoch": 10.871726404436686,
+      "grad_norm": 0.04554521292448044,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 105858
+    },
+    {
+      "epoch": 10.872034507548527,
+      "grad_norm": 0.05292823910713196,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 105861
+    },
+    {
+      "epoch": 10.872342610660368,
+      "grad_norm": 0.05600232630968094,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 105864
+    },
+    {
+      "epoch": 10.87265071377221,
+      "grad_norm": 0.050174783915281296,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 105867
+    },
+    {
+      "epoch": 10.87295881688405,
+      "grad_norm": 0.14416180551052094,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 105870
+    },
+    {
+      "epoch": 10.873266919995892,
+      "grad_norm": 0.07949387282133102,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 105873
+    },
+    {
+      "epoch": 10.873575023107733,
+      "grad_norm": 0.07921165227890015,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 105876
+    },
+    {
+      "epoch": 10.873883126219575,
+      "grad_norm": 0.05337239056825638,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 105879
+    },
+    {
+      "epoch": 10.874191229331416,
+      "grad_norm": 0.033021338284015656,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 105882
+    },
+    {
+      "epoch": 10.874499332443257,
+      "grad_norm": 0.035085279494524,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 105885
+    },
+    {
+      "epoch": 10.874807435555098,
+      "grad_norm": 0.11920043081045151,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 105888
+    },
+    {
+      "epoch": 10.875115538666941,
+      "grad_norm": 0.05265756696462631,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 105891
+    },
+    {
+      "epoch": 10.875423641778783,
+      "grad_norm": 0.09155082702636719,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 105894
+    },
+    {
+      "epoch": 10.875731744890624,
+      "grad_norm": 0.0716506689786911,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 105897
+    },
+    {
+      "epoch": 10.876039848002465,
+      "grad_norm": 0.06458982080221176,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 105900
+    },
+    {
+      "epoch": 10.876347951114306,
+      "grad_norm": 0.09309206157922745,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 105903
+    },
+    {
+      "epoch": 10.876656054226148,
+      "grad_norm": 0.04247670620679855,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 105906
+    },
+    {
+      "epoch": 10.876964157337989,
+      "grad_norm": 0.047627042979002,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 105909
+    },
+    {
+      "epoch": 10.87727226044983,
+      "grad_norm": 0.06475477665662766,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 105912
+    },
+    {
+      "epoch": 10.877580363561671,
+      "grad_norm": 0.042110294103622437,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 105915
+    },
+    {
+      "epoch": 10.877888466673513,
+      "grad_norm": 0.10465336591005325,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 105918
+    },
+    {
+      "epoch": 10.878196569785356,
+      "grad_norm": 0.091881662607193,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 105921
+    },
+    {
+      "epoch": 10.878504672897197,
+      "grad_norm": 0.05411464720964432,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 105924
+    },
+    {
+      "epoch": 10.878812776009038,
+      "grad_norm": 0.040598805993795395,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 105927
+    },
+    {
+      "epoch": 10.87912087912088,
+      "grad_norm": 0.038992274552583694,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 105930
+    },
+    {
+      "epoch": 10.87942898223272,
+      "grad_norm": 0.06576912105083466,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 105933
+    },
+    {
+      "epoch": 10.879737085344562,
+      "grad_norm": 0.08308625966310501,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 105936
+    },
+    {
+      "epoch": 10.880045188456403,
+      "grad_norm": 0.048460930585861206,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 105939
+    },
+    {
+      "epoch": 10.880353291568245,
+      "grad_norm": 0.07270680367946625,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 105942
+    },
+    {
+      "epoch": 10.880661394680086,
+      "grad_norm": 0.12631477415561676,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 105945
+    },
+    {
+      "epoch": 10.880969497791927,
+      "grad_norm": 0.061026062816381454,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 105948
+    },
+    {
+      "epoch": 10.881277600903768,
+      "grad_norm": 0.058033816516399384,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 105951
+    },
+    {
+      "epoch": 10.88158570401561,
+      "grad_norm": 0.10059273988008499,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 105954
+    },
+    {
+      "epoch": 10.881893807127453,
+      "grad_norm": 0.12196524441242218,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 105957
+    },
+    {
+      "epoch": 10.882201910239294,
+      "grad_norm": 0.05642728507518768,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 105960
+    },
+    {
+      "epoch": 10.882510013351135,
+      "grad_norm": 0.048246677964925766,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 105963
+    },
+    {
+      "epoch": 10.882818116462976,
+      "grad_norm": 0.07132484018802643,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 105966
+    },
+    {
+      "epoch": 10.883126219574818,
+      "grad_norm": 0.06366395950317383,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 105969
+    },
+    {
+      "epoch": 10.883434322686659,
+      "grad_norm": 0.077542744576931,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 105972
+    },
+    {
+      "epoch": 10.8837424257985,
+      "grad_norm": 0.09937810897827148,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 105975
+    },
+    {
+      "epoch": 10.884050528910342,
+      "grad_norm": 0.10000995546579361,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 105978
+    },
+    {
+      "epoch": 10.884358632022183,
+      "grad_norm": 0.03962424024939537,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 105981
+    },
+    {
+      "epoch": 10.884666735134024,
+      "grad_norm": 0.03430528566241264,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 105984
+    },
+    {
+      "epoch": 10.884974838245867,
+      "grad_norm": 0.05221163481473923,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 105987
+    },
+    {
+      "epoch": 10.885282941357708,
+      "grad_norm": 0.06595531851053238,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 105990
+    },
+    {
+      "epoch": 10.88559104446955,
+      "grad_norm": 0.0340123176574707,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 105993
+    },
+    {
+      "epoch": 10.88589914758139,
+      "grad_norm": 0.12168629467487335,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 105996
+    },
+    {
+      "epoch": 10.886207250693232,
+      "grad_norm": 0.045555856078863144,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 105999
+    },
+    {
+      "epoch": 10.886515353805073,
+      "grad_norm": 0.11041539162397385,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 106002
+    },
+    {
+      "epoch": 10.886823456916915,
+      "grad_norm": 0.0428503155708313,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 106005
+    },
+    {
+      "epoch": 10.887131560028756,
+      "grad_norm": 0.04097161069512367,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 106008
+    },
+    {
+      "epoch": 10.887439663140597,
+      "grad_norm": 0.055234361439943314,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 106011
+    },
+    {
+      "epoch": 10.887747766252438,
+      "grad_norm": 0.047480981796979904,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 106014
+    },
+    {
+      "epoch": 10.88805586936428,
+      "grad_norm": 0.07451288402080536,
+      "learning_rate": 0.01,
+      "loss": 1.9056,
+      "step": 106017
+    },
+    {
+      "epoch": 10.888363972476123,
+      "grad_norm": 0.09008581936359406,
+      "learning_rate": 0.01,
+      "loss": 1.9668,
+      "step": 106020
+    },
+    {
+      "epoch": 10.888672075587964,
+      "grad_norm": 0.07265052944421768,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 106023
+    },
+    {
+      "epoch": 10.888980178699805,
+      "grad_norm": 0.09329503774642944,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 106026
+    },
+    {
+      "epoch": 10.889288281811647,
+      "grad_norm": 0.07104849815368652,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 106029
+    },
+    {
+      "epoch": 10.889596384923488,
+      "grad_norm": 0.08163333684206009,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 106032
+    },
+    {
+      "epoch": 10.889904488035329,
+      "grad_norm": 0.10383013635873795,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 106035
+    },
+    {
+      "epoch": 10.89021259114717,
+      "grad_norm": 0.06662754714488983,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 106038
+    },
+    {
+      "epoch": 10.890520694259012,
+      "grad_norm": 0.056699350476264954,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 106041
+    },
+    {
+      "epoch": 10.890828797370853,
+      "grad_norm": 0.05268816277384758,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 106044
+    },
+    {
+      "epoch": 10.891136900482694,
+      "grad_norm": 0.037846196442842484,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 106047
+    },
+    {
+      "epoch": 10.891445003594537,
+      "grad_norm": 0.03779904544353485,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 106050
+    },
+    {
+      "epoch": 10.891753106706378,
+      "grad_norm": 0.12322263419628143,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 106053
+    },
+    {
+      "epoch": 10.89206120981822,
+      "grad_norm": 0.07598162442445755,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 106056
+    },
+    {
+      "epoch": 10.892369312930061,
+      "grad_norm": 0.12336615473031998,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 106059
+    },
+    {
+      "epoch": 10.892677416041902,
+      "grad_norm": 0.09707363694906235,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 106062
+    },
+    {
+      "epoch": 10.892985519153743,
+      "grad_norm": 0.04322752729058266,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 106065
+    },
+    {
+      "epoch": 10.893293622265585,
+      "grad_norm": 0.0348566398024559,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 106068
+    },
+    {
+      "epoch": 10.893601725377426,
+      "grad_norm": 0.046460725367069244,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 106071
+    },
+    {
+      "epoch": 10.893909828489267,
+      "grad_norm": 0.06775607168674469,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 106074
+    },
+    {
+      "epoch": 10.894217931601109,
+      "grad_norm": 0.03778929263353348,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 106077
+    },
+    {
+      "epoch": 10.89452603471295,
+      "grad_norm": 0.04769425094127655,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 106080
+    },
+    {
+      "epoch": 10.894834137824793,
+      "grad_norm": 0.05076927691698074,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 106083
+    },
+    {
+      "epoch": 10.895142240936634,
+      "grad_norm": 0.10340415686368942,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 106086
+    },
+    {
+      "epoch": 10.895450344048475,
+      "grad_norm": 0.12192103266716003,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 106089
+    },
+    {
+      "epoch": 10.895758447160317,
+      "grad_norm": 0.06740839779376984,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 106092
+    },
+    {
+      "epoch": 10.896066550272158,
+      "grad_norm": 0.03260725364089012,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 106095
+    },
+    {
+      "epoch": 10.896374653384,
+      "grad_norm": 0.06192610785365105,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 106098
+    },
+    {
+      "epoch": 10.89668275649584,
+      "grad_norm": 0.06303668022155762,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 106101
+    },
+    {
+      "epoch": 10.896990859607682,
+      "grad_norm": 0.1754002720117569,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 106104
+    },
+    {
+      "epoch": 10.897298962719523,
+      "grad_norm": 0.0929434522986412,
+      "learning_rate": 0.01,
+      "loss": 1.899,
+      "step": 106107
+    },
+    {
+      "epoch": 10.897607065831364,
+      "grad_norm": 0.07154718786478043,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 106110
+    },
+    {
+      "epoch": 10.897915168943207,
+      "grad_norm": 0.0400879830121994,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 106113
+    },
+    {
+      "epoch": 10.898223272055048,
+      "grad_norm": 0.034502558410167694,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 106116
+    },
+    {
+      "epoch": 10.89853137516689,
+      "grad_norm": 0.03446902707219124,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 106119
+    },
+    {
+      "epoch": 10.898839478278731,
+      "grad_norm": 0.03258249908685684,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 106122
+    },
+    {
+      "epoch": 10.899147581390572,
+      "grad_norm": 0.04886491969227791,
+      "learning_rate": 0.01,
+      "loss": 1.9781,
+      "step": 106125
+    },
+    {
+      "epoch": 10.899455684502414,
+      "grad_norm": 0.07080874592065811,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 106128
+    },
+    {
+      "epoch": 10.899763787614255,
+      "grad_norm": 0.09572204947471619,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 106131
+    },
+    {
+      "epoch": 10.900071890726096,
+      "grad_norm": 0.0943838506937027,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 106134
+    },
+    {
+      "epoch": 10.900379993837937,
+      "grad_norm": 0.0512869767844677,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 106137
+    },
+    {
+      "epoch": 10.900688096949779,
+      "grad_norm": 0.04730043560266495,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 106140
+    },
+    {
+      "epoch": 10.90099620006162,
+      "grad_norm": 0.18809223175048828,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 106143
+    },
+    {
+      "epoch": 10.901304303173463,
+      "grad_norm": 0.08880333602428436,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 106146
+    },
+    {
+      "epoch": 10.901612406285304,
+      "grad_norm": 0.08753082156181335,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 106149
+    },
+    {
+      "epoch": 10.901920509397145,
+      "grad_norm": 0.0716850608587265,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 106152
+    },
+    {
+      "epoch": 10.902228612508987,
+      "grad_norm": 0.06579985469579697,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 106155
+    },
+    {
+      "epoch": 10.902536715620828,
+      "grad_norm": 0.06625322252511978,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 106158
+    },
+    {
+      "epoch": 10.90284481873267,
+      "grad_norm": 0.04039857164025307,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 106161
+    },
+    {
+      "epoch": 10.90315292184451,
+      "grad_norm": 0.0351543202996254,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 106164
+    },
+    {
+      "epoch": 10.903461024956352,
+      "grad_norm": 0.0355987548828125,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 106167
+    },
+    {
+      "epoch": 10.903769128068193,
+      "grad_norm": 0.03463450074195862,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 106170
+    },
+    {
+      "epoch": 10.904077231180034,
+      "grad_norm": 0.05925464630126953,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 106173
+    },
+    {
+      "epoch": 10.904385334291877,
+      "grad_norm": 0.057822998613119125,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 106176
+    },
+    {
+      "epoch": 10.904693437403719,
+      "grad_norm": 0.15748798847198486,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 106179
+    },
+    {
+      "epoch": 10.90500154051556,
+      "grad_norm": 0.04025212302803993,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 106182
+    },
+    {
+      "epoch": 10.905309643627401,
+      "grad_norm": 0.08656113594770432,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 106185
+    },
+    {
+      "epoch": 10.905617746739242,
+      "grad_norm": 0.0952494889497757,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 106188
+    },
+    {
+      "epoch": 10.905925849851084,
+      "grad_norm": 0.05528434365987778,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 106191
+    },
+    {
+      "epoch": 10.906233952962925,
+      "grad_norm": 0.04796331375837326,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 106194
+    },
+    {
+      "epoch": 10.906542056074766,
+      "grad_norm": 0.04357631504535675,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 106197
+    },
+    {
+      "epoch": 10.906850159186607,
+      "grad_norm": 0.040124353021383286,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 106200
+    },
+    {
+      "epoch": 10.907158262298449,
+      "grad_norm": 0.03544517606496811,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 106203
+    },
+    {
+      "epoch": 10.90746636541029,
+      "grad_norm": 0.058340318500995636,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 106206
+    },
+    {
+      "epoch": 10.907774468522131,
+      "grad_norm": 0.08568999916315079,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 106209
+    },
+    {
+      "epoch": 10.908082571633974,
+      "grad_norm": 0.06406117230653763,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 106212
+    },
+    {
+      "epoch": 10.908390674745815,
+      "grad_norm": 0.059249307960271835,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 106215
+    },
+    {
+      "epoch": 10.908698777857657,
+      "grad_norm": 0.052112944424152374,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 106218
+    },
+    {
+      "epoch": 10.909006880969498,
+      "grad_norm": 0.03925701603293419,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 106221
+    },
+    {
+      "epoch": 10.90931498408134,
+      "grad_norm": 0.14518700540065765,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 106224
+    },
+    {
+      "epoch": 10.90962308719318,
+      "grad_norm": 0.13229291141033173,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 106227
+    },
+    {
+      "epoch": 10.909931190305022,
+      "grad_norm": 0.05290767550468445,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 106230
+    },
+    {
+      "epoch": 10.910239293416863,
+      "grad_norm": 0.044173240661621094,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 106233
+    },
+    {
+      "epoch": 10.910547396528704,
+      "grad_norm": 0.042460594326257706,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 106236
+    },
+    {
+      "epoch": 10.910855499640546,
+      "grad_norm": 0.03345232084393501,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 106239
+    },
+    {
+      "epoch": 10.911163602752389,
+      "grad_norm": 0.03705401346087456,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 106242
+    },
+    {
+      "epoch": 10.91147170586423,
+      "grad_norm": 0.06710290163755417,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 106245
+    },
+    {
+      "epoch": 10.911779808976071,
+      "grad_norm": 0.042429160326719284,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 106248
+    },
+    {
+      "epoch": 10.912087912087912,
+      "grad_norm": 0.0361759252846241,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 106251
+    },
+    {
+      "epoch": 10.912396015199754,
+      "grad_norm": 0.02957993932068348,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 106254
+    },
+    {
+      "epoch": 10.912704118311595,
+      "grad_norm": 0.045638322830200195,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 106257
+    },
+    {
+      "epoch": 10.913012221423436,
+      "grad_norm": 0.06491947174072266,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 106260
+    },
+    {
+      "epoch": 10.913320324535277,
+      "grad_norm": 0.12178110331296921,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 106263
+    },
+    {
+      "epoch": 10.913628427647119,
+      "grad_norm": 0.060336388647556305,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 106266
+    },
+    {
+      "epoch": 10.91393653075896,
+      "grad_norm": 0.0462242066860199,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 106269
+    },
+    {
+      "epoch": 10.914244633870801,
+      "grad_norm": 0.11648637056350708,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 106272
+    },
+    {
+      "epoch": 10.914552736982644,
+      "grad_norm": 0.05701465904712677,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 106275
+    },
+    {
+      "epoch": 10.914860840094486,
+      "grad_norm": 0.06645436584949493,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 106278
+    },
+    {
+      "epoch": 10.915168943206327,
+      "grad_norm": 0.06109488010406494,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 106281
+    },
+    {
+      "epoch": 10.915477046318168,
+      "grad_norm": 0.0647864118218422,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 106284
+    },
+    {
+      "epoch": 10.91578514943001,
+      "grad_norm": 0.16096295416355133,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 106287
+    },
+    {
+      "epoch": 10.91609325254185,
+      "grad_norm": 0.16784384846687317,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 106290
+    },
+    {
+      "epoch": 10.916401355653692,
+      "grad_norm": 0.11912429332733154,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 106293
+    },
+    {
+      "epoch": 10.916709458765533,
+      "grad_norm": 0.10303348302841187,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 106296
+    },
+    {
+      "epoch": 10.917017561877374,
+      "grad_norm": 0.09231548756361008,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 106299
+    },
+    {
+      "epoch": 10.917325664989216,
+      "grad_norm": 0.06794840842485428,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 106302
+    },
+    {
+      "epoch": 10.917633768101059,
+      "grad_norm": 0.12076061964035034,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 106305
+    },
+    {
+      "epoch": 10.9179418712129,
+      "grad_norm": 0.06557746976613998,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 106308
+    },
+    {
+      "epoch": 10.918249974324741,
+      "grad_norm": 0.04917408525943756,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 106311
+    },
+    {
+      "epoch": 10.918558077436582,
+      "grad_norm": 0.09011615812778473,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 106314
+    },
+    {
+      "epoch": 10.918866180548424,
+      "grad_norm": 0.057904936373233795,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 106317
+    },
+    {
+      "epoch": 10.919174283660265,
+      "grad_norm": 0.042555589228868484,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 106320
+    },
+    {
+      "epoch": 10.919482386772106,
+      "grad_norm": 0.10149985551834106,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 106323
+    },
+    {
+      "epoch": 10.919790489883948,
+      "grad_norm": 0.061605051159858704,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 106326
+    },
+    {
+      "epoch": 10.920098592995789,
+      "grad_norm": 0.05120980367064476,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 106329
+    },
+    {
+      "epoch": 10.92040669610763,
+      "grad_norm": 0.07240397483110428,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 106332
+    },
+    {
+      "epoch": 10.920714799219471,
+      "grad_norm": 0.10335461795330048,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 106335
+    },
+    {
+      "epoch": 10.921022902331314,
+      "grad_norm": 0.05146472528576851,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 106338
+    },
+    {
+      "epoch": 10.921331005443156,
+      "grad_norm": 0.09158754348754883,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 106341
+    },
+    {
+      "epoch": 10.921639108554997,
+      "grad_norm": 0.0714024156332016,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 106344
+    },
+    {
+      "epoch": 10.921947211666838,
+      "grad_norm": 0.07200371474027634,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 106347
+    },
+    {
+      "epoch": 10.92225531477868,
+      "grad_norm": 0.033079616725444794,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 106350
+    },
+    {
+      "epoch": 10.92256341789052,
+      "grad_norm": 0.06759501993656158,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 106353
+    },
+    {
+      "epoch": 10.922871521002362,
+      "grad_norm": 0.07132084667682648,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 106356
+    },
+    {
+      "epoch": 10.923179624114203,
+      "grad_norm": 0.03897568956017494,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 106359
+    },
+    {
+      "epoch": 10.923487727226044,
+      "grad_norm": 0.04293280094861984,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 106362
+    },
+    {
+      "epoch": 10.923795830337886,
+      "grad_norm": 0.11670386046171188,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 106365
+    },
+    {
+      "epoch": 10.924103933449729,
+      "grad_norm": 0.12519721686840057,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 106368
+    },
+    {
+      "epoch": 10.92441203656157,
+      "grad_norm": 0.038999710232019424,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 106371
+    },
+    {
+      "epoch": 10.924720139673411,
+      "grad_norm": 0.07099395245313644,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 106374
+    },
+    {
+      "epoch": 10.925028242785253,
+      "grad_norm": 0.05256061628460884,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 106377
+    },
+    {
+      "epoch": 10.925336345897094,
+      "grad_norm": 0.10918042808771133,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 106380
+    },
+    {
+      "epoch": 10.925644449008935,
+      "grad_norm": 0.15344181656837463,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 106383
+    },
+    {
+      "epoch": 10.925952552120776,
+      "grad_norm": 0.07730207592248917,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 106386
+    },
+    {
+      "epoch": 10.926260655232618,
+      "grad_norm": 0.06737470626831055,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 106389
+    },
+    {
+      "epoch": 10.926568758344459,
+      "grad_norm": 0.06340444087982178,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 106392
+    },
+    {
+      "epoch": 10.9268768614563,
+      "grad_norm": 0.08433797955513,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 106395
+    },
+    {
+      "epoch": 10.927184964568141,
+      "grad_norm": 0.059738487005233765,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 106398
+    },
+    {
+      "epoch": 10.927493067679983,
+      "grad_norm": 0.038807209581136703,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 106401
+    },
+    {
+      "epoch": 10.927801170791826,
+      "grad_norm": 0.05405537784099579,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 106404
+    },
+    {
+      "epoch": 10.928109273903667,
+      "grad_norm": 0.04065915197134018,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 106407
+    },
+    {
+      "epoch": 10.928417377015508,
+      "grad_norm": 0.03751445561647415,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 106410
+    },
+    {
+      "epoch": 10.92872548012735,
+      "grad_norm": 0.034337520599365234,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 106413
+    },
+    {
+      "epoch": 10.92903358323919,
+      "grad_norm": 0.07262330502271652,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 106416
+    },
+    {
+      "epoch": 10.929341686351032,
+      "grad_norm": 0.04483049735426903,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 106419
+    },
+    {
+      "epoch": 10.929649789462873,
+      "grad_norm": 0.06375708431005478,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 106422
+    },
+    {
+      "epoch": 10.929957892574715,
+      "grad_norm": 0.0656374990940094,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 106425
+    },
+    {
+      "epoch": 10.930265995686556,
+      "grad_norm": 0.1104300320148468,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 106428
+    },
+    {
+      "epoch": 10.930574098798397,
+      "grad_norm": 0.04479607939720154,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 106431
+    },
+    {
+      "epoch": 10.93088220191024,
+      "grad_norm": 0.09327235817909241,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 106434
+    },
+    {
+      "epoch": 10.931190305022081,
+      "grad_norm": 0.07236454635858536,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 106437
+    },
+    {
+      "epoch": 10.931498408133923,
+      "grad_norm": 0.03589305654168129,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 106440
+    },
+    {
+      "epoch": 10.931806511245764,
+      "grad_norm": 0.034649983048439026,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 106443
+    },
+    {
+      "epoch": 10.932114614357605,
+      "grad_norm": 0.09589678049087524,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 106446
+    },
+    {
+      "epoch": 10.932422717469446,
+      "grad_norm": 0.05428173393011093,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 106449
+    },
+    {
+      "epoch": 10.932730820581288,
+      "grad_norm": 0.06434815376996994,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 106452
+    },
+    {
+      "epoch": 10.933038923693129,
+      "grad_norm": 0.12415088713169098,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 106455
+    },
+    {
+      "epoch": 10.93334702680497,
+      "grad_norm": 0.06227521225810051,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 106458
+    },
+    {
+      "epoch": 10.933655129916811,
+      "grad_norm": 0.06899651139974594,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 106461
+    },
+    {
+      "epoch": 10.933963233028653,
+      "grad_norm": 0.035584982484579086,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 106464
+    },
+    {
+      "epoch": 10.934271336140496,
+      "grad_norm": 0.05347567796707153,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 106467
+    },
+    {
+      "epoch": 10.934579439252337,
+      "grad_norm": 0.04459035024046898,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 106470
+    },
+    {
+      "epoch": 10.934887542364178,
+      "grad_norm": 0.06677372008562088,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 106473
+    },
+    {
+      "epoch": 10.93519564547602,
+      "grad_norm": 0.12824846804141998,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 106476
+    },
+    {
+      "epoch": 10.93550374858786,
+      "grad_norm": 0.05906982347369194,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 106479
+    },
+    {
+      "epoch": 10.935811851699702,
+      "grad_norm": 0.08169703930616379,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 106482
+    },
+    {
+      "epoch": 10.936119954811543,
+      "grad_norm": 0.12834133207798004,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 106485
+    },
+    {
+      "epoch": 10.936428057923385,
+      "grad_norm": 0.053135670721530914,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 106488
+    },
+    {
+      "epoch": 10.936736161035226,
+      "grad_norm": 0.03859506547451019,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 106491
+    },
+    {
+      "epoch": 10.937044264147067,
+      "grad_norm": 0.06961911916732788,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 106494
+    },
+    {
+      "epoch": 10.93735236725891,
+      "grad_norm": 0.1543235182762146,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 106497
+    },
+    {
+      "epoch": 10.937660470370751,
+      "grad_norm": 0.04941339045763016,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 106500
+    },
+    {
+      "epoch": 10.937968573482593,
+      "grad_norm": 0.0709512010216713,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 106503
+    },
+    {
+      "epoch": 10.938276676594434,
+      "grad_norm": 0.04137871414422989,
+      "learning_rate": 0.01,
+      "loss": 1.8939,
+      "step": 106506
+    },
+    {
+      "epoch": 10.938584779706275,
+      "grad_norm": 0.05609670281410217,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 106509
+    },
+    {
+      "epoch": 10.938892882818116,
+      "grad_norm": 0.04416004940867424,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 106512
+    },
+    {
+      "epoch": 10.939200985929958,
+      "grad_norm": 0.0448085255920887,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 106515
+    },
+    {
+      "epoch": 10.939509089041799,
+      "grad_norm": 0.04962565377354622,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 106518
+    },
+    {
+      "epoch": 10.93981719215364,
+      "grad_norm": 0.07378124445676804,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 106521
+    },
+    {
+      "epoch": 10.940125295265481,
+      "grad_norm": 0.08252052217721939,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 106524
+    },
+    {
+      "epoch": 10.940433398377323,
+      "grad_norm": 0.07529515773057938,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 106527
+    },
+    {
+      "epoch": 10.940741501489166,
+      "grad_norm": 0.07012537121772766,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 106530
+    },
+    {
+      "epoch": 10.941049604601007,
+      "grad_norm": 0.0447557158768177,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 106533
+    },
+    {
+      "epoch": 10.941357707712848,
+      "grad_norm": 0.056523241102695465,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 106536
+    },
+    {
+      "epoch": 10.94166581082469,
+      "grad_norm": 0.07511433959007263,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 106539
+    },
+    {
+      "epoch": 10.94197391393653,
+      "grad_norm": 0.04001323878765106,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 106542
+    },
+    {
+      "epoch": 10.942282017048372,
+      "grad_norm": 0.12339727580547333,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 106545
+    },
+    {
+      "epoch": 10.942590120160213,
+      "grad_norm": 0.07011760026216507,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 106548
+    },
+    {
+      "epoch": 10.942898223272055,
+      "grad_norm": 0.08485225588083267,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 106551
+    },
+    {
+      "epoch": 10.943206326383896,
+      "grad_norm": 0.1317882388830185,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 106554
+    },
+    {
+      "epoch": 10.943514429495737,
+      "grad_norm": 0.05222870036959648,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 106557
+    },
+    {
+      "epoch": 10.94382253260758,
+      "grad_norm": 0.060174986720085144,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 106560
+    },
+    {
+      "epoch": 10.944130635719421,
+      "grad_norm": 0.06661222130060196,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 106563
+    },
+    {
+      "epoch": 10.944438738831263,
+      "grad_norm": 0.037411950528621674,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 106566
+    },
+    {
+      "epoch": 10.944746841943104,
+      "grad_norm": 0.046929627656936646,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 106569
+    },
+    {
+      "epoch": 10.945054945054945,
+      "grad_norm": 0.10040893405675888,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 106572
+    },
+    {
+      "epoch": 10.945363048166787,
+      "grad_norm": 0.038712795823812485,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 106575
+    },
+    {
+      "epoch": 10.945671151278628,
+      "grad_norm": 0.12498601526021957,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 106578
+    },
+    {
+      "epoch": 10.945979254390469,
+      "grad_norm": 0.05384666845202446,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 106581
+    },
+    {
+      "epoch": 10.94628735750231,
+      "grad_norm": 0.04159768298268318,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 106584
+    },
+    {
+      "epoch": 10.946595460614152,
+      "grad_norm": 0.1507614552974701,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 106587
+    },
+    {
+      "epoch": 10.946903563725993,
+      "grad_norm": 0.04219367727637291,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 106590
+    },
+    {
+      "epoch": 10.947211666837836,
+      "grad_norm": 0.046481501311063766,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 106593
+    },
+    {
+      "epoch": 10.947519769949677,
+      "grad_norm": 0.05947163328528404,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 106596
+    },
+    {
+      "epoch": 10.947827873061518,
+      "grad_norm": 0.1727290153503418,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 106599
+    },
+    {
+      "epoch": 10.94813597617336,
+      "grad_norm": 0.08525878190994263,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 106602
+    },
+    {
+      "epoch": 10.948444079285201,
+      "grad_norm": 0.058909181505441666,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 106605
+    },
+    {
+      "epoch": 10.948752182397042,
+      "grad_norm": 0.04929728806018829,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 106608
+    },
+    {
+      "epoch": 10.949060285508883,
+      "grad_norm": 0.039295829832553864,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 106611
+    },
+    {
+      "epoch": 10.949368388620725,
+      "grad_norm": 0.04213673621416092,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 106614
+    },
+    {
+      "epoch": 10.949676491732566,
+      "grad_norm": 0.04450298100709915,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 106617
+    },
+    {
+      "epoch": 10.949984594844407,
+      "grad_norm": 0.05633500590920448,
+      "learning_rate": 0.01,
+      "loss": 1.9729,
+      "step": 106620
+    },
+    {
+      "epoch": 10.95029269795625,
+      "grad_norm": 0.042609844356775284,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 106623
+    },
+    {
+      "epoch": 10.950600801068092,
+      "grad_norm": 0.03682399168610573,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 106626
+    },
+    {
+      "epoch": 10.950908904179933,
+      "grad_norm": 0.06465191394090652,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 106629
+    },
+    {
+      "epoch": 10.951217007291774,
+      "grad_norm": 0.05256539583206177,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 106632
+    },
+    {
+      "epoch": 10.951525110403615,
+      "grad_norm": 0.07118343561887741,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 106635
+    },
+    {
+      "epoch": 10.951833213515457,
+      "grad_norm": 0.17668597400188446,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 106638
+    },
+    {
+      "epoch": 10.952141316627298,
+      "grad_norm": 0.045385267585515976,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 106641
+    },
+    {
+      "epoch": 10.952449419739139,
+      "grad_norm": 0.06024050712585449,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 106644
+    },
+    {
+      "epoch": 10.95275752285098,
+      "grad_norm": 0.04597383365035057,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 106647
+    },
+    {
+      "epoch": 10.953065625962822,
+      "grad_norm": 0.035271596163511276,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 106650
+    },
+    {
+      "epoch": 10.953373729074663,
+      "grad_norm": 0.048514969646930695,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 106653
+    },
+    {
+      "epoch": 10.953681832186504,
+      "grad_norm": 0.05543520674109459,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 106656
+    },
+    {
+      "epoch": 10.953989935298347,
+      "grad_norm": 0.06604902446269989,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 106659
+    },
+    {
+      "epoch": 10.954298038410188,
+      "grad_norm": 0.06005027890205383,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 106662
+    },
+    {
+      "epoch": 10.95460614152203,
+      "grad_norm": 0.05095071718096733,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 106665
+    },
+    {
+      "epoch": 10.954914244633871,
+      "grad_norm": 0.1646306812763214,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 106668
+    },
+    {
+      "epoch": 10.955222347745712,
+      "grad_norm": 0.15995948016643524,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 106671
+    },
+    {
+      "epoch": 10.955530450857554,
+      "grad_norm": 0.09556877613067627,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 106674
+    },
+    {
+      "epoch": 10.955838553969395,
+      "grad_norm": 0.04684971272945404,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 106677
+    },
+    {
+      "epoch": 10.956146657081236,
+      "grad_norm": 0.04624011367559433,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 106680
+    },
+    {
+      "epoch": 10.956454760193077,
+      "grad_norm": 0.03910151869058609,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 106683
+    },
+    {
+      "epoch": 10.956762863304919,
+      "grad_norm": 0.08027227967977524,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 106686
+    },
+    {
+      "epoch": 10.957070966416762,
+      "grad_norm": 0.07811252772808075,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 106689
+    },
+    {
+      "epoch": 10.957379069528603,
+      "grad_norm": 0.07080382108688354,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 106692
+    },
+    {
+      "epoch": 10.957687172640444,
+      "grad_norm": 0.03555811569094658,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 106695
+    },
+    {
+      "epoch": 10.957995275752285,
+      "grad_norm": 0.033408600836992264,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 106698
+    },
+    {
+      "epoch": 10.958303378864127,
+      "grad_norm": 0.03809535875916481,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 106701
+    },
+    {
+      "epoch": 10.958611481975968,
+      "grad_norm": 0.08328834921121597,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 106704
+    },
+    {
+      "epoch": 10.95891958508781,
+      "grad_norm": 0.09299707412719727,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 106707
+    },
+    {
+      "epoch": 10.95922768819965,
+      "grad_norm": 0.10310142487287521,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 106710
+    },
+    {
+      "epoch": 10.959535791311492,
+      "grad_norm": 0.11186525225639343,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 106713
+    },
+    {
+      "epoch": 10.959843894423333,
+      "grad_norm": 0.08874673396348953,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 106716
+    },
+    {
+      "epoch": 10.960151997535174,
+      "grad_norm": 0.08445736765861511,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 106719
+    },
+    {
+      "epoch": 10.960460100647017,
+      "grad_norm": 0.07096145302057266,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 106722
+    },
+    {
+      "epoch": 10.960768203758859,
+      "grad_norm": 0.045588068664073944,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 106725
+    },
+    {
+      "epoch": 10.9610763068707,
+      "grad_norm": 0.0643002986907959,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 106728
+    },
+    {
+      "epoch": 10.961384409982541,
+      "grad_norm": 0.12200552970170975,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 106731
+    },
+    {
+      "epoch": 10.961692513094382,
+      "grad_norm": 0.06449657678604126,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 106734
+    },
+    {
+      "epoch": 10.962000616206224,
+      "grad_norm": 0.09041350334882736,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 106737
+    },
+    {
+      "epoch": 10.962308719318065,
+      "grad_norm": 0.04235527664422989,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 106740
+    },
+    {
+      "epoch": 10.962616822429906,
+      "grad_norm": 0.06367935985326767,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 106743
+    },
+    {
+      "epoch": 10.962924925541747,
+      "grad_norm": 0.03904595971107483,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 106746
+    },
+    {
+      "epoch": 10.963233028653589,
+      "grad_norm": 0.09359109401702881,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 106749
+    },
+    {
+      "epoch": 10.963541131765432,
+      "grad_norm": 0.058446332812309265,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 106752
+    },
+    {
+      "epoch": 10.963849234877273,
+      "grad_norm": 0.10020552575588226,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 106755
+    },
+    {
+      "epoch": 10.964157337989114,
+      "grad_norm": 0.050879430025815964,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 106758
+    },
+    {
+      "epoch": 10.964465441100955,
+      "grad_norm": 0.04936183616518974,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 106761
+    },
+    {
+      "epoch": 10.964773544212797,
+      "grad_norm": 0.05340530350804329,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 106764
+    },
+    {
+      "epoch": 10.965081647324638,
+      "grad_norm": 0.1125936210155487,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 106767
+    },
+    {
+      "epoch": 10.96538975043648,
+      "grad_norm": 0.13663361966609955,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 106770
+    },
+    {
+      "epoch": 10.96569785354832,
+      "grad_norm": 0.1450393944978714,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 106773
+    },
+    {
+      "epoch": 10.966005956660162,
+      "grad_norm": 0.09279601275920868,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 106776
+    },
+    {
+      "epoch": 10.966314059772003,
+      "grad_norm": 0.03667345643043518,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 106779
+    },
+    {
+      "epoch": 10.966622162883844,
+      "grad_norm": 0.040375106036663055,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 106782
+    },
+    {
+      "epoch": 10.966930265995687,
+      "grad_norm": 0.05492724850773811,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 106785
+    },
+    {
+      "epoch": 10.967238369107529,
+      "grad_norm": 0.07538987696170807,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 106788
+    },
+    {
+      "epoch": 10.96754647221937,
+      "grad_norm": 0.049806006252765656,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 106791
+    },
+    {
+      "epoch": 10.967854575331211,
+      "grad_norm": 0.04567021504044533,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 106794
+    },
+    {
+      "epoch": 10.968162678443052,
+      "grad_norm": 0.055650562047958374,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 106797
+    },
+    {
+      "epoch": 10.968470781554894,
+      "grad_norm": 0.04961836710572243,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 106800
+    },
+    {
+      "epoch": 10.968778884666735,
+      "grad_norm": 0.08718889951705933,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 106803
+    },
+    {
+      "epoch": 10.969086987778576,
+      "grad_norm": 0.05088066682219505,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 106806
+    },
+    {
+      "epoch": 10.969395090890417,
+      "grad_norm": 0.042699042707681656,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 106809
+    },
+    {
+      "epoch": 10.969703194002259,
+      "grad_norm": 0.04571409523487091,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 106812
+    },
+    {
+      "epoch": 10.970011297114102,
+      "grad_norm": 0.058041494339704514,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 106815
+    },
+    {
+      "epoch": 10.970319400225943,
+      "grad_norm": 0.1141694188117981,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 106818
+    },
+    {
+      "epoch": 10.970627503337784,
+      "grad_norm": 0.06193096935749054,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 106821
+    },
+    {
+      "epoch": 10.970935606449626,
+      "grad_norm": 0.06552765518426895,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 106824
+    },
+    {
+      "epoch": 10.971243709561467,
+      "grad_norm": 0.10089132189750671,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 106827
+    },
+    {
+      "epoch": 10.971551812673308,
+      "grad_norm": 0.04266245290637016,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 106830
+    },
+    {
+      "epoch": 10.97185991578515,
+      "grad_norm": 0.045328978449106216,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 106833
+    },
+    {
+      "epoch": 10.97216801889699,
+      "grad_norm": 0.03314703330397606,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 106836
+    },
+    {
+      "epoch": 10.972476122008832,
+      "grad_norm": 0.10586504638195038,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 106839
+    },
+    {
+      "epoch": 10.972784225120673,
+      "grad_norm": 0.07202047109603882,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 106842
+    },
+    {
+      "epoch": 10.973092328232514,
+      "grad_norm": 0.08412293344736099,
+      "learning_rate": 0.01,
+      "loss": 1.9624,
+      "step": 106845
+    },
+    {
+      "epoch": 10.973400431344357,
+      "grad_norm": 0.041309576481580734,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 106848
+    },
+    {
+      "epoch": 10.973708534456199,
+      "grad_norm": 0.09424412250518799,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 106851
+    },
+    {
+      "epoch": 10.97401663756804,
+      "grad_norm": 0.0540972538292408,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 106854
+    },
+    {
+      "epoch": 10.974324740679881,
+      "grad_norm": 0.04370080307126045,
+      "learning_rate": 0.01,
+      "loss": 1.9816,
+      "step": 106857
+    },
+    {
+      "epoch": 10.974632843791722,
+      "grad_norm": 0.029536878690123558,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 106860
+    },
+    {
+      "epoch": 10.974940946903564,
+      "grad_norm": 0.12666262686252594,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 106863
+    },
+    {
+      "epoch": 10.975249050015405,
+      "grad_norm": 0.07763486355543137,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 106866
+    },
+    {
+      "epoch": 10.975557153127246,
+      "grad_norm": 0.08744452148675919,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 106869
+    },
+    {
+      "epoch": 10.975865256239087,
+      "grad_norm": 0.04688967019319534,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 106872
+    },
+    {
+      "epoch": 10.976173359350929,
+      "grad_norm": 0.044772904366254807,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 106875
+    },
+    {
+      "epoch": 10.976481462462772,
+      "grad_norm": 0.042988039553165436,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 106878
+    },
+    {
+      "epoch": 10.976789565574613,
+      "grad_norm": 0.044973667711019516,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 106881
+    },
+    {
+      "epoch": 10.977097668686454,
+      "grad_norm": 0.07078361511230469,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 106884
+    },
+    {
+      "epoch": 10.977405771798296,
+      "grad_norm": 0.03690226748585701,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 106887
+    },
+    {
+      "epoch": 10.977713874910137,
+      "grad_norm": 0.08343128859996796,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 106890
+    },
+    {
+      "epoch": 10.978021978021978,
+      "grad_norm": 0.0915609672665596,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 106893
+    },
+    {
+      "epoch": 10.97833008113382,
+      "grad_norm": 0.051095981150865555,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 106896
+    },
+    {
+      "epoch": 10.97863818424566,
+      "grad_norm": 0.08984770625829697,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 106899
+    },
+    {
+      "epoch": 10.978946287357502,
+      "grad_norm": 0.033725060522556305,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 106902
+    },
+    {
+      "epoch": 10.979254390469343,
+      "grad_norm": 0.05065962299704552,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 106905
+    },
+    {
+      "epoch": 10.979562493581184,
+      "grad_norm": 0.057545311748981476,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 106908
+    },
+    {
+      "epoch": 10.979870596693026,
+      "grad_norm": 0.04934195429086685,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 106911
+    },
+    {
+      "epoch": 10.980178699804869,
+      "grad_norm": 0.050023820251226425,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 106914
+    },
+    {
+      "epoch": 10.98048680291671,
+      "grad_norm": 0.05062216520309448,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 106917
+    },
+    {
+      "epoch": 10.980794906028551,
+      "grad_norm": 0.035237058997154236,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 106920
+    },
+    {
+      "epoch": 10.981103009140393,
+      "grad_norm": 0.047082848846912384,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 106923
+    },
+    {
+      "epoch": 10.981411112252234,
+      "grad_norm": 0.04227985069155693,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 106926
+    },
+    {
+      "epoch": 10.981719215364075,
+      "grad_norm": 0.09781994670629501,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 106929
+    },
+    {
+      "epoch": 10.982027318475916,
+      "grad_norm": 0.04272658750414848,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 106932
+    },
+    {
+      "epoch": 10.982335421587758,
+      "grad_norm": 0.04117373377084732,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 106935
+    },
+    {
+      "epoch": 10.982643524699599,
+      "grad_norm": 0.0534319207072258,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 106938
+    },
+    {
+      "epoch": 10.98295162781144,
+      "grad_norm": 0.06119696795940399,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 106941
+    },
+    {
+      "epoch": 10.983259730923283,
+      "grad_norm": 0.03843209147453308,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 106944
+    },
+    {
+      "epoch": 10.983567834035124,
+      "grad_norm": 0.04389924556016922,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 106947
+    },
+    {
+      "epoch": 10.983875937146966,
+      "grad_norm": 0.056084759533405304,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 106950
+    },
+    {
+      "epoch": 10.984184040258807,
+      "grad_norm": 0.09229186177253723,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 106953
+    },
+    {
+      "epoch": 10.984492143370648,
+      "grad_norm": 0.10481631010770798,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 106956
+    },
+    {
+      "epoch": 10.98480024648249,
+      "grad_norm": 0.12012122571468353,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 106959
+    },
+    {
+      "epoch": 10.98510834959433,
+      "grad_norm": 0.06024550646543503,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 106962
+    },
+    {
+      "epoch": 10.985416452706172,
+      "grad_norm": 0.06280180811882019,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 106965
+    },
+    {
+      "epoch": 10.985724555818013,
+      "grad_norm": 0.045513831079006195,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 106968
+    },
+    {
+      "epoch": 10.986032658929854,
+      "grad_norm": 0.07376816123723984,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 106971
+    },
+    {
+      "epoch": 10.986340762041696,
+      "grad_norm": 0.05059327185153961,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 106974
+    },
+    {
+      "epoch": 10.986648865153539,
+      "grad_norm": 0.06152288243174553,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 106977
+    },
+    {
+      "epoch": 10.98695696826538,
+      "grad_norm": 0.05427305027842522,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 106980
+    },
+    {
+      "epoch": 10.987265071377221,
+      "grad_norm": 0.08136880397796631,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 106983
+    },
+    {
+      "epoch": 10.987573174489063,
+      "grad_norm": 0.07303925603628159,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 106986
+    },
+    {
+      "epoch": 10.987881277600904,
+      "grad_norm": 0.08733496814966202,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 106989
+    },
+    {
+      "epoch": 10.988189380712745,
+      "grad_norm": 0.09592670202255249,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 106992
+    },
+    {
+      "epoch": 10.988497483824586,
+      "grad_norm": 0.05195619538426399,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 106995
+    },
+    {
+      "epoch": 10.988805586936428,
+      "grad_norm": 0.06398551166057587,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 106998
+    },
+    {
+      "epoch": 10.989113690048269,
+      "grad_norm": 0.10886025428771973,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 107001
+    },
+    {
+      "epoch": 10.98942179316011,
+      "grad_norm": 0.04574966058135033,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 107004
+    },
+    {
+      "epoch": 10.989729896271953,
+      "grad_norm": 0.04682609811425209,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 107007
+    },
+    {
+      "epoch": 10.990037999383794,
+      "grad_norm": 0.10652533173561096,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 107010
+    },
+    {
+      "epoch": 10.990346102495636,
+      "grad_norm": 0.15787062048912048,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 107013
+    },
+    {
+      "epoch": 10.990654205607477,
+      "grad_norm": 0.04114218056201935,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 107016
+    },
+    {
+      "epoch": 10.990962308719318,
+      "grad_norm": 0.05407201871275902,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 107019
+    },
+    {
+      "epoch": 10.99127041183116,
+      "grad_norm": 0.12445944547653198,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 107022
+    },
+    {
+      "epoch": 10.991578514943,
+      "grad_norm": 0.08229243010282516,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 107025
+    },
+    {
+      "epoch": 10.991886618054842,
+      "grad_norm": 0.047715965658426285,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 107028
+    },
+    {
+      "epoch": 10.992194721166683,
+      "grad_norm": 0.04792723432183266,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 107031
+    },
+    {
+      "epoch": 10.992502824278525,
+      "grad_norm": 0.05532047525048256,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 107034
+    },
+    {
+      "epoch": 10.992810927390366,
+      "grad_norm": 0.045810431241989136,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 107037
+    },
+    {
+      "epoch": 10.993119030502209,
+      "grad_norm": 0.11059387773275375,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 107040
+    },
+    {
+      "epoch": 10.99342713361405,
+      "grad_norm": 0.05643268674612045,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 107043
+    },
+    {
+      "epoch": 10.993735236725891,
+      "grad_norm": 0.050066348165273666,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 107046
+    },
+    {
+      "epoch": 10.994043339837733,
+      "grad_norm": 0.06248445808887482,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 107049
+    },
+    {
+      "epoch": 10.994351442949574,
+      "grad_norm": 0.05078605189919472,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 107052
+    },
+    {
+      "epoch": 10.994659546061415,
+      "grad_norm": 0.11182273924350739,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 107055
+    },
+    {
+      "epoch": 10.994967649173256,
+      "grad_norm": 0.09776394814252853,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 107058
+    },
+    {
+      "epoch": 10.995275752285098,
+      "grad_norm": 0.047179535031318665,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 107061
+    },
+    {
+      "epoch": 10.995583855396939,
+      "grad_norm": 0.03766768425703049,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 107064
+    },
+    {
+      "epoch": 10.99589195850878,
+      "grad_norm": 0.04526562616229057,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 107067
+    },
+    {
+      "epoch": 10.996200061620623,
+      "grad_norm": 0.1715056300163269,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 107070
+    },
+    {
+      "epoch": 10.996508164732465,
+      "grad_norm": 0.13546305894851685,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 107073
+    },
+    {
+      "epoch": 10.996816267844306,
+      "grad_norm": 0.050267528742551804,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 107076
+    },
+    {
+      "epoch": 10.997124370956147,
+      "grad_norm": 0.05380922183394432,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 107079
+    },
+    {
+      "epoch": 10.997432474067988,
+      "grad_norm": 0.046985093504190445,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 107082
+    },
+    {
+      "epoch": 10.99774057717983,
+      "grad_norm": 0.0655679777264595,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 107085
+    },
+    {
+      "epoch": 10.99804868029167,
+      "grad_norm": 0.04053609445691109,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 107088
+    },
+    {
+      "epoch": 10.998356783403512,
+      "grad_norm": 0.03527331352233887,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 107091
+    },
+    {
+      "epoch": 10.998664886515353,
+      "grad_norm": 0.033256895840168,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 107094
+    },
+    {
+      "epoch": 10.998972989627195,
+      "grad_norm": 0.09747351706027985,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 107097
+    },
+    {
+      "epoch": 10.999281092739036,
+      "grad_norm": 0.06879878789186478,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 107100
+    },
+    {
+      "epoch": 10.999589195850879,
+      "grad_norm": 0.09804810583591461,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 107103
+    },
+    {
+      "epoch": 10.99989729896272,
+      "grad_norm": 0.08518742024898529,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 107106
+    },
+    {
+      "epoch": 11.008119218910586,
+      "grad_norm": 0.10636568069458008,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 107109
+    },
+    {
+      "epoch": 11.008427543679343,
+      "grad_norm": 0.0721297413110733,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 107112
+    },
+    {
+      "epoch": 11.008735868448099,
+      "grad_norm": 0.06640434265136719,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 107115
+    },
+    {
+      "epoch": 11.009044193216855,
+      "grad_norm": 0.04051525518298149,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 107118
+    },
+    {
+      "epoch": 11.009352517985612,
+      "grad_norm": 0.058860983699560165,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 107121
+    },
+    {
+      "epoch": 11.009660842754368,
+      "grad_norm": 0.0524006150662899,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 107124
+    },
+    {
+      "epoch": 11.009969167523124,
+      "grad_norm": 0.1158260628581047,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 107127
+    },
+    {
+      "epoch": 11.01027749229188,
+      "grad_norm": 0.0887027382850647,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 107130
+    },
+    {
+      "epoch": 11.010585817060637,
+      "grad_norm": 0.054239798337221146,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 107133
+    },
+    {
+      "epoch": 11.010894141829393,
+      "grad_norm": 0.062151018530130386,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 107136
+    },
+    {
+      "epoch": 11.01120246659815,
+      "grad_norm": 0.10690824687480927,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 107139
+    },
+    {
+      "epoch": 11.011510791366906,
+      "grad_norm": 0.09708454459905624,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 107142
+    },
+    {
+      "epoch": 11.011819116135664,
+      "grad_norm": 0.05205899477005005,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 107145
+    },
+    {
+      "epoch": 11.01212744090442,
+      "grad_norm": 0.0874539464712143,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 107148
+    },
+    {
+      "epoch": 11.012435765673176,
+      "grad_norm": 0.054691608995199203,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 107151
+    },
+    {
+      "epoch": 11.012744090441933,
+      "grad_norm": 0.05851634964346886,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 107154
+    },
+    {
+      "epoch": 11.013052415210689,
+      "grad_norm": 0.05712060630321503,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 107157
+    },
+    {
+      "epoch": 11.013360739979445,
+      "grad_norm": 0.07348459213972092,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 107160
+    },
+    {
+      "epoch": 11.013669064748202,
+      "grad_norm": 0.0613148994743824,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 107163
+    },
+    {
+      "epoch": 11.013977389516958,
+      "grad_norm": 0.09934704750776291,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 107166
+    },
+    {
+      "epoch": 11.014285714285714,
+      "grad_norm": 0.04725782200694084,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 107169
+    },
+    {
+      "epoch": 11.01459403905447,
+      "grad_norm": 0.050498757511377335,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 107172
+    },
+    {
+      "epoch": 11.014902363823227,
+      "grad_norm": 0.04003717005252838,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 107175
+    },
+    {
+      "epoch": 11.015210688591983,
+      "grad_norm": 0.11621799319982529,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 107178
+    },
+    {
+      "epoch": 11.01551901336074,
+      "grad_norm": 0.060241155326366425,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 107181
+    },
+    {
+      "epoch": 11.015827338129496,
+      "grad_norm": 0.07701603323221207,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 107184
+    },
+    {
+      "epoch": 11.016135662898252,
+      "grad_norm": 0.10775244981050491,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 107187
+    },
+    {
+      "epoch": 11.016443987667008,
+      "grad_norm": 0.1312934160232544,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 107190
+    },
+    {
+      "epoch": 11.016752312435766,
+      "grad_norm": 0.20268575847148895,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 107193
+    },
+    {
+      "epoch": 11.017060637204523,
+      "grad_norm": 0.15342745184898376,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 107196
+    },
+    {
+      "epoch": 11.017368961973279,
+      "grad_norm": 0.08270037919282913,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 107199
+    },
+    {
+      "epoch": 11.017677286742035,
+      "grad_norm": 0.053513944149017334,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 107202
+    },
+    {
+      "epoch": 11.017985611510792,
+      "grad_norm": 0.051726341247558594,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 107205
+    },
+    {
+      "epoch": 11.018293936279548,
+      "grad_norm": 0.04074995592236519,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 107208
+    },
+    {
+      "epoch": 11.018602261048304,
+      "grad_norm": 0.045796941965818405,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 107211
+    },
+    {
+      "epoch": 11.01891058581706,
+      "grad_norm": 0.07273279130458832,
+      "learning_rate": 0.01,
+      "loss": 1.9717,
+      "step": 107214
+    },
+    {
+      "epoch": 11.019218910585817,
+      "grad_norm": 0.060059960931539536,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 107217
+    },
+    {
+      "epoch": 11.019527235354573,
+      "grad_norm": 0.03590172901749611,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 107220
+    },
+    {
+      "epoch": 11.01983556012333,
+      "grad_norm": 0.09819181263446808,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 107223
+    },
+    {
+      "epoch": 11.020143884892086,
+      "grad_norm": 0.11009673029184341,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 107226
+    },
+    {
+      "epoch": 11.020452209660842,
+      "grad_norm": 0.0914946123957634,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 107229
+    },
+    {
+      "epoch": 11.020760534429598,
+      "grad_norm": 0.12640529870986938,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 107232
+    },
+    {
+      "epoch": 11.021068859198355,
+      "grad_norm": 0.1176491379737854,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 107235
+    },
+    {
+      "epoch": 11.021377183967113,
+      "grad_norm": 0.06920988112688065,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 107238
+    },
+    {
+      "epoch": 11.02168550873587,
+      "grad_norm": 0.034434013068675995,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 107241
+    },
+    {
+      "epoch": 11.021993833504625,
+      "grad_norm": 0.04428716003894806,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 107244
+    },
+    {
+      "epoch": 11.022302158273382,
+      "grad_norm": 0.06214417517185211,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 107247
+    },
+    {
+      "epoch": 11.022610483042138,
+      "grad_norm": 0.053140249103307724,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 107250
+    },
+    {
+      "epoch": 11.022918807810894,
+      "grad_norm": 0.05963686481118202,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 107253
+    },
+    {
+      "epoch": 11.02322713257965,
+      "grad_norm": 0.12098684161901474,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 107256
+    },
+    {
+      "epoch": 11.023535457348407,
+      "grad_norm": 0.06041080132126808,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 107259
+    },
+    {
+      "epoch": 11.023843782117163,
+      "grad_norm": 0.08065960556268692,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 107262
+    },
+    {
+      "epoch": 11.02415210688592,
+      "grad_norm": 0.05811164900660515,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 107265
+    },
+    {
+      "epoch": 11.024460431654676,
+      "grad_norm": 0.11217566579580307,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 107268
+    },
+    {
+      "epoch": 11.024768756423432,
+      "grad_norm": 0.06718678772449493,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 107271
+    },
+    {
+      "epoch": 11.025077081192189,
+      "grad_norm": 0.05535897985100746,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 107274
+    },
+    {
+      "epoch": 11.025385405960945,
+      "grad_norm": 0.07758058607578278,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 107277
+    },
+    {
+      "epoch": 11.025693730729701,
+      "grad_norm": 0.11350016295909882,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 107280
+    },
+    {
+      "epoch": 11.02600205549846,
+      "grad_norm": 0.09810200333595276,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 107283
+    },
+    {
+      "epoch": 11.026310380267216,
+      "grad_norm": 0.1021398976445198,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 107286
+    },
+    {
+      "epoch": 11.026618705035972,
+      "grad_norm": 0.051616981625556946,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 107289
+    },
+    {
+      "epoch": 11.026927029804728,
+      "grad_norm": 0.039984527975320816,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 107292
+    },
+    {
+      "epoch": 11.027235354573484,
+      "grad_norm": 0.05325577035546303,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 107295
+    },
+    {
+      "epoch": 11.02754367934224,
+      "grad_norm": 0.06299860030412674,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 107298
+    },
+    {
+      "epoch": 11.027852004110997,
+      "grad_norm": 0.03828146308660507,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 107301
+    },
+    {
+      "epoch": 11.028160328879753,
+      "grad_norm": 0.04722427949309349,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 107304
+    },
+    {
+      "epoch": 11.02846865364851,
+      "grad_norm": 0.06583574414253235,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 107307
+    },
+    {
+      "epoch": 11.028776978417266,
+      "grad_norm": 0.1033751517534256,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 107310
+    },
+    {
+      "epoch": 11.029085303186022,
+      "grad_norm": 0.08672161400318146,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 107313
+    },
+    {
+      "epoch": 11.029393627954779,
+      "grad_norm": 0.04516211152076721,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 107316
+    },
+    {
+      "epoch": 11.029701952723535,
+      "grad_norm": 0.03711722046136856,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 107319
+    },
+    {
+      "epoch": 11.030010277492291,
+      "grad_norm": 0.033995870500802994,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 107322
+    },
+    {
+      "epoch": 11.030318602261048,
+      "grad_norm": 0.04240257292985916,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 107325
+    },
+    {
+      "epoch": 11.030626927029804,
+      "grad_norm": 0.06180763244628906,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 107328
+    },
+    {
+      "epoch": 11.030935251798562,
+      "grad_norm": 0.13815215229988098,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 107331
+    },
+    {
+      "epoch": 11.031243576567318,
+      "grad_norm": 0.10653617233037949,
+      "learning_rate": 0.01,
+      "loss": 1.9743,
+      "step": 107334
+    },
+    {
+      "epoch": 11.031551901336075,
+      "grad_norm": 0.05722755938768387,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 107337
+    },
+    {
+      "epoch": 11.03186022610483,
+      "grad_norm": 0.044469717890024185,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 107340
+    },
+    {
+      "epoch": 11.032168550873587,
+      "grad_norm": 0.029477711766958237,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 107343
+    },
+    {
+      "epoch": 11.032476875642343,
+      "grad_norm": 0.05988297611474991,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 107346
+    },
+    {
+      "epoch": 11.0327852004111,
+      "grad_norm": 0.11533161252737045,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 107349
+    },
+    {
+      "epoch": 11.033093525179856,
+      "grad_norm": 0.09519381821155548,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 107352
+    },
+    {
+      "epoch": 11.033401849948612,
+      "grad_norm": 0.06721679866313934,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 107355
+    },
+    {
+      "epoch": 11.033710174717369,
+      "grad_norm": 0.051016733050346375,
+      "learning_rate": 0.01,
+      "loss": 1.9784,
+      "step": 107358
+    },
+    {
+      "epoch": 11.034018499486125,
+      "grad_norm": 0.06869174540042877,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 107361
+    },
+    {
+      "epoch": 11.034326824254881,
+      "grad_norm": 0.038322094827890396,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 107364
+    },
+    {
+      "epoch": 11.034635149023638,
+      "grad_norm": 0.17970217764377594,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 107367
+    },
+    {
+      "epoch": 11.034943473792394,
+      "grad_norm": 0.0593324676156044,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 107370
+    },
+    {
+      "epoch": 11.03525179856115,
+      "grad_norm": 0.03376584127545357,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 107373
+    },
+    {
+      "epoch": 11.035560123329908,
+      "grad_norm": 0.03950095176696777,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 107376
+    },
+    {
+      "epoch": 11.035868448098665,
+      "grad_norm": 0.03454350680112839,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 107379
+    },
+    {
+      "epoch": 11.036176772867421,
+      "grad_norm": 0.037530187517404556,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 107382
+    },
+    {
+      "epoch": 11.036485097636177,
+      "grad_norm": 0.13739140331745148,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 107385
+    },
+    {
+      "epoch": 11.036793422404934,
+      "grad_norm": 0.06409517675638199,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 107388
+    },
+    {
+      "epoch": 11.03710174717369,
+      "grad_norm": 0.07674208283424377,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 107391
+    },
+    {
+      "epoch": 11.037410071942446,
+      "grad_norm": 0.04071156308054924,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 107394
+    },
+    {
+      "epoch": 11.037718396711202,
+      "grad_norm": 0.04651397839188576,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 107397
+    },
+    {
+      "epoch": 11.038026721479959,
+      "grad_norm": 0.04479502886533737,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 107400
+    },
+    {
+      "epoch": 11.038335046248715,
+      "grad_norm": 0.03764665499329567,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 107403
+    },
+    {
+      "epoch": 11.038643371017471,
+      "grad_norm": 0.03378015384078026,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 107406
+    },
+    {
+      "epoch": 11.038951695786228,
+      "grad_norm": 0.04425158351659775,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 107409
+    },
+    {
+      "epoch": 11.039260020554984,
+      "grad_norm": 0.07492903620004654,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 107412
+    },
+    {
+      "epoch": 11.03956834532374,
+      "grad_norm": 0.15413755178451538,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 107415
+    },
+    {
+      "epoch": 11.039876670092497,
+      "grad_norm": 0.06271611899137497,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 107418
+    },
+    {
+      "epoch": 11.040184994861255,
+      "grad_norm": 0.08407186716794968,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 107421
+    },
+    {
+      "epoch": 11.040493319630011,
+      "grad_norm": 0.07308407872915268,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 107424
+    },
+    {
+      "epoch": 11.040801644398767,
+      "grad_norm": 0.0837329849600792,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 107427
+    },
+    {
+      "epoch": 11.041109969167524,
+      "grad_norm": 0.07305663079023361,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 107430
+    },
+    {
+      "epoch": 11.04141829393628,
+      "grad_norm": 0.06407970190048218,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 107433
+    },
+    {
+      "epoch": 11.041726618705036,
+      "grad_norm": 0.06141374260187149,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 107436
+    },
+    {
+      "epoch": 11.042034943473793,
+      "grad_norm": 0.12983256578445435,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 107439
+    },
+    {
+      "epoch": 11.042343268242549,
+      "grad_norm": 0.1181797981262207,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 107442
+    },
+    {
+      "epoch": 11.042651593011305,
+      "grad_norm": 0.06571821868419647,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 107445
+    },
+    {
+      "epoch": 11.042959917780061,
+      "grad_norm": 0.047969378530979156,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 107448
+    },
+    {
+      "epoch": 11.043268242548818,
+      "grad_norm": 0.04128817096352577,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 107451
+    },
+    {
+      "epoch": 11.043576567317574,
+      "grad_norm": 0.0404379703104496,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 107454
+    },
+    {
+      "epoch": 11.04388489208633,
+      "grad_norm": 0.07635513693094254,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 107457
+    },
+    {
+      "epoch": 11.044193216855087,
+      "grad_norm": 0.09891491383314133,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 107460
+    },
+    {
+      "epoch": 11.044501541623843,
+      "grad_norm": 0.08441543579101562,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 107463
+    },
+    {
+      "epoch": 11.044809866392601,
+      "grad_norm": 0.10859276354312897,
+      "learning_rate": 0.01,
+      "loss": 1.9794,
+      "step": 107466
+    },
+    {
+      "epoch": 11.045118191161357,
+      "grad_norm": 0.10530184209346771,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 107469
+    },
+    {
+      "epoch": 11.045426515930114,
+      "grad_norm": 0.04014790803194046,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 107472
+    },
+    {
+      "epoch": 11.04573484069887,
+      "grad_norm": 0.09878093749284744,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 107475
+    },
+    {
+      "epoch": 11.046043165467626,
+      "grad_norm": 0.08944501727819443,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 107478
+    },
+    {
+      "epoch": 11.046351490236383,
+      "grad_norm": 0.03847057744860649,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 107481
+    },
+    {
+      "epoch": 11.046659815005139,
+      "grad_norm": 0.06581749767065048,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 107484
+    },
+    {
+      "epoch": 11.046968139773895,
+      "grad_norm": 0.08069849759340286,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 107487
+    },
+    {
+      "epoch": 11.047276464542652,
+      "grad_norm": 0.049780331552028656,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 107490
+    },
+    {
+      "epoch": 11.047584789311408,
+      "grad_norm": 0.06270769983530045,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 107493
+    },
+    {
+      "epoch": 11.047893114080164,
+      "grad_norm": 0.055340807884931564,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 107496
+    },
+    {
+      "epoch": 11.04820143884892,
+      "grad_norm": 0.09283401072025299,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 107499
+    },
+    {
+      "epoch": 11.048509763617677,
+      "grad_norm": 0.0609809048473835,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 107502
+    },
+    {
+      "epoch": 11.048818088386433,
+      "grad_norm": 0.13390469551086426,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 107505
+    },
+    {
+      "epoch": 11.04912641315519,
+      "grad_norm": 0.0418582409620285,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 107508
+    },
+    {
+      "epoch": 11.049434737923946,
+      "grad_norm": 0.06463424861431122,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 107511
+    },
+    {
+      "epoch": 11.049743062692704,
+      "grad_norm": 0.05462968349456787,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 107514
+    },
+    {
+      "epoch": 11.05005138746146,
+      "grad_norm": 0.040294911712408066,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 107517
+    },
+    {
+      "epoch": 11.050359712230216,
+      "grad_norm": 0.07260268926620483,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 107520
+    },
+    {
+      "epoch": 11.050668036998973,
+      "grad_norm": 0.10488167405128479,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 107523
+    },
+    {
+      "epoch": 11.050976361767729,
+      "grad_norm": 0.043960295617580414,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 107526
+    },
+    {
+      "epoch": 11.051284686536485,
+      "grad_norm": 0.0684865415096283,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 107529
+    },
+    {
+      "epoch": 11.051593011305242,
+      "grad_norm": 0.08690311014652252,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 107532
+    },
+    {
+      "epoch": 11.051901336073998,
+      "grad_norm": 0.053025078028440475,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 107535
+    },
+    {
+      "epoch": 11.052209660842754,
+      "grad_norm": 0.07714751362800598,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 107538
+    },
+    {
+      "epoch": 11.05251798561151,
+      "grad_norm": 0.06200997158885002,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 107541
+    },
+    {
+      "epoch": 11.052826310380267,
+      "grad_norm": 0.08136401325464249,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 107544
+    },
+    {
+      "epoch": 11.053134635149023,
+      "grad_norm": 0.047901589423418045,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 107547
+    },
+    {
+      "epoch": 11.05344295991778,
+      "grad_norm": 0.03807402402162552,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 107550
+    },
+    {
+      "epoch": 11.053751284686536,
+      "grad_norm": 0.06316124647855759,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 107553
+    },
+    {
+      "epoch": 11.054059609455292,
+      "grad_norm": 0.05572817102074623,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 107556
+    },
+    {
+      "epoch": 11.05436793422405,
+      "grad_norm": 0.04559632018208504,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 107559
+    },
+    {
+      "epoch": 11.054676258992806,
+      "grad_norm": 0.07237659394741058,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 107562
+    },
+    {
+      "epoch": 11.054984583761563,
+      "grad_norm": 0.092888742685318,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 107565
+    },
+    {
+      "epoch": 11.055292908530319,
+      "grad_norm": 0.08010265231132507,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 107568
+    },
+    {
+      "epoch": 11.055601233299075,
+      "grad_norm": 0.08017858117818832,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 107571
+    },
+    {
+      "epoch": 11.055909558067832,
+      "grad_norm": 0.07664944976568222,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 107574
+    },
+    {
+      "epoch": 11.056217882836588,
+      "grad_norm": 0.04926150292158127,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 107577
+    },
+    {
+      "epoch": 11.056526207605344,
+      "grad_norm": 0.04636576399207115,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 107580
+    },
+    {
+      "epoch": 11.0568345323741,
+      "grad_norm": 0.053231917321681976,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 107583
+    },
+    {
+      "epoch": 11.057142857142857,
+      "grad_norm": 0.06848327815532684,
+      "learning_rate": 0.01,
+      "loss": 1.979,
+      "step": 107586
+    },
+    {
+      "epoch": 11.057451181911613,
+      "grad_norm": 0.05932049825787544,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 107589
+    },
+    {
+      "epoch": 11.05775950668037,
+      "grad_norm": 0.040187012404203415,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 107592
+    },
+    {
+      "epoch": 11.058067831449126,
+      "grad_norm": 0.21725735068321228,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 107595
+    },
+    {
+      "epoch": 11.058376156217882,
+      "grad_norm": 0.10536807775497437,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 107598
+    },
+    {
+      "epoch": 11.058684480986638,
+      "grad_norm": 0.07488951086997986,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 107601
+    },
+    {
+      "epoch": 11.058992805755397,
+      "grad_norm": 0.04247715696692467,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 107604
+    },
+    {
+      "epoch": 11.059301130524153,
+      "grad_norm": 0.033919550478458405,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 107607
+    },
+    {
+      "epoch": 11.05960945529291,
+      "grad_norm": 0.04327520728111267,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 107610
+    },
+    {
+      "epoch": 11.059917780061665,
+      "grad_norm": 0.08149202167987823,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 107613
+    },
+    {
+      "epoch": 11.060226104830422,
+      "grad_norm": 0.04497545585036278,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 107616
+    },
+    {
+      "epoch": 11.060534429599178,
+      "grad_norm": 0.049667440354824066,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 107619
+    },
+    {
+      "epoch": 11.060842754367934,
+      "grad_norm": 0.040313173085451126,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 107622
+    },
+    {
+      "epoch": 11.06115107913669,
+      "grad_norm": 0.042108844965696335,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 107625
+    },
+    {
+      "epoch": 11.061459403905447,
+      "grad_norm": 0.13305234909057617,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 107628
+    },
+    {
+      "epoch": 11.061767728674203,
+      "grad_norm": 0.05455876141786575,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 107631
+    },
+    {
+      "epoch": 11.06207605344296,
+      "grad_norm": 0.07920195162296295,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 107634
+    },
+    {
+      "epoch": 11.062384378211716,
+      "grad_norm": 0.07053566724061966,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 107637
+    },
+    {
+      "epoch": 11.062692702980472,
+      "grad_norm": 0.0445457398891449,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 107640
+    },
+    {
+      "epoch": 11.063001027749229,
+      "grad_norm": 0.044817425310611725,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 107643
+    },
+    {
+      "epoch": 11.063309352517985,
+      "grad_norm": 0.12310293316841125,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 107646
+    },
+    {
+      "epoch": 11.063617677286741,
+      "grad_norm": 0.04152672365307808,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 107649
+    },
+    {
+      "epoch": 11.0639260020555,
+      "grad_norm": 0.07941756397485733,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 107652
+    },
+    {
+      "epoch": 11.064234326824256,
+      "grad_norm": 0.04817092791199684,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 107655
+    },
+    {
+      "epoch": 11.064542651593012,
+      "grad_norm": 0.04424908757209778,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 107658
+    },
+    {
+      "epoch": 11.064850976361768,
+      "grad_norm": 0.1192043200135231,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 107661
+    },
+    {
+      "epoch": 11.065159301130524,
+      "grad_norm": 0.07869096100330353,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 107664
+    },
+    {
+      "epoch": 11.06546762589928,
+      "grad_norm": 0.06656874716281891,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 107667
+    },
+    {
+      "epoch": 11.065775950668037,
+      "grad_norm": 0.04989762231707573,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 107670
+    },
+    {
+      "epoch": 11.066084275436793,
+      "grad_norm": 0.06311410665512085,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 107673
+    },
+    {
+      "epoch": 11.06639260020555,
+      "grad_norm": 0.059550900012254715,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 107676
+    },
+    {
+      "epoch": 11.066700924974306,
+      "grad_norm": 0.058306027203798294,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 107679
+    },
+    {
+      "epoch": 11.067009249743062,
+      "grad_norm": 0.042203981429338455,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 107682
+    },
+    {
+      "epoch": 11.067317574511819,
+      "grad_norm": 0.03726264089345932,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 107685
+    },
+    {
+      "epoch": 11.067625899280575,
+      "grad_norm": 0.06475093215703964,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 107688
+    },
+    {
+      "epoch": 11.067934224049331,
+      "grad_norm": 0.06762620806694031,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 107691
+    },
+    {
+      "epoch": 11.068242548818088,
+      "grad_norm": 0.11405432224273682,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 107694
+    },
+    {
+      "epoch": 11.068550873586846,
+      "grad_norm": 0.07248923927545547,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 107697
+    },
+    {
+      "epoch": 11.068859198355602,
+      "grad_norm": 0.12362180650234222,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 107700
+    },
+    {
+      "epoch": 11.069167523124358,
+      "grad_norm": 0.05302852764725685,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 107703
+    },
+    {
+      "epoch": 11.069475847893115,
+      "grad_norm": 0.05188078433275223,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 107706
+    },
+    {
+      "epoch": 11.06978417266187,
+      "grad_norm": 0.11408820003271103,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 107709
+    },
+    {
+      "epoch": 11.070092497430627,
+      "grad_norm": 0.06259067356586456,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 107712
+    },
+    {
+      "epoch": 11.070400822199383,
+      "grad_norm": 0.042797449976205826,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 107715
+    },
+    {
+      "epoch": 11.07070914696814,
+      "grad_norm": 0.03658578172326088,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 107718
+    },
+    {
+      "epoch": 11.071017471736896,
+      "grad_norm": 0.09516727924346924,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 107721
+    },
+    {
+      "epoch": 11.071325796505652,
+      "grad_norm": 0.0809289962053299,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 107724
+    },
+    {
+      "epoch": 11.071634121274409,
+      "grad_norm": 0.04550709202885628,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 107727
+    },
+    {
+      "epoch": 11.071942446043165,
+      "grad_norm": 0.09415169060230255,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 107730
+    },
+    {
+      "epoch": 11.072250770811921,
+      "grad_norm": 0.11107844114303589,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 107733
+    },
+    {
+      "epoch": 11.072559095580678,
+      "grad_norm": 0.045347124338150024,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 107736
+    },
+    {
+      "epoch": 11.072867420349434,
+      "grad_norm": 0.07700034976005554,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 107739
+    },
+    {
+      "epoch": 11.073175745118192,
+      "grad_norm": 0.07656693458557129,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 107742
+    },
+    {
+      "epoch": 11.073484069886948,
+      "grad_norm": 0.038799382746219635,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 107745
+    },
+    {
+      "epoch": 11.073792394655705,
+      "grad_norm": 0.09669993072748184,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 107748
+    },
+    {
+      "epoch": 11.074100719424461,
+      "grad_norm": 0.06955236941576004,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 107751
+    },
+    {
+      "epoch": 11.074409044193217,
+      "grad_norm": 0.05242108553647995,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 107754
+    },
+    {
+      "epoch": 11.074717368961974,
+      "grad_norm": 0.07128479331731796,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 107757
+    },
+    {
+      "epoch": 11.07502569373073,
+      "grad_norm": 0.1151743233203888,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 107760
+    },
+    {
+      "epoch": 11.075334018499486,
+      "grad_norm": 0.04790006950497627,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 107763
+    },
+    {
+      "epoch": 11.075642343268242,
+      "grad_norm": 0.03786927089095116,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 107766
+    },
+    {
+      "epoch": 11.075950668036999,
+      "grad_norm": 0.16423648595809937,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 107769
+    },
+    {
+      "epoch": 11.076258992805755,
+      "grad_norm": 0.0376586988568306,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 107772
+    },
+    {
+      "epoch": 11.076567317574511,
+      "grad_norm": 0.0358593575656414,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 107775
+    },
+    {
+      "epoch": 11.076875642343268,
+      "grad_norm": 0.0530848503112793,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 107778
+    },
+    {
+      "epoch": 11.077183967112024,
+      "grad_norm": 0.04405958577990532,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 107781
+    },
+    {
+      "epoch": 11.07749229188078,
+      "grad_norm": 0.04691191390156746,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 107784
+    },
+    {
+      "epoch": 11.077800616649537,
+      "grad_norm": 0.09400805830955505,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 107787
+    },
+    {
+      "epoch": 11.078108941418295,
+      "grad_norm": 0.04769006371498108,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 107790
+    },
+    {
+      "epoch": 11.078417266187051,
+      "grad_norm": 0.22307366132736206,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 107793
+    },
+    {
+      "epoch": 11.078725590955807,
+      "grad_norm": 0.08780405670404434,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 107796
+    },
+    {
+      "epoch": 11.079033915724564,
+      "grad_norm": 0.06174111366271973,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 107799
+    },
+    {
+      "epoch": 11.07934224049332,
+      "grad_norm": 0.0673871785402298,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 107802
+    },
+    {
+      "epoch": 11.079650565262076,
+      "grad_norm": 0.04536505043506622,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 107805
+    },
+    {
+      "epoch": 11.079958890030833,
+      "grad_norm": 0.06448676437139511,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 107808
+    },
+    {
+      "epoch": 11.080267214799589,
+      "grad_norm": 0.03986390680074692,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 107811
+    },
+    {
+      "epoch": 11.080575539568345,
+      "grad_norm": 0.03483456373214722,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 107814
+    },
+    {
+      "epoch": 11.080883864337101,
+      "grad_norm": 0.04754864424467087,
+      "learning_rate": 0.01,
+      "loss": 1.9782,
+      "step": 107817
+    },
+    {
+      "epoch": 11.081192189105858,
+      "grad_norm": 0.04833069071173668,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 107820
+    },
+    {
+      "epoch": 11.081500513874614,
+      "grad_norm": 0.044434234499931335,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 107823
+    },
+    {
+      "epoch": 11.08180883864337,
+      "grad_norm": 0.04959580674767494,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 107826
+    },
+    {
+      "epoch": 11.082117163412127,
+      "grad_norm": 0.16906990110874176,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 107829
+    },
+    {
+      "epoch": 11.082425488180883,
+      "grad_norm": 0.04020620882511139,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 107832
+    },
+    {
+      "epoch": 11.082733812949641,
+      "grad_norm": 0.04799540340900421,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 107835
+    },
+    {
+      "epoch": 11.083042137718397,
+      "grad_norm": 0.06791256368160248,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 107838
+    },
+    {
+      "epoch": 11.083350462487154,
+      "grad_norm": 0.06625140458345413,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 107841
+    },
+    {
+      "epoch": 11.08365878725591,
+      "grad_norm": 0.04998164623975754,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 107844
+    },
+    {
+      "epoch": 11.083967112024666,
+      "grad_norm": 0.04162587970495224,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 107847
+    },
+    {
+      "epoch": 11.084275436793423,
+      "grad_norm": 0.049901679158210754,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 107850
+    },
+    {
+      "epoch": 11.084583761562179,
+      "grad_norm": 0.04299694299697876,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 107853
+    },
+    {
+      "epoch": 11.084892086330935,
+      "grad_norm": 0.04808099940419197,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 107856
+    },
+    {
+      "epoch": 11.085200411099692,
+      "grad_norm": 0.09989727288484573,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 107859
+    },
+    {
+      "epoch": 11.085508735868448,
+      "grad_norm": 0.047710586339235306,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 107862
+    },
+    {
+      "epoch": 11.085817060637204,
+      "grad_norm": 0.06054253876209259,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 107865
+    },
+    {
+      "epoch": 11.08612538540596,
+      "grad_norm": 0.05632666498422623,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 107868
+    },
+    {
+      "epoch": 11.086433710174717,
+      "grad_norm": 0.046154700219631195,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 107871
+    },
+    {
+      "epoch": 11.086742034943473,
+      "grad_norm": 0.13304947316646576,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 107874
+    },
+    {
+      "epoch": 11.08705035971223,
+      "grad_norm": 0.06604369729757309,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 107877
+    },
+    {
+      "epoch": 11.087358684480987,
+      "grad_norm": 0.04450514167547226,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 107880
+    },
+    {
+      "epoch": 11.087667009249744,
+      "grad_norm": 0.03248503804206848,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 107883
+    },
+    {
+      "epoch": 11.0879753340185,
+      "grad_norm": 0.054392118006944656,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 107886
+    },
+    {
+      "epoch": 11.088283658787256,
+      "grad_norm": 0.07481544464826584,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 107889
+    },
+    {
+      "epoch": 11.088591983556013,
+      "grad_norm": 0.11516573280096054,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 107892
+    },
+    {
+      "epoch": 11.088900308324769,
+      "grad_norm": 0.06140699237585068,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 107895
+    },
+    {
+      "epoch": 11.089208633093525,
+      "grad_norm": 0.04883519187569618,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 107898
+    },
+    {
+      "epoch": 11.089516957862282,
+      "grad_norm": 0.03312469646334648,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 107901
+    },
+    {
+      "epoch": 11.089825282631038,
+      "grad_norm": 0.10212545096874237,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 107904
+    },
+    {
+      "epoch": 11.090133607399794,
+      "grad_norm": 0.09011369198560715,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 107907
+    },
+    {
+      "epoch": 11.09044193216855,
+      "grad_norm": 0.08168302476406097,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 107910
+    },
+    {
+      "epoch": 11.090750256937307,
+      "grad_norm": 0.09307815134525299,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 107913
+    },
+    {
+      "epoch": 11.091058581706063,
+      "grad_norm": 0.06893011182546616,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 107916
+    },
+    {
+      "epoch": 11.09136690647482,
+      "grad_norm": 0.11134271323680878,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 107919
+    },
+    {
+      "epoch": 11.091675231243576,
+      "grad_norm": 0.06912576407194138,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 107922
+    },
+    {
+      "epoch": 11.091983556012334,
+      "grad_norm": 0.052220139652490616,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 107925
+    },
+    {
+      "epoch": 11.09229188078109,
+      "grad_norm": 0.05896226316690445,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 107928
+    },
+    {
+      "epoch": 11.092600205549846,
+      "grad_norm": 0.059881266206502914,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 107931
+    },
+    {
+      "epoch": 11.092908530318603,
+      "grad_norm": 0.07263971865177155,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 107934
+    },
+    {
+      "epoch": 11.093216855087359,
+      "grad_norm": 0.048972729593515396,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 107937
+    },
+    {
+      "epoch": 11.093525179856115,
+      "grad_norm": 0.03646502271294594,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 107940
+    },
+    {
+      "epoch": 11.093833504624872,
+      "grad_norm": 0.03585082292556763,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 107943
+    },
+    {
+      "epoch": 11.094141829393628,
+      "grad_norm": 0.05213327333331108,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 107946
+    },
+    {
+      "epoch": 11.094450154162384,
+      "grad_norm": 0.056754935532808304,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 107949
+    },
+    {
+      "epoch": 11.09475847893114,
+      "grad_norm": 0.06303158402442932,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 107952
+    },
+    {
+      "epoch": 11.095066803699897,
+      "grad_norm": 0.04736315459012985,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 107955
+    },
+    {
+      "epoch": 11.095375128468653,
+      "grad_norm": 0.04783078655600548,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 107958
+    },
+    {
+      "epoch": 11.09568345323741,
+      "grad_norm": 0.0429544523358345,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 107961
+    },
+    {
+      "epoch": 11.095991778006166,
+      "grad_norm": 0.10795792192220688,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 107964
+    },
+    {
+      "epoch": 11.096300102774922,
+      "grad_norm": 0.051838699728250504,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 107967
+    },
+    {
+      "epoch": 11.096608427543678,
+      "grad_norm": 0.14740125834941864,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 107970
+    },
+    {
+      "epoch": 11.096916752312437,
+      "grad_norm": 0.10696642845869064,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 107973
+    },
+    {
+      "epoch": 11.097225077081193,
+      "grad_norm": 0.04325505718588829,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 107976
+    },
+    {
+      "epoch": 11.09753340184995,
+      "grad_norm": 0.03891993314027786,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 107979
+    },
+    {
+      "epoch": 11.097841726618705,
+      "grad_norm": 0.06790821254253387,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 107982
+    },
+    {
+      "epoch": 11.098150051387462,
+      "grad_norm": 0.0682348757982254,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 107985
+    },
+    {
+      "epoch": 11.098458376156218,
+      "grad_norm": 0.04477721452713013,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 107988
+    },
+    {
+      "epoch": 11.098766700924974,
+      "grad_norm": 0.1636943221092224,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 107991
+    },
+    {
+      "epoch": 11.09907502569373,
+      "grad_norm": 0.10245567560195923,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 107994
+    },
+    {
+      "epoch": 11.099383350462487,
+      "grad_norm": 0.09457534551620483,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 107997
+    },
+    {
+      "epoch": 11.099691675231243,
+      "grad_norm": 0.06260280311107635,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 108000
+    },
+    {
+      "epoch": 11.1,
+      "grad_norm": 0.04717349261045456,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 108003
+    },
+    {
+      "epoch": 11.100308324768756,
+      "grad_norm": 0.0400058813393116,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 108006
+    },
+    {
+      "epoch": 11.100616649537512,
+      "grad_norm": 0.04016717150807381,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 108009
+    },
+    {
+      "epoch": 11.100924974306269,
+      "grad_norm": 0.03915245458483696,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 108012
+    },
+    {
+      "epoch": 11.101233299075025,
+      "grad_norm": 0.050850559026002884,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 108015
+    },
+    {
+      "epoch": 11.101541623843783,
+      "grad_norm": 0.055654145777225494,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 108018
+    },
+    {
+      "epoch": 11.10184994861254,
+      "grad_norm": 0.1390688419342041,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 108021
+    },
+    {
+      "epoch": 11.102158273381296,
+      "grad_norm": 0.044200409203767776,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 108024
+    },
+    {
+      "epoch": 11.102466598150052,
+      "grad_norm": 0.03880887106060982,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 108027
+    },
+    {
+      "epoch": 11.102774922918808,
+      "grad_norm": 0.08580953627824783,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 108030
+    },
+    {
+      "epoch": 11.103083247687564,
+      "grad_norm": 0.08826398104429245,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 108033
+    },
+    {
+      "epoch": 11.10339157245632,
+      "grad_norm": 0.06979897618293762,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 108036
+    },
+    {
+      "epoch": 11.103699897225077,
+      "grad_norm": 0.05558208376169205,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 108039
+    },
+    {
+      "epoch": 11.104008221993833,
+      "grad_norm": 0.07635145634412766,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 108042
+    },
+    {
+      "epoch": 11.10431654676259,
+      "grad_norm": 0.06586254388093948,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 108045
+    },
+    {
+      "epoch": 11.104624871531346,
+      "grad_norm": 0.11627889424562454,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 108048
+    },
+    {
+      "epoch": 11.104933196300102,
+      "grad_norm": 0.05880044028162956,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 108051
+    },
+    {
+      "epoch": 11.105241521068859,
+      "grad_norm": 0.04572506994009018,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 108054
+    },
+    {
+      "epoch": 11.105549845837615,
+      "grad_norm": 0.09277968853712082,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 108057
+    },
+    {
+      "epoch": 11.105858170606371,
+      "grad_norm": 0.03708489239215851,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 108060
+    },
+    {
+      "epoch": 11.10616649537513,
+      "grad_norm": 0.11607236415147781,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 108063
+    },
+    {
+      "epoch": 11.106474820143886,
+      "grad_norm": 0.060427650809288025,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 108066
+    },
+    {
+      "epoch": 11.106783144912642,
+      "grad_norm": 0.062446895986795425,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 108069
+    },
+    {
+      "epoch": 11.107091469681398,
+      "grad_norm": 0.059566475450992584,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 108072
+    },
+    {
+      "epoch": 11.107399794450155,
+      "grad_norm": 0.042464662343263626,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 108075
+    },
+    {
+      "epoch": 11.10770811921891,
+      "grad_norm": 0.13029395043849945,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 108078
+    },
+    {
+      "epoch": 11.108016443987667,
+      "grad_norm": 0.08560604602098465,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 108081
+    },
+    {
+      "epoch": 11.108324768756423,
+      "grad_norm": 0.06263414770364761,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 108084
+    },
+    {
+      "epoch": 11.10863309352518,
+      "grad_norm": 0.04986066371202469,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 108087
+    },
+    {
+      "epoch": 11.108941418293936,
+      "grad_norm": 0.0476389154791832,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 108090
+    },
+    {
+      "epoch": 11.109249743062692,
+      "grad_norm": 0.05869157984852791,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 108093
+    },
+    {
+      "epoch": 11.109558067831449,
+      "grad_norm": 0.04113049432635307,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 108096
+    },
+    {
+      "epoch": 11.109866392600205,
+      "grad_norm": 0.06492377817630768,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 108099
+    },
+    {
+      "epoch": 11.110174717368961,
+      "grad_norm": 0.0680563822388649,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 108102
+    },
+    {
+      "epoch": 11.110483042137718,
+      "grad_norm": 0.09744305908679962,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 108105
+    },
+    {
+      "epoch": 11.110791366906474,
+      "grad_norm": 0.090627022087574,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 108108
+    },
+    {
+      "epoch": 11.111099691675232,
+      "grad_norm": 0.07619655877351761,
+      "learning_rate": 0.01,
+      "loss": 1.9634,
+      "step": 108111
+    },
+    {
+      "epoch": 11.111408016443988,
+      "grad_norm": 0.05335146561264992,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 108114
+    },
+    {
+      "epoch": 11.111716341212745,
+      "grad_norm": 0.05738540366292,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 108117
+    },
+    {
+      "epoch": 11.112024665981501,
+      "grad_norm": 0.12277184426784515,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 108120
+    },
+    {
+      "epoch": 11.112332990750257,
+      "grad_norm": 0.05805842578411102,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 108123
+    },
+    {
+      "epoch": 11.112641315519014,
+      "grad_norm": 0.061712007969617844,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 108126
+    },
+    {
+      "epoch": 11.11294964028777,
+      "grad_norm": 0.045319247990846634,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 108129
+    },
+    {
+      "epoch": 11.113257965056526,
+      "grad_norm": 0.1417672485113144,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 108132
+    },
+    {
+      "epoch": 11.113566289825282,
+      "grad_norm": 0.051083218306303024,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 108135
+    },
+    {
+      "epoch": 11.113874614594039,
+      "grad_norm": 0.03624936193227768,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 108138
+    },
+    {
+      "epoch": 11.114182939362795,
+      "grad_norm": 0.05050056055188179,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 108141
+    },
+    {
+      "epoch": 11.114491264131551,
+      "grad_norm": 0.05254499241709709,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 108144
+    },
+    {
+      "epoch": 11.114799588900308,
+      "grad_norm": 0.04745284095406532,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 108147
+    },
+    {
+      "epoch": 11.115107913669064,
+      "grad_norm": 0.05609699711203575,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 108150
+    },
+    {
+      "epoch": 11.11541623843782,
+      "grad_norm": 0.04675271734595299,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 108153
+    },
+    {
+      "epoch": 11.115724563206578,
+      "grad_norm": 0.11510089039802551,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 108156
+    },
+    {
+      "epoch": 11.116032887975335,
+      "grad_norm": 0.06580482423305511,
+      "learning_rate": 0.01,
+      "loss": 1.9751,
+      "step": 108159
+    },
+    {
+      "epoch": 11.116341212744091,
+      "grad_norm": 0.10699275135993958,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 108162
+    },
+    {
+      "epoch": 11.116649537512847,
+      "grad_norm": 0.10668794810771942,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 108165
+    },
+    {
+      "epoch": 11.116957862281604,
+      "grad_norm": 0.05328281223773956,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 108168
+    },
+    {
+      "epoch": 11.11726618705036,
+      "grad_norm": 0.13215139508247375,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 108171
+    },
+    {
+      "epoch": 11.117574511819116,
+      "grad_norm": 0.12169384956359863,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 108174
+    },
+    {
+      "epoch": 11.117882836587873,
+      "grad_norm": 0.05846225097775459,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 108177
+    },
+    {
+      "epoch": 11.118191161356629,
+      "grad_norm": 0.0506817102432251,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 108180
+    },
+    {
+      "epoch": 11.118499486125385,
+      "grad_norm": 0.04678288847208023,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 108183
+    },
+    {
+      "epoch": 11.118807810894141,
+      "grad_norm": 0.051260873675346375,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 108186
+    },
+    {
+      "epoch": 11.119116135662898,
+      "grad_norm": 0.07663340866565704,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 108189
+    },
+    {
+      "epoch": 11.119424460431654,
+      "grad_norm": 0.09651526063680649,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 108192
+    },
+    {
+      "epoch": 11.11973278520041,
+      "grad_norm": 0.09631019085645676,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 108195
+    },
+    {
+      "epoch": 11.120041109969167,
+      "grad_norm": 0.07368805259466171,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 108198
+    },
+    {
+      "epoch": 11.120349434737925,
+      "grad_norm": 0.04535972326993942,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 108201
+    },
+    {
+      "epoch": 11.120657759506681,
+      "grad_norm": 0.04977213591337204,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 108204
+    },
+    {
+      "epoch": 11.120966084275437,
+      "grad_norm": 0.06690436601638794,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 108207
+    },
+    {
+      "epoch": 11.121274409044194,
+      "grad_norm": 0.03737137094140053,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 108210
+    },
+    {
+      "epoch": 11.12158273381295,
+      "grad_norm": 0.17570669949054718,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 108213
+    },
+    {
+      "epoch": 11.121891058581706,
+      "grad_norm": 0.06054900959134102,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 108216
+    },
+    {
+      "epoch": 11.122199383350463,
+      "grad_norm": 0.052555911242961884,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 108219
+    },
+    {
+      "epoch": 11.122507708119219,
+      "grad_norm": 0.04419514164328575,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 108222
+    },
+    {
+      "epoch": 11.122816032887975,
+      "grad_norm": 0.05516252666711807,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 108225
+    },
+    {
+      "epoch": 11.123124357656732,
+      "grad_norm": 0.0422653891146183,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 108228
+    },
+    {
+      "epoch": 11.123432682425488,
+      "grad_norm": 0.031638115644454956,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 108231
+    },
+    {
+      "epoch": 11.123741007194244,
+      "grad_norm": 0.03453582525253296,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 108234
+    },
+    {
+      "epoch": 11.124049331963,
+      "grad_norm": 0.13449911773204803,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 108237
+    },
+    {
+      "epoch": 11.124357656731757,
+      "grad_norm": 0.10251076519489288,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 108240
+    },
+    {
+      "epoch": 11.124665981500513,
+      "grad_norm": 0.07683219015598297,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 108243
+    },
+    {
+      "epoch": 11.12497430626927,
+      "grad_norm": 0.04399493709206581,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 108246
+    },
+    {
+      "epoch": 11.125282631038028,
+      "grad_norm": 0.04075119271874428,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 108249
+    },
+    {
+      "epoch": 11.125590955806784,
+      "grad_norm": 0.04534753039479256,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 108252
+    },
+    {
+      "epoch": 11.12589928057554,
+      "grad_norm": 0.04419349879026413,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 108255
+    },
+    {
+      "epoch": 11.126207605344296,
+      "grad_norm": 0.0396546870470047,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 108258
+    },
+    {
+      "epoch": 11.126515930113053,
+      "grad_norm": 0.07965173572301865,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 108261
+    },
+    {
+      "epoch": 11.126824254881809,
+      "grad_norm": 0.03837994113564491,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 108264
+    },
+    {
+      "epoch": 11.127132579650565,
+      "grad_norm": 0.05804935842752457,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 108267
+    },
+    {
+      "epoch": 11.127440904419322,
+      "grad_norm": 0.058049287647008896,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 108270
+    },
+    {
+      "epoch": 11.127749229188078,
+      "grad_norm": 0.055208366364240646,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 108273
+    },
+    {
+      "epoch": 11.128057553956834,
+      "grad_norm": 0.10259728878736496,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 108276
+    },
+    {
+      "epoch": 11.12836587872559,
+      "grad_norm": 0.04577815160155296,
+      "learning_rate": 0.01,
+      "loss": 1.98,
+      "step": 108279
+    },
+    {
+      "epoch": 11.128674203494347,
+      "grad_norm": 0.061545539647340775,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 108282
+    },
+    {
+      "epoch": 11.128982528263103,
+      "grad_norm": 0.03951617330312729,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 108285
+    },
+    {
+      "epoch": 11.12929085303186,
+      "grad_norm": 0.06631302833557129,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 108288
+    },
+    {
+      "epoch": 11.129599177800616,
+      "grad_norm": 0.047301799058914185,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 108291
+    },
+    {
+      "epoch": 11.129907502569374,
+      "grad_norm": 0.06022251769900322,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 108294
+    },
+    {
+      "epoch": 11.13021582733813,
+      "grad_norm": 0.07660970091819763,
+      "learning_rate": 0.01,
+      "loss": 1.9745,
+      "step": 108297
+    },
+    {
+      "epoch": 11.130524152106887,
+      "grad_norm": 0.06937503069639206,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 108300
+    },
+    {
+      "epoch": 11.130832476875643,
+      "grad_norm": 0.09794414788484573,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 108303
+    },
+    {
+      "epoch": 11.1311408016444,
+      "grad_norm": 0.05716533213853836,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 108306
+    },
+    {
+      "epoch": 11.131449126413155,
+      "grad_norm": 0.09398681670427322,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 108309
+    },
+    {
+      "epoch": 11.131757451181912,
+      "grad_norm": 0.052096445113420486,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 108312
+    },
+    {
+      "epoch": 11.132065775950668,
+      "grad_norm": 0.09356788545846939,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 108315
+    },
+    {
+      "epoch": 11.132374100719424,
+      "grad_norm": 0.04018421098589897,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 108318
+    },
+    {
+      "epoch": 11.13268242548818,
+      "grad_norm": 0.11394383758306503,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 108321
+    },
+    {
+      "epoch": 11.132990750256937,
+      "grad_norm": 0.09426377713680267,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 108324
+    },
+    {
+      "epoch": 11.133299075025693,
+      "grad_norm": 0.08016916364431381,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 108327
+    },
+    {
+      "epoch": 11.13360739979445,
+      "grad_norm": 0.042565468698740005,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 108330
+    },
+    {
+      "epoch": 11.133915724563206,
+      "grad_norm": 0.11248867213726044,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 108333
+    },
+    {
+      "epoch": 11.134224049331962,
+      "grad_norm": 0.036491721868515015,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 108336
+    },
+    {
+      "epoch": 11.13453237410072,
+      "grad_norm": 0.04385516792535782,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 108339
+    },
+    {
+      "epoch": 11.134840698869477,
+      "grad_norm": 0.05395470932126045,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 108342
+    },
+    {
+      "epoch": 11.135149023638233,
+      "grad_norm": 0.04313792288303375,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 108345
+    },
+    {
+      "epoch": 11.13545734840699,
+      "grad_norm": 0.03667489066720009,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 108348
+    },
+    {
+      "epoch": 11.135765673175746,
+      "grad_norm": 0.04531373456120491,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 108351
+    },
+    {
+      "epoch": 11.136073997944502,
+      "grad_norm": 0.056769829243421555,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 108354
+    },
+    {
+      "epoch": 11.136382322713258,
+      "grad_norm": 0.05138956755399704,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 108357
+    },
+    {
+      "epoch": 11.136690647482014,
+      "grad_norm": 0.048939868807792664,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 108360
+    },
+    {
+      "epoch": 11.13699897225077,
+      "grad_norm": 0.03497758135199547,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 108363
+    },
+    {
+      "epoch": 11.137307297019527,
+      "grad_norm": 0.11892599612474442,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 108366
+    },
+    {
+      "epoch": 11.137615621788283,
+      "grad_norm": 0.04540543258190155,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 108369
+    },
+    {
+      "epoch": 11.13792394655704,
+      "grad_norm": 0.03498394042253494,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 108372
+    },
+    {
+      "epoch": 11.138232271325796,
+      "grad_norm": 0.049198199063539505,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 108375
+    },
+    {
+      "epoch": 11.138540596094552,
+      "grad_norm": 0.1480265110731125,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 108378
+    },
+    {
+      "epoch": 11.138848920863309,
+      "grad_norm": 0.07885235548019409,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 108381
+    },
+    {
+      "epoch": 11.139157245632067,
+      "grad_norm": 0.04114838317036629,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 108384
+    },
+    {
+      "epoch": 11.139465570400823,
+      "grad_norm": 0.03757159411907196,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 108387
+    },
+    {
+      "epoch": 11.13977389516958,
+      "grad_norm": 0.05755399167537689,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 108390
+    },
+    {
+      "epoch": 11.140082219938336,
+      "grad_norm": 0.05723578855395317,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 108393
+    },
+    {
+      "epoch": 11.140390544707092,
+      "grad_norm": 0.13173669576644897,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 108396
+    },
+    {
+      "epoch": 11.140698869475848,
+      "grad_norm": 0.049970634281635284,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 108399
+    },
+    {
+      "epoch": 11.141007194244605,
+      "grad_norm": 0.1683085411787033,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 108402
+    },
+    {
+      "epoch": 11.14131551901336,
+      "grad_norm": 0.10416439175605774,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 108405
+    },
+    {
+      "epoch": 11.141623843782117,
+      "grad_norm": 0.05204487591981888,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 108408
+    },
+    {
+      "epoch": 11.141932168550873,
+      "grad_norm": 0.05820777267217636,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 108411
+    },
+    {
+      "epoch": 11.14224049331963,
+      "grad_norm": 0.0588134303689003,
+      "learning_rate": 0.01,
+      "loss": 1.9735,
+      "step": 108414
+    },
+    {
+      "epoch": 11.142548818088386,
+      "grad_norm": 0.041087377816438675,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 108417
+    },
+    {
+      "epoch": 11.142857142857142,
+      "grad_norm": 0.058787308633327484,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 108420
+    },
+    {
+      "epoch": 11.143165467625899,
+      "grad_norm": 0.06542397290468216,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 108423
+    },
+    {
+      "epoch": 11.143473792394655,
+      "grad_norm": 0.07437827438116074,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 108426
+    },
+    {
+      "epoch": 11.143782117163411,
+      "grad_norm": 0.04234655201435089,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 108429
+    },
+    {
+      "epoch": 11.14409044193217,
+      "grad_norm": 0.08659413456916809,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 108432
+    },
+    {
+      "epoch": 11.144398766700926,
+      "grad_norm": 0.0776323676109314,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 108435
+    },
+    {
+      "epoch": 11.144707091469682,
+      "grad_norm": 0.16545532643795013,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 108438
+    },
+    {
+      "epoch": 11.145015416238438,
+      "grad_norm": 0.1424427479505539,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 108441
+    },
+    {
+      "epoch": 11.145323741007195,
+      "grad_norm": 0.07523380219936371,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 108444
+    },
+    {
+      "epoch": 11.145632065775951,
+      "grad_norm": 0.047919511795043945,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 108447
+    },
+    {
+      "epoch": 11.145940390544707,
+      "grad_norm": 0.04040754586458206,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 108450
+    },
+    {
+      "epoch": 11.146248715313464,
+      "grad_norm": 0.06334877759218216,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 108453
+    },
+    {
+      "epoch": 11.14655704008222,
+      "grad_norm": 0.040293674916028976,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 108456
+    },
+    {
+      "epoch": 11.146865364850976,
+      "grad_norm": 0.11464455723762512,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 108459
+    },
+    {
+      "epoch": 11.147173689619732,
+      "grad_norm": 0.04293007403612137,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 108462
+    },
+    {
+      "epoch": 11.147482014388489,
+      "grad_norm": 0.08018584549427032,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 108465
+    },
+    {
+      "epoch": 11.147790339157245,
+      "grad_norm": 0.06225692853331566,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 108468
+    },
+    {
+      "epoch": 11.148098663926001,
+      "grad_norm": 0.054755352437496185,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 108471
+    },
+    {
+      "epoch": 11.148406988694758,
+      "grad_norm": 0.04388297721743584,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 108474
+    },
+    {
+      "epoch": 11.148715313463516,
+      "grad_norm": 0.1110197901725769,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 108477
+    },
+    {
+      "epoch": 11.149023638232272,
+      "grad_norm": 0.06123514100909233,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 108480
+    },
+    {
+      "epoch": 11.149331963001028,
+      "grad_norm": 0.08198761194944382,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 108483
+    },
+    {
+      "epoch": 11.149640287769785,
+      "grad_norm": 0.06199076399207115,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 108486
+    },
+    {
+      "epoch": 11.149948612538541,
+      "grad_norm": 0.06349356472492218,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 108489
+    },
+    {
+      "epoch": 11.150256937307297,
+      "grad_norm": 0.09270074963569641,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 108492
+    },
+    {
+      "epoch": 11.150565262076054,
+      "grad_norm": 0.046484947204589844,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 108495
+    },
+    {
+      "epoch": 11.15087358684481,
+      "grad_norm": 0.03359349071979523,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 108498
+    },
+    {
+      "epoch": 11.151181911613566,
+      "grad_norm": 0.03490680083632469,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 108501
+    },
+    {
+      "epoch": 11.151490236382323,
+      "grad_norm": 0.03208662569522858,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 108504
+    },
+    {
+      "epoch": 11.151798561151079,
+      "grad_norm": 0.0415281280875206,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 108507
+    },
+    {
+      "epoch": 11.152106885919835,
+      "grad_norm": 0.24639929831027985,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 108510
+    },
+    {
+      "epoch": 11.152415210688591,
+      "grad_norm": 0.13605011999607086,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 108513
+    },
+    {
+      "epoch": 11.152723535457348,
+      "grad_norm": 0.15605586767196655,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 108516
+    },
+    {
+      "epoch": 11.153031860226104,
+      "grad_norm": 0.07057122886180878,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 108519
+    },
+    {
+      "epoch": 11.153340184994862,
+      "grad_norm": 0.04619940370321274,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 108522
+    },
+    {
+      "epoch": 11.153648509763618,
+      "grad_norm": 0.10820236057043076,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 108525
+    },
+    {
+      "epoch": 11.153956834532375,
+      "grad_norm": 0.06613638252019882,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 108528
+    },
+    {
+      "epoch": 11.154265159301131,
+      "grad_norm": 0.042178694158792496,
+      "learning_rate": 0.01,
+      "loss": 1.8919,
+      "step": 108531
+    },
+    {
+      "epoch": 11.154573484069887,
+      "grad_norm": 0.042413171380758286,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 108534
+    },
+    {
+      "epoch": 11.154881808838644,
+      "grad_norm": 0.0663883239030838,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 108537
+    },
+    {
+      "epoch": 11.1551901336074,
+      "grad_norm": 0.04206288233399391,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 108540
+    },
+    {
+      "epoch": 11.155498458376156,
+      "grad_norm": 0.05079186335206032,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 108543
+    },
+    {
+      "epoch": 11.155806783144913,
+      "grad_norm": 0.047072358429431915,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 108546
+    },
+    {
+      "epoch": 11.156115107913669,
+      "grad_norm": 0.06086193770170212,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 108549
+    },
+    {
+      "epoch": 11.156423432682425,
+      "grad_norm": 0.08996503055095673,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 108552
+    },
+    {
+      "epoch": 11.156731757451182,
+      "grad_norm": 0.13274416327476501,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 108555
+    },
+    {
+      "epoch": 11.157040082219938,
+      "grad_norm": 0.21141725778579712,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 108558
+    },
+    {
+      "epoch": 11.157348406988694,
+      "grad_norm": 0.131058931350708,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 108561
+    },
+    {
+      "epoch": 11.15765673175745,
+      "grad_norm": 0.05547711253166199,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 108564
+    },
+    {
+      "epoch": 11.157965056526207,
+      "grad_norm": 0.03413417190313339,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 108567
+    },
+    {
+      "epoch": 11.158273381294965,
+      "grad_norm": 0.03346508368849754,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 108570
+    },
+    {
+      "epoch": 11.158581706063721,
+      "grad_norm": 0.03702901676297188,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 108573
+    },
+    {
+      "epoch": 11.158890030832477,
+      "grad_norm": 0.07465177029371262,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 108576
+    },
+    {
+      "epoch": 11.159198355601234,
+      "grad_norm": 0.04305208474397659,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 108579
+    },
+    {
+      "epoch": 11.15950668036999,
+      "grad_norm": 0.05763426795601845,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 108582
+    },
+    {
+      "epoch": 11.159815005138746,
+      "grad_norm": 0.04999389126896858,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 108585
+    },
+    {
+      "epoch": 11.160123329907503,
+      "grad_norm": 0.03652753308415413,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 108588
+    },
+    {
+      "epoch": 11.160431654676259,
+      "grad_norm": 0.17381641268730164,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 108591
+    },
+    {
+      "epoch": 11.160739979445015,
+      "grad_norm": 0.04396112635731697,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 108594
+    },
+    {
+      "epoch": 11.161048304213772,
+      "grad_norm": 0.03453262522816658,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 108597
+    },
+    {
+      "epoch": 11.161356628982528,
+      "grad_norm": 0.10160734504461288,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 108600
+    },
+    {
+      "epoch": 11.161664953751284,
+      "grad_norm": 0.10512173175811768,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 108603
+    },
+    {
+      "epoch": 11.16197327852004,
+      "grad_norm": 0.0531027652323246,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 108606
+    },
+    {
+      "epoch": 11.162281603288797,
+      "grad_norm": 0.1427735984325409,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 108609
+    },
+    {
+      "epoch": 11.162589928057553,
+      "grad_norm": 0.04474338889122009,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 108612
+    },
+    {
+      "epoch": 11.162898252826311,
+      "grad_norm": 0.1090499684214592,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 108615
+    },
+    {
+      "epoch": 11.163206577595068,
+      "grad_norm": 0.0636020377278328,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 108618
+    },
+    {
+      "epoch": 11.163514902363824,
+      "grad_norm": 0.044732339680194855,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 108621
+    },
+    {
+      "epoch": 11.16382322713258,
+      "grad_norm": 0.05239364132285118,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 108624
+    },
+    {
+      "epoch": 11.164131551901336,
+      "grad_norm": 0.1660080850124359,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 108627
+    },
+    {
+      "epoch": 11.164439876670093,
+      "grad_norm": 0.055751584470272064,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 108630
+    },
+    {
+      "epoch": 11.164748201438849,
+      "grad_norm": 0.042030543088912964,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 108633
+    },
+    {
+      "epoch": 11.165056526207605,
+      "grad_norm": 0.050695545971393585,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 108636
+    },
+    {
+      "epoch": 11.165364850976362,
+      "grad_norm": 0.08683747053146362,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 108639
+    },
+    {
+      "epoch": 11.165673175745118,
+      "grad_norm": 0.05954531580209732,
+      "learning_rate": 0.01,
+      "loss": 1.9831,
+      "step": 108642
+    },
+    {
+      "epoch": 11.165981500513874,
+      "grad_norm": 0.1576981097459793,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 108645
+    },
+    {
+      "epoch": 11.16628982528263,
+      "grad_norm": 0.09284688532352448,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 108648
+    },
+    {
+      "epoch": 11.166598150051387,
+      "grad_norm": 0.07134409248828888,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 108651
+    },
+    {
+      "epoch": 11.166906474820143,
+      "grad_norm": 0.05715997517108917,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 108654
+    },
+    {
+      "epoch": 11.1672147995889,
+      "grad_norm": 0.043932173401117325,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 108657
+    },
+    {
+      "epoch": 11.167523124357658,
+      "grad_norm": 0.02749806083738804,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 108660
+    },
+    {
+      "epoch": 11.167831449126414,
+      "grad_norm": 0.055440399795770645,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 108663
+    },
+    {
+      "epoch": 11.16813977389517,
+      "grad_norm": 0.07681028544902802,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 108666
+    },
+    {
+      "epoch": 11.168448098663927,
+      "grad_norm": 0.14702077209949493,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 108669
+    },
+    {
+      "epoch": 11.168756423432683,
+      "grad_norm": 0.06999138742685318,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 108672
+    },
+    {
+      "epoch": 11.16906474820144,
+      "grad_norm": 0.03188035637140274,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 108675
+    },
+    {
+      "epoch": 11.169373072970195,
+      "grad_norm": 0.03169883415102959,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 108678
+    },
+    {
+      "epoch": 11.169681397738952,
+      "grad_norm": 0.14469371736049652,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 108681
+    },
+    {
+      "epoch": 11.169989722507708,
+      "grad_norm": 0.1052892655134201,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 108684
+    },
+    {
+      "epoch": 11.170298047276464,
+      "grad_norm": 0.06744600832462311,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 108687
+    },
+    {
+      "epoch": 11.17060637204522,
+      "grad_norm": 0.05542672798037529,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 108690
+    },
+    {
+      "epoch": 11.170914696813977,
+      "grad_norm": 0.07598409056663513,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 108693
+    },
+    {
+      "epoch": 11.171223021582733,
+      "grad_norm": 0.06378299742937088,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 108696
+    },
+    {
+      "epoch": 11.17153134635149,
+      "grad_norm": 0.04912390932440758,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 108699
+    },
+    {
+      "epoch": 11.171839671120246,
+      "grad_norm": 0.07536765933036804,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 108702
+    },
+    {
+      "epoch": 11.172147995889002,
+      "grad_norm": 0.05856078118085861,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 108705
+    },
+    {
+      "epoch": 11.17245632065776,
+      "grad_norm": 0.09633874893188477,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 108708
+    },
+    {
+      "epoch": 11.172764645426517,
+      "grad_norm": 0.09800992161035538,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 108711
+    },
+    {
+      "epoch": 11.173072970195273,
+      "grad_norm": 0.08409328013658524,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 108714
+    },
+    {
+      "epoch": 11.17338129496403,
+      "grad_norm": 0.09450411051511765,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 108717
+    },
+    {
+      "epoch": 11.173689619732786,
+      "grad_norm": 0.11592711508274078,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 108720
+    },
+    {
+      "epoch": 11.173997944501542,
+      "grad_norm": 0.09436693042516708,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 108723
+    },
+    {
+      "epoch": 11.174306269270298,
+      "grad_norm": 0.06513006240129471,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 108726
+    },
+    {
+      "epoch": 11.174614594039054,
+      "grad_norm": 0.03724005073308945,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 108729
+    },
+    {
+      "epoch": 11.17492291880781,
+      "grad_norm": 0.03580351173877716,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 108732
+    },
+    {
+      "epoch": 11.175231243576567,
+      "grad_norm": 0.037552379071712494,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 108735
+    },
+    {
+      "epoch": 11.175539568345323,
+      "grad_norm": 0.06251754611730576,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 108738
+    },
+    {
+      "epoch": 11.17584789311408,
+      "grad_norm": 0.059087805449962616,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 108741
+    },
+    {
+      "epoch": 11.176156217882836,
+      "grad_norm": 0.0471821054816246,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 108744
+    },
+    {
+      "epoch": 11.176464542651592,
+      "grad_norm": 0.03444281965494156,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 108747
+    },
+    {
+      "epoch": 11.176772867420349,
+      "grad_norm": 0.16312456130981445,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 108750
+    },
+    {
+      "epoch": 11.177081192189107,
+      "grad_norm": 0.05206234008073807,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 108753
+    },
+    {
+      "epoch": 11.177389516957863,
+      "grad_norm": 0.062144502997398376,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 108756
+    },
+    {
+      "epoch": 11.17769784172662,
+      "grad_norm": 0.04214511439204216,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 108759
+    },
+    {
+      "epoch": 11.178006166495376,
+      "grad_norm": 0.05748814716935158,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 108762
+    },
+    {
+      "epoch": 11.178314491264132,
+      "grad_norm": 0.05643995851278305,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 108765
+    },
+    {
+      "epoch": 11.178622816032888,
+      "grad_norm": 0.07424315065145493,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 108768
+    },
+    {
+      "epoch": 11.178931140801645,
+      "grad_norm": 0.08109375834465027,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 108771
+    },
+    {
+      "epoch": 11.1792394655704,
+      "grad_norm": 0.05265939235687256,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 108774
+    },
+    {
+      "epoch": 11.179547790339157,
+      "grad_norm": 0.03445740416646004,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 108777
+    },
+    {
+      "epoch": 11.179856115107913,
+      "grad_norm": 0.11327429115772247,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 108780
+    },
+    {
+      "epoch": 11.18016443987667,
+      "grad_norm": 0.09449184685945511,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 108783
+    },
+    {
+      "epoch": 11.180472764645426,
+      "grad_norm": 0.0852329358458519,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 108786
+    },
+    {
+      "epoch": 11.180781089414182,
+      "grad_norm": 0.04656582325696945,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 108789
+    },
+    {
+      "epoch": 11.181089414182939,
+      "grad_norm": 0.075565405189991,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 108792
+    },
+    {
+      "epoch": 11.181397738951695,
+      "grad_norm": 0.08553548902273178,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 108795
+    },
+    {
+      "epoch": 11.181706063720453,
+      "grad_norm": 0.03676193952560425,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 108798
+    },
+    {
+      "epoch": 11.18201438848921,
+      "grad_norm": 0.08095156401395798,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 108801
+    },
+    {
+      "epoch": 11.182322713257966,
+      "grad_norm": 0.0733218714594841,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 108804
+    },
+    {
+      "epoch": 11.182631038026722,
+      "grad_norm": 0.038306981325149536,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 108807
+    },
+    {
+      "epoch": 11.182939362795478,
+      "grad_norm": 0.06717819720506668,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 108810
+    },
+    {
+      "epoch": 11.183247687564235,
+      "grad_norm": 0.058772485703229904,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 108813
+    },
+    {
+      "epoch": 11.183556012332991,
+      "grad_norm": 0.05892009660601616,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 108816
+    },
+    {
+      "epoch": 11.183864337101747,
+      "grad_norm": 0.07855512946844101,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 108819
+    },
+    {
+      "epoch": 11.184172661870504,
+      "grad_norm": 0.07210905104875565,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 108822
+    },
+    {
+      "epoch": 11.18448098663926,
+      "grad_norm": 0.15390074253082275,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 108825
+    },
+    {
+      "epoch": 11.184789311408016,
+      "grad_norm": 0.16397853195667267,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 108828
+    },
+    {
+      "epoch": 11.185097636176772,
+      "grad_norm": 0.1202760562300682,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 108831
+    },
+    {
+      "epoch": 11.185405960945529,
+      "grad_norm": 0.056249383836984634,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 108834
+    },
+    {
+      "epoch": 11.185714285714285,
+      "grad_norm": 0.03812577202916145,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 108837
+    },
+    {
+      "epoch": 11.186022610483041,
+      "grad_norm": 0.21584199368953705,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 108840
+    },
+    {
+      "epoch": 11.1863309352518,
+      "grad_norm": 0.048358093947172165,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 108843
+    },
+    {
+      "epoch": 11.186639260020556,
+      "grad_norm": 0.05890496075153351,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 108846
+    },
+    {
+      "epoch": 11.186947584789312,
+      "grad_norm": 0.07943232357501984,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 108849
+    },
+    {
+      "epoch": 11.187255909558068,
+      "grad_norm": 0.08584515005350113,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 108852
+    },
+    {
+      "epoch": 11.187564234326825,
+      "grad_norm": 0.03383355960249901,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 108855
+    },
+    {
+      "epoch": 11.187872559095581,
+      "grad_norm": 0.09473607689142227,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 108858
+    },
+    {
+      "epoch": 11.188180883864337,
+      "grad_norm": 0.03801342844963074,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 108861
+    },
+    {
+      "epoch": 11.188489208633094,
+      "grad_norm": 0.041940245777368546,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 108864
+    },
+    {
+      "epoch": 11.18879753340185,
+      "grad_norm": 0.05744069814682007,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 108867
+    },
+    {
+      "epoch": 11.189105858170606,
+      "grad_norm": 0.08689646422863007,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 108870
+    },
+    {
+      "epoch": 11.189414182939363,
+      "grad_norm": 0.06015917658805847,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 108873
+    },
+    {
+      "epoch": 11.189722507708119,
+      "grad_norm": 0.04627293720841408,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 108876
+    },
+    {
+      "epoch": 11.190030832476875,
+      "grad_norm": 0.10750222951173782,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 108879
+    },
+    {
+      "epoch": 11.190339157245631,
+      "grad_norm": 0.07908863574266434,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 108882
+    },
+    {
+      "epoch": 11.190647482014388,
+      "grad_norm": 0.09672728925943375,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 108885
+    },
+    {
+      "epoch": 11.190955806783144,
+      "grad_norm": 0.10572005808353424,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 108888
+    },
+    {
+      "epoch": 11.191264131551902,
+      "grad_norm": 0.04903019219636917,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 108891
+    },
+    {
+      "epoch": 11.191572456320658,
+      "grad_norm": 0.0401516817510128,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 108894
+    },
+    {
+      "epoch": 11.191880781089415,
+      "grad_norm": 0.050913285464048386,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 108897
+    },
+    {
+      "epoch": 11.192189105858171,
+      "grad_norm": 0.04478805884718895,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 108900
+    },
+    {
+      "epoch": 11.192497430626927,
+      "grad_norm": 0.03856386989355087,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 108903
+    },
+    {
+      "epoch": 11.192805755395684,
+      "grad_norm": 0.05249187722802162,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 108906
+    },
+    {
+      "epoch": 11.19311408016444,
+      "grad_norm": 0.0587519034743309,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 108909
+    },
+    {
+      "epoch": 11.193422404933196,
+      "grad_norm": 0.03957201540470123,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 108912
+    },
+    {
+      "epoch": 11.193730729701953,
+      "grad_norm": 0.04053035378456116,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 108915
+    },
+    {
+      "epoch": 11.194039054470709,
+      "grad_norm": 0.088300421833992,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 108918
+    },
+    {
+      "epoch": 11.194347379239465,
+      "grad_norm": 0.0719873234629631,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 108921
+    },
+    {
+      "epoch": 11.194655704008222,
+      "grad_norm": 0.11401339620351791,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 108924
+    },
+    {
+      "epoch": 11.194964028776978,
+      "grad_norm": 0.12527501583099365,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 108927
+    },
+    {
+      "epoch": 11.195272353545734,
+      "grad_norm": 0.05526786297559738,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 108930
+    },
+    {
+      "epoch": 11.19558067831449,
+      "grad_norm": 0.07284065335988998,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 108933
+    },
+    {
+      "epoch": 11.195889003083249,
+      "grad_norm": 0.07272975146770477,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 108936
+    },
+    {
+      "epoch": 11.196197327852005,
+      "grad_norm": 0.03863842040300369,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 108939
+    },
+    {
+      "epoch": 11.196505652620761,
+      "grad_norm": 0.1017715260386467,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 108942
+    },
+    {
+      "epoch": 11.196813977389517,
+      "grad_norm": 0.05640571564435959,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 108945
+    },
+    {
+      "epoch": 11.197122302158274,
+      "grad_norm": 0.04469192400574684,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 108948
+    },
+    {
+      "epoch": 11.19743062692703,
+      "grad_norm": 0.09084463119506836,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 108951
+    },
+    {
+      "epoch": 11.197738951695786,
+      "grad_norm": 0.032969359308481216,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 108954
+    },
+    {
+      "epoch": 11.198047276464543,
+      "grad_norm": 0.04472844675183296,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 108957
+    },
+    {
+      "epoch": 11.198355601233299,
+      "grad_norm": 0.08740687370300293,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 108960
+    },
+    {
+      "epoch": 11.198663926002055,
+      "grad_norm": 0.06060430780053139,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 108963
+    },
+    {
+      "epoch": 11.198972250770812,
+      "grad_norm": 0.05395602807402611,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 108966
+    },
+    {
+      "epoch": 11.199280575539568,
+      "grad_norm": 0.10777734220027924,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 108969
+    },
+    {
+      "epoch": 11.199588900308324,
+      "grad_norm": 0.04315931722521782,
+      "learning_rate": 0.01,
+      "loss": 1.9791,
+      "step": 108972
+    },
+    {
+      "epoch": 11.19989722507708,
+      "grad_norm": 0.03909776732325554,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 108975
+    },
+    {
+      "epoch": 11.200205549845837,
+      "grad_norm": 0.03314677253365517,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 108978
+    },
+    {
+      "epoch": 11.200513874614595,
+      "grad_norm": 0.07113143801689148,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 108981
+    },
+    {
+      "epoch": 11.200822199383351,
+      "grad_norm": 0.09888097643852234,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 108984
+    },
+    {
+      "epoch": 11.201130524152108,
+      "grad_norm": 0.08767660707235336,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 108987
+    },
+    {
+      "epoch": 11.201438848920864,
+      "grad_norm": 0.046668294817209244,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 108990
+    },
+    {
+      "epoch": 11.20174717368962,
+      "grad_norm": 0.1076827198266983,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 108993
+    },
+    {
+      "epoch": 11.202055498458376,
+      "grad_norm": 0.051028959453105927,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 108996
+    },
+    {
+      "epoch": 11.202363823227133,
+      "grad_norm": 0.050806645303964615,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 108999
+    },
+    {
+      "epoch": 11.202672147995889,
+      "grad_norm": 0.06453768163919449,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 109002
+    },
+    {
+      "epoch": 11.202980472764645,
+      "grad_norm": 0.0641925036907196,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 109005
+    },
+    {
+      "epoch": 11.203288797533402,
+      "grad_norm": 0.042618606239557266,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 109008
+    },
+    {
+      "epoch": 11.203597122302158,
+      "grad_norm": 0.054604362696409225,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 109011
+    },
+    {
+      "epoch": 11.203905447070914,
+      "grad_norm": 0.06285086274147034,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 109014
+    },
+    {
+      "epoch": 11.20421377183967,
+      "grad_norm": 0.07482526451349258,
+      "learning_rate": 0.01,
+      "loss": 1.9796,
+      "step": 109017
+    },
+    {
+      "epoch": 11.204522096608427,
+      "grad_norm": 0.09341461956501007,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 109020
+    },
+    {
+      "epoch": 11.204830421377183,
+      "grad_norm": 0.03731429576873779,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 109023
+    },
+    {
+      "epoch": 11.20513874614594,
+      "grad_norm": 0.14362257719039917,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 109026
+    },
+    {
+      "epoch": 11.205447070914698,
+      "grad_norm": 0.12169114500284195,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 109029
+    },
+    {
+      "epoch": 11.205755395683454,
+      "grad_norm": 0.1270626336336136,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 109032
+    },
+    {
+      "epoch": 11.20606372045221,
+      "grad_norm": 0.07900585979223251,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 109035
+    },
+    {
+      "epoch": 11.206372045220967,
+      "grad_norm": 0.04565434902906418,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 109038
+    },
+    {
+      "epoch": 11.206680369989723,
+      "grad_norm": 0.05286426469683647,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 109041
+    },
+    {
+      "epoch": 11.20698869475848,
+      "grad_norm": 0.044969215989112854,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 109044
+    },
+    {
+      "epoch": 11.207297019527235,
+      "grad_norm": 0.05089700222015381,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 109047
+    },
+    {
+      "epoch": 11.207605344295992,
+      "grad_norm": 0.04107325151562691,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 109050
+    },
+    {
+      "epoch": 11.207913669064748,
+      "grad_norm": 0.04545219987630844,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 109053
+    },
+    {
+      "epoch": 11.208221993833504,
+      "grad_norm": 0.052229829132556915,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 109056
+    },
+    {
+      "epoch": 11.20853031860226,
+      "grad_norm": 0.09277478605508804,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 109059
+    },
+    {
+      "epoch": 11.208838643371017,
+      "grad_norm": 0.06279975920915604,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 109062
+    },
+    {
+      "epoch": 11.209146968139773,
+      "grad_norm": 0.03972071781754494,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 109065
+    },
+    {
+      "epoch": 11.20945529290853,
+      "grad_norm": 0.053854186087846756,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 109068
+    },
+    {
+      "epoch": 11.209763617677286,
+      "grad_norm": 0.046731676906347275,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 109071
+    },
+    {
+      "epoch": 11.210071942446044,
+      "grad_norm": 0.06777119636535645,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 109074
+    },
+    {
+      "epoch": 11.2103802672148,
+      "grad_norm": 0.06705444306135178,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 109077
+    },
+    {
+      "epoch": 11.210688591983557,
+      "grad_norm": 0.08733968436717987,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 109080
+    },
+    {
+      "epoch": 11.210996916752313,
+      "grad_norm": 0.11957607418298721,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 109083
+    },
+    {
+      "epoch": 11.21130524152107,
+      "grad_norm": 0.08926189690828323,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 109086
+    },
+    {
+      "epoch": 11.211613566289826,
+      "grad_norm": 0.08381009101867676,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 109089
+    },
+    {
+      "epoch": 11.211921891058582,
+      "grad_norm": 0.05530963093042374,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 109092
+    },
+    {
+      "epoch": 11.212230215827338,
+      "grad_norm": 0.045270007103681564,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 109095
+    },
+    {
+      "epoch": 11.212538540596094,
+      "grad_norm": 0.045589692890644073,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 109098
+    },
+    {
+      "epoch": 11.21284686536485,
+      "grad_norm": 0.08058987557888031,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 109101
+    },
+    {
+      "epoch": 11.213155190133607,
+      "grad_norm": 0.06342653930187225,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 109104
+    },
+    {
+      "epoch": 11.213463514902363,
+      "grad_norm": 0.0859803557395935,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 109107
+    },
+    {
+      "epoch": 11.21377183967112,
+      "grad_norm": 0.11543281376361847,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 109110
+    },
+    {
+      "epoch": 11.214080164439876,
+      "grad_norm": 0.10394307971000671,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 109113
+    },
+    {
+      "epoch": 11.214388489208632,
+      "grad_norm": 0.04083403944969177,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 109116
+    },
+    {
+      "epoch": 11.21469681397739,
+      "grad_norm": 0.11865125596523285,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 109119
+    },
+    {
+      "epoch": 11.215005138746147,
+      "grad_norm": 0.0339047834277153,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 109122
+    },
+    {
+      "epoch": 11.215313463514903,
+      "grad_norm": 0.09069018065929413,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 109125
+    },
+    {
+      "epoch": 11.21562178828366,
+      "grad_norm": 0.047034863382577896,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 109128
+    },
+    {
+      "epoch": 11.215930113052416,
+      "grad_norm": 0.09295161068439484,
+      "learning_rate": 0.01,
+      "loss": 1.9688,
+      "step": 109131
+    },
+    {
+      "epoch": 11.216238437821172,
+      "grad_norm": 0.057629477232694626,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 109134
+    },
+    {
+      "epoch": 11.216546762589928,
+      "grad_norm": 0.09780970215797424,
+      "learning_rate": 0.01,
+      "loss": 1.8954,
+      "step": 109137
+    },
+    {
+      "epoch": 11.216855087358685,
+      "grad_norm": 0.04168618842959404,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 109140
+    },
+    {
+      "epoch": 11.21716341212744,
+      "grad_norm": 0.10839787125587463,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 109143
+    },
+    {
+      "epoch": 11.217471736896197,
+      "grad_norm": 0.15280303359031677,
+      "learning_rate": 0.01,
+      "loss": 1.981,
+      "step": 109146
+    },
+    {
+      "epoch": 11.217780061664953,
+      "grad_norm": 0.12073913216590881,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 109149
+    },
+    {
+      "epoch": 11.21808838643371,
+      "grad_norm": 0.08217319846153259,
+      "learning_rate": 0.01,
+      "loss": 1.9627,
+      "step": 109152
+    },
+    {
+      "epoch": 11.218396711202466,
+      "grad_norm": 0.06883706897497177,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 109155
+    },
+    {
+      "epoch": 11.218705035971222,
+      "grad_norm": 0.0811753198504448,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 109158
+    },
+    {
+      "epoch": 11.219013360739979,
+      "grad_norm": 0.04602133855223656,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 109161
+    },
+    {
+      "epoch": 11.219321685508735,
+      "grad_norm": 0.03887748718261719,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 109164
+    },
+    {
+      "epoch": 11.219630010277493,
+      "grad_norm": 0.062418244779109955,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 109167
+    },
+    {
+      "epoch": 11.21993833504625,
+      "grad_norm": 0.06071825325489044,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 109170
+    },
+    {
+      "epoch": 11.220246659815006,
+      "grad_norm": 0.04249916225671768,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 109173
+    },
+    {
+      "epoch": 11.220554984583762,
+      "grad_norm": 0.15064731240272522,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 109176
+    },
+    {
+      "epoch": 11.220863309352518,
+      "grad_norm": 0.04599601775407791,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 109179
+    },
+    {
+      "epoch": 11.221171634121275,
+      "grad_norm": 0.04028359055519104,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 109182
+    },
+    {
+      "epoch": 11.221479958890031,
+      "grad_norm": 0.034823331981897354,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 109185
+    },
+    {
+      "epoch": 11.221788283658787,
+      "grad_norm": 0.055492889136075974,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 109188
+    },
+    {
+      "epoch": 11.222096608427544,
+      "grad_norm": 0.057960398495197296,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 109191
+    },
+    {
+      "epoch": 11.2224049331963,
+      "grad_norm": 0.0979379266500473,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 109194
+    },
+    {
+      "epoch": 11.222713257965056,
+      "grad_norm": 0.04197908565402031,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 109197
+    },
+    {
+      "epoch": 11.223021582733812,
+      "grad_norm": 0.09842576831579208,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 109200
+    },
+    {
+      "epoch": 11.223329907502569,
+      "grad_norm": 0.04491826891899109,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 109203
+    },
+    {
+      "epoch": 11.223638232271325,
+      "grad_norm": 0.07851038873195648,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 109206
+    },
+    {
+      "epoch": 11.223946557040081,
+      "grad_norm": 0.05620894581079483,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 109209
+    },
+    {
+      "epoch": 11.22425488180884,
+      "grad_norm": 0.10865885764360428,
+      "learning_rate": 0.01,
+      "loss": 1.9813,
+      "step": 109212
+    },
+    {
+      "epoch": 11.224563206577596,
+      "grad_norm": 0.05403037741780281,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 109215
+    },
+    {
+      "epoch": 11.224871531346352,
+      "grad_norm": 0.06059322878718376,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 109218
+    },
+    {
+      "epoch": 11.225179856115108,
+      "grad_norm": 0.08516819775104523,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 109221
+    },
+    {
+      "epoch": 11.225488180883865,
+      "grad_norm": 0.07061944156885147,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 109224
+    },
+    {
+      "epoch": 11.225796505652621,
+      "grad_norm": 0.039020419120788574,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 109227
+    },
+    {
+      "epoch": 11.226104830421377,
+      "grad_norm": 0.06967667490243912,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 109230
+    },
+    {
+      "epoch": 11.226413155190134,
+      "grad_norm": 0.06568337231874466,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 109233
+    },
+    {
+      "epoch": 11.22672147995889,
+      "grad_norm": 0.09274667501449585,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 109236
+    },
+    {
+      "epoch": 11.227029804727646,
+      "grad_norm": 0.06000722572207451,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 109239
+    },
+    {
+      "epoch": 11.227338129496403,
+      "grad_norm": 0.04201046749949455,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 109242
+    },
+    {
+      "epoch": 11.227646454265159,
+      "grad_norm": 0.03122475929558277,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 109245
+    },
+    {
+      "epoch": 11.227954779033915,
+      "grad_norm": 0.05247949808835983,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 109248
+    },
+    {
+      "epoch": 11.228263103802671,
+      "grad_norm": 0.1298094391822815,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 109251
+    },
+    {
+      "epoch": 11.228571428571428,
+      "grad_norm": 0.05516472086310387,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 109254
+    },
+    {
+      "epoch": 11.228879753340186,
+      "grad_norm": 0.05649101734161377,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 109257
+    },
+    {
+      "epoch": 11.229188078108942,
+      "grad_norm": 0.0960668995976448,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 109260
+    },
+    {
+      "epoch": 11.229496402877698,
+      "grad_norm": 0.09434793889522552,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 109263
+    },
+    {
+      "epoch": 11.229804727646455,
+      "grad_norm": 0.07487189769744873,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 109266
+    },
+    {
+      "epoch": 11.230113052415211,
+      "grad_norm": 0.06864513456821442,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 109269
+    },
+    {
+      "epoch": 11.230421377183967,
+      "grad_norm": 0.12508641183376312,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 109272
+    },
+    {
+      "epoch": 11.230729701952724,
+      "grad_norm": 0.04709652438759804,
+      "learning_rate": 0.01,
+      "loss": 1.9654,
+      "step": 109275
+    },
+    {
+      "epoch": 11.23103802672148,
+      "grad_norm": 0.1833847165107727,
+      "learning_rate": 0.01,
+      "loss": 1.9851,
+      "step": 109278
+    },
+    {
+      "epoch": 11.231346351490236,
+      "grad_norm": 0.06030800938606262,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 109281
+    },
+    {
+      "epoch": 11.231654676258993,
+      "grad_norm": 0.04705330729484558,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 109284
+    },
+    {
+      "epoch": 11.231963001027749,
+      "grad_norm": 0.05176223814487457,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 109287
+    },
+    {
+      "epoch": 11.232271325796505,
+      "grad_norm": 0.036808304488658905,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 109290
+    },
+    {
+      "epoch": 11.232579650565262,
+      "grad_norm": 0.05182893946766853,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 109293
+    },
+    {
+      "epoch": 11.232887975334018,
+      "grad_norm": 0.04240572452545166,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 109296
+    },
+    {
+      "epoch": 11.233196300102774,
+      "grad_norm": 0.06471949815750122,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 109299
+    },
+    {
+      "epoch": 11.233504624871532,
+      "grad_norm": 0.09615979343652725,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 109302
+    },
+    {
+      "epoch": 11.233812949640289,
+      "grad_norm": 0.16381511092185974,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 109305
+    },
+    {
+      "epoch": 11.234121274409045,
+      "grad_norm": 0.10431195050477982,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 109308
+    },
+    {
+      "epoch": 11.234429599177801,
+      "grad_norm": 0.10584813356399536,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 109311
+    },
+    {
+      "epoch": 11.234737923946557,
+      "grad_norm": 0.06652256101369858,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 109314
+    },
+    {
+      "epoch": 11.235046248715314,
+      "grad_norm": 0.054468441754579544,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 109317
+    },
+    {
+      "epoch": 11.23535457348407,
+      "grad_norm": 0.03153740614652634,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 109320
+    },
+    {
+      "epoch": 11.235662898252826,
+      "grad_norm": 0.05361641198396683,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 109323
+    },
+    {
+      "epoch": 11.235971223021583,
+      "grad_norm": 0.16195333003997803,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 109326
+    },
+    {
+      "epoch": 11.236279547790339,
+      "grad_norm": 0.050422552973032,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 109329
+    },
+    {
+      "epoch": 11.236587872559095,
+      "grad_norm": 0.057867612689733505,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 109332
+    },
+    {
+      "epoch": 11.236896197327852,
+      "grad_norm": 0.07480079680681229,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 109335
+    },
+    {
+      "epoch": 11.237204522096608,
+      "grad_norm": 0.05645005404949188,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 109338
+    },
+    {
+      "epoch": 11.237512846865364,
+      "grad_norm": 0.05145471915602684,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 109341
+    },
+    {
+      "epoch": 11.23782117163412,
+      "grad_norm": 0.03488278388977051,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 109344
+    },
+    {
+      "epoch": 11.238129496402877,
+      "grad_norm": 0.050637271255254745,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 109347
+    },
+    {
+      "epoch": 11.238437821171635,
+      "grad_norm": 0.06615863740444183,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 109350
+    },
+    {
+      "epoch": 11.238746145940391,
+      "grad_norm": 0.03594106063246727,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 109353
+    },
+    {
+      "epoch": 11.239054470709148,
+      "grad_norm": 0.07424032688140869,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 109356
+    },
+    {
+      "epoch": 11.239362795477904,
+      "grad_norm": 0.10016637295484543,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 109359
+    },
+    {
+      "epoch": 11.23967112024666,
+      "grad_norm": 0.09463337808847427,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 109362
+    },
+    {
+      "epoch": 11.239979445015416,
+      "grad_norm": 0.041809163987636566,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 109365
+    },
+    {
+      "epoch": 11.240287769784173,
+      "grad_norm": 0.08738263696432114,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 109368
+    },
+    {
+      "epoch": 11.240596094552929,
+      "grad_norm": 0.13975435495376587,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 109371
+    },
+    {
+      "epoch": 11.240904419321685,
+      "grad_norm": 0.059366509318351746,
+      "learning_rate": 0.01,
+      "loss": 1.9703,
+      "step": 109374
+    },
+    {
+      "epoch": 11.241212744090442,
+      "grad_norm": 0.0640646293759346,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 109377
+    },
+    {
+      "epoch": 11.241521068859198,
+      "grad_norm": 0.04212968051433563,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 109380
+    },
+    {
+      "epoch": 11.241829393627954,
+      "grad_norm": 0.04662049189209938,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 109383
+    },
+    {
+      "epoch": 11.24213771839671,
+      "grad_norm": 0.06030857935547829,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 109386
+    },
+    {
+      "epoch": 11.242446043165467,
+      "grad_norm": 0.0616946667432785,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 109389
+    },
+    {
+      "epoch": 11.242754367934223,
+      "grad_norm": 0.06546429544687271,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 109392
+    },
+    {
+      "epoch": 11.243062692702981,
+      "grad_norm": 0.04250052571296692,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 109395
+    },
+    {
+      "epoch": 11.243371017471738,
+      "grad_norm": 0.04504207894206047,
+      "learning_rate": 0.01,
+      "loss": 1.8971,
+      "step": 109398
+    },
+    {
+      "epoch": 11.243679342240494,
+      "grad_norm": 0.044746581465005875,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 109401
+    },
+    {
+      "epoch": 11.24398766700925,
+      "grad_norm": 0.0798160657286644,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 109404
+    },
+    {
+      "epoch": 11.244295991778007,
+      "grad_norm": 0.09753091633319855,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 109407
+    },
+    {
+      "epoch": 11.244604316546763,
+      "grad_norm": 0.1443919986486435,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 109410
+    },
+    {
+      "epoch": 11.24491264131552,
+      "grad_norm": 0.11860787868499756,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 109413
+    },
+    {
+      "epoch": 11.245220966084275,
+      "grad_norm": 0.046349428594112396,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 109416
+    },
+    {
+      "epoch": 11.245529290853032,
+      "grad_norm": 0.03825264796614647,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 109419
+    },
+    {
+      "epoch": 11.245837615621788,
+      "grad_norm": 0.03915859013795853,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 109422
+    },
+    {
+      "epoch": 11.246145940390544,
+      "grad_norm": 0.04815271124243736,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 109425
+    },
+    {
+      "epoch": 11.2464542651593,
+      "grad_norm": 0.06440318375825882,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 109428
+    },
+    {
+      "epoch": 11.246762589928057,
+      "grad_norm": 0.07187019288539886,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 109431
+    },
+    {
+      "epoch": 11.247070914696813,
+      "grad_norm": 0.07239823788404465,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 109434
+    },
+    {
+      "epoch": 11.24737923946557,
+      "grad_norm": 0.08541743457317352,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 109437
+    },
+    {
+      "epoch": 11.247687564234328,
+      "grad_norm": 0.046377550810575485,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 109440
+    },
+    {
+      "epoch": 11.247995889003084,
+      "grad_norm": 0.12575428187847137,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 109443
+    },
+    {
+      "epoch": 11.24830421377184,
+      "grad_norm": 0.14709772169589996,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 109446
+    },
+    {
+      "epoch": 11.248612538540597,
+      "grad_norm": 0.05810967832803726,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 109449
+    },
+    {
+      "epoch": 11.248920863309353,
+      "grad_norm": 0.04419909790158272,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 109452
+    },
+    {
+      "epoch": 11.24922918807811,
+      "grad_norm": 0.06409557908773422,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 109455
+    },
+    {
+      "epoch": 11.249537512846866,
+      "grad_norm": 0.03288624808192253,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 109458
+    },
+    {
+      "epoch": 11.249845837615622,
+      "grad_norm": 0.05867474898695946,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 109461
+    },
+    {
+      "epoch": 11.250154162384378,
+      "grad_norm": 0.08372747153043747,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 109464
+    },
+    {
+      "epoch": 11.250462487153134,
+      "grad_norm": 0.05529605224728584,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 109467
+    },
+    {
+      "epoch": 11.25077081192189,
+      "grad_norm": 0.09608209878206253,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 109470
+    },
+    {
+      "epoch": 11.251079136690647,
+      "grad_norm": 0.04316975548863411,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 109473
+    },
+    {
+      "epoch": 11.251387461459403,
+      "grad_norm": 0.06034945324063301,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 109476
+    },
+    {
+      "epoch": 11.25169578622816,
+      "grad_norm": 0.08070293813943863,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 109479
+    },
+    {
+      "epoch": 11.252004110996916,
+      "grad_norm": 0.1095161959528923,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 109482
+    },
+    {
+      "epoch": 11.252312435765674,
+      "grad_norm": 0.038463711738586426,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 109485
+    },
+    {
+      "epoch": 11.25262076053443,
+      "grad_norm": 0.04902822524309158,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 109488
+    },
+    {
+      "epoch": 11.252929085303187,
+      "grad_norm": 0.03326786682009697,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 109491
+    },
+    {
+      "epoch": 11.253237410071943,
+      "grad_norm": 0.045016270130872726,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 109494
+    },
+    {
+      "epoch": 11.2535457348407,
+      "grad_norm": 0.05962817743420601,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 109497
+    },
+    {
+      "epoch": 11.253854059609456,
+      "grad_norm": 0.07711559534072876,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 109500
+    },
+    {
+      "epoch": 11.254162384378212,
+      "grad_norm": 0.05092623457312584,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 109503
+    },
+    {
+      "epoch": 11.254470709146968,
+      "grad_norm": 0.054487742483615875,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 109506
+    },
+    {
+      "epoch": 11.254779033915725,
+      "grad_norm": 0.03977763652801514,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 109509
+    },
+    {
+      "epoch": 11.25508735868448,
+      "grad_norm": 0.03340889886021614,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 109512
+    },
+    {
+      "epoch": 11.255395683453237,
+      "grad_norm": 0.05879076197743416,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 109515
+    },
+    {
+      "epoch": 11.255704008221993,
+      "grad_norm": 0.1080188900232315,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 109518
+    },
+    {
+      "epoch": 11.25601233299075,
+      "grad_norm": 0.05137190967798233,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 109521
+    },
+    {
+      "epoch": 11.256320657759506,
+      "grad_norm": 0.1083584800362587,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 109524
+    },
+    {
+      "epoch": 11.256628982528262,
+      "grad_norm": 0.05661296844482422,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 109527
+    },
+    {
+      "epoch": 11.256937307297019,
+      "grad_norm": 0.11602769047021866,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 109530
+    },
+    {
+      "epoch": 11.257245632065777,
+      "grad_norm": 0.06640040129423141,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 109533
+    },
+    {
+      "epoch": 11.257553956834533,
+      "grad_norm": 0.054868511855602264,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 109536
+    },
+    {
+      "epoch": 11.25786228160329,
+      "grad_norm": 0.053349707275629044,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 109539
+    },
+    {
+      "epoch": 11.258170606372046,
+      "grad_norm": 0.04147634282708168,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 109542
+    },
+    {
+      "epoch": 11.258478931140802,
+      "grad_norm": 0.04902909696102142,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 109545
+    },
+    {
+      "epoch": 11.258787255909558,
+      "grad_norm": 0.10914880037307739,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 109548
+    },
+    {
+      "epoch": 11.259095580678315,
+      "grad_norm": 0.1410733461380005,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 109551
+    },
+    {
+      "epoch": 11.259403905447071,
+      "grad_norm": 0.07386939227581024,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 109554
+    },
+    {
+      "epoch": 11.259712230215827,
+      "grad_norm": 0.08032076060771942,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 109557
+    },
+    {
+      "epoch": 11.260020554984584,
+      "grad_norm": 0.03709203377366066,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 109560
+    },
+    {
+      "epoch": 11.26032887975334,
+      "grad_norm": 0.0418517105281353,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 109563
+    },
+    {
+      "epoch": 11.260637204522096,
+      "grad_norm": 0.039387185126543045,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 109566
+    },
+    {
+      "epoch": 11.260945529290852,
+      "grad_norm": 0.06155497208237648,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 109569
+    },
+    {
+      "epoch": 11.261253854059609,
+      "grad_norm": 0.1154722049832344,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 109572
+    },
+    {
+      "epoch": 11.261562178828365,
+      "grad_norm": 0.05332230404019356,
+      "learning_rate": 0.01,
+      "loss": 1.974,
+      "step": 109575
+    },
+    {
+      "epoch": 11.261870503597123,
+      "grad_norm": 0.05801251530647278,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 109578
+    },
+    {
+      "epoch": 11.26217882836588,
+      "grad_norm": 0.05469183251261711,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 109581
+    },
+    {
+      "epoch": 11.262487153134636,
+      "grad_norm": 0.042408592998981476,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 109584
+    },
+    {
+      "epoch": 11.262795477903392,
+      "grad_norm": 0.05151085928082466,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 109587
+    },
+    {
+      "epoch": 11.263103802672148,
+      "grad_norm": 0.060142409056425095,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 109590
+    },
+    {
+      "epoch": 11.263412127440905,
+      "grad_norm": 0.10095762461423874,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 109593
+    },
+    {
+      "epoch": 11.263720452209661,
+      "grad_norm": 0.05893385782837868,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 109596
+    },
+    {
+      "epoch": 11.264028776978417,
+      "grad_norm": 0.09010785818099976,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 109599
+    },
+    {
+      "epoch": 11.264337101747174,
+      "grad_norm": 0.08859475702047348,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 109602
+    },
+    {
+      "epoch": 11.26464542651593,
+      "grad_norm": 0.09706424921751022,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 109605
+    },
+    {
+      "epoch": 11.264953751284686,
+      "grad_norm": 0.0626566931605339,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 109608
+    },
+    {
+      "epoch": 11.265262076053443,
+      "grad_norm": 0.04158549755811691,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 109611
+    },
+    {
+      "epoch": 11.265570400822199,
+      "grad_norm": 0.06176656857132912,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 109614
+    },
+    {
+      "epoch": 11.265878725590955,
+      "grad_norm": 0.07009613513946533,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 109617
+    },
+    {
+      "epoch": 11.266187050359711,
+      "grad_norm": 0.07909108698368073,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 109620
+    },
+    {
+      "epoch": 11.266495375128468,
+      "grad_norm": 0.07806423306465149,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 109623
+    },
+    {
+      "epoch": 11.266803699897226,
+      "grad_norm": 0.05793457478284836,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 109626
+    },
+    {
+      "epoch": 11.267112024665982,
+      "grad_norm": 0.05238494277000427,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 109629
+    },
+    {
+      "epoch": 11.267420349434738,
+      "grad_norm": 0.03375493735074997,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 109632
+    },
+    {
+      "epoch": 11.267728674203495,
+      "grad_norm": 0.04535922035574913,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 109635
+    },
+    {
+      "epoch": 11.268036998972251,
+      "grad_norm": 0.04779687523841858,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 109638
+    },
+    {
+      "epoch": 11.268345323741007,
+      "grad_norm": 0.07404086738824844,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 109641
+    },
+    {
+      "epoch": 11.268653648509764,
+      "grad_norm": 0.07996059954166412,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 109644
+    },
+    {
+      "epoch": 11.26896197327852,
+      "grad_norm": 0.11115096509456635,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 109647
+    },
+    {
+      "epoch": 11.269270298047276,
+      "grad_norm": 0.0490814708173275,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 109650
+    },
+    {
+      "epoch": 11.269578622816033,
+      "grad_norm": 0.11355841159820557,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 109653
+    },
+    {
+      "epoch": 11.269886947584789,
+      "grad_norm": 0.029677560552954674,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 109656
+    },
+    {
+      "epoch": 11.270195272353545,
+      "grad_norm": 0.05585773289203644,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 109659
+    },
+    {
+      "epoch": 11.270503597122302,
+      "grad_norm": 0.126790851354599,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 109662
+    },
+    {
+      "epoch": 11.270811921891058,
+      "grad_norm": 0.0652090534567833,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 109665
+    },
+    {
+      "epoch": 11.271120246659814,
+      "grad_norm": 0.04160231351852417,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 109668
+    },
+    {
+      "epoch": 11.271428571428572,
+      "grad_norm": 0.07735766470432281,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 109671
+    },
+    {
+      "epoch": 11.271736896197329,
+      "grad_norm": 0.03592461347579956,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 109674
+    },
+    {
+      "epoch": 11.272045220966085,
+      "grad_norm": 0.042836081236600876,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 109677
+    },
+    {
+      "epoch": 11.272353545734841,
+      "grad_norm": 0.03916572406888008,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 109680
+    },
+    {
+      "epoch": 11.272661870503597,
+      "grad_norm": 0.04746629670262337,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 109683
+    },
+    {
+      "epoch": 11.272970195272354,
+      "grad_norm": 0.052999433130025864,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 109686
+    },
+    {
+      "epoch": 11.27327852004111,
+      "grad_norm": 0.047792401164770126,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 109689
+    },
+    {
+      "epoch": 11.273586844809866,
+      "grad_norm": 0.05411859229207039,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 109692
+    },
+    {
+      "epoch": 11.273895169578623,
+      "grad_norm": 0.041686661541461945,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 109695
+    },
+    {
+      "epoch": 11.274203494347379,
+      "grad_norm": 0.061750467866659164,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 109698
+    },
+    {
+      "epoch": 11.274511819116135,
+      "grad_norm": 0.05879466235637665,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 109701
+    },
+    {
+      "epoch": 11.274820143884892,
+      "grad_norm": 0.07283945381641388,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 109704
+    },
+    {
+      "epoch": 11.275128468653648,
+      "grad_norm": 0.0472678504884243,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 109707
+    },
+    {
+      "epoch": 11.275436793422404,
+      "grad_norm": 0.12681566178798676,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 109710
+    },
+    {
+      "epoch": 11.27574511819116,
+      "grad_norm": 0.033542435616254807,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 109713
+    },
+    {
+      "epoch": 11.276053442959919,
+      "grad_norm": 0.04590669274330139,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 109716
+    },
+    {
+      "epoch": 11.276361767728675,
+      "grad_norm": 0.10272151976823807,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 109719
+    },
+    {
+      "epoch": 11.276670092497431,
+      "grad_norm": 0.06324021518230438,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 109722
+    },
+    {
+      "epoch": 11.276978417266188,
+      "grad_norm": 0.05721462517976761,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 109725
+    },
+    {
+      "epoch": 11.277286742034944,
+      "grad_norm": 0.04144064337015152,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 109728
+    },
+    {
+      "epoch": 11.2775950668037,
+      "grad_norm": 0.041763003915548325,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 109731
+    },
+    {
+      "epoch": 11.277903391572456,
+      "grad_norm": 0.044674765318632126,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 109734
+    },
+    {
+      "epoch": 11.278211716341213,
+      "grad_norm": 0.04908163473010063,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 109737
+    },
+    {
+      "epoch": 11.278520041109969,
+      "grad_norm": 0.06857731193304062,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 109740
+    },
+    {
+      "epoch": 11.278828365878725,
+      "grad_norm": 0.07278558611869812,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 109743
+    },
+    {
+      "epoch": 11.279136690647482,
+      "grad_norm": 0.07408792525529861,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 109746
+    },
+    {
+      "epoch": 11.279445015416238,
+      "grad_norm": 0.04330144822597504,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 109749
+    },
+    {
+      "epoch": 11.279753340184994,
+      "grad_norm": 0.06111234799027443,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 109752
+    },
+    {
+      "epoch": 11.28006166495375,
+      "grad_norm": 0.14030754566192627,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 109755
+    },
+    {
+      "epoch": 11.280369989722507,
+      "grad_norm": 0.05981336161494255,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 109758
+    },
+    {
+      "epoch": 11.280678314491265,
+      "grad_norm": 0.06106378883123398,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 109761
+    },
+    {
+      "epoch": 11.280986639260021,
+      "grad_norm": 0.033522509038448334,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 109764
+    },
+    {
+      "epoch": 11.281294964028778,
+      "grad_norm": 0.03565683215856552,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 109767
+    },
+    {
+      "epoch": 11.281603288797534,
+      "grad_norm": 0.035627976059913635,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 109770
+    },
+    {
+      "epoch": 11.28191161356629,
+      "grad_norm": 0.06138899549841881,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 109773
+    },
+    {
+      "epoch": 11.282219938335047,
+      "grad_norm": 0.08396279811859131,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 109776
+    },
+    {
+      "epoch": 11.282528263103803,
+      "grad_norm": 0.06644977629184723,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 109779
+    },
+    {
+      "epoch": 11.28283658787256,
+      "grad_norm": 0.049319833517074585,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 109782
+    },
+    {
+      "epoch": 11.283144912641315,
+      "grad_norm": 0.0407063327729702,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 109785
+    },
+    {
+      "epoch": 11.283453237410072,
+      "grad_norm": 0.0696980431675911,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 109788
+    },
+    {
+      "epoch": 11.283761562178828,
+      "grad_norm": 0.05926033854484558,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 109791
+    },
+    {
+      "epoch": 11.284069886947584,
+      "grad_norm": 0.04289416968822479,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 109794
+    },
+    {
+      "epoch": 11.28437821171634,
+      "grad_norm": 0.03649679198861122,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 109797
+    },
+    {
+      "epoch": 11.284686536485097,
+      "grad_norm": 0.11007057130336761,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 109800
+    },
+    {
+      "epoch": 11.284994861253853,
+      "grad_norm": 0.051835693418979645,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 109803
+    },
+    {
+      "epoch": 11.28530318602261,
+      "grad_norm": 0.03663384169340134,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 109806
+    },
+    {
+      "epoch": 11.285611510791368,
+      "grad_norm": 0.0417393334209919,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 109809
+    },
+    {
+      "epoch": 11.285919835560124,
+      "grad_norm": 0.08153456449508667,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 109812
+    },
+    {
+      "epoch": 11.28622816032888,
+      "grad_norm": 0.05325230956077576,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 109815
+    },
+    {
+      "epoch": 11.286536485097637,
+      "grad_norm": 0.05065246298909187,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 109818
+    },
+    {
+      "epoch": 11.286844809866393,
+      "grad_norm": 0.144111767411232,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 109821
+    },
+    {
+      "epoch": 11.28715313463515,
+      "grad_norm": 0.09719037264585495,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 109824
+    },
+    {
+      "epoch": 11.287461459403906,
+      "grad_norm": 0.03678983822464943,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 109827
+    },
+    {
+      "epoch": 11.287769784172662,
+      "grad_norm": 0.08810188621282578,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 109830
+    },
+    {
+      "epoch": 11.288078108941418,
+      "grad_norm": 0.05677026882767677,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 109833
+    },
+    {
+      "epoch": 11.288386433710174,
+      "grad_norm": 0.09825364500284195,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 109836
+    },
+    {
+      "epoch": 11.28869475847893,
+      "grad_norm": 0.061190493404865265,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 109839
+    },
+    {
+      "epoch": 11.289003083247687,
+      "grad_norm": 0.060511138290166855,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 109842
+    },
+    {
+      "epoch": 11.289311408016443,
+      "grad_norm": 0.056555476039648056,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 109845
+    },
+    {
+      "epoch": 11.2896197327852,
+      "grad_norm": 0.07167981564998627,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 109848
+    },
+    {
+      "epoch": 11.289928057553956,
+      "grad_norm": 0.08068525046110153,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 109851
+    },
+    {
+      "epoch": 11.290236382322714,
+      "grad_norm": 0.07005833089351654,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 109854
+    },
+    {
+      "epoch": 11.29054470709147,
+      "grad_norm": 0.05481821298599243,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 109857
+    },
+    {
+      "epoch": 11.290853031860227,
+      "grad_norm": 0.0465649776160717,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 109860
+    },
+    {
+      "epoch": 11.291161356628983,
+      "grad_norm": 0.05188848450779915,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 109863
+    },
+    {
+      "epoch": 11.29146968139774,
+      "grad_norm": 0.04130940139293671,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 109866
+    },
+    {
+      "epoch": 11.291778006166496,
+      "grad_norm": 0.10015565901994705,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 109869
+    },
+    {
+      "epoch": 11.292086330935252,
+      "grad_norm": 0.042796485126018524,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 109872
+    },
+    {
+      "epoch": 11.292394655704008,
+      "grad_norm": 0.12834590673446655,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 109875
+    },
+    {
+      "epoch": 11.292702980472765,
+      "grad_norm": 0.17933888733386993,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 109878
+    },
+    {
+      "epoch": 11.29301130524152,
+      "grad_norm": 0.10874367505311966,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 109881
+    },
+    {
+      "epoch": 11.293319630010277,
+      "grad_norm": 0.11151270568370819,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 109884
+    },
+    {
+      "epoch": 11.293627954779033,
+      "grad_norm": 0.06689925491809845,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 109887
+    },
+    {
+      "epoch": 11.29393627954779,
+      "grad_norm": 0.04171884059906006,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 109890
+    },
+    {
+      "epoch": 11.294244604316546,
+      "grad_norm": 0.04404659941792488,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 109893
+    },
+    {
+      "epoch": 11.294552929085302,
+      "grad_norm": 0.05716705322265625,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 109896
+    },
+    {
+      "epoch": 11.294861253854059,
+      "grad_norm": 0.034998469054698944,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 109899
+    },
+    {
+      "epoch": 11.295169578622817,
+      "grad_norm": 0.04066811501979828,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 109902
+    },
+    {
+      "epoch": 11.295477903391573,
+      "grad_norm": 0.10792528092861176,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 109905
+    },
+    {
+      "epoch": 11.29578622816033,
+      "grad_norm": 0.0576433427631855,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 109908
+    },
+    {
+      "epoch": 11.296094552929086,
+      "grad_norm": 0.11625795811414719,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 109911
+    },
+    {
+      "epoch": 11.296402877697842,
+      "grad_norm": 0.03960056230425835,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 109914
+    },
+    {
+      "epoch": 11.296711202466598,
+      "grad_norm": 0.045928679406642914,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 109917
+    },
+    {
+      "epoch": 11.297019527235355,
+      "grad_norm": 0.05385993421077728,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 109920
+    },
+    {
+      "epoch": 11.297327852004111,
+      "grad_norm": 0.059563588351011276,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 109923
+    },
+    {
+      "epoch": 11.297636176772867,
+      "grad_norm": 0.04696774110198021,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 109926
+    },
+    {
+      "epoch": 11.297944501541624,
+      "grad_norm": 0.061480265110731125,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 109929
+    },
+    {
+      "epoch": 11.29825282631038,
+      "grad_norm": 0.07148461043834686,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 109932
+    },
+    {
+      "epoch": 11.298561151079136,
+      "grad_norm": 0.048205867409706116,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 109935
+    },
+    {
+      "epoch": 11.298869475847892,
+      "grad_norm": 0.031730081886053085,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 109938
+    },
+    {
+      "epoch": 11.299177800616649,
+      "grad_norm": 0.04472745582461357,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 109941
+    },
+    {
+      "epoch": 11.299486125385407,
+      "grad_norm": 0.10435371845960617,
+      "learning_rate": 0.01,
+      "loss": 1.8878,
+      "step": 109944
+    },
+    {
+      "epoch": 11.299794450154163,
+      "grad_norm": 0.1116335466504097,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 109947
+    },
+    {
+      "epoch": 11.30010277492292,
+      "grad_norm": 0.04797052964568138,
+      "learning_rate": 0.01,
+      "loss": 1.9701,
+      "step": 109950
+    },
+    {
+      "epoch": 11.300411099691676,
+      "grad_norm": 0.06291238218545914,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 109953
+    },
+    {
+      "epoch": 11.300719424460432,
+      "grad_norm": 0.13573448359966278,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 109956
+    },
+    {
+      "epoch": 11.301027749229188,
+      "grad_norm": 0.042475681751966476,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 109959
+    },
+    {
+      "epoch": 11.301336073997945,
+      "grad_norm": 0.12064820528030396,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 109962
+    },
+    {
+      "epoch": 11.301644398766701,
+      "grad_norm": 0.0992126539349556,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 109965
+    },
+    {
+      "epoch": 11.301952723535457,
+      "grad_norm": 0.03413967788219452,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 109968
+    },
+    {
+      "epoch": 11.302261048304214,
+      "grad_norm": 0.08574753999710083,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 109971
+    },
+    {
+      "epoch": 11.30256937307297,
+      "grad_norm": 0.07113710790872574,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 109974
+    },
+    {
+      "epoch": 11.302877697841726,
+      "grad_norm": 0.06103216111660004,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 109977
+    },
+    {
+      "epoch": 11.303186022610483,
+      "grad_norm": 0.05378776416182518,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 109980
+    },
+    {
+      "epoch": 11.303494347379239,
+      "grad_norm": 0.04234358295798302,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 109983
+    },
+    {
+      "epoch": 11.303802672147995,
+      "grad_norm": 0.044765591621398926,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 109986
+    },
+    {
+      "epoch": 11.304110996916751,
+      "grad_norm": 0.04169373959302902,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 109989
+    },
+    {
+      "epoch": 11.30441932168551,
+      "grad_norm": 0.10466013848781586,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 109992
+    },
+    {
+      "epoch": 11.304727646454266,
+      "grad_norm": 0.0452800951898098,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 109995
+    },
+    {
+      "epoch": 11.305035971223022,
+      "grad_norm": 0.06505489349365234,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 109998
+    },
+    {
+      "epoch": 11.305344295991778,
+      "grad_norm": 0.05290715768933296,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 110001
+    },
+    {
+      "epoch": 11.305652620760535,
+      "grad_norm": 0.03846494108438492,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 110004
+    },
+    {
+      "epoch": 11.305960945529291,
+      "grad_norm": 0.041611991822719574,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 110007
+    },
+    {
+      "epoch": 11.306269270298047,
+      "grad_norm": 0.09771797060966492,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 110010
+    },
+    {
+      "epoch": 11.306577595066804,
+      "grad_norm": 0.0961562916636467,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 110013
+    },
+    {
+      "epoch": 11.30688591983556,
+      "grad_norm": 0.12497589737176895,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 110016
+    },
+    {
+      "epoch": 11.307194244604316,
+      "grad_norm": 0.10389663279056549,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 110019
+    },
+    {
+      "epoch": 11.307502569373073,
+      "grad_norm": 0.055324338376522064,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 110022
+    },
+    {
+      "epoch": 11.307810894141829,
+      "grad_norm": 0.04517209529876709,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 110025
+    },
+    {
+      "epoch": 11.308119218910585,
+      "grad_norm": 0.11330480873584747,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 110028
+    },
+    {
+      "epoch": 11.308427543679342,
+      "grad_norm": 0.03730342537164688,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 110031
+    },
+    {
+      "epoch": 11.308735868448098,
+      "grad_norm": 0.03379646688699722,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 110034
+    },
+    {
+      "epoch": 11.309044193216856,
+      "grad_norm": 0.04477003961801529,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 110037
+    },
+    {
+      "epoch": 11.309352517985612,
+      "grad_norm": 0.05759940668940544,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 110040
+    },
+    {
+      "epoch": 11.309660842754369,
+      "grad_norm": 0.06768691539764404,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 110043
+    },
+    {
+      "epoch": 11.309969167523125,
+      "grad_norm": 0.07362806051969528,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 110046
+    },
+    {
+      "epoch": 11.310277492291881,
+      "grad_norm": 0.07099780440330505,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 110049
+    },
+    {
+      "epoch": 11.310585817060637,
+      "grad_norm": 0.07642699033021927,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 110052
+    },
+    {
+      "epoch": 11.310894141829394,
+      "grad_norm": 0.05950343236327171,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 110055
+    },
+    {
+      "epoch": 11.31120246659815,
+      "grad_norm": 0.04728717356920242,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 110058
+    },
+    {
+      "epoch": 11.311510791366906,
+      "grad_norm": 0.052025727927684784,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 110061
+    },
+    {
+      "epoch": 11.311819116135663,
+      "grad_norm": 0.03810366615653038,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 110064
+    },
+    {
+      "epoch": 11.312127440904419,
+      "grad_norm": 0.12762334942817688,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 110067
+    },
+    {
+      "epoch": 11.312435765673175,
+      "grad_norm": 0.06287941336631775,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 110070
+    },
+    {
+      "epoch": 11.312744090441932,
+      "grad_norm": 0.050556425005197525,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 110073
+    },
+    {
+      "epoch": 11.313052415210688,
+      "grad_norm": 0.052405212074518204,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 110076
+    },
+    {
+      "epoch": 11.313360739979444,
+      "grad_norm": 0.05936199799180031,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 110079
+    },
+    {
+      "epoch": 11.3136690647482,
+      "grad_norm": 0.08959811925888062,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 110082
+    },
+    {
+      "epoch": 11.313977389516959,
+      "grad_norm": 0.06256865710020065,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 110085
+    },
+    {
+      "epoch": 11.314285714285715,
+      "grad_norm": 0.07469283044338226,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 110088
+    },
+    {
+      "epoch": 11.314594039054471,
+      "grad_norm": 0.0618683397769928,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 110091
+    },
+    {
+      "epoch": 11.314902363823228,
+      "grad_norm": 0.05982283130288124,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 110094
+    },
+    {
+      "epoch": 11.315210688591984,
+      "grad_norm": 0.0350305549800396,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 110097
+    },
+    {
+      "epoch": 11.31551901336074,
+      "grad_norm": 0.04055385664105415,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 110100
+    },
+    {
+      "epoch": 11.315827338129496,
+      "grad_norm": 0.044621482491493225,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 110103
+    },
+    {
+      "epoch": 11.316135662898253,
+      "grad_norm": 0.12125648558139801,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 110106
+    },
+    {
+      "epoch": 11.316443987667009,
+      "grad_norm": 0.11933153867721558,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 110109
+    },
+    {
+      "epoch": 11.316752312435765,
+      "grad_norm": 0.08058884739875793,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 110112
+    },
+    {
+      "epoch": 11.317060637204522,
+      "grad_norm": 0.05547872930765152,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 110115
+    },
+    {
+      "epoch": 11.317368961973278,
+      "grad_norm": 0.05052429437637329,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 110118
+    },
+    {
+      "epoch": 11.317677286742034,
+      "grad_norm": 0.03526543080806732,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 110121
+    },
+    {
+      "epoch": 11.31798561151079,
+      "grad_norm": 0.0412910059094429,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 110124
+    },
+    {
+      "epoch": 11.318293936279547,
+      "grad_norm": 0.08957338333129883,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 110127
+    },
+    {
+      "epoch": 11.318602261048305,
+      "grad_norm": 0.12582367658615112,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 110130
+    },
+    {
+      "epoch": 11.318910585817061,
+      "grad_norm": 0.07041830569505692,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 110133
+    },
+    {
+      "epoch": 11.319218910585818,
+      "grad_norm": 0.04017024114727974,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 110136
+    },
+    {
+      "epoch": 11.319527235354574,
+      "grad_norm": 0.05196457728743553,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 110139
+    },
+    {
+      "epoch": 11.31983556012333,
+      "grad_norm": 0.05180789530277252,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 110142
+    },
+    {
+      "epoch": 11.320143884892087,
+      "grad_norm": 0.040728844702243805,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 110145
+    },
+    {
+      "epoch": 11.320452209660843,
+      "grad_norm": 0.056863050907850266,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 110148
+    },
+    {
+      "epoch": 11.3207605344296,
+      "grad_norm": 0.06815367192029953,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 110151
+    },
+    {
+      "epoch": 11.321068859198355,
+      "grad_norm": 0.04380220174789429,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 110154
+    },
+    {
+      "epoch": 11.321377183967112,
+      "grad_norm": 0.12742555141448975,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 110157
+    },
+    {
+      "epoch": 11.321685508735868,
+      "grad_norm": 0.08063594251871109,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 110160
+    },
+    {
+      "epoch": 11.321993833504624,
+      "grad_norm": 0.07678404450416565,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 110163
+    },
+    {
+      "epoch": 11.32230215827338,
+      "grad_norm": 0.12479738891124725,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 110166
+    },
+    {
+      "epoch": 11.322610483042137,
+      "grad_norm": 0.13514432311058044,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 110169
+    },
+    {
+      "epoch": 11.322918807810893,
+      "grad_norm": 0.13325108587741852,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 110172
+    },
+    {
+      "epoch": 11.323227132579651,
+      "grad_norm": 0.12824301421642303,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 110175
+    },
+    {
+      "epoch": 11.323535457348408,
+      "grad_norm": 0.05898331105709076,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 110178
+    },
+    {
+      "epoch": 11.323843782117164,
+      "grad_norm": 0.0602484755218029,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 110181
+    },
+    {
+      "epoch": 11.32415210688592,
+      "grad_norm": 0.060096725821495056,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 110184
+    },
+    {
+      "epoch": 11.324460431654677,
+      "grad_norm": 0.07417149096727371,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 110187
+    },
+    {
+      "epoch": 11.324768756423433,
+      "grad_norm": 0.05343138799071312,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 110190
+    },
+    {
+      "epoch": 11.32507708119219,
+      "grad_norm": 0.03471410274505615,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 110193
+    },
+    {
+      "epoch": 11.325385405960946,
+      "grad_norm": 0.04164749011397362,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 110196
+    },
+    {
+      "epoch": 11.325693730729702,
+      "grad_norm": 0.04597137123346329,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 110199
+    },
+    {
+      "epoch": 11.326002055498458,
+      "grad_norm": 0.051889240741729736,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 110202
+    },
+    {
+      "epoch": 11.326310380267214,
+      "grad_norm": 0.05259767174720764,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 110205
+    },
+    {
+      "epoch": 11.32661870503597,
+      "grad_norm": 0.10912121832370758,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 110208
+    },
+    {
+      "epoch": 11.326927029804727,
+      "grad_norm": 0.09809891879558563,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 110211
+    },
+    {
+      "epoch": 11.327235354573483,
+      "grad_norm": 0.12826496362686157,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 110214
+    },
+    {
+      "epoch": 11.32754367934224,
+      "grad_norm": 0.03926202282309532,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 110217
+    },
+    {
+      "epoch": 11.327852004110998,
+      "grad_norm": 0.08739673346281052,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 110220
+    },
+    {
+      "epoch": 11.328160328879754,
+      "grad_norm": 0.1496817171573639,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 110223
+    },
+    {
+      "epoch": 11.32846865364851,
+      "grad_norm": 0.06414765864610672,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 110226
+    },
+    {
+      "epoch": 11.328776978417267,
+      "grad_norm": 0.04865317791700363,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 110229
+    },
+    {
+      "epoch": 11.329085303186023,
+      "grad_norm": 0.045886069536209106,
+      "learning_rate": 0.01,
+      "loss": 1.9837,
+      "step": 110232
+    },
+    {
+      "epoch": 11.32939362795478,
+      "grad_norm": 0.047968700528144836,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 110235
+    },
+    {
+      "epoch": 11.329701952723536,
+      "grad_norm": 0.04234057292342186,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 110238
+    },
+    {
+      "epoch": 11.330010277492292,
+      "grad_norm": 0.04693339392542839,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 110241
+    },
+    {
+      "epoch": 11.330318602261048,
+      "grad_norm": 0.03062617965042591,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 110244
+    },
+    {
+      "epoch": 11.330626927029805,
+      "grad_norm": 0.04372096061706543,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 110247
+    },
+    {
+      "epoch": 11.33093525179856,
+      "grad_norm": 0.05214737355709076,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 110250
+    },
+    {
+      "epoch": 11.331243576567317,
+      "grad_norm": 0.04116861894726753,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 110253
+    },
+    {
+      "epoch": 11.331551901336073,
+      "grad_norm": 0.13659925758838654,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 110256
+    },
+    {
+      "epoch": 11.33186022610483,
+      "grad_norm": 0.07153276354074478,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 110259
+    },
+    {
+      "epoch": 11.332168550873586,
+      "grad_norm": 0.046991247683763504,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 110262
+    },
+    {
+      "epoch": 11.332476875642342,
+      "grad_norm": 0.03522711619734764,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 110265
+    },
+    {
+      "epoch": 11.3327852004111,
+      "grad_norm": 0.05653287097811699,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 110268
+    },
+    {
+      "epoch": 11.333093525179857,
+      "grad_norm": 0.14457939565181732,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 110271
+    },
+    {
+      "epoch": 11.333401849948613,
+      "grad_norm": 0.04326567426323891,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 110274
+    },
+    {
+      "epoch": 11.33371017471737,
+      "grad_norm": 0.08805881440639496,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 110277
+    },
+    {
+      "epoch": 11.334018499486126,
+      "grad_norm": 0.05980727821588516,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 110280
+    },
+    {
+      "epoch": 11.334326824254882,
+      "grad_norm": 0.08690473437309265,
+      "learning_rate": 0.01,
+      "loss": 1.9748,
+      "step": 110283
+    },
+    {
+      "epoch": 11.334635149023638,
+      "grad_norm": 0.04590797424316406,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 110286
+    },
+    {
+      "epoch": 11.334943473792395,
+      "grad_norm": 0.0751267671585083,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 110289
+    },
+    {
+      "epoch": 11.335251798561151,
+      "grad_norm": 0.06569565087556839,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 110292
+    },
+    {
+      "epoch": 11.335560123329907,
+      "grad_norm": 0.07285059988498688,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 110295
+    },
+    {
+      "epoch": 11.335868448098664,
+      "grad_norm": 0.13581648468971252,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 110298
+    },
+    {
+      "epoch": 11.33617677286742,
+      "grad_norm": 0.051626384258270264,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 110301
+    },
+    {
+      "epoch": 11.336485097636176,
+      "grad_norm": 0.06466205418109894,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 110304
+    },
+    {
+      "epoch": 11.336793422404932,
+      "grad_norm": 0.03319898620247841,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 110307
+    },
+    {
+      "epoch": 11.337101747173689,
+      "grad_norm": 0.04199283570051193,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 110310
+    },
+    {
+      "epoch": 11.337410071942447,
+      "grad_norm": 0.044764723628759384,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 110313
+    },
+    {
+      "epoch": 11.337718396711203,
+      "grad_norm": 0.036410681903362274,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 110316
+    },
+    {
+      "epoch": 11.33802672147996,
+      "grad_norm": 0.06427314132452011,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 110319
+    },
+    {
+      "epoch": 11.338335046248716,
+      "grad_norm": 0.08712683618068695,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 110322
+    },
+    {
+      "epoch": 11.338643371017472,
+      "grad_norm": 0.059931352734565735,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 110325
+    },
+    {
+      "epoch": 11.338951695786228,
+      "grad_norm": 0.03852000832557678,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 110328
+    },
+    {
+      "epoch": 11.339260020554985,
+      "grad_norm": 0.04189864918589592,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 110331
+    },
+    {
+      "epoch": 11.339568345323741,
+      "grad_norm": 0.09058728069067001,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 110334
+    },
+    {
+      "epoch": 11.339876670092497,
+      "grad_norm": 0.16685323417186737,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 110337
+    },
+    {
+      "epoch": 11.340184994861254,
+      "grad_norm": 0.11387660354375839,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 110340
+    },
+    {
+      "epoch": 11.34049331963001,
+      "grad_norm": 0.0847175195813179,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 110343
+    },
+    {
+      "epoch": 11.340801644398766,
+      "grad_norm": 0.05471164360642433,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 110346
+    },
+    {
+      "epoch": 11.341109969167523,
+      "grad_norm": 0.054385025054216385,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 110349
+    },
+    {
+      "epoch": 11.341418293936279,
+      "grad_norm": 0.043967898935079575,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 110352
+    },
+    {
+      "epoch": 11.341726618705035,
+      "grad_norm": 0.06875190883874893,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 110355
+    },
+    {
+      "epoch": 11.342034943473791,
+      "grad_norm": 0.046824581921100616,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 110358
+    },
+    {
+      "epoch": 11.34234326824255,
+      "grad_norm": 0.08238361030817032,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 110361
+    },
+    {
+      "epoch": 11.342651593011306,
+      "grad_norm": 0.11055076867341995,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 110364
+    },
+    {
+      "epoch": 11.342959917780062,
+      "grad_norm": 0.047248125076293945,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 110367
+    },
+    {
+      "epoch": 11.343268242548818,
+      "grad_norm": 0.1021493673324585,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 110370
+    },
+    {
+      "epoch": 11.343576567317575,
+      "grad_norm": 0.048846546560525894,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 110373
+    },
+    {
+      "epoch": 11.343884892086331,
+      "grad_norm": 0.08389388769865036,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 110376
+    },
+    {
+      "epoch": 11.344193216855087,
+      "grad_norm": 0.06787876039743423,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 110379
+    },
+    {
+      "epoch": 11.344501541623844,
+      "grad_norm": 0.0725322738289833,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 110382
+    },
+    {
+      "epoch": 11.3448098663926,
+      "grad_norm": 0.0973624661564827,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 110385
+    },
+    {
+      "epoch": 11.345118191161356,
+      "grad_norm": 0.09873227030038834,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 110388
+    },
+    {
+      "epoch": 11.345426515930113,
+      "grad_norm": 0.08333581686019897,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 110391
+    },
+    {
+      "epoch": 11.345734840698869,
+      "grad_norm": 0.0631309375166893,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 110394
+    },
+    {
+      "epoch": 11.346043165467625,
+      "grad_norm": 0.061351578682661057,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 110397
+    },
+    {
+      "epoch": 11.346351490236382,
+      "grad_norm": 0.06281182169914246,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 110400
+    },
+    {
+      "epoch": 11.34665981500514,
+      "grad_norm": 0.04762888327240944,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 110403
+    },
+    {
+      "epoch": 11.346968139773896,
+      "grad_norm": 0.03745291754603386,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 110406
+    },
+    {
+      "epoch": 11.347276464542652,
+      "grad_norm": 0.04565132409334183,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 110409
+    },
+    {
+      "epoch": 11.347584789311409,
+      "grad_norm": 0.06974771618843079,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 110412
+    },
+    {
+      "epoch": 11.347893114080165,
+      "grad_norm": 0.1146189272403717,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 110415
+    },
+    {
+      "epoch": 11.348201438848921,
+      "grad_norm": 0.1235949769616127,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 110418
+    },
+    {
+      "epoch": 11.348509763617677,
+      "grad_norm": 0.08458071202039719,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 110421
+    },
+    {
+      "epoch": 11.348818088386434,
+      "grad_norm": 0.07701151072978973,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 110424
+    },
+    {
+      "epoch": 11.34912641315519,
+      "grad_norm": 0.07502985745668411,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 110427
+    },
+    {
+      "epoch": 11.349434737923946,
+      "grad_norm": 0.06750281900167465,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 110430
+    },
+    {
+      "epoch": 11.349743062692703,
+      "grad_norm": 0.04019710421562195,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 110433
+    },
+    {
+      "epoch": 11.350051387461459,
+      "grad_norm": 0.03724076971411705,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 110436
+    },
+    {
+      "epoch": 11.350359712230215,
+      "grad_norm": 0.07026968896389008,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 110439
+    },
+    {
+      "epoch": 11.350668036998972,
+      "grad_norm": 0.09067178517580032,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 110442
+    },
+    {
+      "epoch": 11.350976361767728,
+      "grad_norm": 0.07308315485715866,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 110445
+    },
+    {
+      "epoch": 11.351284686536484,
+      "grad_norm": 0.06199457123875618,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 110448
+    },
+    {
+      "epoch": 11.351593011305242,
+      "grad_norm": 0.037689439952373505,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 110451
+    },
+    {
+      "epoch": 11.351901336073999,
+      "grad_norm": 0.033859074115753174,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 110454
+    },
+    {
+      "epoch": 11.352209660842755,
+      "grad_norm": 0.03455094248056412,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 110457
+    },
+    {
+      "epoch": 11.352517985611511,
+      "grad_norm": 0.16149871051311493,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 110460
+    },
+    {
+      "epoch": 11.352826310380268,
+      "grad_norm": 0.04946501553058624,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 110463
+    },
+    {
+      "epoch": 11.353134635149024,
+      "grad_norm": 0.080601267516613,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 110466
+    },
+    {
+      "epoch": 11.35344295991778,
+      "grad_norm": 0.04147493094205856,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 110469
+    },
+    {
+      "epoch": 11.353751284686536,
+      "grad_norm": 0.07370273768901825,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 110472
+    },
+    {
+      "epoch": 11.354059609455293,
+      "grad_norm": 0.04175065830349922,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 110475
+    },
+    {
+      "epoch": 11.354367934224049,
+      "grad_norm": 0.08143223077058792,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 110478
+    },
+    {
+      "epoch": 11.354676258992805,
+      "grad_norm": 0.08016743510961533,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 110481
+    },
+    {
+      "epoch": 11.354984583761562,
+      "grad_norm": 0.06962796300649643,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 110484
+    },
+    {
+      "epoch": 11.355292908530318,
+      "grad_norm": 0.09174045920372009,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 110487
+    },
+    {
+      "epoch": 11.355601233299074,
+      "grad_norm": 0.046844564378261566,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 110490
+    },
+    {
+      "epoch": 11.35590955806783,
+      "grad_norm": 0.08431335538625717,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 110493
+    },
+    {
+      "epoch": 11.356217882836589,
+      "grad_norm": 0.052417635917663574,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 110496
+    },
+    {
+      "epoch": 11.356526207605345,
+      "grad_norm": 0.10628152638673782,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 110499
+    },
+    {
+      "epoch": 11.356834532374101,
+      "grad_norm": 0.0665721446275711,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 110502
+    },
+    {
+      "epoch": 11.357142857142858,
+      "grad_norm": 0.06576160341501236,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 110505
+    },
+    {
+      "epoch": 11.357451181911614,
+      "grad_norm": 0.07959224283695221,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 110508
+    },
+    {
+      "epoch": 11.35775950668037,
+      "grad_norm": 0.09703835844993591,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 110511
+    },
+    {
+      "epoch": 11.358067831449127,
+      "grad_norm": 0.0768088549375534,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 110514
+    },
+    {
+      "epoch": 11.358376156217883,
+      "grad_norm": 0.06530007719993591,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 110517
+    },
+    {
+      "epoch": 11.35868448098664,
+      "grad_norm": 0.06526839733123779,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 110520
+    },
+    {
+      "epoch": 11.358992805755395,
+      "grad_norm": 0.06234097480773926,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 110523
+    },
+    {
+      "epoch": 11.359301130524152,
+      "grad_norm": 0.033938802778720856,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 110526
+    },
+    {
+      "epoch": 11.359609455292908,
+      "grad_norm": 0.04808393120765686,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 110529
+    },
+    {
+      "epoch": 11.359917780061664,
+      "grad_norm": 0.09899698942899704,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 110532
+    },
+    {
+      "epoch": 11.36022610483042,
+      "grad_norm": 0.044449757784605026,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 110535
+    },
+    {
+      "epoch": 11.360534429599177,
+      "grad_norm": 0.0801699161529541,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 110538
+    },
+    {
+      "epoch": 11.360842754367933,
+      "grad_norm": 0.0521024726331234,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 110541
+    },
+    {
+      "epoch": 11.361151079136691,
+      "grad_norm": 0.062242165207862854,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 110544
+    },
+    {
+      "epoch": 11.361459403905448,
+      "grad_norm": 0.06852758675813675,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 110547
+    },
+    {
+      "epoch": 11.361767728674204,
+      "grad_norm": 0.06884782016277313,
+      "learning_rate": 0.01,
+      "loss": 1.9753,
+      "step": 110550
+    },
+    {
+      "epoch": 11.36207605344296,
+      "grad_norm": 0.05993012338876724,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 110553
+    },
+    {
+      "epoch": 11.362384378211717,
+      "grad_norm": 0.057871006429195404,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 110556
+    },
+    {
+      "epoch": 11.362692702980473,
+      "grad_norm": 0.042249929159879684,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 110559
+    },
+    {
+      "epoch": 11.36300102774923,
+      "grad_norm": 0.05203591659665108,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 110562
+    },
+    {
+      "epoch": 11.363309352517986,
+      "grad_norm": 0.037728067487478256,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 110565
+    },
+    {
+      "epoch": 11.363617677286742,
+      "grad_norm": 0.0518614761531353,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 110568
+    },
+    {
+      "epoch": 11.363926002055498,
+      "grad_norm": 0.10485280305147171,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 110571
+    },
+    {
+      "epoch": 11.364234326824254,
+      "grad_norm": 0.07373513281345367,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 110574
+    },
+    {
+      "epoch": 11.36454265159301,
+      "grad_norm": 0.1314626783132553,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 110577
+    },
+    {
+      "epoch": 11.364850976361767,
+      "grad_norm": 0.13870415091514587,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 110580
+    },
+    {
+      "epoch": 11.365159301130523,
+      "grad_norm": 0.05415605381131172,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 110583
+    },
+    {
+      "epoch": 11.36546762589928,
+      "grad_norm": 0.03968094661831856,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 110586
+    },
+    {
+      "epoch": 11.365775950668038,
+      "grad_norm": 0.032541971653699875,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 110589
+    },
+    {
+      "epoch": 11.366084275436794,
+      "grad_norm": 0.07528942078351974,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 110592
+    },
+    {
+      "epoch": 11.36639260020555,
+      "grad_norm": 0.17362824082374573,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 110595
+    },
+    {
+      "epoch": 11.366700924974307,
+      "grad_norm": 0.13993510603904724,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 110598
+    },
+    {
+      "epoch": 11.367009249743063,
+      "grad_norm": 0.05224933475255966,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 110601
+    },
+    {
+      "epoch": 11.36731757451182,
+      "grad_norm": 0.07095824182033539,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 110604
+    },
+    {
+      "epoch": 11.367625899280576,
+      "grad_norm": 0.04492488503456116,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 110607
+    },
+    {
+      "epoch": 11.367934224049332,
+      "grad_norm": 0.06791716068983078,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 110610
+    },
+    {
+      "epoch": 11.368242548818088,
+      "grad_norm": 0.05190029740333557,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 110613
+    },
+    {
+      "epoch": 11.368550873586845,
+      "grad_norm": 0.11406069248914719,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 110616
+    },
+    {
+      "epoch": 11.3688591983556,
+      "grad_norm": 0.13801059126853943,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 110619
+    },
+    {
+      "epoch": 11.369167523124357,
+      "grad_norm": 0.09343323111534119,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 110622
+    },
+    {
+      "epoch": 11.369475847893113,
+      "grad_norm": 0.06386526674032211,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 110625
+    },
+    {
+      "epoch": 11.36978417266187,
+      "grad_norm": 0.03324955701828003,
+      "learning_rate": 0.01,
+      "loss": 1.966,
+      "step": 110628
+    },
+    {
+      "epoch": 11.370092497430626,
+      "grad_norm": 0.043516796082258224,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 110631
+    },
+    {
+      "epoch": 11.370400822199384,
+      "grad_norm": 0.11557954549789429,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 110634
+    },
+    {
+      "epoch": 11.37070914696814,
+      "grad_norm": 0.0785699188709259,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 110637
+    },
+    {
+      "epoch": 11.371017471736897,
+      "grad_norm": 0.08156263828277588,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 110640
+    },
+    {
+      "epoch": 11.371325796505653,
+      "grad_norm": 0.053373608738183975,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 110643
+    },
+    {
+      "epoch": 11.37163412127441,
+      "grad_norm": 0.06310009956359863,
+      "learning_rate": 0.01,
+      "loss": 1.9011,
+      "step": 110646
+    },
+    {
+      "epoch": 11.371942446043166,
+      "grad_norm": 0.10197494179010391,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 110649
+    },
+    {
+      "epoch": 11.372250770811922,
+      "grad_norm": 0.13593943417072296,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 110652
+    },
+    {
+      "epoch": 11.372559095580678,
+      "grad_norm": 0.04047724977135658,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 110655
+    },
+    {
+      "epoch": 11.372867420349435,
+      "grad_norm": 0.06438061594963074,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 110658
+    },
+    {
+      "epoch": 11.373175745118191,
+      "grad_norm": 0.07296453416347504,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 110661
+    },
+    {
+      "epoch": 11.373484069886947,
+      "grad_norm": 0.08513321727514267,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 110664
+    },
+    {
+      "epoch": 11.373792394655704,
+      "grad_norm": 0.13206641376018524,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 110667
+    },
+    {
+      "epoch": 11.37410071942446,
+      "grad_norm": 0.04452472925186157,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 110670
+    },
+    {
+      "epoch": 11.374409044193216,
+      "grad_norm": 0.05014786124229431,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 110673
+    },
+    {
+      "epoch": 11.374717368961972,
+      "grad_norm": 0.039705969393253326,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 110676
+    },
+    {
+      "epoch": 11.37502569373073,
+      "grad_norm": 0.04527021944522858,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 110679
+    },
+    {
+      "epoch": 11.375334018499487,
+      "grad_norm": 0.06424544751644135,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 110682
+    },
+    {
+      "epoch": 11.375642343268243,
+      "grad_norm": 0.038035839796066284,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 110685
+    },
+    {
+      "epoch": 11.375950668037,
+      "grad_norm": 0.043364040553569794,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 110688
+    },
+    {
+      "epoch": 11.376258992805756,
+      "grad_norm": 0.036017462611198425,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 110691
+    },
+    {
+      "epoch": 11.376567317574512,
+      "grad_norm": 0.057449888437986374,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 110694
+    },
+    {
+      "epoch": 11.376875642343268,
+      "grad_norm": 0.18690413236618042,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 110697
+    },
+    {
+      "epoch": 11.377183967112025,
+      "grad_norm": 0.18334275484085083,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 110700
+    },
+    {
+      "epoch": 11.377492291880781,
+      "grad_norm": 0.11479616165161133,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 110703
+    },
+    {
+      "epoch": 11.377800616649537,
+      "grad_norm": 0.08656278252601624,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 110706
+    },
+    {
+      "epoch": 11.378108941418294,
+      "grad_norm": 0.04703566059470177,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 110709
+    },
+    {
+      "epoch": 11.37841726618705,
+      "grad_norm": 0.06349678337574005,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 110712
+    },
+    {
+      "epoch": 11.378725590955806,
+      "grad_norm": 0.10881751030683517,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 110715
+    },
+    {
+      "epoch": 11.379033915724563,
+      "grad_norm": 0.05449318140745163,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 110718
+    },
+    {
+      "epoch": 11.379342240493319,
+      "grad_norm": 0.1894116997718811,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 110721
+    },
+    {
+      "epoch": 11.379650565262075,
+      "grad_norm": 0.049433499574661255,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 110724
+    },
+    {
+      "epoch": 11.379958890030833,
+      "grad_norm": 0.03821530565619469,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 110727
+    },
+    {
+      "epoch": 11.38026721479959,
+      "grad_norm": 0.034824103116989136,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 110730
+    },
+    {
+      "epoch": 11.380575539568346,
+      "grad_norm": 0.05202208086848259,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 110733
+    },
+    {
+      "epoch": 11.380883864337102,
+      "grad_norm": 0.057843755930662155,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 110736
+    },
+    {
+      "epoch": 11.381192189105859,
+      "grad_norm": 0.04548819735646248,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 110739
+    },
+    {
+      "epoch": 11.381500513874615,
+      "grad_norm": 0.08566679805517197,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 110742
+    },
+    {
+      "epoch": 11.381808838643371,
+      "grad_norm": 0.11440742015838623,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 110745
+    },
+    {
+      "epoch": 11.382117163412127,
+      "grad_norm": 0.15535084903240204,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 110748
+    },
+    {
+      "epoch": 11.382425488180884,
+      "grad_norm": 0.0787283256649971,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 110751
+    },
+    {
+      "epoch": 11.38273381294964,
+      "grad_norm": 0.031954240053892136,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 110754
+    },
+    {
+      "epoch": 11.383042137718396,
+      "grad_norm": 0.03369603678584099,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 110757
+    },
+    {
+      "epoch": 11.383350462487153,
+      "grad_norm": 0.08464602380990982,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 110760
+    },
+    {
+      "epoch": 11.383658787255909,
+      "grad_norm": 0.046058326959609985,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 110763
+    },
+    {
+      "epoch": 11.383967112024665,
+      "grad_norm": 0.038871098309755325,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 110766
+    },
+    {
+      "epoch": 11.384275436793422,
+      "grad_norm": 0.040099188685417175,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 110769
+    },
+    {
+      "epoch": 11.38458376156218,
+      "grad_norm": 0.047709397971630096,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 110772
+    },
+    {
+      "epoch": 11.384892086330936,
+      "grad_norm": 0.04168373718857765,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 110775
+    },
+    {
+      "epoch": 11.385200411099692,
+      "grad_norm": 0.15479597449302673,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 110778
+    },
+    {
+      "epoch": 11.385508735868449,
+      "grad_norm": 0.09209232032299042,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 110781
+    },
+    {
+      "epoch": 11.385817060637205,
+      "grad_norm": 0.048776429146528244,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 110784
+    },
+    {
+      "epoch": 11.386125385405961,
+      "grad_norm": 0.04659121483564377,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 110787
+    },
+    {
+      "epoch": 11.386433710174718,
+      "grad_norm": 0.05104237422347069,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 110790
+    },
+    {
+      "epoch": 11.386742034943474,
+      "grad_norm": 0.06261378526687622,
+      "learning_rate": 0.01,
+      "loss": 1.9062,
+      "step": 110793
+    },
+    {
+      "epoch": 11.38705035971223,
+      "grad_norm": 0.06825476884841919,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 110796
+    },
+    {
+      "epoch": 11.387358684480986,
+      "grad_norm": 0.06235522776842117,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 110799
+    },
+    {
+      "epoch": 11.387667009249743,
+      "grad_norm": 0.1005907878279686,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 110802
+    },
+    {
+      "epoch": 11.387975334018499,
+      "grad_norm": 0.12915214896202087,
+      "learning_rate": 0.01,
+      "loss": 1.9083,
+      "step": 110805
+    },
+    {
+      "epoch": 11.388283658787255,
+      "grad_norm": 0.10919107496738434,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 110808
+    },
+    {
+      "epoch": 11.388591983556012,
+      "grad_norm": 0.05322147533297539,
+      "learning_rate": 0.01,
+      "loss": 1.9761,
+      "step": 110811
+    },
+    {
+      "epoch": 11.388900308324768,
+      "grad_norm": 0.04587923735380173,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 110814
+    },
+    {
+      "epoch": 11.389208633093524,
+      "grad_norm": 0.035594645887613297,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 110817
+    },
+    {
+      "epoch": 11.389516957862282,
+      "grad_norm": 0.03522982820868492,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 110820
+    },
+    {
+      "epoch": 11.389825282631039,
+      "grad_norm": 0.04372236505150795,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 110823
+    },
+    {
+      "epoch": 11.390133607399795,
+      "grad_norm": 0.05270487442612648,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 110826
+    },
+    {
+      "epoch": 11.390441932168551,
+      "grad_norm": 0.12330029904842377,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 110829
+    },
+    {
+      "epoch": 11.390750256937308,
+      "grad_norm": 0.07215101271867752,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 110832
+    },
+    {
+      "epoch": 11.391058581706064,
+      "grad_norm": 0.056661464273929596,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 110835
+    },
+    {
+      "epoch": 11.39136690647482,
+      "grad_norm": 0.10824533551931381,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 110838
+    },
+    {
+      "epoch": 11.391675231243577,
+      "grad_norm": 0.05006413906812668,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 110841
+    },
+    {
+      "epoch": 11.391983556012333,
+      "grad_norm": 0.036600951105356216,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 110844
+    },
+    {
+      "epoch": 11.39229188078109,
+      "grad_norm": 0.06521067023277283,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 110847
+    },
+    {
+      "epoch": 11.392600205549845,
+      "grad_norm": 0.1014748141169548,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 110850
+    },
+    {
+      "epoch": 11.392908530318602,
+      "grad_norm": 0.1848376840353012,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 110853
+    },
+    {
+      "epoch": 11.393216855087358,
+      "grad_norm": 0.09469456970691681,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 110856
+    },
+    {
+      "epoch": 11.393525179856114,
+      "grad_norm": 0.09343547374010086,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 110859
+    },
+    {
+      "epoch": 11.393833504624872,
+      "grad_norm": 0.08093136548995972,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 110862
+    },
+    {
+      "epoch": 11.394141829393629,
+      "grad_norm": 0.05411363020539284,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 110865
+    },
+    {
+      "epoch": 11.394450154162385,
+      "grad_norm": 0.047125186771154404,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 110868
+    },
+    {
+      "epoch": 11.394758478931141,
+      "grad_norm": 0.0919274240732193,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 110871
+    },
+    {
+      "epoch": 11.395066803699898,
+      "grad_norm": 0.042707886546850204,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 110874
+    },
+    {
+      "epoch": 11.395375128468654,
+      "grad_norm": 0.08432315289974213,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 110877
+    },
+    {
+      "epoch": 11.39568345323741,
+      "grad_norm": 0.06510323286056519,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 110880
+    },
+    {
+      "epoch": 11.395991778006167,
+      "grad_norm": 0.046135447919368744,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 110883
+    },
+    {
+      "epoch": 11.396300102774923,
+      "grad_norm": 0.10874036699533463,
+      "learning_rate": 0.01,
+      "loss": 1.9633,
+      "step": 110886
+    },
+    {
+      "epoch": 11.39660842754368,
+      "grad_norm": 0.05693401023745537,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 110889
+    },
+    {
+      "epoch": 11.396916752312436,
+      "grad_norm": 0.10287602990865707,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 110892
+    },
+    {
+      "epoch": 11.397225077081192,
+      "grad_norm": 0.14678283035755157,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 110895
+    },
+    {
+      "epoch": 11.397533401849948,
+      "grad_norm": 0.06007491424679756,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 110898
+    },
+    {
+      "epoch": 11.397841726618704,
+      "grad_norm": 0.0790441483259201,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 110901
+    },
+    {
+      "epoch": 11.39815005138746,
+      "grad_norm": 0.07017059624195099,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 110904
+    },
+    {
+      "epoch": 11.398458376156217,
+      "grad_norm": 0.0407567135989666,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 110907
+    },
+    {
+      "epoch": 11.398766700924975,
+      "grad_norm": 0.0886857658624649,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 110910
+    },
+    {
+      "epoch": 11.399075025693731,
+      "grad_norm": 0.06689956784248352,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 110913
+    },
+    {
+      "epoch": 11.399383350462488,
+      "grad_norm": 0.10107925534248352,
+      "learning_rate": 0.01,
+      "loss": 1.9757,
+      "step": 110916
+    },
+    {
+      "epoch": 11.399691675231244,
+      "grad_norm": 0.05348954722285271,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 110919
+    },
+    {
+      "epoch": 11.4,
+      "grad_norm": 0.0453205369412899,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 110922
+    },
+    {
+      "epoch": 11.400308324768757,
+      "grad_norm": 0.037814341485500336,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 110925
+    },
+    {
+      "epoch": 11.400616649537513,
+      "grad_norm": 0.11961714178323746,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 110928
+    },
+    {
+      "epoch": 11.40092497430627,
+      "grad_norm": 0.09395978599786758,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 110931
+    },
+    {
+      "epoch": 11.401233299075026,
+      "grad_norm": 0.06540749222040176,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 110934
+    },
+    {
+      "epoch": 11.401541623843782,
+      "grad_norm": 0.08541125804185867,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 110937
+    },
+    {
+      "epoch": 11.401849948612538,
+      "grad_norm": 0.036478616297245026,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 110940
+    },
+    {
+      "epoch": 11.402158273381295,
+      "grad_norm": 0.0369294211268425,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 110943
+    },
+    {
+      "epoch": 11.40246659815005,
+      "grad_norm": 0.13193678855895996,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 110946
+    },
+    {
+      "epoch": 11.402774922918807,
+      "grad_norm": 0.05666269734501839,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 110949
+    },
+    {
+      "epoch": 11.403083247687563,
+      "grad_norm": 0.04516861215233803,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 110952
+    },
+    {
+      "epoch": 11.403391572456322,
+      "grad_norm": 0.05796361342072487,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 110955
+    },
+    {
+      "epoch": 11.403699897225078,
+      "grad_norm": 0.04144136980175972,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 110958
+    },
+    {
+      "epoch": 11.404008221993834,
+      "grad_norm": 0.04193937033414841,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 110961
+    },
+    {
+      "epoch": 11.40431654676259,
+      "grad_norm": 0.1043611392378807,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 110964
+    },
+    {
+      "epoch": 11.404624871531347,
+      "grad_norm": 0.1311003863811493,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 110967
+    },
+    {
+      "epoch": 11.404933196300103,
+      "grad_norm": 0.05601399391889572,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 110970
+    },
+    {
+      "epoch": 11.40524152106886,
+      "grad_norm": 0.04909517988562584,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 110973
+    },
+    {
+      "epoch": 11.405549845837616,
+      "grad_norm": 0.08073060214519501,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 110976
+    },
+    {
+      "epoch": 11.405858170606372,
+      "grad_norm": 0.07823039591312408,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 110979
+    },
+    {
+      "epoch": 11.406166495375128,
+      "grad_norm": 0.05647626519203186,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 110982
+    },
+    {
+      "epoch": 11.406474820143885,
+      "grad_norm": 0.1711927354335785,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 110985
+    },
+    {
+      "epoch": 11.406783144912641,
+      "grad_norm": 0.1063016802072525,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 110988
+    },
+    {
+      "epoch": 11.407091469681397,
+      "grad_norm": 0.03903825581073761,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 110991
+    },
+    {
+      "epoch": 11.407399794450154,
+      "grad_norm": 0.040212482213974,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 110994
+    },
+    {
+      "epoch": 11.40770811921891,
+      "grad_norm": 0.05461808294057846,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 110997
+    },
+    {
+      "epoch": 11.408016443987666,
+      "grad_norm": 0.05734023079276085,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 111000
+    },
+    {
+      "epoch": 11.408324768756424,
+      "grad_norm": 0.06303809583187103,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 111003
+    },
+    {
+      "epoch": 11.40863309352518,
+      "grad_norm": 0.1410919576883316,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 111006
+    },
+    {
+      "epoch": 11.408941418293937,
+      "grad_norm": 0.07548557221889496,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 111009
+    },
+    {
+      "epoch": 11.409249743062693,
+      "grad_norm": 0.07225187122821808,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 111012
+    },
+    {
+      "epoch": 11.40955806783145,
+      "grad_norm": 0.05371328815817833,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 111015
+    },
+    {
+      "epoch": 11.409866392600206,
+      "grad_norm": 0.10149164497852325,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 111018
+    },
+    {
+      "epoch": 11.410174717368962,
+      "grad_norm": 0.07427545636892319,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 111021
+    },
+    {
+      "epoch": 11.410483042137718,
+      "grad_norm": 0.07735265046358109,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 111024
+    },
+    {
+      "epoch": 11.410791366906475,
+      "grad_norm": 0.10191336274147034,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 111027
+    },
+    {
+      "epoch": 11.411099691675231,
+      "grad_norm": 0.15336443483829498,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 111030
+    },
+    {
+      "epoch": 11.411408016443987,
+      "grad_norm": 0.09829924255609512,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 111033
+    },
+    {
+      "epoch": 11.411716341212744,
+      "grad_norm": 0.07386139780282974,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 111036
+    },
+    {
+      "epoch": 11.4120246659815,
+      "grad_norm": 0.04841221868991852,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 111039
+    },
+    {
+      "epoch": 11.412332990750256,
+      "grad_norm": 0.045958053320646286,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 111042
+    },
+    {
+      "epoch": 11.412641315519013,
+      "grad_norm": 0.09872113913297653,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 111045
+    },
+    {
+      "epoch": 11.41294964028777,
+      "grad_norm": 0.0739109143614769,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 111048
+    },
+    {
+      "epoch": 11.413257965056527,
+      "grad_norm": 0.09770570695400238,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 111051
+    },
+    {
+      "epoch": 11.413566289825283,
+      "grad_norm": 0.1025087758898735,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 111054
+    },
+    {
+      "epoch": 11.41387461459404,
+      "grad_norm": 0.07407236844301224,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 111057
+    },
+    {
+      "epoch": 11.414182939362796,
+      "grad_norm": 0.12390477955341339,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 111060
+    },
+    {
+      "epoch": 11.414491264131552,
+      "grad_norm": 0.1367517113685608,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 111063
+    },
+    {
+      "epoch": 11.414799588900308,
+      "grad_norm": 0.06094538792967796,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 111066
+    },
+    {
+      "epoch": 11.415107913669065,
+      "grad_norm": 0.04376324266195297,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 111069
+    },
+    {
+      "epoch": 11.415416238437821,
+      "grad_norm": 0.057995304465293884,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 111072
+    },
+    {
+      "epoch": 11.415724563206577,
+      "grad_norm": 0.04961060732603073,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 111075
+    },
+    {
+      "epoch": 11.416032887975334,
+      "grad_norm": 0.04857701435685158,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 111078
+    },
+    {
+      "epoch": 11.41634121274409,
+      "grad_norm": 0.0609593540430069,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 111081
+    },
+    {
+      "epoch": 11.416649537512846,
+      "grad_norm": 0.05478176474571228,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 111084
+    },
+    {
+      "epoch": 11.416957862281603,
+      "grad_norm": 0.04392441362142563,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 111087
+    },
+    {
+      "epoch": 11.417266187050359,
+      "grad_norm": 0.08224606513977051,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 111090
+    },
+    {
+      "epoch": 11.417574511819117,
+      "grad_norm": 0.06009015068411827,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 111093
+    },
+    {
+      "epoch": 11.417882836587873,
+      "grad_norm": 0.03522694483399391,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 111096
+    },
+    {
+      "epoch": 11.41819116135663,
+      "grad_norm": 0.08625969290733337,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 111099
+    },
+    {
+      "epoch": 11.418499486125386,
+      "grad_norm": 0.0723060742020607,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 111102
+    },
+    {
+      "epoch": 11.418807810894142,
+      "grad_norm": 0.07083339244127274,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 111105
+    },
+    {
+      "epoch": 11.419116135662899,
+      "grad_norm": 0.05434579402208328,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 111108
+    },
+    {
+      "epoch": 11.419424460431655,
+      "grad_norm": 0.05564429983496666,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 111111
+    },
+    {
+      "epoch": 11.419732785200411,
+      "grad_norm": 0.06538271903991699,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 111114
+    },
+    {
+      "epoch": 11.420041109969167,
+      "grad_norm": 0.12349273264408112,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 111117
+    },
+    {
+      "epoch": 11.420349434737924,
+      "grad_norm": 0.11483631283044815,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 111120
+    },
+    {
+      "epoch": 11.42065775950668,
+      "grad_norm": 0.045582689344882965,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 111123
+    },
+    {
+      "epoch": 11.420966084275436,
+      "grad_norm": 0.051394637674093246,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 111126
+    },
+    {
+      "epoch": 11.421274409044193,
+      "grad_norm": 0.058152392506599426,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 111129
+    },
+    {
+      "epoch": 11.421582733812949,
+      "grad_norm": 0.08196362853050232,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 111132
+    },
+    {
+      "epoch": 11.421891058581705,
+      "grad_norm": 0.0730375200510025,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 111135
+    },
+    {
+      "epoch": 11.422199383350463,
+      "grad_norm": 0.07399123162031174,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 111138
+    },
+    {
+      "epoch": 11.42250770811922,
+      "grad_norm": 0.09269146621227264,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 111141
+    },
+    {
+      "epoch": 11.422816032887976,
+      "grad_norm": 0.06923570483922958,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 111144
+    },
+    {
+      "epoch": 11.423124357656732,
+      "grad_norm": 0.06779024749994278,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 111147
+    },
+    {
+      "epoch": 11.423432682425489,
+      "grad_norm": 0.07309120148420334,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 111150
+    },
+    {
+      "epoch": 11.423741007194245,
+      "grad_norm": 0.05872660502791405,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 111153
+    },
+    {
+      "epoch": 11.424049331963001,
+      "grad_norm": 0.073930524289608,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 111156
+    },
+    {
+      "epoch": 11.424357656731758,
+      "grad_norm": 0.1007046103477478,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 111159
+    },
+    {
+      "epoch": 11.424665981500514,
+      "grad_norm": 0.03730003908276558,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 111162
+    },
+    {
+      "epoch": 11.42497430626927,
+      "grad_norm": 0.17431728541851044,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 111165
+    },
+    {
+      "epoch": 11.425282631038026,
+      "grad_norm": 0.17289651930332184,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 111168
+    },
+    {
+      "epoch": 11.425590955806783,
+      "grad_norm": 0.12980881333351135,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 111171
+    },
+    {
+      "epoch": 11.425899280575539,
+      "grad_norm": 0.06733021885156631,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 111174
+    },
+    {
+      "epoch": 11.426207605344295,
+      "grad_norm": 0.04424450546503067,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 111177
+    },
+    {
+      "epoch": 11.426515930113052,
+      "grad_norm": 0.043618347495794296,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 111180
+    },
+    {
+      "epoch": 11.426824254881808,
+      "grad_norm": 0.06406936794519424,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 111183
+    },
+    {
+      "epoch": 11.427132579650566,
+      "grad_norm": 0.06397773325443268,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 111186
+    },
+    {
+      "epoch": 11.427440904419322,
+      "grad_norm": 0.045785196125507355,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 111189
+    },
+    {
+      "epoch": 11.427749229188079,
+      "grad_norm": 0.036619219928979874,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 111192
+    },
+    {
+      "epoch": 11.428057553956835,
+      "grad_norm": 0.056804507970809937,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 111195
+    },
+    {
+      "epoch": 11.428365878725591,
+      "grad_norm": 0.035350654274225235,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 111198
+    },
+    {
+      "epoch": 11.428674203494348,
+      "grad_norm": 0.09453041106462479,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 111201
+    },
+    {
+      "epoch": 11.428982528263104,
+      "grad_norm": 0.10754123330116272,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 111204
+    },
+    {
+      "epoch": 11.42929085303186,
+      "grad_norm": 0.140550434589386,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 111207
+    },
+    {
+      "epoch": 11.429599177800617,
+      "grad_norm": 0.05969111621379852,
+      "learning_rate": 0.01,
+      "loss": 1.9739,
+      "step": 111210
+    },
+    {
+      "epoch": 11.429907502569373,
+      "grad_norm": 0.07596472650766373,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 111213
+    },
+    {
+      "epoch": 11.43021582733813,
+      "grad_norm": 0.06707892566919327,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 111216
+    },
+    {
+      "epoch": 11.430524152106885,
+      "grad_norm": 0.031418804079294205,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 111219
+    },
+    {
+      "epoch": 11.430832476875642,
+      "grad_norm": 0.11432456225156784,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 111222
+    },
+    {
+      "epoch": 11.431140801644398,
+      "grad_norm": 0.06755537539720535,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 111225
+    },
+    {
+      "epoch": 11.431449126413154,
+      "grad_norm": 0.09963531047105789,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 111228
+    },
+    {
+      "epoch": 11.431757451181912,
+      "grad_norm": 0.14981499314308167,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 111231
+    },
+    {
+      "epoch": 11.432065775950669,
+      "grad_norm": 0.0813409760594368,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 111234
+    },
+    {
+      "epoch": 11.432374100719425,
+      "grad_norm": 0.10392513126134872,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 111237
+    },
+    {
+      "epoch": 11.432682425488181,
+      "grad_norm": 0.05986598879098892,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 111240
+    },
+    {
+      "epoch": 11.432990750256938,
+      "grad_norm": 0.11344392597675323,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 111243
+    },
+    {
+      "epoch": 11.433299075025694,
+      "grad_norm": 0.07161758095026016,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 111246
+    },
+    {
+      "epoch": 11.43360739979445,
+      "grad_norm": 0.0733150988817215,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 111249
+    },
+    {
+      "epoch": 11.433915724563207,
+      "grad_norm": 0.08997683227062225,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 111252
+    },
+    {
+      "epoch": 11.434224049331963,
+      "grad_norm": 0.042780738323926926,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 111255
+    },
+    {
+      "epoch": 11.43453237410072,
+      "grad_norm": 0.055852118879556656,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 111258
+    },
+    {
+      "epoch": 11.434840698869476,
+      "grad_norm": 0.05158184468746185,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 111261
+    },
+    {
+      "epoch": 11.435149023638232,
+      "grad_norm": 0.04345036670565605,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 111264
+    },
+    {
+      "epoch": 11.435457348406988,
+      "grad_norm": 0.03345207870006561,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 111267
+    },
+    {
+      "epoch": 11.435765673175744,
+      "grad_norm": 0.055717889219522476,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 111270
+    },
+    {
+      "epoch": 11.4360739979445,
+      "grad_norm": 0.13665983080863953,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 111273
+    },
+    {
+      "epoch": 11.436382322713257,
+      "grad_norm": 0.14494851231575012,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 111276
+    },
+    {
+      "epoch": 11.436690647482015,
+      "grad_norm": 0.07634695619344711,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 111279
+    },
+    {
+      "epoch": 11.436998972250771,
+      "grad_norm": 0.03860051929950714,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 111282
+    },
+    {
+      "epoch": 11.437307297019528,
+      "grad_norm": 0.0648890882730484,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 111285
+    },
+    {
+      "epoch": 11.437615621788284,
+      "grad_norm": 0.04066958278417587,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 111288
+    },
+    {
+      "epoch": 11.43792394655704,
+      "grad_norm": 0.03933463245630264,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 111291
+    },
+    {
+      "epoch": 11.438232271325797,
+      "grad_norm": 0.06783044338226318,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 111294
+    },
+    {
+      "epoch": 11.438540596094553,
+      "grad_norm": 0.09267598390579224,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 111297
+    },
+    {
+      "epoch": 11.43884892086331,
+      "grad_norm": 0.0766342505812645,
+      "learning_rate": 0.01,
+      "loss": 1.9783,
+      "step": 111300
+    },
+    {
+      "epoch": 11.439157245632066,
+      "grad_norm": 0.07145971804857254,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 111303
+    },
+    {
+      "epoch": 11.439465570400822,
+      "grad_norm": 0.0847540870308876,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 111306
+    },
+    {
+      "epoch": 11.439773895169578,
+      "grad_norm": 0.04288563132286072,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 111309
+    },
+    {
+      "epoch": 11.440082219938335,
+      "grad_norm": 0.03836122527718544,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 111312
+    },
+    {
+      "epoch": 11.44039054470709,
+      "grad_norm": 0.03649865835905075,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 111315
+    },
+    {
+      "epoch": 11.440698869475847,
+      "grad_norm": 0.0543823167681694,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 111318
+    },
+    {
+      "epoch": 11.441007194244605,
+      "grad_norm": 0.04194122180342674,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 111321
+    },
+    {
+      "epoch": 11.441315519013362,
+      "grad_norm": 0.16252778470516205,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 111324
+    },
+    {
+      "epoch": 11.441623843782118,
+      "grad_norm": 0.05408162996172905,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 111327
+    },
+    {
+      "epoch": 11.441932168550874,
+      "grad_norm": 0.04746748507022858,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 111330
+    },
+    {
+      "epoch": 11.44224049331963,
+      "grad_norm": 0.044511113315820694,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 111333
+    },
+    {
+      "epoch": 11.442548818088387,
+      "grad_norm": 0.0402715727686882,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 111336
+    },
+    {
+      "epoch": 11.442857142857143,
+      "grad_norm": 0.04611997306346893,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 111339
+    },
+    {
+      "epoch": 11.4431654676259,
+      "grad_norm": 0.05174090340733528,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 111342
+    },
+    {
+      "epoch": 11.443473792394656,
+      "grad_norm": 0.06841269880533218,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 111345
+    },
+    {
+      "epoch": 11.443782117163412,
+      "grad_norm": 0.0667579174041748,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 111348
+    },
+    {
+      "epoch": 11.444090441932168,
+      "grad_norm": 0.06422604620456696,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 111351
+    },
+    {
+      "epoch": 11.444398766700925,
+      "grad_norm": 0.05790428817272186,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 111354
+    },
+    {
+      "epoch": 11.444707091469681,
+      "grad_norm": 0.060070376843214035,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 111357
+    },
+    {
+      "epoch": 11.445015416238437,
+      "grad_norm": 0.084757000207901,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 111360
+    },
+    {
+      "epoch": 11.445323741007194,
+      "grad_norm": 0.08789457380771637,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 111363
+    },
+    {
+      "epoch": 11.44563206577595,
+      "grad_norm": 0.10299545526504517,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 111366
+    },
+    {
+      "epoch": 11.445940390544708,
+      "grad_norm": 0.05760720372200012,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 111369
+    },
+    {
+      "epoch": 11.446248715313464,
+      "grad_norm": 0.08394305408000946,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 111372
+    },
+    {
+      "epoch": 11.44655704008222,
+      "grad_norm": 0.1427774280309677,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 111375
+    },
+    {
+      "epoch": 11.446865364850977,
+      "grad_norm": 0.11516613513231277,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 111378
+    },
+    {
+      "epoch": 11.447173689619733,
+      "grad_norm": 0.06715381145477295,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 111381
+    },
+    {
+      "epoch": 11.44748201438849,
+      "grad_norm": 0.04250967502593994,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 111384
+    },
+    {
+      "epoch": 11.447790339157246,
+      "grad_norm": 0.1086759939789772,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 111387
+    },
+    {
+      "epoch": 11.448098663926002,
+      "grad_norm": 0.055994849652051926,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 111390
+    },
+    {
+      "epoch": 11.448406988694758,
+      "grad_norm": 0.03699815273284912,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 111393
+    },
+    {
+      "epoch": 11.448715313463515,
+      "grad_norm": 0.05926622822880745,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 111396
+    },
+    {
+      "epoch": 11.449023638232271,
+      "grad_norm": 0.09390268474817276,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 111399
+    },
+    {
+      "epoch": 11.449331963001027,
+      "grad_norm": 0.13381150364875793,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 111402
+    },
+    {
+      "epoch": 11.449640287769784,
+      "grad_norm": 0.17974214255809784,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 111405
+    },
+    {
+      "epoch": 11.44994861253854,
+      "grad_norm": 0.06824097782373428,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 111408
+    },
+    {
+      "epoch": 11.450256937307296,
+      "grad_norm": 0.043500106781721115,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 111411
+    },
+    {
+      "epoch": 11.450565262076054,
+      "grad_norm": 0.03300514072179794,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 111414
+    },
+    {
+      "epoch": 11.45087358684481,
+      "grad_norm": 0.0398101806640625,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 111417
+    },
+    {
+      "epoch": 11.451181911613567,
+      "grad_norm": 0.04787025600671768,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 111420
+    },
+    {
+      "epoch": 11.451490236382323,
+      "grad_norm": 0.03133167326450348,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 111423
+    },
+    {
+      "epoch": 11.45179856115108,
+      "grad_norm": 0.04809502139687538,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 111426
+    },
+    {
+      "epoch": 11.452106885919836,
+      "grad_norm": 0.08147426694631577,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 111429
+    },
+    {
+      "epoch": 11.452415210688592,
+      "grad_norm": 0.058343514800071716,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 111432
+    },
+    {
+      "epoch": 11.452723535457348,
+      "grad_norm": 0.06231757998466492,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 111435
+    },
+    {
+      "epoch": 11.453031860226105,
+      "grad_norm": 0.08571364730596542,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 111438
+    },
+    {
+      "epoch": 11.453340184994861,
+      "grad_norm": 0.04634599760174751,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 111441
+    },
+    {
+      "epoch": 11.453648509763617,
+      "grad_norm": 0.04871790483593941,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 111444
+    },
+    {
+      "epoch": 11.453956834532374,
+      "grad_norm": 0.03777323663234711,
+      "learning_rate": 0.01,
+      "loss": 1.9706,
+      "step": 111447
+    },
+    {
+      "epoch": 11.45426515930113,
+      "grad_norm": 0.08520831167697906,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 111450
+    },
+    {
+      "epoch": 11.454573484069886,
+      "grad_norm": 0.10346395522356033,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 111453
+    },
+    {
+      "epoch": 11.454881808838643,
+      "grad_norm": 0.1196495071053505,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 111456
+    },
+    {
+      "epoch": 11.455190133607399,
+      "grad_norm": 0.12974438071250916,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 111459
+    },
+    {
+      "epoch": 11.455498458376157,
+      "grad_norm": 0.11819576472043991,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 111462
+    },
+    {
+      "epoch": 11.455806783144913,
+      "grad_norm": 0.12511448562145233,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 111465
+    },
+    {
+      "epoch": 11.45611510791367,
+      "grad_norm": 0.10518360882997513,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 111468
+    },
+    {
+      "epoch": 11.456423432682426,
+      "grad_norm": 0.06951570510864258,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 111471
+    },
+    {
+      "epoch": 11.456731757451182,
+      "grad_norm": 0.052723489701747894,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 111474
+    },
+    {
+      "epoch": 11.457040082219939,
+      "grad_norm": 0.040231626480817795,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 111477
+    },
+    {
+      "epoch": 11.457348406988695,
+      "grad_norm": 0.029970921576023102,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 111480
+    },
+    {
+      "epoch": 11.457656731757451,
+      "grad_norm": 0.050749361515045166,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 111483
+    },
+    {
+      "epoch": 11.457965056526207,
+      "grad_norm": 0.06277536600828171,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 111486
+    },
+    {
+      "epoch": 11.458273381294964,
+      "grad_norm": 0.07072249054908752,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 111489
+    },
+    {
+      "epoch": 11.45858170606372,
+      "grad_norm": 0.06058824807405472,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 111492
+    },
+    {
+      "epoch": 11.458890030832476,
+      "grad_norm": 0.1610943228006363,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 111495
+    },
+    {
+      "epoch": 11.459198355601233,
+      "grad_norm": 0.0705665647983551,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 111498
+    },
+    {
+      "epoch": 11.459506680369989,
+      "grad_norm": 0.04693428799510002,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 111501
+    },
+    {
+      "epoch": 11.459815005138745,
+      "grad_norm": 0.11832481622695923,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 111504
+    },
+    {
+      "epoch": 11.460123329907503,
+      "grad_norm": 0.08116341382265091,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 111507
+    },
+    {
+      "epoch": 11.46043165467626,
+      "grad_norm": 0.07210972160100937,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 111510
+    },
+    {
+      "epoch": 11.460739979445016,
+      "grad_norm": 0.06692171096801758,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 111513
+    },
+    {
+      "epoch": 11.461048304213772,
+      "grad_norm": 0.09386149048805237,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 111516
+    },
+    {
+      "epoch": 11.461356628982529,
+      "grad_norm": 0.03938892111182213,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 111519
+    },
+    {
+      "epoch": 11.461664953751285,
+      "grad_norm": 0.15739138424396515,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 111522
+    },
+    {
+      "epoch": 11.461973278520041,
+      "grad_norm": 0.13813142478466034,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 111525
+    },
+    {
+      "epoch": 11.462281603288798,
+      "grad_norm": 0.08947230130434036,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 111528
+    },
+    {
+      "epoch": 11.462589928057554,
+      "grad_norm": 0.05972016230225563,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 111531
+    },
+    {
+      "epoch": 11.46289825282631,
+      "grad_norm": 0.05006679147481918,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 111534
+    },
+    {
+      "epoch": 11.463206577595066,
+      "grad_norm": 0.04381818696856499,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 111537
+    },
+    {
+      "epoch": 11.463514902363823,
+      "grad_norm": 0.03743549436330795,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 111540
+    },
+    {
+      "epoch": 11.463823227132579,
+      "grad_norm": 0.04342208057641983,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 111543
+    },
+    {
+      "epoch": 11.464131551901335,
+      "grad_norm": 0.08246835321187973,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 111546
+    },
+    {
+      "epoch": 11.464439876670092,
+      "grad_norm": 0.052904337644577026,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 111549
+    },
+    {
+      "epoch": 11.46474820143885,
+      "grad_norm": 0.07945089042186737,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 111552
+    },
+    {
+      "epoch": 11.465056526207606,
+      "grad_norm": 0.09860449284315109,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 111555
+    },
+    {
+      "epoch": 11.465364850976362,
+      "grad_norm": 0.07978519052267075,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 111558
+    },
+    {
+      "epoch": 11.465673175745119,
+      "grad_norm": 0.09356561303138733,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 111561
+    },
+    {
+      "epoch": 11.465981500513875,
+      "grad_norm": 0.08849091827869415,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 111564
+    },
+    {
+      "epoch": 11.466289825282631,
+      "grad_norm": 0.07738533616065979,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 111567
+    },
+    {
+      "epoch": 11.466598150051388,
+      "grad_norm": 0.06273981928825378,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 111570
+    },
+    {
+      "epoch": 11.466906474820144,
+      "grad_norm": 0.03618188202381134,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 111573
+    },
+    {
+      "epoch": 11.4672147995889,
+      "grad_norm": 0.03245396167039871,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 111576
+    },
+    {
+      "epoch": 11.467523124357657,
+      "grad_norm": 0.054043613374233246,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 111579
+    },
+    {
+      "epoch": 11.467831449126413,
+      "grad_norm": 0.14490677416324615,
+      "learning_rate": 0.01,
+      "loss": 1.978,
+      "step": 111582
+    },
+    {
+      "epoch": 11.46813977389517,
+      "grad_norm": 0.05632997304201126,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 111585
+    },
+    {
+      "epoch": 11.468448098663925,
+      "grad_norm": 0.13351985812187195,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 111588
+    },
+    {
+      "epoch": 11.468756423432682,
+      "grad_norm": 0.0480462908744812,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 111591
+    },
+    {
+      "epoch": 11.469064748201438,
+      "grad_norm": 0.05255371332168579,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 111594
+    },
+    {
+      "epoch": 11.469373072970196,
+      "grad_norm": 0.050016168504953384,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 111597
+    },
+    {
+      "epoch": 11.469681397738952,
+      "grad_norm": 0.033241692930459976,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 111600
+    },
+    {
+      "epoch": 11.469989722507709,
+      "grad_norm": 0.037866104394197464,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 111603
+    },
+    {
+      "epoch": 11.470298047276465,
+      "grad_norm": 0.10859289765357971,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 111606
+    },
+    {
+      "epoch": 11.470606372045221,
+      "grad_norm": 0.041134439408779144,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 111609
+    },
+    {
+      "epoch": 11.470914696813978,
+      "grad_norm": 0.07839282602071762,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 111612
+    },
+    {
+      "epoch": 11.471223021582734,
+      "grad_norm": 0.09817814826965332,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 111615
+    },
+    {
+      "epoch": 11.47153134635149,
+      "grad_norm": 0.11221611499786377,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 111618
+    },
+    {
+      "epoch": 11.471839671120247,
+      "grad_norm": 0.06966344267129898,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 111621
+    },
+    {
+      "epoch": 11.472147995889003,
+      "grad_norm": 0.06667056679725647,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 111624
+    },
+    {
+      "epoch": 11.47245632065776,
+      "grad_norm": 0.07580654323101044,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 111627
+    },
+    {
+      "epoch": 11.472764645426516,
+      "grad_norm": 0.05771489813923836,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 111630
+    },
+    {
+      "epoch": 11.473072970195272,
+      "grad_norm": 0.05805226042866707,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 111633
+    },
+    {
+      "epoch": 11.473381294964028,
+      "grad_norm": 0.04712773114442825,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 111636
+    },
+    {
+      "epoch": 11.473689619732784,
+      "grad_norm": 0.04592062160372734,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 111639
+    },
+    {
+      "epoch": 11.47399794450154,
+      "grad_norm": 0.06208406761288643,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 111642
+    },
+    {
+      "epoch": 11.474306269270299,
+      "grad_norm": 0.052417706698179245,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 111645
+    },
+    {
+      "epoch": 11.474614594039055,
+      "grad_norm": 0.1250835508108139,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 111648
+    },
+    {
+      "epoch": 11.474922918807811,
+      "grad_norm": 0.04983068257570267,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 111651
+    },
+    {
+      "epoch": 11.475231243576568,
+      "grad_norm": 0.09713703393936157,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 111654
+    },
+    {
+      "epoch": 11.475539568345324,
+      "grad_norm": 0.04548988118767738,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 111657
+    },
+    {
+      "epoch": 11.47584789311408,
+      "grad_norm": 0.0484665147960186,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 111660
+    },
+    {
+      "epoch": 11.476156217882837,
+      "grad_norm": 0.059269923716783524,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 111663
+    },
+    {
+      "epoch": 11.476464542651593,
+      "grad_norm": 0.1312210112810135,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 111666
+    },
+    {
+      "epoch": 11.47677286742035,
+      "grad_norm": 0.04668668285012245,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 111669
+    },
+    {
+      "epoch": 11.477081192189106,
+      "grad_norm": 0.05484083294868469,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 111672
+    },
+    {
+      "epoch": 11.477389516957862,
+      "grad_norm": 0.10938485711812973,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 111675
+    },
+    {
+      "epoch": 11.477697841726618,
+      "grad_norm": 0.055881600826978683,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 111678
+    },
+    {
+      "epoch": 11.478006166495375,
+      "grad_norm": 0.03705210238695145,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 111681
+    },
+    {
+      "epoch": 11.47831449126413,
+      "grad_norm": 0.035369958728551865,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 111684
+    },
+    {
+      "epoch": 11.478622816032887,
+      "grad_norm": 0.03749634325504303,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 111687
+    },
+    {
+      "epoch": 11.478931140801645,
+      "grad_norm": 0.042402010411024094,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 111690
+    },
+    {
+      "epoch": 11.479239465570402,
+      "grad_norm": 0.13027635216712952,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 111693
+    },
+    {
+      "epoch": 11.479547790339158,
+      "grad_norm": 0.09507349133491516,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 111696
+    },
+    {
+      "epoch": 11.479856115107914,
+      "grad_norm": 0.06717098504304886,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 111699
+    },
+    {
+      "epoch": 11.48016443987667,
+      "grad_norm": 0.05186061933636665,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 111702
+    },
+    {
+      "epoch": 11.480472764645427,
+      "grad_norm": 0.05118173733353615,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 111705
+    },
+    {
+      "epoch": 11.480781089414183,
+      "grad_norm": 0.0630086213350296,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 111708
+    },
+    {
+      "epoch": 11.48108941418294,
+      "grad_norm": 0.09794661402702332,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 111711
+    },
+    {
+      "epoch": 11.481397738951696,
+      "grad_norm": 0.04907077178359032,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 111714
+    },
+    {
+      "epoch": 11.481706063720452,
+      "grad_norm": 0.07658752053976059,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 111717
+    },
+    {
+      "epoch": 11.482014388489208,
+      "grad_norm": 0.09202036261558533,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 111720
+    },
+    {
+      "epoch": 11.482322713257965,
+      "grad_norm": 0.06585373729467392,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 111723
+    },
+    {
+      "epoch": 11.482631038026721,
+      "grad_norm": 0.07491261512041092,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 111726
+    },
+    {
+      "epoch": 11.482939362795477,
+      "grad_norm": 0.09483268111944199,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 111729
+    },
+    {
+      "epoch": 11.483247687564234,
+      "grad_norm": 0.15194907784461975,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 111732
+    },
+    {
+      "epoch": 11.48355601233299,
+      "grad_norm": 0.14718058705329895,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 111735
+    },
+    {
+      "epoch": 11.483864337101748,
+      "grad_norm": 0.1479717642068863,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 111738
+    },
+    {
+      "epoch": 11.484172661870504,
+      "grad_norm": 0.0986589714884758,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 111741
+    },
+    {
+      "epoch": 11.48448098663926,
+      "grad_norm": 0.08421244472265244,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 111744
+    },
+    {
+      "epoch": 11.484789311408017,
+      "grad_norm": 0.08169248700141907,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 111747
+    },
+    {
+      "epoch": 11.485097636176773,
+      "grad_norm": 0.08077950030565262,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 111750
+    },
+    {
+      "epoch": 11.48540596094553,
+      "grad_norm": 0.08002752810716629,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 111753
+    },
+    {
+      "epoch": 11.485714285714286,
+      "grad_norm": 0.050435587763786316,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 111756
+    },
+    {
+      "epoch": 11.486022610483042,
+      "grad_norm": 0.043131690472364426,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 111759
+    },
+    {
+      "epoch": 11.486330935251798,
+      "grad_norm": 0.05599942058324814,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 111762
+    },
+    {
+      "epoch": 11.486639260020555,
+      "grad_norm": 0.09633025527000427,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 111765
+    },
+    {
+      "epoch": 11.486947584789311,
+      "grad_norm": 0.07188762724399567,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 111768
+    },
+    {
+      "epoch": 11.487255909558067,
+      "grad_norm": 0.05295489728450775,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 111771
+    },
+    {
+      "epoch": 11.487564234326824,
+      "grad_norm": 0.03551109880208969,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 111774
+    },
+    {
+      "epoch": 11.48787255909558,
+      "grad_norm": 0.09166920185089111,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 111777
+    },
+    {
+      "epoch": 11.488180883864338,
+      "grad_norm": 0.06509532779455185,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 111780
+    },
+    {
+      "epoch": 11.488489208633094,
+      "grad_norm": 0.03466518223285675,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 111783
+    },
+    {
+      "epoch": 11.48879753340185,
+      "grad_norm": 0.06080121546983719,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 111786
+    },
+    {
+      "epoch": 11.489105858170607,
+      "grad_norm": 0.07743913680315018,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 111789
+    },
+    {
+      "epoch": 11.489414182939363,
+      "grad_norm": 0.04427964985370636,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 111792
+    },
+    {
+      "epoch": 11.48972250770812,
+      "grad_norm": 0.049059998244047165,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 111795
+    },
+    {
+      "epoch": 11.490030832476876,
+      "grad_norm": 0.0991068035364151,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 111798
+    },
+    {
+      "epoch": 11.490339157245632,
+      "grad_norm": 0.0421556755900383,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 111801
+    },
+    {
+      "epoch": 11.490647482014388,
+      "grad_norm": 0.03575088083744049,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 111804
+    },
+    {
+      "epoch": 11.490955806783145,
+      "grad_norm": 0.05111589655280113,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 111807
+    },
+    {
+      "epoch": 11.491264131551901,
+      "grad_norm": 0.04898693785071373,
+      "learning_rate": 0.01,
+      "loss": 1.9065,
+      "step": 111810
+    },
+    {
+      "epoch": 11.491572456320657,
+      "grad_norm": 0.16972552239894867,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 111813
+    },
+    {
+      "epoch": 11.491880781089414,
+      "grad_norm": 0.11789322644472122,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 111816
+    },
+    {
+      "epoch": 11.49218910585817,
+      "grad_norm": 0.04322198033332825,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 111819
+    },
+    {
+      "epoch": 11.492497430626926,
+      "grad_norm": 0.05791204422712326,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 111822
+    },
+    {
+      "epoch": 11.492805755395683,
+      "grad_norm": 0.045334842056035995,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 111825
+    },
+    {
+      "epoch": 11.49311408016444,
+      "grad_norm": 0.040358442813158035,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 111828
+    },
+    {
+      "epoch": 11.493422404933197,
+      "grad_norm": 0.11316455900669098,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 111831
+    },
+    {
+      "epoch": 11.493730729701953,
+      "grad_norm": 0.06410171091556549,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 111834
+    },
+    {
+      "epoch": 11.49403905447071,
+      "grad_norm": 0.05099207162857056,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 111837
+    },
+    {
+      "epoch": 11.494347379239466,
+      "grad_norm": 0.13952548801898956,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 111840
+    },
+    {
+      "epoch": 11.494655704008222,
+      "grad_norm": 0.03869643807411194,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 111843
+    },
+    {
+      "epoch": 11.494964028776979,
+      "grad_norm": 0.1269671618938446,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 111846
+    },
+    {
+      "epoch": 11.495272353545735,
+      "grad_norm": 0.08769647032022476,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 111849
+    },
+    {
+      "epoch": 11.495580678314491,
+      "grad_norm": 0.058100420981645584,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 111852
+    },
+    {
+      "epoch": 11.495889003083247,
+      "grad_norm": 0.038575999438762665,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 111855
+    },
+    {
+      "epoch": 11.496197327852004,
+      "grad_norm": 0.04461336135864258,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 111858
+    },
+    {
+      "epoch": 11.49650565262076,
+      "grad_norm": 0.030741844326257706,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 111861
+    },
+    {
+      "epoch": 11.496813977389516,
+      "grad_norm": 0.05148760601878166,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 111864
+    },
+    {
+      "epoch": 11.497122302158273,
+      "grad_norm": 0.11670944094657898,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 111867
+    },
+    {
+      "epoch": 11.497430626927029,
+      "grad_norm": 0.04056670889258385,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 111870
+    },
+    {
+      "epoch": 11.497738951695787,
+      "grad_norm": 0.10150992125272751,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 111873
+    },
+    {
+      "epoch": 11.498047276464543,
+      "grad_norm": 0.08927108347415924,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 111876
+    },
+    {
+      "epoch": 11.4983556012333,
+      "grad_norm": 0.1410294771194458,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 111879
+    },
+    {
+      "epoch": 11.498663926002056,
+      "grad_norm": 0.09127254784107208,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 111882
+    },
+    {
+      "epoch": 11.498972250770812,
+      "grad_norm": 0.0577942319214344,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 111885
+    },
+    {
+      "epoch": 11.499280575539569,
+      "grad_norm": 0.07099872827529907,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 111888
+    },
+    {
+      "epoch": 11.499588900308325,
+      "grad_norm": 0.08721621334552765,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 111891
+    },
+    {
+      "epoch": 11.499897225077081,
+      "grad_norm": 0.049968644976615906,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 111894
+    },
+    {
+      "epoch": 11.500205549845838,
+      "grad_norm": 0.08129146695137024,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 111897
+    },
+    {
+      "epoch": 11.500513874614594,
+      "grad_norm": 0.10893581807613373,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 111900
+    },
+    {
+      "epoch": 11.50082219938335,
+      "grad_norm": 0.10866726189851761,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 111903
+    },
+    {
+      "epoch": 11.501130524152106,
+      "grad_norm": 0.0690096914768219,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 111906
+    },
+    {
+      "epoch": 11.501438848920863,
+      "grad_norm": 0.04975270479917526,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 111909
+    },
+    {
+      "epoch": 11.501747173689619,
+      "grad_norm": 0.07065418362617493,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 111912
+    },
+    {
+      "epoch": 11.502055498458375,
+      "grad_norm": 0.06333848834037781,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 111915
+    },
+    {
+      "epoch": 11.502363823227132,
+      "grad_norm": 0.053953733295202255,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 111918
+    },
+    {
+      "epoch": 11.50267214799589,
+      "grad_norm": 0.06409978866577148,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 111921
+    },
+    {
+      "epoch": 11.502980472764646,
+      "grad_norm": 0.12793174386024475,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 111924
+    },
+    {
+      "epoch": 11.503288797533402,
+      "grad_norm": 0.04191777482628822,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 111927
+    },
+    {
+      "epoch": 11.503597122302159,
+      "grad_norm": 0.10092096775770187,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 111930
+    },
+    {
+      "epoch": 11.503905447070915,
+      "grad_norm": 0.09913140535354614,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 111933
+    },
+    {
+      "epoch": 11.504213771839671,
+      "grad_norm": 0.05205460265278816,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 111936
+    },
+    {
+      "epoch": 11.504522096608428,
+      "grad_norm": 0.10730945318937302,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 111939
+    },
+    {
+      "epoch": 11.504830421377184,
+      "grad_norm": 0.034082312136888504,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 111942
+    },
+    {
+      "epoch": 11.50513874614594,
+      "grad_norm": 0.06130027025938034,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 111945
+    },
+    {
+      "epoch": 11.505447070914697,
+      "grad_norm": 0.07574241608381271,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 111948
+    },
+    {
+      "epoch": 11.505755395683453,
+      "grad_norm": 0.06435610353946686,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 111951
+    },
+    {
+      "epoch": 11.50606372045221,
+      "grad_norm": 0.08691317588090897,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 111954
+    },
+    {
+      "epoch": 11.506372045220965,
+      "grad_norm": 0.11261089891195297,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 111957
+    },
+    {
+      "epoch": 11.506680369989722,
+      "grad_norm": 0.0669609010219574,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 111960
+    },
+    {
+      "epoch": 11.50698869475848,
+      "grad_norm": 0.039548080414533615,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 111963
+    },
+    {
+      "epoch": 11.507297019527236,
+      "grad_norm": 0.1087004542350769,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 111966
+    },
+    {
+      "epoch": 11.507605344295992,
+      "grad_norm": 0.07893752306699753,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 111969
+    },
+    {
+      "epoch": 11.507913669064749,
+      "grad_norm": 0.11534779518842697,
+      "learning_rate": 0.01,
+      "loss": 1.8917,
+      "step": 111972
+    },
+    {
+      "epoch": 11.508221993833505,
+      "grad_norm": 0.0677044466137886,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 111975
+    },
+    {
+      "epoch": 11.508530318602261,
+      "grad_norm": 0.054873283952474594,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 111978
+    },
+    {
+      "epoch": 11.508838643371018,
+      "grad_norm": 0.041117701679468155,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 111981
+    },
+    {
+      "epoch": 11.509146968139774,
+      "grad_norm": 0.099622942507267,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 111984
+    },
+    {
+      "epoch": 11.50945529290853,
+      "grad_norm": 0.0813162699341774,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 111987
+    },
+    {
+      "epoch": 11.509763617677287,
+      "grad_norm": 0.05334705859422684,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 111990
+    },
+    {
+      "epoch": 11.510071942446043,
+      "grad_norm": 0.08059155941009521,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 111993
+    },
+    {
+      "epoch": 11.5103802672148,
+      "grad_norm": 0.07139837741851807,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 111996
+    },
+    {
+      "epoch": 11.510688591983556,
+      "grad_norm": 0.14807172119617462,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 111999
+    },
+    {
+      "epoch": 11.510996916752312,
+      "grad_norm": 0.0448773056268692,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 112002
+    },
+    {
+      "epoch": 11.511305241521068,
+      "grad_norm": 0.06124507263302803,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 112005
+    },
+    {
+      "epoch": 11.511613566289824,
+      "grad_norm": 0.06960859894752502,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 112008
+    },
+    {
+      "epoch": 11.51192189105858,
+      "grad_norm": 0.08753380924463272,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 112011
+    },
+    {
+      "epoch": 11.512230215827339,
+      "grad_norm": 0.15506276488304138,
+      "learning_rate": 0.01,
+      "loss": 1.9687,
+      "step": 112014
+    },
+    {
+      "epoch": 11.512538540596095,
+      "grad_norm": 0.0536036379635334,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 112017
+    },
+    {
+      "epoch": 11.512846865364851,
+      "grad_norm": 0.08961272239685059,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 112020
+    },
+    {
+      "epoch": 11.513155190133608,
+      "grad_norm": 0.058408934623003006,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 112023
+    },
+    {
+      "epoch": 11.513463514902364,
+      "grad_norm": 0.041266631335020065,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 112026
+    },
+    {
+      "epoch": 11.51377183967112,
+      "grad_norm": 0.1152496337890625,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 112029
+    },
+    {
+      "epoch": 11.514080164439877,
+      "grad_norm": 0.07906953990459442,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 112032
+    },
+    {
+      "epoch": 11.514388489208633,
+      "grad_norm": 0.14202702045440674,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 112035
+    },
+    {
+      "epoch": 11.51469681397739,
+      "grad_norm": 0.03130604699254036,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 112038
+    },
+    {
+      "epoch": 11.515005138746146,
+      "grad_norm": 0.058028191328048706,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 112041
+    },
+    {
+      "epoch": 11.515313463514902,
+      "grad_norm": 0.06180107221007347,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 112044
+    },
+    {
+      "epoch": 11.515621788283658,
+      "grad_norm": 0.03522508591413498,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 112047
+    },
+    {
+      "epoch": 11.515930113052415,
+      "grad_norm": 0.040822502225637436,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 112050
+    },
+    {
+      "epoch": 11.51623843782117,
+      "grad_norm": 0.03846633806824684,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 112053
+    },
+    {
+      "epoch": 11.516546762589929,
+      "grad_norm": 0.14067944884300232,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 112056
+    },
+    {
+      "epoch": 11.516855087358685,
+      "grad_norm": 0.04643148556351662,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 112059
+    },
+    {
+      "epoch": 11.517163412127442,
+      "grad_norm": 0.10015133023262024,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 112062
+    },
+    {
+      "epoch": 11.517471736896198,
+      "grad_norm": 0.043383967131376266,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 112065
+    },
+    {
+      "epoch": 11.517780061664954,
+      "grad_norm": 0.05876270309090614,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 112068
+    },
+    {
+      "epoch": 11.51808838643371,
+      "grad_norm": 0.0826994925737381,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 112071
+    },
+    {
+      "epoch": 11.518396711202467,
+      "grad_norm": 0.11554791778326035,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 112074
+    },
+    {
+      "epoch": 11.518705035971223,
+      "grad_norm": 0.07947423309087753,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 112077
+    },
+    {
+      "epoch": 11.51901336073998,
+      "grad_norm": 0.049170155078172684,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 112080
+    },
+    {
+      "epoch": 11.519321685508736,
+      "grad_norm": 0.058392517268657684,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 112083
+    },
+    {
+      "epoch": 11.519630010277492,
+      "grad_norm": 0.03751151263713837,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 112086
+    },
+    {
+      "epoch": 11.519938335046248,
+      "grad_norm": 0.04583292081952095,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 112089
+    },
+    {
+      "epoch": 11.520246659815005,
+      "grad_norm": 0.08585547655820847,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 112092
+    },
+    {
+      "epoch": 11.520554984583761,
+      "grad_norm": 0.07672592252492905,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 112095
+    },
+    {
+      "epoch": 11.520863309352517,
+      "grad_norm": 0.07113534957170486,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 112098
+    },
+    {
+      "epoch": 11.521171634121274,
+      "grad_norm": 0.11068326979875565,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 112101
+    },
+    {
+      "epoch": 11.521479958890032,
+      "grad_norm": 0.060119953006505966,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 112104
+    },
+    {
+      "epoch": 11.521788283658788,
+      "grad_norm": 0.05784304067492485,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 112107
+    },
+    {
+      "epoch": 11.522096608427544,
+      "grad_norm": 0.06138474494218826,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 112110
+    },
+    {
+      "epoch": 11.5224049331963,
+      "grad_norm": 0.06129737198352814,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 112113
+    },
+    {
+      "epoch": 11.522713257965057,
+      "grad_norm": 0.06837484985589981,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 112116
+    },
+    {
+      "epoch": 11.523021582733813,
+      "grad_norm": 0.11184889823198318,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 112119
+    },
+    {
+      "epoch": 11.52332990750257,
+      "grad_norm": 0.1556568741798401,
+      "learning_rate": 0.01,
+      "loss": 1.9804,
+      "step": 112122
+    },
+    {
+      "epoch": 11.523638232271326,
+      "grad_norm": 0.08367830514907837,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 112125
+    },
+    {
+      "epoch": 11.523946557040082,
+      "grad_norm": 0.08855171501636505,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 112128
+    },
+    {
+      "epoch": 11.524254881808838,
+      "grad_norm": 0.1637917459011078,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 112131
+    },
+    {
+      "epoch": 11.524563206577595,
+      "grad_norm": 0.08338747918605804,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 112134
+    },
+    {
+      "epoch": 11.524871531346351,
+      "grad_norm": 0.06042430177330971,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 112137
+    },
+    {
+      "epoch": 11.525179856115107,
+      "grad_norm": 0.12024378031492233,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 112140
+    },
+    {
+      "epoch": 11.525488180883864,
+      "grad_norm": 0.053831133991479874,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 112143
+    },
+    {
+      "epoch": 11.525796505652622,
+      "grad_norm": 0.03246373310685158,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 112146
+    },
+    {
+      "epoch": 11.526104830421378,
+      "grad_norm": 0.03484547138214111,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 112149
+    },
+    {
+      "epoch": 11.526413155190134,
+      "grad_norm": 0.08741461485624313,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 112152
+    },
+    {
+      "epoch": 11.52672147995889,
+      "grad_norm": 0.08762611448764801,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 112155
+    },
+    {
+      "epoch": 11.527029804727647,
+      "grad_norm": 0.055941540747880936,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 112158
+    },
+    {
+      "epoch": 11.527338129496403,
+      "grad_norm": 0.04995638132095337,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 112161
+    },
+    {
+      "epoch": 11.52764645426516,
+      "grad_norm": 0.033121444284915924,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 112164
+    },
+    {
+      "epoch": 11.527954779033916,
+      "grad_norm": 0.10392650216817856,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 112167
+    },
+    {
+      "epoch": 11.528263103802672,
+      "grad_norm": 0.0696452334523201,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 112170
+    },
+    {
+      "epoch": 11.528571428571428,
+      "grad_norm": 0.08160513639450073,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 112173
+    },
+    {
+      "epoch": 11.528879753340185,
+      "grad_norm": 0.0839330330491066,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 112176
+    },
+    {
+      "epoch": 11.529188078108941,
+      "grad_norm": 0.09454485774040222,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 112179
+    },
+    {
+      "epoch": 11.529496402877697,
+      "grad_norm": 0.08612924814224243,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 112182
+    },
+    {
+      "epoch": 11.529804727646454,
+      "grad_norm": 0.08306564390659332,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 112185
+    },
+    {
+      "epoch": 11.53011305241521,
+      "grad_norm": 0.07982834428548813,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 112188
+    },
+    {
+      "epoch": 11.530421377183966,
+      "grad_norm": 0.0803823247551918,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 112191
+    },
+    {
+      "epoch": 11.530729701952723,
+      "grad_norm": 0.06836555898189545,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 112194
+    },
+    {
+      "epoch": 11.53103802672148,
+      "grad_norm": 0.046713683754205704,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 112197
+    },
+    {
+      "epoch": 11.531346351490237,
+      "grad_norm": 0.034670259803533554,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 112200
+    },
+    {
+      "epoch": 11.531654676258993,
+      "grad_norm": 0.112801194190979,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 112203
+    },
+    {
+      "epoch": 11.53196300102775,
+      "grad_norm": 0.11906672269105911,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 112206
+    },
+    {
+      "epoch": 11.532271325796506,
+      "grad_norm": 0.06057396158576012,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 112209
+    },
+    {
+      "epoch": 11.532579650565262,
+      "grad_norm": 0.04204052686691284,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 112212
+    },
+    {
+      "epoch": 11.532887975334019,
+      "grad_norm": 0.08645797520875931,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 112215
+    },
+    {
+      "epoch": 11.533196300102775,
+      "grad_norm": 0.05199515074491501,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 112218
+    },
+    {
+      "epoch": 11.533504624871531,
+      "grad_norm": 0.03792623057961464,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 112221
+    },
+    {
+      "epoch": 11.533812949640287,
+      "grad_norm": 0.08833713829517365,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 112224
+    },
+    {
+      "epoch": 11.534121274409044,
+      "grad_norm": 0.054692838340997696,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 112227
+    },
+    {
+      "epoch": 11.5344295991778,
+      "grad_norm": 0.11953523010015488,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 112230
+    },
+    {
+      "epoch": 11.534737923946556,
+      "grad_norm": 0.05324847996234894,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 112233
+    },
+    {
+      "epoch": 11.535046248715313,
+      "grad_norm": 0.05562381446361542,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 112236
+    },
+    {
+      "epoch": 11.53535457348407,
+      "grad_norm": 0.09408646821975708,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 112239
+    },
+    {
+      "epoch": 11.535662898252827,
+      "grad_norm": 0.07749692350625992,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 112242
+    },
+    {
+      "epoch": 11.535971223021583,
+      "grad_norm": 0.12444134801626205,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 112245
+    },
+    {
+      "epoch": 11.53627954779034,
+      "grad_norm": 0.09990056604146957,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 112248
+    },
+    {
+      "epoch": 11.536587872559096,
+      "grad_norm": 0.045255426317453384,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 112251
+    },
+    {
+      "epoch": 11.536896197327852,
+      "grad_norm": 0.03870481252670288,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 112254
+    },
+    {
+      "epoch": 11.537204522096609,
+      "grad_norm": 0.04264555871486664,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 112257
+    },
+    {
+      "epoch": 11.537512846865365,
+      "grad_norm": 0.1033821702003479,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 112260
+    },
+    {
+      "epoch": 11.537821171634121,
+      "grad_norm": 0.06186060607433319,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 112263
+    },
+    {
+      "epoch": 11.538129496402878,
+      "grad_norm": 0.06803090125322342,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 112266
+    },
+    {
+      "epoch": 11.538437821171634,
+      "grad_norm": 0.07004691660404205,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 112269
+    },
+    {
+      "epoch": 11.53874614594039,
+      "grad_norm": 0.09920225292444229,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 112272
+    },
+    {
+      "epoch": 11.539054470709146,
+      "grad_norm": 0.05515587329864502,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 112275
+    },
+    {
+      "epoch": 11.539362795477903,
+      "grad_norm": 0.14733068645000458,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 112278
+    },
+    {
+      "epoch": 11.539671120246659,
+      "grad_norm": 0.08791196346282959,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 112281
+    },
+    {
+      "epoch": 11.539979445015415,
+      "grad_norm": 0.048449866473674774,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 112284
+    },
+    {
+      "epoch": 11.540287769784173,
+      "grad_norm": 0.054437413811683655,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 112287
+    },
+    {
+      "epoch": 11.54059609455293,
+      "grad_norm": 0.08245435357093811,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 112290
+    },
+    {
+      "epoch": 11.540904419321686,
+      "grad_norm": 0.10597596317529678,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 112293
+    },
+    {
+      "epoch": 11.541212744090442,
+      "grad_norm": 0.14799369871616364,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 112296
+    },
+    {
+      "epoch": 11.541521068859199,
+      "grad_norm": 0.08486896753311157,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 112299
+    },
+    {
+      "epoch": 11.541829393627955,
+      "grad_norm": 0.05308954417705536,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 112302
+    },
+    {
+      "epoch": 11.542137718396711,
+      "grad_norm": 0.05601240321993828,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 112305
+    },
+    {
+      "epoch": 11.542446043165468,
+      "grad_norm": 0.04067224636673927,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 112308
+    },
+    {
+      "epoch": 11.542754367934224,
+      "grad_norm": 0.04340583458542824,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 112311
+    },
+    {
+      "epoch": 11.54306269270298,
+      "grad_norm": 0.038530174642801285,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 112314
+    },
+    {
+      "epoch": 11.543371017471737,
+      "grad_norm": 0.04171457886695862,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 112317
+    },
+    {
+      "epoch": 11.543679342240493,
+      "grad_norm": 0.04869476333260536,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 112320
+    },
+    {
+      "epoch": 11.54398766700925,
+      "grad_norm": 0.11877516657114029,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 112323
+    },
+    {
+      "epoch": 11.544295991778005,
+      "grad_norm": 0.06697926670312881,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 112326
+    },
+    {
+      "epoch": 11.544604316546762,
+      "grad_norm": 0.08970346301794052,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 112329
+    },
+    {
+      "epoch": 11.54491264131552,
+      "grad_norm": 0.052757494151592255,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 112332
+    },
+    {
+      "epoch": 11.545220966084276,
+      "grad_norm": 0.10075823217630386,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 112335
+    },
+    {
+      "epoch": 11.545529290853032,
+      "grad_norm": 0.07792484015226364,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 112338
+    },
+    {
+      "epoch": 11.545837615621789,
+      "grad_norm": 0.08613074570894241,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 112341
+    },
+    {
+      "epoch": 11.546145940390545,
+      "grad_norm": 0.10087897628545761,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 112344
+    },
+    {
+      "epoch": 11.546454265159301,
+      "grad_norm": 0.047122254967689514,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 112347
+    },
+    {
+      "epoch": 11.546762589928058,
+      "grad_norm": 0.053861238062381744,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 112350
+    },
+    {
+      "epoch": 11.547070914696814,
+      "grad_norm": 0.04060134291648865,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 112353
+    },
+    {
+      "epoch": 11.54737923946557,
+      "grad_norm": 0.04421015828847885,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 112356
+    },
+    {
+      "epoch": 11.547687564234327,
+      "grad_norm": 0.050058286637067795,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 112359
+    },
+    {
+      "epoch": 11.547995889003083,
+      "grad_norm": 0.06190727651119232,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 112362
+    },
+    {
+      "epoch": 11.54830421377184,
+      "grad_norm": 0.03759567439556122,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 112365
+    },
+    {
+      "epoch": 11.548612538540596,
+      "grad_norm": 0.04162626340985298,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 112368
+    },
+    {
+      "epoch": 11.548920863309352,
+      "grad_norm": 0.05578555539250374,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 112371
+    },
+    {
+      "epoch": 11.549229188078108,
+      "grad_norm": 0.0695364773273468,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 112374
+    },
+    {
+      "epoch": 11.549537512846864,
+      "grad_norm": 0.0405573733150959,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 112377
+    },
+    {
+      "epoch": 11.549845837615623,
+      "grad_norm": 0.03472146764397621,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 112380
+    },
+    {
+      "epoch": 11.550154162384379,
+      "grad_norm": 0.05320744216442108,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 112383
+    },
+    {
+      "epoch": 11.550462487153135,
+      "grad_norm": 0.07547161728143692,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 112386
+    },
+    {
+      "epoch": 11.550770811921891,
+      "grad_norm": 0.03960610553622246,
+      "learning_rate": 0.01,
+      "loss": 1.899,
+      "step": 112389
+    },
+    {
+      "epoch": 11.551079136690648,
+      "grad_norm": 0.06855163723230362,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 112392
+    },
+    {
+      "epoch": 11.551387461459404,
+      "grad_norm": 0.0669487938284874,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 112395
+    },
+    {
+      "epoch": 11.55169578622816,
+      "grad_norm": 0.05353791266679764,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 112398
+    },
+    {
+      "epoch": 11.552004110996917,
+      "grad_norm": 0.054838988929986954,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 112401
+    },
+    {
+      "epoch": 11.552312435765673,
+      "grad_norm": 0.053157199174165726,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 112404
+    },
+    {
+      "epoch": 11.55262076053443,
+      "grad_norm": 0.04517093673348427,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 112407
+    },
+    {
+      "epoch": 11.552929085303186,
+      "grad_norm": 0.0851716473698616,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 112410
+    },
+    {
+      "epoch": 11.553237410071942,
+      "grad_norm": 0.09431596845388412,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 112413
+    },
+    {
+      "epoch": 11.553545734840698,
+      "grad_norm": 0.0770939439535141,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 112416
+    },
+    {
+      "epoch": 11.553854059609455,
+      "grad_norm": 0.08656878024339676,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 112419
+    },
+    {
+      "epoch": 11.554162384378213,
+      "grad_norm": 0.030203841626644135,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 112422
+    },
+    {
+      "epoch": 11.554470709146969,
+      "grad_norm": 0.09328059107065201,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 112425
+    },
+    {
+      "epoch": 11.554779033915725,
+      "grad_norm": 0.06437768042087555,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 112428
+    },
+    {
+      "epoch": 11.555087358684482,
+      "grad_norm": 0.08344560861587524,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 112431
+    },
+    {
+      "epoch": 11.555395683453238,
+      "grad_norm": 0.05726712942123413,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 112434
+    },
+    {
+      "epoch": 11.555704008221994,
+      "grad_norm": 0.04200645536184311,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 112437
+    },
+    {
+      "epoch": 11.55601233299075,
+      "grad_norm": 0.039023153483867645,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 112440
+    },
+    {
+      "epoch": 11.556320657759507,
+      "grad_norm": 0.12256117910146713,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 112443
+    },
+    {
+      "epoch": 11.556628982528263,
+      "grad_norm": 0.05277597904205322,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 112446
+    },
+    {
+      "epoch": 11.55693730729702,
+      "grad_norm": 0.05291856825351715,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 112449
+    },
+    {
+      "epoch": 11.557245632065776,
+      "grad_norm": 0.050023455172777176,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 112452
+    },
+    {
+      "epoch": 11.557553956834532,
+      "grad_norm": 0.046553801745176315,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 112455
+    },
+    {
+      "epoch": 11.557862281603288,
+      "grad_norm": 0.04273172467947006,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 112458
+    },
+    {
+      "epoch": 11.558170606372045,
+      "grad_norm": 0.05655885860323906,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 112461
+    },
+    {
+      "epoch": 11.558478931140801,
+      "grad_norm": 0.03471823036670685,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 112464
+    },
+    {
+      "epoch": 11.558787255909557,
+      "grad_norm": 0.03444075956940651,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 112467
+    },
+    {
+      "epoch": 11.559095580678314,
+      "grad_norm": 0.05058380216360092,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 112470
+    },
+    {
+      "epoch": 11.559403905447072,
+      "grad_norm": 0.08273141086101532,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 112473
+    },
+    {
+      "epoch": 11.559712230215828,
+      "grad_norm": 0.07805310189723969,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 112476
+    },
+    {
+      "epoch": 11.560020554984584,
+      "grad_norm": 0.055413614958524704,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 112479
+    },
+    {
+      "epoch": 11.56032887975334,
+      "grad_norm": 0.04423193633556366,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 112482
+    },
+    {
+      "epoch": 11.560637204522097,
+      "grad_norm": 0.13389156758785248,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 112485
+    },
+    {
+      "epoch": 11.560945529290853,
+      "grad_norm": 0.05885486677289009,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 112488
+    },
+    {
+      "epoch": 11.56125385405961,
+      "grad_norm": 0.07096994668245316,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 112491
+    },
+    {
+      "epoch": 11.561562178828366,
+      "grad_norm": 0.04965658858418465,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 112494
+    },
+    {
+      "epoch": 11.561870503597122,
+      "grad_norm": 0.07426677644252777,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 112497
+    },
+    {
+      "epoch": 11.562178828365878,
+      "grad_norm": 0.060940932482481,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 112500
+    },
+    {
+      "epoch": 11.562487153134635,
+      "grad_norm": 0.03642065078020096,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 112503
+    },
+    {
+      "epoch": 11.562795477903391,
+      "grad_norm": 0.07831542938947678,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 112506
+    },
+    {
+      "epoch": 11.563103802672147,
+      "grad_norm": 0.05541325733065605,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 112509
+    },
+    {
+      "epoch": 11.563412127440904,
+      "grad_norm": 0.10018374770879745,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 112512
+    },
+    {
+      "epoch": 11.563720452209662,
+      "grad_norm": 0.06554833799600601,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 112515
+    },
+    {
+      "epoch": 11.564028776978418,
+      "grad_norm": 0.04541819915175438,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 112518
+    },
+    {
+      "epoch": 11.564337101747174,
+      "grad_norm": 0.08256267756223679,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 112521
+    },
+    {
+      "epoch": 11.56464542651593,
+      "grad_norm": 0.08194834738969803,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 112524
+    },
+    {
+      "epoch": 11.564953751284687,
+      "grad_norm": 0.0895160362124443,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 112527
+    },
+    {
+      "epoch": 11.565262076053443,
+      "grad_norm": 0.10424204915761948,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 112530
+    },
+    {
+      "epoch": 11.5655704008222,
+      "grad_norm": 0.07442378997802734,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 112533
+    },
+    {
+      "epoch": 11.565878725590956,
+      "grad_norm": 0.07817506790161133,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 112536
+    },
+    {
+      "epoch": 11.566187050359712,
+      "grad_norm": 0.10427911579608917,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 112539
+    },
+    {
+      "epoch": 11.566495375128468,
+      "grad_norm": 0.054581690579652786,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 112542
+    },
+    {
+      "epoch": 11.566803699897225,
+      "grad_norm": 0.04285728186368942,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 112545
+    },
+    {
+      "epoch": 11.567112024665981,
+      "grad_norm": 0.04124976322054863,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 112548
+    },
+    {
+      "epoch": 11.567420349434737,
+      "grad_norm": 0.04103301838040352,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 112551
+    },
+    {
+      "epoch": 11.567728674203494,
+      "grad_norm": 0.1087937131524086,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 112554
+    },
+    {
+      "epoch": 11.56803699897225,
+      "grad_norm": 0.06919058412313461,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 112557
+    },
+    {
+      "epoch": 11.568345323741006,
+      "grad_norm": 0.047698527574539185,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 112560
+    },
+    {
+      "epoch": 11.568653648509764,
+      "grad_norm": 0.04349555820226669,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 112563
+    },
+    {
+      "epoch": 11.56896197327852,
+      "grad_norm": 0.05030469968914986,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 112566
+    },
+    {
+      "epoch": 11.569270298047277,
+      "grad_norm": 0.11689770221710205,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 112569
+    },
+    {
+      "epoch": 11.569578622816033,
+      "grad_norm": 0.08256854861974716,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 112572
+    },
+    {
+      "epoch": 11.56988694758479,
+      "grad_norm": 0.052154459059238434,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 112575
+    },
+    {
+      "epoch": 11.570195272353546,
+      "grad_norm": 0.05697903782129288,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 112578
+    },
+    {
+      "epoch": 11.570503597122302,
+      "grad_norm": 0.04220142588019371,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 112581
+    },
+    {
+      "epoch": 11.570811921891059,
+      "grad_norm": 0.06774673610925674,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 112584
+    },
+    {
+      "epoch": 11.571120246659815,
+      "grad_norm": 0.07773800939321518,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 112587
+    },
+    {
+      "epoch": 11.571428571428571,
+      "grad_norm": 0.1131962239742279,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 112590
+    },
+    {
+      "epoch": 11.571736896197327,
+      "grad_norm": 0.18003375828266144,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 112593
+    },
+    {
+      "epoch": 11.572045220966084,
+      "grad_norm": 0.11840492486953735,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 112596
+    },
+    {
+      "epoch": 11.57235354573484,
+      "grad_norm": 0.05675830692052841,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 112599
+    },
+    {
+      "epoch": 11.572661870503596,
+      "grad_norm": 0.05000123009085655,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 112602
+    },
+    {
+      "epoch": 11.572970195272354,
+      "grad_norm": 0.03676938638091087,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 112605
+    },
+    {
+      "epoch": 11.57327852004111,
+      "grad_norm": 0.03664156049489975,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 112608
+    },
+    {
+      "epoch": 11.573586844809867,
+      "grad_norm": 0.036037057638168335,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 112611
+    },
+    {
+      "epoch": 11.573895169578623,
+      "grad_norm": 0.1087389811873436,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 112614
+    },
+    {
+      "epoch": 11.57420349434738,
+      "grad_norm": 0.035332124680280685,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 112617
+    },
+    {
+      "epoch": 11.574511819116136,
+      "grad_norm": 0.05903634428977966,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 112620
+    },
+    {
+      "epoch": 11.574820143884892,
+      "grad_norm": 0.13460128009319305,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 112623
+    },
+    {
+      "epoch": 11.575128468653649,
+      "grad_norm": 0.1596498340368271,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 112626
+    },
+    {
+      "epoch": 11.575436793422405,
+      "grad_norm": 0.07580243796110153,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 112629
+    },
+    {
+      "epoch": 11.575745118191161,
+      "grad_norm": 0.03767656907439232,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 112632
+    },
+    {
+      "epoch": 11.576053442959918,
+      "grad_norm": 0.03570094704627991,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 112635
+    },
+    {
+      "epoch": 11.576361767728674,
+      "grad_norm": 0.031302034854888916,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 112638
+    },
+    {
+      "epoch": 11.57667009249743,
+      "grad_norm": 0.049241434782743454,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 112641
+    },
+    {
+      "epoch": 11.576978417266186,
+      "grad_norm": 0.15020284056663513,
+      "learning_rate": 0.01,
+      "loss": 1.9032,
+      "step": 112644
+    },
+    {
+      "epoch": 11.577286742034943,
+      "grad_norm": 0.11612813919782639,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 112647
+    },
+    {
+      "epoch": 11.577595066803699,
+      "grad_norm": 0.08976644277572632,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 112650
+    },
+    {
+      "epoch": 11.577903391572455,
+      "grad_norm": 0.0828244537115097,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 112653
+    },
+    {
+      "epoch": 11.578211716341213,
+      "grad_norm": 0.0640794038772583,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 112656
+    },
+    {
+      "epoch": 11.57852004110997,
+      "grad_norm": 0.04499854892492294,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 112659
+    },
+    {
+      "epoch": 11.578828365878726,
+      "grad_norm": 0.03676529601216316,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 112662
+    },
+    {
+      "epoch": 11.579136690647482,
+      "grad_norm": 0.04806596413254738,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 112665
+    },
+    {
+      "epoch": 11.579445015416239,
+      "grad_norm": 0.11217772960662842,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 112668
+    },
+    {
+      "epoch": 11.579753340184995,
+      "grad_norm": 0.06657008826732635,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 112671
+    },
+    {
+      "epoch": 11.580061664953751,
+      "grad_norm": 0.1074846163392067,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 112674
+    },
+    {
+      "epoch": 11.580369989722508,
+      "grad_norm": 0.05520278215408325,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 112677
+    },
+    {
+      "epoch": 11.580678314491264,
+      "grad_norm": 0.055017780512571335,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 112680
+    },
+    {
+      "epoch": 11.58098663926002,
+      "grad_norm": 0.041063256561756134,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 112683
+    },
+    {
+      "epoch": 11.581294964028777,
+      "grad_norm": 0.0543600469827652,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 112686
+    },
+    {
+      "epoch": 11.581603288797533,
+      "grad_norm": 0.07964272797107697,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 112689
+    },
+    {
+      "epoch": 11.58191161356629,
+      "grad_norm": 0.10169428586959839,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 112692
+    },
+    {
+      "epoch": 11.582219938335045,
+      "grad_norm": 0.03501278534531593,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 112695
+    },
+    {
+      "epoch": 11.582528263103804,
+      "grad_norm": 0.08262136578559875,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 112698
+    },
+    {
+      "epoch": 11.58283658787256,
+      "grad_norm": 0.05497727543115616,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 112701
+    },
+    {
+      "epoch": 11.583144912641316,
+      "grad_norm": 0.07225098460912704,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 112704
+    },
+    {
+      "epoch": 11.583453237410072,
+      "grad_norm": 0.037529148161411285,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 112707
+    },
+    {
+      "epoch": 11.583761562178829,
+      "grad_norm": 0.029563959687948227,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 112710
+    },
+    {
+      "epoch": 11.584069886947585,
+      "grad_norm": 0.07537955790758133,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 112713
+    },
+    {
+      "epoch": 11.584378211716341,
+      "grad_norm": 0.06754820793867111,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 112716
+    },
+    {
+      "epoch": 11.584686536485098,
+      "grad_norm": 0.04394010081887245,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 112719
+    },
+    {
+      "epoch": 11.584994861253854,
+      "grad_norm": 0.05102475360035896,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 112722
+    },
+    {
+      "epoch": 11.58530318602261,
+      "grad_norm": 0.053613968193531036,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 112725
+    },
+    {
+      "epoch": 11.585611510791367,
+      "grad_norm": 0.11525541543960571,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 112728
+    },
+    {
+      "epoch": 11.585919835560123,
+      "grad_norm": 0.08309531211853027,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 112731
+    },
+    {
+      "epoch": 11.58622816032888,
+      "grad_norm": 0.07050877809524536,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 112734
+    },
+    {
+      "epoch": 11.586536485097636,
+      "grad_norm": 0.050906527787446976,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 112737
+    },
+    {
+      "epoch": 11.586844809866392,
+      "grad_norm": 0.05442078039050102,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 112740
+    },
+    {
+      "epoch": 11.587153134635148,
+      "grad_norm": 0.06000097841024399,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 112743
+    },
+    {
+      "epoch": 11.587461459403906,
+      "grad_norm": 0.0764593631029129,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 112746
+    },
+    {
+      "epoch": 11.587769784172663,
+      "grad_norm": 0.08361302316188812,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 112749
+    },
+    {
+      "epoch": 11.588078108941419,
+      "grad_norm": 0.07885933667421341,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 112752
+    },
+    {
+      "epoch": 11.588386433710175,
+      "grad_norm": 0.04270821809768677,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 112755
+    },
+    {
+      "epoch": 11.588694758478931,
+      "grad_norm": 0.117026187479496,
+      "learning_rate": 0.01,
+      "loss": 1.8946,
+      "step": 112758
+    },
+    {
+      "epoch": 11.589003083247688,
+      "grad_norm": 0.04440357908606529,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 112761
+    },
+    {
+      "epoch": 11.589311408016444,
+      "grad_norm": 0.05820045620203018,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 112764
+    },
+    {
+      "epoch": 11.5896197327852,
+      "grad_norm": 0.054034579545259476,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 112767
+    },
+    {
+      "epoch": 11.589928057553957,
+      "grad_norm": 0.09348484873771667,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 112770
+    },
+    {
+      "epoch": 11.590236382322713,
+      "grad_norm": 0.053756095468997955,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 112773
+    },
+    {
+      "epoch": 11.59054470709147,
+      "grad_norm": 0.057105984538793564,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 112776
+    },
+    {
+      "epoch": 11.590853031860226,
+      "grad_norm": 0.0945085659623146,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 112779
+    },
+    {
+      "epoch": 11.591161356628982,
+      "grad_norm": 0.07942793518304825,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 112782
+    },
+    {
+      "epoch": 11.591469681397738,
+      "grad_norm": 0.05878401920199394,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 112785
+    },
+    {
+      "epoch": 11.591778006166495,
+      "grad_norm": 0.09565435349941254,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 112788
+    },
+    {
+      "epoch": 11.592086330935253,
+      "grad_norm": 0.06520779430866241,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 112791
+    },
+    {
+      "epoch": 11.592394655704009,
+      "grad_norm": 0.07311799377202988,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 112794
+    },
+    {
+      "epoch": 11.592702980472765,
+      "grad_norm": 0.10966107249259949,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 112797
+    },
+    {
+      "epoch": 11.593011305241522,
+      "grad_norm": 0.03675894811749458,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 112800
+    },
+    {
+      "epoch": 11.593319630010278,
+      "grad_norm": 0.04241945967078209,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 112803
+    },
+    {
+      "epoch": 11.593627954779034,
+      "grad_norm": 0.12033431977033615,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 112806
+    },
+    {
+      "epoch": 11.59393627954779,
+      "grad_norm": 0.11073540151119232,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 112809
+    },
+    {
+      "epoch": 11.594244604316547,
+      "grad_norm": 0.06449783593416214,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 112812
+    },
+    {
+      "epoch": 11.594552929085303,
+      "grad_norm": 0.06169688701629639,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 112815
+    },
+    {
+      "epoch": 11.59486125385406,
+      "grad_norm": 0.03492598980665207,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 112818
+    },
+    {
+      "epoch": 11.595169578622816,
+      "grad_norm": 0.10270565748214722,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 112821
+    },
+    {
+      "epoch": 11.595477903391572,
+      "grad_norm": 0.09160374104976654,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 112824
+    },
+    {
+      "epoch": 11.595786228160328,
+      "grad_norm": 0.0630834549665451,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 112827
+    },
+    {
+      "epoch": 11.596094552929085,
+      "grad_norm": 0.09790679067373276,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 112830
+    },
+    {
+      "epoch": 11.596402877697841,
+      "grad_norm": 0.07473014295101166,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 112833
+    },
+    {
+      "epoch": 11.596711202466597,
+      "grad_norm": 0.043229445815086365,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 112836
+    },
+    {
+      "epoch": 11.597019527235355,
+      "grad_norm": 0.0716719701886177,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 112839
+    },
+    {
+      "epoch": 11.597327852004112,
+      "grad_norm": 0.038993120193481445,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 112842
+    },
+    {
+      "epoch": 11.597636176772868,
+      "grad_norm": 0.0744524821639061,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 112845
+    },
+    {
+      "epoch": 11.597944501541624,
+      "grad_norm": 0.08004055917263031,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 112848
+    },
+    {
+      "epoch": 11.59825282631038,
+      "grad_norm": 0.11334647238254547,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 112851
+    },
+    {
+      "epoch": 11.598561151079137,
+      "grad_norm": 0.10437542200088501,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 112854
+    },
+    {
+      "epoch": 11.598869475847893,
+      "grad_norm": 0.08015387505292892,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 112857
+    },
+    {
+      "epoch": 11.59917780061665,
+      "grad_norm": 0.05252983793616295,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 112860
+    },
+    {
+      "epoch": 11.599486125385406,
+      "grad_norm": 0.03896424174308777,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 112863
+    },
+    {
+      "epoch": 11.599794450154162,
+      "grad_norm": 0.061736226081848145,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 112866
+    },
+    {
+      "epoch": 11.600102774922918,
+      "grad_norm": 0.07320262491703033,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 112869
+    },
+    {
+      "epoch": 11.600411099691675,
+      "grad_norm": 0.07607041299343109,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 112872
+    },
+    {
+      "epoch": 11.600719424460431,
+      "grad_norm": 0.06408149749040604,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 112875
+    },
+    {
+      "epoch": 11.601027749229187,
+      "grad_norm": 0.10191265493631363,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 112878
+    },
+    {
+      "epoch": 11.601336073997945,
+      "grad_norm": 0.13209275901317596,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 112881
+    },
+    {
+      "epoch": 11.601644398766702,
+      "grad_norm": 0.07748264819383621,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 112884
+    },
+    {
+      "epoch": 11.601952723535458,
+      "grad_norm": 0.0734485611319542,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 112887
+    },
+    {
+      "epoch": 11.602261048304214,
+      "grad_norm": 0.05413933843374252,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 112890
+    },
+    {
+      "epoch": 11.60256937307297,
+      "grad_norm": 0.05536697059869766,
+      "learning_rate": 0.01,
+      "loss": 1.9848,
+      "step": 112893
+    },
+    {
+      "epoch": 11.602877697841727,
+      "grad_norm": 0.045468397438526154,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 112896
+    },
+    {
+      "epoch": 11.603186022610483,
+      "grad_norm": 0.10831259936094284,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 112899
+    },
+    {
+      "epoch": 11.60349434737924,
+      "grad_norm": 0.05811365321278572,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 112902
+    },
+    {
+      "epoch": 11.603802672147996,
+      "grad_norm": 0.04905383288860321,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 112905
+    },
+    {
+      "epoch": 11.604110996916752,
+      "grad_norm": 0.037075262516736984,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 112908
+    },
+    {
+      "epoch": 11.604419321685508,
+      "grad_norm": 0.0417492613196373,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 112911
+    },
+    {
+      "epoch": 11.604727646454265,
+      "grad_norm": 0.06051025539636612,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 112914
+    },
+    {
+      "epoch": 11.605035971223021,
+      "grad_norm": 0.11885872483253479,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 112917
+    },
+    {
+      "epoch": 11.605344295991777,
+      "grad_norm": 0.12744250893592834,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 112920
+    },
+    {
+      "epoch": 11.605652620760534,
+      "grad_norm": 0.05316381901502609,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 112923
+    },
+    {
+      "epoch": 11.60596094552929,
+      "grad_norm": 0.039563246071338654,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 112926
+    },
+    {
+      "epoch": 11.606269270298046,
+      "grad_norm": 0.0681186243891716,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 112929
+    },
+    {
+      "epoch": 11.606577595066804,
+      "grad_norm": 0.07162708789110184,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 112932
+    },
+    {
+      "epoch": 11.60688591983556,
+      "grad_norm": 0.06566008925437927,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 112935
+    },
+    {
+      "epoch": 11.607194244604317,
+      "grad_norm": 0.054131586104631424,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 112938
+    },
+    {
+      "epoch": 11.607502569373073,
+      "grad_norm": 0.037236664444208145,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 112941
+    },
+    {
+      "epoch": 11.60781089414183,
+      "grad_norm": 0.11102370172739029,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 112944
+    },
+    {
+      "epoch": 11.608119218910586,
+      "grad_norm": 0.06191109120845795,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 112947
+    },
+    {
+      "epoch": 11.608427543679342,
+      "grad_norm": 0.050658125430345535,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 112950
+    },
+    {
+      "epoch": 11.608735868448099,
+      "grad_norm": 0.03898501396179199,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 112953
+    },
+    {
+      "epoch": 11.609044193216855,
+      "grad_norm": 0.038435082882642746,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 112956
+    },
+    {
+      "epoch": 11.609352517985611,
+      "grad_norm": 0.08997193723917007,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 112959
+    },
+    {
+      "epoch": 11.609660842754367,
+      "grad_norm": 0.10880478471517563,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 112962
+    },
+    {
+      "epoch": 11.609969167523124,
+      "grad_norm": 0.06104841083288193,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 112965
+    },
+    {
+      "epoch": 11.61027749229188,
+      "grad_norm": 0.07381860911846161,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 112968
+    },
+    {
+      "epoch": 11.610585817060636,
+      "grad_norm": 0.11246202141046524,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 112971
+    },
+    {
+      "epoch": 11.610894141829394,
+      "grad_norm": 0.08369491249322891,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 112974
+    },
+    {
+      "epoch": 11.61120246659815,
+      "grad_norm": 0.034752074629068375,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 112977
+    },
+    {
+      "epoch": 11.611510791366907,
+      "grad_norm": 0.03296573460102081,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 112980
+    },
+    {
+      "epoch": 11.611819116135663,
+      "grad_norm": 0.07587042450904846,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 112983
+    },
+    {
+      "epoch": 11.61212744090442,
+      "grad_norm": 0.07403363287448883,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 112986
+    },
+    {
+      "epoch": 11.612435765673176,
+      "grad_norm": 0.0733148530125618,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 112989
+    },
+    {
+      "epoch": 11.612744090441932,
+      "grad_norm": 0.09333302825689316,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 112992
+    },
+    {
+      "epoch": 11.613052415210689,
+      "grad_norm": 0.06466648727655411,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 112995
+    },
+    {
+      "epoch": 11.613360739979445,
+      "grad_norm": 0.03636278957128525,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 112998
+    },
+    {
+      "epoch": 11.613669064748201,
+      "grad_norm": 0.12453662604093552,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 113001
+    },
+    {
+      "epoch": 11.613977389516958,
+      "grad_norm": 0.07740887999534607,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 113004
+    },
+    {
+      "epoch": 11.614285714285714,
+      "grad_norm": 0.057487133890390396,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 113007
+    },
+    {
+      "epoch": 11.61459403905447,
+      "grad_norm": 0.10658985376358032,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 113010
+    },
+    {
+      "epoch": 11.614902363823226,
+      "grad_norm": 0.10348264873027802,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 113013
+    },
+    {
+      "epoch": 11.615210688591983,
+      "grad_norm": 0.08679939061403275,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 113016
+    },
+    {
+      "epoch": 11.615519013360739,
+      "grad_norm": 0.04989555478096008,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 113019
+    },
+    {
+      "epoch": 11.615827338129497,
+      "grad_norm": 0.05296916887164116,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 113022
+    },
+    {
+      "epoch": 11.616135662898253,
+      "grad_norm": 0.06782415509223938,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 113025
+    },
+    {
+      "epoch": 11.61644398766701,
+      "grad_norm": 0.05802762880921364,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 113028
+    },
+    {
+      "epoch": 11.616752312435766,
+      "grad_norm": 0.048324454575777054,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 113031
+    },
+    {
+      "epoch": 11.617060637204522,
+      "grad_norm": 0.04696560651063919,
+      "learning_rate": 0.01,
+      "loss": 1.9646,
+      "step": 113034
+    },
+    {
+      "epoch": 11.617368961973279,
+      "grad_norm": 0.03296945244073868,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 113037
+    },
+    {
+      "epoch": 11.617677286742035,
+      "grad_norm": 0.07478760182857513,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 113040
+    },
+    {
+      "epoch": 11.617985611510791,
+      "grad_norm": 0.11230601370334625,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 113043
+    },
+    {
+      "epoch": 11.618293936279548,
+      "grad_norm": 0.0838407501578331,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 113046
+    },
+    {
+      "epoch": 11.618602261048304,
+      "grad_norm": 0.09206885099411011,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 113049
+    },
+    {
+      "epoch": 11.61891058581706,
+      "grad_norm": 0.04248971492052078,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 113052
+    },
+    {
+      "epoch": 11.619218910585817,
+      "grad_norm": 0.038236189633607864,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 113055
+    },
+    {
+      "epoch": 11.619527235354573,
+      "grad_norm": 0.031842999160289764,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 113058
+    },
+    {
+      "epoch": 11.61983556012333,
+      "grad_norm": 0.030857548117637634,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 113061
+    },
+    {
+      "epoch": 11.620143884892087,
+      "grad_norm": 0.13097026944160461,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 113064
+    },
+    {
+      "epoch": 11.620452209660844,
+      "grad_norm": 0.05664094164967537,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 113067
+    },
+    {
+      "epoch": 11.6207605344296,
+      "grad_norm": 0.09321857243776321,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 113070
+    },
+    {
+      "epoch": 11.621068859198356,
+      "grad_norm": 0.10250440984964371,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 113073
+    },
+    {
+      "epoch": 11.621377183967113,
+      "grad_norm": 0.05741477757692337,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 113076
+    },
+    {
+      "epoch": 11.621685508735869,
+      "grad_norm": 0.036703143268823624,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 113079
+    },
+    {
+      "epoch": 11.621993833504625,
+      "grad_norm": 0.038881104439496994,
+      "learning_rate": 0.01,
+      "loss": 1.9708,
+      "step": 113082
+    },
+    {
+      "epoch": 11.622302158273381,
+      "grad_norm": 0.04341772571206093,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 113085
+    },
+    {
+      "epoch": 11.622610483042138,
+      "grad_norm": 0.04066208004951477,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 113088
+    },
+    {
+      "epoch": 11.622918807810894,
+      "grad_norm": 0.038506485521793365,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 113091
+    },
+    {
+      "epoch": 11.62322713257965,
+      "grad_norm": 0.11907859146595001,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 113094
+    },
+    {
+      "epoch": 11.623535457348407,
+      "grad_norm": 0.050517939031124115,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 113097
+    },
+    {
+      "epoch": 11.623843782117163,
+      "grad_norm": 0.07157903164625168,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 113100
+    },
+    {
+      "epoch": 11.62415210688592,
+      "grad_norm": 0.03911276534199715,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 113103
+    },
+    {
+      "epoch": 11.624460431654676,
+      "grad_norm": 0.06826124340295792,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 113106
+    },
+    {
+      "epoch": 11.624768756423432,
+      "grad_norm": 0.08034567534923553,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 113109
+    },
+    {
+      "epoch": 11.625077081192188,
+      "grad_norm": 0.04130208492279053,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 113112
+    },
+    {
+      "epoch": 11.625385405960946,
+      "grad_norm": 0.0840260237455368,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 113115
+    },
+    {
+      "epoch": 11.625693730729703,
+      "grad_norm": 0.03418037295341492,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 113118
+    },
+    {
+      "epoch": 11.626002055498459,
+      "grad_norm": 0.14465217292308807,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 113121
+    },
+    {
+      "epoch": 11.626310380267215,
+      "grad_norm": 0.11267710477113724,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 113124
+    },
+    {
+      "epoch": 11.626618705035972,
+      "grad_norm": 0.07997629046440125,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 113127
+    },
+    {
+      "epoch": 11.626927029804728,
+      "grad_norm": 0.05365961045026779,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 113130
+    },
+    {
+      "epoch": 11.627235354573484,
+      "grad_norm": 0.0491030178964138,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 113133
+    },
+    {
+      "epoch": 11.62754367934224,
+      "grad_norm": 0.04414552450180054,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 113136
+    },
+    {
+      "epoch": 11.627852004110997,
+      "grad_norm": 0.03378671780228615,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 113139
+    },
+    {
+      "epoch": 11.628160328879753,
+      "grad_norm": 0.05612415820360184,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 113142
+    },
+    {
+      "epoch": 11.62846865364851,
+      "grad_norm": 0.06220652535557747,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 113145
+    },
+    {
+      "epoch": 11.628776978417266,
+      "grad_norm": 0.05417797341942787,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 113148
+    },
+    {
+      "epoch": 11.629085303186022,
+      "grad_norm": 0.07151927798986435,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 113151
+    },
+    {
+      "epoch": 11.629393627954778,
+      "grad_norm": 0.06492269039154053,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 113154
+    },
+    {
+      "epoch": 11.629701952723536,
+      "grad_norm": 0.1323765516281128,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 113157
+    },
+    {
+      "epoch": 11.630010277492293,
+      "grad_norm": 0.06621003150939941,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 113160
+    },
+    {
+      "epoch": 11.630318602261049,
+      "grad_norm": 0.0659116730093956,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 113163
+    },
+    {
+      "epoch": 11.630626927029805,
+      "grad_norm": 0.09065114706754684,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 113166
+    },
+    {
+      "epoch": 11.630935251798562,
+      "grad_norm": 0.0618029460310936,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 113169
+    },
+    {
+      "epoch": 11.631243576567318,
+      "grad_norm": 0.078899085521698,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 113172
+    },
+    {
+      "epoch": 11.631551901336074,
+      "grad_norm": 0.07569435238838196,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 113175
+    },
+    {
+      "epoch": 11.63186022610483,
+      "grad_norm": 0.06244001165032387,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 113178
+    },
+    {
+      "epoch": 11.632168550873587,
+      "grad_norm": 0.12835216522216797,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 113181
+    },
+    {
+      "epoch": 11.632476875642343,
+      "grad_norm": 0.04971390590071678,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 113184
+    },
+    {
+      "epoch": 11.6327852004111,
+      "grad_norm": 0.08454111963510513,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 113187
+    },
+    {
+      "epoch": 11.633093525179856,
+      "grad_norm": 0.03988140821456909,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 113190
+    },
+    {
+      "epoch": 11.633401849948612,
+      "grad_norm": 0.048893485218286514,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 113193
+    },
+    {
+      "epoch": 11.633710174717368,
+      "grad_norm": 0.09177813678979874,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 113196
+    },
+    {
+      "epoch": 11.634018499486125,
+      "grad_norm": 0.04464666545391083,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 113199
+    },
+    {
+      "epoch": 11.634326824254881,
+      "grad_norm": 0.14179858565330505,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 113202
+    },
+    {
+      "epoch": 11.634635149023639,
+      "grad_norm": 0.10474514216184616,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 113205
+    },
+    {
+      "epoch": 11.634943473792395,
+      "grad_norm": 0.05921778455376625,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 113208
+    },
+    {
+      "epoch": 11.635251798561152,
+      "grad_norm": 0.045888129621744156,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 113211
+    },
+    {
+      "epoch": 11.635560123329908,
+      "grad_norm": 0.06556089222431183,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 113214
+    },
+    {
+      "epoch": 11.635868448098664,
+      "grad_norm": 0.044321537017822266,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 113217
+    },
+    {
+      "epoch": 11.63617677286742,
+      "grad_norm": 0.03785913437604904,
+      "learning_rate": 0.01,
+      "loss": 1.9752,
+      "step": 113220
+    },
+    {
+      "epoch": 11.636485097636177,
+      "grad_norm": 0.11449971050024033,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 113223
+    },
+    {
+      "epoch": 11.636793422404933,
+      "grad_norm": 0.0937795415520668,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 113226
+    },
+    {
+      "epoch": 11.63710174717369,
+      "grad_norm": 0.11280496418476105,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 113229
+    },
+    {
+      "epoch": 11.637410071942446,
+      "grad_norm": 0.09897052496671677,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 113232
+    },
+    {
+      "epoch": 11.637718396711202,
+      "grad_norm": 0.08094508945941925,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 113235
+    },
+    {
+      "epoch": 11.638026721479958,
+      "grad_norm": 0.0689573660492897,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 113238
+    },
+    {
+      "epoch": 11.638335046248715,
+      "grad_norm": 0.05370873212814331,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 113241
+    },
+    {
+      "epoch": 11.638643371017471,
+      "grad_norm": 0.06898577511310577,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 113244
+    },
+    {
+      "epoch": 11.638951695786227,
+      "grad_norm": 0.04871980473399162,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 113247
+    },
+    {
+      "epoch": 11.639260020554985,
+      "grad_norm": 0.05161462724208832,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 113250
+    },
+    {
+      "epoch": 11.639568345323742,
+      "grad_norm": 0.055129531770944595,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 113253
+    },
+    {
+      "epoch": 11.639876670092498,
+      "grad_norm": 0.14611971378326416,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 113256
+    },
+    {
+      "epoch": 11.640184994861254,
+      "grad_norm": 0.1649077832698822,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 113259
+    },
+    {
+      "epoch": 11.64049331963001,
+      "grad_norm": 0.04667672887444496,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 113262
+    },
+    {
+      "epoch": 11.640801644398767,
+      "grad_norm": 0.03393012285232544,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 113265
+    },
+    {
+      "epoch": 11.641109969167523,
+      "grad_norm": 0.04141303524374962,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 113268
+    },
+    {
+      "epoch": 11.64141829393628,
+      "grad_norm": 0.0544479675590992,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 113271
+    },
+    {
+      "epoch": 11.641726618705036,
+      "grad_norm": 0.0461701825261116,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 113274
+    },
+    {
+      "epoch": 11.642034943473792,
+      "grad_norm": 0.10449957847595215,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 113277
+    },
+    {
+      "epoch": 11.642343268242549,
+      "grad_norm": 0.07522600889205933,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 113280
+    },
+    {
+      "epoch": 11.642651593011305,
+      "grad_norm": 0.06570401787757874,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 113283
+    },
+    {
+      "epoch": 11.642959917780061,
+      "grad_norm": 0.08367069810628891,
+      "learning_rate": 0.01,
+      "loss": 1.8957,
+      "step": 113286
+    },
+    {
+      "epoch": 11.643268242548817,
+      "grad_norm": 0.08211222290992737,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 113289
+    },
+    {
+      "epoch": 11.643576567317574,
+      "grad_norm": 0.05755218118429184,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 113292
+    },
+    {
+      "epoch": 11.64388489208633,
+      "grad_norm": 0.057840488851070404,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 113295
+    },
+    {
+      "epoch": 11.644193216855088,
+      "grad_norm": 0.05402502417564392,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 113298
+    },
+    {
+      "epoch": 11.644501541623844,
+      "grad_norm": 0.09462658315896988,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 113301
+    },
+    {
+      "epoch": 11.6448098663926,
+      "grad_norm": 0.05124598741531372,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 113304
+    },
+    {
+      "epoch": 11.645118191161357,
+      "grad_norm": 0.04224373400211334,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 113307
+    },
+    {
+      "epoch": 11.645426515930113,
+      "grad_norm": 0.0728943794965744,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 113310
+    },
+    {
+      "epoch": 11.64573484069887,
+      "grad_norm": 0.11200623214244843,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 113313
+    },
+    {
+      "epoch": 11.646043165467626,
+      "grad_norm": 0.08526485413312912,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 113316
+    },
+    {
+      "epoch": 11.646351490236382,
+      "grad_norm": 0.10539497435092926,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 113319
+    },
+    {
+      "epoch": 11.646659815005139,
+      "grad_norm": 0.07505021244287491,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 113322
+    },
+    {
+      "epoch": 11.646968139773895,
+      "grad_norm": 0.06336735934019089,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 113325
+    },
+    {
+      "epoch": 11.647276464542651,
+      "grad_norm": 0.07473690062761307,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 113328
+    },
+    {
+      "epoch": 11.647584789311408,
+      "grad_norm": 0.09957322478294373,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 113331
+    },
+    {
+      "epoch": 11.647893114080164,
+      "grad_norm": 0.028184523805975914,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 113334
+    },
+    {
+      "epoch": 11.64820143884892,
+      "grad_norm": 0.03983794525265694,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 113337
+    },
+    {
+      "epoch": 11.648509763617678,
+      "grad_norm": 0.13572625815868378,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 113340
+    },
+    {
+      "epoch": 11.648818088386435,
+      "grad_norm": 0.12764644622802734,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 113343
+    },
+    {
+      "epoch": 11.64912641315519,
+      "grad_norm": 0.08703610301017761,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 113346
+    },
+    {
+      "epoch": 11.649434737923947,
+      "grad_norm": 0.04160364344716072,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 113349
+    },
+    {
+      "epoch": 11.649743062692703,
+      "grad_norm": 0.03464055061340332,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 113352
+    },
+    {
+      "epoch": 11.65005138746146,
+      "grad_norm": 0.03290557116270065,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 113355
+    },
+    {
+      "epoch": 11.650359712230216,
+      "grad_norm": 0.03284400701522827,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 113358
+    },
+    {
+      "epoch": 11.650668036998972,
+      "grad_norm": 0.04168650880455971,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 113361
+    },
+    {
+      "epoch": 11.650976361767729,
+      "grad_norm": 0.05013224482536316,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 113364
+    },
+    {
+      "epoch": 11.651284686536485,
+      "grad_norm": 0.12323949486017227,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 113367
+    },
+    {
+      "epoch": 11.651593011305241,
+      "grad_norm": 0.11722767353057861,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 113370
+    },
+    {
+      "epoch": 11.651901336073998,
+      "grad_norm": 0.1257123053073883,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 113373
+    },
+    {
+      "epoch": 11.652209660842754,
+      "grad_norm": 0.09780552238225937,
+      "learning_rate": 0.01,
+      "loss": 1.9669,
+      "step": 113376
+    },
+    {
+      "epoch": 11.65251798561151,
+      "grad_norm": 0.0467044822871685,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 113379
+    },
+    {
+      "epoch": 11.652826310380267,
+      "grad_norm": 0.04701404646039009,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 113382
+    },
+    {
+      "epoch": 11.653134635149023,
+      "grad_norm": 0.04993510991334915,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 113385
+    },
+    {
+      "epoch": 11.65344295991778,
+      "grad_norm": 0.03722843900322914,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 113388
+    },
+    {
+      "epoch": 11.653751284686537,
+      "grad_norm": 0.03551566228270531,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 113391
+    },
+    {
+      "epoch": 11.654059609455294,
+      "grad_norm": 0.11122262477874756,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 113394
+    },
+    {
+      "epoch": 11.65436793422405,
+      "grad_norm": 0.06826326251029968,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 113397
+    },
+    {
+      "epoch": 11.654676258992806,
+      "grad_norm": 0.155528724193573,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 113400
+    },
+    {
+      "epoch": 11.654984583761562,
+      "grad_norm": 0.04919079691171646,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 113403
+    },
+    {
+      "epoch": 11.655292908530319,
+      "grad_norm": 0.06651998311281204,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 113406
+    },
+    {
+      "epoch": 11.655601233299075,
+      "grad_norm": 0.05976448580622673,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 113409
+    },
+    {
+      "epoch": 11.655909558067831,
+      "grad_norm": 0.04969658702611923,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 113412
+    },
+    {
+      "epoch": 11.656217882836588,
+      "grad_norm": 0.10790357738733292,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 113415
+    },
+    {
+      "epoch": 11.656526207605344,
+      "grad_norm": 0.06400740146636963,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 113418
+    },
+    {
+      "epoch": 11.6568345323741,
+      "grad_norm": 0.11117127537727356,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 113421
+    },
+    {
+      "epoch": 11.657142857142857,
+      "grad_norm": 0.08646197617053986,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 113424
+    },
+    {
+      "epoch": 11.657451181911613,
+      "grad_norm": 0.11909831315279007,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 113427
+    },
+    {
+      "epoch": 11.65775950668037,
+      "grad_norm": 0.05872794985771179,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 113430
+    },
+    {
+      "epoch": 11.658067831449127,
+      "grad_norm": 0.0353827103972435,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 113433
+    },
+    {
+      "epoch": 11.658376156217884,
+      "grad_norm": 0.09279036521911621,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 113436
+    },
+    {
+      "epoch": 11.65868448098664,
+      "grad_norm": 0.04595791921019554,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 113439
+    },
+    {
+      "epoch": 11.658992805755396,
+      "grad_norm": 0.08540619164705276,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 113442
+    },
+    {
+      "epoch": 11.659301130524153,
+      "grad_norm": 0.0668819323182106,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 113445
+    },
+    {
+      "epoch": 11.659609455292909,
+      "grad_norm": 0.09538742154836655,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 113448
+    },
+    {
+      "epoch": 11.659917780061665,
+      "grad_norm": 0.06846621632575989,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 113451
+    },
+    {
+      "epoch": 11.660226104830421,
+      "grad_norm": 0.05695407837629318,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 113454
+    },
+    {
+      "epoch": 11.660534429599178,
+      "grad_norm": 0.056524164974689484,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 113457
+    },
+    {
+      "epoch": 11.660842754367934,
+      "grad_norm": 0.034500472247600555,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 113460
+    },
+    {
+      "epoch": 11.66115107913669,
+      "grad_norm": 0.046678464859724045,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 113463
+    },
+    {
+      "epoch": 11.661459403905447,
+      "grad_norm": 0.06171998009085655,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 113466
+    },
+    {
+      "epoch": 11.661767728674203,
+      "grad_norm": 0.13738484680652618,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 113469
+    },
+    {
+      "epoch": 11.66207605344296,
+      "grad_norm": 0.07852083444595337,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 113472
+    },
+    {
+      "epoch": 11.662384378211716,
+      "grad_norm": 0.04132244363427162,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 113475
+    },
+    {
+      "epoch": 11.662692702980472,
+      "grad_norm": 0.04207141324877739,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 113478
+    },
+    {
+      "epoch": 11.66300102774923,
+      "grad_norm": 0.03948988765478134,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 113481
+    },
+    {
+      "epoch": 11.663309352517986,
+      "grad_norm": 0.04067898541688919,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 113484
+    },
+    {
+      "epoch": 11.663617677286743,
+      "grad_norm": 0.03814522549510002,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 113487
+    },
+    {
+      "epoch": 11.663926002055499,
+      "grad_norm": 0.0856989324092865,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 113490
+    },
+    {
+      "epoch": 11.664234326824255,
+      "grad_norm": 0.06945549696683884,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 113493
+    },
+    {
+      "epoch": 11.664542651593012,
+      "grad_norm": 0.04732196778059006,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 113496
+    },
+    {
+      "epoch": 11.664850976361768,
+      "grad_norm": 0.0411718524992466,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 113499
+    },
+    {
+      "epoch": 11.665159301130524,
+      "grad_norm": 0.05704187601804733,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 113502
+    },
+    {
+      "epoch": 11.66546762589928,
+      "grad_norm": 0.10712847858667374,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 113505
+    },
+    {
+      "epoch": 11.665775950668037,
+      "grad_norm": 0.09478617459535599,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 113508
+    },
+    {
+      "epoch": 11.666084275436793,
+      "grad_norm": 0.0557820163667202,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 113511
+    },
+    {
+      "epoch": 11.66639260020555,
+      "grad_norm": 0.04023374244570732,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 113514
+    },
+    {
+      "epoch": 11.666700924974306,
+      "grad_norm": 0.09300129860639572,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 113517
+    },
+    {
+      "epoch": 11.667009249743062,
+      "grad_norm": 0.07823445647954941,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 113520
+    },
+    {
+      "epoch": 11.66731757451182,
+      "grad_norm": 0.17606739699840546,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 113523
+    },
+    {
+      "epoch": 11.667625899280576,
+      "grad_norm": 0.07043785601854324,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 113526
+    },
+    {
+      "epoch": 11.667934224049333,
+      "grad_norm": 0.0512007512152195,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 113529
+    },
+    {
+      "epoch": 11.668242548818089,
+      "grad_norm": 0.04862508177757263,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 113532
+    },
+    {
+      "epoch": 11.668550873586845,
+      "grad_norm": 0.05967811495065689,
+      "learning_rate": 0.01,
+      "loss": 1.8976,
+      "step": 113535
+    },
+    {
+      "epoch": 11.668859198355602,
+      "grad_norm": 0.05780831724405289,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 113538
+    },
+    {
+      "epoch": 11.669167523124358,
+      "grad_norm": 0.06208907067775726,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 113541
+    },
+    {
+      "epoch": 11.669475847893114,
+      "grad_norm": 0.07732394337654114,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 113544
+    },
+    {
+      "epoch": 11.66978417266187,
+      "grad_norm": 0.09532228112220764,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 113547
+    },
+    {
+      "epoch": 11.670092497430627,
+      "grad_norm": 0.06125938892364502,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 113550
+    },
+    {
+      "epoch": 11.670400822199383,
+      "grad_norm": 0.09997191280126572,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 113553
+    },
+    {
+      "epoch": 11.67070914696814,
+      "grad_norm": 0.06977102160453796,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 113556
+    },
+    {
+      "epoch": 11.671017471736896,
+      "grad_norm": 0.04314795881509781,
+      "learning_rate": 0.01,
+      "loss": 1.8992,
+      "step": 113559
+    },
+    {
+      "epoch": 11.671325796505652,
+      "grad_norm": 0.06881061941385269,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 113562
+    },
+    {
+      "epoch": 11.671634121274408,
+      "grad_norm": 0.07183831930160522,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 113565
+    },
+    {
+      "epoch": 11.671942446043165,
+      "grad_norm": 0.0504421591758728,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 113568
+    },
+    {
+      "epoch": 11.672250770811921,
+      "grad_norm": 0.09572986513376236,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 113571
+    },
+    {
+      "epoch": 11.672559095580679,
+      "grad_norm": 0.07264112681150436,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 113574
+    },
+    {
+      "epoch": 11.672867420349435,
+      "grad_norm": 0.11083868145942688,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 113577
+    },
+    {
+      "epoch": 11.673175745118192,
+      "grad_norm": 0.04253198578953743,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 113580
+    },
+    {
+      "epoch": 11.673484069886948,
+      "grad_norm": 0.06747709959745407,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 113583
+    },
+    {
+      "epoch": 11.673792394655704,
+      "grad_norm": 0.045510247349739075,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 113586
+    },
+    {
+      "epoch": 11.67410071942446,
+      "grad_norm": 0.05232301354408264,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 113589
+    },
+    {
+      "epoch": 11.674409044193217,
+      "grad_norm": 0.09999211132526398,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 113592
+    },
+    {
+      "epoch": 11.674717368961973,
+      "grad_norm": 0.14111940562725067,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 113595
+    },
+    {
+      "epoch": 11.67502569373073,
+      "grad_norm": 0.050830744206905365,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 113598
+    },
+    {
+      "epoch": 11.675334018499486,
+      "grad_norm": 0.0898144394159317,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 113601
+    },
+    {
+      "epoch": 11.675642343268242,
+      "grad_norm": 0.04930245876312256,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 113604
+    },
+    {
+      "epoch": 11.675950668036998,
+      "grad_norm": 0.08487803488969803,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 113607
+    },
+    {
+      "epoch": 11.676258992805755,
+      "grad_norm": 0.0673738345503807,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 113610
+    },
+    {
+      "epoch": 11.676567317574511,
+      "grad_norm": 0.09414537996053696,
+      "learning_rate": 0.01,
+      "loss": 1.9719,
+      "step": 113613
+    },
+    {
+      "epoch": 11.67687564234327,
+      "grad_norm": 0.04949101433157921,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 113616
+    },
+    {
+      "epoch": 11.677183967112025,
+      "grad_norm": 0.09853395819664001,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 113619
+    },
+    {
+      "epoch": 11.677492291880782,
+      "grad_norm": 0.08439648151397705,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 113622
+    },
+    {
+      "epoch": 11.677800616649538,
+      "grad_norm": 0.04191217198967934,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 113625
+    },
+    {
+      "epoch": 11.678108941418294,
+      "grad_norm": 0.04377821460366249,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 113628
+    },
+    {
+      "epoch": 11.67841726618705,
+      "grad_norm": 0.055352889001369476,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 113631
+    },
+    {
+      "epoch": 11.678725590955807,
+      "grad_norm": 0.05160324648022652,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 113634
+    },
+    {
+      "epoch": 11.679033915724563,
+      "grad_norm": 0.047420211136341095,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 113637
+    },
+    {
+      "epoch": 11.67934224049332,
+      "grad_norm": 0.12467961758375168,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 113640
+    },
+    {
+      "epoch": 11.679650565262076,
+      "grad_norm": 0.06575851887464523,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 113643
+    },
+    {
+      "epoch": 11.679958890030832,
+      "grad_norm": 0.05539676174521446,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 113646
+    },
+    {
+      "epoch": 11.680267214799589,
+      "grad_norm": 0.09138210862874985,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 113649
+    },
+    {
+      "epoch": 11.680575539568345,
+      "grad_norm": 0.06900908797979355,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 113652
+    },
+    {
+      "epoch": 11.680883864337101,
+      "grad_norm": 0.06584549695253372,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 113655
+    },
+    {
+      "epoch": 11.681192189105857,
+      "grad_norm": 0.08189304172992706,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 113658
+    },
+    {
+      "epoch": 11.681500513874614,
+      "grad_norm": 0.050689902156591415,
+      "learning_rate": 0.01,
+      "loss": 1.9677,
+      "step": 113661
+    },
+    {
+      "epoch": 11.681808838643372,
+      "grad_norm": 0.09859631955623627,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 113664
+    },
+    {
+      "epoch": 11.682117163412128,
+      "grad_norm": 0.04112638160586357,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 113667
+    },
+    {
+      "epoch": 11.682425488180884,
+      "grad_norm": 0.08223004639148712,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 113670
+    },
+    {
+      "epoch": 11.68273381294964,
+      "grad_norm": 0.10968174785375595,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 113673
+    },
+    {
+      "epoch": 11.683042137718397,
+      "grad_norm": 0.11927396804094315,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 113676
+    },
+    {
+      "epoch": 11.683350462487153,
+      "grad_norm": 0.058725710958242416,
+      "learning_rate": 0.01,
+      "loss": 1.9773,
+      "step": 113679
+    },
+    {
+      "epoch": 11.68365878725591,
+      "grad_norm": 0.06128184124827385,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 113682
+    },
+    {
+      "epoch": 11.683967112024666,
+      "grad_norm": 0.036178167909383774,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 113685
+    },
+    {
+      "epoch": 11.684275436793422,
+      "grad_norm": 0.03075682558119297,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 113688
+    },
+    {
+      "epoch": 11.684583761562179,
+      "grad_norm": 0.11864864081144333,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 113691
+    },
+    {
+      "epoch": 11.684892086330935,
+      "grad_norm": 0.045278266072273254,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 113694
+    },
+    {
+      "epoch": 11.685200411099691,
+      "grad_norm": 0.07865426689386368,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 113697
+    },
+    {
+      "epoch": 11.685508735868448,
+      "grad_norm": 0.03860568627715111,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 113700
+    },
+    {
+      "epoch": 11.685817060637204,
+      "grad_norm": 0.06591783463954926,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 113703
+    },
+    {
+      "epoch": 11.68612538540596,
+      "grad_norm": 0.04285858944058418,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 113706
+    },
+    {
+      "epoch": 11.686433710174718,
+      "grad_norm": 0.04691053181886673,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 113709
+    },
+    {
+      "epoch": 11.686742034943475,
+      "grad_norm": 0.05432582274079323,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 113712
+    },
+    {
+      "epoch": 11.68705035971223,
+      "grad_norm": 0.048061177134513855,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 113715
+    },
+    {
+      "epoch": 11.687358684480987,
+      "grad_norm": 0.04626024141907692,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 113718
+    },
+    {
+      "epoch": 11.687667009249743,
+      "grad_norm": 0.08235618472099304,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 113721
+    },
+    {
+      "epoch": 11.6879753340185,
+      "grad_norm": 0.11195166409015656,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 113724
+    },
+    {
+      "epoch": 11.688283658787256,
+      "grad_norm": 0.07571295648813248,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 113727
+    },
+    {
+      "epoch": 11.688591983556012,
+      "grad_norm": 0.08173639327287674,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 113730
+    },
+    {
+      "epoch": 11.688900308324769,
+      "grad_norm": 0.06790082156658173,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 113733
+    },
+    {
+      "epoch": 11.689208633093525,
+      "grad_norm": 0.09619564563035965,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 113736
+    },
+    {
+      "epoch": 11.689516957862281,
+      "grad_norm": 0.06219369173049927,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 113739
+    },
+    {
+      "epoch": 11.689825282631038,
+      "grad_norm": 0.12838895618915558,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 113742
+    },
+    {
+      "epoch": 11.690133607399794,
+      "grad_norm": 0.12423622608184814,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 113745
+    },
+    {
+      "epoch": 11.69044193216855,
+      "grad_norm": 0.13346002995967865,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 113748
+    },
+    {
+      "epoch": 11.690750256937307,
+      "grad_norm": 0.13693243265151978,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 113751
+    },
+    {
+      "epoch": 11.691058581706063,
+      "grad_norm": 0.049979254603385925,
+      "learning_rate": 0.01,
+      "loss": 1.9027,
+      "step": 113754
+    },
+    {
+      "epoch": 11.691366906474821,
+      "grad_norm": 0.0416957326233387,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 113757
+    },
+    {
+      "epoch": 11.691675231243577,
+      "grad_norm": 0.05410477891564369,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 113760
+    },
+    {
+      "epoch": 11.691983556012334,
+      "grad_norm": 0.03738997131586075,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 113763
+    },
+    {
+      "epoch": 11.69229188078109,
+      "grad_norm": 0.03813576325774193,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 113766
+    },
+    {
+      "epoch": 11.692600205549846,
+      "grad_norm": 0.04307515546679497,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 113769
+    },
+    {
+      "epoch": 11.692908530318602,
+      "grad_norm": 0.053854089230298996,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 113772
+    },
+    {
+      "epoch": 11.693216855087359,
+      "grad_norm": 0.09748701006174088,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 113775
+    },
+    {
+      "epoch": 11.693525179856115,
+      "grad_norm": 0.08756256103515625,
+      "learning_rate": 0.01,
+      "loss": 1.9728,
+      "step": 113778
+    },
+    {
+      "epoch": 11.693833504624871,
+      "grad_norm": 0.1275547593832016,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 113781
+    },
+    {
+      "epoch": 11.694141829393628,
+      "grad_norm": 0.13405336439609528,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 113784
+    },
+    {
+      "epoch": 11.694450154162384,
+      "grad_norm": 0.15136772394180298,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 113787
+    },
+    {
+      "epoch": 11.69475847893114,
+      "grad_norm": 0.09638786315917969,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 113790
+    },
+    {
+      "epoch": 11.695066803699897,
+      "grad_norm": 0.09290369600057602,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 113793
+    },
+    {
+      "epoch": 11.695375128468653,
+      "grad_norm": 0.10820170491933823,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 113796
+    },
+    {
+      "epoch": 11.695683453237411,
+      "grad_norm": 0.047729067504405975,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 113799
+    },
+    {
+      "epoch": 11.695991778006167,
+      "grad_norm": 0.10129518061876297,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 113802
+    },
+    {
+      "epoch": 11.696300102774924,
+      "grad_norm": 0.0901324599981308,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 113805
+    },
+    {
+      "epoch": 11.69660842754368,
+      "grad_norm": 0.0981982871890068,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 113808
+    },
+    {
+      "epoch": 11.696916752312436,
+      "grad_norm": 0.06594400107860565,
+      "learning_rate": 0.01,
+      "loss": 1.9788,
+      "step": 113811
+    },
+    {
+      "epoch": 11.697225077081193,
+      "grad_norm": 0.035697147250175476,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 113814
+    },
+    {
+      "epoch": 11.697533401849949,
+      "grad_norm": 0.0464530885219574,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 113817
+    },
+    {
+      "epoch": 11.697841726618705,
+      "grad_norm": 0.04814977943897247,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 113820
+    },
+    {
+      "epoch": 11.698150051387461,
+      "grad_norm": 0.07865633070468903,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 113823
+    },
+    {
+      "epoch": 11.698458376156218,
+      "grad_norm": 0.053309258073568344,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 113826
+    },
+    {
+      "epoch": 11.698766700924974,
+      "grad_norm": 0.0638638436794281,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 113829
+    },
+    {
+      "epoch": 11.69907502569373,
+      "grad_norm": 0.11192189902067184,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 113832
+    },
+    {
+      "epoch": 11.699383350462487,
+      "grad_norm": 0.09100265055894852,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 113835
+    },
+    {
+      "epoch": 11.699691675231243,
+      "grad_norm": 0.06564506888389587,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 113838
+    },
+    {
+      "epoch": 11.7,
+      "grad_norm": 0.06351829320192337,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 113841
+    },
+    {
+      "epoch": 11.700308324768756,
+      "grad_norm": 0.0480794832110405,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 113844
+    },
+    {
+      "epoch": 11.700616649537512,
+      "grad_norm": 0.036630935966968536,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 113847
+    },
+    {
+      "epoch": 11.70092497430627,
+      "grad_norm": 0.09170176088809967,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 113850
+    },
+    {
+      "epoch": 11.701233299075026,
+      "grad_norm": 0.10979105532169342,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 113853
+    },
+    {
+      "epoch": 11.701541623843783,
+      "grad_norm": 0.13954408466815948,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 113856
+    },
+    {
+      "epoch": 11.701849948612539,
+      "grad_norm": 0.0569770522415638,
+      "learning_rate": 0.01,
+      "loss": 1.965,
+      "step": 113859
+    },
+    {
+      "epoch": 11.702158273381295,
+      "grad_norm": 0.054077599197626114,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 113862
+    },
+    {
+      "epoch": 11.702466598150052,
+      "grad_norm": 0.042730607092380524,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 113865
+    },
+    {
+      "epoch": 11.702774922918808,
+      "grad_norm": 0.0396852046251297,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 113868
+    },
+    {
+      "epoch": 11.703083247687564,
+      "grad_norm": 0.04677654057741165,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 113871
+    },
+    {
+      "epoch": 11.70339157245632,
+      "grad_norm": 0.060522302985191345,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 113874
+    },
+    {
+      "epoch": 11.703699897225077,
+      "grad_norm": 0.06372793763875961,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 113877
+    },
+    {
+      "epoch": 11.704008221993833,
+      "grad_norm": 0.09572730958461761,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 113880
+    },
+    {
+      "epoch": 11.70431654676259,
+      "grad_norm": 0.03504783287644386,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 113883
+    },
+    {
+      "epoch": 11.704624871531346,
+      "grad_norm": 0.10535631328821182,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 113886
+    },
+    {
+      "epoch": 11.704933196300102,
+      "grad_norm": 0.08261964470148087,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 113889
+    },
+    {
+      "epoch": 11.70524152106886,
+      "grad_norm": 0.09721565246582031,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 113892
+    },
+    {
+      "epoch": 11.705549845837616,
+      "grad_norm": 0.12555786967277527,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 113895
+    },
+    {
+      "epoch": 11.705858170606373,
+      "grad_norm": 0.10910096019506454,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 113898
+    },
+    {
+      "epoch": 11.706166495375129,
+      "grad_norm": 0.08156652003526688,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 113901
+    },
+    {
+      "epoch": 11.706474820143885,
+      "grad_norm": 0.046058304607868195,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 113904
+    },
+    {
+      "epoch": 11.706783144912642,
+      "grad_norm": 0.11909212917089462,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 113907
+    },
+    {
+      "epoch": 11.707091469681398,
+      "grad_norm": 0.06561049818992615,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 113910
+    },
+    {
+      "epoch": 11.707399794450154,
+      "grad_norm": 0.05175124108791351,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 113913
+    },
+    {
+      "epoch": 11.70770811921891,
+      "grad_norm": 0.03363698348402977,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 113916
+    },
+    {
+      "epoch": 11.708016443987667,
+      "grad_norm": 0.05112067237496376,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 113919
+    },
+    {
+      "epoch": 11.708324768756423,
+      "grad_norm": 0.08059346675872803,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 113922
+    },
+    {
+      "epoch": 11.70863309352518,
+      "grad_norm": 0.09412217885255814,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 113925
+    },
+    {
+      "epoch": 11.708941418293936,
+      "grad_norm": 0.05953041836619377,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 113928
+    },
+    {
+      "epoch": 11.709249743062692,
+      "grad_norm": 0.08858895301818848,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 113931
+    },
+    {
+      "epoch": 11.709558067831448,
+      "grad_norm": 0.09069149196147919,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 113934
+    },
+    {
+      "epoch": 11.709866392600205,
+      "grad_norm": 0.06732858717441559,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 113937
+    },
+    {
+      "epoch": 11.710174717368963,
+      "grad_norm": 0.0464613251388073,
+      "learning_rate": 0.01,
+      "loss": 1.9656,
+      "step": 113940
+    },
+    {
+      "epoch": 11.710483042137719,
+      "grad_norm": 0.08977202326059341,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 113943
+    },
+    {
+      "epoch": 11.710791366906475,
+      "grad_norm": 0.11978859454393387,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 113946
+    },
+    {
+      "epoch": 11.711099691675232,
+      "grad_norm": 0.08750269562005997,
+      "learning_rate": 0.01,
+      "loss": 1.9734,
+      "step": 113949
+    },
+    {
+      "epoch": 11.711408016443988,
+      "grad_norm": 0.058178629726171494,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 113952
+    },
+    {
+      "epoch": 11.711716341212744,
+      "grad_norm": 0.12875346839427948,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 113955
+    },
+    {
+      "epoch": 11.7120246659815,
+      "grad_norm": 0.052333295345306396,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 113958
+    },
+    {
+      "epoch": 11.712332990750257,
+      "grad_norm": 0.04038956016302109,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 113961
+    },
+    {
+      "epoch": 11.712641315519013,
+      "grad_norm": 0.09954986721277237,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 113964
+    },
+    {
+      "epoch": 11.71294964028777,
+      "grad_norm": 0.05516171455383301,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 113967
+    },
+    {
+      "epoch": 11.713257965056526,
+      "grad_norm": 0.04130719229578972,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 113970
+    },
+    {
+      "epoch": 11.713566289825282,
+      "grad_norm": 0.04109114035964012,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 113973
+    },
+    {
+      "epoch": 11.713874614594038,
+      "grad_norm": 0.071103036403656,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 113976
+    },
+    {
+      "epoch": 11.714182939362795,
+      "grad_norm": 0.10898397862911224,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 113979
+    },
+    {
+      "epoch": 11.714491264131553,
+      "grad_norm": 0.03577899560332298,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 113982
+    },
+    {
+      "epoch": 11.71479958890031,
+      "grad_norm": 0.07445529848337173,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 113985
+    },
+    {
+      "epoch": 11.715107913669065,
+      "grad_norm": 0.09908154606819153,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 113988
+    },
+    {
+      "epoch": 11.715416238437822,
+      "grad_norm": 0.07953612506389618,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 113991
+    },
+    {
+      "epoch": 11.715724563206578,
+      "grad_norm": 0.05587496981024742,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 113994
+    },
+    {
+      "epoch": 11.716032887975334,
+      "grad_norm": 0.040337421000003815,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 113997
+    },
+    {
+      "epoch": 11.71634121274409,
+      "grad_norm": 0.07240230590105057,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 114000
+    },
+    {
+      "epoch": 11.716649537512847,
+      "grad_norm": 0.0691431537270546,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 114003
+    },
+    {
+      "epoch": 11.716957862281603,
+      "grad_norm": 0.06162949278950691,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 114006
+    },
+    {
+      "epoch": 11.71726618705036,
+      "grad_norm": 0.10445791482925415,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 114009
+    },
+    {
+      "epoch": 11.717574511819116,
+      "grad_norm": 0.041798826307058334,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 114012
+    },
+    {
+      "epoch": 11.717882836587872,
+      "grad_norm": 0.08959610015153885,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 114015
+    },
+    {
+      "epoch": 11.718191161356629,
+      "grad_norm": 0.12138823419809341,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 114018
+    },
+    {
+      "epoch": 11.718499486125385,
+      "grad_norm": 0.066931813955307,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 114021
+    },
+    {
+      "epoch": 11.718807810894141,
+      "grad_norm": 0.05871887132525444,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 114024
+    },
+    {
+      "epoch": 11.719116135662897,
+      "grad_norm": 0.057196423411369324,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 114027
+    },
+    {
+      "epoch": 11.719424460431654,
+      "grad_norm": 0.056774549186229706,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 114030
+    },
+    {
+      "epoch": 11.719732785200412,
+      "grad_norm": 0.050951916724443436,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 114033
+    },
+    {
+      "epoch": 11.720041109969168,
+      "grad_norm": 0.0355374775826931,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 114036
+    },
+    {
+      "epoch": 11.720349434737924,
+      "grad_norm": 0.06845880299806595,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 114039
+    },
+    {
+      "epoch": 11.72065775950668,
+      "grad_norm": 0.09319114685058594,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 114042
+    },
+    {
+      "epoch": 11.720966084275437,
+      "grad_norm": 0.08113586902618408,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 114045
+    },
+    {
+      "epoch": 11.721274409044193,
+      "grad_norm": 0.06692340970039368,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 114048
+    },
+    {
+      "epoch": 11.72158273381295,
+      "grad_norm": 0.10912377387285233,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 114051
+    },
+    {
+      "epoch": 11.721891058581706,
+      "grad_norm": 0.06332485377788544,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 114054
+    },
+    {
+      "epoch": 11.722199383350462,
+      "grad_norm": 0.0489015206694603,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 114057
+    },
+    {
+      "epoch": 11.722507708119219,
+      "grad_norm": 0.06264995783567429,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 114060
+    },
+    {
+      "epoch": 11.722816032887975,
+      "grad_norm": 0.2138553410768509,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 114063
+    },
+    {
+      "epoch": 11.723124357656731,
+      "grad_norm": 0.0857858806848526,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 114066
+    },
+    {
+      "epoch": 11.723432682425488,
+      "grad_norm": 0.06043681129813194,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 114069
+    },
+    {
+      "epoch": 11.723741007194244,
+      "grad_norm": 0.04940197989344597,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 114072
+    },
+    {
+      "epoch": 11.724049331963002,
+      "grad_norm": 0.043699298053979874,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 114075
+    },
+    {
+      "epoch": 11.724357656731758,
+      "grad_norm": 0.05862659215927124,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 114078
+    },
+    {
+      "epoch": 11.724665981500515,
+      "grad_norm": 0.05761629715561867,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 114081
+    },
+    {
+      "epoch": 11.72497430626927,
+      "grad_norm": 0.05093192309141159,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 114084
+    },
+    {
+      "epoch": 11.725282631038027,
+      "grad_norm": 0.0673130676150322,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 114087
+    },
+    {
+      "epoch": 11.725590955806783,
+      "grad_norm": 0.044155821204185486,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 114090
+    },
+    {
+      "epoch": 11.72589928057554,
+      "grad_norm": 0.0731462836265564,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 114093
+    },
+    {
+      "epoch": 11.726207605344296,
+      "grad_norm": 0.18581373989582062,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 114096
+    },
+    {
+      "epoch": 11.726515930113052,
+      "grad_norm": 0.09261856973171234,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 114099
+    },
+    {
+      "epoch": 11.726824254881809,
+      "grad_norm": 0.06688740104436874,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 114102
+    },
+    {
+      "epoch": 11.727132579650565,
+      "grad_norm": 0.04713169485330582,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 114105
+    },
+    {
+      "epoch": 11.727440904419321,
+      "grad_norm": 0.04400784894824028,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 114108
+    },
+    {
+      "epoch": 11.727749229188078,
+      "grad_norm": 0.04737546294927597,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 114111
+    },
+    {
+      "epoch": 11.728057553956834,
+      "grad_norm": 0.03855617716908455,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 114114
+    },
+    {
+      "epoch": 11.72836587872559,
+      "grad_norm": 0.04322519153356552,
+      "learning_rate": 0.01,
+      "loss": 1.9622,
+      "step": 114117
+    },
+    {
+      "epoch": 11.728674203494347,
+      "grad_norm": 0.1041107103228569,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 114120
+    },
+    {
+      "epoch": 11.728982528263105,
+      "grad_norm": 0.04595714807510376,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 114123
+    },
+    {
+      "epoch": 11.729290853031861,
+      "grad_norm": 0.062153588980436325,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 114126
+    },
+    {
+      "epoch": 11.729599177800617,
+      "grad_norm": 0.09231128543615341,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 114129
+    },
+    {
+      "epoch": 11.729907502569374,
+      "grad_norm": 0.08219090104103088,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 114132
+    },
+    {
+      "epoch": 11.73021582733813,
+      "grad_norm": 0.04350368678569794,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 114135
+    },
+    {
+      "epoch": 11.730524152106886,
+      "grad_norm": 0.10275841504335403,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 114138
+    },
+    {
+      "epoch": 11.730832476875642,
+      "grad_norm": 0.04107731208205223,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 114141
+    },
+    {
+      "epoch": 11.731140801644399,
+      "grad_norm": 0.09991445392370224,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 114144
+    },
+    {
+      "epoch": 11.731449126413155,
+      "grad_norm": 0.04435478150844574,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 114147
+    },
+    {
+      "epoch": 11.731757451181911,
+      "grad_norm": 0.08842992782592773,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 114150
+    },
+    {
+      "epoch": 11.732065775950668,
+      "grad_norm": 0.06753559410572052,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 114153
+    },
+    {
+      "epoch": 11.732374100719424,
+      "grad_norm": 0.07544936239719391,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 114156
+    },
+    {
+      "epoch": 11.73268242548818,
+      "grad_norm": 0.11217570304870605,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 114159
+    },
+    {
+      "epoch": 11.732990750256937,
+      "grad_norm": 0.04507189989089966,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 114162
+    },
+    {
+      "epoch": 11.733299075025693,
+      "grad_norm": 0.0769190564751625,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 114165
+    },
+    {
+      "epoch": 11.733607399794451,
+      "grad_norm": 0.12560825049877167,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 114168
+    },
+    {
+      "epoch": 11.733915724563207,
+      "grad_norm": 0.07754191756248474,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 114171
+    },
+    {
+      "epoch": 11.734224049331964,
+      "grad_norm": 0.08842162787914276,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 114174
+    },
+    {
+      "epoch": 11.73453237410072,
+      "grad_norm": 0.043762583285570145,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 114177
+    },
+    {
+      "epoch": 11.734840698869476,
+      "grad_norm": 0.03222280368208885,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 114180
+    },
+    {
+      "epoch": 11.735149023638233,
+      "grad_norm": 0.03388447314500809,
+      "learning_rate": 0.01,
+      "loss": 1.9705,
+      "step": 114183
+    },
+    {
+      "epoch": 11.735457348406989,
+      "grad_norm": 0.03262539952993393,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 114186
+    },
+    {
+      "epoch": 11.735765673175745,
+      "grad_norm": 0.03528260067105293,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 114189
+    },
+    {
+      "epoch": 11.736073997944501,
+      "grad_norm": 0.07946651428937912,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 114192
+    },
+    {
+      "epoch": 11.736382322713258,
+      "grad_norm": 0.059255629777908325,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 114195
+    },
+    {
+      "epoch": 11.736690647482014,
+      "grad_norm": 0.07938528060913086,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 114198
+    },
+    {
+      "epoch": 11.73699897225077,
+      "grad_norm": 0.06482166796922684,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 114201
+    },
+    {
+      "epoch": 11.737307297019527,
+      "grad_norm": 0.03566895052790642,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 114204
+    },
+    {
+      "epoch": 11.737615621788283,
+      "grad_norm": 0.05914255604147911,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 114207
+    },
+    {
+      "epoch": 11.73792394655704,
+      "grad_norm": 0.07074881345033646,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 114210
+    },
+    {
+      "epoch": 11.738232271325796,
+      "grad_norm": 0.09997674077749252,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 114213
+    },
+    {
+      "epoch": 11.738540596094554,
+      "grad_norm": 0.14847432076931,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 114216
+    },
+    {
+      "epoch": 11.73884892086331,
+      "grad_norm": 0.07328097522258759,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 114219
+    },
+    {
+      "epoch": 11.739157245632066,
+      "grad_norm": 0.09382440894842148,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 114222
+    },
+    {
+      "epoch": 11.739465570400823,
+      "grad_norm": 0.09148334711790085,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 114225
+    },
+    {
+      "epoch": 11.739773895169579,
+      "grad_norm": 0.03747795149683952,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 114228
+    },
+    {
+      "epoch": 11.740082219938335,
+      "grad_norm": 0.04483797028660774,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 114231
+    },
+    {
+      "epoch": 11.740390544707092,
+      "grad_norm": 0.07153434306383133,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 114234
+    },
+    {
+      "epoch": 11.740698869475848,
+      "grad_norm": 0.1360405832529068,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 114237
+    },
+    {
+      "epoch": 11.741007194244604,
+      "grad_norm": 0.05981483310461044,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 114240
+    },
+    {
+      "epoch": 11.74131551901336,
+      "grad_norm": 0.06791744381189346,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 114243
+    },
+    {
+      "epoch": 11.741623843782117,
+      "grad_norm": 0.049168311059474945,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 114246
+    },
+    {
+      "epoch": 11.741932168550873,
+      "grad_norm": 0.06850913166999817,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 114249
+    },
+    {
+      "epoch": 11.74224049331963,
+      "grad_norm": 0.09397213160991669,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 114252
+    },
+    {
+      "epoch": 11.742548818088386,
+      "grad_norm": 0.14439097046852112,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 114255
+    },
+    {
+      "epoch": 11.742857142857144,
+      "grad_norm": 0.060273777693510056,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 114258
+    },
+    {
+      "epoch": 11.7431654676259,
+      "grad_norm": 0.05283762887120247,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 114261
+    },
+    {
+      "epoch": 11.743473792394656,
+      "grad_norm": 0.06911233812570572,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 114264
+    },
+    {
+      "epoch": 11.743782117163413,
+      "grad_norm": 0.06413587927818298,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 114267
+    },
+    {
+      "epoch": 11.744090441932169,
+      "grad_norm": 0.032178230583667755,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 114270
+    },
+    {
+      "epoch": 11.744398766700925,
+      "grad_norm": 0.0497891865670681,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 114273
+    },
+    {
+      "epoch": 11.744707091469682,
+      "grad_norm": 0.09936387091875076,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 114276
+    },
+    {
+      "epoch": 11.745015416238438,
+      "grad_norm": 0.05286841839551926,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 114279
+    },
+    {
+      "epoch": 11.745323741007194,
+      "grad_norm": 0.038032785058021545,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 114282
+    },
+    {
+      "epoch": 11.74563206577595,
+      "grad_norm": 0.037610363215208054,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 114285
+    },
+    {
+      "epoch": 11.745940390544707,
+      "grad_norm": 0.03564333915710449,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 114288
+    },
+    {
+      "epoch": 11.746248715313463,
+      "grad_norm": 0.0679815486073494,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 114291
+    },
+    {
+      "epoch": 11.74655704008222,
+      "grad_norm": 0.1300264149904251,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 114294
+    },
+    {
+      "epoch": 11.746865364850976,
+      "grad_norm": 0.06182750687003136,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 114297
+    },
+    {
+      "epoch": 11.747173689619732,
+      "grad_norm": 0.10312792658805847,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 114300
+    },
+    {
+      "epoch": 11.747482014388488,
+      "grad_norm": 0.08729681372642517,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 114303
+    },
+    {
+      "epoch": 11.747790339157246,
+      "grad_norm": 0.04889155924320221,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 114306
+    },
+    {
+      "epoch": 11.748098663926003,
+      "grad_norm": 0.08509171009063721,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 114309
+    },
+    {
+      "epoch": 11.748406988694759,
+      "grad_norm": 0.06404619663953781,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 114312
+    },
+    {
+      "epoch": 11.748715313463515,
+      "grad_norm": 0.05138705298304558,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 114315
+    },
+    {
+      "epoch": 11.749023638232272,
+      "grad_norm": 0.03681953623890877,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 114318
+    },
+    {
+      "epoch": 11.749331963001028,
+      "grad_norm": 0.05085345357656479,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 114321
+    },
+    {
+      "epoch": 11.749640287769784,
+      "grad_norm": 0.10160741955041885,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 114324
+    },
+    {
+      "epoch": 11.74994861253854,
+      "grad_norm": 0.05211213231086731,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 114327
+    },
+    {
+      "epoch": 11.750256937307297,
+      "grad_norm": 0.09597420692443848,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 114330
+    },
+    {
+      "epoch": 11.750565262076053,
+      "grad_norm": 0.05214196443557739,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 114333
+    },
+    {
+      "epoch": 11.75087358684481,
+      "grad_norm": 0.04194926470518112,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 114336
+    },
+    {
+      "epoch": 11.751181911613566,
+      "grad_norm": 0.13959670066833496,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 114339
+    },
+    {
+      "epoch": 11.751490236382322,
+      "grad_norm": 0.10564197599887848,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 114342
+    },
+    {
+      "epoch": 11.751798561151078,
+      "grad_norm": 0.03934367373585701,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 114345
+    },
+    {
+      "epoch": 11.752106885919835,
+      "grad_norm": 0.03357434645295143,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 114348
+    },
+    {
+      "epoch": 11.752415210688593,
+      "grad_norm": 0.0841175839304924,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 114351
+    },
+    {
+      "epoch": 11.75272353545735,
+      "grad_norm": 0.07452627271413803,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 114354
+    },
+    {
+      "epoch": 11.753031860226105,
+      "grad_norm": 0.058551378548145294,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 114357
+    },
+    {
+      "epoch": 11.753340184994862,
+      "grad_norm": 0.07105761021375656,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 114360
+    },
+    {
+      "epoch": 11.753648509763618,
+      "grad_norm": 0.06706007570028305,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 114363
+    },
+    {
+      "epoch": 11.753956834532374,
+      "grad_norm": 0.03882206976413727,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 114366
+    },
+    {
+      "epoch": 11.75426515930113,
+      "grad_norm": 0.05124860629439354,
+      "learning_rate": 0.01,
+      "loss": 1.9655,
+      "step": 114369
+    },
+    {
+      "epoch": 11.754573484069887,
+      "grad_norm": 0.11925289779901505,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 114372
+    },
+    {
+      "epoch": 11.754881808838643,
+      "grad_norm": 0.05063081160187721,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 114375
+    },
+    {
+      "epoch": 11.7551901336074,
+      "grad_norm": 0.07430731505155563,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 114378
+    },
+    {
+      "epoch": 11.755498458376156,
+      "grad_norm": 0.08639290928840637,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 114381
+    },
+    {
+      "epoch": 11.755806783144912,
+      "grad_norm": 0.0768064558506012,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 114384
+    },
+    {
+      "epoch": 11.756115107913669,
+      "grad_norm": 0.0926046296954155,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 114387
+    },
+    {
+      "epoch": 11.756423432682425,
+      "grad_norm": 0.04127130284905434,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 114390
+    },
+    {
+      "epoch": 11.756731757451181,
+      "grad_norm": 0.09220347553491592,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 114393
+    },
+    {
+      "epoch": 11.757040082219937,
+      "grad_norm": 0.038777247071266174,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 114396
+    },
+    {
+      "epoch": 11.757348406988696,
+      "grad_norm": 0.099837526679039,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 114399
+    },
+    {
+      "epoch": 11.757656731757452,
+      "grad_norm": 0.11611324548721313,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 114402
+    },
+    {
+      "epoch": 11.757965056526208,
+      "grad_norm": 0.0463559627532959,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 114405
+    },
+    {
+      "epoch": 11.758273381294964,
+      "grad_norm": 0.07250874489545822,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 114408
+    },
+    {
+      "epoch": 11.75858170606372,
+      "grad_norm": 0.03514222800731659,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 114411
+    },
+    {
+      "epoch": 11.758890030832477,
+      "grad_norm": 0.11197876930236816,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 114414
+    },
+    {
+      "epoch": 11.759198355601233,
+      "grad_norm": 0.06688223779201508,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 114417
+    },
+    {
+      "epoch": 11.75950668036999,
+      "grad_norm": 0.044732652604579926,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 114420
+    },
+    {
+      "epoch": 11.759815005138746,
+      "grad_norm": 0.052017007023096085,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 114423
+    },
+    {
+      "epoch": 11.760123329907502,
+      "grad_norm": 0.045031189918518066,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 114426
+    },
+    {
+      "epoch": 11.760431654676259,
+      "grad_norm": 0.1596817821264267,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 114429
+    },
+    {
+      "epoch": 11.760739979445015,
+      "grad_norm": 0.12803293764591217,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 114432
+    },
+    {
+      "epoch": 11.761048304213771,
+      "grad_norm": 0.10335230082273483,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 114435
+    },
+    {
+      "epoch": 11.761356628982528,
+      "grad_norm": 0.05117227882146835,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 114438
+    },
+    {
+      "epoch": 11.761664953751286,
+      "grad_norm": 0.07433784008026123,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 114441
+    },
+    {
+      "epoch": 11.761973278520042,
+      "grad_norm": 0.10217359662055969,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 114444
+    },
+    {
+      "epoch": 11.762281603288798,
+      "grad_norm": 0.0698227658867836,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 114447
+    },
+    {
+      "epoch": 11.762589928057555,
+      "grad_norm": 0.08645881712436676,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 114450
+    },
+    {
+      "epoch": 11.76289825282631,
+      "grad_norm": 0.0554266981780529,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 114453
+    },
+    {
+      "epoch": 11.763206577595067,
+      "grad_norm": 0.04438130557537079,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 114456
+    },
+    {
+      "epoch": 11.763514902363823,
+      "grad_norm": 0.11492709070444107,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 114459
+    },
+    {
+      "epoch": 11.76382322713258,
+      "grad_norm": 0.045961733907461166,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 114462
+    },
+    {
+      "epoch": 11.764131551901336,
+      "grad_norm": 0.03696111962199211,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 114465
+    },
+    {
+      "epoch": 11.764439876670092,
+      "grad_norm": 0.04073779284954071,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 114468
+    },
+    {
+      "epoch": 11.764748201438849,
+      "grad_norm": 0.0709969773888588,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 114471
+    },
+    {
+      "epoch": 11.765056526207605,
+      "grad_norm": 0.10012876242399216,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 114474
+    },
+    {
+      "epoch": 11.765364850976361,
+      "grad_norm": 0.06984630972146988,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 114477
+    },
+    {
+      "epoch": 11.765673175745118,
+      "grad_norm": 0.0934528261423111,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 114480
+    },
+    {
+      "epoch": 11.765981500513874,
+      "grad_norm": 0.03615093603730202,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 114483
+    },
+    {
+      "epoch": 11.76628982528263,
+      "grad_norm": 0.06740186363458633,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 114486
+    },
+    {
+      "epoch": 11.766598150051387,
+      "grad_norm": 0.1075219139456749,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 114489
+    },
+    {
+      "epoch": 11.766906474820145,
+      "grad_norm": 0.03994523361325264,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 114492
+    },
+    {
+      "epoch": 11.767214799588901,
+      "grad_norm": 0.08029788732528687,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 114495
+    },
+    {
+      "epoch": 11.767523124357657,
+      "grad_norm": 0.05670519918203354,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 114498
+    },
+    {
+      "epoch": 11.767831449126414,
+      "grad_norm": 0.05779879167675972,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 114501
+    },
+    {
+      "epoch": 11.76813977389517,
+      "grad_norm": 0.07612137496471405,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 114504
+    },
+    {
+      "epoch": 11.768448098663926,
+      "grad_norm": 0.05214044824242592,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 114507
+    },
+    {
+      "epoch": 11.768756423432682,
+      "grad_norm": 0.05248599871993065,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 114510
+    },
+    {
+      "epoch": 11.769064748201439,
+      "grad_norm": 0.038120802491903305,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 114513
+    },
+    {
+      "epoch": 11.769373072970195,
+      "grad_norm": 0.04978492483496666,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 114516
+    },
+    {
+      "epoch": 11.769681397738951,
+      "grad_norm": 0.059046100825071335,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 114519
+    },
+    {
+      "epoch": 11.769989722507708,
+      "grad_norm": 0.11214884370565414,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 114522
+    },
+    {
+      "epoch": 11.770298047276464,
+      "grad_norm": 0.05173084884881973,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 114525
+    },
+    {
+      "epoch": 11.77060637204522,
+      "grad_norm": 0.060318879783153534,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 114528
+    },
+    {
+      "epoch": 11.770914696813977,
+      "grad_norm": 0.10231828689575195,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 114531
+    },
+    {
+      "epoch": 11.771223021582735,
+      "grad_norm": 0.08297566324472427,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 114534
+    },
+    {
+      "epoch": 11.771531346351491,
+      "grad_norm": 0.10643540322780609,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 114537
+    },
+    {
+      "epoch": 11.771839671120247,
+      "grad_norm": 0.05867210030555725,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 114540
+    },
+    {
+      "epoch": 11.772147995889004,
+      "grad_norm": 0.039664510637521744,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 114543
+    },
+    {
+      "epoch": 11.77245632065776,
+      "grad_norm": 0.044358786195516586,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 114546
+    },
+    {
+      "epoch": 11.772764645426516,
+      "grad_norm": 0.09489329159259796,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 114549
+    },
+    {
+      "epoch": 11.773072970195273,
+      "grad_norm": 0.06318821012973785,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 114552
+    },
+    {
+      "epoch": 11.773381294964029,
+      "grad_norm": 0.11085253208875656,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 114555
+    },
+    {
+      "epoch": 11.773689619732785,
+      "grad_norm": 0.09740877896547318,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 114558
+    },
+    {
+      "epoch": 11.773997944501541,
+      "grad_norm": 0.037672366946935654,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 114561
+    },
+    {
+      "epoch": 11.774306269270298,
+      "grad_norm": 0.044902145862579346,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 114564
+    },
+    {
+      "epoch": 11.774614594039054,
+      "grad_norm": 0.10807695239782333,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 114567
+    },
+    {
+      "epoch": 11.77492291880781,
+      "grad_norm": 0.0927921012043953,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 114570
+    },
+    {
+      "epoch": 11.775231243576567,
+      "grad_norm": 0.05143541097640991,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 114573
+    },
+    {
+      "epoch": 11.775539568345323,
+      "grad_norm": 0.09573736041784286,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 114576
+    },
+    {
+      "epoch": 11.77584789311408,
+      "grad_norm": 0.11310027539730072,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 114579
+    },
+    {
+      "epoch": 11.776156217882837,
+      "grad_norm": 0.03688127547502518,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 114582
+    },
+    {
+      "epoch": 11.776464542651594,
+      "grad_norm": 0.035128772258758545,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 114585
+    },
+    {
+      "epoch": 11.77677286742035,
+      "grad_norm": 0.039170041680336,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 114588
+    },
+    {
+      "epoch": 11.777081192189106,
+      "grad_norm": 0.06230974569916725,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 114591
+    },
+    {
+      "epoch": 11.777389516957863,
+      "grad_norm": 0.06418593972921371,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 114594
+    },
+    {
+      "epoch": 11.777697841726619,
+      "grad_norm": 0.036100953817367554,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 114597
+    },
+    {
+      "epoch": 11.778006166495375,
+      "grad_norm": 0.037238385528326035,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 114600
+    },
+    {
+      "epoch": 11.778314491264132,
+      "grad_norm": 0.12714260816574097,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 114603
+    },
+    {
+      "epoch": 11.778622816032888,
+      "grad_norm": 0.13031919300556183,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 114606
+    },
+    {
+      "epoch": 11.778931140801644,
+      "grad_norm": 0.04448504000902176,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 114609
+    },
+    {
+      "epoch": 11.7792394655704,
+      "grad_norm": 0.049985919147729874,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 114612
+    },
+    {
+      "epoch": 11.779547790339157,
+      "grad_norm": 0.061246324330568314,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 114615
+    },
+    {
+      "epoch": 11.779856115107913,
+      "grad_norm": 0.04634316638112068,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 114618
+    },
+    {
+      "epoch": 11.78016443987667,
+      "grad_norm": 0.07784038782119751,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 114621
+    },
+    {
+      "epoch": 11.780472764645426,
+      "grad_norm": 0.10071872174739838,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 114624
+    },
+    {
+      "epoch": 11.780781089414184,
+      "grad_norm": 0.04656301811337471,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 114627
+    },
+    {
+      "epoch": 11.78108941418294,
+      "grad_norm": 0.0605628676712513,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 114630
+    },
+    {
+      "epoch": 11.781397738951696,
+      "grad_norm": 0.12403811514377594,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 114633
+    },
+    {
+      "epoch": 11.781706063720453,
+      "grad_norm": 0.037557389587163925,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 114636
+    },
+    {
+      "epoch": 11.782014388489209,
+      "grad_norm": 0.0667833462357521,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 114639
+    },
+    {
+      "epoch": 11.782322713257965,
+      "grad_norm": 0.08555550873279572,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 114642
+    },
+    {
+      "epoch": 11.782631038026722,
+      "grad_norm": 0.1030685231089592,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 114645
+    },
+    {
+      "epoch": 11.782939362795478,
+      "grad_norm": 0.06644846498966217,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 114648
+    },
+    {
+      "epoch": 11.783247687564234,
+      "grad_norm": 0.06920860707759857,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 114651
+    },
+    {
+      "epoch": 11.78355601233299,
+      "grad_norm": 0.07964517176151276,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 114654
+    },
+    {
+      "epoch": 11.783864337101747,
+      "grad_norm": 0.05284479632973671,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 114657
+    },
+    {
+      "epoch": 11.784172661870503,
+      "grad_norm": 0.1048341915011406,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 114660
+    },
+    {
+      "epoch": 11.78448098663926,
+      "grad_norm": 0.035469092428684235,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 114663
+    },
+    {
+      "epoch": 11.784789311408016,
+      "grad_norm": 0.05209077149629593,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 114666
+    },
+    {
+      "epoch": 11.785097636176772,
+      "grad_norm": 0.12196382880210876,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 114669
+    },
+    {
+      "epoch": 11.785405960945528,
+      "grad_norm": 0.047950629144907,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 114672
+    },
+    {
+      "epoch": 11.785714285714286,
+      "grad_norm": 0.0819501280784607,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 114675
+    },
+    {
+      "epoch": 11.786022610483043,
+      "grad_norm": 0.06672423332929611,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 114678
+    },
+    {
+      "epoch": 11.786330935251799,
+      "grad_norm": 0.0418439544737339,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 114681
+    },
+    {
+      "epoch": 11.786639260020555,
+      "grad_norm": 0.04796493425965309,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 114684
+    },
+    {
+      "epoch": 11.786947584789312,
+      "grad_norm": 0.07640409469604492,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 114687
+    },
+    {
+      "epoch": 11.787255909558068,
+      "grad_norm": 0.045105814933776855,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 114690
+    },
+    {
+      "epoch": 11.787564234326824,
+      "grad_norm": 0.13338594138622284,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 114693
+    },
+    {
+      "epoch": 11.78787255909558,
+      "grad_norm": 0.09243103861808777,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 114696
+    },
+    {
+      "epoch": 11.788180883864337,
+      "grad_norm": 0.06404594331979752,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 114699
+    },
+    {
+      "epoch": 11.788489208633093,
+      "grad_norm": 0.07861119508743286,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 114702
+    },
+    {
+      "epoch": 11.78879753340185,
+      "grad_norm": 0.1092933937907219,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 114705
+    },
+    {
+      "epoch": 11.789105858170606,
+      "grad_norm": 0.07028528302907944,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 114708
+    },
+    {
+      "epoch": 11.789414182939362,
+      "grad_norm": 0.07368434220552444,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 114711
+    },
+    {
+      "epoch": 11.789722507708118,
+      "grad_norm": 0.08186358213424683,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 114714
+    },
+    {
+      "epoch": 11.790030832476877,
+      "grad_norm": 0.04825025796890259,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 114717
+    },
+    {
+      "epoch": 11.790339157245633,
+      "grad_norm": 0.05701156705617905,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 114720
+    },
+    {
+      "epoch": 11.79064748201439,
+      "grad_norm": 0.034114472568035126,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 114723
+    },
+    {
+      "epoch": 11.790955806783145,
+      "grad_norm": 0.056312914937734604,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 114726
+    },
+    {
+      "epoch": 11.791264131551902,
+      "grad_norm": 0.07758929580450058,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 114729
+    },
+    {
+      "epoch": 11.791572456320658,
+      "grad_norm": 0.04463218152523041,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 114732
+    },
+    {
+      "epoch": 11.791880781089414,
+      "grad_norm": 0.05125407129526138,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 114735
+    },
+    {
+      "epoch": 11.79218910585817,
+      "grad_norm": 0.08259645104408264,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 114738
+    },
+    {
+      "epoch": 11.792497430626927,
+      "grad_norm": 0.06180672347545624,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 114741
+    },
+    {
+      "epoch": 11.792805755395683,
+      "grad_norm": 0.05911092460155487,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 114744
+    },
+    {
+      "epoch": 11.79311408016444,
+      "grad_norm": 0.1537017822265625,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 114747
+    },
+    {
+      "epoch": 11.793422404933196,
+      "grad_norm": 0.08543099462985992,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 114750
+    },
+    {
+      "epoch": 11.793730729701952,
+      "grad_norm": 0.05843864008784294,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 114753
+    },
+    {
+      "epoch": 11.794039054470709,
+      "grad_norm": 0.04629216343164444,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 114756
+    },
+    {
+      "epoch": 11.794347379239465,
+      "grad_norm": 0.06042463332414627,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 114759
+    },
+    {
+      "epoch": 11.794655704008221,
+      "grad_norm": 0.033288221806287766,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 114762
+    },
+    {
+      "epoch": 11.79496402877698,
+      "grad_norm": 0.060233380645513535,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 114765
+    },
+    {
+      "epoch": 11.795272353545736,
+      "grad_norm": 0.09179306030273438,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 114768
+    },
+    {
+      "epoch": 11.795580678314492,
+      "grad_norm": 0.052546773105859756,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 114771
+    },
+    {
+      "epoch": 11.795889003083248,
+      "grad_norm": 0.06451155245304108,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 114774
+    },
+    {
+      "epoch": 11.796197327852004,
+      "grad_norm": 0.10466645658016205,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 114777
+    },
+    {
+      "epoch": 11.79650565262076,
+      "grad_norm": 0.08434021472930908,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 114780
+    },
+    {
+      "epoch": 11.796813977389517,
+      "grad_norm": 0.06126517429947853,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 114783
+    },
+    {
+      "epoch": 11.797122302158273,
+      "grad_norm": 0.09324466437101364,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 114786
+    },
+    {
+      "epoch": 11.79743062692703,
+      "grad_norm": 0.07028118520975113,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 114789
+    },
+    {
+      "epoch": 11.797738951695786,
+      "grad_norm": 0.06836069375276566,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 114792
+    },
+    {
+      "epoch": 11.798047276464542,
+      "grad_norm": 0.09277386218309402,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 114795
+    },
+    {
+      "epoch": 11.798355601233299,
+      "grad_norm": 0.08132278174161911,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 114798
+    },
+    {
+      "epoch": 11.798663926002055,
+      "grad_norm": 0.05600438639521599,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 114801
+    },
+    {
+      "epoch": 11.798972250770811,
+      "grad_norm": 0.07362427562475204,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 114804
+    },
+    {
+      "epoch": 11.799280575539568,
+      "grad_norm": 0.06464149802923203,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 114807
+    },
+    {
+      "epoch": 11.799588900308326,
+      "grad_norm": 0.1496574431657791,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 114810
+    },
+    {
+      "epoch": 11.799897225077082,
+      "grad_norm": 0.1321372091770172,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 114813
+    },
+    {
+      "epoch": 11.800205549845838,
+      "grad_norm": 0.03908989951014519,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 114816
+    },
+    {
+      "epoch": 11.800513874614595,
+      "grad_norm": 0.03196997568011284,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 114819
+    },
+    {
+      "epoch": 11.80082219938335,
+      "grad_norm": 0.031193269416689873,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 114822
+    },
+    {
+      "epoch": 11.801130524152107,
+      "grad_norm": 0.06302262097597122,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 114825
+    },
+    {
+      "epoch": 11.801438848920863,
+      "grad_norm": 0.06870076805353165,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 114828
+    },
+    {
+      "epoch": 11.80174717368962,
+      "grad_norm": 0.044515084475278854,
+      "learning_rate": 0.01,
+      "loss": 1.9558,
+      "step": 114831
+    },
+    {
+      "epoch": 11.802055498458376,
+      "grad_norm": 0.1153394877910614,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 114834
+    },
+    {
+      "epoch": 11.802363823227132,
+      "grad_norm": 0.03656914830207825,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 114837
+    },
+    {
+      "epoch": 11.802672147995889,
+      "grad_norm": 0.09082802385091782,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 114840
+    },
+    {
+      "epoch": 11.802980472764645,
+      "grad_norm": 0.06535059213638306,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 114843
+    },
+    {
+      "epoch": 11.803288797533401,
+      "grad_norm": 0.1293552815914154,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 114846
+    },
+    {
+      "epoch": 11.803597122302158,
+      "grad_norm": 0.10767047107219696,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 114849
+    },
+    {
+      "epoch": 11.803905447070914,
+      "grad_norm": 0.10255423933267593,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 114852
+    },
+    {
+      "epoch": 11.80421377183967,
+      "grad_norm": 0.08479739725589752,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 114855
+    },
+    {
+      "epoch": 11.804522096608428,
+      "grad_norm": 0.06828516721725464,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 114858
+    },
+    {
+      "epoch": 11.804830421377185,
+      "grad_norm": 0.09660480916500092,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 114861
+    },
+    {
+      "epoch": 11.805138746145941,
+      "grad_norm": 0.05178052559494972,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 114864
+    },
+    {
+      "epoch": 11.805447070914697,
+      "grad_norm": 0.04476309195160866,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 114867
+    },
+    {
+      "epoch": 11.805755395683454,
+      "grad_norm": 0.032808542251586914,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 114870
+    },
+    {
+      "epoch": 11.80606372045221,
+      "grad_norm": 0.04916965961456299,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 114873
+    },
+    {
+      "epoch": 11.806372045220966,
+      "grad_norm": 0.13022355735301971,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 114876
+    },
+    {
+      "epoch": 11.806680369989722,
+      "grad_norm": 0.05587311089038849,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 114879
+    },
+    {
+      "epoch": 11.806988694758479,
+      "grad_norm": 0.10787022858858109,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 114882
+    },
+    {
+      "epoch": 11.807297019527235,
+      "grad_norm": 0.05247826874256134,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 114885
+    },
+    {
+      "epoch": 11.807605344295991,
+      "grad_norm": 0.04825177043676376,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 114888
+    },
+    {
+      "epoch": 11.807913669064748,
+      "grad_norm": 0.04047880694270134,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 114891
+    },
+    {
+      "epoch": 11.808221993833504,
+      "grad_norm": 0.12832796573638916,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 114894
+    },
+    {
+      "epoch": 11.80853031860226,
+      "grad_norm": 0.05020860210061073,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 114897
+    },
+    {
+      "epoch": 11.808838643371018,
+      "grad_norm": 0.07286757975816727,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 114900
+    },
+    {
+      "epoch": 11.809146968139775,
+      "grad_norm": 0.05260704830288887,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 114903
+    },
+    {
+      "epoch": 11.809455292908531,
+      "grad_norm": 0.08480259776115417,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 114906
+    },
+    {
+      "epoch": 11.809763617677287,
+      "grad_norm": 0.07258189469575882,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 114909
+    },
+    {
+      "epoch": 11.810071942446044,
+      "grad_norm": 0.08959133177995682,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 114912
+    },
+    {
+      "epoch": 11.8103802672148,
+      "grad_norm": 0.10010571777820587,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 114915
+    },
+    {
+      "epoch": 11.810688591983556,
+      "grad_norm": 0.06512168049812317,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 114918
+    },
+    {
+      "epoch": 11.810996916752313,
+      "grad_norm": 0.0446767583489418,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 114921
+    },
+    {
+      "epoch": 11.811305241521069,
+      "grad_norm": 0.06724687665700912,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 114924
+    },
+    {
+      "epoch": 11.811613566289825,
+      "grad_norm": 0.07073602825403214,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 114927
+    },
+    {
+      "epoch": 11.811921891058581,
+      "grad_norm": 0.050262439996004105,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 114930
+    },
+    {
+      "epoch": 11.812230215827338,
+      "grad_norm": 0.040208492428064346,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 114933
+    },
+    {
+      "epoch": 11.812538540596094,
+      "grad_norm": 0.04738585278391838,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 114936
+    },
+    {
+      "epoch": 11.81284686536485,
+      "grad_norm": 0.04911480098962784,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 114939
+    },
+    {
+      "epoch": 11.813155190133607,
+      "grad_norm": 0.041393179446458817,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 114942
+    },
+    {
+      "epoch": 11.813463514902363,
+      "grad_norm": 0.1633489429950714,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 114945
+    },
+    {
+      "epoch": 11.81377183967112,
+      "grad_norm": 0.14581318199634552,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 114948
+    },
+    {
+      "epoch": 11.814080164439877,
+      "grad_norm": 0.11440169811248779,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 114951
+    },
+    {
+      "epoch": 11.814388489208634,
+      "grad_norm": 0.08294911682605743,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 114954
+    },
+    {
+      "epoch": 11.81469681397739,
+      "grad_norm": 0.13189250230789185,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 114957
+    },
+    {
+      "epoch": 11.815005138746146,
+      "grad_norm": 0.04286221042275429,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 114960
+    },
+    {
+      "epoch": 11.815313463514903,
+      "grad_norm": 0.04227737709879875,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 114963
+    },
+    {
+      "epoch": 11.815621788283659,
+      "grad_norm": 0.04705207422375679,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 114966
+    },
+    {
+      "epoch": 11.815930113052415,
+      "grad_norm": 0.0436391718685627,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 114969
+    },
+    {
+      "epoch": 11.816238437821172,
+      "grad_norm": 0.03224620968103409,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 114972
+    },
+    {
+      "epoch": 11.816546762589928,
+      "grad_norm": 0.04032877832651138,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 114975
+    },
+    {
+      "epoch": 11.816855087358684,
+      "grad_norm": 0.055777013301849365,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 114978
+    },
+    {
+      "epoch": 11.81716341212744,
+      "grad_norm": 0.08680202811956406,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 114981
+    },
+    {
+      "epoch": 11.817471736896197,
+      "grad_norm": 0.12070533633232117,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 114984
+    },
+    {
+      "epoch": 11.817780061664953,
+      "grad_norm": 0.08620209991931915,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 114987
+    },
+    {
+      "epoch": 11.81808838643371,
+      "grad_norm": 0.080354705452919,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 114990
+    },
+    {
+      "epoch": 11.818396711202467,
+      "grad_norm": 0.07663991302251816,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 114993
+    },
+    {
+      "epoch": 11.818705035971224,
+      "grad_norm": 0.053106699138879776,
+      "learning_rate": 0.01,
+      "loss": 1.9657,
+      "step": 114996
+    },
+    {
+      "epoch": 11.81901336073998,
+      "grad_norm": 0.05605434253811836,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 114999
+    },
+    {
+      "epoch": 11.819321685508736,
+      "grad_norm": 0.04178521782159805,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 115002
+    },
+    {
+      "epoch": 11.819630010277493,
+      "grad_norm": 0.06913168728351593,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 115005
+    },
+    {
+      "epoch": 11.819938335046249,
+      "grad_norm": 0.04784342274069786,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 115008
+    },
+    {
+      "epoch": 11.820246659815005,
+      "grad_norm": 0.0556730218231678,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 115011
+    },
+    {
+      "epoch": 11.820554984583762,
+      "grad_norm": 0.10342864692211151,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 115014
+    },
+    {
+      "epoch": 11.820863309352518,
+      "grad_norm": 0.1111392229795456,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 115017
+    },
+    {
+      "epoch": 11.821171634121274,
+      "grad_norm": 0.06416431069374084,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 115020
+    },
+    {
+      "epoch": 11.82147995889003,
+      "grad_norm": 0.09714426100254059,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 115023
+    },
+    {
+      "epoch": 11.821788283658787,
+      "grad_norm": 0.06005309522151947,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 115026
+    },
+    {
+      "epoch": 11.822096608427543,
+      "grad_norm": 0.09578508883714676,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 115029
+    },
+    {
+      "epoch": 11.8224049331963,
+      "grad_norm": 0.04179423674941063,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 115032
+    },
+    {
+      "epoch": 11.822713257965056,
+      "grad_norm": 0.039336420595645905,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 115035
+    },
+    {
+      "epoch": 11.823021582733812,
+      "grad_norm": 0.0391448512673378,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 115038
+    },
+    {
+      "epoch": 11.82332990750257,
+      "grad_norm": 0.12323050200939178,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 115041
+    },
+    {
+      "epoch": 11.823638232271326,
+      "grad_norm": 0.11971402168273926,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 115044
+    },
+    {
+      "epoch": 11.823946557040083,
+      "grad_norm": 0.09418240189552307,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 115047
+    },
+    {
+      "epoch": 11.824254881808839,
+      "grad_norm": 0.1007615178823471,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 115050
+    },
+    {
+      "epoch": 11.824563206577595,
+      "grad_norm": 0.07259011268615723,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 115053
+    },
+    {
+      "epoch": 11.824871531346352,
+      "grad_norm": 0.04412701353430748,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 115056
+    },
+    {
+      "epoch": 11.825179856115108,
+      "grad_norm": 0.06574788689613342,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 115059
+    },
+    {
+      "epoch": 11.825488180883864,
+      "grad_norm": 0.03828756883740425,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 115062
+    },
+    {
+      "epoch": 11.82579650565262,
+      "grad_norm": 0.044093042612075806,
+      "learning_rate": 0.01,
+      "loss": 1.9702,
+      "step": 115065
+    },
+    {
+      "epoch": 11.826104830421377,
+      "grad_norm": 0.054079849272966385,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 115068
+    },
+    {
+      "epoch": 11.826413155190133,
+      "grad_norm": 0.0788705125451088,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 115071
+    },
+    {
+      "epoch": 11.82672147995889,
+      "grad_norm": 0.059001874178647995,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 115074
+    },
+    {
+      "epoch": 11.827029804727646,
+      "grad_norm": 0.05844485014677048,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 115077
+    },
+    {
+      "epoch": 11.827338129496402,
+      "grad_norm": 0.056450288742780685,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 115080
+    },
+    {
+      "epoch": 11.827646454265158,
+      "grad_norm": 0.04511510208249092,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 115083
+    },
+    {
+      "epoch": 11.827954779033917,
+      "grad_norm": 0.04276560992002487,
+      "learning_rate": 0.01,
+      "loss": 1.9795,
+      "step": 115086
+    },
+    {
+      "epoch": 11.828263103802673,
+      "grad_norm": 0.06847599148750305,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 115089
+    },
+    {
+      "epoch": 11.82857142857143,
+      "grad_norm": 0.09598635882139206,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 115092
+    },
+    {
+      "epoch": 11.828879753340185,
+      "grad_norm": 0.09008855372667313,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 115095
+    },
+    {
+      "epoch": 11.829188078108942,
+      "grad_norm": 0.03705834597349167,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 115098
+    },
+    {
+      "epoch": 11.829496402877698,
+      "grad_norm": 0.1142435371875763,
+      "learning_rate": 0.01,
+      "loss": 1.9625,
+      "step": 115101
+    },
+    {
+      "epoch": 11.829804727646454,
+      "grad_norm": 0.0575905367732048,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 115104
+    },
+    {
+      "epoch": 11.83011305241521,
+      "grad_norm": 0.06506889313459396,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 115107
+    },
+    {
+      "epoch": 11.830421377183967,
+      "grad_norm": 0.04557286947965622,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 115110
+    },
+    {
+      "epoch": 11.830729701952723,
+      "grad_norm": 0.04787040501832962,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 115113
+    },
+    {
+      "epoch": 11.83103802672148,
+      "grad_norm": 0.035487424582242966,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 115116
+    },
+    {
+      "epoch": 11.831346351490236,
+      "grad_norm": 0.037838730961084366,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 115119
+    },
+    {
+      "epoch": 11.831654676258992,
+      "grad_norm": 0.044327255338430405,
+      "learning_rate": 0.01,
+      "loss": 1.9643,
+      "step": 115122
+    },
+    {
+      "epoch": 11.831963001027749,
+      "grad_norm": 0.08877101540565491,
+      "learning_rate": 0.01,
+      "loss": 1.9068,
+      "step": 115125
+    },
+    {
+      "epoch": 11.832271325796505,
+      "grad_norm": 0.03952403739094734,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 115128
+    },
+    {
+      "epoch": 11.832579650565261,
+      "grad_norm": 0.07244662195444107,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 115131
+    },
+    {
+      "epoch": 11.83288797533402,
+      "grad_norm": 0.10626883059740067,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 115134
+    },
+    {
+      "epoch": 11.833196300102776,
+      "grad_norm": 0.08260125666856766,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 115137
+    },
+    {
+      "epoch": 11.833504624871532,
+      "grad_norm": 0.03465457260608673,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 115140
+    },
+    {
+      "epoch": 11.833812949640288,
+      "grad_norm": 0.04090080410242081,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 115143
+    },
+    {
+      "epoch": 11.834121274409044,
+      "grad_norm": 0.056409016251564026,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 115146
+    },
+    {
+      "epoch": 11.8344295991778,
+      "grad_norm": 0.04134396091103554,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 115149
+    },
+    {
+      "epoch": 11.834737923946557,
+      "grad_norm": 0.045888107270002365,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 115152
+    },
+    {
+      "epoch": 11.835046248715313,
+      "grad_norm": 0.06528306007385254,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 115155
+    },
+    {
+      "epoch": 11.83535457348407,
+      "grad_norm": 0.0714694932103157,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 115158
+    },
+    {
+      "epoch": 11.835662898252826,
+      "grad_norm": 0.07249157130718231,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 115161
+    },
+    {
+      "epoch": 11.835971223021582,
+      "grad_norm": 0.06381940096616745,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 115164
+    },
+    {
+      "epoch": 11.836279547790339,
+      "grad_norm": 0.040719326585531235,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 115167
+    },
+    {
+      "epoch": 11.836587872559095,
+      "grad_norm": 0.052242450416088104,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 115170
+    },
+    {
+      "epoch": 11.836896197327851,
+      "grad_norm": 0.07433725148439407,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 115173
+    },
+    {
+      "epoch": 11.83720452209661,
+      "grad_norm": 0.06879228353500366,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 115176
+    },
+    {
+      "epoch": 11.837512846865366,
+      "grad_norm": 0.056126561015844345,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 115179
+    },
+    {
+      "epoch": 11.837821171634122,
+      "grad_norm": 0.051828134804964066,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 115182
+    },
+    {
+      "epoch": 11.838129496402878,
+      "grad_norm": 0.07622215151786804,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 115185
+    },
+    {
+      "epoch": 11.838437821171635,
+      "grad_norm": 0.06668256968259811,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 115188
+    },
+    {
+      "epoch": 11.83874614594039,
+      "grad_norm": 0.11664186418056488,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 115191
+    },
+    {
+      "epoch": 11.839054470709147,
+      "grad_norm": 0.06350251287221909,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 115194
+    },
+    {
+      "epoch": 11.839362795477903,
+      "grad_norm": 0.07587242126464844,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 115197
+    },
+    {
+      "epoch": 11.83967112024666,
+      "grad_norm": 0.11297139525413513,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 115200
+    },
+    {
+      "epoch": 11.839979445015416,
+      "grad_norm": 0.09682600200176239,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 115203
+    },
+    {
+      "epoch": 11.840287769784172,
+      "grad_norm": 0.08607444167137146,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 115206
+    },
+    {
+      "epoch": 11.840596094552929,
+      "grad_norm": 0.06459609419107437,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 115209
+    },
+    {
+      "epoch": 11.840904419321685,
+      "grad_norm": 0.10332488268613815,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 115212
+    },
+    {
+      "epoch": 11.841212744090441,
+      "grad_norm": 0.09943923354148865,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 115215
+    },
+    {
+      "epoch": 11.841521068859198,
+      "grad_norm": 0.04884108528494835,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 115218
+    },
+    {
+      "epoch": 11.841829393627954,
+      "grad_norm": 0.05334654450416565,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 115221
+    },
+    {
+      "epoch": 11.842137718396712,
+      "grad_norm": 0.037848688662052155,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 115224
+    },
+    {
+      "epoch": 11.842446043165468,
+      "grad_norm": 0.05791223794221878,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 115227
+    },
+    {
+      "epoch": 11.842754367934225,
+      "grad_norm": 0.03935936838388443,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 115230
+    },
+    {
+      "epoch": 11.843062692702981,
+      "grad_norm": 0.04526537284255028,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 115233
+    },
+    {
+      "epoch": 11.843371017471737,
+      "grad_norm": 0.0499604307115078,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 115236
+    },
+    {
+      "epoch": 11.843679342240494,
+      "grad_norm": 0.1790214627981186,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 115239
+    },
+    {
+      "epoch": 11.84398766700925,
+      "grad_norm": 0.12452468276023865,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 115242
+    },
+    {
+      "epoch": 11.844295991778006,
+      "grad_norm": 0.08335813879966736,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 115245
+    },
+    {
+      "epoch": 11.844604316546762,
+      "grad_norm": 0.06493064761161804,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 115248
+    },
+    {
+      "epoch": 11.844912641315519,
+      "grad_norm": 0.03797442093491554,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 115251
+    },
+    {
+      "epoch": 11.845220966084275,
+      "grad_norm": 0.04726745933294296,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 115254
+    },
+    {
+      "epoch": 11.845529290853031,
+      "grad_norm": 0.07289905846118927,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 115257
+    },
+    {
+      "epoch": 11.845837615621788,
+      "grad_norm": 0.08225946128368378,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 115260
+    },
+    {
+      "epoch": 11.846145940390544,
+      "grad_norm": 0.09263941645622253,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 115263
+    },
+    {
+      "epoch": 11.8464542651593,
+      "grad_norm": 0.05777449533343315,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 115266
+    },
+    {
+      "epoch": 11.846762589928058,
+      "grad_norm": 0.047394949942827225,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 115269
+    },
+    {
+      "epoch": 11.847070914696815,
+      "grad_norm": 0.046602725982666016,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 115272
+    },
+    {
+      "epoch": 11.847379239465571,
+      "grad_norm": 0.09456834942102432,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 115275
+    },
+    {
+      "epoch": 11.847687564234327,
+      "grad_norm": 0.03587270900607109,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 115278
+    },
+    {
+      "epoch": 11.847995889003084,
+      "grad_norm": 0.059412844479084015,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 115281
+    },
+    {
+      "epoch": 11.84830421377184,
+      "grad_norm": 0.0675816535949707,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 115284
+    },
+    {
+      "epoch": 11.848612538540596,
+      "grad_norm": 0.04191970452666283,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 115287
+    },
+    {
+      "epoch": 11.848920863309353,
+      "grad_norm": 0.039853785187006,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 115290
+    },
+    {
+      "epoch": 11.849229188078109,
+      "grad_norm": 0.1088956668972969,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 115293
+    },
+    {
+      "epoch": 11.849537512846865,
+      "grad_norm": 0.0794067531824112,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 115296
+    },
+    {
+      "epoch": 11.849845837615621,
+      "grad_norm": 0.12512779235839844,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 115299
+    },
+    {
+      "epoch": 11.850154162384378,
+      "grad_norm": 0.11033626645803452,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 115302
+    },
+    {
+      "epoch": 11.850462487153134,
+      "grad_norm": 0.09385468810796738,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 115305
+    },
+    {
+      "epoch": 11.85077081192189,
+      "grad_norm": 0.05897045508027077,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 115308
+    },
+    {
+      "epoch": 11.851079136690647,
+      "grad_norm": 0.05027012526988983,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 115311
+    },
+    {
+      "epoch": 11.851387461459403,
+      "grad_norm": 0.05048549547791481,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 115314
+    },
+    {
+      "epoch": 11.851695786228161,
+      "grad_norm": 0.09872164577245712,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 115317
+    },
+    {
+      "epoch": 11.852004110996917,
+      "grad_norm": 0.04220118373632431,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 115320
+    },
+    {
+      "epoch": 11.852312435765674,
+      "grad_norm": 0.04071341082453728,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 115323
+    },
+    {
+      "epoch": 11.85262076053443,
+      "grad_norm": 0.03733963146805763,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 115326
+    },
+    {
+      "epoch": 11.852929085303186,
+      "grad_norm": 0.053249627351760864,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 115329
+    },
+    {
+      "epoch": 11.853237410071943,
+      "grad_norm": 0.05058839172124863,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 115332
+    },
+    {
+      "epoch": 11.853545734840699,
+      "grad_norm": 0.043892621994018555,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 115335
+    },
+    {
+      "epoch": 11.853854059609455,
+      "grad_norm": 0.07909204810857773,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 115338
+    },
+    {
+      "epoch": 11.854162384378212,
+      "grad_norm": 0.06854996085166931,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 115341
+    },
+    {
+      "epoch": 11.854470709146968,
+      "grad_norm": 0.053314458578825,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 115344
+    },
+    {
+      "epoch": 11.854779033915724,
+      "grad_norm": 0.09922850131988525,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 115347
+    },
+    {
+      "epoch": 11.85508735868448,
+      "grad_norm": 0.08411724865436554,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 115350
+    },
+    {
+      "epoch": 11.855395683453237,
+      "grad_norm": 0.055205248296260834,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 115353
+    },
+    {
+      "epoch": 11.855704008221993,
+      "grad_norm": 0.038524750620126724,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 115356
+    },
+    {
+      "epoch": 11.856012332990751,
+      "grad_norm": 0.039631303399801254,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 115359
+    },
+    {
+      "epoch": 11.856320657759507,
+      "grad_norm": 0.10896901041269302,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 115362
+    },
+    {
+      "epoch": 11.856628982528264,
+      "grad_norm": 0.051289912313222885,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 115365
+    },
+    {
+      "epoch": 11.85693730729702,
+      "grad_norm": 0.1080206036567688,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 115368
+    },
+    {
+      "epoch": 11.857245632065776,
+      "grad_norm": 0.037047237157821655,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 115371
+    },
+    {
+      "epoch": 11.857553956834533,
+      "grad_norm": 0.10592879354953766,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 115374
+    },
+    {
+      "epoch": 11.857862281603289,
+      "grad_norm": 0.06507657468318939,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 115377
+    },
+    {
+      "epoch": 11.858170606372045,
+      "grad_norm": 0.04487865790724754,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 115380
+    },
+    {
+      "epoch": 11.858478931140802,
+      "grad_norm": 0.05850360170006752,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 115383
+    },
+    {
+      "epoch": 11.858787255909558,
+      "grad_norm": 0.08132300525903702,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 115386
+    },
+    {
+      "epoch": 11.859095580678314,
+      "grad_norm": 0.08815014362335205,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 115389
+    },
+    {
+      "epoch": 11.85940390544707,
+      "grad_norm": 0.05203649029135704,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 115392
+    },
+    {
+      "epoch": 11.859712230215827,
+      "grad_norm": 0.07548944652080536,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 115395
+    },
+    {
+      "epoch": 11.860020554984583,
+      "grad_norm": 0.04299336299300194,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 115398
+    },
+    {
+      "epoch": 11.86032887975334,
+      "grad_norm": 0.10916072130203247,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 115401
+    },
+    {
+      "epoch": 11.860637204522096,
+      "grad_norm": 0.06974140554666519,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 115404
+    },
+    {
+      "epoch": 11.860945529290852,
+      "grad_norm": 0.10091980546712875,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 115407
+    },
+    {
+      "epoch": 11.86125385405961,
+      "grad_norm": 0.042180851101875305,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 115410
+    },
+    {
+      "epoch": 11.861562178828366,
+      "grad_norm": 0.08566635102033615,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 115413
+    },
+    {
+      "epoch": 11.861870503597123,
+      "grad_norm": 0.050220876932144165,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 115416
+    },
+    {
+      "epoch": 11.86217882836588,
+      "grad_norm": 0.04333271086215973,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 115419
+    },
+    {
+      "epoch": 11.862487153134635,
+      "grad_norm": 0.039936210960149765,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 115422
+    },
+    {
+      "epoch": 11.862795477903392,
+      "grad_norm": 0.04459049925208092,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 115425
+    },
+    {
+      "epoch": 11.863103802672148,
+      "grad_norm": 0.06823472678661346,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 115428
+    },
+    {
+      "epoch": 11.863412127440904,
+      "grad_norm": 0.08977038413286209,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 115431
+    },
+    {
+      "epoch": 11.86372045220966,
+      "grad_norm": 0.1401505172252655,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 115434
+    },
+    {
+      "epoch": 11.864028776978417,
+      "grad_norm": 0.09276049584150314,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 115437
+    },
+    {
+      "epoch": 11.864337101747173,
+      "grad_norm": 0.06281229853630066,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 115440
+    },
+    {
+      "epoch": 11.86464542651593,
+      "grad_norm": 0.04619922488927841,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 115443
+    },
+    {
+      "epoch": 11.864953751284686,
+      "grad_norm": 0.07578086107969284,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 115446
+    },
+    {
+      "epoch": 11.865262076053442,
+      "grad_norm": 0.0675554946064949,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 115449
+    },
+    {
+      "epoch": 11.8655704008222,
+      "grad_norm": 0.04776846617460251,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 115452
+    },
+    {
+      "epoch": 11.865878725590957,
+      "grad_norm": 0.04917658120393753,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 115455
+    },
+    {
+      "epoch": 11.866187050359713,
+      "grad_norm": 0.04377268627285957,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 115458
+    },
+    {
+      "epoch": 11.86649537512847,
+      "grad_norm": 0.10865142196416855,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 115461
+    },
+    {
+      "epoch": 11.866803699897225,
+      "grad_norm": 0.10919032245874405,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 115464
+    },
+    {
+      "epoch": 11.867112024665982,
+      "grad_norm": 0.14149057865142822,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 115467
+    },
+    {
+      "epoch": 11.867420349434738,
+      "grad_norm": 0.05841014161705971,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 115470
+    },
+    {
+      "epoch": 11.867728674203494,
+      "grad_norm": 0.05625878646969795,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 115473
+    },
+    {
+      "epoch": 11.86803699897225,
+      "grad_norm": 0.048125796020030975,
+      "learning_rate": 0.01,
+      "loss": 1.9667,
+      "step": 115476
+    },
+    {
+      "epoch": 11.868345323741007,
+      "grad_norm": 0.06532830744981766,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 115479
+    },
+    {
+      "epoch": 11.868653648509763,
+      "grad_norm": 0.03691454976797104,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 115482
+    },
+    {
+      "epoch": 11.86896197327852,
+      "grad_norm": 0.054014936089515686,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 115485
+    },
+    {
+      "epoch": 11.869270298047276,
+      "grad_norm": 0.0547962561249733,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 115488
+    },
+    {
+      "epoch": 11.869578622816032,
+      "grad_norm": 0.10535205155611038,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 115491
+    },
+    {
+      "epoch": 11.869886947584789,
+      "grad_norm": 0.035013020038604736,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 115494
+    },
+    {
+      "epoch": 11.870195272353545,
+      "grad_norm": 0.03851677477359772,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 115497
+    },
+    {
+      "epoch": 11.870503597122303,
+      "grad_norm": 0.0499538853764534,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 115500
+    },
+    {
+      "epoch": 11.87081192189106,
+      "grad_norm": 0.0537400096654892,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 115503
+    },
+    {
+      "epoch": 11.871120246659816,
+      "grad_norm": 0.07391097396612167,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 115506
+    },
+    {
+      "epoch": 11.871428571428572,
+      "grad_norm": 0.07708144932985306,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 115509
+    },
+    {
+      "epoch": 11.871736896197328,
+      "grad_norm": 0.05969472602009773,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 115512
+    },
+    {
+      "epoch": 11.872045220966084,
+      "grad_norm": 0.042705025523900986,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 115515
+    },
+    {
+      "epoch": 11.87235354573484,
+      "grad_norm": 0.05245598033070564,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 115518
+    },
+    {
+      "epoch": 11.872661870503597,
+      "grad_norm": 0.06938184052705765,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 115521
+    },
+    {
+      "epoch": 11.872970195272353,
+      "grad_norm": 0.039653230458498,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 115524
+    },
+    {
+      "epoch": 11.87327852004111,
+      "grad_norm": 0.10245396196842194,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 115527
+    },
+    {
+      "epoch": 11.873586844809866,
+      "grad_norm": 0.050254181027412415,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 115530
+    },
+    {
+      "epoch": 11.873895169578622,
+      "grad_norm": 0.09232769906520844,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 115533
+    },
+    {
+      "epoch": 11.874203494347379,
+      "grad_norm": 0.06693422049283981,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 115536
+    },
+    {
+      "epoch": 11.874511819116135,
+      "grad_norm": 0.06777456402778625,
+      "learning_rate": 0.01,
+      "loss": 1.9585,
+      "step": 115539
+    },
+    {
+      "epoch": 11.874820143884891,
+      "grad_norm": 0.046546339988708496,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 115542
+    },
+    {
+      "epoch": 11.87512846865365,
+      "grad_norm": 0.09495719522237778,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 115545
+    },
+    {
+      "epoch": 11.875436793422406,
+      "grad_norm": 0.034482963383197784,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 115548
+    },
+    {
+      "epoch": 11.875745118191162,
+      "grad_norm": 0.060760658234357834,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 115551
+    },
+    {
+      "epoch": 11.876053442959918,
+      "grad_norm": 0.04488319158554077,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 115554
+    },
+    {
+      "epoch": 11.876361767728675,
+      "grad_norm": 0.07081051170825958,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 115557
+    },
+    {
+      "epoch": 11.87667009249743,
+      "grad_norm": 0.03904644399881363,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 115560
+    },
+    {
+      "epoch": 11.876978417266187,
+      "grad_norm": 0.07820060104131699,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 115563
+    },
+    {
+      "epoch": 11.877286742034944,
+      "grad_norm": 0.04241922125220299,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 115566
+    },
+    {
+      "epoch": 11.8775950668037,
+      "grad_norm": 0.06383369117975235,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 115569
+    },
+    {
+      "epoch": 11.877903391572456,
+      "grad_norm": 0.07734183967113495,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 115572
+    },
+    {
+      "epoch": 11.878211716341212,
+      "grad_norm": 0.08373697102069855,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 115575
+    },
+    {
+      "epoch": 11.878520041109969,
+      "grad_norm": 0.0456833653151989,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 115578
+    },
+    {
+      "epoch": 11.878828365878725,
+      "grad_norm": 0.05270833522081375,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 115581
+    },
+    {
+      "epoch": 11.879136690647481,
+      "grad_norm": 0.0652565062046051,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 115584
+    },
+    {
+      "epoch": 11.879445015416238,
+      "grad_norm": 0.09808563441038132,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 115587
+    },
+    {
+      "epoch": 11.879753340184994,
+      "grad_norm": 0.06954976171255112,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 115590
+    },
+    {
+      "epoch": 11.880061664953752,
+      "grad_norm": 0.0871601328253746,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 115593
+    },
+    {
+      "epoch": 11.880369989722508,
+      "grad_norm": 0.033956415951251984,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 115596
+    },
+    {
+      "epoch": 11.880678314491265,
+      "grad_norm": 0.045776210725307465,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 115599
+    },
+    {
+      "epoch": 11.880986639260021,
+      "grad_norm": 0.03411071375012398,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 115602
+    },
+    {
+      "epoch": 11.881294964028777,
+      "grad_norm": 0.09710346907377243,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 115605
+    },
+    {
+      "epoch": 11.881603288797534,
+      "grad_norm": 0.08367155492305756,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 115608
+    },
+    {
+      "epoch": 11.88191161356629,
+      "grad_norm": 0.10393861681222916,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 115611
+    },
+    {
+      "epoch": 11.882219938335046,
+      "grad_norm": 0.12665413320064545,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 115614
+    },
+    {
+      "epoch": 11.882528263103803,
+      "grad_norm": 0.03951384499669075,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 115617
+    },
+    {
+      "epoch": 11.882836587872559,
+      "grad_norm": 0.04121113941073418,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 115620
+    },
+    {
+      "epoch": 11.883144912641315,
+      "grad_norm": 0.034584417939186096,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 115623
+    },
+    {
+      "epoch": 11.883453237410071,
+      "grad_norm": 0.03947766497731209,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 115626
+    },
+    {
+      "epoch": 11.883761562178828,
+      "grad_norm": 0.13804352283477783,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 115629
+    },
+    {
+      "epoch": 11.884069886947584,
+      "grad_norm": 0.04128848761320114,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 115632
+    },
+    {
+      "epoch": 11.884378211716342,
+      "grad_norm": 0.11423856765031815,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 115635
+    },
+    {
+      "epoch": 11.884686536485098,
+      "grad_norm": 0.08762546628713608,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 115638
+    },
+    {
+      "epoch": 11.884994861253855,
+      "grad_norm": 0.058912597596645355,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 115641
+    },
+    {
+      "epoch": 11.885303186022611,
+      "grad_norm": 0.05286059528589249,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 115644
+    },
+    {
+      "epoch": 11.885611510791367,
+      "grad_norm": 0.04823709651827812,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 115647
+    },
+    {
+      "epoch": 11.885919835560124,
+      "grad_norm": 0.03591940179467201,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 115650
+    },
+    {
+      "epoch": 11.88622816032888,
+      "grad_norm": 0.03444457799196243,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 115653
+    },
+    {
+      "epoch": 11.886536485097636,
+      "grad_norm": 0.03756314516067505,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 115656
+    },
+    {
+      "epoch": 11.886844809866393,
+      "grad_norm": 0.0774245411157608,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 115659
+    },
+    {
+      "epoch": 11.887153134635149,
+      "grad_norm": 0.08102896064519882,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 115662
+    },
+    {
+      "epoch": 11.887461459403905,
+      "grad_norm": 0.041952721774578094,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 115665
+    },
+    {
+      "epoch": 11.887769784172662,
+      "grad_norm": 0.04018309712409973,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 115668
+    },
+    {
+      "epoch": 11.888078108941418,
+      "grad_norm": 0.037420839071273804,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 115671
+    },
+    {
+      "epoch": 11.888386433710174,
+      "grad_norm": 0.04719531536102295,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 115674
+    },
+    {
+      "epoch": 11.88869475847893,
+      "grad_norm": 0.1673678457736969,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 115677
+    },
+    {
+      "epoch": 11.889003083247687,
+      "grad_norm": 0.1612764298915863,
+      "learning_rate": 0.01,
+      "loss": 1.9648,
+      "step": 115680
+    },
+    {
+      "epoch": 11.889311408016445,
+      "grad_norm": 0.09129545837640762,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 115683
+    },
+    {
+      "epoch": 11.889619732785201,
+      "grad_norm": 0.07026642560958862,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 115686
+    },
+    {
+      "epoch": 11.889928057553957,
+      "grad_norm": 0.057349856942892075,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 115689
+    },
+    {
+      "epoch": 11.890236382322714,
+      "grad_norm": 0.08883088082075119,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 115692
+    },
+    {
+      "epoch": 11.89054470709147,
+      "grad_norm": 0.0671309158205986,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 115695
+    },
+    {
+      "epoch": 11.890853031860226,
+      "grad_norm": 0.15628202259540558,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 115698
+    },
+    {
+      "epoch": 11.891161356628983,
+      "grad_norm": 0.08999011665582657,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 115701
+    },
+    {
+      "epoch": 11.891469681397739,
+      "grad_norm": 0.036552269011735916,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 115704
+    },
+    {
+      "epoch": 11.891778006166495,
+      "grad_norm": 0.04400331899523735,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 115707
+    },
+    {
+      "epoch": 11.892086330935252,
+      "grad_norm": 0.03440818563103676,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 115710
+    },
+    {
+      "epoch": 11.892394655704008,
+      "grad_norm": 0.03784843161702156,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 115713
+    },
+    {
+      "epoch": 11.892702980472764,
+      "grad_norm": 0.038593459874391556,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 115716
+    },
+    {
+      "epoch": 11.89301130524152,
+      "grad_norm": 0.04749274253845215,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 115719
+    },
+    {
+      "epoch": 11.893319630010277,
+      "grad_norm": 0.13267368078231812,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 115722
+    },
+    {
+      "epoch": 11.893627954779033,
+      "grad_norm": 0.11221770942211151,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 115725
+    },
+    {
+      "epoch": 11.893936279547791,
+      "grad_norm": 0.04560906067490578,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 115728
+    },
+    {
+      "epoch": 11.894244604316548,
+      "grad_norm": 0.059753429144620895,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 115731
+    },
+    {
+      "epoch": 11.894552929085304,
+      "grad_norm": 0.041221734136343,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 115734
+    },
+    {
+      "epoch": 11.89486125385406,
+      "grad_norm": 0.044288646429777145,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 115737
+    },
+    {
+      "epoch": 11.895169578622816,
+      "grad_norm": 0.03386671096086502,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 115740
+    },
+    {
+      "epoch": 11.895477903391573,
+      "grad_norm": 0.04667489603161812,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 115743
+    },
+    {
+      "epoch": 11.895786228160329,
+      "grad_norm": 0.0888177677989006,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 115746
+    },
+    {
+      "epoch": 11.896094552929085,
+      "grad_norm": 0.09287005662918091,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 115749
+    },
+    {
+      "epoch": 11.896402877697842,
+      "grad_norm": 0.05758680775761604,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 115752
+    },
+    {
+      "epoch": 11.896711202466598,
+      "grad_norm": 0.0726981908082962,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 115755
+    },
+    {
+      "epoch": 11.897019527235354,
+      "grad_norm": 0.05644940957427025,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 115758
+    },
+    {
+      "epoch": 11.89732785200411,
+      "grad_norm": 0.14221036434173584,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 115761
+    },
+    {
+      "epoch": 11.897636176772867,
+      "grad_norm": 0.11654707789421082,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 115764
+    },
+    {
+      "epoch": 11.897944501541623,
+      "grad_norm": 0.036727387458086014,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 115767
+    },
+    {
+      "epoch": 11.89825282631038,
+      "grad_norm": 0.04845240339636803,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 115770
+    },
+    {
+      "epoch": 11.898561151079136,
+      "grad_norm": 0.05198315903544426,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 115773
+    },
+    {
+      "epoch": 11.898869475847894,
+      "grad_norm": 0.0603262297809124,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 115776
+    },
+    {
+      "epoch": 11.89917780061665,
+      "grad_norm": 0.037290554493665695,
+      "learning_rate": 0.01,
+      "loss": 1.8963,
+      "step": 115779
+    },
+    {
+      "epoch": 11.899486125385407,
+      "grad_norm": 0.04267965257167816,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 115782
+    },
+    {
+      "epoch": 11.899794450154163,
+      "grad_norm": 0.0385502353310585,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 115785
+    },
+    {
+      "epoch": 11.90010277492292,
+      "grad_norm": 0.07459340244531631,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 115788
+    },
+    {
+      "epoch": 11.900411099691675,
+      "grad_norm": 0.07064307481050491,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 115791
+    },
+    {
+      "epoch": 11.900719424460432,
+      "grad_norm": 0.08032072335481644,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 115794
+    },
+    {
+      "epoch": 11.901027749229188,
+      "grad_norm": 0.04520730301737785,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 115797
+    },
+    {
+      "epoch": 11.901336073997944,
+      "grad_norm": 0.08594592660665512,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 115800
+    },
+    {
+      "epoch": 11.9016443987667,
+      "grad_norm": 0.042838793247938156,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 115803
+    },
+    {
+      "epoch": 11.901952723535457,
+      "grad_norm": 0.03810225799679756,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 115806
+    },
+    {
+      "epoch": 11.902261048304213,
+      "grad_norm": 0.06649664044380188,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 115809
+    },
+    {
+      "epoch": 11.90256937307297,
+      "grad_norm": 0.08439337462186813,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 115812
+    },
+    {
+      "epoch": 11.902877697841726,
+      "grad_norm": 0.10588230937719345,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 115815
+    },
+    {
+      "epoch": 11.903186022610484,
+      "grad_norm": 0.15726368129253387,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 115818
+    },
+    {
+      "epoch": 11.90349434737924,
+      "grad_norm": 0.09451795369386673,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 115821
+    },
+    {
+      "epoch": 11.903802672147997,
+      "grad_norm": 0.048060912638902664,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 115824
+    },
+    {
+      "epoch": 11.904110996916753,
+      "grad_norm": 0.03914719447493553,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 115827
+    },
+    {
+      "epoch": 11.90441932168551,
+      "grad_norm": 0.07605911046266556,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 115830
+    },
+    {
+      "epoch": 11.904727646454266,
+      "grad_norm": 0.029742151498794556,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 115833
+    },
+    {
+      "epoch": 11.905035971223022,
+      "grad_norm": 0.11301050335168839,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 115836
+    },
+    {
+      "epoch": 11.905344295991778,
+      "grad_norm": 0.08454637229442596,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 115839
+    },
+    {
+      "epoch": 11.905652620760534,
+      "grad_norm": 0.0823889970779419,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 115842
+    },
+    {
+      "epoch": 11.90596094552929,
+      "grad_norm": 0.0720774456858635,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 115845
+    },
+    {
+      "epoch": 11.906269270298047,
+      "grad_norm": 0.05574191361665726,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 115848
+    },
+    {
+      "epoch": 11.906577595066803,
+      "grad_norm": 0.06776542216539383,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 115851
+    },
+    {
+      "epoch": 11.90688591983556,
+      "grad_norm": 0.04499393701553345,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 115854
+    },
+    {
+      "epoch": 11.907194244604316,
+      "grad_norm": 0.030573226511478424,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 115857
+    },
+    {
+      "epoch": 11.907502569373072,
+      "grad_norm": 0.05911063402891159,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 115860
+    },
+    {
+      "epoch": 11.907810894141829,
+      "grad_norm": 0.04587172716856003,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 115863
+    },
+    {
+      "epoch": 11.908119218910585,
+      "grad_norm": 0.08495959639549255,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 115866
+    },
+    {
+      "epoch": 11.908427543679343,
+      "grad_norm": 0.11106112599372864,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 115869
+    },
+    {
+      "epoch": 11.9087358684481,
+      "grad_norm": 0.04350137710571289,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 115872
+    },
+    {
+      "epoch": 11.909044193216856,
+      "grad_norm": 0.12991520762443542,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 115875
+    },
+    {
+      "epoch": 11.909352517985612,
+      "grad_norm": 0.0833691954612732,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 115878
+    },
+    {
+      "epoch": 11.909660842754368,
+      "grad_norm": 0.04312315955758095,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 115881
+    },
+    {
+      "epoch": 11.909969167523125,
+      "grad_norm": 0.04676743596792221,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 115884
+    },
+    {
+      "epoch": 11.91027749229188,
+      "grad_norm": 0.0459318682551384,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 115887
+    },
+    {
+      "epoch": 11.910585817060637,
+      "grad_norm": 0.03991396725177765,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 115890
+    },
+    {
+      "epoch": 11.910894141829393,
+      "grad_norm": 0.05674207583069801,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 115893
+    },
+    {
+      "epoch": 11.91120246659815,
+      "grad_norm": 0.08130472898483276,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 115896
+    },
+    {
+      "epoch": 11.911510791366906,
+      "grad_norm": 0.14991524815559387,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 115899
+    },
+    {
+      "epoch": 11.911819116135662,
+      "grad_norm": 0.06426069140434265,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 115902
+    },
+    {
+      "epoch": 11.912127440904419,
+      "grad_norm": 0.046491220593452454,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 115905
+    },
+    {
+      "epoch": 11.912435765673175,
+      "grad_norm": 0.04492336884140968,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 115908
+    },
+    {
+      "epoch": 11.912744090441933,
+      "grad_norm": 0.04138762131333351,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 115911
+    },
+    {
+      "epoch": 11.91305241521069,
+      "grad_norm": 0.07580280303955078,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 115914
+    },
+    {
+      "epoch": 11.913360739979446,
+      "grad_norm": 0.09959438443183899,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 115917
+    },
+    {
+      "epoch": 11.913669064748202,
+      "grad_norm": 0.035286176949739456,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 115920
+    },
+    {
+      "epoch": 11.913977389516958,
+      "grad_norm": 0.089673712849617,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 115923
+    },
+    {
+      "epoch": 11.914285714285715,
+      "grad_norm": 0.09498865902423859,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 115926
+    },
+    {
+      "epoch": 11.914594039054471,
+      "grad_norm": 0.07095696777105331,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 115929
+    },
+    {
+      "epoch": 11.914902363823227,
+      "grad_norm": 0.11000113189220428,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 115932
+    },
+    {
+      "epoch": 11.915210688591984,
+      "grad_norm": 0.049887191504240036,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 115935
+    },
+    {
+      "epoch": 11.91551901336074,
+      "grad_norm": 0.03578448295593262,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 115938
+    },
+    {
+      "epoch": 11.915827338129496,
+      "grad_norm": 0.03840559720993042,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 115941
+    },
+    {
+      "epoch": 11.916135662898252,
+      "grad_norm": 0.08180845528841019,
+      "learning_rate": 0.01,
+      "loss": 1.9774,
+      "step": 115944
+    },
+    {
+      "epoch": 11.916443987667009,
+      "grad_norm": 0.08446424454450607,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 115947
+    },
+    {
+      "epoch": 11.916752312435765,
+      "grad_norm": 0.06296598166227341,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 115950
+    },
+    {
+      "epoch": 11.917060637204521,
+      "grad_norm": 0.09014962613582611,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 115953
+    },
+    {
+      "epoch": 11.917368961973278,
+      "grad_norm": 0.060182347893714905,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 115956
+    },
+    {
+      "epoch": 11.917677286742036,
+      "grad_norm": 0.07497214525938034,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 115959
+    },
+    {
+      "epoch": 11.917985611510792,
+      "grad_norm": 0.0734238550066948,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 115962
+    },
+    {
+      "epoch": 11.918293936279548,
+      "grad_norm": 0.09660092741250992,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 115965
+    },
+    {
+      "epoch": 11.918602261048305,
+      "grad_norm": 0.12239353358745575,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 115968
+    },
+    {
+      "epoch": 11.918910585817061,
+      "grad_norm": 0.09368391335010529,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 115971
+    },
+    {
+      "epoch": 11.919218910585817,
+      "grad_norm": 0.07862232625484467,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 115974
+    },
+    {
+      "epoch": 11.919527235354574,
+      "grad_norm": 0.034651078283786774,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 115977
+    },
+    {
+      "epoch": 11.91983556012333,
+      "grad_norm": 0.12038659304380417,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 115980
+    },
+    {
+      "epoch": 11.920143884892086,
+      "grad_norm": 0.033229418098926544,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 115983
+    },
+    {
+      "epoch": 11.920452209660843,
+      "grad_norm": 0.07518643885850906,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 115986
+    },
+    {
+      "epoch": 11.920760534429599,
+      "grad_norm": 0.08469729870557785,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 115989
+    },
+    {
+      "epoch": 11.921068859198355,
+      "grad_norm": 0.051580049097537994,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 115992
+    },
+    {
+      "epoch": 11.921377183967111,
+      "grad_norm": 0.11890963464975357,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 115995
+    },
+    {
+      "epoch": 11.921685508735868,
+      "grad_norm": 0.04845276474952698,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 115998
+    },
+    {
+      "epoch": 11.921993833504624,
+      "grad_norm": 0.0875236839056015,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 116001
+    },
+    {
+      "epoch": 11.922302158273382,
+      "grad_norm": 0.07505252212285995,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 116004
+    },
+    {
+      "epoch": 11.922610483042138,
+      "grad_norm": 0.06646228581666946,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 116007
+    },
+    {
+      "epoch": 11.922918807810895,
+      "grad_norm": 0.08600006252527237,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 116010
+    },
+    {
+      "epoch": 11.923227132579651,
+      "grad_norm": 0.11209333688020706,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 116013
+    },
+    {
+      "epoch": 11.923535457348407,
+      "grad_norm": 0.10353491455316544,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 116016
+    },
+    {
+      "epoch": 11.923843782117164,
+      "grad_norm": 0.14115461707115173,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 116019
+    },
+    {
+      "epoch": 11.92415210688592,
+      "grad_norm": 0.044179707765579224,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 116022
+    },
+    {
+      "epoch": 11.924460431654676,
+      "grad_norm": 0.03707960247993469,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 116025
+    },
+    {
+      "epoch": 11.924768756423433,
+      "grad_norm": 0.04922756925225258,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 116028
+    },
+    {
+      "epoch": 11.925077081192189,
+      "grad_norm": 0.06139388680458069,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 116031
+    },
+    {
+      "epoch": 11.925385405960945,
+      "grad_norm": 0.09925854951143265,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 116034
+    },
+    {
+      "epoch": 11.925693730729702,
+      "grad_norm": 0.08050132542848587,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 116037
+    },
+    {
+      "epoch": 11.926002055498458,
+      "grad_norm": 0.08102965354919434,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 116040
+    },
+    {
+      "epoch": 11.926310380267214,
+      "grad_norm": 0.11337810754776001,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 116043
+    },
+    {
+      "epoch": 11.92661870503597,
+      "grad_norm": 0.05181996151804924,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 116046
+    },
+    {
+      "epoch": 11.926927029804727,
+      "grad_norm": 0.07451339066028595,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 116049
+    },
+    {
+      "epoch": 11.927235354573485,
+      "grad_norm": 0.04624432325363159,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 116052
+    },
+    {
+      "epoch": 11.927543679342241,
+      "grad_norm": 0.1001310721039772,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 116055
+    },
+    {
+      "epoch": 11.927852004110997,
+      "grad_norm": 0.05606957897543907,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 116058
+    },
+    {
+      "epoch": 11.928160328879754,
+      "grad_norm": 0.08710836619138718,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 116061
+    },
+    {
+      "epoch": 11.92846865364851,
+      "grad_norm": 0.05734976381063461,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 116064
+    },
+    {
+      "epoch": 11.928776978417266,
+      "grad_norm": 0.06889203935861588,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 116067
+    },
+    {
+      "epoch": 11.929085303186023,
+      "grad_norm": 0.07249484211206436,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 116070
+    },
+    {
+      "epoch": 11.929393627954779,
+      "grad_norm": 0.038481082767248154,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 116073
+    },
+    {
+      "epoch": 11.929701952723535,
+      "grad_norm": 0.040560442954301834,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 116076
+    },
+    {
+      "epoch": 11.930010277492292,
+      "grad_norm": 0.04894191026687622,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 116079
+    },
+    {
+      "epoch": 11.930318602261048,
+      "grad_norm": 0.05847233161330223,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 116082
+    },
+    {
+      "epoch": 11.930626927029804,
+      "grad_norm": 0.05613159015774727,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 116085
+    },
+    {
+      "epoch": 11.93093525179856,
+      "grad_norm": 0.03757588565349579,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 116088
+    },
+    {
+      "epoch": 11.931243576567317,
+      "grad_norm": 0.11519763618707657,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 116091
+    },
+    {
+      "epoch": 11.931551901336075,
+      "grad_norm": 0.08751515299081802,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 116094
+    },
+    {
+      "epoch": 11.931860226104831,
+      "grad_norm": 0.14398428797721863,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 116097
+    },
+    {
+      "epoch": 11.932168550873588,
+      "grad_norm": 0.18179289996623993,
+      "learning_rate": 0.01,
+      "loss": 1.9604,
+      "step": 116100
+    },
+    {
+      "epoch": 11.932476875642344,
+      "grad_norm": 0.13266566395759583,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 116103
+    },
+    {
+      "epoch": 11.9327852004111,
+      "grad_norm": 0.061238087713718414,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 116106
+    },
+    {
+      "epoch": 11.933093525179856,
+      "grad_norm": 0.04893723875284195,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 116109
+    },
+    {
+      "epoch": 11.933401849948613,
+      "grad_norm": 0.04873514920473099,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 116112
+    },
+    {
+      "epoch": 11.933710174717369,
+      "grad_norm": 0.07248979061841965,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 116115
+    },
+    {
+      "epoch": 11.934018499486125,
+      "grad_norm": 0.04428212717175484,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 116118
+    },
+    {
+      "epoch": 11.934326824254882,
+      "grad_norm": 0.11622639745473862,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 116121
+    },
+    {
+      "epoch": 11.934635149023638,
+      "grad_norm": 0.10958331823348999,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 116124
+    },
+    {
+      "epoch": 11.934943473792394,
+      "grad_norm": 0.11636814475059509,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 116127
+    },
+    {
+      "epoch": 11.93525179856115,
+      "grad_norm": 0.04917363077402115,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 116130
+    },
+    {
+      "epoch": 11.935560123329907,
+      "grad_norm": 0.04304635524749756,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 116133
+    },
+    {
+      "epoch": 11.935868448098663,
+      "grad_norm": 0.0436469204723835,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 116136
+    },
+    {
+      "epoch": 11.93617677286742,
+      "grad_norm": 0.06456790119409561,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 116139
+    },
+    {
+      "epoch": 11.936485097636178,
+      "grad_norm": 0.09960170090198517,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 116142
+    },
+    {
+      "epoch": 11.936793422404934,
+      "grad_norm": 0.12316549569368362,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 116145
+    },
+    {
+      "epoch": 11.93710174717369,
+      "grad_norm": 0.044478122144937515,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 116148
+    },
+    {
+      "epoch": 11.937410071942447,
+      "grad_norm": 0.048924773931503296,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 116151
+    },
+    {
+      "epoch": 11.937718396711203,
+      "grad_norm": 0.04412318766117096,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 116154
+    },
+    {
+      "epoch": 11.93802672147996,
+      "grad_norm": 0.05070868134498596,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 116157
+    },
+    {
+      "epoch": 11.938335046248715,
+      "grad_norm": 0.06851394474506378,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 116160
+    },
+    {
+      "epoch": 11.938643371017472,
+      "grad_norm": 0.0725923627614975,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 116163
+    },
+    {
+      "epoch": 11.938951695786228,
+      "grad_norm": 0.09985040128231049,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 116166
+    },
+    {
+      "epoch": 11.939260020554984,
+      "grad_norm": 0.09955912828445435,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 116169
+    },
+    {
+      "epoch": 11.93956834532374,
+      "grad_norm": 0.0586288683116436,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 116172
+    },
+    {
+      "epoch": 11.939876670092497,
+      "grad_norm": 0.09830272942781448,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 116175
+    },
+    {
+      "epoch": 11.940184994861253,
+      "grad_norm": 0.08258133381605148,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 116178
+    },
+    {
+      "epoch": 11.94049331963001,
+      "grad_norm": 0.036119941622018814,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 116181
+    },
+    {
+      "epoch": 11.940801644398766,
+      "grad_norm": 0.059807684272527695,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 116184
+    },
+    {
+      "epoch": 11.941109969167524,
+      "grad_norm": 0.06625748425722122,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 116187
+    },
+    {
+      "epoch": 11.94141829393628,
+      "grad_norm": 0.057437337934970856,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 116190
+    },
+    {
+      "epoch": 11.941726618705037,
+      "grad_norm": 0.03686125576496124,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 116193
+    },
+    {
+      "epoch": 11.942034943473793,
+      "grad_norm": 0.050753671675920486,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 116196
+    },
+    {
+      "epoch": 11.94234326824255,
+      "grad_norm": 0.05824119970202446,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 116199
+    },
+    {
+      "epoch": 11.942651593011306,
+      "grad_norm": 0.10610711574554443,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 116202
+    },
+    {
+      "epoch": 11.942959917780062,
+      "grad_norm": 0.04334506765007973,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 116205
+    },
+    {
+      "epoch": 11.943268242548818,
+      "grad_norm": 0.04354944825172424,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 116208
+    },
+    {
+      "epoch": 11.943576567317574,
+      "grad_norm": 0.046755727380514145,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 116211
+    },
+    {
+      "epoch": 11.94388489208633,
+      "grad_norm": 0.06991719454526901,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 116214
+    },
+    {
+      "epoch": 11.944193216855087,
+      "grad_norm": 0.06674511730670929,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 116217
+    },
+    {
+      "epoch": 11.944501541623843,
+      "grad_norm": 0.05555736646056175,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 116220
+    },
+    {
+      "epoch": 11.9448098663926,
+      "grad_norm": 0.05003273859620094,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 116223
+    },
+    {
+      "epoch": 11.945118191161356,
+      "grad_norm": 0.13730725646018982,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 116226
+    },
+    {
+      "epoch": 11.945426515930112,
+      "grad_norm": 0.051087841391563416,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 116229
+    },
+    {
+      "epoch": 11.945734840698869,
+      "grad_norm": 0.07299045473337173,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 116232
+    },
+    {
+      "epoch": 11.946043165467627,
+      "grad_norm": 0.03582729771733284,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 116235
+    },
+    {
+      "epoch": 11.946351490236383,
+      "grad_norm": 0.05510307103395462,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 116238
+    },
+    {
+      "epoch": 11.94665981500514,
+      "grad_norm": 0.07253251224756241,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 116241
+    },
+    {
+      "epoch": 11.946968139773896,
+      "grad_norm": 0.08624812960624695,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 116244
+    },
+    {
+      "epoch": 11.947276464542652,
+      "grad_norm": 0.05577114224433899,
+      "learning_rate": 0.01,
+      "loss": 1.8999,
+      "step": 116247
+    },
+    {
+      "epoch": 11.947584789311408,
+      "grad_norm": 0.044872984290122986,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 116250
+    },
+    {
+      "epoch": 11.947893114080165,
+      "grad_norm": 0.04620710387825966,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 116253
+    },
+    {
+      "epoch": 11.94820143884892,
+      "grad_norm": 0.08223326504230499,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 116256
+    },
+    {
+      "epoch": 11.948509763617677,
+      "grad_norm": 0.06317196786403656,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 116259
+    },
+    {
+      "epoch": 11.948818088386433,
+      "grad_norm": 0.07849643379449844,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 116262
+    },
+    {
+      "epoch": 11.94912641315519,
+      "grad_norm": 0.12053529173135757,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 116265
+    },
+    {
+      "epoch": 11.949434737923946,
+      "grad_norm": 0.13052164018154144,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 116268
+    },
+    {
+      "epoch": 11.949743062692702,
+      "grad_norm": 0.046053748577833176,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 116271
+    },
+    {
+      "epoch": 11.950051387461459,
+      "grad_norm": 0.04677347093820572,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 116274
+    },
+    {
+      "epoch": 11.950359712230217,
+      "grad_norm": 0.04560299590229988,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 116277
+    },
+    {
+      "epoch": 11.950668036998973,
+      "grad_norm": 0.037877749651670456,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 116280
+    },
+    {
+      "epoch": 11.95097636176773,
+      "grad_norm": 0.04126598313450813,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 116283
+    },
+    {
+      "epoch": 11.951284686536486,
+      "grad_norm": 0.038106292486190796,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 116286
+    },
+    {
+      "epoch": 11.951593011305242,
+      "grad_norm": 0.07549608498811722,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 116289
+    },
+    {
+      "epoch": 11.951901336073998,
+      "grad_norm": 0.09480550140142441,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 116292
+    },
+    {
+      "epoch": 11.952209660842755,
+      "grad_norm": 0.051869966089725494,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 116295
+    },
+    {
+      "epoch": 11.952517985611511,
+      "grad_norm": 0.055569011718034744,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 116298
+    },
+    {
+      "epoch": 11.952826310380267,
+      "grad_norm": 0.04779817909002304,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 116301
+    },
+    {
+      "epoch": 11.953134635149024,
+      "grad_norm": 0.05312659218907356,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 116304
+    },
+    {
+      "epoch": 11.95344295991778,
+      "grad_norm": 0.05799674987792969,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 116307
+    },
+    {
+      "epoch": 11.953751284686536,
+      "grad_norm": 0.047829464077949524,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 116310
+    },
+    {
+      "epoch": 11.954059609455292,
+      "grad_norm": 0.08063242584466934,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 116313
+    },
+    {
+      "epoch": 11.954367934224049,
+      "grad_norm": 0.13868576288223267,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 116316
+    },
+    {
+      "epoch": 11.954676258992805,
+      "grad_norm": 0.06510185450315475,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 116319
+    },
+    {
+      "epoch": 11.954984583761561,
+      "grad_norm": 0.05362662300467491,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 116322
+    },
+    {
+      "epoch": 11.955292908530318,
+      "grad_norm": 0.036775220185518265,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 116325
+    },
+    {
+      "epoch": 11.955601233299076,
+      "grad_norm": 0.04017291218042374,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 116328
+    },
+    {
+      "epoch": 11.955909558067832,
+      "grad_norm": 0.033246539533138275,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 116331
+    },
+    {
+      "epoch": 11.956217882836588,
+      "grad_norm": 0.054019495844841,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 116334
+    },
+    {
+      "epoch": 11.956526207605345,
+      "grad_norm": 0.05341818928718567,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 116337
+    },
+    {
+      "epoch": 11.956834532374101,
+      "grad_norm": 0.06354212015867233,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 116340
+    },
+    {
+      "epoch": 11.957142857142857,
+      "grad_norm": 0.11799897998571396,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 116343
+    },
+    {
+      "epoch": 11.957451181911614,
+      "grad_norm": 0.06504609435796738,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 116346
+    },
+    {
+      "epoch": 11.95775950668037,
+      "grad_norm": 0.05881750211119652,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 116349
+    },
+    {
+      "epoch": 11.958067831449126,
+      "grad_norm": 0.03419358655810356,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 116352
+    },
+    {
+      "epoch": 11.958376156217883,
+      "grad_norm": 0.10960660874843597,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 116355
+    },
+    {
+      "epoch": 11.958684480986639,
+      "grad_norm": 0.0514020100235939,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 116358
+    },
+    {
+      "epoch": 11.958992805755395,
+      "grad_norm": 0.11828702688217163,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 116361
+    },
+    {
+      "epoch": 11.959301130524151,
+      "grad_norm": 0.05309384688735008,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 116364
+    },
+    {
+      "epoch": 11.959609455292908,
+      "grad_norm": 0.05590314790606499,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 116367
+    },
+    {
+      "epoch": 11.959917780061666,
+      "grad_norm": 0.04192624241113663,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 116370
+    },
+    {
+      "epoch": 11.960226104830422,
+      "grad_norm": 0.02918892726302147,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 116373
+    },
+    {
+      "epoch": 11.960534429599178,
+      "grad_norm": 0.09773929417133331,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 116376
+    },
+    {
+      "epoch": 11.960842754367935,
+      "grad_norm": 0.09480267763137817,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 116379
+    },
+    {
+      "epoch": 11.961151079136691,
+      "grad_norm": 0.08526931703090668,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 116382
+    },
+    {
+      "epoch": 11.961459403905447,
+      "grad_norm": 0.08976579457521439,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 116385
+    },
+    {
+      "epoch": 11.961767728674204,
+      "grad_norm": 0.13688208162784576,
+      "learning_rate": 0.01,
+      "loss": 1.9683,
+      "step": 116388
+    },
+    {
+      "epoch": 11.96207605344296,
+      "grad_norm": 0.03766091540455818,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 116391
+    },
+    {
+      "epoch": 11.962384378211716,
+      "grad_norm": 0.047475963830947876,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 116394
+    },
+    {
+      "epoch": 11.962692702980473,
+      "grad_norm": 0.03686054050922394,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 116397
+    },
+    {
+      "epoch": 11.963001027749229,
+      "grad_norm": 0.10608810186386108,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 116400
+    },
+    {
+      "epoch": 11.963309352517985,
+      "grad_norm": 0.04265899956226349,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 116403
+    },
+    {
+      "epoch": 11.963617677286742,
+      "grad_norm": 0.04492899030447006,
+      "learning_rate": 0.01,
+      "loss": 1.9031,
+      "step": 116406
+    },
+    {
+      "epoch": 11.963926002055498,
+      "grad_norm": 0.039105210453271866,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 116409
+    },
+    {
+      "epoch": 11.964234326824254,
+      "grad_norm": 0.036893412470817566,
+      "learning_rate": 0.01,
+      "loss": 1.9943,
+      "step": 116412
+    },
+    {
+      "epoch": 11.96454265159301,
+      "grad_norm": 0.10075349360704422,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 116415
+    },
+    {
+      "epoch": 11.964850976361769,
+      "grad_norm": 0.053778357803821564,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 116418
+    },
+    {
+      "epoch": 11.965159301130525,
+      "grad_norm": 0.07513388991355896,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 116421
+    },
+    {
+      "epoch": 11.965467625899281,
+      "grad_norm": 0.12376239150762558,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 116424
+    },
+    {
+      "epoch": 11.965775950668037,
+      "grad_norm": 0.10657428950071335,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 116427
+    },
+    {
+      "epoch": 11.966084275436794,
+      "grad_norm": 0.038908544927835464,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 116430
+    },
+    {
+      "epoch": 11.96639260020555,
+      "grad_norm": 0.062225233763456345,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 116433
+    },
+    {
+      "epoch": 11.966700924974306,
+      "grad_norm": 0.042596325278282166,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 116436
+    },
+    {
+      "epoch": 11.967009249743063,
+      "grad_norm": 0.05816507712006569,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 116439
+    },
+    {
+      "epoch": 11.967317574511819,
+      "grad_norm": 0.05269140750169754,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 116442
+    },
+    {
+      "epoch": 11.967625899280575,
+      "grad_norm": 0.04918801039457321,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 116445
+    },
+    {
+      "epoch": 11.967934224049332,
+      "grad_norm": 0.03559085726737976,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 116448
+    },
+    {
+      "epoch": 11.968242548818088,
+      "grad_norm": 0.06931116431951523,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 116451
+    },
+    {
+      "epoch": 11.968550873586844,
+      "grad_norm": 0.09284847974777222,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 116454
+    },
+    {
+      "epoch": 11.9688591983556,
+      "grad_norm": 0.07549259811639786,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 116457
+    },
+    {
+      "epoch": 11.969167523124357,
+      "grad_norm": 0.043274421244859695,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 116460
+    },
+    {
+      "epoch": 11.969475847893115,
+      "grad_norm": 0.0987868458032608,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 116463
+    },
+    {
+      "epoch": 11.969784172661871,
+      "grad_norm": 0.0849565714597702,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 116466
+    },
+    {
+      "epoch": 11.970092497430628,
+      "grad_norm": 0.09053468704223633,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 116469
+    },
+    {
+      "epoch": 11.970400822199384,
+      "grad_norm": 0.057200171053409576,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 116472
+    },
+    {
+      "epoch": 11.97070914696814,
+      "grad_norm": 0.10704628378152847,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 116475
+    },
+    {
+      "epoch": 11.971017471736896,
+      "grad_norm": 0.04893108084797859,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 116478
+    },
+    {
+      "epoch": 11.971325796505653,
+      "grad_norm": 0.1119956374168396,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 116481
+    },
+    {
+      "epoch": 11.971634121274409,
+      "grad_norm": 0.13471321761608124,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 116484
+    },
+    {
+      "epoch": 11.971942446043165,
+      "grad_norm": 0.039003241807222366,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 116487
+    },
+    {
+      "epoch": 11.972250770811922,
+      "grad_norm": 0.03899799659848213,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 116490
+    },
+    {
+      "epoch": 11.972559095580678,
+      "grad_norm": 0.04584067687392235,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 116493
+    },
+    {
+      "epoch": 11.972867420349434,
+      "grad_norm": 0.08454135060310364,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 116496
+    },
+    {
+      "epoch": 11.97317574511819,
+      "grad_norm": 0.08052065968513489,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 116499
+    },
+    {
+      "epoch": 11.973484069886947,
+      "grad_norm": 0.04771379008889198,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 116502
+    },
+    {
+      "epoch": 11.973792394655703,
+      "grad_norm": 0.09802628308534622,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 116505
+    },
+    {
+      "epoch": 11.97410071942446,
+      "grad_norm": 0.0905313640832901,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 116508
+    },
+    {
+      "epoch": 11.974409044193218,
+      "grad_norm": 0.05095797777175903,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 116511
+    },
+    {
+      "epoch": 11.974717368961974,
+      "grad_norm": 0.0666579008102417,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 116514
+    },
+    {
+      "epoch": 11.97502569373073,
+      "grad_norm": 0.08291205018758774,
+      "learning_rate": 0.01,
+      "loss": 1.9721,
+      "step": 116517
+    },
+    {
+      "epoch": 11.975334018499487,
+      "grad_norm": 0.1152813658118248,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 116520
+    },
+    {
+      "epoch": 11.975642343268243,
+      "grad_norm": 0.03910317271947861,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 116523
+    },
+    {
+      "epoch": 11.975950668037,
+      "grad_norm": 0.07153347879648209,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 116526
+    },
+    {
+      "epoch": 11.976258992805755,
+      "grad_norm": 0.07718818634748459,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 116529
+    },
+    {
+      "epoch": 11.976567317574512,
+      "grad_norm": 0.08614334464073181,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 116532
+    },
+    {
+      "epoch": 11.976875642343268,
+      "grad_norm": 0.10888314247131348,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 116535
+    },
+    {
+      "epoch": 11.977183967112024,
+      "grad_norm": 0.045839011669158936,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 116538
+    },
+    {
+      "epoch": 11.97749229188078,
+      "grad_norm": 0.05105046182870865,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 116541
+    },
+    {
+      "epoch": 11.977800616649537,
+      "grad_norm": 0.06040375679731369,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 116544
+    },
+    {
+      "epoch": 11.978108941418293,
+      "grad_norm": 0.10920973122119904,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 116547
+    },
+    {
+      "epoch": 11.97841726618705,
+      "grad_norm": 0.10301953554153442,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 116550
+    },
+    {
+      "epoch": 11.978725590955808,
+      "grad_norm": 0.08340315520763397,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 116553
+    },
+    {
+      "epoch": 11.979033915724564,
+      "grad_norm": 0.1261022686958313,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 116556
+    },
+    {
+      "epoch": 11.97934224049332,
+      "grad_norm": 0.0827362909913063,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 116559
+    },
+    {
+      "epoch": 11.979650565262077,
+      "grad_norm": 0.059684157371520996,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 116562
+    },
+    {
+      "epoch": 11.979958890030833,
+      "grad_norm": 0.06773815304040909,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 116565
+    },
+    {
+      "epoch": 11.98026721479959,
+      "grad_norm": 0.0493774376809597,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 116568
+    },
+    {
+      "epoch": 11.980575539568346,
+      "grad_norm": 0.06392575055360794,
+      "learning_rate": 0.01,
+      "loss": 1.9695,
+      "step": 116571
+    },
+    {
+      "epoch": 11.980883864337102,
+      "grad_norm": 0.039536163210868835,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 116574
+    },
+    {
+      "epoch": 11.981192189105858,
+      "grad_norm": 0.038751810789108276,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 116577
+    },
+    {
+      "epoch": 11.981500513874614,
+      "grad_norm": 0.06577613949775696,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 116580
+    },
+    {
+      "epoch": 11.98180883864337,
+      "grad_norm": 0.16617251932621002,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 116583
+    },
+    {
+      "epoch": 11.982117163412127,
+      "grad_norm": 0.11082916706800461,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 116586
+    },
+    {
+      "epoch": 11.982425488180883,
+      "grad_norm": 0.09331133961677551,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 116589
+    },
+    {
+      "epoch": 11.98273381294964,
+      "grad_norm": 0.09698370099067688,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 116592
+    },
+    {
+      "epoch": 11.983042137718396,
+      "grad_norm": 0.04064074158668518,
+      "learning_rate": 0.01,
+      "loss": 1.963,
+      "step": 116595
+    },
+    {
+      "epoch": 11.983350462487152,
+      "grad_norm": 0.05459297075867653,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 116598
+    },
+    {
+      "epoch": 11.98365878725591,
+      "grad_norm": 0.050961967557668686,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 116601
+    },
+    {
+      "epoch": 11.983967112024667,
+      "grad_norm": 0.047685325145721436,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 116604
+    },
+    {
+      "epoch": 11.984275436793423,
+      "grad_norm": 0.05707469582557678,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 116607
+    },
+    {
+      "epoch": 11.98458376156218,
+      "grad_norm": 0.08973026275634766,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 116610
+    },
+    {
+      "epoch": 11.984892086330936,
+      "grad_norm": 0.07212896645069122,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 116613
+    },
+    {
+      "epoch": 11.985200411099692,
+      "grad_norm": 0.12053423374891281,
+      "learning_rate": 0.01,
+      "loss": 1.9031,
+      "step": 116616
+    },
+    {
+      "epoch": 11.985508735868448,
+      "grad_norm": 0.07867543399333954,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 116619
+    },
+    {
+      "epoch": 11.985817060637205,
+      "grad_norm": 0.061868395656347275,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 116622
+    },
+    {
+      "epoch": 11.98612538540596,
+      "grad_norm": 0.06762991100549698,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 116625
+    },
+    {
+      "epoch": 11.986433710174717,
+      "grad_norm": 0.04585585370659828,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 116628
+    },
+    {
+      "epoch": 11.986742034943473,
+      "grad_norm": 0.11000943928956985,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 116631
+    },
+    {
+      "epoch": 11.98705035971223,
+      "grad_norm": 0.12769865989685059,
+      "learning_rate": 0.01,
+      "loss": 1.9548,
+      "step": 116634
+    },
+    {
+      "epoch": 11.987358684480986,
+      "grad_norm": 0.0945596843957901,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 116637
+    },
+    {
+      "epoch": 11.987667009249742,
+      "grad_norm": 0.07920830696821213,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 116640
+    },
+    {
+      "epoch": 11.987975334018499,
+      "grad_norm": 0.05533134564757347,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 116643
+    },
+    {
+      "epoch": 11.988283658787257,
+      "grad_norm": 0.06737732887268066,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 116646
+    },
+    {
+      "epoch": 11.988591983556013,
+      "grad_norm": 0.06965016573667526,
+      "learning_rate": 0.01,
+      "loss": 1.9626,
+      "step": 116649
+    },
+    {
+      "epoch": 11.98890030832477,
+      "grad_norm": 0.07038581371307373,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 116652
+    },
+    {
+      "epoch": 11.989208633093526,
+      "grad_norm": 0.0651620402932167,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 116655
+    },
+    {
+      "epoch": 11.989516957862282,
+      "grad_norm": 0.03648030012845993,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 116658
+    },
+    {
+      "epoch": 11.989825282631038,
+      "grad_norm": 0.035522736608982086,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 116661
+    },
+    {
+      "epoch": 11.990133607399795,
+      "grad_norm": 0.03931429609656334,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 116664
+    },
+    {
+      "epoch": 11.990441932168551,
+      "grad_norm": 0.11045116931200027,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 116667
+    },
+    {
+      "epoch": 11.990750256937307,
+      "grad_norm": 0.11161495000123978,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 116670
+    },
+    {
+      "epoch": 11.991058581706064,
+      "grad_norm": 0.12464109063148499,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 116673
+    },
+    {
+      "epoch": 11.99136690647482,
+      "grad_norm": 0.05374126508831978,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 116676
+    },
+    {
+      "epoch": 11.991675231243576,
+      "grad_norm": 0.03395112603902817,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 116679
+    },
+    {
+      "epoch": 11.991983556012332,
+      "grad_norm": 0.045161593705415726,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 116682
+    },
+    {
+      "epoch": 11.992291880781089,
+      "grad_norm": 0.047702956944704056,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 116685
+    },
+    {
+      "epoch": 11.992600205549845,
+      "grad_norm": 0.04149683564901352,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 116688
+    },
+    {
+      "epoch": 11.992908530318601,
+      "grad_norm": 0.0429212860763073,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 116691
+    },
+    {
+      "epoch": 11.99321685508736,
+      "grad_norm": 0.05036528408527374,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 116694
+    },
+    {
+      "epoch": 11.993525179856116,
+      "grad_norm": 0.07086365669965744,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 116697
+    },
+    {
+      "epoch": 11.993833504624872,
+      "grad_norm": 0.11894246190786362,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 116700
+    },
+    {
+      "epoch": 11.994141829393628,
+      "grad_norm": 0.044312190264463425,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 116703
+    },
+    {
+      "epoch": 11.994450154162385,
+      "grad_norm": 0.04942063242197037,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 116706
+    },
+    {
+      "epoch": 11.994758478931141,
+      "grad_norm": 0.04379546642303467,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 116709
+    },
+    {
+      "epoch": 11.995066803699897,
+      "grad_norm": 0.057303521782159805,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 116712
+    },
+    {
+      "epoch": 11.995375128468654,
+      "grad_norm": 0.12316899001598358,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 116715
+    },
+    {
+      "epoch": 11.99568345323741,
+      "grad_norm": 0.0401909202337265,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 116718
+    },
+    {
+      "epoch": 11.995991778006166,
+      "grad_norm": 0.06272529065608978,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 116721
+    },
+    {
+      "epoch": 11.996300102774923,
+      "grad_norm": 0.0627008005976677,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 116724
+    },
+    {
+      "epoch": 11.996608427543679,
+      "grad_norm": 0.041782572865486145,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 116727
+    },
+    {
+      "epoch": 11.996916752312435,
+      "grad_norm": 0.04066125303506851,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 116730
+    },
+    {
+      "epoch": 11.997225077081191,
+      "grad_norm": 0.06177583336830139,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 116733
+    },
+    {
+      "epoch": 11.99753340184995,
+      "grad_norm": 0.13602577149868011,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 116736
+    },
+    {
+      "epoch": 11.997841726618706,
+      "grad_norm": 0.043734993785619736,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 116739
+    },
+    {
+      "epoch": 11.998150051387462,
+      "grad_norm": 0.04399963840842247,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 116742
+    },
+    {
+      "epoch": 11.998458376156218,
+      "grad_norm": 0.05504929646849632,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 116745
+    },
+    {
+      "epoch": 11.998766700924975,
+      "grad_norm": 0.1177234798669815,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 116748
+    },
+    {
+      "epoch": 11.999075025693731,
+      "grad_norm": 0.04470578953623772,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 116751
+    },
+    {
+      "epoch": 11.999383350462487,
+      "grad_norm": 0.05425754934549332,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 116754
+    },
+    {
+      "epoch": 11.999691675231244,
+      "grad_norm": 0.046526480466127396,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 116757
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 0.043274134397506714,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 116760
+    },
+    {
+      "epoch": 11.997842170160297,
+      "grad_norm": 0.19715403020381927,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 116763
+    },
+    {
+      "epoch": 11.998150431565968,
+      "grad_norm": 0.07191906869411469,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 116766
+    },
+    {
+      "epoch": 11.99845869297164,
+      "grad_norm": 0.07266104221343994,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 116769
+    },
+    {
+      "epoch": 11.998766954377311,
+      "grad_norm": 0.04697905853390694,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 116772
+    },
+    {
+      "epoch": 11.999075215782984,
+      "grad_norm": 0.05928608775138855,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 116775
+    },
+    {
+      "epoch": 11.999383477188656,
+      "grad_norm": 0.040938302874565125,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 116778
+    },
+    {
+      "epoch": 11.999691738594327,
+      "grad_norm": 0.04602961242198944,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 116781
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 0.032084643840789795,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 116784
+    },
+    {
+      "epoch": 12.000308261405673,
+      "grad_norm": 0.06682760268449783,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 116787
+    },
+    {
+      "epoch": 12.000616522811344,
+      "grad_norm": 0.06462869793176651,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 116790
+    },
+    {
+      "epoch": 12.000924784217016,
+      "grad_norm": 0.07240908592939377,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 116793
+    },
+    {
+      "epoch": 12.001233045622689,
+      "grad_norm": 0.04780181497335434,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 116796
+    },
+    {
+      "epoch": 12.00154130702836,
+      "grad_norm": 0.040893614292144775,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 116799
+    },
+    {
+      "epoch": 12.001849568434032,
+      "grad_norm": 0.05878365784883499,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 116802
+    },
+    {
+      "epoch": 12.002157829839705,
+      "grad_norm": 0.059600379317998886,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 116805
+    },
+    {
+      "epoch": 12.002466091245376,
+      "grad_norm": 0.053813789039850235,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 116808
+    },
+    {
+      "epoch": 12.002774352651048,
+      "grad_norm": 0.05230226740241051,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 116811
+    },
+    {
+      "epoch": 12.00308261405672,
+      "grad_norm": 0.047241292893886566,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 116814
+    },
+    {
+      "epoch": 12.003390875462392,
+      "grad_norm": 0.11938584595918655,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 116817
+    },
+    {
+      "epoch": 12.003699136868065,
+      "grad_norm": 0.10174132138490677,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 116820
+    },
+    {
+      "epoch": 12.004007398273735,
+      "grad_norm": 0.061686307191848755,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 116823
+    },
+    {
+      "epoch": 12.004315659679408,
+      "grad_norm": 0.04009551927447319,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 116826
+    },
+    {
+      "epoch": 12.00462392108508,
+      "grad_norm": 0.04727621003985405,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 116829
+    },
+    {
+      "epoch": 12.004932182490752,
+      "grad_norm": 0.06283930689096451,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 116832
+    },
+    {
+      "epoch": 12.005240443896424,
+      "grad_norm": 0.035520389676094055,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 116835
+    },
+    {
+      "epoch": 12.005548705302097,
+      "grad_norm": 0.14637045562267303,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 116838
+    },
+    {
+      "epoch": 12.005856966707768,
+      "grad_norm": 0.04466203972697258,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 116841
+    },
+    {
+      "epoch": 12.00616522811344,
+      "grad_norm": 0.0948067232966423,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 116844
+    },
+    {
+      "epoch": 12.006473489519113,
+      "grad_norm": 0.06225452572107315,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 116847
+    },
+    {
+      "epoch": 12.006781750924784,
+      "grad_norm": 0.041357167065143585,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 116850
+    },
+    {
+      "epoch": 12.007090012330456,
+      "grad_norm": 0.03680774196982384,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 116853
+    },
+    {
+      "epoch": 12.00739827373613,
+      "grad_norm": 0.22551889717578888,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 116856
+    },
+    {
+      "epoch": 12.0077065351418,
+      "grad_norm": 0.1491953581571579,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 116859
+    },
+    {
+      "epoch": 12.008014796547473,
+      "grad_norm": 0.11442308872938156,
+      "learning_rate": 0.01,
+      "loss": 1.8926,
+      "step": 116862
+    },
+    {
+      "epoch": 12.008323057953143,
+      "grad_norm": 0.055177319794893265,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 116865
+    },
+    {
+      "epoch": 12.008631319358816,
+      "grad_norm": 0.037357132881879807,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 116868
+    },
+    {
+      "epoch": 12.008939580764489,
+      "grad_norm": 0.08124551177024841,
+      "learning_rate": 0.01,
+      "loss": 1.971,
+      "step": 116871
+    },
+    {
+      "epoch": 12.00924784217016,
+      "grad_norm": 0.06150367483496666,
+      "learning_rate": 0.01,
+      "loss": 1.9681,
+      "step": 116874
+    },
+    {
+      "epoch": 12.009556103575832,
+      "grad_norm": 0.04272659122943878,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 116877
+    },
+    {
+      "epoch": 12.009864364981505,
+      "grad_norm": 0.061403874307870865,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 116880
+    },
+    {
+      "epoch": 12.010172626387176,
+      "grad_norm": 0.03552393987774849,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 116883
+    },
+    {
+      "epoch": 12.010480887792848,
+      "grad_norm": 0.03674667328596115,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 116886
+    },
+    {
+      "epoch": 12.010789149198521,
+      "grad_norm": 0.09467052668333054,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 116889
+    },
+    {
+      "epoch": 12.011097410604192,
+      "grad_norm": 0.10289536416530609,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 116892
+    },
+    {
+      "epoch": 12.011405672009865,
+      "grad_norm": 0.08801929652690887,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 116895
+    },
+    {
+      "epoch": 12.011713933415537,
+      "grad_norm": 0.12483206391334534,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 116898
+    },
+    {
+      "epoch": 12.012022194821208,
+      "grad_norm": 0.08305198699235916,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 116901
+    },
+    {
+      "epoch": 12.01233045622688,
+      "grad_norm": 0.043588075786828995,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 116904
+    },
+    {
+      "epoch": 12.012638717632552,
+      "grad_norm": 0.0356164388358593,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 116907
+    },
+    {
+      "epoch": 12.012946979038224,
+      "grad_norm": 0.05521430820226669,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 116910
+    },
+    {
+      "epoch": 12.013255240443897,
+      "grad_norm": 0.08943873643875122,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 116913
+    },
+    {
+      "epoch": 12.013563501849568,
+      "grad_norm": 0.07061263173818588,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 116916
+    },
+    {
+      "epoch": 12.01387176325524,
+      "grad_norm": 0.0761643648147583,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 116919
+    },
+    {
+      "epoch": 12.014180024660913,
+      "grad_norm": 0.07292278110980988,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 116922
+    },
+    {
+      "epoch": 12.014488286066584,
+      "grad_norm": 0.09125639498233795,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 116925
+    },
+    {
+      "epoch": 12.014796547472256,
+      "grad_norm": 0.10665156692266464,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 116928
+    },
+    {
+      "epoch": 12.015104808877929,
+      "grad_norm": 0.05672309547662735,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 116931
+    },
+    {
+      "epoch": 12.0154130702836,
+      "grad_norm": 0.06380181759595871,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 116934
+    },
+    {
+      "epoch": 12.015721331689273,
+      "grad_norm": 0.08765019476413727,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 116937
+    },
+    {
+      "epoch": 12.016029593094945,
+      "grad_norm": 0.10064586997032166,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 116940
+    },
+    {
+      "epoch": 12.016337854500616,
+      "grad_norm": 0.04949174076318741,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 116943
+    },
+    {
+      "epoch": 12.016646115906289,
+      "grad_norm": 0.05919962376356125,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 116946
+    },
+    {
+      "epoch": 12.016954377311961,
+      "grad_norm": 0.060713063925504684,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 116949
+    },
+    {
+      "epoch": 12.017262638717632,
+      "grad_norm": 0.06562382727861404,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 116952
+    },
+    {
+      "epoch": 12.017570900123305,
+      "grad_norm": 0.10939974337816238,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 116955
+    },
+    {
+      "epoch": 12.017879161528976,
+      "grad_norm": 0.04349961131811142,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 116958
+    },
+    {
+      "epoch": 12.018187422934648,
+      "grad_norm": 0.13677796721458435,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 116961
+    },
+    {
+      "epoch": 12.018495684340321,
+      "grad_norm": 0.10809358954429626,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 116964
+    },
+    {
+      "epoch": 12.018803945745992,
+      "grad_norm": 0.056563131511211395,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 116967
+    },
+    {
+      "epoch": 12.019112207151665,
+      "grad_norm": 0.048835474997758865,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 116970
+    },
+    {
+      "epoch": 12.019420468557337,
+      "grad_norm": 0.038491714745759964,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 116973
+    },
+    {
+      "epoch": 12.019728729963008,
+      "grad_norm": 0.04731836915016174,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 116976
+    },
+    {
+      "epoch": 12.02003699136868,
+      "grad_norm": 0.10551325231790543,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 116979
+    },
+    {
+      "epoch": 12.020345252774353,
+      "grad_norm": 0.05078031122684479,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 116982
+    },
+    {
+      "epoch": 12.020653514180024,
+      "grad_norm": 0.06656154245138168,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 116985
+    },
+    {
+      "epoch": 12.020961775585697,
+      "grad_norm": 0.1516573429107666,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 116988
+    },
+    {
+      "epoch": 12.02127003699137,
+      "grad_norm": 0.07005427777767181,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 116991
+    },
+    {
+      "epoch": 12.02157829839704,
+      "grad_norm": 0.04545412212610245,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 116994
+    },
+    {
+      "epoch": 12.021886559802713,
+      "grad_norm": 0.042291078716516495,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 116997
+    },
+    {
+      "epoch": 12.022194821208386,
+      "grad_norm": 0.0934653952717781,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 117000
+    },
+    {
+      "epoch": 12.022503082614056,
+      "grad_norm": 0.06595387309789658,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 117003
+    },
+    {
+      "epoch": 12.022811344019729,
+      "grad_norm": 0.11432889848947525,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 117006
+    },
+    {
+      "epoch": 12.0231196054254,
+      "grad_norm": 0.08200161904096603,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 117009
+    },
+    {
+      "epoch": 12.023427866831073,
+      "grad_norm": 0.13999858498573303,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 117012
+    },
+    {
+      "epoch": 12.023736128236745,
+      "grad_norm": 0.08538030833005905,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 117015
+    },
+    {
+      "epoch": 12.024044389642416,
+      "grad_norm": 0.04241943359375,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 117018
+    },
+    {
+      "epoch": 12.024352651048089,
+      "grad_norm": 0.06946206837892532,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 117021
+    },
+    {
+      "epoch": 12.024660912453761,
+      "grad_norm": 0.062005434185266495,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 117024
+    },
+    {
+      "epoch": 12.024969173859432,
+      "grad_norm": 0.0390668585896492,
+      "learning_rate": 0.01,
+      "loss": 1.9713,
+      "step": 117027
+    },
+    {
+      "epoch": 12.025277435265105,
+      "grad_norm": 0.038422442972660065,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 117030
+    },
+    {
+      "epoch": 12.025585696670777,
+      "grad_norm": 0.06409028172492981,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 117033
+    },
+    {
+      "epoch": 12.025893958076448,
+      "grad_norm": 0.036864008754491806,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 117036
+    },
+    {
+      "epoch": 12.026202219482121,
+      "grad_norm": 0.11394591629505157,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 117039
+    },
+    {
+      "epoch": 12.026510480887794,
+      "grad_norm": 0.05371858552098274,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 117042
+    },
+    {
+      "epoch": 12.026818742293464,
+      "grad_norm": 0.11123819649219513,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 117045
+    },
+    {
+      "epoch": 12.027127003699137,
+      "grad_norm": 0.04701869934797287,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 117048
+    },
+    {
+      "epoch": 12.02743526510481,
+      "grad_norm": 0.04842498525977135,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 117051
+    },
+    {
+      "epoch": 12.02774352651048,
+      "grad_norm": 0.03712807595729828,
+      "learning_rate": 0.01,
+      "loss": 1.969,
+      "step": 117054
+    },
+    {
+      "epoch": 12.028051787916153,
+      "grad_norm": 0.07579638808965683,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 117057
+    },
+    {
+      "epoch": 12.028360049321824,
+      "grad_norm": 0.10008134692907333,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 117060
+    },
+    {
+      "epoch": 12.028668310727497,
+      "grad_norm": 0.05371668562293053,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 117063
+    },
+    {
+      "epoch": 12.02897657213317,
+      "grad_norm": 0.07945726066827774,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 117066
+    },
+    {
+      "epoch": 12.02928483353884,
+      "grad_norm": 0.035276856273412704,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 117069
+    },
+    {
+      "epoch": 12.029593094944513,
+      "grad_norm": 0.09449401497840881,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 117072
+    },
+    {
+      "epoch": 12.029901356350186,
+      "grad_norm": 0.07702691107988358,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 117075
+    },
+    {
+      "epoch": 12.030209617755856,
+      "grad_norm": 0.1043589785695076,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 117078
+    },
+    {
+      "epoch": 12.030517879161529,
+      "grad_norm": 0.139021098613739,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 117081
+    },
+    {
+      "epoch": 12.030826140567202,
+      "grad_norm": 0.12700006365776062,
+      "learning_rate": 0.01,
+      "loss": 1.9068,
+      "step": 117084
+    },
+    {
+      "epoch": 12.031134401972873,
+      "grad_norm": 0.08406472206115723,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 117087
+    },
+    {
+      "epoch": 12.031442663378545,
+      "grad_norm": 0.058496635407209396,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 117090
+    },
+    {
+      "epoch": 12.031750924784218,
+      "grad_norm": 0.05651369318366051,
+      "learning_rate": 0.01,
+      "loss": 1.9674,
+      "step": 117093
+    },
+    {
+      "epoch": 12.032059186189889,
+      "grad_norm": 0.041309770196676254,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 117096
+    },
+    {
+      "epoch": 12.032367447595561,
+      "grad_norm": 0.08295998722314835,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 117099
+    },
+    {
+      "epoch": 12.032675709001232,
+      "grad_norm": 0.07957513630390167,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 117102
+    },
+    {
+      "epoch": 12.032983970406905,
+      "grad_norm": 0.05058687925338745,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 117105
+    },
+    {
+      "epoch": 12.033292231812577,
+      "grad_norm": 0.034771159291267395,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 117108
+    },
+    {
+      "epoch": 12.033600493218248,
+      "grad_norm": 0.11958864331245422,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 117111
+    },
+    {
+      "epoch": 12.033908754623921,
+      "grad_norm": 0.04502825811505318,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 117114
+    },
+    {
+      "epoch": 12.034217016029594,
+      "grad_norm": 0.1089678481221199,
+      "learning_rate": 0.01,
+      "loss": 1.9693,
+      "step": 117117
+    },
+    {
+      "epoch": 12.034525277435264,
+      "grad_norm": 0.09961727261543274,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 117120
+    },
+    {
+      "epoch": 12.034833538840937,
+      "grad_norm": 0.05590556561946869,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 117123
+    },
+    {
+      "epoch": 12.03514180024661,
+      "grad_norm": 0.03497562184929848,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 117126
+    },
+    {
+      "epoch": 12.03545006165228,
+      "grad_norm": 0.05207972228527069,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 117129
+    },
+    {
+      "epoch": 12.035758323057953,
+      "grad_norm": 0.053018007427453995,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 117132
+    },
+    {
+      "epoch": 12.036066584463626,
+      "grad_norm": 0.08121684193611145,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 117135
+    },
+    {
+      "epoch": 12.036374845869297,
+      "grad_norm": 0.0976773351430893,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 117138
+    },
+    {
+      "epoch": 12.03668310727497,
+      "grad_norm": 0.06670916080474854,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 117141
+    },
+    {
+      "epoch": 12.036991368680642,
+      "grad_norm": 0.11813090741634369,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 117144
+    },
+    {
+      "epoch": 12.037299630086313,
+      "grad_norm": 0.08185863494873047,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 117147
+    },
+    {
+      "epoch": 12.037607891491986,
+      "grad_norm": 0.1361522376537323,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 117150
+    },
+    {
+      "epoch": 12.037916152897656,
+      "grad_norm": 0.09057639539241791,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 117153
+    },
+    {
+      "epoch": 12.038224414303329,
+      "grad_norm": 0.11496086418628693,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 117156
+    },
+    {
+      "epoch": 12.038532675709002,
+      "grad_norm": 0.06829189509153366,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 117159
+    },
+    {
+      "epoch": 12.038840937114673,
+      "grad_norm": 0.05468536168336868,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 117162
+    },
+    {
+      "epoch": 12.039149198520345,
+      "grad_norm": 0.04229513183236122,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 117165
+    },
+    {
+      "epoch": 12.039457459926018,
+      "grad_norm": 0.0614624097943306,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 117168
+    },
+    {
+      "epoch": 12.039765721331689,
+      "grad_norm": 0.06150800362229347,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 117171
+    },
+    {
+      "epoch": 12.040073982737361,
+      "grad_norm": 0.08810736984014511,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 117174
+    },
+    {
+      "epoch": 12.040382244143034,
+      "grad_norm": 0.10887371003627777,
+      "learning_rate": 0.01,
+      "loss": 1.958,
+      "step": 117177
+    },
+    {
+      "epoch": 12.040690505548705,
+      "grad_norm": 0.045244283974170685,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 117180
+    },
+    {
+      "epoch": 12.040998766954377,
+      "grad_norm": 0.040193479508161545,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 117183
+    },
+    {
+      "epoch": 12.04130702836005,
+      "grad_norm": 0.050419118255376816,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 117186
+    },
+    {
+      "epoch": 12.041615289765721,
+      "grad_norm": 0.07721977680921555,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 117189
+    },
+    {
+      "epoch": 12.041923551171394,
+      "grad_norm": 0.04506092891097069,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 117192
+    },
+    {
+      "epoch": 12.042231812577066,
+      "grad_norm": 0.03247934952378273,
+      "learning_rate": 0.01,
+      "loss": 1.8944,
+      "step": 117195
+    },
+    {
+      "epoch": 12.042540073982737,
+      "grad_norm": 0.15020819008350372,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 117198
+    },
+    {
+      "epoch": 12.04284833538841,
+      "grad_norm": 0.04528571665287018,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 117201
+    },
+    {
+      "epoch": 12.04315659679408,
+      "grad_norm": 0.09674827009439468,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 117204
+    },
+    {
+      "epoch": 12.043464858199753,
+      "grad_norm": 0.05445929244160652,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 117207
+    },
+    {
+      "epoch": 12.043773119605426,
+      "grad_norm": 0.04853241890668869,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 117210
+    },
+    {
+      "epoch": 12.044081381011097,
+      "grad_norm": 0.13493533432483673,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 117213
+    },
+    {
+      "epoch": 12.04438964241677,
+      "grad_norm": 0.19780290126800537,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 117216
+    },
+    {
+      "epoch": 12.044697903822442,
+      "grad_norm": 0.1563442051410675,
+      "learning_rate": 0.01,
+      "loss": 1.9658,
+      "step": 117219
+    },
+    {
+      "epoch": 12.045006165228113,
+      "grad_norm": 0.07573375105857849,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 117222
+    },
+    {
+      "epoch": 12.045314426633785,
+      "grad_norm": 0.05705763399600983,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 117225
+    },
+    {
+      "epoch": 12.045622688039458,
+      "grad_norm": 0.03947446122765541,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 117228
+    },
+    {
+      "epoch": 12.045930949445129,
+      "grad_norm": 0.03954556956887245,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 117231
+    },
+    {
+      "epoch": 12.046239210850802,
+      "grad_norm": 0.041128892451524734,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 117234
+    },
+    {
+      "epoch": 12.046547472256474,
+      "grad_norm": 0.0362667441368103,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 117237
+    },
+    {
+      "epoch": 12.046855733662145,
+      "grad_norm": 0.054836705327034,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 117240
+    },
+    {
+      "epoch": 12.047163995067818,
+      "grad_norm": 0.09442821145057678,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 117243
+    },
+    {
+      "epoch": 12.04747225647349,
+      "grad_norm": 0.057980697602033615,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 117246
+    },
+    {
+      "epoch": 12.047780517879161,
+      "grad_norm": 0.04346334934234619,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 117249
+    },
+    {
+      "epoch": 12.048088779284834,
+      "grad_norm": 0.042848844081163406,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 117252
+    },
+    {
+      "epoch": 12.048397040690505,
+      "grad_norm": 0.0719798132777214,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 117255
+    },
+    {
+      "epoch": 12.048705302096177,
+      "grad_norm": 0.04385082796216011,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 117258
+    },
+    {
+      "epoch": 12.04901356350185,
+      "grad_norm": 0.06334295868873596,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 117261
+    },
+    {
+      "epoch": 12.049321824907521,
+      "grad_norm": 0.06021895632147789,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 117264
+    },
+    {
+      "epoch": 12.049630086313194,
+      "grad_norm": 0.07692274451255798,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 117267
+    },
+    {
+      "epoch": 12.049938347718866,
+      "grad_norm": 0.09398198127746582,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 117270
+    },
+    {
+      "epoch": 12.050246609124537,
+      "grad_norm": 0.11060789972543716,
+      "learning_rate": 0.01,
+      "loss": 1.8999,
+      "step": 117273
+    },
+    {
+      "epoch": 12.05055487053021,
+      "grad_norm": 0.055222559720277786,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 117276
+    },
+    {
+      "epoch": 12.050863131935882,
+      "grad_norm": 0.04734884575009346,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 117279
+    },
+    {
+      "epoch": 12.051171393341553,
+      "grad_norm": 0.038412392139434814,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 117282
+    },
+    {
+      "epoch": 12.051479654747226,
+      "grad_norm": 0.0350971519947052,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 117285
+    },
+    {
+      "epoch": 12.051787916152898,
+      "grad_norm": 0.046421583741903305,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 117288
+    },
+    {
+      "epoch": 12.05209617755857,
+      "grad_norm": 0.04922668635845184,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 117291
+    },
+    {
+      "epoch": 12.052404438964242,
+      "grad_norm": 0.05314551666378975,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 117294
+    },
+    {
+      "epoch": 12.052712700369913,
+      "grad_norm": 0.04084093123674393,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 117297
+    },
+    {
+      "epoch": 12.053020961775585,
+      "grad_norm": 0.11658613383769989,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 117300
+    },
+    {
+      "epoch": 12.053329223181258,
+      "grad_norm": 0.07992861419916153,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 117303
+    },
+    {
+      "epoch": 12.053637484586929,
+      "grad_norm": 0.10995413362979889,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 117306
+    },
+    {
+      "epoch": 12.053945745992602,
+      "grad_norm": 0.0791175365447998,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 117309
+    },
+    {
+      "epoch": 12.054254007398274,
+      "grad_norm": 0.08389805257320404,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 117312
+    },
+    {
+      "epoch": 12.054562268803945,
+      "grad_norm": 0.040105294436216354,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 117315
+    },
+    {
+      "epoch": 12.054870530209618,
+      "grad_norm": 0.04215976595878601,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 117318
+    },
+    {
+      "epoch": 12.05517879161529,
+      "grad_norm": 0.04946447163820267,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 117321
+    },
+    {
+      "epoch": 12.055487053020961,
+      "grad_norm": 0.04651471599936485,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 117324
+    },
+    {
+      "epoch": 12.055795314426634,
+      "grad_norm": 0.030955208465456963,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 117327
+    },
+    {
+      "epoch": 12.056103575832307,
+      "grad_norm": 0.04631368815898895,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 117330
+    },
+    {
+      "epoch": 12.056411837237977,
+      "grad_norm": 0.06540189683437347,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 117333
+    },
+    {
+      "epoch": 12.05672009864365,
+      "grad_norm": 0.07952427864074707,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 117336
+    },
+    {
+      "epoch": 12.057028360049323,
+      "grad_norm": 0.06618475168943405,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 117339
+    },
+    {
+      "epoch": 12.057336621454994,
+      "grad_norm": 0.07007245719432831,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 117342
+    },
+    {
+      "epoch": 12.057644882860666,
+      "grad_norm": 0.09179333597421646,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 117345
+    },
+    {
+      "epoch": 12.057953144266337,
+      "grad_norm": 0.04440729320049286,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 117348
+    },
+    {
+      "epoch": 12.05826140567201,
+      "grad_norm": 0.06386969238519669,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 117351
+    },
+    {
+      "epoch": 12.058569667077682,
+      "grad_norm": 0.15889348089694977,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 117354
+    },
+    {
+      "epoch": 12.058877928483353,
+      "grad_norm": 0.08538535237312317,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 117357
+    },
+    {
+      "epoch": 12.059186189889026,
+      "grad_norm": 0.1034465953707695,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 117360
+    },
+    {
+      "epoch": 12.059494451294698,
+      "grad_norm": 0.06334459781646729,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 117363
+    },
+    {
+      "epoch": 12.05980271270037,
+      "grad_norm": 0.045413531363010406,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 117366
+    },
+    {
+      "epoch": 12.060110974106042,
+      "grad_norm": 0.057017140090465546,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 117369
+    },
+    {
+      "epoch": 12.060419235511715,
+      "grad_norm": 0.06852418184280396,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 117372
+    },
+    {
+      "epoch": 12.060727496917385,
+      "grad_norm": 0.12366721779108047,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 117375
+    },
+    {
+      "epoch": 12.061035758323058,
+      "grad_norm": 0.08869688957929611,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 117378
+    },
+    {
+      "epoch": 12.06134401972873,
+      "grad_norm": 0.09724227339029312,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 117381
+    },
+    {
+      "epoch": 12.061652281134402,
+      "grad_norm": 0.07461173087358475,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 117384
+    },
+    {
+      "epoch": 12.061960542540074,
+      "grad_norm": 0.09186264872550964,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 117387
+    },
+    {
+      "epoch": 12.062268803945747,
+      "grad_norm": 0.05991434305906296,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 117390
+    },
+    {
+      "epoch": 12.062577065351418,
+      "grad_norm": 0.08469602465629578,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 117393
+    },
+    {
+      "epoch": 12.06288532675709,
+      "grad_norm": 0.08107118308544159,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 117396
+    },
+    {
+      "epoch": 12.063193588162761,
+      "grad_norm": 0.05645585432648659,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 117399
+    },
+    {
+      "epoch": 12.063501849568434,
+      "grad_norm": 0.048557594418525696,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 117402
+    },
+    {
+      "epoch": 12.063810110974106,
+      "grad_norm": 0.04499780014157295,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 117405
+    },
+    {
+      "epoch": 12.064118372379777,
+      "grad_norm": 0.10070645064115524,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 117408
+    },
+    {
+      "epoch": 12.06442663378545,
+      "grad_norm": 0.07078666239976883,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 117411
+    },
+    {
+      "epoch": 12.064734895191123,
+      "grad_norm": 0.09919533878564835,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 117414
+    },
+    {
+      "epoch": 12.065043156596793,
+      "grad_norm": 0.03838624432682991,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 117417
+    },
+    {
+      "epoch": 12.065351418002466,
+      "grad_norm": 0.06327342987060547,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 117420
+    },
+    {
+      "epoch": 12.065659679408139,
+      "grad_norm": 0.054736699908971786,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 117423
+    },
+    {
+      "epoch": 12.06596794081381,
+      "grad_norm": 0.08838552981615067,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 117426
+    },
+    {
+      "epoch": 12.066276202219482,
+      "grad_norm": 0.0879216119647026,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 117429
+    },
+    {
+      "epoch": 12.066584463625155,
+      "grad_norm": 0.06935036182403564,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 117432
+    },
+    {
+      "epoch": 12.066892725030826,
+      "grad_norm": 0.04287251457571983,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 117435
+    },
+    {
+      "epoch": 12.067200986436498,
+      "grad_norm": 0.08437120169401169,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 117438
+    },
+    {
+      "epoch": 12.067509247842171,
+      "grad_norm": 0.10385788977146149,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 117441
+    },
+    {
+      "epoch": 12.067817509247842,
+      "grad_norm": 0.08448704332113266,
+      "learning_rate": 0.01,
+      "loss": 1.9715,
+      "step": 117444
+    },
+    {
+      "epoch": 12.068125770653515,
+      "grad_norm": 0.09182809293270111,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 117447
+    },
+    {
+      "epoch": 12.068434032059185,
+      "grad_norm": 0.06671719998121262,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 117450
+    },
+    {
+      "epoch": 12.068742293464858,
+      "grad_norm": 0.05060334876179695,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 117453
+    },
+    {
+      "epoch": 12.06905055487053,
+      "grad_norm": 0.04752616211771965,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 117456
+    },
+    {
+      "epoch": 12.069358816276202,
+      "grad_norm": 0.0996180921792984,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 117459
+    },
+    {
+      "epoch": 12.069667077681874,
+      "grad_norm": 0.11132294684648514,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 117462
+    },
+    {
+      "epoch": 12.069975339087547,
+      "grad_norm": 0.06568966805934906,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 117465
+    },
+    {
+      "epoch": 12.070283600493218,
+      "grad_norm": 0.03154284879565239,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 117468
+    },
+    {
+      "epoch": 12.07059186189889,
+      "grad_norm": 0.09147141128778458,
+      "learning_rate": 0.01,
+      "loss": 1.9855,
+      "step": 117471
+    },
+    {
+      "epoch": 12.070900123304563,
+      "grad_norm": 0.06943826377391815,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 117474
+    },
+    {
+      "epoch": 12.071208384710234,
+      "grad_norm": 0.1036934033036232,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 117477
+    },
+    {
+      "epoch": 12.071516646115906,
+      "grad_norm": 0.11400505900382996,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 117480
+    },
+    {
+      "epoch": 12.071824907521579,
+      "grad_norm": 0.055724602192640305,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 117483
+    },
+    {
+      "epoch": 12.07213316892725,
+      "grad_norm": 0.04169715195894241,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 117486
+    },
+    {
+      "epoch": 12.072441430332923,
+      "grad_norm": 0.0361507274210453,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 117489
+    },
+    {
+      "epoch": 12.072749691738593,
+      "grad_norm": 0.05780757591128349,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 117492
+    },
+    {
+      "epoch": 12.073057953144266,
+      "grad_norm": 0.0696556344628334,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 117495
+    },
+    {
+      "epoch": 12.073366214549939,
+      "grad_norm": 0.08824704587459564,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 117498
+    },
+    {
+      "epoch": 12.07367447595561,
+      "grad_norm": 0.11320904642343521,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 117501
+    },
+    {
+      "epoch": 12.073982737361282,
+      "grad_norm": 0.038966935127973557,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 117504
+    },
+    {
+      "epoch": 12.074290998766955,
+      "grad_norm": 0.039885830134153366,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 117507
+    },
+    {
+      "epoch": 12.074599260172626,
+      "grad_norm": 0.03738117218017578,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 117510
+    },
+    {
+      "epoch": 12.074907521578298,
+      "grad_norm": 0.06798634678125381,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 117513
+    },
+    {
+      "epoch": 12.075215782983971,
+      "grad_norm": 0.12045107781887054,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 117516
+    },
+    {
+      "epoch": 12.075524044389642,
+      "grad_norm": 0.05819183215498924,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 117519
+    },
+    {
+      "epoch": 12.075832305795315,
+      "grad_norm": 0.08216798305511475,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 117522
+    },
+    {
+      "epoch": 12.076140567200987,
+      "grad_norm": 0.10765063762664795,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 117525
+    },
+    {
+      "epoch": 12.076448828606658,
+      "grad_norm": 0.10575487464666367,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 117528
+    },
+    {
+      "epoch": 12.07675709001233,
+      "grad_norm": 0.04776548594236374,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 117531
+    },
+    {
+      "epoch": 12.077065351418003,
+      "grad_norm": 0.03476772829890251,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 117534
+    },
+    {
+      "epoch": 12.077373612823674,
+      "grad_norm": 0.06624691188335419,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 117537
+    },
+    {
+      "epoch": 12.077681874229347,
+      "grad_norm": 0.07099944353103638,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 117540
+    },
+    {
+      "epoch": 12.077990135635018,
+      "grad_norm": 0.0620768666267395,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 117543
+    },
+    {
+      "epoch": 12.07829839704069,
+      "grad_norm": 0.04294934868812561,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 117546
+    },
+    {
+      "epoch": 12.078606658446363,
+      "grad_norm": 0.035651784390211105,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 117549
+    },
+    {
+      "epoch": 12.078914919852034,
+      "grad_norm": 0.09945672005414963,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 117552
+    },
+    {
+      "epoch": 12.079223181257706,
+      "grad_norm": 0.07488497346639633,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 117555
+    },
+    {
+      "epoch": 12.079531442663379,
+      "grad_norm": 0.1251220703125,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 117558
+    },
+    {
+      "epoch": 12.07983970406905,
+      "grad_norm": 0.13060495257377625,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 117561
+    },
+    {
+      "epoch": 12.080147965474723,
+      "grad_norm": 0.10690896958112717,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 117564
+    },
+    {
+      "epoch": 12.080456226880395,
+      "grad_norm": 0.10013853758573532,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 117567
+    },
+    {
+      "epoch": 12.080764488286066,
+      "grad_norm": 0.05371091514825821,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 117570
+    },
+    {
+      "epoch": 12.081072749691739,
+      "grad_norm": 0.038389820605516434,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 117573
+    },
+    {
+      "epoch": 12.081381011097411,
+      "grad_norm": 0.06609676778316498,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 117576
+    },
+    {
+      "epoch": 12.081689272503082,
+      "grad_norm": 0.04369485005736351,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 117579
+    },
+    {
+      "epoch": 12.081997533908755,
+      "grad_norm": 0.040801163762807846,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 117582
+    },
+    {
+      "epoch": 12.082305795314427,
+      "grad_norm": 0.058864302933216095,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 117585
+    },
+    {
+      "epoch": 12.082614056720098,
+      "grad_norm": 0.032076817005872726,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 117588
+    },
+    {
+      "epoch": 12.082922318125771,
+      "grad_norm": 0.10731884092092514,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 117591
+    },
+    {
+      "epoch": 12.083230579531442,
+      "grad_norm": 0.04927503317594528,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 117594
+    },
+    {
+      "epoch": 12.083538840937115,
+      "grad_norm": 0.041223879903554916,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 117597
+    },
+    {
+      "epoch": 12.083847102342787,
+      "grad_norm": 0.08759777992963791,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 117600
+    },
+    {
+      "epoch": 12.084155363748458,
+      "grad_norm": 0.09925784170627594,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 117603
+    },
+    {
+      "epoch": 12.08446362515413,
+      "grad_norm": 0.04184839501976967,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 117606
+    },
+    {
+      "epoch": 12.084771886559803,
+      "grad_norm": 0.04358890652656555,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 117609
+    },
+    {
+      "epoch": 12.085080147965474,
+      "grad_norm": 0.054400619119405746,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 117612
+    },
+    {
+      "epoch": 12.085388409371147,
+      "grad_norm": 0.052087198942899704,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 117615
+    },
+    {
+      "epoch": 12.08569667077682,
+      "grad_norm": 0.12438651919364929,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 117618
+    },
+    {
+      "epoch": 12.08600493218249,
+      "grad_norm": 0.07035225629806519,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 117621
+    },
+    {
+      "epoch": 12.086313193588163,
+      "grad_norm": 0.06258494406938553,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 117624
+    },
+    {
+      "epoch": 12.086621454993836,
+      "grad_norm": 0.039971187710762024,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 117627
+    },
+    {
+      "epoch": 12.086929716399506,
+      "grad_norm": 0.05577482283115387,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 117630
+    },
+    {
+      "epoch": 12.087237977805179,
+      "grad_norm": 0.039713747799396515,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 117633
+    },
+    {
+      "epoch": 12.087546239210852,
+      "grad_norm": 0.06807857006788254,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 117636
+    },
+    {
+      "epoch": 12.087854500616523,
+      "grad_norm": 0.09134294092655182,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 117639
+    },
+    {
+      "epoch": 12.088162762022195,
+      "grad_norm": 0.09650267660617828,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 117642
+    },
+    {
+      "epoch": 12.088471023427866,
+      "grad_norm": 0.06829991191625595,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 117645
+    },
+    {
+      "epoch": 12.088779284833539,
+      "grad_norm": 0.08449184894561768,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 117648
+    },
+    {
+      "epoch": 12.089087546239211,
+      "grad_norm": 0.0747109204530716,
+      "learning_rate": 0.01,
+      "loss": 1.9845,
+      "step": 117651
+    },
+    {
+      "epoch": 12.089395807644882,
+      "grad_norm": 0.1121891513466835,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 117654
+    },
+    {
+      "epoch": 12.089704069050555,
+      "grad_norm": 0.058580800890922546,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 117657
+    },
+    {
+      "epoch": 12.090012330456227,
+      "grad_norm": 0.08042570948600769,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 117660
+    },
+    {
+      "epoch": 12.090320591861898,
+      "grad_norm": 0.12063337117433548,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 117663
+    },
+    {
+      "epoch": 12.090628853267571,
+      "grad_norm": 0.04594128206372261,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 117666
+    },
+    {
+      "epoch": 12.090937114673244,
+      "grad_norm": 0.04587708041071892,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 117669
+    },
+    {
+      "epoch": 12.091245376078914,
+      "grad_norm": 0.041933804750442505,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 117672
+    },
+    {
+      "epoch": 12.091553637484587,
+      "grad_norm": 0.10076461732387543,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 117675
+    },
+    {
+      "epoch": 12.09186189889026,
+      "grad_norm": 0.058599699288606644,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 117678
+    },
+    {
+      "epoch": 12.09217016029593,
+      "grad_norm": 0.1215745061635971,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 117681
+    },
+    {
+      "epoch": 12.092478421701603,
+      "grad_norm": 0.07247386127710342,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 117684
+    },
+    {
+      "epoch": 12.092786683107274,
+      "grad_norm": 0.04243781417608261,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 117687
+    },
+    {
+      "epoch": 12.093094944512947,
+      "grad_norm": 0.05693398788571358,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 117690
+    },
+    {
+      "epoch": 12.09340320591862,
+      "grad_norm": 0.04747099429368973,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 117693
+    },
+    {
+      "epoch": 12.09371146732429,
+      "grad_norm": 0.03934578597545624,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 117696
+    },
+    {
+      "epoch": 12.094019728729963,
+      "grad_norm": 0.04611416533589363,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 117699
+    },
+    {
+      "epoch": 12.094327990135636,
+      "grad_norm": 0.05349862575531006,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 117702
+    },
+    {
+      "epoch": 12.094636251541306,
+      "grad_norm": 0.07715880125761032,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 117705
+    },
+    {
+      "epoch": 12.094944512946979,
+      "grad_norm": 0.07728411257266998,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 117708
+    },
+    {
+      "epoch": 12.095252774352652,
+      "grad_norm": 0.08406969159841537,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 117711
+    },
+    {
+      "epoch": 12.095561035758323,
+      "grad_norm": 0.10758381336927414,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 117714
+    },
+    {
+      "epoch": 12.095869297163995,
+      "grad_norm": 0.14954718947410583,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 117717
+    },
+    {
+      "epoch": 12.096177558569668,
+      "grad_norm": 0.07673077285289764,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 117720
+    },
+    {
+      "epoch": 12.096485819975339,
+      "grad_norm": 0.08066879212856293,
+      "learning_rate": 0.01,
+      "loss": 1.9704,
+      "step": 117723
+    },
+    {
+      "epoch": 12.096794081381011,
+      "grad_norm": 0.04644524306058884,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 117726
+    },
+    {
+      "epoch": 12.097102342786684,
+      "grad_norm": 0.061609331518411636,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 117729
+    },
+    {
+      "epoch": 12.097410604192355,
+      "grad_norm": 0.037506479769945145,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 117732
+    },
+    {
+      "epoch": 12.097718865598027,
+      "grad_norm": 0.08549857884645462,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 117735
+    },
+    {
+      "epoch": 12.098027127003698,
+      "grad_norm": 0.04055694490671158,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 117738
+    },
+    {
+      "epoch": 12.098335388409371,
+      "grad_norm": 0.039073679596185684,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 117741
+    },
+    {
+      "epoch": 12.098643649815044,
+      "grad_norm": 0.14250591397285461,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 117744
+    },
+    {
+      "epoch": 12.098951911220714,
+      "grad_norm": 0.0917244404554367,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 117747
+    },
+    {
+      "epoch": 12.099260172626387,
+      "grad_norm": 0.060811594128608704,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 117750
+    },
+    {
+      "epoch": 12.09956843403206,
+      "grad_norm": 0.04312697425484657,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 117753
+    },
+    {
+      "epoch": 12.09987669543773,
+      "grad_norm": 0.032779429107904434,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 117756
+    },
+    {
+      "epoch": 12.100184956843403,
+      "grad_norm": 0.035446010529994965,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 117759
+    },
+    {
+      "epoch": 12.100493218249076,
+      "grad_norm": 0.035251226276159286,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 117762
+    },
+    {
+      "epoch": 12.100801479654747,
+      "grad_norm": 0.0533762201666832,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 117765
+    },
+    {
+      "epoch": 12.10110974106042,
+      "grad_norm": 0.0436546690762043,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 117768
+    },
+    {
+      "epoch": 12.101418002466092,
+      "grad_norm": 0.0903606191277504,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 117771
+    },
+    {
+      "epoch": 12.101726263871763,
+      "grad_norm": 0.059017568826675415,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 117774
+    },
+    {
+      "epoch": 12.102034525277436,
+      "grad_norm": 0.08666199445724487,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 117777
+    },
+    {
+      "epoch": 12.102342786683108,
+      "grad_norm": 0.06040441244840622,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 117780
+    },
+    {
+      "epoch": 12.102651048088779,
+      "grad_norm": 0.04902610555291176,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 117783
+    },
+    {
+      "epoch": 12.102959309494452,
+      "grad_norm": 0.059788379818201065,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 117786
+    },
+    {
+      "epoch": 12.103267570900123,
+      "grad_norm": 0.163142129778862,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 117789
+    },
+    {
+      "epoch": 12.103575832305795,
+      "grad_norm": 0.11989153921604156,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 117792
+    },
+    {
+      "epoch": 12.103884093711468,
+      "grad_norm": 0.054746828973293304,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 117795
+    },
+    {
+      "epoch": 12.104192355117139,
+      "grad_norm": 0.08966901898384094,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 117798
+    },
+    {
+      "epoch": 12.104500616522811,
+      "grad_norm": 0.04606639966368675,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 117801
+    },
+    {
+      "epoch": 12.104808877928484,
+      "grad_norm": 0.04012494906783104,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 117804
+    },
+    {
+      "epoch": 12.105117139334155,
+      "grad_norm": 0.09926046431064606,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 117807
+    },
+    {
+      "epoch": 12.105425400739827,
+      "grad_norm": 0.06846530735492706,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 117810
+    },
+    {
+      "epoch": 12.1057336621455,
+      "grad_norm": 0.06577073037624359,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 117813
+    },
+    {
+      "epoch": 12.106041923551171,
+      "grad_norm": 0.04960223659873009,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 117816
+    },
+    {
+      "epoch": 12.106350184956844,
+      "grad_norm": 0.061768539249897,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 117819
+    },
+    {
+      "epoch": 12.106658446362516,
+      "grad_norm": 0.10435442626476288,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 117822
+    },
+    {
+      "epoch": 12.106966707768187,
+      "grad_norm": 0.043556876480579376,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 117825
+    },
+    {
+      "epoch": 12.10727496917386,
+      "grad_norm": 0.06298083066940308,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 117828
+    },
+    {
+      "epoch": 12.10758323057953,
+      "grad_norm": 0.07426002621650696,
+      "learning_rate": 0.01,
+      "loss": 1.957,
+      "step": 117831
+    },
+    {
+      "epoch": 12.107891491985203,
+      "grad_norm": 0.06376281380653381,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 117834
+    },
+    {
+      "epoch": 12.108199753390876,
+      "grad_norm": 0.12556427717208862,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 117837
+    },
+    {
+      "epoch": 12.108508014796547,
+      "grad_norm": 0.08735141158103943,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 117840
+    },
+    {
+      "epoch": 12.10881627620222,
+      "grad_norm": 0.058545466512441635,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 117843
+    },
+    {
+      "epoch": 12.109124537607892,
+      "grad_norm": 0.07886437326669693,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 117846
+    },
+    {
+      "epoch": 12.109432799013563,
+      "grad_norm": 0.060616493225097656,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 117849
+    },
+    {
+      "epoch": 12.109741060419235,
+      "grad_norm": 0.10220709443092346,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 117852
+    },
+    {
+      "epoch": 12.110049321824908,
+      "grad_norm": 0.0761055052280426,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 117855
+    },
+    {
+      "epoch": 12.110357583230579,
+      "grad_norm": 0.08665987104177475,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 117858
+    },
+    {
+      "epoch": 12.110665844636252,
+      "grad_norm": 0.03652096167206764,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 117861
+    },
+    {
+      "epoch": 12.110974106041924,
+      "grad_norm": 0.04153512045741081,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 117864
+    },
+    {
+      "epoch": 12.111282367447595,
+      "grad_norm": 0.04108879715204239,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 117867
+    },
+    {
+      "epoch": 12.111590628853268,
+      "grad_norm": 0.04862569272518158,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 117870
+    },
+    {
+      "epoch": 12.11189889025894,
+      "grad_norm": 0.05914356932044029,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 117873
+    },
+    {
+      "epoch": 12.112207151664611,
+      "grad_norm": 0.0469847172498703,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 117876
+    },
+    {
+      "epoch": 12.112515413070284,
+      "grad_norm": 0.04443184658885002,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 117879
+    },
+    {
+      "epoch": 12.112823674475955,
+      "grad_norm": 0.030330419540405273,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 117882
+    },
+    {
+      "epoch": 12.113131935881627,
+      "grad_norm": 0.1044810339808464,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 117885
+    },
+    {
+      "epoch": 12.1134401972873,
+      "grad_norm": 0.06196374073624611,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 117888
+    },
+    {
+      "epoch": 12.113748458692971,
+      "grad_norm": 0.05133262276649475,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 117891
+    },
+    {
+      "epoch": 12.114056720098644,
+      "grad_norm": 0.049086328595876694,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 117894
+    },
+    {
+      "epoch": 12.114364981504316,
+      "grad_norm": 0.07143127918243408,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 117897
+    },
+    {
+      "epoch": 12.114673242909987,
+      "grad_norm": 0.13039495050907135,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 117900
+    },
+    {
+      "epoch": 12.11498150431566,
+      "grad_norm": 0.06353791058063507,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 117903
+    },
+    {
+      "epoch": 12.115289765721332,
+      "grad_norm": 0.052343275398015976,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 117906
+    },
+    {
+      "epoch": 12.115598027127003,
+      "grad_norm": 0.10411820560693741,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 117909
+    },
+    {
+      "epoch": 12.115906288532676,
+      "grad_norm": 0.06003390997648239,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 117912
+    },
+    {
+      "epoch": 12.116214549938348,
+      "grad_norm": 0.0739888846874237,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 117915
+    },
+    {
+      "epoch": 12.11652281134402,
+      "grad_norm": 0.05417364090681076,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 117918
+    },
+    {
+      "epoch": 12.116831072749692,
+      "grad_norm": 0.07016511261463165,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 117921
+    },
+    {
+      "epoch": 12.117139334155365,
+      "grad_norm": 0.06602293998003006,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 117924
+    },
+    {
+      "epoch": 12.117447595561035,
+      "grad_norm": 0.03474023938179016,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 117927
+    },
+    {
+      "epoch": 12.117755856966708,
+      "grad_norm": 0.05247775837779045,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 117930
+    },
+    {
+      "epoch": 12.118064118372379,
+      "grad_norm": 0.030670950189232826,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 117933
+    },
+    {
+      "epoch": 12.118372379778052,
+      "grad_norm": 0.05339053273200989,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 117936
+    },
+    {
+      "epoch": 12.118680641183724,
+      "grad_norm": 0.06306292116641998,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 117939
+    },
+    {
+      "epoch": 12.118988902589395,
+      "grad_norm": 0.06941086053848267,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 117942
+    },
+    {
+      "epoch": 12.119297163995068,
+      "grad_norm": 0.04501570016145706,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 117945
+    },
+    {
+      "epoch": 12.11960542540074,
+      "grad_norm": 0.14525742828845978,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 117948
+    },
+    {
+      "epoch": 12.119913686806411,
+      "grad_norm": 0.10927775502204895,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 117951
+    },
+    {
+      "epoch": 12.120221948212084,
+      "grad_norm": 0.06403226405382156,
+      "learning_rate": 0.01,
+      "loss": 1.9649,
+      "step": 117954
+    },
+    {
+      "epoch": 12.120530209617757,
+      "grad_norm": 0.04629103094339371,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 117957
+    },
+    {
+      "epoch": 12.120838471023427,
+      "grad_norm": 0.04517742991447449,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 117960
+    },
+    {
+      "epoch": 12.1211467324291,
+      "grad_norm": 0.045234084129333496,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 117963
+    },
+    {
+      "epoch": 12.121454993834773,
+      "grad_norm": 0.07998780161142349,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 117966
+    },
+    {
+      "epoch": 12.121763255240444,
+      "grad_norm": 0.1172008141875267,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 117969
+    },
+    {
+      "epoch": 12.122071516646116,
+      "grad_norm": 0.05009810999035835,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 117972
+    },
+    {
+      "epoch": 12.122379778051789,
+      "grad_norm": 0.11575902998447418,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 117975
+    },
+    {
+      "epoch": 12.12268803945746,
+      "grad_norm": 0.05342715233564377,
+      "learning_rate": 0.01,
+      "loss": 1.9746,
+      "step": 117978
+    },
+    {
+      "epoch": 12.122996300863132,
+      "grad_norm": 0.0628928691148758,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 117981
+    },
+    {
+      "epoch": 12.123304562268803,
+      "grad_norm": 0.13087360560894012,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 117984
+    },
+    {
+      "epoch": 12.123612823674476,
+      "grad_norm": 0.04477988928556442,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 117987
+    },
+    {
+      "epoch": 12.123921085080148,
+      "grad_norm": 0.11973872035741806,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 117990
+    },
+    {
+      "epoch": 12.12422934648582,
+      "grad_norm": 0.04543408006429672,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 117993
+    },
+    {
+      "epoch": 12.124537607891492,
+      "grad_norm": 0.03628748282790184,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 117996
+    },
+    {
+      "epoch": 12.124845869297165,
+      "grad_norm": 0.09488870948553085,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 117999
+    },
+    {
+      "epoch": 12.125154130702835,
+      "grad_norm": 0.0529656819999218,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 118002
+    },
+    {
+      "epoch": 12.125462392108508,
+      "grad_norm": 0.06371456384658813,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 118005
+    },
+    {
+      "epoch": 12.12577065351418,
+      "grad_norm": 0.03643391653895378,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 118008
+    },
+    {
+      "epoch": 12.126078914919852,
+      "grad_norm": 0.042078692466020584,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 118011
+    },
+    {
+      "epoch": 12.126387176325524,
+      "grad_norm": 0.1325952559709549,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 118014
+    },
+    {
+      "epoch": 12.126695437731197,
+      "grad_norm": 0.05285250023007393,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 118017
+    },
+    {
+      "epoch": 12.127003699136868,
+      "grad_norm": 0.06790726631879807,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 118020
+    },
+    {
+      "epoch": 12.12731196054254,
+      "grad_norm": 0.07077708095312119,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 118023
+    },
+    {
+      "epoch": 12.127620221948211,
+      "grad_norm": 0.05388166382908821,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 118026
+    },
+    {
+      "epoch": 12.127928483353884,
+      "grad_norm": 0.07416366040706635,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 118029
+    },
+    {
+      "epoch": 12.128236744759556,
+      "grad_norm": 0.06110347434878349,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 118032
+    },
+    {
+      "epoch": 12.128545006165227,
+      "grad_norm": 0.06149152293801308,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 118035
+    },
+    {
+      "epoch": 12.1288532675709,
+      "grad_norm": 0.10737799853086472,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 118038
+    },
+    {
+      "epoch": 12.129161528976573,
+      "grad_norm": 0.14195168018341064,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 118041
+    },
+    {
+      "epoch": 12.129469790382243,
+      "grad_norm": 0.09065920114517212,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 118044
+    },
+    {
+      "epoch": 12.129778051787916,
+      "grad_norm": 0.06632434576749802,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 118047
+    },
+    {
+      "epoch": 12.130086313193589,
+      "grad_norm": 0.05222143977880478,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 118050
+    },
+    {
+      "epoch": 12.13039457459926,
+      "grad_norm": 0.050823818892240524,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 118053
+    },
+    {
+      "epoch": 12.130702836004932,
+      "grad_norm": 0.05407403036952019,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 118056
+    },
+    {
+      "epoch": 12.131011097410605,
+      "grad_norm": 0.052906397730112076,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 118059
+    },
+    {
+      "epoch": 12.131319358816276,
+      "grad_norm": 0.08943575620651245,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 118062
+    },
+    {
+      "epoch": 12.131627620221948,
+      "grad_norm": 0.043714381754398346,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 118065
+    },
+    {
+      "epoch": 12.131935881627621,
+      "grad_norm": 0.05486621335148811,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 118068
+    },
+    {
+      "epoch": 12.132244143033292,
+      "grad_norm": 0.05442049726843834,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 118071
+    },
+    {
+      "epoch": 12.132552404438965,
+      "grad_norm": 0.051040273159742355,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 118074
+    },
+    {
+      "epoch": 12.132860665844635,
+      "grad_norm": 0.05163537338376045,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 118077
+    },
+    {
+      "epoch": 12.133168927250308,
+      "grad_norm": 0.12019070982933044,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 118080
+    },
+    {
+      "epoch": 12.13347718865598,
+      "grad_norm": 0.048908084630966187,
+      "learning_rate": 0.01,
+      "loss": 1.9666,
+      "step": 118083
+    },
+    {
+      "epoch": 12.133785450061652,
+      "grad_norm": 0.12288028746843338,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 118086
+    },
+    {
+      "epoch": 12.134093711467324,
+      "grad_norm": 0.12700609862804413,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 118089
+    },
+    {
+      "epoch": 12.134401972872997,
+      "grad_norm": 0.08140483498573303,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 118092
+    },
+    {
+      "epoch": 12.134710234278668,
+      "grad_norm": 0.05661148577928543,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 118095
+    },
+    {
+      "epoch": 12.13501849568434,
+      "grad_norm": 0.04805785417556763,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 118098
+    },
+    {
+      "epoch": 12.135326757090013,
+      "grad_norm": 0.06646308302879333,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 118101
+    },
+    {
+      "epoch": 12.135635018495684,
+      "grad_norm": 0.05786815658211708,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 118104
+    },
+    {
+      "epoch": 12.135943279901356,
+      "grad_norm": 0.04217448830604553,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 118107
+    },
+    {
+      "epoch": 12.136251541307029,
+      "grad_norm": 0.03454460948705673,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 118110
+    },
+    {
+      "epoch": 12.1365598027127,
+      "grad_norm": 0.11412219703197479,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 118113
+    },
+    {
+      "epoch": 12.136868064118373,
+      "grad_norm": 0.08381339907646179,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 118116
+    },
+    {
+      "epoch": 12.137176325524045,
+      "grad_norm": 0.059763435274362564,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 118119
+    },
+    {
+      "epoch": 12.137484586929716,
+      "grad_norm": 0.08242775499820709,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 118122
+    },
+    {
+      "epoch": 12.137792848335389,
+      "grad_norm": 0.04511423036456108,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 118125
+    },
+    {
+      "epoch": 12.13810110974106,
+      "grad_norm": 0.04035963490605354,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 118128
+    },
+    {
+      "epoch": 12.138409371146732,
+      "grad_norm": 0.038142427802085876,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 118131
+    },
+    {
+      "epoch": 12.138717632552405,
+      "grad_norm": 0.10171832144260406,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 118134
+    },
+    {
+      "epoch": 12.139025893958076,
+      "grad_norm": 0.10713452100753784,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 118137
+    },
+    {
+      "epoch": 12.139334155363748,
+      "grad_norm": 0.10755790024995804,
+      "learning_rate": 0.01,
+      "loss": 1.9691,
+      "step": 118140
+    },
+    {
+      "epoch": 12.139642416769421,
+      "grad_norm": 0.04062514007091522,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 118143
+    },
+    {
+      "epoch": 12.139950678175092,
+      "grad_norm": 0.04124769568443298,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 118146
+    },
+    {
+      "epoch": 12.140258939580765,
+      "grad_norm": 0.04127255827188492,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 118149
+    },
+    {
+      "epoch": 12.140567200986437,
+      "grad_norm": 0.03743264824151993,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 118152
+    },
+    {
+      "epoch": 12.140875462392108,
+      "grad_norm": 0.08324801921844482,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 118155
+    },
+    {
+      "epoch": 12.14118372379778,
+      "grad_norm": 0.03775719553232193,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 118158
+    },
+    {
+      "epoch": 12.141491985203453,
+      "grad_norm": 0.12722140550613403,
+      "learning_rate": 0.01,
+      "loss": 1.968,
+      "step": 118161
+    },
+    {
+      "epoch": 12.141800246609124,
+      "grad_norm": 0.06462185084819794,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 118164
+    },
+    {
+      "epoch": 12.142108508014797,
+      "grad_norm": 0.04018726199865341,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 118167
+    },
+    {
+      "epoch": 12.142416769420468,
+      "grad_norm": 0.03868977352976799,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 118170
+    },
+    {
+      "epoch": 12.14272503082614,
+      "grad_norm": 0.06277395039796829,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 118173
+    },
+    {
+      "epoch": 12.143033292231813,
+      "grad_norm": 0.09695149958133698,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 118176
+    },
+    {
+      "epoch": 12.143341553637484,
+      "grad_norm": 0.13503365218639374,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 118179
+    },
+    {
+      "epoch": 12.143649815043156,
+      "grad_norm": 0.14480143785476685,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 118182
+    },
+    {
+      "epoch": 12.143958076448829,
+      "grad_norm": 0.08472104370594025,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 118185
+    },
+    {
+      "epoch": 12.1442663378545,
+      "grad_norm": 0.06492681056261063,
+      "learning_rate": 0.01,
+      "loss": 1.9762,
+      "step": 118188
+    },
+    {
+      "epoch": 12.144574599260173,
+      "grad_norm": 0.046987369656562805,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 118191
+    },
+    {
+      "epoch": 12.144882860665845,
+      "grad_norm": 0.07729876041412354,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 118194
+    },
+    {
+      "epoch": 12.145191122071516,
+      "grad_norm": 0.04762044548988342,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 118197
+    },
+    {
+      "epoch": 12.145499383477189,
+      "grad_norm": 0.044221337884664536,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 118200
+    },
+    {
+      "epoch": 12.145807644882861,
+      "grad_norm": 0.04139994457364082,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 118203
+    },
+    {
+      "epoch": 12.146115906288532,
+      "grad_norm": 0.039472367614507675,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 118206
+    },
+    {
+      "epoch": 12.146424167694205,
+      "grad_norm": 0.08018239587545395,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 118209
+    },
+    {
+      "epoch": 12.146732429099877,
+      "grad_norm": 0.07383769750595093,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 118212
+    },
+    {
+      "epoch": 12.147040690505548,
+      "grad_norm": 0.05636359751224518,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 118215
+    },
+    {
+      "epoch": 12.147348951911221,
+      "grad_norm": 0.04971145838499069,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 118218
+    },
+    {
+      "epoch": 12.147657213316892,
+      "grad_norm": 0.051524318754673004,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 118221
+    },
+    {
+      "epoch": 12.147965474722564,
+      "grad_norm": 0.11375819146633148,
+      "learning_rate": 0.01,
+      "loss": 1.961,
+      "step": 118224
+    },
+    {
+      "epoch": 12.148273736128237,
+      "grad_norm": 0.08179773390293121,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 118227
+    },
+    {
+      "epoch": 12.148581997533908,
+      "grad_norm": 0.10421700030565262,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 118230
+    },
+    {
+      "epoch": 12.14889025893958,
+      "grad_norm": 0.04878521338105202,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 118233
+    },
+    {
+      "epoch": 12.149198520345253,
+      "grad_norm": 0.047480858862400055,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 118236
+    },
+    {
+      "epoch": 12.149506781750924,
+      "grad_norm": 0.04842216521501541,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 118239
+    },
+    {
+      "epoch": 12.149815043156597,
+      "grad_norm": 0.12058722972869873,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 118242
+    },
+    {
+      "epoch": 12.15012330456227,
+      "grad_norm": 0.14690682291984558,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 118245
+    },
+    {
+      "epoch": 12.15043156596794,
+      "grad_norm": 0.04327579587697983,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 118248
+    },
+    {
+      "epoch": 12.150739827373613,
+      "grad_norm": 0.03938378393650055,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 118251
+    },
+    {
+      "epoch": 12.151048088779286,
+      "grad_norm": 0.0566302128136158,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 118254
+    },
+    {
+      "epoch": 12.151356350184956,
+      "grad_norm": 0.06294912099838257,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 118257
+    },
+    {
+      "epoch": 12.151664611590629,
+      "grad_norm": 0.045011408627033234,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 118260
+    },
+    {
+      "epoch": 12.151972872996302,
+      "grad_norm": 0.1352638453245163,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 118263
+    },
+    {
+      "epoch": 12.152281134401973,
+      "grad_norm": 0.052162181586027145,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 118266
+    },
+    {
+      "epoch": 12.152589395807645,
+      "grad_norm": 0.05339041352272034,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 118269
+    },
+    {
+      "epoch": 12.152897657213316,
+      "grad_norm": 0.09388918429613113,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 118272
+    },
+    {
+      "epoch": 12.153205918618989,
+      "grad_norm": 0.06212795898318291,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 118275
+    },
+    {
+      "epoch": 12.153514180024661,
+      "grad_norm": 0.11813616007566452,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 118278
+    },
+    {
+      "epoch": 12.153822441430332,
+      "grad_norm": 0.07211315631866455,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 118281
+    },
+    {
+      "epoch": 12.154130702836005,
+      "grad_norm": 0.07079070061445236,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 118284
+    },
+    {
+      "epoch": 12.154438964241677,
+      "grad_norm": 0.06737685203552246,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 118287
+    },
+    {
+      "epoch": 12.154747225647348,
+      "grad_norm": 0.03841644898056984,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 118290
+    },
+    {
+      "epoch": 12.155055487053021,
+      "grad_norm": 0.09350793063640594,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 118293
+    },
+    {
+      "epoch": 12.155363748458694,
+      "grad_norm": 0.09523818641901016,
+      "learning_rate": 0.01,
+      "loss": 1.9725,
+      "step": 118296
+    },
+    {
+      "epoch": 12.155672009864364,
+      "grad_norm": 0.05099066346883774,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 118299
+    },
+    {
+      "epoch": 12.155980271270037,
+      "grad_norm": 0.05052245408296585,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 118302
+    },
+    {
+      "epoch": 12.15628853267571,
+      "grad_norm": 0.07317576557397842,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 118305
+    },
+    {
+      "epoch": 12.15659679408138,
+      "grad_norm": 0.05431431531906128,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 118308
+    },
+    {
+      "epoch": 12.156905055487053,
+      "grad_norm": 0.038956791162490845,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 118311
+    },
+    {
+      "epoch": 12.157213316892726,
+      "grad_norm": 0.07684636116027832,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 118314
+    },
+    {
+      "epoch": 12.157521578298397,
+      "grad_norm": 0.08905784040689468,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 118317
+    },
+    {
+      "epoch": 12.15782983970407,
+      "grad_norm": 0.05072393640875816,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 118320
+    },
+    {
+      "epoch": 12.15813810110974,
+      "grad_norm": 0.04204133525490761,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 118323
+    },
+    {
+      "epoch": 12.158446362515413,
+      "grad_norm": 0.03856554999947548,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 118326
+    },
+    {
+      "epoch": 12.158754623921086,
+      "grad_norm": 0.06067296117544174,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 118329
+    },
+    {
+      "epoch": 12.159062885326756,
+      "grad_norm": 0.09541289508342743,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 118332
+    },
+    {
+      "epoch": 12.159371146732429,
+      "grad_norm": 0.09871894866228104,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 118335
+    },
+    {
+      "epoch": 12.159679408138102,
+      "grad_norm": 0.06398959457874298,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 118338
+    },
+    {
+      "epoch": 12.159987669543773,
+      "grad_norm": 0.055907636880874634,
+      "learning_rate": 0.01,
+      "loss": 1.8922,
+      "step": 118341
+    },
+    {
+      "epoch": 12.160295930949445,
+      "grad_norm": 0.053581602871418,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 118344
+    },
+    {
+      "epoch": 12.160604192355118,
+      "grad_norm": 0.035246655344963074,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 118347
+    },
+    {
+      "epoch": 12.160912453760789,
+      "grad_norm": 0.05500401556491852,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 118350
+    },
+    {
+      "epoch": 12.161220715166461,
+      "grad_norm": 0.10213566571474075,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 118353
+    },
+    {
+      "epoch": 12.161528976572134,
+      "grad_norm": 0.056656528264284134,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 118356
+    },
+    {
+      "epoch": 12.161837237977805,
+      "grad_norm": 0.05165443941950798,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 118359
+    },
+    {
+      "epoch": 12.162145499383477,
+      "grad_norm": 0.09598568081855774,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 118362
+    },
+    {
+      "epoch": 12.162453760789148,
+      "grad_norm": 0.05595939978957176,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 118365
+    },
+    {
+      "epoch": 12.162762022194821,
+      "grad_norm": 0.09582502394914627,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 118368
+    },
+    {
+      "epoch": 12.163070283600494,
+      "grad_norm": 0.048460979014635086,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 118371
+    },
+    {
+      "epoch": 12.163378545006164,
+      "grad_norm": 0.04140559211373329,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 118374
+    },
+    {
+      "epoch": 12.163686806411837,
+      "grad_norm": 0.11664795130491257,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 118377
+    },
+    {
+      "epoch": 12.16399506781751,
+      "grad_norm": 0.05546128749847412,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 118380
+    },
+    {
+      "epoch": 12.16430332922318,
+      "grad_norm": 0.0834437757730484,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 118383
+    },
+    {
+      "epoch": 12.164611590628853,
+      "grad_norm": 0.10547520220279694,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 118386
+    },
+    {
+      "epoch": 12.164919852034526,
+      "grad_norm": 0.0766419768333435,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 118389
+    },
+    {
+      "epoch": 12.165228113440197,
+      "grad_norm": 0.0726955235004425,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 118392
+    },
+    {
+      "epoch": 12.16553637484587,
+      "grad_norm": 0.08781227469444275,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 118395
+    },
+    {
+      "epoch": 12.165844636251542,
+      "grad_norm": 0.1715017706155777,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 118398
+    },
+    {
+      "epoch": 12.166152897657213,
+      "grad_norm": 0.06803517788648605,
+      "learning_rate": 0.01,
+      "loss": 1.9653,
+      "step": 118401
+    },
+    {
+      "epoch": 12.166461159062885,
+      "grad_norm": 0.06172870472073555,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 118404
+    },
+    {
+      "epoch": 12.166769420468558,
+      "grad_norm": 0.0564359687268734,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 118407
+    },
+    {
+      "epoch": 12.167077681874229,
+      "grad_norm": 0.04325263574719429,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 118410
+    },
+    {
+      "epoch": 12.167385943279902,
+      "grad_norm": 0.04649733752012253,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 118413
+    },
+    {
+      "epoch": 12.167694204685573,
+      "grad_norm": 0.03295613080263138,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 118416
+    },
+    {
+      "epoch": 12.168002466091245,
+      "grad_norm": 0.06244640424847603,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 118419
+    },
+    {
+      "epoch": 12.168310727496918,
+      "grad_norm": 0.06391515582799911,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 118422
+    },
+    {
+      "epoch": 12.168618988902589,
+      "grad_norm": 0.0544745959341526,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 118425
+    },
+    {
+      "epoch": 12.168927250308261,
+      "grad_norm": 0.04499959573149681,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 118428
+    },
+    {
+      "epoch": 12.169235511713934,
+      "grad_norm": 0.040443163365125656,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 118431
+    },
+    {
+      "epoch": 12.169543773119605,
+      "grad_norm": 0.039167050272226334,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 118434
+    },
+    {
+      "epoch": 12.169852034525277,
+      "grad_norm": 0.07148496061563492,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 118437
+    },
+    {
+      "epoch": 12.17016029593095,
+      "grad_norm": 0.04941992461681366,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 118440
+    },
+    {
+      "epoch": 12.170468557336621,
+      "grad_norm": 0.057443637400865555,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 118443
+    },
+    {
+      "epoch": 12.170776818742294,
+      "grad_norm": 0.07597741484642029,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 118446
+    },
+    {
+      "epoch": 12.171085080147966,
+      "grad_norm": 0.13708484172821045,
+      "learning_rate": 0.01,
+      "loss": 1.9692,
+      "step": 118449
+    },
+    {
+      "epoch": 12.171393341553637,
+      "grad_norm": 0.04398206248879433,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 118452
+    },
+    {
+      "epoch": 12.17170160295931,
+      "grad_norm": 0.10531770437955856,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 118455
+    },
+    {
+      "epoch": 12.172009864364982,
+      "grad_norm": 0.05901032313704491,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 118458
+    },
+    {
+      "epoch": 12.172318125770653,
+      "grad_norm": 0.050240326672792435,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 118461
+    },
+    {
+      "epoch": 12.172626387176326,
+      "grad_norm": 0.04151801019906998,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 118464
+    },
+    {
+      "epoch": 12.172934648581997,
+      "grad_norm": 0.10429253429174423,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 118467
+    },
+    {
+      "epoch": 12.17324290998767,
+      "grad_norm": 0.0488281287252903,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 118470
+    },
+    {
+      "epoch": 12.173551171393342,
+      "grad_norm": 0.08701854944229126,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 118473
+    },
+    {
+      "epoch": 12.173859432799013,
+      "grad_norm": 0.07329504191875458,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 118476
+    },
+    {
+      "epoch": 12.174167694204685,
+      "grad_norm": 0.11235913634300232,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 118479
+    },
+    {
+      "epoch": 12.174475955610358,
+      "grad_norm": 0.09730274975299835,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 118482
+    },
+    {
+      "epoch": 12.174784217016029,
+      "grad_norm": 0.06408539414405823,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 118485
+    },
+    {
+      "epoch": 12.175092478421702,
+      "grad_norm": 0.08930247277021408,
+      "learning_rate": 0.01,
+      "loss": 1.9591,
+      "step": 118488
+    },
+    {
+      "epoch": 12.175400739827374,
+      "grad_norm": 0.07743005454540253,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 118491
+    },
+    {
+      "epoch": 12.175709001233045,
+      "grad_norm": 0.033463992178440094,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 118494
+    },
+    {
+      "epoch": 12.176017262638718,
+      "grad_norm": 0.045047443360090256,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 118497
+    },
+    {
+      "epoch": 12.17632552404439,
+      "grad_norm": 0.08373583108186722,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 118500
+    },
+    {
+      "epoch": 12.176633785450061,
+      "grad_norm": 0.1224374994635582,
+      "learning_rate": 0.01,
+      "loss": 1.9779,
+      "step": 118503
+    },
+    {
+      "epoch": 12.176942046855734,
+      "grad_norm": 0.05045723170042038,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 118506
+    },
+    {
+      "epoch": 12.177250308261407,
+      "grad_norm": 0.04794750362634659,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 118509
+    },
+    {
+      "epoch": 12.177558569667077,
+      "grad_norm": 0.08531951904296875,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 118512
+    },
+    {
+      "epoch": 12.17786683107275,
+      "grad_norm": 0.062374114990234375,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 118515
+    },
+    {
+      "epoch": 12.178175092478421,
+      "grad_norm": 0.06713505834341049,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 118518
+    },
+    {
+      "epoch": 12.178483353884094,
+      "grad_norm": 0.11794169247150421,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 118521
+    },
+    {
+      "epoch": 12.178791615289766,
+      "grad_norm": 0.05226500704884529,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 118524
+    },
+    {
+      "epoch": 12.179099876695437,
+      "grad_norm": 0.0920417532324791,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 118527
+    },
+    {
+      "epoch": 12.17940813810111,
+      "grad_norm": 0.05339904874563217,
+      "learning_rate": 0.01,
+      "loss": 1.9631,
+      "step": 118530
+    },
+    {
+      "epoch": 12.179716399506782,
+      "grad_norm": 0.08635959774255753,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 118533
+    },
+    {
+      "epoch": 12.180024660912453,
+      "grad_norm": 0.052172720432281494,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 118536
+    },
+    {
+      "epoch": 12.180332922318126,
+      "grad_norm": 0.04417841508984566,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 118539
+    },
+    {
+      "epoch": 12.180641183723798,
+      "grad_norm": 0.05313883721828461,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 118542
+    },
+    {
+      "epoch": 12.18094944512947,
+      "grad_norm": 0.10818638652563095,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 118545
+    },
+    {
+      "epoch": 12.181257706535142,
+      "grad_norm": 0.12224768847227097,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 118548
+    },
+    {
+      "epoch": 12.181565967940815,
+      "grad_norm": 0.08173675090074539,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 118551
+    },
+    {
+      "epoch": 12.181874229346485,
+      "grad_norm": 0.1057460829615593,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 118554
+    },
+    {
+      "epoch": 12.182182490752158,
+      "grad_norm": 0.10583849996328354,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 118557
+    },
+    {
+      "epoch": 12.182490752157829,
+      "grad_norm": 0.04914642497897148,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 118560
+    },
+    {
+      "epoch": 12.182799013563502,
+      "grad_norm": 0.04464605823159218,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 118563
+    },
+    {
+      "epoch": 12.183107274969174,
+      "grad_norm": 0.0556897297501564,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 118566
+    },
+    {
+      "epoch": 12.183415536374845,
+      "grad_norm": 0.07571946829557419,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 118569
+    },
+    {
+      "epoch": 12.183723797780518,
+      "grad_norm": 0.12731902301311493,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 118572
+    },
+    {
+      "epoch": 12.18403205918619,
+      "grad_norm": 0.03924984857439995,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 118575
+    },
+    {
+      "epoch": 12.184340320591861,
+      "grad_norm": 0.07670131325721741,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 118578
+    },
+    {
+      "epoch": 12.184648581997534,
+      "grad_norm": 0.06920908391475677,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 118581
+    },
+    {
+      "epoch": 12.184956843403207,
+      "grad_norm": 0.08404618501663208,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 118584
+    },
+    {
+      "epoch": 12.185265104808877,
+      "grad_norm": 0.0868164449930191,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 118587
+    },
+    {
+      "epoch": 12.18557336621455,
+      "grad_norm": 0.080803282558918,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 118590
+    },
+    {
+      "epoch": 12.185881627620223,
+      "grad_norm": 0.09213325381278992,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 118593
+    },
+    {
+      "epoch": 12.186189889025894,
+      "grad_norm": 0.09278547763824463,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 118596
+    },
+    {
+      "epoch": 12.186498150431566,
+      "grad_norm": 0.05901975557208061,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 118599
+    },
+    {
+      "epoch": 12.186806411837239,
+      "grad_norm": 0.11545471101999283,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 118602
+    },
+    {
+      "epoch": 12.18711467324291,
+      "grad_norm": 0.1676173359155655,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 118605
+    },
+    {
+      "epoch": 12.187422934648582,
+      "grad_norm": 0.09550116956233978,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 118608
+    },
+    {
+      "epoch": 12.187731196054253,
+      "grad_norm": 0.047323115170001984,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 118611
+    },
+    {
+      "epoch": 12.188039457459926,
+      "grad_norm": 0.047769296914339066,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 118614
+    },
+    {
+      "epoch": 12.188347718865598,
+      "grad_norm": 0.06573311239480972,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 118617
+    },
+    {
+      "epoch": 12.18865598027127,
+      "grad_norm": 0.05319448560476303,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 118620
+    },
+    {
+      "epoch": 12.188964241676942,
+      "grad_norm": 0.046964943408966064,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 118623
+    },
+    {
+      "epoch": 12.189272503082615,
+      "grad_norm": 0.053592875599861145,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 118626
+    },
+    {
+      "epoch": 12.189580764488285,
+      "grad_norm": 0.1196940466761589,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 118629
+    },
+    {
+      "epoch": 12.189889025893958,
+      "grad_norm": 0.044338516891002655,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 118632
+    },
+    {
+      "epoch": 12.19019728729963,
+      "grad_norm": 0.10683481395244598,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 118635
+    },
+    {
+      "epoch": 12.190505548705302,
+      "grad_norm": 0.03649195283651352,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 118638
+    },
+    {
+      "epoch": 12.190813810110974,
+      "grad_norm": 0.10152381658554077,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 118641
+    },
+    {
+      "epoch": 12.191122071516647,
+      "grad_norm": 0.06696424633264542,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 118644
+    },
+    {
+      "epoch": 12.191430332922318,
+      "grad_norm": 0.126417338848114,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 118647
+    },
+    {
+      "epoch": 12.19173859432799,
+      "grad_norm": 0.0509948693215847,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 118650
+    },
+    {
+      "epoch": 12.192046855733663,
+      "grad_norm": 0.06043535843491554,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 118653
+    },
+    {
+      "epoch": 12.192355117139334,
+      "grad_norm": 0.05188688635826111,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 118656
+    },
+    {
+      "epoch": 12.192663378545006,
+      "grad_norm": 0.07013826817274094,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 118659
+    },
+    {
+      "epoch": 12.192971639950677,
+      "grad_norm": 0.08205080032348633,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 118662
+    },
+    {
+      "epoch": 12.19327990135635,
+      "grad_norm": 0.11694313585758209,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 118665
+    },
+    {
+      "epoch": 12.193588162762023,
+      "grad_norm": 0.06345352530479431,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 118668
+    },
+    {
+      "epoch": 12.193896424167693,
+      "grad_norm": 0.11872132122516632,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 118671
+    },
+    {
+      "epoch": 12.194204685573366,
+      "grad_norm": 0.10739465802907944,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 118674
+    },
+    {
+      "epoch": 12.194512946979039,
+      "grad_norm": 0.06278955191373825,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 118677
+    },
+    {
+      "epoch": 12.19482120838471,
+      "grad_norm": 0.07011450082063675,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 118680
+    },
+    {
+      "epoch": 12.195129469790382,
+      "grad_norm": 0.1012188121676445,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 118683
+    },
+    {
+      "epoch": 12.195437731196055,
+      "grad_norm": 0.07303464412689209,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 118686
+    },
+    {
+      "epoch": 12.195745992601726,
+      "grad_norm": 0.059921663254499435,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 118689
+    },
+    {
+      "epoch": 12.196054254007398,
+      "grad_norm": 0.16199304163455963,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 118692
+    },
+    {
+      "epoch": 12.196362515413071,
+      "grad_norm": 0.13744542002677917,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 118695
+    },
+    {
+      "epoch": 12.196670776818742,
+      "grad_norm": 0.07422071695327759,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 118698
+    },
+    {
+      "epoch": 12.196979038224415,
+      "grad_norm": 0.06524713337421417,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 118701
+    },
+    {
+      "epoch": 12.197287299630087,
+      "grad_norm": 0.04250166192650795,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 118704
+    },
+    {
+      "epoch": 12.197595561035758,
+      "grad_norm": 0.04675396904349327,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 118707
+    },
+    {
+      "epoch": 12.19790382244143,
+      "grad_norm": 0.0890340656042099,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 118710
+    },
+    {
+      "epoch": 12.198212083847102,
+      "grad_norm": 0.04863935336470604,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 118713
+    },
+    {
+      "epoch": 12.198520345252774,
+      "grad_norm": 0.04004283621907234,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 118716
+    },
+    {
+      "epoch": 12.198828606658447,
+      "grad_norm": 0.057114001363515854,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 118719
+    },
+    {
+      "epoch": 12.199136868064118,
+      "grad_norm": 0.04745546355843544,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 118722
+    },
+    {
+      "epoch": 12.19944512946979,
+      "grad_norm": 0.053959012031555176,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 118725
+    },
+    {
+      "epoch": 12.199753390875463,
+      "grad_norm": 0.03172319754958153,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 118728
+    },
+    {
+      "epoch": 12.200061652281134,
+      "grad_norm": 0.14972840249538422,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 118731
+    },
+    {
+      "epoch": 12.200369913686806,
+      "grad_norm": 0.09837561845779419,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 118734
+    },
+    {
+      "epoch": 12.200678175092479,
+      "grad_norm": 0.05351947620511055,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 118737
+    },
+    {
+      "epoch": 12.20098643649815,
+      "grad_norm": 0.06797778606414795,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 118740
+    },
+    {
+      "epoch": 12.201294697903823,
+      "grad_norm": 0.04624923691153526,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 118743
+    },
+    {
+      "epoch": 12.201602959309495,
+      "grad_norm": 0.052814073860645294,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 118746
+    },
+    {
+      "epoch": 12.201911220715166,
+      "grad_norm": 0.06401890516281128,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 118749
+    },
+    {
+      "epoch": 12.202219482120839,
+      "grad_norm": 0.07207990437746048,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 118752
+    },
+    {
+      "epoch": 12.20252774352651,
+      "grad_norm": 0.04342376068234444,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 118755
+    },
+    {
+      "epoch": 12.202836004932182,
+      "grad_norm": 0.03639165684580803,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 118758
+    },
+    {
+      "epoch": 12.203144266337855,
+      "grad_norm": 0.04167010635137558,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 118761
+    },
+    {
+      "epoch": 12.203452527743526,
+      "grad_norm": 0.035900626331567764,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 118764
+    },
+    {
+      "epoch": 12.203760789149198,
+      "grad_norm": 0.044128745794296265,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 118767
+    },
+    {
+      "epoch": 12.204069050554871,
+      "grad_norm": 0.14358042180538177,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 118770
+    },
+    {
+      "epoch": 12.204377311960542,
+      "grad_norm": 0.050120823085308075,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 118773
+    },
+    {
+      "epoch": 12.204685573366215,
+      "grad_norm": 0.06396164000034332,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 118776
+    },
+    {
+      "epoch": 12.204993834771887,
+      "grad_norm": 0.09677011519670486,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 118779
+    },
+    {
+      "epoch": 12.205302096177558,
+      "grad_norm": 0.04677804559469223,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 118782
+    },
+    {
+      "epoch": 12.20561035758323,
+      "grad_norm": 0.1319218873977661,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 118785
+    },
+    {
+      "epoch": 12.205918618988903,
+      "grad_norm": 0.14024706184864044,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 118788
+    },
+    {
+      "epoch": 12.206226880394574,
+      "grad_norm": 0.10679846256971359,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 118791
+    },
+    {
+      "epoch": 12.206535141800247,
+      "grad_norm": 0.07363414019346237,
+      "learning_rate": 0.01,
+      "loss": 1.9789,
+      "step": 118794
+    },
+    {
+      "epoch": 12.20684340320592,
+      "grad_norm": 0.0446414016187191,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 118797
+    },
+    {
+      "epoch": 12.20715166461159,
+      "grad_norm": 0.038796547800302505,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 118800
+    },
+    {
+      "epoch": 12.207459926017263,
+      "grad_norm": 0.09463851153850555,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 118803
+    },
+    {
+      "epoch": 12.207768187422934,
+      "grad_norm": 0.045280370861291885,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 118806
+    },
+    {
+      "epoch": 12.208076448828606,
+      "grad_norm": 0.06241309642791748,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 118809
+    },
+    {
+      "epoch": 12.208384710234279,
+      "grad_norm": 0.06056646630167961,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 118812
+    },
+    {
+      "epoch": 12.20869297163995,
+      "grad_norm": 0.043018437922000885,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 118815
+    },
+    {
+      "epoch": 12.209001233045623,
+      "grad_norm": 0.044745851308107376,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 118818
+    },
+    {
+      "epoch": 12.209309494451295,
+      "grad_norm": 0.04173322394490242,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 118821
+    },
+    {
+      "epoch": 12.209617755856966,
+      "grad_norm": 0.03966313600540161,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 118824
+    },
+    {
+      "epoch": 12.209926017262639,
+      "grad_norm": 0.06935320049524307,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 118827
+    },
+    {
+      "epoch": 12.210234278668311,
+      "grad_norm": 0.08756303042173386,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 118830
+    },
+    {
+      "epoch": 12.210542540073982,
+      "grad_norm": 0.14346668124198914,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 118833
+    },
+    {
+      "epoch": 12.210850801479655,
+      "grad_norm": 0.07660556584596634,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 118836
+    },
+    {
+      "epoch": 12.211159062885327,
+      "grad_norm": 0.06557713449001312,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 118839
+    },
+    {
+      "epoch": 12.211467324290998,
+      "grad_norm": 0.05898251757025719,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 118842
+    },
+    {
+      "epoch": 12.211775585696671,
+      "grad_norm": 0.08062247931957245,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 118845
+    },
+    {
+      "epoch": 12.212083847102344,
+      "grad_norm": 0.06156390532851219,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 118848
+    },
+    {
+      "epoch": 12.212392108508014,
+      "grad_norm": 0.11440441757440567,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 118851
+    },
+    {
+      "epoch": 12.212700369913687,
+      "grad_norm": 0.09022848308086395,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 118854
+    },
+    {
+      "epoch": 12.213008631319358,
+      "grad_norm": 0.06540698558092117,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 118857
+    },
+    {
+      "epoch": 12.21331689272503,
+      "grad_norm": 0.13931789994239807,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 118860
+    },
+    {
+      "epoch": 12.213625154130703,
+      "grad_norm": 0.0924806147813797,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 118863
+    },
+    {
+      "epoch": 12.213933415536374,
+      "grad_norm": 0.07151935249567032,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 118866
+    },
+    {
+      "epoch": 12.214241676942047,
+      "grad_norm": 0.05269354209303856,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 118869
+    },
+    {
+      "epoch": 12.21454993834772,
+      "grad_norm": 0.12896299362182617,
+      "learning_rate": 0.01,
+      "loss": 1.8919,
+      "step": 118872
+    },
+    {
+      "epoch": 12.21485819975339,
+      "grad_norm": 0.03468026965856552,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 118875
+    },
+    {
+      "epoch": 12.215166461159063,
+      "grad_norm": 0.0727490782737732,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 118878
+    },
+    {
+      "epoch": 12.215474722564736,
+      "grad_norm": 0.13106340169906616,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 118881
+    },
+    {
+      "epoch": 12.215782983970406,
+      "grad_norm": 0.09597338736057281,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 118884
+    },
+    {
+      "epoch": 12.216091245376079,
+      "grad_norm": 0.06441767513751984,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 118887
+    },
+    {
+      "epoch": 12.216399506781752,
+      "grad_norm": 0.04363396391272545,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 118890
+    },
+    {
+      "epoch": 12.216707768187423,
+      "grad_norm": 0.050265636295080185,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 118893
+    },
+    {
+      "epoch": 12.217016029593095,
+      "grad_norm": 0.03304623067378998,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 118896
+    },
+    {
+      "epoch": 12.217324290998768,
+      "grad_norm": 0.08241475373506546,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 118899
+    },
+    {
+      "epoch": 12.217632552404439,
+      "grad_norm": 0.037594906985759735,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 118902
+    },
+    {
+      "epoch": 12.217940813810111,
+      "grad_norm": 0.04278671368956566,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 118905
+    },
+    {
+      "epoch": 12.218249075215782,
+      "grad_norm": 0.037733662873506546,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 118908
+    },
+    {
+      "epoch": 12.218557336621455,
+      "grad_norm": 0.03884994983673096,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 118911
+    },
+    {
+      "epoch": 12.218865598027127,
+      "grad_norm": 0.056603990495204926,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 118914
+    },
+    {
+      "epoch": 12.219173859432798,
+      "grad_norm": 0.16166208684444427,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 118917
+    },
+    {
+      "epoch": 12.219482120838471,
+      "grad_norm": 0.09817244112491608,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 118920
+    },
+    {
+      "epoch": 12.219790382244144,
+      "grad_norm": 0.06344011425971985,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 118923
+    },
+    {
+      "epoch": 12.220098643649814,
+      "grad_norm": 0.04339836910367012,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 118926
+    },
+    {
+      "epoch": 12.220406905055487,
+      "grad_norm": 0.03249096870422363,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 118929
+    },
+    {
+      "epoch": 12.22071516646116,
+      "grad_norm": 0.03361337259411812,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 118932
+    },
+    {
+      "epoch": 12.22102342786683,
+      "grad_norm": 0.052787475287914276,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 118935
+    },
+    {
+      "epoch": 12.221331689272503,
+      "grad_norm": 0.0435674712061882,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 118938
+    },
+    {
+      "epoch": 12.221639950678176,
+      "grad_norm": 0.045802339911460876,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 118941
+    },
+    {
+      "epoch": 12.221948212083847,
+      "grad_norm": 0.06979796290397644,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 118944
+    },
+    {
+      "epoch": 12.22225647348952,
+      "grad_norm": 0.08848527073860168,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 118947
+    },
+    {
+      "epoch": 12.22256473489519,
+      "grad_norm": 0.055262014269828796,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 118950
+    },
+    {
+      "epoch": 12.222872996300863,
+      "grad_norm": 0.1241009458899498,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 118953
+    },
+    {
+      "epoch": 12.223181257706536,
+      "grad_norm": 0.12001396715641022,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 118956
+    },
+    {
+      "epoch": 12.223489519112206,
+      "grad_norm": 0.04739031195640564,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 118959
+    },
+    {
+      "epoch": 12.223797780517879,
+      "grad_norm": 0.05092400684952736,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 118962
+    },
+    {
+      "epoch": 12.224106041923552,
+      "grad_norm": 0.05298929288983345,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 118965
+    },
+    {
+      "epoch": 12.224414303329223,
+      "grad_norm": 0.06079339236021042,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 118968
+    },
+    {
+      "epoch": 12.224722564734895,
+      "grad_norm": 0.04320069029927254,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 118971
+    },
+    {
+      "epoch": 12.225030826140568,
+      "grad_norm": 0.07928624749183655,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 118974
+    },
+    {
+      "epoch": 12.225339087546239,
+      "grad_norm": 0.13101297616958618,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 118977
+    },
+    {
+      "epoch": 12.225647348951911,
+      "grad_norm": 0.049566250294446945,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 118980
+    },
+    {
+      "epoch": 12.225955610357584,
+      "grad_norm": 0.0518098846077919,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 118983
+    },
+    {
+      "epoch": 12.226263871763255,
+      "grad_norm": 0.035460662096738815,
+      "learning_rate": 0.01,
+      "loss": 1.9724,
+      "step": 118986
+    },
+    {
+      "epoch": 12.226572133168927,
+      "grad_norm": 0.1427997648715973,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 118989
+    },
+    {
+      "epoch": 12.2268803945746,
+      "grad_norm": 0.03751193732023239,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 118992
+    },
+    {
+      "epoch": 12.227188655980271,
+      "grad_norm": 0.04313329607248306,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 118995
+    },
+    {
+      "epoch": 12.227496917385944,
+      "grad_norm": 0.05610892176628113,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 118998
+    },
+    {
+      "epoch": 12.227805178791614,
+      "grad_norm": 0.036053575575351715,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 119001
+    },
+    {
+      "epoch": 12.228113440197287,
+      "grad_norm": 0.06000903621315956,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 119004
+    },
+    {
+      "epoch": 12.22842170160296,
+      "grad_norm": 0.1441926658153534,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 119007
+    },
+    {
+      "epoch": 12.22872996300863,
+      "grad_norm": 0.05757873132824898,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 119010
+    },
+    {
+      "epoch": 12.229038224414303,
+      "grad_norm": 0.039227813482284546,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 119013
+    },
+    {
+      "epoch": 12.229346485819976,
+      "grad_norm": 0.06010231003165245,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 119016
+    },
+    {
+      "epoch": 12.229654747225647,
+      "grad_norm": 0.09603168815374374,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 119019
+    },
+    {
+      "epoch": 12.22996300863132,
+      "grad_norm": 0.04553908482193947,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 119022
+    },
+    {
+      "epoch": 12.230271270036992,
+      "grad_norm": 0.038257114589214325,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 119025
+    },
+    {
+      "epoch": 12.230579531442663,
+      "grad_norm": 0.07678436487913132,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 119028
+    },
+    {
+      "epoch": 12.230887792848335,
+      "grad_norm": 0.11499672383069992,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 119031
+    },
+    {
+      "epoch": 12.231196054254008,
+      "grad_norm": 0.04072137549519539,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 119034
+    },
+    {
+      "epoch": 12.231504315659679,
+      "grad_norm": 0.09349635243415833,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 119037
+    },
+    {
+      "epoch": 12.231812577065352,
+      "grad_norm": 0.08275580406188965,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 119040
+    },
+    {
+      "epoch": 12.232120838471024,
+      "grad_norm": 0.046482153236866,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 119043
+    },
+    {
+      "epoch": 12.232429099876695,
+      "grad_norm": 0.06730115413665771,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 119046
+    },
+    {
+      "epoch": 12.232737361282368,
+      "grad_norm": 0.07506335526704788,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 119049
+    },
+    {
+      "epoch": 12.233045622688039,
+      "grad_norm": 0.07089993357658386,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 119052
+    },
+    {
+      "epoch": 12.233353884093711,
+      "grad_norm": 0.0345221571624279,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 119055
+    },
+    {
+      "epoch": 12.233662145499384,
+      "grad_norm": 0.11468834429979324,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 119058
+    },
+    {
+      "epoch": 12.233970406905055,
+      "grad_norm": 0.1281205713748932,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 119061
+    },
+    {
+      "epoch": 12.234278668310727,
+      "grad_norm": 0.054583653807640076,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 119064
+    },
+    {
+      "epoch": 12.2345869297164,
+      "grad_norm": 0.05678866058588028,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 119067
+    },
+    {
+      "epoch": 12.234895191122071,
+      "grad_norm": 0.04765111580491066,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 119070
+    },
+    {
+      "epoch": 12.235203452527744,
+      "grad_norm": 0.04203588515520096,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 119073
+    },
+    {
+      "epoch": 12.235511713933416,
+      "grad_norm": 0.11908500641584396,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 119076
+    },
+    {
+      "epoch": 12.235819975339087,
+      "grad_norm": 0.09316672384738922,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 119079
+    },
+    {
+      "epoch": 12.23612823674476,
+      "grad_norm": 0.10216987133026123,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 119082
+    },
+    {
+      "epoch": 12.236436498150432,
+      "grad_norm": 0.11083834618330002,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 119085
+    },
+    {
+      "epoch": 12.236744759556103,
+      "grad_norm": 0.06925109773874283,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 119088
+    },
+    {
+      "epoch": 12.237053020961776,
+      "grad_norm": 0.0509336031973362,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 119091
+    },
+    {
+      "epoch": 12.237361282367448,
+      "grad_norm": 0.038560494780540466,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 119094
+    },
+    {
+      "epoch": 12.23766954377312,
+      "grad_norm": 0.07917262613773346,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 119097
+    },
+    {
+      "epoch": 12.237977805178792,
+      "grad_norm": 0.06403060257434845,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 119100
+    },
+    {
+      "epoch": 12.238286066584463,
+      "grad_norm": 0.0364985316991806,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 119103
+    },
+    {
+      "epoch": 12.238594327990135,
+      "grad_norm": 0.04724942147731781,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 119106
+    },
+    {
+      "epoch": 12.238902589395808,
+      "grad_norm": 0.12185391783714294,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 119109
+    },
+    {
+      "epoch": 12.239210850801479,
+      "grad_norm": 0.06551426649093628,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 119112
+    },
+    {
+      "epoch": 12.239519112207152,
+      "grad_norm": 0.04911579191684723,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 119115
+    },
+    {
+      "epoch": 12.239827373612824,
+      "grad_norm": 0.04070357233285904,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 119118
+    },
+    {
+      "epoch": 12.240135635018495,
+      "grad_norm": 0.14026470482349396,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 119121
+    },
+    {
+      "epoch": 12.240443896424168,
+      "grad_norm": 0.07128757983446121,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 119124
+    },
+    {
+      "epoch": 12.24075215782984,
+      "grad_norm": 0.062092967331409454,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 119127
+    },
+    {
+      "epoch": 12.241060419235511,
+      "grad_norm": 0.06296969205141068,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 119130
+    },
+    {
+      "epoch": 12.241368680641184,
+      "grad_norm": 0.09423550218343735,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 119133
+    },
+    {
+      "epoch": 12.241676942046857,
+      "grad_norm": 0.05814823508262634,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 119136
+    },
+    {
+      "epoch": 12.241985203452527,
+      "grad_norm": 0.05670604109764099,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 119139
+    },
+    {
+      "epoch": 12.2422934648582,
+      "grad_norm": 0.041170332580804825,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 119142
+    },
+    {
+      "epoch": 12.24260172626387,
+      "grad_norm": 0.05782140791416168,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 119145
+    },
+    {
+      "epoch": 12.242909987669544,
+      "grad_norm": 0.04377881810069084,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 119148
+    },
+    {
+      "epoch": 12.243218249075216,
+      "grad_norm": 0.03731423243880272,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 119151
+    },
+    {
+      "epoch": 12.243526510480887,
+      "grad_norm": 0.08947639167308807,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 119154
+    },
+    {
+      "epoch": 12.24383477188656,
+      "grad_norm": 0.09892389178276062,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 119157
+    },
+    {
+      "epoch": 12.244143033292232,
+      "grad_norm": 0.13070419430732727,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 119160
+    },
+    {
+      "epoch": 12.244451294697903,
+      "grad_norm": 0.05224129930138588,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 119163
+    },
+    {
+      "epoch": 12.244759556103576,
+      "grad_norm": 0.06416375190019608,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 119166
+    },
+    {
+      "epoch": 12.245067817509248,
+      "grad_norm": 0.03321628272533417,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 119169
+    },
+    {
+      "epoch": 12.24537607891492,
+      "grad_norm": 0.041123420000076294,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 119172
+    },
+    {
+      "epoch": 12.245684340320592,
+      "grad_norm": 0.09715761244297028,
+      "learning_rate": 0.01,
+      "loss": 1.8918,
+      "step": 119175
+    },
+    {
+      "epoch": 12.245992601726265,
+      "grad_norm": 0.05635785683989525,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 119178
+    },
+    {
+      "epoch": 12.246300863131935,
+      "grad_norm": 0.08281838148832321,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 119181
+    },
+    {
+      "epoch": 12.246609124537608,
+      "grad_norm": 0.06842327117919922,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 119184
+    },
+    {
+      "epoch": 12.24691738594328,
+      "grad_norm": 0.044265225529670715,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 119187
+    },
+    {
+      "epoch": 12.247225647348952,
+      "grad_norm": 0.045708850026130676,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 119190
+    },
+    {
+      "epoch": 12.247533908754624,
+      "grad_norm": 0.05862099304795265,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 119193
+    },
+    {
+      "epoch": 12.247842170160295,
+      "grad_norm": 0.0953894555568695,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 119196
+    },
+    {
+      "epoch": 12.248150431565968,
+      "grad_norm": 0.09879028797149658,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 119199
+    },
+    {
+      "epoch": 12.24845869297164,
+      "grad_norm": 0.060379255563020706,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 119202
+    },
+    {
+      "epoch": 12.248766954377311,
+      "grad_norm": 0.04327157512307167,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 119205
+    },
+    {
+      "epoch": 12.249075215782984,
+      "grad_norm": 0.04136524721980095,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 119208
+    },
+    {
+      "epoch": 12.249383477188656,
+      "grad_norm": 0.04830603674054146,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 119211
+    },
+    {
+      "epoch": 12.249691738594327,
+      "grad_norm": 0.07206027954816818,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 119214
+    },
+    {
+      "epoch": 12.25,
+      "grad_norm": 0.11831352114677429,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 119217
+    },
+    {
+      "epoch": 12.250308261405673,
+      "grad_norm": 0.09627742320299149,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 119220
+    },
+    {
+      "epoch": 12.250616522811344,
+      "grad_norm": 0.09025995433330536,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 119223
+    },
+    {
+      "epoch": 12.250924784217016,
+      "grad_norm": 0.041941117495298386,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 119226
+    },
+    {
+      "epoch": 12.251233045622689,
+      "grad_norm": 0.039908573031425476,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 119229
+    },
+    {
+      "epoch": 12.25154130702836,
+      "grad_norm": 0.11591826379299164,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 119232
+    },
+    {
+      "epoch": 12.251849568434032,
+      "grad_norm": 0.03896136209368706,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 119235
+    },
+    {
+      "epoch": 12.252157829839705,
+      "grad_norm": 0.05069341883063316,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 119238
+    },
+    {
+      "epoch": 12.252466091245376,
+      "grad_norm": 0.04808599501848221,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 119241
+    },
+    {
+      "epoch": 12.252774352651048,
+      "grad_norm": 0.03207084536552429,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 119244
+    },
+    {
+      "epoch": 12.25308261405672,
+      "grad_norm": 0.050113700330257416,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 119247
+    },
+    {
+      "epoch": 12.253390875462392,
+      "grad_norm": 0.0495215505361557,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 119250
+    },
+    {
+      "epoch": 12.253699136868065,
+      "grad_norm": 0.12017959356307983,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 119253
+    },
+    {
+      "epoch": 12.254007398273735,
+      "grad_norm": 0.07121411710977554,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 119256
+    },
+    {
+      "epoch": 12.254315659679408,
+      "grad_norm": 0.11534114927053452,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 119259
+    },
+    {
+      "epoch": 12.25462392108508,
+      "grad_norm": 0.05048522725701332,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 119262
+    },
+    {
+      "epoch": 12.254932182490752,
+      "grad_norm": 0.03871830925345421,
+      "learning_rate": 0.01,
+      "loss": 1.9672,
+      "step": 119265
+    },
+    {
+      "epoch": 12.255240443896424,
+      "grad_norm": 0.04501257464289665,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 119268
+    },
+    {
+      "epoch": 12.255548705302097,
+      "grad_norm": 0.03589480742812157,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 119271
+    },
+    {
+      "epoch": 12.255856966707768,
+      "grad_norm": 0.06698665767908096,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 119274
+    },
+    {
+      "epoch": 12.25616522811344,
+      "grad_norm": 0.13130944967269897,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 119277
+    },
+    {
+      "epoch": 12.256473489519113,
+      "grad_norm": 0.10875599831342697,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 119280
+    },
+    {
+      "epoch": 12.256781750924784,
+      "grad_norm": 0.06583302468061447,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 119283
+    },
+    {
+      "epoch": 12.257090012330456,
+      "grad_norm": 0.05685946345329285,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 119286
+    },
+    {
+      "epoch": 12.25739827373613,
+      "grad_norm": 0.10563762485980988,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 119289
+    },
+    {
+      "epoch": 12.2577065351418,
+      "grad_norm": 0.07009796798229218,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 119292
+    },
+    {
+      "epoch": 12.258014796547473,
+      "grad_norm": 0.03915143758058548,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 119295
+    },
+    {
+      "epoch": 12.258323057953143,
+      "grad_norm": 0.03810437023639679,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 119298
+    },
+    {
+      "epoch": 12.258631319358816,
+      "grad_norm": 0.037026748061180115,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 119301
+    },
+    {
+      "epoch": 12.258939580764489,
+      "grad_norm": 0.12063061445951462,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 119304
+    },
+    {
+      "epoch": 12.25924784217016,
+      "grad_norm": 0.04207170382142067,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 119307
+    },
+    {
+      "epoch": 12.259556103575832,
+      "grad_norm": 0.09984170645475388,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 119310
+    },
+    {
+      "epoch": 12.259864364981505,
+      "grad_norm": 0.042393285781145096,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 119313
+    },
+    {
+      "epoch": 12.260172626387176,
+      "grad_norm": 0.06870042532682419,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 119316
+    },
+    {
+      "epoch": 12.260480887792848,
+      "grad_norm": 0.042919475585222244,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 119319
+    },
+    {
+      "epoch": 12.260789149198521,
+      "grad_norm": 0.03628470376133919,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 119322
+    },
+    {
+      "epoch": 12.261097410604192,
+      "grad_norm": 0.1155962273478508,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 119325
+    },
+    {
+      "epoch": 12.261405672009865,
+      "grad_norm": 0.05997050553560257,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 119328
+    },
+    {
+      "epoch": 12.261713933415537,
+      "grad_norm": 0.058512236922979355,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 119331
+    },
+    {
+      "epoch": 12.262022194821208,
+      "grad_norm": 0.04778609424829483,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 119334
+    },
+    {
+      "epoch": 12.26233045622688,
+      "grad_norm": 0.09373581409454346,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 119337
+    },
+    {
+      "epoch": 12.262638717632552,
+      "grad_norm": 0.06936026364564896,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 119340
+    },
+    {
+      "epoch": 12.262946979038224,
+      "grad_norm": 0.12991760671138763,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 119343
+    },
+    {
+      "epoch": 12.263255240443897,
+      "grad_norm": 0.08365713804960251,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 119346
+    },
+    {
+      "epoch": 12.263563501849568,
+      "grad_norm": 0.08354364335536957,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 119349
+    },
+    {
+      "epoch": 12.26387176325524,
+      "grad_norm": 0.053151343017816544,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 119352
+    },
+    {
+      "epoch": 12.264180024660913,
+      "grad_norm": 0.05753325670957565,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 119355
+    },
+    {
+      "epoch": 12.264488286066584,
+      "grad_norm": 0.07821407914161682,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 119358
+    },
+    {
+      "epoch": 12.264796547472256,
+      "grad_norm": 0.0947427898645401,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 119361
+    },
+    {
+      "epoch": 12.265104808877929,
+      "grad_norm": 0.048926010727882385,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 119364
+    },
+    {
+      "epoch": 12.2654130702836,
+      "grad_norm": 0.047043971717357635,
+      "learning_rate": 0.01,
+      "loss": 1.9675,
+      "step": 119367
+    },
+    {
+      "epoch": 12.265721331689273,
+      "grad_norm": 0.03395242989063263,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 119370
+    },
+    {
+      "epoch": 12.266029593094945,
+      "grad_norm": 0.03889608010649681,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 119373
+    },
+    {
+      "epoch": 12.266337854500616,
+      "grad_norm": 0.041535381227731705,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 119376
+    },
+    {
+      "epoch": 12.266646115906289,
+      "grad_norm": 0.07542633265256882,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 119379
+    },
+    {
+      "epoch": 12.266954377311961,
+      "grad_norm": 0.06241780146956444,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 119382
+    },
+    {
+      "epoch": 12.267262638717632,
+      "grad_norm": 0.12422860413789749,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 119385
+    },
+    {
+      "epoch": 12.267570900123305,
+      "grad_norm": 0.09422612190246582,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 119388
+    },
+    {
+      "epoch": 12.267879161528976,
+      "grad_norm": 0.06918896734714508,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 119391
+    },
+    {
+      "epoch": 12.268187422934648,
+      "grad_norm": 0.08683785796165466,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 119394
+    },
+    {
+      "epoch": 12.268495684340321,
+      "grad_norm": 0.13475213944911957,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 119397
+    },
+    {
+      "epoch": 12.268803945745992,
+      "grad_norm": 0.09755492955446243,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 119400
+    },
+    {
+      "epoch": 12.269112207151665,
+      "grad_norm": 0.11172865331172943,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 119403
+    },
+    {
+      "epoch": 12.269420468557337,
+      "grad_norm": 0.06556771695613861,
+      "learning_rate": 0.01,
+      "loss": 1.8946,
+      "step": 119406
+    },
+    {
+      "epoch": 12.269728729963008,
+      "grad_norm": 0.04608120396733284,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 119409
+    },
+    {
+      "epoch": 12.27003699136868,
+      "grad_norm": 0.07218800485134125,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 119412
+    },
+    {
+      "epoch": 12.270345252774353,
+      "grad_norm": 0.061452243477106094,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 119415
+    },
+    {
+      "epoch": 12.270653514180024,
+      "grad_norm": 0.04344675689935684,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 119418
+    },
+    {
+      "epoch": 12.270961775585697,
+      "grad_norm": 0.11082371324300766,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 119421
+    },
+    {
+      "epoch": 12.27127003699137,
+      "grad_norm": 0.06729606539011002,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 119424
+    },
+    {
+      "epoch": 12.27157829839704,
+      "grad_norm": 0.09937182068824768,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 119427
+    },
+    {
+      "epoch": 12.271886559802713,
+      "grad_norm": 0.06512150913476944,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 119430
+    },
+    {
+      "epoch": 12.272194821208386,
+      "grad_norm": 0.08961142599582672,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 119433
+    },
+    {
+      "epoch": 12.272503082614056,
+      "grad_norm": 0.06848619133234024,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 119436
+    },
+    {
+      "epoch": 12.272811344019729,
+      "grad_norm": 0.07923544943332672,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 119439
+    },
+    {
+      "epoch": 12.2731196054254,
+      "grad_norm": 0.09845848381519318,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 119442
+    },
+    {
+      "epoch": 12.273427866831073,
+      "grad_norm": 0.0953758955001831,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 119445
+    },
+    {
+      "epoch": 12.273736128236745,
+      "grad_norm": 0.0558675080537796,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 119448
+    },
+    {
+      "epoch": 12.274044389642416,
+      "grad_norm": 0.07507700473070145,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 119451
+    },
+    {
+      "epoch": 12.274352651048089,
+      "grad_norm": 0.09544411301612854,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 119454
+    },
+    {
+      "epoch": 12.274660912453761,
+      "grad_norm": 0.03666847571730614,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 119457
+    },
+    {
+      "epoch": 12.274969173859432,
+      "grad_norm": 0.10058706998825073,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 119460
+    },
+    {
+      "epoch": 12.275277435265105,
+      "grad_norm": 0.09659945219755173,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 119463
+    },
+    {
+      "epoch": 12.275585696670777,
+      "grad_norm": 0.08124331384897232,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 119466
+    },
+    {
+      "epoch": 12.275893958076448,
+      "grad_norm": 0.03858707845211029,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 119469
+    },
+    {
+      "epoch": 12.276202219482121,
+      "grad_norm": 0.03947792947292328,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 119472
+    },
+    {
+      "epoch": 12.276510480887794,
+      "grad_norm": 0.11748308688402176,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 119475
+    },
+    {
+      "epoch": 12.276818742293464,
+      "grad_norm": 0.05301667004823685,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 119478
+    },
+    {
+      "epoch": 12.277127003699137,
+      "grad_norm": 0.041350193321704865,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 119481
+    },
+    {
+      "epoch": 12.27743526510481,
+      "grad_norm": 0.18285071849822998,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 119484
+    },
+    {
+      "epoch": 12.27774352651048,
+      "grad_norm": 0.06876923143863678,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 119487
+    },
+    {
+      "epoch": 12.278051787916153,
+      "grad_norm": 0.06623870134353638,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 119490
+    },
+    {
+      "epoch": 12.278360049321824,
+      "grad_norm": 0.0681280568242073,
+      "learning_rate": 0.01,
+      "loss": 1.895,
+      "step": 119493
+    },
+    {
+      "epoch": 12.278668310727497,
+      "grad_norm": 0.0857885330915451,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 119496
+    },
+    {
+      "epoch": 12.27897657213317,
+      "grad_norm": 0.055479828268289566,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 119499
+    },
+    {
+      "epoch": 12.27928483353884,
+      "grad_norm": 0.04913216084241867,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 119502
+    },
+    {
+      "epoch": 12.279593094944513,
+      "grad_norm": 0.036898694932460785,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 119505
+    },
+    {
+      "epoch": 12.279901356350186,
+      "grad_norm": 0.049119070172309875,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 119508
+    },
+    {
+      "epoch": 12.280209617755856,
+      "grad_norm": 0.04535537585616112,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 119511
+    },
+    {
+      "epoch": 12.280517879161529,
+      "grad_norm": 0.0731346532702446,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 119514
+    },
+    {
+      "epoch": 12.280826140567202,
+      "grad_norm": 0.08404407650232315,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 119517
+    },
+    {
+      "epoch": 12.281134401972873,
+      "grad_norm": 0.06973934918642044,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 119520
+    },
+    {
+      "epoch": 12.281442663378545,
+      "grad_norm": 0.10510105639696121,
+      "learning_rate": 0.01,
+      "loss": 1.9652,
+      "step": 119523
+    },
+    {
+      "epoch": 12.281750924784218,
+      "grad_norm": 0.04614637792110443,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 119526
+    },
+    {
+      "epoch": 12.282059186189889,
+      "grad_norm": 0.14483202993869781,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 119529
+    },
+    {
+      "epoch": 12.282367447595561,
+      "grad_norm": 0.06402111798524857,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 119532
+    },
+    {
+      "epoch": 12.282675709001232,
+      "grad_norm": 0.035851992666721344,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 119535
+    },
+    {
+      "epoch": 12.282983970406905,
+      "grad_norm": 0.06560065597295761,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 119538
+    },
+    {
+      "epoch": 12.283292231812577,
+      "grad_norm": 0.0462896004319191,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 119541
+    },
+    {
+      "epoch": 12.283600493218248,
+      "grad_norm": 0.07922311872243881,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 119544
+    },
+    {
+      "epoch": 12.283908754623921,
+      "grad_norm": 0.04975975677371025,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 119547
+    },
+    {
+      "epoch": 12.284217016029594,
+      "grad_norm": 0.05062318593263626,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 119550
+    },
+    {
+      "epoch": 12.284525277435264,
+      "grad_norm": 0.1192026287317276,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 119553
+    },
+    {
+      "epoch": 12.284833538840937,
+      "grad_norm": 0.10026537626981735,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 119556
+    },
+    {
+      "epoch": 12.28514180024661,
+      "grad_norm": 0.048863403499126434,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 119559
+    },
+    {
+      "epoch": 12.28545006165228,
+      "grad_norm": 0.034858595579862595,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 119562
+    },
+    {
+      "epoch": 12.285758323057953,
+      "grad_norm": 0.11180103570222855,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 119565
+    },
+    {
+      "epoch": 12.286066584463626,
+      "grad_norm": 0.08020367473363876,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 119568
+    },
+    {
+      "epoch": 12.286374845869297,
+      "grad_norm": 0.0711284652352333,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 119571
+    },
+    {
+      "epoch": 12.28668310727497,
+      "grad_norm": 0.037581104785203934,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 119574
+    },
+    {
+      "epoch": 12.286991368680642,
+      "grad_norm": 0.04725247994065285,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 119577
+    },
+    {
+      "epoch": 12.287299630086313,
+      "grad_norm": 0.052582450211048126,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 119580
+    },
+    {
+      "epoch": 12.287607891491986,
+      "grad_norm": 0.04754595831036568,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 119583
+    },
+    {
+      "epoch": 12.287916152897656,
+      "grad_norm": 0.0919022411108017,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 119586
+    },
+    {
+      "epoch": 12.288224414303329,
+      "grad_norm": 0.08994028717279434,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 119589
+    },
+    {
+      "epoch": 12.288532675709002,
+      "grad_norm": 0.08245561271905899,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 119592
+    },
+    {
+      "epoch": 12.288840937114673,
+      "grad_norm": 0.09249047189950943,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 119595
+    },
+    {
+      "epoch": 12.289149198520345,
+      "grad_norm": 0.044289279729127884,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 119598
+    },
+    {
+      "epoch": 12.289457459926018,
+      "grad_norm": 0.04035840183496475,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 119601
+    },
+    {
+      "epoch": 12.289765721331689,
+      "grad_norm": 0.04231641814112663,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 119604
+    },
+    {
+      "epoch": 12.290073982737361,
+      "grad_norm": 0.037263207137584686,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 119607
+    },
+    {
+      "epoch": 12.290382244143034,
+      "grad_norm": 0.10813697427511215,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 119610
+    },
+    {
+      "epoch": 12.290690505548705,
+      "grad_norm": 0.0699143186211586,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 119613
+    },
+    {
+      "epoch": 12.290998766954377,
+      "grad_norm": 0.1337338536977768,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 119616
+    },
+    {
+      "epoch": 12.29130702836005,
+      "grad_norm": 0.04969757795333862,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 119619
+    },
+    {
+      "epoch": 12.291615289765721,
+      "grad_norm": 0.08740051090717316,
+      "learning_rate": 0.01,
+      "loss": 1.9645,
+      "step": 119622
+    },
+    {
+      "epoch": 12.291923551171394,
+      "grad_norm": 0.08194126188755035,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 119625
+    },
+    {
+      "epoch": 12.292231812577066,
+      "grad_norm": 0.06093340367078781,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 119628
+    },
+    {
+      "epoch": 12.292540073982737,
+      "grad_norm": 0.10872428119182587,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 119631
+    },
+    {
+      "epoch": 12.29284833538841,
+      "grad_norm": 0.12310198694467545,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 119634
+    },
+    {
+      "epoch": 12.29315659679408,
+      "grad_norm": 0.07006506621837616,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 119637
+    },
+    {
+      "epoch": 12.293464858199753,
+      "grad_norm": 0.04315321519970894,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 119640
+    },
+    {
+      "epoch": 12.293773119605426,
+      "grad_norm": 0.03495105728507042,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 119643
+    },
+    {
+      "epoch": 12.294081381011097,
+      "grad_norm": 0.049105752259492874,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 119646
+    },
+    {
+      "epoch": 12.29438964241677,
+      "grad_norm": 0.13561686873435974,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 119649
+    },
+    {
+      "epoch": 12.294697903822442,
+      "grad_norm": 0.0391533300280571,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 119652
+    },
+    {
+      "epoch": 12.295006165228113,
+      "grad_norm": 0.09752475470304489,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 119655
+    },
+    {
+      "epoch": 12.295314426633785,
+      "grad_norm": 0.08919859677553177,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 119658
+    },
+    {
+      "epoch": 12.295622688039458,
+      "grad_norm": 0.057920221239328384,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 119661
+    },
+    {
+      "epoch": 12.295930949445129,
+      "grad_norm": 0.06989624351263046,
+      "learning_rate": 0.01,
+      "loss": 1.9636,
+      "step": 119664
+    },
+    {
+      "epoch": 12.296239210850802,
+      "grad_norm": 0.04143437743186951,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 119667
+    },
+    {
+      "epoch": 12.296547472256474,
+      "grad_norm": 0.03357059881091118,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 119670
+    },
+    {
+      "epoch": 12.296855733662145,
+      "grad_norm": 0.08189679682254791,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 119673
+    },
+    {
+      "epoch": 12.297163995067818,
+      "grad_norm": 0.09445969760417938,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 119676
+    },
+    {
+      "epoch": 12.29747225647349,
+      "grad_norm": 0.07666604220867157,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 119679
+    },
+    {
+      "epoch": 12.297780517879161,
+      "grad_norm": 0.060300711542367935,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 119682
+    },
+    {
+      "epoch": 12.298088779284834,
+      "grad_norm": 0.04202558472752571,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 119685
+    },
+    {
+      "epoch": 12.298397040690505,
+      "grad_norm": 0.05015214905142784,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 119688
+    },
+    {
+      "epoch": 12.298705302096177,
+      "grad_norm": 0.11163275688886642,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 119691
+    },
+    {
+      "epoch": 12.29901356350185,
+      "grad_norm": 0.09391912817955017,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 119694
+    },
+    {
+      "epoch": 12.299321824907521,
+      "grad_norm": 0.035903919488191605,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 119697
+    },
+    {
+      "epoch": 12.299630086313194,
+      "grad_norm": 0.15324540436267853,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 119700
+    },
+    {
+      "epoch": 12.299938347718866,
+      "grad_norm": 0.13761693239212036,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 119703
+    },
+    {
+      "epoch": 12.300246609124537,
+      "grad_norm": 0.10357910394668579,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 119706
+    },
+    {
+      "epoch": 12.30055487053021,
+      "grad_norm": 0.07113146036863327,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 119709
+    },
+    {
+      "epoch": 12.300863131935882,
+      "grad_norm": 0.05279774218797684,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 119712
+    },
+    {
+      "epoch": 12.301171393341553,
+      "grad_norm": 0.03502047434449196,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 119715
+    },
+    {
+      "epoch": 12.301479654747226,
+      "grad_norm": 0.047808703035116196,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 119718
+    },
+    {
+      "epoch": 12.301787916152898,
+      "grad_norm": 0.05250471085309982,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 119721
+    },
+    {
+      "epoch": 12.30209617755857,
+      "grad_norm": 0.041899871081113815,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 119724
+    },
+    {
+      "epoch": 12.302404438964242,
+      "grad_norm": 0.1110948920249939,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 119727
+    },
+    {
+      "epoch": 12.302712700369913,
+      "grad_norm": 0.07395997643470764,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 119730
+    },
+    {
+      "epoch": 12.303020961775585,
+      "grad_norm": 0.09498132765293121,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 119733
+    },
+    {
+      "epoch": 12.303329223181258,
+      "grad_norm": 0.09375674277544022,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 119736
+    },
+    {
+      "epoch": 12.303637484586929,
+      "grad_norm": 0.047957029193639755,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 119739
+    },
+    {
+      "epoch": 12.303945745992602,
+      "grad_norm": 0.038424279540777206,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 119742
+    },
+    {
+      "epoch": 12.304254007398274,
+      "grad_norm": 0.08323843777179718,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 119745
+    },
+    {
+      "epoch": 12.304562268803945,
+      "grad_norm": 0.09454569965600967,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 119748
+    },
+    {
+      "epoch": 12.304870530209618,
+      "grad_norm": 0.12319619208574295,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 119751
+    },
+    {
+      "epoch": 12.30517879161529,
+      "grad_norm": 0.05019475147128105,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 119754
+    },
+    {
+      "epoch": 12.305487053020961,
+      "grad_norm": 0.08038980513811111,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 119757
+    },
+    {
+      "epoch": 12.305795314426634,
+      "grad_norm": 0.09341848641633987,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 119760
+    },
+    {
+      "epoch": 12.306103575832307,
+      "grad_norm": 0.08108554780483246,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 119763
+    },
+    {
+      "epoch": 12.306411837237977,
+      "grad_norm": 0.10190309584140778,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 119766
+    },
+    {
+      "epoch": 12.30672009864365,
+      "grad_norm": 0.07136478275060654,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 119769
+    },
+    {
+      "epoch": 12.307028360049323,
+      "grad_norm": 0.05075160413980484,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 119772
+    },
+    {
+      "epoch": 12.307336621454994,
+      "grad_norm": 0.04689573496580124,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 119775
+    },
+    {
+      "epoch": 12.307644882860666,
+      "grad_norm": 0.04788019880652428,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 119778
+    },
+    {
+      "epoch": 12.307953144266337,
+      "grad_norm": 0.04214687645435333,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 119781
+    },
+    {
+      "epoch": 12.30826140567201,
+      "grad_norm": 0.09224370867013931,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 119784
+    },
+    {
+      "epoch": 12.308569667077682,
+      "grad_norm": 0.07086529582738876,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 119787
+    },
+    {
+      "epoch": 12.308877928483353,
+      "grad_norm": 0.05687602609395981,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 119790
+    },
+    {
+      "epoch": 12.309186189889026,
+      "grad_norm": 0.061124287545681,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 119793
+    },
+    {
+      "epoch": 12.309494451294698,
+      "grad_norm": 0.0497773177921772,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 119796
+    },
+    {
+      "epoch": 12.30980271270037,
+      "grad_norm": 0.03449302166700363,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 119799
+    },
+    {
+      "epoch": 12.310110974106042,
+      "grad_norm": 0.043521273881196976,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 119802
+    },
+    {
+      "epoch": 12.310419235511715,
+      "grad_norm": 0.04101839289069176,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 119805
+    },
+    {
+      "epoch": 12.310727496917385,
+      "grad_norm": 0.13504955172538757,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 119808
+    },
+    {
+      "epoch": 12.311035758323058,
+      "grad_norm": 0.07502903789281845,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 119811
+    },
+    {
+      "epoch": 12.31134401972873,
+      "grad_norm": 0.08481600880622864,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 119814
+    },
+    {
+      "epoch": 12.311652281134402,
+      "grad_norm": 0.07710324972867966,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 119817
+    },
+    {
+      "epoch": 12.311960542540074,
+      "grad_norm": 0.11857612431049347,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 119820
+    },
+    {
+      "epoch": 12.312268803945745,
+      "grad_norm": 0.03913726657629013,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 119823
+    },
+    {
+      "epoch": 12.312577065351418,
+      "grad_norm": 0.06333804130554199,
+      "learning_rate": 0.01,
+      "loss": 1.9678,
+      "step": 119826
+    },
+    {
+      "epoch": 12.31288532675709,
+      "grad_norm": 0.053462814539670944,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 119829
+    },
+    {
+      "epoch": 12.313193588162761,
+      "grad_norm": 0.036863964051008224,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 119832
+    },
+    {
+      "epoch": 12.313501849568434,
+      "grad_norm": 0.05140704661607742,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 119835
+    },
+    {
+      "epoch": 12.313810110974106,
+      "grad_norm": 0.12734730541706085,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 119838
+    },
+    {
+      "epoch": 12.314118372379777,
+      "grad_norm": 0.04494262859225273,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 119841
+    },
+    {
+      "epoch": 12.31442663378545,
+      "grad_norm": 0.10952113568782806,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 119844
+    },
+    {
+      "epoch": 12.314734895191123,
+      "grad_norm": 0.12877106666564941,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 119847
+    },
+    {
+      "epoch": 12.315043156596793,
+      "grad_norm": 0.05764854699373245,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 119850
+    },
+    {
+      "epoch": 12.315351418002466,
+      "grad_norm": 0.07877784222364426,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 119853
+    },
+    {
+      "epoch": 12.315659679408139,
+      "grad_norm": 0.04288757964968681,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 119856
+    },
+    {
+      "epoch": 12.31596794081381,
+      "grad_norm": 0.05566611886024475,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 119859
+    },
+    {
+      "epoch": 12.316276202219482,
+      "grad_norm": 0.03457842767238617,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 119862
+    },
+    {
+      "epoch": 12.316584463625155,
+      "grad_norm": 0.07046034932136536,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 119865
+    },
+    {
+      "epoch": 12.316892725030826,
+      "grad_norm": 0.1305147260427475,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 119868
+    },
+    {
+      "epoch": 12.317200986436498,
+      "grad_norm": 0.14929534494876862,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 119871
+    },
+    {
+      "epoch": 12.317509247842171,
+      "grad_norm": 0.16848324239253998,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 119874
+    },
+    {
+      "epoch": 12.317817509247842,
+      "grad_norm": 0.13916730880737305,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 119877
+    },
+    {
+      "epoch": 12.318125770653515,
+      "grad_norm": 0.04896625503897667,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 119880
+    },
+    {
+      "epoch": 12.318434032059185,
+      "grad_norm": 0.0732089951634407,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 119883
+    },
+    {
+      "epoch": 12.318742293464858,
+      "grad_norm": 0.04600311815738678,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 119886
+    },
+    {
+      "epoch": 12.31905055487053,
+      "grad_norm": 0.057165056467056274,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 119889
+    },
+    {
+      "epoch": 12.319358816276202,
+      "grad_norm": 0.04162990674376488,
+      "learning_rate": 0.01,
+      "loss": 1.9041,
+      "step": 119892
+    },
+    {
+      "epoch": 12.319667077681874,
+      "grad_norm": 0.0429425910115242,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 119895
+    },
+    {
+      "epoch": 12.319975339087547,
+      "grad_norm": 0.06248023360967636,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 119898
+    },
+    {
+      "epoch": 12.320283600493218,
+      "grad_norm": 0.09580297768115997,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 119901
+    },
+    {
+      "epoch": 12.32059186189889,
+      "grad_norm": 0.08072346448898315,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 119904
+    },
+    {
+      "epoch": 12.320900123304563,
+      "grad_norm": 0.09270929545164108,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 119907
+    },
+    {
+      "epoch": 12.321208384710234,
+      "grad_norm": 0.06813566386699677,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 119910
+    },
+    {
+      "epoch": 12.321516646115906,
+      "grad_norm": 0.07171430438756943,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 119913
+    },
+    {
+      "epoch": 12.321824907521579,
+      "grad_norm": 0.056445494294166565,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 119916
+    },
+    {
+      "epoch": 12.32213316892725,
+      "grad_norm": 0.11223911494016647,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 119919
+    },
+    {
+      "epoch": 12.322441430332923,
+      "grad_norm": 0.09491009265184402,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 119922
+    },
+    {
+      "epoch": 12.322749691738593,
+      "grad_norm": 0.06327483057975769,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 119925
+    },
+    {
+      "epoch": 12.323057953144266,
+      "grad_norm": 0.04618639498949051,
+      "learning_rate": 0.01,
+      "loss": 1.9612,
+      "step": 119928
+    },
+    {
+      "epoch": 12.323366214549939,
+      "grad_norm": 0.03715759515762329,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 119931
+    },
+    {
+      "epoch": 12.32367447595561,
+      "grad_norm": 0.03354526311159134,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 119934
+    },
+    {
+      "epoch": 12.323982737361282,
+      "grad_norm": 0.06126822903752327,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 119937
+    },
+    {
+      "epoch": 12.324290998766955,
+      "grad_norm": 0.10685601830482483,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 119940
+    },
+    {
+      "epoch": 12.324599260172626,
+      "grad_norm": 0.07914771139621735,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 119943
+    },
+    {
+      "epoch": 12.324907521578298,
+      "grad_norm": 0.0697379782795906,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 119946
+    },
+    {
+      "epoch": 12.325215782983971,
+      "grad_norm": 0.1079431101679802,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 119949
+    },
+    {
+      "epoch": 12.325524044389642,
+      "grad_norm": 0.03759371489286423,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 119952
+    },
+    {
+      "epoch": 12.325832305795315,
+      "grad_norm": 0.06159079447388649,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 119955
+    },
+    {
+      "epoch": 12.326140567200987,
+      "grad_norm": 0.07581518590450287,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 119958
+    },
+    {
+      "epoch": 12.326448828606658,
+      "grad_norm": 0.06095201149582863,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 119961
+    },
+    {
+      "epoch": 12.32675709001233,
+      "grad_norm": 0.05279630050063133,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 119964
+    },
+    {
+      "epoch": 12.327065351418003,
+      "grad_norm": 0.035489410161972046,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 119967
+    },
+    {
+      "epoch": 12.327373612823674,
+      "grad_norm": 0.037815045565366745,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 119970
+    },
+    {
+      "epoch": 12.327681874229347,
+      "grad_norm": 0.05071322247385979,
+      "learning_rate": 0.01,
+      "loss": 1.9601,
+      "step": 119973
+    },
+    {
+      "epoch": 12.327990135635018,
+      "grad_norm": 0.04019641876220703,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 119976
+    },
+    {
+      "epoch": 12.32829839704069,
+      "grad_norm": 0.09706182032823563,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 119979
+    },
+    {
+      "epoch": 12.328606658446363,
+      "grad_norm": 0.08802255243062973,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 119982
+    },
+    {
+      "epoch": 12.328914919852034,
+      "grad_norm": 0.061536651104688644,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 119985
+    },
+    {
+      "epoch": 12.329223181257706,
+      "grad_norm": 0.07003441452980042,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 119988
+    },
+    {
+      "epoch": 12.329531442663379,
+      "grad_norm": 0.046714432537555695,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 119991
+    },
+    {
+      "epoch": 12.32983970406905,
+      "grad_norm": 0.05683311074972153,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 119994
+    },
+    {
+      "epoch": 12.330147965474723,
+      "grad_norm": 0.05732543021440506,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 119997
+    },
+    {
+      "epoch": 12.330456226880395,
+      "grad_norm": 0.04505174979567528,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 120000
+    },
+    {
+      "epoch": 12.330764488286066,
+      "grad_norm": 0.08947070688009262,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 120003
+    },
+    {
+      "epoch": 12.331072749691739,
+      "grad_norm": 0.12785474956035614,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 120006
+    },
+    {
+      "epoch": 12.331381011097411,
+      "grad_norm": 0.09339133650064468,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 120009
+    },
+    {
+      "epoch": 12.331689272503082,
+      "grad_norm": 0.051951028406620026,
+      "learning_rate": 0.01,
+      "loss": 1.9614,
+      "step": 120012
+    },
+    {
+      "epoch": 12.331997533908755,
+      "grad_norm": 0.04255111515522003,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 120015
+    },
+    {
+      "epoch": 12.332305795314426,
+      "grad_norm": 0.055113568902015686,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 120018
+    },
+    {
+      "epoch": 12.332614056720098,
+      "grad_norm": 0.04617108777165413,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 120021
+    },
+    {
+      "epoch": 12.332922318125771,
+      "grad_norm": 0.06585944443941116,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 120024
+    },
+    {
+      "epoch": 12.333230579531442,
+      "grad_norm": 0.042562294751405716,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 120027
+    },
+    {
+      "epoch": 12.333538840937115,
+      "grad_norm": 0.04126080498099327,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 120030
+    },
+    {
+      "epoch": 12.333847102342787,
+      "grad_norm": 0.03414085879921913,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 120033
+    },
+    {
+      "epoch": 12.334155363748458,
+      "grad_norm": 0.05674465373158455,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 120036
+    },
+    {
+      "epoch": 12.33446362515413,
+      "grad_norm": 0.13386379182338715,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 120039
+    },
+    {
+      "epoch": 12.334771886559803,
+      "grad_norm": 0.05028100311756134,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 120042
+    },
+    {
+      "epoch": 12.335080147965474,
+      "grad_norm": 0.07372204959392548,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 120045
+    },
+    {
+      "epoch": 12.335388409371147,
+      "grad_norm": 0.03808963671326637,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 120048
+    },
+    {
+      "epoch": 12.33569667077682,
+      "grad_norm": 0.06065637245774269,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 120051
+    },
+    {
+      "epoch": 12.33600493218249,
+      "grad_norm": 0.04890544340014458,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 120054
+    },
+    {
+      "epoch": 12.336313193588163,
+      "grad_norm": 0.13426274061203003,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 120057
+    },
+    {
+      "epoch": 12.336621454993836,
+      "grad_norm": 0.0989098846912384,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 120060
+    },
+    {
+      "epoch": 12.336929716399506,
+      "grad_norm": 0.07141082733869553,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 120063
+    },
+    {
+      "epoch": 12.337237977805179,
+      "grad_norm": 0.06102481856942177,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 120066
+    },
+    {
+      "epoch": 12.337546239210852,
+      "grad_norm": 0.060864515602588654,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 120069
+    },
+    {
+      "epoch": 12.337854500616523,
+      "grad_norm": 0.09040584415197372,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 120072
+    },
+    {
+      "epoch": 12.338162762022195,
+      "grad_norm": 0.04068896919488907,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 120075
+    },
+    {
+      "epoch": 12.338471023427866,
+      "grad_norm": 0.10498625785112381,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 120078
+    },
+    {
+      "epoch": 12.338779284833539,
+      "grad_norm": 0.061306800693273544,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 120081
+    },
+    {
+      "epoch": 12.339087546239211,
+      "grad_norm": 0.10269539058208466,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 120084
+    },
+    {
+      "epoch": 12.339395807644882,
+      "grad_norm": 0.07771135121583939,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 120087
+    },
+    {
+      "epoch": 12.339704069050555,
+      "grad_norm": 0.08635908365249634,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 120090
+    },
+    {
+      "epoch": 12.340012330456227,
+      "grad_norm": 0.05086525157094002,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 120093
+    },
+    {
+      "epoch": 12.340320591861898,
+      "grad_norm": 0.06784183531999588,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 120096
+    },
+    {
+      "epoch": 12.340628853267571,
+      "grad_norm": 0.09301754087209702,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 120099
+    },
+    {
+      "epoch": 12.340937114673244,
+      "grad_norm": 0.05019529536366463,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 120102
+    },
+    {
+      "epoch": 12.341245376078914,
+      "grad_norm": 0.07970885932445526,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 120105
+    },
+    {
+      "epoch": 12.341553637484587,
+      "grad_norm": 0.07051090151071548,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 120108
+    },
+    {
+      "epoch": 12.34186189889026,
+      "grad_norm": 0.0799780935049057,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 120111
+    },
+    {
+      "epoch": 12.34217016029593,
+      "grad_norm": 0.053853873163461685,
+      "learning_rate": 0.01,
+      "loss": 1.9618,
+      "step": 120114
+    },
+    {
+      "epoch": 12.342478421701603,
+      "grad_norm": 0.03682436794042587,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 120117
+    },
+    {
+      "epoch": 12.342786683107274,
+      "grad_norm": 0.08985862135887146,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 120120
+    },
+    {
+      "epoch": 12.343094944512947,
+      "grad_norm": 0.09578026086091995,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 120123
+    },
+    {
+      "epoch": 12.34340320591862,
+      "grad_norm": 0.18115627765655518,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 120126
+    },
+    {
+      "epoch": 12.34371146732429,
+      "grad_norm": 0.09127353131771088,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 120129
+    },
+    {
+      "epoch": 12.344019728729963,
+      "grad_norm": 0.042836468666791916,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 120132
+    },
+    {
+      "epoch": 12.344327990135636,
+      "grad_norm": 0.04077746719121933,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 120135
+    },
+    {
+      "epoch": 12.344636251541306,
+      "grad_norm": 0.04519999399781227,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 120138
+    },
+    {
+      "epoch": 12.344944512946979,
+      "grad_norm": 0.04900500923395157,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 120141
+    },
+    {
+      "epoch": 12.345252774352652,
+      "grad_norm": 0.05115954577922821,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 120144
+    },
+    {
+      "epoch": 12.345561035758323,
+      "grad_norm": 0.040632762014865875,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 120147
+    },
+    {
+      "epoch": 12.345869297163995,
+      "grad_norm": 0.13975204527378082,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 120150
+    },
+    {
+      "epoch": 12.346177558569668,
+      "grad_norm": 0.04202760010957718,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 120153
+    },
+    {
+      "epoch": 12.346485819975339,
+      "grad_norm": 0.08207782357931137,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 120156
+    },
+    {
+      "epoch": 12.346794081381011,
+      "grad_norm": 0.052009355276823044,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 120159
+    },
+    {
+      "epoch": 12.347102342786684,
+      "grad_norm": 0.059325072914361954,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 120162
+    },
+    {
+      "epoch": 12.347410604192355,
+      "grad_norm": 0.05349455028772354,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 120165
+    },
+    {
+      "epoch": 12.347718865598027,
+      "grad_norm": 0.06158902868628502,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 120168
+    },
+    {
+      "epoch": 12.348027127003698,
+      "grad_norm": 0.09922099858522415,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 120171
+    },
+    {
+      "epoch": 12.348335388409371,
+      "grad_norm": 0.06710609048604965,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 120174
+    },
+    {
+      "epoch": 12.348643649815044,
+      "grad_norm": 0.11664668470621109,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 120177
+    },
+    {
+      "epoch": 12.348951911220714,
+      "grad_norm": 0.047379445284605026,
+      "learning_rate": 0.01,
+      "loss": 1.9718,
+      "step": 120180
+    },
+    {
+      "epoch": 12.349260172626387,
+      "grad_norm": 0.09358347952365875,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 120183
+    },
+    {
+      "epoch": 12.34956843403206,
+      "grad_norm": 0.11921723186969757,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 120186
+    },
+    {
+      "epoch": 12.34987669543773,
+      "grad_norm": 0.059619396924972534,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 120189
+    },
+    {
+      "epoch": 12.350184956843403,
+      "grad_norm": 0.03857177123427391,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 120192
+    },
+    {
+      "epoch": 12.350493218249076,
+      "grad_norm": 0.06697624921798706,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 120195
+    },
+    {
+      "epoch": 12.350801479654747,
+      "grad_norm": 0.08252769708633423,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 120198
+    },
+    {
+      "epoch": 12.35110974106042,
+      "grad_norm": 0.13171248137950897,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 120201
+    },
+    {
+      "epoch": 12.351418002466092,
+      "grad_norm": 0.05222803354263306,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 120204
+    },
+    {
+      "epoch": 12.351726263871763,
+      "grad_norm": 0.05304306373000145,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 120207
+    },
+    {
+      "epoch": 12.352034525277436,
+      "grad_norm": 0.04716663807630539,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 120210
+    },
+    {
+      "epoch": 12.352342786683106,
+      "grad_norm": 0.0832684114575386,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 120213
+    },
+    {
+      "epoch": 12.352651048088779,
+      "grad_norm": 0.049712538719177246,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 120216
+    },
+    {
+      "epoch": 12.352959309494452,
+      "grad_norm": 0.0388319157063961,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 120219
+    },
+    {
+      "epoch": 12.353267570900123,
+      "grad_norm": 0.0908878967165947,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 120222
+    },
+    {
+      "epoch": 12.353575832305795,
+      "grad_norm": 0.07359366863965988,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 120225
+    },
+    {
+      "epoch": 12.353884093711468,
+      "grad_norm": 0.08322012424468994,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 120228
+    },
+    {
+      "epoch": 12.354192355117139,
+      "grad_norm": 0.08135717362165451,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 120231
+    },
+    {
+      "epoch": 12.354500616522811,
+      "grad_norm": 0.05825061723589897,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 120234
+    },
+    {
+      "epoch": 12.354808877928484,
+      "grad_norm": 0.1178986206650734,
+      "learning_rate": 0.01,
+      "loss": 1.9642,
+      "step": 120237
+    },
+    {
+      "epoch": 12.355117139334155,
+      "grad_norm": 0.04324718937277794,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 120240
+    },
+    {
+      "epoch": 12.355425400739827,
+      "grad_norm": 0.09722001105546951,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 120243
+    },
+    {
+      "epoch": 12.3557336621455,
+      "grad_norm": 0.08974692225456238,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 120246
+    },
+    {
+      "epoch": 12.356041923551171,
+      "grad_norm": 0.05838640034198761,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 120249
+    },
+    {
+      "epoch": 12.356350184956844,
+      "grad_norm": 0.054912663996219635,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 120252
+    },
+    {
+      "epoch": 12.356658446362516,
+      "grad_norm": 0.042390115559101105,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 120255
+    },
+    {
+      "epoch": 12.356966707768187,
+      "grad_norm": 0.09722331166267395,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 120258
+    },
+    {
+      "epoch": 12.35727496917386,
+      "grad_norm": 0.147323876619339,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 120261
+    },
+    {
+      "epoch": 12.357583230579532,
+      "grad_norm": 0.07642398029565811,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 120264
+    },
+    {
+      "epoch": 12.357891491985203,
+      "grad_norm": 0.0503547303378582,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 120267
+    },
+    {
+      "epoch": 12.358199753390876,
+      "grad_norm": 0.04040350764989853,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 120270
+    },
+    {
+      "epoch": 12.358508014796547,
+      "grad_norm": 0.13021105527877808,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 120273
+    },
+    {
+      "epoch": 12.35881627620222,
+      "grad_norm": 0.053368840366601944,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 120276
+    },
+    {
+      "epoch": 12.359124537607892,
+      "grad_norm": 0.05143368989229202,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 120279
+    },
+    {
+      "epoch": 12.359432799013563,
+      "grad_norm": 0.03716684505343437,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 120282
+    },
+    {
+      "epoch": 12.359741060419235,
+      "grad_norm": 0.039309579879045486,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 120285
+    },
+    {
+      "epoch": 12.360049321824908,
+      "grad_norm": 0.05379367247223854,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 120288
+    },
+    {
+      "epoch": 12.360357583230579,
+      "grad_norm": 0.040921930223703384,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 120291
+    },
+    {
+      "epoch": 12.360665844636252,
+      "grad_norm": 0.09452662616968155,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 120294
+    },
+    {
+      "epoch": 12.360974106041924,
+      "grad_norm": 0.0810219794511795,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 120297
+    },
+    {
+      "epoch": 12.361282367447595,
+      "grad_norm": 0.10484972596168518,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 120300
+    },
+    {
+      "epoch": 12.361590628853268,
+      "grad_norm": 0.08036096394062042,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 120303
+    },
+    {
+      "epoch": 12.36189889025894,
+      "grad_norm": 0.0392296239733696,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 120306
+    },
+    {
+      "epoch": 12.362207151664611,
+      "grad_norm": 0.040033504366874695,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 120309
+    },
+    {
+      "epoch": 12.362515413070284,
+      "grad_norm": 0.045406442135572433,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 120312
+    },
+    {
+      "epoch": 12.362823674475955,
+      "grad_norm": 0.09677822142839432,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 120315
+    },
+    {
+      "epoch": 12.363131935881627,
+      "grad_norm": 0.12022221088409424,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 120318
+    },
+    {
+      "epoch": 12.3634401972873,
+      "grad_norm": 0.10691323131322861,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 120321
+    },
+    {
+      "epoch": 12.363748458692971,
+      "grad_norm": 0.10206713527441025,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 120324
+    },
+    {
+      "epoch": 12.364056720098644,
+      "grad_norm": 0.12835240364074707,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 120327
+    },
+    {
+      "epoch": 12.364364981504316,
+      "grad_norm": 0.05794866755604744,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 120330
+    },
+    {
+      "epoch": 12.364673242909987,
+      "grad_norm": 0.0687018483877182,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 120333
+    },
+    {
+      "epoch": 12.36498150431566,
+      "grad_norm": 0.04902690649032593,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 120336
+    },
+    {
+      "epoch": 12.365289765721332,
+      "grad_norm": 0.03381572291254997,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 120339
+    },
+    {
+      "epoch": 12.365598027127003,
+      "grad_norm": 0.04516056180000305,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 120342
+    },
+    {
+      "epoch": 12.365906288532676,
+      "grad_norm": 0.05366626754403114,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 120345
+    },
+    {
+      "epoch": 12.366214549938348,
+      "grad_norm": 0.035903528332710266,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 120348
+    },
+    {
+      "epoch": 12.36652281134402,
+      "grad_norm": 0.06833939999341965,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 120351
+    },
+    {
+      "epoch": 12.366831072749692,
+      "grad_norm": 0.05716986209154129,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 120354
+    },
+    {
+      "epoch": 12.367139334155365,
+      "grad_norm": 0.15210434794425964,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 120357
+    },
+    {
+      "epoch": 12.367447595561035,
+      "grad_norm": 0.08575952798128128,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 120360
+    },
+    {
+      "epoch": 12.367755856966708,
+      "grad_norm": 0.03106115758419037,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 120363
+    },
+    {
+      "epoch": 12.368064118372379,
+      "grad_norm": 0.0336822085082531,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 120366
+    },
+    {
+      "epoch": 12.368372379778052,
+      "grad_norm": 0.09181510657072067,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 120369
+    },
+    {
+      "epoch": 12.368680641183724,
+      "grad_norm": 0.06922073662281036,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 120372
+    },
+    {
+      "epoch": 12.368988902589395,
+      "grad_norm": 0.11679048091173172,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 120375
+    },
+    {
+      "epoch": 12.369297163995068,
+      "grad_norm": 0.06578762084245682,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 120378
+    },
+    {
+      "epoch": 12.36960542540074,
+      "grad_norm": 0.11222466826438904,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 120381
+    },
+    {
+      "epoch": 12.369913686806411,
+      "grad_norm": 0.06960795074701309,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 120384
+    },
+    {
+      "epoch": 12.370221948212084,
+      "grad_norm": 0.05938812717795372,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 120387
+    },
+    {
+      "epoch": 12.370530209617757,
+      "grad_norm": 0.04438195377588272,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 120390
+    },
+    {
+      "epoch": 12.370838471023427,
+      "grad_norm": 0.07959045469760895,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 120393
+    },
+    {
+      "epoch": 12.3711467324291,
+      "grad_norm": 0.0454067625105381,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 120396
+    },
+    {
+      "epoch": 12.371454993834773,
+      "grad_norm": 0.07425530254840851,
+      "learning_rate": 0.01,
+      "loss": 1.8974,
+      "step": 120399
+    },
+    {
+      "epoch": 12.371763255240444,
+      "grad_norm": 0.03199988231062889,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 120402
+    },
+    {
+      "epoch": 12.372071516646116,
+      "grad_norm": 0.10832170397043228,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 120405
+    },
+    {
+      "epoch": 12.372379778051787,
+      "grad_norm": 0.08807660639286041,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 120408
+    },
+    {
+      "epoch": 12.37268803945746,
+      "grad_norm": 0.10143882036209106,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 120411
+    },
+    {
+      "epoch": 12.372996300863132,
+      "grad_norm": 0.06982599198818207,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 120414
+    },
+    {
+      "epoch": 12.373304562268803,
+      "grad_norm": 0.03442126512527466,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 120417
+    },
+    {
+      "epoch": 12.373612823674476,
+      "grad_norm": 0.09746600687503815,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 120420
+    },
+    {
+      "epoch": 12.373921085080148,
+      "grad_norm": 0.11646406352519989,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 120423
+    },
+    {
+      "epoch": 12.37422934648582,
+      "grad_norm": 0.05813951790332794,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 120426
+    },
+    {
+      "epoch": 12.374537607891492,
+      "grad_norm": 0.11898761242628098,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 120429
+    },
+    {
+      "epoch": 12.374845869297165,
+      "grad_norm": 0.10534807294607162,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 120432
+    },
+    {
+      "epoch": 12.375154130702835,
+      "grad_norm": 0.06432823836803436,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 120435
+    },
+    {
+      "epoch": 12.375462392108508,
+      "grad_norm": 0.06572303920984268,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 120438
+    },
+    {
+      "epoch": 12.37577065351418,
+      "grad_norm": 0.0753687247633934,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 120441
+    },
+    {
+      "epoch": 12.376078914919852,
+      "grad_norm": 0.06693238765001297,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 120444
+    },
+    {
+      "epoch": 12.376387176325524,
+      "grad_norm": 0.034933801740407944,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 120447
+    },
+    {
+      "epoch": 12.376695437731197,
+      "grad_norm": 0.04015972092747688,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 120450
+    },
+    {
+      "epoch": 12.377003699136868,
+      "grad_norm": 0.039467908442020416,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 120453
+    },
+    {
+      "epoch": 12.37731196054254,
+      "grad_norm": 0.055875346064567566,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 120456
+    },
+    {
+      "epoch": 12.377620221948213,
+      "grad_norm": 0.10872425884008408,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 120459
+    },
+    {
+      "epoch": 12.377928483353884,
+      "grad_norm": 0.13861580193042755,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 120462
+    },
+    {
+      "epoch": 12.378236744759556,
+      "grad_norm": 0.07626169919967651,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 120465
+    },
+    {
+      "epoch": 12.378545006165227,
+      "grad_norm": 0.03249692916870117,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 120468
+    },
+    {
+      "epoch": 12.3788532675709,
+      "grad_norm": 0.042429469525814056,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 120471
+    },
+    {
+      "epoch": 12.379161528976573,
+      "grad_norm": 0.06338413804769516,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 120474
+    },
+    {
+      "epoch": 12.379469790382243,
+      "grad_norm": 0.0565384179353714,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 120477
+    },
+    {
+      "epoch": 12.379778051787916,
+      "grad_norm": 0.04637407884001732,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 120480
+    },
+    {
+      "epoch": 12.380086313193589,
+      "grad_norm": 0.12896063923835754,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 120483
+    },
+    {
+      "epoch": 12.38039457459926,
+      "grad_norm": 0.08389844745397568,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 120486
+    },
+    {
+      "epoch": 12.380702836004932,
+      "grad_norm": 0.06900922954082489,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 120489
+    },
+    {
+      "epoch": 12.381011097410605,
+      "grad_norm": 0.05321461334824562,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 120492
+    },
+    {
+      "epoch": 12.381319358816276,
+      "grad_norm": 0.04594092443585396,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 120495
+    },
+    {
+      "epoch": 12.381627620221948,
+      "grad_norm": 0.05090029165148735,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 120498
+    },
+    {
+      "epoch": 12.381935881627621,
+      "grad_norm": 0.14940975606441498,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 120501
+    },
+    {
+      "epoch": 12.382244143033292,
+      "grad_norm": 0.12164038419723511,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 120504
+    },
+    {
+      "epoch": 12.382552404438965,
+      "grad_norm": 0.08605052530765533,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 120507
+    },
+    {
+      "epoch": 12.382860665844635,
+      "grad_norm": 0.07495385408401489,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 120510
+    },
+    {
+      "epoch": 12.383168927250308,
+      "grad_norm": 0.03808677941560745,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 120513
+    },
+    {
+      "epoch": 12.38347718865598,
+      "grad_norm": 0.0872986689209938,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 120516
+    },
+    {
+      "epoch": 12.383785450061652,
+      "grad_norm": 0.07244990020990372,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 120519
+    },
+    {
+      "epoch": 12.384093711467324,
+      "grad_norm": 0.07292909175157547,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 120522
+    },
+    {
+      "epoch": 12.384401972872997,
+      "grad_norm": 0.05182334780693054,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 120525
+    },
+    {
+      "epoch": 12.384710234278668,
+      "grad_norm": 0.046150092035532,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 120528
+    },
+    {
+      "epoch": 12.38501849568434,
+      "grad_norm": 0.037887394428253174,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 120531
+    },
+    {
+      "epoch": 12.385326757090013,
+      "grad_norm": 0.03401385247707367,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 120534
+    },
+    {
+      "epoch": 12.385635018495684,
+      "grad_norm": 0.04379456117749214,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 120537
+    },
+    {
+      "epoch": 12.385943279901356,
+      "grad_norm": 0.1705673784017563,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 120540
+    },
+    {
+      "epoch": 12.386251541307029,
+      "grad_norm": 0.054921045899391174,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 120543
+    },
+    {
+      "epoch": 12.3865598027127,
+      "grad_norm": 0.06597351282835007,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 120546
+    },
+    {
+      "epoch": 12.386868064118373,
+      "grad_norm": 0.11311409622430801,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 120549
+    },
+    {
+      "epoch": 12.387176325524045,
+      "grad_norm": 0.044089119881391525,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 120552
+    },
+    {
+      "epoch": 12.387484586929716,
+      "grad_norm": 0.03425237908959389,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 120555
+    },
+    {
+      "epoch": 12.387792848335389,
+      "grad_norm": 0.03652380406856537,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 120558
+    },
+    {
+      "epoch": 12.38810110974106,
+      "grad_norm": 0.04659121856093407,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 120561
+    },
+    {
+      "epoch": 12.388409371146732,
+      "grad_norm": 0.10116490721702576,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 120564
+    },
+    {
+      "epoch": 12.388717632552405,
+      "grad_norm": 0.06742721050977707,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 120567
+    },
+    {
+      "epoch": 12.389025893958076,
+      "grad_norm": 0.05432113632559776,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 120570
+    },
+    {
+      "epoch": 12.389334155363748,
+      "grad_norm": 0.050973620265722275,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 120573
+    },
+    {
+      "epoch": 12.389642416769421,
+      "grad_norm": 0.048542775213718414,
+      "learning_rate": 0.01,
+      "loss": 1.9617,
+      "step": 120576
+    },
+    {
+      "epoch": 12.389950678175092,
+      "grad_norm": 0.03432178869843483,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 120579
+    },
+    {
+      "epoch": 12.390258939580765,
+      "grad_norm": 0.06701638549566269,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 120582
+    },
+    {
+      "epoch": 12.390567200986437,
+      "grad_norm": 0.0790436640381813,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 120585
+    },
+    {
+      "epoch": 12.390875462392108,
+      "grad_norm": 0.0833590179681778,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 120588
+    },
+    {
+      "epoch": 12.39118372379778,
+      "grad_norm": 0.05441255122423172,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 120591
+    },
+    {
+      "epoch": 12.391491985203453,
+      "grad_norm": 0.03294618800282478,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 120594
+    },
+    {
+      "epoch": 12.391800246609124,
+      "grad_norm": 0.03458996117115021,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 120597
+    },
+    {
+      "epoch": 12.392108508014797,
+      "grad_norm": 0.043681446462869644,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 120600
+    },
+    {
+      "epoch": 12.392416769420468,
+      "grad_norm": 0.07503581792116165,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 120603
+    },
+    {
+      "epoch": 12.39272503082614,
+      "grad_norm": 0.0830334946513176,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 120606
+    },
+    {
+      "epoch": 12.393033292231813,
+      "grad_norm": 0.10886512696743011,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 120609
+    },
+    {
+      "epoch": 12.393341553637484,
+      "grad_norm": 0.08003956079483032,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 120612
+    },
+    {
+      "epoch": 12.393649815043156,
+      "grad_norm": 0.06517963856458664,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 120615
+    },
+    {
+      "epoch": 12.393958076448829,
+      "grad_norm": 0.037115540355443954,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 120618
+    },
+    {
+      "epoch": 12.3942663378545,
+      "grad_norm": 0.10049480944871902,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 120621
+    },
+    {
+      "epoch": 12.394574599260173,
+      "grad_norm": 0.052566614001989365,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 120624
+    },
+    {
+      "epoch": 12.394882860665845,
+      "grad_norm": 0.038146406412124634,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 120627
+    },
+    {
+      "epoch": 12.395191122071516,
+      "grad_norm": 0.05074608325958252,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 120630
+    },
+    {
+      "epoch": 12.395499383477189,
+      "grad_norm": 0.04188331216573715,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 120633
+    },
+    {
+      "epoch": 12.395807644882861,
+      "grad_norm": 0.04295455291867256,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 120636
+    },
+    {
+      "epoch": 12.396115906288532,
+      "grad_norm": 0.036794669926166534,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 120639
+    },
+    {
+      "epoch": 12.396424167694205,
+      "grad_norm": 0.07238873839378357,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 120642
+    },
+    {
+      "epoch": 12.396732429099877,
+      "grad_norm": 0.11191675066947937,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 120645
+    },
+    {
+      "epoch": 12.397040690505548,
+      "grad_norm": 0.09227670729160309,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 120648
+    },
+    {
+      "epoch": 12.397348951911221,
+      "grad_norm": 0.0327807180583477,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 120651
+    },
+    {
+      "epoch": 12.397657213316894,
+      "grad_norm": 0.07567586749792099,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 120654
+    },
+    {
+      "epoch": 12.397965474722564,
+      "grad_norm": 0.07382260262966156,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 120657
+    },
+    {
+      "epoch": 12.398273736128237,
+      "grad_norm": 0.13166159391403198,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 120660
+    },
+    {
+      "epoch": 12.398581997533908,
+      "grad_norm": 0.05870798975229263,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 120663
+    },
+    {
+      "epoch": 12.39889025893958,
+      "grad_norm": 0.05218498781323433,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 120666
+    },
+    {
+      "epoch": 12.399198520345253,
+      "grad_norm": 0.06324154138565063,
+      "learning_rate": 0.01,
+      "loss": 1.962,
+      "step": 120669
+    },
+    {
+      "epoch": 12.399506781750924,
+      "grad_norm": 0.0561276376247406,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 120672
+    },
+    {
+      "epoch": 12.399815043156597,
+      "grad_norm": 0.03983089327812195,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 120675
+    },
+    {
+      "epoch": 12.40012330456227,
+      "grad_norm": 0.058458730578422546,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 120678
+    },
+    {
+      "epoch": 12.40043156596794,
+      "grad_norm": 0.10894698649644852,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 120681
+    },
+    {
+      "epoch": 12.400739827373613,
+      "grad_norm": 0.06287480145692825,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 120684
+    },
+    {
+      "epoch": 12.401048088779286,
+      "grad_norm": 0.06695500016212463,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 120687
+    },
+    {
+      "epoch": 12.401356350184956,
+      "grad_norm": 0.04819144681096077,
+      "learning_rate": 0.01,
+      "loss": 1.9635,
+      "step": 120690
+    },
+    {
+      "epoch": 12.401664611590629,
+      "grad_norm": 0.11953316628932953,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 120693
+    },
+    {
+      "epoch": 12.401972872996302,
+      "grad_norm": 0.08536428213119507,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 120696
+    },
+    {
+      "epoch": 12.402281134401973,
+      "grad_norm": 0.051179662346839905,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 120699
+    },
+    {
+      "epoch": 12.402589395807645,
+      "grad_norm": 0.03457881137728691,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 120702
+    },
+    {
+      "epoch": 12.402897657213316,
+      "grad_norm": 0.04182450473308563,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 120705
+    },
+    {
+      "epoch": 12.403205918618989,
+      "grad_norm": 0.08560032397508621,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 120708
+    },
+    {
+      "epoch": 12.403514180024661,
+      "grad_norm": 0.08513845503330231,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 120711
+    },
+    {
+      "epoch": 12.403822441430332,
+      "grad_norm": 0.04573316499590874,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 120714
+    },
+    {
+      "epoch": 12.404130702836005,
+      "grad_norm": 0.04893175885081291,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 120717
+    },
+    {
+      "epoch": 12.404438964241677,
+      "grad_norm": 0.08831377327442169,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 120720
+    },
+    {
+      "epoch": 12.404747225647348,
+      "grad_norm": 0.0819849893450737,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 120723
+    },
+    {
+      "epoch": 12.405055487053021,
+      "grad_norm": 0.04886121675372124,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 120726
+    },
+    {
+      "epoch": 12.405363748458694,
+      "grad_norm": 0.1050790548324585,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 120729
+    },
+    {
+      "epoch": 12.405672009864364,
+      "grad_norm": 0.06852862238883972,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 120732
+    },
+    {
+      "epoch": 12.405980271270037,
+      "grad_norm": 0.09256117045879364,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 120735
+    },
+    {
+      "epoch": 12.40628853267571,
+      "grad_norm": 0.051488909870386124,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 120738
+    },
+    {
+      "epoch": 12.40659679408138,
+      "grad_norm": 0.07905134558677673,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 120741
+    },
+    {
+      "epoch": 12.406905055487053,
+      "grad_norm": 0.08850820362567902,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 120744
+    },
+    {
+      "epoch": 12.407213316892726,
+      "grad_norm": 0.05323689058423042,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 120747
+    },
+    {
+      "epoch": 12.407521578298397,
+      "grad_norm": 0.13145367801189423,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 120750
+    },
+    {
+      "epoch": 12.40782983970407,
+      "grad_norm": 0.0645250603556633,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 120753
+    },
+    {
+      "epoch": 12.40813810110974,
+      "grad_norm": 0.05831822380423546,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 120756
+    },
+    {
+      "epoch": 12.408446362515413,
+      "grad_norm": 0.058313801884651184,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 120759
+    },
+    {
+      "epoch": 12.408754623921086,
+      "grad_norm": 0.0371122881770134,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 120762
+    },
+    {
+      "epoch": 12.409062885326756,
+      "grad_norm": 0.04355745390057564,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 120765
+    },
+    {
+      "epoch": 12.409371146732429,
+      "grad_norm": 0.05011705681681633,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 120768
+    },
+    {
+      "epoch": 12.409679408138102,
+      "grad_norm": 0.13353750109672546,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 120771
+    },
+    {
+      "epoch": 12.409987669543773,
+      "grad_norm": 0.09305182099342346,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 120774
+    },
+    {
+      "epoch": 12.410295930949445,
+      "grad_norm": 0.05511609837412834,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 120777
+    },
+    {
+      "epoch": 12.410604192355118,
+      "grad_norm": 0.05445779114961624,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 120780
+    },
+    {
+      "epoch": 12.410912453760789,
+      "grad_norm": 0.08220420032739639,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 120783
+    },
+    {
+      "epoch": 12.411220715166461,
+      "grad_norm": 0.0428672656416893,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 120786
+    },
+    {
+      "epoch": 12.411528976572134,
+      "grad_norm": 0.03976845368742943,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 120789
+    },
+    {
+      "epoch": 12.411837237977805,
+      "grad_norm": 0.0461377277970314,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 120792
+    },
+    {
+      "epoch": 12.412145499383477,
+      "grad_norm": 0.05648530274629593,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 120795
+    },
+    {
+      "epoch": 12.412453760789148,
+      "grad_norm": 0.06467310339212418,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 120798
+    },
+    {
+      "epoch": 12.412762022194821,
+      "grad_norm": 0.03678126260638237,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 120801
+    },
+    {
+      "epoch": 12.413070283600494,
+      "grad_norm": 0.13990208506584167,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 120804
+    },
+    {
+      "epoch": 12.413378545006164,
+      "grad_norm": 0.04235214367508888,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 120807
+    },
+    {
+      "epoch": 12.413686806411837,
+      "grad_norm": 0.04935699328780174,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 120810
+    },
+    {
+      "epoch": 12.41399506781751,
+      "grad_norm": 0.0406753309071064,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 120813
+    },
+    {
+      "epoch": 12.41430332922318,
+      "grad_norm": 0.03786905109882355,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 120816
+    },
+    {
+      "epoch": 12.414611590628853,
+      "grad_norm": 0.05138446018099785,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 120819
+    },
+    {
+      "epoch": 12.414919852034526,
+      "grad_norm": 0.09492046386003494,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 120822
+    },
+    {
+      "epoch": 12.415228113440197,
+      "grad_norm": 0.03346167132258415,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 120825
+    },
+    {
+      "epoch": 12.41553637484587,
+      "grad_norm": 0.07686081528663635,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 120828
+    },
+    {
+      "epoch": 12.415844636251542,
+      "grad_norm": 0.11712466180324554,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 120831
+    },
+    {
+      "epoch": 12.416152897657213,
+      "grad_norm": 0.047113195061683655,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 120834
+    },
+    {
+      "epoch": 12.416461159062885,
+      "grad_norm": 0.045871369540691376,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 120837
+    },
+    {
+      "epoch": 12.416769420468558,
+      "grad_norm": 0.04527949541807175,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 120840
+    },
+    {
+      "epoch": 12.417077681874229,
+      "grad_norm": 0.03895549476146698,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 120843
+    },
+    {
+      "epoch": 12.417385943279902,
+      "grad_norm": 0.0926399901509285,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 120846
+    },
+    {
+      "epoch": 12.417694204685574,
+      "grad_norm": 0.05789586901664734,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 120849
+    },
+    {
+      "epoch": 12.418002466091245,
+      "grad_norm": 0.0683421865105629,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 120852
+    },
+    {
+      "epoch": 12.418310727496918,
+      "grad_norm": 0.041859518736600876,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 120855
+    },
+    {
+      "epoch": 12.418618988902589,
+      "grad_norm": 0.04999043047428131,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 120858
+    },
+    {
+      "epoch": 12.418927250308261,
+      "grad_norm": 0.04389044642448425,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 120861
+    },
+    {
+      "epoch": 12.419235511713934,
+      "grad_norm": 0.04234243184328079,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 120864
+    },
+    {
+      "epoch": 12.419543773119605,
+      "grad_norm": 0.03945605084300041,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 120867
+    },
+    {
+      "epoch": 12.419852034525277,
+      "grad_norm": 0.049900732934474945,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 120870
+    },
+    {
+      "epoch": 12.42016029593095,
+      "grad_norm": 0.10874069482088089,
+      "learning_rate": 0.01,
+      "loss": 1.9685,
+      "step": 120873
+    },
+    {
+      "epoch": 12.420468557336621,
+      "grad_norm": 0.044814638793468475,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 120876
+    },
+    {
+      "epoch": 12.420776818742294,
+      "grad_norm": 0.08993617445230484,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 120879
+    },
+    {
+      "epoch": 12.421085080147966,
+      "grad_norm": 0.04525793343782425,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 120882
+    },
+    {
+      "epoch": 12.421393341553637,
+      "grad_norm": 0.054966967552900314,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 120885
+    },
+    {
+      "epoch": 12.42170160295931,
+      "grad_norm": 0.08710785955190659,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 120888
+    },
+    {
+      "epoch": 12.422009864364982,
+      "grad_norm": 0.09183228760957718,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 120891
+    },
+    {
+      "epoch": 12.422318125770653,
+      "grad_norm": 0.08628053963184357,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 120894
+    },
+    {
+      "epoch": 12.422626387176326,
+      "grad_norm": 0.09337654709815979,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 120897
+    },
+    {
+      "epoch": 12.422934648581997,
+      "grad_norm": 0.047991879284381866,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 120900
+    },
+    {
+      "epoch": 12.42324290998767,
+      "grad_norm": 0.11279120296239853,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 120903
+    },
+    {
+      "epoch": 12.423551171393342,
+      "grad_norm": 0.08573783189058304,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 120906
+    },
+    {
+      "epoch": 12.423859432799013,
+      "grad_norm": 0.05766313523054123,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 120909
+    },
+    {
+      "epoch": 12.424167694204685,
+      "grad_norm": 0.07933402061462402,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 120912
+    },
+    {
+      "epoch": 12.424475955610358,
+      "grad_norm": 0.03687792271375656,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 120915
+    },
+    {
+      "epoch": 12.424784217016029,
+      "grad_norm": 0.05140282213687897,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 120918
+    },
+    {
+      "epoch": 12.425092478421702,
+      "grad_norm": 0.07529986649751663,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 120921
+    },
+    {
+      "epoch": 12.425400739827374,
+      "grad_norm": 0.06762083619832993,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 120924
+    },
+    {
+      "epoch": 12.425709001233045,
+      "grad_norm": 0.12069932371377945,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 120927
+    },
+    {
+      "epoch": 12.426017262638718,
+      "grad_norm": 0.055894885212183,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 120930
+    },
+    {
+      "epoch": 12.42632552404439,
+      "grad_norm": 0.10258354991674423,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 120933
+    },
+    {
+      "epoch": 12.426633785450061,
+      "grad_norm": 0.058992642909288406,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 120936
+    },
+    {
+      "epoch": 12.426942046855734,
+      "grad_norm": 0.047088347375392914,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 120939
+    },
+    {
+      "epoch": 12.427250308261407,
+      "grad_norm": 0.07983656227588654,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 120942
+    },
+    {
+      "epoch": 12.427558569667077,
+      "grad_norm": 0.11896104365587234,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 120945
+    },
+    {
+      "epoch": 12.42786683107275,
+      "grad_norm": 0.13242436945438385,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 120948
+    },
+    {
+      "epoch": 12.428175092478421,
+      "grad_norm": 0.06254076957702637,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 120951
+    },
+    {
+      "epoch": 12.428483353884094,
+      "grad_norm": 0.044072654098272324,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 120954
+    },
+    {
+      "epoch": 12.428791615289766,
+      "grad_norm": 0.05199466273188591,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 120957
+    },
+    {
+      "epoch": 12.429099876695437,
+      "grad_norm": 0.04316157475113869,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 120960
+    },
+    {
+      "epoch": 12.42940813810111,
+      "grad_norm": 0.11193782091140747,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 120963
+    },
+    {
+      "epoch": 12.429716399506782,
+      "grad_norm": 0.05210939422249794,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 120966
+    },
+    {
+      "epoch": 12.430024660912453,
+      "grad_norm": 0.08012048900127411,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 120969
+    },
+    {
+      "epoch": 12.430332922318126,
+      "grad_norm": 0.06603696197271347,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 120972
+    },
+    {
+      "epoch": 12.430641183723798,
+      "grad_norm": 0.07295702397823334,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 120975
+    },
+    {
+      "epoch": 12.43094944512947,
+      "grad_norm": 0.040298521518707275,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 120978
+    },
+    {
+      "epoch": 12.431257706535142,
+      "grad_norm": 0.0881817564368248,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 120981
+    },
+    {
+      "epoch": 12.431565967940815,
+      "grad_norm": 0.06474234908819199,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 120984
+    },
+    {
+      "epoch": 12.431874229346485,
+      "grad_norm": 0.07105148583650589,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 120987
+    },
+    {
+      "epoch": 12.432182490752158,
+      "grad_norm": 0.09649461507797241,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 120990
+    },
+    {
+      "epoch": 12.432490752157829,
+      "grad_norm": 0.15850776433944702,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 120993
+    },
+    {
+      "epoch": 12.432799013563502,
+      "grad_norm": 0.10326742380857468,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 120996
+    },
+    {
+      "epoch": 12.433107274969174,
+      "grad_norm": 0.06529439985752106,
+      "learning_rate": 0.01,
+      "loss": 1.9572,
+      "step": 120999
+    },
+    {
+      "epoch": 12.433415536374845,
+      "grad_norm": 0.08048941195011139,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 121002
+    },
+    {
+      "epoch": 12.433723797780518,
+      "grad_norm": 0.09297440946102142,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 121005
+    },
+    {
+      "epoch": 12.43403205918619,
+      "grad_norm": 0.0965491309762001,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 121008
+    },
+    {
+      "epoch": 12.434340320591861,
+      "grad_norm": 0.08597447723150253,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 121011
+    },
+    {
+      "epoch": 12.434648581997534,
+      "grad_norm": 0.050102099776268005,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 121014
+    },
+    {
+      "epoch": 12.434956843403207,
+      "grad_norm": 0.035592518746852875,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 121017
+    },
+    {
+      "epoch": 12.435265104808877,
+      "grad_norm": 0.09271007031202316,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 121020
+    },
+    {
+      "epoch": 12.43557336621455,
+      "grad_norm": 0.04439222440123558,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 121023
+    },
+    {
+      "epoch": 12.435881627620223,
+      "grad_norm": 0.03583432734012604,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 121026
+    },
+    {
+      "epoch": 12.436189889025894,
+      "grad_norm": 0.04167827591300011,
+      "learning_rate": 0.01,
+      "loss": 1.9009,
+      "step": 121029
+    },
+    {
+      "epoch": 12.436498150431566,
+      "grad_norm": 0.12448694556951523,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 121032
+    },
+    {
+      "epoch": 12.436806411837239,
+      "grad_norm": 0.07956825941801071,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 121035
+    },
+    {
+      "epoch": 12.43711467324291,
+      "grad_norm": 0.10021093487739563,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 121038
+    },
+    {
+      "epoch": 12.437422934648582,
+      "grad_norm": 0.05917084962129593,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 121041
+    },
+    {
+      "epoch": 12.437731196054255,
+      "grad_norm": 0.08545676618814468,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 121044
+    },
+    {
+      "epoch": 12.438039457459926,
+      "grad_norm": 0.13080301880836487,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 121047
+    },
+    {
+      "epoch": 12.438347718865598,
+      "grad_norm": 0.05625632032752037,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 121050
+    },
+    {
+      "epoch": 12.43865598027127,
+      "grad_norm": 0.0860651507973671,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 121053
+    },
+    {
+      "epoch": 12.438964241676942,
+      "grad_norm": 0.047987114638090134,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 121056
+    },
+    {
+      "epoch": 12.439272503082615,
+      "grad_norm": 0.09193069487810135,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 121059
+    },
+    {
+      "epoch": 12.439580764488285,
+      "grad_norm": 0.07548029720783234,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 121062
+    },
+    {
+      "epoch": 12.439889025893958,
+      "grad_norm": 0.09455464035272598,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 121065
+    },
+    {
+      "epoch": 12.44019728729963,
+      "grad_norm": 0.07177861779928207,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 121068
+    },
+    {
+      "epoch": 12.440505548705302,
+      "grad_norm": 0.049412354826927185,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 121071
+    },
+    {
+      "epoch": 12.440813810110974,
+      "grad_norm": 0.09546947479248047,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 121074
+    },
+    {
+      "epoch": 12.441122071516647,
+      "grad_norm": 0.046247322112321854,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 121077
+    },
+    {
+      "epoch": 12.441430332922318,
+      "grad_norm": 0.09783945232629776,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 121080
+    },
+    {
+      "epoch": 12.44173859432799,
+      "grad_norm": 0.09536667913198471,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 121083
+    },
+    {
+      "epoch": 12.442046855733663,
+      "grad_norm": 0.0748727098107338,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 121086
+    },
+    {
+      "epoch": 12.442355117139334,
+      "grad_norm": 0.04721232131123543,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 121089
+    },
+    {
+      "epoch": 12.442663378545006,
+      "grad_norm": 0.05681651458144188,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 121092
+    },
+    {
+      "epoch": 12.442971639950677,
+      "grad_norm": 0.0716468095779419,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 121095
+    },
+    {
+      "epoch": 12.44327990135635,
+      "grad_norm": 0.07019370794296265,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 121098
+    },
+    {
+      "epoch": 12.443588162762023,
+      "grad_norm": 0.06477268040180206,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 121101
+    },
+    {
+      "epoch": 12.443896424167693,
+      "grad_norm": 0.09084682911634445,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 121104
+    },
+    {
+      "epoch": 12.444204685573366,
+      "grad_norm": 0.0959085151553154,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 121107
+    },
+    {
+      "epoch": 12.444512946979039,
+      "grad_norm": 0.07635615020990372,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 121110
+    },
+    {
+      "epoch": 12.44482120838471,
+      "grad_norm": 0.13731513917446136,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 121113
+    },
+    {
+      "epoch": 12.445129469790382,
+      "grad_norm": 0.13287904858589172,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 121116
+    },
+    {
+      "epoch": 12.445437731196055,
+      "grad_norm": 0.17962051928043365,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 121119
+    },
+    {
+      "epoch": 12.445745992601726,
+      "grad_norm": 0.04233860969543457,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 121122
+    },
+    {
+      "epoch": 12.446054254007398,
+      "grad_norm": 0.03985730558633804,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 121125
+    },
+    {
+      "epoch": 12.446362515413071,
+      "grad_norm": 0.04166532680392265,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 121128
+    },
+    {
+      "epoch": 12.446670776818742,
+      "grad_norm": 0.042059239000082016,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 121131
+    },
+    {
+      "epoch": 12.446979038224415,
+      "grad_norm": 0.036940161138772964,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 121134
+    },
+    {
+      "epoch": 12.447287299630087,
+      "grad_norm": 0.06131014600396156,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 121137
+    },
+    {
+      "epoch": 12.447595561035758,
+      "grad_norm": 0.07934664934873581,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 121140
+    },
+    {
+      "epoch": 12.44790382244143,
+      "grad_norm": 0.14649051427841187,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 121143
+    },
+    {
+      "epoch": 12.448212083847102,
+      "grad_norm": 0.0989823117852211,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 121146
+    },
+    {
+      "epoch": 12.448520345252774,
+      "grad_norm": 0.05163063108921051,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 121149
+    },
+    {
+      "epoch": 12.448828606658447,
+      "grad_norm": 0.054580073803663254,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 121152
+    },
+    {
+      "epoch": 12.449136868064118,
+      "grad_norm": 0.03554552420973778,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 121155
+    },
+    {
+      "epoch": 12.44944512946979,
+      "grad_norm": 0.10882104188203812,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 121158
+    },
+    {
+      "epoch": 12.449753390875463,
+      "grad_norm": 0.08104122430086136,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 121161
+    },
+    {
+      "epoch": 12.450061652281134,
+      "grad_norm": 0.055625855922698975,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 121164
+    },
+    {
+      "epoch": 12.450369913686806,
+      "grad_norm": 0.09399867802858353,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 121167
+    },
+    {
+      "epoch": 12.450678175092479,
+      "grad_norm": 0.04432365298271179,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 121170
+    },
+    {
+      "epoch": 12.45098643649815,
+      "grad_norm": 0.043413758277893066,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 121173
+    },
+    {
+      "epoch": 12.451294697903823,
+      "grad_norm": 0.03953053429722786,
+      "learning_rate": 0.01,
+      "loss": 1.8988,
+      "step": 121176
+    },
+    {
+      "epoch": 12.451602959309495,
+      "grad_norm": 0.1084480807185173,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 121179
+    },
+    {
+      "epoch": 12.451911220715166,
+      "grad_norm": 0.047949500381946564,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 121182
+    },
+    {
+      "epoch": 12.452219482120839,
+      "grad_norm": 0.05586003139615059,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 121185
+    },
+    {
+      "epoch": 12.45252774352651,
+      "grad_norm": 0.09080043435096741,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 121188
+    },
+    {
+      "epoch": 12.452836004932182,
+      "grad_norm": 0.06499528139829636,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 121191
+    },
+    {
+      "epoch": 12.453144266337855,
+      "grad_norm": 0.0984288677573204,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 121194
+    },
+    {
+      "epoch": 12.453452527743526,
+      "grad_norm": 0.045159339904785156,
+      "learning_rate": 0.01,
+      "loss": 1.9546,
+      "step": 121197
+    },
+    {
+      "epoch": 12.453760789149198,
+      "grad_norm": 0.0337919145822525,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 121200
+    },
+    {
+      "epoch": 12.454069050554871,
+      "grad_norm": 0.039271969348192215,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 121203
+    },
+    {
+      "epoch": 12.454377311960542,
+      "grad_norm": 0.049345336854457855,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 121206
+    },
+    {
+      "epoch": 12.454685573366215,
+      "grad_norm": 0.13334542512893677,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 121209
+    },
+    {
+      "epoch": 12.454993834771887,
+      "grad_norm": 0.11365757882595062,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 121212
+    },
+    {
+      "epoch": 12.455302096177558,
+      "grad_norm": 0.055066999047994614,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 121215
+    },
+    {
+      "epoch": 12.45561035758323,
+      "grad_norm": 0.04881853982806206,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 121218
+    },
+    {
+      "epoch": 12.455918618988903,
+      "grad_norm": 0.1275508552789688,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 121221
+    },
+    {
+      "epoch": 12.456226880394574,
+      "grad_norm": 0.042988065630197525,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 121224
+    },
+    {
+      "epoch": 12.456535141800247,
+      "grad_norm": 0.11997077614068985,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 121227
+    },
+    {
+      "epoch": 12.45684340320592,
+      "grad_norm": 0.04036852344870567,
+      "learning_rate": 0.01,
+      "loss": 1.9686,
+      "step": 121230
+    },
+    {
+      "epoch": 12.45715166461159,
+      "grad_norm": 0.08224493265151978,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 121233
+    },
+    {
+      "epoch": 12.457459926017263,
+      "grad_norm": 0.0855056419968605,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 121236
+    },
+    {
+      "epoch": 12.457768187422934,
+      "grad_norm": 0.07086644321680069,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 121239
+    },
+    {
+      "epoch": 12.458076448828606,
+      "grad_norm": 0.0619567334651947,
+      "learning_rate": 0.01,
+      "loss": 1.9707,
+      "step": 121242
+    },
+    {
+      "epoch": 12.458384710234279,
+      "grad_norm": 0.08348061144351959,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 121245
+    },
+    {
+      "epoch": 12.45869297163995,
+      "grad_norm": 0.04736310988664627,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 121248
+    },
+    {
+      "epoch": 12.459001233045623,
+      "grad_norm": 0.12289008498191833,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 121251
+    },
+    {
+      "epoch": 12.459309494451295,
+      "grad_norm": 0.061683282256126404,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 121254
+    },
+    {
+      "epoch": 12.459617755856966,
+      "grad_norm": 0.06045114994049072,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 121257
+    },
+    {
+      "epoch": 12.459926017262639,
+      "grad_norm": 0.042001038789749146,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 121260
+    },
+    {
+      "epoch": 12.460234278668311,
+      "grad_norm": 0.04718760401010513,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 121263
+    },
+    {
+      "epoch": 12.460542540073982,
+      "grad_norm": 0.0723651647567749,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 121266
+    },
+    {
+      "epoch": 12.460850801479655,
+      "grad_norm": 0.05260621756315231,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 121269
+    },
+    {
+      "epoch": 12.461159062885327,
+      "grad_norm": 0.04480955749750137,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 121272
+    },
+    {
+      "epoch": 12.461467324290998,
+      "grad_norm": 0.056711770594120026,
+      "learning_rate": 0.01,
+      "loss": 1.973,
+      "step": 121275
+    },
+    {
+      "epoch": 12.461775585696671,
+      "grad_norm": 0.046238526701927185,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 121278
+    },
+    {
+      "epoch": 12.462083847102344,
+      "grad_norm": 0.0408644899725914,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 121281
+    },
+    {
+      "epoch": 12.462392108508014,
+      "grad_norm": 0.16705437004566193,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 121284
+    },
+    {
+      "epoch": 12.462700369913687,
+      "grad_norm": 0.03730238229036331,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 121287
+    },
+    {
+      "epoch": 12.463008631319358,
+      "grad_norm": 0.0428999587893486,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 121290
+    },
+    {
+      "epoch": 12.46331689272503,
+      "grad_norm": 0.04371247813105583,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 121293
+    },
+    {
+      "epoch": 12.463625154130703,
+      "grad_norm": 0.055939462035894394,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 121296
+    },
+    {
+      "epoch": 12.463933415536374,
+      "grad_norm": 0.062437064945697784,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 121299
+    },
+    {
+      "epoch": 12.464241676942047,
+      "grad_norm": 0.06468332558870316,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 121302
+    },
+    {
+      "epoch": 12.46454993834772,
+      "grad_norm": 0.033661019057035446,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 121305
+    },
+    {
+      "epoch": 12.46485819975339,
+      "grad_norm": 0.04528021067380905,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 121308
+    },
+    {
+      "epoch": 12.465166461159063,
+      "grad_norm": 0.17767322063446045,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 121311
+    },
+    {
+      "epoch": 12.465474722564736,
+      "grad_norm": 0.06002239137887955,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 121314
+    },
+    {
+      "epoch": 12.465782983970406,
+      "grad_norm": 0.04848655313253403,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 121317
+    },
+    {
+      "epoch": 12.466091245376079,
+      "grad_norm": 0.05200524628162384,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 121320
+    },
+    {
+      "epoch": 12.466399506781752,
+      "grad_norm": 0.03162280097603798,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 121323
+    },
+    {
+      "epoch": 12.466707768187423,
+      "grad_norm": 0.03381708264350891,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 121326
+    },
+    {
+      "epoch": 12.467016029593095,
+      "grad_norm": 0.03555064648389816,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 121329
+    },
+    {
+      "epoch": 12.467324290998768,
+      "grad_norm": 0.13290022313594818,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 121332
+    },
+    {
+      "epoch": 12.467632552404439,
+      "grad_norm": 0.08111809194087982,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 121335
+    },
+    {
+      "epoch": 12.467940813810111,
+      "grad_norm": 0.0595904104411602,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 121338
+    },
+    {
+      "epoch": 12.468249075215782,
+      "grad_norm": 0.03929143399000168,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 121341
+    },
+    {
+      "epoch": 12.468557336621455,
+      "grad_norm": 0.045894503593444824,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 121344
+    },
+    {
+      "epoch": 12.468865598027127,
+      "grad_norm": 0.040126699954271317,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 121347
+    },
+    {
+      "epoch": 12.469173859432798,
+      "grad_norm": 0.03402792662382126,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 121350
+    },
+    {
+      "epoch": 12.469482120838471,
+      "grad_norm": 0.04667404294013977,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 121353
+    },
+    {
+      "epoch": 12.469790382244144,
+      "grad_norm": 0.16611348092556,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 121356
+    },
+    {
+      "epoch": 12.470098643649814,
+      "grad_norm": 0.12104840576648712,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 121359
+    },
+    {
+      "epoch": 12.470406905055487,
+      "grad_norm": 0.10649586468935013,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 121362
+    },
+    {
+      "epoch": 12.47071516646116,
+      "grad_norm": 0.05915949493646622,
+      "learning_rate": 0.01,
+      "loss": 1.9557,
+      "step": 121365
+    },
+    {
+      "epoch": 12.47102342786683,
+      "grad_norm": 0.0690794512629509,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 121368
+    },
+    {
+      "epoch": 12.471331689272503,
+      "grad_norm": 0.09581074118614197,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 121371
+    },
+    {
+      "epoch": 12.471639950678176,
+      "grad_norm": 0.05892445519566536,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 121374
+    },
+    {
+      "epoch": 12.471948212083847,
+      "grad_norm": 0.10727337747812271,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 121377
+    },
+    {
+      "epoch": 12.47225647348952,
+      "grad_norm": 0.03369433805346489,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 121380
+    },
+    {
+      "epoch": 12.47256473489519,
+      "grad_norm": 0.052181635051965714,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 121383
+    },
+    {
+      "epoch": 12.472872996300863,
+      "grad_norm": 0.11132442951202393,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 121386
+    },
+    {
+      "epoch": 12.473181257706536,
+      "grad_norm": 0.07367842644453049,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 121389
+    },
+    {
+      "epoch": 12.473489519112206,
+      "grad_norm": 0.07586299628019333,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 121392
+    },
+    {
+      "epoch": 12.473797780517879,
+      "grad_norm": 0.06276009976863861,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 121395
+    },
+    {
+      "epoch": 12.474106041923552,
+      "grad_norm": 0.124395452439785,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 121398
+    },
+    {
+      "epoch": 12.474414303329223,
+      "grad_norm": 0.041389454156160355,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 121401
+    },
+    {
+      "epoch": 12.474722564734895,
+      "grad_norm": 0.11566449701786041,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 121404
+    },
+    {
+      "epoch": 12.475030826140568,
+      "grad_norm": 0.049872953444719315,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 121407
+    },
+    {
+      "epoch": 12.475339087546239,
+      "grad_norm": 0.0904727503657341,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 121410
+    },
+    {
+      "epoch": 12.475647348951911,
+      "grad_norm": 0.12929974496364594,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 121413
+    },
+    {
+      "epoch": 12.475955610357584,
+      "grad_norm": 0.03591210022568703,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 121416
+    },
+    {
+      "epoch": 12.476263871763255,
+      "grad_norm": 0.08779715746641159,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 121419
+    },
+    {
+      "epoch": 12.476572133168927,
+      "grad_norm": 0.06906236708164215,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 121422
+    },
+    {
+      "epoch": 12.4768803945746,
+      "grad_norm": 0.04613060504198074,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 121425
+    },
+    {
+      "epoch": 12.477188655980271,
+      "grad_norm": 0.06489048153162003,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 121428
+    },
+    {
+      "epoch": 12.477496917385944,
+      "grad_norm": 0.03923995792865753,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 121431
+    },
+    {
+      "epoch": 12.477805178791614,
+      "grad_norm": 0.036666881293058395,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 121434
+    },
+    {
+      "epoch": 12.478113440197287,
+      "grad_norm": 0.038181450217962265,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 121437
+    },
+    {
+      "epoch": 12.47842170160296,
+      "grad_norm": 0.03955162689089775,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 121440
+    },
+    {
+      "epoch": 12.47872996300863,
+      "grad_norm": 0.04029517248272896,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 121443
+    },
+    {
+      "epoch": 12.479038224414303,
+      "grad_norm": 0.15464960038661957,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 121446
+    },
+    {
+      "epoch": 12.479346485819976,
+      "grad_norm": 0.11006775498390198,
+      "learning_rate": 0.01,
+      "loss": 1.8956,
+      "step": 121449
+    },
+    {
+      "epoch": 12.479654747225647,
+      "grad_norm": 0.052893463522195816,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 121452
+    },
+    {
+      "epoch": 12.47996300863132,
+      "grad_norm": 0.05330706760287285,
+      "learning_rate": 0.01,
+      "loss": 1.9609,
+      "step": 121455
+    },
+    {
+      "epoch": 12.480271270036992,
+      "grad_norm": 0.032613810151815414,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 121458
+    },
+    {
+      "epoch": 12.480579531442663,
+      "grad_norm": 0.04011857137084007,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 121461
+    },
+    {
+      "epoch": 12.480887792848335,
+      "grad_norm": 0.0690390020608902,
+      "learning_rate": 0.01,
+      "loss": 1.9641,
+      "step": 121464
+    },
+    {
+      "epoch": 12.481196054254008,
+      "grad_norm": 0.0717402994632721,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 121467
+    },
+    {
+      "epoch": 12.481504315659679,
+      "grad_norm": 0.06392014771699905,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 121470
+    },
+    {
+      "epoch": 12.481812577065352,
+      "grad_norm": 0.12007756531238556,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 121473
+    },
+    {
+      "epoch": 12.482120838471024,
+      "grad_norm": 0.13022544980049133,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 121476
+    },
+    {
+      "epoch": 12.482429099876695,
+      "grad_norm": 0.10144117474555969,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 121479
+    },
+    {
+      "epoch": 12.482737361282368,
+      "grad_norm": 0.03754530847072601,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 121482
+    },
+    {
+      "epoch": 12.483045622688039,
+      "grad_norm": 0.03838201239705086,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 121485
+    },
+    {
+      "epoch": 12.483353884093711,
+      "grad_norm": 0.04350865259766579,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 121488
+    },
+    {
+      "epoch": 12.483662145499384,
+      "grad_norm": 0.07184895128011703,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 121491
+    },
+    {
+      "epoch": 12.483970406905055,
+      "grad_norm": 0.05595661699771881,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 121494
+    },
+    {
+      "epoch": 12.484278668310727,
+      "grad_norm": 0.0673314705491066,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 121497
+    },
+    {
+      "epoch": 12.4845869297164,
+      "grad_norm": 0.07136820256710052,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 121500
+    },
+    {
+      "epoch": 12.484895191122071,
+      "grad_norm": 0.05735662207007408,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 121503
+    },
+    {
+      "epoch": 12.485203452527744,
+      "grad_norm": 0.0498378612101078,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 121506
+    },
+    {
+      "epoch": 12.485511713933416,
+      "grad_norm": 0.038448233157396317,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 121509
+    },
+    {
+      "epoch": 12.485819975339087,
+      "grad_norm": 0.039640799164772034,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 121512
+    },
+    {
+      "epoch": 12.48612823674476,
+      "grad_norm": 0.03348913416266441,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 121515
+    },
+    {
+      "epoch": 12.486436498150432,
+      "grad_norm": 0.051115188747644424,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 121518
+    },
+    {
+      "epoch": 12.486744759556103,
+      "grad_norm": 0.1301027238368988,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 121521
+    },
+    {
+      "epoch": 12.487053020961776,
+      "grad_norm": 0.0780315101146698,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 121524
+    },
+    {
+      "epoch": 12.487361282367448,
+      "grad_norm": 0.09338266402482986,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 121527
+    },
+    {
+      "epoch": 12.48766954377312,
+      "grad_norm": 0.07173854112625122,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 121530
+    },
+    {
+      "epoch": 12.487977805178792,
+      "grad_norm": 0.09534495323896408,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 121533
+    },
+    {
+      "epoch": 12.488286066584463,
+      "grad_norm": 0.05919259414076805,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 121536
+    },
+    {
+      "epoch": 12.488594327990135,
+      "grad_norm": 0.1200721487402916,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 121539
+    },
+    {
+      "epoch": 12.488902589395808,
+      "grad_norm": 0.13612700998783112,
+      "learning_rate": 0.01,
+      "loss": 1.9673,
+      "step": 121542
+    },
+    {
+      "epoch": 12.489210850801479,
+      "grad_norm": 0.16138233244419098,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 121545
+    },
+    {
+      "epoch": 12.489519112207152,
+      "grad_norm": 0.06983632594347,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 121548
+    },
+    {
+      "epoch": 12.489827373612824,
+      "grad_norm": 0.034231752157211304,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 121551
+    },
+    {
+      "epoch": 12.490135635018495,
+      "grad_norm": 0.034840893000364304,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 121554
+    },
+    {
+      "epoch": 12.490443896424168,
+      "grad_norm": 0.03223062679171562,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 121557
+    },
+    {
+      "epoch": 12.49075215782984,
+      "grad_norm": 0.057242974638938904,
+      "learning_rate": 0.01,
+      "loss": 1.9565,
+      "step": 121560
+    },
+    {
+      "epoch": 12.491060419235511,
+      "grad_norm": 0.051756180822849274,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 121563
+    },
+    {
+      "epoch": 12.491368680641184,
+      "grad_norm": 0.043041735887527466,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 121566
+    },
+    {
+      "epoch": 12.491676942046857,
+      "grad_norm": 0.04464586824178696,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 121569
+    },
+    {
+      "epoch": 12.491985203452527,
+      "grad_norm": 0.049728620797395706,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 121572
+    },
+    {
+      "epoch": 12.4922934648582,
+      "grad_norm": 0.11435680836439133,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 121575
+    },
+    {
+      "epoch": 12.49260172626387,
+      "grad_norm": 0.049143481999635696,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 121578
+    },
+    {
+      "epoch": 12.492909987669544,
+      "grad_norm": 0.09761693328619003,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 121581
+    },
+    {
+      "epoch": 12.493218249075216,
+      "grad_norm": 0.04102834314107895,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 121584
+    },
+    {
+      "epoch": 12.493526510480887,
+      "grad_norm": 0.07719264179468155,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 121587
+    },
+    {
+      "epoch": 12.49383477188656,
+      "grad_norm": 0.06168089807033539,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 121590
+    },
+    {
+      "epoch": 12.494143033292232,
+      "grad_norm": 0.08931529521942139,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 121593
+    },
+    {
+      "epoch": 12.494451294697903,
+      "grad_norm": 0.04156995564699173,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 121596
+    },
+    {
+      "epoch": 12.494759556103576,
+      "grad_norm": 0.07967917621135712,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 121599
+    },
+    {
+      "epoch": 12.495067817509248,
+      "grad_norm": 0.07035093754529953,
+      "learning_rate": 0.01,
+      "loss": 1.9679,
+      "step": 121602
+    },
+    {
+      "epoch": 12.49537607891492,
+      "grad_norm": 0.09730745106935501,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 121605
+    },
+    {
+      "epoch": 12.495684340320592,
+      "grad_norm": 0.06220466271042824,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 121608
+    },
+    {
+      "epoch": 12.495992601726265,
+      "grad_norm": 0.10313192009925842,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 121611
+    },
+    {
+      "epoch": 12.496300863131935,
+      "grad_norm": 0.057577695697546005,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 121614
+    },
+    {
+      "epoch": 12.496609124537608,
+      "grad_norm": 0.04529568925499916,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 121617
+    },
+    {
+      "epoch": 12.49691738594328,
+      "grad_norm": 0.10593090206384659,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 121620
+    },
+    {
+      "epoch": 12.497225647348952,
+      "grad_norm": 0.05129368230700493,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 121623
+    },
+    {
+      "epoch": 12.497533908754624,
+      "grad_norm": 0.10745473951101303,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 121626
+    },
+    {
+      "epoch": 12.497842170160295,
+      "grad_norm": 0.07244491577148438,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 121629
+    },
+    {
+      "epoch": 12.498150431565968,
+      "grad_norm": 0.14453358948230743,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 121632
+    },
+    {
+      "epoch": 12.49845869297164,
+      "grad_norm": 0.13955679535865784,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 121635
+    },
+    {
+      "epoch": 12.498766954377311,
+      "grad_norm": 0.07900308817625046,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 121638
+    },
+    {
+      "epoch": 12.499075215782984,
+      "grad_norm": 0.07743179053068161,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 121641
+    },
+    {
+      "epoch": 12.499383477188656,
+      "grad_norm": 0.10477512329816818,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 121644
+    },
+    {
+      "epoch": 12.499691738594327,
+      "grad_norm": 0.06475335359573364,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 121647
+    },
+    {
+      "epoch": 12.5,
+      "grad_norm": 0.040137145668268204,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 121650
+    },
+    {
+      "epoch": 12.500308261405673,
+      "grad_norm": 0.045835766941308975,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 121653
+    },
+    {
+      "epoch": 12.500616522811344,
+      "grad_norm": 0.05901820585131645,
+      "learning_rate": 0.01,
+      "loss": 1.9699,
+      "step": 121656
+    },
+    {
+      "epoch": 12.500924784217016,
+      "grad_norm": 0.06410286575555801,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 121659
+    },
+    {
+      "epoch": 12.501233045622689,
+      "grad_norm": 0.10451801866292953,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 121662
+    },
+    {
+      "epoch": 12.50154130702836,
+      "grad_norm": 0.06614912301301956,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 121665
+    },
+    {
+      "epoch": 12.501849568434032,
+      "grad_norm": 0.09007228910923004,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 121668
+    },
+    {
+      "epoch": 12.502157829839703,
+      "grad_norm": 0.0648479238152504,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 121671
+    },
+    {
+      "epoch": 12.502466091245376,
+      "grad_norm": 0.0800272673368454,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 121674
+    },
+    {
+      "epoch": 12.502774352651048,
+      "grad_norm": 0.12876403331756592,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 121677
+    },
+    {
+      "epoch": 12.50308261405672,
+      "grad_norm": 0.08178053051233292,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 121680
+    },
+    {
+      "epoch": 12.503390875462392,
+      "grad_norm": 0.04241476580500603,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 121683
+    },
+    {
+      "epoch": 12.503699136868065,
+      "grad_norm": 0.03111073561012745,
+      "learning_rate": 0.01,
+      "loss": 1.9663,
+      "step": 121686
+    },
+    {
+      "epoch": 12.504007398273735,
+      "grad_norm": 0.03514395281672478,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 121689
+    },
+    {
+      "epoch": 12.504315659679408,
+      "grad_norm": 0.08777007460594177,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 121692
+    },
+    {
+      "epoch": 12.50462392108508,
+      "grad_norm": 0.07255978882312775,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 121695
+    },
+    {
+      "epoch": 12.504932182490752,
+      "grad_norm": 0.11769585311412811,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 121698
+    },
+    {
+      "epoch": 12.505240443896424,
+      "grad_norm": 0.05882567912340164,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 121701
+    },
+    {
+      "epoch": 12.505548705302097,
+      "grad_norm": 0.05234869569540024,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 121704
+    },
+    {
+      "epoch": 12.505856966707768,
+      "grad_norm": 0.07418510317802429,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 121707
+    },
+    {
+      "epoch": 12.50616522811344,
+      "grad_norm": 0.06958342343568802,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 121710
+    },
+    {
+      "epoch": 12.506473489519113,
+      "grad_norm": 0.06797537952661514,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 121713
+    },
+    {
+      "epoch": 12.506781750924784,
+      "grad_norm": 0.08575470000505447,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 121716
+    },
+    {
+      "epoch": 12.507090012330456,
+      "grad_norm": 0.15510590374469757,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 121719
+    },
+    {
+      "epoch": 12.50739827373613,
+      "grad_norm": 0.06296925991773605,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 121722
+    },
+    {
+      "epoch": 12.5077065351418,
+      "grad_norm": 0.1299462765455246,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 121725
+    },
+    {
+      "epoch": 12.508014796547473,
+      "grad_norm": 0.10709226131439209,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 121728
+    },
+    {
+      "epoch": 12.508323057953143,
+      "grad_norm": 0.11783047020435333,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 121731
+    },
+    {
+      "epoch": 12.508631319358816,
+      "grad_norm": 0.09729878604412079,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 121734
+    },
+    {
+      "epoch": 12.508939580764489,
+      "grad_norm": 0.05474984645843506,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 121737
+    },
+    {
+      "epoch": 12.50924784217016,
+      "grad_norm": 0.08836578577756882,
+      "learning_rate": 0.01,
+      "loss": 1.9647,
+      "step": 121740
+    },
+    {
+      "epoch": 12.509556103575832,
+      "grad_norm": 0.042429737746715546,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 121743
+    },
+    {
+      "epoch": 12.509864364981505,
+      "grad_norm": 0.049478959292173386,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 121746
+    },
+    {
+      "epoch": 12.510172626387176,
+      "grad_norm": 0.050765786319971085,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 121749
+    },
+    {
+      "epoch": 12.510480887792848,
+      "grad_norm": 0.043715640902519226,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 121752
+    },
+    {
+      "epoch": 12.510789149198521,
+      "grad_norm": 0.03809693083167076,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 121755
+    },
+    {
+      "epoch": 12.511097410604192,
+      "grad_norm": 0.11676877737045288,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 121758
+    },
+    {
+      "epoch": 12.511405672009865,
+      "grad_norm": 0.13902495801448822,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 121761
+    },
+    {
+      "epoch": 12.511713933415537,
+      "grad_norm": 0.05882659927010536,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 121764
+    },
+    {
+      "epoch": 12.512022194821208,
+      "grad_norm": 0.07055717706680298,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 121767
+    },
+    {
+      "epoch": 12.51233045622688,
+      "grad_norm": 0.06787385791540146,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 121770
+    },
+    {
+      "epoch": 12.512638717632552,
+      "grad_norm": 0.03310607001185417,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 121773
+    },
+    {
+      "epoch": 12.512946979038224,
+      "grad_norm": 0.031156888231635094,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 121776
+    },
+    {
+      "epoch": 12.513255240443897,
+      "grad_norm": 0.12905651330947876,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 121779
+    },
+    {
+      "epoch": 12.513563501849568,
+      "grad_norm": 0.16129325330257416,
+      "learning_rate": 0.01,
+      "loss": 1.9101,
+      "step": 121782
+    },
+    {
+      "epoch": 12.51387176325524,
+      "grad_norm": 0.07566894590854645,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 121785
+    },
+    {
+      "epoch": 12.514180024660913,
+      "grad_norm": 0.05478030815720558,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 121788
+    },
+    {
+      "epoch": 12.514488286066584,
+      "grad_norm": 0.06553813070058823,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 121791
+    },
+    {
+      "epoch": 12.514796547472256,
+      "grad_norm": 0.04844589903950691,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 121794
+    },
+    {
+      "epoch": 12.515104808877929,
+      "grad_norm": 0.04799551144242287,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 121797
+    },
+    {
+      "epoch": 12.5154130702836,
+      "grad_norm": 0.04788275435566902,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 121800
+    },
+    {
+      "epoch": 12.515721331689273,
+      "grad_norm": 0.04693294316530228,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 121803
+    },
+    {
+      "epoch": 12.516029593094945,
+      "grad_norm": 0.05159008875489235,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 121806
+    },
+    {
+      "epoch": 12.516337854500616,
+      "grad_norm": 0.04252525418996811,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 121809
+    },
+    {
+      "epoch": 12.516646115906289,
+      "grad_norm": 0.07126598805189133,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 121812
+    },
+    {
+      "epoch": 12.516954377311961,
+      "grad_norm": 0.2112404853105545,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 121815
+    },
+    {
+      "epoch": 12.517262638717632,
+      "grad_norm": 0.19761815667152405,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 121818
+    },
+    {
+      "epoch": 12.517570900123305,
+      "grad_norm": 0.05613444745540619,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 121821
+    },
+    {
+      "epoch": 12.517879161528978,
+      "grad_norm": 0.044825371354818344,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 121824
+    },
+    {
+      "epoch": 12.518187422934648,
+      "grad_norm": 0.0385623574256897,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 121827
+    },
+    {
+      "epoch": 12.518495684340321,
+      "grad_norm": 0.03992215171456337,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 121830
+    },
+    {
+      "epoch": 12.518803945745992,
+      "grad_norm": 0.05851850286126137,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 121833
+    },
+    {
+      "epoch": 12.519112207151665,
+      "grad_norm": 0.04692927747964859,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 121836
+    },
+    {
+      "epoch": 12.519420468557337,
+      "grad_norm": 0.044795434921979904,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 121839
+    },
+    {
+      "epoch": 12.519728729963008,
+      "grad_norm": 0.03476089611649513,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 121842
+    },
+    {
+      "epoch": 12.52003699136868,
+      "grad_norm": 0.035152047872543335,
+      "learning_rate": 0.01,
+      "loss": 1.9569,
+      "step": 121845
+    },
+    {
+      "epoch": 12.520345252774353,
+      "grad_norm": 0.04816887900233269,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 121848
+    },
+    {
+      "epoch": 12.520653514180024,
+      "grad_norm": 0.07016416639089584,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 121851
+    },
+    {
+      "epoch": 12.520961775585697,
+      "grad_norm": 0.03749404475092888,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 121854
+    },
+    {
+      "epoch": 12.52127003699137,
+      "grad_norm": 0.21347223222255707,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 121857
+    },
+    {
+      "epoch": 12.52157829839704,
+      "grad_norm": 0.08297964930534363,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 121860
+    },
+    {
+      "epoch": 12.521886559802713,
+      "grad_norm": 0.0701962485909462,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 121863
+    },
+    {
+      "epoch": 12.522194821208384,
+      "grad_norm": 0.03970552235841751,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 121866
+    },
+    {
+      "epoch": 12.522503082614056,
+      "grad_norm": 0.03517434746026993,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 121869
+    },
+    {
+      "epoch": 12.522811344019729,
+      "grad_norm": 0.06066088005900383,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 121872
+    },
+    {
+      "epoch": 12.5231196054254,
+      "grad_norm": 0.0656011626124382,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 121875
+    },
+    {
+      "epoch": 12.523427866831073,
+      "grad_norm": 0.10634774714708328,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 121878
+    },
+    {
+      "epoch": 12.523736128236745,
+      "grad_norm": 0.1101708859205246,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 121881
+    },
+    {
+      "epoch": 12.524044389642416,
+      "grad_norm": 0.04546092823147774,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 121884
+    },
+    {
+      "epoch": 12.524352651048089,
+      "grad_norm": 0.082464799284935,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 121887
+    },
+    {
+      "epoch": 12.524660912453761,
+      "grad_norm": 0.03992221876978874,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 121890
+    },
+    {
+      "epoch": 12.524969173859432,
+      "grad_norm": 0.03644541651010513,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 121893
+    },
+    {
+      "epoch": 12.525277435265105,
+      "grad_norm": 0.029982885345816612,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 121896
+    },
+    {
+      "epoch": 12.525585696670777,
+      "grad_norm": 0.16123051941394806,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 121899
+    },
+    {
+      "epoch": 12.525893958076448,
+      "grad_norm": 0.07271135598421097,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 121902
+    },
+    {
+      "epoch": 12.526202219482121,
+      "grad_norm": 0.06443187594413757,
+      "learning_rate": 0.01,
+      "loss": 1.9659,
+      "step": 121905
+    },
+    {
+      "epoch": 12.526510480887794,
+      "grad_norm": 0.03684204816818237,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 121908
+    },
+    {
+      "epoch": 12.526818742293464,
+      "grad_norm": 0.06612155586481094,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 121911
+    },
+    {
+      "epoch": 12.527127003699137,
+      "grad_norm": 0.10207538306713104,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 121914
+    },
+    {
+      "epoch": 12.52743526510481,
+      "grad_norm": 0.08532961457967758,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 121917
+    },
+    {
+      "epoch": 12.52774352651048,
+      "grad_norm": 0.05782786011695862,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 121920
+    },
+    {
+      "epoch": 12.528051787916153,
+      "grad_norm": 0.044351086020469666,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 121923
+    },
+    {
+      "epoch": 12.528360049321824,
+      "grad_norm": 0.053692616522312164,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 121926
+    },
+    {
+      "epoch": 12.528668310727497,
+      "grad_norm": 0.10878470540046692,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 121929
+    },
+    {
+      "epoch": 12.52897657213317,
+      "grad_norm": 0.09176810085773468,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 121932
+    },
+    {
+      "epoch": 12.52928483353884,
+      "grad_norm": 0.10216666758060455,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 121935
+    },
+    {
+      "epoch": 12.529593094944513,
+      "grad_norm": 0.06094343587756157,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 121938
+    },
+    {
+      "epoch": 12.529901356350186,
+      "grad_norm": 0.07060480117797852,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 121941
+    },
+    {
+      "epoch": 12.530209617755856,
+      "grad_norm": 0.04404463618993759,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 121944
+    },
+    {
+      "epoch": 12.530517879161529,
+      "grad_norm": 0.0398835763335228,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 121947
+    },
+    {
+      "epoch": 12.530826140567202,
+      "grad_norm": 0.11280908435583115,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 121950
+    },
+    {
+      "epoch": 12.531134401972873,
+      "grad_norm": 0.10008415579795837,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 121953
+    },
+    {
+      "epoch": 12.531442663378545,
+      "grad_norm": 0.08154910057783127,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 121956
+    },
+    {
+      "epoch": 12.531750924784218,
+      "grad_norm": 0.03601017966866493,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 121959
+    },
+    {
+      "epoch": 12.532059186189889,
+      "grad_norm": 0.039634574204683304,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 121962
+    },
+    {
+      "epoch": 12.532367447595561,
+      "grad_norm": 0.038583431392908096,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 121965
+    },
+    {
+      "epoch": 12.532675709001232,
+      "grad_norm": 0.09992941468954086,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 121968
+    },
+    {
+      "epoch": 12.532983970406905,
+      "grad_norm": 0.08802489191293716,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 121971
+    },
+    {
+      "epoch": 12.533292231812577,
+      "grad_norm": 0.09296849370002747,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 121974
+    },
+    {
+      "epoch": 12.533600493218248,
+      "grad_norm": 0.0878094881772995,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 121977
+    },
+    {
+      "epoch": 12.533908754623921,
+      "grad_norm": 0.04183616116642952,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 121980
+    },
+    {
+      "epoch": 12.534217016029594,
+      "grad_norm": 0.0330495610833168,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 121983
+    },
+    {
+      "epoch": 12.534525277435264,
+      "grad_norm": 0.05904131755232811,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 121986
+    },
+    {
+      "epoch": 12.534833538840937,
+      "grad_norm": 0.11319120228290558,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 121989
+    },
+    {
+      "epoch": 12.53514180024661,
+      "grad_norm": 0.038431279361248016,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 121992
+    },
+    {
+      "epoch": 12.53545006165228,
+      "grad_norm": 0.04068557918071747,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 121995
+    },
+    {
+      "epoch": 12.535758323057953,
+      "grad_norm": 0.0435246042907238,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 121998
+    },
+    {
+      "epoch": 12.536066584463626,
+      "grad_norm": 0.07753629982471466,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 122001
+    },
+    {
+      "epoch": 12.536374845869297,
+      "grad_norm": 0.11436717957258224,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 122004
+    },
+    {
+      "epoch": 12.53668310727497,
+      "grad_norm": 0.15614436566829681,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 122007
+    },
+    {
+      "epoch": 12.536991368680642,
+      "grad_norm": 0.10340051352977753,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 122010
+    },
+    {
+      "epoch": 12.537299630086313,
+      "grad_norm": 0.06114725023508072,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 122013
+    },
+    {
+      "epoch": 12.537607891491986,
+      "grad_norm": 0.044180672615766525,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 122016
+    },
+    {
+      "epoch": 12.537916152897658,
+      "grad_norm": 0.03734464943408966,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 122019
+    },
+    {
+      "epoch": 12.538224414303329,
+      "grad_norm": 0.12132205069065094,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 122022
+    },
+    {
+      "epoch": 12.538532675709002,
+      "grad_norm": 0.044691603630781174,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 122025
+    },
+    {
+      "epoch": 12.538840937114673,
+      "grad_norm": 0.1167520135641098,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 122028
+    },
+    {
+      "epoch": 12.539149198520345,
+      "grad_norm": 0.10806215554475784,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 122031
+    },
+    {
+      "epoch": 12.539457459926018,
+      "grad_norm": 0.0503133200109005,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 122034
+    },
+    {
+      "epoch": 12.539765721331689,
+      "grad_norm": 0.04720020666718483,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 122037
+    },
+    {
+      "epoch": 12.540073982737361,
+      "grad_norm": 0.0499320849776268,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 122040
+    },
+    {
+      "epoch": 12.540382244143034,
+      "grad_norm": 0.07868952304124832,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 122043
+    },
+    {
+      "epoch": 12.540690505548705,
+      "grad_norm": 0.10274378955364227,
+      "learning_rate": 0.01,
+      "loss": 1.964,
+      "step": 122046
+    },
+    {
+      "epoch": 12.540998766954377,
+      "grad_norm": 0.047186024487018585,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 122049
+    },
+    {
+      "epoch": 12.54130702836005,
+      "grad_norm": 0.09629018604755402,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 122052
+    },
+    {
+      "epoch": 12.541615289765721,
+      "grad_norm": 0.08748985081911087,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 122055
+    },
+    {
+      "epoch": 12.541923551171394,
+      "grad_norm": 0.05084213986992836,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 122058
+    },
+    {
+      "epoch": 12.542231812577064,
+      "grad_norm": 0.09220089018344879,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 122061
+    },
+    {
+      "epoch": 12.542540073982737,
+      "grad_norm": 0.13815580308437347,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 122064
+    },
+    {
+      "epoch": 12.54284833538841,
+      "grad_norm": 0.06978030502796173,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 122067
+    },
+    {
+      "epoch": 12.54315659679408,
+      "grad_norm": 0.04032085835933685,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 122070
+    },
+    {
+      "epoch": 12.543464858199753,
+      "grad_norm": 0.06091155484318733,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 122073
+    },
+    {
+      "epoch": 12.543773119605426,
+      "grad_norm": 0.07845517992973328,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 122076
+    },
+    {
+      "epoch": 12.544081381011097,
+      "grad_norm": 0.07486490905284882,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 122079
+    },
+    {
+      "epoch": 12.54438964241677,
+      "grad_norm": 0.09034962207078934,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 122082
+    },
+    {
+      "epoch": 12.544697903822442,
+      "grad_norm": 0.08133091777563095,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 122085
+    },
+    {
+      "epoch": 12.545006165228113,
+      "grad_norm": 0.05244584009051323,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 122088
+    },
+    {
+      "epoch": 12.545314426633785,
+      "grad_norm": 0.04016478732228279,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 122091
+    },
+    {
+      "epoch": 12.545622688039458,
+      "grad_norm": 0.03591429814696312,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 122094
+    },
+    {
+      "epoch": 12.545930949445129,
+      "grad_norm": 0.09179961681365967,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 122097
+    },
+    {
+      "epoch": 12.546239210850802,
+      "grad_norm": 0.05071881040930748,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 122100
+    },
+    {
+      "epoch": 12.546547472256474,
+      "grad_norm": 0.06177901849150658,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 122103
+    },
+    {
+      "epoch": 12.546855733662145,
+      "grad_norm": 0.05903288722038269,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 122106
+    },
+    {
+      "epoch": 12.547163995067818,
+      "grad_norm": 0.06547144055366516,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 122109
+    },
+    {
+      "epoch": 12.54747225647349,
+      "grad_norm": 0.11096128821372986,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 122112
+    },
+    {
+      "epoch": 12.547780517879161,
+      "grad_norm": 0.08324635028839111,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 122115
+    },
+    {
+      "epoch": 12.548088779284834,
+      "grad_norm": 0.10194128006696701,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 122118
+    },
+    {
+      "epoch": 12.548397040690505,
+      "grad_norm": 0.17126311361789703,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 122121
+    },
+    {
+      "epoch": 12.548705302096177,
+      "grad_norm": 0.03643030673265457,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 122124
+    },
+    {
+      "epoch": 12.54901356350185,
+      "grad_norm": 0.061957668513059616,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 122127
+    },
+    {
+      "epoch": 12.549321824907521,
+      "grad_norm": 0.036888238042593,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 122130
+    },
+    {
+      "epoch": 12.549630086313194,
+      "grad_norm": 0.05962397903203964,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 122133
+    },
+    {
+      "epoch": 12.549938347718866,
+      "grad_norm": 0.05716247111558914,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 122136
+    },
+    {
+      "epoch": 12.550246609124537,
+      "grad_norm": 0.09270322322845459,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 122139
+    },
+    {
+      "epoch": 12.55055487053021,
+      "grad_norm": 0.07602955400943756,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 122142
+    },
+    {
+      "epoch": 12.550863131935882,
+      "grad_norm": 0.07429090887308121,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 122145
+    },
+    {
+      "epoch": 12.551171393341553,
+      "grad_norm": 0.06275807321071625,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 122148
+    },
+    {
+      "epoch": 12.551479654747226,
+      "grad_norm": 0.07937415689229965,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 122151
+    },
+    {
+      "epoch": 12.551787916152898,
+      "grad_norm": 0.09188801795244217,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 122154
+    },
+    {
+      "epoch": 12.55209617755857,
+      "grad_norm": 0.07130987197160721,
+      "learning_rate": 0.01,
+      "loss": 1.9541,
+      "step": 122157
+    },
+    {
+      "epoch": 12.552404438964242,
+      "grad_norm": 0.05817567929625511,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 122160
+    },
+    {
+      "epoch": 12.552712700369913,
+      "grad_norm": 0.10335689783096313,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 122163
+    },
+    {
+      "epoch": 12.553020961775585,
+      "grad_norm": 0.07822994142770767,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 122166
+    },
+    {
+      "epoch": 12.553329223181258,
+      "grad_norm": 0.05971822887659073,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 122169
+    },
+    {
+      "epoch": 12.553637484586929,
+      "grad_norm": 0.06985843926668167,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 122172
+    },
+    {
+      "epoch": 12.553945745992602,
+      "grad_norm": 0.0653102770447731,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 122175
+    },
+    {
+      "epoch": 12.554254007398274,
+      "grad_norm": 0.041373997926712036,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 122178
+    },
+    {
+      "epoch": 12.554562268803945,
+      "grad_norm": 0.03778582811355591,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 122181
+    },
+    {
+      "epoch": 12.554870530209618,
+      "grad_norm": 0.05184998735785484,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 122184
+    },
+    {
+      "epoch": 12.55517879161529,
+      "grad_norm": 0.05475650355219841,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 122187
+    },
+    {
+      "epoch": 12.555487053020961,
+      "grad_norm": 0.05435996502637863,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 122190
+    },
+    {
+      "epoch": 12.555795314426634,
+      "grad_norm": 0.06992301344871521,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 122193
+    },
+    {
+      "epoch": 12.556103575832307,
+      "grad_norm": 0.10033873468637466,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 122196
+    },
+    {
+      "epoch": 12.556411837237977,
+      "grad_norm": 0.07717881351709366,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 122199
+    },
+    {
+      "epoch": 12.55672009864365,
+      "grad_norm": 0.07293485105037689,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 122202
+    },
+    {
+      "epoch": 12.557028360049323,
+      "grad_norm": 0.05288426950573921,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 122205
+    },
+    {
+      "epoch": 12.557336621454994,
+      "grad_norm": 0.0352017767727375,
+      "learning_rate": 0.01,
+      "loss": 1.956,
+      "step": 122208
+    },
+    {
+      "epoch": 12.557644882860666,
+      "grad_norm": 0.0325283445417881,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 122211
+    },
+    {
+      "epoch": 12.557953144266339,
+      "grad_norm": 0.11622142791748047,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 122214
+    },
+    {
+      "epoch": 12.55826140567201,
+      "grad_norm": 0.07114896923303604,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 122217
+    },
+    {
+      "epoch": 12.558569667077682,
+      "grad_norm": 0.0963045135140419,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 122220
+    },
+    {
+      "epoch": 12.558877928483353,
+      "grad_norm": 0.09308268874883652,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 122223
+    },
+    {
+      "epoch": 12.559186189889026,
+      "grad_norm": 0.05020532384514809,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 122226
+    },
+    {
+      "epoch": 12.559494451294698,
+      "grad_norm": 0.09539689123630524,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 122229
+    },
+    {
+      "epoch": 12.55980271270037,
+      "grad_norm": 0.05360017716884613,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 122232
+    },
+    {
+      "epoch": 12.560110974106042,
+      "grad_norm": 0.040233734995126724,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 122235
+    },
+    {
+      "epoch": 12.560419235511715,
+      "grad_norm": 0.03760543093085289,
+      "learning_rate": 0.01,
+      "loss": 1.9595,
+      "step": 122238
+    },
+    {
+      "epoch": 12.560727496917385,
+      "grad_norm": 0.06255226582288742,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 122241
+    },
+    {
+      "epoch": 12.561035758323058,
+      "grad_norm": 0.047534991055727005,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 122244
+    },
+    {
+      "epoch": 12.56134401972873,
+      "grad_norm": 0.14942295849323273,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 122247
+    },
+    {
+      "epoch": 12.561652281134402,
+      "grad_norm": 0.11508209258317947,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 122250
+    },
+    {
+      "epoch": 12.561960542540074,
+      "grad_norm": 0.04446358606219292,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 122253
+    },
+    {
+      "epoch": 12.562268803945745,
+      "grad_norm": 0.037333231419324875,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 122256
+    },
+    {
+      "epoch": 12.562577065351418,
+      "grad_norm": 0.061013709753751755,
+      "learning_rate": 0.01,
+      "loss": 1.9526,
+      "step": 122259
+    },
+    {
+      "epoch": 12.56288532675709,
+      "grad_norm": 0.04341999813914299,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 122262
+    },
+    {
+      "epoch": 12.563193588162761,
+      "grad_norm": 0.04809141531586647,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 122265
+    },
+    {
+      "epoch": 12.563501849568434,
+      "grad_norm": 0.04533364251255989,
+      "learning_rate": 0.01,
+      "loss": 1.9651,
+      "step": 122268
+    },
+    {
+      "epoch": 12.563810110974106,
+      "grad_norm": 0.04224034398794174,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 122271
+    },
+    {
+      "epoch": 12.564118372379777,
+      "grad_norm": 0.07633619755506516,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 122274
+    },
+    {
+      "epoch": 12.56442663378545,
+      "grad_norm": 0.03966088593006134,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 122277
+    },
+    {
+      "epoch": 12.564734895191123,
+      "grad_norm": 0.07610847800970078,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 122280
+    },
+    {
+      "epoch": 12.565043156596793,
+      "grad_norm": 0.0485893115401268,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 122283
+    },
+    {
+      "epoch": 12.565351418002466,
+      "grad_norm": 0.052794449031353,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 122286
+    },
+    {
+      "epoch": 12.565659679408139,
+      "grad_norm": 0.1039859801530838,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 122289
+    },
+    {
+      "epoch": 12.56596794081381,
+      "grad_norm": 0.055837295949459076,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 122292
+    },
+    {
+      "epoch": 12.566276202219482,
+      "grad_norm": 0.06777098029851913,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 122295
+    },
+    {
+      "epoch": 12.566584463625155,
+      "grad_norm": 0.037880655378103256,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 122298
+    },
+    {
+      "epoch": 12.566892725030826,
+      "grad_norm": 0.04481416940689087,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 122301
+    },
+    {
+      "epoch": 12.567200986436498,
+      "grad_norm": 0.08036676794290543,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 122304
+    },
+    {
+      "epoch": 12.567509247842171,
+      "grad_norm": 0.09721449762582779,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 122307
+    },
+    {
+      "epoch": 12.567817509247842,
+      "grad_norm": 0.033135443925857544,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 122310
+    },
+    {
+      "epoch": 12.568125770653515,
+      "grad_norm": 0.11335978657007217,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 122313
+    },
+    {
+      "epoch": 12.568434032059185,
+      "grad_norm": 0.07764448970556259,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 122316
+    },
+    {
+      "epoch": 12.568742293464858,
+      "grad_norm": 0.05308235064148903,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 122319
+    },
+    {
+      "epoch": 12.56905055487053,
+      "grad_norm": 0.057582397013902664,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 122322
+    },
+    {
+      "epoch": 12.569358816276202,
+      "grad_norm": 0.050004176795482635,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 122325
+    },
+    {
+      "epoch": 12.569667077681874,
+      "grad_norm": 0.06393969058990479,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 122328
+    },
+    {
+      "epoch": 12.569975339087547,
+      "grad_norm": 0.07683342695236206,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 122331
+    },
+    {
+      "epoch": 12.570283600493218,
+      "grad_norm": 0.05688316002488136,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 122334
+    },
+    {
+      "epoch": 12.57059186189889,
+      "grad_norm": 0.03711773455142975,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 122337
+    },
+    {
+      "epoch": 12.570900123304563,
+      "grad_norm": 0.049031812697649,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 122340
+    },
+    {
+      "epoch": 12.571208384710234,
+      "grad_norm": 0.13783791661262512,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 122343
+    },
+    {
+      "epoch": 12.571516646115906,
+      "grad_norm": 0.10908007621765137,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 122346
+    },
+    {
+      "epoch": 12.571824907521579,
+      "grad_norm": 0.048536717891693115,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 122349
+    },
+    {
+      "epoch": 12.57213316892725,
+      "grad_norm": 0.05644765868782997,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 122352
+    },
+    {
+      "epoch": 12.572441430332923,
+      "grad_norm": 0.04310297593474388,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 122355
+    },
+    {
+      "epoch": 12.572749691738593,
+      "grad_norm": 0.09706427156925201,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 122358
+    },
+    {
+      "epoch": 12.573057953144266,
+      "grad_norm": 0.10384303331375122,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 122361
+    },
+    {
+      "epoch": 12.573366214549939,
+      "grad_norm": 0.0773472934961319,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 122364
+    },
+    {
+      "epoch": 12.57367447595561,
+      "grad_norm": 0.056868892163038254,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 122367
+    },
+    {
+      "epoch": 12.573982737361282,
+      "grad_norm": 0.11514408886432648,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 122370
+    },
+    {
+      "epoch": 12.574290998766955,
+      "grad_norm": 0.047065265476703644,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 122373
+    },
+    {
+      "epoch": 12.574599260172626,
+      "grad_norm": 0.0542309433221817,
+      "learning_rate": 0.01,
+      "loss": 1.9602,
+      "step": 122376
+    },
+    {
+      "epoch": 12.574907521578298,
+      "grad_norm": 0.10700059682130814,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 122379
+    },
+    {
+      "epoch": 12.575215782983971,
+      "grad_norm": 0.10542161762714386,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 122382
+    },
+    {
+      "epoch": 12.575524044389642,
+      "grad_norm": 0.04567668214440346,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 122385
+    },
+    {
+      "epoch": 12.575832305795315,
+      "grad_norm": 0.08345144242048264,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 122388
+    },
+    {
+      "epoch": 12.576140567200987,
+      "grad_norm": 0.06775045394897461,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 122391
+    },
+    {
+      "epoch": 12.576448828606658,
+      "grad_norm": 0.09641405940055847,
+      "learning_rate": 0.01,
+      "loss": 1.9553,
+      "step": 122394
+    },
+    {
+      "epoch": 12.57675709001233,
+      "grad_norm": 0.054200150072574615,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 122397
+    },
+    {
+      "epoch": 12.577065351418003,
+      "grad_norm": 0.0304780974984169,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 122400
+    },
+    {
+      "epoch": 12.577373612823674,
+      "grad_norm": 0.10586915165185928,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 122403
+    },
+    {
+      "epoch": 12.577681874229347,
+      "grad_norm": 0.13238736987113953,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 122406
+    },
+    {
+      "epoch": 12.577990135635018,
+      "grad_norm": 0.07128183543682098,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 122409
+    },
+    {
+      "epoch": 12.57829839704069,
+      "grad_norm": 0.037349313497543335,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 122412
+    },
+    {
+      "epoch": 12.578606658446363,
+      "grad_norm": 0.04454770311713219,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 122415
+    },
+    {
+      "epoch": 12.578914919852034,
+      "grad_norm": 0.04878051206469536,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 122418
+    },
+    {
+      "epoch": 12.579223181257706,
+      "grad_norm": 0.04272807762026787,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 122421
+    },
+    {
+      "epoch": 12.579531442663379,
+      "grad_norm": 0.12098665535449982,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 122424
+    },
+    {
+      "epoch": 12.57983970406905,
+      "grad_norm": 0.06325336545705795,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 122427
+    },
+    {
+      "epoch": 12.580147965474723,
+      "grad_norm": 0.08749137073755264,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 122430
+    },
+    {
+      "epoch": 12.580456226880395,
+      "grad_norm": 0.06324999034404755,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 122433
+    },
+    {
+      "epoch": 12.580764488286066,
+      "grad_norm": 0.07804904133081436,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 122436
+    },
+    {
+      "epoch": 12.581072749691739,
+      "grad_norm": 0.13983319699764252,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 122439
+    },
+    {
+      "epoch": 12.581381011097411,
+      "grad_norm": 0.037088699638843536,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 122442
+    },
+    {
+      "epoch": 12.581689272503082,
+      "grad_norm": 0.046924177557229996,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 122445
+    },
+    {
+      "epoch": 12.581997533908755,
+      "grad_norm": 0.10420291870832443,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 122448
+    },
+    {
+      "epoch": 12.582305795314426,
+      "grad_norm": 0.12177906930446625,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 122451
+    },
+    {
+      "epoch": 12.582614056720098,
+      "grad_norm": 0.07504318654537201,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 122454
+    },
+    {
+      "epoch": 12.582922318125771,
+      "grad_norm": 0.04811587929725647,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 122457
+    },
+    {
+      "epoch": 12.583230579531442,
+      "grad_norm": 0.04745061695575714,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 122460
+    },
+    {
+      "epoch": 12.583538840937115,
+      "grad_norm": 0.09733489900827408,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 122463
+    },
+    {
+      "epoch": 12.583847102342787,
+      "grad_norm": 0.05689321085810661,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 122466
+    },
+    {
+      "epoch": 12.584155363748458,
+      "grad_norm": 0.10812503844499588,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 122469
+    },
+    {
+      "epoch": 12.58446362515413,
+      "grad_norm": 0.041134048253297806,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 122472
+    },
+    {
+      "epoch": 12.584771886559803,
+      "grad_norm": 0.04747617617249489,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 122475
+    },
+    {
+      "epoch": 12.585080147965474,
+      "grad_norm": 0.13469180464744568,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 122478
+    },
+    {
+      "epoch": 12.585388409371147,
+      "grad_norm": 0.04839875176548958,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 122481
+    },
+    {
+      "epoch": 12.58569667077682,
+      "grad_norm": 0.08033574372529984,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 122484
+    },
+    {
+      "epoch": 12.58600493218249,
+      "grad_norm": 0.05126413702964783,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 122487
+    },
+    {
+      "epoch": 12.586313193588163,
+      "grad_norm": 0.09988853335380554,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 122490
+    },
+    {
+      "epoch": 12.586621454993836,
+      "grad_norm": 0.033685505390167236,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 122493
+    },
+    {
+      "epoch": 12.586929716399506,
+      "grad_norm": 0.08883517235517502,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 122496
+    },
+    {
+      "epoch": 12.587237977805179,
+      "grad_norm": 0.044683970510959625,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 122499
+    },
+    {
+      "epoch": 12.587546239210852,
+      "grad_norm": 0.09659519046545029,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 122502
+    },
+    {
+      "epoch": 12.587854500616523,
+      "grad_norm": 0.07233305275440216,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 122505
+    },
+    {
+      "epoch": 12.588162762022195,
+      "grad_norm": 0.04250168055295944,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 122508
+    },
+    {
+      "epoch": 12.588471023427866,
+      "grad_norm": 0.09081467986106873,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 122511
+    },
+    {
+      "epoch": 12.588779284833539,
+      "grad_norm": 0.10205347836017609,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 122514
+    },
+    {
+      "epoch": 12.589087546239211,
+      "grad_norm": 0.12037431448698044,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 122517
+    },
+    {
+      "epoch": 12.589395807644882,
+      "grad_norm": 0.08119291067123413,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 122520
+    },
+    {
+      "epoch": 12.589704069050555,
+      "grad_norm": 0.0845591351389885,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 122523
+    },
+    {
+      "epoch": 12.590012330456227,
+      "grad_norm": 0.036594972014427185,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 122526
+    },
+    {
+      "epoch": 12.590320591861898,
+      "grad_norm": 0.04575011879205704,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 122529
+    },
+    {
+      "epoch": 12.590628853267571,
+      "grad_norm": 0.05228526145219803,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 122532
+    },
+    {
+      "epoch": 12.590937114673244,
+      "grad_norm": 0.0767127051949501,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 122535
+    },
+    {
+      "epoch": 12.591245376078914,
+      "grad_norm": 0.06612443178892136,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 122538
+    },
+    {
+      "epoch": 12.591553637484587,
+      "grad_norm": 0.03941207379102707,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 122541
+    },
+    {
+      "epoch": 12.59186189889026,
+      "grad_norm": 0.04176707565784454,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 122544
+    },
+    {
+      "epoch": 12.59217016029593,
+      "grad_norm": 0.053258270025253296,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 122547
+    },
+    {
+      "epoch": 12.592478421701603,
+      "grad_norm": 0.0759519562125206,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 122550
+    },
+    {
+      "epoch": 12.592786683107274,
+      "grad_norm": 0.06660858541727066,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 122553
+    },
+    {
+      "epoch": 12.593094944512947,
+      "grad_norm": 0.1326006054878235,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 122556
+    },
+    {
+      "epoch": 12.59340320591862,
+      "grad_norm": 0.05958440154790878,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 122559
+    },
+    {
+      "epoch": 12.59371146732429,
+      "grad_norm": 0.04930621758103371,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 122562
+    },
+    {
+      "epoch": 12.594019728729963,
+      "grad_norm": 0.03780602663755417,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 122565
+    },
+    {
+      "epoch": 12.594327990135636,
+      "grad_norm": 0.05326778441667557,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 122568
+    },
+    {
+      "epoch": 12.594636251541306,
+      "grad_norm": 0.07156463712453842,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 122571
+    },
+    {
+      "epoch": 12.594944512946979,
+      "grad_norm": 0.07190526276826859,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 122574
+    },
+    {
+      "epoch": 12.595252774352652,
+      "grad_norm": 0.11505575478076935,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 122577
+    },
+    {
+      "epoch": 12.595561035758323,
+      "grad_norm": 0.06922391802072525,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 122580
+    },
+    {
+      "epoch": 12.595869297163995,
+      "grad_norm": 0.037435878068208694,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 122583
+    },
+    {
+      "epoch": 12.596177558569668,
+      "grad_norm": 0.037485379725694656,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 122586
+    },
+    {
+      "epoch": 12.596485819975339,
+      "grad_norm": 0.04112362116575241,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 122589
+    },
+    {
+      "epoch": 12.596794081381011,
+      "grad_norm": 0.10568088293075562,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 122592
+    },
+    {
+      "epoch": 12.597102342786684,
+      "grad_norm": 0.07157807052135468,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 122595
+    },
+    {
+      "epoch": 12.597410604192355,
+      "grad_norm": 0.08673489093780518,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 122598
+    },
+    {
+      "epoch": 12.597718865598027,
+      "grad_norm": 0.06275806576013565,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 122601
+    },
+    {
+      "epoch": 12.598027127003698,
+      "grad_norm": 0.1066279411315918,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 122604
+    },
+    {
+      "epoch": 12.598335388409371,
+      "grad_norm": 0.13730093836784363,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 122607
+    },
+    {
+      "epoch": 12.598643649815044,
+      "grad_norm": 0.04778913035988808,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 122610
+    },
+    {
+      "epoch": 12.598951911220714,
+      "grad_norm": 0.07837532460689545,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 122613
+    },
+    {
+      "epoch": 12.599260172626387,
+      "grad_norm": 0.07405108213424683,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 122616
+    },
+    {
+      "epoch": 12.59956843403206,
+      "grad_norm": 0.07231219857931137,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 122619
+    },
+    {
+      "epoch": 12.59987669543773,
+      "grad_norm": 0.08693543821573257,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 122622
+    },
+    {
+      "epoch": 12.600184956843403,
+      "grad_norm": 0.07112149894237518,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 122625
+    },
+    {
+      "epoch": 12.600493218249076,
+      "grad_norm": 0.13867275416851044,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 122628
+    },
+    {
+      "epoch": 12.600801479654747,
+      "grad_norm": 0.1485087275505066,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 122631
+    },
+    {
+      "epoch": 12.60110974106042,
+      "grad_norm": 0.09758572280406952,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 122634
+    },
+    {
+      "epoch": 12.601418002466092,
+      "grad_norm": 0.05771038681268692,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 122637
+    },
+    {
+      "epoch": 12.601726263871763,
+      "grad_norm": 0.07908354699611664,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 122640
+    },
+    {
+      "epoch": 12.602034525277436,
+      "grad_norm": 0.050742391496896744,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 122643
+    },
+    {
+      "epoch": 12.602342786683106,
+      "grad_norm": 0.040271371603012085,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 122646
+    },
+    {
+      "epoch": 12.602651048088779,
+      "grad_norm": 0.12437929958105087,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 122649
+    },
+    {
+      "epoch": 12.602959309494452,
+      "grad_norm": 0.0745621845126152,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 122652
+    },
+    {
+      "epoch": 12.603267570900123,
+      "grad_norm": 0.033611565828323364,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 122655
+    },
+    {
+      "epoch": 12.603575832305795,
+      "grad_norm": 0.14124849438667297,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 122658
+    },
+    {
+      "epoch": 12.603884093711468,
+      "grad_norm": 0.10393095016479492,
+      "learning_rate": 0.01,
+      "loss": 1.9736,
+      "step": 122661
+    },
+    {
+      "epoch": 12.604192355117139,
+      "grad_norm": 0.049038369208574295,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 122664
+    },
+    {
+      "epoch": 12.604500616522811,
+      "grad_norm": 0.04277941957116127,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 122667
+    },
+    {
+      "epoch": 12.604808877928484,
+      "grad_norm": 0.03375561535358429,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 122670
+    },
+    {
+      "epoch": 12.605117139334155,
+      "grad_norm": 0.07068139314651489,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 122673
+    },
+    {
+      "epoch": 12.605425400739827,
+      "grad_norm": 0.08763331174850464,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 122676
+    },
+    {
+      "epoch": 12.6057336621455,
+      "grad_norm": 0.0836746096611023,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 122679
+    },
+    {
+      "epoch": 12.606041923551171,
+      "grad_norm": 0.031440578401088715,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 122682
+    },
+    {
+      "epoch": 12.606350184956844,
+      "grad_norm": 0.06475764513015747,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 122685
+    },
+    {
+      "epoch": 12.606658446362516,
+      "grad_norm": 0.13389822840690613,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 122688
+    },
+    {
+      "epoch": 12.606966707768187,
+      "grad_norm": 0.05801908299326897,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 122691
+    },
+    {
+      "epoch": 12.60727496917386,
+      "grad_norm": 0.05795949324965477,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 122694
+    },
+    {
+      "epoch": 12.607583230579532,
+      "grad_norm": 0.030270157381892204,
+      "learning_rate": 0.01,
+      "loss": 1.9671,
+      "step": 122697
+    },
+    {
+      "epoch": 12.607891491985203,
+      "grad_norm": 0.04646274074912071,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 122700
+    },
+    {
+      "epoch": 12.608199753390876,
+      "grad_norm": 0.09027555584907532,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 122703
+    },
+    {
+      "epoch": 12.608508014796547,
+      "grad_norm": 0.08750518411397934,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 122706
+    },
+    {
+      "epoch": 12.60881627620222,
+      "grad_norm": 0.06673325598239899,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 122709
+    },
+    {
+      "epoch": 12.609124537607892,
+      "grad_norm": 0.0680864155292511,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 122712
+    },
+    {
+      "epoch": 12.609432799013563,
+      "grad_norm": 0.15428978204727173,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 122715
+    },
+    {
+      "epoch": 12.609741060419235,
+      "grad_norm": 0.047103870660066605,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 122718
+    },
+    {
+      "epoch": 12.610049321824908,
+      "grad_norm": 0.07863438129425049,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 122721
+    },
+    {
+      "epoch": 12.610357583230579,
+      "grad_norm": 0.11159279197454453,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 122724
+    },
+    {
+      "epoch": 12.610665844636252,
+      "grad_norm": 0.10510515421628952,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 122727
+    },
+    {
+      "epoch": 12.610974106041924,
+      "grad_norm": 0.07300017774105072,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 122730
+    },
+    {
+      "epoch": 12.611282367447595,
+      "grad_norm": 0.06340684741735458,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 122733
+    },
+    {
+      "epoch": 12.611590628853268,
+      "grad_norm": 0.06944388151168823,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 122736
+    },
+    {
+      "epoch": 12.61189889025894,
+      "grad_norm": 0.05677616968750954,
+      "learning_rate": 0.01,
+      "loss": 1.967,
+      "step": 122739
+    },
+    {
+      "epoch": 12.612207151664611,
+      "grad_norm": 0.07102473825216293,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 122742
+    },
+    {
+      "epoch": 12.612515413070284,
+      "grad_norm": 0.10698739439249039,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 122745
+    },
+    {
+      "epoch": 12.612823674475955,
+      "grad_norm": 0.049000632017850876,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 122748
+    },
+    {
+      "epoch": 12.613131935881627,
+      "grad_norm": 0.10147199034690857,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 122751
+    },
+    {
+      "epoch": 12.6134401972873,
+      "grad_norm": 0.08941066265106201,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 122754
+    },
+    {
+      "epoch": 12.613748458692971,
+      "grad_norm": 0.06283801794052124,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 122757
+    },
+    {
+      "epoch": 12.614056720098644,
+      "grad_norm": 0.04824201762676239,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 122760
+    },
+    {
+      "epoch": 12.614364981504316,
+      "grad_norm": 0.05474422872066498,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 122763
+    },
+    {
+      "epoch": 12.614673242909987,
+      "grad_norm": 0.06307218223810196,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 122766
+    },
+    {
+      "epoch": 12.61498150431566,
+      "grad_norm": 0.0631837546825409,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 122769
+    },
+    {
+      "epoch": 12.615289765721332,
+      "grad_norm": 0.06105835363268852,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 122772
+    },
+    {
+      "epoch": 12.615598027127003,
+      "grad_norm": 0.06466460227966309,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 122775
+    },
+    {
+      "epoch": 12.615906288532676,
+      "grad_norm": 0.03602457046508789,
+      "learning_rate": 0.01,
+      "loss": 1.9504,
+      "step": 122778
+    },
+    {
+      "epoch": 12.616214549938348,
+      "grad_norm": 0.03306679427623749,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 122781
+    },
+    {
+      "epoch": 12.61652281134402,
+      "grad_norm": 0.05408303439617157,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 122784
+    },
+    {
+      "epoch": 12.616831072749692,
+      "grad_norm": 0.11107456684112549,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 122787
+    },
+    {
+      "epoch": 12.617139334155365,
+      "grad_norm": 0.04231736809015274,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 122790
+    },
+    {
+      "epoch": 12.617447595561035,
+      "grad_norm": 0.11673669517040253,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 122793
+    },
+    {
+      "epoch": 12.617755856966708,
+      "grad_norm": 0.062019962817430496,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 122796
+    },
+    {
+      "epoch": 12.618064118372379,
+      "grad_norm": 0.07347622513771057,
+      "learning_rate": 0.01,
+      "loss": 1.9726,
+      "step": 122799
+    },
+    {
+      "epoch": 12.618372379778052,
+      "grad_norm": 0.0687282383441925,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 122802
+    },
+    {
+      "epoch": 12.618680641183724,
+      "grad_norm": 0.13374529778957367,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 122805
+    },
+    {
+      "epoch": 12.618988902589395,
+      "grad_norm": 0.1499875783920288,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 122808
+    },
+    {
+      "epoch": 12.619297163995068,
+      "grad_norm": 0.1031130775809288,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 122811
+    },
+    {
+      "epoch": 12.61960542540074,
+      "grad_norm": 0.06197386234998703,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 122814
+    },
+    {
+      "epoch": 12.619913686806411,
+      "grad_norm": 0.045583710074424744,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 122817
+    },
+    {
+      "epoch": 12.620221948212084,
+      "grad_norm": 0.06573253870010376,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 122820
+    },
+    {
+      "epoch": 12.620530209617757,
+      "grad_norm": 0.038534976541996,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 122823
+    },
+    {
+      "epoch": 12.620838471023427,
+      "grad_norm": 0.046993188560009,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 122826
+    },
+    {
+      "epoch": 12.6211467324291,
+      "grad_norm": 0.10635440051555634,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 122829
+    },
+    {
+      "epoch": 12.621454993834773,
+      "grad_norm": 0.033428799360990524,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 122832
+    },
+    {
+      "epoch": 12.621763255240444,
+      "grad_norm": 0.11464046686887741,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 122835
+    },
+    {
+      "epoch": 12.622071516646116,
+      "grad_norm": 0.08591221272945404,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 122838
+    },
+    {
+      "epoch": 12.622379778051787,
+      "grad_norm": 0.07906792312860489,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 122841
+    },
+    {
+      "epoch": 12.62268803945746,
+      "grad_norm": 0.058949537575244904,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 122844
+    },
+    {
+      "epoch": 12.622996300863132,
+      "grad_norm": 0.10783743858337402,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 122847
+    },
+    {
+      "epoch": 12.623304562268803,
+      "grad_norm": 0.10258825868368149,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 122850
+    },
+    {
+      "epoch": 12.623612823674476,
+      "grad_norm": 0.09038569778203964,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 122853
+    },
+    {
+      "epoch": 12.623921085080148,
+      "grad_norm": 0.11617942899465561,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 122856
+    },
+    {
+      "epoch": 12.62422934648582,
+      "grad_norm": 0.14557698369026184,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 122859
+    },
+    {
+      "epoch": 12.624537607891492,
+      "grad_norm": 0.08018018305301666,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 122862
+    },
+    {
+      "epoch": 12.624845869297165,
+      "grad_norm": 0.06026039272546768,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 122865
+    },
+    {
+      "epoch": 12.625154130702835,
+      "grad_norm": 0.05437018349766731,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 122868
+    },
+    {
+      "epoch": 12.625462392108508,
+      "grad_norm": 0.06585533916950226,
+      "learning_rate": 0.01,
+      "loss": 1.9583,
+      "step": 122871
+    },
+    {
+      "epoch": 12.62577065351418,
+      "grad_norm": 0.053909752517938614,
+      "learning_rate": 0.01,
+      "loss": 1.9017,
+      "step": 122874
+    },
+    {
+      "epoch": 12.626078914919852,
+      "grad_norm": 0.051786426454782486,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 122877
+    },
+    {
+      "epoch": 12.626387176325524,
+      "grad_norm": 0.0947243794798851,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 122880
+    },
+    {
+      "epoch": 12.626695437731197,
+      "grad_norm": 0.05719350650906563,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 122883
+    },
+    {
+      "epoch": 12.627003699136868,
+      "grad_norm": 0.14443199336528778,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 122886
+    },
+    {
+      "epoch": 12.62731196054254,
+      "grad_norm": 0.05690363794565201,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 122889
+    },
+    {
+      "epoch": 12.627620221948213,
+      "grad_norm": 0.04786375164985657,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 122892
+    },
+    {
+      "epoch": 12.627928483353884,
+      "grad_norm": 0.09731481224298477,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 122895
+    },
+    {
+      "epoch": 12.628236744759556,
+      "grad_norm": 0.05569015070796013,
+      "learning_rate": 0.01,
+      "loss": 1.96,
+      "step": 122898
+    },
+    {
+      "epoch": 12.628545006165227,
+      "grad_norm": 0.05832977965474129,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 122901
+    },
+    {
+      "epoch": 12.6288532675709,
+      "grad_norm": 0.06494420766830444,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 122904
+    },
+    {
+      "epoch": 12.629161528976573,
+      "grad_norm": 0.05205019563436508,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 122907
+    },
+    {
+      "epoch": 12.629469790382243,
+      "grad_norm": 0.04261380434036255,
+      "learning_rate": 0.01,
+      "loss": 1.9534,
+      "step": 122910
+    },
+    {
+      "epoch": 12.629778051787916,
+      "grad_norm": 0.0320679247379303,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 122913
+    },
+    {
+      "epoch": 12.630086313193589,
+      "grad_norm": 0.13554465770721436,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 122916
+    },
+    {
+      "epoch": 12.63039457459926,
+      "grad_norm": 0.037886518985033035,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 122919
+    },
+    {
+      "epoch": 12.630702836004932,
+      "grad_norm": 0.044243793934583664,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 122922
+    },
+    {
+      "epoch": 12.631011097410605,
+      "grad_norm": 0.046359576284885406,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 122925
+    },
+    {
+      "epoch": 12.631319358816276,
+      "grad_norm": 0.036917366087436676,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 122928
+    },
+    {
+      "epoch": 12.631627620221948,
+      "grad_norm": 0.04713550955057144,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 122931
+    },
+    {
+      "epoch": 12.631935881627621,
+      "grad_norm": 0.0435897521674633,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 122934
+    },
+    {
+      "epoch": 12.632244143033292,
+      "grad_norm": 0.0550503171980381,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 122937
+    },
+    {
+      "epoch": 12.632552404438965,
+      "grad_norm": 0.0765051394701004,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 122940
+    },
+    {
+      "epoch": 12.632860665844635,
+      "grad_norm": 0.11795663833618164,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 122943
+    },
+    {
+      "epoch": 12.633168927250308,
+      "grad_norm": 0.13961495459079742,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 122946
+    },
+    {
+      "epoch": 12.63347718865598,
+      "grad_norm": 0.08773016929626465,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 122949
+    },
+    {
+      "epoch": 12.633785450061652,
+      "grad_norm": 0.06994469463825226,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 122952
+    },
+    {
+      "epoch": 12.634093711467324,
+      "grad_norm": 0.048322346061468124,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 122955
+    },
+    {
+      "epoch": 12.634401972872997,
+      "grad_norm": 0.08110317587852478,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 122958
+    },
+    {
+      "epoch": 12.634710234278668,
+      "grad_norm": 0.06748035550117493,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 122961
+    },
+    {
+      "epoch": 12.63501849568434,
+      "grad_norm": 0.06809863448143005,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 122964
+    },
+    {
+      "epoch": 12.635326757090013,
+      "grad_norm": 0.05909579619765282,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 122967
+    },
+    {
+      "epoch": 12.635635018495684,
+      "grad_norm": 0.04641105979681015,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 122970
+    },
+    {
+      "epoch": 12.635943279901356,
+      "grad_norm": 0.038915254175662994,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 122973
+    },
+    {
+      "epoch": 12.636251541307029,
+      "grad_norm": 0.10991251468658447,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 122976
+    },
+    {
+      "epoch": 12.6365598027127,
+      "grad_norm": 0.0334751196205616,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 122979
+    },
+    {
+      "epoch": 12.636868064118373,
+      "grad_norm": 0.0603213906288147,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 122982
+    },
+    {
+      "epoch": 12.637176325524045,
+      "grad_norm": 0.1104741245508194,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 122985
+    },
+    {
+      "epoch": 12.637484586929716,
+      "grad_norm": 0.036284010857343674,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 122988
+    },
+    {
+      "epoch": 12.637792848335389,
+      "grad_norm": 0.0331115685403347,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 122991
+    },
+    {
+      "epoch": 12.63810110974106,
+      "grad_norm": 0.10184327512979507,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 122994
+    },
+    {
+      "epoch": 12.638409371146732,
+      "grad_norm": 0.07448579370975494,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 122997
+    },
+    {
+      "epoch": 12.638717632552405,
+      "grad_norm": 0.043106839060783386,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 123000
+    },
+    {
+      "epoch": 12.639025893958076,
+      "grad_norm": 0.13510780036449432,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 123003
+    },
+    {
+      "epoch": 12.639334155363748,
+      "grad_norm": 0.07734852284193039,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 123006
+    },
+    {
+      "epoch": 12.639642416769421,
+      "grad_norm": 0.08094023168087006,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 123009
+    },
+    {
+      "epoch": 12.639950678175092,
+      "grad_norm": 0.07922966033220291,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 123012
+    },
+    {
+      "epoch": 12.640258939580765,
+      "grad_norm": 0.052409034222364426,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 123015
+    },
+    {
+      "epoch": 12.640567200986437,
+      "grad_norm": 0.04294409230351448,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 123018
+    },
+    {
+      "epoch": 12.640875462392108,
+      "grad_norm": 0.11415386199951172,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 123021
+    },
+    {
+      "epoch": 12.64118372379778,
+      "grad_norm": 0.03108101338148117,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 123024
+    },
+    {
+      "epoch": 12.641491985203453,
+      "grad_norm": 0.08873893320560455,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 123027
+    },
+    {
+      "epoch": 12.641800246609124,
+      "grad_norm": 0.058484312146902084,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 123030
+    },
+    {
+      "epoch": 12.642108508014797,
+      "grad_norm": 0.1172776147723198,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 123033
+    },
+    {
+      "epoch": 12.642416769420468,
+      "grad_norm": 0.08727061003446579,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 123036
+    },
+    {
+      "epoch": 12.64272503082614,
+      "grad_norm": 0.04476449266076088,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 123039
+    },
+    {
+      "epoch": 12.643033292231813,
+      "grad_norm": 0.0652472972869873,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 123042
+    },
+    {
+      "epoch": 12.643341553637484,
+      "grad_norm": 0.16113439202308655,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 123045
+    },
+    {
+      "epoch": 12.643649815043156,
+      "grad_norm": 0.0810934454202652,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 123048
+    },
+    {
+      "epoch": 12.643958076448829,
+      "grad_norm": 0.036866363137960434,
+      "learning_rate": 0.01,
+      "loss": 1.9694,
+      "step": 123051
+    },
+    {
+      "epoch": 12.6442663378545,
+      "grad_norm": 0.04961570352315903,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 123054
+    },
+    {
+      "epoch": 12.644574599260173,
+      "grad_norm": 0.046355172991752625,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 123057
+    },
+    {
+      "epoch": 12.644882860665845,
+      "grad_norm": 0.05054159089922905,
+      "learning_rate": 0.01,
+      "loss": 1.9564,
+      "step": 123060
+    },
+    {
+      "epoch": 12.645191122071516,
+      "grad_norm": 0.03561927005648613,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 123063
+    },
+    {
+      "epoch": 12.645499383477189,
+      "grad_norm": 0.09091931581497192,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 123066
+    },
+    {
+      "epoch": 12.645807644882861,
+      "grad_norm": 0.06427352130413055,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 123069
+    },
+    {
+      "epoch": 12.646115906288532,
+      "grad_norm": 0.05945998430252075,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 123072
+    },
+    {
+      "epoch": 12.646424167694205,
+      "grad_norm": 0.17813847959041595,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 123075
+    },
+    {
+      "epoch": 12.646732429099877,
+      "grad_norm": 0.0860227569937706,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 123078
+    },
+    {
+      "epoch": 12.647040690505548,
+      "grad_norm": 0.05374440178275108,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 123081
+    },
+    {
+      "epoch": 12.647348951911221,
+      "grad_norm": 0.04365652799606323,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 123084
+    },
+    {
+      "epoch": 12.647657213316894,
+      "grad_norm": 0.042210496962070465,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 123087
+    },
+    {
+      "epoch": 12.647965474722564,
+      "grad_norm": 0.06997524946928024,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 123090
+    },
+    {
+      "epoch": 12.648273736128237,
+      "grad_norm": 0.08367611467838287,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 123093
+    },
+    {
+      "epoch": 12.648581997533908,
+      "grad_norm": 0.05943150445818901,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 123096
+    },
+    {
+      "epoch": 12.64889025893958,
+      "grad_norm": 0.04505249112844467,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 123099
+    },
+    {
+      "epoch": 12.649198520345253,
+      "grad_norm": 0.12502211332321167,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 123102
+    },
+    {
+      "epoch": 12.649506781750924,
+      "grad_norm": 0.11971963942050934,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 123105
+    },
+    {
+      "epoch": 12.649815043156597,
+      "grad_norm": 0.09976421296596527,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 123108
+    },
+    {
+      "epoch": 12.65012330456227,
+      "grad_norm": 0.11778843402862549,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 123111
+    },
+    {
+      "epoch": 12.65043156596794,
+      "grad_norm": 0.06445219367742538,
+      "learning_rate": 0.01,
+      "loss": 1.9475,
+      "step": 123114
+    },
+    {
+      "epoch": 12.650739827373613,
+      "grad_norm": 0.028855200856924057,
+      "learning_rate": 0.01,
+      "loss": 1.897,
+      "step": 123117
+    },
+    {
+      "epoch": 12.651048088779286,
+      "grad_norm": 0.04714740812778473,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 123120
+    },
+    {
+      "epoch": 12.651356350184956,
+      "grad_norm": 0.06030753254890442,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 123123
+    },
+    {
+      "epoch": 12.651664611590629,
+      "grad_norm": 0.059579312801361084,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 123126
+    },
+    {
+      "epoch": 12.651972872996302,
+      "grad_norm": 0.04322392866015434,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 123129
+    },
+    {
+      "epoch": 12.652281134401973,
+      "grad_norm": 0.04606184735894203,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 123132
+    },
+    {
+      "epoch": 12.652589395807645,
+      "grad_norm": 0.12499433755874634,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 123135
+    },
+    {
+      "epoch": 12.652897657213316,
+      "grad_norm": 0.11938849836587906,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 123138
+    },
+    {
+      "epoch": 12.653205918618989,
+      "grad_norm": 0.042211614549160004,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 123141
+    },
+    {
+      "epoch": 12.653514180024661,
+      "grad_norm": 0.0965668112039566,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 123144
+    },
+    {
+      "epoch": 12.653822441430332,
+      "grad_norm": 0.05237396061420441,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 123147
+    },
+    {
+      "epoch": 12.654130702836005,
+      "grad_norm": 0.08860108256340027,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 123150
+    },
+    {
+      "epoch": 12.654438964241677,
+      "grad_norm": 0.07654953002929688,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 123153
+    },
+    {
+      "epoch": 12.654747225647348,
+      "grad_norm": 0.09266380220651627,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 123156
+    },
+    {
+      "epoch": 12.655055487053021,
+      "grad_norm": 0.0614088736474514,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 123159
+    },
+    {
+      "epoch": 12.655363748458694,
+      "grad_norm": 0.04387489706277847,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 123162
+    },
+    {
+      "epoch": 12.655672009864364,
+      "grad_norm": 0.04052634537220001,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 123165
+    },
+    {
+      "epoch": 12.655980271270037,
+      "grad_norm": 0.08114446699619293,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 123168
+    },
+    {
+      "epoch": 12.65628853267571,
+      "grad_norm": 0.057578977197408676,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 123171
+    },
+    {
+      "epoch": 12.65659679408138,
+      "grad_norm": 0.08191578835248947,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 123174
+    },
+    {
+      "epoch": 12.656905055487053,
+      "grad_norm": 0.04840002581477165,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 123177
+    },
+    {
+      "epoch": 12.657213316892726,
+      "grad_norm": 0.09766780585050583,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 123180
+    },
+    {
+      "epoch": 12.657521578298397,
+      "grad_norm": 0.05007852986454964,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 123183
+    },
+    {
+      "epoch": 12.65782983970407,
+      "grad_norm": 0.12118306010961533,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 123186
+    },
+    {
+      "epoch": 12.65813810110974,
+      "grad_norm": 0.12096033990383148,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 123189
+    },
+    {
+      "epoch": 12.658446362515413,
+      "grad_norm": 0.044699691236019135,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 123192
+    },
+    {
+      "epoch": 12.658754623921086,
+      "grad_norm": 0.03887972608208656,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 123195
+    },
+    {
+      "epoch": 12.659062885326756,
+      "grad_norm": 0.044233959168195724,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 123198
+    },
+    {
+      "epoch": 12.659371146732429,
+      "grad_norm": 0.06127122789621353,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 123201
+    },
+    {
+      "epoch": 12.659679408138102,
+      "grad_norm": 0.04868535324931145,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 123204
+    },
+    {
+      "epoch": 12.659987669543773,
+      "grad_norm": 0.041472144424915314,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 123207
+    },
+    {
+      "epoch": 12.660295930949445,
+      "grad_norm": 0.09328923374414444,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 123210
+    },
+    {
+      "epoch": 12.660604192355118,
+      "grad_norm": 0.09349977225065231,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 123213
+    },
+    {
+      "epoch": 12.660912453760789,
+      "grad_norm": 0.13697515428066254,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 123216
+    },
+    {
+      "epoch": 12.661220715166461,
+      "grad_norm": 0.1092478558421135,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 123219
+    },
+    {
+      "epoch": 12.661528976572134,
+      "grad_norm": 0.0793212503194809,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 123222
+    },
+    {
+      "epoch": 12.661837237977805,
+      "grad_norm": 0.03937581181526184,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 123225
+    },
+    {
+      "epoch": 12.662145499383477,
+      "grad_norm": 0.034699954092502594,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 123228
+    },
+    {
+      "epoch": 12.662453760789148,
+      "grad_norm": 0.05393664166331291,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 123231
+    },
+    {
+      "epoch": 12.662762022194821,
+      "grad_norm": 0.07334361970424652,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 123234
+    },
+    {
+      "epoch": 12.663070283600494,
+      "grad_norm": 0.04054495692253113,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 123237
+    },
+    {
+      "epoch": 12.663378545006164,
+      "grad_norm": 0.06787239760160446,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 123240
+    },
+    {
+      "epoch": 12.663686806411837,
+      "grad_norm": 0.04934227466583252,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 123243
+    },
+    {
+      "epoch": 12.66399506781751,
+      "grad_norm": 0.04054686799645424,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 123246
+    },
+    {
+      "epoch": 12.66430332922318,
+      "grad_norm": 0.0366259329020977,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 123249
+    },
+    {
+      "epoch": 12.664611590628853,
+      "grad_norm": 0.04150351881980896,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 123252
+    },
+    {
+      "epoch": 12.664919852034526,
+      "grad_norm": 0.08996198326349258,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 123255
+    },
+    {
+      "epoch": 12.665228113440197,
+      "grad_norm": 0.17594537138938904,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 123258
+    },
+    {
+      "epoch": 12.66553637484587,
+      "grad_norm": 0.1170310527086258,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 123261
+    },
+    {
+      "epoch": 12.665844636251542,
+      "grad_norm": 0.07598567754030228,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 123264
+    },
+    {
+      "epoch": 12.666152897657213,
+      "grad_norm": 0.06181150674819946,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 123267
+    },
+    {
+      "epoch": 12.666461159062885,
+      "grad_norm": 0.055621251463890076,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 123270
+    },
+    {
+      "epoch": 12.666769420468558,
+      "grad_norm": 0.050596028566360474,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 123273
+    },
+    {
+      "epoch": 12.667077681874229,
+      "grad_norm": 0.08173961192369461,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 123276
+    },
+    {
+      "epoch": 12.667385943279902,
+      "grad_norm": 0.052812159061431885,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 123279
+    },
+    {
+      "epoch": 12.667694204685574,
+      "grad_norm": 0.06073348596692085,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 123282
+    },
+    {
+      "epoch": 12.668002466091245,
+      "grad_norm": 0.05775995925068855,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 123285
+    },
+    {
+      "epoch": 12.668310727496918,
+      "grad_norm": 0.06133884936571121,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 123288
+    },
+    {
+      "epoch": 12.668618988902589,
+      "grad_norm": 0.04938175156712532,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 123291
+    },
+    {
+      "epoch": 12.668927250308261,
+      "grad_norm": 0.12475492805242538,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 123294
+    },
+    {
+      "epoch": 12.669235511713934,
+      "grad_norm": 0.04723283275961876,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 123297
+    },
+    {
+      "epoch": 12.669543773119605,
+      "grad_norm": 0.062252528965473175,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 123300
+    },
+    {
+      "epoch": 12.669852034525277,
+      "grad_norm": 0.10509909689426422,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 123303
+    },
+    {
+      "epoch": 12.67016029593095,
+      "grad_norm": 0.08289887756109238,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 123306
+    },
+    {
+      "epoch": 12.670468557336621,
+      "grad_norm": 0.08722329139709473,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 123309
+    },
+    {
+      "epoch": 12.670776818742294,
+      "grad_norm": 0.12085244804620743,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 123312
+    },
+    {
+      "epoch": 12.671085080147966,
+      "grad_norm": 0.16747604310512543,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 123315
+    },
+    {
+      "epoch": 12.671393341553637,
+      "grad_norm": 0.10610677301883698,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 123318
+    },
+    {
+      "epoch": 12.67170160295931,
+      "grad_norm": 0.056399762630462646,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 123321
+    },
+    {
+      "epoch": 12.672009864364982,
+      "grad_norm": 0.0351298563182354,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 123324
+    },
+    {
+      "epoch": 12.672318125770653,
+      "grad_norm": 0.0395541787147522,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 123327
+    },
+    {
+      "epoch": 12.672626387176326,
+      "grad_norm": 0.03527429327368736,
+      "learning_rate": 0.01,
+      "loss": 1.9684,
+      "step": 123330
+    },
+    {
+      "epoch": 12.672934648581997,
+      "grad_norm": 0.03256375715136528,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 123333
+    },
+    {
+      "epoch": 12.67324290998767,
+      "grad_norm": 0.11099425703287125,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 123336
+    },
+    {
+      "epoch": 12.673551171393342,
+      "grad_norm": 0.08047635853290558,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 123339
+    },
+    {
+      "epoch": 12.673859432799013,
+      "grad_norm": 0.06533856689929962,
+      "learning_rate": 0.01,
+      "loss": 1.8991,
+      "step": 123342
+    },
+    {
+      "epoch": 12.674167694204685,
+      "grad_norm": 0.04292716085910797,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 123345
+    },
+    {
+      "epoch": 12.674475955610358,
+      "grad_norm": 0.0538625605404377,
+      "learning_rate": 0.01,
+      "loss": 1.9519,
+      "step": 123348
+    },
+    {
+      "epoch": 12.674784217016029,
+      "grad_norm": 0.04183555766940117,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 123351
+    },
+    {
+      "epoch": 12.675092478421702,
+      "grad_norm": 0.042323462665081024,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 123354
+    },
+    {
+      "epoch": 12.675400739827374,
+      "grad_norm": 0.11391904950141907,
+      "learning_rate": 0.01,
+      "loss": 1.9499,
+      "step": 123357
+    },
+    {
+      "epoch": 12.675709001233045,
+      "grad_norm": 0.1024266704916954,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 123360
+    },
+    {
+      "epoch": 12.676017262638718,
+      "grad_norm": 0.09751450270414352,
+      "learning_rate": 0.01,
+      "loss": 1.9554,
+      "step": 123363
+    },
+    {
+      "epoch": 12.67632552404439,
+      "grad_norm": 0.06176680326461792,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 123366
+    },
+    {
+      "epoch": 12.676633785450061,
+      "grad_norm": 0.06321140378713608,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 123369
+    },
+    {
+      "epoch": 12.676942046855734,
+      "grad_norm": 0.13908283412456512,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 123372
+    },
+    {
+      "epoch": 12.677250308261407,
+      "grad_norm": 0.054892729967832565,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 123375
+    },
+    {
+      "epoch": 12.677558569667077,
+      "grad_norm": 0.05871794372797012,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 123378
+    },
+    {
+      "epoch": 12.67786683107275,
+      "grad_norm": 0.048607565462589264,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 123381
+    },
+    {
+      "epoch": 12.678175092478421,
+      "grad_norm": 0.04668545722961426,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 123384
+    },
+    {
+      "epoch": 12.678483353884094,
+      "grad_norm": 0.05814890190958977,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 123387
+    },
+    {
+      "epoch": 12.678791615289766,
+      "grad_norm": 0.04149680212140083,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 123390
+    },
+    {
+      "epoch": 12.679099876695437,
+      "grad_norm": 0.0385429821908474,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 123393
+    },
+    {
+      "epoch": 12.67940813810111,
+      "grad_norm": 0.07953255623579025,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 123396
+    },
+    {
+      "epoch": 12.679716399506782,
+      "grad_norm": 0.08655081689357758,
+      "learning_rate": 0.01,
+      "loss": 1.9676,
+      "step": 123399
+    },
+    {
+      "epoch": 12.680024660912453,
+      "grad_norm": 0.09242817759513855,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 123402
+    },
+    {
+      "epoch": 12.680332922318126,
+      "grad_norm": 0.06824064254760742,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 123405
+    },
+    {
+      "epoch": 12.680641183723798,
+      "grad_norm": 0.0653814896941185,
+      "learning_rate": 0.01,
+      "loss": 1.9563,
+      "step": 123408
+    },
+    {
+      "epoch": 12.68094944512947,
+      "grad_norm": 0.09969869256019592,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 123411
+    },
+    {
+      "epoch": 12.681257706535142,
+      "grad_norm": 0.061876170337200165,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 123414
+    },
+    {
+      "epoch": 12.681565967940815,
+      "grad_norm": 0.0635654553771019,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 123417
+    },
+    {
+      "epoch": 12.681874229346485,
+      "grad_norm": 0.07676015794277191,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 123420
+    },
+    {
+      "epoch": 12.682182490752158,
+      "grad_norm": 0.0748474970459938,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 123423
+    },
+    {
+      "epoch": 12.682490752157829,
+      "grad_norm": 0.0491030290722847,
+      "learning_rate": 0.01,
+      "loss": 1.9075,
+      "step": 123426
+    },
+    {
+      "epoch": 12.682799013563502,
+      "grad_norm": 0.09475478529930115,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 123429
+    },
+    {
+      "epoch": 12.683107274969174,
+      "grad_norm": 0.060498595237731934,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 123432
+    },
+    {
+      "epoch": 12.683415536374845,
+      "grad_norm": 0.04039103537797928,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 123435
+    },
+    {
+      "epoch": 12.683723797780518,
+      "grad_norm": 0.11218724399805069,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 123438
+    },
+    {
+      "epoch": 12.68403205918619,
+      "grad_norm": 0.11027968674898148,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 123441
+    },
+    {
+      "epoch": 12.684340320591861,
+      "grad_norm": 0.050490401685237885,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 123444
+    },
+    {
+      "epoch": 12.684648581997534,
+      "grad_norm": 0.03872010484337807,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 123447
+    },
+    {
+      "epoch": 12.684956843403207,
+      "grad_norm": 0.03425583988428116,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 123450
+    },
+    {
+      "epoch": 12.685265104808877,
+      "grad_norm": 0.041339047253131866,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 123453
+    },
+    {
+      "epoch": 12.68557336621455,
+      "grad_norm": 0.0473909005522728,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 123456
+    },
+    {
+      "epoch": 12.685881627620223,
+      "grad_norm": 0.057033736258745193,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 123459
+    },
+    {
+      "epoch": 12.686189889025894,
+      "grad_norm": 0.0641959011554718,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 123462
+    },
+    {
+      "epoch": 12.686498150431566,
+      "grad_norm": 0.08935226500034332,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 123465
+    },
+    {
+      "epoch": 12.686806411837239,
+      "grad_norm": 0.0869215652346611,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 123468
+    },
+    {
+      "epoch": 12.68711467324291,
+      "grad_norm": 0.05638349801301956,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 123471
+    },
+    {
+      "epoch": 12.687422934648582,
+      "grad_norm": 0.14322124421596527,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 123474
+    },
+    {
+      "epoch": 12.687731196054255,
+      "grad_norm": 0.09759920090436935,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 123477
+    },
+    {
+      "epoch": 12.688039457459926,
+      "grad_norm": 0.05922066792845726,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 123480
+    },
+    {
+      "epoch": 12.688347718865598,
+      "grad_norm": 0.04995220527052879,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 123483
+    },
+    {
+      "epoch": 12.68865598027127,
+      "grad_norm": 0.043737851083278656,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 123486
+    },
+    {
+      "epoch": 12.688964241676942,
+      "grad_norm": 0.09345188736915588,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 123489
+    },
+    {
+      "epoch": 12.689272503082615,
+      "grad_norm": 0.17204274237155914,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 123492
+    },
+    {
+      "epoch": 12.689580764488285,
+      "grad_norm": 0.09963071346282959,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 123495
+    },
+    {
+      "epoch": 12.689889025893958,
+      "grad_norm": 0.07569556683301926,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 123498
+    },
+    {
+      "epoch": 12.69019728729963,
+      "grad_norm": 0.04148825630545616,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 123501
+    },
+    {
+      "epoch": 12.690505548705302,
+      "grad_norm": 0.06261174380779266,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 123504
+    },
+    {
+      "epoch": 12.690813810110974,
+      "grad_norm": 0.057878635823726654,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 123507
+    },
+    {
+      "epoch": 12.691122071516647,
+      "grad_norm": 0.07493016868829727,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 123510
+    },
+    {
+      "epoch": 12.691430332922318,
+      "grad_norm": 0.04073924198746681,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 123513
+    },
+    {
+      "epoch": 12.69173859432799,
+      "grad_norm": 0.033230170607566833,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 123516
+    },
+    {
+      "epoch": 12.692046855733661,
+      "grad_norm": 0.04627753421664238,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 123519
+    },
+    {
+      "epoch": 12.692355117139334,
+      "grad_norm": 0.04366100952029228,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 123522
+    },
+    {
+      "epoch": 12.692663378545006,
+      "grad_norm": 0.10183031857013702,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 123525
+    },
+    {
+      "epoch": 12.692971639950677,
+      "grad_norm": 0.0781526044011116,
+      "learning_rate": 0.01,
+      "loss": 1.9507,
+      "step": 123528
+    },
+    {
+      "epoch": 12.69327990135635,
+      "grad_norm": 0.11927567422389984,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 123531
+    },
+    {
+      "epoch": 12.693588162762023,
+      "grad_norm": 0.04548252746462822,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 123534
+    },
+    {
+      "epoch": 12.693896424167693,
+      "grad_norm": 0.09688074141740799,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 123537
+    },
+    {
+      "epoch": 12.694204685573366,
+      "grad_norm": 0.042784079909324646,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 123540
+    },
+    {
+      "epoch": 12.694512946979039,
+      "grad_norm": 0.05487271398305893,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 123543
+    },
+    {
+      "epoch": 12.69482120838471,
+      "grad_norm": 0.04063720256090164,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 123546
+    },
+    {
+      "epoch": 12.695129469790382,
+      "grad_norm": 0.04431400075554848,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 123549
+    },
+    {
+      "epoch": 12.695437731196055,
+      "grad_norm": 0.04617541283369064,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 123552
+    },
+    {
+      "epoch": 12.695745992601726,
+      "grad_norm": 0.04244965314865112,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 123555
+    },
+    {
+      "epoch": 12.696054254007398,
+      "grad_norm": 0.058584727346897125,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 123558
+    },
+    {
+      "epoch": 12.696362515413071,
+      "grad_norm": 0.056801773607730865,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 123561
+    },
+    {
+      "epoch": 12.696670776818742,
+      "grad_norm": 0.07578185945749283,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 123564
+    },
+    {
+      "epoch": 12.696979038224415,
+      "grad_norm": 0.09019816666841507,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 123567
+    },
+    {
+      "epoch": 12.697287299630087,
+      "grad_norm": 0.17778441309928894,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 123570
+    },
+    {
+      "epoch": 12.697595561035758,
+      "grad_norm": 0.12761113047599792,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 123573
+    },
+    {
+      "epoch": 12.69790382244143,
+      "grad_norm": 0.08019939064979553,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 123576
+    },
+    {
+      "epoch": 12.698212083847102,
+      "grad_norm": 0.05896785482764244,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 123579
+    },
+    {
+      "epoch": 12.698520345252774,
+      "grad_norm": 0.05120032653212547,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 123582
+    },
+    {
+      "epoch": 12.698828606658447,
+      "grad_norm": 0.04152693226933479,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 123585
+    },
+    {
+      "epoch": 12.699136868064118,
+      "grad_norm": 0.034317269921302795,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 123588
+    },
+    {
+      "epoch": 12.69944512946979,
+      "grad_norm": 0.03374708816409111,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 123591
+    },
+    {
+      "epoch": 12.699753390875463,
+      "grad_norm": 0.05124400556087494,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 123594
+    },
+    {
+      "epoch": 12.700061652281134,
+      "grad_norm": 0.11427992582321167,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 123597
+    },
+    {
+      "epoch": 12.700369913686806,
+      "grad_norm": 0.07133519649505615,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 123600
+    },
+    {
+      "epoch": 12.700678175092479,
+      "grad_norm": 0.11643834412097931,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 123603
+    },
+    {
+      "epoch": 12.70098643649815,
+      "grad_norm": 0.07793018966913223,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 123606
+    },
+    {
+      "epoch": 12.701294697903823,
+      "grad_norm": 0.036932747811079025,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 123609
+    },
+    {
+      "epoch": 12.701602959309495,
+      "grad_norm": 0.03248800337314606,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 123612
+    },
+    {
+      "epoch": 12.701911220715166,
+      "grad_norm": 0.03589079901576042,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 123615
+    },
+    {
+      "epoch": 12.702219482120839,
+      "grad_norm": 0.0629652589559555,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 123618
+    },
+    {
+      "epoch": 12.70252774352651,
+      "grad_norm": 0.07401778548955917,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 123621
+    },
+    {
+      "epoch": 12.702836004932182,
+      "grad_norm": 0.06214204803109169,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 123624
+    },
+    {
+      "epoch": 12.703144266337855,
+      "grad_norm": 0.09164556115865707,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 123627
+    },
+    {
+      "epoch": 12.703452527743526,
+      "grad_norm": 0.05909937992691994,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 123630
+    },
+    {
+      "epoch": 12.703760789149198,
+      "grad_norm": 0.04564003273844719,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 123633
+    },
+    {
+      "epoch": 12.704069050554871,
+      "grad_norm": 0.047293175011873245,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 123636
+    },
+    {
+      "epoch": 12.704377311960542,
+      "grad_norm": 0.09538941085338593,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 123639
+    },
+    {
+      "epoch": 12.704685573366215,
+      "grad_norm": 0.06254024803638458,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 123642
+    },
+    {
+      "epoch": 12.704993834771887,
+      "grad_norm": 0.06853699684143066,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 123645
+    },
+    {
+      "epoch": 12.705302096177558,
+      "grad_norm": 0.09391725808382034,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 123648
+    },
+    {
+      "epoch": 12.70561035758323,
+      "grad_norm": 0.09962349385023117,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 123651
+    },
+    {
+      "epoch": 12.705918618988903,
+      "grad_norm": 0.08067303150892258,
+      "learning_rate": 0.01,
+      "loss": 1.9584,
+      "step": 123654
+    },
+    {
+      "epoch": 12.706226880394574,
+      "grad_norm": 0.04248975217342377,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 123657
+    },
+    {
+      "epoch": 12.706535141800247,
+      "grad_norm": 0.062013447284698486,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 123660
+    },
+    {
+      "epoch": 12.70684340320592,
+      "grad_norm": 0.0528823621571064,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 123663
+    },
+    {
+      "epoch": 12.70715166461159,
+      "grad_norm": 0.058377742767333984,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 123666
+    },
+    {
+      "epoch": 12.707459926017263,
+      "grad_norm": 0.042628709226846695,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 123669
+    },
+    {
+      "epoch": 12.707768187422936,
+      "grad_norm": 0.03786737471818924,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 123672
+    },
+    {
+      "epoch": 12.708076448828606,
+      "grad_norm": 0.03240932896733284,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 123675
+    },
+    {
+      "epoch": 12.708384710234279,
+      "grad_norm": 0.06238216906785965,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 123678
+    },
+    {
+      "epoch": 12.70869297163995,
+      "grad_norm": 0.15355850756168365,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 123681
+    },
+    {
+      "epoch": 12.709001233045623,
+      "grad_norm": 0.06694837659597397,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 123684
+    },
+    {
+      "epoch": 12.709309494451295,
+      "grad_norm": 0.043853238224983215,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 123687
+    },
+    {
+      "epoch": 12.709617755856966,
+      "grad_norm": 0.05808883532881737,
+      "learning_rate": 0.01,
+      "loss": 1.9623,
+      "step": 123690
+    },
+    {
+      "epoch": 12.709926017262639,
+      "grad_norm": 0.05669691041111946,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 123693
+    },
+    {
+      "epoch": 12.710234278668311,
+      "grad_norm": 0.04043015465140343,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 123696
+    },
+    {
+      "epoch": 12.710542540073982,
+      "grad_norm": 0.03637734428048134,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 123699
+    },
+    {
+      "epoch": 12.710850801479655,
+      "grad_norm": 0.03239542618393898,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 123702
+    },
+    {
+      "epoch": 12.711159062885327,
+      "grad_norm": 0.04744594544172287,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 123705
+    },
+    {
+      "epoch": 12.711467324290998,
+      "grad_norm": 0.12527455389499664,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 123708
+    },
+    {
+      "epoch": 12.711775585696671,
+      "grad_norm": 0.039139848202466965,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 123711
+    },
+    {
+      "epoch": 12.712083847102342,
+      "grad_norm": 0.12416050583124161,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 123714
+    },
+    {
+      "epoch": 12.712392108508014,
+      "grad_norm": 0.04227903485298157,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 123717
+    },
+    {
+      "epoch": 12.712700369913687,
+      "grad_norm": 0.05773387476801872,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 123720
+    },
+    {
+      "epoch": 12.713008631319358,
+      "grad_norm": 0.07354748994112015,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 123723
+    },
+    {
+      "epoch": 12.71331689272503,
+      "grad_norm": 0.06634755432605743,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 123726
+    },
+    {
+      "epoch": 12.713625154130703,
+      "grad_norm": 0.03875407576560974,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 123729
+    },
+    {
+      "epoch": 12.713933415536374,
+      "grad_norm": 0.0528351329267025,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 123732
+    },
+    {
+      "epoch": 12.714241676942047,
+      "grad_norm": 0.050535354763269424,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 123735
+    },
+    {
+      "epoch": 12.71454993834772,
+      "grad_norm": 0.056998029351234436,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 123738
+    },
+    {
+      "epoch": 12.71485819975339,
+      "grad_norm": 0.037163373082876205,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 123741
+    },
+    {
+      "epoch": 12.715166461159063,
+      "grad_norm": 0.040953654795885086,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 123744
+    },
+    {
+      "epoch": 12.715474722564736,
+      "grad_norm": 0.09290850162506104,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 123747
+    },
+    {
+      "epoch": 12.715782983970406,
+      "grad_norm": 0.040255237370729446,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 123750
+    },
+    {
+      "epoch": 12.716091245376079,
+      "grad_norm": 0.06343100965023041,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 123753
+    },
+    {
+      "epoch": 12.716399506781752,
+      "grad_norm": 0.049036815762519836,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 123756
+    },
+    {
+      "epoch": 12.716707768187423,
+      "grad_norm": 0.03556976467370987,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 123759
+    },
+    {
+      "epoch": 12.717016029593095,
+      "grad_norm": 0.038292575627565384,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 123762
+    },
+    {
+      "epoch": 12.717324290998768,
+      "grad_norm": 0.05443082004785538,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 123765
+    },
+    {
+      "epoch": 12.717632552404439,
+      "grad_norm": 0.14282263815402985,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 123768
+    },
+    {
+      "epoch": 12.717940813810111,
+      "grad_norm": 0.09722369909286499,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 123771
+    },
+    {
+      "epoch": 12.718249075215782,
+      "grad_norm": 0.04693793132901192,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 123774
+    },
+    {
+      "epoch": 12.718557336621455,
+      "grad_norm": 0.04556123539805412,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 123777
+    },
+    {
+      "epoch": 12.718865598027127,
+      "grad_norm": 0.0539705865085125,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 123780
+    },
+    {
+      "epoch": 12.719173859432798,
+      "grad_norm": 0.08391965180635452,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 123783
+    },
+    {
+      "epoch": 12.719482120838471,
+      "grad_norm": 0.0999051183462143,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 123786
+    },
+    {
+      "epoch": 12.719790382244144,
+      "grad_norm": 0.10758959501981735,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 123789
+    },
+    {
+      "epoch": 12.720098643649814,
+      "grad_norm": 0.04740709438920021,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 123792
+    },
+    {
+      "epoch": 12.720406905055487,
+      "grad_norm": 0.03823184221982956,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 123795
+    },
+    {
+      "epoch": 12.72071516646116,
+      "grad_norm": 0.07436653971672058,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 123798
+    },
+    {
+      "epoch": 12.72102342786683,
+      "grad_norm": 0.04962896183133125,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 123801
+    },
+    {
+      "epoch": 12.721331689272503,
+      "grad_norm": 0.0763697549700737,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 123804
+    },
+    {
+      "epoch": 12.721639950678176,
+      "grad_norm": 0.07073080539703369,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 123807
+    },
+    {
+      "epoch": 12.721948212083847,
+      "grad_norm": 0.07728879153728485,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 123810
+    },
+    {
+      "epoch": 12.72225647348952,
+      "grad_norm": 0.03955766558647156,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 123813
+    },
+    {
+      "epoch": 12.72256473489519,
+      "grad_norm": 0.06398153305053711,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 123816
+    },
+    {
+      "epoch": 12.722872996300863,
+      "grad_norm": 0.0653410479426384,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 123819
+    },
+    {
+      "epoch": 12.723181257706536,
+      "grad_norm": 0.03365045413374901,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 123822
+    },
+    {
+      "epoch": 12.723489519112206,
+      "grad_norm": 0.10590102523565292,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 123825
+    },
+    {
+      "epoch": 12.723797780517879,
+      "grad_norm": 0.05007834732532501,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 123828
+    },
+    {
+      "epoch": 12.724106041923552,
+      "grad_norm": 0.07730080187320709,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 123831
+    },
+    {
+      "epoch": 12.724414303329223,
+      "grad_norm": 0.038783296942710876,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 123834
+    },
+    {
+      "epoch": 12.724722564734895,
+      "grad_norm": 0.028622707352042198,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 123837
+    },
+    {
+      "epoch": 12.725030826140568,
+      "grad_norm": 0.03727482631802559,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 123840
+    },
+    {
+      "epoch": 12.725339087546239,
+      "grad_norm": 0.05076562985777855,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 123843
+    },
+    {
+      "epoch": 12.725647348951911,
+      "grad_norm": 0.14582273364067078,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 123846
+    },
+    {
+      "epoch": 12.725955610357584,
+      "grad_norm": 0.03792782500386238,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 123849
+    },
+    {
+      "epoch": 12.726263871763255,
+      "grad_norm": 0.0417255163192749,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 123852
+    },
+    {
+      "epoch": 12.726572133168927,
+      "grad_norm": 0.12276627868413925,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 123855
+    },
+    {
+      "epoch": 12.7268803945746,
+      "grad_norm": 0.05630599707365036,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 123858
+    },
+    {
+      "epoch": 12.727188655980271,
+      "grad_norm": 0.049009427428245544,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 123861
+    },
+    {
+      "epoch": 12.727496917385944,
+      "grad_norm": 0.060720812529325485,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 123864
+    },
+    {
+      "epoch": 12.727805178791616,
+      "grad_norm": 0.0986635759472847,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 123867
+    },
+    {
+      "epoch": 12.728113440197287,
+      "grad_norm": 0.08428160101175308,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 123870
+    },
+    {
+      "epoch": 12.72842170160296,
+      "grad_norm": 0.053374290466308594,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 123873
+    },
+    {
+      "epoch": 12.72872996300863,
+      "grad_norm": 0.08811751008033752,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 123876
+    },
+    {
+      "epoch": 12.729038224414303,
+      "grad_norm": 0.0746317207813263,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 123879
+    },
+    {
+      "epoch": 12.729346485819976,
+      "grad_norm": 0.08238863945007324,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 123882
+    },
+    {
+      "epoch": 12.729654747225647,
+      "grad_norm": 0.047723446041345596,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 123885
+    },
+    {
+      "epoch": 12.72996300863132,
+      "grad_norm": 0.11083819717168808,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 123888
+    },
+    {
+      "epoch": 12.730271270036992,
+      "grad_norm": 0.06055304408073425,
+      "learning_rate": 0.01,
+      "loss": 1.9714,
+      "step": 123891
+    },
+    {
+      "epoch": 12.730579531442663,
+      "grad_norm": 0.10406802594661713,
+      "learning_rate": 0.01,
+      "loss": 1.9031,
+      "step": 123894
+    },
+    {
+      "epoch": 12.730887792848335,
+      "grad_norm": 0.12105995416641235,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 123897
+    },
+    {
+      "epoch": 12.731196054254008,
+      "grad_norm": 0.050094123929739,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 123900
+    },
+    {
+      "epoch": 12.731504315659679,
+      "grad_norm": 0.054774798452854156,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 123903
+    },
+    {
+      "epoch": 12.731812577065352,
+      "grad_norm": 0.05368786305189133,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 123906
+    },
+    {
+      "epoch": 12.732120838471022,
+      "grad_norm": 0.03528254106640816,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 123909
+    },
+    {
+      "epoch": 12.732429099876695,
+      "grad_norm": 0.04302767664194107,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 123912
+    },
+    {
+      "epoch": 12.732737361282368,
+      "grad_norm": 0.10473338514566422,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 123915
+    },
+    {
+      "epoch": 12.733045622688039,
+      "grad_norm": 0.06902991235256195,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 123918
+    },
+    {
+      "epoch": 12.733353884093711,
+      "grad_norm": 0.06823877990245819,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 123921
+    },
+    {
+      "epoch": 12.733662145499384,
+      "grad_norm": 0.07815282791852951,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 123924
+    },
+    {
+      "epoch": 12.733970406905055,
+      "grad_norm": 0.07565927505493164,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 123927
+    },
+    {
+      "epoch": 12.734278668310727,
+      "grad_norm": 0.04010128229856491,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 123930
+    },
+    {
+      "epoch": 12.7345869297164,
+      "grad_norm": 0.11234023422002792,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 123933
+    },
+    {
+      "epoch": 12.734895191122071,
+      "grad_norm": 0.11952132731676102,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 123936
+    },
+    {
+      "epoch": 12.735203452527744,
+      "grad_norm": 0.07487715035676956,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 123939
+    },
+    {
+      "epoch": 12.735511713933416,
+      "grad_norm": 0.08312923461198807,
+      "learning_rate": 0.01,
+      "loss": 1.9577,
+      "step": 123942
+    },
+    {
+      "epoch": 12.735819975339087,
+      "grad_norm": 0.049645256251096725,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 123945
+    },
+    {
+      "epoch": 12.73612823674476,
+      "grad_norm": 0.0756223052740097,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 123948
+    },
+    {
+      "epoch": 12.736436498150432,
+      "grad_norm": 0.04760962352156639,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 123951
+    },
+    {
+      "epoch": 12.736744759556103,
+      "grad_norm": 0.035297941416502,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 123954
+    },
+    {
+      "epoch": 12.737053020961776,
+      "grad_norm": 0.03172614052891731,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 123957
+    },
+    {
+      "epoch": 12.737361282367448,
+      "grad_norm": 0.03660298138856888,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 123960
+    },
+    {
+      "epoch": 12.73766954377312,
+      "grad_norm": 0.03966713324189186,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 123963
+    },
+    {
+      "epoch": 12.737977805178792,
+      "grad_norm": 0.07121724635362625,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 123966
+    },
+    {
+      "epoch": 12.738286066584463,
+      "grad_norm": 0.06863511353731155,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 123969
+    },
+    {
+      "epoch": 12.738594327990135,
+      "grad_norm": 0.040102001279592514,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 123972
+    },
+    {
+      "epoch": 12.738902589395808,
+      "grad_norm": 0.13240717351436615,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 123975
+    },
+    {
+      "epoch": 12.739210850801479,
+      "grad_norm": 0.08143643289804459,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 123978
+    },
+    {
+      "epoch": 12.739519112207152,
+      "grad_norm": 0.09273307025432587,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 123981
+    },
+    {
+      "epoch": 12.739827373612824,
+      "grad_norm": 0.0630078986287117,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 123984
+    },
+    {
+      "epoch": 12.740135635018495,
+      "grad_norm": 0.0617285780608654,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 123987
+    },
+    {
+      "epoch": 12.740443896424168,
+      "grad_norm": 0.049238421022892,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 123990
+    },
+    {
+      "epoch": 12.74075215782984,
+      "grad_norm": 0.055354706943035126,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 123993
+    },
+    {
+      "epoch": 12.741060419235511,
+      "grad_norm": 0.05170140787959099,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 123996
+    },
+    {
+      "epoch": 12.741368680641184,
+      "grad_norm": 0.04272826388478279,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 123999
+    },
+    {
+      "epoch": 12.741676942046857,
+      "grad_norm": 0.1141861155629158,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 124002
+    },
+    {
+      "epoch": 12.741985203452527,
+      "grad_norm": 0.03754768148064613,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 124005
+    },
+    {
+      "epoch": 12.7422934648582,
+      "grad_norm": 0.04011441394686699,
+      "learning_rate": 0.01,
+      "loss": 1.983,
+      "step": 124008
+    },
+    {
+      "epoch": 12.74260172626387,
+      "grad_norm": 0.13006868958473206,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 124011
+    },
+    {
+      "epoch": 12.742909987669544,
+      "grad_norm": 0.02989177033305168,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 124014
+    },
+    {
+      "epoch": 12.743218249075216,
+      "grad_norm": 0.036035675555467606,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 124017
+    },
+    {
+      "epoch": 12.743526510480887,
+      "grad_norm": 0.0521487332880497,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 124020
+    },
+    {
+      "epoch": 12.74383477188656,
+      "grad_norm": 0.10544262081384659,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 124023
+    },
+    {
+      "epoch": 12.744143033292232,
+      "grad_norm": 0.09543522447347641,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 124026
+    },
+    {
+      "epoch": 12.744451294697903,
+      "grad_norm": 0.048407234251499176,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 124029
+    },
+    {
+      "epoch": 12.744759556103576,
+      "grad_norm": 0.04668554291129112,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 124032
+    },
+    {
+      "epoch": 12.745067817509248,
+      "grad_norm": 0.07983088493347168,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 124035
+    },
+    {
+      "epoch": 12.74537607891492,
+      "grad_norm": 0.09198339283466339,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 124038
+    },
+    {
+      "epoch": 12.745684340320592,
+      "grad_norm": 0.060933612287044525,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 124041
+    },
+    {
+      "epoch": 12.745992601726265,
+      "grad_norm": 0.04717419669032097,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 124044
+    },
+    {
+      "epoch": 12.746300863131935,
+      "grad_norm": 0.04291157424449921,
+      "learning_rate": 0.01,
+      "loss": 1.9017,
+      "step": 124047
+    },
+    {
+      "epoch": 12.746609124537608,
+      "grad_norm": 0.08731654286384583,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 124050
+    },
+    {
+      "epoch": 12.74691738594328,
+      "grad_norm": 0.058710359036922455,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 124053
+    },
+    {
+      "epoch": 12.747225647348952,
+      "grad_norm": 0.05253433808684349,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 124056
+    },
+    {
+      "epoch": 12.747533908754624,
+      "grad_norm": 0.058453548699617386,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 124059
+    },
+    {
+      "epoch": 12.747842170160297,
+      "grad_norm": 0.037066906690597534,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 124062
+    },
+    {
+      "epoch": 12.748150431565968,
+      "grad_norm": 0.03119373321533203,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 124065
+    },
+    {
+      "epoch": 12.74845869297164,
+      "grad_norm": 0.038258105516433716,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 124068
+    },
+    {
+      "epoch": 12.748766954377311,
+      "grad_norm": 0.09073299169540405,
+      "learning_rate": 0.01,
+      "loss": 1.8871,
+      "step": 124071
+    },
+    {
+      "epoch": 12.749075215782984,
+      "grad_norm": 0.09729227423667908,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 124074
+    },
+    {
+      "epoch": 12.749383477188656,
+      "grad_norm": 0.05143263563513756,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 124077
+    },
+    {
+      "epoch": 12.749691738594327,
+      "grad_norm": 0.06965895742177963,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 124080
+    },
+    {
+      "epoch": 12.75,
+      "grad_norm": 0.10010083764791489,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 124083
+    },
+    {
+      "epoch": 12.750308261405673,
+      "grad_norm": 0.060614701360464096,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 124086
+    },
+    {
+      "epoch": 12.750616522811344,
+      "grad_norm": 0.0689188539981842,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 124089
+    },
+    {
+      "epoch": 12.750924784217016,
+      "grad_norm": 0.05061258748173714,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 124092
+    },
+    {
+      "epoch": 12.751233045622689,
+      "grad_norm": 0.12382253259420395,
+      "learning_rate": 0.01,
+      "loss": 1.8836,
+      "step": 124095
+    },
+    {
+      "epoch": 12.75154130702836,
+      "grad_norm": 0.1247350350022316,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 124098
+    },
+    {
+      "epoch": 12.751849568434032,
+      "grad_norm": 0.03914977237582207,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 124101
+    },
+    {
+      "epoch": 12.752157829839703,
+      "grad_norm": 0.09096557646989822,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 124104
+    },
+    {
+      "epoch": 12.752466091245376,
+      "grad_norm": 0.056929972022771835,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 124107
+    },
+    {
+      "epoch": 12.752774352651048,
+      "grad_norm": 0.0355709008872509,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 124110
+    },
+    {
+      "epoch": 12.75308261405672,
+      "grad_norm": 0.0395621620118618,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 124113
+    },
+    {
+      "epoch": 12.753390875462392,
+      "grad_norm": 0.11180397123098373,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 124116
+    },
+    {
+      "epoch": 12.753699136868065,
+      "grad_norm": 0.08548378944396973,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 124119
+    },
+    {
+      "epoch": 12.754007398273735,
+      "grad_norm": 0.042386166751384735,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 124122
+    },
+    {
+      "epoch": 12.754315659679408,
+      "grad_norm": 0.04782962426543236,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 124125
+    },
+    {
+      "epoch": 12.75462392108508,
+      "grad_norm": 0.04724056273698807,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 124128
+    },
+    {
+      "epoch": 12.754932182490752,
+      "grad_norm": 0.11864021420478821,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 124131
+    },
+    {
+      "epoch": 12.755240443896424,
+      "grad_norm": 0.07599248737096786,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 124134
+    },
+    {
+      "epoch": 12.755548705302097,
+      "grad_norm": 0.04107172414660454,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 124137
+    },
+    {
+      "epoch": 12.755856966707768,
+      "grad_norm": 0.13786515593528748,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 124140
+    },
+    {
+      "epoch": 12.75616522811344,
+      "grad_norm": 0.05687346309423447,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 124143
+    },
+    {
+      "epoch": 12.756473489519113,
+      "grad_norm": 0.1486857831478119,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 124146
+    },
+    {
+      "epoch": 12.756781750924784,
+      "grad_norm": 0.05681532993912697,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 124149
+    },
+    {
+      "epoch": 12.757090012330456,
+      "grad_norm": 0.054894618690013885,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 124152
+    },
+    {
+      "epoch": 12.75739827373613,
+      "grad_norm": 0.04187793284654617,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 124155
+    },
+    {
+      "epoch": 12.7577065351418,
+      "grad_norm": 0.052474524825811386,
+      "learning_rate": 0.01,
+      "loss": 1.9559,
+      "step": 124158
+    },
+    {
+      "epoch": 12.758014796547473,
+      "grad_norm": 0.06345690786838531,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 124161
+    },
+    {
+      "epoch": 12.758323057953143,
+      "grad_norm": 0.04370460659265518,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 124164
+    },
+    {
+      "epoch": 12.758631319358816,
+      "grad_norm": 0.033778414130210876,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 124167
+    },
+    {
+      "epoch": 12.758939580764489,
+      "grad_norm": 0.07214024662971497,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 124170
+    },
+    {
+      "epoch": 12.75924784217016,
+      "grad_norm": 0.08046533912420273,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 124173
+    },
+    {
+      "epoch": 12.759556103575832,
+      "grad_norm": 0.1223985031247139,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 124176
+    },
+    {
+      "epoch": 12.759864364981505,
+      "grad_norm": 0.08571730554103851,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 124179
+    },
+    {
+      "epoch": 12.760172626387176,
+      "grad_norm": 0.07246126234531403,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 124182
+    },
+    {
+      "epoch": 12.760480887792848,
+      "grad_norm": 0.06221122667193413,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 124185
+    },
+    {
+      "epoch": 12.760789149198521,
+      "grad_norm": 0.04474806413054466,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 124188
+    },
+    {
+      "epoch": 12.761097410604192,
+      "grad_norm": 0.04277119040489197,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 124191
+    },
+    {
+      "epoch": 12.761405672009865,
+      "grad_norm": 0.05043601989746094,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 124194
+    },
+    {
+      "epoch": 12.761713933415537,
+      "grad_norm": 0.0631195679306984,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 124197
+    },
+    {
+      "epoch": 12.762022194821208,
+      "grad_norm": 0.16869129240512848,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 124200
+    },
+    {
+      "epoch": 12.76233045622688,
+      "grad_norm": 0.07093799114227295,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 124203
+    },
+    {
+      "epoch": 12.762638717632552,
+      "grad_norm": 0.11191149801015854,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 124206
+    },
+    {
+      "epoch": 12.762946979038224,
+      "grad_norm": 0.1054079607129097,
+      "learning_rate": 0.01,
+      "loss": 1.9542,
+      "step": 124209
+    },
+    {
+      "epoch": 12.763255240443897,
+      "grad_norm": 0.08778104186058044,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 124212
+    },
+    {
+      "epoch": 12.763563501849568,
+      "grad_norm": 0.053434088826179504,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 124215
+    },
+    {
+      "epoch": 12.76387176325524,
+      "grad_norm": 0.056754644960165024,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 124218
+    },
+    {
+      "epoch": 12.764180024660913,
+      "grad_norm": 0.030407005921006203,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 124221
+    },
+    {
+      "epoch": 12.764488286066584,
+      "grad_norm": 0.04236002638936043,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 124224
+    },
+    {
+      "epoch": 12.764796547472256,
+      "grad_norm": 0.0928228497505188,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 124227
+    },
+    {
+      "epoch": 12.765104808877929,
+      "grad_norm": 0.05422854796051979,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 124230
+    },
+    {
+      "epoch": 12.7654130702836,
+      "grad_norm": 0.04374406486749649,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 124233
+    },
+    {
+      "epoch": 12.765721331689273,
+      "grad_norm": 0.03411633148789406,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 124236
+    },
+    {
+      "epoch": 12.766029593094945,
+      "grad_norm": 0.03351146727800369,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 124239
+    },
+    {
+      "epoch": 12.766337854500616,
+      "grad_norm": 0.07590338587760925,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 124242
+    },
+    {
+      "epoch": 12.766646115906289,
+      "grad_norm": 0.11562327295541763,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 124245
+    },
+    {
+      "epoch": 12.766954377311961,
+      "grad_norm": 0.10438881814479828,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 124248
+    },
+    {
+      "epoch": 12.767262638717632,
+      "grad_norm": 0.03831292316317558,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 124251
+    },
+    {
+      "epoch": 12.767570900123305,
+      "grad_norm": 0.08192018419504166,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 124254
+    },
+    {
+      "epoch": 12.767879161528978,
+      "grad_norm": 0.060927681624889374,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 124257
+    },
+    {
+      "epoch": 12.768187422934648,
+      "grad_norm": 0.1201568990945816,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 124260
+    },
+    {
+      "epoch": 12.768495684340321,
+      "grad_norm": 0.03554629161953926,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 124263
+    },
+    {
+      "epoch": 12.768803945745992,
+      "grad_norm": 0.0523088201880455,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 124266
+    },
+    {
+      "epoch": 12.769112207151665,
+      "grad_norm": 0.04624674469232559,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 124269
+    },
+    {
+      "epoch": 12.769420468557337,
+      "grad_norm": 0.0483710952103138,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 124272
+    },
+    {
+      "epoch": 12.769728729963008,
+      "grad_norm": 0.06333783268928528,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 124275
+    },
+    {
+      "epoch": 12.77003699136868,
+      "grad_norm": 0.036787934601306915,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 124278
+    },
+    {
+      "epoch": 12.770345252774353,
+      "grad_norm": 0.08958521485328674,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 124281
+    },
+    {
+      "epoch": 12.770653514180024,
+      "grad_norm": 0.05499611794948578,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 124284
+    },
+    {
+      "epoch": 12.770961775585697,
+      "grad_norm": 0.09917958825826645,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 124287
+    },
+    {
+      "epoch": 12.77127003699137,
+      "grad_norm": 0.05552750453352928,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 124290
+    },
+    {
+      "epoch": 12.77157829839704,
+      "grad_norm": 0.11691426485776901,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 124293
+    },
+    {
+      "epoch": 12.771886559802713,
+      "grad_norm": 0.050462447106838226,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 124296
+    },
+    {
+      "epoch": 12.772194821208384,
+      "grad_norm": 0.060034796595573425,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 124299
+    },
+    {
+      "epoch": 12.772503082614056,
+      "grad_norm": 0.052780505269765854,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 124302
+    },
+    {
+      "epoch": 12.772811344019729,
+      "grad_norm": 0.044544320553541183,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 124305
+    },
+    {
+      "epoch": 12.7731196054254,
+      "grad_norm": 0.048675406724214554,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 124308
+    },
+    {
+      "epoch": 12.773427866831073,
+      "grad_norm": 0.03589058294892311,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 124311
+    },
+    {
+      "epoch": 12.773736128236745,
+      "grad_norm": 0.04180203750729561,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 124314
+    },
+    {
+      "epoch": 12.774044389642416,
+      "grad_norm": 0.034893400967121124,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 124317
+    },
+    {
+      "epoch": 12.774352651048089,
+      "grad_norm": 0.03470678627490997,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 124320
+    },
+    {
+      "epoch": 12.774660912453761,
+      "grad_norm": 0.06290122121572495,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 124323
+    },
+    {
+      "epoch": 12.774969173859432,
+      "grad_norm": 0.10993147641420364,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 124326
+    },
+    {
+      "epoch": 12.775277435265105,
+      "grad_norm": 0.045845333486795425,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 124329
+    },
+    {
+      "epoch": 12.775585696670777,
+      "grad_norm": 0.042101263999938965,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 124332
+    },
+    {
+      "epoch": 12.775893958076448,
+      "grad_norm": 0.04230468347668648,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 124335
+    },
+    {
+      "epoch": 12.776202219482121,
+      "grad_norm": 0.0595904216170311,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 124338
+    },
+    {
+      "epoch": 12.776510480887794,
+      "grad_norm": 0.03835110738873482,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 124341
+    },
+    {
+      "epoch": 12.776818742293464,
+      "grad_norm": 0.09960649907588959,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 124344
+    },
+    {
+      "epoch": 12.777127003699137,
+      "grad_norm": 0.04961574450135231,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 124347
+    },
+    {
+      "epoch": 12.77743526510481,
+      "grad_norm": 0.07127340137958527,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 124350
+    },
+    {
+      "epoch": 12.77774352651048,
+      "grad_norm": 0.10124164074659348,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 124353
+    },
+    {
+      "epoch": 12.778051787916153,
+      "grad_norm": 0.0917411595582962,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 124356
+    },
+    {
+      "epoch": 12.778360049321824,
+      "grad_norm": 0.06239021196961403,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 124359
+    },
+    {
+      "epoch": 12.778668310727497,
+      "grad_norm": 0.0381314717233181,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 124362
+    },
+    {
+      "epoch": 12.77897657213317,
+      "grad_norm": 0.060684528201818466,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 124365
+    },
+    {
+      "epoch": 12.77928483353884,
+      "grad_norm": 0.04324139282107353,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 124368
+    },
+    {
+      "epoch": 12.779593094944513,
+      "grad_norm": 0.03981003910303116,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 124371
+    },
+    {
+      "epoch": 12.779901356350186,
+      "grad_norm": 0.044959403574466705,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 124374
+    },
+    {
+      "epoch": 12.780209617755856,
+      "grad_norm": 0.04338802024722099,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 124377
+    },
+    {
+      "epoch": 12.780517879161529,
+      "grad_norm": 0.06274119764566422,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 124380
+    },
+    {
+      "epoch": 12.780826140567202,
+      "grad_norm": 0.10936722904443741,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 124383
+    },
+    {
+      "epoch": 12.781134401972873,
+      "grad_norm": 0.05914433300495148,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 124386
+    },
+    {
+      "epoch": 12.781442663378545,
+      "grad_norm": 0.047597043216228485,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 124389
+    },
+    {
+      "epoch": 12.781750924784218,
+      "grad_norm": 0.03429289162158966,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 124392
+    },
+    {
+      "epoch": 12.782059186189889,
+      "grad_norm": 0.040711697190999985,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 124395
+    },
+    {
+      "epoch": 12.782367447595561,
+      "grad_norm": 0.08311277627944946,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 124398
+    },
+    {
+      "epoch": 12.782675709001232,
+      "grad_norm": 0.1344379186630249,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 124401
+    },
+    {
+      "epoch": 12.782983970406905,
+      "grad_norm": 0.13270977139472961,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 124404
+    },
+    {
+      "epoch": 12.783292231812577,
+      "grad_norm": 0.12600675225257874,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 124407
+    },
+    {
+      "epoch": 12.783600493218248,
+      "grad_norm": 0.052186086773872375,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 124410
+    },
+    {
+      "epoch": 12.783908754623921,
+      "grad_norm": 0.06286251544952393,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 124413
+    },
+    {
+      "epoch": 12.784217016029594,
+      "grad_norm": 0.09777403622865677,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 124416
+    },
+    {
+      "epoch": 12.784525277435264,
+      "grad_norm": 0.03839965909719467,
+      "learning_rate": 0.01,
+      "loss": 1.948,
+      "step": 124419
+    },
+    {
+      "epoch": 12.784833538840937,
+      "grad_norm": 0.06372542679309845,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 124422
+    },
+    {
+      "epoch": 12.78514180024661,
+      "grad_norm": 0.06057681143283844,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 124425
+    },
+    {
+      "epoch": 12.78545006165228,
+      "grad_norm": 0.13042236864566803,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 124428
+    },
+    {
+      "epoch": 12.785758323057953,
+      "grad_norm": 0.1508304327726364,
+      "learning_rate": 0.01,
+      "loss": 1.9605,
+      "step": 124431
+    },
+    {
+      "epoch": 12.786066584463626,
+      "grad_norm": 0.09157309681177139,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 124434
+    },
+    {
+      "epoch": 12.786374845869297,
+      "grad_norm": 0.0452297180891037,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 124437
+    },
+    {
+      "epoch": 12.78668310727497,
+      "grad_norm": 0.11927188187837601,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 124440
+    },
+    {
+      "epoch": 12.786991368680642,
+      "grad_norm": 0.04036303982138634,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 124443
+    },
+    {
+      "epoch": 12.787299630086313,
+      "grad_norm": 0.057232797145843506,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 124446
+    },
+    {
+      "epoch": 12.787607891491986,
+      "grad_norm": 0.058794617652893066,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 124449
+    },
+    {
+      "epoch": 12.787916152897658,
+      "grad_norm": 0.06803694367408752,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 124452
+    },
+    {
+      "epoch": 12.788224414303329,
+      "grad_norm": 0.05735660716891289,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 124455
+    },
+    {
+      "epoch": 12.788532675709002,
+      "grad_norm": 0.059744693338871,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 124458
+    },
+    {
+      "epoch": 12.788840937114673,
+      "grad_norm": 0.05062204971909523,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 124461
+    },
+    {
+      "epoch": 12.789149198520345,
+      "grad_norm": 0.04351005703210831,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 124464
+    },
+    {
+      "epoch": 12.789457459926018,
+      "grad_norm": 0.034482888877391815,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 124467
+    },
+    {
+      "epoch": 12.789765721331689,
+      "grad_norm": 0.13502463698387146,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 124470
+    },
+    {
+      "epoch": 12.790073982737361,
+      "grad_norm": 0.07608455419540405,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 124473
+    },
+    {
+      "epoch": 12.790382244143034,
+      "grad_norm": 0.08391346037387848,
+      "learning_rate": 0.01,
+      "loss": 1.894,
+      "step": 124476
+    },
+    {
+      "epoch": 12.790690505548705,
+      "grad_norm": 0.1317441761493683,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 124479
+    },
+    {
+      "epoch": 12.790998766954377,
+      "grad_norm": 0.13802526891231537,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 124482
+    },
+    {
+      "epoch": 12.79130702836005,
+      "grad_norm": 0.11871980130672455,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 124485
+    },
+    {
+      "epoch": 12.791615289765721,
+      "grad_norm": 0.06530909240245819,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 124488
+    },
+    {
+      "epoch": 12.791923551171394,
+      "grad_norm": 0.06150613725185394,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 124491
+    },
+    {
+      "epoch": 12.792231812577064,
+      "grad_norm": 0.04338151589035988,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 124494
+    },
+    {
+      "epoch": 12.792540073982737,
+      "grad_norm": 0.05015654116868973,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 124497
+    },
+    {
+      "epoch": 12.79284833538841,
+      "grad_norm": 0.0415099561214447,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 124500
+    },
+    {
+      "epoch": 12.79315659679408,
+      "grad_norm": 0.0535995252430439,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 124503
+    },
+    {
+      "epoch": 12.793464858199753,
+      "grad_norm": 0.13396979868412018,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 124506
+    },
+    {
+      "epoch": 12.793773119605426,
+      "grad_norm": 0.04803071543574333,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 124509
+    },
+    {
+      "epoch": 12.794081381011097,
+      "grad_norm": 0.07891876995563507,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 124512
+    },
+    {
+      "epoch": 12.79438964241677,
+      "grad_norm": 0.05627993866801262,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 124515
+    },
+    {
+      "epoch": 12.794697903822442,
+      "grad_norm": 0.10919144004583359,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 124518
+    },
+    {
+      "epoch": 12.795006165228113,
+      "grad_norm": 0.04701576009392738,
+      "learning_rate": 0.01,
+      "loss": 1.9536,
+      "step": 124521
+    },
+    {
+      "epoch": 12.795314426633785,
+      "grad_norm": 0.08555266261100769,
+      "learning_rate": 0.01,
+      "loss": 1.9574,
+      "step": 124524
+    },
+    {
+      "epoch": 12.795622688039458,
+      "grad_norm": 0.042641106992959976,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 124527
+    },
+    {
+      "epoch": 12.795930949445129,
+      "grad_norm": 0.09084159135818481,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 124530
+    },
+    {
+      "epoch": 12.796239210850802,
+      "grad_norm": 0.06366157531738281,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 124533
+    },
+    {
+      "epoch": 12.796547472256474,
+      "grad_norm": 0.031330134719610214,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 124536
+    },
+    {
+      "epoch": 12.796855733662145,
+      "grad_norm": 0.03788667172193527,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 124539
+    },
+    {
+      "epoch": 12.797163995067818,
+      "grad_norm": 0.09419793635606766,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 124542
+    },
+    {
+      "epoch": 12.79747225647349,
+      "grad_norm": 0.061394162476062775,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 124545
+    },
+    {
+      "epoch": 12.797780517879161,
+      "grad_norm": 0.12643764913082123,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 124548
+    },
+    {
+      "epoch": 12.798088779284834,
+      "grad_norm": 0.042745135724544525,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 124551
+    },
+    {
+      "epoch": 12.798397040690505,
+      "grad_norm": 0.03733081370592117,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 124554
+    },
+    {
+      "epoch": 12.798705302096177,
+      "grad_norm": 0.06673793494701385,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 124557
+    },
+    {
+      "epoch": 12.79901356350185,
+      "grad_norm": 0.07428089529275894,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 124560
+    },
+    {
+      "epoch": 12.799321824907521,
+      "grad_norm": 0.17655010521411896,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 124563
+    },
+    {
+      "epoch": 12.799630086313194,
+      "grad_norm": 0.09538134187459946,
+      "learning_rate": 0.01,
+      "loss": 1.9588,
+      "step": 124566
+    },
+    {
+      "epoch": 12.799938347718866,
+      "grad_norm": 0.07083272188901901,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 124569
+    },
+    {
+      "epoch": 12.800246609124537,
+      "grad_norm": 0.10523665696382523,
+      "learning_rate": 0.01,
+      "loss": 1.9529,
+      "step": 124572
+    },
+    {
+      "epoch": 12.80055487053021,
+      "grad_norm": 0.07451805472373962,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 124575
+    },
+    {
+      "epoch": 12.800863131935882,
+      "grad_norm": 0.04025748744606972,
+      "learning_rate": 0.01,
+      "loss": 1.9589,
+      "step": 124578
+    },
+    {
+      "epoch": 12.801171393341553,
+      "grad_norm": 0.03501112014055252,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 124581
+    },
+    {
+      "epoch": 12.801479654747226,
+      "grad_norm": 0.045197147876024246,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 124584
+    },
+    {
+      "epoch": 12.801787916152898,
+      "grad_norm": 0.03901546820998192,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 124587
+    },
+    {
+      "epoch": 12.80209617755857,
+      "grad_norm": 0.043865401297807693,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 124590
+    },
+    {
+      "epoch": 12.802404438964242,
+      "grad_norm": 0.03975968062877655,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 124593
+    },
+    {
+      "epoch": 12.802712700369913,
+      "grad_norm": 0.05233794450759888,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 124596
+    },
+    {
+      "epoch": 12.803020961775585,
+      "grad_norm": 0.049878280609846115,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 124599
+    },
+    {
+      "epoch": 12.803329223181258,
+      "grad_norm": 0.07584605365991592,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 124602
+    },
+    {
+      "epoch": 12.803637484586929,
+      "grad_norm": 0.1152682900428772,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 124605
+    },
+    {
+      "epoch": 12.803945745992602,
+      "grad_norm": 0.0415656641125679,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 124608
+    },
+    {
+      "epoch": 12.804254007398274,
+      "grad_norm": 0.0453491173684597,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 124611
+    },
+    {
+      "epoch": 12.804562268803945,
+      "grad_norm": 0.04551032930612564,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 124614
+    },
+    {
+      "epoch": 12.804870530209618,
+      "grad_norm": 0.04969335347414017,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 124617
+    },
+    {
+      "epoch": 12.80517879161529,
+      "grad_norm": 0.195338174700737,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 124620
+    },
+    {
+      "epoch": 12.805487053020961,
+      "grad_norm": 0.0799102634191513,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 124623
+    },
+    {
+      "epoch": 12.805795314426634,
+      "grad_norm": 0.06339021027088165,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 124626
+    },
+    {
+      "epoch": 12.806103575832307,
+      "grad_norm": 0.0930168554186821,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 124629
+    },
+    {
+      "epoch": 12.806411837237977,
+      "grad_norm": 0.0932261198759079,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 124632
+    },
+    {
+      "epoch": 12.80672009864365,
+      "grad_norm": 0.07090446352958679,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 124635
+    },
+    {
+      "epoch": 12.807028360049323,
+      "grad_norm": 0.07370625436306,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 124638
+    },
+    {
+      "epoch": 12.807336621454994,
+      "grad_norm": 0.057869769632816315,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 124641
+    },
+    {
+      "epoch": 12.807644882860666,
+      "grad_norm": 0.04020111635327339,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 124644
+    },
+    {
+      "epoch": 12.807953144266339,
+      "grad_norm": 0.04522833973169327,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 124647
+    },
+    {
+      "epoch": 12.80826140567201,
+      "grad_norm": 0.0788583904504776,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 124650
+    },
+    {
+      "epoch": 12.808569667077682,
+      "grad_norm": 0.05737563967704773,
+      "learning_rate": 0.01,
+      "loss": 1.9716,
+      "step": 124653
+    },
+    {
+      "epoch": 12.808877928483353,
+      "grad_norm": 0.05475354939699173,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 124656
+    },
+    {
+      "epoch": 12.809186189889026,
+      "grad_norm": 0.04790017381310463,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 124659
+    },
+    {
+      "epoch": 12.809494451294698,
+      "grad_norm": 0.07053875923156738,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 124662
+    },
+    {
+      "epoch": 12.80980271270037,
+      "grad_norm": 0.185977041721344,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 124665
+    },
+    {
+      "epoch": 12.810110974106042,
+      "grad_norm": 0.03858617693185806,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 124668
+    },
+    {
+      "epoch": 12.810419235511715,
+      "grad_norm": 0.044938284903764725,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 124671
+    },
+    {
+      "epoch": 12.810727496917385,
+      "grad_norm": 0.03785492107272148,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 124674
+    },
+    {
+      "epoch": 12.811035758323058,
+      "grad_norm": 0.0343979112803936,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 124677
+    },
+    {
+      "epoch": 12.81134401972873,
+      "grad_norm": 0.0346965529024601,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 124680
+    },
+    {
+      "epoch": 12.811652281134402,
+      "grad_norm": 0.054480504244565964,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 124683
+    },
+    {
+      "epoch": 12.811960542540074,
+      "grad_norm": 0.03773632273077965,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 124686
+    },
+    {
+      "epoch": 12.812268803945745,
+      "grad_norm": 0.10955578088760376,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 124689
+    },
+    {
+      "epoch": 12.812577065351418,
+      "grad_norm": 0.159774050116539,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 124692
+    },
+    {
+      "epoch": 12.81288532675709,
+      "grad_norm": 0.13952012360095978,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 124695
+    },
+    {
+      "epoch": 12.813193588162761,
+      "grad_norm": 0.05918041989207268,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 124698
+    },
+    {
+      "epoch": 12.813501849568434,
+      "grad_norm": 0.04900019243359566,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 124701
+    },
+    {
+      "epoch": 12.813810110974106,
+      "grad_norm": 0.055214982479810715,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 124704
+    },
+    {
+      "epoch": 12.814118372379777,
+      "grad_norm": 0.03945809230208397,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 124707
+    },
+    {
+      "epoch": 12.81442663378545,
+      "grad_norm": 0.07341895997524261,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 124710
+    },
+    {
+      "epoch": 12.814734895191123,
+      "grad_norm": 0.07844100892543793,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 124713
+    },
+    {
+      "epoch": 12.815043156596793,
+      "grad_norm": 0.04210912436246872,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 124716
+    },
+    {
+      "epoch": 12.815351418002466,
+      "grad_norm": 0.04094111919403076,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 124719
+    },
+    {
+      "epoch": 12.815659679408139,
+      "grad_norm": 0.038272298872470856,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 124722
+    },
+    {
+      "epoch": 12.81596794081381,
+      "grad_norm": 0.04964732751250267,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 124725
+    },
+    {
+      "epoch": 12.816276202219482,
+      "grad_norm": 0.16506518423557281,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 124728
+    },
+    {
+      "epoch": 12.816584463625155,
+      "grad_norm": 0.05684073269367218,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 124731
+    },
+    {
+      "epoch": 12.816892725030826,
+      "grad_norm": 0.05659734830260277,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 124734
+    },
+    {
+      "epoch": 12.817200986436498,
+      "grad_norm": 0.04733361303806305,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 124737
+    },
+    {
+      "epoch": 12.817509247842171,
+      "grad_norm": 0.035153668373823166,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 124740
+    },
+    {
+      "epoch": 12.817817509247842,
+      "grad_norm": 0.07535239309072495,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 124743
+    },
+    {
+      "epoch": 12.818125770653515,
+      "grad_norm": 0.07167541980743408,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 124746
+    },
+    {
+      "epoch": 12.818434032059185,
+      "grad_norm": 0.03314826637506485,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 124749
+    },
+    {
+      "epoch": 12.818742293464858,
+      "grad_norm": 0.1441473662853241,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 124752
+    },
+    {
+      "epoch": 12.81905055487053,
+      "grad_norm": 0.10658802092075348,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 124755
+    },
+    {
+      "epoch": 12.819358816276202,
+      "grad_norm": 0.03933339938521385,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 124758
+    },
+    {
+      "epoch": 12.819667077681874,
+      "grad_norm": 0.08993334323167801,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 124761
+    },
+    {
+      "epoch": 12.819975339087547,
+      "grad_norm": 0.08880502730607986,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 124764
+    },
+    {
+      "epoch": 12.820283600493218,
+      "grad_norm": 0.09799978137016296,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 124767
+    },
+    {
+      "epoch": 12.82059186189889,
+      "grad_norm": 0.05939950421452522,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 124770
+    },
+    {
+      "epoch": 12.820900123304563,
+      "grad_norm": 0.036682989448308945,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 124773
+    },
+    {
+      "epoch": 12.821208384710234,
+      "grad_norm": 0.034832343459129333,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 124776
+    },
+    {
+      "epoch": 12.821516646115906,
+      "grad_norm": 0.05991778522729874,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 124779
+    },
+    {
+      "epoch": 12.821824907521579,
+      "grad_norm": 0.10300806164741516,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 124782
+    },
+    {
+      "epoch": 12.82213316892725,
+      "grad_norm": 0.06465953588485718,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 124785
+    },
+    {
+      "epoch": 12.822441430332923,
+      "grad_norm": 0.04358251765370369,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 124788
+    },
+    {
+      "epoch": 12.822749691738593,
+      "grad_norm": 0.091355100274086,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 124791
+    },
+    {
+      "epoch": 12.823057953144266,
+      "grad_norm": 0.0732480138540268,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 124794
+    },
+    {
+      "epoch": 12.823366214549939,
+      "grad_norm": 0.07829828560352325,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 124797
+    },
+    {
+      "epoch": 12.82367447595561,
+      "grad_norm": 0.08272802084684372,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 124800
+    },
+    {
+      "epoch": 12.823982737361282,
+      "grad_norm": 0.10924140363931656,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 124803
+    },
+    {
+      "epoch": 12.824290998766955,
+      "grad_norm": 0.044341158121824265,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 124806
+    },
+    {
+      "epoch": 12.824599260172626,
+      "grad_norm": 0.06467002630233765,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 124809
+    },
+    {
+      "epoch": 12.824907521578298,
+      "grad_norm": 0.09542547166347504,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 124812
+    },
+    {
+      "epoch": 12.825215782983971,
+      "grad_norm": 0.0513361319899559,
+      "learning_rate": 0.01,
+      "loss": 1.9616,
+      "step": 124815
+    },
+    {
+      "epoch": 12.825524044389642,
+      "grad_norm": 0.03738987445831299,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 124818
+    },
+    {
+      "epoch": 12.825832305795315,
+      "grad_norm": 0.04817575588822365,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 124821
+    },
+    {
+      "epoch": 12.826140567200987,
+      "grad_norm": 0.045744240283966064,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 124824
+    },
+    {
+      "epoch": 12.826448828606658,
+      "grad_norm": 0.04282943159341812,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 124827
+    },
+    {
+      "epoch": 12.82675709001233,
+      "grad_norm": 0.05912986397743225,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 124830
+    },
+    {
+      "epoch": 12.827065351418003,
+      "grad_norm": 0.036798182874917984,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 124833
+    },
+    {
+      "epoch": 12.827373612823674,
+      "grad_norm": 0.048573464155197144,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 124836
+    },
+    {
+      "epoch": 12.827681874229347,
+      "grad_norm": 0.12525862455368042,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 124839
+    },
+    {
+      "epoch": 12.827990135635018,
+      "grad_norm": 0.12680740654468536,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 124842
+    },
+    {
+      "epoch": 12.82829839704069,
+      "grad_norm": 0.10475718975067139,
+      "learning_rate": 0.01,
+      "loss": 1.8857,
+      "step": 124845
+    },
+    {
+      "epoch": 12.828606658446363,
+      "grad_norm": 0.03754694387316704,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 124848
+    },
+    {
+      "epoch": 12.828914919852034,
+      "grad_norm": 0.03662050515413284,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 124851
+    },
+    {
+      "epoch": 12.829223181257706,
+      "grad_norm": 0.0618102066218853,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 124854
+    },
+    {
+      "epoch": 12.829531442663379,
+      "grad_norm": 0.03677065297961235,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 124857
+    },
+    {
+      "epoch": 12.82983970406905,
+      "grad_norm": 0.032928451895713806,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 124860
+    },
+    {
+      "epoch": 12.830147965474723,
+      "grad_norm": 0.05401785299181938,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 124863
+    },
+    {
+      "epoch": 12.830456226880395,
+      "grad_norm": 0.08427628874778748,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 124866
+    },
+    {
+      "epoch": 12.830764488286066,
+      "grad_norm": 0.06006439030170441,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 124869
+    },
+    {
+      "epoch": 12.831072749691739,
+      "grad_norm": 0.05316640064120293,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 124872
+    },
+    {
+      "epoch": 12.831381011097411,
+      "grad_norm": 0.0373900830745697,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 124875
+    },
+    {
+      "epoch": 12.831689272503082,
+      "grad_norm": 0.05033502355217934,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 124878
+    },
+    {
+      "epoch": 12.831997533908755,
+      "grad_norm": 0.06176524981856346,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 124881
+    },
+    {
+      "epoch": 12.832305795314426,
+      "grad_norm": 0.038508642464876175,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 124884
+    },
+    {
+      "epoch": 12.832614056720098,
+      "grad_norm": 0.165926992893219,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 124887
+    },
+    {
+      "epoch": 12.832922318125771,
+      "grad_norm": 0.05937818065285683,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 124890
+    },
+    {
+      "epoch": 12.833230579531442,
+      "grad_norm": 0.04595981538295746,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 124893
+    },
+    {
+      "epoch": 12.833538840937115,
+      "grad_norm": 0.04599396139383316,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 124896
+    },
+    {
+      "epoch": 12.833847102342787,
+      "grad_norm": 0.03681658208370209,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 124899
+    },
+    {
+      "epoch": 12.834155363748458,
+      "grad_norm": 0.04873548820614815,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 124902
+    },
+    {
+      "epoch": 12.83446362515413,
+      "grad_norm": 0.04858270660042763,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 124905
+    },
+    {
+      "epoch": 12.834771886559803,
+      "grad_norm": 0.0675751119852066,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 124908
+    },
+    {
+      "epoch": 12.835080147965474,
+      "grad_norm": 0.09886086732149124,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 124911
+    },
+    {
+      "epoch": 12.835388409371147,
+      "grad_norm": 0.08863268047571182,
+      "learning_rate": 0.01,
+      "loss": 1.9518,
+      "step": 124914
+    },
+    {
+      "epoch": 12.83569667077682,
+      "grad_norm": 0.08596405386924744,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 124917
+    },
+    {
+      "epoch": 12.83600493218249,
+      "grad_norm": 0.11636263132095337,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 124920
+    },
+    {
+      "epoch": 12.836313193588163,
+      "grad_norm": 0.07743874937295914,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 124923
+    },
+    {
+      "epoch": 12.836621454993836,
+      "grad_norm": 0.11719123274087906,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 124926
+    },
+    {
+      "epoch": 12.836929716399506,
+      "grad_norm": 0.047170255333185196,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 124929
+    },
+    {
+      "epoch": 12.837237977805179,
+      "grad_norm": 0.10026972740888596,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 124932
+    },
+    {
+      "epoch": 12.837546239210852,
+      "grad_norm": 0.051657043397426605,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 124935
+    },
+    {
+      "epoch": 12.837854500616523,
+      "grad_norm": 0.11636074632406235,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 124938
+    },
+    {
+      "epoch": 12.838162762022195,
+      "grad_norm": 0.03851227089762688,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 124941
+    },
+    {
+      "epoch": 12.838471023427866,
+      "grad_norm": 0.04746010899543762,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 124944
+    },
+    {
+      "epoch": 12.838779284833539,
+      "grad_norm": 0.06377697736024857,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 124947
+    },
+    {
+      "epoch": 12.839087546239211,
+      "grad_norm": 0.06227300688624382,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 124950
+    },
+    {
+      "epoch": 12.839395807644882,
+      "grad_norm": 0.040766604244709015,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 124953
+    },
+    {
+      "epoch": 12.839704069050555,
+      "grad_norm": 0.06182684004306793,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 124956
+    },
+    {
+      "epoch": 12.840012330456227,
+      "grad_norm": 0.04518676921725273,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 124959
+    },
+    {
+      "epoch": 12.840320591861898,
+      "grad_norm": 0.05657326802611351,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 124962
+    },
+    {
+      "epoch": 12.840628853267571,
+      "grad_norm": 0.08930912613868713,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 124965
+    },
+    {
+      "epoch": 12.840937114673244,
+      "grad_norm": 0.05096893385052681,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 124968
+    },
+    {
+      "epoch": 12.841245376078914,
+      "grad_norm": 0.08593285828828812,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 124971
+    },
+    {
+      "epoch": 12.841553637484587,
+      "grad_norm": 0.03875938430428505,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 124974
+    },
+    {
+      "epoch": 12.84186189889026,
+      "grad_norm": 0.11568406969308853,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 124977
+    },
+    {
+      "epoch": 12.84217016029593,
+      "grad_norm": 0.08995538204908371,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 124980
+    },
+    {
+      "epoch": 12.842478421701603,
+      "grad_norm": 0.06632540374994278,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 124983
+    },
+    {
+      "epoch": 12.842786683107274,
+      "grad_norm": 0.06590662896633148,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 124986
+    },
+    {
+      "epoch": 12.843094944512947,
+      "grad_norm": 0.16798827052116394,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 124989
+    },
+    {
+      "epoch": 12.84340320591862,
+      "grad_norm": 0.13125044107437134,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 124992
+    },
+    {
+      "epoch": 12.84371146732429,
+      "grad_norm": 0.05619801953434944,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 124995
+    },
+    {
+      "epoch": 12.844019728729963,
+      "grad_norm": 0.03860598802566528,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 124998
+    },
+    {
+      "epoch": 12.844327990135636,
+      "grad_norm": 0.044260572642087936,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 125001
+    },
+    {
+      "epoch": 12.844636251541306,
+      "grad_norm": 0.04982342943549156,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 125004
+    },
+    {
+      "epoch": 12.844944512946979,
+      "grad_norm": 0.08060560375452042,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 125007
+    },
+    {
+      "epoch": 12.845252774352652,
+      "grad_norm": 0.043235816061496735,
+      "learning_rate": 0.01,
+      "loss": 1.901,
+      "step": 125010
+    },
+    {
+      "epoch": 12.845561035758323,
+      "grad_norm": 0.05255037173628807,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 125013
+    },
+    {
+      "epoch": 12.845869297163995,
+      "grad_norm": 0.0638473629951477,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 125016
+    },
+    {
+      "epoch": 12.846177558569668,
+      "grad_norm": 0.038086581975221634,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 125019
+    },
+    {
+      "epoch": 12.846485819975339,
+      "grad_norm": 0.03848306089639664,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 125022
+    },
+    {
+      "epoch": 12.846794081381011,
+      "grad_norm": 0.08049533516168594,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 125025
+    },
+    {
+      "epoch": 12.847102342786684,
+      "grad_norm": 0.15486104786396027,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 125028
+    },
+    {
+      "epoch": 12.847410604192355,
+      "grad_norm": 0.07169552147388458,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 125031
+    },
+    {
+      "epoch": 12.847718865598027,
+      "grad_norm": 0.040113043040037155,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 125034
+    },
+    {
+      "epoch": 12.848027127003698,
+      "grad_norm": 0.04267123341560364,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 125037
+    },
+    {
+      "epoch": 12.848335388409371,
+      "grad_norm": 0.047227632254362106,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 125040
+    },
+    {
+      "epoch": 12.848643649815044,
+      "grad_norm": 0.03078641928732395,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 125043
+    },
+    {
+      "epoch": 12.848951911220714,
+      "grad_norm": 0.05143728852272034,
+      "learning_rate": 0.01,
+      "loss": 1.9664,
+      "step": 125046
+    },
+    {
+      "epoch": 12.849260172626387,
+      "grad_norm": 0.04470685124397278,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 125049
+    },
+    {
+      "epoch": 12.84956843403206,
+      "grad_norm": 0.06949608027935028,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 125052
+    },
+    {
+      "epoch": 12.84987669543773,
+      "grad_norm": 0.06069021672010422,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 125055
+    },
+    {
+      "epoch": 12.850184956843403,
+      "grad_norm": 0.05496753752231598,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 125058
+    },
+    {
+      "epoch": 12.850493218249076,
+      "grad_norm": 0.04509497433900833,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 125061
+    },
+    {
+      "epoch": 12.850801479654747,
+      "grad_norm": 0.03167354315519333,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 125064
+    },
+    {
+      "epoch": 12.85110974106042,
+      "grad_norm": 0.033969614654779434,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 125067
+    },
+    {
+      "epoch": 12.851418002466092,
+      "grad_norm": 0.06193241477012634,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 125070
+    },
+    {
+      "epoch": 12.851726263871763,
+      "grad_norm": 0.037595510482788086,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 125073
+    },
+    {
+      "epoch": 12.852034525277436,
+      "grad_norm": 0.04467674344778061,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 125076
+    },
+    {
+      "epoch": 12.852342786683106,
+      "grad_norm": 0.04891891032457352,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 125079
+    },
+    {
+      "epoch": 12.852651048088779,
+      "grad_norm": 0.1778205782175064,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 125082
+    },
+    {
+      "epoch": 12.852959309494452,
+      "grad_norm": 0.139647975564003,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 125085
+    },
+    {
+      "epoch": 12.853267570900123,
+      "grad_norm": 0.06531763821840286,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 125088
+    },
+    {
+      "epoch": 12.853575832305795,
+      "grad_norm": 0.08372994512319565,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 125091
+    },
+    {
+      "epoch": 12.853884093711468,
+      "grad_norm": 0.07141024619340897,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 125094
+    },
+    {
+      "epoch": 12.854192355117139,
+      "grad_norm": 0.06588217616081238,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 125097
+    },
+    {
+      "epoch": 12.854500616522811,
+      "grad_norm": 0.059647466987371445,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 125100
+    },
+    {
+      "epoch": 12.854808877928484,
+      "grad_norm": 0.04267524182796478,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 125103
+    },
+    {
+      "epoch": 12.855117139334155,
+      "grad_norm": 0.036257755011320114,
+      "learning_rate": 0.01,
+      "loss": 1.9467,
+      "step": 125106
+    },
+    {
+      "epoch": 12.855425400739827,
+      "grad_norm": 0.07258643954992294,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 125109
+    },
+    {
+      "epoch": 12.8557336621455,
+      "grad_norm": 0.032383304089307785,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 125112
+    },
+    {
+      "epoch": 12.856041923551171,
+      "grad_norm": 0.08107122778892517,
+      "learning_rate": 0.01,
+      "loss": 1.9514,
+      "step": 125115
+    },
+    {
+      "epoch": 12.856350184956844,
+      "grad_norm": 0.09870468825101852,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 125118
+    },
+    {
+      "epoch": 12.856658446362516,
+      "grad_norm": 0.1148717999458313,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 125121
+    },
+    {
+      "epoch": 12.856966707768187,
+      "grad_norm": 0.04018497094511986,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 125124
+    },
+    {
+      "epoch": 12.85727496917386,
+      "grad_norm": 0.07041523605585098,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 125127
+    },
+    {
+      "epoch": 12.857583230579532,
+      "grad_norm": 0.03497825562953949,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 125130
+    },
+    {
+      "epoch": 12.857891491985203,
+      "grad_norm": 0.10776429623365402,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 125133
+    },
+    {
+      "epoch": 12.858199753390876,
+      "grad_norm": 0.04052712395787239,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 125136
+    },
+    {
+      "epoch": 12.858508014796547,
+      "grad_norm": 0.07829517126083374,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 125139
+    },
+    {
+      "epoch": 12.85881627620222,
+      "grad_norm": 0.06916956603527069,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 125142
+    },
+    {
+      "epoch": 12.859124537607892,
+      "grad_norm": 0.037663985043764114,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 125145
+    },
+    {
+      "epoch": 12.859432799013563,
+      "grad_norm": 0.11586187034845352,
+      "learning_rate": 0.01,
+      "loss": 1.8948,
+      "step": 125148
+    },
+    {
+      "epoch": 12.859741060419235,
+      "grad_norm": 0.07533598691225052,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 125151
+    },
+    {
+      "epoch": 12.860049321824908,
+      "grad_norm": 0.04899691045284271,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 125154
+    },
+    {
+      "epoch": 12.860357583230579,
+      "grad_norm": 0.0471162348985672,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 125157
+    },
+    {
+      "epoch": 12.860665844636252,
+      "grad_norm": 0.03868661820888519,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 125160
+    },
+    {
+      "epoch": 12.860974106041924,
+      "grad_norm": 0.05307555943727493,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 125163
+    },
+    {
+      "epoch": 12.861282367447595,
+      "grad_norm": 0.05460083484649658,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 125166
+    },
+    {
+      "epoch": 12.861590628853268,
+      "grad_norm": 0.12687717378139496,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 125169
+    },
+    {
+      "epoch": 12.86189889025894,
+      "grad_norm": 0.07197766751050949,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 125172
+    },
+    {
+      "epoch": 12.862207151664611,
+      "grad_norm": 0.08465447276830673,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 125175
+    },
+    {
+      "epoch": 12.862515413070284,
+      "grad_norm": 0.0753878802061081,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 125178
+    },
+    {
+      "epoch": 12.862823674475955,
+      "grad_norm": 0.08799479156732559,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 125181
+    },
+    {
+      "epoch": 12.863131935881627,
+      "grad_norm": 0.03935414180159569,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 125184
+    },
+    {
+      "epoch": 12.8634401972873,
+      "grad_norm": 0.04997342824935913,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 125187
+    },
+    {
+      "epoch": 12.863748458692971,
+      "grad_norm": 0.12483391165733337,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 125190
+    },
+    {
+      "epoch": 12.864056720098644,
+      "grad_norm": 0.03834403306245804,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 125193
+    },
+    {
+      "epoch": 12.864364981504316,
+      "grad_norm": 0.038831330835819244,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 125196
+    },
+    {
+      "epoch": 12.864673242909987,
+      "grad_norm": 0.05377119407057762,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 125199
+    },
+    {
+      "epoch": 12.86498150431566,
+      "grad_norm": 0.0915791466832161,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 125202
+    },
+    {
+      "epoch": 12.865289765721332,
+      "grad_norm": 0.03884097561240196,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 125205
+    },
+    {
+      "epoch": 12.865598027127003,
+      "grad_norm": 0.03380714729428291,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 125208
+    },
+    {
+      "epoch": 12.865906288532676,
+      "grad_norm": 0.0418735109269619,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 125211
+    },
+    {
+      "epoch": 12.866214549938348,
+      "grad_norm": 0.041301846504211426,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 125214
+    },
+    {
+      "epoch": 12.86652281134402,
+      "grad_norm": 0.0551394484937191,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 125217
+    },
+    {
+      "epoch": 12.866831072749692,
+      "grad_norm": 0.04823874682188034,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 125220
+    },
+    {
+      "epoch": 12.867139334155365,
+      "grad_norm": 0.10362028330564499,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 125223
+    },
+    {
+      "epoch": 12.867447595561035,
+      "grad_norm": 0.06781326979398727,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 125226
+    },
+    {
+      "epoch": 12.867755856966708,
+      "grad_norm": 0.06294917315244675,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 125229
+    },
+    {
+      "epoch": 12.868064118372379,
+      "grad_norm": 0.060097068548202515,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 125232
+    },
+    {
+      "epoch": 12.868372379778052,
+      "grad_norm": 0.04319571703672409,
+      "learning_rate": 0.01,
+      "loss": 1.9533,
+      "step": 125235
+    },
+    {
+      "epoch": 12.868680641183724,
+      "grad_norm": 0.03677972033619881,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 125238
+    },
+    {
+      "epoch": 12.868988902589395,
+      "grad_norm": 0.12327895313501358,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 125241
+    },
+    {
+      "epoch": 12.869297163995068,
+      "grad_norm": 0.05167968198657036,
+      "learning_rate": 0.01,
+      "loss": 1.9431,
+      "step": 125244
+    },
+    {
+      "epoch": 12.86960542540074,
+      "grad_norm": 0.047148581594228745,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 125247
+    },
+    {
+      "epoch": 12.869913686806411,
+      "grad_norm": 0.08654278516769409,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 125250
+    },
+    {
+      "epoch": 12.870221948212084,
+      "grad_norm": 0.06045467033982277,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 125253
+    },
+    {
+      "epoch": 12.870530209617757,
+      "grad_norm": 0.09662190079689026,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 125256
+    },
+    {
+      "epoch": 12.870838471023427,
+      "grad_norm": 0.08902593702077866,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 125259
+    },
+    {
+      "epoch": 12.8711467324291,
+      "grad_norm": 0.042208779603242874,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 125262
+    },
+    {
+      "epoch": 12.871454993834773,
+      "grad_norm": 0.1159537062048912,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 125265
+    },
+    {
+      "epoch": 12.871763255240444,
+      "grad_norm": 0.10170702636241913,
+      "learning_rate": 0.01,
+      "loss": 1.905,
+      "step": 125268
+    },
+    {
+      "epoch": 12.872071516646116,
+      "grad_norm": 0.0402146615087986,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 125271
+    },
+    {
+      "epoch": 12.872379778051787,
+      "grad_norm": 0.12195990979671478,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 125274
+    },
+    {
+      "epoch": 12.87268803945746,
+      "grad_norm": 0.049094632267951965,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 125277
+    },
+    {
+      "epoch": 12.872996300863132,
+      "grad_norm": 0.044050734490156174,
+      "learning_rate": 0.01,
+      "loss": 1.9535,
+      "step": 125280
+    },
+    {
+      "epoch": 12.873304562268803,
+      "grad_norm": 0.040662892162799835,
+      "learning_rate": 0.01,
+      "loss": 1.9423,
+      "step": 125283
+    },
+    {
+      "epoch": 12.873612823674476,
+      "grad_norm": 0.03499949350953102,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 125286
+    },
+    {
+      "epoch": 12.873921085080148,
+      "grad_norm": 0.054730553179979324,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 125289
+    },
+    {
+      "epoch": 12.87422934648582,
+      "grad_norm": 0.04403527081012726,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 125292
+    },
+    {
+      "epoch": 12.874537607891492,
+      "grad_norm": 0.037161070853471756,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 125295
+    },
+    {
+      "epoch": 12.874845869297165,
+      "grad_norm": 0.03912176191806793,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 125298
+    },
+    {
+      "epoch": 12.875154130702835,
+      "grad_norm": 0.09912096709012985,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 125301
+    },
+    {
+      "epoch": 12.875462392108508,
+      "grad_norm": 0.10517314821481705,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 125304
+    },
+    {
+      "epoch": 12.87577065351418,
+      "grad_norm": 0.08175603300333023,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 125307
+    },
+    {
+      "epoch": 12.876078914919852,
+      "grad_norm": 0.04346807673573494,
+      "learning_rate": 0.01,
+      "loss": 1.9758,
+      "step": 125310
+    },
+    {
+      "epoch": 12.876387176325524,
+      "grad_norm": 0.07453333586454391,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 125313
+    },
+    {
+      "epoch": 12.876695437731197,
+      "grad_norm": 0.04787507653236389,
+      "learning_rate": 0.01,
+      "loss": 1.9547,
+      "step": 125316
+    },
+    {
+      "epoch": 12.877003699136868,
+      "grad_norm": 0.03218242898583412,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 125319
+    },
+    {
+      "epoch": 12.87731196054254,
+      "grad_norm": 0.055315811187028885,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 125322
+    },
+    {
+      "epoch": 12.877620221948213,
+      "grad_norm": 0.08303569257259369,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 125325
+    },
+    {
+      "epoch": 12.877928483353884,
+      "grad_norm": 0.07113473862409592,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 125328
+    },
+    {
+      "epoch": 12.878236744759556,
+      "grad_norm": 0.06628505885601044,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 125331
+    },
+    {
+      "epoch": 12.878545006165227,
+      "grad_norm": 0.06445708870887756,
+      "learning_rate": 0.01,
+      "loss": 1.9662,
+      "step": 125334
+    },
+    {
+      "epoch": 12.8788532675709,
+      "grad_norm": 0.05137239024043083,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 125337
+    },
+    {
+      "epoch": 12.879161528976573,
+      "grad_norm": 0.034032564610242844,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 125340
+    },
+    {
+      "epoch": 12.879469790382243,
+      "grad_norm": 0.05968916416168213,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 125343
+    },
+    {
+      "epoch": 12.879778051787916,
+      "grad_norm": 0.14465594291687012,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 125346
+    },
+    {
+      "epoch": 12.880086313193589,
+      "grad_norm": 0.06443560123443604,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 125349
+    },
+    {
+      "epoch": 12.88039457459926,
+      "grad_norm": 0.08430998027324677,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 125352
+    },
+    {
+      "epoch": 12.880702836004932,
+      "grad_norm": 0.04233575984835625,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 125355
+    },
+    {
+      "epoch": 12.881011097410605,
+      "grad_norm": 0.10589230060577393,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 125358
+    },
+    {
+      "epoch": 12.881319358816276,
+      "grad_norm": 0.1286308318376541,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 125361
+    },
+    {
+      "epoch": 12.881627620221948,
+      "grad_norm": 0.09348342567682266,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 125364
+    },
+    {
+      "epoch": 12.881935881627621,
+      "grad_norm": 0.09305001050233841,
+      "learning_rate": 0.01,
+      "loss": 1.9639,
+      "step": 125367
+    },
+    {
+      "epoch": 12.882244143033292,
+      "grad_norm": 0.07719358056783676,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 125370
+    },
+    {
+      "epoch": 12.882552404438965,
+      "grad_norm": 0.04221201315522194,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 125373
+    },
+    {
+      "epoch": 12.882860665844635,
+      "grad_norm": 0.03923868015408516,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 125376
+    },
+    {
+      "epoch": 12.883168927250308,
+      "grad_norm": 0.056949593126773834,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 125379
+    },
+    {
+      "epoch": 12.88347718865598,
+      "grad_norm": 0.05790695548057556,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 125382
+    },
+    {
+      "epoch": 12.883785450061652,
+      "grad_norm": 0.12621572613716125,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 125385
+    },
+    {
+      "epoch": 12.884093711467324,
+      "grad_norm": 0.08982403576374054,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 125388
+    },
+    {
+      "epoch": 12.884401972872997,
+      "grad_norm": 0.07278390973806381,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 125391
+    },
+    {
+      "epoch": 12.884710234278668,
+      "grad_norm": 0.04813414439558983,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 125394
+    },
+    {
+      "epoch": 12.88501849568434,
+      "grad_norm": 0.04034329950809479,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 125397
+    },
+    {
+      "epoch": 12.885326757090013,
+      "grad_norm": 0.03731022775173187,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 125400
+    },
+    {
+      "epoch": 12.885635018495684,
+      "grad_norm": 0.03327002376317978,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 125403
+    },
+    {
+      "epoch": 12.885943279901356,
+      "grad_norm": 0.04345870018005371,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 125406
+    },
+    {
+      "epoch": 12.886251541307029,
+      "grad_norm": 0.2377752810716629,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 125409
+    },
+    {
+      "epoch": 12.8865598027127,
+      "grad_norm": 0.2373015135526657,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 125412
+    },
+    {
+      "epoch": 12.886868064118373,
+      "grad_norm": 0.13731715083122253,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 125415
+    },
+    {
+      "epoch": 12.887176325524045,
+      "grad_norm": 0.06976892054080963,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 125418
+    },
+    {
+      "epoch": 12.887484586929716,
+      "grad_norm": 0.03481016680598259,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 125421
+    },
+    {
+      "epoch": 12.887792848335389,
+      "grad_norm": 0.03333835303783417,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 125424
+    },
+    {
+      "epoch": 12.88810110974106,
+      "grad_norm": 0.039861682802438736,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 125427
+    },
+    {
+      "epoch": 12.888409371146732,
+      "grad_norm": 0.03844046965241432,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 125430
+    },
+    {
+      "epoch": 12.888717632552405,
+      "grad_norm": 0.0711275041103363,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 125433
+    },
+    {
+      "epoch": 12.889025893958076,
+      "grad_norm": 0.039518602192401886,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 125436
+    },
+    {
+      "epoch": 12.889334155363748,
+      "grad_norm": 0.0478103831410408,
+      "learning_rate": 0.01,
+      "loss": 1.9005,
+      "step": 125439
+    },
+    {
+      "epoch": 12.889642416769421,
+      "grad_norm": 0.04707792401313782,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 125442
+    },
+    {
+      "epoch": 12.889950678175092,
+      "grad_norm": 0.03777766972780228,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 125445
+    },
+    {
+      "epoch": 12.890258939580765,
+      "grad_norm": 0.0661175474524498,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 125448
+    },
+    {
+      "epoch": 12.890567200986437,
+      "grad_norm": 0.09127342700958252,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 125451
+    },
+    {
+      "epoch": 12.890875462392108,
+      "grad_norm": 0.06170405074954033,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 125454
+    },
+    {
+      "epoch": 12.89118372379778,
+      "grad_norm": 0.08631531894207001,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 125457
+    },
+    {
+      "epoch": 12.891491985203453,
+      "grad_norm": 0.08351920545101166,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 125460
+    },
+    {
+      "epoch": 12.891800246609124,
+      "grad_norm": 0.09658841788768768,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 125463
+    },
+    {
+      "epoch": 12.892108508014797,
+      "grad_norm": 0.04116319864988327,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 125466
+    },
+    {
+      "epoch": 12.892416769420468,
+      "grad_norm": 0.06576710194349289,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 125469
+    },
+    {
+      "epoch": 12.89272503082614,
+      "grad_norm": 0.08365806192159653,
+      "learning_rate": 0.01,
+      "loss": 1.9632,
+      "step": 125472
+    },
+    {
+      "epoch": 12.893033292231813,
+      "grad_norm": 0.0618315227329731,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 125475
+    },
+    {
+      "epoch": 12.893341553637484,
+      "grad_norm": 0.06475695967674255,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 125478
+    },
+    {
+      "epoch": 12.893649815043156,
+      "grad_norm": 0.04547294229269028,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 125481
+    },
+    {
+      "epoch": 12.893958076448829,
+      "grad_norm": 0.04905145242810249,
+      "learning_rate": 0.01,
+      "loss": 1.9613,
+      "step": 125484
+    },
+    {
+      "epoch": 12.8942663378545,
+      "grad_norm": 0.03583730012178421,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 125487
+    },
+    {
+      "epoch": 12.894574599260173,
+      "grad_norm": 0.03644542396068573,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 125490
+    },
+    {
+      "epoch": 12.894882860665845,
+      "grad_norm": 0.04800483584403992,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 125493
+    },
+    {
+      "epoch": 12.895191122071516,
+      "grad_norm": 0.06762004643678665,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 125496
+    },
+    {
+      "epoch": 12.895499383477189,
+      "grad_norm": 0.08553200960159302,
+      "learning_rate": 0.01,
+      "loss": 1.9041,
+      "step": 125499
+    },
+    {
+      "epoch": 12.895807644882861,
+      "grad_norm": 0.09297972172498703,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 125502
+    },
+    {
+      "epoch": 12.896115906288532,
+      "grad_norm": 0.036268945783376694,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 125505
+    },
+    {
+      "epoch": 12.896424167694205,
+      "grad_norm": 0.07552669942378998,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 125508
+    },
+    {
+      "epoch": 12.896732429099877,
+      "grad_norm": 0.043082501739263535,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 125511
+    },
+    {
+      "epoch": 12.897040690505548,
+      "grad_norm": 0.03717859089374542,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 125514
+    },
+    {
+      "epoch": 12.897348951911221,
+      "grad_norm": 0.13602522015571594,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 125517
+    },
+    {
+      "epoch": 12.897657213316894,
+      "grad_norm": 0.06193123012781143,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 125520
+    },
+    {
+      "epoch": 12.897965474722564,
+      "grad_norm": 0.12001435458660126,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 125523
+    },
+    {
+      "epoch": 12.898273736128237,
+      "grad_norm": 0.06117141991853714,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 125526
+    },
+    {
+      "epoch": 12.898581997533908,
+      "grad_norm": 0.18860256671905518,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 125529
+    },
+    {
+      "epoch": 12.89889025893958,
+      "grad_norm": 0.046133674681186676,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 125532
+    },
+    {
+      "epoch": 12.899198520345253,
+      "grad_norm": 0.034373749047517776,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 125535
+    },
+    {
+      "epoch": 12.899506781750924,
+      "grad_norm": 0.06683547049760818,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 125538
+    },
+    {
+      "epoch": 12.899815043156597,
+      "grad_norm": 0.08050347864627838,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 125541
+    },
+    {
+      "epoch": 12.90012330456227,
+      "grad_norm": 0.08024398982524872,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 125544
+    },
+    {
+      "epoch": 12.90043156596794,
+      "grad_norm": 0.07231173664331436,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 125547
+    },
+    {
+      "epoch": 12.900739827373613,
+      "grad_norm": 0.10628146678209305,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 125550
+    },
+    {
+      "epoch": 12.901048088779286,
+      "grad_norm": 0.03679117560386658,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 125553
+    },
+    {
+      "epoch": 12.901356350184956,
+      "grad_norm": 0.08002956211566925,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 125556
+    },
+    {
+      "epoch": 12.901664611590629,
+      "grad_norm": 0.1332220882177353,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 125559
+    },
+    {
+      "epoch": 12.901972872996302,
+      "grad_norm": 0.0669090673327446,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 125562
+    },
+    {
+      "epoch": 12.902281134401973,
+      "grad_norm": 0.07840543985366821,
+      "learning_rate": 0.01,
+      "loss": 1.9741,
+      "step": 125565
+    },
+    {
+      "epoch": 12.902589395807645,
+      "grad_norm": 0.047904230654239655,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 125568
+    },
+    {
+      "epoch": 12.902897657213316,
+      "grad_norm": 0.05615457519888878,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 125571
+    },
+    {
+      "epoch": 12.903205918618989,
+      "grad_norm": 0.10385458916425705,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 125574
+    },
+    {
+      "epoch": 12.903514180024661,
+      "grad_norm": 0.08210942894220352,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 125577
+    },
+    {
+      "epoch": 12.903822441430332,
+      "grad_norm": 0.05919887498021126,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 125580
+    },
+    {
+      "epoch": 12.904130702836005,
+      "grad_norm": 0.08614572137594223,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 125583
+    },
+    {
+      "epoch": 12.904438964241677,
+      "grad_norm": 0.040114693343639374,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 125586
+    },
+    {
+      "epoch": 12.904747225647348,
+      "grad_norm": 0.06311607360839844,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 125589
+    },
+    {
+      "epoch": 12.905055487053021,
+      "grad_norm": 0.06587512791156769,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 125592
+    },
+    {
+      "epoch": 12.905363748458694,
+      "grad_norm": 0.03994138911366463,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 125595
+    },
+    {
+      "epoch": 12.905672009864364,
+      "grad_norm": 0.04853265359997749,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 125598
+    },
+    {
+      "epoch": 12.905980271270037,
+      "grad_norm": 0.09557167440652847,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 125601
+    },
+    {
+      "epoch": 12.90628853267571,
+      "grad_norm": 0.05322464928030968,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 125604
+    },
+    {
+      "epoch": 12.90659679408138,
+      "grad_norm": 0.05559983104467392,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 125607
+    },
+    {
+      "epoch": 12.906905055487053,
+      "grad_norm": 0.035906389355659485,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 125610
+    },
+    {
+      "epoch": 12.907213316892726,
+      "grad_norm": 0.05089879781007767,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 125613
+    },
+    {
+      "epoch": 12.907521578298397,
+      "grad_norm": 0.0521286278963089,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 125616
+    },
+    {
+      "epoch": 12.90782983970407,
+      "grad_norm": 0.060793764889240265,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 125619
+    },
+    {
+      "epoch": 12.90813810110974,
+      "grad_norm": 0.1368797868490219,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 125622
+    },
+    {
+      "epoch": 12.908446362515413,
+      "grad_norm": 0.05137024074792862,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 125625
+    },
+    {
+      "epoch": 12.908754623921086,
+      "grad_norm": 0.05333459749817848,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 125628
+    },
+    {
+      "epoch": 12.909062885326756,
+      "grad_norm": 0.04691629111766815,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 125631
+    },
+    {
+      "epoch": 12.909371146732429,
+      "grad_norm": 0.047596320509910583,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 125634
+    },
+    {
+      "epoch": 12.909679408138102,
+      "grad_norm": 0.1054455116391182,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 125637
+    },
+    {
+      "epoch": 12.909987669543773,
+      "grad_norm": 0.06832266598939896,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 125640
+    },
+    {
+      "epoch": 12.910295930949445,
+      "grad_norm": 0.09326568990945816,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 125643
+    },
+    {
+      "epoch": 12.910604192355118,
+      "grad_norm": 0.04701947420835495,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 125646
+    },
+    {
+      "epoch": 12.910912453760789,
+      "grad_norm": 0.07189621031284332,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 125649
+    },
+    {
+      "epoch": 12.911220715166461,
+      "grad_norm": 0.07685680687427521,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 125652
+    },
+    {
+      "epoch": 12.911528976572134,
+      "grad_norm": 0.03625324368476868,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 125655
+    },
+    {
+      "epoch": 12.911837237977805,
+      "grad_norm": 0.0910705178976059,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 125658
+    },
+    {
+      "epoch": 12.912145499383477,
+      "grad_norm": 0.11007482558488846,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 125661
+    },
+    {
+      "epoch": 12.912453760789148,
+      "grad_norm": 0.0421065129339695,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 125664
+    },
+    {
+      "epoch": 12.912762022194821,
+      "grad_norm": 0.11193957179784775,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 125667
+    },
+    {
+      "epoch": 12.913070283600494,
+      "grad_norm": 0.05571720749139786,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 125670
+    },
+    {
+      "epoch": 12.913378545006164,
+      "grad_norm": 0.0685516744852066,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 125673
+    },
+    {
+      "epoch": 12.913686806411837,
+      "grad_norm": 0.03462745249271393,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 125676
+    },
+    {
+      "epoch": 12.91399506781751,
+      "grad_norm": 0.040001727640628815,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 125679
+    },
+    {
+      "epoch": 12.91430332922318,
+      "grad_norm": 0.1540547013282776,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 125682
+    },
+    {
+      "epoch": 12.914611590628853,
+      "grad_norm": 0.09268452227115631,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 125685
+    },
+    {
+      "epoch": 12.914919852034526,
+      "grad_norm": 0.050438400357961655,
+      "learning_rate": 0.01,
+      "loss": 1.9549,
+      "step": 125688
+    },
+    {
+      "epoch": 12.915228113440197,
+      "grad_norm": 0.05333781987428665,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 125691
+    },
+    {
+      "epoch": 12.91553637484587,
+      "grad_norm": 0.0557393804192543,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 125694
+    },
+    {
+      "epoch": 12.915844636251542,
+      "grad_norm": 0.04467443376779556,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 125697
+    },
+    {
+      "epoch": 12.916152897657213,
+      "grad_norm": 0.04811103269457817,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 125700
+    },
+    {
+      "epoch": 12.916461159062885,
+      "grad_norm": 0.046122658997774124,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 125703
+    },
+    {
+      "epoch": 12.916769420468558,
+      "grad_norm": 0.0338766910135746,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 125706
+    },
+    {
+      "epoch": 12.917077681874229,
+      "grad_norm": 0.09825889766216278,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 125709
+    },
+    {
+      "epoch": 12.917385943279902,
+      "grad_norm": 0.07365579903125763,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 125712
+    },
+    {
+      "epoch": 12.917694204685574,
+      "grad_norm": 0.057315003126859665,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 125715
+    },
+    {
+      "epoch": 12.918002466091245,
+      "grad_norm": 0.0833062157034874,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 125718
+    },
+    {
+      "epoch": 12.918310727496918,
+      "grad_norm": 0.059123445302248,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 125721
+    },
+    {
+      "epoch": 12.918618988902589,
+      "grad_norm": 0.09031741321086884,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 125724
+    },
+    {
+      "epoch": 12.918927250308261,
+      "grad_norm": 0.057371605187654495,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 125727
+    },
+    {
+      "epoch": 12.919235511713934,
+      "grad_norm": 0.0527326799929142,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 125730
+    },
+    {
+      "epoch": 12.919543773119605,
+      "grad_norm": 0.08562466502189636,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 125733
+    },
+    {
+      "epoch": 12.919852034525277,
+      "grad_norm": 0.09494850784540176,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 125736
+    },
+    {
+      "epoch": 12.92016029593095,
+      "grad_norm": 0.0675438866019249,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 125739
+    },
+    {
+      "epoch": 12.920468557336621,
+      "grad_norm": 0.05136521905660629,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 125742
+    },
+    {
+      "epoch": 12.920776818742294,
+      "grad_norm": 0.04856902360916138,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 125745
+    },
+    {
+      "epoch": 12.921085080147966,
+      "grad_norm": 0.10250970721244812,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 125748
+    },
+    {
+      "epoch": 12.921393341553637,
+      "grad_norm": 0.09212420880794525,
+      "learning_rate": 0.01,
+      "loss": 1.9552,
+      "step": 125751
+    },
+    {
+      "epoch": 12.92170160295931,
+      "grad_norm": 0.08272308111190796,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 125754
+    },
+    {
+      "epoch": 12.922009864364982,
+      "grad_norm": 0.0968143567442894,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 125757
+    },
+    {
+      "epoch": 12.922318125770653,
+      "grad_norm": 0.0924750417470932,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 125760
+    },
+    {
+      "epoch": 12.922626387176326,
+      "grad_norm": 0.03506479784846306,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 125763
+    },
+    {
+      "epoch": 12.922934648581997,
+      "grad_norm": 0.07778924703598022,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 125766
+    },
+    {
+      "epoch": 12.92324290998767,
+      "grad_norm": 0.05012965947389603,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 125769
+    },
+    {
+      "epoch": 12.923551171393342,
+      "grad_norm": 0.11207180470228195,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 125772
+    },
+    {
+      "epoch": 12.923859432799013,
+      "grad_norm": 0.11303342878818512,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 125775
+    },
+    {
+      "epoch": 12.924167694204685,
+      "grad_norm": 0.062038201838731766,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 125778
+    },
+    {
+      "epoch": 12.924475955610358,
+      "grad_norm": 0.03924969211220741,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 125781
+    },
+    {
+      "epoch": 12.924784217016029,
+      "grad_norm": 0.09264468401670456,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 125784
+    },
+    {
+      "epoch": 12.925092478421702,
+      "grad_norm": 0.06861145794391632,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 125787
+    },
+    {
+      "epoch": 12.925400739827374,
+      "grad_norm": 0.08833136409521103,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 125790
+    },
+    {
+      "epoch": 12.925709001233045,
+      "grad_norm": 0.10634084790945053,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 125793
+    },
+    {
+      "epoch": 12.926017262638718,
+      "grad_norm": 0.0582597479224205,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 125796
+    },
+    {
+      "epoch": 12.92632552404439,
+      "grad_norm": 0.0408993661403656,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 125799
+    },
+    {
+      "epoch": 12.926633785450061,
+      "grad_norm": 0.04898262768983841,
+      "learning_rate": 0.01,
+      "loss": 1.9629,
+      "step": 125802
+    },
+    {
+      "epoch": 12.926942046855734,
+      "grad_norm": 0.07243214547634125,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 125805
+    },
+    {
+      "epoch": 12.927250308261407,
+      "grad_norm": 0.04170742630958557,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 125808
+    },
+    {
+      "epoch": 12.927558569667077,
+      "grad_norm": 0.08635149896144867,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 125811
+    },
+    {
+      "epoch": 12.92786683107275,
+      "grad_norm": 0.04802768677473068,
+      "learning_rate": 0.01,
+      "loss": 1.9505,
+      "step": 125814
+    },
+    {
+      "epoch": 12.928175092478421,
+      "grad_norm": 0.04097206890583038,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 125817
+    },
+    {
+      "epoch": 12.928483353884094,
+      "grad_norm": 0.04401004686951637,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 125820
+    },
+    {
+      "epoch": 12.928791615289766,
+      "grad_norm": 0.037511322647333145,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 125823
+    },
+    {
+      "epoch": 12.929099876695437,
+      "grad_norm": 0.09262816607952118,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 125826
+    },
+    {
+      "epoch": 12.92940813810111,
+      "grad_norm": 0.04350706934928894,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 125829
+    },
+    {
+      "epoch": 12.929716399506782,
+      "grad_norm": 0.07727973163127899,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 125832
+    },
+    {
+      "epoch": 12.930024660912453,
+      "grad_norm": 0.05940230190753937,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 125835
+    },
+    {
+      "epoch": 12.930332922318126,
+      "grad_norm": 0.04025878757238388,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 125838
+    },
+    {
+      "epoch": 12.930641183723798,
+      "grad_norm": 0.06040516123175621,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 125841
+    },
+    {
+      "epoch": 12.93094944512947,
+      "grad_norm": 0.13478617370128632,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 125844
+    },
+    {
+      "epoch": 12.931257706535142,
+      "grad_norm": 0.18349310755729675,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 125847
+    },
+    {
+      "epoch": 12.931565967940815,
+      "grad_norm": 0.13208536803722382,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 125850
+    },
+    {
+      "epoch": 12.931874229346485,
+      "grad_norm": 0.03435422480106354,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 125853
+    },
+    {
+      "epoch": 12.932182490752158,
+      "grad_norm": 0.10820911824703217,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 125856
+    },
+    {
+      "epoch": 12.932490752157829,
+      "grad_norm": 0.0809834748506546,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 125859
+    },
+    {
+      "epoch": 12.932799013563502,
+      "grad_norm": 0.1473870873451233,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 125862
+    },
+    {
+      "epoch": 12.933107274969174,
+      "grad_norm": 0.055819351226091385,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 125865
+    },
+    {
+      "epoch": 12.933415536374845,
+      "grad_norm": 0.03641680255532265,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 125868
+    },
+    {
+      "epoch": 12.933723797780518,
+      "grad_norm": 0.039436932653188705,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 125871
+    },
+    {
+      "epoch": 12.93403205918619,
+      "grad_norm": 0.03236067295074463,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 125874
+    },
+    {
+      "epoch": 12.934340320591861,
+      "grad_norm": 0.04350173473358154,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 125877
+    },
+    {
+      "epoch": 12.934648581997534,
+      "grad_norm": 0.10082224756479263,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 125880
+    },
+    {
+      "epoch": 12.934956843403207,
+      "grad_norm": 0.03991500288248062,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 125883
+    },
+    {
+      "epoch": 12.935265104808877,
+      "grad_norm": 0.10700161010026932,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 125886
+    },
+    {
+      "epoch": 12.93557336621455,
+      "grad_norm": 0.07593095302581787,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 125889
+    },
+    {
+      "epoch": 12.935881627620223,
+      "grad_norm": 0.19814598560333252,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 125892
+    },
+    {
+      "epoch": 12.936189889025894,
+      "grad_norm": 0.12118194252252579,
+      "learning_rate": 0.01,
+      "loss": 1.9607,
+      "step": 125895
+    },
+    {
+      "epoch": 12.936498150431566,
+      "grad_norm": 0.08310011774301529,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 125898
+    },
+    {
+      "epoch": 12.936806411837239,
+      "grad_norm": 0.06381826102733612,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 125901
+    },
+    {
+      "epoch": 12.93711467324291,
+      "grad_norm": 0.08685937523841858,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 125904
+    },
+    {
+      "epoch": 12.937422934648582,
+      "grad_norm": 0.0512060709297657,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 125907
+    },
+    {
+      "epoch": 12.937731196054255,
+      "grad_norm": 0.06096792593598366,
+      "learning_rate": 0.01,
+      "loss": 1.9426,
+      "step": 125910
+    },
+    {
+      "epoch": 12.938039457459926,
+      "grad_norm": 0.041577089577913284,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 125913
+    },
+    {
+      "epoch": 12.938347718865598,
+      "grad_norm": 0.027652839198708534,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 125916
+    },
+    {
+      "epoch": 12.93865598027127,
+      "grad_norm": 0.03172867372632027,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 125919
+    },
+    {
+      "epoch": 12.938964241676942,
+      "grad_norm": 0.033684540539979935,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 125922
+    },
+    {
+      "epoch": 12.939272503082615,
+      "grad_norm": 0.02770218439400196,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 125925
+    },
+    {
+      "epoch": 12.939580764488285,
+      "grad_norm": 0.07285824418067932,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 125928
+    },
+    {
+      "epoch": 12.939889025893958,
+      "grad_norm": 0.05254669860005379,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 125931
+    },
+    {
+      "epoch": 12.94019728729963,
+      "grad_norm": 0.1141522154211998,
+      "learning_rate": 0.01,
+      "loss": 1.8849,
+      "step": 125934
+    },
+    {
+      "epoch": 12.940505548705302,
+      "grad_norm": 0.06985414028167725,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 125937
+    },
+    {
+      "epoch": 12.940813810110974,
+      "grad_norm": 0.05042961239814758,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 125940
+    },
+    {
+      "epoch": 12.941122071516647,
+      "grad_norm": 0.09245620667934418,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 125943
+    },
+    {
+      "epoch": 12.941430332922318,
+      "grad_norm": 0.07112108170986176,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 125946
+    },
+    {
+      "epoch": 12.94173859432799,
+      "grad_norm": 0.04745557904243469,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 125949
+    },
+    {
+      "epoch": 12.942046855733661,
+      "grad_norm": 0.09495459496974945,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 125952
+    },
+    {
+      "epoch": 12.942355117139334,
+      "grad_norm": 0.07864931970834732,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 125955
+    },
+    {
+      "epoch": 12.942663378545006,
+      "grad_norm": 0.07978437840938568,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 125958
+    },
+    {
+      "epoch": 12.942971639950677,
+      "grad_norm": 0.06312549114227295,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 125961
+    },
+    {
+      "epoch": 12.94327990135635,
+      "grad_norm": 0.05435273423790932,
+      "learning_rate": 0.01,
+      "loss": 1.9528,
+      "step": 125964
+    },
+    {
+      "epoch": 12.943588162762023,
+      "grad_norm": 0.06605318188667297,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 125967
+    },
+    {
+      "epoch": 12.943896424167693,
+      "grad_norm": 0.10791897028684616,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 125970
+    },
+    {
+      "epoch": 12.944204685573366,
+      "grad_norm": 0.036136701703071594,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 125973
+    },
+    {
+      "epoch": 12.944512946979039,
+      "grad_norm": 0.057532746344804764,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 125976
+    },
+    {
+      "epoch": 12.94482120838471,
+      "grad_norm": 0.05744960159063339,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 125979
+    },
+    {
+      "epoch": 12.945129469790382,
+      "grad_norm": 0.04510439187288284,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 125982
+    },
+    {
+      "epoch": 12.945437731196055,
+      "grad_norm": 0.05059744790196419,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 125985
+    },
+    {
+      "epoch": 12.945745992601726,
+      "grad_norm": 0.081993468105793,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 125988
+    },
+    {
+      "epoch": 12.946054254007398,
+      "grad_norm": 0.04813908040523529,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 125991
+    },
+    {
+      "epoch": 12.946362515413071,
+      "grad_norm": 0.061780817806720734,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 125994
+    },
+    {
+      "epoch": 12.946670776818742,
+      "grad_norm": 0.04033418744802475,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 125997
+    },
+    {
+      "epoch": 12.946979038224415,
+      "grad_norm": 0.03506530076265335,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 126000
+    },
+    {
+      "epoch": 12.947287299630087,
+      "grad_norm": 0.06479054689407349,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 126003
+    },
+    {
+      "epoch": 12.947595561035758,
+      "grad_norm": 0.05530203506350517,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 126006
+    },
+    {
+      "epoch": 12.94790382244143,
+      "grad_norm": 0.05367561802268028,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 126009
+    },
+    {
+      "epoch": 12.948212083847102,
+      "grad_norm": 0.05060233920812607,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 126012
+    },
+    {
+      "epoch": 12.948520345252774,
+      "grad_norm": 0.06014840304851532,
+      "learning_rate": 0.01,
+      "loss": 1.9598,
+      "step": 126015
+    },
+    {
+      "epoch": 12.948828606658447,
+      "grad_norm": 0.11498754471540451,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 126018
+    },
+    {
+      "epoch": 12.949136868064118,
+      "grad_norm": 0.09030990302562714,
+      "learning_rate": 0.01,
+      "loss": 1.9521,
+      "step": 126021
+    },
+    {
+      "epoch": 12.94944512946979,
+      "grad_norm": 0.09651051461696625,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 126024
+    },
+    {
+      "epoch": 12.949753390875463,
+      "grad_norm": 0.11377514153718948,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 126027
+    },
+    {
+      "epoch": 12.950061652281134,
+      "grad_norm": 0.08735734224319458,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 126030
+    },
+    {
+      "epoch": 12.950369913686806,
+      "grad_norm": 0.08239898085594177,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 126033
+    },
+    {
+      "epoch": 12.950678175092479,
+      "grad_norm": 0.049356214702129364,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 126036
+    },
+    {
+      "epoch": 12.95098643649815,
+      "grad_norm": 0.04305762052536011,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 126039
+    },
+    {
+      "epoch": 12.951294697903823,
+      "grad_norm": 0.03519025072455406,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 126042
+    },
+    {
+      "epoch": 12.951602959309495,
+      "grad_norm": 0.03509996086359024,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 126045
+    },
+    {
+      "epoch": 12.951911220715166,
+      "grad_norm": 0.055822983384132385,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 126048
+    },
+    {
+      "epoch": 12.952219482120839,
+      "grad_norm": 0.06479526311159134,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 126051
+    },
+    {
+      "epoch": 12.95252774352651,
+      "grad_norm": 0.061816923320293427,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 126054
+    },
+    {
+      "epoch": 12.952836004932182,
+      "grad_norm": 0.0821833461523056,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 126057
+    },
+    {
+      "epoch": 12.953144266337855,
+      "grad_norm": 0.06200600787997246,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 126060
+    },
+    {
+      "epoch": 12.953452527743526,
+      "grad_norm": 0.10300254821777344,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 126063
+    },
+    {
+      "epoch": 12.953760789149198,
+      "grad_norm": 0.04916617274284363,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 126066
+    },
+    {
+      "epoch": 12.954069050554871,
+      "grad_norm": 0.08358190208673477,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 126069
+    },
+    {
+      "epoch": 12.954377311960542,
+      "grad_norm": 0.050949402153491974,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 126072
+    },
+    {
+      "epoch": 12.954685573366215,
+      "grad_norm": 0.06936567276716232,
+      "learning_rate": 0.01,
+      "loss": 1.9006,
+      "step": 126075
+    },
+    {
+      "epoch": 12.954993834771887,
+      "grad_norm": 0.1449357569217682,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 126078
+    },
+    {
+      "epoch": 12.955302096177558,
+      "grad_norm": 0.09677485376596451,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 126081
+    },
+    {
+      "epoch": 12.95561035758323,
+      "grad_norm": 0.1612035185098648,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 126084
+    },
+    {
+      "epoch": 12.955918618988903,
+      "grad_norm": 0.060679059475660324,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 126087
+    },
+    {
+      "epoch": 12.956226880394574,
+      "grad_norm": 0.10726216435432434,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 126090
+    },
+    {
+      "epoch": 12.956535141800247,
+      "grad_norm": 0.10506871342658997,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 126093
+    },
+    {
+      "epoch": 12.95684340320592,
+      "grad_norm": 0.041960135102272034,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 126096
+    },
+    {
+      "epoch": 12.95715166461159,
+      "grad_norm": 0.08968428522348404,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 126099
+    },
+    {
+      "epoch": 12.957459926017263,
+      "grad_norm": 0.0700458213686943,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 126102
+    },
+    {
+      "epoch": 12.957768187422936,
+      "grad_norm": 0.04659474268555641,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 126105
+    },
+    {
+      "epoch": 12.958076448828606,
+      "grad_norm": 0.09629134833812714,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 126108
+    },
+    {
+      "epoch": 12.958384710234279,
+      "grad_norm": 0.06129566580057144,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 126111
+    },
+    {
+      "epoch": 12.95869297163995,
+      "grad_norm": 0.032240573316812515,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 126114
+    },
+    {
+      "epoch": 12.959001233045623,
+      "grad_norm": 0.08730331808328629,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 126117
+    },
+    {
+      "epoch": 12.959309494451295,
+      "grad_norm": 0.10972293466329575,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 126120
+    },
+    {
+      "epoch": 12.959617755856966,
+      "grad_norm": 0.037579625844955444,
+      "learning_rate": 0.01,
+      "loss": 1.9489,
+      "step": 126123
+    },
+    {
+      "epoch": 12.959926017262639,
+      "grad_norm": 0.08131605386734009,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 126126
+    },
+    {
+      "epoch": 12.960234278668311,
+      "grad_norm": 0.0587998703122139,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 126129
+    },
+    {
+      "epoch": 12.960542540073982,
+      "grad_norm": 0.09930302947759628,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 126132
+    },
+    {
+      "epoch": 12.960850801479655,
+      "grad_norm": 0.07516415417194366,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 126135
+    },
+    {
+      "epoch": 12.961159062885327,
+      "grad_norm": 0.03397361934185028,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 126138
+    },
+    {
+      "epoch": 12.961467324290998,
+      "grad_norm": 0.05763630568981171,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 126141
+    },
+    {
+      "epoch": 12.961775585696671,
+      "grad_norm": 0.07452522218227386,
+      "learning_rate": 0.01,
+      "loss": 1.9637,
+      "step": 126144
+    },
+    {
+      "epoch": 12.962083847102342,
+      "grad_norm": 0.050457656383514404,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 126147
+    },
+    {
+      "epoch": 12.962392108508014,
+      "grad_norm": 0.04492957890033722,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 126150
+    },
+    {
+      "epoch": 12.962700369913687,
+      "grad_norm": 0.032196804881095886,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 126153
+    },
+    {
+      "epoch": 12.963008631319358,
+      "grad_norm": 0.036084260791540146,
+      "learning_rate": 0.01,
+      "loss": 1.9597,
+      "step": 126156
+    },
+    {
+      "epoch": 12.96331689272503,
+      "grad_norm": 0.10411658883094788,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 126159
+    },
+    {
+      "epoch": 12.963625154130703,
+      "grad_norm": 0.07700277864933014,
+      "learning_rate": 0.01,
+      "loss": 1.8948,
+      "step": 126162
+    },
+    {
+      "epoch": 12.963933415536374,
+      "grad_norm": 0.057165950536727905,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 126165
+    },
+    {
+      "epoch": 12.964241676942047,
+      "grad_norm": 0.10003688931465149,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 126168
+    },
+    {
+      "epoch": 12.96454993834772,
+      "grad_norm": 0.03061365708708763,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 126171
+    },
+    {
+      "epoch": 12.96485819975339,
+      "grad_norm": 0.11224297434091568,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 126174
+    },
+    {
+      "epoch": 12.965166461159063,
+      "grad_norm": 0.09531304240226746,
+      "learning_rate": 0.01,
+      "loss": 1.959,
+      "step": 126177
+    },
+    {
+      "epoch": 12.965474722564736,
+      "grad_norm": 0.0589129813015461,
+      "learning_rate": 0.01,
+      "loss": 1.9621,
+      "step": 126180
+    },
+    {
+      "epoch": 12.965782983970406,
+      "grad_norm": 0.05862326920032501,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 126183
+    },
+    {
+      "epoch": 12.966091245376079,
+      "grad_norm": 0.04244602099061012,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 126186
+    },
+    {
+      "epoch": 12.966399506781752,
+      "grad_norm": 0.039083950221538544,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 126189
+    },
+    {
+      "epoch": 12.966707768187423,
+      "grad_norm": 0.04154587909579277,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 126192
+    },
+    {
+      "epoch": 12.967016029593095,
+      "grad_norm": 0.04228314757347107,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 126195
+    },
+    {
+      "epoch": 12.967324290998768,
+      "grad_norm": 0.11571267992258072,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 126198
+    },
+    {
+      "epoch": 12.967632552404439,
+      "grad_norm": 0.08702175319194794,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 126201
+    },
+    {
+      "epoch": 12.967940813810111,
+      "grad_norm": 0.07003472000360489,
+      "learning_rate": 0.01,
+      "loss": 1.9482,
+      "step": 126204
+    },
+    {
+      "epoch": 12.968249075215782,
+      "grad_norm": 0.05063821002840996,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 126207
+    },
+    {
+      "epoch": 12.968557336621455,
+      "grad_norm": 0.034900352358818054,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 126210
+    },
+    {
+      "epoch": 12.968865598027127,
+      "grad_norm": 0.06341734528541565,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 126213
+    },
+    {
+      "epoch": 12.969173859432798,
+      "grad_norm": 0.0889853686094284,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 126216
+    },
+    {
+      "epoch": 12.969482120838471,
+      "grad_norm": 0.042263999581336975,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 126219
+    },
+    {
+      "epoch": 12.969790382244144,
+      "grad_norm": 0.04034658893942833,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 126222
+    },
+    {
+      "epoch": 12.970098643649814,
+      "grad_norm": 0.043879568576812744,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 126225
+    },
+    {
+      "epoch": 12.970406905055487,
+      "grad_norm": 0.10296093672513962,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 126228
+    },
+    {
+      "epoch": 12.97071516646116,
+      "grad_norm": 0.05123838409781456,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 126231
+    },
+    {
+      "epoch": 12.97102342786683,
+      "grad_norm": 0.03683066740632057,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 126234
+    },
+    {
+      "epoch": 12.971331689272503,
+      "grad_norm": 0.09338618069887161,
+      "learning_rate": 0.01,
+      "loss": 1.9524,
+      "step": 126237
+    },
+    {
+      "epoch": 12.971639950678176,
+      "grad_norm": 0.07788629829883575,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 126240
+    },
+    {
+      "epoch": 12.971948212083847,
+      "grad_norm": 0.07458705455064774,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 126243
+    },
+    {
+      "epoch": 12.97225647348952,
+      "grad_norm": 0.0914078876376152,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 126246
+    },
+    {
+      "epoch": 12.97256473489519,
+      "grad_norm": 0.0758683905005455,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 126249
+    },
+    {
+      "epoch": 12.972872996300863,
+      "grad_norm": 0.10591443628072739,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 126252
+    },
+    {
+      "epoch": 12.973181257706536,
+      "grad_norm": 0.15483787655830383,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 126255
+    },
+    {
+      "epoch": 12.973489519112206,
+      "grad_norm": 0.0808720663189888,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 126258
+    },
+    {
+      "epoch": 12.973797780517879,
+      "grad_norm": 0.08866417407989502,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 126261
+    },
+    {
+      "epoch": 12.974106041923552,
+      "grad_norm": 0.04881209880113602,
+      "learning_rate": 0.01,
+      "loss": 1.9611,
+      "step": 126264
+    },
+    {
+      "epoch": 12.974414303329223,
+      "grad_norm": 0.053279809653759,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 126267
+    },
+    {
+      "epoch": 12.974722564734895,
+      "grad_norm": 0.04660681635141373,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 126270
+    },
+    {
+      "epoch": 12.975030826140568,
+      "grad_norm": 0.03146835044026375,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 126273
+    },
+    {
+      "epoch": 12.975339087546239,
+      "grad_norm": 0.058334872126579285,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 126276
+    },
+    {
+      "epoch": 12.975647348951911,
+      "grad_norm": 0.05183687061071396,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 126279
+    },
+    {
+      "epoch": 12.975955610357584,
+      "grad_norm": 0.1583312749862671,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 126282
+    },
+    {
+      "epoch": 12.976263871763255,
+      "grad_norm": 0.09854957461357117,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 126285
+    },
+    {
+      "epoch": 12.976572133168927,
+      "grad_norm": 0.04723304137587547,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 126288
+    },
+    {
+      "epoch": 12.9768803945746,
+      "grad_norm": 0.10145114362239838,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 126291
+    },
+    {
+      "epoch": 12.977188655980271,
+      "grad_norm": 0.04832020401954651,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 126294
+    },
+    {
+      "epoch": 12.977496917385944,
+      "grad_norm": 0.04121669381856918,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 126297
+    },
+    {
+      "epoch": 12.977805178791616,
+      "grad_norm": 0.0549682192504406,
+      "learning_rate": 0.01,
+      "loss": 1.9484,
+      "step": 126300
+    },
+    {
+      "epoch": 12.978113440197287,
+      "grad_norm": 0.10428796708583832,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 126303
+    },
+    {
+      "epoch": 12.97842170160296,
+      "grad_norm": 0.045746032148599625,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 126306
+    },
+    {
+      "epoch": 12.97872996300863,
+      "grad_norm": 0.13235408067703247,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 126309
+    },
+    {
+      "epoch": 12.979038224414303,
+      "grad_norm": 0.06408566981554031,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 126312
+    },
+    {
+      "epoch": 12.979346485819976,
+      "grad_norm": 0.0649254247546196,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 126315
+    },
+    {
+      "epoch": 12.979654747225647,
+      "grad_norm": 0.052825432270765305,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 126318
+    },
+    {
+      "epoch": 12.97996300863132,
+      "grad_norm": 0.03324391692876816,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 126321
+    },
+    {
+      "epoch": 12.980271270036992,
+      "grad_norm": 0.04539883881807327,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 126324
+    },
+    {
+      "epoch": 12.980579531442663,
+      "grad_norm": 0.04607979580760002,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 126327
+    },
+    {
+      "epoch": 12.980887792848335,
+      "grad_norm": 0.036359164863824844,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 126330
+    },
+    {
+      "epoch": 12.981196054254008,
+      "grad_norm": 0.13999003171920776,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 126333
+    },
+    {
+      "epoch": 12.981504315659679,
+      "grad_norm": 0.05416400358080864,
+      "learning_rate": 0.01,
+      "loss": 1.9603,
+      "step": 126336
+    },
+    {
+      "epoch": 12.981812577065352,
+      "grad_norm": 0.09667657315731049,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 126339
+    },
+    {
+      "epoch": 12.982120838471022,
+      "grad_norm": 0.037298351526260376,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 126342
+    },
+    {
+      "epoch": 12.982429099876695,
+      "grad_norm": 0.11069020628929138,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 126345
+    },
+    {
+      "epoch": 12.982737361282368,
+      "grad_norm": 0.06094904616475105,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 126348
+    },
+    {
+      "epoch": 12.983045622688039,
+      "grad_norm": 0.05997653305530548,
+      "learning_rate": 0.01,
+      "loss": 1.9451,
+      "step": 126351
+    },
+    {
+      "epoch": 12.983353884093711,
+      "grad_norm": 0.05035969242453575,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 126354
+    },
+    {
+      "epoch": 12.983662145499384,
+      "grad_norm": 0.061228420585393906,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 126357
+    },
+    {
+      "epoch": 12.983970406905055,
+      "grad_norm": 0.03478321060538292,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 126360
+    },
+    {
+      "epoch": 12.984278668310727,
+      "grad_norm": 0.04280786216259003,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 126363
+    },
+    {
+      "epoch": 12.9845869297164,
+      "grad_norm": 0.05942310765385628,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 126366
+    },
+    {
+      "epoch": 12.984895191122071,
+      "grad_norm": 0.10070284456014633,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 126369
+    },
+    {
+      "epoch": 12.985203452527744,
+      "grad_norm": 0.06833632290363312,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 126372
+    },
+    {
+      "epoch": 12.985511713933416,
+      "grad_norm": 0.0459744967520237,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 126375
+    },
+    {
+      "epoch": 12.985819975339087,
+      "grad_norm": 0.10393807291984558,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 126378
+    },
+    {
+      "epoch": 12.98612823674476,
+      "grad_norm": 0.09013234078884125,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 126381
+    },
+    {
+      "epoch": 12.986436498150432,
+      "grad_norm": 0.0976896733045578,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 126384
+    },
+    {
+      "epoch": 12.986744759556103,
+      "grad_norm": 0.08779832720756531,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 126387
+    },
+    {
+      "epoch": 12.987053020961776,
+      "grad_norm": 0.06239229440689087,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 126390
+    },
+    {
+      "epoch": 12.987361282367448,
+      "grad_norm": 0.05805114284157753,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 126393
+    },
+    {
+      "epoch": 12.98766954377312,
+      "grad_norm": 0.045233603566884995,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 126396
+    },
+    {
+      "epoch": 12.987977805178792,
+      "grad_norm": 0.05403595790266991,
+      "learning_rate": 0.01,
+      "loss": 1.9599,
+      "step": 126399
+    },
+    {
+      "epoch": 12.988286066584463,
+      "grad_norm": 0.04203466325998306,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 126402
+    },
+    {
+      "epoch": 12.988594327990135,
+      "grad_norm": 0.04115534946322441,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 126405
+    },
+    {
+      "epoch": 12.988902589395808,
+      "grad_norm": 0.0468607060611248,
+      "learning_rate": 0.01,
+      "loss": 1.8964,
+      "step": 126408
+    },
+    {
+      "epoch": 12.989210850801479,
+      "grad_norm": 0.09589367359876633,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 126411
+    },
+    {
+      "epoch": 12.989519112207152,
+      "grad_norm": 0.091950923204422,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 126414
+    },
+    {
+      "epoch": 12.989827373612824,
+      "grad_norm": 0.0383729413151741,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 126417
+    },
+    {
+      "epoch": 12.990135635018495,
+      "grad_norm": 0.0655297264456749,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 126420
+    },
+    {
+      "epoch": 12.990443896424168,
+      "grad_norm": 0.08317092061042786,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 126423
+    },
+    {
+      "epoch": 12.99075215782984,
+      "grad_norm": 0.09387842565774918,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 126426
+    },
+    {
+      "epoch": 12.991060419235511,
+      "grad_norm": 0.19405832886695862,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 126429
+    },
+    {
+      "epoch": 12.991368680641184,
+      "grad_norm": 0.08507779985666275,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 126432
+    },
+    {
+      "epoch": 12.991676942046857,
+      "grad_norm": 0.09436507523059845,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 126435
+    },
+    {
+      "epoch": 12.991985203452527,
+      "grad_norm": 0.06602399051189423,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 126438
+    },
+    {
+      "epoch": 12.9922934648582,
+      "grad_norm": 0.06707599759101868,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 126441
+    },
+    {
+      "epoch": 12.99260172626387,
+      "grad_norm": 0.05414499342441559,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 126444
+    },
+    {
+      "epoch": 12.992909987669544,
+      "grad_norm": 0.04891413450241089,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 126447
+    },
+    {
+      "epoch": 12.993218249075216,
+      "grad_norm": 0.03576958179473877,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 126450
+    },
+    {
+      "epoch": 12.993526510480887,
+      "grad_norm": 0.04596221446990967,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 126453
+    },
+    {
+      "epoch": 12.99383477188656,
+      "grad_norm": 0.03685097396373749,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 126456
+    },
+    {
+      "epoch": 12.994143033292232,
+      "grad_norm": 0.1213253065943718,
+      "learning_rate": 0.01,
+      "loss": 1.9594,
+      "step": 126459
+    },
+    {
+      "epoch": 12.994451294697903,
+      "grad_norm": 0.08971472829580307,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 126462
+    },
+    {
+      "epoch": 12.994759556103576,
+      "grad_norm": 0.09155774116516113,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 126465
+    },
+    {
+      "epoch": 12.995067817509248,
+      "grad_norm": 0.05760270729660988,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 126468
+    },
+    {
+      "epoch": 12.99537607891492,
+      "grad_norm": 0.03982146829366684,
+      "learning_rate": 0.01,
+      "loss": 1.9644,
+      "step": 126471
+    },
+    {
+      "epoch": 12.995684340320592,
+      "grad_norm": 0.03316031023859978,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 126474
+    },
+    {
+      "epoch": 12.995992601726265,
+      "grad_norm": 0.06980220228433609,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 126477
+    },
+    {
+      "epoch": 12.996300863131935,
+      "grad_norm": 0.06975305825471878,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 126480
+    },
+    {
+      "epoch": 12.996609124537608,
+      "grad_norm": 0.04797213152050972,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 126483
+    },
+    {
+      "epoch": 12.99691738594328,
+      "grad_norm": 0.07199689000844955,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 126486
+    },
+    {
+      "epoch": 12.997225647348952,
+      "grad_norm": 0.05560987815260887,
+      "learning_rate": 0.01,
+      "loss": 1.9587,
+      "step": 126489
+    },
+    {
+      "epoch": 12.997533908754624,
+      "grad_norm": 0.07067282497882843,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 126492
+    },
+    {
+      "epoch": 12.997842170160297,
+      "grad_norm": 0.0505492277443409,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 126495
+    },
+    {
+      "epoch": 12.998150431565968,
+      "grad_norm": 0.05297574773430824,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 126498
+    },
+    {
+      "epoch": 12.99845869297164,
+      "grad_norm": 0.05754699558019638,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 126501
+    },
+    {
+      "epoch": 12.998766954377311,
+      "grad_norm": 0.07020077109336853,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 126504
+    },
+    {
+      "epoch": 12.999075215782984,
+      "grad_norm": 0.1119004338979721,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 126507
+    },
+    {
+      "epoch": 12.999383477188656,
+      "grad_norm": 0.05641701817512512,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 126510
+    },
+    {
+      "epoch": 12.999691738594327,
+      "grad_norm": 0.09984719753265381,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 126513
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 0.04795163497328758,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 126516
+    },
+    {
+      "epoch": 12.990964164698635,
+      "grad_norm": 0.06837742775678635,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 126519
+    },
+    {
+      "epoch": 12.991272204538454,
+      "grad_norm": 0.0794941857457161,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 126522
+    },
+    {
+      "epoch": 12.991580244378273,
+      "grad_norm": 0.05930837243795395,
+      "learning_rate": 0.01,
+      "loss": 1.9496,
+      "step": 126525
+    },
+    {
+      "epoch": 12.991888284218092,
+      "grad_norm": 0.09570945799350739,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 126528
+    },
+    {
+      "epoch": 12.992196324057911,
+      "grad_norm": 0.06531012803316116,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 126531
+    },
+    {
+      "epoch": 12.99250436389773,
+      "grad_norm": 0.09255339205265045,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 126534
+    },
+    {
+      "epoch": 12.99281240373755,
+      "grad_norm": 0.05466054752469063,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 126537
+    },
+    {
+      "epoch": 12.99312044357737,
+      "grad_norm": 0.07487985491752625,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 126540
+    },
+    {
+      "epoch": 12.993428483417189,
+      "grad_norm": 0.08568083494901657,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 126543
+    },
+    {
+      "epoch": 12.993736523257008,
+      "grad_norm": 0.17832233011722565,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 126546
+    },
+    {
+      "epoch": 12.994044563096827,
+      "grad_norm": 0.0828866958618164,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 126549
+    },
+    {
+      "epoch": 12.994352602936647,
+      "grad_norm": 0.05283142626285553,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 126552
+    },
+    {
+      "epoch": 12.994660642776466,
+      "grad_norm": 0.033367790281772614,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 126555
+    },
+    {
+      "epoch": 12.994968682616285,
+      "grad_norm": 0.033711668103933334,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 126558
+    },
+    {
+      "epoch": 12.995276722456104,
+      "grad_norm": 0.04041490703821182,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 126561
+    },
+    {
+      "epoch": 12.995584762295923,
+      "grad_norm": 0.04604267701506615,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 126564
+    },
+    {
+      "epoch": 12.995892802135742,
+      "grad_norm": 0.06144135072827339,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 126567
+    },
+    {
+      "epoch": 12.996200841975561,
+      "grad_norm": 0.12728659808635712,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 126570
+    },
+    {
+      "epoch": 12.996508881815382,
+      "grad_norm": 0.11559011042118073,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 126573
+    },
+    {
+      "epoch": 12.996816921655201,
+      "grad_norm": 0.04632570222020149,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 126576
+    },
+    {
+      "epoch": 12.99712496149502,
+      "grad_norm": 0.034444134682416916,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 126579
+    },
+    {
+      "epoch": 12.997433001334839,
+      "grad_norm": 0.040618523955345154,
+      "learning_rate": 0.01,
+      "loss": 1.8933,
+      "step": 126582
+    },
+    {
+      "epoch": 12.997741041174658,
+      "grad_norm": 0.03981280326843262,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 126585
+    },
+    {
+      "epoch": 12.998049081014479,
+      "grad_norm": 0.0703684389591217,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 126588
+    },
+    {
+      "epoch": 12.998357120854298,
+      "grad_norm": 0.04806933179497719,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 126591
+    },
+    {
+      "epoch": 12.998665160694117,
+      "grad_norm": 0.05543390288949013,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 126594
+    },
+    {
+      "epoch": 12.998973200533936,
+      "grad_norm": 0.051780570298433304,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 126597
+    },
+    {
+      "epoch": 12.999281240373755,
+      "grad_norm": 0.03488859161734581,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 126600
+    },
+    {
+      "epoch": 12.999589280213574,
+      "grad_norm": 0.07411931455135345,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 126603
+    },
+    {
+      "epoch": 12.999897320053394,
+      "grad_norm": 0.16531667113304138,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 126606
+    },
+    {
+      "epoch": 13.000205359893213,
+      "grad_norm": 0.12784148752689362,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 126609
+    },
+    {
+      "epoch": 13.000513399733032,
+      "grad_norm": 0.07612412422895432,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 126612
+    },
+    {
+      "epoch": 13.000821439572851,
+      "grad_norm": 0.06532411277294159,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 126615
+    },
+    {
+      "epoch": 13.00112947941267,
+      "grad_norm": 0.04270946979522705,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 126618
+    },
+    {
+      "epoch": 13.00143751925249,
+      "grad_norm": 0.0750366672873497,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 126621
+    },
+    {
+      "epoch": 13.00174555909231,
+      "grad_norm": 0.07818207889795303,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 126624
+    },
+    {
+      "epoch": 13.002053598932129,
+      "grad_norm": 0.07291862368583679,
+      "learning_rate": 0.01,
+      "loss": 1.9464,
+      "step": 126627
+    },
+    {
+      "epoch": 13.002361638771948,
+      "grad_norm": 0.0576990582048893,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 126630
+    },
+    {
+      "epoch": 13.002669678611767,
+      "grad_norm": 0.06400927156209946,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 126633
+    },
+    {
+      "epoch": 13.002977718451586,
+      "grad_norm": 0.055741291493177414,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 126636
+    },
+    {
+      "epoch": 13.003285758291407,
+      "grad_norm": 0.04952171444892883,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 126639
+    },
+    {
+      "epoch": 13.003593798131226,
+      "grad_norm": 0.05010899528861046,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 126642
+    },
+    {
+      "epoch": 13.003901837971044,
+      "grad_norm": 0.10982075333595276,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 126645
+    },
+    {
+      "epoch": 13.004209877810863,
+      "grad_norm": 0.08790125697851181,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 126648
+    },
+    {
+      "epoch": 13.004517917650682,
+      "grad_norm": 0.06224539503455162,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 126651
+    },
+    {
+      "epoch": 13.004825957490501,
+      "grad_norm": 0.08626297861337662,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 126654
+    },
+    {
+      "epoch": 13.005133997330322,
+      "grad_norm": 0.0522213838994503,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 126657
+    },
+    {
+      "epoch": 13.005442037170141,
+      "grad_norm": 0.050593301653862,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 126660
+    },
+    {
+      "epoch": 13.00575007700996,
+      "grad_norm": 0.04440055787563324,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 126663
+    },
+    {
+      "epoch": 13.006058116849779,
+      "grad_norm": 0.11652219295501709,
+      "learning_rate": 0.01,
+      "loss": 1.9517,
+      "step": 126666
+    },
+    {
+      "epoch": 13.006366156689598,
+      "grad_norm": 0.06647264957427979,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 126669
+    },
+    {
+      "epoch": 13.006674196529417,
+      "grad_norm": 0.08247080445289612,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 126672
+    },
+    {
+      "epoch": 13.006982236369238,
+      "grad_norm": 0.03975950926542282,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 126675
+    },
+    {
+      "epoch": 13.007290276209057,
+      "grad_norm": 0.08247572928667068,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 126678
+    },
+    {
+      "epoch": 13.007598316048876,
+      "grad_norm": 0.04770240560173988,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 126681
+    },
+    {
+      "epoch": 13.007906355888695,
+      "grad_norm": 0.10593480616807938,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 126684
+    },
+    {
+      "epoch": 13.008214395728514,
+      "grad_norm": 0.07158973813056946,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 126687
+    },
+    {
+      "epoch": 13.008522435568333,
+      "grad_norm": 0.1066482663154602,
+      "learning_rate": 0.01,
+      "loss": 1.9697,
+      "step": 126690
+    },
+    {
+      "epoch": 13.008830475408153,
+      "grad_norm": 0.049158018082380295,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 126693
+    },
+    {
+      "epoch": 13.009138515247972,
+      "grad_norm": 0.09192405641078949,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 126696
+    },
+    {
+      "epoch": 13.009446555087791,
+      "grad_norm": 0.06074890121817589,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 126699
+    },
+    {
+      "epoch": 13.00975459492761,
+      "grad_norm": 0.051200829446315765,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 126702
+    },
+    {
+      "epoch": 13.01006263476743,
+      "grad_norm": 0.10218089818954468,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 126705
+    },
+    {
+      "epoch": 13.01037067460725,
+      "grad_norm": 0.03220054879784584,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 126708
+    },
+    {
+      "epoch": 13.010678714447069,
+      "grad_norm": 0.05471636354923248,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 126711
+    },
+    {
+      "epoch": 13.010986754286888,
+      "grad_norm": 0.0949852392077446,
+      "learning_rate": 0.01,
+      "loss": 1.9516,
+      "step": 126714
+    },
+    {
+      "epoch": 13.011294794126707,
+      "grad_norm": 0.04287450760602951,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 126717
+    },
+    {
+      "epoch": 13.011602833966526,
+      "grad_norm": 0.047532957047224045,
+      "learning_rate": 0.01,
+      "loss": 1.9449,
+      "step": 126720
+    },
+    {
+      "epoch": 13.011910873806345,
+      "grad_norm": 0.04211737960577011,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 126723
+    },
+    {
+      "epoch": 13.012218913646166,
+      "grad_norm": 0.10187740623950958,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 126726
+    },
+    {
+      "epoch": 13.012526953485985,
+      "grad_norm": 0.0747995525598526,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 126729
+    },
+    {
+      "epoch": 13.012834993325804,
+      "grad_norm": 0.08981893211603165,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 126732
+    },
+    {
+      "epoch": 13.013143033165623,
+      "grad_norm": 0.06654146313667297,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 126735
+    },
+    {
+      "epoch": 13.013451073005442,
+      "grad_norm": 0.09594383835792542,
+      "learning_rate": 0.01,
+      "loss": 1.9022,
+      "step": 126738
+    },
+    {
+      "epoch": 13.01375911284526,
+      "grad_norm": 0.07657212018966675,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 126741
+    },
+    {
+      "epoch": 13.014067152685081,
+      "grad_norm": 0.035843364894390106,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 126744
+    },
+    {
+      "epoch": 13.0143751925249,
+      "grad_norm": 0.07398900389671326,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 126747
+    },
+    {
+      "epoch": 13.01468323236472,
+      "grad_norm": 0.06803435832262039,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 126750
+    },
+    {
+      "epoch": 13.014991272204538,
+      "grad_norm": 0.13478237390518188,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 126753
+    },
+    {
+      "epoch": 13.015299312044357,
+      "grad_norm": 0.1437871754169464,
+      "learning_rate": 0.01,
+      "loss": 1.9586,
+      "step": 126756
+    },
+    {
+      "epoch": 13.015607351884176,
+      "grad_norm": 0.06393241882324219,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 126759
+    },
+    {
+      "epoch": 13.015915391723997,
+      "grad_norm": 0.13030607998371124,
+      "learning_rate": 0.01,
+      "loss": 1.9001,
+      "step": 126762
+    },
+    {
+      "epoch": 13.016223431563816,
+      "grad_norm": 0.13317731022834778,
+      "learning_rate": 0.01,
+      "loss": 1.8865,
+      "step": 126765
+    },
+    {
+      "epoch": 13.016531471403635,
+      "grad_norm": 0.0745624303817749,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 126768
+    },
+    {
+      "epoch": 13.016839511243454,
+      "grad_norm": 0.05197130888700485,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 126771
+    },
+    {
+      "epoch": 13.017147551083273,
+      "grad_norm": 0.06029503419995308,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 126774
+    },
+    {
+      "epoch": 13.017455590923094,
+      "grad_norm": 0.058335818350315094,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 126777
+    },
+    {
+      "epoch": 13.017763630762913,
+      "grad_norm": 0.04288724064826965,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 126780
+    },
+    {
+      "epoch": 13.018071670602732,
+      "grad_norm": 0.042841363698244095,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 126783
+    },
+    {
+      "epoch": 13.01837971044255,
+      "grad_norm": 0.06195900961756706,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 126786
+    },
+    {
+      "epoch": 13.01868775028237,
+      "grad_norm": 0.07918534427881241,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 126789
+    },
+    {
+      "epoch": 13.018995790122188,
+      "grad_norm": 0.09109507501125336,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 126792
+    },
+    {
+      "epoch": 13.01930382996201,
+      "grad_norm": 0.07134268432855606,
+      "learning_rate": 0.01,
+      "loss": 1.8977,
+      "step": 126795
+    },
+    {
+      "epoch": 13.019611869801828,
+      "grad_norm": 0.1011328250169754,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 126798
+    },
+    {
+      "epoch": 13.019919909641647,
+      "grad_norm": 0.055137354880571365,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 126801
+    },
+    {
+      "epoch": 13.020227949481466,
+      "grad_norm": 0.08781642466783524,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 126804
+    },
+    {
+      "epoch": 13.020535989321285,
+      "grad_norm": 0.0311215091496706,
+      "learning_rate": 0.01,
+      "loss": 1.8982,
+      "step": 126807
+    },
+    {
+      "epoch": 13.020844029161104,
+      "grad_norm": 0.09589258581399918,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 126810
+    },
+    {
+      "epoch": 13.021152069000925,
+      "grad_norm": 0.0471031628549099,
+      "learning_rate": 0.01,
+      "loss": 1.9488,
+      "step": 126813
+    },
+    {
+      "epoch": 13.021460108840744,
+      "grad_norm": 0.08645718544721603,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 126816
+    },
+    {
+      "epoch": 13.021768148680563,
+      "grad_norm": 0.04973447695374489,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 126819
+    },
+    {
+      "epoch": 13.022076188520382,
+      "grad_norm": 0.08529239147901535,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 126822
+    },
+    {
+      "epoch": 13.0223842283602,
+      "grad_norm": 0.11720240116119385,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 126825
+    },
+    {
+      "epoch": 13.02269226820002,
+      "grad_norm": 0.11393365263938904,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 126828
+    },
+    {
+      "epoch": 13.02300030803984,
+      "grad_norm": 0.07204284518957138,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 126831
+    },
+    {
+      "epoch": 13.02330834787966,
+      "grad_norm": 0.04017645865678787,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 126834
+    },
+    {
+      "epoch": 13.023616387719478,
+      "grad_norm": 0.04806842282414436,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 126837
+    },
+    {
+      "epoch": 13.023924427559297,
+      "grad_norm": 0.1247212216258049,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 126840
+    },
+    {
+      "epoch": 13.024232467399116,
+      "grad_norm": 0.08574271947145462,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 126843
+    },
+    {
+      "epoch": 13.024540507238937,
+      "grad_norm": 0.07768067717552185,
+      "learning_rate": 0.01,
+      "loss": 1.8924,
+      "step": 126846
+    },
+    {
+      "epoch": 13.024848547078756,
+      "grad_norm": 0.03521045297384262,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 126849
+    },
+    {
+      "epoch": 13.025156586918575,
+      "grad_norm": 0.03097418136894703,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 126852
+    },
+    {
+      "epoch": 13.025464626758394,
+      "grad_norm": 0.04022853076457977,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 126855
+    },
+    {
+      "epoch": 13.025772666598213,
+      "grad_norm": 0.037963468581438065,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 126858
+    },
+    {
+      "epoch": 13.026080706438032,
+      "grad_norm": 0.1230914518237114,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 126861
+    },
+    {
+      "epoch": 13.026388746277853,
+      "grad_norm": 0.11064332723617554,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 126864
+    },
+    {
+      "epoch": 13.026696786117672,
+      "grad_norm": 0.03705328330397606,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 126867
+    },
+    {
+      "epoch": 13.02700482595749,
+      "grad_norm": 0.040886882692575455,
+      "learning_rate": 0.01,
+      "loss": 1.8956,
+      "step": 126870
+    },
+    {
+      "epoch": 13.02731286579731,
+      "grad_norm": 0.04579608142375946,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 126873
+    },
+    {
+      "epoch": 13.027620905637129,
+      "grad_norm": 0.055250927805900574,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 126876
+    },
+    {
+      "epoch": 13.027928945476948,
+      "grad_norm": 0.05872594565153122,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 126879
+    },
+    {
+      "epoch": 13.028236985316768,
+      "grad_norm": 0.074102982878685,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 126882
+    },
+    {
+      "epoch": 13.028545025156587,
+      "grad_norm": 0.05230741575360298,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 126885
+    },
+    {
+      "epoch": 13.028853064996406,
+      "grad_norm": 0.049110863357782364,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 126888
+    },
+    {
+      "epoch": 13.029161104836225,
+      "grad_norm": 0.04093809425830841,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 126891
+    },
+    {
+      "epoch": 13.029469144676044,
+      "grad_norm": 0.07468719780445099,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 126894
+    },
+    {
+      "epoch": 13.029777184515863,
+      "grad_norm": 0.10583259165287018,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 126897
+    },
+    {
+      "epoch": 13.030085224355684,
+      "grad_norm": 0.06672538071870804,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 126900
+    },
+    {
+      "epoch": 13.030393264195503,
+      "grad_norm": 0.09136092662811279,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 126903
+    },
+    {
+      "epoch": 13.030701304035322,
+      "grad_norm": 0.06731969863176346,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 126906
+    },
+    {
+      "epoch": 13.03100934387514,
+      "grad_norm": 0.11751905828714371,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 126909
+    },
+    {
+      "epoch": 13.03131738371496,
+      "grad_norm": 0.04479971155524254,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 126912
+    },
+    {
+      "epoch": 13.03162542355478,
+      "grad_norm": 0.03791626915335655,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 126915
+    },
+    {
+      "epoch": 13.0319334633946,
+      "grad_norm": 0.0481899119913578,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 126918
+    },
+    {
+      "epoch": 13.032241503234419,
+      "grad_norm": 0.07406970113515854,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 126921
+    },
+    {
+      "epoch": 13.032549543074238,
+      "grad_norm": 0.0680089145898819,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 126924
+    },
+    {
+      "epoch": 13.032857582914056,
+      "grad_norm": 0.0882832333445549,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 126927
+    },
+    {
+      "epoch": 13.033165622753875,
+      "grad_norm": 0.042513806372880936,
+      "learning_rate": 0.01,
+      "loss": 1.9538,
+      "step": 126930
+    },
+    {
+      "epoch": 13.033473662593696,
+      "grad_norm": 0.04040132462978363,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 126933
+    },
+    {
+      "epoch": 13.033781702433515,
+      "grad_norm": 0.04504537209868431,
+      "learning_rate": 0.01,
+      "loss": 1.9056,
+      "step": 126936
+    },
+    {
+      "epoch": 13.034089742273334,
+      "grad_norm": 0.07481081038713455,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 126939
+    },
+    {
+      "epoch": 13.034397782113153,
+      "grad_norm": 0.08218560367822647,
+      "learning_rate": 0.01,
+      "loss": 1.8901,
+      "step": 126942
+    },
+    {
+      "epoch": 13.034705821952972,
+      "grad_norm": 0.055728938430547714,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 126945
+    },
+    {
+      "epoch": 13.035013861792791,
+      "grad_norm": 0.12384865432977676,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 126948
+    },
+    {
+      "epoch": 13.035321901632612,
+      "grad_norm": 0.053917497396469116,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 126951
+    },
+    {
+      "epoch": 13.03562994147243,
+      "grad_norm": 0.04301830753684044,
+      "learning_rate": 0.01,
+      "loss": 1.8926,
+      "step": 126954
+    },
+    {
+      "epoch": 13.03593798131225,
+      "grad_norm": 0.04783236235380173,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 126957
+    },
+    {
+      "epoch": 13.036246021152069,
+      "grad_norm": 0.06067037209868431,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 126960
+    },
+    {
+      "epoch": 13.036554060991888,
+      "grad_norm": 0.05491422116756439,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 126963
+    },
+    {
+      "epoch": 13.036862100831707,
+      "grad_norm": 0.16311119496822357,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 126966
+    },
+    {
+      "epoch": 13.037170140671527,
+      "grad_norm": 0.10247328877449036,
+      "learning_rate": 0.01,
+      "loss": 1.8934,
+      "step": 126969
+    },
+    {
+      "epoch": 13.037478180511346,
+      "grad_norm": 0.0535706952214241,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 126972
+    },
+    {
+      "epoch": 13.037786220351165,
+      "grad_norm": 0.043741948902606964,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 126975
+    },
+    {
+      "epoch": 13.038094260190984,
+      "grad_norm": 0.11656595766544342,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 126978
+    },
+    {
+      "epoch": 13.038402300030803,
+      "grad_norm": 0.051561418920755386,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 126981
+    },
+    {
+      "epoch": 13.038710339870624,
+      "grad_norm": 0.04763190820813179,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 126984
+    },
+    {
+      "epoch": 13.039018379710443,
+      "grad_norm": 0.0440252348780632,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 126987
+    },
+    {
+      "epoch": 13.039326419550262,
+      "grad_norm": 0.03806642070412636,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 126990
+    },
+    {
+      "epoch": 13.039634459390081,
+      "grad_norm": 0.03395757079124451,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 126993
+    },
+    {
+      "epoch": 13.0399424992299,
+      "grad_norm": 0.11586293578147888,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 126996
+    },
+    {
+      "epoch": 13.040250539069719,
+      "grad_norm": 0.06549695134162903,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 126999
+    },
+    {
+      "epoch": 13.04055857890954,
+      "grad_norm": 0.04970845580101013,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 127002
+    },
+    {
+      "epoch": 13.040866618749359,
+      "grad_norm": 0.055681247264146805,
+      "learning_rate": 0.01,
+      "loss": 1.9006,
+      "step": 127005
+    },
+    {
+      "epoch": 13.041174658589178,
+      "grad_norm": 0.040567461401224136,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 127008
+    },
+    {
+      "epoch": 13.041482698428997,
+      "grad_norm": 0.041462626308202744,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 127011
+    },
+    {
+      "epoch": 13.041790738268816,
+      "grad_norm": 0.04810677096247673,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 127014
+    },
+    {
+      "epoch": 13.042098778108635,
+      "grad_norm": 0.048932626843452454,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 127017
+    },
+    {
+      "epoch": 13.042406817948455,
+      "grad_norm": 0.17227526009082794,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 127020
+    },
+    {
+      "epoch": 13.042714857788274,
+      "grad_norm": 0.13014376163482666,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 127023
+    },
+    {
+      "epoch": 13.043022897628093,
+      "grad_norm": 0.054527293890714645,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 127026
+    },
+    {
+      "epoch": 13.043330937467912,
+      "grad_norm": 0.09164586663246155,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 127029
+    },
+    {
+      "epoch": 13.043638977307731,
+      "grad_norm": 0.045358382165431976,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 127032
+    },
+    {
+      "epoch": 13.04394701714755,
+      "grad_norm": 0.057381246238946915,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 127035
+    },
+    {
+      "epoch": 13.044255056987371,
+      "grad_norm": 0.06120704859495163,
+      "learning_rate": 0.01,
+      "loss": 1.901,
+      "step": 127038
+    },
+    {
+      "epoch": 13.04456309682719,
+      "grad_norm": 0.04124023765325546,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 127041
+    },
+    {
+      "epoch": 13.044871136667009,
+      "grad_norm": 0.0411415621638298,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 127044
+    },
+    {
+      "epoch": 13.045179176506828,
+      "grad_norm": 0.04915084317326546,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 127047
+    },
+    {
+      "epoch": 13.045487216346647,
+      "grad_norm": 0.09298434853553772,
+      "learning_rate": 0.01,
+      "loss": 1.9562,
+      "step": 127050
+    },
+    {
+      "epoch": 13.045795256186468,
+      "grad_norm": 0.07847198098897934,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 127053
+    },
+    {
+      "epoch": 13.046103296026287,
+      "grad_norm": 0.1293015331029892,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 127056
+    },
+    {
+      "epoch": 13.046411335866106,
+      "grad_norm": 0.0400281585752964,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 127059
+    },
+    {
+      "epoch": 13.046719375705925,
+      "grad_norm": 0.08314623683691025,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 127062
+    },
+    {
+      "epoch": 13.047027415545744,
+      "grad_norm": 0.08865764737129211,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 127065
+    },
+    {
+      "epoch": 13.047335455385562,
+      "grad_norm": 0.16069504618644714,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 127068
+    },
+    {
+      "epoch": 13.047643495225383,
+      "grad_norm": 0.041773583739995956,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 127071
+    },
+    {
+      "epoch": 13.047951535065202,
+      "grad_norm": 0.0584380216896534,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 127074
+    },
+    {
+      "epoch": 13.048259574905021,
+      "grad_norm": 0.042118463665246964,
+      "learning_rate": 0.01,
+      "loss": 1.9576,
+      "step": 127077
+    },
+    {
+      "epoch": 13.04856761474484,
+      "grad_norm": 0.05080181732773781,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 127080
+    },
+    {
+      "epoch": 13.04887565458466,
+      "grad_norm": 0.056995194405317307,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 127083
+    },
+    {
+      "epoch": 13.049183694424478,
+      "grad_norm": 0.06421342492103577,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 127086
+    },
+    {
+      "epoch": 13.049491734264299,
+      "grad_norm": 0.048895880579948425,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 127089
+    },
+    {
+      "epoch": 13.049799774104118,
+      "grad_norm": 0.09521444886922836,
+      "learning_rate": 0.01,
+      "loss": 1.9011,
+      "step": 127092
+    },
+    {
+      "epoch": 13.050107813943937,
+      "grad_norm": 0.17821581661701202,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 127095
+    },
+    {
+      "epoch": 13.050415853783756,
+      "grad_norm": 0.07893852144479752,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 127098
+    },
+    {
+      "epoch": 13.050723893623575,
+      "grad_norm": 0.04469249024987221,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 127101
+    },
+    {
+      "epoch": 13.051031933463394,
+      "grad_norm": 0.03402012586593628,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 127104
+    },
+    {
+      "epoch": 13.051339973303214,
+      "grad_norm": 0.0767841711640358,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 127107
+    },
+    {
+      "epoch": 13.051648013143033,
+      "grad_norm": 0.1266629546880722,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 127110
+    },
+    {
+      "epoch": 13.051956052982852,
+      "grad_norm": 0.0934668481349945,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 127113
+    },
+    {
+      "epoch": 13.052264092822671,
+      "grad_norm": 0.04097825288772583,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 127116
+    },
+    {
+      "epoch": 13.05257213266249,
+      "grad_norm": 0.04565742611885071,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 127119
+    },
+    {
+      "epoch": 13.052880172502311,
+      "grad_norm": 0.04591118544340134,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 127122
+    },
+    {
+      "epoch": 13.05318821234213,
+      "grad_norm": 0.039362240582704544,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 127125
+    },
+    {
+      "epoch": 13.053496252181949,
+      "grad_norm": 0.03501315414905548,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 127128
+    },
+    {
+      "epoch": 13.053804292021768,
+      "grad_norm": 0.04087715595960617,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 127131
+    },
+    {
+      "epoch": 13.054112331861587,
+      "grad_norm": 0.07423004508018494,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 127134
+    },
+    {
+      "epoch": 13.054420371701406,
+      "grad_norm": 0.10297241061925888,
+      "learning_rate": 0.01,
+      "loss": 1.9019,
+      "step": 127137
+    },
+    {
+      "epoch": 13.054728411541227,
+      "grad_norm": 0.10126131772994995,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 127140
+    },
+    {
+      "epoch": 13.055036451381046,
+      "grad_norm": 0.05527950078248978,
+      "learning_rate": 0.01,
+      "loss": 1.8919,
+      "step": 127143
+    },
+    {
+      "epoch": 13.055344491220865,
+      "grad_norm": 0.0728352814912796,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 127146
+    },
+    {
+      "epoch": 13.055652531060684,
+      "grad_norm": 0.1472015678882599,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 127149
+    },
+    {
+      "epoch": 13.055960570900503,
+      "grad_norm": 0.08481284976005554,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 127152
+    },
+    {
+      "epoch": 13.056268610740322,
+      "grad_norm": 0.10429725795984268,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 127155
+    },
+    {
+      "epoch": 13.056576650580142,
+      "grad_norm": 0.08946308493614197,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 127158
+    },
+    {
+      "epoch": 13.056884690419961,
+      "grad_norm": 0.05603213980793953,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 127161
+    },
+    {
+      "epoch": 13.05719273025978,
+      "grad_norm": 0.03960242494940758,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 127164
+    },
+    {
+      "epoch": 13.0575007700996,
+      "grad_norm": 0.04127161204814911,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 127167
+    },
+    {
+      "epoch": 13.057808809939418,
+      "grad_norm": 0.08375523239374161,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 127170
+    },
+    {
+      "epoch": 13.058116849779237,
+      "grad_norm": 0.0448022224009037,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 127173
+    },
+    {
+      "epoch": 13.058424889619058,
+      "grad_norm": 0.12804900109767914,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 127176
+    },
+    {
+      "epoch": 13.058732929458877,
+      "grad_norm": 0.0981232151389122,
+      "learning_rate": 0.01,
+      "loss": 1.9615,
+      "step": 127179
+    },
+    {
+      "epoch": 13.059040969298696,
+      "grad_norm": 0.09918200969696045,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 127182
+    },
+    {
+      "epoch": 13.059349009138515,
+      "grad_norm": 0.06237791106104851,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 127185
+    },
+    {
+      "epoch": 13.059657048978334,
+      "grad_norm": 0.11823505908250809,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 127188
+    },
+    {
+      "epoch": 13.059965088818155,
+      "grad_norm": 0.05716945230960846,
+      "learning_rate": 0.01,
+      "loss": 1.905,
+      "step": 127191
+    },
+    {
+      "epoch": 13.060273128657974,
+      "grad_norm": 0.047965697944164276,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 127194
+    },
+    {
+      "epoch": 13.060581168497793,
+      "grad_norm": 0.05010659620165825,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 127197
+    },
+    {
+      "epoch": 13.060889208337612,
+      "grad_norm": 0.09740984439849854,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 127200
+    },
+    {
+      "epoch": 13.06119724817743,
+      "grad_norm": 0.08633100986480713,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 127203
+    },
+    {
+      "epoch": 13.06150528801725,
+      "grad_norm": 0.08889269083738327,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 127206
+    },
+    {
+      "epoch": 13.06181332785707,
+      "grad_norm": 0.07049471884965897,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 127209
+    },
+    {
+      "epoch": 13.06212136769689,
+      "grad_norm": 0.08995243161916733,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 127212
+    },
+    {
+      "epoch": 13.062429407536708,
+      "grad_norm": 0.1442023515701294,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 127215
+    },
+    {
+      "epoch": 13.062737447376527,
+      "grad_norm": 0.056450746953487396,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 127218
+    },
+    {
+      "epoch": 13.063045487216346,
+      "grad_norm": 0.0538800023496151,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 127221
+    },
+    {
+      "epoch": 13.063353527056165,
+      "grad_norm": 0.03840062767267227,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 127224
+    },
+    {
+      "epoch": 13.063661566895986,
+      "grad_norm": 0.05939144268631935,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 127227
+    },
+    {
+      "epoch": 13.063969606735805,
+      "grad_norm": 0.041396938264369965,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 127230
+    },
+    {
+      "epoch": 13.064277646575624,
+      "grad_norm": 0.04435405507683754,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 127233
+    },
+    {
+      "epoch": 13.064585686415443,
+      "grad_norm": 0.04468780383467674,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 127236
+    },
+    {
+      "epoch": 13.064893726255262,
+      "grad_norm": 0.10944616049528122,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 127239
+    },
+    {
+      "epoch": 13.06520176609508,
+      "grad_norm": 0.061596062034368515,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 127242
+    },
+    {
+      "epoch": 13.065509805934902,
+      "grad_norm": 0.06862606853246689,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 127245
+    },
+    {
+      "epoch": 13.06581784577472,
+      "grad_norm": 0.053111523389816284,
+      "learning_rate": 0.01,
+      "loss": 1.9062,
+      "step": 127248
+    },
+    {
+      "epoch": 13.06612588561454,
+      "grad_norm": 0.09948468208312988,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 127251
+    },
+    {
+      "epoch": 13.066433925454358,
+      "grad_norm": 0.04883921146392822,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 127254
+    },
+    {
+      "epoch": 13.066741965294177,
+      "grad_norm": 0.10840325802564621,
+      "learning_rate": 0.01,
+      "loss": 1.9002,
+      "step": 127257
+    },
+    {
+      "epoch": 13.067050005133998,
+      "grad_norm": 0.10582426190376282,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 127260
+    },
+    {
+      "epoch": 13.067358044973817,
+      "grad_norm": 0.035586949437856674,
+      "learning_rate": 0.01,
+      "loss": 1.9571,
+      "step": 127263
+    },
+    {
+      "epoch": 13.067666084813636,
+      "grad_norm": 0.10024740546941757,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 127266
+    },
+    {
+      "epoch": 13.067974124653455,
+      "grad_norm": 0.05136268958449364,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 127269
+    },
+    {
+      "epoch": 13.068282164493274,
+      "grad_norm": 0.05003959685564041,
+      "learning_rate": 0.01,
+      "loss": 1.8979,
+      "step": 127272
+    },
+    {
+      "epoch": 13.068590204333093,
+      "grad_norm": 0.04421168938279152,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 127275
+    },
+    {
+      "epoch": 13.068898244172914,
+      "grad_norm": 0.04345612972974777,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 127278
+    },
+    {
+      "epoch": 13.069206284012733,
+      "grad_norm": 0.052260421216487885,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 127281
+    },
+    {
+      "epoch": 13.069514323852552,
+      "grad_norm": 0.11148987710475922,
+      "learning_rate": 0.01,
+      "loss": 1.8988,
+      "step": 127284
+    },
+    {
+      "epoch": 13.06982236369237,
+      "grad_norm": 0.12439928203821182,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 127287
+    },
+    {
+      "epoch": 13.07013040353219,
+      "grad_norm": 0.05130310729146004,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 127290
+    },
+    {
+      "epoch": 13.070438443372009,
+      "grad_norm": 0.10837828367948532,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 127293
+    },
+    {
+      "epoch": 13.07074648321183,
+      "grad_norm": 0.041389357298612595,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 127296
+    },
+    {
+      "epoch": 13.071054523051648,
+      "grad_norm": 0.039925459772348404,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 127299
+    },
+    {
+      "epoch": 13.071362562891467,
+      "grad_norm": 0.05223574861884117,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 127302
+    },
+    {
+      "epoch": 13.071670602731286,
+      "grad_norm": 0.04174607992172241,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 127305
+    },
+    {
+      "epoch": 13.071978642571105,
+      "grad_norm": 0.059435226023197174,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 127308
+    },
+    {
+      "epoch": 13.072286682410924,
+      "grad_norm": 0.12726831436157227,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 127311
+    },
+    {
+      "epoch": 13.072594722250745,
+      "grad_norm": 0.14294598996639252,
+      "learning_rate": 0.01,
+      "loss": 1.9537,
+      "step": 127314
+    },
+    {
+      "epoch": 13.072902762090564,
+      "grad_norm": 0.04655228182673454,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 127317
+    },
+    {
+      "epoch": 13.073210801930383,
+      "grad_norm": 0.03285637125372887,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 127320
+    },
+    {
+      "epoch": 13.073518841770202,
+      "grad_norm": 0.04555371776223183,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 127323
+    },
+    {
+      "epoch": 13.073826881610021,
+      "grad_norm": 0.040367040783166885,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 127326
+    },
+    {
+      "epoch": 13.074134921449842,
+      "grad_norm": 0.03395960479974747,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 127329
+    },
+    {
+      "epoch": 13.07444296128966,
+      "grad_norm": 0.10771522670984268,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 127332
+    },
+    {
+      "epoch": 13.07475100112948,
+      "grad_norm": 0.05591186508536339,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 127335
+    },
+    {
+      "epoch": 13.075059040969299,
+      "grad_norm": 0.061584796756505966,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 127338
+    },
+    {
+      "epoch": 13.075367080809118,
+      "grad_norm": 0.042033687233924866,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 127341
+    },
+    {
+      "epoch": 13.075675120648937,
+      "grad_norm": 0.0761541873216629,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 127344
+    },
+    {
+      "epoch": 13.075983160488757,
+      "grad_norm": 0.06704878807067871,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 127347
+    },
+    {
+      "epoch": 13.076291200328576,
+      "grad_norm": 0.10433320701122284,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 127350
+    },
+    {
+      "epoch": 13.076599240168395,
+      "grad_norm": 0.04487515613436699,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 127353
+    },
+    {
+      "epoch": 13.076907280008214,
+      "grad_norm": 0.07208144664764404,
+      "learning_rate": 0.01,
+      "loss": 1.9543,
+      "step": 127356
+    },
+    {
+      "epoch": 13.077215319848033,
+      "grad_norm": 0.0855809673666954,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 127359
+    },
+    {
+      "epoch": 13.077523359687852,
+      "grad_norm": 0.04038161039352417,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 127362
+    },
+    {
+      "epoch": 13.077831399527673,
+      "grad_norm": 0.08981285244226456,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 127365
+    },
+    {
+      "epoch": 13.078139439367492,
+      "grad_norm": 0.06221736595034599,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 127368
+    },
+    {
+      "epoch": 13.07844747920731,
+      "grad_norm": 0.08635222166776657,
+      "learning_rate": 0.01,
+      "loss": 1.8968,
+      "step": 127371
+    },
+    {
+      "epoch": 13.07875551904713,
+      "grad_norm": 0.06816978007555008,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 127374
+    },
+    {
+      "epoch": 13.079063558886949,
+      "grad_norm": 0.0782097801566124,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 127377
+    },
+    {
+      "epoch": 13.079371598726768,
+      "grad_norm": 0.14180225133895874,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 127380
+    },
+    {
+      "epoch": 13.079679638566589,
+      "grad_norm": 0.11078017204999924,
+      "learning_rate": 0.01,
+      "loss": 1.8963,
+      "step": 127383
+    },
+    {
+      "epoch": 13.079987678406408,
+      "grad_norm": 0.05348137766122818,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 127386
+    },
+    {
+      "epoch": 13.080295718246226,
+      "grad_norm": 0.06612905114889145,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 127389
+    },
+    {
+      "epoch": 13.080603758086045,
+      "grad_norm": 0.04810092970728874,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 127392
+    },
+    {
+      "epoch": 13.080911797925864,
+      "grad_norm": 0.05613093823194504,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 127395
+    },
+    {
+      "epoch": 13.081219837765685,
+      "grad_norm": 0.03763555735349655,
+      "learning_rate": 0.01,
+      "loss": 1.9021,
+      "step": 127398
+    },
+    {
+      "epoch": 13.081527877605504,
+      "grad_norm": 0.04689381271600723,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 127401
+    },
+    {
+      "epoch": 13.081835917445323,
+      "grad_norm": 0.03601065278053284,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 127404
+    },
+    {
+      "epoch": 13.082143957285142,
+      "grad_norm": 0.10324038565158844,
+      "learning_rate": 0.01,
+      "loss": 1.8979,
+      "step": 127407
+    },
+    {
+      "epoch": 13.082451997124961,
+      "grad_norm": 0.1330144703388214,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 127410
+    },
+    {
+      "epoch": 13.08276003696478,
+      "grad_norm": 0.10919608920812607,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 127413
+    },
+    {
+      "epoch": 13.0830680768046,
+      "grad_norm": 0.09014541655778885,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 127416
+    },
+    {
+      "epoch": 13.08337611664442,
+      "grad_norm": 0.04649462550878525,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 127419
+    },
+    {
+      "epoch": 13.083684156484239,
+      "grad_norm": 0.033851947635412216,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 127422
+    },
+    {
+      "epoch": 13.083992196324058,
+      "grad_norm": 0.04000790789723396,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 127425
+    },
+    {
+      "epoch": 13.084300236163877,
+      "grad_norm": 0.05568542703986168,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 127428
+    },
+    {
+      "epoch": 13.084608276003696,
+      "grad_norm": 0.06895308941602707,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 127431
+    },
+    {
+      "epoch": 13.084916315843516,
+      "grad_norm": 0.03300630673766136,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 127434
+    },
+    {
+      "epoch": 13.085224355683335,
+      "grad_norm": 0.08917191624641418,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 127437
+    },
+    {
+      "epoch": 13.085532395523154,
+      "grad_norm": 0.045750368386507034,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 127440
+    },
+    {
+      "epoch": 13.085840435362973,
+      "grad_norm": 0.05947425588965416,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 127443
+    },
+    {
+      "epoch": 13.086148475202792,
+      "grad_norm": 0.05453433841466904,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 127446
+    },
+    {
+      "epoch": 13.086456515042611,
+      "grad_norm": 0.12132017314434052,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 127449
+    },
+    {
+      "epoch": 13.086764554882432,
+      "grad_norm": 0.08690575510263443,
+      "learning_rate": 0.01,
+      "loss": 1.9582,
+      "step": 127452
+    },
+    {
+      "epoch": 13.087072594722251,
+      "grad_norm": 0.07652682065963745,
+      "learning_rate": 0.01,
+      "loss": 1.9454,
+      "step": 127455
+    },
+    {
+      "epoch": 13.08738063456207,
+      "grad_norm": 0.04850516840815544,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 127458
+    },
+    {
+      "epoch": 13.087688674401889,
+      "grad_norm": 0.054834362119436264,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 127461
+    },
+    {
+      "epoch": 13.087996714241708,
+      "grad_norm": 0.04554072022438049,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 127464
+    },
+    {
+      "epoch": 13.088304754081529,
+      "grad_norm": 0.1364208161830902,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 127467
+    },
+    {
+      "epoch": 13.088612793921348,
+      "grad_norm": 0.05425235629081726,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 127470
+    },
+    {
+      "epoch": 13.088920833761167,
+      "grad_norm": 0.06443481147289276,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 127473
+    },
+    {
+      "epoch": 13.089228873600986,
+      "grad_norm": 0.03747593238949776,
+      "learning_rate": 0.01,
+      "loss": 1.8853,
+      "step": 127476
+    },
+    {
+      "epoch": 13.089536913440805,
+      "grad_norm": 0.04236375913023949,
+      "learning_rate": 0.01,
+      "loss": 1.9075,
+      "step": 127479
+    },
+    {
+      "epoch": 13.089844953280624,
+      "grad_norm": 0.037440817803144455,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 127482
+    },
+    {
+      "epoch": 13.090152993120444,
+      "grad_norm": 0.0762670561671257,
+      "learning_rate": 0.01,
+      "loss": 1.9498,
+      "step": 127485
+    },
+    {
+      "epoch": 13.090461032960263,
+      "grad_norm": 0.05444483831524849,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 127488
+    },
+    {
+      "epoch": 13.090769072800082,
+      "grad_norm": 0.10359358787536621,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 127491
+    },
+    {
+      "epoch": 13.091077112639901,
+      "grad_norm": 0.044725432991981506,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 127494
+    },
+    {
+      "epoch": 13.09138515247972,
+      "grad_norm": 0.10275741666555405,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 127497
+    },
+    {
+      "epoch": 13.09169319231954,
+      "grad_norm": 0.05311112478375435,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 127500
+    },
+    {
+      "epoch": 13.09200123215936,
+      "grad_norm": 0.04492788016796112,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 127503
+    },
+    {
+      "epoch": 13.092309271999179,
+      "grad_norm": 0.06422276049852371,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 127506
+    },
+    {
+      "epoch": 13.092617311838998,
+      "grad_norm": 0.12240847200155258,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 127509
+    },
+    {
+      "epoch": 13.092925351678817,
+      "grad_norm": 0.04907319322228432,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 127512
+    },
+    {
+      "epoch": 13.093233391518636,
+      "grad_norm": 0.05358103662729263,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 127515
+    },
+    {
+      "epoch": 13.093541431358455,
+      "grad_norm": 0.03970174491405487,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 127518
+    },
+    {
+      "epoch": 13.093849471198276,
+      "grad_norm": 0.058363888412714005,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 127521
+    },
+    {
+      "epoch": 13.094157511038095,
+      "grad_norm": 0.03669978678226471,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 127524
+    },
+    {
+      "epoch": 13.094465550877914,
+      "grad_norm": 0.14771434664726257,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 127527
+    },
+    {
+      "epoch": 13.094773590717733,
+      "grad_norm": 0.06578870117664337,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 127530
+    },
+    {
+      "epoch": 13.095081630557551,
+      "grad_norm": 0.13467788696289062,
+      "learning_rate": 0.01,
+      "loss": 1.944,
+      "step": 127533
+    },
+    {
+      "epoch": 13.095389670397372,
+      "grad_norm": 0.0728551521897316,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 127536
+    },
+    {
+      "epoch": 13.095697710237191,
+      "grad_norm": 0.04833552986383438,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 127539
+    },
+    {
+      "epoch": 13.09600575007701,
+      "grad_norm": 0.03024047240614891,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 127542
+    },
+    {
+      "epoch": 13.09631378991683,
+      "grad_norm": 0.03338846564292908,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 127545
+    },
+    {
+      "epoch": 13.096621829756648,
+      "grad_norm": 0.0696599930524826,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 127548
+    },
+    {
+      "epoch": 13.096929869596467,
+      "grad_norm": 0.06427241861820221,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 127551
+    },
+    {
+      "epoch": 13.097237909436288,
+      "grad_norm": 0.0635237917304039,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 127554
+    },
+    {
+      "epoch": 13.097545949276107,
+      "grad_norm": 0.05661383643746376,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 127557
+    },
+    {
+      "epoch": 13.097853989115926,
+      "grad_norm": 0.14469201862812042,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 127560
+    },
+    {
+      "epoch": 13.098162028955745,
+      "grad_norm": 0.11612505465745926,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 127563
+    },
+    {
+      "epoch": 13.098470068795564,
+      "grad_norm": 0.11548126488924026,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 127566
+    },
+    {
+      "epoch": 13.098778108635383,
+      "grad_norm": 0.10726878046989441,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 127569
+    },
+    {
+      "epoch": 13.099086148475203,
+      "grad_norm": 0.07961736619472504,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 127572
+    },
+    {
+      "epoch": 13.099394188315022,
+      "grad_norm": 0.0513598807156086,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 127575
+    },
+    {
+      "epoch": 13.099702228154841,
+      "grad_norm": 0.034828685224056244,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 127578
+    },
+    {
+      "epoch": 13.10001026799466,
+      "grad_norm": 0.034888122230768204,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 127581
+    },
+    {
+      "epoch": 13.10031830783448,
+      "grad_norm": 0.06885156780481339,
+      "learning_rate": 0.01,
+      "loss": 1.95,
+      "step": 127584
+    },
+    {
+      "epoch": 13.100626347674298,
+      "grad_norm": 0.06749074161052704,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 127587
+    },
+    {
+      "epoch": 13.100934387514119,
+      "grad_norm": 0.04102785140275955,
+      "learning_rate": 0.01,
+      "loss": 1.9006,
+      "step": 127590
+    },
+    {
+      "epoch": 13.101242427353938,
+      "grad_norm": 0.047465551644563675,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 127593
+    },
+    {
+      "epoch": 13.101550467193757,
+      "grad_norm": 0.1252158135175705,
+      "learning_rate": 0.01,
+      "loss": 1.9011,
+      "step": 127596
+    },
+    {
+      "epoch": 13.101858507033576,
+      "grad_norm": 0.043445441871881485,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 127599
+    },
+    {
+      "epoch": 13.102166546873395,
+      "grad_norm": 0.09905412793159485,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 127602
+    },
+    {
+      "epoch": 13.102474586713216,
+      "grad_norm": 0.12451162934303284,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 127605
+    },
+    {
+      "epoch": 13.102782626553035,
+      "grad_norm": 0.07141668349504471,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 127608
+    },
+    {
+      "epoch": 13.103090666392854,
+      "grad_norm": 0.04872805252671242,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 127611
+    },
+    {
+      "epoch": 13.103398706232673,
+      "grad_norm": 0.045460961759090424,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 127614
+    },
+    {
+      "epoch": 13.103706746072492,
+      "grad_norm": 0.05963927507400513,
+      "learning_rate": 0.01,
+      "loss": 1.9495,
+      "step": 127617
+    },
+    {
+      "epoch": 13.10401478591231,
+      "grad_norm": 0.10401326417922974,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 127620
+    },
+    {
+      "epoch": 13.104322825752131,
+      "grad_norm": 0.08146955072879791,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 127623
+    },
+    {
+      "epoch": 13.10463086559195,
+      "grad_norm": 0.09373153746128082,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 127626
+    },
+    {
+      "epoch": 13.10493890543177,
+      "grad_norm": 0.03268011659383774,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 127629
+    },
+    {
+      "epoch": 13.105246945271588,
+      "grad_norm": 0.09914293140172958,
+      "learning_rate": 0.01,
+      "loss": 1.9019,
+      "step": 127632
+    },
+    {
+      "epoch": 13.105554985111407,
+      "grad_norm": 0.06838401407003403,
+      "learning_rate": 0.01,
+      "loss": 1.8944,
+      "step": 127635
+    },
+    {
+      "epoch": 13.105863024951226,
+      "grad_norm": 0.07944194972515106,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 127638
+    },
+    {
+      "epoch": 13.106171064791047,
+      "grad_norm": 0.08194562792778015,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 127641
+    },
+    {
+      "epoch": 13.106479104630866,
+      "grad_norm": 0.08502616733312607,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 127644
+    },
+    {
+      "epoch": 13.106787144470685,
+      "grad_norm": 0.03694474324584007,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 127647
+    },
+    {
+      "epoch": 13.107095184310504,
+      "grad_norm": 0.05780287832021713,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 127650
+    },
+    {
+      "epoch": 13.107403224150323,
+      "grad_norm": 0.10309597104787827,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 127653
+    },
+    {
+      "epoch": 13.107711263990142,
+      "grad_norm": 0.03319085016846657,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 127656
+    },
+    {
+      "epoch": 13.108019303829963,
+      "grad_norm": 0.07841961830854416,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 127659
+    },
+    {
+      "epoch": 13.108327343669782,
+      "grad_norm": 0.09356290102005005,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 127662
+    },
+    {
+      "epoch": 13.1086353835096,
+      "grad_norm": 0.04366067424416542,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 127665
+    },
+    {
+      "epoch": 13.10894342334942,
+      "grad_norm": 0.04745535925030708,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 127668
+    },
+    {
+      "epoch": 13.109251463189239,
+      "grad_norm": 0.044011637568473816,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 127671
+    },
+    {
+      "epoch": 13.10955950302906,
+      "grad_norm": 0.030939612537622452,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 127674
+    },
+    {
+      "epoch": 13.109867542868878,
+      "grad_norm": 0.07078239321708679,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 127677
+    },
+    {
+      "epoch": 13.110175582708697,
+      "grad_norm": 0.14157222211360931,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 127680
+    },
+    {
+      "epoch": 13.110483622548516,
+      "grad_norm": 0.17908212542533875,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 127683
+    },
+    {
+      "epoch": 13.110791662388335,
+      "grad_norm": 0.12797129154205322,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 127686
+    },
+    {
+      "epoch": 13.111099702228154,
+      "grad_norm": 0.040857430547475815,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 127689
+    },
+    {
+      "epoch": 13.111407742067975,
+      "grad_norm": 0.03708762302994728,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 127692
+    },
+    {
+      "epoch": 13.111715781907794,
+      "grad_norm": 0.0348348431289196,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 127695
+    },
+    {
+      "epoch": 13.112023821747613,
+      "grad_norm": 0.05099273845553398,
+      "learning_rate": 0.01,
+      "loss": 1.9101,
+      "step": 127698
+    },
+    {
+      "epoch": 13.112331861587432,
+      "grad_norm": 0.05459226667881012,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 127701
+    },
+    {
+      "epoch": 13.11263990142725,
+      "grad_norm": 0.04088655859231949,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 127704
+    },
+    {
+      "epoch": 13.11294794126707,
+      "grad_norm": 0.03913631662726402,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 127707
+    },
+    {
+      "epoch": 13.11325598110689,
+      "grad_norm": 0.03574163094162941,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 127710
+    },
+    {
+      "epoch": 13.11356402094671,
+      "grad_norm": 0.0599394217133522,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 127713
+    },
+    {
+      "epoch": 13.113872060786528,
+      "grad_norm": 0.16355867683887482,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 127716
+    },
+    {
+      "epoch": 13.114180100626347,
+      "grad_norm": 0.18325865268707275,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 127719
+    },
+    {
+      "epoch": 13.114488140466166,
+      "grad_norm": 0.09668468683958054,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 127722
+    },
+    {
+      "epoch": 13.114796180305985,
+      "grad_norm": 0.039592985063791275,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 127725
+    },
+    {
+      "epoch": 13.115104220145806,
+      "grad_norm": 0.03827126696705818,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 127728
+    },
+    {
+      "epoch": 13.115412259985625,
+      "grad_norm": 0.07587017863988876,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 127731
+    },
+    {
+      "epoch": 13.115720299825444,
+      "grad_norm": 0.05452830344438553,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 127734
+    },
+    {
+      "epoch": 13.116028339665263,
+      "grad_norm": 0.10361320525407791,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 127737
+    },
+    {
+      "epoch": 13.116336379505082,
+      "grad_norm": 0.0510588139295578,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 127740
+    },
+    {
+      "epoch": 13.116644419344903,
+      "grad_norm": 0.05836412310600281,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 127743
+    },
+    {
+      "epoch": 13.116952459184722,
+      "grad_norm": 0.04067745804786682,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 127746
+    },
+    {
+      "epoch": 13.11726049902454,
+      "grad_norm": 0.034297723323106766,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 127749
+    },
+    {
+      "epoch": 13.11756853886436,
+      "grad_norm": 0.05775736644864082,
+      "learning_rate": 0.01,
+      "loss": 1.9083,
+      "step": 127752
+    },
+    {
+      "epoch": 13.117876578704179,
+      "grad_norm": 0.07246290147304535,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 127755
+    },
+    {
+      "epoch": 13.118184618543998,
+      "grad_norm": 0.09892375767230988,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 127758
+    },
+    {
+      "epoch": 13.118492658383818,
+      "grad_norm": 0.06460148841142654,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 127761
+    },
+    {
+      "epoch": 13.118800698223637,
+      "grad_norm": 0.07184730470180511,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 127764
+    },
+    {
+      "epoch": 13.119108738063456,
+      "grad_norm": 0.053893525153398514,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 127767
+    },
+    {
+      "epoch": 13.119416777903275,
+      "grad_norm": 0.041147246956825256,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 127770
+    },
+    {
+      "epoch": 13.119724817743094,
+      "grad_norm": 0.04412025958299637,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 127773
+    },
+    {
+      "epoch": 13.120032857582913,
+      "grad_norm": 0.09897839277982712,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 127776
+    },
+    {
+      "epoch": 13.120340897422734,
+      "grad_norm": 0.07074426114559174,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 127779
+    },
+    {
+      "epoch": 13.120648937262553,
+      "grad_norm": 0.07139520347118378,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 127782
+    },
+    {
+      "epoch": 13.120956977102372,
+      "grad_norm": 0.093126080930233,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 127785
+    },
+    {
+      "epoch": 13.121265016942191,
+      "grad_norm": 0.09182556718587875,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 127788
+    },
+    {
+      "epoch": 13.12157305678201,
+      "grad_norm": 0.12576201558113098,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 127791
+    },
+    {
+      "epoch": 13.121881096621829,
+      "grad_norm": 0.06775187700986862,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 127794
+    },
+    {
+      "epoch": 13.12218913646165,
+      "grad_norm": 0.04573152959346771,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 127797
+    },
+    {
+      "epoch": 13.122497176301469,
+      "grad_norm": 0.046152930706739426,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 127800
+    },
+    {
+      "epoch": 13.122805216141288,
+      "grad_norm": 0.09227651357650757,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 127803
+    },
+    {
+      "epoch": 13.123113255981107,
+      "grad_norm": 0.05411921441555023,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 127806
+    },
+    {
+      "epoch": 13.123421295820926,
+      "grad_norm": 0.04880136623978615,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 127809
+    },
+    {
+      "epoch": 13.123729335660746,
+      "grad_norm": 0.04019276052713394,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 127812
+    },
+    {
+      "epoch": 13.124037375500565,
+      "grad_norm": 0.04696383327245712,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 127815
+    },
+    {
+      "epoch": 13.124345415340384,
+      "grad_norm": 0.1439487338066101,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 127818
+    },
+    {
+      "epoch": 13.124653455180203,
+      "grad_norm": 0.042858440428972244,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 127821
+    },
+    {
+      "epoch": 13.124961495020022,
+      "grad_norm": 0.05714735388755798,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 127824
+    },
+    {
+      "epoch": 13.125269534859841,
+      "grad_norm": 0.09066656231880188,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 127827
+    },
+    {
+      "epoch": 13.125577574699662,
+      "grad_norm": 0.06780602782964706,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 127830
+    },
+    {
+      "epoch": 13.12588561453948,
+      "grad_norm": 0.1103760302066803,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 127833
+    },
+    {
+      "epoch": 13.1261936543793,
+      "grad_norm": 0.11341952532529831,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 127836
+    },
+    {
+      "epoch": 13.126501694219119,
+      "grad_norm": 0.06877338886260986,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 127839
+    },
+    {
+      "epoch": 13.126809734058938,
+      "grad_norm": 0.04744340106844902,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 127842
+    },
+    {
+      "epoch": 13.127117773898757,
+      "grad_norm": 0.11649034917354584,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 127845
+    },
+    {
+      "epoch": 13.127425813738578,
+      "grad_norm": 0.07478306442499161,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 127848
+    },
+    {
+      "epoch": 13.127733853578397,
+      "grad_norm": 0.05415070429444313,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 127851
+    },
+    {
+      "epoch": 13.128041893418215,
+      "grad_norm": 0.04640885815024376,
+      "learning_rate": 0.01,
+      "loss": 1.8943,
+      "step": 127854
+    },
+    {
+      "epoch": 13.128349933258034,
+      "grad_norm": 0.03679944947361946,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 127857
+    },
+    {
+      "epoch": 13.128657973097853,
+      "grad_norm": 0.2052852064371109,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 127860
+    },
+    {
+      "epoch": 13.128966012937672,
+      "grad_norm": 0.06634194403886795,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 127863
+    },
+    {
+      "epoch": 13.129274052777493,
+      "grad_norm": 0.0557449609041214,
+      "learning_rate": 0.01,
+      "loss": 1.9497,
+      "step": 127866
+    },
+    {
+      "epoch": 13.129582092617312,
+      "grad_norm": 0.045638907700777054,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 127869
+    },
+    {
+      "epoch": 13.129890132457131,
+      "grad_norm": 0.0392153300344944,
+      "learning_rate": 0.01,
+      "loss": 1.9443,
+      "step": 127872
+    },
+    {
+      "epoch": 13.13019817229695,
+      "grad_norm": 0.05158400535583496,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 127875
+    },
+    {
+      "epoch": 13.130506212136769,
+      "grad_norm": 0.03828379139304161,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 127878
+    },
+    {
+      "epoch": 13.13081425197659,
+      "grad_norm": 0.042600132524967194,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 127881
+    },
+    {
+      "epoch": 13.131122291816409,
+      "grad_norm": 0.057917337864637375,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 127884
+    },
+    {
+      "epoch": 13.131430331656228,
+      "grad_norm": 0.14073272049427032,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 127887
+    },
+    {
+      "epoch": 13.131738371496047,
+      "grad_norm": 0.04800724238157272,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 127890
+    },
+    {
+      "epoch": 13.132046411335866,
+      "grad_norm": 0.047366831451654434,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 127893
+    },
+    {
+      "epoch": 13.132354451175685,
+      "grad_norm": 0.06576810032129288,
+      "learning_rate": 0.01,
+      "loss": 1.9006,
+      "step": 127896
+    },
+    {
+      "epoch": 13.132662491015505,
+      "grad_norm": 0.049865636974573135,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 127899
+    },
+    {
+      "epoch": 13.132970530855324,
+      "grad_norm": 0.046612344682216644,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 127902
+    },
+    {
+      "epoch": 13.133278570695143,
+      "grad_norm": 0.040821097791194916,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 127905
+    },
+    {
+      "epoch": 13.133586610534962,
+      "grad_norm": 0.042939893901348114,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 127908
+    },
+    {
+      "epoch": 13.133894650374781,
+      "grad_norm": 0.06516589224338531,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 127911
+    },
+    {
+      "epoch": 13.1342026902146,
+      "grad_norm": 0.06409568339586258,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 127914
+    },
+    {
+      "epoch": 13.134510730054421,
+      "grad_norm": 0.11601393669843674,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 127917
+    },
+    {
+      "epoch": 13.13481876989424,
+      "grad_norm": 0.047946587204933167,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 127920
+    },
+    {
+      "epoch": 13.135126809734059,
+      "grad_norm": 0.08934775739908218,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 127923
+    },
+    {
+      "epoch": 13.135434849573878,
+      "grad_norm": 0.05754408985376358,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 127926
+    },
+    {
+      "epoch": 13.135742889413697,
+      "grad_norm": 0.11593315005302429,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 127929
+    },
+    {
+      "epoch": 13.136050929253516,
+      "grad_norm": 0.03449985757470131,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 127932
+    },
+    {
+      "epoch": 13.136358969093337,
+      "grad_norm": 0.05720939859747887,
+      "learning_rate": 0.01,
+      "loss": 1.9037,
+      "step": 127935
+    },
+    {
+      "epoch": 13.136667008933156,
+      "grad_norm": 0.08307376503944397,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 127938
+    },
+    {
+      "epoch": 13.136975048772975,
+      "grad_norm": 0.07763773947954178,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 127941
+    },
+    {
+      "epoch": 13.137283088612794,
+      "grad_norm": 0.05950096249580383,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 127944
+    },
+    {
+      "epoch": 13.137591128452613,
+      "grad_norm": 0.036755166947841644,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 127947
+    },
+    {
+      "epoch": 13.137899168292433,
+      "grad_norm": 0.04108569398522377,
+      "learning_rate": 0.01,
+      "loss": 1.9032,
+      "step": 127950
+    },
+    {
+      "epoch": 13.138207208132252,
+      "grad_norm": 0.03706226870417595,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 127953
+    },
+    {
+      "epoch": 13.138515247972071,
+      "grad_norm": 0.0349152497947216,
+      "learning_rate": 0.01,
+      "loss": 1.9013,
+      "step": 127956
+    },
+    {
+      "epoch": 13.13882328781189,
+      "grad_norm": 0.06268630921840668,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 127959
+    },
+    {
+      "epoch": 13.13913132765171,
+      "grad_norm": 0.06357832252979279,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 127962
+    },
+    {
+      "epoch": 13.139439367491528,
+      "grad_norm": 0.05179307609796524,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 127965
+    },
+    {
+      "epoch": 13.139747407331349,
+      "grad_norm": 0.05599494278430939,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 127968
+    },
+    {
+      "epoch": 13.140055447171168,
+      "grad_norm": 0.1250307261943817,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 127971
+    },
+    {
+      "epoch": 13.140363487010987,
+      "grad_norm": 0.06459517776966095,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 127974
+    },
+    {
+      "epoch": 13.140671526850806,
+      "grad_norm": 0.0626378059387207,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 127977
+    },
+    {
+      "epoch": 13.140979566690625,
+      "grad_norm": 0.0509517528116703,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 127980
+    },
+    {
+      "epoch": 13.141287606530444,
+      "grad_norm": 0.11793696135282516,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 127983
+    },
+    {
+      "epoch": 13.141595646370265,
+      "grad_norm": 0.0851455107331276,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 127986
+    },
+    {
+      "epoch": 13.141903686210084,
+      "grad_norm": 0.08695690333843231,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 127989
+    },
+    {
+      "epoch": 13.142211726049903,
+      "grad_norm": 0.07001544535160065,
+      "learning_rate": 0.01,
+      "loss": 1.9551,
+      "step": 127992
+    },
+    {
+      "epoch": 13.142519765889721,
+      "grad_norm": 0.035582929849624634,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 127995
+    },
+    {
+      "epoch": 13.14282780572954,
+      "grad_norm": 0.05355009436607361,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 127998
+    },
+    {
+      "epoch": 13.14313584556936,
+      "grad_norm": 0.05123375728726387,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 128001
+    },
+    {
+      "epoch": 13.14344388540918,
+      "grad_norm": 0.06289832293987274,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 128004
+    },
+    {
+      "epoch": 13.143751925249,
+      "grad_norm": 0.03586145490407944,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 128007
+    },
+    {
+      "epoch": 13.144059965088818,
+      "grad_norm": 0.036939725279808044,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 128010
+    },
+    {
+      "epoch": 13.144368004928637,
+      "grad_norm": 0.03465539589524269,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 128013
+    },
+    {
+      "epoch": 13.144676044768456,
+      "grad_norm": 0.04420056566596031,
+      "learning_rate": 0.01,
+      "loss": 1.8912,
+      "step": 128016
+    },
+    {
+      "epoch": 13.144984084608277,
+      "grad_norm": 0.10149798542261124,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 128019
+    },
+    {
+      "epoch": 13.145292124448096,
+      "grad_norm": 0.11618292331695557,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 128022
+    },
+    {
+      "epoch": 13.145600164287915,
+      "grad_norm": 0.04930877313017845,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 128025
+    },
+    {
+      "epoch": 13.145908204127734,
+      "grad_norm": 0.04495576396584511,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 128028
+    },
+    {
+      "epoch": 13.146216243967553,
+      "grad_norm": 0.10569396615028381,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 128031
+    },
+    {
+      "epoch": 13.146524283807372,
+      "grad_norm": 0.03143114969134331,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 128034
+    },
+    {
+      "epoch": 13.146832323647192,
+      "grad_norm": 0.09436129033565521,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 128037
+    },
+    {
+      "epoch": 13.147140363487011,
+      "grad_norm": 0.12870575487613678,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 128040
+    },
+    {
+      "epoch": 13.14744840332683,
+      "grad_norm": 0.07433728128671646,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 128043
+    },
+    {
+      "epoch": 13.14775644316665,
+      "grad_norm": 0.05593243986368179,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 128046
+    },
+    {
+      "epoch": 13.148064483006468,
+      "grad_norm": 0.08496079593896866,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 128049
+    },
+    {
+      "epoch": 13.148372522846287,
+      "grad_norm": 0.054458409547805786,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 128052
+    },
+    {
+      "epoch": 13.148680562686108,
+      "grad_norm": 0.037430498749017715,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 128055
+    },
+    {
+      "epoch": 13.148988602525927,
+      "grad_norm": 0.05753842741250992,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 128058
+    },
+    {
+      "epoch": 13.149296642365746,
+      "grad_norm": 0.05084479972720146,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 128061
+    },
+    {
+      "epoch": 13.149604682205565,
+      "grad_norm": 0.04421908035874367,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 128064
+    },
+    {
+      "epoch": 13.149912722045384,
+      "grad_norm": 0.056413162499666214,
+      "learning_rate": 0.01,
+      "loss": 1.9441,
+      "step": 128067
+    },
+    {
+      "epoch": 13.150220761885203,
+      "grad_norm": 0.041831210255622864,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 128070
+    },
+    {
+      "epoch": 13.150528801725024,
+      "grad_norm": 0.091091088950634,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 128073
+    },
+    {
+      "epoch": 13.150836841564843,
+      "grad_norm": 0.07474508136510849,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 128076
+    },
+    {
+      "epoch": 13.151144881404662,
+      "grad_norm": 0.08757440745830536,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 128079
+    },
+    {
+      "epoch": 13.15145292124448,
+      "grad_norm": 0.03952433913946152,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 128082
+    },
+    {
+      "epoch": 13.1517609610843,
+      "grad_norm": 0.051735635846853256,
+      "learning_rate": 0.01,
+      "loss": 1.8941,
+      "step": 128085
+    },
+    {
+      "epoch": 13.15206900092412,
+      "grad_norm": 0.055176738649606705,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 128088
+    },
+    {
+      "epoch": 13.15237704076394,
+      "grad_norm": 0.050469979643821716,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 128091
+    },
+    {
+      "epoch": 13.152685080603758,
+      "grad_norm": 0.11509042233228683,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 128094
+    },
+    {
+      "epoch": 13.152993120443577,
+      "grad_norm": 0.08326201140880585,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 128097
+    },
+    {
+      "epoch": 13.153301160283396,
+      "grad_norm": 0.06993751972913742,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 128100
+    },
+    {
+      "epoch": 13.153609200123215,
+      "grad_norm": 0.07246918976306915,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 128103
+    },
+    {
+      "epoch": 13.153917239963036,
+      "grad_norm": 0.1300133764743805,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 128106
+    },
+    {
+      "epoch": 13.154225279802855,
+      "grad_norm": 0.07087678462266922,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 128109
+    },
+    {
+      "epoch": 13.154533319642674,
+      "grad_norm": 0.06616907566785812,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 128112
+    },
+    {
+      "epoch": 13.154841359482493,
+      "grad_norm": 0.0662837028503418,
+      "learning_rate": 0.01,
+      "loss": 1.9027,
+      "step": 128115
+    },
+    {
+      "epoch": 13.155149399322312,
+      "grad_norm": 0.09204954653978348,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 128118
+    },
+    {
+      "epoch": 13.15545743916213,
+      "grad_norm": 0.06275130063295364,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 128121
+    },
+    {
+      "epoch": 13.155765479001952,
+      "grad_norm": 0.07586976140737534,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 128124
+    },
+    {
+      "epoch": 13.15607351884177,
+      "grad_norm": 0.08205139636993408,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 128127
+    },
+    {
+      "epoch": 13.15638155868159,
+      "grad_norm": 0.04885202646255493,
+      "learning_rate": 0.01,
+      "loss": 1.9076,
+      "step": 128130
+    },
+    {
+      "epoch": 13.156689598521409,
+      "grad_norm": 0.1026860922574997,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 128133
+    },
+    {
+      "epoch": 13.156997638361227,
+      "grad_norm": 0.05542922019958496,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 128136
+    },
+    {
+      "epoch": 13.157305678201046,
+      "grad_norm": 0.046177614480257034,
+      "learning_rate": 0.01,
+      "loss": 1.9487,
+      "step": 128139
+    },
+    {
+      "epoch": 13.157613718040867,
+      "grad_norm": 0.09605981409549713,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 128142
+    },
+    {
+      "epoch": 13.157921757880686,
+      "grad_norm": 0.0465361513197422,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 128145
+    },
+    {
+      "epoch": 13.158229797720505,
+      "grad_norm": 0.09830117970705032,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 128148
+    },
+    {
+      "epoch": 13.158537837560324,
+      "grad_norm": 0.05673332139849663,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 128151
+    },
+    {
+      "epoch": 13.158845877400143,
+      "grad_norm": 0.10577160865068436,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 128154
+    },
+    {
+      "epoch": 13.159153917239964,
+      "grad_norm": 0.05297861248254776,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 128157
+    },
+    {
+      "epoch": 13.159461957079783,
+      "grad_norm": 0.054912131279706955,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 128160
+    },
+    {
+      "epoch": 13.159769996919602,
+      "grad_norm": 0.07440555095672607,
+      "learning_rate": 0.01,
+      "loss": 1.9002,
+      "step": 128163
+    },
+    {
+      "epoch": 13.16007803675942,
+      "grad_norm": 0.058643292635679245,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 128166
+    },
+    {
+      "epoch": 13.16038607659924,
+      "grad_norm": 0.040592942386865616,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 128169
+    },
+    {
+      "epoch": 13.160694116439059,
+      "grad_norm": 0.03338607773184776,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 128172
+    },
+    {
+      "epoch": 13.16100215627888,
+      "grad_norm": 0.04402768611907959,
+      "learning_rate": 0.01,
+      "loss": 1.9492,
+      "step": 128175
+    },
+    {
+      "epoch": 13.161310196118698,
+      "grad_norm": 0.05453972890973091,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 128178
+    },
+    {
+      "epoch": 13.161618235958517,
+      "grad_norm": 0.0486614927649498,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 128181
+    },
+    {
+      "epoch": 13.161926275798336,
+      "grad_norm": 0.14312784373760223,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 128184
+    },
+    {
+      "epoch": 13.162234315638155,
+      "grad_norm": 0.03800038620829582,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 128187
+    },
+    {
+      "epoch": 13.162542355477974,
+      "grad_norm": 0.03809847682714462,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 128190
+    },
+    {
+      "epoch": 13.162850395317795,
+      "grad_norm": 0.03336983174085617,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 128193
+    },
+    {
+      "epoch": 13.163158435157614,
+      "grad_norm": 0.10575585067272186,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 128196
+    },
+    {
+      "epoch": 13.163466474997433,
+      "grad_norm": 0.0735940933227539,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 128199
+    },
+    {
+      "epoch": 13.163774514837252,
+      "grad_norm": 0.06410026550292969,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 128202
+    },
+    {
+      "epoch": 13.164082554677071,
+      "grad_norm": 0.14058569073677063,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 128205
+    },
+    {
+      "epoch": 13.164390594516892,
+      "grad_norm": 0.12894608080387115,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 128208
+    },
+    {
+      "epoch": 13.16469863435671,
+      "grad_norm": 0.14890338480472565,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 128211
+    },
+    {
+      "epoch": 13.16500667419653,
+      "grad_norm": 0.11397437751293182,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 128214
+    },
+    {
+      "epoch": 13.165314714036349,
+      "grad_norm": 0.06320411711931229,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 128217
+    },
+    {
+      "epoch": 13.165622753876168,
+      "grad_norm": 0.08112873136997223,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 128220
+    },
+    {
+      "epoch": 13.165930793715987,
+      "grad_norm": 0.06387969106435776,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 128223
+    },
+    {
+      "epoch": 13.166238833555807,
+      "grad_norm": 0.05905725434422493,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 128226
+    },
+    {
+      "epoch": 13.166546873395626,
+      "grad_norm": 0.06875501573085785,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 128229
+    },
+    {
+      "epoch": 13.166854913235445,
+      "grad_norm": 0.05744965374469757,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 128232
+    },
+    {
+      "epoch": 13.167162953075264,
+      "grad_norm": 0.07017608731985092,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 128235
+    },
+    {
+      "epoch": 13.167470992915083,
+      "grad_norm": 0.06479623168706894,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 128238
+    },
+    {
+      "epoch": 13.167779032754902,
+      "grad_norm": 0.04840521886944771,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 128241
+    },
+    {
+      "epoch": 13.168087072594723,
+      "grad_norm": 0.09915796667337418,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 128244
+    },
+    {
+      "epoch": 13.168395112434542,
+      "grad_norm": 0.06504505127668381,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 128247
+    },
+    {
+      "epoch": 13.168703152274361,
+      "grad_norm": 0.05135345086455345,
+      "learning_rate": 0.01,
+      "loss": 1.89,
+      "step": 128250
+    },
+    {
+      "epoch": 13.16901119211418,
+      "grad_norm": 0.06859485059976578,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 128253
+    },
+    {
+      "epoch": 13.169319231953999,
+      "grad_norm": 0.032977066934108734,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 128256
+    },
+    {
+      "epoch": 13.169627271793818,
+      "grad_norm": 0.05988279730081558,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 128259
+    },
+    {
+      "epoch": 13.169935311633639,
+      "grad_norm": 0.06296802312135696,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 128262
+    },
+    {
+      "epoch": 13.170243351473458,
+      "grad_norm": 0.04666455462574959,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 128265
+    },
+    {
+      "epoch": 13.170551391313277,
+      "grad_norm": 0.049062807112932205,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 128268
+    },
+    {
+      "epoch": 13.170859431153096,
+      "grad_norm": 0.10835019499063492,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 128271
+    },
+    {
+      "epoch": 13.171167470992915,
+      "grad_norm": 0.08930020779371262,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 128274
+    },
+    {
+      "epoch": 13.171475510832735,
+      "grad_norm": 0.16498881578445435,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 128277
+    },
+    {
+      "epoch": 13.171783550672554,
+      "grad_norm": 0.11193927377462387,
+      "learning_rate": 0.01,
+      "loss": 1.8956,
+      "step": 128280
+    },
+    {
+      "epoch": 13.172091590512373,
+      "grad_norm": 0.1668013334274292,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 128283
+    },
+    {
+      "epoch": 13.172399630352192,
+      "grad_norm": 0.11954472959041595,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 128286
+    },
+    {
+      "epoch": 13.172707670192011,
+      "grad_norm": 0.08617737889289856,
+      "learning_rate": 0.01,
+      "loss": 1.953,
+      "step": 128289
+    },
+    {
+      "epoch": 13.17301571003183,
+      "grad_norm": 0.07345489412546158,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 128292
+    },
+    {
+      "epoch": 13.173323749871651,
+      "grad_norm": 0.05653364956378937,
+      "learning_rate": 0.01,
+      "loss": 1.8979,
+      "step": 128295
+    },
+    {
+      "epoch": 13.17363178971147,
+      "grad_norm": 0.07379655539989471,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 128298
+    },
+    {
+      "epoch": 13.173939829551289,
+      "grad_norm": 0.07429443299770355,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 128301
+    },
+    {
+      "epoch": 13.174247869391108,
+      "grad_norm": 0.0659106969833374,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 128304
+    },
+    {
+      "epoch": 13.174555909230927,
+      "grad_norm": 0.04974491149187088,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 128307
+    },
+    {
+      "epoch": 13.174863949070746,
+      "grad_norm": 0.04204713925719261,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 128310
+    },
+    {
+      "epoch": 13.175171988910567,
+      "grad_norm": 0.053950510919094086,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 128313
+    },
+    {
+      "epoch": 13.175480028750385,
+      "grad_norm": 0.10668610781431198,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 128316
+    },
+    {
+      "epoch": 13.175788068590204,
+      "grad_norm": 0.10003342479467392,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 128319
+    },
+    {
+      "epoch": 13.176096108430023,
+      "grad_norm": 0.044990260154008865,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 128322
+    },
+    {
+      "epoch": 13.176404148269842,
+      "grad_norm": 0.10324764996767044,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 128325
+    },
+    {
+      "epoch": 13.176712188109661,
+      "grad_norm": 0.04529158025979996,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 128328
+    },
+    {
+      "epoch": 13.177020227949482,
+      "grad_norm": 0.059428367763757706,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 128331
+    },
+    {
+      "epoch": 13.177328267789301,
+      "grad_norm": 0.07299266010522842,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 128334
+    },
+    {
+      "epoch": 13.17763630762912,
+      "grad_norm": 0.10079319775104523,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 128337
+    },
+    {
+      "epoch": 13.177944347468939,
+      "grad_norm": 0.08960142731666565,
+      "learning_rate": 0.01,
+      "loss": 1.9062,
+      "step": 128340
+    },
+    {
+      "epoch": 13.178252387308758,
+      "grad_norm": 0.07033776491880417,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 128343
+    },
+    {
+      "epoch": 13.178560427148579,
+      "grad_norm": 0.03877047449350357,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 128346
+    },
+    {
+      "epoch": 13.178868466988398,
+      "grad_norm": 0.04558166116476059,
+      "learning_rate": 0.01,
+      "loss": 1.8939,
+      "step": 128349
+    },
+    {
+      "epoch": 13.179176506828217,
+      "grad_norm": 0.13322840631008148,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 128352
+    },
+    {
+      "epoch": 13.179484546668036,
+      "grad_norm": 0.0635785311460495,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 128355
+    },
+    {
+      "epoch": 13.179792586507855,
+      "grad_norm": 0.03747203201055527,
+      "learning_rate": 0.01,
+      "loss": 1.9055,
+      "step": 128358
+    },
+    {
+      "epoch": 13.180100626347674,
+      "grad_norm": 0.04630962386727333,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 128361
+    },
+    {
+      "epoch": 13.180408666187494,
+      "grad_norm": 0.08890605717897415,
+      "learning_rate": 0.01,
+      "loss": 1.889,
+      "step": 128364
+    },
+    {
+      "epoch": 13.180716706027313,
+      "grad_norm": 0.06858961284160614,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 128367
+    },
+    {
+      "epoch": 13.181024745867132,
+      "grad_norm": 0.04306871071457863,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 128370
+    },
+    {
+      "epoch": 13.181332785706951,
+      "grad_norm": 0.03864138573408127,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 128373
+    },
+    {
+      "epoch": 13.18164082554677,
+      "grad_norm": 0.049070317298173904,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 128376
+    },
+    {
+      "epoch": 13.18194886538659,
+      "grad_norm": 0.0552310086786747,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 128379
+    },
+    {
+      "epoch": 13.18225690522641,
+      "grad_norm": 0.05121565982699394,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 128382
+    },
+    {
+      "epoch": 13.182564945066229,
+      "grad_norm": 0.09130030870437622,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 128385
+    },
+    {
+      "epoch": 13.182872984906048,
+      "grad_norm": 0.09497486054897308,
+      "learning_rate": 0.01,
+      "loss": 1.8931,
+      "step": 128388
+    },
+    {
+      "epoch": 13.183181024745867,
+      "grad_norm": 0.07303666323423386,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 128391
+    },
+    {
+      "epoch": 13.183489064585686,
+      "grad_norm": 0.051671214401721954,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 128394
+    },
+    {
+      "epoch": 13.183797104425505,
+      "grad_norm": 0.0728432834148407,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 128397
+    },
+    {
+      "epoch": 13.184105144265326,
+      "grad_norm": 0.13195624947547913,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 128400
+    },
+    {
+      "epoch": 13.184413184105145,
+      "grad_norm": 0.0797656923532486,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 128403
+    },
+    {
+      "epoch": 13.184721223944964,
+      "grad_norm": 0.07732376456260681,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 128406
+    },
+    {
+      "epoch": 13.185029263784783,
+      "grad_norm": 0.07109509408473969,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 128409
+    },
+    {
+      "epoch": 13.185337303624602,
+      "grad_norm": 0.058891117572784424,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 128412
+    },
+    {
+      "epoch": 13.185645343464422,
+      "grad_norm": 0.0770486369729042,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 128415
+    },
+    {
+      "epoch": 13.185953383304241,
+      "grad_norm": 0.11541492491960526,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 128418
+    },
+    {
+      "epoch": 13.18626142314406,
+      "grad_norm": 0.047796886414289474,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 128421
+    },
+    {
+      "epoch": 13.18656946298388,
+      "grad_norm": 0.05169094726443291,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 128424
+    },
+    {
+      "epoch": 13.186877502823698,
+      "grad_norm": 0.05594772845506668,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 128427
+    },
+    {
+      "epoch": 13.187185542663517,
+      "grad_norm": 0.05230353772640228,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 128430
+    },
+    {
+      "epoch": 13.187493582503338,
+      "grad_norm": 0.09033312648534775,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 128433
+    },
+    {
+      "epoch": 13.187801622343157,
+      "grad_norm": 0.0845467820763588,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 128436
+    },
+    {
+      "epoch": 13.188109662182976,
+      "grad_norm": 0.08203680068254471,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 128439
+    },
+    {
+      "epoch": 13.188417702022795,
+      "grad_norm": 0.03659658133983612,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 128442
+    },
+    {
+      "epoch": 13.188725741862614,
+      "grad_norm": 0.12295246124267578,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 128445
+    },
+    {
+      "epoch": 13.189033781702433,
+      "grad_norm": 0.0689648985862732,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 128448
+    },
+    {
+      "epoch": 13.189341821542254,
+      "grad_norm": 0.05302094295620918,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 128451
+    },
+    {
+      "epoch": 13.189649861382073,
+      "grad_norm": 0.034731291234493256,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 128454
+    },
+    {
+      "epoch": 13.189957901221891,
+      "grad_norm": 0.04777044057846069,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 128457
+    },
+    {
+      "epoch": 13.19026594106171,
+      "grad_norm": 0.06617261469364166,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 128460
+    },
+    {
+      "epoch": 13.19057398090153,
+      "grad_norm": 0.052258189767599106,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 128463
+    },
+    {
+      "epoch": 13.190882020741348,
+      "grad_norm": 0.04021059349179268,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 128466
+    },
+    {
+      "epoch": 13.19119006058117,
+      "grad_norm": 0.03816143050789833,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 128469
+    },
+    {
+      "epoch": 13.191498100420988,
+      "grad_norm": 0.04497189819812775,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 128472
+    },
+    {
+      "epoch": 13.191806140260807,
+      "grad_norm": 0.1416708528995514,
+      "learning_rate": 0.01,
+      "loss": 1.9002,
+      "step": 128475
+    },
+    {
+      "epoch": 13.192114180100626,
+      "grad_norm": 0.12095382809638977,
+      "learning_rate": 0.01,
+      "loss": 1.8851,
+      "step": 128478
+    },
+    {
+      "epoch": 13.192422219940445,
+      "grad_norm": 0.09808985143899918,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 128481
+    },
+    {
+      "epoch": 13.192730259780266,
+      "grad_norm": 0.05455213785171509,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 128484
+    },
+    {
+      "epoch": 13.193038299620085,
+      "grad_norm": 0.059937696903944016,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 128487
+    },
+    {
+      "epoch": 13.193346339459904,
+      "grad_norm": 0.03418611362576485,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 128490
+    },
+    {
+      "epoch": 13.193654379299723,
+      "grad_norm": 0.06829250603914261,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 128493
+    },
+    {
+      "epoch": 13.193962419139542,
+      "grad_norm": 0.049868740141391754,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 128496
+    },
+    {
+      "epoch": 13.19427045897936,
+      "grad_norm": 0.07731674611568451,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 128499
+    },
+    {
+      "epoch": 13.194578498819181,
+      "grad_norm": 0.06345655769109726,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 128502
+    },
+    {
+      "epoch": 13.194886538659,
+      "grad_norm": 0.03856688365340233,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 128505
+    },
+    {
+      "epoch": 13.19519457849882,
+      "grad_norm": 0.14421236515045166,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 128508
+    },
+    {
+      "epoch": 13.195502618338638,
+      "grad_norm": 0.06367630511522293,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 128511
+    },
+    {
+      "epoch": 13.195810658178457,
+      "grad_norm": 0.046045877039432526,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 128514
+    },
+    {
+      "epoch": 13.196118698018276,
+      "grad_norm": 0.06319709867238998,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 128517
+    },
+    {
+      "epoch": 13.196426737858097,
+      "grad_norm": 0.03876698762178421,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 128520
+    },
+    {
+      "epoch": 13.196734777697916,
+      "grad_norm": 0.033340927213430405,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 128523
+    },
+    {
+      "epoch": 13.197042817537735,
+      "grad_norm": 0.04503343626856804,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 128526
+    },
+    {
+      "epoch": 13.197350857377554,
+      "grad_norm": 0.040798723697662354,
+      "learning_rate": 0.01,
+      "loss": 1.8966,
+      "step": 128529
+    },
+    {
+      "epoch": 13.197658897217373,
+      "grad_norm": 0.047963954508304596,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 128532
+    },
+    {
+      "epoch": 13.197966937057192,
+      "grad_norm": 0.09067251533269882,
+      "learning_rate": 0.01,
+      "loss": 1.8918,
+      "step": 128535
+    },
+    {
+      "epoch": 13.198274976897013,
+      "grad_norm": 0.05490126088261604,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 128538
+    },
+    {
+      "epoch": 13.198583016736832,
+      "grad_norm": 0.04974130913615227,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 128541
+    },
+    {
+      "epoch": 13.19889105657665,
+      "grad_norm": 0.0617697574198246,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 128544
+    },
+    {
+      "epoch": 13.19919909641647,
+      "grad_norm": 0.03912938013672829,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 128547
+    },
+    {
+      "epoch": 13.199507136256289,
+      "grad_norm": 0.06654561311006546,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 128550
+    },
+    {
+      "epoch": 13.19981517609611,
+      "grad_norm": 0.10337626934051514,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 128553
+    },
+    {
+      "epoch": 13.200123215935928,
+      "grad_norm": 0.111897312104702,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 128556
+    },
+    {
+      "epoch": 13.200431255775747,
+      "grad_norm": 0.04952174425125122,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 128559
+    },
+    {
+      "epoch": 13.200739295615566,
+      "grad_norm": 0.04611589014530182,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 128562
+    },
+    {
+      "epoch": 13.201047335455385,
+      "grad_norm": 0.031897012144327164,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 128565
+    },
+    {
+      "epoch": 13.201355375295204,
+      "grad_norm": 0.06147684529423714,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 128568
+    },
+    {
+      "epoch": 13.201663415135025,
+      "grad_norm": 0.1251518726348877,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 128571
+    },
+    {
+      "epoch": 13.201971454974844,
+      "grad_norm": 0.05742456018924713,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 128574
+    },
+    {
+      "epoch": 13.202279494814663,
+      "grad_norm": 0.03674685209989548,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 128577
+    },
+    {
+      "epoch": 13.202587534654482,
+      "grad_norm": 0.08455429971218109,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 128580
+    },
+    {
+      "epoch": 13.2028955744943,
+      "grad_norm": 0.08967318385839462,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 128583
+    },
+    {
+      "epoch": 13.20320361433412,
+      "grad_norm": 0.0961543619632721,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 128586
+    },
+    {
+      "epoch": 13.20351165417394,
+      "grad_norm": 0.09905248880386353,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 128589
+    },
+    {
+      "epoch": 13.20381969401376,
+      "grad_norm": 0.03996612876653671,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 128592
+    },
+    {
+      "epoch": 13.204127733853579,
+      "grad_norm": 0.051846008747816086,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 128595
+    },
+    {
+      "epoch": 13.204435773693397,
+      "grad_norm": 0.0771871954202652,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 128598
+    },
+    {
+      "epoch": 13.204743813533216,
+      "grad_norm": 0.06721936166286469,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 128601
+    },
+    {
+      "epoch": 13.205051853373035,
+      "grad_norm": 0.05308469384908676,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 128604
+    },
+    {
+      "epoch": 13.205359893212856,
+      "grad_norm": 0.12253759801387787,
+      "learning_rate": 0.01,
+      "loss": 1.943,
+      "step": 128607
+    },
+    {
+      "epoch": 13.205667933052675,
+      "grad_norm": 0.048144977539777756,
+      "learning_rate": 0.01,
+      "loss": 1.9425,
+      "step": 128610
+    },
+    {
+      "epoch": 13.205975972892494,
+      "grad_norm": 0.0522279255092144,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 128613
+    },
+    {
+      "epoch": 13.206284012732313,
+      "grad_norm": 0.03817118704319,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 128616
+    },
+    {
+      "epoch": 13.206592052572132,
+      "grad_norm": 0.03967083990573883,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 128619
+    },
+    {
+      "epoch": 13.206900092411953,
+      "grad_norm": 0.0334506630897522,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 128622
+    },
+    {
+      "epoch": 13.207208132251772,
+      "grad_norm": 0.09487753361463547,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 128625
+    },
+    {
+      "epoch": 13.20751617209159,
+      "grad_norm": 0.09752718359231949,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 128628
+    },
+    {
+      "epoch": 13.20782421193141,
+      "grad_norm": 0.12180406600236893,
+      "learning_rate": 0.01,
+      "loss": 1.9009,
+      "step": 128631
+    },
+    {
+      "epoch": 13.208132251771229,
+      "grad_norm": 0.052494414150714874,
+      "learning_rate": 0.01,
+      "loss": 1.8948,
+      "step": 128634
+    },
+    {
+      "epoch": 13.208440291611048,
+      "grad_norm": 0.042598407715559006,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 128637
+    },
+    {
+      "epoch": 13.208748331450868,
+      "grad_norm": 0.05890635773539543,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 128640
+    },
+    {
+      "epoch": 13.209056371290687,
+      "grad_norm": 0.03507623076438904,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 128643
+    },
+    {
+      "epoch": 13.209364411130506,
+      "grad_norm": 0.03071865253150463,
+      "learning_rate": 0.01,
+      "loss": 1.8972,
+      "step": 128646
+    },
+    {
+      "epoch": 13.209672450970325,
+      "grad_norm": 0.03861444815993309,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 128649
+    },
+    {
+      "epoch": 13.209980490810144,
+      "grad_norm": 0.033980824053287506,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 128652
+    },
+    {
+      "epoch": 13.210288530649963,
+      "grad_norm": 0.039954014122486115,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 128655
+    },
+    {
+      "epoch": 13.210596570489784,
+      "grad_norm": 0.03353109210729599,
+      "learning_rate": 0.01,
+      "loss": 1.8782,
+      "step": 128658
+    },
+    {
+      "epoch": 13.210904610329603,
+      "grad_norm": 0.09032177925109863,
+      "learning_rate": 0.01,
+      "loss": 1.8992,
+      "step": 128661
+    },
+    {
+      "epoch": 13.211212650169422,
+      "grad_norm": 0.05219477787613869,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 128664
+    },
+    {
+      "epoch": 13.211520690009241,
+      "grad_norm": 0.11966364085674286,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 128667
+    },
+    {
+      "epoch": 13.21182872984906,
+      "grad_norm": 0.06658205389976501,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 128670
+    },
+    {
+      "epoch": 13.212136769688879,
+      "grad_norm": 0.09207108616828918,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 128673
+    },
+    {
+      "epoch": 13.2124448095287,
+      "grad_norm": 0.06460580974817276,
+      "learning_rate": 0.01,
+      "loss": 1.9031,
+      "step": 128676
+    },
+    {
+      "epoch": 13.212752849368519,
+      "grad_norm": 0.056336041539907455,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 128679
+    },
+    {
+      "epoch": 13.213060889208338,
+      "grad_norm": 0.10067939758300781,
+      "learning_rate": 0.01,
+      "loss": 1.9575,
+      "step": 128682
+    },
+    {
+      "epoch": 13.213368929048157,
+      "grad_norm": 0.036347538232803345,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 128685
+    },
+    {
+      "epoch": 13.213676968887976,
+      "grad_norm": 0.10293001681566238,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 128688
+    },
+    {
+      "epoch": 13.213985008727796,
+      "grad_norm": 0.08132581412792206,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 128691
+    },
+    {
+      "epoch": 13.214293048567615,
+      "grad_norm": 0.038934145122766495,
+      "learning_rate": 0.01,
+      "loss": 1.9445,
+      "step": 128694
+    },
+    {
+      "epoch": 13.214601088407434,
+      "grad_norm": 0.048438649624586105,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 128697
+    },
+    {
+      "epoch": 13.214909128247253,
+      "grad_norm": 0.05083656311035156,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 128700
+    },
+    {
+      "epoch": 13.215217168087072,
+      "grad_norm": 0.03730197995901108,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 128703
+    },
+    {
+      "epoch": 13.215525207926891,
+      "grad_norm": 0.0916329026222229,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 128706
+    },
+    {
+      "epoch": 13.215833247766712,
+      "grad_norm": 0.09122508019208908,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 128709
+    },
+    {
+      "epoch": 13.216141287606531,
+      "grad_norm": 0.09795338660478592,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 128712
+    },
+    {
+      "epoch": 13.21644932744635,
+      "grad_norm": 0.05175530165433884,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 128715
+    },
+    {
+      "epoch": 13.216757367286169,
+      "grad_norm": 0.06675558537244797,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 128718
+    },
+    {
+      "epoch": 13.217065407125988,
+      "grad_norm": 0.06606762111186981,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 128721
+    },
+    {
+      "epoch": 13.217373446965807,
+      "grad_norm": 0.08565639704465866,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 128724
+    },
+    {
+      "epoch": 13.217681486805628,
+      "grad_norm": 0.04588396102190018,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 128727
+    },
+    {
+      "epoch": 13.217989526645447,
+      "grad_norm": 0.08661910146474838,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 128730
+    },
+    {
+      "epoch": 13.218297566485266,
+      "grad_norm": 0.0529467836022377,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 128733
+    },
+    {
+      "epoch": 13.218605606325085,
+      "grad_norm": 0.09541238844394684,
+      "learning_rate": 0.01,
+      "loss": 1.8954,
+      "step": 128736
+    },
+    {
+      "epoch": 13.218913646164903,
+      "grad_norm": 0.06195485219359398,
+      "learning_rate": 0.01,
+      "loss": 1.9578,
+      "step": 128739
+    },
+    {
+      "epoch": 13.219221686004722,
+      "grad_norm": 0.09415952116250992,
+      "learning_rate": 0.01,
+      "loss": 1.9485,
+      "step": 128742
+    },
+    {
+      "epoch": 13.219529725844543,
+      "grad_norm": 0.10664113610982895,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 128745
+    },
+    {
+      "epoch": 13.219837765684362,
+      "grad_norm": 0.0343109630048275,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 128748
+    },
+    {
+      "epoch": 13.220145805524181,
+      "grad_norm": 0.04439496994018555,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 128751
+    },
+    {
+      "epoch": 13.220453845364,
+      "grad_norm": 0.1799716055393219,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 128754
+    },
+    {
+      "epoch": 13.22076188520382,
+      "grad_norm": 0.13082244992256165,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 128757
+    },
+    {
+      "epoch": 13.22106992504364,
+      "grad_norm": 0.07456562668085098,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 128760
+    },
+    {
+      "epoch": 13.221377964883459,
+      "grad_norm": 0.061206087470054626,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 128763
+    },
+    {
+      "epoch": 13.221686004723278,
+      "grad_norm": 0.0474642738699913,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 128766
+    },
+    {
+      "epoch": 13.221994044563097,
+      "grad_norm": 0.036594223231077194,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 128769
+    },
+    {
+      "epoch": 13.222302084402916,
+      "grad_norm": 0.03738519549369812,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 128772
+    },
+    {
+      "epoch": 13.222610124242735,
+      "grad_norm": 0.037344127893447876,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 128775
+    },
+    {
+      "epoch": 13.222918164082555,
+      "grad_norm": 0.04048924520611763,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 128778
+    },
+    {
+      "epoch": 13.223226203922374,
+      "grad_norm": 0.09992200881242752,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 128781
+    },
+    {
+      "epoch": 13.223534243762193,
+      "grad_norm": 0.04822799190878868,
+      "learning_rate": 0.01,
+      "loss": 1.8893,
+      "step": 128784
+    },
+    {
+      "epoch": 13.223842283602012,
+      "grad_norm": 0.03634457662701607,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 128787
+    },
+    {
+      "epoch": 13.224150323441831,
+      "grad_norm": 0.06005654111504555,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 128790
+    },
+    {
+      "epoch": 13.22445836328165,
+      "grad_norm": 0.06848358362913132,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 128793
+    },
+    {
+      "epoch": 13.224766403121471,
+      "grad_norm": 0.04030539467930794,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 128796
+    },
+    {
+      "epoch": 13.22507444296129,
+      "grad_norm": 0.05397363379597664,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 128799
+    },
+    {
+      "epoch": 13.225382482801109,
+      "grad_norm": 0.08214554935693741,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 128802
+    },
+    {
+      "epoch": 13.225690522640928,
+      "grad_norm": 0.052762459963560104,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 128805
+    },
+    {
+      "epoch": 13.225998562480747,
+      "grad_norm": 0.09979503601789474,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 128808
+    },
+    {
+      "epoch": 13.226306602320566,
+      "grad_norm": 0.0754537358880043,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 128811
+    },
+    {
+      "epoch": 13.226614642160387,
+      "grad_norm": 0.07728686183691025,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 128814
+    },
+    {
+      "epoch": 13.226922682000206,
+      "grad_norm": 0.09226962178945541,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 128817
+    },
+    {
+      "epoch": 13.227230721840025,
+      "grad_norm": 0.07268766313791275,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 128820
+    },
+    {
+      "epoch": 13.227538761679844,
+      "grad_norm": 0.06354807317256927,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 128823
+    },
+    {
+      "epoch": 13.227846801519663,
+      "grad_norm": 0.06389658153057098,
+      "learning_rate": 0.01,
+      "loss": 1.9035,
+      "step": 128826
+    },
+    {
+      "epoch": 13.228154841359483,
+      "grad_norm": 0.0607454851269722,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 128829
+    },
+    {
+      "epoch": 13.228462881199302,
+      "grad_norm": 0.08813507854938507,
+      "learning_rate": 0.01,
+      "loss": 1.8779,
+      "step": 128832
+    },
+    {
+      "epoch": 13.228770921039121,
+      "grad_norm": 0.06515578180551529,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 128835
+    },
+    {
+      "epoch": 13.22907896087894,
+      "grad_norm": 0.09099886566400528,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 128838
+    },
+    {
+      "epoch": 13.22938700071876,
+      "grad_norm": 0.10348626971244812,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 128841
+    },
+    {
+      "epoch": 13.229695040558578,
+      "grad_norm": 0.16220329701900482,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 128844
+    },
+    {
+      "epoch": 13.230003080398399,
+      "grad_norm": 0.05998501181602478,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 128847
+    },
+    {
+      "epoch": 13.230311120238218,
+      "grad_norm": 0.054723724722862244,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 128850
+    },
+    {
+      "epoch": 13.230619160078037,
+      "grad_norm": 0.05473949387669563,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 128853
+    },
+    {
+      "epoch": 13.230927199917856,
+      "grad_norm": 0.04675135761499405,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 128856
+    },
+    {
+      "epoch": 13.231235239757675,
+      "grad_norm": 0.06425482779741287,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 128859
+    },
+    {
+      "epoch": 13.231543279597494,
+      "grad_norm": 0.03545411676168442,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 128862
+    },
+    {
+      "epoch": 13.231851319437315,
+      "grad_norm": 0.06792523711919785,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 128865
+    },
+    {
+      "epoch": 13.232159359277134,
+      "grad_norm": 0.06590435653924942,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 128868
+    },
+    {
+      "epoch": 13.232467399116953,
+      "grad_norm": 0.09454898536205292,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 128871
+    },
+    {
+      "epoch": 13.232775438956772,
+      "grad_norm": 0.0366886630654335,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 128874
+    },
+    {
+      "epoch": 13.23308347879659,
+      "grad_norm": 0.036178894340991974,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 128877
+    },
+    {
+      "epoch": 13.23339151863641,
+      "grad_norm": 0.04039180651307106,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 128880
+    },
+    {
+      "epoch": 13.23369955847623,
+      "grad_norm": 0.049927182495594025,
+      "learning_rate": 0.01,
+      "loss": 1.9055,
+      "step": 128883
+    },
+    {
+      "epoch": 13.23400759831605,
+      "grad_norm": 0.09239368140697479,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 128886
+    },
+    {
+      "epoch": 13.234315638155868,
+      "grad_norm": 0.09888288378715515,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 128889
+    },
+    {
+      "epoch": 13.234623677995687,
+      "grad_norm": 0.050970543175935745,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 128892
+    },
+    {
+      "epoch": 13.234931717835506,
+      "grad_norm": 0.06843187659978867,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 128895
+    },
+    {
+      "epoch": 13.235239757675327,
+      "grad_norm": 0.08436018228530884,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 128898
+    },
+    {
+      "epoch": 13.235547797515146,
+      "grad_norm": 0.08341411501169205,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 128901
+    },
+    {
+      "epoch": 13.235855837354965,
+      "grad_norm": 0.051047082990407944,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 128904
+    },
+    {
+      "epoch": 13.236163877194784,
+      "grad_norm": 0.04779675602912903,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 128907
+    },
+    {
+      "epoch": 13.236471917034603,
+      "grad_norm": 0.04439254105091095,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 128910
+    },
+    {
+      "epoch": 13.236779956874422,
+      "grad_norm": 0.14152736961841583,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 128913
+    },
+    {
+      "epoch": 13.237087996714243,
+      "grad_norm": 0.039216648787260056,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 128916
+    },
+    {
+      "epoch": 13.237396036554061,
+      "grad_norm": 0.0804230198264122,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 128919
+    },
+    {
+      "epoch": 13.23770407639388,
+      "grad_norm": 0.03625069186091423,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 128922
+    },
+    {
+      "epoch": 13.2380121162337,
+      "grad_norm": 0.04622931033372879,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 128925
+    },
+    {
+      "epoch": 13.238320156073518,
+      "grad_norm": 0.03691410645842552,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 128928
+    },
+    {
+      "epoch": 13.238628195913337,
+      "grad_norm": 0.05070066079497337,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 128931
+    },
+    {
+      "epoch": 13.238936235753158,
+      "grad_norm": 0.06077568233013153,
+      "learning_rate": 0.01,
+      "loss": 1.9511,
+      "step": 128934
+    },
+    {
+      "epoch": 13.239244275592977,
+      "grad_norm": 0.054304707795381546,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 128937
+    },
+    {
+      "epoch": 13.239552315432796,
+      "grad_norm": 0.032350555062294006,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 128940
+    },
+    {
+      "epoch": 13.239860355272615,
+      "grad_norm": 0.09139896184206009,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 128943
+    },
+    {
+      "epoch": 13.240168395112434,
+      "grad_norm": 0.09659933298826218,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 128946
+    },
+    {
+      "epoch": 13.240476434952253,
+      "grad_norm": 0.04620542749762535,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 128949
+    },
+    {
+      "epoch": 13.240784474792074,
+      "grad_norm": 0.09844648838043213,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 128952
+    },
+    {
+      "epoch": 13.241092514631893,
+      "grad_norm": 0.05950770527124405,
+      "learning_rate": 0.01,
+      "loss": 1.8839,
+      "step": 128955
+    },
+    {
+      "epoch": 13.241400554471712,
+      "grad_norm": 0.04819858446717262,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 128958
+    },
+    {
+      "epoch": 13.24170859431153,
+      "grad_norm": 0.03495674580335617,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 128961
+    },
+    {
+      "epoch": 13.24201663415135,
+      "grad_norm": 0.04327937588095665,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 128964
+    },
+    {
+      "epoch": 13.24232467399117,
+      "grad_norm": 0.07633385062217712,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 128967
+    },
+    {
+      "epoch": 13.24263271383099,
+      "grad_norm": 0.049793582409620285,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 128970
+    },
+    {
+      "epoch": 13.242940753670808,
+      "grad_norm": 0.10919264703989029,
+      "learning_rate": 0.01,
+      "loss": 1.8999,
+      "step": 128973
+    },
+    {
+      "epoch": 13.243248793510627,
+      "grad_norm": 0.07186369597911835,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 128976
+    },
+    {
+      "epoch": 13.243556833350446,
+      "grad_norm": 0.0748286023736,
+      "learning_rate": 0.01,
+      "loss": 1.8939,
+      "step": 128979
+    },
+    {
+      "epoch": 13.243864873190265,
+      "grad_norm": 0.06216063350439072,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 128982
+    },
+    {
+      "epoch": 13.244172913030086,
+      "grad_norm": 0.10752072185277939,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 128985
+    },
+    {
+      "epoch": 13.244480952869905,
+      "grad_norm": 0.055293768644332886,
+      "learning_rate": 0.01,
+      "loss": 1.9021,
+      "step": 128988
+    },
+    {
+      "epoch": 13.244788992709724,
+      "grad_norm": 0.05748097971081734,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 128991
+    },
+    {
+      "epoch": 13.245097032549543,
+      "grad_norm": 0.17047500610351562,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 128994
+    },
+    {
+      "epoch": 13.245405072389362,
+      "grad_norm": 0.10807085037231445,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 128997
+    },
+    {
+      "epoch": 13.245713112229181,
+      "grad_norm": 0.055860165506601334,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 129000
+    },
+    {
+      "epoch": 13.246021152069002,
+      "grad_norm": 0.053709425032138824,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 129003
+    },
+    {
+      "epoch": 13.24632919190882,
+      "grad_norm": 0.06729075312614441,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 129006
+    },
+    {
+      "epoch": 13.24663723174864,
+      "grad_norm": 0.0539906769990921,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 129009
+    },
+    {
+      "epoch": 13.246945271588459,
+      "grad_norm": 0.04343676194548607,
+      "learning_rate": 0.01,
+      "loss": 1.9031,
+      "step": 129012
+    },
+    {
+      "epoch": 13.247253311428278,
+      "grad_norm": 0.1410582959651947,
+      "learning_rate": 0.01,
+      "loss": 1.8672,
+      "step": 129015
+    },
+    {
+      "epoch": 13.247561351268097,
+      "grad_norm": 0.04274624586105347,
+      "learning_rate": 0.01,
+      "loss": 1.899,
+      "step": 129018
+    },
+    {
+      "epoch": 13.247869391107917,
+      "grad_norm": 0.05143105238676071,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 129021
+    },
+    {
+      "epoch": 13.248177430947736,
+      "grad_norm": 0.07548625022172928,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 129024
+    },
+    {
+      "epoch": 13.248485470787555,
+      "grad_norm": 0.05533851310610771,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 129027
+    },
+    {
+      "epoch": 13.248793510627374,
+      "grad_norm": 0.05457909405231476,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 129030
+    },
+    {
+      "epoch": 13.249101550467193,
+      "grad_norm": 0.06756798923015594,
+      "learning_rate": 0.01,
+      "loss": 1.9027,
+      "step": 129033
+    },
+    {
+      "epoch": 13.249409590307014,
+      "grad_norm": 0.04697268828749657,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 129036
+    },
+    {
+      "epoch": 13.249717630146833,
+      "grad_norm": 0.04560460150241852,
+      "learning_rate": 0.01,
+      "loss": 1.8857,
+      "step": 129039
+    },
+    {
+      "epoch": 13.250025669986652,
+      "grad_norm": 0.04431512951850891,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 129042
+    },
+    {
+      "epoch": 13.25033370982647,
+      "grad_norm": 0.05957891792058945,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 129045
+    },
+    {
+      "epoch": 13.25064174966629,
+      "grad_norm": 0.1711960732936859,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 129048
+    },
+    {
+      "epoch": 13.250949789506109,
+      "grad_norm": 0.06780771166086197,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 129051
+    },
+    {
+      "epoch": 13.25125782934593,
+      "grad_norm": 0.034737516194581985,
+      "learning_rate": 0.01,
+      "loss": 1.8966,
+      "step": 129054
+    },
+    {
+      "epoch": 13.251565869185749,
+      "grad_norm": 0.04268951714038849,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 129057
+    },
+    {
+      "epoch": 13.251873909025567,
+      "grad_norm": 0.05026356503367424,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 129060
+    },
+    {
+      "epoch": 13.252181948865386,
+      "grad_norm": 0.043847691267728806,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 129063
+    },
+    {
+      "epoch": 13.252489988705205,
+      "grad_norm": 0.07645340263843536,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 129066
+    },
+    {
+      "epoch": 13.252798028545024,
+      "grad_norm": 0.040714241564273834,
+      "learning_rate": 0.01,
+      "loss": 1.8887,
+      "step": 129069
+    },
+    {
+      "epoch": 13.253106068384845,
+      "grad_norm": 0.04678146541118622,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 129072
+    },
+    {
+      "epoch": 13.253414108224664,
+      "grad_norm": 0.09962925314903259,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 129075
+    },
+    {
+      "epoch": 13.253722148064483,
+      "grad_norm": 0.038441773504018784,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 129078
+    },
+    {
+      "epoch": 13.254030187904302,
+      "grad_norm": 0.043344348669052124,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 129081
+    },
+    {
+      "epoch": 13.254338227744121,
+      "grad_norm": 0.08397742360830307,
+      "learning_rate": 0.01,
+      "loss": 1.9472,
+      "step": 129084
+    },
+    {
+      "epoch": 13.25464626758394,
+      "grad_norm": 0.13360343873500824,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 129087
+    },
+    {
+      "epoch": 13.25495430742376,
+      "grad_norm": 0.14385685324668884,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 129090
+    },
+    {
+      "epoch": 13.25526234726358,
+      "grad_norm": 0.104715995490551,
+      "learning_rate": 0.01,
+      "loss": 1.8901,
+      "step": 129093
+    },
+    {
+      "epoch": 13.255570387103399,
+      "grad_norm": 0.06059815734624863,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 129096
+    },
+    {
+      "epoch": 13.255878426943218,
+      "grad_norm": 0.043850384652614594,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 129099
+    },
+    {
+      "epoch": 13.256186466783037,
+      "grad_norm": 0.05376746878027916,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 129102
+    },
+    {
+      "epoch": 13.256494506622857,
+      "grad_norm": 0.03732164949178696,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 129105
+    },
+    {
+      "epoch": 13.256802546462676,
+      "grad_norm": 0.07699265331029892,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 129108
+    },
+    {
+      "epoch": 13.257110586302495,
+      "grad_norm": 0.09846188127994537,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 129111
+    },
+    {
+      "epoch": 13.257418626142314,
+      "grad_norm": 0.0970764309167862,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 129114
+    },
+    {
+      "epoch": 13.257726665982133,
+      "grad_norm": 0.04339958354830742,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 129117
+    },
+    {
+      "epoch": 13.258034705821952,
+      "grad_norm": 0.031557172536849976,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 129120
+    },
+    {
+      "epoch": 13.258342745661773,
+      "grad_norm": 0.0950339138507843,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 129123
+    },
+    {
+      "epoch": 13.258650785501592,
+      "grad_norm": 0.03336053714156151,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 129126
+    },
+    {
+      "epoch": 13.258958825341411,
+      "grad_norm": 0.08942116051912308,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 129129
+    },
+    {
+      "epoch": 13.25926686518123,
+      "grad_norm": 0.04951188713312149,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 129132
+    },
+    {
+      "epoch": 13.259574905021049,
+      "grad_norm": 0.07134432345628738,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 129135
+    },
+    {
+      "epoch": 13.259882944860868,
+      "grad_norm": 0.13347074389457703,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 129138
+    },
+    {
+      "epoch": 13.260190984700689,
+      "grad_norm": 0.05231053754687309,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 129141
+    },
+    {
+      "epoch": 13.260499024540508,
+      "grad_norm": 0.09639463573694229,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 129144
+    },
+    {
+      "epoch": 13.260807064380327,
+      "grad_norm": 0.09362033009529114,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 129147
+    },
+    {
+      "epoch": 13.261115104220146,
+      "grad_norm": 0.061838217079639435,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 129150
+    },
+    {
+      "epoch": 13.261423144059965,
+      "grad_norm": 0.03951176628470421,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 129153
+    },
+    {
+      "epoch": 13.261731183899784,
+      "grad_norm": 0.04043121263384819,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 129156
+    },
+    {
+      "epoch": 13.262039223739604,
+      "grad_norm": 0.044104333966970444,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 129159
+    },
+    {
+      "epoch": 13.262347263579423,
+      "grad_norm": 0.06577737629413605,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 129162
+    },
+    {
+      "epoch": 13.262655303419242,
+      "grad_norm": 0.05588099732995033,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 129165
+    },
+    {
+      "epoch": 13.262963343259061,
+      "grad_norm": 0.1303408443927765,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 129168
+    },
+    {
+      "epoch": 13.26327138309888,
+      "grad_norm": 0.10937488824129105,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 129171
+    },
+    {
+      "epoch": 13.263579422938701,
+      "grad_norm": 0.09792722016572952,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 129174
+    },
+    {
+      "epoch": 13.26388746277852,
+      "grad_norm": 0.047898069024086,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 129177
+    },
+    {
+      "epoch": 13.264195502618339,
+      "grad_norm": 0.14936259388923645,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 129180
+    },
+    {
+      "epoch": 13.264503542458158,
+      "grad_norm": 0.0617082342505455,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 129183
+    },
+    {
+      "epoch": 13.264811582297977,
+      "grad_norm": 0.06562946736812592,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 129186
+    },
+    {
+      "epoch": 13.265119622137796,
+      "grad_norm": 0.05888977274298668,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 129189
+    },
+    {
+      "epoch": 13.265427661977617,
+      "grad_norm": 0.03986068814992905,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 129192
+    },
+    {
+      "epoch": 13.265735701817436,
+      "grad_norm": 0.04444696754217148,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 129195
+    },
+    {
+      "epoch": 13.266043741657255,
+      "grad_norm": 0.10417811572551727,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 129198
+    },
+    {
+      "epoch": 13.266351781497074,
+      "grad_norm": 0.04169655963778496,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 129201
+    },
+    {
+      "epoch": 13.266659821336892,
+      "grad_norm": 0.032088808715343475,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 129204
+    },
+    {
+      "epoch": 13.266967861176711,
+      "grad_norm": 0.05886637419462204,
+      "learning_rate": 0.01,
+      "loss": 1.8867,
+      "step": 129207
+    },
+    {
+      "epoch": 13.267275901016532,
+      "grad_norm": 0.07081541419029236,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 129210
+    },
+    {
+      "epoch": 13.267583940856351,
+      "grad_norm": 0.059540439397096634,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 129213
+    },
+    {
+      "epoch": 13.26789198069617,
+      "grad_norm": 0.0601552315056324,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 129216
+    },
+    {
+      "epoch": 13.26820002053599,
+      "grad_norm": 0.04202795773744583,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 129219
+    },
+    {
+      "epoch": 13.268508060375808,
+      "grad_norm": 0.039355963468551636,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 129222
+    },
+    {
+      "epoch": 13.268816100215627,
+      "grad_norm": 0.10786538571119308,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 129225
+    },
+    {
+      "epoch": 13.269124140055448,
+      "grad_norm": 0.06146677955985069,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 129228
+    },
+    {
+      "epoch": 13.269432179895267,
+      "grad_norm": 0.08901239931583405,
+      "learning_rate": 0.01,
+      "loss": 1.9509,
+      "step": 129231
+    },
+    {
+      "epoch": 13.269740219735086,
+      "grad_norm": 0.09419004619121552,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 129234
+    },
+    {
+      "epoch": 13.270048259574905,
+      "grad_norm": 0.05888989567756653,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 129237
+    },
+    {
+      "epoch": 13.270356299414724,
+      "grad_norm": 0.09621471166610718,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 129240
+    },
+    {
+      "epoch": 13.270664339254544,
+      "grad_norm": 0.0457032211124897,
+      "learning_rate": 0.01,
+      "loss": 1.9019,
+      "step": 129243
+    },
+    {
+      "epoch": 13.270972379094363,
+      "grad_norm": 0.09191287308931351,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 129246
+    },
+    {
+      "epoch": 13.271280418934182,
+      "grad_norm": 0.08979922533035278,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 129249
+    },
+    {
+      "epoch": 13.271588458774001,
+      "grad_norm": 0.0393611416220665,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 129252
+    },
+    {
+      "epoch": 13.27189649861382,
+      "grad_norm": 0.050079021602869034,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 129255
+    },
+    {
+      "epoch": 13.27220453845364,
+      "grad_norm": 0.06586186587810516,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 129258
+    },
+    {
+      "epoch": 13.27251257829346,
+      "grad_norm": 0.08674393594264984,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 129261
+    },
+    {
+      "epoch": 13.272820618133279,
+      "grad_norm": 0.09615276008844376,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 129264
+    },
+    {
+      "epoch": 13.273128657973098,
+      "grad_norm": 0.14044375717639923,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 129267
+    },
+    {
+      "epoch": 13.273436697812917,
+      "grad_norm": 0.06823724508285522,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 129270
+    },
+    {
+      "epoch": 13.273744737652736,
+      "grad_norm": 0.04253046587109566,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 129273
+    },
+    {
+      "epoch": 13.274052777492555,
+      "grad_norm": 0.07154551148414612,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 129276
+    },
+    {
+      "epoch": 13.274360817332376,
+      "grad_norm": 0.06567098945379257,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 129279
+    },
+    {
+      "epoch": 13.274668857172195,
+      "grad_norm": 0.0839380994439125,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 129282
+    },
+    {
+      "epoch": 13.274976897012014,
+      "grad_norm": 0.05045716091990471,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 129285
+    },
+    {
+      "epoch": 13.275284936851833,
+      "grad_norm": 0.10036513954401016,
+      "learning_rate": 0.01,
+      "loss": 1.9446,
+      "step": 129288
+    },
+    {
+      "epoch": 13.275592976691652,
+      "grad_norm": 0.07544377446174622,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 129291
+    },
+    {
+      "epoch": 13.27590101653147,
+      "grad_norm": 0.06734519451856613,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 129294
+    },
+    {
+      "epoch": 13.276209056371291,
+      "grad_norm": 0.04537443444132805,
+      "learning_rate": 0.01,
+      "loss": 1.8981,
+      "step": 129297
+    },
+    {
+      "epoch": 13.27651709621111,
+      "grad_norm": 0.08838620036840439,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 129300
+    },
+    {
+      "epoch": 13.27682513605093,
+      "grad_norm": 0.05732221528887749,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 129303
+    },
+    {
+      "epoch": 13.277133175890748,
+      "grad_norm": 0.11972025036811829,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 129306
+    },
+    {
+      "epoch": 13.277441215730567,
+      "grad_norm": 0.05065598711371422,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 129309
+    },
+    {
+      "epoch": 13.277749255570388,
+      "grad_norm": 0.06398658454418182,
+      "learning_rate": 0.01,
+      "loss": 1.9013,
+      "step": 129312
+    },
+    {
+      "epoch": 13.278057295410207,
+      "grad_norm": 0.033345576375722885,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 129315
+    },
+    {
+      "epoch": 13.278365335250026,
+      "grad_norm": 0.046490930020809174,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 129318
+    },
+    {
+      "epoch": 13.278673375089845,
+      "grad_norm": 0.04050452262163162,
+      "learning_rate": 0.01,
+      "loss": 1.9394,
+      "step": 129321
+    },
+    {
+      "epoch": 13.278981414929664,
+      "grad_norm": 0.10430192947387695,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 129324
+    },
+    {
+      "epoch": 13.279289454769483,
+      "grad_norm": 0.1070808693766594,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 129327
+    },
+    {
+      "epoch": 13.279597494609304,
+      "grad_norm": 0.04915228858590126,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 129330
+    },
+    {
+      "epoch": 13.279905534449123,
+      "grad_norm": 0.05305736884474754,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 129333
+    },
+    {
+      "epoch": 13.280213574288942,
+      "grad_norm": 0.03477869927883148,
+      "learning_rate": 0.01,
+      "loss": 1.8987,
+      "step": 129336
+    },
+    {
+      "epoch": 13.28052161412876,
+      "grad_norm": 0.08337414264678955,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 129339
+    },
+    {
+      "epoch": 13.28082965396858,
+      "grad_norm": 0.07723475992679596,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 129342
+    },
+    {
+      "epoch": 13.281137693808398,
+      "grad_norm": 0.0686800554394722,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 129345
+    },
+    {
+      "epoch": 13.28144573364822,
+      "grad_norm": 0.11985764652490616,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 129348
+    },
+    {
+      "epoch": 13.281753773488038,
+      "grad_norm": 0.08577185869216919,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 129351
+    },
+    {
+      "epoch": 13.282061813327857,
+      "grad_norm": 0.07968702167272568,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 129354
+    },
+    {
+      "epoch": 13.282369853167676,
+      "grad_norm": 0.07487525790929794,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 129357
+    },
+    {
+      "epoch": 13.282677893007495,
+      "grad_norm": 0.04445855692028999,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 129360
+    },
+    {
+      "epoch": 13.282985932847314,
+      "grad_norm": 0.04486501216888428,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 129363
+    },
+    {
+      "epoch": 13.283293972687135,
+      "grad_norm": 0.03668724000453949,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 129366
+    },
+    {
+      "epoch": 13.283602012526954,
+      "grad_norm": 0.12663684785366058,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 129369
+    },
+    {
+      "epoch": 13.283910052366773,
+      "grad_norm": 0.0821157768368721,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 129372
+    },
+    {
+      "epoch": 13.284218092206592,
+      "grad_norm": 0.12194664031267166,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 129375
+    },
+    {
+      "epoch": 13.28452613204641,
+      "grad_norm": 0.08542870730161667,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 129378
+    },
+    {
+      "epoch": 13.284834171886232,
+      "grad_norm": 0.047838397324085236,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 129381
+    },
+    {
+      "epoch": 13.28514221172605,
+      "grad_norm": 0.030088691040873528,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 129384
+    },
+    {
+      "epoch": 13.28545025156587,
+      "grad_norm": 0.04176461696624756,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 129387
+    },
+    {
+      "epoch": 13.285758291405688,
+      "grad_norm": 0.05294835940003395,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 129390
+    },
+    {
+      "epoch": 13.286066331245507,
+      "grad_norm": 0.03463110327720642,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 129393
+    },
+    {
+      "epoch": 13.286374371085326,
+      "grad_norm": 0.038815587759017944,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 129396
+    },
+    {
+      "epoch": 13.286682410925147,
+      "grad_norm": 0.08738641440868378,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 129399
+    },
+    {
+      "epoch": 13.286990450764966,
+      "grad_norm": 0.05193262919783592,
+      "learning_rate": 0.01,
+      "loss": 1.8981,
+      "step": 129402
+    },
+    {
+      "epoch": 13.287298490604785,
+      "grad_norm": 0.08838549256324768,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 129405
+    },
+    {
+      "epoch": 13.287606530444604,
+      "grad_norm": 0.04979011043906212,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 129408
+    },
+    {
+      "epoch": 13.287914570284423,
+      "grad_norm": 0.05618011951446533,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 129411
+    },
+    {
+      "epoch": 13.288222610124242,
+      "grad_norm": 0.06593050062656403,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 129414
+    },
+    {
+      "epoch": 13.288530649964063,
+      "grad_norm": 0.0682208314538002,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 129417
+    },
+    {
+      "epoch": 13.288838689803882,
+      "grad_norm": 0.04053293541073799,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 129420
+    },
+    {
+      "epoch": 13.2891467296437,
+      "grad_norm": 0.10378430038690567,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 129423
+    },
+    {
+      "epoch": 13.28945476948352,
+      "grad_norm": 0.05505985766649246,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 129426
+    },
+    {
+      "epoch": 13.289762809323339,
+      "grad_norm": 0.12167011946439743,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 129429
+    },
+    {
+      "epoch": 13.290070849163158,
+      "grad_norm": 0.08337932080030441,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 129432
+    },
+    {
+      "epoch": 13.290378889002978,
+      "grad_norm": 0.09331519156694412,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 129435
+    },
+    {
+      "epoch": 13.290686928842797,
+      "grad_norm": 0.03904835879802704,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 129438
+    },
+    {
+      "epoch": 13.290994968682616,
+      "grad_norm": 0.03374563157558441,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 129441
+    },
+    {
+      "epoch": 13.291303008522435,
+      "grad_norm": 0.04161783307790756,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 129444
+    },
+    {
+      "epoch": 13.291611048362254,
+      "grad_norm": 0.056603316217660904,
+      "learning_rate": 0.01,
+      "loss": 1.9006,
+      "step": 129447
+    },
+    {
+      "epoch": 13.291919088202075,
+      "grad_norm": 0.042747288942337036,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 129450
+    },
+    {
+      "epoch": 13.292227128041894,
+      "grad_norm": 0.04726342111825943,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 129453
+    },
+    {
+      "epoch": 13.292535167881713,
+      "grad_norm": 0.045441534370183945,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 129456
+    },
+    {
+      "epoch": 13.292843207721532,
+      "grad_norm": 0.040874212980270386,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 129459
+    },
+    {
+      "epoch": 13.293151247561351,
+      "grad_norm": 0.051837749779224396,
+      "learning_rate": 0.01,
+      "loss": 1.9076,
+      "step": 129462
+    },
+    {
+      "epoch": 13.29345928740117,
+      "grad_norm": 0.10409622639417648,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 129465
+    },
+    {
+      "epoch": 13.29376732724099,
+      "grad_norm": 0.057092078030109406,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 129468
+    },
+    {
+      "epoch": 13.29407536708081,
+      "grad_norm": 0.08772090077400208,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 129471
+    },
+    {
+      "epoch": 13.294383406920629,
+      "grad_norm": 0.06744213402271271,
+      "learning_rate": 0.01,
+      "loss": 1.8968,
+      "step": 129474
+    },
+    {
+      "epoch": 13.294691446760448,
+      "grad_norm": 0.06306581199169159,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 129477
+    },
+    {
+      "epoch": 13.294999486600267,
+      "grad_norm": 0.051581550389528275,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 129480
+    },
+    {
+      "epoch": 13.295307526440086,
+      "grad_norm": 0.05424034968018532,
+      "learning_rate": 0.01,
+      "loss": 1.8862,
+      "step": 129483
+    },
+    {
+      "epoch": 13.295615566279906,
+      "grad_norm": 0.059419360011816025,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 129486
+    },
+    {
+      "epoch": 13.295923606119725,
+      "grad_norm": 0.0704602599143982,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 129489
+    },
+    {
+      "epoch": 13.296231645959544,
+      "grad_norm": 0.06067100539803505,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 129492
+    },
+    {
+      "epoch": 13.296539685799363,
+      "grad_norm": 0.09262153506278992,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 129495
+    },
+    {
+      "epoch": 13.296847725639182,
+      "grad_norm": 0.054725922644138336,
+      "learning_rate": 0.01,
+      "loss": 1.8847,
+      "step": 129498
+    },
+    {
+      "epoch": 13.297155765479001,
+      "grad_norm": 0.052223678678274155,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 129501
+    },
+    {
+      "epoch": 13.297463805318822,
+      "grad_norm": 0.08851344883441925,
+      "learning_rate": 0.01,
+      "loss": 1.9567,
+      "step": 129504
+    },
+    {
+      "epoch": 13.29777184515864,
+      "grad_norm": 0.03212873637676239,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 129507
+    },
+    {
+      "epoch": 13.29807988499846,
+      "grad_norm": 0.045464131981134415,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 129510
+    },
+    {
+      "epoch": 13.298387924838279,
+      "grad_norm": 0.04933739826083183,
+      "learning_rate": 0.01,
+      "loss": 1.9114,
+      "step": 129513
+    },
+    {
+      "epoch": 13.298695964678098,
+      "grad_norm": 0.037281449884176254,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 129516
+    },
+    {
+      "epoch": 13.299004004517919,
+      "grad_norm": 0.03696246072649956,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 129519
+    },
+    {
+      "epoch": 13.299312044357738,
+      "grad_norm": 0.1711316555738449,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 129522
+    },
+    {
+      "epoch": 13.299620084197556,
+      "grad_norm": 0.11211679875850677,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 129525
+    },
+    {
+      "epoch": 13.299928124037375,
+      "grad_norm": 0.05875490605831146,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 129528
+    },
+    {
+      "epoch": 13.300236163877194,
+      "grad_norm": 0.0514705590903759,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 129531
+    },
+    {
+      "epoch": 13.300544203717013,
+      "grad_norm": 0.053770195692777634,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 129534
+    },
+    {
+      "epoch": 13.300852243556834,
+      "grad_norm": 0.08457176387310028,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 129537
+    },
+    {
+      "epoch": 13.301160283396653,
+      "grad_norm": 0.05689070373773575,
+      "learning_rate": 0.01,
+      "loss": 1.8996,
+      "step": 129540
+    },
+    {
+      "epoch": 13.301468323236472,
+      "grad_norm": 0.037381984293460846,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 129543
+    },
+    {
+      "epoch": 13.301776363076291,
+      "grad_norm": 0.04631548002362251,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 129546
+    },
+    {
+      "epoch": 13.30208440291611,
+      "grad_norm": 0.06648321449756622,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 129549
+    },
+    {
+      "epoch": 13.302392442755929,
+      "grad_norm": 0.07430995255708694,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 129552
+    },
+    {
+      "epoch": 13.30270048259575,
+      "grad_norm": 0.08703188598155975,
+      "learning_rate": 0.01,
+      "loss": 1.8941,
+      "step": 129555
+    },
+    {
+      "epoch": 13.303008522435569,
+      "grad_norm": 0.11382687836885452,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 129558
+    },
+    {
+      "epoch": 13.303316562275388,
+      "grad_norm": 0.05929556116461754,
+      "learning_rate": 0.01,
+      "loss": 1.8914,
+      "step": 129561
+    },
+    {
+      "epoch": 13.303624602115207,
+      "grad_norm": 0.04573335871100426,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 129564
+    },
+    {
+      "epoch": 13.303932641955026,
+      "grad_norm": 0.06573352962732315,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 129567
+    },
+    {
+      "epoch": 13.304240681794845,
+      "grad_norm": 0.044627655297517776,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 129570
+    },
+    {
+      "epoch": 13.304548721634665,
+      "grad_norm": 0.053817469626665115,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 129573
+    },
+    {
+      "epoch": 13.304856761474484,
+      "grad_norm": 0.11950349807739258,
+      "learning_rate": 0.01,
+      "loss": 1.8973,
+      "step": 129576
+    },
+    {
+      "epoch": 13.305164801314303,
+      "grad_norm": 0.03673816844820976,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 129579
+    },
+    {
+      "epoch": 13.305472841154122,
+      "grad_norm": 0.0890408605337143,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 129582
+    },
+    {
+      "epoch": 13.305780880993941,
+      "grad_norm": 0.06715387850999832,
+      "learning_rate": 0.01,
+      "loss": 1.9055,
+      "step": 129585
+    },
+    {
+      "epoch": 13.306088920833762,
+      "grad_norm": 0.034186918288469315,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 129588
+    },
+    {
+      "epoch": 13.306396960673581,
+      "grad_norm": 0.06232522055506706,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 129591
+    },
+    {
+      "epoch": 13.3067050005134,
+      "grad_norm": 0.11120866239070892,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 129594
+    },
+    {
+      "epoch": 13.307013040353219,
+      "grad_norm": 0.1457282155752182,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 129597
+    },
+    {
+      "epoch": 13.307321080193038,
+      "grad_norm": 0.1338082104921341,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 129600
+    },
+    {
+      "epoch": 13.307629120032857,
+      "grad_norm": 0.037618059664964676,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 129603
+    },
+    {
+      "epoch": 13.307937159872678,
+      "grad_norm": 0.03888382017612457,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 129606
+    },
+    {
+      "epoch": 13.308245199712497,
+      "grad_norm": 0.04355417564511299,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 129609
+    },
+    {
+      "epoch": 13.308553239552316,
+      "grad_norm": 0.055194199085235596,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 129612
+    },
+    {
+      "epoch": 13.308861279392135,
+      "grad_norm": 0.11141913384199142,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 129615
+    },
+    {
+      "epoch": 13.309169319231954,
+      "grad_norm": 0.042318303138017654,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 129618
+    },
+    {
+      "epoch": 13.309477359071773,
+      "grad_norm": 0.10116101801395416,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 129621
+    },
+    {
+      "epoch": 13.309785398911593,
+      "grad_norm": 0.08044546097517014,
+      "learning_rate": 0.01,
+      "loss": 1.8845,
+      "step": 129624
+    },
+    {
+      "epoch": 13.310093438751412,
+      "grad_norm": 0.057655222713947296,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 129627
+    },
+    {
+      "epoch": 13.310401478591231,
+      "grad_norm": 0.03532911837100983,
+      "learning_rate": 0.01,
+      "loss": 1.8881,
+      "step": 129630
+    },
+    {
+      "epoch": 13.31070951843105,
+      "grad_norm": 0.06472481042146683,
+      "learning_rate": 0.01,
+      "loss": 1.8967,
+      "step": 129633
+    },
+    {
+      "epoch": 13.31101755827087,
+      "grad_norm": 0.053687356412410736,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 129636
+    },
+    {
+      "epoch": 13.311325598110688,
+      "grad_norm": 0.09955917298793793,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 129639
+    },
+    {
+      "epoch": 13.311633637950509,
+      "grad_norm": 0.05217433348298073,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 129642
+    },
+    {
+      "epoch": 13.311941677790328,
+      "grad_norm": 0.03496095910668373,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 129645
+    },
+    {
+      "epoch": 13.312249717630147,
+      "grad_norm": 0.13012070953845978,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 129648
+    },
+    {
+      "epoch": 13.312557757469966,
+      "grad_norm": 0.08615440875291824,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 129651
+    },
+    {
+      "epoch": 13.312865797309785,
+      "grad_norm": 0.0574072040617466,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 129654
+    },
+    {
+      "epoch": 13.313173837149606,
+      "grad_norm": 0.060519132763147354,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 129657
+    },
+    {
+      "epoch": 13.313481876989425,
+      "grad_norm": 0.04307126998901367,
+      "learning_rate": 0.01,
+      "loss": 1.9055,
+      "step": 129660
+    },
+    {
+      "epoch": 13.313789916829244,
+      "grad_norm": 0.0490017868578434,
+      "learning_rate": 0.01,
+      "loss": 1.9337,
+      "step": 129663
+    },
+    {
+      "epoch": 13.314097956669062,
+      "grad_norm": 0.053116172552108765,
+      "learning_rate": 0.01,
+      "loss": 1.9428,
+      "step": 129666
+    },
+    {
+      "epoch": 13.314405996508881,
+      "grad_norm": 0.06051335856318474,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 129669
+    },
+    {
+      "epoch": 13.3147140363487,
+      "grad_norm": 0.11569323390722275,
+      "learning_rate": 0.01,
+      "loss": 1.9455,
+      "step": 129672
+    },
+    {
+      "epoch": 13.315022076188521,
+      "grad_norm": 0.03279983624815941,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 129675
+    },
+    {
+      "epoch": 13.31533011602834,
+      "grad_norm": 0.08274318277835846,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 129678
+    },
+    {
+      "epoch": 13.31563815586816,
+      "grad_norm": 0.07533428817987442,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 129681
+    },
+    {
+      "epoch": 13.315946195707978,
+      "grad_norm": 0.08243946731090546,
+      "learning_rate": 0.01,
+      "loss": 1.8904,
+      "step": 129684
+    },
+    {
+      "epoch": 13.316254235547797,
+      "grad_norm": 0.050875160843133926,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 129687
+    },
+    {
+      "epoch": 13.316562275387616,
+      "grad_norm": 0.10925690084695816,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 129690
+    },
+    {
+      "epoch": 13.316870315227437,
+      "grad_norm": 0.05127814784646034,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 129693
+    },
+    {
+      "epoch": 13.317178355067256,
+      "grad_norm": 0.037685733288526535,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 129696
+    },
+    {
+      "epoch": 13.317486394907075,
+      "grad_norm": 0.039268460124731064,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 129699
+    },
+    {
+      "epoch": 13.317794434746894,
+      "grad_norm": 0.0610152892768383,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 129702
+    },
+    {
+      "epoch": 13.318102474586713,
+      "grad_norm": 0.11556638777256012,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 129705
+    },
+    {
+      "epoch": 13.318410514426532,
+      "grad_norm": 0.1131361722946167,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 129708
+    },
+    {
+      "epoch": 13.318718554266352,
+      "grad_norm": 0.05956597253680229,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 129711
+    },
+    {
+      "epoch": 13.319026594106171,
+      "grad_norm": 0.06298226863145828,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 129714
+    },
+    {
+      "epoch": 13.31933463394599,
+      "grad_norm": 0.053446751087903976,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 129717
+    },
+    {
+      "epoch": 13.31964267378581,
+      "grad_norm": 0.036532558500766754,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 129720
+    },
+    {
+      "epoch": 13.319950713625628,
+      "grad_norm": 0.04398741573095322,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 129723
+    },
+    {
+      "epoch": 13.320258753465449,
+      "grad_norm": 0.0671413242816925,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 129726
+    },
+    {
+      "epoch": 13.320566793305268,
+      "grad_norm": 0.04935973510146141,
+      "learning_rate": 0.01,
+      "loss": 1.9009,
+      "step": 129729
+    },
+    {
+      "epoch": 13.320874833145087,
+      "grad_norm": 0.07472071796655655,
+      "learning_rate": 0.01,
+      "loss": 1.8924,
+      "step": 129732
+    },
+    {
+      "epoch": 13.321182872984906,
+      "grad_norm": 0.0450054369866848,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 129735
+    },
+    {
+      "epoch": 13.321490912824725,
+      "grad_norm": 0.045300472527742386,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 129738
+    },
+    {
+      "epoch": 13.321798952664544,
+      "grad_norm": 0.043478891253471375,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 129741
+    },
+    {
+      "epoch": 13.322106992504365,
+      "grad_norm": 0.14465944468975067,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 129744
+    },
+    {
+      "epoch": 13.322415032344184,
+      "grad_norm": 0.0522492378950119,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 129747
+    },
+    {
+      "epoch": 13.322723072184003,
+      "grad_norm": 0.10883329063653946,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 129750
+    },
+    {
+      "epoch": 13.323031112023822,
+      "grad_norm": 0.03729747235774994,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 129753
+    },
+    {
+      "epoch": 13.32333915186364,
+      "grad_norm": 0.11612371355295181,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 129756
+    },
+    {
+      "epoch": 13.32364719170346,
+      "grad_norm": 0.05548815801739693,
+      "learning_rate": 0.01,
+      "loss": 1.8997,
+      "step": 129759
+    },
+    {
+      "epoch": 13.32395523154328,
+      "grad_norm": 0.04483690857887268,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 129762
+    },
+    {
+      "epoch": 13.3242632713831,
+      "grad_norm": 0.04981400817632675,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 129765
+    },
+    {
+      "epoch": 13.324571311222918,
+      "grad_norm": 0.0443868488073349,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 129768
+    },
+    {
+      "epoch": 13.324879351062737,
+      "grad_norm": 0.12509149312973022,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 129771
+    },
+    {
+      "epoch": 13.325187390902556,
+      "grad_norm": 0.06622964143753052,
+      "learning_rate": 0.01,
+      "loss": 1.949,
+      "step": 129774
+    },
+    {
+      "epoch": 13.325495430742375,
+      "grad_norm": 0.12362387776374817,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 129777
+    },
+    {
+      "epoch": 13.325803470582196,
+      "grad_norm": 0.08693816512823105,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 129780
+    },
+    {
+      "epoch": 13.326111510422015,
+      "grad_norm": 0.10090692341327667,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 129783
+    },
+    {
+      "epoch": 13.326419550261834,
+      "grad_norm": 0.09820685535669327,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 129786
+    },
+    {
+      "epoch": 13.326727590101653,
+      "grad_norm": 0.046172816306352615,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 129789
+    },
+    {
+      "epoch": 13.327035629941472,
+      "grad_norm": 0.04443272203207016,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 129792
+    },
+    {
+      "epoch": 13.327343669781293,
+      "grad_norm": 0.06589057296514511,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 129795
+    },
+    {
+      "epoch": 13.327651709621112,
+      "grad_norm": 0.04592718183994293,
+      "learning_rate": 0.01,
+      "loss": 1.9421,
+      "step": 129798
+    },
+    {
+      "epoch": 13.32795974946093,
+      "grad_norm": 0.04794567450881004,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 129801
+    },
+    {
+      "epoch": 13.32826778930075,
+      "grad_norm": 0.047602783888578415,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 129804
+    },
+    {
+      "epoch": 13.328575829140568,
+      "grad_norm": 0.0432455874979496,
+      "learning_rate": 0.01,
+      "loss": 1.8961,
+      "step": 129807
+    },
+    {
+      "epoch": 13.328883868980387,
+      "grad_norm": 0.1239074170589447,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 129810
+    },
+    {
+      "epoch": 13.329191908820208,
+      "grad_norm": 0.03249470144510269,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 129813
+    },
+    {
+      "epoch": 13.329499948660027,
+      "grad_norm": 0.10205553472042084,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 129816
+    },
+    {
+      "epoch": 13.329807988499846,
+      "grad_norm": 0.04515647515654564,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 129819
+    },
+    {
+      "epoch": 13.330116028339665,
+      "grad_norm": 0.06119481846690178,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 129822
+    },
+    {
+      "epoch": 13.330424068179484,
+      "grad_norm": 0.03687642887234688,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 129825
+    },
+    {
+      "epoch": 13.330732108019303,
+      "grad_norm": 0.04972810298204422,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 129828
+    },
+    {
+      "epoch": 13.331040147859124,
+      "grad_norm": 0.09871906042098999,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 129831
+    },
+    {
+      "epoch": 13.331348187698943,
+      "grad_norm": 0.0817904993891716,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 129834
+    },
+    {
+      "epoch": 13.331656227538762,
+      "grad_norm": 0.040375031530857086,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 129837
+    },
+    {
+      "epoch": 13.33196426737858,
+      "grad_norm": 0.12215966731309891,
+      "learning_rate": 0.01,
+      "loss": 1.9458,
+      "step": 129840
+    },
+    {
+      "epoch": 13.3322723072184,
+      "grad_norm": 0.05876701697707176,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 129843
+    },
+    {
+      "epoch": 13.332580347058219,
+      "grad_norm": 0.06052813306450844,
+      "learning_rate": 0.01,
+      "loss": 1.9002,
+      "step": 129846
+    },
+    {
+      "epoch": 13.33288838689804,
+      "grad_norm": 0.03797772899270058,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 129849
+    },
+    {
+      "epoch": 13.333196426737858,
+      "grad_norm": 0.04099010303616524,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 129852
+    },
+    {
+      "epoch": 13.333504466577677,
+      "grad_norm": 0.07171392440795898,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 129855
+    },
+    {
+      "epoch": 13.333812506417496,
+      "grad_norm": 0.10205614566802979,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 129858
+    },
+    {
+      "epoch": 13.334120546257315,
+      "grad_norm": 0.04092935100197792,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 129861
+    },
+    {
+      "epoch": 13.334428586097136,
+      "grad_norm": 0.12945953011512756,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 129864
+    },
+    {
+      "epoch": 13.334736625936955,
+      "grad_norm": 0.14541998505592346,
+      "learning_rate": 0.01,
+      "loss": 1.8966,
+      "step": 129867
+    },
+    {
+      "epoch": 13.335044665776774,
+      "grad_norm": 0.040134042501449585,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 129870
+    },
+    {
+      "epoch": 13.335352705616593,
+      "grad_norm": 0.07563397288322449,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 129873
+    },
+    {
+      "epoch": 13.335660745456412,
+      "grad_norm": 0.06526198983192444,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 129876
+    },
+    {
+      "epoch": 13.335968785296231,
+      "grad_norm": 0.035391949117183685,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 129879
+    },
+    {
+      "epoch": 13.336276825136052,
+      "grad_norm": 0.08452790975570679,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 129882
+    },
+    {
+      "epoch": 13.33658486497587,
+      "grad_norm": 0.09037191420793533,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 129885
+    },
+    {
+      "epoch": 13.33689290481569,
+      "grad_norm": 0.060933735221624374,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 129888
+    },
+    {
+      "epoch": 13.337200944655509,
+      "grad_norm": 0.058399271219968796,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 129891
+    },
+    {
+      "epoch": 13.337508984495328,
+      "grad_norm": 0.04956258460879326,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 129894
+    },
+    {
+      "epoch": 13.337817024335147,
+      "grad_norm": 0.069740891456604,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 129897
+    },
+    {
+      "epoch": 13.338125064174967,
+      "grad_norm": 0.11041458696126938,
+      "learning_rate": 0.01,
+      "loss": 1.9592,
+      "step": 129900
+    },
+    {
+      "epoch": 13.338433104014786,
+      "grad_norm": 0.0463983379304409,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 129903
+    },
+    {
+      "epoch": 13.338741143854605,
+      "grad_norm": 0.0891149640083313,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 129906
+    },
+    {
+      "epoch": 13.339049183694424,
+      "grad_norm": 0.036531962454319,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 129909
+    },
+    {
+      "epoch": 13.339357223534243,
+      "grad_norm": 0.05816185846924782,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 129912
+    },
+    {
+      "epoch": 13.339665263374062,
+      "grad_norm": 0.10037883371114731,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 129915
+    },
+    {
+      "epoch": 13.339973303213883,
+      "grad_norm": 0.09014111012220383,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 129918
+    },
+    {
+      "epoch": 13.340281343053702,
+      "grad_norm": 0.04957646131515503,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 129921
+    },
+    {
+      "epoch": 13.340589382893521,
+      "grad_norm": 0.063672736287117,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 129924
+    },
+    {
+      "epoch": 13.34089742273334,
+      "grad_norm": 0.09400395303964615,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 129927
+    },
+    {
+      "epoch": 13.341205462573159,
+      "grad_norm": 0.08700636029243469,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 129930
+    },
+    {
+      "epoch": 13.34151350241298,
+      "grad_norm": 0.06137283518910408,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 129933
+    },
+    {
+      "epoch": 13.341821542252799,
+      "grad_norm": 0.06885108351707458,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 129936
+    },
+    {
+      "epoch": 13.342129582092618,
+      "grad_norm": 0.06410050392150879,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 129939
+    },
+    {
+      "epoch": 13.342437621932437,
+      "grad_norm": 0.04796096310019493,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 129942
+    },
+    {
+      "epoch": 13.342745661772256,
+      "grad_norm": 0.05061439797282219,
+      "learning_rate": 0.01,
+      "loss": 1.888,
+      "step": 129945
+    },
+    {
+      "epoch": 13.343053701612074,
+      "grad_norm": 0.04647940397262573,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 129948
+    },
+    {
+      "epoch": 13.343361741451895,
+      "grad_norm": 0.08649050444364548,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 129951
+    },
+    {
+      "epoch": 13.343669781291714,
+      "grad_norm": 0.07439057528972626,
+      "learning_rate": 0.01,
+      "loss": 1.911,
+      "step": 129954
+    },
+    {
+      "epoch": 13.343977821131533,
+      "grad_norm": 0.04305770620703697,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 129957
+    },
+    {
+      "epoch": 13.344285860971352,
+      "grad_norm": 0.0477466955780983,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 129960
+    },
+    {
+      "epoch": 13.344593900811171,
+      "grad_norm": 0.030323054641485214,
+      "learning_rate": 0.01,
+      "loss": 1.905,
+      "step": 129963
+    },
+    {
+      "epoch": 13.34490194065099,
+      "grad_norm": 0.06439530104398727,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 129966
+    },
+    {
+      "epoch": 13.34520998049081,
+      "grad_norm": 0.13896659016609192,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 129969
+    },
+    {
+      "epoch": 13.34551802033063,
+      "grad_norm": 0.06615673005580902,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 129972
+    },
+    {
+      "epoch": 13.345826060170449,
+      "grad_norm": 0.04389958083629608,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 129975
+    },
+    {
+      "epoch": 13.346134100010268,
+      "grad_norm": 0.04844463989138603,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 129978
+    },
+    {
+      "epoch": 13.346442139850087,
+      "grad_norm": 0.04833073541522026,
+      "learning_rate": 0.01,
+      "loss": 1.9465,
+      "step": 129981
+    },
+    {
+      "epoch": 13.346750179689906,
+      "grad_norm": 0.04213207960128784,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 129984
+    },
+    {
+      "epoch": 13.347058219529726,
+      "grad_norm": 0.04052984341979027,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 129987
+    },
+    {
+      "epoch": 13.347366259369545,
+      "grad_norm": 0.07937126606702805,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 129990
+    },
+    {
+      "epoch": 13.347674299209364,
+      "grad_norm": 0.04177657887339592,
+      "learning_rate": 0.01,
+      "loss": 1.8955,
+      "step": 129993
+    },
+    {
+      "epoch": 13.347982339049183,
+      "grad_norm": 0.052641861140728,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 129996
+    },
+    {
+      "epoch": 13.348290378889002,
+      "grad_norm": 0.054449647665023804,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 129999
+    },
+    {
+      "epoch": 13.348598418728823,
+      "grad_norm": 0.07911340147256851,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 130002
+    },
+    {
+      "epoch": 13.348906458568642,
+      "grad_norm": 0.11009012907743454,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 130005
+    },
+    {
+      "epoch": 13.349214498408461,
+      "grad_norm": 0.1320761889219284,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 130008
+    },
+    {
+      "epoch": 13.34952253824828,
+      "grad_norm": 0.14681044220924377,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 130011
+    },
+    {
+      "epoch": 13.349830578088099,
+      "grad_norm": 0.11095894873142242,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 130014
+    },
+    {
+      "epoch": 13.350138617927918,
+      "grad_norm": 0.07749883830547333,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 130017
+    },
+    {
+      "epoch": 13.350446657767739,
+      "grad_norm": 0.12986674904823303,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 130020
+    },
+    {
+      "epoch": 13.350754697607558,
+      "grad_norm": 0.10472730547189713,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 130023
+    },
+    {
+      "epoch": 13.351062737447377,
+      "grad_norm": 0.05852092429995537,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 130026
+    },
+    {
+      "epoch": 13.351370777287196,
+      "grad_norm": 0.03161657974123955,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 130029
+    },
+    {
+      "epoch": 13.351678817127015,
+      "grad_norm": 0.05806410312652588,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 130032
+    },
+    {
+      "epoch": 13.351986856966834,
+      "grad_norm": 0.030168121680617332,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 130035
+    },
+    {
+      "epoch": 13.352294896806654,
+      "grad_norm": 0.06878235191106796,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 130038
+    },
+    {
+      "epoch": 13.352602936646473,
+      "grad_norm": 0.035785697400569916,
+      "learning_rate": 0.01,
+      "loss": 1.9628,
+      "step": 130041
+    },
+    {
+      "epoch": 13.352910976486292,
+      "grad_norm": 0.03621627762913704,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 130044
+    },
+    {
+      "epoch": 13.353219016326111,
+      "grad_norm": 0.07845892757177353,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 130047
+    },
+    {
+      "epoch": 13.35352705616593,
+      "grad_norm": 0.09262198954820633,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 130050
+    },
+    {
+      "epoch": 13.35383509600575,
+      "grad_norm": 0.05965424329042435,
+      "learning_rate": 0.01,
+      "loss": 1.8941,
+      "step": 130053
+    },
+    {
+      "epoch": 13.35414313584557,
+      "grad_norm": 0.08931989222764969,
+      "learning_rate": 0.01,
+      "loss": 1.9001,
+      "step": 130056
+    },
+    {
+      "epoch": 13.354451175685389,
+      "grad_norm": 0.0829695388674736,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 130059
+    },
+    {
+      "epoch": 13.354759215525208,
+      "grad_norm": 0.0733577311038971,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 130062
+    },
+    {
+      "epoch": 13.355067255365027,
+      "grad_norm": 0.05141898989677429,
+      "learning_rate": 0.01,
+      "loss": 1.8948,
+      "step": 130065
+    },
+    {
+      "epoch": 13.355375295204846,
+      "grad_norm": 0.0457218661904335,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 130068
+    },
+    {
+      "epoch": 13.355683335044667,
+      "grad_norm": 0.04116986691951752,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 130071
+    },
+    {
+      "epoch": 13.355991374884486,
+      "grad_norm": 0.09544950723648071,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 130074
+    },
+    {
+      "epoch": 13.356299414724305,
+      "grad_norm": 0.07874602824449539,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 130077
+    },
+    {
+      "epoch": 13.356607454564124,
+      "grad_norm": 0.06750241667032242,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 130080
+    },
+    {
+      "epoch": 13.356915494403943,
+      "grad_norm": 0.038582198321819305,
+      "learning_rate": 0.01,
+      "loss": 1.9068,
+      "step": 130083
+    },
+    {
+      "epoch": 13.357223534243762,
+      "grad_norm": 0.085889533162117,
+      "learning_rate": 0.01,
+      "loss": 1.898,
+      "step": 130086
+    },
+    {
+      "epoch": 13.357531574083582,
+      "grad_norm": 0.07243723422288895,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 130089
+    },
+    {
+      "epoch": 13.357839613923401,
+      "grad_norm": 0.05252539739012718,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 130092
+    },
+    {
+      "epoch": 13.35814765376322,
+      "grad_norm": 0.06544727832078934,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 130095
+    },
+    {
+      "epoch": 13.35845569360304,
+      "grad_norm": 0.05817817151546478,
+      "learning_rate": 0.01,
+      "loss": 1.9056,
+      "step": 130098
+    },
+    {
+      "epoch": 13.358763733442858,
+      "grad_norm": 0.04873080179095268,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 130101
+    },
+    {
+      "epoch": 13.359071773282677,
+      "grad_norm": 0.13894930481910706,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 130104
+    },
+    {
+      "epoch": 13.359379813122498,
+      "grad_norm": 0.10438075661659241,
+      "learning_rate": 0.01,
+      "loss": 1.8987,
+      "step": 130107
+    },
+    {
+      "epoch": 13.359687852962317,
+      "grad_norm": 0.04332970082759857,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 130110
+    },
+    {
+      "epoch": 13.359995892802136,
+      "grad_norm": 0.059613753110170364,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 130113
+    },
+    {
+      "epoch": 13.360303932641955,
+      "grad_norm": 0.0526910126209259,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 130116
+    },
+    {
+      "epoch": 13.360611972481774,
+      "grad_norm": 0.0498511977493763,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 130119
+    },
+    {
+      "epoch": 13.360920012321593,
+      "grad_norm": 0.056588757783174515,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 130122
+    },
+    {
+      "epoch": 13.361228052161414,
+      "grad_norm": 0.04684719070792198,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 130125
+    },
+    {
+      "epoch": 13.361536092001232,
+      "grad_norm": 0.045814819633960724,
+      "learning_rate": 0.01,
+      "loss": 1.8981,
+      "step": 130128
+    },
+    {
+      "epoch": 13.361844131841051,
+      "grad_norm": 0.05945038050413132,
+      "learning_rate": 0.01,
+      "loss": 1.8918,
+      "step": 130131
+    },
+    {
+      "epoch": 13.36215217168087,
+      "grad_norm": 0.1278378814458847,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 130134
+    },
+    {
+      "epoch": 13.36246021152069,
+      "grad_norm": 0.12026727199554443,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 130137
+    },
+    {
+      "epoch": 13.36276825136051,
+      "grad_norm": 0.04181729629635811,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 130140
+    },
+    {
+      "epoch": 13.36307629120033,
+      "grad_norm": 0.03284575790166855,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 130143
+    },
+    {
+      "epoch": 13.363384331040148,
+      "grad_norm": 0.04759792983531952,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 130146
+    },
+    {
+      "epoch": 13.363692370879967,
+      "grad_norm": 0.053606148809194565,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 130149
+    },
+    {
+      "epoch": 13.364000410719786,
+      "grad_norm": 0.06647252291440964,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 130152
+    },
+    {
+      "epoch": 13.364308450559605,
+      "grad_norm": 0.05596812441945076,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 130155
+    },
+    {
+      "epoch": 13.364616490399426,
+      "grad_norm": 0.04482680559158325,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 130158
+    },
+    {
+      "epoch": 13.364924530239245,
+      "grad_norm": 0.11261728405952454,
+      "learning_rate": 0.01,
+      "loss": 1.9366,
+      "step": 130161
+    },
+    {
+      "epoch": 13.365232570079064,
+      "grad_norm": 0.06480871886014938,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 130164
+    },
+    {
+      "epoch": 13.365540609918883,
+      "grad_norm": 0.09778289496898651,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 130167
+    },
+    {
+      "epoch": 13.365848649758702,
+      "grad_norm": 0.05829132720828056,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 130170
+    },
+    {
+      "epoch": 13.36615668959852,
+      "grad_norm": 0.10424979031085968,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 130173
+    },
+    {
+      "epoch": 13.366464729438341,
+      "grad_norm": 0.03707784414291382,
+      "learning_rate": 0.01,
+      "loss": 1.8992,
+      "step": 130176
+    },
+    {
+      "epoch": 13.36677276927816,
+      "grad_norm": 0.0387544222176075,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 130179
+    },
+    {
+      "epoch": 13.36708080911798,
+      "grad_norm": 0.04722850024700165,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 130182
+    },
+    {
+      "epoch": 13.367388848957798,
+      "grad_norm": 0.21787331998348236,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 130185
+    },
+    {
+      "epoch": 13.367696888797617,
+      "grad_norm": 0.18266351521015167,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 130188
+    },
+    {
+      "epoch": 13.368004928637436,
+      "grad_norm": 0.0891261100769043,
+      "learning_rate": 0.01,
+      "loss": 1.8987,
+      "step": 130191
+    },
+    {
+      "epoch": 13.368312968477257,
+      "grad_norm": 0.03555745631456375,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 130194
+    },
+    {
+      "epoch": 13.368621008317076,
+      "grad_norm": 0.03688689321279526,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 130197
+    },
+    {
+      "epoch": 13.368929048156895,
+      "grad_norm": 0.07209073752164841,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 130200
+    },
+    {
+      "epoch": 13.369237087996714,
+      "grad_norm": 0.05605915188789368,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 130203
+    },
+    {
+      "epoch": 13.369545127836533,
+      "grad_norm": 0.06916657835245132,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 130206
+    },
+    {
+      "epoch": 13.369853167676354,
+      "grad_norm": 0.0878639966249466,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 130209
+    },
+    {
+      "epoch": 13.370161207516173,
+      "grad_norm": 0.20460787415504456,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 130212
+    },
+    {
+      "epoch": 13.370469247355992,
+      "grad_norm": 0.17409759759902954,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 130215
+    },
+    {
+      "epoch": 13.37077728719581,
+      "grad_norm": 0.10241883248090744,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 130218
+    },
+    {
+      "epoch": 13.37108532703563,
+      "grad_norm": 0.04647821933031082,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 130221
+    },
+    {
+      "epoch": 13.371393366875449,
+      "grad_norm": 0.036049239337444305,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 130224
+    },
+    {
+      "epoch": 13.37170140671527,
+      "grad_norm": 0.03229028359055519,
+      "learning_rate": 0.01,
+      "loss": 1.89,
+      "step": 130227
+    },
+    {
+      "epoch": 13.372009446555088,
+      "grad_norm": 0.08288008719682693,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 130230
+    },
+    {
+      "epoch": 13.372317486394907,
+      "grad_norm": 0.08158650249242783,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 130233
+    },
+    {
+      "epoch": 13.372625526234726,
+      "grad_norm": 0.09613657742738724,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 130236
+    },
+    {
+      "epoch": 13.372933566074545,
+      "grad_norm": 0.06050723418593407,
+      "learning_rate": 0.01,
+      "loss": 1.8993,
+      "step": 130239
+    },
+    {
+      "epoch": 13.373241605914364,
+      "grad_norm": 0.054775040596723557,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 130242
+    },
+    {
+      "epoch": 13.373549645754185,
+      "grad_norm": 0.03466007485985756,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 130245
+    },
+    {
+      "epoch": 13.373857685594004,
+      "grad_norm": 0.03310166299343109,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 130248
+    },
+    {
+      "epoch": 13.374165725433823,
+      "grad_norm": 0.07772723585367203,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 130251
+    },
+    {
+      "epoch": 13.374473765273642,
+      "grad_norm": 0.038169339299201965,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 130254
+    },
+    {
+      "epoch": 13.37478180511346,
+      "grad_norm": 0.04512426629662514,
+      "learning_rate": 0.01,
+      "loss": 1.9192,
+      "step": 130257
+    },
+    {
+      "epoch": 13.37508984495328,
+      "grad_norm": 0.0583481602370739,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 130260
+    },
+    {
+      "epoch": 13.3753978847931,
+      "grad_norm": 0.09476880729198456,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 130263
+    },
+    {
+      "epoch": 13.37570592463292,
+      "grad_norm": 0.15300163626670837,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 130266
+    },
+    {
+      "epoch": 13.376013964472738,
+      "grad_norm": 0.0430445522069931,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 130269
+    },
+    {
+      "epoch": 13.376322004312557,
+      "grad_norm": 0.07516524195671082,
+      "learning_rate": 0.01,
+      "loss": 1.9083,
+      "step": 130272
+    },
+    {
+      "epoch": 13.376630044152376,
+      "grad_norm": 0.045789092779159546,
+      "learning_rate": 0.01,
+      "loss": 1.8902,
+      "step": 130275
+    },
+    {
+      "epoch": 13.376938083992197,
+      "grad_norm": 0.04921339079737663,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 130278
+    },
+    {
+      "epoch": 13.377246123832016,
+      "grad_norm": 0.04363131523132324,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 130281
+    },
+    {
+      "epoch": 13.377554163671835,
+      "grad_norm": 0.0472864955663681,
+      "learning_rate": 0.01,
+      "loss": 1.9354,
+      "step": 130284
+    },
+    {
+      "epoch": 13.377862203511654,
+      "grad_norm": 0.041092462837696075,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 130287
+    },
+    {
+      "epoch": 13.378170243351473,
+      "grad_norm": 0.050600700080394745,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 130290
+    },
+    {
+      "epoch": 13.378478283191292,
+      "grad_norm": 0.08871981501579285,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 130293
+    },
+    {
+      "epoch": 13.378786323031113,
+      "grad_norm": 0.07451359182596207,
+      "learning_rate": 0.01,
+      "loss": 1.94,
+      "step": 130296
+    },
+    {
+      "epoch": 13.379094362870932,
+      "grad_norm": 0.09753946959972382,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 130299
+    },
+    {
+      "epoch": 13.37940240271075,
+      "grad_norm": 0.04399825260043144,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 130302
+    },
+    {
+      "epoch": 13.37971044255057,
+      "grad_norm": 0.04541864991188049,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 130305
+    },
+    {
+      "epoch": 13.380018482390389,
+      "grad_norm": 0.046489082276821136,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 130308
+    },
+    {
+      "epoch": 13.380326522230208,
+      "grad_norm": 0.03842199593782425,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 130311
+    },
+    {
+      "epoch": 13.380634562070028,
+      "grad_norm": 0.08901941776275635,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 130314
+    },
+    {
+      "epoch": 13.380942601909847,
+      "grad_norm": 0.035034727305173874,
+      "learning_rate": 0.01,
+      "loss": 1.8924,
+      "step": 130317
+    },
+    {
+      "epoch": 13.381250641749666,
+      "grad_norm": 0.03586716204881668,
+      "learning_rate": 0.01,
+      "loss": 1.8976,
+      "step": 130320
+    },
+    {
+      "epoch": 13.381558681589485,
+      "grad_norm": 0.09658779203891754,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 130323
+    },
+    {
+      "epoch": 13.381866721429304,
+      "grad_norm": 0.13156302273273468,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 130326
+    },
+    {
+      "epoch": 13.382174761269123,
+      "grad_norm": 0.06915266066789627,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 130329
+    },
+    {
+      "epoch": 13.382482801108944,
+      "grad_norm": 0.049757279455661774,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 130332
+    },
+    {
+      "epoch": 13.382790840948763,
+      "grad_norm": 0.04229595512151718,
+      "learning_rate": 0.01,
+      "loss": 1.9078,
+      "step": 130335
+    },
+    {
+      "epoch": 13.383098880788582,
+      "grad_norm": 0.03962124511599541,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 130338
+    },
+    {
+      "epoch": 13.383406920628401,
+      "grad_norm": 0.03960268571972847,
+      "learning_rate": 0.01,
+      "loss": 1.9017,
+      "step": 130341
+    },
+    {
+      "epoch": 13.38371496046822,
+      "grad_norm": 0.03750162571668625,
+      "learning_rate": 0.01,
+      "loss": 1.8914,
+      "step": 130344
+    },
+    {
+      "epoch": 13.38402300030804,
+      "grad_norm": 0.0812152773141861,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 130347
+    },
+    {
+      "epoch": 13.38433104014786,
+      "grad_norm": 0.08311905711889267,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 130350
+    },
+    {
+      "epoch": 13.384639079987679,
+      "grad_norm": 0.05700957775115967,
+      "learning_rate": 0.01,
+      "loss": 1.8947,
+      "step": 130353
+    },
+    {
+      "epoch": 13.384947119827498,
+      "grad_norm": 0.045278243720531464,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 130356
+    },
+    {
+      "epoch": 13.385255159667317,
+      "grad_norm": 0.09623616188764572,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 130359
+    },
+    {
+      "epoch": 13.385563199507136,
+      "grad_norm": 0.0714116096496582,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 130362
+    },
+    {
+      "epoch": 13.385871239346956,
+      "grad_norm": 0.150100439786911,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 130365
+    },
+    {
+      "epoch": 13.386179279186775,
+      "grad_norm": 0.0527302548289299,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 130368
+    },
+    {
+      "epoch": 13.386487319026594,
+      "grad_norm": 0.036185190081596375,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 130371
+    },
+    {
+      "epoch": 13.386795358866413,
+      "grad_norm": 0.032587967813014984,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 130374
+    },
+    {
+      "epoch": 13.387103398706232,
+      "grad_norm": 0.04018452763557434,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 130377
+    },
+    {
+      "epoch": 13.387411438546051,
+      "grad_norm": 0.03417624533176422,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 130380
+    },
+    {
+      "epoch": 13.387719478385872,
+      "grad_norm": 0.07745397090911865,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 130383
+    },
+    {
+      "epoch": 13.388027518225691,
+      "grad_norm": 0.08244820684194565,
+      "learning_rate": 0.01,
+      "loss": 1.8937,
+      "step": 130386
+    },
+    {
+      "epoch": 13.38833555806551,
+      "grad_norm": 0.13346752524375916,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 130389
+    },
+    {
+      "epoch": 13.388643597905329,
+      "grad_norm": 0.08453695476055145,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 130392
+    },
+    {
+      "epoch": 13.388951637745148,
+      "grad_norm": 0.04896337538957596,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 130395
+    },
+    {
+      "epoch": 13.389259677584967,
+      "grad_norm": 0.06542571634054184,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 130398
+    },
+    {
+      "epoch": 13.389567717424788,
+      "grad_norm": 0.03692135959863663,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 130401
+    },
+    {
+      "epoch": 13.389875757264607,
+      "grad_norm": 0.03958454728126526,
+      "learning_rate": 0.01,
+      "loss": 1.8957,
+      "step": 130404
+    },
+    {
+      "epoch": 13.390183797104426,
+      "grad_norm": 0.12414499372243881,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 130407
+    },
+    {
+      "epoch": 13.390491836944244,
+      "grad_norm": 0.06378888338804245,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 130410
+    },
+    {
+      "epoch": 13.390799876784063,
+      "grad_norm": 0.04713134095072746,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 130413
+    },
+    {
+      "epoch": 13.391107916623884,
+      "grad_norm": 0.09070335328578949,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 130416
+    },
+    {
+      "epoch": 13.391415956463703,
+      "grad_norm": 0.0798831433057785,
+      "learning_rate": 0.01,
+      "loss": 1.8888,
+      "step": 130419
+    },
+    {
+      "epoch": 13.391723996303522,
+      "grad_norm": 0.09219987690448761,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 130422
+    },
+    {
+      "epoch": 13.392032036143341,
+      "grad_norm": 0.05276596546173096,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 130425
+    },
+    {
+      "epoch": 13.39234007598316,
+      "grad_norm": 0.05043622851371765,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 130428
+    },
+    {
+      "epoch": 13.392648115822979,
+      "grad_norm": 0.04230506345629692,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 130431
+    },
+    {
+      "epoch": 13.3929561556628,
+      "grad_norm": 0.038519054651260376,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 130434
+    },
+    {
+      "epoch": 13.393264195502619,
+      "grad_norm": 0.06899230927228928,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 130437
+    },
+    {
+      "epoch": 13.393572235342438,
+      "grad_norm": 0.11104927211999893,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 130440
+    },
+    {
+      "epoch": 13.393880275182257,
+      "grad_norm": 0.09579530358314514,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 130443
+    },
+    {
+      "epoch": 13.394188315022076,
+      "grad_norm": 0.030721032992005348,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 130446
+    },
+    {
+      "epoch": 13.394496354861895,
+      "grad_norm": 0.08236239850521088,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 130449
+    },
+    {
+      "epoch": 13.394804394701715,
+      "grad_norm": 0.09433204680681229,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 130452
+    },
+    {
+      "epoch": 13.395112434541534,
+      "grad_norm": 0.051662083715200424,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 130455
+    },
+    {
+      "epoch": 13.395420474381353,
+      "grad_norm": 0.11299832165241241,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 130458
+    },
+    {
+      "epoch": 13.395728514221172,
+      "grad_norm": 0.036634862422943115,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 130461
+    },
+    {
+      "epoch": 13.396036554060991,
+      "grad_norm": 0.05035240948200226,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 130464
+    },
+    {
+      "epoch": 13.39634459390081,
+      "grad_norm": 0.035019855946302414,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 130467
+    },
+    {
+      "epoch": 13.396652633740631,
+      "grad_norm": 0.03242092207074165,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 130470
+    },
+    {
+      "epoch": 13.39696067358045,
+      "grad_norm": 0.047839969396591187,
+      "learning_rate": 0.01,
+      "loss": 1.9471,
+      "step": 130473
+    },
+    {
+      "epoch": 13.397268713420269,
+      "grad_norm": 0.06426669657230377,
+      "learning_rate": 0.01,
+      "loss": 1.9596,
+      "step": 130476
+    },
+    {
+      "epoch": 13.397576753260088,
+      "grad_norm": 0.10376609116792679,
+      "learning_rate": 0.01,
+      "loss": 1.9365,
+      "step": 130479
+    },
+    {
+      "epoch": 13.397884793099907,
+      "grad_norm": 0.05030552297830582,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 130482
+    },
+    {
+      "epoch": 13.398192832939728,
+      "grad_norm": 0.040810927748680115,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 130485
+    },
+    {
+      "epoch": 13.398500872779547,
+      "grad_norm": 0.08123403042554855,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 130488
+    },
+    {
+      "epoch": 13.398808912619366,
+      "grad_norm": 0.058148615062236786,
+      "learning_rate": 0.01,
+      "loss": 1.894,
+      "step": 130491
+    },
+    {
+      "epoch": 13.399116952459185,
+      "grad_norm": 0.047627419233322144,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 130494
+    },
+    {
+      "epoch": 13.399424992299004,
+      "grad_norm": 0.08813631534576416,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 130497
+    },
+    {
+      "epoch": 13.399733032138823,
+      "grad_norm": 0.06861664354801178,
+      "learning_rate": 0.01,
+      "loss": 1.9463,
+      "step": 130500
+    },
+    {
+      "epoch": 13.400041071978643,
+      "grad_norm": 0.051999930292367935,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 130503
+    },
+    {
+      "epoch": 13.400349111818462,
+      "grad_norm": 0.04035176709294319,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 130506
+    },
+    {
+      "epoch": 13.400657151658281,
+      "grad_norm": 0.03260912746191025,
+      "learning_rate": 0.01,
+      "loss": 1.8854,
+      "step": 130509
+    },
+    {
+      "epoch": 13.4009651914981,
+      "grad_norm": 0.03398890048265457,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 130512
+    },
+    {
+      "epoch": 13.40127323133792,
+      "grad_norm": 0.04103786125779152,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 130515
+    },
+    {
+      "epoch": 13.401581271177738,
+      "grad_norm": 0.05711282044649124,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 130518
+    },
+    {
+      "epoch": 13.401889311017559,
+      "grad_norm": 0.04243483394384384,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 130521
+    },
+    {
+      "epoch": 13.402197350857378,
+      "grad_norm": 0.04837220162153244,
+      "learning_rate": 0.01,
+      "loss": 1.9101,
+      "step": 130524
+    },
+    {
+      "epoch": 13.402505390697197,
+      "grad_norm": 0.04088788852095604,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 130527
+    },
+    {
+      "epoch": 13.402813430537016,
+      "grad_norm": 0.12206108868122101,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 130530
+    },
+    {
+      "epoch": 13.403121470376835,
+      "grad_norm": 0.046702057123184204,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 130533
+    },
+    {
+      "epoch": 13.403429510216654,
+      "grad_norm": 0.04278607293963432,
+      "learning_rate": 0.01,
+      "loss": 1.8889,
+      "step": 130536
+    },
+    {
+      "epoch": 13.403737550056475,
+      "grad_norm": 0.08654165267944336,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 130539
+    },
+    {
+      "epoch": 13.404045589896294,
+      "grad_norm": 0.046546537429094315,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 130542
+    },
+    {
+      "epoch": 13.404353629736113,
+      "grad_norm": 0.03562478348612785,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 130545
+    },
+    {
+      "epoch": 13.404661669575932,
+      "grad_norm": 0.05516824498772621,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 130548
+    },
+    {
+      "epoch": 13.40496970941575,
+      "grad_norm": 0.04752029478549957,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 130551
+    },
+    {
+      "epoch": 13.405277749255571,
+      "grad_norm": 0.035238225013017654,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 130554
+    },
+    {
+      "epoch": 13.40558578909539,
+      "grad_norm": 0.1231425553560257,
+      "learning_rate": 0.01,
+      "loss": 1.9525,
+      "step": 130557
+    },
+    {
+      "epoch": 13.40589382893521,
+      "grad_norm": 0.06682516634464264,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 130560
+    },
+    {
+      "epoch": 13.406201868775028,
+      "grad_norm": 0.0957551822066307,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 130563
+    },
+    {
+      "epoch": 13.406509908614847,
+      "grad_norm": 0.06120888143777847,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 130566
+    },
+    {
+      "epoch": 13.406817948454666,
+      "grad_norm": 0.031565889716148376,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 130569
+    },
+    {
+      "epoch": 13.407125988294487,
+      "grad_norm": 0.03393023833632469,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 130572
+    },
+    {
+      "epoch": 13.407434028134306,
+      "grad_norm": 0.17877255380153656,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 130575
+    },
+    {
+      "epoch": 13.407742067974125,
+      "grad_norm": 0.04966241866350174,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 130578
+    },
+    {
+      "epoch": 13.408050107813944,
+      "grad_norm": 0.05966407433152199,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 130581
+    },
+    {
+      "epoch": 13.408358147653763,
+      "grad_norm": 0.04085264727473259,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 130584
+    },
+    {
+      "epoch": 13.408666187493582,
+      "grad_norm": 0.14445793628692627,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 130587
+    },
+    {
+      "epoch": 13.408974227333402,
+      "grad_norm": 0.1389753669500351,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 130590
+    },
+    {
+      "epoch": 13.409282267173221,
+      "grad_norm": 0.10714045912027359,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 130593
+    },
+    {
+      "epoch": 13.40959030701304,
+      "grad_norm": 0.08724230527877808,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 130596
+    },
+    {
+      "epoch": 13.40989834685286,
+      "grad_norm": 0.051676928997039795,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 130599
+    },
+    {
+      "epoch": 13.410206386692678,
+      "grad_norm": 0.07044893503189087,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 130602
+    },
+    {
+      "epoch": 13.410514426532497,
+      "grad_norm": 0.03563698008656502,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 130605
+    },
+    {
+      "epoch": 13.410822466372318,
+      "grad_norm": 0.033698372542858124,
+      "learning_rate": 0.01,
+      "loss": 1.9568,
+      "step": 130608
+    },
+    {
+      "epoch": 13.411130506212137,
+      "grad_norm": 0.03243492543697357,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 130611
+    },
+    {
+      "epoch": 13.411438546051956,
+      "grad_norm": 0.030848274007439613,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 130614
+    },
+    {
+      "epoch": 13.411746585891775,
+      "grad_norm": 0.06016855686903,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 130617
+    },
+    {
+      "epoch": 13.412054625731594,
+      "grad_norm": 0.11460036039352417,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 130620
+    },
+    {
+      "epoch": 13.412362665571415,
+      "grad_norm": 0.10069911926984787,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 130623
+    },
+    {
+      "epoch": 13.412670705411234,
+      "grad_norm": 0.06098891422152519,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 130626
+    },
+    {
+      "epoch": 13.412978745251053,
+      "grad_norm": 0.04587498679757118,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 130629
+    },
+    {
+      "epoch": 13.413286785090872,
+      "grad_norm": 0.0319906622171402,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 130632
+    },
+    {
+      "epoch": 13.41359482493069,
+      "grad_norm": 0.11985224485397339,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 130635
+    },
+    {
+      "epoch": 13.41390286477051,
+      "grad_norm": 0.06811452656984329,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 130638
+    },
+    {
+      "epoch": 13.41421090461033,
+      "grad_norm": 0.040213361382484436,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 130641
+    },
+    {
+      "epoch": 13.41451894445015,
+      "grad_norm": 0.04153140261769295,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 130644
+    },
+    {
+      "epoch": 13.414826984289968,
+      "grad_norm": 0.1312190741300583,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 130647
+    },
+    {
+      "epoch": 13.415135024129787,
+      "grad_norm": 0.14672890305519104,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 130650
+    },
+    {
+      "epoch": 13.415443063969606,
+      "grad_norm": 0.10523177683353424,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 130653
+    },
+    {
+      "epoch": 13.415751103809425,
+      "grad_norm": 0.04652473330497742,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 130656
+    },
+    {
+      "epoch": 13.416059143649246,
+      "grad_norm": 0.0431382842361927,
+      "learning_rate": 0.01,
+      "loss": 1.905,
+      "step": 130659
+    },
+    {
+      "epoch": 13.416367183489065,
+      "grad_norm": 0.0417511984705925,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 130662
+    },
+    {
+      "epoch": 13.416675223328884,
+      "grad_norm": 0.0687151849269867,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 130665
+    },
+    {
+      "epoch": 13.416983263168703,
+      "grad_norm": 0.1430179327726364,
+      "learning_rate": 0.01,
+      "loss": 1.9503,
+      "step": 130668
+    },
+    {
+      "epoch": 13.417291303008522,
+      "grad_norm": 0.12186240404844284,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 130671
+    },
+    {
+      "epoch": 13.41759934284834,
+      "grad_norm": 0.10310307145118713,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 130674
+    },
+    {
+      "epoch": 13.417907382688162,
+      "grad_norm": 0.04326725751161575,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 130677
+    },
+    {
+      "epoch": 13.41821542252798,
+      "grad_norm": 0.029291393235325813,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 130680
+    },
+    {
+      "epoch": 13.4185234623678,
+      "grad_norm": 0.050489816814661026,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 130683
+    },
+    {
+      "epoch": 13.418831502207619,
+      "grad_norm": 0.09828450530767441,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 130686
+    },
+    {
+      "epoch": 13.419139542047438,
+      "grad_norm": 0.08631939440965652,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 130689
+    },
+    {
+      "epoch": 13.419447581887258,
+      "grad_norm": 0.13216805458068848,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 130692
+    },
+    {
+      "epoch": 13.419755621727077,
+      "grad_norm": 0.0639142319560051,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 130695
+    },
+    {
+      "epoch": 13.420063661566896,
+      "grad_norm": 0.05188587307929993,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 130698
+    },
+    {
+      "epoch": 13.420371701406715,
+      "grad_norm": 0.04431859403848648,
+      "learning_rate": 0.01,
+      "loss": 1.9075,
+      "step": 130701
+    },
+    {
+      "epoch": 13.420679741246534,
+      "grad_norm": 0.1297181099653244,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 130704
+    },
+    {
+      "epoch": 13.420987781086353,
+      "grad_norm": 0.07664864510297775,
+      "learning_rate": 0.01,
+      "loss": 1.8802,
+      "step": 130707
+    },
+    {
+      "epoch": 13.421295820926174,
+      "grad_norm": 0.044333674013614655,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 130710
+    },
+    {
+      "epoch": 13.421603860765993,
+      "grad_norm": 0.039914291352033615,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 130713
+    },
+    {
+      "epoch": 13.421911900605812,
+      "grad_norm": 0.055848173797130585,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 130716
+    },
+    {
+      "epoch": 13.42221994044563,
+      "grad_norm": 0.05616588145494461,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 130719
+    },
+    {
+      "epoch": 13.42252798028545,
+      "grad_norm": 0.03985327109694481,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 130722
+    },
+    {
+      "epoch": 13.422836020125269,
+      "grad_norm": 0.11752146482467651,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 130725
+    },
+    {
+      "epoch": 13.42314405996509,
+      "grad_norm": 0.12627892196178436,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 130728
+    },
+    {
+      "epoch": 13.423452099804908,
+      "grad_norm": 0.12760494649410248,
+      "learning_rate": 0.01,
+      "loss": 1.9506,
+      "step": 130731
+    },
+    {
+      "epoch": 13.423760139644727,
+      "grad_norm": 0.09585028886795044,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 130734
+    },
+    {
+      "epoch": 13.424068179484546,
+      "grad_norm": 0.05763910710811615,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 130737
+    },
+    {
+      "epoch": 13.424376219324365,
+      "grad_norm": 0.09867516905069351,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 130740
+    },
+    {
+      "epoch": 13.424684259164184,
+      "grad_norm": 0.07108857482671738,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 130743
+    },
+    {
+      "epoch": 13.424992299004005,
+      "grad_norm": 0.037486642599105835,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 130746
+    },
+    {
+      "epoch": 13.425300338843824,
+      "grad_norm": 0.06318271905183792,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 130749
+    },
+    {
+      "epoch": 13.425608378683643,
+      "grad_norm": 0.03398101404309273,
+      "learning_rate": 0.01,
+      "loss": 1.8984,
+      "step": 130752
+    },
+    {
+      "epoch": 13.425916418523462,
+      "grad_norm": 0.054917238652706146,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 130755
+    },
+    {
+      "epoch": 13.426224458363281,
+      "grad_norm": 0.10470902919769287,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 130758
+    },
+    {
+      "epoch": 13.426532498203102,
+      "grad_norm": 0.04670415818691254,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 130761
+    },
+    {
+      "epoch": 13.42684053804292,
+      "grad_norm": 0.0424685999751091,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 130764
+    },
+    {
+      "epoch": 13.42714857788274,
+      "grad_norm": 0.03525495529174805,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 130767
+    },
+    {
+      "epoch": 13.427456617722559,
+      "grad_norm": 0.04420524835586548,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 130770
+    },
+    {
+      "epoch": 13.427764657562378,
+      "grad_norm": 0.04824223741889,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 130773
+    },
+    {
+      "epoch": 13.428072697402197,
+      "grad_norm": 0.1014878898859024,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 130776
+    },
+    {
+      "epoch": 13.428380737242017,
+      "grad_norm": 0.04712039232254028,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 130779
+    },
+    {
+      "epoch": 13.428688777081836,
+      "grad_norm": 0.04444543272256851,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 130782
+    },
+    {
+      "epoch": 13.428996816921655,
+      "grad_norm": 0.034990157932043076,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 130785
+    },
+    {
+      "epoch": 13.429304856761474,
+      "grad_norm": 0.03937904164195061,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 130788
+    },
+    {
+      "epoch": 13.429612896601293,
+      "grad_norm": 0.046606071293354034,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 130791
+    },
+    {
+      "epoch": 13.429920936441112,
+      "grad_norm": 0.12737707793712616,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 130794
+    },
+    {
+      "epoch": 13.430228976280933,
+      "grad_norm": 0.04323772341012955,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 130797
+    },
+    {
+      "epoch": 13.430537016120752,
+      "grad_norm": 0.06813032180070877,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 130800
+    },
+    {
+      "epoch": 13.430845055960571,
+      "grad_norm": 0.10118822753429413,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 130803
+    },
+    {
+      "epoch": 13.43115309580039,
+      "grad_norm": 0.05706276372075081,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 130806
+    },
+    {
+      "epoch": 13.431461135640209,
+      "grad_norm": 0.041412144899368286,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 130809
+    },
+    {
+      "epoch": 13.431769175480028,
+      "grad_norm": 0.05172746255993843,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 130812
+    },
+    {
+      "epoch": 13.432077215319849,
+      "grad_norm": 0.037874769419431686,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 130815
+    },
+    {
+      "epoch": 13.432385255159668,
+      "grad_norm": 0.032217830419540405,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 130818
+    },
+    {
+      "epoch": 13.432693294999487,
+      "grad_norm": 0.04941815137863159,
+      "learning_rate": 0.01,
+      "loss": 1.8946,
+      "step": 130821
+    },
+    {
+      "epoch": 13.433001334839306,
+      "grad_norm": 0.03862486779689789,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 130824
+    },
+    {
+      "epoch": 13.433309374679125,
+      "grad_norm": 0.03585563227534294,
+      "learning_rate": 0.01,
+      "loss": 1.8934,
+      "step": 130827
+    },
+    {
+      "epoch": 13.433617414518945,
+      "grad_norm": 0.05754021927714348,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 130830
+    },
+    {
+      "epoch": 13.433925454358764,
+      "grad_norm": 0.08251040428876877,
+      "learning_rate": 0.01,
+      "loss": 1.9056,
+      "step": 130833
+    },
+    {
+      "epoch": 13.434233494198583,
+      "grad_norm": 0.07171035557985306,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 130836
+    },
+    {
+      "epoch": 13.434541534038402,
+      "grad_norm": 0.11404532939195633,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 130839
+    },
+    {
+      "epoch": 13.434849573878221,
+      "grad_norm": 0.04019588232040405,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 130842
+    },
+    {
+      "epoch": 13.43515761371804,
+      "grad_norm": 0.031964994966983795,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 130845
+    },
+    {
+      "epoch": 13.435465653557861,
+      "grad_norm": 0.06784484535455704,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 130848
+    },
+    {
+      "epoch": 13.43577369339768,
+      "grad_norm": 0.03703048452734947,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 130851
+    },
+    {
+      "epoch": 13.436081733237499,
+      "grad_norm": 0.13901789486408234,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 130854
+    },
+    {
+      "epoch": 13.436389773077318,
+      "grad_norm": 0.040682103484869,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 130857
+    },
+    {
+      "epoch": 13.436697812917137,
+      "grad_norm": 0.09058588743209839,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 130860
+    },
+    {
+      "epoch": 13.437005852756956,
+      "grad_norm": 0.06873542815446854,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 130863
+    },
+    {
+      "epoch": 13.437313892596777,
+      "grad_norm": 0.06866730749607086,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 130866
+    },
+    {
+      "epoch": 13.437621932436596,
+      "grad_norm": 0.1653352677822113,
+      "learning_rate": 0.01,
+      "loss": 1.9306,
+      "step": 130869
+    },
+    {
+      "epoch": 13.437929972276415,
+      "grad_norm": 0.034187354147434235,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 130872
+    },
+    {
+      "epoch": 13.438238012116233,
+      "grad_norm": 0.039564285427331924,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 130875
+    },
+    {
+      "epoch": 13.438546051956052,
+      "grad_norm": 0.06805476546287537,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 130878
+    },
+    {
+      "epoch": 13.438854091795871,
+      "grad_norm": 0.045071039348840714,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 130881
+    },
+    {
+      "epoch": 13.439162131635692,
+      "grad_norm": 0.060436490923166275,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 130884
+    },
+    {
+      "epoch": 13.439470171475511,
+      "grad_norm": 0.05383118987083435,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 130887
+    },
+    {
+      "epoch": 13.43977821131533,
+      "grad_norm": 0.044863227754831314,
+      "learning_rate": 0.01,
+      "loss": 1.9419,
+      "step": 130890
+    },
+    {
+      "epoch": 13.440086251155149,
+      "grad_norm": 0.04008587449789047,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 130893
+    },
+    {
+      "epoch": 13.440394290994968,
+      "grad_norm": 0.03965401649475098,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 130896
+    },
+    {
+      "epoch": 13.440702330834789,
+      "grad_norm": 0.07382699102163315,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 130899
+    },
+    {
+      "epoch": 13.441010370674608,
+      "grad_norm": 0.07308311015367508,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 130902
+    },
+    {
+      "epoch": 13.441318410514427,
+      "grad_norm": 0.08104528486728668,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 130905
+    },
+    {
+      "epoch": 13.441626450354246,
+      "grad_norm": 0.04624982178211212,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 130908
+    },
+    {
+      "epoch": 13.441934490194065,
+      "grad_norm": 0.07705344259738922,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 130911
+    },
+    {
+      "epoch": 13.442242530033884,
+      "grad_norm": 0.1591053605079651,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 130914
+    },
+    {
+      "epoch": 13.442550569873704,
+      "grad_norm": 0.08124366402626038,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 130917
+    },
+    {
+      "epoch": 13.442858609713523,
+      "grad_norm": 0.07179301977157593,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 130920
+    },
+    {
+      "epoch": 13.443166649553342,
+      "grad_norm": 0.07564125210046768,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 130923
+    },
+    {
+      "epoch": 13.443474689393161,
+      "grad_norm": 0.07575230300426483,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 130926
+    },
+    {
+      "epoch": 13.44378272923298,
+      "grad_norm": 0.061750590801239014,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 130929
+    },
+    {
+      "epoch": 13.4440907690728,
+      "grad_norm": 0.04988943785429001,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 130932
+    },
+    {
+      "epoch": 13.44439880891262,
+      "grad_norm": 0.07663754373788834,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 130935
+    },
+    {
+      "epoch": 13.444706848752439,
+      "grad_norm": 0.07344890385866165,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 130938
+    },
+    {
+      "epoch": 13.445014888592258,
+      "grad_norm": 0.05747510865330696,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 130941
+    },
+    {
+      "epoch": 13.445322928432077,
+      "grad_norm": 0.04004649445414543,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 130944
+    },
+    {
+      "epoch": 13.445630968271896,
+      "grad_norm": 0.04760877788066864,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 130947
+    },
+    {
+      "epoch": 13.445939008111715,
+      "grad_norm": 0.06519803404808044,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 130950
+    },
+    {
+      "epoch": 13.446247047951536,
+      "grad_norm": 0.07844406366348267,
+      "learning_rate": 0.01,
+      "loss": 1.9382,
+      "step": 130953
+    },
+    {
+      "epoch": 13.446555087791355,
+      "grad_norm": 0.05578390508890152,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 130956
+    },
+    {
+      "epoch": 13.446863127631174,
+      "grad_norm": 0.052308082580566406,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 130959
+    },
+    {
+      "epoch": 13.447171167470993,
+      "grad_norm": 0.11089842021465302,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 130962
+    },
+    {
+      "epoch": 13.447479207310812,
+      "grad_norm": 0.09027403593063354,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 130965
+    },
+    {
+      "epoch": 13.447787247150632,
+      "grad_norm": 0.10414513200521469,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 130968
+    },
+    {
+      "epoch": 13.448095286990451,
+      "grad_norm": 0.0399368554353714,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 130971
+    },
+    {
+      "epoch": 13.44840332683027,
+      "grad_norm": 0.11621847003698349,
+      "learning_rate": 0.01,
+      "loss": 1.9579,
+      "step": 130974
+    },
+    {
+      "epoch": 13.44871136667009,
+      "grad_norm": 0.04326711595058441,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 130977
+    },
+    {
+      "epoch": 13.449019406509908,
+      "grad_norm": 0.038497406989336014,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 130980
+    },
+    {
+      "epoch": 13.449327446349727,
+      "grad_norm": 0.04257148876786232,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 130983
+    },
+    {
+      "epoch": 13.449635486189548,
+      "grad_norm": 0.04650953412055969,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 130986
+    },
+    {
+      "epoch": 13.449943526029367,
+      "grad_norm": 0.033275306224823,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 130989
+    },
+    {
+      "epoch": 13.450251565869186,
+      "grad_norm": 0.033621493726968765,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 130992
+    },
+    {
+      "epoch": 13.450559605709005,
+      "grad_norm": 0.04494462162256241,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 130995
+    },
+    {
+      "epoch": 13.450867645548824,
+      "grad_norm": 0.06400339305400848,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 130998
+    },
+    {
+      "epoch": 13.451175685388643,
+      "grad_norm": 0.06292519718408585,
+      "learning_rate": 0.01,
+      "loss": 1.9055,
+      "step": 131001
+    },
+    {
+      "epoch": 13.451483725228464,
+      "grad_norm": 0.11436935514211655,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 131004
+    },
+    {
+      "epoch": 13.451791765068283,
+      "grad_norm": 0.08692870289087296,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 131007
+    },
+    {
+      "epoch": 13.452099804908102,
+      "grad_norm": 0.07595816254615784,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 131010
+    },
+    {
+      "epoch": 13.45240784474792,
+      "grad_norm": 0.03356483578681946,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 131013
+    },
+    {
+      "epoch": 13.45271588458774,
+      "grad_norm": 0.04597342014312744,
+      "learning_rate": 0.01,
+      "loss": 1.9532,
+      "step": 131016
+    },
+    {
+      "epoch": 13.453023924427558,
+      "grad_norm": 0.042660098522901535,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 131019
+    },
+    {
+      "epoch": 13.45333196426738,
+      "grad_norm": 0.06123415008187294,
+      "learning_rate": 0.01,
+      "loss": 1.9355,
+      "step": 131022
+    },
+    {
+      "epoch": 13.453640004107198,
+      "grad_norm": 0.10522007942199707,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 131025
+    },
+    {
+      "epoch": 13.453948043947017,
+      "grad_norm": 0.061230067163705826,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 131028
+    },
+    {
+      "epoch": 13.454256083786836,
+      "grad_norm": 0.041935041546821594,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 131031
+    },
+    {
+      "epoch": 13.454564123626655,
+      "grad_norm": 0.10501331835985184,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 131034
+    },
+    {
+      "epoch": 13.454872163466476,
+      "grad_norm": 0.054169557988643646,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 131037
+    },
+    {
+      "epoch": 13.455180203306295,
+      "grad_norm": 0.10532642155885696,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 131040
+    },
+    {
+      "epoch": 13.455488243146114,
+      "grad_norm": 0.05273296684026718,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 131043
+    },
+    {
+      "epoch": 13.455796282985933,
+      "grad_norm": 0.03350077196955681,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 131046
+    },
+    {
+      "epoch": 13.456104322825752,
+      "grad_norm": 0.11782905459403992,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 131049
+    },
+    {
+      "epoch": 13.45641236266557,
+      "grad_norm": 0.03702506422996521,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 131052
+    },
+    {
+      "epoch": 13.456720402505391,
+      "grad_norm": 0.08643093705177307,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 131055
+    },
+    {
+      "epoch": 13.45702844234521,
+      "grad_norm": 0.08253560215234756,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 131058
+    },
+    {
+      "epoch": 13.45733648218503,
+      "grad_norm": 0.0710495188832283,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 131061
+    },
+    {
+      "epoch": 13.457644522024848,
+      "grad_norm": 0.04020102694630623,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 131064
+    },
+    {
+      "epoch": 13.457952561864667,
+      "grad_norm": 0.03451429307460785,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 131067
+    },
+    {
+      "epoch": 13.458260601704486,
+      "grad_norm": 0.05185393616557121,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 131070
+    },
+    {
+      "epoch": 13.458568641544307,
+      "grad_norm": 0.06879854202270508,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 131073
+    },
+    {
+      "epoch": 13.458876681384126,
+      "grad_norm": 0.0692383274435997,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 131076
+    },
+    {
+      "epoch": 13.459184721223945,
+      "grad_norm": 0.07925112545490265,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 131079
+    },
+    {
+      "epoch": 13.459492761063764,
+      "grad_norm": 0.09783706068992615,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 131082
+    },
+    {
+      "epoch": 13.459800800903583,
+      "grad_norm": 0.056679386645555496,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 131085
+    },
+    {
+      "epoch": 13.460108840743402,
+      "grad_norm": 0.0796128362417221,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 131088
+    },
+    {
+      "epoch": 13.460416880583223,
+      "grad_norm": 0.03908248245716095,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 131091
+    },
+    {
+      "epoch": 13.460724920423042,
+      "grad_norm": 0.1095680519938469,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 131094
+    },
+    {
+      "epoch": 13.46103296026286,
+      "grad_norm": 0.04333530738949776,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 131097
+    },
+    {
+      "epoch": 13.46134100010268,
+      "grad_norm": 0.08107000589370728,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 131100
+    },
+    {
+      "epoch": 13.461649039942499,
+      "grad_norm": 0.1015387624502182,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 131103
+    },
+    {
+      "epoch": 13.46195707978232,
+      "grad_norm": 0.13770507276058197,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 131106
+    },
+    {
+      "epoch": 13.462265119622138,
+      "grad_norm": 0.058807794004678726,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 131109
+    },
+    {
+      "epoch": 13.462573159461957,
+      "grad_norm": 0.04877718538045883,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 131112
+    },
+    {
+      "epoch": 13.462881199301776,
+      "grad_norm": 0.06844967603683472,
+      "learning_rate": 0.01,
+      "loss": 1.8877,
+      "step": 131115
+    },
+    {
+      "epoch": 13.463189239141595,
+      "grad_norm": 0.05854065343737602,
+      "learning_rate": 0.01,
+      "loss": 1.9461,
+      "step": 131118
+    },
+    {
+      "epoch": 13.463497278981414,
+      "grad_norm": 0.08441576361656189,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 131121
+    },
+    {
+      "epoch": 13.463805318821235,
+      "grad_norm": 0.04342430457472801,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 131124
+    },
+    {
+      "epoch": 13.464113358661054,
+      "grad_norm": 0.08782751113176346,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 131127
+    },
+    {
+      "epoch": 13.464421398500873,
+      "grad_norm": 0.15467555820941925,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 131130
+    },
+    {
+      "epoch": 13.464729438340692,
+      "grad_norm": 0.04931076616048813,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 131133
+    },
+    {
+      "epoch": 13.46503747818051,
+      "grad_norm": 0.07662250101566315,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 131136
+    },
+    {
+      "epoch": 13.46534551802033,
+      "grad_norm": 0.07537400722503662,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 131139
+    },
+    {
+      "epoch": 13.46565355786015,
+      "grad_norm": 0.08145980536937714,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 131142
+    },
+    {
+      "epoch": 13.46596159769997,
+      "grad_norm": 0.06405560672283173,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 131145
+    },
+    {
+      "epoch": 13.466269637539789,
+      "grad_norm": 0.051541175693273544,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 131148
+    },
+    {
+      "epoch": 13.466577677379608,
+      "grad_norm": 0.059968702495098114,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 131151
+    },
+    {
+      "epoch": 13.466885717219427,
+      "grad_norm": 0.06381983309984207,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 131154
+    },
+    {
+      "epoch": 13.467193757059245,
+      "grad_norm": 0.04236496984958649,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 131157
+    },
+    {
+      "epoch": 13.467501796899066,
+      "grad_norm": 0.04334869608283043,
+      "learning_rate": 0.01,
+      "loss": 1.9418,
+      "step": 131160
+    },
+    {
+      "epoch": 13.467809836738885,
+      "grad_norm": 0.10239947587251663,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 131163
+    },
+    {
+      "epoch": 13.468117876578704,
+      "grad_norm": 0.03539574518799782,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 131166
+    },
+    {
+      "epoch": 13.468425916418523,
+      "grad_norm": 0.11559099704027176,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 131169
+    },
+    {
+      "epoch": 13.468733956258342,
+      "grad_norm": 0.11699347198009491,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 131172
+    },
+    {
+      "epoch": 13.469041996098163,
+      "grad_norm": 0.04169037938117981,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 131175
+    },
+    {
+      "epoch": 13.469350035937982,
+      "grad_norm": 0.06891239434480667,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 131178
+    },
+    {
+      "epoch": 13.4696580757778,
+      "grad_norm": 0.06007331609725952,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 131181
+    },
+    {
+      "epoch": 13.46996611561762,
+      "grad_norm": 0.05303988605737686,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 131184
+    },
+    {
+      "epoch": 13.470274155457439,
+      "grad_norm": 0.05364450067281723,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 131187
+    },
+    {
+      "epoch": 13.470582195297258,
+      "grad_norm": 0.11672054976224899,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 131190
+    },
+    {
+      "epoch": 13.470890235137079,
+      "grad_norm": 0.054327502846717834,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 131193
+    },
+    {
+      "epoch": 13.471198274976897,
+      "grad_norm": 0.10961230844259262,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 131196
+    },
+    {
+      "epoch": 13.471506314816716,
+      "grad_norm": 0.03663426265120506,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 131199
+    },
+    {
+      "epoch": 13.471814354656535,
+      "grad_norm": 0.14347374439239502,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 131202
+    },
+    {
+      "epoch": 13.472122394496354,
+      "grad_norm": 0.14460596442222595,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 131205
+    },
+    {
+      "epoch": 13.472430434336173,
+      "grad_norm": 0.045124586671590805,
+      "learning_rate": 0.01,
+      "loss": 1.9241,
+      "step": 131208
+    },
+    {
+      "epoch": 13.472738474175994,
+      "grad_norm": 0.0590360052883625,
+      "learning_rate": 0.01,
+      "loss": 1.8942,
+      "step": 131211
+    },
+    {
+      "epoch": 13.473046514015813,
+      "grad_norm": 0.07320775836706161,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 131214
+    },
+    {
+      "epoch": 13.473354553855632,
+      "grad_norm": 0.0473875030875206,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 131217
+    },
+    {
+      "epoch": 13.473662593695451,
+      "grad_norm": 0.04100793972611427,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 131220
+    },
+    {
+      "epoch": 13.47397063353527,
+      "grad_norm": 0.059964463114738464,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 131223
+    },
+    {
+      "epoch": 13.474278673375089,
+      "grad_norm": 0.04625070095062256,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 131226
+    },
+    {
+      "epoch": 13.47458671321491,
+      "grad_norm": 0.10319223999977112,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 131229
+    },
+    {
+      "epoch": 13.474894753054729,
+      "grad_norm": 0.10196827352046967,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 131232
+    },
+    {
+      "epoch": 13.475202792894548,
+      "grad_norm": 0.05462782457470894,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 131235
+    },
+    {
+      "epoch": 13.475510832734367,
+      "grad_norm": 0.07545313984155655,
+      "learning_rate": 0.01,
+      "loss": 1.8902,
+      "step": 131238
+    },
+    {
+      "epoch": 13.475818872574186,
+      "grad_norm": 0.0734245628118515,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 131241
+    },
+    {
+      "epoch": 13.476126912414006,
+      "grad_norm": 0.1037333607673645,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 131244
+    },
+    {
+      "epoch": 13.476434952253825,
+      "grad_norm": 0.10400739312171936,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 131247
+    },
+    {
+      "epoch": 13.476742992093644,
+      "grad_norm": 0.03564969822764397,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 131250
+    },
+    {
+      "epoch": 13.477051031933463,
+      "grad_norm": 0.06130120903253555,
+      "learning_rate": 0.01,
+      "loss": 1.9326,
+      "step": 131253
+    },
+    {
+      "epoch": 13.477359071773282,
+      "grad_norm": 0.052632056176662445,
+      "learning_rate": 0.01,
+      "loss": 1.8814,
+      "step": 131256
+    },
+    {
+      "epoch": 13.477667111613101,
+      "grad_norm": 0.04590677469968796,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 131259
+    },
+    {
+      "epoch": 13.477975151452922,
+      "grad_norm": 0.06794480234384537,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 131262
+    },
+    {
+      "epoch": 13.478283191292741,
+      "grad_norm": 0.03867311775684357,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 131265
+    },
+    {
+      "epoch": 13.47859123113256,
+      "grad_norm": 0.04257632791996002,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 131268
+    },
+    {
+      "epoch": 13.478899270972379,
+      "grad_norm": 0.04161803424358368,
+      "learning_rate": 0.01,
+      "loss": 1.8941,
+      "step": 131271
+    },
+    {
+      "epoch": 13.479207310812198,
+      "grad_norm": 0.04685438796877861,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 131274
+    },
+    {
+      "epoch": 13.479515350652017,
+      "grad_norm": 0.09962277859449387,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 131277
+    },
+    {
+      "epoch": 13.479823390491838,
+      "grad_norm": 0.04590221494436264,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 131280
+    },
+    {
+      "epoch": 13.480131430331657,
+      "grad_norm": 0.12976299226284027,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 131283
+    },
+    {
+      "epoch": 13.480439470171476,
+      "grad_norm": 0.1019272729754448,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 131286
+    },
+    {
+      "epoch": 13.480747510011295,
+      "grad_norm": 0.05087721347808838,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 131289
+    },
+    {
+      "epoch": 13.481055549851114,
+      "grad_norm": 0.04486459866166115,
+      "learning_rate": 0.01,
+      "loss": 1.9459,
+      "step": 131292
+    },
+    {
+      "epoch": 13.481363589690933,
+      "grad_norm": 0.05310864374041557,
+      "learning_rate": 0.01,
+      "loss": 1.8971,
+      "step": 131295
+    },
+    {
+      "epoch": 13.481671629530753,
+      "grad_norm": 0.042207080870866776,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 131298
+    },
+    {
+      "epoch": 13.481979669370572,
+      "grad_norm": 0.0481526181101799,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 131301
+    },
+    {
+      "epoch": 13.482287709210391,
+      "grad_norm": 0.03222447633743286,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 131304
+    },
+    {
+      "epoch": 13.48259574905021,
+      "grad_norm": 0.044582564383745193,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 131307
+    },
+    {
+      "epoch": 13.48290378889003,
+      "grad_norm": 0.036356616765260696,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 131310
+    },
+    {
+      "epoch": 13.48321182872985,
+      "grad_norm": 0.07297059148550034,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 131313
+    },
+    {
+      "epoch": 13.483519868569669,
+      "grad_norm": 0.06271197646856308,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 131316
+    },
+    {
+      "epoch": 13.483827908409488,
+      "grad_norm": 0.1310107409954071,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 131319
+    },
+    {
+      "epoch": 13.484135948249307,
+      "grad_norm": 0.04788191244006157,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 131322
+    },
+    {
+      "epoch": 13.484443988089126,
+      "grad_norm": 0.04402579367160797,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 131325
+    },
+    {
+      "epoch": 13.484752027928945,
+      "grad_norm": 0.09526830166578293,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 131328
+    },
+    {
+      "epoch": 13.485060067768766,
+      "grad_norm": 0.04021294414997101,
+      "learning_rate": 0.01,
+      "loss": 1.8964,
+      "step": 131331
+    },
+    {
+      "epoch": 13.485368107608585,
+      "grad_norm": 0.13812986016273499,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 131334
+    },
+    {
+      "epoch": 13.485676147448403,
+      "grad_norm": 0.1184663325548172,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 131337
+    },
+    {
+      "epoch": 13.485984187288222,
+      "grad_norm": 0.04899342730641365,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 131340
+    },
+    {
+      "epoch": 13.486292227128041,
+      "grad_norm": 0.039041101932525635,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 131343
+    },
+    {
+      "epoch": 13.48660026696786,
+      "grad_norm": 0.03660248965024948,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 131346
+    },
+    {
+      "epoch": 13.486908306807681,
+      "grad_norm": 0.05119164660573006,
+      "learning_rate": 0.01,
+      "loss": 1.8924,
+      "step": 131349
+    },
+    {
+      "epoch": 13.4872163466475,
+      "grad_norm": 0.057268720120191574,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 131352
+    },
+    {
+      "epoch": 13.487524386487319,
+      "grad_norm": 0.060897551476955414,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 131355
+    },
+    {
+      "epoch": 13.487832426327138,
+      "grad_norm": 0.08340750634670258,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 131358
+    },
+    {
+      "epoch": 13.488140466166957,
+      "grad_norm": 0.03964683413505554,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 131361
+    },
+    {
+      "epoch": 13.488448506006776,
+      "grad_norm": 0.0516141876578331,
+      "learning_rate": 0.01,
+      "loss": 1.8982,
+      "step": 131364
+    },
+    {
+      "epoch": 13.488756545846597,
+      "grad_norm": 0.057934168726205826,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 131367
+    },
+    {
+      "epoch": 13.489064585686416,
+      "grad_norm": 0.045367445796728134,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 131370
+    },
+    {
+      "epoch": 13.489372625526235,
+      "grad_norm": 0.041196905076503754,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 131373
+    },
+    {
+      "epoch": 13.489680665366054,
+      "grad_norm": 0.07399606704711914,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 131376
+    },
+    {
+      "epoch": 13.489988705205873,
+      "grad_norm": 0.12480009347200394,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 131379
+    },
+    {
+      "epoch": 13.490296745045693,
+      "grad_norm": 0.07524775713682175,
+      "learning_rate": 0.01,
+      "loss": 1.9053,
+      "step": 131382
+    },
+    {
+      "epoch": 13.490604784885512,
+      "grad_norm": 0.0441722646355629,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 131385
+    },
+    {
+      "epoch": 13.490912824725331,
+      "grad_norm": 0.04119295999407768,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 131388
+    },
+    {
+      "epoch": 13.49122086456515,
+      "grad_norm": 0.051573872566223145,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 131391
+    },
+    {
+      "epoch": 13.49152890440497,
+      "grad_norm": 0.13969379663467407,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 131394
+    },
+    {
+      "epoch": 13.491836944244788,
+      "grad_norm": 0.12826821208000183,
+      "learning_rate": 0.01,
+      "loss": 1.9053,
+      "step": 131397
+    },
+    {
+      "epoch": 13.492144984084609,
+      "grad_norm": 0.04658589884638786,
+      "learning_rate": 0.01,
+      "loss": 1.9477,
+      "step": 131400
+    },
+    {
+      "epoch": 13.492453023924428,
+      "grad_norm": 0.033303387463092804,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 131403
+    },
+    {
+      "epoch": 13.492761063764247,
+      "grad_norm": 0.04370022565126419,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 131406
+    },
+    {
+      "epoch": 13.493069103604066,
+      "grad_norm": 0.08708381652832031,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 131409
+    },
+    {
+      "epoch": 13.493377143443885,
+      "grad_norm": 0.06696190685033798,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 131412
+    },
+    {
+      "epoch": 13.493685183283704,
+      "grad_norm": 0.06646513938903809,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 131415
+    },
+    {
+      "epoch": 13.493993223123525,
+      "grad_norm": 0.07811851054430008,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 131418
+    },
+    {
+      "epoch": 13.494301262963344,
+      "grad_norm": 0.08806771785020828,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 131421
+    },
+    {
+      "epoch": 13.494609302803163,
+      "grad_norm": 0.08393225818872452,
+      "learning_rate": 0.01,
+      "loss": 1.8884,
+      "step": 131424
+    },
+    {
+      "epoch": 13.494917342642982,
+      "grad_norm": 0.05147898569703102,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 131427
+    },
+    {
+      "epoch": 13.4952253824828,
+      "grad_norm": 0.09695684909820557,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 131430
+    },
+    {
+      "epoch": 13.49553342232262,
+      "grad_norm": 0.07212649285793304,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 131433
+    },
+    {
+      "epoch": 13.49584146216244,
+      "grad_norm": 0.08241216093301773,
+      "learning_rate": 0.01,
+      "loss": 1.9523,
+      "step": 131436
+    },
+    {
+      "epoch": 13.49614950200226,
+      "grad_norm": 0.06454349309206009,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 131439
+    },
+    {
+      "epoch": 13.496457541842078,
+      "grad_norm": 0.06370728462934494,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 131442
+    },
+    {
+      "epoch": 13.496765581681897,
+      "grad_norm": 0.09012583643198013,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 131445
+    },
+    {
+      "epoch": 13.497073621521716,
+      "grad_norm": 0.07756973803043365,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 131448
+    },
+    {
+      "epoch": 13.497381661361537,
+      "grad_norm": 0.09227123111486435,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 131451
+    },
+    {
+      "epoch": 13.497689701201356,
+      "grad_norm": 0.04524921625852585,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 131454
+    },
+    {
+      "epoch": 13.497997741041175,
+      "grad_norm": 0.05264909937977791,
+      "learning_rate": 0.01,
+      "loss": 1.8997,
+      "step": 131457
+    },
+    {
+      "epoch": 13.498305780880994,
+      "grad_norm": 0.046017859131097794,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 131460
+    },
+    {
+      "epoch": 13.498613820720813,
+      "grad_norm": 0.05455230548977852,
+      "learning_rate": 0.01,
+      "loss": 1.8961,
+      "step": 131463
+    },
+    {
+      "epoch": 13.498921860560632,
+      "grad_norm": 0.05391388759016991,
+      "learning_rate": 0.01,
+      "loss": 1.9173,
+      "step": 131466
+    },
+    {
+      "epoch": 13.499229900400453,
+      "grad_norm": 0.07575694471597672,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 131469
+    },
+    {
+      "epoch": 13.499537940240272,
+      "grad_norm": 0.04195409268140793,
+      "learning_rate": 0.01,
+      "loss": 1.939,
+      "step": 131472
+    },
+    {
+      "epoch": 13.49984598008009,
+      "grad_norm": 0.06630894541740417,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 131475
+    },
+    {
+      "epoch": 13.50015401991991,
+      "grad_norm": 0.05323990806937218,
+      "learning_rate": 0.01,
+      "loss": 1.9101,
+      "step": 131478
+    },
+    {
+      "epoch": 13.500462059759728,
+      "grad_norm": 0.09842026978731155,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 131481
+    },
+    {
+      "epoch": 13.500770099599547,
+      "grad_norm": 0.07389501482248306,
+      "learning_rate": 0.01,
+      "loss": 1.8982,
+      "step": 131484
+    },
+    {
+      "epoch": 13.501078139439368,
+      "grad_norm": 0.09694462269544601,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 131487
+    },
+    {
+      "epoch": 13.501386179279187,
+      "grad_norm": 0.10644526779651642,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 131490
+    },
+    {
+      "epoch": 13.501694219119006,
+      "grad_norm": 0.04853412136435509,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 131493
+    },
+    {
+      "epoch": 13.502002258958825,
+      "grad_norm": 0.038569360971450806,
+      "learning_rate": 0.01,
+      "loss": 1.9606,
+      "step": 131496
+    },
+    {
+      "epoch": 13.502310298798644,
+      "grad_norm": 0.040334656834602356,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 131499
+    },
+    {
+      "epoch": 13.502618338638463,
+      "grad_norm": 0.052357252687215805,
+      "learning_rate": 0.01,
+      "loss": 1.8886,
+      "step": 131502
+    },
+    {
+      "epoch": 13.502926378478284,
+      "grad_norm": 0.038757823407649994,
+      "learning_rate": 0.01,
+      "loss": 1.8839,
+      "step": 131505
+    },
+    {
+      "epoch": 13.503234418318103,
+      "grad_norm": 0.03375411778688431,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 131508
+    },
+    {
+      "epoch": 13.503542458157922,
+      "grad_norm": 0.06393110752105713,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 131511
+    },
+    {
+      "epoch": 13.50385049799774,
+      "grad_norm": 0.09397508203983307,
+      "learning_rate": 0.01,
+      "loss": 1.9407,
+      "step": 131514
+    },
+    {
+      "epoch": 13.50415853783756,
+      "grad_norm": 0.06468804180622101,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 131517
+    },
+    {
+      "epoch": 13.50446657767738,
+      "grad_norm": 0.06706315279006958,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 131520
+    },
+    {
+      "epoch": 13.5047746175172,
+      "grad_norm": 0.11053914576768875,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 131523
+    },
+    {
+      "epoch": 13.505082657357018,
+      "grad_norm": 0.06824685633182526,
+      "learning_rate": 0.01,
+      "loss": 1.8946,
+      "step": 131526
+    },
+    {
+      "epoch": 13.505390697196837,
+      "grad_norm": 0.08001524955034256,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 131529
+    },
+    {
+      "epoch": 13.505698737036656,
+      "grad_norm": 0.07275814563035965,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 131532
+    },
+    {
+      "epoch": 13.506006776876475,
+      "grad_norm": 0.061545051634311676,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 131535
+    },
+    {
+      "epoch": 13.506314816716296,
+      "grad_norm": 0.08902882784605026,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 131538
+    },
+    {
+      "epoch": 13.506622856556115,
+      "grad_norm": 0.07351232320070267,
+      "learning_rate": 0.01,
+      "loss": 1.8999,
+      "step": 131541
+    },
+    {
+      "epoch": 13.506930896395934,
+      "grad_norm": 0.08149990439414978,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 131544
+    },
+    {
+      "epoch": 13.507238936235753,
+      "grad_norm": 0.0546073280274868,
+      "learning_rate": 0.01,
+      "loss": 1.8964,
+      "step": 131547
+    },
+    {
+      "epoch": 13.507546976075572,
+      "grad_norm": 0.13041874766349792,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 131550
+    },
+    {
+      "epoch": 13.507855015915391,
+      "grad_norm": 0.1575966775417328,
+      "learning_rate": 0.01,
+      "loss": 1.9456,
+      "step": 131553
+    },
+    {
+      "epoch": 13.508163055755212,
+      "grad_norm": 0.10986176133155823,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 131556
+    },
+    {
+      "epoch": 13.50847109559503,
+      "grad_norm": 0.1568707525730133,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 131559
+    },
+    {
+      "epoch": 13.50877913543485,
+      "grad_norm": 0.07604028284549713,
+      "learning_rate": 0.01,
+      "loss": 1.9035,
+      "step": 131562
+    },
+    {
+      "epoch": 13.509087175274669,
+      "grad_norm": 0.17703938484191895,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 131565
+    },
+    {
+      "epoch": 13.509395215114488,
+      "grad_norm": 0.07373291254043579,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 131568
+    },
+    {
+      "epoch": 13.509703254954307,
+      "grad_norm": 0.043240737169981,
+      "learning_rate": 0.01,
+      "loss": 1.9494,
+      "step": 131571
+    },
+    {
+      "epoch": 13.510011294794127,
+      "grad_norm": 0.056581102311611176,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 131574
+    },
+    {
+      "epoch": 13.510319334633946,
+      "grad_norm": 0.056114550679922104,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 131577
+    },
+    {
+      "epoch": 13.510627374473765,
+      "grad_norm": 0.05155659466981888,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 131580
+    },
+    {
+      "epoch": 13.510935414313584,
+      "grad_norm": 0.051841139793395996,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 131583
+    },
+    {
+      "epoch": 13.511243454153403,
+      "grad_norm": 0.04535001516342163,
+      "learning_rate": 0.01,
+      "loss": 1.9056,
+      "step": 131586
+    },
+    {
+      "epoch": 13.511551493993224,
+      "grad_norm": 0.0388740599155426,
+      "learning_rate": 0.01,
+      "loss": 1.8833,
+      "step": 131589
+    },
+    {
+      "epoch": 13.511859533833043,
+      "grad_norm": 0.05304255709052086,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 131592
+    },
+    {
+      "epoch": 13.512167573672862,
+      "grad_norm": 0.07619743049144745,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 131595
+    },
+    {
+      "epoch": 13.512475613512681,
+      "grad_norm": 0.09544098377227783,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 131598
+    },
+    {
+      "epoch": 13.5127836533525,
+      "grad_norm": 0.041321638971567154,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 131601
+    },
+    {
+      "epoch": 13.513091693192319,
+      "grad_norm": 0.12591961026191711,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 131604
+    },
+    {
+      "epoch": 13.51339973303214,
+      "grad_norm": 0.057291533797979355,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 131607
+    },
+    {
+      "epoch": 13.513707772871959,
+      "grad_norm": 0.038587216287851334,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 131610
+    },
+    {
+      "epoch": 13.514015812711778,
+      "grad_norm": 0.057038769125938416,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 131613
+    },
+    {
+      "epoch": 13.514323852551597,
+      "grad_norm": 0.07348461449146271,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 131616
+    },
+    {
+      "epoch": 13.514631892391415,
+      "grad_norm": 0.03545389324426651,
+      "learning_rate": 0.01,
+      "loss": 1.8922,
+      "step": 131619
+    },
+    {
+      "epoch": 13.514939932231234,
+      "grad_norm": 0.0512729287147522,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 131622
+    },
+    {
+      "epoch": 13.515247972071055,
+      "grad_norm": 0.03339044749736786,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 131625
+    },
+    {
+      "epoch": 13.515556011910874,
+      "grad_norm": 0.04436996951699257,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 131628
+    },
+    {
+      "epoch": 13.515864051750693,
+      "grad_norm": 0.055637381970882416,
+      "learning_rate": 0.01,
+      "loss": 1.9397,
+      "step": 131631
+    },
+    {
+      "epoch": 13.516172091590512,
+      "grad_norm": 0.04532844200730324,
+      "learning_rate": 0.01,
+      "loss": 1.8958,
+      "step": 131634
+    },
+    {
+      "epoch": 13.516480131430331,
+      "grad_norm": 0.04437865689396858,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 131637
+    },
+    {
+      "epoch": 13.51678817127015,
+      "grad_norm": 0.05323813483119011,
+      "learning_rate": 0.01,
+      "loss": 1.8972,
+      "step": 131640
+    },
+    {
+      "epoch": 13.51709621110997,
+      "grad_norm": 0.07939588278532028,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 131643
+    },
+    {
+      "epoch": 13.51740425094979,
+      "grad_norm": 0.05502822622656822,
+      "learning_rate": 0.01,
+      "loss": 1.8887,
+      "step": 131646
+    },
+    {
+      "epoch": 13.517712290789609,
+      "grad_norm": 0.09964946657419205,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 131649
+    },
+    {
+      "epoch": 13.518020330629428,
+      "grad_norm": 0.06511101871728897,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 131652
+    },
+    {
+      "epoch": 13.518328370469247,
+      "grad_norm": 0.0618147999048233,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 131655
+    },
+    {
+      "epoch": 13.518636410309067,
+      "grad_norm": 0.05449791997671127,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 131658
+    },
+    {
+      "epoch": 13.518944450148886,
+      "grad_norm": 0.11612183600664139,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 131661
+    },
+    {
+      "epoch": 13.519252489988705,
+      "grad_norm": 0.04080687090754509,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 131664
+    },
+    {
+      "epoch": 13.519560529828524,
+      "grad_norm": 0.08392549306154251,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 131667
+    },
+    {
+      "epoch": 13.519868569668343,
+      "grad_norm": 0.09661354124546051,
+      "learning_rate": 0.01,
+      "loss": 1.9395,
+      "step": 131670
+    },
+    {
+      "epoch": 13.520176609508162,
+      "grad_norm": 0.07603606581687927,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 131673
+    },
+    {
+      "epoch": 13.520484649347983,
+      "grad_norm": 0.034152474254369736,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 131676
+    },
+    {
+      "epoch": 13.520792689187802,
+      "grad_norm": 0.03270428627729416,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 131679
+    },
+    {
+      "epoch": 13.521100729027621,
+      "grad_norm": 0.10085827857255936,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 131682
+    },
+    {
+      "epoch": 13.52140876886744,
+      "grad_norm": 0.052235618233680725,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 131685
+    },
+    {
+      "epoch": 13.521716808707259,
+      "grad_norm": 0.10822483152151108,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 131688
+    },
+    {
+      "epoch": 13.522024848547078,
+      "grad_norm": 0.12296050786972046,
+      "learning_rate": 0.01,
+      "loss": 1.9411,
+      "step": 131691
+    },
+    {
+      "epoch": 13.522332888386899,
+      "grad_norm": 0.08823835104703903,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 131694
+    },
+    {
+      "epoch": 13.522640928226718,
+      "grad_norm": 0.10312031954526901,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 131697
+    },
+    {
+      "epoch": 13.522948968066537,
+      "grad_norm": 0.03586811572313309,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 131700
+    },
+    {
+      "epoch": 13.523257007906356,
+      "grad_norm": 0.03647251054644585,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 131703
+    },
+    {
+      "epoch": 13.523565047746175,
+      "grad_norm": 0.05003335699439049,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 131706
+    },
+    {
+      "epoch": 13.523873087585994,
+      "grad_norm": 0.03722973167896271,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 131709
+    },
+    {
+      "epoch": 13.524181127425814,
+      "grad_norm": 0.09022004902362823,
+      "learning_rate": 0.01,
+      "loss": 1.9083,
+      "step": 131712
+    },
+    {
+      "epoch": 13.524489167265633,
+      "grad_norm": 0.08092465996742249,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 131715
+    },
+    {
+      "epoch": 13.524797207105452,
+      "grad_norm": 0.05665673315525055,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 131718
+    },
+    {
+      "epoch": 13.525105246945271,
+      "grad_norm": 0.11822622269392014,
+      "learning_rate": 0.01,
+      "loss": 1.9202,
+      "step": 131721
+    },
+    {
+      "epoch": 13.52541328678509,
+      "grad_norm": 0.057077016681432724,
+      "learning_rate": 0.01,
+      "loss": 1.9019,
+      "step": 131724
+    },
+    {
+      "epoch": 13.525721326624911,
+      "grad_norm": 0.053292274475097656,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 131727
+    },
+    {
+      "epoch": 13.52602936646473,
+      "grad_norm": 0.050330668687820435,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 131730
+    },
+    {
+      "epoch": 13.526337406304549,
+      "grad_norm": 0.08987832069396973,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 131733
+    },
+    {
+      "epoch": 13.526645446144368,
+      "grad_norm": 0.052682794630527496,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 131736
+    },
+    {
+      "epoch": 13.526953485984187,
+      "grad_norm": 0.11335168778896332,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 131739
+    },
+    {
+      "epoch": 13.527261525824006,
+      "grad_norm": 0.05990173667669296,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 131742
+    },
+    {
+      "epoch": 13.527569565663827,
+      "grad_norm": 0.03620849549770355,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 131745
+    },
+    {
+      "epoch": 13.527877605503646,
+      "grad_norm": 0.07759151607751846,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 131748
+    },
+    {
+      "epoch": 13.528185645343465,
+      "grad_norm": 0.06983242928981781,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 131751
+    },
+    {
+      "epoch": 13.528493685183284,
+      "grad_norm": 0.06686399132013321,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 131754
+    },
+    {
+      "epoch": 13.528801725023103,
+      "grad_norm": 0.04652488976716995,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 131757
+    },
+    {
+      "epoch": 13.529109764862921,
+      "grad_norm": 0.0452822670340538,
+      "learning_rate": 0.01,
+      "loss": 1.9024,
+      "step": 131760
+    },
+    {
+      "epoch": 13.529417804702742,
+      "grad_norm": 0.05697820708155632,
+      "learning_rate": 0.01,
+      "loss": 1.8942,
+      "step": 131763
+    },
+    {
+      "epoch": 13.529725844542561,
+      "grad_norm": 0.04186016693711281,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 131766
+    },
+    {
+      "epoch": 13.53003388438238,
+      "grad_norm": 0.04754483327269554,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 131769
+    },
+    {
+      "epoch": 13.5303419242222,
+      "grad_norm": 0.07774531096220016,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 131772
+    },
+    {
+      "epoch": 13.530649964062018,
+      "grad_norm": 0.06691887974739075,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 131775
+    },
+    {
+      "epoch": 13.530958003901837,
+      "grad_norm": 0.10388603806495667,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 131778
+    },
+    {
+      "epoch": 13.531266043741658,
+      "grad_norm": 0.03247929736971855,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 131781
+    },
+    {
+      "epoch": 13.531574083581477,
+      "grad_norm": 0.04648836329579353,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 131784
+    },
+    {
+      "epoch": 13.531882123421296,
+      "grad_norm": 0.04273877665400505,
+      "learning_rate": 0.01,
+      "loss": 1.9083,
+      "step": 131787
+    },
+    {
+      "epoch": 13.532190163261115,
+      "grad_norm": 0.032543834298849106,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 131790
+    },
+    {
+      "epoch": 13.532498203100934,
+      "grad_norm": 0.16048769652843475,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 131793
+    },
+    {
+      "epoch": 13.532806242940755,
+      "grad_norm": 0.12636518478393555,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 131796
+    },
+    {
+      "epoch": 13.533114282780573,
+      "grad_norm": 0.050800494849681854,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 131799
+    },
+    {
+      "epoch": 13.533422322620392,
+      "grad_norm": 0.049428943544626236,
+      "learning_rate": 0.01,
+      "loss": 1.9349,
+      "step": 131802
+    },
+    {
+      "epoch": 13.533730362460211,
+      "grad_norm": 0.1052422821521759,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 131805
+    },
+    {
+      "epoch": 13.53403840230003,
+      "grad_norm": 0.06669687479734421,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 131808
+    },
+    {
+      "epoch": 13.53434644213985,
+      "grad_norm": 0.10331539809703827,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 131811
+    },
+    {
+      "epoch": 13.53465448197967,
+      "grad_norm": 0.07682331651449203,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 131814
+    },
+    {
+      "epoch": 13.53496252181949,
+      "grad_norm": 0.03658591955900192,
+      "learning_rate": 0.01,
+      "loss": 1.894,
+      "step": 131817
+    },
+    {
+      "epoch": 13.535270561659308,
+      "grad_norm": 0.03765729069709778,
+      "learning_rate": 0.01,
+      "loss": 1.9406,
+      "step": 131820
+    },
+    {
+      "epoch": 13.535578601499127,
+      "grad_norm": 0.07123025506734848,
+      "learning_rate": 0.01,
+      "loss": 1.8907,
+      "step": 131823
+    },
+    {
+      "epoch": 13.535886641338946,
+      "grad_norm": 0.06337366998195648,
+      "learning_rate": 0.01,
+      "loss": 1.9015,
+      "step": 131826
+    },
+    {
+      "epoch": 13.536194681178765,
+      "grad_norm": 0.1051003560423851,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 131829
+    },
+    {
+      "epoch": 13.536502721018586,
+      "grad_norm": 0.15537908673286438,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 131832
+    },
+    {
+      "epoch": 13.536810760858405,
+      "grad_norm": 0.10030465573072433,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 131835
+    },
+    {
+      "epoch": 13.537118800698224,
+      "grad_norm": 0.05172615498304367,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 131838
+    },
+    {
+      "epoch": 13.537426840538043,
+      "grad_norm": 0.0509798638522625,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 131841
+    },
+    {
+      "epoch": 13.537734880377862,
+      "grad_norm": 0.034889016300439835,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 131844
+    },
+    {
+      "epoch": 13.53804292021768,
+      "grad_norm": 0.07696577161550522,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 131847
+    },
+    {
+      "epoch": 13.538350960057501,
+      "grad_norm": 0.13341930508613586,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 131850
+    },
+    {
+      "epoch": 13.53865899989732,
+      "grad_norm": 0.16740930080413818,
+      "learning_rate": 0.01,
+      "loss": 1.8935,
+      "step": 131853
+    },
+    {
+      "epoch": 13.53896703973714,
+      "grad_norm": 0.07476458698511124,
+      "learning_rate": 0.01,
+      "loss": 1.8977,
+      "step": 131856
+    },
+    {
+      "epoch": 13.539275079576958,
+      "grad_norm": 0.06607705354690552,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 131859
+    },
+    {
+      "epoch": 13.539583119416777,
+      "grad_norm": 0.051382292062044144,
+      "learning_rate": 0.01,
+      "loss": 1.8887,
+      "step": 131862
+    },
+    {
+      "epoch": 13.539891159256598,
+      "grad_norm": 0.04293932765722275,
+      "learning_rate": 0.01,
+      "loss": 1.9493,
+      "step": 131865
+    },
+    {
+      "epoch": 13.540199199096417,
+      "grad_norm": 0.037694815546274185,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 131868
+    },
+    {
+      "epoch": 13.540507238936236,
+      "grad_norm": 0.04440714418888092,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 131871
+    },
+    {
+      "epoch": 13.540815278776055,
+      "grad_norm": 0.07876510918140411,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 131874
+    },
+    {
+      "epoch": 13.541123318615874,
+      "grad_norm": 0.08295460045337677,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 131877
+    },
+    {
+      "epoch": 13.541431358455693,
+      "grad_norm": 0.0933263748884201,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 131880
+    },
+    {
+      "epoch": 13.541739398295514,
+      "grad_norm": 0.0831962302327156,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 131883
+    },
+    {
+      "epoch": 13.542047438135333,
+      "grad_norm": 0.06657032668590546,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 131886
+    },
+    {
+      "epoch": 13.542355477975152,
+      "grad_norm": 0.10307253897190094,
+      "learning_rate": 0.01,
+      "loss": 1.9508,
+      "step": 131889
+    },
+    {
+      "epoch": 13.54266351781497,
+      "grad_norm": 0.12840963900089264,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 131892
+    },
+    {
+      "epoch": 13.54297155765479,
+      "grad_norm": 0.054330550134181976,
+      "learning_rate": 0.01,
+      "loss": 1.8797,
+      "step": 131895
+    },
+    {
+      "epoch": 13.543279597494609,
+      "grad_norm": 0.063980333507061,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 131898
+    },
+    {
+      "epoch": 13.54358763733443,
+      "grad_norm": 0.05326567962765694,
+      "learning_rate": 0.01,
+      "loss": 1.9422,
+      "step": 131901
+    },
+    {
+      "epoch": 13.543895677174248,
+      "grad_norm": 0.10171520709991455,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 131904
+    },
+    {
+      "epoch": 13.544203717014067,
+      "grad_norm": 0.06023498624563217,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 131907
+    },
+    {
+      "epoch": 13.544511756853886,
+      "grad_norm": 0.02960367314517498,
+      "learning_rate": 0.01,
+      "loss": 1.9566,
+      "step": 131910
+    },
+    {
+      "epoch": 13.544819796693705,
+      "grad_norm": 0.037857554852962494,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 131913
+    },
+    {
+      "epoch": 13.545127836533524,
+      "grad_norm": 0.134146586060524,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 131916
+    },
+    {
+      "epoch": 13.545435876373345,
+      "grad_norm": 0.04209603741765022,
+      "learning_rate": 0.01,
+      "loss": 1.8946,
+      "step": 131919
+    },
+    {
+      "epoch": 13.545743916213164,
+      "grad_norm": 0.056739822030067444,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 131922
+    },
+    {
+      "epoch": 13.546051956052983,
+      "grad_norm": 0.07559722661972046,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 131925
+    },
+    {
+      "epoch": 13.546359995892802,
+      "grad_norm": 0.08239512890577316,
+      "learning_rate": 0.01,
+      "loss": 1.8882,
+      "step": 131928
+    },
+    {
+      "epoch": 13.54666803573262,
+      "grad_norm": 0.1327822059392929,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 131931
+    },
+    {
+      "epoch": 13.546976075572442,
+      "grad_norm": 0.0559154637157917,
+      "learning_rate": 0.01,
+      "loss": 1.9014,
+      "step": 131934
+    },
+    {
+      "epoch": 13.54728411541226,
+      "grad_norm": 0.06256121397018433,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 131937
+    },
+    {
+      "epoch": 13.54759215525208,
+      "grad_norm": 0.054838620126247406,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 131940
+    },
+    {
+      "epoch": 13.547900195091898,
+      "grad_norm": 0.060045551508665085,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 131943
+    },
+    {
+      "epoch": 13.548208234931717,
+      "grad_norm": 0.07156597822904587,
+      "learning_rate": 0.01,
+      "loss": 1.871,
+      "step": 131946
+    },
+    {
+      "epoch": 13.548516274771536,
+      "grad_norm": 0.10272616893053055,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 131949
+    },
+    {
+      "epoch": 13.548824314611357,
+      "grad_norm": 0.09434785693883896,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 131952
+    },
+    {
+      "epoch": 13.549132354451176,
+      "grad_norm": 0.11009252816438675,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 131955
+    },
+    {
+      "epoch": 13.549440394290995,
+      "grad_norm": 0.06729310750961304,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 131958
+    },
+    {
+      "epoch": 13.549748434130814,
+      "grad_norm": 0.08687084168195724,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 131961
+    },
+    {
+      "epoch": 13.550056473970633,
+      "grad_norm": 0.0786261036992073,
+      "learning_rate": 0.01,
+      "loss": 1.9042,
+      "step": 131964
+    },
+    {
+      "epoch": 13.550364513810452,
+      "grad_norm": 0.06976866722106934,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 131967
+    },
+    {
+      "epoch": 13.550672553650273,
+      "grad_norm": 0.06729738414287567,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 131970
+    },
+    {
+      "epoch": 13.550980593490092,
+      "grad_norm": 0.04848663881421089,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 131973
+    },
+    {
+      "epoch": 13.55128863332991,
+      "grad_norm": 0.08129465579986572,
+      "learning_rate": 0.01,
+      "loss": 1.9483,
+      "step": 131976
+    },
+    {
+      "epoch": 13.55159667316973,
+      "grad_norm": 0.06423746049404144,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 131979
+    },
+    {
+      "epoch": 13.551904713009549,
+      "grad_norm": 0.08938214927911758,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 131982
+    },
+    {
+      "epoch": 13.552212752849368,
+      "grad_norm": 0.10618574917316437,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 131985
+    },
+    {
+      "epoch": 13.552520792689188,
+      "grad_norm": 0.06559048593044281,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 131988
+    },
+    {
+      "epoch": 13.552828832529007,
+      "grad_norm": 0.10146395862102509,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 131991
+    },
+    {
+      "epoch": 13.553136872368826,
+      "grad_norm": 0.06789354979991913,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 131994
+    },
+    {
+      "epoch": 13.553444912208645,
+      "grad_norm": 0.052592888474464417,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 131997
+    },
+    {
+      "epoch": 13.553752952048464,
+      "grad_norm": 0.05447304621338844,
+      "learning_rate": 0.01,
+      "loss": 1.8935,
+      "step": 132000
+    },
+    {
+      "epoch": 13.554060991888285,
+      "grad_norm": 0.12737742066383362,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 132003
+    },
+    {
+      "epoch": 13.554369031728104,
+      "grad_norm": 0.041699107736349106,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 132006
+    },
+    {
+      "epoch": 13.554677071567923,
+      "grad_norm": 0.09523919969797134,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 132009
+    },
+    {
+      "epoch": 13.554985111407742,
+      "grad_norm": 0.04702223092317581,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 132012
+    },
+    {
+      "epoch": 13.555293151247561,
+      "grad_norm": 0.0915573313832283,
+      "learning_rate": 0.01,
+      "loss": 1.9032,
+      "step": 132015
+    },
+    {
+      "epoch": 13.55560119108738,
+      "grad_norm": 0.0788947269320488,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 132018
+    },
+    {
+      "epoch": 13.5559092309272,
+      "grad_norm": 0.05678980052471161,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 132021
+    },
+    {
+      "epoch": 13.55621727076702,
+      "grad_norm": 0.08159255981445312,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 132024
+    },
+    {
+      "epoch": 13.556525310606839,
+      "grad_norm": 0.07052342593669891,
+      "learning_rate": 0.01,
+      "loss": 1.9544,
+      "step": 132027
+    },
+    {
+      "epoch": 13.556833350446658,
+      "grad_norm": 0.08300250768661499,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 132030
+    },
+    {
+      "epoch": 13.557141390286477,
+      "grad_norm": 0.036307577043771744,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 132033
+    },
+    {
+      "epoch": 13.557449430126296,
+      "grad_norm": 0.050889384001493454,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 132036
+    },
+    {
+      "epoch": 13.557757469966116,
+      "grad_norm": 0.04341815039515495,
+      "learning_rate": 0.01,
+      "loss": 1.9345,
+      "step": 132039
+    },
+    {
+      "epoch": 13.558065509805935,
+      "grad_norm": 0.14564576745033264,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 132042
+    },
+    {
+      "epoch": 13.558373549645754,
+      "grad_norm": 0.06567149609327316,
+      "learning_rate": 0.01,
+      "loss": 1.9386,
+      "step": 132045
+    },
+    {
+      "epoch": 13.558681589485573,
+      "grad_norm": 0.04719534516334534,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 132048
+    },
+    {
+      "epoch": 13.558989629325392,
+      "grad_norm": 0.04680018126964569,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 132051
+    },
+    {
+      "epoch": 13.559297669165211,
+      "grad_norm": 0.039485737681388855,
+      "learning_rate": 0.01,
+      "loss": 1.9545,
+      "step": 132054
+    },
+    {
+      "epoch": 13.559605709005032,
+      "grad_norm": 0.039566244930028915,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 132057
+    },
+    {
+      "epoch": 13.559913748844851,
+      "grad_norm": 0.12616567313671112,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 132060
+    },
+    {
+      "epoch": 13.56022178868467,
+      "grad_norm": 0.04869133234024048,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 132063
+    },
+    {
+      "epoch": 13.560529828524489,
+      "grad_norm": 0.06983968615531921,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 132066
+    },
+    {
+      "epoch": 13.560837868364308,
+      "grad_norm": 0.061722587794065475,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 132069
+    },
+    {
+      "epoch": 13.561145908204129,
+      "grad_norm": 0.1154983788728714,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 132072
+    },
+    {
+      "epoch": 13.561453948043948,
+      "grad_norm": 0.08315840363502502,
+      "learning_rate": 0.01,
+      "loss": 1.9608,
+      "step": 132075
+    },
+    {
+      "epoch": 13.561761987883767,
+      "grad_norm": 0.08502950519323349,
+      "learning_rate": 0.01,
+      "loss": 1.9444,
+      "step": 132078
+    },
+    {
+      "epoch": 13.562070027723585,
+      "grad_norm": 0.08286229521036148,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 132081
+    },
+    {
+      "epoch": 13.562378067563404,
+      "grad_norm": 0.07582364976406097,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 132084
+    },
+    {
+      "epoch": 13.562686107403223,
+      "grad_norm": 0.07250833511352539,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 132087
+    },
+    {
+      "epoch": 13.562994147243044,
+      "grad_norm": 0.08048056066036224,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 132090
+    },
+    {
+      "epoch": 13.563302187082863,
+      "grad_norm": 0.11056806892156601,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 132093
+    },
+    {
+      "epoch": 13.563610226922682,
+      "grad_norm": 0.050416793674230576,
+      "learning_rate": 0.01,
+      "loss": 1.8978,
+      "step": 132096
+    },
+    {
+      "epoch": 13.563918266762501,
+      "grad_norm": 0.04154177010059357,
+      "learning_rate": 0.01,
+      "loss": 1.9437,
+      "step": 132099
+    },
+    {
+      "epoch": 13.56422630660232,
+      "grad_norm": 0.06463085114955902,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 132102
+    },
+    {
+      "epoch": 13.564534346442139,
+      "grad_norm": 0.03718942403793335,
+      "learning_rate": 0.01,
+      "loss": 1.9723,
+      "step": 132105
+    },
+    {
+      "epoch": 13.56484238628196,
+      "grad_norm": 0.03484407067298889,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 132108
+    },
+    {
+      "epoch": 13.565150426121779,
+      "grad_norm": 0.03363065794110298,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 132111
+    },
+    {
+      "epoch": 13.565458465961598,
+      "grad_norm": 0.04130982607603073,
+      "learning_rate": 0.01,
+      "loss": 1.8804,
+      "step": 132114
+    },
+    {
+      "epoch": 13.565766505801417,
+      "grad_norm": 0.041345711797475815,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 132117
+    },
+    {
+      "epoch": 13.566074545641236,
+      "grad_norm": 0.10748176276683807,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 132120
+    },
+    {
+      "epoch": 13.566382585481055,
+      "grad_norm": 0.09795956313610077,
+      "learning_rate": 0.01,
+      "loss": 1.9062,
+      "step": 132123
+    },
+    {
+      "epoch": 13.566690625320875,
+      "grad_norm": 0.04317321255803108,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 132126
+    },
+    {
+      "epoch": 13.566998665160694,
+      "grad_norm": 0.047823723405599594,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 132129
+    },
+    {
+      "epoch": 13.567306705000513,
+      "grad_norm": 0.07286576926708221,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 132132
+    },
+    {
+      "epoch": 13.567614744840332,
+      "grad_norm": 0.059980206191539764,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 132135
+    },
+    {
+      "epoch": 13.567922784680151,
+      "grad_norm": 0.031100593507289886,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 132138
+    },
+    {
+      "epoch": 13.568230824519972,
+      "grad_norm": 0.042645007371902466,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 132141
+    },
+    {
+      "epoch": 13.568538864359791,
+      "grad_norm": 0.03142551705241203,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 132144
+    },
+    {
+      "epoch": 13.56884690419961,
+      "grad_norm": 0.03679187223315239,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 132147
+    },
+    {
+      "epoch": 13.569154944039429,
+      "grad_norm": 0.03217216953635216,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 132150
+    },
+    {
+      "epoch": 13.569462983879248,
+      "grad_norm": 0.05555422604084015,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 132153
+    },
+    {
+      "epoch": 13.569771023719067,
+      "grad_norm": 0.07838881760835648,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 132156
+    },
+    {
+      "epoch": 13.570079063558888,
+      "grad_norm": 0.045605260878801346,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 132159
+    },
+    {
+      "epoch": 13.570387103398707,
+      "grad_norm": 0.04283960536122322,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 132162
+    },
+    {
+      "epoch": 13.570695143238526,
+      "grad_norm": 0.17408624291419983,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 132165
+    },
+    {
+      "epoch": 13.571003183078345,
+      "grad_norm": 0.0772915706038475,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 132168
+    },
+    {
+      "epoch": 13.571311222918164,
+      "grad_norm": 0.09446059912443161,
+      "learning_rate": 0.01,
+      "loss": 1.9384,
+      "step": 132171
+    },
+    {
+      "epoch": 13.571619262757983,
+      "grad_norm": 0.05909676104784012,
+      "learning_rate": 0.01,
+      "loss": 1.8861,
+      "step": 132174
+    },
+    {
+      "epoch": 13.571927302597803,
+      "grad_norm": 0.04434170946478844,
+      "learning_rate": 0.01,
+      "loss": 1.8865,
+      "step": 132177
+    },
+    {
+      "epoch": 13.572235342437622,
+      "grad_norm": 0.039491549134254456,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 132180
+    },
+    {
+      "epoch": 13.572543382277441,
+      "grad_norm": 0.029434842988848686,
+      "learning_rate": 0.01,
+      "loss": 1.9009,
+      "step": 132183
+    },
+    {
+      "epoch": 13.57285142211726,
+      "grad_norm": 0.030724162235856056,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 132186
+    },
+    {
+      "epoch": 13.57315946195708,
+      "grad_norm": 0.05312420800328255,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 132189
+    },
+    {
+      "epoch": 13.573467501796898,
+      "grad_norm": 0.1140730157494545,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 132192
+    },
+    {
+      "epoch": 13.573775541636719,
+      "grad_norm": 0.03953307867050171,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 132195
+    },
+    {
+      "epoch": 13.574083581476538,
+      "grad_norm": 0.058202117681503296,
+      "learning_rate": 0.01,
+      "loss": 1.8906,
+      "step": 132198
+    },
+    {
+      "epoch": 13.574391621316357,
+      "grad_norm": 0.10719067603349686,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 132201
+    },
+    {
+      "epoch": 13.574699661156176,
+      "grad_norm": 0.08207898586988449,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 132204
+    },
+    {
+      "epoch": 13.575007700995995,
+      "grad_norm": 0.058483630418777466,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 132207
+    },
+    {
+      "epoch": 13.575315740835816,
+      "grad_norm": 0.030889760702848434,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 132210
+    },
+    {
+      "epoch": 13.575623780675635,
+      "grad_norm": 0.11797042191028595,
+      "learning_rate": 0.01,
+      "loss": 1.9021,
+      "step": 132213
+    },
+    {
+      "epoch": 13.575931820515454,
+      "grad_norm": 0.07183902710676193,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 132216
+    },
+    {
+      "epoch": 13.576239860355273,
+      "grad_norm": 0.06486035883426666,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 132219
+    },
+    {
+      "epoch": 13.576547900195092,
+      "grad_norm": 0.05778775364160538,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 132222
+    },
+    {
+      "epoch": 13.57685594003491,
+      "grad_norm": 0.054141197353601456,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 132225
+    },
+    {
+      "epoch": 13.577163979874731,
+      "grad_norm": 0.038162704557180405,
+      "learning_rate": 0.01,
+      "loss": 1.9342,
+      "step": 132228
+    },
+    {
+      "epoch": 13.57747201971455,
+      "grad_norm": 0.10111169517040253,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 132231
+    },
+    {
+      "epoch": 13.57778005955437,
+      "grad_norm": 0.10514866560697556,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 132234
+    },
+    {
+      "epoch": 13.578088099394188,
+      "grad_norm": 0.08045943826436996,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 132237
+    },
+    {
+      "epoch": 13.578396139234007,
+      "grad_norm": 0.08001212775707245,
+      "learning_rate": 0.01,
+      "loss": 1.9109,
+      "step": 132240
+    },
+    {
+      "epoch": 13.578704179073826,
+      "grad_norm": 0.06874321401119232,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 132243
+    },
+    {
+      "epoch": 13.579012218913647,
+      "grad_norm": 0.07667498290538788,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 132246
+    },
+    {
+      "epoch": 13.579320258753466,
+      "grad_norm": 0.06881631910800934,
+      "learning_rate": 0.01,
+      "loss": 1.9009,
+      "step": 132249
+    },
+    {
+      "epoch": 13.579628298593285,
+      "grad_norm": 0.08229288458824158,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 132252
+    },
+    {
+      "epoch": 13.579936338433104,
+      "grad_norm": 0.058794401586055756,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 132255
+    },
+    {
+      "epoch": 13.580244378272923,
+      "grad_norm": 0.11294376105070114,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 132258
+    },
+    {
+      "epoch": 13.580552418112742,
+      "grad_norm": 0.045323677361011505,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 132261
+    },
+    {
+      "epoch": 13.580860457952562,
+      "grad_norm": 0.08761949092149734,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 132264
+    },
+    {
+      "epoch": 13.581168497792381,
+      "grad_norm": 0.0401795357465744,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 132267
+    },
+    {
+      "epoch": 13.5814765376322,
+      "grad_norm": 0.0404195673763752,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 132270
+    },
+    {
+      "epoch": 13.58178457747202,
+      "grad_norm": 0.07799965888261795,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 132273
+    },
+    {
+      "epoch": 13.582092617311838,
+      "grad_norm": 0.1280081570148468,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 132276
+    },
+    {
+      "epoch": 13.58240065715166,
+      "grad_norm": 0.09373471885919571,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 132279
+    },
+    {
+      "epoch": 13.582708696991478,
+      "grad_norm": 0.1325744092464447,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 132282
+    },
+    {
+      "epoch": 13.583016736831297,
+      "grad_norm": 0.08909572660923004,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 132285
+    },
+    {
+      "epoch": 13.583324776671116,
+      "grad_norm": 0.08770900219678879,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 132288
+    },
+    {
+      "epoch": 13.583632816510935,
+      "grad_norm": 0.12634555995464325,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 132291
+    },
+    {
+      "epoch": 13.583940856350754,
+      "grad_norm": 0.0461869053542614,
+      "learning_rate": 0.01,
+      "loss": 1.9473,
+      "step": 132294
+    },
+    {
+      "epoch": 13.584248896190575,
+      "grad_norm": 0.06307772547006607,
+      "learning_rate": 0.01,
+      "loss": 1.8975,
+      "step": 132297
+    },
+    {
+      "epoch": 13.584556936030394,
+      "grad_norm": 0.05029197409749031,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 132300
+    },
+    {
+      "epoch": 13.584864975870213,
+      "grad_norm": 0.050476524978876114,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 132303
+    },
+    {
+      "epoch": 13.585173015710032,
+      "grad_norm": 0.043588001281023026,
+      "learning_rate": 0.01,
+      "loss": 1.9076,
+      "step": 132306
+    },
+    {
+      "epoch": 13.58548105554985,
+      "grad_norm": 0.040362220257520676,
+      "learning_rate": 0.01,
+      "loss": 1.9035,
+      "step": 132309
+    },
+    {
+      "epoch": 13.58578909538967,
+      "grad_norm": 0.052701760083436966,
+      "learning_rate": 0.01,
+      "loss": 1.9206,
+      "step": 132312
+    },
+    {
+      "epoch": 13.58609713522949,
+      "grad_norm": 0.08644449710845947,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 132315
+    },
+    {
+      "epoch": 13.58640517506931,
+      "grad_norm": 0.11038315296173096,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 132318
+    },
+    {
+      "epoch": 13.586713214909128,
+      "grad_norm": 0.04978395625948906,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 132321
+    },
+    {
+      "epoch": 13.587021254748947,
+      "grad_norm": 0.03198670223355293,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 132324
+    },
+    {
+      "epoch": 13.587329294588766,
+      "grad_norm": 0.037590526044368744,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 132327
+    },
+    {
+      "epoch": 13.587637334428585,
+      "grad_norm": 0.053910113871097565,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 132330
+    },
+    {
+      "epoch": 13.587945374268406,
+      "grad_norm": 0.0418432280421257,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 132333
+    },
+    {
+      "epoch": 13.588253414108225,
+      "grad_norm": 0.043975070118904114,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 132336
+    },
+    {
+      "epoch": 13.588561453948044,
+      "grad_norm": 0.12854112684726715,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 132339
+    },
+    {
+      "epoch": 13.588869493787863,
+      "grad_norm": 0.045041751116514206,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 132342
+    },
+    {
+      "epoch": 13.589177533627682,
+      "grad_norm": 0.03786958381533623,
+      "learning_rate": 0.01,
+      "loss": 1.9022,
+      "step": 132345
+    },
+    {
+      "epoch": 13.589485573467503,
+      "grad_norm": 0.04743557795882225,
+      "learning_rate": 0.01,
+      "loss": 1.955,
+      "step": 132348
+    },
+    {
+      "epoch": 13.589793613307322,
+      "grad_norm": 0.04899417236447334,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 132351
+    },
+    {
+      "epoch": 13.59010165314714,
+      "grad_norm": 0.09649678319692612,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 132354
+    },
+    {
+      "epoch": 13.59040969298696,
+      "grad_norm": 0.08850428462028503,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 132357
+    },
+    {
+      "epoch": 13.590717732826779,
+      "grad_norm": 0.040218986570835114,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 132360
+    },
+    {
+      "epoch": 13.591025772666598,
+      "grad_norm": 0.060465436428785324,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 132363
+    },
+    {
+      "epoch": 13.591333812506418,
+      "grad_norm": 0.05747165530920029,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 132366
+    },
+    {
+      "epoch": 13.591641852346237,
+      "grad_norm": 0.05838266387581825,
+      "learning_rate": 0.01,
+      "loss": 1.8824,
+      "step": 132369
+    },
+    {
+      "epoch": 13.591949892186056,
+      "grad_norm": 0.12548838555812836,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 132372
+    },
+    {
+      "epoch": 13.592257932025875,
+      "grad_norm": 0.043225549161434174,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 132375
+    },
+    {
+      "epoch": 13.592565971865694,
+      "grad_norm": 0.03802839294075966,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 132378
+    },
+    {
+      "epoch": 13.592874011705513,
+      "grad_norm": 0.06532235443592072,
+      "learning_rate": 0.01,
+      "loss": 1.8905,
+      "step": 132381
+    },
+    {
+      "epoch": 13.593182051545334,
+      "grad_norm": 0.085301473736763,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 132384
+    },
+    {
+      "epoch": 13.593490091385153,
+      "grad_norm": 0.04738848656415939,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 132387
+    },
+    {
+      "epoch": 13.593798131224972,
+      "grad_norm": 0.11639575660228729,
+      "learning_rate": 0.01,
+      "loss": 1.9404,
+      "step": 132390
+    },
+    {
+      "epoch": 13.59410617106479,
+      "grad_norm": 0.0606863908469677,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 132393
+    },
+    {
+      "epoch": 13.59441421090461,
+      "grad_norm": 0.059407833963632584,
+      "learning_rate": 0.01,
+      "loss": 1.8939,
+      "step": 132396
+    },
+    {
+      "epoch": 13.594722250744429,
+      "grad_norm": 0.1593611091375351,
+      "learning_rate": 0.01,
+      "loss": 1.8978,
+      "step": 132399
+    },
+    {
+      "epoch": 13.59503029058425,
+      "grad_norm": 0.043320685625076294,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 132402
+    },
+    {
+      "epoch": 13.595338330424068,
+      "grad_norm": 0.03582308441400528,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 132405
+    },
+    {
+      "epoch": 13.595646370263887,
+      "grad_norm": 0.05099843069911003,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 132408
+    },
+    {
+      "epoch": 13.595954410103706,
+      "grad_norm": 0.03142688795924187,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 132411
+    },
+    {
+      "epoch": 13.596262449943525,
+      "grad_norm": 0.07230118662118912,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 132414
+    },
+    {
+      "epoch": 13.596570489783346,
+      "grad_norm": 0.08642298728227615,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 132417
+    },
+    {
+      "epoch": 13.596878529623165,
+      "grad_norm": 0.13895951211452484,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 132420
+    },
+    {
+      "epoch": 13.597186569462984,
+      "grad_norm": 0.11629573255777359,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 132423
+    },
+    {
+      "epoch": 13.597494609302803,
+      "grad_norm": 0.17057494819164276,
+      "learning_rate": 0.01,
+      "loss": 1.9502,
+      "step": 132426
+    },
+    {
+      "epoch": 13.597802649142622,
+      "grad_norm": 0.08752473443746567,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 132429
+    },
+    {
+      "epoch": 13.598110688982441,
+      "grad_norm": 0.045257922261953354,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 132432
+    },
+    {
+      "epoch": 13.598418728822262,
+      "grad_norm": 0.03428911417722702,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 132435
+    },
+    {
+      "epoch": 13.59872676866208,
+      "grad_norm": 0.04752839729189873,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 132438
+    },
+    {
+      "epoch": 13.5990348085019,
+      "grad_norm": 0.049963582307100296,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 132441
+    },
+    {
+      "epoch": 13.599342848341719,
+      "grad_norm": 0.1061190664768219,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 132444
+    },
+    {
+      "epoch": 13.599650888181538,
+      "grad_norm": 0.05499035865068436,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 132447
+    },
+    {
+      "epoch": 13.599958928021357,
+      "grad_norm": 0.0994025245308876,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 132450
+    },
+    {
+      "epoch": 13.600266967861177,
+      "grad_norm": 0.08412833511829376,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 132453
+    },
+    {
+      "epoch": 13.600575007700996,
+      "grad_norm": 0.08766642212867737,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 132456
+    },
+    {
+      "epoch": 13.600883047540815,
+      "grad_norm": 0.06333081424236298,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 132459
+    },
+    {
+      "epoch": 13.601191087380634,
+      "grad_norm": 0.03847841918468475,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 132462
+    },
+    {
+      "epoch": 13.601499127220453,
+      "grad_norm": 0.038402970880270004,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 132465
+    },
+    {
+      "epoch": 13.601807167060272,
+      "grad_norm": 0.04149443283677101,
+      "learning_rate": 0.01,
+      "loss": 1.8982,
+      "step": 132468
+    },
+    {
+      "epoch": 13.602115206900093,
+      "grad_norm": 0.04059450700879097,
+      "learning_rate": 0.01,
+      "loss": 1.8974,
+      "step": 132471
+    },
+    {
+      "epoch": 13.602423246739912,
+      "grad_norm": 0.03619808331131935,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 132474
+    },
+    {
+      "epoch": 13.602731286579731,
+      "grad_norm": 0.08096408098936081,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 132477
+    },
+    {
+      "epoch": 13.60303932641955,
+      "grad_norm": 0.05927874520421028,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 132480
+    },
+    {
+      "epoch": 13.603347366259369,
+      "grad_norm": 0.08426184952259064,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 132483
+    },
+    {
+      "epoch": 13.60365540609919,
+      "grad_norm": 0.09298544377088547,
+      "learning_rate": 0.01,
+      "loss": 1.8867,
+      "step": 132486
+    },
+    {
+      "epoch": 13.603963445939009,
+      "grad_norm": 0.14834368228912354,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 132489
+    },
+    {
+      "epoch": 13.604271485778828,
+      "grad_norm": 0.05329186096787453,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 132492
+    },
+    {
+      "epoch": 13.604579525618647,
+      "grad_norm": 0.0344843752682209,
+      "learning_rate": 0.01,
+      "loss": 1.8932,
+      "step": 132495
+    },
+    {
+      "epoch": 13.604887565458466,
+      "grad_norm": 0.10429553687572479,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 132498
+    },
+    {
+      "epoch": 13.605195605298285,
+      "grad_norm": 0.14243310689926147,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 132501
+    },
+    {
+      "epoch": 13.605503645138105,
+      "grad_norm": 0.12058507651090622,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 132504
+    },
+    {
+      "epoch": 13.605811684977924,
+      "grad_norm": 0.060153014957904816,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 132507
+    },
+    {
+      "epoch": 13.606119724817743,
+      "grad_norm": 0.09400785714387894,
+      "learning_rate": 0.01,
+      "loss": 1.9573,
+      "step": 132510
+    },
+    {
+      "epoch": 13.606427764657562,
+      "grad_norm": 0.061529334634542465,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 132513
+    },
+    {
+      "epoch": 13.606735804497381,
+      "grad_norm": 0.0380091592669487,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 132516
+    },
+    {
+      "epoch": 13.6070438443372,
+      "grad_norm": 0.07472321391105652,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 132519
+    },
+    {
+      "epoch": 13.607351884177021,
+      "grad_norm": 0.11282123625278473,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 132522
+    },
+    {
+      "epoch": 13.60765992401684,
+      "grad_norm": 0.04835216701030731,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 132525
+    },
+    {
+      "epoch": 13.607967963856659,
+      "grad_norm": 0.09492181241512299,
+      "learning_rate": 0.01,
+      "loss": 1.9125,
+      "step": 132528
+    },
+    {
+      "epoch": 13.608276003696478,
+      "grad_norm": 0.07662288099527359,
+      "learning_rate": 0.01,
+      "loss": 1.9278,
+      "step": 132531
+    },
+    {
+      "epoch": 13.608584043536297,
+      "grad_norm": 0.0752788856625557,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 132534
+    },
+    {
+      "epoch": 13.608892083376116,
+      "grad_norm": 0.10046999156475067,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 132537
+    },
+    {
+      "epoch": 13.609200123215937,
+      "grad_norm": 0.10675060749053955,
+      "learning_rate": 0.01,
+      "loss": 1.9479,
+      "step": 132540
+    },
+    {
+      "epoch": 13.609508163055756,
+      "grad_norm": 0.06775020062923431,
+      "learning_rate": 0.01,
+      "loss": 1.8779,
+      "step": 132543
+    },
+    {
+      "epoch": 13.609816202895574,
+      "grad_norm": 0.05218996852636337,
+      "learning_rate": 0.01,
+      "loss": 1.9231,
+      "step": 132546
+    },
+    {
+      "epoch": 13.610124242735393,
+      "grad_norm": 0.06487781554460526,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 132549
+    },
+    {
+      "epoch": 13.610432282575212,
+      "grad_norm": 0.05243084207177162,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 132552
+    },
+    {
+      "epoch": 13.610740322415033,
+      "grad_norm": 0.06522991508245468,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 132555
+    },
+    {
+      "epoch": 13.611048362254852,
+      "grad_norm": 0.10282405465841293,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 132558
+    },
+    {
+      "epoch": 13.611356402094671,
+      "grad_norm": 0.11704733222723007,
+      "learning_rate": 0.01,
+      "loss": 1.8992,
+      "step": 132561
+    },
+    {
+      "epoch": 13.61166444193449,
+      "grad_norm": 0.0647767186164856,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 132564
+    },
+    {
+      "epoch": 13.611972481774309,
+      "grad_norm": 0.043058864772319794,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 132567
+    },
+    {
+      "epoch": 13.612280521614128,
+      "grad_norm": 0.04453499987721443,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 132570
+    },
+    {
+      "epoch": 13.612588561453949,
+      "grad_norm": 0.03758270666003227,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 132573
+    },
+    {
+      "epoch": 13.612896601293768,
+      "grad_norm": 0.02714291587471962,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 132576
+    },
+    {
+      "epoch": 13.613204641133587,
+      "grad_norm": 0.0367596372961998,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 132579
+    },
+    {
+      "epoch": 13.613512680973406,
+      "grad_norm": 0.13597920536994934,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 132582
+    },
+    {
+      "epoch": 13.613820720813225,
+      "grad_norm": 0.06245681270956993,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 132585
+    },
+    {
+      "epoch": 13.614128760653044,
+      "grad_norm": 0.055418264120817184,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 132588
+    },
+    {
+      "epoch": 13.614436800492864,
+      "grad_norm": 0.04728015884757042,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 132591
+    },
+    {
+      "epoch": 13.614744840332683,
+      "grad_norm": 0.052111949771642685,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 132594
+    },
+    {
+      "epoch": 13.615052880172502,
+      "grad_norm": 0.09712249040603638,
+      "learning_rate": 0.01,
+      "loss": 1.9122,
+      "step": 132597
+    },
+    {
+      "epoch": 13.615360920012321,
+      "grad_norm": 0.07532146573066711,
+      "learning_rate": 0.01,
+      "loss": 1.9101,
+      "step": 132600
+    },
+    {
+      "epoch": 13.61566895985214,
+      "grad_norm": 0.065547876060009,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 132603
+    },
+    {
+      "epoch": 13.61597699969196,
+      "grad_norm": 0.04966482147574425,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 132606
+    },
+    {
+      "epoch": 13.61628503953178,
+      "grad_norm": 0.05835342779755592,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 132609
+    },
+    {
+      "epoch": 13.616593079371599,
+      "grad_norm": 0.05246637016534805,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 132612
+    },
+    {
+      "epoch": 13.616901119211418,
+      "grad_norm": 0.11848752200603485,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 132615
+    },
+    {
+      "epoch": 13.617209159051237,
+      "grad_norm": 0.10582973808050156,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 132618
+    },
+    {
+      "epoch": 13.617517198891056,
+      "grad_norm": 0.047501660883426666,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 132621
+    },
+    {
+      "epoch": 13.617825238730877,
+      "grad_norm": 0.029313301667571068,
+      "learning_rate": 0.01,
+      "loss": 1.9037,
+      "step": 132624
+    },
+    {
+      "epoch": 13.618133278570696,
+      "grad_norm": 0.05646532401442528,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 132627
+    },
+    {
+      "epoch": 13.618441318410515,
+      "grad_norm": 0.08745972067117691,
+      "learning_rate": 0.01,
+      "loss": 1.9522,
+      "step": 132630
+    },
+    {
+      "epoch": 13.618749358250334,
+      "grad_norm": 0.07873401045799255,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 132633
+    },
+    {
+      "epoch": 13.619057398090153,
+      "grad_norm": 0.1013382226228714,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 132636
+    },
+    {
+      "epoch": 13.619365437929972,
+      "grad_norm": 0.08139097690582275,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 132639
+    },
+    {
+      "epoch": 13.619673477769792,
+      "grad_norm": 0.07273190468549728,
+      "learning_rate": 0.01,
+      "loss": 1.9402,
+      "step": 132642
+    },
+    {
+      "epoch": 13.619981517609611,
+      "grad_norm": 0.05905170738697052,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 132645
+    },
+    {
+      "epoch": 13.62028955744943,
+      "grad_norm": 0.048046860843896866,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 132648
+    },
+    {
+      "epoch": 13.62059759728925,
+      "grad_norm": 0.041154894977808,
+      "learning_rate": 0.01,
+      "loss": 1.9427,
+      "step": 132651
+    },
+    {
+      "epoch": 13.620905637129068,
+      "grad_norm": 0.07600963860750198,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 132654
+    },
+    {
+      "epoch": 13.621213676968887,
+      "grad_norm": 0.08023431152105331,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 132657
+    },
+    {
+      "epoch": 13.621521716808708,
+      "grad_norm": 0.04688793420791626,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 132660
+    },
+    {
+      "epoch": 13.621829756648527,
+      "grad_norm": 0.1276724636554718,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 132663
+    },
+    {
+      "epoch": 13.622137796488346,
+      "grad_norm": 0.15639615058898926,
+      "learning_rate": 0.01,
+      "loss": 1.8968,
+      "step": 132666
+    },
+    {
+      "epoch": 13.622445836328165,
+      "grad_norm": 0.06842445582151413,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 132669
+    },
+    {
+      "epoch": 13.622753876167984,
+      "grad_norm": 0.109409861266613,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 132672
+    },
+    {
+      "epoch": 13.623061916007803,
+      "grad_norm": 0.03898303210735321,
+      "learning_rate": 0.01,
+      "loss": 1.9282,
+      "step": 132675
+    },
+    {
+      "epoch": 13.623369955847624,
+      "grad_norm": 0.0358307883143425,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 132678
+    },
+    {
+      "epoch": 13.623677995687443,
+      "grad_norm": 0.04396601766347885,
+      "learning_rate": 0.01,
+      "loss": 1.8991,
+      "step": 132681
+    },
+    {
+      "epoch": 13.623986035527262,
+      "grad_norm": 0.04737754911184311,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 132684
+    },
+    {
+      "epoch": 13.62429407536708,
+      "grad_norm": 0.049730800092220306,
+      "learning_rate": 0.01,
+      "loss": 1.9068,
+      "step": 132687
+    },
+    {
+      "epoch": 13.6246021152069,
+      "grad_norm": 0.033360932022333145,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 132690
+    },
+    {
+      "epoch": 13.62491015504672,
+      "grad_norm": 0.048622120171785355,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 132693
+    },
+    {
+      "epoch": 13.62521819488654,
+      "grad_norm": 0.05498216673731804,
+      "learning_rate": 0.01,
+      "loss": 1.9413,
+      "step": 132696
+    },
+    {
+      "epoch": 13.625526234726358,
+      "grad_norm": 0.06299851089715958,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 132699
+    },
+    {
+      "epoch": 13.625834274566177,
+      "grad_norm": 0.04300703480839729,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 132702
+    },
+    {
+      "epoch": 13.626142314405996,
+      "grad_norm": 0.03698214516043663,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 132705
+    },
+    {
+      "epoch": 13.626450354245815,
+      "grad_norm": 0.11164549738168716,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 132708
+    },
+    {
+      "epoch": 13.626758394085636,
+      "grad_norm": 0.046303655952215195,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 132711
+    },
+    {
+      "epoch": 13.627066433925455,
+      "grad_norm": 0.03596416115760803,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 132714
+    },
+    {
+      "epoch": 13.627374473765274,
+      "grad_norm": 0.030492739751935005,
+      "learning_rate": 0.01,
+      "loss": 1.9068,
+      "step": 132717
+    },
+    {
+      "epoch": 13.627682513605093,
+      "grad_norm": 0.05169845372438431,
+      "learning_rate": 0.01,
+      "loss": 1.895,
+      "step": 132720
+    },
+    {
+      "epoch": 13.627990553444912,
+      "grad_norm": 0.12154167890548706,
+      "learning_rate": 0.01,
+      "loss": 1.89,
+      "step": 132723
+    },
+    {
+      "epoch": 13.62829859328473,
+      "grad_norm": 0.05729279667139053,
+      "learning_rate": 0.01,
+      "loss": 1.899,
+      "step": 132726
+    },
+    {
+      "epoch": 13.628606633124551,
+      "grad_norm": 0.06351488828659058,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 132729
+    },
+    {
+      "epoch": 13.62891467296437,
+      "grad_norm": 0.10499323159456253,
+      "learning_rate": 0.01,
+      "loss": 1.9335,
+      "step": 132732
+    },
+    {
+      "epoch": 13.62922271280419,
+      "grad_norm": 0.1271558701992035,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 132735
+    },
+    {
+      "epoch": 13.629530752644008,
+      "grad_norm": 0.050670795142650604,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 132738
+    },
+    {
+      "epoch": 13.629838792483827,
+      "grad_norm": 0.08118502795696259,
+      "learning_rate": 0.01,
+      "loss": 1.9393,
+      "step": 132741
+    },
+    {
+      "epoch": 13.630146832323646,
+      "grad_norm": 0.04269060492515564,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 132744
+    },
+    {
+      "epoch": 13.630454872163467,
+      "grad_norm": 0.04747249186038971,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 132747
+    },
+    {
+      "epoch": 13.630762912003286,
+      "grad_norm": 0.035774290561676025,
+      "learning_rate": 0.01,
+      "loss": 1.8916,
+      "step": 132750
+    },
+    {
+      "epoch": 13.631070951843105,
+      "grad_norm": 0.10071831196546555,
+      "learning_rate": 0.01,
+      "loss": 1.8919,
+      "step": 132753
+    },
+    {
+      "epoch": 13.631378991682924,
+      "grad_norm": 0.07193715125322342,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 132756
+    },
+    {
+      "epoch": 13.631687031522743,
+      "grad_norm": 0.11618957668542862,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 132759
+    },
+    {
+      "epoch": 13.631995071362564,
+      "grad_norm": 0.042826756834983826,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 132762
+    },
+    {
+      "epoch": 13.632303111202383,
+      "grad_norm": 0.13055366277694702,
+      "learning_rate": 0.01,
+      "loss": 1.8996,
+      "step": 132765
+    },
+    {
+      "epoch": 13.632611151042202,
+      "grad_norm": 0.03490445390343666,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 132768
+    },
+    {
+      "epoch": 13.63291919088202,
+      "grad_norm": 0.06310755014419556,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 132771
+    },
+    {
+      "epoch": 13.63322723072184,
+      "grad_norm": 0.03780167177319527,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 132774
+    },
+    {
+      "epoch": 13.633535270561659,
+      "grad_norm": 0.04889899864792824,
+      "learning_rate": 0.01,
+      "loss": 1.8903,
+      "step": 132777
+    },
+    {
+      "epoch": 13.63384331040148,
+      "grad_norm": 0.11017850041389465,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 132780
+    },
+    {
+      "epoch": 13.634151350241298,
+      "grad_norm": 0.12095385044813156,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 132783
+    },
+    {
+      "epoch": 13.634459390081117,
+      "grad_norm": 0.06787922233343124,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 132786
+    },
+    {
+      "epoch": 13.634767429920936,
+      "grad_norm": 0.11610075831413269,
+      "learning_rate": 0.01,
+      "loss": 1.9151,
+      "step": 132789
+    },
+    {
+      "epoch": 13.635075469760755,
+      "grad_norm": 0.06578746438026428,
+      "learning_rate": 0.01,
+      "loss": 1.9239,
+      "step": 132792
+    },
+    {
+      "epoch": 13.635383509600574,
+      "grad_norm": 0.06474732607603073,
+      "learning_rate": 0.01,
+      "loss": 1.8915,
+      "step": 132795
+    },
+    {
+      "epoch": 13.635691549440395,
+      "grad_norm": 0.07117334753274918,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 132798
+    },
+    {
+      "epoch": 13.635999589280214,
+      "grad_norm": 0.044577550143003464,
+      "learning_rate": 0.01,
+      "loss": 1.892,
+      "step": 132801
+    },
+    {
+      "epoch": 13.636307629120033,
+      "grad_norm": 0.05026458576321602,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 132804
+    },
+    {
+      "epoch": 13.636615668959852,
+      "grad_norm": 0.17279726266860962,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 132807
+    },
+    {
+      "epoch": 13.63692370879967,
+      "grad_norm": 0.04130767658352852,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 132810
+    },
+    {
+      "epoch": 13.63723174863949,
+      "grad_norm": 0.03637677803635597,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 132813
+    },
+    {
+      "epoch": 13.63753978847931,
+      "grad_norm": 0.035333890467882156,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 132816
+    },
+    {
+      "epoch": 13.63784782831913,
+      "grad_norm": 0.03597143664956093,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 132819
+    },
+    {
+      "epoch": 13.638155868158949,
+      "grad_norm": 0.03420756384730339,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 132822
+    },
+    {
+      "epoch": 13.638463907998768,
+      "grad_norm": 0.04127111658453941,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 132825
+    },
+    {
+      "epoch": 13.638771947838586,
+      "grad_norm": 0.06875152140855789,
+      "learning_rate": 0.01,
+      "loss": 1.899,
+      "step": 132828
+    },
+    {
+      "epoch": 13.639079987678407,
+      "grad_norm": 0.10021661221981049,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 132831
+    },
+    {
+      "epoch": 13.639388027518226,
+      "grad_norm": 0.061660315841436386,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 132834
+    },
+    {
+      "epoch": 13.639696067358045,
+      "grad_norm": 0.15007522702217102,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 132837
+    },
+    {
+      "epoch": 13.640004107197864,
+      "grad_norm": 0.11773552745580673,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 132840
+    },
+    {
+      "epoch": 13.640312147037683,
+      "grad_norm": 0.060646045953035355,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 132843
+    },
+    {
+      "epoch": 13.640620186877502,
+      "grad_norm": 0.041012249886989594,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 132846
+    },
+    {
+      "epoch": 13.640928226717323,
+      "grad_norm": 0.04168090596795082,
+      "learning_rate": 0.01,
+      "loss": 1.8934,
+      "step": 132849
+    },
+    {
+      "epoch": 13.641236266557142,
+      "grad_norm": 0.034505102783441544,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 132852
+    },
+    {
+      "epoch": 13.64154430639696,
+      "grad_norm": 0.03979596495628357,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 132855
+    },
+    {
+      "epoch": 13.64185234623678,
+      "grad_norm": 0.034931741654872894,
+      "learning_rate": 0.01,
+      "loss": 1.9398,
+      "step": 132858
+    },
+    {
+      "epoch": 13.642160386076599,
+      "grad_norm": 0.04210006445646286,
+      "learning_rate": 0.01,
+      "loss": 1.9303,
+      "step": 132861
+    },
+    {
+      "epoch": 13.642468425916418,
+      "grad_norm": 0.08866101503372192,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 132864
+    },
+    {
+      "epoch": 13.642776465756238,
+      "grad_norm": 0.05358554422855377,
+      "learning_rate": 0.01,
+      "loss": 1.8915,
+      "step": 132867
+    },
+    {
+      "epoch": 13.643084505596057,
+      "grad_norm": 0.12009415775537491,
+      "learning_rate": 0.01,
+      "loss": 1.9296,
+      "step": 132870
+    },
+    {
+      "epoch": 13.643392545435876,
+      "grad_norm": 0.04405541718006134,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 132873
+    },
+    {
+      "epoch": 13.643700585275695,
+      "grad_norm": 0.0976574495434761,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 132876
+    },
+    {
+      "epoch": 13.644008625115514,
+      "grad_norm": 0.08781211823225021,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 132879
+    },
+    {
+      "epoch": 13.644316664955333,
+      "grad_norm": 0.047971855849027634,
+      "learning_rate": 0.01,
+      "loss": 1.8868,
+      "step": 132882
+    },
+    {
+      "epoch": 13.644624704795154,
+      "grad_norm": 0.06846671551465988,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 132885
+    },
+    {
+      "epoch": 13.644932744634973,
+      "grad_norm": 0.0916459709405899,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 132888
+    },
+    {
+      "epoch": 13.645240784474792,
+      "grad_norm": 0.047391489148139954,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 132891
+    },
+    {
+      "epoch": 13.645548824314611,
+      "grad_norm": 0.04784681648015976,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 132894
+    },
+    {
+      "epoch": 13.64585686415443,
+      "grad_norm": 0.07404235750436783,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 132897
+    },
+    {
+      "epoch": 13.64616490399425,
+      "grad_norm": 0.08336576074361801,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 132900
+    },
+    {
+      "epoch": 13.64647294383407,
+      "grad_norm": 0.04001960530877113,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 132903
+    },
+    {
+      "epoch": 13.646780983673889,
+      "grad_norm": 0.08146063983440399,
+      "learning_rate": 0.01,
+      "loss": 1.9075,
+      "step": 132906
+    },
+    {
+      "epoch": 13.647089023513708,
+      "grad_norm": 0.054827526211738586,
+      "learning_rate": 0.01,
+      "loss": 1.9401,
+      "step": 132909
+    },
+    {
+      "epoch": 13.647397063353527,
+      "grad_norm": 0.09309622645378113,
+      "learning_rate": 0.01,
+      "loss": 1.9447,
+      "step": 132912
+    },
+    {
+      "epoch": 13.647705103193346,
+      "grad_norm": 0.07018008083105087,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 132915
+    },
+    {
+      "epoch": 13.648013143033166,
+      "grad_norm": 0.13175293803215027,
+      "learning_rate": 0.01,
+      "loss": 1.8875,
+      "step": 132918
+    },
+    {
+      "epoch": 13.648321182872985,
+      "grad_norm": 0.04910729452967644,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 132921
+    },
+    {
+      "epoch": 13.648629222712804,
+      "grad_norm": 0.034327588975429535,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 132924
+    },
+    {
+      "epoch": 13.648937262552623,
+      "grad_norm": 0.11465651541948318,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 132927
+    },
+    {
+      "epoch": 13.649245302392442,
+      "grad_norm": 0.05895157903432846,
+      "learning_rate": 0.01,
+      "loss": 1.9065,
+      "step": 132930
+    },
+    {
+      "epoch": 13.649553342232261,
+      "grad_norm": 0.04770620912313461,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 132933
+    },
+    {
+      "epoch": 13.649861382072082,
+      "grad_norm": 0.04724672809243202,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 132936
+    },
+    {
+      "epoch": 13.650169421911901,
+      "grad_norm": 0.03714083135128021,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 132939
+    },
+    {
+      "epoch": 13.65047746175172,
+      "grad_norm": 0.05976714566349983,
+      "learning_rate": 0.01,
+      "loss": 1.9539,
+      "step": 132942
+    },
+    {
+      "epoch": 13.650785501591539,
+      "grad_norm": 0.05874619260430336,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 132945
+    },
+    {
+      "epoch": 13.651093541431358,
+      "grad_norm": 0.0507487878203392,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 132948
+    },
+    {
+      "epoch": 13.651401581271177,
+      "grad_norm": 0.05467512086033821,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 132951
+    },
+    {
+      "epoch": 13.651709621110998,
+      "grad_norm": 0.04137303680181503,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 132954
+    },
+    {
+      "epoch": 13.652017660950817,
+      "grad_norm": 0.08169732987880707,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 132957
+    },
+    {
+      "epoch": 13.652325700790636,
+      "grad_norm": 0.04622470214962959,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 132960
+    },
+    {
+      "epoch": 13.652633740630455,
+      "grad_norm": 0.07478762418031693,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 132963
+    },
+    {
+      "epoch": 13.652941780470274,
+      "grad_norm": 0.07834535092115402,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 132966
+    },
+    {
+      "epoch": 13.653249820310094,
+      "grad_norm": 0.08142905682325363,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 132969
+    },
+    {
+      "epoch": 13.653557860149913,
+      "grad_norm": 0.04494451731443405,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 132972
+    },
+    {
+      "epoch": 13.653865899989732,
+      "grad_norm": 0.04109867662191391,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 132975
+    },
+    {
+      "epoch": 13.654173939829551,
+      "grad_norm": 0.03387368842959404,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 132978
+    },
+    {
+      "epoch": 13.65448197966937,
+      "grad_norm": 0.04144396260380745,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 132981
+    },
+    {
+      "epoch": 13.65479001950919,
+      "grad_norm": 0.06484666466712952,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 132984
+    },
+    {
+      "epoch": 13.65509805934901,
+      "grad_norm": 0.10235686600208282,
+      "learning_rate": 0.01,
+      "loss": 1.9512,
+      "step": 132987
+    },
+    {
+      "epoch": 13.655406099188829,
+      "grad_norm": 0.10045121610164642,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 132990
+    },
+    {
+      "epoch": 13.655714139028648,
+      "grad_norm": 0.03303813189268112,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 132993
+    },
+    {
+      "epoch": 13.656022178868467,
+      "grad_norm": 0.04118012264370918,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 132996
+    },
+    {
+      "epoch": 13.656330218708286,
+      "grad_norm": 0.11215034127235413,
+      "learning_rate": 0.01,
+      "loss": 1.9001,
+      "step": 132999
+    },
+    {
+      "epoch": 13.656638258548105,
+      "grad_norm": 0.09077531099319458,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 133002
+    },
+    {
+      "epoch": 13.656946298387926,
+      "grad_norm": 0.08898884803056717,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 133005
+    },
+    {
+      "epoch": 13.657254338227744,
+      "grad_norm": 0.0633959025144577,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 133008
+    },
+    {
+      "epoch": 13.657562378067563,
+      "grad_norm": 0.05542381852865219,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 133011
+    },
+    {
+      "epoch": 13.657870417907382,
+      "grad_norm": 0.035190388560295105,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 133014
+    },
+    {
+      "epoch": 13.658178457747201,
+      "grad_norm": 0.038136593997478485,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 133017
+    },
+    {
+      "epoch": 13.65848649758702,
+      "grad_norm": 0.03591368719935417,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 133020
+    },
+    {
+      "epoch": 13.658794537426841,
+      "grad_norm": 0.07388196885585785,
+      "learning_rate": 0.01,
+      "loss": 1.891,
+      "step": 133023
+    },
+    {
+      "epoch": 13.65910257726666,
+      "grad_norm": 0.05979306250810623,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 133026
+    },
+    {
+      "epoch": 13.659410617106479,
+      "grad_norm": 0.04662696644663811,
+      "learning_rate": 0.01,
+      "loss": 1.9076,
+      "step": 133029
+    },
+    {
+      "epoch": 13.659718656946298,
+      "grad_norm": 0.042734142392873764,
+      "learning_rate": 0.01,
+      "loss": 1.898,
+      "step": 133032
+    },
+    {
+      "epoch": 13.660026696786117,
+      "grad_norm": 0.04746732488274574,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 133035
+    },
+    {
+      "epoch": 13.660334736625938,
+      "grad_norm": 0.036459729075431824,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 133038
+    },
+    {
+      "epoch": 13.660642776465757,
+      "grad_norm": 0.10208970308303833,
+      "learning_rate": 0.01,
+      "loss": 1.9011,
+      "step": 133041
+    },
+    {
+      "epoch": 13.660950816305576,
+      "grad_norm": 0.11507239937782288,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 133044
+    },
+    {
+      "epoch": 13.661258856145395,
+      "grad_norm": 0.08767065405845642,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 133047
+    },
+    {
+      "epoch": 13.661566895985214,
+      "grad_norm": 0.04055484011769295,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 133050
+    },
+    {
+      "epoch": 13.661874935825033,
+      "grad_norm": 0.04550919681787491,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 133053
+    },
+    {
+      "epoch": 13.662182975664853,
+      "grad_norm": 0.03845689073204994,
+      "learning_rate": 0.01,
+      "loss": 1.9369,
+      "step": 133056
+    },
+    {
+      "epoch": 13.662491015504672,
+      "grad_norm": 0.03854257985949516,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 133059
+    },
+    {
+      "epoch": 13.662799055344491,
+      "grad_norm": 0.043578118085861206,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 133062
+    },
+    {
+      "epoch": 13.66310709518431,
+      "grad_norm": 0.055198997259140015,
+      "learning_rate": 0.01,
+      "loss": 1.9368,
+      "step": 133065
+    },
+    {
+      "epoch": 13.66341513502413,
+      "grad_norm": 0.11246196925640106,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 133068
+    },
+    {
+      "epoch": 13.663723174863948,
+      "grad_norm": 0.06615752726793289,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 133071
+    },
+    {
+      "epoch": 13.664031214703769,
+      "grad_norm": 0.14438828825950623,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 133074
+    },
+    {
+      "epoch": 13.664339254543588,
+      "grad_norm": 0.07749340683221817,
+      "learning_rate": 0.01,
+      "loss": 1.8954,
+      "step": 133077
+    },
+    {
+      "epoch": 13.664647294383407,
+      "grad_norm": 0.09240418672561646,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 133080
+    },
+    {
+      "epoch": 13.664955334223226,
+      "grad_norm": 0.044705409556627274,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 133083
+    },
+    {
+      "epoch": 13.665263374063045,
+      "grad_norm": 0.041713062673807144,
+      "learning_rate": 0.01,
+      "loss": 1.9059,
+      "step": 133086
+    },
+    {
+      "epoch": 13.665571413902864,
+      "grad_norm": 0.052542347460985184,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 133089
+    },
+    {
+      "epoch": 13.665879453742685,
+      "grad_norm": 0.044348783791065216,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 133092
+    },
+    {
+      "epoch": 13.666187493582504,
+      "grad_norm": 0.041203223168849945,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 133095
+    },
+    {
+      "epoch": 13.666495533422323,
+      "grad_norm": 0.06860839575529099,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 133098
+    },
+    {
+      "epoch": 13.666803573262142,
+      "grad_norm": 0.11049733310937881,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 133101
+    },
+    {
+      "epoch": 13.66711161310196,
+      "grad_norm": 0.037917185574769974,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 133104
+    },
+    {
+      "epoch": 13.667419652941781,
+      "grad_norm": 0.11612064391374588,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 133107
+    },
+    {
+      "epoch": 13.6677276927816,
+      "grad_norm": 0.12070486694574356,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 133110
+    },
+    {
+      "epoch": 13.66803573262142,
+      "grad_norm": 0.12802445888519287,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 133113
+    },
+    {
+      "epoch": 13.668343772461238,
+      "grad_norm": 0.11320973932743073,
+      "learning_rate": 0.01,
+      "loss": 1.9696,
+      "step": 133116
+    },
+    {
+      "epoch": 13.668651812301057,
+      "grad_norm": 0.08258039504289627,
+      "learning_rate": 0.01,
+      "loss": 1.9638,
+      "step": 133119
+    },
+    {
+      "epoch": 13.668959852140876,
+      "grad_norm": 0.08377382904291153,
+      "learning_rate": 0.01,
+      "loss": 1.8929,
+      "step": 133122
+    },
+    {
+      "epoch": 13.669267891980697,
+      "grad_norm": 0.056055765599012375,
+      "learning_rate": 0.01,
+      "loss": 1.9001,
+      "step": 133125
+    },
+    {
+      "epoch": 13.669575931820516,
+      "grad_norm": 0.039423272013664246,
+      "learning_rate": 0.01,
+      "loss": 1.9359,
+      "step": 133128
+    },
+    {
+      "epoch": 13.669883971660335,
+      "grad_norm": 0.040388353168964386,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 133131
+    },
+    {
+      "epoch": 13.670192011500154,
+      "grad_norm": 0.07180473208427429,
+      "learning_rate": 0.01,
+      "loss": 1.8915,
+      "step": 133134
+    },
+    {
+      "epoch": 13.670500051339973,
+      "grad_norm": 0.05298145115375519,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 133137
+    },
+    {
+      "epoch": 13.670808091179792,
+      "grad_norm": 0.09174008667469025,
+      "learning_rate": 0.01,
+      "loss": 1.9242,
+      "step": 133140
+    },
+    {
+      "epoch": 13.671116131019613,
+      "grad_norm": 0.06148224323987961,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 133143
+    },
+    {
+      "epoch": 13.671424170859432,
+      "grad_norm": 0.10041771829128265,
+      "learning_rate": 0.01,
+      "loss": 1.9375,
+      "step": 133146
+    },
+    {
+      "epoch": 13.67173221069925,
+      "grad_norm": 0.03670705854892731,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 133149
+    },
+    {
+      "epoch": 13.67204025053907,
+      "grad_norm": 0.05213388800621033,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 133152
+    },
+    {
+      "epoch": 13.672348290378888,
+      "grad_norm": 0.07948003709316254,
+      "learning_rate": 0.01,
+      "loss": 1.9319,
+      "step": 133155
+    },
+    {
+      "epoch": 13.672656330218707,
+      "grad_norm": 0.05891106277704239,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 133158
+    },
+    {
+      "epoch": 13.672964370058528,
+      "grad_norm": 0.03221432864665985,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 133161
+    },
+    {
+      "epoch": 13.673272409898347,
+      "grad_norm": 0.045262858271598816,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 133164
+    },
+    {
+      "epoch": 13.673580449738166,
+      "grad_norm": 0.05456344783306122,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 133167
+    },
+    {
+      "epoch": 13.673888489577985,
+      "grad_norm": 0.09798543900251389,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 133170
+    },
+    {
+      "epoch": 13.674196529417804,
+      "grad_norm": 0.06923268735408783,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 133173
+    },
+    {
+      "epoch": 13.674504569257625,
+      "grad_norm": 0.038926299661397934,
+      "learning_rate": 0.01,
+      "loss": 1.9065,
+      "step": 133176
+    },
+    {
+      "epoch": 13.674812609097444,
+      "grad_norm": 0.03782898187637329,
+      "learning_rate": 0.01,
+      "loss": 1.9,
+      "step": 133179
+    },
+    {
+      "epoch": 13.675120648937263,
+      "grad_norm": 0.03196824714541435,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 133182
+    },
+    {
+      "epoch": 13.675428688777082,
+      "grad_norm": 0.049436815083026886,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 133185
+    },
+    {
+      "epoch": 13.6757367286169,
+      "grad_norm": 0.11374814808368683,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 133188
+    },
+    {
+      "epoch": 13.67604476845672,
+      "grad_norm": 0.07520898431539536,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 133191
+    },
+    {
+      "epoch": 13.67635280829654,
+      "grad_norm": 0.08074496686458588,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 133194
+    },
+    {
+      "epoch": 13.67666084813636,
+      "grad_norm": 0.046022992581129074,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 133197
+    },
+    {
+      "epoch": 13.676968887976178,
+      "grad_norm": 0.0465850867331028,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 133200
+    },
+    {
+      "epoch": 13.677276927815997,
+      "grad_norm": 0.046240806579589844,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 133203
+    },
+    {
+      "epoch": 13.677584967655816,
+      "grad_norm": 0.04726431891322136,
+      "learning_rate": 0.01,
+      "loss": 1.8947,
+      "step": 133206
+    },
+    {
+      "epoch": 13.677893007495635,
+      "grad_norm": 0.08443351835012436,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 133209
+    },
+    {
+      "epoch": 13.678201047335456,
+      "grad_norm": 0.03419280797243118,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 133212
+    },
+    {
+      "epoch": 13.678509087175275,
+      "grad_norm": 0.043816953897476196,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 133215
+    },
+    {
+      "epoch": 13.678817127015094,
+      "grad_norm": 0.046974197030067444,
+      "learning_rate": 0.01,
+      "loss": 1.8917,
+      "step": 133218
+    },
+    {
+      "epoch": 13.679125166854913,
+      "grad_norm": 0.033940639346838,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 133221
+    },
+    {
+      "epoch": 13.679433206694732,
+      "grad_norm": 0.11976716667413712,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 133224
+    },
+    {
+      "epoch": 13.679741246534551,
+      "grad_norm": 0.06758849322795868,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 133227
+    },
+    {
+      "epoch": 13.680049286374372,
+      "grad_norm": 0.0950511246919632,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 133230
+    },
+    {
+      "epoch": 13.68035732621419,
+      "grad_norm": 0.1029929667711258,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 133233
+    },
+    {
+      "epoch": 13.68066536605401,
+      "grad_norm": 0.08932224661111832,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 133236
+    },
+    {
+      "epoch": 13.680973405893829,
+      "grad_norm": 0.05903936177492142,
+      "learning_rate": 0.01,
+      "loss": 1.9468,
+      "step": 133239
+    },
+    {
+      "epoch": 13.681281445733648,
+      "grad_norm": 0.045178890228271484,
+      "learning_rate": 0.01,
+      "loss": 1.9327,
+      "step": 133242
+    },
+    {
+      "epoch": 13.681589485573468,
+      "grad_norm": 0.03455664590001106,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 133245
+    },
+    {
+      "epoch": 13.681897525413287,
+      "grad_norm": 0.12724003195762634,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 133248
+    },
+    {
+      "epoch": 13.682205565253106,
+      "grad_norm": 0.04221659526228905,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 133251
+    },
+    {
+      "epoch": 13.682513605092925,
+      "grad_norm": 0.050626445561647415,
+      "learning_rate": 0.01,
+      "loss": 1.9016,
+      "step": 133254
+    },
+    {
+      "epoch": 13.682821644932744,
+      "grad_norm": 0.041968297213315964,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 133257
+    },
+    {
+      "epoch": 13.683129684772563,
+      "grad_norm": 0.05025589093565941,
+      "learning_rate": 0.01,
+      "loss": 1.9387,
+      "step": 133260
+    },
+    {
+      "epoch": 13.683437724612384,
+      "grad_norm": 0.03889334201812744,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 133263
+    },
+    {
+      "epoch": 13.683745764452203,
+      "grad_norm": 0.05425889417529106,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 133266
+    },
+    {
+      "epoch": 13.684053804292022,
+      "grad_norm": 0.10900388658046722,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 133269
+    },
+    {
+      "epoch": 13.68436184413184,
+      "grad_norm": 0.07030241936445236,
+      "learning_rate": 0.01,
+      "loss": 1.8929,
+      "step": 133272
+    },
+    {
+      "epoch": 13.68466988397166,
+      "grad_norm": 0.15559154748916626,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 133275
+    },
+    {
+      "epoch": 13.684977923811479,
+      "grad_norm": 0.1169857382774353,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 133278
+    },
+    {
+      "epoch": 13.6852859636513,
+      "grad_norm": 0.10924411565065384,
+      "learning_rate": 0.01,
+      "loss": 1.8964,
+      "step": 133281
+    },
+    {
+      "epoch": 13.685594003491119,
+      "grad_norm": 0.08389702439308167,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 133284
+    },
+    {
+      "epoch": 13.685902043330938,
+      "grad_norm": 0.06294015049934387,
+      "learning_rate": 0.01,
+      "loss": 1.896,
+      "step": 133287
+    },
+    {
+      "epoch": 13.686210083170756,
+      "grad_norm": 0.07205908000469208,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 133290
+    },
+    {
+      "epoch": 13.686518123010575,
+      "grad_norm": 0.08009463548660278,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 133293
+    },
+    {
+      "epoch": 13.686826162850394,
+      "grad_norm": 0.08380520343780518,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 133296
+    },
+    {
+      "epoch": 13.687134202690215,
+      "grad_norm": 0.04630572721362114,
+      "learning_rate": 0.01,
+      "loss": 1.8968,
+      "step": 133299
+    },
+    {
+      "epoch": 13.687442242530034,
+      "grad_norm": 0.07967095077037811,
+      "learning_rate": 0.01,
+      "loss": 1.8965,
+      "step": 133302
+    },
+    {
+      "epoch": 13.687750282369853,
+      "grad_norm": 0.03927162289619446,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 133305
+    },
+    {
+      "epoch": 13.688058322209672,
+      "grad_norm": 0.054481033235788345,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 133308
+    },
+    {
+      "epoch": 13.688366362049491,
+      "grad_norm": 0.07706627249717712,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 133311
+    },
+    {
+      "epoch": 13.688674401889312,
+      "grad_norm": 0.0670505166053772,
+      "learning_rate": 0.01,
+      "loss": 1.893,
+      "step": 133314
+    },
+    {
+      "epoch": 13.68898244172913,
+      "grad_norm": 0.12645144760608673,
+      "learning_rate": 0.01,
+      "loss": 1.9083,
+      "step": 133317
+    },
+    {
+      "epoch": 13.68929048156895,
+      "grad_norm": 0.07617560774087906,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 133320
+    },
+    {
+      "epoch": 13.689598521408769,
+      "grad_norm": 0.07287514954805374,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 133323
+    },
+    {
+      "epoch": 13.689906561248588,
+      "grad_norm": 0.048952165991067886,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 133326
+    },
+    {
+      "epoch": 13.690214601088407,
+      "grad_norm": 0.041913677006959915,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 133329
+    },
+    {
+      "epoch": 13.690522640928227,
+      "grad_norm": 0.10903766751289368,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 133332
+    },
+    {
+      "epoch": 13.690830680768046,
+      "grad_norm": 0.04150333255529404,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 133335
+    },
+    {
+      "epoch": 13.691138720607865,
+      "grad_norm": 0.06722831726074219,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 133338
+    },
+    {
+      "epoch": 13.691446760447684,
+      "grad_norm": 0.12879085540771484,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 133341
+    },
+    {
+      "epoch": 13.691754800287503,
+      "grad_norm": 0.0931577980518341,
+      "learning_rate": 0.01,
+      "loss": 1.906,
+      "step": 133344
+    },
+    {
+      "epoch": 13.692062840127322,
+      "grad_norm": 0.039442114531993866,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 133347
+    },
+    {
+      "epoch": 13.692370879967143,
+      "grad_norm": 0.06459635496139526,
+      "learning_rate": 0.01,
+      "loss": 1.9453,
+      "step": 133350
+    },
+    {
+      "epoch": 13.692678919806962,
+      "grad_norm": 0.038339994847774506,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 133353
+    },
+    {
+      "epoch": 13.692986959646781,
+      "grad_norm": 0.09138669818639755,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 133356
+    },
+    {
+      "epoch": 13.6932949994866,
+      "grad_norm": 0.05956611409783363,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 133359
+    },
+    {
+      "epoch": 13.693603039326419,
+      "grad_norm": 0.09686420857906342,
+      "learning_rate": 0.01,
+      "loss": 1.9515,
+      "step": 133362
+    },
+    {
+      "epoch": 13.693911079166238,
+      "grad_norm": 0.04909735918045044,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 133365
+    },
+    {
+      "epoch": 13.694219119006059,
+      "grad_norm": 0.06500578671693802,
+      "learning_rate": 0.01,
+      "loss": 1.9389,
+      "step": 133368
+    },
+    {
+      "epoch": 13.694527158845878,
+      "grad_norm": 0.10369669646024704,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 133371
+    },
+    {
+      "epoch": 13.694835198685697,
+      "grad_norm": 0.10991252958774567,
+      "learning_rate": 0.01,
+      "loss": 1.952,
+      "step": 133374
+    },
+    {
+      "epoch": 13.695143238525516,
+      "grad_norm": 0.05861903354525566,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 133377
+    },
+    {
+      "epoch": 13.695451278365335,
+      "grad_norm": 0.10776627063751221,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 133380
+    },
+    {
+      "epoch": 13.695759318205155,
+      "grad_norm": 0.10691935569047928,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 133383
+    },
+    {
+      "epoch": 13.696067358044974,
+      "grad_norm": 0.037835754454135895,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 133386
+    },
+    {
+      "epoch": 13.696375397884793,
+      "grad_norm": 0.04589299112558365,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 133389
+    },
+    {
+      "epoch": 13.696683437724612,
+      "grad_norm": 0.06498698890209198,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 133392
+    },
+    {
+      "epoch": 13.696991477564431,
+      "grad_norm": 0.08892109245061874,
+      "learning_rate": 0.01,
+      "loss": 1.896,
+      "step": 133395
+    },
+    {
+      "epoch": 13.69729951740425,
+      "grad_norm": 0.08194664120674133,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 133398
+    },
+    {
+      "epoch": 13.697607557244071,
+      "grad_norm": 0.05720951780676842,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 133401
+    },
+    {
+      "epoch": 13.69791559708389,
+      "grad_norm": 0.03802094981074333,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 133404
+    },
+    {
+      "epoch": 13.698223636923709,
+      "grad_norm": 0.05794553831219673,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 133407
+    },
+    {
+      "epoch": 13.698531676763528,
+      "grad_norm": 0.054694339632987976,
+      "learning_rate": 0.01,
+      "loss": 1.8983,
+      "step": 133410
+    },
+    {
+      "epoch": 13.698839716603347,
+      "grad_norm": 0.04657558724284172,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 133413
+    },
+    {
+      "epoch": 13.699147756443166,
+      "grad_norm": 0.058000653982162476,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 133416
+    },
+    {
+      "epoch": 13.699455796282987,
+      "grad_norm": 0.037103284150362015,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 133419
+    },
+    {
+      "epoch": 13.699763836122806,
+      "grad_norm": 0.07626280188560486,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 133422
+    },
+    {
+      "epoch": 13.700071875962625,
+      "grad_norm": 0.08235161006450653,
+      "learning_rate": 0.01,
+      "loss": 1.9092,
+      "step": 133425
+    },
+    {
+      "epoch": 13.700379915802444,
+      "grad_norm": 0.0640086978673935,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 133428
+    },
+    {
+      "epoch": 13.700687955642262,
+      "grad_norm": 0.04797493666410446,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 133431
+    },
+    {
+      "epoch": 13.700995995482081,
+      "grad_norm": 0.0327463373541832,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 133434
+    },
+    {
+      "epoch": 13.701304035321902,
+      "grad_norm": 0.032630857080221176,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 133437
+    },
+    {
+      "epoch": 13.701612075161721,
+      "grad_norm": 0.11865788698196411,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 133440
+    },
+    {
+      "epoch": 13.70192011500154,
+      "grad_norm": 0.06525228172540665,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 133443
+    },
+    {
+      "epoch": 13.70222815484136,
+      "grad_norm": 0.043781768530607224,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 133446
+    },
+    {
+      "epoch": 13.702536194681178,
+      "grad_norm": 0.08076925575733185,
+      "learning_rate": 0.01,
+      "loss": 1.9371,
+      "step": 133449
+    },
+    {
+      "epoch": 13.702844234520999,
+      "grad_norm": 0.04620203748345375,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 133452
+    },
+    {
+      "epoch": 13.703152274360818,
+      "grad_norm": 0.1338471621274948,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 133455
+    },
+    {
+      "epoch": 13.703460314200637,
+      "grad_norm": 0.07895182818174362,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 133458
+    },
+    {
+      "epoch": 13.703768354040456,
+      "grad_norm": 0.04356776177883148,
+      "learning_rate": 0.01,
+      "loss": 1.8885,
+      "step": 133461
+    },
+    {
+      "epoch": 13.704076393880275,
+      "grad_norm": 0.030977753922343254,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 133464
+    },
+    {
+      "epoch": 13.704384433720094,
+      "grad_norm": 0.044693734496831894,
+      "learning_rate": 0.01,
+      "loss": 1.9113,
+      "step": 133467
+    },
+    {
+      "epoch": 13.704692473559914,
+      "grad_norm": 0.05724677816033363,
+      "learning_rate": 0.01,
+      "loss": 1.8986,
+      "step": 133470
+    },
+    {
+      "epoch": 13.705000513399733,
+      "grad_norm": 0.03263508155941963,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 133473
+    },
+    {
+      "epoch": 13.705308553239552,
+      "grad_norm": 0.05749926716089249,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 133476
+    },
+    {
+      "epoch": 13.705616593079371,
+      "grad_norm": 0.10866115987300873,
+      "learning_rate": 0.01,
+      "loss": 1.8911,
+      "step": 133479
+    },
+    {
+      "epoch": 13.70592463291919,
+      "grad_norm": 0.08124980330467224,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 133482
+    },
+    {
+      "epoch": 13.70623267275901,
+      "grad_norm": 0.11043617129325867,
+      "learning_rate": 0.01,
+      "loss": 1.9039,
+      "step": 133485
+    },
+    {
+      "epoch": 13.70654071259883,
+      "grad_norm": 0.0830722451210022,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 133488
+    },
+    {
+      "epoch": 13.706848752438649,
+      "grad_norm": 0.05270704627037048,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 133491
+    },
+    {
+      "epoch": 13.707156792278468,
+      "grad_norm": 0.03868338465690613,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 133494
+    },
+    {
+      "epoch": 13.707464832118287,
+      "grad_norm": 0.036313995718955994,
+      "learning_rate": 0.01,
+      "loss": 1.9018,
+      "step": 133497
+    },
+    {
+      "epoch": 13.707772871958106,
+      "grad_norm": 0.12680625915527344,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 133500
+    },
+    {
+      "epoch": 13.708080911797925,
+      "grad_norm": 0.09202094376087189,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 133503
+    },
+    {
+      "epoch": 13.708388951637746,
+      "grad_norm": 0.05450097471475601,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 133506
+    },
+    {
+      "epoch": 13.708696991477565,
+      "grad_norm": 0.04995448887348175,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 133509
+    },
+    {
+      "epoch": 13.709005031317384,
+      "grad_norm": 0.10298550128936768,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 133512
+    },
+    {
+      "epoch": 13.709313071157203,
+      "grad_norm": 0.06571313738822937,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 133515
+    },
+    {
+      "epoch": 13.709621110997022,
+      "grad_norm": 0.10729604214429855,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 133518
+    },
+    {
+      "epoch": 13.709929150836842,
+      "grad_norm": 0.0360541008412838,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 133521
+    },
+    {
+      "epoch": 13.710237190676661,
+      "grad_norm": 0.08101950585842133,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 133524
+    },
+    {
+      "epoch": 13.71054523051648,
+      "grad_norm": 0.061308614909648895,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 133527
+    },
+    {
+      "epoch": 13.7108532703563,
+      "grad_norm": 0.0699058249592781,
+      "learning_rate": 0.01,
+      "loss": 1.9486,
+      "step": 133530
+    },
+    {
+      "epoch": 13.711161310196118,
+      "grad_norm": 0.04529229551553726,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 133533
+    },
+    {
+      "epoch": 13.711469350035937,
+      "grad_norm": 0.06092170998454094,
+      "learning_rate": 0.01,
+      "loss": 1.8909,
+      "step": 133536
+    },
+    {
+      "epoch": 13.711777389875758,
+      "grad_norm": 0.14121152460575104,
+      "learning_rate": 0.01,
+      "loss": 1.9439,
+      "step": 133539
+    },
+    {
+      "epoch": 13.712085429715577,
+      "grad_norm": 0.1851249486207962,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 133542
+    },
+    {
+      "epoch": 13.712393469555396,
+      "grad_norm": 0.17182426154613495,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 133545
+    },
+    {
+      "epoch": 13.712701509395215,
+      "grad_norm": 0.04837925732135773,
+      "learning_rate": 0.01,
+      "loss": 1.9581,
+      "step": 133548
+    },
+    {
+      "epoch": 13.713009549235034,
+      "grad_norm": 0.04337107390165329,
+      "learning_rate": 0.01,
+      "loss": 1.9174,
+      "step": 133551
+    },
+    {
+      "epoch": 13.713317589074853,
+      "grad_norm": 0.06445692479610443,
+      "learning_rate": 0.01,
+      "loss": 1.8934,
+      "step": 133554
+    },
+    {
+      "epoch": 13.713625628914674,
+      "grad_norm": 0.04748813062906265,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 133557
+    },
+    {
+      "epoch": 13.713933668754493,
+      "grad_norm": 0.10335930436849594,
+      "learning_rate": 0.01,
+      "loss": 1.8845,
+      "step": 133560
+    },
+    {
+      "epoch": 13.714241708594312,
+      "grad_norm": 0.04327281191945076,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 133563
+    },
+    {
+      "epoch": 13.71454974843413,
+      "grad_norm": 0.07165496051311493,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 133566
+    },
+    {
+      "epoch": 13.71485778827395,
+      "grad_norm": 0.03519191965460777,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 133569
+    },
+    {
+      "epoch": 13.715165828113768,
+      "grad_norm": 0.07492917776107788,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 133572
+    },
+    {
+      "epoch": 13.71547386795359,
+      "grad_norm": 0.07458905875682831,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 133575
+    },
+    {
+      "epoch": 13.715781907793408,
+      "grad_norm": 0.09317105263471603,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 133578
+    },
+    {
+      "epoch": 13.716089947633227,
+      "grad_norm": 0.06646846234798431,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 133581
+    },
+    {
+      "epoch": 13.716397987473046,
+      "grad_norm": 0.05613543465733528,
+      "learning_rate": 0.01,
+      "loss": 1.889,
+      "step": 133584
+    },
+    {
+      "epoch": 13.716706027312865,
+      "grad_norm": 0.096941739320755,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 133587
+    },
+    {
+      "epoch": 13.717014067152686,
+      "grad_norm": 0.05867409333586693,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 133590
+    },
+    {
+      "epoch": 13.717322106992505,
+      "grad_norm": 0.06186516210436821,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 133593
+    },
+    {
+      "epoch": 13.717630146832324,
+      "grad_norm": 0.035699065774679184,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 133596
+    },
+    {
+      "epoch": 13.717938186672143,
+      "grad_norm": 0.03967752680182457,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 133599
+    },
+    {
+      "epoch": 13.718246226511962,
+      "grad_norm": 0.05443970859050751,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 133602
+    },
+    {
+      "epoch": 13.71855426635178,
+      "grad_norm": 0.042695339769124985,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 133605
+    },
+    {
+      "epoch": 13.718862306191602,
+      "grad_norm": 0.12690339982509613,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 133608
+    },
+    {
+      "epoch": 13.71917034603142,
+      "grad_norm": 0.04124609753489494,
+      "learning_rate": 0.01,
+      "loss": 1.9415,
+      "step": 133611
+    },
+    {
+      "epoch": 13.71947838587124,
+      "grad_norm": 0.07924247533082962,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 133614
+    },
+    {
+      "epoch": 13.719786425711058,
+      "grad_norm": 0.06128375232219696,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 133617
+    },
+    {
+      "epoch": 13.720094465550877,
+      "grad_norm": 0.09470294415950775,
+      "learning_rate": 0.01,
+      "loss": 1.9042,
+      "step": 133620
+    },
+    {
+      "epoch": 13.720402505390696,
+      "grad_norm": 0.07386910170316696,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 133623
+    },
+    {
+      "epoch": 13.720710545230517,
+      "grad_norm": 0.07994164526462555,
+      "learning_rate": 0.01,
+      "loss": 1.9363,
+      "step": 133626
+    },
+    {
+      "epoch": 13.721018585070336,
+      "grad_norm": 0.040610119700431824,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 133629
+    },
+    {
+      "epoch": 13.721326624910155,
+      "grad_norm": 0.04326248541474342,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 133632
+    },
+    {
+      "epoch": 13.721634664749974,
+      "grad_norm": 0.09847462922334671,
+      "learning_rate": 0.01,
+      "loss": 1.8997,
+      "step": 133635
+    },
+    {
+      "epoch": 13.721942704589793,
+      "grad_norm": 0.09448638558387756,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 133638
+    },
+    {
+      "epoch": 13.722250744429612,
+      "grad_norm": 0.04267118126153946,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 133641
+    },
+    {
+      "epoch": 13.722558784269433,
+      "grad_norm": 0.03486855700612068,
+      "learning_rate": 0.01,
+      "loss": 1.945,
+      "step": 133644
+    },
+    {
+      "epoch": 13.722866824109252,
+      "grad_norm": 0.06310033798217773,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 133647
+    },
+    {
+      "epoch": 13.72317486394907,
+      "grad_norm": 0.08251602202653885,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 133650
+    },
+    {
+      "epoch": 13.72348290378889,
+      "grad_norm": 0.09197825193405151,
+      "learning_rate": 0.01,
+      "loss": 1.937,
+      "step": 133653
+    },
+    {
+      "epoch": 13.723790943628709,
+      "grad_norm": 0.05254195258021355,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 133656
+    },
+    {
+      "epoch": 13.72409898346853,
+      "grad_norm": 0.0615781806409359,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 133659
+    },
+    {
+      "epoch": 13.724407023308348,
+      "grad_norm": 0.09117575734853745,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 133662
+    },
+    {
+      "epoch": 13.724715063148167,
+      "grad_norm": 0.045430656522512436,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 133665
+    },
+    {
+      "epoch": 13.725023102987986,
+      "grad_norm": 0.03622528910636902,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 133668
+    },
+    {
+      "epoch": 13.725331142827805,
+      "grad_norm": 0.029216891154646873,
+      "learning_rate": 0.01,
+      "loss": 1.8828,
+      "step": 133671
+    },
+    {
+      "epoch": 13.725639182667624,
+      "grad_norm": 0.07972874492406845,
+      "learning_rate": 0.01,
+      "loss": 1.9593,
+      "step": 133674
+    },
+    {
+      "epoch": 13.725947222507445,
+      "grad_norm": 0.0797075554728508,
+      "learning_rate": 0.01,
+      "loss": 1.9403,
+      "step": 133677
+    },
+    {
+      "epoch": 13.726255262347264,
+      "grad_norm": 0.04298889636993408,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 133680
+    },
+    {
+      "epoch": 13.726563302187083,
+      "grad_norm": 0.04494164139032364,
+      "learning_rate": 0.01,
+      "loss": 1.9527,
+      "step": 133683
+    },
+    {
+      "epoch": 13.726871342026902,
+      "grad_norm": 0.04080686718225479,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 133686
+    },
+    {
+      "epoch": 13.727179381866721,
+      "grad_norm": 0.03086329810321331,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 133689
+    },
+    {
+      "epoch": 13.72748742170654,
+      "grad_norm": 0.03140528127551079,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 133692
+    },
+    {
+      "epoch": 13.72779546154636,
+      "grad_norm": 0.05109536647796631,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 133695
+    },
+    {
+      "epoch": 13.72810350138618,
+      "grad_norm": 0.10806011408567429,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 133698
+    },
+    {
+      "epoch": 13.728411541225999,
+      "grad_norm": 0.09576310962438583,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 133701
+    },
+    {
+      "epoch": 13.728719581065818,
+      "grad_norm": 0.08329522609710693,
+      "learning_rate": 0.01,
+      "loss": 1.9232,
+      "step": 133704
+    },
+    {
+      "epoch": 13.729027620905637,
+      "grad_norm": 0.10577517747879028,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 133707
+    },
+    {
+      "epoch": 13.729335660745456,
+      "grad_norm": 0.04874836280941963,
+      "learning_rate": 0.01,
+      "loss": 1.8959,
+      "step": 133710
+    },
+    {
+      "epoch": 13.729643700585276,
+      "grad_norm": 0.03710230812430382,
+      "learning_rate": 0.01,
+      "loss": 1.9462,
+      "step": 133713
+    },
+    {
+      "epoch": 13.729951740425095,
+      "grad_norm": 0.10409806668758392,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 133716
+    },
+    {
+      "epoch": 13.730259780264914,
+      "grad_norm": 0.14586621522903442,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 133719
+    },
+    {
+      "epoch": 13.730567820104733,
+      "grad_norm": 0.06513901054859161,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 133722
+    },
+    {
+      "epoch": 13.730875859944552,
+      "grad_norm": 0.0493028499186039,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 133725
+    },
+    {
+      "epoch": 13.731183899784373,
+      "grad_norm": 0.050208885222673416,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 133728
+    },
+    {
+      "epoch": 13.731491939624192,
+      "grad_norm": 0.041871726512908936,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 133731
+    },
+    {
+      "epoch": 13.73179997946401,
+      "grad_norm": 0.08905645459890366,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 133734
+    },
+    {
+      "epoch": 13.73210801930383,
+      "grad_norm": 0.11692611128091812,
+      "learning_rate": 0.01,
+      "loss": 1.9224,
+      "step": 133737
+    },
+    {
+      "epoch": 13.732416059143649,
+      "grad_norm": 0.032352473586797714,
+      "learning_rate": 0.01,
+      "loss": 1.8914,
+      "step": 133740
+    },
+    {
+      "epoch": 13.732724098983468,
+      "grad_norm": 0.12896639108657837,
+      "learning_rate": 0.01,
+      "loss": 1.9531,
+      "step": 133743
+    },
+    {
+      "epoch": 13.733032138823289,
+      "grad_norm": 0.11440527439117432,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 133746
+    },
+    {
+      "epoch": 13.733340178663108,
+      "grad_norm": 0.04453273490071297,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 133749
+    },
+    {
+      "epoch": 13.733648218502926,
+      "grad_norm": 0.053692251443862915,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 133752
+    },
+    {
+      "epoch": 13.733956258342745,
+      "grad_norm": 0.04075769707560539,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 133755
+    },
+    {
+      "epoch": 13.734264298182564,
+      "grad_norm": 0.06440185755491257,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 133758
+    },
+    {
+      "epoch": 13.734572338022383,
+      "grad_norm": 0.09642770141363144,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 133761
+    },
+    {
+      "epoch": 13.734880377862204,
+      "grad_norm": 0.07496191561222076,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 133764
+    },
+    {
+      "epoch": 13.735188417702023,
+      "grad_norm": 0.040042925626039505,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 133767
+    },
+    {
+      "epoch": 13.735496457541842,
+      "grad_norm": 0.050517965108156204,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 133770
+    },
+    {
+      "epoch": 13.735804497381661,
+      "grad_norm": 0.11992526799440384,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 133773
+    },
+    {
+      "epoch": 13.73611253722148,
+      "grad_norm": 0.030365869402885437,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 133776
+    },
+    {
+      "epoch": 13.736420577061299,
+      "grad_norm": 0.1195293739438057,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 133779
+    },
+    {
+      "epoch": 13.73672861690112,
+      "grad_norm": 0.10183779895305634,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 133782
+    },
+    {
+      "epoch": 13.737036656740939,
+      "grad_norm": 0.05504452809691429,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 133785
+    },
+    {
+      "epoch": 13.737344696580758,
+      "grad_norm": 0.07268170267343521,
+      "learning_rate": 0.01,
+      "loss": 1.9001,
+      "step": 133788
+    },
+    {
+      "epoch": 13.737652736420577,
+      "grad_norm": 0.04448287934064865,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 133791
+    },
+    {
+      "epoch": 13.737960776260396,
+      "grad_norm": 0.044859062880277634,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 133794
+    },
+    {
+      "epoch": 13.738268816100216,
+      "grad_norm": 0.04380045458674431,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 133797
+    },
+    {
+      "epoch": 13.738576855940035,
+      "grad_norm": 0.0548398494720459,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 133800
+    },
+    {
+      "epoch": 13.738884895779854,
+      "grad_norm": 0.0803338885307312,
+      "learning_rate": 0.01,
+      "loss": 1.9372,
+      "step": 133803
+    },
+    {
+      "epoch": 13.739192935619673,
+      "grad_norm": 0.06778112798929214,
+      "learning_rate": 0.01,
+      "loss": 1.9035,
+      "step": 133806
+    },
+    {
+      "epoch": 13.739500975459492,
+      "grad_norm": 0.052715003490448,
+      "learning_rate": 0.01,
+      "loss": 1.8893,
+      "step": 133809
+    },
+    {
+      "epoch": 13.739809015299311,
+      "grad_norm": 0.044049669057130814,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 133812
+    },
+    {
+      "epoch": 13.740117055139132,
+      "grad_norm": 0.04254237189888954,
+      "learning_rate": 0.01,
+      "loss": 1.8912,
+      "step": 133815
+    },
+    {
+      "epoch": 13.740425094978951,
+      "grad_norm": 0.0512581393122673,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 133818
+    },
+    {
+      "epoch": 13.74073313481877,
+      "grad_norm": 0.12065699696540833,
+      "learning_rate": 0.01,
+      "loss": 1.8961,
+      "step": 133821
+    },
+    {
+      "epoch": 13.741041174658589,
+      "grad_norm": 0.1085164025425911,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 133824
+    },
+    {
+      "epoch": 13.741349214498408,
+      "grad_norm": 0.12018072605133057,
+      "learning_rate": 0.01,
+      "loss": 1.9119,
+      "step": 133827
+    },
+    {
+      "epoch": 13.741657254338227,
+      "grad_norm": 0.060278408229351044,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 133830
+    },
+    {
+      "epoch": 13.741965294178048,
+      "grad_norm": 0.05038806051015854,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 133833
+    },
+    {
+      "epoch": 13.742273334017867,
+      "grad_norm": 0.07700009644031525,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 133836
+    },
+    {
+      "epoch": 13.742581373857686,
+      "grad_norm": 0.05274767056107521,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 133839
+    },
+    {
+      "epoch": 13.742889413697505,
+      "grad_norm": 0.0548916831612587,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 133842
+    },
+    {
+      "epoch": 13.743197453537324,
+      "grad_norm": 0.09337977319955826,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 133845
+    },
+    {
+      "epoch": 13.743505493377143,
+      "grad_norm": 0.08361978828907013,
+      "learning_rate": 0.01,
+      "loss": 1.8955,
+      "step": 133848
+    },
+    {
+      "epoch": 13.743813533216963,
+      "grad_norm": 0.07771879434585571,
+      "learning_rate": 0.01,
+      "loss": 1.9271,
+      "step": 133851
+    },
+    {
+      "epoch": 13.744121573056782,
+      "grad_norm": 0.06218159198760986,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 133854
+    },
+    {
+      "epoch": 13.744429612896601,
+      "grad_norm": 0.07486993819475174,
+      "learning_rate": 0.01,
+      "loss": 1.875,
+      "step": 133857
+    },
+    {
+      "epoch": 13.74473765273642,
+      "grad_norm": 0.10061339288949966,
+      "learning_rate": 0.01,
+      "loss": 1.8851,
+      "step": 133860
+    },
+    {
+      "epoch": 13.74504569257624,
+      "grad_norm": 0.06812550872564316,
+      "learning_rate": 0.01,
+      "loss": 1.8978,
+      "step": 133863
+    },
+    {
+      "epoch": 13.74535373241606,
+      "grad_norm": 0.05021774396300316,
+      "learning_rate": 0.01,
+      "loss": 1.9247,
+      "step": 133866
+    },
+    {
+      "epoch": 13.745661772255879,
+      "grad_norm": 0.05543253943324089,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 133869
+    },
+    {
+      "epoch": 13.745969812095698,
+      "grad_norm": 0.05269444361329079,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 133872
+    },
+    {
+      "epoch": 13.746277851935517,
+      "grad_norm": 0.0928499847650528,
+      "learning_rate": 0.01,
+      "loss": 1.917,
+      "step": 133875
+    },
+    {
+      "epoch": 13.746585891775336,
+      "grad_norm": 0.07951610535383224,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 133878
+    },
+    {
+      "epoch": 13.746893931615155,
+      "grad_norm": 0.1270100176334381,
+      "learning_rate": 0.01,
+      "loss": 1.9396,
+      "step": 133881
+    },
+    {
+      "epoch": 13.747201971454976,
+      "grad_norm": 0.14657975733280182,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 133884
+    },
+    {
+      "epoch": 13.747510011294795,
+      "grad_norm": 0.04122904688119888,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 133887
+    },
+    {
+      "epoch": 13.747818051134614,
+      "grad_norm": 0.10457743704319,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 133890
+    },
+    {
+      "epoch": 13.748126090974433,
+      "grad_norm": 0.13485820591449738,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 133893
+    },
+    {
+      "epoch": 13.748434130814251,
+      "grad_norm": 0.09970443695783615,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 133896
+    },
+    {
+      "epoch": 13.74874217065407,
+      "grad_norm": 0.10328401625156403,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 133899
+    },
+    {
+      "epoch": 13.749050210493891,
+      "grad_norm": 0.09621664136648178,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 133902
+    },
+    {
+      "epoch": 13.74935825033371,
+      "grad_norm": 0.04719613119959831,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 133905
+    },
+    {
+      "epoch": 13.74966629017353,
+      "grad_norm": 0.0460328683257103,
+      "learning_rate": 0.01,
+      "loss": 1.9037,
+      "step": 133908
+    },
+    {
+      "epoch": 13.749974330013348,
+      "grad_norm": 0.03773370385169983,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 133911
+    },
+    {
+      "epoch": 13.750282369853167,
+      "grad_norm": 0.03840022161602974,
+      "learning_rate": 0.01,
+      "loss": 1.8983,
+      "step": 133914
+    },
+    {
+      "epoch": 13.750590409692986,
+      "grad_norm": 0.046410925686359406,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 133917
+    },
+    {
+      "epoch": 13.750898449532807,
+      "grad_norm": 0.05189266428351402,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 133920
+    },
+    {
+      "epoch": 13.751206489372626,
+      "grad_norm": 0.041561324149370193,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 133923
+    },
+    {
+      "epoch": 13.751514529212445,
+      "grad_norm": 0.03292752802371979,
+      "learning_rate": 0.01,
+      "loss": 1.9013,
+      "step": 133926
+    },
+    {
+      "epoch": 13.751822569052264,
+      "grad_norm": 0.07799576967954636,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 133929
+    },
+    {
+      "epoch": 13.752130608892083,
+      "grad_norm": 0.06980624794960022,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 133932
+    },
+    {
+      "epoch": 13.752438648731903,
+      "grad_norm": 0.04551766440272331,
+      "learning_rate": 0.01,
+      "loss": 1.9251,
+      "step": 133935
+    },
+    {
+      "epoch": 13.752746688571722,
+      "grad_norm": 0.035989824682474136,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 133938
+    },
+    {
+      "epoch": 13.753054728411541,
+      "grad_norm": 0.04311169683933258,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 133941
+    },
+    {
+      "epoch": 13.75336276825136,
+      "grad_norm": 0.06231575459241867,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 133944
+    },
+    {
+      "epoch": 13.75367080809118,
+      "grad_norm": 0.03330623358488083,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 133947
+    },
+    {
+      "epoch": 13.753978847930998,
+      "grad_norm": 0.100166916847229,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 133950
+    },
+    {
+      "epoch": 13.754286887770819,
+      "grad_norm": 0.046507805585861206,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 133953
+    },
+    {
+      "epoch": 13.754594927610638,
+      "grad_norm": 0.1403961330652237,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 133956
+    },
+    {
+      "epoch": 13.754902967450457,
+      "grad_norm": 0.11719229072332382,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 133959
+    },
+    {
+      "epoch": 13.755211007290276,
+      "grad_norm": 0.04278463497757912,
+      "learning_rate": 0.01,
+      "loss": 1.8903,
+      "step": 133962
+    },
+    {
+      "epoch": 13.755519047130095,
+      "grad_norm": 0.09539931267499924,
+      "learning_rate": 0.01,
+      "loss": 1.9367,
+      "step": 133965
+    },
+    {
+      "epoch": 13.755827086969914,
+      "grad_norm": 0.052098117768764496,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 133968
+    },
+    {
+      "epoch": 13.756135126809735,
+      "grad_norm": 0.0908760279417038,
+      "learning_rate": 0.01,
+      "loss": 1.8889,
+      "step": 133971
+    },
+    {
+      "epoch": 13.756443166649554,
+      "grad_norm": 0.06540768593549728,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 133974
+    },
+    {
+      "epoch": 13.756751206489373,
+      "grad_norm": 0.074335478246212,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 133977
+    },
+    {
+      "epoch": 13.757059246329192,
+      "grad_norm": 0.08555806428194046,
+      "learning_rate": 0.01,
+      "loss": 1.901,
+      "step": 133980
+    },
+    {
+      "epoch": 13.75736728616901,
+      "grad_norm": 0.07242592424154282,
+      "learning_rate": 0.01,
+      "loss": 1.861,
+      "step": 133983
+    },
+    {
+      "epoch": 13.75767532600883,
+      "grad_norm": 0.09046568721532822,
+      "learning_rate": 0.01,
+      "loss": 1.9432,
+      "step": 133986
+    },
+    {
+      "epoch": 13.75798336584865,
+      "grad_norm": 0.04400978982448578,
+      "learning_rate": 0.01,
+      "loss": 1.9186,
+      "step": 133989
+    },
+    {
+      "epoch": 13.75829140568847,
+      "grad_norm": 0.09759508073329926,
+      "learning_rate": 0.01,
+      "loss": 1.898,
+      "step": 133992
+    },
+    {
+      "epoch": 13.758599445528288,
+      "grad_norm": 0.1339149922132492,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 133995
+    },
+    {
+      "epoch": 13.758907485368107,
+      "grad_norm": 0.11595579236745834,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 133998
+    },
+    {
+      "epoch": 13.759215525207926,
+      "grad_norm": 0.06885413825511932,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 134001
+    },
+    {
+      "epoch": 13.759523565047747,
+      "grad_norm": 0.09385296702384949,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 134004
+    },
+    {
+      "epoch": 13.759831604887566,
+      "grad_norm": 0.08676565438508987,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 134007
+    },
+    {
+      "epoch": 13.760139644727385,
+      "grad_norm": 0.04629639908671379,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 134010
+    },
+    {
+      "epoch": 13.760447684567204,
+      "grad_norm": 0.03881304711103439,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 134013
+    },
+    {
+      "epoch": 13.760755724407023,
+      "grad_norm": 0.03458276018500328,
+      "learning_rate": 0.01,
+      "loss": 1.9424,
+      "step": 134016
+    },
+    {
+      "epoch": 13.761063764246842,
+      "grad_norm": 0.042950425297021866,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 134019
+    },
+    {
+      "epoch": 13.761371804086663,
+      "grad_norm": 0.04254752770066261,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 134022
+    },
+    {
+      "epoch": 13.761679843926482,
+      "grad_norm": 0.04615510627627373,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 134025
+    },
+    {
+      "epoch": 13.7619878837663,
+      "grad_norm": 0.06772006303071976,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 134028
+    },
+    {
+      "epoch": 13.76229592360612,
+      "grad_norm": 0.06092317774891853,
+      "learning_rate": 0.01,
+      "loss": 1.893,
+      "step": 134031
+    },
+    {
+      "epoch": 13.762603963445939,
+      "grad_norm": 0.12066809833049774,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 134034
+    },
+    {
+      "epoch": 13.762912003285757,
+      "grad_norm": 0.03660951927304268,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 134037
+    },
+    {
+      "epoch": 13.763220043125578,
+      "grad_norm": 0.0796288251876831,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 134040
+    },
+    {
+      "epoch": 13.763528082965397,
+      "grad_norm": 0.08632009476423264,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 134043
+    },
+    {
+      "epoch": 13.763836122805216,
+      "grad_norm": 0.08966416865587234,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 134046
+    },
+    {
+      "epoch": 13.764144162645035,
+      "grad_norm": 0.07891400158405304,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 134049
+    },
+    {
+      "epoch": 13.764452202484854,
+      "grad_norm": 0.03551961109042168,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 134052
+    },
+    {
+      "epoch": 13.764760242324673,
+      "grad_norm": 0.04799165204167366,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 134055
+    },
+    {
+      "epoch": 13.765068282164494,
+      "grad_norm": 0.03586355969309807,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 134058
+    },
+    {
+      "epoch": 13.765376322004313,
+      "grad_norm": 0.06563351303339005,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 134061
+    },
+    {
+      "epoch": 13.765684361844132,
+      "grad_norm": 0.08807242661714554,
+      "learning_rate": 0.01,
+      "loss": 1.9065,
+      "step": 134064
+    },
+    {
+      "epoch": 13.76599240168395,
+      "grad_norm": 0.043105658143758774,
+      "learning_rate": 0.01,
+      "loss": 1.8927,
+      "step": 134067
+    },
+    {
+      "epoch": 13.76630044152377,
+      "grad_norm": 0.03263378143310547,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 134070
+    },
+    {
+      "epoch": 13.76660848136359,
+      "grad_norm": 0.057539429515600204,
+      "learning_rate": 0.01,
+      "loss": 1.9478,
+      "step": 134073
+    },
+    {
+      "epoch": 13.76691652120341,
+      "grad_norm": 0.11178094893693924,
+      "learning_rate": 0.01,
+      "loss": 1.9408,
+      "step": 134076
+    },
+    {
+      "epoch": 13.767224561043228,
+      "grad_norm": 0.16375568509101868,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 134079
+    },
+    {
+      "epoch": 13.767532600883047,
+      "grad_norm": 0.10242690145969391,
+      "learning_rate": 0.01,
+      "loss": 1.9268,
+      "step": 134082
+    },
+    {
+      "epoch": 13.767840640722866,
+      "grad_norm": 0.06567751616239548,
+      "learning_rate": 0.01,
+      "loss": 1.897,
+      "step": 134085
+    },
+    {
+      "epoch": 13.768148680562685,
+      "grad_norm": 0.03964368626475334,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 134088
+    },
+    {
+      "epoch": 13.768456720402506,
+      "grad_norm": 0.06623902171850204,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 134091
+    },
+    {
+      "epoch": 13.768764760242325,
+      "grad_norm": 0.10227943956851959,
+      "learning_rate": 0.01,
+      "loss": 1.8931,
+      "step": 134094
+    },
+    {
+      "epoch": 13.769072800082144,
+      "grad_norm": 0.03965136781334877,
+      "learning_rate": 0.01,
+      "loss": 1.9358,
+      "step": 134097
+    },
+    {
+      "epoch": 13.769380839921963,
+      "grad_norm": 0.048255518078804016,
+      "learning_rate": 0.01,
+      "loss": 1.9267,
+      "step": 134100
+    },
+    {
+      "epoch": 13.769688879761782,
+      "grad_norm": 0.07890251278877258,
+      "learning_rate": 0.01,
+      "loss": 1.8879,
+      "step": 134103
+    },
+    {
+      "epoch": 13.769996919601601,
+      "grad_norm": 0.05611550807952881,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134106
+    },
+    {
+      "epoch": 13.770304959441422,
+      "grad_norm": 0.11655380576848984,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 134109
+    },
+    {
+      "epoch": 13.77061299928124,
+      "grad_norm": 0.0641232430934906,
+      "learning_rate": 0.01,
+      "loss": 1.913,
+      "step": 134112
+    },
+    {
+      "epoch": 13.77092103912106,
+      "grad_norm": 0.05033102259039879,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 134115
+    },
+    {
+      "epoch": 13.771229078960879,
+      "grad_norm": 0.07937975227832794,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 134118
+    },
+    {
+      "epoch": 13.771537118800698,
+      "grad_norm": 0.08784115314483643,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 134121
+    },
+    {
+      "epoch": 13.771845158640517,
+      "grad_norm": 0.05763769522309303,
+      "learning_rate": 0.01,
+      "loss": 1.9311,
+      "step": 134124
+    },
+    {
+      "epoch": 13.772153198480337,
+      "grad_norm": 0.04057874530553818,
+      "learning_rate": 0.01,
+      "loss": 1.8794,
+      "step": 134127
+    },
+    {
+      "epoch": 13.772461238320156,
+      "grad_norm": 0.033341724425554276,
+      "learning_rate": 0.01,
+      "loss": 1.9203,
+      "step": 134130
+    },
+    {
+      "epoch": 13.772769278159975,
+      "grad_norm": 0.03664823994040489,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 134133
+    },
+    {
+      "epoch": 13.773077317999794,
+      "grad_norm": 0.061253972351551056,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 134136
+    },
+    {
+      "epoch": 13.773385357839613,
+      "grad_norm": 0.041739314794540405,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134139
+    },
+    {
+      "epoch": 13.773693397679434,
+      "grad_norm": 0.24283228814601898,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 134142
+    },
+    {
+      "epoch": 13.774001437519253,
+      "grad_norm": 0.09927505254745483,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 134145
+    },
+    {
+      "epoch": 13.774309477359072,
+      "grad_norm": 0.08252577483654022,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 134148
+    },
+    {
+      "epoch": 13.774617517198891,
+      "grad_norm": 0.031050346791744232,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 134151
+    },
+    {
+      "epoch": 13.77492555703871,
+      "grad_norm": 0.04443134739995003,
+      "learning_rate": 0.01,
+      "loss": 1.9178,
+      "step": 134154
+    },
+    {
+      "epoch": 13.775233596878529,
+      "grad_norm": 0.032365597784519196,
+      "learning_rate": 0.01,
+      "loss": 1.9011,
+      "step": 134157
+    },
+    {
+      "epoch": 13.77554163671835,
+      "grad_norm": 0.046413496136665344,
+      "learning_rate": 0.01,
+      "loss": 1.9434,
+      "step": 134160
+    },
+    {
+      "epoch": 13.775849676558169,
+      "grad_norm": 0.07829392701387405,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 134163
+    },
+    {
+      "epoch": 13.776157716397988,
+      "grad_norm": 0.04649561643600464,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 134166
+    },
+    {
+      "epoch": 13.776465756237807,
+      "grad_norm": 0.037897855043411255,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 134169
+    },
+    {
+      "epoch": 13.776773796077626,
+      "grad_norm": 0.035940542817115784,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 134172
+    },
+    {
+      "epoch": 13.777081835917445,
+      "grad_norm": 0.0427544079720974,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 134175
+    },
+    {
+      "epoch": 13.777389875757265,
+      "grad_norm": 0.031767211854457855,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 134178
+    },
+    {
+      "epoch": 13.777697915597084,
+      "grad_norm": 0.06968454271554947,
+      "learning_rate": 0.01,
+      "loss": 1.9295,
+      "step": 134181
+    },
+    {
+      "epoch": 13.778005955436903,
+      "grad_norm": 0.07139027118682861,
+      "learning_rate": 0.01,
+      "loss": 1.9121,
+      "step": 134184
+    },
+    {
+      "epoch": 13.778313995276722,
+      "grad_norm": 0.05282384157180786,
+      "learning_rate": 0.01,
+      "loss": 1.9374,
+      "step": 134187
+    },
+    {
+      "epoch": 13.778622035116541,
+      "grad_norm": 0.05122571066021919,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 134190
+    },
+    {
+      "epoch": 13.77893007495636,
+      "grad_norm": 0.06055007874965668,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 134193
+    },
+    {
+      "epoch": 13.77923811479618,
+      "grad_norm": 0.06373171508312225,
+      "learning_rate": 0.01,
+      "loss": 1.8905,
+      "step": 134196
+    },
+    {
+      "epoch": 13.779546154636,
+      "grad_norm": 0.09918566048145294,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 134199
+    },
+    {
+      "epoch": 13.779854194475819,
+      "grad_norm": 0.0594269260764122,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 134202
+    },
+    {
+      "epoch": 13.780162234315638,
+      "grad_norm": 0.05142861232161522,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 134205
+    },
+    {
+      "epoch": 13.780470274155457,
+      "grad_norm": 0.030220456421375275,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 134208
+    },
+    {
+      "epoch": 13.780778313995278,
+      "grad_norm": 0.1389419287443161,
+      "learning_rate": 0.01,
+      "loss": 1.9362,
+      "step": 134211
+    },
+    {
+      "epoch": 13.781086353835097,
+      "grad_norm": 0.0764242634177208,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 134214
+    },
+    {
+      "epoch": 13.781394393674915,
+      "grad_norm": 0.04809550940990448,
+      "learning_rate": 0.01,
+      "loss": 1.8808,
+      "step": 134217
+    },
+    {
+      "epoch": 13.781702433514734,
+      "grad_norm": 0.060204897075891495,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 134220
+    },
+    {
+      "epoch": 13.782010473354553,
+      "grad_norm": 0.03459016978740692,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 134223
+    },
+    {
+      "epoch": 13.782318513194372,
+      "grad_norm": 0.037191927433013916,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 134226
+    },
+    {
+      "epoch": 13.782626553034193,
+      "grad_norm": 0.10385231673717499,
+      "learning_rate": 0.01,
+      "loss": 1.9243,
+      "step": 134229
+    },
+    {
+      "epoch": 13.782934592874012,
+      "grad_norm": 0.09311854094266891,
+      "learning_rate": 0.01,
+      "loss": 1.9129,
+      "step": 134232
+    },
+    {
+      "epoch": 13.783242632713831,
+      "grad_norm": 0.08622749894857407,
+      "learning_rate": 0.01,
+      "loss": 1.9353,
+      "step": 134235
+    },
+    {
+      "epoch": 13.78355067255365,
+      "grad_norm": 0.11061009764671326,
+      "learning_rate": 0.01,
+      "loss": 1.9347,
+      "step": 134238
+    },
+    {
+      "epoch": 13.783858712393469,
+      "grad_norm": 0.08668726682662964,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 134241
+    },
+    {
+      "epoch": 13.784166752233288,
+      "grad_norm": 0.054344091564416885,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 134244
+    },
+    {
+      "epoch": 13.784474792073109,
+      "grad_norm": 0.03575201332569122,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 134247
+    },
+    {
+      "epoch": 13.784782831912928,
+      "grad_norm": 0.03980579972267151,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 134250
+    },
+    {
+      "epoch": 13.785090871752747,
+      "grad_norm": 0.07016754895448685,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 134253
+    },
+    {
+      "epoch": 13.785398911592566,
+      "grad_norm": 0.08482769876718521,
+      "learning_rate": 0.01,
+      "loss": 1.9195,
+      "step": 134256
+    },
+    {
+      "epoch": 13.785706951432385,
+      "grad_norm": 0.06293204426765442,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 134259
+    },
+    {
+      "epoch": 13.786014991272204,
+      "grad_norm": 0.04305669665336609,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 134262
+    },
+    {
+      "epoch": 13.786323031112024,
+      "grad_norm": 0.03531415015459061,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 134265
+    },
+    {
+      "epoch": 13.786631070951843,
+      "grad_norm": 0.105159692466259,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 134268
+    },
+    {
+      "epoch": 13.786939110791662,
+      "grad_norm": 0.06086484342813492,
+      "learning_rate": 0.01,
+      "loss": 1.908,
+      "step": 134271
+    },
+    {
+      "epoch": 13.787247150631481,
+      "grad_norm": 0.04095148667693138,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134274
+    },
+    {
+      "epoch": 13.7875551904713,
+      "grad_norm": 0.055096086114645004,
+      "learning_rate": 0.01,
+      "loss": 1.9052,
+      "step": 134277
+    },
+    {
+      "epoch": 13.787863230311121,
+      "grad_norm": 0.13299863040447235,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 134280
+    },
+    {
+      "epoch": 13.78817127015094,
+      "grad_norm": 0.07362045347690582,
+      "learning_rate": 0.01,
+      "loss": 1.9361,
+      "step": 134283
+    },
+    {
+      "epoch": 13.788479309990759,
+      "grad_norm": 0.0877787172794342,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 134286
+    },
+    {
+      "epoch": 13.788787349830578,
+      "grad_norm": 0.12075179815292358,
+      "learning_rate": 0.01,
+      "loss": 1.9187,
+      "step": 134289
+    },
+    {
+      "epoch": 13.789095389670397,
+      "grad_norm": 0.03889086842536926,
+      "learning_rate": 0.01,
+      "loss": 1.9435,
+      "step": 134292
+    },
+    {
+      "epoch": 13.789403429510216,
+      "grad_norm": 0.0332590714097023,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 134295
+    },
+    {
+      "epoch": 13.789711469350037,
+      "grad_norm": 0.04149647057056427,
+      "learning_rate": 0.01,
+      "loss": 1.9143,
+      "step": 134298
+    },
+    {
+      "epoch": 13.790019509189856,
+      "grad_norm": 0.08757143467664719,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 134301
+    },
+    {
+      "epoch": 13.790327549029675,
+      "grad_norm": 0.03478625789284706,
+      "learning_rate": 0.01,
+      "loss": 1.889,
+      "step": 134304
+    },
+    {
+      "epoch": 13.790635588869494,
+      "grad_norm": 0.11735574901103973,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 134307
+    },
+    {
+      "epoch": 13.790943628709313,
+      "grad_norm": 0.12544330954551697,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 134310
+    },
+    {
+      "epoch": 13.791251668549132,
+      "grad_norm": 0.05187934264540672,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 134313
+    },
+    {
+      "epoch": 13.791559708388952,
+      "grad_norm": 0.07519331574440002,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 134316
+    },
+    {
+      "epoch": 13.791867748228771,
+      "grad_norm": 0.05820140987634659,
+      "learning_rate": 0.01,
+      "loss": 1.9469,
+      "step": 134319
+    },
+    {
+      "epoch": 13.79217578806859,
+      "grad_norm": 0.049865126609802246,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 134322
+    },
+    {
+      "epoch": 13.79248382790841,
+      "grad_norm": 0.037818051874637604,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 134325
+    },
+    {
+      "epoch": 13.792791867748228,
+      "grad_norm": 0.051875755190849304,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 134328
+    },
+    {
+      "epoch": 13.793099907588047,
+      "grad_norm": 0.11332733184099197,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 134331
+    },
+    {
+      "epoch": 13.793407947427868,
+      "grad_norm": 0.0697350949048996,
+      "learning_rate": 0.01,
+      "loss": 1.9258,
+      "step": 134334
+    },
+    {
+      "epoch": 13.793715987267687,
+      "grad_norm": 0.04231021925806999,
+      "learning_rate": 0.01,
+      "loss": 1.8921,
+      "step": 134337
+    },
+    {
+      "epoch": 13.794024027107506,
+      "grad_norm": 0.1183324009180069,
+      "learning_rate": 0.01,
+      "loss": 1.8952,
+      "step": 134340
+    },
+    {
+      "epoch": 13.794332066947325,
+      "grad_norm": 0.03632016479969025,
+      "learning_rate": 0.01,
+      "loss": 1.9294,
+      "step": 134343
+    },
+    {
+      "epoch": 13.794640106787144,
+      "grad_norm": 0.06476329267024994,
+      "learning_rate": 0.01,
+      "loss": 1.9149,
+      "step": 134346
+    },
+    {
+      "epoch": 13.794948146626965,
+      "grad_norm": 0.05914657935500145,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134349
+    },
+    {
+      "epoch": 13.795256186466784,
+      "grad_norm": 0.05064254254102707,
+      "learning_rate": 0.01,
+      "loss": 1.9005,
+      "step": 134352
+    },
+    {
+      "epoch": 13.795564226306603,
+      "grad_norm": 0.13330452144145966,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 134355
+    },
+    {
+      "epoch": 13.795872266146421,
+      "grad_norm": 0.046218544244766235,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 134358
+    },
+    {
+      "epoch": 13.79618030598624,
+      "grad_norm": 0.09747438877820969,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 134361
+    },
+    {
+      "epoch": 13.79648834582606,
+      "grad_norm": 0.10136681795120239,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 134364
+    },
+    {
+      "epoch": 13.79679638566588,
+      "grad_norm": 0.0588546022772789,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134367
+    },
+    {
+      "epoch": 13.7971044255057,
+      "grad_norm": 0.045304592698812485,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 134370
+    },
+    {
+      "epoch": 13.797412465345518,
+      "grad_norm": 0.0506785586476326,
+      "learning_rate": 0.01,
+      "loss": 1.9313,
+      "step": 134373
+    },
+    {
+      "epoch": 13.797720505185337,
+      "grad_norm": 0.08301147818565369,
+      "learning_rate": 0.01,
+      "loss": 1.9391,
+      "step": 134376
+    },
+    {
+      "epoch": 13.798028545025156,
+      "grad_norm": 0.04174140468239784,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 134379
+    },
+    {
+      "epoch": 13.798336584864975,
+      "grad_norm": 0.05307769775390625,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 134382
+    },
+    {
+      "epoch": 13.798644624704796,
+      "grad_norm": 0.16477017104625702,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 134385
+    },
+    {
+      "epoch": 13.798952664544615,
+      "grad_norm": 0.07897128909826279,
+      "learning_rate": 0.01,
+      "loss": 1.9269,
+      "step": 134388
+    },
+    {
+      "epoch": 13.799260704384434,
+      "grad_norm": 0.09547238796949387,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 134391
+    },
+    {
+      "epoch": 13.799568744224253,
+      "grad_norm": 0.09013862162828445,
+      "learning_rate": 0.01,
+      "loss": 1.9219,
+      "step": 134394
+    },
+    {
+      "epoch": 13.799876784064072,
+      "grad_norm": 0.045079343020915985,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 134397
+    },
+    {
+      "epoch": 13.80018482390389,
+      "grad_norm": 0.04100290685892105,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 134400
+    },
+    {
+      "epoch": 13.800492863743711,
+      "grad_norm": 0.0522785447537899,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 134403
+    },
+    {
+      "epoch": 13.80080090358353,
+      "grad_norm": 0.05137479305267334,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 134406
+    },
+    {
+      "epoch": 13.80110894342335,
+      "grad_norm": 0.03788720443844795,
+      "learning_rate": 0.01,
+      "loss": 1.9003,
+      "step": 134409
+    },
+    {
+      "epoch": 13.801416983263168,
+      "grad_norm": 0.04322421923279762,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 134412
+    },
+    {
+      "epoch": 13.801725023102987,
+      "grad_norm": 0.04263107106089592,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 134415
+    },
+    {
+      "epoch": 13.802033062942808,
+      "grad_norm": 0.061249397695064545,
+      "learning_rate": 0.01,
+      "loss": 1.9466,
+      "step": 134418
+    },
+    {
+      "epoch": 13.802341102782627,
+      "grad_norm": 0.03674252703785896,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134421
+    },
+    {
+      "epoch": 13.802649142622446,
+      "grad_norm": 0.12491140514612198,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 134424
+    },
+    {
+      "epoch": 13.802957182462265,
+      "grad_norm": 0.04308502376079559,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 134427
+    },
+    {
+      "epoch": 13.803265222302084,
+      "grad_norm": 0.11887374520301819,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 134430
+    },
+    {
+      "epoch": 13.803573262141903,
+      "grad_norm": 0.06694099307060242,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 134433
+    },
+    {
+      "epoch": 13.803881301981724,
+      "grad_norm": 0.15306909382343292,
+      "learning_rate": 0.01,
+      "loss": 1.9556,
+      "step": 134436
+    },
+    {
+      "epoch": 13.804189341821543,
+      "grad_norm": 0.14931873977184296,
+      "learning_rate": 0.01,
+      "loss": 1.881,
+      "step": 134439
+    },
+    {
+      "epoch": 13.804497381661362,
+      "grad_norm": 0.06753832846879959,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 134442
+    },
+    {
+      "epoch": 13.80480542150118,
+      "grad_norm": 0.08120930939912796,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 134445
+    },
+    {
+      "epoch": 13.805113461341,
+      "grad_norm": 0.07243016362190247,
+      "learning_rate": 0.01,
+      "loss": 1.9414,
+      "step": 134448
+    },
+    {
+      "epoch": 13.805421501180819,
+      "grad_norm": 0.06341297924518585,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 134451
+    },
+    {
+      "epoch": 13.80572954102064,
+      "grad_norm": 0.04816992208361626,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 134454
+    },
+    {
+      "epoch": 13.806037580860458,
+      "grad_norm": 0.03846464678645134,
+      "learning_rate": 0.01,
+      "loss": 1.8973,
+      "step": 134457
+    },
+    {
+      "epoch": 13.806345620700277,
+      "grad_norm": 0.04859515279531479,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 134460
+    },
+    {
+      "epoch": 13.806653660540096,
+      "grad_norm": 0.055337030440568924,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 134463
+    },
+    {
+      "epoch": 13.806961700379915,
+      "grad_norm": 0.03773676976561546,
+      "learning_rate": 0.01,
+      "loss": 1.9265,
+      "step": 134466
+    },
+    {
+      "epoch": 13.807269740219734,
+      "grad_norm": 0.06136350706219673,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 134469
+    },
+    {
+      "epoch": 13.807577780059555,
+      "grad_norm": 0.06748557090759277,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 134472
+    },
+    {
+      "epoch": 13.807885819899374,
+      "grad_norm": 0.04549436643719673,
+      "learning_rate": 0.01,
+      "loss": 1.928,
+      "step": 134475
+    },
+    {
+      "epoch": 13.808193859739193,
+      "grad_norm": 0.04872233048081398,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 134478
+    },
+    {
+      "epoch": 13.808501899579012,
+      "grad_norm": 0.057213712483644485,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 134481
+    },
+    {
+      "epoch": 13.80880993941883,
+      "grad_norm": 0.03600826859474182,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 134484
+    },
+    {
+      "epoch": 13.809117979258652,
+      "grad_norm": 0.05736810341477394,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 134487
+    },
+    {
+      "epoch": 13.80942601909847,
+      "grad_norm": 0.18942776322364807,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 134490
+    },
+    {
+      "epoch": 13.80973405893829,
+      "grad_norm": 0.08649734407663345,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 134493
+    },
+    {
+      "epoch": 13.810042098778109,
+      "grad_norm": 0.048332132399082184,
+      "learning_rate": 0.01,
+      "loss": 1.9561,
+      "step": 134496
+    },
+    {
+      "epoch": 13.810350138617927,
+      "grad_norm": 0.040949445217847824,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 134499
+    },
+    {
+      "epoch": 13.810658178457746,
+      "grad_norm": 0.03774237632751465,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 134502
+    },
+    {
+      "epoch": 13.810966218297567,
+      "grad_norm": 0.043185681104660034,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 134505
+    },
+    {
+      "epoch": 13.811274258137386,
+      "grad_norm": 0.03423027694225311,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 134508
+    },
+    {
+      "epoch": 13.811582297977205,
+      "grad_norm": 0.05821787193417549,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 134511
+    },
+    {
+      "epoch": 13.811890337817024,
+      "grad_norm": 0.09472539275884628,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 134514
+    },
+    {
+      "epoch": 13.812198377656843,
+      "grad_norm": 0.14878137409687042,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 134517
+    },
+    {
+      "epoch": 13.812506417496662,
+      "grad_norm": 0.07622846961021423,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 134520
+    },
+    {
+      "epoch": 13.812814457336483,
+      "grad_norm": 0.05293974280357361,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 134523
+    },
+    {
+      "epoch": 13.813122497176302,
+      "grad_norm": 0.05401444807648659,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 134526
+    },
+    {
+      "epoch": 13.81343053701612,
+      "grad_norm": 0.07161324471235275,
+      "learning_rate": 0.01,
+      "loss": 1.8971,
+      "step": 134529
+    },
+    {
+      "epoch": 13.81373857685594,
+      "grad_norm": 0.04347148910164833,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 134532
+    },
+    {
+      "epoch": 13.814046616695759,
+      "grad_norm": 0.09399743378162384,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 134535
+    },
+    {
+      "epoch": 13.814354656535578,
+      "grad_norm": 0.09191381186246872,
+      "learning_rate": 0.01,
+      "loss": 1.9055,
+      "step": 134538
+    },
+    {
+      "epoch": 13.814662696375398,
+      "grad_norm": 0.06622043251991272,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 134541
+    },
+    {
+      "epoch": 13.814970736215217,
+      "grad_norm": 0.0727938637137413,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 134544
+    },
+    {
+      "epoch": 13.815278776055036,
+      "grad_norm": 0.0668262168765068,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 134547
+    },
+    {
+      "epoch": 13.815586815894855,
+      "grad_norm": 0.047191739082336426,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 134550
+    },
+    {
+      "epoch": 13.815894855734674,
+      "grad_norm": 0.13995061814785004,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 134553
+    },
+    {
+      "epoch": 13.816202895574495,
+      "grad_norm": 0.07915446162223816,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 134556
+    },
+    {
+      "epoch": 13.816510935414314,
+      "grad_norm": 0.06913372874259949,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 134559
+    },
+    {
+      "epoch": 13.816818975254133,
+      "grad_norm": 0.06800039112567902,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 134562
+    },
+    {
+      "epoch": 13.817127015093952,
+      "grad_norm": 0.061536066234111786,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 134565
+    },
+    {
+      "epoch": 13.817435054933771,
+      "grad_norm": 0.04014867916703224,
+      "learning_rate": 0.01,
+      "loss": 1.9308,
+      "step": 134568
+    },
+    {
+      "epoch": 13.81774309477359,
+      "grad_norm": 0.11430704593658447,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 134571
+    },
+    {
+      "epoch": 13.81805113461341,
+      "grad_norm": 0.09642425179481506,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 134574
+    },
+    {
+      "epoch": 13.81835917445323,
+      "grad_norm": 0.08144189417362213,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 134577
+    },
+    {
+      "epoch": 13.818667214293049,
+      "grad_norm": 0.03637473285198212,
+      "learning_rate": 0.01,
+      "loss": 1.8997,
+      "step": 134580
+    },
+    {
+      "epoch": 13.818975254132868,
+      "grad_norm": 0.035632647573947906,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 134583
+    },
+    {
+      "epoch": 13.819283293972687,
+      "grad_norm": 0.08469029515981674,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 134586
+    },
+    {
+      "epoch": 13.819591333812506,
+      "grad_norm": 0.10370516031980515,
+      "learning_rate": 0.01,
+      "loss": 1.9099,
+      "step": 134589
+    },
+    {
+      "epoch": 13.819899373652326,
+      "grad_norm": 0.09744898229837418,
+      "learning_rate": 0.01,
+      "loss": 1.9208,
+      "step": 134592
+    },
+    {
+      "epoch": 13.820207413492145,
+      "grad_norm": 0.07495691627264023,
+      "learning_rate": 0.01,
+      "loss": 1.925,
+      "step": 134595
+    },
+    {
+      "epoch": 13.820515453331964,
+      "grad_norm": 0.03808124363422394,
+      "learning_rate": 0.01,
+      "loss": 1.9331,
+      "step": 134598
+    },
+    {
+      "epoch": 13.820823493171783,
+      "grad_norm": 0.03779391571879387,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 134601
+    },
+    {
+      "epoch": 13.821131533011602,
+      "grad_norm": 0.03805067762732506,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 134604
+    },
+    {
+      "epoch": 13.821439572851421,
+      "grad_norm": 0.05452338978648186,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 134607
+    },
+    {
+      "epoch": 13.821747612691242,
+      "grad_norm": 0.13283495604991913,
+      "learning_rate": 0.01,
+      "loss": 1.9329,
+      "step": 134610
+    },
+    {
+      "epoch": 13.822055652531061,
+      "grad_norm": 0.054554685950279236,
+      "learning_rate": 0.01,
+      "loss": 1.9111,
+      "step": 134613
+    },
+    {
+      "epoch": 13.82236369237088,
+      "grad_norm": 0.05358796566724777,
+      "learning_rate": 0.01,
+      "loss": 1.9352,
+      "step": 134616
+    },
+    {
+      "epoch": 13.822671732210699,
+      "grad_norm": 0.04198135435581207,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 134619
+    },
+    {
+      "epoch": 13.822979772050518,
+      "grad_norm": 0.03444984555244446,
+      "learning_rate": 0.01,
+      "loss": 1.9299,
+      "step": 134622
+    },
+    {
+      "epoch": 13.823287811890339,
+      "grad_norm": 0.055453188717365265,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 134625
+    },
+    {
+      "epoch": 13.823595851730158,
+      "grad_norm": 0.13516004383563995,
+      "learning_rate": 0.01,
+      "loss": 1.9007,
+      "step": 134628
+    },
+    {
+      "epoch": 13.823903891569977,
+      "grad_norm": 0.11913404613733292,
+      "learning_rate": 0.01,
+      "loss": 1.9045,
+      "step": 134631
+    },
+    {
+      "epoch": 13.824211931409796,
+      "grad_norm": 0.03697621822357178,
+      "learning_rate": 0.01,
+      "loss": 1.898,
+      "step": 134634
+    },
+    {
+      "epoch": 13.824519971249615,
+      "grad_norm": 0.03736665099859238,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 134637
+    },
+    {
+      "epoch": 13.824828011089433,
+      "grad_norm": 0.11643897742033005,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 134640
+    },
+    {
+      "epoch": 13.825136050929254,
+      "grad_norm": 0.07806897908449173,
+      "learning_rate": 0.01,
+      "loss": 1.9054,
+      "step": 134643
+    },
+    {
+      "epoch": 13.825444090769073,
+      "grad_norm": 0.030000755563378334,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 134646
+    },
+    {
+      "epoch": 13.825752130608892,
+      "grad_norm": 0.04103497788310051,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 134649
+    },
+    {
+      "epoch": 13.826060170448711,
+      "grad_norm": 0.05625663697719574,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 134652
+    },
+    {
+      "epoch": 13.82636821028853,
+      "grad_norm": 0.04921203851699829,
+      "learning_rate": 0.01,
+      "loss": 1.8941,
+      "step": 134655
+    },
+    {
+      "epoch": 13.826676250128349,
+      "grad_norm": 0.06318880617618561,
+      "learning_rate": 0.01,
+      "loss": 1.9253,
+      "step": 134658
+    },
+    {
+      "epoch": 13.82698428996817,
+      "grad_norm": 0.042655784636735916,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 134661
+    },
+    {
+      "epoch": 13.827292329807989,
+      "grad_norm": 0.04156467691063881,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 134664
+    },
+    {
+      "epoch": 13.827600369647808,
+      "grad_norm": 0.05358791723847389,
+      "learning_rate": 0.01,
+      "loss": 1.9098,
+      "step": 134667
+    },
+    {
+      "epoch": 13.827908409487627,
+      "grad_norm": 0.09285346418619156,
+      "learning_rate": 0.01,
+      "loss": 1.9102,
+      "step": 134670
+    },
+    {
+      "epoch": 13.828216449327446,
+      "grad_norm": 0.11332570761442184,
+      "learning_rate": 0.01,
+      "loss": 1.9405,
+      "step": 134673
+    },
+    {
+      "epoch": 13.828524489167265,
+      "grad_norm": 0.12899932265281677,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 134676
+    },
+    {
+      "epoch": 13.828832529007085,
+      "grad_norm": 0.06730727106332779,
+      "learning_rate": 0.01,
+      "loss": 1.8902,
+      "step": 134679
+    },
+    {
+      "epoch": 13.829140568846904,
+      "grad_norm": 0.09343144297599792,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 134682
+    },
+    {
+      "epoch": 13.829448608686723,
+      "grad_norm": 0.09627532958984375,
+      "learning_rate": 0.01,
+      "loss": 1.9379,
+      "step": 134685
+    },
+    {
+      "epoch": 13.829756648526542,
+      "grad_norm": 0.1498269885778427,
+      "learning_rate": 0.01,
+      "loss": 1.9228,
+      "step": 134688
+    },
+    {
+      "epoch": 13.830064688366361,
+      "grad_norm": 0.09698925912380219,
+      "learning_rate": 0.01,
+      "loss": 1.9476,
+      "step": 134691
+    },
+    {
+      "epoch": 13.830372728206182,
+      "grad_norm": 0.05098671838641167,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 134694
+    },
+    {
+      "epoch": 13.830680768046001,
+      "grad_norm": 0.03165208920836449,
+      "learning_rate": 0.01,
+      "loss": 1.895,
+      "step": 134697
+    },
+    {
+      "epoch": 13.83098880788582,
+      "grad_norm": 0.03985311836004257,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 134700
+    },
+    {
+      "epoch": 13.831296847725639,
+      "grad_norm": 0.04467601329088211,
+      "learning_rate": 0.01,
+      "loss": 1.9041,
+      "step": 134703
+    },
+    {
+      "epoch": 13.831604887565458,
+      "grad_norm": 0.06141729652881622,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 134706
+    },
+    {
+      "epoch": 13.831912927405277,
+      "grad_norm": 0.034828633069992065,
+      "learning_rate": 0.01,
+      "loss": 1.898,
+      "step": 134709
+    },
+    {
+      "epoch": 13.832220967245098,
+      "grad_norm": 0.12627530097961426,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 134712
+    },
+    {
+      "epoch": 13.832529007084917,
+      "grad_norm": 0.06956329941749573,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 134715
+    },
+    {
+      "epoch": 13.832837046924736,
+      "grad_norm": 0.032219383865594864,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 134718
+    },
+    {
+      "epoch": 13.833145086764555,
+      "grad_norm": 0.0622720904648304,
+      "learning_rate": 0.01,
+      "loss": 1.9074,
+      "step": 134721
+    },
+    {
+      "epoch": 13.833453126604374,
+      "grad_norm": 0.0415850505232811,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 134724
+    },
+    {
+      "epoch": 13.833761166444193,
+      "grad_norm": 0.039370544254779816,
+      "learning_rate": 0.01,
+      "loss": 1.9183,
+      "step": 134727
+    },
+    {
+      "epoch": 13.834069206284013,
+      "grad_norm": 0.03335817903280258,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 134730
+    },
+    {
+      "epoch": 13.834377246123832,
+      "grad_norm": 0.06952258199453354,
+      "learning_rate": 0.01,
+      "loss": 1.9161,
+      "step": 134733
+    },
+    {
+      "epoch": 13.834685285963651,
+      "grad_norm": 0.07735583931207657,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 134736
+    },
+    {
+      "epoch": 13.83499332580347,
+      "grad_norm": 0.05157242342829704,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 134739
+    },
+    {
+      "epoch": 13.83530136564329,
+      "grad_norm": 0.045335594564676285,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 134742
+    },
+    {
+      "epoch": 13.835609405483108,
+      "grad_norm": 0.044016480445861816,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 134745
+    },
+    {
+      "epoch": 13.835917445322929,
+      "grad_norm": 0.0825684666633606,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 134748
+    },
+    {
+      "epoch": 13.836225485162748,
+      "grad_norm": 0.03913147747516632,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 134751
+    },
+    {
+      "epoch": 13.836533525002567,
+      "grad_norm": 0.08447827398777008,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 134754
+    },
+    {
+      "epoch": 13.836841564842386,
+      "grad_norm": 0.09996984153985977,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 134757
+    },
+    {
+      "epoch": 13.837149604682205,
+      "grad_norm": 0.05911770090460777,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 134760
+    },
+    {
+      "epoch": 13.837457644522026,
+      "grad_norm": 0.06662571430206299,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 134763
+    },
+    {
+      "epoch": 13.837765684361845,
+      "grad_norm": 0.07513116300106049,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 134766
+    },
+    {
+      "epoch": 13.838073724201664,
+      "grad_norm": 0.12169796228408813,
+      "learning_rate": 0.01,
+      "loss": 1.8914,
+      "step": 134769
+    },
+    {
+      "epoch": 13.838381764041483,
+      "grad_norm": 0.04476134106516838,
+      "learning_rate": 0.01,
+      "loss": 1.915,
+      "step": 134772
+    },
+    {
+      "epoch": 13.838689803881302,
+      "grad_norm": 0.09173692017793655,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 134775
+    },
+    {
+      "epoch": 13.83899784372112,
+      "grad_norm": 0.04771685227751732,
+      "learning_rate": 0.01,
+      "loss": 1.9385,
+      "step": 134778
+    },
+    {
+      "epoch": 13.839305883560941,
+      "grad_norm": 0.03183364495635033,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 134781
+    },
+    {
+      "epoch": 13.83961392340076,
+      "grad_norm": 0.03098229318857193,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 134784
+    },
+    {
+      "epoch": 13.83992196324058,
+      "grad_norm": 0.12245805561542511,
+      "learning_rate": 0.01,
+      "loss": 1.9153,
+      "step": 134787
+    },
+    {
+      "epoch": 13.840230003080398,
+      "grad_norm": 0.03875253349542618,
+      "learning_rate": 0.01,
+      "loss": 1.9063,
+      "step": 134790
+    },
+    {
+      "epoch": 13.840538042920217,
+      "grad_norm": 0.06304722279310226,
+      "learning_rate": 0.01,
+      "loss": 1.8953,
+      "step": 134793
+    },
+    {
+      "epoch": 13.840846082760036,
+      "grad_norm": 0.033619582653045654,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 134796
+    },
+    {
+      "epoch": 13.841154122599857,
+      "grad_norm": 0.03246408328413963,
+      "learning_rate": 0.01,
+      "loss": 1.9277,
+      "step": 134799
+    },
+    {
+      "epoch": 13.841462162439676,
+      "grad_norm": 0.0373997837305069,
+      "learning_rate": 0.01,
+      "loss": 1.9452,
+      "step": 134802
+    },
+    {
+      "epoch": 13.841770202279495,
+      "grad_norm": 0.06192956864833832,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 134805
+    },
+    {
+      "epoch": 13.842078242119314,
+      "grad_norm": 0.04907293617725372,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 134808
+    },
+    {
+      "epoch": 13.842386281959133,
+      "grad_norm": 0.07524628192186356,
+      "learning_rate": 0.01,
+      "loss": 1.9006,
+      "step": 134811
+    },
+    {
+      "epoch": 13.842694321798952,
+      "grad_norm": 0.08978945761919022,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 134814
+    },
+    {
+      "epoch": 13.843002361638773,
+      "grad_norm": 0.11512564867734909,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 134817
+    },
+    {
+      "epoch": 13.843310401478591,
+      "grad_norm": 0.18758739531040192,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 134820
+    },
+    {
+      "epoch": 13.84361844131841,
+      "grad_norm": 0.09696117043495178,
+      "learning_rate": 0.01,
+      "loss": 1.903,
+      "step": 134823
+    },
+    {
+      "epoch": 13.84392648115823,
+      "grad_norm": 0.03158370032906532,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 134826
+    },
+    {
+      "epoch": 13.844234520998048,
+      "grad_norm": 0.031874481588602066,
+      "learning_rate": 0.01,
+      "loss": 1.9097,
+      "step": 134829
+    },
+    {
+      "epoch": 13.84454256083787,
+      "grad_norm": 0.045173294842243195,
+      "learning_rate": 0.01,
+      "loss": 1.8971,
+      "step": 134832
+    },
+    {
+      "epoch": 13.844850600677688,
+      "grad_norm": 0.0488600879907608,
+      "learning_rate": 0.01,
+      "loss": 1.9433,
+      "step": 134835
+    },
+    {
+      "epoch": 13.845158640517507,
+      "grad_norm": 0.11225177347660065,
+      "learning_rate": 0.01,
+      "loss": 1.9035,
+      "step": 134838
+    },
+    {
+      "epoch": 13.845466680357326,
+      "grad_norm": 0.09078571945428848,
+      "learning_rate": 0.01,
+      "loss": 1.9285,
+      "step": 134841
+    },
+    {
+      "epoch": 13.845774720197145,
+      "grad_norm": 0.04309966042637825,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 134844
+    },
+    {
+      "epoch": 13.846082760036964,
+      "grad_norm": 0.10900307446718216,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 134847
+    },
+    {
+      "epoch": 13.846390799876785,
+      "grad_norm": 0.08507236838340759,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 134850
+    },
+    {
+      "epoch": 13.846698839716604,
+      "grad_norm": 0.06559781730175018,
+      "learning_rate": 0.01,
+      "loss": 1.8974,
+      "step": 134853
+    },
+    {
+      "epoch": 13.847006879556423,
+      "grad_norm": 0.08693236112594604,
+      "learning_rate": 0.01,
+      "loss": 1.9033,
+      "step": 134856
+    },
+    {
+      "epoch": 13.847314919396242,
+      "grad_norm": 0.07437299937009811,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 134859
+    },
+    {
+      "epoch": 13.84762295923606,
+      "grad_norm": 0.049790412187576294,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 134862
+    },
+    {
+      "epoch": 13.84793099907588,
+      "grad_norm": 0.05421151965856552,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 134865
+    },
+    {
+      "epoch": 13.8482390389157,
+      "grad_norm": 0.04888584837317467,
+      "learning_rate": 0.01,
+      "loss": 1.8943,
+      "step": 134868
+    },
+    {
+      "epoch": 13.84854707875552,
+      "grad_norm": 0.041289784014225006,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 134871
+    },
+    {
+      "epoch": 13.848855118595338,
+      "grad_norm": 0.03770878165960312,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 134874
+    },
+    {
+      "epoch": 13.849163158435157,
+      "grad_norm": 0.08070291578769684,
+      "learning_rate": 0.01,
+      "loss": 1.9041,
+      "step": 134877
+    },
+    {
+      "epoch": 13.849471198274976,
+      "grad_norm": 0.10986967384815216,
+      "learning_rate": 0.01,
+      "loss": 1.8986,
+      "step": 134880
+    },
+    {
+      "epoch": 13.849779238114795,
+      "grad_norm": 0.06043306365609169,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 134883
+    },
+    {
+      "epoch": 13.850087277954616,
+      "grad_norm": 0.059762705117464066,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 134886
+    },
+    {
+      "epoch": 13.850395317794435,
+      "grad_norm": 0.03436395153403282,
+      "learning_rate": 0.01,
+      "loss": 1.8993,
+      "step": 134889
+    },
+    {
+      "epoch": 13.850703357634254,
+      "grad_norm": 0.05521237105131149,
+      "learning_rate": 0.01,
+      "loss": 1.9256,
+      "step": 134892
+    },
+    {
+      "epoch": 13.851011397474073,
+      "grad_norm": 0.11766976118087769,
+      "learning_rate": 0.01,
+      "loss": 1.8848,
+      "step": 134895
+    },
+    {
+      "epoch": 13.851319437313892,
+      "grad_norm": 0.06033182516694069,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 134898
+    },
+    {
+      "epoch": 13.851627477153713,
+      "grad_norm": 0.09229335933923721,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 134901
+    },
+    {
+      "epoch": 13.851935516993532,
+      "grad_norm": 0.07611782103776932,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 134904
+    },
+    {
+      "epoch": 13.85224355683335,
+      "grad_norm": 0.043599940836429596,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 134907
+    },
+    {
+      "epoch": 13.85255159667317,
+      "grad_norm": 0.11440790444612503,
+      "learning_rate": 0.01,
+      "loss": 1.9072,
+      "step": 134910
+    },
+    {
+      "epoch": 13.852859636512989,
+      "grad_norm": 0.08685220777988434,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 134913
+    },
+    {
+      "epoch": 13.853167676352808,
+      "grad_norm": 0.04386351257562637,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 134916
+    },
+    {
+      "epoch": 13.853475716192628,
+      "grad_norm": 0.08440227806568146,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 134919
+    },
+    {
+      "epoch": 13.853783756032447,
+      "grad_norm": 0.052824318408966064,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 134922
+    },
+    {
+      "epoch": 13.854091795872266,
+      "grad_norm": 0.04215586557984352,
+      "learning_rate": 0.01,
+      "loss": 1.9019,
+      "step": 134925
+    },
+    {
+      "epoch": 13.854399835712085,
+      "grad_norm": 0.04561490938067436,
+      "learning_rate": 0.01,
+      "loss": 1.9168,
+      "step": 134928
+    },
+    {
+      "epoch": 13.854707875551904,
+      "grad_norm": 0.048282820731401443,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 134931
+    },
+    {
+      "epoch": 13.855015915391723,
+      "grad_norm": 0.06694139540195465,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 134934
+    },
+    {
+      "epoch": 13.855323955231544,
+      "grad_norm": 0.10877640545368195,
+      "learning_rate": 0.01,
+      "loss": 1.9284,
+      "step": 134937
+    },
+    {
+      "epoch": 13.855631995071363,
+      "grad_norm": 0.037539348006248474,
+      "learning_rate": 0.01,
+      "loss": 1.9665,
+      "step": 134940
+    },
+    {
+      "epoch": 13.855940034911182,
+      "grad_norm": 0.07990458607673645,
+      "learning_rate": 0.01,
+      "loss": 1.8948,
+      "step": 134943
+    },
+    {
+      "epoch": 13.856248074751,
+      "grad_norm": 0.0478433221578598,
+      "learning_rate": 0.01,
+      "loss": 1.8977,
+      "step": 134946
+    },
+    {
+      "epoch": 13.85655611459082,
+      "grad_norm": 0.05559241399168968,
+      "learning_rate": 0.01,
+      "loss": 1.8987,
+      "step": 134949
+    },
+    {
+      "epoch": 13.856864154430639,
+      "grad_norm": 0.07102905213832855,
+      "learning_rate": 0.01,
+      "loss": 1.9245,
+      "step": 134952
+    },
+    {
+      "epoch": 13.85717219427046,
+      "grad_norm": 0.047799259424209595,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 134955
+    },
+    {
+      "epoch": 13.857480234110279,
+      "grad_norm": 0.0574326366186142,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 134958
+    },
+    {
+      "epoch": 13.857788273950097,
+      "grad_norm": 0.04035543277859688,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 134961
+    },
+    {
+      "epoch": 13.858096313789916,
+      "grad_norm": 0.0627676248550415,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 134964
+    },
+    {
+      "epoch": 13.858404353629735,
+      "grad_norm": 0.15775984525680542,
+      "learning_rate": 0.01,
+      "loss": 1.9189,
+      "step": 134967
+    },
+    {
+      "epoch": 13.858712393469556,
+      "grad_norm": 0.08861416578292847,
+      "learning_rate": 0.01,
+      "loss": 1.9377,
+      "step": 134970
+    },
+    {
+      "epoch": 13.859020433309375,
+      "grad_norm": 0.1528318226337433,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 134973
+    },
+    {
+      "epoch": 13.859328473149194,
+      "grad_norm": 0.08797655254602432,
+      "learning_rate": 0.01,
+      "loss": 1.9302,
+      "step": 134976
+    },
+    {
+      "epoch": 13.859636512989013,
+      "grad_norm": 0.057138632982969284,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 134979
+    },
+    {
+      "epoch": 13.859944552828832,
+      "grad_norm": 0.10003890097141266,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 134982
+    },
+    {
+      "epoch": 13.860252592668651,
+      "grad_norm": 0.0484062097966671,
+      "learning_rate": 0.01,
+      "loss": 1.8919,
+      "step": 134985
+    },
+    {
+      "epoch": 13.860560632508472,
+      "grad_norm": 0.03280618414282799,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 134988
+    },
+    {
+      "epoch": 13.86086867234829,
+      "grad_norm": 0.034922242164611816,
+      "learning_rate": 0.01,
+      "loss": 1.9154,
+      "step": 134991
+    },
+    {
+      "epoch": 13.86117671218811,
+      "grad_norm": 0.0517311617732048,
+      "learning_rate": 0.01,
+      "loss": 1.8935,
+      "step": 134994
+    },
+    {
+      "epoch": 13.861484752027929,
+      "grad_norm": 0.09517128020524979,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 134997
+    },
+    {
+      "epoch": 13.861792791867748,
+      "grad_norm": 0.04794229194521904,
+      "learning_rate": 0.01,
+      "loss": 1.902,
+      "step": 135000
+    },
+    {
+      "epoch": 13.862100831707567,
+      "grad_norm": 0.09047180414199829,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 135003
+    },
+    {
+      "epoch": 13.862408871547387,
+      "grad_norm": 0.0789877399802208,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 135006
+    },
+    {
+      "epoch": 13.862716911387206,
+      "grad_norm": 0.04831506684422493,
+      "learning_rate": 0.01,
+      "loss": 1.9252,
+      "step": 135009
+    },
+    {
+      "epoch": 13.863024951227025,
+      "grad_norm": 0.032399944961071014,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 135012
+    },
+    {
+      "epoch": 13.863332991066844,
+      "grad_norm": 0.03341751545667648,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 135015
+    },
+    {
+      "epoch": 13.863641030906663,
+      "grad_norm": 0.04685046896338463,
+      "learning_rate": 0.01,
+      "loss": 1.9037,
+      "step": 135018
+    },
+    {
+      "epoch": 13.863949070746482,
+      "grad_norm": 0.06693561375141144,
+      "learning_rate": 0.01,
+      "loss": 1.9381,
+      "step": 135021
+    },
+    {
+      "epoch": 13.864257110586303,
+      "grad_norm": 0.21458593010902405,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 135024
+    },
+    {
+      "epoch": 13.864565150426122,
+      "grad_norm": 0.18120095133781433,
+      "learning_rate": 0.01,
+      "loss": 1.9123,
+      "step": 135027
+    },
+    {
+      "epoch": 13.864873190265941,
+      "grad_norm": 0.08814983814954758,
+      "learning_rate": 0.01,
+      "loss": 1.932,
+      "step": 135030
+    },
+    {
+      "epoch": 13.86518123010576,
+      "grad_norm": 0.05778368562459946,
+      "learning_rate": 0.01,
+      "loss": 1.9233,
+      "step": 135033
+    },
+    {
+      "epoch": 13.865489269945579,
+      "grad_norm": 0.0430261455476284,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 135036
+    },
+    {
+      "epoch": 13.8657973097854,
+      "grad_norm": 0.03314177319407463,
+      "learning_rate": 0.01,
+      "loss": 1.9378,
+      "step": 135039
+    },
+    {
+      "epoch": 13.866105349625219,
+      "grad_norm": 0.05916096642613411,
+      "learning_rate": 0.01,
+      "loss": 1.8945,
+      "step": 135042
+    },
+    {
+      "epoch": 13.866413389465038,
+      "grad_norm": 0.03325483947992325,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 135045
+    },
+    {
+      "epoch": 13.866721429304857,
+      "grad_norm": 0.06269596517086029,
+      "learning_rate": 0.01,
+      "loss": 1.9163,
+      "step": 135048
+    },
+    {
+      "epoch": 13.867029469144676,
+      "grad_norm": 0.03281661123037338,
+      "learning_rate": 0.01,
+      "loss": 1.9287,
+      "step": 135051
+    },
+    {
+      "epoch": 13.867337508984495,
+      "grad_norm": 0.04659678041934967,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 135054
+    },
+    {
+      "epoch": 13.867645548824315,
+      "grad_norm": 0.07019969075918198,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 135057
+    },
+    {
+      "epoch": 13.867953588664134,
+      "grad_norm": 0.11671337485313416,
+      "learning_rate": 0.01,
+      "loss": 1.8915,
+      "step": 135060
+    },
+    {
+      "epoch": 13.868261628503953,
+      "grad_norm": 0.06813828647136688,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 135063
+    },
+    {
+      "epoch": 13.868569668343772,
+      "grad_norm": 0.10006043314933777,
+      "learning_rate": 0.01,
+      "loss": 1.9169,
+      "step": 135066
+    },
+    {
+      "epoch": 13.868877708183591,
+      "grad_norm": 0.10272083431482315,
+      "learning_rate": 0.01,
+      "loss": 1.9159,
+      "step": 135069
+    },
+    {
+      "epoch": 13.86918574802341,
+      "grad_norm": 0.061462871730327606,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 135072
+    },
+    {
+      "epoch": 13.869493787863231,
+      "grad_norm": 0.09475059062242508,
+      "learning_rate": 0.01,
+      "loss": 1.9066,
+      "step": 135075
+    },
+    {
+      "epoch": 13.86980182770305,
+      "grad_norm": 0.04875073954463005,
+      "learning_rate": 0.01,
+      "loss": 1.93,
+      "step": 135078
+    },
+    {
+      "epoch": 13.870109867542869,
+      "grad_norm": 0.04385347291827202,
+      "learning_rate": 0.01,
+      "loss": 1.9082,
+      "step": 135081
+    },
+    {
+      "epoch": 13.870417907382688,
+      "grad_norm": 0.03330356255173683,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 135084
+    },
+    {
+      "epoch": 13.870725947222507,
+      "grad_norm": 0.06052668020129204,
+      "learning_rate": 0.01,
+      "loss": 1.9179,
+      "step": 135087
+    },
+    {
+      "epoch": 13.871033987062326,
+      "grad_norm": 0.058853231370449066,
+      "learning_rate": 0.01,
+      "loss": 1.8992,
+      "step": 135090
+    },
+    {
+      "epoch": 13.871342026902147,
+      "grad_norm": 0.05946449562907219,
+      "learning_rate": 0.01,
+      "loss": 1.8867,
+      "step": 135093
+    },
+    {
+      "epoch": 13.871650066741966,
+      "grad_norm": 0.09672753512859344,
+      "learning_rate": 0.01,
+      "loss": 1.9138,
+      "step": 135096
+    },
+    {
+      "epoch": 13.871958106581785,
+      "grad_norm": 0.05656097084283829,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 135099
+    },
+    {
+      "epoch": 13.872266146421603,
+      "grad_norm": 0.06717727333307266,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 135102
+    },
+    {
+      "epoch": 13.872574186261422,
+      "grad_norm": 0.0597870871424675,
+      "learning_rate": 0.01,
+      "loss": 1.9115,
+      "step": 135105
+    },
+    {
+      "epoch": 13.872882226101243,
+      "grad_norm": 0.04756862670183182,
+      "learning_rate": 0.01,
+      "loss": 1.9094,
+      "step": 135108
+    },
+    {
+      "epoch": 13.873190265941062,
+      "grad_norm": 0.03676239028573036,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 135111
+    },
+    {
+      "epoch": 13.873498305780881,
+      "grad_norm": 0.11106175184249878,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 135114
+    },
+    {
+      "epoch": 13.8738063456207,
+      "grad_norm": 0.042229536920785904,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 135117
+    },
+    {
+      "epoch": 13.87411438546052,
+      "grad_norm": 0.10879723727703094,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 135120
+    },
+    {
+      "epoch": 13.874422425300338,
+      "grad_norm": 0.0655045285820961,
+      "learning_rate": 0.01,
+      "loss": 1.897,
+      "step": 135123
+    },
+    {
+      "epoch": 13.874730465140159,
+      "grad_norm": 0.04873618111014366,
+      "learning_rate": 0.01,
+      "loss": 1.9255,
+      "step": 135126
+    },
+    {
+      "epoch": 13.875038504979978,
+      "grad_norm": 0.03938988968729973,
+      "learning_rate": 0.01,
+      "loss": 1.9067,
+      "step": 135129
+    },
+    {
+      "epoch": 13.875346544819797,
+      "grad_norm": 0.1056794598698616,
+      "learning_rate": 0.01,
+      "loss": 1.9357,
+      "step": 135132
+    },
+    {
+      "epoch": 13.875654584659616,
+      "grad_norm": 0.10446460545063019,
+      "learning_rate": 0.01,
+      "loss": 1.9091,
+      "step": 135135
+    },
+    {
+      "epoch": 13.875962624499435,
+      "grad_norm": 0.0686206966638565,
+      "learning_rate": 0.01,
+      "loss": 1.9448,
+      "step": 135138
+    },
+    {
+      "epoch": 13.876270664339254,
+      "grad_norm": 0.04981555417180061,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 135141
+    },
+    {
+      "epoch": 13.876578704179074,
+      "grad_norm": 0.11834550648927689,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 135144
+    },
+    {
+      "epoch": 13.876886744018893,
+      "grad_norm": 0.046607282012701035,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 135147
+    },
+    {
+      "epoch": 13.877194783858712,
+      "grad_norm": 0.048484865576028824,
+      "learning_rate": 0.01,
+      "loss": 1.9314,
+      "step": 135150
+    },
+    {
+      "epoch": 13.877502823698531,
+      "grad_norm": 0.04563921317458153,
+      "learning_rate": 0.01,
+      "loss": 1.942,
+      "step": 135153
+    },
+    {
+      "epoch": 13.87781086353835,
+      "grad_norm": 0.04317391663789749,
+      "learning_rate": 0.01,
+      "loss": 1.9416,
+      "step": 135156
+    },
+    {
+      "epoch": 13.87811890337817,
+      "grad_norm": 0.05740208178758621,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 135159
+    },
+    {
+      "epoch": 13.87842694321799,
+      "grad_norm": 0.1365426629781723,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 135162
+    },
+    {
+      "epoch": 13.878734983057809,
+      "grad_norm": 0.04039986804127693,
+      "learning_rate": 0.01,
+      "loss": 1.9236,
+      "step": 135165
+    },
+    {
+      "epoch": 13.879043022897628,
+      "grad_norm": 0.08109308034181595,
+      "learning_rate": 0.01,
+      "loss": 1.9166,
+      "step": 135168
+    },
+    {
+      "epoch": 13.879351062737447,
+      "grad_norm": 0.04718312993645668,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 135171
+    },
+    {
+      "epoch": 13.879659102577266,
+      "grad_norm": 0.04760252311825752,
+      "learning_rate": 0.01,
+      "loss": 1.9338,
+      "step": 135174
+    },
+    {
+      "epoch": 13.879967142417087,
+      "grad_norm": 0.07432134449481964,
+      "learning_rate": 0.01,
+      "loss": 1.9176,
+      "step": 135177
+    },
+    {
+      "epoch": 13.880275182256906,
+      "grad_norm": 0.046237751841545105,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 135180
+    },
+    {
+      "epoch": 13.880583222096725,
+      "grad_norm": 0.041562505066394806,
+      "learning_rate": 0.01,
+      "loss": 1.8957,
+      "step": 135183
+    },
+    {
+      "epoch": 13.880891261936544,
+      "grad_norm": 0.11060908436775208,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 135186
+    },
+    {
+      "epoch": 13.881199301776363,
+      "grad_norm": 0.03289926424622536,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 135189
+    },
+    {
+      "epoch": 13.881507341616182,
+      "grad_norm": 0.1026652455329895,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 135192
+    },
+    {
+      "epoch": 13.881815381456002,
+      "grad_norm": 0.039885275065898895,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 135195
+    },
+    {
+      "epoch": 13.882123421295821,
+      "grad_norm": 0.13234183192253113,
+      "learning_rate": 0.01,
+      "loss": 1.9212,
+      "step": 135198
+    },
+    {
+      "epoch": 13.88243146113564,
+      "grad_norm": 0.15088574588298798,
+      "learning_rate": 0.01,
+      "loss": 1.9318,
+      "step": 135201
+    },
+    {
+      "epoch": 13.88273950097546,
+      "grad_norm": 0.12011145055294037,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 135204
+    },
+    {
+      "epoch": 13.883047540815278,
+      "grad_norm": 0.07452481240034103,
+      "learning_rate": 0.01,
+      "loss": 1.912,
+      "step": 135207
+    },
+    {
+      "epoch": 13.883355580655097,
+      "grad_norm": 0.051672015339136124,
+      "learning_rate": 0.01,
+      "loss": 1.9096,
+      "step": 135210
+    },
+    {
+      "epoch": 13.883663620494918,
+      "grad_norm": 0.038564782589673996,
+      "learning_rate": 0.01,
+      "loss": 1.931,
+      "step": 135213
+    },
+    {
+      "epoch": 13.883971660334737,
+      "grad_norm": 0.051461298018693924,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 135216
+    },
+    {
+      "epoch": 13.884279700174556,
+      "grad_norm": 0.0350487157702446,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 135219
+    },
+    {
+      "epoch": 13.884587740014375,
+      "grad_norm": 0.036131709814071655,
+      "learning_rate": 0.01,
+      "loss": 1.9238,
+      "step": 135222
+    },
+    {
+      "epoch": 13.884895779854194,
+      "grad_norm": 0.13136154413223267,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 135225
+    },
+    {
+      "epoch": 13.885203819694013,
+      "grad_norm": 0.038287337869405746,
+      "learning_rate": 0.01,
+      "loss": 1.8982,
+      "step": 135228
+    },
+    {
+      "epoch": 13.885511859533834,
+      "grad_norm": 0.050868671387434006,
+      "learning_rate": 0.01,
+      "loss": 1.9364,
+      "step": 135231
+    },
+    {
+      "epoch": 13.885819899373653,
+      "grad_norm": 0.051800474524497986,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 135234
+    },
+    {
+      "epoch": 13.886127939213472,
+      "grad_norm": 0.04659080132842064,
+      "learning_rate": 0.01,
+      "loss": 1.9064,
+      "step": 135237
+    },
+    {
+      "epoch": 13.88643597905329,
+      "grad_norm": 0.042710818350315094,
+      "learning_rate": 0.01,
+      "loss": 1.9133,
+      "step": 135240
+    },
+    {
+      "epoch": 13.88674401889311,
+      "grad_norm": 0.042907848954200745,
+      "learning_rate": 0.01,
+      "loss": 1.9105,
+      "step": 135243
+    },
+    {
+      "epoch": 13.88705205873293,
+      "grad_norm": 0.05924863740801811,
+      "learning_rate": 0.01,
+      "loss": 1.8923,
+      "step": 135246
+    },
+    {
+      "epoch": 13.88736009857275,
+      "grad_norm": 0.03258752450346947,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 135249
+    },
+    {
+      "epoch": 13.887668138412568,
+      "grad_norm": 0.15001289546489716,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 135252
+    },
+    {
+      "epoch": 13.887976178252387,
+      "grad_norm": 0.05113843455910683,
+      "learning_rate": 0.01,
+      "loss": 1.9038,
+      "step": 135255
+    },
+    {
+      "epoch": 13.888284218092206,
+      "grad_norm": 0.06070922315120697,
+      "learning_rate": 0.01,
+      "loss": 1.8969,
+      "step": 135258
+    },
+    {
+      "epoch": 13.888592257932025,
+      "grad_norm": 0.06722289323806763,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 135261
+    },
+    {
+      "epoch": 13.888900297771846,
+      "grad_norm": 0.03752301633358002,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 135264
+    },
+    {
+      "epoch": 13.889208337611665,
+      "grad_norm": 0.08959244191646576,
+      "learning_rate": 0.01,
+      "loss": 1.8929,
+      "step": 135267
+    },
+    {
+      "epoch": 13.889516377451484,
+      "grad_norm": 0.06179090589284897,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 135270
+    },
+    {
+      "epoch": 13.889824417291303,
+      "grad_norm": 0.07323700934648514,
+      "learning_rate": 0.01,
+      "loss": 1.9481,
+      "step": 135273
+    },
+    {
+      "epoch": 13.890132457131122,
+      "grad_norm": 0.039218924939632416,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 135276
+    },
+    {
+      "epoch": 13.89044049697094,
+      "grad_norm": 0.03275348246097565,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 135279
+    },
+    {
+      "epoch": 13.890748536810761,
+      "grad_norm": 0.05403297767043114,
+      "learning_rate": 0.01,
+      "loss": 1.9332,
+      "step": 135282
+    },
+    {
+      "epoch": 13.89105657665058,
+      "grad_norm": 0.06202052906155586,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 135285
+    },
+    {
+      "epoch": 13.8913646164904,
+      "grad_norm": 0.10926573723554611,
+      "learning_rate": 0.01,
+      "loss": 1.9081,
+      "step": 135288
+    },
+    {
+      "epoch": 13.891672656330218,
+      "grad_norm": 0.036057569086551666,
+      "learning_rate": 0.01,
+      "loss": 1.9135,
+      "step": 135291
+    },
+    {
+      "epoch": 13.891980696170037,
+      "grad_norm": 0.0959005206823349,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 135294
+    },
+    {
+      "epoch": 13.892288736009856,
+      "grad_norm": 0.10793931782245636,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 135297
+    },
+    {
+      "epoch": 13.892596775849677,
+      "grad_norm": 0.04562230035662651,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 135300
+    },
+    {
+      "epoch": 13.892904815689496,
+      "grad_norm": 0.04484694451093674,
+      "learning_rate": 0.01,
+      "loss": 1.9042,
+      "step": 135303
+    },
+    {
+      "epoch": 13.893212855529315,
+      "grad_norm": 0.03665325790643692,
+      "learning_rate": 0.01,
+      "loss": 1.9305,
+      "step": 135306
+    },
+    {
+      "epoch": 13.893520895369134,
+      "grad_norm": 0.13382597267627716,
+      "learning_rate": 0.01,
+      "loss": 1.9116,
+      "step": 135309
+    },
+    {
+      "epoch": 13.893828935208953,
+      "grad_norm": 0.07904886454343796,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 135312
+    },
+    {
+      "epoch": 13.894136975048774,
+      "grad_norm": 0.058125536888837814,
+      "learning_rate": 0.01,
+      "loss": 1.9043,
+      "step": 135315
+    },
+    {
+      "epoch": 13.894445014888593,
+      "grad_norm": 0.06731139868497849,
+      "learning_rate": 0.01,
+      "loss": 1.9112,
+      "step": 135318
+    },
+    {
+      "epoch": 13.894753054728412,
+      "grad_norm": 0.032721228897571564,
+      "learning_rate": 0.01,
+      "loss": 1.9158,
+      "step": 135321
+    },
+    {
+      "epoch": 13.89506109456823,
+      "grad_norm": 0.03958245366811752,
+      "learning_rate": 0.01,
+      "loss": 1.9417,
+      "step": 135324
+    },
+    {
+      "epoch": 13.89536913440805,
+      "grad_norm": 0.036765206605196,
+      "learning_rate": 0.01,
+      "loss": 1.9298,
+      "step": 135327
+    },
+    {
+      "epoch": 13.895677174247869,
+      "grad_norm": 0.05205739289522171,
+      "learning_rate": 0.01,
+      "loss": 1.8924,
+      "step": 135330
+    },
+    {
+      "epoch": 13.89598521408769,
+      "grad_norm": 0.08795994520187378,
+      "learning_rate": 0.01,
+      "loss": 1.9155,
+      "step": 135333
+    },
+    {
+      "epoch": 13.896293253927508,
+      "grad_norm": 0.0738556981086731,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 135336
+    },
+    {
+      "epoch": 13.896601293767327,
+      "grad_norm": 0.11296192556619644,
+      "learning_rate": 0.01,
+      "loss": 1.8917,
+      "step": 135339
+    },
+    {
+      "epoch": 13.896909333607146,
+      "grad_norm": 0.039331790059804916,
+      "learning_rate": 0.01,
+      "loss": 1.9457,
+      "step": 135342
+    },
+    {
+      "epoch": 13.897217373446965,
+      "grad_norm": 0.03761405870318413,
+      "learning_rate": 0.01,
+      "loss": 1.941,
+      "step": 135345
+    },
+    {
+      "epoch": 13.897525413286786,
+      "grad_norm": 0.047794997692108154,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 135348
+    },
+    {
+      "epoch": 13.897833453126605,
+      "grad_norm": 0.06860291212797165,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 135351
+    },
+    {
+      "epoch": 13.898141492966424,
+      "grad_norm": 0.05428686365485191,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 135354
+    },
+    {
+      "epoch": 13.898449532806243,
+      "grad_norm": 0.056280024349689484,
+      "learning_rate": 0.01,
+      "loss": 1.936,
+      "step": 135357
+    },
+    {
+      "epoch": 13.898757572646062,
+      "grad_norm": 0.06364693492650986,
+      "learning_rate": 0.01,
+      "loss": 1.9046,
+      "step": 135360
+    },
+    {
+      "epoch": 13.899065612485881,
+      "grad_norm": 0.0528084859251976,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 135363
+    },
+    {
+      "epoch": 13.8993736523257,
+      "grad_norm": 0.04407433792948723,
+      "learning_rate": 0.01,
+      "loss": 1.9249,
+      "step": 135366
+    },
+    {
+      "epoch": 13.89968169216552,
+      "grad_norm": 0.1101018413901329,
+      "learning_rate": 0.01,
+      "loss": 1.9286,
+      "step": 135369
+    },
+    {
+      "epoch": 13.89998973200534,
+      "grad_norm": 0.07078106701374054,
+      "learning_rate": 0.01,
+      "loss": 1.9196,
+      "step": 135372
+    },
+    {
+      "epoch": 13.900297771845159,
+      "grad_norm": 0.08511429280042648,
+      "learning_rate": 0.01,
+      "loss": 1.9209,
+      "step": 135375
+    },
+    {
+      "epoch": 13.900605811684978,
+      "grad_norm": 0.04628182575106621,
+      "learning_rate": 0.01,
+      "loss": 1.9023,
+      "step": 135378
+    },
+    {
+      "epoch": 13.900913851524797,
+      "grad_norm": 0.12361085414886475,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 135381
+    },
+    {
+      "epoch": 13.901221891364617,
+      "grad_norm": 0.06140122935175896,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 135384
+    },
+    {
+      "epoch": 13.901529931204436,
+      "grad_norm": 0.06070174649357796,
+      "learning_rate": 0.01,
+      "loss": 1.8937,
+      "step": 135387
+    },
+    {
+      "epoch": 13.901837971044255,
+      "grad_norm": 0.05376163870096207,
+      "learning_rate": 0.01,
+      "loss": 1.9246,
+      "step": 135390
+    },
+    {
+      "epoch": 13.902146010884074,
+      "grad_norm": 0.04665801301598549,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 135393
+    },
+    {
+      "epoch": 13.902454050723893,
+      "grad_norm": 0.03599710017442703,
+      "learning_rate": 0.01,
+      "loss": 1.9061,
+      "step": 135396
+    },
+    {
+      "epoch": 13.902762090563712,
+      "grad_norm": 0.03828664869070053,
+      "learning_rate": 0.01,
+      "loss": 1.8989,
+      "step": 135399
+    },
+    {
+      "epoch": 13.903070130403533,
+      "grad_norm": 0.1455589383840561,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 135402
+    },
+    {
+      "epoch": 13.903378170243352,
+      "grad_norm": 0.04254241660237312,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 135405
+    },
+    {
+      "epoch": 13.90368621008317,
+      "grad_norm": 0.09016977995634079,
+      "learning_rate": 0.01,
+      "loss": 1.9086,
+      "step": 135408
+    },
+    {
+      "epoch": 13.90399424992299,
+      "grad_norm": 0.09002036601305008,
+      "learning_rate": 0.01,
+      "loss": 1.9171,
+      "step": 135411
+    },
+    {
+      "epoch": 13.904302289762809,
+      "grad_norm": 0.09199529886245728,
+      "learning_rate": 0.01,
+      "loss": 1.8907,
+      "step": 135414
+    },
+    {
+      "epoch": 13.90461032960263,
+      "grad_norm": 0.06279487162828445,
+      "learning_rate": 0.01,
+      "loss": 1.9181,
+      "step": 135417
+    },
+    {
+      "epoch": 13.904918369442449,
+      "grad_norm": 0.03712217137217522,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 135420
+    },
+    {
+      "epoch": 13.905226409282267,
+      "grad_norm": 0.03726033493876457,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 135423
+    },
+    {
+      "epoch": 13.905534449122086,
+      "grad_norm": 0.03238022327423096,
+      "learning_rate": 0.01,
+      "loss": 1.899,
+      "step": 135426
+    },
+    {
+      "epoch": 13.905842488961905,
+      "grad_norm": 0.0455828458070755,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 135429
+    },
+    {
+      "epoch": 13.906150528801724,
+      "grad_norm": 0.04420851916074753,
+      "learning_rate": 0.01,
+      "loss": 1.8865,
+      "step": 135432
+    },
+    {
+      "epoch": 13.906458568641543,
+      "grad_norm": 0.05586368963122368,
+      "learning_rate": 0.01,
+      "loss": 1.929,
+      "step": 135435
+    },
+    {
+      "epoch": 13.906766608481364,
+      "grad_norm": 0.08785773813724518,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 135438
+    },
+    {
+      "epoch": 13.907074648321183,
+      "grad_norm": 0.06627316772937775,
+      "learning_rate": 0.01,
+      "loss": 1.9333,
+      "step": 135441
+    },
+    {
+      "epoch": 13.907382688161002,
+      "grad_norm": 0.03906713426113129,
+      "learning_rate": 0.01,
+      "loss": 1.9392,
+      "step": 135444
+    },
+    {
+      "epoch": 13.907690728000821,
+      "grad_norm": 0.06448914110660553,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 135447
+    },
+    {
+      "epoch": 13.90799876784064,
+      "grad_norm": 0.1027592271566391,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 135450
+    },
+    {
+      "epoch": 13.90830680768046,
+      "grad_norm": 0.060778722167015076,
+      "learning_rate": 0.01,
+      "loss": 1.9356,
+      "step": 135453
+    },
+    {
+      "epoch": 13.90861484752028,
+      "grad_norm": 0.09912999719381332,
+      "learning_rate": 0.01,
+      "loss": 1.9204,
+      "step": 135456
+    },
+    {
+      "epoch": 13.908922887360099,
+      "grad_norm": 0.07650015503168106,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 135459
+    },
+    {
+      "epoch": 13.909230927199918,
+      "grad_norm": 0.050851721316576004,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 135462
+    },
+    {
+      "epoch": 13.909538967039737,
+      "grad_norm": 0.07006032019853592,
+      "learning_rate": 0.01,
+      "loss": 1.9117,
+      "step": 135465
+    },
+    {
+      "epoch": 13.909847006879556,
+      "grad_norm": 0.11382680386304855,
+      "learning_rate": 0.01,
+      "loss": 1.9207,
+      "step": 135468
+    },
+    {
+      "epoch": 13.910155046719376,
+      "grad_norm": 0.07516433298587799,
+      "learning_rate": 0.01,
+      "loss": 1.9273,
+      "step": 135471
+    },
+    {
+      "epoch": 13.910463086559195,
+      "grad_norm": 0.08811305463314056,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 135474
+    },
+    {
+      "epoch": 13.910771126399014,
+      "grad_norm": 0.0829676017165184,
+      "learning_rate": 0.01,
+      "loss": 1.905,
+      "step": 135477
+    },
+    {
+      "epoch": 13.911079166238833,
+      "grad_norm": 0.04312453791499138,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 135480
+    },
+    {
+      "epoch": 13.911387206078652,
+      "grad_norm": 0.05931262671947479,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 135483
+    },
+    {
+      "epoch": 13.911695245918473,
+      "grad_norm": 0.03666198253631592,
+      "learning_rate": 0.01,
+      "loss": 1.8998,
+      "step": 135486
+    },
+    {
+      "epoch": 13.912003285758292,
+      "grad_norm": 0.04679884761571884,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 135489
+    },
+    {
+      "epoch": 13.912311325598111,
+      "grad_norm": 0.06975444406270981,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 135492
+    },
+    {
+      "epoch": 13.91261936543793,
+      "grad_norm": 0.12790247797966003,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 135495
+    },
+    {
+      "epoch": 13.912927405277749,
+      "grad_norm": 0.1372372955083847,
+      "learning_rate": 0.01,
+      "loss": 1.9276,
+      "step": 135498
+    },
+    {
+      "epoch": 13.913235445117568,
+      "grad_norm": 0.08832630515098572,
+      "learning_rate": 0.01,
+      "loss": 1.9048,
+      "step": 135501
+    },
+    {
+      "epoch": 13.913543484957387,
+      "grad_norm": 0.10211103409528732,
+      "learning_rate": 0.01,
+      "loss": 1.8774,
+      "step": 135504
+    },
+    {
+      "epoch": 13.913851524797208,
+      "grad_norm": 0.05900106579065323,
+      "learning_rate": 0.01,
+      "loss": 1.8933,
+      "step": 135507
+    },
+    {
+      "epoch": 13.914159564637027,
+      "grad_norm": 0.04350128024816513,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 135510
+    },
+    {
+      "epoch": 13.914467604476846,
+      "grad_norm": 0.06192927807569504,
+      "learning_rate": 0.01,
+      "loss": 1.9198,
+      "step": 135513
+    },
+    {
+      "epoch": 13.914775644316665,
+      "grad_norm": 0.05437299981713295,
+      "learning_rate": 0.01,
+      "loss": 1.927,
+      "step": 135516
+    },
+    {
+      "epoch": 13.915083684156484,
+      "grad_norm": 0.04106966033577919,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 135519
+    },
+    {
+      "epoch": 13.915391723996304,
+      "grad_norm": 0.05908875912427902,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 135522
+    },
+    {
+      "epoch": 13.915699763836123,
+      "grad_norm": 0.12403882294893265,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 135525
+    },
+    {
+      "epoch": 13.916007803675942,
+      "grad_norm": 0.13403476774692535,
+      "learning_rate": 0.01,
+      "loss": 1.947,
+      "step": 135528
+    },
+    {
+      "epoch": 13.916315843515761,
+      "grad_norm": 0.0481337271630764,
+      "learning_rate": 0.01,
+      "loss": 1.9225,
+      "step": 135531
+    },
+    {
+      "epoch": 13.91662388335558,
+      "grad_norm": 0.04039508104324341,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 135534
+    },
+    {
+      "epoch": 13.9169319231954,
+      "grad_norm": 0.05781160667538643,
+      "learning_rate": 0.01,
+      "loss": 1.8933,
+      "step": 135537
+    },
+    {
+      "epoch": 13.91723996303522,
+      "grad_norm": 0.06371809542179108,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 135540
+    },
+    {
+      "epoch": 13.917548002875039,
+      "grad_norm": 0.06895791739225388,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 135543
+    },
+    {
+      "epoch": 13.917856042714858,
+      "grad_norm": 0.052721694111824036,
+      "learning_rate": 0.01,
+      "loss": 1.9288,
+      "step": 135546
+    },
+    {
+      "epoch": 13.918164082554677,
+      "grad_norm": 0.039400748908519745,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 135549
+    },
+    {
+      "epoch": 13.918472122394496,
+      "grad_norm": 0.030469093471765518,
+      "learning_rate": 0.01,
+      "loss": 1.9152,
+      "step": 135552
+    },
+    {
+      "epoch": 13.918780162234317,
+      "grad_norm": 0.06258850544691086,
+      "learning_rate": 0.01,
+      "loss": 1.9182,
+      "step": 135555
+    },
+    {
+      "epoch": 13.919088202074136,
+      "grad_norm": 0.1503921002149582,
+      "learning_rate": 0.01,
+      "loss": 1.9216,
+      "step": 135558
+    },
+    {
+      "epoch": 13.919396241913955,
+      "grad_norm": 0.038195397704839706,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 135561
+    },
+    {
+      "epoch": 13.919704281753774,
+      "grad_norm": 0.08914259076118469,
+      "learning_rate": 0.01,
+      "loss": 1.9474,
+      "step": 135564
+    },
+    {
+      "epoch": 13.920012321593592,
+      "grad_norm": 0.1331617832183838,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 135567
+    },
+    {
+      "epoch": 13.920320361433411,
+      "grad_norm": 0.12877358496189117,
+      "learning_rate": 0.01,
+      "loss": 1.9399,
+      "step": 135570
+    },
+    {
+      "epoch": 13.92062840127323,
+      "grad_norm": 0.07954861223697662,
+      "learning_rate": 0.01,
+      "loss": 1.9272,
+      "step": 135573
+    },
+    {
+      "epoch": 13.920936441113051,
+      "grad_norm": 0.06344691663980484,
+      "learning_rate": 0.01,
+      "loss": 1.9409,
+      "step": 135576
+    },
+    {
+      "epoch": 13.92124448095287,
+      "grad_norm": 0.042949430644512177,
+      "learning_rate": 0.01,
+      "loss": 1.946,
+      "step": 135579
+    },
+    {
+      "epoch": 13.92155252079269,
+      "grad_norm": 0.028374651446938515,
+      "learning_rate": 0.01,
+      "loss": 1.921,
+      "step": 135582
+    },
+    {
+      "epoch": 13.921860560632508,
+      "grad_norm": 0.03140632435679436,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 135585
+    },
+    {
+      "epoch": 13.922168600472327,
+      "grad_norm": 0.04086381196975708,
+      "learning_rate": 0.01,
+      "loss": 1.8984,
+      "step": 135588
+    },
+    {
+      "epoch": 13.922476640312148,
+      "grad_norm": 0.1023203581571579,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 135591
+    },
+    {
+      "epoch": 13.922784680151967,
+      "grad_norm": 0.07584439218044281,
+      "learning_rate": 0.01,
+      "loss": 1.9346,
+      "step": 135594
+    },
+    {
+      "epoch": 13.923092719991786,
+      "grad_norm": 0.0880453884601593,
+      "learning_rate": 0.01,
+      "loss": 1.9316,
+      "step": 135597
+    },
+    {
+      "epoch": 13.923400759831605,
+      "grad_norm": 0.03644146770238876,
+      "learning_rate": 0.01,
+      "loss": 1.9339,
+      "step": 135600
+    },
+    {
+      "epoch": 13.923708799671424,
+      "grad_norm": 0.0340883731842041,
+      "learning_rate": 0.01,
+      "loss": 1.9513,
+      "step": 135603
+    },
+    {
+      "epoch": 13.924016839511243,
+      "grad_norm": 0.03621610626578331,
+      "learning_rate": 0.01,
+      "loss": 1.9162,
+      "step": 135606
+    },
+    {
+      "epoch": 13.924324879351063,
+      "grad_norm": 0.043163832277059555,
+      "learning_rate": 0.01,
+      "loss": 1.9343,
+      "step": 135609
+    },
+    {
+      "epoch": 13.924632919190882,
+      "grad_norm": 0.03737865760922432,
+      "learning_rate": 0.01,
+      "loss": 1.9132,
+      "step": 135612
+    },
+    {
+      "epoch": 13.924940959030701,
+      "grad_norm": 0.0871436670422554,
+      "learning_rate": 0.01,
+      "loss": 1.8803,
+      "step": 135615
+    },
+    {
+      "epoch": 13.92524899887052,
+      "grad_norm": 0.11796832084655762,
+      "learning_rate": 0.01,
+      "loss": 1.926,
+      "step": 135618
+    },
+    {
+      "epoch": 13.92555703871034,
+      "grad_norm": 0.08608805388212204,
+      "learning_rate": 0.01,
+      "loss": 1.9053,
+      "step": 135621
+    },
+    {
+      "epoch": 13.92586507855016,
+      "grad_norm": 0.05203809589147568,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 135624
+    },
+    {
+      "epoch": 13.926173118389979,
+      "grad_norm": 0.07106944173574448,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 135627
+    },
+    {
+      "epoch": 13.926481158229798,
+      "grad_norm": 0.05680394172668457,
+      "learning_rate": 0.01,
+      "loss": 1.9085,
+      "step": 135630
+    },
+    {
+      "epoch": 13.926789198069617,
+      "grad_norm": 0.056109726428985596,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 135633
+    },
+    {
+      "epoch": 13.927097237909436,
+      "grad_norm": 0.08902611583471298,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 135636
+    },
+    {
+      "epoch": 13.927405277749255,
+      "grad_norm": 0.03971083462238312,
+      "learning_rate": 0.01,
+      "loss": 1.9248,
+      "step": 135639
+    },
+    {
+      "epoch": 13.927713317589074,
+      "grad_norm": 0.04642011225223541,
+      "learning_rate": 0.01,
+      "loss": 1.9429,
+      "step": 135642
+    },
+    {
+      "epoch": 13.928021357428895,
+      "grad_norm": 0.08556116372346878,
+      "learning_rate": 0.01,
+      "loss": 1.9172,
+      "step": 135645
+    },
+    {
+      "epoch": 13.928329397268714,
+      "grad_norm": 0.0967288389801979,
+      "learning_rate": 0.01,
+      "loss": 1.8961,
+      "step": 135648
+    },
+    {
+      "epoch": 13.928637437108533,
+      "grad_norm": 0.0669083446264267,
+      "learning_rate": 0.01,
+      "loss": 1.919,
+      "step": 135651
+    },
+    {
+      "epoch": 13.928945476948352,
+      "grad_norm": 0.030001206323504448,
+      "learning_rate": 0.01,
+      "loss": 1.9376,
+      "step": 135654
+    },
+    {
+      "epoch": 13.92925351678817,
+      "grad_norm": 0.06296877562999725,
+      "learning_rate": 0.01,
+      "loss": 1.9005,
+      "step": 135657
+    },
+    {
+      "epoch": 13.929561556627991,
+      "grad_norm": 0.07407094538211823,
+      "learning_rate": 0.01,
+      "loss": 1.938,
+      "step": 135660
+    },
+    {
+      "epoch": 13.92986959646781,
+      "grad_norm": 0.09001734107732773,
+      "learning_rate": 0.01,
+      "loss": 1.9139,
+      "step": 135663
+    },
+    {
+      "epoch": 13.93017763630763,
+      "grad_norm": 0.07464145869016647,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 135666
+    },
+    {
+      "epoch": 13.930485676147448,
+      "grad_norm": 0.03771296888589859,
+      "learning_rate": 0.01,
+      "loss": 1.8994,
+      "step": 135669
+    },
+    {
+      "epoch": 13.930793715987267,
+      "grad_norm": 0.030687393620610237,
+      "learning_rate": 0.01,
+      "loss": 1.9075,
+      "step": 135672
+    },
+    {
+      "epoch": 13.931101755827086,
+      "grad_norm": 0.05820400267839432,
+      "learning_rate": 0.01,
+      "loss": 1.9373,
+      "step": 135675
+    },
+    {
+      "epoch": 13.931409795666907,
+      "grad_norm": 0.03644741699099541,
+      "learning_rate": 0.01,
+      "loss": 1.9164,
+      "step": 135678
+    },
+    {
+      "epoch": 13.931717835506726,
+      "grad_norm": 0.035701751708984375,
+      "learning_rate": 0.01,
+      "loss": 1.8845,
+      "step": 135681
+    },
+    {
+      "epoch": 13.932025875346545,
+      "grad_norm": 0.12128648161888123,
+      "learning_rate": 0.01,
+      "loss": 1.8877,
+      "step": 135684
+    },
+    {
+      "epoch": 13.932333915186364,
+      "grad_norm": 0.10074978321790695,
+      "learning_rate": 0.01,
+      "loss": 1.9177,
+      "step": 135687
+    },
+    {
+      "epoch": 13.932641955026183,
+      "grad_norm": 0.10312608629465103,
+      "learning_rate": 0.01,
+      "loss": 1.9193,
+      "step": 135690
+    },
+    {
+      "epoch": 13.932949994866004,
+      "grad_norm": 0.09056086838245392,
+      "learning_rate": 0.01,
+      "loss": 1.9438,
+      "step": 135693
+    },
+    {
+      "epoch": 13.933258034705823,
+      "grad_norm": 0.07947687804698944,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 135696
+    },
+    {
+      "epoch": 13.933566074545642,
+      "grad_norm": 0.043094437569379807,
+      "learning_rate": 0.01,
+      "loss": 1.9077,
+      "step": 135699
+    },
+    {
+      "epoch": 13.93387411438546,
+      "grad_norm": 0.0779462680220604,
+      "learning_rate": 0.01,
+      "loss": 1.9237,
+      "step": 135702
+    },
+    {
+      "epoch": 13.93418215422528,
+      "grad_norm": 0.051362283527851105,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 135705
+    },
+    {
+      "epoch": 13.934490194065098,
+      "grad_norm": 0.09989504516124725,
+      "learning_rate": 0.01,
+      "loss": 1.9619,
+      "step": 135708
+    },
+    {
+      "epoch": 13.934798233904917,
+      "grad_norm": 0.10736594349145889,
+      "learning_rate": 0.01,
+      "loss": 1.8955,
+      "step": 135711
+    },
+    {
+      "epoch": 13.935106273744738,
+      "grad_norm": 0.09344949573278427,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 135714
+    },
+    {
+      "epoch": 13.935414313584557,
+      "grad_norm": 0.06931407004594803,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 135717
+    },
+    {
+      "epoch": 13.935722353424376,
+      "grad_norm": 0.07043091207742691,
+      "learning_rate": 0.01,
+      "loss": 1.9257,
+      "step": 135720
+    },
+    {
+      "epoch": 13.936030393264195,
+      "grad_norm": 0.08513464778661728,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 135723
+    },
+    {
+      "epoch": 13.936338433104014,
+      "grad_norm": 0.04891832917928696,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 135726
+    },
+    {
+      "epoch": 13.936646472943835,
+      "grad_norm": 0.12543024122714996,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 135729
+    },
+    {
+      "epoch": 13.936954512783654,
+      "grad_norm": 0.046057265251874924,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 135732
+    },
+    {
+      "epoch": 13.937262552623473,
+      "grad_norm": 0.13384684920310974,
+      "learning_rate": 0.01,
+      "loss": 1.9106,
+      "step": 135735
+    },
+    {
+      "epoch": 13.937570592463292,
+      "grad_norm": 0.04958835989236832,
+      "learning_rate": 0.01,
+      "loss": 1.918,
+      "step": 135738
+    },
+    {
+      "epoch": 13.93787863230311,
+      "grad_norm": 0.043337807059288025,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 135741
+    },
+    {
+      "epoch": 13.93818667214293,
+      "grad_norm": 0.07438129186630249,
+      "learning_rate": 0.01,
+      "loss": 1.9307,
+      "step": 135744
+    },
+    {
+      "epoch": 13.93849471198275,
+      "grad_norm": 0.0900125801563263,
+      "learning_rate": 0.01,
+      "loss": 1.9259,
+      "step": 135747
+    },
+    {
+      "epoch": 13.93880275182257,
+      "grad_norm": 0.031092017889022827,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 135750
+    },
+    {
+      "epoch": 13.939110791662388,
+      "grad_norm": 0.056624386459589005,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 135753
+    },
+    {
+      "epoch": 13.939418831502207,
+      "grad_norm": 0.13297538459300995,
+      "learning_rate": 0.01,
+      "loss": 1.8961,
+      "step": 135756
+    },
+    {
+      "epoch": 13.939726871342026,
+      "grad_norm": 0.1548391431570053,
+      "learning_rate": 0.01,
+      "loss": 1.9262,
+      "step": 135759
+    },
+    {
+      "epoch": 13.940034911181847,
+      "grad_norm": 0.052177879959344864,
+      "learning_rate": 0.01,
+      "loss": 1.9199,
+      "step": 135762
+    },
+    {
+      "epoch": 13.940342951021666,
+      "grad_norm": 0.06955404579639435,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 135765
+    },
+    {
+      "epoch": 13.940650990861485,
+      "grad_norm": 0.04738903418183327,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 135768
+    },
+    {
+      "epoch": 13.940959030701304,
+      "grad_norm": 0.049770791083574295,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 135771
+    },
+    {
+      "epoch": 13.941267070541123,
+      "grad_norm": 0.04803701490163803,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 135774
+    },
+    {
+      "epoch": 13.941575110380942,
+      "grad_norm": 0.04551194980740547,
+      "learning_rate": 0.01,
+      "loss": 1.9223,
+      "step": 135777
+    },
+    {
+      "epoch": 13.941883150220761,
+      "grad_norm": 0.05770589038729668,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 135780
+    },
+    {
+      "epoch": 13.942191190060582,
+      "grad_norm": 0.052380964159965515,
+      "learning_rate": 0.01,
+      "loss": 1.9104,
+      "step": 135783
+    },
+    {
+      "epoch": 13.9424992299004,
+      "grad_norm": 0.13396403193473816,
+      "learning_rate": 0.01,
+      "loss": 1.9336,
+      "step": 135786
+    },
+    {
+      "epoch": 13.94280726974022,
+      "grad_norm": 0.11233116686344147,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 135789
+    },
+    {
+      "epoch": 13.943115309580039,
+      "grad_norm": 0.04093320667743683,
+      "learning_rate": 0.01,
+      "loss": 1.9089,
+      "step": 135792
+    },
+    {
+      "epoch": 13.943423349419858,
+      "grad_norm": 0.0822814404964447,
+      "learning_rate": 0.01,
+      "loss": 1.9069,
+      "step": 135795
+    },
+    {
+      "epoch": 13.943731389259678,
+      "grad_norm": 0.03724975138902664,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 135798
+    },
+    {
+      "epoch": 13.944039429099497,
+      "grad_norm": 0.0643593966960907,
+      "learning_rate": 0.01,
+      "loss": 1.914,
+      "step": 135801
+    },
+    {
+      "epoch": 13.944347468939316,
+      "grad_norm": 0.09097941219806671,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 135804
+    },
+    {
+      "epoch": 13.944655508779135,
+      "grad_norm": 0.10440767556428909,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 135807
+    },
+    {
+      "epoch": 13.944963548618954,
+      "grad_norm": 0.07616669684648514,
+      "learning_rate": 0.01,
+      "loss": 1.8999,
+      "step": 135810
+    },
+    {
+      "epoch": 13.945271588458773,
+      "grad_norm": 0.03374413028359413,
+      "learning_rate": 0.01,
+      "loss": 1.9383,
+      "step": 135813
+    },
+    {
+      "epoch": 13.945579628298594,
+      "grad_norm": 0.06170707941055298,
+      "learning_rate": 0.01,
+      "loss": 1.904,
+      "step": 135816
+    },
+    {
+      "epoch": 13.945887668138413,
+      "grad_norm": 0.05592031031847,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 135819
+    },
+    {
+      "epoch": 13.946195707978232,
+      "grad_norm": 0.04099148139357567,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 135822
+    },
+    {
+      "epoch": 13.946503747818051,
+      "grad_norm": 0.13028062880039215,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 135825
+    },
+    {
+      "epoch": 13.94681178765787,
+      "grad_norm": 0.0473533570766449,
+      "learning_rate": 0.01,
+      "loss": 1.9217,
+      "step": 135828
+    },
+    {
+      "epoch": 13.94711982749769,
+      "grad_norm": 0.0578739233314991,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 135831
+    },
+    {
+      "epoch": 13.94742786733751,
+      "grad_norm": 0.05764732509851456,
+      "learning_rate": 0.01,
+      "loss": 1.907,
+      "step": 135834
+    },
+    {
+      "epoch": 13.947735907177329,
+      "grad_norm": 0.03523023799061775,
+      "learning_rate": 0.01,
+      "loss": 1.8918,
+      "step": 135837
+    },
+    {
+      "epoch": 13.948043947017148,
+      "grad_norm": 0.0498935841023922,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 135840
+    },
+    {
+      "epoch": 13.948351986856967,
+      "grad_norm": 0.1308162361383438,
+      "learning_rate": 0.01,
+      "loss": 1.909,
+      "step": 135843
+    },
+    {
+      "epoch": 13.948660026696786,
+      "grad_norm": 0.04596000164747238,
+      "learning_rate": 0.01,
+      "loss": 1.91,
+      "step": 135846
+    },
+    {
+      "epoch": 13.948968066536604,
+      "grad_norm": 0.07514996081590652,
+      "learning_rate": 0.01,
+      "loss": 1.9044,
+      "step": 135849
+    },
+    {
+      "epoch": 13.949276106376425,
+      "grad_norm": 0.09542053937911987,
+      "learning_rate": 0.01,
+      "loss": 1.9227,
+      "step": 135852
+    },
+    {
+      "epoch": 13.949584146216244,
+      "grad_norm": 0.04413442686200142,
+      "learning_rate": 0.01,
+      "loss": 1.9323,
+      "step": 135855
+    },
+    {
+      "epoch": 13.949892186056063,
+      "grad_norm": 0.03635936975479126,
+      "learning_rate": 0.01,
+      "loss": 1.9222,
+      "step": 135858
+    },
+    {
+      "epoch": 13.950200225895882,
+      "grad_norm": 0.06238258630037308,
+      "learning_rate": 0.01,
+      "loss": 1.9024,
+      "step": 135861
+    },
+    {
+      "epoch": 13.950508265735701,
+      "grad_norm": 0.04232063516974449,
+      "learning_rate": 0.01,
+      "loss": 1.9004,
+      "step": 135864
+    },
+    {
+      "epoch": 13.950816305575522,
+      "grad_norm": 0.044490996748209,
+      "learning_rate": 0.01,
+      "loss": 1.9292,
+      "step": 135867
+    },
+    {
+      "epoch": 13.95112434541534,
+      "grad_norm": 0.03435138612985611,
+      "learning_rate": 0.01,
+      "loss": 1.9027,
+      "step": 135870
+    },
+    {
+      "epoch": 13.95143238525516,
+      "grad_norm": 0.040873702615499496,
+      "learning_rate": 0.01,
+      "loss": 1.9073,
+      "step": 135873
+    },
+    {
+      "epoch": 13.951740425094979,
+      "grad_norm": 0.0726858526468277,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 135876
+    },
+    {
+      "epoch": 13.952048464934798,
+      "grad_norm": 0.14251895248889923,
+      "learning_rate": 0.01,
+      "loss": 1.9126,
+      "step": 135879
+    },
+    {
+      "epoch": 13.952356504774617,
+      "grad_norm": 0.12145184725522995,
+      "learning_rate": 0.01,
+      "loss": 1.898,
+      "step": 135882
+    },
+    {
+      "epoch": 13.952664544614438,
+      "grad_norm": 0.06951028108596802,
+      "learning_rate": 0.01,
+      "loss": 1.9028,
+      "step": 135885
+    },
+    {
+      "epoch": 13.952972584454256,
+      "grad_norm": 0.053060855716466904,
+      "learning_rate": 0.01,
+      "loss": 1.92,
+      "step": 135888
+    },
+    {
+      "epoch": 13.953280624294075,
+      "grad_norm": 0.054118603467941284,
+      "learning_rate": 0.01,
+      "loss": 1.9261,
+      "step": 135891
+    },
+    {
+      "epoch": 13.953588664133894,
+      "grad_norm": 0.053629446774721146,
+      "learning_rate": 0.01,
+      "loss": 1.8931,
+      "step": 135894
+    },
+    {
+      "epoch": 13.953896703973713,
+      "grad_norm": 0.05916748568415642,
+      "learning_rate": 0.01,
+      "loss": 1.9279,
+      "step": 135897
+    },
+    {
+      "epoch": 13.954204743813534,
+      "grad_norm": 0.05659320950508118,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 135900
+    },
+    {
+      "epoch": 13.954512783653353,
+      "grad_norm": 0.048919253051280975,
+      "learning_rate": 0.01,
+      "loss": 1.8971,
+      "step": 135903
+    },
+    {
+      "epoch": 13.954820823493172,
+      "grad_norm": 0.04504438862204552,
+      "learning_rate": 0.01,
+      "loss": 1.9328,
+      "step": 135906
+    },
+    {
+      "epoch": 13.955128863332991,
+      "grad_norm": 0.06397843360900879,
+      "learning_rate": 0.01,
+      "loss": 1.9218,
+      "step": 135909
+    },
+    {
+      "epoch": 13.95543690317281,
+      "grad_norm": 0.1494799256324768,
+      "learning_rate": 0.01,
+      "loss": 1.9049,
+      "step": 135912
+    },
+    {
+      "epoch": 13.955744943012629,
+      "grad_norm": 0.09107448160648346,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 135915
+    },
+    {
+      "epoch": 13.956052982852448,
+      "grad_norm": 0.06978949159383774,
+      "learning_rate": 0.01,
+      "loss": 1.9165,
+      "step": 135918
+    },
+    {
+      "epoch": 13.956361022692269,
+      "grad_norm": 0.06903593987226486,
+      "learning_rate": 0.01,
+      "loss": 1.9291,
+      "step": 135921
+    },
+    {
+      "epoch": 13.956669062532088,
+      "grad_norm": 0.05308195948600769,
+      "learning_rate": 0.01,
+      "loss": 1.9266,
+      "step": 135924
+    },
+    {
+      "epoch": 13.956977102371907,
+      "grad_norm": 0.06638754159212112,
+      "learning_rate": 0.01,
+      "loss": 1.8918,
+      "step": 135927
+    },
+    {
+      "epoch": 13.957285142211726,
+      "grad_norm": 0.08520033955574036,
+      "learning_rate": 0.01,
+      "loss": 1.9263,
+      "step": 135930
+    },
+    {
+      "epoch": 13.957593182051545,
+      "grad_norm": 0.09176545590162277,
+      "learning_rate": 0.01,
+      "loss": 1.9001,
+      "step": 135933
+    },
+    {
+      "epoch": 13.957901221891365,
+      "grad_norm": 0.05668408423662186,
+      "learning_rate": 0.01,
+      "loss": 1.951,
+      "step": 135936
+    },
+    {
+      "epoch": 13.958209261731184,
+      "grad_norm": 0.09668008238077164,
+      "learning_rate": 0.01,
+      "loss": 1.9341,
+      "step": 135939
+    },
+    {
+      "epoch": 13.958517301571003,
+      "grad_norm": 0.13489456474781036,
+      "learning_rate": 0.01,
+      "loss": 1.9118,
+      "step": 135942
+    },
+    {
+      "epoch": 13.958825341410822,
+      "grad_norm": 0.10229591280221939,
+      "learning_rate": 0.01,
+      "loss": 1.9201,
+      "step": 135945
+    },
+    {
+      "epoch": 13.959133381250641,
+      "grad_norm": 0.06821364909410477,
+      "learning_rate": 0.01,
+      "loss": 1.9032,
+      "step": 135948
+    },
+    {
+      "epoch": 13.95944142109046,
+      "grad_norm": 0.04750967398285866,
+      "learning_rate": 0.01,
+      "loss": 1.9304,
+      "step": 135951
+    },
+    {
+      "epoch": 13.959749460930281,
+      "grad_norm": 0.052325982600450516,
+      "learning_rate": 0.01,
+      "loss": 1.9491,
+      "step": 135954
+    },
+    {
+      "epoch": 13.9600575007701,
+      "grad_norm": 0.059971246868371964,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 135957
+    },
+    {
+      "epoch": 13.960365540609919,
+      "grad_norm": 0.04755603149533272,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 135960
+    },
+    {
+      "epoch": 13.960673580449738,
+      "grad_norm": 0.09872281551361084,
+      "learning_rate": 0.01,
+      "loss": 1.8948,
+      "step": 135963
+    },
+    {
+      "epoch": 13.960981620289557,
+      "grad_norm": 0.07533475756645203,
+      "learning_rate": 0.01,
+      "loss": 1.8971,
+      "step": 135966
+    },
+    {
+      "epoch": 13.961289660129378,
+      "grad_norm": 0.03325020149350166,
+      "learning_rate": 0.01,
+      "loss": 1.9093,
+      "step": 135969
+    },
+    {
+      "epoch": 13.961597699969197,
+      "grad_norm": 0.06792011857032776,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 135972
+    },
+    {
+      "epoch": 13.961905739809016,
+      "grad_norm": 0.04931830242276192,
+      "learning_rate": 0.01,
+      "loss": 1.9211,
+      "step": 135975
+    },
+    {
+      "epoch": 13.962213779648835,
+      "grad_norm": 0.04478456825017929,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 135978
+    },
+    {
+      "epoch": 13.962521819488654,
+      "grad_norm": 0.051718439906835556,
+      "learning_rate": 0.01,
+      "loss": 1.9297,
+      "step": 135981
+    },
+    {
+      "epoch": 13.962829859328473,
+      "grad_norm": 0.0782901719212532,
+      "learning_rate": 0.01,
+      "loss": 1.916,
+      "step": 135984
+    },
+    {
+      "epoch": 13.963137899168292,
+      "grad_norm": 0.10739661753177643,
+      "learning_rate": 0.01,
+      "loss": 1.9229,
+      "step": 135987
+    },
+    {
+      "epoch": 13.963445939008112,
+      "grad_norm": 0.03352854400873184,
+      "learning_rate": 0.01,
+      "loss": 1.9167,
+      "step": 135990
+    },
+    {
+      "epoch": 13.963753978847931,
+      "grad_norm": 0.1394047886133194,
+      "learning_rate": 0.01,
+      "loss": 1.9275,
+      "step": 135993
+    },
+    {
+      "epoch": 13.96406201868775,
+      "grad_norm": 0.13387511670589447,
+      "learning_rate": 0.01,
+      "loss": 1.9442,
+      "step": 135996
+    },
+    {
+      "epoch": 13.96437005852757,
+      "grad_norm": 0.06781866401433945,
+      "learning_rate": 0.01,
+      "loss": 1.9191,
+      "step": 135999
+    },
+    {
+      "epoch": 13.964678098367388,
+      "grad_norm": 0.05729665979743004,
+      "learning_rate": 0.01,
+      "loss": 1.9021,
+      "step": 136002
+    },
+    {
+      "epoch": 13.964986138207209,
+      "grad_norm": 0.040621280670166016,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 136005
+    },
+    {
+      "epoch": 13.965294178047028,
+      "grad_norm": 0.044031303375959396,
+      "learning_rate": 0.01,
+      "loss": 1.9501,
+      "step": 136008
+    },
+    {
+      "epoch": 13.965602217886847,
+      "grad_norm": 0.031142249703407288,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 136011
+    },
+    {
+      "epoch": 13.965910257726666,
+      "grad_norm": 0.04875698313117027,
+      "learning_rate": 0.01,
+      "loss": 1.8927,
+      "step": 136014
+    },
+    {
+      "epoch": 13.966218297566485,
+      "grad_norm": 0.03648839890956879,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 136017
+    },
+    {
+      "epoch": 13.966526337406304,
+      "grad_norm": 0.11743724346160889,
+      "learning_rate": 0.01,
+      "loss": 1.9108,
+      "step": 136020
+    },
+    {
+      "epoch": 13.966834377246125,
+      "grad_norm": 0.04521479830145836,
+      "learning_rate": 0.01,
+      "loss": 1.9103,
+      "step": 136023
+    },
+    {
+      "epoch": 13.967142417085944,
+      "grad_norm": 0.09294033795595169,
+      "learning_rate": 0.01,
+      "loss": 1.9388,
+      "step": 136026
+    },
+    {
+      "epoch": 13.967450456925762,
+      "grad_norm": 0.14910796284675598,
+      "learning_rate": 0.01,
+      "loss": 1.8865,
+      "step": 136029
+    },
+    {
+      "epoch": 13.967758496765581,
+      "grad_norm": 0.06402064114809036,
+      "learning_rate": 0.01,
+      "loss": 1.9042,
+      "step": 136032
+    },
+    {
+      "epoch": 13.9680665366054,
+      "grad_norm": 0.07508663833141327,
+      "learning_rate": 0.01,
+      "loss": 1.9315,
+      "step": 136035
+    },
+    {
+      "epoch": 13.968374576445221,
+      "grad_norm": 0.0457422137260437,
+      "learning_rate": 0.01,
+      "loss": 1.897,
+      "step": 136038
+    },
+    {
+      "epoch": 13.96868261628504,
+      "grad_norm": 0.06341180950403214,
+      "learning_rate": 0.01,
+      "loss": 1.9185,
+      "step": 136041
+    },
+    {
+      "epoch": 13.96899065612486,
+      "grad_norm": 0.04527623578906059,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 136044
+    },
+    {
+      "epoch": 13.969298695964678,
+      "grad_norm": 0.06024854630231857,
+      "learning_rate": 0.01,
+      "loss": 1.9034,
+      "step": 136047
+    },
+    {
+      "epoch": 13.969606735804497,
+      "grad_norm": 0.03649662062525749,
+      "learning_rate": 0.01,
+      "loss": 1.8911,
+      "step": 136050
+    },
+    {
+      "epoch": 13.969914775644316,
+      "grad_norm": 0.05308336019515991,
+      "learning_rate": 0.01,
+      "loss": 1.9194,
+      "step": 136053
+    },
+    {
+      "epoch": 13.970222815484135,
+      "grad_norm": 0.14693909883499146,
+      "learning_rate": 0.01,
+      "loss": 1.9029,
+      "step": 136056
+    },
+    {
+      "epoch": 13.970530855323956,
+      "grad_norm": 0.07494723796844482,
+      "learning_rate": 0.01,
+      "loss": 1.9047,
+      "step": 136059
+    },
+    {
+      "epoch": 13.970838895163775,
+      "grad_norm": 0.06956320255994797,
+      "learning_rate": 0.01,
+      "loss": 1.9051,
+      "step": 136062
+    },
+    {
+      "epoch": 13.971146935003594,
+      "grad_norm": 0.04551749303936958,
+      "learning_rate": 0.01,
+      "loss": 1.9213,
+      "step": 136065
+    },
+    {
+      "epoch": 13.971454974843413,
+      "grad_norm": 0.03185529634356499,
+      "learning_rate": 0.01,
+      "loss": 1.9175,
+      "step": 136068
+    },
+    {
+      "epoch": 13.971763014683232,
+      "grad_norm": 0.08703722804784775,
+      "learning_rate": 0.01,
+      "loss": 1.9058,
+      "step": 136071
+    },
+    {
+      "epoch": 13.972071054523052,
+      "grad_norm": 0.1031704768538475,
+      "learning_rate": 0.01,
+      "loss": 1.8959,
+      "step": 136074
+    },
+    {
+      "epoch": 13.972379094362871,
+      "grad_norm": 0.06614328175783157,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 136077
+    },
+    {
+      "epoch": 13.97268713420269,
+      "grad_norm": 0.06636031717061996,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 136080
+    },
+    {
+      "epoch": 13.97299517404251,
+      "grad_norm": 0.10501495748758316,
+      "learning_rate": 0.01,
+      "loss": 1.9205,
+      "step": 136083
+    },
+    {
+      "epoch": 13.973303213882328,
+      "grad_norm": 0.04832559451460838,
+      "learning_rate": 0.01,
+      "loss": 1.9127,
+      "step": 136086
+    },
+    {
+      "epoch": 13.973611253722147,
+      "grad_norm": 0.03978385031223297,
+      "learning_rate": 0.01,
+      "loss": 1.922,
+      "step": 136089
+    },
+    {
+      "epoch": 13.973919293561968,
+      "grad_norm": 0.03864763304591179,
+      "learning_rate": 0.01,
+      "loss": 1.9325,
+      "step": 136092
+    },
+    {
+      "epoch": 13.974227333401787,
+      "grad_norm": 0.03821522369980812,
+      "learning_rate": 0.01,
+      "loss": 1.9008,
+      "step": 136095
+    },
+    {
+      "epoch": 13.974535373241606,
+      "grad_norm": 0.12083930522203445,
+      "learning_rate": 0.01,
+      "loss": 1.9134,
+      "step": 136098
+    },
+    {
+      "epoch": 13.974843413081425,
+      "grad_norm": 0.08757197856903076,
+      "learning_rate": 0.01,
+      "loss": 1.9087,
+      "step": 136101
+    },
+    {
+      "epoch": 13.975151452921244,
+      "grad_norm": 0.04910845309495926,
+      "learning_rate": 0.01,
+      "loss": 1.9317,
+      "step": 136104
+    },
+    {
+      "epoch": 13.975459492761065,
+      "grad_norm": 0.04811325669288635,
+      "learning_rate": 0.01,
+      "loss": 1.924,
+      "step": 136107
+    },
+    {
+      "epoch": 13.975767532600884,
+      "grad_norm": 0.05666341632604599,
+      "learning_rate": 0.01,
+      "loss": 1.9436,
+      "step": 136110
+    },
+    {
+      "epoch": 13.976075572440703,
+      "grad_norm": 0.05423150211572647,
+      "learning_rate": 0.01,
+      "loss": 1.9289,
+      "step": 136113
+    },
+    {
+      "epoch": 13.976383612280522,
+      "grad_norm": 0.08029211312532425,
+      "learning_rate": 0.01,
+      "loss": 1.9344,
+      "step": 136116
+    },
+    {
+      "epoch": 13.97669165212034,
+      "grad_norm": 0.07412763684988022,
+      "learning_rate": 0.01,
+      "loss": 1.8923,
+      "step": 136119
+    },
+    {
+      "epoch": 13.97699969196016,
+      "grad_norm": 0.09065306931734085,
+      "learning_rate": 0.01,
+      "loss": 1.9351,
+      "step": 136122
+    },
+    {
+      "epoch": 13.977307731799979,
+      "grad_norm": 0.08047862350940704,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 136125
+    },
+    {
+      "epoch": 13.9776157716398,
+      "grad_norm": 0.07536432892084122,
+      "learning_rate": 0.01,
+      "loss": 1.9283,
+      "step": 136128
+    },
+    {
+      "epoch": 13.977923811479618,
+      "grad_norm": 0.08685136586427689,
+      "learning_rate": 0.01,
+      "loss": 1.9264,
+      "step": 136131
+    },
+    {
+      "epoch": 13.978231851319437,
+      "grad_norm": 0.06133318692445755,
+      "learning_rate": 0.01,
+      "loss": 1.9084,
+      "step": 136134
+    },
+    {
+      "epoch": 13.978539891159256,
+      "grad_norm": 0.028963951393961906,
+      "learning_rate": 0.01,
+      "loss": 1.9032,
+      "step": 136137
+    },
+    {
+      "epoch": 13.978847930999075,
+      "grad_norm": 0.1561354696750641,
+      "learning_rate": 0.01,
+      "loss": 1.9036,
+      "step": 136140
+    },
+    {
+      "epoch": 13.979155970838896,
+      "grad_norm": 0.056513868272304535,
+      "learning_rate": 0.01,
+      "loss": 1.9188,
+      "step": 136143
+    },
+    {
+      "epoch": 13.979464010678715,
+      "grad_norm": 0.10373911261558533,
+      "learning_rate": 0.01,
+      "loss": 1.935,
+      "step": 136146
+    },
+    {
+      "epoch": 13.979772050518534,
+      "grad_norm": 0.11209435760974884,
+      "learning_rate": 0.01,
+      "loss": 1.9071,
+      "step": 136149
+    },
+    {
+      "epoch": 13.980080090358353,
+      "grad_norm": 0.05381729453802109,
+      "learning_rate": 0.01,
+      "loss": 1.9221,
+      "step": 136152
+    },
+    {
+      "epoch": 13.980388130198172,
+      "grad_norm": 0.0462588369846344,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 136155
+    },
+    {
+      "epoch": 13.98069617003799,
+      "grad_norm": 0.030146393924951553,
+      "learning_rate": 0.01,
+      "loss": 1.9334,
+      "step": 136158
+    },
+    {
+      "epoch": 13.981004209877812,
+      "grad_norm": 0.040314652025699615,
+      "learning_rate": 0.01,
+      "loss": 1.933,
+      "step": 136161
+    },
+    {
+      "epoch": 13.98131224971763,
+      "grad_norm": 0.12544989585876465,
+      "learning_rate": 0.01,
+      "loss": 1.9281,
+      "step": 136164
+    },
+    {
+      "epoch": 13.98162028955745,
+      "grad_norm": 0.045409802347421646,
+      "learning_rate": 0.01,
+      "loss": 1.9321,
+      "step": 136167
+    },
+    {
+      "epoch": 13.981928329397268,
+      "grad_norm": 0.11205153912305832,
+      "learning_rate": 0.01,
+      "loss": 1.9145,
+      "step": 136170
+    },
+    {
+      "epoch": 13.982236369237087,
+      "grad_norm": 0.05285382270812988,
+      "learning_rate": 0.01,
+      "loss": 1.9215,
+      "step": 136173
+    },
+    {
+      "epoch": 13.982544409076908,
+      "grad_norm": 0.0466182716190815,
+      "learning_rate": 0.01,
+      "loss": 1.9147,
+      "step": 136176
+    },
+    {
+      "epoch": 13.982852448916727,
+      "grad_norm": 0.09343729168176651,
+      "learning_rate": 0.01,
+      "loss": 1.9214,
+      "step": 136179
+    },
+    {
+      "epoch": 13.983160488756546,
+      "grad_norm": 0.045152872800827026,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 136182
+    },
+    {
+      "epoch": 13.983468528596365,
+      "grad_norm": 0.06445479393005371,
+      "learning_rate": 0.01,
+      "loss": 1.9144,
+      "step": 136185
+    },
+    {
+      "epoch": 13.983776568436184,
+      "grad_norm": 0.03444506973028183,
+      "learning_rate": 0.01,
+      "loss": 1.8771,
+      "step": 136188
+    },
+    {
+      "epoch": 13.984084608276003,
+      "grad_norm": 0.03456961363554001,
+      "learning_rate": 0.01,
+      "loss": 1.9088,
+      "step": 136191
+    },
+    {
+      "epoch": 13.984392648115822,
+      "grad_norm": 0.12383721023797989,
+      "learning_rate": 0.01,
+      "loss": 1.9025,
+      "step": 136194
+    },
+    {
+      "epoch": 13.984700687955643,
+      "grad_norm": 0.0358564555644989,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 136197
+    },
+    {
+      "epoch": 13.985008727795462,
+      "grad_norm": 0.03739408031105995,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 136200
+    },
+    {
+      "epoch": 13.98531676763528,
+      "grad_norm": 0.036199312657117844,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 136203
+    },
+    {
+      "epoch": 13.9856248074751,
+      "grad_norm": 0.08725354075431824,
+      "learning_rate": 0.01,
+      "loss": 1.9157,
+      "step": 136206
+    },
+    {
+      "epoch": 13.985932847314919,
+      "grad_norm": 0.0753713920712471,
+      "learning_rate": 0.01,
+      "loss": 1.923,
+      "step": 136209
+    },
+    {
+      "epoch": 13.98624088715474,
+      "grad_norm": 0.05375002324581146,
+      "learning_rate": 0.01,
+      "loss": 1.9137,
+      "step": 136212
+    },
+    {
+      "epoch": 13.986548926994558,
+      "grad_norm": 0.051483843475580215,
+      "learning_rate": 0.01,
+      "loss": 1.9146,
+      "step": 136215
+    },
+    {
+      "epoch": 13.986856966834377,
+      "grad_norm": 0.050593018531799316,
+      "learning_rate": 0.01,
+      "loss": 1.9057,
+      "step": 136218
+    },
+    {
+      "epoch": 13.987165006674196,
+      "grad_norm": 0.046737831085920334,
+      "learning_rate": 0.01,
+      "loss": 1.9254,
+      "step": 136221
+    },
+    {
+      "epoch": 13.987473046514015,
+      "grad_norm": 0.08650332689285278,
+      "learning_rate": 0.01,
+      "loss": 1.9226,
+      "step": 136224
+    },
+    {
+      "epoch": 13.987781086353834,
+      "grad_norm": 0.034098852425813675,
+      "learning_rate": 0.01,
+      "loss": 1.9128,
+      "step": 136227
+    },
+    {
+      "epoch": 13.988089126193655,
+      "grad_norm": 0.04019639641046524,
+      "learning_rate": 0.01,
+      "loss": 1.954,
+      "step": 136230
+    },
+    {
+      "epoch": 13.988397166033474,
+      "grad_norm": 0.042399823665618896,
+      "learning_rate": 0.01,
+      "loss": 1.9131,
+      "step": 136233
+    },
+    {
+      "epoch": 13.988705205873293,
+      "grad_norm": 0.09482115507125854,
+      "learning_rate": 0.01,
+      "loss": 1.9141,
+      "step": 136236
+    },
+    {
+      "epoch": 13.989013245713112,
+      "grad_norm": 0.11773832887411118,
+      "learning_rate": 0.01,
+      "loss": 1.8883,
+      "step": 136239
+    },
+    {
+      "epoch": 13.989321285552931,
+      "grad_norm": 0.18556223809719086,
+      "learning_rate": 0.01,
+      "loss": 1.9348,
+      "step": 136242
+    },
+    {
+      "epoch": 13.989629325392752,
+      "grad_norm": 0.13001282513141632,
+      "learning_rate": 0.01,
+      "loss": 1.8934,
+      "step": 136245
+    },
+    {
+      "epoch": 13.98993736523257,
+      "grad_norm": 0.042698659002780914,
+      "learning_rate": 0.01,
+      "loss": 1.9309,
+      "step": 136248
+    },
+    {
+      "epoch": 13.99024540507239,
+      "grad_norm": 0.07042375952005386,
+      "learning_rate": 0.01,
+      "loss": 1.8917,
+      "step": 136251
+    },
+    {
+      "epoch": 13.990553444912209,
+      "grad_norm": 0.06746090948581696,
+      "learning_rate": 0.01,
+      "loss": 1.9079,
+      "step": 136254
+    },
+    {
+      "epoch": 13.990861484752028,
+      "grad_norm": 0.04483184218406677,
+      "learning_rate": 0.01,
+      "loss": 1.9095,
+      "step": 136257
+    },
+    {
+      "epoch": 13.991169524591847,
+      "grad_norm": 0.03504500910639763,
+      "learning_rate": 0.01,
+      "loss": 1.9322,
+      "step": 136260
+    },
+    {
+      "epoch": 13.991477564431666,
+      "grad_norm": 0.0382150374352932,
+      "learning_rate": 0.01,
+      "loss": 1.9107,
+      "step": 136263
+    },
+    {
+      "epoch": 13.991785604271486,
+      "grad_norm": 0.05184762552380562,
+      "learning_rate": 0.01,
+      "loss": 1.9124,
+      "step": 136266
+    },
+    {
+      "epoch": 13.992093644111305,
+      "grad_norm": 0.04939550161361694,
+      "learning_rate": 0.01,
+      "loss": 1.9148,
+      "step": 136269
+    },
+    {
+      "epoch": 13.992401683951124,
+      "grad_norm": 0.04401058331131935,
+      "learning_rate": 0.01,
+      "loss": 1.8846,
+      "step": 136272
+    },
+    {
+      "epoch": 13.992709723790943,
+      "grad_norm": 0.05906231701374054,
+      "learning_rate": 0.01,
+      "loss": 1.9197,
+      "step": 136275
+    },
+    {
+      "epoch": 13.993017763630762,
+      "grad_norm": 0.1796485036611557,
+      "learning_rate": 0.01,
+      "loss": 1.9244,
+      "step": 136278
+    },
+    {
+      "epoch": 13.993325803470583,
+      "grad_norm": 0.05688506364822388,
+      "learning_rate": 0.01,
+      "loss": 1.9412,
+      "step": 136281
+    },
+    {
+      "epoch": 13.993633843310402,
+      "grad_norm": 0.05313657596707344,
+      "learning_rate": 0.01,
+      "loss": 1.9234,
+      "step": 136284
+    },
+    {
+      "epoch": 13.993941883150221,
+      "grad_norm": 0.08808745443820953,
+      "learning_rate": 0.01,
+      "loss": 1.9007,
+      "step": 136287
+    },
+    {
+      "epoch": 13.99424992299004,
+      "grad_norm": 0.040147144347429276,
+      "learning_rate": 0.01,
+      "loss": 1.9184,
+      "step": 136290
+    },
+    {
+      "epoch": 13.994557962829859,
+      "grad_norm": 0.030632702633738518,
+      "learning_rate": 0.01,
+      "loss": 1.9156,
+      "step": 136293
+    },
+    {
+      "epoch": 13.994866002669678,
+      "grad_norm": 0.0636623352766037,
+      "learning_rate": 0.01,
+      "loss": 1.9026,
+      "step": 136296
+    },
+    {
+      "epoch": 13.995174042509499,
+      "grad_norm": 0.16742970049381256,
+      "learning_rate": 0.01,
+      "loss": 1.9301,
+      "step": 136299
+    },
+    {
+      "epoch": 13.995482082349318,
+      "grad_norm": 0.0779375210404396,
+      "learning_rate": 0.01,
+      "loss": 1.9235,
+      "step": 136302
+    },
+    {
+      "epoch": 13.995790122189137,
+      "grad_norm": 0.09009319543838501,
+      "learning_rate": 0.01,
+      "loss": 1.8954,
+      "step": 136305
+    },
+    {
+      "epoch": 13.996098162028956,
+      "grad_norm": 0.07694835215806961,
+      "learning_rate": 0.01,
+      "loss": 1.8905,
+      "step": 136308
+    },
+    {
+      "epoch": 13.996406201868774,
+      "grad_norm": 0.07664139568805695,
+      "learning_rate": 0.01,
+      "loss": 1.934,
+      "step": 136311
+    },
+    {
+      "epoch": 13.996714241708595,
+      "grad_norm": 0.04531649872660637,
+      "learning_rate": 0.01,
+      "loss": 1.9012,
+      "step": 136314
+    },
+    {
+      "epoch": 13.997022281548414,
+      "grad_norm": 0.04568890854716301,
+      "learning_rate": 0.01,
+      "loss": 1.9293,
+      "step": 136317
+    },
+    {
+      "epoch": 13.997330321388233,
+      "grad_norm": 0.05323602631688118,
+      "learning_rate": 0.01,
+      "loss": 1.9274,
+      "step": 136320
+    },
+    {
+      "epoch": 13.997638361228052,
+      "grad_norm": 0.09728032350540161,
+      "learning_rate": 0.01,
+      "loss": 1.9324,
+      "step": 136323
+    },
+    {
+      "epoch": 13.997946401067871,
+      "grad_norm": 0.03730779141187668,
+      "learning_rate": 0.01,
+      "loss": 1.9312,
+      "step": 136326
+    },
+    {
+      "epoch": 13.99825444090769,
+      "grad_norm": 0.11414338648319244,
+      "learning_rate": 0.01,
+      "loss": 1.897,
+      "step": 136329
+    },
+    {
+      "epoch": 13.998562480747509,
+      "grad_norm": 0.08878182619810104,
+      "learning_rate": 0.01,
+      "loss": 1.8896,
+      "step": 136332
+    },
+    {
+      "epoch": 13.99887052058733,
+      "grad_norm": 0.04068625718355179,
+      "learning_rate": 0.01,
+      "loss": 1.9136,
+      "step": 136335
+    },
+    {
+      "epoch": 13.999178560427149,
+      "grad_norm": 0.049567922949790955,
+      "learning_rate": 0.01,
+      "loss": 1.8942,
+      "step": 136338
+    },
+    {
+      "epoch": 13.999486600266968,
+      "grad_norm": 0.0645836740732193,
+      "learning_rate": 0.01,
+      "loss": 1.9555,
+      "step": 136341
+    },
+    {
+      "epoch": 13.999794640106787,
+      "grad_norm": 0.039608463644981384,
+      "learning_rate": 0.01,
+      "loss": 1.9142,
+      "step": 136344
+    }
+  ],
+  "logging_steps": 3,
+  "max_steps": 136346,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 14,
+  "save_steps": 250,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.904273444271274e+21,
+  "train_batch_size": 21,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/training_args.bin b/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..c7f8be2ec300beb393c6cd12d2072c70656e6be9
--- /dev/null
+++ b/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33e7564ee02186bf7f4692a0a19fbd45a77ab5660f386f6632aa88dc4d7ea9de
+size 10680
diff --git a/zero_to_fp32.py b/zero_to_fp32.py
new file mode 100644
index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8
--- /dev/null
+++ b/zero_to_fp32.py
@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)